PLACE IN RETURN Box to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

mu 0 6 2008
89 2‘) Ar

 

uvo

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJCIRC/Dateouopss-pJS

IDENTIFICATION AND CHARACTERIZATION OF NOVEL
LATE-NODULTN GENES FROM A MODEL LEGUME LOTUS
JA PONIC US

By

Philipp V. Kapranov

A DISSERTATION

Submitted to
Michigan State University
in partial fulﬁllment of the requirements

for the degree of

DOCTOR OF PHILOSOPHY

Program in Genetics

2000

 

ABSTRACT

IDENTIFICATION AND CHARACTERIZATION OF NOVEL
LATE-NODULIN GENES FROM A MODEL LEGUME LOTUS
JAPONIC US

By
Philipp Kapranov

The formation of nitrogen-ﬁxing nodules represents an unusual example of
externally induced organogenesis that unites the plant host and the symbiotic bacteria in a
microenvironment appropriate for the support of bacterial nitrogen ﬁxation and plant-
mediated assimilation of nitrogen. The process of nodule development can be subdivided
into early and late stages. To study the complexity of genes expressed during in a fully.
functioning, nitrogen-ﬁxing nodule, the mRNA expression proﬁle was examined in
determinate nodules of the diploid legume Lotus japonicus using the differential display
method. A range of novel expressed sequence tags (ESTs) associated with late
developmental events during Lotus japonicus nodule organogenesis was identiﬁed and
three genes, LjNODI6, LjNPPZCl and LjNOD 70 were characterized in detail. The LjPLP
gene family was identiﬁed based on nucleotide sequence homology to a cDNA,
LjNODI6, encoding L. japom'cus late nodulin, Nlj16. We show that LjPLP proteins

consist of an N-terminal domain that shares ~ 40% primary sequence identity with yeast

PITP, and is also capable of complementing a temperature-sensitive phenotype of a yeast
sec] 4 mutant strain, and a C-terminal domain consisting of either Nlj16, or highly related
amino-acid sequence. The Nlj16 C-terminal domain was found to be involved in
targeting PLPs to the cell plasma membrane. Furthermore, nodule-speciﬁc LjNODI6
mRNA is shown to be the result of an unusual transcriptional event, mediated by a
nodule-speciﬁc, bi-directional promoter present in an intron of a member of the LjPLP
gene family. The results obtained suggest that nodulin Nlj16 may exert a dominant
negative effect and function as a component of a mechanism that inactivates the function
of LjPLPs speciﬁcally in nodules. Expression of the LjNPP2C1 gene, encoding a protein
phosphatase type 2C enzyme, was found to increase signiﬁcantly in the mature L.
japom'cus nodules. Expression of the LjNPPZCI gene was found to be drastically altered
in speciﬁc L. japonicus lines carrying monogenic recessive mutations in symbiosis-
related loci, suggesting that the product of the LjNPP2C1 gene may function at both early
and late stages of nodule development. Two major mRNA species corresponding to the
LjNOD70 gene were induced in nodules and shown to be the result of a mechanism
resembling alternative splicing. The longer, presumably unspliced, mRNA species was
shown to contain a single open reading frame (ORF), encoding a polytopic hydrophobic
protein, LjN70, with a predicted molecular mass of 70 kD. The predicted amino-acid
sequence of nodulin LjN7O revealed structural features characteristic of transport
proteins, and was found to share similarity with the oxalate/formate exchange protein of
Oxalobacterformigenes. Therefore, the L. japonicus LjNOD70 gene family is postulated
to encode nodule-speciﬁc transport proteins, which may have evolved as a result of exon-

intron shuffling.

Copyright by
PHILIPP VALERIEVIC H KAPRANOV

2000

To my parents and my friends who believed in me

 

ACKNOWLEDGEMENTS

I want to thank my advisor Frans de Bruijn for giving me the opportunity to work in his
lab on several independent and often esoteric projects. I wish to extend my special
gratitude and appreciation, which is hard to express in words, to Krzysztof Szczyglowski,
who always gave me his friendship, support and guidance, who taught me a lot about
what I currently know about science and molecular biology, in particular, and without
whom this thesis would not have been possible. I want to thank my committee members
Lee McIntosh, Lee Kroos and Doug Gage for always ﬁnding time to discuss science and
life, for their encouragement and suggestions along the way. I owe a lot to all of the past
and present members of the de Bruijn lab, who made it a genuinely pleasant place to
work. I really enjoyed working with this group of people. Especially, I want to thank
Mary Ellen Davey for her support and numerous helpful advices about life and science;
David Silver and Susan Fujimoto, for those many discussions we had about everything;
Jodi Trzebiatowski, Chris Vriezen, Judith Wopereis and Ann Milcamps, without whom
the last few years would have been much less enjoyable; and everybody for being
fantastic people to work with. Thanks to the past and present members of PRL, especially
to Sridhar Venkataraman and Denis Maxwell, who made it a special place to work. To
Xiaoqiong Qin for introducing me to raquetball and sushi, and also helping us to solve a
nasty technical problem. To Ryan Bushey for excellent technical assistance with

generation of transgenic plants. To Marlene and Kurt for excellent artwork.

vi

PREFACE

The experiments in the Chapter 2 were initiated and guided by Dr. Krzysztof
Szczyglowski and represent a truly collective effort of a small team also including Dr.
Dirk Hamburger and the author of the thesis. As to the Chapters 3, 4 and 7, all of the
experiments were done by the author of the thesis.

In the chapter 5, the original differential display product was cloned by the author
of this thesis and the spliced cDNA clone 48-23 was isolated by Drs. Krzysztof
Szczyglowski and Dirk Hamburger. The author of the thesis also contributed the
following: the RT-PCR analysis; the discovery of the spliced, presumably functional,
form of LjNOD70 mRNA; the northern blot analysis of LjNOD 70 expression in the
different tissues; cloning of a related gene from L. japonicus and the generation of the
anti-LjN70 antibody. Dr. Krzysztof Szczyglowski determined the pattern of LjNOD70
expression during nodule developmental and the complexity of the gene family
represented by LjNOD70 in the L. japonicus genome. The experiments to localize the
LjN7O protein to the peribacteroid membrane in L. japonicus nodules were performed by
Jim Guenther and Dr. Dan Roberts, University of Tennessee, Knoxville, using the anti-
LjN7O antibodies generated by the author of the thesis. . The experiments in the Chapter

6 were performed equally by the author and Dr. Krzysztof Szczyglowski.

vii

TABLE OF CONTENTS

LIST OF TABLES ................................................................................. xi
LIST OF FIGURES ............................................................................... xii
CHAPTER 1
SYMBIOTIC NITROGEN F IXATION ........................................................... 1
1.1. Legumes .......................................................................................... 1
2. Nodule development and functioning ........................................................... 3
2.1. Early stages of nodules development ................................................ 3
2.1.1. Signal exchange between free-living rhizobia and
plant host ............................................................................. 3
2.1.2. Perception of the Nod factors and subsequent events ................ 10
2.1.3. Regulation of the early stages of nodule development ............... 12
2.2. Late stages of nodule development ................................................. 14
2.2.1. Formation of the mature nitrogen-ﬁxing nodule ...................... 14
2.2.2. Nitrogenase and nitrogen ﬁxation ....................................... 16
2.2.3. The oxygen paradox ....................................................... 17
2.2.4. Assimilation of ﬁxed nitrogen ............................................ 19
2.2.5. Carbon metabolism in nodules ........................................... 23
2.2.6. Synthesis and functioning of the peribacteroid membrane ........... 25
2.2.7. Regulation of functioning of the mature
nitrogen-ﬁxing nodule .................................................... 29
3. Molecular biology and classical genetics as tools to understand nodule biology ....... 31
3.1. Plant symbiotic mutants and model legumes ..................................... 31
3.2. Identiﬁcation and characterization of late nodulins .............................. 33
4. References ........................................................................................ 37
CHAPTER 2

CONSTRUCTION OF A LOT US JAPONIC US LATE NODULIN
EXPRESSED SEQUENCE TAG LIBRARY AND IDENTIFICATION OF

NOVEL NODULE-SPECIFIC GENES ........................................................ 49
2. 1. Abstract ........................................................................................ 49
2.2. Introduction .................................................................................... 50
2.3. Materials and Methods ....................................................................... 53
2.4. Results .......................................................................................... 59
2.5. Discussion ..................................................................................... 73
2.6. Acknowledgements ........................................................................... 91
2.7. Literature Cited ............................................................................... 92

viii

CHAPTER 3
A PROTEIN PHOSPHATASE 2C GENE, LjNPPZCI, FROM LOTUS

JAPONIC US INDUCED DURING ROOT NODULE DEVELOPMENT ............... 99
3. 1. Abstract ....................................................................................... 90
3.2. Introduction .................................................................................. 101

3.3. Materials and Methods ..................................................................... 103
3.4. Results ....................................................................................... 108
3.5. Discussion ................................................................................... 115
3.6. Acknowledgments .......................................................................... 128
3 .7. Literature .................................................................................... 1 29
CHAPTER 4

ANALYSIS OF THE BIOLOGICAL FUNCTIONS OF LjNPP2C1 ..................... 132
4.1. Abstract ...................................................................................... 132
4.2. Introduction ................................................................................. 133
4.3. Materials and Methods ..................................................................... 138
4.4. Results ....................................................................................... 145
4.5. Discussion ................................................................................... 150
4.6. References ................................................................................... 164
CHAPTER 5

THE LOT US JAPONIC US LjNOD70 NODULIN GENE ENCODES

A PROTEIN WITH SIMILARITIES TO TRANSPORTERS ............................ 168

5.1. Abstract ...................................................................................... 168

5.2. Introduction ................................................................................. 170
5.3. Materials and Methods ..................................................................... 173
5.4. Results ....................................................................................... 178
5.5. Discussion ................................................................................... 187
5.6. Acknowledgments .......................................................................... 200
5.7. References ................................................................................... 201

CHAPTER 6

NOVEL, HIGHLY EXPRESSED LATE NODULIN (1.1'NODI6) FROM

LOT US JAPONIC US ............................................................................ 205

6.1. Abstract ...................................................................................... 205

6.2. Introduction ................................................................................. 207
6.3. Materials and Methods ..................................................................... 210
6.4. Results ....................................................................................... 217
6.5. Discussion ................................................................................... 226
6.6. Acknowledgments .......................................................................... 244
6.7. References ................................................................................... 245

ix

CHAPTER 7
LOT US JAPONIC US PHOSPHATIDYLINOSITOL TRANSFER-LIKE
PROTEINS WITH A PLASMA MEMBRANE TARGETING DOMAIN

EXPRESSED SEPARATELY IN ROOT NODULES ..................................... 250
7.1. Abstract ...................................................................................... 250
7.2. Introduction ................................................................................. 251
7.3. Results ....................................................................................... 254
7.4. Discussion ................................................................................... 267
7.5. Materials and Methods ..................................................................... 274
7.6. References ................................................................................... 299
FUTURE PROSPECTIVES ................................................................... 303

Table 2.1.-

Table 2.2.-

Table 4.1.-

Table 7.1 -

LIST OF TABLES

Sequence similarities detected for the nodule specific LjN clones ....... 66
Sequence similarities detected for randomly sequenced EST clones. . . .71
Number of independent transgenic lines .................................... 145

The nucleotide sequences of the primers used during different
ampliﬁcation procedures ...................................................... 283

xi

Figure 1.1.-

Figure 1.2.-

Figure1.3.-

Figure 1.4.-

Figure 2.1.-

Figure 2.2.-
Figure 2.3.-

Figure 2.4.-

Figure 2.5.-

Figure 2.6.-

Figure 2.7.-
Figure 2.8.-

Figure 3.1.-

Figure 3.2.-
Figure 3.3.-

Figure 3.4.-

Figure 3.5.-

LIST OF FIGURES
Schematic representation of the early events taking place
duringthe symbioticinteractions.............................................7
A generic structure of the Nod factors ....................................... 9

Structure of the indetenninate- and determinate-type
legume nodules ................................................................... 15

Schematic representation of the major pathways of nitrogen
and carbon metabolism in nodules ............................................. 28

RNA gel blot analysis of L. japonicus LjEnod2 and

leghemoglobin (nglbl) gene expression ...................................... 78
Developmental mRNA differential display ................................... 80
Differential colony hybridization ............................................... 82

DNA/protein sequence alignments of LjN77, LjN13,LjN101,

and LjN3 ........................................................................... 84
Protein sequence alignments of LjNP450 and LjN65 ....................... 86
Amino-acid sequences comparison

olej21, Ag13, and PKIWISOI ................................................. 88
Developmental slot blot northern analysis ..................................... 90
RNA gel blot hybridization of LjN3 ............................................ 91
Amino acid sequence alignment of L. japom'cus LjNPP2C1 and

LjPP2C2 and A. thaliana ABIl proteins ..................................... 119
Northern blot analysis of LjNPPZC 1 and LjPPZC 2 expression ........... 120

Protein phosphatase type 2C activity of a GST-LjNPPZC] protein ...... 121

Complementation of the PP2C deﬁcient

yeast mutant strain TM126 (ptclA) ........................................... 123
Northern blot analysis of
developmental LjNPPZCI gene expression..................................124

xii

Figure 3.6.-

Figure 3.7.-

Figure 4.1.-

Figure 4.2.-

Figure 4.3.-

Figure 4.4.-

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 6.1.-

Northern blot analysis of LjNPP2C1 expression
in different tissues of L. japonicus ............................................ 125

Comparison of LjNPPZCI transcript level in different mutants of
L. japonicus ...................................................................... 127

Schematic representation of the T-DNA regions of the constructs
used for transformation of L. japonicus ...................................... 154

LBP interacts speciﬁcally with LjNPP2Cl in the yeast
two-hybrid system ............................................................... 156

The LBP cDNA encodes a protein with a weak
similarity to ginseng ribonuclease 2 .......................................... 159

Expression of LBP mRNA at different stages of nodule
development and in the roots of different Nod' mutant

lines of L. japonr'cus ............................................................ 162
RNA gel blot analysis of L. japonicus LjN—l8 EST ......................... 190
A schematic structure of L. japonicus cDNA 48-23 ........................ 190
Southern blot hybridization of the RT-PCR products... . . . . . . . . . . . . . . . ....191

Nucleotide sequence of genomic- and
RT-PCR derived LjN~I8 products ............................................. 192

Nucleotide and deduced amino-acid sequence
of the 48-23 cDNA containing the 166 bp insert sequence ................ 194

Alignment of the LjN7O protein sequence with the deduced

partial amino-acid sequence of Arabidopsis EST 126K 1 5,

and the oxalate/formate exchange protein

of Oxalobacterformigenes ..................................................... 196
Expression of the LjNOD70 gene in different L. japom'cus tissues ...... 197

Southern hybridization analysis of L. japonicus ecotypes Gifu and

F unakura using the LjNOD 70 gene as probe ................................ 198
Western blot of the isolated peribacteroid membrane fraction ............ 199
Northern blot analysis of PCR5 expression ................................. 231

xiii

Figure 6.2.-

Figure 6.3.-

Figure 6.4.-
Figure 6.5.-

Figure 6.6.-
Figure 6.7.-
Figure 6.8.-
Figure 6.9.-

Figure 6.10.-

Figure 6.11.-

Figure 7.1.-
Figure 7.2.-
Figure 7.3.-

Figure 7.4.-
Figure 7.5.-
Figure 7.6-

Nucleotide and deduced amino acid sequences

of the LjNOD16 cDNA ......................................................... 233
Alignment of Nlj 16 with the deduced aminoacid sequences of

Arabidopsis ESTs 168K8 and llOGl6 ....................................... 235
Expression of LjNOD16 in different tissues of L. japonicus .............. 236
Expression pattern of Arabidopsis EST 168K8 ............................. 236

Organization of the LjNOD16 gene in the genomes
of L. japonicus and other legume species .................................... 238

In situ localization of LjNOD] 6 and leghemoglobin transcripts
in sections of 21-d-old L. japonicus nodules ................................. 240

Expression of putative homologue(s) of LjNOD16 in
different tissues of M. sativa .................................................... 241

Detection of Nlj 16 protein and its putative alfalfa
homologue in nodules of L. japopm'cus and M sativa ...................... 242

Expression of Nlj 16 in different tissues of L. japonicus ................... 242

The levels of Nle6 protein in the nodules from

the different pCRSAnti transgenic lines of L. japonicus ................... 243
Schematic diagrams of LjPLP cDNAs .................................. .....285
Tissue-speciﬁc expression of the LjPLP-l and LjPLP-II genes ............ 286
Tissue-speciﬁc expression of the LjPLP-IV transcripts .................... 287

Schematic diagrams of LjPLP-IV genomic region,
and LjPLP-IV sense- and LjNODI6 transcripts .............................. 288

Amino acid sequence alignment between L. japonicus LjPLP-IV,
nodulin Nlj16, Arabidopsis AtPLP and yeast secl4p protein ............. 290

The nucleotide sequence of the introon-localized
bi-directional promoter of the LjPLP-IV gene .............................. 291

The internal bi-directional promoter of the LjPLP-IV gene

directs the expression of the GUS reporter gene to the central
zone of L. corniculatus nodules ............................................... 293

xiv

Figure 7.8- Subcellular localization of the mGFPS-Nlj 16 fusions in the
onion epidermal cells .......................................................... 295

Figure 7.9- Amino acid sequence alignment of the C-terminal N1jl6-like

Domains of LjPLP II-IV, and Arabidopsis AtPLP, proteins ............. 296
Figure 7.10- The LjPLP proteins complement the temperature-sensitive
phenotype of yeast secH mutant ............................................. 298
xv

 

CHAPTERl

SYMBIOTIC NITROGEN FIXATION

1.1. Legumes.

Plants are an important source of nourishment for the world’s rapidly growing
population. Plant growth and development depends on the availability of nitrogen and
other nutrient compounds in the soil, which are rapidly depleted in modern agricultural
practices. After photosynthesis, the acquisition of nitrogen is considered to be the second
most important process for plant growth (Vance, 1998). To replenish soil nitrogen
content, chemical fertilizers are extensively used. However, this approach suffers from
serious drawbacks, including the large amount of energy consumed during fertilizer
production and its severe negative impact on the environment, manifested by nitrate
contamination of the groundwater. Biological nitrogen ﬁxation by free-living and
symbiotic prokaryotic organisms contributes approximately 120 million tons of ﬁxed
nitrogen per year to the biosphere, almost twice the amount produced by nitrogen
fertilizer industry (Vance, 1998). The ability to fix nitrogen is limited to prokaryotes.
HoWever, a number of families of ﬂowering plants have developed the ability to form a
symbiotic relationship with nitrogen-ﬁxing soil microorganisms. The latter belong to two
distinct groups, gram-negative bacteria of the genera Rhizobium, Bradyrhizobium,
MeZOrhizobr'um, Azorhizobium and Sinorhizobium (collectively referred to as rhizobia)

and gram-positive bacteria of the genus Frankia. The bacterial partner ﬁxes dinitrogen by

converting it to ammonia, which is utilized by the plant host for growth. In retum, the
plant host provides reduced carbon products as a source of energy for the bacterial
endosymbiont and provides an ecological niche.
The legume-Rhizobia symbiosis is the most prominent example of this
relationship, and the most members of this plant family are capable of establishing
symbiotic interactions (Gualtieri and Bisseling, 2000). This symbiosis also has the highest
economical importance, since it has been estimated that that 80% of biologically ﬁxed
nitrogen is generated by the legume-Rhizobia symbiosis (Vance, 1998). Therefore it is not
surprising, that this symbiosis has been the subject of scientiﬁc investigations since the
late 19th century, when Hellriegel and Willfarth conclusively showed that root nodules
allow the assimilation of nitrogen from the air (Hellriegel and Willfarth, 1888). Since then,
the physiological, biochemical and, in past two decades, the molecular aspects of our
understanding of the legume-rhizobia symbiosis have greatly advanced (reviewed in
Gualtieri and Bisseling, 2000; Kaminski et al., 1998; Mylona et al., 1995; Pueppke, 1996;
Hirsch, 1992; Nap and Bisseling, 1990). In particular, signiﬁcant progress has been made
in the development of the molecular genetics of the microsymbiont. Many of the rhizobia]
genes and the biochemical pathways required for the symbiosis have been identiﬁed and
well characterized, in particular the nature of the primary rhizobia] signaling molecules,
the NOd factors have been elucidated in detail (reviewed in Schultze and Kondorosi, 1998;
I)“eppke, 1996; Long, 1996). Moreover, the symbiotic plasmid of the broad-host range
symbiont Rhizobium sp. NGR234, containing many of the genes required for symbiosis,

has been completely sequenced (F reiberg et al., 1997), the sequencing of the entire genome

of Rhizobium meliloti is well advanced (http://cmgmstanford.edu/~mbamett/genome.htm),
and the entire DNA sequence of the symbiosis island of Mesorhr'zobium loti has been
completed (Frans de Bruijn, personal communication). This information is ushering in a
new era in systematic genome-wide analyses of the intricacies of the Rhizobr’al biology,
both in the free-living and symbiotic stages (for example see Perret et al., 1999).

On the other hand, the understanding of the molecular basis of the symbiotic
interactions from the plant side is lagging far behind. Despite the availability of a variety
of plant mutants defective in various stages of symbiosis, only a single gene underlying a
symbiotic phenotype has been cloned thus far (Schauser et al., 1999). However, a vast
number of plant genes, speciﬁcally induced during the interaction with rhizobia have been
identiﬁed, but the biochemical and biological function of many of these genes is still
unknown (see below). Below, I will ﬁrst summarize what is currently known about the

physiology, biochemistry and (molecular) genetics of the legume-rhizobia symbiosis.
2. Nodule development and functioning

2-1- Early stages of nodule development

2°14. Signal exchange between free-living rhizobia and the plant host. Symbiotic
nitrogen ﬁxation represents a unique example of interactions between two very distant
SPeﬁies, and takes place in specialized organ-like structures called root- or stem nodules.
The development of this organ is a complex multistep process which can be subdivided

““0 early and late stages (Sprent, 1989). The early stages of nodule development start

with mutual recognition by the symbiotic partners. This, in turn, sets in motion a cascade
of events leading to the initiation of root cortical cell divisions eventually giving rise to a
nodule primordium. Subsequently, the cells of the nodule primordium are infected and
colonized by rhizobia. At this stage, the nitrogen ﬁxation commences and nodule
development enters into its late stage.

Different stages of nodule organogenesis are hallmarked by the expression of
speciﬁc sets of plant genes, termed nodulin genes, to reﬂect their speciﬁc association with
the processes of nodule development and/or ﬁrnctioning (van Kammen, 1984). Nodulin
genes have been traditionally categorized as "early or late", depending on the time point of
their induction (Nap and Bisseling, 1990). Here I will present a brief overview of the
events taking place during early stages of nodule development. For comprehensive
reviews on this topic see Gualtieri and Bisseling, (2000); Schultze and Kondorosi, (1998);
Geurts and Franssen, (1996); Long, (1996); Mylona et a1, (1995); Hirsh, (1992); Sprent,
(1989).

Nodule morphogenesis is normally initiated by the mutual recognition of speciﬁc
Signal molecules produced by both plant and bacterial symbiotic partners (Fig. 1.1;
Pueppke, 1996; Long, 1996; Geurts and Franssen, 1996). Flavonoid compounds secreted
by the plant host induce in rhizobia the synthesis of a speciﬁc set of morphogenic
liPOChitooligosaccharide molecules, the Nod factors (Fig. 1.1; Spaink, 1992; Mylona et al.,
1995; Long, 1996; Pueppke, 1996). Nod factor molecules, in turn, trigger a cascade of
events resulting in penetration of the plant root by rhizobia and the initiation of nodule

organogenesis (Fig. 1.1; Mylona et al., 1995; Geurts and Franssen, 1996). A typical Nod

factor resembles a short chitin molecule and consists of 3-6 moieties of B-1,4-N-

acetylglucosamine, with a fatty acid linked to the non-reducing end (Fig. 1.2; Schultze and
Kondorosi, 1998; Pueppke, 1996). Different rhizobia species produce one or more
distinct Nod factors, which differ primarily in the spectrum of side groups attached to the
chitin backbone (Fig. 1.2; Pueppke, 1996). The host range of a bacterial endosymbiont is
primarily determined by the repertoire of the Nod factors it can produce, due to the high
speciﬁcity of recognition of a particular Nod factor by a particular plant symbiotic
partner (Spaink, 1992; Spaink and Lugtenberg, 1994; Long, 1996; Pueppke, 1996; Niebel
et al., 1999). Exogeneous application in the absence of rhizobia of the puriﬁed Nod
factors at picomolar concentrations initiates several cellular responses characteristic of
early stages of symbiotic interactions. In addition, it can lead to the induction of
expression of certain nodulin genes and, in some cases, even cause the formation of
nodule-like structures (Spaink et al., 1991; van Brussel et al., 1992; Spaink and

Lugtenberg, 1994; Long, 1996).

Figure 1.1. Schematic representation of the early events taking place during the
symbiotic interactions. Plant cells secrete flavonoid compounds which in turn
stimulate production in rhizobia of speciﬁc signaling compounds, the Nod factors.
The latter interact with yet unidentiﬁed plant receptor and set in a motion a
cascade of cellular events depicted on the diagram which eventually lead to the

formation of nodule primordium. From Hirsch, (1992).

    

 

3

mﬁmcﬂccra:

 

 

+
355%:332
323.830

zeta—.28

(Imp

.
:o :z o: :z o: :z
o o
o

\ :oazo
znomo

re the

it

  

 

 

«

 

 

60ho=< .
‘ l

H
t < £95. H H

 

cor—«Enema:
has :5“ .

I
nzw
10
38am .5 Z 2%
.
0-:
.5 .3..
OO

.
s
I :2 O:
0 ago:
zoazo

:ofu

ﬁOU—uﬁﬂm awn—QM §Q=/

wesemﬁ SSSNSQ

tum
.ctors.
tion a
to the

whoa—:3:
Am—umc:c>a=v

 

F igurel .2. A generic structure of the Nod factors. A typical Nod factor molecule is
based on a oligomeric backbone of 3-6 B-l,4-linked N-acetylglucosamine moieties
containing a spectrum of substitutions at the reducing and non-reducing ends. The

substituting groups identiﬁed in the Nod factors produced by various species of

rhizobia are shown. From Long, (1996).

 

alecule ‘3
. moieties

H
Acetyl

Carbamoyl

O

l“ ' °
Carbamoyll/

16C
20C
(varied

 

Lunsaturation)

\

/N

Acyl group: 'i Methyl]
18C

 

oi

\
H

 

Non-reducing end

HO

OH

 

H 1
Acetate
Sulfate
F ucose
Methylfucose
Sulfo-methylfucose
Acetyl-methylfucose
D-Arabinose

- \ -

 

 

 

H
Glycerol

Reducing end

2.1.2. Perception of the Nod factors and subsequent events. The interactions

of Nod factors with yet hypothetical plant receptor(s) lead to a number of rapid
responses in the host plants, such as the depolarization of the membrane of root
epi dermis cells (Felle et al., 1995; Kurkdjian, 1995), rapid intracellular alkalization (Felle
et al-, 1996), calcium spiking in root hairs (Ehrhardt et al., 1996), and root hair
deformation and curling (Heidstra et al., 1994). There is mounting evidence that signiﬁcant
rearrangements of the root hair cytoskeleton occur within minutes after Nod factor
application (Crdenas et al., 1998). The Nod factors signal transduction pathway(s) in
planta and the molecular mechanisms of their recognition by plant are presently
unknown. However, evidence for the involvement of G-proteins in the Nod factor
Signaling cascades has recently been presented based on pharmacological studies (Pingret
et al-, 1999). In addition, the early nodulin gene rip], which encodes a peroxidase, is
rapidly but transiently induced soon aﬁer plant-Rhizobium interaction, suggesting that
oXidative processes may take place during early stages of symbiotic interactions (Cook et
al., 1995). Root hair curling results in the formation of a structure referred to as a
“Shepherd hook”, which physically entraps the bacteria. Local hydrolysis of the plant
cell wall occurs at the region of rhizobia] attachment to the root hair curl, and invagination

0f the plant plasma membrane allows the bacteria to penetrate the cells of the root
epidermis (Van Spronsen et al., 1994). New plant cell wall material is deposited around
the bacteria as it enters the plant cell, leading to the formation of a tubular structure, the
infecttion thread, in which the bacteria divide and migrate to the nodule primordium

10cated in the root cortex (Kijne, 1992). Two well studied early nodulins, ENOD5 and

10

 

 

ENOD12, are expressed during this process. ENOD5 transcripts are present in cells
containing growing infection threads (Scheres, 1990a). ENOD12 is expressed in root hairs
and root cells which contain growing infection threads, as well as in cells positioned
several cell layers ahead of the growing infection threads (Scheres, 1990b). Both these
nodulin genes encode proline-rich proteins, which are thought to be components of the
infection thread, and, in case of the ENOD12 protein, the cell wall of cortical cells located
in path of the extending infection thread (Schultze et al., 1994). However, it appears that
ENOD12 is not required for successful nodulation in at least some alfalfa varieties
(Csanadi etal., 1994).

Concomitant with root hair deformation and infection thread formation, fully
differentiated cells of the root cortex are induced to divide (Nap and Bisseling, 1990;
Verma, 1992; Mylona et al., 1995; Hirsch, 1992). Generally, temperate legumes such as
alfalfa, pea, and clover will form indeterminate type of nodules, which are characterized
by the presence of a persistent apical meristem (Newcomb et al., 1979). In contrast,
tropical legumes such as soybean, Sesbam’a rostrala, and Lotus japonicus develop
determinate type of nodules, which do not have meristematic activity but are rather
assumed to grow by a cell expansion mechanism (Newcomb et al., 1979). The position of
the root cortical cells which are induced to divide ﬁrst depends on the type of nodules a
particular legume species forms (Hirsch, 1992). In the case of indeterminate nodules,
these are the cells of the inner cortex, while the cells of the outer cortex initiate the
formation of the determinate type of nodules. Division of root cortical cells results in the

formation of a rudimentary nodule structure, the nodule primordium, which, as pointed
ll

—

above, can also be induced by the application of puriﬁed Nod factors (Spaink et al., 1991;
Spaink and Lugtenberg, 1994).

2.1.3. Regulation of the early stages of nodule development. The program of
nodule development is determined primarily by the plant host. The most convincing
evidence for it comes from the existence of a plethora of plant mutants affected at
different stages of nodule development. The mutant phenotypes range from a total lack of
nodules (non-nodulation; Nod"), to the formation of ineffective nodules (NodiFix') with
varying amounts of residual nitrogenase activity, to the production of excessive amounts
of nodules (supemodulation; Nod”, reviewed in Caetano-Anolles and Gresshoff, 1991;
also see Szczyglowski et al., 1998; Schauser et al., 1998). Most of the genes underlying
the symbiotic phenotypes in these plant mutants have not yet been identiﬁed. In fact, the
ﬁrst and, so far, the only plant gene in which a mutation causes a Nod' phenotype has
been cloned from Lotus japonicus nin mutant (Schauser et. al., 1999). Nin mutant plants
do not show the formation of infection threads or cortical cell divisions, however they do
undergo root hair curling and deformation, suggesting that at least some elements of the
Nod factor perception are functional in this mutant line (Schauser et. al., 1999). The
protein product of the nin gene has regional similarity to transcription factors and a
predicted DNA binding/ dimerization domain, suggesting that the nin protein may regulate
the expression of genes involved in early stages of the symbiotic interaction (Schauser et.
al., 1999). Interestingly, the nin mRNA is consitutively expressed in L. japonicus roots

and increases in mature nitrogen-ﬁxing nodules (Schauser et. al., 1999).

12

Not surprisingly, increasing evidence suggests that plant hormones play
signiﬁcant roles in regulating various aspects of the symbiotic interactions (Hirsch and
Fang, 1994). Application of exogenous auxin and cytokinin induces root cortical cell
divisions (Libbenga et al., 1973). Treatment of alfalfa with auxin transport inhibitors

(NPA or TIBA), induces the formation of nodule-like structures or “pseudonodules”, in
which early nodulin genes, such as ENODZ and Nms 30, are expressed (Hirsch et al.,
198 9). Exogenous cytokinin has been shown to induce expression of the early nodulin
gene ENODZ (Dehio and de Druijn, 1992) and ENOD40 (Fang and Hirsch, 1998). A role
of ethylene as an inhibitor of nodule formation has been extensively documented at least
in some legumes. The application of exogenous ethylene, or its precursor 1-
aminocyclopropane-l-carboxylate (ACC), has been shown to be detrimental to nodule
folTnation in pea (Lee and Larue, 1992). On the other hand, inhibitors of ethylene
Synthesis, such as aminoethoxyvinylglycine (AVG), have been shown to have a
stimulatory effect on nodule formation in alfalfa (Peters and Crist-Estes., 1989). A direct
genetic link between ethylene signaling and regulation of nodulation has been established
by the isolation of the sickle mutant of Medicago truncatula, which forms excessive
numher of nodules comparing to wild-type plants and also exhibits insensitivity to
ethylene (Penmetsa and Cook, 1997). However, the role and/or mechanisms of action of
ethylene during the symbiotic interactions may vary among different legume species,
Since an ethylene insensitive soybean mutant has been shown to have a normal number of

nodllles (Schmidt et al., 1999). Ethylene has also been implicated in providing positional

l3

information for the formation of the nodule primordia opposite of protoxylem poles

(Heidstra et al., 1997).

2.2 Late stages of nodule development

2.2.1. Formation of the mature nitrogen-fixing nodule. The late stages of nodule
development commence with the release of bacteria from the infection thread into the cells
of the central region of nodule primordium, via a process resembling endocytosis (Basset
et al., 1977). The bacterial cells replicate and enlarge, ﬁlling up most of the cytoplasm of
the infected cells, and subsequently differentiate into a distinct form, called the bacteroid.
A fraction of the cells in the central region in the nodule primordium remain uninfected
and are typically much smaller than infected cells. Infected and uninfected cells form the
central tissue of nodule, and are surrounded by peripheral tissues. The organization of the
latter varies among the legumes. In the case of indeterminate-type nodules, closest to the
central zone usually lies the nodule vascular system fully embedded in nodule
parenchyma, followed by cell layers of the nodule endodermis and cortex. In the
determinate type of nodules, outside of the parenchyma, cell layers of the sclerenchyma,
nodule endodermis, nodule cortex and periderm are found to be located (Hirsch, 1992).
Nodule peripheral tissues are believed to constitute a barrier for oxygen diffusion (see
below). In indeterminate-type nodules, several developmental zones can be distinguished
in the central tissue, from the nodule apex to the root. Zone I, the meristematic zone, is

located in the apex of the nodule and it contains actively dividing cells which give rise to

14

the remainder of cells. The next zone is invasion zone II, where plant cells are infected by
rhizobia. The N; ﬁxation commences in the adjacent region, interzone II-III, and continues
throughout zone III, the nitrogen ﬁxation zone. Zone IV is the senescence zone where the
cytoplasm of plant cells and bacteroids are degraded. Determinate type of nodules lack a
persistent meristem and, at any given point, the plant cells within a single nodule of this
type are at more or less similar developmental stage. The comparison of the structures of

the determinate- and indeterminate-type nodules is shown in Figure 1.3.

 

 

 

 

Figure 1.3. Structure of the indeterminate- and determinate-type legume nodules. (a)

Structure of indeterminate-type nodule. This type of nodules contains 5 distinctive zones

15

in the central region: zone I-the meristematic zone; zone H-the pre-ﬁxation zone; zone 11-
III- the interzone; zone HI-the nitrogen ﬁxation zone and zone IV-the senescence zone.
(b) Structure of determinate-type nodule. The central region contains cells in a similar
stage of development. Adapted from Hadri et al., (1998).

Bacteria released into plant cells undergo a multistage process of "terrnina "
differentiation ultimately resulting in the formation of their nitrogen-ﬁxing forms, called
bacteroids. The process of differentiation is accompanied by drastic changes in bacterial
morphology and physiology, and has been extensively reviewed by Kaminski et al.,
(1998) and Kahn et al., (1998). Essentially, the metabolism and overall physiology of
bacteroids is adapted to ﬁxing and providing nitrogen for the plant host and utilizing the
carbon compounds provided by the host in a low-oxygen environment.

2.2.2. Nitrogenase and nitrogen ﬁxation. The actual process of symbiotic
nitrogen ﬁxation is catalyzed by a multimeric rhizobial enzyme, called nitrogenase, and

can be summarized as follows:

N2 + 8H“ +8e' + 16Mg-ATP —> 2NH3 + H2 + 16Mg-ADP + 16pI

A typical nitrogenase enzyme complex is composed of two protein components, usually
referred to as the Fe protein and F eMo proteins (reviewed in Howard and Rees, 1996;
Burgess and Lowe, 1996). The Fe protein is a homodimer, containing one Fe4S4 cluster
and two ATP binding sites along the interface of the two subunits, which are encoded by

rhizobial niﬂ-I gene. The FeMo protein is the site of actual nitrogen reduction. It consists

l6

of two subunits (or and B), encoded by the bacterial m‘f D and nifK genes, which interact

with a single homodimer molecule of the Fe protein. The FeMo protein contains two
novel types of FeS clusters: a P cluster and a FeMo cluster, containing molybdenum in a
traditional nitrogenase: in some other variants the Mo atom is replaced with either Fe or V
(Eady, 1996). The Fe protein receives electrons from an electron donor, such as
ferredoxin, and passes them to the F eMo component in a process requiring the hydrolysis
of 2 ATP molecules per single transferred electron (Seefeld and Dean, 1997). The crystal
Structure of a complex between the Fe and FeMo proteins with bound ADP-A1F4' (a
compound resembling a possible intermediate of ATP hydrolysis) has been solved
(Schindelin et al., 1997). The association of Fe and FeMo proteins requires the binding of
2 ATP molecules by the Fe protein. Nucleotide binding induces drastic conformational
Changes in the Fe protein, allowing it to interact with and transfer electrons to the FeMo
Protein (Schindelin et al., 1997). These changes are reversed by the hydrolysis of ATP by
the Fe-FeMo protein complex, since Fe protein alone can not hydrolyze ATP. Nucleotide
hydrolysis is required for the dissociation of the two parts of the nitrogenase complex. X-
ray crystallographic data suggest that the binding and hydrolysis of ATP by the Fe
protein act as a molecular switch in a mode similar to that of GTP-binding proteins, such
as Ras (Schindelin et al., 1997).

2.2.3. The oxygen paradox. A unique and perplexing feature of nitrogenase is its
high sensitivity to molecular oxygen: contact with 02 irreversibly denatures the enzyme.

This property of nitrogenase conﬂicts with the high energetic cost of nitrogen ﬁxation.

l7

The latter requires a high rate of bacteroid respiration which, in turn, relies on a constant
supply of oxygen. The oxygen dilemma is solved by several means. 1. The concentration
of oxygen in the central zone within a nodule is maintained at a very low level, 5-30 nM,

comparing to the equilibrium 0; concentration in aerobic cells (230 uM). Moreover, the

existence of an oxygen diffusion barrier has been demonstrated by directly measuring the
oxygen concentrations inside different regions of a nodule. The concentration of oxygen
decreases signiﬁcantly in the peripheral tissues, suggesting that these tissues provide a
major obstacle for oxygen diffusion (Tjepkema and Yokum, 1974). 2. Specialized plant
proteins, leghemoglobins, bind and transport oxygen to the respiring bacteroids,
decreasing the free 02 concentration in the cytoplasm (Appleby, 1984). Leghemoglobins
are late nodulins that constitute the most abundant group of proteins in nodules (Brisson
et al-, 1982). 3. The rate of bacterial respiration is maintained at a high rate to maximize
the Consumption of oxygen. Specialized bacterial terminal oxidases, with a high afﬁnity
for OXygen, are induced during the nitrogen ﬁxation process (reviewed in Kaminski et al.,
1998). In summary, free oxygen does pose a serious problem for nitrogen ﬁxation but
both symbiotic partners have evolved a set of complex mechanisms to deal with the
problem.

In this respect, it is quite interesting that a free-living nitrogen ﬁxing bacteria,
S("emomyces thermoautotrophicus UBTI, has been recently discovered which contains a
nil:I‘ogenase which is totally insensitive to molecular oxygen (Ribbe et al., 1997).

IVIOreover, this nitrogenase requires superoxide anion radicals (02") as a source of

18

 

electrons for nitrogen ﬁxation and therefore depends on 02" and 02 (Ribbe et al., 1997).
Another unusual properties of the S. thermoautotrophicus enzyme is a lower MgATP
requirement, only 4-12 MgATPs per molecule of N2 ﬁxed, comparing to 16 MgATPs in
case of typical nitrogenases. Moreover, it is unable to reduce acetylene (Ribbe et al.,
1997). The full amino acid sequence of any of the protein components of this novel
nitrogenase is not yet available in the literature. This information together with a detailed
molecular and structural characterization of this enzyme will be invaluable for a better
understanding of the enzymology of nitrogen ﬁxation and perhaps, for engineering more
efficient and less oxygen sensitive nitrogenases in rhizobia and economically important
free-living nitrogen ﬁxing microorganisms. However, by far the most exciting experiment
Would be to express this or a similar nitrogenase enzyme in plants This would circumvent
one of the most severe limitations of expressing a typical nitrogenase in plants: the
sensitivity to oxygen, which is ubiquitous in plant tissues. If successful, this attempt may
allow plants to ﬁx free nitrogen, which has always been a major long-term goal of
nitrogen-ﬁxation research.

2.2.4. Assimilation of fixed nitrogen. It has been long believed that bacteroids
pro\Iide ﬁxed nitrogen to the plant in the form of ammonia (U dvardi and Day, 1997). First
the ﬁxed nitrogen in the form of NH3 diffuses passively through the bacterial membranes
into the peribacteroid space (PBS), where it is converted to NH4“, due to the relatively
low pH of this compartment. The ammonium ion is then transported through the
pel‘ibacteroid membrane (PBM) via a channel or a transporter. A recently identiﬁed

motrovalent cation channel GmSATl from soybean nodules is located in the FEM and is

19

thought to be involved in the transport of NH{ from the PBS into the plant cytosol
(Kaiser et al., 1998). However, recent evidence suggests that alanine, not ammonia, is
transported across bacterial membranes and the PBM in soybean nodules, since alanine
was found to be the major 15N-labeled compound excreted by the puriﬁed soybean
bacteroids during incubation in 15N atmosphere in the presence of D,L-malate (Waters et
al., I 998). The reason for the discrepancies within the previously mentioned studies was
the inclusion of a sucrose density gradient puriﬁcation step by Waters et al., (1998)
which removed the contaminating enzymes of the plant cytosol from the bacteroid
fraction. Addition of plant cytosolic fraction to the bacteroids resulted in a quick
PFOduction of ammonia from alanine (Waters et al., 1998). Incorporation of ammonia in
alanine is most likely mediated by bacterial alanine dehydrogenase, whose activity
irlcreases in the symbiotic state in contrast to the other rhizobial enzymes involved in
ammonia assimilation (Waters et a], 1998; Werner et al., 1980)

The ﬁrst step in the utilization of the ﬁxed nitrogen by plant is the synthesis of
glutamate via the glutamine synthetase/glutamate synthetase (GS/GOGAT) cycle (Fig.
1‘4; Lam et al., 1996). Glutamate is subsequently used for the synthesis of nitrogen
”mISport compounds which, depending on the species of legumes, can be either amides,
predominantly asparagine, or ureides, mostly allantoin or allantoic acid (Fig. 1.4; Venna
and Fortin, 1989). Most, if not all, key plant enzymes involved in nitrogen assimilation in
nodules have been puriﬁed and the corresponding cDNAs and genes have been cloned and
characterized. Nodule speciﬁc GS isoforms has been isolated from a number of legumes

(Bennett et al., 1989; Boron et al., 1989; Boron and Legocki, 1993; Stanford et al., 1993;

20

Temple et al., 1995; Forde et al., 1989). The GOGAT from alfalfa nodules has been
extensively investigated (Gregerson et al., 1993; Vance et al., 1995). The predominant
font) of GOGAT in this tissue is NADH-GOGAT, whose activity increases
substantially during nodule development in parallel with an increase in its mRNA and
protein level (Gregerson et al., 1993; Vance et al., 1995). Plants contain at least one more
GOGAT, the F d-GOGAT, which is most abundant in green tissues, however its activity,
mRN A and protein levels in nodules are undetectable or low comparing to that of
NADH-GOGAT (Vance et al., 1995 and references therein). NADH-GOGAT activity,
mRNA and protein levels are signiﬁcantly downregulated in mutant ineffective nodules
which do not ﬁx nitrogen (Gregerson et al., 1993; Vance et al., 1995). NADH-GOGAT
cOntajns a targeting peptide, however it is not clear to which organelle the protein is
ttargeted (Vance et al., 1995).

The nitrogen from glutamate is incorporated into aspartate and asparagine via the
conSecutive action of aspartate aminotransferase (AAT) and asparagine synthase (AS;
Fig. 1.4; Lam et al., 1995). Asparagine is the major nitrogen transport compound in many
plant species (Lea et al., 1990), and the ﬁnal product of nitrogen ﬁxation in temperate
legume, such as alfalfa (Vance, 1990). Alfalfa nodule AAT and AS have been puriﬁed
and extensively characterized at the molecular level (Gantt et al., 1992; Shi et al., 1997;
VEl-nce and Gantt, 1992; Vance et al., 1994 and references therein). Alfalfa contains two
AAT isoforms: AAT-1 appears to be constitutive while the expression of AAT-2
is0forrn is enhanced 15-20 folds in the mature nodules comparing to other tissues (Gantt

et al., 1992). Analysis of the AAT-2 amino acid sequence and immunogold labeling

21

suggest that it is localized to amyloplast (Robinson et al., 1994). AS is dramatically
induced on the level of mRNA and protein in mature nitrogen-ﬁxing nodules of alfalfa (Shi
et al -, 1997). AS appears to be encoded by a single gene in alfalfa and the corresponding
gene is also induced in leaves, after a dark treatment (Shi et al., 1997). In alfalfa, the
transcripts of NADH-GOGAT, AAT, AS and phosphoenolpyruvate carboxilase (PEPC,
see below) have been detected by in situ hybridization in both infected and uninfected
cells of the alfalfa nodules (Vance et al., 1995; Robinson et al., 1994; Shi et al., 1997).
This suggests that uninfected cells harbor the necessary enzymes for the incorporation of
ﬁxed nitrogen into glutamate and asparagine (Shi et al., 1997; also see below). The reason
for the presence of GS in uninfected cells is still an open question (Temple et al., 1995;
Forde et al., 1989; Shi et a1, 1997).

In tropical legumes, such as soybean, the nitrogen assimilated via the GS/GOGAT
patthray is channeled into the purine biosynthetic pathway, leading to the synthesis of
uric acid in the infected cells (Venna and Fortin, 1989). The enzymology of this pathway
is highly conserved (Zalkin and Dixon, 1992). Several enzymes involved in purine
bi oSynthesis have been identiﬁed in nodules of tropical legumes (Atkins, 1991; Boland
and Schubert, 1983; Schnorr et al., 1996). Uric acid is converted to allantoin and allantoic
acid via the action of uricase and allantoinase, which are localized in the peroxisomes of
uninfected cells (Bergmann et al., 1983; Hanks et al., 1983; Nguyen et al., 1985). Nodules
contain a nodule-speciﬁc form of uricase- uricase H, which consists of four subunits of

noClulin 35 (Bergmann et al., 1983).

22

2.2.5. Carbon metabolism in nodules. Fixation of 1 gram of nitrogen requires
estimated 5-10 grams of carbon (Phillips, 1980), and the cost of N2 ﬁxation can amount
up to 1/3 of the total photosynthate in perennial legumes (Maxwell et al., 1984). In
nodules, the metabolism of carbon is adapted to the speciﬁc requirements of the nitrogen
ﬁxation process. 1. The plant partner provides the appropriate carbon compounds to the
bacteroids for the production of ATP, rapidly consumed during nitrogenase activity, and
for other biosynthetic processes. 2. Assimilation of ﬁxed nitrogen into glutamine,
glutamate and nitrogen transport compounds requires a constant supply of carbon
skeletons. 3. Carbon metabolism in infected cells has to operate under anoxic conditions.

It has been well established that nodules derive carbon from photosynthetic
tissues primarily in the form of sucrose (Reibach and Streeter, 1983), which is broken to
UI)P-glucose and fructose by the enzymatic activity of sucrose synthase (SS). Nodules
contain a speciﬁc isofonn of SS, initially identiﬁed as nodulin 100 (Thummler and Verma,
1987). UDP-glucose is believed to be metabolized to phosphoenolpyruvate (PEP) via the
action of glycolytic enzymes in nodules (Kahn et al., 1998 and references therein).
Ho“lever, sucrose and other sugars do not appear to be major carbon metabolites in
n(>C1ules (reviewed in Kahn et al., 1998). Rather, C4 dicarboxylic organic acids (DAs), such
as Succinate, fumarate and primarily malate, are the most important players in the nodule
cit‘l‘bon metabolism (Fig. 1.4). There is a substantial amount of evidence that DAs are the
actidal source of plant carbon for nitrogen-ﬁxing bacteroids. For example, DAs are efﬁcient

Sub strates for in vitro nitrogen ﬁxation by puriﬁed bacteroids (Bergersen, 1977).

IVIOreover, isolated symbiosomes can readily take up (Udvardi et al., 1988) and

23

metabolize DAs (Salminen and Streeter, 1987). On the other hand, sugars are not
efficiently transported into bacteroids (Udvardi et al., 1990).

The rhizobial DA transport system (dctAB) has been found to be absolutely
required for the establishment of effective nodules (Ronson et al., 1981). A plant DA
transporter activity has been identiﬁed in the peribacteroid membrane (PBM), but the
corresponding gene has not yet been cloned (Udvardi et al., 1988). Bacteroids utilize DAs
via the TCA cycle to generate ATP, reducing equivalents and carbon skeletons for
biosynthetic processes. The TCA cycle requires a constant supply of acetyl-CoA, which
is most likely generated from malate, via the actions of malic enzymes and pyruvate
dehydrogenase (reviewed in Kahn et al., 1988).

The primary route for the generation of malate from oxaloacetate in nodule cells
aPpears to be via the malate dehydrogenase (MDH) pathway (Fig. 1.4; Miller et al.,
1998). It is believed that the TCA cycle does not contribute signiﬁcantly to DA
biOSjynthesis in plant cells (Kahn et al., 1988). A cDNA corresponding to a nodule-
th anced isoform of MDH, neMDH, has been recently cloned from alfalfa (Miller et al.,
1998). The neMDH enzyme has a higher capacity for malate production than any of the
foul‘ other MDH isoforms present in alfalfa (Miller et al., 1998). Taken together with the
Speciﬁc induction of neMDH levels in nodules, this observation suggests that neMDH
may be responsible for the dramatic increase of malate content in this organ (Miller et al.,
1 998).

In nodules, oxaloacetate (0A) is mainly produced by the PEP carboxylase (PEPC)

from H2CO3 and PEP (Fig. 1.4; Vance et al., 1994). PEPC occurs as multiple isoforms in

24

many plant species, and its major role in photosynthetic processes in C4 and CAM
plants is well known (Ting, 1985). However, PEPC is also one of the central enzymes of
nodule carbon- and nitrogen metabolism, primarily because 0A is a direct precursor of the
two key metabolites, malate and aspartate. PEPC activity, mRNA and protein levels are
signiﬁcantly upregulated in the nodules (Pathirana et al., 1997). The structure and
expression of the alfalfa PEPC-7 gene corresponding to the nodule-enhanced isoform has
been well characterized (Pathirana et al., 1992, 1997). The PEPC-7 gene is expressed
throughout the nodule in both infected and uninfected cells, nodule parenchyma, in the
Pel‘icycle of the nodule vascular system, as well as in a number of non-symbiotic tissues
(Pathirana et al., 1997). Immunolocalization studies have shown that PEPC is a cytosolic
Protein (Robinson et al., 1996). The presence of PEPC in the uninfected cells along with
the enzymes of nitrogen metabolism such as NADH-GOGAT, AS, AAT (see above)
suggests that this cell type may also be involved in the assimilation of the ﬁxed nitrogen
into asparagine (Shi et al., 1997)
2.2.6. Synthesis and functioning of the peribacteroid membrane. The
rhi4'?-<>bial endosymbionts are engulfed by a host plasma membrane as they enter plant
cells from the infection thread in a process resembling endocytosis (Basset et al., 1977).
The peribacteroid membrane (PBM), initially derived from the host plasma membrane,
a1"ways separates the endosymbiont from the host cell cytosol. A long standing dogma has
been that a major function of PBM is to separate the prokaryotic endosymbiont from the
en\Iironment of the host cytoplasm, thus possibly preventing the induction of defense

FeSponses by plant cells. The second function proposed for the PBM is to control the

25

ﬂow of metabolites to and from the bacteria. For example, the PBM contains transporters
for C4 dicarboxylic acids, such as malate, and ammonia (see above). On the other hand,
PBM is almost impermeable to sugars (Udvardi et al., 1990). The bacterial cytoplasm is
also enclosed in inner and outer membranes The milieu between the bacterial outer
membrane and the PBM constitutes the peribacteroid space (PBS), where a number of
enzymes, common to the plant vacuole, have been found (Mellor, 1989). Even though the
PBM is derived from the plasmalemma, it also develops its own characteristic features
during bacteroid differentiation. In general, the composition of the PBM resembles both
Plasmalemma and tonoplast membranes (Verma and Hong, 1996). A unique feature of the
PBM is the presence of a set of nodulins (Verma, 1992; Cheon et al., 1993). The best
Studi ed protein from this group is nodulin-26 (Miao and Verma, 1993), which is related to
the a family of water channels (aquaporins) from the major intrinsic protein family
(IVIIP). Nodulin-26 has been shown to be able to transport water and glycerol, suggesting
that it may be involved in osmoregulation (Dean et al., 1999).

A characteristic feature accompanying the process of bacterial multiplication in
infbeted cells is a massive synthesis of peribacteroid membranes, the total area of which
have been estimated to be 30 times that of the plasma membrane (Verma, 1992).
BiOgenesis of the PBM and targeting of proteins and other compounds to and from it is

not well understood (Verma and Hong, 1996). The PBM is associated with smooth and
coated vesicles, however their role in its biogenesis and/or functioning is unclear (Mellor
and Werner, 1987). PBM-associated nodulins such as nodulin-26 and nodulin-24 are not

found in the normal plant plasma membrane, suggesting that they must contain PBM

26

speciﬁc targeting sequences (F ortin et al., 1985), but the identity of the latter is presently

unknown (F ortin et al., 1987). Several small GTP-binding proteins belonging to the Rab
family, known to be involved in regulating the vesicle trafﬁc in eukaryotic cells, have been
shown to be up—regulated in the nodules (Cheon et al., 1993; Borg et al., 1997). These

proteins may, in part, be involved in regulating the vesicle-mediated transport to and/or

from PBM (Cheon et al., 1993).

27

Photosynthate

l

Glucose

   
 

Phosphoenolpyruvate
PEPC

Oxaloacetate

, _IDH
i |T\_—_‘IDI
- N 2 <—§—-Malate Malate-—-‘

:

 

 

‘ Glutam' ne
Asparagine I \Purines

/.Eand [SK (IND
Aspartate xport

AAT

orKetoglutarate

*Oxaloacetate

l- (2) Glutamate

 

‘l

 

Glutamine

 

 

Glutamine
PLASTID

Figure 1.4. Schematic representation of the major pathways of nitrogen and carbon

metabolism in nodules. From Vance et al., (1994).

28

2.2.7. Regulation of the functioning of mature nitrogen-fixing nodules.
Given the high complexity of the cytological, physiological and biochemical events taking
place in plant host cells at late stages of nodule development and functioning, we know
surprisingly little about how these events are regulated. Very few candidates for
regulatory molecules speciﬁcally involved in late stages of nodule development have been
isolated. The nmh7 gene, encoding a MADS-box containing protein, is expressed
speciﬁcally in infected cells of nodules and also in ﬂowers, but not in uninfected root or
leaf tissues (Heard and Dunn, 1995). However, it is not known at what stages of nodule
development the nmh7 gene, and a related nodule-speciﬁc gene nth5, are induced
(Heard and Dunn, 1995; Heard et al., 1997). The nodule-enhanced gene GmNdxl from
soybean encodes a homeobox-containing protein (Jorgensen et al., 1999). The expression
0f the GmNde gene is induced early during symbiotic interactions and gradually increases
during the late stages of nodule development (Jorgensen et al., 1999). GmNde mRNA is
a130 present in dividing cells of other tissues, suggesting that the GmNde product may be
inVolved in regulatory events at both early and late stages of symbiosis, as well as
throughout plant development (Jorgensen et al., 1999). Recently, a number of nodule-
enhanced Expressed Sequence Tags (ESTs) with putative regulatory functions have been
isolated from young Medicago truncatula nodules (Gyorgyey et al., 2000). However, the

temporary and spatial expression proﬁles of these cDNAs are not known (Gyorgyey et

al., 2000).

29

Reversible phosphorylation has also been implicated in the regulation of nodulin

activity. Nodulin-26 has been shown to be phosphorylated on the residue 262 by a
calcium-dependent protein kinase, which changes its voltage-sensitive channel activity
(Weaver and Roberts, 1992; Lee et al., 1995). Phosphorylation of soybean nodule PEPC
has been found to result in a decrease of the inhibition of enzyme activity by L-malate
(Zhang et al., 1995). The phosphorylation status of PEPC has been directly linked to the
photosynthetic status of the plant (Zhang et al., 1995). PEPC kinase has been partially
puriﬁed from soybean nodules (Zhang and Chollet, 1997). Nodule sucrose synthase has
also been shown to be phosphorylated and the corresponding kinase has been puriﬁed
(Zhang and Chollet, 1997).

Regulation of expression of late nodulin genes has been investigated (reviewed in
de Bruijn and Schell, 1993). In particular, the regulation of transcription of the
leghemoglobin genes has been studied in great detail (Stougaard et al., 1987; Szabados et
al., 1990; de Bruijn and Schell, 1993; Szczyglowski et al., 1994). Cis-acting elements
required for the general and nodule-infected-cell-speciﬁc expression have been identiﬁed
and well characterized in the 5’ ﬂanking regions of leghemoglobin genes from different

1egumes (Stougaard et al., 1987; Szabados et al., 1990; de Bruijn and Schell, 1993;
Szczyglowski et al., 1994;). The search for the transcripton factor(s) involved in the

regulation of expression of leghemoglobin or any other late nodulin genes is under way.

30

3. Molecular biology and classical genetics as tools to understand nodule biology.

3.1. Plant symbiotic mutants and model legumes. Molecular and classical
genetics have been widely used to gain basic understanding of the processes taking place
during nodule development. A large number of plant mutants affected at different stages
of nodule development have been identiﬁed in several species of legumes (Caetano-
Annoles and Gresshoff, 1991, also see Szczyglowski et al., 1998; Schauser et al., 1999).
However, as mentioned above, with the exception of the nin mutant of L. japonicus (see
above), none of the genes underlying these mutant phenotypes have been cloned. The
major problem to clone a gene in legumes by a chromosome walking is the huge size and
high complexity of their genomes. The most extensively studied legumes such as soybean,
pea and bean are relatively recalcitrant to Agrobacterium-mediated transformation, and
regeneration from tissue culture. This, in turn, makes it extremely challenging, if not
impossible, to use transgenic plant approaches to study gene function, complement
mutant phenotypes and establish mutagenesis programs based on transposon or T-DNA
gene-tagging approaches. To circumvent these shortcomings in the legume research in the

early 905, two legume species have been proposed as model organisms: Lotus japonicus
and Medicago truncatula (Barker et al., 1990; Handberg and Stougaard, 1992; Cook et al.,
1997; Jiang and Gresshoff, 1993). Both legumes have a number of features making them

well suited for molecular and classical genetic research, and the most important ones are

listed below:

31

1) Both species are diploid, self-fertile, amenable to manual cross-pollination and have
small genomes: ~300-400 Mbp (only 3-4 times that of Arabidopsis), comparing to the
estimated ~3 000-4000 Mbp for pea;

2) Protocols for transformation with Agrobacterium tumefaciens and rhizogenes and
regeneration in tissues culture exist for both legumes (Handberg et al., 1994; Stiller et
a1 ., 1997; Barker et al., 1990)

3) Plants are small and have short generation times (3 months seed to seed), making it
possible to grow large numbers of them in a limited space for mutant screens and
evaluate multiple generations in a relatively short time;

4) Polymorphic ecotypes exist for both species, making genetic mapping of the locus
underlying a particular phenotype possible;

5) Genetic and physical maps and BAC libraries are being actively developed.

The ability to efﬁciently generate large numbers of transgenic Lotus japonicus
1i“es has prompted the initiation of transposon and T-DNA mediated gene tagging efforts
in this legume (Schauser et al., 1998). In total, 1112 transgenic lines have been generated,

among which 16 symbiotic mutants have been found and 2 could be potentially tagged
(SChauser et al., 1998). One mutant line, nin (see above), was subsequently shown to
harbor a transposon insertion in a gene involved in nodulation (Schauser et al., 1999). The
cloning of this gene does not only represent the ﬁrst successful example of a gene tagging

approach in legumes, a common and well established procedure in Arabidopsis and maize,

32

but also the ﬁrst isolation of a gene responsible for a symbiotic phenotype. The isolation
of nin locus clearly demonstrates the full potential of Lotus japonicus as a model legume.
In our laboratory, a screen for novel symbiotic mutants of Lotus japonicus has
been conducted using ethanemethylsulfonate (EMS) as a mutagenic agent. As a result, 20
mutants, comprising at least 14 complementation groups, were found (Szczyglowski et
al . , 1 998). The mutant phenotypes observed could be arranged in several classes: (1) total
absence of any visible signs of nodules and normal root phenotype (N od'); (2) Nod’ with
altered root morphology; (3). formation of small white structures resembling nodules
Wed"); (4) presence of white ineffective nodules (NodTix'); (5) mixed phenotype with
wild-type and ineffective nodules present on the same plant (NodiFix'H); (6)
Supemodulation (Nod++) (Szczyglowski et al., 1998). Presently, some of these mutants
are being subjected to more detailed microscopical, physiological and molecular analyses.
The most extensively characterized mutant line, har-I, forms an excessive number of
noChiles in response to rhizobia and displays an aberrant “bushy” root phenotype in the

n0I1~symbiotic state. (Szczyglowski etal., 1998; Wopereis et al., 2000).

3.2 Identification and characterization of late nodulins. Traditionally, the
molecular biological approach has been the isolation and characterization of nodule-
Speciﬁc genes and cDNAs, with the underlying assumption that genes speciﬁcally
induced in nodules may encode proteins (nodulins) important for symbiosis. A number of
late nodulins have been identiﬁed in this fashion and have been arbitrarily divided in three

groups, based on the level of understanding of their functions in symbiosis. The ﬁrst

33

group comprises nodulins with relatively well understood biochemical and symbiotic
functions. This group mostly represents the proteins involved in maintaining nodule
“housekeeping” functions, such as various enzymes of nitrogen and carbon metabolism,
mentioned above (GS, NADH-GOGAT, PEPC, SS, etc), leghemoglobins and nodulin-26.
The second group comprises nodulins for which the biochemical functions are either
known or a good prediction can be made, but for which exact roles in symbiosis are either
obscure or not ﬁrmly established. This is a very diverse group, with such proteins as
cytochrome P-450 (Szczyglowski et al., 1997), protein phosphatase 2C (Kapranov et al.,
1999), peptide transporter LjNOD65 (Szczyglowski et al., 1997), small GTP-binding
Proteins (Borg et al., 1997), and others. The last group includes nodulins with unknown
functions (Gamas et al., 1996; Szczyglowski et al., 1997; Kapranov et al. 1996;
SZCzyglowski et al., 1998; see also chapters 2-7 of this thesis).

Given the overall complexity of events taking place in mature, fully functioning
nodules, it is strongly anticipated that additional genes are likely to be discovered. In
general, the identiﬁcation of novel late nodulin genes is crucial for a thorough molecular
Understanding of the details of nodule functioning. This is especially true for the most

interesting and under-represented class of late nodulins with regulatory functions.
Therefore, a systematic search aimed at the identiﬁcation and characterization of novel
late mRNA species associated with the late stages of nodule morphogenesis and
functioning in a model legume L. japonicus has been initiated using the RNA differential
display methodology (Liang and Pardee, 1992). Our major goal was to identify putative

regulators of nodule development and functioning. Since mRNAs representing such

34

proteins are often low-abundant, the differential display is particularly suitable for such
endeavor, given the high sensitivity of this PCR-based technology.

In total, we isolated 88 unique differential display products, at least 19 of which
represented transcripts speciﬁcally associated with the late stages of nodule development
as shown by northern blot analysis (Szczyglowski et al., 1997; Kapranov et al., 1997;
Kapranov et al., 1999; Szczyglowski et al., 1998). The remaining ESTs did not yield a
hybridization signal with root and nodule mRNA derived probes, suggesting that these

ESTS may represent genes which are expresses at very low levels (Szczyglowski et al.,
1997). A comprehensive description of the differential display screen and the
characterization of the expressed sequence tags (ESTs) and cDNAs obtained using this
approach also constitutes Chapter 2 of this thesis.
The detailed characterization of three nodulin genes identiﬁed in our late-nodulin
EST screen, LjNPP2C1, LjNOD70, and LjNOD16 constitutes the remaining Chapters of
this thesis and has been published in Szczyglowski et al., (1997); Kapranov et al., (1997);
SZczyglowski et al., (1998); Kapranov et al., (1999). Chapter 3 describes the molecular
and biochemical analysis of a protein phosphatase 2C gene, LjNPP2C1, expression of
Which is induced at the late stages of nodule development. Chapter 4 summarizes the two
main approaches undertaken to further investigate the function of the LjNPP2C1 enzyme:
the generation of transgenic L. japonicus plants (over)expressing sense, antisense or
d(Dminant-negative form of LjNPP2C1 mRNA and the screen for the proteins interacting
With LjNPP2C1 using the yeast two-hybrid system. Chapter 5 describes the isolation and

Characterization of a late-nodulin cDNA, LjNOD70, encoding a putative transporter

35

protein localized to the PBM compartment. The weak but signiﬁcant amino acid sequence
similarity observed between LjNOD70 and the bacterial proteins capable of transporting
organic acid, such as oxalate and fumarate, makes this late nodulin an attractive candidate
for role of the yet unidentiﬁed DA transporter present in the PBM and mentioned above.
Chapter 6 describes the initial cloning and characterization LjNOD16 cDNA
encoding a late nodulin protein Nljl6. The following Chapter 7 shows that LjNOD16 is a

Part of larger gene, LjPLP-IV, and that LjNODM transcripts originate in L. japonicus

nodule tissues as a result of unusual transcription events governed by an intron-localized
Promoter sequence in the LjPLP-I V gene. Furthermore, it is shown that the LjPLP—I V gene

iS a member of a gene family encoding a novel class of plant proteins similar to
Phosphatidylinositol transfer proteins (PITPs) and distinguished from previously

dEScribed PITPs by the presence of a C-terminal extension comprised of either Nljl6 or
0‘;her, highly related to Nlj 16, amino-acid sequences. The roles ofN1j16 either as a domain

Within the PITP-like proteins or as a separately expressed protein are also analyzed and

discussed in this Chapter.

36

REFERENCES

Appleby CA (1984) Leghemoglobin and Rhizobium interaction. Annu Rev Plant Physiol
35: 443-478

Atkins CA (1991) Ammonia assimilation and export of nitrogen from the legume nodule.
In: Dilworth M, Glenn A (eds) Biology and Biochemistry of Nitrogen Fixation, pp293-
319, Elsevier Amsterdam

Barker DG, Bianchi S, London F, Dattee Y, Duc G, Essad S, Flament P, Gallusci P,
Genier G, Guy P, Muel X, Tourneur J, Denarie J, Huguet T (1990) Medicago
truncatula, a model plant for studying the molecular genetics of the Rhizobium-legume
symbiosis. Plant Mol Biol 8: 40-49

Basset B, Goodman RN, Novacky A (1977) Ultrastructure of soybean nodules: Release
of rhizobia from the infection thread. Can J Microbiol 23: 573-582 '

Bennett MJ, Lighftoot DA, Cullimore JV (1989) cDNA sequence and differential
expression of the gene encoding the glutamine synthetase - polypeptide of Phaseolus
vulgaris L. Plant Mol Biol 12: 553-565

Bergmann H, Preddie E, Verma DPS (1983) Nodulin-35: a subunit of speciﬁc uricase
(uricase II) induced and localized in the uninfected cells of soybean nodules. EMBO J 2:
2333-2339

Boland MJ, Schubert KR (1983) Biosynthesis of purines by a proplastid fraction from
soybean nodules. Arch Biochem Biophys 220: 179-187

Borg S, Brandstrup B, Jensen TJ, Poulsen C (1997) Identiﬁcation of new protein
species among 33 different small GTP-binding proteins encoded by cDNAs from Lotus
japonicus, and expression of corresponding mRNAs in developing root nodules. Plant J
11: 237-250

Boron L, Szczyglowski K, Konieczny A, Legocki AB (1989) Glutamine synthetase in
Lupinus luteus. Identiﬁcation and preliminary characterization of nodule-speciﬁc cDNA

clone. Acta Biochim Polon 36: 295-301

Boron L, Legocki AB (1993) Cloning and characterization of a nodule-enhanced
glutamine synthetase-encoding gene from Lupinus luteus. Gene 136: 95-102

Brisson NA, Pombo-Gentile A, Verma DPS (1982) Organization and expression of
leghemoglobin genes. Can J Bot 60: 272-278

37

 

Burgess BK, Lowe DJ (1996) Mechanism of molybdenum nitrogenase. Chem Rev 96:
2983-3011

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

Cheon C-I, Lee N-G, Siddique A-BM, Bal AK, Verma DPS (1993) Roles of plant
homologs of Rablp and Rab7p in the biogenesis of the peribacteroid membrane, a
subcellular compartment formed de novo during root nodule symbiosis. EMBO J 12:
4125-4135

Cook D, Dreyer D, Bonnet D, Howell M, Nony N, VandenBosch K (1995) Transient
induction of a peroxidase gene in Medicago truncatula precedes infection by Rhizobium
meliloti. Plant Cell 7: 43-55

Cook DR, VandenBosch K, de Bruijn F, Huguet T (1997) Model legumes get the nod.
Plant Cell 3: 275-281

Cooper JB, Long SR (1994) Morphogenic rescue of Rhizobium melilotr' nodulation
mutants by trans-zeatin secretion. Plant Cell 6: 215-225

Csanadi G, Szecsi J, Kalo P. Kiss P, Endre G. Kondorosi A, Kondorosi E, and Kiss
GB (1994) ENOD12, an early nodulin gene, is not required for nodule formation and
efﬁcient nitrogen ﬁxation in alfalfa. Plant Cell 6: 201-213

Dean RM, Rivers RL, Zeidel ML, Roberts DM (1999) Puriﬁcation and functional
reconstitution of soybean nodulin 26. An aquaporin with water and glycerol transport
properties. Biochemistry 38: 347-353

de Bruijn FJ, Schell J (1993) Regulation of plant genes speciﬁcally induced in
developing and mature nitrogen-ﬁxing nodules: cis-acting elements and trans-acting
factors. In Verma DPS (ed), Control of plant gene expression. CRC Press, Boca Raton,
pp. 241-258

Eady RR (1996) Structure-function relationships of alternative nitrogenases. Chem Rev
96: 3013-3030

Ehrhardt DW, Wais R, Long SR (1996) Calcium spiking in plant root hairs responding
to Rhizobt'um nodulation signals. Cell 85: 673-681

38

Fang Y, Hirsch AM (1998) Studying early nodulin gene ENOD40 expression and
induction by nodulation factor and cytokinin in transgenic alfalfa. Plant Physiol 116: 53-
68

Felle H, Kondorosi E, Kondorosi A, Schultze M (1995) Nod signal-induced plasma
membrane potential changes in alfalfa root hairs are differentially sensitive to structural
modiﬁcations of the lipo-chitooligosaccharide. Plant J 7: 939-947

Forde BG, Day HM, Turton JF, Shen W-J, Cullimore JV, Oliver JE (1989) Two
glutamine synthetase genes from Phaseolus vulgaris 1. display contrasting developmental
and spatial patterns of expression in transgenic Lotus corniculatus plants. Plant Cell 1:
391-401

Fortin MG, Morison NA, Verma DPS (1987) Nodulin-26, a peribacteroid membrane
nodulin is expressed independently of the development of the peribacteroid compartment.
Nucleic Acids Res 15: 813-824

Fortin MG, Zelechowska M, Verma DPS (1985) Speciﬁc targeting of membrane
nodulins to the bacteroid-enclosing compartment in soybean nodules. EMBO J 4: 3041-
3046

Freiberg C, Fellay C, Bairoch A, Broughton WJ, Rosenthal, Perret X (1997)
Molecular basis of symbiosis between Rhizobium and legumes. Nature 387: 394-401

Gamas P, Niebel FdC, Lescure N, Cullimore J (1996) Use of a substractive
hybridization approach to identify new Medicago truncatula genes induced during root
nodule development. Mol Plant Microbe Interact 9: 233-242

Geurts R Franssen H (1996) Signal transduction in Rhizobium-induced nodule
formation. Plant Physiol 112: 447-453

Gregerson RG, Miller SS, Twary SN, Gantt JS, Vance CP (1993) Molecular
characterization of NADH-dependent glutamate synthase from alfalfa nodules. Plant Cell
5: 215-226

Gualtieri G, Bisseling T (2000) The evolution of nodulation. Plant Mol Biol 42: 181-
194

Gyorgyey J, Vaubert D, Jimenez-Zurdo JI, Charon C, Troussard L, Kondorosi A,
Kondorosi E (2000) Analysis of Medicago truncatula nodule expressed sequence tags.
Mol Plant-Microbe Interact 13: 62-71

39

 

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

Handberg K, Stiller J, Thykjear T, Stougaard J (1994) Transgenic plants:
Agrobacterium mediated transformation of the diploid legume Lotus japonicus. In: Cell
Biology: A Laboratory Handbook, Celis JE, ed, 1: 119-127

Hanks JF, Schubert KR, Tolbert NE (1983) Isolation and characterization of infected
and uninfected cells from soybean nodules. Role of unifected cells in ureide synthesis.
Plant Physiol 71: 869-873

Hadri A-E, Spaink HP, Bisseling T, Brewin NJ (1998) Diversity of root nodulation
and rhizobial infection processes. In Spaink HP, Kondorosi A, Hooykaas PJJ (eds), The
Rhizobiaceae. Kluwer Academic Publishers, Dordrecht, pp.347-3 60.

Heard J, Dunn K (1995) Symbiotic induction of a MADS-box gene during development
of alfalfa root nodules. Proc Natl Acad Sci USA 92: 5273-5277

Heard J, Caspi M, Dunn K (1997) Evolutionary diversity of symbiotically induced
nodule MADS box genes: characterization of nth5, a member of a novel subfamily.
Mol Plant-Microbe Interact 10: 665-676

Heidstra R, Geurts R, F ranssen H, Spaink HP, van Kammen A, Bisseling T (1994)
Root hair deformation activity of nodulation factors and their fate on Vicia sativa. Plant
Physiol 105: 787-797

Heidstra R, Yang WC, Yalcin Y, Peck S, Emons A, van Kammen A, Bisseling T
(1997) Ethylene provides positional information on cortical cell division but is not
involved in Nod factor-induced root hair tip growth in Rhizobium-legume intercation.
Development 124: 1781-1787

Hirsch AM (1992) Tansley Review No. 40. Developmental biology of legume
nodulation, New Phytol 122: 211-237

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

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

40

 

Hong Z, Verma DPS (1994) A phosphatidylinositol 3-kinase is induced during soybean
nodule organogenesis and is associated with membrane proliferation. Proc Natl Acad Sci
USA 91:9617-9621

Howard JB, Rees DC (1996) Structural basis of biological nitrogen ﬁxation. Chem Rev
96: 2965-2982

Jiang Q, Gresshoff PM (1993) Lotus japonicus: a model plant for structure-function
analysis in nodulation and nitrogen ﬁxation. Curr Topics Plant Mol Biol 2: 97-110

Jorgensen J-E, Gronlund M, Pallisgaard N, Larsen K, Marcker KA, Jensen EO
(1999) A new class of plant homeobox genes is expressed in speciﬁc regions of
determinate symbiotic root nodules. Plant Mol Biol 40: 65-77

Kahn ML, McDermott TR, Udvardi MK (1998) Carbon and nitrogen metabolism in
Rhizobia. In Spaink HP, Kondorosi A, Hooykaas PJJ (eds), The Rhizobt‘aceae, Molecular
Biology of Model Plant-Associated Bacteria. Kluwer Academic Publishers, Dordrecht,
pp. 461-485

Kaiser BN, Finnegan PM, Tyerman SD, Whitehead LF, Bergersen FJ, Day DA,
Udvardi MK (1998) Characterization of an ammonium transport protein from he
peribacteroid membrane of soybean nodules. Science 281: 1202-1206

Kaminski PA, Batut J, Boistard P (1998) A survey of symbiotic nitrogen ﬁxation by
Rhizobia. In Spaink HP, Kondorosi A, Hooykaas PJJ (eds), The Rhizobiaceae, Molecular
Biology of Model Plant-Associated Bacteria. Kluwer Academic Publishers, Dordrecht,
pp. 431-460

Kapranov P, de Bruijn FJ, Szczyglowski K (1997) A novel, highly expressed late
nodulin gene LjNODI6 from Lotus japonicus. Plant Physiol 113: 1081-1090

Kapranov P, Jensen TJ, Poulsen C, de Bruijn FJ, Szczyglowski K (1999) A protein
phosphatase 2C gene, LjNPP2C1, from Lotus japonicus induced during root nodule
development. Proc Natl Acad Sci USA 96: 1738-1743

Kijne, J.W. (1992) The Rhizobium infection process. In Stacey G, Burris RH, Evans HJ,
(eds), Biological Nitrogen Fixation. Chapman and Hall, New York, pp. 349-398

Lam H-M, Coschigano KT, Oliveira IC, Melo-Oliveira R, Coruzzi GM (1996) The

molecular-genetics of nitrogen assimilation into amino acids in higher plants. Annu Rev
Plant Physiol Plant Mol Biol 47:569-593

41

Lea PJ, Robinson SA, Stewart GR (1990) The enzymology and metabolism of
glutamine, glutamate and asparagine. In Miﬂin BJ, Lea PJ (eds), The Biochemistry of
Plants, Vol. 16, Intermediary Nitrogen Metabolism. Academic Press, San Diego, pp. 121-
159

Lee JW, Zhang Y, Weaver CD, Shomer NH, Louis CF, Roberts DM (1995)
Phosphorylation of nodulin 26 on serine 262 affects its voltage-sensitive channel activity
in planar lipid bilayers. J Biol Chem 270: 27051-27057

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

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

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

Long SR (1996) Rhizobium symbiosis: Nod factors in perspective. Plant Cell 8: 1885-
1898

Maxwell CA, Vance CP, Heichel GH, Stade S (1984) C02 ﬁxation in alfalfa and
birdsfoot trefoil root nodules and partitioning of 14C to the plant. Crop Sci 24: 257-264

Mellor RB (1989) Bacteroids in the Rhizobr'um-legume symbiosis inhabit a plant internal
lytic compartment: implications for other microbial endosymbioses. J Exp Bot 40: 831-
839

Mellor RB, Werner D (1987) Peribacteroid membrane biogenesis in mature legume root
nodules. Symbiosis 3: 75-100

Miao G-H, Verma DPS (1993) Soybean nodulin-26 gene encoding a channel protein is
expressed only in the infected cells of nodules and is regulated differently in roots of
homologous and heterologous plants. Plant Cell 5: 781-794

Miller SS, Driscoll BT, Gregerson RG, Gantt JS, Vance CP (1998) Alfalfa malate
dehydrogenase (MDH): molecular cloning and characterization of ﬁve different forms

reveals a unique nodule-enhanced MDH. Plant J 15: 173-184

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

42

Nap J-P, Bisseling T (1990) Developmental biology of a plant-prokaryote symbiosis:
The legume root nodule. Science 250, 948-954

Newcomb W, Sippel D, Peterson RL (1979) The early morphogenesis of Glycine max
and Pisum sativum root nodules. Can J Bot 57: 2603-2616

Nguyen T, Zelechowska MG, Foster V, Bergmann H, Verma DPS (1985) Primary
structure of the soybean nodulin-35 gene encoding a nodule-speciﬁc uricase localized in
peroxisomes of uninfected cells of soybean. Proc Natl Acad Sci USA 82: 5040-5044

Niebel A, Gressent F, Bono JJ, Ranjeva R, Cullimore J (1999) Recent advances in
the study of Nod factor perception and signal transduction. Biochimie 81: 669-674

Pathirana SM, Vance CP, Miller SS, Gantt JS (1992) Alfalfa root nodule
phosphoenolpyruvate carboxylase: Characterization of the cDNA and expression in
effective and ineffective nodules. Plant Mol Biol 20: 437-450

Perret X, Freiberg C, Rosenthal A, Broughton WJ, Fellay R (1999) High-resolution
transcriptional analysis of the symbiotic plasmid of Rhizobium sp. NGR234. Mol
Microbiol 32: 415-425

Peters NK, Crist—Estes DK (1989) Nodule formation is stimulated by the ethylene
inhibitor aminoethoxyvinyl glycine. Plant Physiol 91: 690-693

Phillips DA (1980) Efﬁciency of symbiotic nitrogen ﬁxation in legumes. Annu Rev Plant
Physiol Plant Mol Biol 31: 29-49

Pingret J-L, Journet E-P, Barker DG (1998) Rhizobium Nod factor signaling: evidence
for a G protein-mediated transduction mechanism. Plant Cell 10: 659-671

Pueppke SG (1996) The genetic and biochemical basis for nodulation of legumes by
Rhizobia. Crit Rev Biotechnology 16: 1-51

Reibach PH, Streeter JG (1983) Metabolism of 14C-labeled photosynthate and
distribution of enzymes of glucose metabolism in soybean nodules. Plant Physiol 72:
634-640

Ribbe M, Gadkari D, Meyer 0 (1997) N2 ﬁxation by Streptomyces
thermoautotrophicus involves a molybdenum-dinitrogenase and a manganese-superoxide
oxidoreductase that couple N2 reduction to the oxidation of superoxide produced from 02
by a molybdenum-CO dehydrogenase. J. Biol. Chem. 272: 26627-26633

43

Robinson DL, Kahn ML, Vance CP (1994) Cellular localization of nodule enhanced
aspartate amonitransferase in Medicago sativa L. Planta 92:202-210

Robinson DL, Pathirana SM, Gantt JS, Vance CP (1996) Irnmunogold localization of
nodule-enhnanced phosphoenolpyruvate carboxylase in alfalfa. Plant Cell Environ 19:
602-608

Ronson CW, Lyttleton P, Robertson, JG (1981) C4-dicarboxylate transport mutants of
Rhizobium trifolii from ineffective nodules on T rifolium repens. Proc Natl Acad Sci USA
78: 4284-4288

Salminen S0, Streeter JG (1987) Involvement of glutamate in the respiratory
metabolism of Bradyrhizobiumjaponicum bacteroids. J Bacteriol 169: 495—499

Schauser L, Handberg K, Sandal N, Stiller J, Thykjaer T, Pajuelo E, Nielsen A,
Stougaard J (1998) Symbiotic mutants deﬁcient in nodule establishment identiﬁed after
T-DNA transformation of Lotus japonicus. Mol Gen Genet 259: 414-423

Schauser L, Roussis A, Stiller J, Stougaard J (1999) A plant regulator controlling
development of symbiotic root nodules. Nature 402: 191-195

Scheres B, van Engelen F, van den Knaa, E, van de Wiel C, van Kammen A, and
Bisseling T (1990a) Sequential induction of nodulin gene expression in the developing
pea nodule. Plant Cell 2: 687-700

Scheres B, Van de Wiel C., Zalensky A, Horvath B, Spaink H, Van Eck H,
Zwartkruis F, Wolters AM, Gloudemans T, Van Kammen A, and Bisseling T
(1990b) The ENOD12 gene product is involved in the infection process during the pea-
Rlu‘zobium interaction. Cell 60: 281-294

Schindelin H, Kisker C, Schlessman JL, Howard JB, Rees DC (1997) Structure of
ADP-A1F4'-stabilized nitrogenase complex and its implications for signal transduction.
Nature 387 : 370-376

Schmidt JS, Harper JE, Hoffman TK, Bent AF (1999) Regulation of soybean
nodulation independent of ethylene signaling. Plant Physiol 119: 951-960

Schnorr KM, Laloue M, Hirel B (1996) Isolation of cDNAs encoding two purine
biosynthetic enzymes of soybean and expression of the corresponding transcripts in

roots and root nodules. Plant Mol Biol 32: 751-757

Schubert KR (1986) Products of biological nitrogen ﬁxation in higher plants: Synthesis,
transport, and metabolism. Annu Rev Plant Physiol 37: 539-574

44

Schultze M, Kondorosi E (1998) Regulation of symbiotic root nodule development.
Annu. Rev. Genet. 1998: 33-57

Schultze M, Kondorosi E, Ratet P, Buire M, Kondorosi A (1994) Cell and molecular
biology of Rhizobium-plant interactions. Int Rev Cytol 156: 1-75

Shi L, Twary SN, Yoshioka H, Gregerson RG, Miller SS, Samac DA, Gantt J S,
Unkefer PJ, Vance CP (1997) Nitrogen assimilation in alfalfa: isolation and
characterization of an asparagine synthetase gene showing enhanced expression in root
nodules and dark-adapted leaves. Plant Cell 9: 1339-1356

Spaink HP (1992) Rhizobial lipooligosaccharides: answers and questions. Plant Mol Biol
20: 977-986

Spaink HP, Lugtenberg BJJ (1994) Role of rhizobial lipo-chitin oligosaccharide signal
molecules in root nodule organogenesis. Plant Mol Biol 26: 1413-1422

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

Sprent JI (1989) Which steps are essential for the formation of functional legume
nodules? New Phytol. 111: 129-153

Stanford AC, Larsen K, Barker DG, Cullimore JV (1993) Differential expression
within the glutamine synthetase gene family of the model legume Medicago truncatula.
Plant Physiol 103: 73-81

Stiller J, Martirani L, Tuppale S, Chian R-J, Chiurazzi M, Gresshoff PM (1997)
High frequency transformation and regeneration of transgenic plants in the model legume
Lotus japonicus. J Exp Bot 48: 1357-1365

Stougaard J, Sandal NN, Gron A, Kuhle A, Marcker KA (1987) 5’ Analysis of the
soybean leghaemoglobin lc3 gene: regulatory elemnts required for promoter activity and
organ speciﬁcity. EMBO J 6: 3565-3569

Szabados L, Ratet P, Grunenberg B, de Bruijn FJ (1990) Functional anlysis of the

Sesbania rostrata leghemoglobin glb3 gene 5’-upstream region in transgenic Lotus
comiculatus and Nicotiana tabacum plants. Plant Cell 2: 973-986

45

Szczyglowski K, Szabados L, Fujimoto S, Silver D, de Bruijn FJ (1994) Site—speciﬁc
mutagenesis of the nodule-infected cell expression (NICE) element and the AT-rich
element ATRE-BS2* of the Sesbania rostrata leghemoglobin glb3 promoter. Plant Cell 6:
317-332

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

Szczyglowski K, Shaw SR, Wopereis J, Copeland S, Hamburger D, Kasiborski B,
Dazzo FB, de Bruijn FJ (1998a) Nodule organogenesis and symbiotic mutants of the
model legume Lotus japonicus. Mol Plant-Microbe Interact 11: 684-697

Szczyglowski K, Kapranov P, Hamburger D, de Bruijn FJ (1998b) The Lotus
japonicus LjNOD70 gene encodes a protein with similarities to transporters. Plant Mol
Biol 37: 651-661

Temple SJ, Heard J, Ganter G, Dunn K, Sengupta-Gopalan C (1995)
Characterization of a nodule enhanced glutamine synthetase from alfalfa: nucleotide
sequence, in situ localization, and transcript analysis. Mo] Plant-Microbe Interact 8:218-
227

Thummler, F., and Verma, D.P.S. (1987) Nodulin-100 of soybean is the subunit of
sucrose synthetase regulated by the availability of free heme in nodules. J Biol Chem 262:
14730-14736

Ting IP (1985) Crassulacean acid metabolism. Annu Rev Plant Physiol 36: 595-622

Tjepkema JD, Yokum SC (1974) Measurement of oxygen partial pressure within
soybean nodules by oxygen microelectrodes. Planta 119: 351-360

Udvardi MK, Day DA (1997) Metabolite transport across symbiotic membranes of
legume nodules. Annu Rev Plant Physiol Plant Mol Biol 48: 493-523

Udvardi MK, Price GD, Gresshoff PM, Day DA (1988) A dicarboxylate transporter on
the peribacteroid membrane of soybean nodules. FEBS Lett 231: 36-40

Udvardi MK, Ou Yang L-J, Young S, Day DA (1990) Sugar and amino acid transport
across symbiotic membranes from soybean nodules. Mol Plant-Microbe Interact 3: 334-

340

van Kammen A (1984) Suggested nomenclature for plant genes involved in nodulation
and symbiosis. Plant Mol Biol Rep 2: 43-45

46

van Spronsen PC, Bakhuizen R, van Brussel AAN, Kijne JW (1994) Cell wall
degradation during infection thread formation by the root nodule bacterium Rhizobium
leguminosarum is a two-step process. Eur J Cell Biol 64: 88-94

van Brussel AAN, Bakhuizen R, van Spronsen PC, Spaink HP, Tak T, Lugtenberg
BJJ, Kijne JW (1992) Induction of pre-infection thread structures in the leguminous
host plant by mitogenic lipo-oligosaccharides of Rhizobium. Science 257: 70-72

Vance CP (1990) Symbiotic nitrogen ﬁxation: recent genetic advances. In Miﬂin BJ, Lea
PJ (eds), The Biochemistry of Plants, Vol. 16, Intermediary Nitrogen Metabolism.
Academic Press, San Diego, pp. 43-88

Vance CP (1998) Legume symbiotic nitrogen ﬁxation: agronomic aspects. In Spaink HP,
Kondorosi A, Hooykaas PJ J (eds), The Rhizobiaceae, Molecular Biology of Model Plant-
Associated Bacteria. Kluwer Academic Publishers, Dordrecht, pp. 509-530.

Vance CP, Miller SS, Gregerson RG, Samac DA, Robinson DL, Gantt JS (1995)
Alfalfa NADH-dependent glutamate synthase: structure of the gene and importance in
symbiotic N2 ﬁxation. Plant J 8: 345-358

Vance CP, Gantt JS (1992) Control of nitrogen and carbon metabolism in root nodules,
Physiol Plant 85: 266-274

Vance CP, Gregerson RG, Robinson DL, Miller SS, Gantt JS (1994) Primary
assimilation of nitrogen in alfalfa nodules: molecular features of the enzymes involved.
Plant Sci 101: 51-64

Verma DPS, Fortin MG (1989) Nodule development and formation of the
endosymbiotic compartment. In Schell J, Vasil IK, eds, Cell Culture and Somatic Cell
Genetics of Plants, Vol. 6, Molecular Biology of Plant Nuclear Genes. Academic Press,
San Diego, pp. 329-353

Verma DPS, Hong Z (1996) Biogenesis of the peribacteroid membrane in root nodules.
Trends Microbiology 4: 364-368

Waters JK, Hughes BL, Purcell LC, Gerhardt KO, Mawhinney TP, Emerich DW
(1998) Alanine, not ammonia, is excreted from Nz-ﬁxing soybean nodule bacteroids. Proc
Natl Acad Sci USA 95: 12038-12042

Weaver CD, Roberts DM (1992) Determination of the site of phosphorylation of

nodulin 26 by the calcium-dependent protein kinase from soybean nodules. Biochemistry
31 : 8954-8959

47

Weaver CD, Shomer NH, Louis CF, Roberts DM (1994) Nodulin 26, a nodule-speciﬁc
symbiosome membrane protein from soybean, is an ion channel. J Biol Chem 269: 17858-
17862

Werner D, Morschel E, Stripf R, Winchenbach B (1980) Development of nodules of
Glycine max infected with an ineffective strain of Rhizobiumjaponicum.

Zalkin H, Dixon JE (1992) De novo purine biosynthesis. Prog Nucl Acids Res Mol Biol
42: 259-287

Zhang X-Q, Chollet R (1997) Seryl-phosphorylation of soybean nodule sucrose
synthase (nodulin-100) by a Cay-dependent protein kinase. FEBS Lett 410: 126-130

Zhang X-Q, Chollet R (1997) Phosphoenolpyruvate carboxylase protein kinase from

soybean root nodules: partial puriﬁcation, characterization, and up/down-regulation by
photosynthate supply from the shoots. Arch Biochem Biophys 343: 260-268

48

CHAPTER 2

Construction of a Lotus japonicus late nodulin expressed sequence tag

library and identification of novel nodule-speciﬁc genes‘

2.1. ABSTRACT

A range of novel expressed sequence tags (ESTs) associated with late
developmental events during nodule organogenesis in the legume Lotus japonicus were
identiﬁed using mRNA differential display. One hundred ten differentially displayed PCR
products were cloned and analyzed. Of 88 unique cDNAs obtained, 25 shared signiﬁcant
homology to DNA/protein sequences in the respective databases. This group comprises,
among others, a nodule-speciﬁc homologue of protein phosphatase 2C, a peptide
transporter protein, and a nodule-speciﬁc form of cytochrome P450. RNA gel blot
analysis of 18 differentially displayed ESTs conﬁrmed their nodule speciﬁc expression
pattern. The kinetics of mRNA accumulation of the majority of the ESTs analyzed were
found to resemble the expression pattern observed for the L. japonicus leghemoglobin
gene. These results indicate that the newly isolated molecular markers indeed correspond
to genes induced during late developmental stages of L. japonicus nodule organogenesis,

and provide important, novel tools for the study of nodulation.

 

1 This chapter was published in: Szczyglowski K, Hamburger D, Kapranov P, de Bruijn

F (1997) Plant Physiology 114: 1335-1346.

49

2.2. INTRODUCTION

The formation of nitrogen-ﬁxing nodules represents an unusual example of
externally induced organogenesis that unites the plant host and the symbiotic bacteria in a
microenvironment appropriate for the support of bacterial nitrogen ﬁxation and plant-
mediated assimilation of nitrogen. This complex and highly regulated process begins with
the speciﬁc recognition between the plant and bacterial partners, which leads to the
synthesis of morphogenic lipochito-oligosaccharide molecules of bacterial origin (Nod
factors; for a recent review see, Carlson et al., 1995; Spaink, 1996). The Nod factor signal
molecules appear to exert their role in the plant root by inducing root hair curling and
deformation, as well as redirecting the fate of root cortical cells toward the initiation of a
nodule primordium (Spaink, 1996, and references therein). The initial stages of nodule
organogenesis also include infection thread formation and penetration of root cells by
symbiotic bacteria (for a recent review, see Schultze et al., 1994). Speciﬁc plant genes,
called early nodulin genes, are activated during early nodule morphogenetic events
(Schulze et al., 1994; Mylona et al., 1995). Although the exact functions of the majority
of the early nodulin genes identiﬁed to date have yet to be determined, the expression
patterns of some of these genes have been correlated with processes such as pre-infection
(Mtripl; Cook et al., 1995), infection (ENOD5, ENOD12, Scheres et al., 1990a, 1990b),
and nodule meristem initiation or nodule structure formation (cdc2-S5, ENOD40,
ENODZ; Franssen et al., 1987; Miao et al., 1993; Crespi et al., 1994; Matvienko et al.,

1994; van de Sande et al., 1996;).

50

The successful invasion of plant root cells by symbiotic bacteria leads to the ﬁnal
stage of nodule formation, which culminates in the establishment of a fully developed,
nitrogen-ﬁxing nodule. However, the developmental cues and molecular events underlying
the late steps in nodule organogenesis are largely unknown. Several events that occur in
both symbiotic partners at this late stage have been correlated with bacterial release ﬁom
the infection thread, and plant cell colonization (Sprent, 1989). These events include
processes such as proliferation of the membrane system (PBMs; symbiosome
membrane), bacteroid differentiation, and molecular and biochemical alterations that create
and support the physiological environment required for nitrogen ﬁxation and ammonia
assimilation (V erma, 1992; Mylona et al., 1995).

Immediately prior to, or concomitantly with, the initiation of nitrogen ﬁxation, a
group of speciﬁc plant genes, called late nodulin genes, is activated (Verma, 1992; de
Bruijn and Schell, 1992). Since these genes are not expressed in nodules lacking infected
cells, it has been postulated that late nodulin genes may be coordinately expressed as a
result of a single signal related to the release of bacteria from the infection thread (Nap and
Bisseling, 1990; Welters et al., 1993). Typical members of this group include abundantly
expressed genes encoding enzymes, or subunits of enzymes, that function in nitrogen or
carbon metabolism; proteins associated with the peribacteroid membrane; a family of
oxyhemoproteins (leghemoglobins); and a number of proteins the function of which
remains to be elucidated (Delauney and Verma, 1988; de Bruijn and Schell, I992; Mylona
et al., 1995). A role for small GTP-binding proteins (Rablp and Rab7p homologues) in

PBM development has also been proposed (Cheon et al., 1993), and a putative plant

51

transcription factor (NMH7), present speciﬁcally in the infected bacteroid-containing
cells, has been identiﬁed in alfalfa root nodules (Heard and Dunn, 1995). It has been
suggested that NMH7 may be involved in cellular activities speciﬁc to the differentiation
of the infected cells (Heard and Dunn, 1995).

To study the complexity of genes expressed during the transition period between
the development of the nodule structure and formation of a functional nodule, we
examined the mRNA expression proﬁles in determinate nodules of the diploid legume
Lotus japonicus (Handberg and Stougaard, 1992; Jiang and Gresshoff, 1997), using the
differential display method (Liang and Pardee, 1992). Here we describe the isolation of
numerous ESTs corresponding to novel late—nodulin genes, many of which appear to

encode functions that have not previously been implicated in the nodulation process.

52

2.3. MATERIALS AND METHODS

Plant Material

Lotus japonicus GlF U B-129-S9 seeds (kindly provided by Dr. Jens Stougaard,
Aarhus University, Denmark) were surface sterilized and germinated as described by
Handberg et al. (1994). Rhizobium loti strain NZP2235 (Jarvis et al., 1982) was grown for
two days at 28°C in TY medium and used to inoculate seven days old L. japonicus
seedlings immediately before potting. For the initial experiments, a 3:1 mixture of
vermiculite and sand was used as plant growth medium. Later we reﬁned the L. japonicus
grong conditions, resulting in faster and more efficient levels of nodulation (data not
shown). Therefore, for the developmental Northern slot blot experiments shown in Fig.
2.7, a 3:3:1 mixture of grade 2 vermiculite, grade 3 vermiculite, and sand was used. All
plants were grown in a controlled environment in growth chambers (18h/6h day/night
cycle, 250 uE sec"m'2 light intensity, 22°C/180C day/night temperature, and 70%
humidity). B&D solution (Broughton and Dilworth, 1971) supplemented with 0.5 mM
KNO3 was used to water the plants. The relatively low concentration of combined
nitrogen in B&D solution supports growth of the uninoculated control plants, but does
not affect nodule formation on roots of infected L. japonicus plants (data not shown).

Root segments were harvested at various time points after infection, and fully developed

53

nodules were collected at the 21-day time point. Plant tissues were immediately frozen in

liquid nitrogen and stored at -70°C until use.

Nucleic Acid Isolation

R. loti genomic DNA was isolated following the procedure of Marmur (1961),
except that the bacterial cells were washed with 1 M NaCl prior to Pronase E digestion.
For the isolation of plant genomic DNA the method described by Rogers and Bendich
(1988) was used.

Total plant RNA was isolated using the procedure of Verwoerd et al. (1989),
except that the extraction buffer used was as described by Hall et al. (1978). The poly
(A)+ fraction of mRNA was isolated using a mini-spin column kit (5 Prime-3 Prime, Inc.;

Boulder, CO), following the manufacturer’s instructions.

Developmental Differential Display of mRNA

The RNA differential display procedure was canied out using RNAmapTM kits
from the GenHunter Corporation (Brookline, MA), following the manufacturer’s
instructions (see also Goorrnachtig et al., 1995). The cDNA synthesis was performed

using 0.5 pg of total RNA isolated from root segments (7, 11, and 13 days aﬁer

infection), and nodules harvested 21 days after infection. For control experiments, total

54

RNA isolated from 7- and 21-day-old uninfected roots was used. Selected bands were
reampliﬁed and then blunt ended by treatment with the Klenow ﬁagment of DNA
polymerase I (Boehringer Mannheim, Indianapolis, IN), phosphorylated at the 5’ ends
using T4 polynucleotide kinase, and cloned into the SmaI digested plasmid pK18
(Pridmore, 1987). Recombinant plasmids were mobilized into the Escherichia coli strain

InvorF’ (Novagen, Madison, WI), using electrotransformation. For each differentially

displayed PCR-fragment, 48 recombinant colonies were collected and stored individually

in microtiter plates at -70°C, in 30% glycerol.

Southern and Northern Analyses

For Southern blot analyses, 10 pg of plant genomic DNA, or 5 pg of R. loti total
DNA, were digested to completion using EcoRI endonuclease. The digested DNA was
separated on a 0.8% agarose gel, transferred onto Hybond-N nitrocellulose ﬁlters
(Amersham, Arlington Heights, IL), and UV cross-linked, following standard procedures
(Sambrook et al., 1989). Hybridizations and subsequent washes were carried out at 65°C,
using high stringency conditions (Sambrook et al., 1989).

For northern analyses, 10 pg of total RNA, or 4 pg of poly (A)+ RNA, were

separated on a 1.2% denaturing agarose gel (6% formaldehyde, lxMOPS buffer: 20 mM
MOPS, 1.0 mM EDTA, 5.0 mM sodium acetate, pH 7.0) and transferred onto Nitro-Plus

membranes (Fisher Scientiﬁc, Pittsburgh, PA), as described (Sambrook et al., 1989). Ten

55

micrograms of total RNA were used for the slot blot RNA hybridization.
Prehybridizations and hybridizations were performed according to the procedure
described by Church and Gilbert (1984). Filters were washed twice for 15 min in 2X SSC,
0.1% SDS; once for 15 min in 0.3X SSC, 0.1% SDS; and once for 10 min in 0.1x SSC, at

65°C. DNA probes were labeled with a32P-dATP, using the random prime kit from

Boehringer—Mannheim (Indianapolis, IN), following the manufacturer’s instructions.

Differential Colony Hybridization

For differential colony hybridization, microtiter “combo” plates were developed.
A single 96-well microtiter plate combined representatives of eight (A to H) differentially
displayed PCR products, each represented by 12 recombinant colonies (8x12). The
bacteria were transferred from the “combo” plates onto nitrocellulose ﬁlter disks
(Schleicher and Schuell, Keene, NH) and grown over night at 28°C. The colonies on the
ﬁlters were lysed and the liberated DNA was ﬁxed to the ﬁlters (Sambrook et al., 1989).
Two replica ﬁlters were prepared for each microtiter “combo” plate, and hybridized with
radiolabeled cDNA probes derived from poly(A)+ mRNA fractions from uninfected
control roots, or nodules (see below).

Radiolabeled single-stranded cDNA probes were synthesized using 1 pg of poly

(A)+ RNA isolated from 21-day-old uninfected L. japonicus roots, or root nodules,

respectively. The reaction mixtures contained 50 mM Tris HCl pH 8.5, 8 mM MgC12, 30

56

mM KCl, 1 mM DTT, 50 pM of each, dCTP, dGTP and dTTP, 100 pCi a32P-dATP, 10
pM of each of the T12MN primers (RNAmapTM kits; GenHunter Corporation, Brooklyn,

MA), 20 units RNasin, and 50 units of AMV reverse transcriptase. Aﬁer 20 min at 37°C,

2 pl of a mixture containing 10 mM dATP, dCTP, dGTP, and dTTP was added. The
reaction was continued for 40 min and terminated by the addition of 1 pl of 0.5 M EDTA
(pH 8.0). The RNA was hydrolyzed by the addition of 12 pl of 150 mM NaOH,
incubated at 65°C for 1 h, and neutralized with 12 pl 1 M Tris HCl (pH 8.0) and 12 pl 1

N HCl. The radiolabeled, single-stranded cDNAs were separated from unincorporated

dNTPs on Sephadex GSO columns and used directly for differential hybridization.

cDNA Library Screening

The cDNA library from mature nodules of L. japonicus was kindly provided by

Dr. Jens Stougaard, Aarhus, Denmark. The library was constructed with oligo-dT primer

in the lambda UniZAP vector from Stratagene (La Jolla, CA). Screening for full-copy

cDNAs was performed following standard procedures (Sambrook et al., 1989; Stratagene

manual).

DNA Sequencing and Computer Analysis

57

Manual DNA sequencing was performed using the Sequenase 2.0 kit (USB, Inc.,
Cleveland OH), following the manufacturer’s instructions. Computer analyses of DNA
and protein sequences were carried out using the Squd (Applied Biosystems, Inc.,
Foster City, CA), BLAST (Altschul et al., 1990), and GCG (Genetics Computer Group,
Madison, WI) software packages. For automated ﬂuorescent sequencing, Taq
polymerase-mediated cycle-sequencing reactions were performed, according to Newman
et a1. (1994). The plasmid templates were prepared by growing bacterial cultures in
MR2001 broth (MacConnell Research, San Diego, CA) and extracting of DNA using the
Wizard Miniprep Kits (Promega, Madison, WI). One strand of the DNA templates was
sequenced (by the MSU Sequencing Facility; MSU-DOE-Plant Research Laboratory,

East Lansing, MI).

58

2.4. RESULTS

Molecular Characterization of L. japonicus Root Nodule Development

Upon infection with R. loti, L. japonicus plants form determinate nodules
(Handberg and Stougaard, 1992; Jiang and Gresshoff, 1993). However, a considerable
amount of variation in nodulation efﬁciency has been observed with different R. loti
strains (data not shown). Therefore, to develop an efﬁcient system, we examined the
nodulation characteristics of twenty-one different R. loti strains (provided by Dr. D.B.
Scott, Massey University, New Zealand). Out of this collection, a highly efﬁcient R. loti
strain, NZP2235, was selected and used for further nodulation of L. japonicus ecotype
Gifu plants (K. Szczyglowski and F.J. de Bruijn, unpublished data). Under the plant
growth condition used in this study, visible signs of nodule formation could be detected
on the roots of L. japonicus plants between 3 and 5 d after inoculation. Over the next few
days, usually between 7 and 11 days after inoculation, nodule structures enlarged and
turned pink, indicative of leghemoglobin synthesis. To further deﬁne the early and late
stages of L. japonicus nodule development, RNA gel blot analyses were performed with
two well characterized “marker” genes, namely the L. japonicus Enod2 gene (R. Chen, K.
Szczyglowski, F.J. de Bruijn, unpublished data), and a cDNA corresponding to a L.
japonicus leghemoglobin gene (J. Stoltzfus, K. Szczyglowski, F .J . de Bruijn, unpublished

data). In a number of legume species, the expression of the early nodulin gene Enod2 and

59

the leghemoglobin genes has been correlated with early and late stages in nodule
development, respectively (Nap and Bisseling, 1990; Mylona et al., 1995).

Developmental northern blot analysis using total RNA isolated ﬁom roots and
nodules at different time points after L. japonicus inoculation revealed that the Enod2
gene was expressed at a very low level in the uninfected control roots (Fig. 2.1). The level
of Enod2 mRNA gradually increased between 7 and 21 d after inoculation. In comparison,
the expression of the L. japonicus leghemoglobin gene was ﬁrst detected at 11 d after
inoculation. Based on these observations we concluded that the late developmental events
in the L. japonicus nodule morphogenesis process are likely to occur between 7 and 11 d
after inoculation. Therefore, this time period became the focus of our further molecular

analysis.

Construction of a L. japonicus Nodule-Speciﬁc EST Library

The strategy used to construct the L. japonicus nodule-speciﬁc expression
sequence tag (EST) library was based on the mRNA differential display ﬁngerprinting
(Liang and Pardee, 1992). We applied this procedure to detect and clone transcripts that
are speciﬁcally expressed during the transition period between the formation of the
nodule structure and the onset of nitrogen ﬁxation. To minimize false positives (Liang et
al., 1993), and maximize the probability of isolating nodule-speciﬁc ESTs, we compared
the RNA proﬁles derived from four relatively late phases of nodule development. Using

the four degenerate TIZMN primers, in combination with 20 arbitrary decamers, RNA

60

species from the L. japonicus roots harvested at 7, 11, and 13 d after infection and from
21-day-old nodules were displayed and compared to the RNA proﬁles derived from 7 and
21 d old uninfected control roots. Each primer set was found to generate 80 to 150 bands.
Since a total of eighty primer combinations were used, approximately 10,000 PCR
products were displayed for every time point analyzed. A representative example of
these experiments is shown in Fig. 2.2. A visual inspection of all RNA proﬁles obtained
revealed that approximately 1.4% of the bands (137 out of 10,000) appeared to be
present in infected L. japonicus roots and/or nodules, but not in control, uninfected roots
(see Fig. 2.2). Out of 137 bands recovered from the polyacrylamide gels, 110 bands were
successfully reampliﬁed and cloned into the pKl8 vector (Pridmore, 1987). Given the
likelihood that the reampliﬁed PCR products represented a mixture of different mRNA
species (Bauer et al., 1993), 48 recombinant colonies were stored per individual PCR
product in a single microtiter plate. In addition, 8 PCR products, each represented by 12
recombinant colonies were combined in the 96-well microtiter “combo” plate and used for

differential hybridization.

Differential Hybridization and Identification of Nodule-Speciﬁc ESTs

Replica ﬁlters ﬁom each of the 14 microtiter “combo” plates, representing a total
of 110 PCR bands, were differentially hybridized with radiolabeled cDNAs derived from
nodule or root poly(A)+ RNA. This screening procedure permitted the selection of a

group of recombinant colonies, representing 39 differentially displayed PCR products,

61

which hybridized speciﬁcally with the nodule-derived cDNA probe, but not with the
probe from uninfected root. Twenty six colonies, representing 7 different PCR bands,
hybridized with probes from both uninfected roots and nodules, whereas 11 colonies
corresponding to 6 PCR products hybridized only to root cDNAs. The recombinant
colonies associated with the remaining 58 PCR bands did not show detectable signals with
either probe. An example of this differential hybr'rlization analysis is shown in Fig 2.3. In
this example, several different outcomes of the differential hybridization procedure are
shown. The top two rows of nodule-speciﬁc strong hybridization signals in Fig. 2.3
(Panel B, upper ﬁlter), representing a group of LjN50 cDNAs, illustrate an example of
most (if not all) individually picked transformants containing the same cDNA, that most
likely corresponds to a highly expressed nodulin genes. The next two rows and the last
two rows on the upper ﬁlter illustrate cases where no signal was found with either
nodule- or root-speciﬁc RNA. This suggests that the cDNAs contained in these colonies
either represent false positives or correspond to rare mRNA species. A more sensitive
approach will be used to establish their tissue speciﬁcity (see also Discussion). The third
set of rows on the upper ﬁlter illustrate a case of having isolated cDNAs that do not
appear to be nodule speciﬁc. The examples shown on the lower ﬁlter illustrate a case of
having only a limited number of positive cDNAs among the transformants analyzed
(LjN 7 7,LjN50). These results directly show the need to characterize several independent
colonies for each putative cDNA. Based on the differential hybridization analysis, we

selected 39 nodule-speciﬁc L. japonicus PCR products (ESTs).

62

Sequence Analysis of Nodule-Specific ESTs

The entire cDNA inserts of representative members of the 39 nodule-speciﬁc EST
groups were sequenced. A comparative DNA sequence analysis revealed that the 39
sequences represented 18 unique cDNA species. We refer to these as L. japonicus nodulin
(LjN) cDNAs (Table 2.1). All LjN cDNAs were found to contain DNA sequences of the
speciﬁc primer pairs used during the differential display procedure (data not shown). The
majority of the LjN cDNAs were represented by single isolates, whereas in a few cases
(e.g. LjN50, Lle32) the same mRNA species was ampliﬁed more then once, using
different primer combinations (Table 2.1; see also Discussion). Homology searches using
the BLAST algorithm (Altschul et al., 1990) indicated that 6 of the L. japonicus LjN
cDNAs shared signiﬁcant homologies to previously described nodulin genes. The LjN36,
LjN7 7, and Lle32 cDNAs showed homology to different leghemoglobin genes and
appeared to represent three different L. japonicus leghemoglobin loci (Table 2.1, Fig.
2.4A). In fact, both the DNA and predicted amino-acid sequence of LjN 7 7 were identical
to the L. japonicus leghemoglobin cDNA sequence isolated from a nodule-speciﬁc cDNA
library and used for the initial northern blot analysis (data not shown; see Fig. 2.1).

The DNA sequence of Lle3 was similar to Enod40 genes of various legume
species, and LjN] 0] shared signiﬁcant similarity with the Glycine max
coproporphyrinogen oxidase gene (Yang et al., 1993; Madsen et al., 1993). The DNA or
protein alignments for Lle3 and LjN] 01 are shown in Figs. 2.43 and 2.4C, respectively.

In addition, 7 different LjN clones revealed signiﬁcant homology to other DNA/protein

63

sequences stored in the data bases. The translated sequence of LjN3 showed strong
similarity to the Arabidopsis thaliana protein phosphatase 2C (Kuromoni and
Yarnamoto, 1994; Fig. 2.4D; Table 2.1), whereas the deduced LjN 73 protein sequence was
similar to a portion of the Zea mays cytochrome P-450 protein (Frey et al., 1995; Table
2.1). In the latter case, the 1660bp cDNA fragment was isolated from the L. japonicus
nodule-speciﬁc cDNA library and shown to encode a polypeptide (LjNP450) with
signiﬁcant similarity to different cytochrome P-450 proteins. The heme binding domain,
also known as a “signature sequence” characteristic for cytochrome P450 proteins, was
detected at the C-terminal end of the LjNP450 protein. (Nelson et al., 1993; see Fig. 2.5).
An example of the alignment between LjNP450 and the ripening-related cytochrome P-
450 CYP71A1 is shown in Fig. 2.5A (Bozak et al., 1990).

The partial amino-acid sequences deduced from LjN53 and LjN63 were short (60
a and 70 a respectively), and did not show statistically signiﬁcant matches in database
searches. However, the deduced LjN5 3 product was found to share limited similarity with
peptide/amino acid transport proteins from plants. The deduced partial amino acid
sequence of the LjN63 showed a high content (34%) of glutamic acid residues (Table 2.1).
Full-copy cDNAs corresponding to LjN53 and LjN63 were isolated and predicted to
encode a 65 kD protein (Nlj65) and a glutamate-rich (27%) protein of 192 amino acids
(Nlj21), respectively. The deduced amino-acid sequence of Nlj65 showed signiﬁcant
similarity with the peptide transporter AtPTRZ-B from Arabidopsis and other species
(Frommer et al., 1994; Fig. 2.58). The deduced amino-acid sequence of nodulin Nlj21

shared substantial similarity with the Agl3 protein from Alnus glutinosa nodules (Guan

64

et al., 1997), and the pKIWISOl gene speciﬁcally expressed during fruit development of
kiwiﬁuit (Ledger and Gardner, 1994; Fig. 2.6). The expression of the ag13 gene has been
localized to the pericycle of the vascular bundle of A. glutinosa nodules, and in infected
cells that exhibit degradation of the endosymbiont (Guan et al., 1997).

The full-copy cDNA corresponding to the L. japonicus LjN5 cDNA was isolated
and shown to encode a 15.6-kD protein (data not shown). The expression of the
corresponding gene was induced in infected cells of L. japonicus around the time of
initiation of nitrogen ﬁxation, and the deduced protein was found to share signiﬁcant

similarity with predicted or-helical domains of two related anonymous Arabidopsis ESTs

(Kapranov et al., 1997).

The remaining 7 nodule-speciﬁc LjN cDNAs did not show any signiﬁcant
similarities to DNA/ protein sequences stored in the databases. To gain more insights into
the functions of the genes represented by these ESTs, larger cDNA clones were isolated
for each of the 7 LjN clones (Table 2.1). Out of those, six cDNAs showed signiﬁcant
similarities to the sequences in the databases (Table 2.1). Only two cDNAs, LjN93 and
LjN71, were similar to the sequences with known or predicted biochemical functions,
such as tyramine hydroxycinnamoyltransferase (Farmer et al., 1999) and putative UDP-
glucose 4-epimerase (Table 2.1). The cDNAs LjN22, LjN112 and LjN80 were similar to
the genes previously identiﬁed in the screens for cDNAs expressed in a particular tissue
or under certain environmental conditions (Table 2.1). For example, LjN] 12 represented a

homolog of nodulin gene MtN24 (Gamas et al., 1996); LjN22 showed similarity to the

65

gene encoding anoxia-inducible protein (aie) from rice (Huq and Hodges, 1999) and LjN80

was similar to the embryo-speciﬁc genes AT SI and ATS3 (Nuccio and Thomas, 1999).

The product of LjN81 cDNA was highly similar to a putative Arabidopsis protein of

unknown function (Table 2.1). Finally, LjNSO cDNA encoded a novel protein. Thus, out

of 18 unique LjN clones isolated, 14 are likely to represent novel nodulin genes.

Table 2.1. Sequence similarities detected for the nodule speciﬁc LjN clones.

 

 

Clone Accession Number Size range Best homology Signiﬁcancea
name number of (bp) DNAi/protein
isolates
LjN3 AF0003 82 1 359 A. thaliana protein 1.4 x 10"9
phosphatase 2Cl
LjN13 AF0003 83 3 220-462 0. max mRNA 6.9 x 1047“
ENOD40-22

LjN53 AF000392° 1 372 transporter

LjN63 AF000402" 1 342 Glu-rich proteins

LjN73 AF000403" 1 361 z. mays cytochrome 6.2 x 10'25

P4503
LjN77 AFOOO405 2 352-511 M. sativa 3.7 x 10'55
leghemoglobin4
LjN132 AF000390 5 167-170 M. sativa 3.0 x 10-6"
leghemoglobin4
LjN36 AF000406 1 238 C. lineata 3.0 x 10-1 1'
leghemoglobin mRNA5
LjN101 AF000407 2 297-300 G. max 2.1 x 10''6
coproporphyrinogen
oxidase°

66

one

gums

LjN93
cDNA

invuz

cDNA

LjN22
cDNA

undo
cDNA

LMWI

cDNA

gmeo

cDNA

LjN81
cDNA

1J64964°

AF 000404

AF 00039 1

AF 0003 89

AF000388

AF 000408

AF 000387

AF 0003 86

AF000385

607

483

4M
n29

228-415

1014

85
570

162-570
670

428

1480

115

747

477-530
1404

67

Plant PITP-like
proteins7
0. form igenes
oxalate/formate

exchange protein8

NS

N. tabacum tyramine

hydroxycinnamoyl-

transferase9

NS

M. truncatula nodulin

O. sativa anaerobically

inducible early gene 2ll

P. horikoshii probable

A.

nnN2M°
NS

NS
NS, proline-rich
protein

NS

UDP-glucose 4-
epimerasel2
NS
thaliana embryo-
speciﬁc protein 3 '3

NS

A. thaliana

2x10-38

5x10:42

3x104

4x10:24

1x10:16

1x 10"27

hypothetical protein'4

a = probability (P-value) that such match would occur merely by chance as given by
BLAST algorithm; b = full copy or almost full copy cDNA corresponding to the
indicated EST has been deposited to GenBank under this accession number; NS = no
signiﬁcant match for the corresponding EST. The information for the full-length or partial
cDNA clones representing such ESTs is shown below each EST. (1) Kuromoni and
Yarnamoto, 1994; (2) Yang et al., 1993; (3) Frey et al., 1995; (4) Lobler and Hirsch, 1992;
(5) Ace. No. U09671; (6) Madsen et al., 1993; (7) Kapranov et al. 1997, Chapters 3&4 of
this thesis (8) Szczyglowski et al., 1998, Chapter 7 of this thesis; (9) Farmer et al., 1999;
(10) Gamas et al., 1996; (11) Huq and Hodges, 1999; (12) Acc. No. H71145; (13) Nuccio

and Thomas, 1999; (14) Acc. No. AC013483_38.

Expression Analysis of the LjN ESTs

In order to study the temporal expression pattern of selected L. japonicus ESTs,
developmental slot blot northem analyses were performed. Total RNA isolated from
uninfected L. japonicus control roots and root segments or nodules harvested 3, 7, 11, and
21 days after infection, was hybridized with a representative selection of radiolabeled
cDNA inserts (Fig. 2.7). Since the plant material used in this experiment was generated
using slightly modiﬁed grth conditions (see Materials and Methods), the LjN77-
derived insert, encoding one of the L. japonicus leghemoglobins, was used as a marker

gene for late developmental events (Table 2.1; Fig. 2.7). The mRNA species

68

corresponding to the L. japonicus leghemoglobin genes were ﬁrst detectable at 7 d after
infection, which was slightly earlier then the leghemoglobin gene expression pattern
obtained during our initial experiments (compare Figs. 1 and 7). The corresponding
mRNA accumulated gradually to a high level in 21-day-old nodules. All LjN cDNA
inserts analyzed hybridized in a nodule-speciﬁc or enhanced manner, except for LjN3,
which did not give a clearly detectable signal with either root- or nodule-derived total
RNA (see below). The expression of some of the genes analyzed could also be detected at
a very low level in the uninfected roots (e.g. LjN 101, LjN71, LjN93, LjN 73). However, a
signiﬁcant increase in the level of their corresponding mRNAs was apparent around 7 d
after infection, which was similar to the expression patterns of the other L. japonicus
ESTs analyzed (Fig. 2.7).

The expression pattern of the majority of genes analyzed resembled that of the
leghemoglobin gene (LjN 7 7), indicating that they are likely to represent L. japonicus late
nodulin genes. Interestingly, mRNA of the coproporphyrinogen oxidase homolog LjN I 01
accumulated in a slightly different fashion than the mRNA of the leghemoglobin gene
LjN 7 7. The gene corresponding to LjN 101 appeared to be induced to a high level around 7
d after inoculation, with no signiﬁcant changes in the steady-state level of mRNA
accumulation during the later stages of nodule development (Fig. 2.7).

The mRNA corresponding to the LjN13, an ENOD40 homolog (Table 2.1 and Fig.
2.4B), was weakly detectable in uninfected roots, but the hybridization signals were
signiﬁcantly enhanced in infected roots and fully developed 21-day-old nodules. The

poly (A)+ fraction of total RNA from uninfected roots and nodules was used to further

69

analyze tissue-speciﬁc expression of LjN3, the putative protein phosphatase 2C
homologue. The LjN3 insert hybridized with both root and nodule mRNA species of
approximately 1600 nt in length (Fig. 2.8). However, the level of the corresponding
mRNA in L. japonicus nodules was found to be approximately six times higher than in
uninfected roots, conﬁrming the nodule-speciﬁc/enhanced pattern of LjN3gene expression.

Since all EST sequences analyzed were generated using PCR-based procedures,
their plant (L. japonicus) origin needed to be conﬁrmed using Southern blot analysis. All
LjN cDNA inserts hybridized speciﬁcally with L. japonicus genomic DNA, but not with

total DNA isolated from R. loti strain NZP2235 (data not shown).

Further Characterization of the L. japonicus EST Library

To further characterize the L. japonicus EST library, we employed automated
DNA sequencing. Two randomly selected recombinant colonies, both corresponding to a
single differentially displayed PCR product, were used for DNA sequencing analysis. The
recombinant colonies were selected from the collection of L. japonicus EST “combo”
plates, based on their lack of hybridization with the radiolabeled root and nodule cDNA
probes used during the differential hybridization procedure. The nucleotide sequence of
142 cDNA inserts, representing a total of 71 PCR products, was established. DNA
sequence comparisons showed that they corresponded to 69 unique cDNA sequences
(data not shown). The EST nucleotide sequences obtained were compared to the

nucleotide and protein sequences in the data bases by BLASTN and BLASTX searches,

70

respectively (Altschul et al., 1990). Nine of 69 EST sequences analyzed showed a
moderate to high level of similarity to DNA/protein sequences stored in the data bases. A
summary of the results of this analysis is shown in Table 2.2. We refer to these clones as
L. japonicus LjEST cDNAs. All cDNAs, except LjEST59, appeared to encode unique
enzymatic functions, including subtilisin-like protease (Ribeiro et al., 1995),
adenosylosuccinate synthetase (Zeidler et al., 1993), tyrosine decarboxylase (Maldonado-
Mendoza et al., 1996), dehydroquinate dehydratase/shikimate dehydrogenase (Booner
and Jensen, 1994), heme oxygenase (Shibahara et al., 1985), chalcone synthase
(Goormachtig et al., 1995), and glutamine phosphoribosylpyrophosphate
amidotransferase (Kim et al., 1995). However, their speciﬁc involvement in the
nodulation process needs to be conﬁrmed using RNA blot hybridization or other more
sensitive protocols, such as RT-PCR. The remaining 60 EST sequences obtained via the
random sequencing approach did not show any signiﬁcant homologies or similarities and

were therefore classiﬁed as anonymous L. japonicus ESTs.

Table 2.2. Sequence similarities detected for randomly sequenced EST clones.

 

 

Clone Accessio Number Size Best homology Signiﬁcance
name n number of isolates (bp) DNA‘lprotein
LjEST38 AF000393 1 411 Arabidopsis thaliana 3.8 x 10''8

71

subtilisin-like proteasel

LjEST 58 AF 000394 1 527 yeast adenylosuccinate 2.7 x 10'7
synthetase2
LjEST59 AF000395 1 258 unknown Trypanosoma 5.4 x 10'8

brucei protein3
LjEST 66 AF000396 1 245 Papaver somniferum 1.5 x 10'13
tyrosine decarboxylase4
LjEST103 AF000398 1 131 Nicotiana tabacum 1.8 x 10"?

dehydroquinase/shikimate

dehydrogenase mRN A
3’end5
LjEST105 AF000399 1 238 Rat heme oxygenase° 3.4 x 10:5
LjESTII8 AF000400 1 296 Sesbania rostrata chalcone 1.7 x 10'H
reductase7
LjESTIZO AF000401 1 192 Glycine max glutamine 8.0 x 10:8

phosphoribosylpyrophos

phate amidotransferase8

 

(1)Ribeiro et al., 1995; (2) Zeidler et al., 1993; (3) Ace. No. U05313; (4) Maldonado-
Mendoza et al., 1996; (5) Booner and Jensen, 1994; (6) Shibahara et al., 1985; (7)

Goorrnachtig et al., 1995; (8) Kim et al., 1995.

72

2.5. DISCUSSION

We have previously reported the successful application of the mRNA differential
display technique for the identiﬁcation and isolation of early nodulin genes from Sesbania
rostrata stem nodules (Goormachtig et al., 1995). Here, we employed the RT-PCR—based
differential display procedure to identify a range of molecular markers associated with
relatively late developmental events during L. japonicus nodule organogenesis. A well-
characterized nodulin gene, namely the leghemoglobin gene, was used to deﬁne the late
stages in L. japonicus nodule morphogenesis. Using 80 primer combinations, the proﬁles
of approximately 10,000 PCR products were analyzed and 110 differentially displayed
nodule-speciﬁc or enhanced bands were successfully reampliﬁed and cloned. Thus, a
library of differentially displayed L. japonicus ESTs was established. The differential
hybridization procedure revealed relatively abundant mRNA species and allowed the
selection of 39 nodule-speciﬁc PCR products, representing 18 unique cDNA sequences.
For the purpose of this paper, the term “speciﬁc” was used merely to reﬂect the
signiﬁcant differences in the expression levels observed between the uninfected roots, and
infected roots or nodules.

The apparent redundancy found among some of the isolated sequences (between 2
and 9, see Table 2.1) was in part due to the different combinations of primers used during

the differential display procedure. Interestingly, in most cases analyzed, different

73

positions of leMN primers at the 3’ end of the mRNA, in combination with a single
arbitrary decamer primer, gave rise to multiple products derived ﬁ'om the same mRNA
species (e.g. LjN13, LjNIOI, LjN8I, Tablel). This ﬁnding may be explained by the
presence of multiple poly(A) sites in a given plant mRNA species (Wu et al., 1995). In
the case of the LjN50 cDNA group, in addition to differences in the position of the
TIZMN primers, three different arbitrary decamers contributed to multiple independent
isolations of the respective cDNA species (data not shown). The sequence analysis of
nine partial cDNAs belonging to the LjN50 group revealed that all of them shared highly
conserved 3’-terminal sequences of approximately 220 bp. However, signiﬁcant
differences in the corresponding DNA sequences at the 5’ ends were found, indicating
that they may, in fact, correspond to related, but not identical, genes in the L. japonicus
genome (data not shown).

The comparative sequence analysis of representatives of all 18 cDNA groups
established that the majority of them are likely to correspond to novel nodule-speciﬁc
genes. This group of L. japonicus ESTs includes putative homologues of protein
phosphatase 2C and cytochrome P450, both candidates for proteins with regulatory
functions. The protein phosphatase activity in plants, of which the least well-
characterized member is PP2C protein serine/threonine phosphatase, has been implicated
in such processes as signal transduction, hormonal regulation, mitosis, and control of
carbon and nitrogen metabolism (Smith and Walker, 1996). Cytochrome P450 enzymes,
on the other hand, are membrane-bound, heme-containing enzymes, implicated in a

variety of biosynthetic reactions (Nelson et al., 1993; Frey et al., 1995). In plants they

74

are typically involved in the synthesis of chemically diverse secondary metabolites, often
involved in defense mechanism, synthesis of plant hormones, or cell-wall related
substances (Holton et al., 1993; Winkler and Helentjaris, 1995; Frey et al., 1995; Szekeres
et al., 1996). The speciﬁc functions of these interesting new nodulin genes (LjN3 and
LjN73) in L. japonicus nodules remain to be elucidated.

cDNA clones that shared signiﬁcant similarity to already characterized late-
nodulin genes from different legume species were also identiﬁed. The latter group includes
L. japonicus leghemoglobin genes and a putative homologue of soybean
coproporphyrinogen oxidase, an enzyme catalyzing the oxidative decarboxylation of
coproporphyrinogen III to protoporphyrinogen IX, in the heme and chlorophyll
biosynthetic pathway(s) (Madsen et al., 1993). The latter ﬁnding supports the notion
that speciﬁc plant host functions participate in the synthesis of heme, which is needed in
nodules for increased hemoprotein biosynthesis (Madsen et al., 1993).

The kinetics of mRNA accumulation of the L. japonicus ESTs analyzed closely
resemble the pattern observed for leghemoglobin gene expression in L. japonicus nodules,
indicating that a majority of them are almost certainly related to late stages in nodule
development. Since the goal of our analysis was to identify molecular markers associated
speciﬁcally with late developmental stages in nodule organogenesis, our results illustrate
the advantage of using nodules with a determinate developmental pattern for late-nodulin
expression analysis, as opposed to indeterminate nodules, where successive early and late

developmental stages coexists.

75

Since a majority of the L. japonicus EST clones did not hybridize with the root- or
nodule-speciﬁc cDNA probes used during differential hybridization screening, we
employed a random sequencing approach to further characterize the remaining
differentially displayed L. japonicus ESTs. This resulted in the generation of 69 unique
cDNA sequences. Only nine of the cDNAs shared similarity with protein sequences from
the data bases (see Table 2.2). Since selection of all of these sequences was based
exclusively upon the observed differential display patterns, it is not certain whether they
indeed correspond to nodule-speciﬁc or nodule-enhanced genes. Clearly, more in-depth
analyses are needed to unambiguously establish their tissue and or cell speciﬁcity.
However, it is important to note that at least three of them, namely subtilisin-like
protease, chalcone reductase, and glutamine phosphoribosylpyrophosphate
amidotransferase have previously been reported to be induced during nodulation (Ribeiro
et al., 1995; Goormachtig et al., 1995; Kim et al., 1995). Several roles for lipoxygenases
(LOX) in plant-microbe interactions including symbiotic nodule formation have also been
suggested (Croft et al., 1993; Veronesi et al., 1996; Gardner et al., 1996; Perlick et al.,
1996)

In summary, of 110 L. japonicus differential ampliﬁcation products, 88 unique
partial cDNA sequences were obtained. Nineteen of these were further analyzed and
shown to correspond to nodule-speciﬁc genes, with the majority of them likely to encode
novel nodule-speciﬁc functions. Twenty-ﬁve L. japonicus ESTs showed varying degrees
of similarity to different DNA/protein sequences stored in the data bases. For most of

these, solid predictions about their activities could be made. However, their speciﬁc roles

76

in the L. japonicus nodules need to be further analyzed. A group of 59 EST sequences
failed to reveal signiﬁcant homology to any sequences in the data bases, and their tissue
speciﬁcity has not been ﬁrmly established. We are currently analyzing these ESTs with
regard to their nodule-speciﬁc expression. We assume that some of them may, in fact,
correspond to a group of weakly expressed genes with speciﬁc functions during late
stages of L. japonicus nodule development. In fact, a preliminary analysis of 16 different
LjEST clones revealed that 8 of them corresponded to low-abundant mRNAs induced in
L. japonicus nodules (Kasiborski et al., unpublished). The novelty and diversity of the
isolated nodule-speciﬁc genes should greatly facilitate further molecular analyses of the
late stages of nodule development. The collection of ESTs reported here will also be an
indispensible tool for the development of a L. japonicus genetic map, an essential step

toward future map-based cloning of symbiosis-speciﬁc genes in this model legume plant.

77

control

7 11 13 17 21 days after infection
LjEnod2

nglbI

 

 

Figure 2.1. RNA gel blot analysis of L. japonicus LjEnod2 and leghemoglobin
(ngIbI) gene expression. 10 pg of total RNA isolated from 7-day-old uninfected roots
(control), and root segments or nodules harvested 7, ll, 13, and 21 d aﬂer infection were
analyzed. 0t-32P-radiolabeled cDNAs encoding L. japonicus LjEnod2 (upper panel), and

leb1 (lower panel) were used as molecular probes.

78

Figure 2.2. Developmental mRNA differential display. An example is
shown of mRNA ampliﬁcation proﬁles from 21 (lane 1) and 7 (lane 2)
days old uninfected control roots, and root segments and nodules
harvested 7 (lane 3), 11 (lane 4), 13 (lane 5), and 21 (lane 6) days after
infection. The decamer primer AP9 (CGTGGCAATA), in combination
with four 3’- anchoring primers (TIZMG, leMA, leMT, and TlgMC,
respectively), was used to generate the RNA proﬁles shown. The dots (°)
indicate putative nodule-speciﬁc bands. The differentially displayed band
generated using AP9/T12MC primer combination (see lane 6) corresponds
to the LjN 7 7 EST and encodes a L. japonicus leghemoglobin (see Table

2.1).

79

L Primer: AP9

 

R Primer: TnMG TnMA TnMT TnMC
123456 123456 123456 123456

       

~300 nt >

80

Figure 2.3. Differential colony hybridization. The result of differential
hybridization, using recombinant colonies derived from two “combo” microtiter
plates is shown. The ﬁlters shown in Panels A and B were hybridized using
radiolabeled cDNA derived from control 21-day-old uninfected roots and 21-day-

old L. japonicus nodules, respectively. For further details see text.

81

 

 

— LjN50

— LjN50

:1 no signal
:I non-specific

:I no signal

— LjN77

— LjN50
— LjN132

         

5“» _- i. I

 

. a} ' I;
6%

r '
04?:th .

 

 

 

 

Root Nodule

82

Figure 2.4. DNA/protein sequence alignments of LjN77, LjN13, LjN101, and
LjN3. The alignments shown were performed using the BESTFIT program from
the GCG package (Madison, WI). Vertical bars indicate identical amino acids,
whereas colons and periods represent conservative and semiconservative
substitutions, respectively. (A) Amino acid alignment of LjN77 with Medicago
sativa leghemoglobin (Mslgb; Lobler and Hirsh, 1992): 55% identity and 68%
similarity within an overlap of 122 amino acids. (B) DNA alignment of Lle3
with Glycine max ENOD40-2 (GmENOD40-2; Yang et al., 1993): 74.5% identity
within an overlap of 410bp; the homology region 11 is underlined (see Vijn et al.,
1995). (C) Amino-acid alligrnent of LjN101 with Glycine max
coproporphyrinogen oxidase (Gmcopro; Madsen et al., 1993): 100% of amino-
acid sequence identity over 28 amino acids. (D) Amino-acid alligment of L.
japonicus LjN3 with Arabidopsis thaliana protein phosphatase 2C (Atpp2C;
Kuromoni and Yamamoto, 1994): 50% identity and 63% similarity within an

overlap of 79 amino acids.

83

WODI 0 -2

MOD! 0 -2

WODI 0 -2

GIINODIO-Z

MOD! 0 -2

MODl 0 -2

Gncopro

AtppZC

AtppZC

AIVFCSTPOYWKKPQ ........ LLKTCSPSKGFWTHAQSSTP. . . SPML
1:1 :I . :...l. :I....:.|l::. ..l.. 11.]
ALVNSSWESFKQNPGNSVLFYTIILEKAPAAKGMFSFLKDSAGVQDSPKL

..... KRFLDJHAMAAQLLAKGEVTLADASLGAVHVQKAVADPHFAVVKE

| I :. IIII l.l:l.|:|l.||l:l:ll:l.lllllllll
QSHAEKVFGMVRDSAAQLRATGGVVLGDATLGAIHIQKGVVDPHFAVVKE

ALLKTVQAAVGDKWSEELSTAWGVAYDGLAAAIKKAM 122
Illllz... lllll|l|.||l:l|||:ll-l||l|l
ALLKTIKEVSGDKWSEELNTAWEVAYDALATAIKKAM 145

gagagaagctttggctacagcctggcgaaaccggcaagtcac.agaaagg
III II llllllllllllllll lllllllllllllll l lllll
... _ 09. 19 _. oq -. .- ._-. .

O

     
   
   

caatggaccccattaggtttcttatggctatgtatgaatgactcttttgt
llllllll II!!! II I Illlllllllll II III
gag:ggggggcgggggggggtggggggcggtggg..,,.,g;g;;catgt

agttcttcttcaagtagaatgtaataaacaaaaatgttcttcttccttt.

llllllllll Illlllllllllll ll II III lllllllll
agttcttcttgctgtagaatgtaataataaacaaagttggtcttcctttt

.......... gaggtggtgttcattcatatcacccaattttgcagctgac
II 1 II I I II II III llllllllllll
gagaagttaccagcttttgctgtccaaaattactcaa.tttgcagctgac

tagactcccagtttgctcttcagtttcttcgaagatgagtaggtaggtaa
llll | ll 11 Illllllllllll | Illllllllllllll ll
tagaattcc.tttctctcttcagtttct..gcagatgagtaggtaggcaa

.ttatgatcaccttgtctcttctttttctgtg...tctgttttccttttc

ll lllllll ll llll III II lllll III! III
tttgtgatcac....tcccttcccttttcatgtcttctgtgttccctttt

tcatgtgtgtatggttgttgtcttgtggcctatatgcaataatagtattt
llll Ill II I I II II III III! |||| l
ccatgcttgtttgtgttgttagttatgacct.tatgaggaaataaaagaa

gagtgatgtttttgttccttcaagtgtaggattgtgttttgtagaagttg
III ll 1 II III III IIIIIIIII II I I III II
tagtacaattctagtccct..cagtttaggattgtattctattgaacttt

at 410

II
at 537

RWEYDHKPEEGSEEWKLLDACINPKEWI 28
lll|||||||||||llllllllllllll
RWEYDHKPEEGSEEWKLLDACINPKEWI 384

SVTKRSSKDEFLILASDGLWDVISSEMACQVVRKCLNG.QIRRICNENQS
.l|.l...ll Illllllllllt..l ll.l.l.|l.l . :z..:.
TVTDRTDEDECLILASDGLWDVVPNETACGVARMCLRGAGAGDDSDAAHN

RASEAATLLAEIALAKGSRDNTSVIVIELR 79

.|:l| ll..:l|l:.l.|l.|l:l::|l
ACSDAALLLTKLALARQSSDNVSVVVVDLR 390

84

4O

58

85

108

73

202

123

246

172

296

212

345

262

392

308

438

358

487

408

535

49

360

Figure 2.5. Protein sequence alignments of LjNP450 and LjN65. The
alignments shown were performed as describe in Figure 4. (A) L. japonicus
LjNP450 protein with cytochrome P-450 CYP71A1 from avocado fruit
(Bozak at al., 1990): 46% identity and 68% similarity; the heme-binding
domain for cytochrome P-450 (the signature sequence EXXQXXXQXQ;
Nelson et al., 1993) is underlined. (B) L. japonicus Nlj65 with Arabidopsis
thaliana AtPTR2-B transporting protein; 38% identity and 59% similarity

(Frommer et al., 1994).

85

CYP71A1

CYP71A1

CYP71A1

CYP71A1

CYP71A1

CYP71A1

CYP71A1

CYP71A1

CYP7131

1m

CYP7111

111112452

CYP71A1

MAILVSLLF

VALMILRKNLKKPDSIPNIPPGPWKLPIIGSIPHLVGSPPHRKLRD
:.:::|: | .I.:. II:|I:I .IIIII.:.:| I. |I|.I|.
LAIALTFFLLKLN.EKREKKPNLPPSPPNLPIIGNLHQL.GNLPHRSLRS

LAKKYGPLMHLQLGEVIFIIVSSAEYAKEVMKTHDVTFASRPRSLFTDIV

II III: I' II :III II L L II|I IIIII. .
LANELGPLILLHLGHIPTLIVSTABIAEEILKTHDLIFASRPSTTAARRI

FYGSTDIGFSPYGDYWRQVRKICNVELLSMKRVQSLWPIREEEVKNLIQR
II::II::II|II:I||I|II|I :IIII:||I.|.:.IIIII| 2::l
FYDCTDVAFSPYGEYWRQVRKICVLELLSIKRVNSYRSIREEEVGLMMER

I..ASEEGSVVNLSQAIDSLIFTITSRSAFGKRYMEQEE....FISCVRE
I -:--I.-IIII: : I ...-I |I||:| :1II I . . I
ISQSCSTGEAVNLSELLLLLSSGTITRVAFGKKYEGEEERKNKFADLATE

VMKLAGGFNIADLFPSAKWLENLTRMRSKFEYLHQKMDRILETIIDDHKA

. .l Izl ::I.I|l |:: II I .::. I..:| ... :Illl
LTTLMGAFFVGDYFPSFAWVDVLTGMDARLKRNHGELDAFVDHVIDDHLL

NSRTKEGQVEGGEEDLIDVLLKYENSSTDQDFHLTIRNIKAILFDIFIAG
II. :I :.::.l|:||||..:..l. :.|I. .I:||:::|:| :l
.SRKANGSDGVEQKDLVDVLLHLQKDSS.LGVHLNRNNLKAVILDMFSGG

SETSATTINWTMAEMMKDPILLKKAQDEVREIFQRRGKVDETCIYELKYL

:I.|.I::I.III::|.I '.|I|:III z..:::||:|. :.:I.|I
TDTTAVTLEWAMAELIKHPDVMEKAQQEVRRVVGKKAKVEEEDLHQLHYL

KAFINEVLRLHPPGPLVF.RECRQACEINGYHIPAKSTVLVNTFAIGTDS
I :|.I.||III.:II:. II: ... |.|II|III. I::I.:|II I.
KLIIKETLRLHPVAPLLVPRESTRDVVIRGYHIPAKTRVFINAWAIGRDP

KYWAEPERFCPERFIDSSIDYKGTNFEHLPFQAQRRICPQINYGMANVEL

I I.::| I I|lI::.|:I:|I :I: :IIIIIII IIII.:I:..II:
KSWENAEEFLPERFVNNSVDFKGQDFQLIPFGAGRRGCPGIAFGISSVEI

VLALLLYHFDWTLPKGIKNEDLDLTEEFGVTVSKKEDLCLIPSISHPLPST

II III I:|.I| .III|:.|..I:II .I .I |::.
SLANLLYWFNWELPGDLTKEDLDMSEAVGITVHMKFPLQLVAKRHLS

86

9

46

57

96

107

146

157

190

207

240

257

290

305

340

355

389

405

439

455

490

502

uliﬁi M
AtPTRZ-B MGSIEEEARPLIEEGLILQ

NLjﬁﬁ EGKGYTLDGTVDLAGRPVLSSLTGKQKACTYILVYRVLERFAYYGVGANL
|.I l. II.II:.|.I.|.. II. III.:II. . II:III|:::II
AtPTRZ-B EVKLYAEDGSVDFNGNPPLKEKTGNWKACPFILGNECCERLAYYGIAGNL

nljgg VNFMTTQLNKDVVSSITSFNNWSGLATLTPILGAYIADTYTGRFWTITFS
':II.I:.: II. I....I I . III::|I :II.| II:III.
.AtPTRZ-B ITYLTTKLHQGNVSAATNVTTWQGTCYLTPLIGAVLADAYWGRYWTIACF

u;j§§ LLIYAIGLVLLVLTTTLKSLRPA.CENGICREASNLQVALFYTSLYTIAV
II II: |.I...:..I:|l I .::I..I. I I:I:.:II II:
.AtPTRZ-B SGIYFIGMSALTLSASVPALKPAECIGDFCPSATPAQYAMFFGGLYLIAL

ﬂljﬁﬁ GSGAVKPNMSTFGADQFDDFRHEEKEQKVSFFNWWAFNGACGSLMATLFV
|.I::II :I.I|I|IIII I: .I.III|I: |. ..I.I:..
.AtPTRZ-B GTGGIKPCVSSFGADQFDDTDSRERVRKASFFNWFYFSINIGALVSSSLL

uljﬁﬁ VYIQEKNGWGLAYSLSAIGFLLSSIIFFWGSPVYRHKSRQARSPSMNFIR

I: III IIII::::..: : I. II: I: II . :.: II ..
.AtPTRZ-B VWIQENRGWGLGFGIPTVFMGLAIASFFFGTPLYRFQ. KPGGSPITRISQ

£1155 VPLVAFRNRKLQLPCNPSELHEFQLNYYISSGARKIHHTSHFSFLDRAAI
I.:..II...:.:I ::. I.| I . .|.II|.||.. :II:I|:
.AtPTRZ-B VVVASFRKSSVKVPEDATLLYETQDKNSAIAGSRKIEHTDDCQYLDKAAV

Eljﬁﬁ ...... RESNTDLSNPPCTVTQVEGTKLVLGMFQIWLLMLIPTNCWALES

. I . IIIIIII: l::: II II :I ...:I I
.AtPTRZ-B ISEEESKSGDYSNSWRLCTVTQVEELKILIRMFPIWASGIIFSAVYAQMS

ELjﬁﬁ TIFVRQGTTMDRTLGPKFRLPAASLWCFIVLTTLICLPIYDHYFIPFMRR
I:II.|I .I: .:I. |.II:I.| .I . ..zI :|:II::::I: I:
.AtPTRZ-B TMFVQQGRAMNCKIGS.FQLPPAALGTFDTASVIIWVPLYDRFIVPLARK

Eljﬁﬁ RTGNHRGIKLLQRVGIGMAIQVIAMAVTYAVETQRMSVIKKHHIAGPEET
II .:I:. II III: : I: II II I: : .. : ......
.AtPTRZ-B FTGVDKGFTEIQRMGIGLFVSVLCMAAAAIVEIIRLHMANDLGLVESGAP

Eljﬁﬁ VPMSIFWLLPQNIILGVSFAFLATGMLEFFYDQSPEEMKGLGTTLCTSCV

II:I::I :II :III.. .I. .I IIIIIIIII:.I::I...I.
.AtPTRZ-B VPISVLWQIPQYFILGAAEVFYFIGQLEFFYDQSPDAMRSLCSALALLTN

31165 AAGCYINTFLVTMI ....... DKLNWIGNNLNDSQSRITIMPFFSVISAL

| I.|:..:::I: :. .II: :III. . " ..... I
.AtPTRZ-B ALGNYLSSLILTLVTYFTTRNGQEGWISDNLNSGHLDY. FFWLLAGLSLV

ﬂljﬁﬁ NFGVFLWVSSGYIYKKENTSTTEVHDIEMSAEKTVKY

I::I::: .. I II
.AtPTRZ-B NMAVYFFSAARYKQKKASS

87

l

19

51

69

101

119

150

169

200

219

250

268

300

318

344

368

394

417

444

467

494

517

503

567

574

585

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

153.121 ATA VASA SALH - TTKA LKDHEEK E32
A913 ATVEVVSA TALP - TIEIEPIKVHETI 32
pKIwI501 ATVEVTPA TAL TAD VTKP E Q AA 33
8.1321 vv PE EPVSEKTK BEE vr I I' 65
A913 vv1 PPA A sr - . 64
1916111501 VAA P EPV E APE v AI'II 66
111.121 - -EN SIEAET EVVEEV ‘vrvro K I 96
A913 PKPE PLEVET EVVEEA VT--- P If 94
pmso1 AKEIP A ,EVVEE Q P--- 92
m; g' EATETK ' ‘I' ' Q E 129
A913 I I' ETPE A EE EIEATDS 127
pKIWISOl ABA. A GKAI ' Eu. UK A 125
211121 I ASKEE v EPEAKEEV AP& [(162
A913 - I P ”EIPEA- PK AE ----- E VA v154
pKI'WISOl -I'- E’TDAP Pl E ----- EP PEAP153
man K EEKLDTE DE KEiEEVI VAPAEKTE- 1%
A913 P 155va P IA EKVG EAPVEKADIa 185
pumso1 AVG EPEAK KP EAV EA TEVPVQ<TEE 184

 

 

 

 

 

 

 

 

 

Figure 2.6. Amino-acid sequences comparison of Nlj21, Ag13, and PKIWI501 . The
deduced amino-acid sequence of the L. japonicus nodulin Nlj21 is compared with

A. glutinosa Ag13 and kiwifruit pKIW1501 (Guan et a1.,1997; Ledger and Gardner, 1994).

The identical residues, as well as conservative substitutions, are boxed.

88

Figure 2.7. Developmental slot blot northern analysis. 10 pg of total

RNA isolated from uninfected (control), and infected root segments and
nodules at the time points indicated were transferred onto nitrocellulose
ﬁlters using a manifold slot blot vacuum apparatus and hybridized with
radiolabeled cDNA inserts corresponding to 14 unique LjN ESTs. The
cDNAs LjNIIZ and LjN20 were later found to be derived from the same
mRNA species (data not shown). The LjENOD2 and leghemoglobin
(LjN 7 7) cDNAs were included as nodule-developmental markers, whereas

the 18S cDNA was used as a loading control

89

uninfected days after
roots infection

371121371121

llllllllﬂSSrRNA
I I . 4 LjENOD2
.' <LjN77
. <LjN13
I" <LjN53
I ' <LjN63
. <LjN73
' .~ ‘ (LjN101
I_ <LjN48
I I . «LjN93
t <LjN20
I <LjN22
I 8‘ <LjN50
I | . <LjN7I
5 | <LjN80
f "1 <LjN81
I if 'i <LjN112

90

1 2

, W;
.4; L.
. i 1

,

.

- .. s

5.

d" . . u

’5'.» W? E

, r . ‘1

m mm: ;3 eIF4A

Figure 2.8. RNA gel blot hybridization of LjN3. 4 pg of poly(A)+ mRNA from control

uninfected roots (lane 1) and 21-day-old nodules (lane 2) were separated under

denaturating conditions and hybridized with the a32P- labeled cDNA insert from EST

LjN3. The eIF 4A cDNA was used as a loading control (Taylor et al., 1993).

2.6. ACKNOWLEDGEMENTS:
We thank Brian MacSpadden Gardener for comments and helpﬁil discussions. We also

thank Kurt Stepnitz for help in preparing the Figures.

91

2.7. LITERATURE CITED:

Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ (1990) Basic local alignment
search tool. J Mol Biol 215: 403-410.

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

Booner CA, Jensen RA (1994) Cloning of cDNA encoding the bifunctional
dehydroquinase-shikimate dehydrogenase of aromatic-amino-acid biosynthesis in
Nicotiana tabacum. Biochem J 302: 11-14

Bozak KR, Yu H, Sirevag R, Christoffersen, RE (1990) Sequence analysis of ripening-
related cytochrome P-450 cDNAs from avocado fruit. Proc Natl Acad Sci USA 87: 3904-
3908.

Broughton WJ, Dilworth MY (1971) Control of leghemoglobin synthesis in snake
beans. Biochem J 125: 1075-1080

Carlson RW, Price NPJ, Stacey G (1995) The biosynthesis of rhiozobial lipo-
oligosaccharide nodulation signal molecules. Mol. Plant-Microbe Interact 7: 684-695

Cheon C-III, Lee N-G, Siddique A-BM, Bal AK, Verma DPS (1993) Roles of plant
homologs of Rablp and Rab7p in the biogenesis of the peribacteroid membrane, a
subcellular compartment formed de novo during root nodule symbiosis. EMBO J 12:
4125-4135

Church GM, Gilbert W (1984) Genomic sequencing. Proc Natl Acad Sci USA 81: 1991-
1995

Cook D, Dreyer D, Bonnet D, Howell M, Nony N, VandenBosch K (1995) Transient
induction of a peroxidase gene in Medicago truncatula precedes infection by Rhizobium
meliloti. Plant Cell 7: 43-55

Crespi MD, Jurkevitch E, Poiret M, d’Aubenton-Carafa Y, Petrovics G, Kondorosi
E, Kondorosi A (1994) en0d40, a gene expressed during nodule organogenesis, codes for a
non-translatable RNA involved in plant growth. EMBO J 13: 5099-5112

Croft KPC, Juttner F, Slusarenko AJ (1993) Volatile products of the lipoxygenase

pathway evolved from Phaseolus vulgaris (L.) leaves inoculated with Pseudomonas
syringae pv phaseolicola. Plant Physiol 101: 13-24

92

de Bruijn FJ, Schell J (1992) Regulation of plant genes speciﬁcally induced in
developing and mature nitrogen-ﬁxing nodules: cis-acting elements and trans-acting
factors. In DPS Verma, ed, Control of Plant Gene Expression, CRC Press, Boca Raton,
pp 241-258

Delauney AJ, Verma DPS (1988) Cloned nodulin genes for symbiotic nitrogen ﬁxation.
Plant Mol Biol Rep 6: 279-285

Farmer MJ, Czernic P, Michael A, Negrel J (1999) Identiﬁcation and characterization
of cDNA clones encoding hydroxycinnamoyl-COA:tyramine N-
hydroxycinnamoyltransferase from tobacco. Eur J Biochem 263: 686-694

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

Frey M, Kliem R, Saedler H, Gierl A (1995) Expression of a cytochrome P450 gene
family in maize. Mol Gen Genet 246: 100-109

Frommer WB, Hummel S, Rentsch D (1994) Cloning of an Arabidopsis histidine
transporting protein related to nitrate and peptide transporters. F EBS Lett 347 :185-189

Gardner CD, Sherrier DJ, Kardailsky IV, Brewin NJ (1996) Localization of
lipoxygenase in the lumen of infection threads. Mol Plant-Microbe Interact 9: 282-289

Goormachtig S, Valerio—Lepieniec M, Szczyglowski K, Van Montagu M, Holsters
M, de Bruijn FJ (1995) Use of differential display to identify novel Sesbania rostrata
genes enhanced by Azorhizobium caulinodans infection. MOI Plant-Microbe Interact 8:
816-824

Guan Ch, Akkermans AD, van Kammen A, Bisseling T, Pawlowski K (1997) ag13
is expressed in Alnus glutinosa nodules in infected cells during endosymbiont degradation
and in the nodule pericycle. Physiol Plant (in press)

Hall TC, Ma Y, Buchbinder BU, Pyne JW, Sun SM, Bliss FA (1978) Messenger
RNA for G1 protein of French bean seeds: cell-free translation and product
characterization. Proc Natl Acad Sci USA 75: 3196-3200.

Handberg K, Stiller J, Thykjear T, Stougaard J (1994) Transgenic plants:

Agrobacterium mediated transformation of the diploid legume Lotus japonicus. In: Cell
Biology: A Laboratory Handbook, Celis JB, ed, 1: 1 19-127

93

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

Heard J, Dunn K (1995) Symbiotic induction of a MADS-box gene during development
of alfalfa root nodules. Proc Natl Acad Sci USA 92: 5273-5277

Holton TA, Brugliera F, Lester DR, Tanaka Y, Hyland CD, Menting JG, Lu CY,
Farcy E, Stevenson TW, Cornish EC (1993) Cloning and expression of cytochrome
P450 genes controlling ﬂower colour. Nature 366: 276-269

Huq E, Hodges TK (1999) An anaerobically inducible early (aie) gene family from rice.
Plant Mol Biol 40: 591-601

Jiang Q, Gresshoff PM (1993) Lotus japonicus: a model plant for structure-function
analysis in nodulation and nitrogen ﬁxation. Curr Topics Plant Mol Biol 2: 97-110

Jiang Q, Gresshoff PM (1997) Classical and molecular genetics of the model legume
Lotus japonicus. Mol Plant-Microbe Interact 1: 59-68

Jarvis BDW, Pankhurst CE, Patel JJ (1982) Rhizobium loti, a new species of legume
root nodule bacteria. InterJ Syst Bacteriol 32: 378-380

Kapranov P, de Bruijn FJ, Szczyglowski K (1997) A novel, highly expressed late
nodulin gene LjNOD16 from Lotus japonicus. Plant Physiol 113: 1081-1090

Kim JH, Delauney AJ, Verma DPS (1995) Control of de novo purine biosynthesis
genes in ureide-producing legumes: induction of the glutamine
phosphoribosylpyrophosphate amidotransferase gene and characterization of its cDNA
from soybean and Vigna. Plant J 7: 77-86

Kuromoni T, Yamamoto M (1994) Cloning of cDNAs from Arabidopsis thaliana that
encode putative protein phosphatase 2C and a human Dr-l like protein by transformation
of the ﬁssion yeast mutant. Nucleic Acids Res 22: 5296-52301

Ledger SE, Gardner RC (1994) Cloning and characterization of ﬁve cDNAs for genes
differentially expressed during fruit development of kiwifruit (Actinidia deliciosa var.
deliciosa). Plant Mol Biol 25: 877-886

7 Liang P, Averboukh L, Pardee AB (1993) Distribution and cloning of eukaryotic

mRNAs by means of differential display: reﬁnements and optimization. Nucleic Acids
Res 21: 3269-3275

94

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

Lobler M, Hirsch AM (1992) An alfalfa (Medicago sativa L.) cDNA encoding acidic
leghemoglobin (MsLb3). Plant Mol Biol 20: 733-736

Madsen O, Sandal L, Sandal NN, Marcker KA. (1993) A soybean
coproporphyrinogen oxidase gene is highly expressed in root nodules. Plant Mol Biol 23:
35-43

Maldonado-Mendoza IE, Lopez-Meyer M, Galef JR, Burnett, RJ, Nessler CL (1996)
Molecular analysis of a new member of the opium poppy tyrosine/3,4-
dihydroxyphenylalanine decarboxylase gene family. Plant Physiol 110: 43-49

Marmur J (1961) A procedure for the isolation of deoxiribonucleic acid from
microorganisms. J Mol Biol 3: 208-218

Matvienko M, Van De Sande K, Yang W-C, Van Kammen A, Bisseling T, Franssen
H (1994) Comparison of soybean and pea ENOD40 cDNA clones representing genes
expressed during both early and late stages of nodule development. Plant Mol Biol 26:
487-493

Miao G-H, Hong Z, Verma DPS (1993) Two functional soybean genes encoding p34Cdcz
protein kinases are regulated by different plant developmental pathways. Proc Natl Acad
Sci USA 90: 943-947

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

Nap J-P, Bisseling T (1990) Developmental biology of a plant-prokaryote symbiosis:
The legume root nodule. Science 250: 948-954

Nelson DR, Kamataki T, Waxman DJ, Guengerich FP, Estabrook RW, Feyereisen
R, Gonzales FJ, Coon MJ, Gunsalus 1C, Gotoh O, Okuda K, Nebert DW (1993) The
P450 superfamily: Update on new sequences, gene mapping, accession numbers, early
trivial names of enzymes, and nomenclature. DNA Cell Biol 12: 1-51

Newman T, de Bruijn FJ, Green P, Keegstra K, Kende H, McIntosh L, Ohlrogge J,
Raikhel N, Somerville S, Thomashow M, Retzel E, Somerville C (1994) Genes
galore: A summary of methods for accessing results from large-scale partial sequencing of
anonymous Arabidopsis cDNA clones. Plant Physiol 106: 1241-1255

95

 

 

 

Nuccio, ML, Thomas TL (1999)ATS1 and ATS3: two novel embryo-speciﬁc genes in
Arabidopsis thaliana. Plant Mol Biol 39: 1153-1163

Perlick AM, Albus U, Stavridis T, Friihling M, Kiister H, Piihler A (1996) The Vicia
faba lipoxygenase gene VfLOXl is expressed in the root nodule parenchyma. Mol Plant-
Microbe Interact 9: 860-863

Pridmore RD (1987) New and versatile cloning vectors with kanamycin resistance
marker. Gene 56: 309-312

Ribeiro A, Akkermans ADL, van Kammen A, Bisseling T, Pawlowski K (1995) A
nodule-speciﬁc gene encoding a subtilisin-like protease is expressed in early stages of
Actinorhizal nodule deve10pment. Plant Cell 7: 785-794

Rogers SO, Bendich AJ (1988) Extraction of DNA from plant tissues. In: SB Gelvin,
RA Schilperoort, eds, Plant Molecular Biology Manual. Kluwer Academic Publishers,
Dordrecht, A6 1-10

Sambrook J, Fritsch EF, Maniatis T (1989) Molecular Cloning. A Laboratory Manual,
ED2. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY

Scheres B, Van de Wiel C, Zalensky A, Horvath B, Spaink H, Van Eck H,
Zwartkruis F, Wolters AM, Gloudemans T, Van Kammen A, Bisseling T (1990a)
The ENOD12 gene product is involved in the infection process during the pea-Rhizobium
interaction. Cell 60: 281-294

Scheres B, van Engelen F, van den Knaap E, van de Wiel C, van Kammen A,
Bisseling T (1990b) Sequential induction of nodulin gene expression in the developing

pea nodule. Plant Cell 2: 687-700

Schultze M, Kondorosi E, Ratet P, Buire M, Kondorosi A (1994) Cell and molecular
biology of Rhizobium-plant interaction. Int Rev Cytol 156: 1-75

Schibahara S, Muller R, Taguchi H, Yoshida T (1985) Cloning and expression of
cDNA for rat heme oxigenase. Proc Natl Acad Sci USA 82: 7865-7869

Smith RD, and WalkerJC (1996) Plant protein phosphatases. Annu Rev Plant Physiol
Plant Mol Biol 47: 101-125

Spaink HP (1996) Regulation of plant morphogenesis by lipo-chitin oligonucleotides.
Crit Rev Plant Sci 15: 559-582

96

Sprent J] (1989) Which steps are essential for the formation of functional legume
nodules? New Phytol 111:129—153

Szekeres M, Nemeth K, Koncz-Kalman Z, Mathur J, Kauchmann A, Altman T,
Redei GP, Nagy F, Schell J, Koncz C (1996) Brassinosteroids rescue the deﬁciency of
CYP90, a cytochrome P450, controlling cell elongation and de-etiolation in Arabidopsis.
Cell 85: 171-182

Szczyglowski K, Kapranov P, Hamburger D, de Bruijn FJ (1998b) The Lotus
japonicus LjNOD70 gene encodes a protein with similarities to transporters. Plant Mol
Biol 37: 651-661

Taylor CB, Bariola PA, DelCardayre SB, Raines RT, Green PJ (1993) RN82: A
senescence-associated RNase of Arabidopsis that diverged from the S-RNases before
speciation. Proc Natl Acad Sci USA 90: 5118-5122

van de Loo FJ, Turner S, Somerville C (1995) Expressed sequence tags from
developing castor seeds. Plant Physiol 108: 1141-1150

van de Sande K, Pawlowski K, Czaja I, Wieneke U, Schell J, Schmidt J, Walden R,
Matvienko M, Wellink J, van Kammen A, Franssen H, Bisseling T (1996)
Modiﬁcation of phytohormone response by a peptide encoded by ENOD40 of legumes
and nonlegumes. Science 273: 370-3 73

Verma DPS (1992) Signals in root nodule organogenesis and endocytosis of Rhizobium.
Plant Cell 4: 373-3 82

Veronesi C., Rickauer M, F ournier J, Pouenat M-L, Esquerre-Tugaye M-T (1996)
Lipoxygenase gene expression in the tobacco—phytophtora parasitica nicotianae
interaction. Plant Physiol 112: 997-1004

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

Vijn I, Yang W-C, Pallisgard N, Jensen E0, van Kammen A, Bisseling T (1995)
VsENOD5, VsENODIZ and VsENOD40 expression during Rhizobium-induced nodule
formation on Vicia sativa roots. Plant Mol Biol 28: 1111-11 19

Welters P, Metz B, Palme K, Szczyglowski K, de Bruijn FJ (1993) Interaction of a

rhizobial DNA-binding protein with the promoter of a plant leghemoglobin gene. Plant
Physiol 102: 1095-1107

97

Winkler RG, Helentjaris T (1995) The maize Dwarf3 gene encodes a cytochrome
P450-mediated early step in gibberellin biosynthesis. Plant Cell 7: 1307-1317

Wu L, Ueda T, Messing J (1995) The formation of mRNA 3’-ends in plants. Plant J 8:
323-329

Yang WC, Katinakis P, Hendriks P, Smolders A, de Vries F, Spee J, van Kammen
A, Bisseling T, Franssen H (1993) Characterization of the GmENOD40, a gene

showing novel patterns of cell-speciﬁc expression during soybean nodule development.
Plant J 3:573-585

Zeidler R, Hobert O, Johannes L, Faulhammer H, Krauss G (1993) Characterization
of two novel single stranded DNA-speciﬁc autonomously replicating sequence-binding

proteins from Saccharomyces cerevisiae, one of which is adenylosuccinate synthetase. J
Biol Chem 268: 20191-20197

98

CHAPTER 3

A protein phosphatase 2C gene, LjNPP2C1, from Lotus japonicus

induced during root nodule development2

3.1. ABSTRACT

Symbiotic interactions between legumes and compatible strains of rhizobia result
in root nodule formation. This new plant organ provides the unique physiological
environment required for symbiotic nitrogen ﬁxation by the bacterial endosymbiont and
assimilation of this nitrogen by the plant partner. We have isolated two related genes
(LjNPP2C1 and LjPP2C2) from the model legume Lotus japonicus, encoding protein
phosphatases type 2C (PP2C). Expression of the LjNPP2C1 gene was found to be
speciﬁcally enhanced in L. japonicus nodules, while the LjPP2C2 gene was expressed at a
similar level in nodules and roots. A GST-LjNPPZCl fusion protein was shown to have a
Mg2+ or Mn2+ dependent, and okadaic acid insensitive PPZC activity in vitro. A chimeric
construct containing the full length LjNPP2C1 cDNA, under the control of the
Saccharomyces cerevisiae alcohol dehydrogenase promoter, was found to be able to

complement a yeast PP2C deﬁcient mutant (pctI A) The transcript level of the LjNPP2C1

gene was found to increase signiﬁcantly in mature nodules, and its highest expression level

 

2 This chapter was published in Kapranov P, Jensen TJ, Poulsen C, de Bruijn FJ,

Szczyglowski K (1999) Proc. Natl. Acad. Sci. USA 96: 1738-1743.

99

occurred aﬁer leghemoglobin (lb) gene induction, a molecular marker for late
developmental events in nodule organogenesis. Expression of the LjNPP2C1 gene was
found to be drastically altered in speciﬁc L. japonicus lines carrying monogenic recessive
mutations in symbiosis-related loci, suggesting that the product of the LjNPP2C1 gene

may function at both early and late stages of nodule development.

100

3.2. INTRODUCTION

Interaction of legume plants with bacteria belonging to the genera Rhizobium,
Bradyrhizobium, Sinorhizobium, Mesorhizobium, and Azorhizobium, triggers a
developmental program leading to the formation of a highly specialized organ, the nodule.
Nodules provide a unique environment for symbiotic ﬁxation and assimilation of
dinitrogen. The early events during nodule organogenesis are set in motion by molecular
cross-talk between the two symbiotic partners culminating in the synthesis of bacterially
encoded lipochito-oligosaccharide signal molecules, the Nod factors (for recent reviews
see 1, 2). Morphogenic activity of the Nod factor induces diverse cellular responses in the
root of a compatible host plant, resulting in the redirection of a subset of the root cortical
cells towards the formation of a nodule primordium (2). Invasion and subsequent
intracellular colonization of the nodule primordium by the microsymbiont initiate the ﬁnal
differentiation process of plant and bacterial cells, which leads to the formation of a
ﬁinctional nitrogen-ﬁxing nodule (3, 4). The induction of a number of plant host genes,
named nodulins (5, 6), accompanies nodule organogenesis and functioning. We have been
interested in studying the complexity and regulation of plant genes expressed during
relatively late stages of the symbiotic interaction (late nodulin genes; 7, 8). The
components of the signal transduction pathways involved in the late developmental
events of nodule morphogenesis and functioning remain mostly uncharacterized, although
nodule-speciﬁc MADS-box genes have been identiﬁed (9, 10). In order to identify novel

genes with potential regulatory functions during late stages of nodule organogenesis and/or

101

nodule functioning, we have constructed a Lotus japonicus late nodulin expressed
sequence tag (EST) library using the mRNA differential display procedure (8). One of the
ESTs identiﬁed during our initial screening, LjN3, showed strong similarity to protein
phosphatase 2C genes from different organisms (8). Here we describe the molecular
characterization of two homologous L. japonicus genes, LjNPP2C1 and LjPP2C2. The
LjNPP2C1 gene corresponds to EST LjN3 and was found to encode a functional protein
phosphatase 2C (LjNPP2C1). During L. japonicus nodule development, the expression of
the LjNPP2C1 gene was found to correlate with relatively late stages of nodule
organogenesis. In contrast, the LjPP2C2 gene, a member of a small family of related genes
in the L. japonicus genome, was found to be expressed at a similar level in nodules and
uninfected roots. Therefore, we postulate that the LjNPP2C1 gene product may play a
speciﬁc role in a signaling cascade at a late stage of nodule development or functioning.

In addition, two experimental approaches, such as generation of transgenic L.
japonicus plants and a yeast 2-hybrid screen, were used to test this hypothesis and to
further address the biological functions of LjNPP2C1. The results of these experiments

are described in the following chapter.

102

3.3. MATERIALS AND METHODS

Plant Material. The Lotus japonicus GIFU B-129-S9 wild-type line and mutant lines
derived thereof were grown as previously described (11, 12). Mesorhizobium loti wild-
type strain NZP2235 was used for all nodulation experiments (12). Uninoculated control
roots were harvested 7,11, 21, and 35 days after sowing. Roots of wild-type plants
inoculated with M loti NZP2235 were harvested 7 and 11 days after inoculation (dai),
and fully developed nodules were collected 21 and 35 dai. Inoculated roots, and nodule-
like structures from mutant lines were harvested 35 dai. Leaves and stems were harvested

from 21 day old plants, whereas flowers were obtained from mature L. japonicus plants.

Northern Blot Analysis. Northern blot hybridization was carried out in 0.5 M
phosphate buffer (pH 7.2) containing 7% SDS and 1% BSA at 65°C, essentially as
described (8, 11). Filters were washed either at high stringency: 25 min in 2X SSC, 0.1%
SDS; 25 min in 1X SSC, 0.1% SDS at 65°C and 15 min in 0.1X SSC, 0.1% SDS at 65°C.
Poly(A)+ mRNA was isolated using the PolyATtract system (Promega) following the
manufacturer instructions. Signals from northern blot assays were quantiﬁed using
Phosphorlmager analysis (Molecular Dynamics) and were standardized to ubiquitin RNA

levels.

103

cDNA Library Screening and 5’ RACE Analysis. A 32P-radiolabeled LjN3 cDNA
fragment and a partial cDNA corresponding to the LjPP2C2 gene were used to screen a L.
japonicus nodule-speciﬁc cDNA library (kind gift from Dr. Jens Stougaard, Aarhus
University, Denmark), following standard protocols (13). The 5’ RACE procedure (14)

was performed using the 5’ RACE system (Gibco/BRL) following the manufacturer’s

instructions. For the reverse transcription reaction 1 ug of poly(A)+ mRNA derived from

L. japonicus nodules and a LjNPP2C1 gene speciﬁc primer (5’-

CAACCCCGTCACGGTCTCCTC-3’) were used.

In vitro Protein Phosphatase Assay. A portion of the LjNPP2C1 cDNA, corresponding
to amino acids 96-362 of the LjNPP2C1 protein, was inserted into the pGEX-SX-l
vector (Pharmacia) to express a glutathione S-transferase (GST)-LjNPP2C1 fusion
protein in Escherichia coli. Afﬁnity puriﬁcation of the resulting fusion protein was
carried out using glutathione-agarose beads (Sigma, product G-4510), following the
procedure described by Smith and Johnson (15). Puriﬁed protein was aliquoted and
stored in 10% glycerol at -20°C.

Protein phosphatase activity was assayed using phosphorylated casein as a
substrate, essentially as described by McGowan and Cohen (16). Brieﬂy, 5 mg of
dephosphorylated bovine casein (Sigma, product C-4765) was phosphorylated using 50

uCi of [y-3ZP]-ATP (3000 Ci/mmol) and 25 units of the catalytic subunit of CAMP-

dependent protein kinase (Sigma, product P-2645) for 4 hr at 30°C, as described (16). The

104

radiolabelled casein was precipitated with 15% trichloroacetic acid (TCA), washed three
times with cold 20% TCA, twice with cold acetone, air-dried, dissolved in 0.2 M Tris-
HCl (pH 8.0) and stored at 4°C. The puriﬁed GST-LjNPP2C1 fusion protein was
assayed for phosphatase activity in a time-course experiment in 50 pl of incubation
mixture consisting of 50 mM Tris-HCI (pH 7.0), 0.1 mM EGTA, 0.1% 2-
mercaptoethanol, 20 mM magnesium acetate, 1 pM okadaic acid, ~ 0.2 pg of the fusion
protein, and 2x104 cpm (Cerenkov counts) of 32P-labelled casein at 30°C. In the control
reactions the fusion protein was replaced with l-2 pg of puriﬁed GST protein. Divalent
cation dependence was analyzed using similar conditions, except that magnesium acetate
was substituted with either 20 mM MnSO4, 20 mM CaCl;, or 10 mM EDTA. All
reactions were terminated by the addition of 50 pl of cold 30% TCA. Aﬁer 5 min of
incubation on ice, the samples were centrifuged for 10 min at 14,000 g, and the
radioactivity in the supernatant and pellet was measured by Cerenkov counting. All

assays were performed in triplicate.

Site-Directed Mutagenesis of the LjNPP2C1 cDNA. An Arabidopsis abiI—I-type
amino acid substitution (see Results section) was created by site-directed mutagenesis of
the LjNPP2C1 cDNA following the method described by Kunkel (17). An oligonucleotide
(5’-TTTGACGGTCAIGACGGAGCTCAGG-3’), containing the abiI-I type mutation

(G—>A), was used to change the Glyl33 residue to Asp. In addition, a silent C—aT

substitution was introduced two base-pairs upstream from the G-—)A mutation in order to

105

create a unique Bsle restriction site. The presence of the base pair substitutions was

conﬁrmed by DNA sequence analysis.

Functional Complementation Assay in S. cerevisiae. The LjNPP2C1 cDNA fragment
was cloned into the yeast expression vector pDBL2 in the sense (construct pDBL3) and
antisense (construct pDBLS) orientation, with respect to the yeast alcohol-
dehydrogenase promoter ADHl (18). In addition, a mutant LjNPP2C1 construct was
prepared by inserting the LjNPP2C1 cDNA, containing the abiI-I type substitution, into
the pDBL2 vector in the sense orientation to generate plasmid pDBL3M.

Plasmids were introduced into yeast strain TM126 (19), carrying a disruption in
the PT C1 locus encoding yeast protein phosphatase 2C, using a modiﬁed lithium
acetate/PEG method (20). The pDBCl construct, containing the wild-type yeast pth
gene under the control of the ADHl promoter (19), as well as the pDBL2 vector alone,
were also introduced into strain TM126 and used as positive and negative controls,
respectively. Complementation assays were performed essentially as described by
Bertauche et a1. (21). Each transformant was grown to saturation at 28°C in liquid
synthetic SD medium (22), lacking uracil and leucine. For the complementation assay, 3

pl of the yeast suspension culture, containing approximately 104 cells/pl, were replica

plated onto YPD media (22). Replica plates were incubated at either 28°C (pennissive-),
or at 37°C (non-permissive) temperatures, for 30-36 hours. Subsequently, the phenotype

of the individual strains was carefully analyzed to examine the ability of the recombinant

106

plasmids to complement the temperature-sensitive growth phenotype of strain TM126

(19).

107

3.4. RESULTS

The L. japonicus LjNPP2C1 Gene Shares Significant Similarity with Protein
Phosphatase 2C Genes. Using the mRNA differential display procedure (23), we
previously isolated a range of L. japonicus ESTs associated with relatively late stages of
nodule development and/or nodule functioning (8). One of these ESTs, LjN3, was found
to share signiﬁcant similarity with a number of protein phosphatase 2C genes from
different organisms (8). Subsequent screening of a nodule speciﬁc cDNA library, using
radiolabelled LjN3 cDNA as a probe, facilitated the cloning of the corresponding ﬁill
length cDNA. The nucleotide sequence of the longest cDNA clone (LjNPP2C1-16) was
shown to be 1235 bp in length, and was found to contain a 1086 bp long open reading
time (ORF), starting with an ATG codon at base pair 25. Since the ATG25 codon was
not preceded by an in-frame stop codon, the 5’ Rapid Ampliﬁcation of cDNA Ends
(RACE) procedure was used to further characterize the 5’-terminal sequence of the
LjNPP2C1 mRNA. The sequence of the longest 5’-RACE cDNA product extended the 5’
end of the LjNPP2C1-16 cDNA by 83 bp (data not shown). However, no additional in-
frame ATG codons were found in the 5’ sequence. Moreover, the length of the LjNPP2C1
cDNA correlates well with the observed mRNA size on northern blots (~1300 nt). Based
on these observations, we propose that the ATG25 of LjNPP2C1—16 cDNA (ATGlog in
the extended sequence) represents the initiation codon of the LjNPP2C1 gene, and that the

LjNPP2C1-16 cDNA contains the entire coding region for the LjNPP2C1 protein.

108

The amino acid sequence of the deduced LjNPP2C1 protein was found to
correspond to a polypeptide of 362 residues and a predicted molecular mass of ~ 39.5
kD- In agreement with our earlier observations (8), the deduced protein was found to
share a high level of similarity with a number of PP2C proteins from different eukaryotes,

including the A311 and ABIZ proteins from Arabidopsis thaliana (Fig. 3.1).

The L. japonicus LjPP2C2 gene. In the course of an independent research project, a
distinct partial cDNA clone, encoding a putative L. japonicus protein phosphatase type
2C, was fortuitously identiﬁed (data not shown). A corresponding full copy cDNA clone
(LjPP2C2) was isolated from the L. japonicus cDNA library and was found to encode a
POIypeptide (LjPP2C2) of 282 residues and a predicted molecular mass of ~ 30.8 kD.
Similar to the LjNPP2C1 protein, the predicted amino-acid sequence of the LjPP2C2 was

found to share a high level of similarity with a number of PP2C proteins from different
eukaryotes, including L. japonicus LjNPPZC 1, and A. thaliana ABIl (Fig. 3.1). Based on

these results, we conclude that the LjNPP2C1 and LjPP2C2 genes are likely to encode

1W0 Similar, but distinct L. japonicus protein phosphatases type 2C.

LjNPP2C1, but not LjPP2C2 transcripts, accumulate preferentialy in L. japonicus
modules. We previously reported that LjNPP2C1 mRNA levels were enhanced in L.
japonicus nodules versus uninoculated control roots (8). In order to address the question

Whether the LjPP2C2 gene was also preferentialy expressed in nodules, a northern blot

analysis was performed. Poly(A)+ mRNA derived from uninfected L. japonicus control

109

roots and nodules was hybridized sequentially with [a-32P]-dATP labeled LjNPP2C1,

LjPP2C2, and Sesbania rostrata ubiquitin cDNA probes (Fig. 3.2). A signiﬁcantly higher
level of LjNPP2C1 mRNA was detected in nodules than in uninfected L. japonicus
control roots, conﬁrming and extending our earlier observations (8). In contrast, the level
of LjPP2C2 mRNA was found to be equal in both tissues. Based on these results, we
conclude that the LjNPP2C1 is likely to have a unique function(s) associated with
nitrogen ﬁxing nodules. Therefore, our further analysis concentrated on a more detailed

characterization of the LjNPP2C1 gene.

The LjNPP2C1 gene encodes a functional protein phosphatase type 2C. To
investigate whether the LjNPP2C1 gene encoded a functional protein phosphatase 2C, an
in vitro phosphatase assay was carried out. For this purpose, a recombinant protein was
created containing the C-terrninal catalytic domain of LjNPP2C1 fused to the glutathione
S-transferase (GST) protein. The GST-LjNPP2C1 fusion protein was expressed in
Escherichia coli, puriﬁed, and found to display a time-dependent phosphatase activity
using phosphorylated casein as a substrate (Fig. 3.3A). The phosphatase activity of the
GST-LjNPP2C1 fusion protein was found to be insensitive to 1 pM okadaic acid, a
potent inhibitor of protein phosphatases other than type 2C (24), and to be dependent on
the presence of the divalent cations Mg2+ or Mn2+ (Fig. 3.38). The GST protein alone,
expressed and puriﬁed under similar conditions, was found to be inactive in the

phosphatase assay (Fig. 3.3 B).

110

The biochemical evidence for protein phosphatase 2C activity of the LjNPP2C1
gene product was further supported by heterologous genetic complementation
experiments in yeast. Saccharomyces cerevisiae strain TM126 carries a disrupted PP2C

gene (pthA), resulting in a temperature sensitive growth phenotype. Cells of the ptc1A

mutant strain grow slower at 37°C than at 28°C (19). The LjNPP2C1 cDNA, fused to the
yeast ADHI promoter was introduced into S. cerevisiae strain TM126 on plasmid
pDBL3 and was indeed found to complement the yeast [2th mutant phenotype (Fig. 3.4).
A plasmid carrying the wild-type LjNPP2C1 cDNA in the antisense orientation

(pDBLS), failed to complement the temperature sensitive growth phenotype of pthA

yeast cells (Fig. 3.4).

In order to provide further evidence for the designation of the LjNPP2C1 gene
product as a protein phosphatase 2C , we made use of the phenotype of an Arabidopsis
thaliana mutant PP2C protein. The A. thaliana abiI-I mutation is caused by a single
base-pair transition that substitutes amino-acid residue Glyl80 with an Asp residue,
resulting in a signiﬁcant decrease of PP2C activity in vitro, as well as the generation of a
dominant negative mutation leading to abscisic acid insensitivity (21, 25, 26). An abiI-I-

type mutation (Glyl33—>Asp), was created in the wild-type LjNPP2C1 gene by site

directed mutagenesis. The resultant mutant LjNPP2C1 protein was indeed found to be

unable to complement the pthA mutant phenotype of yeast strain TM126 (Fig. 3.4).

111

LjNPP2C1 mRNA Accumulates During Relatively Late Stages of Nodule
Development. To correlate the expression pattern of the LjNPP2C1 gene with (a)
speciﬁc phase(s) in nodule development, a time course northern hybridization analysis
was performed. LjNPP2C1 mRNA was found to be present at a low level in uninoculated
L. japonicus roots, and to be slightly increased at 7 and 11 days after inoculation with M
loti. However a ~ 20 fold increase of LjNPP2C1 mRNA was observed in nodules
harvested 21 days after inoculation (Fig. 3.5). Interestingly, the most dramatic
enhancement of LjNPP2C1 mRNA level was found to occur after the induction of L.
japonicus leghemoglobin (lb) gene expression (a molecular marker for late stages of nodule
development; 7), which was found to occur in nodules 11 days after inoculation (Fig. 3.5;

see also 12).

Tissue-specific expression of the LjNPP2C1 gene. To gain ﬁlrther insight into the
tissue speciﬁcity of LjNPP2C1 gene expression, poly(A)+ mRNA isolated from different
L. japonicus tissues was probed with a radiolabeled LjNPP2C1 cDNA. LjNPP2C1
mRNA was found to accumulate to the highest level in nodules and ﬂowers (Fig. 3.6). A
low level of LjNPP2C1 mRNA could also be found in uninfected root and leaf tissues
(Fig. 3.6), whereas no hybridization signal, even upon prolonged exposure period, was
detected in the L. japonicus stems (data not shown). Therefore, we conclude that
LjNPP2C1 gene expression is not nodule-speciﬁc, but that expression of this gene is

substantially enhanced in fully developed nodules and ﬂowers.

112

LjNPP2C1 Gene Expression is Altered in Symbiotic mutants of L. japonicus. The
expression pattern of the LjNPP2C1 gene was further analyzed by examining six non-
allelic L. japonicus symbiotic mutants. Four non-nodulating lines (Nod‘; LjEMS34,
LjEMS46, LjEMS70, and LjEMS76), which fail to display macroscopically visible signs
of nodulation, and two distinct mutant lines (LjEMS88 and LjEMS75), forming white,
mostly ineffective (Nod+Fix’/+) nodules (12), were examined for LjNPP2C1 expression.
Mutant lines LjEMSSS and LjEMS75 differ with respect to their ability to
support late stages of nodule development. Mutant line LjEMS88 forms only small white
nodule-like structure which do not ﬁx nitrogen, and accumulate only background levels of

leghemoglobin mRNA ($50 fold less than wild type level; 12). In contrast, the LjEMS75

line produces a mixed population of well developed white nodules, in addition to pink
wild-type-like nodules, that ﬁx nitrogen, albeit at a signiﬁcantly diminished level (12).
Both types of the LjEMS75 nodules were harvested and analyzed separately. White
nodules of line LjEMS75 were found to contain a signiﬁcant level of L. japonicus lb
mRNA (only ~ 3 fold lower than the wild type; Fig. 3.7A), which clearly distinguishes
them from white nodules formed on line LjEM888 (see above).

35-days-old wild-type nodules, and pink nodules of mutant line LjEMS75
showed similar, 15-20 times elevated, levels of LjNPP2C1 mRNA, as compared with
uninfected control roots. White nodules of LjEMS75 were found to have a two-fold lower
level of LjNPP2C1 transcripts than wild-type nodules, but an approximately 7 times

higher transcript level than uninfected control reots (Fig. 3.7A). Analysis of LjNPP2C1

113

gene expression in white nodule-like structures formed on the LjEMS88 mutant line
showed only a background level of the corressponding mRNA (Fig. 3.78). A background
level of LjNPP2C1 mRNA was also detected in wild-type uninfected roots and infected
roots of lines LjEMS7O and LjEMS76. In contrast, the infected roots of the non-
nodulating mutant line LjEMS34 showed a relatively modest increase (two times higher
than the wild-type uninoculated roots) of LjNPP2C1 transcript levels, whereas LjEMS46
was found to display a signiﬁcant levels of LjNPP2C1 mRNA: ~50% of that present in

nitrogen ﬁxing nodules.

114

3.5. DISCUSSION

Reversible phosphorylation of proteins has been implicated in the regulation of
cellular processes as diverse as metabolism, transcription and translation, cell division,
membrane transport and secretion, stress response, fertilization, and memory (27). The
role of protein kinases in controlling the level of phosphorylation of proteins has been
well documented (28). The ﬁnding that the activity of protein phosphatases is regulated
in speciﬁc cases, has led to the hypothesis that cellular responses to external stimuli
might result from direct activation or inhibition of protein phosphatases, rather then
through the action of protein kinases (24, 28). So far, three out of the four known major
classes of SerfT hr protein phosphatases have been described in plants (PPl, PP2A and
PP2C), while additional protein phosphatases unrelated to these major groups are thought
to exist (24). In general, PP2C is the least well-characterized class of protein
phosphatases. However, recent studies have suggested an intriguing connection between
PP2C enzyme activity and several cellular processes in different eukaryotes (for recent
reviews see 24, 28, 29). Plant members of this group of protein phosphatases include four
different PP2Cs from A. thaliana: KAPP (30), AtPP2C (31), A811 (25, 26), A812 (32),
and a protein phosphatase 2C (MP2C) from alfalfa (33).

Here we report the identiﬁcation and characterization of a plant gene, LjNPP2C1,
encoding a protein with amino acid similarity to protein phosphatases type 2C, the
expression of which is signiﬁcantly enhanced during L. japonicus root nodule

organogenesis. The following biochemical and genetic experiments have led us to the

115

conclusion that the LjNPP2C1 gene encodes a ﬁlnctional protein phosphatase type 2C. 1.
A recombinant GST-LjNPP2C1 protein is capable of dephosphorylating phosphorylated
casein, a commonly used artiﬁcial substrate for measuring PP2C activity (16). The
activity of the LjNPP2C1 catalytic domain present in the GST-LjNPP2C1 fusion is
insensitive to inhibition by okadaic acid, and dependent on the presence of divalent
cations (Mg2+ or Mn”). 2. The LjNPP2C1 gene is able to complement a yeast PP2C

deﬁcient mutant (pctIA.) 3. The LjNPP2C1 cDNA, carrying an abil type single amino-
acid substitution (Gly——>ASp), is unable to complement the temperature sensitive

phenotype of yeast strain TM126, thereby mimicking the behavior of the A. thaliana
ABIl mutant protein used in similar complementation experiments (21). We also describe
a second gene, LjPP2C2, which encodes a protein sharing signiﬁcant amino-acid similarity
with LjNPP2C1, as well as other PPZC proteins from different eukaryotes. Unlike in the
case of the LjNPP2C1 gene, the LjPP2C2 transcript appears to accumulate to a similar
level in both uninoculated L. japonicus roots and nodules, suggesting that two different
regulatory mechanisms are involved in the expression of these otherwise similar genes.
LjNPP2C1 gene is represented by a single- or low-copy gene number in L. japonicus
genome, while LjPP2C2 gene is a member of a small family of related genes, as evidenced
by Southern blot hybridization (data not shown).

LjNPP2C1 appears to be the only plant protein phosphatase 2C reported thus
far, whose expression is enhanced during plant-microbe interactions in general, and

symbiotic root nodule formation in particular. Moreover, the temporal expression pattern

116

of the LjNPP2C1 gene during rhizobial infection and nodule formation appears to be
unusual in comparison to that of other nodulin genes. Only a limited level of LjNPP2C1
mRNA accumulation is observed at relatively early stages of infection and nodule
morphogenesis. In contrast, a 20-fold higher level of LjNPP2C1 transcripts, as compared
with the control uninoculated roots, is observed in 21 day-old nodules. The prominent
increase in LjNPP2C1 mRN A levels observed is preceded by the developmental transition
between nodule ontogeny and the establishment of a functional, nitrogen ﬁxing nodule, as
indicated by the expression pattern of the L. japonicus leghemoglobin gene, which serves
as a molecular marker for this transition (Fig. 3.5; see also 12). This observation is
supported by results obtained from studies of LjNPP2C1 gene expression in symbiotic
mutants of L. japonicus. The highest levels of LjNPP2C1 transcripts were found in
nodules, or nodule-like structure, already containing a substantial level of leghemoglobin
mRNA. For example, the white-, and pink nodules formed on mutant line LjEMS75
accumulate a high level of both lb and LjNPP2C1 gene transcripts, whereas only a low
level of these mRNA species are found in white, ineffective nodule-like structures formed
on line LjEMSS8 (Fig. 3.7; see also 12).

On the basis of these results, it appears that the LjNPP2C1 protein has a speciﬁc
function(s) in nodules that are in the process of initiating nitrogen ﬁxation, and have
already passed the developmental time-point at which expression of a late nodulin
molecular marker gene lb is activated. The intriguing observation that the roots of two
particular non-nodulating L. japonicus mutant lines, LjEMS34 and LjEMS46, contain

elevated level of LjNPP2C1 mRNA is difﬁcult to interpret. However, it is conceivable

117

that the enhanced expression of LjNPP2C1 gene in these mutant lines is somehow related
to their Nod' phenotypes: it might at least partially be responsible for the observed
phenotypes, especially in the case of LjEMS46 mutant line. It is possible that
inapprOpriate timing or level of LjNPP2C1 induction may have quite different phenotypic
effects at the early vs late stages of nodule development. At present we are investigating
whether the LjEMS46 mutant allele maps at or near the LjNPP2C1 locus. Further
analysis of LjNPP2C1 gene expression, the phenotype(s) of transgenic plants
(over)expressing the LjNPP2C1 transcript in sense- and antisense orientation, and
characterization of the relevant in vivo substrate(s) of LjNPP2C1 will be essential to
resolve the function of the LjNPP2C1 gene during nodulation or symbiotic nitrogen

ﬁxation.

118

   
 
 
 

 

 

    
 
 

 

 

 

 
    

 

 

 

 

 

 

 

 

 

 

 
 
 

 

 

   

 

 

 

 

 

 

 

 

 

LjNPP2C1 7
LjPP2C2 --------------------------------------------

A311 naavspAIAGPPRPPSETQMDF'rGIRLGKGYCNNQYSNQESBNG 44
LjNPPZCI pssznﬁanxﬂnﬁﬂn-Ecoﬁn 5°
LjPP2C2 --------------------------------------------

ABII DLr-PE-“S-S-l-SSHGEEIENSPELNHEA 87
LjNPP2C1 3333333113331 .e. 94
LjPP2C2 ------- MTG 33 1 L a ' 32
A311 DIVVVDISAGBI -- 125
LjNPP2C1 s 513313 MEDAV-V ------- ﬂ -------- Ex 122
LjPP2C2 '1- YR 33311113313 ------------- AQFKTVDNNE 63
A311 3 3 . REE-MBDAVTIPRFLQS- LDGRPDpdEA 169
LjNPP2C1 c YPAVPDGHGGA VA c3331. R VABB ----- @GNGVE 161
LjPP2C2 LG FA IPDGHSG \[ﬂ YL s L ----DNI apnrw ------ 97
A311 A FPGVYDGLIGGSQVA c333 HL-LA33- EKPMLIEDGDTW 213
LjNPP2C1 3v 3c; NMDGBv-A NAA 'rvcs'rAVVAv A :-- 203
LjPP2C2 -TKP xx L3 3 GR GsrAvEAI IN . 140
A311 LEKEKKA FR -3 - p Tves'rSVVAv F'EH- 254
LjNPP2C1 ’VIANCGD RAVI, RGGBA LSSDHKPD - L 12133“;va 247
LjPP2C2 VVANEGD RAVL GEAI stnnE ATE -- 3313 s 182
A311 VANCGD RAvn R K A stnnxpn :- A RI Acexv 298
LjNPP2C1 atrium-R GVL sasrcn YLR ’v1 pav'r x11 ' 239
LjPP2C2 ‘ Ee-Dvnv GELA 331E633 LKK PIEVTVBLIDDD :- 226
A311 . - GVL snsrcn YLK 1 3 DEV Av-x 34o
LjNPP2C1 'PLILAsnchnvrss --------- cﬂlﬁrnﬁdﬂ 324
LjPP2C2 FIILASDGL vusu --------------------- 249
A311 LILAsncvwnvn 3 RILLWHKKNAEA-ASLEAE 334
LjNPP2C1 ENQSR---- LA31A xesnn ‘I‘SVIVIEL GTVT- 352
L5 szcz ------ N v K L '13 AL 313 sap 3 ﬂvv v13 ----- 282
A311 [33311331133 L LA1 aesxn svvvan PRRKL 428
LjNPP2C1 ------ 352
LjPP2C2 ------ 282
A311 KSKPLN 434

Figure 3.1. Amino acid sequence alignment of L. japonicus LjNPP2C1 and LjPP2C2 and
A. thaliana ABIl (25, 26) proteins. The alignment was performed using the PileUp
algorithm from the GCG software package (Genetics Computer Group, Madison, WI).
Identical amino acids are shaded and conservative substitutions are boxed using the Squu

1.1 program. Gaps introduced to allow an optimal alignment are represented by dashes.

119

Nodule

 

—LjNPP2C1
—LjPP2C2

— ubiquitin

 

 

 

 

Figure 3.2. Northern blot analysis of LjNPP2C1 and LjPP2C2 expression. Two
micrograms of poly (A)+ mRNA from uninfected roots and mature nodules were
separated under denaturing conditions, and hybridized sequentially with a-32P-labelled
probes corresponding to the L. japonicus LjNPP2C1, LjPP2C2, and Sesbania rostrata

ubiquitin cDNAs.

120

 

 

 

Release of 32? (%)
Release of 32P (%)

 

 

 

 

 

 

 

 

0 1 % .L r
15 30 60 120 ""8H Mn“ Ci” EDTA 051
Time (min)

 

 

Figure 3.3. Protein phosphatase type 2C activity of a GST-LjNPP2C1 protein. (A)
Experiments showing a time-dependent dephosphorylation of [32P]-labelled casein by a

recombinant GST-LjNPP2C1 protein in the presence of 20 mM magnesium acetate and 1

pM okadaic acid. (B) Divalent cation requirement for the protein phosphatase activity of

LjNPP2C1. The recombinant GST-LjNPP2C1 protein was assayed for its PP2C activity

in the presence of Mg”, Mn2+ or Ca2+, or in the absence of a divalent cation and in the

presence of EDTA. “GST” denotes the assay performed in the presence of the GST

protein alone.

121

Figure 3.4. Complementation of the PP2C deﬁcient yeast mutant strain

TM126 (ptclA). For each strain, 3 pl of cell suspension of each yeast

strain, containing approximately 3x104 cells, was spotted onto agar plates
(see Material and Methods) and incubated at either 28°C or 37°C for two
days. Strain TM126 was transformed with the yeast expression vector
pDBL2 as a control, plasmid pDBCl containing the wild-type yeast
PTCI gene (19), plasmids pDBL3 and pDBL3M containing either wild-
type, or mutant L. japonicus LjNPP2C1 genes, and plasmid pDBLS
containing the LjNPP2C1 gene in the anti-sense orientation, with respect
to the yeast ADH promoter. The poor growth phenotype of strain
TM126 at the non-permissive temperature (37°C) is evidenced by a less
dense (partially translucent) spot; complementation of the temperature
sensitive growth phenotype is evidenced by a denser (non-translucent)

spot.

122

28°C 37°C

 

123

uninoculated days after
root inoculation

7112171121

 

v

_ LjNPP2C1

— leghemoglobin

— Ubiquitin

 

 

 

Figure 3.5. Northern blot analysis of developemntal LjNPP2C1 gene expression. Two
micrograms of poly(A)+ mRNA isolated from uninoculated roots, harvested 7, 11 and 21
days after sowing, and roots harvested from L. japonicus plants 7 and 11 dai, as well as
21 days old nodules, were analysed. The blot was sequentially probed with the
radiolabelled LjNPP2C1 cDNA, the LjN7 7 EST corresponding to a L. japonicus

leghemoglobin gene (8), and the S. rostrata ubiqutin cDNA as a loading control.

124

Root
Nodule

   

. — LjNPP2C1
. i. 3:... — eIF 4A

Figure 3.6. Northern blot analysis of LjNPP2C1 expression in different tissues of L.
japonicus. Four micrograms of poly(A)+ RNA isolated from ﬂowers, leaves, uninoculated
roots and mature nodules of L. japonicus were separated under denaturing conditions, and

probed with radiolabeled probes corresponding to the L. japonicus LjNPP2C1 and

Arabidopsis eIF4A cDNAs.

125

Figure 3.7. Comparison of LjNPP2C1 transcript levels in: (A) wild-type
uninfected roots (control), inoculated roots of non-nodulating mutant lines
LjEMS34, LjEMS46, LjEMS70, and LjEMS76, white (LjEMS75 white)
and pink (LjEMS75 pink) nodules of mutant line LjEMS75, and 35 day-
old wild-type nodules. (B) wild-type uninfected roots (control), white
nodule-like structures induced on the mutant line LjEMSSS, and 35- and
2l-days-old wild-type nodules. The blots shown in panels A and B were

sequentially hybridized with [a-32P]dATP labeled probes corresponding

to the LjNPP2C1 cDNA, the L. japonicus leghemoglobin gene (EST LjN 7 7 ;

8) and a S. rostrata ubiquitin cDNA, as loading control.

126

mo—n—ﬂc: ﬁ—N

3.33.. own

an wim—

3.5.—3

Ame—335 .3
£53 mam—23.:
3.25 maze:
3523.:

@5293
3323.3

3%sz

::.—Ecov .2

LjNPP2C1 _

uh.“

.m
b
h
g
0
m
e
h
g
h

 

 

 

-...

— Ubiquitin -—

Q”.

127

3.6. ACKNOWLEDGMENTS

We are grateﬁll to Dr. Haruo Saito (Harvard University) for kindly providing the
yeast strain TM126, vectors and advice on the yeast complementation experiments. We
wish to thank Kurt Stepnitz for expert photographic assistance. This work was
supported by grants from the Department of Energy (DE-FG02-91ER20021), the

National Science Foundation (N SF-O96301 89) and the US. Department of Agriculture.

128

3.7. LITERATURE

l.

9.

Long, S. R. (1996) Plant Cell 8, 1885-1898.
Spaink, HP. (1996) Crit. Rev. Plant Sci. 15, 559-582.
Verma D. P. S. (1992) Plant Cell 4, 373-382.
Mylona, P., Pawlowski, K. & Bisseling, T. (1995) Plant Cell 7, 869-885.
Legocki, R.P. & Verma, D.P.S. (1980) Cell 20, 153-163.
van Kammen, A. (1984) Plant M01. Biol. Rep. 2, 43—45.
de Bruijn, F. J. & Schell, J. (1992) in Control of Plant Gene Expression, ed. Verma D.
P. S. (CRC Press, Boca Raton), pp. 241-258.
Szczyglowski, K., Hamburger, D., Kapranov, P. & de Bruijn, F. J. (1997) Plant
Physiol. 114, 1335-1346

Heard, J., & Dunn, K. (1995) Proc. Natl. Acad. Sci. USA 92, 5273-5277.

10. Heard, J ., Caspi, M. & Dunn, K. (1997) Mol. Plant-Microbe Interact. 10, 665-676.

11. Kapranov, P., de Bruijn, F. J. & Szczyglowski, K. (1997) Plant Physiol. 113, 1081-

1090.

12. Szczyglowski, K., Shaw, R.S., Wopereis, J ., Copeland, S., Hamburger, D., Kasiborski,

B., Dazzo, F. & de Bruijn, F.J. (1998) Mol. Plant-Microbe Interact. 11, 684-697.

13. Sambrook, J ., Fritsch, E. F. & Maniatis, T. (1989) Molecular Cloning. A Laboratory

Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

14. Frohman, M. A., Dush, M. K. & Martin, G. R. (1988) Proc. Natl. Acad. Sci. USA 85,

8998-9002.

15. Smith, D.B. & Johnson, KS. (1988) Gene 67, 31-40.

129

16. McGowan, C.H. & Cohen, P. (1988) Methods Enzymol. 159, 416-426.

17. Kunkel, T. A. (1985) Proc. Natl. Acad. Sci. USA 82, 488-492.

18. Milne, G. T. & Weaver, D. T. (1993) Genes and Dev. 7, 1755-1765.

19. Maeda, T., Tsai, A. Y. & Saito, H. (1993) Mol. Cell. Biol. 13, 5408-5417.

20. Elble, R. (1992) Biotechniques 13, 18-20.

21. Bertauche, N., Leung, J. & Giraudat, J. (1996) Eur. J. Biochem. 241, 193-200.

22. Sherman, F ., Fink, G. R. & Hicks, J. B. (1983) Methods in Yeast Genetics. Cold Spring
Harbor Laboratory Press, Cold Spring Harbor, New York.

23. Liang, P. & Pardee, A. (1992) Science 257, 967-971.

24. Smith R. D. & Walker, J. C. (1996) Annu. Rev. Plant Physiol. Plant Mol. Biol. 47, 101-

125.

25. Leung, J., Bouvier—Durand, M., Morris, P.-C., Guerrier, D., Chefdor, F. & Giraudat,
J. (1994) Science 264, 1448-1452.

26. Meyer, K., Leube, M.P. & Grill, E. (1994) Science 264, 1452-1455.

27. Cohen, P. (1989) Annu. Rev. Biochem. 58, 453-508.

28. Hunter, T. (1995) Cell 80, 225-236.

29. Luan S. (1998) Trends Plant Sci. 3, 271-275.

30. Stone, J.M., Collinge, M.A., Smith, R.D., Horn, M.A., & Walker, J .C. (1994) Science
266, 793-795.

31. Kuromori, T., & Yamamoto, M. (1994) Nucl. Acids Res. 22, 5296-5301.

32. Leung, J., Merlot, S. & Giraudat, J. (1997) Plant Cell 9, 759-771.

130

33. Meskiene, I., Begre, L., Glaser, W., Balog, J., Brandstotter, M., Zwerger, K.,
Ammerer, G. & Hirt, H. (1998) Proc. Natl. Acad. Sci. USA 95, 1938-1943.

131

CHAPTER 4

Analysis of the biological functions of LjNPP2C1

4.1. ABSTRACT

Protein phosphatases type 2C (PP2Cs) play key roles in regulating multiple biological
processes in eukaryotes. In plants, PP2Cs have been implicated in regulating abscisic acid
signaling, the stress induced MAP kinase pathway and signal transduction via receptor-
like protein kinases. In addition, a novel PP2C from L. japonicus, LjNPP2C1, has been
implicated in regulating the events taking place at late stages of nodule development. This
chapter describes two independent approaches taken to ﬁlrther address the possible role
and signiﬁcance of LjNPP2C1 in L. japonicus nodule development and functioning, as
well as in other aspects of plant physiology. First, transgenic L. japonicus plants were
generated containing chimeric constructs designed to either increase or decrease
endogenous levels of LjNPP2C1 mRNA. In addition, transgenic plants expressing a
dominant-negative isoform of LjNPP2C1 were generated. Second, a yeast two-hybrid
screen was used to identify L. japonicus cDNAs encoding proteins interacting with

LjNPP2C1. One such cDNA, LBP, was identiﬁed and characterized.

132

4.2. INTRODUCTION

Protein Phosphatases type 2C (PP2Cs) belong to the least well-characterized
class of the protein serine/threonine phosphatases, sharing no apparent amino acid
sequence similarity with other major classes of to protein phosphatases, such as PPl,
PP2A and PP2B (Cohen, 1989; Cohen and Cohen, 1989). PP2Cs require Mg2+ or Mn2+
for enzyme activity and exhibit an insensitivity to okadaic acid, a potent inhibitor of
protein phosphatases belonging to classes PP] and PP2A (Cohen, 1989). PP2Cs are
believed to function as monomeric proteins, and no additional regulatory subunits have
been identiﬁed (Cohen, 1989; Smith and Walker, 1996). The distribution of PP2C
enzymes in plants appears to be more restrictive than that of other protein phosphatases
(Smith and Walker, 1996). The lack of genetic approaches and the absence of speciﬁc
inhibitors have greatly hindered the analysis of PP2C functions. However, recent studies
have suggested intriguing connections between PP2C enzymes and several cellular signal
transduction systems. PP2Cs have been implicated in the regulation of mammalian AMP-
activated protein kinases (Davis et al., 1995) and in male sexual development in the
nematode Caenorhabditis elegans (Pilgrim et al., 1995; Chin-Sang and Spence, 1996). In
Saccharomyces cerevisiae, the PT C I and PT C3 genes, encoding PP2C enzymes, have
been shown to be suppressors of the sin] mutation (Maeda et al., 1994). The SLNI gene
encodes a transmembrane protein homologous to histidine kinases and the response
regulator proteins of bacterial two-component systems, which have been identiﬁed in a

wide variety of signaling pathways (Ota and Varshavsky, 1993). In the ﬁssion yeast

133

Sacharomyces pombe , analysis of the three ptc (phosphatase two C) genes (ptcl, ptc2
and ptc3), has led to the hypothesis that they play important roles in the osmo-sensing
MAP kinase signal transduction pathway (Shiozaki and Russell, 1995). Interestingly, the
Ser protein phosphatase SpoIlE from Bacillus subtilis, which acts on SpoIIA triggering
the speciﬁc activation of a sporulation transcription factor, has recently been reported to
have sequence similarity to the PP2C family of eukaryotic Ser/Thr protein phosphatases
(Adler et al., 1997).

Besides LjNPP2C1, several PP2C genes from from plants have been cloned and
characterized (Rodriguez, 1998), including ﬁve from Arabidopsis (PP2C-At, KAPP, A811,
A812 and AtP2C-HA; Kuromori and Yamamoto, 1994; Stone et al., 1994; Leung et
al.,l994; Leung et al.,l997; Meyer et al.,l994; Rodriguez et al., 1998) and one from
Medicago sativa, MPZC (Meskiene et al., 1998). The common feature of plant PP2C
proteins is presence of the N-tenninal extensions of variable lengths in addition to the
PP2C catalytic region (Rodriguez, 1998; Kapranov et al., 1999). The N-terminal regions
of plant PP2Cs typically do not share any sequence similarity with each other or with
any other protein in the databases and have not been found in any of the known ﬁmgal or
mammalian PP2Cs, except for FEM-2 of C. elegans (Smith and Walker, 1996, Pilgrim et
al., 1995).

The protein phosphatase PP2C-At gene was identiﬁed via a complementation
screen of the S. pombe pdel mutant, which is defective in CAMP phosphodiesterase, a
Component of the CAMP-dependent protein kinase cascade (Kuromoni and Yamamoto,

1994). Its role in Arabidopsis, however, remains unknown. The second PP2C cloned,

134

termed “KAPP”, for kinase-associated protein phosphatase, was isolated based on its
ability to interact in vitro with a serine/threonine receptor-like kinase, RLK5, from
Arabidopsis (Stone et al., 1994). RLK5 is a member of a family of related receptor—like
protein kinases (RLKs) that participate in diverse biological functions, including defense
responses, self incompatibility, and plant development (Stone et al., 1994). Moreover,
KAPP interacts in vitro with several RLK catalytic domains, suggesting that this PP2C
may regulate a number of signaling pathways mediated by RLKs (Braun et al., 1997). In
fact, KAPP has been implicated in the negative regulation of CLAVATAI, a receptor-like
kinase from Arabidopsis involved in controlling the size of ﬂoral and apical meristems
(Williams et al., 1997).

The MP2C cDNA was isolated in a screen for negative regulators of the yeast
mitogen-activated protein kinase (MAPK) pathway (Meskiene et al., 1998). Puriﬁed
MP2C protein was found to be able to inactivate the stress-activated MAPK (SAMK) in
plant extracts (Meskiene et al., 1998). The SAMK pathway is normally induced by a
variety of abiotic stresses, such as drought, cold, wounding, etc. (Bogre et al., 1997; Jonak
et al., 1996). Moreover, the MP2C mRNA was transiently induced by wounding and its
expression pattern correlated with the inactivation of the SAMK pathway (Meskiene et
al., 1998). These results suggested that MP2C is a negative regulator of the SAMK
pathway in plants (Meskiene et al., 1998).

Additional evidence that type 2C protein phosphatases are involved in plant
signaling pathways comes from the isolation of the A811 and A812 genes. Originally, a

dominant mutation (abiI-I) conferring abscisic acid (ABA) insensitivity in Arabidopsis

135

was identiﬁed and the corresponding ABII gene shown to encode a PP2C enzyme (Meyer
et al., 1994; Leung et al., 1994; Bertauche et al., 1996). The abiI mutant is severely
impaired in a wide spectrum of ABA responses, including reduced seed dormancy,
excessive water loss and abnormal drought rhizogenesis (Meyer et al., 1994; Leung et al.,
1994, and references therein). Thus, the control of the phosphorylation state of cell
signaling components by the A811 product may be mediating the observed pleiotropic
hormone responses (Meyer et al., 1994; Leung et al., 1994). The abiI-I mutation
converts the Gly residue at position 180, which is embedded in a sizable block of
contiguous sequence identities among different PP2Cs, to an Asp residue and is dominant
over the wild-type ABII allele (Meyer et al., 1994; Leung et al., 1994). The abiI-I
dominant mutant allele of the AB]! gene has been introduced into Nicotiana benthamiana,
and the resulting transgenic plants showed a complementary subset of mutant
phenotypes, including the tendency to wilt and reduced seed dormancy (Amstrong et al.,
1995). These results suggest the presence of a signiﬁcant degree of conservation between
elements in the ABIl signaling cascade(s) across species.

A novel PP2C gene encoding a protein highly related to ABII gne has been isolated
from Arabidopsis and was mapped to the abi2 locus. A mutation in the latter confers
ABA insensitivity similarly with the abil mutation and has dominant properties at least
under certain conditions (Leung et al., 1997 and references therein). Interestingly, the

sequence of the A811 homolog from the abi2 mutant revealed that it had the same Gly—9

Asp substitution identical to that found in the abil -1 allele (Leung et al., 1997). Further

136

sets of experiments conﬁrmed that the identiﬁed ABIl homolog was indeed the A312 gene
as well as a PP2C (Leung et al., 1997). It has been proposed that ABll and ABIZ
function in overlapping but not identical pathways of ABA signaling (Leung et al., 1997).
The above examples illustrate the important roles that PP2Cs play in plant
signaling systems. The LjNPP2C1 cDNA was shown to encode a PP2C enzyme by
biochemical and genetic criteria (Kapranov et al., 1999; chapter 3, this thesis). The
expression pattern of LjNPP2C1 mRNA resembled that of a marker gene for late stages of
nodule development, leghemoglobin (Kapranov et al., 1999; previous chapter in this
thesis). Therefore, LjNPP2C1 was classiﬁed as a late-nodulin gene and was hypothesized
to perform speciﬁc regulatory function(s) during the late stages of nodule development
and/or nodule function. (Kapranov et al, 1999; chapter 3, this thesis). Two independent
approaches were taken to address the function or role of this gene in symbiotic
interactions. 1. Generation of transgenic L. japonicus plants with altered expression of
LjNPP2C1 mRNA or expressing a dominant-negative form of this enzyme. 2. Screen for
proteins interacting with LjNPP2C1 using yeast two-hybrid system. In this chapter, I

will summarize the results of these experiments.

137

4.3. MATERIALS AND METHODS

Plant Growth Conditions

Wild-type L. japonicus ecotype Gifu plants were grown as described (Kapranov
et al., 1997; Szczyglowski et al., 1997). The transgenic L. japonicus plants were
inoculated with Rhizobia loti strain 2235 (Jarvis et al., 1982) before moving from tissue
culture to soil and grown under the same conditions. Transgenic plants were watered
with B&D nutrient solution (Broughton and Dilworth, 1971) supplemented with 1 mM
KNO3. The nodulation and general phenotype of transgenic plants were evaluated starting
from 2-3 weeks aﬁer inoculation. Subsequently the transgenic plants were moved to a rich

soil and grown for seed production.

Isolation of a A phage clone containing the LjNPP2C1 locus
The genomic library from L. japonicus ecotype Gifu was constructed in the FIX II
A phage vector by Stratagene (La Jolla, CA) and was kindly provided by Dr. Jens

Stougaard, Aarhus, Denmark. The library was screened with a LjNPP2C1 cDNA as a

probe following the manufacturer's protocols.

Plasmid Construction for the Transgenic Plants

The LjNPP2C1 cDNA used for the anti-sense and over-expression experiments

contains the entire 1086 bp LjNPP2C1 coding region with additional 23 bp of 5'UTR and

138

129 bp 3'UTR regions. The cDNA clone was originally cloned in pBluescript SK' vector
in the anti-sense orientation with respect to LacZ promoter.

The construct pBIPCR3anti, containing the LjNPP2C1 cDNA in the anti-sense
orientation with respect to 358 CAMV promoter, the cDNA was isolated as a 1.2 kb
BamHI-Kpnl fragment and the Kpnl site was made blunt by treatment with T4 DNA
polymerase. The DNA fragment was then cloned into pBIlZl binary vector (Clontech).
The vector was digested with BamHI-Sacl to remove the GUS gene and the SacI site was
made blunt by treatment with T4 DNA polymerase. For over-expression experiments,
the LjNPP2C1 cDNA was placed behind the 358 CAMV promoter in the sense
orientation in the construct pBIPCR3§ense, engineered as follows. The 1.2 kb
LjNPP2C1 cDNA was isolated as a EcoRI-Kpnl fragment, the Kpnl site was made blunt
by T4 DNA polymerase and the fragment was cloned into pB1121 digested with EcoRI-
Smal.

The abi-I-type substitution was introduced in the LjNPP2C1 cDNA as
previously described (Kapranov et al., 1999; previous chapter in this thesis). The mutant
cDNA was isolated as a 1.2 kb BamHI-Kpnl fragment and the Kpnl site was made blunt
by treatment with T4 DNA Polymarase. The DNA fragment was subsequently
subcloned in pBluescript 11 SK', digested with EcoRV-Sacl, with the Sacl site also blunt-
ended by T4 DNA Polymerase. As, a result, the plasmid pBSPCR3M, containing the
mutant LjNPP2C1 cDNA in the anti-sense orientation with respect to the lacZ promoter,
was selected to facilitate the future cloning steps. The 1.2 kb cDNA fragment was excised

from pBSPCR3M by digestion with EcoRI-Hindlll and the Hindlll site was ﬁlled-in with

139

Klenow DNA Polymerase. The cDNA was then subcloned in pBIlZl previously
digested with EcoRI-Smal to generate construct pBIPCR3M.

A 7.5 kb EcoRI fragment from the L. japonicus genome containing the
LjNPP2C1 gene with 4.3 kb of 5' and 0.5 kb 3' ﬂanking regions was used in all the
constructs designed to over-express LjNPP2C1 mRNA under the control of the cognate

regulatory elements. The 7.5 kb EcoRI fragment was derived from a l. phage isolated

with the LjNPP2C1 cDNA probe. The pBIlOl binary vector was digested with EcoRI-
Xbal to remove the GUS gene, ﬁlled-in with Klenow DNA Polymerase and re-ligated to

generate pBIl01AEX. The 7.5 kb EcoRI fragment was subcloned into pBllOlAEX

digested with EcoRI and de-phosphorylated to generate construct pBIGen2. The 358
CAMV promoter was excised as Hindlll-BamHI fragment from pB1121 and cloned into
pBluescript SK', digested with Hindlll-BamHI, to generate the construct pBS3SS. The
7.5 kb EcoRI fragment representing the LjNPP2C1 gene was ﬁlled-in with Klenow and
ligated into pBS3SS, digested with Xbal-EcoRV. The Xbal site was ﬁlled-in with
Klenow DNA Polymerase and the vector was dephosphorylated. The resulting construct
pBSGen235E contains the 358 CAMV enhancer sequence (-90...-800) fused to the 5'
ﬂanking regions of the LjNPP2C1 gene. The 7.8 kb DNA fragment containing this fusion
was excised with EcoRl-Clal and the Clal site was ﬁlled-in with Klenow DNA
Polymerase. The ﬁ'agment was then cloned into pBIlOl binary vector digested with

SmaI-EcoRl to generate construct pBIGen235E.

140

The 7.5 kb EcoRI fragment representing the LjNPP2C1 gene was subcloned into
pBluescript SK' vector and the abi—I-type substitution was incorporated in it to generate
plasmid pBSGenZSDM. The 7.5 kb region was excised from it by EcoRI digestion and

subcloned into pBIlOlAEX, digested with EcoRI and dephosphorylated, to generate

construct pBISDM. The pBISDM35E construct, containing the 358 CAMV enhancer

fusion to the mutant LjNPP2C1 gene in pBIlOl, was generated in the same way as the

pBIGen235E construct.

Plasmid Construction for the Yeast Two-Hybrid Screen

The 1.2 kb LjNPP2C1 cDNA was excised as a BamHI-Kpnl fragment and treated
with Klenow and T4 DNA Polymerases to blunt both sites. The ﬁagment was then
cloned into the pASZ-l vector (Clontech) digested with BamHI-Pstl, also treated with
Klenow and T4 DNA Polymerases and dephosphorylated. The resulting construct
pASPCR3 contains the GAL4 DNA Binding Domain (BD) fused to the complete
LjNPP2C1 coding sequence (BD-LjNPPZCl). The same BamHI-Kpnl fragment was also
cloned in the pACT2 vector (Clontech), digested with BamHI, ﬁlled-in with Klenow
DNA Polymerase and dephosphorylated. The resulting construct pACTPCR3 contains
the GAL4 Activation Domain (AD)-LjNPP2C 1 fusion (AD-LjNPP2C1).

The pASPCR3 construct was digested with Ndel, ﬁlled-in with Klenow DNA

Polymerase and re-ligated to generate a frame-shiﬁ between the GAL4-BB and

LjNPP2C1 domains (BD-LjNPP2C1(Fr)). The plasmid pASPCRBAN was engineered to

141

contain the GAL4-BD fused to the PP2C catalytic domain of LjNPP2C1, lacking the ﬁrst

96 amino acids of the protein (BD-LjNPPZClAN). To construct it, the 0.9 kb Bsin-

Xhol region of the LjNPP2C1 cDNA was isolated, ﬁlled-in with Klenow DNA
Polymerase and cloned into pASZ-l , digested with Smal and dephosphorylated.

The mutant LjNPP2C1 cDNA with the abi-I-type substitution was isolated as a
1.2 kb Hindlll-EcoRI fragment from pBSPCR3M (see above) and both sites were ﬁlled-in
with Klenow DNA Polymerase. The DNA ﬁagment was then cloned blunt-end into the
pAS2-1 vector digested with Ndel, ﬁlled-in with Klenow DNA Polymerase and
dephosphorylated to generate pASPCR3M construct containing BD-LjNPPZCl-M
fusion.

The construct pASLjPP2C2, containing the fusion between the GAL4BD and
LjPP2C2 (Kapranov et al., 1999; previous chapter), was generated as follows. The
coding region of LjPP2C2 cDNA was ampliﬁed using primers DB623 (5'-
ATGACTGGCAGAGAGATTCTC-3') and DB624 (5'-
TCACTGAAGTCTCACAACAAC-3'). The PCR product, treated with Klenow DNA
Polymerase and T4 polynucleotide kinase, was cloned blunt-end into the pASZ-l vector
digested with BamHI-Pstl, treated with Klenow and T4 DNA Polymerases and
dephosphorylated. To generate the GAL4BD-Nljl6 ﬁlSlOl‘l, the 0.5 kb region of
LjNODI6 cDNA (Kapranov et al., 1997) was excised with BglII-Hindll and treated with
Klenow DNA Polymerase. The DNA fragment was then subcloned blunt-end into the

pASZ-l vector digested with BamHI-Pstl, treated with Klenow and T4 DNA

142

Polymerases and dephosphorylated, to generate construct pASPCR5. The plasmids
containing the BD-laminin and BD-p53 fusions were provided with the
MATCHMAKERII system.

The pAD-LBP plasmid was digested with EcoRI, ﬁlled-in with Klenow DNA
Polymerase and re—ligated to generate a construct containing a frame-shift betWeen the
GAL4AD and the LjNPP2C1 Binding Protein (LBP) domains (AD-LBP(Fr)). The
construct pASLBP, containing the GAL4BD-LBP fusion (BD-LBP), was created as
follows. The 1.1 kb LBP cDNA was excised from pLBP with EcoRI-Xhol and ligated
into pASZ-l digested with EcoRI-Sall.

All constructs were veriﬁed by DNA sequencing.

Northern blot Analysis

Total RNA was extracted from different tissues of L. japonicus as previously
described (Kapranov et al., 1997; Szczyglowski et al., 1997). Ten micrograms of total
RNA was separated on agarose-fonnaldehyde gel, transferred to nitrocellulose and
hybridized with radiolabelled probes under the conditions described previously
(Kapranov et al., 1997; Szczyglowski et al., 1997). The signal intensities were quantiﬁed

using Phospholmager(Molecular Dynamics, Boulder, CO).

Generation of transgenic L. japonicus plants

143

The constructs described above were conjugated into the Agrobacterium
tumefaciens strain LBA4404 using tri-parental mating. The transgenic plants were
generated using the modiﬁcation of the original hypocotyl-transformation method

(Handberg et al., 1994) essentially as described by Stiller et al., 1997.

Yeast Two-Hybrid Screen

The L. japonicus nodule cDNA library, used for the screen, was kindly provided
by Dr. Carsten Poulsen, Aarhus, Denmark. The library was constructed by Stratagene in
the HybriZAP vector and than converted into the pAD-GAL4 library by mass in vivo
excision. The two-hybrid screen was performed using the CG-1945 yeast strain from
MATCHMAKER II system (Clontech, Palo Alto, CA) essentially following the
manufacturer's instructions. The quantitative B-galactosidase assays were performed on
liquid yeast cultures using o-nitrophenyl B-D-galactopyranoside (ONPG) as a substrate
following the manufacturer's protocol. One unit of B-galactosidase was deﬁned by

hydrolysis of 1 pmol of ONPG per minute per cell. All other manipulations with yeasts

were also performed essentially following the manufacturer's protocols.

144

4.4 RESULTS

Transgenic L. japonicus plants do not have an obvious phenotype.
The T-DNA regions of the constructs introduced into transgenic L. japonicus
plants are schematically shown in Figure 1. Table 1 summarizes the number of

independent transgenic lines obtained for each construct.

 

 

 

 

 

 

 

 

 

Table 1

Construct Number of lines
pBIPCR3Anti 69

pBlPCR3 Sense 34

pBIPCR3M 45

pBIGen2 15
pBIGen235E 10

pBlSDM l4
pBISDM35E 8

 

 

 

None of lines displayed any obvious alterations in the symbiotic phenotype
compared to the wild-type L. japonicus that could be correlated with the presence of
transgenes. Several transgenic lines showed morphological abnormalities commonly

observed in the tissue culture-derived plants: partial sterility, dwarﬁsm, abnormal ﬂower

145

morphology etc. Similar abnormalities were also seen in transgenic L. japonicus plants

containing unrelated constructs and most likely can be attributed to somaclonal variation.

Yeast two-hybrid screen

For unknown reasons, the CG-1945 strain harboring pASPCR3 had a much lower
transformation efﬁciency then expected. Therefore, only 150,000 yeast transformants
were screened in total. Unexpectedly, most of the primary transformants could grow and
form variable-size colonies on a media lacking histidine even in the presence of the
inhibitor of HIS3 enzyme, 3-amino-l,2,4-triazole (3-AT; 5-15 mM). A similar
phenomenon was observed when a different PP2C, ABIl, was used as bait in a two-
hybrid screen (Dr. Jeff Leung, personal communication). Therefore, the selection based
on the activation of HIS3 gene could not be reliably used. To circumvent this problem,
the activation of the second reporter gene lacZ was used as a sole indication of an
interaction at the ﬁrst round of screening. The B-galactosidase activity was determined
directly on the plates with the primary transformants using the ﬁlter-lift assay. Two
colonies exhibiting the most intense blue staining where chosen for further analysis. The
pAD-GAL4 plasmids were isolated from these colonies and shown to contain an identical
cDNA insert, designated LBP (LjNPP2C1 Binding Protein). The original pAD-GA14

plasmid containing this insert will be referred to as pAD-LBP.

146

Since the yeast two-hybrid screen is prone to false-positives, a number of
additional controls were performed. Strain CG-1945 was co-transformed with the pairs
of plasmids containing different GAL4BD and AD fusions and the activation of HIS3 and
lacZ genes reporter genes was evaluated as indicated in Fig. 4.2A&B. For each plasmid
pair, four independent transformants were streaked out on the media without histidine
supplemented with 5-15mM 3-AT to score the histidine-independent growth. The
representative results for each GAL4BD and AD fusion pair are shown in Fig. 4.2A. As

an independent measure of interaction, the B-galactosidase acitivity was measured

quantitatively for each of the four independent transformants and the averages for each
pair of GAL4Bd and AD fusions are shown in Fig. 4.28. The expression of HIS3 and
lacZ genes could be activated when in the strain expressing both the AD-LBP fusion and
the fusion between BD and the catalytical PP2C domain of LjNPP2C1 lacking the N-
tenninal extension. LBP and LjNPP2C1 could activate the expression of the reporter
genes when switched to another GAL4 domain, i.e. BD-LBP and AD-LjNPP2C1. LBP
could not activate the expression of HIS3 or lacZ gene by itself in the presence of the
pAS2-1 vector expressing only the GAL4BD. As expected, a frame-shill engineered
between a GAL4 domain and either LBP or LjNPP2C1 abolished the activation.

The abi-I-type mutation is known to signiﬁcantly decrease the catalytical
activity of both A811 and AB12 PP2Cs (Bertauche et al., 1996; Leung et al., 1997).
Interestingly, the abi-I-type substitution, introduced into the LjNPP2C1 domain of the

BD-LjNPPZCl fusion, abolished the activation of HIS3 and lacZ genes when co-

147

expressed with the AD-LBP fusion. The result shows that this mutation may prevent the
interaction between the LjNPP2C1 and LBP proteins.

No activation was observed when the AD-LBP fusion was co-expressed with BD
fused to related PP2C from L. japonicus, LjPP2C2 (Kapranov et al., 1999; previous
chapter) or to any other unrelated protein, such as Nlj l6 (Kapranov et al., 1997), laminin
and p53. Taken together, these results strongly suggest that LBP speciﬁcally interacts

with LjNPP2C1 in the yeast two-hybrid system.

LBP is a novel protein

The pAD-LBP plasmid contains a 1.1 kb cDNA with a major ORF of 183 amino
acids fused in-frame with the GAL4-AD. The BLAST search showed that LBP is related
to a number of hypothetical Arabidopsis proteins (40-70% identity and 59-88%
Similarity, data not shown). Based on the sequence similarity, it is likely that LBP cDNA
contains almost the entire LBP coding region and probably lacks several N-terminal amino
acids (data not shown). LBP did not show a signiﬁcant similarity to proteins with known
functions. It had a weak similarity (E-values 0.095-3.0, Fig. 4.38) to a number of
ribOnucleases and pathogen induced proteins from plants, such ginseng ribonuclease 2
(Moiseyev et al., 1997; Fig. 4.38). The similarity was restricted to a region of
approximately 90 amino acid including a conserved amino acid stretch GDLGIGSVR (Fig.
4°3B)- The latter motif may constitute a P-loop region implicated in binding nucleotides,

SUCh as ATP or GTP (Saraste et al., 1990).

148

LBP mRNA is up-regulated in uninfected roots of L. japonicus

LBP mRNA is expressed in ﬂowers, leaves, stems and nodules of L. japonicus,
with the highest levels in uninfected roots (data not shown). The level of LBP mRNA
actually declines as the nodules mature (Fig. 4.4A&B). Interestingly, the level of LBP
mRNA in the uninfected roots of L. japonicus non-nodulating Nod' mutant LjEMS45
(Szczyglowski et al., 1999) is lower than in the wild-type roots or roots of other Nod'

mutants (Fig. 4.4 A&B).

149

4.5. DISCUSSION

The absence of a mutant phenotype in the transgenic plants containing various
LjNPP2C1 constructs may be due to a variety of different reasons. First of all, the major
alterations in the expression of this gene may be lethal. On the other hand, the LjNPP2C1
enzyme may not be essential for nodulation and other biological processes in L.
japonicus, for example a different PP2C enzyme can fulﬁl the function(s) of LjNPP2C1.
It is also possible that the anti-sense approach was not sufﬁcient to decrease the
expression of LjNPP2C1 gene below a threshold required to maintain the normal levels of
phosphorylation of LjNPP2C1 substrate(s). Whatever the reasons are, it is clear that in
order to address the biological function and signiﬁcance of LjNPP2C1, it is necessary to
generate a complete knock-out mutation in this gene. Only then would it be possible to
make direct, unambiguous conclusions as to the role of this PP2C in nodulation and other
plant processes.

A parallel and complementary approach to address the biological function(s) of
LjNPP2C1 is to identify the substrates of this PP2C. The yeast two-hybrid system
(Fields and Song, 1989) was used to screen for cDNAs encoding proteins that may
interact with LjNPP2C1. As a result, one candidate cDNA clone was isolated. The
protein product of this cDNA was tentatively designated LBP. The latter can interact
with the PP2C catalytic domain of LjNPP2C1 with or without the N-terrninal extension.
LBP and LjNPP2C1 can also interact when their corresponding GAL4 domains are
exchanged. A strong supportive evidence for the speciﬁcity of LjNPP2C1-LBP

interaction comes from observations that LBP can not interact with a related PP2C

150

protein, LjPP2C2, or with any other protein tested in the two-hybrid system.
Interestingly, the abi-I-type substitution can abolish the interaction, suggesting that
correctly folded and active PP2C domain is required for the interaction. Finally, as
expected, the interaction is dependent on the production of the GAL4 fusions with
LjNPP2C1 and LBP, since the frame-shift constructs are not active. All of the above
results strongly argue that LBP represents a bona ﬁde LjNPP2C1-interacting protein, at
least in the yeast two-hybrid system.

Unfortunately, the biological signiﬁcance of this interaction remains unknown.
LBP is a novel protein, with a weak similarity to ginseng ribonuclease l and 2 (Moiseyev
et al., 1997) and a family of proteins including pathogenesis-induced PR-lO proteins
(Wang et al., 1999) and related plant allergens (Acc. No. 049065). The region of
similarity includes a conserved amino acid stretch of LBP (see above) which is similar to
the P-loop regions of proteins binding ATP or GTP nucleotides (Saraste et al., 1990).
The importance of this motif for the biological ﬁrnction of a protein was suggested by the
X-ray crystal structure of a birch pollen allergen Bet v 1, related to the PR-lO proteins
(Gajhede et al., 1996).

The LBP and LjNPP2C1 genes have a different, even somewhat opposite patterns
of expression. The level of LBP mRNA is highest in the uninfected roots of L. japonicus
and decreases signiﬁcantly in the mature L. japonicus nodules, while the level of
LjNPP2C1 mRNA goes up ~20-fold (Kapranov et al., 1999). The activity of LBP may be
controlled by several independent mechanisms, including the regulation at the level of LBP

mRNA abundance and also post-translationally, by protein

151

phosphorylation/dephosphorylation. The optimal functioning of the L. japonicus
nodules may require the decrease in the level of LBP mRNA and concomitant
dephosphorylation of this protein via the increase in the LjNPP2C1 levels. On the other
hand, the LjNPP2C1 mRNA is present, albeit at low levels, in L. japonicus wild-type
roots and induced in the roots of L. japonicus Nod' mutant LjEMS46 (Kapranov et al.,
1999). It is therefore possible, that LjNPP2C1 interacts with LBP in the roots and
perhaps, other L. japonicus tissues as well. The LBP mRNA is also present in the roots
of LjEMS45 mutant line, albeit at slightly lower levels than in the roots of the wild-type
L. japonicus (Fig 4.4A&B). Finally, it is important to state that even if LjNPP2C1
interacts with LBP, the latter may not be a substrate of LjNPP2C1. Alternative
possibilities exist, for example, LBP may bind and regulate in some way the activity of
LjNPP2C1.

At present, it is difﬁcult to address the biological signiﬁcance of LBP-LjNPP2C1
interaction. First and foremost, the loss-of-function mutations in LBP and
LjNPP2C1 genes are needed in order to address the biological functions of both
genes. The transgenic L. japonicus plants expressing LBP anti-sense transcripts behind
the 35S CAMV promoter were generated (data not shown). No alterations in the
symbiotic or other plant phenotypes have been observed so far. Second, the biochemical
function of LBP has to be established, for example, whether it has an RNAase activity or

whether it can binds ATP or GTP nucleotides.

152

Figure 4.1. Schematic representation of the T-DNA regions of the constructs
used for transformation of Lotus japonicus. A, Constructs designed to
(over)express the wild type LjNPP2C1 mRNA under control of the cognate
regulatory elements with and without 35$ CaMV enhancer (pBIGen235E and
pBIGen2, respectively), and under control of a strong constitutive 35$ CaMV
promoter (pBlPCR3Sense). B, Construct with the wild type LjNPP2C1 cDNA
cloned in the anti-sense orientation behind the 355 CaMV promoter. C,
Constructs designed to (over)express the mutant form of LjNPP2C1 mRNA under
control of the cognate regulatory elements with and without 358 CaMV enhancer
(pBlSDM35E and pBlSDM, respectively), and under control of 358 CaMV
promoter (pBIPCR3M). NOSp- nopaline synthetase promoter, NOSt-nopaline
synthetase polyadenylation signal sequence, NPTIl-neomycin
phosphotransferase gene, 3SSp- 353 CaMV promoter, 3SSe- 3SS CaMV
enhancer. LjNPP2C-Gen represents the 7.5 kb fragment containing the LjNPP2C1
coding region (represented by the arrow) and the 5' and 3' ﬂanking regions.
Asterisk denotes the presence of the Glyl33 to Asp substitution (also indicated

by the arrow).

153

A

pBIGen2

pBIGen235E

pBlPCR3sense

B

pBlPCR3Anti

C

pBlSDM

pBlSDM35E

pBlPCR3M

N PTII LjNPP2C-Gen
N PTII LjNPP2C-Gen
NOSp NOSt 3SSen
LjNPP2C
NPTlI CDNA
NOSp NOSt 353p
LjNPP2C
NPTII °DNA
NOSp NOSt 358p N051

LjNPP2C“-Gen

NPTII
NOSp NOSt 7
01331)

LjNPPZ C *-Gen

NPTll
NOSp NOSt 3SSen
Gl33D
LjNPPZC"
NPTll CDNA
NOSp NOSt 358p
61330

154

Figure 4.2. LBP interacts speciﬁcally with LjNPP2C1 in the yeast two-
hybrid system. (A), The growth with and without histidine of the yeast

strains expressing the indicated GAL4 BD and AD fusions.

155

             

Ana-3. + and-am

.::.-A2 + 352.5.

+ mud-QM—

        

52.2.3-3

andun< + Gm

mad-Qt.

+ 2-593235:

.3.—-3
+ €552,235:

 

AME-9a .::-:
+ 5333.3: + 532-3.
2.5.—3-3

.::-:
+ 552.235:

.::.-3
+ 52.2.:

+ 52.2.3-5.

inn

H<im+m=mi

 

156

 

 

 

 

 

 

 

Amzczv .0358 omme_moeom_ew-a

 

Ami—$395
uni—$3335”.

emu-31.22::

mm..-:<+5~.r_3-nm

mmq-:<+z-5§._z.3-am

mag-afeE—urzzﬁdm

cE..5-:<+58az_-_ém
emu-35m
5~ma23¢<¢§qém
..3-:<+z<5~.:z_-_-nm

..5-a<+5~.§z£-am

Figure 4.28. LBP interacts speciﬁcally with L jNPP2C1 in the yeast

two-hybrid system. The average B-galactosidase activity measured for

each pair of the indicated GAL4 BD and AD fusions. The bars represent

conﬁ dence intervals (p<0.05).

157

Figure 4.3. The LBP cDNA encodes a protein with a weak similarity to
ginseng ribonuclease 2. (A) Nucleotide and deduced amino acid sequences
of LBP mRNA. The DNA sequence shown lacks the 5' linker
(GAATTCGGCACGAG) used to clone the cDNA in-frame with the

GAL4-AD region. The asterisk indicates a stop codon.

158

71

141

211

281

351

421

491

561

631

701

771

841

911

981

1051

TGGTGATGAA CCTTACAGTG CTATTGAGTC ACAGTACATT AGGAGACATC ACAAGCATGA ACTCAGGGAC
G D E P Y S A I E S Q Y I R R H H K H E L R D
AATCAGTGCA CTTCTGCACT TGTCAAACAC ATAAAGGCCC CTGTTCATCT TGTGTGGTCT CTGGTTAGAA
N Q C T S A L V K H I K A P V H L V W S L V R
GATTTGATCA GCCTCAGAAG TATAAACCAT TTGTGAGCAG GTGTATCATG CAAGGAGACC TTGGCATTGG
R F D Q P Q K Y K P F V S R C I M Q G D L G I G
AAGTGTTAGA GAAGTGAATG TTAAATCTGG TCTTCCAGCA ACAACTAGCA CTGAGAGGTT GGAACAGCTT
S V R E V N V K S G L P A T T S T E R L E Q L
GATGATGAGG AACACATCCT TGGTATCAGG ATTGTTGGAG GTGATCACAG GCTCAGGAAC TATTGTTCCA
D D E E H I L G I R I V G G D H R L R N Y C S
TAATCACTGT CCATCCGGAG GTCATCGATG GAAGACCCGG TACCATGGTG ATTGAGTCAT TTGTGGTGGA

I I T V

H P E

V I

TGTGCCTGAA GGGAACACCC GAGATGAAAC

V P E

G N T R

D E T

TCTTTGGCTG ATGTCTCAGA GAGGATGGCC

S L A D

TAAGGGAGAG

TTGGATTAGT

GCAACCATCC

TGTAAGATGA

ATTCTTCCTG

GTTTTTCATG

ACTGCCTTGT

CTTGTTTTAG

V S E
TGAAGTGTTG

CCTTCCGTGT

TCCTACAATT

GATTCTGGAT

GTGGTGTGTT

TTTAGTGGCA

TACTTTGGTT

AACAAAAAAG

R M A
GGTATGTGAT

TCCCGAAAGT

TTAGCTGTGT

TTCAAAAGCA

ACTACAGTTA

ACTAGAAACT

TACTTTAAGT

AAAAAAAAAA

D G

R P G

T M V

I E S

TTGTTACTTC GTGGAGGCTC TGATCAGGTG

C Y F
GTGCAGGGTC

V Q G R

CCATTGCATT

TTTGGATGTT

ATTTCTCTTT

CACAGGGATG

TTTGTAAATA

ATTATGTGCT

AATGTTGTAC

AAAAA

159

V E A L

GAA CC GAT CC

T D P

CATTCTGCCC

CATTTTCCTC

TGAACAGCTG

GGAGTGGGGG

TTTTTGTTGG

GGTGGGAGAG

TAAANTGTAA

I R C
GCTCAACCAT

L N H

AGTTCAAGTG

TTATCGTGGA

AGTCTCTCTT

AAGTTGTTTT

AAAAAAAAGA

GTCCCCAAAC

GTTATCTATT

V V D

CAACCTAAGC
N L S

TAAGCATGTC

*

TGGAGATTGG

CACCTTCCCT

CTACAGTTTT

TTACTGCTAA

CATTTCTGTA

AGTAGATAGT

CCAAATTCTT

sp|P80890|RNSZ_PANGI RIBONUCLEASE 2
Length = 153

Score = 38.7 bits (88), Expect = 0.095
Identities = 21/89 (23%), Positives = 47/89 (52%), Gaps = 6/89 (6%)
Frame = +2
Query: 185 IMQGDLGIGSVREVNVKSGLPATTSTERLEQLDDEEHILGIRIVGGDHRLRNYCSIITVH 364
+++G+ G+G+++ V + P T R++ +D+ I+GGD L + I H
Sbjct: 43 VLEGNGGVGTIKNVTLGDATPFNTMKTRIDAIDEHAFTYTYTIIGGD-ILLDIIESIENH 101
Query: 365 PEVIDGRPGTMVIESFVVD ------ VPEGNTRDET 451

+++ G+ + ++ + + +pE N +D T
Sbjct: 102 FKIVPTDGGSTITQTTIYNTIGDAVIPEENIKDAT 136

Figure 4.3. The LBP cDNA encodes a protein with a weak similarity to ginseng
ribonuclease 2. (B) The similarity between LBP (query) and ginseng ribonuclease 2
(subject; Moiseyev et al., 1997) as detected by the BLASTX search algorithm. This
match ranks the highest among the similarities found between LBP and proteins of known

functions. The putative P-loop motif is indicated by the line drawn above it.

160

Figure 4.4. Expression of LBP mRNA at different stages of nodule
development and in the roots different Nod’ mutant lines of L. japonicus.
(A) Total RNA from wild-type L. japonicus roots and nodules harvested
at different times after inoculation with rhizobia, control wild-type
uninoculated roots, and total RNA from uninoculated roots of the 35day-
old LjEMS mutant lines was separated on agarose-formaldehyde gel and
transferred to nitrocellulose. The blot was sequentially hybridized with
the radioactive probes corresponding to LBP and ubiquitin cDNAs. (B)
Quantiﬁcation of the results in panel A. The volume of LBP and ubiquitin
signals was quantiﬁed using Phospholmager (Molecular Dynamics). For
each lane, the volume of LBP hybridization was divided by the volume of
ubiquitin hybridization and expressed in arbitrary units, with the

LBP/ubiquitin ratio for the 7d uninoculated roots set as l.

161

Uninoculated Days after
root (wt) inoculaton (wt)

 

LjEMS 34
LjEMS 46
LjEMS 70

7 ll 21 7 ll 21 35

a: 3,54% ,_v . ' A . v , .' 4 V ' "
’ .—' D 9 wif ‘37- ; ' ~ - . 1 BP

 

162

 

.L._m..“
Aabsnda
11 0000

32:. Panza...“

 

   

0

.5sz
3.55.:

4823.3
3333...: En

8.282: .5
83:85 .3:

cues—=62: E.

EOHN— 506—: :5 6 n N

ﬁvua— 506-: :5 6 w _

682.8255 3

163

4.6. REFERENCES

Adler E, Donella-Deana A, Arigoni F, Pinna LA, Stragier P (1997) Structural
relationship between a bacterial developmental protein and eukariotic PP2C protein
phosphatase. Mol Microbiol 23: 57-62

Armstrong F, Leung J, Grabov A, Brearly J, Giraudat J, Blatt M (1995) Sensitivity
to abscisic acid of guard-cell K+ channels is suppressed by abil -1 , a mutant Arabidopsis
gene encoding a putative protein phosphatase. Proc Natl Acad Sci USA 92: 9520-9524

Bertauche N. Leung J, Giraudat J (1996) Protein phosphatase activity of abscisic acid
insensitive 1 (ABIl) protein from Arabidopsis thaliana. Eur J Biochem 241: 193-200

Bogre L, Ligterink W, Meskiene l, Barker PJ, Heberle-Bors R, Huskisson NS, Hirt
H (1997) Wounding induces the rapid and transient activation of a speciﬁc MAP kinase
pathway. Plant Cell 9: 75-83

Braun DM, Stone JM, Walker JC (1997) Interaction 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

Broughton WJ, Dilworth MY (1971) Control of leghemoglobin synthesis in snake
beans. Biochem J 125: 1075-1080

Chin-Sang, I.D., and Spence, A.M. (1996) Caenorhabditis elegans sex-determining
protein FEM-2 is a protein phosphatase that promotes male development and interacts
directly with F EM-3. Genes and Dev 10: 2314-2325

Cohen P (1989) The structure and regulation of protein phosphatases. Annu Rev
Biochem 58: 453-508

Cohen P, Cohen PTW (1989) Protein phosphatases come of age. J Biol Chem 264:
21435-21438

Davies S P, Helps NR, Cohen PT, Hardie GT (1995) 5’-AMP inhibits
dephosphorylation, as well as promoting phosphorylation, of the AMP-activated protein

kinase. Studies using bacterially expressed human protein phosphatase-2C alpha and
native bovine protein phosphatase-ZAC. F EBS Lett 377: 421-425

164

Fields S, Song 0 (1989) A novel genetic system to detect protein-protein interactions.
Nature 340: 245-246

Gajhede M, Osmark P, Poulsen FM, lpsen H, Larsen JN, van Neerven RJJ, Schou
C, Lowenstein H, Spangfort MD (1996) X-ray and NMR structure of Bet v 1, the
origin of birch pollen allergy. Nature Struct Biol 3: 1040-1045

Handberg, K., Stiller, J., Thykjear, T., and Stougaard, J. (1994) Transgenic plants:
Agrobacterium mediated transformation of the diploid legume Lotus japonicus. In: Cell
Biology: A Laboratory Handbook, Celis JE, ed, 1, 119-127

Jarvis BDW, Pankhurst CE, Patel JJ (1982) Rhizobium loti, a new species of legume
root nodule bacteria. Int J Syst Bact 32: 378-3 80

Jonak C, Kiegerl S, Ligterink W, Barker PJ, Huskisson NS, Hirt H (1996) Stress
signaling in plants: a mitogen-activated protein kinase pathway is activated by cold and
drought. Proc Natl Acad Sci USA 93: 11274-11279

Kapranov P, de Bruijn FJ, Szczyglowski K (1997) A novel, highly expressed late
nodulin gene LjNOD16 from Lotus japonicus. Plant Physiol 113: 1081-1090

Kapranov P, Jensen TJ, Poulsen C, de Bruijn FJ, Szczyglowski K (1999) A protein
phosphatase 2C gene, LjNPP2C1, from Lotus japonicus induced during root nodule
development. Proc Natl Acad Sci USA 96: 1738-1743

Kuromori T, Yamamoto M (1994) Cloning of cDNAs from Arabidopsis thaliana that
encode putative protein phosphatase 2C and a human Dr—l like protein by transformation
of the ﬁssion yeast mutant. Nucl Acids Res 22: 5296-52301

Leung J, Bouvier-Durand M, Morris P-C, Guerrier D, Chefdor F, Giraudat J
(1994) Arabidopsis ABA response gene ABII: features of a calcium-modulated protein
phopshatase. Science 264: 1448-1452

Leung J, Merlot S, Giraudat J (1997) The Arabidopsis Abscisic Acid-Imensitive 2
(A812) and AB]! genes encode homologous protein phosphatases 2C involved in abscisic
acid signal transduction. Plant Cell 9: 759-771

Meskiene I, Bogre L, Glaser W, Balog J, Brandstotter M, Zwerger K, Ammerer G,
Hirt H (1998) MP2C, a plant protein phosphatase 2C, functions as a negative regulator
of mitogen-activated protein kinase pathways in yeast and plants. Proc Natl Acad sci
USA 95: 1938-1943

165

Maeda T, Wurgler-Murphy SM, Saito H (1994) A two-component system that
regulates an osmosensing MAP kinase cascade in yeast. Nature 369: 242-245

Meyer K, Leube MP, Grill E (1994) A protein phosphatase 2C involved in ABA signal
transduction in Arabidopsis thaliana. Science 264: 1452-1455

Moiseyev GP, Fedoreyeva Ll, Zhuravlev YN, Yasnetskaya E, Jekel PA, Beintema JJ
(1997) Primary structure of two ribonucleases from ginseng calluses. FEBS Lett 407: 207-
210

Ota IM, Varshavsky A (1993) A yeast protein similar to bacterial two-component
regulators. Science 262: 566-569

Pilgrim D, McGregor A, Jackle P, Johnson T, Hansen D (1995) The C. elegans sex-
determining gene fem-2 encodes a putative protein phosphatase. Mol Biol Cell 6: 1159-
1171

Rodriguez PL (1998) Protein phosphatase 2C (PP2C) function in higher plants. Plant
Mol Biol 38: 919-927

Rodriguez PL, Leube M, Grill E (1998) Molecular cloning in Arabidopsis thaliana of a
new protein phosphatase 2C (PP2C) with homology to AB“ and ABIZ. Plant Mol Biol
38: 879-883

Saraste M, Sibbald PR, Wittinghofer A (1990) The P-loop- a common motif in ATP-
and GTP-binding proteins. TIBS 15: 430-434

Shiozaki K, Russell P (1995) Conteractive roles of protein phosphatase 2C (PP2C) and
a MAP kinase kinase homolog in the osmoregulation of ﬁssion yeast. EMBO J 14: 492-
502

Smith RD, Walker JC (1996) Plant protein phosphatases. Annu. Rev Plant Physiol
Plant Mol Biol 47: 101-125

Stiller J, Martirani L, Tuppale S, Chian R-J, Chiurazzi M, Gresshoff PM (1997)
High frequency transformation and regeneration of transgenic plants in the model legume
Lotusjaponicus. J Exp Bot 48: 1357-1365

Stone JM, Collinge MA, Smith RD, Horn MA, Walker JC (1994) Interaction of a

protein phosphatase with an Arabidopsis serine-threonine receptor kinase. Science 266:
793-795

166

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

Szczyglowski K, Shaw SR, Wopereis J, Copeland S, Hamburger D, Kasiborski B,
Dazzo F B, de Bruijn FJ (1998a) Nodule organogenesis and symbiotic mutants of the
model legume Lotus japonicus. Mol Plant-Microbe Interact 11: 684-697

Wang CS, Huang JC, Hu JH (1999) Characterization of two subclasses of PR-lO
transcripts in lily anthers and induction of their genes through separate signal transduction
pathways. Plant Mol Biol 40: 807-814

Williams, RW, Wilson JM, Meyerowitz EM (1997) A possible role for kinase-

associated protein phosphatase in the Arabidopsis CLAVATAI signaling pathway. Proc
Natl Acad Sci USA 94: 10467-10472

167

CHAPTER 5

The Lotus japonicus LjNOD 70 nodulin gene encodes a protein with

similarities to transporters3

5.1. ABSTRACT

A novel nodule-speciﬁc gene, LjNOD70, associated with late stages in Lotus
japonicus nodule development and/or functioning was characterized. The LjNOD70 gene
is a member of a small family of closely related L. japonicus genes. Two major mRNA
species corresponding to the LjNOD70 gene were identiﬁed in nodules and shown to be
the result of a mechanism resembling alternative splicing. The longer, presumably
unspliced, mRNA species was shown to contain a single open reading frame (ORF),
encoding a polytopic hydrophobic protein, LjN70, with a predicted molecular mass of 70
kD. The second, presumably spliced, mRNA species was shown to be less abundant in
nodules. The absence of the presumptive “intron” was found to divide the reading ﬁ'ame
into an upstream- and a downstream ORF encoding the partial N- and C- terminal regions
of the LjN70 protein, respectively. The predicted amino-acid sequence of nodulin LjN70
revealed structural features characteristic of transport proteins, and was found to share

similarity with the oxalate/formate exchange protein of Oxalobacter formigenes.

 

3 This chapter was published in Szczyglowski K, Kapranov P, Hamburger (I, de Bruijn

FJ (1998) Plant Mol Biol 37: 651-661.

168

Therefore, we postulate that the L. japonicus LjNOD70 gene family encodes nodule-

speciﬁc transport proteins, which may have evolved as a result of exon—intron shufﬂing.

169

5.2. INTRODUCTION

A carefully orchestrated integration of regulatory mechanisms of plant host cells
and endophytic bacteria of the genera Rhizobium, Bradyrhizobium, Sinorhizobium, and
Azorhizobium is responsible for the formation of nitrogen-ﬁxing nodules. These nodules
develop as a result of a speciﬁc and compatible interaction between plant and microbe and
involve the highly coordinated induction of both bacterial- and plant-encoded genes.
Lipochito-oligosaccharide signal molecules of bacterial origin, the Nod factors, initiate
nodule morphogenesis [3, 36]. This process involves the redirection of differentiated root
cortical cells towards formation of the nodule primordium [36; and references therein].
The induction and/or enhancement of expression of plant genes called early nodulin genes
has been found to be associated with early events in nodule ontogeny [for recent reviews,
see 24, 32]. It has been postulated that early nodulin gene products play a role in
preinfection, infection and nodule structure formation [24]. However, the exact function
and requirement for nodulation of the majority of the early nodulin genes have not yet
been elucidated [4]. Moreover, expression of some of the early nodulin genes has been
shown to persist in fully developed nodules [e.g., ENOD40; 22] suggesting their
involvement in processes, which also occur during late stages of nodule development
and/or in nodule functioning.

The “late” events in nodule organogenesis initiate with the successﬁll colonization

of root cortical cells by symbiotic bacteria. This, in turn, leads to the ﬁnal differentiation

170

of the infected and uninfected cells of the nodule and the commencement of nitrogen
ﬁxation [3 7]. The major molecular and biochemical alterations occurring in nodules prior
to, or concurrent with the initiation of nitrogen ﬁxation, appear to be involved in
supporting the physiological requirements for nitrogen ﬁxation and ammonia assimilation
[24, 44]. Since the physiological environment within the nodule is clearly different from
that in other parts of the plant, several new functions have had to evolve to facilitate
symbiotic nitrogen ﬁxation. On the molecular level, these adaptive changes manifest
themselves in the expression of a number of plant genes associated with late stages of
nodule development and/or nodule functioning. These genes have been classiﬁed as late
nodulin genes [5, 25]. Members of this group include genes encoding enzymes, or
subunits of enzymes, involved in carbon and nitrogen metabolism; proteins associated
with peribacteroid membrane biogenesis and functioning; oxygen carrying leghemoglobins;
putative plant transcription factors; and a number of proteins of unknown function [7, 8,
12,13,17,39]

A continuous molecular cross-talk between plant cells and their microbial
organelle-like structures, the symbiosomes, must occur to ensure an appropriate
partitioning of the molecular clues and metabolic components involved in symbiotic
nitrogen ﬁxation [27, 41]. In addition, an appropriate translocation of organic substances,
derived from photosynthetically active tissues, to nodules is required to provide the
essential carbon and energy sources for symbiotic nitrogen ﬁxation [41, 43]. Mechanisms
for the metabolism and transport of ﬁxed nitrogen from the nodules to other parts of the

plant are also essential [41]. It is apparent that the proteins involved in the transport of

171

different compounds to and from the nodules, as well as within the nodules, are largely
unknown. Recently, it has been suggested that the soybean symbiosome membrane
protein nodulin-26 forms a pore responsible for the uptake of ions or small metabolites
[23, 45-47]. Moreover, a soybean nodulin cDNA encoding a putative sulphate
transporter has been described [18, 31]

We have previously described the identiﬁcation of a range of novel expressed
sequence tags (ESTs) of the model legume Lotus japonicus associated with late
developmental events during nodule organogenesis in [17, 39]. The deduced amino-acid
sequence of one of these nodulins, named Nlj65, revealed a signiﬁcant similarity to the
ATPTRZB peptide transporter from Arabidopsis thaliana [35, 38], suggesting its
involvement in metabolite translocation [39]. Here we report the molecular
characterization of an additional gene from L. japonicus, LjNOD70, encoding a protein
with distinct topological features of transport proteins, that may be involved in nodule-

speciﬁc metabolite transport.

172

5.3. MATERIALS AND METHODS

Plant material

Seeds of Lotus japonicus B-129-S9 ecotype Gifu and B-581 ecotype Funakura
(kindly provided by Dr. Jens Stougaard, Aarhus University, Denmark) were germinated
according to the procedure described by Kapranov et a1. [17]. One week old L. japonicus
Gifu seedlings were inoculated with Rhizobium loti strain NZP2235 [39], and grown
under previously described conditions [17]. For the developmental Northern blot
analysis segments of control (uninfected) roots and corresponding portions of infected
roots or nodules were harvested 7, ll, l3, l7, and 21 days after sowing and/or inoculation
with Rhizobium, frozen in liquid nitrogen, and stored at -80°C. L. japonicus Giﬁl ﬂower,
leaf, stem, root, and nodule tissues were harvested from 21 day old infected plants. The
leaf material used for genomic DNA preparations was collected from mature, 8 weeks old,

L. japonicus Gifu and Funakura plants.

Nucleic acid isolation, Southern and northern analyses
Genomic DNA was extracted from plant tissues by the method of Rogers and
Bendich [26]. Southern blot analysis was carried out following standard protocols [30].

For northern analyses, 10 pg of total RNA was separated on agarose gels under

denaturing conditions, transferred to nitrocellulose membranes (Fisher Scientiﬁc), and

hybridized using previously described conditions [17, 39]. Random primer kits

173

(Boehringer Mannheim) were used to label probes with [a-32P]dATP, following the

manufacturer’s instructions.

RT-PCR procedures

Reverse transcription (RT)-PCR was performed as follows: 0.6 pg of DNAseI

treated total RNA derived from uninfected control roots (7 and 21 day old), and infected
roots and nodules of L. japonicus Gifu (7, 1 l, and 21 days after inoculation) was reverse
transcribed in the presence of 30 pM of a reverse PCR primer (p2) 5’-
GGGTGAAGTTGAAAGATATTTGCA-3’, in RT buffer (100 mM Tris HCl, pH 8.3,
10 mM MgC12, 10 mM DTT, 50 mM KCl), in the presence of 1 mM dNTPs, 40 units of
placental RNAse inhibitor (RNasin, Boehringer Mannheim) and 4 U of AMV reverse

transcriptase (Gibco), in a total volume of 20 pl. Negative control reactions were carried

out under identical conditions, in the absence of AMV-RTase. For the PCR ampliﬁcation

reactions, 3 pl of the ﬁrst strand cDNA synthesis products, 6 pmol of the forward (pl)

5’-CTTAGAAAATGATGTTTTGTGCAC-3’, and reverse (see above) PCR primers,
were incubated in the presence of 100 mM Tris pH 8.3, 75 mM KCl, 1.5 mM MgC12, 0.2
mM dNTPs and 1 unit of Taq polymerase. For semi-quantitative RT-PCR experiments
the ampliﬁcation reactions were carried out for 20 cycles (1 min, 94°C; 1 min, 58°C; 1
min, 72°C,); otherwise the ampliﬁcation was allowed to proceed for 35 cycles. Products
of the semi-quantitative RT-PCR were resolved on 1.2% agarose gels, transferred to

Nylon membranes (DuPont) and hybridized with the radiolabelled insert of cDNA 48-23.

174

Intensities of the hybridizing fragments were compared using a Phospholmager

(Molecular Dynamics)

cDNA library screening and DNA sequence analysis

A L. japonicus cDNA library prepared from mature nodule mRNA, kindly
provided by Dr. Jens Stougaard (Aarhus University, Denmark), was used. The library
was constructed in the UniZAP lambda vector (Stratagene) using an oligo-dT primer.
Screening of the library for full copy cDNAs corresponding to the EST LjN48 product
was carried out following standard techniques [30; Stratagene manual, Stratagene].

DNA sequencing was performed using Sequenase 2.0 kits (United States
Biochemical) following the manufacturer’s instructions. Homology searches were
performed using BLAST software [2]. Sequence alignments were carried out using the
GCG package (Genetics Computer Group, Madison, WI). Hydropathy analysis and
secondary structure predictions of protein sequences were carried out using
LASERGENE (DNASTAR, Madison, WI) and PredictProtein software [29],

respectively.

Generation of anti-LjN70 antibodies
Antibodies were generated against a polypeptide, LjN70-C, comprised of the
amino acids 469-575 of the nodulin LjN70 fused to the histidine-tag. To construct this

fusion, the 0.6 kb BerI-Xhol fragment of the LjNOD70 cDNA was isolated and blunt-

ended with Klenow DNA Polymerase. The fragment was then cloned blunt-end into the

175

pETlSB vector (Novagen), previously digested with Xhol, ﬁlled-in with Klenow DNA
Polymerase and dephosphorylated. The LjN70-C polypeptide was initially puriﬁed by
nickel-afﬁnity chromatography following the standard protocols provided by
manufacturer (Novagen). Subsequently, the polypeptide was ran on a preparative 15%
SDS-PAGE and stained with Coomassie blue. The region of the gel containing the LjN70-
C polypeptide was excised and sent to Cocalico Biologicals (Reamstown, PA) for
antibody production in rabbits following the standard protocols used by this company.
The anti-LjN70 serum was afﬁnity puriﬁed using the LjN70-C polypeptide immobilized

on nitrocellulose membrane as described in Harlow and Lane [1 1].

Preparation of peribacteroid membranes

The following experiments were performed by Jim Guenther in the laboratory of Dr. Dan
Roberts, University of Tennessee, Knoxville. Nodules were harvested from
approximately nine week old plants and processed as in Weaver and Roberts, 1992 [47].
Brieﬂy, the nodules were crushed in ice cold extraction buffer (25 mM Mops NaOh, pH
7.0, 350 mM mannitol, 10 mM MgSO.,,, 5 mM dithiothreitol, 1% (w/v)
polyvinylpyrrolidone-40, l % (w/v) BSA, 17 mM Na-isoacorbate, and 10 mM EDTA)
and the resultant extract was fractionated on a stepwise Percoll gradient as described
previously [45, 47]. The 60-80% Percoll interface was collected and transferred to a 30
m1 Corex tube and symbiosomes were pelleted at low speed (650 x g, 3.5 min).

Bacteroids were then released by vortexing for two minutes, and removed by

176

centrifugation at 7,000 x g for ten minutes at 4° C. Membranes were then pelleted from

the supernatant by centrifugation for 1 hour at 110,000 x g in the Ti 50 rotor.

Western Blotting
Lotus japonicus nodule membrane samples were separated on 15% SDS-PAGE gel and
transferred onto PVDF membranes overnight using a trans blot cell (BioRad) at 120mA.

Blots were blocked with 10% non fat dry milk in PBS for 1.5 hours while shaking at

37°C, then washed three times with PBS-Tween (PBS containing 0.05% Tween 20), ten

minutes each time, at room temperature, while shaking. The blots were placed in a 1:250
dilution of the afﬁnity-puriﬁed anti-LjN7O antibody in PBS Tween with 1% non fat dry

milk for 1 hour while shaking at 37°C. Blots were then washed as above with PBS

Tween, and incubated in a 1:2000 dilution of alkaline phosphatase labeled goat anti rabbit
secondary antibody (supplied by ICN, Aurora western blotting kit), and incubated under
the same conditions as above for the primary antibody incubation. Blots were washed
three times as above in PBS Tween, and then three ﬁnal washes were performed for 15
minutes each, at room temperature, in PBS Tween with .1% Triton X100, 0.05% SDS,
and 1% goat serum. The blots were then incubated in the Aurora kit assay buffer and
chemiluminescent substrate solution containing opti-membraneTM substrate, according to
the manufactures suggestions (AuroraTM Western blot chemiluminescent detection

system, ICN). Blots were visualized by exposure to Biomax ML ﬁlm (Kodak).

177

5.4. RESULTS

EST LjN48 represents a late nodulin gene

A range of novel expressed sequence tags (ESTs) associated with late
developmental events during nodule organogenesis in the- legume L. japonicus has been
identiﬁed, including LjN48 [17, 39].

In order to determinate the expression pattern and the length of the mRNA
species corresponding to the LjN48 EST, developmental northern blot analysis was
performed, using total RNA isolated from L. japonicus roots and nodules harvested at
different time points after infection with Rhizobium (Fig. 5.1). A mRNA species of
approximately 2100 nt was detected in L. japonicus root segments 11 days after
infection, but not in uninfected control roots. The corresponding mRNA was found to
accumulate gradually during nodule development reaching its highest level 21-days after
infection (Fig. 5.1). Thus, the pattern of gene expression observed closely resembles that
of leghemoglobin mRNA and therefore is consistent with our previous classiﬁcation of
EST LjN48 as a nodulin gene associated with late developmental events in nodule

organogenesis [3 9].

Isolation and characterization of a full copy cDNA corresponding to EST LjN48
A L. japonicus nodule speciﬁc cDNA library was screened using radiolabelled

LjN48 EST DNA as a probe. Of 26 independent cDNA clones identiﬁed, the 3’ ends of

11 clones were partially sequenced, and shown to have an identical nucleotide structure

178

(data not shown). The entire DNA sequence of the longest cDNA clone isolated (clone
48-23) was determined. The 1903 bp cDNA sequence obtained was found to contain a
1299 nt open reading frame (frame +3) starting with an ATG codon at nucleotide position
306 (ATG306; relative to the 5’ end of the cDNA), preceeded by several in ﬂame stop
codons. However, a careful analysis of 5’ sequences, upstream of the ATG306 codon,
revealed the presence of two additional short open reading frames. The relative location of
all three ORFs found in the 48-23 cDNA is shown schematically in Figure 2. The ﬁrst
upstream ORF (uORFl; frame +1) was found to be 264 bp in length, starting at the ﬁrst
ATG codon (nucleotide position 46; ATG46) and terminating at a TGA stop codon at
position 310 (Fig. 5.2). The ATG46 codon was found to be preceded by two in ﬁame
stop codons. The second short upstream ORF (uORF 2; ﬂame +3), was found to initiate
at ATG codon at position 123 and to terminate at a TAG codon at position 171.

The amino-acid sequence deduced from the nucleotide sequence of all three ORFs
did not reveal signiﬁcant similarity to sequences in the databases. However, the unusual
architecture of the 48-23 cDNA prompted us to carry out a more detailed molecular

analysis of this gene.

The 48-23 cDNA appears to correspond to a product of alternative splicing.

In order to examine the possibility that the observed structure of the 48-23 cDNA
sequence was a result of a cloning artifact, several experiments were performed. The DNA

sequence of the region containing the overlapping ORFs was determined multiple times

179

on both strands, minimizing the possibility of sequencing error(s) (see below). Moreover,
a 246 bp EcoRI-Hindlll 5’-fragment of the 48-23 cDNA, containing the uORFl, was
used as DNA probe in northern analyses of RNA isolated ﬁ'om 21 day old uninfected
roots and nodules of L. japonicus. The probe speciﬁcally detected a message of 2100 nt
present in nodules but not in uninfected roots, suggesting that this ﬁagment constitutes a
part of the mRNA species recognized by the ﬁlll cDNA clone 48-23 (data not shown).
This conclusion was further conﬁrmed and extended by amplifying L. japonicus mRNA
via reverse transcription (RT)-PCR using a pair of primers (p1; p2) derived ﬁom
sequences ﬂanking the 5’-region of the cDNA (see Fig. 5.2). If the 48-23 cDNA indeed
corresponded to a mRNA species present in L. japonicus nodules, the corresponding RT-
PCR product would be expected to be 587 bp in length. In order to test this hypothesis,
RT-PCR reactions were carried out using total RNA from uninfected and infected roots,
as well as nodules harvested at different time-points aﬁer infection. Control ampliﬁcation
reactions using the same RNA samples, but omitting the reverse transcriptase, were
carried out in parallel. After 35 PCR cycles two DNA products, 5 87 and 753 nt in length,
were detected from RNA of 11 day old infected roots and 21 day old nodules (data not
shown).

In order to estimate the relative abundance of the two distinct RT-PCR products
in nodules (see below), the ampliﬁcation procedure was repeated using a reduced number
of PCR cycles. After 20 PCR cycles two DNA products, 587 and 753 nt in length, were
detected from 21 day-old nodule RNA, whereas a single larger product, 753 bp in length,

was detected with RNA sample derived from root segments, harvested 11 days after

180

infection. Shorter and longer RT-PCR bands were found to hybridize speciﬁcally with the
radiolabelled 48-23 cDNA probe (Fig. 5.3). RT-PCR products derived ﬁ'om 21 day-old
nodules were gel puriﬁed, cloned and the DNA sequence of several independently isolated
inserts was determined. The nucleotide sequence of the shorter, 587 bp long, RT-PCR
product was found to correspond to the 5’ region of the 48-23 cDNA clone. The longer
RT-PCR product (753 bp) contained the same 587 bp long DNA sequence as the shorter
product, but carried a 166 bp insert at the boundary of the short overlap between the
uORFl and dORF. Using the same set of speciﬁc primers ﬂanking the 5’-region of the
cDNA (see above), only the 753 bp long PCR product could be ampliﬁed from L.
japonicus genomic DNA (data not shown). The nucleotide sequence of the genomic
fragment was found to be indistinguishable from that of the 753 bp RT-PCR product, and
to contain the same 166 bp insert (data not shown).

The relative abundance of the two distinct RT-PCR products was estimated using
Phospholmager (Molecular Dynamics, Boulder, CO) analysis. A 5:1 ratio of the longer,
insert containing product versus the shorter, insert lacking RT-PCR product, was found
in RNA samples ﬁ'om 21-old L. japonicus nodules. This result suggests that the 753 bp
long RT-PCR product corresponds to the major form of LjN48 RNA species in L.
japonicus nodules. In spite of the higher abundance of the insert-containing mRNA, we
were unable to isolate its corresponding cDNA clone from the L. japonicus library.
Therefore, the presence of an insert containing mRNA corresponding to the LjN48 of L.

japonicus nodules was further investigated by determining the sequence of the 2069 bp

181

long RT-PCR product ﬁom nodule RNA and found to be identical to the cDNA clone
48-23, including the 166 bp insert (see Fig. 5.5A).

DNA sequence analysis of several independently isolated PCR fragments derived
from both RT-PCR reactions and genomic DNA, revealed that in addition to PCR
products containing the exact 48-23 cDNA sequence, at least one more DNA fragment of
identical length, but containing several nucleotide substitutions was present (Fig. 5.4).
Since such divergent DNA sequence could be generated using both mRNA and genomic
DNA as templates, it is unlikely that it is the result of base pair changes introduced by
the PCR. Therefore, we conclude, that the two distinct classes of PCR products observed
correspond to mRNAs species encoded by different members of a closely related gene
family present in L. japonicus genome. This conclusion was further substantiated by

carrying out Southern blotting analysis (see below).

The 166 bp insert appears to be a plant gene intron

A detailed sequence analysis of the 166 bp intervening sequence revealed that it
displayed several features characteristic of plant introns. The size of the putative intron is
relatively small, which is characteristic of most of the higher plant introns [34]. Virtually
all naturally occuring plant introns have invariant dinucleotide 5’ /GU, and 3’AG/ splice
sites [34], and so does the intervening sequence of the RT- and genomic PCR products of
the LjN48 gene. In addition, the extended 5’- (AG/GUAAGU) and 3’-(UGYAG/GU)
splice sites consensus motifs in higher plants match the sequences at the 5’- and 3’-ends

of the 166 bp insert (Fig. 5.4).

182

The AU-richness, or U-richness, of sequence constitute additional characteristics
of higher plant introns [34]. However, in the case of the 166 bp sequence, the AU content
was found to be 54.8%, and not to be signiﬁcantly different from that of the ﬂanking 5’-,
and 3’- “exon” sequences. The U-richness of the insert sequence (36.7%) was found to be
only slightly higher then that of the ﬂanking regions (30% and 33%, respectively), and the
characteristic 3’ splice site-proximal polypyrimidine tract was not found. Nevertheless,
all the shorter RT-PCR products sequenced revealed the absence of the 166 nt long insert
and to correspond to the shorter form of LjN48 mRNA. Based on these result, we
conclude that the 166 bp long insert is likely to represent an intervening sequence

(intron).

The presumptive unspliced 48-23 mRNA encodes a single polypeptide

The presence of the intron-like sequence in the 48-23 cDNA results in fusion of
the uORF 1 and dORF, to form a single 1725 nucleotide long ORF (Fig. 5.5A). The latter
initiates at the ATG46 codon of uORFl, proceeds through the intron and the dORF and
generates a single 575 amino-acid long putative polypeptide, with a calculated molecular
mass of approximately 70 kD (Fig. 5.5A). PredictProteins software analysis of the
amino-acid sequence of this putative polypeptide revealed a polytopic hydrophobic

protein with 12 putative transmembrane (it-helices (Fig. 5.5A&B). In addition, a central

hydrophilic region of 63 amino-acid residues was detected, located between the predicted

transmembrane domains 7 and 8 (Fig. 5.5B). Homology searches using the BLAST

183

algorithm [2] revealed similarity of the putative 70 kD L. japonicus protein, LjN70, to the
protein product deduced from the Arabidopsis EST 126K15 (accession T44566), and
oxalate/formate exchange protein of Oxalobacter formigenes [1]. The latter protein
represents the vectorial component of a proton-motive metabolic cycle and has been
shown to catalyse the oxalate self-exchange reaction and the electrogenic exchange of
oxalate and formate [1]. A pairwise comparision of the protein sequences using the
BestFit algorithm of the GCG software (version 8.1) revealed that the LjN70 protein
shares 51.5% similarity and 22.2% identity with oxalate/formate exchange protein of
Oxalobacter formigenes, and 58.9% similarity and 33.7% identity with Arabidopsis
protein (data not shown). Figure 6 shows a multiple alignment between these three
proteins.

Based on these observations, we conclude that the intron containing L. japonicus
nodule associated mRNA encodes a 70 kD protein with structural features characteristic
of transport proteins [14, 21]. The corresponding L. japonicus gene was named LjNOD 70

accordingly.

Tissue-specific expression of the LjNOD 70 gene

When total RNA, isolated from different L. japonicus tissues, was probed with a
radiolabelled 48-23 cDNA fragment, hybridization signals could only be detected only in
nodules, but not in the other tissues tested (Fig. 5.7). This result indicates that the L.
japonicus LjNOD70 gene is expressed in a nodule-speciﬁc fashion, hence implicating its

putative protein product(s) in symbiosis speciﬁc processes. In contrast, northern blot

184

hybridization of RNA from different A. thaliana tissues, using EST 126K15 as a probe,
revealed that the approximately 2200 nt mRNA species was present at similar levels in all

tissues tested (data not shown).

LjNOD 70 is a member of a closely related gene family

In order to determine the copy number of the LjNOD70 gene in L. japonicus a
Southern blot analysis was performed. Genomic DNA from two L. japonicus ecotypes,
B-129 Gifu and B-581 Funakura, representing polymorphic partners in the L. japonicus
genome mapping effort [15], was digested using different restriction enzymes and
hybridized under stringent conditions with radiolabelled PCR-48 cDNA [30]. A number
of hybridizing restriction fragments could be detected for each restriction digest tested,
which is indicative of a family of related L. japonicus genes (Fig. 5.8). Interestingly, in the
case of the Funakura plants, a lower number of hybridizing bands was detected as
compared to the ecotype Gifu, suggesting that fewer members of the LjNOD70 related

gene family may be present in this L. japonicus ecotype.

LjN70 is a peribacteroid membrane protein

Since LjN70 appears to be an integral membrane protein with a potential
transporter function (see above), it was quite interesting to test whether the LjN70
protein resides in the peribacteroid membrane (PBM). The protein extract from the
PBM-containing fraction isolated from L. japonicus nodules was subjected to a western

blot analysis with the anti-LjN70 antibodies (see Materials and Methods). As a result,

185

two cross-reacting bands migrating at 64 kD and 27 kD could be detected in this ﬁ'action
(Fig. 5.9A). The upper 64 kD band most likely corresponds to LjN70, while the identity
of the lower band is less clear (see Discussion). In a parallel experiment, the protein
extract from the same PBM-containing ﬁ'action was probed with antibodies against
LIMP2 protein, the L. japonicus homolog of a well-characterized PBM protein, nodulin
26 [45-47]. As expected, LIMP2 could also be detected in the same fraction (Fig. 5.9B).
These results strongly argue that LjN70 is localized in the PBM compartment of the

infected cells of L. japonicus nodules.

186

5.5. DISCUSSION

We have identiﬁed and characterized a novel late nodulin gene (LjNOD 70) from
the model legume L. japonicus. The pattern of LjNOD70 expression was found to be
similar to that of the L. japonicus leghemoglobin gene [39], suggesting that it is likely to
be associated with late stages of nodule development or nodule ﬁmctioning.

The LjNOD 70 gene was found to be a member of a small gene family. Sequencing
of a number of PCR products, generated by RT-PCR or ampliﬁcation of genomic
sequences revealed that the members of this gene family are closely related. The RT-PCR
experiments also revealed the presence of at least two LjNOD70 mRNA species in L.
japonicus nodules. The nucleotide sequence of the ﬁrst mRNA species was identical to
the 5’ region of the 48-23 cDNA, and did not contain a 166 bp insert sequence. The
second, more abundant mRNA species, contained a 166 bp long insert localized in the 5’
region of the mRNA. Since genomic sequences corresponding to the “spliced” mRNA
species was not found, we conclude that the shorter transcript is created by removal of an
166 bp intron-like sequence from the longer mRNA species.

The presence of the intron results in the formation of 3 mRNA species containing
a single long open reading frame (ORF), encoding a 70 kD polypeptide (LjN70). The
amino acid sequence of the LjN70 protein revealed structural features typical of transport
proteins, including the presence of 12 putative transmembrane domains, and a hydrophilic
region located between transmembrane domains 7 and 8. In addition, the LjN70 protein

shares similarity with the oxalate/formate exchange protein of Oxalobacterformigenes [I].

187

High concentrations of oxalate (about 80% of the organic acid content) have been detected
in the organic ﬁaction of broad bean (Vicia faba L.) nodule cytosol and a role for oxalate
as carbon substrate for N2 ﬁxation has been postulated [40]. In general, the dicarboxylic
organic acids, primarily malate and succinate, have been long known to be the source of
plant carbon for the nitrogen ﬁxing bacteroids [reviewed in 16; also see Chapter 1 of this
thesis). The transporter activity for these organic acids have been detected in the PBM,
but the corresponding gene(s) have not yet been cloned [42]. In this respect, the
localization of the LjN70 protein to the PBM compartment suggests that this protein
may be involved in the transport of the organic acids into the nitrogen ﬁxing bacteroids.
The functional signiﬁcance of the postulated alternative splicing of the LjNOD70
mRNA remains unclear. The presence of both short and long forms of the corresponding
mRNA suggests that a speciﬁc splicing mechanism may be involved in the regulation of
LjNOD70 gene expression in L. japonicus nodules. In fact, alternative splicing has been
shown to be involved in the alteration of the function of protein kinases in animal cells
[33], and a similar mechanism has recently been suggested for the control of Arabidopsis
ANPl protein kinase activity [26]. On the other hand, the LjN70 protein appears to be
synthesized from the intron containing template. Since the presence of the intron appears
to unite rather then split the open reading frame of the LjNOD70 gene, it is possible that
the coding region of the LjNOD70 gene has evolved due to an exon-intron shufﬂing event
[9]. The 166 bp intron sequence could have been adopted to function as part of the coding
region of the LjNODI6 gene. The limited level of splicing of the intron sequence observed

(20%), would therefore reﬂect the past, rather than present, function of the 166 bp

188

intron-like sequence. It is also possible that the low efﬁciency of splicing of the 166 bp
sequence observed may be due to its overall low AT content (54.8%) [10, 34]. However,
we cannot rule out the possibility that the spliced form of LjNOD70 mRNA encodes a
functional subdomain of the LjN70 protein, and that the presence of the uORFs is
involved in co-translational regulation of expression of the downstream open reading
frame (dORF) as observed in other cases [6, 20]. It is interesting, in this respect, that the
western blot with the anti-LjN70 antibodies could in fact detect a second protein species
of a lower molecular weight. The latter could potentially be a product of translation of the
spliced form of LjNOD 70 mRNA initiating at an internal AUG codon.

In any event, the identiﬁcation of a novel, late nodulin gene sharing similarities
with transporter proteins, provides us with another useful tool and molecular marker to
elucidate the molecular basis of late nodulation events in this emerging model legume

species.

189

Figure 5.1. RNA gel blot

days after infection ana'ysis 0f L- faponicus
LjN48 EST. Ten micrograms
7 ll 13 17 21

. of total RNA isolated from

control

uninfected roots (control),

 

root segments or nodules
harvested 7, ll, 13, 17, and 21 days after infection were analysed, using the radiolabeled

LjN48 sequence as a probe.

Lotus japonicus 48-23 cDNA (1913 bp)
1—— l l dORF(1299 bp) | 1

 

 

 

 

 

 

  
    

 

 

 

 

 

 

 

 

 

l
I \
I \ \ \
I AUG \ \
1 AUG AUG +148
| +46 +130 \
+133 UG AUG \ \
I 1541:1713 ‘1
frame +1 I ’ ‘ .. - . .. ..
uORF2 '
frame +3
p] +123 +171 p2

 

Figure 5.2. A schematic structure of L. japonicus cDNA 48-23: uORFl and uORF2
denote the upstream open reading frames, whereas dORF refers to the long, continuous

downstream open reading frame. Putative AUG initiation codons for the three ORFs are

190

designated in bold. A cluster of AUG codons in the uORF l is also shown. Positions of

PC R-primers, pl and p2, are indicated.

W
1.2. .3 451. .2 3 4.5.

 

Figure 5.3. Southern blot hybridization of the RT-PCR products. Total RNA (0.6 pg)

derived from 7 (lane 1) and 21 (lane 2) days old uninfected roots, root segments or
nodules harvested 7 (lane 3), 11 (lane 4) and 21 (lane 5) days after infection was subjected
to RT-PCR (see Material and Methods). PCR products of 20 ampliﬁcation cycles were

transfered onto nitrocellulose ﬁlter and hybridized with the a-32P-dATP labelled 48-23

cDNA insert.

191

1 QTTAGAAAATGATGTTTTGTGCACCACATCCAGAAAACGGTTGCAAAAACGTTAAAGGCT 60
2 --------------------------------- G---TA----T--G--TT-C -------

1 TTGCTCTCCAAGTCCTACAAGGGAGATGGTTCATGATGCTTGCATCATTTATGATCATGG 120
2 -------------------------------------- T ---------------------

1 CAGTATCAGGAGCTAGCTACATGTTCAGCCTCTACTCCAGAGAAATCAAGTTCGCTCTCG 180
2 ----------------------------------------------------- T ------

1 GGTATGACCAATCCACTCTCAATTTGTTAAGCTTCTTCAAGGACGTGGGTTCCAACATAG 240
2 -------------------------------------------------- G--T ------

l GCATTCTCTCTGGTCTAATCAATGAthaacgcctccttgggttgtgttatcaatgggtg 300
2 --G ---------------------------------------------------------

l ctgttctaaatttttttgggtttttcatgatttggcttgctgtggccaaaaaaattgcaa 360
2 ---------------------------------------- g ----- a -------- a----

l atccccgcgtgtggcacctgtgcttgtacatcgttattggaagcaattctcattgtttta 420
2 ----------------- t ------------------------------------------

1 ccaacactgcagTTATGGTGACCAGCGTAAAGAATTTCCCGGGCATCCGCGGCATTGTAT 480
1 TAGGCATTTTGGGTGGGTATCTTAGTTTGAGTGCAATAATCATCACTCAACTGTACTATG 540

2 -------------------- G-A --------------------------- T ---------

1 CCTTCTTTATAAATGACTCCCAATCTATGATTTTGATCATGGCATGTCTACCAACAGCCA 600
2 -------------------- A ---------------------------------------

1 CTGCTTTGATTCTCCTACCAGTTATAAAGAACCACAAGAGTATTCAACAGAAAAATGACT 660

1 CCAAAGTTTTCTATAGGTTCATCTACTTGGTACTAGCTCTTGCAGGGTTCCTCATGATTA 720

1 TGATCATTTTGCAAATATCTTTCAAQTTCACCQ 753

Figure 5.4. Nucleotide sequence of genomic- and RT-PCR derived LjN48 products. The
sequences labelled l and 2 correspond to ampliﬁed product derived from L. japonicus
genomic DNA and total RNA from nodules. The presumptive intron sequences are
indicated by lower case letters; dashed lines indicate the identical sequences; The DNA
sequences of the primers used for the ampliﬁcation are underlined, whereas the more

extended 5’- and 3’- splice sites are shown in bold.

192

Figure 5.5. (A) Nucleotide and deduced amino-acid sequences of the 48-23 cDNA
containing the 166 bp insert sequence. The insert sequence is indicated by lower
case letters. The 12 transmembrane domains predicted are underlined, and the

asterisk denotes the predicted stop codon.

193

GT GTC TTCTTTATTAACTTGCTAGCTATATCTTGAAACTTAGAAAATGATGTTTTGTGCACCACA
M M F C A P H
TC CAGAAAAC GGTTGCAAAAACGTTAAAGGCTTTGCTCTCCAAGTCCTACAAGGGAGATGGTTCA
P E: DJ (3 C K N V K G F A L Q V L Q G R W E M
'I‘ GATGC TTGC AT CATTTATGATCATGGCAGTATCAGGAGCTAGCTACATGTTCAGCCTCTACTCC
Li L. IA S E‘ M, I, M, A V S G A S Y AM P S L Y S
AGAGAAATCAAGTTCGCTCTCGGGTATGACCAATCCACTCTCAATTTGTTAAGCTTCTTCAAGGA
I? E: I I< F A L G Y D Q S T L N L L S F F K D
C GTGGGTTCCAACATAGGCATTCTCTCTGGTCTAATCAATGAGg taacgcctcc ttgggt tgtgt
\7 (3 S DJ I G I L S G L I N E V T P P w V V L
t a tcaa tggg t: gctgttctaaatttt tttgggtttttcatgatttggcttgctgtggccaaaaaa
S Li <3 .A V’ L N F F G F F M I w L A V A K K
a t tg caaa t: ccccgcgtgtqgcacctgtgcttgtacatcgttattggaagcaattctcattqttt
1 P1 14 P R V w H L C L Y I, V .17 G Si NLLS H C F
t a c c a a c a c t; g c agTTATGGTGACCAGC GTAAAGAATTTCCCGGGCATCCGCGGCATTGTATTAG
T? hi 1‘ P1 V M V T S V K N F P G I R G I V L G
GCATTTTGGG’I‘GGGTATCTTAGTTTGAGTGCAATAATCATCACTCAACTGTAC TATGCCTTCTTT
1. LL G <3 Y L S L S .A, I I I T O L Y Y A F P
ATAAATGACTCCCAATCTATGATTTTGATCATGGCATGTCTACCAACAGCCACTGCTTTGATTCT
I 'N 'D S Q 8 MA I L I M, AL C L P T A T A L l L
CCTACCAGTTATAAAGAACCACAAGAGTATTCAACAGAAAAATGACTCCAAAGTTTTCTATAGGT
1. E \I I K N H K S I Q Q K N D S K V F Y R F
TCATCTACTTGGTACTAGCTCTTGCAGGGTTCCTCATGATTATGATCATTTTGCAAATATCTTTC
I: Y I, V L A L A G F L M I M I I L O I S F
AACTTCACCCAGAGTGAGTATTATGCCACTACCACTGTGATGCTTCTCTTACTTACCCTTCCACT
N P‘ '3 Q S E Y Y A T, T T V M L L L L T L P L
TTKTTGTTTIFPATTGTGGAAGACTGCAAGATTTGGAAGAGCAAACAGGAACTTATCAATTGTGAAA
A V' \1 41 V E D C K I W K S K Q E L I N C E N
PHKJTTCCGAGACCCGTGGATACAACAACAAAATCTAATGAATTGAAGTCTGAACAAACTATCCCT
P P R P V D T T T K S N E L K S E Q T I P
GAAGGACTTTCCTGTTGGCAAAATATTTTGAGGCACCCGGAAAGAGGTGAAGATCATACAGTACT
E G L S C w Q N I L R H P E R G E D H T V L
TCAAGCCATTTTTAGCCTTGATATGGTTATTCTTTTCTTTGCTACTGTATGTGGATTTGGTAGCA
Q A I E, S L D M V I L F F A, T V CL G FLLG S N
ACCTGACTGTGTACAATAACTTGAGTCAGATTGGGAAGTCCTTAGGATATCCTTCATATACTATA
L T V Y N N L S Q I G K S L G Y P S Y T I
ACCACATTTGTCTCCCTTATGTCCATTTGGATTTTTCTCGGTAAAATTGCACAAGGGGTGCTCTC
T T F V S__L M S I w I F L G K I A, Q G V L S
AGAATTTATGATAACAAAACTGAAACTCCCTAGGCCTCTCATGTTCACGATAGTCCATGTATTGT
E F M I T K L K L P R P L M P T I, V H .V L S
CTTGTATTGGTCACCTTTTAATTGCTTTCAATGTTCCAAATGGTCTATACGCAGCCTCAATTTTT
C I G H L L I A F N V P N G L Y A A S I F
ATTGGGTTTTGCTTGGGAGCAAGCTGGCCAATAATTAATTCCCTCATATCTGAATTGTTTGGTCT
I G F I: L G A S w P I I N S L I S B L F G L
TAAACATTACTCAACATTGTACAATGTTGGGACAGTAGCAAGTCCAATTGGGTCATACTTACTCA
K 11 Y S T L Y N V G T V A S P I G S Y L L N
ATGTGANXHCGCAGGGTATCTGTATGACAGGGAAGCTAGAAGGCAAATGGCAGCATTGGGGCTA
V K \I A G Y L Y D R E A R R Q M A A L G L
CAAAGAMKKTAGGAGAGGAATTGAACTGCAATGGGAGTGACTGCTACAAGTTGGCTTATATCAT
Q R It P G E E L N C N G S D C Y K L A Y I I
AATCACMKAGTTTGCTTATTTGGGGCGCTTGTGTCATTCATTTTGGTGCTTAGGACCAGACAGT
I T‘ A V C L F G A L V S F I L V L R T R Q P
TCTACMUUﬂTGACATATACAAGAAATTCACAGAAGAACCTAGGACTGCTGAAACTAAAATGGTT
Y K 'P D I Y K K F T E E P R T A E T K M V
ATGXHGGNUAGATTGAACAACTTACTGCAACTGAAATGGTTATCTCTTACACTCACTGCAGAT
I P G l( D *
(HANXHTNMGGGGAAACTGTGTTGTCCTTTACTAGTATAGTTTACTCGTTAATGTGTTGCGTA
TAAMGANHATTCATACCTCTTCTACTGCTACAATAGTGAGAGATAACCTCACGACACCAGGCA
TCMHCCNHWGGTACGACTCTGATTACGCGTGAAGATGAGACTTACATTTAAGGGGAGATTATT
GMWCMTAMHGTAATAAAAAATTAAAATAAATAAACATGTTCATATTGGATG

194

65

130

195

260

325

390

455

520

585

650

715

780

845

910

975

1040

1105

1170

1235

1300

1365

1430

1495

1560

1625

1690

1755

1820

1885

1950

2015
2069

 

 

 

 

2‘.

"U 3.1

.E

3- H hilic
5 o P

a:

a O
E thobic
w: -342

>) - _— —_ —— — _— —

z 1 2 3 4 5 6 7 s 9 1011 12

 

I I I I I I I I j I I

50 100 150 200 250 300 350 400 450 500 550

Amino acid number

Figure 5.5. (B) Hydropathy proﬁle of nodulin LjN70 derived using the method of Kyte

and Doolittle [19].

195

 

 

LjN70 1 :MPCAPHPBNEICR VKGPALQVLQGRW nu AEAS 44

At126k15 1 IARTTRBRVK ---------- ::w AEIIG 34

OxlT 1 NNPQTGQSTQLLG ---------- L VLL CNISGVQ 34

LjN70 45 5'3 L YEIR ﬂ? Aaclﬂolﬂs TIN lsﬁllvﬁ NI Is 11‘ 88

At126k15 35 LGSI P 33 N N Ro TLSEF 78
A s!

Oxl’l‘ 35 YSWTLYANPVKDNLGVSLAAVQT AéITLSQVIQ P GGYF D 78

89 umvusmt ~ II:I::I
At126k15 79 L A L s Q ”ll—2M I ”xiii; 122
OxlT LA

79 KPGPRIPLHPGGA TPKGXVDS YALYTLAGAG 122

LjN70 133 Wsacr'r :sv NPP'ﬁ MLGIL 0L1 176

At126k15 123 IGBTYP 1»ve GGA Lsov 166

OxlT 123 VYGIA u .angp RGL GPTA YG VmPPLPL 166

LjN70 177 YAFFINEISEEL IETATIBLILLJ‘ VEIRNRRGI 218

At126k15 167 'ruIRss vﬂ Mew vELIPVGGN 1:19:13 210
GI

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

OxlT 167 anVGAAPNY'rG IILIﬁ PVIR .EPGQQGAK IVVTDKD 210
LjN70 219 KﬁYRPIYLV/LAEGPEIHIILOISP Man 1*qu L 262
At126k15 211 Is TVIYAVCIL AVHLVIDPISE'é'Ioiﬁ IIIH v L I 254
OxlT 211 pNSGRNLRTprNVLwI-ArrSVNPGGLL VANSVPYGRsL A 254
LjN70 263 "I‘ AgsVEDIKIIIKS ------ KOBEINCBNPPRBVDTEKB 300
At126k15 255 L- A'rs FTASTDPCDTLBBP LGDQQGQD GQs 298
03:” 255 Gvﬂ'r 9&1gunmanpnepvsnncayxmsvvpcxuvv 298
LjN70 301 --'--- We MBGLSCWQNIL ----------------- 322
At126k15 299 RC? IFSE HKEVDLLPAVE RimQLQAKL:Q.AAAEG 342
OxlT 299 LALFPTIAALGVAPIAMLAIAPPTWGGS LFPSTNSDIPGTA 342
LjN70 323 --------- EIBIRGBDEE]: :sﬁ NVI HATVC 357
At126k15 343 AVRVRRRRG ~PWLI SLLLE 386
OxlT 343 YSARNYGPPWAAKATASIPGGGL .AIAT GWNTAPLIITAI 386
LjN70 358 Iﬁsg QGLGIIPGYIE: TFVS m1 II’L ﬂh EMVLPH: 401
A1126k15 387 ID G ﬁg DN'ra lepﬂ YPLIV 428
OxIT 387 PIAPAL FVIPRMGRPE VR SPBBKAVH 418

 

 

 

 

LjN70 402 TKLKLPR PLHPTIVHVL8::.LLIﬁl‘NxﬁNGLYAABIPEPCL 445
At126k15 429 RDYAY VAIAVAQLvus 11’? 26 GANRIGILL LGY 472
LjN70 446 ﬁsﬂﬂINsLIGRLRGL Hrsmv GﬁzﬁwﬁmYLLNVKVﬁG 489
At126k15 473 a VPATA RIG LVPSGLI 516
m m "*BLBL°“:E]°‘:E"II"°EI‘E)":EJWEB
At126k15 517 81 B-o-- Q SLP DD R R I 1? Is 1186? 558

LjN70 534 PGALVBFILVLRTRQFYKTDIYKKFTEEPRTAETKMVIPGKD 575
At126k15 559 I 559

 

 

Figure 5.6. Alignment of the LjN70 protein sequence with the deduced

partial amino-acid sequence of Arabidopsis EST 126K15 (accession. No.

 

T44566), and the oxalate/formate exchange protein of Oxalobacter

formigenes [l]. Identical residues are shaded and boxed; conservative

substitutions are shaded.

196

Flower
Leaf
Stem
Root
Nodule

-- 28$
- 18S

“ -188

 

   

Figure 5.7. Expression of the LjNOD70 gene in different L. japonicus tissues. Ten
micrograms of total RNA from L. japonicus ﬂowers, leaves, stems, uninfected roots, and
nodules was probed with radiolabeled 48-23 cDNA insert (Top Panel). Equal RNA
loading was veriﬁed by probing the same ﬁlter with an 18S DNA probe (Bottom Panel).

Positions of the 28S and 18S rRNAs are as indicated.

197

Gifu Funakura

inﬁzigia
35 a 12' S as 13‘
H)
12.2..
9.1-
8.1-

1.0 '

up

 

Figure 5.8. Southern hybridization analysis of L. japonicus ecotypes Gifu and Funakura

using the LjNOD70 gene as probe. Ten micrograms of L. japonicus genomic DNA was

198

digested with the restriction enzymes indicated, and probed with an [a-32P] dATP labeled

LjN48 EST insert.

.- 64le ‘

Figure 5.9. Western blot of the isolated peribacteroid
membrane fraction prepared from 9 weeks-old Lotus
japonicus nodules with antibodies against A. LjN70 and B.

LIMP2. Calculated molecular weights are shown.

199

5.6. ACKNOWLEDGEMENTS

We thank Dr. Dominique Tremousaygue (INRA-CNRS, Tolouse, France), for providing
us with a preliminary sequence of the Arabidopsis EST 126K15. We also thank Kurt
Stepnitz for help with preparing the ﬁgures. The research described in this manuscript
was supported by grant no. DE-FG02-91ER20021 from Department of Energy and grant

no. NSF-09630189 from the National Science Foundation.

200

5.7. REFERENCES

10.

ll.

. Abe K, Ruan Z-S, Maloney PC: Cloning, sequencing, and expression in Escherichia

coli of OxlT, the oxalatezformate exchange protein of Oxalobacterformigenes. J Biol
Chem 27: 6789-6793 (1996).

. Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ: Basic local alignment search

tool. J Mol Biol 215: 403-410 (1990).

. Carlson RW, Price NPJ, Stacey G: The biosynthesis of rhizobial lipo-oligosaccharide

nodulation signal molecules. Mol Plant-Microbe Interact 7: 684-695 (1995).

. Csanadi G, Szecsi J, Kalo P, Kiss P, Endre G, Kondorosi A, Kondorosi E, Kiss GB:

ENOD12, an early nodulin gene, is not required for nodule formation and efﬁcient
nitrogen ﬁxation in alfalfa. Plant Cell 6: 201-213 (1995).

. de Bruijn FJ, Schell J: Regulation of plant genes speciﬁcally induced in developing and

mature nitrogen-ﬁxing nodules: cis-acting elements and trans-acting factors. In Verma
DPS, (ed), Control of Plant Gene Expression, pp. 241-258. CRC Press, Boca Raton
(1992)

. Damiani RD, Wessler SR: An upstream open reading frame represses expression of

Lc, a member of the R/B family of maize transcriptional activators. Proc Natl Acad Sci
USA 90: 8244-8248 (1993).

. Delauney AJ, Verma DPS: Cloned nodulin genes for symbiotic nitrogen ﬁxation. Plant

Mol Biol Rep 6: 279-285 (1988).

. Gamas P, Niebel F, de C, Lescure N, and Cullimore J: Use of a substractive

hybridization approach to identify new Medicago truncatula genes induced during
root nodule development. Mol Plant Microbe Interact 9: 233-242 (1996).

. Gilbert W: Why genes in pieces? Nature 271: 501 (1978).

Goodall GJ, F ilipowicz W: The AU-rich sequences present in the introns of plant
nuclear pre-mRNAs are required for splicing. Cell 58: 473-483 (1989).

Harlow E, Lane D: Antibodies. A laboratory manual. Cold Spring Harbor Laboratory
(1988)

201

12.

13.

Heard J, Caspi M, Dunn K: Evolutionary diversity of symbiotically induced nodule
MADS box genes: characterization of nth5, 3 member of a novel subfamily. Mol
Plant-Microbe Inter 5: 665-676 (1997).

Heard J, Dunn K: Symbiotic induction of a MADS-box gene during development of
alfalfa root nodules. Proc Natl Acad Sci USA 92: 5273-5277 (1995).

14 Henderson PJF: The 12-transmembrane helix transporters. Curr Opin Cell Biol 5: 708-

15.

16.

17.

18.

19.

20.

21.

22.

23.

721(1993)

Jiang Q, Gresshoff PM: Classical and molecular genetics of the model legume Lotus
japonicus. Mo] Plant-Microbe Interact 1: 59-68 (1997).

Kahn ML, McDermott TR, Udvardi MK: Carbon and nitrogen metabolism in
Rhizobia. In Spaink HP, Kondorosi A, Hooykaas PJJ (eds), The Rhizobiaceae.
Kluwer Academic Publishers, Dordrecht, pp. 461-485(1998)

Kapranov P, de Bruijn FJ, Szczyglowski K: A novel, highly expressed late nodulin
gene LjNOD16 from Lotus japonicus. Plant Physiology 1 13:1081-1090 (1997).

Kouchi H, Hata S: Isolation and characterization of novel cDNA representing genes
expressed at early stages of soybean nodule development. Mol Gen Genet: 238: 106-
119(1993)

Kyte J, Doolittle RF: A simple method for displaying the hydropathic character of a
protein. J Mol Biol 157: 105-132 (1982).

Lohmer S, Maddaloni M, Motto M, Salamini F, Thompson RD: Translation of the
mRNA of the maize transcriptional activator opaque-2 is inhibited by upstream open
reading frames present in the leader sequence. Plant Cell 5: 65-73 (1993).

Marger MD, Saier Jr MH: A major superfamily of transmembrane facilitators that
catalyse uniport, symport and antiport. TIBS 18: 13-20 (1993).

Matvienko M, Van De Sande K, Yang W-C, Van Kammen A, Bisseling T, Franssen
H: Comparison of soybean and pea ENOD40 cDNA clones representing genes
expressed during both early and late stages of nodule development. Plant Mol Biol 26:
487-493 (1994).

Miao G-H, Verma DPS: Soybean nodulin-26 gene encoding a channel protein is

expressed only in the infected cells of nodules and is regulated differently in roots of
homologous and heterologous plants. Plant Cell 5: 781-794 (1993).

202

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

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

Nap J-P, Bisseling T: Developmental biology of a plant-prokaryote symbiosis: the
legume root nodule. Science 250: 948-954 (1990).

Nishihama R, Banno H, Kawahara E, Irie K, Machida Y: Possible involvement of
differential splicing in regulation of the activity of Arabidopsis ANPl that is related to
mitogen-activated protein kinase kinases (MAPKKKs). Plant] 12: 39-48 (1997).

Quispel A, Van Brussel AAN, Hooymans JJM, Priem WJE, Staal HJM:
Intersymbiotic transport processes in nitrogen-ﬁxing root nodules. In Ludden PW,
Burris JE (eds), Nitrogen Fixation and C02 Metabolism, pp. 193-202. Elsevier
Science Publishing Co., Inc (1985).

Rogers SO, Bendich AJ: Extraction of DNA from plant tissues. In Gelvin SB,
Schilperoort RA (eds), Plant Molecular Biology Manual, pp. A6/ 1-A6/ 10. Kluwer
Academic Publishers, Dordrecht (1988).

Rost B, Casadio R, Fariselli P, Sander C: Prediction of helical transmembrane
segments at 95% accuracy. Protein Science 4: 521-533 (1995).

Sambrook J, Fritsch EF, Maniatis T: Molecular Cloning. A Laboratory Manual, ED2.
Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY (1989).

Sandal NN, Marcker KA: Similarities between a soybean nodulin, Neurospora crassa
sulphate permease II and a putative human tumor suppressor. Trends Biol Sci 19: 19
(1994).

Schultze M, Kondorosi E, Ratet P, Buire M, Kondorosi A: Cell and molecular biology
of Rhizobium-plant interaction. Int Rev Cytol 156: 1-75 (1994).

Shi E, Kan M, Xu J, Wang F, Hou J, McKeehan WL: Control of ﬁbroblast grth
factor receptor kinase signal transduction by heterodimerization of combinatorial
splice variants. Mol Cell Biol 13: 3907-3918 (1993).

Simpson GG, Filipowicz W: Splicing of precursors to mRNA in higher plants:
mechanism, regulation and sub-nuclear organisation of the spliceosomal machinary.

Plant Mol Biol 32: 1-41 (1996).

Song W, Steiner HY, Zhang L, Naider F, Stacey G, Becker J: Cloning of a second
Arabidopsis peptide transport gene. Plant Physiol: 110: 171-178 (1996).

203

36. Spaink HP: Regulation of plant morphogenesis by lipo-chitin oligonucleotides. Crit
Rev Plant Sci 15: 559-582 (1996).

37. Sprent JI: Which steps are essential for the formation of functional legume nodules?
New Phytol 111:129-153 (1989).

38. Steiner HY, Song W, Zhang L, Naider F, Becker JM, Stacey G: An Arabidopsis
transporter is a member of a novel family of membrane transport proteins. Plant Cell
6: 1289-1299 (1994).

39. Szczyglowski K, Hamburger D, Kapranov P, de Bruijn FJ: Construction of a Lotus
japonicus late nodulin EST library and identiﬁcation of novel nodule-speciﬁc genes.
Plant Physiology 144: 1335-1346 (1997).

40. Trinchant J-C, Guerin V, Rigaud J: Acetylene reduction by symbiosomes and free
bacteroids from broad bean (Vicia faba L.) nodules. Plant Physiol 105: 555-561
(1994)

41. Udvardi MK, Day DA: Metabolite transport across symbiotic membranes of legume
nodules. Annu Rev Plant Physiol Plant Mol Biol 48: 493-523 (1997).

42. Udvardi MK, Price GD, Gresshoff PM, Day DA: A dicarboxylate transporter on the
peribacteroid membrane of soybean nodules. F EBS Lett 231: 36-40 (1988)

43. Vance CP, Heichel GH: Carbon in N2 ﬁxation: limitation or exqusitive adaptation.
Annu Rev Plant Physiol Plant Mol Biol 42: 373-392 (1991).

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

45. Weaver CD, Crombie B, Stacey G, Roberts DM (1991). Calcium-dependent
phosphorylation of symbiosome membrane proteins from nitrogen ﬁxing root
nodules: Evidence for the phosphorylation of nodulin-26. Plant Physiol. 95: 222-227.

46. Weaver CD, Shomer NH, Louis CF, Roberts DM: Nodulin 26, a nodule-speciﬁc
symbiosome membrane protein from soybean, is an ion channel. J Biol Chem 269:
17858-17862 (1994).

47. Weaver CD, Roberts DM (1992) Determination of the site of phosphorylation of

nodulin 26 by the calcium-dependent protein kinase from soybean nodules.
Biochemistry 31: 8954-8959.

204

CHAPTER 6

Novel, highly expressed late nodulin gene (LjNOD16) from Lotus
japonicus4

6.1. ABSTRACT

A Lotus japonicus cDNA was isolated and shown to correspond to a highly abundant,
late nodule-speciﬁc RNA species that encodes a polypeptide with a predicted MW of
15.6 kD. The protein and its corresponding gene were designated Nle6 and LjNOD16,
respectively. The LjNOD16 was found to be expressed only in the infected cells of L.
japonicus nodules. Related DNA sequences could be identiﬁed in the genomes of both
Glycine max and Medicago sativa. In the latter case, a homologous mRNA species was
detected in the nodules. Unlike LjNOD16, its alfalfa homologs appear to represent low-
abundant mRNA species. However, the proteins corresponding to the LjNODI6 and its
alfalfa homolog could be detected at similar levels in nodules, but not in roots of both
legume species. The predicted amino acid sequence analysis of nodulin Nlj16 revealed the
presence of a long a-helical region and a positively charged C-terminus. The former
domain has a very high propensity to form a coiled-coil type structure, indicating that

nodulin Nljl6 may interact with as-yet-unidentiﬁed protein target(s) in the nodule-

infected cells. Homology searches revealed no signiﬁcant similarities to any known

 

4 This chapter was published in Kapranov P, de Bruijn FJ, Szczyglowski K (1997) Plant

Physiology 113: 1081-1090.

205

sequences in the databases, with the exception of two related, anonymous Arabidopsis

ESTs

206

6.2. INTRODUCTION

Symbiotic nitrogen ﬁxation is a unique example of a complex and subtly regulated
biological process that takes place in a specialized plant organ - the nodule. Upon
infection with speciﬁc strains of symbiotic bacteria belonging to the genera Rhizobium,
Bradyrhizobium, and Azorhizobium, root cortical cells of legume plants undergo a
dedifferentiation process that eventually leads to the formation of a nodule meristem. A
highly organized and controlled series of events thereafter culminates in the formation of a
fully functional, nitrogen-ﬁxing nodule. Nodule ontogeny appears to be predominantly
controlled by a plant morphogenetic program (for a recent review see Gresshoff, 1993).
Multiple signals, derived from both the host plant and the symbiotic bacteria, specify the
induction and coordination of this organogenic process which involves the activation of
speciﬁc genes in both symbiotic partners (for a review see Schultze et al., 1994). A
collection of genes encoding plant-nodule-speciﬁc proteins, collectively referred to as
nodulin genes (van Kammen, 1984), has been isolated from various species of legumes
(for a review see Mylona et al., 1995). These genes have been traditionally classiﬁed as
early or late nodulin genes, reﬂecting the developmental time point of their expression
(Nap and Bisseling, 1990). Induction of the early nodulin genes has been correlated with
early morphogenetic processes, such as preinfection, infection, and cortical cell division

(Franssen et al., 1992; Nap and Bisseling, 1990; Cook et al., 1995). However, the

207

functions and requirement for nodulation (Csanadi et al., 1994) of none of the early
nodulin genes have been elucidated.

Signiﬁcant progress has been made in unraveling the cascade of both bacterial and
plant signals, which initiate and direct early morphogenic events. The discovery of a set
of ﬂavonoid inducers of plant origin and Rhizobium-derived lipochito-oligosaccharide
signal molecules, known as Nod factors, has led to exciting progress in this area (Horvath
et al., 1987; Lerouge et al., 1990; Spaink et al., 1987; Spaink, 1992). Although research on
the early stages of nodulation has progressed rapidly, the developmental cues and
molecular events responsible for the ﬁnal steps of nodule formation and functioning
remain largely unknown (de Bruijn and Schell, 1992). These late stages include important
events such as central nodule tissue formation, bacterial release from infection threads and
plant cell colonization, production of peribacteroid membranes (PBM), bacteroid
differentiation, and eventually, commencement of nitrogen ﬁxation (Sprent, 1989). The
latter process is accompanied by major molecular and biochemical alterations, which
create and support the physiological requirements for nitrogen ﬁxation and ammonia
assimilation (Nap and Bisseling, 1990).

In spite of the highly complex nature of the late developmental events, only a
limited number of late nodulin genes have been identiﬁed and characterized. Typical

members of this group include genes encoding enzymes involved in speciﬁc biochemical

pathways, e.g., the y subunit of Gln synthetase involved in ammonia assimilation (Lara et

al., 1983; Gebhardt et al., 1986; Bennett et al., 1989; Boron et al., 1989; Boron and

208

Legocki, 1993), uricase II (Legocki and Verma, 1979; Bergmann et al., 1983), xanthine
dehydrogenase (Triplett, 1985), sucrose synthase (Thummler and Verma, 1987),
peribacteroid membrane proteins (Fortin et al., 1985; 1987; Verma, 1992; Miao and
Verma, 1993), leghemoglobins (Brisson et al., 1982), and a number of proteins the
ﬁrnctions of which remain to be identiﬁed (Delauney and Verma, 1988). Recently, a
symbiotically induced MADS-box containing gene (nmh7) has also been identiﬁed in
alfalfa root nodules (Heard and Dunn, 1995). It has been suggested that the protein
encoded by this gene could be involved in cellular activities speciﬁc to the differentiation
of the infected cells (Heard and Dunn, 1995).

Although a number of late nodulin genes have been isolated, it appears likely that
additional genes remain to be identiﬁed. Further cloning and detailed characterization of
genes coding for novel late nodulins is crucial for understanding the molecular and
biochemical details of late nodule morphogenic events and nodule ﬁmctioning. Therefore,
we have initiated a systematic search for novel late nodule-speciﬁc transcripts in the
model legume Lotus japonicus (Handberg and Stougaard, 1992; Jiang and Gresshoff,
1993), using the RNA differential display technology described by Liang and Pardee,
(1992; and also Szczyglowski et al., 1997). We report here the isolation and molecular

characterization of LjNOD16, a novel late nodulin gene from L. japonicus.

209

6.3. MATERIALS AND METHODS

Plant Material

Lotus japonicus GIFU B-129-S9 seeds were kindly provided by Dr. Jens
Stougaard, Aarhus University , Denmark. Seeds were surface sterilized and germinated as
described by Handberg and Stougaard (1992). One-week-old L. japonicus seedlings were
inoculated with Rhizobium loti strain NZP2235 (Jarvis et al., 1982) and transferred to
pots (30 plants per pot) containing a 6:1 mixture of vermiculite and sand. All plants were
grown in cabinets with a controlled environment: an 18h/6h day/night cycle, a light

intensity of 246 uE sec' m'z, a 220C/180C day/ night temperature regime, and a 70%

humidity level. Both inoculated and uninoculated control plants were watered using B&D
nutrient solution (Broughton and Dilworth, 1971) containing 0.5 mM KNO3. This low
concentration of combined nitrogen supported grth of the uninfected control plant but
did not affect nodule formation on the inoculated L. japonicus plants (data not shown).
For the initial stages (3, 7, and 11 days) root segments were harvested, whereas only
nodules were collected for the 21-day time-point.

The transgenic L. japonicus plants were inoculated with Rhizobia loti strain 2235
(Jarvis et al., 1982) before moving from tissue culture to soil and grown under the same
conditions. Transgenic plants were watered with B&D nutrient solution (Broughton and
Dilworth, 1971) supplemented with 1 mM KNO3. The nodulation and general phenotype

of transgenic plants were evaluated starting from 2-3 weeks after inoculation. Fully

210

mature nodules from each transgenic line were harvested for western blot to assess the
levels of Nlj 16 protein. Subsequently the transgenic plants were moved to a rich soil and
grown for seed production.

The Arabidopsis thaliana Landsberg plants were kindly provided by Jacqueline
Chemys from Dr. Hans Kende’s laboratory, Plant Research Laboratory, Michigan State
University. Medicago sativa Cardinal seeds were surface sterilized and germinated on 1%
agar in water. Three-day-old seedlings were infected with Rhizobium meliloti strain 1021
and transferred to soil [3 vermiculite (grade2): 3 vermiculite (grade 3): 1 sand]. M. sativa
plants were grown under the same controlled environment as L. japonicus plants and
watered with B&D solution or distilled water. M. sativa plants were grown for 21 days
before harvesting. The plant tissues were immediately frozen in liquid nitrogen and stored

at -80°C until use.

Nucleic Acid Isolation, Southern and Northern Analyses

Plant genomic DNA was isolated following the procedure described by Rogers
and Bendich (1988). Total RNA isolation was performed according to the method of
Verwoerd et al. (1989), except that the extraction buffer was as described in Hall et al.
(1978). The poly (A)+ fraction of total RNA was isolated using oligo(dT) cellulose spin

column kits (5 Prime—>3 Prime, Inc., Boulder, CO) following the manufacturer’s

instructions.

211

For Southern blot analysis, 10 pg of genomic DNA was completely digested using
the appropriate restriction enzymes, separated on 0.8% agarose gel, and transferred to
nylon ﬁlters according to standard procedures (Sambrook et al., 1989). Membranes were
prehybridized and hybridized under high stringency conditions (2xSSC) or low stringency

conditions (4xSSC), 0.5% SDS, 5X Denhard’s solution, 100 ug/ml denatured sheared

salmon sperm DNA at 65°C. Washes were carried out under low or high stringency, as
described by Sambrook et a1. (1989).

For Northern analysis, 10 ug of total RNA was separated on 1.2% agarose-
formaldehyde gels in MOPS buffer, as described by Sambrook et al. (1989) and
transferred to Nitro Plus supported nitrocellulose membranes (Fisher Scientiﬁc,
Pittsburgh, PA). Prehybridization and hybridization reactions were performed in 0.5M
phosphate buffer, pH 7.2, 7% SDS and 1% BSA at 650C, according to the procedure
described by Church and Gilbert (1984). Filters were washed twice for 15 min in 2X SSC,

0.1% SDS, and once for 15 min at 0.3X SSC, 0.1% SDS at 65°C. Probes were labeled with

[a-32P]dATP using the random prime kit from Boehringer-Mannheim (Indianapolis, IN),

following the manufacturer’s instructions.

RNA Differential Display
Differential Display was carried out using RNAmapTM kits ﬁ'om the GenHunter
Corporation (Brookline, MA), essentially following the manufacturer’s instructions.

cDNA synthesis was performed using 0.5 pg of total RNA isolated from 21-day-old L.

212

japonicus roots, or nodules harvested 21 days after rhizobial infection. Differentially
expressed bands were puriﬁed from the polyacrylamide gel and reampliﬁed using the
procedure described in the manual of the RNAmap kit. The DNA ﬁagments were
puriﬁed on a 1% agarose gel, blunt ended using the Klenow fragment of DNA polymerase

I, and cloned into the Sma I-digested vector pK18 (Pridmore, 1987).

cDNA Library Screening

The cDNA library from mature nodules of L. japonicus was kindly provided by
Dr. Jens Stougaard, Aarhus, Denmark. The library was constructed with oligo-dT primer
in the lambda UniZAP vector from Stratagene (La Jolla, CA). Screening for full-copy
cDNAs corresponding to the PCR-5 product was performed following standard

procedures (Sambrook et al., 1989; Stratagene manual).

DNA Sequencing and Computer Analyses

DNA sequencing was performed using the Sequenase 2.0 kit from the United
States Biochemical, Inc. (Cleveland, OH), according to the manufacturer’s instructions.
Computer analysis of DNA sequences was carried out using the Squd software (Applied
Biosystems, Foster City, CA). Analysis of predicted protein sequences was performed
using the GCG (Genetics Computer Group, Madison, WI, USA), PHDsec (Rost and
Sander, 1993, 1994) and COILS version 2.2 (Lupas, 1996) programs. Homology searches

were performed using BLAST software (Altschul etal., 1990).

213

In Situ Hybridization

Twenty-one-day-old nodules were ﬁxed, dehydrated, and embedded into parafﬁn

according to procedures described by van de Wiel et al. (1990). Nodule sections of 7 mm

were hybridized with digoxigenin-UTP labeled antisense and sense RNA probes, using
the conditions reported by Engler et al. (1994). The full-copy cDNA encoding nodulin
Nle6 and a cDNA representing a L. japonicus leghemoglobin gene were used to prepare
RNA probes, using the DIG-RNA labeling kit (Boehringer Mannheim, Indianapolis, IN)
following the manufacturer’s instruction. Washing and detection conditions were as
described by Engler et a1. (1994). Dry slides were mounted using polymount
(Polysciences Inc., Warrington, PA) and examined by dark- and bright-ﬁeld microscopy

using an Axiophot microscope (Zeiss, Oberkochen, Germany).

Production of Recombinant Nljl6 Protein in Escherichia coli

The 470-bp Bgl II-Hinc II fragment of the full length LjNOD16 cDNA, containing
the entire coding region of the nodulin LjNOD16 gene was blunt ended using Klenow
enzyme and cloned into the dephosphorylated pET lSB expression vector (Novagen,
Madison, WI), which had been digested with Xho I and blunt ended with Klenow
enzyme. The construct with the insert cloned in the sense orientation was identiﬁed by
digestion with Xho I and Burn HI. In frame fusion of the coding region of LjNOD16 to the
N-terminal histidine tag of pET 15B was conﬁrmed by DNA sequencing. The construct

was transformed into E. coli strain BL21 (DE3) (Novagen, Madison, WI), for expression

214

of the recombinant protein. The His-tag-Nljl6 fusion protein was puriﬁed under
denaturing conditions using His-Bind® Resin (Novagen, Madison, WI), following the
manufacturer’s instructions. The histidine tag was cleaved off by digestion with thrombin
(Novagen, Madison, WI) and the Nle6 recombinant protein was further puriﬁed by
passing it through the His-Bind® Resin column after digestion. The flow through and
wash fractions were collected. N1j16 was refolded by dializing against phosphate-buffered
saline (PBS; Sambrook et al., 1989) overnight. The refolded protein was mixed with

TiterMax® adjuvant (Cthx® Corporation, Norcross, GA) and used for immunization of

rabbits following the instructions of the TiterMax® manual.

Western Blot Analysis

The western blot described in ﬁgure 9A&B was performed as follows. The total
protein extracts from L. japonicus and M. sativa roots and nodules were obtained by
grinding tissues in liquid nitrogen and boiling for 5 min in extraction buffer (IOOmM Tris,
pH 6.8, 5% SDS, 0.5% B-mercaptoethanol). For Western blot analysis, 50 pg of total
protein was separated by SDS-PAGE (Laemmli, 1970), using 15% acrylamide gels, and
electroblotted overnight onto ProtranTM nitrocellulose membranes (Schleicher and Schuell,
Keene, NH) in Towbin buffer (Towbin et al., 1979), supplemented with 0.05% SDS.
Blocking, binding and washes were performed in PBS supplemented with 0.3% Tween 20
(Sigma, St. Louis, MO). Pre-immune and immune sera from a rabbit immunized with

Nle6 recombinant protein were used in a 1: 2000 dilution. Antibody detection was

215

performed using goat anti-rabbit antibodies conjugated to alkaline phosphatase
(Kirkegaard & Perry Laboratories, Gaithersburg, MD).

The western blots in Figures 10 and 11 were performed with the afﬁnity puriﬁed
antibody. The anti-N1j16 serum was afﬁnity-puriﬁed over a column containing
recombinant N1j16 protein conjugated to the Afﬁ-Gel 10 matrix (Bio-Rad Laboratories)
following the manufacturer's instructions. The afﬁnity-puriﬁed antibody was used

in12500 dilution.

Generation of L. japonicus transgenic plants

The LjNODI6 cDNA used for transgenic work contains the entire 423 bp Nljl6
coding region with 21 bp 5'UTR and 404 bp 3'UTR. The cDNA was isolated as EcoRI-
Xhol fragment and both sites were ﬁlled-in with Klenow DNA polymerase. The
LjNOD16 cDNA region was then cloned blunt-end into the pLP14 binary vector
(Szczyglowski et al., 1994) digested with Kpnl-Sacl, treated with T4 DNA Polymerase
to blunt both sites and dephosphorylated. The pLPl4 binary vector contains the 2 kb
promoter sequence of Sesbania rostrata leghemoglobin gene glb3 fused to the GUS
reporter gene (Szczyglowski et al., 1994). The LjNOD16 cDNA was placed under the
control of the glb3 promoter, in place of the GUS gene, in both sense (construct
pCRSSense) and anti-sense (construct pCRSAnti) orientations.

Both constructs were conjugally transferred into A grobacterium tumefaciens strain

LBA4404. The transgenic L. japonicus plants were generated using the modiﬁcation of

216

the original hypocotyl-transformation method (Handberg et al., 1994) essentially as

described by Stiller etal., 1997.

6.4. RESULTS

Identiﬁcation and cloning of the L. japonicus PCR-5 cDNA

The recent development of RNA arbitrarily primed ﬁngerprinting techniques, also
known as RNA differential display, has resulted in an experimental tool with exciting
potential for the detection of differential gene expression during complex biological
processes (Liang and Pardee, 1992; Welsh et al., 1992). We have employed this technique
to construct a library of L. japonicus late nodulin ESTs (Szczyglowski, Hamburger,
Kapranov and de Bruijn, manuscript in preparation). Our experiments involved a simple
comparison of RNA proﬁles from two samples: mature 21-day uninfected roots and fully
developed nodules harvested at 21 days after infection. Thirteen different combinations of
PCR primers were initially used. Five bands appeared to be differentially expressed on
display gels. These were excised, reampliﬁed by PCR, and analyzed using northern
blotting and DNA sequencing. Three out of ﬁve differentially displayed cDNA products
corresponded to nodule-speciﬁc mRNAs, as determined by northern blotting (data not

shown). One of the differentially expressed cDNAs, designated PCR-5, was found to be

217

607 bp long and showed no signiﬁcant homology to DNA sequences in the databases
(data not shown). We chose this cDNA for further detailed molecular characterization.
Northern blot analysis was used to correlate speciﬁc phases in nodule
development with the expression pattern of the gene corresponding to the PCR-5 cDNA.
This experiment revealed that the PCR—5 product corresponded to a highly abundant late
nodulin mRNA species. The mRNA corresponding to PCR-5 was found to be detectable
between seven and eleven days after infection and to accumulate gradually to a high level
in fully developed 21-day-old nodules (see Fig. 6.1). The observed kinetics of mRNA
accumulation resembled very closely the pattern of L. japonicus leghemoglobin gene
expression (data not shown), justifying the categorization of this gene as a late nodulin

gene.

PCR-5 cDNA corresponds to the gene encoding L. japonicus nodulin Nlj16

PCR-5 cDNA was used as a probe to screen a L. japonicus nodule-speciﬁc cDNA
library (kindly provided by Dr. Jens Stougaard, University of Aarhus, Denmark).
Approximately forty hybridizing cDNA clones were puriﬁed. DNA sequence analysis of
the six longest cDNA clones (length range 790 bp to 1010 bp) showed that they all
carried identical DNA sequences, but had different length of 3’-nontranslated regions. Six
different sites of polyA addition were identiﬁed in the cDNAs. The sequence of the
longest cDNA clone is shown in Figure 6.2A (accession number U64964). The different
polyadenylation sites, derived from the sequences of the corresponding cDNA clones, are

as indicated. Further analysis of the DNA sequence of the longest cDNA clone revealed

218

that it contained a 41-bp 5’-untranslated region, followed by an open reading frame 423
nucleotide-long, and a 546-bp 3’-untranslated region. The stop codon (TAA) was found
to be located 6 bp upstream of the putative AUG initiation triplet, indicating that the
cDNA contains the entire coding region (Fig. 6.2A). The deduced protein sequence was
found to correspond to a polypeptide of 141 amino acid residues and a molecular weight
of approximately 15.6 kD. Therefore, we propose that our cDNA represents a late L.
japonicus nodule-speciﬁc gene, designated LjNODI6, encoding a 15.6-kD nodule-speciﬁc

protein (nodulin Nlj 16).

Nodulin Nlj16 contains a long a-helical domain and is homologous to two

anonymous ESTs from Arabidopsis thaliana

A hydropathy proﬁle of nodulin N1jl6 was derived using the algorithm of Kyte
and Doolittle (1982) and LASERGENE software (DNASTAR, Madison, WI). This
analysis indicated that the protein was mostly hydrophilic, except for a hydrophobic
stretch of 16 amino acids, with tetrad repeats of proline residues at the N-terminal end
(Fig. 6.28; see also Fig. 6.2A). Secondary structure analysis using the PHDsec program
predicted with a high probability (7-9 on a scale from 0 to 10) the presence of two

adjacent a-helical regions, divided by a loop or turn (Fig. 6.2A). The relatively long or-

helical domain, spanning 80 out of 141 amino-acid residues of the Nle6 protein, would be
expected to form a coiled-coil type protein structure, as predicted by the COILS version

2.2 program (Lupas, 1996). A homology search using the BLAST algorithm revealed no

219

signiﬁcant similarities to any gene with known ﬁrnction, except for a limited homology to
two anonymous Arabidopsis ESTs: EST168K8 (accession R64923) and 110G16
(accession T42081). The corresponding A. thaliana cDNA clones were obtained from
Tom Newman’s laboratory at the MSU-DOE Plant Research Laboratory and from the
Arabidopsis Biological Resource Center (Ohio State University), respectively, and their
entire DNA sequence was determined. The complete DNA sequences for ESTs 168K8
and 110G16 have been deposited in GenBank under accession numbers U64965 and
U64966, respectively. The deduced amino acid sequences of EST 168K8 (751 bp; A.t.
EST]; Fig. 6.3) and EST 110G16 (644 bp; A.t. EST2; Fig. 6.3), which represent 200 and
150 amino acids, respectively, from the C-terminal ends of the proteins, showed a high
level of homology to each other throughout the whole amino acid sequence (Fig. 6.3). In
addition, a signiﬁcant similarity between the A. thaliana ESTs and the amino acid

sequence of the Nle6 protein, especially in its a-helical domain, was found (Fig. 6.3).
Interestingly, the region of the 168K8 EST with the highest similarity to the a-helical
domain of Nle6 also has a high probability to form two a-helices and to participate in

generating coiled-coil structures, as predicted by the PHDsec and COILS programs. A
computer search of the PROSITE and BLOCKS databases did not reveal any distinct
conserved amino acid motifs in nodulin Nlj 16 or in the protein sequences derived from the

Arabidopsis ESTS.

220

Tissue-specific expression of the LjNODI6 gene

To gain further insight into the tissue speciﬁcity of LjNOD16 gene expression,
total RNA isolated from L. japonicus flower, leaf, stem, root, and fully developed nodule
tissues was analyzed using northern blot hybridization (Fig. 6.4). The LjNOD16 gene was
found to be expressed exclusively in the nodule tissues as revealed after a brief, 7.5-h
exposure of the blot (data not shown). However, a prolonged exposure of the blot, over a
period of 7 d, revealed the presence of a very low level of hybridizing mRNA species in
uninfected roots and ﬂowers. In the latter case, an mRNA species of approximately 2500
nucleotides could be detected, as opposed to the 900- to 1200-nucleotide-long transcripts
consistently found in L. japonicus nodules. Therefore, we conclude that the LjNOD16
gene represents a nodulin gene, and that a related gene(s) may exist in the L. japonicus
genome. Northern blot analysis was also performed using total RNA isolated from
various organs of A. thaliana plants and A.t. ESTl or A.T. EST2 cDNAs were used as
probes. An example of this analysis using A.t. ESTl is shown in Figure 6.5. Both ESTs

hybridized to an approximately 3000 nucleotide-long mRNA in all tissues tested.

Organization of LjNODI6 gene in genomes of L. japonicus and other legume
plants

To determine the number of LjNOD] 6 genes in the L. japonicus genome, Southern
blot analysis was performed. Genomic DNA isolated from L. japonicus plants was

digested to completion using four different restriction enzymes and hybridized with an

221

[a-32P]dATP-labeled probe, representing the PCR-5 cDNA sequence. Under high-

stringency conditions (see Materials and Methods) 3 single hybridizing band was
observed for three out of four different restriction digests (note that a Hind 111 site is
present in the PCR-5 cDNA sequence), indicating that the LjNOD16 gene is represented
by a single or low copy gene number in the L. japonicus genome (Fig. 6.6A). To
determine whether gene(s) homologous to L. japonicus LjNOD16 were present in the
genomes of other legumes, a Southern hybridization was performed using the full-length
cDNA for LjNOD16 as a probe. Several hybridizing bands were observed in genomic
DNA isolated from Lotus corniculatus, Glycine max, and Medicago sativa plants. Three
additional hybridizing fragments were detected in the L. japonicus genome under moderate
stringency conditions using the full-length cDNA as a probe (Fig. 6.6B). The latter results
indicate the presence of LjNOD16-related genes in the L. japonicus genome, providing
additional support to the conclusion drawn above with regard to expression of a related

message of high molecular weight in ﬂowers.

Localization of LjNODI6 transcripts in the L. japonicus nodule

Cellular localization of the LjNOD16 mRNA in nodules was analyzed using non-
radioactive in situ hybridization. Semi-thin sections (7 pm) of L. japonicus 21-d-old
nodules were hybridized with digoxigenin-UTP-labeled RNA probes transcribed from the
LjNODI6 and L. japonicus leghemoglobin cDNAs. Both sense and antisense transcripts

were used as probes. Upon examination of several consecutive sections derived from

222

different L. japonicus nodules, we observed that expression of LjNOD] 6 was restricted to
the infected cells of the nodule (Fig. 6.7 B&C), which closely resembled the expression
pattern of the L. japonicus leghemoglobin mRNA (Fig. 6.7D). No expression of LjNOD16
was detected in uninfected interstitial cells, nor in the peripheral tissues, such as the
cortex and endodermis. In addition, no hybridization signals were observed when
digoxigenin-UTP-labeled sense RNA transcripts corresponding to LjNODI6 was used as

a probe (Fig. 6.7A).

The Medicago sativa LjNODI6 homologue is expressed speciﬁcally in nodules

Since the L. japonicus LjNODI6 cDNA probe hybridized to DNA sequences in
the M. sativa genome, we investigated the tissue-speciﬁc expression of the corresponding
gene(s) using northern blot analysis. Ten micrograms of the total RNA isolated from
different M sativa tissues was hybridized with the radiolabeled L. japonicus LjNODI6
cDNA insert. Unexpectedly, no detectable signal could be found in any of the M sativa
tissues analyzed when total RNA was used (data not shown). We hypothesized,
therefore, that the corresponding gene(s) might be expressed at a low level in the

indeterminate nodules of M sativa plants. We tested this assumption using 4 pg of

poly(A)+ RNA derived from leaves, stems, roots, and nodules of the same 21-d-old M

sativa plant (Fig. 6.8). Two hybridizing bands corresponding to mRNA species of

223

approximately 800 nt and 1800 nt in length, respectively, could be detected speciﬁcally in

nodules, but not in any other M. sativa tissues analyzed.

The L. japonicus N1j16 protein and its M. sativa homologue can be detected in

nodules but not in the other tissues of L. japonicus

Polyclonal antibodies were raised against the recombinant L. japonicus Nlj16
protein, and used to carry out a western blot analysis using total protein extracts from
roots and nodules of L. japonicus and M sativa plants (Fig. 6.9). The recombinant Nlj16
protein was loaded on the SDS-PAGE gel as a positive control. Five to six protein bands
in the size range of 16 to 19 kD could be consistently detected in the extract derived from
L. japonicus nodules, but not from L. japonicus roots (Fig. 6.9A). In comparison, a 10-kD
protein was speciﬁcally detected in M. sativa nodule extract (Fig. 6.9A). Intense protein
bands were also observed in M. sativa root extracts. Since they could also be detected
using the pre-immune serum, it is highly unlikely that they reﬂect speciﬁc reactions with
the anti-LjN16 antibody (Fig. 6.98). The pre-immune serum failed to detect the control
recombinant N1j16 protein. In addition, neither the L. japonicus 16 to l9-kD protein
bands nor the 10-kD protein from M sativa nodule extracts were recognized by pre-
immune serum (Fig. 6.9B). The same nodule proteins were, however, speciﬁcally detected
with the antibodies puriﬁed from the anti-Nlj 16 serum by afﬁnity chromatography over a
column with a recombinant Nlj16 protein (Fig. 6.10). On the other hand, the immuno-

depleted fraction of the serum can no longer efﬁciently recognize the 16-19 kD protein

224

bands in L. japonicus nodules (data not shown). Based on these results, we conclude that
the proteins detected speciﬁcally in the nodule extracts of both plants analyzed are likely
to correspond to nodulin Nle6 and its M sativa homolog, respectively.

To further analyze the distribution of Nle6 in different tissues of L. japonicus,
the western blot with protein extracts from ﬂowers, leaves and stems, uninfected roots
and nodules of L. japonicus was probed with the afﬁnity-puriﬁed anti-Nlj16 antibody
(Fig. 6.10). As expected, the Nle6 protein could be detected only in the nodules (Fig.
6.10). The 40 kD nodule-speciﬁc band represents an unknown Rhizobial protein, also
present in free living Rhizobia loti and E.coli cultures (data not shown). A doublet of
bands around 60 kD could be detected in all the tissues of L. japonicus with a higher level
in ﬂowers (Fig. 6.10, the doublet is very faint in the nodules on this ﬁgure). The 60 kD
protein doublet most likely represents the product(s) of the larger, ~2500 nt transcripts,

cross-hybridizing with the LjNOD] 6 probe (Fig. 6.4, see the next chapter of this thesis)

Analysis of the transgenic L. japonicus plants carrying constructs pCRSAnti and
pCRSSense

In total, 32 independent L. japonicus transgenic lines, carrying pCRS sense
construct, and 44 lines carrying pCR5antisense construct, were generated and analyzed.
The nodules from 4 independent transgenic L. japonicus lines (pCR5Anti-3, 11, 18 and
25) were to found to contain signiﬁcantly diminished levels of Nle6 protein compared
with the wild-type nodules (Fig. 6.11). The nodules from the transgenic line pCRSAnti-

lcontain the typical amount of Nlj 16 protein present in the nodules from the majority of

225

L. japonicus transgenic plants, lower than in the wild-type L. japonicus nodules 21dai
(Fig. 6.11). The nodules from the line pCR5Anti-25 did not contain detectable amounts of
Nle6 protein as judged by the western blot analysis (Fig. 6.11). However, regardless of
the level of Nle6 nodulin, the nodules from all transgenic pCR5Anti lines had a normal
appearance as judged by normal size, shape and a pink color indicative of the presence of
late nodulin leghemoglobin. Examination of the sections from pCR5Anti-18 and
pCR5Anti-25 nodules using light microscopy did not reveal any signiﬁcant structural
differences from the wild-type nodules (data not shown). Also, the plants from both
pCR5Anti-l8 and pCR5Anti-25 transgenic lines did not show chlorosis, a typical sign of
nitrogen starvation, thus further indicating that the nodules on these plants ﬁx normal
amount of nitrogen. Therefore, L. japonicus nodules with signiﬁcantly diminished levels
of Nlj 16 protein appear to develop and function normally. None of the tested pCRSSense
transgenic L. japonicus plants had nodules with elevated levels of Nle6 protein

comparing to the wild-type nodules (data not shown).

6.5. DISCUSSION

We have isolated and characterized a cDNA clone from the model legume L. japonicus
representing a novel late nodulin gene, designated LjNOD16. One intriguing feature of the
LjNODI6 is its very high level of nodule-speciﬁc transcriptional activity, which is similar
to the transcriptional activity of the leghemoglobin genes in L. japonicus nodules. This

property inspired us to perform a more detailed molecular analysis of the LjNODI6 gene.

226

The expression pattern of the LjNODI6 gene, based on the developmental northern blot
analysis, showed distinct characteristics of late-nodulin genes. The corresponding mRNA
was ﬁrst detectable in the L. japonicus roots around 11 d after Rhizobium infection and
accumulated to a very high level in fully developed nodules. This expression pattern may
suggest that the product of the LjNOD16 gene is involved in relatively late stages of
nodule ontogeny and/or nodule functioning. This assumption was further supported by
the localization of transcripts corresponding to the LjNODI6 gene in infected cells of L.
japonicus nodules. Infected-cell-specifrc expression has also been demonstrated for
leghemoglobin (lb) genes (Szczyglowski et al., 1994). Since in situ localization
experiments were performed using fully developed L. japonicus nodules, it is not clear
whether the induction of LjNODI6 gene expression precedes bacterial colonization of the
plant cells, or occurs concomitantly with the release of symbiotic bacteria from the
infection threads and commencement of nitrogen ﬁxation. The latter expression pattern is
characteristic of the majority of late-nodulin genes identiﬁed thus far (Nap and Bisseling,
1990; Govers et al., 1987). Taking into account the time point after Rhizobium infection
when the LjNOD16 gene becomes activated, we propose that it is a late nodulin gene,
appearing to be coordinately induced just prior to, or concomitant with, the
commencement of nitrogenase activity.

Northern blot analysis using different L. japonicus tissues revealed the
presence of very low levels of mRNAs hybridizing to LjNOD16 in uninfected roots and
ﬂowers, in addition to nodules. A faint band corresponding to the size of the LjNODI6

mRNA was detected in the uninfected roots after a prolonged exposure of the blot. A

227

different mRNA species of approximately 2500 nt was detected in ﬂowers, suggesting
that there may be more than one gene related to LjNODI6 in the L. japonicus genome.
This assumption was supported by the data obtained from our Southern blot analyses.
Several hybridizing bands of different intensities were observed under moderately
stringent conditions. However, when high stringency hybridization conditions were used,
and the 3’-portion of the LjNODI6 cDNA (PCR5) was employed as a probe, single
hybridizing bands were detected, indicating that the LjNOD16 gene is likely to be
represented by a single or low-copy gene number in the L. japonicus genome. A full-copy
LjNOD16 cDNA hybridized with several genomic fragments derived from different
legume species. This result suggests that the sequences related to the L. japonicus
LjNODI6 gene are present in the genomes of different legume plants, and may play a role
in both determinate and indeterminate nodules. This assumption was extended by the
result of the northern analysis using mRNAs derived from different M sativa tissues. The
LjNOD16 cDNA probe hybridized speciﬁcally with two distinct nodule-speciﬁc mRNA
species, indicating that homologous transcripts are present in the indeterminate nodules of
M sativa plants.

Interestingly, the level of alfalfa mRNAs appeared to be signiﬁcantly

lower than the level of LjNODI6 mRNA in L. japonicus nodules. In the former case, 4 pg

of poly(A)+ RNA and a long exposure time of 7 to 10 (1 produced detectable
hybridization signals, clearly contrasting with the situation observed in the L. japonicus

nodules. The apparent low abundance of alfalfa NOD] 6 mRNAs is not likely to be due to

228

the fact that a heterologous L. japonicus LjNOD16 probe was used, since the same DNA
insert gave a clearly detectable signal when hybridized with M sativa genomic DNA (Fig.
6.68). Although this remains to be proven, it is possible that a different mode of gene
regulation may account for the observed difference in the abundance of homologous
mRNAs in nodules of L. japonicus and M sativa plants. It is noteworthy here that in the
case of the L. japonicus nodule extract, an array of closely migrating proteins was
consistently detected using anti-Nlj 16 antibody. Although the basis for this phenomenon
is not clear, it may reﬂect the presence of different forms of Nlj16 protein in the L.
japonicus nodules, perhaps due to post-translational modiﬁcations.

A detailed analysis of the amino acid sequence of the L. japonicus protein,
deduced from the full-length cDNA sequence, revealed the presence of several interesting
domains. Nlj16 was predicted to be a soluble protein, as suggested by Kyte-Doolittle
hydropathy analysis. However, N1jl6 contains a hydrophobic domain, consisting of a
regularly spaced prolines, the function of which is not clear. Secondary structure

predictions strongly indicated the presence of an extended (at-helical domain, in which two

a-helices are separated by a putative turn. Homology searches using the predicted amino-

acid sequence of Nlj 16 revealed a signiﬁcant similarity to two closely related anonymous

Arabidopsis ESTs mostly in the predicted a-helical domain. Interestingly, both L.
japonicus and Arabidopsis or-helical sequences showed a very high propensity to form a

coiled-coil-type structure, which may indicate that the nodulin Nljl6, as well as the

Arabidopsis proteins, may interact with as yet unknown proteins in the plant cells.

229

-1I_-

With regard to the size of corresponding mRNAs, the Arabidopsis ESTs seem to
represent a counterpart of the ﬂower-speciﬁc L. japonicus mRNA. The mRNA
corresponding to the nodule-speciﬁc gene is signiﬁcantly smaller than either Arabidopsis
or L. japonicus ﬂower mRNA species. It is tempting, therefore, to speculate that the
nodulin Nljl6 may represent a truncated version of the otherwise constitutively
expressed proteins (as exempliﬁed by the ubiquitous expression of the Arabidopsis
ESTs), primarily harboring the conserved domain, which was adopted during evolution
for speciﬁc symbiotic function(s). However, the exact role of nodulin Nle6 in the
infected cells of L. japonicus nodules remains to be determined.

The L. japonicus nodules containing signiﬁcantly reduced levels of nodulin Nljl6
appear to function normally. However, the residual levels of Nle6 protein may be
sufﬁcient to fulﬁll its normal function(s) in L. japonicus nodules. Therefore to
unequivocally address the importance of Nlj 16 for nodule development and functioning it
is necessary to generate a loss-of-function mutation in the corresponding gene. The over-
expression of Nlj16 protein in L. japonicus nodules was attempted as a complementary
approach to investigate the function of this nodulin. However, an increase in the Nle6
levels was not achieved. Furthemore, an attempt to over-express Nle6 in Arabidopsis
was not successful either. Transgenic Arabidopsis plants carrying the LjNODI6 cDNA
under the control of the 358 CAMV promoter failed to produce any detectable quantities
of Nlj16 protein despite the fact that LjNODI6 mRNA was produced at high levels in

these plants (data not shown). It is therefore possible that the level of Nle6 protein is

230

controlled by post-transcriptional mechanisms, such as regulation of mRNA translation

or protein stability.

_
O
h
«a
=
Q
0

days after
infection

—188

 

Figure 6.1. Northern blot analysis of PCR5 expression. 10 pg of total RNA

isolated from 21-d-old uninfected roots (control), and root segments and nodules harvested
7, ll, 13 and 21 d after infection, were analyzed using the radiolabeled PCR-5 product as a

probe (top panel). The blot was re-probed with 18S ribosomal DNA, as an RNA loading

control (bottom panel).

231

Figure 6.2. Nucleotide and deduced amino acid sequences of the

LjNODI6 cDNA. (A) The positions of the multiple polyadenylation sites

are indicated by the bold underlined nucleotides. Predicted a-helical

domains of nodulin Nljl6 are underlined, and the four prolines of the N-
terminal hydrophobic stretch are indicated in bold. The asterisk denotes

the predicted stop codon.

232

CTCGTCGAATTCGGCACGAGGTTGTATATAGATAAAATTATATGAAGATCTTGCAGCTTGTAGGT
M K I L Q L V G
CCTTCTGAGCATATAGAGTTTGTCCCTGCTGCTAAACTTTCCAAGAACGTGGACGTAATCCCTGT
P S E H I E F V P A A K L S K N V D V I P V
GGCTATCCCCGTGGGTGTCCCGGTGGCTGTTCCAGCGGCTGACAAAAATGCATCGAAGAAAGTTG
A I P V G V P V A V P A A D K N A S K K V G
GTCAGAATGACACAACGTCCAAAGAGTTTACAACTGTGATGAAACGCATGGCTGAGTTGGAAGAG
Q N D T T S K E F T T V M K R M A E L E E
AAAATGACCACCATGAATCATCAGCCTGCTACCATGCCGCCAGAGAAGGAGGAAATGCTGAATGC
K M T T M N H Q P A T M P P E K E E M L N A
TACTATAAGTCGAGCGGATGTCTTAGAGAAACAACTTATGGACACCAAGAAGGCTTTGGAGGATT
T I S R A D V L I K O L M D T K K A L E D S
CGCTTGCTAAGCAAGAGGTGCTTTCAGCTTATGTTGAGAAAAAGAAACAGAAGAAGAAGACGTTT
L A K O E V L S A Y V E K K K O K K K T F
TTCTGCTGTTAAGTGCGAAATAGTGGATGCCAAACAAGAGGTCCATATATTAAAGGAGAGTTGAC
F C C *
TTTTACTTTAAGCTTTTTCAGGAGACTCCCAAAGTGCCCTCTTACCATAAGTGTGCAATATGCGT
ACTAGTAAATCTACATGACGCCAAGAAAAGGATATGCTCAATGTGTGAATTGTATATGCTAATCT
GAATTTTTGTCTCTACTAGGTTAGGACTACTTGCTCTCTATGTATATAGGGCTATGCACAATATG
CAAAGCCAGCCCTGTGATGTCCTTACAGGAACAATACATCTAGTGCAAAATTAGAGAGTATGTTT
GATTTTTATAIACATGGAAATCTACCTCCAACGAATGAGCTCTTTCCTTTTGCTTATGTGACAAG
AAAAAAGCATTACTAGTTGTGAQAAGTACAATTGGCCATATTTGTGTGTQAACTGAGCTGTAAAT
GAAGCAAAGTTTTATTTAAATATAGAAAATGTGACTCTTGIAAAQATTTGTTCAATTACCACAAT
CAGGGGCTGACATGAACTCATGAAGGCTTTTGGCE

233

65

130

195

260

325

390

455

520

585

650

715

780
845

910
975
1010

cu

 

3|!

8

.E 3.46

U

:E thilic
ﬂ

5 o

8'

r. thobic
E, -2.2

I

 

I r f I r r I I I I r fﬁ I
10 20 30 40 50 60 70 80 90 100110120130140

Amino acid number

Figure 6.2. Nucleotide and deduced amino acid sequences of the LjNODI6 cDNA.
(B) Hydropathy proﬁle of nodulin Nlj16 determined by the method of Kyte and

Doolittle (1982) with a window of 7 residues.

234

Nljl6
A.t.ESTl

A.t.ESTZ

Nljl6
A.t.ESTl

A.t.EST2

Nljl6
A.t.ESTl

A.t.ESTZ

Nlj16
A.t.ESTl

A.t.EST2

Figure 6.3. Alignment

51

....... MKI

VDATWKVKPA

LQLVGPSEHI
:l |:..:
IQRVASRGAL
|-|-:|=||
INRAPSKGAH

VGVPVAV .................... PAADK

Ill
MAFLMAVFTF
ll|=||=:|l
MAFVMAILTF

SPVPDLTETD

l-l
....... BAD

151

ADVLEKQLMD
.|.|I :l:.
VDALEAELIA

||l|||||||
VDALEAELIA

FRTVTKKLPA
IIII..::
FRTVSNRV..

.FTTVMKRMA

...I |::.
LLNCVTKKLT
|||=| ||||
LLNSVLKKLT

TKKALEDSLA

lllll =-l
TKKALYEALM

l|||||||||
TKKALYEALM

||...
TTTSSPAETQ

.I. |:..
VTKQLPPPPS

ELEEKMTTMN
l||=|:-|=-
ELEGKIGTLQ

|||=|ll-||
ELEEKIGALQ

KQEVLSAYVE
:II | ll==
RQEELLAYID

|||l|l||||
RQEELLAYID

of Nle6 with the

EFVPAAKLSK

...:I .
MSPTVPKDHE
l-l-Illlll
MPPNVPKDHE

NASKKVGQND
a | k
GNAIELGSNG

.Il
QPQIEGSAAA

HQPATMPPEK

.I.-ll ll
SKPNEMPYEK

Ill-Illlll
SKPSEMPYEK

K .........

RQEEAQFQKM
llloll l
RQEAAQHQ..

50
NVDVIPVAIP
GIKARVLVMF
==-||||| I
SFSARVLVTF

100

TTSKE .....
ll

EGVKEECRPP

I
E .........

150
EEMLNATISR
||=||l.=:|
EELLNAAVCR

Illlllllll
EELLNAAVCR

200
KKQKKKTFFC
11.111 :11
KKKKKKHLFC
||-|=|=:||
KKNKRKQMFC

deduced aminoacid

C

F

|
F

sequences of

Arabidopsis ESTs 168K8 (A.t. ESTI) and 110G16 (A.t. EST2). The alignment was

performed using the GCG package. Vertical bars indicate identical amino acids,

semicolons represent conservative substitutions and dots indicate semiconservative

substitutions.

235

 

 

 

g; 2
.... :3
58522
Ln .4 m a: Z
,. -28S
H -18S
—1ss

 

 

 

 

28S and 18S rRNAs are as indicated.

Flower
Leaf
Stem
Root

 

III! anal anal III!

 

 

 

Figure 6.4. Expression of LjNODI6 in
different tissues of L. japonicus. Ten
micrograms of total RNA from L.

japonicus ﬂowers, leaves, stems,

uninfected roots and nodules

probed with radiolabeled LjNOD16

were

cDNA. The blot was exposed for 7 days

to visualize the less intensely
hybridizing mRNA species from ﬂowers
and uninfected roots, respectively (top
panel). Equal RNA loading was veriﬁed
by probing the same ﬁlter with the 188

DNA probe (bottom panel). Positions of

Figure 6.5. Expression
pattern

EST

of Arabidopsis
168K8. Ten
micrograms of total RNA

from ﬂowers, leaves, stems

-288

and roots of Arabidopsis
thaliana were separated by

agarose-formaldehyde gel,

-18S

transferred to nitrocellulose

and hybridized to 168K8

DNA labelled with [or-32P]dATP. Positions of 28S and 18S rRNAs are as indicated.

236

Figure 6.6. Organization of the LjNODI6 gene in the genomes of L.
japonicus and other legume species. (A) Ten micrograms of L. japonicus
genomic DNA were completely digested with the restriction enzymes indicated,

separated on 0.8% agarose gel and probed with [or-32P]dATP labelled PCR-5

DNA. Hybridization and washes were performed under high stringency
conditions (see Materials and Methods). (B) Southern blot hybridization with

lz't'oRl digested DNA (lOpg) from L. japonicus cv. Gifu, L. corniculatus cv.

Rodeo, Medicago sativa ev. Cardinal and Glycine max cv. Dimon. A full length
Lil-W )1)! 6 cDNA was used as a probe. Hybridization was performed under low
stringency conditions (see Materials and Methods). The ~4.0 kb EcoRI ﬁagment

corresponds to the LjNODI6 gene.

'JJ

RS: .6

333 .3
@533853 4
manuaeﬁax ..N

 

B

p
.D
k
s eav—
m Essa
WEE:
.J
L. 28H

A

 

238

Figure 6.7. In situ localization of LjNODI6 and leghemoglobin transcripts
in sections of 21-d-old L. japonicus nodules. Dark ﬁeld micrographs are shown
in which the hybridization signal appears as pink color. (A) Section hybridized
with sense LjNOD/6 RNA probe. (B) Section hybridized with the anti-sense
LjNOD/6 RNA probe. (C) Detailed view of a region containing both infected and
uninfected cells from the section shown in panel B. (D) Detailed view of a region
containing both infected and uninfected cells from a section hybridized with anti-
sense probe corresponding to the L. japOnicuS leghemoglobin gene. This image is

presented in color.

ﬂ...
W”

n
:‘L‘

 

240

 

 

Nodule

«- E H

c
3 8 o
..1 m a:

 

Figure 6.8. Expression of putative homologue(s) of LjNOD16 in different tissues of
M. sativa. Four micrograms of poly (A)+ fraction of total RNA isolated from leaves,
stems, roots, and nodules of 21-d-old nodulated alfalfa plants were separated on 1.2%
agarose-formaldehyde gel, blotted on nitrocellulose, and probed with radiolabelled
LjNODI6 cDNA fragment. Two hybridizing bands with approximate length of 800 nt
and 1800 nt, detected under low stringency conditions (see Materials and Methods), are
shown. The blot was stripped and rehybridized with an ubiquitin cDNA from Sesbania

rostrata as a loading control.

241

Figure 6.9. Detection of
Nle6 protein and its
putative alfalfa
homologue in nodules of

L. japonicus and M. sativa.

 

Lane 1, recombinant Nljl6
protein (12.5 'ng); lanes 2
and 4, total protein (50 pg) extracted from roots of L. japonicus and M sativa plants,
respectively; lanes 3 and 5, total protein extract (50 pg) from nodules of L. japonicus and
M sativa plants, respectively. (A) Western blot analysis with anti-N1j16 serum. (B).

Western blot analysis with pre-immune serum.

Figure 6.10. Expression of Nle6 in

different tissues of L. japoniucs. Equal

Leaves and Stems

Nodules
Roots
Flowers

amount of total protein extracted from
nodules, uninfected roots, leaves and stems
and ﬂowers of L. japonicus were probed with

the afﬁnity-puriﬁed anti-Nlj 16 antibody.

 

242

pCR5Anti lines
25 1811 3 1

wt nodules

Size (kD)

    

Figure 6.11. The levels of Nle6 protein in the nodules from the different
pCR5Anti transgenic lines of L. japonicus. Equal amounts of total nodule protein from
the indicated lines and wild-type (wt) nodules were probed with the afﬁnity-puriﬁed
anti-Nlj16 antibody. The cross-hybridizing protein species migrating at ~40 kD

represents an unknown Rhizobial protein and can also serve as a loading control.

243

6.6. ACKNOWLEDGMENTS

During the course of this work, Krzysztof Szczyglowski visited the laboratory of Dr.
Ton Bisseling. We thank Drs. Katharina Pawlowski and Ton Bisseling for their help and
many useful suggestions with regard to the in situ hybridization procedure. We wish to
thank Dr. Marcelle Holsters from the University of Gent, Belgium for kindly providing
the ubiquitin clone from Sesbania rostrata. We thank Drs. Lee McIntosh, John Wilson,
and Leslie Kuhn for helpful discussions. We thank Scott Shaw for the preparation of the
plant material used in this work. We also thank Kurt Stepnitz and Marlene Cameron for
their help in preparing the ﬁgures, Karen Bird for expert editorial assistance, and Susan
Fujimoto, David Silver, Dirk Hamburger, and Rujin Chen from our laboratory for their

advice and helpful comments on the manuscript.

244

6.7. REFERENCES

Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ (1990) Basic local alignment
search tool. J Mol Biol 215: 403-410

Bennett MJ, Lighftoot DA, Cullimore JV (1989) cDNA sequence and differential
expression of the gene encoding the glutamine synthetase y-polypeptide of Phaseolus
vulgaris L. Plant Mol Biol 12: 553-565

Bergmann H, Preddie E, Verma DPS (1983) Nodulin-35: a subunit of speciﬁc uricase
(uricaselI) induced and localized in the uninfected cells of soybean nodules. EMBO J 2:
2333-2339

Boron L, Szczyglowski K, Konieczny A, Legocki AB (1989) Glutamine synthetase in
Lupinus luteus. Identiﬁcation and preliminary characterization of nodule-speciﬁc cDNA
clone. Acta Biochim Polon 36: 295-301

Boron L , Legocki AB (1993) Cloning and characterization of a nodule-enhanced
glutamine synthetase-encoding gene from Lupinus luteus. Gene 136: 95-102

Brisson NA, Pombo-Gentile A, Verma DPS (1982) Organization and expression of
leghemoglobin genes. Can J Bot 60: 272-278

Broughton WJ, Dilworth MY (1971) Control of leghemoglobin synthesis in snake
beans. Biochem J 125: 1075-1080

Church GM, Gilbert W (1984) Genomic sequencing. Proc Natl Acad Sci USA 81: 1991-
1995

Cook D, Dreyer D, Bonnet D, Howell M, Nony N, VandenBosch K (1995) Transient
induction of a peroxidase gene in Medicago truncatula precedes infection by Rhizobium
melilati. Plant Cell 7: 43-55

Csanadi G, Szecsi J, Kalo P, Kiss P, Endre G, Kondorosi A, Kondorosi E, Kiss GB
(1994) ENOD12, an early nodulin gene, is not required for nodule formation and efﬁcient
nitrogen ﬁxation in alfalfa. Plant Cell 6: 201-213

de Bruijn FJ, Schell J (1992) Regulation of plant genes speciﬁcally induced in
developing and mature nitrogen-ﬁxing nodules: cis-acting elements and trans-acting
factors. In DPS Verma, ed, Control of Plant Gene Expression, CRC Press, Boca Raton,
pp 241-258

245

Delauney AJ, Verma DPS (1988) Cloned nodulin genes for symbiotic nitrogen ﬁxation.
Plant Mol Biol Rep 6: 279-285

Engler JA, Van Montagu M, Engler G (1994) Hybridization in situ of whole mount
messenger RNA in plants. Plant Mol Biol Rep 12: 321-331

Fortin MG, Zalechowska M, Verma DPS (1985) Speciﬁc targeting of membrane
nodulins to the bacteroid enclosing compartment in soybean nodules. EMBO J 4: 3041-
3046

Fortin MG, Morison NA, Verma DPS (1987) Nodulin-26, a peribacteroid membrane
nodulin is expressed independently of the development of the peribacteroid compartment.
Nucleic Acids Res 15: 813-824

Franssen HJ, Vijn I, Yang WC, Bisseling T (1992) Developmental aspects of the
Rhizobium-legume symbiosis. Plant Mol Biol 19: 89-107

Gebhardt C, Oliver JE, Forde BG, Saarelainen R, Miflin BJ (1986) Primary
structure and differential expression of glutamine synthetase genes in nodules, roots and
leaves of Phaseolus vulgaris. EMBO J 5: 1429-1435

Govers F, Nap JP, van Kammen A, Bisseling T (1987) Nodulins in the developing root
nodule. Plant Physiol Biochem 25: 309-322

Gresshoff PM (1993) Molecular genetic analysis of nodulation genes in soybean. Plant
Breeding Reviews 11: 275-318

Hall TC, Ma Y, Buchbinder BU, Pyne JW, Sun SM, Bliss FA (1978) Messenger
RNA for G1 protein of French bean seeds: cell-free translation and product
characterization. Proc Natl Acad Sci USA 75: 3196-3200

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

Handberg, K., Stiller, J., Thykjear, T., and Stougaard, J. (1994) Transgenic plants:
Agrobacterium mediated transformation of the diploid legume Lotus japonicus. In: Cell

Biology: A Laboratory Handbook, Celis JE, ed, 1, 119-127

Heard J, Dunn K (1995) Symbiotic induction of a MADS-box gene during development
of alfalfa root nodules. Proc Natl Acad Sci USA 92: 5273-5277

246

Horvath B, Bachem CWB, Schell J, Kondorosi A (1987) Host-speciﬁc regulation of
nodulation genes in Rhizobium is mediated by the plant signal, interacting with the nodD
gene product. EMBO J 6: 841-848

Jarvis BDW, Pankhurst CE, Patel JJ (1982) Rhizobium loti, a new species of legume
root nodule bacteria. Int J Syst Bact 32: 378-380

Jiang Q, Gresshoff PM (1993) Lotus japonicus: a model plant for structure-function
analysis in nodulation and nitrogen ﬁxation. Curr Topics Plant Mol Biol 2: 97-110

Kyte J, Doolittle RF (1982) A simple method for displaying the hydropathic character
of a protein. J Mol Biol 157: 105-132

Laemmli UK (1970) Cleavage of structural proteins during the assembly of the head of
bacteriophage T4. Nature 227: 680-685

Lara M, Cullimore JV, Lea PJ, Johnston AWB, Lamb JW (1983) Appearance of a
novel form of plant glutamine synthetase during nodule development in Phaseolus
vulgaris L. Planta 157: 254-258

Legocki RP, Verma DPS (1979) A nodule-speciﬁc plant protein (Nodulin-35) from
soybean. Science 205: 190- 193

Lerouge P, Roche P, F aucher C, Maillet F, Truchet G, Prome JC, Dénarié, J (1990)
Symbiotic host-speciﬁcity of Rhizobium meliloti is determined by a sulfated and acylated
glucosamine oligosaccharide signal. Nature 344: 781-784

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

Lupas A (1996) Prediction and analysis of coiled-coil structures. Methods of
Enzymology 266: 513-525

Miao G-H, Verma DPS (1993) Soybean nodulin-26 gene encoding a channel protein is
expressed only in the infected cells of nodules and is regulated differently in roots of

homologous and heterologous plants. Plant Cell 5: 781-794

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

Nap J-P, Bisseling T (1990) Developmental biology of a plant-prokaryote symbiosis:
the legume root nodule. Science 250: 948-954

247

Pridmore RD (1987) New and versatile cloning vectors with kanamycin resistance
marker. Gene 56: 309-312

Rogers SO, Bendich AJ (1988) Extraction of DNA from plant tissues. In: SB Gelvin,
RA Schilperoort, eds, Plant Molecular Biology Manual. Kluwer Academic Publishers,
Dordrecht, A6 1-10

Rost B, Sander C (1993) Prediction of protein secondary structure at better than 70%
accuracy. J Mol Biol 232: 584-599

Rost B, Sander C (1994) Combining evolutionary information and neural networks to
predict protein secondary structure. Proteins 19: 55-77

Sambrook J, F ritsch EF, Maniatis T (1989) Molecular Cloning. A Laboratory Manual,
ED2. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY

Schultze M, Kondorosi E, Ratet P, Buire M, Kondorosi A (1994) Cell and molecular
biology of Rhizobium-plant interaction. Int Rev Cytol 156: 1-75

Spaink HP, Wijffelman CA, Pees E, Okker RJH, Lugtenberg BJJ (1987) Rhizobium
nodulation gene nodD as a determinant of host speciﬁcity. Nature 328: 337-340

Spaink HP (1992) Rhizobial lipooligosaccharides: answers and questions. Plant Mol Biol
20: 977-986

Sprent J] (1989) Which steps are essential for the formation of functional legume
nodules? New Phytol 111:129-153

Stiller J, Martirani L, Tuppale S, Chian R-J, Chiurazzi M, Gresshoff PM (1997)
High frequency transformation and regeneration of transgenic plants in the model legume
Lotus japonicus. J Exp Bot 48: 1357-1365

Szczyglowski K, Szabados L, F ujimoto S, Silver D, de Bruijn FJ (1994) Site-speciﬁc
mutagenesis of the nodule-infected cell expression (NICE) element and the AT-rich
element ATRE-BSZ* of the Sesbania rostrata leghemoglobin glb3 promoter. Plant Cell 6:
317-332

Thummler F, Verma DPS ( 1987) Nodulin-100 of soybean is the subunit of sucrose
synthetase regulated by the availability of free heme in nodules. J Biol Chem 262: 14730-
14736

Towbin H, Staehelin T, Gordon T (1979) Electrophoretic transfer of proteins from
polyacrylamide gels to nitrocellulose sheets. Proc Natl Acad Sci USA 76: 4350-4354

248

Triplett EW (1985) Intercellular nodule localization and nodule speciﬁcity of xanthine
dehydrogenase in soybean. Plant Physiol. 77: 1004-1005

Van de Wiel C, Scheres B, Franssen H, Van Lierop MJ, Van Lammeren A, Van
Kammen A, Bisseling T (1990) The early nodulin transcript ENOD2 is located in the

nodule parenchyma (inner cortex) of pea and soybean root nodules. EMBO J 9: 1-7

van Kammen A (1984) Suggested nomenclature for plant genes involved in nodulation
and symbiosis. Plant Mol Biol Rep 2: 43-45

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

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

Welsh J, Changa K, Dalal SS, Cheng R, Ralph D, McClelland M (1992) Arbitrarily
primed PCR ﬁngerprinting of RNA. Nucleic Acids Res 20: 4965-4970

249

CHAPTER 7

Lotus japonicus phosphatidylinositol transfer-like proteins with a
plasma membrane targeting domain expressed separately in root

nodules

7.1. ABSTRACT

Phosphatidylinositol transfer proteins (PITPs) are the essential components of
vital cellular functions in eukaryotes, such as phosphoinositol-mediated signaling and
vesicle-traﬁcking. Here the characterization of a gene family from Lotus japonicus,
encoding a novel class of plant PITP-like proteins, called LjPLPs, is described. The
members of this gene family were identiﬁed based on their nucleotide sequence
homology to a cDNA, LjNODI6, encoding L. japonicus late nodulin, Nljl6. LjPLPs
contain an N-terminal domain related to the yeast secl4p PITP that is also capable of
complementing a mutation in the secl4 gene. In addition, LjPLPs contain a C-terminal
domain represented by either Nljl6, or a highly related amino-acid sequence, involved in
targeting LjPLPs to the plasma membrane. Furthermore, nodule-speciﬁc LjNOD16
mRNA was shown to be the result of an unusual transcriptional event, mediated by a
nodule-speciﬁc, bi-directional promoter present in an intron of a member of the LjPLP
gene familly. Finally, we suggest that nodulin Nljl6 may exert a dominant negative effect
in nodules, and discuss its possible role in a mechanism that inactivates the function of

LjPLPs.

250

7.2. INTRODUCTION

Phosphatydilinositol transfer proteins (PlTPs) represent an interesting group of
eukaryotic proteins that play important roles in diverse cellular processes involving
phospholipid metabolism (Cockcroft, 1998; Keams et al., 1998a). PIPTs were originally
discovered based on their unique biochemical properties, such as binding and catalyzing
the exchange of phosphatidylinositol (P1) or phosphatidylcholine (PC) moieties between
membrane bilayers in vitro (Cleves et al., 1991a). The ﬁrst clue about their physiological
functions emerged from genetic studies in Saccharomyces cerevisiae (Aitken et al., 1990;
Bankaitis et al., 1989; Bankaitis et al., 1990), showing that a mutation (secI4-1’s) in the
gene encoding the major PITP transfer protein resulted in the impairment of secretory
function of the Golgi apparatus and seriously affected yeast viability. The sec14 locus
was postulated to have a regulatory role in this process due to its ability to maintain a
critical level diacylglycerol (DAG) pool in the Golgi, which, in turn, has been assumed to
be required for the formation of secretory vesicles in Golgi membranes (McGee et al.,
1994; Keams et al., 1997). Experimental evidence suggests that yeast secl4p PITP acts as
a sensor of phospholipid concentrations in the cell and indirectly helps to preserve a
necessary level of DAG in the Golgi, by modulating the rates of biosynthesis or turnover
of PC and P1. The PC-bound form of secl4p inhibits the rate-limiting enzyme of the PC
biosynthetic pathway, cholinephosphate cytidyltransferase, thus preventing consumption

of Golgi-derived DAG (Skinner et al., 1995). On the other hand, the PI-bound form of

251

Sec14p is believed to enhance the rate of PI turnover, thus increasing DAG levels in
Golgi membranes (Keams et al., 1998a).

Similarly to yeast secl4p, reconstitution studies using perrneabilized cells have
implicated PITPs in trans-Golgi trafﬁcking events in mammals as well (Ohashi et al.,
1995). In addition, PITPs have been postulated to play an important role in cellular
mechanisms that compartmentalize and coordinate the synthesis of
phosphatidylinositol(4,5)biphosphate (PIPZ). The latter compound (PIPZ) is required for
regulated exocytosis in mammalian cells, and has also been shown to constitute a
substrate for phospholipase C (PLC) and PI 3-kinase activities. In agreement with the
postulated function in PIP2 biosynthesis, PITPs have been identiﬁed as “priming “ factors
for exocytosis in mammalian cells, as well as crucial components of PLC and PI 3-kinase
mediated signaling pathways (Hay and Martin, 1993; Hay et al., 1995; Thomas et al.,
1993; Kauffmann-Zeh et al., 1995; Cunningham et al., 1996). One possible function of
PITP in these processes could be the transport of PI lipids from the location of their
synthesis in the endoplasmic reticulum (ER), to the plasma membrane, where PIPZ
synthesis and hydrolysis take place (Cockcroft, 1998; Keams et al., 1998a). However,
strong evidence also exists that the role of PITPs may be more complex than that of a
simple shuttle protein (Cockcroft, 1998; Keams et al., 1998a). In fact, it has been
suggested that PITPs may transfer phospholipids to speciﬁc regions within the cellular
membranes or to speciﬁc protein complexes, mediating differentially regulated signal

transduction events (Cockcroft, 1998; Keams et al., 1998a; see also Discussion).

252

Plant proteins similar to fungal PITPs have also been identiﬁed and characterized.
These include the Sshlp and Ssh2p proteins from soybean and AtSEC14 from
Arabidopsis (Keams et al, 1998b; Jouannic et al., 1998). Plant PITPs share approximately
25% amino acid identity and 36-50% similarity with the S. cerevisiaea secl4p protein,
but do not have any sequence similarity to mammalian PlTPs (Keams et al, 1998b;
Jouannic et al., 1998). Ssh2p and AtSEC14 are capable of transfering PI, but not PC, in
vitro, and thus appear to have different biochemical properties in comparison to the
corresponding yeast and metazoan PITPs (Keams et al, 1998b; Jouannic et al., 1998).
Moreover, Sshlp displayed no detectable P1 or PC transfer activity under the conditions
tested. Interestingly, Sshlp was found to undergo phosphorylation in response to various
environmental stress conditions, including hyperosmotic stress, suggesting its
physiological role in plant osmoprotection (Keams eta1., 1998b).

As a part of an extensive search for expressed sequence tags (ESTs) correlated
with late stages of symbiotic root nodule development in L. japonicus, we previously
identiﬁed and characterized a novel nodule-speciﬁc cDNA, LjNOD16, encoding nodulin
Nlj 16 (Kapranov et al., 1996; Szczyglowski et al., 1997). Here we report that LjNOD16
transcripts originate in nodule tissues as a result of unusual transcription events governed
by an intron-localized promoter sequence in the L. japonicus LjPLP-IV gene.
Furthermore, we show that LjPLP-IV gene is a member of a gene family encoding a
novel class of PITP-like proteins, sharing on average 40% identity and 60% similarity
with the S. cerevisiae secl4p protein and capable of complementing the temperature-

sensitive phenotype of a yeast sec14-l"’mutant strain. This new family of PITP-like

253

proteins can be distinguished from previously described PITPs by the presence of a C-
terrninal extension comprised of either Nlj16 derived or other, highly related amino-acid
sequences. We also present evidence for a role of the Nle6 C-terminal extension in
targeting LjPLPs to the cell plasma membrane. Finally, we suggest that nodulin Nlj16
may exert a dominant negative effect in nodules, and discuss its possible role in a

mechanism that inactivates the function of PLPs.

7.3. RESULTS

LjNODI6 shares a region of similarity with a novel class of PITP-like genes.

We previously described the isolation and preliminary characterization of a Lotus
japonicus cDNA, LjNOD16, corresponding to a highly abundant mRNA species present
in nitrogen ﬁxing root nodules (Kapranov et al., 1997). LjNOD16 mRNA was localized
to the infected, bacteroid containing, cells of L. japonicus nodules, and was found to
encode a 15.5 kD protein, termed nodulin l6 (Nlj l6; Kapranov et al., 1997). Analysis of
the predicted amino-acid sequence of nodulin Nlj 16 revealed several interesting features,
including the presence of two a-helical regions with a high propensity to form coiled-coil
structures, and a positively charged C-terminus (Kapranov et al., 1997). The latter
domain, together with two cystein residues present at the penultimate and ultimate
positions of the Nlj 16 protein, showed a striking resemblance to known plasma

lK’m‘iB’ protein (Hancock et a1. 1991). Interestingly,

membrane targeting motifs of the p2
higher molecular weight mRNA species hybridizing to the LjNODI6 probe were detected

in L. japonicus ﬂowers (Kapranov et al., 1997). In addition, the deduced amino-acid

254

sequence of Nle6 protein was found to share a signiﬁcant level of similarity with C-
terminal regions of much larger proteins represented by two anonymous Arabidopsis
ESTs (Kapranov et al., 1997). Based on these observations we postulated that nodulin
Nlj 16, in addition to its speciﬁc function in nodules, may constitute a functional module
of a much larger protein(s) present in legume and nonlegume plants (Kapranov et al.,
1997). Southern analysis, using LjNOD16 as a probe, conﬁrmed that related genes were
present in the L. japonicus genome (Kapranov et al., 1997), which we set out to isolate
and characterize.

Initially, we focused on the isolation of LjNODI6-related cDNA species from L.
japonicus nodules, and characterized one class of the cDNA clones isolated in detail. The
nucleotide sequence of the longest cDNA clone of this class (pCR5h-24) was found to be
2453 bp long. A conceptual translation of this cDNA, as well as a homology search using
the BLAST algorithm, revealed that the predicted protein was composed of two distinct
amino-acid domains (Fig. 7.1). The C-terrninal domain of the protein showed signiﬁcant
similarity to the entire putative coiled-coil domain of nodulin Nle6 (87% similar and
75% identical on the amino-acid level). The N-terminal domain shared signiﬁcant level
of similarity with a number of PITPs, including the secl4p PITP from S.cerevisiae (39%
identity and 59% similarity; Bankaitis et al., 1989). Based on this observation we
designated this cDNA and its corresponding gene as the L_. japonicus PITP-like protein
gene (LjPLP-I; Fig. 7.1).

An unusual feature of the LjPLP-I cDNA was the presence of an in-frame TAG
stop codon within the PITP-like domain at the codon 222 (TAGm) relative to the ﬁrst
ATG codon. In order to rule out the possibility that the TAG222 stop codon was the
product of a cloning artifact, corresponding genomic DNA fragment, encompassing the
TAG”; codon, was PCR ampliﬁed from L. japonicus genomic DNA via PCR, and the
nucleotide sequence of four independent PCR fragments was determined. Two of these

fragments were shown to be identical to the corresponding region of the LjPLP-I cDNA

255

and to contain TAG stop codons at equivalent nucleotide positions. The two other
genomic PCR products were found to be highly similar but not identical to the LjPLP-I
cDNA and to contain a CAG codon in place of the TAGm stop codon present in LjPLP-
1, suggesting the presence of another LjPLP gene in the L. japonicus genome (LjPLP-II;
Fig. 7.1).

In order to further characterize the LjPLP-II gene and its potential protein
product, a 5.5 kb genomic DNA fragment containing the entire LjPLP-II coding region
was cloned and its DNA sequence determined (see Materials and Methods). The
corresponding region was found to be 98% identical to the LjPLP-I cDNA at the
nucleotide level, and to contain an un-interrupted open reading frame of 550 amino acids,
comprising both PITP- and Nlj16-like domains (Fig. 7.1). Taken together, these results
show that LjPLP-I and -II genes constitute a pair of genes with nearly identical DNA
sequence in the L. japonicus genome. However, while the LjPLP-II appears to encode a
PITP-like protein, the LjPLP-l gene most likely does not encode a functional protein
product due to the presence of the TAGm stop codon within its PITP-like domain(Fig.
7 . 1).

Southern blot hybridization experiment using a DNA probe corresponding to the
LjPLP-I PITP domain was used to estimate the complexity of the LjPLP gene family in
L. japonicus. Low-stringency hybridization revealed that the LjPLP-I and LjPLP-II genes

are members of a gene family (data not shown)

A novel class of PITP-like proteins is present in legumes and non-legume plants
After the initial report describing the isolation of LjNODI6 cDNA (Kapranov et
al., 1997), database search using the LjNODI6 sequence as a query identiﬁed a number
of plant genes encoding PITP-like proteins containing C-terminal domains highly similar
with the putative coiled-coil domain of Nle6 protein (data not shown). All of these

genes, with the exception of a maize EST (accession # AJOO6545), were derived from the

256

Arabidopsis genome sequencing project (e.g., accession # Z99708.1, AC007212.6,
AL023094.2, AC006841_29, T08565). In fact, the two Arabidopsis ESTs (168K8 and
110Gl6), previously shown by us to share similarity to the Nle6 protein (Kapranov et
al., 1997, also see above), were later found to represent PITP-like proteins with the
accession # AC006841_29 and T08565, respectively (data not shown). Thus, the Nlj 16-
type coiled-coil domain, found originally as a major structural component of nodulin
Nljl6, turned out to constitute an integral part of what appears to be a novel class of
PITP-like proteins. The members of this class are distinguished by the presence of the C-
terminal domain, not found in the previously characterized PITP and PITP-like proteins
from plants and other eukaryotes (Fig. 7.1; Cockcroft, 1998; Keams et al., 1997; Kearns
et al., 1998; Jouannic et al., 1998). We therefore, refer to this new class of proteins as

plant PITP-like proteins (PLPs).

Antisense LjPLP transcripts.

To identify additional members of the LjPLP gene family, a L. japonicus nodule-
speciﬁc cDNA library was screened with a DNA probe corresponding to the PITP-like
domain of the LjPLP-I cDNA (see Materials and Methods). Two additional classes of
cDNAs were identiﬁed (LjPLP-III and LjPLP-IV; Fig 7.1).

The longest cDNA corresponding to the LjPLP-III gene was found to be 2256 bp
in length and to contain a 625 amino acids long open reading frame starting with an ATG
codon at nucleotide position 77. The predicted structure of the deduced protein product
was found to be identical to the overall two-domain composition of the PLPs describe
above (Fig. 7.1). Therefore, the LjPLP-III cDNA represents another member of the
LjPLP gene family in L. japonicus.

The second class of cDNAs identiﬁed were unusual and appeared to correspond
to endogenous anti-sense transcripts derived from yet another gene (LjPLP-IV; Fig 7.1).

A search for open reading frames (ORFs) in LjPLP-IV cDNAs revealed the presence of

257

multiple short ORFs on the non-coding strands, interrupted by intron sequences (see
below). Conceptual translation of these ORFs revealed amino acid sequences showing a
high similarity to the LjPLPs, and also to PITPs from other plant species (data not
shown). The regions of similarity were conﬁned to the PITP-like domains of these
proteins (Fig. 7.1). The LjPLP-IV cDNAs varied in size (0.5-1.5 kb), in the positions of

their 5’ and 3’ ends, and contained poly(A) sequence (Fig. 7.1).

Differential expression of the LjPLP genes

Having identiﬁed several different classes of LjPLP genes, LjPLP-I, -II, -III and -
IV, it was of interest to analyze their expression patterns in different L. japonicus tissues.
The LjPLP-I mRNA, containing the TAG222 stop codon, could be conveniently
distinguished from the LjPLP-II mRNA, containing a CAG codon at the equivalent
position (CAGm), by digestion with Ec057I (recognition site for EcoS7I is CTTC_A_G).
RT-PCR ampliﬁcation was carried out using total RNA samples derived from different L.
japonicus tissues, and a pair of primers (see Materials and Methods) designed to
speciﬁcally anneal to the LjPLP-I or LjPLP-II mRNAs, but not to the LjPLP-III or -IV
mRNAs. These primers were predicted to amplify the 400bp region containing either the
TAGm or CAGm codons, respectively. Ec057l was expected to cut the RT-PCR product
derived from the LjPLP-II mRNA in the middle, generating a doublet of approximately
equal size bands of 200 bp. The aliquotes of the PCR reactions were separated on an
agarose gel side by side with the samples digested with the Ec057l endonuclease,
transferred to nitrocellulose ﬁlter and hybridized with a radiolabelled LjPLP-I cDNA
probe under stringent conditions (Fig. 7.2). RT-PCR products of the expected length
(400bp) that hybridized with the probe were present in undigested control samples
derived from L. japonicus ﬂowers, roots and nodules, but not from the shoot tissues.
Interestingly, only the ﬂower-derived RT-PCR products were digestable with Ec057I,

generating two hybridizing bands of 400bp and 200bp. These results show that while the

258

LjPLP-I mRNA is present in different L. japonicus tissues, the expression of LjPLP-II
gene appears to be limited to ﬂowers.

The transcripts hybridizing under stringent conditions to the RNA probe derived
from the LjPLP-III cDNA could be detected at a similar level in L. japonicus ﬂowers,
uninfected roots, and nodules, but were absent in the shoot tissues as determined by
northern blot analysis (data not shown).

The expression pattern of LjPLP-IV gene was investigated using radiolabelled
strand—speciﬁc RNA probes, complementary to either antisense transcripts or the
hypothetical sense transcripts of the LjPLP-IV gene. LjPLP-IV antisense transcripts were
found to be present predominantly in L. japonicus nodules (Fig. 7.3). The broad size
distribution of hybridizing antisense transcripts is in agreement with the heterogeneity in
length observed for the corresponding cDNAs (see above). Interestingly, the apparent
sense transcripts of the same gene could be detected in the L. japonicus ﬂowers and were
found to be represented by a range of closely migrating hybridizing bands (Fig. 7.3). The

reason for the size heterogeneity of the LjPLP-IV sense transcripts is presently unknown.

The LjPLP-IV antisense transcripts and the LjNODI6 mRNA are derived from the
same gene.

We speculated that the antisense LjPLP-IV transcripts and the LjNOD] 6 mRNA
might represent the products of a divergent transcription from a single L. japonicus gene
(LjPLP-IV). This notion was based on the following observations: 1. Both the LjPLP-IV
antisense and LjNOD16 transcripts are present almost exclusively in the same L.
japonicus tissue, the mature nitrogen-ﬁxing nodule. 2. The predicted protein products of
the short ORFs present on the minus strands of the antisense transcripts and the Nle6
protein represent the two domains always found together in the PLP proteins.

Two complementary approaches were undertaken in order to test this hypothesis.

First, a it phage clone, containing the LjNODI6 gene, was isolated from a L. japonicus

259

genomic DNA library. DNA sequence analysis revealed the presence of a region
encoding a PITP-like domain, with nucleotide sequence of the non-coding strand
identical to the LjPLP-IV anti-sense transcripts, immediately upstream of the LjNOD16
gene (Fig. 7.4). This observation established a close physical link between the DNA
regions encoding the PITP-like and Nlj 16 domains in the L. japonicus genome. An RT-
PCR approach was used to conﬁrm that these two domains are present in a single
transcriptional unit. For this purpose, an upstream RT-PCR primer was designed, based
on the nucleotide sequence of the genomic region, 77 bp upstream from the presumed
ATG initiation codon of the PIPT-like domain, while the downstream primer was derived
from nucleotide sequence of the 3’-UTR of the LjNOD16 mRNA (Fig. 7.4; see also
Material and Methods). RT-PCR product of 1903 bp was ampliﬁed from total RNA of L.
japonicus nodule and ﬂower tissues, and was found to contain an ORF of 482 amino
acids. A stop codon (TAA) was found to be located 36 bp upstream from a putative ATG
initiating triplet suggesting that the RT-PCR product contains the entire coding region of
the LjPLP-IV protein. The predicted amino-acid sequence of this protein was found to
encompass the PIPT-like domain, sharing ~40% identity and 60% similarity with yeast
secl4p PITP, and a C-terrninal N1j16 domain, identical to the amino acids 16-141 of
nodulin Nlj 16 (Figures 7.4 and 7.5; see also next section). The ﬁrst exon of the LjNODI6
mRNA, containing the 5' UTR and the amino acids 1-15 of Nle6 protein, was found to
be derived from the intron 10 of the LjPLP-IV gene and was absent from the mRNA

encoding the full LjPLP-IV protein (Fig. 7.4; see also next section).

The exon-intron organization of LjPLP-IV gene
The exon-intron structure of the coding region of LjPLP-IV gene was determined

based on the comparison between corresponding nucleotide sequences of the genomic

260

clone and the product of the RT-PCR ampliﬁcation. The LjPLP-IV gene contains at least
14 exons and 13 introns (Fig. 7.4). The largest intron of the gene (intron number 10),
subdivides the predicted protein structure into an N-terminal PIPT-like and a C-terminal
Nlj16 domain. Since the transcription of the antisense LjPLP-IV RNAs and the LjNOD16
mRNA initiates within the sequence of the intron 10 of the LjPLP-IV gene, we
speculated that a bi-directional nodule-speciﬁc promoter must be present in this intron.
The putative promoter region of the intron was deﬁned as a 581 bp fragment located
between the 5’ ends of the longest antisense LjPLP-IV and LjNODI6 transcripts (Fig.
7.4). Analysis of the nucleotide sequence of this putative promoter region revealed the
presence of several potential regulatory elements (Fig. 7.6). Two TATA-box-like
sequences were found approximately 40 bp upstream of the 5’ termini of the longest
antisense and LjNOD16 cDNAs. Furthermore, a number of DNA sequence motifs,
showing high similarity to nodulin gene consensus sequence, 5’-TTGT_C__T_C_LT-3’, were
present within this putative promoter sequence (Fig. 7.6; Szczyglowski et al., 1994). The
latter motifs, and especially the CTCTT core sequences, have been shown to be
indispensable for nodule infected-cell-speciﬁc expression of late nodulin genes, such as
the leghemoglobin genes (Ramlov et al., 1993; Szczyglowski et al., 1994). Similar to
TATA box-like sequences, the nodulin-box motifs are located on both strands of the
promoter sequence, coinciding with the presumed orientations of the bi-directional gene

transcription (Fig. 7.6).

A functional promoter is present within intron 10 of the L. japonicus LjPLP-IV

gene.

The presence of an active nodule-speciﬁc promoter sequence within intron 10 of

the LjPLP-IV gene was conﬁrmed using a transgenic plant approach. A 581 bp DNA

261

fragment, encompassing the predicted promoter region of the intron, was fused, in both

orientations, to the coding region of a uidA reporter gene, encoding B-glucuronidase

(GUS). Thus, the p-For construct contained the GUS coding region fused to the
LjNODI6 side of the intron-derived fragment, while in the p-Rev construct the position
of this putative promoter region was reversed.

GUS staining of hand-cut nodule sections revealed that the intron fragment
directed GUS expression only to the central, infected zone, of the nodules (Fig. 7.7).
Other plant tissues, including L. corniculatus roots, leaves and ﬂowers, showed no
cytological staining for GUS activity (data not shown). The intron sequence was found to
be capable of activating the reporter gene expression in an orientation-independent
manner (see Fig. 7.7). However, in contrast with the p-For construct reporter gene
construct, the p-Rev construct showed a strong histochemical staining also in the nodule
vascular bundles (Fig. 7.7). The promoterless uia'A construct, used as a negative control
showed no detectable staining in the central zone of the nodules (Fig. 7.7). However, a
relatively weak staining in nodule vascular bundles could be detected in some of these

control transgenic lines (data not shown).

The Nlj16 domain contains a functional plasma membrane targeting motif.
A number of proteins are known to be attached to cellular membranes via lipid
modiﬁcation of their C-terrninal cysteine residues (reviewed in Zhang and Casey, 1996;

Rodriguez-Concepcion et al., 1999). In addition, a polybasic region, present at the C-

262

terminus of K-Ras(B) protein, has been shown to be required for the localization of this
prenylated protein to the inner surface of the plasma membrane (Hancock et al., 1991).
Since both of these motifs, namely a polybasic domain of six lysine residues and two
terminal cysteine residues, were found to be present at the C-terminal end of the Nle6
protein we hypothesized that they may also constitute a plasma membrane targeting
signal (Figures 7.6 and 7.9). In order to test this hypothesis, cell extracts derived from L.
japonicus nodules were fractionated by differential centrifugation. The majority of Nlj16
protein was found to be associated with membrane-enriched fractions obtained after
differential centrifugation at 10,000g and 100,000g (data not shown). N1j16 protein could
not be detected in the supernatant fraction collected after centrifugation at 100,000g,
consistent with a predicted localization of Nle6 to a cellular membrane compartment. In
a parallel experiment, we failed to detect Nlj 16 protein in symbiosome- or symbiosome
membrane-enriched fractions derived from L. japonicus nodule extracts, suggesting that
this protein may be anchored to a different membrane compartment in the infected cells
of L. japonicus nodules (data not shown).

In order to investigate the subcellular localization of Nle6 and LjPLP-IV proteins
in more details, the mGFP5-Nlj 16 and mGFP5-LjPLP-IV fusion proteins were transiently
expressed in onion epidermal cells. Control cells, expressing mGFP5 alone, displayed a
broad distribution of mGFP5-ﬂuorescence in the cell cytoplasm, nucleus, and
transvacuolar strands (Fig. 7.8A). In contrast, the C-terminal N1j16 extension was found
to be able to target the mGFP5-Nlj16 chimeric protein to the periphery of the cells only,

indicating localization of the Nle6 protein to the plasma membrane (Fig. 7.88).

263

Furthermore, a small subset of transformed onion cells expressing the mGFP5-Nljl6
fusion protein was found to undergo plasmolysis. In these cells the mGFP5 ﬂuorescence
was always detected at the periphery of the protoplasts, and was not associated with cell
walls or vacuolar membranes (Fig. 7.8D & E). The mGFP5-LjPLP-IV fusion protein was
also found to be localized to the plasma membrane in the manner identical to the mGFP5-
Nlj 16 fusion (data not shown).

The contributions of the polybasic domain and the two C-terminal cystein
residues of the Nle6 protein to targeting the chimeric mGFP5-Nljl6 protein to the

plasma membrane were also evaluated. A chimeric protein deprived of the two C-

terminal cysteines (mGFP5-Nljl6ACC) was localized non-speciﬁcally to all accessible

intracellular compartments in a manner similar to the mGFP5 alone (Fig. 7.8C). Although
required for the proper localization of the chimeric protein, the two cysteine residues
were not sufﬁcient for targeting when fused to the C-terrninal end of the mGFP5 (data
not shown). Instead the localizion of the mGF P5 to the plasma membrane required both,
the presence of the polybasic region and the two cysteine residues (mGFP5+ amino acids
KKKQKKKTFFCC; data not shown).

As mentioned above, the Nle6 domain of LjPLP-IV is highly related to the C-
terminal domains of the other PLP proteins from L. japonicus and other plant species
(Fig. 7.9). The most notable differences include the variations in the nature of the amino
acids present at the very C-terrnini of these proteins, e.g. the CC residues of LjPLP-IV are

substituted to the CW residues in LjPLP-III or residues WA in the LjPLP-II (Fig. 7.9). It

264

is therefore possible that the Nlj16-like domains of the PLP proteins may have different
requirements for the plasma mebrane localization. Thus, we evaluated the contribution of
the Nljl6-like domain of the LjPLP-III protein (Nlj16-III) to the plasma membrane
localization of this protein in the transient expression experiments. As expected, the
mGFP5-N1jl6—III fusion was localized to the plasma mebrane in a manner identical to the
mGFP5-N1jl6 fusion (Fig. 7.8B, data not shown). Also, the deletion of the last two

residues (ACW) from the C-terminus of the Nlj l6-III domain caused de-localization of
the mGFP5-Nljl6-IIIACW fusion from plasma membrane to the interior of the onion

cells, in a manner identical to the mGFP5-Nlj16ACC fusion (Fig. 7.8C, data not shown).

However, the C-terminal basic domain of the Nlj16-III region (amino acids
RQAEAKLRKKRFCW) was not sufﬁcient to localize the heterologous protein mGFP5
to the plasma membrane (data not shown), in contrast to the analogous region derived

from Nle6 domain (see above).

Functional characterization of the LjPLP-IV protein

Complementation of the growth defects caused by the mutant yeast sec14
alleles has been a powerful tool to isolate and functionally characterize novel
heterologous PITPs (Keams et al., 1998; Skinner et al., 1993; Tanaka and Hosaka, 1994).
Expression of the LjPLP-IV protein from the yeast PGK-promoter failed to rescue a
temperature-sensitive phenotype of the CTY1079 yeast strain, carrying sec14-I"’ allele
(Phillips et al., 1999), when grown under non-permissive condition (37°C; construct

pGK-IV; Fig. 7.10). Based on our targeting experiments, we postulated that the observed

265

lack of functional complementation could be due to sequestration of intracellular pool of
LjPLP-IV protein to the yeast cell plasma membrane, thereby precluding this protein
from substituting for the functions of predominantly cytosolic yeast secl4p protein
(Bankaitis et al., 1989). To test this hypothesis, a construct (pGK-IVACC) was
engineered to express a truncated LjPLP-IVACC protein lacking the two C-terminal
cysteine residues required for the plasma membrane localization in the plant cells (see
above). Expression of this mutant protein could in fact rescue the grth of the CTY1079
yeast strain at the non-permissive temperature (Fig. 7.10). A truncated form of a related

protein, LjPLP-III, lacking the entire Nlj16-like domain (LjPLP-IIIA) could also

complement the temperature sensitive growth of strain CTY1079 (construct pGK-IIIA;

Fig. 7.10). This result suggests that the Nljl6-like domain is not required for functioning
of the LjPLP proteins in yeast. In addition to the secI4-1"’ allele, the CTY1079 strain also
contains a deletion for the phospholipase D (PLD) gene (Phillips et al., 1999), therefore
the PLD activity is not required for the observed rescue of the temperature-sensitive

phenotype by the LjPLPs.

266

 

7.4. DISCUSSION

The biological functions of fungal and animal phosphatidylinositol transfer
proteins (PITP) and PIPT-like proteins have been a subject of intensive investigations for
more then a decade (reviewed in Cockcroft, 1998; Keams et al., 1998). On the other
hand, plant PlTPs, are only beginning to be characterized (Keams et al., 1998; Jouannic
et al., 1998). We report here on identiﬁcation and characterization of a novel class of
plant PITP-like proteins, (PLPs). We show that these proteins are present in both legume
and non-legume plants, and describe a set of intriguing observations that suggest that PLP
proteins may be connected to the development and/or functioning of L. japonicus
nodules.

Generally, little is known about the role of plant phospholipid metabolism in the
legume-rhizobium symbiosis. Hong and Verma, (1994) have been the only authors to
touch on this topic when they describe the expression patterns and biochemical properties
of root- and nodule-speciﬁc isoforms of soybean PI 3-kinase. PI 3-kinase carries out the
phosphorylation of P1P», PI, and PI4P, giving rise to 3-phosphorylated inositol lipids that
serve as secondry cellular messengers. Interestingly, the expression of the PI 3-kinase
gene was shown to be associated with the proliferation of peribacteroid membranes in
infected cells of the nodules, suggesting that lipid metabolism may play a role in this
process (Hong and Verma, 1994).

We previously identiﬁed a highly abundant mRNA species from L. japonicus
nodules expressed during the developmental transition between nodule ontogeny and the

commencement of nitrogen ﬁxation, and its corresponding cDNA (LjNOD! 6; Kapranov

267

et al., 1997; Szczyglowski et al., 1997). We show here that the LjNOD16 sequence,
encoding late nodulin Nlj l 6, constitutes a C-terminal portion of a much larger
transcriptional unit, namely the L. japonicus LjPLP-IV gene. Furthermore, we
demonstrate that LjPLP-IV is a member of a small gene family in L. japonicus, encoding
a novel class of phosphatidylinositol transfer-like proteins (LjPLPs).

With exception of the LjPLP-l gene, which contains the TAG stop codon within
its presumed coding region, the other members of this family isolated thus far (LjPLP 11-
IV) are predicted to encode PLP proteins that share a two-domain protein structures. The
N-terminal portions of these proteins show a signiﬁcant level of similarity with
previously described PlTPs from yeast, Arabidopsis, and soybean, whereas their C-
terrninal regions are composed of either NLj16, or highly related amino acid sequences.

The LjPLP-III or IV proteins, when deprived of their plasma membrane targeting
signals (see below), are able to complement the temperature sensitive phenotype of yeast
secl4-1” mutant. These proteins also contain the conserved amino-acid residues (K85,
E227, and K259), known to be necessary for PI binding by the yeast secl4p protein (Sha
et al., 1998; Phillips et al., 1999). Furthermore, they are unlike other previously
characterized plant PITPs, since they do not require a functional yeast PLD gene for their
complementation activity (Phillips et al., 1999). Therefore, it appears that LjPLPs
function as phosphatidylinositol transfer protein, in spite of the fact that evidence for their
enzymatic function, binding and transport properties, yet remains to be established. We
have as yet been unable to isolate sufﬁcient amount of these protein for this purpose, due

to their strong toxicity, which prevents their production in E. coli.

268

The presence of Nle6 or Nljl6-1ike C-terminal domains of LjPLPs appears to
constitute a hallmark of this new class of plant PITP-like proteins. With the exception of
the retinal degeneration-B (rng) group of PITPs, which in Dr050phila have been shown
to represent integral membrane proteins, and to contain carboxy-terminal region
encompassing six potential transmembrane domains (Vihtelic at el., 1991, 1993), all
other previously characterized PITPs are cytosolic, and lack C-terminal extensions
(Cockcroft, 1998). We demonstrate here that Nle6 and Nlj l6-III domains of LjPLP-IV
and LjPLP-III proteins, respectively, contain functional targeting signals, sufﬁcient for
the localization of chimeric GFP-N1jl6 or GFP-Nljl6-III proteins to the plasma
membrane of onion epidermal cells. These results are consistent with the predicted
localization of Nle6 to cellular membrane compartments in nodules, and suggest a role
of these C-terminal amino-acid sequences in localization of LjPLPs to the plasma
membrane.

It is noteworthy that the Nlj16-III domain contains the most divergent amino-acid
sequence when compared with the equivalent regions of the other LjPLP proteins. Both
the C-terminal polybasic region and two terminal cystein residues (amino-acids 471-482)
of the Nlj16 domain are required and sufﬁcient for speciﬁc targeting of GP P to the
plasma membrane, while the equivalent region of the Nlj16-III domain (amino-acids 612-
625) is necessary, but not sufﬁcient. These results indicate that in addition to the
polybasic region and the CW motif, the Nlj l6-III domain may contain other, as yet not

identiﬁed signals that are required for plasma membrane targeting activity.

269

The L. japonicus LjPLP-IV gene appears to be the most unusual member of this
novel family of phospatidylinositol-transfer-like proteins since it encodes at least three
different mRNA species, expressed in L. japonicus nodules and/or ﬂowers. A full-length
1.9 kb sense LjPLP-IV transcript, encoding the LjPLP-IV protein, accumulates
predominantly in L. japonicus ﬂowers. The same transcript is also present in nodules,
albeit at a signiﬁcantly diminished level. In contrast two distinct classes of mRNA
species derived from the same LjPLP-IV gene, anti-sense LjPLP-IV transcripts and
LjNOD] 6, accumulate to relatively high levels in L. japonicus nodules. These transcripts
originate as a result of the tissue-speciﬁc activity of an internal bi-directional promoter,
localized within the largest intron (intron No. 10) of the LjPLP-IV gene. By using a
transgenic plant approach we were able to show that this intron-derived promoter
sequence is sufﬁcient to direct a high level of GUS reporter gene activity speciﬁcally to
the central, infected cell containing, zone of L. corniculatus nodules. The latter
observation is entirely consistent with our previous in-situ hybridization results, showing
a localization of LjNODI6 mRNA to the infected cell of L. japonicus nodules (Kapranov
et al., 1996). The nodule-speciﬁc nature of this promoter sequence is further indicated by
the presence of the nodulin gene promoter consensus-like sequence, CTCTT, which has
been shown to be essential for infected-cell speciﬁc expression of leghemoglobin (lb)
genes in nodules (Ramlov et al., 1993; Szczyglowski et al., 1994). Interestingly, the
CTCTT motifs, as well as putative TATA box sequences, are present on both strands of

the intron-localized promoter region of the LjPLP-IV gene. We postulate, therefore, that

270

these putative regulatory elements determine the bi-directional and tissue-speciﬁc nature
of the intron-bom promoter.

The biological function of LjPLP-proteins remains to be determined. The data
presented here strongly suggest that Lj PLP proteins are plasma membrane-localized. It is
likely that these proteins bind and present phospholipids to plasma membrane-associated
proteins or protein complexes, containing enzymes of downstream phospholipid
metabolism pathways and/or components of phospholipid-mediated signaling. In this
context, it is tempting to speculate that if the recruitment of LjPLPs to a speciﬁc location
on plasma membrane indeed occurs, it could be mediated, at least in part, by a

mechanism involving the a—helical regions of Nlj 16 or Nlj16-like domains, since they are

predicted to have a high propensity to form coil-coiled structures, and thus are likely to
be involved in protein-protein interactions (Kapranov et al., 1997).

Interestingly, in the infected cells of nodules, the Nle6 domain of LjPLP-IV gene
is expressed from the intron-bome bi—directional promoter as an independent entity,
namely nodulin Nlj l 6. Since nodulin Nle6 is targeted to plasma membrane, we speculate
that it may occupy (a) speciﬁc location(s) therein, normally destined to actively interact
with LjPLP proteins, thus preventing the latter from exerting their biological function(s)
in this particular subcellular compartment.

The biological role of LjPLPs in nodules remains elusive. However, it has been
suggested that soybean PITP (Sshlp) functions as a component of a stress response

pathway, protecting adult plants from osmotic insults (Keams et al., 1998). Hyperosmotic

271

conditions have been shown to induce phosphorylation of Sshlp, and its concomitant
mobilization from the plasma membrane to a cytosolic location. In its new cellular
location, the cytoplasm, the phosphorylated form of Sshl, called Ssh1p*, has been
proposed to assume an active role in the plant osmoprotective response (Keams et al.,
1998). In this context, it is interesting to consider a possible role of LjPLPs in
mechanisms governing osmoregulation in L. japonicus nodule tissues. The infected cells
of the nodules contain high concentration of sugars, amino-acids, and organic acids that
results in four- to ﬁve-fold higher osmoticum in these cells (Verma et al., 1978).
Consequently, the active employment of osmoregulation may be necessary for optimum
functioning of this tissue (Delauney and Verma, 1996), and the LjPLP proteins can
potentially be involved in this process. However, the presence of anti-sense LjPLP-IV
transcripts in nodules argues against this hypothesis. Rather then being actively involved
in osmoprotection, the expression of LjPLPs genes in nodules seems to be vigorously
inactivated. Whether such an inactivation would facilitate, in any way, the
osmoprotective response in nodules remains to be established.

There may be yet another explanation for the detection of sense and anti-sense
LjPLP-IV transcripts in L. japonicus nodules. Assuming that both of those transcripts are
present in infected cells of the nodules, this could lead to the formation of double-
stranded RNAs species (dsRNA). It is tempting to speculate that such dsRNA molecules,
in addition to their presumed inhibitory role in LjPLP gene expression in nodule, could
serve as signaling molecules involved in long distance coordination of plant

developmental processes related to symbiotic nitrogen ﬁxation (e.g. signaling between

272

 

 

nodules and ﬂowers), in a manner similar to a phenomenon of posttranscriptional gene
silencing (PTGS; Waterhouse et al., 1998). A more detailed comparative analysis of
LjPLP-IV gene expression in nodules and ﬂowers, and at the different developmental

stages, will be required to test this hypothesis.

273

7.5. MATERIALS AND METHODS

Plant material and growth conditions.

L. japonicus ecotype B-129-S9 Gifu plants were germinated and grown as
described previously (Kapranov et al., 1997; Szczyglowski et al., 1997). Nodules, leaves,
and stems of L. japonicus plants inoculated with Mezorhizobium loti strain NZP2235
were harvested 35 days after inoculation (dai), while control uninoculated roots were
collected from axenically grown L. japonicus plants of the same age. L. japonicus ﬂowers
were obtained from 2 to 3 month-old plants.

Transgenic Lotus corniculatus plants were inoculated with M loti strain 2037
(Jarvis et al., 1982) and subsequently grown in a 6:1 mixture of verrniculate and sand

under controlled environmental conditions (18-/6-h day/night cycle, 250 pE s'l m‘z,

22/18°C day/night temperature). B&D solution (Broughton and Dilworth, 1971),
supplemented with 1 mM KNO3, was used to water these plants. Fully mature nodules,
leaves and root segments were harvested from transgenic plants 42-45 dai, and used

directly for histochemical analyses.

Screening of L. japonicus genomic DNA- and nodule-speciﬁc cDN A libraries.
A L. japonicus genomic DNA library, and a cDNA library from mature nodules
of the same plant species were kindly provided by Dr. Jens Stougaard (Aarhus

University, Denmark). The genomic library was constructed in the FIX II A vector

(Stratagene), while the cDNA library was constructed with oligo(dT) primers in the A-

274

 

UniZAP vector (Strategene). Filters carrying the libraries were pre-hybridized and
hybridized in a buffer containing 0.5 M sodium phosphate pH 7.2, 7% SDS, and 1%
BSA, at 65°C. The ﬁlters were washed either at low-stringency (last wash in 2 X SSC,
0.1% SDS at 65°C for 15 minutes), or high-stringency (last wash in 0.1 X SSC, 0.1%
SDS at 65°C for 15 minutes) conditions, as speciﬁed.

To isolate LjNOD16-related genes, L. japonicus genomic and cDNA libraries
were initially screened under low-stringency conditions, using a 530 bp EcoRI-Hina'll
fragment of the LjNODI6 cDNA, representing the entire coding region ofN1j16, as a
probe. Hybridizing phage plaques were also hybridized with a gene-speciﬁc probe
corresponding to the 3’UTR of LjNODI6 mRNA (the 370 bp HindII-Xhol fragment of
LjNODI6 cDNA) under high-stringency conditions. Plaques hybridizing to the probe
derived from the coding region, but not to the gene-speciﬁc probe, were assumed to
represent L. japonicus genes related to, but not identical, to LjNODI6. In addition, to
isolate LjPLP-III and -IV cDNAs, the L. japonicus cDNA library was screened with the
PIPT-like domain-containing fragment of the LjPLP-I cDNA (base pairs 44-1282), under

low-stringency conditions.

Nucleic Acid Isolation and Northern Analyses

Genomic DNA, and total RNA from different L. japonicus tissues were isolated
as described by Kapranov et al., (1997) and Szczyglowski et al., (1997). Northern blot
analyses were performed essentially as described (Kapranov et al., 1997; Szczyglowski et

al., 1997). For hybridization with strand-speciﬁc RNA probes, the ﬁlters were pre-

275

hybridized in 100 mM potassium phosphate buffer pH 6.8, 5X SSC, 1X Denhardt's, 0.1%

SDS, 100 pg/ml denatured salmon sperm DNA, at 50°C, for 4 hours. Hybridization was

carried out in 70 mM potassium phosphate buffer pH 6.8, 3.6 X SSC, 0.7 X Denhardt's,
7.0 % dextran sulphate, 71 pg/ml denatured salmon sperm DNA, and 50% deionized
formamide, at 65°C. The ﬁlters were washed for 15 minutes in 2X SSC, 0.1% SDS, 15
minutes in 1X SSC, 0.1% SDS, and 15 minutes in 0.1X SSC, 0.1% SDS, at 65°C.
Radiolabelled RNA probes were prepared as followed. Template DNA (0.5-1 pg),
was linearized and incubated in a buffer containing 40 mM Tris pH 7.5, 8 mM MgC12, 2
mM spermidine, 25 mM NaCl, 10 mM DTT, 40 Units Placental RNAse Inhibitor (BMB),

0.5 mM ATP, 0.5 mM GTP, 0.5 mM CTP, 15 pM UTP, 50 pCi [or-32P]UTP, and 20 units
of T3 or T7 RNA polymerase (BMB), in a total volume of 20 pL. The labeling reactions

were performed for 1 hr, at 37°C. The DNA template was removed from the reaction mix
by adding 10 units of RNAse-free DNAse I (BMB), and incubation at 37°C, for 15 min.
Radiolabelled RNA probes were puriﬁed on Bio-Spin 6 Chromatography Columns,

following the manufacturer’s instructions (Bio-Rad).

PCR amplifications and DNA sequencing.

The 400 bp genomic PCR fragments encompassing the regions of the
LjPLP-I or -11 genes containing TAG or CAG codons, respectively, were PCR-ampliﬁed
from the L. japonicus genomic DNA using a pair of common primers, DB588 and DB544

(see Table 1). PCR reactions were performed for 30 cycles with the following cycle

276

 

proﬁle: 1 min. DNA denaturation step at 94°C, l min annealing step at 55 °C, and l min.
extension step at 72 °C.

A 5.5 kb genomic DNA fragment corresponding to the entire coding region of the
LjPLP class II gene was PCR-ampliﬁed from the corresponding phage lysate using
forward (D8562) and reverse (D8561) primers (see Table I). These primers were
designed based on the nucleotide sequences of the 5' and 3' UTRs of LjPLP-I cDNA,
respectively. The PCR was carried out for 30 cycles using Pfu DNA polymerase
(Promega) under the following cycling conditions: 1 min. denaturation step at 94°C, 1
min. annealing step at 55°C, and 7 min. extension step at 72°C. The PCR product was

cloned and sequenced.

RT-PCR procedure.

Total RNA (5 pg) from different L. japonicus tissues was denatured for 10
minutes at 65°C, and reverse transcribed for 1 hr at 42°C, in a reaction mix containing 50
mM Tris-HCL pH 8.3, 75 mM KCl, 3 mM MgC12, 10 mM DTT, 1 mM of each dNTPs,
40 units of placental RNAse inhibitor (BMB), 100 ng of D8544 primer, and 200 units of
Superscript II reverse transcriptase (Gibco), in a total volume of 20 pL. A subsequent
PCR ampliﬁcation step was performed using 4 pL of the original cDNA reaction mix,

and 40ng of forward (D8588) and reverse (D8544) primers, in a total volume of 25 pL.

The primers (D8588, D8544; see Table I) were designed to amplify a 400bp long DNA

fragments encompassing a portion of the PIPT-like domain of both the LjPLP-I and -II

277

 

genes. The PCR reactions were performed essentially as described above, except that 20
PCR ampliﬁcation cycles were used. Aliquots of each PCR reaction were digested with
Ec057l endonuclease, and separated on 1% agarose gel along with the same amount of
undigested DNA. The resulting DNA fragments were transferred to a nylon membrane

and hybridized to [a32P]-radiolabelled LjPLP-I cDNA probe under stringent condition

described above. The relative intensities of hybridizing bands were evaluated by using
Phospholmager (Molecular Dynamics).

The RT-PCR approach was also used to amplify a full copy of the LjPLP-IV
mRNA. The forward N16-4 primer was designed based on the nucleotide sequence of a
genomic region position 77bp upstream from the putative ATG codon of the LjPLP-IV
protein. The reverse D8641 primer was designed to be complementary to the nucleotide
sequence of the 3'UTR of LjNODI6 cDNA (see Table 1). RT-PCR was carried out using
total RNA derived from L. japonicus ﬂowers and nodules, essentially as described above,
except that two consecutive rounds of PCR ampliﬁcation were performed. The 1.9 kb
PCR product was cloned into the eyast YePlac195PGK expression vector (see below),

and its nucleotide sequence was determined.

Chimeric gene constructs and generation of transgenic L. corniculatus plants.
Chimeric gene constructs were prepared using standard molecular
techniques and the exact details of their construction are available from the authors upon

request. Brieﬂy, the 5 81bp long DNA fragment derived from intron 10 of LjPLP-IV gene

278

was PCR ampliﬁed and cloned in both orientations into the unique BamHI restriction site
of p81 101 (Clontech) derived binary vector. This resulted in the construction of the p-For
and p-Rev binary vectors, carrying the intron sequence in forward (p-F or), or reverse (p-
Rev) orientation, with respect to the direction of the GUS coding region.

The binary vectors described above were independently transferred into
Agrobacterium rhizogenes A4 (Tempe and Casse-Delbart, 1989), by using the freeze-
thaw method of Hofgen and Willmitzer, (1988). Transgenic Lotus corniculatus cv. Rodeo
plants were generated as previously described (Szabados etal., 1990, Szczyglowski et al.,
1994). GUS activity in the nodule hand sections and other L. corniculatus tissues was
analysed histochemically (Jefferson et al., 1987; Szczyglowski et al., 1994), using
condition described by Malamy and Benfey (1997). Stained tissues were examined using
a Wild Heerburgg M420 stereoscope. The images of stained nodule sections were
generated using Kodak DC 120 digital camera and processed using Adobe Photoshop 5.02

software.

Subcellular localization of mGFP5-Nlj16 fusion in onion epidermal cells

In order to generate an mGFP5-Nlj16 chimeric protein, the BglII-Hina'll fragment
(470 bp), encompassing almost the entire coding region (except for the ﬁrst two
aminoacids) of LjNOD] 6 cDNA, was isolated and blunt ended using Klenow fragment of
Polymerase I. The fragment was then fused, in-frame, to the C-terminal end of the coding
region of mGFP5 (Siemering et al., 1996) in vector pAVA393. The latter is based on the

vector pAVA319 (von Amim et al., 1998), in which the g/p cDNA was replaced with the

279

cDNA encoding mGF P5, and kindly provided by Dr. von Amim, the University of

Tennessee, Knoxville. The mGFP5-Nljl6ACC fusion was constructed in a similar

manner as mGFP5-Nljl6, except that a 464 bp BglII-Bbsl fragment of LjNOD16 cDNA,
deprived of the two C-terminal codons encoding cystein residues, was used.

To generate mGFP5 protein with either two C-terminal cysteins residues
(mGFP5+CC) or KKKQKKKTFFCC sequence (mGFP5+KCC), the corresponding
nucleotide sequences of the LjNODI6 cDNA were PCR-ampliﬁed using two sets of
speciﬁc primer pairs, NODl6-l/NOD16-5 and NODl6-2/NODl6-5, respectively (see
Table I). The resulting PCR products were digested with BamHI-Xbal and cloned into
vector pAVA393. The Nlj 16-1ike domain of the LjPLP-III protein (amino acids 513-612)
and its derivatives were used to construct the fusions mGFP5-Nlj16(III), mGFP5-
N1j16(lII)ACW and mGFP5-basic domain of Nlj16(III), in a manner similar to the
constructs described above.

Plasmid DNA from each constructs was delivered into the onion epidermal cells
using 1.6 pM gold particles and the BIOLISTIC PDS-IOOO gene transformation system
(Dupont), essentially as described by Varagona et al., (1992). The bombarded onion
epidermal explants were incubated for 18-24 hrs on light, on a solid MS media (Gibco)

containing 30 g/L sucrose and 180 mg/L KHZPO4. Fluorescence of mGFP5 protein

fusions was analyzed using a Zeiss Axiophot epiﬂuorescent microscope under a ﬁlter

with following parameters: emission at 470i20 nm, beam splitter at 510 nm and

excitation at 540i20 nm. Images were obtained using a Kodak DC120 digital camera and

280

Adobe Photoshop 5.02 software. Each transfection experiment was repeated at least two

times.

Yeast complementation experiments

The yeast strain CTY1079 (MA Ta ura3-52, lys2-801, Ahis3-200, secI4-1”

Asp014::HIS3; Phillips et al., 1999) was used for the complementation experiments. The

YePlacl95PGK yeast expression vector was created by cloning a 2 kb Hindlll fragment
containing the yeast PGK promoter cassette from pHVX2 (Volschenk et al., 1997) into
the Hindlll site of the YePlac195 plasmid (Gietz and Sugino, 1988). A 1.9 kb RT-PCR
product containing the entire LjPLP-IV coding region (see above) was cloned into
YePlacl95PGK under the control of the PGK promoter to generate construct pGK-IV. To
remove the two cysteine residues from the C-terminus of LjPLP-IV protein, the 1.5 kb
region of the LjPLP-IV cDNA was ampliﬁed using primers SFX and SRX. The PCR
product was inserted into the YePlacl95PGK vector, downstream of the PGK promoter

to generate construct pGK-IVACC. To express a truncated LjPLP-111A protein, construct

pGK-IIIA was generated as follows. A 1.6 kb DNA fragment was PCR ampliﬁed from

the LjPLP-III cDNA with the primer pair D8637-DB638 and inserted into vector
YePlacl95PGK downstream of the PGK promoter.

Plasmids YePlacl95PGK, pGK-IV, pGK-IVACC or pGK-IIIA were introduced

into the yeast strains following the procedure of Elble, (1992). CTY1079 transformants

were initially selected at 28°C on deﬁned yeast media lacking uracil. Complementation of

281

 

the temperature-sensitive (ts) phenotype of the sec14-1" allele in the strain CTY1079 was

 

evaluated after 4 days of growth on selective media at 37°C. Five independent CTY1079

transformants containing each plasmid were evaluated.

 

282

 

Table 7.1. The nucleotide sequences of the primers used during different ampliﬁcation

 

 

 

 

 

 

 

procedures.

D8568 GAACTTCAACAAACATGCCAG
D8544 CAAGCAATTTGCTTTGATAC
D8561 GGGAAGTAGCATTTGGAAAGC
D8562 CATATTAAAATTCAGCAGAAGC
D8588 GTGTACATTGAGAATATAGGC
D8640 GACGACCCGTGTACATAGAGC
D8641 CTTGTCACATAAGCAAAAGG

 

N l 6-5'-4 GGGAGTGCTTTTGTTCTCTGC

 

NOD 1 6-1 GAAGAAGACGGGATCCTGCTGTTAAGTG

 

NOD 1 6-2 TCAGCTTATGGATCCAAAAAGAAAC

 

NOD 1 6-5 CACACTTATGTCTAGAGGGCAC I I I G

 

 

 

SRX ATTTCGCACTCGAGATTAGAAAAACGTC
SFX CTTTTGTTCTCGAGCTCGCATGAT
D8637 CATGATTAGATGTTATTGCAGG

 

 

D8638 CAACGCTGCACAAATTCTAGG

 

 

283

 

Figure 7.1. Schematic diagrams of LjPLP cDNAs. Boxes represent the
coding regions, while lines correspond to the 5' and 3' UTRs of the cDNAs. The
stippled boxes represent either PITP or PITP-like domains, the dotted boxes
denotes the N1j16-like domains, and the white boxes correspond to the regions
with no apparent sequence similarity between different cDNAs. The positions of
the stop TAGzzz codon in LjPLP-l, and the corresponding CAGm codon in
LjPLP-II genes, are shown. The asterisk indicates the fact that the LjPLP-II
cDNA sequence was deduced from the nucleotide sequence of the
corresponding 1.. japonicus genomic region. The short open reading frames
(ORFs), present in the LjPLP-IV antisense transcripts, are represented by the
stippled boxes, and the intron sequences are represented by lines. Arrows

indicate the direction of ORFs.

 

 

LjNODI6

S. cerevisiae

secl4p PITP WW

     

     

PITP-like domain ”£1,323“
LjPLP-1 IWW- - (A)n
PITP-like domain ”$1,133"
LjPLP-11* IWW- 1-
CAG
LjPLP-III PITP-like domain Ndiimiiiie

 

LjPLP-IV ORF ORF ORF
antisense (A)
transcripts n

285

 

- Eco 57I + Eco 57I

 

f5

8
a:

.....- o » ~-
200bp— _'

Figure 7.2. Tissue-speciﬁc expression of the LjPLP-I and LjPLP-II genes. The 400 bp

Nodules
Flowers
Shoots

Nodules

Flowers
Shoots
Roots

fragments, encompassing either UAGm stop codon-, or CAGm triplet-containing
regions of LjPLP-I or LjPLP-II mRNAs, respectively, were ampliﬁed using RT-PCR
approach. An aliquot of each PCR reaction was digested with Eco 57I (+Eco 57I),
separated on agarose gel, along with the undigested control sample (-Eco 571), and
hybridized to the radiolabelled LjPLP-I cDNA probe. The 400 and 200 bp fragments of

the digested samples correspond to LjPLP-I and LjPLP-II transcript, respectively.

286

SENSE ANTI-SENSE

In W m m
bami’ Stan-‘3
333% 383%
.2.:°° 2.:°°
mmmz Lair/2&2

— 28$
— 18$

 

Figure 7.3. Tissue-speciﬁc expression of the LjPLP-IV transcripts. Ten
micrograms of total RNA derived from L. japonicus ﬂowers, leaves, uninoculated
roots, and fully mature nodules, was separated on the formaldehyde-agarose gel
and hybridized with radiolabeled, strand-speciﬁc, RNA probes, complementary to

the presumed sense or antisense transcripts of the LjPLP-IV gene. The positions

of 288 and 185 ribosomal RNAs are indicated.

287

 

Anti-sense transcripts

e LjNODI6 mRNA

 

 

 

 

 

 

LjPLP-IV gene

 

 

Nlj16 domain
, PITP-like domain ..
LjPLP-IV mRNA 1W I -. A)“
LjNODI6 mRNA —[D——-(A)n

Nlj16

Figure 7.4. Schematic diagrams of LjPLP-IV genomic region, and LjPLP-IV
sense- and LjNODI6 transcripts. Exons are numbered and represented by boxes,
introns are symbolized by the thin lines, whereas thick lines indicate the 5' and 3’
UTRs. Differently shaded boxes correspond to either PIPT-like- or N1j16
domains, as describe in the legend to Figure 7.1. The solid box represents the ﬁrst
exon of the LjNOD16 transcript. The position of different cDNA clones,
corresponding to either LjPLP-IV antisense- or LjNOD16 transcripts, is indicated
by arrows above the diagram of the LjPLP-IV gene. The localization of the bi-
directional promoter (P) within the intron 10 is indicated. The positions of primers

used to amplify the LjPLP-IV mRNA are indicated by arrowheads.

288

 

Fig. 7.5. Amino acids sequence alignment between L. japonicus LjPLP-IV,
nodulin Nljl6, Arabidopsis AtPLP (Ace. No. 2997081), and yeast secl4p
proteins. The alignment was generated using the PileUp function of the Genetics
Computer Group package (GCG, Madison, Wisconsin) and the Squu software
version 1.0.1. Similar residues are boxed, while the identical aminoacids are
boxed and shaded. Dashes represent gaps in the sequences, and asterisks indicate
the conserved amino-acid residues required for PI binding activity of yeast secl4p

protein (Sha et al., 1998).

289

 

 

LjPLP-IV
AtPLP
secl4p

LjPLP-IV
AtPLP
secl4p

LjPLP-IV
AtPLP
secl4p

LjPLP-IV
AtPLP
secl4p

LjPLP-IV
AtPLP
secl4p

LjPLP-IV
AtPLP
secl4p

LjPLP-IV
AtPLP
secl4p

LjPLP-IV
AtPLP
secl4p

LjPLP-IV
AtPLP
secl4p

LjPLP-IV
AtPLP
Nljl6

LjPLP-IV
AtPLP
Nlj16

LjPLP-IV
AtPLP
Nlj16

LjPLP-IV
AtPLP
Nljl6

LjPLP-IV
AtPLP
N1j16

LjPLP-IV
AtPLP
Nlj16

[ElsaLlsse'rD-AI as
MKRFFSSLFCYLLVLDVVLC DAB KPRMG -;::--- s
I
unkﬁu LTRK RRSSKVMSVB-IBDVHDABELKAVE FRQ
KN YEumxnl 5 av I DA xav a
TQQEKEFLESYPQ cppn PGTPGNL E A x

 

 

 

. , * I
ALISDDLLP KHDDYHMMLRFLKARKFEIDKSKQMWSDML
LILD D MLRFLKARKFDLE TKQM
- E GFIB L -ST FE CE

 

 

KWRKBFGADTIV EFEFKEIDEVLKYYPQGHHGVDKEGRP
RWRKEPGADTVM EPD'LE DIV KYYPQ .Ht DKEGRP
2 ';n - .I'HIDZVPL -;' '9 YaKT-;I :-

 

VYIB QVDATKLMQVTTMDRYI YH’K FEKTFDLK A
V? E LVD L VTT DR V M PB N P
LHE NK M NL W SV LP

 

 

acsIxaxxnrDQSTTILDVUGVGLKSFNK RELVTRIEKC
acs AA IDQs TILDVQGVGL N I

a GHL T c s ~AYSVMSY REASY

*

 

VDGDNYPETLNRMFIINAGSGPRILWNTVKSFLDPKTTAK
VDCDiYPE LnRvPIINA :GFtu .t ' FLDPlT AK
SQth'uRu :' .:PF uSTAFRLPuP' 93V; '

 

INVLGNKYDTKLLBIID SBLPEFLGGTCTCTD -GGCMR°

 

LGN YQ LL I LP 66 D -GG
VKF

 

KQ P E KSEVD s

 

 

 

snxspwx a ILR’ VQN A pllncnalaxpElsx
8D GPWK P M R Du. S3;ENS T PK'D
I PKY pas PEAF‘ .
Tgh'lgl- EIQV PVC Q PAA ---------- . -- - \
D‘JT p KA KEV I PAAIPAWNMPEAHI us:
uﬁIL- L --:EHIeFI-.. ---------- n :---;
vnvr -------------- a ----- Ipvc‘ --------
EVEAHQEACNNATTEGGRS‘ FTGVM LEMGVVTMIKVTK
‘ -------------- 1“ -------------

.............. pvavaADKNASKKVGQNDT----r8

 

 

 

 

NVPRKLTESTLYSsevEcnohﬁﬁhﬁkﬁnobakpbvparso

 

 

 

 

ATMPPBKEEMLNATIS
Tﬁhppoxaannuaars

anhIMxRMABLEQx ﬂﬁp

 

BFTTVMKRMABLBEKMTTMNHQP
v 4E9?
KBFTTVMKRMAELBEKMTTMNH pa MPPBKEBMLNATI

 

 

   

 

RADVLEKQLMDTKKALEDSLAKQEVLSAYVEKKﬂ%KKKTP
RSNVLEEMLEIWKKALDDSLGRQEEhVAYIBKKK KKKEh
:. .' :- :9: .2“ ~; :12; ::1 '

_
" VD: ..

 

      

   

      

am

m
w

w
W
a

W
I”
w

I”
1%
rm

1”
1%
rm

N9
2%
I”

2$
2%
2”

2%
3H
2”

3”
3%
3M

3%
ﬂ

3W
45
M

3%
05
ﬂ

”9
MS
%

U9

5%
In

“2
5%
1M

 

 

.3832:

8a £88 85 62%-:235: 23 $2638 ox__-xon-<H<H mo gags—etc
Badman use 8363 2C. 838.8 3 @8865 2a 39838“ w R523 can
353% >73qu 05 mo coumtomcmb mo 283086 2:. .ocew >73qu 2: mo
68 ESQ 38886-5 woszooTSbE 05 mo 8:253 028226 2F 6.5 .wE

{ZN—E @Nggzva ¢ BOBUOBU§UBU§¢09<G§§BBU<¢UOU

<U¢UU<OBBU<UBBBU¢BUBBBBBBB<¢OBBO00

ﬁdU<UUBOUBEE<B¢OBUU<U<O¢¢BBUU<¢<¢O<¢O¢¢OOUU¢B<BUB<¢<<B00¢48040<¢<¢B¢O<<U<B¢U<¢

BBQBOO<UG<¢<B<BU¢UOBOBUBB<¢UDBBBEUBBUBBUDUOB<B<O¢BBBBdOUBEdUBUBBEBdBUBBUBdBOBE

<<BUBBB¢BBBBUBBOO<B<B0049UUU<<<UBUU<¢OO¢BU<UBBGUUUUBBU<UUBB<¢<B<¢O<OBU¢80404<B
BE¢U<<<B<<<<U<<UUBdB<OOB<OOUBEBU<UDEEUUB<UBO<<UUOUO¢¢OEOU<4BBBs BUBU<OB=UBUBB<

ill
U<¢U<BEOU¢¢OUBB¢BUU¢U¢40<w<¢OBBO€OB<U<UOUBdUOOU<OO<O<B<BBOBOOBUO¢<OBOOBBBUBBU¢

DEBUB<¢UUBEUO4¢B<OOBOBBOBUBBU<¢UBUdBGBOUO<BOUUOBUUBUB<E¢¢U¢UU<OUBBU¢UU¢¢40¢dOB

     

<O<UUEBDOOBB¢OUB<O<<O€O<O<<BE<<<U<O<U¢B¢9999004<O<B4OBUOUBEOBOdOddOUBU<U<OUB¢B
BUBOOQﬁUUUdﬁEUD<B BUBUBUB ..99909UBOEdB<¢¢dDUBBUB¢BU¢OUO¢¢U<UBUPBUO<OBUBUO<B¢
All 1'17
<BBO¢UUU<B<OBB¢DO¢BOOUU<BUBU4<O<BO¢B¢O¢BOBU¢OOBUUUs BUBUBUB BBBOUO¢B<O<OOOOO¢
94408000948U<¢EUUB<UUUOB¢O<UBBUE<UB<BUB<04UBUU<U®OBG<O<O¢O<<U<€4UUUB<BUBUUUUUB
Ill' ll'
oeeeeoeeoueeeeeoeeouoaeooeouaeooouu< eoeueue «ooeeawoeuoaooeue eouueoe capo

UBB<¢U<¢UUB<EB<UE<OOOEBOO<UOBBUUUUOBO<<UEO<O<¢OBOU¢¢B0060UBUOﬂUBOddUOOdUddOBdO

1'
BB<OUB¢¢UO<BU<EBBU<<<UBBUB€B¢<OB¢O<<<OUBO<0004¢00<OBBBOOEOBUOBUdOUBOéOﬁOUBOdUU

<4800488UUBdOEdddOBBBU¢<0d9<BBU<BUBBBUO<UBUUUBBUUBU<dduU¢U<004OEUO<UBUBUO<UBUO

bvm
mow
Ham
mam
mmm
bmﬂ

mm

a;

33.—8.3.:
8:82.;

3.24.3

291

Fig. 7.7. The internal bi-directional promoter of the LjPLP-IV gene directs the
expression of the GUS reporter gene to the central zone of L. corniculatus
nodules. Histochemical localization of B-glucuronidase activity in nodule hand-
sections derived from transgenic L. corniculatus plants harboring the promoter-

GUS fusions. This image is presented in color.

292

 

i ,
«-
...

u-
I
‘4

i L ‘r ".13;
..
m -.;.j,:: '5‘

m l-

I
”www.5-
., j,

 

293

 

Fig. 7.8. Subcellular localization of the mGFP5-Nljl6 fusions in the onion
epidermal cells. Fluorescent images of onion cells expressing mGF P5 alone (A),

mGFP5-Nlj16 fusion (8 & D) and mGFP5-N1j16ACC fusion (C). (E) Bright ﬁeld

image ofthe cell shown in the panel D. The onion cell shown in the panels D & E
has undergone plasmolysis and the GFP ﬂuorescence can be seen on the periphery

of the protoplast. This image is presented in color.

294

 

 

LjPLP-IV
LjPLP-II]
LjPLP-II
AtPLP

LjPLP-IV
LjPLP-[II
LjPLP-I]

402
542
471
478

 

 

 

 

 

P EKEEMLNATI
P EKEELLNAAV
PPEKEEMLNAAIS

 
     
   
 

 

’mKKAL DSL KQE
mxxa EA QE
mKKALEDSLAKQE

440
581
508
516

477
621
545

 

 

 

 

 

AtPLP 553

 

482
625
550
558

LjPLP-IV
LjPLP-III
LjPLP-II
AtPLP

 

Fig. 7.9. Amino acid sequence alignment of the C-terminal N1j16-like domains of

 

LjPLP Il-IV and Arabidopsis AtPLP (accession # Z99708.l) proteins. The
alignment was generated using the PileUp function of the Genetics Computer
Group package (GCG, Madison, Wisconsin) and the Squu software version
1.0.1. Similar residues are boxed, while the identical aminoacids are boxed and

shaded. Dashes represent gaps in the sequences.

 

296

Fig. 7.10. The LjPLP proteins complement the temperature-sensitive phenotype
of yeast sec14 mutant. The growth phenotype of the yeast strain CTY1079
(sec14-I"’ Asp014) harboring the indicated constructs is shown (see Materials and
Methods). Individual transformants were streaked on a selective (-Ura) media and

grown at permissive 28°C and non-permissive 37°C temperatures.

297

 

 

37°C

pGK-leCC

pGKinCI

 

298

 

7.6. REFERENCES:

Aitken,J.F., van Heusden,G.P.H., Temkin,M. and Dowhan,W. (1990) The gene encoding
the phosphatidylinositol transfer protein is essential for cell growth. J. Biol. Chem., 265,
4711-4717.

Bankaitis,V.A., Aitken,J.F., Cleves,A.E. and Dowhan,W. (1990) An essential role for a
phospholipid transfer protein in yeast Golgi function. Nature, 347, 561-562.

Bankaitis,V.A., Malehorn,D.E., Emr,S.D. and Greene,R. (1989) The Saccharomyces
cerevisiae SEC14 gene encodes a cytosolic factor that is required for transport of
secretory proteins from the yeast Golgi complex. .1. Cell. Biol., 108, 1271-1281.

Broughton,W.J. and Dilworth,M.Y. (1971) Control of leghemoglobin synthesis in snake
beans. Biochem. J., 125, 1075-1080.

Cockcroﬁ,S. (1998) Phosphatidylinositol transfer proteins: a requirement in signal
transduction and vesicle trafﬁc. BioEssays, 20, 423-432.

Cunningham,E., Tan,S.W., Swigart,P., Hsuan,J., Bankaitis,V. and Cockcroft,S. (1996)
The yeast and mammalian isoforms of phosphatidylinositol transfer protein can all
restore phospholipase C-mediated inositol lipid signaling in cytosol-depleted RBL-2H3
and HL60 cells. Proc. Natl. Acad. Sci. USA, 93, 6589-6593.

Delauney,A.J. and Verma,D.P.S. (1996) Improvement of soybean for nitrogen ﬁxation:
molecular genetics of nodulation. In Verma,D.P.S. and Shoemaker,R.C. (eds.), Soybean
Genetics, Molecular Biology and Biotechnology. CAN International, Wallingford, UK,
pp. 219-248.

Dickeson,S.K., Lim,C.N., Schulyer,G.T., Dalton,T.P., Helmkamp,G.M.,Jr. and
Yarbrough,L.R. (1989) Isolation and sequence of cDNA clones encoding rat
phosphatidylinositol transfer protein. J. Biol. Chem, 264, 16557-16564.

Elble,R. (1992) A simple and efﬁcient procedure for transformation of yeasts.
Biotechniques, 13, 18-20.

Gietz,R.D. and Sugino,A. (1988) New yeast-Escherichia coli shuttle vectors constructed

with in vitro mutagenized yeast genes lacking six-base pair restriction sites. Gene, 74,
527-534.

299

 

 

 

Hara,S., Swigart,P., Jones,D. and Cockcroft,S. (1997) The ﬁrst 5 amino acids of the
carboxy terminus of phosphatidylinositol transfer protein or (PITPor) play a critical role in

inositol lipid signaling: transfer activity of PITP is essential but not sufﬁcient for
restoration of phospholipase C signaling. J. Biol. Chem, 272, 14909-14913.

Hay,J.C., Fisette,P.L., Jenkins,G.H., Fukami,K., Takenawa,T., Anderson,R.E. and
Martin,T.F.J. (1995) ATP-dependent inositide phosphorylation required for Ca2+-
activated secretion. Nature, 372, 173-177.

Hay,J.C. and Martin,T.F.J. (1993) Phosphatidylinositol transfer protein required for ATP-
dependent priming of CaZ+-activated secretion. Nature, 366, 572-575.

Hong,Z. and Verma,D.P.S. (1994) A phosphatidylinositol 3-kinase is induced during
soybean nodule organogenesis and is associated with membrane proliferation. Proc. Natl.
Acad. Sci. USA, 91, 9617-9621.

Jouannic,N., Lepetit,M., Vergnolle,C., Cantrel,C., Gardies,A.-M., Kader,J.-C. and
Arondel,V. (1998) Isolation of a cDNA from Arabidopsis thaliana that complements the
sec14 mutant of yeast. Eur. J. Biochem, 258, 402-410.

Kapranov,P., de Bruijn,F.J. and Szczyglowski,K. (1997) Novel, highly expressed late
nodulin gene (LjNOD16) from Lotus japonicus. Plant Physiol., 113, 1081-1090.

Kauffmann-Zeh,A., Thomas,G.M.H., Ball,A., Prosser,S., Cunningham,E., Cockcroft,S.
and Hsuan,J.J. (1995) Requirement for phosphatidylinositol transfer protein in epidermal
grth factor signaling. Science, 268, 1188-1190.

Keams,8.G., Alb,J.G.,Jr., and Bankaitis,V.A. (1998a) Phosphatidylinositol transfer
proteins: the long and winding road to physiological function. Trends Cell Biol., 8, 276-
282.

Keams,M.A., Monks,D.E., Fang,M., Rivas,M.P., Courtney,P.D., Chen,J.,
Prestwich,G.D., Theibert,A.B., Dewey,R.E. and Bankaitis,V.A. (1998b) Novel
developmentally regulated phosphoinositide binding proteins from soybean whose

expression bypasses the requirements for an essential phosphatidylinositol transfer
protein in yeast. EMBO J., 17, 4004-4017.

Keams,M.A., McGee,T.P., Mayinger,P., Gedvilaite,A., Phillips,S.E., Kagiwada,S. and

Bankaitis,V.A. (1997) An essential role for diacylglycerol in protein transport from the
yeast Golgi complex. Nature, 387, 101-105.

300

 

Lopez,M.C., Nicaud,J.-M., Skiner,H.B., Vergnolle,C., Kader,J.-C., Bankaitis,V.A. and
Gaillardin,C. (1994) A phoshatidylinositol/phosphatidylcholine transfer protein is

required for differentiation of the dimorphic yeast Yarrowia Iipolytica from the yeast to
the mycelial form. J. Cell. Biol., 125, 113-127.

Malamy,J.E. and Benfey,P.N. (1997) Organization and cell differentiation in lateral roots
of Arabidopsis thaliana. Development, 124, 33-44.

McGee, T.P., Skinner,H.8., Whitters,E.A., Henry,S.A. and Bankaitis,V.A. (1994) A
phosphatidylinositol transfer protein controls the phosphatidylcholine content of yeast
Golgi membranes. J. Cell. Biol., 124, 273-287.

Ohashi,M., de Vries,J.K., Frank,R., Snoek,G., Bankaitis,V., Wirtz,K. and Huttner,W.B.
(1995) A role for phosphatidylinositol transfer protein in secretory vesicle formation.
Nature, 377, 544-547.

Phillips,S.E., Sha,8., Topalof,L., Xie,Z., Alb,J.G., Klenchin,V.A., Swigart,P.,
Cockcroft,S., Martin,T.F.J., Luo,M. and Bankaitis,V.A. (1999) Yeast secl4p deﬁcient in
phosphatidylinositol transfer activity is functional in vivo. Mol. Cell, 4, 187-197.

Ramlov,K.8., Laursen,N.B., Stougaard,J. and Marcker,K.A. (1993) Site-directed
mutagenesis of the organ-speciﬁc element in the soybean leghemoglobin lbc3 gene
promoter. Plant J., 4, 577-5 80.

Rodriguez-Concepcion,M., Yalovsky,S., Zik,M., Fromm,H. and Gruissem,W. (1999) The
prenylation status of a novel plant calmodulin directs plasma membrane or nuclear
localization of the protein. EMBO J., 18, 1996-2007.

Sha,8., Phillips,S.E., Bankaitis,V.A. and Luo,M. (1998) Crystal structure of the
Saccharomyces cerevisiae phosphatidylinositol-transfer protein. Nature, 391, 506-510.

Siemering,K.R., Golbik,R., Sever,R. and Haseloff,J. (1996) Mutations that suppress the
thermosensitivity of green ﬂuorescent protein. Curr. Biol., 6, 1653-1663.

Skinner,H.8., McGee,T.P., McMaster,C.R., Fry,M.R., Bell,R.M. and Bankaitis,V.A.
(1995) Phosphatidylinositol transfer protein stimulates yeast Golgi function by inhibiting
choline-phosphate cytidyltransferase activity. Proc. Natl. Acad. Sci. USA, 92, 112-116.

Szczyglowski,K., Hamburger,D., Kapranov,P. and de Bruijn,F.J. (1997) Construction of

a Lotus japonicus late nodulin expressed sequence tag library and identiﬁcation of novel
nodule-speciﬁc genes. Plant Physiol., 114, 1335-1346.

301

 

Szczyglowski,K., Szabados,L., Fujimoto,S.Y., Silver,D. and de Bruijn,F.J. (1994) Site-
speciﬁc mutagenesis of the nodule-infected cell expression (NICE) element and the AT-

rich element ATRE-BSZ* of the Sesbania rostrata leghemoglobin glb3 promoter. Plant
Cell, 6, 317-332.

Thomas,G.M.H., Cunningham,E., F ensome,A., Ball,A., Totty,N.F., Troung,O., Hsuan,J.J.
and Cockcroft, S. (1993) An essential role for phosphatidylinositol transfer protein in
phospholipase C-mediated inositol lipid signaling. Cell, 74, 919-928.

Vanhee-Brossollet,C. and Vaquero,C. (1998) Do natural antisense transcripts make sense
in eukaryotes? Gene, 211, 1-9.

Varagona,M.J., Schmidt,R.J. and Raikhel,N.V. (1992) Nuclear localization signal(s)
required for nuclear targeting of the maize regulatory protein Opaque-2. Plant Cell, 4,
1213-1227.

Verma,D., Kazain, V., Zogbie,V. and Bal., (1978) Isolation and characterization of the
membrane envelope enclosing the bacteroids in soybean root nodules. J. Cell Biol., 78,
919-939.

Vihtelic,T.S., Goebl,M., Milligan,S., O'Tousa,S.E. and Hyde,D.R. (1993) Localization of
Drosophila retinal degeneration B, a membrane-associated phosphatidylinositol transfer
protein. J. Cell Biol., 122, 1013-1022.

Volschenk,H., Viljoen,M., Grobler,J., Petzold,B., Bauer,F., Subden,R.E., Young,R.A.,
Lonvaud,A., Denayrolles,M. and van Vuuren,H.J.J. (1997) Engineering pathways for
malate degradation in Saccharomyces serevisiae. Nature Biotech., 15, 253-257.

von Amim,A.G., Deng,X.-W. and Stacey,M.G. (1998) Cloning vectors for the expression
of green ﬂuorescent protein fusion proteins in transgenic plants. Gene, 221, 35-43.

Waterhouse,P.M., Graham,M. and Wang,M.-B. (1998) Virus resistance and gene

silencing in plants can be induced by simultaneous expression of sense and antisense
RNA. Proc. Natl. Acad. Sci. USA, 95, 13959-13964.

302

 

FUTURE PERSPECTIVES

The research described in this thesis has led to identiﬁcation and characterization
of a number of novel late nodulin genes from a model legume, Lotus japonicus. Three
nodulin genes, LjNPP2C], LjNOD70 and LjNODI6, were chosen for detailed
characterization, leading to the formulation of separate projects that yielded their own
hypotheses and conclusions. These projects also generated a number of questions to be
addressed in future, the main one being the biological role of these genes in nodules and
other tissues of L. japonicus. This question was addressed to various extents for each of
these genes using various molecular and biochemical approaches.

In my opinion, the disruptions of, or loss-of-function mutations in, the genes
encoding LjNPP2C1 and its potential interacting partner (LBP) will be absolutely
required to understand the biological functions of these proteins and assess the relevance
of their interaction. A second route to address the function of LjNPP2C1 is to conduct a
more extensive yeast two-hybrid screen for putative binding proteins, followed by genetic
experiments to test in vivo the signiﬁcance of any observed interactions.

Nodulin LjN70 could potentially be an organic acid transporter in the
peribacteroid membrane. It is therefore crucial to functionally characterize the
biochemical properties of this protein, to establish whether it has transporter-like activity
and to identify its substrates. Such experiments are in progress in collaboration with the
laboratory of Dr. Dan Roberts. The University of Tennessee, Knoxville. In addition, it
will also be extremely interesting to know if the products of the LjNOD70 gene and

related members of this gene family are required for normal functioning of L. japonicus

303

 

nodules. The anti sense experiments with transgenic L. japonicus plants are in progress to
down regulate the expression of LjNOD70 and related genes. It is also important to
establish whether the spliced form of the LjNOD70 mRNA has a speciﬁc function in
nodules, and, whether it encodes a protein product.

The function of the LjPLP gene family and LjNODI6 late nodulin is quite
intriguing. The apparent nodule-ﬂower dichotomy in the gene expression patterns of the
different members of this family is very interesting and clearly warrants further
investigation. Furthermore, the detailed biochemical characterization of the lipid binding
and transfer properties of the LjPLP proteins is necessary to gain an understanding of
their function. These experiments are underway in collaboration with the laboratory of
Dr. Vytas Bankaitis, University of Alabama, Birmingham. In addition, several important
questions remain. What are the functions of Nle6 protein and the LjPLP-IV antisense
transcripts? Are they a part of a mechanism designed to inactivate the LjPLP-IV and
other LjPLP genes in nodules? If yes, then why is this gene family inactivated in
nodules? What are the functions of the members of LjPLP gene family in other plant

tissues?

304