:1“

43. -
_- -fl 5-: $5.13:
.:'.»-f‘ ‘ '
v’
v',’
'P
A .-
i=3
..
...
Vxé-Jg‘
a

---v

I “in“.
‘11:. 1""lflfi tit,- J, ‘2:
319‘. t
. Q f‘;“l:" “1
‘L.)v 9"", .21": :31 “l
~o  1

V -"...4
.. V‘i‘. :: V‘“
. 2V: ”m‘::“-‘ I: -
,.,.- 0:: ' 'A‘J/n'
' '6 '3.

$7152».

,4 I" '.

«53.53) ,3}
'7

'F‘.

in

L5": 4:
33%»?! ‘Ifiq'.

.v
”'u’
, .

« u .
“123‘. “1' . ‘
”‘ 'ht ,? 12%.}: 'ds'
5“ egfigl‘l‘." ‘S" I

n.“

v1...

1

4%?-

mac' ‘ 2W ‘. ‘

' {13. "01, .- WIT
WWW?- LuJ} I "55:?
. 4:12- 531'? MGM"!

-. f “
..

ou-

253%} i kit, “"vm‘EIZ'f;

W.
‘52:

. 91‘
{“3”
11595:”,‘1‘ “imfl

'i'l‘yi '; |
r‘

‘M
0+ W 3"? )dt

 

W? WWWWWWW

3 1293 01688 5430

This is to certify that the
dissertation entitled
Identification of gibberellin-induced

genes in deepwater rice and the role of
these genes in plant growth

presented by

Esther Klazina Maria van der Knaap

has been accepted towards fulfillment
of the requirements for

PhD degree in Genetics

[44w weak

7
M ajor professor

Date May 13, 1998

 

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

 

 

LIBRARY

Mtchigan State
Universlty

 

 

 

PLACE IN RETURN BOX
to remove this checkout from your record.
To AVOID FINES return on or before date due.

 

MTE DUE MTE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1/98 chIRWDuopes-p.“

IDENTIFICATION OF GIBBERELLIN-INDUCED
GENES IN DEEPWATER RICE AND THE ROLE OF
THESE GENES IN PLANT GROWTH

By

Esther Klazina Maria van der Knaap

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

DOCTOR OF PHILOSOPHY

Program in Genetics

1998

ABSTRACT

IDENTIFICATION OF GIBBERELLIN-INDUCED GENES IN
DEEPWATER RICE AND THE ROLE OF THESE GENES IN
PLANT GROWTH

By

Esther Klazina Maria van der Knaap

Several gibberellin (GA)-regulated genes were identified by differential
display in the intercalary meristem of deepwater rice internodes. Some of
these genes encoded proteins that played a role in the cell cycle. Other genes
encoded proteins whose function in plant growth was not known. A GA-
induced increase in the transcript levels of a histone H3 gene correlated with
the increased rate of DNA synthesis. This suggests that histone H3 expression
is a marker for the S-phase of the cell cycle. Another GA-regulated gene with
sequence similarity to replication protein A1 (RPAl) was shown to
complement a yeast rpal mutant. The transcript level of Os-RPAI increased
earlier after GA application than that of histone H3, indicating a role for this
gene in the late G1 phase of the cell cycle. Os-RPAl is likely to play a role in
DNA replication during plant growth. However, it may also play a role in
coordinating gene expression that accompanies growth. A leucine-rich repeat
receptor-like kinase, OsTMK, was identified, and showed 59% amino acid
identity to TMKl from Arabidopsis thaliana. The expression of OsTMK was
very high in rapidly expanding regions, and transcript levels increased during
GA treatment. This suggests a role for this gene in plant growth. The kinase

domain of OsTMK was found to be an active, serine/threonine kinase. A

putative downstream component in OsTMK signaling was a kinase-
associated protein phosphatase (OsKAPP). It was shown in vitro that the
kinase-interaction domain of OsKAPP was a substrate for phophorylation by
OsTMK. The tissue-specific expression of the OSKAPP gene correlated well
with that of OsTMK, suggesting an in viva interaction. GA treatment also
resulted in increased transcript levels for OsDD3 and OsDD4. No homologs of
these genes were identified in the databases. However, amino acid sequence
features predicted that OsDD3 might be a type 1a receptor located at the
plasma membrane. OsDD4 was partially localized in the nucleus and could
regulate transcription of genes in the GA response pathway. The expression of
both genes was found to be higher in tissues containing meristems than in
other tissues, and OsDD4 seemed to be expressed exclusively in meristems.
The role of OsDD4 in plant growth was investigated by overexpression of this
gene in Arabidopsis thaliana. Transgenic plants were delayed in bolting, and
several inflorescence stems showed fasciation followed by bifurcation. The
severe phenotype showed flowering before bolting, reduced inflorescence
stem elongation and reduced apical dominance. The phenotype of the flowers
from the severe lines ranged from partially unfused carpels to flowers in

which all organs displayed leaf-like carpels.

To my parents and brothers

iv

ACKNOWLEDGEMENTS

I want to thank my advisor, Hans Kende, for his support, guidance and
for letting me develop this thesis project. My committee members: Frans de
Bruijn, Dave DeWitt and Mike Thomashow for their encouraging and
constructive comments along the way. Thanks to past and current Kende lab
members, as well as other members of the Plant Research Lab for insightful
discussions and help with various experiments. In particular Sandrine
Jagoueix, for an amazing job in search of more differentially expressed genes,
and Becca Wilford and Nate Krivitzky for your help with experiments.
Thanks to Jim Klug for keeping the rice growing even faster than under
submerged conditions; Kurt Stepnitz and Marlene Cameron for making
beautiful last minute slides and images; ”Sparkles” for emptying trash cans,
changing light bulbs and keeping me informed about conditions inside and
outside the building. My swimming buddies Beth, Antje, Nikki and Sigrun
for dragging me out of the lab and making me exercise (running around the
lab and biking year-round do not qualify!). Thank you Scott & Antje for being
dear friends and helping me start my life in Michigan. Art & Marlene, Kirsten
& Soren for camping and stamping parties. Sue, Steve & Dagan for the
dinner/ hot tub / hurt-stories on Sunday nights. Thanks Merideth for being
our favorite neighbor, the Monday night blues and the walks in the woods.
Frank, Marc, Anja, Ios & Riki, Isgouhi for keeping the friendships predating
graduate school alive. Thanks to my family for their unconditional love and

'Il

support (”we thought you working on plant nematodes ) and above all
thanks to Eric Stockinger for allowing me to be who I am and without whom

life would not be as special.

PREFACE

All experiments described in Chapter 2, e.g., RNA isolations, control
reactions, Northern blot analysis, differential display I, and library
construction, were performed by the author of this thesis. The differential
display II experiment and the cDN A library screen to obtain the full length
Os-DDIZ cDNA were performed by Dr. Sandrine Jagoueix.

With regard to Chapter 3, the Os-RPAI (Os-DDIZ) gene was originally
identified by Dr. Sandrine Iagoueix, who also generated the constructs for the
yeast complementation experiments. All other experiments, such as RNA
isolation, Northern blot analysis, and the yeast complementation were
performed by the author of this thesis. This chapter has appeared in the
Proceedings of the National Acadamy of Science of the USA 94:9979-9983.

With regard to Chapter 4, the OsKIN gene was originally identified by
Margret Sauter, and the experiment shown in Figure 4.2 was carried out in
collaboration with her. All other experiments, such as RNA and DNA
isolations, Northern and Southern blot analysis, library screening,
overexpression and phosphorylation studies were performed by the author of
this thesis.

With regard to Chapter 5, the DD4::GUS construct for nuclear
localization studies was constructed by the author of this thesis. The onion
bombardment experiments were performed by Emily Avila in Dr. M.
Varagona's laboratory at New Mexico State University in Las Crucas. All
other experiments, such as the generation of transgenic Arabidopsis plants,
were performed by the author of this thesis.

With regard to Chapter 6, all experiments were performed by the

author of this thesis.

vi

TABLE OF CONTENTS

LIST OF TABLES ............................................................................................................. ix
LIST OF FIGURES ............................................................................................................ x
CHAPTER 1
REGULATION OF SUBMERGENCE—INDUCED GROWTH IN DEEPWATER
RICE AND ROLE OF GIBBERELLIN IN VEGETATIVE GROWTH ..................... 1
1.1. Rice ......................................................................................................................... 1
1.1.1. Deepwater rice ............................................................................................ 2
1.2. Developmental and genetic analysis of rice internodal elongation ........ 3
1.3. Physiology of the elongation of deepwater rice internodes ....................... 5
1.4. Role of GA in plant growth and development ............................................ 7
1.4.1. GA biosynthesis .......................................................................................... 8
1.4.2. GA signal transduction ............................................................................. 9
1.4.3. GA-regulated gene expression during elongation ............................ 14
1.5. The anatomy of the rice internode and cell wall changes during
rapid growth .............................................................................................................. 20
1.6. Events initiated by GA in the IM of elongating rice internodes ............. 23
1.7 Literature ............................................................................................................. 26
CHAPTER 2

MODIFICATION OF THE DIFFERENTIAL DISPLAY METHOD AND
IDENTIFICATION OF DIFFERENTIALLY EXPRESSED GENES IN

GIBBERELLIN-TREATED DEEPWATER RICE INTERNODES .......................... 35
2.1. Abstract ................................................................................................................ 35
2.2. Introduction ....................................................................................................... 36
2.3. Materials and Methods .................................................................................... 38
2.4. Results ................................................................................................................. 51
2.5. Discussion ........................................................................................................... 65
2.6. Literature ............................................................................................................ 73

CHAPTER 3

EXPRESSION OF AN ORTHOLOG OF REPLICATION PROTEIN A1 (RPAl)

IS INDUCED BY GIBBERELLIN IN DEEPWATER RICE ...................................... 78
3.1. Abstract ................................................................................................................ 78
3.2. Introduction ....................................................................................................... 79
3.3. Materials and Methods .................................................................................... 81
3.4. Results ................................................................................................................. 84

vii

3.5. Discussion .......................................................................................................... 92
3.6. Literature ............................................................................................................ 94

CHAPTER 4
IDENTIFICATION OF A GIBBERELLIN-INDUCED LEUCINE-RICH REPEAT
RECEPTOR-LIKE KINASE IN DEEPWATER RICE AND ITS INTERACTION

WITH A KINASE-ASSOCIATED PROTEIN PHOSPHATASE ............................ 99
4.1. Abstract ................................................................................................................ 99
4.2. Introduction ..................................................................................................... 100
4.3. Materials and Methods .................................................................................. 102
4.4. Results ............................................................................................................... 107
4.5. Discussion ......................................................................................................... 124
4.6. Literature .......................................................................................................... 130

CHAPTER 5

A GIBBERELLIN-INDUCED GENE FROM DEEPWATER RICE CONFERS
DELAYED BOLTING, FASCIATED STEMS AND GROSS ALTERATIONS IN
FLOWER MORPHOLOGY WHEN OVEREXPRESSED IN ARABIDOPSIS....136

5.1. Abstract .............................................................................................................. 136
5.2. Introduction ..................................................................................................... 137
5.3. Materials and Methods .................................................................................. 138
5.4. Results ............................................................................................................... 143
5.5. Discussion ......................................................................................................... 166
5.6. Literature .......................................................................................................... 174

CHAPTER 6

A PUTATIVE TYPE 1a PLASMA MEMBRANE RECEPTOR IS INDUCED

BY GIBBERELLIN IN DEEPWATER RICE ............................................................. 178
6.1. Abstract .............................................................................................................. 178
6.2. Introduction ..................................................................................................... 179
6.3. Materials and Methods .................................................................................. 179
6.4. Results ............................................................................................................... 182
6.5. Discussion ......................................................................................................... 191
6.6. Literature .......................................................................................................... 196

CONCLUSION .............................................................................................................. 199

viii

Table 1.1.

Table 1.2.

Table 2.1.

Table 4.1.

LIST OF TABLES

GA signal transduction mutants ........................................................ 11
GA-regulated genes in vegetative growth ....................................... 15
Results of differential display analyses ............................................. 58

Amino acid identity between leucine-rich repeat receptor-
like kinases ............................................................................................ 113

ix

Figure 1.1.

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 4.1.

Figure 4.2.

Figure 4.3.

LIST OF FIGURES

Longitudinal median section through a 20-cm-long rice stem
section ....................................................................................................... 21

Flow chart for two differential display protocols used in this

study .......................................................................................................... 43
Differential display using different RNA concentrations or

PCR cycles ................................................................................................ 52
Differential display using 32P end-labeled primers ........................ 54

Differential display using RNA isolated by different methods
and reverse transcriptases .................................................................... 56

Differential display using RNA isolated from the IM of rice
stem sections treated for 0.5, 2.5, and 6.5 h with GA or kept in
water ......................................................................................................... 60

Northern blots showing changes in transcript levels of
differentially expressed genes in the IM in response to GA ......... 63

Amino acid sequence and phylogenetic analysis of DD12

(Os-RPAl) ................................................................................................ 85
Complementation of a yeast rpal mutant with pDSQ ................... 87
Tissue-specific expression of Os-RPAI in rice ................................. 89

Change in Os-RPAI transcript levels during GA treatment of
rice stem sections and during submergence of whole plants ....... 91

Sequence comparison between OsTMK, TMK1, and

AC000103 ................................................................................................ 109
Comparison of conserved features in the putative extracellular

domains of OsTMK, TMK1, and AC000103 .................................... 111
Expression of OsTMK in GA-treated rice stem sections .............. 115

Figure 4.4.
Figure 4.5.
Figure 4.6.
Figure 4.7.
Figure 4.8.
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 5.11.-

Southern blot analysis of OsTMK in rice ........................................ 116

Autophosphorylation of MBP-TKD ................................................ 118
Phosphorylation assay using GST-OsKID as substrate ................. 120
Decrease in autophosphorylation of MBP-OsTKD ....................... 122
Tissue-specific expression of OsTMK and OsKAPP in rice ......... 125
Amino acid sequence of OsDD4 ........................................................ 144
Alignment of QLQ and WRC motifs present in OsDD4 ............. 146
Expression of OsDD4 and D40170 in GA-treated rice stem

sections ................................................................................................... 149
Expression of OsDD4 in submerged rice plants ............................. 150
Tissue-specific expression of OsDD4 in rice ................................... 152
Southern blot analysis of OsDD4 in rice ......................................... 153

Overexpression of 05004 in T1 lines of A. thaliana
Columbia ............................................................................................... 156

Overexpression of OsDD4 in flowers of T1 lines of A.
thaliana Columbia ............................................................................... 159

Overexpression of OsDD4 in T1 lines of A. thaliana
Landsberg erecta ................................................................................... 162

OsDD4 copy number in T1 lines of A. thaliana Columbia .......... 165

OsDD4 transcript levels in T1 lines of A. thaliana

Columbia ............................................................................................... 167
Figure 6.1. - GA-induced expression of OsDDB in the IM of rice stem

sections ................................................................................................... 183
Figure 6.2. - Submergence-induced expression of the OsDD3, histone H3

and cchsI genes .................................................................................. 184
Figure 6.3. - Tissue-specific expression of OsDD3 in rice ................................... 187

xi

Figure 6.4. - Southern blot analysis of OsDD3 in rice ......................................... 188

Figure 6.5. - Amino acid sequence of OsDD3 ....................................................... 190

Figure 6.6. - In-vitro translation of OsDD3 ........................................................... 192

xii

CHAPTER 1

Regulation of submergence-induced growth in deepwater rice and

role of gibberellin in vegetative growth
1.1. RICE

Rice (Oryza spp.) is one of the most important food crops and a primary
food source for more than a third of the world's population. More than 90%
of the world's rice is grown and consumed in Asia, where rice in the diet can
account for over 70% of the daily calorie intake (Hossain, 1996; Khush, 1997).
Rice is a major factor in national economies as well, providing up to 30 to
50% of the agricultural value added in the poorer countries in Asia (Hossain,
1996)

The genus Oryza comprises about 20 different species and belongs to a
group of ancient grasses. Two major cultigens, the Asian 0. sativa and the
African 0. glaberrima, together with their wild progenitors O. rufipogon, O.
nivara and O. longistaminata, O. bartii, respectively, belong to the AA
genome (2n=24) of the O. sativa complex (Oka, 1988). Cultivation of rice for at
least 7000 years (Oka, 1988) has resulted in the generation of ~ 120,000 distinct
rice varieties (Khush, 1997). These cultivars display several distinctive
characteristics and show more genetic and morphological diversity than most
other cultivated crops. Rice is cultivated in five different ecosystems in which
the flooding pattern (i.e., depth and duration) is a major determinant. The
ecosystems are irrigated, rainfed lowland, deepwater, upland, and tidal
wetlands (IRRI, 1984). The most productive rice growing system is the
irrigated environment, yielding 75% of the world's rice production (IRRI,

1990). This high productivity is mainly achieved by using high-yielding,

short-stemmed modern varieties, which are also highly responsive to
fertilizer. The productivity in the other ecosystems is lower, in part, because
the high-yielding modern rice varieties do not tolerate less-controlled water

regimes.

1.1.1. DEEPWATER RICE

Rice grown in the deepwater ecosystem distinguishes itself from most
modern rice varieties by its ability to survive in water depths of 50 cm or
more for at least one month during the growing season (Catling et al., 1988).
Deepwater rice areas in Asia are found in the floodplains and river deltas of
the Ganges-Brahmaputra in North-East India and in Bangladesh, of the
Irrawaddy in Myanmar, of the Chao Phraya in Thailand, and of the Mekong
in Vietnam and Cambodia. Although the rice production from deepwater
environments is only 6% of the worldwide production (IRRI, 1990), it
provides the major calorie intake of 100 million people living along the
rivers and is the only crop that can be grown by subsistence farmers during
the monsoon season (Catling, 1992).

The two strategies for survival of rice under deepwater conditions are
tall stature and ability to elongate. Two groups can be distinguished among
the deepwater rice varieties: traditional tall and floating rices. The traditional
tall rices grow in water depths between 50 to 100 cm and are relatively tall
even when not flooded. Because they have limited or no ability to elongate
when submerged, traditional tall rices are not suited for sustained water
depths of more than 100 cm. Under these circumstances, the elongation
ability of the floating rices is essential (Catling, 1992). When well established,

floating rices are able to elongate at rates of up to 25 cm per day, resulting in

plants that are up to 7 m tall (Vergara et al., 1977; Catling, 1992). Almost all
wild rices show floating ability, often exceeding that of the best elongating rice
cultivars. Therefore, the elongation ability was probably retained from wild

progenitors (Oka, 1988).

1.2. DEVELOPMENTAL AND GENETIC ANALYSIS OF RICE
INTERNODAL ELONGATION

Post-embryonic plant development can be divided in a juvenile and an
adult phase. The adult phase is further divided in a vegetative and
reproductive phase (Catling, 1992). The juvenile phase starts after
germination and lasts until the plant is established (3 to 5 weeks after
germination). The vegetative adult phase follows the juvenile phase and lasts
until the beginning of the reproductive phase, the length of which varies
with each cultivar and is often influenced by day length. The reproductive
phase begins at the initiation of panicle formation and lasts until flowering.
Elongation growth in both the juvenile phase (Raskin and Kende, 1983) and
the reproductive phase (Morishima, 1975; Vergara, 1985; Keith et al., 1986) are
similar in deepwater and modern rice varieties. For example, elongation at
the onset of the reproductive phase occurs and is further increased by
submergence in all rice cultivars (Keith et al., 1986). What distinguishes
deepwater rice from most modern rice varieties is the vegetative adult phase.
The ability to grow rapidly out of the water when partially submerged is
primarily attributed to enhanced internodal elongation. Modern rice varieties
and most traditional tall varieties do not respond at this stage to rapidly rising

water levels, and partial submergence severely damages the plants.

The genetic basis for enhanced internodal elongation has received little
attention. Studies have shown that this trait is controlled by a few major and
many minor genes (Catling, 1992). Suge (1988) has indicated that elongation
during submergence is based on the ability of an intemode to elongate, as well
as the degree of elongation. In this study, one dominant gene was identified
conferring elongation ability (in addition to a locus controlling gibberellin
[GA] biosynthesis; Suge, 1988; see below). In another study, one dominant
gene responsible for early nodal differentiation was identified (Tripathi and
Balakrishna Rao, 1985). This trait was linked to nodal rooting and tillering,
which are typical characteristics of floating rice. Whether the same locus was
identified in both studies is not known.

Another locus for elongation ability was identified in a study where a
modern non-elongating rice cultivar was used as the recurrent parent for
introgression of a locus from a perennial O. rufipogon accession. O. rufipogon
is the progenitor of O. sativa, exhibiting great floating abilities. The near-
isogenic line thus created indicated the presence of a recessive allele
conferring internodal elongation in response to submergence in the
vegetative adult phase (Eiguchi et al., 1993). These authors may have
identified a heterochronic gene that changes the timing of internodal
elongation depending on environmental conditions. This is a major trait
distinguishing modern, non-elongating from elongating varieties
(Morishima, 1975; Keith et al., 1986). The above results indicate the presence
of two or three major genes conferring internodal elongation ability.
However, to survive 2- to 4-m-deep floodwaters, the degree of elongation, i.e.,
rapid increase in growth rate and total elongation, is essential and the degree

of elongation is probably controlled by many genes (Suge, 1988).

Yields of modern rice cultivars average 6 tons per hectare, whereas
those of deepwater rice are 2 to 3 tons per hectare (Catling, 1992; Setboonsarng,
1996). The efforts to improve the yield and grain quality of deepwater rice
varieties have been met with limited success. Breeding for short-stemmed
rice varieties with conditional elongation ability (possibly involving the loci
described above) has resulted in the generation of varieties that fared well
under moderate flooding conditions (up to 150 cm; Catling, 1992). However,
increases in yield and retention of the floating ability has never been
combined in one single cultivar. Improvement of floating rice cultivars,
which represent the most extreme example of adaptation to flooding, may not
be feasible according to breeders (Catling, 1992) and economists (Setboonsarng,
1996). Studies on how rapid increases in rates of elongation are achieved in
these cultivars will be very useful, however, to improve cultivars grown in

other ecosystems.

1.3. PHYSIOLOGY OF THE ELONGATION OF DEEPWATER
RICE INTERNODES

Partial submergence of floating rice varieties leads to rapid increases in
the growth rates of rice internodes (Métraux and Kende, 1983; Vergara et al.,
1977). This increased growth is also observed when plants are exposed to 1
ul/l ethylene, which is the natural concentration found in submerged
internodes (Raskin and Kende, 1984a; Stfinzi and Kende, 1989). Moreover,
plants treated with aminoethoxyvinylglycine, an inhibitor of ethylene
biosynthesis, do not show internodal elongation upon submergence (Métraux

and Kende, 1983; Raskin and Kende, 1984a). Ethylene levels increase in

submerged internodes, and this rise precedes internodal elongation (Rose-
Iohn and Kende, 1985). The initial increase in internodal ethylene levels is
due to accumulation from basal production and the subsequent entrapment
of the gas under submerged conditions (Zarembinski and Theologis, 1997).
Long-term increases in ethylene levels are likely to be due to enhanced
ethylene biosynthesis. Enzyme and transcript levels of the gene encoding 1-
aminocyclopropane—l-carboxylate (ACC) oxidase increases several fold after 8
h of submergence (Mekhedov and Kende, 1996). ACC synthase 1 transcript
levels marginally increases after 12 h of submergence only in the growing
region of the internode, while in other regions of the internode the levels
decreases (Zarembinski and Theologis, 1997). This indicates that the increase
in ethylene biosynthesis is primarily due to increased transcript levels and
increased activity of ACC oxidase, the terminal enzyme in ethylene
biosynthesis (Mekhedov and Kende, 1996).

The internode that can grow and respond rapidly to rising water levels
is the uppermost internode. Rice stem sections containing the uppermost
internode respond to submergence and high ethylene concentrations in a
manner similar to that observed in whole plants (Raskin and Kende, 1984a).
The growth response of rice stem sections induced by submergence or
ethylene is blocked when the plants and sections are treated with the GA
biosynthesis inhibitor tetcyclacis (Raskin and Kende, 1984b), or with
ancymidol (Suge, 1985), prior to ethylene treatment. Addition of GA to the
tetcyclacis-treated sections restores the growth response. GA is more effective
in inducing growth in an atmosphere containing ethylene than in air,
indicating that ethylene increases the sensitivity of the tissue to GA (Raskin
and Kende, 1984b). The enhancement of the sensitivity of the tissue to GA is

thought to be mediated through a decrease of the abscisic acid (ABA) levels in

the plant. ABA has been shown to antagonize the growth-promoting action
of GA in rice stem sections (Hoffmann-Benning and Kende, 1992). Whole
plants submerged or treated with ethylene for 3 h show a 75% reduction in
ABA levels; therefore, the responsiveness to GA may be a function of ABA
content (Hoffmann-Benning and Kende, 1992). The induction of growth by
either submergence, ethylene or GA has been measured with an angular
transducer attached to the growing internode. Growth induced by
submergence of rice stem sections shows a 3 h 20 min lag phase. The lag phase
is 60 min for treatment with ethylene and low 02 (Rose-John and Kende,
1985) and 40 min for GA (Sauter and Kende, 1992b). Furthermore, using a rice
GA biosynthetic mutant line, GA has been shown to be a prerequisite for
elongation ability during submergence (Suge, 1988). Therefore, the
physiological and genetic studies and the short lag phase all indicate that GA
is the ultimate hormone eliciting the growth response. Submergence
increases the level of ethylene, which enhances the sensitivity of the
internode to GA, at least in part through a decrease in ABA content. This
effect is not unique to rice. Gibberellins have also been shown to mediate the
stimulatory effect of ethylene on stem elongation in other semi-aquatic plants

(Musgrave et al., 1972; Rijnders et al., 1997).

1.4. ROLE OF GA IN PLANT GROWTH AND DEVELOPMENT

Gibberellins were first identified and purified from the fungus
Gibberella fujikuroi which causes infected rice plants to grow excessively tall
(Yabuta and Sumiki, 1938). Gibberellins have been implicated in many plant

processes, such as germination and mobilization of reserves immediately

after germination of cereal grains, juvenile-to-adult and vegetative-to-
reproductive phase changes, apical dominance, flower initiation, in anther,
seed and fruit development, sex determination and, above all, shoot and leaf

elongation.

1.4.1. GA BIOSYNTHESIS

The role of GA in plant growth and development has been
documented in several plant species via the generation of dwarf mutants,
whose phenotypes are reversed upon GA treatment. The characterization of
these mutants has also allowed the elucidation of the GA biosynthetic
pathway. Recently, many of the GA biosynthetic genes have been cloned. GA
biosynthesis branches from the general terpene biosynthetic pathway at
geranylgeranyl pyrophosphate (GGPP). The first committed step is the
formation of ent-kaurene from GGPP, which is catalyzed by ent-kaurene
synthases A and B (KSA and KSB, respectively). The gene encoding KSA has
been identified from Arabidopsis (Sun et al., 1992) and the gene encoding KSB
from pumpkin (Yamaguchi et al., 1996). Transcript levels of the KSA and KSB
genes is detected in immature and growing regions of the plant, and the
enzymes are most likely located in proplastids of meristematic cells
(Silverstone et al., 1997a; Aach et al., 1997). The next steps in the biosynthetic
pathway are the stepwise oxidation of ent-kaurene to GA12-aldehyde by
membrane-bound P450 monooxygenases. The final step is the conversion of
GA12-aldehyde to bioactive GAs (GA1, GA3 and GA4) and is mediated by at
least two enzymes which belong to the class of 2-oxoglutarate-dependent
dioxygenases located in the cytosol (Kende and Zeevaart, 1997). GA 20-oxidase

is responsible for the oxidation and subsequent elimination of C-20, a pivotal

step in the pathway. Several genes encoding GA 20-oxidases have been cloned
from pumpkin (Lange, 1997), spinach (Wu et al., 1996) and Arabidopsis (Xu et
al., 1995; Phillips et al., 1995). These enzymes were shown to have broad
substrate specificity. Genes encoding GA 20-oxidases are part of a gene family,
and Northern blot analysis has indicated tissue-specific expression patterns
for three of the Arabidopsis GA 20-oxidase genes (Phillips et al., 1995).
Application of GA results in a decrease in transcript levels, indicating
negative feedback regulation (Phillips et al., 1995). Feedback regulation of the
gene encoding 3B-hydroxylase, the last step in GA biosynthesis, has been
observed in Arabidopsis (Chiang et al., 1995) and in pea as well (Martin et al.,
1997). While 38-hydroxylation is essential in the formation of bioactive GA,
the 28-hydroxylation irreversibly inactivates GA. The "slender" mutant of
pea is the only mutant known to be blocked in catabolism of bioactive GA.
However, the gene encoding this 2B-hydroxylase has not yet been identified

(Ross et al., 1995).

1.4.2. GA SIGNAL TRAN SDUCT ION

Plant hormone responses are mediated by both the amount of
hormone and by tissue responsiveness. The ability of GA to induce growth in
pea internodes, for example, has been correlated with the amount of GA, as
well as with tissue responsiveness (Ross and Reid, 1992). The importance of
tissue responsiveness to GA is also evident in the internodal elongation of
rice, where modern varieties are unresponsive to both submergence and GA
during the vegetative adult phase (Inouye, 1985; Keith et al., 1986, Eiguchi et

al., 1993). Whereas much is known about GA biosynthesis (see above), less is

known about GA signal transduction. Recently, however, significant progress
has been made in this area of research.

There are a few mutants with altered stature whose phenotypes appear
to correlate with an increase or a decrease in GA signal transduction (see
Table 1.1). In case of these mutants, many or all of the pleiotropic effects of GA
or the lack thereof (i.e., effects on germination and male fertility) have been
observed. The phenotype of an increase in GA signaling resembles that of a
wild-type plant treated with saturating doses of GA. All mutants disturbed in
this process are recessive, indicating a loss of function of a repressor-like
molecule. The sln mutant in barley (slender; Chandler, 1988) and the la cryS
double mutant in pea (Potts et al., 1985) show no change in height regardless
of endogenous GA levels. Both loci in pea, LA andCRY, need to be mutated
to confer the slender phenotype. The spy mutant (spindly) in Arabidopsis
confers a slender phenotype as well, but this mutant is somewhat responsive
to GA. The spy mutant was identified in a screen for GA-independent
germination by using an inhibitor of GA biosynthesis, paclobutrazol (Jacobsen
and Olszewski, 1993). The SPY gene was found to encode a protein containing
a tetratricopeptide repeat, which may be involved in protein-protein
interactions, and a domain with O-linked N-acetylglucosamine transferase
activity (Jacobsen et al., 1996; Olszewski, 1997). Several mutant alleles of SPY
have been identified and indicate the importance of both the tetratricopeptide
repeat as well as the O-linked N-acetylglucosamine transferase activity in GA
signaling (Olszewski, 1997). Whether the SLN (barley) or LA and CRY (pea)
genes encode SPY orthologs is unknown.

The mutants which display a decrease in GA signaling are dwarf in
appearance, reminiscent of GA deficiency. All mutants are semi-dominant or

dominant and do not respond to applied GA. However, paclobutrazol further

10

 

mam?

 

 

.o___ZoEom ucm com=>> m $385864 Em Co 539qu mcmm
moor 4m um 9.99025 6 .903 cozatomcm: $3023; mémm Co 53659 «99 m9
.3?
55.8.“. new 52E; 6 Eco o>:_mcomc_-<0 mp
mm? {m E 2mm w “mos; o>=_mcomc_-<0 SE
39 :E Co 95¢ a 58$ cozatomcmz flamingo. o>_=mcomc_-<0 .ch
mwmr :E Co gen. a mom Enema «Co E
33 .56:ch 8 >253 Enema Sm
33 ._xm2,m~m_o ommeohwcmz oEEmmooagbmom Emu new mémm
”0mm? gm .8 588.2. .2 “393:6 6cm “mono. 85me $3023; Co 53069 .5956 En
mocmcmymm 8:82 5305 836:3 $6on oazococd E922
.88? 58>

emu cm> new Coochoxv EmSE twin ozmcoamo. <0 .m.m>om m m_ mémm .mEmSE cozoznmcmb .mc9m <0 é... 29a...

11

decreases the height in the plants, which is reversible upon GA treatment.
This indicates that the ability to respond to GA exists, but also that the
response is saturated under physiological conditions. Genetic analysis has
shown that the d8 locus in maize (dwarf 8; Winkler and Freeling, 1994), rht3
in wheat (reduced height 3; Gale et al., 1975) and gai in Arabidopsis (GA-
insensitive; Wilson and Somerville, 1995) are "gain-of-function" mutants.
Recently, the GAI gene has been identified (Peng et al., 1997), and found to
have high amino acid similarity to the protein encoded by the SCR
(SCARECROW) gene. SCR is required for asymmetric cell division during
root development in Arabidopsis (Di Laurenzio et al., 1996). The proteins
encoded by the GAI and SCR genes may belong to a new class of plant
transcription factors (Peng et al., 1997). A gene related to GAI is GRS and the
proteins encoded by these genes are 83% identical. Interestingly, both the GAI
and RG3 genes were identified previously in a screen for Arabidopsis genes
able to rescue a yeast strain deficient in the production of a transcription factor
regulating nitrogen metabolism (Truong et al., 1997). It is unknown whether
there is a link between nitrogen metabolism and GA signaling.

A screen for suppressor mutations of gai have resulted, in many cases,
in the generation of intragenic suppressor mutants, indicating that the wild-
type function of the GAI protein is dispensible (Peng and Harberd, 1993;
Wilson and Somerville, 1995). However, two extragenic loci have been
identified, that partially suppress the gai phenotype. The gasl mutant gene
(gai suppressor) is an allele of SPY, while garZ (gai revertant) appears to be a
mutated allele at a new locus involved in GA signaling (Wilson and
Somerville, 1995; Carol et al., 1995; Peng et al., 1997). Interestingly, the spy
mutant has not only been identified as a suppressor of gai, but also as

suppressor of gal-3, a severe GA biosynthesis mutant in Arabidopsis

12

(Silverstone et al., 1997b). In the latter study, a mutant allele at another locus
was identified, rga (repressor of gal-3; Silverstone et al., 1997b). The gene
affected in the rga mutant appears to be GRS, previously identified because of
its high amino acid sequence similarity to GAI (see above; Silverstone et al.,
1998). In the gal-3 mutant, the recessive mutant alleles at the RCA (GRS)
locus partially restores stem growth and apical dominance to wild-type levels,
while germination and male fertility are not restored (Silverstone et al.,
1997b). This indicates that RGA (GRS) affects only part of the GA response
pathway. The spy mutation on the other hand, partially restores all defects
caused by GA deficiency (Silverstone et al., 1997b).

The studies described above have shown that GA most likely controls
stem elongation not through activation of the signal transduction pathway
but by derepression. The repressor function of GAI may be relieved by GA,
while the mutant protein has lost the ability to be derepressed. The SPY
protein appears to be a negative regulator of GA signaling and the spy
mutation is epistatic to the gai mutation (Wilson and Somerville, 1995;
Iacobsen et al., 1996). At the molecular level, it is possible that SPY post-
translationally modifies GAI and RGA (GRS), thereby altering the activity of
these putative transcription factors (Olszewski, 1997; Peng et al., 1997;
Silverstone et al., 1998). While the spy mutant partially reverts gai to wild-
type, the triple homozygous mutant, gai spy garZ, completely suppresses the
phenotype displaying a decrease in GA signaling (Peng et al., 1997). The
phenotype displayed by the GA biosynthesis mutant gal-3 is also completely
suppressed by additional mutations at the SPY and RCA (GRS) loci
(Silverstone et al., 1997b). These studies indicate that stem elongation in both

homozygous triple mutants appears to be mediated independently of GA.

13

1.4.3. GA-REGULATED GENE EXPRESSION DURING
ELONGATION

Complementary to genetic analyses of GA-regulated processes is the
identification of genes whose expression is specifically regulated by GA.
Previously, two-dimensional gel electrophoresis studies have revealed
changes in protein profiles of GA-deficient maize and pea shoots after
treatment with GA (Chory et al., 1987). While these latter studies have led to
the identification of several proteins whose level increase or decrease, a later
study, using the stems of a GA-deficient tomato mutant line have displayed
only proteins whose level decrease upon GA treatment (Jacobsen et al., 1994).
Using leaves of a GA-deficient barley plant, two (Speulman and Salamini,
1995) or no (Zwar and Chandler, 1995) proteins have been identified, which
change in abundance after GA application.

Differential screening methods have been used to identify genes
regulated by GA (see Table 1.2). The first gene identified in stems of a GA-
deficient tomato line is GASTI (GA stimulated transcript 1). GASTI transcript
levels accumulate after GA treatment, while ABA, auxin and cytokinin have
no effect (Shi et al., 1992). The GASTI gene encodes a putative 112-amino acid
protein containing a signal sequence, which indicate that it may be secreted
(Shi et al., 1992). A Petunia homolog, GIP (GA-induced gene), has been
identified and the protein it encodes shows 82% amino acid identity with
GASTl including the signal sequence. High levels of GIP transcript are
present in elongating corollas and young stem internodes, and their level
increases further by GA treatment (Ben—Nissan and Weiss, 1996). Another
homolog of GASTl is RSI-1 (root system inducible-1; 79% amino acid identity

without signal sequence) from tomato (Taylor and Scheuring, 1994). The gene

14

mam?

 

._c_Em_mm ccm cmEBmam ”Em 2 3_Em__E_m 3:8 Eo_ocmn-<0 mmmobE Emacmz 3mm
82 4m a. 5:5 Ex 9 SEE; 3:2 220.335 8855 Ea
82 4m 5 5:5 Ex 5.8 2283-6 3852 :kxm
59 :E Co ccoo m. mom Eo_o:ou-<0 ommEoE Emacmb UTOmQ
vmmr 4m Cm £00 8 won Em_ocou-<0 ommmbE Emacm: «rwmQ
mam? gm. Cm 3>> owmto>£ mma Eo_ozmn-<0 mmmmbE Eoficm: 39:95

£903
3? i=5: ucm 85:5 _mccmco .662, 623:2 mxmqonsmcq. Embaoutqo mmmocos mix
82 a a mesa: £96 2 35:55 emqousmé 2288-5 3855 Emacs“ $36
82 .a do Bum: F55 2 33:56 maqonsmé 220:8th $855 Emacs; 39$

82

.mm_o>> new 532-com :95 9 3_.m__E_m $2:th ommmbE Q5
Nam? :E Co Em £205 “326.com BmEB Eo_ocon-<0 owmocoE Emficm: Emiw
mocoemhmm 5382 59.0.5. 6:: EmE w_m>m_ acumen; meow

 

.5265 o>=£omm> E $ch 69939-4.6 .m... 633

15

 

8? 4m a 853;
82 :3 a 8____;n_
82 gm E 352a

mam?
.EmBoNflo new cmwnoomw

mmmr
._v_m>>mNm_O ncm 5808..

0mm P
._xmam~m_0 new cmmnoomw

952 .283. new 96
£2: .288. 2m 26

mam?
.mucmx ucm 3ch Eu cm>

mam? 4m 3 55mm
5mm? gmfimw
5mg cmuzmm
ommw :E E 0:050

39
._c_Em_mm ucm :95:QO

$85-8 <0
35x98 <0

mmmExoém <0

8359.3ch
.9305 2 3_.m__E_m

wmmmcmgxofi Emucmamu
-mum.m~:_moxo-w 2 btgém

22525
$2653 9 3:386

Emcmaxm

Emcmaxm

m: 95E:

mono 2 3tm__E_m
c__o>o 032::
5.28 032::

w

5.. 01:52.6

wagonsmé‘ “$28-5
muqousmé 2283-5

wagonsmé‘ Em_o%u-<w

29:2 Em_ocmu-<o

99:0... Em_ocmu-<0

99:8 Em_ocmu-<0
mo: .mfizfimmv

mo: .mumzammn

mo: hmfizammu
mo: .Bmzammn
mo: 552503
8: 56383

538308

>259 Em_oc%-<o

3.980%
$383

$883

mwmmbmn

mmmmbmu

mmmmbwn
@8905

$862

$8.65

$399:

$885

3980:.

$885

mwmmgoE “C0559:

©0513»
mmmwzx

5mm;

mQ<®

NQ<®

ch‘O
VQXméO

NQXWéO

mm @220 mom
«-mOmobo
mmOQG
3098

SEED

N<mm

16

encoded by RSI-1 is identified because of increase in transcript levels in lateral
root primordia, especially after auxin treatment. While GASTI is
predominantly expressed in the shoot, RSI-1 is predominantly expressed in
the root. Whether GA plays a role in RSI-1 transcript accumulation is
unknown. Four less related genes have been identified in Arabidopsis,
GASAI to GASA4 (Herzog et al., 1995). These genes display a tissue-specific
expression pattern, and transcript levels of GASAI and GASA4 are transiently
increased by GA. The sequence similarity between GASAI, -2, -3, and -4 on the
one hand, and GASTI, RSI-1 and GIP on the other hand, is restricted to the C—
terminal half of the protein. However, all putative proteins are predicted to
contain a signal sequence, and hydrophobic cluster analysis indicates the
presence of similar domains in all of them (Herzog et al., 1995).

In seedlings of a (EA-deficient pea line, increased transcript and enzyme
activity levels of an invertase have been observed after GA treatment (Wu et
al., 1993), as well as a marginal, transient increase in the transcript levels of
two genes, PsG-A and PsG-F, whose sequences are not yet known (Cohn et al.,
1994). GA treatment of (SA-deficient Arabidopsis plants have been found to
result in the increase in transcript levels of a gene encoding a tonoplast water
channel protein, y-TIP (Phillips and Huttly, 1994). In the leaf blades of a GA-
deficient barley line, an increase in transcript levels of two genes whose
encoded proteins share sequence similarity to xyloglucan
endotransglycosylase enzymes (XET; Smith et al., 1996), EXTll and PMZ, have
been observed after GA treatment. During rapid growth, cells expand and
have to maintain turgor pressure. Invertases and water channel proteins are
implicated in this process. As will be described below, XET may be involved
in cell wall expansion. In another study, transcript levels of two genes, BSA]

and ESAZ, accumulate in response to GA in the leaf blades of (SA-deficient

17

barley plants (Speulman and Salamini, 1995). Interestingly, ESAI has
potentially two ORFs, one of which has similarity to epidermal growth factor.
ESAZ has similarity to LEA (late embryogenesis abundant) proteins induced
during dehydration and by ABA. The ESAI and ESAZ genes are, however,
neither auxin nor ABA regulated (Speulman and Salamini, 1995). In
cucumber seedlings, a gene has been identified, CRG16, whose transcript
levels increase marginally after GA application. No amino acid sequence
similarity to known proteins was found. However, the promoter of this gene
contains a putative GA responsive element (GARE) found in promoters of a-
amylase genes whose transcription is induced by GA in the aleurone of cereal
grains (Chono et al., 1996).

The genes decribed above show an increase in transcript levels after GA
treatment. Jacobsen and Olszewski (1996) have identified three genes (GA
down: GADI to GAD3) whose transcript levels decrease upon GA treatment.
The protein encoded by the GADI gene shows similarity to proteinase
inhibitors, and transcript levels of this same gene are found to be increased
during auxin-induced lateral root initiation as well (Taylor et al., 1993). GADZ
has similarity to 2-oxoglutarate—dependent dioxygenases but is not related to
GA 20-oxidases or SIS-hydroxylases involved in GA biosynthesis. GAD3 has
similarity to alcohol dehydrogenases. The decrease in transcript levels of
these three GAD genes is not specific to GA; auxin and ethylene also decrease
the transcript levels of the CAD genes (Jacobsen and Olszewski, 1996). As was
described above, a decrease in transcript levels is also observed for three genes

encoding GA ZO-oxidases involved in GA biosynthesis (Phillips et al., 1995).

Plant elongation is not only regulated by GA but also by other plant

hormones and environmental factors. This is particularly well illustrated by

18

mutants in pea (lkb and lk) and tomato (d), which were thought to belong to
the reduced GA signaling class but have recently been shown to have defects
in brassinolide biosynthesis (BishOp et al., 1996; Nomura et al., 1997). Light
has a profound effect on elongation via the action of phytochrome and some
phytochrome mutants look like GA-treated wild-type plants. Dose-response
curves with Arabidopsis hypocotyls show that the double mutant phyB gal-3
is more responsive to applied GA than the single mutant ga1-3, indicating
that the phyB mutant magnifies the response to GA (Reed et al., 1996).

It is important to carefully select the system in which to study GA-
regulated processes. Most of the gene expression studies described above were
done with GA-deficient seedlings. In seedlings, many developmental
processes are taking place concurrently, and several plant hormones, as well
as environmental factors, play an intricate role in the outcome. Also,
treatment of "severe" GA-deficient mutants with GA may have a similar
effect as removal of hydroxyurea or colchicine from cell cultures to release the
block of progression through the cell cycle. Suddenly, several genes
transiently change their expression pattern of which many may have nothing
to do with the GA-regulated process per se.

Knowledge gained from one system cannot automatically be applied to
another system. For example, both barley leaf blades and leaf sheaths show
increased elongation upon GA treatment. However, the transcript levels of
the ESAI and ESAZ genes increase only in leaf blades (Speulman and
Salamini, 1995). Promotion of growth in cucumber leaves is clearly promoted
by GA, but transcript levels of the CRG16 gene do not increase in leaves as
they do in hypocotyls (Chono et al., 1996). Lastly, transcript levels of the same

gene can increase in young roots and hypocotyls (Taylor et al., 1993), while

19

decreasing in more mature tissue in response to the same stimulus (Jacobsen
and Olszewski, 1996).

Deepwater rice internodes in the vegetative adult phase are induced to
elongate rapidly during partial submergence. The increase in elongation rate
is, ultimately, mediated by GA, while auxin, another growth-promoting
hormone, has no effect in these internodes. The following paragraphs
describe in detail the analyses of submergence- and GA-mediated rapid

growth response in deepwater rice internodes.

1.5. THE ANATOMY OF THE RICE INTERNODE AND CELL
WALL CHANGES DURING RAPID GROWTH

Based on cell size measurements and [3H]thymidine incorporation
studies (Métraux and Kende, 1984; Bleecker et al., 1986), the internode can be
divided into three regions, as shown in Figure 1.1. At the base of the
internode is the intercalary meristem (IM), 2-3 mm in length where cells
divide; the elongation zone (EZ) where cells reach their final length; and the
differentiation zone (DZ) where secondary wall and xylem formation takes
place. The high growth rates attained upon submergence, ethylene or GA
treatment are achieved by a 3-fold increase in the rate at which cells are
produced in the IM and a 3- to 4-fold increase in the final length of cells in the
EZ (Raskin and Kende, 1984b; Métraux and Kende, 1984; Bleecker et al., 1986).

Upon submergence, differences in cell wall composition are observed
between newly elongated parts of the internode and pre-existing parts or air-
grown internodes (Rose-John and Kende, 1984; Azuma et al., 1996). Primary

cell walls consist of cellulose microfibrils interconnected by matrix

20

Leaves

 

First node

 

Differentiation
zone

Elongation zone

Intercalary meristem
Second node

20 mm

 

Figure 1.1. Longitudinal median section through a 20-cm-Iong rice stem section.
The second node is separated from the first node by the youngest internode.
The locations of the intercalary meristem, the elongation zone, and the differen-
tiation zone are indicated along the internode. The stem section above the first
node consists of leaf sheaths and developing leaves. On the right are scanning
electron micrographs of parenchyma cells from a stem section treated with 50
pM GA3 for 24 h. The top micrograph are cells from the differentiation zone, the
middle micrograph are cells from the elongation zone and the bottom micrograph
are cells from the intercalary meristem. From Sauter and Kende, 1992b.

21

polysaccharides. The direction of cell growth is determined by the
arrangement of the cellulose microfibrils which, in turn, is determined by the
orientation of cortical microtubules. Elongation is favored when the the
cellulose microfibrils are oriented transversly to the direction of growth. In
rice internodes, the orientation of the cellulose microfibrils is transverse in
the growing zone, and this orientation does not change in the parenchyma
cell walls even outside the growing region. In the outer epidermal cell walls,
however, the orientation is oblique in the non-growing zones, indicating that
the epidermis is a growth-limiting structure (Kutschera and Kende, 1988;
Sauter et al., 1993). In GA-treated rapidly growing internodes, the change from
transverse to oblique is more gradual and extends over the length of the
elongation zone (Sauter et al., 1993). One of the major matrix polysaccharide

' in monocots is (1->3, 1->4)-B-glucan, which is thought to be involved in cell
growth (Carpita and Gibeaut, 1993). Higher levels of B-glucan are found in cell
walls of elongating cells than in walls of non-growing cells. Submergence
greatly extends the EZ as well as the zone of high B-glucan levels (Sauter and
Kende, 1992a). In contrast, lignin content and activity of coniferyl alcohol
dehydrogenase and phenylalanine ammonia-lyase, two enzymes in the lignin
biosynthesis pathway, are up to six-fold lower in rapidly growing rice
internodes than in air-grown rice internodes. It seems that the process of
lignification is suppressed to permit rapid growth (Sauter and Kende, 1992a).
The above results indicate that GA extends the E2 in rice internodes by
delaying cell wall processes involved in cessation of growth and permitting
cell wall processes that favor growth to continue over a longer distance. This
effect of GA has also been observed during wheat leaf blade elongation, where
the initial relative elemental growth rate (REGR) and maximal REGR remain

unchanged whether GA is applied or not (Tonkinson et al., 1995). However,

22

the extension zone beyond the point of maximum REGR is extended in the
presence of GA as compared to control. The addition of paclobutrazol, an
inhibitor of GA biosynthesis, decreases the extension zone further; this effect
is reversed by the addition of GA, indicating an important role for GA in
extending the EZ (Tonkinson et al., 1997). Although GA increases the length
of the EZ and the length of the cells in the EZ, it cannot change the fate of cells
already in the EZ at the start of GA treatment. For example, GA does not
reorient cellulose microfibrils from oblique to transverse in rice internodes.
Furthermore, stem sections in which the IM has been removed do not
respond to GA treatment (Sauter et al., 1993). Thus, the tissue that responds to
GA is the IM, and only cells that move out of the IM into the EZ during GA

treatment will increase further in size.

1.6. EVENTS INITIATED BY GA IN THE IM OF ELONGATING
RICE INTERNODES

Research in deepwater rice has shown several early effects of GA-
induced growth. After 2 h of GA application, the cells in the IM have
increased their average length (Sauter and Kende, 1992b). Cell elongation is
driven by water uptake into the central vacuole. In rice, increased internodal
elongation after submergence is not due to increased hydrolic conductance or
decrease in osmotic potential, but to changes in yielding properties of the cell
wall (Kutschera and Kende, 1988). The stress relaxation of the cell wall is
manifested by an increase in extensibility after submergence (Kutschera and
Kende, 1988; Cho and Kende, 1997a). GA also increases cell wall extensibility

as has been shown in oat internodal segments (Adams et al., 1975) and in

23

immature wheat leaves (Keyes et al., 1990). Relaxation of cell wall stress is
thought to result from the action of cell wall-loosening factors. Amongst
them are cell wall enzymes that break covalent bonds, such as hydrolases and
endotransglycosylases. In particular, the xyloglucan endotransglycosylase
enzyme (XET) has been implicated in the cell wall loosening process (Fry et
al., 1992) because of a correlation between XET activity and tissue growth rate.
XET activity also provides a mechanism to cause wall loosening without
weakening of the wall. GA has been shown to increase XET activity in
internodes of a GA biosynthesis mutant pea line (Potter and Fry, 1993) and in
leaf blades of a GA biosynthetic barley mutant plant (Smith et al., 1996). The
mRNA transcript levels for EXTII, encoding XET, is also increased after GA
treatment (Smith et al., 1996). However, the role of XET in cell growth is
unclear since the enzyme has no effect on extension of isolated cucumber cell
walls (McQueen-Mason et al., 1993).

Another group of cell wall loosening factors are the expansins, which
act by breaking hydrogen bonds between the cellulose microfibrils and the
matrix polysaccharides (McQueen-Mason and Cosgrove, 1994). In rice,
expansin protein is detected mainly in the growing region of the internode.
During submergence, the amount of expansin increases as does the sensitivity
of the tissue to expansin action (Cho and Kende, 1997a). Furthermore, the
mRNA level of most expansin genes was highest in growing regions of the
plant. One expansin gene, Os-EXP4, shows an increase in transcript levels
within one hour after GA-treatment and submergence, while that of Os-EXPZ
has increased after 6 h of submergence (Cho and Kende, 1997b). Whether the
onset of increased transcript levels for Os-EXP4 correlate with the onset of
increased extensibility remains to be determined. However, the onset of

increased mRNA accumulation of Os-EXP4 correlates very well with the 40-

24

min lag phase of GA-induced growth (Sauter and Kende, 1992b; Cho and
Kende, 1997b).

Besides cell enlargement, GA application also leads to changes in cell
cycle activity. The transcript levels of cchOs-Z are increased 1 h after addition
of GA (Sauter et al., 1995). After 4 h, the number of cells in the G2 phase of the
cell cycle decreases, indicating an accelerated rate of mitosis (Sauter and
Kende, 1992b). This change coincides with an increase in the in vitro
phosphorylation of histone H1 and an increase in the expression of two
mitotic cyclin genes (Sauter et al., 1995). After 7 h of GA treatment, the rate of
DNA synthesis increases. This increase coincides with an increase in
expression of another gene encoding a histone H3, which is correlated with
the S-phase of the cell cycle (Van der Knaap and Kende, 1995). These results
lead us to conclude that GA promotes cell elongation and subsequently cell
division in the IM. In addition, the cell cycle appears to be first induced at the
G2 / M phase transition followed by an increase in DNA synthesis (Sauter and

Kende, 1992b).

Internodal elongation of deepwater rice during the vegetative adult
phase is a very good system to study growth and GA-regulated gene
expression. The response is physiologically very important for survival of the
plant, it is fast, and specific for GA. The internode of deepwater rice has been
studied extensively to identify early GA-regulated events in the IM, which is
the primary site of action for GA. The identification of early GA-regulated
genes from the IM of deepwater rice internodes may help us understand how
this remarkable growth response is achieved. The following Chapters describe
the experiments designed to identify and characterize early GA-regulated

genes. The identification of these genes will allow an assessment of their role

25

in GA-regulated processes in the M of rice internodes in terms of how they
regulate cell elongation and / or the cell cycle. After identification of early-
regulated genes we can also address how GA regulates changes in transcript

levels.

1.7. LITERATURE

Aach, H., Bode, H., Robinson, D.G. and Graebe, LE. (1997) ent-Kaurene
synthase is located in proplastids of meristematic shoot tissues. Planta
202:211-219.

Adams, P.A., Montague M.I., Tepfer, M., Rayle, D.L., Ikuma, H. and Kaufman,
PB. (1975) Effect of gibberellic acid on the plasticity and elasticity of Avena
stem segments. Plant Physiol 56:757-760.

Azuma, T., Sumida, Y., Kaneda, Y., Uchida, N. and Yasuda, T. (1996) Changes
in cell wall polysaccharides in the internodes of submerged floating rice.
Plant Growth Reg 19:183-187.

Ben-Nissan, G. and Weiss, D. (1996) The petunia homologue of tomato gastl:
transcript accumulation coincides with gibberellin-induced corolla cell
elongation. Plant Mol Biol 32:1067-1074.

Bishop, G.J., Harrison, K. and Jones, ].D.G. (1996) The tomato dwarf gene
isolated by heterologous transposon tagging encodes the first member of a
new cytochrome P450 family. Plant Cell 8:959-969.

Bleecker, A.B., Schuette, ].L. and Kende, H. (1986) Anatomical analyisis of
growth and developmental patterns in the internode of deepwater rice.
Planta 169:490-497.

Carol, P., Peng, J. and Harberd, NP. (1995) Isolation and preliminary
characterization of gasI-I, a mutation causing partial suppression of the
phenotype conferred by the gibberellin-insensitive (gai) mutation in
Arabidopsis thaliana (L.) Heyhn. Planta 197:414-417.

Carpita, NC. and Gibeaut, D.M. (1993) Structural models of primary cell walls

in flowering plants: consistency of molecular structure with the physical
properties of the walls during growth. Plant] 3:1-30.

26

Catling, H.D., Puckridge, D.W. and HilleRisLambers, D. (1988) The
environment of Asian deepwater rice. In Proceedings of the 1987
International Deepwater Rice Workshop, Bangkok, Thailand.
International Rice Research Institute, Manila, Philippines, pp 11—34

Catling, H.D. (1992) Rice in Deep Water. The Macmillan Press Ltd, London
and Basingstoke, UK.

Chandler, PM. (1988) Hormonal regulation of gene expression in the
"slender" mutant of barley (Hordeum vulgare L.). Planta 175:115-120.

Chiang, H.-H., Hwang, I. and Goodman, H.M. (1995) Isolation of the
Arabidopsis GA4 locus. Plant Cell 7:195-201.

Cho, H.—T. and Kende, H. (1997a) Expansins and internodal growth of
deepwater rice. Plant Physiol 113:1145-1151.

Cho, H.-T. and Kende, H. (1997b) Expression of expansin genes is correlated
with growth in deepwater rice. Plant Cell 9:1661-1671.

Chono, M., Yamauchi, T., Yamaguchi, S., Yamane, H. and Murofushi, N.
(1996) cDNA cloning and characterization of a gibberellin-responsive gene
in hypocotyls of Cucumis sativus L. Plant Cell Physiol 37:686-691.

Chory, ]., Voytas, D.F., Olszewski, N .E. and Ausubel, RM. (1987) Gibberellin-
induced changes in the populations of translatable mRNAs and

accumulated polypeptides in dwarfs of maize and pea. Plant Physiol 83:15-
23.

Cohn, N.S., Zhang, L., Mitchell, ].P. and Vierheller, C.—Z.]. (1994) Gibberellin-
stimulated changes in abundance of two mRNAs in the developing shoot
of dwarf peas (Pisum sativum L.). Int] Plant Sci 155:498-505.

Di Laurenzio, L., Wysocka-Diller, ]., Malamy, ].E., Pysh, L., Helariutta, Y.,
Freshour, G., Hahn, M.G., Feldmann, K.A. and Benfey, RN. (1996) The
SCARECROW gene regulates an asymmetric cell division that is essential
for generating the radial organization of the Arabidopsis root. Cell 86:423-
433.

Eiguchi, M., Sano, R., Hirano, H.-Y. and Sano, Y. (1993) Genetic and
developmental bases for phenotypic plasticity in deepwater rice. I Heredity
84:201-205.

Fry, S.C., Smith, R.C., Renwick, K.F., Martin, D.]., Hodge, SK. and Matthews,

KJ. (1992) Xyloglucan endotransglycosylase, a new wall-loosening enzyme
activity from plants. Biochem J 282:821-828.

27

Gale, M.D., Law, C.N., Marshall, GA. and Worland, A.]. (1975) The genetic
control of gibberellic acid insensitivity and coleoptile length in a "dwarf"
wheat. Heredity 34:393-399.

Herzog, M., Dorne, A.-M. and Grellet, F. (1995) GAS/l, a gibberellin-regulated
gene family from Arabidopsis thaliana reated to the tomato GASTI gene.
Plant Mol Biol 27:743-752.

Hoffrnann-Benning, S. and Kende, H. (1992) On the role of abscisic acid and
gibberellin in the regulation of growth in rice. Plant Physiol 99:1156-1161.

Hossain, M. (1996) Recent developments in the Asian rice economy:
challenges for rice research. In Rice Research in Asia: Progress and
Priorities. Eds: Evenson, R.E., Herdt, R.W. and Hossain, M. CAB
INTERNATIONAL and International Rice Research Institute, University
Press Cambridge, UK pp 17-33.

Inouye, J. (1985) Varietion of elongation ability in the Asian floating rice (L.)
Jap Agri Res Quart 19:86-91.

International Rice Research Institute. (1984) Terminology for rice growing
environments. IRRI, Los Banos, Philippines.

International Rice Research Institute. (1990) World rice database (RICESTAT).
IRRI, Los Banos, Philippines.

Jacobsen, SE. and Olszewski, NE. (1993) Mutations at the SPINDLY locus of
Arabidopsis alter gibberellin signal transduction. Plant Cell 52887-896.

Jacobsen, S.E., Shi, L., Xin, Z. and Olszewski, N .E. (1994) Gibberellin-induced
changes in the translatable mRNA populations of stamens and shoots of
gibberellin-deficient tomato. Planta 192:372-378.

Jacobsen, SE. and Olszewski, N .E. (1996) Gibberellins regulate the abundance
of RNAs with sequence similarity to proteinase inhibitors, dioxygenases
and dehydrogenases. Planta 198278-86.

Jacobsen, S.E., Binkowski, K.A. and Olszewski, NE. (1996) SPINDLY, a
tetratricopeptide repeat protein involved in gibberellin signal transduction
in Arabidopsis. Proc Natl Acad Sci USA 93:9292-9296.

Keith, K.A., Raskin, I. and Kende, H. (1986) A comparison of the submergence
response of deepwater and non-deepwater rice. Plant Physiol 80:479-482.

28

Kende, H. and Zeevaart, ].A.D. (1997) The five "classical" plant hormones.
Plant Cell 921197-1210.

Keyes, G., Sorrells, M.E. and Setter, TL. (1990) Gibberellic acid regulates cell
wall extensibility in wheat (Triticum aestivum L.). Plant Physiol 92:242-
245.

Khush, GS. (1997) Origin, dispersal, cultivation and variation of rice. Plant
Mol Biol 35:25-34.

Koornneef, M. and van der Veen, ].H. (1980) Induction and analysis of
gibberellin-sensitive mutants in Arabidopsis thaliana (L.) Heynh. Theor
Appl Genet 58:257-263.

Kutschera, U. and Kende, H. (1988) The biophysical basis of elongation growth
in internodes of deepwater rice. Plant Physiol 88:361-366.

Lange, T. (1997) Cloning gibberellin dioxygenase genes from pumpkin
endosperm by hetrologous expression of enzyme activities in Escherichia
coli. Proc Natl Acad Sci USA 94:6553-6558.

Martin, D.N., Proebsting, W.M. and Hedden, P. (1997) Mendel's dwarfing
gene: cDNAs from the Le alleles and function of the expressed proteins.
Proc Natl Acad Sci USA 94:8907-8911.

McQueen-Mason, S.]., Fry, S.C., Durachko D.M. and Cosgrove, DJ. (1993) The
relationship between xyloglucan endotranglycosylase and in-vitro cell wall
extension in cucumber hypocotyls. Planta 190:327-331.

McQueen-Mason, SJ. and Cosgrove, DJ. (1994) Disruption of hydrogen-
bonding between plant-cell wall polymers by proteins that induce wall
extension. Proc Natl Acad Sci USA 91:6574-6578.

Mekhedov, S.L. and Kende, H. (1996) Submergence enhances expression of a
gene encoding 1-aminocyclopropane-1-carboxylate oxidase in deepwater
rice. Plant Cell Physiol 37:531-537.

Métraux, ].-P. and Kende, H. (1983) The role of ethylene in the growth
response of submerged deepwater rice. Plant Physiol 72:441-446.

Métraux, ].-P. and Kende, H. (1984) The cellular basis of the elongation
response in submerged deepwater rice. Planta 160:73-77.

Morishima, H. (1975) Floating ability as an adaptive character of rice and its
measuring method. In Proceedings of the 1974 International Seminar on

29

Deepwater Rice. Bangladesh Rice Research Institute, Dacca, Bangladesh, pp
109-114.

Musgrave, A., Jackson, MB. and Ling, E. (1972) Callitriche stem elongation is
controlled by ethylene and gibberellin. Nature New Biol 238:93-96.

Nomura, T., Nakayama, M., Reid, J.B., Takeuchi, Y. and Yokota, T. (1997)
Blockage of brassinosteroid biosynthesis and sensitivity causes dwarfism
in garden pea. Plant Physiol 113:31-37.

Oka, HI. (1988) Origin of cultivated rice. In: Developments in crop science 14.
Japan Scientific Societies Press, Tokyo and Elsevier, Amsterdam, 254 pp.

Olszewski, N. (1997) Evidence that dynamic O-linked N-acetylglucosamine
modification of proteins plays a role in gibberellin signal transduction. 8th
International conference on Arabidopsis research. Madison, Wisconsin.

Peng, J. and Harberd, N .P. (1993) Derivative alleles of the Arabidopsis
gibberellin-insensitive (gai) mutation confer a wildtype phenotype. Plant
Cell 52351-360.

Peng, J., Carol, P., Richards, D.E., King, K.E., Cowling, R.J., Murphy, GP. and
Harberd NP. (1997) The Arabidopsis GAI gene defines a signaling pathway
that negatively regulates gibberellin responses. Genes Dev 11:3194-3205.

Phillips, AL. and Huttly, AK. (1994) Cloning of two gibberellin-regulated
cDNAs from Arabidopsis thaliana by subtractive hybridization: expression

of the tonoplast water channel, y-TIP, is increased by GA3. Plant Mol Biol
24:603-615.

Phillips, A.L., Ward, D.A., Uknes, S., Appleford, N.E.J., Lange, T., Huttly, A.K.,
Gaskin, P., Graebe, J.B. and Hedden, P. (1995) Isolation and expression of
three gibberellin 20-oxidase cDNA clones from Arabidopsis. Plant Physiol
108:1049-1057.

Potter, 1. and Fry, SC. (1993) Xyloglucan endotranglycosylase activity in pea
internodes. Effects of applied gibberellic acid. Plant Physiol 103:235-241.

Potts, W.C., Reid, J.B. and Murfet, LC. (1985) Internode length in Pisum.
Gibberellins and the slender phenotype, Physiol Plant 63:357-364.

Raskin, I. and Kende, H. (1983) Regulation of growth in rice seedlings. J Plant
Growth Regul 2:193-203.

Raskin, I. and Kende, H. (1984a) Regulation of growth in stem sections of
deepwater rice. Planta 160:66-72.

30

Raskin, I. and Kende, H. (1984b) Role of gibberellin in the growth response of
submerged deepwater rice. Plant Physiol 76:947-950.

Reed, J.W., Foster, K.R., Morgan, P.W. and Chory, J. (1996) Phytochrome B
affects responsiveness to gibberellins in Arabidopsis. Plant Physiol.
112:337-342.

Rijnders, J.G.H.M., Yang, Y.—Y., Kayima, Y., Takahashi, N., Barendse, G.W.M.,
Blom, C.W.P.M. and Voesenek, L.A.C.J. (1997) Ethylene enhances
gibberellin levels and petiole sensitivity in flooding-tolereant Rumex
palustris but not in flooding-intolerant R. acetosa. Planta 203:20-25.

Rose-John, S. and Kende, H. (1984) Effect of submergence on the cell wall
composition of deepwater rice internodes. Plant Physiol 76:106-111.

Rose-John, S. and Kende, H. (1985) Short-term growth response of deepwater
rice to submergence and ethylene. Plant Science 38:129-134.

Ross, J]. and Reid, J.B. (1992) Ontogenetic and environmental effects on GA1
levels and the implications for the control of internode length. In:
Progress in plant growth regulation. Eds. Karssen, van Loon and
Vreugdenhil, Kluwer Academic Publishers, Dordrecht, the Netherlands
pp 180-187.

Ross, J.J., Reid, J.B., Swain, S.M., Hasan, O., Poole, A.T., Hedden, P. and Willis,
CL. (1995) Genetic regulation of gibberellin deactivation in Pisum. Plant]
72513-523.

Sauter, M. and Kende, H. (1992a) Levels of b-glucan and lignin in elongating
internodes of deepwater rice. Plant Cell Physiol 33:1089-1097.

Sauter, M. and Kende, H. (1992b) Gibberellin-induced growth and regulation
of the cell division cycle in deepwater rice. Planta 188:362-368.

Sauter, M., Seagull, R.W. and Kende, H. (1993) Internodal elongation and
orientation of cellulose microfibrils and microtubules in deepwater rice.
Planta 190:354-362.

Sauter, M., Mekhedov, S.L. and Kende, H. (1995) Gibberellin promotes
histone H1 kinase activity and the expression of cdc2 and cyclin genes
during the induction of rapid growth in deepwater rice internodes. Plant J
7:623-632.

31

Sauter, M. (1997) Differential expression of a CAK (cdc2-activating kinase)-like
protein kinase, cyclins and cdc2 genes from rice during the cell cycle and in
response to gibberellin. Plant] 11:181-190.

Setboonsarng, S. (1996) Rice research priorities in Thailand. In Rice Research
in Asia: Progress and Priorities. Eds: Evenson, R.E., Herdt, R.W. and
Hossain, M. CAB INTERNATIONAL and International Rice Research
Institute, University Press Cambridge, UK pp 217-238.

Shi, L., Gast, R.T., Gopalraj, M. and Olszewski, NE. (1992) Characterization of
a shoot-specific, GA3- and ABA-regulated gene from tomato. Plant] 2:153-
159.

Silverstone, A.L., Chang, C.-w., Krol, E. and Sun, T.-p. (1997a) Developmental
regulation of the gibberellin biosynthetic gene GA1 in Arabidopsis
thaliana. Plant] 12:9-19.

Silverstone, A.L., Mak, P.Y.A., Martinez, EC. and Sun, T.-p. (1997b) The new
RGA locus encodes a negative regulator of gibberellin response in
Arabidopsis thaliana. Genetics 146:1087-1099.

Silverstone, A.L., Ciampaglio, C.N. and Sun, T.-p. (1998) The Arabidopsis
RGA gene encodes a transcriptional regulator repressing the gibberellin
signal transduction pathway. Plant Cell 10:155-169.

Smith, R.C, Matthews, P.R., Schunmann P.H.D. and Chandler, RM. (1996)
The regulation of leaf elongation and xyloglucan endotransglycosylase by

gibberellin in 'Himalaya' barley (Hordeum vulgare L.). J Exp Bot 47:1395-
1404.

Speulman, E. and Salamini, F. (1995) GA3-regulated cDNAs from Hordeum
vulgare leaves. Plant Mol Biol 28:915-926.

Stiinzi, ].T. and Kende, H. (1989) Gas composition in the internal air spaces of
deepwater rice in relation to growth induced by submergence. Plant Cell
Physiol 30:49-56.

Suge, H. (1985) Ethylene and gibberellin: regulation of internodal elongation
and nodal root development in floating rice. Plant Cell Physiol 26:607-614.

Suge, H. (1988) Physiological genetics of internode elongation in submerged
deepwater rice. In Proceedings of the 1987 International Deepwater Rice
Workshop, Bangkok, Thailand. International Rice Research Institute,
Manila, Philippines pp 275-285.

32

Sun, T.-p., Goodman, H.M. and Ausubel, FM. (1992) Cloning the Arabidopsis
gal-2 locus by genomic subtraction. Plant Cell 4:119-128.

Swain, SM. and Olszewski, NE. (1996) Genetic analysis of gibberellin signal
transduction. Plant Physiol 112211-17.

Taylor, B.H., Young, R.]. and Scheuring, CF. (1993) Induction of a proteinase
inhibitor II-class gene by auxin in tomato roots. Plant Mol Biol 23:1005-
1014.

Taylor, B.H. and Scheuring, CF. (1994) A molecular marker for lateral root
initiation: the RSI-1 gene of tomato (Lycopersicon esculentum Mill) is
activated in early lateral root primordia. Mol Gen Genet 243:148-157.

Tonkinson, C.L., Lyndon, R.F., Arnold, GM. and Lenton, JR. (1995) Effect of
the Rht3 dwarfing gene on dynamics of cell extension in wheat leaves, and
its modification by gibberellic acid and paclobutrazol. J Exp Bot 46:1085-
1092.

Tonkinson, C.L., Lyndon, R.F., Arnold, GM. and Lenton, JR. (1997) The
effects of temperature and the Rht3 dwarfing gene on growth, cell
extension, and gibberellin content and responsiveness in the wheat leaf. J
Exp Bot 48:963-970.

Tripathi, RS. and Balakrishna Rao, M]. (1985) Inheritance studies of
characters associated with floating habit and their linkage relationhip in
rice. Euphytica 34:875-881.

Truong, H.-N., Caboche, M. and Daniel-Vedele, F. (1997) Sequence and
characterization of two Arabidopsis thaliana cDNAs isolated by functional
complementation of a yeast gln3 gth mutant. FEBS letters 410:213-218.

Van der Knaap, E. and Kende, H. (1995) Identification of a gibberellin-induced
gene in deepwater rice using differential display of mRNA. Pl Mol Biol
28:589-592.

Vergara, B.S., Jackson, B. and De Datta, SK. (1977) Deep water rice and its
response to deep water stress. In Climate and Rice: proceedings of the 1976

symposium. International Rice Research Institute, Los Banos, Philippines,
pp 301-319.

Vergara, BS. (1985) Growth and development of the deepwater rice plant.

IRRI res paper ser no 103. International Rice Research Institute, Manila,
Philippines 38 pp.

33

Wilson, RN. and Somerville, CR. (1995) Phenotypic suppression of the

gibberellin-insensitive mutant (gai) of Arabidopsis. Plant Physiol 108:495-
502.

Winkler, R.G. and Freeling, M. (1994) Physiological genetics of the dominant
gibberellin-nonresponsive maize dwarfs, Dwarf8 and Dwarf9. Planta
193:341-348.

Wu, K., Li, L., Gage, DA. and Zeevaart, J.A.D. (1996) Molecular cloning and
photoperiod-regulated expression of gibberellin 20-oxidase form the long-
day plant spinach. Plant Physiol 110:547-554.

Wu, L.-L., Mitchell, ].P., Cohn, NS. and Kaufman, PB. (1993) Gibberellin
(GA3) enhances cell wall invertase activity and mRNA levels in
elongating dwarf pea (Pisum sativum) shoots. Int] Plant Sci 154:280-289.

Xu, Y.-L., Li, L., Wu, K., Peeters, A.].M., Gage, DA. and Zeevaart, J.A.D. (1995)
The GAS locus of Arabidopsis thaliana encodes a multifunctional
gibberellin 20-oxidase: Molecular cloning and functional expression. Proc
Natl Acad Sci USA 92:6640-6644.

Yabuta, T. and Sumiki, Y. (1938) On the crystal of gibberellin, a substance to
promote plant growth. J Agr Chem Soc Jpn 14:1526.

Yamaguchi, S., Saito, T., Abe, H., Yamane, H., Murofushi, N. and Kamiya, Y.
(1996) Molecular cloning and characterization of a cDNA encoding the
gibberellin biosynthetic enzyme ent-kaurene synthase B from pumpkin
(Cucurbita maxima L.). Plant] 10:203-213.

Zarembinski, T.I. and Theologis, A. (1997) Expression characteristics of OS-
ACSI and OS-ACSZ, two members of the 1-aminocyclopropane-1-
carboxylate synthase family in rice (Oryza sativa L. cv. Habiganj Aman II)
during partial submergence. Plant Mol Biol 33:71-77.

Zwar, J.A. and Chandler, RM. (1995) a-Amylase production and leaf protein
synthesis in a gibberellin-responsive dwarf mutant of "Himalaya" barley
(Hordeum vulgare L.). Planta 197:39-48.

34

CHAPTER 2
Modification of the differential display method and identification of
differentially expressed genes in gibberellin-treated deepwater rice

internodes

2.1. ABSTRACT

Differential display was performed to identify genes differentially
expressed following gibberellin (GA) treatment of rice stem sections. The
technique was optimized and resulted in the identification of four
differentially expressed genes. The transcript levels of dd3, dd4, dd12 and
histone H3 increased after GA treatment, while transcript levels remained
constant in control stem sections. Another differentially expressed gene,
OsTMK was identified serendipitously. A cDNA library was constructed from
RNA isolated from intercalary meristems of rice internodes, and full-length

cDNA clones of dd3, dd4, dd12 and OsTMK were isolated.

35

2.2. INTRODUCTION

Survival of deepwater rice during flooding is based on its capacity for
rapid internodal elongation when it becomes submerged. The signal for
accelerated growth is an increase in ethylene levels, which enhances the
responsiveness of the internode to GA (Raskin and Kende, 1984a). Enhanced
growth is initiated in the intercalary meristem (IM) at the base of the growing
internode and is based on increased production of cells and increased
elongation of these cells after they emerge from the meristem into the
elongation zone (EZ) (Bleecker et al., 1986; Sauter and Kende, 1992). Rice stem
sections containing the growing internode can be isolated and respond to
submergence, ethylene and GA in the same fashion as whole plants (Raskin
and Kende, 1984b). The lag phase of GA-induced growth in stem sections is 40
min (Sauter and Kende, 1992). After 2 h of GA application, the cells in the IM
have increased their average length. After 4 h, the number of cells in the G2
phase of the cell cycle has decreased indicating an accelerated rate of mitosis
(Sauter and Kende, 1992). After 7 h of GA treatment, the rate of DNA
synthesis increases. These results have led to the conclusion that GA
promotes cell elongation and then cell division in the IM (Sauter and Kende,
1992).

We were interested in identifying genes whose transcript levels are
altered in response to GA. There are three principal methods to identify
differentially expressed genes. (i) In differential screening, duplicate filters
containing a cDNA library are screened with a probe derived either from
RNA isolated from induced or uninduced tissues. By comparing
hybridization signal intensity of the duplicate filters, cDNA clones can be

identified whose transcript levels are differentially expressed (St. John and

36

Davis, 1979). Many differentially expressed genes have been identified with
this method. However, this method can only be used for the analysis of
relatively abundant genes with large differences in transcript levels. (ii) A
more widely used technique is subtractive hybridization. Genes with similar
transcript levels in both induced and uninduced tissues are removed by
hybridization to each other, leaving behind single stranded transcripts
specifically present in induced tissue (Sargent and Dawid, 1983). Although
this technique is more sensitive than differential screening, it is technically
challenging and suffers from a lack of reproducibility in that different genes
may be identified at different times. Also, genes whose transcript levels
increase or decrease cannot be identified at the same time. (iii) Recently, a
new technique has been developed: differential display of mRNA. RNA
isolated from induced and uninduced tissue is reverse transcribed and
amplified with an oligo dT containing and a random decamer primer in the
presence of a radioactive nucleotide. The amplified products are separated on
a DNA sequencing gel, and the amplified cDNA bands derived from induced
and uninduced RNA can be compared. Bands disappearing or appearing as a
result of a particular induction can be isolated and further analyzed (Liang
and Pardee, 1992). The advantage of this technique is that initial screening is
fast, the technique requires only small amounts of RNA, detects transcripts of
low abundance and may allow the identification of all differentially expressed
genes in the cell. Moreover, genes whose transcripts levels increase or
decrease can be detected simultanously, and RNA samples from different
time points and/ or treatments can be analyzed in the same experiment.

The role of GA-regulated genes in vegetative growth has received
relatively little attention. The identification of genes whose transcript levels

are increased or decreased early after GA treatment are likely candidates for

37

key regulatory steps leading to accelerated growth. Therefore, the
identification of these genes would allow an assessment of their role in GA-
regulated processes in the IM of rice internodes in terms of how they regulate
cell elongation and / or the cell cycle. After identification of early-regulated
genes we would also be able to address how GA regulates changes in
transcript levels. These analyses, in turn, would be complementary to a
genetic dissection of the signal transduction pathway leading from GA
application to changed gene expression. As described above, GA promotes
growth of rice internodes within 40 min. Therefore, we chose to perform a
differential display analysis at early timepoints after GA application, namely
0.5, 2.5 and 6.5 h after start of treatment for the pilot study (DD I), and 2 h after

start of treatment in case of the larger screen (DD II).

2.3. MATERIALS AND METHODS

PLANT MATERIAL AND GROWTH CONDITIONS

Seeds of deepwater rice (Oryza sativa L., cv. Pin Gaew 56) were obtained from
the International Rice Research Institute (Los Banos, Philippines). Seeds were
germinated in darkness in Petri dishes on moist filter paper at 30°C for three
days. Two seedlings were sown in one-quart plastic pots containing Baccto.
Ten pots were placed in 6-cm-deep trays, and seedlings were carefully and
sparingly watered from the top. After 3 weeks, trays were filled on alternating
days with water or half-strength Hoagland solution. Gravel was placed on top
of the soil to prevent algal growth, and the plants were staked. After 6 weeks,

trays were filled with half-strength Hoagland solution every morning and

38

with water every afternoon. The plants were grown in an environmental
chamber under the following conditions: 13 h light, 27°C; 11 h dark 20°C;
relative humidity 80%; 400 umol rn‘2 3'1. Twenty-cm-long stem sections
containing the highest internode were excised from the main stems and
tillers of 11- to 13-week-old plants. The sections were cut in such a fashion
that that lower node was 2 cm above the basal cut. Only sections in which the
growing internode was 3 to 9 cm long were used. All leaf sheaths originating
from the nodes not included in the sections were peeled off. The sections
were placed in 250-ml beakers containing 30 ml of distilled water and placed
in a 2.5-L plastic cylinder through which water-saturated, ethylene-free air at
80 ml min'1 was passed. After preincubation for 3 h, GA3 was added to half of
the sections to a final concentration of 50 11M, while the other half was kept in
H20. Incubation was allowed to proceed for the times indicated under
continuous light after which the IM (a 3- to 4-mm-long region above the
second highest node) was excised, frozen immediately in liquid nitrogen, and

stored at -80°C until use.

RNA ISOLATION

In all cases, except when specified otherwise, RNA was isolated by the
guanidine/acid-phenol method (Chomczynski and Sacchi, 1987). The tissue
(< 1 g FW) was ground in liquid nitrogen and thawed in 5 ml 4 M guanidine
thiocyanate, 25 mM Na-citrate, 0.5% sarkosyl, and 0.1 M B-mercaptoethanol.
After addition of 0.5 m1 2 M Na-acetate, pH 4.0, RNA was purified by organic
extraction with 5 ml of water-saturated phenol and 1 ml of chloroform. The
phases were separated by centrifugation at 12,0003 for 10 min at 4°C. The

RNA in the aquaous phase was precipitated by addition of 5 ml of

39

isopropanol. After centrifugation at 12.0003 for 10 min, the RNA pellet was
transferred to a 2-ml Eppendorf tube. The RNA was dissolved in 1.5 ml of 4 M
LiCl by vortexing for 15 min. It was precipitated by centrifugation at 12,0003
for 15 min, and the RNA pellet was resuspended again in 1 ml of 4 M LiCl.
After vortexing for 15 min, the RNA was precipitated by centrifugation. To
remove residual proteins, the RNA pellet was dissolved in 0.6 ml buffer
consisting of 10 mM Tris-HCl, pH 7.5, 0.5% SDS, and 1 mM EDTA by
vortexing for 10 min and placing the sample at 50°C for 10 min. RNA was
extracted with 0.6 ml of 10 mM Tris-HCl, pH 8.0, 1 mM EDTA-equilibrated-
phenol (TE-phenol). After centrifugation, the phenol-phase was extracted
once again with 0.2 ml 10 mM Tris-HCl, pH 7.5, 0.5% SDS, and 1 mM EDTA.
The aqueous phases were pooled and extracted twice with 0.8 ml chloroform,
after which the RNA was precipitated with 80 ul 3 M Na-acetate, pH 5.2 and
0.8 ml iSOpropanol. The pellet was washed with 70% ethanol, dried, and RNA
was dissolved in the appropriate amount of water.

The second method for RNA isolation was the phenol method. Tissue (FW <
1 g) was ground in liquid nitrogen and thawed in 5 ml extraction buffer
consisting of 50 mM Tris-HCl, pH 7.5, 5 mM EDTA, 100 mM LiCl, 1% SDS, 1%
B-mercaptoethanol, and 2.5 ml TE-phenol. The tissue was further
homogenized in a Polytron for 3 min. After addition of 7.5 ml chloroform,
the organic phase was separated from the aqueous phase by centrifugation at
10,0003 for 10 min at 4°C. The aqueous phase was extracted once more with
chloroform, the RNA was precipitated by addition of 2.5 ml°8 M LiCl, and
collected by centrifugation at 12,0003 for 10 min at 4°C. The RNA pellet was
dissolved in 900 pl TE. To remove traces of DNA, the RNA was precipitated
by addition of 300 11L 8 M LiCl and collected by centrifugation at 12,0003 for 10

min at 4°C. To remove the LiCl, which can inhibit reverse transcription, the

40

RNA was dissolved in 200 pl TE and precipitated again in the presence of 0.3
M Na-acetate, pH 5.2 and 2.5 times the volume of ethanol.

The third RNA isolation method was the TriReagentTM method (Molecular
Research Center, Cincinnati, OH). Tissue (FW < 1 g) was ground in liquid
nitrogen and thawed in 10 ml of TriReagent solution. The tissue was
homogenized in Polytron for 3 min. After addition of 2 ml chloroform, the
organic phase was separated from the aqueous phase by centrifugation at
10,0003 for 10 min at 4°C. The aqueous phase was extracted once more with
chloroform, and the RNA was precipitated by addition of an equal volume of
isopropanol. Although this RNA should be ready for Northern blot analysis
and reverse transcriptase reactions, the pellet dissolved poorly in 600 111 TE
and contained many particulates. RNA was, therefore, precipitated again by
addition of 600 pl 4 M LiCl and collected by centrifugation for 10 min at
12,0003 at 4°C. To remove residual proteins, the RNA pellet was dissolved in
0.6 ml 10 mM Tris-HCI, pH 7.5, 0.5% SDS, and 1 mM EDTA by vortexing for 10
min and placing the sample at 50°C for 10 min, after which the sample was
extracted with 0.6 ml of TE-phenol. After centrifugation the phenol-phase was
extracted once again with 0.2 ml 10 mM Tris-HCl, pH 7.5, 0.5% SDS, and 1 mM
EDTA. The pooled aqueous phases were extracted twice with 0.8 ml of
chloroform, and the RNA was precipitated with 80 ul 3 M Na-acetate, pH 5.2,
and 0.8 ml isopropanol. The pellet was washed in 70% ethanol, dried, and the

RNA was dissolved in the appropriate amount of water.

DIFFERENTIAL DISPLAY

To remove traces of DNA, 100 ug of RNA was treated with 100 U DNase I
(Boehringer Mannheim Biochemicals [BMB]) in 100 mM Na-acetate, pH 6.2

41

and 5 mM MgC12 prior to reverse transcription. The RNA was purified by
organic extraction with TE-phenol/ chloroform and precipitated with ethanol.
The RNA was dissolved in the appropriate amount H20, and its integrity was

checked by formaldehyde-agarose gel electrophoresis (see below).

Differential display I

Differential display analysis was performed using two different variations of
the method (see Figure 2.1). A pilot experiment was set up to study the
feasiblity of the technique and to perform control reactions. The primers used
in differential display I (DD 1) were as follows: one degenerate anchored oligo
dT primer T12MG (M stands for A, C and G) synthesized at the
Macromolecular Facility at Michigan State University, East Lansing, and 20
random decamer primers OPA 01 to OPA 20 purchased from Operon
Technologies (Alameida, CA). Total RNA, 0.4 or 0.8 ug, was annealed to 1 uM
of one anchored oligo dT primer and reverse transcribed in a final volume of
20 ul with 100 U of M-MLV reverse transcriptase (Promega) in the presence of
20 uM dNTP, 50 mM Tris-HCI, pH 8.3, 75 mM KCl, 3 mM MgC12, and 10 mM
DTT for 1 h at 37°C. Alternatively, when SUPERSCRIPT 11 RT (Gibco-BRL) was
used, the reaction mixture was incubated for 10 min at 37°C, followed by 50
min at 42°C, under otherwise identical conditions. One-tenth of the reaction
mixture was amplified by PCR with 1 11M of the same anchored oligo dT
primer, end-labeled with 1.7 uCi of 'y-[37-PJATP (6000 Ci / mmol, New England
Nuclear [NEN]) per reaction using polynucleotide kinase, and 0.2 11M of one
random 10-mer in a final volume of 20 [.11 containing 10 mM Tris-HCl, pH 8.3,
50 mM KCl, 1 mM MgC12, 4 uM dNTP, and 2 U Taq polymerase. After 40

cycles, each consisting of 94°C for 30 sec, 40°C for 2 min, and 72°C for 30 sec,

42

Total RNA

DDI I DD II

 

 

 

 

 

 

 

 

TidVlG. 3 time points Reverse Transcription H-T11M, 1 time point
of subset of mRNA

32F endlabeled T1 ZMG PCR P33incorporation into

and OPAOi-OPAZO —_ PCR product, H-T1 {VI and

(20 primer combinations) H_Ap1 _ H_Ap30

(240 primer combinations)

 

Separate products on 8% Identification Separate products on 6%
sequencing gel, expose, . . Genomyx programmable
cut out band, elute, pre- 0] differentially sequencing gel, expose,
cipitate DNA, and reamplify displayed bands cutout band, and reamplify

/ \ “‘”°}\

Probe labeling Clone in pUC19

by FOR I Probe labeling Clone in pGEM-T

by random prime

Dot blots/colony lifts
to identify likely inserts

Northerns . . . Northerns
Verification

 

 

 

 

Figure 2.1 . Flow chart for the two differential display protocols used
in this study.

43

the PCR products were electrophoretically separated on a 8% denaturing
polyacrylamide gel, after which the gel was dried without fixing. X-ray film

was exposed to the gel overnight at room temperature.

Reamplification and clonin3 of DD I products

Differentially displayed products were excised from the gel and eluted by
boiling in 100 pl H20 for 15 min. The DNA was precipitated with ethanol in
the presence of 40 ug glycogen as a carrier. The DNA was reamplified without
addition of radioactive label under the same reaction conditions as described
above, except that the final concentration of dNTP was 20 [1M in a volume of
40 pl. After PCR-mediated amplification, the DNA products were incubated
in the presence of 20 ug of proteinase K for 1 h at 65°C, purified by TE-
phenol/ chloroform extraction, and precipitated with ethanol. The generation
of blunt-end DNA fragments was mediated by incubation with Klenow
enzyme (BMB) for 30 min at room temperature in the presence of 20 pM
dNTP. The DNA fragments were phosphorylated with 300 11M ATP and
polynucleotide kinase (BMB) for 30 min at 37°C. The products were
electrophoretically separated on a non-denaturing polyacrylamide gel. DNA
fragments of expected size were isolated by shaking the corresponding gel-
slice overnight in 7.5 M NH4-acetate, 10 mM Mg-acetate, 1 mM EDTA, and
0.1% SDS. DNA was precipitated with ethanol and either labeled for use as a
probe or inserted into the dephosphorylated SmaI site of pUC 19.

Labeling of DD I products for use as probes

A one-hundredth dilution of differentially displayed products was
reamplified under the same reaction conditions as described above, except
that 5 pl of a-[32PJdCTP (3000 Ci / mmol, NEN) was added to the PCR, and the

final concentration of non-radioactive dCTP was 2.5 pM.

Differential display II

The second differential display method (DD 11, see Figure 2.1) was performed
by Sandrine Jagoueix, who carried out a screen with 240 primer combinations
using 10 RNAimageTM kits (GenHunter Corp, Nashville, TN). The
experiments were performed with slight modifications of the manufacturer’s
specifications. Briefly, 0.2 pg of RNA was reverse transcribed in a total
volume of 20 pl in the presence of 20 pM dNTP and 0.2 pM H-T11M (16-mer)
in which the H stands for a HindIII recognition sequence and M stands for
either A, C or G. Two pl of cDNA was amplified with the same H-T11M (0.2
pM) and 0.2 pM H-AP primer (12-mer), which is an arbitrary primer
containing a H indIII recognition site, in the presence of 4 pM dNTP and 0.25
pl a-[33P]dATP (2000 Ci/mmol, 10 mCi/ ml, NEN) in a total volume of 20 pl.
PCR conditions were: 95°C for 30 sec, 40°C for 2 min, 72°C for 1 min over 40
cycles. PCR products were separated on a 6% DNA sequencing gel in a
GenomnyR programmable DNA sequencer (Genomyx, Foster City, CA). The
gel was dried on the glassplate, and PCR products were visualized by

autoradiography.

45

Reamplification and cloning of DD II products.

Differentially displayed products were excised from the gel and reamplified
directly in PCR buffer containing 20 pM dNTP in a total volume of 40 pl
under otherwise conditions as described above. DNA fragments were
separated on 1.5% agarose gel and isolated from the gel using a WizardTM PCR
prep DNA purification system (Promega). Part of the reamplified cDNA was
used for random prime probe labeling, and the remaining cDNA was ligated

directly into the pGEM®-T vector (Promega).

NORTHERN BLOT ANALYSIS

Twenty pg of total RNA was electrophoretically separated in a 1.2% agarose
gel containing 1.3% formaldehyde, 20 mM MOPS [3-(N-morpholino)-
propanesulfonic acid], 5 mM Na-acetate, 1 mM EDTA, pH 7, and 0.2 pg/ m1
ethidium bromide. After visualization of ribosomal RNA by illumination
with UV light, the gel was washed three times for 10 min each in H20 and
equilibrated in 10x SSC for 45 min. RNA was transferred to Hybond-N
membrane (Amersham) in 20x SSC transfer buffer. DNA fragments to be
labeled as probe were isolated from agarose gels by B-agarase digestion (New
England Biolabs [NEB]), after which the DNA was precipitated with ethanol.
Fifty ng of template DNA was labeled with or-[32P]dCTP (3000 Ci / mmol, NEN)
using a random primer DNA labeling kit (BMB). Northern blots were
prehybridized for 4 h at 42°C in 5x SSC, 10x Denhardt's solution, 0.1% SDS, 0.1
M K-PO4 pH 6.8, and 100 pg/ ml denatured salmon sperm DNA and
hybridized overnight at 42°C in 5x SSC, 10x Denhardt's solution, 0.1 M K-PO4,
pH 6.8, 100 pg/ ml denatured salmon sperm DNA, 10% dextrane sulfate, and

46

30% formamide. The blots were washed twice in 2x SSC and 0.5% SDS and
twice in 0.1x SSC and 0.1% SDS at 65°C for 30 min each. The radioactivity was

quantified with a Phosphorlmager (Molecular Dynamics, Sunnyvale, CA).

DOT BLOT AND COLONY LIFT HYBRIDIZATION

Plasmids (50 ng) were denatured in 0.5 M NaOH, 1.5 M NaCl and spotted onto
Hybond N+ membrane (Amersham) in a BioRad dot blot apparatus. After
neutralization in 1.5 M N aCl, 0.5 M Tris-HCl, pH 7.2, and 1 mM EDTA, the
DNA was crosslinked to the membrane by baking for 2 h at 80°C.
Recombinant E. coli cultures were streaked out on selective medium and
grown overnight. The next day, a nitrocellulose filter was placed on top of the
colonies after which the filter was peeled off. DNA was denatured by placing
the filters on Whatman paper soaked in 0.5 M NaOH, neutralized in 1 M Tris-
HCl, pH 8, washed in 1 M Tris-HCl, pH 8, 1.5 M NaCl, and washed once more
in 2x SSC for 2 min each. The filters were air-dried and baked for 2 h at 80°C.
Dot blot and colony lifts were prehybridized for 4 h and hybridized with
random prime labeled probe overnight in 6x SSC, 0.5% SDS, 5x Denhardt's
solution, and 100 pg / ml denatured salmon sperm DNA at 65°C. Excess probe
was washed off twice with 2x SSC and 0.5% SDS and twice with 0.2x SSC and
0.1% SDS at 65°C for 30 min each.

cDNA LIBRARY CONSTRUCTION

Total RNA isolated from the IM of stem sections treated with GA for 0.5, 2.5

and 6.5 h was used for polyadenylated RNA isolation. RNA, 150 pg of each

timepoint, was pooled, and NaCl was added to a final concentration of 0.5 M.

47

Polyadenylated RNA was allowed to bind to BioMag Oligo (dT)20 beads
(Advanced Magnetics, Cambridge, MA), after which the bound
polyadenylated RNA was magnetically separated from the non-
polyadenylated RNA. Magnetic particles were washed in 7 mM Tris—HCl, pH
8, 170 mM NaCl, and polyadenylated RNA was eluted with H20 at 55°C for 2
min. The recovery of RNA was spectrophotometrically determined to be 4 pg.
The integrety of the RNA was checked by Northern blot analysis prior to
library construction. A SUPERSCRIPTTM Choice System for cDNA synthesis was
obtained (Gibco-BRL), and the manufacturer's specifications were followed.
Briefly, 3 pg of polyadenylated RNA was reverse transcribed with 400 U
SUPERSCRIPT II RT (RNase H-) in the presence of 1 pg oligo (dT)12-13 and 50 ng
random hexamers in a total volume of 20 pl at 37°C for 20 min followed by
45°C for 30 min. Second-strand synthesis was performed in a total volume of
150 pl in the presence of 10 U E. coli DNA ligase, 40 U E. coli DNA polymerase
I, 2 U E. coli RNase H, 1 pl a-[32PJdCTP (3000 Ci/mmol, NEN) for 2 h at 16°C.
To blunt-end cDNA, 10 U of T4 DNA polymerase was added to second-strand
synthesis reaction, and the incubation was continued for an additional 5 min
at 16°C. The cDNA was purified by organic extraction with an equal volume
of TE-phenol / chloroform, and the cDNA was precipitated with ethanol. The
percent conversion from mRNA to second-strand cDNA was 35% as
determined by trichloroacetic acid precipitation, resulting in 0.998 pg of
cDNA. After ligation of EcoRI (NotI, SalI) adapters to both ends of the cDNA,
the products were size-fractionated on SEPHACRYL® S-500 HR prepacked
columns. Fractions were pooled to two final sizes: one containing 0.6 to 6.6 kb
inserts (3.6 kb average) and one containing 0.4 to 4.4 kb inserts (2.4 kb
average). Two hundred ng of each size was ligated into 5 pg of EcoRI-digested
kgt11 arms (Promega) for 3 h at 22°C and packaged for 2 h at 21°C in

48

Gigapack® 111 Gold Packaging Extract (Stratagene). The extract was diluted 10
times in SM buffer (50 mM Tris-HCl, pH 7.5, 10 mM NaCl, 8 mM MgSO.;, and
0.01% gelatin). The total number of plaque forming units (pfu) of the 3.6-kb
average-size library was 6.5 x 106 and the pfu of the 2.4-kb average-size library
was 1.65 x 107. One million pfu of both libraries were amplified once and

stored at 4°C and, after addition of 7% DMSO, at -80°C.
Screening of cDNA libraries

Host E. coli strain LE392 was subcultured from an overnight culture in LB
medium supplemented with 0.2% maltose and 10 mM MgSO4 until A600 ~ 0.6
to 0.8. For the primary screen, 5 x 104 pfu in 200 pl SM buffer per plate were
incubated with 200 pl LE392 at 37°C for 20 min. Subsequently, 7 ml of top agar
(1 g bacto-tryptone, 0.5 g N aCl, 0.8 g agar, 10 mM MgSO4 per 100 ml) at 48°C
was added to the phage-E. coli mix, which was immediately plated out on
prewarmed LB-plates (150 mm). The phages were grown overnight at 37°C.
Plates were cooled to 4°C, after which nitrocellulose filters were placed on top
of the agar. Filters were marked with India ink and peeled off. After air drying
for 10 min, DNA on filters was denatured for 1 to 2 min in 0.2 M NaOH, 1.5 M
NaCl, neutralized for 1 to 2 min in 0.4 M Tris-HCl, pH 7.6, 2x SSC, and washed
for 1 to 2 min in 2x SSC. DNA was crosslinked to nitrocellulose by baking for
2 h at 80°C. The filters were prehybridized for 4 h and hybridized with
random prime labeled probe overnight in 6x SSC, 0.5% SDS, 5x Denhardt's
solution, and 100 pg/ ml denatured salmon sperm DNA at 62°C. Excess probe
was washed off twice with 2x SSC and 0.5% SDS and twice with 0.2x SSC and
0.1% SDS for 30 min each at 62°C. X-ray film was exposed to the filters for 1 to

3 days at -80°C. Plaques were picked with the wide end of a Pasteur pipet and

49

eluted in 1 ml SM buffer. The secondary screen was performed using the
above described conditions, except that 90-mm Petri plates and 3 ml of top
agar were used. Purification was continued until the probe hybridized to 100%

of phages on one plate.

Phage purification

Phages were eluted from a single plaque overnight at 4°C in 100 pl SM buffer.
Two hundred and fifty pl of an overnight culture of LE392 grown in the
presence of 0.2% maltose and 10 mM MgSO4 was incubated with 4 pl phage
suspension for 20 min at 37°C. This was added to 50 ml of LB supplemented
with 10 mM MgSO4 and shaken until lysis occurred (5 to 7 h). Ten pg/ ml
RNase A and 1 U / ml DNase I were added, and the lysed culture was
incubated at room temperature for 20 min. To remove cellular RNA and
DNA, an equal volume of DE52 resin equilibrated in LB was mixed with the
phage lysate. After centrifugation, the phage supernatant was treated with 20
mM EDTA, 50 pg/ ml proteinase K, and 0.1% SDS at 45°C for 20 min. DNA
was purified from the phage heads by repeated organic extraction with an
equal volume of TE-phenol until the interphase was clean. The aqueous
phase was once extracted with chloroform, and the DNA was precipitated by
addition of 1/ 10 of volume 3 M Na-acetate, pH 6, and 1 volume of
isopropanol. The DNA pellet was dissolved in 100 pL TE at 4°C.

PCR analysis of phage insert size

The insert sizes of phages were checked after the primary screen using a gene-

specific primer and either primer of the Agtll arms. The sequence of the

50

vector primer were HK115: 5’-CATATGGGGATTGGTGGCGACGACTCC-B’
and HK116: 5’-CCAGACCAACTGGTAATGGTAGCGACC-3’. Gene-specific
primers were: for dd3, HK130: 5’-
AGGTGGTGAAAGTAGCCGAGCCGATCCAG-3’; for dd4, HK131: 5’—
GTCCACAAGCTGCTGCAGATATTACAAGG-3’; for dd12, HK159: 5’-
AAGGCTAATACTAGCTAGCATCTATTGCAG-3’; for OsTMK, 5’-
AACATCCGAAAGAGTAGAGAGCACATTGACAG-B’. Two pl of eluted
phage was amplified in the presence of 10 mM Tris-HCl pH 8.3, 50 mM KCl,
1.5 mM MgC12, 50 pM dNTP, and 0.2 pM of each of the primers. After 35
cycles, each consisting of 94°C for 40 sec, 60°C for 1 min, and 72°C for 2 min,
the PCR products were electrophoretically separated and visualized on a 1%

agarose gel.

2.4. RESULTS

OPTIMIZATION OF DD I

At the start of the pilot study, the reaction conditions used in the
original protocol were applied (Liang and Pardee, 1992). However, several
control experiments proved to be necessary to optimize the technique.
Therefore, first, the concentration of RNA and the number of PCR cycles were
varied. As can be seen in Figure 2.2, the banding pattern did not change
qualitatively whether 0.1 pg of polyadenylated RNA, 2 pg of total RNA or 0.2
pg of total RNA was used for the reverse transcription reaction. For selected
primer combinations, only 30 cycles were necessary to obtain sufficient

amplification using 0.1 pg of polyadenylated RNA or 2 pg total RNA (Fig.

51

 

Primer T1zGA T12CA
RNA 0.1 pg A+ 2 pg total 0.2 pg total 0.1 pg A+ 2 pg total 0.2 pg total
PCR cycles 30 35 40 30 35 40 30 35 40 30 35 40 30 35 40 30 35 4O

 

 

Figure 2.2. Differential display using different RNA concentrations or PCR
cycles. Differential display was performed using RNA isolated from GA-treated
stem sections at the concentrations indicated above the lanes. The cDNA was
amplified in the presence of a-[358]dATP and 2 pM dNTP at an annealing
temperature of 42'C and number of PCR cycles as indicated above the lanes.
Othenwise, conditions were as described in Materials and Methods.

(A) Differential display using T12GA and 5'-GCGGAGGTA-3’ as primer pair.
(B) Differential display using T12CA and 5'-GCGGAGGTA-3' as primer pair.

52

2.2A). For other primer combinations, at least 35 cycles were necessary (Fig.
2.2B). An additional number of amplification cycles did not change the
banding pattern. Therefore, the number of PCR cycles was maintained at 40.
Using 0.2 pg of total RNA did not always give consistent banding patterns.
This was shown in Figure 2.2A, where 0.2 pg of total RNA yielded more
bands after 35 cycles than after 40 cycles. In Figure 2.23, some amplification
after 30 cycles was observed, none after 35, and sufficient amplification after 40
cycles starting with 0.2 pg of total RNA. By increasing the total RNA
concentration to 0.4 pg in the reverse transcription and the final
concentration of dNTP from 2 pM to 4 pM in the PCR step, amplification and
reproducibility were greatly enhanced (data not shown).

We also observed that the a-[35SJdATP used to label the products
during amplification severely contaminated the thermal cycler. The use of 32P
end-labeled primers was found to alleviate contamination and would allow
us to address the question whether both 3’ and 5’ primers contributed equally
to the amplification of the products. Therefore, each of the primers were end-
labeled separately with y—[32PJrATP, and differential display was performed
under otherwise identical conditions. Fewer bands were observed with end-
labeled T12AG primer than with end-labeled OPA16 primer at PCR conditions
described in the original protocol (Liang and Pardee, 1992; data not shown).
This suggested that the PCR cycling conditions were too stringent for the
T12AG primer. The experiment was repeated by lowering the annealing
temperature by 2°C to 40°C. As can be seen in Figure 2.3A, in this case a
similar banding pattern was obtained when either end-labeled T12AG primer
or end-labeled OPA16 primer was used. We repeated this experiment with
other primer combinations (T12MG-OPA03, T12MG-OPA16, and T12AG-

OPA04) and found that the banding patterns were similar. We concluded that,

53

A B

End-labeled primer T12AG OPA16 T12MG T,ZCG

Treatment GA C GA C GA C GA C

 

 

 

 

 

 

Figure 2.3. Differential display using 32P end-labeled primers. RNA was
isolated from the IM of stem sectionstreated for 6.5 h with 6A3, or kept in

water as control. (A) Comparison of banding patterns obtained using either

the oligo dT primer (T 12AG) or the decamer (OPA16) as end-labeled primer. (B)
Comparison of banding patterns obtained using either end-labeled degenerate
oligo dT primer (T 1 2MG) or end-labeled specific oligo dTprimer (T 1 206) and
OPAO4 as decamer.

under these conditions, both primers contributed equally to the PCR. End-
labeling of the oligo dT primer gave the sharpest bands and the most
reproducible results and, therefore, this protocol was used in further
experiments.

We also compared the banding patterns of degenerate (T12MG) and
specific (T12CG) oligo dT primers. As can be seen in Figure 2.3B, the banding
pattern obtained with each of the oligo dT primers showed similar as well as
different bands. One would expect that more bands will be seen with
degenerate than with specific primers. However, this was not the case. With
T12MG, more pronounced bands were observed than with T12CG. Therefore,
degenerate oligo dT primers were used for the DD I protocol.

The quality of RNA is essential for successful differential display
analysis. To test which RNA isolation generated the highest quality RNA for
differential display, RNA was isolated by three different methods as decribed
in Materials and Methods. Stem sections were treated with GA or kept in
water as control, after which differential display was performed with OPA04
and T12MG as primers. In Figure 2.4, the guanidine/acid-phenol method
(lanes Gu) is compared to the phenol (lanes Ph) and the TriReagent (lanes
Tri) methods. By comparing the amplification products, it is evident that
different RNA isolations did not yield differences in banding patterns. This
result was not primer-pair specific as the OPA16-T12MG primer pair showed
the same results (data not shown).

Besides the quality of RNA, reproducible reverse transcription is also
essential in differential display analysis. The banding pattern created by two
different reverse transcriptases is shown in Figure 2.4. RNA in the first six
lanes was reverse transcribed with Promega M-MLV, while RNA in the

second six lanes was reverse transcribed with Gibco-BRL SUPERSCRIPT H RT.

55

Reverse transcriptase M-MLV SUPERSCRIPT

RNA isolation Gu Ph Tri Gu Ph Tri

GACGACGAC GAC GAC GAC

 

 

 

 

 

Figure 2.4. Differential display using RNA isolated by different methods and
reverse transcriptases. RNA was isolated from the IM of stem sections
treated for 6.5 h with GA3 or kept in water as control, using 321’ end-labeled
T12MG and OPA04 as primer pair. Gu = Guanidine/acid-phenol method; Ph =
phenol method; Tri = TriReagent method as described in Materials and
Methods. RNA was reverse transcribed using either M-MLV or SUPERSCRIPT
ll RT. Star indicates differentially displayed band present preferentially when
using SUPERSCRIPT ll RT. Arrow indicates displayed band present
preferentially when using M-MLV.

56

As can be seen in Figure 2.4, the two different reverse transcriptases resulted
in slightly different banding patterns. For example, two differentially
displayed bands, denoted by the arrow, are barely differentially displayed
when SUPERSCRIPT H RT was used instead of M-MLV. On the other hand,
some differentially displayed bands, denoted by the star, appeared more
clearly when SUPERSCRIPT H RT was used. Slightly different banding patterns
were observed as well with OPA02-T12MG and OPA10-T12MG as primer pairs
(data not shown). Since banding patterns changed little (with different reverse
transcriptases) or not at all (with different RNA isolation methods), the
differential display analysis was continued with RNA isolated by
guanidine/acid-phenol method and Promega M-MLV reverse transcriptase, a

technique and enzyme standardly used in our laboratory.

IDENTIFICATION OF GA-REGULATED GENES VIA DD I

RNA was isolated from the IM of stem sections treated for 0.5, 2.5 or 6.5
h with GA or kept for the same amount of time in water as control. To
eliminate false positives, each differential display was repeated with RNA
isolated in a separate experiment. For DD 1, 20 primer combinations using
T12MG and OPA1 - OPA20 primers were employed. Very few bands were
found to be obviously differentially displayed. Therefore, some of the less
convincing bands were also analyzed (Table 2.1). Thirty-eight differentially
displayed products were reamplified. Fourteen PCR products were
reamplified in the presence of a-[32PJdCTP and used as probe for Northern
blot analysis. None were found to contain DNA sequences representing
differentially expressed genes. The differential display analysis was repeated

using the same primers that led to the identification of those 14 bands. The

57

 

805 mm :85
66:26 53, .mfibmcm 83 56582 >6 bout?
.mocom nommoaxo >__m=co.ot_u 69236340

303605 33:35 23:55:? 66:20

.06ch 663.86 >__m=:o_ot_u 6.9636036”.

 

or 0

mV «.4

mm mm

mocom comwmaxo >__m=co.ot_u uQmSdeG

mocom 66.8993 232.2935

395 mm £03605 mun. £25 .mfibmcm £03
56582 B 586.93 .mzcoeoti ho concoct“;

mucus 626.36 >=m=c2¢£u noEEEmom

 

=8 .00

 

.mom>_mcm >335 .mzcouotfi .6 8.361 ._..~ 638.

58

repeat experiment indicated that six of the 14 differentially displayed bands
were not reproducibly differentially displayed. In the case of three bands, the
labeling intensity on the differential display gel remained constant during GA
treatment, while the intensity decreased in the control reaction. For one band,
the intensity decreased during GA treatment. The remaining three bands,
whose labeling intensity had reproducibly increased during GA treatment, are
indicated in Figure 2.5 as PCR clone 98, dd3 and dd4. All three were derived
from the same primer pair OPA04-T12MG.

Previously, Northern blot analysis, performed as described above, had
failed to show that dd3 and dd4 were derived from a gene whose transcript
levels had changed during GA treatment. Because dd3 and dd4 were
consistently differentially displayed, the PCR products were cloned first, after
which individual inserts were analyzed. The most convincing differentially
displayed product was PCR clone 93 (whose PCR product was not used for
Northern blot analysis as described above). After reamplification of PCR clone
9B, cloning and transformation, 45 colonies were picked and analyzed for
plasmid insert. All 45 plasmids contained an insert of the expected size; of
these, 38 contained the same insert as verified by dot-blot hybridization
(results not shown). One of these inserts was used as hybridization probe on a
Northern blot of RNA isolated from the IM of rice stem sections treated with
GA, and of RNA isolated from control IM. As shown in Figure 2.6, transcript
levels for PCR clone 9B had increased during GA treatment. Sequence
analysis of PCR clone 98 indicated that the clone contained the oligo dT
primer at the 3’ end of the gene and OPA04 in the coding region. Except for
one unidentified amino acid, the sequence of clone 9B is identical to the
corresponding region of histone H3 of rice and other organisms (Van der

Knaap and Kende, 1995).

59

0.5 2.5 6.5 0.5 2.5 6.5 h

M GACGACGAC GACGACGAC

 

 

 

 

587-
434-
<:l PCR clone 9B
267-
<.‘: dd3
234' <2: dd4
213-
192-
1 84-

 

 

 

 

Figure 2.5. Differential display using RNA isolated from the N of stem sections
treated for 0.5 h, 2.5 h and 6.5 h with GA3 or kept in water. 32F end-labeled
T12MG and OPA04 were used as primer pair. The second panel is a repeat of
the first one, using RNA isolated in a separate experiment. Star indicates a
differentially displayed band which is not reproducibly differentially displayed.
Arrows indicate differentially displayed bands which are reproducibly
differentially displayed.

60

After the PCR products corresponding to dd3 and dd4 were cloned, a
method similar to that described for PCR clone QB was used to identify the
most commonly cloned insert. Instead of DNA dot blots, a colony lift
identified the most likely insert. This insert was subsequently used as probe
for Northern blot analysis. As can be seen in Figure 2.6, the expression of dd3
increased during GA treatment. For dd4, initial Northern blot analysis with
total RNA gave no signal. However, a signal was detected on a Northern blot
of polyadenylated RNA isolated from GA-treated stem sections. The
Northern blot analysis indicated that the transcript levels of the gene encoded
by dd4 increased during GA treatment (data not shown). Figure 2.6 shows the
induction pattern of dd4, using a longer clone isolated from the cDNA library
as probe. Inserts of dd3 and dd4 contained the oligo dT primer at the 3’ end of
the gene and the OPA04 primer at the 5’ end of the gene. Sequence analysis of
dd3 and dd4 did not indicate the function of these genes.

Also shown in Figure 2.6 is the signal detected by a cDNA the
corresponding mRNA levels of which did not change during the course of
the experiment. On a differential display using primer pair T12MG and
OPA07, one prominent band, PCR clone D, showed equal intensity in all
treatments (data not shown). After isolation from the gel, the corresponding
cDNA was reamplified and inserted into the pUC19 cloning vector. Sequence
analysis indicated 65% amino acid identity to E37, a chloroplast inner-
envelope protein isolated from spinach (Dreses-Werringloer et al., 1991; Van

der Knaap and Kende, 1995).

61

IDENTIFICATION OF GA-REGULATED GENES BY DD H

The pilot experiment resulted in the identification of few differentially
expressed genes. Further optimization did not increase the number of clear
bands per lane, nor reduce the often high background in the lanes. To
perform a complete screen, primers and reagents provided in the RNAimage
kit were used (GenHunter). To reduce the magnitude of this screen,
differential display was performed on RNA isolated at one time point,
namely from GA-treated stem sections 2 h after start of treatment and RNA
isolated from stem sections incubated for the same amount of time in water
as control. Also, instead of using oligo dT primers end-labeled with y-
[32P]rATP, a-[33PJdATP was used in the PCR reaction to label the amplified
products. A Genomyx programmable DNA sequencer was purchased, which
yielded higher quality sequencing gels and provided cDNA bands of
reproducible sharpness. This screen was performed by Dr. Sandrine Jagoueix,
and her results are summarized below.

Following the screen with 240 primer combinations, 63 bands were
reamplified, 45 of which were tested on Northern blots (Table 2.1). Twenty-six
were not differentially expressed, and four gave no signal on Northern blots.
The remaining 15 PCR products were differentially expressed, but only four of
them had interesting expression patterns: dd12, dd14, (M35 and dd53. In the
case of seven, the intensity of the bands stayed the same in GA-treated stem
sections but decreased in water-treated stem sections. For four, the band
intensity decreased faster in GA-treated stem sections than in control stem
sections. PCR products of dd12 and dd 14 were purified from agarose gel and
inserted into the pGEM-T cloning vector. Northern blot analysis identified

inserts from both dd 12 and dd 14 to be clones derived from differentially

62

GA C

 

0.5 2.5 6.5 0.5 2.5 6.5 h

 

PCR clone 98/
histone H3

 

 

 

dd3

 

 

dd4

 

 

 

‘ WW ' . dd12

 

 

 

 

. - OsTMK

 

 

 

PCR clone D/
E37

 

Figure 2.6. Northern blots showing changes in transcript levels of differential-
ly expressed genes in the IM in response to GA. Each lane contains 20 pg
of total RNA, isolated from stem sections treated for 0.5 h, 2.5 h and 6.5 h
With GA3 or kept in water as control. Probes for histone H3, dd12 and E37
were random prime labeled clones of their respective differentially displayed
products. Probes for dd4 and OsTMK were derived from clones isolated from
the cDNA library. The probe for dd3 was riboprobe derived from the 3’ UTR
as described in Chapter 6.

63

expressed genes (Fig. 2.6). However, sequence analysis showed that dd12 and
dd14 were derived from the same gene. They both contained the arbitrary
primer H-AP54, but contained different oligo dT primers. The sequence of
dd12 did not identify its function. The PCR product of dd35 could not be
cloned, and repeated differential display failed to show the band again. The
PCR product of dd53 was cloned, but none of the three different inserts tested
showed differential expression (data not shown).

The pilot study (DD 1) had shown that, if a band is reproducibly
differentially displayed, it is mostly likely derived from a gene whose RNA
levels have changed. Of the 30 bands that were not differentially expressed or
gave no signal on Northern blots, 13 were tested again by differential display
with the same primer pairs that had originally identified them. Differential
display was reproduced for one band only. After reamplification and cloning,
an insert was identified showing differential expression on a Northern blot.
However, a time course experiment showed that the transcript levels
decreased in control stem sections, but remained constant in GA—treated stem
sections. The primer pair that identified dd 12 showed the differentially

displayed band dd 12 repeatedly, indicating that its detection was reproducible.

cDNA LIBRARY SCREENS

Inserts, corresponding to the differentially displayed products dd3, dd4
and dd 12, were used to screen the kgtll cDNA library. The library was also
screened to obtain the full-length clone for OsTMK, whose transcript levels
increased during GA treatment of stem sections as shown in Figure 2.6.
Northern blot analysis had indicated the approximate size of the full-length

clones. After the primary phages were identified, PCR with gene-specific

64

primers was performed to eliminate truncated cDNA clones. Inserts for dd3
and dd4 were used to screen the 2.4-kb average size-fractionated library. For
dd3, four phages were purified to homogeneity that had inserts ranging from
1 to 1.2 kb in size. For dd4, ten phages were purified to homogeneity, three of
which carried full-length inserts of 1.8 kb. The 3.6-kb average size-fractionated
library was screened with inserts corresponding to dd 12 and OsTMK. Four
phages carrying the OsTMK insert, were purified to homogeneity, three of
which had full-length inserts of 3.1 kb. For dd3, dd4, and OsTMK, the PCR
strategy with gene-specific primers was used to eliminate the purification of
severly truncated cDNA clones. It failed, however, to identify full-length
clones. For dd12 PCR analysis identified 12 of 18 primary phages to have full-
length inserts. One of these 12 phages was purified to homogeneity and
shown to carry a 2.3 kb insert. The sequence analysis of all four clones is given

in the following chapters.

2. 5. DISCUSSION

By means of differential display of mRNA, four clones were identified
whose transcript levels increased during GA treatment of stem sections. A
fifth clone, OsTMK, was identified serendipitously by Margret Sauter during a

library screen for cyclin genes.

OPTIMIZATION OF DD I

Initial control experiments were necessary to optimize the differential

display technique for our purpose. As little as 0.4 pg total RNA gave

65

reproducible and reamplifiable banding patterns although 0.02 pg of total
RNA is sufficient in some systems (Liang et al., 1993). The optimum dNTP
concentration in the PCR was 4 pM instead of 2 pM as suggested in the
original protocol (Liang and Pardee, 1992, Liang et al., 1993, Liang et al., 1994).
This change in concentration was also necessary in DD H to avoid laddering of
bands (see below). An often overlooked control is to assess whether the PCR
cycling conditions are Optimal. Thermal cyclers from different manufacturers
have different ramping times and may reach different final temperatures.
Using the suggested cycling conditions, we found that 42°C was too stringent
an annealing temperature for the anchored oligo dT primer. This was also
recently reported by Vielle—Calzada et al. (1996). While these authors used an
annealing temperature of 35°C to identify differentially displayed bands, for
us an annealing temperature of 40°C was sufficient for both primers to
contribute equally to the reaction.

After establishing that both primers contributed equally to the
amplification of PCR products, the use of 32P end-labeled oligo dT was
continued for several reasons. By replacing a—[35S]dATP with y—[37-P]ATP end-
labeled oligo dT, severe contamination of the room and the thermal cycler
was prevented. a-[35S]dATP forms high levels of volatile 355 -labeled
decomposition products during PCR (Trentmann et al., 1995). Also, the use of
end-labeled oligo dT primers gave sharper bands than end-labeled decamers
(Fig. 2.2). Less sharp bands were observed as well using a-[32P]dCTP in the
PCR to label products (Trentmann et al., 1995). Others have found that
successful differential display depended on amplification of products derived
from the 3' UTR (Tokuyama and Takeda, 1995; Graf et al., 1997). This may be
because differentially displayed products from the 3' UTR minimize artifacts

arising from variations in cDNA quality. Sequence analysis of many false

66

positives indicated the presence of the arbitrary primer sequence or an
unknown sequence at either end of the fragment, and the absence of the
anchored oligo dT primer sequence (Tokuyama and Takeda, 1995). The
authors alleviated the problem with 33P end-labeled anchored oligo dT
primers and concluded that this modification ensured amplification of the 3'
UTR sequences by both the anchored oligo dT primer and the decamer. In
another study, far fewer differences were observed than anticipated, and the
few differentially displayed products isolated proved to contain the decamer
on both ends of the fragment (Graf et al., 1997). By designing arbitrary primers
with an A / T content of 60-80%, which corresponds more closely to the A/ T
content of the 3' UTR, many more differentially expressed genes were
identified. In hindsight, this is an additional reason to use end-labeled oligo
dT primers to ensure amplification of 3' UTR. Liang et al. (1993) found no
change in banding pattern whether degenerate or specific anchored oligo dT
primers were used. As can be seen in Figure 2.3B, this was not the case in our
hands and may be due to using 32P end-labeled oligo dT primers as opposed to
a-[3SSJdATP in the PCR (Liang et al., 1993).

Further optimization did not lead to improved differential display
analysis. For example, different RNA isolation methods did not result in any
change in banding patterns as can be seen in Figure 2.4. The guanidine/acid-
phenol method used mostly in this thesis resulted in high-quality RNA based
on the facts that no degradation products were observed on Northern blots
and that several full-length clones of 2 to 3 kb were isolated from the cDNA
library. The reverse transcriptases used in this study had different specificities
for the template as can be seen in Figure 2.4. Some studies had suggested that
the use of two different reverse transcriptases may eliminate false positives

(Sung and Denman, 1997). In our case, this would have led to the elimination

67

of two out of three differentially expressed genes identified during DD I (the
arrows in Figure 2.4 correspond to histone H3 and dd3 respectively).
SUPERSCRIPT 11 RT may be a superior reverse transcriptase for generating
cDNA libraries. The rice genome has a high GC content, which can create
secondary structure problems in the RNA strand through which reverse
transcriptase cannot read. SUPERSCRIPT 11 RT can perform at higher
temperatures and has high processivity through GC-rich sequences. This
resulted in long cDNAs as evidenced by the many 2 to 3 kb full-length clones
obtained from the library. For differential display, however, amplification of
3’ UTR sequences is more preferable (see above). Therefore, the advantage of

SUPERSCRIPT II RT over M-MLV was lost.

COMPARISON BETWEEN DD I AND DD 11

Although the banding patterns in DD I were reproducible, the bands
looked often fuzzy and the lanes contained high background, suggesting
suboptimal conditions for running of sequencing gels. Higher quality band
patterns were obtained by using a Genomyx programmable sequencer
(Averboukh et al., 1996). This resulted in consistenly sharp and reproducible
banding patterns, while also being easy to use. The use of longer primers in
the differential display (Zhao et al., 1995) yielded more reproducible banding
patterns and better separation of longer products on this programmable
sequencer (Averboukh et al., 1996). A new generation of GenHunter
differential display kits consisted of larger primers (Liang et al., 1994), and this
kit was used to perform a complete search for differentially expressed genes.

Both methods for differential display differ in the source of label and in

the kind of primers used. DD 11 made use of RNAimage kits which included

68

all reaction components except Taq polymerase. During PCR, a-[33P1dATP was
incorporated into the reaction products. Compared to DD I using 32P end-
labeled oligo dT primer, this may have resulted in a slight increase in number
of bands. This is because both strands carried label and may run differently on
a denaturing polyacrylamide gel. The advantage of using a-[33P]dATP is that
one less manipulation is necessary to complete differential display analysis.
The banding pattern in DD I was very reproducible but could vary slightly
from day to day. Using DD H, many spurious bands (differentially displayed
products between two control RNA samples on the same gel) were obtained
with several primer combinations (data not shown). This may be due to the
use of only one-base anchored oligo dT primer as compared to two-base
anchored oligo dT in DD 1. Also, during DD II, a laddering of bands was
visible, which was due to either concatemerized primers or concatemerized
short PCR products. This concatemerization may be the result of the
palindromic HindHI site at the very end of both primers. By increasing the
dNTP concentration to 4 pM in the PCR, the concatemerization was
suppressed, but the spurious bands remained (data not shown). While DD 11
should have given more efficient cDNA amplification and should have
minimized underrepresentation of certain RNA species (Liang et al., 1994), it
yielded, in our hands, differential display results that were inferior to those

obtained with DD 1.

FALSE-POSITIVES IN DIFFERENTIAL DISPLAY

Differential display analysis resulted in the identification of few
differentially expressed genes and many false positives. The number of false

positives compared to the number of differentially expressed genes was

69

inflated because, in DD H, 11 more genes were identified to be differentially
expressed. However, their transcript level stayed constant during GA
treatment but decreased in water-treated control tissue. It is difficult to
envision what their role would be in GA-regulated growth. However, a
significant number of such 'true' false positives still remained. Since few
convincing differentially displayed products were observed, many weak bands
were analyzed as well. Initially, dd3 and dd4 were identified as false positives.
The reamplified and labeled differentially displayed product dd4 gave no
signal on a Northern blot of total RNA (data not shown). However, it was
reproducibly differentially displayed and, therefore, reisolated and cloned.
With the insert from one of the colonies as probe, a signal could only be
detected on blots containing polyadenylated RNA. For dd3, at least three
different products were contained in the amplified differentially displayed
band because three bands were visible on a Northern blot. Two bands were
sharp and not differentially expressed. A third band was differentially
expressed, but was broad and faint. Therefore, this band was considered
background (data not shown). As can be seen in Figure 2.6, the authentic
signal for dd3 is broad and may have been detected on the initial Northern
blot.

Although all differential display was repeated side by side (as in Fig. 2.5)
with the same primer pair but different batches of RNA, this procedure was
not reproducible enough. For example, the band denoted by the star in Figure
2.4 was visible three times with three different RNA isolation methods but
was not observed at other times in differential display using the same primer
pair. The band was cloned and shown to be a false positive. The bands

denoted by the star in Figure 2.5 were false positives also. On repeated

70

differential display with the same primer pair, the bands were not observed
consistently (see Figure 2.4 around 213 bp marker).

In conclusion, instead of chasing down faint bands, it is better to repeat
differential display several times. After differentially displayed bands have
been identified reproducibly, it is better to clone and to analyze individual
inserts instead of using reamplified differentially displayed products as probe
for Northern blot analysis. This approach of repeating differentially display
and analyzing individual inserts is not feasible when many differential
displayed bands have to be evaluated. One approach to deal with this problem
is reverse Northern analysis with reverse-transcribed RNA from the different
treatments as probe. One can check for true positives either prior to cloning
(Zegzouti et al., 1997), or after cloning (Szczyglowski et al., 1997; Poirier et al.,
1997). However, differentially displayed products derived from rare
transcripts cannot be labeled as true or false positives directly because the

signal derived from those RNAs will be too weak.

COMPARISON OF DIFFERENTIAL DISPLAY AND SUBTRACTIVE
HYBRIDIZATION: FUTURE ANALYSIS OF DIFFERENTIAL GENE
EXPRESSION

Although an extensive search for differentially displayed genes was
performed, possibly covering >95% of all genes expressed (Liang and Pardee,
1992; Bauer et al., 1993), not all differentially regulated genes were identified.
Transcript levels for two cyclin genes increasing early during GA treatment
(Sauter et al., 1995) were not identified. A gene encoding expansin (Os-EXP4)
was not identified either, although its transcript levels clearly increased 2 h

after start of treatment (Cho and Kende, 1997). The reason for not detecting

71

the cyclin genes and Os-EXP4 may lay in the fact that differential display
cannot always discern between 3- to 6-fold changes in mRNA abundance.
Also, differential display may not detect every gene that is differentially
expressed. In a study comparing subtractive hybridization to differential
display indicated that both methods identified different sets of genes with
only two out of 56 clones being identified by both techniques (Wan et al.,
1996). Although the differential display was not exhaustive in that case, these
authors concluded that amplification in the differential display depended
mostly on the primer sequence and the product rather than on abundance,
whereas subtractive hybridization mostly dependended on mRNA
abundance.

Several new techniques for identifying genes with altered expression
patterns have been developed since the start of this work. Although
differential display has been a great improvement over differential expression
and subtractive hybridization and although this technique can be partially
automated (Bauer et al., 1993; Ito et al., 1994; Luehrsen et al., 1997), differential
display has its shortcomings (see above). Serial analysis of gene expression
(SAGE) can be employed to evaluate the patterns of gene expression in a
highly quantitative way with a 97% chance of detecting transcripts as low as
one copy per cell (Velculescu et al., 1995). In this method, short sequence tags
(11 to 15 base pairs) generated from defined positions within each mRNA
molecule are used, and expression patterns are deduced from the abundance
of individual tags. The sequence tags generated are long enough to
unequivocally determine from which gene they were derived either by
database searching (when working with an organism whose genome is
extensively studied or entirely sequenced), or by using these tags to identify

corresponding cDNA clones by screening a library. SAGE was used in the

72

analysis of cell-cycle-specific gene expression in yeast (Velculescu et al., 1997),
in studying gene expression profiles in normal and cancer cells (Zhang et al.,
1997), and in studying the p53-mediated gene regulation in humans (Polyak et
al., 1997), and in rat (Madden et al., 1997).

Recent developments made possible by technological improvements
and by using whole genome and EST databases have revived the use of
differential screening as a tool to identify genes with altered expression
patterns (Schena et al., 1995; Chee et al., 1996). Automated scanning for
predicted ORFs on each yeast chromosome was used to automatically select
PCR primers to amplify each ORF separately. The PCR products were spotted
onto microscope slides using high-speed robotics (Lashkari et al., 1997). These
so-called microarrays, containing up to 6100 clones or PCR products per 1.8
cm2, can be used in reverse Northern experiments. RNA isolated from
treated and untreated tissue is reverse transcribed and each is labeled with a
different fluorescent tag. The probes are hybridized to the same slide, and
signals are quantified using laser confocal scanning microscopy, giving a
detection limit as high as 1 mRNA molecule in 500,000 (Schena et al., 1996).
This technology will allow the identification of all differentially expressed

genes in a particular tissue under certain conditions in a matter of hours.

2.6. LITERATURE

Bauer, D., Muller, H., Reich, J., Riedel, H., Ahrenkiel, V., Warthoe, P. and
Strauss, M. (1993) Identification of differentially expressed mRNA species
by an improved display technique (DDRT-PCR). Nucleic Acids Res 21:4272-
4280.

73

Bleecker, A.B., Schuette, ].L. and Kende, H. (1986) Anatomical analysis of
growth and developmental patterns in the internode of deepwater rice.
Planta 169:490-497.

Chee, M., Yang, R., Hubbell, E., Berno, A., Huang, X.C., Stern, D., Winkler, ].,
Lockhart, D.]., Morris, MS. and Fodor, S.P.A. (1996) Accessing genetic
information with high-density DNA arrays. Science 274:610-614.

Cho, H.-T. and Kende, H. (1997) Expression of expansin genes is correlated
with growth in deepwater rice. Plant Cell 9:1661-1671.

Chomczynski, P. and Sacchi, N. (1987) Single-step method of RNA isolation
by acid guanidinium thiocyanate phenol chloroform extraction. Anal
Biochem 162:156-159.

Dreses-Werringloer, U., Fischer, K., Wachter, E., Link, TA. and Fluegge, UL
(1991) cDNA sequence and deduced amino acid sequence of the precursor
of the 37-kDa inner envelope membrane polypeptide from spinach
chloroplasts--Its transit peptide contains an amphiphilic alpha-helix as the
only detectable structural element. Eur J Biochem 195:361-368.

Graf, D., Fisher, AG. and Merkenschlager, M. (1997) Rational primer design
greatly improves differential display-PCR (DD-PCR). Nucleic Acids Res
25:2239-2240.

Ito, T., Kito, K., Adati, N., Mitsui, Y., Hagiwara, H. and Sakaki, Y. (1994)
Fluorescent differential display: arbitrarily primed RT-PCR fingerprinting
on an automated DNA sequencer. FEBS Letters 351:231-236.

Lashkari, D.A., DeRisi, ].L., McCusker, J.H., Namath, A.F., Gentile, C., Hwang,
S.Y., Brown, PO. and Davis, R.W. (1997) Yeast microarrays for genome

wide parallel genetic and gene expression analysis. Proc Natl Acad Sci USA
94:13057-13062.

Liang, P., and Pardee, AB. (1992) Differential display of eukaryotic messenger
RNA by means of the polymerase chain reaction. Science 257:967-971.

Liang, P., Averboukh, L. and Pardee, AB. (1993) Distribution and cloning of
eukaryotic mRNAs by means of differential display: refinements and
optimization. Nucleic Acids Res 21:3269-3275.

Liang, P., Zhu, W., Zhang, X., Guo, Z., O'Connell, R.P., Averboukh, L., Wang,

F. and Pardee, AB. (1994) Differential display using one-base anchored
oligo dT primers. Nucleic Acids Res 22:5763-5764.

74

Luehrsen, K. R., Marr, L.L., van der Knaap, E. and Cumberledge, S. (1997)
Analysis of differential display RT-PCR products using fluorescent primers
and GeneScan software. BioTechniques 22:168-174.

Madden, S.L., Galella, E.A., Zhu, J., Bertelsen, AH. and Beaudry, GA. (1997)
SAGE transcript profiles for p53-dependent growth regulation. Oncogene
15:1079-1085.

Métraux, JR, and Kende, H. (1983) The role of ethylene in the growth
response of submerged deepwater rice. Plant Physiol 72:441-446.

Poirier, G.M.-C., Pyati, J., Wan, ].S. and Erlander, MG. (1997) Screening
differentially expressed cDNA clones obtained by differential display using
amplified RNA. Nucleic Acids Res 25:913-914.

Polyak, K., Xia, Y., Zweier, J.L., Kinzler, K.W. and Vogelstein, B. (1997) A
model for p53-induced apoptosis. Nature 389:300-305.

Raskin, I., and Kende, H. (1984a) Role of gibberellin in the growth response of
submerged deep water rice. Plant Physiol 76:947-950.

Raskin, I., and Kende, H. (1984b) Regulation of growth in stem sections of
deep-water rice. Planta 160:66-72.

Sargent, TD. and Dawid, LB. (1983) Differential gene expression in the
gastrula of X. laevis. Science 222:135-139.

Sauter, M. and Kende, H. (1992) Gibberellin-induced growth and regulation of
the cell division cycle in deepwater rice. Planta 188:362-368.

Sauter, M., Mekhedov, S.L. and Kende, H. (1995) Gibberellin promotes
histone H1 kinase activity and the expression of cdc2 and cyclin genes

during the induction of rapid growth in deepwater rice internodes. Plant]
7:623-632.

Schena, M., Shalon, D., Davis, R.W. and Brown PO. (1995) Quantitative
monitoring of gene expression patterns with a complementary DNA
microarray. Science 270:467-470.

Schena, M., Shalon, D., Heller, R., Chai, A., Brown, PO. and Davis, R.W.
(1996) Parallel human genome analysis: microarray-based expression
monitoring of 1000 genes. Proc Natl Acad Sci USA 93:10614-10619.

St. John, T.P. anf Davis, R.W. (1979) Isolation of galactose-inducible DNA

sequences from S. cerevisiae by differential plaque filter hybridization. Cell
16:443-452.

75

Sung, Y.]., and Denman, RB. (1997) Use of two reverse transcriptases
eliminates false-positive results in differential display. BioTechniques
23:462-464.

Szczyglowski, K., Hamburger, D., Kapranov, P. and de Bruijn, F]. (1997)
Construction of a Lotus japonicus late nodulin expressed sequence tag
library and identification of novel nodule-specific genes. Plant Physiol
114:1335-1346.

Tokuyama, Y. and Takeda, J. (1995) Use of 33P-labeled primer increases the
sensitivity and specificity of mRNA differential display. BioTechniques
18:424-425.

Trentmann, S.M., van der Knaap, E. and Kende, H. (1995) Alternatives to 35S
as a label for the differential display of eukaryotic messenger RNA. Science
267:1186-1187.

Van der Knaap, E. and Kende H. (1995) Identification of a gibberellin-induced
gene in deepwater rice using differential display of mRNA. Plant Mol Biol
28:589-592.

Velculescu, V.E., Zhang, L., Vogelstein, B. and Kinzler, K.W. (1995) Serial
analysis of gene expression. Science 270:484-487.

Velculescu, V.E., Zhang, L., Zhou, W., Vogelstein, J., Basrai, M.A., Bassett Jr,
D.E., Hieter, P., Vogelstein, B. and Kinzler, K.W. (1997) Characterization of
the yeast transcriptome. Cell 88:243-251.

Vielle-Calzada, J.P., Nuccio, M.L., Budiman, M.A., Thomas, T.L., Burson, B.L.,
Hussey, MA. and Wing, RA. (1996) Comparative gene expression in

sexual and apomictic ovaries of Pennisetum ciliare (L.) Link. Plant Mol
Biol 32:1085-1092.

Wan, J.S., Sharp, 8]., Poirier, G.M.-C., Wagaman, P.C., Chambers, J., Pyati, ].,
Hom, Y.-L., Galindo, J.E., Huvar, A., Peterson, P.A., Jackson, MR. and
Erlander, MG. (1996) Cloning differentially expressed mRNAs. Nature
Biotech 14:1685-1691.

Zegzouti, H., Marty, C., Jones, B., Bouquin, T., Latche, A., Pech, J.C. and
Bouzayen, M. (1997) Improved screening of cDNAs generated by mRNA
differential display enables the selection of true positives and the isolation
of weakly expressed messages. Plant Mol Biol Rep 15:236-245.

76

Zhang, L., Zhou, W., Velculescu, V.E., Kern, S.E., Hruban, R.H., Hamilton,
S.R., Vogelstein, B. and Kinzler, K.W. (1997) Gene expression profiles in
normal and cancer cells. Science 276:1268-1272.

Zhao, S., Ooi, S.L. and Pardee, AB. (1995) New primer strategy improves
precision of differential display. BioTechniques 18:842-850.

77

CHAPTER 3

Expression of an ortholog of replication protein A1 (RPA1) is induced
by gibberellin in deepwater rice

3.1. ABSTRACT

Internodes of deepwater rice are induced to grow rapidly when plants
become submerged. This adaptation enables deepwater rice to keep part of its
foliage above the rising flood waters during the monsoon season and to avoid
drowning. This growth response is, ultimately, elicited by the plant hormone
gibberellin (GA). The primary target tissue for GA action is the intercalary
meristem of the internode. Using differential display of mRNA, we have
isolated a number of genes whose expression in the intercalary meristem is
regulated by GA. The product of one of these genes was identified as an
ortholog of replication protein A1 (RPA1). RPA is a heterotrimeric protein
involved in DNA replication, recombination, and repair and also in
regulation of transcription. A chimeric construct, in which the single-
stranded DNA-binding domain of rice RPA1 was spliced into the
corresponding region of yeast RPA1, was able to complement a yeast rpal
mutant. The transcript level of rice RPA1 is high in tissues containing
dividing cells. RPA1 mRNA levels increase rapidly in the intercalary
meristem during submergence and treatment with GA prior to the increase

in the level of histone H3 mRNA, a marker for DNA replication.

78

3.2. INTRODUCTION

Deepwater rice (Oryza sativa L.) is a subsistence crop in regions of
Southeast Asia that are flooded during the monsoon season (Catling, 1992).
To avoid drowning, it has evolved the capacity to elongate very rapidly when
it becomes submerged. This adaptation permits deepwater rice to keep part of
its foliage above the rising flood waters. In the flood plains of Bangladesh,
elongation rates of up to 25 cm day‘1 have been reported (Vergara et al., 1976);
in our laboratory, we have measured growth rates of up to 5 m h’1 (Stiinzi
and Kende, 1989). Understanding the physiological and molecular basis of the
growth response in deepwater rice is important for two reasons. It may help
in identifying the gene(s) that could confer elongation capacity onto modern,
high-yielding rice cultivars. Deepwater rice is also an excellent object in which
to study basic aspects of plant growth because of its unusually high growth
rate, which is under environmental and hormonal control.

In earlier work, we examined the environmental and hormonal
regulation of the growth response in deepwater rice and the cellular basis of
rapid internodal elongation. The plant hormone ethylene accumulates in
submerged internodes because of enhanced synthesis under reduced partial
pressures of Oz and because of its low rate of diffusion from the plant into the
surrounding water (Raskin and Kende, 1984a). The interaction of ethylene
and two other plant hormones i.e. abscisic acid, and gibberellin (GA),
determines the growth rate of the plant. Ethylene renders the internode more
responsive to GA (Raskin and Kende, 1984b), at least in part by lowering the
level of endogenous abscisic acid, a potent antagonist of GA action in rice
(Hoffmann-Benning and Kende, 1992). Growth of the internode is,
ultimately, promoted by GA (Raskin and Kende, 1984b).

79

The primary target tissue of GA is the intercalary meristem of the
internode, where GA enhances cell division activity and cell elongation
(Raskin and Kende, 1984b; Sauter and Kende, 1992b; Sauter et al., 1993). The
intercalary meristem is a zone of about 3 mm in length and is located at the
base of the internode (Bleecker et al., 1986). Cells are displaced from the
intercalary meristem into the elongation zone, where they reach their final
size. In GA—treated internodes, the final cell length is about three to four
times longer than in control internodes (Raskin and Kende, 1984b; Sauter and
Kende, 1992b). Correspondingly, the length of the elongation zone expands
from about 10 to 35 mm upon treatment with GA (Sauter et al., 1993). Growth
stops above the elongation zone in the differentiation zone, where
lignification of developing metaxylem and cortical sclerenchyma takes place
(Bleecker et al., 1986; Sauter and Kende, 1992a).

We have investigated the effect of GA on cell division activity in the
intercalary meristem and have correlated the progression of cells through the
cell cycle to molecular events that regulate it. The fraction of meristematic
cells in the G2 phase declined within 4 h of GA treatment, indicating that
these cells had entered mitosis (Sauter and Kende, 1992b). The expression of
two cyclin genes in the intercalary meristem was enhanced by GA, and the
time course of induction was compatible with a role for both cyclins in
regulating the transition from the G2 to M phase (Sauter et al., 1995). Between
4 and 7 h of incubation in GA, the rate of [3H]thymidine incorporation into
DNA doubled, showing an increase in DNA synthesis (Sauter and Kende,
1992b). In a screen for GA-regulated gene expression using differential display
of mRNA (Liang and Pardee, 1992), we identified a histone H3 gene whose
transcript level increased in parallel with the GA-induced rise in DNA

synthesis (Van der Knaap and Kende, 1995). In this report, we are describing

80

the results of further screening, which led to the identification of a gene that
encodes an ortholog of replication protein A1 (RPA1). In the internode, this
gene is expressed in the intercalary meristem, and its transcript level increases

as a result of GA treatment and submergence.

3.3. MATERIALS AND METHODS
PLANT MATERIAL

Deepwater rice (Oryza sativa L., cv. Pin Gaew 56) was obtained from the
International Rice Research Institute, Los Bafios, Philippines, and grown as
described (Stiinzi and Kende, 1989). For submergence experiments, 12-week-
old plants were partially immersed in deionized water (Métraux and Kende,
1983) and kept under continuous light. Stem sections containing the growing
internode were excised and treated with 50 pM gibberellin A3 (gibberellic acid,
GA3) for the periods indicated (Raskin and Kende, 1984a).

DIFFERENTIAL DISPLAY OF mRNA

RNA was isolated according to Puissant and Houdebine (1990) from the
intercalary meristem of internodes treated for 2 h with GA3 or distilled water
as control. Prior to reverse transcription, the RNA was treated with DNase I
(Boehringer Mannheim) to remove residual DNA. Differential display (Liang
and Pardee, 1992) was performed with RNAimageTM kits (GenHunter,
Nashville, TN) with slight modifications of the manufacturer's specifications.

Briefly, 0.2 pg of RNA was reverse transcribed in a total volume of 20 pl in

81

the presence of 20 pM dNTP and 0.2 pM H-T11M for 1 h. Two pl of cDNA was
amplified with the same H-T11M (0.2 pM) primer and 0.2 pM H-AP primer in
the presence of 4 pM dNTP and 0.25 pl a-[33PJdATP (2000 Ci / mmol, New
England Nuclear) in a total volume of 20 pl. PCR conditions were: 95°C for 30
sec, 40°C for 2 min, 72°C for 1 min over 40 cycles. PCR products were separated
on a 6% DNA sequencing gel in a GenomnyR programmable DNA sequencer
(Foster City, CA) and visualized by autoradiography. Using the primers 5'-
AAGCTTTTTTTTTTTA-B' and 5'-AAGCTTTTGAGGT-3', a differentially
displayed cDNA band, dd12, was identified. After re-amplification using the
above PCR conditions but at a dNTP concentration of 20 pM in a total
volume of 40 pl, the dd 12 cDNA was ligated into the pGEM-T vector
(Promega). Differential expression of the dd12 transcript was confirmed by

Northern blotting.

NORTHERN BLOT ANALYSIS

Twenty pg of total RNA was loaded on 1.2% agarose-formaldehyde gels
(Ausubel et al., 1987) and transferred to Hybond N membrane (Amersham).
Blots were prehybridized in 5x SSC, 10x Denhardt's solution (Sambrook et al.,
1989), 0.1% SDS, 0.1 M K-phosphate, pH 6.8, and 100 pg/ ml denatured salmon
sperm DNA for 4 h at 42°C, and hybridized in 5x SSC, 10x Denhardt's
solution, 0.1 M K-phosphate, pH 6.8, 10% dextran sulphate, 30% formamide,
and 100 pg/ ml denatured salmon sperm DNA overnight at 42°C with a probe
prepared in the presence of a-[32P]dCTP (New England Nuclear) using a
random prime labeling kit (Boehringer Mannheim). High-stringency washes
were performed with 0.1x SSC and 0.1% SDS at 65°C twice for 30 min. The

radioactivity on blots was quantified by Phosphorlmager analysis (Molecular

82

Dynamics, Sunnyvale, CA). All values were normalized for equal loading
with E37, a cDNA corresponding to a transcript whose expression does not

change during treatment with GA3 (Van der Knaap and Kende, 1995).

SEQUENCE ANALYSIS

A full-length cDNA clone corresponding to the PCR product dd12 and
henceforth called DD12 was isolated from a rice intercalary meristem cDNA
library and cloned into the EcoRI site of pBluescript SK(-) phagemid
(Stratagene). Sequence analysis was performed at the W.M. Keck facility at

Yale University.
IN VITRO MUTAGENESIS

DD12 was subcloned into the EcoRI-Sphl site of pALTERR-l (Promega). The
Sphl site of DD12 was located 150 bp from the 3' end of DDIZ within the 3'
untranslated region. To create an NdeI cloning site at the 5' end of the
nucleotide sequence encoding the single-stranded DNA-binding domain
SBD-A and a BamHI site at the 3' end of the nucleotide sequence encoding
SBD-B, 5'-GGCGCGGTTGCATATGACGAGAAGAGTT-3' and 5'-
GCTTGGTATGACGGATCCGGCAAGGGTACT-3', respectively, were used as
primers for mutagenesis (Fig. 3.1A). The internal NdeI site was eliminated by
mutagenesis using 5'-AGCTAGGGCCTTATGTTGGTG-3' as primer. The
construct thus formed was called pALTER-Sl. In vitro mutagenesis was
performed according to manufacturer's (Promega) specifications.

Mutagenesis over the primer regions was verified by sequence analysis.

83

YEAST COMPLEMENTATION

pDSI, a yeast shuttle vector containing wild-type yeast RPA1 on pRS415
(LEU2) (Philipova et al., 1996) was digested with NdeI and BamHI to release
the region of the yeast SBD-A and SBD—B domains. pALTER-Sl was digested
with NdeI and BamHI, and the rice SBD-A and SBD-B region was cloned into
the NdeI-BamHI site of pDSI. This resulted in the replacement of the yeast
SBD-A and SBD—B with the rice SBD-A and SBD-B, and this construct was
called pDS9. The yeast strain SBY102 (MATa, adeZ-I, can1-100, leu2-3,112,
his3-11, ura3-1, Arpal::TRP1;) (Philipova et al., 1996) containing wild-type
RPA1 on the shuttle vector YCp50 (URA3) was transformed with the
appropriate construct by a modified LiAc method (Schiestl and Gietz, 1989)
and selected on synthetic complete medium without leucine (Sherman, 1991).
To remove the wild-type yeast RPA1 on YCp50, the colonies were selected on

the same medium containing 5-fluoroorotic acid (Boeke et al., 1987).

3.4. RESULTS

IDENTIFICATION OF A GA-REGULATED GENE

We employed differential display of mRNA (Liang and Pardee, 1992;
Van der Knaap and Kende, 1995) to identify genes whose transcript level in
the intercalary meristem was altered within 2 h of GA treatment. A 239-bp
differentially displayed PCR product, dd12, appeared in GA-treated tissue and
was further investigated. The cDNA was reamplified, cloned, and used as

probe to verify by Northern blot analysis that the corresponding transcript

84

A

 

 

 

 

 

 

 

 

 

 

 

 

 

msnmmpummspmrmxrmmmrmmx 60
mmmmsmgmmmmmqmnmsnmm 120
nomnusrmsmmmrumgmmgmonmm 180
.. .. ._ .. .. W . . 340
300
360
frmnmagmmmnmsmsnmmrvmnuxmg 420
immsmmsmmnvommsmmsnnsmmmsm no
560
nsmmmummmmm 600
oumsmvnnmumnmmc 630
B J x. Iaevis
l H. sapiens
D. melanogaster
S. cerevr‘siae
S. pombe
. C. fasciculata
‘1 0. sativa
C. elegans

 

Figure 3.1. Amino acid sequence and phylogenetic analysis of DD12 (Os-RPA1).
(A) The lightly shaded amino acid sequences comprise SBD-A and SBD-B, the
darker shaded sequence the zinc finger domain. The location of the primers used

for in vitro mutagenesis are underlined. (B) Phylogenetic analysis of RPA1

proteins from the species indicated. The dendogram was constructed using the
Clustal method with the PAM250 residue weight table. The accession numbers

are: 001588 (Xenopus Iaevis), P27694 (Homo sapiens), 270277 (Drosophila

melanogaster), P22336 (Saccharomyces cerevisiae), U59385 (Schizosaccharo-
myces pombe), $38458 (Cn'thidia fasciculata), AF009179 (Oryza sativa), U41535

(Caenorhabditis elegans).

85

indeed accumulated as a result of GA treatment (results not shown, see also
Fig. 3.4). A rice internode-specific cDNA library was screened with the
differentially displayed and subcloned dd12 PCR product. A full-length clone
of 2.3 kb was isolated whose sequence showed an open reading frame from
position 55 to 1944, encoding a protein of 69.6 kDa predicted molecular mass
(Fig. 3.1A). Database searches indicated amino acid similarity to RPA1 from
other organisms. RPA complexes are heterotrimers with subunits of
approximately 70 (RPA1), 3O (RPAZ), and 14 (RPA3) kDa (Wold, 1997). RPA
was first identified as a factor necessary to support SV4O replication (Wobbe et
al., 1987; Fairman and Stillman, 1988; Wold and Kelly, 1988). Later, it was also
found to be necessary for recombination (Heyer et al., 1990; Moore et al., 1991)
and for DNA repair (Coverley et al., 1991; 1992). DD12 encodes a protein
containing two contiguous single-stranded DNA-binding domains, SBD—A
and SBD—B (Fig. 3.1A, lightly shaded amino acid sequences), which share
similarity with E. coli single-stranded DNA-binding domains (Philipova et
al., 1996). DD12 also encodes a zinc finger motif (Fig. 3.1A, dark-shaded amino
acid sequences), which is conserved in all RPA1 proteins but whose function
is unknown. The phylogenetic relationships between all known RPA1 genes
in the data base are given in Figure 3.18. The percentage amino acid identity
between rice and other RPA1 proteins ranges from 33.3% (H. sapiens) to 24.5%
(C. fasciculata) based on pairwise comparisons using ALIGN. The percentage
amino acid identity of the SBD-A and SBD-B region varies from 44.9% (H.
sapiens) to 33.1% (C. elegans).

YEAST COMPLEMENTATION.

It has been suggested that species-specific interactions between RPA and

86

A

pDSQ . .
rice rice
SBD-A SBD-B
182 287 426 630

 

Figure 3.2. Complementation of a yeast ma1 mutant with p089. (A) Schematic
representation of the chimeric protein encoded by p089 containing the rice
SBD-A and SBD-B domains (shaded) between the yeast N-tenninal and C-ter-
minal regions. The numbers above the diagram represent the amino acid posi-
tions of Os-RPA1. (B) Transformed yeast was plated out on complete synthetic
medium (minus leucine, plus 5-fluorootic acid). Sector a, SBY102 non-transfor-
med; sector b, SBY102 transformed with pDS1, which contains the wild-type
RPA1 of yeast, sector c, SBY102 transformed with p089, sector d, SBY102
transformed with pJM241; which contains the wild- -type RPA2 of yeast on
pFiS415 (LEU2).

87

other cellular components account for the inability of yeast RPA to function
in SV4O DNA replication (Brill and Stillman, 1989) and for the failure of
human RPAZ to complement a yeast rpaZ mutant (Brill and Stillman, 1991).
The C-terminal and N-terminal regions of RPA1 interact with other cellular
factors (see below) and are less conserved among each other than are SBD—A
and SBD-B. Replacement of yeast SBD—A and SBD—B with human SBD—A and
SBD—B was shown before to rescue a yeast rpa1 mutant (Philipova et al., 1996).
We constructed a chimeric clone, pDSQ, encoding a protein with the rice SBD-
A and SBD-B domains between the yeast N-terminal and C-terminal domains
(Fig. 3.2A). Figure 3.23 shows the result of the complementation experiment.
Both pDSI containing the wild-type yeast RPA1 (sector b), and pDSQ
containing the rice-yeast chimera (sector c) rescued the yeast mutant, whereas
le241 containing yeast RPAZ (sector (1) did not. After selection, the colonies
in sectors b and c were not able to grow on synthetic complete medium
without uracil, which indicates loss of the original yeast RPA1 gene on the
YCp50 vector (data not shown). Also, dot blot analysis showed the presence of
the yeast-rice chimera in colonies from sector c (data not shown). These
results confirmed that the protein encoded by D012 is an ortholog of RPA1.
Therefore, D012 will henceforth be called Os-RPAI.

TISSUE-SPECIFIC EXPRESSION OF Os—RPAI

Os-RPAI mRNA was detected in tissues that contain dividing cells,
namely in the highest node, which includes the apical meristem; in the
sheath of the second youngest leaf; in the youngest leaf; in root tips; and in
the coleoptile (Fig. 3.3). Os-RPAI transcript was also expressed in the

intercalary meristem of the internode; a trace of Os-RPAI mRNA was also

88

Internode

N2 N1 L2b L23 L1 Co Flo 0-3 3-8 8-18 old

 

 

 

“”“nflmm ”can“ E37

 

 

 

Figure 3.3. Tissue-specific expression of Os-RPA1 in rice. N2, second
highest node; N1, highest node containing the shoot apex; L2b, basal

2 cm of second youngest leaf blade; L23, basal 2 cm of second
youngest leaf sheath; L1, youngest leaf; Co, coleoptile 3 days after
germination; Flo, root 3 days after germination; 0-3, internodal region 0-
3 mm above N2 containing the IM; 3-8, internodal region 3-8 mm above
N2 containing mostly the EZ; 8-18, internodal region 8-18 mm above N2
containing the upper part of the EZ and the DZ; old, oldest part of the
internode. The upper panel shows hybridization signals with Os-RPA1
as probe and the lower panel shows hybridization signals with E37 as
internal loading control.

89

evident in the internodal region just above it, which probably still contains
some meristematic cells at its base but which, otherwise, consists of
elongating cells. No Os-RPAI transcript was detected in the differentiation

zone and in the oldest part of the internode.

THE TIME COURSE OF Os-RPAI EXPRESSION AND ITS CORRELATION
WITH THE CELL CYCLE.

The level of Os-RPAI transcript in the intercalary meristem increased
after 2 to 3 h of treatment with GA3 and reached a maximum after 8 h (Fig.
3.4A, top panel). The increase in transcript level of Os-RPAI was not observed
in control stem sections (data not shown). Submergence of whole plants also
caused an increase in accumulation of Os-RPAI mRNA (Fig. 3.43, top panel).
These same blots were also hybridized to 1337 as loading control (Fig. 3.4A and
B; bottom panels).

Because RPA is involved in DNA replication, we were interested in
correlating the increase in transcript level of Os—RPAI to that of histone H3,
which is a marker for the S-phase of the cell cycle (Van der Knaap and Kende,
1995). For this purpose, the Northern blots shown in Figure 3.4A and B were
hybridized to the histone H3 probe (Fig. 3.4A and B, center panels). The
signals were quantified by Phosphorlmager analysis and normalized for equal
loading using E37 as internal standard. Taking a three-fold higher mRNA
level over the O-h time point as a significant increase, we found that the rise
in Os-RPAI transcript level preceded that of histone H3 by 2 h in GA-treated
internodes (Fig. 3.4C) and by 4 h in internodes of submerged plants (Fig. 3.4D).

9O

A B-

0123456781524 02468101524

 

 

 

 

 

 

 

    
 

    
 

 

 

 

 

16 .. so ..
E 14 .. 3
4o ..
i: 12 -r 5
£3 10 « + Os-FiPA1 .‘é- 30 0
8 4- Histone H3 3 -0- Os-RPAt
5 8 5 -°- Histone H3
3 6 .. b 20 1
3 .3
a: 4 a
'5 a 10 '
a: 2 .. a:
O t i 'r 4 O u 3
o 10 20 30 0 1o 20 30
9 GA treatment (h) Submergence (h)

Figure 3.4. Change in Os-FiPA1 transcript levels during GA treatment of rice
stem sections and during submergence of whole plants. (A) Northern blot
analysis of RNA from the intercalary meristem of stem sections treated with GA3
for the periods in hours indicated above the lanes. The same blot was hybridized
to Os-RPA 1, the histone H3 cDNA probe, and E37. (B) Northern blot analysis of
RNA from the intercalary meristem of plants submerged for the periods in hours
indicated above the lanes. The same blot was hybridized to Os-FiPA 1, the
histone H3 cDNA probe, and E37. (C) Quantitative analysis of the Northern blot
shown in (A) using a Phosphorlmager. All values were normalized to the E37
loading control. (D) Quantitative analysis of the Northern blot shown in (A).

91

3.5. DISCUSSION

In our search for (SA-regulated genes in deepwater rice, we identified a
gene, DD12, whose transcript level increased early in response to GA
treatment and submergence. Sequence analysis of DDIZ indicated similarity to
RPA1 genes from several organisms. Replacement of the SBD domain of
yeast RPA1 with the homologous domain from DD12 yielded a construct that
was used successfully to complement a yeast rpa1 mutant. Based on these
results, we concluded that DD12 is an ortholog of RPA1 and called it Os-RPAI.
To date, Os-RPAI is the only identified plant RPA1 gene in the database,
although apparent Arabidopsis and maize homologs exist in the EST
database.

In synchronized yeast cells, increased transcript levels of RPA1
correlated with the late G1 to S—phase (Brill and Stillman, 1991). This is also
the case for the expression of several other yeast replication genes whose
transcript levels increased prior to the accumulation of histone H2A-HZB
mRNA (Lowndes et al., 1991). We found a similar trend in the intercalary
meristem of rice internodes. Particularly during submergence (Fig. 3.43 and
D), the increase in Os-RPAI mRNA levels preceded the onset of DNA
replication, as indicated by the accumulation of histone H3 mRNA (Van der
Knaap and Kende, 1995). On the basis of these results, it appears that in rice, as
in yeast, expression of replication proteins is regulated differently than that of
histones.

RPA1 encodes the largest subunit of the heterotrimeric complex RPA
and contains three functional domains (Figs. 3.1A and 3.2A). The C-terminal
region mediates the interaction with the other two subunits, RPA2 and RPA3
(33—35). The primary function of SBD-A and SBD-B is to bind single-stranded

92

DNA (Philipova et al., 1996; Kim et al., 1996; Lin et al., 1996; Bochkarev et al.,
1997); however SBD is also able to bind to damaged and double-stranded DNA
(see below, and Clugston et al, 1992; He et al., 1995; Burns et al., 1996). The N-
terminal domain is important for protein-protein interactions, e.g., with the
transcriptional activators GAL4 and VP16 (He et al., 1993; Li and Botchan,
1993), with the tumor suppressor p53 (Dutta et al., 1993), and possibly with
other proteins of DNA metabolism. However, the precise role of RPA in
these interactions is unclear (for a recent comprehensive review of RPA
structure and function, see Wold, 1997).

In addition to its role in DNA replication, repair, and recombination,
RPA is implicated in transcriptional regulation of several genes. RPA was
identified as a protein factor that bound specifically to the upstream
repression sequence (URS) of the promoter of the yeast CARI gene, which is
involved in nitrogen metabolism (Luche et al., 1993), to a similar element in
the promoter of the yeast MAG gene involved in DNA repair (Singh and
Samson, 1995), and to a similar element in the promoter of the yeast P OX3
gene, which is required for peroxisome functioning (Einderhand et al., 1995).
Based on sequence similarity, several cis elements, to which RPA binds, were
identified in the promoters of many more genes involved in basic cell
metabolism. Functional analysis of these cis elements showed that most of
them act as upstream repression sequences, and some as upstream activation
sequences. This indicates a role for RPA in both transcriptional repression
and activation and in coordination of gene expression. The involvement of
RPA in transcriptional regulation is found not only in yeast. The
transcription of the human metallothionine HA gene is repressed by RPA
both in vitro and in viva (Tang et al., 1996).

In conclusion, rice RPA is probably involved in submergence- and GA-

93

enhanced DNA replication. In addition, it may also play a role in

coordinating general transcription that accompanies accelerated growth.

3.6. LITERATURE

Ausubel, F.M., Brent, R., Kingston, R.E., Moore, D.D., Seidman, I.G., Smith,
J.A. & Struhl, K. (1987) Current Protocols in Molecular Biology. Wiley,
New York, pp. 4.9.1.-4.9.8.

Bleecker, A.B., Schuette, IL. and Kende, H. (1986) Anatomical analyisis of
growth and developmental patterns in the internode of deepwater rice.
Planta 169:490-497.

Bochkarev, A., Pfuetzner, R.A., Edwards, A.M. and Frappier, L. (1997)
Structure of the single-stranded-DNA-binding domain of replication
protein A bound to DNA. Nature 285:176-181.

Boeke, ].D., Trueheart, ]., Natsoulis, G. and Fink, GR. (1987) S-Fluoroorotic
acid as a selective agent in yeast molecular genetics. Methods Enzymol
154:164-175.

Brill, S.]. and Stillman, B. (1991) Replication factor-A from Saccharomyces
cerevisiae is encoded by three essential genes coordinately expressed at S
phase. Genes Dev 5:1589-1600.

Brill, SJ. and Stillman, B. (1989) Yeast replication factor-A functions in the
unwinding of the SV40 origin of DNA replication. Nature 342:92-95.

Burns, J.L., Guzder, S.N., Sung, P., Prakash, S. and Prakash, L. (1996) An
affinity of human replication protein A for ultraviolet-damaged DNA.
Implications for damage recognition in nucleotide excision repair. J Biol
Chem 271:11607-11610.

Catling, D. (1992) Rice in Deep Water. The MacMillan Press Ltd, London and
Basingstoke, UK.

Clugston, C.K., McLaughlin, K., Kenny, MK. and Brown, R. (1992) Binding of
the human single-stranded-DNA binding~protein to DNA damaged by the

anticancer drug cis-diamminedichloroplatinum (H). Cancer Res 52:6375-
6379.

Coverley, D., Kenny, M.K., Munn, M., Rupp, W.D., Lane, DP. and Wood, RD.
(1991) Requirement for the replication protein SSB in human excision

94

repair. Nature 349:538-541.

Coverley, D., Kenny, M.K., Lane, DP. and Wood, RD. (1992) A role for the
human single-stranded-DNA binding-protein HSSB/RPA in an early stage
of nucleotide excision repair. Nucleic Acids Res 20:3873-3880.

Dutta, A., Ruppert, ].M., Aster, ].C. and Winchester, E. (1993) Inhibition of
DNA replication factor RPA by p53. Nature 365279-82.

Einerhand, A.W., Kos, W., Smart, W.C., Kal, A.]., Tabak, H.F. and Cooper,
T.G. (1995) The upstream region of the FOX3 gene encoding peroxisomal 3-
oxoacyl-coenzyme A thiolase in Saccharomyces cerevisiae contains ABFl-
and replication protein A-binding sites that participate in its regulation by
glucose repression. Mol Cell Biol 15:3405-3414.

Fairman, MP and Stillman, B. (1988) Cellular factors required for multiple
stages of SV40 DNA-replication in vitro. EMBO] 7:1211-1218.

Gomes, X.V. and Wold, MS. (1995) Structural analysis of human replication
protein A. Mapping fucntional domains of the 70—kDa subunit. J Biol
Chem 270:4534-4543.

He, Z.G., Brinton, B.T., Greenblatt, ]., Hassell, J.A. and Ingles CJ. (1993) The
transactivator proteins VP16 and GAL4 bind replication factor A. Cell
73:1223-1232.

He, Z.G., Henricksen, L.A., Wold, MS. and Ingles, CJ. (1995) RPA
involvement in the damage-recognition and incision steps of nucleotide
excision-repair. Nature 374:566-569.

Heyer, W.-D., Rao, M.R.S., Erdile, L.F., Kelly, TJ. and Kolodner, RD. (1990)
An essential Saccharomyces cerevisiae single-stranded DNA binding

protein is homologous to the large subunit of human RP-A. EMBO J
9:2321-2329.

Hoffmann-Benning, S. and Kende, H. (1992) On the role of abscisic acid and
gibberellin in the regulation of growth in rice. Plant Physiol 99:1156-1161.

Kim, D.K., Stigger, E. and Lee, SH. (1996) Role of the 70-kDa subunit of
human replication protein A (I). Single-stranded DNA binding activity,

but not polymerase stimulatory activity, is required for DNA replication. J
Biol Chem 271:15124-15129.

Li, R. and Botchan, MR. (1993) The acidic transcriptional activation domains
of VP16 and p53 bind the cellular replication protein A and stimulate in
vitro BPV-l DNA replication. Cell 73:1207-1221.

95

Liang, P., and Pardee, AB. (1992) Differential display of eukaryotic messenger
RNA by means of the polymerase chain reaction. Science 257:967-971.

Lin, Y.L., Chen, C., Keshav, K.E., Winchester, E. and Dutta, A. (1996)
Dissection of functional domains of the human DNA replication protein
complex Replication Protein A. J Biol Chem 271:17190-17198.

Lowndes, N.F., Johnson, AL. and Johnston L.H. (1991) Coordination of
expression of DNA synthesis genes in budding yeast by a cell-cycle
regulated trans factor. Nature 350:247-250.

Luche, R.M., Smart, W.C., Marion, T., Tillman, M., Sumrada, RA. and
Cooper, T.G. (1993) Saccharomyces cerevisiae BUF protein binds to
sequences participating in DNA replication in addition to those mediating
transcriptional repression (URSl) and activation. Mol Cell Biol 13:5749-
5761.

Métraux, ].-P. and Kende, H. (1983) The role of ethylene in the growth
response of submerged deepwater rice. Plant Physiol 72:441-446.

Moore, S.P., Erdile, L., Kelly, T. and Fishel, R. (1991) The human homologous
pairing protein HPP-l is specifically stimulated by the cognate single-
stranded binding-protein hRP-A. Proc Natl Acad Sci USA 88:9067-9071.

Philipova, D., Mullen, ].R., Maniar, H.S., Lu, L., Gu, C. & Brill SJ. (1996) A
hierarchy of SSB protomers in replication protein A. Genes Dev 10:2222-
2233.

Puissant, C. and Houdebine, L.M. (1990) An improvement of the single-step
method of RNA isolation by acid guanidium thiocyanate-phenol-
chloroform extraction. BioTechniques 8:148-149.

Raskin, I. and Kende, H. (1984a) Regulation of growth in stem sections of
deepwater rice. Planta 160166-72.

Raskin, I. and Kende, H. (1984b) Role of gibberellin in the growth response of
submerged deepwater rice. Plant Physiol 762947-950.

Sambrook, ]., Fritsch, E.F. & Maniatis, T. (1989) Molecular Cloning: A

Laboratory Manual. Cold Spring Harbor Lab. Press, Plainview, N.Y., 2nd
Ed., p. B15.

Sauter, M. and Kende, H. (1992a) Levels of B-glucan and lignin in elongating
internodes of deepwater rice. Plant Cell Physiol 33:1089-1097.

96

Sauter, M. and Kende, H. (1992b) Gibberellin-induced growth and regulation
of the cell division cycle in deepwater rice. Planta 188:362-368.

Sauter, M., Seagull, R.W. and Kende, H. (1993) Internodal elongation and
orientation of cellulose microfibrils and microtubules in deepwater rice.
Planta 190:354-362.

Sauter, M., Mekhedov, S.L. and Kende, H. (1995) Gibberellin promotes
histone H1 kinase activity and the expression of cdc2 and cyclin genes
during the induction of rapid growth in deepwater rice internodes. Plant J
7:623-632.

Schiestl, RH. and Gietz, RD. (1989) High efficiency transformation of intact
yeast cells using single stranded nucleic acids as a carrier. Curr Genet
16:339-346.

Sherman, F. (1991) Getting started with yeast. Methods Enzymol 194:3-21.

Singh, K.K. and Samson, L. (1995) Replication protein A binds to regulatory
elements in yeast DNA repair and DNA metabolism genes. Proc Natl Acad
Sci USA 92:4907-4911.

Stiinzi, IT and Kende, H. (1989) Gas composition in the internal air spaces of
deepwater rice in relation to growth induced by submergence. Plant Cell
Physiol 30:49-56.

Tang, C.M., Tomkinson, A.E., Lane, W.S., Wold, MS. and Seto, E. (1996)
Replication protein A is a component of a complex that binds the human
metallothionein IIA gene transcription start site. J Biol Chem 271:21637-
21644.

Van der Knaap, E. and Kende, H. (1995) Identification of a gibberellin-induced
gene in deepwater rice using differential display of mRNA. Pl Mol Biol
28:589-592.

Vergara, B.S., Jackson, B. & De Datta SK. (1977) Deep water rice and its
response to deep water stress. In Climate and Rice: proceedings of the 1976

symposium. International Rice Research Institute, Los Banos, Philippines,
pp. 301-319.

Wobbe, C.R., Weissbach, L., Borowiec, J.A., Dean, F.B., Murakami, Y., Bullock,
P. and Hurwitz. J. (1987) Replication of simian virus-40 origin-containing
DNA in vitro with purified proteins. Proc Natl Acad Sci USA 84:1834-1838.

Wold, MS. and Kelly, T. (1988) Purification and characterization of

replication protein-A, a cellular protein required for in vitro replication of

97

simian virus-4O DNA. Proc Natl Acad Sci USA 85:2523-2527.
Wold, MS. (1997) Replication Protein A1: a heterotrimeric, single-stranded

DNA-binding protein required for eukaryotic DNA metabolism. Annu
Rev Biochem 66:61-91.

98

CHAPTER 4

Identification of a gibberellin-induced leucine-rich-repeat receptor-
like kinase in deepwater rice and its interaction with a kinase-

associated protein phosphatase

4.1. ABSTRACT

A leucine-rich-repeat receptor—like kinase, OsTMK, was identified in
deepwater rice. The transcript levels of OsTMK increased in response to
gibberellin (GA) in the internode of rice. Furthermore, the expression was
high in regions undergoing cell division and elongation, suggesting a role for
this gene in growth. The kinase domain of OsTMK constituted an active
kinase, autophosphorylating on serine and threonine residues. A putative
downstream signal transduction component is the rice homolog of kinase-
associated protein phosphatase (KAPP). The kinase interaction domain of
OsKAPP was efficiently phosphorylated by the kinase domain of OsTMK. In
rice, the expression patterns of OsKAPP and OsTMK were found to be similar,

suggesting an in vivo interaction.

99

4.2. INTRODUCTION

The succesful existence of all plants is dependent upon their ability to
coordinate complex developmental changes and to sense and respond to
fluctuations in their environment. A stimulus is perceived by the cell, a
signal is generated and transmitted (signal transduction), and a biochemical
change is elicited (the response). Deepwater rice belongs to a group of rice
cultivars whose survival during flooding is based on its capacity for rapid
increase of internodal elongation when it becomes submerged. Under field
conditions growth rates of up to 25 cm/ day have been reported which result
in plants that are up to 6 m tall (Catling, 1992). The signal for accelerated
growth is an increase in ethylene levels which, via a decrease in abscisic
levels (Hoffman-Benning and Kende, 1992), enhances the responsiveness of
the internode to GA (Raskin and Kende, 1984a). While ethylene, in part,
signals the change in the environment, the growth response is, ultimately,
elicited by GA. Recently, several signal transduction components have been
identified that mediate responses to GA. Two putative transcription factors
with high sequence similarity to each other, GAI (Peng et al., 1997) and RGA
(Silverstone et al., 1998), were shown to mediate responses to GA. Epistasis
experiments have indicated that downstream of GAI is SPY. SPY encodes a
protein with O-linked N-acetylglucosamine transferase activity which may
modify post-translationally target proteins in the GA signaling pathway
(J acobsen et al., 1996; Olszewski, 1997). As of yet, no specific receptor, kinase,
phosphatase, channel protein, or heterotrimeric G-protein has been identified
and shown to mediate responses to GA. However, in the cereal aleurone
system, studied extensively to understand the role of GA and ABA in

synthesis and secretion of hydrolytic enzymes, several signal transduction

100

components have been implicated (Bethke et al., 1997; Ritchie and Gilroy,
1998; Jones et al., 1998). Furthermore, circumstantial evidence has pointed to
the plasma membrane as the site of GA perception (Hooley et al., 1991; Gilroy
and Jones, 1994).

Many signals are initially perceived by transmembrane receptors, a
large number of which function by activation of an intrinsic protein kinase
domain. In recent years, many plant receptor-like kinases (RLK) have been
identified. While the majority of animal receptor kinases autophosphorylate
on tyrosine residues, the majority of plant RLKs autophosphorylate on serine
and / or threonine residues (Braun and Walker, 1996), and to date no plant
kinase has been identified that exclusively autophosporylates on tyrosine
residues. One RLK however, PRKl (Mu et al., 1994), confers dual-specificity,
phosphorylating on tyrosine as well as serine residues. A related group of
plant RLKs shows sequence similarity to members of the two-component
signal transduction systems from bacteria. The intracellular domain of these
RLKs shows sequence similarity to histidine kinases, and this group
comprises the ethylene receptors (Chang et al., 1992; Schaller and Bleecker,
1995; Wilkinson et al., 1995) and a gene, CKI 1, whose encoded protein is
involved in cytokinin signaling (Kakimoto, 1996). The red / far-red light-
absorbing phytochromes are not transmembrane receptors but are located in
the cytosol. However, they also show limited sequence similarity to histidine
kinases (Elich and Chory, 1997). Autophosphorylation of these receptor-like
histidine kinases has not been conclusively shown and, in the case of
phytochrome, may not be at a conserved histidine residue (Elich and Chory,
1997)

The group of RLKs, which, based on sequence similarity, are predicted

or have been shown to have substrate specificity for serine and threonine

101

residues, can be classified according to the amino acid sequence features in the
proposed extracellular domains. The two largest subclasses are: (i) RLKs
whose extracellular domain contain leucine-rich repeats (LR), and (ii) RLKs
whose extracellular domain shows sequence similarity to the S-locus
glycoprotein. The extracellular domain of several other plant RLKs identified
are unique.

We are interested in the role of GA in vegetative growth in the
internodes of deepwater rice. In a search for GA-regulated transcripts, a
leucine-rich repeat receptor-like kinase (LRR-RLK), OsTMK, was identified.
Direct involvement of this RLK in GA signaling cannot be confirmed at this
time. However, the expression of OsTMK increased during GA treatment of
rice stem sections, which suggests a role for this gene in plant growth. A
potential downstream signal transduction component is a kinase-associated
protein phosphatase (KAPP), originally identified in Arabidopsis by its in
vitro interaction with RLK5, a LRR-RLK (Stone et al., 1994). Interaction
between the kinase domain of OsTMK and the kinase interaction domain of

OsKAPP was, therefore, also investigated.

4.3. MATERIALS AND METHODS

PLANT MATERIAL

Seeds of deepwater rice (Oryza sativa L., cv. Pin Gaew 56) were obtained from
the International Rice Research Institute (Los Banos, Philippines). Plants were
grown as described in Stiinzi and Kende (1989). Twenty-cm-long stem sections

containing the growing internode were excised and treated with 50 uM GA3

102

(Raskin and Kende, 1984b). Incubation was allowed to proceed for the periods
indicated, after which the different regions were excised, frozen immediately,

and stored at -80°C until use.

IDENTIFICATION OF OsTMK

A partial cDNA clone, OsKIN of 913 bp, was serendipitously identified in our
laboratory by Margret Sauter during a library screen for cyclin genes. This
clone had no similarity to cyclins but had similarity to protein kinases. A 305—
bp fragment at the 5'-end of OsKIN was used to screen a intercalary meristem-
specific, unamplified cDNA library to obtain a full-length clone, OsTMK. The
phage insert was inserted into the Nail site of pBluescript SK(-) phagemid
(Stratagene) and the sequence was determined by the Biochemistry Facility of
the Plant Research Laboratory at Michigan State University, East Lansing, MI;
by the WM Keck Facility at Yale University, New Haven, CT; and by
Genomyx, Fostercity, CA. The sequences were aligned using Sequencher,

version 3.0 (Gene Codes Corporation).

NORTHERN BLOT ANALYSIS

Twenty ug of total RNA, isolated according to Puissant and Houdebine (1990),
was electrophoretically separated in a 1.2% formaldehyde-agarose gel
(Ausubel et al., 1987) and transferred to Hybond-N membrane (Amersham).
DNA fragments containing the OsKIN insert, the ribosomal SS RNA binding
protein insert (RL5; Kim and Wu, 1993), or the E37 insert (Van der Knaap and
Kende, 1995) were isolated from agarose gels by digestion with B-agarase

(NEB). Fifty ng of template DNA was labeled in the presence of a-[32P]dCTP

103

(3000 Ci/mmol, NEN), using a random prime labeling kit (BMB). Northern
blots were prehybridized for 4 h at 42°C in 5x SSC, 10x Denhardt's solution,
0.1% SDS, 0.1 M K-PO4, pH 6.8, and 100 ug/ ml denatured salmon sperm DNA
and hybridized overnight at 42°C in 5x SSC, 10x Denhardt's solution, 0.1 M K-
PO4, pH 6.8, 100 ug / ml denatured salmon sperm DNA, 10% dextrane sulfate,
and 30% formamide (BMB). The blots were washed twice in 2x SSC and 0.5%
SDS and twice in 0.1x SSC and 0.1% SDS at 65°C for 30 min each. For OsKAPP,
an RNA probe was made from the region encoding the kinase interaction
domain (KID; Song et al., 1998), in the presence of a-[32P]UTP (NEN). Blots
were prehybridized and hybridized in 3x SSPE, 10x Denhardt's solution, 0.5%
SDS, 50 ug/ ml denatured salmon sperm DNA, and 50% formamide at 65°C
overnight. Blots were washed twice for 5 min each in 2x SSC and 0.5% SDS at
65°C followed by twice in 0.1x SSC and 0.5% SDS for 30 min each. The
radioactivity on blots was quantified by Phosphorlmager analysis (Molecular

Dynamics).

SOUTHERN BLOT ANALYSIS

Rice genomic DNA was isolated from a CsCl gradient according to Ausubel et
al. (1987). Four ug of genomic DNA was digested with the appropriate
restriction enzyme, and the fragments were separated on a 0.8% agarose gel
for 20 h at 30 V. The gel was treated for 15 min in 0.25 N HCl, 30 min in 0.5 N
NaOH and 1.5 M NaCl and 15 min in 1 M Tris-HCl, pH 8 and 1.5 M NaCl,
after which the DNA was transferred to Hybond N+ (Amersham). A probe
was made using the insert of OsKIN in the presence of a-[32P]dCTP (NEN)
using a random prime labeling kit (BMB). Blots were prehybridized and
hybridized in 6x SSC, 5x Denhardt's solution, and 1% SDS overnight at 65°C.

104

Non-specific hybridization was removed by stringent washes in 0.1x SSC and

0.1% SDS at 65°C.

OVEREXPRESSION OF THE KINASE DOMAIN OF OsTMK IN E. coli

To facilitate the cloning of the kinase domain of OsTMK in frame to maltose
binding protein (MBP), a restriction enzyme site was introduced via PCR. The
primers used were 5'-ATGGAATTCTCAATTCAAGTCCTC-B' and the reverse
primer present in pBluescript SK (-). The product was amplified from a
plasmid containing the full-length clone with Pwo polymerase (BMB). The
PCR product was digested with EcoRI and HindIII and inserted directionally
in the same sites of pMAL-cRI (NEB). The construct thus created, pMBP-
OsTKD, was sequenced over the primer region to verify that the proper
fusion construct had been obtained. pMBP-OsTKD was introduced in E. coli
host ER2508 (NEB). The cells were grown in rich medium (10 g bacto tryptone,
5 g yeast extract, 5 g NaCl, and 2 g glucose per liter) supplemented with 1 mM
MnClz and 50 ug/ ml carbenicillin (Sigma) at 37°C until an OD600 of 0.6 was
reached. The cells were induced by addition of 50 uM isopropyl B-D-
thiogalactoside for 2 h at room temperature, after which the cells were
harvested by centrifugation. The bacterial pellet was resuspended in lysis
buffer (10 mM Tris-HCl, pH 7.3, 150 mM NaCl, 1 mM DTT, 0.1% Tween-20,
and lmM phenylmethylsulfonyl fluoride), and stored at -20°C overnight. The
cells were lysed with a cup sonicator at 30% duty cycle, 4-5 probe setting, or
with a French press. MBP-OsTKD fusion protein (83 kD) was allowed to bind
to amylose resin (NEB) for 20 min, and the resin was washed several times
with lysis buffer. The final wash was performed in lysis buffer without

Tween-20, and the fusion proteins were eluted in 10 mM maltose and 10 mM

105

Tris-HCl, pH 7. The GST-OsKID fusion construct was kindly provided by W.-
Y. Song and PC. Ronald (University of California at Davis) and contained the
rice KID fused in-frame to GST (Song et al., 1998). Overexpression of GST-
OsKID (55 kD) was performed essentially the same as described for MBP-
OsTKD, except that MnClz was omitted from the rich medium. After lysis of
the cells, the fusion proteins were purified by affinity binding to glutathione
agarose resin (Sigma). After several washes, the fusion protein was eluted in
10 mM reduced glutathione and 10 mM Tris-HCl, pH 7. Protein
concentrations were determined by comparison of band intensity to known

standards on a Coomassie blue-stained polyacrylamide gel.

AUTO- AND TRANSPHOSPHORYLATION ASSAYS AND
PHOSPHOAMINO ACID ANALYSIS

Purified MBP-OsTKD in the presence or absence of GST-OsKID, was incubated
in a total volume of 15 pl with 10 uCi y-[37-P]ATP (6000 Ci / mmol, NEN) + 20
uM non-radioactive ATP, 50 mM Tris-HCl, pH 7.3, 1 mM DTT, and 10 mM
MnClz. The reaction was allowed to proceed for the appropriate time at room
temperature or at 30°C, after which the reaction was stopped by addition of 35
ul of ice-cold 10% trichloroacetic acid (TCA). The radiolabeled proteins were
collected by centrifugation, washed with 50 ul of ice-cold 10% TCA, and
resuspended in 4 ul Tris-base and 5 p.11 2x SDS sample buffer (BioRad). The
resuspended proteins were directly loaded on a 10 or 12% polyacrylamide gel
and electrophoretically separated with a constant current of 15 mA. The gel
was stained in Coomassie blue to verify equal loading, dried, and X-ray film
was exposed to the gel at room temperature. The radioactivity was quantified

by Phosphorlmager analysis. For phosphoamino acid analysis, the

106

aut0phosphorylated MBP-OsTKD was eluted from polyacrylamide gel
overnight in 50 mM NH4HCO3. The sample was precipitated by TCA and acid
hydrolyzed in 6 N HCl at 110°C for 1 h. The HCl was evaporated, and the
pellet was resuspended in pH 1.9 electrophoresis buffer (2.2% formic acid and
7.8% acetic acid) containing phosphoamino acid standards, and applied to a
thin-layer cellulose plate (TLC, Merck) as described by Boyle et al. (1991).
Samples were subjected to electophoresis at 1.5 kV for 20 min in pH 1.9 buffer
in the first dimension followed by electrophoresis in pH 3.5 buffer (5% acetic
acid and 0.5% pyridine) at 1.3 kV for 16 min in the second dimension.
Phosphoamino acids standards were visualized by spraying the plate with
0.25% ninhydrin in acetone and incubating the plate at 65°C for 30 min. X-ray

film was exposed to the TLC plate for 2 days at -80°C.

4. 4 RESULTS

SEQUENCE ANALYSIS OF OsTMK

Screening of the rice cDNA library with the partial clone OsKIN
resulted in the isolation of three independent inserts of approximately 3100
bp. This size was similar to the size expected from Northern blot analysis,
indicating that a full-length clone was isolated. DNA sequence analysis of the
longest cDNA insert (3123 bp) showed an ORF from position 50 to 2935,
yielding a predicted protein of 101.6 kD (see Fig. 4.1). No in frame stop codon
was observed upstream of the putative start methionine. However, one rice
EST, identical to OsTMK (D41598), was found to be extended by 5 nucleotides

at the 5' end and showed an in frame stopcodon. The putative signal

107

sequence is followed by an extracellular leucine-rich repeat (LRR) domain
containing eleven complete and two incomplete repeats (see Fig. 4.2A), and
eight potential N-glycosylation sites (consensus N-X-S/ T). The first ten LRRs
are flanked by two cysteine residues spaced eleven and eight amino acids
apart, respectively. One cysteine pair (spaced eight amino acids apart) is found
at the N-terminal side of the last three LRRs. LRRs are believed to play a role
in protein-protein interaction, and several LRR domains are flanked by
cysteine clusters (Kobe and Deisenhofer, 1994; see Fig. 4.23). The extracellular
domain is followed by a putative transmembrane region, and the C-terminal
intracellular portion of the protein containing the 12 characteristic kinase
subdomains; the conserved residues of which indicate serine/threonine
kinase activity (Hanks and Quinn, 1991). Database searches with the BLAST
program (Altschul et al., 1990) indicated high amino acid similarity to RLKs
from plants. The highest similarity was to TMK1 from Arabidopsis (Chang et
al., 1992), followed by an ORF (locus 2213607 on BAC F21J9) with accession
number AC000103. We, therefore, named the gene encoded by this insert
OsTMK. As can be seen in Figure 4.1, the alignment of OsTMK and TMK1
shows high amino acid identity and almost no gaps. In contrast, the
alignment of AC000103 to OsTMK and to TMK1 shows gaps. The
serine/glycine-rich stretch between the last LRRs and the transmembrane
region is missing. Also, less amino acid sequence similarity is found
immediately outside the conserved kinase domains. The spacing of the LRRs
and the presence of the cysteine pairs is conserved in all three proteins. Figure
4.2A shows the alignment of the LRRs in the three proteins. The consensus,
which is shown below the alignment, is commonly found in other plant
proteins containing LRRs (Li and Chory, 1997), except for the presence of S/ T

and S at position 5 and 6, respectively. Figure 4.23 shows the alignment of the

108

mmw
ham
wan

mom
bow
mmw

mmv
ham
bmm

omv
va
ave

ovm
mmm
mmm

0mm
mvm
wmm

owH
vma
owa

on
«D
om

 

IIIIHIIme om >IH m oIIIq m x > m IIIQIIIomHHmmII
IIIquIoIIqIIm mm m .. . a m .III<mIIu¢HIIzII
oanmqumeqmqummxzmmqmomqeoomz>m>>qHmmzooqu>oqqm>gzmmm>xeq>< .xmmzqoxzwz>o« V<erwom

     

 

 

 

> >H HHH HH
IIII .HI MWQII>I <9 llllllll HIIdI DQI dI ZO mmOmII> OEUAZmHIIIdeOIImOOOIIZ
IIIII HI IDmIIIIIImIIIIIIIJIII¢IIZOI ZQO>WI BquI VBDmH Du mm mmz > 2w:
amox l>2mamm229>zmd>OHmH>SZUEM>>IHDmmwmm<0m>fimm<<400 Z>ZOO<>BHX>EDQOOm$mmr>>Z<mmm
xxWOZIxIDJhIH<>OHIQQ¢AAHIHIII>I¥IIZIm¢xxI QZIZmUQ mmIOEBZIHBPZI

mmOmImIImXII30I>IdOHIhHmJJUIQIII>IIIHmFmIMZImmDMIIZHImOmOOImmlmlmmIJWWXIIIIDIIIIBZI>II mmI
O>m0mmx0xmx>0¥m04¢¢QQO>O<H<O>>m0<HHO¢mmmmmm0 ZmOQOOZOOQm{ZmOOmOOmOmm4m<QXOHZQZOZ¢M>J>Z¥<<W

WMI>mIoImIxI>IqexI2mIMIquIHIIquIOmIzH>mIm<mQ¢mmIIIeIquqszIHIIIoeoeIeHI>3Im ....... memz>
hoI>IImIMImI>Ia§IzIIeSIoIHzIIqIIHoIIHmIImxéImmIIIeISmemqu IIIII 2mI<HIIzze IIIIIII mmIImm
9mqzo>4222mqo>meq<mq¢<>mxm>eoeszoaqHaxoqeeHon<mmHucmmozm<mza>eHzo<ouo>oHwonomozoxgzo<qm<

EII, .

IIoI<m>HmIIez ..... meIIIoIIImZJozaomxHom<eIz IIIIIIIII oI¢IomIIoImmIImIIaImIIImI>2mmmII> IIIII .
IIoImmIHaIImm IIIII mIImIIIImZmoxoaI>mImmmI>I>IIIImIIIIzI>xIImImIIIa IIIII mmIIIImImmImx IIIII .
m>omo<<>mqqam>mmoomoomemaummmeem>>o<x>xo4wxmemcoupzzeqe>xeamoaxqamonoooeawmnmamqnmqmncwmmc.

mIIImImIooIe>zeIIIIzoqIIIrIqummeIo-IZI>I>mwommzzImmo>qumexIesIImmoxommqsIIHmIquuz>III.
wIIIme IIIII >IonIIqu>eIaIea onII-ImI>0moIIIqumImazIIIImIHmIoImmo mIanIImImIIIm .....
wumoeunzmmazqwoamezzmummHmeszZmoH02253u<oqm¢e<qm¢m>momzxzm¢qmamoammqaeI;ommomq<0e>zaez<mn

wo>III<zmImmIH>Iw IIIIIIIII :oImmIImIIImI>wIIoIaz<w>a>IIm IIIIII HmIaqumz>ImzeImIIuoqumIzImHo
IoI I¢m<zmI mmI HmI mx IIIII Hm- mo mIIIoI I>I mIImoI zIIquI >III«IImIII>mIIIzI zoIIIImI mmIIIqu m- eI o- o
azeqmsooaqn<mSeamammzzQQm>¢¢gaegoxmmommHmemozzm>qq>oammqo¢ammeOmHvzomqmqm<q<eqzm>mm<umoea

 

r

IIoIHIHIInv2m<IImez mstIImoIIZmzuquImomaIHI>ImImmmm >z>q<HmuquqmszHz
“mmIHIHmIIIxeweI>HmeI xIIImnIIImmmmszmququIqu oomI Ixmaquaqmemmmqqme IIxxI
wm20>o>aepmo<mooe>w<3<mmmmumooeu3c;axcaL gacuaaeccaaao>>qz><

 

 

mOH00004
HMS?
KEBmO

moHoooUC
axZB
xEFmO

moaoooU¢
aXZF
XEEMO

moaoooU<
HxEB
XEEmC

moaooou<
HxEB
xzemo

moaoooUd
HXSB
XSEmO

moHoooU4
HXEE
XZEmO

MOH00004
H¥ZB
XSEwO

109

mmw
mhw
mmw

vv:
mm:
mom

3mm? 6:50 :8 9.85 :85 :_ 82mo_n:_ 9m
m:_mEou 88:} m5 :_ moocmacmm 828:8 of. .wowmsx £99: :_ 8:2 288038 N? 85 2 88: :85

:_ £895: :25: 9:. .8:__8_u:: 83:8 88. 98: .8885 m oz» .885 :_ 8806:. w_ EmEon m”: 8:...
282688: .EmEoo m:_::mam m:mEEmEm:m= 88.85 9: :8 88:08 .m:m_m 8285: 9.: 9m moxon 8:5
9:. .EmE:m__m 85 :_ mama 2865 80me 8:0 Ewan m 3 88062 88 8:8 05:8 .8382 N... >> .935

>3 8:95 882, 808.68 9:. .morooooxx :8 5:2... 22.50 :8283 82.368 88:8w ._..v 959”.

mmm IIoonIemmqmmIImIIIIqIII

oImomqugIIImeIwIIIIHImIImmmm
mvm III> IIIIIIII w IIIIII z IIIIII mIqIIImmmqumoI> IIIIIIIIII m IIIIIII HImmzo I

mmm moo<memmm¢20mmmemHmeoezoqm<qmmme¢oommmmomm<ozqu<oaqezoqoHo>moamom o
eIIIo>>Im>IIII>IzII umIIImmIoz IIIII HzHImIIIIz>mz¢mIIIImIm wIsz IIIIII <II>mmImm<> IIIIIII qI
IIIIqm>ImIIH IIIIIII I> IIIIIIIIIIIIII m IIIIIII oIIIIeIIIIxIm<mIzH>IIxIImIIIHImIIImIIIm IIIIII

mm¥3>deBmA>Z><IOEQ mmmm<UUIO<Jm4>Bm>m<AEmMBJOHBQQHdXOL onmqozmmk39>dm2mammOfimoqumOEHa

 

HK X

IIII I-» > IIIIIIIIIIIIIIIII HxIIHI I0 9. m IIIIII IIIIm IIIIIII IIIIm :IIe IIIIIIIII
IIII II» > IIIIIIIII . IIIIIIII HIIIHI o IIm IIIII IIIIm IIIIIII IIIImI zIIo IIIIIIIII
mquc .<xee>moe>< .L<q>wmeo<qmem>m>uxooam<qm>. e<>m<szooquyv:m:. mHmeoo<qmmq>m>wm<>

 

 

 

HHH> HH> DH> MH>

MOH00004
HXEB
MZBmO

moaooou<
HXEB
mZBmO

moaoooud
HMEB
KEBWO

110

 

      
  

   

 

 

 

1
2
i
.i 3
5
i
l' 4
I!
c
A 5
A
I A
F M 6
F M
s LA I F
5 LA G v 7
_N__FS D v
I SN M D
QN M D 8
QK M D
so a; DS
so .1 As 9
so 53
LR KF
LS VF 10
FE NF
PA
PE 11
_ PR
I ox I KE
GA I _} QB 12
AD F QE
v AA ‘1‘]?
TT 1'; GF 13
I AK ' RF
E xx L gs Lxx Ex Lxx ng §G xaP xx

consensus :

Figure 4.2. Comparison of conserved features in the putative extracellular
domains of OsTMK, TMK1, and ACOOO103.

111

N-terrninal C-terrninal

 

 

OsTMK

TMK1

AC000103

Xa21
BRIl
CLV1
ER
RLKS
PRKl
SERK
RPK1

 

OMWUH

 

 

 

 

 

 

 

XEC xW xnyCxx xgcg xxx§ CxP
H

consensus. I F L L

Figure 4.2. Comparison of conserved features in the putative extracellular
domains of OsTMK, TMK1, and AC000103. Amino acid residues with
identity to the consensus are indicated by the grey boxes. The “x” in the
consensus indicates any amino acid; the “a" indicates an aliphatic amino
acid. (A) Alignment of the eleven complete and two incomplete LRRs in
OsTMK, TMK1, and ACOOO103. The first LRR of all three proteins are
grouped as indicated by the horizontal bar and the number behind the
repeat. The alignment is followed by the second LRR and so forth. The gap
between LRR 10 and 11 is not shown. (B) On the left, alignment of the
cysteine pairs present N-terminal of the LRRs; on the right, alignment of the
cysteine pairs present C-terminal of the LRRs. The references are: TMK1,
Chang et al., 1992; ACOOO103, accession number for genomic sequence of
Arabidopsis BAC F21J9; Xa21, Song et al., 1995; BRl1, Li and Chory,
1997; CLV1, Clark etal., 1997; ER, Torii etal., 1996; RLK5, Walker, 1993;
PRK1, Mu etal., 1994; DcSERK, Schmidt etal., 1997; RPK1, Hong etal.,
1997. The cysteine pair in RPK1 is located between the first and second
LRR.

112

Table 4.1 . Amino acid identity between leucine-rich repeat receptor-like
kinases. The numbers indicate overall percent identity between the pairs,
based on pairwise comparisons using ALIGN. The last column shows the
number of repeats found in the presumed extracellular domain of the
kinases. The references are: TMK1, Chang et al., 1992; AC000103,
accession number for genomic sequence of Arabidopsis BAC F21J9; Xa21,
Song et al., 1995;BRI1, Li and Chory, 1997; CLV1, Clark et al., 1997;
ERECTA, Torii et al.,1996; RLK5, Walker, 1993; DcSERK, Schmidt et al.,
1997; PRK1, Mu et al., 1994.

 

 

OsTMK TMK1 number of LRR
OsTMK 13
TMK1 59.0 13
ACOOO103 49.2 53.3 13
Xa21 24.9 26.5 23
BRI1 24.4 25.2 25
CLV1 25.6 27.9 21
ERECTA 25.6 28.2 19
RLK5 25.8 27.2 21
DcSERK 23.3 22.6 5
PRK1 20.9 21.2 5

 

113

sequences surrounding the two cysteine residues located either N-terminal or
C-terminal of the LRRs in the different LRR-RLKs. Often, the cysteines are
spaced seven amino acids apart; for OsTMK, TMK1 and AC000103 they are
predominantly eight amino acids apart. The percent amino acid identity,
which includes the signal sequences and the transmembrane regions,
between several LRR-RLKs is shown in Table 4.1. While the amino acid
identity between OsTMK, TMK1 and AC000103 is approximately 55%, the
identity between all other LRR-RLK5 is around 25%. The kinase domain of
OsTMK is 79% identical to TMK1 and 63% to AC000103, while the LRR
domain of OsTMK is 49% identical to TMK1 and 40% to AC000103. This
indicates that OsTMK, TMK1 and AC000103 belong to a subgroup within the
larger LRR-RLK family.

EXPRESSION OF OsTMK IN GA-TREATED RICE INTERNODES AND
SOUTHERN BLOT ANALYSIS

We were interested to know whether this gene plays a role in GA-
mediated growth. In rice stem sections, GA was shown to increase cell size
and the rate of cell production in the intercalary meristem (IM), which is
located at the base of the growing internode. GA also resulted in a 3- to 4-fold
increase in final cell size of cells in the elongation zone (EZ) after cells emerge
out of the IM (Bleecker et al., 1986; Sauter and Kende, 1992). Northern blot
analysis shown in Figure 4.3, indicated that the transcript levels for OsTMK
changed in response to GA in different regions of the internode. Figure 4.3A
shows a Northern blot containing RNA from the IM and the lower part of the
' EZ of GA-treated stem sections. The top panel shows a blot hybridized to

OsTMK, while in the bottom panel the same blot was hybridized to a gene

114

A0123456781524h

 

OsTMK

 

 

 

RL5

 

 

 

 

OsTMK

 

 

 

RL5

 

 

 

 

OsTMK

 

 

RL5

 

 

 

 

Figure 4.3. Expression of OsTMK in GA-treated stem sections. Stem sections
were incubated in 50 11M 6A3 for the times indicated above the lanes. (A) Nor-
thern blot containing RNA isolated from the 0-5 mm region above the second
highest node, including the IM and the lower part of the EZ. (B) 5-10 mm
above the second highest node, including the EZ. (C) 10-20 mm above the
second highest node, including the upper part of the EZ and DZ. To verify
equal loading, all blots were hybridized to RL5.

115

 

14.2- ., __

 

3.6-

2.3-
1.9-

 

 

 

 

Figure 4.4. Southern blot analysis of OsTMK in rice. Blot containing
genomic DNA, digested with either BamHl (B), EcoRl (E), Hindlll (H),
or Pstl (P). The blot was probed with random prime labeled OsKIN
insert, which corresponds to the kinase domain of the gene.

116

whose expression did not change significantly during treatment, RL5. The
signals were quantified by Phosphorlmager analysis and normalized for equal
loading with the signals derived from RL5. OsTMK transcript levels increased
after 3 h and continued to increase 9-fold, 15 h after the start of treatment.
Figure 4.33 and C show Northern blots containing RNA from the EZ and the
differentiation zone (DZ), respectively. The expression of OsTMK is very low
in the oldest part of the internode and did not change during GA treatment
(data not shown). The transcript levels for OsTMK did not increase either in
control stem sections (data not shown). As is evident from these analyses, the
transcript levels of OsTMK increase in growing tissues in response to GA.
Southern blot analysis shows that, following high stringency
hybridizations and washes, only one band was detected in each lane (see Fig.
4.4). An additional faint band is seen in the lane containing genomic DNA
digested by HindIII. This was expected since the probe used had an internal
site for this restriction enzyme. The size of the smallest band is 2 kb, which
eliminated the possibility that one fragment contained two linked copies of
OsTMK in the rice genome. The probe used was derived from the region
encoding the kinase domain, which, together with the presence of a single
band in each digest on the Southern blot, indicated the presence of a single

gene in the rice genome.

PHOSPHORYLATION ASSAYS WITH THE KINASE DOMAIN OF OsTMK

To examine whether OsTMK constituted an active kinase and to
determine which amino acid residues were phosphorylated, the kinase
domain was cloned in frame to MBP, and overexpressed in E. coli. The fusion

protein was purified and subjected to a phosphorylation assay. As can be seen

117

 

 

 

 

 

Relative phosphorylation

 

 

Relative amount of MBP-OsTKD

Figure 4.5. Autophosphorylation of MBP-TKD. (A) Purified MBP-OsTKD (500
ng) was allowed to autophosphorylate for 20 min at room temperature. The
right panel is an autoradiogram of the Coomassie blue-stained gel in the left
panel. (B) Phosphoamino acid analysis of aut0phosphorylated MBP-OsTKD.
(C) Autophosphorylation as a function of enzyme concentration. The amount
of MBP-OsTKD used ranged from 10 to 810 nM (12.5 ng to 1 pg per 15 pl
reaction). The phosphorylation reaction was performed at 309C for 20 min. The
open symbols reflect a curve expected for a first or second order reaction,
respectively. The closed symbols reflect the curve that was observed.

118

in Figure 4.5A, the overexpressed protein was purified to homogeneity and
was able to autophosphorylate. The aut0phosphorylated fusion protein was
eluted from the acrylamide gel and acid hydrolyzed to determine the substrate
specificity (Fig. 4.53). 32P labeled spots corresponding to the positions of
phosphothreonine and phosphoserine were detected, while no label
corresponding to phosphotyrosine was detected. This indicated that OsTMK,
like TMK1, is a serine/threonine protein kinase. Most RLKs undergo
intermolecular phosphorylation (second order with respect to enzyme
concentration; Horn and Walker, 1995). A first order reaction, or
intramolecular phosphorylation mechanism, would result in a linear
increase of phosporylation depending on enzyme concentration. As shown in
Figure 4.5C, MBP-OsTKD can phosphorylate itself via an intermolecular
mechanism. At low enzyme concentrations, the level of autophosphorylation
displayed a first order reaction mechanism, while at higher enzyme
concentrations, the level of autophosphorylation increased more than the
increase in MBP-OsTKD concentration. This suggests that the
autophosphorylation of OsTMK is dependent on receptor concentration and

is accomplished via a higher order reaction mechanism.

A potential downstream component of several RLKs in plants is KAPP,
originally identified in Arabidopsis as a protein that interacts in vitro with
the kinase domain of RLK5 (Stone et al., 1994). Arabidopsis and maize KAPP
have been shown to interact with TMK1 (Braun et al., 1997) via the kinase
interaction domain (KID). We were interested to find out whether the KID of
OsKAPP is a substrate for phosphorylation by OsTMK. As shown in Figure
4.6, MBP-OsTKD phosphorylated GST-OsKID, while it did not phosphorylate
GST. GST or GST-OsKID did not autophosphorylate or phosphorylate each

119

.3082: 8.88586 6.5835 Sm

mucus 826_ 8:... .o_v_mo-...m0 6:5 othOIAms. .m 88. Fww ucm 9:80-de .m 8:6.
5:56.82 9. no .4 25 Home 2.... 028.50 .m 28. 628.50 9. no .~ 2....
”.50 m... n... .F o8. ”853:8 ucm 2393:.2 E00: 8 :.E cm :8 8009: 2 826......
mm; 8.88: 8:... .88: to. 85 :_ .8 8538.833 8.88880 85 he 86528.28 :m
m. .88: £9: 8:... 88:38 mm 05.89.50 9.»: >88 5.83.5585. 6.8 952“.

 

 
 

 

 

 

 

50. I... II
.8 5!...3‘
8.8.50. 85 _
9.30-3.2- V
o m 4 m N F m m 4 m

Em:mo.8.o§< 8.88580

 

120

other. This indicates that GST-OsKID is a substrate for MBP-OsTKD and that
these proteins interact in vitro.

MBP-OsTKD and GST-OsKID are both substrates for phosphorylation.
Therefore, the level of phosphorylation was determined by using increasing
amount of fusion proteins. As shown in Figure 4.7A, increasing amounts of
MBP-OsTKD led to increase in autophosphorylation of MBP-OsTKD and
increase in phosphorylation of GST-OsKID. However, the level of
autophosphorylation was greatly increased by decreasing amounts of GST-
OsKID. Decrease in levels of autophosphorylation may be due, in part, to a
transfer of phosphate from enzyme to substrate. After MBP-OsTKD was fully
aut0phosphorylated, free ATP was removed. MBP-OsTKD was allowed to
further incubate in the presence of GST-OsKID. As can be seen in Figure 4.73,
no phosphate turnover of MBP-OsTKD was observed (compare lane 1 with
lanes 2 to 5 in Fig. 4.73). Also, no transfer of phosphate from MBP-OsTKD to
GST-OsKID was detected. OsKID became only labeled when it was incubated
in the presence of y—[32PJATP, as shown in lane 5 of Figure 4.73.
Autophosphorylation can be inhibited by increasing amounts of inactive
kinase (Horn and Walker, 1995; Williams et al., 1997). To investigate whether
inhibition of in vitro autophosphorylation of MBP-OsTKD was specific to its
substrates, a phosphorylation assay with MBP-OsTKD either in the presence
of OsKID, GST, or BSA was performed. As shown in Figure 4.7C,
autophosphorylation of MBP-OsTKD and phosphorylation of GST-OsKID
were inhibited at high protein concentrations, whether the proteins were

substrates or not.

121

MBP-OSTKD ‘ I

66 200 600 600 600 ng

E—T\

 

 

 

 

 

GST'OSK'D 200 200 200 66 22 ng
-MBP-OsTKD
-GST-OsKlD
B
Coomassie Autoradiogram
1 2 3 4 5 1 2 3 4 5
V“ ” -MBP-OsTKD
-GST-OsKlD

 

 

 

 

Figure 4.7. Decrease in autophosphorylation of MBP-OsTKD.

122

 

   

Coomassie Autoradiogram
1 2 3 4 1 2 3 4
”N we ,_ -MBP-OsTKD
, . -GST—OsKID
-GST

 

 

Figure 4.7. Decrease in autophosphorylation of MBP-OsTKD. (A)
Phosphorylation assay in the presence of different amounts of fusion proteins
as indicated above the lanes. The reactions were incubated at room
temperature for 20 min. (B) MBP-OsTKD, 3.5 pg, was allowed to
autophosphorylate at 30°C for 1 h. Free y[32P]ATP was removed by Sephadex
G-50 gel filtration and phosphorylated MBP-OsTKD was distributed in aliquots
over five Eppendorf tubes. Lane 1, MBP-OsTKD was precipitated by TCA
directly. MBP-OsTKD present in all other lanes was incubated at 30°C for an
additional 20 min in the presence of: lane 2, 20 pM ATP and 10 mM MnClz;
lane 3, 250 ng GST-OsKID and 10 mM MnClz; lane 4, 250 ng GST-OsKID,

20 pM ATP, and 10 mM MnClz; lane 5, 250 ng GST-OsKlD, 20 pM ATP +

10 pCi v-[32PJATP . and 10 mM MnCl2_ The amount of MBP-OsTKD per
reaction was approximately 150 ng. (C) Inhibition of MBP-OsTKD activity.
Phosphorylation assay with 500 ng MBP-OsTKD and 500 ng GST-OsKID in
the presence of: lane 1, no additions; lane 2, 3 pg GST; lane 3, 2.5 pg BSA;
lane 4, additional 750 ng of GST-OsKID. The reactions were incubated at
30°C for 20 min.

123

EXPRESSION OF OsKAPP IN RICE

KAPP was shown to interact in vitro with several RLKs (Braun et al.,
1997). Whether KAPP interacts in vivo is, in most cases, unclear. An
exception may be the interaction between CLV1 and KAPP (Williams et al.,
1997). To determine whether OsTMK and OsKAPP may interact in vivo, the
expression patterns of these genes were investigated by Northern blot
analysis. High transcript levels of OsTMK were detected in many tissues in
rice, in particular in regions containing dividing and elongating cells (see Fig.
4.8). Lower levels of expression were detected in the basal part of the second
youngest leaf blade and in the oldest part of the internode; those tissues show
reduced or no growth, respectively. The transcript levels for OsKAPP mirror
those of OsTMK almost perfectly. This indicated that OsTMK and OsKAPP are
expressed in the same tissue and, therefore, may interact in vivo. While the
transcript levels of OsTMK clearly increased during GA treatment of stem
sections, the increase in transcript levels of OsKAPP in response to GA was

small, reaching only a 3-fold increase after 24 h (data not shown).

4.5. DISCUSSION

In deepwater rice, a LRR-RLK that was highly expressed in growing
regions of the plant was identified. The gene was named OsTMK for several
reasons. (i) The highest amino acid sequence identity is to TMK1 from
Arabidopsis (Chang et al., 1992). (ii) The expression of OsTMKI is ubiquitous
and is high in regions undergoing cell proliferation as is the expression of

TMK1 in Arabidopsis (Chang et al., 1992; AB. Bleecker, personal

124

lntemode

 

N2 N1 L2b L25 L1 Co Ro 0-3 3-8 8-18 old

 

-OsTMK

 

 

 

 

-OsKAPP

 

 

 

 

 

Figure 4.8. Tissue-specific expression of OsTMK and OsKAPP in rice. N2,
second highest node; N1, highest node containing the shoot apex; L2b, basal

2 cm of second youngest leaf blade; L25, basal 2 cm of second youngest

leaf sheath; L1, youngest leaf; Co, coleoptile 3 days after germination; Ro, root
3 days after germination; 0-3, internodal region 0-3 mm above N2 containing
the IM; 3-8, internodal region 3-8 mm above N2 containing mostly the E2; 818,
internodal region 8-18 mm above N2 containing the upper part of the EZ and
the D2; old, oldest part of the internode. The upper panel shows hybridization
signals with OsTMK as probe, the middle panel shows hybridization signals with
OsKAPP as probe, the lower panel shows hybridization signals with E37 as
internal loading control.

125

communications). Although Southern blot analysis indicated that TMK1 is a
unique gene in Arabidopsis, a related gene showing 53.3% amino acid identity
to TMK1 was identified by the genome sequencing project. Southern blot
analysis with rice genomic DNA and the region corresponding to the OsTMK
kinase domain as probe also indicated the presence of only one gene in the
rice genome. One rice EST (D39936) was identified, however, with high
sequence similarity to a part of the LRR domain of OsTMK. Therefore, it is
possible that homologs of OsTMK exist in rice as well. Arabidopsis AC000103
and TMK1 are likely to be evolutionarily related. The position of a unique
intron (84 nucleotides in both TMK1 and AC000103) is conserved between the
first and the second nucleotide of the codon corresponding to valine at
postion 766 in TMK1, immediately after kinase subdomain VIII (Chang et al.,
1992). For OsTMK the intron position(s) have not been determined. However,
the lane containing Hindlll-digested genomic DNA was expected to show a
band of 1.7 kb, based on the cDNA sequence. Instead, the band detected was 2
kb, suggesting the presence of an intron in this fragment. This fragment spans
the region containing the intron in AC000103 and TMK1.

Based on amino acid sequence similarity, TMK1 is more related to
OsTMK than to AC000103. Several gaps exist in the sequence alignment of
TMK1 and AC000103. In Arabidopsis, TMK1 was in the membrane fraction
and was glycosylated in vivo (Schaller and Bleecker, 1995). Six of the eight
potential glycosylation sites in the presumed extracellular domain were
conserved between OsTMK and TMK1. Between AC000103 and OsTMK or
TMK1, only two of the eight potential glycosylation sites were conserved.
Furthermore, phylogenetic analysis by the Clustal method with the PAM250
residue weight table, showed that TMK1 and OsTMK were more related to

each other than to AC000103 (data not shown). The high sequence

126

conservation between two RLKs from a monocot and a dicot species and the
similar expression pattern suggests an important role for this gene in the

plant and suggests functional relatedness.

The transcript levels of OsTMK were induced by GA in deepwater rice
internodes. Furthermore, the expression of this gene is particularly high in all
regions undergoing cell division and elongation, whereas low transcript
levels were detected in the non-growing region of the internode. This
suggests a role for this gene in plant growth and possibly in GA signal
transduction. Evidence for a role of OsTMK in growth is substantiated by the
dwarfed phenotype observed in Arabidopsis plants overexpressing TMK1 in
the antisense orientation (E. Maher, personal communication). GA is a
hydrophobic molecule and may pass through the plasma membrane by
simple diffusion. It may then interact with cytosolic receptors. However,
evidence in the cereal aleurone pointed to a receptor at the plasma membrane
(Hooley et al., 1991) and not in the cytosol (Gilroy and Jones, 1994). Recently, a
LRR-RLK, BRIl (Li and Chory, 1997), has been identified to play a role in
brassinosteroid (BR) signaling. Structurally, GA and BK are related, suggesting
that an LRR-RLK may be involved in GA signaling as it is in BR signaling.

The transcript levels of some genes encoding RLKs in plants are
mediated by hormones and external stimuli. The transcript levels of SFR, a S-
locus-like RLK found in Brassica oleracea, increases after wounding and
bacterial infection (Pastuglia et al., 1997). Also, the transcript levels of RPK1,
encoding a LRR-RLK from Arabidopsis, increases after abscisic acid treatment
and in response to a variety of environmental stresses (Hong et al., 1997).
Inducible gene expression has been observed for some receptor kinases in

animals as well. Interestingly, epidermal growth factor (EGF) and platelet-

127

derived growth factor were able to increase the transcript levels of their
respective receptors (Clark et al., 1985; Ericksson et al., 1991). Ligand-induced
changes in receptor transcript levels have been observed in plants as well.
The transcript levels for the ethylene receptor NR in tomato (Wilkinson et
al., 1995) and ERS in pea (Peck and Kende, manuscript in preparation),
increase in response to ethylene. The transcript levels corresponding to
phytochrome A, one of the red/ far-red light receptors, decreases in response
to red light (Somers et al., 1991). Therefore, it is not unprecedented that
ligands change the transcript levels of their respective receptors. It will be
interesting to know whether GA by itself or via an accessory protein can bind
to OsTMK and, therefore, is responsible as its ligand for changes in transcript
levels. Alternatively, increased transcript levels of OsTMK may be a result of

growth in general and may not be involved in GA signaling.

The kinase domain of OsTMK is an active kinase autophosphorylating
primarily on threonine and to a lesser extent on serine residues, as shown for
TMK1 (Chang et al., 1992). The autophosphorylation mechanism of MBP-
OsTKD is complex, and suggests that autophosphorylation is accomplished
via a higher order reaction mechanism. Complex autophosphorylation was
found as well for RLK5. Phosphorylation of the inactive RLK5 kinase domain
by active RLK5 kinase showed that RLK5 uses primarily an intermolecular
mechanism (Horn and Walker, 1994). In animals, it is known that inactive
receptor monomers are in equilibrium with active receptor dimers, and
ligand binding stabilizes the active dimeric form (Ullrich and Schlessinger,
1990). Dimerization (or higher order oligomerization) is responsible for
activation of the intrinsic protein kinase activity and for

autophosphorylation; both processes are mediated by an intermolecular

128

mechanism (Lemmon and Schlessinger, 1994). For animal tyrosine kinase
receptors, phosphorylation of a conserved tyrosine residue within the kinase
domain leads to an increase in kinase activity. Phosphorylation of residues
outside the kinase domain will create docking sites for downstream signal
transduction molecules (Heldin, 1995). Therefore, the complex reaction order
observed may reflect changes in activity of the kinase due to a different level
of phosphorylation.

One candidate for a signal transduction component downstream of
several RLKs is, KAPP. Far-western analysis has indicated that KAPP interacts
only with active, phosphorylated kinase (Stone et al., 1994) and that the KID
of Arabidopsis KAPP interacts in vitro with TMK1 (Braun et al., 1997). Here,
the interaction between the rice homologs was indicated by phosphorylation
of GST-OsKID by MBP-OsTKD. No rapid turnover of the phosphates on MBP-
OsTKD occurred even in the presence of ATP, and phosphates were not
transferred from MBP-OsTKD to GST-OsKID. This was expected because of the
low-energy phosphoester bond in phosphoserine and phospothreonine. This
is contrary to the high-energy phosphoester bond of phosphohistidine and
phosphoaspartate found in bacterial two-component systems, which all signal
by phosphorelay (reviewed in Wurgler-Murphy and Saito, 1997).

Increasing amounts of inactive kinases were shown in vitro to inhibit
autophosphorylation of active kinases (Horn and Walker, 1994; Williams et
al., 1997). In vivo coexpression of an active and an inactive EGF receptor
kinase led to changes in the mitogenic response towards EGF (Honegger et al.,
1990), possibly because of the formation of partially inactive heterodimers.
Here it is shown that autophosphorylation of MBP-OsTKD can be inhibited
also by increasing amounts of another substrate, GST-OsKID. However, the in

vitro inhibition of autophosphorylation needs to be interpreted with caution.

129

Kinase activity was not only inhibited by its substrate, but also by BSA and
GST. This indicated that in vitro inhibition of MBP-OsTKD activity is not

specific to its substrate.

KAPP may function in a number of signaling pathways. This is
indicated by its interaction with several RLKs in vivo and in vitro, by the
ubiquitous expression of KAPP and by the presence of a single gene in both
Arabidopsis and maize (this work; Braun et al., 1997; Williams et al., 1997).
The expression pattern of OsKAPP was very similar to that of OsTMK, which
indicated that in vivo interaction is possible and that OsKAPP may function
in the OsTMK signaling pathway. The in vivo role of this interaction is
unknown. As was the case for another RLK, SRK-A14, it is expected that more
proteins are interacting directly with the kinase domain of OsTMK (Braun et

al., 1997; Gu et al., 1998) to signal changes in the environment.

4.6 LITERATURE

Altschul, S.F., Warren, G., Gish, W., Myers, E.W. and Lipman, DJ. (1990) Basic
local alignment search tool. J Mol Biol 215:403-410.

Ausubel, F.M., Brent, R., Kingston, RE, Moore, D.D., Seidman, J.G., Smith,
J.A. and Struhl, K. (1987) Current Protocols in Molecular Biology (Wiley,
New York).

Bethke, P.C., Schuurink, R. and Jones, KL. (1997) Hormonal signalling in
cereal aleurone. J Exp Bot 48:1337-1356.

Bleecker, A.B., Schuette, IL. and Kende, H. (1986) Anatomical analysis of
growth and developmental patterns in the internode of deepwater rice.
Planta 169:490-497.

Boyle, W.L., van der Geer, P. and Hunter, T. (1991) Phosphopeptide mapping

and phosphoamino acid analysis by two-dimensional separation on thin-
layer cellulose plates. Methods Enzymol 201:110-149.

130

Braun, D. M. and Walker, J.C. (1996) Plant transmembrane receptors: new
pieces in the signaling puzzle. Trends Biochem Sci 21:70-73.

Braun, D.M., Stone, J.M. and Walker, J.C. (1997) Interactions of the maize and
Arabidopsis kinase interaction domains with a subset of receptor-like
protein kinases: implications for transmembrane signaling in plants. Plant
J 12:83-95.

Catling, H.D. (1992) Rice in Deepwater. The Macmillan Press Ltd, London and
Basingstoke, UK.

Chang, C., Schaller, GE, Patterson, S.E., Kwok, S.F., Meyerowitz, EM. and
Bleecker, AB. (1992) The TMK1 gene from Arabidopsis codes for a protein

with structural and biochemical characteristics of a receptor protein kinase.

Plant Cell 4:1263-1271.

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

Clark, A.].L., Ishii, S., Richert, N., Merlino, GT and Pastan, I. (1985)
Epidermal growth factor regulates the expression of its own receptor. Proc
Natl Acad Sci USA 82:8374-8378.

Clark, S.E., Williams, R.W. and Meyerowitz, EM. (1997) The CLA VAT/11
gene encodes a putative receptor kinase that controls shoot and floral
meristem size in Arabidopsis. Cell 89:575-585.

Elich, TD. and Chory, J. (1997) Phytochrome: if it looks and smells like a
histidine kinase, is it a histidine kinase? Cell 91:713-716.

Erickkson, A., Nistér, M., Leveen, P., Westermark, 3., Heldin, C.-H. and
Claesson-Welsh, L. (1991) Induction of platelet-derived growth factor a-

and B-receptor mRNA and protein by platelet-derived growth factor 33. J
Biol Chem 266:21138-2144.

Gilroy, S. and Jones, KL. (1994) Perception of gibberellin and abscisic acid at
the external face of the plasma membrane of barley (Hordeum vulgare L.)
aleurone protoplasts. Plant Physiol 104:1185-1192.

Gu, T., Mazzurco, M., Sulaman, W., Matias, DD. and Goring, DR. (1998)

Binding of an arm repeat protein to the kinase domain of the S-locus
receptor kinase. Proc Natl Acad Sci USA 95:382-387.

131

Hanks, SK. and Quinn, A.M. (1991) Protein kinase catalytic domain sequence
database: identification of conserved features of primary structure and
classification of family members. Methods Enzymol 200:38-62.

Heldin, C.-H. (1995) Dimerization of cell surface receptors in signal
transduction. Cell 80:213-223.

Hoffmann-Benning, S. and Kende, H. (1992) On the role of abscisic acid and
gibberellin in the regulation of growth in rice. Plant Physiol 99:1156-1161.

Honegger, A.M., Schmidt, A., Ullrich, A. and Schlessinger, J. (1990) Evidence
for epidermal growth factor (EGF)-induced intermolecular
autophosphorylation of the EGF receptors in living cells. Mol Cell Biol
10:4035-4044.

Hong, S.W., Jon, J.H., Kwak, J.M. and Nam, H.G. (1997) Identification of a
receptor-like protein kinase gene rapidly induced by abscisic acid,
dehydration, high salt, and cold treatments in Arabidopsis thaliana. Plant
Physiol 113:1203-1212.

Hooley, R., Beale, M.H. and Smith, S]. (1991) Gibberellin perception at the
plasma membrane of Avena fatua aleurone protoplasts. Planta 183:274-
280.

Horn, MA. and Walker, J.C. (1994) Biochemical properties of the
autophosphorylation of RLK5, a receptor-like protein kinase from
Arabidopsis thaliana. Biochim Biophys Acta 1208:65-74.

Jacobsen, S.E., Binkowski, K.A. and Olszewski, N .E. (1996) SPINDLY, a
tetratricopeptide repeat protein involved in gibberellin signal transduction
in Arabidopsis. Proc Natl Acad Sci USA 93:9292-9296.

Jones, H.D., Smith, S.J., Desikan, R., Plakidou-Dymock, S., Lovegrove, A. and
Hooley, R. (1998) Heterotrimeric G-proteins are implicated in gibberellin

induction of (Jr-Amylase gene expression in wild oat aleurone. Plant Cell
10:245-253.

Kakimoto, T. (1996) CKIl, a histidine kinase homolog implicated in cytokinin
signal transduction. Science 274:982-985.

Kim, J.-K. and Wu, R. (1993) A rice (Oryza sativa L.) cDNA encodes a protein
sequence homologous to the eukaryotic ribosomal SS RNA-binding
protein. Plant Mol Biol 23:409-413.

Kobe, B. and Deisenhofer, J. (1994) The leucine-rich repeat: a versatile binding
motif. Trends Biochem Sci 19:415-421.

132

Lemmon, MA. and Schlessinger, J. (1994) Regulation of signal transduction
and signal diversity by receptor oligomerization. Trends Biochem Sci
19:459-463.

Li, J. and Chory, J. (1997) A putative leucine-rich repeat receptor kinase
involved in brassinosteroid signal transduction. Cell 90:929-938.

Mu, J.-H., Lee, H.-S. and Kao, T.-h. (1994) Characterization of a pollen-
expressed receptor-like kinase gene of Petunia inflata and the activity of its
encoded kinase. Plant Cell 6:709—721.

Olszewski, N. (1997) Evidence that dynamic O-linked N-acetylglucosamine
modification of proteins plays a role in gibberellin signal transduction. 8th
International conference on Arabidopsis research. Madison, Wisconsin.

Pastuglia, M. Roby, D., Dumas, C. and Cock, J.M. (1997) Rapid induction by
wounding and bacterial infection of an S gene family receptor-like kinase
gene in Brassica oleracea. Plant Cell 9:49-60.

Peng, J., Carol, P., Richards, D.E., King, K.E., Cowling, R.J., Murphy, GP. and
Harberd N .P. (1997) The Arabidopsis GAI gene defines a signaling pathway
that negatively regulates gibberellin responses. Genes Dev 11:3194-3205.

Puissant, C. and Houdebine, L.M. (1990) An improvement of the single-step
method of RNA isolation by acid guanidium thiocyanate-phenol-
chloroform extraction. BioTechniques 8:148-149.

Raskin, I. and Kende, H. (1984a) Role of gibberellin in the growth response of
submerged deepwater rice. Plant Physiol 762947-950.

Raskin, I. and Kende, H. (1984b) Regulation of growth in stem sections of
deepwater rice. Planta 160:66-72.

Ritchie, S. and Gilroy, S. (1998) Calcium-dependent protein phosphorylation
may mediate the gibberellic acid response in barley aleurone. Plant Physiol
116:765-776.

Sauter, M. and Kende, H. (1992) Gibberellin-induced growth and regulation of
the cell division cycle in deepwater rice. Planta 188:362-368.

Schaller, GE. and Bleecker, AB. (1993) Receptor-like kinase activity in
membranes of Arabidopsis thaliana. FEBS Letters 333:306-310.

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

133

Schmidt, E.D.L., Guzzo, F., Toonen, M.A.J. and de Vries, SC. (1997) A leucine-
rich repeat containing receptor-like kinase marks somatic plant cells
competent to form embryos. Development 124:2049-2062.

Silverstone, A.L., Ciampaglio, C.N. and Sun, T.-p. (1998) The Arabidopsis
RGA gene encodes a transcriptional regulator repressing the gibberellin
signal transduction pathway. Plant Cell 10:155-169.

Somers, D.E., Sharrock, R.A., Tepperman, J.M. and Quail, RH. (1991) The hy3
long hypocotyl mutant of Arabidopsis is deficient in phytochrome 3. Plant
Cell 3:1263-1274.

Song, W.-Y., Wang, G.-L., Chen, L.-l., Kim, H.-S., Pi, L.-Y., Holsten, T.,
Gardner, J., Wang, 3., Zhai, W.-X., Zhu, L.-H., Fauquet, C. and Ronald, P.
(1995) A receptor kinase-like protein encoded by the rice disease resistance
gene, Xa21. Science 270:1804-1806.

Song, W.-Y., Pi, L.-Y., Ruan, D.-L., Braun, D.M., Walker, J.C. and Ronald, PC.
(1998) The rice disease resitance gene, Xa21, encodes an active
serine/threonine protein kinase and does not interact with a rice kinase-
associated protein phosphatase. Submitted

Stone, J.M., Collinge, M.A., Smith, R.D., Horn, MA. and Walker, J.C. (1994)
Interaction of a protein phosphatase with an Arabidopsis serine-threonine
receptor kinase. Science 266:793-795.

Stiinzi, J.T. and Kende, H. (1989) Gas composition in the internal air spaces of
deepwater rice in relation to growth induced by submergence. Plant Cell
Physiol 30:49-56.

Torii, K.U., Mitsukawa, N., Oosumi, T., Matsuura, Y., Yokoyama, R.,
Whittier, RF. and Komeda, Y. (1996) The Arabidopsis ERECTA gene

encodes a putative receptor protein kinase with extracellular leucine-rich
repeats. Plant Cell 81735-746.

Ullrich, A. and Schlessinger, J. (1990) Signal transduction by receptors with
tyrosine kinase activity. Cell 61:203-212.

Van der Knaap, E. and Kende, H. (1995) Identification of a gibberellin-induced
gene in deepwater rice using differential display of mRNA. Plant Mol Biol
28:589-592.

Walker, J.C. (1993) Receptor-like protein kinase genes of Arabidopsis thaliana.
Plant] 32451-456.

134

Wilkinson, J.Q., Lanahan, M.B., Yen, H.-C., Giovannoni, J.J. and Klee, H].
(1995) An ethylene-inducible component of signal transduction encoded
by Never-ripe. Science 270:1807-1809.

Williams, R.W., Wilson, J.M. and Meyerowitz EM. (1997) A possible role for
kinase-associated protein phosphatase in the Arabidopsis CLAVATAl
signaling pathway. Proc Natl Acad Sci USA 94:10467-10472.

Wurgler-Murphy, SM. and Saito, H. (1997) Two-component signal
transducers and MAPK cascades. Trends Biochem Sci 22:172-176.

135

CHAPTER 5
A gibberellin-induced gene from deepwater rice confers delayed

bolting, fasciated stems and gross alterations in flower morphology

when overexpressed in Arabidopsis

5.1. ABSTRACT

In search for differentially expressed genes, a novel gene, OsDD4, was
identified whose transcript levels increased in response to gibberellin (GA) in
the intercalary meristem (IM) of deepwater rice internodes. Its tissue-specific
expression indicated that OsDD4 is exclusively expressed in regions
containing meristems. The encoded protein is novel and is localized to the
nucleus. OsDD4 was overexpressed in Arabidopsis thaliana and individual T1
lines were analyzed. Overexpression of OsDD4 led to delayed bolting and
fasciated stems. The severe phenotype showed reduction in inflorescence
stem elongation and apical dominance. The severe lines also displayed
altered leaf and flower morphology. In particular the development of the
gynoecia was impaired. At later stages of plant development, changes in floral

organ number, and organ and floral meristem identity were observed.

136

5.2. INTRODUCTION

Deepwater rice belongs to a group of rice cultivars whose survival
during flooding is based on its capacity for rapid increase of internodal
elongation when it becomes submerged. Under field conditions growth rates
of up to 25 cm / day have been reported resulting in plants that are up to 6 m
tall (Catling, 1992). The signal for accelerated growth is an increase in ethylene
levels (Métraux and Kende, 1983) which, via a decrease in abscisic levels
(Hoffman-Benning and Kende, 1992), enhances the responsiveness of the
internode to GA (Raskin and Kende, 1984a). While ethylene signals, in part,
the change in the environment, the growth response is, ultimately, elicited by
GA. In rice stems, the primary site of action of GA is the IM located at the base
of the growing internode (Sauter et al., 1993). GA has been shown to increase
cell size and the rate of cell production in the IM. GA also results in a 3- to 4-
fold increase in final cell size of cells that emerge from the IM into the
elongation zone (EZ) (Bleecker et al., 1986; Sauter and Kende, 1992). Recently,
several signal transduction components that mediate responses to GA have
been identified. Two putative transcription factors with high sequence
similarity to each other, GAI (Peng et al., 1997) and RGA (Silverstone et al.,
1998), have been shown to mediate responses to GA. The gai mutant is a GA-
insensitive dwarf, and rga was identified for its ability to partially suppress
some of the defects observed in the severe GA biosynthesis mutant, gal-3.
Epistasis analyses have indicated that downstream of GAI is SPY (Wilson and
Somerville, 1995; Jacobsen et al., 1996). The spy mutant partially suppresses
both gai (Wilson and Somerville, 1995) and gal-3 (Silverstone et al., 1997).

SPY encodes a protein with O-linked N-acetylglucosamine transferase activity

137

which may modify post-translationally target proteins in the GA signaling
pathway (J acobsen et al., 1996; Olszewski, 1997).

GAI, RGA, and SPY are negative regulators of the GA signal
transduction pathway. So far, no positive regulator in GA signaling has been
identified. A search for early GA-induced genes was initiated to identify
potential positive regulators. Deepwater rice provides an excellent system to
study GA-mediated growth. Changes in the internodal growth rate are
observed 40 min after application of GA (Sauter and Kende, 1992), which
results in a maximum growth rate of 5 mm / h (Stiinzi and Kende, 1989).
Using differential display (Liang and Pardee, 1992), we have identified a gene,
OsDD4, whose transcript level increases rapidly in the IM of deepwater rice
internodes in response to GA and submergence. The protein encoded by
OsDD4 is novel and is partially localized in the nucleus. To investigate its role
in GA-mediated growth, OsDD4 was overexpressed in Arabidopsis thaliana.
This has led to what appeared to be a dominant negative phenotype in which

several processes mediated by GA are affected.

5.3. MATERIALS AND METHODS

PLANT MATERIAL

Seeds of deepwater rice (Oryza sativa L., cv. Pin Gaew 56) were obtained from
the International Rice Research Institute (Los Bafios, Philippines). Plants were
grown as described in Stiinzi and Kende (1989). For submergence
experiments, 12-week-old plants were partially submerged under continuous

light (Métraux and Kende, 1983). Twenty-cm-long stem sections containing

138

the growing internode were excised and treated with 50 pM GA3 (Raskin and
Kende, 1984b). Incubation was allowed to proceed for the periods indicated,
after which the IM was excised, frozen immediately, and stored at -80°C until

use.

IDENTIFICATION OF OsDD4 AND D40170

A differentially displayed 22 bp cDNA band, dd4, was identified using the
primers T12MG and OPA04 (Van der Knaap and Kende, 1995). The differential
displayed product was inserted in pUC19, and the cloned DNA fragment was
used to screen a IM-specific cDNA library to obtain a full-length cDNA insert,
OsDD4. The phage insert was cloned into the Nail site of pBluescript SK(-)
phagemid (Stratagene), and sequence analysis was performed at the W.M.
Keck Facility at Yale University, New Haven, CT. The sequences were aligned
using Sequencher, version 3.0 (Gene Codes Corporation).

One rice EST, D40170, with deduced amino acid sequence similarity to OsDD4
was obtained from Japan, and sequenced entirely in one orientation. The EST
contained a partial cDNA since approximately 500 bp at the 5'end were

missing.

NUCLEAR LOCALIZATION OF OsDD4

To facilitate the insertion of OsDD4 in frame to the reporter protein B-
glucuronidase (GUS), restriction enzyme sites were introduced by PCR with
Pwo polymerase (BMB). The primer sequences were: 5’-
TCGGTCTAGAGGCGGTCGGTCGACGCTGAA-B’ and 5’-
TCATTGTGGATCCGGGAGGTGGTGGTGATC-3’. The PCR product that

139

contained the full ORF of OsDD4, except for the last 5 amino acids, was
digested with XbaI and BamHI and inserted in the same sites of pBS::G 1.15.3
(Varagona et al., 1992). Sequence analysis over the primer regions in
pBS::DD4::GUS was performed to verify that the proper fusion construct had
been obtained. pBS::DD4::GLIS was digested with SalI and ClaI and inserted
into the XhoI and ClaI sites of pMP6, a monocot-specific transient assay
transformation vector. The onion epidermal cell transformation was
performed as described (Varagona et al., 1992) by Emily Avila in Dr. M.

Varagona's laboratory (New Mexico State University, Las Crucas).

NORTHERN BLOT ANALYSIS

Twenty pg of total RNA, isolated according to Puissant and Houdebine (1990),
was electrophoretically separated in a 1.2% formaldehyde-agarose gel
(Ausubel et al., 1987) and transferred to Hybond-N+ membrane (Amersham).
DNA fragments containing the insert derived from the 3' 737 bp of OsDD4, or
the full-length D40170, the E37, and the histone H3 insert (Van der Knaap and
Kende, 1995) were isolated from agarose gels by digestion with B-agarase
(NEB). Fifty ng of template DNA was labeled in the presence of a-[32P]dCTP
(3000 Ci/mmol, NEN), using a random prime labeling kit (BMB). Northern
blots were prehybridized for 4 h at 42°C in 5x SSC, 10x Denhardt's solution,
0.2% SDS, 0.1 M K-PO4, pH 6.8, and 100 pg / ml denatured salmon sperm DNA
and hybridized overnight at 42°C in 5x SSC, 10x Denhardt's solution, 0.1%
SDS, 0.1 M K-PO4, pH 6.8, 100 pg/ ml denatured salmon sperm DNA, 10%
dextran sulfate, and 30% formamide (BMB). The blots were washed twice in
2x SSC and 0.5% SDS and twice in 0.1x SSC and 0.3% SDS at 65°C for 30 min

each. For cchsI, the RNA probe was prepared in the presence of or-[37-P]UTP

140

(800 Ci / mmol, NEN) and contained the 3'UTR and most of the coding region
(Sauter et al., 1995). The RNA probe used for the bottom panel of Figure 5.11
was derived from the 5’ UTR region (107 bp) of OsDD4. The blots were
prehybridized and hybridized in 3x SSPE, 10x Denhardt's solution, 0.5% SDS,
50 pg/ ml denatured salmon sperm DNA, and 50% deionized formamide
(BMB) at 65°C overnight. Blots were washed twice for 5 min each in 2x SSC
and 0.5% SDS followed by two washes in 0.1x SSC and 0.5% SDS for 30 min
each at 65°C. The radioactivity on blots was quantified by Phosphorlmager

analysis (Molecular Dynamics).

SOUTHERN BLOT ANALYSIS

Rice genomic DNA was isolated from a CsCl gradient according to Ausubel et
al. (1987). Arabidopsis genomic DNA was isolated by grinding two lyophilized
leaves in liquid nitrogen. The tissue was thawed in hot
hexadecyltrimethylammonium bromide (CTAB) buffer (2% CTAB, 1.4 M
NaCl, 20 mM EDTA, 100 mM Tris-HCl, pH 8.0, and 0.2% B-mercaptoethanol at
65°C) and incubated at 65°C for 30 min. After addition of an equal volume of
chloroform, the organic phase was separated from the aqueous phase by
centrifugation. The DNA in the aqueous phase was precipitated with 0.6
volumes of isopropanol. The pellet was dissolved in TE and the RNA was
removed by RNase A treatment (10 pg/ ml) at 37°C for 30 min. After organic
extraction, the DNA was precipitated with 1/ 3 volume of 7.5 M NH4-acetate
and 2.5 volumes of ethanol and subsequently dissolved in the appropriate
amount of TE. Five pg of rice DNA and 500 ng of Arabidopsis DNA were
digested overnight, and fragments were separated on a 0.8% agarose gel at 30

V for 20 h. The gel was then treated for 15 min in 0.25 N HCl; 30 min in 0.5 N

141

NaOH and 1.5 M NaCl; and 15 min in 1 M Tris-HCl, pH 8 and 1.5 M NaCl;
after which the DNA was transferred to Hybond N+. The blots were
prehybridized and hybridized in the same buffer and conditions as described
above for Northern blot analysis. For the rice Southern blot, the probe used
was derived from the 3' 737 bp of OsDD4. For the Arabidopsis Southern blot,
the probe used was derived from the full-length OsDD4.

OVEREXPRESSION OF OsDD4 IN A. THALIANA

OsDD4 was inserted in the sense orientation into the XbaI-Clal site of the
Agrobacterium binary vector pGA643 (An et al., 1988). This vector contains a
neomycin phophotransferase II gene conferring kanamycin resistance. For
cloning purposes, the construct thus created, pGA::DD4, lacked the 3' 428 bp of
OsDD4, leaving only 74 bp of the 3' UTR. pGA::DD4 was transformed into
Agrobacterium strain GV3101 (C58C1 RifR) pMP90 (GmR). pGA::DD4 and
pGA643 were each vacuum infiltrated into 40 A. thaliana ecotype Columbia
and 40 ecotype Landsberg erecta plants as described (Van Hoof and Green,
1996). Seeds obtained were surface sterilized and plated on MS medium
containing 1% sucrose, 10 mM MES, 0.8% phytagar, 500 pg/ ml vancomycin
(Clinical Center, Michigan State University), and 50 pg/ m1 kanamycin
(Sigma). After two weeks, transgenic plants were transferred to soil and

grown in the growth chamber at 20°C, 16h light, 8 h dark, 100 pmol rn’2 5'1.

142

5.4. RESULTS

IDENTIFICATION OF OsDD4 AND AMINO ACID SEQUENCE ANALYSIS

We used differential display of mRNA to identify genes whose
transcript had changed in the IM in response to GA (Van der Knaap and
Kende, 1995). A 222-bp differentially displayed product, dd4, appeared 2.5 h
after start of GA treatment and was investigated further. After cloning, one
insert was used as probe and hybridized to a Northern blot containing
polyadenylated RNA isolated from GA-treated stem sections (data not shown,
see also Fig. 5.4). The results from this experiment indicated that the dd4
insert identified was derived from a differentially expressed gene. A rice IM-
specific cDN A library was screened with dd4, and a full-length clone of 1882
bp was isolated. The sequence showed an ORF from position 193 to 1380,
yielding a 43-kD protein (Fig. 5.1). The protein encoded by OsDD4 contains
homopolymeric sequences of proline, threonine, alanine and glutamine. It
has a histidine-rich region as well as a region rich in acidic residues.
Furthermore, a putative bipartite nuclear localization signal (NLS) was
identified (Hicks and Raikhel, 1995). These features are reminiscent of
transcription factors (Mitchell and Tjian, 1989). Database searches indicated
indeed weak similarities to a range of transcription factors, transcriptional
activators and homeo box proteins. With the BLASTP program (Altschul et
al., 1997), the highest similarity was found to a domain of unknown function
in the yeast SW12/ SNF2 and related genes in human, chicken and
Drosophila. The alignment of this domain is shown in Figure 5.2A. Because
of the conserved spacing of the glutamine and leucine residues as

QXXXLXXQ, we refer to this region as the QLQ domain. Searches through EST

143

MSGRPSGGAGGGR PFTASQWQELEHQAL IYKYMASGTP I P SDLI LP

, .. OO _ _ , ,0. 00,
-Dsmrmmmrsmrwmmmmmm

oo o_.. - _.o, o o, .oo

SNSSAGVAPTTTTTSSPAPSY8RPAPHDAAPYQALYGGPIAAAIARTPAA

AAIHAQVSPFHLHIDTTHPHPPPSYYSMDHKEYRYGHAIREVHGBHAPF8

LSKEDDDBKERRQQQQQQQQHCFLLGADLRLEKRAGHDHAAAAQKPLRR!

FDIWPHEKNSRGSWMGLEGETQL8M8IPHAANDLPITTTSRIHNDB

50

100

150

200

250

300

350

396

Figure 5.1. Amino acid sequence of OsDD4. The QLQ domain is indicated

the dark grey box, the WRC domain in the light grey box (see Fig. 5.2).
The histidine-rich region is underlined by dashes; the acidic region is

underlined with a solid line. Residues potentially involved in nuclear

localization are found in the WRC domain and are indicated by the circles

above the amino acid.

144

and genomic databases indicated that the QLQ domain was also found in
other plant genes, namely in the rice EST D40170 and in two putative proteins
identified by the Arabidopsis genome sequencing project. The adjacent
domain in OsDD4 contains a putative NLS, and database searches with this
sequence identified high sequence similarity to several plant genes. The
alignment of this domain, which we refer to as the WRC domain for its
conserved core, is shown in Figure 5.23. Sofar, this motif has not been found
outside the plant kingdom. The position of the QLQ and WRC motifs in the
different proteins is variable as shown in Figure 5.2C. In most cases, except for
the SW12 / SNF2 orthologs and for AC000106, the proteins containing the QLQ
and / or the WRC domain are small, ranging from 269 to 430 amino acids. The

Arabidopsis AC002387 contains two WRC domains.

NUCLEAR LOCALIZATION OF OsDD4

The features of the protein encoded by OsDD4 including a putative
NLS, are found in transcriptional regulators. To address the cellular
localization, a construct containing OsDD4 fused in frame to the reporter
protein GUS was made. After biolistic bombardment of onion epidermal cells
with this construct, the localization of the fusion proteins was determined by
histochemical staining, with the GUS substrate 5-bromo-4-chloro-3-indoly1-B-
D-glucuronide (X-gluc). Blue staining was found associated with the nucleus,
a result confirmed by coincident staining of the nucleus with the DNA-
specific dye 4,6-diamidino-2-phenylindole (DAPI) (data not shown). Some
level of staining was also observed in the cytoplasm, however, GUS itself is
exclusively localized in the cytoplasm. Of the 44 onion cells counted, 30%

showed exclusive nuclear localization, 64% showed nuclear and cytoplasmic

145

Figure 5.2. Alignment of QLQ and WRC motifs present in 03004.
(A) Amino acid sequence alignment of proteins containing

the QLQ domain. The numbers indicate the position of the domain

in the protein (B) Amino acid sequence alignment of proteins
containing the WRC domain. (C) Position of QLQ and WRC domains
in the different proteins. Numbers indicate the length of the protein in
amino acid residues. 03., Oryza sativa; A.t., Arabidopsis thaliana;
H.s., Homo sapiens; G.g., Gallus gallus; D.m., Drosophila melanogaster,
S.c., Saccharomyces cerevisiae. The locus for AC002387 is 2583107;
for AC002343 is 2262102; for AC000106 is 2342679; for U90439 is
1871180.

146

A

OsDD4
D40170
AC002387
AC002343
P51532
X91637
845251
X91638
P25439
P22082

OsDD4
D40170
AC002387A
AC002343
AC002387B
U90439
AC000106

C

-I—l:l——[:l——
__.._E]— 430 A. t. AC002343
.__________£::]___.
—I:
l

529959555399

0
O
A
A.
A
A
A.

QELQUIKQinr‘tt‘tinfn

n n n n n m m

81
69
74

295
151
39

12

11

16
111
171
171
173
171
173
246

EMGFGRKAED..
CLDFGKNPE...
PVTHIDTLE...
ESSASNNTAD..
QGVETDN .....

EKKKAKKSGGGLEP

 

      
     

 
   

PEP RCRRTDGKKWRCS WIZYCEKHMHRG I
PEP RCRRTDGKKWRC - WKYCERHIVIHRGRICR E
EPERCRRTDGKKWRCSI T IIF‘EIHIYCERIIP‘IHRGRKRSP ES
EP RCERTDGKKWRCSRI PDOKYCERIIEHKEEZER-
.EP RCRRTDGKKWRCS ID SGOKYCDKHMHRGI’II'

I I RCKHSDGI’O Rc

  

   

‘l—CJ— 396 Cs. OsDD4

166 0.5. D40170

424 A. t. A0002387

269 At. U90439

// 950 At. ACOOO106

 

4'

l/ 1647 H.s. P51532

 

6’

Figure 5.2. Alignment of QLQ and WRC motifs present in OsDD4.

147

localization and 7% showed cytoplasmic staining (data not shown). This
indicated that OsDD4 contains a functional NLS that can redirect GUS to the

nucleus.

EXPRESSION OF OsDD4 IN RICE AND SOUTHERN BLOT ANALYSIS

GA treatment of rice stem sections containing the growing internode
led to an increase in transcript abundance of OsDD4 as shown in Figure 5.3A.
The signals were quantified and normalized for equal loading using a cDNA,
E37, corresponding to a gene whose transcript levels did not change over the
course of the experiment. Figure 5.33 shows that the level of 05004 in the [M
of GA-treated stem sections had increased more than 3-fold 3 h after start of
GA treatment and 8-fold 4 h after start of treatment. Since the rice EST D40170
contained the QLQ and the WCR domains, we were interested in
investigating its expression in GA-treated stem sections. The probe was
hybridized to the same Northern blot, but the transcript levels of D40170 were
not found to change in response to GA (Fig. 5.3A).

The expression pattern of OsDD4 was also investigated in submerged
plants as shown in Figure 5.4. The transcript levels had increased after 4 h,
reaching a maximum 8 h after start of treatment. The increase in transcript
levels for OsDD4 was compared to the increase in transcript level for histone
H3 and cchsI. Histone H3 is a marker for the S-phase of the cell cycle (Van
der Knaap and Kende, 1995) and cchsI for the transition from the G2 to M
phase (Sauter et al., 1995; Sauter, 1997). The signals for histone H3 and cchsI
were quantified by Phosphorlmager and normalized for equal loading with
E37. A greater than 3-fold increase in histone H3 transcript level was detected

8 h after start of treatment, while that of cchsI was detected after 16 h.

148

 

012345678152411

M”

 

 

 

 

 

 

 

 

 

 

 

 

B

c 10'1-

.2

H

E 81-

E‘ A. ..

2

a 6!.

In

2

a 4'

2

a: 2:

.5!

o

m o. . : . - -
o 5 1o 15 20 25

GA treatment (h)

Figure 5.3. Expression of 03004 and D40170 in GA-treated rice stem
sections. Stem sections were incubated in 50 pM GA, for the times indicated
above the lanes, after which RNA was isolated from the IM. (A) Northern
blot containing 20 pg of RNA was hybridized to OsDD4, D40170, and E37.
(B) Quantitative analysis of the induction of 03004 transcript levels using a
Phosphorlmager. The values were normalized to the E37 loading control.

149

02 4 6 8101624h

 

 

 

 

 

histone H3

 

 

cchs1

 

 

 

ll 33...“. 93.85.51: 58:8 5 .‘m E37

 

 

Figure 5.4. Expression of 03004 in submerged rice plants. Plants
were submerged for the times indicated above the lanes, after which
RNA was isolated from the IM. Northern blots containing RNA
isolated from the same submergence experiment were hybridized to
03004, histone H3, and cchs1. Equal loading was verified by
hybridization of the same blots to E37.

150

Because the signal for OsDD4 could not be detected at the 0 h timepoint by
Phosphorlmager analysis, the signals were not quantified. It is clear however,
that the increase in expression of OsDD4 preceded that of histone H3 and of
cchsI. Interestingly, a transient decrease in OsDD4 transcript levels was
observed in both stem sections and in submerged plants. The significance of
this is unknown but it has been observed reproducibly in different
experiments.

The tissue-specific expression of OsDD4 is shown in Figure 5.5. The
highest transcript levels were found in the N1, which contained the shoot
apex, and in the IM. Slightly lower expression was detected in the coleoptile
and in the youngest leaf. Weak expression was detected in the N2, in the basal
part of the second youngest leaf sheath, and in the root. Along the internode,
OsDD4 was expressed in the IM, while no transcript was detected in the 32, in
the differentiation zone (DZ), and in the oldest part of the internode. This
indicated that OsDD4 is expressed exclusively in the IM, which is the primary
site of GA action in the internode (Sauter et al., 1993).

Southern blot analysis was employed to address the question whether
genes related to OsDD4 exist in rice. The probe was derived from the 3' region
of OsDD4 that does not overlap with the regions corresponding to the QLQ
and WRC domains. As can be seen in Figure 5.6, three digests produced one
predominant band. The lane containing BamHI-digested DNA showed two
additional faint bands, while the lanes containing EcoRI- and Hindlll-digested
DNA showed one additional faint band. Four bands were observed in the
lane containing PstI-digested DNA. The probe used contained a PstI
restriction enzyme site, which explains the presence of more bands. This
indicates that OsDD4 is not a member of a large gene family. However, the

rice genome contains at least one and maybe two additional homologs of

151

Internode

 

N2 N1 L2b L25 L1 Co R0 0-3 3-8 8-18 old

 

   

" I   ' OsDD4

 

 

 

E37

 

Figure 5.5. Tissue-specific expression of 03004 in rice. N2, second highest
node; N1, highest node containing the shoot apex; L25, basal 2 cm of second
youngest leaf blade; L2s, basal 2 cm of second youngest leaf sheath; L1,
youngest leaf; Co, coleoptile 3 days after germination; Ro, root 3 days after
germination; 0-3; internodal region 0-3 mm above N2 containing the IM; 3-8,
internodal region 3-8 mm above N2 containing mostly the EZ; 8-18, inter-
nodal region 8-18 mm above N2 containing the upper part of the EZ and the
D2; old, oldest part of the internode. The upper panel shows hybridization
signals with 03004 as probe, the lower panel shows hybridization signals
with E37 as internal loading control.

152

 

 

 

M
14.2- . __ _ p
4'2‘ a
2.3-

0.7-

 

 

 

Figure 5.6. Southern blot analysis of 05004 in rice. Southern blot
containing rice genomic DNA, digested with either BamHl (B),
EcoRl (E), Hinolll (H), or Psll (P). The blot was probed with random
prime labeled insert derived from the 3’ 737 bp region of 08004.

153

OsDD4. The same Southern blot was also hybridized to D40170 (data not
shown). The pattern obtained was, however, different from the pattern
obtained with OsDD4 as probe. None of the faint bands on the Southern blot
shown in Figure 5.6, hybridized to D40170.

OVEREXPRESSION OF OsDD4 IN A. THALIANA

The rapid increase in transcript levels of OsDD4 in response to GA and
to submergence indicated that OsDD4 is involved in early GA-regulated
growth. The expression pattern in the internode showed that OsDD4 is
expressed exclusively in the IM, which is the region of primary
responsiveness to GA. Moreover, the protein encoded by OsDD4 is nuclear
localized suggesting a regulatory role in GA-mediated growth. To test this
hypothesis further, OsDD4 was introduced into Arabidopsis, and transgenic
seedlings were selected by plating the seeds on MS medium containing
kanamycin. To ensure that the lines obtained were independent, one seedling
per transformed plant was transplanted to soil. In cases where two seedlings
were transplanted, subsequent Southern blot analysis was performed to prove
that the lines were indeed independent (data not shown; see Fig. 5.10). In A.
thaliana ecotype Columbia, 23 independent kanamycin-resistant T1 lines
were obtained. The overexpression of OsDD4 resulted in a range of
phenotypes from wild type to severe. The phenotype became apparent in the
early vegetative adult phase when Arabidopsis grows as a rosette. Six
independent lines exhibiting a severe phenotype are described in detail in the
section below. The less severe phenotypes displayed delayed flowering and
bolting (see Fig 5.7A) and, in some cases, a fused and flattened stem

(fasciation).

154

Figure 5.7A shows the range of phenotypes observed when OsDD4 is
overexpressed in A. thaliana ecotype Columbia. Control transformed plants
had started bolting five weeks after germination, whereas, the lines
overexpressing OsDD4 were delayed in bolting. Six severe lines developed
flowers around the same time as did control transformed plants, but the
inflorescence stem did not elongate. However, the pedicel elongation was
normal in these lines, as shown in Figure 5.73. An additional phenotype
observed in the lines displaying a severe phenotype, was curly rosette leaves.
The nature of this phenotype has not been investigated, but the presence of
curly leaves predicted the severity of the phenotype.

Eventually, some stem elongation was observed in all six severe lines,
as shown in Figure 5.7C. As a result of reduced apical dominance, many
secondary inflorescence stems had elongated. The bushy appearance was
further enhanced by the severely fasciated and bifurcated inflorescence stems
as shown in Figure 5.7E. In Figure 5.7D, the inflorescence stem of an ten-
week-old overexpressing severe line was compared to the inflorescence stem
of a four-week-old wild type plant. In the overexpressing lines, stem
elongation between siliques was variable, but mostly reduced. In wild type
Arabidopsis, leaves and flowers are initiated in a spiral phyllotaxis, with
successive organs being offset by approximately 140° (see Fig. 5.7D; Furner and
Pumfrey, 1992). Figure 5.7E shows the altered phyllotaxy in the
overexpressing lines, indicating that the initiation of the flowers is highly
irregular compared to wild type. The apparent alteration in phyllotaxy may be
due to the fasciated stems. It is possible that the fused stems each display
correct phyllotaxy. Irregular phyllotaxy may then occur as a result of the
fusion of individual stems into one flattened stem. Changes in phyllotaxy of

the rosette leaves were not recorded.

155

 

A. thaliana Columbia

35S::DD4

 

Figure 5.7. Overexpression of 05004 in T1 lines
of A. thaliana Columbia.

156

 

 

Figure 5.7. Overexpression of 05004 in T1 lines of A. thaliana Columbia.

(A) Five-week-old Arabidopsis plants transformed with 05004 (left six pots) or
transformed with vector without insert (right four pots). (B) Five-week-old plants
comprising the severe class. Note the curly leaves and the flowering prior to
bolting. (C) Plants displaying the severe phenotype, 10 weeks after germination.
(D) A ten-week-old stem from a severe line (left) and a four-week-old stem from
a wild type plant (right). (E) Close-up of a stem from a severe line. Note the
fasciation of the stem and the altered phyllotaxy compared to wild type in (D).

157

As can be seen in Figure 5.7D and E, the formation of siliques was impaired in
the severe lines overexpressing OsDD4. These lines showed complete female
sterility and impaired male fertility, although later in development a few
seed-bearing siliques were obtained from two lines. The flower organs in
Arabidopsis are formed in four concentric rings or whorls. The innermost
whorl contains the gynoecium which is composed of two fused carpels, a
short style and stigmatic tissue on top of the style (Clark and Meyerowitz,
1994). In Figures 5.8A to C, the development of wild type flowers and siliques
is shown. That of the overexpressing lines is shown in Figures 5.8D to H. The
most noticeable phenotype was the partially unfused carpels, visible at flower
stage 12 as shown in Figure 5.8D (Smyth et al., 1990). The ovules, which
develop at the edge of each carpel, became exposed at later stages of flower
development. In some cases instead of an ovule a filamentous structure was
formed, as indicated most clearly in Figure 5.8G. At the time when wild type
stigmas were pollinated and the developing siliques were elongating, the
siliques of the mutant flowers were not elongating, as shown in Figures 5.8G
and H. The unfused carpels were observed in all flowers in all six lines
displaying the severe phenotype.

During development of the plant, the carpel defect worsened. Early in
development, the gynoecium showed a relatively normal stigma which,
however, was not as symmetrical shaped as wild type (see Fig. 5.8E). These
gynoecia developed into siliques as shown in Figure 5.8G, and displayed
various growth alterations. Flowers formed later in development showed
changes in identity of the inner whorl. The stigmatic tissue was replaced by a
leaf-like outgrowth as shown in Figure 5.8F, and the top of the gynoecium

was completely open. Siliques developing from these gynoecia are shown in

158

Figure 5.8. Overexpression of 03004 in T1 lines of A. thaliana Columbia
flowers. Development of wild type flowers and siliques (A to C) and the
development of flowers and siliques in the severe lines (D to K). (A) and
(D) stage 12 flower. (B), (E) and (F) flower at time of pollination. (C), (G)
and (H) maturing siliques. The arrow in (G) indicates the replacement of
an ovule with a filamentous structure. (I) and (J) Inflorescence stem of an
14-week-old severe line. (K) Inflorescence stem of an 18-week-old severe
line. The size of the images in Figures (A) to (H) are on the same scale,
as are the images in Figures (I) to (K).

159

Figure 5.8.

Overexpression of 03004 in
A. thaliana Columbia flowers

 
   

160

Figure 5.8H. The leaf-like structure continued elongating somewhat as if the
carpel were reverting to a leaf. Even later in development, the leaf-like carpel
phenotype had extended to the other flower organs. In Figure 5.81, a transition
from relatively normal flowers to flowers containing only leaf-like carpels is
shown. The oldest flower showed all organs and appeared normal, except for
the unfused carpels. The second oldest flower showed two petals which were
reduced in size, while the other two appeared normal. The third oldest flower
showed a decrease in floral organ number and a change in organ identity,
while in the youngest flower all organs displayed a leaf-like carpel structure.
A partial reversion of this progression is shown in Figure 5.8]. Also shown in
Figure 5.8] is the absence of the pedicel in the leaf-like carpel flowers. Instead,
the organs in this altered flower were emerging in a spiral phyllotaxy, and
elongation between the organs was observed (see also the oldest flower in Fig.
5.8K). This is a pattern reminiscent of the inflorescence stem. As shown in
Figure 5.8K, many flowers eventually displayed a complete change to leaf-like
carpel flowers in which stem elongation between the successive flowers was
severly reduced. This suggested that overexpression of OsDD4 may lead to a
change in the identity from a floral meristem to an inflorescence or even a
vegetative meristem.

OsDD4 was also overexpressed in A. thaliana ecotype Landsberg erecta,
and 16 independent T1 lines were obtained. The phenotypes ranged from
wild type to severe. Two Landsberg erecta lines showed an extreme severe
phenotype, similar to the phenotypes observed in the Columbia background
described above. However, flowering was severely delayed, and the
inflorescence stem never elongated. In one line, the only flowers observed
were reminiscent of the leaf-like carpel flowers shown in Figure 5.8K. In the

other line, the flowers contained partially unfused carpels (data not shown).

161

Figure 5.9. Overexpression of 03004 in T1 lines of A. thaliana Landsberg
erecta. (A) Four-week-old Arabidopsis plants transformed with 03004
(left four pots) or transformed with vector without insert (right two pots).
(B) Four-week-old plants comprising the less-severe class. Note the curly
leaves and the enhanced compact inflorescence. (C) Inflorescence stem
of an eight-week-old plant in the less-severe class. Note the fasciated
stem.

162

 

_\ :1;

,‘kkfiyrk—r—k—r—r—r—W —————— fi1’1-g'i-tii‘. /
I ( ..
r I,

l I.

 

/
I

 

A- "man" Landswg mm A. thaliana Landsberg erecta

35S::DD4 ‘ CONTROL

 

 

Figure 5.9. Overexpression of 05004 in T1 lines of A. thaliana Landsberg erecta.

163

As for the Columbia lines overexpressing OsDD4, the less severe Landsberg
erecta T1 lines showed delayed bolting compared to control transformed
plants (see Fig. 5.9A), and showed fasciated stems (see Fig. 5.9C). The
difference between the phenotypes observed in the Columbia versus
Landsberg erecta background was that the phenotype in Landsberg erecta was
more severe at an earlier age (in the early vegetative adult phase) than in
Columbia. In contrast to Columbia, the Landsberg erecta lines (except for the
two described above) overcame the dwarf phenotype and bore fertile flowers,
as shown in Figures 5.93 and C. In Figure 5.93 the curly leaves and very
compact inflorescence are shown, although the plants have started bolting.
Shown in Figure 5.9C is the stem fasciation and the presence of siliques

developed from later flowers.

SOUTHERN AND NORTHERN BLOT ANALYSES OF A. THALIANA
OVEREXPRESSING OsDD4

Southern and Northern blot analyses were performed with 12 T1 lines
overexpressing OsDD4 in the Columbia background. This included the six
severe lines. In Figure 5.10, Southern blot analysis indicated that these lines
were independent, and carried one to several copies of the transgene. The
genomic DNA was digested with BamHI, which cuts once within the T-DNA
inserted into the Arabidopsis genome. The presence of more than one band
unequivocally determines the presence of more than one insert in the
genome. Interestingly, Southern blot analysis indicated that a single band was
present in the six severe lines 11, 12, 13, 16, 22.2, and 23, suggesting the
presence of one copy of the transgene. The remaining lines contained at least

two inserts and showed the less severe phenotype. A band of 9.4 kb on the

164

5 7 8 11 12 13 15 16 18 22.1222 23

 

Figure 5.10. 05004 copy number in T1 lines of A. thaliana Columbia.

The Southern blot contains BamHl-digested DNA. The blot was hybridized
with random prime labeled probe derived from the full length 05004 cDNA.
Numbers above the lanes indicate the individual lines.

165

Southern blot indicates the presence of two T-DNA copies inserted head-to-
head in the same location. This band was observed in the severe lines 13 and
16. Therefore, the presence of one or two T-DNA inserts in these lines cannot
be determined from the data presented in Figure 5.10. Northern blot analysis
of the same lines was performed to correlate the severity of the observed
phenotype with the OsDD4 transcript levels. In Figure 5.11, the upper panel is
a Northern blot hybridized with a probe derived from the 3' 737-bp region of
OsDD4, while the bottom panel is hybridized with a probe derived from the 5'
107-bp region of OsDD4. Equal RNA loading was confirmed by ethidium
bromide staining of the ribosomal RNA (data not shown). Based on intensity
of the bands in Figure 5.10, lines containing many copies of the transgene
(lines 5, 7, and 8) showed high levels of transcript, ranging from full-length to
partial-size transcripts. The lines containing one or few copies showed less
RNA. The level of full-length OsDD4 could, however, not be correlated to the
phenotype. Only the presence of an RNA species, approximately 100 bp
shorter than full-length shown in lines 5, 7, 8, 15, 18, and 22.1, correlated with

the less severe phenotype.

5.5. DISCUSSION

We have identified a novel gene, OsDD4, whose expression is rapidly
induced in the IM of rice internodes by GA and submergence. Southern blot
analysis indicated that OsDD4 is part of a small gene family, and database
searches suggest that OsDD4 may belong to a larger class of regulatory proteins

containing QLQ and WRC domains. The overexpression of OsDD4 in

166

5 7 811 1213151618 22.122.223

 

 

 

 

 

 

Figure 5.11. 05004 transcript levels in T1 lines of A. thaliana
Columbia. Northern blot containing RNA isolated from mature
leaves of A. thaliana ecotype Columbia T1 lines. The top panel

is hybridized with random prime labeled probe derived from the

3’ 737 bp region of 05004. The bottom panel is hybridized with
an RNA probe derived from the 5' 107 bp region of 05004. The
arrow indicates the full length RNA (~1.8 kb). The numbers above
the lanes indicate the individual lines.

167

Arabidopsis led to a pleiotropic phenotype, suggesting that overexpression

resulted in changes in several developmental pathways.

The amino acid sequence of OsDD4 indicated several features
reminiscent of transcription factors, including a putative NLS. The most
conserved region comprises the WRC domain, which is found in several
plant genes downstream from the less conserved QLQ domain. The function
of the WRC domain-containing plant genes is unknown, since most of those
genes were identified by genome sequencing. While most of these genes show
no similarity to known proteins, AC000106 does exhibit high sequence
similarity to ENBP1, a factor thought to be involved in transcriptional
regulation of ENODIZ (Christiansen et al., 1996). ENBPl and AC000106
contain two conserved zinc-finger motifs, which, in ENBP1, were shown to
bind DNA. Although the WRC domain itself is not found in ENBPl but only
in AC000106, this suggests that WRC domains may be found in nuclear
proteins.

The less conserved QLQ domain is present in proteins related to
SWIZ/SNFZ from yeast. SW12/ SNF2 is a subunit of a large 2 MDa complex
that uses the energy of ATP hydrolysis to drive transcription factors onto
nucleosomal transcription-factor-binding sites (Peterson and Tamkun, 1995).
While the orthologs are highly related in the C-terminal region, the
homology at the N-terminus between the yeast, Drosophila, human and
chicken is restricted to two short regions. These regions were called domain I
and H in brahma, the Drosophila ortholog (Tamkun et al., 1992) and are
present in the N-terminally located proline-rich and highly charged domain,
respectively, of the SW12/ SNF2 orthologs (Peterson and Tamkun, 1995). The

QLQ domain shows similarity to domain I, and this domain appears to be

168

conserved from yeast to plants to mammals. Genetically, the yeast SWI—SNF
complex was shown to be essential for nuclear receptor-activated
transcription in yeast (Yoshinaga et al., 1992). Using the yeast two-hybrid
system, the region of the human orthologs of SWIZ/SNFZ important for
ligand-dependent interaction with the estrogen receptor was found to span
the two conserved domains, including QLQ (Ichinose et al., 1997). This
suggests that QLQ itself may be involved in the interaction. Gibberellins are
lipophilic molecules that may traverse membranes by diffusion. Although
evidence from other GA-regulated processes suggests that GA is perceived at
the plasma membrane (Hooley et al., 1991), it is also possible that GA is
perceived by cytoplasmic receptors. In analogy to the animal nuclear receptors
that, after ligand binding, activate transcription of target genes, OsDD4 may
provide the coactivator of transcription of genes that are regulated by GA-
receptor transcription factors.

OsDD4 contains a functional NLS and this motif is most likely located
in the conserved WRC domain. The distribution of the OsDD4-GUS fusion
protein in both the cytoplasm and the nucleus indicates that the NLS is not
very efficient and that other factors may be involved in regulating the cellular
localization of OsDD4. It will be interesting to know whether the cellular
localization of OsDD4 is regulated by GA. Usually, the amino acid sequence
surrounding NLSs is not highly conserved. The amino acids surrounding the
putative NLS in OsDD4 comprise the most conserved motif found in this
protein. For transcription factors in the basic-domain leucine-zipper class
(bZIP), a functional NLS is present in the conserved DNA binding domain
(Varagona and Raikhel, 1994). In several other classes of transcription factors,

NLSs were found overlapping with or adjacent to DNA-binding domains as

169

well (Hicks and Raikhel, 1995). This suggests that one of the functions of the
WRC domain may be in DNA binding.

Overexpression of OsDD4 in Arabidopsis leads to delayed bolting, to
fasciated stems and altered phyllotaxy. In addition, in severe lines it also leads
to reduced apical dominance, flowering before bolting, impaired inflorescence
stem elongation, altered leaves and flowers, impaired male fertility, and
complete female sterility. Later in development, changes in flower organ
identity and number, and changes in floral meristem identity were observed.
GA positively regulates flowering, stem elongation (bolting), apical
dominance, male fertility, and vegetative phase change. Although flowering
time was not affected in the severe Columbia lines, the other processes were
negatively regulated in the lines overexpressing OsDD4. This would suggest
that OsDD4 is a negative regulator of GA-regulated processes, much like GAI
and SPY. However, the rapid increase in transcript levels of OsDD4 in the IM
of rice internodes in response to GA and submergence suggest a positive role
of this gene in GA-regulated processes. The alternative explanation then for
the observed phenotype is that overexpression leads to a dominant negative
phenotype. The overexpression of the rice DD4 in Arabidopsis may inactivate
the larger protein complex in which the endogenous DD4 is an essential
component. Alternatively, overexpression of OsDD4 may titrate out all
components important for endogenous DD4 to function properly. Reduced
stem elongation, reduced apical dominance, reduced male fertility, and
delayed bolting are observed as well in the gal-3/rga double mutant
(Silverstone et al., 1997) and in gai (except male fertility; Koomneef et al.,
1985). The ga1-3 mutant is a severe GA biosynthesis mutant whose

phenotype is partially restored by a mutation in the RCA gene. Application of

170

GA leads to full reversion of the mutant phenotype to wild type (Silverstone
et al., 1997). The gai mutant confers GA insensitivity and GA application will
not restore the phenotype (Koornneef et al., 1985). The phenotype observed in
the severe lines overexpressing OsDD4 was not restored by GA application
either.

Not all GA-regulated processes were affected. For example the rate of
germination and the length of the pedicel were similar to control
transformed plants. On the other hand, some processes not known to be
regulated by GA were affected. For example, extreme fasciation was observed
in the severe Columbia lines and, to a lesser extent, in the weaker Columbia
and Landsberg erecta lines. The few fasciated mutants described in
Arabidopsis display altered phyllotaxy, reduced stem elongation, a change in
floral organ number (Leyser and Furner, 1992), and delayed bolting (Medford
et al., 1992). The fasciation in those mutants was associated with an enlarged
apical meristem, and it is, therefore, likely that overexpression of OsDD4 leads
to an enlarged meristem. Fasciation can be the result of a disease caused by
Corynebacterium fascians and has been linked to increased cytokinin levels
(Agrios, 1978). It is, therefore, possible that overexpression of OsDD4 interferes
with cytokinin signaling. Another process not known to be regulated by GA is
gynoecium development. Partially unfused carpels and replacement of
ovules for filamentous structures were observed in all flowers of the six
severe Columbia lines and one severe Landsberg erecta line. A null mutation
at the TOUSLED (Roe et al., 1993; 1997) locus leads to impaired carpel fusion,
delayed flowering, and a loss of floral organs. Mutations at the LE UNIG locus
(Liu and Meyerowitz, 1995) also affect carpel fusion, and mutations at the
AINTEGUMENTA locus lead to similar gynoecium and ovule development

defects as observed for severe lines overexpressing OsDD4 (Elliot et al., 1996).

171

This suggests that OsDD4 overexpression interferes with pathways in which
TOUSLED, LEUNIG, and/ or AINTEGUMENTA play an important role.

The pleiotropic phenotypes observed suggest that OsDD4 plays a role in
some GA-regulated processes as well as in other processes in plant
development. On the other hand, overexpression of OsDD4 under a general
and highly active promoter could potentially affect several developmental
pathways due to the inappropriate expression of the transgene. It is possible,
for example, that overexpression of OsDD4 interferes with the function of
other proteins containing the QLQ and WRC domains. Ectopic expression of a
gene can help reveal its in viva role as shown for the maize homeo box gene,
KNOTTED (Sinha et al., 1993). In the case of OsDD4, for example, the putative
change in floral meristem identity due to overexpression may not indicate a
role for this gene in floral meristems but in vegetative meristems, where it
may play a role in delaying vegetative phase change. During submergence of
deepwater rice, the IM becomes more sensitive to GA and this leads to
increased growth (Raskin and Kende, 1984a). Flowering in deepwater rice is
under strict photoperiod control. However, in Arabidopsis, which is a
facultative long day plant, flowering time under short days can be reduced by
application of GA. Flowering time in deepwater rice is correlated with the
receding water levels at the end of the growing season. This is a very
important trait since plants that flower prematurely will drown, because of
the change from an indeterminate to a determinate shoot meristem.
Therefore, increased sensitivity of the rice tissue to GA should not lead to
flowering but only to enhanced internodal elongation. Maintenance of the
apical meristem in the vegetative phase will accomplish this.

The transgene copy number correlated with the phenotype, suggesting

that, in the case of two or more inserted T-DNAs, partial or complete

172

cosuppression of the transgenes was observed. Cosuppression is manifested at
the transcriptional or the post-transcriptional level, resulting in reduced
RNA levels (Matzke et al., 1996). However, in the lines examined, the
transcript levels did not correlate to the phenotype. Therefore, it is unclear
how the observed phenotypes are attained. However, the identification of six
independent T1 lines in Columbia and two in Landsberg erecta displaying
similar severe phenotypes suggests that the phenotype observed is not due to

position effects of the transgene in the Arabidopsis genome.

In conclusion, overexpression of OsDD4 in Arabidopsis leads to
negative regulation of several growth-related processes. To understand
whether this is achieved via a dominant negative interference or via the
overexpression of a negative regulator of growth will require the analysis of
transgenic plants which down-regulate the endogenous DD4. The phenotypes
observed can be classified in three groups: (i) Fasciation, which, in severe
cases, resulted in reduced height, altered phyllotaxy and altered organ
numbers. (ii) Altered organ morphology, which resulted in altered leaves and
partially unfused carpels. (iii) Changes in organ and meristem identity, which
resulted in altered flower organs, a floral meristem identity switch, and
delayed vegetative phase change. All these phenotypes combined suggest that
overexpression leads to a change in apical meristem structure and function.
However, our current knowledge about OsDD4 is too limited to distinguish
between an enhancement of its in viva role or an interference in an
unrelated signaling pathway. To understand the role of OsDD4 in the
meristem and to understand its role in GA signaling, it will be important to

know more about OsDD4 in rice and about its ortholog in Arabidopsis.

173

5.6. LITERATURE

Altschul, S.F., Madden, T.L., Schaffer, A.A., Zhang, J., Zhang, Z., Miller, W.
and Lipman, D.]. (1997) Gapped BLAST and PSI-BLAST: a new generation
of protein database search programs. Nucleic Acids Res 25:3389-3402.

Agrios, G.N. (1978) Plant Pathology. Academic Press, New York, New York,
USA.

An, G., Ebert, P.R., Mitra, A. and Ha, SB. (1988) Binary vectors. In Plant
Molecular Biology Manual A3. Kluwer Acad Publishers, Dordrecht, the
Netherlands, pp 1-19.

Ausubel, F.M., Brent, R., Kingston, R.E., Moore, D.D., Seidman, J.G., Smith,
J.A. and Struhl, K. (1987) Current Protocols in Molecular Biology (Wiley,
New York).

Bleecker, A.B., Schuette, ].L. and Kende, H. (1986) Anatomical analysis of
growth and developmental patterns in the internode of deepwater rice.
Planta 169:490-497.

Catling, H.D. (1992) Rice in Deepwater. The Macmillan Press Ltd, London and
Basingstoke, UK.

Christiansen, H., Hansen, A.C., Vijn, I., Pallisgaard, N ., Larsen, K., Yang, W.-
C., Bisseling, T., Marcker, K.A. and Jensen, ED. (1996) A novel type of
DNA-binding protein interacts with a conserved sequence in an early
nodulin ENODIZ promoter. Plant Mol Biol 32:809-821.

Clark, SE. and Meyerowitz, EM. (1994) Flower development. In Arabidopsis.
E.M. Meyerowitz and CR. Somerville, eds. Cold Spring Harbor, New York:
Cold Spring Harbor Laboratory, pp 435-466.

Elliot, R.C., Betzner, A.S., Huttner, E., Oakes, M.P., Tucker, W.Q.J., Gerentes,
D., Perez, P. and Smyth, DR. (1996) AINTEGUMENTA, an APETALAZ-
like gene of Arabidopsis with pleiotropic roles in ovule development and
floral organ growth. Plant Cell 8:155-168.

Furner, 1.]. and Pumfrey, ].E. (1992) Cell fate in the shoot apical meristem of
Arabidopsis thaliana. Development 115:755-764.

Hicks, GR. and Raikhel, NV. (1995) Protein import into the nucleus: an
integrated view. Annu Rev Cell Dev Biol 11:155-188.

174

Hoffmann-Benning, S. and Kende, H. (1992) On the role of abscisic acid and
gibberellin in the regulation of growth in rice. Plant Physiol 99:1156-1161.

Hooley, R., Beale, M.H. and Smith, 8.]. (1991) Gibberellin perception at the
plasma membrane of Avena fatua aleurone protoplasts. Planta 183:274-
280.

Ichinose, H., Gamier, ].-M., Chambon, P. and Losson, R. (1997) Ligand-
dependent interaction between the estrogen receptor and the human
homologues of SWIZ/SNFZ. Gene 188:95-100.

Jacobsen, S.E., Binkowski, K.A. and Olszewski, NE. (1996) SPINDLY, a
tetratricopeptide repeat protein involved in gibberellin signal transduction
in Arabidopsis. Proc Natl Acad Sci USA 93:9292-9296.

Koornneef, M., Elgersma, A., Hanhart, C.]., van Loenen-Martinet, E.P., van
Rijn, L. and Zeevaart, J.A.D. (1985) A gibberellin insensitive mutant of
Arabidopsis thaliana. Plant Physiol 65:33-39.

Leyser, H.M.O. and Furner, 1.]. (1992) Characterization of three shoot apical
meristem mutants of Arabidopsis thaliana. Development 116:397-403.

Liang, P. and Pardee, A3. (1992) Differential display of eukaryotic messenger
RNA by means of the polymerase chain reaction. Science 257:967-971.

Liu, Z. and Meyerowitz, EM. (1995) LE UNIG regulates AGAMOLIS expression
in Arabidopsis flowers. Development 121:975-991.

Matzke, M.A., Matzke, A.].M. and Eggleston, W.B. (1996) Paramutation and
transgene silencing: a common response to invasive DNA. Trends Plant
Sci 1:382-388.

Medford, ].I., Behringer, F.]., Callos, JD. and Feldmann, K.A. (1992) Normal
and abnormal development in the Arabidopsis vegetative shoot apex.
Plant Cell 4:631-643.

Métraux, J.-P. and Kende, H. (1983) The role of ethylene in the growth
response of submerged deepwater rice. Plant Physiol 72:441-446.

Mitchell, P.]. and Tjian, R. (1989) Transcriptional regulation in mammalian
cells by sequence-specific DNA binding proteins. Science 245:371-378.

Olszewski, N. (1997) Evidence that dynamic O—linked N-acetylglucosamine

modification of proteins plays a role in gibberellin signal transduction. 8th
International conference on Arabidopsis research. Madison, Wisconsin.

175

Peng, ]., Carol, P., Richards, D.E., King, K.E., Cowling, R.J., Murphy, GP. and
Harberd NP. (1997) The Arabidopsis GAI gene defines a signaling pathway
that negatively regulates gibberellin responses. Genes Dev 11:3194-3205.

Peterson, CL. and Tamkun, J.W. (1995) The SWI-SNF complex: a chromatin
remodeling machine? Trends Biochem Sci 20:143-146.

Puissant, C. and Houdebine, L.M. (1990) An improvement of the single-step
method of RNA isolation by acid guanidium thiocyanate-phenol-
chloroform extraction. BioTechniques 8:148-149.

Raskin, I. and Kende, H. (1984a) Role of gibberellin in the growth response of
submerged deepwater rice. Plant Physiol 76:947-950.

Raskin, I. and Kende, H. (1984b) Regulation of growth in stem sections of
deepwater rice. Planta 160:66-72.

Roe, ].L., Rivin, C.]., Sessions, R.A., Feldmann, K.A. and Zambryski, RC.
(1993) The Tausled gene in A. thaliana encodes a protein kinase homolog
that is required for leaf and flower development. Cell 75:939-950.

Roe, J.L., Nemhauser, ].L. and Zambryski, RC. (1997) TOUSLED participates in
apical tissue formation during gynoecium development in Arabidopsis.
Plant Cell 9:335-353.

Sauter, M. and Kende, H. (1992) Gibberellin-induced growth and regulation of
the cell division cycle in deepwater rice. Planta 188:362-368.

Sauter, M., Seagull, R.W. and Kende, H. (1993) Internodal elongation and
orientation of cellulose microfibrils and microtubules in deepwater rice.
Planta 190:354-362.

Sauter, M., Mekhedov, S.L. and Kende, H. (1995) Gibberellin promotes
histone H1 kinase activity and the expression of cdc2 and cyclin genes

during the induction of rapid growth in deepwater rice internodes. Plant]
7:623-632.

Sauter, M. (1997) Differential expression of a CAK (cdc2-activating kinase)-like
protein kinase, cyclins and cdc2 genes from rice during the cell cycle and in
response to gibberellin. Plant] 112181-190.

Silverstone, A.L., Mak, P.Y.A., Martinez, EC. and Sun, T.-p. (1997) The new

RGA locus encodes a negative regulator of gibberellin response in
Arabidopsis thaliana. Genetics 146:1087-1099.

176

Silverstone, A.L., Ciampaglio, C.N. and Sun, T.-p. (1998) The Arabidopsis
RGA gene encodes a transcriptional regulator repressing the gibberellin
signal transduction pathway. Plant Cell 10:155-169.

Sinha, N.R., Williams, RE. and Hake, S. (1993) Overexpression of the maize
homeo box gene, KN OTTED-I, causes a switch from determinate to
indeterminate cell fates. Genes Dev 7:787-795.

Smyth, D.R., Bowman, ].L. and Meyerowitz, EM. (1990) Early flower
development in Arabidopsis. Plant Cell 2:755-767.

Stiinzi, J.T. and Kende, H. (1989) Gas composition in the internal air spaces of
deepwater rice in relation to growth induced by submergence. Plant Cell
Physiol 30:49-56.

Tamkun, J.W., Deuring, R., Scott, M.P., Kissinger, M., Pattatucci, A.M.,
Kaufman, TC. and Kennison, ].A. (1992) brahma: A regulator of
Drosophila homeotic genes structurally related to the yeast transcriptional
activator SNF2/ SW12. Cell 68:561-572.

Van der Knaap, E. and Kende, H. (1995) Identification of a gibberellin-induced
gene in deepwater rice using differential display of mRNA. Plant Mol Biol
28:589-592.

Van Hoof, A. and Green, P.]. (1996) Premature nonsense codons decrease the
stability of phytohemagglutinin mRNA in a position-dependent manner.
Plant] 10:415-424.

Varagona, M.]., Schmidt, R.]. and Raikhel, N.V. (1992) Nuclear localization
signal(s) required for nuclear targeting of the maize regualtory protein
Opaque-2. Plant Cell 4:1213-1227.

Varagona, M]. and Raikhel, N.V. (1994) The basic domain in the bZIP
regulatory protein Opaque2 serves two independent functions: DNA
binding and nuclear localization. Plant] 5:207-214.

Wilson, RN. and Somerville, CR. (1995) Phenotypic suppression of the

gibberellin-insensitive mutant (gai) of Arabidopsis. Plant Physiol 1082495-
502.

Yoshinaga, S.K., Peterson, C.L., Herskowitz, I. and Yamamoto, KR. (1992)

Roles of SW11, SW12, and SW13 proteins for transciptional enhancement
by steroid receptors. Science 258:1598-1604.

177

CHAPTER 6

A putative type 1a plasma membrane receptor is induced by

gibberellin in deepwater rice

6.1. ABSTRACT

In search for differentially expressed genes, a novel gene was identified
whose transcript levels had increased in response to gibberellin (GA) in the
internodes of deepwater rice. The expression was high in regions undergoing
cell division and lower in the elongation and differentiation zones. The gene
may encode a type 1a receptor with an extracellular domain, a single

transmembrane domain and a short cytoplasmic domain.

178

6.2. INTRODUCTION

Survival of deepwater rice during flooding is based on its capacity for
rapid internodal elongation when it becomes submerged. The signal for
accelerated growth is an increase in ethylene levels, which enhances the
responsiveness of the internode to GA (Raskin and Kende, 1984a). Enhanced
growth is initiated in the intercalary meristem (IM) at the base of the growing
internode and is based on increased production of cells and increased
elongation of these cells after they emerge from the meristem into the
elongation zone (EZ) (Bleecker et al., 1986; Sauter and Kende, 1992). To
understand the molecular events taking place during growth, a search was
initiated to identify genes that are differentially regulated by GA in the 1M of
rice internodes. By means of differential display (Liang and Pardee, 1992),
several genes have been identified whose transcript levels increase in
response to GA (Van der Knaap and Kende, 1995; Van der Knaap et al., 1997).
Here, we report on the identification of OsDD3, a gene whose transcript levels
increase in response to GA and to submergence in the 1M of deepwater rice
internodes. The gene product may encode a novel type 1a receptor located at

the plasma membrane.

6.3. MATERIALS AND METHODS

PLANT MATERIAL

Seeds of deepwater rice (Oryza sativa L., cv. Pin Gaew 56) were obtained from

the International Rice Research Institute (Los Bafios, Philippines). Plants were

179

grown as described in Stiinzi and Kende (1989). For submergence
experiments, 12-week-old plants were partially immersed under continuous
light (Métraux and Kende, 1983). Twenty-cm-long stem sections containing
the growing internode were excised and treated with 50 pM GA3 (Raskin and
Kende, 1984b). Incubation was allowed to proceed for the periods indicated,
after which the 1M was excised, frozen immediately, and stored at -80°C until

use.

IDENTIFICATION OF OsDD3

A differentially displayed 28 bp cDNA band, dd3, was identified using
primers T12MG and OPA04 (Van der Knaap and Kende, 1995). The differential
displayed product was inserted in pUC19, and the cloned cDNA was used to
screen a IM-specific cDNA library to obtain a full-length cDNA insert, OsDD3.
The phage insert was cloned into the EcaRI site of pBluescript SK(-) phagemid
(Stratagene) and sequence analysis was performed at the Biochemistry Facility
of the Plant Research Laboratory at Michigan State University. The sequences

were aligned using Sequencher, version 3.0 (Gene Codes Corporation).

NORTHERN BLOT ANALYSIS

Twenty pg of total RNA, isolated according to Puissant and Houdebine (1990),
was electrophoretically separated in a 1.2% formaldehyde-agarose gel
(Ausubel et al., 1987) and transferred to Hybond-N+ membrane (Amersham).
For OsDD3, the RNA probe was made from the BstXI - EcoRI fragment
containing the 3' UTR (see Fig. 6.5) in the presence of aI-[32P]UTP (NEN). For
cchsI, the RNA probe contained the 3' UTR and most of the coding region

180

(Sauter et al., 1995). Blots were prehybridized and hybridized in 3x SSPE, 10x
Denhardt's solution (Sambrook et al., 1989), 0.5% SDS, 50 pg/ ml denatured
salmon sperm DNA, and 50% deionized formamide (BMB) at 65°C
overnight. Blots were washed twice for 5 min each in 2x SSC and 0.5% SDS at
65°C followed by two washes in 0.1x SSC and 0.5% SDS for 30 min each.
Probes for histone H3 and E37 were prepared by random prime labeling and
hybridized to the blots as described in Van der Knaap et al. (1997). The
radioactivity on blots was quantified by Phosphorlmager analysis (Molecular

Dynamics).

SOUTHERN BLOT ANALYSIS

Rice genomic DNA was isolated from a CsCl gradient according to Ausubel et
al. (1987). Five pg of DNA was digested overnight, and fragments were
separated on a 0.8% agarose gel for 20 h at 30 V. The gel was then treated for 15
min in 0.25 N HCl, 30 min in 0.5 N NaOH and 1.5 M NaCl and 15 min in 1 M
Tris-HCI, pH 8 and 1.5 M NaCl after which the DNA was transferred to
Hybond N+ (Amersham). Blots were prehybridized in 5x SSC, 10x Denhardt's
solution, 0.2% SDS, 0.1 M K-phosphate, pH 6.8 and 100 pg/ ml denatured
salmon sperm for 4 h and hybridized in 5x SSC, 10x Denhardt's solution, 0.1%
SDS, 0.1 M K-phosphate, pH 6.8, 10% dextran sulphate, 30% deionized
formamide, and 100 pg/ ml denatured salmon sperm overnight at 42°C. A
probe derived from the BstXI - EcoRI fragment (see Fig. 6.5) was prepared in
the presence of at-[32P]dCTP (NEN) with a random prime labeling kit (BMB).
Blots were washed twice for 30 min each in 2x SSC and 0.5% SDS at 65°C
followed by two washes in 0.1x SSC and 0.5% SDS for 30 min each.

181

IN-VITRO TRANSLATION REACTION

A plasmid containing a full-length OsDD3 1 10.1, was linearized, and sense
RNA was prepared in an in-vitro transcription reaction with T3 RNA
polymerase (BMB). RNA was purified by organic extractions and precipitated
with ethanol. Approximately 100 ng of RNA was translated in a rabbit
reticulocyte lysate (Promega) in the presence of [355]methionine (1,200

Ci / mmol; NEN) or [3H]leucine (100-200 Ci/mmol; NEN) in a total volume of
25 pl. The translation reaction was allowed to proceed at 30°C for 1 h, either
in the presence or absence of 1.8 pl canine pancreatic microsomal membranes
(Promega). Three pl of translation products were separated on 15% SDS-
PAGE. Prior to drying, the gel was fixed in 30% methanol and 10% acetic acid
and subsequently incubated in 1M Na-salicylate, pH 6, to enhance the signal

for autoradiography.

6.4. RESULTS

GA treatment of rice stem sections, containing the growing internode,
led to an increase in transcript abundance of OsDD3 as shown in Figure 6.1A.
The signals were quantified and normalized for equal loading using a cDNA,
E37, corresponding to a gene whose transcript levels did not change over the
course of the experiment. Figure 6.13 shows that the level of OsDD3 in the 1M
of GA-treated stem sections had increased over 3-fold 4 h after start of GA
treatment. The expression pattern of OsDD3 was also investigated in
submerged plants as shown in Figure 6.2A. The transcript levels had

increased after 8 h, reaching a 6-fold increase 24 h after start of treatment (Fig.

182

0123456781524h

 

OsDD3

 

 

 

 

 

 

.h

N
omAmwam-h'm
l

(A)
1 I

—L

J: l

Relative transcript levels
0

 

 

I l I I I I

10 15 2o 25 30
GA treatment (h)

C
01

Figure 6.1. GA-induced expression of 05003 in the IM of rice stem
sections. Rice stem sections were incubated in 50 pM GAa for the times
indicated above the lanes, after which RNA was isolated from the IM.
(A) Northern blot, containing 20 pg of RNA, was hybridized to 05003
and 537. (B) Quantitative analysis of the Northern blot shown in (A)
using a Phosphorlmager. All values were normalized to the E37 loading
control.

183

Figure 6.2. Submergence-induced expression of the 03003, histone
H3 and cchs1 genes. Plants were submerged for the times indicated
above the lanes, after which RNA was isolated from the IM. (A) The
Northern blots were hybridized to 03003, histone H3, cchs1 and
E37. (B) Quantitative analysis of the Northern blots by Phosphor-
lmager. All values were normalized to the E37 loading control
obtained from the same blot.

l8-l

02468101624h

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

OsDD3
Histone H3
cchs1
E37
18 -
In
E 16 -
g 14 -
EL 12 -
3
g 10-
In 8 -
b
2 6'
._ _ -°-OsDD3
E 4 +histone Ha
I» 2 - +cyc0s1
I | '
0 7 I l l l T!
0 5 10 15 20 25 30

Submergence (h)

Figure 6.2. Submergence-induced expression of the 05003, histone H3 and
cchs1 genes.

185

6.23). The increase in transcript level for OsDDB was compared to the increase
in transcript level for histone H3 and cchsI. Histone H3 is a marker for the
S-phase of the cell cycle (Van der Knaap and Kende, 1995) and cchs1 is a
marker for the transition of the G2 to M phase (Sauter et al., 1995; Sauter,
1997). As shown in Figure 6.23, the increase in expression of OsDDB coincided
with an increase in histone H3 transcript levels. This was followed by an

increase in cchsI transcript levels.

The tissue-specific expression of OsDD3 is shown in Figure 6.3. The
transcript for OsDD3 was detected in many tissues of rice. However, the
highest level was found in the basal part of the sheath of the second youngest
leaf, the youngest leaf and in the 1M. Slightly lower expression was detected in
the coleoptile containing the youngest leaf and in the root tip. Along the
internode, OsDD3 was expressed at much lower levels in the EZ and in the
differentiation zone, than in the IM, while no transcript was detected in the

oldest part of the internode.

Southern blot analysis was employed to address the question whether
the probe used in this study is gene specific and whether close homologs of
OsDD3 exist in rice. As can be seen in Figure 6.4, all digests produced one

band, indicating that the probe used is gene specific.

A full-length clone encoding OsDD3 was obtained from an IM-specific
cDNA library, of which the largest clone (7: 5.1, see Fig. 6.53) was sequenced.
The cDNA was 1188 nucleotides long and showed an ORF of 196 amino acids
yielding a predicted protein of 20 kD (Fig. 6.5A). Searches to find homologous

sequences in the non-redundant or EST database using the BLAST program

186

Internode

 

N2 N1 L2b L28 L1 Co Flo 0-3 3-8 8-18 old

 

OsDDS

 

 

 

E37

 

 

 

Figure 6.3. Tissue-specific expression of 03003 in rice. N2, second
highest node; N1, highest node containing the apical meristem; L2b,
basal 2 cm of second youngest leaf blade; L2s, basal 2 cm of second
youngest leaf sheath; L1, youngest leaf; Co, coleoptile 3 days after
germination; Ro, root 3 days after germination; 0-3, internodal region
03 mm above N2 containing the IM; 3-8, internodal region 3-8 mm
above N2 containing mostly the EZ; 8-18, internodal region 8-18 mm
above N2 containing the upper part of the EZ and the DZ; old, oldest
part of the internode. The upper panel shows hybridization signals with
03003 as probe, the lower panel shows hybridization signals with 537
as internal loading control.

187

 

23.1-

9.4-

6.6-

 

4.4-

 

 

 

Figure 6.4. Southern blot analysis of 05003 in rice. Southern blot
containing rice genomic DNA, digested with either BamHl (B), EcoRl
(E), Hindlll (H) or Pstl (P). The blot was probed with random prime
labeled insert derived from the 3’ UTR of 05003.

188

(Altschul et al., 1990) did not identify related genes. As is indicated in Figure
6.5A, the protein contains a putative signal sequence, with a potential
cleavage site between residue 33 and 34 (von Heijne, 1986). PSORT

(httpz/ /psort.nibb.ac.jp/) predicted a type 1a membrane protein, located at the
plasma membrane. The putative transmembrane region is between residue
167 and 183, with the C-terminal tail inserted in the cytoplasm. Motif searches
using PrositeScan (http:/ /u1rec3.unil.ch/ software / PSTSCAN__form. htrnl)
indicated two consensus protein kinase C phosphorylation sites as well as a
RGD motif found in proteins mediating cell adhesion (D'Souza et al., 1991).
Besides the RGD motif, the predicted extracellular domain contains a
cysteine-rich region, a homopolymeric sequence of alanines, as well as several
proline and serine residues, but no consensus N-glycosylation site. Other
inserts obtained from the cDNA library were shorter than 71 5.1. As can be
seen in Figure 6.53, this was, in part, because of different additions of the
poly(A) tail. While 1 10.1 had the same addition of the poly(A) tail as 1 5.1,
the poly(A) tail of 10.4 and 5.4 were 36 bp and 91 bp respectively upstream of
that in 1 5.1. The original differentially displayed product, dd3, was amplified
by the oligo dT primer at the 3'end indicating that this product was derived
from an mRNA species containing the poly(A) tail 143 bp upstream of the
one in 1 5.1. All four inserts identified from the cDNA library contained the

same putative ORF.

In-vitro translation reactions were performed to determine whether
OsDD3 encodes a protein of approximately 20 kD. Also, the presence of the
signal sequence was tested by allowing the reactions to take place in the
presence or absence of dog pancreatic microsomes. A functional signal

sequence will direct the translating ribosomes to the microsomes. After

189

A

 

 

 

 

 

 

 

 

 

 

50
AVAWAGEEEKVRLGSSPPSQYSKCYG: SPDVAVQVPTLSAPSVPAAAAA 1 o 0
AARRWQLQAARVEVPVPRPPVRPLILRRARPVARRGVAWRVHG 150
GARARALAVNYGVCGR ACPAAHGAALLLMvajIESLssBBAERDC 196
B
(pi: Probe
5' BstXl 3'
I 1. 5.1
i I. 5.4
I 1 10.1
I 1 10.4
I
__ l
200 bp

Figure 6.5. Amino acid sequence of OsDD3. (A) Putative ORF of 03003. The
signal sequence is double underlined and the potential transmembrane region

is boxed. Motifs found in OsDD3 are two potential protein kinase C

phosphorylation sites (single underlined) and one RGD motif (dark boxed).

Cysteine residues that may participate tertiary structure, in ligand binding, or in
dimerization via disulfide bridges are shown as outline letters. (B) Independent
clones isolated from an lM-specific cDNA library showing differences in 5’ ends
and addition of the polyadenylated tall at the 3’ end. The location of the ORF

and the region used as probe are indicated.

190

transfer into the microsomes, the signal sequence will be cleaved off. As
shown in Figure 6.6, the in-vitro translated control mRNA, encoding [3-
lactamase, was processed from 31.5 kD to 29 kD. The in-vitro translated
OsDD3 mRNA showed a product of expected size, which, in the presence of
microsomes, was processed from 20 kD to 17 kD. This indicated that a
functional signal sequence is present in OsDD3 and that OsDD3 enters the

secretory pathway.

6.5. DISCUSSION

A gene, OsDD3, whose transcript levels increased during GA treatment
of stem sections and during submergence of whole plants, was identified by
differential display. The difference in timing of the increase in transcript
levels between the two treatments was expected since the lag phase of GA-
induce growth is 40 min (Sauter and Kende, 1992), and that of submergence-
induced growth is 3 h 20 min (Rose-John and Kende, 1985). Although the
increase in transcript levels was only 3- to 4-fold and 6-fold in stem sections
and in submerged plants, respectively, this response was reproducible. The
early events leading to growth are more clearly separated time-wise in
submerged plants than in GA-treated stem sections (Lorbiecke and Sauter,
1998). Therefore, submerged plants were used to compare the increase in
transcript levels of OsDD3 to markers of the cell cycle. Changes in expression
of histone H3 and cchsI were shown before to correlate well with the S-
phase and the G2 to M-phase transition, respectively (Sauter et al., 1995; Van
der Knaap and Kende, 1995; Lorbieke and Sauter, 1998). The increase in

OsDD3 transcript level coincided with the first noticeable change in cell cycle

191

RNA B-lactamase OsDD3

Microsomes - + - +

 

M
66-

39-

26-

21-

14-

 

 

 

 

Figure 6.6. ln-vitro translation of 05003. Control RNA, encoding
B-Iactamase, and 03003 were in-vitro translated in the absence or
presence of canine pancreatic microsomal membranes as indicated
above the lanes. B-lactamase RNA was in-vitro translated in the
presence of [358]methionine and 05003 was in-vitro translated in
the presence of [3H]leucine.

192

activity, as was indicated by histone H3 expression. This indicates that the
transcript levels of OsDD3 increase early during submergence-induced
growth. However, this was not as early as that of Os-RPAI and OsDD4, whose
transcript levels increased 4 h after start of submergence (Van der Knaap et al.,

1997; Chapter 5).

The tissue-specific transcript accumulation indicated that OsDD3 is
expressed in most tissues examined but that its expression is highest in
actively growing regions, i.e., the basal part of the second youngest leaf sheath,
the youngest leaf, and the IM. Interestingly, the band pattern was smeary in
most tissues examined. It is unlikely that the RNA was partially degraded
since the signal for E37 was sharp. A more likely explanation is the size
difference in mRNA due to the difference in addition of the poly(A) tail, as
was indicated by the different clones isolated from the cDNA library. The
addition of the poly(A) tail may be under tissue-specific control since the lane
containing RNA isolated from the root tip showed a sharp band in contrast to
RNA isolated from other tissues. Whether the 5' end of OsDD3 is
heterogeneous as well cannot be concluded from the clones isolated from the

cDNA library, and this question has not been further investigated.

Southern blot analysis showed one band in all lanes. The size of the
smallest band on the Southern blot was 6.2 kb and was found in the lane
containing PstI-digested DNA. If the genomic region spanning OsDD3
contained no intron, it is possible that the fragments on the Southern blot
contained two linked copies of OsDD3. The presence of genes related to the
protein encoded by OsDD3 can not be determined since the probe used was

entirely derived from the 3' UTR.

193

No significant similarity was detected between the nucleotide sequence
of OsDD3 and the protein it encodes to entries in the DNA and protein
databases. One rice EST, D23940, has 25.8% amino acid identity and 37.4%
similarity to OsDD3. The significance of this similarity is unclear since the
available sequence was only 264 nucleotides long, and the amino acid
sequence alignment showed gaps between OsDD3 and amino acid sequence
derived from the rice EST. BLASTP searches in the non-redundant databases
identified potential homologs of low probability. They ranged from
membrane proteins to antifreeze proteins and transcriptional activators. One
protein with limited homology to OsDD3 is an allergen from Japanese cedar
pollen (Namba et al., 1994). The significance of that finding can be debated as
well since the sequence similarity is over a short region, and this region is not
present in the mature allergen protein.

Analysis by PSORT predicts that the OsDD3 protein is a type 1a
membrane protein that spans the plasma membrane and contains a short C-
terminal tail in the cytoplasm. The presence of a functional signal sequence
provides support for the plasma membrane localization of OsDD3. It may
function there as a receptor and signal transmitter between the cell wall
matrix and the cytoplasm. During plant growth, communication between the
cell wall and in the cytoplasm is likely to be essential. Interestingly, database
searches indicated some sequence similarity to CD8 01 chain. CD8 or and B are
glycoproteins expressed on T cells and contain an immunoglobulin-like
extracellular domain, a membrane spanning segment, and a short
cytoplasmic tail (Leahy, 1995). The extracellular domain recognizes the
appropriate major histocompatibility complex molecules while the

cytoplasmic tail interacts with a src-like tyrosine kinase, p561Ck. The

194

significance of this similarity is mostly structural since OsDD3 has no
similarity to the immunoglobulin-like extracellular domain, but only to the
region preceding and spanning the plasma membrane, as well as the C-
terminal region. Overall amino acid identity between CD8 or and OsDD3 is
22.6% while the similarity is 44.4%. Like CD8, OsDD3 may signal through
cytoplasmic kinases which interact with its C-terminal tail. The serines
located in the C-terminal domain of OsDD3 (see Fig. 6.5A) may be targets for
phophorylation.

The proposed extracellular domain of OsDD3 bears no sequence
similarity to any known or suspected receptors in either animals or plants. It
contains several cysteine residues, which may be involved in the tertiary
structure, in ligand binding, or in dimerization via disulfide bridges. Another
motif found in the presumed extracellular domain is the RGD motif found in
proteins mediating cell adhesion in animals (D'Souza et al., 1991). Fibronectin
is an extracellular glycoprotein and is the prototype cell adhesion molecule.
The minimum functional unit for cell adhesion is RGD, which is recognized
by receptors belonging to the integrin family of cell adhesion molecules. The
role of integrins is to integrate the extracellular matrix outside the cell with
the actin-containing cytoskeleton inside the cell. In plants, effects of peptides
containing the RGD motif have been observed (Schindler et al., 1989). In
soybean suspension cells, addition of these peptides led to enhanced growth
rates and aberrant cell wall - plasma membrane interactions. The relevance of
this is unknown. In plants, several proteins have been identified to contain
an RGD domain. However, the function of this domain in any of the plant
proteins is unknown.

Many extracellular plant proteins are glycosylated, and proline residues

may be hydroxylated (Carpita and Gibeaut, 1993). While no consensus N-

195

linked glycosylation site is present in OsDD3, the serine and potentially
hydroxylated proline residues may be glycosylated via O-linkage to their

hydroxyl group.

The in-vitro translation experiment in the presence or absence of
microsomes confirmed the existence of a functional signal sequence in
OsDD3. This indicated that OsDD3 enters the secretory pathway. Motifs
known to be transport or retrieval signals between compartments along the
vesicular transport pathway (Rothman and Wieland, 1996) were not found in
OsDD3. This strengthens the possibility that OsDD3 is plasma membrane
localized. However, before the role of OsDD3 in GA-regulated growth can be

elucidated, the precise location of OsDD3 in the cell needs to be addressed.

6.6. LITERATURE

Altschul, S.F., Warren, G., Gish, W., Myers, E.W. and Lipman, D.]. (1990) Basic
local alignment search tool. ] Mol Biol 215:403-410.

Ausubel, F.M., Brent, R., Kingston, RE, Moore, D.D., Seidman, J.G., Smith,
].A. and Struhl, K. (1987) Current Protocols in Molecular Biology (Wiley,
New York).

Bleecker, A.B., Schuette, ].L. and Kende, H. (1986) Anatomical analysis of

growth and developmental patterns in the internode of deepwater rice.
Planta 169:490-497.

D'Souza, S.E., Ginsberg, M.H. and Plow, E.F. (1991) Arginyl-glycyl-aspartic acid
(RGD): a cell adhesion motif. Trends Biochem Sci 16:246-250.

Carpita, NC. and Gibeaut, D.M. (1993) Structural models of primary cell walls
in flowering plants: consistency of molecular structure with the physical

properties of the walls during growth. Plant] 321-30.

Leahy, D.]. (1995) A structural View of CD4 and CD8. FASEB J 9:17-25.

196

Liang, P. and Pardee, A3. (1992) Differential display of eukaryotic messenger
RNA by means of the polymerase chain reaction. Science 257:967-971.

Lorbiecke, R. and Sauter, M. (1998) Induction of cell growth and cell division
in the intercalary meristem of submerged deepwater rice (Oryza sativa L.).
Planta 204:140-145.

Métraux, ].-P. and Kende, H. (1983) The role of ethylene in the growth
response of submerged deepwater rice. Plant Physiol 72:441-446.

Namba, M., Kurose, M., Torigoe, K., Hino, K., Taniguchi, Y., Fukuda, S., Usui,
M. and Kurimoto, M. (1994) Molecular cloning of the second major
allergen, Cry 11, from Japanese cedar pollen. FEBS Letters 353:124-128.

Puissant, C. and Houdebine, L.M. (1990) An improvement of the single-step
method of RNA isolation by acid guanidium thiocyanate-phenol-
chloroform extraction. BioTechniques 8:148-149.

Raskin, I. and Kende, H. (1984a) Role of gibberellin in the growth response of
submerged deepwater rice. Plant Physiol 76:947-950.

Raskin, I. and Kende, H. (1984b) Regulation of growth in stem sections of
deepwater rice. Planta 160:66-72.

Rose-John, S. and Kende, H. (1985) Short-term growth response of deepwater
rice to submergence and ethylene. Plant Sci 38:129-134.

Rothman, ].E. and Wieland, ET. (1996) Protein sorting by transport vesicles.
Science 272:227-234.

Sambrook, J., Fritsch, E.F. and Maniatis, T. (1989) Molecular Cloning: A
Laboratory Manual (Cold Spring Harbor Laboratory Press, Plainview, NY),
2nd ed.

Sauter, M. and Kende, H. (1992) Gibberellin-induced growth and regulation of
the cell division cycle in deepwater rice. Planta 188:362-368.

Sauter, M., Mekhedov, S.L. and Kende, H. (1995) Gibberellin promotes
histone H1 kinase activity and the expression of cdc2 and cyclin genes

during the induction of rapid growth in deepwater rice internodes. Plant]
7:623-632.

Sauter, M. (1997) Differential expression of a CAK (cdc2-activating kinase)-like

protein kinase, cyclins and cdc2 genes from rice during the cell cycle and in
response to gibberellin. Plant] 11:181-190.

197

Schindler, M., Meiners, S. and Cheresh, DA. (1989) RGD-dependent linkage
between plant cell wall and plasma membrane: consequences for growth. ]
Cell Biol 108:1955-1965.

Stiinzi, J.T. and Kende, H. (1989) Gas composition in the internal air spaces of
deepwater rice in relation to growth induced by submergence. Plant Cell
Physiol 30:49-56.

Van der Knaap, E. and Kende, H. (1995) Identification of a gibberellin-induced

gene in deepwater rice using differential display of mRNA. Plant Mol Biol
28:589-592.

Van der Knaap, E., Jagoueix, S. and Kende, H. (1997) Expression of an ortholog
of replication protein A1 (RPA1) is induced by gibberellin in deepwater
rice. Proc Natl Acad Sci USA 94:9979-9983.

Von Heijne, G. (1986) A new method for predicting signal sequence cleavage
sites. Nucleic Acids Res 14:4683-4690.

198

CONCLUSION

This thesis reports on the identification of five genes encoding histone
H3, Os-RPAI, OsTMK, OsDD3, and OsDD4, whose transcript levels increase in
response to gibberellin (GA) and to submergence in deepwater rice
internodes. In addition, several other GA-regulated genes in deepwater rice
have been identified in our laboratory. Broadly, the genes can be grouped in

three categories:

(i) Genes involved in the cell cycle. GA increases the rate at which new
cells are produced in the intercalary meristem. Therefore, an increase in
transcript levels for cell cycle-specific genes is expected. Histone H3 and Os-
RPAI were identified by differential display, while two cyclin genes, cchsI
and cchsZ, were identified by other means. Changes in the cell cycle, as
measured by flow cytometry and [3H]thymidine labeling, correlate well with
the changes in expression of histone H3, Os-RPAI and the two cyclin genes.
Throughout this thesis, we have used the change in transcript levels for
histone H3 and cchsl as markers for the different phases of the cell cycle and

compared their increase with that of the other genes identified.

(ii) Genes involved in cell wall property changes. Os-EXPZ and Os-EXP4
encode two expansin proteins, and their transcript levels are increased by GA.
Expansins were identified originally because of their cell wall loosening
property. During growth, the cell wall has to yield or loosen, and expansins
are thought to play a critical role in this process by disrupting the hydrogen

bonds between cellulose microfibrils and hemicelluloses.

199

(iii) Genes whose role in plant growth and development is unknown.
This thesis reports on the identification of three genes in this class: OsTMK,
encoding a leucine-rich receptor-like kinase; OsDDB, encoding a putative type
1a plasma membrane receptor; and OsDD4, encoding a putative transcription
factor. OsTMK transcript levels increase during GA treatment, and the
expression is high in growing regions, suggesting a role for this gene in plant
growth. A potential downstream component of OsTMK is a kinase-associated
protein phosphatase. The expression of OsDD4 is restricted to regions
containing meristems. Overexpression of OsDD4 in Arabidopsis leads to a

pleiotropic phenotype in which several growth related processes are affected.

Data shown in this thesis indicate that the transcript levels for Os-
RPA1 and OsDD4 increase before changes in cell cycle activity are observed.
This suggests that Os-RPAI and OsDD4 may be primary response genes. It will
be interesting to understand how GA regulates the increase in transcript
levels. If is this accomplished through an increase in transcription, the
promoters of these genes may help elucidate important components of the
GA signal transduction pathway. The transcription of Os-RPAI and OsDD4
may be regulated by GA1 and/ or RGA, two putative transcription factors in

the GA signaling pathway.

Studies to learn more about the function of the proteins encoded by
OsDD3 and OsDD4 will be important. These two genes are interesting since no
homologs have been identified and they may encode plant-specific genes.
Before the role for OsDD3 can be elucidated, the precise localization of this

protein in the cell needs to be addressed. For the partially nuclear-localized

200

OsDD4, it will be interesting to know whether it can bind DNA, activate
transcription, and which domains are responsible for those two activities. It
will be intriging to find out whether GA plays a role in the cell-specific
location of OsDD4 and whether GA can modulate DNA binding or
transcriptional regulation. If OsDD4 plays a role in GA signal transduction, it
is possible that it is a substrate for SPY, a protein containing a domain of N-
acetylglucosamine transferase activity. It has been hypothesized, for example,

that GAI and RGA are likely substrates of SPY.

The role of OsTMK, OsDD3, and OsDD4 in GA-regulated growth in the
whole plant should be addressed, at least in part, via experiments designed to
increase or decrease the transcript levels of the respective genes. For OsDD4,
the expression is relatively high in the intercalary meristem of growing
internodes and may be limiting in modern rice varieties that do not display
early internodal elongation. Overexpression of OsDD4 in modern rice
varieties may indicate whether the protein encoded by this gene plays a
critical role in internodal elongation. Alternatively, the function of these
genes can be addressed in a genetically more amenable system such as
Arabidopsis. The identification of the Arabidopsis orthologs and subsequent
antisense and overexpression experiments may allow a faster indication for

the role of these genes in plant growth and development.

201

   

"‘IIIIIIIIIIIIIIIIIII