In search of extramolluscan FMRFamide-related peptides


Material Information

In search of extramolluscan FMRFamide-related peptides
Physical Description:
Krajniak, Kevin Gerard, 1958-
Publication Date:

Record Information

Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 23233311
oclc - 25216563
System ID:

Table of Contents
    Title Page
        Page i
        Page ii
    Table of Contents
        Page iii
    List of Tables
        Page iv
    List of Figures
        Page v
        Page vi
        Page vii
        Page viii
        Page ix
        Page x
        Page xi
    Chapter 1. Introduction
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
    Chapter 2. The FMRFamide-related peptide of nereis virens
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
    Chapter 3. The FMRFamide-related peptide of callinectes sapidus
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
    Chapter 4. Summary and conclusions
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
    Biographical sketch
        Page 100
        Page 101
        Page 102
        Page 103
Full Text








I would like to thank Dr. M.J. Greenberg for his

guidance and support throughout this work. I am grateful

to Dr. D.A. Price for his suggestions and criticisms of

experiments, especially those concerning peptide

purification and sequencing. I am also grateful to Drs.

W.R. Kem, T.C. Rowe, W.E. Carr, and B.M. Dunn for their

participation on my committee. Furthermore I wish to

thank Dr. T.D. Lee (Beckman Research Institute, City of

Hope, Duarte, California) for fast atom bombardment mass

spectrometric analysis of peptide samples, and Dr. B.M.

Dunn, Benne Parten, and Alicia Alvarez for microsequencing

and synthesis of peptides. Finally I would like to thank

Mr. John Young for his assistance in using computers and

software, and Mrs. Lynn Milstead and Mr. Jim Netherton III

for their help in preparing the figures. This work was

supported by NIH grant HL-28440 to M.J. Greenberg.



ACKNOWLEDGEMENTS...................................... ii

LIST OF TABLES ....................................... iv

LIST OF FIGURES ........................................... v

ABBREVIATIONS ........................................ vii

ABSTRACT ............................................. ix


1 INTRODUCTION........................................ 1

Background ..................................... .... .. 1
Objectives.......................................... 9

NEREIS VIRENS ................................. .. 10

Introduction .................................. 10
Methods ....... .... . ........................ ...... 13
Results ........................................ 20
Discussion .......................................... 42

CALLINECTES SAPIDUS ........................... 53

Introduction..................................... 53
Methods ........................................ 58
Results ........................................ 63
Discussion.......................................... 79

4 SUMMARY AND CONCLUSIONS ........................ 86

REFERENCES ........................................... 90

BIOGRAPHICAL SKETCH.................................. 100



TABLE 1-1. INVERTEBRATE FaRPs ........................ 2


CALLINECTES SAPIDUS FaRP ............................ 71

SAPIDUS ............................................. 76

SAPIDUS ............................................. 78

Figure 2-1. The isolated Nereis esophagus bioassay... 22

Figure 2-2. The HPLC purification and sequencing
of the Nereis FaRP.................................. 24

Figure 2-3. The central nervous system of Nereis
virens............................................... 27

Figure 2-4. Photomicrographs of FMRFamide
immunoreactivity in the brain (supraesophageal
ganglion) of Nereis virens........................... 30

Figure 2-5. The positions of the 26 paired nuclei of the
brain (supraesophageal ganglion) superimposed on the
pattern of FMRFamide immunoreactive cell bodies
in Nereis virens....................................... 31

Figure 2-6. Photomicrographs of the FMRFamide
immunoreactivity in the subesophageal ganglion
and ventral nerve cord of Nereis virens............. 33

Figure 2-7. Photomicrographs of immunoreactive
FMRFamide in a wholemount and cross section of an
intersegmental ganglion of Nereis virens............ 36

Figure 2-8. Photomicrographs of FMRFamide
immunoreactivity in the gut of Nereis virens........ 38

Figure 2-9. Photomicrographs of FMRFamide
immunoreactive varicosities in the peripheral
tissue of Nereis virens............................... 40

Figure 2-10. The effects of FMRFamide on the
isolated, spontaneously active esophagus of
Nereis virens........................................ 41

Figure 2-11. The effects of FMRFamide on the
isolated, electrically stimulated esophagus
of Nereis virens .................................... 44

Figure 2-12. Dose-response curves of five
different Nereis, showing the effects of
FMRFamide on the rate of relaxation................. 45

Figure 2-13. The effects of methysergide (UML)
and benzoquinonium (BQ) on FMRFamide
relaxation of the spontaneously active
esophagus of Nereis virens........................... 46

Figure 3-1. A schematic of the blue crab
Callinectes sapidus................................. 57

Figure 3-2. The perfused Callinectes heart
isolated in a bioassay apparatus..................... 62

Figure 3-3. The HPLC purification and FAB-mass
spectrum of the Callinectes FaRP isolated from
512 animals .......................................... 65

Figure 3-4. The HPLC purification and sequencing
of the Callinectes FaRP from 200 animals............ 68

Figure 3-5. The effects of increasing doses of
GYNRSFLRFamide on the isolated heart of the
blue crab Callinectes sapidus........................ 73

Figure 3-6. A dose-response curve, showing the
effects of GYNRSFLRFamide on the heart of
Callinectes sapidus................................. 75


Amino Acids

























































































Other Abbreviations


Benzoquinonium chloride

Crustacean cardioactive peptide


Central nervous system

Counts per minute

Carboxy terminal

Ethylenediaminetetraacetic acid

Fast atom bombardment

FMRFamide-related peptide

Fluorescein isothiocyanate-conjugated

goat anti-rabbit globulin


N'-[2-ethanesulfonic acid]

Heptafluorobutyric acid

High Performance Liquid Chromatography


Not determined

No response

Amino terminal


Trifluoroacetic acid




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



Kevin Krajniak

August, 1990

Chairman: Michael J. Greenberg
Major Department: Pharmacology and Therapeutics

FMRFamide-related peptides (FaRPs) have been found

primarily in molluscs and arthropods. Since these phyla

are thought to have evolved, with annelids, from a common

ancestor, annelids should contain FaRPs. Indeed, acetone

extracts of the polychaete annelid, Nereis virens,

contained a single tetrapeptide, FMRFamide. Each worm

contained 100 to 600 fmol of peptide, levels 10 to 100

times lower than in molluscs.

Immunohistochemical techniques were used to localize

the peptide. FMRFamidergic cells and fibers were found in

the supraesophageal (brain) and subesophageal ganglia, as

well as in the intersegmental ganglia of the ventral nerve

cord. Immunoreactive fibers were present in the neuropile

of, and the connectives between, the supraesophageal,

subesophageal, and intersegmental ganglia. In the

periphery FMRFamidergic fibers and a few cell bodies were

observed in the gut. Sparse fibers were also seen in the

body wall, parapodia, and cephalic palps. When the

antiserum was preabsorbed with FMRFamnide, no specific

immunoreactivity was detected.

The esophagus of Nereis, isolated and suspended in a

tissue bath, responded to FMRFamide with a dose-dependent

relaxation; threshold was between 30 and 300 nM, and the

EC50 was 1.550.60 p.M. Neither methysergide nor

benzoquinonium modified this response.

In conclusion, FMRFamide is a neurotransmitter in

both the central and peripheral nervous systems of Nereis

virens and may be involved in controlling digestive tract


In arthropods, the FaRPs fall into three subgroups

based upon sequence; two of these groups occur in insects,

but only one in crustaceans. Still, FaRPs from only a

single crustacean species have been characterized.

Therefore, to determine the variability of this peptide

family in Crustacea, FaRPs from the pericardial organs and

thoracic ganglia of the blue crab Callinectes sapidus were

isolated and sequenced. Multiple peaks of

inimunoreactivity were present, and one yielded the

sequence, GYNRSFLRFamide. Each animal contained between 7

and 13 pmol.

The peptide caused a dose-dependent increase in heart

rate; threshold was 10 to 30 nM, and the EC50 was

32362 nM. A structure-activity study of the crab heart

suggests that, for full potency, the peptide should be at

least a heptapeptide with the sequence XXZFLRFamide, where

X is any amino acid and Z is either asparagine or serine.



Neuropeptides can be grouped in subsets based upon

various criteria; e.g., structure, function, or the tissue

or species in which they occur. A group of neuropeptides

with similar sequences constitute a family which, in turn,

can be subdivided into a nuclear- and extended

sub-families. Members of the nuclear family contain a

common core sequence which is required for biological

activity; the remainder of the sequence differs from the

others in the sub-family in only a few extra amino acid

residues. In peptides of the extended sub-family, the

common core sequence is much reduced, and the remainder of

the sequence is much more variable.

The family of the FMRFamide-related peptides (FaRPs)

is exemplary. The eponymous tetrapeptide

Phe-Met-Arg-Phe-NH2 (FMRFamide) was first isolated from

the ganglia of a clam, Macrocallista nimbosa (Price and

Greenberg, 1977), and is present in all classes of the

phylum Mollusca (Price et al., 1987). The nuclear family

in molluscs includes, at present, ten additional peptides

in four groups (Table 1-1). One group, comprises only











Price et al., 1987

Price et al., 1987

Price et al., 1990

Martin & Voigt, 1987

Homarus americanus

Leucophaea maderae

Schistocerca gregaria

Drosophila melanogaster


Ascaris suum

Panagrellus redivivus

Trimmer et al., 1987

Holman et al., 1986

Robb et al., 1989

Nambu et al., 1988

Cowden & Stretton, 1989

Price, personal


FMRFamide and another tetrapeptide -- FLRFamide -- in

which the methionine has been replaced by a leucine

residue. This group is ubiquitous. Four heptapeptides of

the form XDPFLRFamide (where X is Gly, Ser, Asn, or pGlu)

are present in only pulmonate gastropods (e.g., land

snails and slugs); all are N-terminally extended analogs

of FLRFamide, and they only differ in their N-terminal

residue (Price et al., 1987). The beginning of an

analogous set of pulmonate peptides XXPYLRFamide (where XX

is Ser-Glu or Asn-Asp) has been discovered in Helix (Price

et al., 1990). Finally, the cephalopod mollusc Octopus

seems to contain AFLRFamide and TFLRFamide (Voigt and

Martin, 1986; Martin and Voigt, 1987).

Members of this nuclear sub-family are also present

in other invertebrate phyla (Table 1-1). In the

Arthropoda, peptides containing the core sequence

FXRFamide (where X is Leu, Met, or Ile) have been

sequenced: TNRNFLRFamide and SDRNFLRFamide from the

lobster Homarus americanus (Trimmer et al., 1987);

pQDVDHVFLRFamide from the cockroach Leucophaea maderae

(Holman et al., 1986); PDVDHVFLRFamide from the locust

Schistocerca gregaria (Robb et al., 1989); and

DPKQDFMRFamide from the fruit fly Drosophila melanogaster

(Nambu et al., 1988). Finally, among the nematodes,

Ascaris suum contains the neuropeptide KNEFIRFamide

(Cowden et al., 1989) and Panagrellus redivivus,

SADPNFLRFamide and SDPNFLRFamide (D. Price, personal

communication). Although these peptides of the nuclear

sub-family are all quite similar, no particular sequence

has ever appeared without modification in two different


Molluscs and arthropods, along with annelids are

believed to have evolved from a common flatworm-like

ancestor (Barnes, 1974; Vagvolgyi, 1967; Stasek, 1972).

This notion is based mainly on similarities of embryology

and anatomy. Embryologically, members of all three phyla

are characterized by spiral, determinate cleavage. The

first opening to the digestive tract to develop during

gastrulation is the mouth, or protostome. Annelids and

molluscs also share a very similar larval form, the


Anatomically, annelids and arthropods display

segmentation, as do very primitive molluscs, like the

monoplacophoran Neopilina galatheae (Lemche and

Wingstrand, 1959). The metamerism in Neopilina involves,

not only the musculature, but also the nervous system

(Lemche and Wingstrand, 1959). However, the ontogeny of

segmentation is different in each phylum and appears to

have been a recent evolutionary acquisition.

Annelid and arthropod nervous systems are very

similar. They consist anteriorly, of a dorsally located

brain (supraesophageal ganglion) united by


circumesophageal connectives to a subesophageal ganglion;

the ganglia and connectives constitute the

circumesophageal nerve ring. The ring is connected, in

turn, to a ventral nerve cord with ganglionic swellings in

each segment (Manton, 1970). Primitive molluscs, e.g.,

monoplacophorans and polyplacophorans chitonss), also have

a circumesophageal nerve ring, but their nerve cords lack

arthropod-like segmental ganglia.

According to recent comparisons of 18S ribosomal RNA,

annelidan RNA sequences are more similar to the molluscan

ones than to those of arthropods (Field et al., 1988).

The resulting evolutionary tree suggests that the

arthropods diverged from annelids and molluscs before the

latter phyla diverged from each other.

The basis for the suggestion that these three phyla

evolved from a common flatworm-like ancestor comes from:

(1) the similarities of flatworm embryology to that of the

annelids, molluscs, and to a lesser extent arthropods, and

(2) the simpler, more primitive organization of flatworm


Genes encoding FaRP precursors have been isolated

from and sequenced in three gastropod and two arthropod

species. In the gastropods, three different genes have

been identified: in Aplysia, a gene encoding 28 copies of

FMRFamide and one of FLRFamide (Taussig and Scheller,

1986); in Lymnaea, a tetrapeptide gene encoding nine


copies of FMRFamide and two of FLRFamide (Linacre et al.,

1989); and in Helix, a single gene encoding ten copies of

FMRFamide and one copy of FLRFamide (Lutz et al., 1990).

Also in Helix, a second separate gene has been found to

encode multiple copies of some of the pulmonate

heptapeptides (Lutz, personal communication). Thus in

gastropod molluscs, two separate genes seem to encode the

tetrapeptides and heptapeptides.

The FaRP genes of the congeneric insects Drosophila

melanogaster and Drosophila virilis contain copies of

several N-terminally extended analogs of FMRFamide

(Schneider and Taghert, 1988; Taghert and Schneider,

1990). They have two characteristics in common with all of

the molluscan genes studied so far: first, they encode for

at least two different peptides; and second, they include

multiple copies of at least one of the peptide sequences.

There are still other peptides the genes of which have yet

to be isolated. The discovery of such genes will probably

point to further sequences which have not yet yielded to

classical peptide isolation, purification, and sequencing


Immunocytochemistry and bioassay suggest several

biological functions for the FaRPs in the nuclear

sub-family. Initially characterized as cardioregulators,

these peptides modulate both cardiac and non-cardiac

muscles of molluscs (Painter and Greenberg, 1982; Lehman


and Greenberg, 1987; Krajniak and Bourne, 1987; Payza,

1987), arthropods (Trimmer et al., 1987; Cuthbert and

Evans, 1989, Mercier et al., 1989), and annelids (Kuhlman

et al., 1985a; 1985b; Li and Calabrese, 1987; Diaz-Miranda

et al., 1989; Fujii et al., 1989; Calabrese and Norris,

1989), as well as the musculature of nematodes (Smart et

al., 1990). In addition, they appear to control the

digestive and reproductive states of pulmonate molluscs

(reviewed in Krajniak et al., 1989). The FaRPs also

affect nerve cells in molluscs (reviewed in Kobayashi and

Muneoka, 1989), arthropods (Weimann and Marder, 1989),

annelids (Calabrese and Norris, 1989), and nematodes

(Cowden et al., 1989).

The extended sub-family of FaRPs contains peptides

that end in Arg-Phe-NH2 (RFamide); this C-terminal

dipeptide is the minimal portion of the core required for

cross-reactivity with FMRFamide antisera. Their

cross-reactivity in standard FMRFamide bioassays (e.g.,

clam heart) is usually very poor. To date, these peptides

have been identified primarily in cnidarians and

chordates: pQGRFamide and pQLLGGRFamide are present in the

cnidarians (Grimmelikhuijzen and Graff, 1986;

Grimmelikhuijzen et al., 1988); the vertebrate peptides

include LPLRFamide in the chicken brain (Dockray et al.,

1983), gamma-1-MSH (YVMGHFRWDRFamide) in the ox pituitary

(O'Donahue et al., 1984; Triepel and Grimmelikhuijzen,


1984), and the morphine modulating peptides, FLFQPQRFamide

and AGELSSPFWSLAAPQRFamide in the cow brain (Yang et al.,


Three members of the extended family occur in

protostomous invertebrates: the L5 peptide of the mollusc

Aplysia californica that ends in QGRFamide (Shyamala et

al., 1986); and two peptides found in mosquito heads,

pQRP[hP]SLKTRFamide and TRFamide (Matsumoto et al., 1989).

A variety of other peptides which are shirttail

relations of the FaRP family (Price and Greenberg, 1989).

They include predicted peptide sequences from genes like

pQDPFLRIamide from the Helix heptapeptide gene, and

SRPQDPVRSamide from the gene of Drosophila. Other members

of this group come from a family of peptides in starfish

and include SALMFamide, GFNSALMFamide, and

SGPYSFNSGLTFamide. These peptides were isolated using an

antiserum to the heptapeptide pQDPFLRFamiide (Elphick et

al., 1989). Their cross-reactivity with the FMRFamide

receptor is likely to be extremely poor.

In conclusion, the nuclear sub-family FaRPs are

restricted to protostomes, whereas the extended sub-family

is ubiquitous in metazoans. Since most of the parasites

and pests that plague man and commercially important

vertebrates are protostomes, studies of different FaRPs

and their receptors could yield new forms of

antihelimenthics, insecticides, and other regulatory agents.


The first objective of this study was to extend the

boundaries of the protostomous nuclear sub-family of

FaRPs. Since no peptides of annelid worms have been

sequenced, extracts of the polychaete Nereis virens were

examined for novel immunoreactive peptides.

The second objective was to determine the

physiological function of the nereid FaRP. To this end,

the distribution of the peptide was mapped

immunohistochemically in both nervous and non-nervous

tissues. Also, bioassays were performed on immunoreactive

tissues, particularly the esophagus.

The third objective was to explore the variations in

sequence in the nuclear sub-family of FaRPs in the

superphylum, Arthropoda. Several species of insect, but

only one species of crustacean, Homarus americanus, have

yielded peptide sequences. Therefore, FaRPs of the blue

crab Callinectes sapidus were isolated and sequenced.

Since the crustacean FaRPs are cardioexcitatory, the

final objective was to study the structure-activity

relationships of these peptides on the crustacean heart.




Most of the known invertebrate FaRPs have been

discovered in molluscs and arthropods. The annelids, are

a closely related phylum, and ample immunohistochemical

and pharmacological data suggests that FaRPs are present

in all three classes of Annelida (Kuhlman et al., 1985a,b;

Li and Calabrese, 1987; Porchet and Dhainaut-Courtois,

1988; Diaz-Miranda et al., 1988; 1989; Fujii et al., 1989;

Calabrese and Norris, 1989). Nevertheless, no FMRFamide-

related peptides in annelids have actually been sequenced.

All annelids are segmented worms, but the three

classes in the phylum -- Polychaeta, Oligochaeta, and

Hirudinea -- are quite different (Clark, 1978). The most

variable of these classes is the Polychaeta, which

includes mostly marine worms, distributed in many

families, and characterized by a mixture of primitive and

advanced features (Fauchild, 1977). The Oligochaeta

comprise mostly terrestrial and freshwater worms, are much

less diverse in structure than the polychaetes, and can

readily be systematized according to their embryology and

the organization of their reproductive systems


(Brinkhurst, 1982). The evolutionary connection between

the oligochaetes and polychaetes is unclear, but the

leeches, constituting the class Hirudinea are very closely

allied to the oligochaetes, from which they diverged

relatively recently. The members of this class are the

least diverse and the most specialized of the three.

The role of FMRFamide in the leech Hirudo medicinalis

has been well studied (Kuhlman et al., 1985a,b; Li and

Calabrese, 1987; and Calabrese and Norris, 1989).

FMRFamidergic cell bodies, processes, and varicosities are

present in the brain and all segmental ganglia. The

amount of ir-FMRFamide declines in the two ganglia nearest

each end of the animal (head and tail). The function of

many of the neurons in each leech ganglion have been

identified: the immunoreactive cells include heart

excitor motor neurons, heart accessory motor neurons,

swim-initiating interneurons, excitatory body wall motor

neurons, and the rostral and lateral penile everter motor

neurons (Kuhlman et al., 1985a,b; Li and Calabrese, 1987;

and Calabrese and Norris, 1989). Furthermore, a peak of

immunoreactive-FMRFamide (ir-FMRFamide) has been extracted

from nervous tissue of the leech and chromatographically

characterized as FMRFamide; but it has yet to be sequenced

(Li and Calabrese, 1987).

Application of exogenous FMRFamide to the CNS mimics

the effects of the heart motor neurons and the


swim-initiating interneurons, both of which can accelerate

the central pattern generator of the heartbeat.

Furthermore, the peptide also stimulates the heart in the

same way that the heart accessory motor neuron does. Bath

application of FaRPs to isolated body wall longitudinal

muscles causes a dose-dependent increase in tone. Thus,

FMRFamide is involved both centrally and peripherally in

the control of the leech heart and body wall musculature.

Hirudo medicinalis is a highly evolved, specialized

worm, and the distribution and actions of FMRFamide in the

leech may not, therefore represent those in the remaining

two classes of annelids. The polychaetes are considered

more primitive, and an examination of their peptides would

create a more general understanding of the sequences,

distributions, and functions of FaRPs in annelids.

Because FMRFamide immunoreactive cell bodies were found in

the brain of the polychaete Nereis diversicolor (Porchet

and Dhainaut-Courtois, 1988), I began to isolate the FaRPs

from the related species Nereis virens, a polychaete that

is commercially available in very large numbers.

As an approach to the physiological roles of the

nereid FaRP, the distribution of the peptide was mapped

immunohistochemically. Furthermore, bioassays were

performed on the esophagus, a tissue containing FMRFamide-





Nereis virens was purchased from F.H. Hammond

Wholesale (Wiscasset, ME). The animals were maintained in

flowing natural seawater at 15 C until used.

Peptide Extraction, Purification, and Sequencing

The anterior 3 cm of the worms (total length, about

20 cm) were cut off and placed in 4 volumes of acetone;

the tissue was allowed to steep overnight at -20 C. The

anterior 3 cm was used because it contained the

circumesophageal nerve ring and, in a preliminary

experiment, very little immunoreactivity was extracted

from the remaining 17 cm of the body.

The acetone was decanted and driven off by rotary

evaporation. The remaining aqueous residue was clarified

by centrifugation and filtration. An equal volume of

water was added to this fluid and then separated by HPLC.

Two extracts were made and purified: a preliminary one of

500 worms, and a definitive one of 2000 worms.

The aqueous residue from the 500 animal extract

(200 ml) was loaded onto a Waters Radial Pak C-18 reverse-

phase HPLC column (8 x 100 mm) at a flow rate of 10

ml/min. The material was then eluted from the column with

a gradient of acetonitrile (ACN) (16 to 32% in 20 min)

containing 0.1% trifluoroacetic acid (TFA). One-minute

fractions were collected and the ir-FMRFamide determined


by radioimmunoassay (RIA; described below). A rabbit

polyclonal antiserum (S253) made to YGGFMRFaimide was used

in the RIA (Price et al., 1987). The few fractions

containing the most immunoreactivity were diluted with

water and pumped onto a Waters Novapak C-18 reverse-phase

HPLC column (3.9 x 150 mm) which was eluted with a

gradient of n-butanol (2 to 8% in 20 min) containing 0.1%

TFA. Fractions were collected every 30 seconds, and the

pooled immunoreactive fractions were further purified on

the same column, except that the ACN/TFA gradient already

described was used. The two buffer systems were used

thus, in alternation, until the immunoreactive peak

corresponded to a single peak of UV absorbance at 210 nm.

The sample was then sequenced by the Protein Chemistry

Laboratory of the University of Florida Interdisciplinary

Center for Biotechnology Research.

The 2000 worm extraction yielded a liter of aqueous

residue which was purified, 200 ml at a time, on a

Brownlee Prep 10 Aquapore Octyl C-8 reverse-phase HPLC

column (10 x 100 mm). The same ACN/TFA gradient was used,

but with a flow rate of 4 ml/min. The immunoreactive

fractions from all five runs were pooled and loaded on the

same column once again. All subsequent steps were

performed on a Brownlee RP-300 Aquapore Octyl reverse-

phase HPLC column (2.1 x 220 mm) with a flow rate of

0.5 ml/min.


Since alternation of the ACN/TFA and butanol/TFA

systems could not separate a peptide to purity, an

additional step was added. The sample was neutralized to

pH 7 with phosphate buffer; it was loaded onto the RP-300

column, and the immunoreactive material was eluted from

the column with a gradient from 5 mM sodium phosphate (pH

7.0) to 5 mM phosphoric acid (pH 3.0), both containing 25%

ACN. Although this step was inefficient -- a large

portion of the immunoreactivity being eluted with the load

fraction -- it gave a high degree of purification.

Nevertheless, the retained immunoreactive peak was still

not pure, as evidenced by the presence of 280 nm

absorbance in the immunoreactive fraction. Therefore, the

fractions were oxidized with 50 gl/ml of hydrogen peroxide

(30%) for 15 min at room temperature. The oxidized

peptide eluted at 4.51 min, several minutes earlier than

the unoxidized peptide, and well before the impurities at

6.5 min.

The immunoreactivity in the final peak was measured

with both the standard S253 antiserum and another

polyclonal antiserum (Q2) raised in a rabbit to


The purified, oxidized peptide was divided into two

equal portions; one was sequenced at the University of

Florida, and the other was analyzed by fast atom

bombardment mass spectrometry (T.D. Lee, Division of

Immunology, Beckman Research Institute of the City of

Hope, Duarte, California), as previously described

(Bulloch et al.,1988).


The antisera used in the RIA were raised in rabbits

to either YGGFMRFamide conjugated with thyroglobulin

(identified as S253; Price et al., 1987) or pQDPFLRFamide

conjugated with thyroglobulin (identified as Q2). The

trace was iodinated pQYPFLRFamide.

The cross-reactivity of S253 and, to a lesser extent,

Q2 have been examined with a number of peptides (Greenberg

et al., 1988; Elphick et al., 1989). With S253 FMRFamide

and the pulmonate heptapeptides are equipotent, whereas

FLRFamide is one order of magnitude less potent. Other

substitutions of the amino acids in tetrapeptide structure

also decrease the potency of the analog. Replacing the

N-terminal phenylalanine decreases the potency by two to

three orders of magnitude, while replacement of the

methionine in position 2 causes only a decrease of one

order of magnitude. Substitution of the C-terminal

phenylalanine also drops the potency by three orders of

magnitude. Peptides without the C-terminal amide have no

activity. Unrelated peptides, like SCPB which has a

C-terminal sequence of FPRMamide, show no cross-

reactivity. Thus, S253 binds preferentially to FMRFamide


and to N-terminally extended peptides with the C-terminal

sequence FXRFamide (where X is leucine or methionine).

Q2 reacts only poorly with FMRFamide-like peptides

containing methionine sulfoxide in the second position.

It also cross-reacts with SALMFamide, GFNSALMFamide, and

SGPYSFNSGLTFamide which are shirttail FaRPs from

echinoderms. Thus, the specificity of Q2 is for peptides

with the C-terminal sequence LXFamide, where X is any

amino acid.

The RIA was performed as follows. An aliquot (2 il)

of each HPLC fraction was transferred to a glass test tube

with an automatic diluter (Micromedic model 3000) that

added 48 4l RIA buffer to give a final sample volume of

50 gl. The RIA buffer included 0.01 M sodium phosphate

containing 1% bovine serum albumen, 0.9% sodium chloride,

0.01% merthiolate, and 0.025 M sodium EDTA, and was

adjusted to pH 7.5. For sample dilution, a buffer without

albumen was used to avoid contaminating the fraction with

anything containing amino acids. The trace (10,000 cpm)

in 100 4l of buffer was added to each tube along with

100 gl of diluted antiserum (1:10,000 for S253, and 1:500

for Q2). All of the tubes were left overnight in the

refrigerator (4 C), and 1 ml of charcoal solution was

added in the morning. The charcoal solution contained

0.25% charcoal, 0.025% dextran, and 0.01% merthiolate in

0.1 M sodium phosphate, pH 7.5; it was stirred overnight

before the first use, and was kept in the refrigerator

thereafter. After 15 minutes, the charcoal was

centrifuged at 2500 x g for 15 minutes, and the

supernatant was then decanted and counted.


The desired tissues were removed and fixed in Bouin's

fixative overnight in the refrigerator. Following the

fixation, most of the tissues were embedded in paraffin

(Humason, 1967), sectioned, rehydrated, and stained by the

indirect immunofluorescence method (Beltz and Burd, 1989).

Ganglia to be wholemounted were desheathed the day after

fixation, washed, and stained according to the

wholemounting procedures outlined in Beltz and Burd

(1989), except that 1% goat serum was added to both the

primary and secondary antisera to block non-specific

immunoreactivity. All tissues were stained with a

different polyclonal serum raised in rabbits to FMRFamide

(1:1000 dilution; supplied by Dr. E. Weber, Oregon Health

Science University, Portland, OR). Control tissues were

stained with the same antiserum (1:1000 dilution)

preabsorbed with FMRFamide (1 mg/ml). The secondary

antibody in all cases was fluorescein

isothiocyanate-conjugated goat anti-rabbit gamma globulin

(FITC-GAR, 1:50 dilution).

The specificity of the Weber antiserum differs from

those of S253 and Q2 in that it cross-reacts poorly with

even close analogs of FMRFamide or YGGFMRFamide (Price,

personal communication). Peptides containing the

C-terminal sequence FXRFamide, where X is leucine or

proline, are at least two orders of magnitude less potent

when this serum is used in the FMRFamide RIA.

Nevertheless, this antiserum always shows more

immunoreactivity than S253 in immunohistochemistry

(personal observation), and is widely used.


The esophagus, unlike other parts of the digestive

tract, is unattached to the body wall and was therefore

chosen for the bioassay. Contractions of the isolated

esophagus were recorded as follows. The portion of the

esophagus distal to the pharynx anteriorly, and proximal

to the intestine posteriorly, was removed; the esophageal

ceca (digestive glands) were left attached. Ligatures

were then tied around both ends of the tissue, and the

posterior end was secured to a stationary rod, while the

anterior was connected to a force-displacement transducer

(Grass FT-03) coupled to a Grass Model 7 Polygraph. In

some experiments, the esophagus was stimulated with

concentric platinum electrodes on the stationary rod

connected to a Grass stimulator (2 Hz, 1 ms, 20 V) (Figure

2-1). All drugs were added directly to the organ bath

from stock solutions; the dose reported was the final

molar concentration in the bath.


Peptide Extraction, Purification, and Seqcuencin'

The initial fractionation of the extract yielded a

single peak of immunoreactivity, irrespective of the

column used. But subsequent steps often showed a minor

peak of immunoreactivity corresponding to the oxidized

form of the peptide. The initial peak from the first

batch of 500 worms contained about 2.0 nmol of

ir-FMRFamide. The purified peak of immunoreactivity

(Fig. 2-2a) was equivalent to about 600 fmol/worm, but the

UV absorbance was more indicative of about 300 fmol/worm,

and the sequencing levels were those expected for about

150 fmol/worm.

From the second extract, containing 2000 worms, the

initial amount of ir-FMRFamide from the pooled first run

was 15.4 nmol. In the final step (Fig. 2-2c), I recovered

about 300 fmol/worm according to the RIA data, 250

fmol/worm from the UV data, and 100 fmol/worm from

sequencing. In this instance, the amount of peptide is

smaller than that in the 500 worm batch, but much of the

peptide was lost when the sample was separated on the pH


Figure 2-1. The isolated Nereis esophagus bioassay
apparatus. The figure shows a schematic of the isolated
esophagus suspended in a tissue bath connected at its
posterior end to an hook in the electrode support and at
its anterior end to a transducer. Concentric electrodes
surround the tissue and were used with a Grass stimulator
to stimulate a tonic contraction (2 Hz, 1 ms, 20 V). In
this drawing the recording device shown is a smoked-drum
kymograph, but in reality a Grass transducer and recorder
were used. The same apparatus was used whether the tissue
was electrically stimulated or not. All drugs were added
directly to the organ bath from stock solutions; the dose
reported was the final molar concentration in the bath.






0,005 AU

5 10 minutes
24 28 % ACN .
0.1% TFA

I01 'U

4 51
-AI *

0 5 10 minutes

E' 1.2


LL 0.4

0 5 10 minutes
16 24 % ACN
d 50 0.1% TFA
5 Ol

- Met
I 30

U 20
:2 A
S0 ^Pe

2 4 6 8 10 2 4 6 8 10
SeQuencer Cycle Seouencer Cycle

e oo

; 80


20 J0
S0- 1I

400 600 800 1000


The antisera S253 and Q2 indicated different levels

of ir-FMRFamide in the major peak of the 2000 animal

extraction. The level assayed with S253 was 0.16 nmol,

whereas it was 1.6 pmol with the Q2 antiserum, a

difference of 100-fold. The same differential occurs when

the methionine in FMRFamide is oxidized to the sulfoxide.

Thus, these data suggested that the peak contained an

oxidized peptide with the C-terminal sequence of


The sequence of both samples can be seen in Figures

2-2b and 2-2d. The sequence from the 500 animal extract

clearly contains the tripeptide, Phe-Met-Arg. Since the

S253 antiserum requires an amidated C-terminal

phenylalanine for binding in the RIA (Price, 1983;

Greenberg et al., 1988; Price, 1987), this tripeptide

could not constitute the complete sequence. Indeed, in

the analysis of the oxidized peak from the 2000 worm

extract, the peak had the sequence Phe-Met-Arg-Phe. The

presence of the C-terminal amide was indicated by the

weight of the molecular ion of the oxidized sample as

determined by FAB-mass spectrometry (Fig. 2-2e). The

observed molecular ion was 615.32 which is extremely close

to the calculated weight 615.31 of oxidized FMRF-NH2 (the

calculated molecular ion of oxidized FMRF-OH is 616.31).

Thus, the FaRP of Nereis virens is FMRFamide itself.


The central nervous system of Nereis virens consists

of a brain (supraesophageal ganglion) connected to a

subesophageal ganglion via the paired circumesophageal

connectives (Figure 2-3). The subesophageal ganglion is

continuous with the ventral nerve cord which joins the

intersegmental ganglia. The subesophageal ganglion itself

is composed of three fused intersegmental ganglia.

Immunoreactive FMRFamide was found in cell somas

throughout the central nervous system. These bodies were

distributed in a bilaterally symmetrical pattern in all

ganglia (supraesophageal, subesophageal, and

intersegmental). The cell bodies ranged from 10 to 85 nm

in diameter, depending upon their locations, and each

ganglion appeared to have at least 50 or more

immunoreactive cells. Assuming that the anterior 3 cm of

the worm contained 1000 spherical cells in the

circumesophageal ring and ventral nerve cord, an average

cell diameter of about 35 pm, and an intracellular

concentration of about imM FMRFamide, the total amount of

peptide in this section of the worm would have been about

180 fmol. This estimation was within the range determined

by RIA of acetone extracts (i.e., 100-600 fmol/worm).

Thus, the immunohistochemical data were consistent with

those from peptide isolation.


circumesophageal -Ke
nerve ring

-- ---- subesophageal

--- intersegmental
g a ganglion

Figure 2-3. The central nervous system of Nereis virens.
At the anterior end is the circumesophageal nerve ring
composed of the brain (supraesophageal ganglion), the
subesophageal ganglion, and the circumesophageal
connectives. Posteriorly, the subesophageal ganglion is
continuous with the ventral nerve cord which comprises the
intersegmental ganglia and connectives.

In the brain (supraesophageal ganglion), both large

and small cell bodies were stained, as were processes and

varicosities in the neuropile (Figures 2-4a,b). The cell

bodies were localized in the periphery of the ganglion and

ranged in size from 10 to 70 jim. Immunoreactive

varicosities and processes were seen in the connectives

which lead to the subesophageal ganglion and some of the

cephalic nerves.

Holmgren (1916) divided the nereid brain into 26

paired ganglionic cell groups or nuclei based upon their

location in frontal sections. The first three nuclei

densely packed mushroom shaped groups of cells referred to

as Globuli I, II, III; the remaining 23 nuclei are

non-mushroom shaped groups of cells and are designated by

arabic numbers. When the outlines of these nuclei (taken

from Holmgren) are superimposed upon the pattern of

immunoreactive cell bodies in the nereid brain (Figure

2-5) the FMRFamidergic cells appear in Globuli I and III

and nuclei 4, 5, 6, 7, 8, 10, 14, 15, 16, 17, 18, 19, 20,

22, and 23. FMRFamidergic cells may also be in nuclei 9

which is overlapped by nuclei 14, but frontal sections

will be required to determine the exact nuclear location

of the FMRFamidergic cell bodies.

Many immunoreactive cell bodies were present in the

subesophageal ganglion (Figures 2-6). Also stained were

the processes and varicosities in the ganglionic neuropile

Figure 2-4. Photomicrographs of FMRFamide immunoreactivity
in the brain (supraesophageal ganglion) of Nereis virens.
a. The ir-FMRFamide cell bodies and varicosities present
in a wholemounted brain. Scale bar: 100 jim. b. A cross
section of the brain showing immunoreactivity in both the
cell bodies and neuropile. Scale bar: 100 im. c. A cross
section of the brain stained with preabsorbed FMRFamide
antiserum. The fluorescence seen the upper left corner of
2-4b and c is non-specific fluorescence of the body wall
covering the brain. Scale bar: 100 gm.


nil, II





Figure 2-5. The positions of the 26 paired nuclei of the
brain (supraesophageal ganglion) superimposed on the
pattern of FMRFamide imimunoreactive cell bodies. Only one
of the symmetrically distributed, paired nuclei are shown
on the right side of the ganglion. The first three
nuclei, referred to as Globuli I, II, and III are marked
in Roman numerals, while the remaining nuclei are marked
in arabic numbers. The roots of some of the larger
supraesophageal ganglionic nerves (nI, nil, nVI, nVII,
nVIII, and nX ) and the connective to the subesophageal
ganglion (con) are shown. (The outline of the nuclei was
adapted from Holmgren, 1916).

Figure 2-6. Photomicrographs of the FMRFamide
immunoreactivity in the subesophageal ganglion and ventral
nerve cord of Nereis virens. a. A wholemounted
subesophageal ganglion and ventral nerve cord showing
ir-FMRFamide cell bodies and varicosities. Scale bar:
100 im. b. A close up of another wholemounted
subesophageal ganglion showing FMRFamide immunoreactivity
in the circumesophageal connectives. Scale bar: 100 gm.
c. A close up of the last fused segment of the
subesophageal ganglion in 2-6b, showing ir-FMRFamide cell
bodies and processes. Scale bar: 100 gm.


and the connectives to the supraesophageal ganglion

(Figure 2-6b). The cell bodies ranged in size from 10 to

60 pm.

The intersegmental ganglia of the ventral nerve cord

contained immunoreactive cell bodies and processes, both

contralateral and longitudinal (Figures 2-6a, 2-7a,c).

Cell body sizes ranged from 15 to 85 pm. Four major

nerves leave from each ganglion (Smith, 1955, 1957) and

FMRFamidergic fibers were present in each one. The

interganglionic connectives also displayed many

longitudinal FMRFamidergic processes.

Immunoreactive varicosities were also found in the

gut (Figures 2-8a,b), parapodium (Figure 2-9a), body wall

(Figure 2-9b), and palps (Figure 2-9c). The gut also

contained a few cell bodies (Figure 2-8c).


The isolated nereid esophagus was spontaneously

active, displaying a complex pattern of rhythmic

contractions (Fig 2-10). Moderate to high doses of

FMRFamide relaxed three of the four spontaneously active

preparations studied, but the effects of low

concentrations were difficult to ascertain because the

spontaneous relaxations of the tissue occur unpredictably.

Therefore, the esophagus was pre-contracted either by

application of excitatory neurotransmitters, or transmural

stimulation. Acetylcholine and serotonin caused small

Figure 2-7. Photomicrographs of immunoreactive FMRFamide
in a wholemount and cross section of an intersegmental
ganglion of Nereis virens. a. A cross section of an
intersegmental ganglion showing the presence of FMRFamide
in cell bodies and varicosities in the neuropile. Scale
bar: 100 p-m. b. A wholemount of an intersegmental
ganglion showing FMRFamide immunoreactive processes
projecting contralaterally through the ganglion.
Scale bar: 100 pm.


Figure 2-8. Photomicrographs of FMRFamide immunoreactivity
in the gut of Nereis virens. a. A sagittal section of the
esophagus showing FMRFamide immunoreactive varicosities.
Scale bar: 50 uim. b. A cross section of the gut showing
ir-FMRFamide in varicosities along the muscular layer. The
arrows point out some of the varicosities in the in the
nerve fiber. Scale bar: 50 gm. c. FMRFamide
immunoreactive cell bodies in the same cross section of
the gut as in 2-8b. The arrows are pointing to these cell
bodies. Scale bar: 50 nm.


Figure 2-9. Photomicrographs of FMRFamide immunoreactive
varicosities in the peripheral tissue of Nereis virens.
a. Ir-FMRFamide varicosities in a cross section of a
parapodium. Some of the varicosities along the path of the
fiber are indicated by the three arrows. The brightly
fluorescing objects below the nerve fiber are setae.
Scale bar: 50 pm. b. FMRFamide immunoreactive
varicosities in a cross section of the body wall. Some of
the varicosities in the two fibers shown in this photo are
indicated by the three arrows. Scale bar: 50 pmn.
c. FMRFamide immunoreactive varicosities in a cross
section of a cephalic palp. Some of the varicosities
along the nerve fiber are pointed to by the three arrows).
Scale bar: 50 gm.


I 7
10-8 M

3xl 08 M i

107 M

3x10.7 M

3x10-6 M

Wx 0- 6 M

105 M

1 min
Figure 2-10. The effects of FMRFamide on the isolated,
spontaneously active esophagus of Nereis virens. The down
arrow indicates that FMRFamide was added to the tissue
bath. The downward movement of the trace indicates a
relaxation of the esophagus, while an upward movement
indicates a contraction.

transient esophageal contractions that were not useful for

the purposes of the bioassay (data not shown). But

transmural electrical stimulation (2 Hz, 1 ms, 20 V)

caused a more tonic increase in muscle tone, and was

therefore used to contract the esophagus in the remaining

experiments, although spontaneous relaxation still

occurred in three of the eight preparations examined.

In both the spontaneously active and electrically

stimulated preparations, FMRFamide relaxed the muscle tone

with a threshold between 30 and 300 nM (Figures 2-10 and

2-11). The slope of the initial phase of relaxation of

the electrically stimulated muscle increased in a

dose-dependent manner (Figure 2-12) and the EC50 was

1.550.60 p-M (N=5). Addition of neither the serotonin

antagonist methysergide (UML) nor the acetylcholine

antagonist benzoquinonium modified the effects of the

peptide (Figure 2-13).

The tetrapeptide FMRFamide has now been identified

unequivocally in a non-molluscan species. This is the

first time that exactly the same neuropeptide sequence has

been identified in two different invertebrate phyla. The

levels of FMRFamide in Nereis and molluscs are, however,

quite different: i.e., 100-600 fmol per worm versus more

than 10 pmol per animal in the gastropod mollusc Aplysia

brasiliana (Lehman et al., 1984), a difference of 20-100

Figure 2-11. The effects of FMRFamide on the isolated,
electrically stimulated esophagus of Nereis virens. The
line under the trace indicates the duration of the
electrical stimulation (2 Hz, 1 ms, 20 V). The down arrow
indicates the addition of FMRFamide to the tissue bath.
The up arrow indicates the washing of the tissue.
Controls, electrical stimulation alone, were performed
before and after the series of FMRFamide doses were given.


AA^A.-^,A A-'

Control 1 .

108 M AAA-fVi fW /.,
"-''- """"I--'" \

l 0-7 M A^ ''. .

10-5 m
h I
3x10O7 M .

3xl 0.6 M ''- ,,,r.

Control 2 M . "..,._

1--- -----

120 -

100 3 3

0 80 -

a) 60
o m! i \

I 40 -


10-7 10-6 10-5

FMRFamide (M)

Figure 2-12. Dose-response curves of five different
Nereis showing the effects of FMRFamide on the rate of
relaxation of each electrically-stimulated esophagus. The
slope of the initial relaxation of the esophagus was
determined for each dose of peptide and then normalized to
the maximum slope attained in that preparation. All
points are adjusted to the maximal rate of relaxation for
each animal. Each symbol represents one animal.

3x10-6 M UML 3x10-6 M Fa

3x 10-6 M BQ 3x10-6 M Fa

3x10-6 M Fa

1 min
Figure 2-13. The effects of methysergide (UML) and
benzoquinonium (BQ) on FMRFamide relaxation of the
spontaneously active esophagus of Nereis virens. The down
arrows indicate the addition of the antagonists and
FMRFamide. The downward movement of the trace indicates a
relaxation of the esophagus, while an upward movement
indicates a contraction.

fold. This difference may be real, or it may reflect the

difference in animal size, since the values were not

normalized to the weight of tissue or the level of

protein. Another phyletic difference is that FLRFamide,

though present in all molluscs (Price and Greenberg,

1989), was not detected in the worms. Yet the presence of

this peptide (or one or more other analogs) in Nereis

cannot be ruled out; as in Aplysia, it may be present at

very low levels relative to FMRFamide (Price and

Greenberg, 1989).

All of the known genes encoding FaRP precursors

contain two or more different peptides, and for at least

one of these analogs, multiple copies are encoded

(Schaefer et al., 1985; Taussig and Scheller, 1986; Nambu

et al., 1988; Schneider and Taghert, 1988; Linacre et al.,

1989; Lutz et al., 1990; Taghert and Schneider, 1990).

Therefore, the Nereis gene probably also encodes for

FMRFamide and one or more analogs. If the ratio of

FLRFamide (or any other analog) to FMRFamide is 1/28-1/5

in Nereis, as it is in molluscs, then the level of this

peptide would be 3.5-120 fmol per worm and it would

probably not be seen. In summary, then other analogs in

the precursor might not be detectable.

Phylogenetically the annelids, molluscs, and

arthropods have many characteristics which suggest that

they have a common evolutionary lineage. The data from

Nereis are consistent with this hypothesis and suggest

that the FMRFamide tetrapeptide gene may have arisen in

the ancestry of that lineage.

The presence of ir-FMRFamide in the cell bodies and

processes of the circumesophageal nerve ring and ventral

nerve cord agrees with the preliminary data of Porchet and

Dhainaut-Courtois (1988). They showed that immunoreactive

cell bodies occurred in the supraesophageal ganglion of

Nereis diversicolor in nuclei 7, 10, and 13.

Many of the 26 nuclei in the nereid brain contain

both secretary and non-secretory cell types, based upon

the selective staining of secretary cells with paraldehyde

fuschin (Golding, 1967; Engelhardt et al., 1982).

Staining the same nereid tissue sections with both

paraldehyde fuschin and antibodies to the vertebrate

peptide cholecystokinin (CCK), Engelhardt et al. (1982)

showed that CCK immunoreactivity was associated with both

cell types, and suggested that a CCK-like peptide might

function as both a neurotransmitter and neurohormone.

FMRFamidergic cells appear to be in some of the same

nuclei, so they may also have both a neurohumoral and a

neurotransmitter function, although further investigation

would be required to confirm this.

The intersegmental ganglia each contain about 200

neurons in a crescent-shaped cortex along the ventral and

lateral edges of the nerve cord (Smith, 1957). Of these

200 cells, only 7 pairs are motor neurons; the remainder

are interneurons with their processes in the ganglionic

neuropile. Since these ganglia contain at least 40

immunoreactive cell bodies, some of them must be FMRFamide

containing interneurons.

Motor neurons in the intersegmental ganglia may also

be FMRFamidergic. To confirm this, however, the entire

neuron must be traced from its cell body in the cortex,

along its axon in the neuropile, to the ganglionic nerve

through which the axon exits, and to the target organ. No

cells could be so traced, but FMRFamidergic processes do

cross the ganglia contralaterally, as some motor neurons

have been shown to do (Smith, 1955; 1957). Some

immunoreactive processes do appear in each of the four

ganglionic nerves, but these could either be sensory or

motor fibers. Nevertheless, the occurrence of

FMRFamidergic varicosities in the body wall muscles

strongly suggests that at least some of these fibers are

motor in function.

All of the FMRFamide extracted in acetone came from

the first 3 cm of the animal and the levels observed were

consistent with those estimated from the

immunohistochemical data. When the remaining 17 cm of the

animal were extracted, however, no peptide was found,

suggesting that the level of peptide declined toward the

more posterior ganglia. Indeed, this distribution pattern

appears to occur in two other annelids; i.e., the number

of ganglionic immunoreactive cells decreased toward the

tail in both the leech (Kuhlman et al., 1985a; Calabrese

and Norris, 1990) and the earthworm (Fujii, 1989).

The anatomical and physiological data indicate that

FMRFamide is involved in the control of nereid intestinal

motility. FMRFamidergic varicosities and cell bodies were

observed in the gut. The polychaete somatogastric nervous

system is composed of a network of ganglionic and sensory

neurons, including their processes in the wall of the gut

(Whitear, 1953). The functions of the cells containing

the peptide is unclear. In the sedentary, burrow-dwelling

polychaete, Arenicola, the esophagus is innervated by the

somatogastric nervous system, and when it is isolated from

the rest of the worm, the tissue contracts with a distinct

rhythm suggesting the presence of an endogenous pacemaker

(Wells, 1937). In the nereid esophagus, FMRFamide at low

concentrations inhibits the rhythmicity of some

preparations, suggesting that the peptide may be involved

in modulation of the pacemaker. The failure of

serotonergic and cholinergic antagonists to inhibit the

response to FMRFamide suggests that the peptide is not

acting on the esophagus by releasing serotonin or

acetylcholine from nerve terminals. The involvement of

FMRFamide in intestinal motility may be a general function

in annelids since FMRFamidergic fibers are present in the


gut of the earthworm Eisenia foetida (Fujii et al., 1989).

In the closely related phylum Mollusca, FMRFamide

inhibits the feeding and gastric motility of gastropods by

acting within the CNS (central nervous system) as well as

the musculature of the alimentary tract (Krajniak et al.,

1989). The involvement of FMRFamide in feeding at the

level of the CNS in annelids remains to be studied.

The localization of FMRFamidergic varicosities in the

muscles of the body wall could suggest that the peptide

plays a role in body movement. But it could further the

proposed role for FMRFamide in the intestinal motility.

In Nereis, the musculature of the body wall (as well as

that of the gut) is involved in the propulsion of food

through the animal. The septal muscles dilate the

intestine and cooperate with the suspensory muscles and

body wall muscles in the complicated intestinal movements

(Michel and Devillez, 1978). In any event, FMRFamide

immunoreactivity has been found in the body walls of

Hirudo medicinalis (Kuhlman et al., 1985a), Sabellastarte

magnifica (Diaz-Miranda et al., 1989), and Eisenia foetida

(Fujii et al., 1989). The effects of FMRFamide on the

body wall of Nereis were not examined because its

attachment to the intestine via the septa between each

body segment precludes a clean, isolated body wall

preparation. In S. magnifica, however, an isolated body

wall preparation has been made, and FMRFamide relaxes a

dopamine-induced contraction of musculature (Diaz-Miranda

et al., 1989).

FMRFamide immunoreactivity was also observed in the

palps and in the supraesophageal ganglionic nerve roots

that supply them. The palps are sensory structures, but

the FMRFamidergic varicosities within them could be

associated with motor neurons or interneurons, as well as

sensory neurons.




The occurrence of FMRFamide-related peptides (FaRPs)

in the superphylum Arthropoda is now firmly established.

Five peptides, all N-terminal extensions of the

tetrapeptide core (FXRFamide, where X is Leu, Met, or lie)

have been isolated and sequenced. Another 12 peptides are

encoded in the genes of two species of Drosophila.

The arthropodan FaRPs can be divided further into

three structurally distinct groups, two in insects and

another in crustaceans and insects (Table 3-1). Group 1

includes two insect peptides, one from the cockroach

Leucophaea maderae (Holman et al., 1986) and another from

the locust Schistocerca gregaria (Robb et al., 1989);

both contain the common C-terminal core, Val-Phe-Leu-Arg-

Phe-NH2. Group 2 includes peptides from two species of

fruit fly, Drosophila melanogaster and D. virilis (Nambu

et al., 1988; Taghert and Schneider, 1990) which share the

C-terminal sequence Asp-Phe-X-Arg-Phe-NH2 (where X is

either Met or Val). Group 3 comprises peptides from the

two species of Drosophila and the crustacean Homarus


Nuclear Sub-Family Sequence Species

1 pGlu-Asp-Val-Asp-His-Val-Phe-Leu-Arg-Phe-NH2 L. maderae
Pro-Asp-Val-Asp-His-Val-Phe-Leu-Arg-Phe-NH2 S. gregaria

2 Asp-Pro-Lys-Gln-Asp-Phe-Met-Arg-Phe-NH2* D. melanogaster/
Thr-Pro-Ala-Glu-Asp-Phe-Met-Arg-Phe-NH2 D. virilis

3 Ser-Asp-Asn-Phe-Met-Arg-Phe-NH2 D. melanogaster/
Pro-Asp-Asn-Phe-Met-Arg-Phe-NH2 D. virilis

Ser-Asp-Arg-Asn-Phe-Leu-Arg-Phe-NH2 H. americanus
*The only Drosophila gene product which has also been isolated
and sequenced as a peptide. All other Drosophila peptides listed
are predicted by the gene.


americanus (Trimmer et al., 1987); they contain the common

C-terminal sequence Asn-Phe-X-Arg-Phe-NH2 (where X is

Met, Leu, or lie).

The arthropodan FaRPs have many biological effects.

In insects, the peptides of group 2 potentiate the twitch

ofi/the neurally evoked locust extensor tibiae muscle

(Cuthbert and Evans, 1989) and inhibit the contractions of

cockroach hindgut (Holman et al., 1986) and the locust

heart (Cuthbert and Evans, 1989; Robb et al, 1989). The

group 3 FaRPs stimulate the lobster heart (Trimmer et

al.,1987), modulate crayfish neuromuscular synapses

(Mercier et al., 1989), and stimulate the crab

somatogastric ganglion (Weimann and Marder, 1989). Hooper

and Marder (1984) showed that FMRFamide, a

non-arthropodan FaRP, also stimulated the somatogastric


Four different insects have been examined and FaRPs

representative of all three arthropodan peptide groups

have been found. But only one crustacean has been

studied, and only peptides from group 3 were found

(Trimmer et al., 1987). I therefore chose to study a

second decapod crustacean to determine the sequence

variability within this order. I hoped further to

determine whether crustaceans would also contain peptides

with the characteristics of groups 1 or 2.

The lobster FaRPs were isolated from the pericardial

organs (Trimmer et al., 1987), and the highest

concentrations of ir-FMRFamide throughout the entire

nervous system were in these organs (Kobierski et al.,

1987). The cells innervating the pericardial organs have

their somas in the thoracic ganglion; the organs

themselves are neurohemal release sites that lie in the

venous cavity surrounding the crustacean heart (Cooke and

Sullivan, 1982) (Figure 3-1). Hemolymph flows through the

gossamer structures of the pericardial organs and into the

heart, making them the perfect release point for

cardioactive neurohormones. Many other cardioexcitatory

substances are found in this tissue, including such

peptides as, proctolin (Sullivan, 1979) and crustacean

cardioexcitatory peptide (CCAP)(Stangier et al., 1987),

and the classic neurotransmitters serotonin (Beltz and

Kravitz, 1983), octopamine (Livingstone et al., 1981), and

dopamine (Cooke and Sullivan, 1982).

Therefore, I have isolated and sequenced peptides

from the pericardial organs and thoracic ganglia of the

blue crab Callinectes sapidus, which is commercially

available in large numbers. Furthermore, since Trimmer et

al. (1987) indicated that the Homarus FaRPs were

cardioexcitatory, the structure-activity relationship of

the blue crab cardiac receptor was examined.


Figure 3-1. A schematic of the blue crab Callinectes
sapidus. The drawing is a dorsal view of an animal with
part of its dorsal carapace removed showing the placement
of the heart, thoracic ganglion, pericardial organs and
gills (redrawn after Maynard, 1961).




Blue crabs, Callinectes sapidus, were purchased from

a commercial supplier and maintained in flowing, filtered

seawater for at least one week prior to experimentation.

The animals were fed squid at least twice a week.

Peptide Extraction, Purification, and SequencinQ

The pericardial organs or thoracic ganglia were

dissected from the crabs and placed in 10 volumes of

acetone. The mixture was maintained at -20 C and kept

until the tissues from 100 to 500 animals had accumulated.

Once the desired amount of tissue was acquired, the

acetone was decanted, and the extract was processed as

described in Chapter 2.

The processed aqueous residue was loaded onto either

a Waters Novapak C-18 reverse-phase HPLC column (3.9 x 150

mm) with a flow rate of 2 ml/min, or a Brownlee Prep 10

Aquapore Octyl C-8 reverse-phase HPLC column (10 x 100 mm)

with a flow rate of 4 ml/min. The columns were eluted

with a gradient of ACN (0 to 40% in 30 min) containing

0.1% TFA. Fractions were collected every 30 s, and a 2 il

aliquot from each was assayed in an RIA for FMRFamide; the

assay employed both the S253 and Q2 antisera (details in

Chapter 2).

The fractions from the first separation were further

purified on either the Waters Novapak C-18 HPLC column or

a Brownlee RP-300 Aquapore Octyl reverse-phase HPLC column

(2.1 x 220 mm) with a flow rate of 0.5ml/min. A second

solvent system, of either isopropanol with 0.1% TFA (10 to

40% in 20 min) or ACN with 0.1% heptafluorobutyric acid

(HFBA) (16 to 32% in 20 min), was used in alternation with

the ACN/TFA system (16 to 32% in 20 min) on these columns

until each immunoreactive peak was coincident with a

single peak at 210 nm absorbance.

Material form the peaks of the purified peptide were

submitted for microsequencing to the Protein Chemistry

Laboratory of the University of Florida Interdisciplinary

Center for Biotechnology Research, and for FAB-mass

spectroscopy to Dr. T.D. Lee, Division of Immunology,

Beckman Research Institute of the City of Hope, Duarte,



A crab was removed from the holding tank and placed

in ice for at least 30 min. Once anesthetized, the dorsal

carapace was opened and the heart removed. With the aid

of dissecting microscope, I secured a cannula was secured

in the sternal artery with thread. The cannula was

connected with the reservoir of crab saline suspended

above a level of the heart. Most of the other arteries

were ligated, but the posterior aortas were left patent so

that the heart could empty during systole. The heart was

suspended in a temperature controlled tissue bath (Figure


3-2) maintained at 15 C, and perfused with a crustacean

saline at a constant pressure of 42 cm H20 (4.2 kPa) and a

flow rate of 3 ml/min (Fig. 3-2). The saline contained

458.6 mM sodium chloride, 13.6 mM potassium chloride, 13.4

mM calcium chloride, 13.6 mM magnesium chloride, 1.4 mM

sodium sulfate, 3 mM HEPES (N-[2-Hydroxyethyl]piperazine-

N'-[2-ethanesulfonic acid]), adjusted to pH 7.4 with

sodium hydroxide (Mulloney and Selverston, 1974).

Contractions of the heart were recorded with a Grass

force/displacement transducer attached to the anterior

aorta and monitored on a Grass recorder.

All peptides were diluted to the desired

concentration in crab saline (final volume of 0.5 ml) and

were introduced into the heart via the sample loop.

Addition of peptide by this method did not disturb the

flow of saline to the heart. Prior to the addition of any

peptides, 0.5 ml of saline were loaded into the sample

loop and injected to ensure that the mechanical components

of the system had no effect on the heart. Each heart was

exposed to a test peptide and the standard,

GYNRSFLRFamide. Starting with the lowest concentrations,

samples of these two were added alternately. The sample

loop was replaced after each peptide, and saline was

injected prior to the next peptide to control the

Figure 3-2. The perfused Callinectes heart isolated in a
bioassay apparatus. The crab heart is suspended in a
15 C tissue bath and is connected to the saline reservoir
by a cannula through its thoracic artery. The saline
reservoir is a Mariotte bottle which kept the perfusion
pressure constant at 42 cm HO20 (4.2 kPa) and the flow rate
at 3 ml/min. The cannula is connected to both the sample
loop and the shunt loop which are regulated by the two
valves. While the shunt loop is open to the heart, the
sample loop which is open to the waste reservoir can be
filled with peptide without disturbing the perfusion. The
sample loop is then closed off to the waste line and the
shunt loop is closed off to the heart without interrupting
saline flow rate or pressure. The perfusate passes through
the heart and into the inner chamber of the tissue bath to
help buoyantly support the heart and maintain the
temperature. The bottom of the inner chamber is connected
via rubber tubing to the same Mariotte bottle saline
reservoir so that the entire chamber can be flushed. A
needle attached to a source of compressed air is inserted
into this rubber tubing to oxygenate the tissue. Excess
waste water flows out of the top of the chamber to a waste
reservoir. The anterior aorta is connected to the
force/displacement transducer which allows the
contractions to be recorded.








- Waste




possibility that some of the previous dose was still in

the line.

The peptides always affected heart rate and the

change in frequency was therefore the parameter used to

quantify the effects of each peptide. When comparing a

peptide to GYNRSFLRFamide, doses of each peptide were used

to construct the linear portions of their curves around

their ECs50s. The relative potency of the test peptide to

GYNRSFLRFamide was determined by calculating the

difference between the two lines at the EC50 (Tallarida

and Murray, 1981). In some experiments full dose-response

curves were generated, and the maximal effect of the test

peptide was compared with that of GYNRSFLRFamide, so that

the relative maximal effect (efficacy) could be


Synthetic peptides were either purchased from

Peninsula, Sigma, or synthesized by Dr. Ben Dunn and

Alicia Alvarez of the Protein Chemistry Laboratory of the

University of Florida Interdisciplinary Center for

Biotechnology Research.


Peptide Extraction, Purification, and Sequencinc

In one experiment with 512 animals, a peak of

immunoreactivity with absorbance at both 210 nm and 280 nm

was purified (Figure 3-3). When analyzed by FAB-mass

spectrometry, it yielded a major ion at m/z 1159. Thus,

Figure 3-3. The HPLC purification and FAB-mass spectrum of
the Callinectes FaRP isolated from 512 animals, a. The UV
absorbance at 210 nm (solid line) and 280 run (dotted line)
of the final HPLC purification step. The dashed line shows
the gradient of ACN with 0.1% HFBA. b. The amount of FaRP
determined by RIA with antisera Q2 and S253 in the UV
absorbance peak in 3-3a. The lag between the peaks of UV
absorbance and immunoreactivity reflects the dead space
between the detector and the fraction collector, c. Mass
spectrometry of the irmmunoreactive fraction in 3-3b,
showing a normal (magnetic sector scan) positive-ion
FAB- mass spectrum for the peptide. The other ions in the
spectrum are due to sample matrix.


j 210 nm
// 1"" 280 nm

80 1

C 601

-E 0
> 40 -. .- ..

0 I0 20

Time (min)



E 0.04- |1

A 0
>, 18 20 22
0.12 S5

700 C '0 2 I
6Du E
c 0.04
c 60"
<0 0
40" 18 20 22
> 40 Time (mm)

:: 20-
1 159

700 800 900 1000 1100 1200 1300 1400

the peak contained an immunoreactive peptide with a

molecular weight of 1158.

In another experiment (200 animals) a similar

immunoreactive peak with absorbance at 210 and 280 nm was

observed (Figure 3-4). It eluted off the column at the

same time as the peak in the previous experiment when the

same HPLC gradients and buffer systems were used (data not

shown). (Note that different buffer systems and gradients

were used in the experiments illustrated in Figures 3-3a

and 3-4a, so the peaks elute at different times.) When

the purified peak was sequenced, the identity of the first

amino acid was unclear because the levels of all amino

acids in this cycle were high, including that of glycine,

a common contaminant (Figure 3-4a). The remainder of the

sequence was determined and indicated a peptide with the

partial sequence X-Tyr-Asn-Ser-Phe-Leu-Arg, where X

represents an unknown amino acid in the first position

(Fig 3-4a). This incomplete sequence also lacked a

C-terminal phenylalanine-amide which is seen in all other

FaRPs. The calculated molecular weight of this fragment

was 955, and the difference between it and the observed

molecular weight of 1158 (taken from the peak in the

previous experiment) was 203.

Only two amino acid combinations, plus an amide,

could account for this difference: (cysteine, threonine)

and (glycine, phenylalanine). Cysteine was not present in

Figure 3-4. The HPLC purification and sequence of the
Callinectes FaRP from 200 animals, a. The UV absorbance
at 210 nm (solid line) and 280 runm (dotted line) of the
final HPLC purification step of the FaRP. The dashed line
shows the gradient of ACN with 0.1% TFA. b. The amount of
FaRP determined by RIA with antiserum Q2 in the UV peak of
3-4a. The lag between the peaks of UV absorbance and
immunoreactivity reflects the dead space between the
detector and the fraction collector, c. The microsequence
of the immunoreactive fraction in 3-4b, showing the
changes in glycine, tyrosine, asparagine, arginine,
serine, phenylalanine, and leucine in each cycle.


S 1 210 Onm
80- ...... 280 nm

o 60 .

40 -------------


0 5 10 15 20 25

Time (min)
S0.5- b
E 0.4


B 0.2-
c 0.1 -
E 0
0 5 15 20 25
Time (min)

20 C

15- 01y

o 10
E \Tyr

5he Asn Ar

1 2 3 4 5 6 7 8 9 10

Sequencer Cycle

any of the sequencer cycles, so the missing residues must

be glycyl and phenylalanyl. Thus the peptide sequence

predicted was Gly-Tyr-Asn-Arg-Ser-Phe-Leu-Arg-Phe-NH2

(GYNRSFLRFamide). The presence of the C-terminal Phe-NH2

was further indicated by the binding of the peptide to

antiserum S253, which requires an amidated phenylalanine

in that position (Price, 1983; Price 1987; Greenberg et

al., 1988). Furthermore, the sequence Phe-Leu-Arg-Phe was

found in yet another purification of pericardial organs,

and could be a breakdown product of GYNRSFLRFamide (Table

3-1). The predicted weight of the protonated peptide is

1158.6, which is very close to the observed ion of 1159.

Synthetic GYNRSFLRFamide has the same elution time on the

HPLC as the native peptide, when either the ACN/TFA or the

ACN/HFBA buffer systems are used.

The levels of acetone-extracted peptide are very low

in these animals. In the 200 animal experiment, the

initial yield of peptide estimated by RIA in the

GYNRSFLRFamide peak is 2.67 nmol (13.4 pmol/animal). The

purified peak yielded only 410 pmol by RIA with antiserum

Q2 (2 pmol/animal), 8 pmoles by UV absorbance

(40 fmol/animal), and 7 pmol by microsequencing

(35 fmol/animal). In the 512 animal experiment, the

initial peak was 3.4 nmoles (6.6 pmol/animal) by RIA. The

purified peak contained 113 pmol (50 pmol with Q2) by RIA

(from 98 to 220 fmol/animal) and 12 pmol by UV absorbance


(23 fmol/animal). One reason for the such a small yield

of peptide may be the method of extraction. Trimmer et al.

(1987) reported that acidified methanol extracted more

pericardial organ peptides than pure acetone. However,

acidified methanol extracted more impurities (indicated by

the increased amount of UV absorbance during the initial

HPLC purification step) and required more purification

steps with associated losses in peptide.

In addition to the experiment described above,

multiple batches of pericardial organs and thoracic

ganglia were extracted, purified, and sequenced. In most

cases, the initial separation of the extract yielded at

least two peaks of FMRFamide immunoreactivity. Upon

completion of the HPLC purification, partial amino acid

sequences were observed in all instances (Table 3-2). The

array of sequences suggests that the crab contains several

different analogs of GYNRSFLRFamide; and that the

sequences Phe-Leu-Arg-Phe, Asn-Phe-Leu-Arg-Phe, and

Ser-Phe-Leu-Arg are probably fragments of these larger

native peptides.

Structure-Activity Relationship of the Cardiac Receptor

The isolated heart of Callinectes sapidus beat

spontaneously in the tissue bath at 15 C. The average

resting heart rate prior to any peptide administration was

26.61.0 beats per minute (beats/min) (N=132). The

average heart rate of intact animals at 15 C is 62.310.0















organ X -Tyr-Asn-Arg-Ser-Phe-Leu-Arg

organ Ser-Phe-Leu-Arg

organ X X -His-Lys-Asn-Tyr-Leu-Arg-Phe

organ X X -Met-Asn-Phe-Leu-Arg

organ Gly-Asn-Arg-Asn-Phe-Leu-Arg

organ X -Arg-Asn-Phe-Leu-Arg

organ Arg-Asn-Phe-Leu-Arg

organ X X -Asn-Phe-Leu-Arg

organ Asn-Phe-Leu-Arg-Phe

organ/Thoracic ganglion Phe-Leu-Arg-Phe

* X represents an unknown amino acid in the sequence.

Figure 3-5. The effects of increasing doses of
GYNRSFLRFamide on the isolated heart of the blue crab
Callinectes sapidus. The figure shows the effects of
increasing doses (indicated on the left) of GYNRSFLRFamide
on heart rate and contraction amplitude. The down arrow
indicates when the sample loop was opened to the heart.

10-9M ,

l, ll (i, i ll!l, i, i1l llli.lllll




3x10-7 M ,

10-6 M


10-5 M
,,!l!t~lt~~ f~isgsmmm

1 min


beats/min (N=6) (deFur and Mangum, 1979). Such

differences are to be expected, since both heart rate and

muscle tone vary with perfusion pressure (Maynard, 1960;

Kuramoto and Ebara, 1984; 1988), and the conditions under

which the heart is isolated do not approximate those in

the intact animal.

GYNRSFLRFamide excited the heart in a dose dependent

manner (Figure 3-5). The excitation always consisted of

an increase in heart rate. Contraction amplitude and tone

either increased, decreased, or remained unchanged.

Because of this inconsistency, heart rate was chosen as

the measured variable. GYNRSFLRFamide had a threshold of

between 10 and 30 nM and an ECs50 of 32362 nM (N = 11)

(Figure 3-6).

Sixteen analogs of GYNRSFLRFamide were tested on the

isolated crab heart bioassay (Table 3-3). All of the

C-terminally amidated FaRPs were cardioexcitatory,

increasing heart rate and changing amplitude and tone in a

manner similar to that of the native Callinectes peptide.

The Homarus peptide TNRNFLRFamide, the most potent analog,

was one order of magnitude more effective than

GYNRSFLRFamide. The other known lobster peptide

(SDRNFLRFamide) and three other synthetic N-terminally

extended FaRPs (VNRNFLRFamide, LRNFLRFamide, and

VNNFLRFamide) were roughly equipotent with the crab FaRP.

,N 100


l 8O -


I I i

20 0

o .0

10-9 10-~8 10-7 10-6 10-5


Figure 3-6. A dose response curve, showing the effects of
GYNRSFLRFamide on the heart of Callinectes sapidus. The
curve shows the change in heart rate plotted against the
concentration of GYNRSFLRFamide injected into the heart.
These data come from the experiment illustrated in
Figure 3-5.




1 2 3 4 5 6 7 8 9+










pGlu-Asn-Phe- I le-Arg-Phe-NH2









22.4 13.6 16

4.960.53 11

4.841.91 3

1.350.15 8

4.352.76 4

0.170.02 7

0.480.14 6

0.390.09 6

0.220.03 2

0.020.01 6

0.090.05 6

0.070.04 6

0.300.01 7

0.060.01 10

0.020.01 6

No Response* 3

a Standard error.
+ The numbers designate the amino acid positions in
GYNRSFLRFamide for the discussion of amino acid
* At concentrations up to 100 p-M.


Since the lobster peptides and at least one partially

sequenced Callinectes peptide (Table 3-2) have the

sequence RNFLRFamide in common, this peptide and several

different analogs with amino acid substitutions in either

position 4 or 5 (as numbered in GYNRSFLRFamide) were

examined, especially to study the importance of the

arginine and the asparagine. When analogs with the

sequence XNFLRFamide (where X is Arg, Gly, pGlu, or Glu)

were used, they were about an order of magnitude less

potent on the heart than GYNRSFLRFamide. The same was

true for the pentapeptide NFLRFamide. When the sequence

was changed to RXFLRFamide (where X is Leu or Gln), the

potency decreased further relative to GYNRSFLRFamide. But

note that the substitutions of glutamine and serine

substitutions are not nearly so adverse as that of the

leucine, which was expected since glutamine and serine

are, like asparagine, hydrophilic. Similarly the core

tetrapeptides, FLRFamide and its analog FMRFamide, were

also about two orders of magnitude less potent than

GYNRSFLRFamide. The unamidated tetrapeptide FLRF-OH had

no biological activity at concentrations up to 100 gM.

The relative efficacies of the analogs ranged from 1.25

for TNRNFLRFamide, to 0.80 for ENFLRFamide (Table 3-4),

indicating that all the peptides in this study could reach

approximately the same maximal response as GYNRSFLRFamide.



Gly-Tyr-Asn-Arg-Ser-Phe-Leu-Arg-Phe-NH2 1.00

Thr-Asn-Arg-Asn-Phe-Leu-Arg-Phe-NH2 1.25 1

Ser-Asp-Arg-Asn-Phe-Leu-Arg-Phe-NH2 1.19 3

Val-Asn-Arg-Asn-Phe-Leu-Arg-Phe-NH2 1.00 1

Leu-Arg-Asn-Phe-Leu-Arg-Phe-NH2 0.98 2

Val-Asn-Asn-Phe-Leu-Arg-Phe-NH2 1.00 1

Arg-Asn-Phe-Leu-Arg-Phe-NH2 0.84 3

Glu-Asn-Phe-Leu-Arg-Phe-NH2 0.80 3

Gly-Asn-Phe-Leu-Arg-Phe-NH2 N.D.+ 0

pGlu-Asn-Phe-Ile-Arg-Phe-NH2 N.D.+ 0

Arg-Leu-Phe-Leu-Arg-Phe-NH2 0.84 3

Arg-Gln-Phe-Leu-Arg-Phe-NH2 0.86 2

Arg-Ser-Phe-Leu-Arg-Phe-NH2 0.89 2

Asn-Phe-Leu-Arg-Phe-NH2 0.83 3

Phe-Leu-Arg-Phe-NH2 0.85 4

Phe-Met-Arg-Phe-NH2 0.86 2

Phe-Leu-Arg-Phe-OH N.R.* 3

* No response at concentrations up to 100 pM.
+ Not determined.



I report here the sequence of a new crustacean FaRP:

GYNRSFLRFamide. GYNRSFLRFamide is more similar in

structure to the two Homarus peptides, TNRNFLRFamide and

SDRNFLRFamide, than it is to other arthropodan peptides

isolated from insects, although it lacks the asparagine

present in all the group 3 peptides. Phyletically,

Homarus and Callinectes are closely related. They are

both decapods in the suborder Pleocyemata (Bowman and

Abele, 1982). They diverge at the next taxanomic

division, infraorder, Homarus belonging to the Astacidae,

and Callinectes to the Brachyura. The Astacidae

originated during the late Permian, while the Brachyura

originated more recently, during the Jurassic, a

difference of 100 million years (Schram, 1982). The class

Crustacea originated in the Cambrian (Barnes, 1974),

whereas the class Insecta arose much later, during the

Mississippian (Meglitsch, 1967), a difference of 265

million years.

The Drosophila genes encode for group 2 and 3

peptides which contain asparagine and aspartic acid,

respectively, in the position just adjacent to the

tetrapeptide core. The codons for both of these amino

acids differ by only a single nucleotide. In the blue

crab, the amino acids in the same position are serine and

asparagine (seen in the partial sequences), both of which

also differ by only one nucleotide in the genetic code.

Therefore, the gene encoding the FaRPs in the decapod

Crustacea may be similar to those of the fruit fly and the

pulmonate molluscs, containing copies of multiple

peptides. But the crustacean FaRPs have peptides with a

C-terminal SFLRFamide, as well as NFLRFamide.

The group 1 peptides are unlike those in crustaceans

and a single change in a codon cannot change a valine to

either a serine or an asparagine. The group 1 and 3

peptides also have very different biological effects. In

a single experiment (not shown in Table 3-3), the insect

peptide pQDVDHVFLRFamide had a threshold of 10,000 times

higher than that of the lobster peptide TNRNFLRFamide in

stimulating the heart. This difference is reciprocal; the

locust heart is potently inhibited by its native peptide,

whereas TNRNFLRFamide is cardioexcitatory (Cuthbert and

Evans, 1989; Robb et al., 1989).

The peptide isolation and identification data clearly

indicates that several different FaRPs are present in

Callinectes sapidus, including the new sequence

GYNRSFLRFamide and several partial ones with the

N-terminal sequence NFLRFamide. As with the peptides in

Homarus americanus (Trimmer et al., 1987), they are

N-terminal extensions of FLRFamide. However, there is

evidence for one peptide in which a tyrosine has been

substituted for the N-terminal phenylalanine

(X-X-His-Lys-Asn-Tyr-Leu-Arg-Phe) (Table 3-2). Two

peptides containing the C-terminal sequence YLRFamide have

been isolated from the gastropod mollusc, Helix aspersa

(Price et al., 1990).

Many of the blue crab FaRP sequences appear to be

fragmented, and Trimmer et al. (1987) also reported

finding a fragment of the TNRNFLRFamide sequence. Yet

different methods of extraction were employed with Homarus

(Trimmer et al., 1987) and Callinectes (present study).

If the crustaceans contained a hardy protease, which was

not inhibited by either acetone or methanol, then some

digestion of the peptides could occur during one of the

extraction steps. Also, an enzyme might coelute with the

peptide during the first HPLC separation and act at this

point, though it would have to be acetonitrile resistant

and function in the presence of TFA. An enzyme released

during dissection might degrade the peptide before the

pericardial organs and thoracic ganglia could be immersed

in the acetone. Another possible explanation is that the

Asn-Phe and Ser-Phe bonds are labile under the conditions

of extraction, purification, and storage. But the

nematode peptides, SDPNFLRFamide and SADPNFLRFamide were

processed and purified in the same manner in this

laboratory, and no fragments were observed.

The crustacean heart beat is neurogenic; a cardiac

ganglion innervates the myocardium and initiates the

contractions (Maynard, 1960; Hartline, 1967; Kuramoto and

Yamagishi, 1990). Nine neurons make up the lobster

cardiac ganglion, five large motor neurons at its anterior

end, and four small driver neurons posteriorly (Maynard,

1953; 1955; Hagiwara, 1961). Only the large neurons

innervate the myocardium; the small cells send processes

only to the large ones. Electrical recordings show that

the small cells trigger a burst of spike activity in the

large cells, suggesting that the small cells constitute

the endogenous pacemaker (Friesen, 1975). Extrinsic

excitatory and inhibitory neurons send their axons into

the cardiac ganglion and the myocardium and regulate the

beat (Hagiwara, 1961; Wiens, 1982). GYNRSFLRFamide

primarily increases heart rate; its effects on muscle tone

and amplitude seem to be indirect. Therefore, the peptide

probably acts on the pacemaker cells in the ganglion.

Studies on the heart and cardiac ganglion have been

performed with three other cardioactive products of the

pericardial organ. Proctolin, another peptide found in

pericardial organs, stimulates the intact, isolated

lobster heart and the isolated cardiac ganglion (Miller

and Sullivan, 1981; Sullivan and Miller, 1984). Serotonin

increases the burst frequency of the small neurons, the

heart rate, and to a lesser extent, the burst frequency of

the large motor neurons (Kuramoto and Yamagishi, 1990).


Dopamine also increases the burst frequency of the cardiac

ganglion (Miller et al., 1981; Cooke and Sullivan, 1982).

Peptide structure is related to the relative potency.

The tetrapeptides are very weak agonists. When the

asparagine is placed in position 5 (as numbered in

GYNRSFLRFamide) (Table 3-3), the pentapeptide potency is

only one order of magnitude less than GYNRSFLRFamide. The

further addition of any amino acid, including arginine, in

position 4 (as numbered in GYNRSFLRFamide) of this

pentapeptide does not greatly change the potency.

Arginine in position 4 is found in all native crustacean

peptides and therefore might be necessary for receptor

activation. When arginine is placed in position 4, and

the asparagine in position 5 is replaced by other

hydrophilic amino acids, like glutamine and serine, the

peptides are slightly less potent than RNFLRFamide.

However, when a hydrophobic amino acid like leucine is

substituted for the asparagine, the potency of the

resulting analog decreases by almost an order of magnitude

compared to RNFLRFamide. When the asparagine is retained

in position 5, but the arginine is replaced by either

glutamate, which is negatively charged rather than

positively charged, or glycine, which is small and

uncharged, or pyroglutamate, which is large and polar, the

analogs are equipotent with RNFLRFamide. When two or more

amino acids are added to the N-terminal of the

pentapeptide, NFLRFamide, it becomes equipotent with

GYNRSFLRFamide. Only one peptide TNRNFLRFamide is as much

as one order of magnitude more potent than GYNRSFLRFamide.

For full receptor activation, therefore, the peptide

should have the FLRFamide core, an asparagine in position

5, and amino acids in positions 3 and 4 (as numbered in


Some of the partially sequenced crab peptides have

the asparagine proximal to the FLRFamide core and are

expected to have full biological activity on the heart

bioassay. But GYNRSFLRFamide lacks the asparagine in

position 5, and yet it is equipotent with the analogs that

do possess an asparagine. In the native crab peptide,

serine occupies position 5. RSFLRFamide is about half as

potent as RNFLRFamide. A comparison of serine and

asparagine shows that the hydrogen of the hydroxyl group

in serine will have a slight positive charge, as will the

nitrogen in the amide group of asparagine (Roberts and

Caserio, 1977). If the amide nitrogen on the asparagine

binds to a nucleophilic region on the receptor, then the

hydrogen of the serine hydroxyl group might also be

positioned close enough to bind in the same way. This

could also explain why RQFLRFamide is also half as potent

as RNFLRFamide. When glutamine is placed in position 5,

its amide group is located one carbon atom farther away

from the peptide backbone than in asparagine, and its

nitrogen would therefore be too far away to bind optimally

to that nucleophilic region of the receptor. Thus,

peptides like GYNRSFLRFamide, with serine in position 5,

would be equipotent with peptides containing asparagine in

position 5.



Both the polychaete annelid Nereis virens, and the

crustacean Callinectes sapidus, contain FMRFamide-related

peptides. In Nereis, the native FaRP is the tetrapeptide

FMRFamide, the first molluscan FaRP to be found in an

extra-molluscan species. Immunohistochemically it is

localized in cells and processes of the circumesophageal

nerve ring and the ganglia of the ventral nerve cord, as

well as the gut and other non-nervous tissues. FMRFamide

relaxes the spontaneously active and electrically

stimulated esophagus in a dose-dependent manner,

suggesting that it is involved in regulating gastric

motility. In Callinectes, the native FaRP is

GYNRSFLRFamide. It is found in the pericardial organs and

is a potent cardioexcitatory peptide. A structure-

activity study revealed the requirements for full

biological potency.

The polychaete peptide, FMRFamide, is identical to

the tetrapeptide sequence found in all molluscs. All

other extra-molluscan peptide sequences have been

N-terminal extensions of the tetrapeptide core, FXRFamide

(where X is L, M, or I). Annelids and molluscs have many

characteristics which suggest a common ancestry (Stasek,

1972; Field et al., 1988). The peptide sequence is

consistent with this, and suggests that the tetrapeptide

gene may have arisen in the primordial ancestor.

FMRFamide is found throughout the central nervous

system of the worm. It is also localized in the

periphery, including the body wall and digestive system.

Furthermore, exogenous FMRFamide causes a dose-dependent

relaxation of the isolated esophagus. These data indicate

that FMRFamide is a neurotransmitter in the central and

peripheral nervous system and is involved in the control

of gastric motility.

Unlike the peptide in Nereis, the blue crab peptide,

GYNRSFLRFamide is an N-terminally extended FaRP. It and

the other partially sequenced peptides are similar to the

other two known crustacean sequences, suggesting that

there are at least three distinct sub-populations of

arthropodan FaRPs: two in insects, and another in both

insects and crustaceans. The crustacean peptides are also

similar to the peptides found in the nematode, Panagrellus

redivivus, SADPNFLRFamide and SDPNFLRFamide. Both of

these peptides contain an asparagine in the position

adjacent to the FLRFamide core; the same residue is also

found in the same position in many of the crustacean and

Drosophila peptides and affects the potency of the

peptides in the crustacean heart bioassay. Nematodes are

ancient and probably evolved from the same primordial

group of flatworms that gave rise to the annelids,

arthropods, and molluscs (Field et al., 1988). The

similarity in peptide sequence suggests that the

asparagine-containing FaRPs may have developed early in


The Callinectes peptide, GYNRSFLRFamide, is found in

the pericardial organs, part of the crustacean

neuroendocrine system. It is readily accessible to the

heart and causes a dose-dependent increase in heart rate.

This suggests that GYNRSFLRFamide and the other native

FaRPs have a role in controlling the circulation and

possibly other organs of Callinectes sapidus.

The structure and activity study of the crustacean

cardiac receptor shows that the receptor requires more

than the FLRFamide tetrapeptide core. For full potency

the peptide should have the sequence XXZFLRFamide, where X

is any amino acid and Z is either asparagine or serine.

This is unlike any other FaRP receptor studied to date,

although the similarity of the crustacean peptide

sequences to some Drosophila and nematode peptides, i.e.

the C-terminal sequence NFLRFamide, may indicate a

similarity in receptors.

The results of this dissertation suggests several

lines of research that could be pursued. The FaRP genes

of both Nereis virens and Callinectes sapidus should be


isolated and sequenced. The nereid gene may reveal other

less abundant peptide sequences undetected by the peptide

isolation and sequencing techniques used here. The crab

gene would identify the missing amino acids in the partial

sequences isolated in this study.

In Nereis, a structure-activity study of the

esophageal relaxation response would reveal whether the

annelid receptor is similar to the molluscan tetrapeptide


In Callinectes, a receptor binding assay should be

developed to determine whether the in vitro bioassay data

corresponds to binding affinities of the FaRPs studied.