3...:

$1.2:

:1
l. . . ».
V7.9:

1, .vun

drul... D
.5." J:
. l :.,IH-I..1A.
E t V

 

 

 

 

 

 

 

 

 

 

 

r.. ...

 

 

THESIS

WNWIHIIHIHWlUlWill”I‘HIIIHHIHHI

\ 31293 01050 8806

LIBRARY

Michigan State
University

         

 

 

 

 

This is to certify that the

dissertation entitled

The Sesbania rostrata Early Nodulin
Gene SrEnodZ As A Marker For Cytokinin
Signal Transduction

presented by

David L. Silver

has been accepted towards fulfillment
of the requirements for

PhD degree in Genet 1‘ cs .

 

 

ajor professor

bate 1996

 

MS U is an Affirmative Action/Equal Opportunity Institution 0-12771

PLACE ll RETURN BOXto monthl- ehockout from your record.
TO AVOID FINES Mum on or baton date duo.

DATE DUE DATE DUE DATE DUE

       

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

THE SESBANIA ROSTRATA EARLY NODULIN GENE SRENODZ AS A
MARKER FOR CYTOKININ SIGNAL TRANSDUCTION

By

David L. Silver

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Program in Genetics

1996

ABSTRACT

THE SESBANIA ROSTRA TA EARLY NODULIN GENE SRENODZ AS A MARKER
FOR CYTOKININ SIGNAL TRANSDUCI'ION

By
David L. Silver

The Sesbania rostrata early nodulin gene SrEnodZ encodes a proline-rich protein
which is expressed tissue-Specifically in the nodule. Additionally, the SrEnodZ mRNA
accumulates in roots in response to cytokinin application. This accumulation occurs in the
absence of infection by the microsymbiont Azorhizobium caulinodans. Nuclear run-on
assays using isolated root nuclei indicated that SrEnodZ mRN A accumulation in response
to cytokinin application occurs postuanscriptionally. Analysis of nuclear RNA revealed
that this cytokinin enhancement occurs primarily in the cytoplasm and not in the nucleus.
Application of the translational inhibitor, cycloheximide, was found to cause its rapid
decay. It was also found that okadaic acid and staurosporine, inhibitors of protein
phosphatases and kinases, respectively, inhibited cytokinin enhancement of SrEnodZ
mRN A accumulation. Arabidopsis thaliana was used to study the mechanism of cytokinin-
mediated SrEnodZ mRNA accumulation. It was demonstrated that a chimeric SrEnodZ 5’-
gus-SrEnodZ 3’construct can be specifically induced by cytokinin and expressed in
Arabidopsis roots in the vascular tissue and emerging lateral roots, which closely resembles
the pattern seen in the legume Lotus japonicus. In addition, expression was found to be
localized to the shoot apical meristem, newly expanding leaves, and trichomes of
Arabidopsis. The observed mode of regulation was shown to be dependent on the
SrEnodZ 3’ region. A cross between Arabidopsis plants harboring the SrEnodZ 5’-gus-
SrEnodZ 3’ construct and the cytokinin-resistant mutant cyrl (Deikman and Ulrich, 1995)
yielded F2 progeny in which GUS activity could not be induced upon cytokinin treatment.
This data provides genetic evidence that the SrEnodZ 5’-gus-SrEnod2 3’construct is

regulated by a conserved cytokinin signal transduction pathway. The underlying
hypothesis is that conserved regulatory protein(s) is interacting with the SrEnodZ 3’ region
which regulates root/apical shoot meristem-specific expression and regulation by cytokinin.
To test this idea genetically, the SrEnodZ 3’ region was overexpressed under the control of
the CaMV 358 promoter in Arabidopsis, and the resulting transgenic plants were crossed to
plants harboring the SrEnodZ 5’-gus-SrEnod2 3’construct. The assumption was that if the
SrEnodZ 3’region can titrate out an important regulatory factor(s), then expression of the
gas reporter construct should be downregulated. Five out of six F2 plants analyzed from
this cross did not show cytokinin-enhanced GUS expression, thus supporting the existence
of a factor(s) which interact with the SrEnodZ 3’ region in Arabidopsis. In addition, a

genetic screen for the isolation of the gene(s) which may encode this factor(s) is presented.

ToIrma

iv

ACKNOWLEDGMENTS

I would like to thank my advisor Frans J. de Bruijn for supporting me in my studies as
well as creating an environment in the lab which encouraged me to think creatively. He
constantly challenged my ideas and helped me to think critically about my research. I like
to thank my committee members Kenneth Keegstra, Hans Kende, Michael Thomashaw,
and Natasha Raikhel for there many helpful criticisms and discussions. I would especially
like to thank Hans for his critical review of my manuscripts, as well as teaching me one of
Anton Lang’s golden rules, which I will never forget. I would like to thank Krzyzstof
Szczyglowski for not only teaching me a great deal of molecular biology, but teaching me
that humor is perhaps the greatest achievement of the human spirit. I thank Susan Fujimoto
for her friendship during these years. I will not only miss her company, but I will miss her
chicken ter'iyaki! I thank Rujin Chen for many helpful discussions. Thanks to Philipp
Kapranov, whom I worked closely with on a number of projects. I’m happy that he
learned the difference between “being in somebody’ s pants” and “being in somebody’s
shoes.” I thank all members past and present in the dc Bruijn lab for their helpful
suggestions and for their friendship. Above all, I thank Irma Velez for her love and

support. She gave meaning to many years of madness.

TABLE OF CONTENTS

LIST OF TABLES .................................................................................. ix
LIST OF FIGURES ................................................................................. x
CHAPTER 1
Introduction ........................................................................................... 1
The pleiotropic effects of cytokinins on plant development ............................ 2
A role of cytokinin in flowering and
flower development ................................................................ 7
Apical dominance ................................................................... 8
Leaf senescence ..................................................................... 9
The Agrobacterium paradigm ............................................................... 9
Altering cytokinin levels in transgenic plants ........................................... l3
Altering cytokinin sensitivity .................................................... 16
Cytokinins in stress responses .................................................. 18
The effects of ipt gene expression
on flower development ........................................................... 19
Cytokinin-altered mutants ................................................................. 20
Molecular responses to cytokinin ........................................................ 23
A role for cytokinin in nodule development? ........................................... 27
References .................................................................................. 33
CHAPTER2
Posttranscriptional regulation of the Sesbania rostrata early
nodulin gene SrEnod2 by cytokinin .............................................................. 45
Abstract ..................................................................................... 46
Introduction ................................................................................. 47
Methods ........ ' ............................................................................. 50
Results
Transcription of the SrEnodZ gene and accumulation
of its mRNA in response to cytokinin .......................................... 53

SrEnodZ mRNA accumulation 11 response to cytokinin

occurs primarily in the cytoplasm ............................................... 56
SrEnodZ mRN A accumulation appears to be

a long lived process .............................................................. 59
Cycloheximide inhibits SrEnodZ mRN A accumulation in
response to cytokinin ............................................................. 59

vi

Both cellular protein phosphatases and protein kinases may be

required for the accumulation of SrEnodZ mRNA by cytokinin ............. 68
Discussion ................................................................................... 72
Acknowledgments .......................................................................... 77
References ................................................................................... 78
CHAPTER 3
Regulation of the Lotus japonicus LjEnodZ gene by ethylene ................................. 82
Abstract ...................................................................................... 8 3
Introduction ................................................................................. 84
Methods ..................................................................................... 86
Results ....................................................................................... 88
Water treatment induces LjEnodZ mRNA accumulation
in L. japonicus roots via ethylene production .................................. 88
Ethylene effects on S. rostrata SrEnodZ gene
expression .......................................................................... 93
Discussion ................................................................................. 102
References ................................................................................. 104
CHAPTER 4
The SrEnodZ gene is controlled by a conserved cytokinin
signal tranduction pathway ....................................................................... 106
Abstract .................................................................................... 107
Introduction ............................................................................... 108
Methods .................................................................................... 1 1 1
Results ..................................................................................... l 15
Tissue specificity of the SrEnodZ gene in Lotus japonicus:
Requirement of the SrEnodZ 3’ region ........................................ 1 15
Expression of the SrEnodZ 3’ GUS chimeric gene in
Arabidopsis roots: Tissue specificity and cytokinin induction ............. 121
Expression of the SrEnodZ 3’ GUS construct in Arabidopsis
is enhanced specifically by cytokinin, and inhibited by ethylene ........... 127
The SrEnodZ 3’ UTR specifies cytokinin inducibility and tissue-
specificity ......................................................................... 128
Expression of the SrEnodZ 3’ GUS construct in the cytokinin-
resistant mutant cyrI and auxin-resistant mutant axr2 ....................... 131
Discussion ................................................................................. 134
Acknowledgments ...................................................................... 143

vii

References ................................................................................. 144

CHAPTER 5
Genetic evidence for the existence of SrEnodZ 3’-interacting factors ....................... 149
Abstract .................................................................................... 150
Introduction ............................................................................... 151
Methods ................................................................................... 1 5 3
Results ..................................................................................... 156
Overexpression of the SrEnodZ 3’ downstream region in
Arabidopsis causes a novel developmental phenotype ...................... 156
Genetic evidence of the existence of SrEnodZ 3’-
interacting factors ................................................................ 168
Discussion ................................................................................. 180
Aknowledgments ......................................................................... 183
References ................................................................................. 184
CHAPTER 6
Future perspectives ................................................................................ 186
References ................................................................................. 189

viii

LIST OF TABLES

Table 1.1. Cytokinin-producing bacteria .................................................... 1 1

Table 1.2. Examples of cytokinin-regulated genes ..................................... 24

ix

Figure 2.1.

Figure 2.2.

Figure 2.3.

Figure 2.4.

Figure 2.5.

Figure 2.6.

Figure 3.1.

Figure 3.2.

Figure 3.3.

Figure 3.4.

Figure 3.5.

Figure 3.6.
Figure 4.1.
Figure 4.2.

Figure 4.3.

LIST OF FIGURES

Comparison of nuclear run-on transcription with
mRNA accumulation .......................................................... 55

Abundance of SrEnodZ mRN A derived from total
RNA and mRNA ............................................................... 5 8

Fluctuation in SrEnodZ mRN A accumulation during
a 36-h period ................................................................... 61

Effect of cycloheximide on SrEnodZ mRNA
accumulation ................................................................... 64

Decay of SrEnodZ mRNA in the presence of

cycloheximide .................................................................. 67
Effect of okadaic acid and staurosporine on

SrEnodZ mRNA accumulation ............................................... 70
LjEnodZ mRN A levels in roots treated with

cytokinin ........................................................................ 9O
LjEnodZ mRN A levels in roots treated with H20, AVG and

C o C l 2 ............................................................................ 9 2
Ethylene levels in S. rostrata, and L. japonicus roots

treated with AVG ............................................................... 95
LjEnodZ mRN A levels in roots from seedlings treated

with ethylene .................................................................... 97
SrEnodZ mRNA levels in roots of seedlings treated

with H20, BAP, ethylene and CoCl2 ........................................ 99
SrEnodZ mRNA levels in roots treated with AVG ........................ 101
Structure of the chimeric SrEnodZ-gus reporter gene .................... 1 18

Histochemical localization of GUS activity in transgenic
L. japom'cus and Arabidopsis plants plants harboring the
SrEnodZ 3’ and nos 3’ constructs ........................................... 120

Histochemical staining of GUS activity in Arabidopsis seedlings
harboring the SrEnodZ 3’ and nos 3’ constructs treated
with cytokinin .................................................................. 124

Figure 4.4.

Figure 4.5.

Figure 4.6.

Figure 5.1.
Figure 5.2.
Figure 5.3.

Figure 5.4.
Figure 5.5.
Figure 5.6.
Figure 5.7.

Figure 5.8.

Figure 5.9.

Figure 5.10.

Figure 6.1.

GUS activity of Arabidopsis plants harboring the
SrEnodZ 3 construct after hormone treatments ........................... 126

GUS activity of Arabidopsis plants harboring the
SrEnodZ 3’UTR construct after cytokinin treatment ..................... 130

Regulation of the SrEnodZ 3’ construct in the cytokinin-resistant
mutant cyrI , and the dominant auxin-resistant

mutant (1er ................................................................... 133
In vivo titration scheme ...................................................... 158
Hypothetical mechanism for in vivo titration .............................. 160

Phenotypes of seedlings harboring the 358 SrEnodZ 3’
constructs ...................................................................... 163

Scanning Electron Microscopy of 10—day-old plants harboring

the 358 SrEnodZ 3’ construct exhibiting Siamese and Flag

phenotypes .................................................................... 165
Diagram of constructs used as controls .................................... 167

Northern blot analysis of Arabidopsis plants harboring
the 358 SrEnodZ 3’ construct ............................................... 170

Accumulation of the SrEnodZ 3’UTR mRNA ............................ 172
GUS expression in F, progeny from crosses between

plants harboring both of the SrEnodZ 3’

and 358 SrEnodZ 3’ constructs ............................................. 175

GUS expression in F2 progeny from crosses between
plants harboring both of the SrEnodZ 3’

and 35S SrEnodZ 3’ constructs ............................................. 177
Diagram illustrating normal Arabidopsis apical

meristem development ....................................................... 179
Genetic selection scheme .................................................... 19]

xi

Chapter 1

INTRODUCTION

The plant hormone cytokinin comprises a group of growth substances which are
derivatives of adenine first isolated by Miller et al. (1955) from autoclaved herring sperm
DNA and later by Letham (1963) from plants. The aim of this chapter is to present
information gathered from key studies indicating the roles of cytokinins in plant
development and gene expression, with the goal of presenting a perspective on cytokinin
action. However, this chapter will not cover the metabolism and biochemistry of
cytokinins, as this information can be found in a recent review (Brzobohaty et al.,l994).
Here, I will focus primarily on two aspects of cytokinin action in plants, the first being the
effects of cytokinins on plant development. This section will focus on studies involving
the application of cytokinins and other hormones to plants, which may share common
signal transduction pathways with cytokinins. Second, I will discuss the Agrobacterium
system and transgenic expression studies using Agrobacterium cytokinin biosynthetic
genes. Third, information will be presented on plant cytokinin response mutants. Fourth,
the molecular action of cytokinins will be reviewed. Lastly, nodule. development and the

roles of cytokinin, as well as other hormones, in nodule organogenesis will be discussed.
The pleiotropic effects of cytokinins on plant development

One mode of studying cytokinin action has been through the application of
cytokinin to intact plants or isolated tissues. Cytokinins were first identified by their ability
to stimulate cell division in tobacco pith cells (Miller et al., 1955; Miller et al., 1956). In
these studies, it was shown by Miller and co—workers that auxin stimulated cell elongation,
and that upon the addition of kinetin, a synthetic cytokinin, the pith cells began to divide.
This work was the first to demonstrate the involvement of more than one phytohormone in
inducing a biological response. This is now recognized as a general feature of plant
hormone action. After the original work, demonstrating the influence of cytokinin on cell

division, Skoog and co-workers set out to test the effects of cytokinins on plant

development. It was discovered that cytokinins influence the formation of flowers and
fruits, activities of enzymes (metabolism), and the appearance of chlorOplasts. Cytokinins
also delay the onset of leaf senescence, as well as play a role in the resistance to adverse
environmental stresses. Even today the mechanisms by which cytokinins influence these
processes are not well understood.

For the past 40 years, the sole method available to studying cytokinin action has
been the application of cytokinin to excised tissues, cells, and intact plants. Although the
results obtained from such direct application experiments are often correlative and
circumstantial, these types of experiments have built the framework for current work on
cytokinin action and have led to the elucidation of the general characteristics of cytokinin
action in plants. The result of the application of cytokinin depends largely on the
concentration and type of cytokinin used and the kind of plant tissue it is applied to. In
general, effects observed include the release of axillary buds from apical dominance, the
accumulation of anthocyanins, the inhibition of root and hypocotyl growth, the greening of
etiolated leaves, and a delay in senescence.

The effects of cytokinin application on plant development is not always directly due
to cytokinin, but may be through the effects of ethylene and light. It is known that
cytokinins stimulate ethylene production (Fuchs and Lieberman, 1968; Radin and Loomis,
1969). Bertell and Eliasson (1992) demonstrated in pea roots that cytokinin application
inhibited root elongation, the formation of lateral roots, and stimulated swelling of the root
tips. These effects were obtained at a benzylarninopurine (BAP) concentration as low as
0.01' M. They further demonstrated that BAP caused up to a four-fold increase in
ethylene levels in roots. The application of cobalt ions to inhibit ethylene production
counteracted both the inhibition of elongation and the swelling at the root tip caused by
BAP. In addition, it was shown that BAP treatment increased the levels of IAA per root tip
approximately two—fold, whereas treatment with the ethylene precursor 1-

aminocyclopropane-l-carboxylic acid (ACC) caused a 50% reduction in IAA levels. This

study indicates that cytokinins can influence growth processes in roots via multiple
pathways, including through ethylene and auxin. Cytokinin-induced radial expansion of
hypocotyls has also been shown to be cytokinin-mediated (Corriveau and Krul, 1986).
Cytokinins produce effects in dark-grown Arabidopsis seedlings which are similar to the
effects caused by ethylene, known as the “triple response.” The “triple response” is
characterized by the inhibition of hypocotyl growth, the curling of the apical hook, and the
expansion of the hypocotyl base (Crocker et al., 1913). In addition, ethylene inhibits
primary root elongation (Guzman and Ecker, 1990). This “triple response” has been
exploited in the isolation of ethylene resistant mutants (Guzman and Ecker, 1990), and
mutants which produce a constitutive triple response (Kieber et al., 1993). The constitutive
mutants are divided into two groups, based on whether or not the phenotype is repressed
by inhibitors of ethylene biosynthesis or action. The ctr mutants are not repressed by
inhibitors, whereas the ethylene overproducer (eta) mutants are repressed (Guzman and
Ecker, 1990). Cary et a1. (1995) tested genetically, by the use of mutants in ethylene
responses and action, the hypothesis put forth by Lieberman (1979) that cytokinin action is
coupled to ethylene action in seedlings. They found that the inhibitory effects of BAP on
root and hypocotyl elongation were partially blocked by the action of ethylene inhibitors or
in the ethylene—resistant mutations einl -I and ein2-1 (Guzman and Ecker, 1990).
Furthermore, the finding that cytokinin and ethylene responses are coupled was reinforced
by the demonstration that the cytokinin-resistant mutant ckrI (Su and Howell, 1992) is
allelic to ein2 (Cary et al., 1995).

Cytokinin and light interact in processes such as anthocyarrin accumulation
(Kasemir and Mohr, 1982), betacyanin synthesis (Koehler, 1972), hypocotyl elongation
(Cohen et al., 1991), and chloroplast development (Feierabend and de Boer, 1978). From
these studies, it appears that cytokinin can mimic some of the effects produced by light in
photomorphogenesis, but it remained unclear whether cytokinin action was dependent on

light or vice versa, or whether they act independently. Tong et al. (1983) showed using

mustard plants that the effects of cytokinin and light are additive with respect to increases in
cotyledon size, carotenoid contents, levels of glyceraldehyde-B-phosphate dehydrogenase,
and anthocyanin formation. These effects were observed regardless of the order of
treatment with cytokinin and light, or a simultaneous treatment with both. More recent
work has provided genetic evidence for the independence of 1i ght- and cytokinin-mediated
action on photomorphogenesis (Chory et al., 1994; Su and Howell, 1995). A class of
Arabidopsis mutants has been identified which shows many characteristics of light-grown
plants when grown in complete darkness. These mutants have been designated det (de-
etiolated) because of the de—etiolated phenotype in the dark, as compared with wild-type
seedlings (Chory et al., 1989, 1991b; Cabrera et al., 1993). detI and det2 mutants grown
in the light are small and have reduced apical dominance and fertility as compared with
wild-type plants (Chory and Peto, 1990; Chory et al., 1991). This indicates that the gene
products of the detI and det2 genes play a role in light-grown as well as dark-grown
plants. Recently, the detZ gene has been cloned and shown to have significant homology
with mammalian steroid Sa-reductases, which may function in the brassinolide
biosynthetic pathway (Li et al., 1996). In support of this idea, the application of
brassinolide to dark-grown det2 mutants partially suppressed the mutant phenotype (Li et
al., 1996). Cytokinins applied to wild-type dark-grown seedlings resulted in a phenocopy
of the detI mutant, which includes inhibition of hypocotyl elongation, promotion of
cotyledon expansion and leaf development. In addition, thylakoid-containing plastids are
formed in the cytokinin treated seedlings in much the same fashion as those formed in detI
mutants (Chory et al., 1989). The light-regulated genes cab, chs, and rch are also seven-
to eight-fold more active in the cytokinin-treated dark-grown seedlings, as compared to
untreated seedlings. This enhancement of the expression of light-regulated genes was also
found in dark grown detI seedlings (Chory etal., 1994). Interestingly, the cytokinin
levels in wild-type and detI seedlings (dark- or light-grown) were found to be the same.

However, in a detached leaf experiment to measure senescence, detI and detZ detached

leaves had a significant delay in senescence, as compared to wild—type leaves. In addition, .
dot] and det2 root and leaf explants in tissue culture continuously produced callus while
wild-type explants formed roots under the hormone conditions used, suggesting that detI
and det2 have a different requirement for cytokinins to initiate a normal developmental
response. This work indicates that cytokinin can partially overcome the requirement of
light to induce leaf and chloroplast development, as well as the expression of light-induced
genes, and that a simple additive relationship of light and cytokinin may not be true, at least
for the process of de-etiolation in Arabidopsis, but may take place in other
photomorphogenetic processes. Ultimately, the action of light and cytokinin appears to be
connected by some signal transduction pathways.

Cytokinin and light both inhibit hypocotyl elongation in a Ca2’—dependent manner
(Cohen et al., 1991), and the effects of cytokinin are mediated primarily through the action
of ethylene (Cary et al., 1995). It was recently shown by Su and Howell (1995) that the
effects of ethylene and cytokinin and light on the inhibition of hypocotyl elongation are '
independent and additive. This was demonstrated by the use of the Arabidopsis hypocotyl
elongation mutants (hy) (Koomneef et al., 1980; Liscum and Hangarter, 1993) which
comprise a group of mutants insensitive to light. A number of genes represented by these
mutants have been cloned (Parks and Quail, 1991; Somers et al., 1991; Reed et al., 1993;
Ahmad and Cashmore, 1993; Koomneef et al., 1980; Chory et al., 1989; Chory, 1992;
Parks and Quail, 1993). The inhibition of hypocotyl elongation by the application of
cytokinin was similar in the hy mutants as in wild-type plants, indicating that cytokinin acts
independently of light. In contrast, the ckrI/einZ mutant, which exhibits a normal response
to light, did not respond to cytokinin by inhibition of hypocotyl elongation, indicating that

the action of cytokinin in this process is coupled to ethylene action.

A role of cytokinin in flowering and flower development

The identity of the floral stimulus, or “florigen” is unknown. However, many
factors are known which influence the transition to flowering, and among them are
carbohydrate, light, and cytokinin. The mustard plant Sinapis alba has been used as a
model to study the transition to flowering (Bemier et al., 1977). It has been demonstrated
that a long day treatment will stimulate S. alba to flower, but a single low dose of cytokinin
to the apical meristem will evoke a partial flowering phenotype (Bemier et al., 1977;
Havelange et al., 1986), consisting of an increase in the mitotic index of meristem cells,
halving of the size of DNA replication units, and the splitting of vacuoles (Bemier et al.,
1977; Havelange et al., 1986; Houssa et al., 1990). In addition to cytokinin, sucrose may
play a signaling role in the evocation to flower. During the exposure of S. alba plants to a
long-day light treatment, sucrose accumulates very early in the apical meristem of induced
plants (Bodson and Outlaw, 1985). This increase in sucrose precedes mitotic activation.
The mobilization of sucrose stores to the apical meristem may be related to the export of
cytokinin from the roots, as suggested by the work of Bernier et al. (1993) who showed
that the removal of phloem by girdling at 8 h after the start of the inductive long day
treatment inhibited the transition to flowering, but surprisingly a treatment of cytokinin to
the apical meristem at 16 h after girdling reversed this inhibition. This result indicates that
the mobilization of cytokinin from root to shoot may play a role in the induction of
flowering. It was also demonstrated that a transient increase in cytokinin levels in roots
occurs 1 h after the long day treatment, and that elevated levels of cytokinin in mature
leaves could be detected 16 h after induction with a long—day treatment (Bemier et al.,
1981). In an experiment designed to abolish the postulated export of cytokinins from root
to shoot, plants were grown in 100% relative humidity to prevent transpiration, which is

believed to be the force of cytokinin movement in plants. This treatment completely

abolished the induction to flower (Bemier et al., 1993), although a 100% relative humidity
treatment may have many pleiotropic affects.

The application of cytokinin to the developing inflorescence of Arabidopsis has
been shown to result in increases in floral organ number, formation of abnormal floral
organs and production of secondary floral buds in the axils of sepals (V englat and
Sawhney, 1996). These abnormalities resemble the Arabidopsis floral mutants cle, ap2,
ap3 and apI (Leyser and Fumer, 1992; Okamuro et al., 1993). Although cytokinin was
applied at a high local concentration for a relatively long time period, this work suggests
that cytokinins play a role in normal floral organ development, perhaps by affecting the
regulation of floral organ identity genes. Transgenic plant studies will be presented later

which further support this idea.

Apical dominance

One of the known effects of cytokinin action in plants, other than effects on cell
division, is the role cytokinins seem to play in regulating apical dominance. The
application of cytokinin to axillary buds stimulates bud growth in many plant species
including apple, Cuscuta, Macadamia, oats, peas and soybeans (Cline, 1991). This raises
the question whether there is a correlation between cytokinin levels in buds and their ability
to grow and develop. Sossuountzov et al. (1988) have shown by the use of
irmnunolabelling of cytokinins that in the Craigella sideshootless tomato mutant (Cls),
which does not have axillary bud growth, the highest levels of cytokinins were found in the
apical bud closest to the meristem and decreased basipetally in the normal isogenic parental
line. The Cls mutant has been shown to have strikingly lower levels of cytokinin in the
terminal apical bud and in all subapical buds, whereas the root apical meristem had equal
levels of cytokinin as compared to the parental line. This work should be interpreted with

caution, however, as only the levels of one type of active and two types of inactive

cytokinins were measured. In a separate study using the aquatic fern Marsilea drummondii
A. Br., it was shown that the apical meristem contained the highest levels of IAA and
cytokinins. Upon decapitation of M. drummondii A. Br. plants, the subapical bud will be
the most rapidly growing bud (Pilate et al., 1989). In non-decapitated plants, the highest
levels of iPA, a precursor of zeatin, are found in this bud (Pilate etal., 1989). This
suggests that the subapical bud is in a “standby state” for release from apical dominance,

and that this process depends on the local cytokinin levels.

Leaf senescence

Cytokinins have also been implicated in the control of leaf senescence (Richmond
and Lang, 1957; Nooden and Leopold, 1978). Little is understood of the hormonal
involvement in sequential leaf senescence, which is the senescence from older, lower
leaves toward the younger leaves near the apex. Singh et al. (1992) have shown that the
levels of cytokinin bases and cytokinin ribosides are lower in older than upper younger
leaves in tobacco. Moreover, application of cytokinin to leaves was found to be effective in
retarding senescence, independent of metabolite mobilization in leaves (Singh et al., 1992).
It has also been observed that the application of nitrogenous compounds to tobacco leaves
greatly increases cytokinin levels, and retards leaf senescence (Singh et al., 1992). These
studies lend support to the idea that endogenous cytokinin levels are involved in the control
of sequential leaf senescence. Recent work using transgenic plants indicates that this is

indeed the case (see below).

The Agrobacterium paradigm

The morphological and physiological effects of cytokinin application have been the

basis for understanding cytokinin action in plants. However, in order to better understand

10

cytokinin action in plants, the need arose to understand the effects of changing endogenous
levels of cytokinins on plant processes. The genes responsible for plant cytokinin
biosynthesis have not been isolated, even though biosynthetic enzyme activities have been
described (Chen and Melitz, 1979; Chen and Leisner, 1984). The discovery that
Agrobacterium tumefaciens and A. rhizogenes produce cytokinins has proven immensely
valuable for the in viva manipulation of hormone levels in plants. A. tumefaciens is the
causative agent of crown gall tumor disease, which is the formation of a hyperplasia,
although in some plants shooty tumors are formed (Morris, 1995). It was first shown by
Braun (1958) that cell division factors are responsible for the formation of hyperplasias
caused by A. tumefaciens. Work by Willmitzer et a1. (1983) gave the first indication that
A. tumefaciens contained a gene responsible for hormone production which contribute to
alter hormone levels in crown galls. It was subsequently shown that A. tumefaciens
contains genes for auxin and cytokinin biosynthesis on a large plasmid, termed the Ti
plasmid (Akiyoshi et al., 1983; Barry et al., 1984; Klee et al., 1984; Schroder et al., 1984;
Kemper et al., 1985; Yamada, et al., 1985). Agrobacterz’um is able to transfer these genes
into the plant nucleus. The mechanism of Agrobacterium—mediated transformation will not
be discussed here as this information can be found elsewhere in more comprehensive
reviews (Lessl and Lanka, 1994; Zupan and Zambryski, 1995; Tinland and Hohn, 1995).
It can be seen from Table 1.1 that both phytopathogens and symbionts produce
cytokinins. The phytopathogens can be divided into two fundamentally distinct groups
based on their mode of forming hyperplasias. One group, comprised of A. tumefaciens
and A. rhizogenes (although rhizogenes does not from hyperplasias, but proliferation of
malformed roots on stems), transform dicotyledonous plants (De Cleene, 1988) with a
region of the Ti-plasmid DNA, the T-DNA, (Zaenen et al., 1974), harboring the genes
necessary for auxin and cytokinin biosynthesis, or genes which enhance the sensitivity of
the plant cell to hormones (Spena et al., 1987; Estruch et al., 1991a). Therefore, after T-

DNA transfer the presence of the bacteria is no longer required for the

11

and...” 6.2.33.5 u..:o..:o< .032: .8555: .::E .26. 8.25 E :_ mam. Om $.82 :5: 22.5.: «:3 2:5...

.25— .._: .0 5:59:52
33—
...ou._.o..:8.::.p. 1:: 20.2

as. .953 ..er. e2:...

3:. r... .o 8:35
mm:— .»..02 2.: 2.32m
3a. ...: .o 5:33;:
“we. .._: .o 8:o_.-._oo:o::m
L5:5.

2.2 ..o__:... .5: .::>o..:.m
:3. .._: .o $.52
.52 ...: .u EELS:—
23. .._e .e .4832
5.2 ...: 3 8.5m
:5. .._e .o 28...?

3.3. 5:99: 1:: ...5m

moo—5.5.“:—

efia. ....
:55.
.e. 5.2.5. SE; .N .N..

E33. x: .N

:33. .d. .Nci .N
:53.
e. .N
:55. .e. .Nss. .N
2:033:79 5.2.5—
....mefem p.53. .NEEEN
:55. .... .No
:. .Nze .N NESSE.
N
mazes... .Nos... Sea. .N
N

$35. ..= .NEo- .N

22.3030

25:5 .m:_>: 8...
12:633.: o.c._..m.§_._.

32:68:: 82.38....

m::m 2:30.83
82:5: 229:?
3:63: 52:5»...
3:50: 229:»:
3.35: 332...?
8.25: 522...»...
8:8... :58: 8.3.3.
.828 2.25:
:3 3.06::
m=:w 3::
8:92: .8. DE:

23...... =:w :32:

momdfiagoma—w—p

.7. 2.5.5
...:....:tce.:... ......=£:.::<
£3533... ..e..o.5e.e~<

2:. .2331... u
2. 53.5.2. 52.3.5

SEEE
$.30. :EEeNEQ
am 2533...:
2.35....t5... ..=..=.e~.:s.:~<
...=.......e.€... 5:33..5.9:...
2.39% c......oe.o£5.E
:Semb...» 2555553.;
5:333:53. 2:5..53E:
..c..e=.e..5. 2:52.353...
3.33.3.2 ..=......:.o.5=..u<

aziefiefiz. .==.....:....5=.&<

SEED—mm _

5.28m mics—55-.....3.2.6 ..._ 92:...

12

maintenance of the grown gall or root proliferation. The second group is comprised of
bacteria which do not introduce DNA into the plant cell nucleus, but secrete hormones and
therefore need to be present in close association with the plant for the maintenance of the
gall tissue. These bacteria include Erwinia herbicola, Pseudomonas savastanoi and
Rhodococcus fascians.

Cytokinin biosynthesis by A. tumefaciens is controlled by a single gene, called tmr.
These genes encode proteins catalyzing the rate-limiting step in cytokinin biosynthesis,
namely the transfer of an dimethylallylpyrophosphate onto the purine ring of adenine. This
gene, also known as the isopentyl transferase gene (ipt), is regulated by a plant promoter
and has plant termination and polyadenylation sequences. Therefore, the transformed plant
is capable of synthesizing cytokinins independent of bacterial control. The ipt gene has
provided a way to test in vivo the effects of altering cytokinin levels in plants, either at the
whole-plant level or in individual organs, tissues and cells.

A. rhizogenes is the etiological agent of the hairy-root disease (Riker et al., 1930).
This disease develops as a result of expression of several oncogenes located on the T-DNA
of Ri plasmids (Chilton et al., 1982). Unlike A. tumefaciens, A. rhizogenes does not
transform the plant cell by transferring auxin and cytokinin biosynthetic genes, but
introduces so called rol genes (named rol for root loci) (White et al., 1985). The rol A, B,
and C genes have been shown to be necessary for the induction of hairy roots in tobacco
plants (Jouanin et al., 1987; Schmulling et al., 1988). It has been demonstrated in vitro
that the R01 C protein hydrolyzes cytokinin glucosides (Estruch et al., 1991a). Although
recently Faiss et al. ( 1996) failed to detect changes in the endogenous pool of different
cytokinin glucosides in planta in rat C-expressing tobacco plants. They propose that other
low molecular weight signals, such as oligosaccharins, may be the in viva substrate for the
R01 C protein, rather than cytokinin glucosides. The expression of the rat B gene in plants
has been implicated in auxin action (Cardarelli et al., 1987; Filippini et al., 1994). The

substrate for the R01 B protein has never been conclusively determined. However,

13

recently it has been demonstrated in vitro that recombinant Rol B protein has tyrosine
phosphatase activity and is localized in the plasma membrane of transformed plants
(Filippini et al., 1996). These data indicate a role of kinase/phosphatase cascades in auxin
signal transduction. Therefore, the rol genes may interfere with plant hormone metabolism
rather than de novo synthesis, as observed in A. tumefaciens transformed plants. Further

rol C expression studies in transgenic plants will be presented below.

Altering cytokinin levels in transgenic plants

The effect of applied cytokinin on plant processes may not be the same as that of
endogenous cytokinin. Problems of cytokinin uptake, transport, metabolism, and
tissue/cell-specific hormone concentrations and sensitivities may produce numerous
secondary effects. To avoid these problems, many groups have utilized the cytokinin
biosynthesis gene from A. tumefaciens to manipulate endogenous cytokinin levels in
transgenic plants, to test the effects of altered hormone levels on plant physiology,
development, and gene expression. Overall, most of these studies in transgenic plants have
confirmed the previous results from experiments using applied cytokinins.

In the first of these studies, the ipt gene was placed under the control of the CaMV
35S promoter to achieve high levels of endogenous cytokinins (Ooms et al., 1983; Binns et
al., 1987b; Smigocki and Owens, 1988; Smigocki and Owens, 1989). All transformants
displayed extreme phenotypes correlated with cytokinin action, such as profuse shoot
development and little or no root development. Aside from these classic phenotypes, the
transgenic tobacco plants constitutively expressing ipt also showed auxin-autonomous
growth. Tobacco cell lines in which the ipt gene was overexpressed were also found to be
auxin autonomous, while non-transformed tobacco cells grown in the presence of cytokinin
are auxin-requiring (Binns et al., 1987). This indicates that applied cytokinin does not

completely mimic the effects of the endogenous production of cytokinin. Smigocki and

l4

Owens (1989) measured an increase of cytokinin levels of up to 300-fold in transgenic
tobacco plants harboring a CaMV 35S-ipt fusion with no significant increase in IAA levels.
In addition, a 24- to over a 2,000—fold increase in cytokinin-to-auxin ratios was observed,
which may explain the morphogenic changes and the auxin-autonomous growth of tissues
in vitro.

To better control the levels of cytokinins in planta, which would allow the
regeneration of transgenic plants with roots, experiments were conducted using inducible
promoters. Medford et al. (1989) placed the ipt gene under the control of the maize heat
shock promoter hsp70 and generated transgenic tobacco and Arabidopsis plants harboring
this construct. They found that under noninducing conditions, the levels of zeatin riboside,
and zeatin riboside 5’-monophosphate increased 3 and 7 times, respectively. After heat
induction, the levels of zeatin, zeatin riboside, and zeatin riboside 5’-monophosphate were
found to increase 52-, 23-, and 2—fold respectively. The small increase in cytokinin levels.
in plants under non-inducing conditions already caused dramatic affects on plant
development, such as reduction in stature, release of axillary buds, generation of smaller
stem and leaf areas, reduced xylem production, and the generation of a reduced root system
with short and thicker roots with more root hairs. Heat treatment of the transgenic plants
did not lead to further alterations in plant development, despite large increases in cytokinin
levels. These results indicate that there is a threshold level at which cytokinins are
perceived and act. Similar results were obtained by Schmulling et al. (1989) and Smigocki
(1991) using the Drosophila melanogaster hsp70 promoter. Moreover, Ainley et al. (1993)
performed similar experiments using the soybean heat shock promoter, which was shown
to be tightly regulated and mediated a high level of cytokinin production at elevated
temperatures (Ainley and Key, 1990). In contrast to experiments using the maize and
Drosophila heat shock promoters, transgenic plants harboring the ipt gene under the control
of the soybean heat shock promoter showed no alterations in phenotype under non-

inducing temperatures, but exhibited phenotypic alterations only after heat treatment. Heat

15

treatment had a strong effect only on developing leaves from transgenic plants, but not on
fully developed leaves, indicating that certain deve10ping tissues may be susceptible to
cytokinin, whereas mature tissues may not be.

One additional phenotypic alteration, which has not previously been reported as a
cytokinin effect, is the production of leaf chlorosis. The chlorosis effect is in direct
contrast to the observed increase in greening of leaves upon cytokinin treatment (reviewed
by Thomas and Stoddart, 1980). The authors explain that chlorosis may be due to the
inhibition of vascular tissue development in cytokinin-overproducing transgenic plants, as
observed by Medford et al. (1989). This inhibition may lead to a limitation in the transport
of aSsimilates to expanding leaves and could severely limit the development of leaves, as
well as cause chlorosis. Smart et al. (1991), using the soybean heat shock promoter fused
to the ipt gene in tobacco, showed that transgenic plants at non-inducing temperatures did
show marked phenotypic differences as compared to untransformed plants. These plants
were shorter in stature, had an increase in side shoot production, and remained green for
longer time periods than untransformed plants. Differences were more pronounced after
several heat shock treatments. These results again suggest a role of cytokinins in the delay
of leaf senescence. The developmental and morphological alterations seen in these
transgenic plants complicate the interpretatiOn of a direct role of cytokinins in leaf
senescence. In an elegant approach to address this problem Gan and Amasino (1995)
placed the ipt gene under the control of a senescence-specific promoter, SAG12, in
transgenic Arabidopsis plants. The idea was to autoregulate the production of cytokinin in
leaves during senescence. The SAGlZ (Senescence—Associated Gene) was isolated as a
gene specifically expressed during leaf senescence (Lohman et al., 1994). Transgenic
plants did not exhibit any developmental alterations. As wild-type, non-transgenic plants
aged, leaf senescence progressed sequentially from the bottom to the top leaves. In
contrast, identically aged transgenic plants showed no sign of leaf senescence. In a

detached leaf assay, transgenic plant leaves showed no sign of senescence after more than

16

40 days, whereas wild-type plant leaves began to senescence after 10 days. The
autoregulatory nature of pSAGlZ-ipt expression was analyzed by placing the gas reporter
gene under the control of pSAG12 promoter and monitoring gus expression during '
senescence in plants with or without the pSAGlZ-ipt gene construct. The levels of GUS
activity in pSAGlZ-gus plants, increased only in leaves undergoing senescence. In the
pSAGlZ-gus/pSAGlZ-ipt plants GUS levels were over 1,000—fold lower. Thus, the
pSAGlZ promoter is strongly autoregulated. This work confirms that cytokinins delay leaf
senescence and that cytokinins are able to negatively affect the senescence program through

the regulation of senescence-specific genes.

Altering cytokinin sensitivity

The overexpression of the rat C gene in transgenic tobacco and potato plants shows
similar effects as the overexpression of the ipt gene (Schmulling et al., 1988; Fladung,
1990). Phenotypic changes observed include the reduction of apical dominance, generation
of male sterile flowers, reduced leaf pigment content (seen only in the ipt study by Ainley
and Key, 1990) and, surprisingly, dwarfism. The reduced leaf pigment content has been
used as a phenotypic marker to determine whether the R01 C protein acts in a cell-
autonomous fashion. Transgenic tobacco plants, harboring the rat C gene under the
control of the CaMV 358 promoter interrupted by an Ac element, gave rise to sections of
yellowing cells on leaves upon Ac excision. This type of clonal analysis indicates that the
rol C product acts in a cell-autonomous fashion (Spena et al., 1989).

It was shown that the overexpression of rol C in transgenic potato, but not in
tobacco plants, led to a 4-fold increase in the free cytokinin content, which may explain
some of the phenotypic alterations seen in potato plants (Schmulling et al., 1993).
Interestingly, alterations in the levels of other hormones were found, such as an up to 50%

reduction in ABA content in leaves and a 100% increase in roots. The dwarfism phenotype

17

could be correlated with a 28-60% reduction in GAI in the apical shoots of both transgenic
tobacco and potato plants, since this phenotype could be suppressed by the application of
GA). to the apical shoot of rol C transgenic plants (Schmulling et al., 1993). It should also
be noted that the rol C related phenotype could not be phenocopied by the application of
any hormone. The sensitivity of rat C—expressing transgenic tobacco seedlings to
externally supplied hormones in a germination assay was also different as compared to
hormone sensitivity in wild-type seedlings. rol C-overexpressing seedlings were found to
have an increased resistance to auxins and ABA, and a higher sensitivity to cytokinins, the
ethylene precursor ACC, as well as the auxin transport inhibitor TIBA. In addition,
Schmulling et al. (1993) crossed rol C overexpressing plants with ipt overexpressing
transgenic plants and found that the phenotype exhibited by rol C overexpression is
dominant to the phenotypes obtained by overexpression of the ipt gene, in that the plants
were now able to root, and were dwarfed. The rolC/ipt plants as well as the ipt plants did
have normal chlorophyll levels, indicating that at least this phenotype is dominant to the
phenotype of rol C overexpressing plants having reduced chlorophyll levels. Overall, this
work suggests that, although Rol C acts by releasing active cytokinins from inactive forms
in vitro (Estruch et al., 1991a), the regulation of free hormone levels in plants is much
more complex, as shown by the pleiotropic morphological alterations in rolC plants, as
well as the tissue-specific changes in various hormone levels and sensitivities.

In an interesting variation to the regulated expression studies of the ipt gene in
plants, Hewelt et al. (1994) used a promoterless ipt gene to utilize endogenous plant
promoters to regulate cytokinin production in a developmental and tissue-specific manner.
A wide variety of phenotypic alterations were observed. Although, it cannot be determined
in this study which tissues do not respond to changes in cytokinin levels since a reporter
gene was not included in this work to follow tissue-specific expression in transformed, but

phenotypically normal plants. It was seen that not all ipt lines showed dosage effects of the

18

trangene on plant phenotype, indicating that gene dosage effects are dependent on the tissue

and/or cell type as well as its developmental state.
Cytokinins in stress responses

Cytokinins may play a role in plant responses to certain physiological stresses.
Application of cytokinin to certain plants can mimic salt-induced responses (Thomas et al.,
1992; Thomas and Bohnert, 1993), namely the accumulation of proline and an osmotin-like
protein, although endogenous cytokinin levels tend to decrease under salt stress (Kupier et
al., 1990; Thomas et al., 1992). This effect was reconfirmed in planta by expressing the
ipt gene in tobacco under the control of the light-inducible rch-3A promoter from pea.
Under high light conditions, the transgenic plants accumulated appreciable amounts of
proline and osmotin, although the plants also showed dramatic morphological alterations
(Thomas et al., 1995). Another stress which may involve modulations of endogenous
cytokinin levels is the plant response to pathogen attack. N on-rooting shoot lines of
tobacco overexpressing the ipt gene were found to exhibit an increase in the expression of
defense-related mRNAs (Memelink et al., 1987). The proteins encoded by these genes are
coordinately induced by wounding and pathogenic attack (Chen and Vamer, 1985; Ward et
al., 1991). In a more recent study, the ipt gene was placed under the control of the
proteinase inhibitor II promoter and introduced into tobacco. Upon attack by the insect
larvae Manduca sexta , transgenic PI-II-ipt plants were 70% less susceptible to
consumption by the larva than were control plants (Smigocki et al., 1993). It is not clear
what the mode of action of the ipt gene in resistance. The authors propose that an increase
in endogenous cytokinin levels may cause an increase in the production of secondary
metabolites with insecticidal properties (Binns et al., 1987a; Orr and Lynn, 1992;
Teutonico et al, 1991 ).

19

The effects of ipt gene expression on flower development

One of the most intriguing phenotypes resulting from endogenous alteration of
cytokinin levles is the production of viviparous leaves and epiphyllous floral bud
development (Estruch et al., 1991b; Estruch et al., 1993). In these studies, the ipt gene
was placed under the control of the CaMV 35$ promoter, interrupted by the maize
transposable element Ac. Upon somatic transposon excision, the 358 promoter activated
the expression of the ipt gene. Surprisingly, in one transgenic line, the vascular
parenchyma of leaves re-differentiated into vegetative buds at the leaf midrib. This change
in cell fate, related to the alteration in cytokinin levels, is the first example of developmental
switches caused by alterations of endogenous cytokinin levels. In a more striking example
of this type of developmental change, Estruch and co-workers (1993) showed that the same
epiphyllous bud-producing tobacco plants were capable of producing both normal and
abnormal epiphyllous floral buds. Epiphyllous floral buds developed only after the normal
apical vegetative buds underwent the transition to floral bud development. Epiphyllous
buds which were produced prior to this developmental switch remained vegetative. The
abnormal epiphyllous floral buds had fused organs and were characterized by a local
activation of the ipt gene, resulting in a 100- to 1000-fold increase in zeatin riboside
equivalents. In contrast, the normal epiphyllous floral buds had cytokinin levels equivalent
to those found in normal apical floral buds. In addition, there was a decrease in the mRN A
steady-state levels of the tobacco homologues of the homeotic genes DEFA (Sommer et al.,
1990), GLO (Schwarz-Sommer et al., 1992)and PLENA (Bradley et al., 1993) of
Antirrhinum majus. This study lends support to the physiological evidence showing that
the application of cytokinin provokes floral development (see above), as well as alters
normal floral development. It is not known how cytokinins act to induce epiphyllous

vegetative and floral bud development. However, what is clear is that the development of

20

either type of bud relies on the same signals that the normal apical buds respond to for
vegetative and floral bud development. This suggests that cytokinins are not acting only to

reactivate the cell cycle, but are triggering a complex ectopic developmental program.

Cytokinin-altered mutants

Two approaches have been taken toward the isolation of mutants altered in
cytokinin production or action. One has been to screen for mutants altered in sensitivity to
cytokinins, while the other has been based on a screen for plants with characteristic
changes in morphology and development that have been correlated with cytokinin action.

A classic result of cytokinin application to seedlings is the inhibition of root growth,
increase in root hair production, and root tip swelling. Using these criteria, Blonstein et al.
(1991) isolated a cytokinin-resistant mutant of N. plumbaginifolia. This mutant was
originally named ckrI and is characterized by a reduction in root development, cytokinin
resistance during seedling development, and wiltiness of the shoot. The wiltiness was
found to be caused by a defect in stomatal closure. In a study it was shown that the ckrI
mutant is deficient in abscisic acid biosynthesis (Rousselin et al., 1992), and that the
mutation affects the conversion of ABA-aldehyde to ABA, the final step in the ABA
biosynthetic pathway (Parry et al., 1991). Thus, the mutant was renamed AbaI.

In a similar study by Su and Howell (1992), Arabidopsis mutants were isolated
based on resistance to low levels of cytokinin, in order to avoid the isolation of mutants in
general stress responses. Five independent mutants were isolated which comprise a single
complementation group, named ckrI . In a later study it was shown that the ckrI mutant is
allelic to the ethylene insensitive mutant, einZ (Cary et al., 1995). The isolation of ckrI
indicates that the pathways for ethylene and cytokinin responses overlap. These studies
underscore the problem associated with isolating cytokinin-response mutants, namely that

cytokinin action is very pleiotropic and may be mediated through other factors. In a more

21

recent effort to isolate cytokinin-resistant mutants, Deikman and Ulrich (1995) found the
Arabidopsis cyrI mutant (cytokinin-resistant 1). The cyrI mutant is characterized by a 10-
fold reduction in sensitivity to benzyladenine in a root-elongation assay, but not to ACC,
IAA or ABA. Rather, cyrI has an increased sensitivity to ABA. The phenotype of cyrI
includes abbreviated shoot development, limited leaf production, reduction in cotyledon
and leaf expansion, reduced chlorophyll accumulation, failure to accumulate anthocyanins
in response to cytokinin treatment (a typical response to cytokinins; Pecket and Bassim,
1974; Ozeki and Komamine, 1981), and the formation of a single infertile flower. All of
these traits are consistent with a mutation in cytokinin perception rather than biosynthesis.
In support of the idea that cyrI is a true mutant in cytokinin perception, it was demonstrated
that expression of an SrEnodZ-GUS chimeric construct, which is cytokinin-enhanced in
wild-type Arabidopsis, was found not to be cytokinin-enhanced in the cyrI mutant (see
Chapter 4). The complex phenotype exhibited by cyrI appears to be due to a mutation in a
single gene, indicating that a single gene required for normal cytokinin responses can have
diverse effects on plant growth and development.

An Arabidopsis mutant deficient in adenine phophoribosyltransferase (APRT)
activity (apt), was originally isolated in a screen for purine metabolism mutants (Moffatt et
al., 1991). The apt mutant has approximately 1% of the APRT activity found in wild-type
Arabidopsis plants. This mutant has normal vegetative morphology, grows more slowly
than wild type, and is male sterile. In both in vivo and in vitro tests, the apt mutant was
found to be unable to convert benzyladenine (BA) to benzyladenine-monophosphate
(BAMP), indicating that APRT is the main enzyme which converts BA to its nucleotide
form in young Arabidopsis plants. It is not known whether the slow growth and the male
sterility are caused by the alteration in cytokinin metabolism in the apt plants.

A group of tobacco mutants originally isolated on the basis of their resistance to
cytokinin define three complementation groups, zeal, zeaZ, and zea3 (Jullien et al., 1992).

The zea3 mutant has a particularly complex phenotype in that it is highly sensitive to a high

22

carbon/nitrogen ratio, as well as to cytokinin, but only during germination at the jointed-
cotyledon developmental stage (Faure et al., 1994). Under low nitrate conditions, zea3
accumulates three times more sucrose and 5 times more amino acids than wild-type
seedlings. The zea3 mutant is able to germinate under high cytokinin concentrations,
whereas the wild-type is completely inhibited. In addition, cytokinin causes the
development of leaf hypertrophies in zea3 mutant plants. Faure et al. (1994) have
proposed that in zea3 export or translocation of photoassimilates is perturbed causing the
cotyledons to act as a sink instead of a source organ, thereby competing with the apical
meristem for import of sucrose and amino acids. Also, the authors suggest that the
cytokinin-induced hypertrophies may be related to the large import of photoassimilates into
cotyledons, resulting in an increase of turgor pressure. This cytokinin-induced
hypertrophy has been recently shown to be specific to cytokinins in the zea] group of
mutants, and proposed as a bioassay for cytokinins (N ogue et al., 1995).

Using an entirely different approach, Chaudhury et al. (1993) screened for
Arabidopsis mutants with novel developmental phenotypes. Their objective was to test
developmentally altered plants for changes in the levels of plant growth regulators. The
result was the isolation of the amp] mutant (altered meristem program). The amp] mutant
is characterized by the frequent occurrence of polycoty (20% of total plants), bushiness of
shoots, increased life span, floral abnormalities, such as siliques made of three or four
carpels, and semi-sterility. Other abnormalities include a four-fold increase in rosette
leaves formed before flowering as compared to wild-type, a significant decrease in the time
to flower, and de-etiolation in the dark. The effect of a lack of phytochrome on amp]
mutation was investigated in the double mutant hyZ amp]. hy2 mutants lack the
phytochrome chromophore and are deficient in the production of all phytochromes (Parks
and Quail, 1991). The hy2 mutant has a longer hypocotyl and exhibits an increase in apical
dominance over wild-type plants. The amp] hy2 double mutant displays an intermediate

phenotype with respect to hypocotyl length in the dark and apical dominance. This

23

suggests that the AMP] product is required for the hyZ phenotype. Another interesting
characteristic of the amp] mutant is that it displays a 6—fold increase in cytokinin levels.
Therefore, amp] represents the only known cytokinin overproducing mutant of
Arabidopsis. The observed 6-fold increase in cytokinin levels is well within the range
measured in transgenic plants expressing the ipt gene, which gave rise to similar
phenotypic variations (Ainley et al., 1993; Hewelt et al., 1994). Therefore, the pleiotrOpic
phenotypes of amp] can most likely be explained by an elevated level of cytokinin,
although some of the phenotypes of the amp] mutant have not been reported before in ipt
expressing transgenic plants. The authors further propose that AMPl may be a regulator of
cytokinin biosynthesis or metabolism. It is known that several genes whose expression is
enhanced by application of cytokinin are, in fact, constitutively expressed in the amp]
mutant, and that the amp] mutation appears to affect primarily the shoot and not the root
(E. Dennis, personal communication). The determination of the identity of AMP]
promises to yield exciting information, and map-based cloning efforts of the corresponding

locus are in progress (J .-D. Faure, personal communication).
Molecular responses to cytokinin

Cytokinin is capable of modulating the expression of a wide variety of genes, as shown in
Table 1.2. It can be seen in Table 1.2 that cytokinin can act either at a transcriptional or
posttranscriptional level, depending on the gene in question. Unfortunately, little is known
about the molecular mechanisms of cytokinin action. Great progress has been made in
understanding ethylene signal transduction through the isolation of Arabidopsis mutants

and the identification of the corresponding genes. One of these genes ET R1 possesses all

24

Table 1.2. Examples of Cytokinin-Regulated Genes“

 

 

 

 

 

 

 

 

 

Gene name 1 level of gene 1 mode of 1 time of 1 other influencing reference
9 expression 9 regulation I response 5 factors 1.
wheat protein 1 mRNAT 1 ND 1‘ 24 h 1 light, nutrients 1 Sano and
kinase. ka4 g 1 E 1 Youssefian. 1994
L. gibba rch 1 mRNAT 1 P 1 24 h 1 light 1 Flores and Tobin.
1 = 1 l 1988
L. gibba cab 1 mRNAT P 1 24 h 1 light 1 Flores and Tobin.
- 1 : 1988
tobacco defense- mRNAT ND 1 3 weeks 1 ND Memelink. et al..
related genes 1987
maize PEPC; C4ppcl mRNAT TIP 2 h light, nitrogen Suzuki et al.,
1994
barley nr mRNAT T 15 min light. nitrogen. ABA Lu et al.. 1990
soybean pollen mRNAT ND 4 h auxin Crowell, 1994
allergen ciml
Arabidopsis ebs mRNAT T 3 b light Deikman and
Hammer. 1995
Arabidopsis pall. mRNAT P 10 d light Deikman and
chi Hammer. 1995
Arabidopsis dfr mRNAT T 10 d light Deikman and
Hammer. 1995
Arabidopsis cyclin D mRNAT ND 4 h sucrose Soni et al., 1995
homolog 53
Arabidopsis cdca GUST T 72 h auxin, wounding Hemerly et al..
1 1993
19 unidentified mRNAT ND 1 4 h auxin Crowell ct al..
soybean cDNAs 1 i g 1990
tobacco multiple mRNAT ND 1 < 10 h auxin 1 Dorninov et al.,
stimulus response 1 1 1992
gene 131.8216 3 i 5 3
s. rostrata Enod2 mRNAT P ; 2 h 3 none found ‘ Dehio and de
g i Bruijn. 1992
Alfalfa EnodlZ. 1 mRNAT 1 ND 6 h nod factor Hirsch and Fang.
Enod40 1 i i i 1994
rice fl-gluennase 1 mRNAT 1 ND 1 > 24 h 1 ethylene, wounding Simmons et al..
Gnsl 1 1 salicyclic acid.fungal 1992
1 1 3 1 elicitors
tobacco msr gene. ; GUST 1 T g 18 h g pathogen attack. auxin. , Gough et al..
str 246C 1 1 ; salicyclic acid 1995
Spirodela polyrrhiza 1 mRNAl. 1 ND 3 24—72 h 3 ABA g Chaloupkova and
L basic peroxidase g 2 3 3 Smart. 1994
pumpkin bpt ; mRNAT g T ; 1.5 h ; ND 1 Anderson ct al..
1 g ‘ é 1996
tobacco class I B- 1 GUSl ’ T 1 4 d auxin, ethylene. 1 Vogeli-Lange ct
1.3-glucanase glb 1 9 pathogen attack al.. 1994
cucumber catalase. ; mRNAl ND 1-4 h ; ND g Toyama et al..
HMGR. Lectin i = i 1995

 

 

25

of the characteristics of an ethylene receptor (Schaller and Bleecker, 1995). In addition,
there are multiple genes which have been genetically determined to code for proteins which
lie downstream of ETRI in the ethylene response pathway (Kieber et al., 1993; Hua et al.,
1995). In analogy, it has generally been assumed that cytokinin also interacts with a
specific receptor protein. There are reports of cytokinin-binding proteins, although there is
little evidence that any act a receptor (Mitsue et al., 1993; Mitsue and Sugiura, 1993;
Palme, 1993). The question remains as to why cytokinin receptor mutants have not been
isolated to date. One possibility, other than lethality caused by such mutations, is that there
are no specific receptor proteins, but rather that cytokinin interacts with multiple receptors,
which feed into multiple signal transduction pathways. The identification of the genes
involved in cytokinin-resistance (Deikman and Ulrich, 1995; Nogue et al., 1995) and
overproduction mutants (Chaudhury et al., 1993) promises to shed light on this topic.

A confounding problem when using a specific gene as a probe or reporter to study
cytokinin action is that most genes which are regulated by cytokinin are co-regulated by
other factors, such as light, nutrients, and other hormones. Many of these cytokinin-
regulated genes are regulated by auxin and light, which is supported by data showing that
cytokinin, auxin and light interact to affect plant development (Miller et al., 1955; Miller et
al., 1956; Tong et al., 1983; Su and Howell, 1995). The complex interaction of cytokinin
and other factors in gene expression is best exemplified by the nitrate reductase gene.
Treatment of etiolated barley leaves with light, nitrate, and cytokinin greatly enhances the
accumulation of the nr mRNA, primarily through transcriptional activation (Lu et al.,
1990), and ABA negatively regulates this enhancement (Lu et al., 1992). The requirement
for both nitrate and light for nr mRNA accumulation has not been found in the case of the
nr gene from Agrostema githago, in which nitrate reductase activity, as well as mRNA
accumulation, are enhanced solely upon cytokinin treatment, and seems to occur primarily
at the posttranscriptional level (Kende et al., 1974; White, 1996). In addition, the
accumulation of the A. githago nr mRN A upon cytokinin treatment appeals to be inhibited

26

by ethylene (White, 1996). The cytokinin-specific enhancement of the A. githago nr
mRNA accumulation occurs only in embryos and not at any other stage of plant
development. It is interesting to note that although the expression of both nr genes from
barley and A. githago is enhanced by cytokinin, the mechanisms by which this occurs
appear to be quite different, as are the tissues in which the nr genes are expressed. More
recently, it has been shown that the genes of the anthocyanin biosynthesis pathway also are
cytokinin-induced in a light dependent manner (Deikman and Hammer, 1995), which now
explains old observations of anthocyanin accumulation as a typical effect of cytokinin
application. One gene which may be an exception to this complex interaction of cytokinin
and other factors in regulating gene expression is the SrEnodZ gene from Sesbania rostrata.
The expression of the SrEnod2 gene appears to be enhanced solely by cytokinin (Dehio and
dc Bruijn, 1993). The SrEnodZ gene may be a candidate marker gene for studying
cytokinin signal transduction (Silver et al., 1996; and Chapter 4).

One of the best known affects of cytokinin is its effect on cell division. It has been
shown that the Arabidopsis cdc2a gene is transcriptionally induced by cytokinin after long
exposures to the hormone (Hemerly et al., 1993), although its expression is not directly
coupled to cell division, but always precedes it. Hemerly et al. (1993) propose that
multiple signals may be involved in the triggering of cells to divide or to be competent for
division, and that cytokinin, acting through the cdc2a protein, may be one of these signals.
The cyclin gene 63 from Arabidopsis, which is homologous to the human D-type cyclins,
is expressed at the 61/8 transition and is proposed to play a role in regulating this transition
in a similar way as the CLN 1 and CLN2 cyclins in yeast (Richardson etal., 1989;
Wittenberg et al., 1990; Soni et al., 1995). Interestingly, the expression of the 63 gene is
rapidly induced by cytokinin after 4 h of treatment. In a separate study by John et al.
(1993) using tobacco pith cells, a tissue classically known to require cytokinin and auxin
for cell division, it was shown that cdc2 expression was induced by treatment with auxin

alone, but that the protein was nonfunctional unless cytokinin was also present in the

27

growth medium. They speculate that the cytokinin-induced component required for the
activation of cdc2 in G1 cells might be cyclin 83. The activation of the cell cycle is critical
for the deve10pment of new plant organs, such as lateral roots and nodules (see below),
and determining how cytokinin acts on the cell cycle will be of great importance to
understanding the mechanisms that regulate cell division during plant development.
Overall, these studies reveal that, in some instances, the effects of cytokinin on

plant development and physiology can now be partially explained at the molecular level.
A role for cytokinin in nodule development?

In the case of most plant organogenesis events, a role for cytokinin has been
postulated, although in several cases it has been very difficult to prove this hypothesis
directly. This also appears to be the case for the involvement of cytokinin in the
development of nitrogen-fixing nodules on legume plants. Nodule development has been
recently extensively reviewed (V errna, 1992; Mylona et al., 1995), and will only be
reviewed briefly here. The development of nitrogen fixing nodules involves the highly
specific interaction of rhizobia with the legume plant root, or stem as in the case of the
tropical legume Sesbam'a rostrata. At the onset of this interaction, rhizobia are induced by
plant phenolics to produce a specific chito-lipooligosaccharide molecule, known as the Nod
factor (Peters and Verma, 1990). The Nod factor, in turn, seems to interact with an as yet
unidentified receptor at the root epidermis, resulting in root hair deformation or curling.
This root hair curling is part of the uptake mechanism of the bacteria into the plant cell.
Following, is the production of an infection thread produced by the plant harboring the
bacteria, and the initiation of cortical cell division in the inner or outer cortex, depending on
the type of legume infected (determinate versus indeterminate nodules). Once the bacteria
make their way to the dividing cortical cells, via the infection threads, they are taken up into

large plant cells known as infected cells, in which they will differentiate into nitrogen-fixing

28

bacteroids. During this entire process, the expression of specific plant genes is being
induced. These plant genes are termed nodulins, since they are induced or their expression
is enhanced in nodules, although some are known to be expressed elsewhere in the plant as
well. The nodulins have been classified based on their time point of expression. Those
which are expressed early during nodule development are termed early nodulins, and those
which are expressed in fully developed nodules are called late nodulins. The early
nodulins, such as Enod5, 12, 40, and 2 are believed to play a role in nodule ontogeny,
although no exact function has yet been assigned to most of them. The late nodulins such
as glutamine synthetase, sucrose synthetase, and leghemoglobin all play a role in nodule
functioning. A growing number of other early and late nodulin genes are being isolated,
for example by the differential display of mRN As (Goorrnachtig et al., 1995; K.
Szczyglowski and F. J. de Bruijn, unpublished data), and these genes promise to yield
exciting novel information regarding nodule development and functioning.

Nodule development is primarily determined by a plant genetic program and not by
the presence of the infecting bacteria, as evidenced by the discovery of spontaneous
nodulating alfalfa plants (Truchet et al., 1989). In addition, alfalfa plants can be induced to
form nodule-like structures upon treatment with auxin transport inhibitors (Allen et al.,
1953; Hirsch et al., 1989), or purified Nod factor (T ruchet et al., 1991; Mergaert et al.,
1993; Stokkermans et al., 1994). It has been postulated that the ability of the rhizobial Nod
factor to induce the plant nodule ontogony program is related to hormone action, although
the evidence for the involvement of hormones in nodule development is circumstantial.
Thimann (1936) first proposed a role of auxin in nodule development, and postulated that
nodule development may be related to lateral root development Libbenga et al. (197 3),
using an in vitro approach, treated pea root cortical explants with auxin and found that cell
division took place in the pericycle, the location of lateral root initiation. However,
division of cortical cells, the location of nodule initiation, occurred upon the addition of

both auxin and cytokinin to the media. It has been shown that rhizobia secrete cytokinins

29

into the culture medium (Morris, 1986; Sturtevarlt and Taller, 1989; Taller and Sturtevarlt,
1991; Upadhyaya et al., 1991), but again the significance of bacterially produced cytokinin
on nodule development remains unclear. Since cytokinin biosynthesis genes have not yet
been identified in rhizobium to date, the effects of mutations in these genes are unknown.
Probably the most conclusive evidence for a role of cytokinins in nodule development
comes from the work by C00per and Long (1994). They expressed the Agrobacterium
cytokinin biosynthetic gene 12s in a Rhizobium meliloti strain carrying a mutation in the nod
structural genes, preventing the synthesis of the Nod factor. This t2; expressing
Rhizobium was capable of inducing nodule-like structures on alfalfa, supporting the idea
that localized cytokinin production may be involved in nodule development and may be able
to “substitute for” Nod factor action. This work does not prove that cytokinin secreted
from wild-type rhizobia is involved in triggering of nodulation. In fact, it seems more
likely that the source of cytokinin is plant derived, given the existence of spontaneously
nodulating alfalfa plants.

One of the first events in nodule development involves cortical cell divisions. It has
been shown that purified Nod factor is capable of eliciting cortical cell divisions (Spaink et
al., 1991; Truchet et al., 1991; Relic et al., 1993). This process of cortical cell division has
been extensively studied by Yang et al. (1994). It has been determined that cells
susceptible to Nod factor are arrested in the 60/61 stage of the cell cycle, and not the
G2/M stage as previously believed. Therefore, Nod factor-susceptible cortical cells are
arrested in the same stage as all other cortical cells. Those cortical cells which divide are
opposite protoxylem poles. Interestingly, a positive regulator of nodulation has been
isolated from the xylem and shown to be able to replace cytokinin in an in vitro pea cortical
cell assay (Libbenga et al., 1973). This factor, called stele factor, has been identified to be
uridine (H. Spaink, personal communication). As reviewed above, cytokinin also clearly
plays an important role in cortical cell divisions, as exemplified by the classical experiments

of Miller and Skoog (1957), as well as more recently by Yang et al. (1994). At the

3O

molecular level, it has been shown that the expression of the early nodulin genes Enod 12
and Enod40 are induced by both Nod factor and cytokinin in alfalfa, whereas the EnodZ
gene from Sesbania rostrata and an EnodZ-like gene of alfalfa are induced only by
cytokinin (Dehio and de Bruijn, 1992; Hirsch and Fang, 1994). It is not known whether,
in fact, cytokinin is the direct regulator of the expression of these genes during nodule
development. It has been proposed that Nod factor does not act directly on nodule
development but acts through altering the endogenous cytokinin/auxin ratio (Mylona et al.,
1995). Support for this idea comes from the expression of the rhizobial nodA and nodB
genes in transgenic tobacco plants. nodA and nodB encode proteins involved in the
production of Nod factor in Rhizobium (John et al., 1993; Rohrig et al., 1994). These
transgenic tobacco plants displayed phenotypes similar to plants with an imbalance in
hormone levels, such as epinastic leaves, and increased apical dominance (Schmidt et al.,
1993). Another line of evidence is that the non-nodulating alfalfa line MN 1008 can be
induced to form nodule-like structures expressing early nodulin genes upon treatment with
auxin transport inhibitors, independent of Nod factor application (Hirsch and Fang, 1994).
The corollary to this idea is that the Nod factor mimics the action of a plant N od-like factor
which, in turn, triggers the nodule developmental signal transduction pathway. Although
active plant N od—like factors have not been isolated, with the exception of Nod-like factor
molecules in plant secondary cell walls (Spaink et al., 1993), application of Nod factor to
non-legume plants has yielded a surprising result. For example, a mutant carrot cell line
arrested in development can be complimented by the application of Nod factor (De Jong et
al., 1993). In addition, Nod factor have been observed to cause suspension-cultured
tomato cells to trigger the alkalization of the culture medium (Staehelin et al., 1994).

It can be observed from the above discussion that the role of cytokinin in nodule
development is still poorly understood. There is considerable evidence for a role of auxin
in nodule development. This idea goes back to Thimann (1936) in which he proposed that

a relationship exists between lateral root development and nodule development. The

31

primary evidence for this comes from the nodule forming non-legume Paraspania.
Parasponia nodules, which are true nodules harboring Rhizobia, form from pericycle cells,
as do lateral roots, rather than from cortical cells (Marvel et al., 1987). In further support
of a role for auxin, an auxin-sensitive alfalfa line A2 (Borgre et al., 1990) was shown to
form significantly more nodules and undergo earlier nodule initiation than the genetically
related wild-type line R15 (Kondorosi et al., 1993). In addition, Alfalfa plants transgenic
for the rolB gene, expression of which is correlated with auxin-sensitivity (Shen et al.,
1988), produced considerably more nodules and in a shorter time than non-transformed
alfalfa plants (Kondorosi et al., 1993). The rolB expressing plants also had an increase in
root production, which correlates with the effects of rolB expression on root production in
tobacco (Schmulling, 1988). Interestingly, expression of Enod40 in tobacco, which is
induced by nod factor in both root pericycle cells and in dividing cortical cells, caused a
phenotype similar to a hormone affect. Expression of Enod40 in tobacco protoplasts
caused auxin-insensitive growth at concentrations which are inhibitory to non-transformed
protoplasts (Van de Sande et al., 1996). Mylona et al. (1995) proposes that Enod40
expression causes a change in the auxin/cytokinin ratio in cortical cells leading to mitotic
reactivation. Therefore, Nod factor may act through certain nodulin genes, such as
Enod40, to trigger nodule development and or regulation of development.

At the present, an effort is being made to understand the relationship between lateral
root development and nodule development, and the involvment of hormones in these
processes. K. Szczygwoski and F. J. de Bruijn (unpublished data) have isolated a Lotus
japonicus mutant bar] which forms a hyper-amount of root nodules when inocculated with
Rhizobia similar to the soybean supemodulating mutant nts (Carroll et al., 1985).
Surpisingly, in the absence of Rhizobr'a, harI forms a profuse amount of lateral roots. It is
proposed that had represents a gene which is a negative regulator of both nodule
initiation/development as well as lateral root initiation/development (K. Szczygwoski,

personal communication). The superroot mutant, sup], of Arabidopsis forms an excess of

32

lateral roots and contains increased levels auxin (Boerjan et al., 1995). The alfI mutant of
Arabidopsis is also characterized by a profuse production of lateral roots, and is believed to
be due to the overproduction of auxin (Celenza et al., 1995). It remains to be determined
whether the phenotypes of the had mutant is caused by an increase in auxin levels or
sensitivity, and does harI represent the legume equivalent of alfI or sup] . In a distinct
approach to these questions, another group is studying the Arabidopsis mutants nut] and 2
which fail to form lateral roots but instead form nodule-like structures (Cheng et al., 1995)
in the hope of elucidating the mechanisms underlying lateral root and nodule developmental
programs (R. Wilson, personal communication). The further characterization of nodulation
mutants, such as harI, as well as non-legume root mutants, will shed light on the
understanding of the involvement of hormones in nodulation and the relationship with
lateral root development. Towards this goal, it is proposed in this thesis that the SrEnodZ
gene can be utilized in Arabidopsis for the isolation of trans-acting factors involved in
regulating the cytokinin-enhancement of SrEnodZ gene expression. Any genes which code
for such trans-acting factors can be tested for their relevance in regulating nodulin genes

during nodule development.

33
REFERENCES

Ahmad M, Cashmere AR (1993) H Y4 gene of A. thaliana encodes a protein with
characteristics of a blue light photoreceptor. Nature 336: 162-165

Ainley WM, Key JL (1990) Development of a heat shock inducible expression cassette
for plants: characterization of parameters for its use in transient expression assays. Plant
Mol Biol 14: 949—967

Ainley WM, McNeil KJ, Hill JW, Lingle WL, Simpson RB, Brenner ML,
Nagao RT, Key JL (1993) Regulatable endogenous production of cytokinins up to
‘toxic’levels in transgenic plants and plant tissues. Plant Mol Biol 22: 13-23

Akiyoshi DE, Morris R0, Hinz R, Mischke BS, Kosuge T (1983) Cytokiin-
auxin balsnce in crown gall tumors is regulated by specific loci in the T-DNA. Proc Natl
Acad Sci USA 80: 407-411

Allen EK, Allen ON, Newman AS (1953) Pseudonodulation of leguminous plants
induced by 2-bromo-3,5-dichlorobenzoic acid. Amer J Bot 40: 429-435

Barry GF, Rogers SG, Fraley RT, Brand L (1984) Identification of a cloned
cytokinin biosynthetic gene. Proc Natl Acad Sci USA 81: 4776-4780

Bernier G, Havelange A, Houssa C, Petitjean A, Lejeune P (1993)
Physiological signals that induce flowering. Plant Cell 5: 1147-1155

Bernier G, Kinet J-M, Jacqmard A, Havelange A, Bodson M (1977) Cytokinin
as a possible component of the floral stimulus in Sinapis alba. Plant Physiol 60: 282-
285

Bernier G, Linet J-M, Sachs RM (1981) The physiology of flowering, Vol 11 (Boca
Raton: CRC Press)

Bertell G, Eliasson L (1992) Cytokinin effects on root growth and possible
interactions with ethylene and indole-3-acetic acid. Physiol Plant 84: 255-261

Binns AN, Chen RH, Wood HN, Lynn DG (1987a) Cell division promoting
activity of naturally occuring dehydrodiconiferyl glucosides: Do cell wall components
control cell division. Proc Natl Acad Sci USA 84: 980-984

Binns AN, Labriola J, Black RC (1987b) Initiation of auxin autonomy in Nicotiana
glutinosa cells by the cytokinin-biosynthesis gene from Agrobacterium tumefaciens.
Planta 171: 539-548

Blonstein AD, Parry AD, Horgan R, King PJ (1991) A cytokinin-resistant mutant
of Nicotiana plumbaginifolia is wilty. Planta 183: 244-250

Bodson M, Outlaw WH Jr (1985) Elevation in the sucrose content of the shoot apical
meristem of Sinapis alba at floral evocation. Plant Physiol 79: 420—424

Boerjan W, Cervera M-T, Delarue M, Beeckman T, Dewitte W, Bellini C,
Caboche M, Van Onckelen H, Van Montagu M, Inze D (1995) superroot, a

34

Recessive mutation in Arabid0psis, confers auxin overproduction. Plant Cell 7: 1405-
1419

Bogre L, Stefanov 1, Abraham M, Somogyi I, Dudits D (1990) Differences in
responses to 2,4-dichlorophenoxy acetic acid (2,4—D) treatment between embryogenic
and non-embryogenic lines of alfalfa. In HJJ Nijkarnp, LHW Van der Plas, J Van
Aartrijk, eds, Progress in Plant Cellular and Molecular Biology, Kluwer Academic
Publishers, Dordrecht pp 427-436

Bradley D, Carpenter R, Sommer H, Hartley N, Coen E (1993) Complementary
floral homeotic phenotypes result from opposite orientations of a transposon at the plena
locus of Antirrhinum. Cell 72: 85-95

Braun AC (1958) A physiological basis for the autonomous growth of the crown gall
tumor cell. Proc Natl Acad Sci USA 44: 344-349

Brzobohaty B, Moore I, Palme K (1994) Cytokinin metablolism: Implications for
regulation of plant growth and development. Plant Mol Biol 26: 1483-1497

Cabrera H, Peto C, Chory J (1993) A mutation in the Arabidopsis DET3 gene
uncouples photregulated leaf development from gene expression and chlorplast
biogenesis. Plant J 4: 671-682

Cardarelli M, Spano‘ L, Mariotti D, Mauro ML, Van Sluys MA, Costantino
P (1987) The role of auxin in hairy root induction. Mol Gen Genet 208: 457-463

Carroll BJ, McNeil DL, Gresshoff PM (1985) Isolation and properties of soybean
(Glycine max) mutants that nodulate in the presence of high nitrate concentrations. Proc
Natl Acad Sci USA 82: 4164-4166

Cary AJ, Wennuan L, Howell SH (1995) Cytokinin action is coupled to ethylene in
its effects on the inhibition of root and hypocotyl elongation in Arabidopsis thaliana
seedlings. Plant Physiol 107: 1075-1082

Celenza JL Jr, Grisafi PL, Fink GR (1995) A pathway for lateral root formation in
Arabidopsis thaliana. Genes Dev 9: 2131-2142

Chaudhury AM, Letham S, craig S, Dennis ES (1993) amp] - a mutant with high
cytokinin levels and altered embryonic pattern, faster vegetative growth, constitutive
photomorphogenesis and precocious flowering. Plant J 4: 907-916

Chen CM, Leisner SM (1984) Modification of cytokinins by cauliflower microsomal
enzymes. Plant Physiol 75: 442-446

Chen CM, Melitz DK (1979) Cytokinin biosynthesis in a cell-free system from
cytokiin-autotrophic tobacco tissue cultures. FEBS Lett 107: 15-20

Chen J, Varner JE (1985) An extracellular matrix protein in plants: Characterization of
a genomic clone for carrot extensin. EMBO J 4: 2145-2151

Cheng JC, Seeley KA, Sung ZR (1995) RMLl and RML2, Arabidopsis genes
required for cell proliferation at the root tip. Plant Physiol 107: 365-376

35

Chilton MD, Tepfer DA, Petit A, David C, Casse-Delbart F, Tempe J (1982)
Agrobacterium rhizogens inserts T-DN A into the genomes of the host plant root cells.
Nature 295: 432-434

Chory J (1992) A genetic model for light-regulated seedling deve10pment in Arabidopsis.
Development 115: 337-354

Chory J, Nagpal P, Peto CA (1991) Phenotypic and genetic analysis of detZ, a new
mutant that affects light-regulated seedling development in Arabidopsis. Plant Cell 3:
445-459

Chory J, Peto CA (1990) Mutations in the detI gene affect cell-type-specific expression
of light-regulated genes and chloroplast development in Arabidopsis. Proc Natl Acad Sci
USA 87: 8776-8780

Chory J, Peto C, Feinbaum R, Pratt L, Ausubel F (1989) Arabidopsis thaliana
mutant develops as a light-grown plant in the absence of light. Cell 58: 991-999

Chory J, Reinecke D, Sim S, Washhurn T, Brenner M (1994) A role for
cytokinins in de-etiolation in Arabidopsis det mutants have an altered response to
cytokinins. Plant Physiol 104: 339-347

Cline MJ (1991) Apical dominance. Bot Rev 57: 318-358

Cohen L, Gepstein S, Horwitz BA (1991) Similarity between cytokinin and blue
light inhibition of cucumber hypocotyl elongation. Plant Physiol 95: 77-81

Cooper JB, Long SR (1994) Morphogenetic rescue of Rhizobium meliloti nodulaiton
mutants by trans-zeatin secretion. Plant Cell 6: 215-225

Corriveau JL, Km] WR (1986) Inhibition of benzyladenine-induced radial expansion
in soybean hypocotyls by silver ion. J Plant Physiol 126: 297-300

Cracker W, Knight LI, Rose RC (1913) A delicate seedling test. Science 37: 380-
381

De Cleene M (1988) The susceptibility of plants to Agrobacterium: A discussion of the
role of phenolic compounds. FEMS Microbiol Rev 54: 1-8

De Jong AJ, Heidstra R, Spaink HP, Hartog MV, Meijer EA, Hendriks T,
Lo Schiavo F, Terzi M, Bisseling T, Van Kammen A, De Vries SC (1993)
Rhizobium lipooligosaccharides rescue a carrot somatic embryo mutant Plant Cell 5:
615-620

Dehio C, de Bruijn FJ (1992) The early nodulin gene SrEnodZ from Sesbania rostrata
is inducible by cytokinin. Plant J 2: 117-128

Deikman J, Hammer PE (1995) Induction of anthocyanin accumulation by cytokinin in
Arabidopsis thaliana. Plant Physiol 108: 47-57

Deikman J, Ulrich M (1995) A novel cytokinin-resistant mutant of Arabidopsis with
abbreviated shoot development. Planta 195: 440-449

36

Estruch JJ, Chriqui D, Grossmann K, Schell J, Spena A (1991a) The plant
oncogene rolC is responsible for the release of cytokinins from glucoside conjugates.
EMBO 10: 2889-2895

Estruch JJ, Granell A, Hansen G, Prinsen E, Redig P, Van Onckelen H,
Schwarz-Sommer Z, Sommer H, Spena A (1993) Flora] development and
expression of floral homeotic genes ae influenced by cytokinin. Plant J 4: 379-384

Estruch JJ, Prinsen E, Van Onckelen H, Schell J, Spena A (1991b) Viviparous
leaves produced by somatic activation of an inactive cytokinin-synthesizing gene.
Science 254: 1364-1367

Faiss M, Strnad M, Redig P, Dolezal K, Hanus J, Van Onckelen H,
Schmulling T (1996) Chemically induced expression of the rolC-encoded B-
glucosidase in transgenic tobacco plants and analysis of cytokinin metabolism: rolC does
not hydrolyze endogenous cytokinin glucosides in planta Plant J 10: 33-46

Faure J-D, Jullien M, Caboche M (1994) Zea3: A pleiotrOpic mutation affecting
cotyledon development, cytokinin resistance and carbon-nitrogen metablolism. Plant J
5: 481-491

Feierahend J, de Boer I (1978) Comparative analysis of the action of cytokinin and
light on the formation of RBPC and plastid biogenesis. Planta 142: 75-82

Filippini F, Rossi V, Marin O, Trovato M, Costantino P, Downey PM, Lo
Schiavo F, Terzi M (1996) A plant oncogene as a phosphatase. Nature 379: 499- -
500

Fladung M (1990) Transformation of diploid and tetraploid potato clones with the rolC

gene of Agrobacterium rhizogenes and characterization of transgenic plants. Plant
Breeding 104: 295-304

Fuchs Y, Lieberman M (1968) Effects of kinetin, IAA and gibberellin on ethylene
production and their interactions. Plant Physiol 43: 2029-2036

Gan S, Amasino RM (1995) Inhibition of leaf senescence by autoregulated production
of cytokinin. Science 270: 1986-1988

Goormachtig S, Valerio-Lepiniec M, Szczyglowski K, Van Montagu M,
Holsters M, de Bruijn FJ (1995) Use of differential display to identify novel
Sesbania rostrata genes enhanced by Azorhizobium caulinodans infection. Mol Plant-
Microbe Interact 6: 816—824

Guzman P, Ecker JR (1990) Exploiting the triple response of Arabidopsis to identify
ethylene-related mutants. Plant Cell 2: 513-524

Havelange A, Bodson M, Bernier G (1986) Partial floral evocation by exogenous
cytokinin in the long-day plant Sinapis alba. Physiol. Plant 67: 695-701

Hemerly AS, Ferreira P, de Almeida Engler J, Van Montagu M, Engler G,
Inze D (1993) cdc2a expression in Arabidopsis is linked with competence for cell
division. Plant Cell 5: 1711-1723

37

Hewelt A, Prinsen E, Schell J, Van Onckelen H, Schmulling T (1994)
Promoter tagging with a promoterless ipt gene leads to cytokinin-induced phenotypic

variability in transgenic tobacco plants: Implications of gene dosage effects. Plant J 6:
879-891 .

Hirsch AM, Bhuvaneswari TV, Torrey JG, Bisseling T (1989) Early nodulin
genes are induced in alfalfa root outgrowths elicited by auxin transport inhibitors. Proc
Natl Acad Sci USA 86: 1244-1248

Hirsch AM, Fang Y (1994) Plant hormones and nodulation: what’s the connection?
Plant Mol Biol 26: 5-9

Houssa C, Jacqmard A, Bernier G (1990) Activation of replication origins as a
possible target for cytokinins in shoot meristems of Sinapis. Planta 181: 324-326

Hua J, Chang C, Sun Q, Meyerowitz EM (1995) Ethylene insensitivity conferred
by Arabidopsis ERS gene. Science 269: 1712-1714

John PA, Rohrig H, Schmidt J, Wieneke U, Schell J (1993) Rhizobium NodB
protein involved in nodulation signal synthesis is a chitooligosaccharide deacetylase.
Proc Natl Acad Sci USA 90: 625-629

John PCL, Zhang K, Dong C, Diederich L, Wightman F (1993) p34cdc2 related
proteins in control of cell cycle progression, the switch between division and
differentiation in tissue development, and stimulation of division by auxin and
cytokinin. Aust J Plant Physiol 20: 503-526

Jouanin L, Vilaine F, Tourneur J, Tourneur C, Pautot V, Muller JF,
Caboche M (1987) Transfer of a 4.3-kb fragment of the TL-DNA of Agrobacterium
rhizogenes strain A4 confers the pRi transformed phenotype to regenerated tobacco
plants. Plant Sci 53: 53-63

Jullien M, Lesueur D, Laloue M, Caboche M (1992) Isolation and preliminary
characterization of cytokinin-resistant mutants of Nicotiana plumbaginifolia. In
Kamineck M, Mok DWS, Zazimalova E, eds, Proc Int Symp on Physiology and
Biochemistry of Cytokinins in Plants. The Hague: SPB Academic Publishing BV, The
Netherlands pp 157-162

Kasemir H, Mohr H (1982) Coaction of three factors controlling chlorophyll and
anthocyanin synthesis. Planta 156: 282-288

Kemper E, Waffenschmidt S, Weiler EW, Rausch T, Schroder J (1985) T-
DNA-encoded auxin formation in crown-gall cells. Planta 163: 257-262

Kende H, Fukuyama-Dilworth M, deZacks R (1974) On the control of nitrate
reductase by nitrate and benzyladenine in Agrostemma githago embryos. In Tamura S,
ed, Plant Growth Substances-1973. Tokyo, Hirokawa Pub] pp 675-682

Kieber JJ, Rothenberg M, Roman G, Feldmann KA, Ecker JR (1993) CTR] , a
negative regulator of the ethylene response pathway in Arabidopsis, encodes a member
of the raf family of protein kinases. Cell 72: 427-41

Klee H, Montoya A, Horodyski F, Lichtenstein C, Garfinkel D, Fuller S,
Flores C, Peschon J, Nester E, Gordon M (1984) Nucleotide sequence of the

38

Mrs genes of the pTiA6NC octopine Ti plasmid: 2 gene products involved in plant
tumorigenesis. Proc Natl Acad Sci USA 81: 1728-1730

Koehler K-H (1972) Photocontrol of betacyanin synthesis in Amaranthus caudatus
seedlings in the presence of kinetin. Phytochemistry l 1: 133-137

Kondorosi E, Schultze M, Savoure A, Hoffmann B, Dudits (1, Pierre M,
Allison L, Bauer P, Kiss GB, Kondorosi A (1993) Control of nodule induction
and plant cell growth by nod factors. In EW Nester, DPS Verma ,eds, Advances in
Molecular Genetics of Plant-Microbe Interactions, Kluwer Academic Publishers,
Dordrecht, The Netherlands pp 143-150

Koornneef M, Rolff E, Spruit CJP (1980) Genetic control of light-inhibited
hypocotyl elongation in Arabidopsis thaliana L Heynh. Z Pflanzenphysiol 100: 147-160

Kupier D, Schuit J, Kupier PJC (1990) Actual cytokinin concentrations in plant
tissue as an indicator for salt resistance in cereals. Plant Soil 123: 243-250

Less] M, Lanka E (1994) Common mechanisms in bacterial conjugation and Ti-
mediated T-DNA transfer to plant cells. Cell 77: 321-324

Letham DS (1963) Zeatin, a factor inducing cell division from Zea Mays. Life Sci 8: 569-
573

Leyser HMO, Fumer H (1992) Characterization of three shoot apical meristem mutants
of Arabidopsis thaliana. Development 16: 397-403

Li J, Nagpal P, Vitart V, McMorris TC, Chory J (1996) A role for
brassinosteroids in light-dependent development of Arabidopsis. Science 272: 398-401

Libbenga KR, van Iren F, Bogers RJ, Schraag-Lamers MF (1973) The role of
hormones and gradients in the initiation of cortex proliferation and nodule formation in
Pisum sativum L. Planta 114: 29-39

Lieberman M (1979) Biosynthesis and action of ethylene. Annu rev Plant Physiol 30:
533-591

Liscum E, Hangarter RP (1993) Genetic evidence that the red-absorbing from of
phytochromone B mondulates gravitropism in Arabidopsis thaliana. Plant Physiol 103:
15-19

Lehman KN, Gan S, John MC, Amasino RM (1994) Molecular analysis of natural
leaf senescence in Arabidopsis thaliana. Physiol Plant 92: 322-328

Lu J-I, Ertl JR, Chen C-M (1990) Cytokinin enhancement of the light induction of
nitrate reductase transcript levels in etiolated barley leaves. Plant Mol Biol 14: 585-594

Lu J -I, Ertle JR, Chen C-M (1992) Transcriptional regulation of nitrate reductase
mRNA levels by cytokinin-abscisic acid interations in etiolated barley leaves. Plant
Physiol 98: 1255-1260

Marvel DJ, Torrey JG, Ausubel FM (1987) Rhizobium symbiotic genes required for
nodulation of legume and nonlegume hosts. Proc Natl Acad Sci USA 84: 1319-1323

39

Medford JI, Horgan R, El-Sawi Z, Klee JH (1989) Alterations of endogenous
cytokinins in transgenic plants using a chimeric isopentenyltransferase gene. Plant Cell
1: 403-413

Memelink J, Hoge JHC, Schilperoort RA (1987) Cytokinin stress changes the
developmental regulation of several defence-related genes in tobacco. EMBO J 6: 3579-
3583

Mergaert P, van Montagu M, Prome’ J-C, Holsters M (1993) Three unusual
modifications, a D-arabinosyl, an N-methyl, and a carbomoyl group are present on Nod
factors of Azorhizobium caulinodans strain ORSS71. Proc Natl Acad Sci USA 90:

155 1-1555

Miller CO, Skoog F, Okomura FS, von Saltza MH, Strong FM (1956)
Isolation, structure and synthesis of kinetin, a substance promoting cell division. J Am
Chem Soc 78: 1345-1350

Miller CO, Skoog F, von Saltza MH, Strong FM (1955) Kinetin, a cell division
factor from deoxyribonucleic acid. J Am Chem Soc 77 : 1329-1334

Mitsui S, Sugiura M (1993) Purification and properties of a cytokinin-binding protein
from tobacco leaves. Plant Cell Physiol 34: 543-547

Mitsui S, Wakasugi T, Sugiura M (1993) A cDNA encoding the 57KDa subunit of a
cytokinin-bindin g protein complex from tobacco-the subunit has high homology to S-
adenosyl-homocysteine hydrolase. Plant Cell Physiol 34: 1089-1096

Moffatt B, Pethe C, Laloue M (1991) Metabolism of benzyladenine is impaired in a
mutant of Arabidopsis thaliana lacking adenine phosphoribosyltransferase activity. Plant
Physiol 95: 900-908

Morris R0 (1986) Genes specifying auxin and cytokinin biosynthesis in phytopathogens.
Annu Rev Plant Physiol 37: 509-538

Morris R0 (1995) Molecular aspects of hormone synthesis and action. Genes specifying
auxin and cytokinin biosynthesis in prokaryotes. In P] Davies, ed, Plant Hormones
Physiology, Biochemistry and Molecualar Biology. Kluwer Academic Publishers,
Dordrecht, The Netherlands, pp 318-339

Mylona P, Pawlowski K, Bisseling T (1995) Symbiotic nitrogen fixation. Plant
Cell 7: 869-885

Nogue F, Jullien M, Mornet R, Laloue M (1995) The response of a cytokinin
resistant mutant is highly specific and permits a new cytokinin bioassay. Plant Growth
Regul 17: 87-94 ‘

Nooden L, Leopold A (1978) Phytohormones and the endogenous regulation of
senescence and abscission. In Phytohormones and related compounds: A
Comprehensive Treatise, Vol 2, 329-369. Elsevier/North Holland, Amsterdam

Okamuro JK, den Boer BGW, Jofuku KD (1993) Regulation of Arabidopsis
flower development. Plant Cell 5: 1183-1193

40

Ooms G, Darp A, Roberts J (1983) From tumour to tuber: Tumour cell characteristics
and chromosome numbers of crown gall-derived tetraploid potato plants (Solanum
tuberosum cv. Maris Bard). Theor Appl Genet 66: 169-172

Orr JD, Lynn DG (1992) Biosynthesis of dehydrodiconiferyl alcohol glucosides:
Implications for the control of tobacco cell growth. Plant Physiol 98: 343-352

Ozeki Y, Komamine A (1981) Induction of anthocyanin synthesis in relation to
embryogenesis in a carrot suspension culture: Correlation of metabolic differentiation
with morphological differentiation. Physiol Plant 53: 570-577

Palme K (1993) From binding proteins to hormone receptors? J Plant Growth Regul 12:
17 1- 178

Parks BM, Quail PH (1991) Phytochrome-deficient by] and hy2 long hypocotyl
mutants of Arabidopsis are defective in phytochrome chromophore biosynthesis. Plant
Cell 3: 1177-1186

Parks BM, Quail PH (1993) Hy8 a new class of Arabidopsis long hypocotyl mutants
deficient in functional phytochrome A. Plant Cell 5: 39-48

Parry AD, Blonstein AD, Babiano M], King PJ, Horgan R (1991) Abscisic
acid metabolism in a wilty mutant of Nicotiana plumbaginifolia. Planta 183: 237-243

Pecket RC, Bassim TAH (1974) The effect of kinetin in relation to photocontrol of
anthocyanin biosynthesis in Brassica oleracea. Phytochemistry 13: 1395-1399

Peters NK, Verma DPS (1990) Phenolic compounds as regulators of gene expression
in plant-microbe interactions. Mol Plant-Microbe Interact 3: 4-8

Pilate G, Sossountzov L, Miginiac E (1989) Hormone levels and apical dominance
in the aquatic fern Marsilea drummondii A Br. Plant Physiol 90: 907-912

Radin JW, Loomis RS (1969) Ethylene and carbon dioxide in the growth and
development of cultired radish roots. Plant Physiol 44: 1584- 1589

Reed JW, Nagpal P, Poole DS, Furuya M, Chory J (1993) Mutations in the gene
for the red/far-red light receptor phytochrome B alter cell elongation and physiological
responses throughout Arabidopsis development. Plant Cell 5: 147-157

Relic B, Talmont F, Kopcinska J, Golinowski W, Prome’ J -C, Broughton
WJ (1993) Biological activity of Rhizobium sp NGR234 Nod-factors on Macroptilium
atropurpureum. Mol Plant-Microbe Interact 6: 764-774

Richardson HE, Wittenberg C, Cross F, Reed SI ( 1989) An essential G1
function for cyclin-like proteins in yeast. Cell 59: 1127-1133

Richmond AE, Lang A B (1957) Effect of kinetin on protein content and survival of
detached Xanrhium leaves. Science 125: 650-651

Riker AJ, Banfield WM, Wright WH, Keitt WH, Sagen HE (1930) Studies of
Infectious Hairy Root of nursery apple trees. J Agr Res 41: 507-540

41

Rohrig H, Schmidt J, Wieneke U, Kondorosi E, Barlier I, Schell J, John M
(1994) Biosynthesis of lipooligosaccharide nodulation factors: Rhizobium N odA

protein is involved in N-acylation of the chitooligosaccharide backbone. Proc Natl Acad
Sci USA 91: 3122-3126

Rousselin P, Kraepiel Y, Maldiney R, Miginiac E, Caboche M (1992)
Characterization of three hormone mutants of Nicotiana plumbaginifolia: Evidence for a
common ABA deficiency. Theor Appl Genet 85: 213-221

Schaller GE, Bleecker AB (1995) Ethylene-binding sites generated ln yeast expressing
the Arabidopsis ETRI gene. Science 270: 1809- 1811

Schmidt J, Rohrig H, John M, Wieneke U, Stacey G, Koncz C, Schell J
(1993) Alteration of plant growth and development by Rhizobium nodA and nodB genes
involved in the synthesis of oligosaccharide signal molecules. Plant J 4: 651-658

Schmulling T, Beinsberger S, De Greef J, Schell J, Van Onckelen H,

Spena A (1989) Construction of a heat-inducible chimeric gene to increae the cytokinin
content in transgenic plant tissue. FEBS Lett 249: 401-406

Schmulling T, Fladung M, Grossmann K, Schell (1993) Hormonal content and
sensitivity of transgenic tobacco and potato plants expressing single rol genes of
Agrobacterium rhizogenes T-DNA. Plant J 3: 371-382

Schmulling T, Schell J, Spena A (1988) Single genes from Agrobacrerium
rhizogenes influence plant development. EMBO J 7: 2621-2629

Schroder G, Waffenschmidt S, Weiler EW, Schroder J (1984) The t-region of
Ti plasmids codes for an enzyme synthesizing indole-3-acetic acid. Eur J Biochem 138:
387-391

Schwarz-Sommer Z, Huijser P, Nacken W, Saedler H, Sommer H (1992)
Characterization of the Antirrhinum floral homeotic MADS-box gene deficiens: Evidence
for DNA binding and autoregulation of its persistent expression throughout flower
development. EMBO J 11: 251-263

Shen WH, Petit A, Guern A, Tempe J (1988) Hairy roots are more sensitive to
auxin than normal roots. Proc Natl Acad Sci USA 85: 3417-3421

Silver DL, Pinaev A, Chen R, de Bruijn FJ (1996) Posttranscriptional regulation
of the Sesbania rostrata early nodulin gene SrEnodZ by cytokinin. Plant Physiol 112:
559-567

Singh S, Letham DS, Zhang X, Palni LMS (1992) Cytokinin biochemistry in
relation to leaf senescence. VI Effect of nitrogenous nutrients on cytokinin levels and
senescence of tobacco leaves. Physiol Plant 84: 262-268

Smart CM, Scofield SR, Bevan MW, Dyer TA (1991) Delayed leaf senescence in
tobacco plants transformed with rmr, a gene for cytokinin production in Agrobacten'um.
Plant Cell 3: 647- 656

Smigocki AC (1991) Cytokinin content and tissue distribution in plants transformed by a
reconstructed isopentenyl transferase gene. Plant Mol Biol 16: 105-115

42

Smigocki AC, Neal JW Jr, McCanna I, Douglass L (1993) Cytokinin-mediated
insect rgsistance in Nicotiana plants transformed with the ipt gene. Plant Mol Biol 23:
325-33

Smigocki AC, Owens LD (1988) Cytokinin gene fused with a strong promoter
enhances shoot organogenesis and zeatin levels in transformed plant cells. Proc Natl
Acad Sci USA 85: 5131-5135

Smigocki AC, Owens LB (1989) Cytokinin-to-auxin ratios and morphology of shoots
and tissues transformed by a chimeric isopentenyl transferase gene. Plant Physiol 91:
808-811

Somers DE, Sharrock RA, Tepperman JM, Quail PH (1991) The hy3 long
hypocotyl mutant of Arabidopsis is deficient in phytochrome B. Plant Cell 3: 1263-1274

Sommer H, Beltran J-P, Huijser P, Pape H, Lonnig W-E, Saedler H,
Schwarz-Sommer Z (1990) Deficiens, a homeotic gene involved in the control of

flower morphogenesis in Antirrhinum majus: The protein shows homology to
transcription factors. EMBO J 9: 605-613

Soni R, Carmichael JP, Shah ZH, Murray JA (1995) A family of cyclin D
homologs from plants differentially controlled by growth regulators and containing the
conserved retinoblastoma protein interaction motif. Plant Cell 7: 85-103

Sossountzov L, Maldiney R, Sotta B, Sabhagh I, Hahricot Y, Bonnet M,
Miginiac E (1988) Immunocytochemical localization of cytokinins in Craigella tomato
and a sideshootless mutant. Planta 175: 291-304

Spaink HP, Aarts A, Bloemberg GV, Folch J, Geiger O, Schlaman HRM,
Thomas-Oates JE, van Brussel AAN, van de Sande K, van Spronsen P,
Wijfjes AHM, Lugtenberg BJJ (1993) Rhizobial lipo-oligosaccharide signals:
Their biosynthesis and their role in the plant. In EW Nester, DPS Verma, eds, Advances
in Molecular-Genetics of Plant-Microbe Interactions. Kluwer Academic Publishers,
Dordrecht,The Netherlands pp 1 5 1 - 162

Spaink HP, Sheely DM, van Brussel AAN, Glushka J, York WS, Tak T,
Geiger 0, Kennedy EP, Reinhold VN, Lugtenberg BJJ (1991) A novel
highly unsaturated fatty acid moiety of lipo-oligosaccharide signals determines host
specificity of Rhizobium. Nature 354: 125-130

Spena A, Aalen RB, Schulze S (1989) Cell-autonomous behavior of the MIC gene
of Agrobacterium rhizogenes during leaf development: A visual assay for transposon
excision in transgenic plants. Plant Cell 1: 1157-1164

Spena A, Schmulling T, Koncz C, Schell J (1987) Independent and synergistic
activity of rat A, B and C loci in stimulating abnormal growth in plants. EMBO J 6:
3891-3899

Staehelin C, Granado J, Muller J, Wiemken A, Mellor RB, Felix G,
Regenaas M, Broughton WJ, Boller T (1994) Perception of Rhizobium
nodulation factors by tomato cells and inactivation by root chitinases. Proc Natl Acad
Sci USA 91: 2196-2200

43

Stokkermans TJW, Peters NK (1994) Bradyrhizobium elkanii lipooligosaccharide
signal induces complete nodule structures on Glycine soja Siebold et Zucc. Planta 193:
413-420

Sturtevant DB, Taller BJ (1989) Cytokinin production by Bradyrhizobium japonicum.
Plant Physiol. 89: 1247-1252

Su W, Howell SH (1992) A single genetic locus, ckrI, defines Arabidopsis mutants in
which root growth is resistant to low concentrations of cytokinin. Plant Physiol 99:
1569-1574

Su W, Howell SH (1995) The effects of cytokinin and light on hypocotyl elongation in
Arabidopsis seedlings are independent and additive. Plant Physiol 108: 1423-1430

Taller BJ, Sturtevant DB (1991) In H Hennecke, DPS Verma, eds, Advances in
Molecular Genetics of Plant-Microbe Interactions. Kluwer Academic Publishers,
Dordrecht, The Netherlands pp 215-221

Teutonico RA, Dudley MW, Orr JD, Lynn DG, Binns AN (1991) Activity and
accumulation of cell division-promoting phenolics in tobacco tissue cultures. Plant
Physiol 97: 288-297

Thimann KV (1936) On the physiology of the formation of nodules on legume roots.
Proc Natl Acad Sci USA 22: 511-514

Thomas H, Stoddart JL (1980) Leaf Senescence. Annu Rev Plant Physiol 31: 83-112

Thomas J C, Bohnert HJ (1993) Salt stress perception and Plant growth regulators in
the halophyte Mesembryanthemum crystallinum. Plant Physiol 103: 1299-1304

Thomas J C, McElwain EF, Bohnert HJ (1992) Convergent induction of osmotic
stress-responses. Abscisic acid, cytokinin, and the effects of NaCl. Plant Physiol 100:
416-423

Thomas JC, Smigocki AC, Bohnert HJ (1995) Light-induced expression of ipt
from Agrobacterium tumefaciens results in cytokinin accumulation and osmotic stress
symptoms in transgenic tobacco. Plant Mol Biol 27: 225-235

Tinland B, Hohn B (1995) Recombination between prokaryotic and eukaryotic DNA:
integration of Agrobacterium tumefaciens T-DNA into the plant cell. Genet Eng (NY)
17: 209-229

Tong Z, Kasemir H, Mohr H (1983) Coaction of light and cytokinin in
photomorphogenesis. Planta 159: 136-142

Truchet G, Barker DG, Camut S, de Billy F, Vasse J, Huguet T (1989)
Alfalfa nodulation in the absence of Rhizobium. Mol Gen Genet 219: 65-68

Truchet G, Roche P, Lerouge P, Vasse J, Camut S, de Billy F, Prome’ JC,
Denarie J (1991) Sulphated lipooligosaccharide signals from Rhizobium meliloti elicit
root nodule organogenesis in alfalfa. Nature 351: 670-673

44

Upadhyaya NM, Letham DS, Parker CW, Hocart CH, Dart PJ (1991) Do
rhizobia produce cytokinins? Biochem Int 24: 123-30

Van de Sande S, Pawlowski K, Czaja I, Wieneke U, Schell J, Schmidt J,
Walden R, Matvienko M, Wellink J, van Kammen A, Franssen F,
Bisseling T (1996) Modification of phytohormone response by a peptide encoded by
Enod40 of legumes and a nonlegume. Science 273: 370-373

Venglat SP, Sawhney VK (1996) Benzylarninopurine induces phenoc0pies of floral
meristem and organ identity mutants in wild-type Arabidopsis plants. Planta 198: 480-
487

Verma DPS (1992) Signals in root nodule organogenesis and endocytosis of Rhizobium.
Plant Cell 4: 373-382

Ward ER, Uknes JJ, Williams SC, Dincher SS, Wiederhold DL, alexander
DC, Ahl-Goy P, Metraux J-P, Ryals JA (1991) Coordinate gene activity in
response to agents that induce systemic acquired resistance. Plant Cell 3: 1085-1094

White FF, Taylor BH, Huffman GA, Gordon MP, Nester EW (1985)
Molecular and genetic analysis of the transferred DNA regions of the root-inducing
plasmid of Agrobacterium rhizogenes. J Bacteriol 164: 33-44

White R (1996) Cytokinin and nitrate regulation of nitrate reductase in Agrostemma
githago. PhD thesis, Michigan State University, East Lansing, Michigan

Willmitzer L, Dhaese P, Schreier PH, Schmalenbach W, Van Montagu M,
Schell J (1983) Size, location and polarity of T-DNA-encoded transcripts in nopaline
crown gall tumors; common transcripts in octopine and opaline tumors. Cell 32: 1045-
1056

Wittenherg C, Sugimoto K, Reed SI (1990) Gl-specific cyclins of S cerevisiae cell
cycle periodicity reguation by mating pheromone and association with the p34m
protein kinase. Cell 62: 225-237

Yamada T, Palm CJ, Brooks B, Kosuge T (1985) Nucleotide sequences of the
Pseudomonas savastanoi indoleacetic acid genes show homology with Agrobacterium
tumefaciens DNA. Proc Natl Acad Sci USA 82: 6522-6526

Yang W-C, de Blank C, Meskiene I, Hirt H, Bakker J, van Kammen A,
Franssen H, Bisseling T (1994) Rhizobium Nod factors reactivate the cell cycle
during infection and nodule primordium formation, but the cycle is only completed in
primordium formation. Plant Cell 6: 1415-1426

Zaenen 1, Van Larabeke N, Teuchy H, Van Montagu M, Schell J (1974)
Supercoiled circular DNA in crown gall inducing Agrobacterium strains. J Mol Biol 86:
109-127

Zupan JR, Zambryski P (1995) Transfer of T-DNA from Agrobacterium to the plant
cell. Plant Physiol 107: 1041-1047

Chapter 2

POSTRANSCRIPTIONAL REGULATION OF THE SESBANIA
KROSTRATA EARLY NODULIN GENE SRENODZ BY CYTOKININ

The contents of this chapter have been published in Plant Physiol, 1996 (Silver, Pinaev, and de Bruijn)

45

46

ABSTRACT

The mRN A from the Sesbania rostrata early nodan gene SrEnodZ accumulates in
response to cytokinin application. Nuclear mn-on assays using isolated root nuclei have
shown that this accumulation occurs posttranscriptionally, and northern blot analysis of
nuclear and total RNA levels revealed that it occurs primarily in the cytoplasm and not in
the nucleus. After cytokinin enhancement of SrEnodZ mRN A accumulation and the
subsequent removal of cytokinin, the levels of SrEnodZ mRNA did not return to basal
levels, but oscillated over a 36—h time course. Application of the translational inhibitor
cycloheximid, was found to inhibit the enhancement of SrEnodZ mRN A accumulation by
cytokinin and to cause its rapid decay. Okadaic acid and staurosporine, inhibitors of
protein phosphatases and kinases, respectively, also inhibited cytokinin enhancement of
SrEnodZ mRN A accumulation. In addition, okadaic acid was found to cause a decrease in
SrEnodZ mRN A levels. These results provide evidence for a posttranscriptional
mechanism of cytokinin enhancement of SrEnodZ mRN A accumulation, which appears to
require concurrent protein synthesis, to involve protein phosphatases and kinases, and to

occur primarily in the cyt0plasm of the plant cell.

47

INTRODUCTION

The plant hormone cytokinin comprises of a group of plant growth substances that
are derived from adenine. It has been shown that cytokinins induce cell division and
organogenesis in cell cultures (Skoog and Miller, 1957), and also affect other physiological
and developmental plant processes (Evans, 1984; Brzobohaty et al., 1994; Davies 1995).
Cytokinin and auxin have been shown to play a central role in photomorphogenesis and
elongation growth (Hobbie et al., 1994). The available information on auxin signal
nansduction has been accumulating rapidly with the identification of auxin-induced
mRN As and auxin-binding proteins, the cloning of putative auxin receptors, and auxin-
responsive DNA elements, as well as the characterization of mutants in auxin responses
(for review see Hobbie et al., 1994), and the cloning of a gene responsible for one of the
auxin resistant mutant phenotypes (Leyser et al., 1993).

Only a limited number of cytokinin response mutants have been isolated (Moffatt et
al., 1991; Chaudhury et al., 1993; Deikman and Ulrich., 1995), and the genes
corresponding to these mutant loci have yet to be identified. Putative cytokinin-binding
proteins have been purified (Brzobohaty et al., 1994), but the demonstration of their
biological activity is still lacking. Overall, little is known about the molecular mechanisms
of cytokinin signal transduction.

At the molecular level, cytokinin has been shown to modulate enzyme activities
(Treharne et al., 1970; Chatfield and Armstrong, 1986), and transcript levels of a variety of
genes. mRNAs whose accumulation is enhanced by cytokinins, including those encoded
by a wheat protein kinase gene (Sano and Youssefian, 1994 ), the gene for the small
subunit of Rubisco (Rch) (Flores and Tobin, 1988), the chlorophyll a/b binding protein
gene (Cab) (Flores and Tobin, 1988), defense-related genes (Memelink et al., 1987), the
PEP carboxylase gene (PepC) (Suzuki et al., 1994), nitrate reductase genes (Lips and
Roth-Bejerano, 1969; Dilworth and Kende, 1974; Lu et al., 1990), the pollen allergen gene

48

CimI (Crowell, 1994), the multiple stimulus response gene pLS216 (Dominov et al.,
1992), genes of the anthocyanin biosynthetic pathway (Deikman and Hammer, 1995),
cyclin D homologs (Soni et al., 1995), and genes for a number of unidentified cDNAs
(Crowell et al., 1990). It was shown by nuclear run-on assays that the expression of genes
encoding PEP carboxylase (Suzuki et al., 1994), nitrate reductase (Lu et al., 1990),
chalcone synthase, and dihydrofolate reductase (Deikman and Hammer, 1995) are
enhanced primarily at the transcriptional level by cytokinin. On the contrary, nuclear run-
on assays show that some genes are regulated by cytokinin primarily at the post-
transcriptional level. These genes include those encoding the chlorophyll a/b binding
protein, the small subunit of Rubisco (Flores and Tobin, 1988), chalcone isomerase, and
Phe ammoniumlyase l (Deikman and Hammer, 1995).

The genes involved in the cytokinin-induced accumulation of mRN A are diverse, as
are the mechanisms of cytokinin induction. Cytokinin appears to enhance the transcription
of genes or enhance mRNA accumulation posttranscriptionally. It is important to note that
in all the examples reported thus far, cytokinin enhancement of gene expression is never
exclusively the result of cytokinin action, but generally co-mediated by other environmental
factors, such as light, nitrogen, carbon, or other plant hormones. This has complicated the
analysis of the molecular basis of cytokinin action. One exception is the Sesbania rostrata
early nodulin gene SrEnodZ. The SrEnodZ gene encodes a Pro-rich protein expressed in a
cell-specific manner in nodules of legumes (van de Wiel et. al., 1990). This cell layer,
which surrounds the cells infected with nitrogen-fixing bacteria, is called the nodule
parenchyma. It has been previously shown by Dehio and de Bruijn (1992) that the
SrEnodZ mRNA accumulates in S. rostrata roots in the absence of rhizobia in a time- and
concentration-dependent manner in response to cytokinin treatment. The root cell type in
which the SrEnodZ gene is expressed has not been determined. This accumulation of
SrEnodZ mRNA in unnodulated roots occurs primarily in the primary root, and to a lesser

extent in fully developed lateral roots, which correlates well with the observed GUS

49

expression pattern in transgenic Lotus japonicus plants harboring SrEnodZ-GUS fusions
(see Chapter 4 of this thesis). SrEnodZ is a good gene with which to study cytokinin
signal transduction, since its mRNA accumulation is stimulated solely by cytokinin (Dehio
and de Bruijn, 1992; Hirsch and Fang, 1994). We sought to determine the most important
parameters affecting this process. We demonstrate that SrEnodZ mRNA accumulation is
posttranscriptionally enhanced by cytokinin, and that this mechanism requires ongoing
protein synthesis, involves protein phosphatases and kinases, and occurs primarily in the

cytoplasm.

50

METHODS
Plant material and treatments

Sesbania rostrata seeds were germinated and seedlings grown in soil composed of
Metromix (Hummert International, Earth City, MO) and sand (2: 1) at 30°C, with a 18-h
light, 28°C/6-h dark, 22°C regime for 2 weeks in growth chambers with 75% RH. For all
chemical treatments, plants were washed free of soil and incubated in a 1:2 dilution of
Murashige-Skoog minimal organic medium (Murashige-Skoog, Gibco-BRL), along with
the appropriate chemicals, under normal growth conditions. 6-BAP (Sigma),
cycloheximide (Sigma), okadaic acid (Gibco-BRL), and staurosporine (Sigma) were used
at concentrations of 10 uM, 140 M, 0.5 uM, and 10 11M, respectively. Following

chemical treatments, root tissues were excised, frozen in liquid nitrogen, and stored at -

80°C.
Isolation of nuclei and analysis of nuclear run-on transcripts.

Frozen roots were ground in liquid nitrogen to a fine powder. The powder was
resuspended in nuclei isolation buffer (20 mM MES, pH 6.5; 2.5% Ficoll 400; 2.5%
Dextran 40000 [Sigma]; 50 mM KCl; 0.44 M sucrose; 0.1% thiodiglycol; 0.5 mM
spermidine; 0.1 mM sperrnine; 0.5 mM EDTA; 0.5% Triton X-100; 5 rig/ml aprotinin,
leupeptin, and leupeptin [Sigma]). The resuspended material was passed through four
layers of cheesecloth, two layers of Miracloth (Calbiochem), 1 layer of loo-um mesh, and
spun at 2,500 rpm for 15 min (I-IB4 rotor, Sorvall). The pellet was resuspended in nuclei
isolation buffer, passed through a 20-um mesh, and spun at 2,000 rpm for 15 min. The
pellet was resuspended in nuclei isolation buffer. Nuclei were counted using the DNA

stain 4’,6-diarnidino-2-phenylindole dihydrochloride, and aliquots of 2 x 106 nuclei were

51

frozen at -80° C. Nuclear run-on assays were performed as described by DeRocher and
Bohnert (1993). Slot blot filters containing 5 ug of linearized plasmid DNA containing the
SrEnodZ coding region (Dehio and de Bruijn, 1992), pUC19 (New England BioLabs), and
the BATPase gene ( Boutry and Chua, 1985) on nitrocellulose membrane (Bio-Rad) were
used to hybridize with equal counts of transcripts (5 x 107 cpm). Filters were washed at
65°C in 2.0 x SSC (0.3 M NaCl, 0.03 M sodium citrate), 0.1% SDS for 20 min; 0.5 x
SSC, 0.1% SDS for 20 min; 0.1 x SSC, 0.1% SDS for 30 min. The signals were
quantified using phosphorimager analysis (Model 400B, Molecular Dynamics, Sunnyvale,
CA)

Isolation of nuclei for RNA extraction.

Nuclei were isolated as described by Peters and Silverthome (1995) with the
following modification: B—mercaptoethanol was replaced by 10 mM of the RNase inhibitor

ribonucleoside-vanadyl complex (Gibco-BRL). Nuclei were resuspended in RNA
extraction buffer, and RNA isolated as described by Verwoerd et al. (1989).

Northern blot analysis.

10 pg of RNA was electrophoresed in 1.2% (w/v) agarose gels in Mops buffer
(20mM MOPS, 1.0 mM EDTA, 5.0 mM sodium acetate, pH 7.0) containing 5.4% (v/v)
formaldehyde. Gels were blotted onto 0.22 M NitroPlus nitrocellulose membrane
(Micron Separation Inc., Westborough, MA). Membranes were probed with a [32P] dATP
labeled DNA fragment containing the SrEnodZ coding region generated with random
priming (Boehringer Mannheim). Filters were reprobed with an 18S rRNA DNA probe as

a loading control. The use of the flATPase gene (Boutry and Chua, 1985) and the soybean

52

actin gene, pSAC3 (Shah 1982) as further controls are indicated in the Results section. All
filters were washed at 65°C in 2.0 x SSC, 0.1% SDS for 20 min; 0.5 x SSC, 0.1% SDS
for 20 min; and 0.1 x SSC, 0.1% SDS for 15 min. The signals were quantified using
phosphorimager analysis (Model 400B, Molecular Dynamics, Inc.).

53

RESULTS

Transcription of the SrEnodZ gene and accumulation of its mRNA in

response to cytokinin

Run-on transcription assays were performed using isolated nuclei to determine
whether SrEnodZ mRN A is regulated at the transcriptional and/or post-transcriptional level
by cytokinin. Two-week-old S. rostrata seedlings were treated with the cytokinin
benzylarninopurine (BAP) for time periods of 0 (no BAP), 2, 4, and 6 h, after which nuclei
and RNA were isolated from root tissues. A northern blot analysis of the RNA samples is
shown in Figure 2.1A. SrEnod2 mRNA accumulated to levels approximately 4-fold higher
than those of the control over the time course of BAP treatment, as shown in Figure 2.1A.
A second mRNA smaller than SrEnodZ was frequently observed on northern blots (Figure
2.1A), the origin of which is unclear. This second mRNA probably does not represent a
second SrEnodZ gene, since Southern blot analysis indicates that SrEnodZ exists as a
single-copy gene in S. rostrata (Dehio, 1989), and the northern blots were washed at high
stringency. More likely, this smaller RNA may be a processing product derived from the
SrEnodZ mRN A. Radiolabelled transcripts from the nuclei were hybridized with an
immobilized SrEnodZ DNA probe on slot blots. No change in transcription of the SrEnodZ
gene was observed over the time course of BAP treatment (Figure 2.1B), as indicated by
the SrEnodZ signal relative to the BATPase signal using phosphorimager analysis. A
comparison of the northern blot and nuclear run-on data (Figure 2.1C) showed that
although SrEnodZ mRN A accumulates over time, no detectable change in transcription was
evident. These data suggest an involvement of posttranscriptional processes in SrEnodZ
mRN A accumulation in response to cytokinin. To examine the possibility that SrEnodZ
mRN A stability was altered in the presence of cytokinin, we examined SrEnodZ mRN A

half-life using the cellular RNA synthesis inhibitor actinomycin-D. However, the detection

54

Figure 2.1. Comparison of nuclear run-on transcription with mRNA
accumulation. A, Northern blot analysis of SrEnodZ mRN A accumulation enhanced by
cytokinin. Two-week-old S. rostrata seedlings were incubated in the presence of 10 “M
BAP for time periods of 0 h (untreated), 2 h, 4 h, and 6 h, and the total RNA was then
extracted from roots. B, Radiolabeled run-on transcripts from root nuclei isolated from the
same seedlings treated in A were hybridized with immobilized SrEnodZ, pUC19 as
background plasmid control, and fiATPase probes on slot blots. C, Signals from northern
blot and run-on transcription assays were quantified by phosphorimager analysis and
plotted as the increase (-fold) in signal as standardized to BATPase mRNA levels. The SD
of two independent nuclear run-on transcription assays is shown.

Circle, Run-on transcription; Square, mRNA accumulation.

A

55

B

 

Time (h) Time (h)
0 2 4 6 o 2 4 6
.. _ _ SrEnod2 .1 - SrEnod2
‘ fl. --- ‘ pUCl9
.-_, a... 188 i p - BATPase

0

Fold increase (arbitrary units)

  

 

 

 

 

 

 

 

 

 

6
—0— run-on transcription
5 ,_ _.._- mRNA accumulation )- ___________ 4.
[I
I
[I
4 ' /
I
/
I
I

> I,

3 1’
I
I
[I
2 ” /’/
’I
I
__L
1 I - q\f
l

o i

o 2 4 6

Time (h)

56

of differences in SrEnodZ mRNA stability in response to cytokinin treatment has so far not
been possible, because it was found that actinomycin-D stabilizes the SrEnodZ mRN A
(Silver and de Bruijn, unpublished results), making this type of analysis impossible. The
effect of mRNA stabilization due to transcription inhibitors has also been observed in the
case of the PhyA (Seeley et al., 1992), rch (Fritz et al., 1991), PvPrPI (Zhang et al.,
1993), and Fd mRNAs (Dickey et al., 1994).

SrEnodZ mRNA accumulation in response to cytokinin occurs primarily in

the cytoplasm

The nuclear run-on data suggest that SrEnodZ mRN A accumulates
posttranscriptionally, but do not provide insight into the question whether SrEnodZ mRN A
accumulation is a nuclear and/ or cytoplasmic event. To better understand the cytokinin
signal transduction pathway, it was important to determine if SrEnodZ mRNA accumulated
in the nucleus or in the cytoplasm. Events which occur in the nucleus after transcription
include pre-mRN A processing, turnover, and transport to the cytoplasm. To investigate
the fate of SrEnodZ mRNA in the nucleus, mRNA was isolated from S. rostrara roots
treated and untreated with BAP, and analyzed by northern blot hybridization (Figure 2.2A).
There was only a 1.2-fold increase in SrEnodZ mRNA accumulation in the nucleus,
compared with an approximately 4-fold increase in total cellular RNA (Figure 2.2B). It is
important to note here that the levels of SrEnodZ mRN A in the nucleus constitute
approximately 38% of the total SrEnodZ RNA, indicating that most of the SrEnodZ
transcripts are located in the cytoplasm. Therefore, the accumulation of SrEnodZ mRN A

appears to occur primarily in the cytoplasm and not in the nucleus.

57

Figure 2.2. Abundance of SrEnodZ mRN A derived from total RNA and
nRNA. A, Representative northern blot analysis of total RNA and nRNA isolated from
roots of S. rostrata seedlings treated with or without 10 “M BAP for 4 h. B,
Quantification of SrEnod2 mRN A accumulation in both total and nRN A. Signals from two
independent experiments were quantified using phosphorimager analysis. SrEnodZ signals

were standardized to 188 rRNA signals. The SD is shown.

58

<72 322.2

<ZM :38.

A

+ BAP

S
8
1

 

 

 

 

4.5

5.
3

3 5. 2 5
2 1..

A38: b23543 5:965 Eon

1

 

0.5 “*

Total RNA Nuclear RNA

59

SrEnod2 mRNA accumulation appears to be a long-lived process

To better define the link between SrEnod2 mRN A accumulation and cytokinin
application, we determined the level to which SrEnod2 mRN A would increase, and to what
extent the mRNA would be present over time after the removal of cytokinin to determine if
this process was a short-lived or a long-lived response to cytokinin. S. rostrata seedlings
were incubated in the presence of BAP for 4 h, then divided into two groups. One group
was washed extensively in water to remove BAP and was placed in BAP-free medium.
Another group was further incubated in media with BAP. A third group of S. rostrata
seedlings was not exposed to BAP and was placed continuously in BAP-free medium. All
three groups of seedlings were incubated for 36 h under these conditions. Roots were
harvested at 4-h intervals, and RNA was isolated and analyzed by northern blot
hybridization. Neither group of seedlings exposed to BAP exhibited SrEnod2 mRN A
decay below induced levels during the 36 h time course (Figure 2.3). In addition, a
rhythmic oscillation with essentially the same pattern of oscillation was observed in both
groups treated with BAP. However, the level of SrEnod2 mRNA in samples treated with
BAP for only 4 h appeared to be slightly higher than in samples treated for 36 h. The
reason for this apparent difference is unclear, but may be due to signal attenuation in
samples continuously exposed to cytokinin. These data may indicate that the process by
which SrEnod2 mRN A accumulates is long-lived. Whether the observed oscillation is due
in part to a change in stability or synthesis of the SrEnod2 mRNA is unknown.

Cycloheximide inhibits SrEnod2 mRNA accumulation in response to

cytokinin

It is a common observation that the levels of mRNA, which is unstable, is increased

by the translational inhibitor cycloheximide (Zhang et al., 1993; Dickey et al., 1994; Gil et

60

Figure 2.3. Fluctuation in SrEnod2 mRNA accumulation during a 36-h
period. Two-week-old S. rostrata seedlings were divided into three groups according to
the following treatments: 36 h no BAP (circle); 36 h 10 11M BAP (square); 4 h 10 M
BAP then washed free of BAP and incubated for 32 h (triangle). RNA from roots of 5
seedlings were isolated every 4 h for 36 h. Signals were quantified using phosphorimager
analysis. SrEnod2 signals were standardized to fiATPase mRNA signals.

Relative mRNA level

61

 

0.9 -

0.7 -

 

 

 

 

"-0" 36hno BAP
--l--36hBAP
+4 h BAP. 32 h no BAP;

 

 

 

62

al., 1994). There are three possible reasons for this. Cycloheximide may inhibit the
accumulation of unstable components of the mRN A degradation machinery (Brewer and
Ross, 1989), or translation of the specific mRNA in question may be required for its
degradation (Aharon and Schneider, 1993; Byme et al., 1993; Tanzer and Meagher, 1994).
Another possibility is that cycloheximide, through a secondary effect, increases
transcription (Edwards and Mahadevan, 1992). Therefore, it was of interest to determine if
cytokinin could directly stimulate SrEnod2 mRN A accumulation in the absence of
translation, and/or if the SrEnod2 mRN A exhibited properties common to unstable
transcripts in the presence of cycloheximide. A typical concentration of cycloheximide
used in plants is between 50 to 70 11M in cell culture(Heinhorst et a1 1985; Gil et al 1994),
and up to 300 M (for 2-h pretreatment) in intact plants (Satoru et a1, 1995). In our system
70 11M cycloheximide inhibited the accumulation of SrEnod2 mRNA, but required slightly
more than 1 h of pretreatment to be effective. A concentration of 140 1.1M was effective
after a pretreatment of between 30 min to 1 h. (data not shown). It may be that a higher
concentration of cycloheximide is needed to elicit an effect in whole plants than is needed in
cell culture due to the timing of uptake of the substance. Two-week-old S. rostrata
seedlings were pretreated for 1 h with 140 M cycloheximide. The seedlings were then
treated with BAP for 4 h, after which RNA was isolated and SrEnod2 mRN A levels
examined by northern blot hybridization. It was observed that 1 h of cycloheximide
pretreatment prevented subsequent stimulation of SrEnod2 mRN A accumulation by BAP
(Figure 2.4), suggesting that protein synthesis may be involved in the observed cytokinin
stimulation of SrEnod2 mRN A accumulation. However, one must be cautious when
interpreting results from experiments involving cycloheximide, because of its effect on

global translation.

63

Figure 2.4. Effect of cycloheximide on SrEnod2 mRNA accumulation.

Two-week-old S. rostrata seedlings were pretreated for l h with 140 “M cycloheximide
(CHX), and then extensively washed and further incubated for an additional 4 h with or
without 10 11M BAP. Control seedlings were untreated with cycloheximide. Following

the treatments, RNA from roots were isolated and analyzed by northern blot hybridization.

64

nvm<m+.xmof
:vfim-.xmvc_
avg<m+
2:235-

SrEnod2

188

that

65

In addition to the inhibition of SrEnod2 mRN A accumulation, a slight decrease in
SrEnod2 mRN A was observed in samples treated with cycloheximide with or without BAP
present. Because of the observed slight decrease in levels of SrEnod2 mRNA, it was
subsequently examined whether cycloheximide could cause the decay of SrEnod2 mRN A
over time after an initial BAP treatment of 4 h. Plants were treated with BAP for 4 h, after
which cycloheximide was added, and RNA was harvested every hour for 4 h. Control
plants were not treated with BAP, but were treated with cycloheximide. Northern blot
analysis revealed that the SrEnod2 mRNA from control plants and from plants treated with
BAP decayed upon treatment with cycloheximide (Figure 2.5), whereas the SrEnod2
mRNA from plants treated with BAP but not with cycloheximide increased approximately
2-fold over this same time period of 4 h (refer to Figure 2.3 between 4 h and 8 h).

An interesting observation was made after analyzing the samples from BAP-treated
, plants at the 0- and 2-h time points after cycloheximide addition (Figure 2.5B). Between
the 0- and 2-h time points, the SrEnod2 mRN A increased slightly, and then decayed
dramatically. The apparent increase between 0 and l h may be explained by a lag time
required for cycloheximide to be effective, but the reason for the rapid decay after the 1-h
time point is not known. In addition, between 2 and 4 h after cycloheximide application,
SrEnod2 mRN A from both BAP-treated and untreated samples decayed similarly. To
exclude the possibility that cycloheximide altered the levels of gene expression in general,
we reprobed the northern blots with the BATPase gene. We found that the levels of
fiATPase mRN A were not affected by cycloheximide, suggesting that the observed effects
of cycloheximide on SrEnod2 mRN A accumulation may be specific to the SrEnod2 gene
(data not shown). Although direct evidence for an inhibition of cellular translation, and/or
translation of the SrEnod2 mRN A by cycloheximide has not been demonstrated under the

conditions used in these experiments, these data may suggest that ongoing protein

66

Figure 2.5. Decay of SrEnod2 mRNA in the presence of cycloheximide. A,

Two-week-old S. rostrata seedlings were initially treated with (+) or without (-) 10 1.1M

BAP for 4 h after which 140 M cycloheximide (CHX) was added and RNA from roots of
5 seedlings isolated every hour for 4 h and analyzed by northern blot hybridization. B,
Signals were quantified using phosphorimager analysis. A value of 100% on the ordinate
represents the maximum mRNA levels detected, as standardized to 188 rRNA signals.
Square, With BAP; circle, without BAP.

67

A

BAP'----+++++
CHX-++++-++++

Time(h) o 1 2 3

4 0 1 2
h- -. . . . ~— SrEnod2
Ha , a , u

Unuuauuuv" ’83

03

 

% mRNA remaining

 

 

 

 

Time (h)

68

synthesis is required for the continued accumulation of both basal and cytokinin-stimulated

levels of SrEnod2 mRNA.

Both cellular protein phosphatases and protein kinases may be required for

the accumulation of SrEnod2 mRNA by cytokinin

Since cycloheximide was found to inhibit the enhancement of SrEnod2 mRNA
accumulation by cytokinin, we postulated that cytokinin may act far upstream of the actual
event that actually causes the accumulation of SrEnod2 mRNA. It has been shown that
cytokinins are involved in protein phosphorylation cascades (Sano and Youssefian, 1994).
We sought to determine if protein phosphatases and protein kinases, the mediators of signal
transduction pathways (Bowler and Chua, 1994), may be involved in this mechanism, as
has been documented for ethylene signal transduction pathways (Raz and Fluhr, 1993). To
test this hypothesis, the pharmacological agents okadaic acid and staurosporine were used.
Okadaic acid is an inhibitor of protein phosphatases type PPl and PP2B at submicromolar

concentrations and of PP2A at higher concentrations (Hunter, 1995). Okadaic acid and
staurosporine, a broad spectrum protein kinase inhibitor (Tarnaoki, 1991), were applied at
concentrations previously used for inhibitor studies in plants (Dominov et a1, 1992; Raz
and Fluhr, 1993; Crowell, 1994) to two-week-old S. rostrata seedlings in the presence or
absence of BAP for 4 h. Although okadaic acid and staurosporine are useful tools for
studying signal transduction, it should be noted that the nature of their in vivo action is not
known, so results obtained using these inhibitors should be interpreted with caution.
Okadaic acid was found to abolish SrEnod2 mRN A accumulation in response to BAP and
to cause a dramatic decrease in the levels of SrEnod2 mRN A in both BAP-treated and
untreated S. rostrata seedlings below levels observed in seedlings not treated with BAP
(basal level; Figure 2.6). Staurosporine was also found to inhibit SrEnod2 mRN A

accumulation in response to BAP, but did not reduce the amount of SrEnod2 mRNA below

69

Figure 2.6. Effect of okadaic acid and staurosporine on SrEnod2 mRNA
accumulation. Two-week-old S rostrata seedlings were preincubated for 1 h with 0.5
11M okadaic acid or 10 M staurosporine, then treated with (+) or without (-) 10 M BAP.

RNA was isolated from roots and analyzed by northern blot hybridization.

70

a
m
w a.

S

6:50 1

6283235

 

+

+
Bow 0398.0
+ 1..

 

Cine-0.. 18S

71

basal levels. Levels of both flATPase and actin mRNA were found to be unaffected
by the inhibitor treatments (data not shown).

The observation that both inhibitors of protein phosphatases and kinases also inhibit
the accumulation of SrEnod2 mRN A in the presence of cytokinin suggests the involvement

of these classes of proteins in the cytokinin signal transduction pathway.

72

DISCUSSION

In this paper we present evidence for a posttranscriptional mechanism of
enhancement of SrEnod2 mRN A accumulation in response to cytokinin. Nuclear run-on
experiments demonstrated that the SrEnod2 transcription rate cannot account for the
increase in SrEnod2 mRN A accumulation in response to cytokinin treatment. In addition,
the northern blot hybridization analysis of nuclear versus total RNA accumulation showed
that cytokinin enhancement of mRN A accumulation appears to occur primarily in the
cytoplasm. There is an approximately 4-fold enhancement in SrEnod2 mRN A
accumulation in total RNA and an approximately 1.2-fold enhancement in nRN A
populations, which may be due to a small increase in transcription and/or altered mRN A
stability in the nucleus. It has been documented that mRN A stability in the nucleus plays
an important role in gene expression (Belgrader and Maquat, 1994; Peters and
Silverthome, 1995). Of course, it also cannot be ruled out that the observed small
enhancement in nuclear SrEnod2 mRN A accumulation is due to contamination from
cytoplasmic RNA. Nevertheless, these data correlate well with the nuclear run-on data,
suggesting the involvement of mRN A stability changes in response to cytokinin treatment
in the cytoplasm (Jackson, 1993; Sullivan and Green, 1993; Beelman and Parker, 1995).

Cytokinin seems to be acting indirectly on SrEnod2 mRNA levels, since it was
found that the inhibition of translation by cycloheximide counteracted the enhancement of
SrEnod2 mRN A accumulation by cytokinin. cytokinin enhancement of the light-
dependent expression of the nitrate reductase gene has been shown by nuclear run-on
assays to be partially regulated at the transcriptional level. In addition, nitrate reductase
mRNA levels do not appear to be affected by cycloheximide (Lu et al., 1990). Lu et al.
(1990) proposed that concurrent protein synthesis is not required for cytokinin
enhancement of the gene for nitrate reductase (NR) mRN A accumulation. This appears to

be in direct contrast to the enhancement of SrEnod2 mRN A accumulation by cytokinin

73

observed here, which appears to require concurrent protein synthesis and to occur primarily
at the posttranscriptional level. Surprisingly, cycloheximide was found to cause the rapid
decay of SrEnod2 mRN A from plants treated with BAP and from those that were not
treated with BAP. This could be the result of an the inhibition of translation of a labile
protein required for SrEnod2 mRNA accumulation, or translation of the SrEnod2 mRN A
itself may be required for its stabilization.

In contrast to the observed rapid decay of SrEnod2 mRNA caused by
cycloheximide application, the SrEnod2 mRN A levels of plants treated with BAP and then
washed free of the hormone did not return to basal levels, but oscillated for 36 h in a
manner similar to those of plants continuously treated with BAP for 36 h. A circadian
fluctuation of mRN A accumulation enhancement was recently reported in genes of the
anthocyanin biosynthesis pathway (Deikman and Hammer, 1995). It that study, cytokinin
treatment caused a dramatic dampening in the diurnal fluctuations of mRN A accumulation
compared with plants grown without cytokinin, which suggests that cytokinin acts .
independently of the mechanism responsible for the circadian rhythm. This is in contrast to
our observation that an increase in amplitude of SrEnod2 mRN A occurs in the presence of
cytokinin. It remains to be determined whether this oscillation in SrEnod2 mRNA is
regulated in a circadian manner. In a study by Pilgrim et al. (1993) on nitrate reductase
mRNA, expression of which is known to be induced by cytokinin (Dilworth and Kende,
1974; Lu et al., 1990), it was found that nitrate reductase mRN A accumulation also
oscillates in a circadian fashion. These oscillations in mRN A accumulation were shown by
nuclear run-on assays to occur at a posttranscriptional level; therefore, the involvement of
posttranscriptional mechanisms for the oscillation in SrEnod2 mRNA accumulation may,
also-be plausible.

Okadaic acid and staurosporine were found to cause different effects on SrEnod2
mRN A accumulation. Okadiac acid inhibited cytokinin enhancement of SrEnod2 mRNA

accumulation, causing a dramatic decrease in its accumulation. We postulated that if

74

okadaic acid was able to inhibit cytokinin enhancement of SrEnod2 mRN A accumulation,
then an inhibitor of prorein kinases would probably cause an enhancement of SrEnod2
mRN A accumulation in the absence of cytokinin. This was found not to be the case, since
staurosporine inhibited SrEnod2 mRN A accumulation in response to cytokinin, but did not
cause a reduction in basal levels, as okadaic acid did. These data suggest the involvement
of both dephosphorylation and phosphorylation processes, perhaps acting on different
proteins of the cytokinin signal transduction pathway. The observation that okadiac acid
and cycloheximide cause a dramatic decrease in SrEnod2 mRN A accumulation in samples
treated or untreated with BAP may indicate that both inhibitors act on the same protein(s).
In addition, these data suggest that both basal and cytokinin-enhanced levels of SrEnod2
mRNA may be due to the same mechanism of mRNA accumulation. It may be that basal
levels of SrEnod2 mRN A are caused by a small flux through the signal transduction
pathway. The idea of a “flux”, as described by Bowler and Chua (1994), relates to basal
level activity of signaling intermediates, for example, protein kinases or phosphatases. In
the case of SrEnod2, endogenous cytokinin levels could be responsible for maintaining a
basal level of SrEnod2 mRNA, as observed by northern blot analysis, and this increase in
cytokinin levels above a threshold could cause an increase in SrEnod2 mRN A
accumulation.

Crowell (1994) showed for cytokinin-enhanced expression of the soybean allergen
gene, okadaic acid inhibits cytokinin induction, whereas for the cytokinin-enhanced
accumulation of pLS'216 RNA, staurosporine inhibits enhancement of mRNA accumulation
by cytokinin, and okadaic acid only slightly enhances mRNA accumulation (Dominov et
al., 1992). These studies, along with the observations presented here, suggest that the
phosphorylation state of proteins are important for cytokinin signal transduction.
However, differences may exist in the requirement for phosphorylation among different
pathways for accumulation of mRN As enhanced by cytokinin, as is the case for the signal

transduction pathways for other plant hormones (Bowler and Chua, 1994). The evidence

75

presented in this study strongly suggests that SrEnod2 is stabilized in the cytoplasm in
response to cytokinin. However, direct proof of regulated mRN A stability of SrEnod2
mRNA by cytokinin remains to be provided, as does the regulated mRN A stability of any
mRNA by a plant hormone. The use of a posttranscriptional mechanism of regulation of the
SrEnod2 gene may provide a more rapid means of controlling gene expression than the use
of transcriptional regulation (Green 1993).

There are questions that remain to be answered. For example, what is the role of
the Enod2 protein in nodulation and the significance of cytokinin regulation of SrEnod2
gene expression? It has previously been postulated by van de Wiel et al. (1990) that the
Enod2 protein may play a role in creating an oxygen diffusion barrier in the parenchyma
cell layer necessary for protecting the oxygen-sensitive nitrogenase enzyme in the infected
cells of the nodule. A direct test of this hypothesis has not yet been concluded, so the role
of the Enod2 protein in nodule development and functioning remains unclear. With regard
to cytokinin regulation of SrEnod2 gene expression, previous studies aimed at
understanding the involvement of cytokinin in nodule development have indirectly shown

an involvement of cytokinin in nodule development. One such study by Cooper and Long

(1994) demonstrated the partial rescue of a nonnodulating Rhizobium strain (N od' ) by the

overexpression of the Agrobacterium tumefaciens cytokinin biosynthesis gene( 12s). The

Nod' Rhizobium overexpressing the tzs gene was capable of eliciting the production of

nodule-er structures on alfalfa roots, indicating, at least in part, that a localized production
of cytokinin can produce a phenocopy of nodule morphogenesis.

Preliminary transgenic work has shown the requirement of the SrEnod2 3’
downstream region for tissue-specific expression in uninfected roots of the legume Lotus
japonicu (D.L. Silver and R]. de Bruijn, unpublished data), which may also point to a
posttranscriptional mechanism for the regulation of SrEnod2 gene expression by cytokinin.

It now remains to be determined whether DNA elements of the SrEnod2 gene required for

76

cytokinin regulation are the same as those required for tissue-specific expression in the
nodule. This type of analysis may shed some light on the relationship between cytokinin
regulation and the tissue-specific expression pattern of the SrEnod2 gene in nodules and in
roots. In addition, we are trying to develop the use of the SrEnod2 gene as a model system
for understanding cytokinin signal transduction, and to isolate trans-acting factors

responsible for cytokinin enhancement of plant gene expression.

77

ACKNOWLEDGMENTS

I thank Alexandr Pinaev for developing the protocol for nuclei isolation from roots
of S. rostrata plants. I thank Drs. Pamela Green and Hans Kende for comments on the
manuscript as well as all members of the de Bruijn laboratory for helpful discussions. I
thank Dr. Jay DeRocher for helpful suggestions with the nuclear run-on transcription
assays. I thank Dr. Jane Silverthome for providing us with the nuclear RNA isolation
protocol as well as helpful suggestions concerning this protocol. I also like to thank
Ambro van Hoof for his suggestion of using ribonucleoside-vanadyl complex for nuclear

RNA isolation.

78
REFERENCES

Aharon T, Schneider RJ (1993) Selective destabilization of short-lived mRNAs with
the granulocyte-macrophage colony-stimulating factor AU-rich 3’ noncoding region is
mediated by a cotranslational mechanism. Mol Cell Biol 13: 1971-1980

Beelman CA, Parker R (1995) Degradation of mRNA in eukaryotes. Cell 81: 179-183

Belgrader P, Maquat LE (1994) Nonsense but not missense mutations can decrease
the abundance of nuclear mRNA for the mouse major urinary protein, while both types
of mutations can facilitate exon skipping. Mol Cell Biol 14: 6326-6336

Boutry M, Chua N-H (1985) A nuclear gene encoding the beta subunit of the
mitochondrial ATP synthase in Nicotiana plumbaginifolia.. EMBO J 4: 2159-2165

Bowler C, Chua N-H (1994) Emerging themes of plant signal transduction. Plant Cell
6: 1529-1541

Brewer G, Ross J (1989) Regulation of c—myc mRNA stability by a labile destabilizer
with an essential nucleic acid component. Mol Cell Biol 9: 1996-1989

Brzobohaty B, Moore 1, Palme K (1994) Cytokinin metabolism: implications for
regulation of plant growth and development. Plant Mol Biol 26: 1483-1497

Byme DH, Seeley KA, Colbert JT (1993) Half-lives of oat mRNAs in vivo and in a
polysome—based in vitro system. Planta 189: 249-256

Chatfield MJ, Armstrong DJ (1986) Regulation of cytokinin oxidase activity in callus
tissues of Phaseolus vulgaris L. cv Great Northern. Plant Physiol 80: 493-499

Chaudhury AM, Letham S, Craig S, Dennis ES (1993) ampI-a mutant with high
cytokinin levels and altered embryonic pattern, faster vegetative growth, constitutive
photomorphogenesis and precocious flowering. Plant J 4: 907-916

Cooper JB, Long SR (1994) Morphogenetic rescue of Rhizobium meliloti nodulaiton
mutants by trans-zeatin secretion. Plant Cell 6: 215-225

Crowell DN (1994) Cytokinin regulation of a soybean pollen allergen gene. Plant Mol
Biol 25: 829-835

Crowell DN, Kadlecek AT, John MC, Amasino RM (1990) Cytokinin-induced
mRNAs in cultured soybean cells. Proc Natl Acad Sci USA 87: 8815-8819

Davies PJ (1988) The plant hormones: Their nature, occurrence, and functions. In: P]
Davies,ed, Plant Hormones and Their Role in Plant Growth and Development, Kluwer
Academic Publishers, Dordrecht, The Netherlands pp 1-12

Dehio C (1989) Isolierung und charakterisierung des knollchenspezifisch experimierten
gens ENOD2 aus der tropischen Leguminose Sesbania rostrata. PhD thesis. Max-
Planck-Institut fur Zuchtungsforschung Abteilung genetische Grundlagen der
Pflanzenzuchtung, Koln, FRG

Dehio C, l
is induc
Deikman
in Arab

Deikman,
abbtel".

DeRoche
differe

Dickey l
ferredt

Dilwortl
Agros

Domino
Cytol'
cells 1

Edward

super
repre

Evans 1
Scon
Pym

Flores
invo

Fritz (
rch
Sci ‘

Gill P
Cha

tool

GTEen
106

Heinl
are

Hirsc
Pia

HObh

79

Dehio C, de Bruiin FJ (1992) The early nodulin gene SrEnod2 from Sesbania rostrata
is inducible by cytokinin. Plant J 2: 117-128

Deikman J, Hammer PE (1995) Induction of anthocyanin accumulation by cytokinins
in Arabidopsis thaliana. Plant Physiol 108: 47-57

Deikman J, Ulrich M (1995) A novel cytokinin-resistant mutant of Arabidopsis with
abbreviated shoot development. Planta 195: 440—449

DeRocher EJ, and Bohnert HJ (1993) Development and environmental stress employ
different mechanisms in the expression of a plant gene family. Plant Cell 5: 1611-1625

Dickey LF, Nguyen T-T, Allen GC, Thompson WF (1994) Light modulation of
ferredoxin mRNA abundance requires an open reading frame. Plant Cell 6: 1171-1176

Dilworth MF, Kende H (1974) Comparative studies on nitrate reductase in
Agrostemma githago induced by nitrate and benzyladenine. Plant Physiol 54: 821-825

Dominov JA, Stenzler L, Lee S, Schwarz JJ, Leisner S, Howell SH (1992)
Cytokinins and auxins control the expression of a gene in Nicotiana plumbaginifolia
cells by feedback regulation. Plant Cell 4: 451-461

Edwards DR, Mahadevan LC (1992) Protein synthesis inhibitors differentially
superinduce c-fos and c-jun by three distinct mechanisms: lack of evidence for labile
repressors. EMBO J 11: 2415-2424

Evans ML (1984) Functions of hormones at the cellular level of organization. In: TK
Scott,ed, Hormonal Regulation of Development 11, vol 10, Encyclopedia of Plant
Pysiology, Springer-Verlag, Berlin, pp 23-79

Flores S, Tobin EM (1988) Cytokinin modulation of LHCP mRNA levels: the
involvement of post-transcriptional regulation. Plant Mol Biol 11: 409-415

Fritz CC, Herget T, Wolter FP, Schell J (1991) Reduced steady-state levels of
rch mRNA in plants kept in the dark are due to differential degradation. Proc Natl Acad
Sci USA 88: 4458-4462

Gill P, Liu Y, Orbovic V, Verkamp E, Poff KL, Green PJ (1994)
Characterization of the auxin-inducible SAUR-ACl gene for use as a molecular genetic
tool in Arabidopsis. Plant Physiol 104: 777-784

Green PJ (1993) Control of mRNA stability in higher plants. Plant Physiology 102:
1065-1070

Heinhorst S, Cannon G, Weissbach A (1985) Plastid and nuclear DNA synthesis
are not coupled in suspension cells of Nicotiana tabacum . Plant Mol Biol 4: 3-12

Hirsch MA, Fang Y (1994) Plant hormones and nodulation: what’s the connection?
Plant Mol Biol 26: 5-9

Hobbie L, Timpte C, Estelle M (1994) Molecular genetics of auxin and cytokinin.
Plant Mol Biol 26: 1499-1519

Hunter T (1995) Protein kinases and phosphatases: The yin and yang of protein
phosphorylation and signaling. Cell 80: 225-236

80

Jackson RJ (1993) Cytoplasmic regulation of mRNA function: The importance of the 3’
untranslated region. Cell 74: 9—14

Leyser HMO, Lincoln CA, Timpte C, Lammer D, Turner J, Estelle M (1993)
Arabidopsis auxin-resistance gene AXRl encodes a protein related to ubiquitin-
activating enzyme El. Nature 364: 161-164

Lips SH, Roth-Bejerano N (1969) Light and hormones: interchangeability in the
induction of nitrate reductase. Science 166: 109-110

Lu J-L, Ertl JR, Chen C-M (1990) Cytokinin enhancement of the light induction of
nitrate reductase transcript levels in etiolated barley leaves. Plant Mol Biol 14: 585-594

Memelink J, Hoge JHC, Schilperoort RA (1987) Cytokinin stress changes the
developmental regulation of several defence-related genes in tobacco. EMBO J 6: 3579-
3583

Mita S, Suzuki-fujii K, Nakamura K (1995) Sugar-inducible expression of a gene
for beta-amylase in Arabidopsis thaliana. Plant Physiol 107: 895-904

Moffatt B, Pethe C, Laloue M (1991) Metabolism of benzyladenine is impaired in a
mutant of Arabidopsis thaliana lacking adenine phosphoribosyltransferase activity.

Peters JL, Silverthome J (1995) Organ-specific stability of two Lemna rch mRNAs
is determined primarily rn the nuclear compartment. Plant Cell 7: 131- 140

Pilgrim ML, Caspar T, Quail PH, McClung CR (1993) Circadian and light-
regulated expression of nitrate reductase in Arabidopsis. Plant Mol Biol 23: 349-364

Raz V, Fluhr R (1993) Ethylene signal is transduced via protein phosphorylation events
in plants. Plant Cell 5: 523-530

Sano H, Youssefian S (1994) Light and nutritional regulation of transcripts encoding a
wheat protein kinase homolog is mediated by cytokinins. Proc Natl Acad Sci USA 91:
2582-2586

Satoru M, Suzuki-Fujii K, Nakamura K (1995) Sugar-inducible expression of a
gene for B-amylase in Arabidopsis thaliana. Plant Physiol 107: 895-904

Seeley KA, Byrne DH, Colbert JT (1992) Red light-independent instability of oat
phytochrome mRNA in vivio. Plant Cell 4: 29-38

Shah DM, Hightower RC, Meagher RB (1982) Complete nucleotide sequence of a
soybean actin gene. Proc Natl Acad Sci USA 79: 1022-1026

Skoog F, Miller C (1957) Chemical regulation of growth and organ formation in plant
tissues cultured in vitro. Soc Expt Biol Symp 11: 188-231

Soni R, Carmichael JP, Shah ZH, Murray JAH (1995) A family of cyclin D
homologs from plants differentially controlled by growth regulators and containing the
conserved retinoblastoma protein interaction motif. Plant Cell 7: 85-103

81

Sullivan ML, Green JP (1993) Post-transcriptional regulation of nuclear-encoded
genes in higher plants: the roles of mRNA stability and translation. Plant Mol Biol 23:
109 1- 1 104

Suzuki 1, Cretin C, Omata T, and Sugiyama T (1994) Transcriptional and
posttranscriptional regulation of nitrogen-responding expression of
phosphoenolpyruvate carboxylase gene in maize. Plant Physiol 105: 1223-1229

Tamaoki T (1991) Use and specificity of stauroporine, UCN-Ol and calphostin C as
protein kinase inhibitors. In T Hunter, BW Sefton, eds,Methods in Enzymology, Vol.
201B, NY Academic Press, pp 340-347

Tanzer MM, and Meagher BR (1994) Faithful degradation of soybean rch mRNA in
vitro. Mol Cell Biol 14: 2640-2650

Treharne KJ, Stoddart J L, Pughe J, Paraniothy K, and Wareing PF (1970)
Effects of gibberellin and cytokinins on the activity of photosynthetic enzymes and
plastid ribosomal RNA synthesis in Phaseolus vulgaris L. Nature 228: 129-131

van de Wiel C, Scheres B, Franssen H, van Lierop MJ, van Lammeren A,
van Kammen A, and Bisseling T (1990) The early nodulin transcript ENOD2 is
located in the nodule parenchyma (inner cortex) of pea and soybean root nodules.
EMBO J 9: 1-7

Verwoerd TC, Dekker BMM, and Hoekema A (1989) A small-scale procedure for
the rapid isolation of plant RNAs. Nuc Ac Res 17: 2362

Zhang S, Sheng J, Liu Y, Mehdy MC (1993) Fungal elicitor-induced bean proline-
lich protein mRNA down-regulation is due to destabilization that is transcription and
translation dependent. Plant Cell 5: 1089-1099

Chapter 3

REGULATION OF LJENODZ GENE EXPRESSION BY ETHYLENE

82

83

ABSTRACT

The early nodulin gene Enod2 encodes a Pro-rich protein which is expressed in the
nodule parenchyma. Expression of most Enod2 genes is restricted to the nodule.
However, expression of the SrEnod2 gene from Sesbania rostrata is enhanced in roots
upon application of cytokinin. Here, we present evidence indicating that the Lotus
japonicus LjEnodZ gene is not regulated by cytokinin, but by ethylene. Ethylene gas or
submergence of L japonicus roots in water caused an enhancement of LjEnodZ mRN A
accumulation. The accumulation in submerged roots was partially inhibited by AVG, and
strongly inhibited by cobalt chloride. In addition, a comparison was made between the
effect of ethylene on SrEnod2 and LjEnodZ gene expression.

84

INTRODUCTION

The early nodulin gene SrEnod2 from the tropical legume Sesbania rostrata encodes
a Pro-rich protein which is expressed in a cell-specific manner in the nodule parenchyma.
The nodule parenchyma is composed of cytoplasmically dense cells which surround the
cells harboring the nitrogen-fixing bacteroids (Van de Wiel et al., 1990). The function of
the SrEnod2 gene remains to be determined. However, it has been proposed to play a role
in establishing an oxygen diffusion barrier (Tjepkema and Yocum, 1974; Witty et al.,
1986; Van de Wiel et al., 1990). The SrEnod2 gene is regulated both in a tissue-specific
manner and by the plant hormone cytokinin (Dehio and de Bruijn, 1992; Chen et al.,
1996). SrEnod2 tissue-specific expression in the nodule parenchyma has been recently
shown to require sequences located in the 3’UT'R of the SrEnod2 gene, and not in its
promoter (Chen et al., 1996).

The expression of the SrEnod2 has been shown to be specifically induced in roots
in a time- and concentration-dependent manner by the application of cytokinin, in the
absence of inoculation by the microsymbiont Azorhizobium caulinodans (Dehio and de
Bruijn 1992). The mechanism underlying this induction appeals to involve
posttranscriptional events (Silver et al. 1996). The tissue-specific expression of Enod2
genes from several legumes is conserved. However, the phenomenon of cytokinin
regulation is not. The only two legumes known in which Enod2 gene expression in roots
is induced by cytokinin are Sesbania rostrata and alfalfa (Hirsch and Fang, 1994), although
the alfalfa Enod2 message which accumulates in roots upon cytokinin treatment is of a
different molecular weight than is the Enod2 message from nodules of the same plant (A.
Hirsch, personal communication).

The effects of cytokinin on plant development have been shown in some instances
to mediated through the action of ethylene. For example, Bertell and Eliasson (1992)

demonstrated in pea roots that the application of cytokinin inhibits root elongation and

85

formation of lateral roots and stimulates swelling of the root tips. They further
demonstrated that BAP caused an up to a 4-fold increase in ethylene levels in roots. These
effects were found to be counteracted by the application of cobalt ions to inhibit ethylene
production. Although there is no direct evidence that cytokinin induces the synthesis of
ACC synthase mRNA or ACC activity, it is known that cytokinins act to increase auxin
levels, which, in turn, can induce ACC mRNA accumulation (Yu et al., 1979; Yoshii and
Irnaseki, 1981; Bertell and Eliasson, 1992). Su et al. (1995) genetically tested the
hypothesis that cytokinin action is coupled to ethylene action (Lieberman, 1979). They
demonstrated that the inhibitory effects of BA on root and hypocotyl elongation were
partially blocked in the ethylene-resistant mutants einI-I and einZ-I , as well as by
inhibitors of ethylene action (Cary et al., 1995). Furthermore, the finding that cytokinin
and ethylene responses are coupled was reinforced by the demonstration that the cytokinin-
resistant mutant ckrI (Su and Howell, 1992) is allelic to ein2 (Cary et al., 1995). Reports
of the effects of ethylene on the process of nodule development are limited (Lee and LaRue,
1992a, 1992b). In general, ethylene appears to be inhibitory to early stages of nodule
development (Lee and LaRue 1992b).

In a search for other legumes in which the Enod2 gene is inducible by cytokinin,
and to determine if ethylene is indeed involved in cytokinin-mediated regulation, I initiated
a study of the Enod2 gene of the model legume Lotus japonicus. In this report I show that
expression of the LjEnodZ gene is not regulated by cytokinin, but, surprisingly, is
regulated by ethylene. In addition, I present data comparing the effects of ethylene on the
expession of the cytokinin-induced SrEnod2 gene to the ethylene-induced LjEnodZ gene.

To our knowledge this is the first report of an ethylene-inducible nodulin gene.

100

Eth:

86

METHODS
Plant material and growth conditions

Lotus japonicus GIFU seeds were germinated on B5 media (Sigma)/ 0.8% agar for
one week and then grown for two weeks in pots containing a sterilized vermiculite mix
(fine vermiculite 2: coarse vermiculite 3: sand 1) in a growth chamber. The growth chamber
conditions were 18 h light/ 6 h dark; 23° c day/ 18° c night. Plants in pots were watered
with Hoaglands media once a week. S. rostrata seeds were germinated on wet gel blot
paper (Schleicher and Schuell, Keene, NH) and seedlings grown in soil composed of
Metromix (Hummert International, Earth City, MO) and sand (2:1) at 30°C, with a 18 h
light, 28°C/6 h dark, 22°C regime for two weeks in growth chambers with 75 % RH. For
all chemical treatments, plants were washed free of soil and incubated in water containing 1
11M Benzylarninopurine (BAP) [Sigma], 0.5 mM CoCl2 (Sigma), and/or100 uM
Aminoethoxyvinyl-glycine (AVG) [Sigma]. Ethylene treatments were performed as
follows: Seedlings of L japonicus plants in pots were placed in desiccating jars injected
with 40ppm ethylene gas. S. rostrata seedlings were washed free of soil, placed upright in
a glass beaker lined with gel blot paper (Schleicher and Schuell, Keene, NH) soaked in
water or water containing 1 uM BAP. Seedlings in beakers were placed in desiccating jars
containing 40 nIJL ethylene gas. Control seedlings of L. japonicus and S. rostrata were
treated in the same manner, and placed in desiccating jars containing solid KMnO4 (Sigma)
to absorb endogenously produced ethylene from plants. After all chemical treatments,

roots were isolated and frozen in liquid nitrogen for RNA isolation.

Ethylene gas measurements

87

Root tissue was isolated from seedlings, and individual roots were placed in 10 ml
glass tubes sealed with rubber stoppers. The roots were incubated for 1 h, then lml of gas
volume was removed and injected into a gas chromatograph (V arian 3400 gas
chromatograph, Sugar Land, TX). Measurements were expressed as nL of ethylene
evolved per hour per gram of fresh weight of tissue.

Northern blot analysis.

10 ug of RNA was electrophoresed in 1.2% (w/v) agarose gels in Mops buffer
(20mM Mops, 1.0 mM EDTA, 5.0 mM sodium acetate, pH 7.0) containing 5.4% (v/v)
formaldehyde. Gels were blotted onto 0.22 W NitroPlus nitrocellulose membrane
(Micron Separation, Inc., Westborough, MA). Membranes were probed with a [32P]
dATP labeled DNA fragment containing the SrEnod2 coding region generated by random
priming (Boehringer Mannheim). Filters were reprobed with a Sesbania rostrata ubiquitin
coding gene (ubi-I) DNA probe as a loading control. All filters were washed at 65°C in 2.0
x SSC, 0.1% SDS for 20 min; 0.5 x SSC, 0.1% SDS for 20 min; and 0.1 x SSC, 0.1%
SDS for 15 min. The signals were quantified using phosphorimager analysis (400B model
Phosphorlrnager, Molecular Dynamics, Inc., Sunyvale, Ca.).

88

RESULTS

Water treatment induces LjEnodZ mRNA accumulation in L. japonicus roots

via ethylene production

Twenty-one-day-old L. japonicus plants were removed from the soil and incubated
in water for 4 and 8 h in the presence of 1 uM BAP; control plants were incubated in water
without BAP for 8 h, LjEnodZ mRN A from roots of plants treated with BAP accumulated
to the same levels as in the control plants (Figure 3.1). In order to determine if incubation
in water alone was the cause of LjEnodZ mRN A accumulation, 21-day-old L. japonicus
plants grown in soil were incubated for 4 and 8 h in water, and RNA was isolated from
their roots. Control plants were not incubated in water, and RNA was isolated from roots
immediately after removal from soil. As shown in Figure 3.2, over an 8 h period, LjEnodZ
mRN A accumulated to a level nearly 17-fold higher than in the control plants. Therefore,
accumulation of LjEnodZ mRN A appears to be the result of submergence of roots in water,
rather than to cytokinin action, as seen in S. rostrata (Dehio and de Bruijn, 1992).

Next I sought to determine whether ethylene production during flooding (Bradford
and Dilley, 1978; Bradford and Yang, 1980) could be responsible for the enhanced
accumulation of LjEnodZ mRNA. To test this hypothesis, roots were treated with
arninoethoxyvinyl glycine (AVG) to inhibit the activity of ACC synthase, the rate limiting
enzyme in ethylene biosynthesis. In addition, cobalt chloride was used to inhibit the
activity of ACC oxidase. Plants were incubated in water containing 100 MM AVG for 4
and 8 h. Separate plants were incubated in water containing 0.5 mM cobalt chloride. The
concentrations of inhibitors correspond to those reported in the literature (Yu and Yang,
1979). Control plants were incubated in water without inhibitors for 4 and 8 b. As shown

in Figure 3.2, AVG caused a greater than 2-fold reduction in LjEnodZ mRNA

89

Figure 3.1. LjEnodZ mRNA levels in roots treated with cytokinin. Northern
blot analysis showing LjEnod2 mRNA accumulation after the indicated treatments. 21-day-
old L japonicus plants were treated with 1 11M BAP for 4 and 8 h by submergence of

roots in water. Roots of intact control plants were incubated in water for 8 h without BAP.

The blot was reprobed with the ubi-I gene as a loading control.

 

90

-BAP
+BAP

 

Time (h) 8 4 8

_ a u LjEnodZ

ubi-I

91

Figure 3.2. LjEnodZ mRNA levels in roots treated with H20, AVG and
C00,. 100 M AVG and 0.5 mM CoCl2 for the times indicated, then total RNA was
isolated from roots. Treatment with water alone was carried out as a control. A, Northern
blot analysis showing LjEnodZ mRNA accumulation after the indicated treatments. The
blot was reprobed with the ubi-I gene as a loading control. B, Quantification of the

northern blot shown in A using phosphorimager analysis. Values are reported as the ratio

of LjEnodZ/ ubi-I signals.

 

 

 

 

 

 

 

 

 

 

 

 

 

A
N
o <3 ""
N > L5
I _<5_ _9_
,, Tlme(h) 0 4 8 4 8 8
'35 — am- LjEn0d2
than
The
-- In Ml .- mi ubi-I
trifle ’ "
B
0.18
78; 0.16 .. E] H20
5 014 AVG
<1 . - COC12
% 0.12 ~
% 0.1
3
g 0.08l-
d)
.5 0.06 ~-
.53
a; 0.04 ~-
0.02--
0 ['_I -
0 4 8 4 8 8

Time (h)

93

accumulation in response to flooding. This remaining level of LjEnodZ mRN A
accumulation at 8 h in the presence of AVG was probably due to the fact that AVG did not
completely abolish the synthesis of ethylene, as verified by gas chromatography (Figure
3.3). In contrast, cobalt chloride drasticly reduced the accumulation of LjEnodZ mRNA to
levels comparable to the 0-h control (Figure 3.2).

To determine whether ethylene could induce the accumulation of LjEnodZ mRNA
directly, L japonicus plants grown in soil were treated with 40 1.1le ethylene gas for 4 and
8 h, respectively. Control plants were treated under similar conditions for 8 h without
ethylene. As shown in Figure 3.4 LjEnodZ mRN A was found to accumulate in ethylene
treated plants to a level 4-fold higher than that in control plants. The level of accumulation
was found to be lower than that seen in plants removed from soil and incubated in water

(Figure 3.2).
Ethylene effects on S. rostrata SrEnod2 gene expression

To compare the effects of ethylene on SrEnod2 to LjEnodZ gene expression, two-
week-old S. rostrata seedlings were removed from the soil and treated for 8 h with 40 11le
ethylene gas with or without the addition of 1 uM BAP. The controls consisted of S.
rostrata seedlings treated on wetted bloting paper (air control) or with their roots submerged
in water. Treatment with ethylene in the absence of BAP resulted in a 2-fold increase of
SrEnod2 mRNA levels, as compared to the air control (Figure 3.5). Seedlings which were
submerged in water with or without BAP, accumulated overall lower levels of SrEnod2
mRN A when compared to the air controls. Treatment of two-week-old S. rostrata plants
with 100 [AM AVG showed no significant effect on SrEnod2 mRN A accumulation (Figure
3.6), although ACC synthase activity was nearly abolished, as indicated by ethylene gas
measurements (Figure 3.3). However, treatment of S. rostrata plants with 0.5 mM cobalt
chloride significantly reduced SrEnod2 mRN A levels to below control levels, which is

likely an effect by the non-specific action of cobalt ion.

94

Figure 3.3. Ethylene levels in S. rostrata, and L. japonicus roots treated
with AVG. Plants were incubated with or without 100 11M AVG for the times indicated.
Roots were isolated, and ethylene production was determined by gas chromatagraphy.
Ethylene levels are reported as the average ethylene produced (nL ethylene per h per g fresh

weight) from three independent root systems. The error bars represent the SD.

nL ethyle
O N k 0\ 00

95

 

 

 

- L. japonicus

S. rostrata

 

 

 

96

Figure 3.4. LjEnodZ mRNA levels in roots from seedlings treated with
ethylene. 21-day-old L. japom'cus plants were treated with 40 1'1le ethylene gas for the
times indicated. Control (C) seedlings were incubated under similar conditions for 8 h
without ethylene gas. A, Northern blot analysis showing LjEnodZ mRNA accumulation
after the indicated treatments. The blot was reprobed with the ubi—I gene as a loading
control. B, Quantification of northern blot shown in A using phosphorimager analysis.
Values are reported as the fold increase in the ratio of LjEnodZ/ ubi-I signals over the

control level, which has been set at zero.

97

4 8

C

Time (h)

at... ”I ] LjEnodZ

.-

- ubi-I

 

 

6.
l

 

by
-

421854.20
1111 0000

292 .4sz meeamseésem

98

Figure 3.5. SrEnod2 mRNA levels in roots of seedlings treated with H20 ,
BAP, ethylene and C00,. Two-week-old S. rostrata seedlings were incubated for 8 h
with or without 1 11M BAP in the presence of 40 nIJL ethylene gas, and/or 0.5 mM CoClz.
Air and water treatments for 8 h served as controls. A, Northern blot analysis showing
SrEnod2 mRNA accumulation after the indicated treatments. The blot was reprobed with
the ubi-I gene as a loading control. B, Quantification of the northern blot shown in A

using phosphorimager analysis. Values are reported as the ratio of SrEnodZ/ ubi-I signals.

99

~68

of

25735

22.

BAP

SrEnod2

,A

“.8.“-

ubfl]

 

 

- BAP
I +BAP

 

 

5.
2

2 5. l 5
1

<ZmE magmam mo £32 ozfiom

ET H20 CoC12

Air

100

Figure 3.6. SrEnod2 mRNA levels in roots treated with AVG. Two-week-old
S. rostrata seedlings were incubated with or without 100 pM AVG for the times indicated,
then total RNA was isolated from roots. A, Northern blot analysis showing SrEnod2
mRNA accumulation after the indicated treatments. The blot was reprobed with the ubi-I
gene as a loading control. B, Quantification of the northern blot shown in A using

phosphorimager analysis. Values are reported as the ratio of SrEnod2! ubi-I signals.

1.8

Relative SrEnod2 mRNA levels

101

 

 

. - "w SrEnod2

 

1.6 .
1.4 ~
1.2 .,

0.8 “
0.6 “
0.4 1'
0.2 *’

 

8h 4h

 

 

’fi‘i‘fff - BAP ..
i::;::i:é l + BAP

............

 

 

4m

8h 4h 8h 4h

 

-AVG

 

8h

 

+AVG

 

102

DISCUSSION

The results presented here demonstrate that expression of the LjEnodZ gene is, at
least in part, regulated by ethylene. However, the level of increase in LjEnodZ mRNA by
ethylene treatment was lower compared to levels observed after submergence in water.

One explanation for this difference may be that the higher increase in ethylene production in
the roots may be caused by stress from removal from the pots (wounding) and flooding of
the roots. Another possible explanation is that the roots beneath the soil do not encounter
high enough levels of ethylene gas due to low diffusion of ethylene gas through soil. Still
another explanation may be that some other factor related to water stress is causing a higher
accumulation of LjEnodZ mRN A. Nonetheless, these results show that ethylene gas can
indeed cause the accumulation of LjEnodZ mRN A. This raises important questions
regarding the involvement of ethylene in nodulation, since a role for ethylene as a positive
regulator of nodan gene expression has never been reported. In fact, all reports thus far
indicate an inhibitory role of ethylene in nodulation of pea, but not of other legumes (Feam
and LaRue, 1991; Lee and LaRue, 1992; Guinel and LaRue, 1992).

The tissue-specific expression pattern of the LjEnodZ gene is similar to that seen for
the SrEnod2 gene (D. Silver and F. J. de Bruijn, unpublished data), as well as for other
Enod2 genes (Hirsch et al., 1989; van de Wiel et al., 1990). However, the response of
Enod2 genes from different legumes to hormones is very different. The expression of
most Enod2 genes is not regulated by cytokinin, but only induced during the process of
nodulation, while expression of the SrEnod2 and the LjEnodZ genes appears to be
mediated by two different plant growth regulators. Why is the expression of the LjEnodZ
gene inducible by ethylene, and the SrEnod2 gene by cytokinin? The difference in habitat
of these plants may present an explanation. S. rostrata naturally grows in water-logged
soil, conditions which can induce ethylene production, while L. japonicus grows in semi-

dry soil. Therefore, S. rostrata may have adopted the cytokinin signal transduction

103

pathway to regulate its Enod2 gene, since it may have a high tolerance for ethylene, or a
decreased sensitivity to ethylene.

Feam and LaRue (1991) have shown that the low level of nodulation by the sym 5
mutant of pea can be restored to nearly normal levels by treatment of the plants with Co”,
AVG and Ag’. However, the plants’ ACC and ethylene levels in sym 5 were not found to
be significantly different from wild-type plants. Ethylene inhibitors were also found to
partially restore nodulation by another nodulation mutant of pea, E 107 (Guinel and LaRue,
1992). Moreover, application of ethylene to wild-type pea plants inhibits nodulation on the
primary and lateral roots, and this inhibition is relieved by application of Ag+ (Lee and
LaRue, 1992). On close examination, it was determined that ethylene may inhibit very
early stages of nodulation, particularly the advancement of infection threads through the
root cell epidermis and outer cortex (Lee and LaRue, 1992). However, it is not known if
ethylene plays an inhibitory role in other legume plants. Recently, it was found by
Heidstra et al. that ACC oxidase is expressed in cells opposite protophloem poles in pea
roots (S. Peck, personal communication). They postulate (S. Peck, personal
communication) that during the early events of nodulation, in which cortical cells opposite
protoxylem poles begin to divide, expression of ACC oxidase in adjacent cells opposite
protophloem cells will set up an ethylene gradient. In turn, the ethylene gradient will reflect
an inhibition gradient for the formation of nodule primordia. Nodule primordia which
reach a certain stage of development may then become insensitive to ethylene. In order to
better understand the roles ethylene may play in nodulation of pea and in other legumes, it
will be important to determine the expression pattern of ACC oxidase during nodulation of

different species of legumes, and to identify other ethylene-regulated nodulin genes.

104
REFERENCES

Bertell G, Eliasson L (1992) Cytokinin effects on root growth and possible
interactions with ethylene and indole-3-acetic acid. Physiol Plant 84: 255-261

Bradford KJ, Dilley DR (1978) Effects of root anaerobiosis on ethylene production,
epinasty, and growth of tomato plants. Plant Physiol 61: 506-509

Bradford KJ, Yang SF (1980) Xylem transport of l-aminocyclopmpane-l-carboxylic
acid, an ethylene precursor, in waterlogged tomato plants. Plant Physiol 65: 322-326

Cary AJ, Wennuan L, Howell SH (1995) Cytokinin action is coupled to ethylene in
its effects on the inhibition of root and hypocotyl elongation in Arabidopsis thaliana
seedlings. Plant Physiol. 107: 1075-1082

Chen R, Silver DL, de Bruijn FJ (1996) The nodule parenchyma-specific expression
of the Sesbania rostrata early noduhn gene SrEnod2 is mediated by its 3’ untranslated
region (3’UTR). Plant Cell; submitted

Dehio C, de Bruijn F J (1992) The early nodulin gene SrEnod2 from Sesbania rostrata
is inducible by cytokinin. Plant J. 2: 117-128

Feam, J C, LaRue, TA (1991) Ethylene inhibitors restore nodulation to sym5 mutants
of Pisum sativum L. cv Sparkle. Plant Physiol 96: 239-244

Guinel, FC, LaRue, TA (1992) Ethylene inhibitors partly restore nodulation to pea
mutant E107 (brz). Plant Physiol 99: 515-518

Hirsch, AM, Bhuvaneswari, TV, Torrey, JG, Bisseling, T (1989) Early
nodulin genes are induced in alfalfa root outgrowths elicited by auxin transport
inhibitors. Proc Natl Acad Sci USA 86: 1244-1248.

Hirsch AM, Fang Y (1994) Plant hormones and nodulation: what’s the connecton?
Plant Mol Biol 26: 5-9

Lee KH, LaRue TA (1992a) Ethylene as a possible mediator of light- and nitrate-
induced inhibition of nodulation of Pisum sativum L. cv Sparkle. Plant Physiol 100:
1 334-1 338

Lee KH, LaRue TA (1992b) Exogenous ethylene inhibits nodulation of Pisum sativum
L. cv Sparkle. Plant Physiol 100: 1759-1763

Lieberman M.(1979) Biosynthesis and action of ethylene. Annu Rev Plant
Physiol 30: 533-591

Su W, Howell SH (1992) A single genetic locus, ckrI, defines Arabidopsis mutants in
which root growth is resistant to low concentrations of cytokinins. Plant Physiol 99:

1569-1574

Silver DL, Pinaev A, Chen R, de Bruijn FJ (1996). Posttranscriptional regulation
of the Sesbania rostrata early nodulin gene SrEnod2 by cytokinin. Plant Physiol 112:
559-567 .

105

Tjepkema JD, Yocum CS (1974) Measurement of oxygen partial pressure within
soybean nodules by oxygen mircoelectrodes. Planta 119: 351-360

Van de Wiel C, Scheres B, Franssen H, van Lierop MJ, van Lammeren A,
van Kammen A, Bisseling T (1990) The early nodulin transcript ENOD2 is located
in the nodule parenchyma (inner cortex) of pea and soybean root nodules. EMBO J 9: 1-
7

Witty JF, Minchin FR, Skot L, Sheehy JE (1986) Nitrogen fixation and oxygen in
legume nodules. Oxford Surv. Plant Mol Cell Biol 3: 275-314

Yoshii H, Imaseki H (1981) Biosynthesis of auxin-induced ethylene. Effects of indole-
3-acetic acid, benzyladenine and abscisic acid on endogenous levels of 1-
arninocyclopropane-l-carboxylic acid (ACC) and ACC synthase. Plant Cell Physiol 22:

369-379

Yu Y-B, Yang SF (1979) Auxin-induced ethylene production and its inhibition by
arninoethoxyvinylglycine and cobalt ion. Plant Physiol 64: 1074-1077

Chapter 4

THE SRENODZ GENE IS CONTROLLED BY A CONSERVED
CYTOKININ SIGNAL TRANSDUCTION PATHWAY

The contents of this chapter have been submitted for publication in Plant Physiol, 1996 (Silver, Deikman,
and de Bnrijn)

106

107

ABSTRACT

Here we report the use of Arabidopsis thaliana to study the mechanism of
cytokinin-mediated mRN A accumulation from the Sesbania rostrata early nodan gene
SrEnod2. A chimeric gus gene, comprised of 1.9 kb of the SrEnod2 5' promoter region
fused to the gus reporter gene and terminated by 2.5 kb of the SrEnod2 3' downstream
region, was found to be expressed in roots of transgenic Arabidopsis plants in an identical
tissue-specific pattern as in roots of the legume Lotus japonicus. Moreover, the expression
of this construct was found to be specifically enhanced by cytokinin, but repressed by
ethylene, in roots of transgenic Arabidopsis plants. Auxin, GA, or ABA, did not affect
the expression of the chimeric SrEnod2 5 ’-gus-SrEnod2 3 ’ construct. The SrEnod2 3'
downstream region was found to be essential for both root-specific and cytokinin
enhancement of expression in transgenic Arabidopsis plants. Cytokinin enhancement of
SrEnod2 5 ’-gus-SrEnod2 3 ’ gene expression was found to be abolished in the Arabidopsis
cytokinin-resistant mutant cyrI, but was regulated normally in the auxin resistant mutant,
axr2. These results indicate the existence of a conserved cytokinin signal transduction

pathway between legumes and Arabidopsis, and open the way for its genetic dissection.

108

INTRODUCTION

Since the discovery of cytokinins (Miller et al., 1955), a considerable amount of
research has been dedicated to understanding the physiological, developmental and
molecular mechanisms of cytokinin action in plants. Cytokinins appear to play a crucial
role in a large number of developmental processes, including regulation of cell division,
nutrient mobilization, chloroplast development, and apical dominance (Binns, 1994). The
complexity of cytokinin effects has hindered the elucidation of the mechanism of cytokinin
action. Many genes have been described which are regulated by cytokinin (Chen et al.,
1993; Hobbie et al., 1994), but the majority of these genes are also co-regulated by other
factors, such as other hormones, light, nutrient status, and environmental stresses (Chen et
a1, 1993). One exception may be the SrEnod2 gene from the tropical legume Sesbania
rostrata. This gene was shown to be specifically regulated in response to physiologically
relevant concentrations of cytokinin, without requiring other hormones, physical or
chemical stresses as co-regulating factors (Dehio and de Bruijn, 1992). The steady state
level of SrEnod2 mRN A is increased within 2 hours of treatment of non-nodulated roots
with cytokinin, and this relatively short response time suggests that the mRNA increase
could be a primary effect of cytokinin treatment (Dehio and de Bruijn, 1992). Study of the
regulation of this gene by cytokinins may reveal general mechanisms of cytokinin action.

The SrEnod2 gene codes for a proline-rich protein, which is expressed during
nodule formation, specifically in the nodule parenchyma (Van de Wiel et al., 1990), and is
highly conserved among legume plants (Franssen et al., 1987; Dickstein et al., 1988;
Govers et al., 1990; Van de Wiel et al., 1990; Szczyglowski and Legocki, 1990; Dehio and
de Bruijn, 1992). The nodule parenchyma is typically comprised of a few cytoplasmically
dense cell layers that surround the central cells of the nodule, which harbor the nitrogen
fixing bacteroids (Van de Wiel et al., 1990). Based on the location of expression of the
Enod2 gene in pea and soybean, it has been proposed that the Enod2 protein may function

109

as part of an oxygen diffusion barrier (T jepkema and Yocum, 1974; Witty et al., 1986; Van
de Wiel et al., 1990). Such an oxygen diffusion barrier might protect the bacterial nitrogen
fixing enzyme nitrogenase, which is highly oxygen sensitive, in the infected core cells of
the nodule. However, the function of the Enod2 protein still remains to be determined.

The results of nuclear run-on assays using nuclei from cytokinin-treated roots of S.
rostrata suggested that SrEnod2 mRN A accumulation in response to cytokinin is due to
posttranscriptional regulation (Silver et al., 1996). The accumulation of SrEnod2 mRNA
appears to occur primarily in the cytoplasm, and is inhibited by the translational inhibitor
cycloheximide, inhibitors of protein phosphatases, and kinases (Silver et al., 1996). In
addition, SrEnod2 mRN A levels decay rapidly upon treatment with cycloheximide,
suggesting that ongoing protein synthesis is required for the accumulation, as well as
maintenance, of SrEnod2 mRN A levels (Silver et al., 1996). In order to better understand
the mechanism of this regulatory process, it is necessary to delimit the sequence elements
of the SrEnod2 locus which are required for the observed cytokinin response and to isolate
proteins involved in the regulation of cytokinin-modulated mRN A stability.

In this report, I have developed a model system using Arabidopsis thaliana to study
the mechanism of SrEnod2 mRN A accumulation mediated by cytokinin. I chose
Arabidopsis, in part, because of the existence of mutants affected in hormone response that
would be useful for investigating the regulation of this gene, in particular, the cytokinin
resistant I (cyrI ) mutant, which appears to be specifically resistant to cytokinin based on a
root-elongation assay (Deikman and Ulrich, 1995). The cyrI mutation results in a ten-fold
reduction in sensitivity of roots to BAP, but does not affect sensitivity to IAA and ACC.
However, this mutant has increased sensitivity to ABA. The cyrl mutation is pleiotropic,
causing shoot abnormalities, including limited leaf production, and the development of a
single infertile flower. Some aspects of this phenotype are consistent with a defect in

cytokinin response, such as a failure of the cotyledons to expand, reduced accumulation of

110

chlor0phyll, and failure to accumulate anthocyanins in response to cytokinins. However,
the cyrI gene has not been isolated, and the nature of the defect is not precisely known.

Many plant developmental processes require the action of both cytokinin and auxin.
The classical example is the initiation of cell division in tobacco pith cells, which will
expand upon auxin treatment, but only divide in the presence of cytokinin (Miller et al.,
1955, 1956). Root nodule formation has also been shown to involve both auxin and
cytokinin action (Allen et al., 1953; Hirsch et al., 1989; Cooper and Long, 1994).
Therefore, I also used the auxin resistant 2 (aer) mutant in this study. The aer mutation
(Wilson et al., 1990) is a dominant mutation which confers auxin-resistant root growth, as
well as root growth resistance to inhibition by ethylene and ABA. The axr2 mutation is
also pleiotropic in that aer mutants are dwarfs, affected in both shoot and root gravitropic
growth, and lack root hairs.

In this report, I demonstrate that a chimeric gene consisting of the SrEnod2 5’
upstream region, including its 5’ UTR, fused to the gus reporter gene, and terminated by
the SrEnod2 3’ downstream region, is specifically regulated by cytokinin, and is expressed
in roots of transgenic Arabidopsis plants in an identical fashion as in the legumes S.
rostrata and lotus japonicus. In addition, expression was found to be localized to newly
expanding leaves, trichomes of these leaves, and stipules of Arabidopsis. Both tissue-
specific and cytokinin responsive regulation was shown to be dependent on the SrEnod2 3’
downstream region. I generated cyrI mutants that harbored the SrEnod2 5’-gus-SrEnod2
3’ gene, and found that GUS activity was not induced upon cytokinin treatment in these
plants. This data provides genetic evidence for the hypothesis that the SrEnod25’-gus-
SrEnod23’ construct is regulated by a cytokinin signal transduction pathway that is

conserved between legumes and Arabidopsis.

1]]

METHODS

Plant material and growth conditions

Transgenic Lotus japonicus GIFU seeds were germinated on B5 media (Sigma)/
0.8% agar for one week and then grown for two weeks in pots containing a sterilized
vermiculite mix (fine vermiculite 2: coarse vermiculite 3: sand 1) in a growth chamber. The
growth chamber conditions were 18 h light/ 6 h dark; 23°C day/ 18° c night. Plants in
pots were watered with Hoagland’s media once a week. Transgenic Arabidopsis thaliana
Columbia plants as well as the Arabidopsis mutant lines cyrl and aer (obtained from the
Arabidopsis Biological Resource Center at The Ohio State University) were germinated and
grown on MS media (Gibco BRL)/ 0.8% agar for 2 weeks in a growth chamber.
Arabidopsis plants were grown in Arabidopsis Mix (Hummert International, Earth City,
MO) in a growth chamber for approximately one month after which plants were allowed to
dry, and seeds were harvested. Growth chamber conditions were 18 h light/ 6h dark; 210
c day/18° C night.

For cytokinin treatments of transgenic L. japonicus plants, three-week-old plants
were washed free of soil and incubated in distilled water containing 1 11M BAP for 4 h, 8
h, 12 h, and 24 h, respectively. Control plants were incubated only in distilled water for
the same time period. Hormone treatments of transgenic Arabidopsis plants were carried
out as follows: Two-week-old transgenic Arabidopsis plants grown on solid MS plates, as
described above, were transferred to solid MS plates containing 1 11M BAP for the time
period indicated in the Results section. Their roots were isolated and frozen in liquid .

nitrogen. Treatments with 10 11M GA3 (Abbott Labs, Chicago, IL), 10 11M ABA (Sigma),

0.02 11M NAA (Sigma), and 40 uL/L ethylene were performed for 24 h under the same

112

condition as the cytokinin treatments. Ethylene incubations were canied out in a

desiccating jar.

Chimeric gus reporter gene construction

The construction of the SrEnod2 3’ and nos 3’ constructs was carried out as follows: A
5.3 kb EcoRI fragment containing 1.9 kb of SrEnod2 5’ upstream region, 1.0 kb

SrEnod2 coding region, and 2.5 kb SrEnod2 3’ downstream sequence was subcloned from
the genomic clone KCDl (Dehio, 1989) into pBluescript KS- vector (Clonetech, Palo Alto,
CA). BamHI and SacI restriction sites were introduced immediately proceeding the
translational initiation site and after the translational stop codon, respectively, as described
by Kunkel et al. (1985). The oligonucleotide for introducing the BamHI site was
(5’)GAGTAGTGTAGAGA-W CT ATCT AT(3’). The location of the original
translational start codon is indicated in bold. The oligonucleotide for introducing a Sacl site
was (5’)GTAGTGGTA-GTGGTTGWAATITTITITGGQ’). The
translational stop codon TAA is indicated in bold. Both the 1.9 kb EcoRI-BamHI SrEnod2
5’ fragment and the 2.5 kb SacI-EcoRl SrEnod2 3’ fragment were subcloned into the
pBllOl .1 vector (Clonetech) flanking the gus reporter gene. This construct was designated
as SrEnod2 3’. The nos 3’ construct was made by subcloning only the 1.9 kb EcoRI-
BamHI SrEnod2 5’ fragment into pBIlOl.l 5’ in front of the gus reporter gene which is

terminated by the nopaline synthase 3’ terminator (nos 3’).

Plant transformation

The SrEnod2 3’ gus and nos 3’ constructs were introduced into the genome of L.

japonicus GIFU seedlings according to the method of Handberg et al. (1992). A. thaliana

113

Columbia seedlings were transformed with the SrEnod2 3’ gus and nos 3’ constructs using

vacuum infiltration, according to the procedure described by Bent et al. (1994).

RNA isolation and northern blot analysis

Total RNA was isolated from roots of transgenic Arabidopsis plants according to
the method of Verwoerd et al., (1989). Poly(A) RNA was isolated from total RNA by
oligo-dT column chromatography as described by the manufacturer’s instructions (5 prime-
3 prime, Inc., Boulder, Co). RNA gel blot analysis was performed as according to
Maniatis et al. (1982). Approximately 3 ug of poly(A) RNA was used for analysis. The
coding region of the E. coli B-glucuronidase gene (uidA; gus) was used as a DNA probe
for RNA gel analysis. A DNA fragment encompassing part of the eif4 coding region

(Taylor et al., 1993) was used as a loading control on northern blots. [32P]-dATP labelled
gus and eif4 DNA probes were generated by random priming (Boehringer Mannheim,
Indianapolis, IN). Hybridization and washing conditions of RNA gel blots were as
previously described (Silver et al., 1996). gus and eif4 mRN A levels were quantified
using phosphorimager analysis (Model 400B Phosphorlrnager, Molecular Dynamics,
Sunnyvale, CA). Radioactive blots were exposed to autoradiography film (Dupont,

Wilmington, DE) at -75°C.

Quantification of GUS enzymatic activity

Quantification of GUS enzymatic activity was performed according to the procedure
described by Jefferson et al. (1987). The fluorometric assay was performed using a
fluorescence spectrophotometer (Model F-2000; Hitachi, Tokyo, Japan). GUS acitivity

was reported as picomoles of 4—methylumbelliferone produced per minute per milligram of

114

protein in total protein extracts. Total protein concentrations in the extracts were measured

using the Bradford assay, with BSA as a standard (Bradford, 1976).

GUS histochemical staining

GUS histochemistry was performed according to the procedure described by
Jefferson et al. (1987). Stained plant tissues were photographed under bright-field using a
Wild Makroskop M420 light microscope (W. Nuhsbaum, Inc., McHenry, Illinois).
Microscopical analysis of stained root cross sections were performed as follows: Root
tissues were embedded in historesin (Reichert-Jung; Cambridge Instruments, Heidelberg,
Germany) as described by De Block and DeBrouwer (1992). 10 uM sections were made
using an ultramicrotome. Sections were then photographed under bright-field using an

Axiophot microscope (Zeiss; Oberkochen, Germany).

Genetic crosses

Arabidopsis plants heterozygous for the cyrI mutation (CyrI/cyrl) were crossed to
kanamycin-resistant wild-type plants (CyrI/Cyrl) harboring the SrEnod2 3' gus construct.
Kanamycin-resistant Fl plants were allowed to self-pollinate, and kanamycin-resistant F2
seedlings were selected. Kanamycin-resistant cyrI homozygotes were identified based on
their phenotype (Diekman and Ulrich, 1995). Homozygous axr2 mutants (axr2/axr2) were
crossed to kanamycin-resistant wild-type plants (Axr2/Axr2) harboring the SrEnod2 3' gus
construct. Kanamycin-resistant Fl plants were allowed to self-pollinate, and kanamycin-
resistant F2 seedlings were selected. Kanamycin-resistant aer mutants were identified

based on their phenotype (Wilson et al., 1990).

115

RESULTS

Tissue Specificity of SrEnod2 gene expression in L. japonicus :

Requirement of the SrEnod2 3’ region.

We previously demonstrated that the enhancement of SrEnod2 mRN A accumulation
by cytokinin is regulated at a posttranscriptional level (Silver et al., 1996). Based on this
study I sought to define the SrEnod2 sequence(s) that are necessary for SrEnod2 mRN A
accumulation in response to cytokinin, and to determine whether the SrEnod2 gene is
expressed in a tissue-specific manner in roots, as it is in nodules (Van de Wiel et al., 1990;
Chen et al., 1996). To this end, we generated a fusion of a 1.9 kb SrEnod2 5’ region,
including its 5’ UTR, to the gas reporter gene, flanked by 2.5 kb of the SrEnod2 3’ region,
and designated this construct “SrEnod2 3’” (Figure 4.1). The 3’ region includes
sequences immediately downstream of the stop codon of the SrEnod2 gene. A second
construct was prepared in which the SrEnod2 3’ region was replaced with the 3’ terminator
of the nopaline synthase gene, and was designated “nos 3’.” These two constructs were
introduced into the genome of L japonicus via Agrobacterium tumefaciens-mediated
transformation. Since transformation of S. rostrata is not yet possible, L japonicus was
used because it is a legume which is self-pollinating, diploid, and amenable to molecular
genetics (Handberg et al., 1992).

Three independent lines harboring the constructs described in Figure 4.1 were
recovered, and GUS histochemistry was performed. GUS activity in plants harboring the
SrEnod2 3’ construct was localized to specific regions in the root system (Figure 4.2A and
B), and was not detected in any other tissues (Data not shown). GUS activity was seen
mainly along the primary root in the vascular cylinder and at emerging lateral root sites.
The only emerging lateral roots which displayed GUS activity were those located on the

primary root, and primary or secondary fully developed lateral roots (Data not shown).

116

GUS activity was not found in lateral roots produced later during development. Even in
the primary root, GUS expression was not observed along the full length of the root and
was never found in root tips (Data not shown). Thus, the reporter gene expression pattern
was found to be restricted to the primary root and the first emerging lateral roots in a
gradient-like fashion. L japonicus plants transformed with the nos 3’ construct failed to
show detectable GUS activity in any tissues (Figure 4.2C, and Data not shown). These
data indicate that the SrEnod2 3’ region is required for root-specific expression in L
japonicus. To determine which particular root cell types expressed the SrEnod2 3’
construct, cross sections of stained primary roots were examined. As shown in Figure
4.21, GUS activity was localized in the pericycle cells and in the pith parenchyma cells of
transgenic L japonicus roots.

Next, I wanted to determine whether the SrEnod2 3’ and the nos 3’ chimeric genes
would respond to cytokinin treatment. Roots from intact two-week-old transgenic plants
harboring the SrEnod2 3’ and nos 3 ’ constructs were incubated in a solution containing 1
uM BAP for 4 h, 8 h, 12 h , and 24 h, respectively. Control plants were incubated in
solution without BAP for the same time periods. Treatment with BAP was not found to
lead to an enhancement of GUS activity in plants harboring the SrEnod2 3’ construct, as
determined by GUS fluorometric assays (Data not shown). Plants transformed with the
nos 3’ construct failed to show GUS activity in the presence or absence of cytokinin (Data
not shown). These results are consistent with the observations that in most legumes, other
than S. rostrata and alfalfa, Enod2 genes are not regulated by application of cytokinin
(Dehio and de Bruijn, 1992; Hirsch and Fang, 1994).

117

Figure 4.1. Structure of the chimeric SrEnod2-gas reporter gene constructs.
The SrEnod2 3’ construct is composed of 1.9 kb of the SrEnod2 5’ upstream region
(sequences from -1900 to +23), fused to the gus reporter gene (cross hatched box), and
terminated by 2.5 kb of the SrEnod2 3’ downstream region (sequences from +1017 to
+2500). The black box represents the SrEnod2 3’ UTR (250 bp). The nos 3’ construct is
identical to the SrEnod2 3’ construct except that the SrEnod2 3’ downstream region was

replaced by the 300 bp nopaline synthase 3’ end (nos 3’).

118

Constructs

4900 SrEnod2 5 ' gus SrEnod2 3' +2500

some 3' m

_1900 SrEnod2 5' gas +250

542042 3' m m—

_1900 SrEnod2 5 gus nos 3

119

Figure 4.2. Histochemical localization of GUS activity in transgenic L.
japonicus and Arabidopsis plants harboring the SrEnod2 3’ and nos 3’
constructs. A, GUS staining pattern of 21-day-old L. japonicus roots. B, GUS
staining of L japonicus emerging lateral roots. C, Absence of GUS staining in L.
japonicus plants harboring the nos 3’ construct. D, GUS staining in emerging leaves and
stipules of two-week-old Arabidopsis seedlings harboring the SrEnod2 3’ construct. E,
GUS staining in stipules of two-week-old Arabidopsis seedlings harboring the nos 3’
construct. F, GUS expression in emerging lateral roots of Arabidopsis seedlings harboring
the SrEnod2 3’ construct. G, Absence of GUS expression in the meristematic region of
etiolated plants harboring the nos 3’ construct. H, GUS expression in the meristematic
region of etiolated plants harboring the SrEnod2 3’ construct. I and J, Cross sections of
primary roots from 21-day -old L japonicus and two-week-old Arabidopsis plants
harboring the SrEnod2 3’ construct, respectively. Sections were generated from the
primary root within 2 centimeters from the hypocotyl/root junctions in both plant species.

GUS expression can be seen in the vascular cylinder and pericyle cells of both plant

species.

caves 1111
tract 1.
nos 3‘

[gs Wilt
region of
stem
cctionsof
3115

m the

m specits

plant

120

 

 

121

Expression of the SrEnod2 3’ gas chimeric gene in Arabidopsis roots:

Tissue specificity and cytokinin induction.

Since the SrEnod2 3’ construct was not regulated by cytokinin in L japonicus, I
investigated whether Arabidopsis could be used to study cytokinin regulation of SrEnod2
expression. To this end, I transformed Arabidopsis plants with the SrEnod2 3’ and nos 3’
constructs, and recovered seven independent transformants for each construct.
Arabidopsis plants bearing the SrEnod2 3’ construct expressed GUS in a tissue-specific
manner similar to that observed in L japonicus. That is, expression was confined to the
vascular cylinder in the hypocotyl-root junction and to the first few emerging lateral roots
along the primary root, and weak expression was observed in the upper part of the primary
root (Figure 4.2F and Data not shown). Interestingly, GUS expression in emerging lateral
roots was found to be confined to the primary root, and first and second fully developed
lateral roots while lateral roots that developed later did not show GUS activity, as had been
observed in transgenic L japonicus plants (Figure 4.2F, 4.3B and Data not shown). In
contrast to L japonicus, GUS activity was also detected in certain parts of the shoot of
Arabidopsis plants. GUS activity was visible in emerging leaves, stipules and trichomes
(Figure 4.2D) and in the meristematic region of etiolated transgenic plants harboring the
SrEnod2 3’ construct (Figure 4.2H). GUS activity was not found in the meristematic
region of transgenic plants harboring the nos3 ’ construct (Figure 4.2G). GUS expression
decreased as the leaves expanded, and was completely absent in fully developed leaves
(Data not shown). No GUS expression was detected in stems, flowers, siliques, and
rosette leaves of mature Arabidopsis plants harboring the SrEnod2 3’ construct (Data not
shown). As in L japonicus, GUS expression was localized to cells of the root pericycle
and pith parenchyma (Figure 4.2J). No GUS expression was seen in any tissues of
transgenic Arabidopsis plants harboring the nos 3’ construct, with the exception of stipules

(Figure 4.2E and Data not shown).

122

To examine whether the SrEnod2 3’ and nos 3’ constructs were regulated by
cytokinin in Arabidopsis, two-week-old transgenic plants were placed onto solid media
containing 1 uM BAP for 12 h. Controls plants were incubated for the same time period
on solid media lacking BAP. The plants were then stained for GUS activity. The intensity
of GUS staining in roots of plants bearing the SrEnod2 3’ construct was found to be
substantially increased in response to cytokinin treatment (compare Figure 4.3A and B).
However, cytokinin treatment did not result in detectable GUS staining in transgenic plants
harboring the nos 3’ construct (Data not shown). In plants transformed with the SrEnod2
3’ construct, cytokinin treatment extended GUS expression from the hypocotyl-root
junction down the primary root (Figure 4.3B). An increase in GUS activity was detectable
by histochemical staining after 4 h of cytokinin treatment of plants harboring the SrEnod2
3’ construct (Data not shown). However, the time required for staining was found to be
twice that of the tissue shown in Figure 4.38, indicating that expression was significantly
lower at 4 h than at 12 h.

To quantify the observed changes in expression of the SrEnod2 3’ construct in
transgenic Arabidopsis plants in response to cytokinin treatment, proteins were extracted
from the roots of plants incubated with or without 1 11M BAP for 12, 24, and 30 h, and
GUS activity was measured. GUS activity was found to increase approximately 2.8-fold
in plants treated with cytokinin for 12 h, and at 30 h the GUS activity level was found to be
approximately 3.5-fold greater than that in control plants (Figure 4.4). Again, no GUS
activity could be detected in extracts of roots harboring the nos 3’ construCt (Data not
shown).

The increase in GUS activity in response to cytokinin appeared greater when intact
roots were stained for GUS activity compared to the results obtained with the fluorometric
assay (Figure 4.3A and B). This apparent discrepancy may be the result of the fact that the

SrEnod2 3’ construct is only expressed in a small portion of the total root tissues, since

123

Figure 4.3. Histochemical staining of GUS activity in Arabidopsis
seedlings harboring the SrEnod2 3’ and nos 3’ constructs treated with
cytokinin. A, Arabidopsis seedlings harboring the SrEnod2 3’ construct incubated on
media without BAP for 12 h. B, Arabidopsis seedlings harboring the SrEnod2 3’
construct treated with 1 11M BAP for 12 h. The shoots from seedlings in A and B were
removed prior to photography in order to visualize GUS staining in the upper part of the

I'OOI.

124

125

Figure 4.4. GUS activity of Arabidopsis plants harboring the SrEnod2 3’
construct after hormone treatments. Two-week-old seedlings were treated with 1
11M BAP for the time periods indicated. Control seedlings (C) were incubated for 30 h on
media without hormones. Separate seedlings were treated for 24 h with 0.02 M NAA,
and/or 40 11le ethylene. Roots were then isolated for GUS measurements. The mean
GUS activity from 3 independent experiments is shown and expressed as pmol
MU/min/mg protein. Each experiment included 5 seedlings. The error bars represent the
SD.

126

 

 

”T

 

 

m.

_ _ p — P
_ d J _ q
0 O 0
a m m m s

7508a max -58 DE 8:5

ET/
BAP

30 h NAA NAA/ ET
BAP

24h

 

12h

BAP

127

lateral roots fail to stain. Therefore, the use of the entire root system for GUS activity
measurements, may have diluted the amount of GUS protein in the total root protein
extract.

In order to rule out the possibility that cytokinin was acting non-specifically in
causing an increase in GUS activity, poly(A) RNA was isolated from roots of transgenic
Arabidopsis plants harboring the SrEnod2 3’ and nos 3’ constructs, incubated with or
without BAP for 12 h. gus mRN A levels were determined by northern blot hybridization.
Treatment with BAP was not found to affect the accumulation of gas mRNA in plants
transformed with the nos 3’ construct (Data not shown). However, in plants transformed
with the SrEnod2 3’ construct, treatment with BAP resulted in gus mRN A levels more than
4—fold higher than in the control (Data not shown). These data indicate that cytokinin
treatment results in accumulation of the gas mRNA, and that the SrEnod2 3’ region is

required for this response.

Expression of the SrEnod2 3’ gas construct in Arabidopsis is enhanced

specifically by cytokinin, and inhibited by ethylene.

To determine whether the SrEnod2 3’ gas construct was regulated solely by
cytokinin or whether its expression is also influenced by other hormones, two-week-old
transgenic Arabidopsis plants harboring this construct were incubated for 24 h in the
presence of BAP, the auxin NAA, and ethylene, alone or in combination. Control plants
were incubated under the same conditions without hormones. NAA was found not to
affect expression of the SrEnod2 3’ gus construct, and not to affect the cytokinin
enhancement of GUS activity (Figure 4.4). In contrast, ethylene appeared to have a
significant effect on expression of the SrEnod2 3’ gus construct, since in ethylene treated
plants, GUS activity was approximately 2-fold greater than in untreated controls, but

ethylene inhibited further induction of GUS activity by BAP (Figure 4.4). Similar effects

128

of ethylene on SrEnod2 mRNA levels have been observed in S. rostrata plants (D.L. Silver
and F .J . de Bruijn, unpublished data). In addition, GA3 and ABA were tested and did not

have any effect on GUS activity (Data not shown).

The SrEnod2 3’ UTR specifies cytokinin inducibility and tissue-specificity

To define the region of the SrEnod2 3’ locus which is responsible for tissue-
specificity and cytokinin inducibility, a deletion of the SrEnod2 3’ sequence was
constructed, so that only 250 bp of the 3’ region, comprising the SrEnod2 3’ UTR,
remained. The corresponding gus fusion was designated as “SrEnod2 3’A7” (Figure 4.1).
This construct was introduced into Arabidopsis, and six independent transgenic lines were
obtained. Two-week-old transgenic plants harboring the SrEnod2 3’A7 construct were
incubated for 12 h and 24 h on solid media with or without 1 1.1M BAP, and GUS activity
was measured using the fluorometric assay. Tissue-specific expression of the reporter
gene fusion was found to be identical to that in plants expressing the full SrEnod2 3’
construct (Data not shown). The level of GUS activity was enhanced about two-fold after
12 h and 24 h of BAP treatment (Figure 4.5), although the overall level of expression of
this construct was lower than that observed with the SrEnod2 3’ construct containing the
full 2.5 kb SrEnod2 3’ region. This decrease in total level of expression was also observed
by histochemical staining (Data not shown). These results indicate that sequences outside
of the 3’ UTR are required for optimal cytokinin inducibility of SrEnod2 expression.
However, these data do delimit a cytokinin response element within the 250 bp 3’ UTR of
the SrEnod2 gene.

129

Figure 4.5. GUS activity of Arabidopsis plants harboring the SrEnod2
3’UTR construct after cytokinin treatment. Two-week-old Arabidopsis plants
harboring the SrEnod2 3’UTR construct were treated for 12 h and 24 h with 1 |.IM BAP.
Roots were isolated for GUS activity measurements. Control seedlings were incubated for
24 h on media without BAP. The mean GUS activity from 3 independent experiments is
shown. Each experiment included 5 seedlings. Error bars represent the SD. The asterik

indicates a p < 0.05.

130

 

 

 

 

00 00 0
mesmosmmzmo

.EBEQ we .56 :2 SEQ

131

Expression of the SrEnod2 3’ gas construct in the cytokinin-resistant

mutant cyrl and auxin-resistant mutant aer

To further examine cytokinin regulation of the SrEnod2 3’ gus construct, I
examined the expression of this reporter gene fusion in Arabidopsis mutants defective in
cytokinin or auxin response. Transgenic Arabidopsis cyrI mutant plants harboring the
SrEnod2 3’ construct were generated (see Materials and Methods) and stained for GUS
activity. cyrI/cyrI plants harboring the construct failed to show GUS staining in roots
(Data not shown). Furthermore, the application of BAP did not cause an increase in GUS
activity in cyrI mutants bearing the SrEnod2 3’ construct (Figure 4.6). These plants did
have GUS staining in stipules, as did plants harboring the nos 3’ construct (Figure 4.2),
but not in expanding leaves or trichomes (Data not shown).

Two-week-old auxin-resistant axr2 Arabidopsis plants harboring the SrEnod2 3’
gus construct were also placed onto solid media with or without 1 11M BAP for 24 h. In
aer mutants, there appeared to be a slight increase in the level of expression of the
SrEnod2 3’ construct in the absence of BAP, but this increase was not statistically
significant (Figure 4.6). GUS activity increased in response to BAP in the aer mutant
plants qualitatively similar to wild type (Figure 4.6). The tissue specific pattern of GUS
staining in aer plants was the same as in wild-type plants (Data not shown), indicating that
the aer mutation does not affect the localized expression pattern of the SrEnod2 3’ gas

construct in Arabidopsis.

132

Figure 4.6. Regulation of the SrEnod2 3’ construct in the cytokinin-
resistant mutant cyrl, and the dominant auxin-resistant mutant aer. Two-
week-old wild-type, cyrl/cyrl, and axr2 seedlings harboring the SrEnod2 3’ construct
were treated for 24 h with 111M BAP. Control plants were not treated with BAP. Roots
were then isolated for GUS activity measurements. The mean GUS activity from 3
independent experiments is shown. Each experiment included 5 seedlings. Error bars
represent the SD.

133

 

 

160

 

WT cyrI/cyrl axr2

   

.000
mmwm42
1

140 ”
120 ”

£88m we -58 DZ BBQ

134

DISCUSSION

Conserved tissue-specificity and cytokinin regulation of the early nodulin

gene SrEnod2 in Arabidopsis

Here it is demonstrated that a chimeric reporter gene construct composed of the 5’
and 3’ flanking regions of the SrEnod2 locus of S. rostrata and the gas reporter gene is
expressed in a tissue-specific manner in both the legume plant L japonicus and in
Arabidopsis. The expression pattern in the roots of these divergent species was found to
be identical, namely in the vascular parenchyma, and in emerging lateral roots. However,
while transgenic L japonicus plants harboring this construct displayed no detectable
expression in shoots, transgenic Arabidopsis plants harboring this construct showed
distinct GUS expression in emerging leaves, stipules and trichomes (Figure 4.2D). In
transgenic Arabidopsis plants, the expression of this construct was enhanced by treatment.
with cytokinin to approximately the same degree as the accumulation of the endogenous
SrEnod2 mRNA of S. rostrata (Dehio and de Bruijn, 1992). Expression of the SrEnod2
3’-gus-SrEnod2 5’ construct was found not to be enhanced by cytokinin in the legume L
japonicus. It is possible that regulation by a nodule-specific developmental program
supercedes cytokinin regulation in L japonicus, and that cytokinins are not involved in
nodule development of L japonicus as they are in S. rostrata. Indeed, we have found that
expression of the endogenous LjEnodZ gene in L. japonicus was not enhanced by
cytokinin (D.L. Silver, R. Chen, P. J. de Bruijn, unpublished data). However, it is also
possible that changes in sensitivity to endogenous cytokinin are involved in regulation of
Enod2 genes in L japonicus, but that the cytokinin response is saturated in developed roots
of this species.

Our previous study on the cytokinin enhancement of SrEnod2 mRN A accumulation

in S. rostrata suggested a posttranscriptional mechanism. I now show that the tissue-

135

specific and cytokinin responsive expression of the chimeric gus fusion requires the
SrEnod2 3’ region. A chimeric construct which contains the nopaline 3’ terminator instead
of the SrEnod2 3’ region, failed to be expressed in roots of either L japonicus or
Arabidopsis transgenic plants (Figure 4.2C and Figure 4.3C), and could not be enhanced
by cytokinin in Arabidopsis plants (Figure 4.3D). Furthermore, the 250 bp UTR from the
SrEnod2 gene was found to be sufficient for the cytokinin response in Arabidopsis plants
(Figure 4.5). The requirement of the SrEnod2 3’ UTR for nodule parenchyma-specific
expression (Chen et al., 1996), root-specific expression in L japonicu and Arabidopsis,
and cytokinin enhancement of expression in Arabidopsis, further suggests a
posttranscriptional mechanism. 3’ UTRs contain signals needed for transcript processing
(for a review see Proudfoot, 1991), and have been shown to contain mRNA stabilizing
(Klausner et al., 1993; Weiss and Liebhaber, 1995), or de-stabilizing elements (Newman et
al., 1993; Zhang and Mehdy, 1994; Riedl and J acobs-Lorena, 1996; DeMaria and Brewer,
1996). In our studies we cannot fully rule out the involvement of the SrEnod2 5’ region in
tissue specific- or cytokinin-regulated expression patterns, although Chen et al. (1996)
have presented strong evidence that nodule-parenchyma-specific expression of the SrEnod2
gene is mediated by its 3’UTR. It remains possible that both the SrEnod2 5’ region and its
3’ UTR are essential for posttranscriptional regulation. In fact, there is increasing evidence
in the literature supporting an interaction of 5’ UTRs and 3’ UTRs of mRN As in gene
regulation (Munroe and Jacobson, 1990; Muhlrad et al., 1994; Tarun and Sachs, 1995).
These interactions have been shown to be important for posttranscriptional events such as
translation and mRN A stability (for review see Beelman and Parker, 1995). One scenario
for the mechanism of regulation of theSrEnodZ gene would be that (a) protein(s)
interacting with the 3’UTR, stabilizes the mRNA, as well as confer(s) tissue specificity. A
posttranscriptional mechanism for regulating SrEnod2 expression may afford the cell with a
rapid means of regulating SrEnod2 message levels (Green, 1993). Work is in progress to

further delimit the 3’UTR sequences which are involved in this level of regulation.

136

Evidence that SrEnod2 is regulated by cytokinin in vivo

To learn more about the role of cytokinin in regulation of the SrEnod2 gene, I
examined expression of the SrEnod2 3’ gas construct in transgenic Arabidopsis plants that
contain a mutation that affects sensitivity to cytokinin; cyrI (Deikman and Ulrich, 1995).
The cyrI mutation, which appears to have pleiotropic effects, was isolated using a root
elongation assay which examined root sensitivity to cytokinin. Roots of cyrI mutant
plants were found to have a 10-fold reduction in cytokinin sensitivity, as compared to wild-
type plants (Deikman and Ulrich, 1995). The expression of the SrEnod2 3’ construct was
found not to be affected by cytokinin treatment in cyrI plants (Figure 4.6). This interesting
observation is consistent with the cytokinin-insensitive nature of the mutant. Furthermore,
this construct was not expressed in any tissues of cyrI transgenic plants except for stipules.
These data suggest that tissue-specific regulation of the SrEnod2 3’ construct in roots,
expanding leaves, and trichomes in Arabidopsis requires a functional cytokinin-sensing
pathway.

The failure of the SrEnod2 3’ construct to respond to cytokinin in cyrI mutant
plants provides the first molecular evidence supporting the proposed cytokinin insensitivity
of this mutant, and underscores the value of this mutant for studying cytokinin responses.
In addition, these data suggest that the cyrI gene is part of a conserved cytokinin signal
transduction pathway, found both in Arabidopsis and in legume plants. To our knowledge
these data are the first to show a link between cytokinin induction of gene expression and
the existence of an in vivo cytokinin signal transduction pathway.

Another Arabidopsis mutant (amp! ) has been characterized which produces
elevated levels of cytokinin (Chaudhury et al., 1993). ampl exhibits a severe shoot
phenotype, in that it produces an overabundance of rosette leaves, and has loss of apical

dominance, phenotypes which would be expected for a cytokinin overproducer

137

(Chaudhury et al., 1993). However, amp] produces normal roots, suggesting that AMPl
acts only in the shoot. Expression of the SrEnod2 3’ gas fusion in amp] roots appeared to
be unaffected by the mutation, consistent with the idea that the amp] mutation has its

primary effects on the shoot (D.L. Silver and F. J. de Bruijn, unpublished data).

SrEnod2 3' gas expression in an auxin-resistant aer mutant of

Arabidopsis

A number of cytokinin regulated genes are also regulated by auxin, such as the gene
represented by pL8216 from Nicotiana plumbaginzfolia (Dominov et al., 1992), the cimI
. gene of tobacco (Crowell, 1994), and others (Crowell et al., 1990; Hemerly et al., 1993;
Gough et al., 1995). It has been shown that the expression of the SrEnod2 gene in S.
rostrata is not regulated by the application of auxins (Dehio and de Bnrijn, 1992).
Similarly, our studies indicate that exogenous auxins do not stimulate the expression of the
SrEnod2 3' gas construct in Arabidopsis, and it does not affect the cytokinin-enhanced
expression of this gene in transgenic plants (Figure 4.4). However, it is possible that
endogenous levels of auxin are important for cytokinin enhancement of gene expression.
For example, auxin dependence of cytokinin induction of gene expression has been
suggested for the regulation of the cdc2a gene in Arabidopsis. Expression of a cdc2a gus
fusion has been shown to be induced by cytokinin alone only in the upper part of the
primary root in intact Arabidopsis plants (Hemerly et al., 1993), while auxin is required for
cytokinin induction of expression of the cdc2a gus fusion in excised roots of the same
plants. Based on these data, Hemerly et al. (1993) suggest that auxin transport from the
shoot is required for cytokinin induction of the cdc2a gene.

To determine whether a functional auxin signal transduction pathway was necessary
for the cytokinin induction of the SrEnod2 3’ gus chimeric gene, I examined expression of

this gene in the auxin resistant mutant axr2. Cytokinin-enhanced expression of the

138

SrEnod2 3’ construct was not affected by the aer mutation (Figure 4.6). Taken together,
the lack of effect of either exogenous auxin or the aer mutation on expression of the
SrEnod2 3' gus construct suggests that cytokinins act independently of auxin in the

regulation of SrEnod2 gene expression.
Role of ethylene in regulation of the SrEnod2 gene

Treatment of transgenic SrEnod2 3’ Arabidopsis plants with ethylene alone caused
a slight increase in GUS activity compared to control plants, but application of ethylene and
cytokinin together appeared to prevent full cytokinin enhancement of gene expression
(Figure 4.4). Similarly, we found that SrEnod2 mRNA accumulated approximately two-
fold over controls in ethylene-treated S. rostrata plants, and that further SrEnod2 rnRN A
accumulation was inhibited in the presence of both ethylene and cytokinin (D.L. Silver and
F. J. de Bruijn, unpublished data). Many physiological responses to cytokinin in roots are
known to be mediated through ethylene, and application of cytokinins has been shown to
be able to increase the production of ethylene in the root (Fuchs and Lieberman, 1968;
Radin and Loomis, 1969; Bertell and Eliasson, 1992). The inhibition of root growth by
cytokinin in roots of Arabidopsis seedlings has been shown to be mediated through
ethylene (Cary et al., 1995). The mechanism by which ethylene acts to inhibit cytokinin-
induced expression of the SrEnod2 3' gus construct in Arabidopsis roots remains to be
determined. One possibility is that BAP acts to enhance expression of the SrEnod2 3’
construct by stimulating the production of ethylene, but that the relatively high level of
ethylene used in our study (40 til/l) was inhibitory for gene expression. It will be
necessary to determine the dose dependence of the response to both ethylene and cytokinin
to resolve this question. A genetic approach to better understand the role that ethylene may
play in regulating the SrEnod2 3' gas construct would be to analyze GUS expression in

ethylene insensitive mutants (Guzman and Ecker, 1990; Chang et al., 1993; Hua et al.,

139

1995 ; Roman et al., 1995), or ethylene overproducing plants (Guzman and Ecker, 1990).

These experiments are being initiated presently.

Gradient-like expression pattern of the SrEnod2 3' gas construct

The expression of the SrEnod2 3’ construct in only a limited number of emerging
lateral roots is interesting. In S. rostrata the exact cellular location of SrEnod2 expression
in roots is unknown. However, RNA gel blot analysis indicates that SrEnod2 is expressed
most highly in the primary root of S. rostrata plants (D.L. Silver and R]. de Bruijn,
unpublished data). It is tempting to speculate that this pattern of expression is due to the
level of cytokinin, which is being transported to the shoot from the root, the proposed site
of cytokinin biosynthesis (Torrey 1976; Van Staden and Davey, 1979).

The identical tissue-specific expression patterns and levels of the SrEnod2 3’ gas
construct in both L japonicus and Arabidopsis plants, namely highest at the hypocotyl-root
junction and decreasing toward the root tip, may be indicative of a signal gradient. It is
possible that a gradient of cytokinin concentration forms as it is transported from the root
tip to the shoot, and in combination with tissue-specific signals, is responsible for the
observed pattern of expression. Alternatively, a signal may be transported from the shoot
to the root, which allows only root tissues near the shoot junction to perceive the cytokinin
signal. An obvious candidate for such a signal would be auxin, which is transported from
the shoot. However, the evidence presented in this report is not consistent with auxin as
the signal, since exogenous auxin had no significant effect on GUS expression, and the
aer mutation did not affect cytokinin inducibility of the SrEnod2 3' gas fusion. There is
precedence for the role of a shoot to root signal molecule in nodulation. It has been found
that during rhizobial infection, a regulatory signal is transported from the shoot to the root
through the stele, which may be involved in the autoregulation of nodule initiation and

development (reviewed by Caetano-Anolles and Gresshoff, 1991; Hirsch and Fang, 1994).

140

A third possibility is that a negative factor produced in the root apical meristem generates an
opposing gradient to expression of the SrEnod2 3’ construct.

Expression of SrEnod2 3’ in both L japonicus and Arabidopsis was found to be
located in the pericycle and vascular pith cells of roots. This cell-specific expression
pattern of SrEnod2 3' gus fusion may also be related to cytokinin transport. Evidence in
the literature indicates that cytokinins are synthesized in the root apical meristem and
transported to the shoot via the vascular system (Torrey 1976; Van Staden and Davey,

197 9). It may be that inactive, conjugated cytokinins are transported from the root
meristem to more distal regions of the root and eventually into the shoot where they become
active through deconjugation in tissues which require cytokinin action. An intriguing
question arises as to the relationship between the cell- and tissue-specific expression pattern
seen for the SrEnod2 3' gas fusion in the root and that seen in nodules. It is known that
pericycle cells and outer cortical cells are the site of nodule primordia initiation in
determinant-type nodules (Hirsch, 1992). Further investigation is needed to discover the
identity of the endogenous signal(s) involved in forming this gradient-like pattern of gene

expression.

The role of cytokinin in root nodule formation

An important question remains as to the relationship between cytokinin
enhancement of expression of the SrEnod2 gene in roots, and its expression in nodules.
Specifically, it is not known whether SrEnod2 gene expression is under cytokinin control
during nodule development. However, we have found that the SrEnod2 3’ UTR is
required both for the tissue-specific expression pattern seen in nodules (Chen et al., 1996)
and cytokinin inducibility in roots. The cytokinin inducibility of the SrEnod2 gene in S.
rostrata roots, and of the SrEnod2 3’ chimeric gus fusion in Arabidopsis, suggests that

legume plants have evolved a mechanism to regulate SrEnod2 gene expression by making

141

use of a general cytokinin signal transduction pathway, in addition to tissue-specific
factors.

The evidence for the involvement of cytokinin in nodule development is
circumstantial. Thimann (1936) first proposed a role for auxin in nodule development as
well as the idea that nodule development may be related to lateral root development.
Libbenga et al. (1973) in an in vitro approach using pea root cortical explants, found that
cell divisions took place in the pericycle, the place of lateral root development, if auxin was
added. Cortical cell division, the location of nodule initiation, was shown to occur upon
the addition of both auxin and cytokinin. It should be noted that nodule development is
primarily determined by a plant genetic program and not by the bacteria, as evidenced by
the discovery of spontaneous nodulating alfalfa plants (Truchet et al., 1989). In addition,
alfalfa plants can be induced to form nodule-like structures upon treatment with auxin
transport inhibitors (Allen et al., 1953; Hirsch et al., 1989), or Nod factors (T ruchet et al.,
1991; Mergaert et al., 1993; Stokkerrnans et al., 1994). However, it has also been shown
that rhizobia secrete cytokinins into the culture medium (Morris, 1986; Sturtevant and
Taller, 1989; Taller and Sturtevant, 1991; Upadhyaya et al., 1991) The significance of this
source of cytokinin on nodule development remains to be proven, and no cytokinin
biosynthetic genes or mutants have been identified in rhizobium to date. Probably the most
conclusive study suggesting a role of cytokinins in nodule development comes from the
work by Cooper and Long (1994). They expressed the Agrobacterium cytokinin
biosynthetic gene, 12s, in a Rhizobium meliloti strain mutant for the nod structural genes.
This tzs-expressing Rhizobium was capable of inducing nodule-like structures on alfalfa
which expressed the Enod2 gene, supporting the idea that localized cytokinin production
may be important for nodule development. This work does not prove that cytokinin
secreted from wild-type rhizobia is required for nodule development, but rather that a
localized production of cytokinin may play a role in nodule development, be it from

Rhizobium or the plant.

142

The SrEnod2 gene as a marker for a cytokinin signal transduction pathway

Expression of the SrEnod2 3’ construct in wild type and cyr] mutant transgenic
Arabidopsis indicates that the cytokinin signal transduction pathway that regulates the
expression of this gene is conserved between legume and non-legume plants. Thus, this
chimeric gene may be a useful tool to probe the general mechanism of cytokinin signal
transduction in plants. The evidence presented here implies that a conserved protein(s)
interacts with the SrEnod2 3' mRN A region in Arabidopsis and in S. rostrata. Therefore,
endogenous Arabidopsis mRN As must share conserved binding sites with SrEnod2 3'
mRN A sequences and/or secondary mRN A structures. In order to identify the
corresponding trans-acting factor(s), we have initiated a genetic screen using the transgenic
Arabidopsis plants harboring the SrEnod2 3’construct, and are searching for RNA binding
proteins in vitro. These studies may shed light on the molecular basis of cytokinin

regulation of plant gene expression in general.

143

ACKNOWLEDMENTS

I thank Dr. Jill Deikman for her very helpful comments and suggestions on the
manuscript. I thank Dr. Rujin Chen for supplying me with the published constructs
SrEnod2 3 ’, nos 3’, and the SrEnod2 3 ’ A7.

144
REFERENCES

Allen EK, Allen ON, Newman AS (1953) Pseudonodulation of leguminous plants
induced by 2-Bromo-3,5-dichlorobenzoic acid. Amer J Bot 40: 429-435

Beelman CA, Parker R (1995). Degradation of mRNA in eukaryotes. Cell 81: 179-
183

Bent A, Kunkel BN, Dahlbeck D, Brown KL, Schmidt R, Giraudat J,
Leung J, Staskawicz BJ (1994) RPSZ of Arabidopsis thaliana: A leucine-rich repeat
class of plant disease resistance genes. Science 265: 1856-1860

Bertell G, Eliasson L (1992) Cytokinin effects on root growth and possible
interactions with ethylene and indole-3-acetic acid. Physiol Plant 84: 255-261

Binns, AN (1994) Cytokinin accumulation and action: biochemical, genetic, and
molecular approaches. Annu Rev Plant Physiol Mol Biol 45: 173-196

Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram
quantities of protein utilizing the principle of protein-dye-binding. Anal Biochem 72:
248-254

Caetano-Anolles G, Gresshoff PM (1991) Plant genetic control of nodulation. Annu
Rev Microbiol 45 : 345-382

Cary AJ, Wennuan L, Howell SH (1995) Cytokinin action is coupled to ethylene in
its effects on the inhibition of root and hypocotyl elongation in Arabidopsis thaliana
seedlings. Plant Physiol 107: 1075-1082

Chang C, Kwok SF, Bleecker AB, Meyerowitz EM (1993) Arabidopsis ethylene-
response gene ETRl: similarity of product to two-component regulators. Science 262:
539-544

Chaudhury AM, Letham S, Craig S, Dennis ES (1993) amp] -a mutant with high
cytokinin levels and altered embryonic pattern, faster vegetative growth, constitutive
photomorphogenesis and precocious flowering. Plant J 4: 907-916

Chen C-M, Jin G, Andersen BR, Ertl JR (1993) Modulation of plant gene
expression by cytokinins. Aust J Plant Physiol 20: 609-619

Chen R, Silver DL, de Bruijn FJ (1996) The nodule parenchyma-specific expression
of the Sesbania rostrata early nodulin gene SrEnod2 is mediated by its 3’ untranslated
region (3’UTR). Plant Cell; submitted

Cooper JB, Long SR (1994) Morphogenetic rescue of Rhizobium meliloti nodulaiton
mutants by trans-zeatin secretion. Plant Cell 6: 215-225

Crowell DN (1994) Cytokinin regulation of a soybean pollen allergen gene. Plant Mol
Biol 25: 829-835

Crowell DN, Kadlecek AT, John MC, Amasino RM (1990) Cytokinin-induced
mRNAs in cultured soybean cells. Proc Natl Acad Sci USA 87 : 8815-8819

145

De Block M, DeBrouwer D (1992) In-situ enzyme histochemistry on plastic embedded
plant material: the development of an artifact-free B-glucuronidase assay. Plant J 2: 261-
266

Dehio C (1989) Isolierung und Charakterisierung des Knollchenspezifisch exprimierten
Gens ENOD2 aus der tropischen Leguminose Sesbania rostrata. PhD thesis. Max-
Planck-Institut fur Ziichtungsforschung, Abteilung genetische Grundlagen der
Pflanzenzr’ichtung, Koln, FRG

Dehio C, de Bruijn FJ (1992) The early nodulin gene SrEnod2 from Sesbania rostrata
is inducible by cytokinin. Plant J 2: 117-128

Deikman J, Ulrich M (1995) A novel cytokinin-resistant mutant of Arabidopsis with
abbreviated shoot development. Planta 195: 440-449

DeMaria C., Brewer G (1996) AUFl binding affinity to A+U-rich elements correlates
with rapid mRNA degredation. J Biol Che. 271: 12179-12184

Dickstein RT, Bisseling T, Reinhold VN, Ausubel FM (1988) Expression of
nodule-specific genes in alfalfa root nodules blocked at an early stage of development
Genes Devel 2: 677-678

Dominov JA, Stenzler L, Lee S, Schwarz JJ, Leisner S, Howell SH (1992)
Cytokinins and auxins control the expression of a gene in Nicotiana plumbaginifolia
cells by feedback regulation. Plant Cell 4: 451-461

Franssen HJ, Nap J -P, Goudemans T, Stiekema W, van Dam H, Govers F,
Louwerse J, van Kammen A, Bisseling T (1987) Characterization of cDNA for
nodulin-75 of soybean: a gene product involved in early stages of root nodule
development. Proc Natl Acad Sci USA 84: 4495-4499

Fuchs Y, Lieberman M (1968) Effects of kinetin, IAA and gibberellin on ethylene
production and their interactions. Plant Physiol 43: 2029-2036

Gough C, Hemon P, Tronchet M, Lacomme C, Marco Y, Roby D (1995)
Developmental and pathogen-induced activation of an msr gene, str 246C, from tobacco
involves multiple regulatory elements. Mol Gen Genet 247: 323-337

Govers F, Franseen HJ, Pieterse C, Wilmer J, Bisseling T (1990) Function
and regulation of the early nodulin gene ENOD2. In GW Lycett, DW Grierson, eds,
Genetic Engineering of Crop Plants. Butterworths, London pp 259-269

Green PJ (1993) Control of mRNA stability in higher plants. Plant Physiol 102: 1065-
1070

Guzman P, Ecker JR (1990) Exploiting the triple response of Arabidopsis to identify
ethylene-related mutants. Plant Cell 2: 513-523

Handberg K, Stougaard J (1992) Lotus japonicus, an autogarnous, diploid legume
species for classical and molecular genetics. Plant J 2: 487-496

Hemerly AS, Ferreira P, de Almeida Engler J, Van Montagu M, Engler G,
Inze D (1993) cdc2a expression in Arabidopsis is linked with competence for cell
division. Plant Cell 5: 1711-1723

146

Hirsch AM (1992) Developmental biology of legume nodulaton. New Phytol 122: 211-
237

Hirsch AM, Bhuvaneswari TV, Torrey JG, Bisseling T (1989) Early nodulin
genes are induced in alfalfa root outgrowths elicited by auxin transport inhibitors. Proc
Natl Acad Sci USA 86: 1244-1248

Hirsch AM, Fang Y (1994) Plant hormones and nodulation: what’s the connecton?
Plant Mol Biol 26: 5-9

Hobbie L, Timpte C, Estelle M (1994). Molecular genetics of auxin and cytokinin.
Plant Mol Biol 26: 1499-1519

Hua J, Chang C, Sun Q, Meyerowitz EM (1995) Ethylene insensitivity conferred
by Arabidopsis ERS gene. Science 269: 1712-1714

Jefferson R., Kavanagh TA, Bevan MW (1987) GUS fusions: B—glucuronidase as a
sensitive and versatile gene fusion marker in higher plants. EMBO J 6: 3901-3907

Klausner RD, Rouault TA, Harford JB (1993) Regulating the fate of mRNA: The
control of cellular rron metabolism. Cell 72: 19-28

Kunkel, TA ( 1985) Rapid and efficient site-specific mutagenesis without phenotypic
selection. Proc Natl Acad Sci USA 82: 488-489

Libbenga KR, van Iren F, Bogers RJ, Schraag-Lamers MF (1973). The role of
hormones and gradients in the initiation of cortex proliferation and nodule formation in
Pisum sativum L. Planta 114: 29-39

Maniatis T, Fritsch EF, Sambrook J (1989) Molecular cloning: A laboratory
manual, 2nd ed, Cold Spring Harbor, NY, USA Cold Spring Harbor Laboratory Press

Mergaert P, van Montagu M, Prome’ J-C, Holsters M (1993) Three unusual
modifications, a D-arabinosyl, an N-methyl, and a carbomoyl group are present on Nod
factors of Azorhizobium caulinodans strain ORSS71. Proc Natl Acad Sci USA 90:
1551-1555

Miller CO, Skoog F, Okomura FS, von Saltza MH, Strong FM (1956)
Isolation, structure and synthesis of kinetin, a substance promoting cell division. J Am
Chem Soc 78: 1345-1350

Miller CO, Skoog F, von Saltza MH, Strong FM (1955) Kinetin, a cell division
factor from deoxyribonucleic acid. J Am Chem Soc 77: 1329-1334

Morris R0 (1986) Genes specifying auxin and cytokinin biosynthesis in phytOpathogens.
Annu Rev Plant Physiol 37 : 509-538

Muhlrad D, Decker CJ, Parker R (1994) Deadenylation of the unstable mRNA
encoded by the yeast MFA2 gene leads to decapping followed by 5’ to 3’ digestion of
the transcript. Genes Dev 8: 855-866

147

Munroe D, Jacobson A (1990) mRNA poly(A) tail, a 3’ enhancer of translational
initiation. Mol Cell Biol 10: 3441-3455

Newman TC, Ohme-Takagi M, Talylor CB, Green PJ (1993) DST sequences,
highly conserved among plant SAUR genes, target reporter transcripts for rapid decay in
tobacco. Plant Cell 5: 701-714

Proudfoot NJ (1991) Poly(A) signals. Cell 64: 671-674

Radin JW, Loomis RS (1969) Ethylene and carbon dioxide in the growth and
development of cultired radish roots. Plant Physiol 44: 1584-1589

Riedl A, Jacobs-Lorena M (1996) Determinants of Drosophila fushi tarazu mRN A
instability. Mol Cell Biol 16: 3047-3053

Roman G, Lubarsky B, Kieber JJ, Rothenberg M, Ecker JR (1995) Genetic
analysis of ethylene signal transduction in Arabidopsis thaliana: five novel mutant loci
integrated into a stress response pathway. Genetics 139: 1393-1409

Silver DL, Pinaev A, Chen R, de Bruijn FJ (1996) Posttranscriptional regulation
of the Sesbania rostrata early nodulin gene SrEnod2 by cytokinin. Plant Physiol 112:
559-567

Stokkermans TJW, Peters NK (1994) Bradyrhizobium elkanii lipooligosaccharide
signal induces complete nodule structures on Glycine soja Siebold et Zucc. Planta 193:
413-420

Sturtevant DB, Taller BJ (1989) Cytokinin production by Bradyrhizobium japonicum.
Plant Physiol 89: 1247-1252

Szczyglowski K, Legocki AB (1990) Isolation and nucleotide sequence of cDNA

clone encoding nodule-specific (hydroxy) proline-rich protein LEnodZ from yellow
lupin. Plant Mol Biol 15: 361-363

Taller BJ, Sturtevant DB (1991) Cytokinin production by rhizobia. In Advances in
molecular genetics of plant-microbe interactions, H Hennecke, DPS Verma,eds, Kluwer
Acadenric Publishers, Dordrecht, The Netherlands, pp 215-221

Tarun SZ, Sachs AB (1995) A common function for mRNA 5’ and 3’ ends in
translation initiation in yeast. Genes Dev 9: 2997-3007

Taylor CB, Bariola PA, DelCardayre SB, Raines RT, Green PJ (1993) RN82:

A senescence-associated RNase of Arabidopsis that diverged from the S—RNases before
speciation. Proc Natl Acad Sci USA 90: 5118-5122

Thimann KV (1936) On the physiology of the formation of nodules on legume roots.
Proc Natl Acad Sci USA 22 511-514

Tjepkema JD, Yocum CS (1974) Measurement of oxygen partial pressure within
soybean nodules by oxygen rnicroelectrodes. Planta 119: 351-360

Torrey JG (197 6) Root hormones and plant growth. Annu Rev Plant Physiol 27 : 435-
459

148

Truchet G, Barker DG, Camut S, de Billy F, Vasse J, Huguet T (1989)
Alfalfa nodulation in the absence of Rhizobium. Mol Gen Genet 219: 65-68

Truchet G, Roche P, Lerouge P, Vasse J, Camut S, de Billy f, Prome J C,

Denarie J (1991) Sulphated lipooligosaccharide signals from Rhizobium meliloti elicit
root nodule organogenesis in alfalfa. Nature 351: 670-673

Upadhyaya NM, Letham DS, Parker CW, Hocart CH, Dart PJ (1991) Do
rhizobia produce cytokinins? Biochem Int 24, 123-30

Van de Wiel C, Scheres B, Franssen H, van Lierop MJ, van Lammeren A,
van Kammen A, Bisseling T (1990). The early nodulin transcript ENOD2 is
located in the nodule parenchyma (inner cortex) of pea and soybean root nodules.
EMBO J 9: 1-7

Van Staden, J., and Davey J.E. (1979). The synthesis, tranSport and metabolism of
endogenous cytokinins. Plant Cell Environ. 2, 93-106.

Verwoerd, T.C., Dekker, B.M.M., and Hoekema, A. (1989). A small-scale
procedure for the rapid isolation of plant RNAs. Nucleic Acids Res. 17, 2362.

Weiss, I.M., and Liebhaber, S.A. (1995). Erythroid cell-specific mRNA stability
elements in the alpha 2-globin 3’ nontranslated region. Mol. Cell. Biol. 15, 2457-2465.

Wilson, A.K., Pickett, F.B., Turner, J.C., and Estelle, M. (1990). A
dominant mutation in Arabidopsis confers resistance to auxin, ethylene and abscisic
acid. Mol. Gen. Genet. 222, 377-383.

Witty JF, Minchin FR, Skot L, Sheehy JE (1986) Nitrogen fixation and oxygen in
legume nodules. Oxford Surv Plant Mol Cell Biol 3: 275-314

Zhang, S., and Mehdy, M.C. (1994). Binding of a 50-kd protein to a U-rich
sequence in an mRN A encoding a proline-rich protein that is destablized by fungal
elicitor. Plant Cell 6, 135-145.

Chapter 5

Genetic Evidence For the Existence of
SrEnod2 3’-Interacting Factors

149

150

ABSTRACT

It has previously been shown that expression of a gas reporter gene under the
control of the SrEnod2 5’ upstream and SrEnod2 3’ downstream regions (SrEnod2 5’-gus-
SrEnod2 3’ construct) can be cytokinin-enhanced in Arabidopsis, and that sequences
necessary for this cytokinin-mediated enhancement of expression are located in the 3’
nontranslated region of the SrEnod2 gene. This indicates that conserved regulatory
proteins exist between Arabidopsis and legume plants. Here, it is reported that
overexpression of the nontranslated SrEnod2 3’ downstream region via the CaMV 35S
promoter (35$ SrEnod2 3’ construct) in Arabidopsis gives rise to a novel developmental
abnormality, namely the production of more than one shoot apical meristem. This
developmental abnormality indicates that the factor(s) which regulate the SrEnod2 5’-gus-
SrEnod2 3’ construct in Arabidopsis plays an important role in development. To provide
genetic evidence for the existence of these factors, an in vivo titration experiment was
performed in Arabidopsis. The expression of the SrEnod2 5’-gus—SrEnod2 3’ construct in
plants harboring the 35S SrEnod2 3’ construct was not found to be cytokinin enhanced.
Therefore, it can be postulated that the factor(s) which interact with the SrEnod2 3’ region

are required for both cytokinin-mediated expression and for regulating meristem number.

151

INTRODUCTION

Without a doubt cytokinins play an important role in plant development, but
unfortunately little is known about their action at the molecular level. As in the studies of
other plant hormones, a genetic approach to mutant isolation in Arabidopsis has been
undertaken towards the understanding of cytokinin signal transduction. Surprisingly, very
few mutants have been isolated which appear to be specific to cytokinin. It may be that
there exists a large redundancy in genes coding for cytokinin receptors and other interacting
proteins in the cytokinin signal transduction pathway, for example.

Studies of signal transduction pathways in plants as well as in animals has utilized
marker genes to gain a foothold into determining the components of signal transduction
pathways. In conjuction with mutants in a specific pathway, marker genes can serve as a
probe to help order the genes comprising a pathway. It is proposed in this thesis that the
SrEnod2 gene may serve as a molecular probe to dissect a cytokinin signal transduction
pathway. In this thesis, evidence is presented showing that the cytokinin-enhanced
SrEnod2 mRNA accumulation occurs posttranscriptionally, requires protein synthesis,
protein phosphatases and kinases. Furthermore, it is shown that expression of the chimeric
gus reporter gene SrEnod2 5’-gus-SrEnod2 3’ is enhanced specifically by cytokinin, and in
a tissue-specific manner in roots of Arabidopsis plants. It is also tissue-specifically
expressed in L japonicus, but not cytokinin regulated. Cytokinin regulation was shown to
require the SrEnod2 3’ UTR. In addition, it was demonstrated in Chapter 4 that the
SrEnod2 5’-gus-SrEnod2 3’construct is not cytokinin-enhanced in the cytokinin-resistant
Arabidopsis mutant cyr] , thus providing genetic evidence that the SrEnod2 5’-gus-
SrEnod2 3’construct is regulated by an in vivo cytokinin signal transduction pathway.

Here, genetic evidence is presented which may indicate the existence of trans-acting

factors which are required for the functioning of the SrEnod2 3’ downstream region in

152

Arabidopsis, and that overexpression of the SrEnod2 3 ’ downstream region in Arabidopsis

causes a novel developmental phenotype.

153
METHODS

Plant material and growth conditions

Transgenic Arabidopsis thaliana Columbia and Arabidopsis amp] plants were
germinated and grown on MS media (Gibco BRL)/ 0.8% agar for 2 weeks in a growth
chamber. Arabidopsis plants were grown in Arabidopsis Mix (Hummert International,
Earth City, MO) in a growth chamber for approximately one month after which plants
were allowed to dry, and seeds harvested. Growth chamber conditions were 18 h light!
6h dark; 21° c day/18° c night.

Cytokinin treatments of transgenic Arabidopsis plants were as follows: Two-
week-old transgenic Arabidopsis plants grown on solid MS plates, as described above,
were then transferred to solid MS plates containing 1 1.1M BAP for 24 h. Roots and

green tissues were then isolated and frozen in liquid nitrogen.

Chimeric gus reporter gene construction

The 358 SrEnod2 3’ construct was prepared as follows: The CaMV 358
promoter was isolated as a HindIII-Xbal fragment from pB1121 (Clonetech). The 358
promoter (H indIII-Xbal) was then subcloned into the SacI-EcoRI digested pBIB vector
to make pBIB3SS. The 2.5 kb SrEnod2 3’ region was removed from the SrEnod2 3’
construct (see Chapter 4 of this thesis) as a SacI-EcoRI fragment and subcloned into the
pBIB35S digested with SacI-EcoRI. The SrEnod2 3’A7 deletion was made using the
Eon1 digestion protocol as described by the manufacturer (Erase A Base, Boeringer

Mannheim). The 3’A7 deletion was subconed into pBIB3SS as a SacI-EcoRI

fragment, making the construct 35S SrEnod2 3’A7.

154

Plant Transformation

Transformation of Arabidopsis thaliana Columbia plants with the 35S SrEnod2
3’ 358 SrEnod2 3’A7, and SrEnod2 5’-codA-SrEnod2 3’ contructs were perfomed

using vacuum infiltration according to the procedure described by Bent et al. (1994).

RNA isolation and northern blot analysis

Total RNA was isolated from roots of transgenic Arabidopsis plants according
to the method of Verwoerd et al., (1989). Poly(A) RNA was then isolated from total
RNA by oligo-dT column chromatography as described by the manufacturer’s
instructions (5 prime- 3 prime, Inc., Boulder, Co). Northern blot analysis was
performed as according to Maniatis et al. (1982). Approximately 3 ug of poly(A) RNA
was used for analysis. The 2.5 kb SrEnod2 3’ region, and the 250bp SrEnod2 3’ UTR
was used as a DNA probe for northern analysis. [32P]-dATP labelled SrEnod2 3’ DNA
probes were generated by random priming (Boehringer Mannheim). Hybridization and

washing conditions of Northern blots were as previously described (Silver et al.,

1996). Radioactive blots were exposed to autoradiography film (Dupont) at -750 C.

Quantification of GUS enzymatic activity

Quantification of GUS enzymatic acitivity was performed according to the
procedure described by Jefferson et al. (1987). The fluorometric assay was performed
using a fluorescence spectrophotometer (Model F-2000; Hitachi, Tokyo, Japan). GUS
acitivity was reported as picomoles of 4-methylumbelliferone produced per minute per
milligram of protein in total protein extract. Total protein concentrations in the extracts

were measured using the Bradford assay with BSA as a standard (Bradford, 1976).

155

GUS histochemical staining

GUS histochemistry was performed according to the procedure described by
Jefferson et al. (1987). Stained plant tissues were photographed under bright-field
using an Axiophot microscope (Zeiss; Oberkochen, Germany).

Genetic crosses

T2 generation Arabidopsis plants harboring the 358 SrEnod2 3’ construct
(hygromycin resistant) were crossed to plants harboring the SrEnod2 5’-gus-SrEnod2
3’ construct. Kanamycin/hygromycin resistant F, plants were allowed to self-pollinate,
and F2 Kanarnycin/hygromycin resistant seedlings were selected, and analyzed for
GUS activity. Plants harboring the 35S SrEnod2 3’ construct were crossed to the
amp] mutants (Chaudhury et al., 1995) harboring the SrEnod2 5 ’-gus-SrEnod2 3’
construct. Kanamycin/hygromycin resistant F, plants were allowed to self-pollinate,
and F2 Kanamycin/hygromycin resistant seedlings were selected, and will be used for
further analysis.

156

RESULTS

Overexpression of the SrEnod2 3’ downstream region in Arabidopsis

causes a novel developmental phenotype.

It has been shown that the SrEnod2 3’ UTR is required for the cytokinin-enhanced
expression of the SrEnod2 5’-gus-SrEnod2 3’construct in Arabidopsis (Chapter 4).
Overall, tissue-specificity and cytokinin-enhancement of gene expression appear to be
conserved between S. rostrata, L japonicus (tissue-specificity only), and Arabidopsis. It
is tempting to speculate that factors which regulate the expression of the SrEnod2 5’-gus-
SrEnod2 3’construct in ArabidOpsis are homologous to factors which regulate the SrEnod2
gene in the legumes S. rostrata, and L japonicus. To demonstrate the existence of such
factors, an in vivo titration experiment was designed as diagrarnmed in Figure 5.1. The
aim was to overexpress the SrEnod2 3’ region under the control of a constitutive promoter
and genetically cross this construct into a homozygous line expressing the SrEnod2 5’-gus-
SrEnodZ 3’construct in Arabidopsis (Figure 5.1). The expectation is that if expression of
the SrEnod2 3’ region can titrate out SrEnod2 3’-interacting factors, then GUS expression
from the SrEnod2 5’-gus-SrEnod2 3’ construct will be signifcantly reduced (Figure 5.2).

To overexpress the SrEnod2 3’ downstream region in Arabidopsis, the 2.5kb
SrEnod2 3 ’ region was fused to the CaMV 35$ promoter and introduced into Arabidopsis
via Agrobacterium tumefaciens. Seven independent lines were isolated harboring the 35S
SrEnod2 3’ construct. Surprisingly, plants expressing these 358 SrEnod2 3’ constructs
exhibited a novel developmental phenotype. As shown in Figure 5.3, plants harboring the
35S SrEnod2 3’ construct are initially arrested in the production of leaves during the first
10 d after germination while wild-type plants have produced four leaves. After 10 (1 post-

germination, leaf primordia begin to form from more than one

157

Figure 5.1. In vivo titration scheme. A, Diagram of the 358 SrEnod2 3’ and
SrEnod2 3’ construct used to determine whether proteins interact with the SrEnod2

3’UTR. The 2.5 kb of the SrEnod2 3’ region was overexpressed under the control of the
CaMV 358 promoter. The SrEnod2 3’ construct contains 1.9 kb of SrEnod2 5’ region
ligated to the gus reporter gene (hatched box) fused to the 2.5 kb SrEnod2 3’ region. The
black box represents the SrEnod2 3’UTR ending at the poly(A) addition site. B, in vivo
titration experiment was performed by a genetic cross between Arabidopsis plants
harboring the 35S SrEnod2 3’ construct and plants harboring the SrEnod2 3’ construct. F2

progeny were analyzed for GUS activity.

 

 

158

A

Comma

CaMV 35s SrEnod2 3' +2500

   

358 SrEnod2 3'

_,900 SrEnod2 5' gas SrEnod2 3' +2500

share 3' “—

35S SrEnod2 3' X SrEnod2 3'

GUS activity quantification

159

Figure 5.2. Hypothetical mechanism for in viva titration. The panel on the
left represents a cell in which a protein(s) interacts with the gas-SrEnod2 3’UTR mRNA
resulting in GUS activity. The panel on the right represents a cell overexpressing the
SrEnod2 3’UTR which titrates out the SrEnod2 3’UTR-binding protein(s) resulting in a
decrease in the levels of gas-SrEnod2 3’UTR mRNA, and thus a decrease in GUS activity.

160

in vivo Titration

GUS’I‘

 

' . -__Szffi:_'AAAAA
NW“
ABQ‘NSL
AAAAA
\m jg/ “AAA

AAAAA

WA

Jig fin...

AAAAA

_\,

Gusw

161

meristem (Figure 5.3). These plants were named Siamese, for the presence of two
meristems (Figure 5.3 and 5.4). Infrequently, three or four meristems were formed. A
second phenotype was less frequently observed among the seven lines, which were named
Flag. Flag plants produced a single, large leaf at the stage in which wild-type plants
produced two primary leaves. Further leaf production was arrested until about day 10,
then two possible fates of the meristem occured. One fate is that two meristems form,
giving rise to the Siamese phenotype. The second fate is that a normal, single meristem
forms, giving rise to a wild-type phenotype, except for production of a the single, large
primary leaf. Both Siamese and Flag plants produce normal leaves, trichomes, roots,
stems, flowers, and a normal seed set. However, at a low frequency, a more severe
phenotype was seen in multiple independent lines. This phenotype exhibited a biforcated
or triforcated root system as well as shoot abnormalities (Figure 5.3). To determine if the
Siamese/Flag phenotype was due to overexpression of the SrEnod2 3’ UTR, and / or
sequences downstream of the poly(A) addition site, a deletion of the SrEnod2 3’ region
containing 170 bp of the 3’ UTR was fused to the CaMV 358 promoter and was designated
as 358 SrEnod2 3’A6 (Figure 5.5). This construct contains 110 bp of an internal fragment
of the SrEnod2 3’ region as a result of the deletion procedure. Sixteen independent lines
were generated harboring the 35 S SrEnod2 3 ’A6 construct. Eight T2 independent lines
harboring the SrEnod2 3’A6 construct were examined, and all exhibited the Siamese/Flag
phenotype. Thus, the Siamese/Flag phenotype is likely due to expression of the SrEnod2
3’UTR and not sequences downstream. Three other control constructs were produced to
determine whether the Siamese/Flag phenotype was specifically due to the overexpression
of the SrEnod2 3’ UTR or can be caused by the overexpression of any UTR. To
overexpress a different 3’UTR, the nopaline synthase 3’ terminator (nos 3’) was fused to
the 35S promoter. Two other constructs contained promoterless nos 3’ and SrEnodZ 3’
regions (Figure 5.5). These constructs have recently been transformed into Arabidopsis

and await further analysis.

162

Figure 5.3. Phenotypes of seedlings harboring the 358 SrEnod2 3’
construct. A, Two-week-old non-transformed wild-type plant. B, One-week old
Siamese plant lacking visible leaves. C, Two-week-old Siamese plant. Arrows indicate the
two independent whorls of leaves. D, Two-week-old Flag plant. Arrow indicates a single
primary leaf. E, Biforcated root system of a two-week-old Siamese plant F, Root system
of a non-transformed wild-type plant. G, Siamese plants in soil. H, Non-transfonned

wild-type plants in soil.

 

 

 

 

163

 

164

Figure 5.4. Scanning electron microscopy of lO-day-old plants harboring
the 358 SrEnod2 3’ construct exhibiting Siamese and Flag phenotypes. A,
The first two leaf primordia emerging from the apex of a wild-type plant are visible, as
indicated by asteriks. (450X magnification). B, A ten-day-old Siamese plant exhibiting
multiple structures in the apical region (120X). C, A higher magnifcation (370X) of the
apical region of the plant shown in panel B. Multiple structures (leaf primordia and or
meristems) are indicated by asteriks. D, The apical region of a Flag plant which has
developed two meristems.(220X magnification). Meristems are indicated by arrows; The

asterik indicates leaf primoridia

 

165

m.

ti

 

s1!

166

Figure 5.5. Diagram of constructs used as controls. ' The 358 S rEnodZ 3’A6
construct contains 170 bp of the SrEnod2 3’UTR along with 110 bp of an internal region
from downstream sequence and was used to determine whether the Siamese/Flag
phenotype was due to the overexpression of the SrEnod2 3’UTR or sequences
downstream. The 35S nos 3’ construct contains the nopaline synthase 3’ terminator (nos
3’) under the control of the CaMV 35S promoter. Promoterless SrEnod2 3’ and nos 3’

sequences were transformed into Arabidopsis as additional controls.

 

 

Constructs

35S SrEnod2 3' A6

35S nos 3'

pBIB SrEnod2 3'

pBIB

167

SrEnod2 3 '

 

 

+2500

 

168

Inheritence of the Siamese and Flag phenotype across three generations was in a
non-Mendelian fashion (data not shown). Plants exhibiting a wild-type phenotype gave
rise to Siamese/Flag plants, and Siamese/Flag plants gave rise to plants with wild-type
phenotype. Therefore, it was of interest to determine whether the expression level of the
SrEnod2 3 ’ region correlates with phenotype. Total RNA from two-week-old wild-type
and Siamese plants were isolated for nothem blot anlaysis. A roughly 300 nt RNA was
detected when northern blots were probed with either a full length SrEnod2 3’ region or
only the SrEnod2 3’ UTR (Figure 5.6), indicating that the 2.5 kb 3’ end may be processed
at the normally used poly(A) addition site (Dehio and de Bruijn, 1992). This idea was
supported by finding this same size transcript in the poly(A)’ RNA fraction (Figure 5.7).
Overall, the expression levels of the 35S SrEnod2 3’ construct in multiple independent
lines did not correlate with phenotype (Figure 5.6).

Genetic Evidence for the existence of SrEnod2 3’-interacting factors

Genetic crosses were made between twelve independent lines harboring the 35S
SrEnod2 3’ construct and an inbred line ( T4 generation) harboring a single copy of the
SrEnod2 5 ’-gus-SrEnod2 3’construct (Figure 5.1). Figure 5.8 shows the results of GUS
staining of F, plants from this cross. In general, F, plants showed a dramatic decrease in
expression in apical and root tissues. Plants harboring the SrEnod2 5’-gus-SrEnod2
3’construct typically show GUS staining in trichomes of newly expanding leaves and
infrequently in the expanding leaf (Figure 5.8). However, F, progeny showed reduced or
absence of GUS staining in trichomes of newly expanding leaves. In addition, tissue-
specific staining in roots not treated with BAP was significantly reduced or absent. F2

plants were then analyzed for GUS activity in response to cytokinin. Two-week-old

169

Figure 5.6. Northern blot analysis of Arabidopsis plants harboring the 355
SrEnod2 3’ construct. Lines WT7-6 and WTl-9 did not segregate for the
Siamese/Flag phenotype. The other lines shown segregated for the Siamese/Flag
phenotype. C = non-transformed wild-type plant. Total RNA was isolated from entire
two-week-old seedlings and probed with 2.5 kb SrEnod2 3’ DNA. A ubiquitin DNA
probe (ubi-I) was used as a loading control. Lines were named based as described in the
following example: WT1-9; WT refers to the phenotype of the parent line (F2) from which
the F3 line was derived. 9 refers to the parental F, vacuum infiltrated plant. 1 denotes the
individual line derived from plant 9. WT = wild-type phenotype, Sia = Siamese
phenotype. Six additional lines were analysed and yielded similar results.

17C)

lgmefi

H... F3 Line WT7-6 Sia7-7 WT1-9 WTl-l
W Phenotype C WT. Sia WT WT Sia WT

y.“ ’3 Si’EI’lOdZ 3' UTR
“gains . _. '

- ‘t 9 no-1 p” ? ‘N-fi H -~ “bi-1

 

171

Figure 5.7 . Accumulation of the SrEnod2 3’UTR mRNA. Total RNA from
Line 1 segregating for the Siamese (Sia) and wild-type (wild type) phenotype was probed
with a 250 bp DNA probe encompasing the SrEnod2 3’UTR (3’UTR). C = non-
transformed wild-type plant. Poly(A)+ RNA from Line 7 segregating for the Siamese/WT
pheotype was probed with 2.5 kb of the SrEnod2 3’ downstream region (3’end). The
location of the 288, 188 and 5S ribosomal RNAs as size markers are indicated.

 

a: T132
311'

r . '1'
141253213

til. 1':

probe

F2 Line
Phenotype

172

<
<
z 3.
°‘ <
7‘9 E»
E3 8.
3'UTR 3'end
1 7
Sia WT C 3,, Wt

288
188

"' " " .. SrEnod2 3' UTR

SS

173

Siamese and wild-type phenotype F2 plants and plants harboring the SrEnod2 5’-gus-
SrEnod2 3’ construct were placed onto solid media with or without 1 uM BAP for 24 h.
Protein was extracted from roots and GUS activity was quantified. Figure 5.9 shows that
the levels of GUS staining in roots was absent in both Siamese and wild-type phenotype F2
plants in five out of six lines tested (only line WT7-3 showed staining). In addition, GUS
staining was not enhanced in five out of 6 F2 lines treated with BAP (again, only line
WT7-3 showed staining). Also, GUS staining of trichomes was reduced or absent in three
out of six F2 lines (lines: Sia1-9, Sial-10, Sia7-7). gus reporter gene expression, as
measured by GUS acitivty in roots of both Siamese and wild-type phenotype F2 plants,
was not cytokinin enhanced except for line WT'7-3 (Figure 5.9). Indeed, this decrease in
GUS activity may define a form of cosuppression, in which the titration of trans-acting
factors accounts for the decrease in GUS expression. However, other proposed
mechanisms of cosuppression are conceivable (Matzke et al., 1994; Van Blokland et al.,
1994; de Carvalho Neibel et al., 1995), and would need to be investigated. Thus, this in
vivo tiratration experiment provides evidence for the existence of SrEnod2 3 ’-interacting

factors involved in cytokinin control of gene expression.

174

Figure 5.8. GUS expression in F, progeny from crosses between plants
harboring both of the SrEnod2 3’ and 35S SrEnod2 3’ constructs. A and B,
GUS staining in two-week-old seedlings harboring the SrEnod2 3’ construct not treated

and treated with 1 uM BAP for 24 h, respectively. C and D, GUS staining of F, progeny.

 

175

 

176

Figure 5.9. GUS expression in F, progeny from crosses between plants
harboring both of the SrEnod2 3’ and 358 SrEnod2 3’ constructs. A, Top
panels show plants not treated with 1 11M BAP, and bottom panels show plants of the same
line treated with BAP for 24 h. Panels a,b,c,d,e show a control plant and representitive F,
progeny of lines WT7-3 WT, Sia1-5 WT, Sia1-5 Sia, Sia7-2 Sia, respectively. GUS
staining was seen only in plants in panels a and b. B, GUS activity measurements of six
F, lines. C = plants harboring only the SrEnod2 3’ construct; asteriks denotes samples
which showed BAP-enhanced GUS staining. Refer to Figure 5.6 for explanation of

nomenclature for lines.

 

 

llarts
in 10;-
i it in
$111361,
313

Is of 511
laps

1101'

177

 

B

-1

pmol MU min '1 mg protein

 

Phenotype:
Line: C

3]:

WT WT Sia
WT7-3 Sin] -5

3]:

 

I -BAP
I+BAP

 

WT Sia WT Sia WT Sin Sia7-7
Sial-9 Sia] -lO Sia7-2 Sin

178

Figure 5.10. Diagram illustrating normal ArabidOpsis apical meristem
development. Central circle represents the shoot apical meristem. The first two leaf
primordia are formed opposite each other at nearly the same time. Subsequent leaf
primordia are formed in succession giving rise to a spiral phyllotaxy. Numbers denote the

chronological order in which leaf primordia form.

 

 

 

179

Arabidopsis Apical Meristem Development

 

180

DISCUSSION

Future characterization of the plants expressing the 358 SrEnod2 3’

construct.

Many examples for the involvement of 3’UTRs in the regulation of Drosophila
development have been reported (for review see Johnston, 1995). For example, the
anterior determinant bicoid, a homeodomain protein, binds to the 3’UTR of its posterior
counterpart caudal, a homeodomain transcription factor, and negatively regulates its
translation in a posterior to anterior gradient (Dubnau and Struhl, 1996; Rivera-Pomar et
al., 1996). In embryos lacking bicoid expression, caudal protein will accumulate evenly
throughout the embryo, giving rise to deletion of head and thoracic segmentation (Mlodzik,
et al., 1990). Another important example comes from the study of C. elegans
development. Expression of the fem-3 gene in the hermaphrodite germ line allows the
switch to spermatogenesis and negative regulation of fem-3 expression is required for the
switch to oogenesis (Barton et al., 1987). Negative regulation of fem-3 expression was
shown to localize to the 3’UTR of fem-3 (Ahringer and Kimble, 1991). In an in vivo
tritration experiment, overexpression of the 3’UTR of fem-3 caused 21% masculinization
of the hermaphrodite germ line.

In the SrEnod2 3’ in vivo titration experiment, the overexpression of the SrEnod2
3 ’UTR in Arabidopsis resulted in an unexpected novel developmental phenotype, the
Siamese/Flag phenotype. It has been been shown that the SrEnod2 5’-gus-SrEnod2 3’
construct is expressed at a lower level in the meristematic region of Arabidopsis (see
chapter 4), as compared to expression in the root system. These alterations in the
determination of meristem number of Siamese plants may reflect the level of expression of
SrEnod2 3’-interacting proteins in these tissues as compared to the root system, where a

root phenotype is only rarely observed. It may be that this lower level of expression of the

181

SrEnod2 5’-gus-SrEnod2 3’ construct reflects a low level of expression of SrEnod2 3’-
interacting proteins in these tissues. Therefore, expression of these SrEnod2 3 ’-interacting
proteins in the meristem may be low enough that they can be titrated away from their
normal in vivo targets, whereas the higher expression of the SrEnod2 3 ’-interacting
proteins in roots may not allow for a significant decrease in their effective concentrations by
titration. Analysis of control plants harboring the promoterless nos 3 ’ and SrEnod2 3 ’
regions are in progress. To better understand the Siamese phenotype, a detailed
developmental anlaysis will need to be performed using scanning electron microsc0py. It
is not known at what stage of meristem development is effected to give rise to the
Siamese/Flag phenotype. The only evidence which supports a postembryonic effect is the
existence of Flag plants. Figure 5.10 shows a diagram of normal Arabidopsis meristem
development The first two leaf primordia are formed opposite each other at approximately
the same time within the first two days after germination. Following the development of
these primary leaves, single primordia develop in a clockwise or counter-clockwise
fashion, until a mature whorl of leaves is formed (Medford et al., 1992). The presence of
a single primary leaf in Flag plants suggests that the meristem was functional at an early
stage after germination, but soon changed fate. A close examination of the meristem in the
embryo, and during germination may determine whether this developmental abnormality
occurs embryonically or postembryonically.

The existence of multiple idependent lines makes it likely that the Siamese
phenotype is indeed due to expression of the 358 SrEnod2 3’ construct, and not do to
random insertional mutagenesis. A question remains as to the possible involvement of
components of a cytokinin signal transduction pathway in the Siamese phenotype. To gain
some insight into this question, the cytokinin-overproducing mutant amp] (Chaudhury et
al., 1995) harboring the SrEnod2 5’-gus-SrEnod2 3’ construct was crossed to plants
harboring the 358 SrEnod2 3’ construct. The amp] mutation gives rise to an increase in

leaf production, as well as a loss of apical dominance (Chaudhury et al., 1995). F, plants

182

of being analysed for phenotype and GUS activity. Another question is what is the
connection between this meristem phenotype and SrEnod2 expression in roots and
nodules? The answer to this question will lie in the isolation of trans-acting factors and the
study of their function. One can only speculate that the protein(s) which may be titrated
away from its normal in vivo target is a negative regulator of a gene(s) involved in
meristem development (perhaps through translational regulation).

It is important to note that overexpression of only the SrEnod2 3’UTR gives rise to
the Siamese/Flag phenotype, and downregulation of the gus reporter construct. This
phenotype and measurement of GUS activity may be usefull assays for the further
delimitation of the SrEnod2 3 ’UTR for sequences important for its function. In
conjunction with these assays, deletions or linker-scanning mutations of the SrEnod2
3 ’UTR-gus fusions may allow for the identification of sequences important for cytokinin
regulation of the SrEnod2 gene. It will be interesting to determine whether overexpression
of sequences giving rise to the Siamese/Flag phenotype are the same as those which are
important for cytokinin regulation.

The utility of this in vivo titration experiment may extend to answering questions
regarding the existence of SrEnod2 3’-binding proteins in nodules, as well as the function
of the SrEnod2 protein in nodules. The latter idea is based on the assumption that titration
of SrEnod2 3’-binding proteins may down regulate the SrEnod2 gene, thereby giving a
cosuppression effect. Multiple independent lines of Lotus comiculatus plants harboring the
358 SrEnod2 3' construct have been generated and are now in the process of being

analyzed for alterations in endogenous LcEnodZ gene expression.

183

ACKNOWLEDGMENTS

I thank Philipp Kapranov for helping develop the in vivo titration experiment, as
well as for making the pBIB SrEnod2 3 ’, and 35S nos 3’ constructs. I thank Carol Flegler

for all work involving scanning electron microscopy analysis.

184

REFERENCES

Ahringer J, Kimble J (1991) Control of the sperm-oocyte switch in Caenorhabditis
elegans herrnaphrodites by the fem-3 3’ untranslated region. Nature 349: 346-348

Barton MK, Schedl TB, Kimble J (1987) gain-of—function mutations of fem-3, a
sex-determination gene in Caenorhabditis elegans. Genetics 115: 107-19

Bent A, Kunkel BN, Dahlbeck D, Brown KL, Schmidt R, Giraudat J,
Leung J, Staskawicz BJ (1994) RPSZ of Arabidopsis thaliana: A leucine-rich repeat
class of plant disease resistance genes. Science 265: 1856-1860

Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram
quantities of protein utilizing the principle of protein-dye-binding. Anal Biochem 72:
248-254

Chaudhury AM, Letham S, Craig S, Dennis ES (1993) amp] -a mutant with high
cytokinin levels and altered embryonic pattern, faster vegetative growth, constitutive
photomorphogenesis and precocious flowering. Plant J 4: 907-916

de Carvalho Niebel F, Frendo P, Van Montagu M, Cornelissen M (1995)
Post-transcriptional cosuppression of B-l-3-glucanase genes does not affect accumulation
of transgene nuclear mRNA. Plant Cell 7: 347-358

Dehio C, de Bruijn FJ (1992) The early nodulin gene SrEnod2 from Sesbania rostrata
is inducible by cytokinin. Plant J: 2 117-128

Dubnau J, Struhl G (1996) RNA recognition and translational regulation by a
homeodomain protein. Nature 379: 694-699

Jefferson RA, Kavanagh TA, Bevan MW (1987) GUS fusions: B-glucuronidase as
a sensitive and versatile gene fusion marker in higher plants. EMBO J 6: 3901-3907

Johnston D St (1995) The intracellular localization of messenger RNAs. Cell 81: 161-
170

Maniatis T, Fritsch EF, Sambrook J (1989) Molecular cloning: A laboratory
manual, 2nd ed, Cold Spring Harbor, NY, USA Cold Spring Harbor Laboratory Press

Matzke AJ M, Neuhuber F, Park YD, Ambros PF, Matzke MA (1994)
Homology-dependent gene silencing in transgenic plants: Epistatic silencing loci contain
multiple copies of methylated transgenes. Mol Gen Genet 244: 219-229

Medford JI, Behringer FJ, Callos JD, Feldmann KA (1992) Normal and
abnormal development in the Arabidopsis vegetative shoot apex. Plant Cell 4: 631-643

Mlodzik M, Gibson G, Gehring WJ (1990) Effects of ectopic expression of caudal
during Drosophila development. Development 109: 271-277

Rivera-Pomar R, Niessing D, Schmidt-Ott U, Gehring WJ, Jackle H (1996)
RNA binding and translational suppression by bicoid. Nature 379: 746-749

185

Silver DL, Pinaev A, Chen R, de Bruijn FJ (1996) Posttranscriptional regulation
of the Sesbania rostrata early nodulin gene SrEnod2 by cytokinin. Plant Physiol 112:
559-567

Stougaard J (1993) Substrate-dependent negative selection in plants using a bacterial
cytosine deaminase gene. Plant J 3: 755-761

Van Blockland R, Van der geest N, Mol JNM, Kooter JM (1994) Transgene-
mediated suppression of chalcone synthase expression in Petunia hybrida results from an
increase in RNA turnover. Plant J 6: 861-877

Van de Wiel C, Scheres B, Franssen H, van Lierop MJ, van Lammeren A,
van Kammen A, Bisseling T (1990). The early nodulin transcript ENOD2 is
located in the nodule parenchyma (inner cortex) of pea and soybean root nodules.
EMBO J 9: l-7

Verwoerd TC, Dekker BMM, Hoekema A (1989) A small-scale procedure for the
rapid isolation of plant RNAs. Nucleic Acids Res 17: 2362

 

Chapter 6

FUTURE PERSPECTIVES

186

187

A central problem to the study of the molecular action of cytokinins has been the
lack of a gene(s) which is specifically regulated by cytokinin to be used to dissect cytokinin
signal transduction pathways. The work presented in this thesis was based on that by
Dehio and de Bruijn (1992) showing that expression of the SrEnod2 gene can be
specifically enhanced by application of cytokinin. Their work led me to further determine
the important parameters of this regulation by cytokinin and encouraged me to try to
develop a system in Arabidopsis utilizing the SrEnod2 gene as a marker for cytokinin
signal transduction. Surprisingly, expression of the SrEnod2 gene is specifically enhanced
by cytokinin in Arabidopsis, although it is not in legumes other than S. rostrata and alfalfa.
In addition, genetic evidence was presented in Chapter 5 which indicates the existence of
SrEnod2 3’-interacting proteins in Arabidpsis. This thesis opens the way for the genetic
dissecton of a cytokinin signal transduction pathway in Arabidopsis, and specifically the
isolation of SrEnod2 3’-interacting factors.

A selection scheme was devised based on the cytokinin-specific enhancement of
expression of the SrEnod2 5’-gus-SrEnod2 3’ construct in Arabidopsis. This scheme
utilizes the E. coli cytosine deaminase gene (codA). The codA gene functions to deaminate
cytosine as part of the pyrimidine salvage pathway in E. coli, whereas higher eukaryotes
lack the ability to deaminate cytosine (Danielsen et al., 1992). It has been shown that
Arabidopsis plants expressing the codA gene under the control of the 35S promoter grown
in the presence of 5-fluorocytosine were severely inhibited in growth (Petra et al., 1993).
This inhibition is due to the deamination of 5-fluorocytosine to 5-fluorouricil, an inhibitor
of thymidilate synthase. Arabidopsis plants which harbor a single copy of the SrEnod25’-
gas-SrEnod2 3’ construct were transformed with the SrEnod25’-codA-SrEnod23’
construct. Work is in progress to create inbred lines which harbor both constructs. A
schematic diagram of this selection scheme is shown in Figure 6.1. Plants which harbor

both GUS and codA genes under the control of the SrEnod2 5’ and 3’ regions will be

188

mutagenized with EMS, and M, plants selected for growth on plates containing 5-
fluorocytosine. To eliminate mutations in the SrEnod25’-codA-SrEnod2 3’ T-DNA, which
would give rise to 5-fluorocytosine resistant plants, plants which survive this selection will
be assayed for GUS expression. Future work will focus on the characterization of mutants
from this screen and the isolation of their corresponding genes. Any gene which can be
isolated from such mutants, will be used as a probe to isolate possible legume homologs,
and used to study function in the legume plant. Alternative approachs to consider for the
isolation of SrEnod2 3’ RNA binding proteins are by screening a nodule-specific cDNA
expression library using a translational repression system in E. coli (Jain and Belasco,

1996) or by the three-hybrid system in yeast (SenGupta, 1996).

189

REFERENCES

Danielsen S, Kilstrup M, Barilla K, Jochimsen B, Neuhard J (1992)
Characterization of the Escherichia coli codBA operon encoding cytosine perrnease and
cytosine deaminase. Mol Microbiol 6: 1335-1344

Dehio C, de Bruijn FJ (1992) The early nodulin gene SrEnod2 from Sesbania rostrata
is inducible by cytokinin. Plant J: 2 117-128

Jain C, Belasco JG (1996) A structural model for the HIV-1 Rev-RRE complex
deduced from altered-specificity Rev varients isolated by a rapid genetic strategy. Cell
87 : 115-125

Perera RJ, Linard CG, Signer ER (1993) Cytosine deaminase as a negative selective
marker for Arabidopsis. Plant Mol Biol 23: 793-799

SenGupta DJ, Zhang B, Kraemer B, Pochart P, Fields S, Wickens M
(1996). A three-hybrid system to detect RNA-protein interactions in vivo. Proc Natl Acad
Sci USA 93: 8496-8501

190

Figure 6.1. Genetic selection scheme. Arabidopsis plants harboring both the
SrEnod2 5’-gus-SrEn0d2 3’ and SrEnod2 5’-c0dA-SrEn0d2 3’ construct will be
mutagenized with EMS and F,plants selected for growth on 5-Fluorocytosine. 5-
Fluorocytosine-resistant plants will then be analysed for lack of GUS activity. The
construction of the SrEnod2 5’-codA-SrEnod2 3’ was as follows: The codA coding region
was subcloned as a ClaI-XbaI fragment (1285 bp) from the pNE3 vector (Stougaard,

1993) into pBluescript SK+ (Stratagene). The codA coding region in SK+ was subcloned
as a SacI-HindII fragment into a BamHI (made blunt ended)-Sacl digested SrEnod2 5’-
gus-SrEnodZ 3’ construct (with gas gene removed). The construction of the SrEnod2 5’-

gus-SrEnodZ 3’construct was as described in Chapter 4.

 

 

191

Selection Scheme

SrEnod2 5 ’-gus-SrEn0d2 3 ’
SrEnod2 5 ’-c0dA-SrEn0d2 3 ’

 

j EMS mutagenize
M1
M2
. 5 Select
. 1' 5-fluorocytosine
resistance

 

Test for lack of
GUS activity

"Illlllllllllllll