Citation
Estrogen-induced hepatic contributions to ovarian follicle development in Fundulus heteroclitus : vitellogenins and choriogenins

Material Information

Title:
Estrogen-induced hepatic contributions to ovarian follicle development in Fundulus heteroclitus : vitellogenins and choriogenins
Alternate title:
Vitellogenins and choriogenins
Creator:
LaFleur, Gary James, 1963-
Publication Date:
Language:
English
Physical Description:
viii, 142 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Amino acids ( jstor )
Codons ( jstor )
Complementary DNA ( jstor )
Egg proteins ( jstor )
Generally accepted auditing standards ( jstor )
Liver ( jstor )
Oocytes ( jstor )
Proteins ( jstor )
RNA ( jstor )
Vertebrates ( jstor )
Amino Acid Sequence ( mesh )
Chorion -- physiology ( mesh )
DNA, Complementary ( mesh )
Department of Anatomy and Cell Biology thesis Ph.D ( mesh )
Dissertations, Academic -- College of Medicine -- Department of Anatomy and Cell Biology -- UF ( mesh )
Egg Proteins -- chemistry ( mesh )
Egg Proteins -- genetics ( mesh )
Estradiol -- physiology ( mesh )
Killifishes ( mesh )
Liver -- physiology ( mesh )
Ovum -- growth & development ( mesh )
Ovum -- physiology ( mesh )
Vitellogenin -- chemistry ( mesh )
Vitellogenin -- genetics ( mesh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1996.
Bibliography:
Includes bibliographical references (leaves 125-140).
Additional Physical Form:
Also available online.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Gary James LaFleur.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright [name of dissertation author]. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
028193188 ( ALEPH )
49818829 ( OCLC )

Downloads

This item has the following downloads:


Full Text











ESTROGEN-INDUCED HEPATIC CONTRIBUTIONS
TO OVARIAN FOLLICLE DEVELOPMENT IN FUNDULUS HETEROCLITUS:
VITELLOGENINS AND CHORIOGENINS











by

GARY JAMES LaFLEUR, JR.

















A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

1996

































Copyright 1996

by

Gary James LaFleur, Jr.














ACKNOWLEDGMENTS



Much of the data presented in this dissertation was gained through cooperative

projects. The original cDNA librarians, Marion Byrne, Josna Kanungo, and Laura

Nelson, added a great gris-gris to the library. The N-terminal sequencing work was done

by the ICBR Protein Core under the supervision of Nancy Denslow, with special

contributions by Hung Nguyen and Sara Reynolds, who were particularly helpful and

patient in their protein isolation wizardry. Invaluable skill in artwork and photography

was contributed by Lynn Milstead and Jim Netherton.

I would like to thank the faculty members that invested an extra amount of

support to my academic training. Kelly Selman gave me my first view of what yolk and

the vitelline envelope really look like. Kyle Rarey was instrumental in my decision to

join the Dept. of Anatomy and Cell Biology, and offered a safe haven for my qualifying

exam studies. Gill Small introduced me to library screening techniques and library

screaming techniques. Dave Price was always willing to field questions on a wide

variety of subjects from specific molecular interactions to cooking recipes for

invertebrates.

My committee members were very supportive. Chris West walked me through

my first isolation of DNA; he was also the best Mardi Gras king I have ever seen. Paul

Linser was generous with space and comradery in his kind-hearted lab, the chicken wing.

iii









Michael Greenberg was truly inspirational, playing the King Arthur role at Camelot.

Rob Greenberg created a phenomenally free atmosphere in the molecular suite that was

very conducive to hard work and good fun. My adviser Robin Wallace was nothing less

than the perfect mentor. I have never learned so much from someone who said so little.

Robin never gave unsolicited advice. He let me grovel and groan and goof and grow,

taking delight in my triumphs and not noticing my failures. I will admire him always.

The molecular suite at the Whitney Lab, supervised by Rob Greenberg, has been

my incubator for five years. It offered a wonderful mix of data and good friends and

Rock and Roll. Primary fellow cloners that contributed to my journey included Bill

Buzzi, Bernd Eschweiler, Mike Jeziorski, Steve Munger, Clay Smith, and Chuck

Peterson. My late night comrade Sean Boyle deserves special mention for his Southern

kindness and great knack for developing shortcut protocols.

My mother and father's enthusiasm and unwavering support provided a

cornerstone of stability for all of my years as a student. Thus, their parental investment

in me, the product of their germ cells, has far-outdone that of any ordinary somatic

contribution.

Finally, I would like thank Susanna, a Texas girl that stole my Louisiana heart.

She is an excellent scientist and a perfect mother, but her real talent lies in her knack for

swinging the world by the tail, bouncing over the white clouds, and killing the blues.









iv
















TABLE OF CONTENTS

ACKNOWLEDGMENTS ...................................iii

TABLE OF CONTENTS ................................... v

ABSTRACT ............................................ vii

CHAPTERS
1 GENERAL INTRODUCTION ............................ 1
The Demands of the Germ Cell ........... .............. 1
The Original Emphasis: Vitellogenins ....................... 5
A New Emphasis: Estrogen-Induced Liver Proteins ... .. ...... 7

2 FUNDULUS HETEROCLITUS VITELLOGENIN:
THE DEDUCED PRIMARY STRUCTURE OF A
PISCINE PRECURSOR TO NON-CRYSTALLINE,
LIQUID-PHASE YOLK PROTEIN ........................ 11
Introduction .................... .................. 11
Material and Methods .................................. 15
Results ...................... ................... 22
Discussion ................... ................... 37

3 SEQUENCE COMPARISON OF FUNDULUS HETEROCLITUS
VITELLOGENINS I AND II............................. 46
Introduction ....... ........... .................... 46
Material and Methods ............... ................. 48
Results ...................... .................. 58
Discussion ..... .......... .. ..... ...... ........... 64

4 PRECURSOR-PRODUCT RELATIONSHIP OF VTGS I AND II
TO THE YOLK PROTEINS OF FUNDULUS HETEROCLITUS ...... 70
Introduction .............. ... ........... ........... 70
Material and Methods .... ........................... 72
Results ................. ...... .. ............... 76
Discussion ........... ........ .................... 81




v









5 FUNDULUS HETEROCLITUS CHORIOGENINS:
LIVER-DERIVED COMPONENTS OF THE VITELLINE
ENVELOPE AND CHORION SHARING SEQUENCE
IDENTITY WITH MAMMALIAN ZP PROTEINS .............. 87
Introduction ...................................... 87
Material and Methods ..................... ............ 90
Results ................. ......................... 98
Discussion ........................................ 111



6 GENERAL SUMMARY ................................ 121

REFERENCES ................................... 125

BIOGRAPHICAL SKETCH ........................... 141
































vi















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

ESTROGEN-INDUCED HEPATIC CONTRIBUTIONS TO
OVARIAN FOLLICLE DEVELOPMENT IN FUNDULUS HETEROCLITUS:
VITELLOGENINS AND CHORIOGENINS

By

Gary James LaFleur, Jr.

May 1996


Chairperson: Professor Robin A. Wallace, Ph.D.
Major Department: Anatomy and Cell Biology

We have increased our chances of isolating cDNAs that code for estrogen-induced

proteins by constructing a liver cDNA library from the poly A+RNA of Fundulus

heteroclitus treated with estradiol-17j3. We report cDNAs coding for two vitellogenins

(Vtg I and Vtg II) and three novel proteins that share identity with mammalian ZP

proteins. We have designated the latter proteins as "choriogenins" to highlight their role

as components of the vitelline envelope and chorion, yet to emphasize their site of

synthesis as being extra-ovarian, and thus different from that of the mammalian ZP

proteins.

Conceptual translations of the F. heteroclitus Vtg I and II cDNAs share 60%

sequence identity with each other and 30% identity with other reported vertebrate Vtgs.


vii









The N-terminus of a 69 kDa yolk protein matched the predicted N-terminus of Vtg II

(minus a signal peptide), verifying that Vtg II is expressed without being N-terminally

blocked. Six other yolk proteins were mapped to the predicted Vtg I sequence,

confirming that Vtg I represents the major yolk protein precursor. A 125-kDa yolk

protein that is specifically degraded during final maturation was mapped to a region of

the Vtg I sequence that contained a PEST site, suggesting an explanation for its

preferential break-down.

The three choriogenins were referred to as Chg 500, Chg 427, and Chg 553,

according to the number of amino acids predicted for each protein. Chg 500 and 553

were found to be 58% identical to a flounder "zp gene product", and 30% identical with

the mouse ZP1 protein. Chgs 500 and 553 contain proline-glutamine-rich repeating

regions that resemble a PXX motif reported in other extracellular matrix proteins. Chg

427 was found to be 67% identical to a medaka "L-SF protein" and 30% identitical to

the mouse ZP3 protein that has been implicated as the primary sperm receptor. Besides

reporting the sequences of five hepatically-derived proteins that contribute to the

development of the ovarian follicle, we emphasize that the estrogen-induced library is an

excellent strategy to screen for reproductively significant cDNAs.













viii














CHAPTER 1
GENERAL INTRODUCTION


The Demands of the Germ Cell


Reflecting on reproductive strategies of vertebrates, I am reminded of a once

familiar phrase used by an automobile repair shop: You can pay me now...or you can

pay me later." This is a fitting slogan, I think, to describe two alternative relationships

between germ cells and somatic cells, as manifested by different vertebrate groups.

Although the germ cells of all vertebrates are bound to receive an investment from their

associated somatic cells, this investment can be delivered either sooner or later according

to the specific developmental programs. As adults, ourselves, we may consider the

investment made by mothers to their young as an opportunity that is chosen by the

mother, voluntarily. However, this point of view has been described by some as "adult

chauvinism," biased toward the attitudes and experiences of the adult (Wallace, 1983).

An alternative view would be that the mother, or the somatic cells, are essentially held

captive by the germ cells, and (if healthy) have no choice but to respond when called

upon for support. As an illustration, consider the physiological state of the mummichog,

Fundulus heteroclitus (Fig 1.1). When the days of winter begin to grow long, and the

water temperature rises, the female mummichog does not have much say in the matter,

but her ovary begins to grow dramatically, mainly by the incorporation and storage of


1









2








































Figure 1.1 The mummichog, Fundulus heteroclitus, an estuarine teleost of the
Order Cyprinodontiformes as drawn by Lynn Milstead of the
Whitney Laboratory. The top fish is the female; the bottom fish,
displaying more pigment is the male.








3


yolk by the oocytes (vitellogenesis) (Taylor et al., 1977; Hsiao et al., 1994). The origin

of the yolk proteins can be traced to a cascade of events resulting in the maternal liver

synthesizing a suite of secreted proteins, primarily consisting of the yolk precursor,

vitellogenin (Vtg), but also containing riboflavin- and vitamin- binding proteins (White,

1987; White and Merrill, 1988) and most recently discovered, precursors of the vitelline

envelope (Hamazaki et al., 1985; Murata, et al., 1991; Hyllner et al., 1991). Thus, the

oocytes, or germ cells, demand an investment by the maternal or somatic cells. They

are saying, "Pay me now." This extensive investment begins long before fertilization,

without the adult knowing whether the eggs will actually ever be spawned or fertilized.

Once the oocytes are expelled, the female, having already surrendered a substantial

amount of energy and material, is relieved of any further investment (until the next clutch

of oocytes begins its demands).

On the other hand, in mammals the germ cells present more of a "Pay me later"

scenario. Mammalian oocytes appear to not receive any yolk at all, with synthesis of

vitellogenin presumed (but not proven) to be totally nonexistent in mammals (except in

the egg laying monotremes) (Eckelbarger, 1994). The investment, then, comes mainly

after fertilization, with support and nourishment provided first by a modification of the

uterus into the chorionic villi, and secondly through lactation, where protein nourishment

continues to be demanded by the progeny, and thus supplied by the adult.

The work contained in this dissertation provides an example of the "Pay me now"

demands of the oocyte on its somatic surroundings. We provide evidence of at least five

distinct proteins that are made by the maternal liver, in response to estradiol, and










__ 4




j r






%"# 'l) *1








4j






IIf





_e ~ 1z A











Figure 1. 2 A transmission electron micrograph providing an ultrastructural view of
the environment surrounding the oocyte membrane. To the bottom left is
the cytoplasm of the oocyte including a yolk sphere (arrow) containing
processed yolk proteins, derived from Vtg. Distal to the oocyte
membrane is the stratified appearance of the vitelline envelope (bracket),
containing components derived from the choriogenins. This micrograph
was kindly provided by Kelly Selman (X 12,200).








5

transported to the ovary, to be used by the germ cells and their descendants. Two of

these proteins, Vtg I and Vtg II, are endocytosed by the oocyte, processed, and stored

as yolk (Fig. 1.2), mainly to be used as a nutrient source by the developing embryo.

The three remaining proteins, designated the choriogenins (Chgs), are also synthesized

by the estrogen-induced liver, and transported to the ovary. However, rather than being

endocytosed, the Chgs are laid down as an extracellular matrix between the oocyte and

follicle cells forming the vitelline envelope (Fig. 1.2, in brackets).


The Original Emphasis: Vitellogenins


One of the initial goals of this project was to establish a definitive precursor-

product relationship between vitellogenin and the processed yolk proteins. It was decided

that primary sequence information would be needed for this goal and that the best method

to gain the amino acid sequence of vitellogenin was to use a molecular approach, produce

a cDNA library, screen for Vtg with degenerate primers designed from yolk proteins,

and sequence the cDNA clone. Before the lengthy Vtg sequence was completed, the

original research team disbanded. I subsequently joined the Wallace lab and thereby

"inherited" the Vtg sequencing project. Influenced by the dissertation of Byrne (1989)

describing the evolution of yolk proteins, I became interested in the evolutionary aspects

of Vtg, particularly in the independently evolving phosvitin domain. The lack of a

phosvitin domain in the Caenorhabditis elegans Vtgs (Speith et al., 1991) prompted the

idea that phosvitin may be an exclusively vertebrate inclusion within the Vtg gene (Byrne








6

et al., 1989). We theorized that an interesting oviparous model may be provided by the

protochordate Branchiosotma floridae, the Florida lancelet.

The original aims of my project were, thus, to complete the F. heteroclitus Vtg

cDNA sequence, and thereafter use the piscine cDNA as a heterologous probe to isolate

phylogenetically primitive Vtgs. I succeeded in the former goal, and the results of that

work are provided in Chapter 2. I began screening a cDNA library synthesized from the

MRNA of spawning female amphioxus, B. floridae, by a PCR-based method that utilized

degenerate primers designed by aligning the currently known Vtg protein sequences.

Before long, a new Vtg cDNA was successfully isolated. However, the new Vtg was

isolated from the "control" F. heteroclitus library template, rather than the targeted

amphioxus library template (Fig. 1.3). At that time, no two Vtgs from one vertebrate

species had yet been completely sequenced, and so this appeared to be a worthwhile

challenge. Additionally, the sequence of two F. heteroclitus Vtgs would provide

information presumably necessary to continue mapping out the precursor-product

relationships of the Vtgs and the yolk proteins. As a result, phylogenetic aspects of Vtg

evolution were shelved in order to consider the variations of Vtg that might be

encountered from within one species, F. heteroclitus. The second primary aim of my

project was thus to complete the Vtg II cDNA; this data is provided in Chapter 3.

While completing the two Vtg cDNA sequences, N-terminal sequences of the yolk

proteins were also being obtained. This work represented a collaborative effort that

included data collected from three students at the Whitney lab (including myself) plus a

considerable effort by the ICBR Protein Core facility. Eventually we established a









7

scheme mapping out the specific processing of two Vtgs into several separate yolk

protein products. We have further submitted a hypothesis implicating a PEST site found

within the predicted YP 125 sequence as a possible factor influencing its extensive

degradation. This study is presented in Chapter 4.


A New Emphasis: Estrogen-Induced Reproductive Proteins


While completing the Vtg II cDNA sequence, using a PCR-based screening

method, other non-target cDNAs were often isolated. This is a common phenomenon

in cloning that is usually dismissed as misfortune. However, because our template was

an estrogen-induced cDNA library, the non-target cDNAs stood a likely chance of

representing reproductively significant molecules. This was exactly the case concerning

the Chgs. All three of the Chgs cDNAs were isolated by a fortuitous mis-priming event

that occurred while screening for Vtg II cDNAs (Fig. 1.3). Only recently had a

hypothesis been submitted that ascribed the origin of the major proteins of the teleost

vitelline envelope to the estrogen-induced liver (Hamazaki et al., 1987b). This ran

counter to the mammalian literature that had established the oocyte as the primary site

of synthesis for the proteins of the mammalian zona pellucida (Wassarman, 1988a).

Nevertheless, our data verifies that several Chgs are in fact synthesized by the liver,

transported to the ovary, and laid down as the vitelline envelope between the oocyte and

the follicle cells. These proteins have been referred to by several names. When isolated

from the ovarian follicle, they are usually called vitelline envelope proteins

(VEPs)(Hyllner et al., 1991). Another nomenclature based on isolating the proteins from










8








Vtg 'I 4650 CTGGATAGAGAGGCCAGACTGTGGGCTCTGCGGAAAGGCCGACGGGGAAGTCAGACAGG 4693
TGTGGICTCTGCGGIAAIAACGA
ROW 19 (degenerate) C G T CG T


Chg 500 918 TCCTGGACCTCTGCGGTGGGAGCTCAGGCTTGGGAATGGAGAGTGTTCTGTCAAGGGTT 975
***w**** ** *
ROW 45 GAGCTCAGTCTGTACACTGCT


Chg 427 669 CAGCCTTCCTCTGGATCCCCTTTGGGTCCCATTCTCTGCAGTTAAGATGGCTGAGGAGT 698

ROW 55 CATTCTGAAACT.TGAAGACCC


Chg 553 820 CTCATTGTTGGGAGGAGGTCAAGGCTGTACACATGTTGACCCCAATTC.CTTTTTGCCA 878
*ww *** ***** **
ROW 45 GAGCTCA-GTCTGTACACTGCT


























Figure 1.3 Four accounts of fortuitous annealing that resulted in the eventual
isolation of cDNAs coding for Vtg II, Chg 500, Chg 427, and Chg
553. Vtg II was discovered using the degenerate primer ROW 19.
Chg 500 and Chg 553 were discovered using ROW 45, originally
designed for annealing to Vtg II. Chg 427 was isolated using
ROW 55, also designed to anneal to Vtg II.








9

the blood of spawning females used the terms "low molecular weight spawning female

specific substance (L-SF), and high molecular weight spawning female-specific substance

(H-SF)" (Hamazaki et al., 1987a). Still other groups that concentrated on sequence

identity between their teleost proteins and the published mammalian ZP proteins, referred

to their sequences as teleost ZPs (Lyons et al., 1993). We designated the cDNAs and

coded proteins described here as choriogenins (Chgs), precursor proteins of the vitelline

envelope and chorion. We feel that this name accentuates the role of these molecules as

structural components of the vitelline

envelope and chorion, yet emphasizes their origin as being extra-ovarian and thus

different from the homologous ZP proteins of mammals. In Chapter 5, we present the

cDNA and protein sequences of three Chgs, as well as a partial characterization of F.

heteroclitus VEPs. The Chg data represent the most novel aspect of the dissertation,

with the hypothesis of liver-derived vitelline envelope components still fairly recent. One

of the remaining paradoxes presented by the Chg sequences is the comparative disparity

between the mammalian and teleostean systems for producing the extracellular matrix that

surrounds the oocyte. Because the sequence identity between the Chgs and mammalian

ZP proteins suggests an ancestral relationship, the differences in gene regulation, site of

synthesis, and functional roles offer a wealth of interesting questions for future

investigations.

By providing the structure of five previously unsequenced molecules that

contribute to the architecture of the ovarian follicle, we have contributed to our

understanding of reproductive processes in F. heteroclitus. However, we are even more








10

impressed by the remarkable resource proven to lie within the estrogen-induced liver

library that was used to isolate these cDNAs. Rather than being the means to an end,

the library has rather been venerated as possibly the most important attribute of the

project. We expect that other estrogen-induced liver products can be easily isolated from

it, and modifications of this strategy can be used in the future to investigate other

inductive hormone effects on other tissues.















CHAPTER 2
FUNDULUS HETEROCLITUS VITELLOGENIN:
THE DEDUCED PRIMARY STRUCTURE OF A PISCINE PRECURSOR TO
NON-CRYSTALLINE, LIQUID-PHASE YOLK PROTEINS





Introduction


Vitellogenin (Vtg) is a large phosphoglycoprotein (-200 Kda) used by most

oviparous animals as a maternally derived yolk precursor (Pan et al., 1969; Kunkel and

Nordin, 1985; Wallace, 1985; Selman and Wallace 1989). It is synthesized by either the

liver (vertebrates), fat body (insects), or intestine (nematodes) under hormonal induction

and transported to growing oocytes via the blood (Flickinger and Rounds, 1956; Wallace

and Jared, 1969). Vtg is incorporated into oocytes by receptor-mediated endocytosis

(Opresko et al., 1980; Opresko and Wiley, 1987; Shen et al., 1993) and is stored for

later use by the developing embryo (Flickinger, 1960; Yamagami, 1960; Karasaki,

1963b; Selman and Pawsey, 1965; Murakami et al., 1990). Once inside the oocyte, Vtg

is processed into smaller yolk proteins consisting of lipovitellins (Lvl and Lv2),

phosvitins (Pv), and phosvettes, that may in turn be degraded into even smaller cleavage

products (Flickinger and Rounds, 1956; Taborsky, 1967; Wallace and Selman, 1985;

Gerber-Huber et al., 1987; Greeley et al., 1986).


11








12

The now familiar Vtg gene family (Wahli et al., 1979, 1991; Tata et al., 1980;

Blumenthal et al., 1984; Nardelli et al., 1987; Byrne et al., 1989; Speith et al., 1991)

encompasses Vtgs synthesized by a wide range of metazoans including the nematode

Caenorhabditis elegans (Speith et al., 1985), the boll weevil Anthonomus grandis

(Trewitt et al., 1992), the silkworkm Bombyx mori (Yano et al., 1994), the mosquito

Aedes aegypti (Chen et al., 1994) the cyclostome Ichthyomyzon unicuspis (Sharrock et

al., 1992), the anuran Xenopus laevis (Germond et al., 1984; Gerber-Huber et al., 1987),

and the chicken Gallus domesticus (van het Schip et al., 1987). Additionally, two human

cDNAs, those encoding von Willebrand factor (- 250 kDa) (Baker, 1988a) and

apolipoprotein B-100 (-510 Kda) (Baker, 1988b), have also been reported as distantly

related members of the Vtg gene family. Exceptions to a Vtg-derived yolk precursor

system have been reported in at least two dipteran species: Drosophila melanogaster

(Hovemann et al., 1981) and Ceratitis capitata (Rina and Savakis, 1991) where yolk

precursors, often called Vtgs, do not, in fact, share significant sequence identity with the

"Vtg gene family" setting a precedent for the use of alternative molecules in the

production of yolk (Terpestra and AB, 1988; Bownes, 1992).

A large component of vertebrate Vtgs, the Pv region, was found to be nonexistent

in both C. elegans and the boll weevil Vtg (Nardelli et al., 1987; Trewitt et al., 1992),

inspiring the notion that Pv was an element unique to vertebrate Vtgs. The apparent

absence of the Pv region from invertebrate Vtgs (see Discussion), along with studies

documenting the ability of the phosphate groups of Pv to bind and transport large

amounts of divalent cations, especially Ca"+ (Urist et al., 1958; Urist and Schjeide,









13

1961; Taborsky, 1980), have led to the speculation that Pv may be important in

embryonic bone formation (Mecham and Olcott, 1949; Rabinowitz, 1962; Taborsky,

1974; Lange, 1981; Wallace and Begovac, 1985; Nardelli et al., 1987; Byrne et al.,

1989). Of additional interest is the hypothesis that evolutionary changes in the Pv region

have occurred at a faster rate than in the two flanking regions, Lvl and Lv2 (Byrne et

al., 1989). To address comparative and evolutionary questions about Vtg, we sought to

characterize a Vtg cDNA that was phylogenetically intermediate to the meager collection

of currently reported sequences. Complete Vtg sequences from the superclass

Gnathostomata have been reported from only two tetrapods (Xenopus and chicken)

leaving several entire lower vertebrate classes unrepresented. Since at least half of all

vertebrates are contained within the subclass Teleostei (Nelson, 1984), the absence of a

teleostean Vtg sequence leaves a substantial gap in our understanding of Vtg evolution,

diversity, and function.

For the present study, we chose as a model the estuarine teleost, Fundulus

heteroclitus, which possesses a non-specialized body plan with a fairly typical

reproductive system, in the hopes of obtaining a piscine Vtg that could be considered as

representative of most teleosts. Much work has already been reported on F. heteroclitus

describing vitellogenesis (Wallace and Selman, 1978, 1981; Selman and Wallace, 1983;

Kanungo et al., 1990), the resulting yolk proteins (Wallace and Begovac, 1985; Wallace

and Selman, 1985; Greeley et al., 1986), and oocyte maturation (Wallace and Selman,

1978, 1980). Besides the advantages of F. heteroclitus possessing many typical

teleostean traits, there are at least two characteristics of its yolk that presented additional









14

motivation for our comparative analyses. First, the yolk proteins of F. heteroclitus

oocytes and eggs remain in a liquid form throughout oocyte growth and maturation

(Wallace et al., 1966; Wallace and Begovac, 1985); this is in marked contrast to the

more typical observation that vertebrate yolk proteins are organized into a specific

crystalline lattice as was reported in lamprey (Karasaki, 1967; Raag et al., 1988),

sturgeon (Lange and Kilarski, 1986), several amphibians (Karasaki, 1963a), and the

reptile, tuatuara (Lange and Kilarski, 1986; reviews by Lange, 1985, Banaszak et al.,

1991). Second, whereas Xenopus and chicken yolk remains in the form of three primary

Vtg cleavage products, Lv,, Pv, and Lv2 plus a few minor peptides or phosvettes (Wiley

and Wallace, 1981; Wallace and Morgan, 1986a, 1986b; Wallace et al. 1990), F.

heteroclitus yolk proteins undergo substantially more processing, resulting in a complex

suite of smaller Vtg-derived cleavage products (Wallace and Begovac, 1985; Wallace and

Selman, 1985; Greeley et al., 1986). We hoped that by obtaining the primary structure

of a teleostean Vtg we would not only confirm regions that are ubiquitously conserved

among oviparous organisms, but would also reveal novel sequence differences that play

a role in the yolk processing events unique to F. heteroclitus.

In this paper we present the predicted primary structure of F. heteroclitus Vtg.

By aligning the F. heteroclitus Vtg sequence to other vertebrate Vtgs, we found that the

most significant differences occurred within the polyserine domain. These differences

may account for some of the molecular phenomena specifically associated with F.

heteroclitus yolk, such as the perpetuation of a liquid phase yolk in both oocytes and

eggs, or the substantial amount of proteolytic processing which occurs in the growing









15

oocytes. Although the polyserine domain is indeed a polymorphic region, a conserved

genetic pattern (Byme et al., 1984, 1989) persists in all of the vertebrates thus far

examined: TCX repeats at the 5' end and a larger group of AGY repeats towards the 3'

end, suggesting an ancient origin of the linkage between these two clusters of

trinucleotide repeats.


Materials and Methods


Chemicals


Estradiol-173 was obtained from Sigma Chemical Co. (St. Louis, MO).

Radioisotopes, [Ca-32P] dCTP and [a-35S] dATP, were purchased from New England

Nuclear (Boston MA). Lambda gtlO vector and cDNA synthesis reagents were obtained

from Promega (Madison, WI). The subcloning plasmid pGem-3Z was purchased from

Promega, pT7BLUE from Novagen (Madison, WI), and pCR1000 from Invitrogen (San

Diego, CA). All sequencing gels were cast using Sequagel-8 (National Diagnostics,

Atlanta) polyacrylamide reagents. Amplification reactions were performed using

Thermophilus aquaticus DNA polymerase (Promega). Sequenase version 2.0 DNA

polymerase and dideoxy sequencing reagents were obtained from US Biochemicals

(Cleveland, OH). Reagents for random-primed labeling of probes were purchased from

Pharmacia (Piscataway, NJ). Both Nytran nylon and S&S NC nitrocellulose transfer

membranes were purchased from Schleicher and Schuell (Keene, NH). Amino acid N-

terminal sequencing and synthesis of oligonucleotide primers were performed by the

University of Florida Interdisciplinary Center for Biotechnology Research core facility.








16


Induction of vitellogenin synthesis


Male Fundulus heteroclitus were collected from the estuarine creeks adjacent to

the Whitney Laboratory, and were maintained in running seawater tanks under 14L: 10D

photoperiod conditions at 25 + 20C. Fish were maintained for at least one month before

being used for RNA collections.

In order to increase the proportion of Vtg RNA within the total RNA pool,

vitellogenin synthesis was artificially induced in six males (8-10 g body weight) by two

intraperitoneal injections of estradiol-170 (0.01 mg/g body weight) dissolved in 50 1

peanut oil (Kanungo et al., 1990). Five control males were sham-injected with peanut

oil alone. The first injection was performed on day 1, the second injection on day 4,

followed by sacrifice and liver dissection on day 8.


Isolation of liver poly A+ RNA


Livers from both groups of fish were collected and immediately placed in 0C

guanidinium thiocyanate solution (5M guanidinium thiocyanate, 50 mM Tris-Hcl, 25 mM

EDTA, 8% v/v mercaptoethanol, pH 7.4) and homogenized by one thirty-second

polytron (Brinkman) blast. RNA was isolated by the guandidinium thiocyanate method

according to MacDonald et al. (1987). One gram of liver from estrogen-treated fish

yielded an average of 0.536 mg RNA, with an average O.D. 260/280 ratio of 2.03,

while a gram of liver from sham-treated fish yielded an average of 0.318 mg RNA with

an O.D. 260/280 of 1.87. Total RNA samples were combined into two pools: one








17

from six estrogen-treated males and the other from five sham-treated males.

Oligo-dT cellulose chromatography was used to isolate poly A+ RNA from the

two initial pools of total RNA (Aviv and Leder, 1972). Of the 2.1 mg total RNA from

estrogen-treated fish, 46.3 sg poly A+ RNA was recovered (2.2% recovery). Poly A'

RNA from both experimental and control fish was analyzed by northern blot analysis to

verify that Vtg transcripts were included in the poly A+ RNA fraction. The poly A+

RNA was dissolved in deionized glyoxal/DMSO (1:1) and electrophoresed through an

agarose gel (McMaster and Carmichael, 1977) and transferred by capillary action onto

a nylon membrane. The membrane was probed with a "2P end-labeled 17-mer

oligonucleotide, MB6 (degeneracy = 32) which was designed from the N-terminal amino

acid sequence of a small yolk peptide isolated from F. heteroclitus oocytes: His-Lys-

Lys-Met-Val-Ala. Autoradiography of northern blots revealed an MB6 positive, 6 kb

transcript found in the estrogen-treated fish which was absent in sham-treated male fish.

This transcript size was consistent with Vtg cDNA previously reported from chicken

(Cozens et al., 1980; Arnberg et al., 1981; van het Schip et al., 1987), frog (Whali et

al., 1979), and rainbow trout (Le Guellec et al., 1988).


cDNA library construction, screening, and sequencing


Synthesis of cDNA was performed by annealing 2 jug poly A+ RNA with oligo

dT primers, and using AMV reverse transcriptase and T4 DNA pol I for first and second

strand synthesis respectively. Eco R1 adapters were ligated to the two ends of the cDNA

transcripts using T4 DNA ligase (these Eco RI adapters were later found to have become








18

compromised). After phosphorylation of adapter ends, the cDNA transcripts were ligated

into the bacteriophage vector X gtlO (Promega). Once the X particles were packaged,

the primary library was plated using host E. coli strain C600HFL, resulting in an initial

library titer of 6 x 10 total plaque forming units.

Two 24.5 cm2 petri dishes were used for plating phage-transfected cells

(-400,000 total plaques). Plaques were lifted onto nylon membranes. Hybridization

was performed at 39"C using lX Denhardt's solution (Denhardt, 1966), 6X SSC (150

Mm NaCl and 15 Mm sodium citrate, pH = 7) with the same end-labelled

oligonucleotide probe as described earlier, MB6. The primary screening resulted in 30%

of the plaques testing positive for the degenerate MB6 probe. By following four plaque

clones (X5, X20, X21, X16) through two more rounds of positive screening, four final X

clones were set aside for subcloning. The clone (X21) containing the largest insert

(-5000 bp) was subjected to endonuclease digestion with EcoR1, which was expected

to free the entire cDNA insert. Unfortunately the EcoR1 sites had inadvertently been

modified so that when digested with EcoR1, one end of the insert remained attached to

the vector. Alternatively, the enzymes HindIll and BglI, which like EcoR1 cleave at

rare sites, were used to digest X21. Two large fragments were released (1900 bp from

EcoR1/BglII and a 2060 bp fragment from EcoRl/HindIII) and these were subcloned into

the sequencing plasmid PGEM 3Z resulting in subclones pMMB6 and pMMB1,

respectively. Because the size of these two fragments did not add up to the total putative

insert size (6000 bp), digestion of another clone, X5, was performed in order to provide








19

an overlapping sequence. Digestion with HindIII and EcoR1 yielded a third fragment,

1610 bp, which was subcloned as pMMB9.

Dideoxynucleotide chain termination sequencing of these three clones revealed

that there were two remaining nucleotide stretches that were needed to complete the

entire cDNA: a small 5' portion which included the initial methionine codon and a 300

bp overlap between pMMB9 and pMMBl. Both of these additional portions were

retrieved from the cDNA library by PCR techniques. First, the initiating methionine was

retrieved by using an exact forward primer (NEB #1231) complementary to the Xgtl0

primer adapter sequence and an exact reverse primer, ROW 1, 195 base pairs internal

to the existing 5' end. The resulting product was gel-purified and ligated into the

sequencing plasmid pT7BLUE by the T/A cloning method.

The overlap between pMMB9 and pMMB1 was retrieved in a similar fashion by

using two exact internal primers, ROW 12 and ROW 13, made according to the existing

ends of PMMB9 and PMMB1. The resulting product was gel-purified and ligated into

a similar T/A plasmid, pCR1000. These two PCR inserts were sequenced and found to

overlap with the already existing sequence resulting in a 5112 bp open reading frame

from which we have deduced the complete primary structure of the putative Fundulus

heteroclitus Vtg polypeptide.


Sequence analysis


Sequencing data were organized and examined using PC\GENE software

(Intelligenetics, Mountain View, CA) including the following analyses: predictions









20











CONTIGUOUS cDNA SEQUENCE 5112 bp

ofigo MB6 X #21
pGL8 -- I-I-I
pMMB6 --.---- oligo MB6 X#5
SpMM--B9 P--
3sman plasmid insert pMMBI
lambda insert
-. degenerate probe



















Figure 2.1 Cloning strategy used in isolating Fundulus heteroclitus Vtg
cDNA. Lambda gtlO bacteriophage clones #5 and #21 were
isolated by tertiary screening with degenerate 17mer, MB6.
pGEM 3Z subclones (pMMB6 and pMMBl) were constructed
from digestion products of X21 and pMMB9 originated from X5.
Two remaining sections, pGL8 and pGL5, were isolated by
anchored PCR, using a 5-p1 aliquot of the cDNA library as
template and exact primers, and then inserted into pT7Blue and
pCR1000 vectors, respectively.








21

of post-translational modification sites by PROSITE, signal peptide prediction by

PSIGNAL, antigenic determinant analysis using ANTIGEN, codon usage statistics by

CDUSAGE.

Protein sequence alignments were performed using two programs: ClustalV

(Higgins et al., 1992), utilizing the PAM 250 matrix, gap penalty=3, K-tuple= 1, no.

of top diagonals=5; window size=5) for the multiple alignment, and ALIGN Plus (S&E

Software, State Line, PA) for pairwise alignments. To normalize domain comparisons,

we defined a "polyserine domain" within the Vtg sequence by choosing two well-aligned

termini as the exterior boundaries, thereby including all of the poorly-aligned polyserine

tracts on the interior. Because we do not have yolk protein data to map the exact region

which is processed into Pv, we have chosen this "polyserine domain" to represent a

hypothetical Pv domain. The chicken and Xenopus Pv termini, which have been

documented (Clark, 1973; Gerber-Huber et al., 1987), lie to the inside of our

boundaries, verifying our convention.

A phylogram was drawn to compare Vtg sequences from eight species. Although

multiple isoforms of Vtg have been identified from several organisms, nomenclature

formally separating these isoforms into subfamilies has not yet been proposed. For our

tree analysis we selected only one Vtg sequence from each species. In species which

contain multiple Vtgs, we chose either the only complete Vtg available from Genbank

databases, as in chicken and Xenopus, or the Vtg which is considered the "major" yolk

protein precursor, as in C. elegans (Speith et al., 1985). An optimal tree was chosen by

importing a ClustalV alignment into the program PAUP (Swofford, 1983) and performing









22



bootstrap analysis of 100 replicates in a branch-and-bound search. C. elegans Vtg 5 was

defined as the outgroup and chicken Vtg was designated as the reference sequence.

Accession codes of sequences used for alignments are as follows: Chicken,

Gallus domesticus Vtg II, EMBL:X13607; Xenopus laevis Vtg A2, GB:M18061; silver

lamprey, Ichthyomyzon unicuspis Vtg, GB:M88749; white sturgeon, Acipenser

transmontanus, partial Vtg, GB:U00455; rainbow trout, Oncorhynchus mykiss partial

Vtg GB:M27651; tilapia, Oreochromis aureus partial Vtg, number not available (Ding

et al., 1990); boll weevil, Anthonomus grandis Vtg, GB:M72980; nematode,

Caenorhabditis elegans Vtg 5, EMBL:X56213; mosquito, Aedes aegypti Vtg,

GB:U02548; and finally, our own mummichog, Fundulus heteroclitus Vtg, GB:U07055.


Results


Cloning


A summary of our cloning strategy is presented in Figure 2.1 Three restriction

products of two MB6-positive lambda clones (#21 and #5) were subcloned into PGEM

3Z (PMMB1, PMMB6, and PMMB9). Two smaller clones pGL8 and pGL5 were

amplified by PCR directly from the cDNA library. The five subclones were sequenced

in both directions for a final overlapping cDNA sequence of 5198 bp. The overlapping

cDNA sequence contained an open reading frame of 5112 bp and a poly-A tail of

undetermined length beginning 11 nucleotides after a poly-adenylation site (AATAAA)

denoted by underlining in Figure 2.2.















Figure 2.2 Translated amino acid sequence (1,704 residues) of the putative F.
heteroclitus Vtg polypeptide. Two separate signal peptide
predictions are presented. The first was obtained by an alignment
with other fish Vtg signal peptides (Folmar et al., 1995) and is
denoted by shaded lettering. The second prediction was obtained
by the computer analysis method of von Heijne (1986) and is
denoted by asterisks. The nucleotide stretch corresponding to the
degenerate oligonucleotide MB6, used to screen the library, is
shown by double underlining and bold letters. Five predicted
antigenic determinants are depicted by shaded lettering with
average hydrophilicity values (Ah) indicated underneath. A
polyadenylation site (AATAAA) is located 53 nucleotides past the
stop codon and denoted by underlining.










24




1 ATG AAA GCG GTT GTG CTT GCC CTG ACT CTG GCC TTC GTG GCT GGA CAA AAT TTT
SK A V V L A L T LG Q N F

55 GCC CCT GAA TTT GCT GCT GGT AAG ACC TAC GTA TAT AAG TAT GAA GCG CTC ATC
A P E F A A G K T Y V Y K Y E A L I

109 CTG GGC GGT CTT CCT GAG GAA GGT TTG GCA AGA GCT GGA TTG AAA ATC AGC ACC
L G G L P E E G L A R A G L K I S T

163 AAA CTT CTA CTC AGT GCA GCT GAC CAA AAT ACT TAT ATG CTG AAG CTT GTG GAA
K L L L S A A D Q N T Y M L K L V E

217 CCT GAG CTC TCT GAG TAC AGC GGC ATT TGG CCA AAG GAC CCA GCA GTG CCA GCA
P E L S E Y S G I W P K D P A V P A

271 ACC AAG TTG ACA GCA GCC CTT CAC CTC AGC TCG CAA TTC CCA TCA AGT TTG AAT
T K L T A A L H L S S Q F P S S L N

325 ACA CCA ATG GTG TTT GTT GGT AAA GTC TTT GCT CCT GAG GAA GTC TCG ACT TTG
T P M V F V G K V F A P E E V S T L

379 GTG CTG AAC ATC TAC AGA GGC ATC CTG AAT ATT CTC CAG CTG AAC ATC AAG AAG
V L N I Y R G I L N I L Q L N I K K

433 ACC CAC AAA GTC TAT GAC TTG CAG GAG GTT GGA ACT CAG GGT GTG TGC AAG ACC
T H K V Y D L Q E V G T Q G V C K T

487 CTC TAT TCC ATC AGT GAA GAT GCA CGA ATT GAG AAC ATC CTT CTG ACC AAG ACC
L Y S I S E D A R I E N I L L T K T

541 AGG GAC CTG AGC AAC TGC CAG GAA AGA CTC AAT AAG GAC ATC GGG TTG GCA TAC
R D L S N C Q E R L N K 0 I G L A Y

595 ACT GAG AAA TGC GAC AAG TGC CAG GAG GAA ACT AAA AAC TTG AGA GGT ACC ACA
T E K C D K C Q E E T K N L R G T T

649 ACA TTA AGT TAC GTC TTG AAA CCA GTC GCC GAT GCC GTC ATG ATC CTG AAG GCG
T L S Y V L K ? V A D A V M I L K A

703 TAC GTT AAT GAG CTG ATC CAG TTT TCA CCT TTC TCT GAG GCT AAC GGA GCT GCC
Y V N E L I Q F S P F S E A N G A A

757 CAG ATG AGG ACC AAG CAG TCT TTG GAG TTC CTT GAA ATT GAG AAA GAA CCC ATT
Q M R T K Q S L E F L E I E K 2 P I

811 CCA TCT GTC AAG GCT GAA TAT CGT CAC CGT GGA TCT CTC AAA TAC GAG TTC TCC
P S V K A E Y R H R G S L K Y E F S

865 GAT GAA CTT CTT CAG ACA CCC CTT CAG CTG ATC AAG ATC AGT GAT GCA CCA GCC
D E L L Q T P L Q L I K I S D A P A

919 CAG GTT GCA GAG GTC CTG AAG CAC CTG GCT ACC TAC AAC ATT GAG GAT GTT CAT
Q V A E V L K H L A T Y N I E D V H

973 GAA AAT GCA CCT TTG AAG TTT TTG GAA CTG GTA CAA CTC CTC CGT ATT GCC CGC
E N A P L K F L E L V Q L L R I A R

1027 TAT GAA GAT TTG GAA ATG TAC TGG AAC CAG TAC AAA AAG ATG TCT CCC CAC AGA
Y E D L E M Y W N Q Y K K M S P H R










25


1081 CAC TGG TTC TTG GAC ACT ATT CCT GCC ACT GGT ACC TTC GCT GGT CTC AGA TTC
H W F L D T I P A T G T F A G L R F

1135 ATC AAA GAG AAG TTC ATG GCT GAG GAA ATA ACC ATC GCT GAG GCA GCT CAG GCT
I K E K F M A E E I T I A E A A Q A

1189 TTC ATT ACA GCT GTG CAC ATG GTG ACT GCT GAC CCT GAG GTT ATC AAG CTG TTT
F I T A V H M V T A D P E V I K L F

1243 GAG AGC CTG GTA GAC AGC GAC AAA GTA GTG GAA AAC CCA CTT CTG CGT GAG GTT
E S L V D S D K V V E N P L L R E V

1297 GTC TTC CTT GGA TAT GGA ACA ATG GTT AAC AAA TAC TGC AAT AAG ACA GTT GAT
V F L G Y G T M V N K Y C N K T V D

1351 TGT CCT GTT GAA CTC ATA AAG CCT ATT CAA CAA CGA CTG TCA GAC GCC ATT GCA
C P V E L I K P I Q Q R L S D A I A

1405 AAG AAC GAG GAA GAG AAC ATC ATC CTG TAC ATA AAG GTT TTG GGA AAT GCC GGC
MA 201)aim I I L Y I K V L G N A G
(Ah -2.07)
1459 CAT CCA TCT AGC TTC AAG TCA CTC ACT AAG ATC ATG CCC ATC CAT GGC ACT GCT
H P S S F K S L T K I M P I H G T A

1513 GCT GTA TCT CTG CCA ATG ACA ATC CAT GTT GAA GCC ATC ATG GCT CTG AGG AAC
A V S L P M T I H V E A I M A L R N

1567 ATT GCA AAG AAG GAG TCC AGA ATG GTC CAG GAA CTG GCT CTC CAG CTC TAC ATG
I A K K E S R M V Q E L A L Q L Y M

1621 GAC AAG GCT CTC CAC CCA GAG CTC CGT ATG CTG TCC TGC ATT GTT CTC TTC GAG
D K A L H P E L R M L S C I V L F E

1675 ACA AGT CCT TCT ATG GGT TTG GTG ACA ACT GTT GCC AAC TCT GTG AAA ACC GAG
T S P S M G L V T T V A N S V K T E

1729 GAG AAT TTG CAG GTG GCC AGC TTC ACT TAC TCT CAC ATG AAG TCC CTA AGC AGG
E N L Q V A S F T Y S H M K S L S R

1783 AGC CCC GCA ACC ATC CAT CCC GAT GTT GCT GCC GCA TGC AGC GCC GCC ATG AAG
S P A T I H P D V A A A C S A A M K

1837 ATC TTG GGT ACA AAG CTG GAC AGA CTG AGC CTG CGT TAT AGC AAA GCT GTA CAT
I L G T K L D R L S L R Y S K A V H

1891 GTG GAC CTC TAC AAC AGT TCC TTG GCG GTC GGT GCT GCT GCA ACT GCT TTT TAC
V D L Y N S S L A V G A A A T A F Y

1945 ATC AAC GAT GCT GCC ACC TTT ATG CCA AAA TCC TTT GTT GCA AAG ACC AAA GGC
I N D A A T F M P K S F V A K T K G

1999 TTC ATC GCT GGA AGT ACT GCT GAA GTC CTG GAG ATT GGA GCG AAT ATT GAA GGA
F I A G S T A E V L E I G A N I E G

2053 CTG CAG GAG CTG ATT CTG AAA AAC CCT GCT CTC TCT GAA ACT ACT GAC AGG ATC
L Q E L I L K N P A L S E S T D R I

2107 ACC AAA ATG AAG CGA GTC ATT AAG GCT CTG TCA GAA TGG AGA TCC TTG CCC ACC
T K M K R V I K A L S E W R S L P T

2161 AGC AAA CCC CTA GCC TCT GTC TAT GTT AAG TTC TTT GGA CAA GAG ATT GGC TTT
S K P L A S V Y V K F F G Q E I G F

Figure 2.2--continued












26



2215 GCT AAC ATT GAC AAA CCC ATG ATC GAT AAG GCT GTC AAG TTT GGC AAG GAA TTA
A N I D K P M I D K A V K F G K E L

2269 CCC ATT CAG GAA TAT GGA AGA GAG GCT CTC AAG GCT CTG CTC CTG TCT GGC ATC
P I Q E Y G R E A L K A L L L S G I

2323 AAC TTC CAC TAC GCT AAG CCA GTG CTG GCT GCT GAG ATG CGA CGC ATT CTT CCT
N F H Y A K P V L A A E M R R I L P

2377 ACC GTC GCT GGT ATT CCA ATG GAA CTC AGT CTG TAC AGT GCT GCT GTG GCT GCA
T V A G I P M E L S L Y S A A V A A

2431 GCC TCT GTT GAA ATC AAG CCC AAC ACG TCA CCA CGT CTG TCA GCG GAC TTC GAC
A S V E I K P N T S P R L S A D F D

2485 GTA AAG ACT CTG CTG GAG ACA GAC GTT GAG CTC AAG GCT GAG ATC AGA CCA ATG
V K T L L E T D V E L K A E I R P M

2539 GTT GCC ATG GAC ACA TAT GCC GTT ATG GGA CTT AAC ACC GAC ATC TTC CAG GCT
V A M D T Y A V M G L N T D I F Q A

2593 GCT TTG GTA GCT CGC GCT AAA CTG CAC TCT GTT GTG CCA GCC AAA ATA GCT GCA
A L V A R A K L H S V V P A K I A A

2647 AGA CTT AAT ATC AAA GAG GGT GAC TTT AAG CTT GAA GCT CTT CCT GTT GAT GTG
R L N I K E G D F K L E A L P V D V

2701 CCT GAA AAC ATC ACA TCC ATG AAT GTT ACA ACC TTT GCT GTA GCA AGA AAC ATC
P E N I T S M N V T T F A V A R N I

2755 GAG GAA CCT TTG GTT GAG AGA ATC ACT CCT CTT CTC CCC ACC AAA GTT TTG GTA
E E P L V E R I T P L L P T K V L V

2809 CCC ATC CCA ATC AGG AGA CAC ACA TCC AAG CTT GAT CCC ACT CGC AAT AGC ATG
P I P I R R H T S K L D P T R N S M

2863 TTA GAC TCC TCA GAA CTC CTT CCC ATG GAA GAA GAA GAT GTA GAG CCC ATT CCT
L D S S E L L P M I P I P
(Ah 2.25)
2917 GAA TAC AAG TTC CGT CGA TTT GCC AAA AAG TAC TGC GCT AAG CAC ATT GGT GTT
E Y K F R R F A K K Y C A K H I G V

2971 GGA CTG AAG GCC TGT TTC AAG TTT GCC AGT CAA AAT GGA GCC TCC ATC CAA GAC
G L K A C F K F A S Q N G A S I Q D

3025 ATT GTC CTG TAC AAA CTG GCT GGT AGC CAC AAC TTC TCT TTC TCT GTG ACA CCA
I V L Y K L A G S H N F S F S V T P

3079 ATT GAA GGA GAA GTT GTT GAG AGA TTG GAG ATG GAG GTT AAA GTC GGA GCA AAG
I E G E V V E R L E M E V K V G A K

3133 GCT GCA GAG AAG CTT GTT AAA CGC ATC AAC CTG AGT GAG GAC GAA GAA ACT GAA
A A E K L V K R I N L S JI
(Ah 2.43)
3187 GAA GGA GGT CCA GTC CTG GTG AAG CTC AAC AAA ATC CTG TCT TCA AGA CGG AAC
| G G P V L V K L N K I L S S R R N

3241 AGC TCC TCA TCT TCC TCC TCC AGC TCC AGC AGC TCT TCT GAG AGC CGT TCT TCA
S S S S S S S S S S S S S E S R S S

3295 AGG TCC TCC TCT TCC TCC TCC TCT TCA TCT CGC TCC AGC CGT AAG ATT GAC CTT
R S S S S S S S S S R S S R K I D L
Figure 2.2--continued













27







3349 GCA GCC AGG ACC AAT AGC AGC AGC AGC AGC AGT AGC CGT CGC AGC AGA AGC AGC
A A R T N S S S S S S S R R S R S S

3403 AGC AGC AGC AGC AGC AGC AGT AGC AGT AGC C A C AGC AGC A AGC AGC AGC AGC
S S S S S S S S S S S S S S S S S S

3457 AGG AGA AGC AGC AGC AGC AGC AGT AGT AGC AGC AGC AGC AGC AGT AGG AGC AGC
R R S S S S S S S S S S S S S R S S

3511 AGG AGA GTC AAC TCA ACA AGA TCC AGC AGC AGT TCA AGT AGG ACC AGC TCT GCA
R R V N S T R S S S S S S R T S S A

3565 TCA AGC CTT GCA TCT TTC TTC AGT GAC AGC TCA AGC TCT TCT AGC TCC AGT GAT
S S L A S F F S D S S S S S S S S i

3619 CGT CGC TCA AAG GAA GTG ATG GAG AAG TTC CAG AGG TTA CAC AAG AAA ATG GTC
R R S K EI V M E K F Q R L H K K M V
(Ah 2.55)
3673 GCC TCC GGT AGC AGT GCC TCA AGC GTT GAA GCC ATC TAC AAA GAG AAA AAA TAT
A S G S S A S S V E A I Y K E K K Y

3727 CTT GGC GAG GAA GAA GCC GTT GTG GCA GTG ATT CTC CGT GCT GTC AAA GCT GAC
L G E E E A V V A V I L R A V K A D

3781 AAG AGG ATG GTG GGA TAC CAG CTT GGT TTC TAC CTT GAC AAA CCA AAT GCC AGA
K R M V G Y Q L G F Y L D K P N A R

3835 GTT CAG ATC ATT GTC GCC AAC ATT TCT TCT GAT AGC AAC TGG AGG ATC TGT GCT
V Q I I V A N I S S D S N W R I C A

3889 GAT GCA GTT GTG TTG AGC AAG CAC AAA GTT ACA ACC AAG ATT TCC TGG GGA GAA
D A V V L S K H K V T T K I S W G E

3943 CAG TGC AGG AAA TAC AGC ACC AAT GTT ACA GGA GAG ACT GGT ATT GTT TCT TCA
Q C R K Y S T N V T G E T G I V S S

3997 AGC CCT GCC GCT CGC CTC AGA GTG TCC TGG GAA AGA CTG CCT TCT ACC CTG AAA
S P A A R L R V S W E R L P S T L K

4051 CGC TAT GGA AAG ATG GTT AAC AAG TAC GTT CCT GTT AAA ATA TTG TCT GAC TTG
R Y G K M V N K Y V P V K I L S D L

4105 ATC CAC ACA AAG AGA GAA AAC AGC ACC AGG AAT ATC TCA GTC ATT GCA GTT GCC
I H T K R E N S T R N I S V I A V A

4159 ACA TCT GAA AAG ACA ATT GAC ATC ATA ACC AAA ACT CCA ATG AGC TCT GTC TAC
T S E K T I D I I T K T P M S S V Y

4213 AAT GTC ACT ATG CAT CTT CCC ATG TGT ATT CCC ATT GAT GAG ATC AAA GGT CTC
N V T M H L P M C I P I D E I K G L

4267 AGC CCC TTT GAT GAA GTC ATT GAC AAG ATC CAC TTC ATG GTT TCT AAG GCT GCT
S P F D E V I D K I H F M V S K A A

4321 GCA GCT GAA TGC AGC TTC GTC GAA GAC ACA CTC TAC ACA TTC AAC AAC AGG AGC
A A E C S F V E D T L Y T F N N R S

4375 TAC AAG AAC AAG ATG CCT TCC TCT TGC TAC CAG GTT GCA GCA CAG GAC TGC ACA
Y K N K M P S S C Y Q V A A Q D C T

4429 GAT GAG CTG AAA TTC ATG GTT CTC CTG AGG AAG GAT TCG TCC GAA CAA CAC CAC
D E L K F M V L L R K D Si S ii Q H H
(Ah = 2.1)



Figure 2.2--continued












28


4483 ATC AAT GTC AAG ATT TCT GAG ATC GAT ATT GAC ATG TTT CCA AAG GAC GAC AAC
I N V K I S E I D I D M F P K D D N

4537 GTC ACT GTG AAG GTC AAC GAA ATG GAA ATA CCC CCA CCA GCC TGC CTT ACC GCC
V T V K V N E M E I P P P A C L T A

4591 ACC CAA CAG CTT CCA TTG AAG ATC AAG ACA AAG CGG AGA GGA CTT GCT GTC TAT
T Q Q L P L K I K T K R R G L A V Y

4645 GCA CCC AGC CAC GGT CTC CAA GAA GTC TAC TTT GAC AGG AAG ACA TGG AGG ATC
A P S H G L Q E V Y F D R K T W R I

4699 AAA GTT GCT GAC TGG ATG AAA GGA AAG ACC TGT GGA CTC TGT GGA AAG GCT GAT
K V A D W M K G K T C G L C G K A D

4753 GGA GAG ATC AGA CAG GAG TAC CAC ACT CCC AAC GGA CGC GTG GCC AAG AAC TCG
G E I R Q E Y H T P N G R V A K N S

4807 ATC AGC TTT GCT CAC TCC TGG ATT CTT CCT GCT GAA AGC TGC AGG GAT GCA TCT
I S F A H S W I L P A E S C R D A S

4861 GAG TGC CGT CTG AAA CTT GAA TCT GT CG CAG CTG GAG AAA CAG TTG ACC ATC CAC
E C R L K L E S V Q L E K Q L T I H

4915 GGT GAG GAC TCC ACA TGC TTC TCA GTT GAG CCT GTA CCT CGT TGT CTG CCC GGT
G E D S T C F S V E P V P R C L P G

4969 TGC TTG CC" GTC AAG ACC ACA CCT GTC ACT GTT GGT TTC AGC TGC CTG GCA TCT
C L P V K T T P V T V G F S C L A S

5023 GAT CCT CAG ACC AGT GTC TAT GAC AG AGT GTG GAT CTA AGA CAA ACT ACC CAG
D P Q T S V Y D R S V D L R Q T T Q

5077 GCT CAC CTG GCT TGC AGC TGC AAC ACC AAG TGC TCT TAA ACA TAA GAT TTC CTT
A H L A C S C N T K C S -

5131 GAA GTC ACT ACT ATG TGT AAG TTT TAT CTG TAA CAA TAA ATA AAC TGC ATC TGA

5185 AAA TAA AAA AAA AA

























Figure 2.2--continued









29

F. heteroclitus Vtg Sequence

A conceptual translation of the 5112 bp open reading frame resulted in a 1704-

amino acid protein sequence (Fig. 2.2). A signal peptide was predicted (underlined) by

aligning the F. heteroclitus sequence with the N-terminal sequences of several other

piscine Vtgs (Folmar et al. in press). This prediction can be compared to that resulting

from the method of von Heijne (1986), represented in Figure 2.2 by asterisks. We made

several attempts to determine the signal peptide sequence through N-terminal sequencing

of Vtg isolated from the blood of estrogen-treated male F. heteroclitus, all of which

resulted in inconclusive residue readings, suggesting that the secreted Fundulus Vtg is

N-terminally blocked. Five internal peptide sequences predicted to offer high antigenicity

by the method of Hopp and Woods (1981) are represented by shaded lettering in Figure

2.2. The end of the cDNA sequence was revealed by a poly-adenylation site

(AATAAA), beginning at bp 5165 and denoted by underlining.

A scan of the sequence for post-translational modification sites of the putative

protein revealed 16 potential N-glycosylation sites, 13 potential N-myristoylation sites,

and potential phosphorylation sites for the following kinases: 7 for CAMP- and CGMP-

dependent protein kinase; 39 for protein kinase C; 23 for casein kinase II; and finally,

a single site for tyrosine kinase (Fig. 2.3). We have highlighted the polyserine domain

in Figure 2.3 with asterisks. The asterisks signify that, in addition to the predicted

phosphorylation sites for the above mentioned kinases, past studies in F. heteroclitus

(Wallace and Begovac, 1985) and in other non-mammalian vertebrates (Mecham and

Olcott, 1949, Mano and Lipmann, 1966, Wiley and Wallace, 1981; Byrne et al., 1984)








30













Fundulus heteroclitus Vitellogenin



PREDICTED...
SN-glycosylation site
SN-myristoylation site
Sphosphorylation site













Figure 2.3 A schematic representation of potential sites for posttranslational
modifications of the putative F. heteroclitus Vtg protein as predicted by
the Prosite program (Bairoch et al., 1995). Phosphorylation sites
represent potential targets for the following kinases: c-AMP- and g-AMP-
dependent kinase, protein kinase C, casein kinase II, and tyrosine kinase.
The region denoted by asterisks represents the polyserine domain. Past
studies suggest that in addition to the sites displayed by the above-
mentioned kinases, every serine residue in this region undergoes
phosphorylation by an as-yet unidentified "vitellogenin kinase."








31

suggest that almost all serine residues within the phosvitin region are phosphorylated by

an as yet uncharacterized "vitellogenin kinase" activity.


Protein Alignments


Alignment of the F. heteroclitus Vtg sequence with other selected vertebrate Vtgs

is shown in Figure 2.4. Partial Vtg cDNA translations published from three other fish

species are included. Pairwise comparisons of these vertebrate Vtg sequences against the

F. heteroclitus sequence result in similar degrees of identity: Gallus, 38%; Xenopus,

39%; Acipenser, 38%; and Ichthyomyzon, 37%. Against the two smaller teleost

sequences, the F. heteroclitus sequence shares 50% identity with rainbow trout,

Oncorhynchus but only 30% with Oreochromis. These last two values should be

considered only preliminary until more sequence information becomes available.

Attempting to find an obvious difference between the F. heteroclitus Vtg and that of the

other vertebrates, we compared several types of predicted structural analysis scales

including those by the methods of Hopp and Woods (1981), Kyte and Doolittle (1982),

and Janin (1979). There were no striking differences revealed by these methods that

might account for the greater solubility of the F. heteroclitus yolk proteins (data not

shown).

The phylogram in Figure 2.5 was created using the program PAUP (Swofford,

1993) from an alignment (not shown) containing the first five vertebrate Vtgs listed in

Figure 2.4, plus three invertebrate Vtgs from boll weevil, Anthonomus grandis,

mosquito, Aedes aegypti, and finally Vtg 5 from C. elegans, defined as an outgroup. In











Figure 2.4 Alignment of the putative F. heteroclitus Vtg sequence (gi:459202)
with other vertebrate Vtgs: the chicken Gallus domesticus Vtg II
(van het Schip et al., 1987); Xenopus laevis Vtg A2 (Gerber-Huber
et al., 1987); the white sturgeon Acipenser transmontanus Vtg
(Bidwell and Carlson, 1995); the silver lamprey Ichthyomyzon
unicuspis Vtg (Sharrock et al., 1992); and the C-termini from the
rainbow trout Oncorhynchus mykiss Vtg (LeGuellec et al., 1988)
and the tilapia Oreochromis aureus (Ding et al., 1990) Vtg as
constructed by ClustalV (Higgins et al., 1992) and modified by
eye. Our defined polyserine domain, which includes putative Pv
regions, is labeled and underscored with a triple dashed line.
Residues identical in at least four of the aligned sequences are
denoted by shaded lettering. Sequence gaps are represented as
dashes.










33






Fundulus MKAVVL-ALTLAFVAGQ--NFAPEFAAGKTYVYKYEAL ILGGLPEEGLARAGLKISTKLL 57
Gallus MRGIIL-ALVLTLVGSQKFDIDPGFNSRRSYLYNYEGSMLNGLQDRSLGKAGVRLSSKLE 59
Xenopus MKGIVL-ALLLALAGSERTHIEPVFSESKISVYNYEAVILNGFPESGLSRAGIKINCKVE 59
Acipenser -------- -TIALVGSQQTKYEPSFSGSKTYQYKYEGVILTGLPEKGLARAGLKVHCKVE 52
Ichthyomyzon MWKLLLVALAFAADAQ------- FQPGKVRYSYDAFSISGLPEPGVNRAGLSGEMKIE 53

Fundulus LSAADQNTYMLKLVEPELSEYSGIWPKDPAVPATKLTAALHLSSQFPSSLNTPMVFVGKV 117
Gallus ISGLPENAYLLKVRSPQVEEYNGVWPRDPFTRSSKITQVISSCFTRLFKFEYSSGRIGNI 119
Xenopus ISAYAQRSYFLKIQSPEIKEYNGVWPKDPFTRSSKLTQALAEQLTKPARFEYSNGRVGDI 119
Acipenser ISEVAQKTYLLKILNPEIQEYINGIWPKAPFYPASKLTQALASQLTQPIKFQYRNGQVGDI 112
Ichthyonyzon IHGHTHNQATLKITQVNLKYFLGPWPSDSFYPLTAGYDHFIQQLEVPVRPDYSAGRIGDI 113

Fundulus FAPEEVSTLVLNIYRGILN I LQLNIKKTHKVYDLQEVGTQGVCKTLYSISEDAR I ENILL 177
Gallus YAPEDCPDLCVNIVRGILNMFQMTIKKSQNVYELQEAGIGGICHARYVIQEDRKNSRIYV 179
Xenopus FVADDVSDTVANIYRGILNLLQVTIKKSQDVYDLQESSVGGICHTRYVIQEDKRGDQIRI 179
Acipenser FASEDVSDTVLNIQRGILNMLQLTIKTTQNVYGLQENGIAGICEASYVIQEDRKANKXIV 172
Ichthyomyzon YAPPQVTDTAVNIIVRGILNLFQLSLKKNQQTFELQETGVEGICQTTYVVQEGYRTNEMAV 173

Fundulus TKTRDLSNCQERLNKDIGLAYTEKCDKCQEETKNLRGTTTLSYVLKPVADAVMILKAYVN 237
Gallus TRTVDLNNCQEKVQKSIGMAYIYPCPVDVMKERLTKGTTAFSYKLKQSDSGTLITDVSSR 239
Xenopus IKSTDFNNCQDKVSKTIGLELAEFCHSCKQLNRVIQGAATYTYKLKGRDQGTVIMEVTAR 239
Acipenser TKSKDLNNCNEKIKMDIGMAYSHTCSNCRKIRKNTRGTAAYTYILKPTDTGTLITQATSQ 232
Ichthyomwon VKTKD1NCDHKVYKTMGTAYAERCPTCQKMNKNLRSTAVYNYAIFDEPSGYI IKSAHSE 233

Fundulus ELIQFSPFSEANGAAQMRTKQSLEFLEIEKEPIPSVKAEYRHRGSLKYEFSDELLQTPLQ 297
Gallus QVYQI SPFNEPTGVAVMEARQQTLVEVRSERGSAPDVPMQNYGSLRYRFPAVLPQMPLQ 299
Xenopus QVLQVTPFAERHGAATMESRQVLAWVGSKSGQLTPPQIQLKNRGNLHYQFASELHQMPIH 299
Acipenser EVHQLTPFNEMTGAAITEARQKLVLEDAKVIHVTVPEQELKNRGSIQYQFASEILQTPIQ 292
Ichthyomyzon EIQQLSVEDIKEGNVVIESRQKLILEGIQSAPAASQAASLQNRGGLMYKFPSSAITKMSS 293

Fundulus LI--KISDAPAQVAEVLKHLATYNIEDVHENAPLKFLELVQLLRIARYEDLEMYWNQYKK 355
Gallus LI -KTKNPEQRIVETLQHIVLNNQQDFHDDVSYRFLEVVQLCRIANADNLESIWRQVSD 357
Xenopus LM- -KTKSPEAQAVEVLQHLVQDTQQHIREDAPAKFLQLVQLLRASNFENLQALWKQFAQ 357
Acipenser LF- -KTRSPETKIKEVLQHLVQNNQQQVQSDAPSKFLQLTQLLRACTHENIEGIWRQYEK 350
Ichhvomnyzon LFVTKGKNLESE IHTVLKHLVENNQLSVHEDAPAKFLRLTAFLRNVDAGVLQSIWHKLHQ 353

Fundulus MSPHRHWFLDTIPATGTFAGLRFIKEKFMAEEITIAEAAQAFITAVHMVTADPEVIKLFE 415
Gallus KPRYRRWLLSAVSASGTTETLKFLKNRIRNDDLNYIQTLLTVSLTLHLLQADEHTLPIAA 417
Xenopus RTQYRRCLLDALPMAGTVDCLKFIKQLIHNEELTTQEAAVLITFAMRSARPGQRNFQISA 417
Acipenser TQLYRRWILDALPAAATPTAFRPITQRIMKRDLTDAEAIQTLVTAMHLVQTNHQIVQMAA 410
Ichthyomyzon QKDYRRWILDAVPAMATSEALLFLKRTLASEQLTSAEATQIVYSTLSNQQATRESLSYAR 413

Fundulus SLVDSDKVVENPLLREVVFLGYGTMVNKYCNKTVDCPVELI KIQQRLSDAIAKNEEENI 475
Gallus DLMTSSRIQKNPVLQQVACLGYSSVVNRYCSQTSACPKEALQPIHDLADEAISRGREDKM 477
Xenopus DLVQDSKVQKYSTVHKAAILAYGTMVRRYCDQLSSCPEHALEPLHELAAEAANKGHYEDI 477
Acipenser ELVFDRANLKCPVLRKHAVLAYGSMVNRYCAETLNCREEALKPLHDFANDAISRAHEEET 470
Ichthyomvzon ELLHTSFIRNRPILRKTAVLGYGSLVFRYCANTVSCPDELLQPLHDLLSQSSDRADEEEI 473











34





Fundulus ILYIKVLGNAGHPSSFKSLTKIMPIHGTAAVSLPMTIHVEAIMALRNIAKKESRMVQELA 535
Gallus KLALKCIGNMGEPASLKRILKFLPISSSSAADIPVHIQIDAITALKKIAWKDPKTVQGYL 537
Xenopus ALALKALGNAGQESIKRIQKFLPGFSSSADQLPVRIQTDVMALRNIAEDRKQEIL 537
Acipenser VLALKALGGQSSIKRIQKCLPGFSSGASQLPVKIQViDAV ELT 530
Ichrhyomyzon VLALKALGNAGQPNSIKKIQRFLPGQGKSLDEYSTRVQAEAIM NAKRDPRKVQEIV 533

Fundulus LQLYMDKALHPELRMLSCIVLFEBSPSMGLVTTVANSVKTEE- -NQVAS TYSHMKSLS 593
Gallus IQILADQSLPPBVRMMACAVIFETRPALALITTIANVAMKESKTNMQVASFVYSHMKSLS 597
Xenopus LQIFMDRDVRTEVRMMACLALETRPGLATVTAIANVAARESKTNLQLASTFSQMKAL 597
Acipenser MQLFMDHQLHSEVRMVASMVLLETRPSMALVATLAEALLK- -TSQVASFSTYHMKAIT 588
Ichthyomyzon LPIFLNVAIKSELRIRSCIVFFESKPSVALVSMVAVRLRRP NLQVASFVYSQMRSLS 591

Fundulus RSPATIHPDVAAACSAAMKILGTKLDRLSLRYSKAVHVDLYNSSLAVGAAATAPYINDAA 653
Gallus KSRLPFMYNISSACNIALKLLSPKLDSMSYRYSKVIRADTYFDNYRVGATGE I VNSPR 657
Xenopus KSSVPHLE PLAAACVALKILNPSLDNLGYRYSKVMRVDTFKYNLMAGAAAKVIMNSAN 657
Acipenser RSTAPENHALSSACNVAVKLLSRKDRLSYRYSKAMHMDTFKYPLMA AANIHIINNAA 648
Ichihyomyzon RSSNPEFRDVAAACSVAIKMLGSKLRLGCRYSKAVHVTFNARTMAGVSADYRINSPS 651

Fundulus TFMPKSFVAKTKGFIAGSTAEVLEIGANIEGLQELILKNPALSESTDR------------ 701
Gallus TMFPSAIISKLMANSAGSVADLVEVGIRVEGADVIMKRNIPFAEYPT ------------ 705
Xenopus TMFPVFILAKFREYTSLVENDDIBIGIRGEGIEEFLRKQNIQFANFPM------------ 705
Acipenser SILPSAVVMKFQAYILSATADPLEIGLHTEGLQEVLMQNHEHIDQMPS- ----------- 696
Ichthyomyzon GPLPRAVAAKIRGQGMGYASDIVEFGLRAEGLQELLYRGSQEQDAYGTALDRQTLLRSGQ 711

Fundulus - - ITKIKRVIKALSEWRSLPTSKPLASVYVKFFGQEIGFANIDKPMIDKAVKFGKELP 757
Gallus -- YKQIKELGIKAIQGWKELPTETPLVSAYLKILGQEVAFININKELLQQVMKTVVEPA 761
Xenopus ---RKKISQIVKSLLGFKGLPSQVPLISGYIKLFGQEIAFTELNKEVIQNTIQALNQPA 761
Acipenser ----AGKIQQIMMLSGWKSVPSEKTLASAYIKLFGQEISPSRLDKKTIQEALQAVREPV 752
Ichlhyomyzon ARSHVSS IHDTLRKLSDWKSVPEERPLASGYVKVHGQEVVFAELDKKMMQRISQLWHSAR 771

Fundulus IQEYG----REALKALLLSGINFHYAKPVLAAEMRRILP TVAGIPMELSLYSAAVAAASV 813
Gallus DRNAA- --- IKRIANQILNSIAGQWTQPVWMGELRYVVPSCLGLPLEYGSYTTALARAAV 817
Xenopus ERHTM ---- IRNVLNKLLNGVVGQYARRWMTWEYRH II PTTVGLPAELSYQS IVHAAV 817
Acipenser ERQTV - -IKRVVNQLERGAAAQLSKPLLVAEVRRI LPTCIGLPMEMSLYVSAVTTADI 808
Ichthyomzon SHHAAAQEQIRAVVSKLEQGMDVLLTKGYVVSEVRYMQPVCI I MDLNLLVSGVTTNRA 831

Fundulus E I KPNTS PRLSADFDVKTLLETDVELKAE I RPMVAMDTYAVMLNTD I FQAALVARAKLH 873
Gallus SVEGKMTPPLTGDFRLSQLLESTMQIRSDLKPSLYVHTVAfTMGVNTEYFQHAVEIQGEVQ 877
Xenopus NSDVKVKPTPSGBDSAAQLLESQIQLNGEVKPSVLVHVATMGINSPLFQAGIEFHGKVH 877
Acipenser NVQAHITPSPTNDFNVAQLLNSNIVLHTDVTS IAMIHTIAVMGINTHVIQTGVELHVKAR 868
Ichthyomyzon NLHASFSQSLPADMKLADLLATNIELRVAATTSMSQHAVAIMGLTTDLAKAGMQTHYKTS 891

Fundulus SVVPAKIAARLNIKEGDFKLEALPVDVPENITSMNVTTFAVARNIEEPLVERITPLLPTK 933
Gallus TRMPMKFDAKIDVKLKNLKIETNPCREETEIVVGRHKAFAVSRNIGELGVEKRTS ILED 937
Xenopus AHLPAKFTAFLDMKDRNFKIETPPFQQENHLVEIRAQTFAFTRNIADLDSARKTLVVPRN 937
Acipenser TTVPMKFTAKIDLKEKNFKIESEPCQQETEVLSLSAQAFAISRNVEDLDAAKKNPLLPEE 928
Ichlhyomyzon AGLGVNGKIEMNARESNFKASLKPFQQKTVVVLSTMESIVFVR-- -DPSGSRILPVLPPK 948

Fundulus VLVP- I-------PIRRHTSKLDPTRNSMLDSSEL--LPMEEEDVEPIPEYKF -RRFA 980
Gallus APLD- -VTEEPFQTSERASREH -----FAMQGPDS- -MPRKQSHSSREDLRRSTGKRAHK 988
Xenopus NEQN--ILKKHFETTGRTSAE---GASMMEDSSEM--GPKKYSAEPGHHQYAPN--- INS 987
Acipenser AVRN- ILNEQFNSGTEDSNERERAGKFARPSAEM- -MSQELMNSGEHQNRKGA -HAT 981
Ichthyomyzon MTLDKGLISQQQQQPHHQQQPHQHGQDQARAAYQRPWASHEFSPAEQKQIHDIMTARPVM 1008







Figure 2.4--continued










35






Fundulus KK- -YCAKHIGVGLKACFKFASGAQGSIQDIVLYKLAGSHNFSFSVTPIEGE- -VVERLE 1036
Gallus RD- ICLKMHHIGCQLCFSRRSRDASFIQNTYLHKLIGEHEAKIVLMPVHT-DADIDKIQ 1045
Xenopus YD- -ACTKFSKAQVHLCIQCKTHNAASRRNTIFYQAVGEHDFKLTMKPAHT-EGAIEKLQ 1044
Acipenser RS--A AKAKNFGFEVCFEGKSENVAFLRDSPLYKIIGQHHCKIALKPSHSSEATTEKIQ 1039
Ichthyomyzon RRKQHCSKSAALSSKVCFSARLRNAAFIRNALLYKITGDYVSKVYVQPT-SSKAQIQKVE 1067

Fundulus MEVKVGAKAAEKLVKRINLSEDEETEEG--GPVLVKLNKI -------------------- 1075
Gallus LEIQAGSRAAARIITEVNPESEEEDESSPYEDIQAKLKRItLGIDSMFKVANKTRHPKNRP 1105
Xenopus LEITAGPKAASKIMGLVEVEGTEGEPMDE-TAVTKRLKMILGIDESRKDTNETALYRSKQ 1103
Acipenser LELQTGNKAASKIIRVVAMQSLAEADEMK-GNILKKLNKLLTVDGE ------------- 1084
Ichthyomyzon LELQAGPQAAEKVIRMVELVAKASKKSKKNSTITEEGVGETIISQLKKILSSDKDK - 1123
>smaPOLYSERINE DOMAIN=-

Fundulus ------------------------------------------------------------ 1075
Gallus SKKGNTVLAEFGTEPDAKTSSSSSSASSTATSSASSSASSPNRKKPMDEEENDQVKQARN 1165
Xenopus KKKNKI------------------------------------- HNRRLDAE----VVEARK 1123
Acipenser -----------------------------------------------------------T 1085
Ichthyomyzon -------------- DAKKPPGSSSSSSSSSSSSSSSSSSDKSGKKTPRQGSTVNLAAKR 1168
POLYSERINE DOMAIN==============

Fundulus --------SSRRNSSSSSSSSSSSSSESRSSRSSSSSSSSSRSSRKIDLAARTNSSSSSS 1127
Gallus KDASSSSRSSKSSNSSKRRSSKSSNSSKRSSSSSSSSSSSSSSSSSSSSS- --- SSNSK 1220
Xenopus QQSSLSSSSSSSS SSSSSSSSSSSPSSSSSSSYSKRSKRREHNPHHQRESSS-S 1182
Acipenser QDSTLRGFKRRSSSSSSSSSSSSSSSSSSSSSSQQSRMEKRMEQDKLTENLERDRDHMR 1145
Ichthyomnyzon ASKKQRGKDSSSSSSSSSSSSDSSKSPHK--HGGAKRQHAGHGAPHLGPQSHSSSSSSSS 1226
E~ss=ssaassaasPOLYSERINE DOMAINamm----i-msma =s------

Fundulus SRRSRSS- --------------SSSSSSSSSSSSSSSSSSRRSSS SSSSSSSSSRSSRR 1172
Gallus SSSSSKSSSSSSRSRSSSKSSSSSSSSSSSSSSKSSSRSSSSSKSSSHHSHSHHSGH 1280
Xenopus SSQEQNKKRNLQENRKHGQKGMSSSSSSSS SSSSSSSS$SSSSSEENRPHKNRQ 1242
Acipenser GKQSKNKKQEWKNKQKKHHKQLPSSSSSSSSSSGSNSSSSSSSSSSSSS --RSHNHRN 1202
Ichthyomyzon SSSSSSASKSFSTVKPPMTRKPRPARSSSSSSSSDSSSSSSSSSSSSSSSSSSS ----- 1281
--=====am=======-=-mmPOLYSERINE DOMAIN-=aaa--mm-=-a.------

Fundulus VNSTRSSSSSSRTSSASSLASFFSDSSSSSSSSD--------RRSKEVME-KFQRLHK-K 1222
Gallus LNGSSSSSSSSRSVSHHSHEHHSGHLEDDSSSSSSSSVLSKIWGRHEIYQYRFRSAHR-Q 1339
Xenopus --- HDNKQAKMQSNQHQQKKNKFSESSSSSSSSSSSEMWNKKKHHRNFYDLNFRRTAR-T 1298
Acipenser -- -NTRTLSK---------SKRYQNNNNSSSSSGSSSSSEEIQKNPEIFAYRFRSHRD-K 1249
Ichthyomyzon --------------------------- SSSESKSLEWLAVKDVNQSAFYNFKYVPQRKPQ 1314
ass^5Em =E;sesrsEPOLYSERINE DOMAINms=s-=-grs-r~ s s

Fundulus MV---------- -ASGSSASSVEAIYKEK-- ----KYLGEEEA-VVAVILRAVKADKRMV 1264
Gallus --- EFPKRKLPGDRATSRYSSTRSSHDTSRAASWPKFLGDIKTVLAAFLHGISNNKKTG 1396
Xenopus KGTEHRGSRLSSSSESSSSSSESAY ---RHKA---KFLGDKEPVLVVTFKAVRNDNTKQ 1352
Acipenser LGFQNKRGRMSSSSSSSSSSSSQSTLNSKQDA- ---KFLGDSSPPIFAFVAR-kRSDGLQQ 1306
Ichthyomyzon TSRRHTPASSSSSSSSSSSSSSSSSSSDSDMTVSAESFEKHSKPKVVIVLRAVRADGKQQ 1374
am==me===~=iPOLYSERINE DOMAIN.s==.===== =mm<

Fundulus GYQLGFYLD---KPNARVQIIVANISSDSNWRIAAVVLSKHKVTTKISWGEQCRKYST 1321
Gallus GLQLWYAD -TDSVRPRVQVFVTNLTDSSKWKLCADASVRNAPQAVAYVKWGWDCRDYKV 1455
Xenopus GYQMVVYQE-YHSSKQQIQAYVMDI -SKTRWAACFDAVVVNPHEAQASLKWGQNCQDKI 1410
Acipenser GYQVAAYTD NRVSRPRVQLLATEIIEKSRWQICADAILASNYKAMALMRWGEECQDYKV 1365
Ichthyomyzon GLQTTLYYGLTSNGLPKAKIVAVELSDLSVWKLCAKFRLSAHMKAKAAIGWGKNCQQYRA 1434
Oncorhynchus LGRPKTTSDEPNIITAALDENDNWKLCADGVLLSKHKVNAKIAWGAGCKDYNT 53





Figure 2.4--continued










36



Fundulus NVTGETGIVSSSPAARLRVSWERLPSTLKRYGK-MVNKYVP-VKILSDLIHTKRENSTRN 1379
Gallus STELVTGRFAGHPAAQVKLEWPKVPSNVRSVVE-WFYEFVPGAAFMLGFSERMDKNPSRQ 1514
Xenopus NMKAETGNFGNQPALRVTANWPK I PSKWKSTGK VVGEYVPGAMYMMGFQGEYKRNSQRQ 1469
Acipenser AVSAVTGRLASHPSLQIKAKWSRIPRAAKQTQN- ILAEYVPGAAFMLGFSQKEQRNPSKQ 1424
Ichthvomyzon MLEASTGNLQSHPAARVDIKWGRLPSSLQRAKNALLENKAPVIASKLEMEIMPKKNQKHQ 1494
Oncorhyncus FITAETGLVGPSPAVRLLDKLPKVPKAVWRYVRIVSEFIPGHIPYYtADLVPMQKDKNSE 113

Fundulus ISVIAVATSEKTIDIITKTPMSSVYNVTMHLPMCIBIDE--I---KGLSP--FDEVIDKI 1432
Gallus ARMVVALTSPRTCDVVVKLPDIILYQKAVRLPLSLVGP --RIPASELQPPIW-NVFAEA 1571
Xenopus VKLVFALSSPRTCDVVIR IRLTVYYRALRLPVPIRVGH- -HAKENVLQTPTW-NIFAEA 1526
Acipenser FKIILAVTSPNTIDTLIKAPKITLFKQAVQIPVQIPMEP--SDAER--RSPGLASIMNEI 1480
Ichihyomnyon VSVILAAMTPRRMNIIVKLPKVTYFQQGILLPFTFPSPRFWDRPEGSQSDSLPAQIASAF 1554
Oncorhynchus KQFTVATSERTLDVILKTPKMTLTKTGVNLPCSLPFESMTDLSPFDDNIVNKIHYL -F 171

Fundulus HFMVSKAAAECSFVEDTLYTFrNRSYKNKMPSSCYQVAAQDCTDELFMVLLRK --DSS 1490
Callus PSAVLENLKARCSVSYNKI KTFNEVKFNYSMPANCYHILVQDCSSSLKLVMMKSAGEAT 1631
Xenopus PKLIMDSIQGECKVAQDQITTFNGVDLASALPENCYNVAQDCSPEMKF4VLMRNSKESP 1586
Acipenser PFLIEEATKSKVAQENKFITFDGVKFSYQMPGGCYHILAQRSKVRF LKQASMSK 1540
Ichthyomyzon SGIVQDPVASACELNEQSLTTFNGAFFNYDMPESCYHLAQECSSRPPFIVLIKLDSERR 1614
Oncorhynchus S---- EVNAVKCSMVRDTLTTNNKKYKINMPLSCYQVLAQDCTT LKYMVSAEEGSVHL 227

Fundulus EQHHINVKISEIDIDMF-PKDDNVTVKVNEMEIPPPA-CLTATQQLPLKIKTKRRGLAVY 1548
Gallus NLKAINIKIGSHEIDM-HPVNGQVKLLVDGAESPTANISLIS-AGASLWIHNENQGFAIA 1689
Xenopus NHKDINVKLGEYDIDMYYSA-DAFKMKINNLEVSEEHLPYKSFNYPTVEIKKKGNGVSLS 1645
Acipenser NLRAVNAKIYNKDIDILPTTKGSVRLLINNNEIPLSQLPFTD-SSGNIHIKRADEGVSVS 1599
Ichthyomyzon I--SLELQLDDKKVKIVSRND---- IRVDGEKV#LRRLSQKN - -QYGFLVLDAGVHLL 1664
Oncorhynchus NKTTSNVKISDIDVDLYTQDHGVIVKVNEMEVSNEQLPYKDPSG-SIKIDRKKGEGVSLY 286

Fundulus APSHGLQEVYFDRKTWRIKVADWMKGKTCGLCGKADG IRQEYHTPNGRVAKNS ISFHS 1608
Gallus APGHGIDKLYFDGKTITIQVPLWMAGKTCGICGKYDAECEQEYRMPNGYLAKNAVSFGHS 1749
Xenopus ASEYGIDSLDYDGLTFKFRPTIWMKGKTCGICGHNDDESEKELQMPDGSVAKDQMRFIHS 1705
Acipenser AQQYGLESLYFDGKTVQVKVTSEMRGKTCLCGHNDGERRKSFRPDRQARGP------ 1653
Ichthyoinyzon LKYKDL-RVSFNSSSVQVWVPSSLKGQTCGLCGRNDDELVTEMRMPNLEVKDFTSFAHS 1723
Oncorhynchus APSHGLQKVYFDKYSWKIKVVDWMKGQTCGQLCKADGENRQgYRTPSGRLTiSSVSFAHS 346
Oreochromis FFFSLVFHAVS 11

Fundulus WILPAESCRDASECRLKLESVQLEKQLTIHGEDSTCFSVEPVPRCLPGCLPVKTTPVTVG 1668
Gallus WILEEAPCRGA- -CKLHRSFVKLEKTVQLAGVDSKCYSTEPVLRCAKGCSATKTTPVTVG 1807
Xenopus WILPAESCSEG- -CNLKHTLVKLEKAIATDGAKAKCYSVQPVLRCAKGCSPVKTVEVSTG 1763
Acipenser -----------------------------------------------------SVSPTPG 1660
Ichthyomyzon WIAPDETCGGACALSRQ- -TVHKESTSVISGSRENCYSTEPIMRCPATCSASRSVPVSVA 1781
Oncorhynchus WVIPSDRC-DASEC-LM- -KLEKQVIVDD-RESK-CYSVEPVLRCLPGCPVRTTPITIG 400
Oreochromis KKLQNHYSLRLLKEKVKS ---- ELMVPI LKVSEPNATLLSPCCSACPACIPVRTTTVNVG 67

Fundulus -FSCLASDPQ-------TSVYD-RSVDLRQTTQAHLACSCNTK-CS- 1704
Gallus -FHCLPADSANSLTDKQ-MKYDQKSEDMQDTVDAHTTCSCENEECST 1852
Xenopus -FHCLPSDVSLDLPEGQ IRLE-KSEDFSEKVEAHTACSCETSPCAA 1807
Acipenser --LCLEKTATEAASFCVIM 1677
Ichihyomyzon -MHCLPAESEAISLAMSEGRPFSLSGKSEDLVTEMEAHVSCVA 1823
Oncorhynchus -HCLPFDSNLNRSEGLSSIY-EKSVDLMEKAEAHVACRCSEQ-CM 442
Oreochromis FYGCLPSDTT-VDRSGLSSFF-EKSIDLRDTAEAHLACRCTPQ-CA 110











Figure 2.4--continued









37

the single best tree F. heteroclitus Vtg was placed on an independent branch,

intermediate to the positions of sturgeon and lamprey Vtgs. The Ichthyomyzon sequence

was the vertebrate Vtg determined to lie furthest from the reference sequence, thereby

placing it nearest to the outgroup. One of the more significant relationships provided by

the tree is indicated by the bootstrapping values at the Acipenser branch (in parentheses,

Fig. 2.5): through 100 bootstrap replicates, sturgeon Vtg was partitioned with the two

tretrapod Vtgs 95 % of the time, substantially more than the Vtgs of either F. heteroclitus

(67%) or Ichthyomyzon (31% not shown).


Polyserine Domain


We have designated a polyserine domain from each of the aligned Vtgs

(underscored with a triple dotted line in Fig. 2.4; see Materials and Methods) and

compared them in regard to size, relative serine composition and serine codon usage

(Fig. 2.6). Of the Vtgs listed here, F. heteroclitus Vtg contains the smallest polyserine

domain (171 a.a.); it also contains the highest relative serine composition (57.6%). We

compared the serine codon usage from each of the domains and found a consistent

pattern: TCX repeats are more prevalent at the 5'end while AGY codons are more

prevalent at the 3' end. Finally, of the six possible serine codons, AGC was invariably

the dominant codon in all five vertebrate polyserine domains.



Discussion


We present the first complete teleost Vtg cDNA sequence along with its translated









38

primary structure. F. heteroclitus Vtg shares 37% 38% identity with other vertebrate

Vtgs and it includes the characteristic N-terminal Lvl region, an internal Pv region and

a C-terminal Lv2 region. The genetic organization of the polyserine domain is consistent

with that found in other vertebrates, from lamprey to chicken, suggesting, at the latest,

a pre-gnathostome arrival of this domain into the Vtg gene. In contrast to other

vertebrate Vtgs, F. heteroclitus Vtg is predicted to be 100 amino acids shorter, and

contains a polyserine region with a 10-20% higher relative serine composition than the

other vertebrates Vtgs. We suspect that the occurrence of liquid phase yolk in F.

heteroclitus is in part due to differences within its Vtg polyserine domain as compared

with the polyserine domains of insoluble yolk producers. The higher than usual relative

serine composition would eventually be modified into a polyphosphoserine domain,

endowing the resulting Pv yolk protein with an uncommonly strong hydrophilic potential.

On examination of the alignment in Figure 2.4, the conserved organization of

vertebrate Vtg is evident: two well-aligned termini interrupted by a polymorphic

polyserine domain. The degree of Vtg conservation among several oviparous species is

further resolved by the phylogenetic tree analysis presented in Figure 2.5. The results

of the branch-and-bound tree search suggest that the present structure of F. heteroclitus

Vtg represents a substantial history of divergence from the ancestral osteichthyean Vtg.

Although, phylogenetically, F. heteroclitus and A. transmontanus are considered

monophyletic as actinopterygian fishes (Nelson, 1989), the Vtg structure of A.

transmontanus was found to be more closely related to the Vtgs of the two tetrapods than

it was to that of F. heteroclitus. Indeed many character traits of the genus Acipenser









39

have long been recognized as tetrapod-like, ie. a holoblastic embryonic cleavage and

anuran-like gastrulation (Balinksky, 1965; Beer, 1981; Conte et al., 1988), an acrosome-

capped spermatozoan (Conte et al., 1988), and development of oviducts from true

Miillerian ducts (Conte et al., 1988). We suggest that the structure of F. heteroclitus Vtg

represents a derived, perhaps more specialized, example of Vtg structure in contrast to

the tetrapod/chondrostean Vtg, which more likely resembles the Vtg of an ancestral

osteichthyean. If this is the case, we would predict that the structure of an elasmobranch

Vtg (especially from a less derived species) would also resemble the tetrapod Vtgs more

closely than it would a teleostean Vtg. Whether the structure of lamprey Vtg represents

an independent derivation, or an even earlier, prototypical vertebrate Vtg, is difficult to

surmise. This question will be more easily answered once a protochordate or

invertebrate deuterostome Vtg (from within the "Vtg family) has been sequenced. Within

the invertebrate outgroup of our phylogram, the two insect Vtgs appear to be highly

derived versions of Vtg structure as compared to the C. elegans Vtg. The C. elegans

Vtg is substantially more similar to vertebrate Vtgs than are the Vtgs of the two insects,

suggesting a faithfulness of the nematode Vtg to an ancestral form originating in a

predecessor common to both vertebrates and platyhelminthes.

In reference to past alignments between multiple Xenopus and chicken Vtgs,

Byrne (1989) described Pv as an independently evolving domain within Vtg. Our

alignment confirms this suggestion. While the two Lv domains of Vtg can be well

aligned among several organisms, the polyserine domain exists in a wide range of sizes










40

568
(65%) Gallus Vtg II
234
(95%) 469
278 Xenopus Vtg A2

(67%)
428
341 Acipenser Vtg

(100%)
665 3 Fundulus Vtg



581
Ichthyomyzon Vtg


836
Caenorhabdiris Vtg 5


642
(100o) Anthonomus Vtg
896

591
Aedes Vtg




Figure 2.5 Branch-and-bound phylogenetic tree analysis comparing selected Vtgs
spanning 600 million years of divergence (Raff et al., 1989). PAUP
(Swofford, 1992) analysis was done on a ClustalV alignment (Higgins et
al., 1992) containing five of the vertebrate cDNAs from Fig. 2.4.:chicken
Gallus domesticus Vtg II; clawed frog Xenopus laevis Vtg A2; white
sturgeon Acipenser transmontanus Vtg; mummichog Fundulus hereroclitus
Vtg; silver lamprey Ichthyomyzon unicuspis Vtg; plus three invertebrate
Vtg cDNAs; nematode Caenorhabditis elegans Vtg 5; boll weevil
Anthonomus grandis Vtg; and mosquito Aedes aegypti Vtg. The Gallus
Vtg was designated as the reference sequence and the C. elegans Vtg was
defined as the outgroup. The number of reconstructed changes in amino
acid sequence occurring along each branch are shown without parentheses;
bootstrap data are depicted at partition boundaries as percentages in
parentheses.









41

from being completely absent in C. elegans (not shown; Speith et al., 1985), to a small

size in F. heteroclitus (99 Ser within 171 a.a. region), to a larger size in the chicken

Gallus (132 Ser within 291 a.a. region). A trend emerges in consideration of these data:

as one proceeds up the vertebrate phylogenetic ladder, Vtg polyserine domains appear

to increase in size. However, at least two exceptions to this trend have been reported:

the lamprey, Ichthyomyzon Vtg possesses a polyserine domain larger than that of F.

heteroclitus (113 Ser within 238 a.a. region; Fig. 2.2) and the

Gallus Vtg III (Byrne et al., 1989) contains a small polyserine domain (37 Ser; not

shown).

Although the vertebrate Vtg polyserine domains vary in size and serine content

as described above, their genetic organizations have sustained an element of similarity.

At the DNA level, the F. heteroclitus polyserine domain contains a distinct cluster of

TCX serine codons directly preceding a larger cluster of AGY serine codons (Fig. 2.2),

a pattern that is found in all other vertebrate Vtg cDNAs. When this cluster organization

was observed by Byrne et al. (1989) in Xenopus and chicken Vtgs, it was speculated that

a non-tetrapod Vtg would perhaps contain a cluster of only one type of serine codon,

representing the original trinucleotide repeating unit, and thus the original Vtg polyserine

domain. However, the polyserine domains presented here from the lamprey, sturgeon,

and mummichog are all dominated by the same two serine codons as is seen in Xenopus

and chicken, suggesting that these two codon clusters have been present within the Vtg

gene since before the divergence of agnathans and












42












Vtg POLYSRUNE DOMAIN Toa % Scrine SAGY %TCX
Codons cadons adons codon
Funduls 444 =;dz& AGY Codons 171 58 63 36
"""""a ""' TCX Codons

.Aci, enser* AGY Codons 212 37 51 49
Acipener .. "' "' TCX Codon

Icduhvomyzan *" ** AGY Codons 238 48 55 45
ul at * TCX Codons

Xenopus '" AGY Codomn 249 39 53 47
S' a" TCX Codons

r, .a umn Bm..... AGY Codons 291 45 80 20
GaU . "' TCXCodons

= one AGY serine codon
= one TCX serine codon 20 codons


















Figure 2.6. A comparison of the serine codon usage in the polyserine domains (see
Fig. 2.4) of five vertebrate Vtgs. Although the number of trinucleotide
repeats vary, the overall codon structure is conserved: a cluster of TCX
codons at the 5' end precedes a larger cluster of AGY codons. Only TCX
or AGY codons are shown. Relative lengths of polyserine domains are
drawn to scale.









43

gnathostomes, over 400 million years ago (Levtrup, 1977). Chen et al. recently

described a mosquito cDNA sequence which codes for a Vtg containing three separate

polyserine domains; 82% of the serines in these domains are coded for by the TCX

codon. Since insects are a highly derived group, it remains unclear whether the TCX

repeats represent the conservation of a primitive polyserine coding domain or an

incidence of convergent evolution between separate Vtg clades.

It has been theorized that the phosphoserine clusters of Pv, documented to bind

Ca in a 1:1 stoichiometric ratio in Xenopus (Follet and Redshaw, 1968; Munday et al.,

1968; Wallace, 1970) are necessary for early bone mineralization in vertebrate embryos.

Even more speculative is the idea that the phosphoserine tracts of Vtg were a necessary

pre-adaptation allowing the original evolutionary emergence of ossified bone in ancestral

chordates. Both the lamprey and the sturgeon are examples of cartilaginous vertebrates,

albeit with bony ancestors (Jarvik, 1980), that have retained their Vtg polyserine

domains. Thus, the possession of a Vtg polyserine domain is not universally concomitant

with the possession of a bony skeleton. Indeed, it appears that polyserine domains can

no longer be considered an exclusive vertebrate Vtg characteristic. Recent reports by

Chen et al. (1994) describing a mosquito (Aedes aegypti) Vtg cDNA and Yano et al.

(1994) describing a silkworm (Bombyx moni) Vtg cDNA, provide invertebrate sequences

containing various arrangements of polyserine tracts. These findings suggests a pre-

chordate origin of Vtg polyserine domains and challenges the hypothesis of Pvs being

unique to chordates. However, polyserine domains are not synonymous with true Pv

domains. Whether these invertebrate polyserine tracts are highly phosphorylated and









44

cleaved, as are bona fide Pv proteins, has not yet been reported. It is possible that

polyserine domains have existed within Vtgs since before the emergence of chordates,

but that Pv proteins, per se, remain a unique chordate trait, representing a novel

modification and utilization of these polyserine tracts.

Though we know little of why Vtg polyserine domains vary in size, findings from

studies of heritable disease may offer clues as to how these size differences originated.

The aberrant amplification of trinucleotide repeats from one generation to another has

recently been coupled to the occurrence of several human genetic diseases including

Huntington's Disease (Huntington's Disease Collaborative Research Group, 1993; review

by Caskey et al., 1992). An increased potential for trinucleotide amplification may

explain the faster rate of evolution attributed to the Pv region in comparison to its two

flanking Lv regions (Byrne et al., 1989). We are aware of very few descriptions of

"yolk-based diseases" in fish (Olin and von der Decken, 1989), and in these it was

neither suspected nor tested whether the disease was caused by aberrant amplification of

the Pv polyserine domain. Diseases aside, novel duplications or omissions in the Pv

polyserine domain may certainly have affected the evolution of specific yolk structures

or functions. F. heteroclitus, possessing the smallest polyserine domain of our

alignment, produces a yolk which remains totally soluble throughout oocyte development.

As the smaller yet serine-enriched polyserine domain of F. heteroclitus Vtg is

phosphorylated and finally processed into a more soluble Pv yolk protein, it may

somehow be prevented from re-combining with the Lv yolk proteins and forming the

insoluble yolk complexes of other vertebrates. Another possible explanation for the









45

persistence of a liquid phase yolk in F. heteroclitus oocytes is that the high proteolytic

activity documented by Greeley et al. (1986) prevents the recombination of Pv and the

Lvs into their usual insoluble particles. By obtaining more examples of Vtg protein

structure from other liquid phase yolk producers, a more substantial and, hopefully,

causal difference between soluble and non-soluble yolk will materialize.

In conclusion, the F. heteroclitus Vtg cDNA along with its amino acid translation

represents the first complete Vtg sequence documented from a teleost fish. The predicted

primary structure suggests to us that a heightened proportion of phosphoserine in the

polyserine domain endows the F. heteroclitus Pv yolk proteins with a higher solubility

preventing the formation of non-soluble yolk particles as is seen in many other

vertebrates. Knowledge of the complete primary structure of F. heteroclitus Vtg

provides us with useful information for mapping the extensive proteolytic processing of

native Vtg into its respective yolk proteins. We hope that this sequence will aid

investigators of other vertebrate Vtgs by providing a piscine model for molecular probes

and antibodies. Finally, we have provided yet another example of the evolutionary

independence of Pv within the Vtg gene, where the codon cluster organization is

preserved, yet the size of the serine clusters and intervening regions remains quite

unpredictable.














CHAPTER 3
SEQUENCE COMPARISON OF FUNDULUS HETEROCLITUS
VITELLOGENINS I AND II



Introduction


Vitellogenin gene families have been described from various metazoan species

including Xenopus laevis (Wahli et al., 1979; Wiley and Wallace, 1980; Tata et al. 1980),

Caenorhabditis elegans (Blumenthal et al., 1984), and chicken (Evans et al., 1987; Byrne

et al., 1989). These small gene families from individual species have likewise been

shown to share genomic organization and sequence identity, establishing these related Vtg

genes as members of an ancient gene superfamily (Speith et al., 1985; Nardelli et al.,

1987; Byrne et al., 1989; Speith et al., 1991).

The existence of four X. laevis Vtg genes can be partially explained by the

hypothesis that an ancient duplication occurred in the X. laevis genome (Thiebaud and

Fischberg, 1977; Bisbee et al., 1977). Tata et al. (1980) reported the extraordinary

occurrence of twelve to sixteen Vtg genes in X. laevis, stating that only four to six of

them were in an expressible form, and that the rest were nonexpressible or "silent".

Another example of a silent Vtg gene has been documented among the six Vtg genes of

C. elegans; whereas vit-2 to vit-6 have been shown to encode specific YP proteins, vit-1

has been described as a pseudogene (Speith et al., 1985) with no apparent translation


46








47

product. In contrast, the three Vtgs genes reported from the chicken include no apparent

silent genes. The primary translation products VtgI, VtgII, and VtgHI were found to be

present in blood in a ratio of 0.33 : 1.00 : 0.08 confirming chicken Vtgl as the major

yolk protein precursor (Wang et al., 1983). Recently two Vtgs have been reported in

related tilapia species, Oreochromis aureus (Ding et al., 1989) and 0. mossambicus

(Kishida and Specker, 1992). These studies established the occurrence of two piscine

Vtgs (180 kDa and 130 kDa) using an immunological approach. The immunological data

from 0. aureus was additionally complemented by a small nucleotide sequence from the

C-terminus of one of the purported Vtgs, probably the larger (Ding et al., 1990).

Though the existence of multiple Vtgs has been established in these species, it remains

unclear as to why several Vtgs would be functionally necessary.

We have recently reported the cDNA sequence and predicted primary structure

of Fundulus heteroclitus Vtg I, as a precursor to non-crystalline, liquid phase yolk

proteins (LaFleur et al.,1995). Here we describe a second F. heteroclitus Vtg cDNA and

protein sequence that we have designated as Vtg II. The predicted primary structure

shares 45% identity with Vtg I (with regions as high as 65%) and contains the same

general domain profile: a large lipoveitellin 1 region, followed by a serine-rich, phosvitin

region and terminating in lipovitellin 2 region. We have confirmed Vtg II MRNA

expression as well as a derived yolk protein cleavage product, verifying that Vtg II

represents a separate but functional Vtg. This report therefore, establishes the existence

of a bona fide Vtg gene family in F. heteroclitus that acts as a precursor to liquid phase

yolk proteins.









48


Material and Methods


Chemicals


Estradiol-170 was obtained from Sigma Chemical Co. (St. Louis, MO).

Radioisotopes, [a-2P]dCTP and [a-35S]dATP, were purchased from New England

Nuclear (Boston MA). Lambda gtlO vector, cDNA synthesis reagents, the subcloning

plasmid Pgem-T, and T4 ligase were obtained from Promega (Madison, WI). All

amplification reactions were performed using a 50:1 mixture of Taq DNA

polymerase:cloned Pyrococcus furiosus DNA polymerase (Stratagene, La Jolla, CA).

All sequencing gels were cast using Sequagel-8 (National Diagnostics, Atlanta)

polyacrylamide reagents. Sequenase version 2.0 DNA polymerase and dideoxy

sequencing reagents were obtained from US Biochemicals (Cleveland, OH). Reagents

for random-primed labeling of probes were purchased from Pharmacia (Piscataway, NJ).

Magna nylon transfer membranes were used for nucleic acid transfers and purchased

from MSI (Westboro, MA). Amino acid N-terminal sequencing, synthesis of

oligonucleotide primers, and a limited amount of DNA sequencing were performed by

the University of Florida Interdisciplinary Center for Biotechnology Research core

facilities.


Cloning strategy using an estrogen-induced liver cDNA library


Seven of the eight overlapping clones resulting in the contiguous cDNA sequence

were isolated from a XgtlO liver library whose synthesis has been previously described









49

(LaFleur et al. 1995) In brief, the library was constructed from the pooled mRNA of

six male Fundulus heteroclitus that had been treated with two IP injections of estradiol-

179/ (0.01 mg/g body weight). The library contained an initial titer of only 6 X 10' total

plaque-forming units, and had been amplified twice.

The initial Vtg II clone was discovered using the degenerate primer ROW 19, and

the vector primer NEB 1231, with 5 j1 of the Xgtl0 library as template in a PCR

reaction utilizing a 50:1 mixture of Taq DNA polymerase:Pfu DNA polymerase. ROW

19 was designed to match a conserved region of Vtgs, ranging from C. elegans to

chicken (Fig. 3.1). A 550 bp band was isolated, inserted into pGem-T, sequenced and

revealed to be a second Vtg cDNA that we designated as Vtg II. This insert was then

isolated and used to generate a random primed 3P-labeled probe. The library was plated

out on 150-mm petri dishes by transfecting E. coli C600hfl cells, and overlaying them

in agarose atop agar plates containing 25 /g/ml tetracycline. Duplicate plaque lifts were

carried out using Magna nylon membranes, and these were probed at 650C in 0.05 X

BLOTTO, 6 X SSC (150 mM NaCI, 15 mM sodium citrate, pH 7) overnight. A large

proportion of the plaques were found to be positive, and 20 agarose plugs were isolated

and stored in SM buffer (.1 M NaCI, 8 mM MgSO4, 50 mM Tris, 2% gelatin) at 4oC

with a drop of chloroform. Thereafter, plug lysates from these Vtg II positive plugs

were used in amplification reactions targeting Vtg II positive clones in a successively

overlapping 5' direction. Six more Vtg II clones were isolated in this manner

approaching the initial methionine codon, but several attempts at targeting the last few

nucleotides to include the initial methionine failed.









50












Fundulus heteroclitus Vitellogenin II cDNA 5195 bp
-- ROW 55 pFhv2a-
pFh2h ROW 19-


pFhv2e
pFhv2d
pFhv2e --------------
pFhv2d --------------























Figure 3.1 Cloning strategy used in isolating the F. heteroclitus Vtg II cDNA (5166
bp). Seven inserts (pFhv2a thru g) were isolated from the Xgtl0 liver
library by anchored PCR with indicated oligonucleotide primers and
inserted into the pGem-T cloning vector. The final cDNA (pFhv2h) was
isolated by RACE using reverse primer ROW 55.









51

A protocol for rapid amplification of cDNA ends (RACE; Frohman et al., 1988),

was performed to retrieve this region using total RNA (described below) isolated from

the liver of an individual reproductively active female F. heteroclitus. A first strand

synthesis reaction was performed using 0.5 /g total RNA, the primer ROW 55 and

Superscript RT, followed by addition of a "poly-C tail" using 4 1l of 1.0 mM dCTP, and

10 units of terminal deoxynucleotidyl transferase (BRL). Then, an amplification reaction

was carried out using the forward primer ROG 51, which targeted the poly C-tail, along

with the reverse primer ROW 55, and the Taq DNA polymerase. Through this effort

we successfully isolated a 230-bp band that was inserted into pGem-T, sequenced and

found to include a valid methionine codon, preceded by a short region that fit the criteria

for a transcription start site (Kozak, 1991).


Estrogen treatment, RNA isolation and analysis


Male and female F. heteroclitus were collected from the estuarine creeks adjacent

to the Whitney Laboratory, and were maintained in running seawater tanks under

14L:10D photoperiod conditions at 25 2C. Fish were maintained for at least one

month before being used for RNA collections.

Experimental groups of fish were subjected to two intraperitoneal injections of

estradiol-173 (0.01 mg/g body weight) dissolved in 50 l1 coconut oil (Kanungo et al.

1990). Control groups were sham-injected with coconut oil alone. The first injection

was performed on day 1, the second injection on day 4, followed by sacrifice and liver

dissection on day 8.









52

Total RNA was isolated from livers by extraction with RNA Stat-60 reagents

(Tel-Test "B", Inc. Friendswood, TX). Tissues were dissected and immediately frozen

in 1.5-ml tubes containing 500 ld of RNA Stat-60 emulsion by immersion in liquid

nitrogen. Tissues were homogenized at 20oC using a Kontes pestle and motor.

Typically, a 300 mg liver yielded 0.350 mg total RNA, with O.D. 260/280 ratios

consistently above 1.8. Total RNA samples were resuspended in diethyl pyrocarbonate-

treated water and stored at -800C until used in analyses.

Before electrophoresis, aliquots of 15 ig total RNA were precipitated in

isopropanol, and denatured in 2.2 M formaldehyde, 50% formamide, 50 mM MOPS (pH

7.0) for 30 min at 65oC. Samples were electrophoresed through gels containing 1.0%

agarose, 0.6 M formaldehyde, 50 mM MOPS, and 1 mM EDTA for 2.0 hours at 3.5

V/cm gel in 50 mM MOPS, 1 mM EDTA running buffer. RNA was blotted onto Magna

nylon membranes by capillary action with 20 X SSC, immobilized by U.V. crosslinking

and visualized by staining briefly with methylene blue.

Random-primed [32P]probes were made for resolving Vtg I and Vtg II RNA

transcripts. The Vtg I probe was synthesized from a PCR product off of the template

pMMB1 using primers ROW 5 and MB 13, resulting in a 639-bp cDNA probe from

nucleotide 4284 to 4923 of the Vtg I cDNA. The Vtg II probe was made from pFhv2a

using primers ROW 19 and ROW 33, yielding a 277-bp probe from nucleotide 4692 to

4969 of the Vtg II cDNA. After random prime labeling, oligonucleotide probes were

separated from non-incorporated [3P]dCTP by size chromatography through Stratagene













Figure 3.2 Translated amino acid sequence (1687 residues) of the putative F.
heteroclitus Vtg II polypeptide. The signal peptide, predicted by
the method of von Heijne (1986) is indicated by underlining, and
verified by the N-terminal sequence obtained from an isolated 69-
kDa yolk protein (shaded lettering). The annealing site of ROW
19, used to isolate the initial insert, is indicated by double
underlining. A polyadenylation site is indicated by underlining.












54









aatccaccagcc 12

ATGAGGGTGCTTGTGCTGGCTCTCACTGTGGCCCTTGTGGCCOGGAACCAGGTGAGCTATGCCCCA 78
: R V L T V. A L, V.- A- G N Q V S Y A P

GAATTTGCCCCTGGAAAGACCTACGAGACAAGAGAAGGTAGT ATTCTGGGTGGCCTGCCGAG 144
E F A PG K T Y E Y KY E G Y I L G G L P

GAGGGCCTGGCAAAGGCTGGGGTGAAGATCCAGAGCAAAGTCTTGATCGGTGCAGCAGGTCCTGAC 210
E G L A K A G V K I Q S K V L I G A A G P D

AGCTACATTCTGAAACTTGAAGACCCTGTCATCTCGGGGTACAGTGGCATTAGGCCTAAAGAGGTT 276
S Y I L K L E D P V I S G Y S G I W P K E V

TTCCACCCTGCCACAAAGCTCACCTCAGCTCTCTCGCTCAGCTCTTGACACCCGTCAAGTTTGAG 342
F H P A T K L T S A L S A Q L L T P V K F E

TATGCCAACGGAGTGATCGGAAAAGTGTTCGCACCTCCAGGCATCTCTACAAATGTGCTGAATGTC 408
Y A N G V I G K V F A P P G I S T N V L N V

TTCAGGGGACTCCTCAACATGTTCAGATGAACATCAAGAAGACTCAGAATGTGTATGACCTGCAA 474
F R G L L N M F Q M N : K K T Q N V Y DL Q

GAGACTGGAGTAAAAGGTGTGTGCAAGACACACTATATCCTTCATGAGGACTCCAAGGCTGATCGC 540
E T G V K G V C K T HY I L H E D S K A D R

CTCCACTTGACGAAAACCACAGACCTGAATCACTGCACCGACAGCATCCACATGGATGTTGGCATG 606
L H L T K T T 0 L N H C T D S I H M D V G M

GCTGGTTATACGGAAAAATGTGCAGAGTGCATGGCTCGGGGAAAAACTCTTTCAGGAGCAATTTCT 672
A G Y T E K C A E C M A R G K T L S G A I S

GTCAACTACATCATGAAGCCGTCTGCCTCTGGCACCTTGATCCTAGAGGCAACCGCCACTGAGCTT 738
V N Y I M K P S A S G T L I L E A T A T E L

CTCCAGTACTCGCCCGTCAACATTGTAAATGGAGCTGTCCAGATGGAGGCTAAGCAGACCGTGACC 804
L Q Y S P V N I V N G A V Q M E A K Q T V T

TTCGTGGACATCAGGAAGACCCCATTAGAGCCCCTCAAAGCAGACTATATTCCCCCTGGATCGCTC 870
F V D I R K T P L E P L K A D Y I P R G S L

AAGTACGAGTTAGGCACTGAATTCCTACAGACACCAATTCAGCTTCTGAGGATCACCAATGTCGAG 936
K Y E L G T E F L Q T P I Q L L R I T N V E

GCTCAGATTGTTGAGTCTCTGAACAACCTAGTGAGCCTCAATATGGGCCATGCCCATGAGGATTCC 1002
A Q I V E S L N N L V S L N M G H A H E D S

CCTCTGAAGTTTATTGAGCTCATCCAGCTGCTGCGTGTGGCCAAGTATGAGAGCATTGAAGCTCTC 1068
P L K F I E L I Q L L R V A K Y E S I E A L

TGGAGTCAGTTTAAAACCAAAATTGATCACAGGCACTGGTTGCTGAGCTCTATCCCTGCCATTGGT 1134
W S Q F K T K I D H R H W L L S S I P A I G

ACTCATGTTGCTCTCAAGTTCATCAGGAGAGATCGTTGCTGGTGAAGTCACTGCCGCTGAGGCT 1200
T H V A L K F I K E K I V A G E V T A A E A

GCTCAGGCCACATGTC2TTACACACTGGTGAAGGCCGACCTGGAGGCAATCAAGCTTCAGGAG 1266
A Q A I M S S T H L V K A O L E A I K L Q E

GGCCTGGCTGTGACCCCTAATATTCGGGAAAATGCAGGTTTGCGTGAACTCGTTATGCTGGGCTTT 1332
G L A V T P N I R E N A G L R E L V M L G F











55




GGCATCATGGTT~ACAAATACTGTGTGGAGAAC=CTCATGTCCATCTGAGCTGGTCAGGCCAGTT 1398
G : M V H K Y C V E N ? S C P S E L V R P V

CATGACATTATTGCCAAGGCTCTTGAGAAACGCGACAATGATGAGCTCTCCTGGCTCTCAAAGTT 1464
H OD I A K A L E K R D N D E L S L A L K V

CTGGGTAATGCCGGACATCCCAGCAGCCTGAAGCCAATCATGAAACrCTCCTGGCTTTGGCAGC 1530
L G N A G H P S S L K P I M K L L P G F G S

TCTGCCTCCGAACTTGAGCTCAGAGTTCACATTGACGCTACACTGGCGCTGAGGAAAATTGGCAAG 1596
S A S S L E L R v H I D A T L A L R K I G K

AGAGAACCCAAGATGATTCAGGATGTGGCCCTTCAGCTCTTCATGGACAGGACTCGACCCAGAG 1662
R E P K M I Q D V A L Q L F M D R T L D P E

CTCCGTATGGTGCTGTTGTTGTGCTGTTTGATACCAAGCTACCTATGGGTCTGATAACCACTCTC 1728
L R M V A V V V L F D T K L P M G L I T T L

GCTCAGAGTCTCCTGAAACAGCCAAACCTGCAGGTCCTTAGCTTTGTCTCTTACATGAAGGCC 1794
A Q S L L K E P N L Q V L S F V Y S Y M K A

TTCACCAAGACCACCACCCCGGACCATTCCACTGTAGCCGCTGCCTGCAATGTTGCCATCAGGATC 1860
F T K T T T P D H S T V A A A C N V A I R I

CTCAGCCCAAGATTCGAAAGACTGAGCTACCGCTACAGCCGAGCTTTCCATTATGACCACTATCAT 1926
L S P R F E R L S Y R Y S R A F H Y D H Y H

AATCCTTGGATGCTGGGAGCTGCTGCCAGCGAT ACATCAATGATGCCGCGACGTATTGCCA 1992
N P W M L G A A A S A F Y I N D A A T V L P

AAAAACATCATGGCAAAAGCTCGCGTTTACCTCTCTGGAGTGTCTGTTGATGTTCTGGAGTTTGGA 2058
K N I M A K A R V Y L S G V S V D V L E F G

GCCAGAGCTGAAGGAGTGCAAGAGGCCCTTTTGAAAGCCCGTGATGTTCCTGAGAGTGCAGACAGG 2124
A R A E G V Q E A L L K A R D V P E S A D R

CTCACCAAGATAAGCAAGCTTAAGGCTCTGACTGAGTGGAGGGCCAATCCTTCCCGCCAGCCT 2190
L T K M K Q A L K A L T E W R A N P S R Q P

CTCGGCTCT~CGTACGTGAAGGTTCTTGGGCAGGATGTTGCTTGCAAACATCGACAAAGAAATG 2256
L G S L Y V K V L G Q O V A F A N I D K E M

GTTGAGAAGATCATTGAGTTTGCAACTGGACCTGAAATCCGCACCCGTGGCAAAAAGGCCTTGGAC 2322
V E K I I E F A T G P E I R T R G K K A L D

GCCCTGCTTGGTTACTCTATGAAATACTCCAAGCCAATGTCGGCCATTGAGGTCCGTCACATC 2388
A L L S G Y S M K Y S K P M S A I E V R H I

TTCCCCACCTCTCTTGGTTTACCCATGGAGCTCAGTCTGTACACTGCTGCCGTGACAGCCGCATCC 2454
F P T S L G L P M E L S L Y T A A V T A A S

GTTGAAGTACAAGCCACCATTTCACCACCACTTCCCGAGGACTTCCATCCTGCCCACCTACTGAAG 2520
V E V Q A T I S P P L P E D F H P A H L L K

TCTGATTTTCCATGAAGGCTTCAGTCACTCCAAGTGTATCTTTGCACACCTATGGAGTTATGGGA 2586
S D : S M K A S V T P S V S L H T Y G V M G

GTGAATAGTCC CTCATCCAGGCTTCTGTGCTGTCAAGAGCCAAAGACCATGCAGCCTCCCAAA 2652
V N S P F I Q A S V L S R A K D H A A L P K

AAGATGGAGGCAAGACTTGACATAGTCAAGGGTTACTTTAGCTACCAGTTCCTGCCTGTTGAGGGT 2718
K M E A R L D I V K G Y F S Y Q F L P V E G








Figure 3.2--continued














56






GTTAAAACAATGCCATCTGCTCGTCTTGAAACAGTTGCCATTGCAAGAGATGTTGAAGGCCTCGCT 2784
V K T I A S AR L E T V A I AR D V G LA

GCTGCCAAAGTCACACCGGTTGTCCCATATGAGCTATTGTGAGCAAGAACGCCACTTTAAATCTT 2850
A A K VT P V VP YE P V S K N AT L N L

TCACAGATGTCTTACTATCTGAATGATAGCATATCAGCATCATCTGAACTTCTTCCTCT"TCGCTG 2916
S Q M S Y Y L ND SI S A S S E LL P F S L

CAAAGGCAAACTGGCAAAAATAAAATCCCCAAGCCCATTGTGAAGAAAATGTGTGCAACAACGTAT 2982
Q R Q T G K N K I P K P IV KK M C A T T Y

ACGTATGGGATTGAGGGCTGCGTTGACATTTGGTCTCGCAATGCAACCTTCCTCAGAAACACCCCC 3048
T Y G I E G C VD I W SR N A T F L N T P

ATCTACGCCATAATTGGAAACCACTCTCTTTGGTTAATGTTACCCCAGCTGCTGGACCGTCCATC 3114
SY A I I G N H S L L VN V T P A A PS I

GAAAGGATCGAAATCGAGGTTCAGTTTGGTGAACAAGCAGCAGAAAAGATCCTTAAAGAGGTTTAC 3180
ER I E I E V Q F G Q A A E K I L K V Y

CTGAATGAGGAGGAAGAAGTACTTGAAGACAAAAACGTCCTTATGAAGCTGAAGAAGATTCTGTCT 3246
L N E E E V L E D K V L K L K K I L S

CCGGTCTGAAGAACAGACAAAGCTTCATCCCTAGTTCGGGCAGCTCTCGC'TCCAGTAGATCT 3312
P G L K N S T K A S S S SS G S S R S S R S

CGCTCCAGCAGCTCCAGCAGCTCCAGCAGCTCCAGCAGCTCCAGCCGTTCCTCCTCTAGCTCTTCC 3378
R S S S S S S S S S S S S S RS S S S S S

AGGAGCTCTTCCTCTTGCGCCGCATAGCCAAGATGTTGGATCTTGCGATCCCCTCAACATAACA 3444
R S S S S L R R N S K M L D L A P N T

TCAAAGAGATCCTCCAGCAGCTCCTCCAGCTCCAGCTCCTCCAGACTCTCCTCCCAGCTCCCT 3510
S K R S S S S SS S S S S S S S S S S S S S

TCTAGCTCCAAGACCAAGTGGCAGCTGCACGAAAGGAACT7CACCAAGGATCACATCCACCAGCAT 3576
S S S K T K W Q L H ER N F T K D H H Q H

TCCGTCTCAAAAGAACGTCTTAACAGCAAGAGCAGTGCGAGAGCTTTGAATCCATTTACAACAAG 3642
S V S K ER L N S K S S A S S F ES I Y N K

ATCACATACCTGTCTAACATCGTCAGCCCAGTGGTCACAGTCCTTGTCCGTGCCATCAGAGCTGAC 3708
I T Y L S N I VS P V V TV L V R A I RA

CACAAGAACCAGGGGATCAGATCGCTGTGTACTAGACAAACTCACTACCAGAGTGCAGATCATT 3774
H K N Q G Y Q I AV Y Y D K L T T R V Q I I

GTGGCCAACCTCACTGAAGATGACAACTGGAGAATCTGTTCTGACAGCATGATGCTCAGCCACCAC 3840
VAN L T ED ON W R I C SD S M M L S H H

AAAGTGATGACTCGAGTCACCTGGGGCATTGGATGCAAGCAGTACAACACCACGATCGTGGCCGAA 3906
K V M T R V T W G I G C K Q Y N T T I VA E

ACTGGTCGCGTTGAGAAGGAGCCTGCCGTCCGTGTGAAGCTGGCCTGGGCCAGACTCCCTACTTAC 3972
T G R VE K E P AV R V KL AW A R L PT Y

ATCAGGGATTATGCAAGAAGAGTGTCCAGGTACATTTCCCGCGTCGCTGAGGACAATGGAGTGAAC 4038
I RD Y A R R V S R Y I S R V A N G V N

AGGACAAAGGTCGCCAGTAAACCCAAAGAGATCAAACTGACTGTAGCTGTTGCCAACGAGACAAGC 4104
RT K V A S K P K EI KL T V A V A NE T S






Figure 3.2--continued











57








CTGATGTCACGCTGAATACACCAAAGAACACC-T-TTCAAACTGGGATGGGTTCT-CCC-TTTAC 4170
L N V T L N T P K N T F F K L G W V L P F Y

CTACCAATTAACAACACTGCTGCTGAGCTGCAGGCATTCCAGGGCAGGTGGATGGACCAGGTCACA 4236
L P I N N T A A E L Q A F Q G R W M 0 Q V T

TACATGCTCACCAAGTCTGCTGGCAGCGAGTGCACCGTGGTTGAAGACACAGTGGTCACTTTCAAC 4302
Y M L T K S A A A E C T V V E D T V V T F N

AACAGGAAGACA CGGAGACGCCCCACTCTGCCATCAGGTCTGGCTCAAGATTGCACATCT 4368
N R K Y K T E T P H S C H Q V L A Q D C T S

GAAATCAAATCATAGTG CTGAAGAGGGATCAAACAGCAGAACGGAATGAGA TTCAGTATTAAG 4434
E I K F I V LL K R D Q T A E R N E I S I K

ATTGAAAACATTGTGTTGACATGTATCCCAAGGACAACGCTGT2GTGGTGAAGGTTAATGGAGTA 4500
I E N I D V D M Y P K D N A V V V K V N G V

GAAATTCCTCACCAACCTGCCATATCAGCATCCAACAGGCAACATACAGAT CCGACAAAGAGAA 4566
E I P L T N L P Y Q H P T G N I Q I R Q R E

GAGGGCATCTCTCTGCATGCTCCCAGTCATGGCCTTCAGGGTCTTCCTCAGTT.AAACAAAGTG 4632
M G I S L H A P S H G L Q Z V F L S L N K V

CAGGTTAAGTTGTTGACTGATGAGAGGCCAGACGTTGGGGCCCGGAAAGGCCGACGGGGAA 4698
Q V K '. V D W M R G Q T C G L C G K A 0 G E

GTCAGACAGGAGTACAGCACTCCCAATGAACGGGTGTCCAGGAACGCAACCAGCTTCGCTCATTCC 4764
V R Q E Y S T P N E R V S R N A T S F A H S

TGGGTGCTGCCCG GAGCTGCCGTGACGCCTCAGAGTGCTACATGCAACTGAATCGGTGAAG 4830
W V L P A K S C R 0 A S E C Y M Q L E S V K

CTCGAGAAACAGATCAGCCTGGAAGGCGAGGAATCCAAATGCTACTCAGTCGAACCTGTCTGGCGC 4896
L E K Q I S L E G E E S K C Y S V E P V W R

TGTCTCCCTGGCCGTGCACCAGTGAGAACCACCTCCGTCACTGTCGGGCTACCATGCGTGTCTCTG 4962
C L P G C A P V R T T S V T V G L P C V S L

GATTCAACCTGAATCGCTCTGATAGTCTCAGCAGCATCTATCAGAAGAGCGTTGACGTGAGCGAG 5028
D S N L N R S D S L S S I Y Q K S V D V S E

ACGGCAGAGTCCCACGGCGTCGCTGCACTCCTC AGGCGTGCCTAAacgtgttgc cctgact 5094
T A E S H L A C R C T P Q C A -

ttcgttec-gt-tttgqgttatatggatgctctaaactaaaataaagaaqcaataaaaaaaaaaa 5160
aaaaaattcagcstggactuaaccaggcgaactt 5195

















Figure 3.2-continued









58

Nuc-Trap columns. All RNA hybridizations were carried out at 65oC in 1 X Denhardt's

solution, 6 X SSC, and 0.1 % SDS without formamide (Denhardt 1966).

Autoradiographs were analyzed using the Bio Image Whole Band Analyzer system

(Millipore, Ann Arbor). For estimating amounts of Vtg RNA visualized on gels, RNA

was transcribed from the Vtg I plasmid pMMB1 and the Vtg II plasmid pFhv2a, using

Ambion reagents. Transcribed RNA yields were measured spectrophotometrically, and

diluted to a concentration of 66.7 pg//il. For RNA standards, 133 pg transcribed RNA

from both pMMB1 and pFhv2a was loaded onto each gel.

Results


The complete cDNA sequence (5166 bp) of a Vtg mRNA, encoding a protein

designated as Vtg II is provided in Figure 3.2. The eight overlapping pGem-T clones

that were used to complete the sequence are represented in Figure 3.1. A ClustalV

alignment of Vtgs I and II by the method of Swofford et al. (1993) revealed 45%

sequence identity between the two amino acid sequences (Fig. 3.3).

In general, the two sequences share the same profile as other reported Vtgs: a

large lipovitellin 1 region that is followed by a polyserine domain (assumed to represent

phosvitin) that, in turn, is followed by a lipovitellin 2 region containing a

substantial amount of conserved cysteines. Like Vtg I, Vtg n contains several predicted

N-glycosylation (16), phosphorylation (45), and N-myristoylation sites (16), agreeing

with our expectations for a lipophosphoglycoprotein. The smaller length of the Vtg II

a.a. sequence (1687) compared to that of Vtg I (1704) can be primarily attributed to gaps

in the polyserine domain. A graphical comparison of the polyserine domains of Vtg













Figure 3.3 ClustalV alignment of F. heteroclitus Vtg I and Vtg II. A
polyserine domain defined according to a previously published
alignment (LaFleur et al. 1995) is indicated by shaded lettering.
Identical residues are denoted by asterisks. Vtg I and Vtg II share
45% overall sequence identity.













60





Vtg : MKAVVLATLAFVAGQN-- 'apEFAAGKTYVYKyXEAL:LGGLPEEGLARA 48
Vtg :I MRVLVLALIVALVAGNQVSiAPEAPGKTYEYKYEGY'L3GLSEEGLAKA 50


Vtg I GLKISTTKLLLSAADQNTYMLKLVEPELSEYSGIWPKDPAVPATKLTAALH 98
Vtg :I G/KI:QSKVL GAAGPDSYILXLEDPVISGYSGIWPKEVFHPATKLTSALS 100
we en tee a ***l5* *5W

Vtg I LSSQFPSSLNTPMVFVGKVFAPEEVSTL'rVLNIYRG:NI LQLNIKXXTHI 148
Vtg ; AQLLTPVKFEYANGVIGKVFAPPGISTNVLNVFRGLL'NMFQMNIKXTQNV 150


Vtg I YDLQEVGTQGVCKTILYSISEDARIENILLTXTRDLSNCQERLNMD IGLA- 197
Vtg I; YDLQETGVKGVCKT.YILHEDSKADRLHLTKTTDLNHCTD SIMDVGMAG 200
a*nt a t*a a fe **t ** tt *

Vtg I YTSEKCKCQEETXNLRGTTTLSYVLXPVADAVMILKAYVNELIQFSPFSE 247
Vtgq : YTEKCAECMARGKTLSGAISVNYIMKPSASGTL:LEATATELLQYSPVNI 250
*.... n * a .. *a. .

Vtg I ANGAAQMRTKQSLEFLEIEKEPIPSVKAEYRHRGSLKYEFSDELLQTPLQ 297
Vtg I: VNGAVQMEAKQTVTFVDIRKTPLEPLKADYIPRGSLKYELGTEFLQTPIQ 300
*.. .I *. It * *** *.w. *

Vtg I LIKISDAPAQVAEVLKHLATYNIEDVHENAPLKFLELVQLLRIARYEDLE 347
Vtg II LLRITNVEAQIVESLNNLVSLNMGHAHEDSPLKFIELIQLLRVAKYS SI 350
as s en .* ** n* n **

Vtg MYWNQYKKMSPHRHWFLDTIPATGTFAGLRFIKEKFMAEE TIAEAAQAF 397
Vtg II ;ALWSQFKTKIDHRHWLLSSIPAIGTHVALKFIKEKIVAGEVTAAEAAQAI 400
* a *sweet *Inst *t *1** ** * l*.*

Vtg : ITAVHMVTADPEVIKLFESLVDSDKVENPLLaREVVLGYGTMVNKYCTNK 447
Vtg I: MSSTHLVKA.DLLAIKLQEGLAVTPNIRENAGLRELVMLGFGIMVHKYC'/E 450
a ea* fi * at Ine ft* n s *fe

Vtg I TVDCPVEL KPIQQRLSDAIAKNEEENIILYIKVLGNAGHPSSFKSLTKI 497
Vtg I= NPSCPSE.L'RPVHDIIAKALEKRDNDELSLALKVLGNAGHPSSLKPIMKL 500


Vtg ; MPISGTAAVSLPMT:HVEAIMALRNIAKXESRMVQELALQLYMDKALHPE 547
Vtg :I LPGFGSSASELELAVHIDATLALRKIGKREPKMIQDVALQLMDRTLDPE 550


Vtg I LRMLSCI'/LFETSPSMGLVTTVANSVKTEENLQVASFTYSHMKSLSRSPA 597
Vtg II LRMVAVVVLFDTKLPMGLITTLAQSLL:ZPNLQVLSFVYSYMKAFT.TTT 600
5fl esn n Ie t a n e w ta ase w

Vtg I TIHPDVAAACSAAMKILGTKLDRLSLRYSKAVHVDLYNSSLAVGAAATAF 647
Vtq II PDHSVIAAACNVAIRILSPRFERLSYRYSRAFHYDHYHNPWMLGAAASAF 650


Vtg I YINDAATFMPKSFVAKTKGFIAGSTAEVLE GANIEGLQELILKNPALSE 697
Vtg II Y=NDAATVLPKNIMAKARVYLSGVSVDVLEFGARAEGVQEALL VPE 700
nan..** fi a. at *. an ** e a a

Vtg I STDRITKMRVIKALSEWRSLPTSKPLASVYVKFFGQEIGFANIDKPMID 747
Vtg II SADRLTKMKQALKALTEWRANPSRQPLGSLYVKVLGQDVAFANIDKEMVE 750
a **t. *nn** .f ** a a a. *. ***enn *

Vtg I KAVKFGKELPIQEYGREALKALLLSGINFHYAKPVLAAEMRRILPTV AGI 797
Vtg II K:IEFATGPEIRTRGKiALDALL-SGYSYSM KPMSAIEVRHIFPTSLGL 799
~* a a fi *fi* at t ta w a

Vtg I PMELSLYSAAVAAASVEIKPNTSPRLSADFDVKTLLETDVELKAEIRPMV 847
Vtg II PMELSLYTAAVTAASVEVQATISPPLPEDFHPAHL LKSDISMKASVTPSV 849


Vtg I AMDTYAVMGLNT IFQAALVARAKLHSVVPAKIAARLNIKEGDFKLEALp 897
Vtg II SLHTYGVMGVNSPFIQASVLSRAKDHAALPKKMEARLDIVKGYFSYQFLP 899
*i nan a ne a** a * a f a a a*n











61



VtgI VDVEIi-SM3NVTTFAVARNIW~ PLVER2ITPLTKVV PIPIRRHTSiK 947
Vtg II 'EGVK:AARGLAAAAKVTPWVP--- A SPIVSKiNATNL 946
a a a ** ** at *

Vtg I DPTR----NSMLDSSELLPME---EDVEPS iYKFRRFAiKYCAKHIGV 990
Vtg I; SQMSYYLNDSISASSELLPFSLQRQTGKKIP ---- KPIVKKMCATTYTY 992


Vtg I GLKACFIFASQNGASIQD IVLYLAGSHNFSFSVTPIEGEVVERLMEVK 1040
Vtg II GIEGC'D IWSRNATFLRNTPIYAIIGNHSLLVNVTPAAGPS ERISIEVQ 1042


Vtg I VGAKAAEXi'VRINLSEDEETEEGGPVLVXLKiLfISSRNSSSSSSSSS 1090
Vtg II FGzQAAEK: LKEYLEEEEEvLZDKNVLMKLKi.asGPExssa SSSS 1092
Satit t t at *I t ** t*t ttt.t



a. a t nt a et a.ttt at s tw tt *

Vt; 2 1190


Vtg I SS 1240
Vtg II LSaDssanSSSSSSSSSS --'SSSSSSSSat 1211




Vtq I KLGEE- EAWAVILRAVKADKRMVGYQLGFYLD KPNARVQIIVyANISSD 1289
Vtg II TYSN IVSPVVTVLVRAIRADHKNQGYQIAVYYDKLTTRVQIIVANLTED 1261
a t ** *t ** **et at *atattt t


Vtg : SNWRICADAVVLSKHKVTTKISWGEQCRKYSTNVTGZTGIVSSSPAARLR 1339
Vtg II DNWRICSDSMMLSHHKVMTRVTWGIGCKQYNTTIVAETGRVEKEPAVRVK 1311


Vtq I VSWERLP STLKRYGXMVNKYVP-VKILSDLIHTEENSTRNISVIAVATS 1388
Vtqg I LAWARL PTY RDYARRVSRY SRVAEDNGVNRTKVASKPXE KLT'VAVAN 1361


Vtg I EKTD IITKTPMSSVYNVTMHLPMCIPIDEIKG-LSPFDEV- DKIHFMV 1436
Vtg II ETSLNVTLUTPKNTFFKLGWVLPFYLPINNTAAELQAFQGRWMDQVTYML 1411


Vtg I SKAAAAECS FEDTLYTFNNRSYKNKMP SSCYQVAAQDCTD ELK2XMVLLR 1486
Vtqg I TKSAAAECTVIEDT/VTFNNRKYXTP HSCHQVLAQDCTSE FIVL.L 1461
*ta ttt twa tWate w ttb a a e t a a tat

Vtg I KD-SSEQHHINVKISEDIDIMFPKDONVTVKVNEME:PPPACLTATQQLP 1535
Vtg II RDQTAERNEZSIKIENIDVDMYPKDNAVVVKVNGVEIPLTNLPYQHPTGN 1511



Vtg I LKIKTKRRGLAVYAPSHGLQEVYFDRKTWRI2KVADWMKGKTCGLCGKADG 1585
Vtg II IQIRQREEGISLHAPSHGLQEVFLSLNKVQVKVVDWMRGQTCGLCGKADG 1561


Vtg I EIRQEYHTPNGRVANS ISFAHSWILPAESCRDASECRLKLESVQLEKQL 1635
Vtq II EVRQEYSTPNERVSRNATS AHSWVLPAKSCrDASECYMQLESVKLEKQI 1611
a tan. at we a tttttt tat ttttttat tttt Wttt

Vtq I TIHGEDSTCFSVEPVPRCLPGCLPVKTPVTVGFSCLA-------SDPQT 1678
Vtg II SLEGEESKCYSVEPVWRCLPGCAPVRTTSVTVGL.PCVSLDSNLNRSDSLS 1661
an a a aetna ananna at at attn a an

Vtg I SVYDRSVDLRQTTQAHLACSCNTKCS 1704
Vtg II S YQKSVDVSETAESHLACRCTPQCA 1687




Figure 3.3--continued








62

I and II (Fig. 3.4) reveals a departure from a trend that had previously been noted

concerning serine codon usage in Vtg I (LaFleur et al., 1995) and other vertebrate Vtgs

(Byrne et al., 1989) Whereas the polyserine domains of most vertebrate Vtgs contain a

cluster of TCX codons at the 5' side of the polyserine coding domain and a cluster of

AGY codons at the 3' side, the Vtg II polyserine domain appears to have these codons

equally dispersed, with no obvious clustering.

Northern blot analyses showed that the mRNA of Vtg II transcript can be found

in both estrogen-treated males and spawning females, at an approximate size of 6.0 Kb

(Fig 3.5). By analysis of duplicate blots with separate Vtg I and Vtg II cDNA probes,

it was found that Vtg II transcripts numbered ten times less than those of Vtg I. Vtg I

probes did not cross-hybridize with RNA transcribed from Vtg II clones and vice versa,

confirming that two separate mRNAs for Vtg I and Vtg II were indicated (Fig 3.5).

The N-terminal amino acid sequence of a 69 kDa protein band isolated from the

yolk protein of ovulated eggs was determined as to be N Q V S Y A P E F A P G x T

Y, where "x" was undetermined ("YP 69" indicated in Chapter 4). Allowing the

predicted K residue in the unidentified "x" position, this sequence provides a perfect

match for the N-terminus of Vtg II after cleavage of the predicted signal peptide (Fig.

3.2, shaded lettering) and indicates that Vtg II is not blocked as is the case with the N-

terminus of Vtg I (LaFleur et al. 1995). These data verify that the Vtg II protein is in

fact expressed, transported, and incorporated as a yolk protein precursor in oocytes of

F. heteroclitus.








63









Polyserine Domain Number of
Ser codons

vtgl ,,n^l,,,1, J1 AGY 62

Vtgl 1 AGY 29
vtgI Wl1 TCX 40

S= one AGY serine codon i--.
I one TCX serine codon 20 codons


















Figure 3.4 A comparison of the serine codon usage in the polyserine domains (see
Fig. 3.3) of F. heteroclirus Vtg I and Vtg II. Whereas the TCX and AGY
codons of the Vtg I polyserine domain are clustered into two separate
groups, the TCX and AGY codons of Vtg II show no apparent clustering.
Only serine codons are shown, with relative lengths of the domains drawn
to scale.









64



Discussion


F. heteroclitus Vtg II cDNA and predicted amino acid sequence are provided in

Figure 3.2. Vtg II mRNA is present in the liver of estrogen-treated males and normal,

spawning females (Fig. 3.5). The N-terminal sequence of a yolk protein isolated from

ovulated eggs was found to be identical to the predicted N-terminus of the putative Vtg

II translation product. Taken together these data indicate that the yolk proteins of F.

heteroclitus are derived from a mixture of at least two estrogen-induced liver precursors,

Vtg I and Vtg II, establishing the existence of a Vtg gene family in F. heteroclitus.

These two cDNAs represent the first two Vtg sequences from a single vertebrate species

to have been completely sequenced, offering a unique perspective into the possible

variance between Vtg isoforms occurring in single species.

Examination of the alignment of Vtg I and Vtg II reveals typical conservation of

lipovitellin regions seen among other vertebrate Vtgs. As previously described for other

vertebrate Vtgs (LaFleur et al., 1995) poor alignment occurred in the polyserine

domains. Although the tandem repeats of serine can be aligned in small stretches, the

overall lengths and intervening amino acid sequences are highly variable, resulting in a

region whose conservation is difficult to interpret. In an attempt to compare and

visualize these polyserine domains, hypothetical boundaries were drawn up according to

those used in a previous report (LaFleur et al., 1995), and a graphical representation was

created showing relative domain length as well as serine codon usage (Fig. 3.4).

Whereas the serine codons (TCX and AGY) of the Vtg I polyserine domain appear to be














Figure 3.5 Northern blot analysis comparing relative expression of F.
heteroclitus Vtg I and Vtg II mRNAs.

A) Methylene blue staining of duplicate samples transferred to
nylon membranes before hybridization, showing equal loading of
lanes, as indicated by 28s and 18s rRNA bands. Lanes a and a'
contain 300 pg Vtg I RNA translated from plasmid cDNA
(pMMB1); lanes b and b' contain 300 pg Vtg II RNA translated
from plasmid cDNA (pFhv2a); lanes c and c' contain 15 /g total
liver RNA from an estrogen-treated male; lanes d and d' contain
15 1g total liver RNA from a female four days before spawning;
lanes e and e' contain 15 /g total liver RNA from a female 4 days
after spawning. RNA markers (kb) are indicated with arrows on
the left.

B) Autoradiographs of the membranes shown above, indicating
bands hybridizing to Vtg I (left side) and Vtg II (right side) DNA
probes. Note that the Vtg I probe did not hybridize to the Vtg II
control RNA (lane b) and Vtg II probe did not hybridize to the Vtg
I control RNA (lane a').










66










A Pre-Hyb Pre-Hyb
Vtg I Vg II
a b c d e a' b' c' d' e'
RNA
kb
7.46-

4.4 10- -.* -28s

2.37- 18s
t 18s
1.35-








B Vtg I Vtg II
RNA a b c d e a' b' c' d' e'
kb
7.46
4.43

2.37-

1.35 r





0.24








67

separated into two general clusters, those of the Vtg II polyserine domain are

randomly interspersed. This arrangement of polyserine codons again confirms

the observations noted by Byrne et al., (1989) that the polyserine, or phosvitin,

domain is an independently evolving domain within the Vtg gene, showing more

variability than its flanking lipovitellin regions.

The predicted post-translational modifications of Vtg II are in agreement

with expectations for a lipophosphoglycoprotein. Although 45 phosphorylation

sites may appear to be high, we expect an even higher amount of phosphorylation

than is predicted. Seven of the 45 predicted phosphorylation sites occur within

the polyserine domain (all protein kinase C sites), however, from previously

published accounts, it is likely that every serine residue in this domain is

phosphorylated, resulting in a very hydrophilic domain with a highly negative

charge. Phosvitin yolk proteins have been described as possessing the highest

amount of phosphorylation of any known proteins. Unfortunately, the hepatic

Vtg kinase responsible for phosphorylating the extensive polyserine domains of

Vtg has not yet been isolated or characterized, so that an algorithm predicting its

target sites is not yet available.

Considering the ratio of expression of Vtg I and Vtg II, our data suggest

that Vtg I is the major yolk protein precursor. Vtg II mRNA is present in the

liver of spawning females at ratio of 1:10 with respect to Vtg I RNA, as

evidenced by northern blots. In SDS PAGE analysis, YP 69, which was mapped

to Vtg II, is hardly discernable (not shown here) when compared to the Vtg I-









68

derived yolk proteins, YP 125 and YP 105 (Chapter 4), agreeing well with the

mRNA expression data. This may suggest a difference in the interaction between

the estrogen-estrogen receptor complex with the estrogen response elements

(ERE) suspected to lie upstream of the Vtg I and Vtg II coding regions. Isolation

and characterization of the ERE from each Vtg gene should offer valuable

insights into ERE mechanics. Another explanation for the difference in amounts

of Vtg transcript may involve RNA stabilization, rather than gene transcription.

It has been shown that the half-life of Vtg transcripts increases dramatically in the

presence of estrogen (Brock and Shapiro, 1983). A recent report suggests that

an estrogen inducible protein that binds to the 3' untranslated region of Xenopus

Vtg may be responsible for this stabilization (Dodson et al.,1995). It would be

interesting to compare protein-RNA interactions of this protein with two closely

related Vtg mRNAs, F. heteroclitus Vtgs I and II. Furthermore, our estrogen-

induced liver library would offer an excellent template to screen for such cDNAs

that might code for this protein.

By completing the sequences of two Vtgs, we now have the basic

information and tools to molecularly dissect the process of yolk formation in F.

heteroclitus. We may begin to answer questions about the functional significance

of possessing multiple Vtgs. Antibodies produced against non-conserved regions

of the two Vtgs may indicate differences in receptor mediated endocytosis,

compartmentalization, or catabolism by the embryo. Cycling controls involving

Vtg may also be studied; for instance, Vtg cDNA probes can be used to document









69

the fine-tuned expression of Vtg that must occur in a sequentially spawning

animal. Besides being used as tools to specifically investigate F. heteroclitus

reproduction, the Vtg I and II cDNAs represent valuable bio-markers for assaying

the reproductive health of naturally occurring fish. As examples of mRNAs and

proteins that are normally induced by estrogens, Vtg I and II will be particularly

valuable in testing for the estrogenic effects of environmental contaminants such

as polychlorinated biphenyls (Bergeron et al. 1994; Guillette et al. 1994).














CHAPTER 4
PRECURSOR-PRODUCT RELATIONSHIP OF
VITELLOGENINS I AND II TO THE YOLK PROTEINS
OF FUNDULUS HETEROCLITUS



Introduction


Current views concerning the origin and processing of yolk proteins in oviparous

vertebrates were formed through a slow, and controversial suite of biochemical studies

that eventually elucidated two unexpected aspects concerning the origin of yolk proteins

(reviews by Wallace 1978, 1985; Eckelbarger 1994). First, it was shown that yolk

proteins originated "hetero-synthetically" in the liver, rather than the ovary. Secondly,

it was shown that yolk proteins were not synthesized individually, but rather as a large

protein precursor, that was subsequently processed into bona fide yolk proteins. This

yolk protein precursor, vitellogenin (Vtg), has now been documented to appear in the

blood of estrogen-treated males or spawning females from countless oviparous vertebrates

(Wallace and Jared, 1969). Additionally, Vtg has been documented to be incorporated

into growing oocytes by receptor-mediated endocytosis (Wallace and Jared, 1969b;

Opresko et al., 1980; Stifani et al., 1990; Shen et al., 1993; Shibata et al., 1993), and

processed into yolk proteins (Wallace and Jared, 1969). Although isotopic and

immunologic tracking studies have established the connection between yolk proteins and

Vtg, direct sequence data, mapping the precursor-product relationship of Vtg to the

70









71

derived yolk proteins, have been scarce (Clark, 1973; Bergink and Wallace, 1974; Byrne

et al., 1984; Gerber-Huber et al., 1987; Wallace et al., 1990b; Yamamura et al., 1995)

and especially lacking from teleosts (Matsubaro and Sawano, 1995). Recent studies

focusing on Vtg genes and cDNAs have documented that many animals possess multiple

Vtg genes and proteins (Wahli et al., 1979; Blumenthal et al., 1984; review by Byrne

et al., 1989) offering an even more challenging puzzle to workers seeking to map these

relationships.

Obtaining a clear synopsis of precursor-product relationships in many teleosts, is

further complicated by the extensive yolk protein processing that occurs in teleost yolk

as compared to the yolk of tetrapods. The most striking difference in yolk content

documented in F. heteroclitus concerns the disappearance of a 125-kDa yolk protein (YP

125), and the concomitant appearance of smaller yolk protein bands immediately prior

to oocyte ovulation (Wallace and Begovac, 1985; Wallace and Selman, 1985; Greeley

et al., 1986). This enhanced proteolytic processing may be connected to a unique pre-

ovulatory process that occurs in some teleost oocytes, termed hydration. Near the time

of germinal vesicle breakdown, a rapid increase in oocyte volume occurs, usually

attributed to the uptake of water (Fulton 1898; reviewed in Selman and Wallace, 1989).

In F. heteroclitus, a substrate spawner, post-maturational oocytes possess twice the

volume of pre-maturational oocytes (Wallace and Selman, 1985, Greeley et al., 1991;

McPherson et al.), but in the oocytes of pelagic spawners, oocyte volumes can increase

over four times the original volume, in as little as twelve hours (Wallace and Selman,

1981; Watanabe and Kuo, 1986; Craik and Harvey, 1987; LaFleur and Thomas, 1991).








72

Several possible factors have been hypothesized to drive hydration, ranging from the

osmotic balance of ions (Hirose, 1976; Watanabe and Kuo, 1986; LaFleur and Thomas,

1991; Greeley et al., 1991; Wallace et al., 1992), ionic balance via gap junction control

(Cerdi et al., 1993), and the colligative osmotic contribution of cleavage peptides and

free amino acids (Oshiro and Hibiya, 1981; Wallace and Selman, 1985; Greeley et al.,

1987; Thorsen et al. 1993). With these issues in mind we sought to characterize the

precursor-product relationship between Vtg and yolk proteins in F. heteroclitus, with

emphasis on the processing of YP 125. By completing the cDNA and putative protein

sequences of two F. heteroclitus Vtgs (LaFleur et al., 1996; chapter 3), we obtained the

necessary blueprint for comparison of microsequencing data. In this paper we document

internal and N-terminal amino acid sequences from seven isolated yolk proteins, all of

which can be positioned within the Vtg I and Vtg II predicted protein sequences. Our

data suggest that the majority of yolk proteins are derived from Vtg I, and that a small

amount are derived from Vtg II. Additionally, we suggest that the rapid processing of

YP 125 during hydration is associated with the presence of a PEST site (Rogers et al.

1986) near its predicted C-terminus.



Materials and Methods


Ovarian follicles were dissected from the ovaries of reproductively active F.

heteroclitus. Up to 20 prematurational follicles or up to 10 ovulated eggs were aliquoted

into a 1.5 ml eppendorf tube containing 500-750 /l of sample buffer (0.1 M Tris, pH

6.8, 2% SDS, 64 Mm dithiothreitol, 10% glycerol) on ice. The follicles were









73
immediately ground with a Kontes pestle and heated for 10 min at 100 oC. The

homogenate was then briefly centrifuged at 12,000 g for 1 min., separating the dissolved

yolk from insoluble cellular debris. The supernatant was aliquoted to a fresh tube and

stored at -200C until electrophoresis. Samples were diluted again by as much as 1:50

with sample buffer before loading onto gels.

Sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) was

carried out according to Laemmli (1970), using 125 X 140 X 1.5 mm slab gels

containing a 3.5% stacking gel overlaying a separating gels ranging from 7% for larger

YPs to 12% for smaller YPs, with modifications based on the protocol of Schagger and

von Jagow (1987) using Tris-tricine running buffers.

Proteins in electrophorese gels were electroblotted onto PVDF membranes in

buffer containing 10 Mm MES, Ph 6, and 20% methanol at 20 V overnight. Protein

bands were visualized by brief staining in 0.02% Coomassie blue in 40% methanol plus

5% acetic acid, destained in 40% methanol plus 5% acetic acid, and rinsed in distilled

water. Membranes were dried and stored at -20 oC until individual bands were cut out

and submitted for sequencing. N-terminal amino acid analyses were performed on PVDF

bound proteins using an Applied Biosystems Model 473a Sequencer (LeGendre and

Matsudaira, 1988) by the Protein Chemistry Core Facility of the University of Florida.

The two largest N-terminally blocked yolk proteins (YP 125 and YP 105) were

again electrophoresed, blotted onto PVDF and subjected to in situ cleavage (Scott et al.,

1988) by endoproteinase LysC (Endo LysC)(0.003 units/jg protein, Promega), in 50 mM

Tris, Ph 8.8, 0.2M ammonium bicarbonate, and 0.1% SDS, 0.1 Mm EDTA. Protein








74



Oocyte Egg


YP 125 -
YP 105-
YP 83
YP 80 YP
YP 77



00
-: YP 45 b YP39
O" d YP 39




YP 29


YP 20 -







Figure 4.1 Major yolk proteins isolated from oocytes and eggs of Fundulus
heteroclitus. The major yolk proteins shown here were resolved by an
SDS-PAGE gradient gel (7%-20%) enabling the resolution of a wide
range of proteins ranging from 125 kDa to 20 kDa. For N-terminal
sequencing, however, straight gels were used at various acrylamide
concentrations allowing optimal resolution of yolk proteins at specific size
classes. Yolk proteins that were isolated for N-terminal sequencing are
indicated with our designated labels. Note YP 125 appears as a robust
band, when isolated from pre-maturational oocytes, but is hardly visible
in yolk isolated from ovulated eggs. (Photo courtesy of R. McPherson,
Clarion University)








75

















Mw, kDa


21.5>




13 kDa -- 13 kDa
LFESLVDSDKVV... LFESLVDSDKVV ...
YEFSDELLQTPL... YEFSDELLQTPL...
KYxAKHIGVGLK... .
6.5--












Figure 4.2 Endo LysC digestion products of YP 125 and YP 105. After partial
digestion with Endo LysC, polypeptide fragments were electroblotted onto
a PVDF membrane and silver stained. Positions of the 13 kDa bands
(presumed to be identical) from each digestion are indicated. Molecular
weight standards (kDa) are shown on the left.









76

fragments were then separated by Tris-tricine gels, blotted to PVDF, and visualized by

silver staining (Wray et al., 1981). Similar bands of 13 kDa (presumed to be identical)

were isolated from both the YP 125, and YP 105 digestion and once again submitted to

the Protein Chemistry Core for N-terminal amino acid sequencing.

Sequencing data, including Vtg I and II cDNAs, along with microsequencing

results were organized using the PC/GENE software package (Intelligenetics, Mountain

View, CA). Prediction of signal peptides was carried out according to von Heijne

(1986). PEST sites were designated according to the algorithm described by Rogers et

al. (1986). Other Vtg sequences referred to in this paper include chicken Vtg II

(gi:63887; van het Schip et al. 1987), Xenopus laevis Vtg A2 (gi: 139636, Gerber-Huber

et al. 1987), lamprey, Ichthyomyzon unicuspus Vtg (gi:213312, Sharrock et. al. 1992)

and sturgeon, Acipenser transmontanus Vtg (gi:437051, Bidwell and Carlson, 1995).


Results


The yolk proteins typically found in F. heteroclitus oocytes and eggs are

demonstrated in Figure 4.1, along with our designations of certain bands according

to their apparent molecular mass. At least nine yolk proteins were resolved by Tris-

tricine SDS-PAGE, and these were blotted onto PVDF membranes and submitted for

protein sequencing by Edman degradation. Four yolk proteins appeared to be N-

terminally blocked, while five yielded N-terminal sequences (Table 4.1).

By aligning the yolk protein N-terminal sequences against the predicted amino

acid sequences of Vtg I and Vtg II, we successfully mapped the five sequenced yolk










77






Table 4.1. N-Terminal Sequences of F. heteroclitus Yolk Proteins

Protein Source N-Terminal Sequence Source n


YP 125 Oocyte Blocked ? 3

YP 105 Oocyte/egg Blocked ? 2

YP 83 Egg Blocked ? 1

YP 80 Oocyte/egg Blocked ? 1

YP 77 Egg Blocked ? 1

YP 69 Egg NCVSY APEFA PGCTY EYKYE Vtg II 1

YP 45 Oocyte HIKMV AxGxx A Vtg I 2

YP 39 Egg EEEAV VAVIL RAVKA D Vtg I 2

YP 29 Oocyte AAAAE xSFVE DTLYT FN Vtg I 1

YP 20 Oocyte EEDVE PIPEY KFRRF AKCY.C Vtg I 2


ELC 13 YP 125 YEFSD ELLQT PLQLI KISD Vtg I 1

ELC 13 YP 125 LFESL VDSDK VVENP LLREV Vtg I 1

ELC 13 YP 125 KYCAK HIGVG LKACF KFASQ Vtg I 1

ELC 13 YP 105 YEFSD ELLQT PLQLI KISD Vtg I 1

ELC 13 YP 105 LFESL VDSDK VVENP LLREV Vtg I 1


Mapped to Vtg I (982-1001), C-terminal to a PEST site

ELC denotes N-terminus of products cleaved with Endo Lys C (.003 units/ug protein)









78

protein products at internal positions within their respective precursors (Fig. 4.2). Of

these five sequences, the most notable was that of YP 69, lining up to the N-terminus of

Vtg II, verifying the expression of this secondary Vtg as well as demonstrating that the

signal peptide cleavage site had been correctly predicted. The data from YP 69 also

indicate that the N-terminus in Vtg II is unblocked in contrast to the apparently blocked

N-terminus of Vtg I.

In order to identify the origin of YP 105 and YP 125, the protein bands were

again blotted onto PVDF membranes and proteolytically cleaved with Endo Lys C

(0.003 units/lsg protein, Promega) in 50 Mm Tris, Ph 8.8, 0.2 M ammonium

bicarbonate, and 0.1% SDS, 0.1 Mm EDTA. The digestion products were again

separated by Tris-tricine gels, and visualized by silver staining. The reaction with Endo

LysC was confirmed as only a partial digestion by the isolation of peptide products larger

than those predicted if cleavage had occurred at every lysine residue. The pattern of

electrophoresed digestion products from YP 125 and YP 105 initially appeared to be

identical, indicating that the two yolk proteins originated from the same precursor

molecule (Fig. 4.3). However, a difference between the digestion products was

discovered when the 13-Kda peptides derived from YP 125 and YP 105 were sequenced.

The 13-Kda band isolated from YP 105 digestion contained two peptides, mapping near

the N-terminal region of Vtg I. The 13-Kda band isolated from YP 125 contained the

exact two peptides found in the YP 105 digestion, plus a third peptide (K Y C A K H

I G V G L K A C F K F A S Q), that mapped much further along the Vtg I sequence,

to residue 982 (Figs. 4.4 and 4.5). We interpret these data as evidence that YP 105 and








79




Vtg I Vtg 11
predicted 188 kD predicted 185 kD
signci g snci
NQVSYAPEFAPG
YP 69













SEEDVEPIPEYKF
YP18

Polyserine Poyserine
domain domain
HKKMVAxG
YP 45 EEEAWAV1LRA
YP 39

I AAAAExSFVEDT
YP 29








Figure 4.3 A graphical representation of F. heteroclitus yolk proteins positioned
along the length of the Vtg I and Vtg II. Length and positions along the
Vtg molecules are drawn to scale according to alignments of N-termini
data with cDNA translations. C-termini of yolk proteins were calculated
according to molecular weight estimations and should be regarded as
putative. The signal peptides and polyserine domains as predicted from
cDNA translations are indicated.









80







Vtg 188 kD
signal
YP 105 YP125


ELC13-- -I-ELC 13

ELC 13 1 ELC 13







PESTI
P ELC 13


Polyserine
domain



200 ca




C-TERMINUS









Figure 4.4 A graphic representation of the 13 kDa digestion products and their
positions in reference to YP 125, YP 105 and Vtg I. Note that the third
digestion product of YP 125 lies beyond the calculated C-terminus of YP
105. The indicated PEST site was found in YP 125, but is truncated, and
thus invalidated in YP 105.









81

YP 125 are identical Vtg I-derived yolk proteins except for a short 20 Kda extension at

the C-terminus of YP 125 that contains the third Endo LysC digestion product (Fig. 4.3).

The C-terminus of YP 105 was predicted to lie at (or before) residue Ser 962 of

Vtg I, using the estimated mass of YP 105 and the masses of the individual residues

predicted from the Vtg I cDNA. This places the YP 105 C-terminus only 2 residues

away from the N-terminal residue obtained from YP 20 (Glu 965) suggesting that YP 105

and YP 20 result from cleavage of YP 125. The estimated juncture between YP 105 and

YP 20 lies at the exact midpoint of a predicted PEST site (residues 952-974, receiving

a score of 6.9, where 5.0 and above is considered a site). This purported cleavage site

bisects the predicted PEST site, leaving the two resulting protein sequences with termini

that do not surpass the cutoff value for valid PEST sites. Thus, although YP 125

contains a PEST site, neither of its cleavage products, YP 105, nor YP 20 do.


Discussion


We have presented precursor-product relationships to account for the origin of

seven yolk proteins isolated from oocytes and eggs of F. heteroclitus. Likewise, the

sequences determined from these yolk proteins verify the expression, transport, and

incorporation of both the yolk protein precursors Vtg I and Vtg II, whose cDNA

sequences are provided in Chapters 2 and 3, respectively.

We had initially assumed that YP 125 and YP 105, the major bands in oocyte

extracts, were derived separately from Vtg I and II, but the internal sequences indicated

that both yolk proteins originate from Vtg I. We can thus surmise that Vtg I is truly the









Figure 4.5 A summary of the precursor-product relationship of Vtg I to
derived yolk proteins. The entire translated amino acid sequence
of the Vtg I cDNA sequence (LaFleur et al., 1995) is presented,
separated into sections representing yolk proteins as indicated by
brackets on the right. N-terminal sequences of isolated yolk
proteins are indicated by double underlining. Internal sequences
obtained from Endo LysC digestion products are indicated by
shaded lettering. The residues of the PEST site are represented by
bold face lettering. The predicted polyserine domain (no N-
terminal sequencing data) is shown in brackets.













83






1 MKAVVLALTLA FVAGQNFAPE FAAGKTYVY
31 KYEALI L G GLPEGLARAGLKI STKLLLSA
61AD Q NTYMLKLVE PELS Y S GI W P K DPAVPA
91 TKLTAALH LSSQ F PSS LNT PMV F V GKV FAP
121 EE V S T LV L N I Y R G I LN I L Q L N I K K T H K VY
151 L Q V GT Q G VC KT L Y S I S E D A R I E N I L L T K T
181 R D L S N C QE R LN K D I G LA T K C K C Q EE T K
211 NL R G T T T L S Y VL K P VA D A V MI L KA Y V NELI
241 Q FS P F S E A N G A A Q M R T K QS LE F L I E KE P I
271 PS V KAE YR H RGS L K. f k 4 LO> L:8 TL ~
301 i1 DA P A Q V A E V L K H L A T Y N I E D V H E N A P L K
331 FL L V Q L L I AR Y E D L MY W N Q Y KKM S P H R
361 HW F LD T I PAT GT FAG L F I K E K FMAE E I T I
391 AE A A Q A F I T A V H M V T A D P E V I K
421 DK Vi~ PL R. V V~ G T M V N KY C N K TV D
451 C PVE L I K P I QQ R L S DA I A K N E E E N I I LY I K
481 VL G N A G H P S S F K S L T K I P I H G T A A V S LP M
511TI HVEAIMALRNIAKKE S RMVQE LAL Q LY M YP 105
541 D KALH P E L RML S C I V L FE T S PS MG V TV A
571 NS V K TE E N L Q V A S F T Y S HM K S L S R S PA T I H
601 PD V A A A C S A A M K I L G T K L D R L S L R Y S K A V H YP 125
631 V D L Y N S S L A V G A A A T A F Y I N D A A T FM P K S F
661 V A K T K G F IAG S TAE V L E I GA N I E G L Q L I L
691 KN P A L S E S T D R I T K M K V I K A LS E W S L P T
721 SK PLAS V YV K FF G Q E I G FAN I D K PMIDKAV
751 K F GK E L P I Q E Y GRE AL K AL L LS G I N F H YAK
781 P V L A A EMRR I L P T V A G I P M E L S LY S A A V A A
811 AS VE I K PNT S PR LSAD F D V K T L LETD VE LK
841 AE I R PM V A M D T Y A V M G L N T D I F Q A A L V A R A
871 K LH S V V P A K I A A R L N I KE G D F K LE A L PV D V
901 PEN I T S M NV T T F A V A RN I E E P LV E I T P LL
931 P T KV LV P I P I R R H T S K L D PT R
952 N SMLD S SZ L L PME PE
T Psite


991 B E.FD V P I P E N IQ D IVLY K LAGSH N F
1021 S F S V TP IE G EV V E R LE M E V K V G A K A A E K L V 20
1051 K R I N LS E D E E T E E G G P V L V KL N K I LS S R R N
1081 SS S S S S S S S S S S E S R S R S S S S S S S S S R S
1111 S R K I D L A A R T N S S S S S S S R RS R S S S S S polyserine
1141 SSSSS S S S S S RR S S S S S S S S S S S S S S domain
1171 R R V N S T R S S SSS S S R T S S AS S A S F F S D S S S
1201 S S S S S D R R S K E VM E K F Q R L


1220 H K K M V A S GS SA S S VEAI Y K E K KY L GE
YP 45

1246 E E A V V A V I L R A V K A
1261 K RMV G Y Q L G F Y LD K PNARVQ I I V A N I S S DS YP 39
1291 NW R I CAD A V V L SK H K VT T K I SW GE Q C R KY S
1321 T N V T GE TG I VS S S PAAR L R V S W ER L PS T L K
1351 R Y G KMV N K Y V P V K I L S D L I HT K R E N S T R N I
1381 S V I A V A T SE KT I I I T K T PMS S V Y N V T H L
1411 PMCI P I DE I KG L S P F DE VI D K I H FMVS K


1439 AAAAECSFV EDT L Y T F N NR S Y KN KMPSS C Y
1469 QVAA Q DCTDELK FMVLLRKDSS EQ H HI NVK YP29
1599 I S E I D I DM F P K D D N V T V K V N M I P P P A C L
1529 TAT QQ L P L K I K T K G LA V Y A P H GL Q E V Y
1559 F DRK T WR I K V A D W M KG K T C G L C G KAD G E I R
1589 Q E Y HT P N G R V A K N S IS F A H SWI L PAE S CR D
1619 AS E C R L K L E S V Q L E K Q LT I H G E D S T C FS V E
1649 PVPR C L PGC L PV K TTPVTVG FS C LASDPQT
1679 S VY D R S VD L R Q TT Q A H LA C S CNT K CS









84

major yolk protein precursor in F. heteroclitus. Our finding that most of the yolk protein

is derived from Vtg I agrees with northern blot analyses that suggested ten times more

Vtg I than Vtg II message is present in total liver RNA (Chapter 3; LaFleur et al.,

1996).

A major factor that prevents construction of a definitive map accounting for all

yolk proteins derived from the Vtgs is the difficulty in isolating and microsequencing

peptides derived from the phosvitin domain (Wallace and Begovac, 1986; Wallace et al.

1990). Although we expect that the polyserine repeats represented in both Vtg I and II

cDNAs are processed into true phosvitins, we have been unable to verify this by N-

terminal sequencing. The highly negative charge of phosvitin prevents it from staining

with Coomassie blue, as well as adhering to PVDF membranes for sequencing. Because

phosvitin can be visualized using Stains-all, it has been documented as a single 25-30

kDa band in prematurational oocytes, with at least four smaller phosvitin-like bands

(phosvettes) appearing in preparations from ovulated eggs (Wallace and Begovac, 1985).

We estimate that the C-terminus of YP 20 (and presumably, the C-terminus of YP 125)

lies adjacent to the N-terminus of phosvitin, as predicted by the position of the Vtg I

cDNA polyserine repeating region. Likewise, the sequence obtained from YP 45,

sharing identity with residues 1220-1230 of Vtg I, most likely abuts the C-terminal

cleavage site of phosvitin.

As previously mentioned, one of the most pronounced changes observed to occur

in F. heteroclitus yolk proteins is the disappearance of YP 125 during the transformation

of oocytes to mature, ovulated eggs (Fig. 4.1). A possible explanation for this rapid and









85

rather selective proteolysis is the occurrence of a PEST site within the C-terminal tail of

YP 125. The apparently longer-lived YP 105 is identical to YP 125 except for lacking

the C-terminal tail where the PEST site occurs. PEST sites were initially defined as a

conserved clustering of amino acids that was observed to occur in proteins known to be

rapidly degraded. Common to all PEST site are high local concentrations of Pro, Glu,

Ser, and Thr, and to a lesser extent Asp. Of the other five vertebrate Vtg sequences

contained in Genbank, chicken Vtg I (residues 1058-1080 and 931-951) and lamprey Vtg

(residues 1161-1182 and 1360-1393) contain two PEST sites, while Xenopus Vtg A2

contains a sequence (residues 953-969) with a score (4.71) very close to the cutoff value

of 5. The lack of proteolysis during oocyte maturation in such animals may indicate

either the absence of an appropriate proteolytic mechanism or the inaccessibility of the

cleavage sites in the granular yolk of these animals (Wallace, 1985). The Vtg of

sturgeon, a chondrostean fish, does not contain a PEST site.

The proteolytic processing of YP 125 has been implicated as part of the hydration

mechanism of F. heteroclitus oocytes, with the generated small peptides and free amino

acids providing the osmotic potential to drive an uptake of water into the oocyte (Wallace

and Begovac, 1985; Wallace and Selman, 1985). More recent data suggest that

hydration in F. heteroclitus is primarily due to K' fluxes via the gap junctions between

oocytes and follicle cells (Wallace et al., 1992; Cerdi et al., 1993), but the possibility

of some contributions to hydration resulting from yolk cleavage has not yet been

abandoned. So far, complete Vtg sequences have been reported from no other teleosts

besides F. heteroclitus. However, as more sequences are completed, it will be









86

interesting to see whether PEST sites are found in other teleostean Vtgs, especially those

of pelagic spawners in which both oocyte hydration and yolk proteolysis are especially

pronounced.















CHAPTER 5
FUNDULUS HETEROCLITUS CHORIOGENINS: LIVER-DERIVED
COMPONENTS OF THE VITELLINE ENVELOPE AND CHORION
SHARING SEQUENCE IDENTITY WITH MAMMALIAN ZP PROTEINS



Introduction


The spawned eggs of the estuarine teleost Fundulus heteroclitus are exposed to

quite a different environment than the ovulated eggs of mammals. Whereas mammalian

eggs are protected from infection, desiccation, and predation by the safe surroundings

of the uterus, F. heteroclitus eggs are released and fertilized during the tumultuous spring

tides, and deposited into empty mussel shells or onto the leaves of marsh grass, where

they remain actually stranded above the water line for fourteen days until the embryos

emerge by hatching during the next spring tide (Taylor et al., 1977; Hsiao et al., 1994).

Though exposed to extremely different environments, both of these vertebrate eggs are

protected by a quasi-similar layer of extracellular matrix (ECM). In mammals this

translucent layer of ECM is termed the zona pellucida (ZP), but in fish and many other

invertebrates it is often referred to as the vitelline envelope or chorion.

In this paper we adhere to the definitions of Dumont and Brummett (1980)

regarding the vitelline envelope and chorion. They stated that the term "vitelline

envelope" referred to the highly structured acellular layer that appears and encloses the



87









88

teleost oocyte during its development, while the term "chorion" referred to the

structurally and perhaps chemically transformed vitelline envelope that surrounds the

ovulated egg, separates from the egg at the time of fertilization, and encloses the embryo

until hatching. Implicit in these definitions is the assumption that the proteinaceous

structure of the vitelline envelope comprises a substantial component of the chorion.

The structure of the teleostean vitelline envelope has been well documented in

several cyprinodont species (Yamamoto, 1963; Fliigel, 1967; Dumont and Brummett,

1980) as well as in many other teleosts (reviewed by Dumont and Brummett, 1985;

Selman and Wallace, 1989). Early biochemical characterizations of the vitelline envelope

and chorion concentrated on the formation of the vitelline envelope during oocyte

development (Chaudry, 1956; Yamamoto, 1963; Flegler, 1977; Tesoriero, 1977), as well

as the breakdown of the chorion by the proteolytic enzymes of the hatching embryo

(Yamamoto and Yamagami, 1975; Kaighn, 1964, Hagenmaier, 1985). In earlier works

it had been assumed, but not proven that the major vitelline envelope proteins (VEPs)

were synthesized by the ovarian follicle -- the site of synthesis residing in either the

oocyte or surrounding follicle cells (Anderson, 1967). More recent investigations

targeting VEP synthesis include studies by Tesoriero (1978) using [3H]proline

incorporation, and by Begovac and Wallace (1989) in which incorporation of

[3S]methionine combined with immunohistochemistry provided evidence that at least one

of the VEPs from the pipefish, Syngnathus scovelli, originated from within the ovarian

follicle.

A new direction towards understanding vitelline envelope formation was launched








89

by research concentrating on the chemistry of hatching enzymes in the medaka, Oryzias

latipes. When polyclonal antibodies directed against protein fragments of the lysed

chorion were used as probes on medaka tissues, Hamazaki et al. (1984) found that tissues

other than the ovary were recognized by the antibody. By 1989 Hamazaki et al. (1989a)

had isolated an estrogen-induced glycoprotein from the liver that could be localized to

the inner layer of the vitelline envelope. Since then additional reports have verified these

findings in several other fish (Hyllner et al., 1991; Murata et al., 1991; Oppen-Berntsen

et al., 1992a, 1992b; Larsson et al., 1994). Additionally, Hyllner et al. (1991) showed

that the synthesis of VEPs could be induced by estrogen treatment in males of the

rainbow trout (Oncorhynchus mykiss), brown trout (Salmo trutta), and turbot

(Scophthalmus maximus), providing convincing evidence that the major VEPs in these

species could be synthesized without any contribution by the ovary.

So far, only two nucleotide and protein sequences representing piscine VEPs have

been published. Lyons et al. (1993) reported a gene sequence (wf) from the flounder,

Pseudopleuronectes americanus, that they described as a "teleostean homolog of a

mammalian ZP gene." The predicted amino acid sequence contained a novel PQQ

repeating region near the N-terminus, resembling a motif found in extracellular matrix

proteins. Murata et al. (1995) reported a cDNA sequence (L-SF) from medaka that also

shared identity with mammalian ZP proteins. The predicted amino acid sequence of the

medaka L-SF protein shared more identity with mouse ZP3 (37.9%; Ringuette et al.

1988) than it did with the flounder ZP sequence (18%) previously mentioned, suggesting

the presence of at least two distinct groups of teleost ZP homologs.









90

In this paper we present the predicted primary structure of three proteins that

share identity with mammalian ZP proteins. Furthermore, we have isolated cDNAs

encoding these sequences from a liver library rather than an ovarian library, followed by

northern analyses revealing liver rather than ovarian transcripts. Lastly, the amino acid

compositions predicted from our cDNAs are similar to the composition of VEPs isolated

from F. heteroclitus follicles. Therefore, we conclude that the proteins encoded by these

cDNAs are synthesized by the liver, transported to the ovary, and incorporated into the

vitelline envelope. We further suggest that as major constituents of the vitelline

envelope, these proteins eventually contribute to the structure of the hardened chorion,

where they remain until finally degraded by embryonic hatching enzymes. We designate

these cDNAs and the proteins that they encode as "choriogenins" (Chgs) to emphasize

their role as proteins of the vitelline envelope and chorion, yet to underscore their site

of synthesis as being extra-ovarian, and thus different from that of the mammalian ZP

proteins. Although the teleostean chorion and mammalian zona pellucida have different

appearances, functions, and, as this study verifies, origins of synthesis, we provide

evidence that the constituent molecules appear to have evolved from a set of common

ancestral proteins


Materials and Methods


Reagents


Estradiol-170 was obtained from Sigma Chemical Co. (St. Louis, MO).

Radioisotopes, [ca-32P]dCTP and [a-"S]dATP, were purchased from New England









91

Nuclear (Boston, MA). Lambda gtlO vector and cDNA synthesis reagents were obtained

from Promega (Madison, WI). The subcloning plasmid pGem-T was a product of

Promega. All "ROW" oligonucleotide primers were synthesized by the University of

Florida Interdisciplinary Center for Biotechnology Research (ICBR) oligonucleotide core

facility, while primers labelled "GL" were synthesized by Bio-Synthesis (Freindswood,

TX). Sequenase version 2.0 DNA polymerase and dideoxy sequencing reagents were

obtained from US Biochemicals (Cleveland, OH). In-house sequencing gels were cast

using Sequagel-8 (National Diagnostics, Atlanta) polyacrylamide reagents. Some cDNA

sequences, especially through repeating regions or when verifications were needed, were

performed by The University of Florida ICBR DNA Sequencing Core. Amplification

reactions were performed using a 1:50 mixture of cloned pfu DNA polymerase and

Thermophilus aquaticus DNA polymerase (Stratagene and Promega, respectively).

Reagents for random-primed labeling of probes were purchased from Pharmacia

(Piscataway, NJ). Magna nylon and PVDF transfer membranes were obtained from MSI

(Westboro, MA) and Millipore Corp. (Bedford, MA), respectively.


Cloning Strategy


A liver cDNA library was constructed from poly A-RNA pooled from five F.

heteroclitus males that had been treated with two injections of estradiol-173, as

previously described (LaFleur et al., 1995). While screening the Xgt 10 library for Vtg

cDNAs using anchored PCR, we isolated several non-target cDNAs. Three of these non-

Vtg cDNAs were revealed by BLAST analysis to code for protein sequences that









92










I Chg 500
pChg1a ROW 45
ROW 52 pChg1b


Chg 427

pChg2a ROW 55
ROW 65 pChg2b


|I Chg 553
ROW 45
pChg3a -
GL1 pChg3b
200 bp














Figure 5.1 Strategy for cloning Chg 500, 427 and 553 cDNAs. Boxes indicate
relative sizes of contiguous cDNA sequences coding for Chgs 500, 427,
and 553. Thin black lines represent individual cDNA isolates obtained by
anchored PCR or RACE and inserted into pGem-T. Arrows indicate
gene-specific primers that were used in initial amplifications of individual
clones. The legend indicates relative length of 200 bp.




Full Text
Figure 5.2 Nucleotide and conceptually translated amino acid sequences of F.
heteroclitus choriogenins.
A) The Chg 500 cDNA (1641 bp) codes for a 500 amino acid
protein sequence, containing a predicted signal peptide (von
Heijne, 1986) from residues 1-22, indicated by shading. The poly-
adenylation signal is indicated by underlining (beginning at
nucleotide 1607).
B) The Chg 427 cDNA (1672 bp) codes for a 427 amino acid
protein sequence, containing a predicted signal peptide from
residues 1-22 indicated by shading. A poly-adenylation signal is
represented by underlining (beginning at nucleotide 1637).
C)The Chg 553 cDNA (1816 bp) codes for a 553 amino acid
protein sequence, containing a predicted signal peptide from
residues 1-25 indicated by shading. A poly-adenylation signal is
represented by underlining (beginning at nucleotide 1767).


130
Harris JD, Hibler DW, Fontenot GK, Hsu KT, Yurewicz EC, Sacco AG (1994) Cloning
and characterizaion of zona pellucida genes and cDNAs from a variety of
mammalian species: The ZPA, ZPB and ZPC gene families. DNA Sequence
4:361-393
Hart NF (1990) Fertilization in teleost fishes: mechanism of sperm-egg interactions. Int
Rev Cytol 121:1-66
Higgins DG, Bleasby AJ, Fuchs R (1992) CLUSTAL V: improved software for multiple
sequence alignment. Comput Appl Biosci 8:189-191
Hirose K (1976) Endocrine control of ovulation in medaka (Oryzias latipes) and ayu
(Plecoglossus altivelis). J Fish Res Bd Can 33:989-994
Hopp TK, Woods KR (1981) Prediction of protein antigenic determinants from amino
acid sequences. Proc Natl Acad Sci USA 78:3824-3828.
Hovemann B, Galler R, Walldorf U, Kupper H, Bautz EK (1981) Vitellogenin in
Drosophila melanogaster: sequence of the yolk protein I gene and its flanking
regions. Nucleic Acids Res 9:4721-4734
Hsiao S-M, Greeley MS Jr., Wallace RA (1994) Reproductive cycling in female
Fundulus heteroclitus. Biol Bull 186:271-284
The Huntingtons Disease Collaborative Research Group (1993) A novel gene containing
a trinucleotide repeat that is expanded and unstable on Huntingtons disease
chromosomes. Cell 72: 971-983
Hyllner SJ, Oppen-Bemtsen DO, Helvik JV, Walther BT, Haux C (1991) Oestradiol-17/3
induces the major vitelline envelope proteins in both sexes in telosts. J Endocrinol
131:229-336
Hyllner SJ, Barber HF-P, Larsson DGJ, Haux C (1995) Amino acid composition and
endocrine control of vitelline envelope proteins in European sea bass
(Dicentrarchus labrax) and gilthead sea bream (Spams aurata) Mol Repro Dev
41:339-347
Iuchi I, Yamagami K (1976) Major glycoproteins solubilized from the teleostean egg
membrane by the action of the hatching enzyme. Biochim Biophys Acta 453: 240-
249
Janin J (1979) Surface and inside volumes in globular proteins. Nature 277:491-492


110
Pre-Hyb Pre-Hyb Pre-Hyb
Chg 500 Chg 427 Chg 553
28s
18s
B
kb
4.4 >
2.37)
1.35 =
Chg 500 Chg 427
Chg 553
i ii i
a b c d e f
i i
g h i


68
derived yolk proteins, YP 125 and YP 105 (Chapter 4), agreeing well with the
mRNA expression data. This may suggest a difference in the interaction between
the estrogen-estrogen receptor complex with the estrogen response elements
(ERE) suspected to lie upstream of the Vtg I and Vtg II coding regions. Isolation
and characterization of the ERE from each Vtg gene should offer valuable
insights into ERE mechanics. Another explanation for the difference in amounts
of Vtg transcript may involve RNA stabilization, rather than gene transcription.
It has been shown that the half-life of Vtg transcripts increases dramatically in the
presence of estrogen (Brock and Shapiro, 1983). A recent report suggests that
an estrogen inducible protein that binds to the 3 untranslated region of Xenopus
Vtg may be responsible for this stabilization (Dodson et al.,1995). It would be
interesting to compare protein-RNA interactions of this protein with two closely
related Vtg mRNAs, F. heteroclitus Vtgs I and II. Furthermore, our estrogen-
induced liver library would offer an excellent template to screen for such cDNAs
that might code for this protein.
By completing the sequences of two Vtgs, we now have the basic
information and tools to molecularly dissect the process of yolk formation in F.
heteroclitus. We may begin to answer questions about the functional significance
of possessing multiple Vtgs. Antibodies produced against non-conserved regions
of the two Vtgs may indicate differences in receptor mediated endocytosis,
compartmentalization, or catabolism by the embryo. Cycling controls involving
Vtg may also be studied; for instance, Vtg cDNA probes can be used to document


The N-terminus of a 69 kDa yolk protein matched the predicted N-terminus of Vtg II
(minus a signal peptide), verifying that Vtg II is expressed without being N-terminally
blocked. Six other yolk proteins were mapped to the predicted Vtg I sequence,
confirming that Vtg I represents the major yolk protein precursor. A 125-kDa yolk
protein that is specifically degraded during final maturation was mapped to a region of
the Vtg I sequence that contained a PEST site, suggesting an explanation for its
preferential break-down.
The three choriogenins were referred to as Chg 500, Chg 427, and Chg 553,
according to the number of amino acids predicted for each protein. Chg 500 and 553
were found to be 58% identical to a flounder "zp gene product", and 30% identical with
the mouse ZP1 protein. Chgs 500 and 553 contain proline-glutamine-rich repeating
regions that resemble a PXX motif reported in other extracellular matrix proteins. Chg
427 was found to be 67% identical to a medaka "L-SF protein" and 30% identitical to
the mouse ZP3 protein that has been implicated as the primary sperm receptor. Besides
reporting the sequences of five hepatically-derived proteins that contribute to the
development of the ovarian follicle, we emphasize that the estrogen-induced library is an
excellent strategy to screen for reproductively significant cDNAs.
Vlll


Figure 2.4 Alignment of the putative F. heteroclitus Vtg sequence (gi:459202)
with other vertebrate Vtgs: the chicken Gallus domesticus Vtg II
(van het Schip et al., 1987); Xenopus laevis Vtg A2 (Gerber-Huber
et al., 1987); the white sturgeon Acipenser transmontanus Vtg
(Bidwell and Carlson, 1995); the silver lamprey Ichthyomyzon
unicuspis Vtg (Sharrock et al., 1992); and the C-termini from the
rainbow trout Oncorhynchus mykiss Vtg (LeGuellec et al., 1988)
and the tilapia Oreochromis aureus (Ding et al., 1990) Vtg as
constructed by ClustalV (Higgins et al., 1992) and modified by
eye. Our defined polyserine domain, which includes putative Pv
regions, is labeled and underscored with a triple dashed line.
Residues identical in at least four of the aligned sequences are
denoted by shaded lettering. Sequence gaps are represented as
dashes.


54
aacrcaccagcc 12
ATGAGGGTGCTTGTGCTGGCTCTCACTGTGGCCCTTGTGGCCGGGAACCAGGTGAGCTATGCCCCA 78
W R V t V L A.. L._ T V A £ V" A G.. N Q 7 S A P
GAATTTGCCCCTGGAAAGACCTACGAGTACAAGTATGAAGGTTATATTCTGGGTGGCCTGCCTGAG 144
EFAPGKTYE7K7EG7ILGGLPE
GAGGGCCTGGCAAAGGCTGGGGTGAAGATCCAGAGCAAAGTCTTGATCGGTGCAGCAGGTCCTGAC 210
EGLAKAG7KIQSK7LIGAAGPD
AGCTACATTCTGAAACTTGAAGACCCTGTCATCTCGGGGTACAGTGGCATTTGGCCTAAAGAGGTT 276
SilLKLEOPVISGX'SGXWPKSV
TTCCACCCTGCCACAAAGCTCACCTCAGCTCTCTCTGCTCAGCTCTTGACACCCGTCAAGTTTGAG 342
FHPATKLTSALSAQLLTP7KFE
TATGCCAACGGAGTGATCGGAAAAGTGTTCGCACCTCCAGGCATCTCTACAAATGTGCTGAATGTC 408
YANG7IGK7FAPPGISTN7LN7
TTCAGGGGACTCCTCAACATGTTTCAGATGAACATCAAGAAGACTCAGAATG7GTATGACCTGCAA 474
FRGLLNMFQMNXKKTQNVyDLQ
GAGACTGGAGTAAAAGGTGCGTGCAAGACACACTATATCCTTCATGAGGACTCCAAGGCTGATCGC 540
ETGVKGVCKTH5TILHEDSKADR
CTCCACTTGACGAAAACCACAGACCTGAATCACTGCACCGACAGCATCCACATGGATGTTGGCATG 606
LHLTKTTOLNHCTDSIHMD7GM
GCTGGTTATACGGAAAAATGTGCAGAGTGCATGGCTCGGGGAAAAACTC7TT CAGGAGCAATTTCT 672
AGKTEKCAECMARGKTLSGAIS
GTCAACTACATCATGAAGCCGTCTGCCTCTGGCACCTTGATCCTAGAGGCAACCGCCACTGAGCTT 738
VNXIMKPSASGTLILEATATEI.
CTCCAGTACTCGCCCGTCAACATTGTAAATGGAGCTG7CCAGATGGAGGCTAAGCAGACCGTGACC 304
LQYSPVNIVNGAVQMEAKQTVT
I
TTCGTGGACATCAGGAAGACCCCATTAGAGCCCCTCAAAGCAGACTATATTCCCCGTGGATCGCTC 870
FVOIRKTPLEPLKADYIPRGSL
AAGTACGAGTTAGGCACTGAATTCCTACAGACACCAATTCAGCTTCTGAGGATCACCAATG7CGAG 936
KYELGTEFLQTPIQLLRITilVE
GCTCAGATTGTTGAGTCTCTGAACAACCTAGTGAGCCTCAATATGGGCCATGCCCATGAGGATTCC 1002
AQI7ESLNNL7SLNMGHAHEDS
CCTCTGAAGTTTATTGAGCTCATCCAGCTGCTGCGTGTGGCCAAGTATGAGAGCATTGAAGCTCTC 1063
PLKFIELIQLLR7AK2ESIEAX.
TGGAG7CAGTTTAAAACCAAAATTGATCACAGGCACTGGTTGCTGAGCTCTATCCCTGCCATTGGT 1134
WSQFKTKIDHRHWLLSSIPAIG
ACTCATGTTGCTCTCAAGTTCATCAAGGAGAAGATCGTTGCTGGTGAAGTCACTGCTGCTGAGGCT 1200
TH7ALKFIKEKI7AGE7TAASA
GCTCAGGCCATCATGTCATCTACACACTTGGTGAAGGCCGACCTGGAGGCAATCAAGCTTCAGGAG 1266
AQAIMSSTHL7KA0LEAIKLQE
GGCCTGGCTG7GACCCCTAATATTCGGGAAAATGCAGGTTTGCGTGAACTCGTTATGCTGGGCTTT 1332
GLA7TPNIRSNAGLREL7MLGF


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E4OW70TP8_FYIPRQ INGEST_TIME 2014-10-20T21:02:20Z PACKAGE AA00025778_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES


ACKNOWLEDGMENTS
Much of the data presented in this dissertation was gained through cooperative
projects. The original cDNA librarians, Marion Byrne, Josna Kanungo, and Laura
Nelson, added a great gris-gris to the library. The N-terminal sequencing work was done
by the ICBR Protein Core under the supervision of Nancy Denslow, with special
contributions by Hung Nguyen and Sara Reynolds, who were particularly helpful and
patient in their protein isolation wizardry. Invaluable skill in artwork and photography
was contributed by Lynn Milstead and Jim Netherton.
I would like to thank the faculty members that invested an extra amount of
support to my academic training. Kelly Sel man gave me my first view of what yolk and
the vitelline envelope really look like. Kyle Rarey was instrumental in my decision to
join the Dept, of Anatomy and Cell Biology, and offered a safe haven for my qualifying
exam studies. Gill Small introduced me to library screening techniques and library
screaming techniques. Dave Price was always willing to field questions on a wide
variety of subjects from specific molecular interactions to cooking recipes for
invertebrates.
My committee members were very supportive. Chris West walked me through
my first isolation of DNA; he was also the best Mardi Gras king I have ever seen. Paul
Linser was generous with space and comradery in his kind-hearted lab, the chicken wing.
iii


99
protein products as Chg 500, Chg 427 and Chg 553, according to the number of residues
in the predicted amino acid sequence.
The cDNA encoding Chg 500 is 1641 bp long, including a 1500 bp open reading
frame (Fig. 5.2a). The calculated molecular weight after subtracting the weight of a
predicted signal peptide (residue 1-22) is 53,125. The most notable region of the
predicted primary structure is a proline-rich repeating domain near the N-terminus,
including five repeats of (PQQ PQQ PQY PSK). Other proteins that share sequence
identity with Chg 500 include the flounder ZP gene product (58%; Lyons et al., 1993),
which also contains a proline-rich repeating domain (Figs. 5.3a and 5.3b), and several
mammalian ZP proteins, including mouse ZP1 (32%; Epifano et al., 1995), cat ZPB
(35%; Harris et al., 1994), and human ZPB (34%; Harris et al., 1994).
The Chg 427 is encoded by a cDNA of 1751 bp (Fig. 5.2b). Subtracting the
weight of the predicted signal peptide (residues 1-24) leaves a calculated molecular
weight of 44,892. This protein sequence does not include a substantial repeating domain,
although residues 28-46 (PGK PSK PQS PPT QNQ QQL Q) are reminiscent of the
proline-rich repeat previously described for Chg 500. The predicted N-terminus
possesses three in-frame methionine codons, but the first codon agrees best with the
context and positional environment for initiation of translation as described by Kozak
(1991). Alignment analyses revealed that Chg 427 shares highest identity (67%) with a
medaka female-specific protein termed "L-SF" (Murata et al., 1995) (Fig. 5.4). The
next highest identity comes from sequences recently deposited for two cyprinid fishes
(42% from C. auratus ZP3, 43% from C. Carpi ZP3i, and 44% with C. carpi ZP3ii).


114
evidence that the Chgs are indeed expressed in the liver, but not the ovary of spawning
females. We additionally showed that the Chgs can be induced in males by injection
with estradiol (Fig. 5.5). From the cDNA sequences, we expected the sizes of the
mRNAs encoding Chgs 500, 427, and 553 to be 1.64, 1.67 and 1.82 kb, respectively.
In northern blots, however, the Chg 500 probe hybridized to a band estimated to be 1.9
kb, while the Chg 427 hybridized to two bands, at 1.7 and 1.4 kb, and the Chg 553
probe hybridized to two bands at 2.3 and 1.8 kb. Indication of doublet mRNAs by
hybridization to Chgs 427 and 553 probes was observed from repeated stringent
hybridizations using different individual samples as
well as with probes representing different sections of the cDNA (not shown). We
interpret these data to suggest that two isoforms, possibly splicing variants of Chgs 427
and 553, are present in the liver total RNA. We also suggest that our cDNA clones
probably did not contain the total amount of 5 untranslated sequence, consistent with a
conservative estimate of mRNA sizes as compared with actual mRNAs indicated by the
gels.
The Predicted Structure of Chgs 500 and 553
The proline-glutamine-rich domains found in Chg 500 and Chg 553, along with
that of the flounder ZP (Lyons et al., 1993) represent a novel protein domain for
vertebrates. Although high proline and glutamic acid/glutamine compositions had long
been predicted through amino acid composition analyses of VEPs from F. heteroclitus
(Kaighn, 1964) and other fish (Young and Smith, 1956; Iuchi and Yamagami, 1976,


Pre-Hyb Pre-Hyb
Vtg I Vtg II
a b c d e a' b' c' d' e'
RNA
kb
7.46
4.4
2.37
1.35
28s
18s
0.24


I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy
Robin A Wallace, Chair
Professor of Anatomy and Cell Biology
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Assistant Scientist
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
A\.
Christopher West
Associate Professor
of Anatomy and Cell Biology
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy. "7
Paul J. Linse
Associate Professor
of Anatomy and Cell Biology


39
have long been recognized as tetrapod-like, ie. a holoblastic embryonic cleavage and
anuran-like gastrulation (Balinksky, 1965; Beer, 1981; Conte et al., 1988), an acrosome-
capped spermatozoan (Conte et al., 1988), and development of oviducts from true
Mllerian ducts (Conte et al., 1988). We suggest that the structure of F. heteroclitus Vtg
represents a derived, perhaps more specialized, example of Vtg structure in contrast to
the tetrapod/chondrostean Vtg, which more likely resembles the Vtg of an ancestral
osteichthyean. If this is the case, we would predict that the structure of an elasmobranch
Vtg (especially from a less derived species) would also resemble the tetrapod Vtgs more
closely than it would a teleostean Vtg. Whether the structure of lamprey Vtg represents
an independent derivation, or an even earlier, prototypical vertebrate Vtg, is difficult to
surmise. This question will be more easily answered once a protochordate or
invertebrate deuterostome Vtg (from within the "Vtg family) has been sequenced. Within
the invertebrate outgroup of our phylogram, the two insect Vtgs appear to be highly
derived versions of Vtg structure as compared to the C. elegans Vtg. The C. elegans
Vtg is substantially more similar to vertebrate Vtgs than are the Vtgs of the two insects,
suggesting a faithfulness of the nematode Vtg to an ancestral form originating in a
predecessor common to both vertebrates and platyhelminthes.
In reference to past alignments between multiple Xenopus and chicken Vtgs,
Byrne (1989) described Pv as an independently evolving domain within Vtg. Our
alignment confirms this suggestion. While the two Lv domains of Vtg can be well
aligned among several organisms, the polyserine domain exists in a wide range of sizes


98
the insolubility of the VEPs. Successful transfer of the VEPs was accomplished in 0.01
M MES, 10% methanol, and .01% SDS. After transfer, the PVDF membrane was
stained with 0.02% Coomassie blue in 40% methanol and 5% acetic acid to indicate
protein bands. Duplicate membrane transfers containing VEP 69, 60, and 46 were
submitted for amino acid composition analysis and N-terminal amino acid analysis using
an Applied Biosystems Model 473a Sequencer (LeGendre and Matsudaira, 1988) at the
Protein Chemistry Core Facility of the University of Florida Interdisciplinary Center for
Biotechnology Research.
The N-terminal sequences initially obtained from the three bands consisted of
overlapping and weak signals that were only five residues long (data not shown);
therefore the VEP 69, 60, and 46 were isolated again and subjected to in-gel digestion
with endoproteinase Lys C (0.003 units//xg protein, Promega), in 10 mM Tris, pH 8.8,
0.2M ammonium bicarbonate, and 0.1 % SDS, 0.1 mM EDTA. Protein fragments were
separated by Tris-tricine gels, blotted onto PVDF and visualized by silver staining. Two
of the best resolved bands from each digestion were once again submitted to the Protein
Chemistry Core for N-terminal amino acid sequencing.
Results
Choriogenin cDNA Sequences
The nucleotide and translated amino acid sequences from three estrogen-induced
liver cDNAs are presented in Figure 5.2. We have designated the cDNAs and predicted


104
Mammalian ZP proteins sharing high identity with Chg 427 include mouse ZP3 (30%;
Ringuette et al., 1988), cat ZPC (32%; Harris et al., 1994) and human ZP3A (32%;
Chamberlin and Dean, 1990)(Fig. 5.4). Chg 427 shares little identity (18%) with
Chgs500 and Chg 553 (not shown). A Prosite scan predicted only one N-glycosylation
site from residue 184-187 (Fig. 5.2b).
Chg 553 was translated from a 1817-bp cDNA (Fig. 5.2c). Subtraction of a
predicted signal peptide (residue 1-26) resulted in a calculated molecular weight of
58,290. Chg 553 is 62% identical to Chg 500, and likewise shares identity with the
flounder ZP (52%) (Fig. 5.3a), mouse ZP1 (30%; Epifano et al., 1995), cat ZPB (29%,
Harris et al., 1995), and human ZPB (28%; Harris et al., 1994). The N-terminal region
of Chg 553 contains a proline-rich repeating domain that differs from that of Chg 500
by containing only half as many glutamine residues.
A ClustalV alignment (not shown) containing the ZP domains (Bork and Sander,
1992) of seventeen reported sequences, including the three Chg sequences, and five other
reported sequences from fish, plus three mouse, three cat, and three human ZPs, was
used in parsimony analysis. The shortest tree resulting from a heuristic search with 100
bootstrap replicates is presented in Figure 5.8. The resulting unrooted tree was drawn
according to the format of the Fitch analysis program in order to emphasize relatedness
among sequences rather than a deduced ancestral relationship. Bootstrap values are
indicated adjacent to the appropriate nodes. The results of the analysis suggest that three
major groups of ZP proteins can be described, each one containing a separate set of
mouse, cat, and human ZP sequences. In this paper we refer to these groups according


Figure 5.6 Northern blot analysis testing ovary vs. liver expression of Chgs.
A) Methylene blue staining of a nylon membrane blot of lanes
containing 2.0 /xg of total liver RNA next to loads of 20 /xg of
total ovarian RNA, from two identically-treated female fish.
Lanes a, e, and i contain liver RNA from fish 1, while lanes b, f,
and j, contain ten times more RNA isolated form the ovary of fish
1. Likewise, lanes c, g, and k contain liver RNA from fish 2,
while lanes d, h, and 1 contain ten times more total RNA, isolated
from the ovary of fish 2. RNA kb markers are indicated on the
left with 28s and 18s rRNA bands indicated on the right.
B) Autoradiograph showing the same nylon membrane after being
cut into three pieces and hybridized (65C) with the random
primed [32P] Chg probe indicated above the blot. Although ten
times more ovarian RNA than liver RNA was loaded onto the gel,
only bands from the lanes containing liver RNA hybridizing to the
Chg probes. Absolutely no hybridization was seen in the lanes
containing ovary RNA.


5
transported to the ovary, to be used by the germ cells and their descendants. Two of
these proteins, Vtg I and Vtg II, are endocytosed by the oocyte, processed, and stored
as yolk (Fig. 1.2), mainly to be used as a nutrient source by the developing embryo.
The three remaining proteins, designated the choriogenins (Chgs), are also synthesized
by the estrogen-induced liver, and transported to the ovary. However, rather than being
endocytosed, the Chgs are laid down as an extracellular matrix between the oocyte and
follicle cells forming the vitelline envelope (Fig. 1.2, in brackets).
The Original Emphasis: Vitellogenins
One of the initial goals of this project was to establish a definitive precursor-
product relationship between vitellogenin and the processed yolk proteins. It was decided
that primary sequence information would be needed for this goal and that the best method
to gain the amino acid sequence of vitellogenin was to use a molecular approach, produce
a cDNA library, screen for Vtg with degenerate primers designed from yolk proteins,
and sequence the cDNA clone. Before the lengthy Vtg sequence was completed, the
original research team disbanded. I subsequently joined the Wallace lab and thereby
"inherited" the Vtg sequencing project. Influenced by the dissertation of Byrne (1989)
describing the evolution of yolk proteins, I became interested in the evolutionary aspects
of Vtg, particularly in the independently evolving phosvitin domain. The lack of a
phosvitin domain in the Caenorhabditis elegans Vtgs (Speith et al., 1991) prompted the
idea that phosvitin may be an exclusively vertebrate inclusion within the Vtg gene (Byrne


89
by research concentrating on the chemistry of hatching enzymes in the medaka, Oryzias
latipes. When polyclonal antibodies directed against protein fragments of the lysed
chorion were used as probes on medaka tissues, Hamazaki et al. (1984) found that tissues
other than the ovary were recognized by the antibody. By 1989 Hamazaki et al. (1989a)
had isolated an estrogen-induced glycoprotein from the liver that could be localized to
the inner layer of the vitelline envelope. Since then additional reports have verified these
findings in several other fish (Hyllner et al., 1991; Murata et al., 1991; Oppen-Bemtsen
et al., 1992a, 1992b; Larsson et al., 1994). Additionally, Hyllner et al. (1991) showed
that the synthesis of VEPs could be induced by estrogen treatment in males of the
rainbow trout (Oncorhynchus mykiss), brown trout (Salmo trutta), and turbot
(Scophthalmus maximus), providing convincing evidence that the major VEPs in these
species could be synthesized without any contribution by the ovary.
So far, only two nucleotide and protein sequences representing piscine VEPs have
been published. Lyons et al. (1993) reported a gene sequence (wf) from the flounder,
Pseudopleuronectes americanus, that they described as a "teleostean homolog of a
mammalian ZP gene." The predicted amino acid sequence contained a novel PQQ
repeating region near the N-terminus, resembling a motif found in extracellular matrix
proteins. Murata et al. (1995) reported a cDNA sequence (L-SF) from medaka that also
shared identity with mammalian ZP proteins. The predicted amino acid sequence of the
medaka L-SF protein shared more identity with mouse ZP3 (37.9%; Ringuette et al.
1988) than it did with the flounder ZP sequence (18%) previously mentioned, suggesting
the presence of at least two distinct groups of teleost ZP homologs.


3
yolk by the oocytes (vitellogenesis) (Taylor et al., 1977; Hsiao et al., 1994). The origin
of the yolk proteins can be traced to a cascade of events resulting in the maternal liver
synthesizing a suite of secreted proteins, primarily consisting of the yolk precursor,
vitellogenin (Vtg), but also containing riboflavin- and vitamin- binding proteins (White,
1987; White and Merrill, 1988) and most recently discovered, precursors of the vitelline
envelope (Hamazaki et al., 1985; Murata, et al., 1991; Hyllner et al., 1991). Thus, the
oocytes, or germ cells, demand an investment by the maternal or somatic cells. They
are saying, "Pay me now." This extensive investment begins long before fertilization,
without the adult knowing whether the eggs will actually ever be spawned or fertilized.
Once the oocytes are expelled, the female, having already surrendered a substantial
amount of energy and material, is relieved of any further investment (until the next clutch
of oocytes begins its demands).
On the other hand, in mammals the germ cells present more of a "Pay me later"
scenario. Mammalian oocytes appear to not receive any yolk at all, with synthesis of
vitellogenin presumed (but not proven) to be totally nonexistent in mammals (except in
the egg laying monotremes) (Eckelbarger, 1994). The investment, then, comes mainly
after fertilization, with support and nourishment provided first by a modification of the
uterus into the chorionic villi, and secondly through lactation, where protein nourishment
continues to be demanded by the progeny, and thus supplied by the adult.
The work contained in this dissertation provides an example of the "Pay me now"
demands of the oocyte on its somatic surroundings. We provide evidence of at least five
distinct proteins that are made by the maternal liver, in response to estradiol, and


Figure 2.2 Translated amino acid sequence (1,704 residues) of the putative F.
heteroclitus Vtg polypeptide. Two separate signal peptide
predictions are presented. The first was obtained by an alignment
with other fish Vtg signal peptides (Folmar et al., 1995) and is
denoted by shaded lettering. The second prediction was obtained
by the computer analysis method of von Heijne (1986) and is
denoted by asterisks. The nucleotide stretch corresponding to the
degenerate oligonucleotide MB6, used to screen the library, is
shown by double underlining and bold letters. Five predicted
antigenic determinants are depicted by shaded lettering with
average hydrophilicity values (Ah) indicated underneath. A
polyadenylation site (A AT A A A) is located 53 nucleotides past the
stop codon and denoted by underlining.


106
A)
Chg 500
Chg 553
Flounder2?
1 MT MKLIYCCLLAVAIHG YLVG- AQPGKPQ YPSKPQ
1 MASHWSVTRWAA-LALL-CCLAGXGAEA QKGSY?PQPQKPSYPQNPQTPSY?
1 MAKRWSANSLVAQWLIYLVWTNVEVLGSRRRS RS SERGGR1XVQQTGHYH?AGKGQRYV
Chg 500
Chg 553
FlounderZP
3 5 -QPQQPQYPSKPQQPQQPQYPSKPQQPQQPQYPQQPQQPQQPQ YPSXPQY?SKPQQP
51 QQPQKPSYPQNPQTPSYPQYPQTPSNPQQPQYPQTPSNPQYPQT? SYPQNPQTPSYPQN?
6 1 QQRRRLHHDF SPQNPG- AEPPQTPQQPTYPQQPQQPQQPQQPKYPQQPQ QPQQP
Chg 500
Chg 553
FlounderZP
9 1 QQPQYPSXPQQPQ QPQYPQKPQQPQQPQYPQ KPQTP.TE -
111 QTPSYPQNPQQPQLSWDFSXPTKPQYPKPQRPPSKPQYPRPQTP P SKPQYPRPQTPQQPG
114 QQPKYPQQPQQPQ QPQQPKYPQQPQQPQQPQQPKYPQQPQQPKNPQPKNPQPPQ
Chg 500
Chg 553
FlounderZP
129 TFHTCDVP AP FRIQCGAPT XSNTECEAINCCFDGRMC
17 1 KKQWDDTKTPNVP SKRPEAPGVPTPKSCDVEVASRVPCGASAVSATXC2ARDCCFDGQ SC
163 PQKNPQPTXQQVSDDRI FCGVDPYLRIQCOVDDXTAAZCSALXCCFEGYQC
Chg 500
Chg 553
FlounderZP
166 YYGKSVTLQCTXDGQFIIWARDATLPHXDL3SXSLLGGGPNCGPVGTT3AFAXYQFFAD
231 YFAXGVTVQCTXDGHFXVVVAXDVTLPHXDLXTXSLLGGGQGCTHVDPNSLPAX^YFPVT
219 FFGKAVTVQCTXDAQFVVVVAKDATLPNLIINTXSI.QaEaQQCTAVDSNSEFAIFQ.FPVL
Chg 500
Chg 553
FlounderZP
226 CCGTIMTESPOVIXYSNRMASSYSVAVaPYGAXTRDSQYSLFVQCRYXaTSXEALVXEV-
291 ACGTVVMESPGVIMYENRMTSSYSVGVGPLGAITRDSTYSLLFQCRYXGTSVBTLWEVL
27 9 ACGSVVTEEPGT XXYSNRMTSSYBVDVGPNGVXTRDSFFXLQFQCRYTGLSXZTVVIEIL
Chg 500
Chg 553
FlounderZP
285 GLLPPPPGVAAPGPLRVELRLGNGECSVKGCTSEQVAYT3YYTDADYPVTKXLRDPVYVE
351 PLDNPPPAVAELGPIRVALRLAN0QCATKGCNEAZVAYTSYYLDSDY2ITKILRDPVYVE
33 9 P SNTPPRPVAALGPIRVQLRLGNGECETKGCNEVEAAYTSYYTEGDYPVTKVLRDPVYVE
Chg 500
Chg 553
FlounderZP
345 VRXLXRTDPNXVLTLGRCWATASPFPQSLPQWDLLXNGCPYQDDRYRTNLXPVDSSSGCL
411 VQLLZKTDPALVLTLGRCWATTSPNPRSLPQWDILXDGCPYTDDRYLSTLVPVDASSGLQ
39 9 VRLLZKRDPNLVLTLGRCWVTNSPNPHHQPQWDLLXDGCPYADDRYXSSLVPVG? SSGVN
Chg 500
Chg 553
FlounderZP
40 5 FPTHYRRFVFXMFTFVSGGGGASDATKKTP SDP SWNPLHXKVYXHCDAAVCQ PSMTNSCS
47 1 FPSHYRRFTFKMFTFVDTTAM DPLRZNVYXHCSTAVCVPGQGVSCS
4 5 9 FPTHYXRFIFKMFTFVDSSTLZPQRRR CTFTV
Chg 500
Chg 553
FlounderZP
46 5 PSCGRXXRZX SGSTKMX SRZEATXVSSK3WFTAT- Z
517 PSCNRXGXRDTZAAEQRKVZPKVVVSSG3VIMTAPQZ
491 VQLS ALVTQAAPVSRHATG
B)
Chg 500
FlounderZP
Chg 500
FlounderZP
33 PQQPQQPQY PSKPQQPQQPQYPSKPQQPQQ
97 CCTCAGCAACCCCAGCAGCCTCAGTATCCTTCCAAGCCTCAGCAACCCCAGCAGCCTCAGTATCCTTCGAAGCCTCACCAACCCCAGCAG
238 CCACAGACTCCACAGCAACCAACGTACCCACACCAACCACAGCAGCCACAGCAACCACAGCAACCAAAGTACCCACAGCAACCACAGCAG
BO PQTPQQPTyPQQPQgPQQPQQPKYPQQPQQ
62 PQYPQQPQQPQQPQYPSKPQYPSKPQQPQQ
186 CCTCAGTATCCCCAGCAGCCTCAACAACCCCAGCAGCCTCAGTATCC7TCGAAGCCTCAGTATCCTTCGAAGCCTCAGCAACCCCAGCAG
327 CCACAGCAACCACAGCAACCAAAGTACCCACAGCAACCACAGCAACCACAGCAACCACAGCAACCAAAGTACCCACAGCAACCACAGCAA
109 PyQPQ'3PKY PQQPQQPQQPQQPKY PQQPQQ


132
LaFleur GJ Jr, Byrne MB, Kanungo J, Nelson LD, Greenberg RM, Wallace RA (1995)
Fundulus heteroclitus vitellogenin: The deduced primary structure of a piscine
precursor to noncrystalline, liquid phase yolk protein. J Mol Evol 41:505-521
LaFleur GJ Jr, Byrne MB, Greenberg RM, Haux C, Wallace RA (1996) Liver-derived
cDNAs: vitellogenins and vitelline envelope protein precursors (choriogenins). In:
Thomas P, Goetz F (eds) Proceedings of the Fifth International Symposium on
the Reproductive Physiology of Fish, (in press) University of Texas Press,
Austin, TX
LaFleur GJ Jr, Thomas P (1991) Evidence for a role of Na+, K+- ATPase in the
hydration of Atlantic croaker and spotted seatrout oocytes during final maturatin
J Exp Zool 258:126-136
Lange RH (1981) Are yolk phosvitins carriers for specific cations? Comparative
microanalysis in vertebrate yolk platelets. Z Naturforsch 36:686-687
Lange RH (1985) The vertebrate yolk-platelet crystal: Comparative analysis of an in vivo
crystalline aggregate. Intemat Rev Cytol 87:133-181
Lange RH, Kilarski W (1986) Similarity in yolk-patelet structure of an ancient bony fish
(Acipenser) and an ancient reptile (Sphenodon). Tissue Cell 1:117-124
Larsson DGJ, Hyllner SJ, Haux C (1994) Induction of vitelline envelope proteins by
estradiol-17(8 in ten teleost species. Gen Comp Endocrinol 96:445-450
Lee K-S, Yasumasu S, Nomura K, Iuchi I (1994) HCE, a constituent of the hatching
enzymes of Oryzias latipes embryos, releases unique proline-rich polypeptieds
from its natural substrate, the hardened chorion. FEBS Lett 339:281-284
LeGendre N, Matsudaira, O (1988) Direct protein microsequencing from Immobilon-P
transfer membranes. Biotechniques 6:154-159
LeGuellec K, Lawless K, Valotaire Y, Dress M, Tenniswood M (1988) Vitellogenin
gene expression in male rainbow trout (Salmo gairdneri) Gen Comp Endocrinol
71: 359-371
Liang L-F, Chamow SM, Dean J (1990) Oocyte specific expression of mouse Zp-2:
Developmental regulation of the zona pellucida genes. Mol Cell Biol 10:1507-
1515
Liang L-F, Dean J (1993) Conservation of mammalian secondary sperm receptor genes
enables the promoter of the human gene to function in mouse oocytes. Dev Biol
156:399-408


76
fragments were then separated by Tris-tricine gels, blotted to PVDF, and visualized by
silver staining (Wray et al., 1981). Similar bands of 13 kDa (presumed to be identical)
were isolated from both the YP 125, and YP 105 digestion and once again submitted to
the Protein Chemistry Core for N-terminal amino acid sequencing.
Sequencing data, including Vtg I and II cDNAs, along with microsequencing
results were organized using the PC/GENE software package (Intelligenetics, Mountain
View, CA). Prediction of signal peptides was carried out according to von Heijne
(1986). PEST sites were designated according to the algorithm described by Rogers et
al. (1986). Other Vtg sequences referred to in this paper include chicken Vtg II
(gi:63887; van het Schip et al. 1987), Xenopus laevis Vtg A2 (gi: 139636, Gerber-Huber
et al. 1987), lamprey, Ichthyomyzon unicuspus Vtg (gi:213312, Sharrock et. al. 1992)
and sturgeon, Acipenser transmontanus Vtg (gi:437051, Bidwell and Carlson, 1995).
Results
The yolk proteins typically found in F. heteroclitus oocytes and eggs are
demonstrated in Figure 4.1, along with our designations of certain bands according
to their apparent molecular mass. At least nine yolk proteins were resolved by Tris-
tricine SDS-PAGE, and these were blotted onto PVDF membranes and submitted for
protein sequencing by Edman degradation. Four yolk proteins appeared to be N-
terminally blocked, while five yielded N-terminal sequences (Table 4.1).
By aligning the yolk protein N-terminal sequences against the predicted amino
acid sequences of Vtg I and Vtg II, we successfully mapped the five sequenced yolk


44
cleaved, as are bona fide Pv proteins, has not yet been reported. It is possible that
polyserine domains have existed within Vtgs since before the emergence of chordates,
but that Pv proteins, per se, remain a unique chordate trait, representing a novel
modification and utilization of these polyserine tracts.
Though we know little of why Vtg polyserine domains vary in size, findings from
studies of heritable disease may offer clues as to how these size differences originated.
The aberrant amplification of trinucleotide repeats from one generation to another has
recently been coupled to the occurrence of several human genetic diseases including
Huntingtons Disease (Huntingtons Disease Collaborative Research Group, 1993; review
by Caskey et al., 1992). An increased potential for trinucleotide amplification may
explain the faster rate of evolution attributed to the Pv region in comparison to its two
flanking Lv regions (Byrne et al., 1989). We are aware of very few descriptions of
"yolk-based diseases" in fish (Olin and von der Decken, 1989), and in these it was
neither suspected nor tested whether the disease was caused by aberrant amplification of
the Pv polyserine domain. Diseases aside, novel duplications or omissions in the Pv
polyserine domain may certainly have affected the evolution of specific yolk structures
or functions. F. heteroclitus, possessing the smallest polyserine domain of our
alignment, produces a yolk which remains totally soluble throughout oocyte development.
As the smaller yet serine-enriched polyserine domain of F. heteroclitus Vtg is
phosphorylated and finally processed into a more soluble Pv yolk protein, it may
somehow be prevented from re-combining with the Lv yolk proteins and forming the
insoluble yolk complexes of other vertebrates. Another possible explanation for the


48
Material and Methods
Chemicals
Estradiol-17/3 was obtained from Sigma Chemical Co. (St. Louis, MO).
Radioisotopes, [c*-32P]dCTP and [a-35S]dATP, were purchased from New England
Nuclear (Boston MA). Lambda gtlO vector, cDNA synthesis reagents, the subcloning
plasmid Pgem-T, and T4 ligase were obtained from Promega (Madison, WI). All
amplification reactions were performed using a 50:1 mixture of Taq DNA
polymerase:cloned Pyrococcus furiosus DNA polymerase (Stratagene, La Jolla, CA).
All sequencing gels were cast using Sequagel-8 (National Diagnostics, Atlanta)
polyacrylamide reagents. Sequenase version 2.0 DNA polymerase and dideoxy
sequencing reagents were obtained from US Biochemicals (Cleveland, OH). Reagents
for random-primed labeling of probes were purchased from Pharmacia (Piscataway, NJ).
Magna nylon transfer membranes were used for nucleic acid transfers and purchased
from MSI (Westboro, MA). Amino acid N-terminal sequencing, synthesis of
oligonucleotide primers, and a limited amount of DNA sequencing were performed by
the University of Florida Interdisciplinary Center for Biotechnology Research core
facilities.
Cloning strategy using an estrogen-induced liver cDNA library
Seven of the eight overlapping clones resulting in the contiguous cDNA sequence
were isolated from a XgtlO liver library whose synthesis has been previously described


18
compromised). After phosphorylation of adapter ends, the cDNA transcripts were ligated
into the bacteriophage vector X gtlO (Promega). Once the X particles were packaged,
the primary library was plated using host E. coli strain C600HFL, resulting in an initial
library titer of 6 x 104 total plaque forming units.
Two 24.5 cm2 petri dishes were used for plating phage-transfected cells
(-400,000 total plaques). Plaques were lifted onto nylon membranes. Hybridization
was performed at 39C using IX Denhardts solution (Denhardt, 1966), 6X SSC (150
Mm NaCl and 15 Mm sodium citrate, pH = 7) with the same end-labelled
oligonucleotide probe as described earlier, MB6. The primary screening resulted in 30%
of the plaques testing positive for the degenerate MB6 probe. By following four plaque
clones (X5, X20, X21, XI6) through two more rounds of positive screening, four final X
clones were set aside for subcloning. The clone (X21) containing the largest insert
(-5000 bp) was subjected to endonuclease digestion with EcoRl, which was expected
to free the entire cDNA insert. Unfortunately the EcoRl sites had inadvertently been
modified so that when digested with EcoRl, one end of the insert remained attached to
the vector. Alternatively, the enzymes Hindlll and Bglil, which like EcoRl cleave at
rare sites, were used to digest X21. Two large fragments were released (1900 bp from
EcoRMBglll and a 2060 bp fragment from EcoR 1 /Hindlll) and these were subcloned into
the sequencing plasmid PGEM 3Z resulting in subclones pMMB6 and pMMBl,
respectively. Because the size of these two fragments did not add up to the total putative
insert size (6000 bp), digestion of another clone, X5, was performed in order to provide


55
GGCATCATGGTrCACAAATACTGTGTGGAGAACCCTTCATGTCCATCrGAGCTGGTCAGGCCAGTT 1398
GXMVHKYCVENPSCPSELVRPV
CATGACATTATTGCCAAGGCTCTTGAGAAACGCGACAATGATGAGCTCTCCCTGGCrCTCAAAGTT 1464
HDIIAKALEKRDNDELSLALKV
CTGGGTAATGCCGGACATCCCAGCAGCCTGAAGCCAATCATGAAACT'rCTTCCTGGCTTTGGCAGC 1530
LGNAGHPSSLKPXMKLLPGFGS
TCTGCCTCCGAACTTGAGCTCAGAGTTCACATTGACGCTACACTGGCGCrGAGGAAAATTGGCAAG 1596
SASELELRVHIDATLALRKIGK
AGAGAACCCAAGATGATrCAGGATGTGGCCCTTCAGCTCTTCATGGACAGGACTCTTGACCCAG AG 1662
RSPKMIQDVALQLFMDRTLOPS
CTCCGTATGGTTGCTGTTGTTGTGCTGTTTGATACCAAGCTACCTATGGGTCTGATAACCACTCTC 1728
LRM7A777LFDTKLPMGLITTL
GC'rCAGAGTCTCCTGAAACAGCCAAACCTGCAGGTCCTTAGCTTTGTCTACrCTTACATGAAGGCC 1794
AQSLLKSPNI.Q7LSF7YSYMKA
TTCACCAAGACCACCACCCCGGACCATTCCACTGTAGCCGCTGCCTGCAATGTTGCCATCAGGATC 1860
FTKTTTPOaSTVAAACNVAIRI
CTCAGCCCAAGATTCGAAAGACTGAGCTACCGCTACAGCCGAGCTTTCCATTATGACCACTATCAT 1926
LSPRFERLSYRYSRAFaYDHYB
AATCCTTGGATGCTGGGAGCTGCTGCCAGCGCATTTTACATCAATGATGCCGCGACTGTATTGCCA 1992
NPWMLGAAASAFYINDAATVLP
AAAAACATCATGGCAAAAGCTCGCGTTTACCTCTCTGGAGTGTCTGTTGATGTTCTGGAG7TTGGA 2053
KNIMAKAR7YLSG7S7D7LEFG
GCCAGAGCTGAAGGAGTGCAAGAGGCCCTTTTGAAAGCCCGTGATGTTCCTGAGAGTGCAGACAGG 2124
ARASG7QEAI.I.KARD7PESADR
CrrCACCAAGATGAAGCAAGCTCTTAAGGCTCTGACTGAGTGGAGGGCCAATCCTTCCCGCCAGCCT 2190
OTKMKQALKALTEWRANPSRQP
CTCGGCTC7CTGTACGTGAAGG7TCTTGGGCAGGATGTTGCTTTGCAAACATCGACAAAGAAATG 2256
LGSLY7K7LGQD7AFANIDKEM
GTTGAGAAGATCATTGAGTTTGCAACTGGACCTGAAATCCGCACCCGTGGCAAAAAGGCCTTGGAC 2322
72KIIEFATGPEIRTRGKKALD
GCCCTGTTGTCTGGTTACTCTATGAAATACTCCAAGCCAATGTCGGCCATTGAGGTCCG7CACATC 2388
ALLSGYSMKYSKPMSAIS7Rai
TTCCCCACCTCTCTTGGTTTACCCATGGAGCTCAGTCTGTACACTGCTGCCGTGACAGCCGCATCC 2454
FPTSLGLPMSLSLYTAA7TAAS
G7TGAAGTACAAGCCACCATTTCACCACCACTTCCCGAGGACTTCCATCCTGCCCACCTAC7GAAG 2520
7E7QATISPPLPEDFHPAHLLK
TCTGATATTTCCATGAAGGCT7CAGTCACTCCAAGTGTATC7TTGCACACCTATGGAGTTATGGGA 2586
SO ISMKAS7TPS7SLBTYG7MG
GTGAATAGTCCTTTCATCCAGGCrrCTGTGCTGTCAAGAGCCAAAGACCATGCAGCTCTTCCCAAA 2652
7NSPFIQAS7LSRAKDBAALPK
AAGATGGAGGCAAGACTTGACATAGTCAAGGGTTACTTTAGCTACCAGTTCCTGCCTGTTGAGGGT 2718
KMEARLDI7KGYFSYQFLP7EG
Figure 3.2--continued


12
The now familiar Vtg gene family (Wahli et al., 1979, 1991; Tata et al., 1980;
Blumenthal et al., 1984; Nardelli et al., 1987; Byrne et al., 1989; Speith et al., 1991)
encompasses Vtgs synthesized by a wide range of metazoans including the nematode
Caenorhabditis elegans (Speith et al., 1985), the boll weevil Anthonomus granis
(Trewitt et al., 1992), the silkworkm Bombyx mori (Yano et al., 1994), the mosquito
Aedes aegypti (Chen et al., 1994) the cyclostome Ichthyomyzon unicuspis (Sharrock et
al., 1992), the anuran Xenopus laevis (Germond et al., 1984; Gerber-Huber et al., 1987),
and the chicken Gallus domesticus (van het Schip et al., 1987). Additionally, two human
cDNAs, those encoding von Willebrand factor ( 250 kDa) (Baker, 1988a) and
apolipoprotein B-100 (-510 Kda) (Baker, 1988b), have also been reported as distantly
related members of the Vtg gene family. Exceptions to a Vtg-derived yolk precursor
system have been reported in at least two dipteran species: Drosophila melanogaster
(Hovemann et al., 1981) and Ceratitis capitata (Rina and Savakis, 1991) where yolk
precursors, often called Vtgs, do not, in fact, share significant sequence identity with the
"Vtg gene family" setting a precedent for the use of alternative molecules in the
production of yolk (Terpestra and AB, 1988; Bownes, 1992).
A large component of vertebrate Vtgs, the Pv region, was found to be nonexistent
in both C. elegans and the boll weevil Vtg (Nardelli et al., 1987; Trewitt et al., 1992),
inspiring the notion that Pv was an element unique to vertebrate Vtgs. The apparent
absence of the Pv region from invertebrate Vtgs (see Discussion), along with studies
documenting the ability of the phosphate groups of Pv to bind and transport large
amounts of divalent cations, especially Ca++ (Urist et al., 1958; Urist and Schjeide,


81
YP 125 are identical Vtg I-derived yolk proteins except for a short 20 Kda extension at
the C-terminus of YP 125 that contains the third Endo LysC digestion product (Fig. 4.3).
The C-terminus of YP 105 was predicted to lie at (or before) residue Ser 962 of
Vtg I, using the estimated mass of YP 105 and the masses of the individual residues
predicted from the Vtg I cDNA. This places the YP 105 C-terminus only 2 residues
away from the N-terminal residue obtained from YP 20 (Glu 965) suggesting that YP 105
and YP 20 result from cleavage of YP 125. The estimated juncture between YP 105 and
YP 20 lies at the exact midpoint of a predicted PEST site (residues 952-974, receiving
a score of 6.9, where 5.0 and above is considered a site). This purported cleavage site
bisects the predicted PEST site, leaving the two resulting protein sequences with termini
that do not surpass the cutoff value for valid PEST sites. Thus, although YP 125
contains a PEST site, neither of its cleavage products, YP 105, nor YP 20 do.
Discussion
We have presented precursor-product relationships to account for the origin of
seven yolk proteins isolated from oocytes and eggs of F. heteroclitus. Likewise, the
sequences determined from these yolk proteins verify the expression, transport, and
incorporation of both the yolk protein precursors Vtg I and Vtg II, whose cDNA
sequences are provided in Chapters 2 and 3, respectively.
We had initially assumed that YP 125 and YP 105, the major bands in oocyte
extracts, were derived separately from Vtg I and II, but the internal sequences indicated
that both yolk proteins originate from Vtg I. We can thus surmise that Vtg I is truly the


64
Discussion
F. heteroclitus Vtg II cDNA and predicted amino acid sequence are provided in
Figure 3.2. Vtg II mRNA is present in the liver of estrogen-treated males and normal,
spawning females (Fig. 3.5). The N-terminal sequence of a yolk protein isolated from
ovulated eggs was found to be identical to the predicted N-terminus of the putative Vtg
II translation product. Taken together these data indicate that the yolk proteins of F.
heteroclitus are derived from a mixture of at least two estrogen-induced liver precursors,
Vtg I and Vtg H, establishing the existence of a Vtg gene family in F. heteroclitus.
These two cDNAs represent the first two Vtg sequences from a single vertebrate species
to have been completely sequenced, offering a unique perspective into the possible
variance between Vtg isoforms occurring in single species.
Examination of the alignment of Vtg I and Vtg II reveals typical conservation of
lipovitellin regions seen among other vertebrate Vtgs. As previously described for other
vertebrate Vtgs (LaFleur et al., 1995) poor alignment occurred in the polyserine
domains. Although the tandem repeats of serine can be aligned in small stretches, the
overall lengths and intervening amino acid sequences are highly variable, resulting in a
region whose conservation is difficult to interpret. In an attempt to compare and
visualize these polyserine domains, hypothetical boundaries were drawn up according to
those used in a previous report (LaFleur et al., 1995), and a graphical representation was
created showing relative domain length as well as serine codon usage (Fig. 3.4).
Whereas the serine codons (TCX and AGY) of the Vtg I poly serine domain appear to be


142
coral reefs and cloud forests near Veracruz, where LaFleur gained an appreciation for
the structure and diversity of the oft-ignored invertebrates.
For his thesis research, LaFleur examined the role of Na+, K+- ATPase in the
hydration of Atlantic croaker oocytes (a maturation-associated phenomenon that had
gained renewed attention through the publications of Robin Wallace). LaFleur earned
his M.S. degree from CCSU in May, 1989.
Having been strongly influenced by the work of Robin Wallace, LaFleur sought
out a position in his lab at the University of Florida Whitney Laboratory. Under the
mentorship of Wallace, and with the support of Rob Greenberg in the Molecular Biology
Suite, LaFleur began investigating cDNAs isolated from an estrogen-induced liver library
in Fundulus heteroclitus. His doctoral dissertation was entitled "Estrogen-induced hepatic
contributions to ovarian follicle development in Fundulus heteroclitus: Vitellogenins and
Choriogenins." He completed his degree in May, 1996.
While studying in Port Aransas LaFleur met and fell in love with Susanna Lee
Lamers. They moved to Gainesville together and were married in St. Augustine on Leap
Day, 1992. In 1993 Susanna established Gene Genie, a company specializing in
computer analysis of genetic data. Their daughter, Hannah Alyse was bom on May 1,
1995. In March, 1996, LaFleur began work as a research scientist with Gary Wessel at
Brown University studying the molecular biology of cortical granule biogenesis in
urchins.


107
Chg 427 1
MadakaL-S? 1
Mouse 2? 3 1
Cat ZPC 1
Human ZP3A 1
Chg 427 60
MadakaL-S? 60
Mouse ZP 3 3 2
Cat ZPC 29
Human Z? 3A 3 2
Chg 427 113
MadakaL-S? 113
Mouse ZP 3 81
Cat ZPC 79
Human Z? 3A 81
Chg 427 178
MadakaL-S? 178
Mouse ZP 3 140
Cat ZPC 139
Human Z? 3 A 141
Chg 427 238
MedaJcaL-S? 238
Mouse ZP3 200
Cat ZPC 199
Human ZP 3 A 201
Chg 427 295
MadakaL-S? 295
Mouse ZP 3 259
Cat ZPC 256
Human ZP 3A 258
Chg 427 352
MedaJcaL-SF 352
Mouse ZP 3 319
Cat ZPC 316
Human ZP 3A 318
Chg 427 400
MadakaL-S? 393
Mouse ZP 3 379
Cat ZPC 376
Human ZP 3 A 378
MMMKWTV?CWALALLGS?CDAQ-GYAXPGKP SK?QS ? PTQNQQQLQTFZKZLTWXYPDD
MM-KFTAVCLVVLALLDG?CDAQHNYGK?SYPPTGSXTPQDPTQQKQLHZXZLTWKYPAD
MASSYFLFLCLLXCGGPELCNSQT LWLLPGG
MGLSYGLFICFLLWAGTGLCYP PT T TED
MELS YRLFICLLLWGS TELCYPQP LWLLQGG
PQPDPKPNVPFSLXYPVPAATVAVECRSSIAHVEVXXDMFGTGQPXNPNDLTLGN CA?
PQ PEAX?VVPFSQRYPVPAATVAVECREDLAHVSAXXDL7GIGQFXDFADLTLGT CPP
TPTPVGSSS? VKVZCLZAZLVVTVSRDL7GTGXLVQPGDLTLGSEGCQ?
KTHPSLP SSP S VVVECRH AWL VVNVSKNL7GTGRLVRPADLTLGP ESJCE?
ASHPETSVQP VLVECQEATLMVMVSXDL7GTGKLXRAADLTLGPEACE?.
VGZDSAAQVLXY3ASLHQCGSQLMMTNDALVYTFVLNYNPTPLGSVFWRTSQAAVXVEC
5 ASD? AAQVLXF3S PIiQNCGSVLTMTEDSLVYT3!TL2IYNPK?LGSAPVyRTSQAVVXYEC
RVSVDTD-WRFNAQLHECSSRVQMTKDAI.VYSTFLLHDPRPVSGLSILIlTNRVEV!PISq
LISGDSDDTVRFXVELHKCGNSVQVTEDAXVYSTFLLHNPRPMGNLSILRTNREVPISC
LVSMDTEDVVR73VGLHECGNSMQVTDDALVYST FLLHDPRPVQNL 3IVRTNRAEI?ISC
HYPRXHMVSSLPLDPLWVPFSAVKMASEFLYFTMXLMTDDWMY QRPSTQYFLGDLIRXEV
ZY?RZHNVSSLALDPLWV??3AAKMA22?LYFILXLT7DD?Q?3H2SJ1Y?L3DL>3ISA
R YPRQ GNVSSHPI 3 PTWVP FRATVS 33ZXLA7S LRLMEZNWNTZX3A? 7 FHLQEYAHLQA
RYPRHSNV3 SEA!LPTWVPFRTTML 3ZZKLAFS LRLME2DWG3 2KQ SPT 7QLGDLAHLQA
RYPRQGNVSSQAILPTWL2FRTTVFSEZXLTFSLRL2IEENWNAEXRSPTFHEGDAAHLQA
TVXQYFHVPLRVYVDRCVATLSP DVTSSPNYAFXDMFGCEIDARXTGSDSXE-MARTQ
TVXQ.YFHVPLRVYVDRCVATESP--DANSSPSYAFIDNYGCIiLDGRXTaSDSKF-VSRPA
3VQTGSHLPLQL7VDHCVATPSPLPDPNSSPYHFIVDFHGCLVDG-LSZ3FSAFQVPRPR
2VHTGRHIPLRLFVDYCVATLT PDQNASPHHTIVDFHGCEVDG-LSDAS SA7KAPRPR
ZIHTGSHVPLRLFVDHCVAXPT PDQNASPYHTIVDFHGCLVDG-LTDASSAFKVPRPG
SNHLQFQLZAFRFOMSDSGVXYXTCYLXATSTSQAIDSQHRACSYT GGWRSASGVDG
3NKLDFQLZAFRF' GADSGMXYXTCHLXATSAAY ? LDAEHRACSYI QGWXSVSGADP
PETLQFTVDVFHF.MSSRNTLYXTCHLKVAPANQIPDKLNKACSFNKTSQSWLPVEGDAD
PETLQFT7DTFHFAMDPRNMXYXTCHLJCVTPA3RVPDQLNKAC3FIKS SNRWFPVEGPAD
? DTLQFTVDVFHFAMDSRNMXYXTCHLXVTLAEQDPDELNXACSFSK? SNSWFPVEGPAD
ACGSCSTNVTP YTAP A VTFASPPVVVTDGGOVTLPAPGS - PKVP YNPRX
I CAS CSS GG FEVHA NAVVSHGT STLSGGGHGTGKPSD- ? SRK
ICDCCSHGNCSNSSS SQFQIHGPRQWSXLVSRNRRHVTDEADVTVGPLIFLGXANDQTVE
ICNCCNKGSCGLQGRSWRLSHLDRPWHKMASRNRRHVTEEADITVGPLIFLGXAADRGVE
ICQCCNKGDCGTP SHSRRQPHVMSQWSRSASRNRRHVTEEADVTVGPLI7LDRRGDHEVE
VRDVTQAZILZWEGV VSLGPIPIMEKXL
TREAAKTEVLZWSGD VTLGPIPIEERRV
GWTASAQTSVAL-GLGLATVAFLTLAAIVLAVTRKCHSSS- YLVSLPQ
GSTSPHTS--VMVGIGLATVLS LTLATIVLGLARRHHTASRPMICPVSASQ
QWAL? SDTSVVLLGVGLAVVVSLTLTAVILVLTRRCRTASHP VSASE
Figure 5.4 A ClustalV alignment of Chg 427, against the medaka L-SF protein
(Murata et al. 1995), the mouse ZP3 (Ringuette et al., 1988), cat ZPB
(Harris et al. 1994), and human ZP3A (Chamberlin and Dean 1990)A
conserved core region sharing sequence identity with other ZP proteins
and designated as a "ZP domain" (Bork and Sander, 1992) is denoted by
a dark line. This is the core domain used for drawing the tree shown in
Figure 5.8.


103
C)
actaactagaccagacagcttcgaggt 27
ATGGCAAGTCACTCGAGTGTCACCCGTTCGGCCGCGCTGGCTCTGCTATGCTGCTTAGCTGGGAAA 93
HASHWSVTRWAALALLCCLAGK22
GGAGCAGAGGCTCAGAAGGGTTCGTATCCTCCGCAACCTCAAAAGCCTTCGTACCCTCAGAATCCT 159
G pgpgA QKOSYPPQPQKPSYPQNP44
CAAACGCCTTCGTATCCTCAGCAACCTCAAAAGCCTTCGTACCCTCAGAATCCTCAAACGCCTTCG 225
QTPSYPQQPQKPSYPQNPQTPS66
TACCCTCAGTATCCTCAAACACCTTCAAACCCTCAGCAACCTCAGTATCCTCAAACACCTTCAAAC 291
YPQYPQTPSNPQQPQYPQTPSN88
CCTCAGTATCCTCAAACGCCTTCGTACCCTCAGAATCCTCAAACGCCTTCGTACCCTCAGAATCCT 357
PQYPQTPSYPQNPQTPSYPQNP 110
CAAACGCCTTCGTACCCTCAGAACCCTCAGCAACCTCAATTGTCGTGGGATTTTTCAAAGCCTACA 423
QTPSYPQNPQQPQLSWDFSKPT 132
AAACCTCAATATCCTAAGCCCCAAAGGCCTCCATCAAAACCTCAATATCCTAGGCCCCAAACGCCT 489
KPQYPXPQRPPSKPQYPRPQTP 154
CCTTCAAAACCTCAATATCCTAGGCCTCAAACGCCCCAACAACCTGGAAAAAAACAATGGGATGAT 555
PSKPQYPRPQTPQQPGKKQWDD 176
ACAAAGACTCCGAATGTCCCTTCCAAGAGACCAGAGGCCCCTGGAGTTCCCACCCCTAAAAGTTGT 621
TKTPNVPSKRPEAPGVPTPKSC 198
GACGTGGAAGTAGCTTCAAGAGTCCCCTGTGGAGCTTCTGCCGTCTCTGCTACTGAATGTGAGGCC 687
DVEVASRVPCGASAVSATECEA 220
AGAGACTGTTGCTTTGATGGCCAGTCATGCTACTTTGCAAAAGGAGTGACAGTCCAGTGTACCAAG 753
RDCCFDGQSCYFAKGVTVQCTK 242
GATGGCCATTTTATCGTTGTTGTGGCCAAAGATGTCACCCTGCCACACATTGACCTTGAAACAATC 819
DGHFIVVVAKDVTLPHIDLETI 264
TCATTGTTJGGAGGAGGTCAAGGCTGTACACATGTTGACCCCAATTCACTTTTTGCCATCTACTAC 885
SLL.GGGQGCTHVDPNSLFAIYY 286
TTTCCCGTTACTGCTTGTGGGACTGTTGTCATGGAGGAGCCTGGCGTTATAATGTATGAGAATCGG 951
FPVTACGTVVMEEPGVIMYENR 309
ATGACCTCCTCATATGAAGTAGGAGTTGGGCCTCTTGGAGCCATTACCAGGGACAGCACCTACGAA 1017
MTSSYEVGVGPLGAITRDSTYE 330
TTGCTCTTCCAGTGTAGGTACATTGGCACCTCAGTTGAAACTTTGGTGGTCGAAGTGCTGCCATTA 1083
LLFQCRY IGTSVETLVVEVLPL 352
GACAATCCTCCTCCAGCAGTTGCTGAGCTCGGACCGATCAGAGTGGCCCTTAGGTTGGCCAATGGC 1149
DNPPPAVAELGPIRVALRLANG 374
CAGTGTGCTACAAAGGGTTGCAACGAAGCGGAGGTAGCCTACACCTCCTACTATTTGGACTCAGAC 1215
QCATKGCNEAEVAYTSYYLDSD 396
TATCCGATTACCAAGATACTGAGGGATCCCGTGTATGTGGAGGTTCAGCTCCTTGAAAAGACAGAT 1281
YPITKILRDPVYVEVQLLEKTD 418
CCCGCTCTGGTTCTGACTCTTGGACGTTGTTGGGCAACCACTAGCCCCAATCCTCACAGCTTGCCC 1347
PALVLTLGRCWATTSPNPHSLP 440
CAGTGGGACATTCTGATTGACGGATGTCCCTACACGGATGATCGTTACCTCTCCACACTGGTTCCA 1413
QWDILIDGCPYTDDRYLSTLVP 462
GTGGACGCCTCTTCTGGTCTGCAATTTCCAAGTCACTACCGGCGTTTCACTTTCAAAATGTTTACC 1479
VDASSGLQFPSHYRRFTFKMFT 484
TTTGTGGACACCACTGCAATGGACCCCCTGAGGGAAAATGTGTACATTCACTGTAGCACAGCTGTG 1545
FVDTTAMDPLRENVYIHCSTAV 506
TGCGTGCCAGGACAGGGTGTCAGCTGCGAACCATCATGCAACAGAAAAGGAAAGAGAGACACTGAG 1611
CVPGQGVSCEPSCNRKGKRDTE 528
GCTGCAGAGCAGAGGAAGGTCGAACCAAAGGTTGTGGTTTCGTCCGGAGAAGTGATCATGACCGCT 1677
AAEQRKVEPKVVVSSGEVIMTA 550
CCTCAGGAGTAAtctgggacaagctcaggaattcatctgggaacatttagacaaaactctttgaaa 1743
P Q E 553
atcaacaaqqttqttqaacaqtaaataaaaatgtcaccctaaqtaaaaaaaaaaaaaaaaaaaaaa 1809
aaaaaaa 1816
Figure 5.2continued


34
Fundulus ILYIKVLGNAGHPSSFKSLTKIMPIHGTAAVSLPMTIHVEAIMAXRNIAKKESRMVQELA 53 5
Callus KLALKCIGNMGEPASLKRILKFLPISSSSAADIPVHIQIDAITALKKIAWKDPKTVQGYL 53 7
Xenopus ALALKALGNAGQPESIKRIQKFEPGFSSSADQLPVRIQTDAVMALRNIAKEDPRKVQEIL 537
Acipenser VLALKALGNAGQPSSIKRIQKCLPGFSSGASQLPVKIQVDAVMALRNIAKKEPGKVQELT 53 0
Ichihyomyzon VLALKALGNAGQPNSIKKIQRFLPGQGKSLDEYSTRVQAEAIMALRNIAKRDPRKVQIV 53 3
Furuiulus LQLYMDKALHPELRMLSCIVLFETSPSMGLVTTVANSVKTEE- -NLQVASFTYSHMKSLS 593
Callus IQILADQSLPPEVRMMACAVIFETRPALALITTIANVAMKESKTNMQVASFVYSHMKSLS 597
Xenopus LQIFMDRDVRTEVRMMACLALFETRPGLATVTAIANVAARESKTNLQLASFTFSQMKALS 597
Acipenser MQLFMDHQLHSEVRMVASMVLLETRPSMALVATLAEALLKE- -TS LQVASFTYSHMKAIT 588
Ichihyomyzon LPIFLNVAIKSELRIRSCIVFFESKPSVALVSMVAVRLRREP- -NLQVASFVYSQMRSLS 591
Furuiulus RSPATIHPDVAAACSAAMKILGTKLDRESLRYSKAVHVDLYNSSLAVGAAATAFYINDAA 653
Gallus KSRLPFMYNISSACNIALKLLSPKLDSMSYRYSKVIRADTYFDNYRVGATGEIFWNSPR 657
Xenopus KSSVPHLEPLAAACSVALKUjNPSLDNBGYRYSKVMRVDTFKYNLMAGAAAKVFIMNSAN 657
Acipenser RST APENHALS S ACNVAVKLIiS RKLDRLS YRYS KAMHMDT F KYPLMAGAAANIH11NNAA 648
Ichihyomyzon RSSNPEFRDVAAACSVAIKMLGSKLDRLGCRYSKAVHVDTFNARTMAGVSADYFRINSPS 651
Fundulus TFMPKSFVAXTKGFIAGSTAEVLEIGANISGLCELILKNPALSESTDR- 701
Gallus TMFPSAIISRLMANSAGSVADLVEVGIRVEGEADVIMKRNIPFAEYPT 705
Xenopus TMFPVFILAKFREYTSLVENDDIEIGIRGEGIEEFLRKQNIQFANFPM 705
Acipenser SILPSAWMKFQAYILSATADPLEXGLHTGliQBVLMQNHEHIDQMPS 6 96
Ichihyomyzon GPLPRAVAAKIRGQGMGYASDIVEFGLRAEGLQELLYRGSQEQDAYGTALDRQTLLRSGQ 711
Fundulus ITKflKRVIKAESEWRSLPTSKPIiASVYVKFFGQEIGFANIDKPMIDKAVKFGKELP 757
Gallus YKQIKELGKALQGWKELPTETPLVSAYLKILGQEVAFININKELLQQVMKTWEPA 761
Xenopus RKKISQIVKSLLGFKGLPSQVPLISGYIKLFGQEIAFTELNKEVIQNTIQALNQPA 761
Acipenser AGKXQQIMKMLSGWKSVPSEKTLASAYIKLFGQEISFSRLDKKTIQEALQAVREPV 752
Ichihyomyzon ARSHVSSIHDTLRKLSDWKSVPEERPLASGYVKVHGQEWFAELDKKMMQRISQLWHSAR 771
Fundulus IQEYG REALKALLLSGINFHYAKPVLAAEMRRILPTVAGIPMBLSLYSAAVAAASV 813
Gallus DRNAA JKRIANQILNSIAGQWTQPVWMGELRYWPSCLGLPLBYGSYTTALARAAV 817
Xenopus ERHTM 1RNVLNKLLNGWGQYARRWMTWEYRH11PTTVGLPAELS LYQS AIVHAAV 817
Acipenser ERQTV 1KRWNQLERGAAAQLSKPLLVAEVRRILPTCIGLPMEMSLYVSAVTTADI 808
Ichihyomyzon SHHAAAQEQI RAWS KLEQGMDVLLTKGYWSEVRYMQPVCIGIPMDLNLLVSGVTTNRA 831
Fundulus EIKPNTSPRLSADFDVKTLLETDVELKAEIRPMVAMDTYAVMGLNTDIFQAALVARAKLH 873
Gallus SVEGKMTPPLTGDFRLSQLLESTMQIRSDLKPSLYVHTVATMGVNTEYFQHAVEIQGEVQ 877
Xenopus NSDVKVKPTPSGDFSAAQLLESQIQLNGEVKPSVLVHTVATMGINSPLFQAGIEFHGKVH 877
Acipenser NVQAHITPSPTNDFNVAQLLNSNIVLHTDVTP3IAMHTIAVMGINTHVIQTGVELHVKAR 868
Ichihyomyzon NLHASFSQSLPADMKLADLLATNIELRVAATTSMSCHAVAIMGLTTDLAKAGMQTHYKTS 8 91
Fundulus SWPAKXAARLNIKEGDFKLEALPVDVPENITSMNVTTFAVARNIEEPLVERITPLIiPTK 933
Gallus TRMPMKFDAKIDVKLKNLKIETNPCREETEIVVGRHKAFAVSRNIGELGVEKRTSILPED 937
Xenopus AHLPAKFTAFLDMKDKNFKIETPPFQQENHLVEIRAQTFAFTRNIADLDSARKTLWPRN 937
Acipenser TTVPMKFTAKIDLKEKNFKIESEPCQQETEVLSLSAQAFAISRNVEDLDAAKKNPLLPEE 928
Ichihyomyzon AGLGVNGK EMNARESNFKASLKPFQQKTWVLSTMESlVFVR- -DPSGSRILPVLPPK 948
Fundulus VLVP -I PIRRHTSKLDPTRNSMLDSSEL - LPMEEEDVEPIPEYKF RRFA 980
Gallus APLD- VTEEPFQTSERASREH FAMQGPDS -MPRKQSHSSREDLRRSTGKRAHK 988
Xenopus NEQN- ILKKHFETTGRTSAE GASMMEDSSEM- -GPKKYSAEPGHHQYAPN INS 987
Acipenser AVRN- ILNEQFNSGTEDSNERERAGKFARPSAEM- -MSQELMNSGEHQNRKGA HAT 981
Ichihyomyzon MTLDKGLISQQQQQPHHQQQPHQHGQDQARAAYQRPWASHEFSPAEQKQIHDIMTARPVM 1008
Figure 2.4-continued


84
major yolk protein precursor in F. heteroclitus. Our finding that most of the yolk protein
is derived from Vtg I agrees with northern blot analyses that suggested ten times more
Vtg I than Vtg II message is present in total liver RNA (Chapter 3; LaFleur et al.,
1996).
A major factor that prevents construction of a definitive map accounting for all
yolk proteins derived from the Vtgs is the difficulty in isolating and microsequencing
peptides derived from the phosvitin domain (Wallace and Begovac, 1986; Wallace et al.
1990). Although we expect that the polyserine repeats represented in both Vtg I and II
cDNAs are processed into true phosvitins, we have been unable to verify this by N-
terminal sequencing. The highly negative charge of phosvitin prevents it from staining
with Coomassie blue, as well as adhering to PVDF membranes for sequencing. Because
phosvitin can be visualized using Stains-all, it has been documented as a single 25-30
kDa band in prematurational oocytes, with at least four smaller phosvitin-like bands
(phosvettes) appearing in preparations from ovulated eggs (Wallace and Begovac, 1985).
We estimate that the C-terminus of YP 20 (and presumably, the C-terminus of YP 125)
lies adjacent to the N-terminus of phosvitin, as predicted by the position of the Vtg I
cDNA polyserine repeating region. Likewise, the sequence obtained from YP 45,
sharing identity with residues 1220-1230 of Vtg I, most likely abuts the C-terminal
cleavage site of phosvitin.
As previously mentioned, one of the most pronounced changes observed to occur
in F. heteroclitus yolk proteins is the disappearance of YP 125 during the transformation
of oocytes to mature, ovulated eggs (Fig. 4.1). A possible explanation for this rapid and


Ill
faint to isolate. Amino acid analysis revealed that extraordinarily high proline and
glutamine compositions were present in VEP 69 and 60 (Table 5.1) agreeing well with
reported VEP compositions from other teleosts (Hyliner et al., 1991, 1995; Hamazaki
et al., 1987). The amino acid compositions of the isolated VEPs are compared to those
predicted from Chg cDNA translations in Table 5.1.
Discussion
We present the predicted primary structure of three liver-derived proteins, Chg
500, Chg 423, and Chg 553 (Fig. 5.2). We have shown that mRNAs hybridizing to
cDNA probes from each Chg occur in the liver RNA of estrogen-treated males and
spawning females, but are not detectable from ovarian RNA. Furthermore, the predicted
amino acid compositions of the Chgs are similar to the profiles of three VEPs isolated
from ovarian follicles (Table 5.1). We submit that although the Chgs differ from
mammalian ZP proteins by way of being estrogen-induced and synthesized in the liver,
they are in fact, related groups of proteins, as evidenced by the shared identity of a ZP
domain. Chg 500 and 553 can be more specifically grouped as homologs to the
mammalian ZP1 subfamily of molecules, while Chg 427 can be grouped with the
mammalian ZP3 subfamily (Fig. 5.8).
Northern Analyses
By showing no indication of Chg mRNA from 20 /g of ovarian RNA compared
with the ample Chg signals from only 2 /xg of liver RNA (Fig. 5.6), we provided strong


123
the more frequently used, were grouped at the 3 end of the domain. As the first teleost
Vtg to be completely sequenced, this report represented an important step toward in
gaining comparative information on Vtg variability.
In Chapter 3, the sequence of Vtg II cDNA and predicted protein structure was
presented. Vtg II shares 45% overall identity with Vtg I, and 30-40% identity with the
other vertebrate Vtgs. The polyserine domain of Vtg II is slightly smaller than that of
Vtg I, but more surprising is the polyserine codon organization. The trend that had been
observed previously for Vtg I and other vertebrate Vtgs was not apparent in the
polyserine domain of Vtg II. Rather than a clustering of TCX and AGY codons, each
type were interspersed throughout the length of the domain. In a comparison of mRNA
expression, Vtg I transcripts were 10 times more prevalent than those of Vtg II from total
liver RNA isolated from spawning females and estrogen-induced males. According to
these data, we suggest that Vtg I is the primary and Vtg II a secondary yolk precursor
in F. heteroclitus.
N-terminal sequences of isolated yolk proteins were provided in Chapter 4. We
were able to map out a precursor-product relationship for seven yolk proteins by
comparing obtained N-terminal sequences with the predicted amino acid sequences
derived from Vtg I and Vtg II. A PEST site found in the Vtg region mapping to YP 125
was hypothesized as a possible factor influencing the proteolytic processing of YP 125
during oocyte maturation as compared to YP 105, which does not contain a PEST site.
In Chapter 5 the cDNA and predicted protein sequences of the choriogenins was
presented. The Chg mRNAs were shown to be expressed by the liver of reproductive


25
1081
CAC
TGG
TTC
TTG
GAC
ACT
ATT
CCT
GCC
ACT
GGT
ACC
TTC
GCT
GGT
CTC
AGA
TTC
H
W
F
L
D
T
I
P
A
T
G
T
F
A
G
L
R
F
1135
ATC
AAA
GAG
AAG
TTC
ATG
GCT
GAG
GAA
ATA
ACC
ATC
GCT
GAG
GCA
GCT
CAG
GCT
I
K
E
K
F
M
A
E
E
I
T
I
A
E
A
A
2
A
1189
TTC
ATT
ACA
GCT
GTG
CAC
ATG
GTG
ACT
GCT
GAC
CCT
GAG
GTT
ATC
AAG
CTG
TTT
F
I
T
A
V
H
M
V
T
A
D
P
E
V
I
K
L
F
1243
GAG
AGC
CTG
GTA
GAC
AGC
GAC
AAA
GTA
GTG
GAA
AAC
CCA
CTT
CTG
CGT
GAG
GTT
E
S
L
V
D
S
D
K
V
V
E
N
P
L
L
R
E
V
1297
GTC
TTC
CTT
GGA
TAT
GGA
ACA
ATG
GTT
AAC
AAA
TAC
TGC
AAT
AAG
ACA
GTT
GAT
V
F
L
G
Y
G
T
M
V
N
K
Y
C
N
K
T
V
D
1351
TGT
CCT
GTT
GAA
CTC
ATA
AAG
CCT
ATT
CAA
CAA
CGA
CTG
TCA
GAC
GCC
ATT
GCA
C
P
V
E
L
I
K
P
I
2
Q
R
L
S
D
A
I
A
1405
AAG
AAC
GAG
GAA
GAG
AAC
ATC
ATC
CTG
TAC
ATA
AAG
GTT
TTG
GGA
AAT
GCC
GGC
K R B E ^B R
; -
I
I
L
Y
I
K
V
L
G
N
A
G
(Ah
- 2.07)
1459
CAT
CCA
TCT
AGC
TTC
AAG
TCA
CTC
ACT
AAG
ATC
ATG
CCC
ATC
CAT
GGC
ACT
GCT
H
P
S
S
F
K
S
L
T
K
I
M
p
I
H
G
T
A
1513
GCT
GTA
TCT
CTG
CCA
ATG
ACA
ATC
CAT
GTT
GAA
GCC
ATC
ATG
GCT
CTG
AGG
AAC
A
V
S
L
P
M
T
I
H
V
E
A
I
M
A
L
R
N
1567
ATT
GCA
AAG
AAG
GAG
TCC
AGA
ATG
GTC
CAG
GAA
CTG
GCT
CTC
CAG
CTC
TAC
ATG
I
A
K
K
E
S
R
M
V
Q
E
L
A
L
2
L
Y
M
1621
GAC
AAG
GCT
CTC
CAC
CCA
GAG
CTC
CGT
ATG
CTG
TCC
TGC
ATT
GTT
CTC
TTC
GAG
D
K
A
L
H
P
E
L
R
M
L
S
C
I
V
L
F
E
1675
ACA
AGT
CCT
TCT
ATG
GGT
TTG
GTG
ACA
ACT
GTT
GCC
AAC
TCT
GTG
AAA
ACC
GAG
T
S
P
S
M
G
L
V
T
T
V
A
N
S
V
K
T
E
1729
GAG
AAT
TTG
CAG
GTG
GCC
AGC
TTC
ACT
TAC
TCT
CAC
ATG
AAG
TCC
CTA
AGC
AGG
E
N
L
Q
V
A
S
F
T
Y
S
H
M
K
S
L
S
R
1783
AGC
ccc
GCA
ACC
ATC
CAT
CCC
GAT
GTT
GCT
GCC
GCA
TGC
AGC
GCC
GCC
ATG
AAG
S
p
A
T
I
H
p
D
V
A
A
A
C
S
A
A
M
K
1837
ATC
TTG
GGT
ACA
AAG
CTG
GAC
AGA
CTG
AGC
CTG
CGT
TAT
AGC
AAA
GCT
GTA
CAT
I
L
G
T
K
L
D
R
L
S
L
R
Y
S
K
A
V
H
1891
GTG
GAC
CTC
TAC
AAC
AGT
TCC
TTG
GCG
GTC
GGT
GCT
GCT
GCA
ACT
GCT
TTT
TAC
V
D
L
Y
N
S
S
L
A
V
G
A
A
A
T
A
F
Y
1945
ATC
AAC
GAT
GCT
GCC
ACC
TTT
ATG
CCA
AAA
TCC
TTT
GTT
GCA
AAG
ACC
AAA
GGC
I
N
D
A
A
T
F
M
P
K
S
F
V
A
K
T
K
G
1999
TTC
ATC
GCT
GGA
AGT
ACT
GCT
GAA
GTC
CTG
GAG
ATT
GGA
GCG
AAT
ATT
GAA
GGA
F
I
A
G
S
T
A
E
V
L
E
I
G
A
N
I
E
G
2053
CTG
CAG
GAG
CTG
ATT
CTG
AAA
AAC
CCT
GCT
CTC
TCT
GAA
AGT
ACT
GAC
AGG
ATC
L
Q
E
L
I
L
K
N
P
A
L
S
E
S
T
D
R
I
2107
ACC
AAA
ATG
AAG
CGA
GTC
ATT
AAG
GCT
CTG
TCA
GAA
TGG
AGA
TCC
TTG
CCC
ACC
T
K
M
K
R
V
I
K
A
L
S
E
W
R
S
L
P
T
2161
AGC
AAA
CCC
CTA
GCC
TCT
GTC
TAT
GTT
AAG
TTC
TTT
GGA
CAA
GAG
ATT
GGC
TTT
S
K
p
L
A
S
V
Y
V
K
F
F
G
Q
E
I
G
F
Figure 2.2-continued


73
immediately ground with a Kontes pestle and heated for 10 min at 100 C. The
homogenate was then briefly centrifuged at 12,000 g for 1 min., separating the dissolved
yolk from insoluble cellular debris. The supernatant was aliquoted to a fresh tube and
stored at -20C until electrophoresis. Samples were diluted again by as much as 1:50
with sample buffer before loading onto gels.
Sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) was
carried out according to Laemmli (1970), using 125 X 140 X 1.5 mm slab gels
containing a 3.5% stacking gel overlaying a separating gels ranging from 7% for larger
YPs to 12% for smaller YPs, with modifications based on the protocol of Schagger and
von Jagow (1987) using Tris-tricine running buffers.
Proteins in electrophorese gels were electroblotted onto PVDF membranes in
buffer containing 10 Mm MES, Ph 6, and 20% methanol at 20 V overnight. Protein
bands were visualized by brief staining in 0.02% Coomassie blue in 40% methanol plus
5% acetic acid, destained in 40% methanol plus 5% acetic acid, and rinsed in distilled
water. Membranes were dried and stored at -20 C until individual bands were cut out
and submitted for sequencing. N-terminal amino acid analyses were performed on PVDF
bound proteins using an Applied Biosystems Model 473a Sequencer (LeGendre and
Matsudaira, 1988) by the Protein Chemistry Core Facility of the University of Florida.
The two largest N-terminally blocked yolk proteins (YP 125 and YP 105) were
again electrophoresed, blotted onto PVDF and subjected to in situ cleavage (Scott et al.,
1988) by endoproteinase LysC (Endo LysC)(0.003 units//xg protein, Promega), in 50 mM
Tris, Ph 8.8, 0.2M ammonium bicarbonate, and 0.1% SDS, 0.1 Mm EDTA. Protein


119
group is homologous to mouse ZP2, their ZPB group is homologous to mouse ZP1, and
their ZPC group is homologous to mouse ZP3. Thus, this new organization scheme does
not improve on the original grouping offered by Wassarman et al. (1988a,b) based on
the three mouse ZPs: ZP1; ZP2; and ZP3. Therefore in our discussions of similarity to
mammalian ZP proteins we refer to three major groups of mammalian ZPs, according
to molecular identities shared with mouse ZP1, ZP2 or ZP3. These three "subfamilies"
are nevertheless contained within a larger family of related ZP proteins that can be
recognized by the possession of a conserved region designated as the ZP domain (Bork
and Sander, 1992).
Combining the present molecular data describing Chgs with the reports by Lyons
et al. (1993), Murata et al. (1995), and the Genbank entries for the carp ZP3 molecules,
it appears that the eight liver-derived VEP precursors described for teleosts can be
separated into two distinct groups: one containing Chg 500, Chg 553, plus the flounder
ZP protein; and another containing Chg 427, the medaka L-SF, and the three carp ZP3
sequences. Furthermore, these two fish groups share identity with two distinct
mammalian ZP subfamilies: Chgs 500 and 553 are grouped with the mouse ZP1
subfamily while Chg 427 is grouped with mouse ZP3 subfamily. Bootstrapping values
indicate a high confidence value associated with the Chg 427-mouse ZP3 subfamily
division, whereas the Chgs 500 and 553 might be expected to group with the mouse ZP2
or ZP1 subfamily. Whether Chgs 500 and 553 are closer related to the mouse ZP1 or
ZP2 subfamily, the trend remains that all reported piscine molecules separate into two
rather than three subdivisions. The lack of a third subtype of teleost homolog may


140
Yano K-I, Toriyama-Sakurai M, Watabe S, Izumi S, Tomino S (1994) Structure and
expression of mRNA for vitellogenin in Bombyx mor. Biochim. Biophys. Acta
(In Press)
Young EG and Smith DG (1956) The amio acid in the ichthulokeratin of salmon eggs.
J Biol Chem 219:161-164


120
reflect a difference in the construction and function of the teleostean vitelline envelope
from that of mammals, but is more likely due to a lack of targeted investigation. For
instance, although, we report the amino acid compositions of three VEPs: 69, 60, and
46, there is at least one other F. heteroclitus VEP, with estimated molecular weight of
65,000 that could be visualized but was stained too faintly to isolate. Thus, other minor
F. heteroclitus VEPs still remain that may represent a third subclass of teleost VEPs.
It is also possible that a third sub-type may be synthesized by the ovary rather than the
liver of teleosts, explaining why the studies investigating liver cDNAs have not yet
discovered it. Results obtained with the pipefish Syngnathus scovelli (Wallace and
Begovac, 1989) as well as a preliminary study in F. heteroclitus (Hamazaki et al., 1989b)
suggest that at least one VEP may indeed be synthesized within the ovarian follicle.
By this preliminary characterization of Chg 500, 427, and 553, estrogen-induced,
liver-derived precursors to the vitelline envelope, we hope to set the stage for further
investigations of the regulation, structure and function of the vitelline envelope and
chorion. Besides being used to study development of the ovarian follicle, the Chgs
should provide excellent biomarkers to indicate either naturally occurring, or
toxicological states of estrogen induction in fish.


21
of post-translational modification sites by PROSITE, signal peptide prediction by
PSIGNAL, antigenic determinant analysis using ANTIGEN, codon usage statistics by
CDUSAGE.
Protein sequence alignments were performed using two programs: ClustalV
(Higgins et al., 1992), utilizing the PAM 250 matrix, gap penalty=3, K-tuple=l, no.
of top diagonals=5; window size=5) for the multiple alignment, and ALIGN Plus (S&E
Software, State Line, PA) for pairwise alignments. To normalize domain comparisons,
we defined a "polyserine domain" within the Vtg sequence by choosing two well-aligned
termini as the exterior boundaries, thereby including all of the poorly-aligned polyserine
tracts on the interior. Because we do not have yolk protein data to map the exact region
which is processed into Pv, we have chosen this "polyserine domain" to represent a
hypothetical Pv domain. The chicken and Xenopus Pv termini, which have been
documented (Clark, 1973; Gerber-Huber et al., 1987), lie to the inside of our
boundaries, verifying our convention.
A phylogram was drawn to compare Vtg sequences from eight species. Although
multiple isoforms of Vtg have been identified from several organisms, nomenclature
formally separating these isoforms into subfamilies has not yet been proposed. For our
tree analysis we selected only one Vtg sequence from each species. In species which
contain multiple Vtgs, we chose either the only complete Vtg available from Genbank
databases, as in chicken and Xenopus, or the Vtg which is considered the "major" yolk
protein precursor, as in C. elegans (Speith et al., 1985). An optimal tree was chosen by
importing a ClustalV alignment into the program PAUP (Swofford, 1983) and performing


47
product. In contrast, the three Vtgs genes reported from the chicken include no apparent
silent genes. The primary translation products Vtgl, Vtgll, and VtgUI were found to be
present in blood in a ratio of 0.33 : 1.00 : 0.08 confirming chicken Vtgll as the major
yolk protein precursor (Wang et al., 1983). Recently two Vtgs have been reported in
related tilapia species, Oreochromis aureus (Ding et al., 1989) and O. mossambicus
(Kishida and Specker, 1992). These studies established the occurrence of two piscine
Vtgs (180 kDa and 130 kDa) using an immunological approach. The immunological data
from O. aureus was additionally complemented by a small nucleotide sequence from the
C-terminus of one of the purported Vtgs, probably the larger (Ding et al., 1990).
Though the existence of multiple Vtgs has been established in these species, it remains
unclear as to why several Vtgs would be functionally necessary.
We have recently reported the cDNA sequence and predicted primary structure
of Fundulus heteroclitus Vtg I, as a precursor to non-crystalline, liquid phase yolk
proteins (LaFleur et al., 1995). Here we describe a second F. heteroclitus Vtg cDNA and
protein sequence that we have designated as Vtg II. The predicted primary structure
shares 45% identity with Vtg I (with regions as high as 65%) and contains the same
general domain profile: a large lipoveitellin 1 region, followed by a serine-rich, phosvitin
region and terminating in lipovitellin 2 region. We have confirmed Vtg II MRNA
expression as well as a derived yolk protein cleavage product, verifying that Vtg II
represents a separate but functional Vtg. This report therefore, establishes the existence
of a bona fide Vtg gene family in F. heteroclitus that acts as a precursor to liquid phase
yolk proteins.


568
40
(6555)
234
(9555)
278
(6755)
341
(100%)
665
Gall LIS Vt2 II
469
428
Xenopus Vtg A2
Acipenser Vtg
563
Fundulus Vtg
581
836
Ichthyomyzon Vtg
Caenorhabditis Vtg 5
642
(100%)
896
Anthonomus Vtg
591
Aedes Vtg
Figure 2.5 Branch-and-bound phylogenetic tree analysis comparing selected Vtgs
spanning 600 million years of divergence (Raff et al., 1989). PAUP
(Swofford, 1992) analysis was done on a ClustalV alignment (Higgins et
al., 1992) containing five of the vertebrate cDNAs from Fig. 2.4.:chicken
Gallus domesticus Vtg II; clawed frog Xenopus laevis Vtg A2; white
sturgeon Acipenser transmontanus Vtg; mummichog Fundulus heteroclitus
Vtg; silver lamprey Ichthyomyzon unicuspis Vtg; plus three invertebrate
Vtg cDNAs; nematode Caenorhabditis elegans Vtg 5; boll weevil
Anthonomus granis Vtg; and mosquito Aedes aegypti Vtg. The Gallus
Vtg was designated as the reference sequence and the C. elegans Vtg was
defined as the outgroup. The number of reconstructed changes in amino
acid sequence occurring along each branch are shown without parentheses;
bootstrap data are depicted at partition boundaries as percentages in
parentheses.


122
library, primarily due to fortuitous annealing events, that led to the discovery of cDNAs
coding for Vtg II and the three choriogenins (Chgs).
In order to complete the cDNA sequence of Vtg I, two small overlapping regions
were isolated out of the library using anchored PCR (Fig 2.1). The resulting cDNA
(5112 bp) and predicted amino acid sequences (1704 residues) of Vtg I (gi:459202) were
described in Chapter 2 (Fig 2.2). Alignment of the F. heteroclitus Vtg against the other
known vertebrate Vtgs revealed 30%-40% sequence identity being shared among the
proteins (Fig. 2.4). The sturgeon Vtg sequence was found to share more identity with
chicken and Xenopus Vtgs than with the F. heteroclitus Vtg, suggesting that the F.
heteroclitus Vtg reflects a more derived rather than ancestral vertebrate protein (Fig 2.5).
We had hoped that by comparing the sequence of F. heteroclitus Vtg with the Vtgs of
other vertebrates, we might find an explanation for why F. heteroclitus yolk proteins
remain in a non-crystalline liquid phase. Analyses predicting secondary structure showed
no obvious differences to account for structural disparity. Although the poly serine
domain of the F. heteroclitus Vtg was the shortest of the five vertebrates, it possessed
the highest relative composition of serine. We hypothesized, assuming these serines were
phosphorylated, that the poly serine domain of F. heteroclitus Vtg may be more
hydrophilic and polar than that of the other Vtgs, perhaps preventing the recombination
of phosvitin and lipovitellin that occurs in granular or crystalline yolk. A graphical
representation was used to emphasize the codon usage of the polyserine domains from
sequenced vertebrate Vtgs (Fig. 2.6). A specific clustering of codons was observed:
TCX codons generally occurred near the 5 end of the domain, and AGY codons, often


5 FUND UL US HETER0CL1TUS CHORIOGENINS:
LIVER-DERIVED COMPONENTS OF THE VITELLINE
ENVELOPE AND CHORION SHARING SEQUENCE
IDENTITY WITH MAMMALIAN ZP PROTEINS 87
Introduction 87
Material and Methods 90
Results 98
Discussion Ill
6 GENERAL SUMMARY 121
REFERENCES 125
BIOGRAPHICAL SKETCH 141
vi


17
from six estrogen-treated males and the other from five sham-treated males.
Oligo-dT cellulose chromatography was used to isolate poly A+ RNA from the
two initial pools of total RNA (Aviv and Leder, 1972). Of the 2.1 mg total RNA from
estrogen-treated fish, 46.3 ig poly A+ RNA was recovered (2.2% recovery). Poly A+
RNA from both experimental and control fish was analyzed by northern blot analysis to
verify that Vtg transcripts were included in the poly A+ RNA fraction. The poly A+
RNA was dissolved in deionized glyoxal/DMSO (1:1) and electrophoresed through an
agarose gel (McMaster and Carmichael, 1977) and transferred by capillary action onto
a nylon membrane. The membrane was probed with a 32P end-labeled 17-mer
oligonucleotide, MB6 (degeneracy = 32) which was designed from the N-terminal amino
acid sequence of a small yolk peptide isolated from F. heteroclitus oocytes: His-Lys-
Lys-Met-Val-Ala. Autoradiography of northern blots revealed an MB6 positive, ~ 6 kb
transcript found in the estrogen-treated fish which was absent in sham-treated male fish.
This transcript size was consistent with Vtg cDNA previously reported from chicken
(Cozens et al., 1980; Amberg et al., 1981; van het Schip et al., 1987), frog (Whali et
al., 1979), and rainbow trout (Le Guellec et al., 1988).
cDNA library construction, screening, and sequencing
Synthesis of cDNA was performed by annealing 2 /ig poly A+ RNA with oligo
dT primers, and using AMV reverse transcriptase and T4 DNA pol I for first and second
strand synthesis respectively. Eco R1 adapters were ligated to the two ends of the cDNA
transcripts using T4 DNA ligase (these Eco R1 adapters were later found to have become


113
Pre-Hyb Pre-Hyb Pre-Hyb
Chg 500 Chg 427 Chg 553
i ii ii
abcdef ghijkl
RNA
kb
4.4-
2.37
1.35-
LOLOLOLOLOLO
Chg 500 Chg 427 Chg 553
i ii ii 1
abcdefghii kl
RNA *
kb
4.4-
2.37
1.35-
28s
18s
LOLOLOLOLO LO


REFERENCES
Amberg A, Meijlink FCPW, Mulder J, van Bruggen EFJ, Gruber M, AB G (1981)
Isolation and characterization of genomic clones covering the chicken vitellogenin
gene. Nucleic Acids Res 9:3271-3286
Aviv H, P Leder (1972) Purification of biologically active globin messenger RNA by
chromatography on oligothymidylic acid-cellulose. Proc Nat Acad Sci USA
69:1408-1412
Bairoch A, Bucher P, Hofmann K (1995) The Prosite database, its status in 1995.
Nucleic Acids Res 24:189-196
Baker ME (1988a) Invertebrate vitellogenin is homologous to human von Willebrand
factor. Biochem J 256:1059-1063
Baker ME (1988b) Is vitellogenin an ancestor of apolipoprotein B-100 of human low
density lipoprotein and human lipoprotein lipase? Biochem J 255:1057-1060
Balinsky BI (1965) In: An introduction to embryology. Second Edition. W.B. Saunders
Co., Philadelphia, p 137
Banaszak LJ, Sharrock W, Timmins P (1991) Structure and function of a lipoprotein:
Lipovitellin. Ann Rev Biophys Biophys Chem 20:221-246
Beer KE (1981) Embryonic and larval development of white sturgeon (Acipenser
transmontanus). M.S. thesis, University of California, Davis. 93 pp
Begovac PC, Wallace RA (1989) Major vitelline envelope proteins in pipefish oocytes
originate within the follicle and are associated with the Z3 layer. J Exp Zool
251:56-73
Bergeron JM, Crews D, McLachlan JA (1994) PCBs as environmental estrogens: turtle
sex determination as a biomarker of environmetnal contamination. Environ Health
Perspect 102:780-781
Bergink EW, Wallace RA (1974) Precursor-product relationship between amphibian
vitellogenin and the yolk proteins, lipovitellin and phosvitin. J Biol Chem
249:2897-2903
125


CHAPTER 4
PRECURSOR-PRODUCT RELATIONSHIP OF
VITELLOGENINS I AND II TO THE YOLK PROTEINS
OF FUNDULUS HETEROCLITUS
Introduction
Current views concerning the origin and processing of yolk proteins in oviparous
vertebrates were formed through a slow, and controversial suite of biochemical studies
that eventually elucidated two unexpected aspects concerning the origin of yolk proteins
(reviews by Wallace 1978, 1985; Eckelbarger 1994). First, it was shown that yolk
proteins originated "hetero-synthetically" in the liver, rather than the ovary. Secondly,
it was shown that yolk proteins were not synthesized individually, but rather as a large
protein precursor, that was subsequently processed into bona fide yolk proteins. This
yolk protein precursor, vitellogenin (Vtg), has now been documented to appear in the
blood of estrogen-treated males or spawning females from countless oviparous vertebrates
(Wallace and Jared, 1969). Additionally, Vtg has been documented to be incorporated
into growing oocytes by receptor-mediated endocytosis (Wallace and Jared, 1969b;
Opresko et al., 1980; Stifani et al., 1990; Shen et al., 1993; Shibata et al., 1993), and
processed into yolk proteins (Wallace and Jared, 1969). Although isotopic and
immunologic tracking studies have established the connection between yolk proteins and
Vtg, direct sequence data, mapping the precursor-product relationship of Vtg to the
70


94
Sequence Analyses
Nucleotide sequencing data was organized and assembled using the sequence
analysis software package PC/GENE (Intelligenetics, Mountain View, CA). A search
for post-translational modifications and signature sequences was done with the Prosite
program (Bairoch et al., 1995) available from the world wide web ExPaSy molecular
biology server (http://expasy.hcuge.ch/www/expasy.top.html). Protein alignments were
performed with the ClustalV program (Higgins et al. 1992), utilizing a Pam 250 matrix
with fixed gap and floating gap penalties = 10. In order to compare Chg sequences with
a large number of ZP Genbank entries a preliminary ClustalV alignment containing
complete sequences was performed. Whereas the N- and C-termini from Chgs differ
greatly with those of mammalian ZP proteins, a core region of conserved sequence was
observed where all three Chgs, as well as all other reported ZPs could be aligned with
a minimum number of gaps when anchored to five strictly conserved cysteines. This
region has previously been defined by Bork and Sander (1992) as the "ZP domain" and
is included in the Prosite program (Bairoch et al., 1995) available on the ExPaSy
molecular biology server. For parsimonious tree analysis, a new ClustalV alignment was
performed including only the ZP domains from each entry, providing a well conserved
region on which to base our distance analysis. Parsimonious tree analysis was done by
importing a ClustalV alignment in phylip 3.4 format into the PAUP 3.1 program
(Swofford, 1993) available from the Center for Biodiversity (Champagne IL). The
unrooted tree presented in Figure 5.8. resulted from running 100 bootstrap replicates of
a heuristic search. All entries used in alignment and tree analysis were retrieved from


50
Fundulus heteroclitus Vitellogenin II cDNA 5195 bp
ROW 55
pFhv2h
pFhv2a
ROW 19
pFhv2b
pFhv2g pFhv2f pFhv2c
pFhv2e
pFhv2d
Figure 3.1 Cloning strategy used in isolating the F. heteroclitus Vtg II cDNA (5166
bp). Seven inserts (pFhv2a thru g) were isolated from the XgtlO liver
library by anchored PCR with indicated oligonucleotide primers and
inserted into the pGem-T cloning vector. The final cDNA (pFhv2h) was
isolated by RACE using reverse primer ROW 55.


85
rather selective proteolysis is the occurrence of a PEST site within the C-terminal tail of
YP 125. The apparently longer-lived YP 105 is identical to YP 125 except for lacking
the C-terminal tail where the PEST site occurs. PEST sites were initially defined as a
conserved clustering of amino acids that was observed to occur in proteins known to be
rapidly degraded. Common to all PEST site are high local concentrations of Pro, Glu,
Ser, and Thr, and to a lesser extent Asp. Of the other five vertebrate Vtg sequences
contained in Genbank, chicken Vtg II (residues 1058-1080 and 931-951) and lamprey Vtg
(residues 1161-1182 and 1360-1393) contain two PEST sites, while Xenopus Vtg A2
contains a sequence (residues 953-969) with a score (4.71) very close to the cutoff value
of 5. The lack of proteolysis during oocyte maturation in such animals may indicate
either the absence of an appropriate proteolytic mechanism or the inaccessibility of the
cleavage sites in the granular yolk of these animals (Wallace, 1985). The Vtg of
sturgeon, a chondrostean fish, does not contain a PEST site.
The proteolytic processing of YP 125 has been implicated as part of the hydration
mechanism of F. heteroclitus oocytes, with the generated small peptides and free amino
acids providing the osmotic potential to drive an uptake of water into the oocyte (Wallace
and Begovac, 1985; Wallace and Selman, 1985). More recent data suggest that
hydration in F. heteroclitus is primarily due to K+ fluxes via the gap junctions between
oocytes and follicle cells (Wallace et al., 1992; Cerd et al., 1993), but the possibility
of some contributions to hydration resulting from yolk cleavage has not yet been
abandoned. So far, complete Vtg sequences have been reported from no other teleosts
besides F. heteroclitus. However, as more sequences are completed, it will be


BIOGRAPHICAL SKETCH
Gary LaFleur, Jr. was bom June 11, 1963, in Eunice, La. He was raised in
Eunice with his three sisters, Michelle, Holly, and Claire, and attended St. Edmund
Elementary and High School, as had his parents: Amanda Lee Rozas and Gary J.
LaFleur, M.D. After high school he attended Louisiana State University at Eunice for
two years, receiving an Associate of Science degree (1983), and then completed his
Bachelor of Science at L.S.U. in Baton Rouge (1984). His first encounter with the study
of estuarine teleosts was under the tuteledge of Dr. John Sharp during an Ichthyology
course offered at the Gulf Coast Research Laboratory in Ocean Springs, MS. After
completing his B.S. he became involved with the Ocean Research and Education Society,
in Gloucester, MA, first as a student, and then as a teaching assistant.
In 1986, he accepted a position as a technician in the lab of Peter Thomas at the
University of Texas Marine Science Institute in Port Aransas, TX. Under the mentorship
of Thomas, LaFleur began to study the repoductive physiology of sciaenid fishes. His
work as a technician was incorporated into a Master of Science degree from Corpus
Christi State University. Although LaFleurs research under Thomas consisted mainly
of experimental biochemistry, the field-oriented coursework at CCSU directed by J. Wes
Tunnel, left a lasting impression on him. Particularly infuential were field trips to the
141


129
Greeley, MS Jr, Hols H, Wallace RA (1991) Changes in size, hydration and low
molecular weight osmotic effectors during meiotic maturation of Fundulus oocytes
in vivo. Comp Biochem Physiol 100A:639-647
Germond JE, Walker P, ten Heggeler B, Brown-Luedi M, de Bony E, Wahli W. (1984)
Evolution of vitellogenin genes: Comparative analysis of the nucleotide sequences
downstream of the transcription initiation site of four Xenopus laevis and one
chicken gene. Nucleic Acids Res 12:8595-8609
Guraya SS (1986) The cell and molecular biology of fish oogenesis. Monographs in
developmental biology, Vol 18 Karger, Basel
Guillette U Jr, Gross TM, Masson GR, Matter JM, Percival HF, Woodward AR (1994)
Developmental abnormalities of the gonad and abnormal sex hormone
concentrations in juvenile alligators from contaminated and control lakes in
Florida. Environ Health Perspect 102:680-688
Hagenmaier HE (1985) The hatching process in fish embryos VIII. The chemical
composition of the trout chorion (zona radiata) and its modification by the action
of the hatching enzyme. Zool Jb Physiol 89:509-520
Hamazaki T, Iuchi I, Yamagami K (1984) Chorion glycoprotein-like immunoreactivity
in some tissues of adult female medaka. Zool Sci 12:148-150
Hamazaki T, Iuchi I, Yamagami K (1985) A spawning female-specific substance reactive
to anti-chorion (egg envelope) glycoprotein antibody in the teleost, Oryzias
latipes. J Exp Zool 235:269-279
Hamazaki TS, Iuchi I, Yamagami K (1987a) Isolation and partial characterization of a
"spawning female-specific substance" in the teleost, Oryzias latipes. J Exp. Zool
242:343-349
Hamazaki TS, Iuchi I, Yamagami K (1987b) Production of a "spawning female-specific
substance" in hepatic cells and its acumulation in the ascites of the estrogen-
treated adult fish, Oryzias latipes J Exp Zool 242:325-332
Hamazaki TS, Nagahama Y, Iuchi I, Yamagami K (1989a) A glycoprotein from the liver
constitutes the inner layer of the egg envelope (zona pellucida interna) of the fish
Oryzias latipes. Dev Biol 133: 101-110
Hamazaki TS, Selman K, Wallace RA (1989b) Major components of the vitelline
envelope of the fish, Fundulus heteroclitus. Develop Growth Differ 31:407


43
gnathostomes, over 400 million years ago (Lovtrup, 1977). Chen et al. recently
described a mosquito cDNA sequence which codes for a Vtg containing three separate
polyserine domains; 82% of the serines in these domains are coded for by the TCX
codon. Since insects are a highly derived group, it remains unclear whether the TCX
repeats represent the conservation of a primitive polyserine coding domain or an
incidence of convergent evolution between separate Vtg clades.
It has been theorized that the phosphoserine clusters of Pv, documented to bind
Ca++ in a 1:1 stoichiometric ratio in Xenopus (Folletand Redshaw, 1968; Munday et al.,
1968; Wallace, 1970) are necessary for early bone mineralization in vertebrate embryos.
Even more speculative is the idea that the phosphoserine tracts of Vtg were a necessary
pre-adaptation allowing the original evolutionary emergence of ossified bone in ancestral
chordates. Both the lamprey and the sturgeon are examples of cartilaginous vertebrates,
albeit with bony ancestors (Jarvik, 1980), that have retained their Vtg polyserine
domains. Thus, the possession of a Vtg polyserine domain is not universally concomitant
with the possession of a bony skeleton. Indeed, it appears that polyserine domains can
no longer be considered an exclusive vertebrate Vtg characteristic. Recent reports by
Chen et al. (1994) describing a mosquito (Aedes aegypti) Vtg cDNA and Yano et al.
(1994) describing a silkworm (Bombyx mori) Vtg cDNA, provide invertebrate sequences
containing various arrangements of polyserine tracts. These findings suggests a pre-
chordate origin of Vtg polyserine domains and challenges the hypothesis of Pvs being
unique to chordates. However, polyserine domains are not synonymous with true Pv
domains. Whether these invertebrate polyserine tracts are highly phosphorylated and


I certify that I have read this study and that in
acceptable standards of scholarly presentation and i* full
as a dissertation for the degree of Doctor of Philos*
This dissertation was submitted to the Graduate Faculty of the College of
Medicine and to the Graduate School and was accepted as partial fulfillment of the
requirements for the degree of Doctor of Philosophy.
May, 1996
K
ean, College of Medicine
Dean, Graduate School


116
Ohzu and Kusa, 1981; Kobayashi, 1982; Begovac and Wallace, 1989; Hyllner et al.,
1991, 1995), the extensive PQX repeat is nonetheless extraordinary. The finding that
(Pro-Glx-X) peptides were specifically released from the lysed chorions of medaka (Lee
et al., 1994) offers evidence consistent with the notion that components of Chg 500 and
553 contribute to the structure of the hardened chorion. Insights into the mechanism of
chorion hardening have been provided by the studies of Hagenmaier et al. (1985) and
Oppen-Bemtsen et al. (1990) in which Glx-Lys crosslinks were discovered in the
chorions of fertilized but not unfertilized eggs. A somewhat similar crosslinking
phenomenon has been suggested from the proline-rich repeats of mussel adhesive
proteins, where highly repetitive motifs containing hydroxyproline, lysine and tyrosine
(also modified to 3,4-dihydroxyphenylalanine) are involved in the formation of
underwater adhesives (Rzepecki et al., 1991). The exact mechanism whereby the
vitelline envelope of F. heteroclitus is hardened into a rigid chorion remains a mystery;
however we suggest that the high content of Pro, Gin, Lys, and Tyr found within the
PQX repeating region of Chg 500 and 553 are likely to play significant roles in this
process.
The predicted structure of Chg 427
The shortest of the three sequences reported here is that of Chg 427. It does not
contain an extensive repeating region as do the other Chgs; however the short sequence
(PGK PSK PQS PPT QNQ QQL Q) contains the high proline and glutamine content
characteristic of the repeats of Chgs 500 and 553. Chg 427 is the only sequence of the


137
Terpstra P, AB G (1988) Homology of Drosophila yolk proteins and the triacylglycreol
lipase family. J Mol Biol 663-665
Tesonero JV (1977) Formation of the (zona pellucida) in the teleost, Oryzias latipes. I.
Morphology of early oogenesis. J Ultrastruct Res 59:282-191
Tesonero JV (1978) Formation of the chorion (zona pellucida) in the teleost, Oryzias
latipes. HI. Autoradiography of [3H] proline incorporation. J. Ultrastruct Res
64:315-326
Thiebaud CH, Fischberg M (1977) DNA content in the genus Xenopus. Chromosoma
59:253-257
Thorsen A, Fyhn HJ, Wallace RA (1993) Free amino acids as osmotic effectors for
oocyte hydration in marine fishes. In: Walther BT and Fyhn HJ (eds)
Physiological and biochemical aspects of fish development. University of Bergen,
Norway
Trewitt PM, Heilmann U, Degrugillier SS, Kumaran AK (1992) The boll weevil
vitellogenin gene: nucleotide sequence, structure, and evolutionary relationship
to nematode and vertebrate vitellogenin genes. J Mol Evol 34:478-492
Urist MR, Schjeide OA, Mclean FC (1958) The partition and binding of calcium in the
serum of the laying hen and of the estrogenized rooster. Endocrinol 63:570-585
Urist MR, Schjeide AO (1961) The partition of calcium and protein in the blood of
oviparous vertebrates during estrus. J Gen Physiol 44:743-756
van het Schip F, Samallo J, Broos J, Ophuis J, Mojet M, Gruber M, and AB G (1987)
Nucleotide sequence of a chicken vitellogenin gene and derived amino acid
sequence of the encoded yolk precursor protein. J Mol Biol 196:245-260
von Heijne G (1986) A new method for predicting signal sequence cleavage sites.
Nucleic Acids Res 14:4683-4690
Wahli W. (1988) Evolution and expression of vitellogenin genes. TIG 4:227-232
Wahli W, Dawid IB, Wyler T, Jaggi RB, Weber R, Ryffel GU (1979) Vitellogenin in
Xenopus laevis is encoded in a small family of genes. Cell 16:535-549
Wallace RA (1978) Oocyte growth in nonmammalian vertebrates In: Jones RE (ed) The
Vertebrate ovary. Plenum Publishing, New York, pp 469-502


CHAPTER 5
FUNDULUS HETEROCLITUS CHORIOGENINS: LIVER-DERIVED
COMPONENTS OF THE VITELLINE ENVELOPE AND CHORION
SHARING SEQUENCE IDENTITY WITH MAMMALIAN ZP PROTEINS
Introduction
The spawned eggs of the estuarine teleost Fundulus heteroclitus are exposed to
quite a different environment than the ovulated eggs of mammals. Whereas mammalian
eggs are protected from infection, desiccation, and predation by the safe surroundings
of the uterus, F. heteroclitus eggs are released and fertilized during the tumultuous spring
tides, and deposited into empty mussel shells or onto the leaves of marsh grass, where
they remain actually stranded above the water line for fourteen days until the embryos
emerge by hatching during the next spring tide (Taylor et al., 1977; Hsiao et al., 1994).
Though exposed to extremely different environments, both of these vertebrate eggs are
protected by a quasi-similar layer of extracellular matrix (ECM). In mammals this
translucent layer of ECM is termed the zona pellucida (ZP), but in fish and many other
invertebrates it is often referred to as the vitelline envelope or chorion.
In this paper we adhere to the definitions of Dumont and Brummett (1980)
regarding the vitelline envelope and chorion. They stated that the term "vitelline
envelope" referred to the highly structured acellular layer that appears and encloses the
87


139
Wallace RA, Greeley MS Jr, McPherson R (1992) Analytical and experimental studies
on the relationship between Na+, K+, and water uptake during volume increases
associated with Fundulus oocyte maturation in vitor. J Comp Physiol B 162:241-
248
Wang S, Smith DE, Williams DL (1983) Purification of avian vitellogenin III:
Comparison with vitellogenins I and II. Biochemistry 22:6206-6212
Wassarman PM (1988a) Zona pellucida glycoproteins. Ann Rev Biochem 57: 415-442
Wassarman PM (1988b) Fertilization in mammals. Sci Amer 256:78-84
Watanabe WO, Kuo C-M (1986) Water and ion balance in hydrating oocytes of the grey
mullet, Mugil cephalus (L.) during hormone-induced final maturation. J. Fish
Biol 28:425-337
White HB III (1987) Vitamin-binding proteins in the nutrtion of the avian embryo. J Exp
Zool Supplement 1:53-63
White HB III, Merrill AH Jr (1988) Riboflavin-binding proteins. Ann Rev Nutr 8:279-
299
Wiley HS, Wallace RA (1981) Multiple vitellogenins in Xenopus laevis give rise to
multiple forms of the yolk proteins. J Biol Chem 256:8626-8634
Wray W, Boulikas T, Wray VP, Hancock R (1981) Silver staining of proteins in
polyacrylamide gels. Anal Biochem 118:197-203
Yamagami K (1960) Phosphorous metablolism in fish eggs II. Transfer of some
phosphorous compounds from egg yolk into embryonic tissues in Salmo irideus
during development. Sci Papers Coll Gen Ed Univ Tokyo 11:153-161
Yamamoto M (1963) Electron microscopy of fish develpment. II. Oocyte-follicle cell
relationship and formation of chorion in Oryzias latipes. J Fac Sci Univ Tokyo
10:123-127
Yamamoto M, Yamagami K (1975) Electron microscopic studies on choriolysis by the
hatching enzyme ofthe teleost, Oryzias latipes. Dev. Biol 43:313-321
Yamamura J, Adachi T, Aoki A, Nakajima H, Nakamura R, Matsuda T (1995)
Precursor-product relationship between chicken vitellogenin and the yolk proteins:
the 40 kDa yolk plasma glycoprotein is derived from the C-terminal cysteine-rich
domain of vitellogeinin II. Biochem Biophys Acta 1244: 384-394


35
Fundulus
Callus
Xenopus
Acipenser
Ichlhyomyzon
Fundulus
Gallus
Xenopus
Acipenser
Ichiliyomyzon
Fundulus
Gallus
Xenopus
Acipenser
Ichthyomyzon
Fundulus
Gallus
Xenopus
Acipenser
Ichlhyomyzon
Fundulus
Gallus
Xenopus
Acipenser
Ichlhyomyzon
Fundulus
Gallus
Xenopus
Acipenser
Ichthyomyzon
Fundulus
Gallus
Xenopus
Acipenser
Ichlhyomyzon
Fundulus
Gallus
Xenopus
Acipenser
Ichlhyomyzon
Oncorhynchus
KK- -YCAKHIGVGLKACFKFASQNGASIQDIVLYKLAGSHNFS FSVTPIEGE- -WERLE
RD--ICLKMHHIGCQLCFSRRSRDASFIQNTYLHKLIGEHEAKIVLMPVHT-DADIDKIQ
YD--ACTKFSKAGVHLCIQCKTHNAASRRNTIFYQAVGEHDFKLTMKPAHT-EGAIEKLQ
RS--ACAKAKNFGFEVCFEGKSENVAFLRDSPLYKIIGQHHCKIALKPSHSSEATIEKIQ
RRKQHCSKSAALSSKVCFSARLRNAAFIRNALLYKITGDYVSKVYVQPT-SSKAQIQKVE
MEVKVGAKAAEKLVKRINI.SEDEETEEG--GPVLVKLNKI
LEIQAGSRAAARIITEVNPESEEEDESSPYEDIQAKliKRILGIDSMFKVANKTRHPKNRP
LEITAGPKAASKTMGLVEVEGTEGEPMDE-TAVTKRLKMILGIDESRKDTNETALYRSKQ
LELQTGNKAASKI IRWAMQSLAEADEMK GNI LKKNKLItVDGE
ELQAGPQAEKVIRMVELVAKASKKSKKNSTITEEGVGETIISQLKKILSSDKDK
>===POLYSERINE DOMAINss
SKKGNTVLAEFGTEPDAKTSSSSSSASSTATSSASSSASSPNRKKPMDEEENDQVKQARN
KKKNKI HNRRLDAE WEARK
T
DAKKPPGSSSSSSSSSSSSSSSSSSDKSGKKTPRQGSTVNLAAKR
====~=============s==POLYSERINE DOMAIN=====================
SSRRNSSSSSSSSSSSSSESRSSRSSSSSSSSSRSSRKIDLAARTNSSSSSS
KDASSSSRSSKSSNSSKRSSSKSSNSSKRSSSSSSSSSSSSRSSSSSSSS SSNSK
QQSSLSSSSSSSSSSSSSSSSSSSSSSSSSPSSSSSSSYSKRSKRREHNPHHQRESSS-S
QDSTLRGFKRRSSSSSSS3SSSSSSSSSSSSSSSQQSRMEKRMEQDKLTENLERDRDHMR
AS KKQRGKDSSSSSSSSSSSSDSSKSPHK- -HGGAKRQHAGHGAPHLGPQSHSSSSSSSS
s=======2=============POLYSERINE DOMAIN=====================
SRRSRSS- SSSSSSSSSSSSSSSSSSRRSSSSSSSSSSSSSRSSRR
SSSSSSKSSSSSSRSRSSSKSSSSSSSSSSSSSSKSSSSRSSSSSSKSSSHHSHSHHSGH
SSQEQNKKRNLQENRKHGQKGMSSSSSSSSSSSSSSSSSSSSSSSSSSSSEENRPHKNRQ
GKQSKNKKQEWKNKQKKHHKQLPSSSSSSSSSSSGSNSSSSSSSSSSSSS---RSHNHRN
SSSSSSASKSFSTVKPPMTRKPRPARSSSSSSSSDSSSSSSSSSSSSSSSSSSSS
======s==!3============POLYSERINE DOMAIN=5=s=================
VNSTRSSSSSSRTSSASSLASFFSDSSSSSSSSD RRSKEVME-KFQRLHK-K
LNGSSSSSSSSRSVSHHSHEHHSGHLEDDSSSSSSSSVLSKIWGRHEIYQYRFRSAHR-Q
HDNKQAKMQSNQHQQKKNKFSESSSSSSSSSSSEMWNKKKHHRNFYDLNFRRTAR-T
NTRTLSK SKRYQNNNNSSSSSGSSSSSEEIQKNPEI FAYRfRSHRD K
SSSSKSEEWLAVKDVNQSAFYNFKYVPQRKPQ
===ss===========3==ss=POLYSERINE DOMAINh=======~===5=======
MV ASGSSASSVEAIYKEK KYLGEEEA- WAVILRAVKADKRMV
EFPKRKLPGDRATSRYSSTRSSHDTSRAASWPKFLGDIKTPVLAAFLHGISNNKKTG
KGTEHRGSRLSSSSESSSSSSESAY RHKA KFLGDKEPPVLWTFKAVRNDNTKQ
LGFQNKRGRMSSSSSSSSSSSSQSTLNSKQDA KFLGDSSPPIFAFVARAVRSDGLQQ
TSRRHTPASSSSSSSSSSSSSSSSSSSDSDMTVSAESFEKHSKPKWIVLRAVRADGKQQ
============POLYSERINE DOMAINh==========<
GYQLGFYLD KPNARVQIIVANISSDSNWRICADAWLSKHKVTTKISWGEQCRKYST
GLQLWYAD-TDSVRPRVQVFVTNLTDSSKWKLCADASVRNAPQAVAYVKWGWDCRDYKV
GYQMWYQE YHSSKQQIQAYVMDI -SKTRWAACFDAVWNPHEAQASLKWGQNCQDYKI
GYQVAAYTD-NRVSRPRVQLLATEIIEKSRWQICADAILASNY KAMALMRWGEECQDYKV
GLQTTLYYGLTSNGLPKAKIVAVELSDLSVWKLCAKFRLSAUMKAKAAIGWGKNCQQYRA
LGRPKTTSDEPNIITAALDENDNWKLCADGVLLSKHKVNKIAWGAGCKDYNT
1036
1045
1044
1039
1067
1075
1105
1103
1084
1123
1075
1165
1123
1085
1168
1127
1220
1182
1145
1226
1172
1280
1242
1202
1281
1222
1339
1298
1249
1314
1264
1396
1352
1306
1374
1321
1455
1410
1365
1434
53
Figure 2.4--continued


Vtg :
Vtg II
MXAWLALTLAFVAGQN F APEFAAGXTYVYXYEAL ILGGLPSEGI.ARA
MRVLVLALTVAI.VAGNQVS1APEFAPGXTYEYKYEG7II.GGI.PSEGLAXA
48
50
Vtg I
Vtg II
GLXISTXLLLSAADQNTYMLXLVEPELSEYSGIWPKDPAVPATXLTAALH 98
GVKIQSKVLIGAAG? D S YILXLEDPVISGYSGIWPK2VFHPATXLTSALS 100
IT* *#*
Vtg I
Vtg II
LSSQFPSSLJJTPMVFVGXVFAPEEVSTLVLNIYRGIUJILQLNIXXTHKV 148
AQLL TP VKFS Y ANG VIGKVFAP PGISTNVLNVFRGLLNMFQMNIXXTQNV 150
****** * * ** ** ***** *
Vtg I
Vtg II
YDLQEVGTQGVCKTLYSISEDARISNILLTX7RDI.SNCQERLNKDIGEA
YDLQETGVXGVCXTHYILHEDSXADRI.HI.TXTTDLNHCTDSIHMDVGMAG
197
200
Vtg I
Vtg 13
YTSXCORCQESTXNLRGTTTLSYVLXPVADAVMII.XAYVNEI.IQFSPFSE
YTSXCAECMARGXTLSGAISVNYIMXPSASG7LILEATATE2I.QYSPVNI
247
250
Vtg I
Vtg II
AIIGAAQMRTXQSI.2FLSIEXEPIPSVKAEYRHRGSLXYEFSDELI.QTPI.Q 297
VNGAVQMEAXQTVTFVDIRKTPL3PLXADYIPRGSLXYELG7EFLQTP IQ 300
* * ******* *** *
* * * *
Vtg I
Vtg II
LIXISDAPAQVAE'/LXHLATYNIEDVHENAPLXFLEI.VQLLRIARYEDLS
LLRITNVEAQIVESLNNLVSLHMGHAHEDSPLXFIELIQLLRVAXYESIS
347
350
Vtg I
Vtg II
MYWNQYKXMSPHRHWFLDTIPATG7FAG.RFIXEXFMAEEITIAEAAQAF 397
ALWSQFKTXIDHRHWLLSSIPAIGTHVALXFIXEXIVAGEVTAAEAAQAI 400
* * **** *** ** ***** * ******
Vtg I
Vtg II
17AVHMVTAD P EVIXLFE SL VO S D XWENPLUIE WFLG Y GTMVNXY CUR
MSSTHLVKADLEAIXLQEGI.AVTPNIRENAGI.REI.VMLGFGIMVHXYCVE
447
450
Vtg I
Vtg II
TVDCPVELIXPIQQRLSDAIAXNEEENIILYIXVLGNAGHPSSFXSLTXI 497
NPSCPSELVRPVHDIIAXALZXRDNDELSLALXVLGNAGHPSSLXPIMKL 500
* TT # * ****** *
Vtg I
Vtg II
MPIHG7AAVSLPMTIHVEAIMAI.RNIAXXESRMVQELALQLYMDKALHPS 547
LPGFGSSASELELRVHIDATLALRRIGXREPKMIQDVALQLFMDRTLOPS 550
* * * *** * *
* * *
Vtg I
Vtg II
LRMLSCIVX.FE7SPSMGLVTTVANSVKTSENLQVASF7YSHMXSLSRSPA
I.RMVAWVLFDTXLPMGLITTLAQSI.I.X2PNI.QVLSFVYSYMKAFTXTTT

597
600
Vtg I
Vtg II
71HPDVAAACSAAMXILGTXLD RLSLRYSXAVHVD L YNS SLAVGAAATAF 647
POHSTVAAACrrVAIRILSPRFSRLSYRYSRAFHYDHYHNPWMLGAAASAF 650
* **** ** * ** # *
Vtg I
Vtg II
YINDAATFMPXSFVAXTXGFIAGSTAEVLEIGANIEGIQELILXNPALSE
YIIIDAAT/I.PKNIHAXARVYLSGVSVDVLEFGARAEGVQEALIXARDVPE
697
700
Vtg I
Vtg II
S70RI7XMKRVIXALSEWRSLPTSXPEASVYVRFFGQEIGFANIDXPMID 747
SADRLTXMXQALXALTEWRANPSRQPLGSLYVKVLGQDVAFANIDREMVE 750
** ** ** ** ** *** * ****** *
Vtg I
Vtg II
XAVKFGXELPIQEYGREAI.XALLLSGINFHYAXPVIAAEMRRILPTVAGI 797
XIIEFATGPEIRTRGXXALDALI,-SGYSMXYSXPMSAIEVRHIFPTSI.GL 799
* * ** *** ** * * *
Vtg I
Vtg II
PMELSLYSAAVAAASVEIXPNTSPRLSADFDVKTLI.ETDVELXAEIRPMV
PMELSI/YTAAVTAASVEVQATISPPLPEOFHPAHLLXSDISMXASVTPSV
847
849
Vtg I
Vtg II
AMDTYAVMGLNTDIFQAALVARAKLHSWPAXIAARLNIKEGDFKLEALP 397
SLHTYGVMGVNSPFIQASVLSRAXDHAALPKXMEARLDIVKGYFSYQFI.? 899


127
Chen J-S, Cho W-L, Raikhel AS (1994) Analysis of mosquito vitellogenin cDNA,
similarity with vertebrate phosvitins and arthropod serum proteins. J Mol Biol
237:641-647
Clark RC (1973) Amino acid sequence of a cyanogen bromide cleavage peptide from
hens egg phosvitin. Biochim Biophys Acta 310:174-187
Conte FS, Doroshov SI, Lutes PB, Strange EM (1988) In: Hatchery manual for the
white sturgeon. Publication 3322. The Regents of the University of California
Division of Agriculture and Natural Resources, Oakland CA
Cozens PJ, Cato AC, Jost JP (1980) Characterization of cloned complementary DNA
covering more than 6000 nucleotides (97%) of avian vitellogenin mRNA. Eur
J Biochem 112:443-450
Craik JCA, Harvey SM (1984) Phosphorous metabolism and water uptake during final
maturation of ovaries of teleosts with pelagic and demersal eggs. Mar Biol
90:285-289
Craik JCA, Harvey SM (1986) The causes of buoyancy in eggs of marine teleosts. J Mar
Biol Assoc 67:169-182
Denhardt DT (1966) A membrane-filter technique for the detection of complementary
DNA. Biochem Biophys Res Commun 23:641-646
Ding JL, Hee PL, Lam TJ (1989) Two forms of vitellogenin in the plasma and gonads
of male Oreochromis aureus. Comp Biochem Physiol 93B:363-370
Ding JL, Ho B, Valotaire Y, LeGuellec K, Lim EH, Tay SP, Lam TJ (1990) Cloning,
characterization and expression of vitellogenin gene of Oreochromis aureus
(Teleostei, Cichlidae). Biochem Int 20:843-852
Dodson RE, Acea MR, Shapiro DJ (1995) Tissue distribution, hormone regulation and
evidence for a human homologue of the estrogen-inducible Xenopus laevis
vitellogenin mRNA binding protein. J Steroid Biochem Molec Biol 52:505-515
Dumont JN, Brummett AR (1980) The vitelline envelope, chorion and micropyle of
Fundulus heteroclitus eggs. Gamete Res. 3:25-44
Dumont JN, Brummett AR (1985) Egg envelopes in vertebrates. In: Browder LW (ed)
Developmental biology. Vol 1. Plenum Press, New York pp 235-278
Eckelbarger KJ (1994) Diversity of metazoan ovaries and viellogenic mechanisms:
implications for life history theory. Proc Biol Soc Wash 107: 193-218


58
Nuc-Trap columns. All RNA hybridizations were carried out at 65C in 1 X Denhardts
solution, 6 X SSC, and 0.1 % SDS without formamide (Denhardt 1966).
Autoradiographs were analyzed using the Bio Image Whole Band Analyzer system
(Millipore, Ann Arbor). For estimating amounts of Vtg RNA visualized on gels, RNA
was transcribed from the Vtg I plasmid pMMBl and the Vtg II plasmid pFhv2a, using
Ambion reagents. Transcribed RNA yields were measured spectrophotometrically, and
diluted to a concentration of 66.7 pg//xl. For RNA standards, 133 pg transcribed RNA
from both pMMBl and pFhv2a was loaded onto each gel.
Results
The complete cDNA sequence (5166 bp) of a Vtg mRNA, encoding a protein
designated as Vtg II is provided in Figure 3.2. The eight overlapping pGem-T clones
that were used to complete the sequence are represented in Figure 3.1. A ClustalV
alignment of Vtgs I and II by the method of Swofford et al. (1993) revealed 45%
sequence identity between the two amino acid sequences (Fig. 3.3).
In general, the two sequences share the same profile as other reported Vtgs: a
large lipovitellin 1 region that is followed by a polyserine domain (assumed to represent
phosvitin) that, in turn, is followed by a lipovitellin 2 region containing a
substantial amount of conserved cysteines. Like Vtg I, Vtg II contains several predicted
N-glycosylation (16), phosphorylation (45), and N-myristoylation sites (16), agreeing
with our expectations for a lipophosphoglycoprotein. The smaller length of the Vtg II
a.a. sequence (1687) compared to that of Vtg I (1704) can be primarily attributed to gaps
in the polyserine domain. A graphical comparison of the polyserine domains of Vtg


117
three Chgs that contains a predicted N-glycosylation site. Rather than sharing identity
with Chgs 500 and 553 of F. heteroclitus, Chg 427 is most similar to that of the medaka
L-SF protein, followed closely by three carp "ZP3" sequences (not shown). These five
fish sequences contain ZP domains that share highest identity to the mouse ZP3
subfamily of mammalian ZPs (Fig. 5.8). Because the mouse ZP3 subfamily of molecules
is implicated as the primary sperm receptor in mammals, it is tempting to postulate a
similar role for Chg 427 as a likely candidate for sperm interaction in F. heteroclitus.
However, this postulation is significantly hindered by the well established existence of
a micropyle on F. heteroclitus eggs (Dumont and Brummett, 1980; Selman and Wallace,
1986). The micropyle of teleost eggs is essentially a narrow channel through the chorion
that provides homologous spermatozoa with direct access to the oocyte membrane
(Dumont and Brummett, 1980; reviews by Guraya, 1986; Hart 1990). Its existence
dismisses the necessity for most of the fertilization-associated interactions that have been
documented to occur in urchins and mammals, including sperm binding, induction of the
acrosome reaction, and the burrowing of sperm through the ZP en route to the oocyte
surface. Therefore, although Chg 427 shares identity with ZP3 molecules, it remains
unclear as to what function it fulfills.
The ZP Family of Proteins
A recent review by Harris et al. (1994) has attempted to lend order to the
currently confusing ZP nomenclature by separating all ZP proteins into three groups:
ZPA; ZPB; and ZPC, according to comparisons by protein alignments. Their ZPA


108
to the mouse sequence that they contain, thus the ZP1, ZP2, and ZP3 subdivisions. The
fish sequences represented in the tree were separated into two major subdivisions: one
containing Chg 427, medaka L-SF, and the three carp sequences, that was grouped with
the mouse ZP3 subdivision; and another containing the Chg 500, Chg 553, and the
flounder ZP that was grouped with the mouse ZP1 subdivision. Of the eight fish
sequences analyzed, none showed significant relatedness with the mammalian ZP2
subdivision; however, bootstrap values at the node dividing the ZP2 and ZP3
subdivisions were the lowest on the tree, arguing against a weighted interpretation
concerning this delineation.
Northern Blot Analysis
Northern blot analysis using three separate random-primed [32P]probes for Chg
550, 427, and 553 revealed Chg mRNAs present in liver RNA from both estrogen-treated
males and spawning females (Fig. 5.5). Furthermore, when 20.0 ig of ovary RNA was
blotted next to 2.0 /ug of liver RNA, Chg transcripts were apparent only in RNA from
the liver (Fig 5.6).
Vitelline envelope proteins
VEPs were isolated from ovarian follicles and resolved by SDS-PAGE into three
major Coomassie blue-staining bands at estimated molecular weights of 69,000, 60,000,
and 46,000, designated as VEP 69, VEP 60, and VEP 46, respectively (Fig. 5.7). At
least one other band could be visualized between VEP 60 and VEP 69, but appeared too


126
Bidwell CA, Carlson DM (1995) Characterization of vitellogenin from white sturgeon
Acipenser transmontanas. J Mol Evol 41:104-112
Bleil JD, Wassarman PM (1980) Structure and function of the zona pellucida:
Identification and characterization of the proteins of the mouse oocytes zona
pellucida. Dev Biol 76:185-202
Blumenthal T, Squire M, Kirtland S, Cane J, Donegan M, Speith J and Sharrock W
(1984) Cloning of a yolk protein gene family from Caenorhabditis elegans. J Mol
Biol 174:1-18
Bork P, Sander C (1992) A large domain common to sperm receptors (Zp2 and Zp3) and
TGF-/3 type III receptor. FEBS Lett 300:237-240
Bownes M (1992) Why is there sequence similarity between insect yolk proteins and
vertebrate liases? J Lipid Res 33:777-790
Brock ML, Shapiro DJ (1983) Estrogen regulates the absolute rate of transcription of the
Xenopus laevis vitellogenin genes. Cell 34:207-214
Byrne BM (1989) In: Phosvitin, an independent domain in vitellogenin genes. Ph.D
dissertation, University of Groningen, The Netherlands 122 pp
Byrne BM, van het Schip F, van de Klundert JAM, Amberg AC, Gruber M, and AB,
G (1984) Amino acid sequence of phosvitin derived from the nucleotide sequence
of part of the chicken vitellogenin gene. Biochemistry 23:4275-4279
Byrne BM, Gruber M, AB G (1989) The evolution of egg yolk proteins. Prog Biophys
Molec Biol 53:33-69
Caskey CT, Pizzuti A, Fu Y-H, Fenwick RG, Nelson DL (1992) Triplet repeat
mutations in human disease. Science 256: 784-789
Cerd JL, Petrino TR, Wallace RA (1993) Functional heterologous gap junctions in
Fundulus ovarian follicles maintain meiotic arrest and permit hydration during
oocyte maturation. Dev Biol 160:228-235
Chamberlin ME, Dean J (1990) Human homolog of the mouse sperm receptor. Proc Natl
Acad Sci USA 87:6014-6018
Chaudry HS (1956) The origin and structure of the zona pellucida in the ovarian eggs
of teleost. Z. Zellforschung 43:478-485


49
(LaFleur et al. 1995) In brief, the library was constructed from the pooled mRNA of
six male Fundulus heteroclitus that had been treated with two IP injections of estradiol-
17/3 (0.01 mg/g body weight). The library contained an initial titer of only 6 X 104 total
plaque-forming units, and had been amplified twice.
The initial Vtg II clone was discovered using the degenerate primer ROW 19, and
the vector primer NEB 1231, with 5 ¡A of the XgtlO library as template in a PCR
reaction utilizing a 50:1 mixture of Taq DNA polymerase:Pfu DNA polymerase. ROW
19 was designed to match a conserved region of Vtgs, ranging from C. elegans to
chicken (Fig. 3.1). A 550 bp band was isolated, inserted into pGem-T, sequenced and
revealed to be a second Vtg cDNA that we designated as Vtg II. This insert was then
isolated and used to generate a random primed 32P-labeled probe. The library was plated
out on 150-mm petri dishes by transfecting E. coli C600hfl cells, and overlaying them
in agarose atop agar plates containing 25 tg/ml tetracycline. Duplicate plaque lifts were
carried out using Magna nylon membranes, and these were probed at 65C in 0.05 X
BLOTTO, 6 X SSC (150 mM NaCl, 15 mM sodium citrate, pH 7) overnight. A large
proportion of the plaques were found to be positive, and 20 agarose plugs were isolated
and stored in SM buffer (. 1 M NaCl, 8 mM MgS04, 50 mM Tris, 2% gelatin) at 4C
with a drop of chloroform. Thereafter, plug lysates from these Vtg II positive plugs
were used in amplification reactions targeting Vtg II positive clones in a successively
overlapping 5 direction. Six more Vtg II clones were isolated in this manner
approaching the initial methionine codon, but several attempts at targeting the last few
nucleotides to include the initial methionine failed.


7
scheme mapping out the specific processing of two Vtgs into several separate yolk
protein products. We have further submitted a hypothesis implicating a PEST site found
within the predicted YP 125 sequence as a possible factor influencing its extensive
degradation. This study is presented in Chapter 4.
A New Emphasis: Estrogen-Induced Reproductive Proteins
While completing the Vtg II cDNA sequence, using a PCR-based screening
method, other non-target cDNAs were often isolated. This is a common phenomenon
in cloning that is usually dismissed as misfortune. However, because our template was
an estrogen-induced cDNA library, the non-target cDNAs stood a likely chance of
representing reproductively significant molecules. This was exactly the case concerning
the Chgs. All three of the Chgs cDNAs were isolated by a fortuitous mis-priming event
that occurred while screening for Vtg II cDNAs (Fig. 1.3). Only recently had a
hypothesis been submitted that ascribed the origin of the major proteins of the teleost
vitelline envelope to the estrogen-induced liver (Hamazaki et al., 1987b). This ran
counter to the mammalian literature that had established the oocyte as the primary site
of synthesis for the proteins of the mammalian zona pellucida (Wassarman, 1988a).
Nevertheless, our data verifies that several Chgs are in fact synthesized by the liver,
transported to the ovary, and laid down as the vitelline envelope between the oocyte and
the follicle cells. These proteins have been referred to by several names. When isolated
from the ovarian follicle, they are usually called vitelline envelope proteins
(VEPs)(Hyllner et al., 1991). Another nomenclature based on isolating the proteins from


26
2215
GCT
AAC
ATT
GAC
AAA
CCC
ATG
ATC
GAT
AAG
GCT
GTC
AAG
TTT
GGC
AAG
GAA
TTA
A
N
I

K
P
M
I
D
K
A
V
K
F
G
K
E
L
2269
CCC
ATT
CAG
GAA
TAT
GGA
AGA
GAG
GCT
CTC
AAG
GCT
CTG
CTC
CTG
TCT
GGC
ATC
P
I
Q
E
Y
G
R
E
A
L
K
A
L
L
L
S
G
I
2323
AAC
TTC
CAC
TAC
GCT
AAG
CCA
GTG
CTG
GCT
GCT
GAG
ATG
CGA
CGC
ATT
CTT
CCT
N
F
H
Y
A
K
P
V
L
A
A
E
M
R
R
I
L
P
2377
ACC
GTC
GCT
GGT
ATT
CCA
ATG
GAA
CTC
AGT
CTG
TAC
AGT
GCT
GCT
GTG
GCT
GCA
T
V
A
G
I
P
M
E
L
S
L
Y
S
A
A
V
A
A
2431
GCC
TCT
GTT
GAA
ATC
AAG
CCC
AAC
ACG
TCA
CCA
CGT
CTG
TCA
GCG
GAC
TTC
GAC
A
S
V
E
I
K
p
N
T
S
P
R
L
S
A
0
F
D
2485
GTA
AAG
ACT
CTG
CTG
GAG
ACA
GAC
GTT
GAG
CTC
AAG
GCT
GAG
ATC
AGA
CCA
ATG
V
K
T
L
L
E
T
D
V
E
L
K
A
E
I
R
P
M
2539
GTT
GCC
ATG
GAC
ACA
TAT
GCC
GTT
ATG
GGA
CTT
AAC
ACC
GAC
ATC
TTC
CAG
GCT
V
A
M
0
T
Y
A
V
M
G
L
N
T
D
I
F
Q
A
2593
GCT
TTG
GTA
GCT
CGC
GCT
AAA
CTG
CAC
TCT
GTT
GTG
CCA
GCC
AAA
ATA
GCT
GCA
A
L
V
A
R
A
K
L
H
S
V
V
P
A
K
X
A
A
2647
AGA
CTT
AAT
ATC
AAA
GAG
GGT
GAC
TTT
AAG
CTT
GAA
GCT
CTT
CCT
GTT
GAT
GTG
R
L
N
I
K
E
G
0
F
K
L
E
A
L
P
V
D
V
2701
CCT
GAA
AAC
ATC
ACA
TCC
ATG
AAT
GTT
ACA
ACC
TTT
GCT
GTA
GCA
AGA
AAC
ATC
P
E
N
I
T
S
M
N
V
T
T
F
A
V
A
R
N
I .
2755
GAG
GAA
CCT
TTG
GTT
GAG
AGA
ATC
ACT
CCT
CTT
CTC
CCC
ACC
AAA
GTT
TTG
GTA
E
E
P
L
V
E
R
I
T
P
L
L
P
T
K
V
L
V
2809
CCC
ATC
CCA
ATC
AGG
AGA
CAC
ACA
TCC
AAG
CTT
GAT
CCC
ACT
CGC
AAT
AGC
ATG
p
I
P
I
R
R
H
T
S
K
L
D
P
T
R
N
S
M
2863
TTA
GAC
TCC
TCA
GAA
CTC
CTT
CCC
ATG
GAA
GAA
GAA
GAT
GTA
GAG
CCC
ATT
CCT
L
0
S
S
E
L
I>
p
M
MU
; s
v E i'
D
V
E
p
I
P
(Ah = 2.25)
2917
GAA
TAC
AAG
TTC
CGT
CGA
TTT
GCC
AAA
AAG
TAC
TGC
GCT
AAG
CAC
ATT
GGT
GTT
E
Y
K
F
R
R
F
A
K
K
Y
C
A
K
H
I
G
V
2971
GGA
CTG
AAG
GCC
TGT
TTC
AAG
TTT
GCC
AGT
CAA
AAT
GGA
GCC
TCC
ATC
CAA
GAC
G
L
K
A
C
F
K
F
A
S
Q
N
G
A
S
I
<2
0
3025
ATT
GTC
CTG
TAC
AAA
CTG
GCT
GGT
AGC
CAC
AAC
TTC
TCT
TTC
TCT
GTG
ACA
CCA
I
V
L
Y
K
L
A
G
S
H
N
F
S
F
S
V
T
P
3079
ATT
GAA
GGA
GAA
GTT
GTT
GAG
AGA
TTG
GAG
ATG
GAG
GTT
AAA
GTC
GGA
GCA
AAG
I
E
G
E
V
V
E
R
L
E
M
E
V
K
V
G
A
K
3133
GCT
GCA
GAG
AAG
CTT
GTT
AAA
CGC
ATC
AAC
CTG
AGT
GAG
GAC
GAA
GAA
ACT
GAA
A
A
E
K
L
V
K
R
I
N
L
S
B i
E
E T E
(Ah = 2.43)
3187
GAA
GGA
GGT
CCA
GTC
CTG
GTG
AAG
CTC
AAC
AAA
ATC
CTG
TCT
TCA
AGA
CGG
AAC
m
G
G
P
V
L
V
K
L
N
K
I
L
S
S
R
R
N
3241
AGC
TCC
TCA
TCT
TCC
TCC
TCC
AGC
TCC
AGC
AGC
TCT
TCT
GAG
AGC
CGT
TCT
TCA
S
S
S
S
S
S
S
S
S
S
S
S
S
E
S
R
S
S
3295
AGG
TCC
TCC
TCT
TCC
TCC
TCC
TCT
TCA
TCT
CGC
TCC
AGC
CGT
AAG
ATT
GAC
CTT
R
s
S
S
S
S
S
S
S
S
R
S
S
R
K
I
D
L
Figure 2.2continued


6
et al., 1989). We theorized that an interesting oviparous model may be provided by the
protochordate Branchiosotma floridae, the Florida lancelet.
The original aims of my project were, thus, to complete the F. heteroclitus Vtg
cDNA sequence, and thereafter use the piscine cDNA as a heterologous probe to isolate
phylogenetically primitive Vtgs. I succeeded in the former goal, and the results of that
work are provided in Chapter 2. I began screening a cDNA library synthesized from the
MRNA of spawning female amphioxus, B. floridae, by a PCR-based method that utilized
degenerate primers designed by aligning the currently known Vtg protein sequences.
Before long, a new Vtg cDNA was successfully isolated. However, the new Vtg was
isolated from the "control" F. heteroclitus library template, rather than the targeted
amphioxus library template (Fig. 1.3). At that time, no two Vtgs from one vertebrate
species had yet been completely sequenced, and so this appeared to be a worthwhile
challenge. Additionally, the sequence of two F. heteroclitus Vtgs would provide
information presumably necessary to continue mapping out the precursor-product
relationships of the Vtgs and the yolk proteins. As a result, phylogenetic aspects of Vtg
evolution were shelved in order to consider the variations of Vtg that might be
encountered from within one species, F. heteroclitus. The second primary aim of my
project was thus to complete the Vtg II cDNA; this data is provided in Chapter 3.
While completing the two Vtg cDNA sequences, N-terminal sequences of the yolk
proteins were also being obtained. This work represented a collaborative effort that
included data collected from three students at the Whitney lab (including myself) plus a
considerable effort by the ICBR Protein Core facility. Eventually we established a


138
Wallace, RA (1983) Interactions between somatic cells and the growing oocyte of
Xenopus laevis In: McLaren A, Wylie CC (eds) Current problems in germ cell
differentiation. Symposium of British society for developmental biology.
Cambridge University Press, Great Britain, pp 285-306
Wallace, RA (1985) Vitellogenesis and oocyte growth in nonmammalian vertebrates. In:
Browder LW (ed) Developmental Biology, vol 1. Plenum Press, New York, pp
127-177
Wallace RA, Begovac P (1985) Phosvitins in Fundulus oocytes and eggs. J Biol Chem
260:11268-11274
Wallace RA, Jared DW, Eisen AZ (1966) A general method for the isolation and
purification of phosvitin from vertebrate eggs. Can J Biochem 44:1647-1655
Wallace RA, Jared DW (1969a) Estrogen induces lipophosphoprotein in serum of male
Xenopus laevis. Science 160:91-92
Wallace RA and Jared DW (1969b) Studies on amphibian yolk. VII. Serum-
phosphoprotein synthesis by vitellogenic females and estrogen-treated males of
Xenopus laevis. Can J Biochem 46:953-959
Wallace RA, Morgan JP (1986a) Isolation of phosvitin: retention of small molecular
weight species and staining characteristics on electrophoretic gels. Anal Biochem
157:256-261
Wallace RA, Morgan JP (1986b) Chromatographic resolution of chicken phosvitin.
Biochem J 240:871-878
Wallace RA, Selman K (1980) Oogenesis in Fundulus heteroclitus II. The transition
from vitellogenesis to maturation. Gen Comp Endocrinol. 42:3454-354
Wallace RA, Selman K (1981) Cellular and dynamic aspects of oocyte growth in
teleosts. Amer Zool 21:325-343
Wallace RA, Selman K (1985) Major protein changes during vitellogenesis and
maturation of Fundulus oocytes. Dev Biol 110:492-498
Wallace RA, Camevali O, Hollinger TG (1990a) Preparation and rapid resolution of
Xenopus phosvitins and phosvettes by hight-performance liquied chromatography.
J Chromatography 519:75-86
Wallace RA, Hoch KL, Camevali O (1990b) Placement of small lipovitellin subunits
within the vitellogenin precursor in Xenopus laevis. J Mol Biol 213:407-409


29
F. heteroclitus Vtg Sequence
A conceptual translation of the 5112 bp open reading frame resulted in a 1704-
amino acid protein sequence (Fig. 2.2). A signal peptide was predicted (underlined) by
aligning the F. heteroclitus sequence with the N-terminal sequences of several other
piscine Vtgs (Folmar et al. in press). This prediction can be compared to that resulting
from the method of von Heijne (1986), represented in Figure 2.2 by asterisks. We made
several attempts to determine the signal peptide sequence through N-terminal sequencing
of Vtg isolated from the blood of estrogen-treated male F. heteroclitus, all of which
resulted in inconclusive residue readings, suggesting that the secreted Fundulus Vtg is
N-terminally blocked. Five internal peptide sequences predicted to offer high antigenicity
by the method of Hopp and Woods (1981) are represented by shaded lettering in Figure
2.2. The end of the cDNA sequence was revealed by a poly-adenylation site
(AATAAA), beginning at bp 5165 and denoted by underlining.
A scan of the sequence for post-translational modification sites of the putative
protein revealed 16 potential N-glycosylation sites, 13 potential N-myristoylation sites,
and potential phosphorylation sites for the following kinases: 7 for CAMP- and CGMP-
dependent protein kinase; 39 for protein kinase C; 23 for casein kinase II; and finally,
a single site for tyrosine kinase (Fig. 2.3). We have highlighted the polyserine domain
in Figure 2.3 with asterisks. The asterisks signify that, in addition to the predicted
phosphorylation sites for the above mentioned kinases, past studies in F. heteroclitus
(Wallace and Begovac, 1985) and in other non-mammalian vertebrates (Mecham and
Olcott, 1949, Mano and Lipmann, 1966, Wiley and Wallace, 1981; Byrne et al., 1984)


134
Nardelli D, Gerber-Huber S, van het Schip F, Gruber M, AB G, Wahli W (1987a)
Vertebrate and nematode genes coding for yolk proteins are derived from a
common ancestor. Biochemistry 26:6397-6402
Nelson G (1989) Phylogeny of major fish groups. In: Femholm B, Bremer K, Jmvall
H (eds) The hierarchy of life. Elsevier, New York p 330
Nelson JS (1984) In: Fishes of the world. Wiley-Interscience, New York, p 87
Ohzu E, Kusa M (1981) Amino acid composition of the egg chorion of rainbow trout.
Annot Zool Japn 54:241-244
Olin T, von der Decken A (1989) Yolk proteins in salmon (Salmo salar) oocytes, eyed
eggs, and alevins differing in viability. Can J Zool 68:895-900
Oppen-Bemtsen DO, Gram-Jensen E, Walther BT (1992b) Zona radiata proteins are
synthesized by rainbow trout (Oncorhynchus mykiss) hepatocytes in response to
oestradiol-17/3. J Endocrinol 135:293-302
Oppen-Bemtsen DO, Helvik JV, Walther BT (1990) The major structural proteins of cod
{Gadus morhua) eggshells and protein crosslinking during teleost egg hardening.
Dev Biol 137:258-265
Oppen-Bemtsen DO, Hyllner SJ, Haux C, Helvik JV, Walther BT (1992a) Eggshell zona
radiata-proteins from cod (Gadus morhua): extra-ovarian origin and induction by
estradiol-17/3. Int J Dev Biol 36:247-254
Oppen-Bemtsen DO, Olsen SO, Rong CJ, Taranger GL, Swanson P, Walther BT (1994)
Plasma levels of eggshell zr-proteins, estradiol-17/3, and gonadotropins during an
annual reproductive cycle of Atlantic salmon {Salmo salar). J Exp Zool 268:59-70
Opresko LK, Wiley HS, Wallace RA (1980) Differential postendocytotic
compartmentation in Xenopus oocytes is mediated by a specifically bound ligand.
Cell 22:47-57
Opresko LK, Wiley HS (1987) Receptor-mediated endocytosis in Xenopus oocytes. I.
Characterization of the vitellogenin receptor system. J Biol Chem 262: 4109-
4115
Oshiro T, Hibiya T (1981) Relationship of yolk globules fusion to oocyte water
absorption in the plaice Limando yokohamae during meiotic maturation. Bull Jpn
Soc Sci Fish 47:1123-1130


62
I and II (Fig. 3.4) reveals a departure from a trend that had previously been noted
concerning serine codon usage in Vtg I (LaFleur et al., 1995) and other vertebrate Vtgs
(Byrne et al., 1989) Whereas the polyserine domains of most vertebrate Vtgs contain a
cluster of TCX codons at the 5 side of the polyserine coding domain and a cluster of
AGY codons at the 3 side, the Vtg II polyserine domain appears to have these codons
equally dispersed, with no obvious clustering.
Northern blot analyses showed that the mRNA of Vtg II transcript can be found
in both estrogen-treated males and spawning females, at an approximate size of 6.0 Kb
(Fig 3.5). By analysis of duplicate blots with separate Vtg I and Vtg II cDNA probes,
it was found that Vtg II transcripts numbered ten times less than those of Vtg I. Vtg I
probes did not cross-hybridize with RNA transcribed from Vtg II clones and vice versa,
confirming that two separate mRNAs for Vtg I and Vtg II were indicated (Fig 3.5).
The N-terminal amino acid sequence of a 69 kDa protein band isolated from the
yolk protein of ovulated eggs was determined astobeNQVSYAPEFAPGxT
Y, where "x" was undetermined ("YP 69" indicated in Chapter 4). Allowing the
predicted K residue in the unidentified "x" position, this sequence provides a perfect
match for the N-terminus of Vtg II after cleavage of the predicted signal peptide (Fig.
3.2, shaded lettering) and indicates that Vtg II is not blocked as is the case with the N-
terminus of Vtg I (LaFleur et al. 1995). These data verify that the Vtg II protein is in
fact expressed, transported, and incorporated as a yolk protein precursor in oocytes of
F. heteroclitus.


Figure 3.2 Translated amino acid sequence (1687 residues) of the putative F.
heteroclitus Vtg II polypeptide. The signal peptide, predicted by
the method of von Heijne (1986) is indicated by underlining, and
verified by the N-terminal sequence obtained from an isolated 69-
kDa yolk protein (shaded lettering). The annealing site of ROW
19, used to isolate the initial insert, is indicated by double
underlining. A polyadenylation site is indicated by underlining.


13
1961; Taborsky, 1980), have led to the speculation that Pv may be important in
embryonic bone formation (Mecham and Olcott, 1949; Rabinowitz, 1962; Taborsky,
1974; Lange, 1981; Wallace and Begovac, 1985; Nardelli et al., 1987; Byrne et al.,
1989). Of additional interest is the hypothesis that evolutionary changes in the Pv region
have occurred at a faster rate than in the two flanking regions, Lvl and Lv2 (Byrne et
al., 1989). To address comparative and evolutionary questions about Vtg, we sought to
characterize a Vtg cDNA that was phylogenetically intermediate to the meager collection
of currently reported sequences. Complete Vtg sequences from the superclass
Gnathostomata have been reported from only two tetrapods (Xenopus and chicken)
leaving several entire lower vertebrate classes unrepresented. Since at least half of all
vertebrates are contained within the subclass Teleostei (Nelson, 1984), the absence of a
teleostean Vtg sequence leaves a substantial gap in our understanding of Vtg evolution,
diversity, and function.
For the present study, we chose as a model the estuarine teleost, Fundulus
heteroclitus, which possesses a non-specialized body plan with a fairly typical
reproductive system, in the hopes of obtaining a piscine Vtg that could be considered as
representative of most teleosts. Much work has already been reported on F. heteroditus
describing vitellogenesis (Wallace and Selman, 1978, 1981; Selman and Wallace, 1983;
Kanungo et al., 1990), the resulting yolk proteins (Wallace and Begovac, 1985; Wallace
and Selman, 1985; Greeley et al., 1986), and oocyte maturation (Wallace and Selman,
1978, 1980). Besides the advantages of F. heteroditus possessing many typical
teleostean traits, there are at least two characteristics of its yolk that presented additional


97
Isolation and Partial Characterization of the Major VEPs
VEPs were isolated following the protocols of Oppen-Bemsten et al. (1990) and
Hyllner et al. (1991) with slight modifications. Ovarian follicles were dissected
from the ovary of a reproductively active F. heteroclitus. Up to 30 individual unovulated
follicles were placed in a 1.5 ml Eppendorf tube containing an ice-cold solution of 0.1
M EDTA and 0.5 M NaCl. The follicles were gently ground with a Kontes pestle. The
intact vitelline envelopes were collected by a low speed spin (150 g) for 5 min, and the
supernatant containing mainly yolk was discarded. The insoluble vitelline envelope
material was washed over 24 hours with at least five changes of ice-cold 0.5 M NaCl
followed by five changes of Milli Q water, each time collecting the material by
centrifugation at 150 g. VEPs were solubilized in a Tris-buffered extraction buffer (0.1
M Tris-HCl, pH 8.8; 2% SDS; 0.3 M 2-mercaptoethanol; 0.1 M EGTA by heating to
70C for 30 min.
For electrophoresis of VEPs, samples were diluted at least 1:4 in sample buffer
(0.06 M Tris-HCl, pH 6.8; 2% SDS; 0.3 M 2-mercaptoethanol; 10% glycerol; without
bromophenol blue) and heated to 95C for 5 min. Sodium dodecyl sulphate-
polyacrylamide gel electrophoresis (SDS-PAGE) was carried out according to Laemmli
(1970) using 125 mm X 110 mm X 1.5 mm slab gels containing a 3.5% stacking gel
overlaying a 10% w/v separating gel, with modifications based on the protocol of
Schgger and von Jagow (1987), using Tris-tricine running buffers.
Initial attempts to transfer VEPs onto membranes using buffers containing 0.01
M morpholinoethane sulphonic acid (MES) and 20% methanol failed, probably due to


Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
ESTROGEN-INDUCED HEPATIC CONTRIBUTIONS TO
OVARIAN FOLLICLE DEVELOPMENT IN FUNDULUS HETEROCL1TUS:
VITELLOGENINS AND CHORIOGENINS
By
Gary James LaFleur, Jr.
May 1996
Chairperson: Professor Robin A. Wallace, Ph.D.
Major Department: Anatomy and Cell Biology
We have increased our chances of isolating cDNAs that code for estrogen-induced
proteins by constructing a liver cDNA library from the poly A+RNA of Fundulus
heteroclitus treated with estradiol- 17/S. We report cDNAs coding for two vitellogenins
(Vtg I and Vtg II) and three novel proteins that share identity with mammalian ZP
proteins. We have designated the latter proteins as "choriogenins" to highlight their role
as components of the vitelline envelope and chorion, yet to emphasize their site of
synthesis as being extra-ovarian, and thus different from that of the mammalian ZP
proteins.
Conceptual translations of the F. heteroclitus Vtg I and II cDNAs share 60%
sequence identity with each other and 30% identity with other reported vertebrate Vtgs.
Vll


31
suggest that almost all serine residues within the phosvitin region are phosphorylated by
an as yet uncharacterized "vitellogenin kinase" activity.
Protein Alignments
Alignment of the F. heteroclitus Vtg sequence with other selected vertebrate Vtgs
is shown in Figure 2.4. Partial Vtg cDNA translations published from three other fish
species are included. Pairwise comparisons of these vertebrate Vtg sequences against the
F. heteroclitus sequence result in similar degrees of identity: Gallus, 38%; Xenopus,
39%; Acipenser, 38%; and Ichthyomyzon, 37%. Against the two smaller teleost
sequences, the F. heteroclitus sequence shares 50% identity with rainbow trout,
Oncorhynchus but only 30% with Oreochromis. These last two values should be
considered only preliminary until more sequence information becomes available.
Attempting to find an obvious difference between the F. heteroclitus Vtg and that of the
other vertebrates, we compared several types of predicted structural analysis scales
including those by the methods of Hopp and Woods (1981), Kyte and Doolittle (1982),
and Janin (1979). There were no striking differences revealed by these methods that
might account for the greater solubility of the F. heteroclitus yolk proteins (data not
shown).
The phylogram in Figure 2.5 was created using the program PAUP (Swofford,
1993) from an alignment (not shown) containing the first five vertebrate Vtgs listed in
Figure 2.4, plus three invertebrate Vtgs from boll weevil, Anthonomus granis,
mosquito, Aedes aegypti, and finally Vtg 5 from C. elegans, defined as an outgroup. In


135
Pan ML, Bell WJ, Telfer WH (1969) Vitellogenic blood protein synthesis by insect fat
body. Science 165:393-394
Raag R, Appelt K, Xuong, N-H, Banaszak L (1988) Structure of the lamprey yolk lipid-
protein complex lipovitellin-phosvitin at 2.8 resolution. J Mol Biol 200:553-569
Rabinowitz M (1962) In: Boyer PD, Lardy H, Myrbck K (eds) The Enzymes. Vol. 6,
Academic Press, N.Y. p 119
Raff RA, Field KG, Olsen GJ, Giovannoni SJ, Lane DJ, Ghiselin MT, Pace NR, Raff
EC (1989) Metazoan phylogeny based on analysis of 18S ribosomal RNA. In:
Femholm B, Bremer K, Jomvall H (eds) The hierarchy of life. Elsevier, New
York, pp 247-260
Rina M, Savakis C (1981) A cluster of vitellogenin genes in the mediterranean fruit fly
Ceratitis capitata: sequence and structural conservation in dipteran yolk proteins
and their genes. Genetics 127:769-780
Ringuette MJ, Chamberlin ME, Baur AW, Sobieski DA, Dean J (1988) Molecular
analysis of a cDNA coding for ZP3, a sperm binding protein of the mouse zona
pellucida. Dev Biol 127:287-295
Rogers S, Wells R, Rechsteiner M (1986) Amino acid sequences common to rapidly
degraded proteins: the PEST hypothesis. Science 234:364-368
Rzepecki LM, Chin SS, Waite JH, and Lavin MF (1991) Molecular diversity of marine
glues: polyphenolic proteins from five mussel species. Mol Mar Biol Biotech
1:78-88
Schagger H, von Jagow, G (1987) Tricine-sodium dodecil sulfate-polyacrylamide ge
electrophoresis for the seperation of proteins in the range from 1 to 100 kDa.
Anal Biochem 166:368-379
Scott MG, Crimmins DL, McCourt DW, Tarrand JJ, Eyerman MC, Nahm MH (1988)
A simple in situ cyanogen cleavage method to obtain internal amino acid sequence
of proteins electroblotted to polyvinyldifluoride membranes. Biochem Biophys
Res Comm 155:1353-1359
Selman GG, Pawsey GJ (1965) The utilization of yolk platelets by tissues of Xenopus
embryos studied by a safranin staining method. J Embryol Exp Morph 14:191-
212
Selman K, Wallace RA (1983) Oogenesis in Fundulus heteroclitus.III. Vitellogenesis.
J Exp Zool 226:441-457


Figure 3.3 ClustalV alignment of F. heteroclitus Vtg I and Vtg II. A
polyserine domain defined according to a previously published
alignment (LaFleur et al. 1995) is indicated by shaded lettering.
Identical residues are denoted by asterisks. Vtg I and Vtg II share
45% overall sequence identity.


19
an overlapping sequence. Digestion with HindiII and EcoR 1 yielded a third fragment,
1610 bp, which was subcloned as pMMB9.
Dideoxynucleotide chain termination sequencing of these three clones revealed
that there were two remaining nucleotide stretches that were needed to complete the
entire cDNA: a small 5 portion which included the initial methionine codon and a 300
bp overlap between pMMB9 and pMMBl. Both of these additional portions were
retrieved from the cDNA library by PCR techniques. First, the initiating methionine was
retrieved by using an exact forward primer (NEB #1231) complementary to the XgtlO
primer adapter sequence and an exact reverse primer, ROW 1, 195 base pairs internal
to the existing 5 end. The resulting product was gel-purified and ligated into the
sequencing plasmid pT7BLUE by the T/A cloning method.
The overlap between pMMB9 and pMMBl was retrieved in a similar fashion by
using two exact internal primers, ROW 12 and ROW 13, made according to the existing
ends of PMMB9 and PMMBl. The resulting product was gel-purified and ligated into
a similar T/A plasmid, pCRIOOO. These two PCR inserts were sequenced and found to
overlap with the already existing sequence resulting in a 5112 bp open reading frame
from which we have deduced the complete primary structure of the putative Fundulus
heteroclitus Vtg polypeptide.
Sequence analysis
Sequencing data were organized and examined using PC\GENE software
(Intelligenetics, Mountain View, CA) including the following analyses: predictions


56
G77AAAACAA77GCA7C7GC7CG7C77GAAACAG77GCCA77GCAAGAGA7G77GAAGGCC7CGC7 2784
VK7IASARLE7VAIAR0VEGLA
GC7GCCAAAG7CACACCGG77G7CCCA7A7GAGCC7A77G7GAGCAAGAACGCCAC777AAA7C77 2 8 S 0
aakvtpvvpyepivsknatlml
TCACAGA7G7C77AC7A7C7GAA7GA7AGCA7A7CAGCA7CA7C7GAAC77C77CC77777CGC7G 2916
SQMSYYI,NDSISASSELLPFSL
CAAAGGCAAACTGGCAAAAATAAAATCCCCAAGCCCATTGTGAAGAAAATGTGTGCAACAACGTAT 2982
QRQ7GKNKIPKPIVKKMCA77Y
ACG7A7GGGA77GAGGGC7GCG77GACA777GG7C7CGCAA7GCAACC77CC7CAGAAACACCCCC 3048
7YGIEGCVDIWSRNATFI.RN7P
A7C7ACGCCA7AA77GGAAACCAC7C7C7777GG77AA7G77ACCCCAGC7GC7GGACCG7CCA7C 3114
IYAIIGNHSLLVNVTPAAGPSI
GAAAGGA7CGAAA7CGAGG77CAG777GG7GAACAAGCAGCAGAAAAGA7CC77AAAGAGG777AC 3180
ERIEISVQFGEQAAEKILKEVY
C7GAA7GAGGAGGAAGAAG7AC77GAAGACAAAAACG7CC77A7GAAGC7GAAGAAGA7TC7G7C7 3246
LNEESEVLEDKNVLMKLKKILS
CC7GG7C7GAAGAACAGCACCAAAGC77CA7CC7C7AG77CGGGCAGC7C7CGC7CCAG7AGA7C7 3312
PGLKNS7KASSSSSGSSRSSRS
CGC7CCAGCAGC7CCAGCAGC7CCAGCAGC7CCAGCAGC7CCAGCCG77CC7CC7C7AGC7C77CC 3378
RSSSSSSSSSSSSSSRSSSSSS
AGGAGC7C77CC7C777GCGCCGCAA7AGCAAGA7G77GGA7C77GCCGA7CCCC7CAACA7AACA 3444
RSSSSLRRNSKMLDLADPLNIT
7CAAAGAGA7CC7CCAGCAGC7CC7CCAGC7CCAGC7CC7CCAGC7CC7CCAGC7CC7CCAGC7CC 3510
SKRSSSSSSSSSSSSSSSSSSS
7CCAGC7CCAAGACCAAG7GGCAGC7GCACGAAAGGAAC77CACCAAGGA7CACA7CCACCAGCA7 3576
sssktkwqlhernftkohihqh
7CCG7C7CAAAAGAACGTC77AACAGCAAGAGCAG7GCGAGCAGC777GAA7CCA777ACAACAAG 3642
SVSKERLNSKSSASSFESIYNK
A7CACA7ACC7G7C7AACA7CG7CAGCCCAG7GG7CACAG7CC77G7CCG7GCCA7CAGAGC7GAC 3708
I7YLSNIVSPVV7VLVRAIRAD
CACAAGAACCAGGGG7A7CAGA7CGC7G7G7AC7A7GACAAAC7CAC7ACCAGAG7GCAGA7CA77 3774
HKNQGYQIAVYYDKLTTRVQII
G7GGCCAACC7CAC7GAAGA7GACAAC7GGAGAA7C7G77C7GACAGCA7GA7GC7CAGCCACCAC 3840
VANL7EDDNWRICSDSMMLSH H
AAAG7GA7GAC7CGAG7CACC7GGGGCA77GGA7GCAAGCAG7ACAACACCACGA7CG7GGCCGAA 3906
KVM7RV7WGIGCKQYN77IVAE
AC7GG7CGCG77GAGAAGGAGCC7GCCG7CCG7G7GAAGC7GGCC7GGGCCAGAC7CCC7AC77AC 3972
7GR7EKEPAVRVKLAWARLPTY
A7CAGGGA77A7GCAAGAAGAG7G7CCAGG7ACA777CCCGCG7CGC7GAGGACAA7GGAG7GAAC 4038
IRDYARRVSRYISRVAEOMGVN
AGGACAAAGG7CGCCAG7AAACCCAAAGAGA7CAAAC7GAC7G7AGC7G77GCCAACGAGACAAGC 4104
R7KVASKPKEIKL7VAVANE7S
Figure 3.2--continued


TABLE OF CONTENTS
ACKNOWLEDGMENTS iii
TABLE OF CONTENTS v
ABSTRACT vii
CHAPTERS
1GENERAL INTRODUCTION 1
The Demands of the Germ Cell 1
The Original Emphasis: Vitellogenins 5
A New Emphasis: Estrogen-Induced Liver Proteins 7
2 FUNDULUS HETEROCL1TUS VITELLOGENIN:
THE DEDUCED PRIMARY STRUCTURE OF A
PISCINE PRECURSOR TO NON-CRYSTALLINE,
LIQUID-PHASE YOLK PROTEIN 11
Introduction 11
Material and Methods 15
Results 22
Discussion 37
3 SEQUENCE COMPARISON OF FUNDULUS HETEROCLITUS
VITELLOGENINS I AND II 46
Introduction 46
Material and Methods 48
Results 58
Discussion 64
4 PRECURSOR-PRODUCT RELATIONSHIP OF VTGS I AND II
TO THE YOLK PROTEINS OF FUNDULUS HETEROCLITUS 70
Introduction 70
Material and Methods 72
Results 76
Discussion 81
v


4
Figure 1.2 A transmission electron micrograph providing an ultrastructural view of
the environment surrounding the oocyte membrane. To the bottom left is
the cytoplasm of the oocyte including a yolk sphere (arrow) containing
processed yolk proteins, derived from Vtg. Distal to the oocyte
membrane is the stratified appearance of the vitelline envelope (bracket),
containing components derived from the choriogenins. This micrograph
was kindly provided by Kelly Selman (X 12,200).


14
motivation for our comparative analyses. First, the yolk proteins of F. heteroclitus
oocytes and eggs remain in a liquid form throughout oocyte growth and maturation
(Wallace et al., 1966; Wallace and Begovac, 1985); this is in marked contrast to the
more typical observation that vertebrate yolk proteins are organized into a specific
crystalline lattice as was reported in lamprey (Karasaki, 1967; Raag et al., 1988),
sturgeon (Lange and Kilarski, 1986), several amphibians (Karasaki, 1963a), and the
reptile, tuatuara (Lange and Kilarski, 1986; reviews by Lange, 1985, Banaszak et al.,
1991). Second, whereas Xenopus and chicken yolk remains in the form of three primary
Vtg cleavage products, Lvj, Pv, and Lv2 plus a few minor peptides or phosvettes (Wiley
and Wallace, 1981; Wallace and Morgan, 1986a, 1986b; Wallace et al. 1990), F.
heteroclitus yolk proteins undergo substantially more processing, resulting in a complex
suite of smaller Vtg-derived cleavage products (Wallace and Begovac, 1985; Wallace and
Selman, 1985; Greeley et al., 1986). We hoped that by obtaining the primary structure
of a teleostean Vtg we would not only confirm regions that are ubiquitously conserved
among oviparous organisms, but would also reveal novel sequence differences that play
a role in the yolk processing events unique to F. heteroclitus.
In this paper we present the predicted primary structure of F. heteroclitus Vtg.
By aligning the F. heteroclitus Vtg sequence to other vertebrate Vtgs, we found that the
most significant differences occurred within the polyserine domain. These differences
may account for some of the molecular phenomena specifically associated with F.
heteroclitus yolk, such as the perpetuation of a liquid phase yolk in both oocytes and
eggs, or the substantial amount of proteolytic processing which occurs in the growing


Table 5.1 Amino Acid Composition, Percent of Total
Chq 427 Cha 500 Chq 553
VEP46
VEP 60
VEP 69
ASN
3.7
2.1
3.0
ASP
5.2
3.8
4.6
ASX
8.9
5.9
7.6
11.4
8.9
7.8
GLN
5.7
11.3
8.7
GLU
5.2
5.0
4.6
GLX
10.9
16.3
13.3
11.2
14.4
18.3
SER
6.7
7.1
7.4
7.6
7.8
7.0
GLY
5.7
6.3
5.1
7.6
7.5
7.0
HIS
1.7
1.0
1.1
1.0
2.0
0.5
ARG
3.7
4.0
3.3
4.1
3.5
3.0
THR
6.9
6.7
8.0
9.9
8.9
8.1
ALA
7.2
5.0
4.9
8.5
5.1
4.9
PRO
9.4
13.4
15.0
8.5
13.3
15.9
TYR
4.7
5.4
5.7
3.5
5.3
5.8
VAL
9.4
5.9
8.0
9.0
7.3
5.7
MET
2.2
1.3
1.1
0.0
1.1
0.0
CYS
2.5
4.0
3.4
0.0
0.2
0.0
ILE
3.7
5.0
2.7
2.8
2.8
4.8
LEU
6.9
4.8
5.3
6.5
5.5
5.2
PHE
4.0
2.9
2.3
4.1
2.8
2.7
LYS
4.4
4.4
4.6
4.7
4.2
4.0
TRP
1.2
0.6
0.8


121
CHAPTER 6
GENERAL SUMMARY
In this dissertation, I have presented the nucleotide and predicted amino acid
sequences of five Fundulus heteroclitus cDNAs: two vitellogenins (Vtg I and Vtg II) and
three choriogenins (Chg 500, Chg 427, and Chg 553). All five of these protein products
are synthesized and secreted by the liver under estrogen induction, and transported by
the blood to the ovary. Vtgs I and II are endocytosed by the oocyte and processed into
liquid phase yolk proteins. In contrast, the Chgs are probably not taken up by the
oocyte, but rather laid down as components of the vitelline envelope and thus eventually
contribute to the structure of the chorion.
As an introduction, I described in Chapter 1, the historical context and initial
goals of the project. Probably the one most essential task accomplished in this work, the
construction of an estrogen-induced liver library, was completed before I became
affiliated with the study, by Marion Byrne, Jyotshnabala Kanungo, and Laura Nelson.
They constructed the library to obtain the primary structure of F. heteroclitus Vtg,
hoping to answer evolutionary as well as biochemical questions. After the initial
investigators disbanded, I became involved with the project, first as a compiler of
sequence data, but eventually leading to a role as primary caretaker of the library.
Chapter 1 concluded with a description of some of my unexpected adventures with the


actaactagaccagacagcttcgaggt 27
101
A)
ATGGCAAGTCACTGGAGTGTCACCCGTTGGGCCGCGCTGGCTCTGCTATGCTGCTTAGCTGGGAAA 93
MASHWSVTRWAALALLCCLAG K22
GGAGCAGAGGCTCAGAAGGGTTCGTATCCTCCGCAACCTCAAAAGCCTTCGTACCCTCAGAATCCT 159
GAEAQKGSYPPQPQKPSYPQNP44
CAAACGCCTTCGTATCCTCAGCAACCTCAAAAGCCTTCGTACCCTCAGAATCCTCAAACGCCTTCG 225
QTPSYPQQPQKPSYPQNPQTPS66
TACCCTCAGTATCCTCAAACACCTTCAAACCCTCAGCAACCTCAGTATCCTCAAACACCTTCAAAC 291
YPQYPQTPSNPQQPQYPQTPSN88
CCTCAGTATCCTCAAACGCCTTCGTACCCTCAGAATCCTCAAACGCCTTCGTACCCTCAGAATCCT 357
PQYPQTPSYPQNPQTPSYPQNP 110
CAAACGCCTTCGTACCCTCAGAACCCTCAGCAACCTCAATTCTCGTGGGATTTTTCAAAGCCTACA 423
QTPSYPQNPQQPQLSWDFSKPT 132
AAACCTCAATATCCTAAGCCCCAAAGGCCTCCATCAAAACCTCAATATCCTAGGCCCCAAACGCCT 489
KPQYPKPQRPPSKPQYPRPQTP 154
CCTTCAAAACCTCAATATCCTAGGCCTCAAACGCCCCAACAACCTGGAAAAAAACAATGGGATGAT 555
PSKPQYPRPQTPQQPGKKQWDD 176
ACAAAGACTCCGAATGTCCCTTCCAAGAGACCAGAGGCCCCTGGAGTTCCCACCCCTAAAAGTTGT 621
TKTPNVPSKRPEAPGVPTPKSC 198
GACGTGGAAGTAGCTTCAAGAGTCCCCTGTGGAGCTTCTGCCGTCTCTGCTACTGAATGTGAGGCC 687
DVEVASRVPCGASAVSATECEA 220
AGAGACTGTTGCTTTGATGGCCAGTCATGCTACTTTGCAAAAGGAGTGACAGTCCAGTGTACCAAG 753
RDCCFDGQSCYFAKGVTVQCTK242
GATGGCCATTTTATCGTTGT1GTGGCCAAAGATGTCACCCTGCCACACATTGACCTTGAAACAATC 819
DGHFIVVVAKDVTLPHIDLETI 264
TCATTGTTGGGAGGAGGTCAAGGCTGTACACATGTTGACCCCAATTCACTTTTTGCCATCTACTAC 885
SLLGGGQGCTHVDPNSLFAIYY 286
TTTCCCGTTACTGCTTGTGGGACTGTTGTCATGGAGGAGCCTGGCGTTATAATGTATGAGAATCGG 951
FPVTACGTVVMEEPGVIMYENR 309
ATGACCTCCTCATATGAAGTAGGAGTTGGGCCTCTTGGAGCCATTACCAGGGACAGCACCTACGAA 1017
MTSSYEVGVGPLGAITRDSTYE 330
TTGCTCTTCCAGTGTAGGTACATTGGCACCTCAGTTGAAACTTTGGTGGTCGAAGTGCTGCCATTA 1083
LLFQCRYIGTSVETLVVEVLPL 352
GACAATCCTCCTCCAGCAGTTGCTGAGCTCGGACCGATCAGAGTGGCCCTTAGGTTGGCCAATGGC 1149
DNPPPAVAELGPIRVALRLANG 374
CAGTGTGCTACAAAGGGTTGCAACGAAGCGGAGGTAGCCTACACCTCCTACTATTTGGACTCAGAC 1215
QCATKGCNEAEVAYTSYYLDSD 396
TA7CCGATTACCAAGATACTGAGGGATCCCGTGTATGTGGAGGTTCAGCTCCTTGAAAAGACAGAT 1281
YPITKILRDPVYVEVQLLEKTD 418
CCCGCTCTGGTTCTGACTCTTGGACGTTGTTGGGCAACCACTAGCCCCAATCCTCACAGCTTGCCC 1347
PALVLTLGRCWATTSPNPHSLP 440
CAGTGGGACATTCTGATTGACGGATGTCCCTACACGGATGATCGTTACCTCTCCACACTGGTTCCA 1413
QWDILIDGCPYTDDRYLSTLVP 462
GTGGACGCCTCTTCTGGTCTGCAATTTCCAAGTCACTACCGGCGTTTCACTTTCAAAATGTTTACC 1479
VDASSGLQFPSHYRRFTFKMFT 484
TTTGTGGACACCACTGCAATGGACCCCCTGAGGGAAAATGTGTACATTCACTGTAGCACAGCTGTG 1545
FVDTTAMDPLRENVYIHCSTAV 506
TGCGTGCCAGGACAGGGTGTCAGCTGCGAACCATCATGCAACAGAAAAGGAAAGAGAGACACTGAG 1611
CVPGQGVSCEPSCNRKGKRDTE 528
GCTGCAGAGCAGAGGAAGGTCGAACCAAAGGTTGTGGTTTCGTCCGGAGAAGTGATCATGACCGCT 1677
AAEQRKVEPKVVVSSGEVIMTA 550
CCTCAGGAGTAAtctgggacaagctcaggaattcatctgggaacatttagacaaaactctttgaaa 1743
P Q E 553
atcaacaaggttgttgaacagtaaataaaaatgtcaccctaagtaaaaaaaaaaaaaaaaaaaaaa 1809
aaaaaaa 1816


33
Fundulus MKAWIi-ALTLAFVAGQ- -NFAPEFAAGKTYVYKYEALILGGLPEEGLARAGLKISTKLL 57
Gallus MRGIIL-ALVLTLVGSQKFDIDPGFNSRRSYLYNYEGSMLNGLQDRSLGKAGVRLSSKLE 59
Xenopus MKGIVL-ALLLALAGSERTHIEPVFSESKISVYNYEAVILNGFPESGLSRAGIKINCKVE 59
Acipenser -LTIALVGSQQTKYEPSFSGSKTYQYKYEGVILTGLPEKGLARAGLKVHCKVE 52
Ichthyomyzon MWKLLLVALAFALADAQ FQPGKVYRYSYDAFSISGLPEPGVNRAGLSGEMKIE 53
Fundulus LSAAIDQNTYMLKLVEPELSEYSGIWPKSPAVPATKLTAALHLSSQFPSSLNTPMVFVGKV 117
Gallus ISGLPENAYLLKVRSPQVEEYNGVWPRDPFTRSSKITQVISSCFTRLFKFEYSSGRIGNI 119
Xenopus ISAYAQRSYFLKIQSPEIKEYNGVWPKDPFTRSSKLTQALAEQLTKPARFEYSNGRVGDI 119
Acipenser ISEVAQKTYLLKILNPEIQEYNGIWPKAPFYPASKLTQALASQLTQPIKFQYRNGQVGDI 112
Ichthyomyzon IHGHTHNQATLKITQVNLKYFLGPWPSDSFYPLTAGYDHFIQQLEVPVREJDYSAGRIGDI 113
Fundulus FAPEEVSTLVLNIYRGILNILQLNIKKTHKVYDLQEVGTQGVCKTLYSISEDARIENILL 177
Gallus YAPEDCPDLCVNIVRGILNMFQMTIKKSQNVYELQEAGIGGICHARYVIQEDRKNSRIYV 179
Xenopus FVADDVSDTVANIYRGILNLLQVTIKKSQDVYDLQESSVGGICHTRYVIQEDKRGDQIRI 179
Acipenser FASEDVSDTVLNIQRGILNMLQLTIKTTQNVYGLQENGIAGICEASYVIQEDRKANKIIV 172
Ichthyomyzon YAPPQVTDTAVNIVRGILNLFQLSLKKNQQTFELQETGVEGICQTTYWQEGYRTNEMAV 173
Fundulus TKTRDLSNCQERLNKDIGLAYTEKCDKeQEETKNLRGTTTLSYVLKPVADAVMJLKAYVN 23 7
Gallus TRTVDLNNCQEKVQKSIGMAYIYPCPVDVMKERLTKGTTAFSYKLKQSDSGTLITDVSSR 23 9
Xenopus IKSTDFNNCQDKVSKTIGLELAEFCHSCKQLNRVIQGAATYTYKLKGRDQGTVIMEVTAR 239
Acipenser TKSKDLNNCNEKIKMDIGMAYSHTCSNCRKIRKNTRGTAAYTYILKPTDTGTLITQATSQ 232
Ichthyomyzon VKTKDLNNCDHKVYKTMGTAYAERCPTCQKMNKNLRSTAVYNYAIFDEPSGYIIKSAHSE 23 3
Fundulus ELIQFSPFSEA-NGAAQMRTKQSLEFLEIEKEPIPSVKAEYRHRGSIjKYEFSDELLQTPLQ 297
Gallus QVYQISPFNEPTGVAVMEARQQLTLVEVRSERGSAPDVPMQNYGSLRYRFPAVLPQMPLQ 2 9 9
Xenopus QVLQVTPFAERHGAATMESRQVLAWVGSKSGQLTPPQIQLKNRGNLHYQFASELHQMPIH 299
Acipenser EVHQLTPFNEMTGAAITEARQKLVLEDAKVIHVTVPEQELKNRGSIQYQFASEILQTPIQ 2 92
Ichthyomyzon EIQQLSVFDIKEGNWIESRQKLILEGIQSAPAASQAASLQNRGGLMYKFPSSAITKMSS 293
Fundulus LI --KISDAPAQVAEVLKHLATYNIEDVHENAPLKFLELVQLLRIARYEDLEMYWNQYKK 355
Gallus LI -KTKNPEQRIVETLQHIVLNNQQDFHDDVSYRFLEWQLCRIANADNLESIWRQVSD 357
Xenopus LM--KTKS PEAQAVEVLQHLVQDTQQHIREDAPAKFLQLVQLLRASNFENLQALWKQFAQ 3 57
Acipenser LF- -KTRSPETKIKEVLQHLVQNNQQQVQSDAPSKFLQLTQLLRACTHENIEGIWRQYEK 350
Ichthyomyzon LFVTKGKNLESEIHTVLKHLVENNQLSVHEDAPAKFLRLTAFLRNVDAGVLQSIWHKLHQ 3 53
Fundulus MSPHRHWFLDTIBATGTFAGLRFIKEKFMAEEITIAEAAQAFITAVHMVTADPEVIKLFE 415
Gallus KPRYRRWLLSAVSASGTTETLKFLKNRIRNDDLNYIQTLLTVSLTLHLLQADEHTLPIAA 417
Xenopus RTQYRRCLLDALPMAGTVDCLKFIKQLIHNEELTTQEAAVLITFAMRSARPGQRNFQISA 417
Acipenser TQLYRRWILDALPAAATPTAFRFITQRIMKRDLTDAEAIQTLVTAMHLVQTNHQIVQMAA 410
Ichthyomyzon QKDYRRWILDAVPAMATSEALLFLKRTLASEQLTSAEATQIVYSTLSNQQATRESLSYAR 413
Fundulus SLVDSDKWENPLLREYVFLGYGTMWKYCNKTVDCPVEMKPIQQRL3DAIAKNEEENI 475
Gallus DLMTSSRIQKNPVLQQVACLGYSSWNRYCSQTSACPKEALQPIHDLADEAISRGREDKM 4 77
Xenopus DLVQDSKVQKYSTVHKAAILAYGTMVRRYCDQLSSCPEHALEPLHELAAEAANKGHYEDI 477
Acipenser ELVFDRANLKCPVLRKHAVLAYGSMVNRYCAETLNCREEALKPLHDFANDAISRAHEEET 4 70
Ichthyomyzon ELLHTSFIRNRPILRKTAVLGYGSLVFRYCANTVSCPDELLQPLHDLLSQSSDRADEEEI 4 73


133
Lyons CE, Payette KL, Price JL, Huang RCC (1993) Expression and structrual analysis
of a teleost homog of a mammalian zona pellucida gene. J Biol Chem 268:21351-
21358
MacDonald RJ, Swift GH, Przybyla AE, and Chirgwin JM (1987) Isolation of RNA
using guanidinium salts. Methods Enzymol 152:219-227
Mano Y, Lipmann F (1966) Enzymatic phosphorylation of fish phosvitin. J Biol Chem
241:3822-3833
Masuda K, Iuchi I, Yamagami K (1991) Analysis of hardening of the egg envelope
(chorion) of the fish, Oryzias latipes. Develop Growth Differ 33:75-83
Matsubara T, Sawano K (1995) Proteolytic cleavage of vitellogenin and yolk proteins
during vitellogenin uptake and oocyte maturation in barfin flounder (Verasper
moseri). J Exp Zool 272:34-45
McMaster GK, Carmichael GG (1977) Analysis of single- and double-stranded nucleic
acids on polyacrylamide and agarose gels by using glyoxal and acridine orange.
Proc Natl Acad Sci USA 74:4835-4838
McPherson R, Greeley MS Jr, Wallace RA (1989) The influence of yolk protein
proteolysis on hydration in the oocytes of Fundulus heteroclitus. Develop Growth
& Differ 31:475-483
Mecham DK, Olcott HS (1949) Phosvitin, the principal phosphoprotein of egg yolk. J
Amer Chem Soc 71:3670-3679
Munday KA, Ansari AQ, Oldroyd D, Akhtar M (1968) Oestrogen-induced calcium
binding protein in Xenopus laevis. Biochim Biophys Acta 166:748-751
Murakami M, Iuchi I, Yamagami K (1990) Yolk phosphoprotein metabolism during early
development of the fish, Oryzias latipes. Develop Growth Difieren 32:619-627
Murata K, Hamazaki TS, Iuchi I, Yamagami K (1991) Spawning female-specific egg
envelope glycoprotein-like substances in Oryzias latipes. Develop. Growth Differ.
33:553-562
Murata K, Sadaki T, Yasumasu S, Iuchi I, Enami J, Yasumasu I, Yamagami K (1995)
Cloning of cDNAs for the precursor protein of a low-molecular-weight subunit
of the inner layer of the egg envelope (chorion) of the fish Oryzias latipes. Dev
Biol 167:9-17


Figure 5.5 Northern blot analysis using Chg 500, 427, and 553 as probes.
A) Methylene blue staining of a nylon membrane indicating
equivalent loading of six lanes with total RNA. Lanes a, d, and
i contain RNA kb markers, lanes b, e, and g each contain 15 /xg
of the same total liver RNA isolated from a single estrogen-treated
male. Lanes c, f, and h contain 15 ng of total liver RNA isolated
from a single female, approximately four days before spawning.
28s and 18s ribosomal RNA bands are indicated in total RNA
lanes, suggesting RNA preparations lacking in RNAse
contamination.
B) Autoradiograph of the same nylon membrane after being cut
into three pieces and hybridized (65C) to the 32P-labeled random-
primed Chg probes indicated above the blot. Positions of RNA
markers are shown to the left, indicating 4.4, 2.37, and 1.35 kb
RNA.


86
interesting to see whether PEST sites are found in other teleostean Vtgs, especially those
of pelagic spawners in which both oocyte hydration and yolk proteolysis are especially
pronounced.


>
ESTROGEN-INDUCED HEPATIC CONTRIBUTIONS
TO OVARIAN FOLLICLE DEVELOPMENT IN FUNDULUS HETEROCL1TUS:
VITELLOGENINS AND CHORIOGENINS
by
GARY JAMES LaFLEUR, JR.
DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1996

Copyright 1996
by
Gary James LaFleur, Jr.

ACKNOWLEDGMENTS
Much of the data presented in this dissertation was gained through cooperative
projects. The original cDNA librarians, Marion Byrne, Josna Kanungo, and Laura
Nelson, added a great gris-gris to the library. The N-terminal sequencing work was done
by the ICBR Protein Core under the supervision of Nancy Denslow, with special
contributions by Hung Nguyen and Sara Reynolds, who were particularly helpful and
patient in their protein isolation wizardry. Invaluable skill in artwork and photography
was contributed by Lynn Milstead and Jim Netherton.
I would like to thank the faculty members that invested an extra amount of
support to my academic training. Kelly Sel man gave me my first view of what yolk and
the vitelline envelope really look like. Kyle Rarey was instrumental in my decision to
join the Dept, of Anatomy and Cell Biology, and offered a safe haven for my qualifying
exam studies. Gill Small introduced me to library screening techniques and library
screaming techniques. Dave Price was always willing to field questions on a wide
variety of subjects from specific molecular interactions to cooking recipes for
invertebrates.
My committee members were very supportive. Chris West walked me through
my first isolation of DNA; he was also the best Mardi Gras king I have ever seen. Paul
Linser was generous with space and comradery in his kind-hearted lab, the chicken wing.
iii

Michael Greenberg was truly inspirational, playing the King Arthur role at Camelot.
Rob Greenberg created a phenomenally free atmosphere in the molecular suite that was
very conducive to hard work and good fun. My adviser Robin Wallace was nothing less
than the perfect mentor. I have never learned so much from someone who said so little.
Robin never gave unsolicited advice. He let me grovel and groan and goof and grow,
taking delight in my triumphs and not noticing my failures. I will admire him always.
The molecular suite at the Whitney Lab, supervised by Rob Greenberg, has been
my incubator for five years. It offered a wonderful mix of data and good friends and
Rock and Roll. Primary fellow cloners that contributed to my journey included Bill
Buzzi, Bemd Eschweiler, Mike Jeziorski, Steve Munger, Clay Smith, and Chuck
Peterson. My late night comrade Sean Boyle deserves special mention for his Southern
kindness and great knack for developing shortcut protocols.
My mother and fathers enthusiasm and unwavering support provided a
cornerstone of stability for all of my years as a student. Thus, their parental investment
in me, the product of their germ cells, has far-outdone that of any ordinary somatic
contribution.
Finally, I would like thank Susanna, a Texas girl that stole my Louisiana heart.
She is an excellent scientist and a perfect mother, but her real talent lies in her knack for
swinging the world by the tail, bouncing over the white clouds, and killing the blues.
IV

TABLE OF CONTENTS
ACKNOWLEDGMENTS iii
TABLE OF CONTENTS v
ABSTRACT vii
CHAPTERS
1GENERAL INTRODUCTION 1
The Demands of the Germ Cell 1
The Original Emphasis: Vitellogenins 5
A New Emphasis: Estrogen-Induced Liver Proteins 7
2 FUNDULUS HETEROCL1TUS VITELLOGENIN:
THE DEDUCED PRIMARY STRUCTURE OF A
PISCINE PRECURSOR TO NON-CRYSTALLINE,
LIQUID-PHASE YOLK PROTEIN 11
Introduction 11
Material and Methods 15
Results 22
Discussion 37
3 SEQUENCE COMPARISON OF FUNDULUS HETEROCLITUS
VITELLOGENINS I AND II 46
Introduction 46
Material and Methods 48
Results 58
Discussion 64
4 PRECURSOR-PRODUCT RELATIONSHIP OF VTGS I AND II
TO THE YOLK PROTEINS OF FUNDULUS HETEROCLITUS 70
Introduction 70
Material and Methods 72
Results 76
Discussion 81
v

5 FUND UL US HETER0CL1TUS CHORIOGENINS:
LIVER-DERIVED COMPONENTS OF THE VITELLINE
ENVELOPE AND CHORION SHARING SEQUENCE
IDENTITY WITH MAMMALIAN ZP PROTEINS 87
Introduction 87
Material and Methods 90
Results 98
Discussion Ill
6 GENERAL SUMMARY 121
REFERENCES 125
BIOGRAPHICAL SKETCH 141
vi

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
ESTROGEN-INDUCED HEPATIC CONTRIBUTIONS TO
OVARIAN FOLLICLE DEVELOPMENT IN FUNDULUS HETEROCL1TUS:
VITELLOGENINS AND CHORIOGENINS
By
Gary James LaFleur, Jr.
May 1996
Chairperson: Professor Robin A. Wallace, Ph.D.
Major Department: Anatomy and Cell Biology
We have increased our chances of isolating cDNAs that code for estrogen-induced
proteins by constructing a liver cDNA library from the poly A+RNA of Fundulus
heteroclitus treated with estradiol- 17/S. We report cDNAs coding for two vitellogenins
(Vtg I and Vtg II) and three novel proteins that share identity with mammalian ZP
proteins. We have designated the latter proteins as "choriogenins" to highlight their role
as components of the vitelline envelope and chorion, yet to emphasize their site of
synthesis as being extra-ovarian, and thus different from that of the mammalian ZP
proteins.
Conceptual translations of the F. heteroclitus Vtg I and II cDNAs share 60%
sequence identity with each other and 30% identity with other reported vertebrate Vtgs.
Vll

The N-terminus of a 69 kDa yolk protein matched the predicted N-terminus of Vtg II
(minus a signal peptide), verifying that Vtg II is expressed without being N-terminally
blocked. Six other yolk proteins were mapped to the predicted Vtg I sequence,
confirming that Vtg I represents the major yolk protein precursor. A 125-kDa yolk
protein that is specifically degraded during final maturation was mapped to a region of
the Vtg I sequence that contained a PEST site, suggesting an explanation for its
preferential break-down.
The three choriogenins were referred to as Chg 500, Chg 427, and Chg 553,
according to the number of amino acids predicted for each protein. Chg 500 and 553
were found to be 58% identical to a flounder "zp gene product", and 30% identical with
the mouse ZP1 protein. Chgs 500 and 553 contain proline-glutamine-rich repeating
regions that resemble a PXX motif reported in other extracellular matrix proteins. Chg
427 was found to be 67% identical to a medaka "L-SF protein" and 30% identitical to
the mouse ZP3 protein that has been implicated as the primary sperm receptor. Besides
reporting the sequences of five hepatically-derived proteins that contribute to the
development of the ovarian follicle, we emphasize that the estrogen-induced library is an
excellent strategy to screen for reproductively significant cDNAs.
Vlll

CHAPTER 1
GENERAL INTRODUCTION
The Demands of the Germ Cell
Reflecting on reproductive strategies of vertebrates, I am reminded of a once
familiar phrase used by an automobile repair shop: You can pay me now...or you can
pay me later." This is a fitting slogan, I think, to describe two alternative relationships
between germ cells and somatic cells, as manifested by different vertebrate groups.
Although the germ cells of all vertebrates are bound to receive an investment from their
associated somatic cells, this investment can be delivered either sooner or later according
to the specific developmental programs. As adults, ourselves, we may consider the
investment made by mothers to their young as an opportunity that is chosen by the
mother, voluntarily. However, this point of view has been described by some as "adult
chauvinism," biased toward the attitudes and experiences of the adult (Wallace, 1983).
An alternative view would be that the mother, or the somatic cells, are essentially held
captive by the germ cells, and (if healthy) have no choice but to respond when called
upon for support. As an illustration, consider the physiological state of the mummichog,
Fundulus heteroclitus (Fig 1.1). When the days of winter begin to grow long, and the
water temperature rises, the female mummichog does not have much say in the matter,
but her ovary begins to grow dramatically, mainly by the incorporation and storage of
1

2
The mummichog, Fundulus heteroclitus, an estuarine teleost of the
Order Cyprinodontiformes as drawn by Lynn Milstead of the
Whitney Laboratory. The top fish is the female; the bottom fish,
displaying more pigment is the male.
Figure 1.1

3
yolk by the oocytes (vitellogenesis) (Taylor et al., 1977; Hsiao et al., 1994). The origin
of the yolk proteins can be traced to a cascade of events resulting in the maternal liver
synthesizing a suite of secreted proteins, primarily consisting of the yolk precursor,
vitellogenin (Vtg), but also containing riboflavin- and vitamin- binding proteins (White,
1987; White and Merrill, 1988) and most recently discovered, precursors of the vitelline
envelope (Hamazaki et al., 1985; Murata, et al., 1991; Hyllner et al., 1991). Thus, the
oocytes, or germ cells, demand an investment by the maternal or somatic cells. They
are saying, "Pay me now." This extensive investment begins long before fertilization,
without the adult knowing whether the eggs will actually ever be spawned or fertilized.
Once the oocytes are expelled, the female, having already surrendered a substantial
amount of energy and material, is relieved of any further investment (until the next clutch
of oocytes begins its demands).
On the other hand, in mammals the germ cells present more of a "Pay me later"
scenario. Mammalian oocytes appear to not receive any yolk at all, with synthesis of
vitellogenin presumed (but not proven) to be totally nonexistent in mammals (except in
the egg laying monotremes) (Eckelbarger, 1994). The investment, then, comes mainly
after fertilization, with support and nourishment provided first by a modification of the
uterus into the chorionic villi, and secondly through lactation, where protein nourishment
continues to be demanded by the progeny, and thus supplied by the adult.
The work contained in this dissertation provides an example of the "Pay me now"
demands of the oocyte on its somatic surroundings. We provide evidence of at least five
distinct proteins that are made by the maternal liver, in response to estradiol, and

4
Figure 1.2 A transmission electron micrograph providing an ultrastructural view of
the environment surrounding the oocyte membrane. To the bottom left is
the cytoplasm of the oocyte including a yolk sphere (arrow) containing
processed yolk proteins, derived from Vtg. Distal to the oocyte
membrane is the stratified appearance of the vitelline envelope (bracket),
containing components derived from the choriogenins. This micrograph
was kindly provided by Kelly Selman (X 12,200).

5
transported to the ovary, to be used by the germ cells and their descendants. Two of
these proteins, Vtg I and Vtg II, are endocytosed by the oocyte, processed, and stored
as yolk (Fig. 1.2), mainly to be used as a nutrient source by the developing embryo.
The three remaining proteins, designated the choriogenins (Chgs), are also synthesized
by the estrogen-induced liver, and transported to the ovary. However, rather than being
endocytosed, the Chgs are laid down as an extracellular matrix between the oocyte and
follicle cells forming the vitelline envelope (Fig. 1.2, in brackets).
The Original Emphasis: Vitellogenins
One of the initial goals of this project was to establish a definitive precursor-
product relationship between vitellogenin and the processed yolk proteins. It was decided
that primary sequence information would be needed for this goal and that the best method
to gain the amino acid sequence of vitellogenin was to use a molecular approach, produce
a cDNA library, screen for Vtg with degenerate primers designed from yolk proteins,
and sequence the cDNA clone. Before the lengthy Vtg sequence was completed, the
original research team disbanded. I subsequently joined the Wallace lab and thereby
"inherited" the Vtg sequencing project. Influenced by the dissertation of Byrne (1989)
describing the evolution of yolk proteins, I became interested in the evolutionary aspects
of Vtg, particularly in the independently evolving phosvitin domain. The lack of a
phosvitin domain in the Caenorhabditis elegans Vtgs (Speith et al., 1991) prompted the
idea that phosvitin may be an exclusively vertebrate inclusion within the Vtg gene (Byrne

6
et al., 1989). We theorized that an interesting oviparous model may be provided by the
protochordate Branchiosotma floridae, the Florida lancelet.
The original aims of my project were, thus, to complete the F. heteroclitus Vtg
cDNA sequence, and thereafter use the piscine cDNA as a heterologous probe to isolate
phylogenetically primitive Vtgs. I succeeded in the former goal, and the results of that
work are provided in Chapter 2. I began screening a cDNA library synthesized from the
MRNA of spawning female amphioxus, B. floridae, by a PCR-based method that utilized
degenerate primers designed by aligning the currently known Vtg protein sequences.
Before long, a new Vtg cDNA was successfully isolated. However, the new Vtg was
isolated from the "control" F. heteroclitus library template, rather than the targeted
amphioxus library template (Fig. 1.3). At that time, no two Vtgs from one vertebrate
species had yet been completely sequenced, and so this appeared to be a worthwhile
challenge. Additionally, the sequence of two F. heteroclitus Vtgs would provide
information presumably necessary to continue mapping out the precursor-product
relationships of the Vtgs and the yolk proteins. As a result, phylogenetic aspects of Vtg
evolution were shelved in order to consider the variations of Vtg that might be
encountered from within one species, F. heteroclitus. The second primary aim of my
project was thus to complete the Vtg II cDNA; this data is provided in Chapter 3.
While completing the two Vtg cDNA sequences, N-terminal sequences of the yolk
proteins were also being obtained. This work represented a collaborative effort that
included data collected from three students at the Whitney lab (including myself) plus a
considerable effort by the ICBR Protein Core facility. Eventually we established a

7
scheme mapping out the specific processing of two Vtgs into several separate yolk
protein products. We have further submitted a hypothesis implicating a PEST site found
within the predicted YP 125 sequence as a possible factor influencing its extensive
degradation. This study is presented in Chapter 4.
A New Emphasis: Estrogen-Induced Reproductive Proteins
While completing the Vtg II cDNA sequence, using a PCR-based screening
method, other non-target cDNAs were often isolated. This is a common phenomenon
in cloning that is usually dismissed as misfortune. However, because our template was
an estrogen-induced cDNA library, the non-target cDNAs stood a likely chance of
representing reproductively significant molecules. This was exactly the case concerning
the Chgs. All three of the Chgs cDNAs were isolated by a fortuitous mis-priming event
that occurred while screening for Vtg II cDNAs (Fig. 1.3). Only recently had a
hypothesis been submitted that ascribed the origin of the major proteins of the teleost
vitelline envelope to the estrogen-induced liver (Hamazaki et al., 1987b). This ran
counter to the mammalian literature that had established the oocyte as the primary site
of synthesis for the proteins of the mammalian zona pellucida (Wassarman, 1988a).
Nevertheless, our data verifies that several Chgs are in fact synthesized by the liver,
transported to the ovary, and laid down as the vitelline envelope between the oocyte and
the follicle cells. These proteins have been referred to by several names. When isolated
from the ovarian follicle, they are usually called vitelline envelope proteins
(VEPs)(Hyllner et al., 1991). Another nomenclature based on isolating the proteins from

8
Vtg IX 4650 CTGGATGAGAGGCCAGACGTG7GGGCTCTGCGGAAAGGCCGACGGGGAAGTCAGACAGG 4693
***** ******** ** ***
TGTGGICTCTGCGGIAAIAACGA
ROW 19 (degenerara) C G T CG T
Chg 500
ROW 45
913 TCCTGGACCTCTGCGTGTGGAGCTCAGGCTTGGGAATGGAGAGTG7TCTGTCAAGGGTT 975
******** ** **
GAG CT CAGTCTGTACACTGCT
Chg 427
ROW 55
669 CAGCCTTCCTCTGGATCCCCTTTGGGTCCCATTCTCTGCAGTTAAGATGGCTGAGGAGT 693
ft** ***
CATTCTGAAACTTGAAGACCC
Chg 553
ROW 45
320 C'rCATTGTTGGGAGGAGGTCAAGGCIGTACACATGTTGACCCCAATTCACTTTTTGCCA 373
*** *** ft******* *
GAGCTCA-GTCTGTACACTGCT
Figure 1.3 Four accounts of fortuitous annealing that resulted in the eventual
isolation of cDNAs coding for Vtg II, Chg 500, Chg 427, and Chg
553. Vtg II was discovered using the degenerate primer ROW 19.
Chg 500 and Chg 553 were discovered using ROW 45, originally
designed for annealing to Vtg II. Chg 427 was isolated using
ROW 55, also designed to anneal to Vtg II.

9
the blood of spawning females used the terms "low molecular weight spawning female
specific substance (L-SF), and high molecular weight spawning female-specific substance
(H-SF)" (Hamazaki et al., 1987a). Still other groups that concentrated on sequence
identity between their teleost proteins and the published mammalian ZP proteins, referred
to their sequences as teleost ZPs (Lyons et al., 1993). We designated the cDNAs and
coded proteins described here as choriogenins (Chgs), precursor proteins of the vitelline
envelope and chorion. We feel that this name accentuates the role of these molecules as
structural components of the vitelline
envelope and chorion, yet emphasizes their origin as being extra-ovarian and thus
different from the homologous ZP proteins of mammals. In Chapter 5, we present the
cDNA and protein sequences of three Chgs, as well as a partial characterization of F.
heteroclitus VEPs. The Chg data represent the most novel aspect of the dissertation,
with the hypothesis of liver-derived vitelline envelope components still fairly recent. One
of the remaining paradoxes presented by the Chg sequences is the comparative disparity
between the mammalian and teleostean systems for producing the extracellular matrix that
surrounds the oocyte. Because the sequence identity between the Chgs and mammalian
ZP proteins suggests an ancestral relationship, the differences in gene regulation, site of
synthesis, and functional roles offer a wealth of interesting questions for future
investigations.
By providing the structure of five previously unsequenced molecules that
contribute to the architecture of the ovarian follicle, we have contributed to our
understanding of reproductive processes in F. heteroclitus. However, we are even more

10
impressed by the remarkable resource proven to lie within the estrogen-induced liver
library that was used to isolate these cDNAs. Rather than being the means to an end,
the library has rather been venerated as possibly the most important attribute of the
project. We expect that other estrogen-induced liver products can be easily isolated from
it, and modifications of this strategy can be used in the future to investigate other
inductive hormone effects on other tissues.

CHAPTER 2
FUNDULUS HETER0CL1TUS VITELLOGENIN:
THE DEDUCED PRIMARY STRUCTURE OF A PISCINE PRECURSOR TO
NON-CRYSTALLINE, LIQUID-PHASE YOLK PROTEINS
Introduction
Vitellogenin (Vtg) is a large phosphoglycoprotein ( 200 Kda) used by most
oviparous animals as a maternally derived yolk precursor (Pan et al., 1969; Kunkel and
Nordin, 1985; Wallace, 1985; Selman and Wallace 1989). It is synthesized by either the
liver (vertebrates), fat body (insects), or intestine (nematodes) under hormonal induction
and transported to growing oocytes via the blood (Flickinger and Rounds, 1956; Wallace
and Jared, 1969). Vtg is incorporated into oocytes by receptor-mediated endocytosis
(Opresko et al., 1980; Opresko and Wiley, 1987; Shen et al., 1993) and is stored for
later use by the developing embryo (Flickinger, 1960; Yamagami, 1960; Karasaki,
1963b; Selman and Pawsey, 1965; Murakami et al., 1990). Once inside the oocyte, Vtg
is processed into smaller yolk proteins consisting of lipovitellins (Lvl and Lv2),
phosvitins (Pv), and phosvettes, that may in turn be degraded into even smaller cleavage
products (Flickinger and Rounds, 1956; Taborsky, 1967; Wallace and Selman, 1985;
Gerber-Huber et al., 1987; Greeley et al., 1986).
11

12
The now familiar Vtg gene family (Wahli et al., 1979, 1991; Tata et al., 1980;
Blumenthal et al., 1984; Nardelli et al., 1987; Byrne et al., 1989; Speith et al., 1991)
encompasses Vtgs synthesized by a wide range of metazoans including the nematode
Caenorhabditis elegans (Speith et al., 1985), the boll weevil Anthonomus granis
(Trewitt et al., 1992), the silkworkm Bombyx mori (Yano et al., 1994), the mosquito
Aedes aegypti (Chen et al., 1994) the cyclostome Ichthyomyzon unicuspis (Sharrock et
al., 1992), the anuran Xenopus laevis (Germond et al., 1984; Gerber-Huber et al., 1987),
and the chicken Gallus domesticus (van het Schip et al., 1987). Additionally, two human
cDNAs, those encoding von Willebrand factor ( 250 kDa) (Baker, 1988a) and
apolipoprotein B-100 (-510 Kda) (Baker, 1988b), have also been reported as distantly
related members of the Vtg gene family. Exceptions to a Vtg-derived yolk precursor
system have been reported in at least two dipteran species: Drosophila melanogaster
(Hovemann et al., 1981) and Ceratitis capitata (Rina and Savakis, 1991) where yolk
precursors, often called Vtgs, do not, in fact, share significant sequence identity with the
"Vtg gene family" setting a precedent for the use of alternative molecules in the
production of yolk (Terpestra and AB, 1988; Bownes, 1992).
A large component of vertebrate Vtgs, the Pv region, was found to be nonexistent
in both C. elegans and the boll weevil Vtg (Nardelli et al., 1987; Trewitt et al., 1992),
inspiring the notion that Pv was an element unique to vertebrate Vtgs. The apparent
absence of the Pv region from invertebrate Vtgs (see Discussion), along with studies
documenting the ability of the phosphate groups of Pv to bind and transport large
amounts of divalent cations, especially Ca++ (Urist et al., 1958; Urist and Schjeide,

13
1961; Taborsky, 1980), have led to the speculation that Pv may be important in
embryonic bone formation (Mecham and Olcott, 1949; Rabinowitz, 1962; Taborsky,
1974; Lange, 1981; Wallace and Begovac, 1985; Nardelli et al., 1987; Byrne et al.,
1989). Of additional interest is the hypothesis that evolutionary changes in the Pv region
have occurred at a faster rate than in the two flanking regions, Lvl and Lv2 (Byrne et
al., 1989). To address comparative and evolutionary questions about Vtg, we sought to
characterize a Vtg cDNA that was phylogenetically intermediate to the meager collection
of currently reported sequences. Complete Vtg sequences from the superclass
Gnathostomata have been reported from only two tetrapods (Xenopus and chicken)
leaving several entire lower vertebrate classes unrepresented. Since at least half of all
vertebrates are contained within the subclass Teleostei (Nelson, 1984), the absence of a
teleostean Vtg sequence leaves a substantial gap in our understanding of Vtg evolution,
diversity, and function.
For the present study, we chose as a model the estuarine teleost, Fundulus
heteroclitus, which possesses a non-specialized body plan with a fairly typical
reproductive system, in the hopes of obtaining a piscine Vtg that could be considered as
representative of most teleosts. Much work has already been reported on F. heteroditus
describing vitellogenesis (Wallace and Selman, 1978, 1981; Selman and Wallace, 1983;
Kanungo et al., 1990), the resulting yolk proteins (Wallace and Begovac, 1985; Wallace
and Selman, 1985; Greeley et al., 1986), and oocyte maturation (Wallace and Selman,
1978, 1980). Besides the advantages of F. heteroditus possessing many typical
teleostean traits, there are at least two characteristics of its yolk that presented additional

14
motivation for our comparative analyses. First, the yolk proteins of F. heteroclitus
oocytes and eggs remain in a liquid form throughout oocyte growth and maturation
(Wallace et al., 1966; Wallace and Begovac, 1985); this is in marked contrast to the
more typical observation that vertebrate yolk proteins are organized into a specific
crystalline lattice as was reported in lamprey (Karasaki, 1967; Raag et al., 1988),
sturgeon (Lange and Kilarski, 1986), several amphibians (Karasaki, 1963a), and the
reptile, tuatuara (Lange and Kilarski, 1986; reviews by Lange, 1985, Banaszak et al.,
1991). Second, whereas Xenopus and chicken yolk remains in the form of three primary
Vtg cleavage products, Lvj, Pv, and Lv2 plus a few minor peptides or phosvettes (Wiley
and Wallace, 1981; Wallace and Morgan, 1986a, 1986b; Wallace et al. 1990), F.
heteroclitus yolk proteins undergo substantially more processing, resulting in a complex
suite of smaller Vtg-derived cleavage products (Wallace and Begovac, 1985; Wallace and
Selman, 1985; Greeley et al., 1986). We hoped that by obtaining the primary structure
of a teleostean Vtg we would not only confirm regions that are ubiquitously conserved
among oviparous organisms, but would also reveal novel sequence differences that play
a role in the yolk processing events unique to F. heteroclitus.
In this paper we present the predicted primary structure of F. heteroclitus Vtg.
By aligning the F. heteroclitus Vtg sequence to other vertebrate Vtgs, we found that the
most significant differences occurred within the polyserine domain. These differences
may account for some of the molecular phenomena specifically associated with F.
heteroclitus yolk, such as the perpetuation of a liquid phase yolk in both oocytes and
eggs, or the substantial amount of proteolytic processing which occurs in the growing

15
oocytes. Although the polyserine domain is indeed a polymorphic region, a conserved
genetic pattern (Byrne et al., 1984, 1989) persists in all of the vertebrates thus far
examined: TCX repeats at the 5 end and a larger group of AGY repeats towards the 3
end, suggesting an ancient origin of the linkage between these two clusters of
trinucleotide repeats.
Materials and Methods
Chemicals
Estradiol-17/3 was obtained from Sigma Chemical Co. (St. Louis, MO).
Radioisotopes, [a-32P] dCTP and [a-35S] dATP, were purchased from New England
Nuclear (Boston MA). Lambda gtlO vector and cDNA synthesis reagents were obtained
from Promega (Madison, WI). The subcloning plasmid pGem-3Z was purchased from
Promega, pT7BLUE from Novagen (Madison, WI), and pCRIOOO from Invitrogen (San
Diego, CA). All sequencing gels were cast using Sequagel-8 (National Diagnostics,
Atlanta) polyacrylamide reagents. Amplification reactions were performed using
Thermophilus aquaticus DNA polymerase (Promega). Sequenase version 2.0 DNA
polymerase and dideoxy sequencing reagents were obtained from US Biochemicals
(Cleveland, OH). Reagents for random-primed labeling of probes were purchased from
Pharmacia (Piscataway, NJ). Both Nytran nylon and S&S NC nitrocellulose transfer
membranes were purchased from Schleicher and Schuell (Keene, NH). Amino acid N-
terminal sequencing and synthesis of oligonucleotide primers were performed by the
University of Florida Interdisciplinary Center for Biotechnology Research core facility.

16
Induction of vitellogenin synthesis
Male Fundulus heteroclitus were collected from the estuarine creeks adjacent to
the Whitney Laboratory, and were maintained in running seawater tanks under 14L: 10D
photoperiod conditions at 25 2C. Fish were maintained for at least one month before
being used for RNA collections.
In order to increase the proportion of Vtg RNA within the total RNA pool,
vitellogenin synthesis was artificially induced in six males (8-10 g body weight) by two
intraperitoneal injections of estradiol-17/3 (0.01 mg/g body weight) dissolved in 50 /xl
peanut oil (Kanungo et al., 1990). Five control males were sham-injected with peanut
oil alone. The first injection was performed on day 1, the second injection on day 4,
followed by sacrifice and liver dissection on day 8.
Isolation of liver poly A + RNA
Livers from both groups of fish were collected and immediately placed in 0C
guanidinium thiocyanate solution (5M guanidinium thiocyanate, 50 mM Tris-Hcl, 25 mM
EDTA, 8% v/v mercaptoethanol, pH 7.4) and homogenized by one thirty-second
polytron (Brinkman) blast. RNA was isolated by the guandidinium thiocyanate method
according to MacDonald et al. (1987). One gram of liver from estrogen-treated fish
yielded an average of 0.536 mg RNA, with an average O.D. 260/280 ratio of 2.03,
while a gram of liver from sham-treated fish yielded an average of 0.318 mg RNA with
an O.D. 260/280 of 1.87. Total RNA samples were combined into two pools: one

17
from six estrogen-treated males and the other from five sham-treated males.
Oligo-dT cellulose chromatography was used to isolate poly A+ RNA from the
two initial pools of total RNA (Aviv and Leder, 1972). Of the 2.1 mg total RNA from
estrogen-treated fish, 46.3 ig poly A+ RNA was recovered (2.2% recovery). Poly A+
RNA from both experimental and control fish was analyzed by northern blot analysis to
verify that Vtg transcripts were included in the poly A+ RNA fraction. The poly A+
RNA was dissolved in deionized glyoxal/DMSO (1:1) and electrophoresed through an
agarose gel (McMaster and Carmichael, 1977) and transferred by capillary action onto
a nylon membrane. The membrane was probed with a 32P end-labeled 17-mer
oligonucleotide, MB6 (degeneracy = 32) which was designed from the N-terminal amino
acid sequence of a small yolk peptide isolated from F. heteroclitus oocytes: His-Lys-
Lys-Met-Val-Ala. Autoradiography of northern blots revealed an MB6 positive, ~ 6 kb
transcript found in the estrogen-treated fish which was absent in sham-treated male fish.
This transcript size was consistent with Vtg cDNA previously reported from chicken
(Cozens et al., 1980; Amberg et al., 1981; van het Schip et al., 1987), frog (Whali et
al., 1979), and rainbow trout (Le Guellec et al., 1988).
cDNA library construction, screening, and sequencing
Synthesis of cDNA was performed by annealing 2 /ig poly A+ RNA with oligo
dT primers, and using AMV reverse transcriptase and T4 DNA pol I for first and second
strand synthesis respectively. Eco R1 adapters were ligated to the two ends of the cDNA
transcripts using T4 DNA ligase (these Eco R1 adapters were later found to have become

18
compromised). After phosphorylation of adapter ends, the cDNA transcripts were ligated
into the bacteriophage vector X gtlO (Promega). Once the X particles were packaged,
the primary library was plated using host E. coli strain C600HFL, resulting in an initial
library titer of 6 x 104 total plaque forming units.
Two 24.5 cm2 petri dishes were used for plating phage-transfected cells
(-400,000 total plaques). Plaques were lifted onto nylon membranes. Hybridization
was performed at 39C using IX Denhardts solution (Denhardt, 1966), 6X SSC (150
Mm NaCl and 15 Mm sodium citrate, pH = 7) with the same end-labelled
oligonucleotide probe as described earlier, MB6. The primary screening resulted in 30%
of the plaques testing positive for the degenerate MB6 probe. By following four plaque
clones (X5, X20, X21, XI6) through two more rounds of positive screening, four final X
clones were set aside for subcloning. The clone (X21) containing the largest insert
(-5000 bp) was subjected to endonuclease digestion with EcoRl, which was expected
to free the entire cDNA insert. Unfortunately the EcoRl sites had inadvertently been
modified so that when digested with EcoRl, one end of the insert remained attached to
the vector. Alternatively, the enzymes Hindlll and Bglil, which like EcoRl cleave at
rare sites, were used to digest X21. Two large fragments were released (1900 bp from
EcoRMBglll and a 2060 bp fragment from EcoR 1 /Hindlll) and these were subcloned into
the sequencing plasmid PGEM 3Z resulting in subclones pMMB6 and pMMBl,
respectively. Because the size of these two fragments did not add up to the total putative
insert size (6000 bp), digestion of another clone, X5, was performed in order to provide

19
an overlapping sequence. Digestion with HindiII and EcoR 1 yielded a third fragment,
1610 bp, which was subcloned as pMMB9.
Dideoxynucleotide chain termination sequencing of these three clones revealed
that there were two remaining nucleotide stretches that were needed to complete the
entire cDNA: a small 5 portion which included the initial methionine codon and a 300
bp overlap between pMMB9 and pMMBl. Both of these additional portions were
retrieved from the cDNA library by PCR techniques. First, the initiating methionine was
retrieved by using an exact forward primer (NEB #1231) complementary to the XgtlO
primer adapter sequence and an exact reverse primer, ROW 1, 195 base pairs internal
to the existing 5 end. The resulting product was gel-purified and ligated into the
sequencing plasmid pT7BLUE by the T/A cloning method.
The overlap between pMMB9 and pMMBl was retrieved in a similar fashion by
using two exact internal primers, ROW 12 and ROW 13, made according to the existing
ends of PMMB9 and PMMBl. The resulting product was gel-purified and ligated into
a similar T/A plasmid, pCRIOOO. These two PCR inserts were sequenced and found to
overlap with the already existing sequence resulting in a 5112 bp open reading frame
from which we have deduced the complete primary structure of the putative Fundulus
heteroclitus Vtg polypeptide.
Sequence analysis
Sequencing data were organized and examined using PC\GENE software
(Intelligenetics, Mountain View, CA) including the following analyses: predictions

20
CONTIGUOUS cDNA SEQUENCE 5112 bp
pGLS
pMMB BBRHBS
pMMB9
seaeow> plasmid insert
lambda insert
I i degenerate probe
czi
oligo MB6
X #21
oligo MB6
X #5
pGL5
pMMBl
Figure 2.1 Cloning strategy used in isolating Furululus heteroclitus Vtg
cDNA. Lambda gtlO bacteriophage clones #5 and #21 were
isolated by tertiary screening with degenerate 17mer, MB6.
pGEM 3Z subclones (pMMB and pMMBl) were constructed
from digestion products of X21 and pMMB9 originated from X5.
Two remaining sections, pGLS and pGL5, were isolated by
anchored PCR, using a 5-/xl aliquot of the cDNA library as
template and exact primers, and then inserted into pT7Blue and
pCRIOOO vectors, respectively.

21
of post-translational modification sites by PROSITE, signal peptide prediction by
PSIGNAL, antigenic determinant analysis using ANTIGEN, codon usage statistics by
CDUSAGE.
Protein sequence alignments were performed using two programs: ClustalV
(Higgins et al., 1992), utilizing the PAM 250 matrix, gap penalty=3, K-tuple=l, no.
of top diagonals=5; window size=5) for the multiple alignment, and ALIGN Plus (S&E
Software, State Line, PA) for pairwise alignments. To normalize domain comparisons,
we defined a "polyserine domain" within the Vtg sequence by choosing two well-aligned
termini as the exterior boundaries, thereby including all of the poorly-aligned polyserine
tracts on the interior. Because we do not have yolk protein data to map the exact region
which is processed into Pv, we have chosen this "polyserine domain" to represent a
hypothetical Pv domain. The chicken and Xenopus Pv termini, which have been
documented (Clark, 1973; Gerber-Huber et al., 1987), lie to the inside of our
boundaries, verifying our convention.
A phylogram was drawn to compare Vtg sequences from eight species. Although
multiple isoforms of Vtg have been identified from several organisms, nomenclature
formally separating these isoforms into subfamilies has not yet been proposed. For our
tree analysis we selected only one Vtg sequence from each species. In species which
contain multiple Vtgs, we chose either the only complete Vtg available from Genbank
databases, as in chicken and Xenopus, or the Vtg which is considered the "major" yolk
protein precursor, as in C. elegans (Speith et al., 1985). An optimal tree was chosen by
importing a ClustalV alignment into the program PAUP (Swofford, 1983) and performing

22
bootstrap analysis of 100 replicates in a branch-and-bound search. C. elegans Vtg 5 was
defined as the outgroup and chicken Vtg was designated as the reference sequence.
Accession codes of sequences used for alignments are as follows: Chicken,
Gallus domesticus Vtg II, EMBL:X13607; Xenopus laevis Vtg A2, GB:M18061; silver
lamprey, lchthyomyzon unicuspis Vtg, GB:M88749; white sturgeon, Acipenser
transmontanus, partial Vtg, GB:U00455; rainbow trout, Oncorhynchus mykiss partial
Vtg GB:M27651; tilapia, Oreochromis aureus partial Vtg, number not available (Ding
et al., 1990); boll weevil, Anthonomus granis Vtg, GB:M72980; nematode,
Caenorhabditis elegans Vtg 5, EMBL:X56213; mosquito, Aedes aegypti Vtg,
GB:U02548; and finally, our own mummichog, Fundulus heteroclitus Vtg, GB:U07055.
Results
Cloning
A summary of our cloning strategy is presented in Figure 2.1 Three restriction
products of two MB6-positive lambda clones (#21 and #5) were subcloned into PGEM
3Z (PMMB1, PMMB6, and PMMB9). Two smaller clones pGL8 and pGL5 were
amplified by PCR directly from the cDNA library. The five subclones were sequenced
in both directions for a final overlapping cDNA sequence of 5198 bp. The overlapping
cDNA sequence contained an open reading frame of 5112 bp and a poly-A tail of
undetermined length beginning 11 nucleotides after a poly-adenylation site (AATAAA)
denoted by underlining in Figure 2.2.

Figure 2.2 Translated amino acid sequence (1,704 residues) of the putative F.
heteroclitus Vtg polypeptide. Two separate signal peptide
predictions are presented. The first was obtained by an alignment
with other fish Vtg signal peptides (Folmar et al., 1995) and is
denoted by shaded lettering. The second prediction was obtained
by the computer analysis method of von Heijne (1986) and is
denoted by asterisks. The nucleotide stretch corresponding to the
degenerate oligonucleotide MB6, used to screen the library, is
shown by double underlining and bold letters. Five predicted
antigenic determinants are depicted by shaded lettering with
average hydrophilicity values (Ah) indicated underneath. A
polyadenylation site (A AT A A A) is located 53 nucleotides past the
stop codon and denoted by underlining.

1
55
109
163
217
271
325
379
433
487
541
595
649
703
7S7
811
865
919
973
1027
24
*
*
*
*
*

*
*
*
*
*
*

#
*
ATG
AAA
GCG
GTT
GTG
CTT
GCC
CTG
ACT
CTG
GCC
TTC
GTG
GCT
GGA
CAA
AAT
TTT
M
X
A
V
V
L
A
L
t*
L
A
p
V
A
G
Q
N
p
GCC
CCT
GAA
GCT
GCT
GGT
AAG
ACC
TAC
GTA
TAT
AAG
TAT
GAA
GCG
CTC
ATC
A
?
E
F
A
A
G
X
T
Y
V
Y
X
Y
E
A
L
I
CTG
GGC
GGT
CCT
GAG
GAA
GGT
TTG
GCA
AGA
GCT
GGA
TTG
AAA
ATC
AGC
ACC
L
G
G
L
P
E
E
G
L
A
R
A
G
L
X
I
S
T
AAA
CTT
CTA
CTC
AGT
GCA
GCT
GAC
CAA
AAT
ACT
TAT
ATG
CTG
AAG
Qmm
GTG
GAA
K
L
L
L
S
A
A
D
Q
N
T
Y
M
L
X
L
V
E
CCT
GAG
CTC
TCT
GAG
TAC
AGC
GGC
ATT
TGG
CCA
AAG
GAC
CCA
GCA
GTG
CCA
GCA
P
E
L
S
E
Y
S
G
I
W
P
X
D
P
A
V
P
A
ACC
AAG
TTG
ACA
GCA
GCC
CTT
CAC
CTC
AGC
TCG
CAA
TTC
CCA
TCA
AGT
TTG
AAT
T
K
L
T
A
A
L
H
L
S
S
Q
F
P
S
S
L
N
ACA
CCA
ATG
GTG
TTT
Uii
GGT
AAA
GTC
TTT
GCT
CCT
GAG
GAA
GTC
TCG
ACT
TTG
T
P
M
V
F
V
G
X
V
F
A
P
S
E
V
S
T
L
GTG
CTG
AAC
ATC
TAC
AGA
GGC
ATC
CTG
AAT
ATT
CTC
CAG
CTG
AAC
ATC
AAG
AAG
V
L
N
I
Y
R
G
I
L
N
I
L
Q
L
N
I
X
X
ACC
CAC
AAA
GTC
TAT
GAC
TTG
CAG
GAG
GTT
GGA
ACT
CAG
GGT
GTG
TGC
AAG
ACC
T
H
X
V
Y
D
L
Q
E
V
G
T
Q
G
V
C
X
T
CTC
TAT
TCC
ATC
AGT
GAA
GAT
GCA
CGA
ATT
GAG
AAC
ATC
CTT
CTG
ACC
AAG
ACC
L
Y
s
I
S
E
D
A
R
I
S
N
I
L
L
T
X
T
AGG
GAC
CTG
AGC
AAC
TGC
CAG
GAA
AGA
CTC
AAT
AAG
GAC
ATC
GGG
TTG
GCA
TAC
R
D
L
S
N
C
Q
E
R
L
N
X

I
G
L
A
Y
ACT
GAG
AAA
TGC
GAC
AAG
TGC
CAG
GAG
GAA
ACT
AAA
AAC
TTG
AGA
GGT
ACC
ACA
T
E
K
C
D
X
C
Q
S
E
T
X
N
L
R
G
T
T
ACA
TTA
AGT
TAC
GTC
TTG
AAA
CCA
GTC
GCC
GAT
GCC
GTC
ATG
ATC
CTG
AAG
GCG
T
L
S
Y
V
L
X
o
V
A
0
A
V
M
X
L
X
A
TAC
AAT
GAG
CTG
ATC
CAG
TTT
TCA
CCT
TTC
TCT
GAG
GCT
AAC
GGA
GCT
GCC
Y
V
N
E
L
I
Q
F
S
P
F
S
E
A
N
G
A
A
CAG
ATG
AGG
ACC
AAG
CAG
TCT
TTG
GAG
TTC
CTT
GAA
ATT
GAG
AAA
GAA
CCC
ATT
Q
M
R
T
X
Q
S
L
E
F
L
E
I
E
X
E
p
I
CCA
TCT
GTC
AAG
GCT
GAA
TAT
CGT
CAC
CGT
GGA
TCT
CTC
AAA
TAC
GAG
TTC
TCC
P
S
V
K
A
E
Y
R
H
R
G
S
L
X
Y
E
F
S
GAT
GAA
Qmn*
CTT
CAG
ACA
CCC
CTT
CAG
CTG
ATC
AAG
ATC
AGT
GAT
GCA
CCA
GCC
D
E
L
L
Q
T
P
L
<2
L
I
X
I
S

A
P
A
CAG
Qwm
GCA
GAG
GTC
CTG
AAG
CAC
CTG
GCT
ACC
TAC
AAC
ATT
GAG
GAT
GTT
CAT
Q
V
A
E
V
L
X
H
L
A
T
Y
N
I
E
D
V
H
GAA
AAT
GCA
CCT
TTG
AAG
TTT
TTG
GAA
CTG
GTA
CAA
CTC
CTC
CGT
ATT
GCC
CGC
E
N
A
P
L
X
F
L
E
L
V
Q
I
L
R
I
A
R
TAT
GAA
GAT
TTG
GAA
ATG
TAC
TGG
AAC
CAG
TAC
AAA
AAG
ATG
TCT
CCC
CAC
AGA
Y
E
D
L
E
M
Y
W
N
Q
Y
X
X
M
S
p
H
R

25
1081
CAC
TGG
TTC
TTG
GAC
ACT
ATT
CCT
GCC
ACT
GGT
ACC
TTC
GCT
GGT
CTC
AGA
TTC
H
W
F
L
D
T
I
P
A
T
G
T
F
A
G
L
R
F
1135
ATC
AAA
GAG
AAG
TTC
ATG
GCT
GAG
GAA
ATA
ACC
ATC
GCT
GAG
GCA
GCT
CAG
GCT
I
K
E
K
F
M
A
E
E
I
T
I
A
E
A
A
2
A
1189
TTC
ATT
ACA
GCT
GTG
CAC
ATG
GTG
ACT
GCT
GAC
CCT
GAG
GTT
ATC
AAG
CTG
TTT
F
I
T
A
V
H
M
V
T
A
D
P
E
V
I
K
L
F
1243
GAG
AGC
CTG
GTA
GAC
AGC
GAC
AAA
GTA
GTG
GAA
AAC
CCA
CTT
CTG
CGT
GAG
GTT
E
S
L
V
D
S
D
K
V
V
E
N
P
L
L
R
E
V
1297
GTC
TTC
CTT
GGA
TAT
GGA
ACA
ATG
GTT
AAC
AAA
TAC
TGC
AAT
AAG
ACA
GTT
GAT
V
F
L
G
Y
G
T
M
V
N
K
Y
C
N
K
T
V
D
1351
TGT
CCT
GTT
GAA
CTC
ATA
AAG
CCT
ATT
CAA
CAA
CGA
CTG
TCA
GAC
GCC
ATT
GCA
C
P
V
E
L
I
K
P
I
2
Q
R
L
S
D
A
I
A
1405
AAG
AAC
GAG
GAA
GAG
AAC
ATC
ATC
CTG
TAC
ATA
AAG
GTT
TTG
GGA
AAT
GCC
GGC
K R B E ^B R
; -
I
I
L
Y
I
K
V
L
G
N
A
G
(Ah
- 2.07)
1459
CAT
CCA
TCT
AGC
TTC
AAG
TCA
CTC
ACT
AAG
ATC
ATG
CCC
ATC
CAT
GGC
ACT
GCT
H
P
S
S
F
K
S
L
T
K
I
M
p
I
H
G
T
A
1513
GCT
GTA
TCT
CTG
CCA
ATG
ACA
ATC
CAT
GTT
GAA
GCC
ATC
ATG
GCT
CTG
AGG
AAC
A
V
S
L
P
M
T
I
H
V
E
A
I
M
A
L
R
N
1567
ATT
GCA
AAG
AAG
GAG
TCC
AGA
ATG
GTC
CAG
GAA
CTG
GCT
CTC
CAG
CTC
TAC
ATG
I
A
K
K
E
S
R
M
V
Q
E
L
A
L
2
L
Y
M
1621
GAC
AAG
GCT
CTC
CAC
CCA
GAG
CTC
CGT
ATG
CTG
TCC
TGC
ATT
GTT
CTC
TTC
GAG
D
K
A
L
H
P
E
L
R
M
L
S
C
I
V
L
F
E
1675
ACA
AGT
CCT
TCT
ATG
GGT
TTG
GTG
ACA
ACT
GTT
GCC
AAC
TCT
GTG
AAA
ACC
GAG
T
S
P
S
M
G
L
V
T
T
V
A
N
S
V
K
T
E
1729
GAG
AAT
TTG
CAG
GTG
GCC
AGC
TTC
ACT
TAC
TCT
CAC
ATG
AAG
TCC
CTA
AGC
AGG
E
N
L
Q
V
A
S
F
T
Y
S
H
M
K
S
L
S
R
1783
AGC
ccc
GCA
ACC
ATC
CAT
CCC
GAT
GTT
GCT
GCC
GCA
TGC
AGC
GCC
GCC
ATG
AAG
S
p
A
T
I
H
p
D
V
A
A
A
C
S
A
A
M
K
1837
ATC
TTG
GGT
ACA
AAG
CTG
GAC
AGA
CTG
AGC
CTG
CGT
TAT
AGC
AAA
GCT
GTA
CAT
I
L
G
T
K
L
D
R
L
S
L
R
Y
S
K
A
V
H
1891
GTG
GAC
CTC
TAC
AAC
AGT
TCC
TTG
GCG
GTC
GGT
GCT
GCT
GCA
ACT
GCT
TTT
TAC
V
D
L
Y
N
S
S
L
A
V
G
A
A
A
T
A
F
Y
1945
ATC
AAC
GAT
GCT
GCC
ACC
TTT
ATG
CCA
AAA
TCC
TTT
GTT
GCA
AAG
ACC
AAA
GGC
I
N
D
A
A
T
F
M
P
K
S
F
V
A
K
T
K
G
1999
TTC
ATC
GCT
GGA
AGT
ACT
GCT
GAA
GTC
CTG
GAG
ATT
GGA
GCG
AAT
ATT
GAA
GGA
F
I
A
G
S
T
A
E
V
L
E
I
G
A
N
I
E
G
2053
CTG
CAG
GAG
CTG
ATT
CTG
AAA
AAC
CCT
GCT
CTC
TCT
GAA
AGT
ACT
GAC
AGG
ATC
L
Q
E
L
I
L
K
N
P
A
L
S
E
S
T
D
R
I
2107
ACC
AAA
ATG
AAG
CGA
GTC
ATT
AAG
GCT
CTG
TCA
GAA
TGG
AGA
TCC
TTG
CCC
ACC
T
K
M
K
R
V
I
K
A
L
S
E
W
R
S
L
P
T
2161
AGC
AAA
CCC
CTA
GCC
TCT
GTC
TAT
GTT
AAG
TTC
TTT
GGA
CAA
GAG
ATT
GGC
TTT
S
K
p
L
A
S
V
Y
V
K
F
F
G
Q
E
I
G
F
Figure 2.2-continued

26
2215
GCT
AAC
ATT
GAC
AAA
CCC
ATG
ATC
GAT
AAG
GCT
GTC
AAG
TTT
GGC
AAG
GAA
TTA
A
N
I

K
P
M
I
D
K
A
V
K
F
G
K
E
L
2269
CCC
ATT
CAG
GAA
TAT
GGA
AGA
GAG
GCT
CTC
AAG
GCT
CTG
CTC
CTG
TCT
GGC
ATC
P
I
Q
E
Y
G
R
E
A
L
K
A
L
L
L
S
G
I
2323
AAC
TTC
CAC
TAC
GCT
AAG
CCA
GTG
CTG
GCT
GCT
GAG
ATG
CGA
CGC
ATT
CTT
CCT
N
F
H
Y
A
K
P
V
L
A
A
E
M
R
R
I
L
P
2377
ACC
GTC
GCT
GGT
ATT
CCA
ATG
GAA
CTC
AGT
CTG
TAC
AGT
GCT
GCT
GTG
GCT
GCA
T
V
A
G
I
P
M
E
L
S
L
Y
S
A
A
V
A
A
2431
GCC
TCT
GTT
GAA
ATC
AAG
CCC
AAC
ACG
TCA
CCA
CGT
CTG
TCA
GCG
GAC
TTC
GAC
A
S
V
E
I
K
p
N
T
S
P
R
L
S
A
0
F
D
2485
GTA
AAG
ACT
CTG
CTG
GAG
ACA
GAC
GTT
GAG
CTC
AAG
GCT
GAG
ATC
AGA
CCA
ATG
V
K
T
L
L
E
T
D
V
E
L
K
A
E
I
R
P
M
2539
GTT
GCC
ATG
GAC
ACA
TAT
GCC
GTT
ATG
GGA
CTT
AAC
ACC
GAC
ATC
TTC
CAG
GCT
V
A
M
0
T
Y
A
V
M
G
L
N
T
D
I
F
Q
A
2593
GCT
TTG
GTA
GCT
CGC
GCT
AAA
CTG
CAC
TCT
GTT
GTG
CCA
GCC
AAA
ATA
GCT
GCA
A
L
V
A
R
A
K
L
H
S
V
V
P
A
K
X
A
A
2647
AGA
CTT
AAT
ATC
AAA
GAG
GGT
GAC
TTT
AAG
CTT
GAA
GCT
CTT
CCT
GTT
GAT
GTG
R
L
N
I
K
E
G
0
F
K
L
E
A
L
P
V
D
V
2701
CCT
GAA
AAC
ATC
ACA
TCC
ATG
AAT
GTT
ACA
ACC
TTT
GCT
GTA
GCA
AGA
AAC
ATC
P
E
N
I
T
S
M
N
V
T
T
F
A
V
A
R
N
I .
2755
GAG
GAA
CCT
TTG
GTT
GAG
AGA
ATC
ACT
CCT
CTT
CTC
CCC
ACC
AAA
GTT
TTG
GTA
E
E
P
L
V
E
R
I
T
P
L
L
P
T
K
V
L
V
2809
CCC
ATC
CCA
ATC
AGG
AGA
CAC
ACA
TCC
AAG
CTT
GAT
CCC
ACT
CGC
AAT
AGC
ATG
p
I
P
I
R
R
H
T
S
K
L
D
P
T
R
N
S
M
2863
TTA
GAC
TCC
TCA
GAA
CTC
CTT
CCC
ATG
GAA
GAA
GAA
GAT
GTA
GAG
CCC
ATT
CCT
L
0
S
S
E
L
I>
p
M
MU
; s
v E i'
D
V
E
p
I
P
(Ah = 2.25)
2917
GAA
TAC
AAG
TTC
CGT
CGA
TTT
GCC
AAA
AAG
TAC
TGC
GCT
AAG
CAC
ATT
GGT
GTT
E
Y
K
F
R
R
F
A
K
K
Y
C
A
K
H
I
G
V
2971
GGA
CTG
AAG
GCC
TGT
TTC
AAG
TTT
GCC
AGT
CAA
AAT
GGA
GCC
TCC
ATC
CAA
GAC
G
L
K
A
C
F
K
F
A
S
Q
N
G
A
S
I
<2
0
3025
ATT
GTC
CTG
TAC
AAA
CTG
GCT
GGT
AGC
CAC
AAC
TTC
TCT
TTC
TCT
GTG
ACA
CCA
I
V
L
Y
K
L
A
G
S
H
N
F
S
F
S
V
T
P
3079
ATT
GAA
GGA
GAA
GTT
GTT
GAG
AGA
TTG
GAG
ATG
GAG
GTT
AAA
GTC
GGA
GCA
AAG
I
E
G
E
V
V
E
R
L
E
M
E
V
K
V
G
A
K
3133
GCT
GCA
GAG
AAG
CTT
GTT
AAA
CGC
ATC
AAC
CTG
AGT
GAG
GAC
GAA
GAA
ACT
GAA
A
A
E
K
L
V
K
R
I
N
L
S
B i
E
E T E
(Ah = 2.43)
3187
GAA
GGA
GGT
CCA
GTC
CTG
GTG
AAG
CTC
AAC
AAA
ATC
CTG
TCT
TCA
AGA
CGG
AAC
m
G
G
P
V
L
V
K
L
N
K
I
L
S
S
R
R
N
3241
AGC
TCC
TCA
TCT
TCC
TCC
TCC
AGC
TCC
AGC
AGC
TCT
TCT
GAG
AGC
CGT
TCT
TCA
S
S
S
S
S
S
S
S
S
S
S
S
S
E
S
R
S
S
3295
AGG
TCC
TCC
TCT
TCC
TCC
TCC
TCT
TCA
TCT
CGC
TCC
AGC
CGT
AAG
ATT
GAC
CTT
R
s
S
S
S
S
S
S
S
S
R
S
S
R
K
I
D
L
Figure 2.2continued

27
3349
GCA
GCC
AGG
ACC
AAT
AGC
AGC
AGC
AGC
AGC
AGT
AGC
CGT
CGC
AGC
AGA
AGC
AGC
A
A
R
T
N
S
S
S
s
S
S
s
R
R
s
R
S
S
3403
AGO
AGC
AGC
AGC
AGC
AGC
AGT
AGC
AGT
AGC
AGC
AGC
AGC
AGC
AGC
AGC
AGC
AGC
s
S
s
S
S
S
s
S
S
s
S
S
S
s
S
S
S
S
3457
AGG
AGA
AGC
AGC
AGC
AGC
AGC
AGT
AGT
AGC
AGC
AGC
AGC
AGC
AGT
AGG
AGC
AGC
R
R
s
S
S
S
s
s
s
S
s
s
s
S
s
R
S
S
3511
AGG
AGA
GTC
AAC
TCA
ACA
AGA
TCC
AGC
AGC
AGT
TCA
AGT
AGG
ACC
AGC
TCT
GCA
R
R
V
N
S
T
R
S
S
S
S
S
S
R
T
S
S
A
3565
TCA
AGC
CTT
GCA
TCT
TTC
TTC
AGT
GAC
AGC
TCA
AGC
TCT
TCT
AGC
TCC
AGT
GAT
s
S
L
A
s
F
F
S
D
S
s
s
s
s
S
S
S
D
3619
CGT
CGC
TCA
AAG
GAA
GTG
ATG
GAG
AAG
TTC
CAG
AGG
TTA
CAC
AAG
AAA
ATG
GTC
R
R
S
K
E
V
M
S
K
F
Q
R
L
H
K
K
M
V
(Ah = 2.55)
3673
GCC
TCC
GGT
AGC
AGT
GCC
TCA
AGC
GTT
GAA
GCC
ATC
TAC
AAA
GAG
AAA
AAA
TAT
A
S
G
S
S
A
S
S
V
E
A
I
Y
K
E
K
K
Y
3727
CT^
GGC
GAG
GAA
GAA
GCC
GTT
GTG
GCA
GTG
ATT
CTC
CGT
GCT
GTC
AAA
GCT
GAC
L
G
E
E
E
A
V
V
A
V
I
L
R
A
V
K
A
0
3781
AAG
AGG
ATG
GTG
GGA
TAC
CAG
CTT
GGT
TTC
TAC
CTT
GAC
AAA
CCA
AAT
GCC
AGA
K
R
M
V
G
Y
Q
L
G
F
Y
L
D
K
P
N
A
R
3835
GTT
CAG
ATC
ATT
GTC
GCC
AAC
ATT
TCT
TCT
GAT
AGC
AAC
TGG
AGG
ATC
TGT
GCT
V
Q
I
I
V
A
N
i
s
S
D
s
N
w
R
I
C
A
3889
GAT
GCA
GTT
GTG
TTG
AGC
AAG
CAC
AAA
GTT
ACA
ACC
AAG
ATT
TCC
TGG
GGA
GAA
D
A
V
V
L
S
K
H
K
V
T
T
K
I
S
W
G
E
3943
CAG
TGC
AGG
AAA
TAC
AGC
ACC
AAT
GTT
ACA
GGA
GAG
ACT
GGT
ATT
GTT
TCT
TCA
Q
c
R
K
Y
S
T
N
V
T
G
E
T
G
I
V
S
S
3997
AGC
CCT
GCC
GCT
CGC
CTC
AGA
GTG
TCC
TGG
GAA
AGA
CTG
CCT
TCT
ACC
CTG
AAA
S
P
A
A
R
L
R
V
s
w
E
R
L
P
S
T
L
K
4051
CGC
TAT
GGA
AAG
ATG
GTT
AAC
AAG
TAC
GTT
CCT
GTT
AAA
ATA
TTG
TCT
GAC
TTG
R
Y
G
K
M
V
N
K
Y
V
P
V
K
I
L
S
D
L
4105
ATC
CAC
ACA
AAG
AGA
GAA
AAC
AGC
ACC
AGG
AAT
ATC
TCA
GTC
ATT
GCA
GTT
GCC
I
H
T
K
R
E
N
S
T
R
N
I
S
V
I
A
V
A
4159
ACA
TCT
GAA
AAG
ACA
ATT
GAC
ATC
ATA
ACC
AAA
ACT
CCA
ATG
AGC
TCT
GTC
TAC
T
S
E
K
T
I
D
I
X
T
K
T
P
M
S
s
V
Y
4213
AAT
GTC
ACT
ATG
CAT
CTT
CCC
ATG
TGT
ATT
CCC
ATT
GAT
GAG
ATC
AAA
GGT
CTC
N
V
T
M
H
L
p
M
c
I
p
I
D
E
I
K
G
L
4267
AGC
CCC
TTT
GAT
GAA
GTC
ATT
GAC
AAG
ATC
CAC
TTC
ATG
GTT
TCT
AAG
GCT
GCT
S
p
F
D
E
V
I
D
K
I
H
F
M
V
S
K
A
A
4321
GCA
GCT
GAA
TGC
AGC
TTC
GTC
GAA
GAC
ACA
CTC
TAC
ACA
TTC
AAC
AAC
AGG
AGC
A
A
E
C
S
F
V
E
D
T
L
Y
T
F
N
N
R
S
4375
TAC
AAG
AAC
AAG
ATG
CCT
TCC
TCT
TGC
TAC
CAG
GTT
GCA
GCA
CAG
GAC
TGC
ACA
Y
K
N
K
M
p
s
s
C
Y
Q
V
A
A
Q
D
C
T
4429
GAT
GAG
CTG
AAA
TTC
ATG
GTT
CTC
CTG
AGG
AAG
GAT
TCG
TCC
GAA
CAA
CAC
CAC
D
E
L
K
F
M
V
L
L
R
K
D
s
s
E
Q
H
H
(Ah 2.1)
Figure 2.2-continued

28
4433
ATC
AAT
GTC
AAG
ATT
tct
GAG
ATC
GAT
ATT
GAC
ATG
TTT
CCA
AAG
GAC
GAC
AAC
I
N
V
X
I
s
E
I
D
I
D
M
F
P
X
D
D
N
4537
GTC
ACT
GTG
AAG
GTC
AAC
GAA
ATG
GAA
ATA
CCC
CCA
CCA
GCC
TGC
CTT
ACC
GCC
V
T
V
K
V
N
E
M
E
X
p
P
P
A
C
L
T
A
4591
ACC
CAA
CAG
Q>ryv
CCA
TTG
AAG
ATC
AAG
ACA
AAG
CGG
AGA
GGA
CTT
GCT
GTC
TAT
T
Q
Q
L
P
L
K
I
K
T
X
R
R
G
L
A
V
Y
4645
GCA
ccc
AGC
CAC
GGT
CTC
CAA
GAA
GTC
TAC
TTT
GAC
AGG
AAG
ACA
TGG
AGG
ATC
A
p
S
H
G
L
Q
E
V
Y
F
D
R
X
T
W
R
I
4699
AAA
GTT
GCT
GAC
TGG
ATG
AAA
GGA
AAG
ACC
TGT
GGA
CTC
TGT
GGA
AAG
GCT
GAT
K
V
A

W
M
K
G
K
T
C
G
L
C
G
X
A
D
4753
GGA
GAG
ATC
AGA
CAG
GAG
TAC
CAC
ACT
CCC
AAC
GGA
CGC
GTG
GCC
AAG
AAC
TCG
G
E
I
R
<2
E
Y
H
T
P
N
G
R
V
A
X
M
S
4807
ATC
AGC
r^rr*rn
GCT
CAC
TCC
TGG
ATT
CTT
CCT
GCT
GAA
AGC
TGC
AGG
GAT
GCA
TCT
I
S
F
A
H
S
W
I
L
P
A
E
S
C
R

A
S
4361
GAG
TGC
CGT
CTG
AAA
CTT
GAA
TCT
GTG
CAG
CTG
GAG
AAA
CAG
TTG
ACC
ATC
CAC
E
C
R
L
K
L
S
S
V
Q
L
E
X
Q
L
T
I
H
4915
GGT
GAG
GAC
TCC
ACA
TGC
TTC
TCA
GTT
GAG
CCT
GTA
CCT
CGT
TGT
CTG
CCC
GGT
G
E
D
S
T
C
F
S
V
E
P
V
P
R
c
L
p
G
4969
TGC
TTG
CCT
GTC
AAG
ACC
ACA
CCT
GTC
ACT
GTT
GGT
TTC
AGC
TGC
CTG
GCA
TCT
C
L
P
V
K
T
T
P
V
T
V
G
F
S
C
L
A
S
5023
GAT
CCT
CAG
ACC
AGT
GTC
TAT
GAC
AGA
AGT
GTG
GAT
CTA
AGA
CAA
ACT
ACC
CAG
D
P
Q
T
S
V
Y
D
R
S
V
0
L
R
Q
T
T
2
5077
GCT
CAC
CTG
GCT
TGC
AGC
TGC
AAC
ACC
AAG
TGC
TCT
TAA
ACA
TAA
GAT
TTC
CTT
A
H
L
A
C
S
C
N
T
K
C
s
-
5131
GAA
GTC
ACT
ACT
ATG
TGT
AAG
TTT
TAT
CTG
TAA
CAA
TAA
ATA
AAC
TGC
ATC
TGA
5185
AAA
TAA
AAA
AAA
AA
Figure 2.2--continued

29
F. heteroclitus Vtg Sequence
A conceptual translation of the 5112 bp open reading frame resulted in a 1704-
amino acid protein sequence (Fig. 2.2). A signal peptide was predicted (underlined) by
aligning the F. heteroclitus sequence with the N-terminal sequences of several other
piscine Vtgs (Folmar et al. in press). This prediction can be compared to that resulting
from the method of von Heijne (1986), represented in Figure 2.2 by asterisks. We made
several attempts to determine the signal peptide sequence through N-terminal sequencing
of Vtg isolated from the blood of estrogen-treated male F. heteroclitus, all of which
resulted in inconclusive residue readings, suggesting that the secreted Fundulus Vtg is
N-terminally blocked. Five internal peptide sequences predicted to offer high antigenicity
by the method of Hopp and Woods (1981) are represented by shaded lettering in Figure
2.2. The end of the cDNA sequence was revealed by a poly-adenylation site
(AATAAA), beginning at bp 5165 and denoted by underlining.
A scan of the sequence for post-translational modification sites of the putative
protein revealed 16 potential N-glycosylation sites, 13 potential N-myristoylation sites,
and potential phosphorylation sites for the following kinases: 7 for CAMP- and CGMP-
dependent protein kinase; 39 for protein kinase C; 23 for casein kinase II; and finally,
a single site for tyrosine kinase (Fig. 2.3). We have highlighted the polyserine domain
in Figure 2.3 with asterisks. The asterisks signify that, in addition to the predicted
phosphorylation sites for the above mentioned kinases, past studies in F. heteroclitus
(Wallace and Begovac, 1985) and in other non-mammalian vertebrates (Mecham and
Olcott, 1949, Mano and Lipmann, 1966, Wiley and Wallace, 1981; Byrne et al., 1984)

30
M /
Fundulus heteroclitus Vitellogenin
PREDICTED...
f N-glycosylation site
^ N-myristoylation site
^ phosphorylation site
irtttttt
Figure 2.3 A schematic representation of potential sites for posttranslational
modifications of the putative F. heteroclitus Vtg protein as predicted by
the Prosite program (Bairoch et al., 1995). Phosphorylation sites
represent potential targets for the following kinases: c-AMP- and g-AMP-
dependent kinase, protein kinase C, casein kinase n, and tyrosine kinase.
The region denoted by asterisks represents the polyserine domain. Past
studies suggest that in addition to the sites displayed by the above-
mentioned kinases, every serine residue in this region undergoes
phosphorylation by an as-yet unidentified "vitellogenin kinase."

31
suggest that almost all serine residues within the phosvitin region are phosphorylated by
an as yet uncharacterized "vitellogenin kinase" activity.
Protein Alignments
Alignment of the F. heteroclitus Vtg sequence with other selected vertebrate Vtgs
is shown in Figure 2.4. Partial Vtg cDNA translations published from three other fish
species are included. Pairwise comparisons of these vertebrate Vtg sequences against the
F. heteroclitus sequence result in similar degrees of identity: Gallus, 38%; Xenopus,
39%; Acipenser, 38%; and Ichthyomyzon, 37%. Against the two smaller teleost
sequences, the F. heteroclitus sequence shares 50% identity with rainbow trout,
Oncorhynchus but only 30% with Oreochromis. These last two values should be
considered only preliminary until more sequence information becomes available.
Attempting to find an obvious difference between the F. heteroclitus Vtg and that of the
other vertebrates, we compared several types of predicted structural analysis scales
including those by the methods of Hopp and Woods (1981), Kyte and Doolittle (1982),
and Janin (1979). There were no striking differences revealed by these methods that
might account for the greater solubility of the F. heteroclitus yolk proteins (data not
shown).
The phylogram in Figure 2.5 was created using the program PAUP (Swofford,
1993) from an alignment (not shown) containing the first five vertebrate Vtgs listed in
Figure 2.4, plus three invertebrate Vtgs from boll weevil, Anthonomus granis,
mosquito, Aedes aegypti, and finally Vtg 5 from C. elegans, defined as an outgroup. In

Figure 2.4 Alignment of the putative F. heteroclitus Vtg sequence (gi:459202)
with other vertebrate Vtgs: the chicken Gallus domesticus Vtg II
(van het Schip et al., 1987); Xenopus laevis Vtg A2 (Gerber-Huber
et al., 1987); the white sturgeon Acipenser transmontanus Vtg
(Bidwell and Carlson, 1995); the silver lamprey Ichthyomyzon
unicuspis Vtg (Sharrock et al., 1992); and the C-termini from the
rainbow trout Oncorhynchus mykiss Vtg (LeGuellec et al., 1988)
and the tilapia Oreochromis aureus (Ding et al., 1990) Vtg as
constructed by ClustalV (Higgins et al., 1992) and modified by
eye. Our defined polyserine domain, which includes putative Pv
regions, is labeled and underscored with a triple dashed line.
Residues identical in at least four of the aligned sequences are
denoted by shaded lettering. Sequence gaps are represented as
dashes.

33
Fundulus MKAWIi-ALTLAFVAGQ- -NFAPEFAAGKTYVYKYEALILGGLPEEGLARAGLKISTKLL 57
Gallus MRGIIL-ALVLTLVGSQKFDIDPGFNSRRSYLYNYEGSMLNGLQDRSLGKAGVRLSSKLE 59
Xenopus MKGIVL-ALLLALAGSERTHIEPVFSESKISVYNYEAVILNGFPESGLSRAGIKINCKVE 59
Acipenser -LTIALVGSQQTKYEPSFSGSKTYQYKYEGVILTGLPEKGLARAGLKVHCKVE 52
Ichthyomyzon MWKLLLVALAFALADAQ FQPGKVYRYSYDAFSISGLPEPGVNRAGLSGEMKIE 53
Fundulus LSAAIDQNTYMLKLVEPELSEYSGIWPKSPAVPATKLTAALHLSSQFPSSLNTPMVFVGKV 117
Gallus ISGLPENAYLLKVRSPQVEEYNGVWPRDPFTRSSKITQVISSCFTRLFKFEYSSGRIGNI 119
Xenopus ISAYAQRSYFLKIQSPEIKEYNGVWPKDPFTRSSKLTQALAEQLTKPARFEYSNGRVGDI 119
Acipenser ISEVAQKTYLLKILNPEIQEYNGIWPKAPFYPASKLTQALASQLTQPIKFQYRNGQVGDI 112
Ichthyomyzon IHGHTHNQATLKITQVNLKYFLGPWPSDSFYPLTAGYDHFIQQLEVPVREJDYSAGRIGDI 113
Fundulus FAPEEVSTLVLNIYRGILNILQLNIKKTHKVYDLQEVGTQGVCKTLYSISEDARIENILL 177
Gallus YAPEDCPDLCVNIVRGILNMFQMTIKKSQNVYELQEAGIGGICHARYVIQEDRKNSRIYV 179
Xenopus FVADDVSDTVANIYRGILNLLQVTIKKSQDVYDLQESSVGGICHTRYVIQEDKRGDQIRI 179
Acipenser FASEDVSDTVLNIQRGILNMLQLTIKTTQNVYGLQENGIAGICEASYVIQEDRKANKIIV 172
Ichthyomyzon YAPPQVTDTAVNIVRGILNLFQLSLKKNQQTFELQETGVEGICQTTYWQEGYRTNEMAV 173
Fundulus TKTRDLSNCQERLNKDIGLAYTEKCDKeQEETKNLRGTTTLSYVLKPVADAVMJLKAYVN 23 7
Gallus TRTVDLNNCQEKVQKSIGMAYIYPCPVDVMKERLTKGTTAFSYKLKQSDSGTLITDVSSR 23 9
Xenopus IKSTDFNNCQDKVSKTIGLELAEFCHSCKQLNRVIQGAATYTYKLKGRDQGTVIMEVTAR 239
Acipenser TKSKDLNNCNEKIKMDIGMAYSHTCSNCRKIRKNTRGTAAYTYILKPTDTGTLITQATSQ 232
Ichthyomyzon VKTKDLNNCDHKVYKTMGTAYAERCPTCQKMNKNLRSTAVYNYAIFDEPSGYIIKSAHSE 23 3
Fundulus ELIQFSPFSEA-NGAAQMRTKQSLEFLEIEKEPIPSVKAEYRHRGSIjKYEFSDELLQTPLQ 297
Gallus QVYQISPFNEPTGVAVMEARQQLTLVEVRSERGSAPDVPMQNYGSLRYRFPAVLPQMPLQ 2 9 9
Xenopus QVLQVTPFAERHGAATMESRQVLAWVGSKSGQLTPPQIQLKNRGNLHYQFASELHQMPIH 299
Acipenser EVHQLTPFNEMTGAAITEARQKLVLEDAKVIHVTVPEQELKNRGSIQYQFASEILQTPIQ 2 92
Ichthyomyzon EIQQLSVFDIKEGNWIESRQKLILEGIQSAPAASQAASLQNRGGLMYKFPSSAITKMSS 293
Fundulus LI --KISDAPAQVAEVLKHLATYNIEDVHENAPLKFLELVQLLRIARYEDLEMYWNQYKK 355
Gallus LI -KTKNPEQRIVETLQHIVLNNQQDFHDDVSYRFLEWQLCRIANADNLESIWRQVSD 357
Xenopus LM--KTKS PEAQAVEVLQHLVQDTQQHIREDAPAKFLQLVQLLRASNFENLQALWKQFAQ 3 57
Acipenser LF- -KTRSPETKIKEVLQHLVQNNQQQVQSDAPSKFLQLTQLLRACTHENIEGIWRQYEK 350
Ichthyomyzon LFVTKGKNLESEIHTVLKHLVENNQLSVHEDAPAKFLRLTAFLRNVDAGVLQSIWHKLHQ 3 53
Fundulus MSPHRHWFLDTIBATGTFAGLRFIKEKFMAEEITIAEAAQAFITAVHMVTADPEVIKLFE 415
Gallus KPRYRRWLLSAVSASGTTETLKFLKNRIRNDDLNYIQTLLTVSLTLHLLQADEHTLPIAA 417
Xenopus RTQYRRCLLDALPMAGTVDCLKFIKQLIHNEELTTQEAAVLITFAMRSARPGQRNFQISA 417
Acipenser TQLYRRWILDALPAAATPTAFRFITQRIMKRDLTDAEAIQTLVTAMHLVQTNHQIVQMAA 410
Ichthyomyzon QKDYRRWILDAVPAMATSEALLFLKRTLASEQLTSAEATQIVYSTLSNQQATRESLSYAR 413
Fundulus SLVDSDKWENPLLREYVFLGYGTMWKYCNKTVDCPVEMKPIQQRL3DAIAKNEEENI 475
Gallus DLMTSSRIQKNPVLQQVACLGYSSWNRYCSQTSACPKEALQPIHDLADEAISRGREDKM 4 77
Xenopus DLVQDSKVQKYSTVHKAAILAYGTMVRRYCDQLSSCPEHALEPLHELAAEAANKGHYEDI 477
Acipenser ELVFDRANLKCPVLRKHAVLAYGSMVNRYCAETLNCREEALKPLHDFANDAISRAHEEET 4 70
Ichthyomyzon ELLHTSFIRNRPILRKTAVLGYGSLVFRYCANTVSCPDELLQPLHDLLSQSSDRADEEEI 4 73

34
Fundulus ILYIKVLGNAGHPSSFKSLTKIMPIHGTAAVSLPMTIHVEAIMAXRNIAKKESRMVQELA 53 5
Callus KLALKCIGNMGEPASLKRILKFLPISSSSAADIPVHIQIDAITALKKIAWKDPKTVQGYL 53 7
Xenopus ALALKALGNAGQPESIKRIQKFEPGFSSSADQLPVRIQTDAVMALRNIAKEDPRKVQEIL 537
Acipenser VLALKALGNAGQPSSIKRIQKCLPGFSSGASQLPVKIQVDAVMALRNIAKKEPGKVQELT 53 0
Ichihyomyzon VLALKALGNAGQPNSIKKIQRFLPGQGKSLDEYSTRVQAEAIMALRNIAKRDPRKVQIV 53 3
Furuiulus LQLYMDKALHPELRMLSCIVLFETSPSMGLVTTVANSVKTEE- -NLQVASFTYSHMKSLS 593
Callus IQILADQSLPPEVRMMACAVIFETRPALALITTIANVAMKESKTNMQVASFVYSHMKSLS 597
Xenopus LQIFMDRDVRTEVRMMACLALFETRPGLATVTAIANVAARESKTNLQLASFTFSQMKALS 597
Acipenser MQLFMDHQLHSEVRMVASMVLLETRPSMALVATLAEALLKE- -TS LQVASFTYSHMKAIT 588
Ichihyomyzon LPIFLNVAIKSELRIRSCIVFFESKPSVALVSMVAVRLRREP- -NLQVASFVYSQMRSLS 591
Furuiulus RSPATIHPDVAAACSAAMKILGTKLDRESLRYSKAVHVDLYNSSLAVGAAATAFYINDAA 653
Gallus KSRLPFMYNISSACNIALKLLSPKLDSMSYRYSKVIRADTYFDNYRVGATGEIFWNSPR 657
Xenopus KSSVPHLEPLAAACSVALKUjNPSLDNBGYRYSKVMRVDTFKYNLMAGAAAKVFIMNSAN 657
Acipenser RST APENHALS S ACNVAVKLIiS RKLDRLS YRYS KAMHMDT F KYPLMAGAAANIH11NNAA 648
Ichihyomyzon RSSNPEFRDVAAACSVAIKMLGSKLDRLGCRYSKAVHVDTFNARTMAGVSADYFRINSPS 651
Fundulus TFMPKSFVAXTKGFIAGSTAEVLEIGANISGLCELILKNPALSESTDR- 701
Gallus TMFPSAIISRLMANSAGSVADLVEVGIRVEGEADVIMKRNIPFAEYPT 705
Xenopus TMFPVFILAKFREYTSLVENDDIEIGIRGEGIEEFLRKQNIQFANFPM 705
Acipenser SILPSAWMKFQAYILSATADPLEXGLHTGliQBVLMQNHEHIDQMPS 6 96
Ichihyomyzon GPLPRAVAAKIRGQGMGYASDIVEFGLRAEGLQELLYRGSQEQDAYGTALDRQTLLRSGQ 711
Fundulus ITKflKRVIKAESEWRSLPTSKPIiASVYVKFFGQEIGFANIDKPMIDKAVKFGKELP 757
Gallus YKQIKELGKALQGWKELPTETPLVSAYLKILGQEVAFININKELLQQVMKTWEPA 761
Xenopus RKKISQIVKSLLGFKGLPSQVPLISGYIKLFGQEIAFTELNKEVIQNTIQALNQPA 761
Acipenser AGKXQQIMKMLSGWKSVPSEKTLASAYIKLFGQEISFSRLDKKTIQEALQAVREPV 752
Ichihyomyzon ARSHVSSIHDTLRKLSDWKSVPEERPLASGYVKVHGQEWFAELDKKMMQRISQLWHSAR 771
Fundulus IQEYG REALKALLLSGINFHYAKPVLAAEMRRILPTVAGIPMBLSLYSAAVAAASV 813
Gallus DRNAA JKRIANQILNSIAGQWTQPVWMGELRYWPSCLGLPLBYGSYTTALARAAV 817
Xenopus ERHTM 1RNVLNKLLNGWGQYARRWMTWEYRH11PTTVGLPAELS LYQS AIVHAAV 817
Acipenser ERQTV 1KRWNQLERGAAAQLSKPLLVAEVRRILPTCIGLPMEMSLYVSAVTTADI 808
Ichihyomyzon SHHAAAQEQI RAWS KLEQGMDVLLTKGYWSEVRYMQPVCIGIPMDLNLLVSGVTTNRA 831
Fundulus EIKPNTSPRLSADFDVKTLLETDVELKAEIRPMVAMDTYAVMGLNTDIFQAALVARAKLH 873
Gallus SVEGKMTPPLTGDFRLSQLLESTMQIRSDLKPSLYVHTVATMGVNTEYFQHAVEIQGEVQ 877
Xenopus NSDVKVKPTPSGDFSAAQLLESQIQLNGEVKPSVLVHTVATMGINSPLFQAGIEFHGKVH 877
Acipenser NVQAHITPSPTNDFNVAQLLNSNIVLHTDVTP3IAMHTIAVMGINTHVIQTGVELHVKAR 868
Ichihyomyzon NLHASFSQSLPADMKLADLLATNIELRVAATTSMSCHAVAIMGLTTDLAKAGMQTHYKTS 8 91
Fundulus SWPAKXAARLNIKEGDFKLEALPVDVPENITSMNVTTFAVARNIEEPLVERITPLIiPTK 933
Gallus TRMPMKFDAKIDVKLKNLKIETNPCREETEIVVGRHKAFAVSRNIGELGVEKRTSILPED 937
Xenopus AHLPAKFTAFLDMKDKNFKIETPPFQQENHLVEIRAQTFAFTRNIADLDSARKTLWPRN 937
Acipenser TTVPMKFTAKIDLKEKNFKIESEPCQQETEVLSLSAQAFAISRNVEDLDAAKKNPLLPEE 928
Ichihyomyzon AGLGVNGK EMNARESNFKASLKPFQQKTWVLSTMESlVFVR- -DPSGSRILPVLPPK 948
Fundulus VLVP -I PIRRHTSKLDPTRNSMLDSSEL - LPMEEEDVEPIPEYKF RRFA 980
Gallus APLD- VTEEPFQTSERASREH FAMQGPDS -MPRKQSHSSREDLRRSTGKRAHK 988
Xenopus NEQN- ILKKHFETTGRTSAE GASMMEDSSEM- -GPKKYSAEPGHHQYAPN INS 987
Acipenser AVRN- ILNEQFNSGTEDSNERERAGKFARPSAEM- -MSQELMNSGEHQNRKGA HAT 981
Ichihyomyzon MTLDKGLISQQQQQPHHQQQPHQHGQDQARAAYQRPWASHEFSPAEQKQIHDIMTARPVM 1008
Figure 2.4-continued

35
Fundulus
Callus
Xenopus
Acipenser
Ichlhyomyzon
Fundulus
Gallus
Xenopus
Acipenser
Ichiliyomyzon
Fundulus
Gallus
Xenopus
Acipenser
Ichthyomyzon
Fundulus
Gallus
Xenopus
Acipenser
Ichlhyomyzon
Fundulus
Gallus
Xenopus
Acipenser
Ichlhyomyzon
Fundulus
Gallus
Xenopus
Acipenser
Ichthyomyzon
Fundulus
Gallus
Xenopus
Acipenser
Ichlhyomyzon
Fundulus
Gallus
Xenopus
Acipenser
Ichlhyomyzon
Oncorhynchus
KK- -YCAKHIGVGLKACFKFASQNGASIQDIVLYKLAGSHNFS FSVTPIEGE- -WERLE
RD--ICLKMHHIGCQLCFSRRSRDASFIQNTYLHKLIGEHEAKIVLMPVHT-DADIDKIQ
YD--ACTKFSKAGVHLCIQCKTHNAASRRNTIFYQAVGEHDFKLTMKPAHT-EGAIEKLQ
RS--ACAKAKNFGFEVCFEGKSENVAFLRDSPLYKIIGQHHCKIALKPSHSSEATIEKIQ
RRKQHCSKSAALSSKVCFSARLRNAAFIRNALLYKITGDYVSKVYVQPT-SSKAQIQKVE
MEVKVGAKAAEKLVKRINI.SEDEETEEG--GPVLVKLNKI
LEIQAGSRAAARIITEVNPESEEEDESSPYEDIQAKliKRILGIDSMFKVANKTRHPKNRP
LEITAGPKAASKTMGLVEVEGTEGEPMDE-TAVTKRLKMILGIDESRKDTNETALYRSKQ
LELQTGNKAASKI IRWAMQSLAEADEMK GNI LKKNKLItVDGE
ELQAGPQAEKVIRMVELVAKASKKSKKNSTITEEGVGETIISQLKKILSSDKDK
>===POLYSERINE DOMAINss
SKKGNTVLAEFGTEPDAKTSSSSSSASSTATSSASSSASSPNRKKPMDEEENDQVKQARN
KKKNKI HNRRLDAE WEARK
T
DAKKPPGSSSSSSSSSSSSSSSSSSDKSGKKTPRQGSTVNLAAKR
====~=============s==POLYSERINE DOMAIN=====================
SSRRNSSSSSSSSSSSSSESRSSRSSSSSSSSSRSSRKIDLAARTNSSSSSS
KDASSSSRSSKSSNSSKRSSSKSSNSSKRSSSSSSSSSSSSRSSSSSSSS SSNSK
QQSSLSSSSSSSSSSSSSSSSSSSSSSSSSPSSSSSSSYSKRSKRREHNPHHQRESSS-S
QDSTLRGFKRRSSSSSSS3SSSSSSSSSSSSSSSQQSRMEKRMEQDKLTENLERDRDHMR
AS KKQRGKDSSSSSSSSSSSSDSSKSPHK- -HGGAKRQHAGHGAPHLGPQSHSSSSSSSS
s=======2=============POLYSERINE DOMAIN=====================
SRRSRSS- SSSSSSSSSSSSSSSSSSRRSSSSSSSSSSSSSRSSRR
SSSSSSKSSSSSSRSRSSSKSSSSSSSSSSSSSSKSSSSRSSSSSSKSSSHHSHSHHSGH
SSQEQNKKRNLQENRKHGQKGMSSSSSSSSSSSSSSSSSSSSSSSSSSSSEENRPHKNRQ
GKQSKNKKQEWKNKQKKHHKQLPSSSSSSSSSSSGSNSSSSSSSSSSSSS---RSHNHRN
SSSSSSASKSFSTVKPPMTRKPRPARSSSSSSSSDSSSSSSSSSSSSSSSSSSSS
======s==!3============POLYSERINE DOMAIN=5=s=================
VNSTRSSSSSSRTSSASSLASFFSDSSSSSSSSD RRSKEVME-KFQRLHK-K
LNGSSSSSSSSRSVSHHSHEHHSGHLEDDSSSSSSSSVLSKIWGRHEIYQYRFRSAHR-Q
HDNKQAKMQSNQHQQKKNKFSESSSSSSSSSSSEMWNKKKHHRNFYDLNFRRTAR-T
NTRTLSK SKRYQNNNNSSSSSGSSSSSEEIQKNPEI FAYRfRSHRD K
SSSSKSEEWLAVKDVNQSAFYNFKYVPQRKPQ
===ss===========3==ss=POLYSERINE DOMAINh=======~===5=======
MV ASGSSASSVEAIYKEK KYLGEEEA- WAVILRAVKADKRMV
EFPKRKLPGDRATSRYSSTRSSHDTSRAASWPKFLGDIKTPVLAAFLHGISNNKKTG
KGTEHRGSRLSSSSESSSSSSESAY RHKA KFLGDKEPPVLWTFKAVRNDNTKQ
LGFQNKRGRMSSSSSSSSSSSSQSTLNSKQDA KFLGDSSPPIFAFVARAVRSDGLQQ
TSRRHTPASSSSSSSSSSSSSSSSSSSDSDMTVSAESFEKHSKPKWIVLRAVRADGKQQ
============POLYSERINE DOMAINh==========<
GYQLGFYLD KPNARVQIIVANISSDSNWRICADAWLSKHKVTTKISWGEQCRKYST
GLQLWYAD-TDSVRPRVQVFVTNLTDSSKWKLCADASVRNAPQAVAYVKWGWDCRDYKV
GYQMWYQE YHSSKQQIQAYVMDI -SKTRWAACFDAVWNPHEAQASLKWGQNCQDYKI
GYQVAAYTD-NRVSRPRVQLLATEIIEKSRWQICADAILASNY KAMALMRWGEECQDYKV
GLQTTLYYGLTSNGLPKAKIVAVELSDLSVWKLCAKFRLSAUMKAKAAIGWGKNCQQYRA
LGRPKTTSDEPNIITAALDENDNWKLCADGVLLSKHKVNKIAWGAGCKDYNT
1036
1045
1044
1039
1067
1075
1105
1103
1084
1123
1075
1165
1123
1085
1168
1127
1220
1182
1145
1226
1172
1280
1242
1202
1281
1222
1339
1298
1249
1314
1264
1396
1352
1306
1374
1321
1455
1410
1365
1434
53
Figure 2.4--continued

36
Fundillos NVTGETGIVSSSPAARLRVSWERLPSTLKRYGK-MVNKYVP-VKILSDLIHTKRENSTRN 1379
Gallus STELVTGRFAGHPAAQVKLEWPKVPSNVRSWE-WFYEFVPGAAFMLGFSERMDKNPSRQ 1514
Xenopus NMKAETGNFGNQPALRVTANWPKIPSKWKSTGK-WGEYVPGAMYMMGFQGEYKRNSQRQ 14 6 9
Acipenser AVSAVTGRLASHPSLQIKAKWSRIPRAAKQTQN- ILAEYVPGAAFMLGFSQKEQRNPSKQ 14 24
Ichihvomyzon MLEASTGNLQSHPAARVDIKWGRLPSSLQRAKNALLENKAPVIASKLEMEIMPKKNQKHQ 14 94
Oncorhyncus FITAETGLVGPSPAVRLLDKLPKVPKAVWRYVRIVSEFIPGHIPYYLADLVPMQKDKNSE 113
Fundulus IS VIAVATS E KTID11TKTPMSS VYNVTMHLPMCIPIDE--I KGLSP--FDEVIDKI 14 32
Gallus ARM WALTS PRTCD WVKLPDI1LYQKAVRLPLS LPVG P--RIPASELQPPIW NVFAEA 1571
Xenopus VKLVFALSSPRTCDWIRI PR LTVYYRALRLPVPIPVGH -HAKENVLQTPTW-NIFAEA 1526
Acipenser FK11LAVTSPNTIDTLIKAPKITLFKQAVQIPVQIPMEP--SDAER- -RSPGLASIMNEI 1480
Ichlhyomyzon VSVILAAMTPRRMNIIVKLPKVTYFQQGILLPFTFPSPRFWDRPEGSQSDSLPAQIASAF 1554
Oncorhynchus KQFTWATSERTLDVILKTPKMTLTKTGVNIPCSLPFESMTDLSPFDDNIVNKIHYL- -F 171
Fundulus HFMVSKAAAAECSFVEDTLYTFMNRSYKNKMPSSCYQVAAQDCTDELKFMVLLRK- -DSS 14 90
Gallus PSAVLENLKARCSVSYNKIKTFNEVKFNYSMPANCYHILVQDCSSELKFLVMMKSAGEAT 1631
Xenopus PKLIMDSIQGECKVAQDQITTFNGVDLASALPENCYNVLAQDCS PEMKFMVLMRNS KESP 1586
Acipenser PFLIEEATKSKCVAQENKFITFDGVKFSYQMPGGCYHILAQDCRSKVRFMVMLKQASMSK 154 0
Ichlhyomyzon SGIVQDPVASACELNEQSLTTFNGAFFNYDMPESCYHVLAQECSSRPPFIVLIKLDSERR 1614
Oncorhynchus S EVNAVKCSMVRDTLTTFNNKKYKINMPLSCYQVLAQDCTTELKFMVSAEEGSVHL 227
Fundulus EQHHIHVKTSEIDTDMF- PKDDNVTVKWEMEIPPPA- CLTATQQLPLKIKTKRRGLAVY 154 8
Gallus NI.KAINIKIGSHEIDM-H PVNGQVKLLVDGAES PTANIS LIS AGAS LWIHNENQGF ALA 168 9
Xenopus NHKDINVKLGEYDXDMYYSA-DAFKMKINNLBVSEEHLPYKSFNYPTVEIKKKGNGVSLS 1645
Acipenser NLRAVHAKIYNKDXDILPTTKGSVRLLINNNEIPLSQLPFTD-SSGNIHKRADEGVSVS 1599
Ichlhyomyzon I- -SLELQLDDKKVKIVSRND IRVDfJEKVPLRRLSQKN QYGFLVLDAGVHLL 1664
Oncorhynchus NKTTSNVKISDIBVpLYTQDHGVIVKVNEMEVSNEQLPYKDPSG-SIKIDRKKGEGVSLY 28 6
Fundulus APSHGLQEVYFDRKTWRIKVADWMKGKTCGLCGKADGEIRQEYHTPNGRVAKNSISFAHS 16 08
Gallus APGHGIDKLYFDGKTITIQVPLWMAGKTCGICGKYDAECEQEYRMPNGYLAKNAVSFGHS 1749
Xenopus ASEYGIDSLDYDGLTFKFRPTIWMKGKTCGICGHNDDESEKELQMPDGSVAKDQMRFIHS 1705
Acipenser AQQYGLESLYFDGKTVQVKVTSEMRGKTCGLCGHNDGERRKEFRMPDGRQARGP 1653
Ichlhyomyzon LKYKDL-RVSFNSSSVQVWVPSSLKGQTCGLCGRNDDELVTEMRMPNLEVAKDFTSFAHS 1723
Oncorhynchus APSHGLQKVYFDKYSWKIKWDWMKGQTCGLCGKADGENRQEYRTPSGRLTKSSVSFAHS 34 6
Oreochromis FFFSLVFHAVS 11
Fundulus WILPAESCRDASECRLKLESVQLEKQLTIHGEDSTCFSVEPVPRCLPGCLPVKTTPVTVG 16 6 8
Gallus WILEEAPCRGA- -CKLHRSFVKLEKTVQLAGVDSKCYSTEPVLRCAKGCSATKTTPVTVG 1807
Xenopus WILPAESCSEG--CNLKHTLVKLEKAIATDGAKAKCYSVQPVLRCAKGCSPVKTVEVSTG 1763
Acipenser SVSPTPG 1660
Ichlhyomyzon WIAPDETCGGACALSRQ--TWKESTSVISGSRENCYSTEPIMRCPATCSASRSVPVSVA 1781
Oncorhynchus WVLPSDRG-DASEG-LM-- KLEKQVIVDD-REEK-CYSVEPVLRCLPGCSPVRTTPITIG 400
Oreochromis KKLQNHYSLRLLKEKVKS ELMVPILKVSEPNATLLSPCCSACPACIPVRTTTVNVG 67
Fundulus -FSCLASDPQ TSVYD RSVDLRQTTQAHLACSCNTK- CS 1704
Gallus -FHCLPADSANSLTDKQ-MKYDQKSEDMQDTVDAHTTCSCENEECST 1852
Xenopus FHCLPSDVSLDLPEGQ IRLE- KSEDFSEKVEAI1TACSGETSPCAA 1807
Acipenser --LGLEKTATEAASFCVIM 1677
Ichlhyomyzon -MHCLPAESEAISLAMSEGRPFSLSGKSEDLVTEMEAHVSCVA 1823
Oncorhynchus - HCLPFDSNLNRSEGLSSIY EKSVDLMF.KAEAHVACRCSEQ- CM 442
Oreochromis FYGCLPSDTT VDRSGLSSFF EKSIDLRDTAEAHLACRCTPQ-CA 110
Figure 2.4--continued

37
the single best tree F. heteroclitus Vtg was placed on an independent branch,
intermediate to the positions of sturgeon and lamprey Vtgs. The Ichthyomyzon sequence
was the vertebrate Vtg determined to lie furthest from the reference sequence, thereby
placing it nearest to the outgroup. One of the more significant relationships provided by
the tree is indicated by the bootstrapping values at the Acipenser branch (in parentheses,
Fig. 2.5): through 100 bootstrap replicates, sturgeon Vtg was partitioned with the two
tretrapod Vtgs 95% of the time, substantially more than the Vtgs of either F. heteroclitus
(67%) or Ichthyomyzon (31% not shown).
Polvserine Domain
We have designated a polyserine domain from each of the aligned Vtgs
(underscored with a triple dotted line in Fig. 2.4; see Materials and Methods) and
compared them in regard to size, relative serine composition and serine codon usage
(Fig. 2.6). Of the Vtgs listed here, F. heteroclitus Vtg contains the smallest polyserine
domain (171 a.a.); it also contains the highest relative serine composition (57.6%). We
compared the serine codon usage from each of the domains and found a consistent
pattern: TCX repeats are more prevalent at the 5end while AGY codons are more
prevalent at the 3 end. Finally, of the six possible serine codons, AGC was invariably
the dominant codon in all five vertebrate polyserine domains.
Discussion
We present the first complete teleost Vtg cDNA sequence along with its translated

38
primary structure. F. heteroclitus Vtg shares 37% 38% identity with other vertebrate
Vtgs and it includes the characteristic N-terminal Lvl region, an internal Pv region and
a C-terminal Lv2 region. The genetic organization of the polyserine domain is consistent
with that found in other vertebrates, from lamprey to chicken, suggesting, at the latest,
a pre-gnathostome arrival of this domain into the Vtg gene. In contrast to other
vertebrate Vtgs, F. heteroclitus Vtg is predicted to be 100 amino acids shorter, and
contains a polyserine region with a 10-20% higher relative serine composition than the
other vertebrates Vtgs. We suspect that the occurrence of liquid phase yolk in F.
heteroclitus is in part due to differences within its Vtg polyserine domain as compared
with the polyserine domains of insoluble yolk producers. The higher than usual relative
serine composition would eventually be modified into a polyphosphoserine domain,
endowing the resulting Pv yolk protein with an uncommonly strong hydrophilic potential.
On examination of the alignment in Figure 2.4, the conserved organization of
vertebrate Vtg is evident: two well-aligned termini interrupted by a polymorphic
polyserine domain. The degree of Vtg conservation among several oviparous species is
further resolved by the phylogenetic tree analysis presented in Figure 2.5. The results
of the branch-and-bound tree search suggest that the present structure of F. heteroclitus
Vtg represents a substantial history of divergence from the ancestral osteichthyean Vtg.
Although, phylogenetically, F. heteroclitus and A. transmontanus are considered
monophyletic as actinopterygian fishes (Nelson, 1989), the Vtg structure of A.
transmontanus was found to be more closely related to the Vtgs of the two tetrapods than
it was to that of F. heteroclitus. Indeed many character traits of the genus Acipenser

39
have long been recognized as tetrapod-like, ie. a holoblastic embryonic cleavage and
anuran-like gastrulation (Balinksky, 1965; Beer, 1981; Conte et al., 1988), an acrosome-
capped spermatozoan (Conte et al., 1988), and development of oviducts from true
Mllerian ducts (Conte et al., 1988). We suggest that the structure of F. heteroclitus Vtg
represents a derived, perhaps more specialized, example of Vtg structure in contrast to
the tetrapod/chondrostean Vtg, which more likely resembles the Vtg of an ancestral
osteichthyean. If this is the case, we would predict that the structure of an elasmobranch
Vtg (especially from a less derived species) would also resemble the tetrapod Vtgs more
closely than it would a teleostean Vtg. Whether the structure of lamprey Vtg represents
an independent derivation, or an even earlier, prototypical vertebrate Vtg, is difficult to
surmise. This question will be more easily answered once a protochordate or
invertebrate deuterostome Vtg (from within the "Vtg family) has been sequenced. Within
the invertebrate outgroup of our phylogram, the two insect Vtgs appear to be highly
derived versions of Vtg structure as compared to the C. elegans Vtg. The C. elegans
Vtg is substantially more similar to vertebrate Vtgs than are the Vtgs of the two insects,
suggesting a faithfulness of the nematode Vtg to an ancestral form originating in a
predecessor common to both vertebrates and platyhelminthes.
In reference to past alignments between multiple Xenopus and chicken Vtgs,
Byrne (1989) described Pv as an independently evolving domain within Vtg. Our
alignment confirms this suggestion. While the two Lv domains of Vtg can be well
aligned among several organisms, the polyserine domain exists in a wide range of sizes

568
40
(6555)
234
(9555)
278
(6755)
341
(100%)
665
Gall LIS Vt2 II
469
428
Xenopus Vtg A2
Acipenser Vtg
563
Fundulus Vtg
581
836
Ichthyomyzon Vtg
Caenorhabditis Vtg 5
642
(100%)
896
Anthonomus Vtg
591
Aedes Vtg
Figure 2.5 Branch-and-bound phylogenetic tree analysis comparing selected Vtgs
spanning 600 million years of divergence (Raff et al., 1989). PAUP
(Swofford, 1992) analysis was done on a ClustalV alignment (Higgins et
al., 1992) containing five of the vertebrate cDNAs from Fig. 2.4.:chicken
Gallus domesticus Vtg II; clawed frog Xenopus laevis Vtg A2; white
sturgeon Acipenser transmontanus Vtg; mummichog Fundulus heteroclitus
Vtg; silver lamprey Ichthyomyzon unicuspis Vtg; plus three invertebrate
Vtg cDNAs; nematode Caenorhabditis elegans Vtg 5; boll weevil
Anthonomus granis Vtg; and mosquito Aedes aegypti Vtg. The Gallus
Vtg was designated as the reference sequence and the C. elegans Vtg was
defined as the outgroup. The number of reconstructed changes in amino
acid sequence occurring along each branch are shown without parentheses;
bootstrap data are depicted at partition boundaries as percentages in
parentheses.

41
from being completely absent in C. elegans (not shown; Speith et al., 1985), to a small
size in F. heteroclitus (99 Ser within 171 a.a. region), to a larger size in the chicken
Gallus (132 Ser within 291 a.a. region). A trend emerges in consideration of these data:
as one proceeds up the vertebrate phylogenetic ladder, Vtg polyserine domains appear
to increase in size. However, at least two exceptions to this trend have been reported:
the lamprey, Ichthyomyzon Vtg possesses a poly serine domain larger than that of F.
heteroclitus (113 Ser within 238 a.a. region; Fig. 2.2) and the
Gallus Vtg III (Byrne et al., 1989) contains a small polyserine domain (37 Ser; not
shown).
Although the vertebrate Vtg polyserine domains vary in size and serine content
as described above, their genetic organizations have sustained an element of similarity.
At the DNA level, the F. heteroclitus polyserine domain contains a distinct cluster of
TCX serine codons directly preceding a larger cluster of AGY serine codons (Fig. 2.2),
a pattern that is found in all other vertebrate Vtg cDNAs. When this cluster organization
was observed by Byrne et al. (1989) in Xenopus and chicken Vtgs, it was speculated that
a non-tetrapod Vtg would perhaps contain a cluster of only one type of serine codon,
representing the original trinucleotide repeating unit, and thus the original Vtg polyserine
domain. However, the polyserine domains presented here from the lamprey, sturgeon,
and mummichog are all dominated by the same two serine codons as is seen in Xenopus
and chicken, suggesting that these two codon clusters have been present within the Vtg
gene since before the divergence of agnathans and

42
Vtg POLYSERINE DOMAIN
Total
% Serine
SAGY
STCX
Funduius
ATtY Codons
codons
171
codons
58
codons
63
codons
36
jiiiriudlHuw I44 iHtd 14
TCX Codons
Acipertser
f MMf < f
m i AGY Codons
212
37
51
49
i M UMHMM4 t l MMi
i* TCX Codons
Ichihyomyzon
1
i AfrY CdonS
238
48
55
45
'mu ¡titt u :ui
* TCX Codons
Xenopus
f f i'??7?'?r?7iT?|
m AGY Codons
249
39
53
47
4
' TCX Codons
Gxiilus
, ,anwwm 291
45
80
20
* m4
' TCX Codons
i =
one AGY serine codon
1 =
one TCX serine codon
20 codons
Figure 2.6. A comparison of the serine codon usage in the polyserine domains (see
Fig. 2.4) of five vertebrate Vtgs. Although the number of trinucleotide
repeats vary, the overall codon structure is conserved: a cluster of TCX
codons at the 5 end precedes a larger cluster of AGY codons. Only TCX
or AGY codons are shown. Relative lengths of polyserine domains are
drawn to scale.

43
gnathostomes, over 400 million years ago (Lovtrup, 1977). Chen et al. recently
described a mosquito cDNA sequence which codes for a Vtg containing three separate
polyserine domains; 82% of the serines in these domains are coded for by the TCX
codon. Since insects are a highly derived group, it remains unclear whether the TCX
repeats represent the conservation of a primitive polyserine coding domain or an
incidence of convergent evolution between separate Vtg clades.
It has been theorized that the phosphoserine clusters of Pv, documented to bind
Ca++ in a 1:1 stoichiometric ratio in Xenopus (Folletand Redshaw, 1968; Munday et al.,
1968; Wallace, 1970) are necessary for early bone mineralization in vertebrate embryos.
Even more speculative is the idea that the phosphoserine tracts of Vtg were a necessary
pre-adaptation allowing the original evolutionary emergence of ossified bone in ancestral
chordates. Both the lamprey and the sturgeon are examples of cartilaginous vertebrates,
albeit with bony ancestors (Jarvik, 1980), that have retained their Vtg polyserine
domains. Thus, the possession of a Vtg polyserine domain is not universally concomitant
with the possession of a bony skeleton. Indeed, it appears that polyserine domains can
no longer be considered an exclusive vertebrate Vtg characteristic. Recent reports by
Chen et al. (1994) describing a mosquito (Aedes aegypti) Vtg cDNA and Yano et al.
(1994) describing a silkworm (Bombyx mori) Vtg cDNA, provide invertebrate sequences
containing various arrangements of polyserine tracts. These findings suggests a pre-
chordate origin of Vtg polyserine domains and challenges the hypothesis of Pvs being
unique to chordates. However, polyserine domains are not synonymous with true Pv
domains. Whether these invertebrate polyserine tracts are highly phosphorylated and

44
cleaved, as are bona fide Pv proteins, has not yet been reported. It is possible that
polyserine domains have existed within Vtgs since before the emergence of chordates,
but that Pv proteins, per se, remain a unique chordate trait, representing a novel
modification and utilization of these polyserine tracts.
Though we know little of why Vtg polyserine domains vary in size, findings from
studies of heritable disease may offer clues as to how these size differences originated.
The aberrant amplification of trinucleotide repeats from one generation to another has
recently been coupled to the occurrence of several human genetic diseases including
Huntingtons Disease (Huntingtons Disease Collaborative Research Group, 1993; review
by Caskey et al., 1992). An increased potential for trinucleotide amplification may
explain the faster rate of evolution attributed to the Pv region in comparison to its two
flanking Lv regions (Byrne et al., 1989). We are aware of very few descriptions of
"yolk-based diseases" in fish (Olin and von der Decken, 1989), and in these it was
neither suspected nor tested whether the disease was caused by aberrant amplification of
the Pv polyserine domain. Diseases aside, novel duplications or omissions in the Pv
polyserine domain may certainly have affected the evolution of specific yolk structures
or functions. F. heteroclitus, possessing the smallest polyserine domain of our
alignment, produces a yolk which remains totally soluble throughout oocyte development.
As the smaller yet serine-enriched polyserine domain of F. heteroclitus Vtg is
phosphorylated and finally processed into a more soluble Pv yolk protein, it may
somehow be prevented from re-combining with the Lv yolk proteins and forming the
insoluble yolk complexes of other vertebrates. Another possible explanation for the

45
persistence of a liquid phase yolk in F. heteroclitus oocytes is that the high proteolytic
activity documented by Greeley et al. (1986) prevents the recombination of Pv and the
Lvs into their usual insoluble particles. By obtaining more examples of Vtg protein
structure from other liquid phase yolk producers, a more substantial and, hopefully,
causal difference between soluble and non-soluble yolk will materialize.
In conclusion, the F. heteroclitus Vtg cDNA along with its amino acid translation
represents the first complete Vtg sequence documented from a teleost fish. The predicted
primary structure suggests to us that a heightened proportion of phosphoserine in the
polyserine domain endows the F. heteroclitus Pv yolk proteins with a higher solubility
preventing the formation of non-soluble yolk particles as is seen in many other
vertebrates. Knowledge of the complete primary structure of F. heteroclitus Vtg
provides us with useful information for mapping the extensive proteolytic processing of
native Vtg into its respective yolk proteins. We hope that this sequence will aid
investigators of other vertebrate Vtgs by providing a piscine model for molecular probes
and antibodies. Finally, we have provided yet another example of the evolutionary
independence of Pv within the Vtg gene, where the codon cluster organization is
preserved, yet the size of the serine clusters and intervening regions remains quite
unpredictable.

CHAPTER 3
SEQUENCE COMPARISON OF FUNDULUS HETEROCL1TUS
VITELLOGENINS I AND II
Introduction
Vitellogenin gene families have been described from various metazoan species
including Xenopus laevis (Wahli et al.,1979; Wiley and Wallace, 1980; Tata et al. 1980),
Caenorhabditis elegans (Blumenthal et al., 1984), and chicken (Evans et al., 1987; Byrne
et al., 1989). These small gene families from individual species have likewise been
shown to share genomic organization and sequence identity, establishing these related Vtg
genes as members of an ancient gene superfamily (Speith et al., 1985; Nardelli et al.,
1987; Byrne et al., 1989; Speith et al., 1991).
The existence of four X. laevis Vtg genes can be partially explained by the
hypothesis that an ancient duplication occurred in the X laevis genome (Thiebaud and
Fischberg, 1977; Bisbee et al., 1977). Tata et al. (1980) reported the extraordinary
occurrence of twelve to sixteen Vtg genes in X. laevis, stating that only four to six of
them were in an expressible form, and that the rest were nonexpressible or "silent".
Another example of a silent Vtg gene has been documented among the six Vtg genes of
C. elegans; whereas vit-2 to vit-6 have been shown to encode specific YP proteins, vit-1
has been described as a pseudogene (Speith et al., 1985) with no apparent translation
46

47
product. In contrast, the three Vtgs genes reported from the chicken include no apparent
silent genes. The primary translation products Vtgl, Vtgll, and VtgUI were found to be
present in blood in a ratio of 0.33 : 1.00 : 0.08 confirming chicken Vtgll as the major
yolk protein precursor (Wang et al., 1983). Recently two Vtgs have been reported in
related tilapia species, Oreochromis aureus (Ding et al., 1989) and O. mossambicus
(Kishida and Specker, 1992). These studies established the occurrence of two piscine
Vtgs (180 kDa and 130 kDa) using an immunological approach. The immunological data
from O. aureus was additionally complemented by a small nucleotide sequence from the
C-terminus of one of the purported Vtgs, probably the larger (Ding et al., 1990).
Though the existence of multiple Vtgs has been established in these species, it remains
unclear as to why several Vtgs would be functionally necessary.
We have recently reported the cDNA sequence and predicted primary structure
of Fundulus heteroclitus Vtg I, as a precursor to non-crystalline, liquid phase yolk
proteins (LaFleur et al., 1995). Here we describe a second F. heteroclitus Vtg cDNA and
protein sequence that we have designated as Vtg II. The predicted primary structure
shares 45% identity with Vtg I (with regions as high as 65%) and contains the same
general domain profile: a large lipoveitellin 1 region, followed by a serine-rich, phosvitin
region and terminating in lipovitellin 2 region. We have confirmed Vtg II MRNA
expression as well as a derived yolk protein cleavage product, verifying that Vtg II
represents a separate but functional Vtg. This report therefore, establishes the existence
of a bona fide Vtg gene family in F. heteroclitus that acts as a precursor to liquid phase
yolk proteins.

48
Material and Methods
Chemicals
Estradiol-17/3 was obtained from Sigma Chemical Co. (St. Louis, MO).
Radioisotopes, [c*-32P]dCTP and [a-35S]dATP, were purchased from New England
Nuclear (Boston MA). Lambda gtlO vector, cDNA synthesis reagents, the subcloning
plasmid Pgem-T, and T4 ligase were obtained from Promega (Madison, WI). All
amplification reactions were performed using a 50:1 mixture of Taq DNA
polymerase:cloned Pyrococcus furiosus DNA polymerase (Stratagene, La Jolla, CA).
All sequencing gels were cast using Sequagel-8 (National Diagnostics, Atlanta)
polyacrylamide reagents. Sequenase version 2.0 DNA polymerase and dideoxy
sequencing reagents were obtained from US Biochemicals (Cleveland, OH). Reagents
for random-primed labeling of probes were purchased from Pharmacia (Piscataway, NJ).
Magna nylon transfer membranes were used for nucleic acid transfers and purchased
from MSI (Westboro, MA). Amino acid N-terminal sequencing, synthesis of
oligonucleotide primers, and a limited amount of DNA sequencing were performed by
the University of Florida Interdisciplinary Center for Biotechnology Research core
facilities.
Cloning strategy using an estrogen-induced liver cDNA library
Seven of the eight overlapping clones resulting in the contiguous cDNA sequence
were isolated from a XgtlO liver library whose synthesis has been previously described

49
(LaFleur et al. 1995) In brief, the library was constructed from the pooled mRNA of
six male Fundulus heteroclitus that had been treated with two IP injections of estradiol-
17/3 (0.01 mg/g body weight). The library contained an initial titer of only 6 X 104 total
plaque-forming units, and had been amplified twice.
The initial Vtg II clone was discovered using the degenerate primer ROW 19, and
the vector primer NEB 1231, with 5 ¡A of the XgtlO library as template in a PCR
reaction utilizing a 50:1 mixture of Taq DNA polymerase:Pfu DNA polymerase. ROW
19 was designed to match a conserved region of Vtgs, ranging from C. elegans to
chicken (Fig. 3.1). A 550 bp band was isolated, inserted into pGem-T, sequenced and
revealed to be a second Vtg cDNA that we designated as Vtg II. This insert was then
isolated and used to generate a random primed 32P-labeled probe. The library was plated
out on 150-mm petri dishes by transfecting E. coli C600hfl cells, and overlaying them
in agarose atop agar plates containing 25 tg/ml tetracycline. Duplicate plaque lifts were
carried out using Magna nylon membranes, and these were probed at 65C in 0.05 X
BLOTTO, 6 X SSC (150 mM NaCl, 15 mM sodium citrate, pH 7) overnight. A large
proportion of the plaques were found to be positive, and 20 agarose plugs were isolated
and stored in SM buffer (. 1 M NaCl, 8 mM MgS04, 50 mM Tris, 2% gelatin) at 4C
with a drop of chloroform. Thereafter, plug lysates from these Vtg II positive plugs
were used in amplification reactions targeting Vtg II positive clones in a successively
overlapping 5 direction. Six more Vtg II clones were isolated in this manner
approaching the initial methionine codon, but several attempts at targeting the last few
nucleotides to include the initial methionine failed.

50
Fundulus heteroclitus Vitellogenin II cDNA 5195 bp
ROW 55
pFhv2h
pFhv2a
ROW 19
pFhv2b
pFhv2g pFhv2f pFhv2c
pFhv2e
pFhv2d
Figure 3.1 Cloning strategy used in isolating the F. heteroclitus Vtg II cDNA (5166
bp). Seven inserts (pFhv2a thru g) were isolated from the XgtlO liver
library by anchored PCR with indicated oligonucleotide primers and
inserted into the pGem-T cloning vector. The final cDNA (pFhv2h) was
isolated by RACE using reverse primer ROW 55.

51
A protocol for rapid amplification of cDNA ends (RACE; Frohman et al., 1988),
was performed to retrieve this region using total RNA (described below) isolated from
the liver of an individual reproductively active female F. heteroclitus. A first strand
synthesis reaction was performed using 0.5 ng total RNA, the primer ROW 55 and
Superscript RT, followed by addition of a "poly-C tail" using 4 f of 1.0 mM dCTP, and
10 units of terminal deoxynucleotidyl transferase (BRL). Then, an amplification reaction
was carried out using the forward primer ROG 51, which targeted the poly C-tail, along
with the reverse primer ROW 55, and the Taq DNA polymerase. Through this effort
we successfully isolated a 230-bp band that was inserted into pGem-T, sequenced and
found to include a valid methionine codon, preceded by a short region that fit the criteria
for a transcription start site (Kozak, 1991).
Estrogen treatment. RNA isolation and analysis
Male and female F. heteroditus were collected from the estuarine creeks adjacent
to the Whitney Laboratory, and were maintained in running seawater tanks under
14L:10D photoperiod conditions at 25 + 2C. Fish were maintained for at least one
month before being used for RNA collections.
Experimental groups of fish were subjected to two intraperitoneal injections of
estradiol-17/3 (0.01 mg/g body weight) dissolved in 50 /fi coconut oil (Kanungo et al.
1990). Control groups were sham-injected with coconut oil alone. The first injection
was performed on day 1, the second injection on day 4, followed by sacrifice and liver
dissection on day 8.

52
Total RNA was isolated from livers by extraction with RNA Stat-60 reagents
(Tel-Test "B", Inc. Friendswood, TX). Tissues were dissected and immediately frozen
in 1.5-ml tubes containing 500 /zl of RNA Stat-60 emulsion by immersion in liquid
nitrogen. Tissues were homogenized at 20C using a Kontes pestle and motor.
Typically, a 300 mg liver yielded 0.350 mg total RNA, with O.D. 260/280 ratios
consistently above 1.8. Total RNA samples were resuspended in diethyl pyrocarbonate-
treated water and stored at -80C until used in analyses.
Before electrophoresis, aliquots of 15 ¡xg total RNA were precipitated in
isopropanol, and denatured in 2.2 M formaldehyde, 50% formamide, 50 mM MOPS (pH
7.0) for 30 min at 65C. Samples were electrophoresed through gels containing 1.0%
agarose, 0.6 M formaldehyde, 50 mM MOPS, and 1 mM EDTA for 2.0 hours at 3.5
V/cm gel in 50 mM MOPS, 1 mM EDTA running buffer. RNA was blotted onto Magna
nylon membranes by capillary action with 20 X SSC, immobilized by U.V. crosslinking
and visualized by staining briefly with methylene blue.
Random-primed [32P]probes were made for resolving Vtg I and Vtg II RNA
transcripts. The Vtg I probe was synthesized from a PCR product off of the template
pMMBl using primers ROW 5 and MB 13, resulting in a 639-bp cDNA probe from
nucleotide 4284 to 4923 of the Vtg I cDNA. The Vtg II probe was made from pFhv2a
using primers ROW 19 and ROW 33, yielding a 277-bp probe from nucleotide 4692 to
4969 of the Vtg II cDNA. After random prime labeling, oligonucleotide probes were
separated from non-incorporated [32P]dCTP by size chromatography through Stratagene

Figure 3.2 Translated amino acid sequence (1687 residues) of the putative F.
heteroclitus Vtg II polypeptide. The signal peptide, predicted by
the method of von Heijne (1986) is indicated by underlining, and
verified by the N-terminal sequence obtained from an isolated 69-
kDa yolk protein (shaded lettering). The annealing site of ROW
19, used to isolate the initial insert, is indicated by double
underlining. A polyadenylation site is indicated by underlining.

54
aacrcaccagcc 12
ATGAGGGTGCTTGTGCTGGCTCTCACTGTGGCCCTTGTGGCCGGGAACCAGGTGAGCTATGCCCCA 78
W R V t V L A.. L._ T V A £ V" A G.. N Q 7 S A P
GAATTTGCCCCTGGAAAGACCTACGAGTACAAGTATGAAGGTTATATTCTGGGTGGCCTGCCTGAG 144
EFAPGKTYE7K7EG7ILGGLPE
GAGGGCCTGGCAAAGGCTGGGGTGAAGATCCAGAGCAAAGTCTTGATCGGTGCAGCAGGTCCTGAC 210
EGLAKAG7KIQSK7LIGAAGPD
AGCTACATTCTGAAACTTGAAGACCCTGTCATCTCGGGGTACAGTGGCATTTGGCCTAAAGAGGTT 276
SilLKLEOPVISGX'SGXWPKSV
TTCCACCCTGCCACAAAGCTCACCTCAGCTCTCTCTGCTCAGCTCTTGACACCCGTCAAGTTTGAG 342
FHPATKLTSALSAQLLTP7KFE
TATGCCAACGGAGTGATCGGAAAAGTGTTCGCACCTCCAGGCATCTCTACAAATGTGCTGAATGTC 408
YANG7IGK7FAPPGISTN7LN7
TTCAGGGGACTCCTCAACATGTTTCAGATGAACATCAAGAAGACTCAGAATG7GTATGACCTGCAA 474
FRGLLNMFQMNXKKTQNVyDLQ
GAGACTGGAGTAAAAGGTGCGTGCAAGACACACTATATCCTTCATGAGGACTCCAAGGCTGATCGC 540
ETGVKGVCKTH5TILHEDSKADR
CTCCACTTGACGAAAACCACAGACCTGAATCACTGCACCGACAGCATCCACATGGATGTTGGCATG 606
LHLTKTTOLNHCTDSIHMD7GM
GCTGGTTATACGGAAAAATGTGCAGAGTGCATGGCTCGGGGAAAAACTC7TT CAGGAGCAATTTCT 672
AGKTEKCAECMARGKTLSGAIS
GTCAACTACATCATGAAGCCGTCTGCCTCTGGCACCTTGATCCTAGAGGCAACCGCCACTGAGCTT 738
VNXIMKPSASGTLILEATATEI.
CTCCAGTACTCGCCCGTCAACATTGTAAATGGAGCTG7CCAGATGGAGGCTAAGCAGACCGTGACC 304
LQYSPVNIVNGAVQMEAKQTVT
I
TTCGTGGACATCAGGAAGACCCCATTAGAGCCCCTCAAAGCAGACTATATTCCCCGTGGATCGCTC 870
FVOIRKTPLEPLKADYIPRGSL
AAGTACGAGTTAGGCACTGAATTCCTACAGACACCAATTCAGCTTCTGAGGATCACCAATG7CGAG 936
KYELGTEFLQTPIQLLRITilVE
GCTCAGATTGTTGAGTCTCTGAACAACCTAGTGAGCCTCAATATGGGCCATGCCCATGAGGATTCC 1002
AQI7ESLNNL7SLNMGHAHEDS
CCTCTGAAGTTTATTGAGCTCATCCAGCTGCTGCGTGTGGCCAAGTATGAGAGCATTGAAGCTCTC 1063
PLKFIELIQLLR7AK2ESIEAX.
TGGAG7CAGTTTAAAACCAAAATTGATCACAGGCACTGGTTGCTGAGCTCTATCCCTGCCATTGGT 1134
WSQFKTKIDHRHWLLSSIPAIG
ACTCATGTTGCTCTCAAGTTCATCAAGGAGAAGATCGTTGCTGGTGAAGTCACTGCTGCTGAGGCT 1200
TH7ALKFIKEKI7AGE7TAASA
GCTCAGGCCATCATGTCATCTACACACTTGGTGAAGGCCGACCTGGAGGCAATCAAGCTTCAGGAG 1266
AQAIMSSTHL7KA0LEAIKLQE
GGCCTGGCTG7GACCCCTAATATTCGGGAAAATGCAGGTTTGCGTGAACTCGTTATGCTGGGCTTT 1332
GLA7TPNIRSNAGLREL7MLGF

55
GGCATCATGGTrCACAAATACTGTGTGGAGAACCCTTCATGTCCATCrGAGCTGGTCAGGCCAGTT 1398
GXMVHKYCVENPSCPSELVRPV
CATGACATTATTGCCAAGGCTCTTGAGAAACGCGACAATGATGAGCTCTCCCTGGCrCTCAAAGTT 1464
HDIIAKALEKRDNDELSLALKV
CTGGGTAATGCCGGACATCCCAGCAGCCTGAAGCCAATCATGAAACT'rCTTCCTGGCTTTGGCAGC 1530
LGNAGHPSSLKPXMKLLPGFGS
TCTGCCTCCGAACTTGAGCTCAGAGTTCACATTGACGCTACACTGGCGCrGAGGAAAATTGGCAAG 1596
SASELELRVHIDATLALRKIGK
AGAGAACCCAAGATGATrCAGGATGTGGCCCTTCAGCTCTTCATGGACAGGACTCTTGACCCAG AG 1662
RSPKMIQDVALQLFMDRTLOPS
CTCCGTATGGTTGCTGTTGTTGTGCTGTTTGATACCAAGCTACCTATGGGTCTGATAACCACTCTC 1728
LRM7A777LFDTKLPMGLITTL
GC'rCAGAGTCTCCTGAAACAGCCAAACCTGCAGGTCCTTAGCTTTGTCTACrCTTACATGAAGGCC 1794
AQSLLKSPNI.Q7LSF7YSYMKA
TTCACCAAGACCACCACCCCGGACCATTCCACTGTAGCCGCTGCCTGCAATGTTGCCATCAGGATC 1860
FTKTTTPOaSTVAAACNVAIRI
CTCAGCCCAAGATTCGAAAGACTGAGCTACCGCTACAGCCGAGCTTTCCATTATGACCACTATCAT 1926
LSPRFERLSYRYSRAFaYDHYB
AATCCTTGGATGCTGGGAGCTGCTGCCAGCGCATTTTACATCAATGATGCCGCGACTGTATTGCCA 1992
NPWMLGAAASAFYINDAATVLP
AAAAACATCATGGCAAAAGCTCGCGTTTACCTCTCTGGAGTGTCTGTTGATGTTCTGGAG7TTGGA 2053
KNIMAKAR7YLSG7S7D7LEFG
GCCAGAGCTGAAGGAGTGCAAGAGGCCCTTTTGAAAGCCCGTGATGTTCCTGAGAGTGCAGACAGG 2124
ARASG7QEAI.I.KARD7PESADR
CrrCACCAAGATGAAGCAAGCTCTTAAGGCTCTGACTGAGTGGAGGGCCAATCCTTCCCGCCAGCCT 2190
OTKMKQALKALTEWRANPSRQP
CTCGGCTC7CTGTACGTGAAGG7TCTTGGGCAGGATGTTGCTTTGCAAACATCGACAAAGAAATG 2256
LGSLY7K7LGQD7AFANIDKEM
GTTGAGAAGATCATTGAGTTTGCAACTGGACCTGAAATCCGCACCCGTGGCAAAAAGGCCTTGGAC 2322
72KIIEFATGPEIRTRGKKALD
GCCCTGTTGTCTGGTTACTCTATGAAATACTCCAAGCCAATGTCGGCCATTGAGGTCCG7CACATC 2388
ALLSGYSMKYSKPMSAIS7Rai
TTCCCCACCTCTCTTGGTTTACCCATGGAGCTCAGTCTGTACACTGCTGCCGTGACAGCCGCATCC 2454
FPTSLGLPMSLSLYTAA7TAAS
G7TGAAGTACAAGCCACCATTTCACCACCACTTCCCGAGGACTTCCATCCTGCCCACCTAC7GAAG 2520
7E7QATISPPLPEDFHPAHLLK
TCTGATATTTCCATGAAGGCT7CAGTCACTCCAAGTGTATC7TTGCACACCTATGGAGTTATGGGA 2586
SO ISMKAS7TPS7SLBTYG7MG
GTGAATAGTCCTTTCATCCAGGCrrCTGTGCTGTCAAGAGCCAAAGACCATGCAGCTCTTCCCAAA 2652
7NSPFIQAS7LSRAKDBAALPK
AAGATGGAGGCAAGACTTGACATAGTCAAGGGTTACTTTAGCTACCAGTTCCTGCCTGTTGAGGGT 2718
KMEARLDI7KGYFSYQFLP7EG
Figure 3.2--continued

56
G77AAAACAA77GCA7C7GC7CG7C77GAAACAG77GCCA77GCAAGAGA7G77GAAGGCC7CGC7 2784
VK7IASARLE7VAIAR0VEGLA
GC7GCCAAAG7CACACCGG77G7CCCA7A7GAGCC7A77G7GAGCAAGAACGCCAC777AAA7C77 2 8 S 0
aakvtpvvpyepivsknatlml
TCACAGA7G7C77AC7A7C7GAA7GA7AGCA7A7CAGCA7CA7C7GAAC77C77CC77777CGC7G 2916
SQMSYYI,NDSISASSELLPFSL
CAAAGGCAAACTGGCAAAAATAAAATCCCCAAGCCCATTGTGAAGAAAATGTGTGCAACAACGTAT 2982
QRQ7GKNKIPKPIVKKMCA77Y
ACG7A7GGGA77GAGGGC7GCG77GACA777GG7C7CGCAA7GCAACC77CC7CAGAAACACCCCC 3048
7YGIEGCVDIWSRNATFI.RN7P
A7C7ACGCCA7AA77GGAAACCAC7C7C7777GG77AA7G77ACCCCAGC7GC7GGACCG7CCA7C 3114
IYAIIGNHSLLVNVTPAAGPSI
GAAAGGA7CGAAA7CGAGG77CAG777GG7GAACAAGCAGCAGAAAAGA7CC77AAAGAGG777AC 3180
ERIEISVQFGEQAAEKILKEVY
C7GAA7GAGGAGGAAGAAG7AC77GAAGACAAAAACG7CC77A7GAAGC7GAAGAAGA7TC7G7C7 3246
LNEESEVLEDKNVLMKLKKILS
CC7GG7C7GAAGAACAGCACCAAAGC77CA7CC7C7AG77CGGGCAGC7C7CGC7CCAG7AGA7C7 3312
PGLKNS7KASSSSSGSSRSSRS
CGC7CCAGCAGC7CCAGCAGC7CCAGCAGC7CCAGCAGC7CCAGCCG77CC7CC7C7AGC7C77CC 3378
RSSSSSSSSSSSSSSRSSSSSS
AGGAGC7C77CC7C777GCGCCGCAA7AGCAAGA7G77GGA7C77GCCGA7CCCC7CAACA7AACA 3444
RSSSSLRRNSKMLDLADPLNIT
7CAAAGAGA7CC7CCAGCAGC7CC7CCAGC7CCAGC7CC7CCAGC7CC7CCAGC7CC7CCAGC7CC 3510
SKRSSSSSSSSSSSSSSSSSSS
7CCAGC7CCAAGACCAAG7GGCAGC7GCACGAAAGGAAC77CACCAAGGA7CACA7CCACCAGCA7 3576
sssktkwqlhernftkohihqh
7CCG7C7CAAAAGAACGTC77AACAGCAAGAGCAG7GCGAGCAGC777GAA7CCA777ACAACAAG 3642
SVSKERLNSKSSASSFESIYNK
A7CACA7ACC7G7C7AACA7CG7CAGCCCAG7GG7CACAG7CC77G7CCG7GCCA7CAGAGC7GAC 3708
I7YLSNIVSPVV7VLVRAIRAD
CACAAGAACCAGGGG7A7CAGA7CGC7G7G7AC7A7GACAAAC7CAC7ACCAGAG7GCAGA7CA77 3774
HKNQGYQIAVYYDKLTTRVQII
G7GGCCAACC7CAC7GAAGA7GACAAC7GGAGAA7C7G77C7GACAGCA7GA7GC7CAGCCACCAC 3840
VANL7EDDNWRICSDSMMLSH H
AAAG7GA7GAC7CGAG7CACC7GGGGCA77GGA7GCAAGCAG7ACAACACCACGA7CG7GGCCGAA 3906
KVM7RV7WGIGCKQYN77IVAE
AC7GG7CGCG77GAGAAGGAGCC7GCCG7CCG7G7GAAGC7GGCC7GGGCCAGAC7CCC7AC77AC 3972
7GR7EKEPAVRVKLAWARLPTY
A7CAGGGA77A7GCAAGAAGAG7G7CCAGG7ACA777CCCGCG7CGC7GAGGACAA7GGAG7GAAC 4038
IRDYARRVSRYISRVAEOMGVN
AGGACAAAGG7CGCCAG7AAACCCAAAGAGA7CAAAC7GAC7G7AGC7G77GCCAACGAGACAAGC 4104
R7KVASKPKEIKL7VAVANE7S
Figure 3.2--continued

57
C7GAA7G7CACGC7GAA7ACACCAAAGAACACC77777CAAAC7GGGA7GGG77C77CCC7777AC 4170
LNVTLNTPXNTFFXLGWVLPFY
C7ACCAA77AACAACAC7GC7GC7GAGC7GCAGGCA77CCAGGGCAGG7GGATGGACCAGG7CACA 4236
L? INNTAAELQAFQGRWMDQVT
7ACA7GC7CACCAAG7CTGC7GCAGC7GAG7GCACCG7GG77GAAGACACAG7GG7CAC777CAAC 4302
YMLTXSAAAECTVVEDTVVTFN
AACAGGAAG7ACAAAACGGAGACGCCCCAC7C77GCCATCAGG7C77GGC7CAAGAT7GCACA7C7 4368
¡RXYXTSTPHSCHQVLAQDCTS
GAAA7CAAAT7CA7AG7GC7GC7GAAGAGGGA7CAAACAGCAGAACGGAA7GAGA7CAG7ATTAAG 4434
EIXFIVLLXRDQTAERNEISIX
A7TGAAAACATTGA7G77GACA7GTATCCCAAGGACAACGC7G77GTGG7GAAGG77AATGGAG7A 4500
IENIOVDMYPXDNAVVVXVNGV
GAAAT7CC7C7CACCAACC7GCCATATCAGCATCCAACAGGCAACATACAGATCCGACAAAGAGAA 4566
EIPLTNLPYQHPTGNIQIRQRE
GAGGGCA7C7C7C7GCATGC7CCCAG7CA7GGCC77CAGGAGG7C77CC7CAG777AAACAAAG7G 4632
EGISLHAPSHGLQSVFX.SLriKV
CAGG77AAAG77G77GAC7GGATGAGAGGCCAGACG7G7GGGC7C7GCGGAAAGGCCGACGGGGAA 4698
QVXVVDWMRGQTCGLCGXADGE
G7CAGACAGGAG7ACAGCAC7CCCAATGAACGGG7G7CCAGGAACGCAACCAGC77CGC7CAT7CC 4764
VRQEYSTPNERVSRNATSFAHS
TGGG7GC7GCC7GCAAAGAGC7GCCG7GACGCC7CAGAG7GC7ACATGCAAC77GAATCGG7GAAG 4830
WVLPAXSCRDASECYMQLESVX
C7CGAGAAACAGA7CAGCC7GGAAGGCGAGGAA7CCAAATGC7AC7CAG7CGAACC7GTC7GGCGC 4896
LSXQISLEGEESXCYSVEPVWR
7G7C7CCC7GGC7G7GCACCAG7GAGAACCACC7CCG7CAC7G7CGGGC7ACCA7GCGTG7C7C7G 4962
CLPGCAPVRTTSVTVGLPCVSL
GA77CAAACC7GAA7CGC7C7GATAGTC7CAGCAGCATC7A7CAGAAGAGCG77GACGTGAGCGAG S 0 2 8
DS.VLNRSOSLSSIYQXSVDVSE
ACGGCAGAG7CCCACC7GGCC7G7CGC7GCAC7CC7CAG7G7GCC7AAacgtgttgcctccrgac= 5094
TASSHLACRCTPQCA-
tttcg'ctctg'cttttgrgttacatggacgctcgcaaactaaaataaagaagcaactaaaaaaaaaaa 5160
aaaaaattcagctttggacttaacoaggctgaacct 5195
Figure 3.2--continued

58
Nuc-Trap columns. All RNA hybridizations were carried out at 65C in 1 X Denhardts
solution, 6 X SSC, and 0.1 % SDS without formamide (Denhardt 1966).
Autoradiographs were analyzed using the Bio Image Whole Band Analyzer system
(Millipore, Ann Arbor). For estimating amounts of Vtg RNA visualized on gels, RNA
was transcribed from the Vtg I plasmid pMMBl and the Vtg II plasmid pFhv2a, using
Ambion reagents. Transcribed RNA yields were measured spectrophotometrically, and
diluted to a concentration of 66.7 pg//xl. For RNA standards, 133 pg transcribed RNA
from both pMMBl and pFhv2a was loaded onto each gel.
Results
The complete cDNA sequence (5166 bp) of a Vtg mRNA, encoding a protein
designated as Vtg II is provided in Figure 3.2. The eight overlapping pGem-T clones
that were used to complete the sequence are represented in Figure 3.1. A ClustalV
alignment of Vtgs I and II by the method of Swofford et al. (1993) revealed 45%
sequence identity between the two amino acid sequences (Fig. 3.3).
In general, the two sequences share the same profile as other reported Vtgs: a
large lipovitellin 1 region that is followed by a polyserine domain (assumed to represent
phosvitin) that, in turn, is followed by a lipovitellin 2 region containing a
substantial amount of conserved cysteines. Like Vtg I, Vtg II contains several predicted
N-glycosylation (16), phosphorylation (45), and N-myristoylation sites (16), agreeing
with our expectations for a lipophosphoglycoprotein. The smaller length of the Vtg II
a.a. sequence (1687) compared to that of Vtg I (1704) can be primarily attributed to gaps
in the polyserine domain. A graphical comparison of the polyserine domains of Vtg

Figure 3.3 ClustalV alignment of F. heteroclitus Vtg I and Vtg II. A
polyserine domain defined according to a previously published
alignment (LaFleur et al. 1995) is indicated by shaded lettering.
Identical residues are denoted by asterisks. Vtg I and Vtg II share
45% overall sequence identity.

Vtg :
Vtg II
MXAWLALTLAFVAGQN F APEFAAGXTYVYXYEAL ILGGLPSEGI.ARA
MRVLVLALTVAI.VAGNQVS1APEFAPGXTYEYKYEG7II.GGI.PSEGLAXA
48
50
Vtg I
Vtg II
GLXISTXLLLSAADQNTYMLXLVEPELSEYSGIWPKDPAVPATXLTAALH 98
GVKIQSKVLIGAAG? D S YILXLEDPVISGYSGIWPK2VFHPATXLTSALS 100
IT* *#*
Vtg I
Vtg II
LSSQFPSSLJJTPMVFVGXVFAPEEVSTLVLNIYRGIUJILQLNIXXTHKV 148
AQLL TP VKFS Y ANG VIGKVFAP PGISTNVLNVFRGLLNMFQMNIXXTQNV 150
****** * * ** ** ***** *
Vtg I
Vtg II
YDLQEVGTQGVCKTLYSISEDARISNILLTX7RDI.SNCQERLNKDIGEA
YDLQETGVXGVCXTHYILHEDSXADRI.HI.TXTTDLNHCTDSIHMDVGMAG
197
200
Vtg I
Vtg 13
YTSXCORCQESTXNLRGTTTLSYVLXPVADAVMII.XAYVNEI.IQFSPFSE
YTSXCAECMARGXTLSGAISVNYIMXPSASG7LILEATATE2I.QYSPVNI
247
250
Vtg I
Vtg II
AIIGAAQMRTXQSI.2FLSIEXEPIPSVKAEYRHRGSLXYEFSDELI.QTPI.Q 297
VNGAVQMEAXQTVTFVDIRKTPL3PLXADYIPRGSLXYELG7EFLQTP IQ 300
* * ******* *** *
* * * *
Vtg I
Vtg II
LIXISDAPAQVAE'/LXHLATYNIEDVHENAPLXFLEI.VQLLRIARYEDLS
LLRITNVEAQIVESLNNLVSLHMGHAHEDSPLXFIELIQLLRVAXYESIS
347
350
Vtg I
Vtg II
MYWNQYKXMSPHRHWFLDTIPATG7FAG.RFIXEXFMAEEITIAEAAQAF 397
ALWSQFKTXIDHRHWLLSSIPAIGTHVALXFIXEXIVAGEVTAAEAAQAI 400
* * **** *** ** ***** * ******
Vtg I
Vtg II
17AVHMVTAD P EVIXLFE SL VO S D XWENPLUIE WFLG Y GTMVNXY CUR
MSSTHLVKADLEAIXLQEGI.AVTPNIRENAGI.REI.VMLGFGIMVHXYCVE
447
450
Vtg I
Vtg II
TVDCPVELIXPIQQRLSDAIAXNEEENIILYIXVLGNAGHPSSFXSLTXI 497
NPSCPSELVRPVHDIIAXALZXRDNDELSLALXVLGNAGHPSSLXPIMKL 500
* TT # * ****** *
Vtg I
Vtg II
MPIHG7AAVSLPMTIHVEAIMAI.RNIAXXESRMVQELALQLYMDKALHPS 547
LPGFGSSASELELRVHIDATLALRRIGXREPKMIQDVALQLFMDRTLOPS 550
* * * *** * *
* * *
Vtg I
Vtg II
LRMLSCIVX.FE7SPSMGLVTTVANSVKTSENLQVASF7YSHMXSLSRSPA
I.RMVAWVLFDTXLPMGLITTLAQSI.I.X2PNI.QVLSFVYSYMKAFTXTTT

597
600
Vtg I
Vtg II
71HPDVAAACSAAMXILGTXLD RLSLRYSXAVHVD L YNS SLAVGAAATAF 647
POHSTVAAACrrVAIRILSPRFSRLSYRYSRAFHYDHYHNPWMLGAAASAF 650
* **** ** * ** # *
Vtg I
Vtg II
YINDAATFMPXSFVAXTXGFIAGSTAEVLEIGANIEGIQELILXNPALSE
YIIIDAAT/I.PKNIHAXARVYLSGVSVDVLEFGARAEGVQEALIXARDVPE
697
700
Vtg I
Vtg II
S70RI7XMKRVIXALSEWRSLPTSXPEASVYVRFFGQEIGFANIDXPMID 747
SADRLTXMXQALXALTEWRANPSRQPLGSLYVKVLGQDVAFANIDREMVE 750
** ** ** ** ** *** * ****** *
Vtg I
Vtg II
XAVKFGXELPIQEYGREAI.XALLLSGINFHYAXPVIAAEMRRILPTVAGI 797
XIIEFATGPEIRTRGXXALDALI,-SGYSMXYSXPMSAIEVRHIFPTSI.GL 799
* * ** *** ** * * *
Vtg I
Vtg II
PMELSLYSAAVAAASVEIXPNTSPRLSADFDVKTLI.ETDVELXAEIRPMV
PMELSI/YTAAVTAASVEVQATISPPLPEOFHPAHLLXSDISMXASVTPSV
847
849
Vtg I
Vtg II
AMDTYAVMGLNTDIFQAALVARAKLHSWPAXIAARLNIKEGDFKLEALP 397
SLHTYGVMGVNSPFIQASVLSRAXDHAALPKXMEARLDIVKGYFSYQFI.? 899

vtg :
vtg ::
VDVPSNirSMNVTTFAVARNIEEPLVERITPLLPTXVLVPIPIRRHTSKI.
VEGVK7IASARLETVAIARDVEGLAAAXVTPWP YSPIVSKNATLNL
* * * ** * * ** *
947
946
Vtg I
vtg n
DPTR MSMLDSSEPME SSDVEPIPEYXFRRFAKKYCAKHIGV
SQMSYYLNDSISASSELLPFSLQRQTGXNXI? X? IVXKMCATTYTY
* ****** ** ** **
990
992
vtg I
Vtg XI
GLKACFXFASQNGASIQDIVLYXLAGSHNFSFSVTPIEGEWERI.2MEVX
GIEGCVDIWSRNATFLRNTPIYAIIGNHSLLVNVTPAAGPSIERISIEVQ
o .*
1040
1042
Vtg I
Vtg II
VGAJCAAEXiVKSINLSEDEETSEGGPVLVXLNKtSSHRNSSSSSSSlSsS
FGEQAAEXILXEVYLNEESEVLSDKNVLMKI.XXILSPGt,XNSTXASSSSS
* *** * ** ** ** *** *****
1090
1092
Vtg I
Vtg II
SSSSSaSSRSSSSSSSSSRSSRlCIDEAARTSSSSSSSSRRSRSSSSSSSS
GSSRSSSSRSSSSSSSSSSSS SSRSSSSSSRSSSSLRRNSK
* * W******* ** ***** ** *
1140
1133
Vtg I
Vtg II
SSSSSSSSSSSSRRSSSSSSSSSSSSSRSSRRVNSTRSSSSSSRTSSASS
MLBLAPgEMITSXRSSSSSSSESSSSSSSS: SSSSSSXTKWQLH
vw**' "w** * * ** ir" * jir-r ****** *
1190
1176
Vtg I
Vtg II
LASFFSDSSSSSSSSORRSkXVMEXFQRLHXKMVASGSSASSVSAIYXEX
SRNF ; tX0HIHQHSVSK-3RUSK--S3ASSF5SI?nCX
1240
1211
Vtg I
Vtg II
XYLGEE- SAWAVILRAVKADKRMVGYQLG FYLD KPMARVQ11 VAN X SSD
TYLSNIVSPWTVLVRAIRADHXNQGYQIAVYYDXLTTRVQIIVANLTED
1289
1261
vtg I
vtg II
SNWRICADAWLSKHKVTTKISWGEQCRKYSTNVTGETGIVSSSPAARLR
DNWRICSDSMMLSHHKVMTRVTWGIGCXQYNTTIVAETGRVEXSPAVRVK
***** * * * * *** ** *
1339
1311
Vtg I
Vtg II
vswerlpstlxrygxmvnkyvp-vkilsolihtxrenstrnisviavats
LA WARL ?T YIRD Y ARR VS R Y1S RVAED NGVNRTXVASX? X2IXLT/A VAN
# * n * * *
1388
1361
Vtg I
Vtg II
EXTIOIITXTPMSSVYNVTMHLPMCIPIDEIXG-LSPFDEVIDKIHFMV
etslnvtlntpxntffxlgwvlpfylpinntaaelqafqgrwmdqvty.ml
* * * V *
1436
1411
Vtg I
Vtg II
S XAAAAE CS FVEDTL YTFNNRS YXNKMP S S CYQVAAQD CTD ELXFMVLLR
TXSAAAECTWEDTWTFNNRKYXTSTPHSCHQVLAQDCTSEIXFIVLLX
1486
1461
Vtg I
Vtg II
KD-SSEQHHINVXISEIDIDMFPXDDNVTVXVNEMEIPPPACLTATQQL?
RDQTAERNEISIXIENIDVDMYPXDNAVWKVNGVEIPLTNLPYQHPTGN
* * ** ** ** *** **** ***
1S3S
1311
vtg I
vtg II
LKIXTXRRGLAVYAPSHGLQEVYFDRXTWRIKVADWMKGXTCGLCGXADG
IQIRQREEGISLHAPSHGLQEVFLSLNXVQVKWDWMRGQTCGLCGXAflG
isas
1561
Vtg I
Vtg II
EIRQE YHTPNGRVAXNSIS F AHSWIL? AESCRD AS ECRLXLESVQLEXQL
3VRQEYSTPNERVSRNATSFAHSWVLPAXSCRDASECYMQL2SVKI.EXQI
1635
1611
Vtg I
Vtg II
TIHGEDSTCFSVEPVPRCLPGCLPVXTTPVTVGFSCLA SDPQT
SLEGESSXCYSVEPVWRCLPGCAPVRTTSVTVGI.PCVSLDSNLNRSOSLS
1678
1661
Vtg I
Vtg II
SVYD RSVD LRQT7QAHLACS CNTKCS
SIYQKSVDVSETAESHLACRCTPQCA
* ** **** *
1704
1637
Figure 3.3--continued

62
I and II (Fig. 3.4) reveals a departure from a trend that had previously been noted
concerning serine codon usage in Vtg I (LaFleur et al., 1995) and other vertebrate Vtgs
(Byrne et al., 1989) Whereas the polyserine domains of most vertebrate Vtgs contain a
cluster of TCX codons at the 5 side of the polyserine coding domain and a cluster of
AGY codons at the 3 side, the Vtg II polyserine domain appears to have these codons
equally dispersed, with no obvious clustering.
Northern blot analyses showed that the mRNA of Vtg II transcript can be found
in both estrogen-treated males and spawning females, at an approximate size of 6.0 Kb
(Fig 3.5). By analysis of duplicate blots with separate Vtg I and Vtg II cDNA probes,
it was found that Vtg II transcripts numbered ten times less than those of Vtg I. Vtg I
probes did not cross-hybridize with RNA transcribed from Vtg II clones and vice versa,
confirming that two separate mRNAs for Vtg I and Vtg II were indicated (Fig 3.5).
The N-terminal amino acid sequence of a 69 kDa protein band isolated from the
yolk protein of ovulated eggs was determined astobeNQVSYAPEFAPGxT
Y, where "x" was undetermined ("YP 69" indicated in Chapter 4). Allowing the
predicted K residue in the unidentified "x" position, this sequence provides a perfect
match for the N-terminus of Vtg II after cleavage of the predicted signal peptide (Fig.
3.2, shaded lettering) and indicates that Vtg II is not blocked as is the case with the N-
terminus of Vtg I (LaFleur et al. 1995). These data verify that the Vtg II protein is in
fact expressed, transported, and incorporated as a yolk protein precursor in oocytes of
F. heteroclitus.

63
Polyserine Domain
Number of
Ser codons
I = one AGY serine codon | j
j one TCX serine codon 20 codons
Figure 3.4 A comparison of the serine codon usage in the poly serine domains (see
Fig. 3.3) of F. heteroclitus Vtg I and Vtg II. Whereas the TCX and AGY
codons of the Vtg I poly serine domain are clustered into two separate
groups, the TCX and AGY codons of Vtg II show no apparent clustering.
Only serine codons are shown, with relative lengths of the domains drawn
to scale.

64
Discussion
F. heteroclitus Vtg II cDNA and predicted amino acid sequence are provided in
Figure 3.2. Vtg II mRNA is present in the liver of estrogen-treated males and normal,
spawning females (Fig. 3.5). The N-terminal sequence of a yolk protein isolated from
ovulated eggs was found to be identical to the predicted N-terminus of the putative Vtg
II translation product. Taken together these data indicate that the yolk proteins of F.
heteroclitus are derived from a mixture of at least two estrogen-induced liver precursors,
Vtg I and Vtg H, establishing the existence of a Vtg gene family in F. heteroclitus.
These two cDNAs represent the first two Vtg sequences from a single vertebrate species
to have been completely sequenced, offering a unique perspective into the possible
variance between Vtg isoforms occurring in single species.
Examination of the alignment of Vtg I and Vtg II reveals typical conservation of
lipovitellin regions seen among other vertebrate Vtgs. As previously described for other
vertebrate Vtgs (LaFleur et al., 1995) poor alignment occurred in the polyserine
domains. Although the tandem repeats of serine can be aligned in small stretches, the
overall lengths and intervening amino acid sequences are highly variable, resulting in a
region whose conservation is difficult to interpret. In an attempt to compare and
visualize these polyserine domains, hypothetical boundaries were drawn up according to
those used in a previous report (LaFleur et al., 1995), and a graphical representation was
created showing relative domain length as well as serine codon usage (Fig. 3.4).
Whereas the serine codons (TCX and AGY) of the Vtg I poly serine domain appear to be

Figure 3.5 Northern blot analysis comparing relative expression of F.
heteroclitus Vtg I and Vtg II mRNAs.
A) Methylene blue staining of duplicate samples transferred to
nylon membranes before hybridization, showing equal loading of
lanes, as indicated by 28s and 18s rRNA bands. Lanes a and a
contain 300 pg Vtg I RNA translated from plasmid cDNA
(pMMBl); lanes b and b contain 300 pg Vtg II RNA translated
from plasmid cDNA (pFhv2a); lanes c and c contain 15 fig total
liver RNA from an estrogen-treated male; lanes d and d contain
15 fig total liver RNA from a female four days before spawning;
lanes e and e contain 15 fig total liver RNA from a female 4 days
after spawning. RNA markers (kb) are indicated with arrows on
the left.
B) Autoradiographs of the membranes shown above, indicating
bands hybridizing to Vtg I (left side) and Vtg II (right side) DNA
probes. Note that the Vtg I probe did not hybridize to the Vtg II
control RNA (lane b) and Vtg II probe did not hybridize to the Vtg
I control RNA (lane a).

Pre-Hyb Pre-Hyb
Vtg I Vtg II
a b c d e a' b' c' d' e'
RNA
kb
7.46
4.4
2.37
1.35
28s
18s
0.24

67
separated into two general clusters, those of the Vtg II polyserine domain are
randomly interspersed. This arrangement of polyserine codons again confirms
the observations noted by Byrne et al., (1989) that the polyserine, or phosvitin,
domain is an independently evolving domain within the Vtg gene, showing more
variability than its flanking lipovitellin regions.
The predicted post-translational modifications of Vtg II are in agreement
with expectations for a lipophosphoglycoprotein. Although 45 phosphorylation
sites may appear to be high, we expect an even higher amount of phosphorylation
than is predicted. Seven of the 45 predicted phosphorylation sites occur within
the polyserine domain (all protein kinase C sites), however, from previously
published accounts, it is likely that every serine residue in this domain is
phosphorylated, resulting in a very hydrophilic domain with a highly negative
charge. Phosvitin yolk proteins have been described as possessing the highest
amount of phosphorylation of any known proteins. Unfortunately, the hepatic
Vtg kinase responsible for phosphorylating the extensive polyserine domains of
Vtg has not yet been isolated or characterized, so that an algorithm predicting its
target sites is not yet available.
Considering the ratio of expression of Vtg I and Vtg II, our data suggest
that Vtg I is the major yolk protein precursor. Vtg II mRNA is present in the
liver of spawning females at ratio of 1:10 with respect to Vtg I RNA, as
evidenced by northern blots. In SDS PAGE analysis, YP 69, which was mapped
to Vtg II, is hardly discemable (not shown here) when compared to the Vtg I-

68
derived yolk proteins, YP 125 and YP 105 (Chapter 4), agreeing well with the
mRNA expression data. This may suggest a difference in the interaction between
the estrogen-estrogen receptor complex with the estrogen response elements
(ERE) suspected to lie upstream of the Vtg I and Vtg II coding regions. Isolation
and characterization of the ERE from each Vtg gene should offer valuable
insights into ERE mechanics. Another explanation for the difference in amounts
of Vtg transcript may involve RNA stabilization, rather than gene transcription.
It has been shown that the half-life of Vtg transcripts increases dramatically in the
presence of estrogen (Brock and Shapiro, 1983). A recent report suggests that
an estrogen inducible protein that binds to the 3 untranslated region of Xenopus
Vtg may be responsible for this stabilization (Dodson et al.,1995). It would be
interesting to compare protein-RNA interactions of this protein with two closely
related Vtg mRNAs, F. heteroclitus Vtgs I and II. Furthermore, our estrogen-
induced liver library would offer an excellent template to screen for such cDNAs
that might code for this protein.
By completing the sequences of two Vtgs, we now have the basic
information and tools to molecularly dissect the process of yolk formation in F.
heteroclitus. We may begin to answer questions about the functional significance
of possessing multiple Vtgs. Antibodies produced against non-conserved regions
of the two Vtgs may indicate differences in receptor mediated endocytosis,
compartmentalization, or catabolism by the embryo. Cycling controls involving
Vtg may also be studied; for instance, Vtg cDNA probes can be used to document

69
the fine-tuned expression of Vtg that must occur in a sequentially spawning
animal. Besides being used as tools to specifically investigate F. heteroclitus
reproduction, the Vtg I and II cDNAs represent valuable bio-markers for assaying
the reproductive health of naturally occurring fish. As examples of mRNAs and
proteins that are normally induced by estrogens, Vtg I and II will be particularly
valuable in testing for the estrogenic effects of environmental contaminants such
as polychlorinated biphenyls (Bergeron et al. 1994; Guillette et al. 1994).

CHAPTER 4
PRECURSOR-PRODUCT RELATIONSHIP OF
VITELLOGENINS I AND II TO THE YOLK PROTEINS
OF FUNDULUS HETEROCLITUS
Introduction
Current views concerning the origin and processing of yolk proteins in oviparous
vertebrates were formed through a slow, and controversial suite of biochemical studies
that eventually elucidated two unexpected aspects concerning the origin of yolk proteins
(reviews by Wallace 1978, 1985; Eckelbarger 1994). First, it was shown that yolk
proteins originated "hetero-synthetically" in the liver, rather than the ovary. Secondly,
it was shown that yolk proteins were not synthesized individually, but rather as a large
protein precursor, that was subsequently processed into bona fide yolk proteins. This
yolk protein precursor, vitellogenin (Vtg), has now been documented to appear in the
blood of estrogen-treated males or spawning females from countless oviparous vertebrates
(Wallace and Jared, 1969). Additionally, Vtg has been documented to be incorporated
into growing oocytes by receptor-mediated endocytosis (Wallace and Jared, 1969b;
Opresko et al., 1980; Stifani et al., 1990; Shen et al., 1993; Shibata et al., 1993), and
processed into yolk proteins (Wallace and Jared, 1969). Although isotopic and
immunologic tracking studies have established the connection between yolk proteins and
Vtg, direct sequence data, mapping the precursor-product relationship of Vtg to the
70

71
derived yolk proteins, have been scarce (Clark, 1973; Bergink and Wallace, 1974; Byrne
et al., 1984; Gerber-Huber et al., 1987; Wallace et al., 1990b; Yamamura et al., 1995)
and especially lacking from teleosts (Matsubaro and Sawano, 1995). Recent studies
focusing on Vtg genes and cDNAs have documented that many animals possess multiple
Vtg genes and proteins (Wahli et al., 1979; Blumenthal et al., 1984; review by Byrne
et al., 1989) offering an even more challenging puzzle to workers seeking to map these
relationships.
Obtaining a clear synopsis of precursor-product relationships in many teleosts, is
further complicated by the extensive yolk protein processing that occurs in teleost yolk
as compared to the yolk of tetrapods. The most striking difference in yolk content
documented in F. heteroclitus concerns the disappearance of a 125-kDa yolk protein (YP
125), and the concomitant appearance of smaller yolk protein bands immediately prior
to oocyte ovulation (Wallace and Begovac, 1985; Wallace and Selman, 1985; Greeley
et al., 1986). This enhanced proteolytic processing may be connected to a unique pre
ovulatory process that occurs in some teleost oocytes, termed hydration. Near the time
of germinal vesicle breakdown, a rapid increase in oocyte volume occurs, usually
attributed to the uptake of water (Fulton 1898; reviewed in Selman and Wallace, 1989).
In F. heteroclitus, a substrate spawner, post-maturational oocytes possess twice the
volume of pre-maturational oocytes (Wallace and Selman, 1985, Greeley et al., 1991;
McPherson et al.), but in the oocytes of pelagic spawners, oocyte volumes can increase
over four times the original volume, in as little as twelve hours (Wallace and Selman,
1981; Watanabe and Kuo, 1986; Craik and Harvey, 1987; LaFleur and Thomas, 1991).

72
Several possible factors have been hypothesized to drive hydration, ranging from the
osmotic balance of ions (Hirose, 1976; Watanabe and Kuo, 1986; LaFleur and Thomas,
1991; Greeley et al., 1991; Wallace et al., 1992), ionic balance via gap junction control
(Cerd et al., 1993), and the colligative osmotic contribution of cleavage peptides and
free amino acids (Oshiro and Hibiya, 1981; Wallace and Selman, 1985; Greeley et al.,
1987; Thorsen et al. 1993). With these issues in mind we sought to characterize the
precursor-product relationship between Vtg and yolk proteins in F. heteroclitus, with
emphasis on the processing of YP 125. By completing the cDNA and putative protein
sequences of two F. heteroditus Vtgs (LaFleur et al., 1996; chapter 3), we obtained the
necessary blueprint for comparison of microsequencing data. In this paper we document
internal and N-terminal amino acid sequences from seven isolated yolk proteins, all of
which can be positioned within the Vtg I and Vtg II predicted protein sequences. Our
data suggest that the majority of yolk proteins are derived from Vtg I, and that a small
amount are derived from Vtg II. Additionally, we suggest that the rapid processing of
YP 125 during hydration is associated with the presence of a PEST site (Rogers et al.
1986) near its predicted C-terminus.
Materials and Methods
Ovarian follicles were dissected from the ovaries of reproductively active F.
heteroditus. Up to 20 prematurational follicles or up to 10 ovulated eggs were aliquoted
into a 1.5 ml eppendorf tube containing 500-750 ri of sample buffer (0.1 M Tris, pH
6.8, 2% SDS, 64 Mm dithiothreitol, 10% glycerol) on ice. The follicles were

73
immediately ground with a Kontes pestle and heated for 10 min at 100 C. The
homogenate was then briefly centrifuged at 12,000 g for 1 min., separating the dissolved
yolk from insoluble cellular debris. The supernatant was aliquoted to a fresh tube and
stored at -20C until electrophoresis. Samples were diluted again by as much as 1:50
with sample buffer before loading onto gels.
Sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) was
carried out according to Laemmli (1970), using 125 X 140 X 1.5 mm slab gels
containing a 3.5% stacking gel overlaying a separating gels ranging from 7% for larger
YPs to 12% for smaller YPs, with modifications based on the protocol of Schagger and
von Jagow (1987) using Tris-tricine running buffers.
Proteins in electrophorese gels were electroblotted onto PVDF membranes in
buffer containing 10 Mm MES, Ph 6, and 20% methanol at 20 V overnight. Protein
bands were visualized by brief staining in 0.02% Coomassie blue in 40% methanol plus
5% acetic acid, destained in 40% methanol plus 5% acetic acid, and rinsed in distilled
water. Membranes were dried and stored at -20 C until individual bands were cut out
and submitted for sequencing. N-terminal amino acid analyses were performed on PVDF
bound proteins using an Applied Biosystems Model 473a Sequencer (LeGendre and
Matsudaira, 1988) by the Protein Chemistry Core Facility of the University of Florida.
The two largest N-terminally blocked yolk proteins (YP 125 and YP 105) were
again electrophoresed, blotted onto PVDF and subjected to in situ cleavage (Scott et al.,
1988) by endoproteinase LysC (Endo LysC)(0.003 units//xg protein, Promega), in 50 mM
Tris, Ph 8.8, 0.2M ammonium bicarbonate, and 0.1% SDS, 0.1 Mm EDTA. Protein

74
Oocyte Egg
CO
3
O CO
o c
q3 B
o o
r- -
Q_
3 P
T3 ^
C
l
YP 29-
-t yp 83
-r YP 77
^ YP 69
-< YP 39
YP 20^
Figure 4.1 Major yolk proteins isolated from oocytes and eggs of Fundulus
heteroclitus. The major yolk proteins shown here were resolved by an
SDS-PAGE gradient gel (7%-20%) enabling the resolution of a wide
range of proteins ranging from 125 kDa to 20 kDa. For N-terminal
sequencing, however, straight gels were used at various acrylamide
concentrations allowing optimal resolution of yolk proteins at specific size
classes. Yolk proteins that were isolated for N-terminal sequencing are
indicated with our designated labels. Note YP 125 appears as a robust
band, when isolated from pre-maturational oocytes, but is hardly visible
in yolk isolated from ovulated eggs. (Photo courtesy of R. McPherson,
Clarion University)

75
Mw, kDa
21.5^
LFESLVDSDKW.
YEFSDELLQTPL.
KYxAKHIGVGLK.
13 kDa
LFESLVDSDCW. .
YEFSDELLQTPL...
Figure 4.2 Endo LysC digestion products of YP 125 and YP 105. After partial
digestion with Endo LysC, polypeptide fragments were electroblotted onto
a PVDF membrane and silver stained. Positions of the 13 kDa bands
(presumed to be identical) from each digestion are indicated. Molecular
weight standards (kDa) are shown on the left.

76
fragments were then separated by Tris-tricine gels, blotted to PVDF, and visualized by
silver staining (Wray et al., 1981). Similar bands of 13 kDa (presumed to be identical)
were isolated from both the YP 125, and YP 105 digestion and once again submitted to
the Protein Chemistry Core for N-terminal amino acid sequencing.
Sequencing data, including Vtg I and II cDNAs, along with microsequencing
results were organized using the PC/GENE software package (Intelligenetics, Mountain
View, CA). Prediction of signal peptides was carried out according to von Heijne
(1986). PEST sites were designated according to the algorithm described by Rogers et
al. (1986). Other Vtg sequences referred to in this paper include chicken Vtg II
(gi:63887; van het Schip et al. 1987), Xenopus laevis Vtg A2 (gi: 139636, Gerber-Huber
et al. 1987), lamprey, Ichthyomyzon unicuspus Vtg (gi:213312, Sharrock et. al. 1992)
and sturgeon, Acipenser transmontanus Vtg (gi:437051, Bidwell and Carlson, 1995).
Results
The yolk proteins typically found in F. heteroclitus oocytes and eggs are
demonstrated in Figure 4.1, along with our designations of certain bands according
to their apparent molecular mass. At least nine yolk proteins were resolved by Tris-
tricine SDS-PAGE, and these were blotted onto PVDF membranes and submitted for
protein sequencing by Edman degradation. Four yolk proteins appeared to be N-
terminally blocked, while five yielded N-terminal sequences (Table 4.1).
By aligning the yolk protein N-terminal sequences against the predicted amino
acid sequences of Vtg I and Vtg II, we successfully mapped the five sequenced yolk

Table 4.1. N-Terminal Sequences of F. heteroclitus Yolk Proteins
Protein
Source
N-Terminal Sequence
Source
n
Y?
125
Oocyte
Blocked
?
3
Y?
105
Oocyte/egg
Blocked
?
2
YP
83
Egg
Blocked
?
1
YP
30
Oocyte/egg
Blocked
?
1
YP
77
Egg
Blocked
7
1
YP
69
Egg
NQVSY APE PA PGXTY
SYXYE
Vtg II
1
Y?
45
Oocyte
HKKMV AxGxx A
Vtg I
2
YP
39
Egg
EEEAV VAVIL RAVKA
D
Vtg I
2
YP
29
Oocyte
AAAAE xSFVE DTLYT
PN
Vtg I
1
YP
20
Oocyte
EEDVE PIPEY KFRRF
AKKYC
Vtg I
2
ELC
13
YP 125
YEPSD ELLQT PLQLI
KISD
Vtg I
1
ELC
13
YP 125
LPESL VDSDK WENP
LLREV
Vtg I
1
ELC
13
YP 125
KYCAK HIGVG LXACP
KPASQ
Vtg I
1
ELC
13
YP 105
YEPSD ELLQT PLQLI
KISD
Vtg I
1
ELC
13
YP 105
LPESL VDSDK WENP
LLREV
Vtg I
1
* Mapped to Vtg I (982-1001), C-terminal to a PEST site
ELC denotes N-terminus of products cleaved with Endo Lys C (.003 units/ug protein)

78
protein products at internal positions within their respective precursors (Fig. 4.2). Of
these five sequences, the most notable was that of YP 69, lining up to the N-terminus of
Vtg II, verifying the expression of this secondary Vtg as well as demonstrating that the
signal peptide cleavage site had been correctly predicted. The data from YP 69 also
indicate that the N-terminus in Vtg II is unblocked in contrast to the apparently blocked
N-terminus of Vtg I.
In order to identify the origin of YP 105 and YP 125, the protein bands were
again blotted onto PVDF membranes and proteolytically cleaved with Endo Lys C
(0.003 units/jig protein, Promega) in 50 Mm Tris, Ph 8.8, 0.2 M ammonium
bicarbonate, and 0.1% SDS, 0.1 Mm EDTA. The digestion products were again
separated by Tris-tricine gels, and visualized by silver staining. The reaction with Endo
LysC was confirmed as only a partial digestion by the isolation of peptide products larger
than those predicted if cleavage had occurred at every lysine residue. The pattern of
electrophoresed digestion products from YP 125 and YP 105 initially appeared to be
identical, indicating that the two yolk proteins originated from the same precursor
molecule (Fig. 4.3). However, a difference between the digestion products was
discovered when the 13-Kda peptides derived from YP 125 and YP 105 were sequenced.
The 13-Kda band isolated from YP 105 digestion contained two peptides, mapping near
the N-terminal region of Vtg I. The 13-Kda band isolated from YP 125 contained the
exact two peptides found in the YP 105 digestion, plus a third peptide (K Y C A K H
IGVGLKACFKFAS Q), that mapped much further along the Vtg I sequence,
to residue 982 (Figs. 4.4 and 4.5). We interpret these data as evidence that YP 105 and

79
Vtgl
predicted 188 kD
signal
Pclyserine
domain
HKKMVAxG
YP 45
I
EEDVEPIPEYKF
YP 18
EEEAWAVILRA
YP 39
AAAAExSP/EDT
YP 29
Vtgll
predicted 185 kD
signal

NGVSYAPEFAPG
YP 69
Poiyserine
domain
Figure 4.3 A graphical representation of F. heteroclitus yolk proteins positioned
along the length of the Vtg I and Vtg n. Length and positions along the
Vtg molecules are drawn to scale according to alignments of N-termini
data with cDNA translations. C-termini of yolk proteins were calculated
according to molecular weight estimations and should be regarded as
putative. The signal peptides and polyserine domains as predicted from
cDNA translations are indicated.

80
Vtg I 1 88 kD
YP 125
ELC 13
ELC 13
ELC 13
T
200 aa
C-TERMINUS
Figure 4.4 A graphic representation of the 13 kDa digestion products and their
positions in reference to YP 125, YP 105 and Vtg I. Note that the third
digestion product of YP 125 lies beyond the calculated C-terminus of YP
105. The indicated PEST site was found in YP 125, but is truncated, and
thus invalidated in YP 105.

81
YP 125 are identical Vtg I-derived yolk proteins except for a short 20 Kda extension at
the C-terminus of YP 125 that contains the third Endo LysC digestion product (Fig. 4.3).
The C-terminus of YP 105 was predicted to lie at (or before) residue Ser 962 of
Vtg I, using the estimated mass of YP 105 and the masses of the individual residues
predicted from the Vtg I cDNA. This places the YP 105 C-terminus only 2 residues
away from the N-terminal residue obtained from YP 20 (Glu 965) suggesting that YP 105
and YP 20 result from cleavage of YP 125. The estimated juncture between YP 105 and
YP 20 lies at the exact midpoint of a predicted PEST site (residues 952-974, receiving
a score of 6.9, where 5.0 and above is considered a site). This purported cleavage site
bisects the predicted PEST site, leaving the two resulting protein sequences with termini
that do not surpass the cutoff value for valid PEST sites. Thus, although YP 125
contains a PEST site, neither of its cleavage products, YP 105, nor YP 20 do.
Discussion
We have presented precursor-product relationships to account for the origin of
seven yolk proteins isolated from oocytes and eggs of F. heteroclitus. Likewise, the
sequences determined from these yolk proteins verify the expression, transport, and
incorporation of both the yolk protein precursors Vtg I and Vtg II, whose cDNA
sequences are provided in Chapters 2 and 3, respectively.
We had initially assumed that YP 125 and YP 105, the major bands in oocyte
extracts, were derived separately from Vtg I and II, but the internal sequences indicated
that both yolk proteins originate from Vtg I. We can thus surmise that Vtg I is truly the

Figure 4.5 A summary of the precursor-product relationship of Vtg I to
derived yolk proteins. The entire translated amino acid sequence
of the Vtg I cDNA sequence (LaFleur et al., 1995) is presented,
separated into sections representing yolk proteins as indicated by
brackets on the right. N-terminal sequences of isolated yolk
proteins are indicated by double underlining. Internal sequences
obtained from Endo LysC digestion products are indicated by
shaded lettering. The residues of the PEST site are represented by
bold face lettering. The predicted polyserine domain (no N-
terminal sequencing data) is shown in brackets.

1
M
K
A
V
V
L
A
L
T
L
A
F
V
A
G
<2
N
F
A
P
E
F
A
A
G
K
T
Y
V
Y
31
K
Y
E
A
L
I
L
G
G
L
P
£
E
G
L
A
R
A
G
L
K
i
S
T
K
L
L
L
s
A
61
A
D
Q
N
T
Y
M
L
K
L
V
E
P
E
L
S
E
Y
S
G
I
W
P
K
D
P
A
V
p
A
91
T
K
L
T
A
A
L
H
L
S
S
Q
F
P
S
S
L
N
T
P
M
V
F
V
G
K
V
F
A
P
121
E
E
V
S
T
L
V
L
N
I
Y
R
G
I
L
N
i
L
Q
L
N
I
K
K
T
H
K
V
Y
D
151
L
Q
E
V
G
T
Q
G
V
C
K
T
L
Y
S
I
S
E
D
A
R
I
E
N
X
L
L
T
K
T
181
R
D
L
s
N
C
Q
E
R
L
N
K
D
I
G
L
A
Y
T
E
K
c
D
K
c
Q
E
E
T
K
211
N
L
R
G
T
T
T
L
S
Y
V
L
K
P
V
A
D
A
V
M
I
L
K
A
Y
V
N
E
L
i
241
Q
F
S
P
F
S
E
A
N
G
A
A
Q
M
R
T
K
<2
S
L
E
F
L
E
I
E
K
E
P
I
271
P
S
V
K
A
E
Y
R
H
R
G
S
L
K
V E F 3
D
E
it,It
Q
T P.
. 1. Q. ,vK<
301
1
3
D-
A
P
A
Q
V
A
E
V
L
K
H
L
A
T
Y
N
I
E
D
V
H
E
N
A
P
L
K
331
F
L
E
L
V
Q
L
L
R
I
A
R
Y
E
D
L
E
M
Y
w
N
Q
Y
K
K
M
S
P
H
R
361
H
W
F
L
D
T
I
P
A
T
G
T
F
A
G
L
R
F
I
K
E
K
F
M
A
E
E
I
T
I
391
A
E
A
A
<2
A
F
I
T
A
V
H
M
V
T
A
D
P
E
V
I
K
i>
F E
s
X. V D S.
421
D
K
mi

e
N
Si
m.
X.
R
E
V
V
F
h
,s:
m
G
T
M
V
N
K
Y
C
N
K
T
V
D
451
c
P
V
E
L
I
K
s
I
Q
Q
R
L
s
D
A
I
A
K
N
E
E
E
N
i
I
L
Y
i
K
481
V
L
G
N
A
G
H
P
s
s
F
K
S
L
T
K
i
M
P
I
H
G
T
A
A
V
s
L
P
M
511
T
I
H
V
E
A
I
M
A
L
R
N
I
A
K
K
E
S
R
M
V
Q
E
L
A
L
Q
L
Y
M
541
D
K
A
L
H
P
E
L
R
M
L
S
C
I
V
L
F
E
T
S
P
s
M
G
L
V
T
T
V
A
571
N
S
V
K
T
E
E
N
L
Q
V
A
s
F
T
Y
S
H
M
K
s
L
S
R
S
p
A
T
I
H
601
P
D
V
A
A
A
C
S
A
A
M
K
I
L
G
T
K
L
D
R
L
S
L
R
Y
s
K
A
V
H
631
V
D
L
Y
N
S
S
L
A
V
G
A
A
A
T
A
F
Y
I
N
D
A
A
T
F
M
P
K
s
F
661
V
A
K
T
K
G
F
I
A
G
S
T
A
E
V
L
E
I
G
A
N
I
E
G
L
Q
E
L
I
L
691
K
N
P
A
L
S
E
S
T
D
R
I
T
K
M
K
R
V
I
K
A
L
S
E
W
R
S
L
p
T
721
S
K
P
L
A
s
V
Y
V
K
F
F
G
Q
E
I
G
F
A
N
I
D
K
P
M
I
D
K
A
V
751
K
F
G
K
E
L
P
I
Q
E
Y
G
R
E
A
L
K
A
L
L
L
S
G
I
N
F
H
Y
A
K
781
P
V
L
A
A
E
M
R
R
I
L
P
T
V
A
G
I
P
M
E
L
S
L
Y
S
A
A
V
A
A
811
A
S
V
E
I
K
P
N
T
S
P
R
L
S
A
D
F
D
V
K
T
L
L
E
T
D
V
E
L
K
841
A
E
I
R
P
M
V
A
M
D
T
Y
A
V
M
G
L
N
T
D
I
F
Q
A
A
L
V
A
R
A
871
K
L
H
S
V
V
P
A
K
I
A
A
R
L
N
I
K
E
G
D
F
K
L
E
A
L
P
V
D
V
901
P
E
N
I
T
s
M
N
V
T
T
F
A
V
A
R
N
I
E
E
P
L
V
E
R
I
T
P
L
L
931
P
T
K
V
L
V
P
I
P
i
R
R
H
T
S
K
L
D
P
T
R
952
N
3
M
L
D
3
3
E
L
L
P
M
E
964
E
E
D
V
2
p
I
P
2
Y
K
F
R
R
F
A
K
K Y C A
K
H
i
0
V
991
G
L
K
A
C
F
K F
A
3
2
N
G
A
S
i
<2
D
i
V
L
Y
K
L
A
G
S
H
N
F
1021
S
P
s
V
T
p
I
E
G
E
V
V
E
R
L
E
M
E
V
K
V
G
A
K
A
A
E
K
L
V
1051
K
R
I
N
L
s
E
D
E
S
T
E
E
G
G
P
V
L
V
K
L
N
K
I
L
S
S
R
R
N
1081
S
S
s
S
S
s
S
S
S
S
S
S
S
E
S
R
S
S
R
S
S
S
S
S
S
S
s
S
R
S
1111
S
R
K
I
D
L
A
A
R
T
N
S
s
S
S
S
s
S
R
R
S
R
S
S
S
S
s
S
S
S
1141
S
S
S
S
S
s
S
S
S
S
S
s
R
R
S
S
s
s
S
S
S
S
s
s
s
S
s
R
S
s
1171
R
R
V
N
S
T
R
S
S
S
S
s
S
R
T
S
s
A
s
S
I.
A
s
F
F
s
D
S
S
s
1201
S
S
s
S
s
D
R
R
S
K
E
V
M
E
K
F
<2
R
L
1220
H
K
K
M
V
A
S
G
S
S
A
s
S
V
E
A
I
Y
K
E
K
K
Y
L
G
E
1246
E_
E
A
V
V
A
V
I
L
R
A
7
K
A
D
1261
K
R
M
V
G
Y
2
L
G
F
Y
L
0
K
P
N
A
R
V
Q
I
I
V
A
N
I
s
s
D
S
1291
N
W
R
I
C
A
D
A
V
V
L
S
K
H
K
V
T
T
K
I
S
W
G
E
Q
C
R
K
Y
S
1321
T
N
V
T
G
E
T
G
X
V
S
S
S
P
A
A
R
L
R
V
S
W
E
R
L
P
s
T
L
K
1351
R
Y
G
K
M
V
N
K
Y
V
P
V
K
I
L
S
D
L
I
H
T
K
R
E
N
s
T
R
N
I
1381
S
V
I
A
V
A
T
S
E
K
T
I
D
I
I
T
K
T
P
M
S
S
V
Y
N
V
T
M
H
L
1411
P
M
C
I
P
I
D
E
I
K
G
L
S
P
F
D
E
V
I
D
K
I
H
F
M
V
S
K
1439
A
A
A
A
E
C
s
F
V
E
D
T
L
Y
T
F
N
N
R
s
Y
K
N
K
M
P
S
S
c
Y
1469
<2
V
A
A
<2
D
c
T
D
E
L
K
F
M
V
L L
R
K
D
S
S
E
Q
H
H
I
N
V
K
1599
I
S
E
I
D
I
D
M
F
P
K
D
0
N
V
T
V
K
V
N
E
M
E
I
P
P
P
A
c
L
1529
T
A
T
Q
Q
L
P
L
K
I
K
T
K
R
R
G
L
A
V
Y
A
P
S
H
G
L
Q
E
V
Y
1559
F
D
R
K
T
W
R
I
K
V
A
D
W
M
K
G
K
T
c
G
L
C
G
K
A
D
G
E
I
R
1589
Q
E
Y
H
T
P
N
G
R
V
A
K
N
S
I
S
F
A
H
S
W
I
L
P
A
E
S
C
R
D
1619
A
S
E
C
R
L
K
L
E
s
V
2
L
E
K
Q
L
T
I
H
G
E
D
S
T
C
F
S
V
E
1649
P
V
P
R
C
L
P
G
C
L
P
V
K
T
T
P
V
T
V
G
F
S
C
L
A
S
D
P
Q
T
1679
S
V
Y
D
R
S
V
D
L
R
Q
T
T
Q
A
H
L
A
c
S
C
N
T
K
C
3
YP 105
YP 125
} polyserine
domain
PEST
siie
YP 45
YP 39
YP 29

84
major yolk protein precursor in F. heteroclitus. Our finding that most of the yolk protein
is derived from Vtg I agrees with northern blot analyses that suggested ten times more
Vtg I than Vtg II message is present in total liver RNA (Chapter 3; LaFleur et al.,
1996).
A major factor that prevents construction of a definitive map accounting for all
yolk proteins derived from the Vtgs is the difficulty in isolating and microsequencing
peptides derived from the phosvitin domain (Wallace and Begovac, 1986; Wallace et al.
1990). Although we expect that the polyserine repeats represented in both Vtg I and II
cDNAs are processed into true phosvitins, we have been unable to verify this by N-
terminal sequencing. The highly negative charge of phosvitin prevents it from staining
with Coomassie blue, as well as adhering to PVDF membranes for sequencing. Because
phosvitin can be visualized using Stains-all, it has been documented as a single 25-30
kDa band in prematurational oocytes, with at least four smaller phosvitin-like bands
(phosvettes) appearing in preparations from ovulated eggs (Wallace and Begovac, 1985).
We estimate that the C-terminus of YP 20 (and presumably, the C-terminus of YP 125)
lies adjacent to the N-terminus of phosvitin, as predicted by the position of the Vtg I
cDNA polyserine repeating region. Likewise, the sequence obtained from YP 45,
sharing identity with residues 1220-1230 of Vtg I, most likely abuts the C-terminal
cleavage site of phosvitin.
As previously mentioned, one of the most pronounced changes observed to occur
in F. heteroclitus yolk proteins is the disappearance of YP 125 during the transformation
of oocytes to mature, ovulated eggs (Fig. 4.1). A possible explanation for this rapid and

85
rather selective proteolysis is the occurrence of a PEST site within the C-terminal tail of
YP 125. The apparently longer-lived YP 105 is identical to YP 125 except for lacking
the C-terminal tail where the PEST site occurs. PEST sites were initially defined as a
conserved clustering of amino acids that was observed to occur in proteins known to be
rapidly degraded. Common to all PEST site are high local concentrations of Pro, Glu,
Ser, and Thr, and to a lesser extent Asp. Of the other five vertebrate Vtg sequences
contained in Genbank, chicken Vtg II (residues 1058-1080 and 931-951) and lamprey Vtg
(residues 1161-1182 and 1360-1393) contain two PEST sites, while Xenopus Vtg A2
contains a sequence (residues 953-969) with a score (4.71) very close to the cutoff value
of 5. The lack of proteolysis during oocyte maturation in such animals may indicate
either the absence of an appropriate proteolytic mechanism or the inaccessibility of the
cleavage sites in the granular yolk of these animals (Wallace, 1985). The Vtg of
sturgeon, a chondrostean fish, does not contain a PEST site.
The proteolytic processing of YP 125 has been implicated as part of the hydration
mechanism of F. heteroclitus oocytes, with the generated small peptides and free amino
acids providing the osmotic potential to drive an uptake of water into the oocyte (Wallace
and Begovac, 1985; Wallace and Selman, 1985). More recent data suggest that
hydration in F. heteroclitus is primarily due to K+ fluxes via the gap junctions between
oocytes and follicle cells (Wallace et al., 1992; Cerd et al., 1993), but the possibility
of some contributions to hydration resulting from yolk cleavage has not yet been
abandoned. So far, complete Vtg sequences have been reported from no other teleosts
besides F. heteroclitus. However, as more sequences are completed, it will be

86
interesting to see whether PEST sites are found in other teleostean Vtgs, especially those
of pelagic spawners in which both oocyte hydration and yolk proteolysis are especially
pronounced.

CHAPTER 5
FUNDULUS HETEROCLITUS CHORIOGENINS: LIVER-DERIVED
COMPONENTS OF THE VITELLINE ENVELOPE AND CHORION
SHARING SEQUENCE IDENTITY WITH MAMMALIAN ZP PROTEINS
Introduction
The spawned eggs of the estuarine teleost Fundulus heteroclitus are exposed to
quite a different environment than the ovulated eggs of mammals. Whereas mammalian
eggs are protected from infection, desiccation, and predation by the safe surroundings
of the uterus, F. heteroclitus eggs are released and fertilized during the tumultuous spring
tides, and deposited into empty mussel shells or onto the leaves of marsh grass, where
they remain actually stranded above the water line for fourteen days until the embryos
emerge by hatching during the next spring tide (Taylor et al., 1977; Hsiao et al., 1994).
Though exposed to extremely different environments, both of these vertebrate eggs are
protected by a quasi-similar layer of extracellular matrix (ECM). In mammals this
translucent layer of ECM is termed the zona pellucida (ZP), but in fish and many other
invertebrates it is often referred to as the vitelline envelope or chorion.
In this paper we adhere to the definitions of Dumont and Brummett (1980)
regarding the vitelline envelope and chorion. They stated that the term "vitelline
envelope" referred to the highly structured acellular layer that appears and encloses the
87

88
teleost oocyte during its development, while the term "chorion" referred to the
structurally and perhaps chemically transformed vitelline envelope that surrounds the
ovulated egg, separates from the egg at the time of fertilization, and encloses the embryo
until hatching. Implicit in these definitions is the assumption that the proteinaceous
structure of the vitelline envelope comprises a substantial component of the chorion.
The structure of the teleostean vitelline envelope has been well documented in
several cyprinodont species (Yamamoto, 1963; Fliigel, 1967; Dumont and Brummett,
1980) as well as in many other teleosts (reviewed by Dumont and Brummett, 1985;
Selman and Wallace, 1989). Early biochemical characterizations of the vitelline envelope
and chorion concentrated on the formation of the vitelline envelope during oocyte
development (Chaudry, 1956; Yamamoto, 1963; Flegler, 1977; Tesoriero, 1977), as well
as the breakdown of the chorion by the proteolytic enzymes of the hatching embryo
(Yamamoto and Yamagami, 1975; Kaighn, 1964, Hagenmaier, 1985). In earlier works
it had been assumed, but not proven that the major vitelline envelope proteins (VEPs)
were synthesized by the ovarian follicle the site of synthesis residing in either the
oocyte or surrounding follicle cells (Anderson, 1967). More recent investigations
targeting VEP synthesis include studies by Tesoriero (1978) using [3H]proline
incorporation, and by Begovac and Wallace (1989) in which incorporation of
[35S]methionine combined with immunohistochemistry provided evidence that at least one
of the VEPs from the pipefish, Syngnathus scovelli, originated from within the ovarian
follicle.
A new direction towards understanding vitelline envelope formation was launched

89
by research concentrating on the chemistry of hatching enzymes in the medaka, Oryzias
latipes. When polyclonal antibodies directed against protein fragments of the lysed
chorion were used as probes on medaka tissues, Hamazaki et al. (1984) found that tissues
other than the ovary were recognized by the antibody. By 1989 Hamazaki et al. (1989a)
had isolated an estrogen-induced glycoprotein from the liver that could be localized to
the inner layer of the vitelline envelope. Since then additional reports have verified these
findings in several other fish (Hyllner et al., 1991; Murata et al., 1991; Oppen-Bemtsen
et al., 1992a, 1992b; Larsson et al., 1994). Additionally, Hyllner et al. (1991) showed
that the synthesis of VEPs could be induced by estrogen treatment in males of the
rainbow trout (Oncorhynchus mykiss), brown trout (Salmo trutta), and turbot
(Scophthalmus maximus), providing convincing evidence that the major VEPs in these
species could be synthesized without any contribution by the ovary.
So far, only two nucleotide and protein sequences representing piscine VEPs have
been published. Lyons et al. (1993) reported a gene sequence (wf) from the flounder,
Pseudopleuronectes americanus, that they described as a "teleostean homolog of a
mammalian ZP gene." The predicted amino acid sequence contained a novel PQQ
repeating region near the N-terminus, resembling a motif found in extracellular matrix
proteins. Murata et al. (1995) reported a cDNA sequence (L-SF) from medaka that also
shared identity with mammalian ZP proteins. The predicted amino acid sequence of the
medaka L-SF protein shared more identity with mouse ZP3 (37.9%; Ringuette et al.
1988) than it did with the flounder ZP sequence (18%) previously mentioned, suggesting
the presence of at least two distinct groups of teleost ZP homologs.

90
In this paper we present the predicted primary structure of three proteins that
share identity with mammalian ZP proteins. Furthermore, we have isolated cDNAs
encoding these sequences from a liver library rather than an ovarian library, followed by
northern analyses revealing liver rather than ovarian transcripts. Lastly, the amino acid
compositions predicted from our cDNAs are similar to the composition of VEPs isolated
from F. heteroclitus follicles. Therefore, we conclude that the proteins encoded by these
cDNAs are synthesized by the liver, transported to the ovary, and incorporated into the
vitelline envelope. We further suggest that as major constituents of the vitelline
envelope, these proteins eventually contribute to the structure of the hardened chorion,
where they remain until finally degraded by embryonic hatching enzymes. We designate
these cDNAs and the proteins that they encode as "choriogenins" (Chgs) to emphasize
their role as proteins of the vitelline envelope and chorion, yet to underscore their site
of synthesis as being extra-ovarian, and thus different from that of the mammalian ZP
proteins. Although the teleostean chorion and mammalian zona pellucida have different
appearances, functions, and, as this study verifies, origins of synthesis, we provide
evidence that the constituent molecules appear to have evolved from a set of common
ancestral proteins
Materials and Methods
Reagents
Estradiol-17/3 was obtained from Sigma Chemical Co. (St. Louis, MO).
Radioisotopes, [a-32P]dCTP and [a-35S]dATP, were purchased from New England

91
Nuclear (Boston, MA). Lambda gtlO vector and cDNA synthesis reagents were obtained
from Promega (Madison, WI). The subcloning plasmid pGem-T was a product of
Promega. All "ROW" oligonucleotide primers were synthesized by the University of
Florida Interdisciplinary Center for Biotechnology Research (ICBR) oligonucleotide core
facility, while primers labelled "GL" were synthesized by Bio-Synthesis (Freindswood,
TX). Sequenase version 2.0 DNA polymerase and dideoxy sequencing reagents were
obtained from US Biochemicals (Cleveland, OH). In-house sequencing gels were cast
using Sequagel-8 (National Diagnostics, Atlanta) polyacrylamide reagents. Some cDNA
sequences, especially through repeating regions or when verifications were needed, were
performed by The University of Florida ICBR DNA Sequencing Core. Amplification
reactions were performed using a 1:50 mixture of cloned pfu DNA polymerase and
Thermophilus aquaticus DNA polymerase (Stratagene and Promega, respectively).
Reagents for random-primed labeling of probes were purchased from Pharmacia
(Piscataway, NJ). Magna nylon and PVDF transfer membranes were obtained from MSI
(Westboro, MA) and Millipore Corp. (Bedford, MA), respectively.
Cloning Strategy
A liver cDNA library was constructed from poly A+-RNA pooled from five F.
heteroclitus males that had been treated with two injections of estradiol-17)8, as
previously described (LaFleur et al., 1995). While screening the Xgt 10 library for Vtg
cDNAs using anchored PCR, we isolated several non-target cDNAs. Three of these non-
Vtg cDNAs were revealed by BLAST analysis to code for protein sequences that

92
Chq 500
pChgla
ROW 45
ROW 52
pChglb
Chq 427
PChg2a RW55
ROW 65
pChg2b
Chq 553
pChg3a
ROW 45
200 bp
GL1
pChg3b
Figure 5.1 Strategy for cloning Chg 500, 427 and 553 cDNAs. Boxes indicate
relative sizes of contiguous cDNA sequences coding for Chgs 500, 427,
and 553. Thin black lines represent individual cDNA isolates obtained by
anchored PCR or RACE and inserted into pGem-T. Arrows indicate
gene-specific primers that were used in initial amplifications of individual
clones. The legend indicates relative length of 200 bp.

93
resembled mammalian ZP proteins. The clones containing these initial cDNAs were used
as probes and to design primers that would target additional cDNAs in order to complete
the Chg coding regions (Fig. 5.1).
The first Chg cDNA, pChgla, was isolated by anchored PCR with a Vtg II-
targeted reverse primer, ROW 45, and the XgtlO vector primer, NEB 1231. To retrieve
a cDNA containing the poly-A tail, primer ROW 52 was designed from the
3 side of pChgla. An overlapping clone (pChglb), containing a poly-A tail, was
isolated by anchored PCR, completing the translated region of Chg 500.
The second choriogenin cDNA, pChg2a, was isolated by anchored PCR with
anchor primer NEB 1232 and reverse primer ROW 55, also designed to target Vtg II
sequence. Blast analysis on the sequence of pChg2a revealed that it shared 67 % identity
with the medaka L-SF protein (Murata et al. 1995). The primer ROW 65 was then
designed from the 3 region of pChg2a to retrieve an overlapping cDNA that contained
a poly-A tail. The sequence from the resulting clone (pChg2b) completed the translated
portion of the second Chg 427.
The third Chg clone, (pChg3a), was also isolated with ROW 45, along with the
initial pChgla, but it remained unrecognized as a novel clone until further review of the
sequences. An overlapping clone (pChg3b) containing the poly-A tail was isolated by
anchored PCR with the forward primer GL2. A third clone containing a short segment
5 to pChg3a including the initial methionine codon was isolated using a rapid
amplification of cDNA ends protocol (RACE) (Frohman, 1992) with reverse primer
GL1, and 3 fig total liver RNA as template.

94
Sequence Analyses
Nucleotide sequencing data was organized and assembled using the sequence
analysis software package PC/GENE (Intelligenetics, Mountain View, CA). A search
for post-translational modifications and signature sequences was done with the Prosite
program (Bairoch et al., 1995) available from the world wide web ExPaSy molecular
biology server (http://expasy.hcuge.ch/www/expasy.top.html). Protein alignments were
performed with the ClustalV program (Higgins et al. 1992), utilizing a Pam 250 matrix
with fixed gap and floating gap penalties = 10. In order to compare Chg sequences with
a large number of ZP Genbank entries a preliminary ClustalV alignment containing
complete sequences was performed. Whereas the N- and C-termini from Chgs differ
greatly with those of mammalian ZP proteins, a core region of conserved sequence was
observed where all three Chgs, as well as all other reported ZPs could be aligned with
a minimum number of gaps when anchored to five strictly conserved cysteines. This
region has previously been defined by Bork and Sander (1992) as the "ZP domain" and
is included in the Prosite program (Bairoch et al., 1995) available on the ExPaSy
molecular biology server. For parsimonious tree analysis, a new ClustalV alignment was
performed including only the ZP domains from each entry, providing a well conserved
region on which to base our distance analysis. Parsimonious tree analysis was done by
importing a ClustalV alignment in phylip 3.4 format into the PAUP 3.1 program
(Swofford, 1993) available from the Center for Biodiversity (Champagne IL). The
unrooted tree presented in Figure 5.8. resulted from running 100 bootstrap replicates of
a heuristic search. All entries used in alignment and tree analysis were retrieved from

95
the Entrez document retrieval system, Release 20.0, available from NCBI (NIH,
Bethesda, MD). Sequences referred to in this paper include the flounder
Pseudopleuronectes americanus ZP or wf (gi: 425355; Lyons et al., 1993); medaka, O.
latipes L-SF (gi: 563774; Murata et al., 1995); goldfish, Carrasius auratus ZP3
(gi:763073; unpublished); carp, Cyprinus carpi ZP3i (gi:763078; unpublished) and
ZP3ii (gi:763080; unpublished); mouse, Mus musculus ZP1 (gi: 972946; Epifano et al.,
1995), ZP2 (gi: 202460; Liang et al., 1990), and ZP3 (gi: 141726; Ringuette et al.,
1988); human ZP3A (gi: 141724; Chamberlin and Dean, 1990), ZPB (gi: 458279; Harris
et al., 1994), and ZP2 (gi: 466206; Liang and Dean, 1993); cat, Felis catus ZPA
(gi:458269), ZPB (gi:458271), and ZPC (gi: 458273; Harris et al., 1994).
Northern Blot Analyses
Male and female F. heteroclitus were collected from the estuarine creeks adjacent
to the Whitney Laboratory, and were maintained in running seawater tanks under
14L:10D photoperiod conditions at 25 + 2C. After approximately two weeks in
captivity, fish began spawning in laboratory tanks on a 14-day cycle (Hsiao et al., 1994).
By monitoring amounts of eggs spawned each day, we were able to calculate the 14-day
cycle of separate tanks and thus predict what phase of the spawning cycle individual fish
were in before sacrifice (Hsiao et al., 1996). In this paper two northern blots were
performed using a female fish that was predicted to be in a pre-maturational phase, four
days prior to spawning. Fish were maintained for at least one month before being used
for RNA collections.

96
Experimental groups of fish were subjected to two intraperitoneal injections of
estradiol-17/3 (0.01 mg/g body weight) dissolved in 50 /ri coconut oil (Kanungo et al.,
1990). Control groups were sham-injected with coconut oil alone. The first
injection was performed on day 1, the second injection on day 4, followed by sacrifice
and liver dissection on day 8.
Total RNA was isolated from livers and ovaries by extraction with RNA Stat-60
reagents (Tel-Test "B", Inc. Friendswood, TX). Tissues were dissected and immediately
frozen in 1.5-ml tubes containing 500 /I of RNA Stat-60 emulsion, by immersion in
liquid nitrogen. Tissues were homogenized at 20C using a Kontes pestle and motor.
Typically, a 300 mg liver yielded 0.35 mg total RNA, with O.D. 260/280 ratios
consistently above 1.8. Total RNA samples were resuspended in DEP-C-treated water
and stored at -80C until used in analyses.
Before electrophoresis, aliquots of 15 /xg total RNA were precipitated in
isopropanol and denatured in 2.2 M formaldehyde, 50% formamide, 50 mM MOPS (pH
7.0) for 30 min at 65C. Samples were electrophoresed through gels containing 2.0%
agarose, 0.6 M formaldehyde, 50 mM MOPS, and 1 mM EDTA for 1.5 hours at 3.5
V/cm gel in 50 mM MOPS, 1 mM EDTA running buffer. RNA was blotted onto Magna
nylon membranes by capillary action with 20 X SSC, immobilized by U.V. crosslinking
and visualized by staining briefly with methylene blue. All hybridizations were carried
out at 65C in 1 X Denhardts solution, 6 X SSC, and 0.1% SDS without formamide.

97
Isolation and Partial Characterization of the Major VEPs
VEPs were isolated following the protocols of Oppen-Bemsten et al. (1990) and
Hyllner et al. (1991) with slight modifications. Ovarian follicles were dissected
from the ovary of a reproductively active F. heteroclitus. Up to 30 individual unovulated
follicles were placed in a 1.5 ml Eppendorf tube containing an ice-cold solution of 0.1
M EDTA and 0.5 M NaCl. The follicles were gently ground with a Kontes pestle. The
intact vitelline envelopes were collected by a low speed spin (150 g) for 5 min, and the
supernatant containing mainly yolk was discarded. The insoluble vitelline envelope
material was washed over 24 hours with at least five changes of ice-cold 0.5 M NaCl
followed by five changes of Milli Q water, each time collecting the material by
centrifugation at 150 g. VEPs were solubilized in a Tris-buffered extraction buffer (0.1
M Tris-HCl, pH 8.8; 2% SDS; 0.3 M 2-mercaptoethanol; 0.1 M EGTA by heating to
70C for 30 min.
For electrophoresis of VEPs, samples were diluted at least 1:4 in sample buffer
(0.06 M Tris-HCl, pH 6.8; 2% SDS; 0.3 M 2-mercaptoethanol; 10% glycerol; without
bromophenol blue) and heated to 95C for 5 min. Sodium dodecyl sulphate-
polyacrylamide gel electrophoresis (SDS-PAGE) was carried out according to Laemmli
(1970) using 125 mm X 110 mm X 1.5 mm slab gels containing a 3.5% stacking gel
overlaying a 10% w/v separating gel, with modifications based on the protocol of
Schgger and von Jagow (1987), using Tris-tricine running buffers.
Initial attempts to transfer VEPs onto membranes using buffers containing 0.01
M morpholinoethane sulphonic acid (MES) and 20% methanol failed, probably due to

98
the insolubility of the VEPs. Successful transfer of the VEPs was accomplished in 0.01
M MES, 10% methanol, and .01% SDS. After transfer, the PVDF membrane was
stained with 0.02% Coomassie blue in 40% methanol and 5% acetic acid to indicate
protein bands. Duplicate membrane transfers containing VEP 69, 60, and 46 were
submitted for amino acid composition analysis and N-terminal amino acid analysis using
an Applied Biosystems Model 473a Sequencer (LeGendre and Matsudaira, 1988) at the
Protein Chemistry Core Facility of the University of Florida Interdisciplinary Center for
Biotechnology Research.
The N-terminal sequences initially obtained from the three bands consisted of
overlapping and weak signals that were only five residues long (data not shown);
therefore the VEP 69, 60, and 46 were isolated again and subjected to in-gel digestion
with endoproteinase Lys C (0.003 units//xg protein, Promega), in 10 mM Tris, pH 8.8,
0.2M ammonium bicarbonate, and 0.1 % SDS, 0.1 mM EDTA. Protein fragments were
separated by Tris-tricine gels, blotted onto PVDF and visualized by silver staining. Two
of the best resolved bands from each digestion were once again submitted to the Protein
Chemistry Core for N-terminal amino acid sequencing.
Results
Choriogenin cDNA Sequences
The nucleotide and translated amino acid sequences from three estrogen-induced
liver cDNAs are presented in Figure 5.2. We have designated the cDNAs and predicted

99
protein products as Chg 500, Chg 427 and Chg 553, according to the number of residues
in the predicted amino acid sequence.
The cDNA encoding Chg 500 is 1641 bp long, including a 1500 bp open reading
frame (Fig. 5.2a). The calculated molecular weight after subtracting the weight of a
predicted signal peptide (residue 1-22) is 53,125. The most notable region of the
predicted primary structure is a proline-rich repeating domain near the N-terminus,
including five repeats of (PQQ PQQ PQY PSK). Other proteins that share sequence
identity with Chg 500 include the flounder ZP gene product (58%; Lyons et al., 1993),
which also contains a proline-rich repeating domain (Figs. 5.3a and 5.3b), and several
mammalian ZP proteins, including mouse ZP1 (32%; Epifano et al., 1995), cat ZPB
(35%; Harris et al., 1994), and human ZPB (34%; Harris et al., 1994).
The Chg 427 is encoded by a cDNA of 1751 bp (Fig. 5.2b). Subtracting the
weight of the predicted signal peptide (residues 1-24) leaves a calculated molecular
weight of 44,892. This protein sequence does not include a substantial repeating domain,
although residues 28-46 (PGK PSK PQS PPT QNQ QQL Q) are reminiscent of the
proline-rich repeat previously described for Chg 500. The predicted N-terminus
possesses three in-frame methionine codons, but the first codon agrees best with the
context and positional environment for initiation of translation as described by Kozak
(1991). Alignment analyses revealed that Chg 427 shares highest identity (67%) with a
medaka female-specific protein termed "L-SF" (Murata et al., 1995) (Fig. 5.4). The
next highest identity comes from sequences recently deposited for two cyprinid fishes
(42% from C. auratus ZP3, 43% from C. Carpi ZP3i, and 44% with C. carpi ZP3ii).

Figure 5.2 Nucleotide and conceptually translated amino acid sequences of F.
heteroclitus choriogenins.
A) The Chg 500 cDNA (1641 bp) codes for a 500 amino acid
protein sequence, containing a predicted signal peptide (von
Heijne, 1986) from residues 1-22, indicated by shading. The poly-
adenylation signal is indicated by underlining (beginning at
nucleotide 1607).
B) The Chg 427 cDNA (1672 bp) codes for a 427 amino acid
protein sequence, containing a predicted signal peptide from
residues 1-22 indicated by shading. A poly-adenylation signal is
represented by underlining (beginning at nucleotide 1637).
C)The Chg 553 cDNA (1816 bp) codes for a 553 amino acid
protein sequence, containing a predicted signal peptide from
residues 1-25 indicated by shading. A poly-adenylation signal is
represented by underlining (beginning at nucleotide 1767).

actaactagaccagacagcttcgaggt 27
101
A)
ATGGCAAGTCACTGGAGTGTCACCCGTTGGGCCGCGCTGGCTCTGCTATGCTGCTTAGCTGGGAAA 93
MASHWSVTRWAALALLCCLAG K22
GGAGCAGAGGCTCAGAAGGGTTCGTATCCTCCGCAACCTCAAAAGCCTTCGTACCCTCAGAATCCT 159
GAEAQKGSYPPQPQKPSYPQNP44
CAAACGCCTTCGTATCCTCAGCAACCTCAAAAGCCTTCGTACCCTCAGAATCCTCAAACGCCTTCG 225
QTPSYPQQPQKPSYPQNPQTPS66
TACCCTCAGTATCCTCAAACACCTTCAAACCCTCAGCAACCTCAGTATCCTCAAACACCTTCAAAC 291
YPQYPQTPSNPQQPQYPQTPSN88
CCTCAGTATCCTCAAACGCCTTCGTACCCTCAGAATCCTCAAACGCCTTCGTACCCTCAGAATCCT 357
PQYPQTPSYPQNPQTPSYPQNP 110
CAAACGCCTTCGTACCCTCAGAACCCTCAGCAACCTCAATTCTCGTGGGATTTTTCAAAGCCTACA 423
QTPSYPQNPQQPQLSWDFSKPT 132
AAACCTCAATATCCTAAGCCCCAAAGGCCTCCATCAAAACCTCAATATCCTAGGCCCCAAACGCCT 489
KPQYPKPQRPPSKPQYPRPQTP 154
CCTTCAAAACCTCAATATCCTAGGCCTCAAACGCCCCAACAACCTGGAAAAAAACAATGGGATGAT 555
PSKPQYPRPQTPQQPGKKQWDD 176
ACAAAGACTCCGAATGTCCCTTCCAAGAGACCAGAGGCCCCTGGAGTTCCCACCCCTAAAAGTTGT 621
TKTPNVPSKRPEAPGVPTPKSC 198
GACGTGGAAGTAGCTTCAAGAGTCCCCTGTGGAGCTTCTGCCGTCTCTGCTACTGAATGTGAGGCC 687
DVEVASRVPCGASAVSATECEA 220
AGAGACTGTTGCTTTGATGGCCAGTCATGCTACTTTGCAAAAGGAGTGACAGTCCAGTGTACCAAG 753
RDCCFDGQSCYFAKGVTVQCTK242
GATGGCCATTTTATCGTTGT1GTGGCCAAAGATGTCACCCTGCCACACATTGACCTTGAAACAATC 819
DGHFIVVVAKDVTLPHIDLETI 264
TCATTGTTGGGAGGAGGTCAAGGCTGTACACATGTTGACCCCAATTCACTTTTTGCCATCTACTAC 885
SLLGGGQGCTHVDPNSLFAIYY 286
TTTCCCGTTACTGCTTGTGGGACTGTTGTCATGGAGGAGCCTGGCGTTATAATGTATGAGAATCGG 951
FPVTACGTVVMEEPGVIMYENR 309
ATGACCTCCTCATATGAAGTAGGAGTTGGGCCTCTTGGAGCCATTACCAGGGACAGCACCTACGAA 1017
MTSSYEVGVGPLGAITRDSTYE 330
TTGCTCTTCCAGTGTAGGTACATTGGCACCTCAGTTGAAACTTTGGTGGTCGAAGTGCTGCCATTA 1083
LLFQCRYIGTSVETLVVEVLPL 352
GACAATCCTCCTCCAGCAGTTGCTGAGCTCGGACCGATCAGAGTGGCCCTTAGGTTGGCCAATGGC 1149
DNPPPAVAELGPIRVALRLANG 374
CAGTGTGCTACAAAGGGTTGCAACGAAGCGGAGGTAGCCTACACCTCCTACTATTTGGACTCAGAC 1215
QCATKGCNEAEVAYTSYYLDSD 396
TA7CCGATTACCAAGATACTGAGGGATCCCGTGTATGTGGAGGTTCAGCTCCTTGAAAAGACAGAT 1281
YPITKILRDPVYVEVQLLEKTD 418
CCCGCTCTGGTTCTGACTCTTGGACGTTGTTGGGCAACCACTAGCCCCAATCCTCACAGCTTGCCC 1347
PALVLTLGRCWATTSPNPHSLP 440
CAGTGGGACATTCTGATTGACGGATGTCCCTACACGGATGATCGTTACCTCTCCACACTGGTTCCA 1413
QWDILIDGCPYTDDRYLSTLVP 462
GTGGACGCCTCTTCTGGTCTGCAATTTCCAAGTCACTACCGGCGTTTCACTTTCAAAATGTTTACC 1479
VDASSGLQFPSHYRRFTFKMFT 484
TTTGTGGACACCACTGCAATGGACCCCCTGAGGGAAAATGTGTACATTCACTGTAGCACAGCTGTG 1545
FVDTTAMDPLRENVYIHCSTAV 506
TGCGTGCCAGGACAGGGTGTCAGCTGCGAACCATCATGCAACAGAAAAGGAAAGAGAGACACTGAG 1611
CVPGQGVSCEPSCNRKGKRDTE 528
GCTGCAGAGCAGAGGAAGGTCGAACCAAAGGTTGTGGTTTCGTCCGGAGAAGTGATCATGACCGCT 1677
AAEQRKVEPKVVVSSGEVIMTA 550
CCTCAGGAGTAAtctgggacaagctcaggaattcatctgggaacatttagacaaaactctttgaaa 1743
P Q E 553
atcaacaaggttgttgaacagtaaataaaaatgtcaccctaagtaaaaaaaaaaaaaaaaaaaaaa 1809
aaaaaaa 1816

gaacttttcagatcacttgtgtttgtgaagcc 32
102
B)
ATGATGATGAAGTGGACTGTCTTTTGCGTTGTGGCGCTGGCTTTGCTTGGCAGCTTCTGTGATGCT 98
MMMKWTVFCVVALALLGSPCDA22
CAGGGGTACGCGAAACCTGGTAAGCCATCAAAACCCCAATCACCACCTACGCAAAACCAACAGCAA 164
QGYAKPGKPSKPQSPPTQNQQQ44
TTGCAGACATTTGAGAAAGAGCTCACCTGGAAGTACCCCGACGATCCCCAGCCAGACCCCAAGCCT 230
LQTFEKELTWKYPDDPQPDPKP66
AATGTGCCATTTGAGTTGAGATACCCTGTTCCTGCTGCAACCGTTGCTGTTGAGTGCAGAGAGAGC 296
NVPFELRYPVPAATVAVECRES88
ATAGCTCACGTGGAGGTCAAGAAAGACATGTTTGGCACCGGCCAGCCGATCAATCCAAATGACCTC 362
IAHVEVKKDMFGTGQPINPNDL 110
ACCCTGGGTAACTGTGCGCCTGTTGGAGAGGATAGTGCCGCTCAAGTGTTGATTTATGAAGCTGAA 428
TLGNCAPVGEDSAAQVLIYEAE 132
CTGCATCAATGCGGAAGCCAGCTGATGATGACAAATGATGCTCTCGTCTACACCTTCGTTTTGAAC 494
LHQCGSQLMMTNDALVYTFVLN 154
TATAACCCTACGCCTTTGGGATCGGTTCCTGTTGTGAGAACCTCCCAAGCTGCTGTGATCGTGGAA 560
YNPTPLGSVPVVRTSQAAVIVE 176
TGCCACTACCCAAGGAAGCACAATGTGAGCAGCCTTCCTCTGGATCCCCTTTGGGTCCCATTCTCT 626
CHYPRKHNVSSLPLDPLWVPFS 198
GCAGTTAAGATGGCTGAGGAGTTCCTGTACTTCACTATGAAACTCATGACTGATGACTGGATGTAC 692
AVKMAEEFLYFTMKLMTDDWMY 220
CAGAGGCCAAGCTACCAGTS.TTTCCTGGGAGACCTGATCCGTATAGAGGTTACTGTCAAGCAATAC 758
QRPSYQYFLGDLIRIEVTVKQY 242
TTCCATGTACCCCTGCGTGTTTACGTGGACAGATGTGTGGCAACCCTCTCTCCTGATGTAACCTCA 824
FHVPLRVYVDRCVATLSPDVTS 264
AGCCCCAACTATGCCTTCATTGATAACTTTGGGTGTTTGATTGACGCCAGAATCACAGGCTCTGAC 890
SPNYAFIDNFGCLIDARITGSD 286
TCAAAGTTCATGGCTCGCACCCAAGAGAACCACCTTCAGTTCCAGCTGGAGGCCTTCAGGTTCCAG 1956
SKFMARTQENHLQFQLEAFRFQ 309
AATTCTGACAGTGGAGTGATCTACATCACCTGCTACTTGAAGGCAACGTCTACTAGCCAGGCCATA 1022
NSDSGVIYITCYLKATSTSQAI 330
GACAGCCAGCACAGAGCTTGTTCCTACACTGGCGGATGGAGGGAGGCCAGTGGAGTTGATGGAGCT 1088
DSQHRACSYTGGWREASGVDGA 352
TGTGGTTCTTGTGAGACCAACGTGACGCCGTACACCGCTCCAGCAGTTACATTCGCTTCACCACCT 1154
CGSCETNVTPYTAPAVTFASPP 374
GTCGTTGTTACTGATGGTGGTGGAGTAACGCTTCCAGCTCCAGGCAGTCCAAAAGTCCCTTATAAT 1220
VVVTDGGGVTLPAPGSPKVPYN 396
CCGAGGAAAGTCCGTGACGTCACCCAAGCCGAAATTTTGGAATGGGAAGGCGTTGTCTCTCTGGGC 1286
PRKVRDVTQAE ILEWEGVVSLG 418
CCCATCCCCATCATGGAGAAGAAACTCTGAaaaacagaagtgtaacatgatattccgccgtagcca 1352
PIPIMEKKL- 427
tgaacaccataataaaaagtatcattggttcatatcgctgtctatgttatgcctatgtctcatggt 1418
agattttcttaaacaagtaacaaacccccacttagtctcttaaatctgcttaaaattttaaatatt 1484
gacaaatttccaaaaaattgtagaggtctttttttaggggggagggataaatgaaggaaaacttgt 1550
cttagattcccttttatgtaatggtaaggcagtgtgtggacccccatgtgtccagcaccataatct 1616
gtaaccctecttttcatgaaaataaaattcgcaactataaaaaaaaaaaaaaaaaa 1672
Figure 5.2-continued

103
C)
actaactagaccagacagcttcgaggt 27
ATGGCAAGTCACTCGAGTGTCACCCGTTCGGCCGCGCTGGCTCTGCTATGCTGCTTAGCTGGGAAA 93
HASHWSVTRWAALALLCCLAGK22
GGAGCAGAGGCTCAGAAGGGTTCGTATCCTCCGCAACCTCAAAAGCCTTCGTACCCTCAGAATCCT 159
G pgpgA QKOSYPPQPQKPSYPQNP44
CAAACGCCTTCGTATCCTCAGCAACCTCAAAAGCCTTCGTACCCTCAGAATCCTCAAACGCCTTCG 225
QTPSYPQQPQKPSYPQNPQTPS66
TACCCTCAGTATCCTCAAACACCTTCAAACCCTCAGCAACCTCAGTATCCTCAAACACCTTCAAAC 291
YPQYPQTPSNPQQPQYPQTPSN88
CCTCAGTATCCTCAAACGCCTTCGTACCCTCAGAATCCTCAAACGCCTTCGTACCCTCAGAATCCT 357
PQYPQTPSYPQNPQTPSYPQNP 110
CAAACGCCTTCGTACCCTCAGAACCCTCAGCAACCTCAATTGTCGTGGGATTTTTCAAAGCCTACA 423
QTPSYPQNPQQPQLSWDFSKPT 132
AAACCTCAATATCCTAAGCCCCAAAGGCCTCCATCAAAACCTCAATATCCTAGGCCCCAAACGCCT 489
KPQYPXPQRPPSKPQYPRPQTP 154
CCTTCAAAACCTCAATATCCTAGGCCTCAAACGCCCCAACAACCTGGAAAAAAACAATGGGATGAT 555
PSKPQYPRPQTPQQPGKKQWDD 176
ACAAAGACTCCGAATGTCCCTTCCAAGAGACCAGAGGCCCCTGGAGTTCCCACCCCTAAAAGTTGT 621
TKTPNVPSKRPEAPGVPTPKSC 198
GACGTGGAAGTAGCTTCAAGAGTCCCCTGTGGAGCTTCTGCCGTCTCTGCTACTGAATGTGAGGCC 687
DVEVASRVPCGASAVSATECEA 220
AGAGACTGTTGCTTTGATGGCCAGTCATGCTACTTTGCAAAAGGAGTGACAGTCCAGTGTACCAAG 753
RDCCFDGQSCYFAKGVTVQCTK 242
GATGGCCATTTTATCGTTGTTGTGGCCAAAGATGTCACCCTGCCACACATTGACCTTGAAACAATC 819
DGHFIVVVAKDVTLPHIDLETI 264
TCATTGTTJGGAGGAGGTCAAGGCTGTACACATGTTGACCCCAATTCACTTTTTGCCATCTACTAC 885
SLL.GGGQGCTHVDPNSLFAIYY 286
TTTCCCGTTACTGCTTGTGGGACTGTTGTCATGGAGGAGCCTGGCGTTATAATGTATGAGAATCGG 951
FPVTACGTVVMEEPGVIMYENR 309
ATGACCTCCTCATATGAAGTAGGAGTTGGGCCTCTTGGAGCCATTACCAGGGACAGCACCTACGAA 1017
MTSSYEVGVGPLGAITRDSTYE 330
TTGCTCTTCCAGTGTAGGTACATTGGCACCTCAGTTGAAACTTTGGTGGTCGAAGTGCTGCCATTA 1083
LLFQCRY IGTSVETLVVEVLPL 352
GACAATCCTCCTCCAGCAGTTGCTGAGCTCGGACCGATCAGAGTGGCCCTTAGGTTGGCCAATGGC 1149
DNPPPAVAELGPIRVALRLANG 374
CAGTGTGCTACAAAGGGTTGCAACGAAGCGGAGGTAGCCTACACCTCCTACTATTTGGACTCAGAC 1215
QCATKGCNEAEVAYTSYYLDSD 396
TATCCGATTACCAAGATACTGAGGGATCCCGTGTATGTGGAGGTTCAGCTCCTTGAAAAGACAGAT 1281
YPITKILRDPVYVEVQLLEKTD 418
CCCGCTCTGGTTCTGACTCTTGGACGTTGTTGGGCAACCACTAGCCCCAATCCTCACAGCTTGCCC 1347
PALVLTLGRCWATTSPNPHSLP 440
CAGTGGGACATTCTGATTGACGGATGTCCCTACACGGATGATCGTTACCTCTCCACACTGGTTCCA 1413
QWDILIDGCPYTDDRYLSTLVP 462
GTGGACGCCTCTTCTGGTCTGCAATTTCCAAGTCACTACCGGCGTTTCACTTTCAAAATGTTTACC 1479
VDASSGLQFPSHYRRFTFKMFT 484
TTTGTGGACACCACTGCAATGGACCCCCTGAGGGAAAATGTGTACATTCACTGTAGCACAGCTGTG 1545
FVDTTAMDPLRENVYIHCSTAV 506
TGCGTGCCAGGACAGGGTGTCAGCTGCGAACCATCATGCAACAGAAAAGGAAAGAGAGACACTGAG 1611
CVPGQGVSCEPSCNRKGKRDTE 528
GCTGCAGAGCAGAGGAAGGTCGAACCAAAGGTTGTGGTTTCGTCCGGAGAAGTGATCATGACCGCT 1677
AAEQRKVEPKVVVSSGEVIMTA 550
CCTCAGGAGTAAtctgggacaagctcaggaattcatctgggaacatttagacaaaactctttgaaa 1743
P Q E 553
atcaacaaqqttqttqaacaqtaaataaaaatgtcaccctaaqtaaaaaaaaaaaaaaaaaaaaaa 1809
aaaaaaa 1816
Figure 5.2continued

104
Mammalian ZP proteins sharing high identity with Chg 427 include mouse ZP3 (30%;
Ringuette et al., 1988), cat ZPC (32%; Harris et al., 1994) and human ZP3A (32%;
Chamberlin and Dean, 1990)(Fig. 5.4). Chg 427 shares little identity (18%) with
Chgs500 and Chg 553 (not shown). A Prosite scan predicted only one N-glycosylation
site from residue 184-187 (Fig. 5.2b).
Chg 553 was translated from a 1817-bp cDNA (Fig. 5.2c). Subtraction of a
predicted signal peptide (residue 1-26) resulted in a calculated molecular weight of
58,290. Chg 553 is 62% identical to Chg 500, and likewise shares identity with the
flounder ZP (52%) (Fig. 5.3a), mouse ZP1 (30%; Epifano et al., 1995), cat ZPB (29%,
Harris et al., 1995), and human ZPB (28%; Harris et al., 1994). The N-terminal region
of Chg 553 contains a proline-rich repeating domain that differs from that of Chg 500
by containing only half as many glutamine residues.
A ClustalV alignment (not shown) containing the ZP domains (Bork and Sander,
1992) of seventeen reported sequences, including the three Chg sequences, and five other
reported sequences from fish, plus three mouse, three cat, and three human ZPs, was
used in parsimony analysis. The shortest tree resulting from a heuristic search with 100
bootstrap replicates is presented in Figure 5.8. The resulting unrooted tree was drawn
according to the format of the Fitch analysis program in order to emphasize relatedness
among sequences rather than a deduced ancestral relationship. Bootstrap values are
indicated adjacent to the appropriate nodes. The results of the analysis suggest that three
major groups of ZP proteins can be described, each one containing a separate set of
mouse, cat, and human ZP sequences. In this paper we refer to these groups according

Figure 5.3
Alignments of Chg 500, Chg 553 and the flounder ZP protein.
A) A ClustalV (Higgins et al. 1992) alignment including predicted
amino acid sequences of Chg 500, Chg 553, and a flounder ZP
protein (Lyons et al. 1992). A conserved core region sharing
sequence identity with other ZP proteins and designated as a "ZP
domain" (Bork and Sander, 1992) is denoted by dark line. This
is the core domain used for drawing the tree shown in Figure 5.8.
B) A ClustalV alignment modified by eye of the proline-glutamine
rich repeating region from Chg 500 and the flounder ZP protein.
A Pro-Gln-X triplet is strongly conserved throughout the region.

106
A)
Chg 500
Chg 553
Flounder2?
1 MT MKLIYCCLLAVAIHG YLVG- AQPGKPQ YPSKPQ
1 MASHWSVTRWAA-LALL-CCLAGXGAEA QKGSY?PQPQKPSYPQNPQTPSY?
1 MAKRWSANSLVAQWLIYLVWTNVEVLGSRRRS RS SERGGR1XVQQTGHYH?AGKGQRYV
Chg 500
Chg 553
FlounderZP
3 5 -QPQQPQYPSKPQQPQQPQYPSKPQQPQQPQYPQQPQQPQQPQ YPSXPQY?SKPQQP
51 QQPQKPSYPQNPQTPSYPQYPQTPSNPQQPQYPQTPSNPQYPQT? SYPQNPQTPSYPQN?
6 1 QQRRRLHHDF SPQNPG- AEPPQTPQQPTYPQQPQQPQQPQQPKYPQQPQ QPQQP
Chg 500
Chg 553
FlounderZP
9 1 QQPQYPSXPQQPQ QPQYPQKPQQPQQPQYPQ KPQTP.TE -
111 QTPSYPQNPQQPQLSWDFSXPTKPQYPKPQRPPSKPQYPRPQTP P SKPQYPRPQTPQQPG
114 QQPKYPQQPQQPQ QPQQPKYPQQPQQPQQPQQPKYPQQPQQPKNPQPKNPQPPQ
Chg 500
Chg 553
FlounderZP
129 TFHTCDVP AP FRIQCGAPT XSNTECEAINCCFDGRMC
17 1 KKQWDDTKTPNVP SKRPEAPGVPTPKSCDVEVASRVPCGASAVSATXC2ARDCCFDGQ SC
163 PQKNPQPTXQQVSDDRI FCGVDPYLRIQCOVDDXTAAZCSALXCCFEGYQC
Chg 500
Chg 553
FlounderZP
166 YYGKSVTLQCTXDGQFIIWARDATLPHXDL3SXSLLGGGPNCGPVGTT3AFAXYQFFAD
231 YFAXGVTVQCTXDGHFXVVVAXDVTLPHXDLXTXSLLGGGQGCTHVDPNSLPAX^YFPVT
219 FFGKAVTVQCTXDAQFVVVVAKDATLPNLIINTXSI.QaEaQQCTAVDSNSEFAIFQ.FPVL
Chg 500
Chg 553
FlounderZP
226 CCGTIMTESPOVIXYSNRMASSYSVAVaPYGAXTRDSQYSLFVQCRYXaTSXEALVXEV-
291 ACGTVVMESPGVIMYENRMTSSYSVGVGPLGAITRDSTYSLLFQCRYXGTSVBTLWEVL
27 9 ACGSVVTEEPGT XXYSNRMTSSYBVDVGPNGVXTRDSFFXLQFQCRYTGLSXZTVVIEIL
Chg 500
Chg 553
FlounderZP
285 GLLPPPPGVAAPGPLRVELRLGNGECSVKGCTSEQVAYT3YYTDADYPVTKXLRDPVYVE
351 PLDNPPPAVAELGPIRVALRLAN0QCATKGCNEAZVAYTSYYLDSDY2ITKILRDPVYVE
33 9 P SNTPPRPVAALGPIRVQLRLGNGECETKGCNEVEAAYTSYYTEGDYPVTKVLRDPVYVE
Chg 500
Chg 553
FlounderZP
345 VRXLXRTDPNXVLTLGRCWATASPFPQSLPQWDLLXNGCPYQDDRYRTNLXPVDSSSGCL
411 VQLLZKTDPALVLTLGRCWATTSPNPRSLPQWDILXDGCPYTDDRYLSTLVPVDASSGLQ
39 9 VRLLZKRDPNLVLTLGRCWVTNSPNPHHQPQWDLLXDGCPYADDRYXSSLVPVG? SSGVN
Chg 500
Chg 553
FlounderZP
40 5 FPTHYRRFVFXMFTFVSGGGGASDATKKTP SDP SWNPLHXKVYXHCDAAVCQ PSMTNSCS
47 1 FPSHYRRFTFKMFTFVDTTAM DPLRZNVYXHCSTAVCVPGQGVSCS
4 5 9 FPTHYXRFIFKMFTFVDSSTLZPQRRR CTFTV
Chg 500
Chg 553
FlounderZP
46 5 PSCGRXXRZX SGSTKMX SRZEATXVSSK3WFTAT- Z
517 PSCNRXGXRDTZAAEQRKVZPKVVVSSG3VIMTAPQZ
491 VQLS ALVTQAAPVSRHATG
B)
Chg 500
FlounderZP
Chg 500
FlounderZP
33 PQQPQQPQY PSKPQQPQQPQYPSKPQQPQQ
97 CCTCAGCAACCCCAGCAGCCTCAGTATCCTTCCAAGCCTCAGCAACCCCAGCAGCCTCAGTATCCTTCGAAGCCTCACCAACCCCAGCAG
238 CCACAGACTCCACAGCAACCAACGTACCCACACCAACCACAGCAGCCACAGCAACCACAGCAACCAAAGTACCCACAGCAACCACAGCAG
BO PQTPQQPTyPQQPQgPQQPQQPKYPQQPQQ
62 PQYPQQPQQPQQPQYPSKPQYPSKPQQPQQ
186 CCTCAGTATCCCCAGCAGCCTCAACAACCCCAGCAGCCTCAGTATCC7TCGAAGCCTCAGTATCCTTCGAAGCCTCAGCAACCCCAGCAG
327 CCACAGCAACCACAGCAACCAAAGTACCCACAGCAACCACAGCAACCACAGCAACCACAGCAACCAAAGTACCCACAGCAACCACAGCAA
109 PyQPQ'3PKY PQQPQQPQQPQQPKY PQQPQQ

107
Chg 427 1
MadakaL-S? 1
Mouse 2? 3 1
Cat ZPC 1
Human ZP3A 1
Chg 427 60
MadakaL-S? 60
Mouse ZP 3 3 2
Cat ZPC 29
Human Z? 3A 3 2
Chg 427 113
MadakaL-S? 113
Mouse ZP 3 81
Cat ZPC 79
Human Z? 3A 81
Chg 427 178
MadakaL-S? 178
Mouse ZP 3 140
Cat ZPC 139
Human Z? 3 A 141
Chg 427 238
MedaJcaL-S? 238
Mouse ZP3 200
Cat ZPC 199
Human ZP 3 A 201
Chg 427 295
MadakaL-S? 295
Mouse ZP 3 259
Cat ZPC 256
Human ZP 3A 258
Chg 427 352
MedaJcaL-SF 352
Mouse ZP 3 319
Cat ZPC 316
Human ZP 3A 318
Chg 427 400
MadakaL-S? 393
Mouse ZP 3 379
Cat ZPC 376
Human ZP 3 A 378
MMMKWTV?CWALALLGS?CDAQ-GYAXPGKP SK?QS ? PTQNQQQLQTFZKZLTWXYPDD
MM-KFTAVCLVVLALLDG?CDAQHNYGK?SYPPTGSXTPQDPTQQKQLHZXZLTWKYPAD
MASSYFLFLCLLXCGGPELCNSQT LWLLPGG
MGLSYGLFICFLLWAGTGLCYP PT T TED
MELS YRLFICLLLWGS TELCYPQP LWLLQGG
PQPDPKPNVPFSLXYPVPAATVAVECRSSIAHVEVXXDMFGTGQPXNPNDLTLGN CA?
PQ PEAX?VVPFSQRYPVPAATVAVECREDLAHVSAXXDL7GIGQFXDFADLTLGT CPP
TPTPVGSSS? VKVZCLZAZLVVTVSRDL7GTGXLVQPGDLTLGSEGCQ?
KTHPSLP SSP S VVVECRH AWL VVNVSKNL7GTGRLVRPADLTLGP ESJCE?
ASHPETSVQP VLVECQEATLMVMVSXDL7GTGKLXRAADLTLGPEACE?.
VGZDSAAQVLXY3ASLHQCGSQLMMTNDALVYTFVLNYNPTPLGSVFWRTSQAAVXVEC
5 ASD? AAQVLXF3S PIiQNCGSVLTMTEDSLVYT3!TL2IYNPK?LGSAPVyRTSQAVVXYEC
RVSVDTD-WRFNAQLHECSSRVQMTKDAI.VYSTFLLHDPRPVSGLSILIlTNRVEV!PISq
LISGDSDDTVRFXVELHKCGNSVQVTEDAXVYSTFLLHNPRPMGNLSILRTNREVPISC
LVSMDTEDVVR73VGLHECGNSMQVTDDALVYST FLLHDPRPVQNL 3IVRTNRAEI?ISC
HYPRXHMVSSLPLDPLWVPFSAVKMASEFLYFTMXLMTDDWMY QRPSTQYFLGDLIRXEV
ZY?RZHNVSSLALDPLWV??3AAKMA22?LYFILXLT7DD?Q?3H2SJ1Y?L3DL>3ISA
R YPRQ GNVSSHPI 3 PTWVP FRATVS 33ZXLA7S LRLMEZNWNTZX3A? 7 FHLQEYAHLQA
RYPRHSNV3 SEA!LPTWVPFRTTML 3ZZKLAFS LRLME2DWG3 2KQ SPT 7QLGDLAHLQA
RYPRQGNVSSQAILPTWL2FRTTVFSEZXLTFSLRL2IEENWNAEXRSPTFHEGDAAHLQA
TVXQYFHVPLRVYVDRCVATLSP DVTSSPNYAFXDMFGCEIDARXTGSDSXE-MARTQ
TVXQ.YFHVPLRVYVDRCVATESP--DANSSPSYAFIDNYGCIiLDGRXTaSDSKF-VSRPA
3VQTGSHLPLQL7VDHCVATPSPLPDPNSSPYHFIVDFHGCLVDG-LSZ3FSAFQVPRPR
2VHTGRHIPLRLFVDYCVATLT PDQNASPHHTIVDFHGCEVDG-LSDAS SA7KAPRPR
ZIHTGSHVPLRLFVDHCVAXPT PDQNASPYHTIVDFHGCLVDG-LTDASSAFKVPRPG
SNHLQFQLZAFRFOMSDSGVXYXTCYLXATSTSQAIDSQHRACSYT GGWRSASGVDG
3NKLDFQLZAFRF' GADSGMXYXTCHLXATSAAY ? LDAEHRACSYI QGWXSVSGADP
PETLQFTVDVFHF.MSSRNTLYXTCHLKVAPANQIPDKLNKACSFNKTSQSWLPVEGDAD
PETLQFT7DTFHFAMDPRNMXYXTCHLJCVTPA3RVPDQLNKAC3FIKS SNRWFPVEGPAD
? DTLQFTVDVFHFAMDSRNMXYXTCHLXVTLAEQDPDELNXACSFSK? SNSWFPVEGPAD
ACGSCSTNVTP YTAP A VTFASPPVVVTDGGOVTLPAPGS - PKVP YNPRX
I CAS CSS GG FEVHA NAVVSHGT STLSGGGHGTGKPSD- ? SRK
ICDCCSHGNCSNSSS SQFQIHGPRQWSXLVSRNRRHVTDEADVTVGPLIFLGXANDQTVE
ICNCCNKGSCGLQGRSWRLSHLDRPWHKMASRNRRHVTEEADITVGPLIFLGXAADRGVE
ICQCCNKGDCGTP SHSRRQPHVMSQWSRSASRNRRHVTEEADVTVGPLI7LDRRGDHEVE
VRDVTQAZILZWEGV VSLGPIPIMEKXL
TREAAKTEVLZWSGD VTLGPIPIEERRV
GWTASAQTSVAL-GLGLATVAFLTLAAIVLAVTRKCHSSS- YLVSLPQ
GSTSPHTS--VMVGIGLATVLS LTLATIVLGLARRHHTASRPMICPVSASQ
QWAL? SDTSVVLLGVGLAVVVSLTLTAVILVLTRRCRTASHP VSASE
Figure 5.4 A ClustalV alignment of Chg 427, against the medaka L-SF protein
(Murata et al. 1995), the mouse ZP3 (Ringuette et al., 1988), cat ZPB
(Harris et al. 1994), and human ZP3A (Chamberlin and Dean 1990)A
conserved core region sharing sequence identity with other ZP proteins
and designated as a "ZP domain" (Bork and Sander, 1992) is denoted by
a dark line. This is the core domain used for drawing the tree shown in
Figure 5.8.

108
to the mouse sequence that they contain, thus the ZP1, ZP2, and ZP3 subdivisions. The
fish sequences represented in the tree were separated into two major subdivisions: one
containing Chg 427, medaka L-SF, and the three carp sequences, that was grouped with
the mouse ZP3 subdivision; and another containing the Chg 500, Chg 553, and the
flounder ZP that was grouped with the mouse ZP1 subdivision. Of the eight fish
sequences analyzed, none showed significant relatedness with the mammalian ZP2
subdivision; however, bootstrap values at the node dividing the ZP2 and ZP3
subdivisions were the lowest on the tree, arguing against a weighted interpretation
concerning this delineation.
Northern Blot Analysis
Northern blot analysis using three separate random-primed [32P]probes for Chg
550, 427, and 553 revealed Chg mRNAs present in liver RNA from both estrogen-treated
males and spawning females (Fig. 5.5). Furthermore, when 20.0 ig of ovary RNA was
blotted next to 2.0 /ug of liver RNA, Chg transcripts were apparent only in RNA from
the liver (Fig 5.6).
Vitelline envelope proteins
VEPs were isolated from ovarian follicles and resolved by SDS-PAGE into three
major Coomassie blue-staining bands at estimated molecular weights of 69,000, 60,000,
and 46,000, designated as VEP 69, VEP 60, and VEP 46, respectively (Fig. 5.7). At
least one other band could be visualized between VEP 60 and VEP 69, but appeared too

Figure 5.5 Northern blot analysis using Chg 500, 427, and 553 as probes.
A) Methylene blue staining of a nylon membrane indicating
equivalent loading of six lanes with total RNA. Lanes a, d, and
i contain RNA kb markers, lanes b, e, and g each contain 15 /xg
of the same total liver RNA isolated from a single estrogen-treated
male. Lanes c, f, and h contain 15 ng of total liver RNA isolated
from a single female, approximately four days before spawning.
28s and 18s ribosomal RNA bands are indicated in total RNA
lanes, suggesting RNA preparations lacking in RNAse
contamination.
B) Autoradiograph of the same nylon membrane after being cut
into three pieces and hybridized (65C) to the 32P-labeled random-
primed Chg probes indicated above the blot. Positions of RNA
markers are shown to the left, indicating 4.4, 2.37, and 1.35 kb
RNA.

110
Pre-Hyb Pre-Hyb Pre-Hyb
Chg 500 Chg 427 Chg 553
28s
18s
B
kb
4.4 >
2.37)
1.35 =
Chg 500 Chg 427
Chg 553
i ii i
a b c d e f
i i
g h i

Ill
faint to isolate. Amino acid analysis revealed that extraordinarily high proline and
glutamine compositions were present in VEP 69 and 60 (Table 5.1) agreeing well with
reported VEP compositions from other teleosts (Hyliner et al., 1991, 1995; Hamazaki
et al., 1987). The amino acid compositions of the isolated VEPs are compared to those
predicted from Chg cDNA translations in Table 5.1.
Discussion
We present the predicted primary structure of three liver-derived proteins, Chg
500, Chg 423, and Chg 553 (Fig. 5.2). We have shown that mRNAs hybridizing to
cDNA probes from each Chg occur in the liver RNA of estrogen-treated males and
spawning females, but are not detectable from ovarian RNA. Furthermore, the predicted
amino acid compositions of the Chgs are similar to the profiles of three VEPs isolated
from ovarian follicles (Table 5.1). We submit that although the Chgs differ from
mammalian ZP proteins by way of being estrogen-induced and synthesized in the liver,
they are in fact, related groups of proteins, as evidenced by the shared identity of a ZP
domain. Chg 500 and 553 can be more specifically grouped as homologs to the
mammalian ZP1 subfamily of molecules, while Chg 427 can be grouped with the
mammalian ZP3 subfamily (Fig. 5.8).
Northern Analyses
By showing no indication of Chg mRNA from 20 /g of ovarian RNA compared
with the ample Chg signals from only 2 /xg of liver RNA (Fig. 5.6), we provided strong

Figure 5.6 Northern blot analysis testing ovary vs. liver expression of Chgs.
A) Methylene blue staining of a nylon membrane blot of lanes
containing 2.0 /xg of total liver RNA next to loads of 20 /xg of
total ovarian RNA, from two identically-treated female fish.
Lanes a, e, and i contain liver RNA from fish 1, while lanes b, f,
and j, contain ten times more RNA isolated form the ovary of fish
1. Likewise, lanes c, g, and k contain liver RNA from fish 2,
while lanes d, h, and 1 contain ten times more total RNA, isolated
from the ovary of fish 2. RNA kb markers are indicated on the
left with 28s and 18s rRNA bands indicated on the right.
B) Autoradiograph showing the same nylon membrane after being
cut into three pieces and hybridized (65C) with the random
primed [32P] Chg probe indicated above the blot. Although ten
times more ovarian RNA than liver RNA was loaded onto the gel,
only bands from the lanes containing liver RNA hybridizing to the
Chg probes. Absolutely no hybridization was seen in the lanes
containing ovary RNA.

113
Pre-Hyb Pre-Hyb Pre-Hyb
Chg 500 Chg 427 Chg 553
i ii ii
abcdef ghijkl
RNA
kb
4.4-
2.37
1.35-
LOLOLOLOLOLO
Chg 500 Chg 427 Chg 553
i ii ii 1
abcdefghii kl
RNA *
kb
4.4-
2.37
1.35-
28s
18s
LOLOLOLOLO LO

114
evidence that the Chgs are indeed expressed in the liver, but not the ovary of spawning
females. We additionally showed that the Chgs can be induced in males by injection
with estradiol (Fig. 5.5). From the cDNA sequences, we expected the sizes of the
mRNAs encoding Chgs 500, 427, and 553 to be 1.64, 1.67 and 1.82 kb, respectively.
In northern blots, however, the Chg 500 probe hybridized to a band estimated to be 1.9
kb, while the Chg 427 hybridized to two bands, at 1.7 and 1.4 kb, and the Chg 553
probe hybridized to two bands at 2.3 and 1.8 kb. Indication of doublet mRNAs by
hybridization to Chgs 427 and 553 probes was observed from repeated stringent
hybridizations using different individual samples as
well as with probes representing different sections of the cDNA (not shown). We
interpret these data to suggest that two isoforms, possibly splicing variants of Chgs 427
and 553, are present in the liver total RNA. We also suggest that our cDNA clones
probably did not contain the total amount of 5 untranslated sequence, consistent with a
conservative estimate of mRNA sizes as compared with actual mRNAs indicated by the
gels.
The Predicted Structure of Chgs 500 and 553
The proline-glutamine-rich domains found in Chg 500 and Chg 553, along with
that of the flounder ZP (Lyons et al., 1993) represent a novel protein domain for
vertebrates. Although high proline and glutamic acid/glutamine compositions had long
been predicted through amino acid composition analyses of VEPs from F. heteroclitus
(Kaighn, 1964) and other fish (Young and Smith, 1956; Iuchi and Yamagami, 1976,

Table 5.1 Amino Acid Composition, Percent of Total
Chq 427 Cha 500 Chq 553
VEP46
VEP 60
VEP 69
ASN
3.7
2.1
3.0
ASP
5.2
3.8
4.6
ASX
8.9
5.9
7.6
11.4
8.9
7.8
GLN
5.7
11.3
8.7
GLU
5.2
5.0
4.6
GLX
10.9
16.3
13.3
11.2
14.4
18.3
SER
6.7
7.1
7.4
7.6
7.8
7.0
GLY
5.7
6.3
5.1
7.6
7.5
7.0
HIS
1.7
1.0
1.1
1.0
2.0
0.5
ARG
3.7
4.0
3.3
4.1
3.5
3.0
THR
6.9
6.7
8.0
9.9
8.9
8.1
ALA
7.2
5.0
4.9
8.5
5.1
4.9
PRO
9.4
13.4
15.0
8.5
13.3
15.9
TYR
4.7
5.4
5.7
3.5
5.3
5.8
VAL
9.4
5.9
8.0
9.0
7.3
5.7
MET
2.2
1.3
1.1
0.0
1.1
0.0
CYS
2.5
4.0
3.4
0.0
0.2
0.0
ILE
3.7
5.0
2.7
2.8
2.8
4.8
LEU
6.9
4.8
5.3
6.5
5.5
5.2
PHE
4.0
2.9
2.3
4.1
2.8
2.7
LYS
4.4
4.4
4.6
4.7
4.2
4.0
TRP
1.2
0.6
0.8

115
Mw, kD
69:
46
VEP 69
VEP 60
VEP 46
Figure 5.7 Isolation of three vitelline envelope proteins (VEPs) by SDS-PAGE. VEP
69, VEP 60, and VEP 46 were visualized by Coomassie blue staining,
indicating migration patterns according to estimated molecular weights of
69 kDa, 60 kDa, and 46 kDa. After resolution by SDS-PAGE, in Tris-
tricine buffers, the three VEPs were electroblotted to PVDF, and
submitted for protein analyses. An additional VEP, indicated by a weaker
staining band near 65 kDa, could be visualized, but was not isolated for
characterization.

116
Ohzu and Kusa, 1981; Kobayashi, 1982; Begovac and Wallace, 1989; Hyllner et al.,
1991, 1995), the extensive PQX repeat is nonetheless extraordinary. The finding that
(Pro-Glx-X) peptides were specifically released from the lysed chorions of medaka (Lee
et al., 1994) offers evidence consistent with the notion that components of Chg 500 and
553 contribute to the structure of the hardened chorion. Insights into the mechanism of
chorion hardening have been provided by the studies of Hagenmaier et al. (1985) and
Oppen-Bemtsen et al. (1990) in which Glx-Lys crosslinks were discovered in the
chorions of fertilized but not unfertilized eggs. A somewhat similar crosslinking
phenomenon has been suggested from the proline-rich repeats of mussel adhesive
proteins, where highly repetitive motifs containing hydroxyproline, lysine and tyrosine
(also modified to 3,4-dihydroxyphenylalanine) are involved in the formation of
underwater adhesives (Rzepecki et al., 1991). The exact mechanism whereby the
vitelline envelope of F. heteroclitus is hardened into a rigid chorion remains a mystery;
however we suggest that the high content of Pro, Gin, Lys, and Tyr found within the
PQX repeating region of Chg 500 and 553 are likely to play significant roles in this
process.
The predicted structure of Chg 427
The shortest of the three sequences reported here is that of Chg 427. It does not
contain an extensive repeating region as do the other Chgs; however the short sequence
(PGK PSK PQS PPT QNQ QQL Q) contains the high proline and glutamine content
characteristic of the repeats of Chgs 500 and 553. Chg 427 is the only sequence of the

117
three Chgs that contains a predicted N-glycosylation site. Rather than sharing identity
with Chgs 500 and 553 of F. heteroclitus, Chg 427 is most similar to that of the medaka
L-SF protein, followed closely by three carp "ZP3" sequences (not shown). These five
fish sequences contain ZP domains that share highest identity to the mouse ZP3
subfamily of mammalian ZPs (Fig. 5.8). Because the mouse ZP3 subfamily of molecules
is implicated as the primary sperm receptor in mammals, it is tempting to postulate a
similar role for Chg 427 as a likely candidate for sperm interaction in F. heteroclitus.
However, this postulation is significantly hindered by the well established existence of
a micropyle on F. heteroclitus eggs (Dumont and Brummett, 1980; Selman and Wallace,
1986). The micropyle of teleost eggs is essentially a narrow channel through the chorion
that provides homologous spermatozoa with direct access to the oocyte membrane
(Dumont and Brummett, 1980; reviews by Guraya, 1986; Hart 1990). Its existence
dismisses the necessity for most of the fertilization-associated interactions that have been
documented to occur in urchins and mammals, including sperm binding, induction of the
acrosome reaction, and the burrowing of sperm through the ZP en route to the oocyte
surface. Therefore, although Chg 427 shares identity with ZP3 molecules, it remains
unclear as to what function it fulfills.
The ZP Family of Proteins
A recent review by Harris et al. (1994) has attempted to lend order to the
currently confusing ZP nomenclature by separating all ZP proteins into three groups:
ZPA; ZPB; and ZPC, according to comparisons by protein alignments. Their ZPA

118
Figure 5.8 Parsimonious tree analysis of ZP domains from seventeen vertebrate ZP
homologs. The unrooted tree was obtained by running 100 bootstrap
replicates of a heuristic search (PAUP 3.1; Swofford, 1993) through a
ClustalV alignment, containing only the ZP domains of selected proteins,
and drawn according to the format of Fitch parsimony program. Three
divisions of ZP homologs resulted, each designated by a separate mouse
ZP protein: ZP1 division, ZP2 division, and ZP3 division. Three piscine
proteins, including Chg 500 and 553 were grouped within the ZP1
division, while five piscine proteins, including Chg 427 were grouped
within the ZP3 division. Bootstrap percentage values are indicated
adjacent to appropriate nodes.

119
group is homologous to mouse ZP2, their ZPB group is homologous to mouse ZP1, and
their ZPC group is homologous to mouse ZP3. Thus, this new organization scheme does
not improve on the original grouping offered by Wassarman et al. (1988a,b) based on
the three mouse ZPs: ZP1; ZP2; and ZP3. Therefore in our discussions of similarity to
mammalian ZP proteins we refer to three major groups of mammalian ZPs, according
to molecular identities shared with mouse ZP1, ZP2 or ZP3. These three "subfamilies"
are nevertheless contained within a larger family of related ZP proteins that can be
recognized by the possession of a conserved region designated as the ZP domain (Bork
and Sander, 1992).
Combining the present molecular data describing Chgs with the reports by Lyons
et al. (1993), Murata et al. (1995), and the Genbank entries for the carp ZP3 molecules,
it appears that the eight liver-derived VEP precursors described for teleosts can be
separated into two distinct groups: one containing Chg 500, Chg 553, plus the flounder
ZP protein; and another containing Chg 427, the medaka L-SF, and the three carp ZP3
sequences. Furthermore, these two fish groups share identity with two distinct
mammalian ZP subfamilies: Chgs 500 and 553 are grouped with the mouse ZP1
subfamily while Chg 427 is grouped with mouse ZP3 subfamily. Bootstrapping values
indicate a high confidence value associated with the Chg 427-mouse ZP3 subfamily
division, whereas the Chgs 500 and 553 might be expected to group with the mouse ZP2
or ZP1 subfamily. Whether Chgs 500 and 553 are closer related to the mouse ZP1 or
ZP2 subfamily, the trend remains that all reported piscine molecules separate into two
rather than three subdivisions. The lack of a third subtype of teleost homolog may

120
reflect a difference in the construction and function of the teleostean vitelline envelope
from that of mammals, but is more likely due to a lack of targeted investigation. For
instance, although, we report the amino acid compositions of three VEPs: 69, 60, and
46, there is at least one other F. heteroclitus VEP, with estimated molecular weight of
65,000 that could be visualized but was stained too faintly to isolate. Thus, other minor
F. heteroclitus VEPs still remain that may represent a third subclass of teleost VEPs.
It is also possible that a third sub-type may be synthesized by the ovary rather than the
liver of teleosts, explaining why the studies investigating liver cDNAs have not yet
discovered it. Results obtained with the pipefish Syngnathus scovelli (Wallace and
Begovac, 1989) as well as a preliminary study in F. heteroclitus (Hamazaki et al., 1989b)
suggest that at least one VEP may indeed be synthesized within the ovarian follicle.
By this preliminary characterization of Chg 500, 427, and 553, estrogen-induced,
liver-derived precursors to the vitelline envelope, we hope to set the stage for further
investigations of the regulation, structure and function of the vitelline envelope and
chorion. Besides being used to study development of the ovarian follicle, the Chgs
should provide excellent biomarkers to indicate either naturally occurring, or
toxicological states of estrogen induction in fish.

121
CHAPTER 6
GENERAL SUMMARY
In this dissertation, I have presented the nucleotide and predicted amino acid
sequences of five Fundulus heteroclitus cDNAs: two vitellogenins (Vtg I and Vtg II) and
three choriogenins (Chg 500, Chg 427, and Chg 553). All five of these protein products
are synthesized and secreted by the liver under estrogen induction, and transported by
the blood to the ovary. Vtgs I and II are endocytosed by the oocyte and processed into
liquid phase yolk proteins. In contrast, the Chgs are probably not taken up by the
oocyte, but rather laid down as components of the vitelline envelope and thus eventually
contribute to the structure of the chorion.
As an introduction, I described in Chapter 1, the historical context and initial
goals of the project. Probably the one most essential task accomplished in this work, the
construction of an estrogen-induced liver library, was completed before I became
affiliated with the study, by Marion Byrne, Jyotshnabala Kanungo, and Laura Nelson.
They constructed the library to obtain the primary structure of F. heteroclitus Vtg,
hoping to answer evolutionary as well as biochemical questions. After the initial
investigators disbanded, I became involved with the project, first as a compiler of
sequence data, but eventually leading to a role as primary caretaker of the library.
Chapter 1 concluded with a description of some of my unexpected adventures with the

122
library, primarily due to fortuitous annealing events, that led to the discovery of cDNAs
coding for Vtg II and the three choriogenins (Chgs).
In order to complete the cDNA sequence of Vtg I, two small overlapping regions
were isolated out of the library using anchored PCR (Fig 2.1). The resulting cDNA
(5112 bp) and predicted amino acid sequences (1704 residues) of Vtg I (gi:459202) were
described in Chapter 2 (Fig 2.2). Alignment of the F. heteroclitus Vtg against the other
known vertebrate Vtgs revealed 30%-40% sequence identity being shared among the
proteins (Fig. 2.4). The sturgeon Vtg sequence was found to share more identity with
chicken and Xenopus Vtgs than with the F. heteroclitus Vtg, suggesting that the F.
heteroclitus Vtg reflects a more derived rather than ancestral vertebrate protein (Fig 2.5).
We had hoped that by comparing the sequence of F. heteroclitus Vtg with the Vtgs of
other vertebrates, we might find an explanation for why F. heteroclitus yolk proteins
remain in a non-crystalline liquid phase. Analyses predicting secondary structure showed
no obvious differences to account for structural disparity. Although the poly serine
domain of the F. heteroclitus Vtg was the shortest of the five vertebrates, it possessed
the highest relative composition of serine. We hypothesized, assuming these serines were
phosphorylated, that the poly serine domain of F. heteroclitus Vtg may be more
hydrophilic and polar than that of the other Vtgs, perhaps preventing the recombination
of phosvitin and lipovitellin that occurs in granular or crystalline yolk. A graphical
representation was used to emphasize the codon usage of the polyserine domains from
sequenced vertebrate Vtgs (Fig. 2.6). A specific clustering of codons was observed:
TCX codons generally occurred near the 5 end of the domain, and AGY codons, often

123
the more frequently used, were grouped at the 3 end of the domain. As the first teleost
Vtg to be completely sequenced, this report represented an important step toward in
gaining comparative information on Vtg variability.
In Chapter 3, the sequence of Vtg II cDNA and predicted protein structure was
presented. Vtg II shares 45% overall identity with Vtg I, and 30-40% identity with the
other vertebrate Vtgs. The polyserine domain of Vtg II is slightly smaller than that of
Vtg I, but more surprising is the polyserine codon organization. The trend that had been
observed previously for Vtg I and other vertebrate Vtgs was not apparent in the
polyserine domain of Vtg II. Rather than a clustering of TCX and AGY codons, each
type were interspersed throughout the length of the domain. In a comparison of mRNA
expression, Vtg I transcripts were 10 times more prevalent than those of Vtg II from total
liver RNA isolated from spawning females and estrogen-induced males. According to
these data, we suggest that Vtg I is the primary and Vtg II a secondary yolk precursor
in F. heteroclitus.
N-terminal sequences of isolated yolk proteins were provided in Chapter 4. We
were able to map out a precursor-product relationship for seven yolk proteins by
comparing obtained N-terminal sequences with the predicted amino acid sequences
derived from Vtg I and Vtg II. A PEST site found in the Vtg region mapping to YP 125
was hypothesized as a possible factor influencing the proteolytic processing of YP 125
during oocyte maturation as compared to YP 105, which does not contain a PEST site.
In Chapter 5 the cDNA and predicted protein sequences of the choriogenins was
presented. The Chg mRNAs were shown to be expressed by the liver of reproductive

124
females and estrogen-induced males. It was further shown that Chg MRNA was not
indicated in RNA isolated from the ovaries of reproductive females. The F. heteroclitus
Chgs were recognized as homologs to the ZP proteins of mammals by their possession
of a ZP domain. A parsimonious tree analysis of the ZP domains from the three Chgs,
five other fish homologs, and nine other mammalian ZP proteins separated the molecules
into three major subdivisions. Chg 427 was grouped with mouse ZP3 and its homologs.
Chg 500 and Chg 533 were grouped with the mouse ZP1 subdivision, but not
significantly separated from the ZP2 subdivision. We isolated vitelline envelope proteins
(VEPs) from ovarian follicles and obtained their amino acid compositions for comparison
with the predicted compositions of the three Chgs. Although similarities existed between
the Chgs and the VEPs, we are currently awaiting N-terminal sequence analysis data to
provide unambiguous matches, verifying that Chgs are processed into bona fide VEPs.
The sequences of these five liver-derived molecules provide us with a substantial
amount of new information regarding the hepatic contribution to oocyte development.
Not only do we have data describing the primary structure of five important components
of the oocyte, but we have convincing evidence that they originate heterosynthetically in
the liver and are produced under estrogen induction. Although these results provide
exciting opportunities for further study, the flames of our ambition are truly fed by the
estrogen-induced liver library as an excellent tool to aid in elucidation of molecules that
contribute to reproductive processes.

REFERENCES
Amberg A, Meijlink FCPW, Mulder J, van Bruggen EFJ, Gruber M, AB G (1981)
Isolation and characterization of genomic clones covering the chicken vitellogenin
gene. Nucleic Acids Res 9:3271-3286
Aviv H, P Leder (1972) Purification of biologically active globin messenger RNA by
chromatography on oligothymidylic acid-cellulose. Proc Nat Acad Sci USA
69:1408-1412
Bairoch A, Bucher P, Hofmann K (1995) The Prosite database, its status in 1995.
Nucleic Acids Res 24:189-196
Baker ME (1988a) Invertebrate vitellogenin is homologous to human von Willebrand
factor. Biochem J 256:1059-1063
Baker ME (1988b) Is vitellogenin an ancestor of apolipoprotein B-100 of human low
density lipoprotein and human lipoprotein lipase? Biochem J 255:1057-1060
Balinsky BI (1965) In: An introduction to embryology. Second Edition. W.B. Saunders
Co., Philadelphia, p 137
Banaszak LJ, Sharrock W, Timmins P (1991) Structure and function of a lipoprotein:
Lipovitellin. Ann Rev Biophys Biophys Chem 20:221-246
Beer KE (1981) Embryonic and larval development of white sturgeon (Acipenser
transmontanus). M.S. thesis, University of California, Davis. 93 pp
Begovac PC, Wallace RA (1989) Major vitelline envelope proteins in pipefish oocytes
originate within the follicle and are associated with the Z3 layer. J Exp Zool
251:56-73
Bergeron JM, Crews D, McLachlan JA (1994) PCBs as environmental estrogens: turtle
sex determination as a biomarker of environmetnal contamination. Environ Health
Perspect 102:780-781
Bergink EW, Wallace RA (1974) Precursor-product relationship between amphibian
vitellogenin and the yolk proteins, lipovitellin and phosvitin. J Biol Chem
249:2897-2903
125

126
Bidwell CA, Carlson DM (1995) Characterization of vitellogenin from white sturgeon
Acipenser transmontanas. J Mol Evol 41:104-112
Bleil JD, Wassarman PM (1980) Structure and function of the zona pellucida:
Identification and characterization of the proteins of the mouse oocytes zona
pellucida. Dev Biol 76:185-202
Blumenthal T, Squire M, Kirtland S, Cane J, Donegan M, Speith J and Sharrock W
(1984) Cloning of a yolk protein gene family from Caenorhabditis elegans. J Mol
Biol 174:1-18
Bork P, Sander C (1992) A large domain common to sperm receptors (Zp2 and Zp3) and
TGF-/3 type III receptor. FEBS Lett 300:237-240
Bownes M (1992) Why is there sequence similarity between insect yolk proteins and
vertebrate liases? J Lipid Res 33:777-790
Brock ML, Shapiro DJ (1983) Estrogen regulates the absolute rate of transcription of the
Xenopus laevis vitellogenin genes. Cell 34:207-214
Byrne BM (1989) In: Phosvitin, an independent domain in vitellogenin genes. Ph.D
dissertation, University of Groningen, The Netherlands 122 pp
Byrne BM, van het Schip F, van de Klundert JAM, Amberg AC, Gruber M, and AB,
G (1984) Amino acid sequence of phosvitin derived from the nucleotide sequence
of part of the chicken vitellogenin gene. Biochemistry 23:4275-4279
Byrne BM, Gruber M, AB G (1989) The evolution of egg yolk proteins. Prog Biophys
Molec Biol 53:33-69
Caskey CT, Pizzuti A, Fu Y-H, Fenwick RG, Nelson DL (1992) Triplet repeat
mutations in human disease. Science 256: 784-789
Cerd JL, Petrino TR, Wallace RA (1993) Functional heterologous gap junctions in
Fundulus ovarian follicles maintain meiotic arrest and permit hydration during
oocyte maturation. Dev Biol 160:228-235
Chamberlin ME, Dean J (1990) Human homolog of the mouse sperm receptor. Proc Natl
Acad Sci USA 87:6014-6018
Chaudry HS (1956) The origin and structure of the zona pellucida in the ovarian eggs
of teleost. Z. Zellforschung 43:478-485

127
Chen J-S, Cho W-L, Raikhel AS (1994) Analysis of mosquito vitellogenin cDNA,
similarity with vertebrate phosvitins and arthropod serum proteins. J Mol Biol
237:641-647
Clark RC (1973) Amino acid sequence of a cyanogen bromide cleavage peptide from
hens egg phosvitin. Biochim Biophys Acta 310:174-187
Conte FS, Doroshov SI, Lutes PB, Strange EM (1988) In: Hatchery manual for the
white sturgeon. Publication 3322. The Regents of the University of California
Division of Agriculture and Natural Resources, Oakland CA
Cozens PJ, Cato AC, Jost JP (1980) Characterization of cloned complementary DNA
covering more than 6000 nucleotides (97%) of avian vitellogenin mRNA. Eur
J Biochem 112:443-450
Craik JCA, Harvey SM (1984) Phosphorous metabolism and water uptake during final
maturation of ovaries of teleosts with pelagic and demersal eggs. Mar Biol
90:285-289
Craik JCA, Harvey SM (1986) The causes of buoyancy in eggs of marine teleosts. J Mar
Biol Assoc 67:169-182
Denhardt DT (1966) A membrane-filter technique for the detection of complementary
DNA. Biochem Biophys Res Commun 23:641-646
Ding JL, Hee PL, Lam TJ (1989) Two forms of vitellogenin in the plasma and gonads
of male Oreochromis aureus. Comp Biochem Physiol 93B:363-370
Ding JL, Ho B, Valotaire Y, LeGuellec K, Lim EH, Tay SP, Lam TJ (1990) Cloning,
characterization and expression of vitellogenin gene of Oreochromis aureus
(Teleostei, Cichlidae). Biochem Int 20:843-852
Dodson RE, Acea MR, Shapiro DJ (1995) Tissue distribution, hormone regulation and
evidence for a human homologue of the estrogen-inducible Xenopus laevis
vitellogenin mRNA binding protein. J Steroid Biochem Molec Biol 52:505-515
Dumont JN, Brummett AR (1980) The vitelline envelope, chorion and micropyle of
Fundulus heteroclitus eggs. Gamete Res. 3:25-44
Dumont JN, Brummett AR (1985) Egg envelopes in vertebrates. In: Browder LW (ed)
Developmental biology. Vol 1. Plenum Press, New York pp 235-278
Eckelbarger KJ (1994) Diversity of metazoan ovaries and viellogenic mechanisms:
implications for life history theory. Proc Biol Soc Wash 107: 193-218

128
Epifano O, Liang LF, Familari M, Moos MC, Dean J (1995) Coordinate expression of
the three zona pellucida genes during mouse oogenesis. Development 121:1947-
1956
Evans AJ, Burley RW (1987) Proteolysis of apoprotein B during the transfer of very low
density lipoprotein from hens blood to egg yolk. J Biol Chem 262:501-504
Flegler C (1977) Electron microscopic studies on the development of the chorion of
viviparous teleost Dermogenys pusillus (Hemirhamphidae). Cell Tissue Res
179:255-270
Flickinger RA (1960) The relation of phosphoprotein phosphatase activity to yolk platelet
utilisation in the amphibian embryo. J Exp Zool 131:307-332
Flickinger RA, Rounds DE (1956) The maternal synthesis of egg yolk proteins as
demonstrated by isotopic and serological means. Biochim Biophys Acta 22:38-42
Flgel H (1967) Licht- und elektronmikroskopische Untersuchungen an Oozyten und Eim
einiger Knochenfische. Z. Zellforsch 83:82-116
Follett BK, Redshaw MR (1968) The effects of oestrogen and gonadotrophins on lipid
and protein metabolism in Xenopus laevis Daudin. J Endocrinol 40:439-585
Folmar LC, Denslow ND, Wallace RA, LaFleur GJ, Bonomelli S, Sullivan CV (1995)
A highly conserved N-terminal sequence for teleost vitellogenin with potential
value to the biochemistry, molecular biology and pathology of vitellogenesis. J
Fish Biol 46:255-263
Frohman MA, Dush MK, Martin GR (1988) Rapid production of full-length cDNAs
from rare transcripts: amplification using a single gene-specific oligo-nucleotide
primer. Proc Natl Acad Sci USA 85:8998-9002.
Fulton TW (1898) On the growth and maturation of the ovarian egs of teleostean fishes.
16th Annu Rep Fish Bd Scotl 3:88-134
Gerber-Huber S, Nardelli D, Haefliger J-A, Cooper DN, Givel F, Germond J-E, Engel
J, Green M, Wahli W (1987) Precursor-product relationship between vitellogenin
and the yolk proteins as derived from the complete sequence of a Xenopus
vitellogenin gene. Nucl Acids Res 15: 4737-4760
Greeley MS Jr., Calder DR, Wallace RA (1986) Changes in teleost yolk proteins during
oocyte maturation: correlation of yolk proteolysis with oocyte hydration. Comp
Biochem Physiol 84B:l-9

129
Greeley, MS Jr, Hols H, Wallace RA (1991) Changes in size, hydration and low
molecular weight osmotic effectors during meiotic maturation of Fundulus oocytes
in vivo. Comp Biochem Physiol 100A:639-647
Germond JE, Walker P, ten Heggeler B, Brown-Luedi M, de Bony E, Wahli W. (1984)
Evolution of vitellogenin genes: Comparative analysis of the nucleotide sequences
downstream of the transcription initiation site of four Xenopus laevis and one
chicken gene. Nucleic Acids Res 12:8595-8609
Guraya SS (1986) The cell and molecular biology of fish oogenesis. Monographs in
developmental biology, Vol 18 Karger, Basel
Guillette U Jr, Gross TM, Masson GR, Matter JM, Percival HF, Woodward AR (1994)
Developmental abnormalities of the gonad and abnormal sex hormone
concentrations in juvenile alligators from contaminated and control lakes in
Florida. Environ Health Perspect 102:680-688
Hagenmaier HE (1985) The hatching process in fish embryos VIII. The chemical
composition of the trout chorion (zona radiata) and its modification by the action
of the hatching enzyme. Zool Jb Physiol 89:509-520
Hamazaki T, Iuchi I, Yamagami K (1984) Chorion glycoprotein-like immunoreactivity
in some tissues of adult female medaka. Zool Sci 12:148-150
Hamazaki T, Iuchi I, Yamagami K (1985) A spawning female-specific substance reactive
to anti-chorion (egg envelope) glycoprotein antibody in the teleost, Oryzias
latipes. J Exp Zool 235:269-279
Hamazaki TS, Iuchi I, Yamagami K (1987a) Isolation and partial characterization of a
"spawning female-specific substance" in the teleost, Oryzias latipes. J Exp. Zool
242:343-349
Hamazaki TS, Iuchi I, Yamagami K (1987b) Production of a "spawning female-specific
substance" in hepatic cells and its acumulation in the ascites of the estrogen-
treated adult fish, Oryzias latipes J Exp Zool 242:325-332
Hamazaki TS, Nagahama Y, Iuchi I, Yamagami K (1989a) A glycoprotein from the liver
constitutes the inner layer of the egg envelope (zona pellucida interna) of the fish
Oryzias latipes. Dev Biol 133: 101-110
Hamazaki TS, Selman K, Wallace RA (1989b) Major components of the vitelline
envelope of the fish, Fundulus heteroclitus. Develop Growth Differ 31:407

130
Harris JD, Hibler DW, Fontenot GK, Hsu KT, Yurewicz EC, Sacco AG (1994) Cloning
and characterizaion of zona pellucida genes and cDNAs from a variety of
mammalian species: The ZPA, ZPB and ZPC gene families. DNA Sequence
4:361-393
Hart NF (1990) Fertilization in teleost fishes: mechanism of sperm-egg interactions. Int
Rev Cytol 121:1-66
Higgins DG, Bleasby AJ, Fuchs R (1992) CLUSTAL V: improved software for multiple
sequence alignment. Comput Appl Biosci 8:189-191
Hirose K (1976) Endocrine control of ovulation in medaka (Oryzias latipes) and ayu
(Plecoglossus altivelis). J Fish Res Bd Can 33:989-994
Hopp TK, Woods KR (1981) Prediction of protein antigenic determinants from amino
acid sequences. Proc Natl Acad Sci USA 78:3824-3828.
Hovemann B, Galler R, Walldorf U, Kupper H, Bautz EK (1981) Vitellogenin in
Drosophila melanogaster: sequence of the yolk protein I gene and its flanking
regions. Nucleic Acids Res 9:4721-4734
Hsiao S-M, Greeley MS Jr., Wallace RA (1994) Reproductive cycling in female
Fundulus heteroclitus. Biol Bull 186:271-284
The Huntingtons Disease Collaborative Research Group (1993) A novel gene containing
a trinucleotide repeat that is expanded and unstable on Huntingtons disease
chromosomes. Cell 72: 971-983
Hyllner SJ, Oppen-Bemtsen DO, Helvik JV, Walther BT, Haux C (1991) Oestradiol-17/3
induces the major vitelline envelope proteins in both sexes in telosts. J Endocrinol
131:229-336
Hyllner SJ, Barber HF-P, Larsson DGJ, Haux C (1995) Amino acid composition and
endocrine control of vitelline envelope proteins in European sea bass
(Dicentrarchus labrax) and gilthead sea bream (Spams aurata) Mol Repro Dev
41:339-347
Iuchi I, Yamagami K (1976) Major glycoproteins solubilized from the teleostean egg
membrane by the action of the hatching enzyme. Biochim Biophys Acta 453: 240-
249
Janin J (1979) Surface and inside volumes in globular proteins. Nature 277:491-492

131
Jarvik E (1980) In: Basic structure and evolution of vertebrates Vol. 1. Academic Press,
NY pp 439-446
Kaighn ME (1964) A biochemical study of the hatching process in Fundulus heteroclitus.
Dev Biol 9:56-80
Kanungo J, Petrino TR, Wallace RA (1990) Oogenesis in Fundulus heteroclitus. VI.
Establishment and verification of conditions for vitellogenin incorporation by
oocytes in vitro. J Exp Zool 25:313-321
Karasaki S (1963a) Studies on amphibian yolk. 5. Electron microscopic observations on
the utilization of yolk platelets during embryogenesis. J Ultrastruct Res 9:225-
247
Karasaki S (1963b) Studies on amphibian yolk. l.The ultrastructure of the yolk platelet.
J Cell Biol 18:135-151
Karasaki S (1967) An electron microscope study on the crystalline structure of the yolk
platelets of the lamprey egg. J Ultrastruct Res 18:377-390
Kishida M, Specker JL (1993) Vitellogenin in tilapia (Oreochromis mossambicus):
Induction of two froms by estradiol, quantification in plasma and characterization
in oocyte extract. Fish Physiol Biochem 12: 171-182
Kishida M, Specker JL (1994) Vitellogenin in the surface mucus of tilapia (Oreochromis
mossambicus): Possibility for uptake by the free swimming embryos. J Exp Zool
268:258-268
Kobayashi W and Yamamoto TS (1993) Factors inducing closure of the micropylar canal
in the chum salmon egg. J Fish Biol 42:385-394
Kozak M (1991) Structural features in eukaryotic mRNAs that modulate the initiation of
translation. J Biol Chem 266:19867-19870
Kunkel JG, Nordin JH (1985) Yolk proteins. In: Kerkut GA, Gilbert LI (eds)
Comprehensive insect physiology, biochemistry and pharmacology. Pergamon
Press, New York, pp 83-111
Kyte J, Doolittle RF (1982) A simple method for displaying the hydropathic character
of a protein. J Mol Biol 157:105-132
Laemmli, UK (1970) Cleavage of structural proteins during the assembly of the head of
bacteriophage T4. Nature 227 680-685

132
LaFleur GJ Jr, Byrne MB, Kanungo J, Nelson LD, Greenberg RM, Wallace RA (1995)
Fundulus heteroclitus vitellogenin: The deduced primary structure of a piscine
precursor to noncrystalline, liquid phase yolk protein. J Mol Evol 41:505-521
LaFleur GJ Jr, Byrne MB, Greenberg RM, Haux C, Wallace RA (1996) Liver-derived
cDNAs: vitellogenins and vitelline envelope protein precursors (choriogenins). In:
Thomas P, Goetz F (eds) Proceedings of the Fifth International Symposium on
the Reproductive Physiology of Fish, (in press) University of Texas Press,
Austin, TX
LaFleur GJ Jr, Thomas P (1991) Evidence for a role of Na+, K+- ATPase in the
hydration of Atlantic croaker and spotted seatrout oocytes during final maturatin
J Exp Zool 258:126-136
Lange RH (1981) Are yolk phosvitins carriers for specific cations? Comparative
microanalysis in vertebrate yolk platelets. Z Naturforsch 36:686-687
Lange RH (1985) The vertebrate yolk-platelet crystal: Comparative analysis of an in vivo
crystalline aggregate. Intemat Rev Cytol 87:133-181
Lange RH, Kilarski W (1986) Similarity in yolk-patelet structure of an ancient bony fish
(Acipenser) and an ancient reptile (Sphenodon). Tissue Cell 1:117-124
Larsson DGJ, Hyllner SJ, Haux C (1994) Induction of vitelline envelope proteins by
estradiol-17(8 in ten teleost species. Gen Comp Endocrinol 96:445-450
Lee K-S, Yasumasu S, Nomura K, Iuchi I (1994) HCE, a constituent of the hatching
enzymes of Oryzias latipes embryos, releases unique proline-rich polypeptieds
from its natural substrate, the hardened chorion. FEBS Lett 339:281-284
LeGendre N, Matsudaira, O (1988) Direct protein microsequencing from Immobilon-P
transfer membranes. Biotechniques 6:154-159
LeGuellec K, Lawless K, Valotaire Y, Dress M, Tenniswood M (1988) Vitellogenin
gene expression in male rainbow trout (Salmo gairdneri) Gen Comp Endocrinol
71: 359-371
Liang L-F, Chamow SM, Dean J (1990) Oocyte specific expression of mouse Zp-2:
Developmental regulation of the zona pellucida genes. Mol Cell Biol 10:1507-
1515
Liang L-F, Dean J (1993) Conservation of mammalian secondary sperm receptor genes
enables the promoter of the human gene to function in mouse oocytes. Dev Biol
156:399-408

133
Lyons CE, Payette KL, Price JL, Huang RCC (1993) Expression and structrual analysis
of a teleost homog of a mammalian zona pellucida gene. J Biol Chem 268:21351-
21358
MacDonald RJ, Swift GH, Przybyla AE, and Chirgwin JM (1987) Isolation of RNA
using guanidinium salts. Methods Enzymol 152:219-227
Mano Y, Lipmann F (1966) Enzymatic phosphorylation of fish phosvitin. J Biol Chem
241:3822-3833
Masuda K, Iuchi I, Yamagami K (1991) Analysis of hardening of the egg envelope
(chorion) of the fish, Oryzias latipes. Develop Growth Differ 33:75-83
Matsubara T, Sawano K (1995) Proteolytic cleavage of vitellogenin and yolk proteins
during vitellogenin uptake and oocyte maturation in barfin flounder (Verasper
moseri). J Exp Zool 272:34-45
McMaster GK, Carmichael GG (1977) Analysis of single- and double-stranded nucleic
acids on polyacrylamide and agarose gels by using glyoxal and acridine orange.
Proc Natl Acad Sci USA 74:4835-4838
McPherson R, Greeley MS Jr, Wallace RA (1989) The influence of yolk protein
proteolysis on hydration in the oocytes of Fundulus heteroclitus. Develop Growth
& Differ 31:475-483
Mecham DK, Olcott HS (1949) Phosvitin, the principal phosphoprotein of egg yolk. J
Amer Chem Soc 71:3670-3679
Munday KA, Ansari AQ, Oldroyd D, Akhtar M (1968) Oestrogen-induced calcium
binding protein in Xenopus laevis. Biochim Biophys Acta 166:748-751
Murakami M, Iuchi I, Yamagami K (1990) Yolk phosphoprotein metabolism during early
development of the fish, Oryzias latipes. Develop Growth Difieren 32:619-627
Murata K, Hamazaki TS, Iuchi I, Yamagami K (1991) Spawning female-specific egg
envelope glycoprotein-like substances in Oryzias latipes. Develop. Growth Differ.
33:553-562
Murata K, Sadaki T, Yasumasu S, Iuchi I, Enami J, Yasumasu I, Yamagami K (1995)
Cloning of cDNAs for the precursor protein of a low-molecular-weight subunit
of the inner layer of the egg envelope (chorion) of the fish Oryzias latipes. Dev
Biol 167:9-17

134
Nardelli D, Gerber-Huber S, van het Schip F, Gruber M, AB G, Wahli W (1987a)
Vertebrate and nematode genes coding for yolk proteins are derived from a
common ancestor. Biochemistry 26:6397-6402
Nelson G (1989) Phylogeny of major fish groups. In: Femholm B, Bremer K, Jmvall
H (eds) The hierarchy of life. Elsevier, New York p 330
Nelson JS (1984) In: Fishes of the world. Wiley-Interscience, New York, p 87
Ohzu E, Kusa M (1981) Amino acid composition of the egg chorion of rainbow trout.
Annot Zool Japn 54:241-244
Olin T, von der Decken A (1989) Yolk proteins in salmon (Salmo salar) oocytes, eyed
eggs, and alevins differing in viability. Can J Zool 68:895-900
Oppen-Bemtsen DO, Gram-Jensen E, Walther BT (1992b) Zona radiata proteins are
synthesized by rainbow trout (Oncorhynchus mykiss) hepatocytes in response to
oestradiol-17/3. J Endocrinol 135:293-302
Oppen-Bemtsen DO, Helvik JV, Walther BT (1990) The major structural proteins of cod
{Gadus morhua) eggshells and protein crosslinking during teleost egg hardening.
Dev Biol 137:258-265
Oppen-Bemtsen DO, Hyllner SJ, Haux C, Helvik JV, Walther BT (1992a) Eggshell zona
radiata-proteins from cod (Gadus morhua): extra-ovarian origin and induction by
estradiol-17/3. Int J Dev Biol 36:247-254
Oppen-Bemtsen DO, Olsen SO, Rong CJ, Taranger GL, Swanson P, Walther BT (1994)
Plasma levels of eggshell zr-proteins, estradiol-17/3, and gonadotropins during an
annual reproductive cycle of Atlantic salmon {Salmo salar). J Exp Zool 268:59-70
Opresko LK, Wiley HS, Wallace RA (1980) Differential postendocytotic
compartmentation in Xenopus oocytes is mediated by a specifically bound ligand.
Cell 22:47-57
Opresko LK, Wiley HS (1987) Receptor-mediated endocytosis in Xenopus oocytes. I.
Characterization of the vitellogenin receptor system. J Biol Chem 262: 4109-
4115
Oshiro T, Hibiya T (1981) Relationship of yolk globules fusion to oocyte water
absorption in the plaice Limando yokohamae during meiotic maturation. Bull Jpn
Soc Sci Fish 47:1123-1130

135
Pan ML, Bell WJ, Telfer WH (1969) Vitellogenic blood protein synthesis by insect fat
body. Science 165:393-394
Raag R, Appelt K, Xuong, N-H, Banaszak L (1988) Structure of the lamprey yolk lipid-
protein complex lipovitellin-phosvitin at 2.8 resolution. J Mol Biol 200:553-569
Rabinowitz M (1962) In: Boyer PD, Lardy H, Myrbck K (eds) The Enzymes. Vol. 6,
Academic Press, N.Y. p 119
Raff RA, Field KG, Olsen GJ, Giovannoni SJ, Lane DJ, Ghiselin MT, Pace NR, Raff
EC (1989) Metazoan phylogeny based on analysis of 18S ribosomal RNA. In:
Femholm B, Bremer K, Jomvall H (eds) The hierarchy of life. Elsevier, New
York, pp 247-260
Rina M, Savakis C (1981) A cluster of vitellogenin genes in the mediterranean fruit fly
Ceratitis capitata: sequence and structural conservation in dipteran yolk proteins
and their genes. Genetics 127:769-780
Ringuette MJ, Chamberlin ME, Baur AW, Sobieski DA, Dean J (1988) Molecular
analysis of a cDNA coding for ZP3, a sperm binding protein of the mouse zona
pellucida. Dev Biol 127:287-295
Rogers S, Wells R, Rechsteiner M (1986) Amino acid sequences common to rapidly
degraded proteins: the PEST hypothesis. Science 234:364-368
Rzepecki LM, Chin SS, Waite JH, and Lavin MF (1991) Molecular diversity of marine
glues: polyphenolic proteins from five mussel species. Mol Mar Biol Biotech
1:78-88
Schagger H, von Jagow, G (1987) Tricine-sodium dodecil sulfate-polyacrylamide ge
electrophoresis for the seperation of proteins in the range from 1 to 100 kDa.
Anal Biochem 166:368-379
Scott MG, Crimmins DL, McCourt DW, Tarrand JJ, Eyerman MC, Nahm MH (1988)
A simple in situ cyanogen cleavage method to obtain internal amino acid sequence
of proteins electroblotted to polyvinyldifluoride membranes. Biochem Biophys
Res Comm 155:1353-1359
Selman GG, Pawsey GJ (1965) The utilization of yolk platelets by tissues of Xenopus
embryos studied by a safranin staining method. J Embryol Exp Morph 14:191-
212
Selman K, Wallace RA (1983) Oogenesis in Fundulus heteroclitus.III. Vitellogenesis.
J Exp Zool 226:441-457

136
Selman K, Wallace RA (1986) Gametogenesis in Fundulus heteroclitus. Amer Zool
26:173-192
Selman K, Wallace RA (1989) Cellular aspects of oocyte growth in teleosts. Zool Sci
6:211-231
Sharrock WJ, Rosenwasser TA, Gould J, Knott J, Hussey D, Gordon JI, Banaszak L
(1992) Sequence of lamprey vitellogenin. Implications for the lipovitellin crystal
structure. J Mol Biol 226: 903-907
Shen X, Steyrer E, Retzek H, Sanders EJ, Schneider WJ (1993) Chicken oocyte growth:
receptor-mediated yolk deposition. Cell Tiss Res 272:459-471
Shibata N, Yoshikuni M, Nagahama Y (1993) Vitellogenin incorporation into oocytes of
rainbow trout, Oncorhynchus my kiss, in vitro: effect of hormones on denuded
oocytes. Develop Growth & Differ 35: 115-121
Speith J, Denison K, Zucker E, Blumenthal T (1985) The nucleotide sequence of a
nematode vitellogenin gene. Nucleic Acids Res 13:7129-7138
Speith J, Nettleton M, Zucker-Aprison E, Lea K, Blumenthal T (1991) Vitellogenin
motifs conserved in nematodes and vertebrates. J Mol Evol 32:429-438
Stifani S, Le Menn F, Nunez Rodriguez J, Schneider W (1990) Regulation of oogenesis:
the piscine receptor for vitellogenin. Biochim Biophys Acta 1045:271-279
Swofford DL (1993) In: PAUP: Phylogenetic analysis using parsimony, version 3.1.
Computer program distributed by the Illinois Natural History Survey, Champaign,
II
Taborsky G (1974) Phosphoproteins. Adv Prot Chem 28:1-210
Taborsky G (1980) Iron binding by phosvitin and its conformational consequences. J.
Biol. Chem. 255:2976-2985
Taborsky G, Mok C-C (1967) Phosvitin. J Biol Chem 242:1495-1501
Tata JR, Baker BS, Deeley JV (1980) Vitellogenin as a multigene family. Not all
Xenopus vitellogenin genes may be in an "expressible" configuration. J Biol
Chem 255:6721-6726
Taylor MA, DiMichele L, Leach GJ (1977) Egg stranding in the life cycle of the
mummichog, Fundulus heteroclitus. Copeia 1977:397-399

137
Terpstra P, AB G (1988) Homology of Drosophila yolk proteins and the triacylglycreol
lipase family. J Mol Biol 663-665
Tesonero JV (1977) Formation of the (zona pellucida) in the teleost, Oryzias latipes. I.
Morphology of early oogenesis. J Ultrastruct Res 59:282-191
Tesonero JV (1978) Formation of the chorion (zona pellucida) in the teleost, Oryzias
latipes. HI. Autoradiography of [3H] proline incorporation. J. Ultrastruct Res
64:315-326
Thiebaud CH, Fischberg M (1977) DNA content in the genus Xenopus. Chromosoma
59:253-257
Thorsen A, Fyhn HJ, Wallace RA (1993) Free amino acids as osmotic effectors for
oocyte hydration in marine fishes. In: Walther BT and Fyhn HJ (eds)
Physiological and biochemical aspects of fish development. University of Bergen,
Norway
Trewitt PM, Heilmann U, Degrugillier SS, Kumaran AK (1992) The boll weevil
vitellogenin gene: nucleotide sequence, structure, and evolutionary relationship
to nematode and vertebrate vitellogenin genes. J Mol Evol 34:478-492
Urist MR, Schjeide OA, Mclean FC (1958) The partition and binding of calcium in the
serum of the laying hen and of the estrogenized rooster. Endocrinol 63:570-585
Urist MR, Schjeide AO (1961) The partition of calcium and protein in the blood of
oviparous vertebrates during estrus. J Gen Physiol 44:743-756
van het Schip F, Samallo J, Broos J, Ophuis J, Mojet M, Gruber M, and AB G (1987)
Nucleotide sequence of a chicken vitellogenin gene and derived amino acid
sequence of the encoded yolk precursor protein. J Mol Biol 196:245-260
von Heijne G (1986) A new method for predicting signal sequence cleavage sites.
Nucleic Acids Res 14:4683-4690
Wahli W. (1988) Evolution and expression of vitellogenin genes. TIG 4:227-232
Wahli W, Dawid IB, Wyler T, Jaggi RB, Weber R, Ryffel GU (1979) Vitellogenin in
Xenopus laevis is encoded in a small family of genes. Cell 16:535-549
Wallace RA (1978) Oocyte growth in nonmammalian vertebrates In: Jones RE (ed) The
Vertebrate ovary. Plenum Publishing, New York, pp 469-502

138
Wallace, RA (1983) Interactions between somatic cells and the growing oocyte of
Xenopus laevis In: McLaren A, Wylie CC (eds) Current problems in germ cell
differentiation. Symposium of British society for developmental biology.
Cambridge University Press, Great Britain, pp 285-306
Wallace, RA (1985) Vitellogenesis and oocyte growth in nonmammalian vertebrates. In:
Browder LW (ed) Developmental Biology, vol 1. Plenum Press, New York, pp
127-177
Wallace RA, Begovac P (1985) Phosvitins in Fundulus oocytes and eggs. J Biol Chem
260:11268-11274
Wallace RA, Jared DW, Eisen AZ (1966) A general method for the isolation and
purification of phosvitin from vertebrate eggs. Can J Biochem 44:1647-1655
Wallace RA, Jared DW (1969a) Estrogen induces lipophosphoprotein in serum of male
Xenopus laevis. Science 160:91-92
Wallace RA and Jared DW (1969b) Studies on amphibian yolk. VII. Serum-
phosphoprotein synthesis by vitellogenic females and estrogen-treated males of
Xenopus laevis. Can J Biochem 46:953-959
Wallace RA, Morgan JP (1986a) Isolation of phosvitin: retention of small molecular
weight species and staining characteristics on electrophoretic gels. Anal Biochem
157:256-261
Wallace RA, Morgan JP (1986b) Chromatographic resolution of chicken phosvitin.
Biochem J 240:871-878
Wallace RA, Selman K (1980) Oogenesis in Fundulus heteroclitus II. The transition
from vitellogenesis to maturation. Gen Comp Endocrinol. 42:3454-354
Wallace RA, Selman K (1981) Cellular and dynamic aspects of oocyte growth in
teleosts. Amer Zool 21:325-343
Wallace RA, Selman K (1985) Major protein changes during vitellogenesis and
maturation of Fundulus oocytes. Dev Biol 110:492-498
Wallace RA, Camevali O, Hollinger TG (1990a) Preparation and rapid resolution of
Xenopus phosvitins and phosvettes by hight-performance liquied chromatography.
J Chromatography 519:75-86
Wallace RA, Hoch KL, Camevali O (1990b) Placement of small lipovitellin subunits
within the vitellogenin precursor in Xenopus laevis. J Mol Biol 213:407-409

139
Wallace RA, Greeley MS Jr, McPherson R (1992) Analytical and experimental studies
on the relationship between Na+, K+, and water uptake during volume increases
associated with Fundulus oocyte maturation in vitor. J Comp Physiol B 162:241-
248
Wang S, Smith DE, Williams DL (1983) Purification of avian vitellogenin III:
Comparison with vitellogenins I and II. Biochemistry 22:6206-6212
Wassarman PM (1988a) Zona pellucida glycoproteins. Ann Rev Biochem 57: 415-442
Wassarman PM (1988b) Fertilization in mammals. Sci Amer 256:78-84
Watanabe WO, Kuo C-M (1986) Water and ion balance in hydrating oocytes of the grey
mullet, Mugil cephalus (L.) during hormone-induced final maturation. J. Fish
Biol 28:425-337
White HB III (1987) Vitamin-binding proteins in the nutrtion of the avian embryo. J Exp
Zool Supplement 1:53-63
White HB III, Merrill AH Jr (1988) Riboflavin-binding proteins. Ann Rev Nutr 8:279-
299
Wiley HS, Wallace RA (1981) Multiple vitellogenins in Xenopus laevis give rise to
multiple forms of the yolk proteins. J Biol Chem 256:8626-8634
Wray W, Boulikas T, Wray VP, Hancock R (1981) Silver staining of proteins in
polyacrylamide gels. Anal Biochem 118:197-203
Yamagami K (1960) Phosphorous metablolism in fish eggs II. Transfer of some
phosphorous compounds from egg yolk into embryonic tissues in Salmo irideus
during development. Sci Papers Coll Gen Ed Univ Tokyo 11:153-161
Yamamoto M (1963) Electron microscopy of fish develpment. II. Oocyte-follicle cell
relationship and formation of chorion in Oryzias latipes. J Fac Sci Univ Tokyo
10:123-127
Yamamoto M, Yamagami K (1975) Electron microscopic studies on choriolysis by the
hatching enzyme ofthe teleost, Oryzias latipes. Dev. Biol 43:313-321
Yamamura J, Adachi T, Aoki A, Nakajima H, Nakamura R, Matsuda T (1995)
Precursor-product relationship between chicken vitellogenin and the yolk proteins:
the 40 kDa yolk plasma glycoprotein is derived from the C-terminal cysteine-rich
domain of vitellogeinin II. Biochem Biophys Acta 1244: 384-394

140
Yano K-I, Toriyama-Sakurai M, Watabe S, Izumi S, Tomino S (1994) Structure and
expression of mRNA for vitellogenin in Bombyx mor. Biochim. Biophys. Acta
(In Press)
Young EG and Smith DG (1956) The amio acid in the ichthulokeratin of salmon eggs.
J Biol Chem 219:161-164

BIOGRAPHICAL SKETCH
Gary LaFleur, Jr. was bom June 11, 1963, in Eunice, La. He was raised in
Eunice with his three sisters, Michelle, Holly, and Claire, and attended St. Edmund
Elementary and High School, as had his parents: Amanda Lee Rozas and Gary J.
LaFleur, M.D. After high school he attended Louisiana State University at Eunice for
two years, receiving an Associate of Science degree (1983), and then completed his
Bachelor of Science at L.S.U. in Baton Rouge (1984). His first encounter with the study
of estuarine teleosts was under the tuteledge of Dr. John Sharp during an Ichthyology
course offered at the Gulf Coast Research Laboratory in Ocean Springs, MS. After
completing his B.S. he became involved with the Ocean Research and Education Society,
in Gloucester, MA, first as a student, and then as a teaching assistant.
In 1986, he accepted a position as a technician in the lab of Peter Thomas at the
University of Texas Marine Science Institute in Port Aransas, TX. Under the mentorship
of Thomas, LaFleur began to study the repoductive physiology of sciaenid fishes. His
work as a technician was incorporated into a Master of Science degree from Corpus
Christi State University. Although LaFleurs research under Thomas consisted mainly
of experimental biochemistry, the field-oriented coursework at CCSU directed by J. Wes
Tunnel, left a lasting impression on him. Particularly infuential were field trips to the
141

142
coral reefs and cloud forests near Veracruz, where LaFleur gained an appreciation for
the structure and diversity of the oft-ignored invertebrates.
For his thesis research, LaFleur examined the role of Na+, K+- ATPase in the
hydration of Atlantic croaker oocytes (a maturation-associated phenomenon that had
gained renewed attention through the publications of Robin Wallace). LaFleur earned
his M.S. degree from CCSU in May, 1989.
Having been strongly influenced by the work of Robin Wallace, LaFleur sought
out a position in his lab at the University of Florida Whitney Laboratory. Under the
mentorship of Wallace, and with the support of Rob Greenberg in the Molecular Biology
Suite, LaFleur began investigating cDNAs isolated from an estrogen-induced liver library
in Fundulus heteroclitus. His doctoral dissertation was entitled "Estrogen-induced hepatic
contributions to ovarian follicle development in Fundulus heteroclitus: Vitellogenins and
Choriogenins." He completed his degree in May, 1996.
While studying in Port Aransas LaFleur met and fell in love with Susanna Lee
Lamers. They moved to Gainesville together and were married in St. Augustine on Leap
Day, 1992. In 1993 Susanna established Gene Genie, a company specializing in
computer analysis of genetic data. Their daughter, Hannah Alyse was bom on May 1,
1995. In March, 1996, LaFleur began work as a research scientist with Gary Wessel at
Brown University studying the molecular biology of cortical granule biogenesis in
urchins.

I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy
Robin A Wallace, Chair
Professor of Anatomy and Cell Biology
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Assistant Scientist
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
A\.
Christopher West
Associate Professor
of Anatomy and Cell Biology
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy. "7
Paul J. Linse
Associate Professor
of Anatomy and Cell Biology

I certify that I have read this study and that in
acceptable standards of scholarly presentation and i* full
as a dissertation for the degree of Doctor of Philos*
This dissertation was submitted to the Graduate Faculty of the College of
Medicine and to the Graduate School and was accepted as partial fulfillment of the
requirements for the degree of Doctor of Philosophy.
May, 1996
K
ean, College of Medicine
Dean, Graduate School



20
CONTIGUOUS cDNA SEQUENCE 5112 bp
pGLS
pMMB BBRHBS
pMMB9
seaeow> plasmid insert
lambda insert
I i degenerate probe
czi
oligo MB6
X #21
oligo MB6
X #5
pGL5
pMMBl
Figure 2.1 Cloning strategy used in isolating Furululus heteroclitus Vtg
cDNA. Lambda gtlO bacteriophage clones #5 and #21 were
isolated by tertiary screening with degenerate 17mer, MB6.
pGEM 3Z subclones (pMMB and pMMBl) were constructed
from digestion products of X21 and pMMB9 originated from X5.
Two remaining sections, pGLS and pGL5, were isolated by
anchored PCR, using a 5-/xl aliquot of the cDNA library as
template and exact primers, and then inserted into pT7Blue and
pCRIOOO vectors, respectively.


42
Vtg POLYSERINE DOMAIN
Total
% Serine
SAGY
STCX
Funduius
ATtY Codons
codons
171
codons
58
codons
63
codons
36
jiiiriudlHuw I44 iHtd 14
TCX Codons
Acipertser
f MMf < f
m i AGY Codons
212
37
51
49
i M UMHMM4 t l MMi
i* TCX Codons
Ichihyomyzon
1
i AfrY CdonS
238
48
55
45
'mu ¡titt u :ui
* TCX Codons
Xenopus
f f i'??7?'?r?7iT?|
m AGY Codons
249
39
53
47
4
' TCX Codons
Gxiilus
, ,anwwm 291
45
80
20
* m4
' TCX Codons
i =
one AGY serine codon
1 =
one TCX serine codon
20 codons
Figure 2.6. A comparison of the serine codon usage in the polyserine domains (see
Fig. 2.4) of five vertebrate Vtgs. Although the number of trinucleotide
repeats vary, the overall codon structure is conserved: a cluster of TCX
codons at the 5 end precedes a larger cluster of AGY codons. Only TCX
or AGY codons are shown. Relative lengths of polyserine domains are
drawn to scale.


Copyright 1996
by
Gary James LaFleur, Jr.


72
Several possible factors have been hypothesized to drive hydration, ranging from the
osmotic balance of ions (Hirose, 1976; Watanabe and Kuo, 1986; LaFleur and Thomas,
1991; Greeley et al., 1991; Wallace et al., 1992), ionic balance via gap junction control
(Cerd et al., 1993), and the colligative osmotic contribution of cleavage peptides and
free amino acids (Oshiro and Hibiya, 1981; Wallace and Selman, 1985; Greeley et al.,
1987; Thorsen et al. 1993). With these issues in mind we sought to characterize the
precursor-product relationship between Vtg and yolk proteins in F. heteroclitus, with
emphasis on the processing of YP 125. By completing the cDNA and putative protein
sequences of two F. heteroditus Vtgs (LaFleur et al., 1996; chapter 3), we obtained the
necessary blueprint for comparison of microsequencing data. In this paper we document
internal and N-terminal amino acid sequences from seven isolated yolk proteins, all of
which can be positioned within the Vtg I and Vtg II predicted protein sequences. Our
data suggest that the majority of yolk proteins are derived from Vtg I, and that a small
amount are derived from Vtg II. Additionally, we suggest that the rapid processing of
YP 125 during hydration is associated with the presence of a PEST site (Rogers et al.
1986) near its predicted C-terminus.
Materials and Methods
Ovarian follicles were dissected from the ovaries of reproductively active F.
heteroditus. Up to 20 prematurational follicles or up to 10 ovulated eggs were aliquoted
into a 1.5 ml eppendorf tube containing 500-750 ri of sample buffer (0.1 M Tris, pH
6.8, 2% SDS, 64 Mm dithiothreitol, 10% glycerol) on ice. The follicles were


1
M
K
A
V
V
L
A
L
T
L
A
F
V
A
G
<2
N
F
A
P
E
F
A
A
G
K
T
Y
V
Y
31
K
Y
E
A
L
I
L
G
G
L
P
£
E
G
L
A
R
A
G
L
K
i
S
T
K
L
L
L
s
A
61
A
D
Q
N
T
Y
M
L
K
L
V
E
P
E
L
S
E
Y
S
G
I
W
P
K
D
P
A
V
p
A
91
T
K
L
T
A
A
L
H
L
S
S
Q
F
P
S
S
L
N
T
P
M
V
F
V
G
K
V
F
A
P
121
E
E
V
S
T
L
V
L
N
I
Y
R
G
I
L
N
i
L
Q
L
N
I
K
K
T
H
K
V
Y
D
151
L
Q
E
V
G
T
Q
G
V
C
K
T
L
Y
S
I
S
E
D
A
R
I
E
N
X
L
L
T
K
T
181
R
D
L
s
N
C
Q
E
R
L
N
K
D
I
G
L
A
Y
T
E
K
c
D
K
c
Q
E
E
T
K
211
N
L
R
G
T
T
T
L
S
Y
V
L
K
P
V
A
D
A
V
M
I
L
K
A
Y
V
N
E
L
i
241
Q
F
S
P
F
S
E
A
N
G
A
A
Q
M
R
T
K
<2
S
L
E
F
L
E
I
E
K
E
P
I
271
P
S
V
K
A
E
Y
R
H
R
G
S
L
K
V E F 3
D
E
it,It
Q
T P.
. 1. Q. ,vK<
301
1
3
D-
A
P
A
Q
V
A
E
V
L
K
H
L
A
T
Y
N
I
E
D
V
H
E
N
A
P
L
K
331
F
L
E
L
V
Q
L
L
R
I
A
R
Y
E
D
L
E
M
Y
w
N
Q
Y
K
K
M
S
P
H
R
361
H
W
F
L
D
T
I
P
A
T
G
T
F
A
G
L
R
F
I
K
E
K
F
M
A
E
E
I
T
I
391
A
E
A
A
<2
A
F
I
T
A
V
H
M
V
T
A
D
P
E
V
I
K
i>
F E
s
X. V D S.
421
D
K
mi

e
N
Si
m.
X.
R
E
V
V
F
h
,s:
m
G
T
M
V
N
K
Y
C
N
K
T
V
D
451
c
P
V
E
L
I
K
s
I
Q
Q
R
L
s
D
A
I
A
K
N
E
E
E
N
i
I
L
Y
i
K
481
V
L
G
N
A
G
H
P
s
s
F
K
S
L
T
K
i
M
P
I
H
G
T
A
A
V
s
L
P
M
511
T
I
H
V
E
A
I
M
A
L
R
N
I
A
K
K
E
S
R
M
V
Q
E
L
A
L
Q
L
Y
M
541
D
K
A
L
H
P
E
L
R
M
L
S
C
I
V
L
F
E
T
S
P
s
M
G
L
V
T
T
V
A
571
N
S
V
K
T
E
E
N
L
Q
V
A
s
F
T
Y
S
H
M
K
s
L
S
R
S
p
A
T
I
H
601
P
D
V
A
A
A
C
S
A
A
M
K
I
L
G
T
K
L
D
R
L
S
L
R
Y
s
K
A
V
H
631
V
D
L
Y
N
S
S
L
A
V
G
A
A
A
T
A
F
Y
I
N
D
A
A
T
F
M
P
K
s
F
661
V
A
K
T
K
G
F
I
A
G
S
T
A
E
V
L
E
I
G
A
N
I
E
G
L
Q
E
L
I
L
691
K
N
P
A
L
S
E
S
T
D
R
I
T
K
M
K
R
V
I
K
A
L
S
E
W
R
S
L
p
T
721
S
K
P
L
A
s
V
Y
V
K
F
F
G
Q
E
I
G
F
A
N
I
D
K
P
M
I
D
K
A
V
751
K
F
G
K
E
L
P
I
Q
E
Y
G
R
E
A
L
K
A
L
L
L
S
G
I
N
F
H
Y
A
K
781
P
V
L
A
A
E
M
R
R
I
L
P
T
V
A
G
I
P
M
E
L
S
L
Y
S
A
A
V
A
A
811
A
S
V
E
I
K
P
N
T
S
P
R
L
S
A
D
F
D
V
K
T
L
L
E
T
D
V
E
L
K
841
A
E
I
R
P
M
V
A
M
D
T
Y
A
V
M
G
L
N
T
D
I
F
Q
A
A
L
V
A
R
A
871
K
L
H
S
V
V
P
A
K
I
A
A
R
L
N
I
K
E
G
D
F
K
L
E
A
L
P
V
D
V
901
P
E
N
I
T
s
M
N
V
T
T
F
A
V
A
R
N
I
E
E
P
L
V
E
R
I
T
P
L
L
931
P
T
K
V
L
V
P
I
P
i
R
R
H
T
S
K
L
D
P
T
R
952
N
3
M
L
D
3
3
E
L
L
P
M
E
964
E
E
D
V
2
p
I
P
2
Y
K
F
R
R
F
A
K
K Y C A
K
H
i
0
V
991
G
L
K
A
C
F
K F
A
3
2
N
G
A
S
i
<2
D
i
V
L
Y
K
L
A
G
S
H
N
F
1021
S
P
s
V
T
p
I
E
G
E
V
V
E
R
L
E
M
E
V
K
V
G
A
K
A
A
E
K
L
V
1051
K
R
I
N
L
s
E
D
E
S
T
E
E
G
G
P
V
L
V
K
L
N
K
I
L
S
S
R
R
N
1081
S
S
s
S
S
s
S
S
S
S
S
S
S
E
S
R
S
S
R
S
S
S
S
S
S
S
s
S
R
S
1111
S
R
K
I
D
L
A
A
R
T
N
S
s
S
S
S
s
S
R
R
S
R
S
S
S
S
s
S
S
S
1141
S
S
S
S
S
s
S
S
S
S
S
s
R
R
S
S
s
s
S
S
S
S
s
s
s
S
s
R
S
s
1171
R
R
V
N
S
T
R
S
S
S
S
s
S
R
T
S
s
A
s
S
I.
A
s
F
F
s
D
S
S
s
1201
S
S
s
S
s
D
R
R
S
K
E
V
M
E
K
F
<2
R
L
1220
H
K
K
M
V
A
S
G
S
S
A
s
S
V
E
A
I
Y
K
E
K
K
Y
L
G
E
1246
E_
E
A
V
V
A
V
I
L
R
A
7
K
A
D
1261
K
R
M
V
G
Y
2
L
G
F
Y
L
0
K
P
N
A
R
V
Q
I
I
V
A
N
I
s
s
D
S
1291
N
W
R
I
C
A
D
A
V
V
L
S
K
H
K
V
T
T
K
I
S
W
G
E
Q
C
R
K
Y
S
1321
T
N
V
T
G
E
T
G
X
V
S
S
S
P
A
A
R
L
R
V
S
W
E
R
L
P
s
T
L
K
1351
R
Y
G
K
M
V
N
K
Y
V
P
V
K
I
L
S
D
L
I
H
T
K
R
E
N
s
T
R
N
I
1381
S
V
I
A
V
A
T
S
E
K
T
I
D
I
I
T
K
T
P
M
S
S
V
Y
N
V
T
M
H
L
1411
P
M
C
I
P
I
D
E
I
K
G
L
S
P
F
D
E
V
I
D
K
I
H
F
M
V
S
K
1439
A
A
A
A
E
C
s
F
V
E
D
T
L
Y
T
F
N
N
R
s
Y
K
N
K
M
P
S
S
c
Y
1469
<2
V
A
A
<2
D
c
T
D
E
L
K
F
M
V
L L
R
K
D
S
S
E
Q
H
H
I
N
V
K
1599
I
S
E
I
D
I
D
M
F
P
K
D
0
N
V
T
V
K
V
N
E
M
E
I
P
P
P
A
c
L
1529
T
A
T
Q
Q
L
P
L
K
I
K
T
K
R
R
G
L
A
V
Y
A
P
S
H
G
L
Q
E
V
Y
1559
F
D
R
K
T
W
R
I
K
V
A
D
W
M
K
G
K
T
c
G
L
C
G
K
A
D
G
E
I
R
1589
Q
E
Y
H
T
P
N
G
R
V
A
K
N
S
I
S
F
A
H
S
W
I
L
P
A
E
S
C
R
D
1619
A
S
E
C
R
L
K
L
E
s
V
2
L
E
K
Q
L
T
I
H
G
E
D
S
T
C
F
S
V
E
1649
P
V
P
R
C
L
P
G
C
L
P
V
K
T
T
P
V
T
V
G
F
S
C
L
A
S
D
P
Q
T
1679
S
V
Y
D
R
S
V
D
L
R
Q
T
T
Q
A
H
L
A
c
S
C
N
T
K
C
3
YP 105
YP 125
} polyserine
domain
PEST
siie
YP 45
YP 39
YP 29


CHAPTER 3
SEQUENCE COMPARISON OF FUNDULUS HETEROCL1TUS
VITELLOGENINS I AND II
Introduction
Vitellogenin gene families have been described from various metazoan species
including Xenopus laevis (Wahli et al.,1979; Wiley and Wallace, 1980; Tata et al. 1980),
Caenorhabditis elegans (Blumenthal et al., 1984), and chicken (Evans et al., 1987; Byrne
et al., 1989). These small gene families from individual species have likewise been
shown to share genomic organization and sequence identity, establishing these related Vtg
genes as members of an ancient gene superfamily (Speith et al., 1985; Nardelli et al.,
1987; Byrne et al., 1989; Speith et al., 1991).
The existence of four X. laevis Vtg genes can be partially explained by the
hypothesis that an ancient duplication occurred in the X laevis genome (Thiebaud and
Fischberg, 1977; Bisbee et al., 1977). Tata et al. (1980) reported the extraordinary
occurrence of twelve to sixteen Vtg genes in X. laevis, stating that only four to six of
them were in an expressible form, and that the rest were nonexpressible or "silent".
Another example of a silent Vtg gene has been documented among the six Vtg genes of
C. elegans; whereas vit-2 to vit-6 have been shown to encode specific YP proteins, vit-1
has been described as a pseudogene (Speith et al., 1985) with no apparent translation
46


38
primary structure. F. heteroclitus Vtg shares 37% 38% identity with other vertebrate
Vtgs and it includes the characteristic N-terminal Lvl region, an internal Pv region and
a C-terminal Lv2 region. The genetic organization of the polyserine domain is consistent
with that found in other vertebrates, from lamprey to chicken, suggesting, at the latest,
a pre-gnathostome arrival of this domain into the Vtg gene. In contrast to other
vertebrate Vtgs, F. heteroclitus Vtg is predicted to be 100 amino acids shorter, and
contains a polyserine region with a 10-20% higher relative serine composition than the
other vertebrates Vtgs. We suspect that the occurrence of liquid phase yolk in F.
heteroclitus is in part due to differences within its Vtg polyserine domain as compared
with the polyserine domains of insoluble yolk producers. The higher than usual relative
serine composition would eventually be modified into a polyphosphoserine domain,
endowing the resulting Pv yolk protein with an uncommonly strong hydrophilic potential.
On examination of the alignment in Figure 2.4, the conserved organization of
vertebrate Vtg is evident: two well-aligned termini interrupted by a polymorphic
polyserine domain. The degree of Vtg conservation among several oviparous species is
further resolved by the phylogenetic tree analysis presented in Figure 2.5. The results
of the branch-and-bound tree search suggest that the present structure of F. heteroclitus
Vtg represents a substantial history of divergence from the ancestral osteichthyean Vtg.
Although, phylogenetically, F. heteroclitus and A. transmontanus are considered
monophyletic as actinopterygian fishes (Nelson, 1989), the Vtg structure of A.
transmontanus was found to be more closely related to the Vtgs of the two tetrapods than
it was to that of F. heteroclitus. Indeed many character traits of the genus Acipenser


27
3349
GCA
GCC
AGG
ACC
AAT
AGC
AGC
AGC
AGC
AGC
AGT
AGC
CGT
CGC
AGC
AGA
AGC
AGC
A
A
R
T
N
S
S
S
s
S
S
s
R
R
s
R
S
S
3403
AGO
AGC
AGC
AGC
AGC
AGC
AGT
AGC
AGT
AGC
AGC
AGC
AGC
AGC
AGC
AGC
AGC
AGC
s
S
s
S
S
S
s
S
S
s
S
S
S
s
S
S
S
S
3457
AGG
AGA
AGC
AGC
AGC
AGC
AGC
AGT
AGT
AGC
AGC
AGC
AGC
AGC
AGT
AGG
AGC
AGC
R
R
s
S
S
S
s
s
s
S
s
s
s
S
s
R
S
S
3511
AGG
AGA
GTC
AAC
TCA
ACA
AGA
TCC
AGC
AGC
AGT
TCA
AGT
AGG
ACC
AGC
TCT
GCA
R
R
V
N
S
T
R
S
S
S
S
S
S
R
T
S
S
A
3565
TCA
AGC
CTT
GCA
TCT
TTC
TTC
AGT
GAC
AGC
TCA
AGC
TCT
TCT
AGC
TCC
AGT
GAT
s
S
L
A
s
F
F
S
D
S
s
s
s
s
S
S
S
D
3619
CGT
CGC
TCA
AAG
GAA
GTG
ATG
GAG
AAG
TTC
CAG
AGG
TTA
CAC
AAG
AAA
ATG
GTC
R
R
S
K
E
V
M
S
K
F
Q
R
L
H
K
K
M
V
(Ah = 2.55)
3673
GCC
TCC
GGT
AGC
AGT
GCC
TCA
AGC
GTT
GAA
GCC
ATC
TAC
AAA
GAG
AAA
AAA
TAT
A
S
G
S
S
A
S
S
V
E
A
I
Y
K
E
K
K
Y
3727
CT^
GGC
GAG
GAA
GAA
GCC
GTT
GTG
GCA
GTG
ATT
CTC
CGT
GCT
GTC
AAA
GCT
GAC
L
G
E
E
E
A
V
V
A
V
I
L
R
A
V
K
A
0
3781
AAG
AGG
ATG
GTG
GGA
TAC
CAG
CTT
GGT
TTC
TAC
CTT
GAC
AAA
CCA
AAT
GCC
AGA
K
R
M
V
G
Y
Q
L
G
F
Y
L
D
K
P
N
A
R
3835
GTT
CAG
ATC
ATT
GTC
GCC
AAC
ATT
TCT
TCT
GAT
AGC
AAC
TGG
AGG
ATC
TGT
GCT
V
Q
I
I
V
A
N
i
s
S
D
s
N
w
R
I
C
A
3889
GAT
GCA
GTT
GTG
TTG
AGC
AAG
CAC
AAA
GTT
ACA
ACC
AAG
ATT
TCC
TGG
GGA
GAA
D
A
V
V
L
S
K
H
K
V
T
T
K
I
S
W
G
E
3943
CAG
TGC
AGG
AAA
TAC
AGC
ACC
AAT
GTT
ACA
GGA
GAG
ACT
GGT
ATT
GTT
TCT
TCA
Q
c
R
K
Y
S
T
N
V
T
G
E
T
G
I
V
S
S
3997
AGC
CCT
GCC
GCT
CGC
CTC
AGA
GTG
TCC
TGG
GAA
AGA
CTG
CCT
TCT
ACC
CTG
AAA
S
P
A
A
R
L
R
V
s
w
E
R
L
P
S
T
L
K
4051
CGC
TAT
GGA
AAG
ATG
GTT
AAC
AAG
TAC
GTT
CCT
GTT
AAA
ATA
TTG
TCT
GAC
TTG
R
Y
G
K
M
V
N
K
Y
V
P
V
K
I
L
S
D
L
4105
ATC
CAC
ACA
AAG
AGA
GAA
AAC
AGC
ACC
AGG
AAT
ATC
TCA
GTC
ATT
GCA
GTT
GCC
I
H
T
K
R
E
N
S
T
R
N
I
S
V
I
A
V
A
4159
ACA
TCT
GAA
AAG
ACA
ATT
GAC
ATC
ATA
ACC
AAA
ACT
CCA
ATG
AGC
TCT
GTC
TAC
T
S
E
K
T
I
D
I
X
T
K
T
P
M
S
s
V
Y
4213
AAT
GTC
ACT
ATG
CAT
CTT
CCC
ATG
TGT
ATT
CCC
ATT
GAT
GAG
ATC
AAA
GGT
CTC
N
V
T
M
H
L
p
M
c
I
p
I
D
E
I
K
G
L
4267
AGC
CCC
TTT
GAT
GAA
GTC
ATT
GAC
AAG
ATC
CAC
TTC
ATG
GTT
TCT
AAG
GCT
GCT
S
p
F
D
E
V
I
D
K
I
H
F
M
V
S
K
A
A
4321
GCA
GCT
GAA
TGC
AGC
TTC
GTC
GAA
GAC
ACA
CTC
TAC
ACA
TTC
AAC
AAC
AGG
AGC
A
A
E
C
S
F
V
E
D
T
L
Y
T
F
N
N
R
S
4375
TAC
AAG
AAC
AAG
ATG
CCT
TCC
TCT
TGC
TAC
CAG
GTT
GCA
GCA
CAG
GAC
TGC
ACA
Y
K
N
K
M
p
s
s
C
Y
Q
V
A
A
Q
D
C
T
4429
GAT
GAG
CTG
AAA
TTC
ATG
GTT
CTC
CTG
AGG
AAG
GAT
TCG
TCC
GAA
CAA
CAC
CAC
D
E
L
K
F
M
V
L
L
R
K
D
s
s
E
Q
H
H
(Ah 2.1)
Figure 2.2-continued


Michael Greenberg was truly inspirational, playing the King Arthur role at Camelot.
Rob Greenberg created a phenomenally free atmosphere in the molecular suite that was
very conducive to hard work and good fun. My adviser Robin Wallace was nothing less
than the perfect mentor. I have never learned so much from someone who said so little.
Robin never gave unsolicited advice. He let me grovel and groan and goof and grow,
taking delight in my triumphs and not noticing my failures. I will admire him always.
The molecular suite at the Whitney Lab, supervised by Rob Greenberg, has been
my incubator for five years. It offered a wonderful mix of data and good friends and
Rock and Roll. Primary fellow cloners that contributed to my journey included Bill
Buzzi, Bemd Eschweiler, Mike Jeziorski, Steve Munger, Clay Smith, and Chuck
Peterson. My late night comrade Sean Boyle deserves special mention for his Southern
kindness and great knack for developing shortcut protocols.
My mother and fathers enthusiasm and unwavering support provided a
cornerstone of stability for all of my years as a student. Thus, their parental investment
in me, the product of their germ cells, has far-outdone that of any ordinary somatic
contribution.
Finally, I would like thank Susanna, a Texas girl that stole my Louisiana heart.
She is an excellent scientist and a perfect mother, but her real talent lies in her knack for
swinging the world by the tail, bouncing over the white clouds, and killing the blues.
IV


63
Polyserine Domain
Number of
Ser codons
I = one AGY serine codon | j
j one TCX serine codon 20 codons
Figure 3.4 A comparison of the serine codon usage in the poly serine domains (see
Fig. 3.3) of F. heteroclitus Vtg I and Vtg II. Whereas the TCX and AGY
codons of the Vtg I poly serine domain are clustered into two separate
groups, the TCX and AGY codons of Vtg II show no apparent clustering.
Only serine codons are shown, with relative lengths of the domains drawn
to scale.


96
Experimental groups of fish were subjected to two intraperitoneal injections of
estradiol-17/3 (0.01 mg/g body weight) dissolved in 50 /ri coconut oil (Kanungo et al.,
1990). Control groups were sham-injected with coconut oil alone. The first
injection was performed on day 1, the second injection on day 4, followed by sacrifice
and liver dissection on day 8.
Total RNA was isolated from livers and ovaries by extraction with RNA Stat-60
reagents (Tel-Test "B", Inc. Friendswood, TX). Tissues were dissected and immediately
frozen in 1.5-ml tubes containing 500 /I of RNA Stat-60 emulsion, by immersion in
liquid nitrogen. Tissues were homogenized at 20C using a Kontes pestle and motor.
Typically, a 300 mg liver yielded 0.35 mg total RNA, with O.D. 260/280 ratios
consistently above 1.8. Total RNA samples were resuspended in DEP-C-treated water
and stored at -80C until used in analyses.
Before electrophoresis, aliquots of 15 /xg total RNA were precipitated in
isopropanol and denatured in 2.2 M formaldehyde, 50% formamide, 50 mM MOPS (pH
7.0) for 30 min at 65C. Samples were electrophoresed through gels containing 2.0%
agarose, 0.6 M formaldehyde, 50 mM MOPS, and 1 mM EDTA for 1.5 hours at 3.5
V/cm gel in 50 mM MOPS, 1 mM EDTA running buffer. RNA was blotted onto Magna
nylon membranes by capillary action with 20 X SSC, immobilized by U.V. crosslinking
and visualized by staining briefly with methylene blue. All hybridizations were carried
out at 65C in 1 X Denhardts solution, 6 X SSC, and 0.1% SDS without formamide.


115
Mw, kD
69:
46
VEP 69
VEP 60
VEP 46
Figure 5.7 Isolation of three vitelline envelope proteins (VEPs) by SDS-PAGE. VEP
69, VEP 60, and VEP 46 were visualized by Coomassie blue staining,
indicating migration patterns according to estimated molecular weights of
69 kDa, 60 kDa, and 46 kDa. After resolution by SDS-PAGE, in Tris-
tricine buffers, the three VEPs were electroblotted to PVDF, and
submitted for protein analyses. An additional VEP, indicated by a weaker
staining band near 65 kDa, could be visualized, but was not isolated for
characterization.


78
protein products at internal positions within their respective precursors (Fig. 4.2). Of
these five sequences, the most notable was that of YP 69, lining up to the N-terminus of
Vtg II, verifying the expression of this secondary Vtg as well as demonstrating that the
signal peptide cleavage site had been correctly predicted. The data from YP 69 also
indicate that the N-terminus in Vtg II is unblocked in contrast to the apparently blocked
N-terminus of Vtg I.
In order to identify the origin of YP 105 and YP 125, the protein bands were
again blotted onto PVDF membranes and proteolytically cleaved with Endo Lys C
(0.003 units/jig protein, Promega) in 50 Mm Tris, Ph 8.8, 0.2 M ammonium
bicarbonate, and 0.1% SDS, 0.1 Mm EDTA. The digestion products were again
separated by Tris-tricine gels, and visualized by silver staining. The reaction with Endo
LysC was confirmed as only a partial digestion by the isolation of peptide products larger
than those predicted if cleavage had occurred at every lysine residue. The pattern of
electrophoresed digestion products from YP 125 and YP 105 initially appeared to be
identical, indicating that the two yolk proteins originated from the same precursor
molecule (Fig. 4.3). However, a difference between the digestion products was
discovered when the 13-Kda peptides derived from YP 125 and YP 105 were sequenced.
The 13-Kda band isolated from YP 105 digestion contained two peptides, mapping near
the N-terminal region of Vtg I. The 13-Kda band isolated from YP 125 contained the
exact two peptides found in the YP 105 digestion, plus a third peptide (K Y C A K H
IGVGLKACFKFAS Q), that mapped much further along the Vtg I sequence,
to residue 982 (Figs. 4.4 and 4.5). We interpret these data as evidence that YP 105 and


52
Total RNA was isolated from livers by extraction with RNA Stat-60 reagents
(Tel-Test "B", Inc. Friendswood, TX). Tissues were dissected and immediately frozen
in 1.5-ml tubes containing 500 /zl of RNA Stat-60 emulsion by immersion in liquid
nitrogen. Tissues were homogenized at 20C using a Kontes pestle and motor.
Typically, a 300 mg liver yielded 0.350 mg total RNA, with O.D. 260/280 ratios
consistently above 1.8. Total RNA samples were resuspended in diethyl pyrocarbonate-
treated water and stored at -80C until used in analyses.
Before electrophoresis, aliquots of 15 ¡xg total RNA were precipitated in
isopropanol, and denatured in 2.2 M formaldehyde, 50% formamide, 50 mM MOPS (pH
7.0) for 30 min at 65C. Samples were electrophoresed through gels containing 1.0%
agarose, 0.6 M formaldehyde, 50 mM MOPS, and 1 mM EDTA for 2.0 hours at 3.5
V/cm gel in 50 mM MOPS, 1 mM EDTA running buffer. RNA was blotted onto Magna
nylon membranes by capillary action with 20 X SSC, immobilized by U.V. crosslinking
and visualized by staining briefly with methylene blue.
Random-primed [32P]probes were made for resolving Vtg I and Vtg II RNA
transcripts. The Vtg I probe was synthesized from a PCR product off of the template
pMMBl using primers ROW 5 and MB 13, resulting in a 639-bp cDNA probe from
nucleotide 4284 to 4923 of the Vtg I cDNA. The Vtg II probe was made from pFhv2a
using primers ROW 19 and ROW 33, yielding a 277-bp probe from nucleotide 4692 to
4969 of the Vtg II cDNA. After random prime labeling, oligonucleotide probes were
separated from non-incorporated [32P]dCTP by size chromatography through Stratagene


16
Induction of vitellogenin synthesis
Male Fundulus heteroclitus were collected from the estuarine creeks adjacent to
the Whitney Laboratory, and were maintained in running seawater tanks under 14L: 10D
photoperiod conditions at 25 2C. Fish were maintained for at least one month before
being used for RNA collections.
In order to increase the proportion of Vtg RNA within the total RNA pool,
vitellogenin synthesis was artificially induced in six males (8-10 g body weight) by two
intraperitoneal injections of estradiol-17/3 (0.01 mg/g body weight) dissolved in 50 /xl
peanut oil (Kanungo et al., 1990). Five control males were sham-injected with peanut
oil alone. The first injection was performed on day 1, the second injection on day 4,
followed by sacrifice and liver dissection on day 8.
Isolation of liver poly A + RNA
Livers from both groups of fish were collected and immediately placed in 0C
guanidinium thiocyanate solution (5M guanidinium thiocyanate, 50 mM Tris-Hcl, 25 mM
EDTA, 8% v/v mercaptoethanol, pH 7.4) and homogenized by one thirty-second
polytron (Brinkman) blast. RNA was isolated by the guandidinium thiocyanate method
according to MacDonald et al. (1987). One gram of liver from estrogen-treated fish
yielded an average of 0.536 mg RNA, with an average O.D. 260/280 ratio of 2.03,
while a gram of liver from sham-treated fish yielded an average of 0.318 mg RNA with
an O.D. 260/280 of 1.87. Total RNA samples were combined into two pools: one


vtg :
vtg ::
VDVPSNirSMNVTTFAVARNIEEPLVERITPLLPTXVLVPIPIRRHTSKI.
VEGVK7IASARLETVAIARDVEGLAAAXVTPWP YSPIVSKNATLNL
* * * ** * * ** *
947
946
Vtg I
vtg n
DPTR MSMLDSSEPME SSDVEPIPEYXFRRFAKKYCAKHIGV
SQMSYYLNDSISASSELLPFSLQRQTGXNXI? X? IVXKMCATTYTY
* ****** ** ** **
990
992
vtg I
Vtg XI
GLKACFXFASQNGASIQDIVLYXLAGSHNFSFSVTPIEGEWERI.2MEVX
GIEGCVDIWSRNATFLRNTPIYAIIGNHSLLVNVTPAAGPSIERISIEVQ
o .*
1040
1042
Vtg I
Vtg II
VGAJCAAEXiVKSINLSEDEETSEGGPVLVXLNKtSSHRNSSSSSSSlSsS
FGEQAAEXILXEVYLNEESEVLSDKNVLMKI.XXILSPGt,XNSTXASSSSS
* *** * ** ** ** *** *****
1090
1092
Vtg I
Vtg II
SSSSSaSSRSSSSSSSSSRSSRlCIDEAARTSSSSSSSSRRSRSSSSSSSS
GSSRSSSSRSSSSSSSSSSSS SSRSSSSSSRSSSSLRRNSK
* * W******* ** ***** ** *
1140
1133
Vtg I
Vtg II
SSSSSSSSSSSSRRSSSSSSSSSSSSSRSSRRVNSTRSSSSSSRTSSASS
MLBLAPgEMITSXRSSSSSSSESSSSSSSS: SSSSSSXTKWQLH
vw**' "w** * * ** ir" * jir-r ****** *
1190
1176
Vtg I
Vtg II
LASFFSDSSSSSSSSORRSkXVMEXFQRLHXKMVASGSSASSVSAIYXEX
SRNF ; tX0HIHQHSVSK-3RUSK--S3ASSF5SI?nCX
1240
1211
Vtg I
Vtg II
XYLGEE- SAWAVILRAVKADKRMVGYQLG FYLD KPMARVQ11 VAN X SSD
TYLSNIVSPWTVLVRAIRADHXNQGYQIAVYYDXLTTRVQIIVANLTED
1289
1261
vtg I
vtg II
SNWRICADAWLSKHKVTTKISWGEQCRKYSTNVTGETGIVSSSPAARLR
DNWRICSDSMMLSHHKVMTRVTWGIGCXQYNTTIVAETGRVEXSPAVRVK
***** * * * * *** ** *
1339
1311
Vtg I
Vtg II
vswerlpstlxrygxmvnkyvp-vkilsolihtxrenstrnisviavats
LA WARL ?T YIRD Y ARR VS R Y1S RVAED NGVNRTXVASX? X2IXLT/A VAN
# * n * * *
1388
1361
Vtg I
Vtg II
EXTIOIITXTPMSSVYNVTMHLPMCIPIDEIXG-LSPFDEVIDKIHFMV
etslnvtlntpxntffxlgwvlpfylpinntaaelqafqgrwmdqvty.ml
* * * V *
1436
1411
Vtg I
Vtg II
S XAAAAE CS FVEDTL YTFNNRS YXNKMP S S CYQVAAQD CTD ELXFMVLLR
TXSAAAECTWEDTWTFNNRKYXTSTPHSCHQVLAQDCTSEIXFIVLLX
1486
1461
Vtg I
Vtg II
KD-SSEQHHINVXISEIDIDMFPXDDNVTVXVNEMEIPPPACLTATQQL?
RDQTAERNEISIXIENIDVDMYPXDNAVWKVNGVEIPLTNLPYQHPTGN
* * ** ** ** *** **** ***
1S3S
1311
vtg I
vtg II
LKIXTXRRGLAVYAPSHGLQEVYFDRXTWRIKVADWMKGXTCGLCGXADG
IQIRQREEGISLHAPSHGLQEVFLSLNXVQVKWDWMRGQTCGLCGXAflG
isas
1561
Vtg I
Vtg II
EIRQE YHTPNGRVAXNSIS F AHSWIL? AESCRD AS ECRLXLESVQLEXQL
3VRQEYSTPNERVSRNATSFAHSWVLPAXSCRDASECYMQL2SVKI.EXQI
1635
1611
Vtg I
Vtg II
TIHGEDSTCFSVEPVPRCLPGCLPVXTTPVTVGFSCLA SDPQT
SLEGESSXCYSVEPVWRCLPGCAPVRTTSVTVGI.PCVSLDSNLNRSOSLS
1678
1661
Vtg I
Vtg II
SVYD RSVD LRQT7QAHLACS CNTKCS
SIYQKSVDVSETAESHLACRCTPQCA
* ** **** *
1704
1637
Figure 3.3--continued


124
females and estrogen-induced males. It was further shown that Chg MRNA was not
indicated in RNA isolated from the ovaries of reproductive females. The F. heteroclitus
Chgs were recognized as homologs to the ZP proteins of mammals by their possession
of a ZP domain. A parsimonious tree analysis of the ZP domains from the three Chgs,
five other fish homologs, and nine other mammalian ZP proteins separated the molecules
into three major subdivisions. Chg 427 was grouped with mouse ZP3 and its homologs.
Chg 500 and Chg 533 were grouped with the mouse ZP1 subdivision, but not
significantly separated from the ZP2 subdivision. We isolated vitelline envelope proteins
(VEPs) from ovarian follicles and obtained their amino acid compositions for comparison
with the predicted compositions of the three Chgs. Although similarities existed between
the Chgs and the VEPs, we are currently awaiting N-terminal sequence analysis data to
provide unambiguous matches, verifying that Chgs are processed into bona fide VEPs.
The sequences of these five liver-derived molecules provide us with a substantial
amount of new information regarding the hepatic contribution to oocyte development.
Not only do we have data describing the primary structure of five important components
of the oocyte, but we have convincing evidence that they originate heterosynthetically in
the liver and are produced under estrogen induction. Although these results provide
exciting opportunities for further study, the flames of our ambition are truly fed by the
estrogen-induced liver library as an excellent tool to aid in elucidation of molecules that
contribute to reproductive processes.


28
4433
ATC
AAT
GTC
AAG
ATT
tct
GAG
ATC
GAT
ATT
GAC
ATG
TTT
CCA
AAG
GAC
GAC
AAC
I
N
V
X
I
s
E
I
D
I
D
M
F
P
X
D
D
N
4537
GTC
ACT
GTG
AAG
GTC
AAC
GAA
ATG
GAA
ATA
CCC
CCA
CCA
GCC
TGC
CTT
ACC
GCC
V
T
V
K
V
N
E
M
E
X
p
P
P
A
C
L
T
A
4591
ACC
CAA
CAG
Q>ryv
CCA
TTG
AAG
ATC
AAG
ACA
AAG
CGG
AGA
GGA
CTT
GCT
GTC
TAT
T
Q
Q
L
P
L
K
I
K
T
X
R
R
G
L
A
V
Y
4645
GCA
ccc
AGC
CAC
GGT
CTC
CAA
GAA
GTC
TAC
TTT
GAC
AGG
AAG
ACA
TGG
AGG
ATC
A
p
S
H
G
L
Q
E
V
Y
F
D
R
X
T
W
R
I
4699
AAA
GTT
GCT
GAC
TGG
ATG
AAA
GGA
AAG
ACC
TGT
GGA
CTC
TGT
GGA
AAG
GCT
GAT
K
V
A

W
M
K
G
K
T
C
G
L
C
G
X
A
D
4753
GGA
GAG
ATC
AGA
CAG
GAG
TAC
CAC
ACT
CCC
AAC
GGA
CGC
GTG
GCC
AAG
AAC
TCG
G
E
I
R
<2
E
Y
H
T
P
N
G
R
V
A
X
M
S
4807
ATC
AGC
r^rr*rn
GCT
CAC
TCC
TGG
ATT
CTT
CCT
GCT
GAA
AGC
TGC
AGG
GAT
GCA
TCT
I
S
F
A
H
S
W
I
L
P
A
E
S
C
R

A
S
4361
GAG
TGC
CGT
CTG
AAA
CTT
GAA
TCT
GTG
CAG
CTG
GAG
AAA
CAG
TTG
ACC
ATC
CAC
E
C
R
L
K
L
S
S
V
Q
L
E
X
Q
L
T
I
H
4915
GGT
GAG
GAC
TCC
ACA
TGC
TTC
TCA
GTT
GAG
CCT
GTA
CCT
CGT
TGT
CTG
CCC
GGT
G
E
D
S
T
C
F
S
V
E
P
V
P
R
c
L
p
G
4969
TGC
TTG
CCT
GTC
AAG
ACC
ACA
CCT
GTC
ACT
GTT
GGT
TTC
AGC
TGC
CTG
GCA
TCT
C
L
P
V
K
T
T
P
V
T
V
G
F
S
C
L
A
S
5023
GAT
CCT
CAG
ACC
AGT
GTC
TAT
GAC
AGA
AGT
GTG
GAT
CTA
AGA
CAA
ACT
ACC
CAG
D
P
Q
T
S
V
Y
D
R
S
V
0
L
R
Q
T
T
2
5077
GCT
CAC
CTG
GCT
TGC
AGC
TGC
AAC
ACC
AAG
TGC
TCT
TAA
ACA
TAA
GAT
TTC
CTT
A
H
L
A
C
S
C
N
T
K
C
s
-
5131
GAA
GTC
ACT
ACT
ATG
TGT
AAG
TTT
TAT
CTG
TAA
CAA
TAA
ATA
AAC
TGC
ATC
TGA
5185
AAA
TAA
AAA
AAA
AA
Figure 2.2--continued


gaacttttcagatcacttgtgtttgtgaagcc 32
102
B)
ATGATGATGAAGTGGACTGTCTTTTGCGTTGTGGCGCTGGCTTTGCTTGGCAGCTTCTGTGATGCT 98
MMMKWTVFCVVALALLGSPCDA22
CAGGGGTACGCGAAACCTGGTAAGCCATCAAAACCCCAATCACCACCTACGCAAAACCAACAGCAA 164
QGYAKPGKPSKPQSPPTQNQQQ44
TTGCAGACATTTGAGAAAGAGCTCACCTGGAAGTACCCCGACGATCCCCAGCCAGACCCCAAGCCT 230
LQTFEKELTWKYPDDPQPDPKP66
AATGTGCCATTTGAGTTGAGATACCCTGTTCCTGCTGCAACCGTTGCTGTTGAGTGCAGAGAGAGC 296
NVPFELRYPVPAATVAVECRES88
ATAGCTCACGTGGAGGTCAAGAAAGACATGTTTGGCACCGGCCAGCCGATCAATCCAAATGACCTC 362
IAHVEVKKDMFGTGQPINPNDL 110
ACCCTGGGTAACTGTGCGCCTGTTGGAGAGGATAGTGCCGCTCAAGTGTTGATTTATGAAGCTGAA 428
TLGNCAPVGEDSAAQVLIYEAE 132
CTGCATCAATGCGGAAGCCAGCTGATGATGACAAATGATGCTCTCGTCTACACCTTCGTTTTGAAC 494
LHQCGSQLMMTNDALVYTFVLN 154
TATAACCCTACGCCTTTGGGATCGGTTCCTGTTGTGAGAACCTCCCAAGCTGCTGTGATCGTGGAA 560
YNPTPLGSVPVVRTSQAAVIVE 176
TGCCACTACCCAAGGAAGCACAATGTGAGCAGCCTTCCTCTGGATCCCCTTTGGGTCCCATTCTCT 626
CHYPRKHNVSSLPLDPLWVPFS 198
GCAGTTAAGATGGCTGAGGAGTTCCTGTACTTCACTATGAAACTCATGACTGATGACTGGATGTAC 692
AVKMAEEFLYFTMKLMTDDWMY 220
CAGAGGCCAAGCTACCAGTS.TTTCCTGGGAGACCTGATCCGTATAGAGGTTACTGTCAAGCAATAC 758
QRPSYQYFLGDLIRIEVTVKQY 242
TTCCATGTACCCCTGCGTGTTTACGTGGACAGATGTGTGGCAACCCTCTCTCCTGATGTAACCTCA 824
FHVPLRVYVDRCVATLSPDVTS 264
AGCCCCAACTATGCCTTCATTGATAACTTTGGGTGTTTGATTGACGCCAGAATCACAGGCTCTGAC 890
SPNYAFIDNFGCLIDARITGSD 286
TCAAAGTTCATGGCTCGCACCCAAGAGAACCACCTTCAGTTCCAGCTGGAGGCCTTCAGGTTCCAG 1956
SKFMARTQENHLQFQLEAFRFQ 309
AATTCTGACAGTGGAGTGATCTACATCACCTGCTACTTGAAGGCAACGTCTACTAGCCAGGCCATA 1022
NSDSGVIYITCYLKATSTSQAI 330
GACAGCCAGCACAGAGCTTGTTCCTACACTGGCGGATGGAGGGAGGCCAGTGGAGTTGATGGAGCT 1088
DSQHRACSYTGGWREASGVDGA 352
TGTGGTTCTTGTGAGACCAACGTGACGCCGTACACCGCTCCAGCAGTTACATTCGCTTCACCACCT 1154
CGSCETNVTPYTAPAVTFASPP 374
GTCGTTGTTACTGATGGTGGTGGAGTAACGCTTCCAGCTCCAGGCAGTCCAAAAGTCCCTTATAAT 1220
VVVTDGGGVTLPAPGSPKVPYN 396
CCGAGGAAAGTCCGTGACGTCACCCAAGCCGAAATTTTGGAATGGGAAGGCGTTGTCTCTCTGGGC 1286
PRKVRDVTQAE ILEWEGVVSLG 418
CCCATCCCCATCATGGAGAAGAAACTCTGAaaaacagaagtgtaacatgatattccgccgtagcca 1352
PIPIMEKKL- 427
tgaacaccataataaaaagtatcattggttcatatcgctgtctatgttatgcctatgtctcatggt 1418
agattttcttaaacaagtaacaaacccccacttagtctcttaaatctgcttaaaattttaaatatt 1484
gacaaatttccaaaaaattgtagaggtctttttttaggggggagggataaatgaaggaaaacttgt 1550
cttagattcccttttatgtaatggtaaggcagtgtgtggacccccatgtgtccagcaccataatct 1616
gtaaccctecttttcatgaaaataaaattcgcaactataaaaaaaaaaaaaaaaaa 1672
Figure 5.2-continued


92
Chq 500
pChgla
ROW 45
ROW 52
pChglb
Chq 427
PChg2a RW55
ROW 65
pChg2b
Chq 553
pChg3a
ROW 45
200 bp
GL1
pChg3b
Figure 5.1 Strategy for cloning Chg 500, 427 and 553 cDNAs. Boxes indicate
relative sizes of contiguous cDNA sequences coding for Chgs 500, 427,
and 553. Thin black lines represent individual cDNA isolates obtained by
anchored PCR or RACE and inserted into pGem-T. Arrows indicate
gene-specific primers that were used in initial amplifications of individual
clones. The legend indicates relative length of 200 bp.


90
In this paper we present the predicted primary structure of three proteins that
share identity with mammalian ZP proteins. Furthermore, we have isolated cDNAs
encoding these sequences from a liver library rather than an ovarian library, followed by
northern analyses revealing liver rather than ovarian transcripts. Lastly, the amino acid
compositions predicted from our cDNAs are similar to the composition of VEPs isolated
from F. heteroclitus follicles. Therefore, we conclude that the proteins encoded by these
cDNAs are synthesized by the liver, transported to the ovary, and incorporated into the
vitelline envelope. We further suggest that as major constituents of the vitelline
envelope, these proteins eventually contribute to the structure of the hardened chorion,
where they remain until finally degraded by embryonic hatching enzymes. We designate
these cDNAs and the proteins that they encode as "choriogenins" (Chgs) to emphasize
their role as proteins of the vitelline envelope and chorion, yet to underscore their site
of synthesis as being extra-ovarian, and thus different from that of the mammalian ZP
proteins. Although the teleostean chorion and mammalian zona pellucida have different
appearances, functions, and, as this study verifies, origins of synthesis, we provide
evidence that the constituent molecules appear to have evolved from a set of common
ancestral proteins
Materials and Methods
Reagents
Estradiol-17/3 was obtained from Sigma Chemical Co. (St. Louis, MO).
Radioisotopes, [a-32P]dCTP and [a-35S]dATP, were purchased from New England


8
Vtg IX 4650 CTGGATGAGAGGCCAGACGTG7GGGCTCTGCGGAAAGGCCGACGGGGAAGTCAGACAGG 4693
***** ******** ** ***
TGTGGICTCTGCGGIAAIAACGA
ROW 19 (degenerara) C G T CG T
Chg 500
ROW 45
913 TCCTGGACCTCTGCGTGTGGAGCTCAGGCTTGGGAATGGAGAGTG7TCTGTCAAGGGTT 975
******** ** **
GAG CT CAGTCTGTACACTGCT
Chg 427
ROW 55
669 CAGCCTTCCTCTGGATCCCCTTTGGGTCCCATTCTCTGCAGTTAAGATGGCTGAGGAGT 693
ft** ***
CATTCTGAAACTTGAAGACCC
Chg 553
ROW 45
320 C'rCATTGTTGGGAGGAGGTCAAGGCIGTACACATGTTGACCCCAATTCACTTTTTGCCA 373
*** *** ft******* *
GAGCTCA-GTCTGTACACTGCT
Figure 1.3 Four accounts of fortuitous annealing that resulted in the eventual
isolation of cDNAs coding for Vtg II, Chg 500, Chg 427, and Chg
553. Vtg II was discovered using the degenerate primer ROW 19.
Chg 500 and Chg 553 were discovered using ROW 45, originally
designed for annealing to Vtg II. Chg 427 was isolated using
ROW 55, also designed to anneal to Vtg II.


15
oocytes. Although the polyserine domain is indeed a polymorphic region, a conserved
genetic pattern (Byrne et al., 1984, 1989) persists in all of the vertebrates thus far
examined: TCX repeats at the 5 end and a larger group of AGY repeats towards the 3
end, suggesting an ancient origin of the linkage between these two clusters of
trinucleotide repeats.
Materials and Methods
Chemicals
Estradiol-17/3 was obtained from Sigma Chemical Co. (St. Louis, MO).
Radioisotopes, [a-32P] dCTP and [a-35S] dATP, were purchased from New England
Nuclear (Boston MA). Lambda gtlO vector and cDNA synthesis reagents were obtained
from Promega (Madison, WI). The subcloning plasmid pGem-3Z was purchased from
Promega, pT7BLUE from Novagen (Madison, WI), and pCRIOOO from Invitrogen (San
Diego, CA). All sequencing gels were cast using Sequagel-8 (National Diagnostics,
Atlanta) polyacrylamide reagents. Amplification reactions were performed using
Thermophilus aquaticus DNA polymerase (Promega). Sequenase version 2.0 DNA
polymerase and dideoxy sequencing reagents were obtained from US Biochemicals
(Cleveland, OH). Reagents for random-primed labeling of probes were purchased from
Pharmacia (Piscataway, NJ). Both Nytran nylon and S&S NC nitrocellulose transfer
membranes were purchased from Schleicher and Schuell (Keene, NH). Amino acid N-
terminal sequencing and synthesis of oligonucleotide primers were performed by the
University of Florida Interdisciplinary Center for Biotechnology Research core facility.


118
Figure 5.8 Parsimonious tree analysis of ZP domains from seventeen vertebrate ZP
homologs. The unrooted tree was obtained by running 100 bootstrap
replicates of a heuristic search (PAUP 3.1; Swofford, 1993) through a
ClustalV alignment, containing only the ZP domains of selected proteins,
and drawn according to the format of Fitch parsimony program. Three
divisions of ZP homologs resulted, each designated by a separate mouse
ZP protein: ZP1 division, ZP2 division, and ZP3 division. Three piscine
proteins, including Chg 500 and 553 were grouped within the ZP1
division, while five piscine proteins, including Chg 427 were grouped
within the ZP3 division. Bootstrap percentage values are indicated
adjacent to appropriate nodes.


88
teleost oocyte during its development, while the term "chorion" referred to the
structurally and perhaps chemically transformed vitelline envelope that surrounds the
ovulated egg, separates from the egg at the time of fertilization, and encloses the embryo
until hatching. Implicit in these definitions is the assumption that the proteinaceous
structure of the vitelline envelope comprises a substantial component of the chorion.
The structure of the teleostean vitelline envelope has been well documented in
several cyprinodont species (Yamamoto, 1963; Fliigel, 1967; Dumont and Brummett,
1980) as well as in many other teleosts (reviewed by Dumont and Brummett, 1985;
Selman and Wallace, 1989). Early biochemical characterizations of the vitelline envelope
and chorion concentrated on the formation of the vitelline envelope during oocyte
development (Chaudry, 1956; Yamamoto, 1963; Flegler, 1977; Tesoriero, 1977), as well
as the breakdown of the chorion by the proteolytic enzymes of the hatching embryo
(Yamamoto and Yamagami, 1975; Kaighn, 1964, Hagenmaier, 1985). In earlier works
it had been assumed, but not proven that the major vitelline envelope proteins (VEPs)
were synthesized by the ovarian follicle the site of synthesis residing in either the
oocyte or surrounding follicle cells (Anderson, 1967). More recent investigations
targeting VEP synthesis include studies by Tesoriero (1978) using [3H]proline
incorporation, and by Begovac and Wallace (1989) in which incorporation of
[35S]methionine combined with immunohistochemistry provided evidence that at least one
of the VEPs from the pipefish, Syngnathus scovelli, originated from within the ovarian
follicle.
A new direction towards understanding vitelline envelope formation was launched


69
the fine-tuned expression of Vtg that must occur in a sequentially spawning
animal. Besides being used as tools to specifically investigate F. heteroclitus
reproduction, the Vtg I and II cDNAs represent valuable bio-markers for assaying
the reproductive health of naturally occurring fish. As examples of mRNAs and
proteins that are normally induced by estrogens, Vtg I and II will be particularly
valuable in testing for the estrogenic effects of environmental contaminants such
as polychlorinated biphenyls (Bergeron et al. 1994; Guillette et al. 1994).


36
Fundillos NVTGETGIVSSSPAARLRVSWERLPSTLKRYGK-MVNKYVP-VKILSDLIHTKRENSTRN 1379
Gallus STELVTGRFAGHPAAQVKLEWPKVPSNVRSWE-WFYEFVPGAAFMLGFSERMDKNPSRQ 1514
Xenopus NMKAETGNFGNQPALRVTANWPKIPSKWKSTGK-WGEYVPGAMYMMGFQGEYKRNSQRQ 14 6 9
Acipenser AVSAVTGRLASHPSLQIKAKWSRIPRAAKQTQN- ILAEYVPGAAFMLGFSQKEQRNPSKQ 14 24
Ichihvomyzon MLEASTGNLQSHPAARVDIKWGRLPSSLQRAKNALLENKAPVIASKLEMEIMPKKNQKHQ 14 94
Oncorhyncus FITAETGLVGPSPAVRLLDKLPKVPKAVWRYVRIVSEFIPGHIPYYLADLVPMQKDKNSE 113
Fundulus IS VIAVATS E KTID11TKTPMSS VYNVTMHLPMCIPIDE--I KGLSP--FDEVIDKI 14 32
Gallus ARM WALTS PRTCD WVKLPDI1LYQKAVRLPLS LPVG P--RIPASELQPPIW NVFAEA 1571
Xenopus VKLVFALSSPRTCDWIRI PR LTVYYRALRLPVPIPVGH -HAKENVLQTPTW-NIFAEA 1526
Acipenser FK11LAVTSPNTIDTLIKAPKITLFKQAVQIPVQIPMEP--SDAER- -RSPGLASIMNEI 1480
Ichlhyomyzon VSVILAAMTPRRMNIIVKLPKVTYFQQGILLPFTFPSPRFWDRPEGSQSDSLPAQIASAF 1554
Oncorhynchus KQFTWATSERTLDVILKTPKMTLTKTGVNIPCSLPFESMTDLSPFDDNIVNKIHYL- -F 171
Fundulus HFMVSKAAAAECSFVEDTLYTFMNRSYKNKMPSSCYQVAAQDCTDELKFMVLLRK- -DSS 14 90
Gallus PSAVLENLKARCSVSYNKIKTFNEVKFNYSMPANCYHILVQDCSSELKFLVMMKSAGEAT 1631
Xenopus PKLIMDSIQGECKVAQDQITTFNGVDLASALPENCYNVLAQDCS PEMKFMVLMRNS KESP 1586
Acipenser PFLIEEATKSKCVAQENKFITFDGVKFSYQMPGGCYHILAQDCRSKVRFMVMLKQASMSK 154 0
Ichlhyomyzon SGIVQDPVASACELNEQSLTTFNGAFFNYDMPESCYHVLAQECSSRPPFIVLIKLDSERR 1614
Oncorhynchus S EVNAVKCSMVRDTLTTFNNKKYKINMPLSCYQVLAQDCTTELKFMVSAEEGSVHL 227
Fundulus EQHHIHVKTSEIDTDMF- PKDDNVTVKWEMEIPPPA- CLTATQQLPLKIKTKRRGLAVY 154 8
Gallus NI.KAINIKIGSHEIDM-H PVNGQVKLLVDGAES PTANIS LIS AGAS LWIHNENQGF ALA 168 9
Xenopus NHKDINVKLGEYDXDMYYSA-DAFKMKINNLBVSEEHLPYKSFNYPTVEIKKKGNGVSLS 1645
Acipenser NLRAVHAKIYNKDXDILPTTKGSVRLLINNNEIPLSQLPFTD-SSGNIHKRADEGVSVS 1599
Ichlhyomyzon I- -SLELQLDDKKVKIVSRND IRVDfJEKVPLRRLSQKN QYGFLVLDAGVHLL 1664
Oncorhynchus NKTTSNVKISDIBVpLYTQDHGVIVKVNEMEVSNEQLPYKDPSG-SIKIDRKKGEGVSLY 28 6
Fundulus APSHGLQEVYFDRKTWRIKVADWMKGKTCGLCGKADGEIRQEYHTPNGRVAKNSISFAHS 16 08
Gallus APGHGIDKLYFDGKTITIQVPLWMAGKTCGICGKYDAECEQEYRMPNGYLAKNAVSFGHS 1749
Xenopus ASEYGIDSLDYDGLTFKFRPTIWMKGKTCGICGHNDDESEKELQMPDGSVAKDQMRFIHS 1705
Acipenser AQQYGLESLYFDGKTVQVKVTSEMRGKTCGLCGHNDGERRKEFRMPDGRQARGP 1653
Ichlhyomyzon LKYKDL-RVSFNSSSVQVWVPSSLKGQTCGLCGRNDDELVTEMRMPNLEVAKDFTSFAHS 1723
Oncorhynchus APSHGLQKVYFDKYSWKIKWDWMKGQTCGLCGKADGENRQEYRTPSGRLTKSSVSFAHS 34 6
Oreochromis FFFSLVFHAVS 11
Fundulus WILPAESCRDASECRLKLESVQLEKQLTIHGEDSTCFSVEPVPRCLPGCLPVKTTPVTVG 16 6 8
Gallus WILEEAPCRGA- -CKLHRSFVKLEKTVQLAGVDSKCYSTEPVLRCAKGCSATKTTPVTVG 1807
Xenopus WILPAESCSEG--CNLKHTLVKLEKAIATDGAKAKCYSVQPVLRCAKGCSPVKTVEVSTG 1763
Acipenser SVSPTPG 1660
Ichlhyomyzon WIAPDETCGGACALSRQ--TWKESTSVISGSRENCYSTEPIMRCPATCSASRSVPVSVA 1781
Oncorhynchus WVLPSDRG-DASEG-LM-- KLEKQVIVDD-REEK-CYSVEPVLRCLPGCSPVRTTPITIG 400
Oreochromis KKLQNHYSLRLLKEKVKS ELMVPILKVSEPNATLLSPCCSACPACIPVRTTTVNVG 67
Fundulus -FSCLASDPQ TSVYD RSVDLRQTTQAHLACSCNTK- CS 1704
Gallus -FHCLPADSANSLTDKQ-MKYDQKSEDMQDTVDAHTTCSCENEECST 1852
Xenopus FHCLPSDVSLDLPEGQ IRLE- KSEDFSEKVEAI1TACSGETSPCAA 1807
Acipenser --LGLEKTATEAASFCVIM 1677
Ichlhyomyzon -MHCLPAESEAISLAMSEGRPFSLSGKSEDLVTEMEAHVSCVA 1823
Oncorhynchus - HCLPFDSNLNRSEGLSSIY EKSVDLMF.KAEAHVACRCSEQ- CM 442
Oreochromis FYGCLPSDTT VDRSGLSSFF EKSIDLRDTAEAHLACRCTPQ-CA 110
Figure 2.4--continued


37
the single best tree F. heteroclitus Vtg was placed on an independent branch,
intermediate to the positions of sturgeon and lamprey Vtgs. The Ichthyomyzon sequence
was the vertebrate Vtg determined to lie furthest from the reference sequence, thereby
placing it nearest to the outgroup. One of the more significant relationships provided by
the tree is indicated by the bootstrapping values at the Acipenser branch (in parentheses,
Fig. 2.5): through 100 bootstrap replicates, sturgeon Vtg was partitioned with the two
tretrapod Vtgs 95% of the time, substantially more than the Vtgs of either F. heteroclitus
(67%) or Ichthyomyzon (31% not shown).
Polvserine Domain
We have designated a polyserine domain from each of the aligned Vtgs
(underscored with a triple dotted line in Fig. 2.4; see Materials and Methods) and
compared them in regard to size, relative serine composition and serine codon usage
(Fig. 2.6). Of the Vtgs listed here, F. heteroclitus Vtg contains the smallest polyserine
domain (171 a.a.); it also contains the highest relative serine composition (57.6%). We
compared the serine codon usage from each of the domains and found a consistent
pattern: TCX repeats are more prevalent at the 5end while AGY codons are more
prevalent at the 3 end. Finally, of the six possible serine codons, AGC was invariably
the dominant codon in all five vertebrate polyserine domains.
Discussion
We present the first complete teleost Vtg cDNA sequence along with its translated


93
resembled mammalian ZP proteins. The clones containing these initial cDNAs were used
as probes and to design primers that would target additional cDNAs in order to complete
the Chg coding regions (Fig. 5.1).
The first Chg cDNA, pChgla, was isolated by anchored PCR with a Vtg II-
targeted reverse primer, ROW 45, and the XgtlO vector primer, NEB 1231. To retrieve
a cDNA containing the poly-A tail, primer ROW 52 was designed from the
3 side of pChgla. An overlapping clone (pChglb), containing a poly-A tail, was
isolated by anchored PCR, completing the translated region of Chg 500.
The second choriogenin cDNA, pChg2a, was isolated by anchored PCR with
anchor primer NEB 1232 and reverse primer ROW 55, also designed to target Vtg II
sequence. Blast analysis on the sequence of pChg2a revealed that it shared 67 % identity
with the medaka L-SF protein (Murata et al. 1995). The primer ROW 65 was then
designed from the 3 region of pChg2a to retrieve an overlapping cDNA that contained
a poly-A tail. The sequence from the resulting clone (pChg2b) completed the translated
portion of the second Chg 427.
The third Chg clone, (pChg3a), was also isolated with ROW 45, along with the
initial pChgla, but it remained unrecognized as a novel clone until further review of the
sequences. An overlapping clone (pChg3b) containing the poly-A tail was isolated by
anchored PCR with the forward primer GL2. A third clone containing a short segment
5 to pChg3a including the initial methionine codon was isolated using a rapid
amplification of cDNA ends protocol (RACE) (Frohman, 1992) with reverse primer
GL1, and 3 fig total liver RNA as template.


75
Mw, kDa
21.5^
LFESLVDSDKW.
YEFSDELLQTPL.
KYxAKHIGVGLK.
13 kDa
LFESLVDSDCW. .
YEFSDELLQTPL...
Figure 4.2 Endo LysC digestion products of YP 125 and YP 105. After partial
digestion with Endo LysC, polypeptide fragments were electroblotted onto
a PVDF membrane and silver stained. Positions of the 13 kDa bands
(presumed to be identical) from each digestion are indicated. Molecular
weight standards (kDa) are shown on the left.


CHAPTER 2
FUNDULUS HETER0CL1TUS VITELLOGENIN:
THE DEDUCED PRIMARY STRUCTURE OF A PISCINE PRECURSOR TO
NON-CRYSTALLINE, LIQUID-PHASE YOLK PROTEINS
Introduction
Vitellogenin (Vtg) is a large phosphoglycoprotein ( 200 Kda) used by most
oviparous animals as a maternally derived yolk precursor (Pan et al., 1969; Kunkel and
Nordin, 1985; Wallace, 1985; Selman and Wallace 1989). It is synthesized by either the
liver (vertebrates), fat body (insects), or intestine (nematodes) under hormonal induction
and transported to growing oocytes via the blood (Flickinger and Rounds, 1956; Wallace
and Jared, 1969). Vtg is incorporated into oocytes by receptor-mediated endocytosis
(Opresko et al., 1980; Opresko and Wiley, 1987; Shen et al., 1993) and is stored for
later use by the developing embryo (Flickinger, 1960; Yamagami, 1960; Karasaki,
1963b; Selman and Pawsey, 1965; Murakami et al., 1990). Once inside the oocyte, Vtg
is processed into smaller yolk proteins consisting of lipovitellins (Lvl and Lv2),
phosvitins (Pv), and phosvettes, that may in turn be degraded into even smaller cleavage
products (Flickinger and Rounds, 1956; Taborsky, 1967; Wallace and Selman, 1985;
Gerber-Huber et al., 1987; Greeley et al., 1986).
11


67
separated into two general clusters, those of the Vtg II polyserine domain are
randomly interspersed. This arrangement of polyserine codons again confirms
the observations noted by Byrne et al., (1989) that the polyserine, or phosvitin,
domain is an independently evolving domain within the Vtg gene, showing more
variability than its flanking lipovitellin regions.
The predicted post-translational modifications of Vtg II are in agreement
with expectations for a lipophosphoglycoprotein. Although 45 phosphorylation
sites may appear to be high, we expect an even higher amount of phosphorylation
than is predicted. Seven of the 45 predicted phosphorylation sites occur within
the polyserine domain (all protein kinase C sites), however, from previously
published accounts, it is likely that every serine residue in this domain is
phosphorylated, resulting in a very hydrophilic domain with a highly negative
charge. Phosvitin yolk proteins have been described as possessing the highest
amount of phosphorylation of any known proteins. Unfortunately, the hepatic
Vtg kinase responsible for phosphorylating the extensive polyserine domains of
Vtg has not yet been isolated or characterized, so that an algorithm predicting its
target sites is not yet available.
Considering the ratio of expression of Vtg I and Vtg II, our data suggest
that Vtg I is the major yolk protein precursor. Vtg II mRNA is present in the
liver of spawning females at ratio of 1:10 with respect to Vtg I RNA, as
evidenced by northern blots. In SDS PAGE analysis, YP 69, which was mapped
to Vtg II, is hardly discemable (not shown here) when compared to the Vtg I-


91
Nuclear (Boston, MA). Lambda gtlO vector and cDNA synthesis reagents were obtained
from Promega (Madison, WI). The subcloning plasmid pGem-T was a product of
Promega. All "ROW" oligonucleotide primers were synthesized by the University of
Florida Interdisciplinary Center for Biotechnology Research (ICBR) oligonucleotide core
facility, while primers labelled "GL" were synthesized by Bio-Synthesis (Freindswood,
TX). Sequenase version 2.0 DNA polymerase and dideoxy sequencing reagents were
obtained from US Biochemicals (Cleveland, OH). In-house sequencing gels were cast
using Sequagel-8 (National Diagnostics, Atlanta) polyacrylamide reagents. Some cDNA
sequences, especially through repeating regions or when verifications were needed, were
performed by The University of Florida ICBR DNA Sequencing Core. Amplification
reactions were performed using a 1:50 mixture of cloned pfu DNA polymerase and
Thermophilus aquaticus DNA polymerase (Stratagene and Promega, respectively).
Reagents for random-primed labeling of probes were purchased from Pharmacia
(Piscataway, NJ). Magna nylon and PVDF transfer membranes were obtained from MSI
(Westboro, MA) and Millipore Corp. (Bedford, MA), respectively.
Cloning Strategy
A liver cDNA library was constructed from poly A+-RNA pooled from five F.
heteroclitus males that had been treated with two injections of estradiol-17)8, as
previously described (LaFleur et al., 1995). While screening the Xgt 10 library for Vtg
cDNAs using anchored PCR, we isolated several non-target cDNAs. Three of these non-
Vtg cDNAs were revealed by BLAST analysis to code for protein sequences that


74
Oocyte Egg
CO
3
O CO
o c
q3 B
o o
r- -
Q_
3 P
T3 ^
C
l
YP 29-
-t yp 83
-r YP 77
^ YP 69
-< YP 39
YP 20^
Figure 4.1 Major yolk proteins isolated from oocytes and eggs of Fundulus
heteroclitus. The major yolk proteins shown here were resolved by an
SDS-PAGE gradient gel (7%-20%) enabling the resolution of a wide
range of proteins ranging from 125 kDa to 20 kDa. For N-terminal
sequencing, however, straight gels were used at various acrylamide
concentrations allowing optimal resolution of yolk proteins at specific size
classes. Yolk proteins that were isolated for N-terminal sequencing are
indicated with our designated labels. Note YP 125 appears as a robust
band, when isolated from pre-maturational oocytes, but is hardly visible
in yolk isolated from ovulated eggs. (Photo courtesy of R. McPherson,
Clarion University)


95
the Entrez document retrieval system, Release 20.0, available from NCBI (NIH,
Bethesda, MD). Sequences referred to in this paper include the flounder
Pseudopleuronectes americanus ZP or wf (gi: 425355; Lyons et al., 1993); medaka, O.
latipes L-SF (gi: 563774; Murata et al., 1995); goldfish, Carrasius auratus ZP3
(gi:763073; unpublished); carp, Cyprinus carpi ZP3i (gi:763078; unpublished) and
ZP3ii (gi:763080; unpublished); mouse, Mus musculus ZP1 (gi: 972946; Epifano et al.,
1995), ZP2 (gi: 202460; Liang et al., 1990), and ZP3 (gi: 141726; Ringuette et al.,
1988); human ZP3A (gi: 141724; Chamberlin and Dean, 1990), ZPB (gi: 458279; Harris
et al., 1994), and ZP2 (gi: 466206; Liang and Dean, 1993); cat, Felis catus ZPA
(gi:458269), ZPB (gi:458271), and ZPC (gi: 458273; Harris et al., 1994).
Northern Blot Analyses
Male and female F. heteroclitus were collected from the estuarine creeks adjacent
to the Whitney Laboratory, and were maintained in running seawater tanks under
14L:10D photoperiod conditions at 25 + 2C. After approximately two weeks in
captivity, fish began spawning in laboratory tanks on a 14-day cycle (Hsiao et al., 1994).
By monitoring amounts of eggs spawned each day, we were able to calculate the 14-day
cycle of separate tanks and thus predict what phase of the spawning cycle individual fish
were in before sacrifice (Hsiao et al., 1996). In this paper two northern blots were
performed using a female fish that was predicted to be in a pre-maturational phase, four
days prior to spawning. Fish were maintained for at least one month before being used
for RNA collections.


10
impressed by the remarkable resource proven to lie within the estrogen-induced liver
library that was used to isolate these cDNAs. Rather than being the means to an end,
the library has rather been venerated as possibly the most important attribute of the
project. We expect that other estrogen-induced liver products can be easily isolated from
it, and modifications of this strategy can be used in the future to investigate other
inductive hormone effects on other tissues.


Figure 5.3
Alignments of Chg 500, Chg 553 and the flounder ZP protein.
A) A ClustalV (Higgins et al. 1992) alignment including predicted
amino acid sequences of Chg 500, Chg 553, and a flounder ZP
protein (Lyons et al. 1992). A conserved core region sharing
sequence identity with other ZP proteins and designated as a "ZP
domain" (Bork and Sander, 1992) is denoted by dark line. This
is the core domain used for drawing the tree shown in Figure 5.8.
B) A ClustalV alignment modified by eye of the proline-glutamine
rich repeating region from Chg 500 and the flounder ZP protein.
A Pro-Gln-X triplet is strongly conserved throughout the region.


45
persistence of a liquid phase yolk in F. heteroclitus oocytes is that the high proteolytic
activity documented by Greeley et al. (1986) prevents the recombination of Pv and the
Lvs into their usual insoluble particles. By obtaining more examples of Vtg protein
structure from other liquid phase yolk producers, a more substantial and, hopefully,
causal difference between soluble and non-soluble yolk will materialize.
In conclusion, the F. heteroclitus Vtg cDNA along with its amino acid translation
represents the first complete Vtg sequence documented from a teleost fish. The predicted
primary structure suggests to us that a heightened proportion of phosphoserine in the
polyserine domain endows the F. heteroclitus Pv yolk proteins with a higher solubility
preventing the formation of non-soluble yolk particles as is seen in many other
vertebrates. Knowledge of the complete primary structure of F. heteroclitus Vtg
provides us with useful information for mapping the extensive proteolytic processing of
native Vtg into its respective yolk proteins. We hope that this sequence will aid
investigators of other vertebrate Vtgs by providing a piscine model for molecular probes
and antibodies. Finally, we have provided yet another example of the evolutionary
independence of Pv within the Vtg gene, where the codon cluster organization is
preserved, yet the size of the serine clusters and intervening regions remains quite
unpredictable.


CHAPTER 1
GENERAL INTRODUCTION
The Demands of the Germ Cell
Reflecting on reproductive strategies of vertebrates, I am reminded of a once
familiar phrase used by an automobile repair shop: You can pay me now...or you can
pay me later." This is a fitting slogan, I think, to describe two alternative relationships
between germ cells and somatic cells, as manifested by different vertebrate groups.
Although the germ cells of all vertebrates are bound to receive an investment from their
associated somatic cells, this investment can be delivered either sooner or later according
to the specific developmental programs. As adults, ourselves, we may consider the
investment made by mothers to their young as an opportunity that is chosen by the
mother, voluntarily. However, this point of view has been described by some as "adult
chauvinism," biased toward the attitudes and experiences of the adult (Wallace, 1983).
An alternative view would be that the mother, or the somatic cells, are essentially held
captive by the germ cells, and (if healthy) have no choice but to respond when called
upon for support. As an illustration, consider the physiological state of the mummichog,
Fundulus heteroclitus (Fig 1.1). When the days of winter begin to grow long, and the
water temperature rises, the female mummichog does not have much say in the matter,
but her ovary begins to grow dramatically, mainly by the incorporation and storage of
1


22
bootstrap analysis of 100 replicates in a branch-and-bound search. C. elegans Vtg 5 was
defined as the outgroup and chicken Vtg was designated as the reference sequence.
Accession codes of sequences used for alignments are as follows: Chicken,
Gallus domesticus Vtg II, EMBL:X13607; Xenopus laevis Vtg A2, GB:M18061; silver
lamprey, lchthyomyzon unicuspis Vtg, GB:M88749; white sturgeon, Acipenser
transmontanus, partial Vtg, GB:U00455; rainbow trout, Oncorhynchus mykiss partial
Vtg GB:M27651; tilapia, Oreochromis aureus partial Vtg, number not available (Ding
et al., 1990); boll weevil, Anthonomus granis Vtg, GB:M72980; nematode,
Caenorhabditis elegans Vtg 5, EMBL:X56213; mosquito, Aedes aegypti Vtg,
GB:U02548; and finally, our own mummichog, Fundulus heteroclitus Vtg, GB:U07055.
Results
Cloning
A summary of our cloning strategy is presented in Figure 2.1 Three restriction
products of two MB6-positive lambda clones (#21 and #5) were subcloned into PGEM
3Z (PMMB1, PMMB6, and PMMB9). Two smaller clones pGL8 and pGL5 were
amplified by PCR directly from the cDNA library. The five subclones were sequenced
in both directions for a final overlapping cDNA sequence of 5198 bp. The overlapping
cDNA sequence contained an open reading frame of 5112 bp and a poly-A tail of
undetermined length beginning 11 nucleotides after a poly-adenylation site (AATAAA)
denoted by underlining in Figure 2.2.


80
Vtg I 1 88 kD
YP 125
ELC 13
ELC 13
ELC 13
T
200 aa
C-TERMINUS
Figure 4.4 A graphic representation of the 13 kDa digestion products and their
positions in reference to YP 125, YP 105 and Vtg I. Note that the third
digestion product of YP 125 lies beyond the calculated C-terminus of YP
105. The indicated PEST site was found in YP 125, but is truncated, and
thus invalidated in YP 105.


30
M /
Fundulus heteroclitus Vitellogenin
PREDICTED...
f N-glycosylation site
^ N-myristoylation site
^ phosphorylation site
irtttttt
Figure 2.3 A schematic representation of potential sites for posttranslational
modifications of the putative F. heteroclitus Vtg protein as predicted by
the Prosite program (Bairoch et al., 1995). Phosphorylation sites
represent potential targets for the following kinases: c-AMP- and g-AMP-
dependent kinase, protein kinase C, casein kinase n, and tyrosine kinase.
The region denoted by asterisks represents the polyserine domain. Past
studies suggest that in addition to the sites displayed by the above-
mentioned kinases, every serine residue in this region undergoes
phosphorylation by an as-yet unidentified "vitellogenin kinase."


9
the blood of spawning females used the terms "low molecular weight spawning female
specific substance (L-SF), and high molecular weight spawning female-specific substance
(H-SF)" (Hamazaki et al., 1987a). Still other groups that concentrated on sequence
identity between their teleost proteins and the published mammalian ZP proteins, referred
to their sequences as teleost ZPs (Lyons et al., 1993). We designated the cDNAs and
coded proteins described here as choriogenins (Chgs), precursor proteins of the vitelline
envelope and chorion. We feel that this name accentuates the role of these molecules as
structural components of the vitelline
envelope and chorion, yet emphasizes their origin as being extra-ovarian and thus
different from the homologous ZP proteins of mammals. In Chapter 5, we present the
cDNA and protein sequences of three Chgs, as well as a partial characterization of F.
heteroclitus VEPs. The Chg data represent the most novel aspect of the dissertation,
with the hypothesis of liver-derived vitelline envelope components still fairly recent. One
of the remaining paradoxes presented by the Chg sequences is the comparative disparity
between the mammalian and teleostean systems for producing the extracellular matrix that
surrounds the oocyte. Because the sequence identity between the Chgs and mammalian
ZP proteins suggests an ancestral relationship, the differences in gene regulation, site of
synthesis, and functional roles offer a wealth of interesting questions for future
investigations.
By providing the structure of five previously unsequenced molecules that
contribute to the architecture of the ovarian follicle, we have contributed to our
understanding of reproductive processes in F. heteroclitus. However, we are even more


128
Epifano O, Liang LF, Familari M, Moos MC, Dean J (1995) Coordinate expression of
the three zona pellucida genes during mouse oogenesis. Development 121:1947-
1956
Evans AJ, Burley RW (1987) Proteolysis of apoprotein B during the transfer of very low
density lipoprotein from hens blood to egg yolk. J Biol Chem 262:501-504
Flegler C (1977) Electron microscopic studies on the development of the chorion of
viviparous teleost Dermogenys pusillus (Hemirhamphidae). Cell Tissue Res
179:255-270
Flickinger RA (1960) The relation of phosphoprotein phosphatase activity to yolk platelet
utilisation in the amphibian embryo. J Exp Zool 131:307-332
Flickinger RA, Rounds DE (1956) The maternal synthesis of egg yolk proteins as
demonstrated by isotopic and serological means. Biochim Biophys Acta 22:38-42
Flgel H (1967) Licht- und elektronmikroskopische Untersuchungen an Oozyten und Eim
einiger Knochenfische. Z. Zellforsch 83:82-116
Follett BK, Redshaw MR (1968) The effects of oestrogen and gonadotrophins on lipid
and protein metabolism in Xenopus laevis Daudin. J Endocrinol 40:439-585
Folmar LC, Denslow ND, Wallace RA, LaFleur GJ, Bonomelli S, Sullivan CV (1995)
A highly conserved N-terminal sequence for teleost vitellogenin with potential
value to the biochemistry, molecular biology and pathology of vitellogenesis. J
Fish Biol 46:255-263
Frohman MA, Dush MK, Martin GR (1988) Rapid production of full-length cDNAs
from rare transcripts: amplification using a single gene-specific oligo-nucleotide
primer. Proc Natl Acad Sci USA 85:8998-9002.
Fulton TW (1898) On the growth and maturation of the ovarian egs of teleostean fishes.
16th Annu Rep Fish Bd Scotl 3:88-134
Gerber-Huber S, Nardelli D, Haefliger J-A, Cooper DN, Givel F, Germond J-E, Engel
J, Green M, Wahli W (1987) Precursor-product relationship between vitellogenin
and the yolk proteins as derived from the complete sequence of a Xenopus
vitellogenin gene. Nucl Acids Res 15: 4737-4760
Greeley MS Jr., Calder DR, Wallace RA (1986) Changes in teleost yolk proteins during
oocyte maturation: correlation of yolk proteolysis with oocyte hydration. Comp
Biochem Physiol 84B:l-9


2
The mummichog, Fundulus heteroclitus, an estuarine teleost of the
Order Cyprinodontiformes as drawn by Lynn Milstead of the
Whitney Laboratory. The top fish is the female; the bottom fish,
displaying more pigment is the male.
Figure 1.1


41
from being completely absent in C. elegans (not shown; Speith et al., 1985), to a small
size in F. heteroclitus (99 Ser within 171 a.a. region), to a larger size in the chicken
Gallus (132 Ser within 291 a.a. region). A trend emerges in consideration of these data:
as one proceeds up the vertebrate phylogenetic ladder, Vtg polyserine domains appear
to increase in size. However, at least two exceptions to this trend have been reported:
the lamprey, Ichthyomyzon Vtg possesses a poly serine domain larger than that of F.
heteroclitus (113 Ser within 238 a.a. region; Fig. 2.2) and the
Gallus Vtg III (Byrne et al., 1989) contains a small polyserine domain (37 Ser; not
shown).
Although the vertebrate Vtg polyserine domains vary in size and serine content
as described above, their genetic organizations have sustained an element of similarity.
At the DNA level, the F. heteroclitus polyserine domain contains a distinct cluster of
TCX serine codons directly preceding a larger cluster of AGY serine codons (Fig. 2.2),
a pattern that is found in all other vertebrate Vtg cDNAs. When this cluster organization
was observed by Byrne et al. (1989) in Xenopus and chicken Vtgs, it was speculated that
a non-tetrapod Vtg would perhaps contain a cluster of only one type of serine codon,
representing the original trinucleotide repeating unit, and thus the original Vtg polyserine
domain. However, the polyserine domains presented here from the lamprey, sturgeon,
and mummichog are all dominated by the same two serine codons as is seen in Xenopus
and chicken, suggesting that these two codon clusters have been present within the Vtg
gene since before the divergence of agnathans and


79
Vtgl
predicted 188 kD
signal
Pclyserine
domain
HKKMVAxG
YP 45
I
EEDVEPIPEYKF
YP 18
EEEAWAVILRA
YP 39
AAAAExSP/EDT
YP 29
Vtgll
predicted 185 kD
signal

NGVSYAPEFAPG
YP 69
Poiyserine
domain
Figure 4.3 A graphical representation of F. heteroclitus yolk proteins positioned
along the length of the Vtg I and Vtg n. Length and positions along the
Vtg molecules are drawn to scale according to alignments of N-termini
data with cDNA translations. C-termini of yolk proteins were calculated
according to molecular weight estimations and should be regarded as
putative. The signal peptides and polyserine domains as predicted from
cDNA translations are indicated.


Table 4.1. N-Terminal Sequences of F. heteroclitus Yolk Proteins
Protein
Source
N-Terminal Sequence
Source
n
Y?
125
Oocyte
Blocked
?
3
Y?
105
Oocyte/egg
Blocked
?
2
YP
83
Egg
Blocked
?
1
YP
30
Oocyte/egg
Blocked
?
1
YP
77
Egg
Blocked
7
1
YP
69
Egg
NQVSY APE PA PGXTY
SYXYE
Vtg II
1
Y?
45
Oocyte
HKKMV AxGxx A
Vtg I
2
YP
39
Egg
EEEAV VAVIL RAVKA
D
Vtg I
2
YP
29
Oocyte
AAAAE xSFVE DTLYT
PN
Vtg I
1
YP
20
Oocyte
EEDVE PIPEY KFRRF
AKKYC
Vtg I
2
ELC
13
YP 125
YEPSD ELLQT PLQLI
KISD
Vtg I
1
ELC
13
YP 125
LPESL VDSDK WENP
LLREV
Vtg I
1
ELC
13
YP 125
KYCAK HIGVG LXACP
KPASQ
Vtg I
1
ELC
13
YP 105
YEPSD ELLQT PLQLI
KISD
Vtg I
1
ELC
13
YP 105
LPESL VDSDK WENP
LLREV
Vtg I
1
* Mapped to Vtg I (982-1001), C-terminal to a PEST site
ELC denotes N-terminus of products cleaved with Endo Lys C (.003 units/ug protein)


Figure 3.5 Northern blot analysis comparing relative expression of F.
heteroclitus Vtg I and Vtg II mRNAs.
A) Methylene blue staining of duplicate samples transferred to
nylon membranes before hybridization, showing equal loading of
lanes, as indicated by 28s and 18s rRNA bands. Lanes a and a
contain 300 pg Vtg I RNA translated from plasmid cDNA
(pMMBl); lanes b and b contain 300 pg Vtg II RNA translated
from plasmid cDNA (pFhv2a); lanes c and c contain 15 fig total
liver RNA from an estrogen-treated male; lanes d and d contain
15 fig total liver RNA from a female four days before spawning;
lanes e and e contain 15 fig total liver RNA from a female 4 days
after spawning. RNA markers (kb) are indicated with arrows on
the left.
B) Autoradiographs of the membranes shown above, indicating
bands hybridizing to Vtg I (left side) and Vtg II (right side) DNA
probes. Note that the Vtg I probe did not hybridize to the Vtg II
control RNA (lane b) and Vtg II probe did not hybridize to the Vtg
I control RNA (lane a).


Figure 4.5 A summary of the precursor-product relationship of Vtg I to
derived yolk proteins. The entire translated amino acid sequence
of the Vtg I cDNA sequence (LaFleur et al., 1995) is presented,
separated into sections representing yolk proteins as indicated by
brackets on the right. N-terminal sequences of isolated yolk
proteins are indicated by double underlining. Internal sequences
obtained from Endo LysC digestion products are indicated by
shaded lettering. The residues of the PEST site are represented by
bold face lettering. The predicted polyserine domain (no N-
terminal sequencing data) is shown in brackets.


1
55
109
163
217
271
325
379
433
487
541
595
649
703
7S7
811
865
919
973
1027
24
*
*
*
*
*

*
*
*
*
*
*

#
*
ATG
AAA
GCG
GTT
GTG
CTT
GCC
CTG
ACT
CTG
GCC
TTC
GTG
GCT
GGA
CAA
AAT
TTT
M
X
A
V
V
L
A
L
t*
L
A
p
V
A
G
Q
N
p
GCC
CCT
GAA
GCT
GCT
GGT
AAG
ACC
TAC
GTA
TAT
AAG
TAT
GAA
GCG
CTC
ATC
A
?
E
F
A
A
G
X
T
Y
V
Y
X
Y
E
A
L
I
CTG
GGC
GGT
CCT
GAG
GAA
GGT
TTG
GCA
AGA
GCT
GGA
TTG
AAA
ATC
AGC
ACC
L
G
G
L
P
E
E
G
L
A
R
A
G
L
X
I
S
T
AAA
CTT
CTA
CTC
AGT
GCA
GCT
GAC
CAA
AAT
ACT
TAT
ATG
CTG
AAG
Qmm
GTG
GAA
K
L
L
L
S
A
A
D
Q
N
T
Y
M
L
X
L
V
E
CCT
GAG
CTC
TCT
GAG
TAC
AGC
GGC
ATT
TGG
CCA
AAG
GAC
CCA
GCA
GTG
CCA
GCA
P
E
L
S
E
Y
S
G
I
W
P
X
D
P
A
V
P
A
ACC
AAG
TTG
ACA
GCA
GCC
CTT
CAC
CTC
AGC
TCG
CAA
TTC
CCA
TCA
AGT
TTG
AAT
T
K
L
T
A
A
L
H
L
S
S
Q
F
P
S
S
L
N
ACA
CCA
ATG
GTG
TTT
Uii
GGT
AAA
GTC
TTT
GCT
CCT
GAG
GAA
GTC
TCG
ACT
TTG
T
P
M
V
F
V
G
X
V
F
A
P
S
E
V
S
T
L
GTG
CTG
AAC
ATC
TAC
AGA
GGC
ATC
CTG
AAT
ATT
CTC
CAG
CTG
AAC
ATC
AAG
AAG
V
L
N
I
Y
R
G
I
L
N
I
L
Q
L
N
I
X
X
ACC
CAC
AAA
GTC
TAT
GAC
TTG
CAG
GAG
GTT
GGA
ACT
CAG
GGT
GTG
TGC
AAG
ACC
T
H
X
V
Y
D
L
Q
E
V
G
T
Q
G
V
C
X
T
CTC
TAT
TCC
ATC
AGT
GAA
GAT
GCA
CGA
ATT
GAG
AAC
ATC
CTT
CTG
ACC
AAG
ACC
L
Y
s
I
S
E
D
A
R
I
S
N
I
L
L
T
X
T
AGG
GAC
CTG
AGC
AAC
TGC
CAG
GAA
AGA
CTC
AAT
AAG
GAC
ATC
GGG
TTG
GCA
TAC
R
D
L
S
N
C
Q
E
R
L
N
X

I
G
L
A
Y
ACT
GAG
AAA
TGC
GAC
AAG
TGC
CAG
GAG
GAA
ACT
AAA
AAC
TTG
AGA
GGT
ACC
ACA
T
E
K
C
D
X
C
Q
S
E
T
X
N
L
R
G
T
T
ACA
TTA
AGT
TAC
GTC
TTG
AAA
CCA
GTC
GCC
GAT
GCC
GTC
ATG
ATC
CTG
AAG
GCG
T
L
S
Y
V
L
X
o
V
A
0
A
V
M
X
L
X
A
TAC
AAT
GAG
CTG
ATC
CAG
TTT
TCA
CCT
TTC
TCT
GAG
GCT
AAC
GGA
GCT
GCC
Y
V
N
E
L
I
Q
F
S
P
F
S
E
A
N
G
A
A
CAG
ATG
AGG
ACC
AAG
CAG
TCT
TTG
GAG
TTC
CTT
GAA
ATT
GAG
AAA
GAA
CCC
ATT
Q
M
R
T
X
Q
S
L
E
F
L
E
I
E
X
E
p
I
CCA
TCT
GTC
AAG
GCT
GAA
TAT
CGT
CAC
CGT
GGA
TCT
CTC
AAA
TAC
GAG
TTC
TCC
P
S
V
K
A
E
Y
R
H
R
G
S
L
X
Y
E
F
S
GAT
GAA
Qmn*
CTT
CAG
ACA
CCC
CTT
CAG
CTG
ATC
AAG
ATC
AGT
GAT
GCA
CCA
GCC
D
E
L
L
Q
T
P
L
<2
L
I
X
I
S

A
P
A
CAG
Qwm
GCA
GAG
GTC
CTG
AAG
CAC
CTG
GCT
ACC
TAC
AAC
ATT
GAG
GAT
GTT
CAT
Q
V
A
E
V
L
X
H
L
A
T
Y
N
I
E
D
V
H
GAA
AAT
GCA
CCT
TTG
AAG
TTT
TTG
GAA
CTG
GTA
CAA
CTC
CTC
CGT
ATT
GCC
CGC
E
N
A
P
L
X
F
L
E
L
V
Q
I
L
R
I
A
R
TAT
GAA
GAT
TTG
GAA
ATG
TAC
TGG
AAC
CAG
TAC
AAA
AAG
ATG
TCT
CCC
CAC
AGA
Y
E
D
L
E
M
Y
W
N
Q
Y
X
X
M
S
p
H
R


51
A protocol for rapid amplification of cDNA ends (RACE; Frohman et al., 1988),
was performed to retrieve this region using total RNA (described below) isolated from
the liver of an individual reproductively active female F. heteroclitus. A first strand
synthesis reaction was performed using 0.5 ng total RNA, the primer ROW 55 and
Superscript RT, followed by addition of a "poly-C tail" using 4 f of 1.0 mM dCTP, and
10 units of terminal deoxynucleotidyl transferase (BRL). Then, an amplification reaction
was carried out using the forward primer ROG 51, which targeted the poly C-tail, along
with the reverse primer ROW 55, and the Taq DNA polymerase. Through this effort
we successfully isolated a 230-bp band that was inserted into pGem-T, sequenced and
found to include a valid methionine codon, preceded by a short region that fit the criteria
for a transcription start site (Kozak, 1991).
Estrogen treatment. RNA isolation and analysis
Male and female F. heteroditus were collected from the estuarine creeks adjacent
to the Whitney Laboratory, and were maintained in running seawater tanks under
14L:10D photoperiod conditions at 25 + 2C. Fish were maintained for at least one
month before being used for RNA collections.
Experimental groups of fish were subjected to two intraperitoneal injections of
estradiol-17/3 (0.01 mg/g body weight) dissolved in 50 /fi coconut oil (Kanungo et al.
1990). Control groups were sham-injected with coconut oil alone. The first injection
was performed on day 1, the second injection on day 4, followed by sacrifice and liver
dissection on day 8.


>
ESTROGEN-INDUCED HEPATIC CONTRIBUTIONS
TO OVARIAN FOLLICLE DEVELOPMENT IN FUNDULUS HETEROCL1TUS:
VITELLOGENINS AND CHORIOGENINS
by
GARY JAMES LaFLEUR, JR.
DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1996


71
derived yolk proteins, have been scarce (Clark, 1973; Bergink and Wallace, 1974; Byrne
et al., 1984; Gerber-Huber et al., 1987; Wallace et al., 1990b; Yamamura et al., 1995)
and especially lacking from teleosts (Matsubaro and Sawano, 1995). Recent studies
focusing on Vtg genes and cDNAs have documented that many animals possess multiple
Vtg genes and proteins (Wahli et al., 1979; Blumenthal et al., 1984; review by Byrne
et al., 1989) offering an even more challenging puzzle to workers seeking to map these
relationships.
Obtaining a clear synopsis of precursor-product relationships in many teleosts, is
further complicated by the extensive yolk protein processing that occurs in teleost yolk
as compared to the yolk of tetrapods. The most striking difference in yolk content
documented in F. heteroclitus concerns the disappearance of a 125-kDa yolk protein (YP
125), and the concomitant appearance of smaller yolk protein bands immediately prior
to oocyte ovulation (Wallace and Begovac, 1985; Wallace and Selman, 1985; Greeley
et al., 1986). This enhanced proteolytic processing may be connected to a unique pre
ovulatory process that occurs in some teleost oocytes, termed hydration. Near the time
of germinal vesicle breakdown, a rapid increase in oocyte volume occurs, usually
attributed to the uptake of water (Fulton 1898; reviewed in Selman and Wallace, 1989).
In F. heteroclitus, a substrate spawner, post-maturational oocytes possess twice the
volume of pre-maturational oocytes (Wallace and Selman, 1985, Greeley et al., 1991;
McPherson et al.), but in the oocytes of pelagic spawners, oocyte volumes can increase
over four times the original volume, in as little as twelve hours (Wallace and Selman,
1981; Watanabe and Kuo, 1986; Craik and Harvey, 1987; LaFleur and Thomas, 1991).


136
Selman K, Wallace RA (1986) Gametogenesis in Fundulus heteroclitus. Amer Zool
26:173-192
Selman K, Wallace RA (1989) Cellular aspects of oocyte growth in teleosts. Zool Sci
6:211-231
Sharrock WJ, Rosenwasser TA, Gould J, Knott J, Hussey D, Gordon JI, Banaszak L
(1992) Sequence of lamprey vitellogenin. Implications for the lipovitellin crystal
structure. J Mol Biol 226: 903-907
Shen X, Steyrer E, Retzek H, Sanders EJ, Schneider WJ (1993) Chicken oocyte growth:
receptor-mediated yolk deposition. Cell Tiss Res 272:459-471
Shibata N, Yoshikuni M, Nagahama Y (1993) Vitellogenin incorporation into oocytes of
rainbow trout, Oncorhynchus my kiss, in vitro: effect of hormones on denuded
oocytes. Develop Growth & Differ 35: 115-121
Speith J, Denison K, Zucker E, Blumenthal T (1985) The nucleotide sequence of a
nematode vitellogenin gene. Nucleic Acids Res 13:7129-7138
Speith J, Nettleton M, Zucker-Aprison E, Lea K, Blumenthal T (1991) Vitellogenin
motifs conserved in nematodes and vertebrates. J Mol Evol 32:429-438
Stifani S, Le Menn F, Nunez Rodriguez J, Schneider W (1990) Regulation of oogenesis:
the piscine receptor for vitellogenin. Biochim Biophys Acta 1045:271-279
Swofford DL (1993) In: PAUP: Phylogenetic analysis using parsimony, version 3.1.
Computer program distributed by the Illinois Natural History Survey, Champaign,
II
Taborsky G (1974) Phosphoproteins. Adv Prot Chem 28:1-210
Taborsky G (1980) Iron binding by phosvitin and its conformational consequences. J.
Biol. Chem. 255:2976-2985
Taborsky G, Mok C-C (1967) Phosvitin. J Biol Chem 242:1495-1501
Tata JR, Baker BS, Deeley JV (1980) Vitellogenin as a multigene family. Not all
Xenopus vitellogenin genes may be in an "expressible" configuration. J Biol
Chem 255:6721-6726
Taylor MA, DiMichele L, Leach GJ (1977) Egg stranding in the life cycle of the
mummichog, Fundulus heteroclitus. Copeia 1977:397-399


131
Jarvik E (1980) In: Basic structure and evolution of vertebrates Vol. 1. Academic Press,
NY pp 439-446
Kaighn ME (1964) A biochemical study of the hatching process in Fundulus heteroclitus.
Dev Biol 9:56-80
Kanungo J, Petrino TR, Wallace RA (1990) Oogenesis in Fundulus heteroclitus. VI.
Establishment and verification of conditions for vitellogenin incorporation by
oocytes in vitro. J Exp Zool 25:313-321
Karasaki S (1963a) Studies on amphibian yolk. 5. Electron microscopic observations on
the utilization of yolk platelets during embryogenesis. J Ultrastruct Res 9:225-
247
Karasaki S (1963b) Studies on amphibian yolk. l.The ultrastructure of the yolk platelet.
J Cell Biol 18:135-151
Karasaki S (1967) An electron microscope study on the crystalline structure of the yolk
platelets of the lamprey egg. J Ultrastruct Res 18:377-390
Kishida M, Specker JL (1993) Vitellogenin in tilapia (Oreochromis mossambicus):
Induction of two froms by estradiol, quantification in plasma and characterization
in oocyte extract. Fish Physiol Biochem 12: 171-182
Kishida M, Specker JL (1994) Vitellogenin in the surface mucus of tilapia (Oreochromis
mossambicus): Possibility for uptake by the free swimming embryos. J Exp Zool
268:258-268
Kobayashi W and Yamamoto TS (1993) Factors inducing closure of the micropylar canal
in the chum salmon egg. J Fish Biol 42:385-394
Kozak M (1991) Structural features in eukaryotic mRNAs that modulate the initiation of
translation. J Biol Chem 266:19867-19870
Kunkel JG, Nordin JH (1985) Yolk proteins. In: Kerkut GA, Gilbert LI (eds)
Comprehensive insect physiology, biochemistry and pharmacology. Pergamon
Press, New York, pp 83-111
Kyte J, Doolittle RF (1982) A simple method for displaying the hydropathic character
of a protein. J Mol Biol 157:105-132
Laemmli, UK (1970) Cleavage of structural proteins during the assembly of the head of
bacteriophage T4. Nature 227 680-685


57
C7GAA7G7CACGC7GAA7ACACCAAAGAACACC77777CAAAC7GGGA7GGG77C77CCC7777AC 4170
LNVTLNTPXNTFFXLGWVLPFY
C7ACCAA77AACAACAC7GC7GC7GAGC7GCAGGCA77CCAGGGCAGG7GGATGGACCAGG7CACA 4236
L? INNTAAELQAFQGRWMDQVT
7ACA7GC7CACCAAG7CTGC7GCAGC7GAG7GCACCG7GG77GAAGACACAG7GG7CAC777CAAC 4302
YMLTXSAAAECTVVEDTVVTFN
AACAGGAAG7ACAAAACGGAGACGCCCCAC7C77GCCATCAGG7C77GGC7CAAGAT7GCACA7C7 4368
¡RXYXTSTPHSCHQVLAQDCTS
GAAA7CAAAT7CA7AG7GC7GC7GAAGAGGGA7CAAACAGCAGAACGGAA7GAGA7CAG7ATTAAG 4434
EIXFIVLLXRDQTAERNEISIX
A7TGAAAACATTGA7G77GACA7GTATCCCAAGGACAACGC7G77GTGG7GAAGG77AATGGAG7A 4500
IENIOVDMYPXDNAVVVXVNGV
GAAAT7CC7C7CACCAACC7GCCATATCAGCATCCAACAGGCAACATACAGATCCGACAAAGAGAA 4566
EIPLTNLPYQHPTGNIQIRQRE
GAGGGCA7C7C7C7GCATGC7CCCAG7CA7GGCC77CAGGAGG7C77CC7CAG777AAACAAAG7G 4632
EGISLHAPSHGLQSVFX.SLriKV
CAGG77AAAG77G77GAC7GGATGAGAGGCCAGACG7G7GGGC7C7GCGGAAAGGCCGACGGGGAA 4698
QVXVVDWMRGQTCGLCGXADGE
G7CAGACAGGAG7ACAGCAC7CCCAATGAACGGG7G7CCAGGAACGCAACCAGC77CGC7CAT7CC 4764
VRQEYSTPNERVSRNATSFAHS
TGGG7GC7GCC7GCAAAGAGC7GCCG7GACGCC7CAGAG7GC7ACATGCAAC77GAATCGG7GAAG 4830
WVLPAXSCRDASECYMQLESVX
C7CGAGAAACAGA7CAGCC7GGAAGGCGAGGAA7CCAAATGC7AC7CAG7CGAACC7GTC7GGCGC 4896
LSXQISLEGEESXCYSVEPVWR
7G7C7CCC7GGC7G7GCACCAG7GAGAACCACC7CCG7CAC7G7CGGGC7ACCA7GCGTG7C7C7G 4962
CLPGCAPVRTTSVTVGLPCVSL
GA77CAAACC7GAA7CGC7C7GATAGTC7CAGCAGCATC7A7CAGAAGAGCG77GACGTGAGCGAG S 0 2 8
DS.VLNRSOSLSSIYQXSVDVSE
ACGGCAGAG7CCCACC7GGCC7G7CGC7GCAC7CC7CAG7G7GCC7AAacgtgttgcctccrgac= 5094
TASSHLACRCTPQCA-
tttcg'ctctg'cttttgrgttacatggacgctcgcaaactaaaataaagaagcaactaaaaaaaaaaa 5160
aaaaaattcagctttggacttaacoaggctgaacct 5195
Figure 3.2--continued