‘ 7.
knuilfnmhfiw rs
b o» h
‘- I
k "rt.
J.

.
.5

a .‘l
15.1 In.“
an . .r

5
tr

.

k n

s...
. l‘}
E»?
i 518...”... z... .
£3me
Jan . a
Ems...“

. 3.5!; t).
1 I

WWW” z .
:45“. 1,.» {husband‘s

TC. .31.!!!
: 3.1. .u- . Ilballl
li

a,

[54

Jun:

h”

I. k l.
. z: .

.v 1.,(,\(:‘!lr.

 

 

“

will

1L

Ililllllflllllfllllfill

31293 01 0

 

LIBRARY
Michigan State
University

 

 

 

This is to certify that the

dissertation entitled

THE RNS FAMILY OF S-LIKE RIBONUCLEASES
0F ARABIDOPSIS THALIANA:
STRUCTURES, {EXPRESSION AND FUNCTIONS

 

presented by

Pauline Anne Bariola

has been accepted towards fulfillment
of the requirements for

Ph.D. . Biochemistry
degree in

 

 

 

Date /%//%/7(0

MS U is an Affirmative Action/Equal Opportunity Institution

0-12771

PLACE N RETURN BOX to romovo this checkout from your rooord.
TO AVOID FINES mum on or before date duo.

      

DATE DUE DATE DUE DATE DUE

r—(E:\£3 “C 'l i

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU loAnAfflmotlvo Action/Equal Opportunity Inotltwon
mm:

 

THE RNS FAMILY OF S-LIKE RIBONUCLEASES OF ARABIDOPSIS THALL4NA:
STRUCTURES, EXPRESSION AND FUNCTIONS

By

Pauline Anne Bariola

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

DOCTOR OF PHILOSOPHY

Department of Biochemistry

1996

U'MI Number: 9718808

 

UMI Microform 9718808
Copyright 1997, by UMI Company. All rights reserved.

This microform edition is protected against unauthorized
copying under Title 17, United States Code.

 

UMI

300 North Zeeb Road
Ann Arbor, MI 48103

ABSTRACT

THE RNS FAMILY OF S-LIKE RIBONUCLEASES OF ARABIDOPSIS
T HALIANA: STRUCTURES, EXPRESSION AND FUNCTIONS

By

Pauline Anne Ban'ola

In the past decade, ribonucleases (RNases) have been found to be involved in many
unexpected processes in different biological systems. One of these processes is self-
incompatibility in some types of plants, in which the involvement of S-RNases has been
shown to be crucial. This phenomenon prompted the identification of three genes
homologous to the S-RNases in Arabidopsis thaliana, a self-compatible plant. The project
described in this thesis involves the characterization of these genes and their expression, as
well as experiments designed to provide insight into the roles of these proteins, termed S-
like RNases, in self-compatible plants and plants in general. At the outset of the project,
cDNAs for the three genes, RNS] , RNS2 and RNS3, were isolated and sequenced.
Extensive comparisons of the deduced amino acid sequences to those of other S-RNases
and S-like RNases were made, revealing several regions where residues of either group
were found to be unique in comparison to the other group. The RNS cDNAs were
expressed in yeast and the resulting proteins assayed for RNase activity, which showed that

the genes encode active RNases. Expression studies showed that although closely related at

the sequence level, these genes have quite different expression patterns. All are expressed
in flowers, but among other organs expression varies. All three are induced to some extent
during senescence, but only RNS] and RNS2 are induced during starvation for phosphate
(Pi). These observations led to the proposal that RN81 and RN82 have roles in Pi
remobilization. Antibodies that recognize RN82 specifically were made for use in the
subcellular localization of RN82. Immunogold electron microscopy of leaves and petals
showed RN82 present in the extracellular space, a location consistent with a role in Pi
remobilization and possibly defense against pathogens. The final aspect of the project
involved the generation of antisense and overexpression transgenic plants in which the ideas
about the roles of these proteins could be tested. Attempts to generate antisense RN82
plants led to lines in which RNSZ expression was only mildly inhibited. In contrast,
antisense RNS] plants were obtained in which induction of RNS] in flowers and during Pi
starvation is reduced substantially. These lines also exhibit high anthocyanin levels, which
is a symptom of Pi starvation and stress. Further characterization will be required to
determine the mechanism leading to induction of anthocyanins. However, these results
demonstrate that decreasing the expression of one member of the RNS gene family is

sufficient to induce an identifiable phenotype.

ACKNOWLEDGEMENTS

Thanks are due to many, many people for contributions to this thesis project and my
experience at Michigan State. First of all, I would like to thank Pam Green for her
boundless enthusiasm, energy and confidence in me. I also wish to thank my committee
members, Zach Burton, Lee Kroos, Lee McIntosh, Jack Preiss and Mike Thomashow, for
support and helpful advice. My project would not have been nearly as much firn without
working with my co-authors, Crispin Taylor and Christie Howard. I would like to
acknowledge the contributions of all my collaborators, including Ron Raines and Steve del
Cardayré, who taught me to use their yeast expression system; Yves Poirier, who provided
seeds of the phol mutant as well as advice about phosphate experiments; Natasha Raikhel
and Olga Borkhsenious, who assisted with irnmunogold electron microscopy; and Marcel
Bucher, who initiated the RNSI root overexpression project. Several talented
undergraduates also contributed to this project, including Andre Dandridge, who picked
numerous Arabidopsis petals for Figure 2-7, Michael Verburg, who contributed to the
sequencing of the RNS] cDNA, and Vanita Jaglan, who ran the gel shown in Figure 3-13.
Thanks are also due to many people who have helped me with techniques during the course
of this project, including Jun Tsuji and Xinnian Dong for pathogen studies, Dave Shintani

and Jennifer Gorlach for assistance with I-IPLC, and Andrew Bent and David Bouchez for

iv

advice on vacuum infiltration. This project would not have been possible without the
contributions of Don Herrington, who cared for my rabbits expertly, Joe Leykam, who
synthesized peptides, and Kurt Stepnitz and Marlene Cameron, who provided photographic
assistance throughout my project. As I have asked advice of people in numerous
laboratories in the Biochemisu'y and Botany departments, as well as just about every
laboratory in the PRL, there are too many to mention, but I would like to thank everyone
sincerely.

Finally, I would like to thank all the friends I have made at Michigan State for their
support throughout these (many) years. Included in this group are all present and former
Green lab members, who made our lab a wonderful place to work, and in particular Mike
Abler, Jay De Rocher, Mike Sullivan, and Tom Newman for their advice, humor and ability
to calm me down in stressful situations. Finally I would like to thank Susan Fujimoto, who
has been a great friend and roommate, and my parents, who made me happy by learning

how to pronounce Arabidopsis.

TABLE OF CONTENTS

LIST OF FIGURES ................................................................................................................. ix
CHAPTER 1
INTRODUCTION: PLANT RIBONUCLEASES .................................................................. 1
Classes of Plant RNases ............................................................................................... 3
Classifications Based on Early Biochemical Work ......................................... 3
Plant RNases in the T2 Family ......................................................................... 4
Pathogenesis-Related Protein Group PR- 1 O .................................................. 12
Group V Allergens ......................................................................................... 13
Bifunctional Nucleases .................................................................................. 14
Regulation and Functions of Plant RNases ................................................................ l7
Phosphate Remobilization ............................................................................. 17
Senescence ...................................................................................................... 19
Cell Death Pathways ...................................................................................... 21
Defense Against Pathogens ............................................................................ 22
RNA Processing and Decay ........................................................................... 25
Dissertation Topic and Thesis Overview ................................................................... 29
CHAPTER 2
RN82: A SENESCENCE-ASSOCIATED RNASE OF ARABIDOPSIS THAT
DIVERGED FROM THE S-RNASES BEFORE SPECIATION ......................................... 30
Abstract ....................................................................................................................... 31
Introduction ................................................................................................................. 33
Results and Discussion ............................................................................................... 34
Comparison of RN 82 to Related RNases ...................................................... 37
Control of RNSZ Expression .......................................................................... 47
Conclusions .................................................................................................... 54
Materials and Methods ............................................................................................... 56
Isolation and Sequencing of RN82 cDNAs ................................................... 56
Expression of RN82 in Saccharomyces cerevisiae ....................................... 56
Multiple Sequence Alignment and Gene Genealogy .................................... 57
Expression Analyses ...................................................................................... 57

vi

CHAPTER 3
STUDIES ON RNS] and RNS3: RNS] IS TIGHTLY CONTROLLED

IN RESPONSE TO PHOSPHATE STARVATION ............................................................. 60
Abstract ....................................................................................................................... 61
Introduction ................................................................................................................. 62
Results ......................................................................................................................... 65

Features of RNS] and RN83 Sequences ........................................................ 65
Comparison of RN81 and RN 83 Protein Sequences With Those of Other 8-

Like RNases ....................................................................................... 71

Ribonuclease Activity of RN81 and RN 83 ................................................... 72

RNS] and RN83 Expression during Normal Development ........................... 74

RNS] and RNS3 Expression in Response to Phosphate Limitation .............. 76

RNase Profiles of Pi-starved and phoI Plants ............................................... 82
Discussion ................................................................................................................... 86
Structural Features of RN81 and RN83 ........................................................ 86
Induction of the RNS Genes During Senescence and Pi Starvation is
Differential and not a General Nutrient Starvation Response .......... 87

Roles of the RNS Gene Products in Higher Plants ........................................ 89
Materials and Methods ............................................................................................... 92
cDNA Isolation .............................................................................................. 92
Expression Analyses ...................................................................................... 92
Expression of the RNS cDNAs in Yeast ........................................................ 94

RNase Activity Gels ....................................................................................... 95

CHAPTER 4

IMMUNOLOCALIZATION OF RN82 ................................................................................ 96
Abstract ....................................................................................................................... 97
Introduction ................................................................................................................. 98
Results ....................................................................................................................... 100

Production of Antibodies that Specifically Recognize RN82 .................... 100
Patterns of RN82 Expression ....................................................................... 105
Immunolocalization of RN 82 ...................................................................... 108
Discussion ................................................................................................................. 112
Materials and Methods ............................................................................................. 115
Anti-RN 82 Antibody Production ................................................................ 115
Protein Gels and Immunoblot Analysis ....................................................... 117
Immunocytochemistry ................................................................................. 1 18

CHAPTER 5

GENERATION OF TRANSGENIC PLANTS WITH DECREASED AND INCREASED

AMOUNTS OF RNS] AND RN82 ..................................................................................... 120
Abstract ..................................................................................................................... 121
Introduction ............................................................................................................... 122
Results and Discussion ............................................................................................. 124

vii

Antisense Inhibition of the RNS] Gene ....................................................... 124

Antisense Inhibition of the RN82 Gene ....................................................... 138
Overexpression of RNS] in Roots ............................................................... 143
Materials and Methods ............................................................................................. 149
Plasmid Constructions ................................................................................. 149
Generation of Transgenic Plants .................................................................. 151
Anti-RN81 Antibody Preparation ................................................................ 152
Expression Analyses .................................................................................... 153
Anthocyanin Assays ..................................................................................... 157

CHAPTER 6
CONCLUSIONS AND FUTURE PROSPECTS ................................................................ 158
LIST OF REFERENCES ...................................................................................................... 162

viii

LIST OF FIGURES

Figure 2-1 - Strategy for sequencing of RNS2 cDNA ............................................................ 35
Figure 2-2 - Primary structure and deduced amino acid sequence of RNS2 cDNAs ............ 36
Figure 2-3 - Expression of RN 82 in yeast .............................................................................. 38
Figure 2-4 - Alignment of S- and S-like RNase amino acid sequences ................................. 39
Figure 2-5 - Gene genealogy of S— and S-like RNases ........................................................... 45
Figure 2-6 - Expression of RNS2 in different organs of Arabidopsis .................................... 48
Figure 2-7 - Induction of RNS2 during senescence and phosphate starvation ....................... 49
Figure 2-8 - Induction of RNS2 by pathogens ........................................................................ 52
Figure 2-9 - Time course of RN82 induction by pathogens ................................................... 53
Figure 3-1 - Strategy for sequencing of RNS] and RN83 cDNAs ........................................ 66
Figure 3-2 - Primary structure and deduced arrrino acid sequence of RNS] cDNA .............. 67
Figure 3-3 - Primary structure and deduced amino acid sequence of RN83 cDNAs ............ 68

Figure 3-4 - Comparison of deduced amino acid sequences of RN81 and RN83 to
sequences of RN ases LE and LX of tomato .............................................................. 69

Figure 3-5 - Amino acid similarities between the RN 8 proteins and RNases LE and LX ...70

Figure 3-6 - Expression of RN 8 proteins in Saccharomyces cerevisiae ............................... 73
Figure 3-7 - RNS expression in organs of soil-grown Arabidopsis ........................................ 75
Figure 3-8 - RNS expression during senescence in leaves of Arabidopsis ............................ 77

Figure 3-9 - RNS expression in plants grown under Pi-rich or Pi-deficient conditions ......... 78

Figure 3-10 - RNS] expression in plants starved for various nutrients .................................. 80
Figure 3-11 - RNS] expression in plants grown on media containing no potassium ............ 81
Figure 3-12 - RNS Expression in the phoI mutant of Arabidopsis ........................................ 83
Figure 3-13 - RNase activity profiles of Pi-starved plants and p110] mutant plants .............. 85
Figure 4-1 - Synthetic peptides used for producing anti-RNS2 antibodies ......................... 101
Figure 4-2 - Immunoblot characterization of anti-RNS2 antibodies ................................... 104
Figure 4-3 - Distribution of RN 82 among various organs of Arabidopsis .......................... 106
Figure 4-4 - Increase in RN82 abundance during phosphate starvation .............................. 107
Figure 4-5 - Immunocytochemical localization of RN S2 in leaves of Arabidopsis using
irnmunogold labeling ................................................................................................ 1 10
Figure 4-6 - Immunocytochemical localization of RN S2 in petals of Arabidopsis flowers
using irnmunogold labeling ...................................................................................... 111
Figure 5-1 - Attempts to prepare anti-RN81 antibodies using a peptide antigen ................ 125

Figure 5-2 - Anti-RN81 antibody preparation using heterologously-produced RN 81 as an

antigen ....................................................................................................................... 127
Figure 5-3 - Structure of antisense RNS] plant transformation vector ................................ 128
Figure 5-4 - Decreased RN81 activity in RNS] antisense lines ........................................... 131
Figure 5-5 - RNS] mRNA levels in RNS] antisense lines ................................................... 133
Figure 5-6 - RN81 protein levels in RNS] antisense lines ................................................... 134
Figure 5-7 - Quantitation of anthocyanin levels in seedlings of RNS] antisense lines ....... 136
Figure 5-8 - Structure of antisense RN82 plant transformation vectors ............................... 139
Figure 5-9 - RN82 mRNA levels in RNS2 antisense lines ................................................... 142
Figure 5-10 - Structure of RNS] root overexpression transformation vector ...................... 144

Figure 5-11 - RNS] mRNA levels in RNS] root overexpression lines ................................ 146

Figure 5-12 - Effect of Pi starvation on RNS] mRNA levels and RN81 activity in roots .. 147

xi

CHAPTER 1

INTRODUCTION: PLANT RIBONUCLEASES

Portions of this chapter are in press:

Bariola PA, Green P] (1997) Plant Ribonucleases. In: D'Alessio G, Riordan IF (eds)
Ribonucleases: Structure and Functions. Academic Press, Inc., Orlando, Florida. In
press.

2

During the past few years, the study of plant ribonucleases (RNases) has advanced
considerably due to renewed interest in the field and to technological advances. Prior to
the early 1980's, the regulation and biochemistry of plant RNases were actively
investigated as reviewed by Farkas (1982) and Wilson (1982), but subsequently interest
in the field subsided because the methods available at the time were insufficient to
elucidate the biological firnctions of individual enzymes. Much renewed interest was
prompted by the discovery that genotype-specific ribonucleases, the S-RNases, are
critical components of self-incompatibility (SI) mechanisms in some Solanaceous plants
(McClure et al. 1989b; Lee et a1. 1994). These findings, which exploited gene cloning
and transgenic plant technologies, emphasized that plant RNases could participate in
diverse and unexpected processes. Moreover, the discovery of the S-RNases led to the
identification of a large number of related plant RNases that do not function in self-
incompatibility, called the S-like RNases.

The goal of this chapter is to highlight current knowledge of the S-like RNases
and other RNA-degrading enzymes in higher plants. Rather than attempt a
comprehensive review, this chapter concentrates on findings of the last decade with
emphasis on molecular analyses and work reported since the last review of this topic
(Green 1994). The chapter is divided into sections dealing with the classes of plant

RNases and with their regulation and functions.

3

CLASSES OF PLANT RNASES

Classifications Based on Early Biochemical Work

Before the widespread use of molecular biology and protein microsequencing,
plant RNases were extensively characterized on the basis of their biochemical properties.
An abundance of reports on RNases from a variety of plants facilitated the classification
of plant RNases into four main groups: RNase I, RNase II, Nuclease I and Exonuclease I
(Farkas 1982; Wilson 1982). RNase I proteins, or acid RNases, are RNA-specific,
soluble endonucleases with molecular weights from 20 to 25 kDa and pH optima between
5.0 and 6.0. They are insensitive to EDTA and produce 3'-phospho (3'-P) nucleotides as
end products. RNase II enzymes are also RNA-specific endonucleases, with molecular
weights between 17 and 25 kDa. They too are EDTA-insensitive and produce 3'-P
nucleotide end products, but unlike RNase I enzymes, they have pH optima of 6.0 to 7.0.
They differ most from RNase I-type enzymes by their microsomal location. Both RNase
I and RNase II enzymes preferentially cleave bonds adjacent to guanine. Nuclease I
proteins degrade both RNA and ssDNA endonucleolytically, with a preference for bonds
adjacent to adenine, and produce 5'-P nucleotide end products. Highly sensitive to
EDTA, they have molecular weights of 31 to 39 kDa and pH optima of 5.0 to 6.5. Lastly,
Exonuclease I enzymes are large exonucleases of more than 100 kDa, are capable of
degrading both RNA and ssDNA, have pH optima of 7.0 to 9.0 and a high sensitivity to

EDTA, and produce 5'-P end products.

4

The above designations are not comprehensive and did not include all known
RNases even at the time of their publication (Farkas 1982; Wilson 1982). The
classification of RNases was further complicated by common problems such as
proteolytic degradation and aggregate formation during purification, as well as
inconsistency in the assay procedures used in different laboratories. In addition, not
every defining characteristic is investigated for each RNase reported. Despite these
problems, many RNases reported more recently fit well into one of the four groups. Both
Arabidopsis thaliana (Yen and Green 1991) and barley (Yen and Baenziger 1993) appear
to have representatives of the RNase I, RNase II, and nuclease I classes, based on the
properties of the RNA-degrading enzymes observed using activity gels. Additional
reports from barley describe RNase I-type enzymes (Prentice and Heisel 1985; Kenefick
and Blake 1986) and nucleases (Prentice and Heisel 1986; Brown and Ho 1987). Other
RNase I-type enzymes have been identified in wheat (Blank and McKeon 1991a) and
tomato (Abel and Glund 1987). A protein from wheat is a recent addition to the RNase 11
family (Yen and Baenziger 1993), and nucleases have been found recently in wheat
(Kuligowska et a1. 1988; Blank and McKeon 1989; Yen and Baenziger 1993), rye
(Siwecka et a1. 1989), zinnia (Thelen and Northcote 1989), and lentil (Kefalas and

Yupsanis 1995).

Plant RNases in the T2 Family
The family of plant RNases best characterized molecularly is a subset of the

family of enzymes typified by the fungal RNase T2 (Irie 1997). This family is the most

5

widespread RNase family known, with representatives in viruses (Schneider et al. 1993;
Hime et al. 1995), bacteria (Meador, III and Kennel] 1990; Favre et al. 1993), fungi (Irie
1996), slime mold (Inokuchi et a1. 1993), Drosophila (Hime et al. 1995), oyster

(Watanabe et al. 1993), cow (Irie 1993), and chicken (Irie 1993).

S-RNases: Plant members of the T2 family were first identified when sequences of
proteins genetically associated to gametophytic self-incompatibility in the Solanaceae
family (Kao and Huang 1994) were determined and found to be similar to fungal
ribonucleases (McClure et al. 1989b). Tenned S-RNases, numerous examples of these
proteins have been characterized in tobacco, tomato, potato, and petunia, each
Solanaceous species (Newbigin et al. 1993). More recently, S-RNase families have been
identified in snapdragon (the Scrophulariaceae family) (Xue et al. 1996), and in apple
(Broothaerts et al. 1995) and pear (Norioka et al. 1995) of the Rosaceae family.

However, other types of self-incompatibility are not controlled by S-RNases, such as that
exhibited by the Brassicaceae family (Dodds et al. 1996b). The production of
extracellular S-RNases in styles is both necessary and sufficient to confer plants with
self-incompatibility, as was shown by transgenic plant studies in petunia (Lee et al. 1994)
and tobacco (Murfett et al. 1994). It has been shown that the ribonuclease activity of the
S-proteins is essential for the self-incompatibility phenotype (Huang et al. 1994). In
vivo, rRNA is degraded in incompatible pollen tubes (McClure et al. 1990), leading to
two popular models of ribonuclease action during self-incompatibility. In the first model,

S-RNases are taken up into pollen tubes by a receptor whose gene is presumably closely

6

linked to the S-allele, degrading the RNA in the pollen tube, halting growth and thus
preventing pollination. In the second model, S-RNases are allowed to enter pollen tubes
non-specifically but inactivated once inside, unless the interaction is incompatible (Dodds

et al. 1996b). Much work needs to be done to provide evidence for either model.

Other Plant Members of the T2 family: The identification of the S-RNases led to the
discovery of related proteins in a variety of self-compatible plant species. Based on their
similarity to S-RNases, the latter group of enzymes was referred to as "S-like RNases"
(Taylor and Green 1991; Taylor et al. 1993).

The S-like RNases have the two conserved histidine residues shown to be
necessary for catalysis in RNase Rh, a related fungal RNase (Ohgi et al. 1992), and the
five boxes of conserved sequence characteristic of the S-RNases (Ioerger et al. 1991),
although these boxes in the S-like RNases tend to be less highly conserved. In general,
the S-like RNases have molecular weights between 21 and 29 kDa, and most have been
shown or predicted to be secretory proteins. Although the S-RNases and S-like RNases
share many structural features (Green 1994), each group contains highly conserved
residues not found in the other (Taylor et al. 1993). In fact, gene genealogies of these
RNases indicate that most S-like RNases form a lineage distinct from that of the
S-RNases (Bariola et al. 1994) (also see Chapter 2). Initially the term "S-like RNases"
was used to describe RNases that are structurally similar to S-RNases but found in
species not known to exhibit self-incompatibility. However, as more of these proteins

were found in a wide variety of plant species, it became apparent that S-like RNases are

7

likely to play roles distinct from self-incompatibility, but fundamental to plants in
general. Based on this hypothesis, self-incompatible plants would be expected to contain
proteins of both the S-RNase and S-like RNase families. This is now known to be the
case, as discussed below. In addition, some species that exhibit self-incompatibility
contain proteins very closely related to S-RNases that are not involved in the
self-incompatibility of the plant, as will be discussed. These enzymes lack the
distinguishing structural features associated with S-like RNases and appear to have arisen
as part of the S-RNase lineage afier it diverged from the S-like RNase lineage. To avoid
confusion, we propose that S-like RNases be defined as proteins from self-incompatible
or self-compatible plants whose structures allow their placement into the evolutionary
lineage shown in Figure 2-5 (see Chapter 2).

Comparison of the amino acid sequences of the proteins that form the S-like
RNase lineage to those of the S-RNases reveals numerous residues that are conserved in
S-like RNases but not in S-RNases (Green 1994). Conserved residues unique to the 8-
like RNases are mainly found between the conserved boxes that contain the two active-
site histidine residues, as well as at the N-terminal end (see Green 1994, Figure 1).

The first S-like RNase genes to be identified were RNS], RNS2, and RNS3 of
Arabidopsis thaliana (Taylor and Green 1991). These were cloned via the polymerase
chain reaction (PCR) based on their containing the conserved active site regions of the T2
family. The initial PCR products isolated in this reaction contained several residues
conserved in the S-RNases. When the full sequences of their cDNAs were determined,

however, several differences from the S-RNases were apparent (Taylor et al. 1993; Green

8

1994), laying the foundation for the designation of the S-like RNases as a separate group.
One of the RNS genes, RNS2, is expressed in all organs examined with highest expression
in flowers (Taylor et al. 1993). Immunocytological evidence indicates that RN82 is
extracellular (Chapter 4); RN81 and RN83 are also expected to be extracellular because
they appear to contain typical N-terminal signal sequences for entry into the secretory
pathway (Bariola et al. 1994).

RN81 and RN 83 are quite closely related to RNases LE (Jost et al. 1991) and LX
(Loffler et al. 1993) (see Figure 2-5) of the self-compatible tomato Lycopersicon
esculentum, both of which were isolated from cell cultures due to their appearance upon
starvation for phosphate (Pi) (Niirnberger et al. 1990; Loffler et al. 1992). Three other Pi-
starvation induced RNases from tomato (Loffler et al. 1992) have also recently been
sequenced at the protein level (Kock et al. 1995). These enzymes, designated LVl , LV2,
and LV3, were isolated from tomato vacuoles. Interestingly, LV3 appears to be identical
to RNase LE, an extracellular protein, and the regions of LVl and LV2 that have been
sequenced are identical to sequences of RNase LX. The latter enzyme has been shown to
have an intracellular but extravacuolar location (Loffler et al. 1992). RNase LX contains
a putative C-tenninal endoplasmic reticulum retention signal (Loffler et al. 1993), which
is missing in the LV2 peptide sequence (Kock et al. 1995). This observation may
indicate that in the absence of this putative ER-retention signal, the protein is targeted to
the vacuole. It has not been reported whether these proteins sharing long sequences are

products of the same genes. Whether or not this is the case, regulating the location of

9
these RNases could be a novel mechanism for control of RNase activities and possibly
RNase functions.

Another member of the Solanaceae, the self-compatible species Nicotiana
sylvestris, was found to contain a stylar RNase with homology to the T2 family (J. Golz,
M. Anderson, E. Newbigin, manuscript in preparation). Tubers from the lotus species
Nelumbo nucifera gave rise to another S-like RNase, identified as a storage protein (G.
Day, Z. Chen, T. Chow, unpublished, Genbank accession number M83668). Two S-like
RNases were identified in cultured zinnia leaf mesophyll cells undergoing xylogenesis
(Ye and Droste 1996). These genes, ZRNaseI and ZRNaseII, are very closely related to
the tomato and Arabidopsis RNases described above, but have different expression
patterns; neither is induced in response to P, starvation (Ye and Droste 1996).

Other studies have shown that the seeds of several cucurbit species contain S-like
RNases. The first of these enzymes to be identified was RNase MCI from the seeds of
the bitter gourd Momordica charantia (Ide et al. 1991). Subsequently two RNases were
isolated on the basis of their translational inhibitory properties in cell-free systems:
cusativin, from cucumber seeds (Rojo et al. 1994a), and melonin, from seeds of the melon
Cucumis melo (Rojo et al. 1994b). Both of these enzymes were found to have sequences
characteristic of S-like RNases. Cusativin is known to accumulate only in the coat and
cotyledons of dry seeds (Rojo et al. 1994a). Recently two related S-like RNases, LC]
and LC2, have been cloned from the seeds of Lufla cylindrica, the sponge gourd (T.

Nakamura, K. Sasaki, G. Funatsu, submitted). As all of the above cucurbit species are

10

self-compatible it was suggested that the RNases play a role in protection of the seeds
against pathogens ((Rojo et al. 1994a); T. Nakamura, K. Sasaki, G. Funatsu, submitted).

It is likely that S-like RNases will be found to be widespread among
monocotyledonous plants as well. This is evidenced by the recent identification in rice,
which is self compatible, of several cDNA sequences corresponding to S-like RNases by
investigators participating in the Rice Genome Project (Genbank accession numbers
D21885, D22272, D2364], and D24884). Based on these partial sequences it is not yet
clear whether monocot S-like RNases will form their own sublineage within the S-like
RNase lineage.

Interestingly, self-compatible species have been found to contain proteins
structurally similar to the S-RNases. Two RNases, 8x and So, have been identified in a
self-compatible cultivar of Petunia hybrida (Ai et al. 1992). Both their structural
similarity to the S-RNases and breeding behavior suggest that these RNases are defunct
S-RNases, possibly selected for in the breeding process that generated this
self-compatible cultivar from its self-incompatible ancestors (Ai et al. 1992). A
self-compatible variant of the potato Lycopersicon peruvianum contains a protein, Sc,
whose sequence identifies it as a member of the S-RNase family, but which lacks RNase
activity (Royo et al. 1994). A mutation of one of the highly conserved histidines in
RNase Sc is thought to be responsible for the lack of activity and thus the
self-compatibility of this normally self-incompatible plant. Like RNases 8x and So, the SC

protein is structurally more closely related to the S-RNases than the S-like RNases.

11

Several Tz-related enzymes other than the S-RNases have also been identified in
self-incompatible plants. RNase NE, an S-like RNase with a sequence quite similar to
those of RNase LE of tomato and RNSl of Arabidopsis, was identified via PCR in
anthers of Nicotiana alata (Dodds et al. 1996a). Similarly, styles of Nicotiana alata
contain a member of the T2 family, RNase M81, thought to be unassociated with
self-incompatibility (Kuroda et al. 1994), but the sequence of this enzyme has not been
reported so it is unclear if this protein is an S-RNase or an S-like RNase. Other members
of the T2 family in self-incompatible plants bear stronger resemblance to the S-RNases,
such as RNase X2 from Petunia inflata (Lee et al. 1992). Although abundant in pistils,
the protein is not associated with self-incompatibility. RNase X2 may have diverged
from the S-RNases, or alternately evolved from a common ancestor (Lee et al. 1992).

Few of the plant RN ases described in this section have been categorized according
to the traditional biochemical classifications referred to earlier. However, RNases LE,
LX, LVl, LV2 and LV3 from tomato are all considered RNase I-type enzymes, based on
their biochemical properties (Niimberger et al. 1990; Jost et al. 1991; Loffler et al. 1992;
Loffler et al. 1993). Most of the S-like RNases described in this section have molecular
weights in the 20 to 25 kDa range specified for RNase I-type enzymes, but RNS2 from
Arabidopsis, with a deduced molecular weight of 27.2 and two potential N-glycosylation
sites (Taylor et al. 1993), is an exception. The tertiary structure of the S-like RNases has
been investigated to a limited extent. Preliminary observations indicate that the tertiary
structure of RNase LE (M. Irie, M. Keck, K. Glund, unpublished) exhibits several

differences from that of RNase Rh fi'om the fungus Rhizopus niveus (Kurihara et al. 1992;

12

Kurihara et al. 1996), the only other member of the T2 family whose tertiary structure is
currently available. In addition, RNase MCI was recently crystallized (De and Funatsu

1992), so additional data on this enzyme may be forthcoming.

Pathogenesis-related protein group PR-10

Pathogenesis-related proteins (PR proteins) are enzymes induced in plants upon
pathogen attack or in related situations (van Loon et al. 1994), and presumably have roles
in the defense of plants against pathogens. The numerous families of PR proteins include
such diverse members as chitinases (Legrand et al. 1987), glucanases (Kauffrnann et al.
1987), and proteinase inhibitors (Geoffroy et al. 1990). PR proteins were recently linked
to RNases with the report that a ginseng RNase with non-specific activity (Moiseyev et
al. 1994) has protein sequence homology to two PR proteins from parsley (Somssich et
al. 1988; van de Locht et al. 1990). The parsley proteins in turn are known to be
members of a large PR-protein family recently designated as PR-10 (van Loon et a1.
1994; Walter et al. 1996 and references therein). The members of this family are present
in a variety of plants, have molecular weights of 17 to 18 kDa, and are considered
intracellular. In addition, proteins of a related group, the Bet v I family, the major pollen
allergens in birch, are constitutively present at high levels in pollen (Swoboda et al.
1994). These pollen allergens constitute a large isoform family in birch and other plants
(Swoboda et al. 1995).

Besides the ginseng RNase preparation (Moiseyev et al. 1994), which has recently

been shown to be a mixture of two related proteins (G. Moiseyev, J. Beintema,

l3

unpublished), the only other protein in the PR-lO family that has been reported to have
RNase activity is Bet v I, reported in two separate papers (Swoboda et al. 1996; Bufe et
al. 1996). Efforts to demonstrate activity for members of this family in potato (Constabel
and Brisson 1995), parsley (1. Somssich, unpublished), and asparagus (S.A.J. Warner, J.
Draper, unpublished) have been unsuccessful. Until more proteins of this type are shown
to exhibit RNase activity, it may be premature to designate this family as another major
family of plant RNases. However, if activity can be demonstrated in other PR-lO
proteins, this family would be the first group of intracellular, non-specific RNases
characterized molecularly in plants. In addition, it may then be possible to establish a

direct link between specific RNase activities and plant defense against pathogens.

Group V Allergens

RNase activity has also been associated with a protein from timothy grass, Phl p
Vb (Bufe et al. 1995), a member of a group of grass pollen allergens molecularly distinct
from the Bet v I family described above. These proteins, the group V allergens, generally
have molecular weights of 32 to 38 kDa, and include proteins targeted to the arnyloplasts
in rye grass (Singh et al. 1991; Knox 1993). However, the RNase activity of Phi p Vb
should be interpreted with caution, as the activity is inhibited by human placental RNase
inhibitor, which was previously found to inhibit only animal RNase A-type proteins
among RNase groups tested (Lee and Vallee 1993). The sequences of group V allergens
exhibit no readily apparent similarity to proteins in the RNase A superfamily. (Similar

concerns about the use of human placental RNase inhibitor are also associated with the

l4

cucumber S-like RNase cusativin (Rojo et al. 1994a), as well as with one of the Bet v I
studies (Bufe et al. 1996), since the PR-lO proteins also have no obvious sequence
similarities to RNase A). Whether the group V allergens comprise another major RNase

family has yet to be determined.

Bifunctional Nucleases

Although many plant enzymes with the characteristics of nuclease I enzymes have
been identified, little sequence information is available to confirm the relatedness of these
proteins. However, there is a limited region of sequence identity between two nucleases
from distantly related plants. One of these proteins is a 39 kDa nuclease I secreted from
barley aleurone layers (Brown and Ho 1987). The secretion of this nuclease is induced
by gibberelic acid (Brown and Ho 1986), a plant hormone which induces the secretion of
a range of hydrolytic enzymes from aleurone layers to mobilize seed endospenn reserves
for the germinating seedling (Jacobsen et al. 1995). Another nuclease, from zinnia, is
induced during the differentiation of xylem elements (Thelen and Northcote 1989).
Although the zinnia nuclease technically cannot be classified as a nuclease I (Wilson
1982) due to its 43 kDa molecular weight, its N-terminal sequence is similar to that of the
barley nuclease, implying that these two enzymes are related. It has been noted (Fraser
and Low 1993) that the partial barley nuclease sequence is similar to the amino-termini of
81 and P1 nuclease from the fungi Aspergillus oryzae (Iwamatsu et al. 1991) and
Penicillium citrum (Maekawa et al. 1991), respectively. These firngal nucleases are part

of a large family of single-strand-specific bifunctional (degrade both RNA and DNA)

15

nucleases (Gite and Shankar 1995) based on their biochemical properties. Mung bean
nuclease (Laskowski 1980) is also considered to be a member of this family. The
biochemical characteristics that define this group (Gite and Shankar 1995) are much
broader than those that define the plant nuclease I enzymes (Wilson 1982), and nuclease I
enzymes could easily be classified as part of the single-strand-specific nuclease family.

(See Gite and Shankar, 1995, for an extensive discussion of the biochemical properties of
the enzymes of this family.) Recently, a protein with limited homology to nuclease P1
has been purified from spinach chloroplasts (Yang et al. 1996); this protein will be
discussed later.

In Arabidopsis, nucleases shown to have some of the properties of nuclease I
enzymes have been identified using activity gels (Yen and Green 1991). A doublet of
about 33 kDa that appears on both RNase and DNase activity gels led to the suggestion
that the same 33 kDa enzymes can degrade both RNA and DNA. Recently this idea was
confirmed through the analysis of altered RNase profile (arp) mutants of Arabidopsis.
Six arp mutants that either lack or overproduce one or both of the 33 kDa doublet RN ase
activities were examined on DNase activity gels and shown to exhibit phenotypes
identical to those on RNase activity gels (M.L. Abler, P.J. Green, manuscript in
preparation). This result demonstrates genetically that the 33 kDa doublet represents a
pair of bifunctional nuclease activities. The availability of these mutants may help
elucidate the biological functions of these nucleases and reveal whether they are encoded

by the same gene.

16

In addition to the nuclease I enzymes mentioned earlier (in the "Classifications
based on Early Biochemical Work" section), a number of activities with some similarities
to a tobacco pollen nuclease I (Matousek and Tupy 1984) have been identified in pollen
from various species, with a nuclease from Pinus nigra best characterized (Matousek and
Tupy 1985). Likewise, a nuclease with similar properties was identified in tobacco
anthers (Matousek and Tupy 1987). Single—strand-specific nucleases have recently been
found in spinach (Strickland et al. 1991), scallion (Uchida et a1. 1993), wheat chloroplasts
(Monko et al. 1994), and pea seeds (Naseem and Hadi 1987) and chloroplasts (Kumar et
al. 1995). It will be interesting to determine if the nucleases described in this section, the
plant enzymes classified as nuclease I mentioned earlier, and the single-strand-specific

nucleases from fungi and plants are members of the same or multiple molecular families.

l7

REGULATION AND FUNCTIONS OF PLANT RNASES

Phosphate Remobilization

Several S-like RNases have been shown to be upregulated in response to
starvation for inorganic phosphate (P,). The first of these reports showed that RNase LE
from tomato is secreted in response to P, limitation (Niirnberger et al. 1990).
Subsequently, increased levels of the other tomato S-like RNases, LX, LVl, LV2, and
LV3, all intracellular enzymes, were found during Pi starvation (Loffler et al. 1992).
Most of this work was done at the protein level, but recently starvation for this nutrient
has been shown to induce RNases LE and LX at the mRNA level (Keck et al. 1995).
These results mirrored previous reports of two Arabidopsis RNase genes, RNS] (Bariola
et al. 1994) and RNS2 (Taylor et al. 1993), which were also found to be Pi-starvation
inducible. In particular, the RNS] mRNA is dramatically upregulated from a low basal
level (Bariola et al. 1994). This induction of RNS] also appears to occur at the protein
level, because an RNase activity that comigrates with RN81 produced in yeast increases
in parallel with RNS] mRNA (Bariola et al. 1994; OJ. Howard and P.J. Green,
manuscript in preparation). In addition, Pi starvation has recently been found to induce
the gene for Nicotiana alata RNase NE (Dodds et al. 1996a). However, at least three 8-
like RNase genes, RNS3 of Arabidopsis (Bariola et al. 1994) and ZRNaseI and ZRNaseII
of Zinnia (Ye and Droste 1996), do not respond to P, limitation, so Pi-starvation
inducibility should not be considered characteristic of the S-like RNases. Limitation for

P, also leads to a large increase in the activity of a surface membrane-associated nuclease

18

in the trypanosome Crithidia luciliae (Gottlieb et al. 1988). Although the sequence of
this protein is not available, the nuclease is considered similar to plant nuclease I
enzymes due to its enzymatic properties (Neubert and Gottlieb 1990). It is conceivable
that plant nuclease I genes could also be induced by P, starvation. Indeed, extracts of
plants grown on Pi-deficient medium exhibit increased RNase activity on activity gels in
the area of the 33 kDa nuclease doublet (refer to Bariola et al. 1994, Figure 10).

It has been proposed that under Pi-limiting conditions, RNases could degrade
RNA in conjunction with phosphatases and phosphodiesterases to release Pi, making it
available for the plant to use in other processes (Glund and Goldstein 1993). This
response is likely only a part of a broader effort by the plant to optimize Pi availability
under conditions of scarcity (Goldstein et al. 1989). RNases could increase the efficiency
of the plant to scavenge P, in several ways. RNases secreted from roots into the soil
could make Pi previously sequestered in RNA in organic matter in the soil available for
uptake. The role of scavenging P, from RNA in the growth environment is believed to be
a main function of two fungal nucleases (Fraser and Low 1993). Extracellular RNases
within the plant could rescue P, from RNA that arises in the extracellular space from cells
that have lysed due to senescence, damage or programmed cell death (discussed below).
Finally, vacuolar RNases may participate in some aspect of intracellular RNA
degradation, since short pieces of RNA exist in the vacuoles of plant cells (Abel et al.
1990). It is generally recognized that one main function of the plant vacuole is the
turnover of cellular macromolecules, analogous to animal lysosomes (Boller and

Wiemken 1986; Wink 1994). Since autophagy of small amounts of cytoplasm may occur

19

in plant cells (Boller and Wiemken 1986), this process may be a route for vacuolar uptake
of cytoplasmic RNA. An increase in the RNase concentration in the vacuole could speed
the process of the recycling of components of RNA.

Remobilization of Pi may take place during normal plant development. In barley
seeds, a nuclease I is secreted from the aleurone layer upon treatment with gibberellin
(Brown and Ho 1986), a hormone associated with seed germination (Jacobsen et al.
1995). The aleurone layer lays outside the endosperm, which stores macromolecules
such as proteins and carbohydrates. Germination triggers the secretion of several
hydrolytic enzymes (Jacobsen et a1. 1995), which release nutrients in forms able to be
used quickly by germinating seedlings. The nuclease I is proposed to degrade nucleic
acids in endosperm, in conjunction with acid phosphatases, to release nucleosides and
phosphate for use in new RNA synthesis in the seedling (Brown and Ho 1986). The
widespread increase in RNase activities in germinating seeds (Farkas 1982) suggests that

this may be a universal phenomenon in plants.

Senescence

The effect of senescence on plant RNase activity has been studied extensively. In
general, RNase activities increase in plants during senescence, but in different systems
the timing and extent of the increases vary (Farkas 1982). Many of these variations are
likely due to differences in systems and experimental design (e.g. studies in attached
leaves or excised leaves), but different patterns of induction may also contribute (Farkas

1982). It is now clear that senescence dramatically upregulates individual RNase

20

activities and genes. In wheat leaves, single-strand-specific nuclease activity increases
during senescence (Blank and McKeon 1989), as does the activity of three RNases of 20
to 27 kDa (Blank and McKeon 1991b), monitored both biochemically and in activity
gels. The three Arabidopsis S-like RNase genes, RNS], RNSZ, and RNS3, are each
induced to different extents in leaves during senescence: RNS] mRNA levels increase
only slightly (Bariola et al. 1994), whereas those of RN52 and RNS3 increase more
dramatically (Taylor et al. 1993; Bariola et al. 1994). RNS2 is also known to be induced
in senescing petals (Taylor et al. 1993).

During senescence, cellular structures are disassembled and macromolecules in
certain plant organs are broken down, freeing nutrients for relocation to other organs
(Stoddart and Thomas 1982). This process is thought to occur to conserve minerals and
nutrients for reuse. The breakdown and redistribution can occur both during vegetative
growth, such as the rescue of minerals from senescing cotyledons, and also during
reproductive growth, when in some plants all organs except the reproductive structures
senesce and the vegetative organs serve as a source of nutrients for the reproductive
structures (Noodén 1988a). RNases are likely some of the variety of hydrolytic enzymes
induced during senescence (Borochov and Woodson 1989) that facilitate the breakdown
of cellular components. The actions of RNases during senescence could lead to the

recycling of Pi.

21

Cell Death Pathways

Programmed cell death is an essential part of developmental patterns and
physiological processes in many organisms (V aux 1993). In plants, cell death pathways
are only beginning to be investigated (Greenberg 1996). Programmed cell death is
associated with sex determination in maize (DeLong et al. 1993), the hypersensitive
response (HR) against plant pathogens (Greenberg et al. 1994), and possibly pollination,
senescence, and various developmental processes (Greenberg 1996). One of the best
studied processes regarding cell death in plants is the differentiation of xylem, a major
component of the plant vascular system. When isolated leaf mesophyll cells from zinnia
are cultured in the presence of appropriate concentrations of the hormones auxin and
cytokinin, they differentiate synchronously into elongated, lignified tracheary elements,
the building blocks of the xylem (Fukuda 1996). The last part of the differentiation
process involves strengthening of the cell wall and hydrolysis of the end walls between
two differentiating cells to form a tube. Vacuoles lyse several hours after the secondary
cell wall is formed, leading to degradation of cytoplasmic macromolecules. Finally, lysis
of the protoplast occurs so that the cell wall can serve as a link in the channel that forms
the xylem (F ukuda 1996).

A single-strand-specific nuclease of 43 kDa with homology to a nuclease I of
barley (Brown and Ho 1987) appears during xylogenesis in the zinnia system (Thelen and
Northcote 1989). In addition, several RNase activities of 17 to 25 kDa appear in zinnia
cell extracts, and a 37 kDa nuclease accumulates in the culture medium. One of these

RNases may correspond to ZRNase I, an S-like RNase with a predicted molecular weight

22

of 27 kDa, whose cDNA was isolated from zinnia mesophyll cells differentiating into
xylem elements (Ye and Droste 1996). In zinnia cultured mesophyll cells, its mRNA first
appears at high levels after about 48 hours of culture in differentiation-inducing medium,
relatively late in the differentiation process (Ye and Droste 1996). This result was
confirmed by tissue-print hybridization, in which ZRNaseI mRNA appears in
differentiating xylem elements of stems (Ye and Droste 1996). A protease gene is also
strongly induced at 48 hours of culture (Ye and Varner 1993). It is possible that these
and other hydrolytic enzymes are involved either in killing the cell directly or in
degrading cytoplasmic components during and after lysis in order to clear the xylem
channel and facilitate nutrient reutilization. Nucleases could be involved in RNA
degradation as well as the fragmentation of DNA that is one of the hallmarks of apoptotic
cell death in animal systems (Zhivotovsky et al. 1994). Nuclear DNA fragmentation has
been observed in pea root xylem cells undergoing cell death (Mittler and Lam 1995a).

Less is known about the association of RNases and nucleases with other cell death
pathways in plants. However, recent studies indicate that anther nucleases are highest
during the first microspore division in tobacco (Matousek et al. 1994). This led to the
suggestion that anther nucleases could participate in tapetal cell degeneration (Matousek

et al. 1994).

Defense Against Pathogens
Increases in RNase activities in diseased plants are well-documented (Farkas

1982; Green 1994; Bama et al. 1989; Lusso and Kuc 1995). Several roles can be

23

proposed for RNases in plant disease and defense. First, if RNases are involved in cell
death pathways in plants, as discussed in the previous section, they could play a role in
the hypersensitive response of plants, which appears to involve programmed cell death
(Greenberg et al. 1994). The HR involves death of plant cells shortly after pathogen
infection in the immediate vicinity of the infection site, and the localized cell death is
thought to contribute to the resistance of the plant to the disease (Keen 1992). Recently a
DNase activity has been found to be induced in tobacco nuclei during cell death due to
the HR (Mittler and Lam 1995b). This DNase, NUCIII, has been characterized as an
endonuclease that cleaves both single-stranded and double-stranded DNA, but it has not
been reported whether the protein has RNase activity (Mittler and Lam 1995b). It seems
probable that DNA fragmentation occurs during the HR and the NUCIII could play a role
in this process. In tobacco, an HR-like response that was for some reason induced by
overexpression of a bacterial proton pump gene provided evidence for DNA
fragmentation (Mittler et al. 1995).

A second role for plant RNases during pathogen attack could be to act as defense
proteins in tissues potentially susceptible to infection. For example, the pistil, which is
penetrated by the pollen tube during the fertilization process, is rich in extracellular
nutrients that should make it susceptible to pathogen invasion. However, it is rarely
infected. This resistance may be due to defense-related proteins that are present
extracellularly in the pistil, such as proteinase inhibitors (Atkinson et al. 1993). Several
Tz-type RNase genes have been shown to be expressed in pistils, so they are also

candidate defense proteins. These include X2 of Petunia inflata (Lee et al. 1992), RNS2

24

of Arabidopsis (Taylor et al. 1993), and NE of Nicotiana alata (Dodds et al. 1996a). It
has been proposed that gametophytic self-incompatibility may have arisen via the
recruitment of defense-related pistil RNases (Lee et al. 1992). Extracellular RNases may
also play a part in the plant's defense against RNA viruses. Finally, RNases are known to
accumulate in plant vacuoles (Farkas 1982; Wilson 1982). The defense-related proteins
chitinase and [3-1,3-glucanase, which can degrade fungal cell walls, are also known to be
sequestered in vacuoles, increasing in abundance during pathogen attack (Mauch and
Staehelin I989). Mauch and Staehelin (1989) propose a model in which accumulation of
large amounts of these proteins in vacuoles is an advantage: when fungi invade, cells
lyse either due to the HR or direct pathogen invasion of the cell. Upon lysis, fungal
hyphae would be flooded with defense-related proteins in high enough concentrations to
lyse the hyphae. RNases may well be components of this onslaught of hydrolytic defense
enzymes released from the vacuoles upon cell lysis.

Plants interpret mechanical wounding as a signal of attack, since pathogen
infection and chewing by insects or other herbivores often result in wounding of tissues.
Wounding induces defense-related genes in plants, such as those of proteinase inhibitors
(Ryan 1990), peroxidases (Bowles 1990), and chitinases (Bowles 1990). Wounding is
known to induce rapid increases in RNase activities (Farkas 1982). One zinnia S-like
RNase gene, ZRNaseII, is rapidly induced upon mechanical wounding (Ye and Droste
1996). In Arabidopsis, wounding of stems resulted in induction of an RNase activity of
about 34 kDa (M. Saitoh, M.L. Abler, P.J. Green, unpublished). Like the 33 kDa

bifunctional nuclease activities discussed in the "Bifunctional Nucleases" section, the 34

25

kDa RNase comigrates with a DNase activity induced under the same conditions, so the
protein is likely a bifunctional nuclease. Further examination of this enzyme and its

wound-inducibility may provide insights into the roles of RNases in defense responses.

RNA Processing and Decay

Nuclear Activities: A number of events that take place in plant nuclei likely involve
RNases, but very few such enzymatic activities have been identified. Presumably the
major degradative process in the nucleus is the decay of introns and other sequences
removed from precursors of mature mRNAs, rRNAs, and tRNAs. There are a few early
reports of nuclear-associated RNase activities that might be involved in these processes in
plants (reviewed in Farkas, 1982), but this area of research should be revisited because it
is underdeveloped not only in plants but also in other eukaryotes (Stevens 1993; Ross
1995). The most convincing data for an RNase located in the nucleus comes from work
on the 7-2/MRP RNA, which is known to be the RNA component of RNase MRP in
mammalian cells. In plants and mammalian cells, 7-2/MRP RNA is found in nucleoli
where it is likely to be involved in rRNA processing (Kiss and Filipowicz 1992; Kiss et

al. 1992; Morrissey and Tollervey 1995; Lygerou et al. 1996).

Chloroplast and Mitochondrial Activities: Most of the organellar RNase activities that
have been identified participate in the maturation of 5' and 3' ends of chloroplast and
mitochondrial transcripts such as tRNAs. The ribonucleoprotein RNase P is responsible

for the processing of the 5' end of pre-tRNAs (Altman et al. 1993); in plants, RNase P

26

enzymes have been identified in chloroplasts (Gegenheimer 1996) and mitochondria
(Marchfelder and Brennicke 1994). Interestingly, the spinach chloroplast RNase P may
not contain an RNA component (Wang et al. 1988; Gegenheimer 1996). Maturation of
the 3' end of pre-tRNAs in eukaryotes involves an endonucleolytic cleavage in most
cases, in contrast to the prokaryotic mechanism (Deutscher 1993b). 3' tRNA processing
activities are detectable in plant nuclei, mitochondria, and chloroplasts (Oommen et al.
1992; Marchfelder and Brennicke 1994; Gegenheimer 1996). Preliminary
characterization of RNase activities that affect 3' end maturation has also been achieved
(Chen and Stern 1991). These include one or more 3' to 5' exoribonucleases and an
endoribonuclease. The latter has been shown to cleave the spinach petD mRNA at the
termination codon and at the mature RNA 3' end. Recently, an encoribonuclease has
been purified from this system and its cDNA cloned (Yang et al. 1996). Interestingly, the
protein has a region of similarity to nuclease P1 (Maekawa et al. 1991), placing it in the
category of plant nucleases. In maize, a nuclear mutation, crpI, has been isolated that
blocks the processing of the polycistronic precursor of petD mRNA, which appears to
inhibit translation (Barkan et al. 1994). It is possible that the CRPI gene encodes a

processing RN ase or a protein that regulates such an activity.

Activities Implicated in Cytoplasmic mRNA Decay: Little is known about the RNase
activities that facilitate the degradation of most plant mRNAs, since most degrade
without generating easily identifiable intermediates. Two exceptions are the soybean

rch (Tanzer and Meagher 1994; Tanzer and Meagher 1995), and the PHYA (Higgs and

27

Colbert 1994) mRNAs. Discrete fragments of the rch mRNA are observed in vivo and
are produced in an in vitro decay system. The structures of these fragments suggest that
they are produced by a stochastic endonuclease followed by exonuclease digestion in the
5' to 3' or 3' to 5' direction (Tanzer and Meagher 1994; Tanzer and Meagher 1995). In
contrast, a continuous population of lower molecular weight fragments is observed for the
PHYA mRNA, rather than discrete intermediates. Characterization of these fragments
indicates that they most likely arise through the action of 5' to 3' and 3' to 5'
exoribonucleases, although endoribonuclease models cannot be ruled out (Higgs and
Colbert 1994). In yeast, many mRNAs are known to be degraded by a pathway involving
deadenylation, possibly by a poly(A) nuclease, followed by decapping and digestion by
XRNI, a 5' to 3' exoribonuclease. This type of pathway may explain the decay of about
25% of the PHYA mRNA, but the remainder (Higgs and Colbert 1994), as well as the
rch mRNA (Tanzer and Meagher 1995), appears to degrade independent of
deadenylation. Deadenylation-independent mRN A decay pathways, some of which
involve cleavage by sequence-specific endoribonucleases (Brown and Harland 1990;
Binder et al. 1994), also exist in yeast and animal systems (Beelman and Parker 1995).
Recent reports have identified sequences that can trigger rapid decay of reporter
mRNAs in plants. These include the 3' untranslated region (UTR) of the Arabidopsis
SA UR-ACI transcript (Gil and Green 1996), a dimer of the DST element (Newman et al.
1993), which is conserved among the 3' UTRs of unstable SA UR transcripts (McClure et
al. 1989a), and AUUUA repeats (Ohme-Takagi et al. 1993), which also trigger mRNA

decay in mammalian cells (Shaw and Kamen 1986; Vakalopoulou et al. 1991). However,

28

it is unknown whether these sequences serve as endonuclease-sensitive sites in plants or
whether they trigger exonuclease digestion. By examining the effect of these instability
elements in the Arabidopsis mutants that alter the RN ase profile (arp mutants, described
in the "Plant Nuclease" section) and other mutants that may become available, it may be
possible to identify some of the RNases involved in general and specific mRNA decay
pathways. In addition, in vitro systems (Byrne et a1. 1993; Tanzer and Meagher 1994;
Tanzer and Meagher 1995) may facilitate purification of RNases that act on specific
mRN A transcripts, particularly in the case of the PHYA and rch transcripts.

Finally, it has been suggested that antisense RNA effects in plants may be
mediated in part by the action of a double-stranded RNA degrading activity that is
presumed to degrade the sense-antisense hybrid (Nellen and Lichtenstein 1993). This
idea is based on the observation that in many cases accumulation of the sense RNA is
diminished in plants engineered to produce a corresponding antisense RNA (Nellen and
Lichtenstein 1993; Bourque 1995). In one case, this effect has been shown to be due to
rapid decay of the sense RNA (Jiang et al. 1994). Although a considerable amount of
dsRNase activity in plants may be extracellular (Matousek et al. 1994), perhaps the
enzyme corresponding to an Arabidopsis expressed sequence tag (Genbank accession
number 218464) (Hofte et al. 1993), which has homology to RNase III, a bacterial

intracellular dsRNase (Robertson 1982), is involved in this process.

29

DISSERTATION TOPIC AND THESIS OVERVIEW

The relationships among different RNases have become much clearer now that
many have been cloned and sequenced. It seems apparent that the S-like RNases are
among the major, if not the major, class of RNA-degrading enzymes in higher plants.
Based on their gene expression and activity levels, the S-like RNases are likely to
participate in fundamental physiological processes such as senescence, phosphate
starvation responses and cell death pathways. My thesis project has involved
characterizing three S-like RNases of Arabidopsis thaliana and using techniques
designed to provide insight into their roles and purposes in plants. Chapters 2 and 3
detail my contributions to the initial analysis of S-like RNase structure and regulation
mentioned in this introduction. My efforts to determine the subcellular location of RN82
are described in Chapter 4, and the generation of transgenic plants with altered levels of

RN81 and RN82 is described in Chapter 5.

CHAPTER 2

RN82: A SENESCENCE-ASSOCIATED RNASE OF ARABIDOPSIS THAT
DIVERGED FROM THE S-RNASES BEFORE SPECIATION

Portions of this chapter were published in Proceedings of the National Academy of
Sciences USA:

Taylor CB, Bariola PA, del Cardayré SB, Raines RT, Green PJ (1993) RNS2: A

senescence-associated RNase of Arabidopsis that diverged from the S-RNases before
speciation. Proc. Natl. Acad. Sci USA 90:51 18-5122

30

31

ABSTRACT

Several self-compatible species of higher plants, such as Arabidopsis thaliana, have
been found to contain S-like RNases. These S-like RNases are homologous to the S-
RNases that are involved in self-incompatibility in Solanaceous species. However, the
relationship of the S-like RNases to the S-RNases is unknown, and their roles in self-
compatible plants are not yet understood. To address these questions, we have investigated
the RNS2 gene, which encodes an S-like RNase of Arabidopsis. Amino acid sequence
comparisons indicate that RN82 and other S-like RNases make up a subclass within an
RNase superfarrrily, which is distinct from the subclasses formed by the S-RNases. RN 82
is quite similar to RNases LE (Jost et al. 1991) and LX (Loffler et al. 1993) of Lycopersicon
esculentum, and to RNase NE of Nicotiana alata (Dodds et al. 1996a); both of these are
Solanaceous species. The fact that RNases LE, LX and NE are more similar to RN82 than
to the S-RNases from other Solanaceous plants indicates that the S-like RNases diverged
from the S-RNases prior to speciation. Like the S-RNase genes, RNSZ is most highly
expressed in flowers, but unlike the S-RNase genes, RN82 is also expressed in roots, stems,
and leaves of Arabidopsis. Moreover, the expression of RNS2 is increased in both leaves
and petals of Arabidopsis during senescence. Phosphate starvation can also induce the
expression of RNS2. Finally, RNS2 is induced during infection by a bacterial pathogen,
during both compatible and incompatible infection, although to a higher extent during

compatible infection. On the basis of these observations, we suggest that one role of RNS2

32

in Arabidopsis may be to remobilize phosphate, particularly when cells senesce or when

phosphate becomes limiting.

33

INTRODUCTION

The identification of three Tz/S family RNase genes in a self-compatible plant,
Arabidopsis thaliana (Taylor and Green 1991), was unprecedented. At the time, the only
RNases of this type known were either fungal enzymes or enzymes associated with
gametophytic self-incompatibility in certain Solanaceous plants. The logical first steps after
the initial identification of the three RNS genes were to isolate the full-length cDNAs,
examine their sequences, and determine their expression properties. These were intended to
give clues as to the function of this type of RNase, which we termed "S-like RNases" due to

their similarity to S-RNases, in self-compatible plants.

34

RESULTS AND DISCUSSION

RNSZ was initially identified as a PCR product amplified from an Arabidopsis
cDNA library using primers corresponding to the regions most conserved between the S-
RNases and a class of fungal RNases (Taylor and Green 1991). This PCR product was used
as a hybridization probe to isolate RNSZ cDNA clones from the same library. 39 positive
clones were initially identified in a screen of 250,000 plaques, and four were partially
sequenced. Once the clone with the longest 3' untranslated region was identified, it was
sequenced completely from both 5' to 3' and 3' to 5' directions (Figure 2-1). The nucleotide
and deduced amino acid sequences of the longest clone containing an open reading frame
have been deposited in the GenBank data base (accession number M98336) and are shown
in Figure 2-2. Analysis of several independent cDNA clones showed that. transcripts fiom
the RNS2 gene can be polyadenylated at multiple sites (Figure 2-2), a feature common to
many plant genes (Dean et al. 1986). The 19 amino acids at the N terminus of the RN82
protein are typical of a eukaryotic secretion signal sequence, as defrned by a statistical
analysis of known signal sequences (von Heijne 1986). This suggests that RNS2 is targeted
to the secretory pathway in Arabidopsis, similar to the Solanaceous S-RNases, which are
secreted enzymes.

To confirm that RN82 is indeed an RNase, the coding sequence was expressed in
Saccharomyces cerevisiae under the control of the PH05 promoter (Thill et al. 1983).
RNase activity secreted into the culture medium by yeast transformed with the vector

control (CON) or the RN82 expression construct (RN S2) was then detected following

35

 

 

 

 

 

 

A
A

Figure 2-1 - Strategy for sequencing of RNS2 cDNA. The rectangle represents the RNS2
cDNA sequence, and arrows represent individual sequencing reactions read. Each base of
the RN82 cDNA was read at least once from each direction.

36

1 ATCGAATTAAAGTCAATGGCGTCACGTTTATGTCTTCTCCTTCTCGTTGCGTGTATCGCC
1 M A 8 R L C L L L L V A C I A

V
61 GGAGCATTTGCCGGAGACGTCATCGAACTCAATCGATCTCAGAGGGAGTTCGATTATTTC
16 G A F A G D V I E L H R 8 Q R F F D Y F

121 GCTCTATCTCTTCAATGGCCTGGAACCTATTGCCGTGGAACTCGCCATTGTTGCTCCAAA
36 A L S L Q W P G T Y C R G T R H C C S K

181 AACGCTTGCTGCAGAGGCTCCGATGCTCCAACTCAATTCACAATTCATGGGTTATGGCCT
56 N A C C R G S D A P T Q F T I H G L W P

241 GACTATAACGATGGTTCGTGGCCTTCATGTTGTTATCGATCTGACTTTAAAGAGAAGGAG
76 D Y N D G 8 W P 8 C C Y R 8 D F R 8 K E

301 ATTTCAACGTTGATGGATGGTCTTGAGAAGTACTGGCCTAGTCTCAGTTGTGGTTCTCCA
96 I 8 T L M D G L R K Y W P 8 L 8 C G S P

361 TCATCATGCAATGGTGGGAAAGGGTCATTTTGGGGCCACGAGTGGGAGAAACATGGGACT
116 8 8 C N G G K G 8 F W G H R W E K H G T

421 TGTTCTTCTCCTGTTTTTCATGATGAGTATAATTACTTCCTTACCACACTTAATCTCTAC
136 C S S P V F H D I Y N Y F L T T L N L Y

481 TTGAAGCATAATGTCACGGATGTCCTTTATCAAGCTGGCTATGTTGCTTCCAACAGTGAA
156 L K H H V T D V L Y Q A G Y V A 8 N 8 R

541 AAGTATCCTCTAGGAGGTATCGTAACAGCCATTCAGAATGCATTTCATATCACCCCTGAA
176 K Y P L G G I V T A I Q N A F H I T P F

601 GTGGTTTGCAAAAGAGATGCAATCGATGAAATACGTATATGCTTCTATAAAGATTTTAAG
196 V V C K R D A I D E I R I C F Y K D F K

661 CCCAGGGACTGTGTTGGTTCACAAGATTTGACATCTAGAAAGTCATGCCCCAAGTACGTA
216 P R D C V G S Q D L T S R K S C P R Y V

721 AGTTTGCCGGAATACACGCCATTAGATGGTGAAGCTATGGTTCTGAAGATGCCAACAGAA
236 8 L P I Y T P L D G R A H V L K M P T F

781 AGAGAAGCTCTTTGAATCGGAAAAGATGGGAGCTTTGTTATCTTCTGAGAGACAATACAT
256 R E A L *

841 ACATGTCTCTGATGTTGTAACTTTACTACCAAAACCTATAAAGATTGGCTTATTTCGTTC
X
901 TATTGGATATGTATCATCATTACTGGTAAATCAAGTTTCTTTCTAATAATGTAGAAGATC

‘A St
961 AGAAAATCCATAAGAAGATATCAACATTTGAGTTCTATGGTAAAAAAAAAAA

Figure 2-2 - Primary structure and deduced amino acid sequence of RN82 cDNAs.
Deduced amino acid residues are shown in one-letter notation below the nucleotide
sequence. Triangle indicates the putative N-terminal end of the mature protein predicted by
a statistical analysis (von Heijne 1986). Arrows indicate polyadenylation sites used in
different cDNA isolates. The two putative N-glycosylation sites are underlined.

37

electrophoresis on RNase activity gels (Yen and Green 1991) (Figure 2-3). Under inducing
conditions, two bands of RN82 activity were observed that have apparent molecular masses
of 28-33 kDa, which is slightly higher than the predicted molecular mass of 27 kDa
(assuming cleavage of the N-terrrrinal signal sequence). This slight difference in molecular
mass and the presence of two RN82 bands may result from processing of the RN 82 signal
sequence at multiple sites (Ohgi et al. 1991) or differences in glycosylation (Tague and
Chrispeels 1987) that are known to affect the mobility of heterologous proteins produced in
yeast. It should also be noted that the RNase activity gels are run under nonreducing
conditions (Yen and Green 1991), which may contribute to the differences. However, both
RNase bands are specific to the RN82 clone and correlate with the induction of the PH05

promoter, as expected. This demonstrates that the RNS2 gene encodes an active RNase.

Comparison of RN 82 to Related RNases

To compare the deduced amino acid sequence of RN 82 with those of related plant
RNases, the alignment shown in Figure 2-4 was generated as described in the Materials and
Methods. The alignment demonstrates that the similarity of RN S2 to the S-RNases is
dispersed throughout the coding region (Figure 2-4). Moreover, each of the five regions
most conserved among the S-RNases [numbered C1-C5 by Kao and coworkers (Kheyr-
Pour et al. 1990; Ioerger et al. 1991) and boxed in Figure 2-4] is also evident in RN 82. At
the original time of publication of this sequence comparison (Taylor et al. 1993), RN82 was
compared to two other S-like RNases and 15 S-RNases, which comprised all of the plant

Tz/S type RNases for which sequences were known. At present, the sequences of 12 S-like

38

CON RN32
NI l NI I

43.7 -

28.9 -

18.4-

 

Figure 2-3 - Expression of RN82 in yeast. Yeast cultures transformed with the control
pWL (CON) or RNS2 (RNS2) constructs were grown under conditions that do not induce
(NI) or do induce (I) transcription from the PH05 promoter. Supematants from these
cultures were run on RNase activity gels as described in the Materials and Methods.
Positions of molecular mass markers (kDa) are shown to the left of the gel. The arrowhead
indicates the RN82 activity bands.

39

Legend to Figure 2-4 - Alignment of 8- and S-like RNase amino acid sequences. S-like
RNase sequences are LE (Jost et al. 1991) and LX (Loffler et al. 1993) of Lycopersicon
esculentum, RN81 and RNS3 of Arabidopsis thaliana (Bariola et a1. 1994), RN82 of
Arabidopsis thaliana (Taylor et al. 1993), NE of Nicotiana alata (Dodds et al. 1996a),
ZRN l and ZRN2 of Zinnia elegans (Ye and Droste 1996), MCI of Momordica charantia
(Ide et al. 1991), LCl and LC2 of Lufla cylindrica (T. Nakamura, K Sasaki, G Funatsu,
submitted), and NNUC of Nelumbo nucifera (G Day, Z Chen, T Chow, unpublished,
Genbank accession number M83668). S-RNase sequences are lStu, rlStu, and 28m of
Solanum tuberosum (Kaufinann et al. 1991), lPet, 2Pet and 3Pet of Petunia inflata (Ai et al.
1990), Ps2A, Ps3A and PslB of Petunia hybrida (Clark et al. 1990), 5Lyc of Lycopersicon
peruvianum (Tsai et al. 1992), a, Z, F11 and lNic of N. alata (Kheyr-Pour et al. 1990),
XPet and OPet of self-compatible P. hybrida (Ai et a1. 1992), 2Nic, 3Nic and 6Nic of N.
alata (Anderson et a1. 1989), 2801 and 3801 of Solanum chacoense (Xu et al. 1990), 1 18c of
S. chacoense (Saba-El-Leil et al. 1994), 5Lp of L. peruvianum (Rivers et al. 1993), lle,
12Lp and 13Lp of L. peruvianum (Chung et al. 1994), fMdo of Malus domestica (Sassa et
al. 1996), 2Mdo and 3Mdo of M domestica (Broothaerts et al. 1995), 2Pp and 4Pp of Pyrus
pyrifolia (Norioka et al. 1995), and 2Ant, 4Ant and 5Ant of Antirrhinum species (Xue et a1.
1996). X2 of P. inflata (Lee et al. 1992) is not an S-RNase but was placed with this group
because it is more closely related to the S-RNases than the S-like RNases (Lee et al. 1992).
The sequences are aligned and numbered from predicted mature N-tennini of the N. alata
S-RNases (see Haring et al. 1990). Heavy-bordered boxes enclose conserved regions C1-
C5 (Ioerger et al. 1991). Light shading indicates residues that are identical or functionally
identical in at least 35 of the 47 sequences. Light-bordered boxes enclose residues that are
identical or functionally identical in at least 9 of the 12 S-like RNase sequences but not
highly conserved among the S-RNases. Dark shading indicates residues that are identical or
functionally identical in at least 26 of the 35 S-RNase sequences but not highly conserved
among the S-like RNases. Functionally identical residues are grouped as follows: A, S, T;
I, L, M, V; H, K, R; F, W, Y; D, E; Q, N.

40

m0

$3838 28 265 0872 85% can -m .«o €08ng - YN Pia...—

NU FU

 

Legend to Figure 2-4 - Alignment of 8- and S-like RNase amino acid sequences (second
page). For the complete legend please see page 39.

42

mquHem<¢mummmmmmH.
m."

acmmmemzzq>2<mwnqmewmmqmo
mamas
mmnmanzemmmmumu.
mmmuma.
«sauna.
_
ow~ onN

 

 

    

CNN

 

Gwen 288$ 80858 Eon 8.58 0872 85% v5 -m we “Sam? - TN 953%

m0 vU

43

RNases have been reported, and the number of completely or partially sequenced S-RNases
is approaching fifty. Figure 2-4 is an updated sequence comparison, including all of the
known S-like RNase sequences and a representative majority of the S-RNase sequences.
All of the features noted in the original published comparison (Taylor et al. 1993) are still
valid with the newer sequences added.

For example, the histidine residues at positions 42 and 108 are absolutely conserved
in all of these RNases. These residues have been shown to be important for catalysis in
RNase T2 of Aspergillus oryzae (Kawata et a1. 1990). Other groups of notable residues are
the conserved cysteines at positions 58 and 111 and positions 177 and 217 that have been
shown to form disulfide bonds that are critical for maintaining the structure of RNase T2
(Kawata et al. 1988). Residues at a number of positions are conserved in mutually
exclusive sets, either only among the S-RNases or only among the S-like RNases. Many of
the residues that distinguish these subclasses are clustered and therefore correspond to
regions within the S-RNases and the S-like RNases that are potentially related to their
disparate functions. The majority of the residues that are conserved among the S-like
RNases but absent from the S-RNases fall between the histidine residues of the putative
active site (Figure 2-4). Some of the residues that are highly conserved among the 8-
RNases but not highly conserved in the S-like RNases are clustered between positions 134
and 167 (Figure 2-4). This region may constitute a domain required for the specialized

fLlrrctions of the S-RNases in self-incompatibility.

44

The most striking difference between the amino acid sequence of RN82 and those
of most of the other enzymes is the C-tenninal extension of 20 arrrino acids. This sequence
has some features in common with C-tenninal vacuolar-targeting signals from other plant
proteins, such as lectins and seed storage proteins--namely, a preponderance of hydrophobic
amino acids, especially in stretches of three to four (Chrispeels and Raikhel 1992; Neuhaus
1996). This idea will be discussed further in Chapter 4.

A broader illustration of the relationship of RN82 to the other related RNases was
obtained by constructing a gene genealogy based on the deduced amino acid sequences in
Figure 2-4. Once again, the genealogy shown in the original publication of these results
(Taylor et al. 1993) has been updated with new RNases (Figure 2-5), but the same trends
are observed even with the many new sequences added. In the original figure, the S-
RNases and S-like RNases were placed in separate lineages, which indicates that they form
distinct categories of RNases in plants. This is consistent with the theory that the S-RNases
and the S-like RNases derive from the same ancestral RNase, the S-RNases having acquired
a specialized function in self-incompatibility (Taylor et al. 1993). In the updated figure, the
S-like RNases still form a separate lineage. The S-RNases also form separate lineages;
however, there are now three groups instead of one. These correspond to RNases from
three plant families: the Solanaceae, the Rosaceae (apple and pear RNases), and the

Scrophulariaceae (snapdragon RNases) (Figure 2-5). In the original figure, the lineage
l‘eferred to as the S-RNase lineage was made up of only Solanaceous S-RNase sequences,
thus forming only one lineage (Taylor et al. 1993). In contrast to this clear separation of

Proteins from different plant families, the S-like RNase lineage contains RNases from five

45

 

i 1.01
MC

L02

 

 

 

 

 

ZRN1

-— l-| i—Chi S-like RNases

RNS1
ZRN2
RN83
NNUC
RN82
1 Stu
r1 Stu

i 11Sc

— f1 1
2Pet

PSZA
Ps3A
3Pet
1 Pet

-——1: 25‘“
5Lyc

XPet
_{

 

 

 

 

 

 

 

 

 

 

F]

 

 

l J.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1— P818
__ OPet

13Lp
1Ni<= S-RNases
2Nic

_[
Ll _. — _‘|:{ 3Nic

 

 

 

 

 

 

 

 

 

 

6Nic

 

 

 

12Lp
SSol
2So|
5Lp

[ 2Ant
5Ant
4Ant
4Pp
Fl tho
L[ 2Mdo

3Mdo
2Pp

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 2-5 - Gene genealogy of S- and S-like RNases. The Genetics Computer Group
Program PILEUP (Devereux et al. 1984) was used to create the dendogram. Abbreviations
of RNases are as described in the legend of Figure 2-4.

46

separate families, the Cucurbitaceae, Solanaceae, Asteraceae, Brassicaceae and
Nelumbonaceae. This grouping is notable because it implies that their sequences, and thus
presumably their function, may be evolutionarily conserved across a broad range of plant
species. This is most clearly illustrated by the Solanaceous S-like RNases, LE (Jost et al.
1991) and LX (L'offler et al. 1993) of Lycopersicon esculentum and NE of Nicotiana alata
(Dodds et al. 1996a). These RNases are placed on the same branch as S-like RNases from
four other plant families and are therefore more closely related to these RNases than to any
of the S-RNases isolated from Solanaceous species. The placing of the S-like RNases on a
separate branch of the genealogy from those of the S-RNases and the close relationship of
the Solanaceous S-like RNases to other non-Solanaceous S-like RNases strongly indicate
that these two groups of RN ases diverged prior to speciation. It should be noted that the
plant S-like RNases form a lineage separate from not only that of the S-RNases, but also
from those of other Tz-type proteins from viruses, bacteria, fungi, and animals, although
only the plant lineage is shown in Figure 2-5.

As described in Chapter 1, we have since defined placement in this lineage indicated
in Figure 2-5 as a requirement for categorization as an S-like RNase. This definition
became necessary with the discoveries in self-incompatible plants of RNases very closely
related to the S-like RNases, as well as enzymes more closely related structurally to the 8-

RNases but not playing roles in self-incompatibility.

47

Control of RNS2 Expression

As a first step toward elucidating the function of RNS2, the expression of RNS2 in
roots, leaves, stems, and flowers of Arabidopsis was investigated by Northern blotting
(Figure 2-6). Similar to the S-RNases of the Solanaceae (Cornish et al. 1987), RNS2 is
most highly expressed in flowers of Arabidopsis. However, RNSZ is also expressed in other
organs, notably leaf and stem, albeit at a much lower level than in flowers. The expression
of RNS2 in all organs that were examined implies that RN82 is a fimdamental component of
the RNA degradation machinery in Arabidopsis. To localize further RNS2 expression in
flowers of Arabidopsis, RNA was isolated from Arabidopsis pistils (stigma and style) and
petals harvested at anthesis. The results shown in Figure 2-6 demonstrate the RNS2 is
expressed in both of these flower organs, with slightly higher expression apparent in petals.
This is in contrast to the expression of the S-RNases, which is restricted to the gynoecium
and is most prominent in the transmitting tissue of the style (Cornish et al. 1987). Thus, in
analogy with the sequence data described above, there are some distinct similarities
between the expression properties of RNS2 and the S-RNases but also some significant
differences.

The presence of RNS2 transcripts in petals indicates that RNS2 may contribute in
part to the increase in RNase activity that is known to occur in petals during senescence in
plants (Matile and Winkenbach 1971). RNA was therefore isolated from Arabidopsis
petals harvested at anthesis [stages 13 and 14 in Smyth et al. 1990] and during senescence
[stage 16 in Smyth et al. 1990] and probed for the RN82 transcript. A clear increase in

RNS2 expression in senescing petals was observed in these experiments, as shown in Figure

48

 

Figure 2-6 - Expression of RN82 in different organs of Arabidopsis. (A) Samples of 12 ug
of total RNA isolated from roots (R), stems (8), leaves (L), and flowers (F) were hybridized
to the RNS2 probe following Northern blotting. (B) Samples of 5 ug of total RNA isolated
from style and stigma (St) and petals (P) dissected from Arabidopsis flowers were subjected
to Northern blotting and hybridization to the RNS2 probe. The RNS2 transcript is indicated
by the arrowhead.

These experiments were performed by Crispin Taylor

49

A B

PETAL LEAF PHOS
N S N S + —

. ..« uQ<RNs2

C D

nu.-- < .. <e|F4A

Figure 2-7 - Induction of RNS2 during senescence and phosphate starvation. (A) Samples
of 5 ug of total RNA isolated from nonsenescing (N) and senescing (S) petals and leaves, as
indicated above the lanes, were hybridized to the RNS2 probe following Northern blotting.
The RNS2 transcript is indicated by the arrowhead. (B) Etiolated Arabidopsis seedlings
were incubated in the presence (+) or absence (-) of phosphate as described in the Materials
and Methods. Samples of 10 ug of total RNA isolated from these seedlings were
hybridized to the RNS2 probe following Northern blotting. (C and D) The Northern blots
shown in A and B were stripped of the RNS2 probe and hybridized with an eIF 4A probe
(see Materials and Methods) to generate C and D, respectively. The arrowhead shows the
eIF 4A transcript.

These experiments were performed by Crispin Taylor

50

2-7A. The increase in RNS2 expression during senescence is even more evident in leaves,
where the basal level of the RNS2 transcript is lower than in petals (Figure 2-7A). As a
control, the blot shown in Figure 2-7A was stripped of the RNS2 probe and hybridized with
a probe for the translation factor e1F4A from Arabidopsis (see Materials and Methods). The
levels of the e1F4A transcript are approximately equal in each pair of samples (Figure 2-
7C). These results confirm that the senescence-induced accumulation of RNS2 mRNA is a
specific effect. RN82 is therefore likely to be a component of the major change in gene
expression that is associated with the onset of senescence (Woodson 1987). This includes
the induction of a large number of hydrolytic enzymes (Borochov and Woodson 1989),
which are thought to be involved in the recycling of nutrients from the vegetative to the
reproductive organs (Kelly and Davies 1988).

A role for RNase LE in phosphate starvation rescue has been suggested because it is
secreted from tomato cells following phosphate starvation (Niirnberger et al. 1990; Loffler
et al. 1992). To test whether phosphate limitation could induce the expression of the RN82
gene, RNA was isolated from Arabidopsis seedlings that had been placed in phosphate-free
medium or in medium containing 1.25 mM phosphate for 12 hours. As shown in the
Northern blot in Figure 2-7B, accumulation of the RNS2 transcript increases markedly
following phosphate limitation, while the level of the e1F4A transcript remains constant
(Figure 2-7D). Examples of other enzymatic activities induced by phosphate starvation
have been described in plants (Duff et al. 1991; Usuda and Shimogawara 1992), but in these

cases it is unknown whether regulation is exerted at the mRNA or protein levels.

51

Ribonuclease activities are known to be higher in diseased plants than in healthy
plants (F arkas 1982; Green 1994). To investigate whether an increase in RN 82 activity
may contribute to this defense response, Arabidopsis plants were inoculated with one of two
types of pathogens: avirulent pathogens, which induce the plant defense response termed
the hypersensitivity response, or virulent pathogens, which can avoid activating the defense
response and thus infect the plant. Arabidopsis leaves were infiltrated with Pseudomonas
syringae pv. syringae, which is avirulent in this plant, P. syringae pv. maculicola, which is
virulent in this plant, or buffer. After 48 hours, leaves infiltrated with P. syringae pv.
syringae exhibited classic hypersensitivity response symptoms of necrotic dry lesions with
the surrounding tissue appearing healthy. Leaves infected with P. syringae pv. maculicola
had larger, water-soaked lesions and were chlorotic in the surrounding tissue. Control
leaves appeared normal. Levels of RNS2 mRNA appear to increase slightly over control
leaves upon inoculation with the P. syringae strains (Figure 2-8), 1.7 times control with P.
syringae pv. syringae and 2.7 times control with P. syringae pv. maculicola. In another
experiment, levels of RNS2 mRNA were examined in Arabidopsis leaves infiltrated with
similar pathogens and harvested at different time points. P. syringae pv. maculicola was
again used as the virulent strain, but the avirulent strain used was P. syringae pv. tomato.
Leaves were infected with various titers of bacteria and harvested fi'om 6 to 48 hours after
infection. RNS2 was once again induced slightly by attack by pathogens of both strains
tested (Figure 2-9), with the maximum level of induction 2-2.4 times control in all cases.
The avirulent P. syringae pv. tomato strain produced the highest levels of RN82 mRN A at

24 hours after infection. Maximal levels of RNS2 mRNA in plants infected with P.

52

 

Figure 2-8 - Induction of RNS2 by pathogens. Leaves of 4-week-old Arabidopsis plants
were infiltrated with (1) buffer, (2) Pseudomonas syringae pv syringae, or (3) P. syringae
pv maculicola. After 48 hours infiltrated leaves were harvested and their RNA was
isolated. The RNA was subjected to Northern blot analysis and the resulting blot
hybridized to the RN82 probe as described in Figure 2-6.

53

RN52

hr $151.22.“?

Control :

1065 - Qt
Low

1065 .. .+..
High

4236 . g . . ,.,

Low

4236 ,_
High .

Figure 2-9 - Time course of RNS2 induction by pathogens. Leaves of 4-week-old
Arabidopsis plants were infiltrated with buffer (control) or low titer (OD600=0.01) or high
titer (OD600=O.1) of P. syringae pv tomato strain 1065 or P. syringae pv maculicola strain
4236. Leaves were harvested at the times indicated and their RNA was isolated. The RNA
was subjected to Northern blot analysis and the blot hybridized to the RNS2 probe as
described in Figure 2-6.

This experiment was performed by X innian Dong

54

syringae pv. maculicola depended on the titer of bacteria infiltrated, with a higher titer of
infiltrated cells producing a later peak. Overall, there is a modest but reproducible
induction of RN82 in Arabidopsis by all of the pathogens used in both experiments. The
induction seems to appear only afier 24 hours, suggesting that RNS2 is not directly induced
by a specific bacterial component. It is possible that a peak in RN82 induction may have
been missed because it occurs very soon after infection, as is true for phenylalanine-
arnmonia lyase (Dong et a1. 1991). Another possibility, taking into account the induction of
RN82 by senescence seen in Figure 2-7A, is that RNS2 may be induced in infected leaves
solely due to the senescence that accompanies bacterial infection. RNS2 is induced to a
similar level during these pathogen experiments (Figures 2-8 and 2-9) as during senescence
(Figure 2-7A), and is higher during infection with the virulent pathogen, which produces

more senescence in the affected plant.

Conclusions

RNS2 was the first senescence-associated RNase gene identified in higher plants.
Senescence-associated RNase activites have been demonstrated in a number of plant species
(Matile and Winkenbach 1971; Blank and McKeon I991b), but it is not known whether
they are encoded by S-like RNase genes. The idea that an RNase can be induced to rescue
phosphate from RNA in the plant is supported by the induction of RN82 during senescence,
phosphate starvation, and pathogen attack, during all of which the plant has a need to make

the most efficient use of the resources it has available. Plants ofien grow under phosphate-

55

limiting conditions (Fried and Brosehart 1967), so the presence of such an induction

pathway is not surprising. These ideas will be examined further in Chapter 3.

56

MATERIALS AND METHODS

Isolation and Sequencing of RNS2 cDNAs

The Arabidopsis cDNA library in lZAP (Short et al. 1988) that was initially used to
detect the RNS genes by PCR amplification (Taylor and Green 1991) was used to screen for
RNS2 cDNA clones by plaque hybridization (Sambrook et al. 1989) using the RNS2 PCR
product as a probe. Positive plaques were purified; the cDNA clones were converted into

plasmid form (protocol provided by Stratagene) and sequenced (Sanger et al. 1977).

Expression of RNS2 in Saccharomyces cerevisiae

An RNS2 cDNA covering the entire coding region including the signal sequence
was inserted between the yeast PH05 promoter and GADPH terminator (Rosenberg et al.
1984), and the resulting PH05-RNS2-GADPH gene was then cloned into the yeast shuttle
vector pWL (Del Cardayré et al. 1995). Transformants of Saccharomyces cerevisiae strain
BJ2168 (MATa prcI-407 prb1-1122 pep4-3 leu2 trpI ura3-52) harboring these plasmids
were grown in minimal dextrose liquid lacking tryptophan and containing 0.2 mM KH2P04
(which induces expression from the PH05 promoter) (Thill et al. 1983) or 11 mM KHZPO4
(which does not induce the PH05 promoter) for two days at 30° C. Five pl of the culture

supernatants were assayed by electrophoresis on RNase activity gels (Yen and Green 1991).

57

.Multiple Sequence Alignment and Gene Genealogy

The deduced anrino acid sequences of RN 82, as well as those of RN 81 and RN83
(see Chapter 3), were aligned with the amino acid sequences of all other S-Iike RNases
known at present and sequences of the majority of S-RNases for which complete sequence
data are available (see legend to Figure 2-4 for a list of the sequences aligned). The
alignment was performed using the Genetics Computer Group program PILEUP (Devereux
et al. 1984) with a gap weight of 3.0 and a gap length weight of 0.1, and adjustments to this
alignment were made by eye. The RNase sequences were aligned and numbered fiom the
positions corresponding to the presumptive mature N-termini of the N. alata S-RNases
(Haring et al. 1990). The gene genealogy was also generated using PILEUP, which
produces a similarity score for each possible pair of sequences. The genealogy is a
representation of these similarity scores, which are used to order the alignment. The
horizontal branch distances in the genealogy are proportional to the similarities between the

sequences.

Expression Analyses

For organ analyses, Arabidopsis thaliana (L.) Heynh. ecotype RLD was grown in a
1:121 mixture of Sphagnum moss, perlite, and fine vermiculite under conditions of 16 hr
light / 8 hr dark with a relative humidity of 50% at 20° C. Roots, stems, leaves, and flowers
of 4- to 5-week-old plants were collected. For the senescence experiments, flowers were
staged on the basis of morphological characteristics as defined (Smyth et a1. 1990).

Senescing leaves were those showing visible signs of senescence, including chlorosis at the

58

leaf margins and wilting. To starve Arabidopsis for phosphate, one-week-old etiolated
seedlings were removed from solid AGM medium [MS salts (Sigma) at 4.3 g/liter, sucrose
at 30 g/liter, pyridoxine at 0.5 mg/liter, nicotinic acid at 0.5 mg/liter, thiamine
hydrochloride at 0.1 mg/liter, buffered with 2.5 mM MES at pH 5.7] and shaken in liquid
medium with or without 1.25 mM KH2P04 as described (Tewes et al. 1984) for 12 hr in the
dark. For pathogen studies, Arabidopsis thaliana plants of ecotype RLD were grown in
Metromix medium at 20° C under a 14 hour light / 10 hour dark photoperiod.
Pseudomonas syringae pv syringae strain van Hall (Vincent and Fulbright 1983) and
Pseudomonas syringae pv maculicola strain 4326 (Whalen et al. 1991) were grown
overnight in King's B medium (King et al. 1954), washed twice with 10 mM potassium
phosphate buffer at pH 6.9, and adjusted to OD640=0.1 in the same buffer. Four-week-old
Arabidopsis plants with well-expanded rosettes that had not yet bolted or that had bolts less
than 2 cm tall were infiltrated with the above Pseudomonas strains as described (Tsuji et al.
1991). Infiltrated leaves were harvested after 48 hours. All of the above tissues were frozen
in liquid nitrogen immediately upon collection and stored at -80° C before extraction.

Total RNA was isolated essentially as described (Puissant and Houdebine I990).
RN As were denatured and separated on fonnaldehyde/agarose gels and transferred to
Biotrace HP membrane (Gelman). The blots were hybridized to a 0.7 kb gene-specific
probe for RNS2, corresponding to the Eco RI-Xba I fiagment of the longest RNS2 cDNA
clone. This probe includes 15 bp of the 5' untranslated region and the coding region up to
nucleotide position 696. The probe was established to be specific to RN82 by hybridization

to DNA gel blots containing Arabidopsis genomic DNAs treated with various restriction

59

endonucleases, and subsequent comparison to patterns of blots probed with RN81 and
RNS3 fragments. Prehybridization was for five hours at 52° C in a buffer containing 5X
88C, 10X Denhardt's solution (Sambrook et al. 1989), 0.1% SDS, 0.1 M potassium
phosphate (pH 6.8), and 100 ug/ml denatured salmon sperm DNA. Hybridization was for
16 hours at 52° C in the same buffer, except that 10 % (w/v) dextran sulfate and 30% (v/v)
formamide were added, and the SDS was omitted. For use as an internal standard, an
Arabidopsis probe for the ubiquitous, highly expressed translation initiation factor e1F4A
(Owttrim et al. 1991) was generated by using PCR. This probe corresponds to amino acids
197-323 of Nicotiana tabacum eIF4A2, and its deduced amino acid sequence is 96.5%
identical to the latter (C. B. Taylor, P. J. Green, unpublished). The probes were prepared
using a random-primed labeling kit (Boehringer Mannheim) and hybridized to the
membranes at final concentrations of 1 x 106 to 2 x 106 dpm / ml. The membranes were
washed, with the final washes in 0.5X SSC, 0.1% SDS at 650 C, and then exposed to X-
Omat AR film (Kodak) with intensifying screens. Radioactive bands were quantitated

using a PhosphorImager model 400B.

CHAPTER 3

STUDIES ON RNS] and RNS3: RNS] IS TIGHTLY CONTROLLED
IN RESPONSE TO PHOSPHATE STARVATION

Portions of this chapter were published in The Plant Journal:

Bariola PA, Howard CJ, Taylor CB, Verburg MT, Jaglan VD, Green P] (1994) The
ArabidOpsis ribonuclease gene RNSI is tightly controlled in response to phosphate
limitation. Plant J 6:673-685

60

61

ABSTRACT

Two stimuli that have been associated with nutrient remobilization in plants are
phosphate (Pi) starvation and senescence. Little is known about how the nutrient
remobilization machinery is induced at the molecular level, but in the case of Pi starvation,
ribonucleases are considered to play important roles in the remobilization process. Here we
investigate the control of two closely related ribonuclease genes of Arabidopsis, RNS] and
RNS3. The RNS] gene is sharply induced during starvation for Pi, an effect specific among
the major macronutrients, whereas RNS3 transcript levels remain relatively constant. RN81
and RN 83 produced in yeast comigrate with Arabidopsis ribonuclease activities that exhibit
the same induction properties as the transcripts in both wild-type plants and the phoI
mutant, which is defective in xylem loading of Pi. In contrast to what occurs during Pi
starvation, both RNS] and RNS3 are modestly induced during senescence, indicating that
the two stimuli could trigger different signal transduction pathways. The characterization of
RNS] , in particular, provides an important first step towards elucidating the mechanisms by

which plants sense and respond to P, limitation, a prominent condition in many soil types.

62

INTRODUCTION

Plants often grow under Pi-limiting conditions (Fried and Brosehart 1967).
Phosphorus is commonly the most limiting nutrient in soil, afier nitrogen (Salisbury and
Ross 1992), and in some areas, Pi deficiency is a chronic situation (Sanchez and Uehara
1980). Despite the importance of this nutrient for plant growth, few details are known
about the mechanisms by which plants sense and respond to the availability of Pi. It has
been suggested that the extracellular rather than the intracellular Pi concentration elicits the
induction of RN ase activities in tomato (Lycopersicon esculentum). This proposal is based
on the observation that following the transfer of cultured tomato cells to Pi-deficient
medium, excretion of a Pi-starvation-inducible RNase (RNase LE) occurred well before
intracellular P, concentrations dropped (Glund and Goldstein 1993). The RNases induced
under these conditions are thought to be part of a P, starvation rescue system proposed to
exist in higher plants (Goldstein et al. 1989). One fimction of this system would be to
liberate P, from RNA to facilitate its remobilization. In addition to RNase LE, P, starvation
has been reported to induce secretion of both phosphatases and phosphodiesterases in plant
cells (Duff et al. 1991; Goldstein et al. 1988; Loffler et al. 1992; Ninomiya et al. 1977; Ueki
and Sato 1971). The combined actions of these enzymes could contribute to the release of
Pi (Loffler et al. 1992), in a form utilizable by plants.

A number of other plant enzymatic activities have also been reported to be affected
by P, limitation. In maize (Zea mays), Pi starvation results in altered activities of several

enzymes involved in photosynthetic carbon metabolism (U suda and Shimogawara 1992).

63

Studies of P, starvation in cell cultures of black mustard (Brassica nigra) and periwinkle
(Catharanthus roseus), as well as in tobacco plants, have demonstrated the enhancement or
repression of the activities of numerous enzymes involved in respiratory metabolism
(Nagano and Ashihara 1993; Paul and Stitt 1993; Theodorou and Plaxton 1993). Some of
these findings conflict with each other, perhaps due to differences in culture methods or
plant materials. In any case, very little information exists concerning whether regulation of
activities affected by P, starvation is exerted at the mRNA or protein levels in these studies
as few of the corresponding genes have been isolated. Prior to this work only one cDNA
had been found to be induced by P, starvation, a tomato cDNA, but its protein product has
yet to be identified (Glund and Goldstein 1993).

We have previously reported the identification of S-like RN ase genes in
Arabidopsis thaliana, a self-compatible plant (Taylor and Green 1991). One of these genes,
RNS2, is induced during senescence and P, starvation (Taylor et al. 1993) (also see Chapter
2). These observations suggest RN82 may degrade RNA during senescence so that its P,
can be remobilized to non-senescing organs, as well as participate in P, starvation rescue.
However, it was unclear from this study whether or not induction was the result of a general
pathway that could be triggered by starvation for other macronutrients. A fairly high level
of RNS2 expression was observed in the absence of induction, indicating that RNS2
expression is not entirely dependent on either senescence or P, starvation (Taylor et al.
1993). In this study we describe the cDNA cloning and expression of two related RNS
genes: RNS], which is highly inducible by P, starvation, and RNS3, which is relatively

insensitive to this stimulus. Among the major macronutrients, RNS] was preferentially

64

induced in response to starvation for P,; other results indicate that the mechanisms leading
to induction by P, starvation and senescence may differ. The tight control of RNS] mRN A
and associated RN ase activity in response to P, availability indicate this gene will provide

an effective means for elucidating P, starvation signal transduction pathways in plants.

65

RESULTS

Features of RNS] and RNS3 Sequences

To isolate RNS] and RNS3 cDNA clones, a Lambda ZAP cDNA library of aerial
tissue of Arabidopsis (ecotype Columbia) was screened with gene-specific PCR products
corresponding to RNS] and RNS3 as described in the Materials and Methods. The sequence
strategies for the cDNAs are shown in Figure 3-1, and the deduced amino acid sequences of
the longest clones identified are shown in Figure 3-2 (RNS!) and Figure 3-3 (RNS3). The
RNS] and RNS3 cDNAs contain open reading frames of 230 and 222 arrrino acids,
respectively. Both contain putative signal sequences at their N-terrninal ends, but no other
known targeting sequences, and therefore are likely extracellular proteins. When the
putative signal peptide sequences are removed at the sites predicted by statistical analysis
(von Heijne 1986), the proteins have deduced molecular weights of 23.0 kDa for RN81 and
23.3 kDa for RNS3. Neither protein contains potential N-glycosylation sites. The protein
sequences of RN81 and RN83 are quite similar (Figure 3-4), with 61% identity and 75%
similarity between them (Figure 3-5). In comparison, RN82 (Taylor et al. 1993) is only
37% identical to RN81 and 36% identical to RN83. A notable feature of RN81 is a
putative P-loop sequence containing the sequence GINPDGKS between residues 133-141
(Figure 3-4). P-loops, which form ATP or GTP binding sites, have a consensus sequence of
GXXXXGK[T/S] (where X refers to any amino acid), although the presence of this
sequence is not necessarily confirmation of a nucleotide binding site (reviewed in Saraste et

a1. 1990).

66

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

a- ‘A 4'
J 5' RN81 3'}
+~—-- — > ;—————>"
I 5' RN83 3+

 

 

 

A
A

 

 

A
A

Figure 3-1 - Strategy for sequencing of RNS] and RNS3 cDNAs. The large rectangles
represent the RNS] and RNS3 sequences, and arrows represent sequences read from
individual primers. Each base of the RNS] and RNS3 cDNAs were read at least once from
each direction.

121
12

181
32

241
52

301
72

361
92

421
112

481
132

541
152

601
172

661
192

721
212

781

841

901

Figure 3-2 - Primary structure and deduced amino acid sequence of RNS] cDNA. Deduced
anrino acid residues are shown in one-letter notation below the nucleotide sequence. The
triangle indicates the putative N-terrninal end of the mature protein predicted by a statistical

67

ATTTCTCTTTATATATATTCACCCATTAACCATCTCAATCTTATAACCCTCAAAATCACA

ATCTTCTCTTACAAAAAACTTTGAAAGATGAAGATTCTTCTAGCATCATTGTGTTTGATC
N K I L L A S L C L I

AGTCTTCTCGTAATCTTGCCTTCTGTCTTCTCTGCTTCTTCTTCCTCTGAAGATTTTGAT
8 L L V I L P 8 V F 8 A 8 8 8 8 F D F D

TTCTTCTACTTCGTCCAACAATGGCCAGGATCATACTGTGACACACAGAAGAAGTGTTGT
F F Y F V Q Q W P G S Y C D T Q R K C C

TATCCAAATTCAGGCAAACCAGCTGCTGATTTTGGCATTCATGGTCTTTGGCCTAACTAC
Y P N 8 G R P A A D F G I H G L W P N Y

AAAGATGGAACTTATCCATCTAACTGTGATGCCTCTAAACCATTCGATAGCTCAACGATA
K D G T Y P 8 N C D A 8 R P F D 8 8 T I

TCAGATCTTCTCACTTCAATGAAGAAGAGCTGGCCAACACTGGCTTGCCCAAGCGGTTCA
8 D L L T 8 N K K 8 W P T L A C P 8 G S

GGTGAAGCGTTTTGGGAGCACGAATGGGAGAAGCATGGTACTTGCTCTGAATCGGTTATC
G R A F W R H F W I R H G T C 8 F 8 V I

GATCAACATGAATATTTCCAAACCGCTCTTAACCTTAAACAGAAAACCAATCTCCTTGGA
D Q R F Y F Q T A L N L R Q R T N L L G

GCTCTAACCAAAGCCGGGATTAATCCGGATGGAAAATCTTACTCTTTGGAGAGCATAAGA
A L T K A G I N P D G K 8 Y S L F 8 I R

GATTCGATAAAAGAGTCAATTGGTTTCACTCCTTGGGTTGAGTGTAACAGAGATGGTTCT
D S I K R S I G F T P W V R C N R D G 8

GGTAATAGCCAATTGTACCAAGTCTATCTTTGTGTTGACCGGTCTGGTTCCGGTTTAATC
G N 8 Q L Y Q V Y L C V D R 8 G 8 G L I

GAATGTCCGGTTTTCCCACATGGGAAATGTGGAGCTGAGATCGAATTCCCTTCTTTTTAG
R C P V F P H G K C G A I I F F P 8 F *

TTTCCAATTTCTAAATTCAGTCTCTTGAACCGGCATCGATCTTTTGTTCATGGAGATTGG

TTTGGTCAAAAAGAAGATTTAATGTAATAAAAATTTATGGGATATTGGATAATGATATAG

‘A
TTCAATTTCAATAAAAAAAAAAAAA

analysis (von Heijne 1986). The arrow indicates the polyadenylation site used.

121
27

181
47

241
67

301
87

361
107

421
127

481
147

541
167

601
187

661
207

721

781

841

Figure 3-3 - Primary structure and deduced amino acid sequence of RNS3 cDNA. Deduced
amino acid residues are shown in one-letter notation below the nucleotide sequence. The
triangle indicates the putative N-terminal end of the mature protein predicted by a statistical
analysis (von Heijne 1986). Arrows indicate polyadenylation sites used in different cDNA

_ isolates.

68

CGACAAATTTGATTCCCACAAATTATTTGATATCTTGAGGAAATGAAATTCTTCATTTTT
M K F F I F

ATTCTAGCGTTACAACAACTCTACGTACAAAGTTTCGCCCAAGATTTCGATTTCTTCTAC
I L A L Q Q L Y V Q 8 F A Q D F D F F Y

TTCGTTTTACAGTGGCCTGGAGCGTATTGTGATTCAAGACATAGTTGTTGCTATCCACAA
F V L Q W P G A Y C D S R H 8 C C Y P Q

ACCGGTAAACCAGCTGCAGATTTTGGAATTCACGGTCTTTGGCCTAACTACAAAACCGGT
T G K P A A D F G I H G L W P N Y K T G

GGATGGCCGCAAAATTGTAATCCTGACAGTCGATTTGATGATTTACGGGTTTCTGATCTG
G W P Q N C N P D 8 R F D D L R V S D L

ATGAGCGATTTACAAAGAGAATGGCCAACGTTGTCGTGTCCGAGCAATGATGGTATGAAG
N 8 D L Q R R W P T L 8 C P 8 N D G M K

TTTTGGACACATGAGTGGGAGAAACACGGTACGTGCGCTGAGTCCGAGCTTGACCAACAC
F W T H F W F K H G T C A F 8 F L D Q R

GATTACTTCGAAGCTGGTCTCAAGCTCAAACAGAAAGCTAATCTCCTTCATGCCCTTACC
D Y F R A G L K L R Q R A N L L H A L T

AATGCTGGGATCAAACCGGATGATAAATTTTATGAAATGAAAGATATTGAGAATACGATC
N A G I X P D D K F Y 3 N K D I F N T I

AAACAAGTAGTTGGGTTTGCTCCGGGCATTGAATGTAACAAAGATTCGTCACATAATAGC
R Q V V G F A P G I F C N R D 8 8 H N S

CAACTCTATCAGATTTATTTATGTGTTGACACTTCGGCCTCCAAATTCATCAACTGTCCT
Q L Y Q I Y L C V D T 8 A 8 K F I N C P

GTTATGCCGCATGGTCGATGCGACTCTCGAGTTCAATTTCCCAAGTTCTAAATTTTATGG
V M P H G R C D 8 R V Q F P K F *

CTACATTTTTTCTATGTATTTACATTCTACCTATTAATTTTTATTTGTGTTTCTTGCAAA
SA
ATTTAGTTGTAATAAACAGTGTATAATTGGACGATACCCGTGATTTGCATGGTAAAAAGA

‘A
TGATAAAGGTTACTAAAAAAAAA

69

8289 2 82258 moo—d 3383 2F A2662 BEE .8 “Seams“ BEES: 8 @8882. 22?» e228
22> Quorum .3 poignancy ENG .825an So“ =« 5 :83qu 8:22: 2&in 82on 603w .872 98 52% we 80:253.

Bama 2682a 05 @5280 A32 .8 8 8596 VS 98 A33 .130 “we: m: 882% 0:52: me 2662 Eu 2: “a mammcn E2533
2:. 92:3 me VS 98 mg 882% me 822268 8 mmzm 98 572 Mo 80538 28 05:8 B2628 mo 588800 - FM 95»:—

ex ..x .mw> a o x3
22. .23.?“ S
Lz .3538. 85.
.8. .me ma 35.
.w or

on

 

   

70

RNSl RNS2 RNS3 LE LX

 

 

 

 

 

RN83 ~ 51

 

 

36 ' - 75 , 73
1‘ 7,, 1*" . 7 +7 . .,

V 1.-

 

 

 

 

 

LX643562‘65-

Figure 3-5 - Amino acid similarities between the RNS proteins and RNases LE and LX.
Sequences were compared using the Genetics Computer Group program GAP (Devereux et
al. 1984). Shaded values indicate percent similarity; non-shaded values denote percent
identity.

71
Comparison of RN81 and RNS3 Protein Sequences With Those of Other S-Like
RN ases

Inspection of the RN 81 and RNS3 sequences reveals that both proteins are highly
conserved members of the T2/S family of plant ribonucleases. Many residues are conserved
in the five regions of major homology (Ioerger et al. 1991), including histidines at positions
39, 92 and 97 and a carboxylic acid residue at position 93 (Figure 3-4), all of which were
shown to be necessary for catalysis in RNase Rh of Rhizopus niveus (Ohgi et al. 1992; Ohgi
et al. 1993). In addition, they contain several conserved cysteines, some of which are
involved in disulfide bond formation in another homolog, RNase T2 of Aspergillus oryzae
(Kawata et al. 1988). (For comparison to other sequences see Figure 2-4). Also striking is
the similarity of RN81 and RN83 to RNases LE and LX (Figures 3-4 and 3-5), for which
protein and recently cDNA sequences have been determined. RNases LE and LX are
induced in suspension-cultured cells of the self-compatible tomato Lycopersicon
esculentum upon starvation for P, (Lbffler et al. 1992; Niimberger et al. 1990). When
compared on a percentage similarity basis, RN81 and RNS3 are approximately as similar to
RNases LE and LX as they are to each other (Figure 3-5).

The sequences of RN81 and RNS3 have been added to the updated gene genealogy
tree shown in Figure 2-5. As expected, when added to the dendrogram, these sequences fall
into the S-like RNase lineage (Figure 2-5). Consistent with the data in Figure 3-5, this
analysis demonstrates that RN81 and RNS3 are more closely related to RNases LE and LX
than to RN82, suggesting that divergence within the S-like RNase lineage, as well as within

the plant T2/S family, may have preceded speciation.

72

Ribonuclease Activity of RN 81 and RNS3

RN81 and RN83 have significant similarity to proteins known to have ribonuclease
activity. To test whether they encode active RNases, their cDNAs were expressed in
Saccharomyces cerevisiae using the yeast expression vector pWL (Del Cardayré et al. 1995;
Taylor et al. 1993), as done for Figure 2-3. The proteins were targeted for secretion to
avoid possible deleterious effects fiom expressing an RNase intracellularly. Because a
heterologous signal sequence has been shown to function inefficiently in yeast (Ohgi et al.
1991), the sequences encoding the putative RN81 and RN83 signal peptides were replaced
by that of the yeast or-factor protein (Brake et a1. 1984). As described in the Materials and
Methods, the promoter used in these constructs can be repressed or derepressed depending
on the culture conditions. Following growth of yeast transforrnants under both conditions,
the electrophoresis of media samples containing secreted proteins on RN ase activity gels
(Yen and Green 1991) revealed that RN81 and RNS3 are RNases capable of degrading bulk
RNA (Figure 3-6). As expected, expression levels are high under derepressing conditions
and low under repressing conditions. No induction of any yeast RNase activity is visible.
The bands of RN81 and RNS3 activity, at 21.4 and 23.2 kDa, respectively, correspond well
with the predicted values of 23.0 and 23.3 kDa for RN81 and RN83 taking into account
removal of the putative signal peptides according to a statistical analysis (von Heijne 1986).
The slight differences may be due to the non-reducing conditions of the RN ase activity gels.

RN81 and RNS3 comigrate with bands in the RNase profile of Arabidopsis aerial
tissues (Figure 3-6). These bands appear to correspond to those referred to previously as the

22.6 and 23.7 kDa RNases (Yen and Green 1991). RN82, which appears as a doublet when

73

(a)

 

iPHos N ans I am—
(b)

repressed derepressed
total

C123eer123

43.7 >

28.9 >

 

 

Figure 3-6 - Expression of RNS proteins in Saccharomyces cerevisiae.

(a) Schematic representation of yeast expression constructs. RNS cDNAs were fused
between the yeast PH05 promoter and GAPDH terminator (Rosenberg et al. 1984) in the
plasmid pWL (Del Cardayré et al. 1995) as described in the Materials and Methods. Hatch
marks indicate the position of the signal peptide sequence included in each construct to
facilitate secretion of the RNS gene products.

(b) RNase activity gel of the RNS proteins. Yeast cells containing the RNS expression
constructs described in (a) were grown in liquid minimal dextrose medium (Thill et al.
1983) low in P, to induce the PH05 promoter ("derepressed"), or medium high in P, as
controls ("repressed"). Samples of culture medium were electrophoresed on RNase activity
gels. Lane C, vector control; lane 1, RN81; lane 2, RN82; lane 3, RNS3; total ext, protein
extract of Arabidopsis above-ground tissues. Molecular weights of standards in kDa are
shown to the left of the gel.

74

expressed in yeast (Figures 2-3 and 3-6), does not comigrate with any bands in the profile.
This discrepancy may result from differential processing of the RN82 protein in yeast, via
glycosylation (possibly at its potential N-glycosylation sites), removal of the signal peptide,
or other modifications that could lead to a molecular weight different from that of the
endogenous plant protein. RN81 and RN 83 have no putative N-glycosylation sites, and
modifications other than glycosylation and signal peptide removal have not been reported
for other members of the Tz/S RNase family. Therefore, it is likely that the Arabidopsis

RNases that comigrate with the yeast-expressed proteins are RN81 and RN83.

RNS] and RNS3 Expression During Normal Development

To examine the expression patterns of RNSI and RNS3 in different organs, blots of
RNA isolated fi'om roots, inflorescence stems, leaves, and flowers were hybridized to RNS]
and RNS3 probes. As shown in Figure 3-7, the RNS] transcript is present primarily in
flowers and nearly undetectable in the other organs. In contrast, substantial levels of the
RNS3 transcript are observed in roots, inflorescence stems and flowers, but not in leaves.
Both of these patterns differ from that of the RNS2 transcript, which is present at a
significant level in all four organs and most abundant in flowers (Figure 2-6). Although all
three RNS genes are highly expressed in flowers, these results demonstrate that the genes
are differentially controlled in other organs.

It was also of interest to test whether RN81 and RN 83 might contribute to the well-
documented increase in RNase activity reported to take place in senescing plant organs

(reviewed in Farkas 1982 and Green 1994). Total RNA was isolated from healthy and

75

R S I F
RN51» a
B R S L F

RN53» ii an .

 

Figure 3-7 - RNS expression in organs of soil-grown Arabidopsis. Total RNA was isolated
fi'om roots (R), inflorescence stems (8), leaves (L), and flowers (F) of Arabidopsis. RNA

gel blots containing 12 pg of these samples per lane were hybridized with RNS] (a) or
. RNS3 (b) probes.

76

senescing Arabidopsis leaves at the stages described in Chapter 2. As shown in the gel
blots in Figure 3-8, both RNS] and RNS3 are induced during senescence. Although the
levels of RNS] and RNS3 transcripts in senescing leaves are modest compared to that of the
RNS2 transcript (Figure 2-7A), the induction of RNS] and RNS3 during senescence is
nevertheless reproducible. When the same blot was rehybridized with a probe for the
eukaryotic translation factor e1F4A from Arabidopsis (CB Taylor, PJ Green, unpublished),
no induction of the corresponding transcript was observed, indicating that equal amounts of
RNA were loaded and that induction of the RNS transcripts during senescence cannot be
accounted for by a general effect, such as an overall change in the ratio of mRNA to

ribosomal RNA in senescing tissues.

RNS] and RNS3 Expression in Response to P, Limitation

As discussed earlier, the S-like RNases induced during senescence may play a role
in P, remobilization. A similar argument can be made for those induced during P,
starvation. Previous reports of increases in RNase activities in tomato (Loffler et al. 1992;
Ntirnberger et al. 1990) and RNS2 gene expression in Arabidopsis (Chapter 2) during P,
starvation could indicate that induction in response to this stimulus is common to all S-like
RNases. To test this hypothesis with respect to RNS] and RNS3, total RNA was isolated
from green seedlings grown on P,-rich or P,-deficient media after germination on P,-rich
medium. Gel blots of this RNA were hybridized with probes for RNSI , RN82, RNS3, and
the internal standard eIF 4A. Figure 3-9 shows that RNS] transcript levels are extremely low

in plants grown in P,-rich medium, but are induced to a high level by P, starvation. In

77

NS S
< RN81
B NS 8
< RNS3
C
1! 09‘. < OIFM

Figure 3-8 - RNS expression during senescence in leaves of Arabidopsis. Total RNA was
isolated from non-senescing (N S) or senescing (8) leaves of Arabidopsis. An RNA gel blot
with 5 ug of these samples per lane was hybridized sequentially to RNS] (a) and RNS3 (b)
V probes, as well as to the translation initiation factor e1F4A (c) probe as an internal standard.

78

RNSI RNS2 RNS3

e1F4A» . . . g ‘ .“

Figure 3-9 - RNS expression in plants grown under P,-rich or P,-deficient conditions.
Arabidopsis plants were germinated on AGM medium, transferred three days afier
germination to media rich (+) or deficient (-) in P,, and grown for an additional seven days.
Total RNA was isolated from these plants, and RNA gel blots containing 10 ug of these
samples per lane were hybridized to the RNS] , RNS2, or RNS3 probes as indicated, and
subsequently to the e1F4A probe.

Parts of this experiment were performed by Christie Howard and Crispin Taylor

79

contrast, RNS3 transcript levels remain nearly unchanged during P, starvation. RNS2,
previously found to be induced in etiolated seedlings starved for P,, is induced to essentially
the same level in green seedlings starved for P, (compare Figure 3-9 with Figure 2-7B).

Seedlings starved for P, grow more slowly than those grown on P,-rich medium, and
appear somewhat stunted when compared to normal seedlings of the same age.
Accordingly, RNS] induction during P, starvation could be a general nutrient starvation
response, not specific to P,. This possibility was tested by subjecting seedlings to starvation
for two other macronutrients, nitrogen and potassium, using methods analogous to the P,
starvation experiments. RNA gel blot analysis of RNA from these plants revealed that
unlike P, starvation, neither nitrogen nor potassium deficiencies lead to any large induction
of RNS] expression (Figure 3-10). RNS3 and e1F4A transcript accumulation is insensitive
to starvation for all three macronutrients (Figure 3-10). The seedling samples used for
Figures 3-9 and 3-10 were subjected to seven days of nutrient starvation. The RNS]
transcript was also induced in seedlings starved for P, for five and nine days, whereas
seedlings starved for nitrogen or potassium for the same periods of time exhibited no
change in RNS] transcript levels relative to unstarved plants (CJ Howard, PJ Green,
unpublished).

Since the potassium starvation media described above contained a small amount of
potassium, the effect of media containing no potassium were also tested on Arabidopsis
seedlings. As shown in Figure 3-11, RNSI is induced to a small extent in seedlings grown
on medium containing no potassium. The induction is 3.7-fold, as compared to the

approximately 75-fold induction by starvation for P, (Figure 3-11) (figures calculated after

80

$119,. ‘,~ ’tl‘v ( RNS]

< RNS3

“0 no . < e|F4A

Figure 3-10 - RNS] and RNS3 expression in plants starved for various nutrients. Plants
were germinated on AGM medium, transferred three days after germination to media rich
(+) or deficient (-) in phosphate (P,), nitrogen (N), or potassium (K), and grown for seven
more days. Total RNA was isolated from these seedlings, and an RNA gel blot containing
10 ug of these samples per lane was hybridized sequentially to the RNS] (a), RNS3 (b), and
e1F4A (c) probes.

Parts of this experiment were performed by Christie Howard and Crispin Taylor

81

RN51) I. ,

e1F4A > . -..

Figure 3-11 - RNS] expression in plants grown on media containing no potassium. Plants
were germinated on AGM medium, transferred three days afier germination to media rich
(+) or deficient (-) in phosphate (P,) or potassium (K), and grown for seven more days.
Total RNA was isolated from these seedlings, and an RNA gel blot containing 10 ug of
these samples per lane was hybridized sequentially to the RNS] and e1F4A probes.

Parts of this experiment were performed by Christie Howard

82

normalization to the signal of the e1F4A transcript in each lane). Potassium is a major
cation involved in maintaining the cation-anion balance in plant cells (Poirier et al. 1991),
therefore one possible explanation for the small induction of RNS] during potassium
starvation is that lack of potassium allows less phosphate to be taken up into plants.

Poirier and co-workers (Poirier et al. 1991) have isolated an Arabidopsis mutant,
phol, impaired in delivery of P, into the xylem after its uptake by root epidermal cells. Leaf
P, levels in this mutant are much lower than in wild type, so its leaves are constantly
deprived of P,. To examine the response of the RNS genes to the P, limitation inherent in
this mutant, total RNA was isolated from healthy, non-senescing leaves of the phoI mutant
and of Columbia wild type. The expression patterns of RNS], RNS2, and RNS3 in wild type
and the phoI mutant reflect their expression patterns in plants grown on P,-rich and P,-
deficient media: RNS] is highly induced, RN82 is moderately induced, and RNS3 transcript
levels are nearly unchanged (Figure 3-12). These results show that exogenously applied

and endogenous P, deficiency result in similar RNS gene expression patterns.

RN ase Profiles of P,-starved and phol Plants

RN81 and RNS3 proteins produced in yeast corrrigrate with major activities in the
Arabidopsis RNase profile (Figure 3-6). If the comigrating Arabidopsis RNases are indeed
RN 81 and RN83, then they might be expected to exhibit the same responses to P, starvation
observed for the RNS] and RNS3 genes. To investigate this possibility, protein extracts
from seedlings grown on P,-rich and P,-deficient media were resolved on RNase activity

gels. Samples from the P,-starved plants exhibit a strong band of activity comigrating with

83

RNSI RNS2 RN53
W1 phol wt phol wt phol

    

. a...

e1F4A) -. .. ..

Figure 3-12 - RNS expression in the phol mutant of Arabidopsis. Columbia wild-type (wt)
and phol plants were germinated on AGM medium and transferred to soil ten to twelve
days after germination. Total RNA was isolated from leaves of these plants nine to ten days
after transfer to soil. RNA gel blots containing 10 pg of these samples per lane were
hybridized to the indicated probes as in Figure 3-9. The same blot was hybridized with
both the RNS] and RN83 probes.

Parts of this experiment were performed by Christie Howard and Crispin Taylor

84

yeast-produced RN 81 (Figure 3-13). Since the band is not visible in samples from plants
grown on P,-rich medium, it likely corresponds to an increase in the amount of RN81
protein present in P,-starved plants, coinciding with the increase in RNS] mRN A. A similar
result was obtained when protein samples from phol mutant plants were examined. These
plants, which have increased levels of RNS] mRNA (Figure 3-12), exhibit increased
activity of the band comigrating with yeast-produced RN81, as compared to wild type.
Changes are less obvious in the area of the RNS3 band in both P,-starved and phol samples,
which was expected because little change in RNS3 transcript levels is observed in these
plants (Figures 3-9 and 3-12). However, it should be noted that there are multiple activities
in this region. The apparent increases in activity of the band comigrating with yeast-
produced RN81 in P,-starved and phol mutant plants support the contention that this RNase
is the RNS] gene product. These experiments also suggest that induction of RNS] could be

part of the plant P, starvation rescue response.

85

—- or
a) U)
Z2
ex :2

wtCol
pholCol

 

Figure 3-13 - RNase activity profiles of P,-starved plants and phol mutant plants. Protein
extracts were prepared from the same plant materials used for RNA analysis (see the
Materials and Methods) and resolved on RNase activity gels. Samples were: protein
extracts from seedlings grown on media rich (P+) or deficient (P-) in P,; RN81 and RNS3
expressed in yeast; and protein extracts of leaves from Columbia wild-type (wt Col) or phol
(phol Col) plants. Molecular weights of standards in kDa are shown to the lefi of the gel.

Parts of this experiment were performed by Vanita Jaglan and Christie Howard

86

DISCUSSION

In this study we examined two Arabidopsis genes, RNS] and RNS3, which encode
RNases in the T2/8 superfamily, a major family of RNA-degrading enzymes in higher
plants. Our results demonstrate that control of RNS] differs from that of RNS3 in non-floral
organs and in response to P, starvation, but is similar during senescence. The marked
induction of RNS] under P,-lirniting conditions was of particular interest because of its

potential for providing insight into P, starvation-inducible signal transduction mechanisms.

Structural Features of RN 8] and RNS3

Comparison of the amino acid sequences of RN81 and RN 83 revealed many
similarities with other members of the Tz/S RNase family. Most significant was the finding
that RN81 and RNS3 are more closely related to S-like RNases from tomato, zinnia, and
tobacco than to any other known RNases, including RN82 of Arabidopsis. This result
indicates that the lineages giving rise to the former enzymes and RNS2 (Figure 2-5)
diverged before speciation. RN81 and RNase LE are 71% identical at the amino acid level
and both are induced by P, starvation. Because tomato and Arabidopsis are in different
families, the high degree of similarity between RNases LE and RN81 suggests that
homologs of these RNases may be fundamental components involved in RNA degradation
in diverse plant species.

RN 81 contains a putative P-loop sequence, which constitutes an interesting

distinction between RN81 and all other TZ/S RNases with published sequences. P-loops are

87

present in several families of proteins that specifically associate with nucleoside
triphosphates, such as some ATP synthases, kinases, elongation factors, and myosins
(Saraste et al. 1990). In addition, an unrelated RNase, 2-5A-dependent RNase, utilizes P-
loops to bind an oligoadenylate activator (Zhou et al. 1993). It is not known how a P-loop
could function in RN81. One possibility is that RN81 activity is controlled by a nucleotide
effector; alternatively, a P-loop could conceivably contribute to an RNA binding domain on

RN81.

Induction of the RNS Genes During Senescence and P, Starvation is Differential and
not a General Nutrient Starvation Response

Our analyses have shown that all three Arabidopsis RNS genes are induced by
senescence, albeit to different extents. The effect of senescence is most apparent in the case
of RN82, which is strongly induced in aging flower petals and leaves (Figure 2-7A).
Induction of RNS] and RNS3 in senescing leaves of Arabidopsis is modest in comparison to
RNS2, but nevertheless significant.

A second stimulus shown to induce RNS genes in Arabidopsis is P, starvation. It
was important to examine whether this effect was related to the control of RNS genes during
senescence, because mineral deficiency often induces senescence-like responses in plants
(Noodén 1988). If both P, starvation and senescence trigger the same signal transduction
pathway, then one might expect the hierarchy of control of the three genes to be similar in
response to both stimuli. However, this does not appear to be the case. The effect of P,

starvation is large for RNS], which is highly induced from a low basal level, and minimal

88

for RNS3, which is expressed at similar levels with or without the stimulus of P, starvation.
RN82 exhibits an intermediate effect; it is clearly induced during P, starvation, but also
displays a fairly high basal level. In contrast, during senescence RNSI and RNS3 are both
modestly induced from a low basal level. RNS2 is highly expressed during senescence, but
the induction ratio is modest due to a relatively high basal level. The simplest way to
explain the differential effects of senescence and P, starvation on RNS gene expression is
that the two stimuli operate via different signal transduction mechanisms, perhaps with
some common components. However, alternative models involving multiple pathways for
one or both stimuli or gene-specific repressors of a common pathway cannot be excluded.

It was also of interest to investigate whether the effects of P, starvation on RNS]
expression were mediated through a general pathway that responds to multiple forms of
nutrient deficiency in Arabidopsis. To this end, we examined transcript levels following
nitrogen and potassium starvation. As shown in Figures 3-10 and 3-11, neither treatment
resulted in large induction of the RNS] or the RNS3 transcripts. Induction due to nitrogen
deficiency might have been expected because RNA could be viewed as a source of nitrogen
as well as P,. Nevertheless, RNS] induction was preferentially induced under P, starvation
conditions relative to the other nutrients tested. This argues that the strong induction of
RNS] during P, starvation is not a general nutrient starvation response.

In spite of the importance of P, as a nutrient for plant growth, very little is known
about the molecular components that mediate the response of plants to P,-lirniting
conditions. The regulatory properties of RNS] indicate that this gene will provide an

excellent entry point into elucidation of P, starvation-inducible signal transduction

89

pathways. For example, if regulation is controlled by the 5' flanking sequences of RNS],
then the RNS] promoter may be an effective tool to identify the transcription factors that
mediate responses to P, starvation. Other signal transduction components could be
identified genetically through the isolation of mutants that alter RNS] expression under P,-
rich or P,-deficient conditions. Another important advantage of using RNS] in these studies
is that regulation can be examined at both the mRNA and protein levels. Figure 3-13
indicates that the RNS] gene product is likely responsible for the band of RNase activity
induced in phol and P,-starved wild-type plants that comigrates with RN81 produced in
yeast. Although we cannot rule out the possibility that the RNase activity of RN81 does not
change in P,-starved plants and another RNase the same size as RN81 is induced under

identical conditions, this seems unlikely.

Roles of the RNS Gene Products in Higher Plants

The expression properties of the RNS genes strongly suggest that the corresponding
RNases participate in P, remobilization in Arabidopsis. Induction of all three RNS genes
during senescence is consistent with previous observations that senescence induces a
number of hydrolytic enzymes (Borochov and Woodson 1989). These enzymes
presumably degrade macromolecules of dying cells, fleeing them for remobilization to
reproductive structures (Kelly and Davies 1988). Senescence has been correlated with the
induction of RNase and nuclease activities in other plant systems (F arkas 1982), but the
genes for these RNases have not been isolated so it is unclear whether regulation occurs at

the RNA or protein levels. Another situation in which P, remobilization could play a key

90

role is under conditions of P, limitation, when RNS] and RNS2 are induced. As discussed
earlier, plants commonly grow under P,-limiting conditions, so understanding their
responses to this situation is of fundamental importance. It has been proposed, on the basis
of the experiments performed primarily with cultured tomato cells, that plants have a P,
starvation rescue system involving not only RN ases, but also phosphodiesterases and
phosphatases (Goldstein et al. 1989). Presumably, this system would allow plants to
establish different priorities for P, use during conditions of nutrient limitation and
abundance.

Beyond their role in P, remobilization, the RNS gene products may also participate
in other processes in plants, including cell death and defense against pathogens.
Programmed cell death in animal systems is often associated with the induction of
nucleolytic activities (Collins and Rivas 1993), and this has also been reported to occur in
plants during xylogenesis. When mesophyll cells of Zinnia elegans are induced to
differentiate into xylem cells, a nuclease and several RNases accumulate (Thelen and
Northcote 1989). A recent screen for xylogenesis-associated cDNAs (Ye and Varner 1993)
led to isolation of zinnia homologs of RN81 and RN83 (Ye and Droste 1996); this may
suggest that RN81 and/or RN83 participate in xylem maturation in Arabidopsis. With
respect to defense against pathogens, RNase X2 of Petunia inflata has been suggested to
participate in plant defense because it is specifically localized to the pistil (Lee et al. 1992).
The pistil is a floral structure that is potentially vulnerable to infection but is rarely invaded,
perhaps due to the presence of RNases, protease inhibitors, and other proteins that could be

deleterious to pathogens (Lee et al. 1992; Atkinson et al. 1993 and references therein).

91

RNS] , RNS2, and RNS3 could contribute to this effect because all three genes are expressed
in flowers, and RN82 expression has been further localized to Arabidopsis pistils (Chapter

2).

92

MATERIALS AND METHODS

cDNA Isolation

RNS] and RNS3 were first identified via the polymerase chain reaction (PCR) from
rescued plasmid DNA of a Lambda ZAP library as previously described (Taylor and Green
1991). The PCR products were then used as hybridization probes to screen the same
Lambda ZAP library for full-length cDNAs as described in Chapter 2. cDNAs were
sequenced by the dideoxy chain termination method (Sanger et a1. 1977). Multiple clones
of RNS] and RNS3 were isolated; DNA sequences of the longest RNS] and RNS3 cDNAs
identified were deposited into the EMBL, Genbank, and DDBJ databases with accession
numbers of U05206 (RNS!) and U05207 (RNS3). Nucleotide position numbers referred to

in this chapter correspond to these sequences.

Expression Analyses

Plant Material: A. thaliana (L.) Heynh. ecotype RLD was grown in soil as described in
Chapter 2. For organ analysis, samples of roots, inflorescence stems, leaves, and flowers of
four- to five-week-old plants were collected. Senescing leaves were harvested as described
in Chapter 2. For nutrient starvation analyses, seeds were sterilized and plated on solid
AGM medium (see Chapter 2 for formulation) on a layer of Nitex 300 um nylon mesh
(Tetko Inc., Briarcliff Manor, New York). Plating density was 100-200 seeds per 100 x 25
mm plate. Plants were germinated under conditions of 16 hour light/8 hour dark, at 20°C.

Three days after plating, when radicles had appeared but no shoot growth was yet apparent,

93

the seedlings were transferred on the nylon mesh to growth medium rich or deficient in
phosphate, nitrogen or potassium (see formulations below), allowed to grow for seven days,
and then harvested. Seeds of the P,-uptake mutant phol, mutant line PL9 (Poirier et al.
1991), a derivative of the Columbia ecotype, were kindly provided by Yves Poirier. For
examination of RNS expression in the phol mutant, seeds of both this line and Columbia
wild-type were surface-sterilized and sown on solid AGM medium. After ten to twelve
days, seedlings were transferred to soil and grown under conditions for soil-grown plants
described in Chapter 2. Nine to ten days after transfer, healthy leaves of each line were
harvested. All plant tissue was frozen immediately in liquid nitrogen and stored at -80°C

until use.

Media Formulations for Nutrient Starvation Experiments: The composition of media for
the nutrient starvation experiments was as described for AGM medium (Chapter 2) with
several changes. The minimal organics salts of Linsmaier and Skoog (Linsmaier and Skoog
1965) were substituted for the MS salts, with the following modifications: In all cases,
CuSO4 and CoC12 were omitted and the F e804 x 7H20 concentration was increased to 0.05
g l", as recommended by Tewes et al. (1984). In P,-deficient medium, the KHZPO4 was
omitted. In both N-rich and N-deficient media, 1.4 g 1'1 KCl was substituted for the KNO3.
N-rich medium also contained 1.7 g 1'1 (NH4)NO3. In both K-rich and K-deficient media,
1.6 g 1'1 NaNO3 was substituted for the KNO3. K-rich medium also contained 1.4 g 1'1 KCl.
In K-deficient medium, 0.17 g 1" NaH2P04 was substituted for the KH2P04. Medium with

no potassium was the same as K-deficient medium except MES was omitted and the molar

94

equivalent amount of NaI was substituted for the K1. The pH of all media was adjusted to
5.7, using KOH for media rich and deficient in P, and N, and NaOH for K-rich and K-

deficient media.

RNA Manipulation: Total RNA was isolated and Northern blots were prepared and probed
as described in Chapter 2. The RNS] probe used was a 0.76 kb EcoRI fragment from the 5'
end of the cDNA, including 87 bp of the 5' untranslated region and 673 bp of the coding
region. RNS] and RNS3 probes were prepared using 32P-labeled dCTP and a random-
primed labeling kit (Boehringer Mannheim). For the RNS3 probe, a 0.66 kb EcoRV-Xhol
fragment from the 5' end of the cDNA, consisting of 14 bp of the 5' untranslated region and
643 bp of the coding region, was used. On genomic DNA blots, both probes showed the
same hybridization patterns as the gene-specific RNS] and RNS3 PCR probes (Taylor and
Green 1991, and unpublished results). As an internal standard, the blots were hybridized
with a probe for the Arabidopsis translation initiation factor e1F4A (CB Taylor, PJ Green,
unpublished). For sequential hybridizations, blots were stripped between hybridizations in
a solution of 0.1X SSC, 0.1% SDS that was brought to greater than 90°C and shaken for 20
min at room temperature. Blots were checked for residual activity using Phosphorimager

analysis before rehybridization.

Expression of the RNS cDNAs in Yeast
RNS] and RNS3 cDNAs were inserted into the yeast expression vector pWL (Del

Cardayré et al. 1995) under the control of the inducible PH05 promoter (Arima et al. 1983;

95

Vogel and Hinnen 1990) and the GAPDH terminator (Rosenberg et al. 1984). Sequences
encoding the putative signal peptides of RN81 and RN83 upstream of nucleotide positions
135 and 84, respectively, were replaced by the signal peptide of the yeast or-factor signal
sequence (Brake et al. 1984). Saccharomyces cerevisiae cells transformed with these
constructs via the lithium acetate method (Ito et al. 1983) were grown in liquid minimal
dextrose high-P, (repressing conditions) and low-P, (derepressing conditions) media (Thill
et al. 1983) as described in Chapter 2. RNase activities secreted into the culture medium

were then assayed as described below.

RNase activity gels

RNase activity gels were electrophoresed and developed as described (Yen and
Green 1991). Plant material was grown, harvested and stored in the same manner as for
RNA analyses, and protein extracts were prepared essentially as described (Yen and Green
1991), except that the concentrated extraction buffer consisted of 250 mM NaPO4 pH 7.4, 5
mM EDTA, 4 mM PMSF, 25 pg ml'l leupeptin, and 25 pg ml"l antipain. Arabidopsis
above-ground tissues consisted of all tissues growing above the soil. Each lane contained
100 pg of Arabidopsis protein samples. For yeast expression studies, cultures of cells
expressing a given RNS gene or control were centrifuged for two minutes at 14,000 g, and
the supematants were harvested. Supernatant samples of 5 pl for RN81 and 10 pl for

RN82, RN 83, and control were then resolved on RNase activity gels.

CHAPTER 4

IMNIUNOLOCALIZATION OF RN82

96

97

ABSTRACT

As discussed in Chapter 2, RN82 is a highly abundant RNase whose gene is
induced by both senescence and P, starvation. The deduced amino acid sequence of RN 82
contains a C-terrninal extension that has some features in common with plant C-tenninal
vacuolar targeting signals. To examine whether or not RN S2 is vacuolar, antibodies that
recognize RN82 were produced for use in determining its subcellular location. Generated
against a peptide corresponding to a region unique to RNS2, the antibodies recognize the
protein specifically. Use of the antibodies in irnmunoblots showed that RN82 is present in
all major organs of the plant, and increases in abundance during P, starvation, as predicted
by RNA gel blot analysis (Chapter 2). Irnmunogold electron microscopy of leaf and petal
sections using these antibodies revealed that RN82 is present in the cell wall. This
extracellular location is consistent with a role for RNS2 in remobilization of P, fi'om RNA
released from lysed cells or RNA stored as reserves in seeds, as well as in defense against

pathogens.

98

INTRODUCTION

As described in Chapter 2, the deduced amino acid sequence of RN82 contains, in
addition to an N-terminal secretory signal sequence, a C-terrninal extension of about 20
amino acids as compared to the C-tenninal ends of other 8- and S-like RNases (Figure 2-4).
When coupled with an N-tenninal secretory signal sequences, some C-tenninal extensions
target plant proteins to the vacuole (Bednarek and Raikhel 1992). At the onset of the
experiments described in this chapter, no consensus sequence had been deduced for plant C-
terrninal vacuolar targeting sequences, but a common feature among those known was short
stretches of hydrophobic amino acids, and it was therefore thought that this feature could
form the core of the signal recognized by the vacuolar sorting system in plants (Bednarek
and Raikhel 1992). The RN 82 C-tenninal extension has such stretches of hydrophobic
arrrino acids (Figure 2-4), and it was thus proposed that RN82 could be a vacuolar protein.
This idea seemed plausible, since plant vacuoles contain large amounts of RNase activity
(Boller and Kende 1979; Abel and Glund 1987). More recently, two C-tenninal vacuolar
targeting sequences from plant proteins were analyzed by mutagenesis, and the hydrophobic
stretches present in both were found not to be required for proper targeting (Dombrowski et
al. 1993; Neuhaus et al. 1994). Since these reports, progress has been slow in determining
whether any consensus sequence or structure is necessary for vacuolar targeting by C-
terrninal extensions (Neuhaus 1996).

One of the major goals of this thesis project was to elucidate roles of the RNS

proteins in plants. Since determining the subcellular location of a protein can aid greatly in

99

deducing its biological role, it was decided that determining the subcellular location of
RN82 was essential. To do this, antibodies recognizing RN82 were needed, so the initial

efforts for this project were to obtain antisera specific for RN82.

100

RESULTS

Production of Antibodies that Specifically Recognize RN 82

In order to determine the subcellular location of RNS2, antibodies that specifically
recognized this protein were necessary. One method to obtain antibodies specific to one
protein out of a family of closely related proteins is to use as antigens synthetic peptides
with regions unique to one protein. In the initial attempt to obtain such antibodies for
RN82, three peptide regions were chosen from RN 82 based on their uniqueness in
comparison to the protein sequences of RN81 and RN83. Peptides SP616 and SP618
correspond to intemal sequences of RN82, whereas JK374 encompasses its C-tenninal
extension (Figure 4-1). These peptides (Figure 4-1) were synthesized with a lysine residue
added at their N-terrninal ends to facilitate coupling to carrier proteins. Afier coupling to
keyhole limpet hemocyanin, the peptides were injected into rabbits for antibody production.
Sera were tested for reactivity against the antigens using dot blots containing samples of the
peptides.

Injection of peptides SP616, SP618 and JK374 led to the production of antibodies
that reacted specifically with each peptide. Unfortunately, the antibodies that recognized
peptides SP616 and SP618 on dot blots did not recognize either RN82 produced
heterologously in yeast (see Chapter 2) or RNS2 in Arabidopsis extracts (data not shown).
Antibodies generated against peptide JK374 recognized yeast-produced RN82, but reacted
with no proteins in Arabidopsis extracts (data not shown). The latter observation raises the

possibility that the C-terrninal extension of RN82 is removed in viva, but this idea was not

101

mi. -- ..
m:
was

m1
RN82
m:

"I
m2
m3

 

Figure 4-1 - Synthetic peptides used for producing anti-RN82 antibodies. Deduced amino
acid sequences of RN81, RN82 and RN83 are aligned, beginning with the first residues
shown in Figure 2-2. Peptides are shaded and their designations indicated above the
sequences. Boxes highlight the conserved regions described in Figure 2-2.

102

explored firrther. The failure of antibodies generated against synthetic peptides to recognize
the whole protein is not uncommon, as the conformation of the free peptide or peptide-
carrier protein conjugate may be different than that of the corresponding region in the
protein (Harlow and Lane 1988).

After the above efforts, an attempt was made to generate antibodies against
heterologously-produced RN 82, in the hopes that the structure of the RN 82 protein is
different enough from RN81 and RN83 so that specific antibodies would be possible. The
system chosen to produce RN 82 in this case was E. coli, utilizing pET-based vectors that
added a 6X-histidine tail to RN82 (see Materials and Methods), for ease in the purification
of large amounts of protein. In this system, RN 82 formed preferentially an insoluble,
inactive form; however, the protein could be solubilized in 6 M urea and purified on a
nickel column. After elution from the column, RN82 was further purified by SDS-PAGE
and electroelution. Injection of this RN 82 preparation into rabbits did not lead to the
production of anti-RN82 antibodies. This failure was probably due to the presence in the
RN S2 preparation of some component that inhibited strong emulsion formation with the
adjuvant. A poorly-formed emulsion fails to protect the antigen from rapid catabolism,
leading to a shortened time of exposure of the antigen to the immune system and thus
possibly a poor immune response (Harlow and Lane 1988). The emulsion-inhibiting
component could not be removed from the preparation by passing it through a desalting
column.

Finally, another attempt was made to produce specific anti-RN82 antibodies using a

peptide antigen. The sequence corresponding to peptide PGI was chosen using the criteria

103

of high hydrophilicity, increasing the chances that the sequence appears on the surface of
the protein (Harlow and Lane 1988), and uniqueness in relation to RN81 and RNS3. In
addition, the peptide was chosen such that it began with a cysteine, which had two
advantages: it can be used in coupling to carrier proteins, and this cysteine in particular is
part of the true sequence of the peptide, potentially increasing accuracy in the antibody
formation. At the time of synthesis this peptide sequence was unique in the Genbank
database. Injection of P61 coupled to keyhole limpet hemocyanin into rabbits led to the
production of sera rich in antibodies that strongly recognize RNS2, both yeast-produced and
in Arabidopsis extracts (Figure 4-2). These antibodies do not recognize RN81 or RN 83
(Figure 4-2), and pre-irnmune serum fi'om the same rabbit exhibits very low reactivity with
proteins in Arabidopsis extracts (data not shown). The anti-RN82 antibodies detect a single
band of approximately 32 kDa in Arabidopsis protein extracts (Figure 4-2). This size
corresponds well with the predicted molecular weight of RN82 of 27.2 kDa, given removal
of the putative secretion signal sequence and glycosylation at one or both of the potential N-
glycosylation sites (see Chapter 2). Since extensive efforts were made using PCR to
identify any additional RNases of the RNS family in Arabidopsis (data not shown), and, as
of November 1996, no other RNS-like sequences have appeared in the Genbank database as
a result of the large-scale Arabidopsis cDNA sequencing projects, it is reasonably certain
that RN81, RN82 and RNS3 comprise the entire gene family of this type of RNase in

Arabidopsis. Therefore, it is likely that these antibodies recognize only RN82.

104

F0100
(DUDCD-Q
zzze
EIE<
1(44
<19

 

Figure 4-2 - hnmunoblot characterization of anti-RN82 antibodies. Proteins were resolved
by SDS-PAGE, transferred to PVDF membrane, and irnmunodetected with anti-RN82
serum. RN81, RN82, RNS3: approximately 300 ng each of the indicated RNase, in
supernatant from RNase-expressing yeast cells (see Chapter 3). Arab: 50 pg of proteins
extracted fiom above-ground tissues of five-week old Arabidopsis Columbia wild-type
plants.

105

Patterns of RN82 Expression

Protein extracts were made from several organs of Arabidopsis to determine where
in the plant RNS2 is present. Immunoblot analysis showed that RN S2 is present in all
major organs of the plant (Figure 4-3A). RN82 is abundant in roots, stems, leaves and
flowers, as predicted by analysis of RN82 mRNA levels in Chapter 2 (Figure 2-6). Among
these organs, it appears that RNS2 is most abundant in roots as a percentage of total protein -
in each extract (Figure 4-3A), which conflicts with the relative RN82 abundance seen at the
mRNA level (Figure 2-6). However, it is known that a large percentage of the proteins in
green tissues is RuBisCo protein, for example up to 50% of the soluble protein in leaves
(Kung 1976) (note the large bands of about 55 and 15 kDa in leaf, stem and flower extracts
in Figure 4-3B). Roots contain no chloroplasts and therefore no RuBisCo protein. When
comparisons are made using equal protein amounts, in extracts of green tissues such a large
percentage of the proteins are RuBisCo that a smaller proportion of all other proteins can be
loaded in relation to the proportion in root extracts. Thus it may appear that RN82 is more
abundant in roots, whereas the relative amounts of RN82 per cell may be very different
from the relative amounts in total soluble protein of different organs. In addition to the
above organs, RN82 is also present in significant amounts in extracts of siliques and seeds.
This observation indicates that RN82 may have a role in the reproductive process.

RN82 abundance was examined in protein extracts from seedlings grown on media
rich or deficient in P,. As shown in Figure 4-4A, RN82 is more abundant in P,-deprived
seedlings. This confirms the assumption made in Chapter 2, that induction of the RNS2

gene during P, starvation leads to increased accumulation of RNS2 protein in Arabidopsis.

106

RSLFSISd

RSLF SISd

v~--
it

a (97
' <66

 

Figure 4-3 - Distribution of RN82 among various organs of Arabidopsis. Protein extracts
(30 pg per lane) were made from organs of four- to five-week-old Columbia wild-type
plants and resolved by SDS-PAGE. (A) Protein extracts transferred to PVDF membrane
and irnmunodetected with anti-RN82 serum. R, roots; 8, stems; L, leaves; F, flowers; 8],
siliques; 8d, seeds. (B) Protein extracts visualized with Coomassie staining. Molecular
weights of standards in kDa are shown to the sides of the gels.

107

P+ P- P+ P-

 

44>
29> "-
19>
22)
14> --

Figure 4-4 - Increase in RN 82 abundance during phosphate starvation. Protein extracts (25
pg per lane) were made from seedlings grown on media rich (P+) or deficient (P-) in P, and
resolved on SDS-PAGE. (A) Protein extracts transferred to PVDF membrane and
irnmunodetected with anti-RNS2 serum. (B) Protein extracts visualized with Coomassie

staining. For both A and B, molecular weights of standards in kDa are shown to the left of
the gels.

108

Extensive efforts were made to determine the relative abundance of RN82 in senescing and
non-senescing leaves, analogous to the experiments done at the mRNA level in Figure 2-
7A. No conclusions could be made since, during senescence, proteins are being rapidy
degraded (N oodén 1988b), which made accurate protein quantitation and thus equal loading
of proteins very difficult (data not shown).

The anti-RNS2 antibodies generated using peptide PGl as an antigen recognize a
protein of approximately the same size as the predicted RNS2 protein, which increases in
abundance during P, starvation, as does RN82 mRNA. These antibodies do not recognize
RN81 or RN83, and were produced in high titer. For these reasons it was decided that they
were of sufficient quality to use in irnmunogold electron microscopy to determine the

subcellular location of RN 82.

Immunolocalization of RN82

The anti-RN82 antibodies described earlier were used in irnmunogold labeling at
the ultrastructural level to determine the subcellular location of RN 82. In the initial
attempts, Arabidopsis tissues were fixed with a typical fixation mixture of glutaraldehyde
and formaldehyde (see Materials and Methods). Treatment of these sections with anti-
RNS2 antibodies produced very little labeling. This problem was traced to the use of
glutaraldehyde as a fixative. Glutaraldehyde reacts with free amino groups (Harlow and
Lane 1988). Since the PGI antigen, against which the anti-RN82 antibodies were
generated, contains two lysine residues, treatment of the Arabidopsis tissues with

glutaraldehyde likely resulted in the modification of the PG] epitope(s) on RNS2 such that

109

the antibodies were unable to bind to the protein. Glutaraldehyde also effectively masked
RN82 detection on irnmunoblots (data not shown).

To overcome this problem, Arabidopsis tissues were treated with a fixation mixture
containing formaldehyde. Omission of the glutaraldehyde at this step allowed labeling of
RN82 as well as good preservation of cell structures. Extensive labeling of RN82 was
observed in leaf tissues (Figures 5A and 5B). Unexpectedly, RN82 labeling appeared in
cell walls. Treatment of leaf tissues with pre-irnmune serum resulted in a low background
level of labeling; however, a small amount of labeling, much less extensive than that seen
with anti-RNS2 serum, was seen in cell walls (Figures 5C and 5D). RN82 labeling was
also observed in petal tissue (Figure 6A and 6B), although not as strongly as in leaf tissue.
Again, the labeling is specific, as treatment with pre-irnmune serum results in very little
labeling (Figures 6C and 6D).

The appearance of RN82 in cell walls in electron micrographs of leaf and petal
tissues does not imply that RNS2 is a cell-wall intrinsic protein, since small extracellular
proteins are able to move freely within the porous structure of the plant cell wall. Since
RN82 is found in substantial quantities in the soluble fraction of crude plant extracts, it is
likely that this protein is a free-floating extracellular protein. Experiments are underway to
confrrm by cell fractionation studies the extracellular localization of RN82 by electron

microscopy.

110

   
  

  

'Z-zi',;,'..."-,_ '
seer... ..
,. , M, "-"’ ‘ 4
.7 5 flat,
1. ‘ -.r_" .

I
\

 

Figure 4-5 - Immunocytochemical localization of RN82 in leaves of Arabidopsis using
irnmunogold labeling. Sections of leaf cells were treated with (A & B) anti-RN82
antibodies or (C & D) pre-irnmune serum. Cw, cell wall; V, vacuole; Cp, chloroplast. Bars
= 0.5 pm.

lll

 

 

Figure 4-6 - Immunocytochemical localization of RN82 in petals of Arabidopsis flowers
using irnmunogold labeling. Sections of petal cells were treated with (A & B) anti-RN82
antibodies or (C & D) pro-immune serum. Cw, cell wall; V, vacuole. Bars = 0.5 pm.

112

DISCUSSION

A surprising finding in determining the subcellular location of RN82 was that RN 82
appears to be an extracellular protein. Although unexpected, this result is not inconsistent
with data from other Tz/S type proteins in plants. For example, the S-RNase 82 of
Nicotiana alata has been localized to the intercellular matrix and cell wall in pistils using
methods like those described in this chapter (Anderson et al. 1989), and the S-like RNase
LE of tomato was purified from culture medium after secretion from suspension-cultured
cells (Nt‘rrnberger et al. 1990).

The plant cell wall contains numerous species of hydrolytic enzymes under normal
conditions, among them proteases, glycosidases, and phosphatases (Cassab and Varner
1988), as well as some types of ribosome-inactivating proteins (Barbieri et a1. 1993). Roles
in defense against pathogens have been proposed for many of the above enzymes (Cassab
and Varner 1988), and proven in some cases (Alexander et al. 1994). Since the cell wall is
one of the first baniers pathogens encounter upon infection of a plant, a cell wall location
for defense-related enzymes seems appropriate. Bolstering this idea is the observation that
several types of proteins that are specifically upregulated during defense responses are
extracellular proteins (Alexander et al. 1994).

Following this logic, since RN S2 is a cell-wall localized hydrolytic enzyme, it is
possible that it participates in the defense response of plants against pathogens. A direct
role in attack on pathogens seems difficult to reconcile, as the protein would be separated

from substrate RNAs by a physical barrier. However, RN82 could act in concert with other

113

defense-related enzymes. For example, [MB-glucanases, which attack fungal cell walls,
may initiate fungal cell lysis, after which RN82 may degrade firngal RNA. Indeed, the
cytotoxicities of a barley ribosome-inactivating protein, a chitinase and a [3-1 ,3-glucanase
against fungal hyphae are synergistically enhanced when their actions are combined (Leah
et al. 1991). Altemately, RN82 could aid in the protection of plants against infection by
RNA-based viruses, although a mechanism by which RN82 could gain access to the viral
RNA is obscure. It should be noted that the S-RNase 82 of Nicotiana alata, which is
known to be a secreted protein (Anderson et al. 1989), is taken up into the cytoplasm of
pollen tubes when applied in vitro (Gray et al. 1991). The exact mechanism by which 8-
RNases contribute to self-incompatibility is unknown, but one current model incorporates
the idea of allele-specific uptake of S-RNases into the pollen tube cytoplasm (Kao and
McCubbin 1996). Although the idea that RNS2 could be taken up into cells of plant
pathogens seems unlikely, it cannot be ruled out, due to the lack of knowledge about the
target(s) of action of most T2/8 type RNases.

As described above, it is possible that RNS2 may be a defense-related enzyme, but
the observation that the RN82 gene is not induced strongly by attack with some pathogens
(Figures 2-8 and 2-9) suggest that RNS2 may have alternate or additional roles. The RN82
protein is relatively abundant in all organs exanrined (Figure 4-3), and increases in
abundance during P, starvation (Figure 4-4). In addition, the RNS2 gene is induced during
senescence (Figure 2-7). As discussed in Chapter 2, these results are consistent with a role
for RN82 in the recovery of phosphate from RNA. However, since RNA is not normally

present extracellularly, this idea seems unusual. Interestingly, acid phosphatase is the most

 

ll4

abundant cell wall hydrolase in some tissues (Larnport and Catt 1981), and, like RNA, the
substrates for acid phosphatases are not usually found outside of cells (Cassab and Varner
1988). As plants often grow under nutrient-limiting conditions, they may maintain a
hydrolytic extracellular environment to maximize recycling of cellular components released
during such processes as programmed cell death (see Chapter 1) and wounding. Another
role is suggested by the presence of RN82 in dry seeds (Figure 4-3). The RNase may
participate in the mobilization of nucleotides and phosphate upon germination fiom RNA
stored in the endosperm.

In conclusion, RN82 is an abundant extracellular protein that may have multiple
roles in plants. To attempt to confirm some of these ideas, it will be necessary to examine
transgenic plants with decreased RN 82 levels. Progress toward these experiments is

described in Chapter 5.

llS

MATERIALS AND METHODS

Anti-RN82 Antibody Production

Antigen Preparation: Synthetic peptides SP616, SP618 and JK374 were prepared at the W.
M. Keck Facility at Yale University. Synthetic peptide PGI was prepared at the Michigan
State University Department of Biochemistry Macromolecular Structure Facility. The
pruity of peptides was monitored by analytical reversed-phase HPLC and mass
spectrometry. Peptide sequences were as follows: SP616,
KRGTRHCCSKNACCRGSDAP; SP618, KGSPSSCNGGKGSFWG; JK374,
KYTPLDGEAMVLKMPTEREAL; PGI, CYRSDFKEKE. Peptides SP616, SP618 and
JK374 were coupled to keyhole limpet hemocyanin (KLH) (Pierce) using the crosslinkers
disuccinimidyl suberate (D88) and m-maleirnidobenzoyl-N-hydroxysuccinirnide ester
(MBS) (Pierce), in separate reactions. Peptide PGl was coupled to KLH using maleirnide-
activated KLH (Pierce).

Heterologously-produced RN S2 was made using the E. coli pET system. The RNS2
cDNA was altered via site-directed mutagenesis (Kunkel et a1. 1987) to create an Nco I site
just after the putative signal sequence (see Chapter 2) and an Xho I site at the 3' end of the
coding sequence. The mutagenized cDNA was inserted into the pET-22b+ vector
(N ovagen) between the corresponding sites. The resulting gene encodes a fission protein
with a pelB leader and a 6X-histidine tail. E. coli strain BL21 (DE3) (Novagen) containing
the above plasmid was grown in Luria broth containing 100 mg/l carbenicillin until mid-log

phase, at which point IPTG was added to a final concentration of 1 mM. After two hours,

116

cells were collected by centrifugation. Growth at either 37° C or 28° C resulted in RN82
forming in insoluble inclusion bodies (data not shown). To purify RN82, the cells were
lysed in a solution of 10 mg/ml lysozyme, 1% Triton X-100, 50 mM Tris pH 7.5, 2 mM
EDTA at 30° C for 30 minutes, followed by passing the lysate repeatedly through a 22-
gauge needle and centrifuging at 20,000 g. The resulting pellet was solubilized in 6 M urea,
5 mM imidazole, 0.5 M NaCl, 20 mM Tris pH 7.9 and purified on a nickel column (His-
Bind Resin; Novagen) according to manufacturer's instructions. Eluted RN82 was further
purified by separation on SDS-PAGE, visualization with CuClz (Lee et al. 1987), isolation
of the band corresponding to RN82, and electroelution. The purity of the protein was
monitored by SDS-PAGE and silver staining (Bio-Rad Silver Stain kit). The protein was
desalted on a PD-lO column (Pharmacia), concentrated, and the buffer replaced with
phosphate-buffered saline (PBS) (Sambrook et al. 1989) before injection.

Anti-peptide antibodies were isolated from whole sera via affinity purification on
Affi-Gel 10 or Affr-Gel 15 columns (Bio-Rad), depending on the pl of the peptide.
Peptides were coupled to the support in DMSO as per the manufacturer's suggestions, with
7.5-20 pmol peptide per ml of resin. Anti-peptide antibodies were bound to and eluted
from the affinity column as described (Harlow and Lane 1988). After elution, antibodies
were concentrated in Centricon-IO units (Anricon) to the same volume as the serum

originally applied to the column.

Injection of Antigens and Antiboay Isolation: Female New Zealand white rabbits were used

for all antibody production, and pre-irnmune serum was collected from each rabbit. All

117

antigens were emulsified in Titer Max adjuvant (V axcel, Inc.) for subcutaneous injection.
Approximately three weeks after the initial injections, rabbits were re-inj ected with the
same antigens as boosts, followed ten days later by collection of blood. Rabbits were bled
at two-week intervals thereafter. Peptides SP616, SP618 and JK374 coupled to KLH via
D88 or MBS were injected in amounts of 40-160 pg total protein, in both initial and boost
injections. These peptides were also injected uncoupled, 5 mg per initial injection or boost.
Approximately 100 pg of RN82 purified via the pET system was injected initially, followed
by 140 pg boosts. Peptide PGl coupled to KLH was administered at 267 pg in the first
injection and 400 pg in boosts. Sera were screened by testing various dilutions on

irnmunoblots containing yeast-produced RN82 (Chapter 2).

Protein Gels and Immunoblot Analysis

All Arabidopsis tissues described in this chapter are from the Columbia ecotype.
Roots, stems, leaves and flowers from soil-grown plants, above-ground tissues from soil-
grown plants, and plants grown on P,-rich and P,-deficient media were grown and harvested
as described in Chapter 3. Green siliques from 5 to 15 mm in length were collected from
four-week-old plants. Seeds used for protein extraction were viable, dry seeds that had been
stored at room temperature for approximately one year. Proteins were extracted from
harvested tissues as described in Chapter 3. Glycerol was added to 10% v/v in all protein
extracts and samples were frozen at -80° C in small aliquots.

Protein extracts were mixed with sample buffer and boiled for five minutes before

being separated on 11% SDS-PAGE gels (Laemmli 1970) cast with an

118

acrylarnidezbisacrylamide ratio of 30:0.8. Proteins were blotted onto PVDF membrane
(lmmobilon-P; Millipore) using semi-dry electrophoretic transfer as described (Harlow and
Lane 1988). After transfer but before drying, membranes were autoclaved in transfer buffer
for 20 minutes at 120° C as described (Swerdlow et al. 1986), since autoclaving increases
the signal strength of RN 82 detection by the antibodies described above (data not shown).
Blots were processed for RN 82 signal detection using one of two protocols: one with a
TBS-Tween buffer (Birkett et al. 1985) and another with a Blotto/T ween buffer (Harlow
and Lane 1988) as blocking agents. Anti-PGI (anti-RN 82) serum (described above) from
rabbit 11192 was diluted 1:2000 for detection, and goat-anti-rabbit Ingalkaline
phosphatase conjugate was used as the secondary antibody. Signals were developed using
nitroblue tetrazoliurn and 5-chloro-4-bromo-3-indolyl phosphate (Harlow and Lane 1988),
incubating in development buffer for 10 minutes. Dot blots with peptides were prepared by
spotting 1 pl of 5 mg/ml solutions of peptides onto strips of nitrocellulose membrane. For
signal detection, the dot blots were incubated with various dilutions of sera and processed as

described above.

Immunocytochemistry

In initial experiments, Arabidopsis tissues (ecotype Columbia) from healthy, soil-
grown plants were fixed in 2% formaldehyde/ 1% glutaraldehyde in 50 mM sodium
phosphate buffer with 0.1 M sucrose (pH 7.2) and vacuum infiltrated for two hours at room
temperature. In subsequent experiments, Arabidopsis leaves and petals were fixed in 4%

formaldehyde in the same buffer and vacuum infiltrated as described above. After fixation,

119

all tissues were washed three times in 10 mM sodium phosphate buffer with 0.5 M sucrose
(pH 7.2) for 10 rrrinutes each. The tissue was postfixed in 1% 0504 + 10 mM sodium
phosphate + 0.05 M sucrose (pH 7.2) for one hour and then rinsed in distilled water three
times for 10 minutes each. Following dehydration in an ethanol series, the tissue was
infiltrated with London Resin White acrylic resin (Polysciences, Warrington, PA) and
polymerized at 58° C under vacuum overnight. Thin sections were prepared on an Ultracut
E nricrotome (Reichert-Jung, Vienna, Austria) and mounted on fonnvar-coated nickel grids
(Polysciences, Warrington, PA). Immunocytochemistry was performed as described
(Schroeder et al. 1993). Affinity purified anti-PGI (anti-RN82) antibodies or pre-irnmune
serum were diluted 1 to 4. Goat anti-rabbit IgG (1:1 dilution) was used as an amplifying
bridge. Protein A conjugated to colloidal gold of 15 nm diameter (EY Lab Inc., San Mateo,
CA) was diluted 1 to 50. Thin sections were examined on a JEOL 100CXII transmission

microscope (Tokyo, Japan).

CHAPTER 5

GENERATION OF TRANSGENIC PLANTS WITH DECREASED AND
INCREASED AMOUNTS OF RNS] AND RNS2

120

121

ABSTRACT

To gain insight into the in vivo roles of the RN81 and RNS2 proteins, antisense
constructs for the RNS] and RNS2 genes were transformed into plants and the transfonnants
screened for decreased protein or RNA of the corresponding gene. Several constructions
were used in the attempt to produce antisense RNS2 lines. The most affected antisense lines
had leaf RN82 mRNA levels of 39-60% that of control plants and exhibit no obvious altered
phenotype. In contrast, antisense RNS] lines were obtained that had RNS] mRNA levels as
low as 10% of that of control plants under P,-deficient conditions. These lines have high
anthocyanin contents, which is a response seen during starvation for P,. A project was also
begun to overexpress RNS] in roots such that RN81 would be secreted into the soil,
degrading rhizospheric RNA in combination with phosphatases and thus increasing the P,
available to the plant. Transgenic plants with increased root RNS] expression were
obtained, but the expression level is comparable to that of the endogenous RNS] gene under

P,-deficient conditions.

122

INTRODUCTION

To determine the function of a protein in any biological system, some of the best
approaches to use are gene inactivation techniques. The most definitive methods of this
type, targeted gene replacement or deletion, have not been extensively developed for use in
plants. Instead, the method most widely used in plants involves the techniques of antisense
RNA (Bourque 1995; Dougherty and Parks 1995). This method has the advantage that
expression can be downregulated over a wide range, from completely abolished to only
slightly affected, making it possible to not only see the effects of varying amounts of the
protein but also to obtain plants affected in proteins whose complete suppression would be
lethal. In addition, expression of a family of closely related genes can sometimes be
suppressed en masse with antisense.

In Chapters 3 and 4, it was proposed that RN 81 and RN82 had roles in the
mobilization and recycling of P, during such phenomena as senescence and P, starvation.
Obtaining antisense plants with decreased expression of RNS] and RN82 seemed an
attractive first step to test these theories. Antisense techniques have recently been used to
show that S-RNases are essential in the rejection of self pollen during the self-
incompatibility response (Lee et al. 1994), demonstrating that this method can succeed with
genes of the T2/8 farme of RNases in plants. Since no clear guidelines exist regarding the
best way to construct antisense gene fusions, as a starting point the full-length RNS} and
RNS2 cDNAs were fused in reverse orientation to a strong promoter and transformed into

Arabidopsis.

123

Another project using similar techniques but with a different objective was begun in
collaboration with Dr. Marcel Bucher to overexpress RNS] in roots of Arabidopsis. Some
soils contain abundant organic matter and therefore contain RNA, a rich source of
phosphate. Plants are known to secrete organic acids into the rhizosphere, increasing the
solubility of inorganic phosphates and thus their availability to the plant (Johnson et al.
1996a and references therein). In addition, secretion of acid phosphatases into the
rhizosphere by plant roots is thought to aid in the mobilization of P, from organic
phosphates (Duff et al. 1994). Some fungi are known to secrete nucleases, thought to
facilitate absorption of P, from their environment (Fraser and Low 1993). However, it is
not known whether plants secrete RNases or nucleases from their roots. The secretion of
RNases fi'om roots should give plants a competitive advantage in extracting P, from soils
that are rich in organic matter but otherwise poor in P,. To release P, from RNA, RNase
action needs to be combined with phosphatase action, and there is evidence that
phosphatases are indeed secreted from Arabidopsis roots (M. L. Abler, unpublished). An
observation was made that RNS], dramatically upregulated in seedlings during P, starvation,
is induced to a lesser extent in roots of P,-starved plants (C. J. Howard, P. J. Green,
unpublished). Increasing RN81 secretion fi'om roots may give plants the competitive
advantage described above. To test this theory, the RNS] cDNA was firsed to a root-
specific promoter and Arabidopsis plants transformed with this construction, in the hopes
that transformed plants would secrete from their roots abundant RN81 protein. This project
was designed to explore whether such engineering would be useful for crop plants, so that

less chemical fertilizer use would be needed.

124

RESULTS AND DISCUSSION

Antisense Inhibition of the RNS] Gene
Production of anti-RNS] antibodies: To confirm that putative antisense RNS] plants
contained low levels of RN81 protein, anti-RN81 antibodies were essential. Because the
use of a unique synthetic peptide as an antigen worked well with RN82 (Chapter 4), this
approach was also tried for RN81. The same region of the protein that was used to design
the synthetic peptide for anti-RNS2 antibodies was used to make a peptide specific for
RN81. The peptide, PG2, shown in Figure 5-1A, was unique in the Genbank database at
the time of synthesis, with its closest relative a ribonuclease fiom zinnia, ZRNaseI (Ye and
Droste 1996). Peptide PG2 was prepared and injected just as done for the anti-RN82
peptide PGI , but the antibodies generated against it produced unexpected results. These
antibodies recognize a protein of about 17 kDa, which is significantly smaller than the
predicted molecular weight of 23.0 kDa for RN81 (Figure 5-lB). In addition, the
abundance of this protein actually decreases during P, starvation, in stark contrast to the
strong induction of the RNS] gene during this stimulus (Figure 3-9). Finally, the protein
appears abundant in stems and leaves, unlike the RNS] transcript (Figure 3-7). Since it
appears that the anti-PG2 antibodies recognize a protein that is not RN81, these antibodies
were not used in further studies. This type of problem is common when using synthetic
peptides as antigens (Harlow and Lane 1988).

The next approach used to make anti-RN81 antibodies was to use as an antigen

RN81 protein produced in a heterologous system. The yeast overexpression system used to

125

alter WNW . T [ japan-smart
nus: MATCJJMBC. .
ruler nuc. .rs. and l I _ _,_

RN82
RN83 PTLSC..PS. "M I

RN81 ...... GK. .CGAIIII'PSF
RN82 TAIQMITPIVVCKRDAI . . +3.
RN83 .-

RSLF din.

44> , I .,"<44
29) "<29
<19

Figure 5-1 - Attempts to prepare anti-RN81 antibodies using a peptide antigen. (A)
Location of peptide PG2 in the amino acid sequence of RN81. Deduced amino acid
sequences of RN81, RNS2 and RNS3 are aligned, beginning with the first residues shown
in Figure 2-2. Boxes highlight the conserved regions described in Figure 2-2. (B) Protein
samples were resolved by SDS-PAGE, transferred to PVDF membrane, and
immunodetected with anti-PG2 antiserum. Roots (R), stems (8), leaves (L), and flowers (F)
were prepared as for Figure 4-3, and seedlings grown on P,-rich (P+) or P,-deficient (P-)
media were prepared as for Figure 4-4.

126

produce RN81 (described in Chapter 3) was chosen because RN81 can be generated in
quantities up to 9 mg/liter of supernatant, and RN81 is the dominant protein secreted so it
can be easily purified. Concentrated supernatant in which yeast cells secreting the RN 81
protein were grown was passed over an anion exchange column and bound proteins eluted
with an increasing salt gradient. At pH 6.0, RN81 bound to the anion exhange column and
eluted at a fairly narrow range, such that it was possible to separate the protein from all
major contaminant proteins. The purity of the eluted RN81 preparation was estimated at
>95%, as monitored by one-dirnensional SDS-PAGE and silver staining (Figure 5-2A) as
well as isoelectric focusing (data not shown). Injection of RN 81 protein into rabbits led to
the production of antisera of high titer that recognize a protein of about 25 kDa, close to the
predicted size of 23.0, in extracts of P,-starved Arabidopsis seedlings (Figure 5-ZB). These
antibodies are not entirely specific for RN81; RNS3 is also detected at approximately 10-
fold lower efficiency (data not shown; the RNS3 band is faintly visible in Figure 5-2B).
However, RN81 and RN 83 have slightly different electrophoretic mobilities, so these

antibodies are adequate for use in detection of RN81 on irnmunoblots.

Generation of antisense RN81 plants: To obtain antisense RNS] plants, a T-DNA
transformation vector was constructed containing the entire RNS] cDNA fused in reverse
orientation between the strong cauliflower mosaic virus 358 promoter and the nos
terminator (Figure 5-3). Transgenic plants were made using Agrobacterium tumefaciens-
mediated transformation with the method of vacuum infiltration. With this method, large

numbers of independent transfonnants can be obtained easily, which was an advantage for

127

v- N ('3
9 9 9
123 iimmm
97)
.v . < 44
66) x
45) ' .< 29
- --
3" . ' -< 19
22)

14)

Figure 5-2 - Anti-RN81 antibody preparation using heterologously-produced RN81 as an
antigen. (A) Various amounts of a preparation of yeast-produced RN81 that was purified
by anion exchange chromatography, resolved by SDS-PAGE and silver stained. 1, 200 pg ;
2, 600 ng; 3, 200 ng. (B) Protein samples were resolved by SDS-PAGE, transferred to
PVDF membrane, and immunodetected with anti-RN81 antiserum. RN81, RN82, RNS3:
approximately 70 ng each of the indicated RNase, in supematant from RNase-expressing
yeast cells (see Chapter 3). P+, P-: 30 pg of total proteins from Columbia wild-type
seedlings grown on media rich (P+) or deficient (P-) in P, as described for Figure 5-2. For
both (A) and (B), molecular weights of standards in kDa are shown to the sides of the gels.

128

925 1

/..—-QF°S.1<M>°9 .358 ' ISNH “:99 ; .353 ‘ GUS [19$ ,

 

 

 

 

 

 

 

 

 

 

 

p1448 I

Figure 5-3 - Structure of antisense RNS] plant transformation vector. The entire RNS]
cDNA is fused to the cauliflower mosaic virus 358 promoter and a nos terminator. nos,
nopaline synthase; NPTII, neomycin phosphotransferase; GUS, B—glucuronidase; RB and
LB, right border and left border, respectively, of the T-DNA. All elements except for RNS]
are as described (Jefferson I987). Shaded elements indicate promoters, hatched elements
indicate terminators, and empty boxes indicate coding regions.

129

this project since it was expected that many transforrnants would need to be screened to
obtain plants in which the antisense effect was very strong. Transgenic control lines were
made by transforming plants with the pBI 121 vector.

Under normal growth conditions, the flower is the only major organ of the plant in
which RNS] mRNA is abundant (see Figure 3-7). For this reason, flowers were selected as
the organ to assay for decreased RN81 activity. (Screening for lowered RNS] mRNA
levels was ruled out due to the difficulty of gathering enough flowers from individual plants
to obtain sufficient total flower RNA.) The initial generation of transgenic plants was
grown in soil and proteins extracted from flowers gathered from each of 120 independently
transformed plants were resolved on RNase activity gels. Lowered RN81 activity was
judged by comparing the intensity of the RN81 activity band in plants containing the
antisense RNS] construction with that of control plants. As discussed in Chapter 3, this
band is very likely to be that resulting from the RN81 protein. Extensive efforts to prove
this were done by attempting to transfer proteins from stained RN ase activity gels to blots
and detect RN81 with anti-RN81 antibodies (described above), but these did not succeed
due to technical problems. However, to date all fluctuations in RN81 activity observed due
to various growth conditions are rrrirrored at the mRNA level (C. J. Howard, P. J. Green,
manuscript in preparation); therefore the use of this band to assay RN81 activity in putative
antisense plants seemed justified. Of the 120 individual transfonnants screened in this way,
13 putative RNS] antisense plants were identified.

To confirm lowered RN 81 activity in the 13 putative antisense RNS] lines, one

antibiotic-resistant progeny plant (T2 generation) fi'om each line was grown in soil and its

130

flowers assayed as done for the T1 generation. Of these lines, five displayed lowered RN81
activity in the T2 generation. T3 plants of several of these lines were assayed in the same
way, which eliminated two of the lines due to normal levels of RN81 activity. The final
test done to confirm lowered RN 81 levels lines was to grow T3 and T4 seedlings of these
lines on media rich or deficient in P,, as described for wild-type seeds in Chapter 3.
Seedlings were harvested after seven days of grth on these media, taking care to remove
antibiotic selection-sensitive seedlings, and these plants were analyzed for RN81 activity on
RNase activity gels.

The RN ase activity profiles of the lines most affected in RNS] are shown in Figure
5-4. The antisense effect is most easily seen in extracts of seedlings grown under P,-
deficient conditions, due to the low level of RN81 activity in seedlings (antisense and
control) grown with abundant P,. Lines 8d.5.2, 8d.5.3 and 23g.4 all have significantly
lower RN 81 activity than controls when grown on P,-deficient medium. 8d.5.2 and 8d.5.3
have a similar amount of RN81 activity under these conditions, probably due to their being
sibling lines. 23 g.4 exhibits even lower RN81 activity under P,-deficient conditions, and in
addition appears to have less activity in high-P, conditions than all the other lines in Figure
5-4. The profiles of these lines are shown in comparison to several controls, including the
wild-type plants and two transgenic lines. Because all the antisense lines were grown in the
presence of kanamycin, the wild-type is not the best control line for this situation, but it is
shown for comparison. Line 34d.3, a transgenic line containing only the pB1121 vector,
exhibits the expected large amount of RN81 activity during P, starvation. Line 36h.3.2

contains the antisense RNS] construction but has near-normal levels of RN 81 activity in

131

control 1 antisense control 2

 

wtCol 34d.3 8d.5.2 8d.5.3 239.4 36h.3.2

+-+-+-+-+-+-

RN81

 

Figure 5-4 - Decreased RN81 activity in RNS] antisense lines. T3 or T4 seedlings of
antisense RNS] lines were germinated on AGM medium, transferred two days after
germination to media rich (+) or deficient (-) in P,, and grown for an additional seven days.
Protein extracts were prepared from all kanamycin-resistant seedlings and 50 pg of each
sample was resolved on RNase activity gels. Control 1, transgenic line containing pBIl2l
vector; control 2, transgenic line containing construction p1448 but which has a near—
norrnal amount of RN81 activity; RNS], RN81 protein produced in yeast. The RN81
activity band is indicated. Molecular weights of standards in kDa are shown to the left of
the gel.

132
low-P, conditions, like the majority of the lines containing this construction that were
screened.

RNS] gene expression in the above lines was examined by hybridizing total RNA
isolated from the same batches of tissue harvested for Figure 5-4 with an RNS] RNA probe
that hybridizes only with the sense RNS] mRNA (Figure 5-5A). Once again, the antisense
effect is most clearly seen in P,-deprived seedlings. As RNS] levels seem to be similar
under high-P, conditions in all lines shown, antisense suppression of RNS] was quantitated
by calculating the fold induction of RNS] by P, starvation (after normalization with e1F4A).
As shown in Figure 5-5B, the 34d.3 control line is induced 11.3 fold by this stimulus,
whereas the 8d.5.2, 8d.5.3 and 23g.4 lines are only induced 3.1, 2.4 and 1.1 fold,
respectively. Although RNSI expression is not completely abolished in 23g.4, its RNS]
mRNA level during P, starvation is only 10% that of the 34d.3 control, making this line an
excellent choice for future studies.

The reduction in RN81 protein levels expected in the antisense RNS] lines was
confirmed using the anti-RN81 antibodies described earlier with irnmunoblots containing
the same protein extracts used in Figure 5-4. RN81 protein is clearly detected in P,-starved
extracts of the wild-type, 34d.3 and 36h.3.2 lines (Figure 5-6), all of which have more
RN81 activity (Figure 5-4) and RNSI mRNA (Figure 5-5) than the antisense lines.
However, little, if any, labeling of RN81 is observed in extracts of the three antisense lines
(Figure 5-6). This lack of labeling, which is in contrast to the reduced but visible levels of

RN81 activity seen in Figure 5-4, is probably due to the differences in sensitivity between

133

antisense controlt control 2
8d.5.2 8d.5.3 239.4 34d.3 36h.3.2

+-+-+-+-+-

 

 

 

RN81) . has

 

elF4A >

 

B

 

.5
N
r l

8
1

Fold induction of RN81 by Pl starvation
G
r

 

 

 

Figure 5-5 - RNS] mRNA levels in RNS] antisense lines. Seedlings of antisense RNS]
lines were grown as described in Figure 5-4, and total RNA was isolated from all
kanamycin-resistant seedlings. (A) RNA gel blots containing 10 pg of these samples per
lane were hybridized to an e1F4A probe and subsequently to the antisense RNS] probe. +,
seedlings grown on P,-rich medium; -, seedlings grown on P,-deficient medium; antisense,
antisense lines; control 1, empty vector control line; control 2, line containing construction
p1448 but with a near-normal level of RN81 activity. (B) RNS] and e1F4A counts were
quantitated by Phosphorimager analysis, and RNS] counts were divided by e1F4A counts
for each lane. The normalized counts for the - lanes were divided by those of the + lanes,
and the result plotted to show relative differences in RNS] mRNA levels in these lines.

134

oontrolt 301139039 control2
WICOI 34d.3 8d.5.2 8d.5.3 23g.4 36h.3.2

+-+-+-+-+-+-

 

 

RN81

44>

29>.

s. - < RN81
19>

Figure 5-6 - RN81 protein levels in RNS] antisense lines. Seedlings of antisense RNS]
lines were grown as described in Figure 5-4, and proteins were isolated from all kanarnycin-
resistant seedlings. 100 pg of each plant protein extract were separated by SDS-PAGE,
transferred to PVDF membrane, and immunodetected with anti-RN81 antiserum. +, grown
on P,-rich medium; -, grown on P,-deficient medium; RN81, yeast-produced RN81 control
(see Chapter 3). The RN81 band is indicated. Molecular weights of standards in kDa are
shown to the left of the gel.

135

RNase activity gels and irnmunoblots. In any case, it is clear that the antisense lines have
drastically reduced amounts of RN81 protein.

An interesting phenotype displayed by the three antisense RNS] lines is increased
accumulation of anthocyanins. When grown for seven days on P,-deficient medium, the
antisense lines contain from 2.6 to 5.1 times the amount of anthocyanins present in the
wild-type and 34d.3 control plants (Figure 5-7). In addition, lines 8d.5.2 and 8d.5.3, when
grown on P,-rich medium, contain amounts of anthocyanins comparable to those seen in P,-
deprived control plants. Interestingly, line 36h.3.2, which has RNS] mRNA levels between
those of the antisense lines and the control lines (see Figure 5-5), also has anthocyanin
levels between those of the two other groups of plants. The chalcone synthase gene, which
encodes an enzyme involved in anthocyanin biosynthesis, is induced by many stress-related
stimuli, such as high light intensity, pathogen attack, and wounding (reviewed in M01 et a1.
1996). However, anthocyanin accumulation is also a symptom of P, starvation
(Dedaldechamp et al. 1995; Marschner 1986). Although the elevated anthocyanin levels in
the antisense RNSI lines is not proof that their tissues are more starved for P, than control
plants are under the same conditions, the possibility is intriguing.

Line 23 g.4 has a phenotype in addition to the increased anthocyanin accumulation
described above. In the original T1 plant, as well as in some, but not all, of the descendants
of this plant, the following traits were observed in comparison to wild-type: smaller leaves,
shorter and thinner stems, reduced seed set, and a preference for multiple stems to appear
during the initial phase of bolting instead of the one dominant bolt that usually appears.

Interestingly, the phol mutant, examined in Chapter 3, has a similar phenotype, including

136

 

 

 

 

p+

Alfresh weight
it]
s
‘P

 

 

 

WW

 

 

s\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\‘

 

 

 

 

l
Colwt 8d.5.2 8d.5.3 239.4 34d.3 36h.3.2

Line

Figure 5-7 - Quantitation of anthocyanin levels in seedlings of RNS] antisense lines.
Seedlings of antisense RNS] lines were grown as described in Figure 5-4, and anthocyanins
were extracted as described in the Materials and Methods. Each bar encompasses four
independent readings. Absorbance at 530 nm minus absorbance at 657 nm was taken as a
measure of anthocyanin content, and was normalized to the fresh weight of each sample.
Empty bars, seedlings grown on P,-rich medium; hatched bars, seedlings grown on P,-
deficient medium.

I37

elevated anthocyanin levels, smaller leaves, thinner stalks, reduced seed set, less secondary
inflorescences and delayed flowering (Poirier et al. 1991). The similarities in these
phenotypes are an intriguing indication that the tissues of plants in the 23 g.4 line may be P,
starved.

As mentioned above, this phenotype has incomplete penetrance in the 23g line. It is
not known how many copies of the transgene were present in the intial 23 g T1 plant, but
segregation data suggest that it contained more than one transgene locus. One of three
progeny plants (T2 generation) grown in soil, a sibling plant to the 23 g.4 line, displayed the
altered phenotype, although the 23 g.4 plant itself did not. Finally, of three descendants (T3
generation) of the 23 g.4 plant grown in soil, one displayed the altered phenotype. These
observations are consistent with the effects of varying transgene dosage. Since the
molecular mechanisms of transfonnation by vacuum infiltration have not been elucidated, it
is not known whether this method gives rise to plants that are hemizygous or homozygous
for the transgene(s). A plant herrrizygous for multiple transgenes may give rise to progeny
that contain more or less transgenes than it had itself, and if antisense effects are
proportional to the amount of antisense transcript, as some studies suggest (Dougherty and
Parks 1995), progeny with a greater number of transgenes should display a more affected
phenotype. This idea is a simple explanation for the transient nature of the 23g phenotype,
but more complicated models cannot be ruled out. The above model would also explain
another phenomenon observed with both the 8d and 23g lines: only some of the progeny of
these lines exhibit reduced amounts of RN81 activity. It may be possible to test this model

using Southern blot analysis.

138

The isolation of the antisense RNS] lines described in this section is an important
step in determining the role of RN81 during P, starvation in plants. The visible phenotypes
of increased anthocyanin accumulation and, for the 23 g line, a phol-like phenotype, are
further clues that RN81 could be a major part of the phosphate rescue and recycling system
that has been proposed to exist in plants (Goldstein et al. 1989). These phenotypes also
suggest that the three RNS proteins do not have redundant functions, since reducing the
expression of only RNS] has a marked effect. One priority for the near future is to select a
line homozygous for the transgene(s), in which the antisense RNS] effect is strong and
stably inherited. Such a line could then be crossed to other pertinent lines in the firture, for
example an antisense RN82 line. Another interesting experiment would be the quantitation
of free P, in the antisense RNS] lines. If the tissues of these lines are indeed starved for P,,
there may be an effect on the amount of free P, in the plant as compared to the amount of P,

sequestered in nucleic acids.

Antisense Inhibition of the RNS2 Gene

The same approach used to produce antisense RNSI plants was used in the initial
attempt to obtain antisense RNS2 plants. As shown in Figure 5-8A, the entire RNS2 cDNA
was fused in reverse orientation into a T-DNA construction analagous to that made for the
antisense RNSI construction. 119 independent transforrnants of this plasmid were screened
by assaying leaf proteins of soil-grown plants for lowered RN 82 levels with irnmunoblots

and anti-RN82 antibodies (see Chapter 4). In this population of transgenic plants, RN82

139

A

020

1
/, (Inosl NPTI|(kan mlhosj { 353

 

 

more

1
asuu RU [ ass 1 GUS nos} 6 \

 

 

p1449 I

977 329

R}-—1~1PTll(k ethics} [ 35's zsuu 6H 1 s] " '7 LB
an . . ,
/-- \i’ . l . -°,;.-°l“l 35. , GUS Fifi/UN

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

p1525 1

l

\ /

424 227
R816 _. .- V. , LB

.. __<j Enos: NPTIItkan R) nos i 358 ZSNB 998p 35$ GUS nosl,_<:|.____ , \
/ \‘1
p1527 1

I

Figure 5-8 - Structure of antisense RNS2 plant transformation vectors. Various portions of
the RNS2 cDNA are fused to the cauliflower mosaic virus 358 promoter and a nos
terminator, in three separate constructions. All other elements are as described in Figure 5-
1. Shaded elements indicate promoters, hatched elements indicate terminators, and empty
boxes indicate coding regions. A) Construction p1449. B) Construction p1525. C)
Construction pl 527.

140

levels varied to a small extent, but no plants with dramatically lowered amounts of RN82
were observed.

There is no consensus in the literature as to which portions of genes to use to obtain
the best results with antisense techniques in plants (Bourque 1995). The size and portion of
an open reading frame which results in the greatest antisense effects appears to be gene-
dependent. For this reason, the above results were not interpreted to mean that no antisense
RN82 plants could be obtained, and so two additional portions of the RNS2 cDNA were
fused into antisense constructions for plant transformation in the hopes that these portions
would be more successful in suppressing RN 32 production. The first construction, p1525
(Figure 5-8B) included the final two-thirds of the RNSZ cDNA, a fragment of about 650
base pairs, and the second, p1527 (Figure 5-8C), incorporated a fragment of approximately
200 base pairs between and including the conserved active site regions (see Chapter 2).

Independent transforrnants, 74 of p1525 and 63 of p1527, were screened by
analyzing leaf RNAs of soil-grown plants for lowered amounts of RNS2 mRNA. This
method replaced the irnmunoblotting method used earlier because of the greater relative
ease of quantitation and normalization of the RNS2 signal. RN52 signal strength was
divided by that of the e1F4A signal in the same sample, and this number was compared to
the corresponding number from a control sample on the same blot. Of the 137 plants
screened, only 19 had leaf RN82 mRNA levels less than or equal to 70% that of wild-type
plants, and of the 19, only four had less than 35% that of wild-type. Several antibiotic-
resistant progeny plants (1? generation) from each of these 19 plants were grown in soil and

leaf mRN As were analyzed for RNS2 as done for the T1 generation. Progeny of only four

 

141

of the 19 T1 plants showed lowered leaf RNS2 mRNA levels; RNA gel blot analysis of
several of these progeny plants is shown in Figure 5-9. In these plants, levels of RNS2
mRNA range from 39% to 68% of that in wild-type. Three of these four lines, resulted
from transformation with plasmid pl 527.

The lines shown in Figure 5-9 display no unusual phenotype when grown under
normal conditions in soil. Progeny from two of the four lines were observed to have leaf
anthocyanin levels slightly higher than normal, but this effect was not examined
quantitatively. Due to a lack of time, it was not possible to test the reactions of these plants
to phenomena known to affect RNS2 expression, such as Pi starvation and inoculation with
pathogens. The lack of obvious phenotypes in these plants may be due to the incomplete
suppression of RNS2 expression: the inhibition of RNS2 to only 39 to 68% that of wild-
type may not be sufficient to affect the plant because an adequate amount of RN82 protein
is still being produced. Altemately, the reduction of RNSZ levels may be compensated by
the actions of other RNases in plants. In E. coli, for example, it is known that molecularly
distinct RNases can participate in similar processes, and that the knockout of all the
functionally overlapping RNases is required to produce a measurable effect (Deutscher
1993a).

It is possible that the lines shown in Figure 5-9 could be manipulated to contain
lower levels of RNSZ than they do at present. It is unknown whether these 12 lines are
hemizygous or homozygous for the transgene, although all are progeny of hemizygous T1
plants. Selection of homozygous progeny of these lines could lead to plants in which

antisense effects are stronger due to a higher transgene dosage. Along similar lines, it has

142

wt

74! 4
79¢ 3
796.5
66c 6
790 4
790 6

CID...- (RN82

...... . <eIF4A

 

.5
l

  

W

   
 

l

   

/////////////

      

Relative RN82 mRNA level

 

 

WW

7///////////////////////
W

W

   

Figure 5-9 - RNS2 mRNA levels in RNS2 antisense lines. (A) Total RNA was extracted
from leaves of four-week-old, kanarnycin resistant, soil-grown antisense RNSZ lines. RNA
gel blots containing 6 pg of these samples per lane were hybridized to an antisense RNS2
probe and subsequently to the e1F4A probe. (B) RNS2 and e1F4A counts for the bands
shown in (A) were quantitated by Phosphorimager analysis. RNS2 counts were divided by
e1F4A counts for each lane, and these results were divided by the RNS2/eIF 4A ratio for the
wild-type control lane to show relative differences in RN82 mRNA levels. wt, Col wt:
Columbia wild-type samples.

143

been shown that crossing of two independent antisense plants and thus increasing transgene
dosage can result in a greater antisense effect (Pennisi 1996). It remains to be seen whether

these techniques will lower RNSZ levels further.

Overexpression of RNS] in Roots

To overexpress RNS] in roots, the full-length RNS] cDNA was fused to a root-
specific promoter from a soybean proline-rich protein (Suzuki et a1. 1993). Specific
expression in the root was desired to avoid any potential deleterious effects of
overexpression of RNSl in other organs of the plant. It was expected that the endogenous
RNSl N-terminal sequence would be sufficient to target the protein for secretion. Although
it has not been proven that RNS] is an extracellular protein, its N-teiminal sequence has a
high score as a secretory signal sequence according to a statistical analysis (von Heijne
1986). In addition, RNSl has been purified from culture medium of Arabidopsis seedlings
grown under Pi-deficient conditions (J. Paz-Ares, personal communication).

The vector (shown in Figure 5-10) was transformed into Arabidopsis as described
above for the antisense constructions. Primary transforrnants were grown in soil, their seeds
collected, and this generation (T2 generation) screened for increased RNS] transcript levels
in roots. Pooled T2 seedlings from each T1 line were grown on standard growth medium
containing antibiotic selection, with the plates placed vertically to maximize root growth
such that more tissue was available for analysis. After about four weeks of growth, roots
were collected and total RNA extracted for RNA gel blot analysis. All of the lines screened

contained measurably higher RNS] mRNA levels than control plants. A selection of lines

144

 

 

 

/,--m ‘sbp’am' ‘ RN31 i-~;’,I5§$1'.I.3ii;~

 

 

 

l
BinAR-SbPRPt-RNS1 }

Figure 5-10 - Structure of RNS] root overexpression transformation vector. The entire
RNS] cDNA is fused to the root-specific SbPRPl promoter and an ocs terminator.
SbPRPl , soybean proline-rich protein promoter (Suzuki et al. 1993); ocs, octopine
synthase. Shaded box indicates a promoter, hatched box indicates a terminator, empty box
indicates a coding region.

This plasmid was constructed by Dr. Marcel Bucher

145

with a representative range of the RNS] expression levels seen is shown in Figure 5-11:
these lines express RNS] between a range of 2.3 to 6.6 times that of the control line. None
of the lines showed an unusual phenotype when grown in soil, although lines 583 and 59C,
the lines with the highest root RNS] expression, did have slightly smaller rosettes than other
lines.

To determine whether lines with these amounts of RNS] expression would be more
competitive in Pi-poor soils, it was first necessary to compare their expression to the
endogenous expression of RNS] in roots of Pi-starved plants. Arabidopsis wild-type seeds
were grown on media rich or deficient in Pi (media formations given in Chapter 3) on plates
placed vertically as done for the RNS] root overexpression lines. Seedlings grown on P,-
rich medium appeared to grow normally, much like control plants growing on AGM
medium, but seedlings grown on Pi-deficient medium had both stunted green tissue and
stunted roots. Pi-starved roots were approximately 20% the length of roots grown on Pi-rich
medium, and often grew partially off the medium into the air, an unusual response. It is
known that Pi starvation can have large effects on root morphology in other plants (Johnson
et al. 1996b). After about four weeks of growth, roots were harvested and RNS] expression
was measured using RNA gel blots as described above, and shown in Figure 5-12A. Under
Pi-deficient conditions, RNSI is induced about 6.6-fold, much less than the induction of
approximately 75-fold that is seen in total seedling RNA (compare with Figure 3-11). This
level of induction is very similar to that of the most efficient RNS] root overexpression lines

shown in Figure 5-10. Proteins from root tissues of the same batch were resolved on RNase

146

control
54E
57A
588
590

RN81) a p .

 

ielF4A> .

——- 1 u‘ “- ‘— — .-

 

llllllllLl¥LL

 

 

Relative RN81 mRNA level
0)
L
W
W

8

Control 548 57A 5

O

85

Line

Figure 5-11 - RNS] mRNA levels in RNS] root overexpression lines. (A) Total RNA was
extracted from roots of four- to five-week-old transgenic plants containing the RNS! root
overexpression construction that were grown on kanamycin-containing medium. RNA gel
blots containing 9 pg of these samples per lane were hybridized to the RNS] probe and
subsequently to the e1F4A probe. (B) RNS] and e1F4A counts for the bands shown in (A)
were quantitated by Phosphorimager analysis. RNS] counts were divided by e1F4A counts
for each lane, and these results were divided by the RNS] /eIF 4A ratio for the wild-type
control lane to show relative differences in RNS] mRNA levels. Control, transgenic line
34d.3, which contains the pB1121 vector.

147

P+
p-
RN31
P+

<43

<29

RN81) u.

 

(18

e|F4A > . .

Figure 5-12 - Effect of Pi starvation on RNS] mRNA levels and RNSl activity in roots.
(A) Total RNA was extracted from roots of wild-type plants grown on Pi-rich or Pi-deficient
media for four to five weeks. RNA gel blots containing 10 pg of these samples per lane
were hybridized to the RNS] probe and subsequently to the e1F4A probe. (B) Proteins were
extracted from the same batches of roots and 50 pg of each sample were resolved on an

RNase activity gel. RNSl, RNSl protein expressed from yeast. Molecular weights of
standards in kDa are shown to the right of the gel.

148

activity gels, as shown in Figure 5-12B, and the extent of the induction of RNSl activity is
comparable to that observed at the RNA level.

Since the attempt to increase RNS] expression in Arabidopsis roots resulted in
plants that express RNS] at about the same level as wild-type plants under Pi-deficient
conditions, this strategy may not be useful in engineering crop plants able to tolerate low-Pi
conditions in organic soils. However, RNS] root expression in these lines may be different
under more natural conditions, such as in soil, or in other plants. At present, the effect of
the RNS] overexpression construction is being tested in tobacco plants by Dr. Marcel
Bucher. In addition, this idea may be more feasible if one were to use a stronger root-
specific promoter, should one become available.

An interesting observation made during these studies can be seen in Figure 5-12.
An RN ase activity of about 33 kDa is induced strongly, and to a much greater extent than
RNSl, in Pi-starved roots. This activity is in the range of a pair of bifunctional nuclease
activities, believed to be secreted, that are observed in Arabidopsis stem extracts. The
strong upregulation of this activity during Pi starvation may constitute part of the
endogenous plant system to scavenge P, from the rhizosphere. This possibility merits

further study.

149

MATERIALS AND METHODS

Plasmid Constructions

p1448: The entire RNS] cDNA was excised with Barn HI and Sal I from plasmid p1184
and inserted into pT7T3a19 (BRL) between the Barn HI and Sal I sites to make p1392. A
Pst I fi‘agment of p1392, containing the RNS] cDNA, was fused between the Pst I sites of
p1079 (provided by Wan-ng Chiu), and a plasmid with the cDNA in the antisense
orientation relative to the 35 S promoter was designated pl398. p1398 was partially
digested with Hind HI and the ends of linearized forms of the plasmid converted to blunt
ends with T4 DNA polymerase (Sambrook et al. 1989), after which the DNA was
recircularized. A plasmid was selected such that the Hind 111 site flanking the cDNA was
destroyed; this plasmid was designated p1444. The Hind III cassette of p1444, containing
the RNS] cDNA firsed between the 35S promoter and nos terminator, was inserted into the
Hind 111 site of pB1121 (Jefferson 1987). A plasmid in which the RNS] cassette was

oriented in the same direction relative to the GUS cassette of pB1121 was designated p1448.

p1449: The entire RNS2 cDNA was excised with Barn HI and Kpn I from plasmid pl 127
and inserted into pUC118 (Sambrook et al. 1989) between the Barn HI and Kpn I sites to
make p1393. A Sal I fragment of p1393, containing the RN52 cDNA, was fused between
the Sal I sites of p1079, and a plasmid with the cDNA in the antisense orientation relative to
the 358 promoter was designated p1399. p1399 was partially digested with Hind III and the

ends of linearized forms of the plasmid converted to blunt ends with T4 DNA polymerase

150

(Sambrook et al. 1989), after which the DNA was recircularized. A plasmid was selected
such that the Hind 111 site flanking the cDNA was destroyed; this plasmid was designated
p1445. The Hind III cassette of p1445, containing the RNS2 cDNA fiised between the 358
promoter and nos terminator, was inserted into the Hind 111 site of pBIlZl (Jefferson 1987).
A plasmid in which the RNS2 cassette was oriented in the same direction relative to the

GUS cassette of pBI121 was designated p1449.

p1525: A 648 bp Sca I-Eco RV fiagment of the RNS2 cDNA, comprising approximately
the last two-thirds of the cDNA and including sequences encoding the conserved regions
C3-C5 (see Figure 2-4), was excised from plasmid p1127. This fragment was inserted into
the p1079 backbone, which had been digested with Pst I and whose ends had been
converted to blunt ends using T4 DNA polymerase (Sambrook et a1. 1989). A plasmid
containing the fragment inserted in the antisense orientation relative to the 35S promoter
was designated p1524. The Hind III cassette of pl 524, containing the above fragment fused
between the 358 promoter and nos terminator, was inserted into the Hind III site of pB1121
(Jefferson 1987). A plasmid in which the cassette was oriented in the same direction

relative to the GUS cassette of pB1121 was designated pl 525.

p152 7: A 208 bp Eco RI-Xho I fiagment of the RNS2 cDNA, comprising sequences
encoding the conserved regions C2 and C3 (see Figure 2-4) and the sequences between
them, was excised from plasmid p1010-8. This ends of this fragment were converted to

blunt ends using T4 DNA polymerase (Sambrook et al. 1989), and it was inserted into the

151

p1079 backbone between blunted Pst I sites. A plasmid containing the fragment inserted in
the antisense orientation relative to the 358 promoter was designated p1526. The Hind III
cassette of p1526, containing the above fragment fused between the 358 promoter and nos
terminator, was inserted into the Hind 111 site of pBIlZl (Jefferson 1987). A plasmid in
which the cassette was oriented in the same direction relative to the GUS cassette of pB1121

was designated p1527.

BinAR-SbPRPI -RNS1 .' Plasmid provided by Marcel Bucher. See Figure 5-10.

Generation of Transgenic Plants

Plasmids p1448, p1449, p1525, p1527 and BinAR-SbPRPl-RNSl were
transformed into Agrobacterium tumefaciens strain GV3101 C58C1 Rif (pMP90) (Koncz
and Schell 1986) via electroporation using a Gene-Pulser apparatus (Bio-Rad) as per
manufacturer's recommendations. The plasmids were inserted into Arabidopsis thaliana
ecotype Columbia with a vacuum infiltration method of Agrobacterium-mediated
transformation. Protocols for this method by N. Bechtold (Bechtold et al. 1993), A. Bent
(Bent et al. 1994), and T. Araki were modified or combined. Briefly, rosettes of four-week
old plants with bolts of 5-15 cm were submerged in a solution of A. tumefaciens containing
the plasmid of interest and subjected to a vacuum of 400 mm Hg for five minutes. The
vacuum was quickly broken and the plants were allowed to recover and set seed under
normal growth conditions. Details of this protocol can be viewed on the World Wide Web

(http://www.bch.msu.edu/pamgreen/green.htm#prot). Seeds fi'om these plants were plated

152

on kanamycin-containing medium, and one antibiotic-resistant seedling from each plant that
was originally infiltrated was transferred to soil, to ensure that all plants analde were the
results of unique integration events. To ensure unbiased selection of transfonnants to be
transferred to soil, the antibiotic-resistant plant closest to the edge of the plate was selected

in each case.

Anti-RNS] Antibody Preparation

Synthetic peptide PGZ was prepared at the Michigan State University Department of
Biochemistry Macromolecular Structure Facility, and its purity was monitored by analytical
reversed-phase HPLC and mass spectrometry. Peptide PG2 was coupled to maleimide-
activated KLH (Pierce) before injection. RNS] protein used as an antigen was
heterologously produced in yeast as described in Chapter 3. 250 ml of liquid minimal
dextrose low-Pi medium (Thill et al. 1983) were inoculated with S. cerevisiae strain
BJ2168 containing plasmid p1270 and grown until saturation (2.5 days). Cells were
removed by centrifugation, the supernatant was concentrated to 0.5% of its original volume
in Centriprep-lO units (Amicon), and the buffer replaced with 20 mM MES pH 6.0. The
predicted amino acid sequence of RNSl is predicted to bind to an anion exchange column
at pH 6.0 (analysis done using the Prosis program). The proteins in the supernatant were
bound to a Mono Q FPLC column (Pharmacia), eluted with a gradient of 0 to 0.25 M NaCl,
and fractions analyzed by SDS-PAGE and silver staining. RNSl eluted between 0.15-0.18

M NaCl. Pooling of the RNSl-containing fractions resulted in a preparation >95% pure in

153

RNSl, as monitored by SDS-PAGE and silver staining. Purity was also checked by
isoelectric focusing and silver staining.

Proteins were injected subcutaneously into female New Zealand white rabbits and
blood collected as described in Chapter 4. Two rabbits were used for each antigen. The
first injections of PG2 coupled to KLH were of 280 pg, and after three weeks, boosts of 350
pg were administered. The first injections of heterologously-produced RNSl protein were
of 80 pg, followed by 300 pg boosts. Sera were screened for anti-RNS] binding ability by

testing various dilutions on irnmunoblots containing yeast-produced RNSl (Chapter 3).

Expression Analyses

Antisense RNS] plants: 120 kanamycin-resistant independent transforrnants of p1448 were
analyzed for reduced RN S1 activity. Original transforrnants (T1 generation) as well as
wild-type plants were grown in soil under conditions described in Chapter 3. Flowers were
collected from four- to five-week-old plants on one day only per plant, collecting all flowers
on the plant that were in the range of development from buds with petals showing to fully-
open flowers that do not yet have developing siliques protruding. Flowers were frozen on
dry ice and stored at -80° C. Flower proteins were extracted as described for other tissues in
Chapter 3. RN 81 activity was analyzed by electrophoresing 20 pg of flower proteins from
each transfonned line on RNase activity gels (Yen and Green 1991); the intensity of the
RNase activity band likely to be RNSl was compared in transformed lines and wild-type.
For those lines with lower flower RNSl activity, seeds were collected, one kanamycin-

resistant progeny plant (T2 generation) was grown in soil and their flowers analyzed for

154

RN 8] activity just as done for the T1 generation. Seeds from those lines that still appeared
to have lowered flower RNSI activity were once again selected for kanamycin resistance,
and several resistant plants (T3 generation) were moved to soil and flowers screened for
lowered RN 8] activity. In some cases, the T4 generation was screened in the same way.
Seeds of promising T3 or T4 lines were plated on AGM containing kanamycin on
mesh circles, moved after two days to kanamycin-containing Pi-rich or Pi-deficient media,
and harvested seven days later, as described for wild-type seeds in Chapter 3. Control lines
included in these experiments were Columbia wild-type (grown on media without
kanamycin) and transgenic lines containing just the pBIlZl vector. Any kanamycin-
sensitive seedlings were removed prior to harvesting. Harvested seedlings were fi'ozen in
liquid nitrogen for RNA extraction as described in Chapter 2, or on dry ice for protein
extraction, as described in Chapter 3. RNA gels were prepared and blotted to membrane as
in Chapter 2, and then RNA gel blots were probed first with the e1F4A probe as described in
Chapter 2. The blots were then stripped and re-probed with an RNS] RNA probe such that
only sense RNA strands would be detected. The RNA probe was made using a Riboprobe
kit (Promega), with plasmid p1184 linearized with Bam HI, and T7 RNA polymerase.
Quantitation of signal in RNSI and e1F4A bands was done using Phosphorimager analysis.
Proteins were extracted fi'om seedlings and electrophoresed on RN ase activity gels as
described in Chapter 3, and were electrophoresed and blotted to membrane for irnmunoblots
as described in Chapter 4. For irnmunoblots, a 1:1000 dilution of serum from rabbit 55315

was used.

155

Antisense RN S2 plants: For the first strategy, 119 kanamycin-resistant independent
transforrnants of p1449 were analyzed for lowered amounts of RN82 by irnmunoblotting.
Original transforrnants (Tl generation) as well as wild-type plants were grown in soil under
conditions described in Chapter 3. Several healthy, non-senescing leaves were collected
from four- to five-week-old plants on one day only per plant, frozen on dry ice and stored at
-80° C. Proteins were extracted from the leaves as described in Chapter 3. 50 pg of leaf
proteins fiom each transformed line, as well as wild-type, were electrophoresed on SDS-
PAGE gels, blotted to PVDF membrane and immunodetected with anti-RN82 antibodies as
described in Chapter 4. Judgements were made by eye as to whether RN S2 bands for each
lane were less intense than the wild-type RN 82 band on the same blot.

In the next screen for antisense RNS2 plants, 74 kanamycin-resistant independent
transforrnants of p1525 and 63 of p1527 were screened for lowered RNS2 mRNA levels by
RNA gel analysis. Original transforrnants (T1 generation) and wild-type plants were grown
in soil and leaves harvested as for the p1449 transfonnants, except that the harvested leaves
were frozen in liquid nitrogen. Total RNA was extracted from leaves and RNA gel blots
prepared as described in Chapter 2. RNA gel blots were probed first with the e1F4A probe
as described in Chapter 2, and were then stripped and re-probed with an RNS2 RNA probe
such that only sense RNA strands would be detected. This probe was made using a
Riboprobe kit (Promega), with plasmid p1127 linearized with Eco RI, and T3 RNA
polymerase. Levels of RNS2 and e1F4A mRN A were measured by Phosphorimager
analysis, the RNS2/e1F4A ratio was calculated for each line, and these numbers were

divided by the RNS2/e1F4A ratio of the wild-type sample on the same blot to calculate the

156

relative level of RNS2 mRNA in putative antisense RNS2 lines. Lines in which the level of
RNS2 mRNA was less than or equal to 70% of that in wild-type were selected for re-
screening. Several kanamycin-resistant progeny (T2 generation) were grown in soil, leaves

harvested, and relative RNS2 mRNA levels calculated as done for the T1 generation.

RNS] root overexpression plants: Independent initial transformed lines (T1 generation)
containing the BinAR-SbPRPl-RNS] construction were grown in soil. Seeds (T2
generation) from ten of these lines were screened for overexpression of RNS] in roots, as
compared to a transgenic line containing only the pBIlZl vector. Approximately 200
seedlings from each line were grown on AGM medium containing kanamycin; the plates
were placed vertically such that roots grew vertically down the surface of the medium.
After four to five weeks of growth at 22° C and a daylength of 16 hours light/8 hours dark,
roots were excised, frozen in liquid nitrogen and stored at -80° C. RNA extraction was
done and RNA gel blots were prepared as described in Chapter 2. RNA gel blots were
probed first with an RNS] random-primed probe, and then stripped and re-probed with the
e1F4A probe, as described in Chapter 3. Levels of RNS] and e1F4A mRN A were measured
by Phosphorimager analysis, the RNSI/eIF 4A ratio calculated for each line, and these
numbers divided by the RNSI/eIF 4A ratio of the control line on the same blot to calculate
the relative level of RNS] mRN A in putative RNS] root overexpression lines.

To compare endogenous RNS] expression levels in roots with the transgenic lines
described above, Columbia wild-type seeds were grown on Pi-rich and Pi-deficient media

(Chapter 3) on plates placed vertically under the same conditions as above. Roots were

157

collected from four- to five-week-old plants. From one portion of each set, RNA was
extracted and RNA gel blots were prepared and probed as described above. Proteins were
extracted from another portion of each set and resolved on RNase activity gels as described

in Chapter 3.

Anthocyanin Assays

For assay of anthocyanin content in antisense RNS] lines, seeds of T3 or T4 lines
were grown on Pi-rich or Pi-deficient media and harvested seven days afier transfer as
described in the previous section. Fresh weight was recorded for each sample. Seedlings
were frozen in liquid nitrogen, lyophilized in 13 ml plastic test tubes (Sarstedt) and
pulverized with 3 mm diameter glass beads (Fisher). Anthocyanin content from each line
was measured using a protocol based on the methods of Rabino and Mancinelli (1986),
Feinbaum and Ausubel (1988), and Kubasek et al (1992). Ground tissue was shaken in 2.5
ml of 1% HCl/methanol for two hours at room temperature. 2 ml of chloroform were added
and the mixture vortexed, after which 5 ml of H20 were added and the vortex repeated.
After separating the phases by centrifugation, 1 ml of the aqueous/methanol phase was
assayed using a Beckman DU Series 600 spectrophotometer. Absorbance at 530 nm minus
absorbance at 657 nm was used as a measure of anthocyanin content; values were
normalized to the fresh weight of each sample. Two separate experiments were done, each
including four plates for each line (two on Pi-rich and two on Pi—deficient media) such that

four readings were incorporated in each instance.

CHAPTER 6

CONCLUSIONS AND FUTURE PROSPECTS

158

159

The discovery of genes for S-like RNases in plants (Taylor and Green 1991) was the
first indication that T2/ S type RNases are present in self-compatible plants, and opened up
many questions about the roles of these types of RNases in plants. An initial goal of this
thesis project was to compare the protein structures and gene expression of the RNS family
in Arabidopsis with those of each other and other plant T2/S RNases. These studies
revealed important structural differences between the S-RNases and the S-like RNases
(Chapter 2). Moreover, the RNS genes are induced during senescence, and RNSI and RNS2
are induced during Pi starvation (Chapters 2 and 3). The induction of RNS2 by starvation
for P, was one of the first reports of a Pi-starvation inducible transcript, and the first for
which a physiological role could be proposed (in remobilization of Pi). The RNS] and
RNS2 cDNAs have since been used in several other laboratories as molecular markers for Pi
starvation and senescence (e. g. Callard et al. 1996).

These observations led to the next phase of this thesis project, in which experiments
designed to further elucidate potential roles of the RN S proteins were begun. The
localization of RN82 to the cell wall (Chapter 4) is an important first step in pinpointing the
cellular role(s) of RN82. If antisense plants with lower amounts of RNS2 than those
described in Chapter 5 can be obtained, possibly by crossing separate antisense lines or
selection of homozygous lines, the proposals that RN82 is involved in phosphate and/or
nucleoside remobilization, and possibly plant defense, can be more easily tested. A
selection of bacterial and fungal plant pathogens should be tested, since defense reactions to
different pathogens can vary. Should it not be possible to obtain antisense plants with

significantly decreased amounts of RN82, it may be feasible to obtain a transgenic line

160

whose RNS2 gene has been disrupted by T-DNA insertional mutagenesis, using a recently
developed method (Krysan et al. 1996).

The RNS] gene is much more strongly induced during Pi starvation than the RN82
gene, and so may have a more crucial role in the response of the plant to this stress
condition. Antisense plants with low amounts of RNSl have been obtained and preliminary
characterization on the lines performed (Chapter 5). Some characteristics of these lines,
namely increased anthocyanin deposition and a sometimes-seen phoI -like phenotype,
indicate that these plants may be impaired in the Pi starvation response, but more
characterization of the lines is needed. Possible future experiments would be the
determination of free Pi levels in antisense and control plants grown in media containing
varying concentrations of P,, or perhaps media in which RNA is included as the sole F,
source. Similarly, a dose-response experiment in which plants would be grown on media
containing decreasing amounts of Pi would be useful for monitoring with other molecular
markers that indicate Pi starvation, such as the PAP] acid phosphatase gene of Arabidopsis
(T. McKnight, personal communication). In this way, it may be possible to determine if
antisense RNS] plants begin to show symptoms of starvation for Pi at higher P, levels than
control plants.

With the completion earlier this year of the sequencing of the yeast genome, a yeast
member of the T2/ S RN ase family has recently been identified. Targeted inactivation
(Rothstein 1991) of this gene is in progress. Characterization of the effects of this
inactivation, if any, and subsequent complementation with the plant RNS genes, may

provide insight into the cellular roles of the RNS group of RNases.

161

The study of plant RNases has the potential to provide much information on plant
RNA metabolism and nutrient remobilization. In addition to these contributions to basic
sciences, RNases could have applied significance. For example, male sterile maize
(Mariani et al. 1990) and corresponding restorer lines (Mariani et al. 1992) have been
created by expressing heterologous RN ase and RNase inhibitor genes in appropriate cell
types. Heterologous expression of a double-stranded RNA activated RN ase system in
plants has shown promise in protecting plants against RNA viruses (Ogawa et al. 1996).
RNases is also being explored as anti-tumor (Wu et al. 1993) and anti-AIDS virus (Saxena
et al. 1996) agents. As most of the families of plant RNases have yet to be characterized
molecularly, fiirther research in this field should contribute to basic and possibly applied

science.

LIST OF REFERENCES

162

LIST OF REFERENCES

Abel S, Blume B, Glund K (1990) Evidence for RNA-oligonucleotides in plant vacuoles
isolated from cultured tomato cells. Plant Physiol 94:1163-1171
purification and molecular

Abel S, Glund K (1987) Ribonuclease in plant vacuoles:
properties of the enzyme from cultured tomato cells. Planta 172:71-78

Ai Y, Singh A, Coleman CE, Ioerger TR, Kheyr-Pour A, Kao T-H (1990) Self-
incompatibility in Petunia inflata: isolation and characterization of cDNAs
encoding three S-allele-associated proteins. Sex Plant Reprod 3: 130-138

Ai Y, Tsai D-S, Kao T-H (1992) Cloning and sequencing of cDNAs encoding two S
proteins of a self-compatible cultivar of Petunia hybrida. Plant Mol Biol 19:523-

528
Alexander D, Lawton K, Uknes S, Ward E, Ryals J (1994) Defense-related gene induction

in plants. In: Setlow JK (ed) Genetic Engineering. Plenum Press, New York, pp
195-212
Altman S, Kirsebom L, Talbot S (1993) Recent studies of ribonuclease P. F ASEB J 7:7-14

Anderson MA, McFadden GI, Bematzky R, Atkinson AH, Orpin T, Dedman H, Tregear G,
F ernley R, Clarke AE (1989) Sequence variability of three alleles of the self-

incompatibility gene of Nicotiana alata. Plant Cell 1:483-491

Arima K, Oshima T, Kubota I, Nakamura N, Mizunaga T, Toh-e A (1983) The nucleotide
sequence of the yeast PH05 gene: a putative precursor of repressible acid

phosphatase contains a signal peptide. Nucleic Acids Res 11:1657-1672

Atkinson AH, Heath RL, Simpson RJ, Clarke AE, Anderson MA (1993) Proteinase
inhibitors in Nicotiana alata stigrnas are derived from a precursor protein which is

processed into five homologous inhibitors. Plant Cell 5:203-213

Barbieri L, Battelli MG, Stirpe F (1993) Ribosome-inactivating proteins from plants.
Biochim Biophys Acta 1154:237-282

163

Bariola PA, Howard CJ, Taylor CB, Verburg MT, Jaglan VD, Green PJ (1994) The
Arabidopsis ribonuclease gene RNS] is tightly controlled in response to phosphate
limitation. Plant J 6:673-685

Barkan A, Walker M, Nolasco M, Johnson D (1994) A nuclear mutation in maize blocks
the processing and translation of several chloroplast mRN As and provides evidence
for the differential translation of alternative mRNA forms. EMBO J 13:3170-3181

Bama B, Ibentlral WD, Heitefuss R (1989) Extracellular RNase activity in healthy and rust
infected wheat leaves. Physiol Mol Plant Pathol 35: 1 51-160

Bechtold N, Ellis J, Pelletier G (1993) In planta Agrobacterium mediated gene transfer by
infiltration of adult Arabidopsis thaliana plants. C R Acad Sci Paris 316:1194-1 199

Bednarek SY, Raikhel NV (1992) Intracellular trafficking of secretory proteins. Plant Mol
Biol 20:133-150

Beelrnan CA, Parker R (1995) Degradation of mRNA in eukaryotes. Cell 81 :179-1 83

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

Binder R, Horowitz JA, Basilion JP, Koeller DM, Klausner RD, Harford JB (1994)
Evidence that the pathway of transferrin receptor mRN A degradation involves an

endonucleolytic cleavage within the 3' UTR and does not involve poly(A) tail
shortening. EMBO J 13:1969-1980

Birkett CR, Foster KE, Johnson L, Gull K (1985) Use of monoclonal antibodies to analyse
the expression of a multi-tubulin family. FEBS Lett 187:211-218

Blank A, McKeon TA (1989) Single-strand-preferring nuclease activity in wheat leaves is
increased in senescence and is negatively photoregulated. Proc Natl Acad Sci USA
86:3169-3173

Blank A, McKeon TA (1991a) Three RNases in senescent and nonsenescent wheat leaves.
Characterization by activity staining in sodium dodecyl sulfate-polyacrylarnide gels.
Plant Physiol 97: 1402-1408

Blank A, McKeon TA (1991b) Expression of three RNase activities during natural and
dark-induced senescence of wheat leaves. Plant Physiol 97:1409-1413

164

Boller T, Kende H (1979) Hydrolytic enzymes in the central vacuole of plant cells. Plant
Physiol 63:1123-1 132

Boller T, Wiemken A (1986) Dynamics of vacuolar compartrnentation. Annu Rev Plant
Physiol 37:137-164

Borochov A, Woodson WR (1989) Physiology and biochemistry of flower petal
senescence. Hort Rev 11:15-43

Bourque JE (1995) Antisense strategies for genetic manipulations in plants. Plant Sci
105:125-149

Bowles DJ (1990) Defense-related proteins in higher plants. Annu Rev Biochem 59:873-
907

Brake AJ, Merryweather JP, Coit DG, Heberlein UA, Masiarz FR, Mullenbach GT, Urdea
MS, Valenzuela P, Barr PJ (1984) or-factor-directed synthesis and secretion of

mature foreign proteins in Saccharomyces cerevisiae. Proc Natl Acad Sci USA
81 :4642-4646

Broothaerts W, Janssens GA, Proost P, Broekaert WF (1995) cDNA cloning and molecular
analysis of two self-incompatibility alleles from apple. Plant Mol Biol 27 :499-51 1

Brown BD, Harland RM (1990) Endonucleolytic cleavage of a maternal homeo box mRN A
inXenopus oocytes. Genes Dev 4:1925-1935

Brown PH, Ho T-HD (1986) Barley aleurone layers secrete a nuclease in response to
gibberellic acid. Plant Physiol 82:801-806

Brown PH, Ho T-HD (1987) Biochemical properties and hormonal regulation of barley
nuclease. Eur J Biochem 168:357-364

Bufe A, Schramm G, Keown MB, Schlaak M, Becker W-M (1995) Major allergen Phl p
Vb in timothy grass is a novel pollen RNase. FEBS Lett 363:6-12

Bufe A, Spangfort MD, Kahlert H, Schlaak M, Becker WM (1996) The major birch pollen
allergen, Bet v 1, shows ribonuclease activity. Planta 199:413-415

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

Callard D, Axelos M, Mazzolini L (1996) Novel molecular markers for late phases of the
growth cycle of Arabidopsis thaliana cell-suspension cultures are expressed during
organ senescence. Plant Physiol 112:705-715

165

Cassab GI, Varner JE (1988) Cell wall proteins. Annu Rev Plant Physiol Plant Mol Biol
39:321-353

Chen H-C, Stern DB (1991) Specific ribonuclease activities in spinach chloroplasts promote
mRNA maturation and degradation. J Biol Chem 266:24205-24211

Chrispeels MJ, Raikhel NV (1992) Short peptide domains target proteins to plant vacuoles.
Cell 68:613-616

Chung I-K, Ito T, Tanaka H, Ohta A, Nan HG, Takagi M (1994) Molecular diversity of
three S-allele cDNAs associated with gametophytic self-incompatibility in
Lycopersicon peruvianum. Plant Mol Biol 26:757-762

Clark KR, Okuley JJ, Collins PD, Sims TL (1990) Sequence variability and developmental
expression of S-alleles in self-incompatible and pseudo-self-compatible petunia.
Plant Cell 21815-826

Collins MKL, Rivas AL (1993) The control of apoptosis in mammalian cells. Trends
Biochem Sci 18:307-309

Constabel CP, Brisson N (1995) Stigrna- and vascular-specific expression of the PR-IOa
gene of potato: A novel pattern of expression of a pathogenesis-related gene. Mol
Plant Microbe Interact 8:104-113

Cornish EC, Pettitt JM, Bonig 1, Clarke AE (1987) Developmentally controlled expression
of a gene associated with self-incompatibility in Nicotiana alata. Nature 326:99-102

De A, Funatsu G (1992) Crystallization and preliminary X-ray diffraction analysis of a
plant ribonuclease from the seeds of the bitter gourd Momordica charantia. J Mol
Biol 228:1271-1273

Dean C, Tamaki S, Dunsmuir P, Favreau M, Katayarna C, Dooner H, Bedbrook J (1986)
mRN A transcripts of several plant genes are polyadenylated at multiple sites in vivo.
Nucleic Acids Res 14:2229-2240

Dedaldechamp F, Uhel C, Macheix JJ (1995) Enhancement of anthocyanin synthesis and
dihydroflavonol reductase (DFR) activity in response to phosphate deprivation in
grape cell suspensions. Phytochemistry 40: 1 357-1 360

Del Cardayré SB, Ribo M, Yokel EM, Quirk DJ, Rutter WJ, Raines RT (1995) Engineering
ribonuclease A: Production, purification and characterization of wild-type enzyme
and mutants at Glnl 1. Protein Eng 8:261-273

166

DeLong A, Calderon-Unea A, Dellaporta SL (1993) Sex determination gene
T ASSELSEEDZ of maize encodes a short-chain alcohol dehydrogenase required for
stage-specific floral organ abortion. Cell 74:757-768

Deutscher MP (1993a) Ribonuclease multiplicity, diversity, and complexity. J Biol Chem
268:13011-13014

Deutscher MP (1993b) RNA maturation nucleases. In: Linn SM, Lloyd RS, Roberts RJ
(eds) Nucleases: Second Edition. Cold Spring Harbor Laboratory Press, Cold
Spring Harbor, NY, pp 377-406

Devereux J, Haeberli P, Snrithies O (1984) A comprehensive set of sequence analysis
programs for the VAX. Nucleic Acids Res 12:387-395

Dodds PN, Clarke AE, Newbigin E (1996a) Molecular characterisation of an S-like RNase
of Nicotiana alata that is induced by phosphate starvation. Plant Mol Biol 31:227-
238

Dodds PN, Clarke AB, Newbigin E (1996b) A molecular perspective on pollination in
flowering plants. Cell 85:141-144

Dombrowski JE, Schroeder MR, Bednarek SY, Raikhel NV (1993) Determination of the
functional elements within the vacuolar targeting signal of barley lectin. Plant Cell
5:587-596

Dong X, Mindrinos M, Davis KR, Ausubel FM (1991) Induction of Arabidopsis defense
genes by virulent and avirulent Pseudomonas syringae strains and by a cloned
avirulence gene. Plant Cell 3:61-72

Dougherty WG, Parks TD (1995) Transgenes and gene suppression: telling us something
new? Curr Opin Cell Biol 7:399-405

Duff SMG, Plaxton WC, Lefebvre DD (1991) Phosphate-starvation response in plant cells:
De novo synthesis and degradation of acid phosphatases. Proc Natl Acad Sci USA
88:9538-9542

Duff SMG, Sarath G, Plaxton WC (1994) The role of acid phosphatases in plant
phosphorus metabolism. Physiol Plant 90:791-800

Farkas GL (1982) Ribonucleases and ribonucleic acid breakdown. In: Parthier B, Boulter D
(eds) Encyclopedia of Plant Physiology. Springer Verlag, Berlin, pp 224-262

167

Favre D, Ngai PK, Tirnnris KN (1993) Relatedness of a periplasnric, broad-specificity
RNase from Aeromonas hydrophila to RNase I of Escherichia coli and to a family
of eukaryotic RNases. J Bacteriol 175:3710-3722

Feinbaum RL, Ausubel FM (1988) Transcriptional regulation of the Arabidopsis thaliana
chalcone synthase gene. Mol Cell Biol 8:1985-1992

Fraser MJ, Low RL (1993) Fungal and mitochondrial nucleases. In: Linn SM, Lloyd RS,
Roberts RJ (eds) Nucleases, Second Edition. Cold Spring Harbor Laboratory Press,
Cold Spring Harbor, NY, pp 171-207

Fried M, Brosehart H (1967) The Soil-Plant System in Relation to Organic Mineral
Nutrition, Academic Press, New York

Fukuda H (1996) Xylogenesis: Initiation, progression, and cell death. Annu Rev Plant
Physiol Plant Mol Biol 47:299-325

Gegenheimer P (1996) Structure, mechanism and evolution of chloroplast transfer RNA
processing systems. Mol Biol Rep 22:147-150

Geoffroy P, Legrand M, Fritig B (1990) Isolation and characterization of a proteinaceous
inhibitor of microbial proteinases induced during the hypersensitive reaction of
tobacco to tobacco mosaic virus. Mol Plant Microbe Interact 3:327—333

Gil P, Green PJ (1996) Multiple regions of the Arabidopsis SAUR-ACI gene control
transcript abundance: The 3' untranslated region functions as an mRNA instability
determinant. EMBO J 15:1678-1686

Gite SU, Shankar V (1995) Single-strand-specific nucleases. Crit Rev Microbiol 21:101-
122

Glund K, Goldstein AH (1993) Regulation, synthesis, and excretion of a phosphate
starvation inducible RNase by plant cells. In: Verrna DPS (ed) Control of Plant
Gene Expression. CRC Press, Boca Raton, FL, pp 311-323

Goldstein AH, Baertlein DA, McDaniel RG (1988) Phosphate sstarvation inducible
metabolism in L. esculentum. I. Excretion of acid phosphatase by tomato plants and
suspension cultured cells. Plant Physiol 87:711-715

Goldstein AH, Baertlein DA, Danon A (1989) Phosphate starvation stress as an
experimental system for molecular analysis. Plant Mol Biol Rep 7:7-16

Gottlieb M, Mackow MC, Neubert TA (1988) Crithidia Iuciliae: factors affecting the
expression of 3'-nucleotidase/nuclease activity. Exp Parasitol 66: 108-1 17

168

Gray JE, McClure BA, Bonig I, Anderson MA, Clarke AB (1991) Action of the style
product of the self-incompatibility gene of Nicotiana alata (S-RNase) on in vitro-
grown pollen tubes. Plant Cell 3:271-283

Green PJ (1994) The ribonucleases of higher plants. Annu Rev Plant Physiol Plant Mol
Biol 45:421-445

Greenberg JT, Guo A, Klessig DF, Ausubel FM (1994) Programmed cell death in plants: A
pathogen-triggered response activated coordinately with multiple defense functions.
Cell 77:551-563

Greenberg JT (1996) Programmed cell death: A way of life for plants. Proc Natl Acad Sci
USA 93: 12094-12097

Haring V, Gray JE, McClure BA, Anderson MA, Clarke AE (1990) Self-incompatibility: a
self-recognition system in plants. Science 250:937-941

Harlow E, Lane D (1988) Antibodies: A Laboratory Manual, Cold Spring Harbor
Laboratory Press, Cold Spring Harbor, New York

Higgs DC, Colbert JT (1994) Oat phytochrome A mRNA degradation appears to occur via
two distinct pathways. Plant Cell 6: 1007-1019

Hime G, Prior L, Saint R (1995) The Drosophila melanogaster genome contains a member
of the Rh/TZIS-glycoprotein family of ribonuclease-encoding genes. Gene 158:203-
207

Hofte H, Desprez T, Amselem J, Chiapello H, Caboche M, Moisan A, Jourjon M-F,
Charpenteau J-L, Berthomieu P, Guerrier D, Giraudat J, Quigley F, Thomas F, Yu
D-Y, Mache R, Raynal M, Cooke R, Grellet F, Delseny M, Parrnentier Y, de
Marcillac G, Gigot C, Fleck J, Philipps G, Axelos M, Bardet C, Tremousaygue D,
Lescure B (1993) An inventory of 1152 expressed sequence tags obtained by partial
sequencing of cDNAs from Arabidopsis thaliana. Plant J 4: 1051-1061

Huang S, Lee H-S, Karunanandaa B, Kao T (1994) Ribonuclease activity of Petunia inflata
S proteins is essential for rejection of self-pollen. Plant Cell 6:1021-1028

Ide H, Kimura M, Arai M, Funatsu G (1991) The complete amino acid sequence of
ribonuclease from the seeds of bitter gourd (Momordica charantia). FEBS Lett
284:161-164

169

Inokuchi N, Koyarna T, Sawada F, Irie M (1993) Purification, some properties, and primary
structure of base non-specific ribonucleases from Physarum polycephalum. J
Biochem (Tokyo) 1 13 :425-432

Ioerger TR, Gohlke JR, Xu B, Kao T-H (1991) Primary structural features of the self-
incompatibility protein in solanaceae. Sex Plant Reprod 4:81-87

Irie M (1993) Structure-fimction relationship and distribution of RNase T2 family enzymes.
Ribonucleases: Chemistry, Biology, Biotechnology, 3rd Int Meet, Capri, Italy.
Abstract L22

Irie M (1997) RNase Tl/RNase T2 family RNases. In: Riordan JF, D'Alessio G (eds)
Ribonucleases: Structure and Function. Academic Press, Orlando, FL

Ito H, Fukuda Y, Murata K, Kumura A (1983) Transformation of intact yeast cells treated
with alkali cations. J Bacteriol 153:163-168

Iwarrratsu A, Aoyarna H, Dibo G, Tsunasawa S, Sakiyama F (1991) Amino acid sequence
of nuclease S] from Aspergillus oryzae. J Biochem (Tokyo) 110:151-158

Jacobsen JV, Gubler F, Chandler PM (1995) Gibberellin action in germinated cereal grains.
In: Davies PJ (ed) Plant Hormones: Physiology, Biochemistry and Molecular
Biology. Kluwer Academic Publishers, Dordrecht/Boston/London, pp 246-71

Jefferson RA (1987) Assaying chimeric genes in plants: the GUS gene fusion system. Plant
Mol Biol Rep 5:387-405

Jiang C-Z, Kliebenstein D, Ke N, Rodermel S (1994) Destabilization of rch sense
transcripts by antisense RNA. Plant Mol Biol 25:569-576

Johnson JF, Allan DL, Vance CP, Weiblen G (1996a) Root carbon dioxide fixation by
phosphorus-deficient Lupinus albus - Contribution to organic acid exudation by
proteoid roots. Plant Physiol 112:19-30

Johnson JF, Vance CP, Allan DL (1996b) Phosphorus deficiency in Lupinus albus -
Altered lateral root development and enhanced expression of phosphoenolpyruvate
carboxylase. Plant Physiol 1 12:31-41

Jost W, Bak H, Glund K, Terpstra P, Beintema JJ (1991) Amino acid sequence of an
extracellular, phosphate-starvation-induced ribonuclease from cultured tomato
(Lycopersicon esculentum) cells. Eur J Biochem 198:1-6

Kao T, Huang S (1994) Gametophytic self-incompatibility: A mechanism for self/nonself
discrimination during sexual reproduction. Plant Physiol 105:461-466

170

Kao T-H, McCubbin AG (1996) How flowering plants discriminate between self and non-
self pollen to prevent inbreeding. Proc Natl Acad Sci USA 93: 12059-12065

Kauffrnann S, Legrand M, Geoffroy P, Fritig B (1987) Biological function of pathogenesis-
related proteins: four PR proteins of tobacco have 1,3-B-glucanase activity. EMBO
J 6:3209-3212

Kaufrnann H, Salarnini F, Thompson RD (1991) Sequence variability and gene structure at
the self-incompatibility locus of Solanum tuberosum. Mol Gen Genet 226:457-466

Kawata Y, Sakiyama F, Tamaoki H (1988) Anrino-acid sequence of ribonuclease T2 from
Aspergillus omae. Eur J Biochem 176:683-697

Kawata Y, Sakiyama F, Hayashi F, Kyogoku Y (1990) Identification of two essential
histidine residues of ribonuclease T2 from Aspergillus oryzae. Eur J Biochem
187:255-262

Keen NT (1992) The molecular biology of disease resistance. Plant Mol Biol 19: 109-122

Kefalas PS, Yupsanis T (1995) Properties and specificity of a calcium dependent
endonuclease from genninated lentil (Lens culinaris). J Plant Physiol 146:1-9

Kelly MO, Davies PJ (1988) The control of whole plant senescence. Crit Rev Plant Sci
7:139—173

Kenefick DG, Blake TK (1986) Low ribonuclease I activity prior to cold acclimation in
freeze selected winter barley. Crop Sci 26: 1099-1 103

Kheyr-Pour A, Bintrim SB, Ioerger TR, Remy R, Hammond SA, Kao T-H (1990) Sequence
diversity of pistil S-proteins associated with garnetophyticself-incompatibility in
Nicotiana alata. Sex Plant Reprod 3:88-97

King EO, Ward MK, Raney DE (1954) Two simple media for the demonstration of
phycocyanin and fluorescin. J Lab Clin Med 44:301-307

Kiss T, Marshallsay C, Filipowicz W (1992) 7-2/MRP RNAs in plant and mammalian cells:
Association with higher order structures in the nucleolus. EMBO J 11:3737—3746

Kiss T, Filipowicz W (1992) Evidence against a mitochondrial location of the 7-2/MRP
RNA in mammalian cells. Cell 70:11-16

Knox RB (1993) Grass pollen, thunderstorms and asthma. Clin Exp Allergy 232354-359

171

Koncz C, Schell J (1986) The promoter of TL-DNA gene 5 controls the tissue-specific
expression of chirnaeric genes carried by a novel type of Agrobacterium binary
vector. Mol Gen Genet 204:383-396

Kock M, Lofiler A, Abel S, Glund K (1995) cDNA structure and regulatory properties of a
family of starvation-induced ribonucleases fi'om tomato. Plant Mol Biol 27:477-485

Krysan PJ, Young JC, Tax F, Sussman MR (1996) Identification of transferred DNA
insertions within Arabidopsis genes involved in signal transduction and ion
transport. Proc Natl Acad Sci USA 93:8145-8150

Kubasek WL, Shirley BA, McKillop A, Goodman HM, Briggs W, Ausubel FM (1992)
Regulation of flavonoid biosynethetic genes in germinating Arabidopsis seedlings.
Plant Cell 4: 1229-1236

Kuligowska E, Klarkowska D, Szarkowski JW (1988) Purification and properties of
endonuclease from wheat chloroplasts, specific for single-stranded DNA.
Phytochemistry 27: 1275-1279

Kumar D, Mukherjee S, Reddy MK, Tewari KK (1995) A novel single-stranded DNA-
specific endonuclease from pea chloroplasts. J Exp Bot 46:767-776

Kung SD (1976) Tobacco fraction 1 protein: a unique genetic marker. Science 191:429-434

Kunkel TA, Roberts JD, Zakour RA (1987) Rapid and efficient site-specific mutagenesis
without phenotypic selection. Methods Enzymol 154:367-3 82

Kurihara H, Mitsui Y, Ohgi K, Irie M, Mizuno H, Nakamura KT (1992) Crystal and
molecular structure of RNase Rh, a new class of microbial ribonuclease from
Rhizopus niveus. FEBS Lett 306:189-192

Kurihara H, Nonaka T, Mitsui Y, Ohgi K, Irie M, Nakamura KT (1996) The crystal
structure of ribonuclease Rh from Rhizopus niveus at 2.0A resolution. J Mol Biol
255:310-320

Kuroda S, Norioka S, Mitta M, Kato I, Sakiyama F (1994) Primary structure of a novel
stylar RN ase unassociated with self-incompatibility in tobacco plant, Nicotiana
alata. J Protein Chem 13:438-439

Laemmli UK (1970) Cleavage of structural proteins during the assembly of the head of
bacteriophage T4. Nature 227:680-685

172

Lamport DTA, Catt JW (1981) Glycoproteins and enzymes of the cell wall. In: Tanner W,
Holmes FA (eds) Encyclopedia of Plant Physiology (NS), Plant Carbohydrates 11.
Springer-Verlag, New York, pp 133-165

Laskowski MS (1980) Purification and properties of the mung bean nuclease. Methods
Enzymol 65:263-276

Leah R, Tommerup H, Svendsen I, Mundy J (1991) Biochemical and molecular
characterization of three barley seed proteins with antifungal properties. J Biol
Chem 266:1564-1573

Lee C, Levin A, Branton D (1987) Copper staining: a five-miute protein stain for sodium
dodecyl sulfate-polyacrylarnide gels. Anal Biochem 166:308-312

Lee FS, Vallee BL (1993) Structure and action of mammalian ribonuclease (angiogenin)
inhibitor. Prog Nucleic Acid Res Mol Biol 44:1-30

Lee H-S, Singh A, Kao T (1992) RNase X2, a pistil-specific ribonuclease from Petunia
inflata, shares sequence similarity with solanaceous S proteins. Plant Mol Biol
20:1131-1141

Lee H-S, Huang S, Kao T-H (1994) S proteins control rejection of incompatible pollen in
Petunia inflata. Nature 367:560-566

Legrand M, Kauffmann S, Geoffroy P, F ritig B (1987) Biological function of pathogenesis-
related proteins: four tobacco pathogenesis-related proteins are chitinases. Proc Natl
Acad Sci USA 84:6750-6754

Linsmaier EM, Skoog F (1965) Organic growth factor requirements of tobacco tissue
cultures. Physiol Plant 18:100-127

Loffler A, Abel S, Jost W, Beintema JJ, Glund K (1992) Phosphate-regulated induction of
intracellular ribonucleases in cultured tomato (Lycopersicon esculentum) cells. Plant
Physiol 98: 1472-1478

Lofller A, Glund K, Irie M (1993) Amino acid sequence of an intracellular, phosphate-
starvation-induced ribonuclease from cultured tomato (Lycopersicon esculentum)
cells. Eur J Biochem 214:627-633

Lusso M, Kuc J (1995) Increased activities of ribonuclease and protease after challenge in
tobacco plants with induced systemic resistance. Physiol Mol Plant Pathol 47:419-
428

173

Lygerou Z, Allrnang C, Tollervey D, Séraphin B (1996) Accurate processing of a
eukaryotic precursor ribosomal RNA by ribonuclease MRP in vitro. Science
272:268-270

Maekawa K, Tsunasawa S, Dibo G, Sakiyama F (1991) Primary structure of nuclease Pl
from Penicillium citrinum. Eur J Biochem 200:651-661

Marchfelder A, Brennicke A (1994) Characterization and partial purification of tRN A
processing activities from potato mitochondria. Plant Physiol 105:1247-1254

Mariani C, De Beuckeleer M, Truettner J, Leemans J, Goldberg RB (1990) Induction of
male sterility in plants by a chimaeric ribonuclease gene. Nature 347:737-741

Mariani C, Gossele V, De Beuckeleer M, De Block M, Goldberg RB, De Greef W,
Leemans J (1992) A chimaeric ribonuclease-inhibitor gene restores fertility to male
sterile plants. Nature 357:384-387

Marschner H (1986) Function of mineral nutrients: macronutrients. In: Marscher H (ed)
Mineral Nutrition of Higher Plants. Academic Press, London,

Matile P, Winkenbach F (1971) Function of lysosomes and lysosomal enzymes in the
senescing corolla of the morning glory (Ipomoea purpurea). J Exp Bot 22:759-771

Matousek J, Tmena L, Oberhauser R, Lichtenstein CP, Nellen W (1994) dsRNA degrading
nucleases are differentially expressed in tobacco anthers. Biol Chem Hoppe Seyler
375:261-269

Matousek J, Tupy J (1984) Purification and properties of extracellular nuclease from
tobacco pollen. Biol Plant 26:62-73

Matousek J, Tupy J (1985) The release and some properties of nuclease from various pollen
species. J Plant Physiol 119:169-178

Matousek J, Tupy J (1987) Developmental changes in nuclease and other phosphohydrolase
activities in anthers of Nicotiana tabacum L. J Plant Physiol 129:351-362

Mauch F, Staehelin LA (1989) Functional implications of the subcellular localization of
ethylene-induced chitinase and B-l,3-glucanase in bean leaves. Plant Cell 1:447-457

McClure BA, Hagen G, Brown CS, Gee MA, Guilfoyle TJ (1989a) Transcription,
organization, and sequence of an auxin-regulated gene cluster in soybean. Plant Cell
1:229-239

174

McClure BA, Haring V, Ebert PR, Anderson MA, Simpson RJ, Sakiyama F, Clarke AE
(1989b) Style self-incompatibility gene products of Nicotiana alata are
ribonucleases. Nature 342:955-958

McClure BA, Gray JE, Anderson MA, Clarke AE (1990) Self-incompatibility in Nicotiana
alata involves degradation of pollen rRNA. Nature 347:757-760

Meador J, IH, Kennel] D (1990) Cloning and sequencing the gene encoding Escherichia
coli ribonuclease I: Exact physical mapping using the genome library. Gene 95: 1-7

Mittler R, Shulaev V, Lam E (1995) Coordinated activation of programmed cell death and
defense mechanisms in transgenic tobacco plants expressing a bacterial proton
pump. Plant Cell 7:29-42

Mittler R, Lam E (1995a) In situ detection of nDNA fiagmentation during the
differentiation of tracheary elements in higher plants. Plant Physiol 108:489-493

Mittler R, Lam E (1995b) Identification, characterization, and purification of a tobacco
endonuclease activity induced upon hypersensitive response cell death. Plant Cell
7: 195 1-1962

Moiseyev GP, Beintema JJ, Fedoreyeva LI, Yakovlev GI (1994) High sequence similarity
between a ribonuclease from ginseng calluses and fungus-elicited proteins from
parsley indicates that intracellular pathogenesis-related proteins are ribonucleases.
Planta 193 :470-472

Mol J, Jenkins G, Schafer E, Weiss D (1996) Signal perception, transduction, and gene
expression involved in anthocyanin biosynthesis. Crit Rev Plant Sci 15:525-557

Monko M, Kuligowska E, Szarkowski JW (1994) A single-strand-specific nuclease from a
fiaction of wheat chloroplast stromal protein. Phytochemistry 37:301-305

Morrissey JP, Tollervey D (1995) Birth of the snoRNPs: The evolution of RNase MRP and
the eukaryotic pre-rRNA-processing system. Trends Biochem Sci 20:78-82

Murfett J, Atherton TL, Mou B, Gasser CS, McClure BA (1994) S-RNase expressed in
transgenic Nicotiana causes S-allele-specific pollen rejection. Nature 367:563-566

Nagano M, Ashihara H (1993) Long-terrn phosphate starvation and respiratory metabolism
in suspension-cultured Catharanthus roseus cells. Plant Cell Physiol 34:1219-1228

Naseem I, Hadi SM (1987) Single-strand-specific nuclease of pea seeds: glycoprotein
nature and associated nucleotidase activity. Arch Biochem Biophys 255:437-445

175

Nellen W, Lichtenstein C (1993) What makes an mRNA anti-sense-itive? Trends Biochem
Sci 18:419-423

Neubert TA, Gottlieb M (1990) An inducible 3'-nucleotidase/nuclease from the

trypanosomatid Crithidia Iuciliae: purification and characterization. J Biol Chem
265:7236-7242

Neuhaus JM, Pietrzak M, Boller T (1994) Mutation analysis of the vacuolar targeting
peptide of tobacco chitinase indicates a low sequence specificity of the sorting
system. Plant J 5:45-54

Neuhaus JM (1996) Protein targeting to the plant vacuole. Plant Physiol Biochem 34:217-
221

Newbigin B, Anderson MA, Clarke AE (1993) Gametophytic self-incompatibility systems.
Plant Cell 521315-1324

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

Ninorrriya Y, Ueki K, Sato S (1977) Chromatographic separation of extracellular acid
phosphatase of tobacco cells cultured under Pi-supplied and omitted conditions.
Plant Cell Physiol 18:413-420

Noodén LD (1988a) Whole plant senescence. In: Noode'n LD, Leopold AC (eds)
Senescence and Aging in Plants. Academic Press, Inc. San Diego, CA, pp 391-439

Noodén LD (1988b) The phenomena of senescence and aging. In: Noodén LD, Leopold AC
(eds) Senescence and Aging in Plants. Academic Press, San Diego, pp 1-50

Norioka N, Ohnishi Y, Norioka S, Ishimizu T, Nakanishi T, Sakiyama F (1995) Nucleotide
sequences of cDNAs encoding S2- and S4-RNases (D49527 and D49528 for
EMBL) from Japanese pear (Pyrus pyrifolia Nakai 1) (PGR95-020). Plant Physiol
108:1343

Nfimberger T, Abel S, Jost W, Glund K (1990) Induction of an extracellular ribonuclease in
cultured tomato cells upon phosphate starvation. Plant Physiol 92:970-976

Ogawa T, Hori T, Ishida I (1996) Virus-induced cell death in plants expressing the
mammalian 2', 5' oligoadenylate system. Nature Biotech 14: 1 566-1 569

176

Ohgi K, Horiuchi H, Watanabe H, Takagi M, Yano K, Irie M (1991) Expression of RNase
Rh fiom Rhizopus niveus in yeast and characterization of the secreted proteins. J
Biochem (Tokyo) 109:776-785

Ohgi K, Horiuchi H, Watanabe H, Iwama M, Takagi M, Irie M (1992) Evidence that three
histidine residues of a base non-specific and adenylic acid preferential ribonuclease

from Rhizopus niveus are involved in the catalytic function. J Biochem (Tokyo)
112:132-138

Ohgi K, Horiuchi H, Watanabe H, Iwama M, Takagi M, Irie M (1993) Role of Asp51 and
Glu105 in the enzymatic activity of a ribonuclease from Rhizopus niveus. J Biochem
(Tokyo) 1 13:219-224

Ohme-Takagi M, Taylor CB, Newman TC, Green PJ (1993) The effect of sequences with
high AU content on mRNA stability in tobacco. Proc Natl Acad Sci USA 90:11811-
1 1815

Oommen A, Li X, Gegenheimer P (1992) Cleavage specificity of chloroplast and nuclear
tRNA 3'-processing nucleases. Mol Cell Biol 12:865-875

Owttrim GW, Hofrnann S, Kuhlemeier C (1991) Divergent genes for translation initiation
factor eIF-4A are coordinately expressed in tobacco. Nucleic Acids Res 19:5491-
5496

Paul MJ, Stitt M (1993) Effects of nitrogen and phosphorus deficiencies on levels of
carbohydrates, respiratory enzymes and metabolites in seedlings of tobacco and
their response to exogenous sucrose. Plant Cell Environ 16:1047-1057

Pennisi E (1996) Chemical shackles for genes? Science 273:574-575

Poirier Y, Thoma S, Somerville C, Schiefelbein J (1991) A mutant of Arabidopsis deficient
in xylem loading of phosphate. Plant Physiol 97:1087-1093 '

Prentice N, Heisel S (1985) Purification and characterization of a ribonuclease from barley
roots. Phytochemistry 24: 1451-1457

Prentice N, Heisel S (1986) Characterization of a nuclease from barley shoots.
Phytochemistry 25:2057-2062 .

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

177

Rabino I, Mancinelli AL (1986) Light, temperature, and anthocyanin production. Plant
Physiol 81 :922-924

Rivers BA, Bematzky R, Robinson SJ, Jahnen-Dechent W (1993) Molecular diversity at the
self-incompatibility locus is a salient feature in natural populations of wild tomato
(Lycopersicon peruvianum). Mol Gen Genet 238:419-427

Robertson HD (1982) Escherichia coli ribonuclease III cleavage sites. Cell 30:669-672

Rojo MA, Arias FJ, Iglesias R, Ferreras JM, Munoz R, Escarmis C, Soriano F, Lopez-
Fando J, Méndez E, Girbés T (1994a) Cusativin, a new cytidine-specific
ribonuclease accumulated in seeds of Cucumis sativus L. Planta 194:328-338

Rojo MA, Arias FJ, Iglesias R, Ferreras JM, Soriano F, Méndez E, Escarrrris C, Girbés T
(1994b) Enzynric activity of melonin, a translational inhibitor present in dry seeds
of Cucumis melo L. Plant Sci 103:127-134

Rosenberg S, Barr PJ, Najarian RC, Hallewell RA (1984) Synthesis in yeast of a functional
oxidation-resistant mutant of human al-antitrypsin. Nature 312:77-80

Ross J (1995) mRNA stability in mammalian cells. Microbiol Rev 59:423-450

Rothstein R (1991) Targeting, disruption, replacement, and allele rescue: integrative DNA
transformation in yeast. Methods Enzymol 194:281-301

Royo J, Kunz C, Kowyarna Y, Anderson M, Clarke AE, Newbigin E (1994) Loss of a
histidine residue at the active site of S-locus ribonuclease is associated with self-
compatibility in Lycopersicon peruvianum. Proc Natl Acad Sci USA 91 :651 1-6514

Ryan CA (1990) Protease inhibitors in plants: genes for improving defenses against insects
and pathogens. Annu Rev Phytopathol 28:425-449

Saba-El-Leil MK, Rivard S, Morse D, Cappadocia M (1994) The S1] and S13 self
incompatibility alleles in Solanum chacoense Bitt. are remarkably similar. Plant
Mol Biol 24:571-583

Salisbury FB, Ross CW (1992) Plant Physiology, 4th edn. Wadsworth Publishing
Company, Belmont, CA

Sambrook J, Fritsch EF, Maniatis T (1989) Molecular Cloning: A Laboratory Manual,
Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York

Sanchez PA, Uehara G (1980) Management considerations for acid soils with high
phosphorus fixation capacity. In: Khasawneh FE, Sample EC, Karnprath EJ (eds)

178

The Role of Phosphorus in Agriculture. American Society of Agronomy, Madison,
WI, pp 471-514

Sanger F, Nicklen S, Coulson AR (1977) DNA sequencing with chain-terminating
inhibitors. Proc Natl Acad Sci USA 74:5463-5467

Saraste M, Sibbald PR, Wittinghofer A (1990) The P-loop--a common motif in ATP- and
GTP-binding proteins. Trends Biochem Sci 14:430-434

Sassa H, Nishio T, Kowyama Y, Hirano H, Koba T, Ikehashi H (1996) Self incompatibility
(S) alleles of the Rosaceae encode members of a distinct class of the T2/S
ribonuclease superfamily. Mol Gen Genet 250:547—557

Saxena SK, Gravel] M, Wu YN, Mikulski SM, Shogen K, Ardelt W, Youle RJ (1996)
Inhibition of HIV-1 production and selective degradation of viral RNA by an
amphibian ribonuclease. J Biol Chem 271:20783-20788

Schneider R, Unger G, Stark R, Schneider-Schema E, Thiel H-J (1993) Identification of a
structural glycoprotein of an RNA virus as a ribonuclease. Science 261:1169-1171

Schroeder MR, Borkhsenious ON, Matsuoka K, Nakamura K, Raikhel NV (1993)
Colocalization of barley lectin and sporarnin in vacuoles of transgenic tobacco
plants. Plant Physiol 101:451-458

Shaw G, Kamen R (1986) A conserved AU sequence fiom the 3' untranslated region of
GM-CSF mRNA mediates selective mRNA degradation. Cell 46:659-667

Short JM, Fernandez JM, Sorge JA, Huse WD (1988) Lambda ZAP: a bacteriophage
lambda expression vector with in vivo excision properties. Nucleic Acids Res
16:7583-7600

Singh MB, Hough T, Theerakulpisut P, Avjioglu A, Davies S, Smith PM, Taylor P,
Simpson RJ, Ward LD, McCluskey J, Puy R, Knox RB (1991) Isolation of cDNA
encoding a newly identified major allergenic protein of rye-grass pollen:
intracellular targeting to the arnyloplast. Proc Natl Acad Sci USA 88:1384-1388

Siwecka MA, Ryte] M, Szarkowski JW (1989) Purification and characterization of nuclease
I associated with rye germ ribosomes. Acta Biochim Pol 36:45-62

Smyth DR, Bowman JL, Meyerowitz EM (1990) Early flower development in Arabidopsis.
Plant Cell 2:755-767

179

Somssich IE, Schmelzer E, Kawalleck P, Hahlbrock K (1988) Gene structure and in situ

transcript localization of pathogenesis-related protein 1 in parsley. Mol Gen Genet
213:93-98

Stevens A (1993) Eukaryotic nucleases and mRNA turnover. In: Belasco JG, Brawerrnan G
(eds) Control of Messenger RNA Stability. Academic Press, Inc. San Diego, CA, pp
449-471

Stoddart JL, Thomas H (1982) Leaf Senescence. In: Boulter D, Parthier B (eds)
Encyclopedia of plant physiology. Springer, Berlin, Heidelberg, New York, pp 592-
636

Strickland JA, Marzilli LG, Puckett JMJ, Doetsch PW (1991) Purification and properties of
nuclease SP. Biochemistry 30:9749-9756

Suzuki H, Fowler TJ, Tierney ML (1993) Deletion analysis and localization of SbPRPl, a
soybean cell wall protein gene, in roots of transgenic tobacco and cowpea. Plant
MolBi0121:109-119

Swerdlow PS, Finley D, Varshavsky A (1986) Enhancement of irnmunoblot sensitivity by
heating of hydrated filters. Anal Biochem 156:147-153

Swoboda I, Scheiner O, Kraft D, Breitenbach M, Heberle-Bors E, Vicente O (1994) A birch
gene family encoding pollen allergens and pathogenesis-related proteins. Biochim
Biophys Acta 1219:457-464

Swoboda I, Jilek A, Ferreira F, Engel E, Hoffinann-Sommergruber K, Scheiner O, Kraft D,
Breiteneder H, Pittenauer E, Schmid E, Vicente O, Heberle-Bors E, Ahom H,
Breitenbach M (1995) Isoforrns of Bet v 1, the major birch pollen allergen, analyzed
by liquid chromatography, mass spectrometry, and cDNA cloning. J Biol Chem
270:2607-2613

Swoboda I, Hoffrnann-Sommergruber K, O'Riordain G, Scheiner O, Heberle-Bors E,
Vicente O (1996) Bet v 1 proteins, the major birch pollen allergens and members of
a family of conserved pathogenesis-related proteins, show ribonuclease activity in
vitro. Physiol Plant 96:433-438

Tague BW, Chrispeels MJ (1987) The plant vacuolar protein, phytohemagglutinin, is
transported to the vacuole of transgenic yeast. J Cell Biol 105:1971-1979

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

 

180

Tanzer MM, Meagher RB (1995) Degradation of the soybean ribulose-l,5-bisphosphate
carboxylase small-subunit mRNA, SRS4, initiates with endonucleolytic cleavage.
Mol Cell Biol 15:6641-6652

Taylor CB, Bariola PA, DelCardayré SB, Raines RT, Green PJ (1993) RNS2: A
senescence-associated RNase of Arabidopsis that diverged from the S-RNases
before speciation. Proc Natl Acad Sci USA 90:5118-5122

Taylor CB, Green P] (1991) Genes with homology to fungal and S-gene RNases are
expressed in Arabidopsis thaliana. Plant Physiol 96:980-984

Tewes A, Glund K, Walther R, Reinbothe H (1984) High yield isolation and rapid recovery
of protoplasts from suspension cultures of tomato (Lycopersicon esculentum). Z
Pflanzenphysiol 1 13: 141-150

Thelen MP, Northcote DH (1989) Identification and purification of a nuclease from Zinnia
elegans L.: A potential molecular marker for xylogenesis. Planta 179:181-195

Theodorou ME, Plaxton WC (1993) Metabolic adaptations of plant respiration to nutritional
phosphate deprivation. Plant Physiol 101:339-344

Thill GP, Kramer RA, Turner KJ, Bostian KA (1983) Comparative analysis of the 5'-end
regions of two repressible acid phosphatase genes in Saccharomyces cerevisiae.
Mol Cell Biol 31570-579

Tsai D-S, Lee H-S, Post LC, Kreiling KM, Kao T-H (1992) Sequence of an S-protein of
Lycopersicon peruvianum and comparison with other solanaceous S-proteins. Sex
Plant Reprod 5:263

Tsuji J, Somerville SC, Hammerschmidt R (1991) Identification of a gene in Arabidopsis
thaliana that controls resistance to Xanthomonas campestris pv campestris. Physiol
Mol Plant Pathol 38:57-65

Uchida H, Wu Y-D, Takadera M, Miyashita S, Nomura A (1993) Purification and some
properties of plant endonuclease from scallion bulbs. Biosci Biotechnol Biochem
57:2139-2143

Ueki K, Sato S (197]) Effect of inorganic phosphate on the extracellular acid phosphatase
activity of tobacco cells cultured in vitro. Physiol Plant 24:506-511

Usuda H, Shimogawara K (1992) Phosphate deficiency in maize. III. Changes in enzyme
activities during the course of phosphate deprivation. Plant Physiol 99: 1680-1685

181

Vakalopoulou E, Schaack J, Shenk T (1991) A 32-kilodalton protein binds to AU-rich
domains in the 3' untranslated regions of rapidly degraded mRNAs. Mol Cell Biol
11:3355-3364

van de Locht U, Meier I, Hahlbrock K, Somssich IE (1990) A 125 bp promoter fragment is
sufficient for strong elicitor-mediated gene activation in parsley. EMBO J 92945-
2950

van Loon LC, Pierpoint WS, Boller T, Conejero V (1994) Recommendations for naming
plant pathogenesis-related proteins. Plant Mol Biol Rep 12:245-264

Vaux DL (1993) Toward an understanding of the molecular mechanisms of physiological
cell death. Proc Natl Acad Sci USA 90:786-789

Vincent JR, Fulbright DW (1983) Transfer of pRD] to Pseudomonas syringae and
evidence for its integration into the chromosome. J Bacteriol 156: 1349-1351

Vogel K, Hinnen A (1990) The yeast phosphatase system. Mol Microbiol 4:2013-2017

von Heijne G (1986) A new method for predicting signal sequence cleavage sites. Nucleic
Acids Res 14:4683-4690

Walter MH, Liu JW, Wfinn J, Hess D (1996) Bean ribonuclease-like pathogenesis-related
protein genes (Ypr10) display complex patterns of developmental, dark-induced and
exogenous-stimulus—dependent expression. Eur J Biochem 239:281-293

Wang MJ, Davis NW, Gegenheimer P (1988) Novel mechanisms for maturation of
chloroplast transfer RNA precursors. EMBO J 7:1567-1574

Watanabe H, Narumi H, Inaba T, Ohgi K, Irie M (1993) Purification, some properties, and
primary structure of a base non-specific ribonuclease from oyster (Crussdstrea
grigus). J Biochem (Tokyo) 114:800-807

Whalen MC, Innes RW, Bent AF, Staskawicz BJ (1991) Identification of Pseudomonas
syringae pathogens of Arabidopsis and a bacterial locus determining avirulence on

both Arabidopsis and soybean. Plant Cell 3:49-59

Wilson CM (1982) Plant nucleases: biochemistry and development of multiple molecular
forms. Isozymes: Curr Top Biol Med Res 6:33-54

Wink M (1994) The plant vacuole: a multifunctional compartment. J Exp Bot 44:231-246

Woodson WR (1987) Changes in protein and mRNA populations during the senescence of
carnation petals. Physiol Plant 71:495-502

182

Wu Y, Mikulski SM, Ardelt W, Rybak SM, Youle RJ (1993) A cytotoxic ribonuclease.
Study of the mechanism of onconase cytotoxicity. J Biol Chem 268:]0686-10693

Xu B, Mu J, Nevins DL, Grun P, Kao T-H (1990) Cloning and sequencing of cDNAs
encoding two self-incompatibility associated proteins in Solanum chacoense. Mol
Gen Genet 2242341-346

Xue YB, Carpenter R, Dickinson HG, Coen ES (1996) Origin of allelic diversity in
antirrhinum S locus RNases. Plant Cell 8:805-814

Yang JJ, Schuster G, Stem DB (1996) CSP41, a sequence-specific chloroplast mRNA
binding protein, is an endoribonuclease. Plant Cell 8: 1409-1420

Ye Z-H, Varner JE (1993) Gene expression patterns associated with in vitro tracheary
element formation in isolated single mesophyll cells of Zinnia elegans. Plant
Physiol 103:805-813

Ye ZH, Droste DL (1996) Isolation and characterization of cDNAs encoding xylogenesis-
associated and wounding-induced ribonucleases in Zinnia elegans. Plant Mol Biol
30:697-709

Yen Y, Baenziger PS (1993) Identification, characterization, and comparison of RNA-
degrading enzymes of wheat and barley. Biochem Genet 31:133-146

Yen Y, Green PJ (1991) Identification and properties of the major ribonucleases of
Arabidopsis thaliana. Plant Physiol 97:1487-1493

Zhivotovsky B, Wade D, Nicotera P, Orrenius S (1994) Role of nucleases in apoptosis. Int
Arch Allergy Immunol 105:333-338

Zhou A, Hassel BA, Silverrnan RH (1993) Expression cloning of a 2-5A-dependent
RNAase: a uniquely regulated mediator of interferon action. Cell 72:7 53-765

HICH RN TA

16 S

31293

TE UNIV. LIBRARIES
ll]Ill"lllllllllllllllllllll
010506420