Title: Diverse localization of 14-3-3 proteins in Arabidopsis thaliana
CITATION PDF VIEWER THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00100828/00001
 Material Information
Title: Diverse localization of 14-3-3 proteins in Arabidopsis thaliana
Physical Description: Book
Language: English
Creator: DeLille, Justin Marcus, 1977-
Publisher: University of Florida
Place of Publication: Gainesville Fla
Gainesville, Fla
Publication Date: 2001
Copyright Date: 2001
 Subjects
Subject: Plant Molecular and Cellular Biology thesis, M.S   ( lcsh )
Dissertations, Academic -- Plant Molecular and Cellular Biology -- UF   ( lcsh )
Genre: government publication (state, provincial, terriorial, dependent)   ( marcgt )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )
 Notes
Summary: ABSTRACT: The 14-3-3 family of proteins was discovered in the late 1960s during an extensive study of cytosolic bovine brain proteins. Once thought to be limited to mammalian brain tissue, studies over the last 15 years revealed 14-3-3s to be ubiquitous. The 14-3-3s are a family of acidic, soluble proteins that are commonly found as multi-isoform families within an organism. Specifically in plants, 14-3-3s are associated with many essential processes including transcription complexes, carbon and nitrogen metabolism, and protein-protein interactions. The Arabidopsis thaliana family of 14-3-3 proteins consists of ten distinct and well-characterized isoforms as well as three newly recognized isoforms. The question of functional specificity among the isoforms has led us to our current project. Using a set of isoform-specific antibodies and C-terminal green fluorescent protein fusions, we show that individual 14-3-3s differ in subcellular localization in Arabidopsis thaliana root tissue.
Summary: KEYWORDS: Arabidopsis thaliana, 14-3-3, green fluorescent protein
Thesis: Thesis (M.S.)--University of Florida, 2001.
Bibliography: Includes bibliographical references (p. 126-130).
System Details: System requirements: World Wide Web browser and PDF reader.
System Details: Mode of access: World Wide Web.
Statement of Responsibility: by Justin Marcus DeLille.
General Note: Title from first page of PDF file.
General Note: Document formatted into pages; contains viii, 131 p.; also contains graphics.
General Note: Vita.
 Record Information
Bibliographic ID: UF00100828
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 48991869
alephbibnum - 002763575
notis - ANP1597

Downloads

This item has the following downloads:

master ( PDF )


Full Text











DIVERSE LOCALIZATION OF 14-3-3 PROTEINS IN Arabidopsis thaliana


By

JUSTIN MARCUS DeLILLE












A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA


2001





























Copyright 2001

by

Justin Marcus DeLille



























This thesis is dedicated to all the friends and family who have helped me over the years.















ACKNOWLEDGMENTS


So many people have helped me that I am not sure where to start. I would first

like to thank my wife, Angela. She has been supportive and loving in everything I have

wanted to do. I know that this work would not have been possible without her. I would

also like to thank my family and specifically my dad for always being there for me when

I needed advice.

I would also like to thank Dr. Ralph Henry and the soon-to-be Dr. Eric Peterson.

Both of these men have probably had the greatest impact on my graduate career. Dr.

Henry generously allowed me to work in his lab as an undergraduate. He donated his

time and effort at a time when he really had none to give. He saw something in me that

no one else saw and for that I will always be thankful. Eric has always been a good

friend and someone I have always enjoyed talking science with (mostly while drinking).

I would next like to thank all those who make up the Ferl lab, both past and

present: Beth Laughner, Dr. Paul Sehnke, Dr. Anna-Lisa Paul, Michael Manak, Matthew

Reyes, and the new Dr. Carla Lyerly-Linebarger. This group of people really has helped

me in every way possible. Whether it was bouncing an idea around or watering plants,

they have always been happy to help.

I would like to thank my committee members. Dr. Charles Guy and Dr. Gloria

Moore. They generously donated their time and effort when they were not obligated. I

thank them for that.









Finally, I would like to thank Dr. Ferl. I was not sure what I really wanted to do

with my life when I entered the lab. However, he helped point me in a direction, whether

directly or indirectly. He told me once, "We {professors} are here for the students, not

the other way around." Well, he has certainly been there for me and I will always be

indebted. I thank him.

















TABLE OF CONTENTS

page

A C K N O W L E D G M E N T S ................................................................................................. iv

A B STRA CT .............. .... ..................................................................................... viii

CHAPTERS

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

2 REVIEW OF LITERA TURE ........................................ ................................. 3

H isto ry .................................................................. 3
N om enclature ...................................................................................... 4
Function ..................... .................................. ........ 5
A rabidopsis thaliana 14-3-3 Proteins .. .................................................... .............. 7


3 Arabidopsis thaliana 14-3-3 FAMILY OF SIGNALING
R E G U L A T O R S ....................................................... 12


4 WHOLE-MOUNT IMMUNOFLUORESCENCE STUDIES .......................................30

Introduction....................... ............... ..... ............. 30
M materials and M methods ..................................................... 32
R e su lts ................... ................... ...................3.........4
D iscu ssio n ..................................................... 3 7


5 USING GREEN FLUORESCENT PROTEIN FOR 14-3-3
L O C A L IZ A T IO N ........................................................................................57

Introduction....................... ............... ..... ............. 57
M materials and M ethods................................... .............. 58
R e su lts ....................................................................... 6 2
D iscu ssio n ..................................................... 6 8









6 CON CLU SION S ....................................................... ...... ............ .. 97

A P P E N D IX ................................................................................................................. 1 0 6

L IST O F R E FE R E N C E S ......................................................................... ...................126

BIOGRAPHICAL SKETCH ............................................................. ..................131















Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

DIVERSE LOCALIZATION OF 14-3-3 PROTEINS IN Arabidopsis thaliana


By

Justin Marcus DeLille

August, 2001


Chairman: Dr. Robert J. Ferl
Major Department: Plant Molecular and Cellular Biology

The 14-3-3 family of proteins was discovered in the late 1960s during an

extensive study of cytosolic bovine brain proteins. Once thought to be limited to

mammalian brain tissue, studies over the last 15 years revealed 14-3-3s to be ubiquitous.

The 14-3-3s are a family of acidic, soluble proteins that are commonly found as multi-

isoform families within an organism. Specifically in plants, 14-3-3s are associated with

many essential processes including transcription complexes, carbon and nitrogen

metabolism, and protein-protein interactions. The Arabidopsis thaliana family of 14-3-3

proteins consists of ten distinct and well-characterized isoforms as well as three newly

recognized isoforms. The question of functional specificity among the isoforms has led

us to our current project. Using a set of isoform-specific antibodies and C-terminal green

fluorescent protein fusions, we show that individual 14-3-3s differ in subcellular

localization in Arabidopsis thaliana root tissue.















CHAPTER 1
INTRODUCTION

This is truly an exciting era for science and technology. Buzzwords such as

genomics, cloning, and proteomics are regularly used by non-scientists and have nearly

become household terms. The last year brought huge advancements in sequencing

projects. We began the millennium with a grand feat in sequencing the entire human

genome (Lander et al., 2001). For the first time, humans will be able to study in great

detail genes that make us human. Many lessons were learned during the human genome

project: some scientific and some political, but none more important than 'if we work

hard enough we can accomplish anything.'

The human genome project also sparked many other sequencing projects. The

Arabidopsis thaliana genome project is of particular interest to many scientists because it

is used as a model organism for plant studies. Its relatively small genome size, ease of

propagation, and short life cycle make it a prime candidate for study by a wide range of

scientists with very different interests. The completion of the Arabidopsis thaliana

genome project late last year finally gives scientists the opportunity to speak with

confidence about the genome as a whole (Arabidopsis Genome Initiative, 2000).

Arabidopsis thaliana also provides us with the opportunity to study the 14-3-3

proteins. The 14-3-3s are a multi-isoform, acidic, soluble family of proteins that were

once thought to be found only in mammalian brain tissue. However, more extensive

studies have proven 14-3-3s to be ubiquitous among eukaryotes. The 14-3-3s perform









multiple functions within the cell and are grouped under an overall theme of protein-

protein interaction. Plant 14-3-3s are known to associate with kinases, to regulate key

enzymes of carbon and nitrogen metabolism, and even to associate with transcription

factors.

The goal of this thesis is to characterize the subcellular localization of nine

specific isoforms within the root tip of Arabidopsis thaliana using laser scanning confocal

microscopy. Isoform-specific antibodies and in-frame green fluorescent protein fusions

are used.

In characterizing the subcellular localization of these isoforms, we are hoping to

answer several questions. However, the key question is this: do the isoforms have

functional specificity or do they play redundant roles within the cell? If the isoforms do

have specificity, then we expect to see different localization patterns within the cell

among the nine isoforms. If there is no isoform specificity, then we expect to see a

general localization pattern emerge for all the isoforms.















CHAPTER 2
REVIEW OF LITERATURE


History

The 14-3-3 proteins are a family of acidic, soluble proteins that were discovered

in 1967 by Blake Moore and Vernon Perez during an extensive study of cytosolic bovine

brain proteins (Moore and Perez, 1967). The 14-3-3 proteins were named based on

fractionation on DEAE cellulose and electrophoretic mobility in starch gel

electrophoresis (Moore and Perez, 1967). Further studies over the next twenty years

revealed that the 14-3-3 family of proteins are conserved among mammals. Still it was

thought that the 14-3-3s were primarily limited in their location and function to brain

tissue. However, upon extended examination starting in the late-1980s, 14-3-3s were

found to be present in a plethora of organisms and tissues (Ichimura et al., 1991). To

date, 14-3-3 members have been discovered in many organisms including cow (Moore

and Perez, 1967), sheep (Token et al., 1990), chicken, human, rat, mouse, turtle, frog,

goldfish (Ichimura et al., 1991), spinach (Hirsch et al., 1992), Arabidopsis thaliana (Lu et

al., 1992), maize (de Vetten et al., 1992), yeast (van Heusden et al., 1992), barley (Brandt

et al., 1992), Drosophila (Swanson and Ganguly, 1992), rice (Kidou et al., 1993), tomato

(Laughner et al., 1994), tobacco (Chen et al., 1994), oat (Korthout and de Boer, 1994).

14-3-3s have been found in a variety of tissues as well, including testis, intestine, liver,

kidney, and skeletal muscles of mammals (Ichimura et al., 1991). It is important to note









that 14-3-3s are present in different organisms and tissues in varying concentrations and

locations.


Nomenclature

The nomenclature regarding 14-3-3s has been difficult and often confusing for

several reasons. First, many scientists characterized and named proteins based on

function or other characteristics; only later discovering the protein to be a 14-3-3 family

member. Therefore, the protein or group of proteins lacks the 14-3-3 moniker and is thus

misnamed. Second, the speed at which proteins are being associated as 14-3-3 family

members has and continues to be rapid. The rate of discovery combined with the lack of

a formal naming scheme has led to much confusion in the literature regarding 14-3-3

proteins. In the late 1980s, 14-3-3 members found in the bovine brain were further

characterized and designated as Greek letters starting at the beginning of the alphabet

(Ichimura et al., 1988). Seven 14-3-3 members were identified in that study and were

named based upon elution during reverse-phase HPLC. The mammalian isoforms are

designated alpha, beta, gamma, delta, epsilon, zeta, and eta. The list of mammalian

isoforms grew again to include tau and sigma in 1991 and 1993, respectively (Nielsen,

1991; Leffers et al., 1993). Additionally, further studies revealed the alpha and delta

isoforms to be the phosphorylated forms of beta and zeta, respectively (Aitken et al.

1995). In keeping with the mammalian trend, the Arabidopsis thaliana 14-3-3s are

designated starting with the end of the Greek alphabet and working forward as they are

discovered. However, there is not a strict adherence to the naming system in arabidopsis.

Therefore, overlap in designations does exist between mammalian and arabidopsis 14-3-3

isoforms. The Arabidopsis thaliana 14-3-3 isoforms are omega, psi, chi, phi, upsilon,









rho, pi, omicron, nu, mu, lambda, kappa, and epsilon (Lu et al., 1994a; Wu et al., 1997;

Rosenquist et al., 2000; DeLille et al., 2001). Table 2-1 lists the mammalian and

Arabidopsis thaliana 14-3-3 Greek names and symbols for reference.


Function

The 14-3-3 family of proteins is found in every eukaryotic organism where this

specific family is sought (Robinson et al., 1994). The conservation of 14-3-3 proteins

among a wide variety of species implies an essential and necessary role, yet 14-3-3

members are associated with a variety of functions. Two general themes have emerged

regarding the roles of 14-3-3s; protein kinase activities and protein-protein interactions.

The earliest function of 14-3-3s was recognized in 1987 by Ichimura et al. during a study

of acidic brain proteins (Ichimura et al., 1987). The 14-3-3s Ichimura studied were

remarkably similar in amino acid sequence to a protein known to activate tyrosine and

tryptophan hydroxylases. Further studies revealed the activator protein and the 14-3-3s

to be indistinguishable in amino acid sequence. Thus, the first role of 14-3-3s was

established as an activator associated with the kinase-dependent regulation of enzyme

activity. The 14-3-3s are also associated with other protein kinase activities including the

regulation of protein kinase C, stimulation of calcium-dependent exocytosis in adrenal

chromaffin cells, activation of c-Raf-1, and the regulation of Bcr-Abl (Aitken, 1995;

Morgan and Burgoyne, 1992; Fantl et al., 1994; Reuther et al., 1994).

The protein-protein interaction theme is more general and therefore less well

defined. The 14-3-3 family of proteins has been shown to be associated with large

complexes. These complexes involve protein kinases, DNA binding complexes and

chaperones (Ferl, 1996). Multiple recognition motifs have been proposed and the









common theme among them is the presence of a phosphoserine. Two of the motifs are

RX(Y/F)XpSXP, RXXSXpSXP, where X is any amino acid and pS denotes a

phosphoserine. However, other binding motifs have also been proposed that do not

contain a phosphoserine, such as WLDLE, RXSX(S/T)XP, XXXSXXSXXXSXXSX

(Pan et al., 1999).

Specifically in plants, 14-3-3s are associated with several specialized functions

including a role in fusicoccin binding sites and the regulation of nitrate reductase and

sucrose phosphate synthase (Korthout and de Boer, 1994; Finnie et al., 1999). Fusicoccin

(FC) is a fungal toxin excreted by Fusicoccum amygdali that targets the plant plasma

membrane HT-ATPase proton pump (Chung et al., 1999). Studies have shown that 14-3-

3s complex with the H -ATPase proton pump and thus constitute the FC receptor

(Sehnke and Ferl, 2000). FC binding to the receptor causes the proton pump to stay "on,"

thus causing turgor pressure to decrease in guard cells and the plant leaves to wilt.

Key enzymes of nitrogen and carbon metabolism are also regulated by 14-3-3s,

and currently form the paradigm for 14-3-3 regulation of enzyme activity. Nitrate

reductase (NR) and sucrose phosphate synthase (SPS) are up-regulated during

photosynthesis. However, in the dark, NR is inactivated via a two-step mechanism

(Figure 2-1) (Chung et al., 1999). In the first step, a specific serine is phosphorylated in

the target enzyme. In the second step, Mg2 binds an inhibitor protein, causing a

conformational change, thus allowing the inhibitor protein to bind to the target enzyme.

One or more 14-3-3s make up this inhibitor protein (Bachmann et al., 1996). Sucrose

phosphate synthase is also regulated by light and phosphorylation. However, 14-3-3









binding to SPS has resulted in both activation (Moorhead et al., 1999) and inactivation

(Toroser et al., 1998).

The 14-3-3s in plants are also associated with a class of transcription factors

called G-box factors (GBFs) (Lu et al., 1992). G-box factors are a class of bZIP

transcriptional activators involved in stress and abscisic acid activation of genes.

Additionally, Arabidopsis thaliana 14-3-3s have been shown to interact with other

transcription factors, TBP and transcription factor IIB (Pan et al., 1999).

The overall amino acid conservation of 14-3-3s is relatively well conserved both

within and among species. The N- and C-terminal regions are the most variable in both

mammals and plants. The Arabidopsis thaliana isoforms share a core region in which

51% of the amino acids are conserved. The amino acids in the N-terminal region are

only conserved to a degree of 14%, while there is no amino acid conservation in the C-

terminal region among the isoforms (Chung et al., 1999).


Arabidopsis thaliana 14-3-3 Proteins

One of the largest and best-characterized 14-3-3 families in plants is found in

Arabidopsis thaliana. The well-characterized Arabidopsis thaliana 14-3-3 isoforms are

omega, psi, chi, phi, upsilon, nu, mu, lambda, kappa, and epsilon (Lu et al., 1992; Lu et

al., 1994b; Wu et al., 1997). The 14-3-3 family in Arabidopsis thaliana was initially

discovered while studying the regulation of the alcohol dehydrogenase gene and a cis

regulatory element called the G-box (Lu et al., 1992). Using monoclonal antibodies

against the partially purified G-box binding factor, Lu et al. isolated cDNA clones that

encoded 14-3-3 isoforms omega, psi, chi, phi, and upsilon (Lu et al., 1992; Lu et al.,

1994b). Using the chi isoform as bait in a yeast two-hybrid screen, Wu et al. identified









the kappa, lambda, and epsilon isoforms (Wu et al., 1997). Finally, the isoforms mu and

nu were identified by searching the Arabidopsis thaliana EST database using the eight

known isoforms as a query (Wu et al., 1997). Recently, 3 new isoforms, omicron, rho,

and pi, were identified by database analysis (Rosenquist et al., 2000; DeLille et al., 2001).

A full-length cDNA has been identified for omicron and a truncated cDNA has been

identified for the rho isoform. The third isoform, pi, remains putative at this point as no

cDNA or EST has been found.

Determining the subcellular localization of the individual 14-3-3s is essential to

understanding their true roles in the cell. Bihn et al. reported the general subcellular

location of the Arabidopsis thaliana 14-3-3s (Bihn et al., 1997). The authors used a suite

of three monoclonal antibody cell lines and localized 14-3-3s to the nucleus and

cytoplasm. However, the localization of specific isoforms was not revealed.

Thel4-3-3s are generally found as multi-isoform families in eukaryotic

organisms. This begs the question: Do all the isoforms play the same role or does

functional specificity exist among the isoforms? Studies have identified subsets of

isoforms in various organelles within the cell. For example, in addition to being present

in the cytoplasm, epsilon, mu, nu and upsilon are also present in chloroplasts (Sehnke et

al., 2000). Functional specificity can also be seen with nitrate reductase. Omega, chi,

and upsilon demonstrate decreasing affinity, while the isoforms psi and phi have no

affinity for the enzyme.

The goal of this thesis is to determine the localization patterns of nine well-

characterized arabidopsis 14-3-3 isoforms using a combination of isoform specific

antibodies and in-frame green fluorescent (GFP) fusions. It is the hope that differences in






9


localization patterns among the isoforms do indeed occur, thus providing evidence for

functional specificity.
























Table 2-1. Mammalian and Arabidopsis thaliana 14-3-3 Proteins. Currently nine are
found in mammalian tissue and thirteen in arabidopsis. Mammalian isoforms
are designated starting at the beginning of the Greek alphabet. Arabidopsis
14-3-3s are designated from the end of the alphabet, working forward. Some
overlap in naming does exist between mammalian and arabidopsis isoforms.
The Greek names and symbols are listed for reference.


Mammalian 14-3-3 Proteins


Greek Name
alpha
beta
gamma
delta
epsilon
zeta
eta
tau
sigma


arA bidopsis


Greek
Letter

K


1
V




U


X

0)


14-3-3 Proteins


Greek Name
epsilon
kappa
lambda
mu
nu
omicron
pi
rho
upsilon
phi
chi
psi
omega


Greek
Letter
a



6



T
sC
(
n
1
a


.......... r" ... ... .........












14-3-3


Kinase (CDPK)


Phosphatase


e


Mg2


<--_-


0 0


Key enzymes in carbon and nitrogen metabolism in plants in a reversible
two-step mechanism. The first step is the phosphorylation of a specific
serine in the target enzyme by calmodulin-domain protein kinase (CDPK).
In the second step, Mg2 binds 14-3-3 proteins causing a conformational
change that allows the 14-3-3 to bind the target enzyme and thus complete
the inactivation. ATP = adenosine triphosphate.


MAtvnzm


Figure 2-1.


OKI















CHAPTER 3
Arabidopsis thaliana 14-3-3 FAMILY OF SIGNALING REGULATORS

The 14-3-3 family of proteins has received much attention in the literature during

the last 10 years. The current interest is not surprising given the number of diverse

organisms in which 14-3-3s have been identified and the important role that they play in

signal transduction. Moore and Perez initially catalogued the 14-3-3 proteins in 1967

during an extensive study in which bovine brain proteins were given numerical

designations based on column fractionation and electrophoretic mobility (Moore and

Perez, 1967). The 14-3-3 family was thought to be limited to nervous tissue and largely

conserved among mammals during the late 1960s and 1970s. However, studies over the

last 15 years have proven 14-3-3s to be ubiquitous, found in virtually every eukaryotic

organism and tissue (Ichimura et al., 1987; Robinson et al., 1994). In any given organism,

the 14-3-3 family usually consists of multiple genes and protein isoforms. Multiple

isoforms and multiple functions, coupled with the large number of different organisms

that have been studied, have led to potential confusion regarding 14-3-3 nomenclature

and function. (The 14-3-3s are currently designated by Greek letters, with the mammalian

isoform names generally chosen from the beginning of the alphabet and the plant

isoforms chosen from the end of the alphabet). Recent completion of the Arabidopsis

thaliana genome project provides a unique opportunity to examine a complete 14-3-3

family within a single higher eukaryotic organism and to present a framework to codify

the understanding of plant 14-3-3 functional diversity and constraint.









The 14-3-3 proteins play key functional roles in many critical physiological

pathways that are regulated by phosphorylation. Their role is to complete the signal

transduction process by binding to the phosphorylated target, which completes a change

in structure that regulates activity. This core functional characteristic is deeply engrained

in the highly conserved structural core of the 14-3-3 dimer, which provides grooves for

binding two specifically phosphorylated peptides. The primary diversity among 14-3-3

isoforms lies in the N- and C-termini, with the C-terminal region potentially able to form

a flexible hinge guarding access to the central core region (Sehnke and Ferl, 2000).

Plants require a battery of regulators and corresponding responses to deal with

complex environmental and developmental changes, a situation that seems consistent

with the presence of a large and diverse 14-3-3 family. Localization of 14-3-3 family

members inside organelles such as the chloroplast (Sehnke et al., 2000), nucleus (Bihn et

al., 1997) and mitochondria (Sehnke and Ferl, 2000), in addition to the cytoplasm (Bihn

et al., 1997), further demonstrates both their global regulatory potential and their apparent

need for diversity in expression and function. The list of the processes controlled by 14-

3-3s includes the fundamental nitrogen and carbon assimilation pathways, which are

executed by the light and substrate-regulated metabolic enzymes nitrate reductase and

sucrose phosphate synthase (Sehnke and Ferl, 2000). Other enzymes under the control of

14-3-3s include starch synthase (Sehnke et al., 2001), glutamate synthase, Fl ATP

synthase, ascorbate peroxidase, and affeate O-methyl transferase (Finnie et al., 1999).

Additionally, the control of the plant's turgor pressure via regulation of at least one form

of a plasma membrane H+ ATPase is accomplished by 14-3-3 proteins (Korthout and

deBoer, 1994; Marra et al., 1994; Oecking et al., 1994). Less understood, yet equally









bona fide 14-3-3 binding partners include transcriptional machinery such as the G-Box

complex, core transcription factors; TBP, TFIIB and EMBP (Chung et al., 1999). The

specific 14-3-3 isoforms required by each of these pathways has not been fully

characterized, however a conserved mechanism of plant 14-3-3s binding is the

requirement for divalent cations to "charge" the 14-3-3s via a structural reorientation of

the C-termini (Lu et al., 1994b). Interestingly, only a subset of the Arabidopsis thaliana

14-3-3 isoforms possess this EF hand-like divalent cation-binding motif in the C-terminal

region.

The Arabidopsis thaliana genome project for the first time provides reasonable

certainty about the number and diversity of 14-3-3 family members within a plant

species. The Arabidopsis thaliana 14-3-3 family consists of 13 members. Ten of the

members, omega, phi, chi, psi, upsilon, nu, mu, lambda, kappa, and epsilon are well

characterized and present as ESTs and cDNAs (Lu et al., 1992; Lu et al., 1994b; Wu et

al., 1997). Three of the members, omicron, rho, and pi are recently recognized 14-3-3

family members. The omicron isoform was identified by Rosenquist et al. (Rosenquist et

al., 2000) and subsequently a full-length cDNA was found by the same authors (AC #

AF323920). A C-terminally truncated cDNA relative to the genomic sequence has been

found for the rho isoform (AC# AF335544). The pi isoform remains a putative 14-3-3

member at this point as no cDNA or EST has been found. Two other putative members

exist, however one isoform is badly truncated and would likely not be functional

(Accession # AC007264), and the other isoform is annotated in Genbank as containing a

one base pair intron (Accession #AC068562). We believe the Genbank entry for the later

isoform is annotated incorrectly and the intron is actually a single base pair insertion that









causes a frame shift and the loss of the last 45 amino acids. Thus we are designating both

of these isoforms as "14-3-3-like proteins." Figure 3-1 represents the intron-exon

structures of all the Arabidopsis thaliana 14-3-3s. All of the reference information

regarding Arabidopsis thaliana 14-3-3s is found in Table 3-1.

An alignment of the thirteen isoforms reveals some interesting information

(Figure 3-2). The isoforms range in length from 241-286 amino acids. The isoforms all

share a conserved core region, with the N-termini and C-termini being the most

divergent. In fact, the amino acids in the N-termini are conserved to a degree of only 14%

and there is very little amino acid conservation at the C-termini (Chung et al., 1999).

Phylogenetic analyses based on amino acid sequence data and gene structure

provides a robust tree upon which to hang descriptions of family member function and

localization (Figure 3-3). The family members break into two major evolutionary

branches, the Epsilon group and the Non-Epsilon group. This clear delineation at the

trunk of the tree is ubiquitous among plants and animals possessing multiple isoforms,

indicating that the initial formation of two isoforms is a fundamental and ancient

divergence. The Epsilon group is itself split into the isoforms epsilon, mu, omicron, rho,

and pi. The Non-Epsilon group is made up of the isoforms kappa, lambda, phi, chi,

omega, psi, nu, and upsilon. The Epsilon group breaks into two sub-branches, with

epsilon and pi on one sub-branch; and omicron, rho and mu in the second sub-branch.

The Non-Epsilon group comprises three very distinct sub-branches. Kappa and lambda

make up one sub-branch: phi, chi, and omega make up a second sub-branch; and psi, nu,

and upsilon make up the third sub-branch. The Non-Epsilon group members contain the

previously mentioned EF hand-like divalent cation-binding motif (Lu et al., 1994b).









The Non-Epsilon and Epsilon groupings are also well supported by intron-exon

structure. The Non-Epsilon members all contain four exons and three introns that are

highly conserved in placement. Psi, nu and upsilon contain an extra intron in the 5' leader

(Wu et al., 1997). The Epsilon members all possess an intron-exon structure distinct from

the Non-Epsilon group, having two additional N-terminal exons. (Presently, the pi

isoform contains five exons and four introns. The first N-terminal exon is not annotated

in Genbank; however it is indeed present on inspection.) Several genes of the Epsilon

group also appear to have additional C-terminal exons. However, the extreme divergence

of the C-terminal regions prohibits exon identification based solely on sequence data. The

pi isoform is not present as cDNA or an EST. Therefore its structure remains putative at

this point and the cDNA available for rho is believed to be truncated.

The complexity of this phylogenetic tree raises an important question. Why are so

many 14-3-3 genes present within a single organism? One possible answer is that there is

a need to ensure that 14-3-3 activity is present in every compartment of every cell of the

organism, suggesting that diversity is simply a reflection of developmental evolution and

sophistication. Using current prediction programs, there are no obvious subcellular

targeting signals associated with any of the isoforms. So the large number of isoforms is

not obviously linked to diversifying the subcellular location. It has been observed,

however, that unicellular organisms contain relatively few isoforms while multicellular

organisms have many; and certain organelles contain only subsets of the isoforms

(Rosenquist et al., 2000). Another possible answer is that each isoform plays a specific

and essential biochemical role, suggesting that general diversity reflects functional

divergence. They all share a relatively conserved core region, which could point to the









conservation of a general theme, yet subtle changes in the core and the divergent termini

could give each isoform its specific function by dictating affinity over a range of possible

targets.

The structure of this tree does provide an evolutionary perspective that should

contribute to answers to these questions based on emerging data. For example, only

epsilon, mu, nu, and upsilon are present in chloroplast stroma, in addition to the

cytoplasm (Sehnke et al., 2000), demonstrating that subcellular localization could be

consistent with their position on the phylogenetic tree. Omega, chi, and upsilon

demonstrate decreasing affinity for nitrate reductase, while phi and psi show no affinity

(Bachmann et al., 1996). Isoforms omega, kappa, and lambda demonstrate a decreasing

affinity for the proton ATPase (Rosenquist et al., 2000). These examples provide

evidence that functional affinity for targets could also be consistent with the phylogenetic

tree, but both the localization and function data sets are far from complete. The

Arabidopsis thaliana 14-3-3 family should provide a well-developed and inclusive

framework for comparative 14-3-3 biology.



























Arabidopsis thaliana 14-3-3 gene structures and reference information.
Colored boxes represent exons, which are drawn to scale and color coded as
to their similarity between genes. The thin lines represent introns, which are
not to scale and are shown to denote position only. The purple boxes
represent 5' leader and 3' trailer sequences. The asterisks identify putative
exons, cases where cDNA sequences are not yet available. Current
information is outlined to the right of each 14-3-3 gene. Rho, omicron, and pi
are the least well characterized of the Arabidopsis thaliana 14-3-3s, thus very
little information is available. An expanded version of this table is available
at http://www.hos.ufl.edu/ferllab/. (a) Genomic structures of the Arabidopsis
thaliana 14-3-3 family. (b) Genomic structures of the epsilon group members
containing rho, omicron, mu, epsilon and pi. (c) Genomic structures of one
branch of the non-epsilon group containing kappa and lambda. (d) Genomic
structures of one branch of the non-epsilon group containing phi, chi, and
omega. (e) Genomic structures of one branch of the non-epsilon group
containing psi, upsilon and nu members. (f) Genomic structures of the "14-
3-3 like" isoforms.


Figure 3-1.








19



94 b74 b 76 b 83b 9 b 57 b

Rho V-A.A.
27AA 102AA 30AA 36AA 50AA 29AA 12AA
S7 84b98bp

Omicron 241 A.A.
22AA 102AA 30AA 37AA 50AA
97b 937b 84b

Mu 2 263 A.A.
24AA 103 AA 29AA 37 AA 49AA 21AA STOP


Epsilon 254 A.A.
22AA 103AA 29AA 36AA 50AA 14AA
EXOn nOtf atfd~ 337 b 94 bo 1161 189b 9 b

Pi M M I241A.A.
23AA 98AA 30AA 36AA 54AA 9AA ?AA


Kappa *-- 248 A.A.
162 AA 41AA 39AA 6AA


Lambda 248 A.A.
162 AA 41AA 39AA 6AA


Phi 1 267 A.A.
165 AA 41AA 39AA 22AA


1Chi A 4267 A.A.
164 AA 41AA 39AA 23AA


Omega 259 A.A.
159 AA 41AA 39AA 20AA
438 78 8"/1 76 b



Upsion 255A.A.
16158 AA 41AA 39AA 17AA


Upsilon 268 A.A.
156 A 441AA 39AA 27AA
1611AA


V 265 A.A.
156 A-A 44AA 39AA 26AA

AC8562 241 A.A.
AC068562
22AA 106 AA 23AA 25AA 18AA 47AA
2 142bAA 80 b
AC# 7 Y4f 82 A.A.
AC007264
20AA 31AA 31AA


Legend:


S- Triangles Represent Introns



M Colored Boxes Represent Exons


Figure 3-la














007






co
Cl CC

0 --
00





















pp



o-


po -
U, o
'] <.<',
Ct


hi .,. 00

















00 00

(N (N






































-o -o




o-
S 0



































0\0
I~I









a
Cl





00
M~ 0







.


T- c 0
( H Q
0


M
Sp




UM
SC
0\ O







































<

























a a















d ^d ^



























0 -












o









INI












P o
W o-
N

S< <















Table 3-1. Reference information for the Arabidopsis thaliana proteins and genes.
A truncated cDNA is available for rho. Pi remains a putative family member
at this point.


14-3-3 PROTEINS AND GENES
GRF Gene Genbank Genbank Genbank
Greek Greek and Accession # Locus # for cDNA Accession # Original
Letter Name Accession # for BAC cDNA clone Designation for cDNA Publication
GRF10 ATU36446 Wu et al.
s epsilon AF145302 AC068562 1100 bp 2hybr231 U36446 1997
GRF8 ATU36447 Wu et al.
K kappa AF145300 AB011479 1023 bp 2hybr31 U36447 1997
GRF6 ATU68545 Jarillo et al.
X lambda AF145298 AL353995 1040 bp 2hybr 171 U68545 1994
GRF9 ATU60444 Wu et al.
[L mu AF145301 AC007087 1103 bp EST D U60444 1997
GRF7 ATU60445 Wu et al.
v nu AF145299 AC021640 1069 bp EST H U60445 1997
Rosenquist
o omicron GRF 11 AC007894 AF323920 et al. 2000
7t pi AC012680 Not Available
p rho GRF 12 AC013427 AF335544
GRF5 ATHGFUPS Lu et al.
u upsilon AF001415 AL391145 990 bp GF14-19 L09109 1994
GRF4 ATHGFPHI Lu et al.
d) phi AF001414 AC079605 1126 bp GF14-10 L09111 1994
GRF1 ATHGFCHI Lu et al.
x chi U09377 AL161513 1126 bp GF14-5 L09112 1994
GRF3 AB005248 ATHGFPSI Lu et al.
__ psi U09375 AB005231 A 1108bp GF14-4 L09110 1994
GRF2 ATHGF14A Lu et al.
co omega U09376 AC013430 1133 bp GF14-14 M96855 1992,1994
14-3-3 LIKE PROTEINS
Rosenquist
AC007264 et al. 2000
AC068562
































ClustalW alignment of all thirteen functional 14-3-3 proteins. The proteins
range in size from 241-286 amino acids. The blue color denotes a match
with the consensus sequence. The red color denotes an identical match
among all the isoforms.


Figure 3-2.








27



Epsilon (1) ----- MENEREKQVYLAKLSEQTERYDEMVEAMKKVAQ--- LDVELTVEERNLVSVGYKNVIGARRASWRI
Kappa (1) ---- MATTLS RDQYVYMAKLAE QAERYE EMVQ FMEQLVSGAT PAGE LTVEERNLLSGAYKNPIGSLRAAWRI
Lambda (1) -- MAATLGRDQYVYMAKLAE QAERYEEMVQFMEQLVTGATPAEELTVEERNLSVAYKNIGSLRAAWR
Mu (1) ---- MGSGKERDTFVYLAKLSEQAERYEEMVESMKSVAK--- LNVDLTVEERNLLS KIGSRRASWRI
Nu (1) ----- MSSS REENVYLAKLAEQAERYEEMVE ME KVAK- TVDTDELTVEERNL
Upsilon (1) ----MSSDSS REENVYLAKLAEQAERYEEMVEFMEKVAK-TVETEELTVEERNL
Phi (1) MAAPPAS SSAREEFVYLAKLAE QAERYEEMVEFMEKVAE-AVDKDELTVEERNLLSVAYKN IGARRASWR
Chi (1) -MAT PGAS SARDE FVYMAKLAE QAERYE EMVE FMEKVAK-AVDKDE LTVEERNLSVAYKNIGARRASWRI
Psi (1) ------MST REENVYMAKLAEQAERYE EMVEFMEKVAK-TVDVEELSVEERNL
Omega (1) ------ MASGREEFVYMAKLAEQAERYEEMVEFMEKVSA-AVDGDELTVEERNLLSVAY VIGARRASWRI
Omicron (1) ------MENERAKQVYLAKLNEQAERYDEMVEAMKKVAA--- LDVELTIEERNLLSVGYKNVIGARRASWRI
Rho (1) -MSSSGSDKERETFVYMAKLSEQAERYDEMVETMKKVAR---VNSELTVEERNLIGARRASWRI
Pi (1) ----- MENEREKLIYLAKLGCQAGRYDGMRKVCELDIE ------- LSEEERDLLTT
Consensus (1) RE FVYLAKLAEQAERYEEMVEFMEKVA DELTVEERNLLSVAYKNVIGARRASWRI

Epsilon ( 64) LSSIEQKEESKGNDENVKRLKNYRKRVEDELAKVCNDILSVIDKHLIPSSNAV-ESTVFFYKMKGDYYRY
Kappa i(69) VS SLEQKEESRKNEEHVSLVKDYRSKVETELSSIFSGILRLLDSHLIPSATAIR-ESKVFYLI KGDHRYLA
Lambda (69) VSSIEQKE ESRKNDEHVSLVKDYRSKVE SELSSVCSGILKLLDSHL I P SAGAS-ESKFYLKMKGDY
Mu (66) FSSIEQKEAVKGNDVNVKRIKEYMEKVELELSNICIDIMSVLDEHLIPSASEG-ESTVFFNKMKGDYYRYLA
Nu (66) ISSIEQKEESRGNDDHVSIIKDYRGKIETELSKICDGILNLLDSHLVPTASLA-
Upsilon (68) ISSIEQKEDSRGNSDHVSIIKDYRGKIETELSKICDGILNLLEAHLIPAASLA-E
Phi (72) ISSIEQKEESRGNDDHVTTIRDYRSKIESELSKICDGILKLLDTRLVPASANG-
Chi (71) ISSIEQKEESRGNDDHVSLIRDYRSKIETELSDICDGILKLLDTILVPAAASG-D
Psi (65) ISSIEQKEESKGNEDHVAIIKDYRGEIESELSKICDGILNVLEAHLIPSASPA-E
Omega (66) ISSIEQKEESRGNDDHVTAIREYRSKIETELSGICDGILKLLDSRLIPAAASG-D
Omicron ( 64) LSSIEQKEESKGNEQNAKRIKDYRTKVEEELSKICYDILAVIDKHLVPFATSG-ESTVFYYKMKGDYFRYLA
Rho (69) MSSIEQKEES KGNESNVKQI KGYRQKVEDELANICQDILTI I DQHLIPHATSG-ATVFYYKMKGDYYRY
Pi (60) ISSIEKMEDSKGNDQNVKLIKGQQEMVKYEFFNVCNDILSLIDSHLIPSTTTNVESIVLFNR YFRYMA
Consensus 73 ISSIEQKEESRGNDDHVSLIKDYR KVETELS IC GIL LLDSHLIPSASSG ESKFYLKMKGDYHRYL

Epsilon (135) EFSSGzAERKEAADQSLEAYKAAVAAAENGLAPTHPVRLGLALNFSVFYYEILNSPESACQLAKQAFDDAIAE
Kappa EFKSGDERKTAAEDAMIAYKAAQDVAVADLAP TH PI RLGLALN FSV FYYE ILN SSEKCSMAKQAFEEAIAE
ambda (14 0) EFKSGDERKTAAEDTMLAYKAAQDIAAADMAPTHPIRLGLALNFSVFYYEILNSSDKACNMAKQAFEEAIAE
1Mu (137) EFKSGNERKEAADQSL KAYEIATTAAEAKLPPTHPIRLGLALNFSVFYYEIMNAPERACHLAKQAFDEAISE
Nu (137) EFKTGlAERKEAAESTLVAYKSAQDIALADLAPTHPIRLGLALNFSVFYYEILNSPDRACSLAKQAFDEAISE
Upsilon (13 9) EFKTGAERKEAAESTLVAYKSAQDIALADLAPTHPIRLGLALNFSVFYYEILNSSDRACSLAKQAFDEAISE
Phi (143) EFKTGQERKDAAEHTLTAYKAAQDIANAELAPTHPIRLGLALNFSVFYYEILNSPDRACNLAKQAFDEAIAE
Chi (142) EFKSGQERKDAAEHTLTAYKAAQDIANSELAPTHPIRLGLALNFSVFYYEILNSPDRACNLAKQAFDEAIAE
Psi (136 ) EFKA,GAERKEAAESTLVAYKSASDIATAELAPTHPIRLGLALNFSVFYYEILNSPDRACSLAKQAFDDAIAE
Omega (137) EFKTGQERKDAAEHTLAAYKSAQDIANAELAPTHPIRLGLALNFSVFYYEILNSPDRACNLAKQAFDEAIAE
Omi crown i(135) EFKSGADREEAADLSL KAYEAATSSASTELSTTHPIRLGLALNFSVFYYEILNSPERACHLAKRAFDEAIAE
Rho (140 ) EFKTEQERKEAAEQSLKGYEAATQAASTELPSTHPIRLGLALNFSVFYYEIMNSPERACHLAKQAFD
Pi (13 2 E FGSDAERKENADNSL DAYKVAMEMAEN SLAP TNMVRLGLALN FSI FNYE IHKS I E SACKLVKAYDEAITE
Consensus (145) EFKSG ERKEAAE TL AYKAAQDIA AELAPTHPIRLGLALNFSVFYYEILNSPDPAC LAKQAFDEAIAE

Epsilon (2 07) LDSLNEESYKDSTLIMQLLRDNLTLWTSDLNEEG---DERTKGADEPQDEN--------------------
Kappa ( 212) LDTLGEESYKDSTLIMQLLRDNLTLWTSDMQEQMDEA--------------- ------------
Lambda (212) E LDTLGEESYKDSTLIMQLLRDNLTLWTSDMQEQMDEA----------------
Mu (209) LDTLSEESYKDSTLIMQLLRDNLTLWTSDISEEGGDDAHKTNGSAKPGAGGDDA
Nu i(209) LDTLGEESYKDSTLIMQLLRDNLTLWNSDINDEAGGDEIKEASKHEPEEGKPAETGQ ----------
Upsilon (211) LDTLGEESYKDSTLIMQLLRDNLTLWTSDLNDEAG-DDIKEAPKEVQKVDEQAQ-------
Phi (215) E LDTLGEESYKDSTLIMQLLRDNLTLWTSDMQDESP-EEIKEAA--APKPAEEQKEI--- --------
Chi -214) LDTLGEESYKDSTLIMQLLRDNLTLWASDMQDDVA-DDIKEAAPAAAKPADEQQS
Psi (208)i LDTLGEESYKDSTLIMQLLRDNLTLWTSDMTDEAG-DEIKEASKPDGAE --------- ------
Omega (209)i LDTLGEESYKDSTLIMQLLRDNLTLWTSDMQDDAA-DEIKEAA--APKPTEEQQ -----------
Omicron ,207) LDSLNE DSYKDSTLIMQLLRDNLTLWTSDLEEGGK------------------
Rho (2_12) LDTLSEESYKDSTLIMQLLRDNLTLWTSDLPEDGGEDNIKTEESKQEQAKPADATEWIHFKMRTDQRAWK
Pi (204) LDGLDKNICEESMYIIEMLKYNLSTWTSGDGNGNKTDG---------------
Consensus (217) LDTLGEESYKDSTLIMQLLRDNLTLWTSDMQDEAG DEIK

Epsilon (255) ---
Kappa (249) ---
Lambda (249) ---
Mu (264) ---
Nu (266) ---
Upsilon (269) ---
Phi (268) ---
Chi (268) ---
Psi (256) ---
Omega (260) ---
Omicron (242) ---
Rho (284) NEI
Pi (242) ---
Consensus (289)































Figure 3-3. Rooted tree of the Arabidopsis thaliana 14-3-3 family. This
rectangular cladogram summarizes the relationships among the 14-3-3
isoforms. A human 14-3-3 was included as an outgroup, and essentially roots
the tree between the Epsilon and Non-Epsilon groupings. The basic branch
topology presented in this neighbor-joining tree is essentially supported by
bootstrapped parsimony analysis, however the confidence of the branching
patterns within the Epsilon group is less well supported than the branching
within the highly supported Non-Epsilon group. The major branches of the
tree are also clearly supported by the gene structures, which is reflected by
the numbers of exons and introns as indicated on the branches. Bolded
isoforms are present in the chloroplast as well as the cytoplasm. The
underlined isoform is putative in that, as yet, they have not been isolated as
cDNAs or ESTs.

































Epsilon
- members



















Non-
_ Epsilon
members


Figure 3-3.


Human


Mu


Epsilon


Pi


Rho


Omicron


Kappa


Lambda


Psi


Nu


Upsilon


Omega


Phi


Chi















CHAPTER 4
WHOLE-MOUNT IMMUNOFLUORESCENCE STUDIES


Introduction

The 14-3-3s are a family of proteins that were initially discovered during an

extensive study of bovine brain proteins by Moore and Perez in 1967 (Moore and Perez,

1967). The 14-3-3s are named based on fractionation and electrophoretic mobility in

starch gel electrophoresis. It was long thought that 14-3-3s were limited to mammalian

brain tissue, however studies over the past fifteen years have proven 14-3-3s to be

ubiquitous. The 14-3-3s are found in greatly diverse organisms from drosophilia and

plants to frogs and humans (Finnie et al., 1999). This family of proteins has been found

in almost every eukaryotic organism that scientists have studied (Robinson et al., 1994).

The 14-3-3s are usually found as multi-isoform families within a single organism.

For example, there are nine isoforms found in mammalian tissue and are designated as

Greek letters (alpha, beta, gamma, delta, epsilon, zeta, eta, tau, and sigma) (Ichimura et

al., 1988; Patel et al., 1994). The Arabidopsis thaliana 14-3-3 members are designated in

a similar fashion starting at the end of the Greek alphabet and working forward. There

are currently thirteen isoforms found in Arabidopsis thaliana and are designated as

omega, psi, chi, phi, upsilon, rho, pi, omicron, nu, mu, kappa, lambda, and epsilon (Lu et

al., 1992; Lu et al. 1994b; Wu et al., 1997; Rosenquist et al., 2001; DeLille et al., 2001).

The 14-3-3s are involved in a number of diverse functions within the cell. They

are known to regulate key enzymes in metabolic processes such as nitrogen metabolism









and carbon metabolism (Chung et al., 1999), and to associate with the plasma membrane

HF-ATPase (Sehnke and Ferl, 2000) and transcription factors in plants (Lu et al., 1992).

The 14-3-3s are also involved in protein kinase activities (Ferl, 1996). In most of these

associations, 14-3-3s participate by binding to phosphorylated proteins in order to

consummate signal transduction events.

To fully understand the functions and interaction partners of the Arabidopsis

thaliana 14-3-3 family of proteins it is essential to know where the specific isoforms are

localized at the subcellular level. The biological question that this thesis attempts to

answer is this: Do the 14-3-3s have functional specificity? As previously mentioned, the

14-3-3s usually exist as multi-isoform families within a particular organism. Do multiple

isoforms exist simply to insure their presence in the proper tissues or does each isoform

play an essential and necessary role? Studies have been performed that give clues as to

specificity. For example, in Arabidopsis thaliana only isoforms epsilon, mu, nu, and

upsilon are present in chloroplasts (Sehnke et al., 2000). Why those four isoforms?

There is nothing obviously inherent about these isoforms that would make them

chloroplast localized. Mu and epsilon are the only two well-characterized isoforms

making up the epsilon sub-group. Nu and upsilon are closely related members of the

non-epsilon group. There are no obvious signal or targeting sequences in any of the

arabidopsis 14-3-3s. Therefore, they must be localizing to the chloroplast via interaction

with other proteins. It is the unpredictable nature of the 14-3-3s that makes localization

studies so important. We must know where the isoforms are localized.

Using laser scanning confocal microscopy, Bihn et al. reported the localization of

14-3-3s in the nucleus and cytoplasm using a mixture of ascites from three GF14









monoclonal antibody cell lines (Bihn et al., 1997). These cell lines combined recognize

at least ten of the thirteen 14-3-3 isoforms in Arabidopsis thaliana. However, the authors

did not localize individual 14-3-3 isoforms within the cell.

In the present study, we use laser scanning confocal microscopy along with

isoform specific polyclonal antibodies to study the subcellular localization of 14-3-3s

within the root tip of Arabidopsis thaliana. We find that individual isoform localization

occurs in two distinct groups, a nuclear localized group and a non-nuclear group.


Materials and Methods

Whole Mount Procedure

Arabidopsis thaliana seeds, ecotype Wassileweskija (WS), were surface sterilized

by washing with 70% ethanol for five minutes, further rinsed with 50% bleach/water

solution with a drop of Tween-20 for eight to ten minutes. The seeds were then rinsed

with sterile water five to eight times in order to remove detergent and bleach and plated

on germination media consisting of 4.4 g/L MS salts, 0.5 g/L MES, 10.0 g/L sucrose, 1.0

mL/L Gamborg's vitamin solution (1000x), pH 5.7, 4.0 g/L phytagel. After 48 hours at

4C in the dark, the plates were taken to the Apopka room and grown vertically under

continuous illumination for 3-10 days at 230C. The plants were removed from the plate

and fixed according to Shaw and Bourdonck protocol (Bihn et al., 1997). All

manipulations of the plant tissue starting with fixation were performed in a 6 or 12 well

microtiter plate. Solutions were added and removed to the dish and the plants were not

removed until mounted on a microscope slide. The plants were placed in freshly made

4% paraformaldehyde (prepared by dissolving 8% (w/v) solid paraformaldehyde in water

and then diluted with an equal volume of 2X PEM/0.4% Nonidet P-40 (NP-40) buffer









(PEM: 50 mM PIPES/KOH pH 6.9; 5 mM EGTA; 5 mM MgSO4/500 mL)) for 1 hour at

room temperature. The plants were then washed twice with 1X PEM/0.2% NP-40 and

then washed twice with PEM alone. The tissue was digested with 0.05%

cellulase/0.025% Y23 Pectolyase/1.0% Driselase in PEM for ten minutes and then

washed with PEM/0.2% NP-40 for ten minutes. The plant tissue was blocked with 3%

BSA in PEM for 1.5 hours at room temperature. The plants were subsequently incubated

with primary antibody in blocking buffer at 40C overnight. The next day the tissue was

washed in blocking buffer for one hour and then incubated with secondary antibody in

blocking buffer for two hours at 370C (dark). The plant tissue was finally washed several

times with PEM/0.2% NP-40 and then overnight in PEM alone. The next day, the plant

tissue was removed from the solution using a plastic disposable pipette and then placed

on the microscope slide. The excess liquid was removed and a small drop of Vectashield

mountant (Vector Laboratories Inc., Burlingame, CA 94010, USA) was applied and then

coversliped.

Immunofluorescence

Plant tissues were incubated with a polyclonal antibody recognizing a specific 14-

3-3 isoform, monoclonal antibody #65 that recognizes eight out of the 13 isoforms (mu,

epsilon, omicron, rho, and pi are excluded) (Bihn et al., 1997), Histone H1; 1:1000. They

were secondarily incubated with a FITC conjugated antibody (Jackson Laboratories,

Inc.); 1:2500.

Laser Scanning Confocal Microscopy and Image Analysis

Plant tissue samples were examined with an Olympus IX70 inverted microscope

mounted to a Bio-Rad MRC 1024ES laser scanning confocal microscope with a









krypton/argon laser. Collected images were Kalman averaged four to five times. The

images were collected using the Laser Sharp software package (Bio-Rad). The images

were transferred to a PC and cataloged using Thumbs Plus 4 (Cerious software).

Confocal Assistant 4.02 (Bio-Rad) was used to view collected Z-series files and convert

files from a .pic (Bio-Rad format) to an .avi format (suitable for Windows viewing of Z-

series).


Results

Polyclonal antibodies were generated by Dr. Paul Sehnke against specific 14-3-3

isoforms in arabidopsis. The arabidopsis 14-3-3s share a very conserved core region and

differ primarily at their amino and carboxyl termini. The polyclonal antibodies were

therefore made to peptides derived from either the N- or C-termini. At the time the

antibodies were generated, we did not yet know of the existence of isoforms omicron,

rho, and pi. Also, the psi isoform lacks enough unique sequence to generate an antibody

that would not cross react with the other isoforms. Therefore, nine polyclonal antibodies

were generated with each one recognizing a specific 14-3-3 isoform; phi, kappa, chi,

lambda, epsilon, upsilon, mu, nu, and omega.

Initial studies were performed to optimize three factors: antibody concentration,

age of plant tissue, and type of microscope to use. The proper antibody concentration

was determined empirically during several experiments. It was determined that a primary

antibody concentration of 1:1000 was sufficient for the 14-3-3 polyclonal antibodies.

The proper concentration of FITC conjugated secondary antibody was determined to be

1:2500. The combination of these two antibody dilutions provided enough fluorescence









to be seen, but not show a high fluorescence background. These studies were performed

using an epifluorescence microscope without image capture capabilities.

The 14-3-3 localization studies were performed on plants subjected to 48 hours at

4C in the dark and then grown vertically for 3 or 10 days. The goal was to determine

which tissues would work the best and if the localization of the isoforms within the tissue

changed over time. It was determined that the root tip provided the best tissue to work

with and there were no visible changes in localization among any of the 14-3-3 isoforms

(Figure 4-1). There was no difference in localization between the two ages. In fact, as

the plants progressed in age, they became more difficult to work with. Therefore, it was

determined that three day-old seedlings were the best candidates for further study.

The type of microscope to use was also an initial area of concern. The choice was

between an epifluorescence microscope with a mounted camera and a laser scanning

confocal microscope. We wanted to balance quality of information with cost. The

fluorescence microscope had a problem in that we could not get the light source low

enough in power to eliminate autofluorescence of the plant tissue. The laser scanning

confocal microscope eliminated this issue because it uses a laser and any light not in the

focal plane is discarded at image formation. Although the cost of the confocal

microscope is relatively high, it was clearly the best choice for our purposes.

Confocal microscopy allows several types of image collection. A single image or

optical slice can be captured and viewed. Also, a series of images through the sample can

also be captured and viewed as a movie. Both of these types of data are presented in this

work.









Upon examination of nine 14-3-3 isoforms in 4-5 day old Arabidopsis thaliana

root tips using the polyclonal antibodies and laser scanning confocal microscopy, we

were able to identify two distinct patterns of localization, a nuclear group and non-

nuclear group. Figure 4-2 is thumbnail images of nine 14-3-3 isoforms and control

antibodies for comparison, and individual enlargements are presented in Figures 4-3 and

4-4.

Isoforms phi, kappa, chi, and lambda make up the nuclear group (Figure 4-2 and

Figures 4-3a-d, respectively). The fluorescent signal is predominately localized to the

nucleus in these isoforms. A significant amount of signal can also be seen throughout the

cytoplasm. No other subcellular organelles are visible in this group of isoforms. Also,

there are no compartments in the cell devoid of fluorescence, such as vacuoles or other

structures. The intensity of the fluorescent signal cannot be directly compared between

images, but the signal can be compared in the same image, i.e. the nucleus is more

intense than the cytoplasm.

Isoforms epsilon, upsilon, mu, nu, and omega make up the non-nuclear group of

isoforms (Figure 4-2 and Figures 4-4e, respectively). The fluorescent signal is present

throughout the cell and no organelle shows a predominance of signal. Again, there are no

compartments within the cell devoid of fluorescence. However, in all the isoforms, gaps

are visible between the cells. This is possibly a product of the fixation process, rather

than an absence of signal in the cell wall. The plasma membrane could have pulled away

from the cell wall, leaving the gaps. However, the reason for the gaps remains unknown.

Several antibodies are used as controls to verify subcellular location (Figure 4-5).

Antibody #65 (Figure 4-5a) is a monoclonal antibody that recognizes eight out of the









thirteen well-characterized 14-3-3 isoforms (mu and epsilon are excluded and the

antibody has not been tested against the omicron, rho, or pi isoforms). Antibody #65 is

predominately found in the nucleus, but some signal can also be seen in the cytoplasm.

An antibody to Histone H1 (Figure 4-2 and Figure 4-5b) that associates with

nuclear material also serves as a control for nuclear localization. This antibody is clearly

localized to the nucleus. It provides a pattern of localization very similar to that of

antibody #65 and the nuclear 14-3-3 group.

Wildtype (WS) arabidopsis with no antibody treatment is used as a negative

control for antibody and as a control for autofluorescence. An image could not be

acquired for wildtype plants with no antibody treatment although both the iris and gain

were maximized during image capture (Figure 4-5c). Finally, plants treated with only the

secondary antibody serve as a control for non-specific antibody binding. Again, no

image could be acquired although the iris and gain were maximized at the time of image

capture (Figure 4-5d).


Discussion

The localization studies reveal two distinct patterns, nuclear and non-nuclear.

The isoforms phi, kappa, chi, and lambda make up the nuclear group. The isoforms

epsilon, upsilon, mu, nu, and omega make up the non-nuclear group. Both groups are

also localized to the cytoplasm. The nuclear vs. non-nuclear localization could be an

indication of isoform specificity. However, the polyclonal antibodies themselves pose

several problems. First, the way in which the antibodies were made posses a problem.

The antibodies to isoforms phi, kappa, chi, and lambda were all generated to the N-

terminus. The antibodies to isoforms epsilon, upsilon, mu, nu, and omega were all









generated to the C-terminus. All of the N-terminal antibodies make up the nuclear group

and all the C-terminal antibodies make up the non-nuclear group. It is possible that the

groupings are purely coincidental, although highly unlikely. Second, the process is lethal

to the plant. To use the antibodies, the plant must be fixed in 4% paraformaldehyde.

This process limits what can be done to the plant, i.e. time course experiments with the

same plant. Third, a series of enzymes must be used to degrade the cell wall. There is no

way to tell how much of the cell wall has or has not been degraded or what other

unforeseen effect the enzymes may have on the plant tissue.

The clean groups that the 14-3-3 antibodies fall into according to termini is highly

suspicious. I suspect that there is some unknown interaction that is blocking either the N-

or C-terminus of the 14-3-3 and causing the antibodies to localize in this particular

manner. The 14-3-3s commonly form homo- and hetero-dimers via their N-termini

forming a clamp like structure. The inner groove of the clamp forms the interaction site

for other proteins. If the 14-3-3s were dimerized, this might cause enough hindrance, so

that the antibodies would not be able to recognize the isoform, thus giving the false

appearance of distinct nuclear and non-nuclear groups. The individual isoforms might in

fact have a very distinct pattern of localization, but the antibodies are unable to recognize

the isoforms.

Penetration of the 14-3-3 antibodies into the deeper cell layers is also an

important issue. Figure 4-6 is a graphical representation of an Arabidopsis thaliana root

tip. The cell layers from outermost to innermost are lateral root cap, epidermis, cortex,

endodermis, pericycle, and vascular bundle. The 14-3-3 localizations in this study are

limited to the lateral root cap and possibly the epidermis. Figure 4-7 shows a series of









four images through an Arabidopsis thaliana root tip treated with monoclonal antibody

#65. These images were taken from a more detailed series, but the overall meaning is

clear. Image 7a shows clear fluorescent labeling of the nuclei, however as you progress

through the root tip (4-7b and 4-7c), no nuclei are labeled. Yet, in 4-7b and 4-7c, nuclei

in the outer cell layers are still visible. Finally, in 4-7d, the nuclei reappear on the

opposite side of the root tip. This effect can be seen with all of the 14-3-3 isoforms to

some degree. It is easier to see with the nuclear localized isoforms rather than the non-

nuclear isoforms. It is my hypothesis that the antibody cannot get into these inner layers

of the root tip, and thus no labeling is occurring. Perhaps with longer incubation times in

the digestion mix, it would be possible to visualize the inner cell layers.






























4-1a




















4-lb


Oea(w


Figure 4-1. Arabidopsis thaliana root tips treated with a 14-3-3 polyclonal antibody. The
images were collected using laser scanning confocal microscopy. (a) Three
day-oroot tip treated with a polyclonal antibody to the omega 14-3-3 isoform.
The magnification is 51x. (b) Ten day-old root tip treated with a polyclonal
antibody to the omega 14-3-3 isoform. The magnification is 47x.






























Thumbnail images of Arabidopsis thaliana root tips treated with nine 14-3-3
polyclonal antibodies and control antibodies. The thumbnails are labeled as to the
specific isoform used. The 14-3-3s fall into two distinct groups, a nuclear and non-
nuclear group. Phi, kappa, chi and lambda make up the nuclear group. Epsilon,
upsilon, mu, nu, and omega make up the non-nuclear group. Two controls are also
shown. The image labeled "monoclonal" is a single monoclonal antibody that
recognizes eight out of the ten well-characterized isoforms (mu and epsilon are
excluded). Histone H1 is a monoclonal antibody that is included as a control to
identify nuclei.


Figure 4-2.



























pin W













Lambda













Mu (lt)


Kappa (Y,)













Epsilon (s)


Chi(z)


Mooloa


Fhstone H I




















































Figure 4-3a. Arabidopsis thaliana root tip treated with a polyclonal antibody to the phi
14-3-3 isoform. The image was collected using laser scanning confocal
microscopy at a magnification of 57x. The localization pattern is
predominately nuclear, although some signal can be seen in the cytoplasm.



















































Figure 4-3b. Arabidopsis thaliana root tip treated with a polyclonal antibody to the kappa
14-3-3 isoform. The image was collected using laser scanning confocal
microscopy at a magnification of 45x. The localization pattern is
predominately nuclear, although some signal can be seen in the cytoplasm.




















































Figure 4-3c. Arabidopsis thaliana root tip treated with a polyclonal antibody to the chi
14-3-3 isoform. The image was collected using laser scanning confocal
microscopy at a magnification of41x. The localization pattern is
predominately nuclear, although some signal can be seen in the cytoplasm.



















































Figure 4-3d. Arabidopsis thaliana root tip treated with a polyclonal antibody to the
lambda 14-3-3 isoform. The image was collected using laser scanning
confocal microscopy at a magnification of 61x. The localization pattern is
predominately nuclear, although some signal can be seen in the cytoplasm.





















































Figure 4-4a. Arabidopsis thaliana root tip treated with a polyclonal antibody to the
epsilon 14-3-3 isoform. The image was collected using laser scanning
confocal microscopy at a magnification of 32x. The localization pattern is
predominately cytoplasmic. No subcellular organelles are visible.




















































Figure 4-4b. Arabidopsis thaliana root tip treated with a polyclonal antibody to the
upsilon isoform. The image was collected using laser scanning confocal
microscopy at a magnification of 34x. The localization pattern is
predominately cytoplasmic. No subcellular organelles are visible.









































9U(t
28


Figure 4-4c. Arabidopsis thaliana root tip treated with a polyclonal antibody to the mu
14-3-3 isoform. The image was collected using laser scanning confocal
microscopy at a magnification of 28x. The localization pattern is
predominately cytoplasmic. No subcellular organelles are visible.





















































Figure 4-4d. Arabidopsis thaliana root tip treated with a polyclonal antibody to the nu
14-3-3 isoform. The image was collected using laser scanning confocal
microscopy at a magnification of 45x. The localization pattern is
predominately cytoplasmic. No subcellular organelles are visible.





















































Figure 4-4e.


Arabidopsis thaliana root tip treated with a polyclonal antibody to the
omega 14-3-3 isoform. The image was collected using laser scanning
confocal microscopy at a magnification of 5 Ix. The localization pattern is
predominately cytoplasmic. No subcellular organelles are visible.



















































Figure 4-5a. Arabidopsis thaliana root tip treated with a monoclonal antibody
recognizing eight out of the ten well-characterized 14-3-3 isoforms (mu and
epsilon excluded). The image was collected using laser scanning confocal
microscopy at a magnification of 35x. This is a compression image of a
larger series of images from a z-series. The localization pattern is
predominately nuclear, although some signal is present in the cytoplasm.






















































Figure 4-5b. Arabidopsis thaliana root tip treated with a monoclonal antibody to Histone
H1. The image was collected using laser scanning confocal microscopy at
a magnification of 40x. Histone H1 is located in the nucleus. As expected,
the localization pattern is predominately nuclear.















































Figure 4-5. Arabidopsis thaliana root tips treated with either no antibody or
FITC conjugated secondary antibody only. The images were collected using
laser scanning confocal microscopy a magnification of 20x. (a) Wildtype
plants with no antibody treatment. (b) Wildtype plants treated with FITC
conjugated secondary antibody alone.


Wildtype

4-5c 20x













Secondary
Only

4-5c 20x









































Figure 4-6. Cell layers of an Arabidopsis thaliana root tip. Six cell layers make up the
root tip: lateral root cap, epidermis, cortex, endodermis, pericycle, and
vascular bundle, form outermost to innermost. Additionally, there are four
layers of columella cells immediately in front of the root meristem. This
diagram was taken directly from the website of Dr. Jim Haseloff,
http://www.plantsci.cam.ac.uk/Haleloff/Arabidopsis/architecture.



















































Arabidopsis thaliana root tip treated with a monoclonal antibody recognizing
eight of the ten well-characterized 14-3-3 isoforms (mu and epsilon
excluded). The image was collected using laser scanning confocal
microscopy at a magnification of 20x. This is a set of four image taken from
a larger z-series through the root tip. The nuclear signal is only present in the
first or second cell layers. The lack of signal in the interior cell layers maybe
due to inaccessibility of the antibody.


Figure 4-7.















CHAPTER 5
USING GREEN FLUORESCENT PROTEIN FOR 14-3-3 LOCALIZATION


Introduction

The 14-3-3s are a family of acidic, soluble proteins that were initially discovered

in 1967 during a study of cytosolic bovine brain proteins (Moore and Perez, 1967). Once

thought to be limited to brain tissue, studies over the last fifteen years have proven 14-3-

3s to be ubiquitous. The 14-3-3s usually consist of multiple isoforms per organism and

have been found in virtually every eukaryotic organism where scientists have looked for

them (Robinson et al., 1994).

Once the importance of 14-3-3s was established in the late 1980s, the original

mammalian isoforms were re-characterized. There are nine isoforms in mammals and

they are designated as Greek letters starting at the beginning of the alphabet (Ichimura et

al., 1988). In keeping with that trend, the Arabidopsis thaliana 14-3-3s were named

starting at the end of the Greek alphabet as they were discovered. The arabidopsis 14-3-3

family consists of ten well-characterized isoforms and three newly recognized members.

The established 14-3-3 isoforms are omega, psi, chi, phi, upsilon, lambda, kappa, mu, nu

and epsilon (Lu et al., 1992; Lu et al., 1994b; Wu et al., 1997). The newly recognized

members are omicron, rho, and pi (Rosenquist et al., 2000; DeLille et al., 2001).

The functions of 14-3-3s are many. The general theme of protein-protein

interaction seems to fit well. 14-3-3s are known to regulate key enzymes in metabolic

pathways such as nitrate reductase and sucrose phosphate synthase (Finnie et al., 1999).









Specifically in plants, 14-3-3s are also involved in transcription complexes (Lu et al.,

1992). These are just a few of the many known functions of 14-3-3s.

The reason for the large number of 14-3-3 isoforms per organism has always been

a key question. Arabidopsis has the largest number of isoforms with 13 and tomato is a

close second with 10 members. Is there a specific function for each isoform? Or do the

isoforms serve redundant functions and exist in multiple isoforms to insure their presence

in every tissue? The question remains largely unanswered.

To try to answer the multiple isoform question, we must know where all the

isoforms are located on the subcellular level as well as whole tissue. This chapter

attempts to localize seven of the arabidopsis 14-3-3 on the subcellular level using in-

frame C-terminal green fluorescent fusions.


Materials and Methods

Oligonucleotide Design

Oligonucleotides were designed to produce a full-length 14-3-3 isoform with the

stop codon removed and either an Spel for the omega isoform or an Xbal restriction site

for all other isoforms on both the 5' and 3' ends (Gibco BRL and Integrated DNA

Technologies, Inc). The omega isoform used Spel restriction sites due to the presence of

internal XbaI sites in the omega coding region.

Cloning of C-terminal Green Fluorescent Protein Fusions

The coding sequence for the 14-3-3 isoforms phi, psi, kappa, chi, lambda, epsilon,

upsilon, omega, and nu were amplified in a polymerase chain reaction (PCR) (lx

polymerase buffer, 0.1 mM dNTPs, -500 ng of DNA template, 3.8 [tg/mL primer, and

2.5 U Vent polymerase. The templates for the PCR reactions were the individual 14-3-3









isoform cDNAs in the expression vector, pET-15b, with the exception of omega, which

was in pUC18. The PCR product was gel purified using the QIAquick gel extraction kit

(Qiagen). The PCR products were then cloned into the pCR-blunt vector (Invitrogen) as

an Xbal/Xbal or SpeI/Spel fragment lacking the stop codon, and transformed into

DH10B competent cells for large-scale plasmid preparations. The presence of the

isoform in pCR-blunt was confirmed by restriction analysis. Each pCR-blunt vector

containing an individual 14-3-3 isoform was restricted using Xbal or Spel, releasing the

entire isoform from the pCR-blunt vector. The proper size fragment was gel purified and

further subcloned into pBI12135sGFP (S65T) and transformed into DH10B competent

cells. This puts the entire construct under the control of the CaMV 35S promoter and

GFP is fused to the C-terminus of the 14-3-3. The pBI12135sGFP (S65T) construct was

made by Dr. Paul Sehnke and Michael Manak. The 14-3-3/GFP fusions were subjected to

restriction analysis to confirm orientation of the isoforms, and sequence analysis to verify

that they were truly fusions. The University of Florida ICBR sequencing core performed

all sequence analysis. The 14-3-3/GFP fusions in pBI12135SGFP (S65T) were then

transformed into agrobacterium (GV 3101). An additional construct made by Michael

Manak includes pBI1O1GFP (S65T), which is a promoterless construct and serves as a

negative control.

Vacuum Infiltration Protocol

This protocol was adapted from the Masters thesis of Elizabeth Bihn, formally of

the Ferl lab. Arabidopsis thaliana, ecotype Wassilevskija (WS), seeds were planted in a

plastic four pack with soil, covered with screen door mesh and secured with two rubber

bands. The pots were placed at 40C for two days in the dark. The muffins were then









moved to the Apopka room, where they received constant light at 230C. Once the plants

started to germinate, they were thinned down to -20 plants per muffin. The plants

continued to grow until they started to bolt. The bolts were cut everyday for the next 3-4

days with scissors. On the fifth day, the bolts were not cut and vacuum infiltration was

performed.

For infiltration, one 500 mL culture of agrobacterium strain GV 3101 per clone

(nine total) containing the 14-3-3/GFP fusion were grown in 2YT media ( 10 g/L yeast

extract, 16 g/L tryptone, 5 g/L NaC1, pH 7.3) for several days at 280C. Cultures were

spun down in a centrifuge at 7,000 rpm for ten minutes and re-suspended in 500 mL of

infiltration media (2.15 g/L Murashige and Skoog salts, ImL/L Gamborg's B5 vitamins,

50 g/L sucrose, 10 gL/L ofbenzylamino purine (1 mg/mL stock), 50 ptL Silwet L-77).

The suspension was placed in a 600 mL beaker, which was placed inside a vacuum

chamber. Plants were placed upside down in the infiltration media suspension and the lid

was placed on the chamber. A vacuum was drawn for seven minutes. Plants were

removed from the suspension, placed on their sides, covered with saran wrap, and

allowed to recover overnight under constant light at 230C. The next day, the plants were

righted, allowed to grow normally and set seed. The seed was plated onto MS media

plates (4.4 g/L Murashige and Skoog salts, 0.5 g/L MES, 1 mL 1000X Gamborg's B5

vitamins, pH 5.7 with KOH, 4.0 g/L phytagel) containing carbenicillin (250 mg/L) and

kanamyacin sulfate (50mg/L) to kill agrobacteria and select for transformants,

respectively.









Protein Extraction and Western Blot Analysis

Fresh tissue (100-200 mg) was used in protein extractions. The tissue was placed

into a pre-chilled 1.5 mL tube with equal volume of cold extraction buffer (50 mM Tris

pH 7.6, 150 mM NaC1, 0.1% Tween-20, 25 mM NaF, 0.1 mM EDTA). Tissue was

ground with a polytron grinder at a setting of forty for two minutes and immediately

placed back on ice. Samples were centrifuged at 14,000 rpm for five minutes at 40C. 60

tiL of supernatant was removed and an equal volume of pre-warmed 2x SDS sample

buffer was added. All samples were heated in boiling water for two minutes. 30 ptL of

sample was loaded per lane and subject to 10% SDS-PAGE. The proteins were electro-

transferred to a nitrocellulose membrane and taken through a standard Western-blotting

procedure. Rabbit anti-GFP (Santa Cruz Biotechnology) was used as a primary antibody

at a dilution of 1:400 and polyclonal 14-3-3 antibodies were used at either 1:1000 or

1:2000. Goat anti-rabbit horseradish peroxidase was used as secondary antibody at a

dilution of 1:1500.

Whole Mount Sample Preparation

Transgenic seeds were surface sterilized and plated onto MS media. After two

days at 40C in the dark, the seedlings were placed under constant light at 230C and grown

vertically for 3-7 days. The entire plant was removed from the media plate and placed on

a regular microscope slide with a drop of water and coversliped. Some samples were

treated with propidium iodide (10 ptg/mL) for ten minutes to stain cell walls.

Laser Scanning Confocal Microscopy and Image Analysis

Plant tissue samples were examined with an Olympus IX70 inverted microscope

mounted to a Bio-Rad MRC 1024ES laser scanning confocal microscope with a









krypton/argon laser. Collected images were Kalman averaged four to five times. The

images were collected using the Laser Sharp software package (Bio-Rad). The images

were transferred to a PC and cataloged using Thumbs Plus 4 (Cerious software).

Confocal Assistant 4.02 (Bio-Rad) was used to view collected z-series files and convert

files from a .pic (Bio-Rad format) to an .avi format suitable for Windows viewing of z-

series.


Results

The individual 14-3-3 isoform coding sequences for omega, chi, psi, phi, upsilon,

lambda, kappa, epsilon, and nu were amplified using PCR based methods. All of the

primers were designed to remove the stop codons from each isoform and to put an Xbal

restriction site on the 5' and 3' ends. Omega used a Spel restriction site due to internal

Xbal sites within the omega coding region (Table 5-1). The nine isoforms were blunt end

cloned into the pCR-blunt vector (Invitrogen). The isoforms were then restricted using

either Xbal or Spel, thus releasing the entire isoform. The 14-3-3s were then cloned into

the vector pBI12135SGFP (S65T). The result is an individual 14-3-3 isoform C-

terminally fused in-frame with GFP. The constitutive CaMV 35S promoter drives the

expression of the fusion construct. A graphical representation of the fusion construct is

presented in Figure 5-1. All nine isoforms were successfully fused in this manner as

confirmed by sequence analysis of the 14-3-3/GFP junction.

The pBI12135S14-3-3GFP (S65T) fusion constructs were transformed into

agrobacterium for plant transformation. Arabidopsis thaliana, ecotype Wassilevskija

(WS), was transformed with each of the nine 14-3-3 constructs. The transgenic plants

were allowed to set seed and were ultimately taken through at least two generations.









Two of the constructs, chi and psi, did not express GFP in any of the Ro seed.

The 14-3-3/GFP fusions themselves were verified by sequence analysis. Therefore, it is

my opinion that the transformation process itself was not successful. The other remote

possibility is that they are undergoing gene silencing. However, this is unlikely given the

nature of the 35S promoter. Re-transformation of these constructs is planned in future

experiments.

Western blot analysis was performed on the remaining seven 14-3-3/GFP fusions.

Whole plant tissue was used in a protein extraction protocol found in materials and

methods. The protein was subjected to a 10% SDS-PAGE and further subjected to

electro-transfer to a nitrocellulose membrane. The membrane was incubated with a

rabbit anti-GFP antibody, and a goat-anti-rabbit secondary antibody conjugated to

horseradish peroxidase. The results of the western can be seen in Figure 5-2a. All of the

isoforms have a band present at -55 kDa. GFP is -27 kDa and the 14-3-3 proteins are

between 27-30 kDa. Therefore, you would expect a protein of -55 kDa if they were true

fusions. In a separate experiment, the same procedure was followed and the membrane

was incubated with 14-3-3 polyclonal antibodies, kappa, omega, lambda, phi, epsilon,

upsilon, and nu. The membrane was secondarily incubated with goat-anti-rabbit

horseradish peroxidase antibody. The results of the Western can be seen in Figure 5-2b.

Two distinct bands are visible for omega, epsilon, upsilon, and nu isoforms. The lower

band is endogenous 14-3-3 and is -27-30 kDa. The upper band at -55-60 kDa is the 14-

3-3/GFP fusion. A single band at -55 kDa is present for the lambda isoform and a faint

band is visible for the kappa isoform. There are no visible bands present for the phi

isoform. Figure 5-2c is a Western blot analysis of wildtype plants, plants transformed









with pBI12135SGFP (S65T) only and plants transformed with epsilon/GFP and

kappa/GFP fusion constructs. Rabbit anti-GFP was used as the primary antibody. No

bands are visible for the wildtype lane. A band at -27 kDa is visible for the GFP only

lane and a band at -55 kDa is visible for both the epsilon/GFP and kappa/GFP lanes. The

amount of tissue used was not consistent between constructs and no effort was made to

load the samples equally in either experiment. The Western blots were used simply to

demonstrate that the plants were transformed with GFP and that they were fused to a 14-

3-3.

Three to four lines per construct were examined to confirm that there were no

major differences in localization due to position effects. Normally many more lines per

construct would be used, however GFP and kanamyacin selection allow fewer lines to be

examined. False-positives can be eliminated by visual inspection of GFP expression in a

large number of plants. Transgenic plants were grown vertically on MS media under

constant light for 3-7 days. The plants were removed from the media plates and mounted

in water and viewed with a laser scanning confocal microscope. The seven 14-3-3/GFP

fusions demonstrate variable locations within the cell.

Epsilon/GFP is localized predominately to the nuclear envelope, plasma

membrane and cytoplasm. Figure 5-3a shows cells at the root tip. A small ring of

brighter fluorescence can be seen wrapped around the nucleus and is thought to be the

nuclear membrane. No signal can be seen in the nucleolus and only a slight signal,

compared to the rest of the cell, can be seen in the nucleus. Figure 5-3b is a magnified

view of cells not at the true root tip, but further up the root. In these cells, a clear ring

around the nucleus can be seen and also structures streaming from the nucleus can be









seen. These streaming structures could be endoplasmic reticulum or golgi apparatus,

however the data is difficult to interpret. Again, no signal is present in the nucleolus.

Figure 5-4 shows single images taken from a larger series of images through a root tip

expressing the epsilon/GFP fusion. A contrast can be seen between the cells of the true

root tip versus cells higher up the root. The cells higher up are elongated and appear to

be less cytoplasmic and more vacuolated. The cells at the true root tip appear to be less

vacuolated and more cytoplasmic in nature. However, the location of the epsilon/GFP

signal within the cell does not change.

Kappa/GFP is predominately localized to the nucleus, plasma membrane and cell

wall. A small amount of fluorescence can be seen in the cytoplasm as well. Figure 5-5a

is a compressed image of a complete z-series through the root sample. In short, the Laser

Sharp software takes all of the individual images (65 images in this series) and combines

them into one complete image. The results a three-dimensional looking image. Figure

5-5b is a higher magnification of a root tip expressing kappa/GFP. There is no signal in

the nucleolus and no signal in what appear to be vacuoles. Figure 5-6a is a root tip

expressing kappa/GFP using the FITC filter set. Figure 5-6b is the same root tip that has

been counter stained with propidium iodide and viewed using the Texas Red filter set.

This stain predominately labels the cell walls, however longer incubation times allow the

nuclei to become labeled. Figure 5-6c is a dual image of the green filter and the red filter

combined. The result is overlap of the fluorescence of GFP and the propidium iodide in

what appears to be the cell wall, which creates a yellow color. Figure 5-7 is a series of

single images taken from a larger z-series that shows the progression through the root tip.

The fluorescence in the nucleus, plasma membrane and cell wall do not change.









Figure 5-8a is of lambda/GFP showing localization in the nucleus, plasma

membrane and cytoplasm. Figure 5-8b is a higher magnification of the same sample.

The nucleoli as well as the vacuoles appear to be absent of any fluorescence. There are

not any subcellular structures that are more intensely labeled than another with this

isoform. The fluorescent signal appears to be evenly distributed among the structures.

Figure 5-9 is a series of four images taken from a more detailed z-series. These images

are from a different sample than those of Figure 5-8. The overall fluorescence is higher

in Figure 5-9, however the ratio of fluorescence among subcellular structures remains the

same.

Figure 5-10a is of omega/GFP demonstrating localization in the cytoplasm,

plasma membrane, and nucleus. Again, signal is absent from the nucleolus and what

appear to be vacuoles. Figure 5-10b is a higher magnification of a different sample

expressing omega/GFP. This figure shows what could be endoplasmic reticulum or the

golgi apparatus expressing GFP as well. Figure 5-10c represents the true root tip

extremely well. The columella root cap and meristem are clearly visible.

Phi/GFP is predominately localized to the cytoplasm, nuclear membrane and

plasma membrane (Figure 5-1 la). However, slight fluorescence can also be seen in the

nucleus. Structures that appear to be vacuoles and the nucleolus lack fluorescence in

Figure 5-11a. Figure 5-11b is a higher magnification image of a different sample

expressing phi/GFP. Other unknown structures are also labeled and visible in Figure 5-

1 lb. However, without a cellular marker, it is difficult to identify them with any level of

confidence. Figure 5-12 is a series of five images taken from a larger, more detailed z-

series through the root tip. The first image, Figure 5-12a, starts high and the last image,









Figure 5-12g, ends at the true root tip. The common theme of elongated, less cytoplasmic

cells higher up the root tip, proceeding to heavily cytoplasmic cells at the true root tip is

demonstrated well here.

Upsilon/GFP is also predominately localized to the cytoplasm, nuclear membrane

and plasma membrane (Figure 5-13a). A higher magnification of a different sample

expressing upsilon/GFP can be seen in Figure 5-13b. Again, fluorescent signal is absent

from the nucleolus and what appear to be vacuoles. A slight fluorescent signal can be

seen in the nucleus, but it is not as intense as the signal in the cytoplasm. Figure 5-14 is a

series of four images taken from a larger z-series. These images show the progression

through a root tip. The GFP signal does not change location in the cytoplasm, nuclear

membrane, or plasma membrane. As the images progress through the root tip, many

unknown cellular structures show fluorescence.

Nu/GFP is predominately localized to the cytoplasm, plasma membrane and

nuclear membrane (Figure 5-15a). Figure 5-15b represents a higher magnification of

Nu/GFP. The nucleolus is void of any signal and the nucleus is only slightly fluorescing.

Several unknown structures are also visible with this isoform as well. Figure 5-16 is a

series of four images taken from a larger series of images through the root tip. As the

images progress, no major changes in localization can be seen.

Plants were also transformed with the pBI12135SGFP (S65T) vector with no 14-

3-3 present as a control. These plants demonstrate localization largely in the nucleus and

some signal is present in the cytoplasm (Figure 5-17a). Signal is absent from the

nucleolus and appears to be absent from the vacuoles. A vector was also transformed into

arabidopsis that contains GFP (S65T), but lacks a promoter, pBI1O1GFP. An image









could not be acquired although the gain and iris were maximized at the time of image

capture (Figure 5-17b). Finally, wildtype arabidopsis is used as a control for

autofluorescence (Figure 5-17c). Again, the iris and gain were maximized, but no image

was visible.


Discussion

Nine constructs were made C-terminally fusing an individual Arabidopsis 14-3-3

member to green fluorescent protein. The successfully fused 14-3-3s are kappa, omega,

lambda, phi, epsilon, upsilon, nu, chi, and psi. All fusion constructs were verified by

sequence analysis before being transformed into agrobacterium. All nine constructs were

subsequently transformed into arabidopsis. Upon screening the Ro seed, no visible

fluorescence was seen for the chi and psi isoforms, therefore they were dropped from

further study. Chi and psi will be re-transformed into arabidopsis at a later date.

Western analysis of the remaining seven constructs revealed some interesting

results. All seven isoforms reveal a proper size band at -55 kDa when treated with anti-

GFP antibody. In the blot treated with the 14-3-3 antibodies, the results of omega,

epsilon, upsilon, and nu are exactly as expected. The specific 14-3-3 antibody recognizes

two different size bands, the endogenous 14-3-3 and the 14-3-3/GFP fusion. The lambda,

kappa, and phi 14-3-3 antibodies provide a much different result. A single band is visible

for the kappa and lambda isoforms at -55 kDa. However, no endogenous 14-3-3 is

visible. This result could be an indication that the endogenous kappa and lambda

isoforms are in low abundance. There is not an endogenous or fusion band visible for the

phi isoform. Multiple attempts to show this isoforms presence, using the 14-3-3

antibody, have failed. Varying concentration of antibody and tissue amount has had no









effect. This is curious considering a clear band is visible at the proper size when treating

with the anti-GFP antibody. The lack of both an endogenous and fusion band lead me to

believe there is a problem with the antibody recognizing the 14-3-3, rather than a

problem with the 14-3-3/GFP fusion. Finally, it is important to show that the GFP

antibody is specific. No bands are visible for wildtype plants using anti-GFP antibody.

Additionally, a size shift can be seen between plants transformed with pBI12135SGFP

(S65T) only and those fused to a 14-3-3 as in Figure 5-2c.

The 14-3-3/GFP fusions demonstrate a dynamic localization pattern in the root tip

of Arabidopsis. Table 5-2 is a summary of the predominate localization patterns for

seven of the 14-3-3/GFP isoform fusions. The location is ordered from strongest to

weakest. The isoforms appear to localize along evolutionary lines in some instances. For

example, the kappa and lambda isoforms make up one sub-branch of the non-epsilon

isoforms in a neighbor-joining parsimony tree. Furthermore, both the kappa and lambda

isoforms appear to be present strongly in the nucleus and plasma membrane/cell wall.

No other isoforms have this particular localization pattern. This could point to a specific

and evolutionarily conserved role for these two isoforms. Also, the presence of the kappa

isoform in the cell wall is a unique finding, in that no other arabidopsis 14-3-3 has been

localized to the cell wall. This result, however, is not conclusive. An argument could be

made that although the signal appears to be yellow, it could because the green in the

plasma membrane is so close to the red of the cell wall. Clearly, this result needs to be

investigated further.

The remaining five isoforms have a similar localization pattern among

themselves, although there is some variance. For example, the epsilon isoform looks









predominately localized to the nuclear envelope and to a lesser degree in the plasma

membrane and cytoplasm. The phi isoform is predominately localized to the cytoplasm,

however, it is also localized to the nuclear membrane and plasma membrane to a lesser

extent. However, evidence is present of these isoforms falling along evolutionary lines as

well. Upsilon and nu are on the same sub-branch in a neighbor- joining parsimony tree

and they also have very similar localization patterns. Both are predominately localized to

the cytoplasm and to a lesser extent the nuclear membrane and plasma membrane.

Although not as dramatic as lambda and kappa, these isoforms could also be localized

according to evolutionary relatedness.

The GFP fusions provide a unique look at 14-3-3 localization within the cell. The

GFP fusions also provide a valuable tool for future studies. Other tissues within the

plants are clearly expressing the 14-3-3/GFP fusion and need to be examined in more

detail. These seven constructs can also be used in future environmental stress

experiments to examine changes in 14-3-3 localization. Further, native promoters can

easily replace the current constitutive promoter for in-depth studies.

















Table 5-1. Primers used to amplify the individual 14-3-3 isoforms from expression
vectors. Each primer has an Xbal restriction site incorporated into the 5' and
3' ends, with the exception of omega which uses Spel. Additionally, the
primers were designed to remove the stop codon at the 3' end of each isoform.



Epsilon Forward: 5' GCTCTAGAACAATGGAGAATGAGAGGGAAAAGCAG 3'
Reverse: 5' TCTAGAGTTCTCATCTTGAGGCTCATCAGCACC 3'

Kappa Forward: 5' GCTCTAGAACAATGGCGACGACCTTAAGCAGA 3'
Reverse: 5' TCTAGAGGCCTCATCCATCTGCTCCTGCATATC 3'

Lambda Forward: 5' GCTCTAGAACAATGGCGGCGACATTAGGC 3'
Reverse: 5' TCTAGAGGCCTCGTCCATCTGCTCCTGCAT 3'

Omega Forward: 5' GCACTAGTACAATGGCGTCTGGGCGTGAA3'
Reverse: 5' ACTAGTCTGCTGTTCCTCGGTCGGTTTTGG 3'

Psi Forward: 5' GCTCTAGAACAATGTCGACAAGGGAAGAGAATG 3'
Reverse: 5' TCTAGACTCGGCACCATCGGGCTTTGATGC 3'

Nu Forward: 5' GCTCTAGAACAATGTCGTCTTCTCGGGAAGAG 3'
Reverse: 5' TCTAGACTGCCCTGTCTCAGCTGGTTTCCC 3'

Phi Forward: 5' GCTCTAGAACAATGGCGGCACCACCAGCA 3'
Reverse: 5' TCTAGAGATCTCCTTCTGTTCTTCAGCAGG 3'

Upsilon Forward: 5' GCTCTAGAACAATGTCTTCTGATTCGTCCCGG 3'
Reverse: 5' TCTAGACTGCGAAGGTGGTGGTTGGGC 3'

Chi Forward: 5' GCTCTAGAACAATGGCGACACCAGGAGCTT 3'
Reverse: 5' TCTAGAGGATTGTTGCTCGTCAGCGGGTTT 3'


























35S promoter


Figure 5-1.


General representation of the completed Arabidopsis thaliana 14-3-3/GFP
fusion constructs. Nine individual Arabidopsis thaliana 14-3-3 isoforms
were fused to green fluorescent protein in this manner. The constitutive
Cauliflower Mosaic 3 5 S promoter drives the expression of the fusion
constructs. The vector also contains the Kanamyacin resistance gene
(NPT II).

















69 kDa 1-


46 kDa --


Figure 5-2a


a1
o
a

jr

M


69 kDa --

46 kDa 1


Figure 5-2b


Protein extraction and Western blot analysis. (a) Proteins were extracted
from whole plant tissue and subjected to SDS-PAGE. Proteins were electro-
transferred to a nitrocellulose membrane and treated with anti-GFP antibody.
A single band at -55 kDa is visible for all seven constructs demonstrating
that GFP is fused to another protein (14-3-3). (b) A similar procedure was
followed as in (a). The nitrocellulose membrane was treated with polyclonal
antibodies to the Arabidopsis thaliana 14-3-3s. A band at -55 kDa can be
seen for all constructs except phi. Additionally, the isoforms omega, epsilon,
upsilon, and nu have a band at -30 kDa. This lower band is endogenous 14-
3-3. No bands are present for the phi isoform.


Figure 5-2.

























-





69 kDa

46 kDa -







Figure 5-2c. Protein extraction and Western blot analysis. Proteins were extracted from
whole plant tissue and subjected to SDS-PAGE. Proteins were electro-
transferred to a nitrocellulose membrane and treated with anti-GFP
antibody. No bands are present for the lane containing wildtype plants. A
single band at -27 kDa is visible for the lane labeled "GFP (S65T) only."
Finally a single band at -55 kDa can be seen for both epsilon/GFP and
kappa/GFP fusion constructs.































Epsilon/GFP fusion construct demonstrates localization predominately in the
nuclear envelope, plasma membrane and cytoplasm. No signal can be seen
in the vacuoles or nucleolus. (a) Image of root tip at 1 1Ox magnification.
(b) Image of cells further up the root tip at 220x magnification.


Figure 5-3.









































Esio (0 ,
5-3 gl























Series of four images taken from a larger z-series of epsilon/GFP. The
images show that cells distal to the root tip are elongated and less
cytoplasmic, however that cells at the true root tip are more cytoplasmic.
The images are at 115x magnification.


5-a E sln(


Figure 5-4.


i 5-4b
ism Epsi





5-4d Epsilon (E;)





























Kappa/GFP fusion construct demonstrates localization predominately in the
nucleus, plasma membrane, and cell wall. No signal can be seen in the
vacuoles or nucleolus. (a) A compressed image of a complete z-series
through the root tip at 60x magnification. (b) A single optical slice of cells
in the root tip of the same plant at 1 1Ox magnification.


Figure 5-5.



















































Kappm(K

5-5 gl
















































Kappa/GFP fusion in the root tip. (a) Image of a root tip expressing
kappa/GFP collected using a FITC filter set. (b) Image of the same root tip
using a Texas Red filter set after staining with propidium iodide to identify
the cell walls. (c) Image of the same root tip using a dual filter set that
collects both green and red signals. Any overlap in the two colors results in
yellow. There appears to be a yellow signal in the cell walls, indicating 14-
3-3/GFP presence. All images are at 60x magnification.


Figure 5-6.

















































Series of four images taken from a larger z-series of kappa/GFP. The images
show no change in signal as the z-series progresses. Again, localization is
predominately in the nucleus, plasma membrane, and cell wall. These
images are at 60x magnification.


Figure 5-7.


V











5-7a Kappa (K)
















5-7c Kappa (K)


5-7b Kappa (K)
















5-7d Kappa (K)





























Lambda/GFP fusion construct demonstrates localization predominately in the
nucleus, plasma membrane and cytoplasm. No signal can be seen in the
vacuoles or nucleolus. (a) Magnification at 60x. (b) Magnification at 110x.


Figure 5-8.



































gaba k



















































Series of four images taken from a larger z-series of lambda/GFP. The
overall fluorescence is more intense in these images, but the ratio of intensity
within the cell doesn't change from that of (a) and (b). These images were
collected at a magnification of 115x. No signal is visible in the vacuoles or
nucleolus.


5-cLmdSX


Figure 5-9.


5-9b Lambda (X)
















5-9d Lambda (X)






























5-10 Omg (o


Figure 5-10.


Omega/GFP fusion construct demonstrates localization predominately in
the cytoplasm, plasma membrane, and nucleus. (a) Image collected at a
magnification of 60x. (b) Image collected at a magnification of 180x. (c)
Image collected at a magnification of 97x. No signal is visible in the
vacuoles or nucleolus in any of these images. The columella root cap and
meristem are shown well in (c).


5-10b Omega (co)


5-10c Omega (o))































Figure 5-11. Phi/GFP fusion construct demonstrates localization predominately in the
cytoplasm, nuclear membrane, and plasma membrane. (a) A single optical
slice through the root tip at a magnification of 106x. (b) A single optical
slice through the root tip at a magnification of 166x. No signal is visible in
the vacuoles or nucleolus in either image.














































Ph
5-1 106








































5-12a P i


5-12b Ph


Figure 5-12. Series of four images taken from a larger z-series of phi/GFP. The signal
does not change in any of the images as the z-series progresses. These
images are the same sample as 5-11 and are at a magnification of 106x. No
signal can be seen in the vacuoles or nucleolus in these images.


































Figure 5-13.


Upsilon/GFP fusion construct demonstrates localization predominately in
the cytoplasm, nuclear membrane, and plasma membrane. (a) A single
optical slice at a magnification of 97x. (b) A single optical slice at a
magnification of 148x. No signal can be seen in the vacuoles or nucleolus
in these images.































Upio 0u
5-13a 97























Usio 0u
5-13b 148







































51aUpio (u


Figure 5-14. Series of four images taken from a larger z-series of upsilon/GFP. The
signal does not change in any of the images as the z-series progresses.
These images are at a magnification of 97x. No signal can be seen in the
vacuoles or nucleolus in these images.


5-14b Upsilon (u)
















5-14d Upsilon (u)

































Figure 5-15. Nu/GFP fusion construct demonstrates localization predominately in the
cytoplasm, plasma membrane, and nuclear membrane. (a) A single optical
slice at a magnification of 110x. (b) A single optical slice at a
magnification of 193x. No signal can be seen in the vacuoles or nucleolus
in any of these images.




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs