ABSTRACT
Title of Dissertation:FUNCTIONAL ANALYSES OF ARABIDOPSIS
RIBONUCLEOTIDE REDUCTASE SMALL SUBUNIT
GENE FAMILY
Chunxin Wang, Doctor of Philosophy, 2004
Dissertation Directed By:Associate Professor Zhongchi Liu,
Department of Cell Biology and Molecular Genetics
A fundamental question in plant development is how cell division events are
coordinately regulated in the context of growth and development. To address this
question, I chose to study a pleiotropic mutant, tso2, which exhibited developmental
defects including callus-like floral organs and fasciated shoot meristem. I isolated the
TSO2 gene and showed it encodes the small subunit of ribonucleotide reductase (RNR).
RNR catalyzes a rate-limiting step in the production of deoxyribonucleotides needed for
DNA synthesis. Subsequently, I showed that tso2 mutants reduce the dNTP levels.
To understand why tso2 mutants, defective in this essential process, are still
viable, I identified two homologs of TSO2, R2A and R2B in Arabidopsis. Mutations in
R2A and R2B were isolated using a reverse genetic approach. While r2a, r2b single
mutants or r2a r2b double mutants fail to display any visible phenotype, r2a and r2b
mutations can enhance tso2 genetically, resulting in seedling lethality and embryonic
lethality in tso2 r2a and tso2 r2b, respectively. Overexpression of either R2A or R2B can
rescue the tso2 mutants, suggesting that the three R2 genes are functionally redundant.
In addition to the developmental defects, tso2 mutants were more sensitive to HU
(hydroxyurea) and UV-C, indicating that TSO2 plays a major role in DNA repair. In tso2
r2a double mutant seedlings, increased DNA damage accumulates, leading to massive
programmed cell death. In addition, release of transcriptional gene silencing was
observed in tso2 r2a double mutants, suggesting that DNA damage can lead to epigenetic
instability. To further identify regulators of RNR and novel components of plant DNA
damage response pathways, 18 independent tso2 suppressors were isolated in a genetic
screen. These suppressors fall into at least four different complementation groups.
My genetic and molecular characterization of TSO2 is the first functional study of
RNR in plants. My results indicated that plants could initiate programmed cell death in
response to DNA damage. The developmental defects in tso2 mutants are caused by
epigenetic instability and aberrant cell division. The isolation of potential tso2
suppressors will be crucial to the understanding of plant DNA damage response pathway,
an understudied area in plant biology.
FUNCTIONAL ANALYSES OF ARABIDOPSIS RIBONUCLEOTIDE
REDUCTASE SMALL SUBUNIT GENE FAMILY
By
Chunxin Wang
Dissertation submitted to the Faculty of the Graduate School of the
University of Maryland, College Park, in partial fulfillment
of the requirements for the degree of
Doctor of Philosophy
2004
Advisory Committee:
Professor Zhongchi Liu, Chair
Professor Eric H. Baehrecke
Professor Caren Chang
Professor Gerald F. Deitzer
Professor Stephan M. Mount
? Copyright by
Chunxin Wang
2004
ii
Dedication
To my parents
iii
Acknowledgements
To my advisor Dr. Zhongchi Liu for her guidance, encouragement, patience and
motivation, for allowing me to explore a new field, and for giving me an
opportunity to pursue my dream and become a Terp.
To my graduate committee members Drs. Eric Baehrecke, Caren Chang, Gerald Deitzer,
and Steve Mount for their continuous guidance, support and inspiration during the
period of my study.
To Drs. Caren Chang and Steve Mount for having me as their TA. What a great
experience!
To Dr. Eric Baehrecke for his support that helped me go through some difficult time in
my life.
To Dr. Joshua Levin for providing tso2 alleles, Dr. Peter Doerner for seeds of
pcycB1::GUS,  Dr. Roy  Beatrice for advice and protocol in dNTP measurement,
Dr. Rebecca Stevens for cell suspension line, Dr. Tim Mougel for microscopy
assistance, Dr. David Straney for providing sequencing equipment, Arabidopsis
TILLING Project for identifying r2a-1allele, the Arabidopsis Biological Resource
Center (BBRC) and Kazusa DNA Research Institute for DNA clones and seed
stocks.
To Drs. Robert Franks, Yingjie Wu and Xiaodong Yao for their wonderful friendship,
encouragement and advice. They teach me how to be a good person.
To Terrapins for giving me numerous great experience and excitement.
iv
List of Tables
Table 1-1. Core cell cycle genes in Arabidopsis????????????????21
Table 1-2. Overview of the characteristics of the ribonucleotide
     reductase classes??????????????????...?????35
Table 1-3. Overview of RNR genes in eukaryotes???????????????41
Table 2-1. Polymorphisms in the R2B gene between Col and Ler?????????63
Table 2-2. E2F binding sites in the promoters of various RNR genes???????..65
Table 2-3. Genotyping among the progeny of tso2-1/tso2-1 r2b-1/+ parent?????67
Table 3-1. Complementation test of 9 suppressor lines????????????..100
Table 3-2. Complementation test of 7 suppressor lines????????????..100
Table 3-3. Potential complementation groups of the 18 suppressors???????..103
Table A1. CAPS and dCAPS markers used in this study????????????123
Table A2. Gene-specific primers used for RT-PCR?????????????....124
v
List of Figures
Figure 1-1. Schematic diagram of organization of SAM?????????.????4
Figure 1-2. Mutants affecting shoot apical meristem development?????????.7
Figure 1-3. A model for CLV3 action as an intercellular signal in the SAM?????..9
Figure 1-4. Schematic diagram of cell cycle progression in plants????????...16
Figure 1-5. The model for activation of E2F-pRb pathway???????????...18
Figure 1-6. Structures of E2F gene family?????????????????.?23
Figure 1-7. Components of DNA damage checkpoints in yeast and human????.?28
Figure 1-8.  Diagram of de novo and salvage pathways for biosynthesis of dNTPs?.?32
Figure 1-9. Schematic diagram of RNR structure?????????????.??37
Figure 1-10. Schematic diagram of regulation of RNR activity????..????.?.38
Figure 1-11. Three-dimensional structure of RNR???????????????39
Figure 1-12. Translocation of human RNR subunits upon DNA damage??????.46
Figure 1-12. A model illustrating CRT1 mediated repression of RNR genes????...49
Figure 2-1.  Phenotype of tso2 mutants???????????????????.58
Figure 2-2. Scanning electronic microscopic analysis of tso2 mutants???????.59
Figure 2-3. Semi-thin cross section of wild type and tso2-1 mutant leaves and siliques..60
Figure 2-4. Schematic diagram of map-based cloning of the TSO2 gene??????.62
Figure 2-5. Gene structures of the three R2 genes in Arabidopsis?????????.64
Figure 2-6. Genetic interaction among the three R2 genes???????????....66
Figure 2-7. tso2-1 is hypersensitive to HU treatment??????????????67
Figure 2-8.  Relative dNTP level in Ler and tso2-1 mutants???????????.69
Figure 2-9. mRNA expression of TSO2, R2A and R2B?????????????.70
Figure 2-10. DNA damage induction of RNR genes??????????????.72
Figure 2-11. Cell cycle arrest in tso2-1 mutants indicated by pcycB1::GUS expression..73
Figure 2-12. Increased sensitivity to UV-C and epigenetic misregulation in r2 mutants.75
Figure 2-13.  Increased DNA damage in tso2-1 r2a-1 seedlings?????????..77
Figure 2-14.  Expression of cell death markers in tso2-1 r2a-1 mutants??????..79
Figure 2-15. tso2-5 null allele display a wild type phenotype??????????..80
Figure 2-16. tso2-1 r2a and tso2-5 r2a exhibit similar phenotype?????????81
Figure 3-1. Representative suppressor lines of tso2-1?????????????..99
Figure 3-2. DS41 might be semi-dominant?????????????????..101
Figure 3-3. Phenotype of sua1-1 suppressor (DS101) in tso2-1 background????..102
Figure 3-4. Diagram of complementation test of suppressor lines????????..106
Figure 4-1. Eyes of Drosophila and leaves of Arabidopsis???????????.111
vi
Abbreviation
4-NQO4-nitroquinoline-1-oxide
APCanaphase-protomoting complex
ATM ataxia-telangiectasia mutated
ATR ataxia-telangiectasia and Rad3-related
bru1 brushy1
CAK CDK activating kinase
CAPScleavage amplified polymorphic sequence
cdc cell-division-cycle
CDK cyclin-dependent kinase
Cidcaffeine-induced death resistant
CKI CDK inhibitor
CKS CDK subunit
CLVclavata
Col Columbia
CRT constitutive RNR3 transcription
CZ central zone
dAK deoxyadenosine kinase
dCAPS     degenerate cleavage amplified polymorphic sequence
dCK deoxyguanosine kinase
DEL DP-E2F-like
dGK deoxycytidine kinase
DHF dihydrofolate
DHFR dihydrofolate reductase
DNDP deoxyribonucleoside diphosphate
DNMT DNA methyltransferase
dNTP deoxyribonucleoside triphosphate
DP DRTF-1 polypeptide
DSBdouble strand break
EMS ethyl methanesulphonate
fas fasciated
fis fertilization-independent seed
FM floral meristem
GST glutathionine-S-Transferase
GUS ?-glucuronidase
HDAC histone deacetylase
HMTase histone methyltransferase
HR hypersensitive response
HU hydroxyurea
Ler Landsberg erecta
LUC luciferase
MAPK mitogen-activating protein kinase
vii
MDS mitochondrial DNA-depletion syndromes
MMC mitomycin C
MMS methyl methanesulfonate
NDP ribonucleoside diphosphate
ORFopen reading frame
OC organizing center
PARP-2 poly (ADP-ribose) polymerase-2
PCD programmed cell death
PCR polymerase chain reaction
PR pathogen-related
PRXc peroxidase C
PZ peripheral zone
RB retinoblastoma
RBRretinoblastoma-related
RNRribonucleotide reductase
ROS                reactive oxygen species
RZ rib zone
SAG senescence-associated-gene
SAM  shoot apical meristem
SCG single cell gel electrophoresis
SEM scanning electron microscopy
sin short integuments
Sml  suppressor of mec1 lethality
Ssnsuppressor of snf1
sua shuai
TB trypan blue
TGS transcriptional gene silencing
THF 5,10-methylenetetrahydrofolate
TILLING        targeting induced local lesions in genomes
TK thymidine kinase
TS thymidylate synthase
WS Wassilewskija
wuswuschel
viii
Table of Content
Dedication???????????????????????????????ii
Acknowledgement???????????????????????????..iii
List of Tables?????????????????????????????..iv
List of Figures?????????????????????????????..v
List of Abbreviation???????????????????????????vi
Table of Contents???????????????????????????..viii
Chapter one Literature review???????????????????????.1
Overview????????????????????????????..1
Organization of shoot apical meristem?????????????????2
Maintenance of shoot apical meristem?????????????????5
CLV independent pathway????????????????????.?11
Cell division?????????????????????????.?..14
Cell cycle??????????????????????...??14
Progression of cell cycle??????????????????....15
Retinoblastoma (RB)????????????????????.17
E2F transcription factor???????????????????.19
Core cell cycle genes in Arabidopsis??????????????.20
Cell division and development???????????????????...24
DNA damage checkpoint?????????????????????...27
Programmed cell death??????????????????????..29
Ribonucleotide reductase (RNR)??????????????????...30
Biosynthesis and metabolism of deoxyribonucleotide???????..30
Salvage pathway?????????????????????....32
History of ribonucleotide reductase??????????????..34
Structure of ribonucleotide reductase??????????????36
Unusual organization of RNR genes in yeast???????????39
Regulation of RNR genes?????????????????.?.42
Controlled degradation of RNR protein?????????????42
Specified isoform of R2 for DNA repair????????????...43
Translocalization of RNR proteins in response to DNA damage???..44
Control of RNR mRNA stability???????????????...46
Protein inhibitor of RNR??????????????????...47
Repression of RNR gene expression??????????????.48
dNTP pool as a signal????????????????????50
Chemotherapy???????????????????????51
Chapter two Functional analyses of Arabidopsis ribonucleotide reductase small subunit
gene family??????????????????????.??..54
Abstract???????????????????????????.?.54
Introduction???????????????????????????55
Results?????????????????????????????57
tso2-1 is a pleiotropic mutant?????????????????.57
ix
Chloroplast development is defective in tso2-1??????????.60
Molecular cloning of the TSO2 gene??????????????.61
Arabidopsis genome has three R2 genes???????????.?...62
Genetic interaction among the three R2 genes?????????.?..65
tso2-1 is sensitive to hydroxyurea (HU)?????????????67
dNTP pool size is reduced in tso2-1 mutants???????????.68
The mRNA expression patterns of TSO2, R2A and R2B?????..?.69
DNA damage induction of the RNR genes????????????71
tso2-1 exhibit defects in cell cycle progression??????????.73
tso2-1 seedlings show increased sensitivity to UV-C????????74
tso2-1 r2a-1 double mutants exhibit defects in epigenetic regulation,
accumulate DNA damage, and undergo programmed cell death
(PCD)???????????????????????.76
tso2-1 is an recessive antimorphic allele????????????...79
Discussion???????????????????????????..81
Differential regulation of the three R2 genes??????????.?82
Induction of RNR genes by DNA damage??????????.??83
Decreased dNTP levels affect organelle replication????????..85
Fasciation is an indirect consequence of epigenetic mis-regulation in tso2
mutants?????????????????????.?..86
Cell division and development???????????????..?.89
tso2-1 is a recessive antimorphic allele???????????.??89
tso2-mediated PCD?????????????????????91
Chapter three Isolation of suppressors of tso2-1??????????????.?...93
Abstract???????????????????????????..?93
Introduction?????????????????????????..?..93
Results???????????????????????????..?..98
Discussion???????????????????????????104
Chapter four Concluding remarks?????????????????????107
Cell division and development???????????????????.107
Compensatory proliferation????????????????????.110
Chapter five Materials and methods????????????????????114
Plant growth, mutant strains, and materials??????????????.114
Genetic analyses????????????????????????..115
Microscopic and histological analysis????????????????.115
Map-based cloning of TSO2????????????????????116
Molecular analyses of TSO2, R2A and R2B????.????????.?...117
DNA damage induction of RNR expression??????????????118
Measurement of dNTP pool levels?????????????????..119
Comet assay??????????????????????????.119
UV-C sensitivity assay??????????????????????.120
DNA damage sensitivity assay???????????????????120
Cell death assays????????????????????????..121
x
Mutagenesis of tso2-1 seeds????????????????????121
Complementation test??????????????????????..122
Appendix??????????????????????????????..123
Bibliography?????????????????????????????125
1
Chapter one
Literature Review
Overview
The longevity of plants is beyond the reach of animals. One of the unique features
of plants is that plant growth continues throughout the lifespan and is modular. This
powerful postembryonic growth originates from the special organization of the shoot
apical meristem (SAM), the center for repetitive organ initiation. Due to the thick cell
wall, the movement of plant cells is forbidden. This emphasizes the importance of control
of cell division orientation and rate and cell expansion in plant development. How plant
cell proliferation is regulated within the context of development has been a subject of
debate for many years.
 Another unique property of plants is their sedentary life style and their
dependence on sunlight for photosynthesis. Therefore, plants are obliged to be constantly
exposed to environmental mutagens, including UV irradiation and reactive oxygen
species (ROS). The completed Arabidopsis genome revealed that plants have a large
number of DNA repair genes (The Arabidopsis Genome Initiative, 2000). For example,
Arabidopsis has six replication protein A (RPA1), four Rad23 for nucleotide excision
repair, sixteen DNA base glycosylases with roles in base excision repair, suggesting that
plants have a higher repair capacity in response to constant exposure of DNA damaging
agents. However, when DNA damage is beyond repair, apoptosis-like programmed cell
death can occur in plants. In addition, plants have evolved some active defense
2
mechanisms to effectively protect themselves from pathogen attack, such as
hypersensitive (HR)-mediated programmed cell death.
One key event in cell division and in DNA damage repair is DNA synthesis, in
which DNA strands are duplicated with high fidelity. In all organisms, ribonucleotide
reductase (RNR) provides the dNTPs for DNA synthesis by converting the
ribonucleoside diphosphates (NDPs) into deoxyribonucleoside diphosphates (dNDPs).
An ample and balanced dNTP pool is required for high fidelity DNA replication. Due to
its vital role in DNA synthesis, RNR has been the target site for chemotherapy for cancer
or infectious diseases.
This literature review aims to give relatively detailed background information for
the Ph.D. research reported in this thesis. First, the organization of shoot apical meristem
(SAM) and the maintenance mechanism underlying SAM will be described. Second, the
progression of cell cycle in plants and DNA damage checkpoint will be described and
discussed. Third, the regulation of cell division in the context of plant development will
be discussed. Finally, the history, structure, regulation and application of RNR will be
presented.
Organization of the shoot apical meristem (SAM)
Unlike animals whose body plan is already established in embryos, the
elaboration of plant architecture occurs largely post-embryonically. All above-ground
parts of flowering plants are derived from a group of cells positioned at the tip of the
3
stem, called shoot apical meristem (SAM). On one hand, the shoot apical meristem acts
as a self-renewing source of undifferentiated, pluripotent stem cells. On the other hand, it
is the reservoir of cells that will undergo differentiation and be utilized for organogenesis.
As a consequence, continuous and repetitive formation of new structures and organs such
as leaves and flowers results from the modular nature of the plant body plan. The
amazing regenerative power of the SAM enables plants to live up to several thousand
years as seen in some bristlecone pines.
How does the SAM maintain its regenerative ability for such a long time? Clonal
analyses demonstrated that all of the postembryonic structures of the plants are derived
from about three stem cells in every layer of the meristem (Stewart and Dermen, 1970).
To reduce the chance of accumulating mutations during many rounds of cell divisions,
stem cells in the SAM normally divide slowly. After cell division, some daughter cells
will stay in the center and remain as new stem cells, other daughter cells are displaced to
the periphery of the SAM to undergo differentiation into specific organs or secondary
shoots. As shown in Fig. 1-1, the stem cells are positioned in the central zone (CZ) of the
SAM. Surrounding the central zone is the peripheral zone (PZ), where cells divide
quickly and will become incorporated into lateral organs. Underneath the central zone is
the rib zone (RZ), where cell divisions give rise to the pith and the vascular system,
contributing to the growth of the stem. Cells in these different regions were first
distinguished by their cytohistological characteristics (Steves and Sussex, 1989), one of
which is the different cell proliferation rates measured by their mitotic index or
incorporation rates of radioactive thymidine. As noted earlier, cells divide more slowly in
4
the CZ than in the PZ and RZ. Obviously, in a growing meristem, a continuous flow of
cells will transit from the CZ to the PZ or the RZ.
The highly ordered structure of the SAM can also be divided into different layers.
The outermost two layers, called L1 and L2, remain clonally separated from the
underneath L3 cells as cells in L1 and L2 layers divide exclusively anticlinally (Fig. 1-1).
Cells in L3 layer apparently divide in random orientations.  The allocation of cells in the
meristem into separate clonal cell layers suggests cell-lineage-dependent mechanism of
development, which is, however, not the case. Studies with genetic mosaics showed that
cell fate in plants is determined by the position of a cell, not by its clonal origin. These
studies using genetic mosaics also showed that all three layers contribute to organ
formation and growth of the stem, indicating coordinated cell proliferation and cell fate
specification in different layers.
CZ
L1
RZ
PZ
PZ
CZ
PZ
PZ
RZ
L2
L3
SAMFMFM
    Figure 1-1. Schematic diagram of organization of shoot apcial meristem
(SAM). The tunica corresponds to L1 and L2 layers and corpus corresponds to
the internal L3 layer. The black outlines represent the approximate boundaries
between the different meristematic zones. Although the cells of the PZ and the
CZ are histologically distinguishable, there is no sharp boundary between each
zone (from Carles and Fletcher, 2003).
5
The intercellular communication could involve transfer of informational
molecules through plasmodesmata, which establish a cytoplasmic continuity between
neighboring cells. Therefore, the cells can act as a syncytium or symplasm. By tracking
the movement of an inflorescent tracer dye, Rinne and van der Schoot (1998) found that
the birch shoot meristem is compartmentalized into a central and a peripheral symplastic
field, which could restrict the diffusion of small potential morphogens to the cells inside
their boundaries. Gisel et al. (1999) reported that the symplastic connections between the
phloem and shoot apex change over developmental time and the extent of the symplastic
fields are dynamic.
Maintenance of shoot apical meristem
During the lifetime of a plant, the size of shoot apical meristem remains largely
the same. However, the proliferation of the stem cells and the rate of organ formation
vary in response to environmental stresses and developmental age. What is the
underlying molecular mechanism to maintain a consistent number of stem cells?
Mutations that cause a lack of shoot apical meristem or enlarged SAM led to the isolation
of several regulators of SAM for initiation and maintenance. Mutations in three clavata
loci (clv1, clv2, clv3) (Latin clavatus, meaning ?shaped like a club?) cause the plants to
produce a broader and distorted meristem, a phenomena called fasciation (Latin fascia,
meaning ?bundle?) (Fig.1-2B). clv mutants produce many extra floral meristems (FM)
and many extra floral organs (Carles and Fletcher, 2003; Clark et al., 1993, 1995; Kayes
and Clark, 1998). A simple explanation for this fasciation phenotype is that the three CLV
6
genes normally promote the transition of meristem cells from the central zone into the
peripheral zone. Alternatively, CLV may inhibit the proliferation rate in the central zone.
Since the mitotic index is not increased in the central zone in clv mutants, the increased
size in SAMs and FMs is likely caused by the accumulation of cells that wait to be
recruited by the developing organs in SAMs and FMs (Laufs et al., 1998).
Genetic and biochemical studies have demonstrated that the three CLV genes act
in the same pathway by forming a complex. CLV1 and CLV2 encode a receptor-like
kinase (RLK) and a receptor-like protein (RLP), respectively, whereas CLV3 encodes a
secreted polypeptide (Clark et al., 1997; Jeong et al., 1999; Fletcher et al., 1999). CLV3 is
mainly expressed in the L1 and L2 of the central zone while CLV1 mRNA is mostly
detected in the L3 of the central zone (Clark et al., 1997; Fletcher et al., 1999). The
expression domain of the CLV3 gene coincides with the location of stem cells and the
CLV3 gene has been used as a stem-cell marker. Consistent with its potential role as a
ligand for CLV1/CLV2 receptors, immunological studies showed that CLV3 is secreted
from the overlying L1 and L2 cell layers and moves into the underlying L3 and lateral
neighbors (Fig. 1-3). This is supported by genetic studies that the function of CLV3 is
dependent on both CLV1 and CLV2.
7
wuschel (wus) mutants display an opposite phenotype from clv. In wus mutants,
the shoot apical meristem is prematurely terminated, indicating that WUS gene is required
for the maintenance of SAM (Laux et al., 1996). Weak wus mutants can undergo iterative
processes of meristem initiation and premature arrest, leading to the production of
disorganized groups of leaves and shoots, like ?tousled hair? (the German meaning of
wuschel) (Fig. 1-2C). WUS encodes a homeobox protein and the mRNA is only found in
a few cells, which constitute the so-called ?organizing center (OC)? in the L3 layer of the
central zone, just beneath the CLV3 expression domain (Mayer et al., 1998).
Overexpression of WUS represses organ formation, induces CLV3 expression and forms
seedlings with an apex consisting entirely of undifferentiated meristematic cells (Schoof
et al., 2000). This suggests that WUS mis-expression is necessary and sufficient to induce
C D
    Figure 1-2. Mutants affecting shoot apical meristem development. (A)
Wild type. (B) clv3-2. (C) a weak wus-1 allele. (D) tso1-1.
8
CLV3 transcription and stem cell identity. Since WUS is not expressed within the stem
cells but in the underlying organizing center, WUS activity in the OC was proposed to
signal and specify the overlying cells as stem cells. Why are only the overlying cells
induced to become stem cells? One possibility is that only these cells are competent to
respond to the WUS-dependent signal. This is supported by observations that ubiquitously
expressed WUS induces CLV3 expression only in some cell types (Haecker and Laux,
2001). Alternatively, the WUS-mediated signal may be communicated via plasmodesmata
in one direction only. The existence of symplasm or syncitum makes it possible that the
movement or diffusion of small signal molecules is restricted (Rinne and van der Schoot,
1998).
wus mutants are epistatic to clv because clv wus double mutants display the wus
mutant phenotype. In clv3 mutants, the WUS expression domain expands laterally and
into the L2 layer, suggesting that the CLV pathway normally restricts WUS transcription
to the OC. Overexpression of CLV3 throughout the SAM can mimic the wus loss-of-
function phenotype, suggesting that CLV3 activity is sufficient to repress WUS in SAM.
Why is WUS only expressed in the OC in wild type plants? Perhaps, all CLV3 protein
produced in the apical stem cells is bound by CLV1 in the third cell layer, preventing it
from entering the OC cells underneath (Fig. 1-3). This is supported by the observation
that 75% of the total amount of CLV3 protein in cauliflower meristem extracts was found
to be associated with CLV1 (Trotochaud et al., 2000). In vivo evidence came from
transgenic studies. Expression of CLV3 from the epidermis using a L1-specific promoter
ATML1 could repress WUS in the OC non cell-autonomously and caused a wus mutant
9
phenotype. However, coexpressing CLV3 and CLV1 under the control of the ATML1
promoter exhibited a wild type phenotype. In contrast, coexpressing CLV3 and a mutant
CLV1 (clv1-4) under the control of the ATML1 promoter still lead to a wus-like
phenotype (Lenhard and Laux, 2003). Since clv1-4 is a mutant form of CLV1 that is
predicted to bind less CLV3 protein (Trotochaud et al., 1999), these results strongly
suggest that CLV1 can restrict the range of CLV3 action via sequestration of the ligand.
Therefore, CLV1 fulfills a dual function: On one hand, it relays the CLV3-dependent
signal into the receiving cells and ultimately causes repression of WUS transcription in
apical stem cells. On the other hand, CLV1 protects the underlying cells of the OC from
CLV3 by sequestering the ligand and thus, allows WUS expression in OC.
The feedback loop (Fig. 1-3) between CLV3 and WUS thus, underlies the self-
regulatory property of the SAM. If the WUS-dependent signal is too weak, CLV3
expression will be decreased, leading to specification of fewer stem cells. When CLV3
Figure 1-3. A model for CLV3 action as an intercellular signal in the SAM.
(cited from Lenhard and Laux, 2003)
CLV3CLV1WUS
10
level is low, the CLV3-dependent signal for WUS repression is also weakened. As a
consequence, the WUS expression domain will expand and thus specify more stem cells.
Conversely, if the WUS-dependent signal is too strong, CLV3 expression will be
increased and more stem cells will be specified. This, again, will lead to a strong
repression of WUS and reduce WUS expression domain, leading to fewer stem cells.
Thus, the size of the stem cell population in the SAM is continuously monitored and
adjusted by the WUS-CLV feedback loop, leading to a constant size of the SAM.
The above model does not answer the question about how the organizing center is
maintained, since new cells are continuously entering and leaving the WUS expression
domain (the organizing center). Periclinal divisions of the overlying L3 stem cells result
in a flow of cells through the OC: cells that enter the OC activate WUS expression,
whereas cells leaving the OC towards the rib zone switch WUS off. One explanation is
that the stem cells also send a positive signal inducing WUS expression in the OC. As the
strength of the repressive signal drops sharply due to the sequestration of CLV3 by CLV1
at the outmost L3 layer, the activating signal can reach OC. Alternatively, the position of
the OC is dependent on signals from underlying or lateral cells. It was shown that
meristem maintenance requires the presence of young leaf primordia and is influenced by
genes expressed in the leaves and vasculature (Moussian et al., 1998; Waites and Hudson,
1995).
11
CLV independent pathway
What happens if the SAM has more than one organizing center? What
mechanisms prevent plants from forming additional OCs in their SAM?  Arabidopsis tso1
mutants seemed to be defective in this process. tso1-1 also exhibited a fasciation
phenotype resulted from extensive bifurcation or multiple splitting of the inflorescence
meristems (Fig.1-2D). This is in contrast to the fasciaton in clv mutants, which is caused
by a single enlarged SAM. TSO1 encodes a putative DNA-binding protein with cysteine-
rich repeats, bearing similarity with Drosophila Enhancer of zeste (Song et al., 2000).
Song et al. proposed that one of the TSO1 functions might be to limit central zone
initiation either by repressing the expression of genes that normally activate central zone
formation or by mediating communications between the central zone and its neighbors to
prevent additional central zone initiation. Since cell division and cytokinesis defects were
also observed in tso1 mutants including partially formed cell walls and increased DNA
ploidy (Liu and Meyerowitz, 1997), it is possible that the communications between the
central zone and its neighbors are disrupted. It will be interesting to see if the CLV-WUS
genes are ectopically expressed in those bifurcated meristems.
Recently, this type of fasciation was reported in a set of mutants that are defective
in chromatin/DNA replication or DNA damage checkpoint.  For instance, AtCAP-E1 and
AtCAP-E2 are two condensing genes that are essential for mitotic chromosome
condensation. While E1
-/-
E2
-/-
 plants are embryo lethal, E1
-/+
E2
-/-
 plants exhibited stem
bifurcation associated with fasciation (Siddiqui et al., 2003). It is possible that the
chromatin remains relatively decondensated, which could activate a checkpoint control to
arrest the cell cycle. Actually, it has been shown that condensing is also involved in S-
12
phase checkpoint activation in other organisms (Aono et al., 2002). Since overexpression
of cyclin D3 results in a variety of developmental defects including meristem
disorganization (Riou-Khamlichi et al., 1999), this suggests that altered cell cycle can
affect meristem development. Alternatively, condensins and other factors involved in
DNA/chromatin dynamics can epigenetically alter gene expression programs. That is, in
E1
-/+
 E2
-/-
 mutants, expression of certain genes like WUS might be altered.
Another example is Arabidopsis MRE11 gene that encodes a component of the
MRX complex (Mre11, Rad50, Xrs2/Nbs1) with roles in repairing double-strand breaks
(DSB) and maintaining telomere length (Bundock and Hooykaas, 2002). It might be also
involved in DNA damage checkpoint (D?Amours and Jackson, 2002).  mre11 mutants are
hypersensitive to DNA damage and  fail to grow directly in soil. The most severely
affected seedlings grown on medium died after 8-weeks. Only the fittest mre11 seedlings
can survive in soil to form mature plants, which often showed bifurcation in the stems.
In fasciated (fas1, fas2) mutants, the clear histological distinction between the
peripheral zone and central zone was lost, and the regular cell arrangement in L1 and L2
was also disrupted (Kaya et al., 2001). The expression domain of WUS expanded laterally
but not uniformly in fas1 mutants. Sometimes the WUS expression domain also shifted
and/or expanded to outer cell layers, in some extreme cases, even to L1 cells. This is
different from clv mutants, in which the WUS expression domain laterally expands and
shifts one cell layer up but remains as two cell layers (Schoof et al., 2000). Genetic
studies showed that FAS and CLV genes function in different pathways (Leyser and
13
Furner, 1992). FAS1 and FAS2 genes encode the two subunits of the chromosome
assembly factor, which functions in assembling nucleosomes onto nascent DNA. It is
likely that the role of FAS genes is to facilitate stable maintenance of gene expression
states in the apical meristem.  In fas mutants, the epigenetic repression of WUS might be
lost, leading to ectopic expression of WUS.
The discovery of bru1 mutants leads to a novel link between reponse to DNA
damage and epigenetic gene silencing. bru1-1 was isolated from the screen for mutants
with elevated sensitivity to DNA damaging agents and bru1-2 was obtained from the
screen for mutations that release transcriptional gene silencing (TGS) at a transgenic
luciferease marker gene (Takeda et al., 2004). In bru1 mutants, the frequency of intra-
chromosomal homologous recombinantion is elevated and genotoxic stress responses are
constitutively activated, indicated by 2~3-fold induction of the poly (ADP-ribose)
polymerase-2 (PARP-2) gene under normal growth conditions. The release of TGS in
bru1 is stochastic, indicated by the expression pattern of LUC and GUS reporter genes. It
probably reflects random deregulation of silencing followed by mitotic transmission of
the newly acquired active state (Takeda et al., 2004). bru1 plants exhibited fasicated and
often bifurcated shoot meristems. This is reminiscent of fas, mre11 and AtCAP-E1
-
/+
AtCAP-E2
-/-
 mutants that are defective in chromatin/DNA replication and S-phase DNA
damage checkpoints of the cell cycle. Takeda et al. (2004) realized that there might be a
link between chromatin/DNA replication, TGS and meristem maintenance. They found
that fas mutants were hypersensitive to the DNA damaging agent MMS, and that fas,
mre11 and AtCAP-E1
-/+
AtCAP-E2
-/-
 mutants all abnormally express transcriptionally
14
silent information (TSI) (Steimer et al. 2000). TSI are pericentromeric repeats normally
silent in wild type plants. In addition, WUS was also ectopically expressed in a
stochastical manner in bru1 mutants. These results suggest that defects in the S-phase
DNA damage checkpoint or inaccuracies during chromatin/DNA replication cause
instability of epigenetic states, which is often manifested by meristem fasciation due to
mis-expression of meristem regulators such as WUS.
Cell division
Cell cycle
    To better understand the role of cell division in plant development, we need to
have good knowledge of cell cycle regulation in plants. At first glance, cell division is as
simple as duplicating (S phase) and dividing the DNA (M phase). The problem is: all the
genetic information of a cell has to be accurately copied and precisely segregated into
two genetically identical daughter cells. In addition, cells have to grow and double the
mass of proteins and organelles during the cell cycle. This is done in the so-called gap
phases, G
1
 and G
2
.  The gap phases also provide time for cells to check and ensure that
the external and internal conditions are suitable for the next phase, and that the previous
phase has been accurately and fully completed. In mammals, a cell can also enter a
specialized phase called G
0
 (G zero), in which cells permanently rest their division.
During G
1
, cells must integrate external and internal signals before making the
decision to initiate DNA replication, a commitment not only to the S phase but also to the
completion of the cell cycle. Once committed, cells cannot revert back to G1. During
15
DNA synthesis, there is a mechanism to assure that the DNA duplication occurs only
once. During G
2
, a checkpoint is turned on to ensure that the onset of mitosis occurs only
upon the completion of DNA replication and repair. During mitosis, the duplicated DNA
first condenses into chromosomes, and mitotic spindles begin to form (prophase). When
the nuclear envelope breaks down, the spindle microtubules are now able to interact with
the chromosomes. Then, the duplicated chromosomes are aligned on the mitotic spindle,
a characteristic of metaphase. During anaphase, the chromosomes are pulled to the
opposite poles of the cell. In telophase, chromosomes are completely separated into two
daughter cells. The nuclear envelope re-forms around the daughter nuclei. In animals,
cytokinesis occurs by constricting the cell membrane between the two daughter nuclei
until the final separation of the two cells. In plants, a structure called ?phragmoplast?
forms at the site of the future new cell wall, which directs vesicles carrying cell wall
materials to the center between the two daughter cells. Vesicle fusion results in new cell
wall formation. Phragmoplast assembly starts centrally and expands toward the parental
cell wall to divide the cell into two.
Progression of cell cycle
How is the cell cycle regulated at the molecular level? Extensive studies in yeast
and mammals have shown that the activation of a special class of serine-threonine protein
kinases, CDKs, is the core of cell cycle progression (Fig. 1-4). The activity of CDK
(cyclin-dependent kinase) rises and falls as the cells progress through the cell cycle.
CDKs are regulated at multiple levels, mainly by association and dissociation with
different sets of cyclins during different cell cycle phases. The variation in CDK-activity
16
leads to changes in the phosphorylation state of proteins that initiate or regulate cell cycle
events such as DNA replication, mitosis and cytokinesis. The CDKs are only active when
bound to cyclins. Cyclins undergo a cycle of synthesis and degradation in each cell cycle.
Consequently, the cyclic assembly and activation of the cyclin-Cdk complexes trigger
different events in the cell cycle.
In plants, in response to cytokinin, brassinosteriods and sucrose, cyclin D3 and
D2 are up-regulated, leading to activation of CDK-a. ABA can increase the level of
inhibitor KPP to repress the CDK-a activity and thus, inhibit cell division. Rising levels
of CycD?CDK-a kinase activity lead to phosphorylation of retinoblastoma-related
M
G2
G1
S
CycB
CDK-a/b
T161
T14
Y15
CycB
CDK-a/b
T161
T14
Y15
CycB
CDK-a/b
T161
T14
Y15
CycB
CDK-a/b
T161
T14
Y15
CDC25?
P
PP
P
P
ProphaseMeta/anaphaseTelophaseG1
Inactive
Inactive
Active
Active
Target to
APC
Fzr/ccs52-like
Proteolytic cycB
destruction
CDK-a/b
CycD/A
CDK-a
T161
CycD/A
Cytokinin
Sucrose
Brassinosteroids
(CycD3)
(CycD2)
CycD/A
T161
P
CAK
WEE1
KRPs
ABA
RBRE2Fs
CDK-a
CDC6
CDT1
ORC
MCM
RNR
S phase
DELs
Proteolytic
destruction
AXR1Auxin
Figure 1-4. Schematic diagram of cell cycle progression in plants. CDK activities peak
during G1?S and G2 ?M transitions. External and intrinsic signals have different
targets. This model is based on the mammalian cell cycle and adapted from den Boer
and Murray (2000). Most factors have been identified in plants except the cdc25
phosphatase.
17
protein (RBR) and inactivate RBR. Once RBR is inactive, the E2F transcription factor is
no longer repressed and S-phase genes are transcribed. DELs are E2F-like proteins but
act as repressors (see later). Auxin can up-regulate AXR1 protein, which promotes the
degradation of DEL protein. As a result, auxin promotes cell division. When cells
approaching M-phase, CycB ? CDK-a/b complex is inactive. The activation requires
both CAK kinase and removal of inhibitory phosphates on T14 and Y15 by a CDC25-like
phosphatase. The active complex triggers entry into prophase. A further control operating
between telophase and G1 requires the destruction of mitotic CDK activity by anaphase-
promoting complex (APC).
Retinoblastoma (RB) proteins
RB is the first tumor suppressor gene to be identified by its role in a rare pediatric
cancer called retinoblastoma. The RB gene is mutated in a wide range of cancers in at
least one third of all human tumors (Weinberg, 1992). RB is the main regulatory gene for
the G1-S transition of the cell cycle. In early G1, RB is found in an underphosphorylated
form. At a time close to the R (restriction) point, which demarcates the transition from a
serum-dependent to serum-independent state (Pardee, 1989), the G1 cyclins bind to RB
and direct their partner CDKs to phosphorylate and inactivate RB (Fig. 1-5b).
18
As shown in Figure 1-5a, RB harbors more than a dozen distinct sites of
phosphorylation on either serine or threnonine residue. The A/B pocket domains
physically interact with a set of protein factors with LXCXE motif including E2Fs and
HDAC. The C-terminal region is responsible for binding to cyclins through R/KXL
motif. In growth-arrested cells and during early G1, hypophosphorylated RB binds to
E2F/DP dimmer and consequently repress the E2F/DP transcription activity by recruiting
a set of chromatin-remodeling factors such as SWI-SNF (modification of nucleosome
structures), HDAC (deacetylation of histones), HMTase (methylation of histones) and
DNMT (methylation of DNA) to the promoters of E2F/DP target genes (Shen, 2002).
    Figure 1-5. The model for activation of E2F-pRb pathway. (a) Structure of
Arabidopsis retinoblastoma protein. Asterisks indicate the potential phosphorylation
sites by CDK-cyclin complex. (b) During G0 or early G1 phase, hypophosphorylated
Rb binds E2F/DP dimers and inactivates the E2F activity. During late G1 and early S
phase, Rb is hyperphosphorylated, first by CDK-CycD and then by CDK-CycA,
resulting in the dissociation of Rb from E2F. The released E2F/DP becomes active
and promotes transcription of E2f-target genes in cell-cycle regulation, DNA
synthesis and replication, and chromatin assembly(cited from Shen, 2002).
19
However, once hyperphosphorylated by the CDK/cyclin complex during late G1 and
early S phase, RB is no longer able to bind E2F/DP dimers as phosphorylation causes RB
to loose its grip on E2F, enabling E2F to activate a cohort of client genes (Chellappan et
al., 1991). Since E2F regulates the expression of numerous genes needed for cell cycle
entry and DNA synthesis, loss-of-function of RB will result in constitutive activation of
E2F, thus leading to loss of control of cell proliferation.
E2F transcription factors
As mentioned above, the key target of RB gene is E2F transcription factor. E2F
was a critical cellular factor in the E1A-mediated activation of the Adenovirus E2
promoter, thus, the name E2F (Nevins, 1992). It was observed that E1A caused a cellular
protein (later shown to be the RB) to dissociate from E2F and activate E2F by
sequestering RB. E2F is identical to the differentiation-regulated transcription factor
DRTF. In a variety of organisms, E2F is found to exist as a gene family. So far, there are
seven E2Fs reported in human.  In addition, a distantly related protein, called DP (first
identified as DRTF-1 polypeptide) (Girling et al., 1993), always forms a heterodimer
with E2F and binds to DNA.  Therefore, people very often combine the two, and use
E2F/DP to describe their function in literature.
The most intriguing aspect of the E2F gene family is the existence of a ?sibling
rivalry? among them. E2F1, E2F2 and E2F3 are activators whereas E2F4, E2F5, E2F6
and E2F7 are repressors. While activating E2Fs are specifically regulated by their
association with RB in normal cells, the repressive E2Fs are mainly associated with the
20
related pocket proteins (p107 or p130) or lack the binding domain for any of the pocket
proteins (see below) (Trimarchi and Lees, 2002). Nonetheless, they all recognize the
same consensus site TTTCCCGC. Hence, different transcription responses invoked by
the individual E2F/DP species depend on the identity of the E2F moiety as well as
proteins associated with them.  Overexpression of each activating E2F is sufficient to
irreversibly commit cells to re-enter the cell cycle, while the combined mutations of
E2F1, E2F2 and E2F3 are sufficient to completely block cellular proliferation (Lukas et
al., 1996;Wu et al., 2001). In addition, de-regulated E2F activity can trigger apoptosis
through either p53-dependent or ?independent mechanisms when the total pool of
transcriptional activity contributed by the activating E2Fs exceeds a higher threshold
(Trimarchi and Lees, 2001). Although the three activating E2Fs share redundant
functions, they might play different roles in normal development shown by the tissue-
specific defects of corresponding E2F-difficiency mice (Trimarchi and Lees, 2001).
Core cell cycle genes in Arabidopsis
Although cell cycle has been extensively studied in yeast and animals, little is
known about the regulation of cell cycle in plants. The high level of sequence similarity
shared between the plant and animal cell cycle regulators such as E2F/DP, RB proteins,
and other core cell cycle regulators, suggests the conservation of cell cycle mechanism.
Recent high-quality, homology-based annotation of the Arabidopsis genome revealed that
there are 61 core cell cycle genes, belonging to seven selected families (Vandepoele et al,
2002) (Table 1-1). In yeast, one CDK is sufficient to drive cells through all cell cycle
phases whereas in multicellular organisms, a family of related CDKs with specific
functions are involved. The Arabidopsis genome has eight CDKs with one A-type, four
21
B-type, two C-type, one E-type and four CDK-activating kinases (CAK) (three D-type
and one F-type) (Vandepoele et al, 2002). A-type CDKs have been shown to regulate
both the G1-to-S and G2-to-M transitions whereas B-type CDKs is only involved in
controlling G2-to-M checkpoint (Porceddu et al., 2001). The large number of CDK genes
suggests the complexity of cell cycle regulation in plants.
Major playersSubtypesNo. of genesNote
CDKCDK-A1 G1-S and G2-M
CDK-B4 G2-M
CDK-C2 ?
CDK-E1 ?
CAKCAK-D3 CDK activation kinase
CAK-F1
CYCACYCA12 mitotic cyclins
CYCA24
CYCA34
CYCBCYCB14 mitotic cyclins
CYCB24
CYCB31
CYCDCYCD11 G1 cyclins
CYCD21
CYCD33
CYCD42
CYCD51
CYCD61
CYCD71
CYCHno 1 regulate CAK activity
CKSno 2 docking factros for CDK
KRPno 7 CDK inhibitor
WEE1no 1 negative regulator of CDK/cyclin
RBRno 1 tumor suppressor
E2FE2Fa1 transcription factor (G1-S transition)
E2Fb1
E2Fc1
DP DPa 1 binding partners for E2F
DPb1
DELDEL11 DP-E2F-like, repressor
DEL21
DEL31
Table 1-1. Core cell cycle genes in Arabidopsis
22
The timing of CDK activation is determined mainly by a large number of cyclins,
whose levels fluctuate in the cell cycle. There are two major groups of cyclins:  mitotic
cyclins (A- and B-type cyclins) and G1-specific cyclins (D-type cyclins). H-type cyclins
regulate the CAK activity. Arabidopsis has ten A-type cyclins, nine B-type cyclins, ten
D-type (seven subclasses: one D1-type, one D2-type, three D3-type, two D4-type and one
D5, D6, D7-type). There is only one H-type cyclin in Arabidopsis genome.
There are two CKS (CDK subunit) proteins (CKS1 and CKS2) in Arabidopsis,
which act as docking factors to mediate the interaction between CDKs and their
substrates or regulatory proteins. CDK inhibitor (CKI) proteins can bind to specific
CDKs and inactivate CDKs. In Arabidopsis, all seven CKI genes belong to the group of
Kip/Cip CKIs and hence, are designated KRP1 to KRP7 (De Veylder et al., 2001).
CDK/cyclin activity is negatively regulated by WEE1 kinase, which phosphorylates CDK
subunit. This inhibitory effect can be removed by CDC25 phosphatase. There is one
WEE1 homolog gene but no CDC25 homolog in Arabidopsis.
RB (retinoblastoma) and the E2F/DP proteins are key regulators that control the
start of DNA replication. While only one Rb gene exists in the Arabidopsis genome, eight
E2F/DP genes are found, which can be divided into three groups. The first group
comprises E2Fa, E2Fb, and E2Fc, which are most similar to the mammalian E2F factors
(46% overall similarity). The second group consists of two DP genes (DPa and DPb). The
third group contains three genes with 59% similarity among themselves, 21% and 18%
similarity to E2F and DP genes, respectively, which is named DP-E2F-like (DEL1,
23
DEL2, DEL3). They were previously named E2Fd, E2Fe and E2Ff, respectively. Unlike
the E2F and DP genes that harbor DNA binding, dimerization, Marked and Rb binding
domains, DEL genes only contain two duplicated DNA binding domains that are highly
similar to the E2F DNA binding domain (Fig. 1-6). The DNA binding domain of
E2F/DP/DEL recognize the same consensus motif (TTTSSCGSS, S being C or G).
Based on the DNA binding site, Ramirez-Parra et al (2003) conducted a genome-
wide analysis of E2F-regulated genes in Arabidopsis. In addition to a large number of
genes belonging to cell cycle and DNA replication (31.3%), and transcription (21.7%), a
variety of putative E2F target genes are found to be involved in other processes such as
defense, signal transduction, and metabolism. This is consistent with the observation in
animals that E2F/DP transcription factors regulate gene expression not only in
proliferating cells but also during differentiation, development and apoptotic response
(Muller et al., 2001; Ren et al., 2002; Wells et al., 2002). Further evidence came from a
E2Fa
E2Fc
E2Fb
DPa
DPb
DEL1
DEL2
DEL3
355
360
404
387
292
483
396
4691
1
1
1
1
1
1
1
DNA binding
domain
Dimerization
domain
Marked
box
Rb binding
domain
    Figure 1-6. Structures of E2F gene family. E2Fa, E2Fb and E2Fc are in the same class
while DPa and DPb belong to the second class. DEL1, DEL2 and DEL3 are unique in
that they lack a dimerization domain and act more likely as repressors (cited from
Vandepoele et al, 2002).
24
genome-wide microarray analysis of E2Fa-DPa overexpression in plants (Vlieghe et al.,
2003), in which a number of up-regulated genes are involved in cell wall biosynthesis
and nitrate assimilation.
Cell division and development
A fundamental question in plant development is how cell division events are
coordinated and integrated with cell expansion, leading to the normal growth patterns.
The basic cell cycle machinery in plants is controlled by a variety of internal and external
signals (Hemerly et al, 1999; Mironov et al, 1999). Upstream controls can act at distinct
levels such as modulating CDK level and activation, cell cycle timing and cell division
pattern and frequency to generate the final plant architecture. Previous studies showed
that plant morphogenesis is not strictly dependent on rates and precise planes of cell
division (Smith et al. 1996; Reynolds et al. 1998; Mironov et al. 1999). For example, the
maize tangled mutant has misaligned cell division orientations in developing leaves, but
its overall leaf shape is normal (Smith et al., 1996). In Arabidopsis short integuments 2
(sin2) mutants, cell number in the outer integuments of ovules is reduced 5-10-fold, but
the overall morphology of the sin2 is relatively normal (Broadhvest et al., 2000). It was
suggested that when the basic cell cycle machinery is somewhat perturbed, it can be
locally sensed via cell-cell communication and finally balanced with cell expansion to
maintain normal growth patterns (Meyerowitz 1996; Scheres and Heidstra 1999).
Obviously, this kind of tolerance of altered cell number or size has its limit.
25
When specifically expressed in embryos under the promoter of 2S2 albumin gene,
the dominant-negative cdc2aAt mutant causes a distorted apical-basal pattern, suggesting
that cell division events are essential for the elaboration of the apical-basal pattern and
morphogenesis of embryos (Hemerly et al., 2000). Constitutive expression of E2F/DP in
tobacco altered plant morphology with curled leaves, round petals and shortened pistils.
Mature leaves of the transgenic plants contained increased numbers of small cells
(Kosugi and Ohashi, 2003).
Overexpression of E2Fa/DPa in Arabidopsis also results in uncontrolled cell
division and delayed differentiation. Transgenic plants arrest growth early with several
curled leaves (De Veylder et al., 2002). This is in contrast to ectopical expression of
cyclin B1 and cyclin D2, which promote plant growth (Doerner et al., 1996; Cockcroft et
al., 2000). It suggests that overexpression of E2Fa/DPa overrides signals that regulate cell
differentiation and that the balance between division and differentiation is vital for
normal plant development. Arabidopsis cyclin D3 is transcriptionally regulated by
sucrose and cytokinins (Soni et al., 1995), suggesting that cyclin D3 could relay
developmental signals to the cell cycle machinery. Overexpression of cyclin D3 results in
developmental aberrations such as extensive leaf curling, disorganized shoot meristem
and delayed senescence (Riou-Khamlichi et al., 1999).
However, most of these studies are limited to ovexpression of cell cycle genes and
these gain-of-function studies might not reveal the normal function of each of the cell
cycle genes in plant development. On one hand, extreme redundancies among the cell
26
cycle genes in plants prevent single loss-of-function mutants from exhibiting any
developmental defects. On the other hand, some cell cycle genes play such essential roles
that they are indispensable for viability. For instance, the anaphase-promoting complex
(APC) regulates mitotic progression and exit by controlling the stability of cell cycle
regulatory proteins such as mitotic cyclins. Arabidopsis apc2 is female gametophytic
lethal and arrested mainly at one or two-nucleus stage of megagametogenesis (Capron et
al., 2003). In hobbit mutants, the organization of cell layers in shoot apical meristem is
disrupted due to unordered cell division patterns. Cell division ceases after an initial
proliferation phase, leading to seedling lethality (Billou et al., 2002). HOBBIT encodes a
homolog of the CDC27 subunit of APC. The deregulation of auxin reporter genes and
repressors AXR3/IAA17 in hobbit mutants suggests that HOBBIT/CDC27B controls
auxin-mediated cell division and differentiation responses.
Mutations in plant retinoblastoma protein (RB) homolog also cause a
gametophytic lethality (Ebel et al., 2004). Arabidopsis is currently the only plant known
to contain a single retinoblastoma-related gene (RBR1). In rbr1 mutants, the mature
unfertilized megagametophyte fails to arrest mitosis and undergoes excessive nuclear
proliferation in the embryo sac, which is in consistent with its negative role in cell cycle
progression. In addition, the central cell nucleus in rbr1 mutants can initiate autonomous
endosperm development reminiscent of fertilization-independent seed (fis) mutants (Ebel
et al., 2004), suggesting that correctly exiting the cell cycle is required for normal
gametogenesis and repression of autonomous endosperm development.
27
DNA damage checkpoint
Progression of the cell cycle is not always smooth as cells are facing constant
physical and chemical assaults including reactive metabolic products, which can induce
heritable mutations or DNA strand breaks. A surveillance mechanism has evolved so that
the integrity of DNA is monitored during the cell cycle. The DNA damage checkpoint
will slow or arrest the cell cycle and lead to induction of genes participating in DNA
repair. Increasing evidence suggests that these checkpoints are not specific points during
the cell cycle at which genomic integrity is assessed, instead they continuously monitor
the integrity of the genome and control cell cycle progression accordingly (Nasmyth,
1996). The same proteins involved in regulating the orderly progression through the cell
cycle are also involved in the checkpoint responses. Conversely, DNA damage
checkpoint proteins like ATM and ATR, previously thought to be only involved in DNA
damage responses, have been shown to regulate normal DNA replication in unperturbed
cells (Shechter et al., 2004).
The DNA damage checkpoint has three components; sensor, signal transducer,
and effectors (Sancar et al., 2004). The sensors monitor DNA for structural abnormalities
and then initiate the checkpoint signal. Transducers further transmit and amplify this
signal. Effectors execute the action leading to the biological consequences. However,
there is no absolute demarcation between the various components of the checkpoint. For
instance, the damage sensor ATM also acts as a signal transducer (Sancar et al., 2004). A
fourth class of checkpoint proteins, called mediators, is conceptually placed between
sensors and signal transducers. In addition, many components of the DNA damage
checkpoint pathways are shared with G1/S, intra-S, and G2/M checkpoint pathways.
28
Checkpoint proteins are well conserved from yeast to human, indicating that the
basic organization of these pathways has been preserved throughout evolution. For
example, yeast Mec1 and human ATM/ATR are members of the evolutionarily conserved
subfamily of phosphatidylinositol 3-kinase (PI3-kinase) (Elledge, 1996). In yeast, the
ATM/ATR
CHK1/CHK2
p53Cdc25
Rad9
Rad53
Mec1
Dun1
Crt1
Sensors
Transducers
Effectors
CDC28SWI6
Budding yeastHuman
DNA damage
Stalled replication
PDS1
Induction of
DNA repair
G1 arrest
Metaphase
arrest
Apoptosis
G1/S checkpoint
Intra-S checkpoint
G2 checkpoint
Cellular
responses
?
BRCA1p53BPMediators
    Figure 1-7. Components of the DNA damage checkpoints in yeast and human
cells. The damage is detected by sensors. With the aid of mediators, sensors
transduce the signal to transducers, which, in turn, activate or inactivate other
proteins. These effectors directly participate in various cellular responses including
cell cycle arrest and damage repair. Proteins shown in same color are conserved from
yeast to human.
29
Rad24/Rad17 subclass of genes and Rad9 are the two major proteins involved in sensing
DNA damage (Lowndes and Murguia, 2000). Although Mec1, Rad53 and Dun1 are
regarded as the signal transducer, Mec1 is actually required for phosphorylation of Rad9
(Vialard et al., 1998). Rad53 can recognize and specifically bind to hyperphosphorylated
Rad9 (Durocher, et al., 1999). Since Rad9 remains bound to chromation after DNA
damage, this mechanism would recruit Rad53 to lesions, allowing localized amplification
of the checkpoint signal (Lowndes and Murguia, 2000).
In mammalian cells, the tumor suppressor p53 protein is the core of the DNA
damage checkpoint pathway. As a transcription factor once activated by DNA damage,
p53 can turn on p21 transcription, which in turn inactivates CDK4. As a consequence,
cell cycle is arrested at G1/S phase. P53 also regulates a large number of target genes
including those involved in DNA repair. In addition, p53 can monitor the extent of DNA
damage and trigger apoptosis to remove damaged cells. Cells undergoing apoptosis
shrink, develop bubble-like blebs on their surface, and degrade their chromatin. The
mitochondria of apoptotic cells eventually break down into small, membrane-wrapped
fragments and are engulfed by phagocytic cells (Lam et al., 2001). The pattern of events
is so orderly that the process is often called programmed cell death (PCD).
Programmed cell death
Programmed cell death (PCD) is needed not only to destroy cells that represent a
threat to the integrity of the organism, such as those damaged cells, but also intrinsic to
the cell for proper development. For instance, apoptosis occurs during the resorption of
30
the tadpole tail at the time of its metamorphosis.  The formation of the fingers and toes of
the fetus requires removal of the tissues between them, which is carried out by apoptosis.
In plants, PCD is needed for terminal differentiation of xylem vessels. Senescence is an
active programmed cell death process to recycle the energy and resources to other parts.
Hypersensitive response (HR)-mediated PCD is an important mechanism in defense
against pathogens.
Several morphological and biochemical similarities were found between animal
cells undergoing apoptosis and plant cells undergoing PCD, such as condensation and
shrinkage of the cytoplasm and nucleus, caspase-like proteolytic activity, cytochrome c
release from mitochondria, the formation of DNA-containing (apoptotic-like) bodies and
genomic DNA degradation or DNA laddering (Pennell and Lamb, 1997; Danon et al.,
2000). However, plant cells do not engulf their dead neighbors due to the presence of cell
wall. Instead, in xylogenesis, the dead plant cells become part of the very architecture of
the plant performing crucial functions such as water conducting and mechanical support
(Greenberg, 1996).
Ribonucleotide reductase (RNR)
Biosynthesis and metabolism of deoxyribonucleotide
Deoxyribonucleotides are of paramount importance for DNA replication and
repair. Two processes ? the conversion of ribose to deoxyribose and the conversion of
uracil to thymine, are essential to de novo deoxyribonucleotide biosynthesis. Not
31
surprisingly, both processes are target sites for chemotherapy for cancer or infectious
diseases. Thus, enzymes catalyzing both steps are under intensive study.
Ribonucleotide reductase (RNR) catalyzes the reduction of all four common
ribonucleotides to the corresponding 2?- deoxyribonucleotides in all organisms. RNR acts
upon ribonucleoside diphosphate (NDPs) substrates. Once reduced, dADP, dGDP and
dCDP are converted directly to the corresponding triphosphate by nucleoside diphosphate
kinase. Biosynthesis of dTTP is completed partly from dUDP and partly from dCDP (Fig.
1-8). First, dCDP is dephosphorylated to dCMP, which then undergoes deamination by
dCMP deaminase to become dUMP. Second, dUDP is phosphorylated to dUTP, which is
then cleaved by highly active dUTPase, to produce dUMP. This step is very important to
exclude uracil from DNA.  dUMP can arise in DNA either by spontaneous deamination
of dCMP, which will lead to mutations, or by dUTP incorporation. Since the cell cannot
tell whether a dUMP arises from the deamination of dCMP or from natural incorporation,
a surveillance system has evolved to excise dUMP no matter how it arises.
After dUMP is formed, it is converted to dTMP by thymidylate synthase (TS).
5,10-methylenetetrahydrofolate (THF) serves as a redox cofactor to give rise to
dihydrofolate (DHF) as a byproduct of the reaction (Fig. 1-8). In some species, such as
plants, dihydrofolate reductase (DHFR) and thymidylate synthase (TS) are encoded by a
single gene, called DHFR-TS. dTMP, once formed, is converted to dTTP by two
successive phosphorylations.
32
Salvage Pathway
Conversion of NDPs into dNDPs is not the only pathway for dNTP synthesis.
Free deoxypurine nucleosides and deoxypyrimidine nucleosides can also be utilized for
dNTPs production, which is called the salvage pathway (Fig. 1-8). This is accomplished
by the corresponding deoxyribonucleoside kinases including thymidine kinase (TK),
deoxyadenosine kinase (dAK), deoxyguanosine kinase (dGK) and deoxycytidine kinase
ADPdADPdATP
GDPdGDPdGTP
CDPdCDPdCTP
dUMP
dCMP
HU
(1)
(1)
(1)
(1)
dTMPdTDPdTTP
UDPdUDPdUTP
THFDHF
FDUR
DNA
dNTP
dNDP
dNMP
(2)
MTX
dU
dAdCdG
T
Nucleic acid metabolism
(5)
(4)
De Novo synthesis
(1) Ribonucleotide reductase
(2) Thymidylate synthase
(3) Dihydrofolate reductase
(4) DCMP deaminase
(5) Salvage pathway
Thymidine kinase
Deoxyadenosine kinase
Deoxycytidine kinase
(3)
    Figure 1-8.  Diagram of de novo and salvage pathways for biosynthesis of dNTPs. The
salvage pathway allows the reutilization of material obtained from degradation of DNA,
regulates the flux of deoxyribonucleotides in and out of cells and together with the de novo
pathway, contributes to the setting of the intracellular levels of dNTPs.
33
(dCK). Salvage pathways vary from species to species, partly in their contents of
deoxyribonucleoside kinases. For example, human cells contain only four
deoxyribonucleoside kinases including a cytosolic dCK, a mitochondria-localized dGK
and two isoforms of thymidine kinase: one localized in cytosol (TK1) and one in
mitochondria (TK2). The dCK can act on deoxyadenosine and deoxyguanosine as well as
deoxycytidine. The TK2 can also act upon deoxycytidine and deoxyuridine. Of the four
deoxyribonucleoside kinases, only TK1 is cell cycle-regulated with peak expression at S-
phase, resembling enzymes of the de novo deoxyribonucleotide synthesis. Since
mitochondrial DNA is constantly replicating even in quiescent cells, the constant supply
of dNTPs is vital to the maintenance of the mitochondrial genome. Hence, the main
supply of dNTPs for mtDNA synthesis comes from the salvage pathway initiated by dGK
and TK2, which are constitutively expressed.
The importance of salvage pathway for dNTP synthesis is best illustrated by an
autosomal recessive disorder called mitochondrial DNA-depletion syndromes (MDS).
MDS patients show progressive muscle weakness or liver failure, and the majority of
affected individuals die during the first year of life (Mandel et al. 2001). The
mtDNA/nuclear DNA ratio could be reduced to 8-39% of the control mean in MDS
patients, mainly in liver and brain tissues. MDS is caused by defective dGK (Mandel et
al. 2001). Saada and his colleagues (2001) showed that mutations in TK2 are responsible
for the mitochondrial DNA depletion myopathy.
34
History of ribonucleotide reductase
 The first ribonucleotide reductase (RNR) was cloned in E.coli in 1961, a big
shock to the field since back then, organic chemists were not aware of reactions in which
a carbon-bound OH-group could be replaced directly by hydrogen (Jordan and Reichard,
1998). RNR was the first protein free radical to be discovered: an organic free radical,
identified as tyrosyl-122, forms part of the polypeptide structure of the small subunits of
the enzyme and is essential for its activity (Ehrenberg and Reichard, 1972; Sjoberg et al.,
1978). The radical is so stable that it survived the two-week period taken to purify the
enzyme.
The E.coli RNR became the prototype of class I enzymes to which all eukaryotic
reductases belong. Later, class I reductases were subdivided into two subgroups. One
class (Ia) has the original E.coli enzyme as the prototype, which also includes the
eukaryotic enzymes. The second class (Ib) is present exclusively in bacteria. RNR in
Lactobacillus leichmannii was found to require adenosylcobalamin, leading researchers
to believe that the reductases were B
12
 enzymes. This is because pernicious anemia, a
disease caused by defective DNA synthesis due to malfunctional RNR, can be cured by
administration of vitamin B
12
 (Blakley and Barker, 1964). The requirement of
adenosylcobalamin is an exclusive feature of class II reductases (see Table 1-2). Since the
generation of the tyrosyl radical of class I enzymes requires oxygen for its formation, the
discovery of a second ribonucleotide reductase in anaerobically growing E.coli came as
no surprise (Fontecave et al., 1989). The anaerobic enzyme, the prototype of class III
reductases, contains an iron-sulfur cluster and uses this cluster and S-adenosylmethionine
to generate the glycyl radical (Ollagnier et al., 1996).
35
The three RNR classes share a common tertiary structure in spite of large
differences in the primary sequences. The known three-dimensional structures of class I
and class III show overwhelming structural similarity. Sequence comparison between
class I and class II enzymes shows that they contain identical, strategically placed amino
acids involved in catalysis and allosteric regulation. With the elucidation of an increasing
number of primary sequences of class II and class III enzymes, there is now almost no
doubt that the R1 large subunits of the three classes share a common ancestor.
As shown in Table1-2 (adapted from Jordan and Reichard, 1998), all classes but
class II enzymes consist of two homodimeric proteins, R1 (?
2
) and R2 (?
2
). The large R1
subunit of class I enzyme harbors the catalytic site and binding sites for allosteric
effectors whereas the small R2 subunit of class I enzymes contains an oxygen-linked
diferric center and a stable tyrosyl free radical. In contrast, a stable oxygen-sensitive
Class IaClass IbClass IIClass III
Oxygen dependenceAerobicAerobicAerobic/anaerobicAnaerobic
Structurea2b2a2b2 a(a2) a2b2
GenesnrdABnrdEF nrdDG
RadicalTyr?CysTyr?CysAdB12?CysAdoMet?Gly?Cys?
Metal siteFe-O-FeFe-O-FeCo Fe-S
Mn-O-Mn
SubstrateNDPNDPNDP/NTPNTP
ReductantThioredoxinNrdH-redoxinThioredoxinFormate
GlutaredoxinGlutaredoxin
Allosteric sites/2 1 1 2
  polypeptide chain
dATP inhibitionYes No No Yes
OccurrenceEukaryotesArchaebacteriaArchaebacteria
EubacteriaEubacteriaEubacteriaEubacteria
BacteriophagesBacterophagesBacterophages
Viruses
PrototypeE.ColiS.typhimuriumL.leishmanniiE.Coli
MouseC.ammoniagenes
Table 1-2. Overview of the characteristics of the ribonucleotide reductase classes
36
glycyl radical is located on the large subunit of class III enzymes whereas the small
subunit contains an iron-sulfur cluster that together with S-adenosylmethionine generate
the glycyl radical. Since the glycyl radical of class III is rapidly destroyed by oxygen
(King and Reichard, 1995), class III enzymes are strictly anaerobic. However, radical
formation from adnosylcobalamin of class II enzymes does not require oxygen nor is the
radical sensitive to oxygen. Therefore, class II enzymes function in both aerobic and
anaerobic organisms.
Structure of ribonucleotide reductase
Class I RNR is a tetramer protein complex (Fig. 1-9 and Fig.1-10), consisting of
two nonidentical homodimers: the large R1 subunit and the small R2 subunit. The large
homodimer mediates both catalysis and allosteric interactions and the small homodimer
provide the tyrosine free radical for catalysis. The large subunit has three binding sites.
The first site binds ATP/dATP to control the overall activity. The negative feedback
regulation of RNR by dATP is a mechanism to make sure that cells will not use up NTPs,
which are the backbones for RNA synthesis. The second site binds ATP, dATP, dGTP
and dTTP to control the specificity. This level of regulation leads to the production of a
balanced dNTP pool (illustrated in Fig. 1-9). The third site binds the substrates (NDPs) as
a catalytic site. In addition, there is a redox site where thioreddoxin or glutaredoxin
reduces the disulfide bonds generated during catalysis. The ultimate reducing power
comes from NADPH.
37
How does one enzyme catalyze the reduction of four different ribonucleotides and
how does it manage to produce a balanced dNTP pool? The three binding sites for
activity, specificity and catalysis in the R1 large subunit provide an intricate regulation of
the order in which ribonucleotides are bound and reduced (Fig. 1-10). The rationale is
that cell division will not occur unless the cell?s energy state is high. When the cell?s
energy state is high, the concentration of ATP is high but the dNTP level is low. So, ATP
will bind to the activity site of R1, turning on the RNR activity. ATP will also binds to
the substrate specificity site, leading to the production of dUDP and dCDP. As noted
earlier, both dUDP and dCDP are precursors for dTTP. When the concentrations of
dUDP and dCDP go up, the concentration of dTTP also rises. Then, dTTP is able to
compete with ATP for the substrate specificity site. When dTTP is bound, then GDP
becomes the substrate and dGDP is made. dGDP is rapidly converted into dGTP, which,
    Figure 1-9. Schematic diagram of RNR structure. R1 dimer are shown as blue
spheres while R1 dimer in red. The diferric iron center is linked to oxygen and is
essential for radical generation on the neighboring tyrosine. (cited from Mathews,
van Holde and Ahern (2000) Biochemistry, 3
rd
 edition p810)
38
upon reaching a certain concentration, will compete with dTTP for the specificity site.
When dGTP is bound to the specificity site of R1, ADP becomes the substrate and dADP
is produced. dADP is again converted into dATP, which, upon reaching a certain
concentration, will compete with ATP for the activity site. When dATP is bound the
RNR is turned off.
As shown in Fig. 1-11, the R1 dimer (on the top) is an unstable structure as both
subunits and domains can move with respect to each other. This may form a structural
basis for the allosteric regulation and catalysis.  The binding of R2 can stabilize the R1
dimer. The shape of the R1 dimer is complementary to the upper part of the heart-shaped
structure of the R2 dimer (on the bottom), allowing the R1 dimer to sit like a saddle on
the top of the R2 dimer. The upper part of R2 fits neatly into the cavities of R1, in
agreement with the location of conserved residues in this part of the R2. There is still
space for the substrate to pass into the active site, but only nucleoside diphosphates are
allowed to bind as substrates as the ribose moiety of a nucleoside triphosphate would not
be able to bind close to the active site (shown as yellow spheres in Fig. 1-11).
CDP
UDP
GDP
ADP
dCDP
ATP
dTTP
dATP
dUDP
dGTPdGDP
dADP
dCTP
ATP
+
+
+
+
 Figure 1-10. Schematic
diagram of regulation of
RNR activity. The red
rectangle bar designates
the activity site, which is
activated by ATP and
inhibited by dATP. The
red dashed line indicates
inhibition and green line
indicates activation.
39
The C-terminus of R2 protein (at most 30 residues) accounts for all of its
interactions with R1 in E.coli.  Like the N-terminus, this carboxyl domain is not
conserved between species, reflecting the species-specific interaction between R1 and
R2. Surprisingly, the last two residues of R2 are always Asp and Phe in most species. As
short as 7 polypeptides, corresponding to the C-terminus of R2, can strongly compete
with the full-length R2 (375 aa) for binding of R1. Substitution of one amino acid in the
C-terminus of R2, such as E350A or Y356A can still bind to R1 tightly. However, E350A
exhibits a low activity (240 times less active than the wild type), and Y356A is
completely inactive (Climent et al, 1992), suggesting that the conformation of the
holoenzyme complex might be changed.
Unusual organization of RNR genes in yeast
The large subunit and the small subunit are encoded by two genes in yeast, RNR1
and RNR3 for R1, and RNR2 and RNR4 for R2 (see Table 1-3). Unlike other yeast RNR
    Figure 1-11. Three-dimensional structure of RNR. The R1 dimer sits like a
saddle on the top of R2 dimer, which is heart-shaped. (cited from Uhlin  and
Eklund, 1994)
40
genes, expression of RNR3 is undetectable under normal growth conditions but can be
induced up to 100-fold by DNA damage (Elledge and Davis, 1990), suggesting that
RNR3 is mainly involved in the DNA repair pathway. Nevertheless, the RNR3 mutant has
no obvious phenotype, even under DNA damage conditions. Domkin et al (2002) found
that the in vitro activity of RNR3 is less than 1% of the RNR1 activity even though the
two proteins share 80% identity. This is partly because RNR3 does not readily dimerize.
In vivo, the protein levels of RNR3 after DNA damage never reach more than 10% of the
RNR1 levels. However, a strong synergism between the RNR3 and RNR1 was observed.
A catalytically inactive RNR1-C428A mutant can increase the endogenous activity of
RNR3 by at least 10-fold. This, in combination with the fact that RNR3 lacks a functional
allosteric activity site, may provide a selective advantage for yeast growing in natural
ecological niches under permanent DNA-damaging conditions when the levels of RNR1
are limiting for ribonucleotide reduction.
RNR4 reveals another intriguing story. RNR4 and RNR2 are only 54% identical.
In addition, RNR2 contains a 51-amino-acid N-terminal peptide that is absent in RNR4.
Most strikingly, six of 16 residues that are absolutely conserved in all R2 proteins from
E.coli to mammals are changed in RNR4, including the two histidines and one glutamate
that are directly involved in the coordination of the iron complex (Huang and Elledge,
1997). This suggests that RNR4 is very likely catalytically inactive, or at least its function
is not dependent on the tyrosyl radical. Since Tyr131 is an invariant residue that is
responsible for generating the free radical in the R2 subunit, Wang et al (1997) replaced
the Tyr131 in RNR4 with Phe and found that RNR4-Y131F was still able to complement
41
the RNR4-?1 deletion allele. However, like RNR2, RNR4 is also essential for mitotic
viability as rnr4 null mutants are lethal. To make things more complicated,
overexpression of RNR2 cannot rescue the rnr4 mutant phenotype and vice verse,
indicating that the two small subunit genes are not functionally redundant. Huang and
Elledge (1997) speculated that RNR4 might play a structural or a regulatory
nonenzymatic role in RNR activity.
Chabes et al (2000) examined in detail the structure and biochemical properties of
the two yeast small subunits. They demonstrated that the crucial role of RNR4 is to fold
correctly and stabilize the radical-storing RNR2 by forming a stable 1:1 RNR2/RNR4
complex. No RNR activity was observed with RNR2 alone or with RNR4 alone in the
presence of RNR1, suggesting that RNR2/RNR4 heterodimer is the only active form.
SpeciesGeneSubunitCell cycle 
regulation
DNA damage 
inducibility
StructureSubcellular 
localization
Rnr1Large10-fold5-fold
?
2
???
majority in 
cytoplasm
Rnr2Small2-fold25-fold?????? majority in 
cytoplasm
Rnr3Large? 100-fold
Rnr4SmallNo yes majority in 
nucleus
RRM1Large4-fold9-fold
?
2
?
2
cytoplasm
RRM2Small4-fold13-fold
?
2
??
2
cytoplasm
p53R2Small yes nucleus
RNR1-1Large
RNR1-2Large3.3-fold
RNR2Small6.6-fold
R1Large
?
2
????
TSO2Smallyes
?
2
???      
?
2
?????    
?
2
?????
R2aSmall
R2bSmall
Arabidopsis
Table 1-3. Overview of  RNR genes in eukaryotes
S.cerevisiae
Human
Tobacco
42
Regulation of RNR genes
Considering the essential role of RNR in DNA synthesis, the multileveled and
complicated regulation of RNR expression and activity is not surprising. First, the RNR
activity is cell cycle-regulated. This could be executed at transcriptional (mRNA
induction), posttranscriptional (mRNA stability), translational (protein induction) and/or
posttranslational (protein degradation or subcellular localization) levels. Either one of the
two subunits or both subunits are under stringent regulation in different species. Second,
in addition to its role in DNA replication, RNR also provides dNTPs for DNA damage
repair. Therefore, RNR genes are regulated by DNA damage checkpoints.
Controlled degradation of RNR protein
In mammalian cells, transcription of the R1 and R2 genes is cell cycle-regulated
with maximal levels during S-phase. However, because of its long half-life (20 h), the R1
protein is in excess and is present constantly during all phases of the cell cycle (Kolberg
et al. 2004). In contrast, the R2 protein and its enzyme activity are present at high levels
during S-phase and at lower levels in G0 and G1. Although neither DNA damage nor
replication blocks promote the R2 gene transcription, they could delay the degradation of
the R2 protein and hence, increase the overall RNR activity. Interestingly, mammalian
cells evolved an additional R2 gene, called p53R2, which seems to be specifically
employed in response to DNA damage (see later).
What is the mechanism underlying the R2 protein degradation in mitosis? Chabes
and her colleagues (2003) found that mouse R2 protein contains a conserved KEN box at
its N-terminus and demonstrated that it is a new target for anaphase-promoting complex
43
(APC)-Cdh1-mediated proteolysis. R2 protein with mutated KEN box or without the
KEN box is not degraded in late mitosis. Since the p53R2 protein lacks the N-terminal
33-aa residues containing KEN box in R2 protein, it easily escapes the degradation
mediated by the APC-Cdh1 complex. Why is it so important for the cell to degrade the
R2 protein in late mitosis? As the R1 protein is present in all phases of the cell cycle, the
undegraded R2 protein in mitosis will keep RNR activity high, which may cause
unscheduled DNA replication. In addition, increasing evidence suggests that a
deregulated R2 protein might be a tumor progressor determinant (see later).  Thus
unwanted R2 protein must be removed from the cells.
Specified isoform of R2 for DNA repair
 In yeast, all four RNR genes can be induced by DNA damage outside S-phase.
Mammalian cells use a different strategy. As mentioned earlier, a special R2 gene
(p53R2) has evolved. Another distinction is that the p53R2 protein is nuclear localized
while all other R2s are cytoplasmic. This p53R2 is a direct target of p53 and can be
induced by p53 in response to various genotoxic stresses. Tanaka and his colleagues
(2000) speculated that there might be two independent pathways for supply of dNTPs,
one involving normal maintenance of dNTPs for DNA replication during S phase and the
other involving supplying dNTPs for urgent repair of DNA damage.  They found that
after DNA damage, the level of p53R2 protein is increased whereas the level of R2 is
decreased.  The RNR activity after DNA damage is exclusively dependent on p53R2 as
RNR activity can be abolished by p53R2 antisense interference (Yamaguchi et al. 2001).
Moreover, a point mutation in p53R2 was found in a cancer cell line HCT116, which
44
causes a substitution of a highly conserved valine into leucine. This mutation not only
abolished the RNR activity after DNA damage, but also caused remarkable apoptosis
after ?-irradiation. p53AIP1, a pivotal mediator of p53-dependent apoptosis, is induced in
this p53R2 mutant cell line. These results suggest that inactivation of p53R2 may lead to
accumulation of too much DNA damags that can not be repaired, which eventually
triggers the p53-dependent apoptotic pathway to eliminate those damaged cells.
Is the RNR activity still induced in p53 mutants where p53R2 is no longer active?
The answer is yes. While both p53R2 protein level and the p53R2 binding to hRRM1
(human R1 protein) is down-regulated in cells with mutant p53, hRRM2 (human R2)
protein level and its binding to hRRM1 was up-regulated. In addition, RNR activity is
still increased after UV treatment in cells with mutant p53, suggesting that hRRM2
complements p53R2 to form RNR holoenzyme and maintain RNR activity after UV
treatment when p53 is inactive (Zhou et al., 2003). Consistently, expression of antisense
hRRM2 in cells with mutant p53 led to decreased hRRM2 level and resulted in greater
sensitivity to UV than observed in cells with active p53. There might be a mechanism
that can only switch on hRRM2 expression when p53 is inactive. Lin et al. (2004) also
found that hRRM2 can be employed to supply dNTPs for repair of DNA damage in
HCT-116 cells (p53-/-) with impaired p53-dependent induction of p53R2.
Translocalization of RNR proteins in response to DNA damage
In yeast, RNR1 and RNR3 are predominantly localized to the cytoplasm whereas
RNR2 and RNR4 are predominantly present in the nucleus during the normal cell cycle.
45
However, under DNA damage and replication stress, RNR2 and RNR4 are redistributed
to the cytoplasm in a checkpoint Mec1/Rad53/Dun1-dependent manner (Yao et al, 2003).
In fission yeast, a small protein, Spd1, acts to anchor the R2 to the nucleus (Liu et al.
2003). In the presence of DNA damage, the checkpoint pathway inactivates Spd1 via the
COP9/signalosome. The removal of Spd1 resulted in the R2 staying in cytoplasm,
allowing the formation of holoenzyme with the R1 subunit. In spd1 mutants, the R2
protein will be always present in cytoplasm and remain active. In csn1 or csn2 mutants
(Csn1 and Csn2 genes encode two subunits of COP9/signalosome), Spd1 will not be
degraded, locking the R2 protein in nucleus. Since R1 is predominantly present in
cytoplasm, the compartmentalization of R2 to the nucleus results in the absence of or less
RNR holoenzyme activity. The fact that csn1mutants are sensitive to DNA damage and
spd1 mutants can suppress the DNA damage-sensitivity of csn1 suggests that the CSN-
dependent as well as the checkpoint-dependent relocalization of R2 is required for an
efficient DNA damage response.
This relocalization mechanism displays an opposite direction in different species.
Tobacco RNR1a can be relocated from cytoplasm into nuclei upon UV-C treatment
(Lincker et al., 2004). In mammalian cells, all three RNR subunits (hRRM1, hRRM2 and
p53R2) are located in cytoplasm. However, both p53R2 and hRRM2 physically bind to
p53 and hence, cannot bind to hRRM1. Upon UV-C treatment, the binding affinity
between p53 and the two small RNR subunits decreases. All three RNR subunits
translocate to the nucleus and the active tetrameric RNR complex forms (Fig. 1-12).
46
Control of RNR mRNA stability
Saitoh et al (2002) found that Cid13 encodes a cytoplasmic Poly(A) polymerase
regulating RNR mRNA in fission yeast. Cid13 shares a high level of sequence similarity
to Trf4/5 family of DNA polymeras. However, Cid13 is a cytoplasmic protein and lacks
detectable DNA polymerase activity. Overexpression of Cid13 rescues hydroxyurea (HU)
sensitivity of many checkpoint mutants (Saitoh et al., 2002). In addition, ?cid13 cells are
HU sensitive and have reduced dNTP pools, suggesting a role of Cid13 in controlling
RNR activity. The RNR2 gene in fission yeast, suc22 mRNA poly(A) tail was 10
    Figure 1-12. Translocation of human RNR subunits upon DNA damage.
hRRM1, the large subunit, is shown as M1 and hRRM2, the small subunit as M2.
Once DNA damage occurs, the signal transduction pathway leads to the
dissociation of p53 from hRRM2 and p53R2. Meanwhile, genes in nucleotide
excision repair are up-regulated and the repair proteins move into the nucleus,
recognize and bind to the damage sites. All three RNR subunits translocate into
the nucleus and form active holoenzyme, producing dNTPs for repair.
p53
M2
p53
M1
p53R2
M1M1
M1M1
M1M1
p53R2
p53R2p53R2
M2M2
M2
UV stress
Signal transduction
Nucleotide
excision repair
dNTP
NDP
dNDP
DNA repair
complex
UV damage
Nucleus
47
nucleotides shorter in ?cid13 cells than that in wild type. Cid13 associates with poly(A)
binding proteins in vivo and can polyadenylate mRNA in vitro. The onset of DNA
replication marks a point of rapid consumption of dNTPs. DNA damage or heat shock
may also strain the capacity to synthesize dNTPs. In these circumstances, suc22 mRNA
polyadenylation by Cid13 may provide a mechanism of rapidly enhancing dNTP
synthesis, bypassing steps of promoter activation, synthesis and processing of mRNAs,
and mRNA nuclear export. Such a capacity may be crucial for survival during times of
stress or to ensure optimal dNTP availability during DNA replication.
Protein inhibitor of RNR
Added to the already complex regulation of RNR is the discovery of a novel small
protein (104 aa), Sml1, as a negative regulator of dNTP pools by binding directly to yeast
RNR1. Sml1 was isolated as a suppressor of two essential checkpoint genes MEC1 and
RAD53. sml1 mutants exhibited increased levels of dNTPs (about 2.5-fold) without
altering RNR genes transcription. Chabes et al. (1999) reported that Sml1 binds to RNR1
but not to RNR2, in a 1:1 ratio. Zhao et al. (2000) further dissected the Sml1 protein by
randomly mutagenizing the Sml1 ORF. They identified seven mutations that did not
affect protein expression and the mutant protein could still suppress lethality caused by
mec1 and rad53. All seven mutations abolished the interaction between Sml1 and RNR1
or human R1. This demonstrates that the binding mechanism between Sml1 and the RNR
large subunit is conserved from yeast to human and the Sml1/R1 interaction is required
for Sml1 to inhibit R1. However, no homolog of Sml1 in other species so far has been
identified based on the sequence similarity.
48
In the model proposed by Zhao et al. (1998), Sml1 binds RNR1 to inhibit RNR
activity when DNA synthesis is not required. During S phase or after DNA damage, the
checkpoint pathway inactivates the Sml1 to allow RNR activation. Consistent with this
model, Sml1 protein levels fluctuate during the cell cycle, being lowest during the S
phase. The decrease of Sml1 levels in S phase and after DNA damage and HU treatment
is dependent on Mec1/Rad53 and other checkpoint genes. Later, it was found that Dun1,
a downstream kinase of Mec1/Rad53, genetically and physically interacts with Sml1 in
vivo and can phosphorylate Sml1 in vitro. Such posttranslational regulation of RNR can
be achieved in the absence of protein synthesis, resulting in a more rapid change in RNR
activity than that gained by transcriptional regulation.
Repression of RNR gene expression
During the screen for mutations that cause constitutive expression of RNR3 under
normal growth conditions, two large groups of CRT (constitutive RNR transcription)
genes were isolated (Zhou and Elledge, 1992). The first group includes CRT3/RNR4,
CRT5/POL1/CDC17, CRT6/RNR2, CRT7/RNR1, and CRT9/TMP1/CDC21. Mutations
in genes of this group activate RNR3 transcription by generating an endogenous DNA
damage signal. The second group includes CRT1, CRT4/TUP1, and CRT8/SSN6. SSN6
and TUP1 are two global repressors recruited to particular promoters by interacting with
target-specific DNA-binding proteins.
49
CRT1 encodes a homolog of the mammalian RFX family of DNA-binding
proteins. It recognizes a conserved DNA sequence of 13 nucleotides, called the X-box.
More than one X-box (with different strengths) was found in the promoters of RNR2,
RNR3 and RNR4 as well as CRT1 itself, which is required for the CRT1-mediated
repression in vivo. Among the different crt mutants, crt1, tup1 and ssn6 mutants exhibit
the strongest derepression of both RNR3 and RNR2 (from 50 to 200-fold) (Zhao and
Elledge, 1992). CTR1, TUP1 and SSN6 form a complex to mediate the repression of
RNR2 (Huang et al., 1998), which was supported by a pull-down assay. In the model
proposed by Huang et al. (1998) (Fig. 1-13), under normal conditions, CRT1 binds to the
X-boxes and recruits SSN6 and TUP1 co-repressors to the promoter of the target genes to
repress their transcription.  In response to replication blocks or DNA damage, CRT1
Tup1
Ssn6
Crt1
Crt1
X-box
X-box
CRT1, RNR, etc
Tup1
Ssn6
Crt1
X-box
CRT1, RNR, etc
DNA damage or
Replication blocks
Mec1-Rad53-Dun1
Pi
PiPi
Pi
Pi
RNR and other
repair proteins
Crt1 protein
Repair
Pi
Recovery
Repressed
Induced
Repressed
CRT1, RNR, etc
    Figure 1-13. A model illustrating CRT1 mediated repression
of RNR genes.(cited from Huang et al., 1998)
50
becomes phosphorylated by Mec1-Rad53-Dun1 kinase cascade and loses its ability to
bind X-boxes. This leads to transcriptional induction of its target genes including RNR as
well as CRT1 itself. Accordingly, overexpressiong of CRT1 inhibits RNR induction. As
DNA damage is repaired or replication blocks are removed, the kinase cascade becomes
inactive. The increasing level of CRT1 will bind to X-boxes again, resulting in rapid
restoration of the repressed state.
dNTP pool as a signal
In yeast, Mec1 and Rad53 are responsible for the transcriptional and cell cycle
response to both DNA damage and replication blocks, thus either mec1 or rad53 null
mutants are lethal. To one?s surprise, overexpression of RNR1 or RNR3 (but not RNR2 or
RNR4) can suppress ?rad53 and ?mec1 lethality. Desany et al. (1998) found that even a
twofold increase in RNR1 gene dosage is sufficient for suppression. As overexpression of
RNR1 has no effect on the timing of S-phase completion or the overall rate of DNA
synthesis, it suggests that the essential function of Rad53 and Mec1 is to maintain an
adequate dNTPs supply. The fact that low amounts of exogenously supplied RNR1 can
efficiently suppress lethality suggests that the defect responsible for lethality is just below
the threshold for survival.
A nucleotide sensor must be activated to prevent replication in response to the
low dNTP pools. Inhibition of replication fork progression by low dNTP pools leads to
prolonged S phase associated with increased chromosome breakage (Windle et al., 1991;
Di Leonardo et al. 1993), which may be the initial lesions required for gene
51
amplification, a form of genetic instability (Ma et al. 1993; Kuo et al. 1994). It was
reported that synthesis of DNA from limiting levels of dNTPs could lead to accumulation
of short DNA fragments and chromosome breakages, and cell death after a few divisions
(Reichard, 1988). However, the nature of the nucleotide sensor remains unknown.
Although limiting levels of dNTPs could be error-prone for DNA synthesis,
overproduced dNTPs are also detrimental to DNA synthesis. Chabes et al. (2003)
reported that an increased dNTP concentration leads to a higher mutation rate in yeast
but, at the same time, helps yeast cells to survive DNA damage. They speculate that
excessive dNTP concentrations allow more efficient bypass of DNA lesions by DNA
polymerases, which is beneficial to individual cells whose survival needs surpass the risk
of higher mutation rate. However, in multi-cellular higher eukaryotes, accumulation of
mutations will eventually lead to disasters. This survival mechanism in yeast may find no
place in higher eukaryotes.
  
Chemotherapy
RNR has long been the target site for chemotherapy for cancer due to its essential
role in DNA synthesis. Hydroxyurea is a well-known therapeutic agent produced by
Bristol-Myers Squibb as Hydrea and by Roxane as Hydroxyurea Capsule, which is a
specific but reversible inhibitor of the small subunit R2.  In addition, the species-specific
interaction between the large subunit and the small subunit also provides a good strategy
to develop a peptidomimetic drug that competes specifically with the binding of the viral
R2 to R1.
52
Recently, several studies provided renewed interest in targeting RNR in the
development of anticancer therapeutics. It was found that RNR activity, and the level of
R2, are markedly altered in malignant cells exposed to the growth factor TGF-? in rodent
and human tumor cells. These observations suggest that RNR, and particularly the rate-
limiting R2 subunit, may be critically involved in mechanisms controlling malignant
progression.  Fan et al (1996) reported that expression of R2 in benign BALB/c 3T3 and
NIH 3T3 cells leads to a greatly increased frequency of focus formation in cooperation
with H-ras transformation. Moreover, stable expression of R2 leads to significant
increases in the membrane-associated Raf-1 protein and mitogen-activating protein
kinase-2 (MAPK-2) activity. Since deregulated R2 expression causes no obvious
differences in growth rates or cell cycle-phase distributions, their results demonstrated
that the R2 protein can participate in other critical cellular functions in addition to
ribonucleotide reduction and that deregulated R2 is a novel tumor progressor
determinant.
What about deregulation of R1, the large subunit? Surprisingly, stable expression
of R1, with or without R2 expression, led to suppression of tumorgenic and/or metastatic
potential (Fan et al, 1997), suggesting that R1 and R2 play opposite roles in determining
malignancy and a balance in the levels of R1 and R2 might be a very important control
point.
As a new strategy, Lee et al. (2003) screened and identified a 20-mer
oligonucleotide complementary to the coding region of R2, GTI-2040. They showed that
53
GTI-2040 specifically inhibits expression of R2 mRNA in a number of cell lines. The
antisense oligo also displays sequence-specific and dose-dependent antitumor activity in
vivo with superior efficacy compared with known RNR-based therapeutic agents, such as
5-fluorouracil (FU), gemcitabine and vinblastine.
Eklund et al. (2001) discussed the possibility of designing specific inhibitors to
normal RNR, especially the R2, based on the fact that human cells have an additional
p53R2 for DNA repair. Since cancers often have mutations in the p53 pathway, they are
presumably unable to make p53R2. Therefore, cancer cells would die if the normal R2
were inhibited, whereas normal cells with p53 could still survive on the dNTPs supplied
by p53R2. The discovery that p53R2 is 158-fold more susceptible to the iron chelator
deferoxamine mesylate than the human R2 and 2.5-fold less sensitive to hyroxyurea than
R2 (Shao et al. 2004) makes this strategy very attractive. Unfortunately, the real picture is
always more complicated as an additional p53-independent induction of p53R2 might
exist. Kolber et al. (2004) found that in some breast cancer cell lines with mutations in
p53, p53R2 is present throughout the cell cycle in relatively high amounts.
54
Chapter two
Functional Analyses of Arabidopsis Ribonucleotide Reductase
Small Subunit Gene Family
Abstract
Ribonucleotide reductase (RNR), comprising two large subunits (R1) and two
small subunits (R2), catalyzes a rate-limiting step in the production of
deoxyribonucleotides needed for DNA synthesis.  Previous studies in yeast and mammals
indicated that defective RNR often led to cell cycle arrest, growth retardation and p53-
dependent apoptosis.  Given that plants are constantly exposed to environmental
mutagens and plant cells are totipotent, an understanding of RNR function in plants is of
great importance.  We isolated and characterized mutations in all three R2 genes (TSO2,
R2A and R2B) in Arabidopsis.  We showed that while neither r2a, r2b nor r2a r2b double
mutants display any visible phenotype, tso2 mutants, with reduced dNTP levels,
exhibited developmental defects including callus-like floral organs and fasciated shoot
apical meristem. The TSO2 gene is cell cycle-regulated and can be induced by HU and
DNA damaging agents such as MMS. tso2 mutants were more sensitive to HU and UV-C
and  tso2 r2a seedlings exhibited increased DNA damage and massive programmed cell
death. The release of transcriptional gene silencing observed in tso2 r2a mutants supports
a link between defective DNA replication and epigenetic stability.  Our data suggests that
plants can initiate programmed cell death upon DNA damage despite the absence of p53
in its genome.
55
Introduction
Fundamental to all life forms is the ability to replicate and repair DNA.
Ribonucleotide reductase (RNR), comprising two large subunits (R1) and two small
subunits (R2), catalyzes a rate-limiting step in the production of dNDP needed for DNA
synthesis (Elledge et al., 1992).  The R1 subunit binds the nucleoside diphosphate
substrates and allosteric effectors and the R2 subunit contains the di-iron tyrosyl radical
cofactor essential for the reduction of NDP to dNDP (Elledge et al., 1992). RNR
enzymatic activity can be efficiently inhibited by hydroxyurea (HU), a chemical used to
block DNA synthesis and cell cycle progression (Reichard, 1988). The enzyme activity
and the mRNA expression of the RNR genes are cell cycle-regulated in different species
with maximal values in the S phase (Chaboute et al., 1998; Elledge et al., 1992;
Reichard, 1988). In addition to regulation by the cell cycle, RNR genes are induced by
DNA damage when DNA repair is required outside S-phase. It has been shown that
aberrant expression of R2 directly alters the malignant potential of tumor cells (Amara et
al., 1996). A failure to control the size of dNTP pools and /or their relative amounts leads
to cell death or genetic abnormalities (Reichard, 1988).
Genetic analyses of rnr mutants in yeast and mammals indicated that defective
RNR often led to cell cycle arrest, growth retardation and p53-depedent apoptosis
(Elledge et al., 1992; Kimura et al., 2003). Defective feedback regulation of RNR led to
elevate dNTP levels resulting in increased mutation rates but enhanced resistance to
mutagens (Chabes et al., 2003).  These genetic studies demonstrated that RNR plays a
key role in maintaining genomic stability by providing proper dNTP pools.
56
RNR is one of the best-studied transcriptional targets of the DNA damage
checkpoint pathway in yeast and mammals.  In mammals, p53R2, a RNR small subunit,
is a direct target of the tumor suppressor protein p53 (Tanaka et al., 2000).  In yeast,
which does not have p53, the checkpoint pathway induces RNR via several mechanisms
including alleviation of the Crt1-mediated transcriptional repression (Huang et al., 1998),
phosphorylation-mediated removal of an inhibitory protein Sml1, (Zhao et al., 2001;
Zhao et al., 1998), and redistribution of RNR2 and RNR4 from nucleus to cytoplasm
(Yao et al., 2003).  The existence of multiple mechanisms regulating RNR activities
underscores the importance of proper dNTP pool for the fitness and survival of an
organism.
In contrast to yeast and mammals, our knowledge of RNR function in plants is
limited. R1 and R2 genes have been previously isolated from tobacco and Arabidopsis
(Chaboute et al., 1998; Philipps et al., 1995) and their S-phase specific expression was
shown to depend on the E2F-like motifs in their promoters (Chaboute et al., 2002;
Chaboute et al., 2000).  Transient translocation of a GFP-R1 protein from cytoplasm to
nucleus in response to UV irradiation was recently reported (Lincker et al., 2004).
However, no mutation in any of the plant RNR genes has been reported.  The function of
RNR in maintaining plant genomic stability and genetic variability remains unknown.
Here we report the first functional dissection of RNR in higher plants.
Arabidopsis TSO1 and TSO2 genes were previously identified in the same genetic screen
by mutations that cause the formation of callus-like floral organs.  These two loci were
57
named TSO, meaning ugly in Chinese (Liu et al., 1997).  TSO1 was shown to encode a
novel nuclear protein with two cysteine-rich repeats and tso1 mutants exhibited defects in
cell division (Liu et al., 1997; Song et al., 2000).  The molecular isolation of TSO2
reported here revealed a direct role of TSO2 in the cell division as TSO2 encodes one of
the three R2 genes in Arabidopsis.  Subsequent identification and characterization of
mutations in all three R2 genes in the Arabidopsis genome provided the first insights into
RNR function for plant growth and response to environmental mutagens.
Results
 tso2-1 is a pleiotropic mutant
Four tso2 mutants (tso2-1, 2, 3, 4) were isolated from EMS mutagenesis and they
all are recessive and phenotypically similar. Therefore, tso2-1 was chosen for further
analysis. In tso2-1 mutants, the morphology of roots and the first four rosette leaves are
normal. Subsequently, tso2 mutants display abnormal leaf and floral morphology (Fig. 2-
1A,B), including white sectors in green organs (Fig. 2-1C, D), uneven thickness, rough
surface, and irregular margins of leaves or floral organs. Very often, the sepals are
wrinkled and have irregular margins (Fig. 2-1E, F). The siliques of tso2 mutants are
shorter, wrinkled and have white patches or sectors (Fig. 2-7J). In addition, embryo
development inside the siliques exhibits a high frequency of abortion (Fig. 2-1J-M).  In
the wild type, normal embryos turn green (usually five days after pollination). However,
in tso2-1 mutants, white aborted embryos (Fig. 2-1K) or brown wrinkled, abnormal
embryos (Fig. 2-1M) are often observed. We counted forty siliques (at late stage) from
58
both wild type and tso2-1 mutants. While none of the wild type siliques contains white
embryos, all tso2 siliques contain white aborted embryos ranging from 5% to 60%.
Occasionally, tso2-1 mutant stems broaden and flatten and split into multiple
shoots. This fasciation occurs both in reproductive stage and vegetative development
stages (Fig. 2-1H and I). Sometimes, the main inflorescence remains arrested, even when
the cauline leaves are fully developed. Occasionally the main inflorescence meristem is
arrested but the secondary shoots develop normally (Fig. 2-1H, indicated by arrow).
    Figure 2-1.  Phenotype of tso2 mutants. (A) A wild type (ler) plant. (B) tso2-1
mutant. (C) A wild type inflorescence. (D) tso2-1 inflorescence. Note the
white/green leaves start to exhibit abnormal morphology such as the irregular
marginal white tissues (arrow). (E) A wild type flower. (F) A tso2-1 flower. Note the
wrinkled sepals and sepals with white margins. (G) A wild type shoot.  (H) A tso2-1
shoot showing arrested shoot apex (indicated by an arrow) and bifurcation. (I) A
fasciated tso2-1 plant, showing the multiple bifurcation. (J) A wild type silique at
late developmental stage. Note the mature seeds turning green. (K) A tso2-1 silique.
Note the seeds at different stages. (L) A tso2-1 silique. Two seeds are arrested at
very early stage (indicated by arrows). (M) seeds removed from one tso2-1 silique.
Note the brown defective seeds (indicated by an arrow).
59
The wild type shoot apical meristem (SAM) is a dome-like structure with floral
primordia on the periphery (Fig. 2-2A).  In tso2-1 mutants, occasionally the SAM splits
into multiple apical meristems (Fig. 2-2B). Sometimes the fasciated shoots are flattened
(Fig. 2-2C). In wild type, the sepal epidermal cells are well-organized in the same plane
while in tso2-1 mutants, sepal epidermal cells are disorganized with protruding cell
clusters (Fig. 2-2F, G). In addition, the sepals very often display irregularly serrated
margins (Fig. 2-2H). Occasionally, petals are also serrated (Fig. 2-2K). Sometimes, we
observed asynchronized development for petals (Fig. 2-2J) and unfused or twisted carpels
    Figure 2-2. Scanning electronic microscopic analysis of tso2 mutants. (A)
wild type (Ler), showing the inflorescence apex.  (B, C) tso2-1 mutants showing
the fasciated shoot apex. (D) A flower of tso2 mutants. Two sepals and petals
are removed to show a twin anther. (E) Wild type sepal epidermis.  Immature
stomata (arrow) are interspersed with long cells.  (F) tso2-1 sepal epidermis.
Clusters of epidermal cells are projected above the surface.  (G) A close-up of
sepal epidermal cells showing enlarged cells. (H) A tso2-1 flower, showing the
unfused capels. (I) A wild type flower. (J) A tso2-1 flower. Three petals are at
different developmental stages. (K) A serrated petal in tso2 flowers. (L) Partial
homeotic transformation in tso2-1 flowers. Note the stigmatic-like tissues
(indicated by an arrow) on top of an anther.
60
(Fig. 2-2H, L). Partial homeotic transformation of stamens into carpels is also observed:
on the top of the anther, stigmatic-like tissues are formed (Fig. 2-2L).
Chloroplast development is defective in tso2-1
The appearance of white sectors in green organs could be due to a lack of
chloroplasts in mesophyll cells or due to a lack of mesophyll cells. Histological analysis
of leaf and silique tissues was performed in wild type and tso2-1. Occasionally, one valve
of the silique is completely white in tso2-1, which occurred more often in tso2-1 r2b/+
double mutants (see later, Fig.2-7K). The siliques were cut in half and the white and
green sectors were fixed separately. In wild type and in the green sectors of tso2-1
A B C
D E F
    Figure 2-3. 2 ?m cros section f wild type and tso2-1 mutant leaves
(top anel) and silques (botm panel). (A) wil type. Mesophyl cells
are well organized and contain chloroplasts (stained dark purple). (B, C)
tso2-1. Note the unsual arge intracelular ir space and isorganized
mesophyll cell layers indicated by the red bracket. (D) wild type silique.
(E, F) tso2-1 silque. Subepidermal cels in the white sector egion
(shown i Figure 2-7K) contain fewr or o clorplast. The white
sector region also has larger intracellular air space (indicated by yellow
arrows). Note some extremely smaller cells (indicated by red arrows).
61
siliques, the subepidermal cells contain a large number of chloroplasts outlining the cell
wall, which are stained in dark purple. In contrast, the majority of the subepidermal cells
in tso2-1 white siliques do not contain chloroplasts. Therefore, the white sector is caused
partly by a lack of chloroplasts in mesophyll cells (Fig.2-3E) and partly by abnormal
airspaces, indicating a lack of mesophyll cells (Fig. 2-3F). In addition, a large number of
small cells are present and unevenly distributed (Fig.2-3E and F). The uneven thickness
and rough surface of the leaves, floral organs and siliques are probably caused by
unevenly distributed cell layers and unusually large intracellular air space.
Molecular cloning of the TSO2 gene
A mapping population of 358 tso2 plants selected from F2 generation of a cross
between tso2-1 (in Ler background) and sup-5 (in Col background) was used to map the
TSO2 gene. TSO2 was gradually mapped to a 20-kb region (from 32-53kb) located within
one BAC clone, MOJ10 (Fig.2-4). A cosmid library was constructed from this BAC and
overlapping cosmids spanning this 20-kb region were identified. Two overlapping
cosmids (D and G) were transformed into tso2 mutants. Cosmid D rescued the tso2
mutants while cosmid G failed to rescue the mutant.
Cosmid D covers the 39-62 kb region of MOJ10 (Fig.2-4). Two recombinants at
53-kb region indicated that tso2 is within this 39-53 kb region of MOJ10, where four
genes reside. All four genes have been sequenced in tso2 alleles. Only the RNR
(ribonucleotide reductase) small subunit gene (at3g27060) was found to have a mutation
in all four tso2 alleles. The invariant aspartic acid (D) at residue 49, which is involved in
62
binding to the large subunit of ribonucleotide reductase (Philipps, et al., 1995), is
changed to asparagine (N) in both tso2-1 and tso2-4 alleles. In tso2-2, Glycine 170 is
substituted by serine while arginine 97 is replaced with cysteine in tso2-3 (Fig.2-5).
Arabidopsis genome has three R2 genes
Since tso2 mutants are all viable and fertile, other redundant genes in the
Arabidopsis genome may compensate for tso2.  A close examination of the Arabidopsis
genome revealed two additional R2 genes named R2A (At3g23580) and R2B
(At5g40942) (Fig. 2-5). R2A was previously isolated by Philipps et al. (1995), while R2B
was annotated first as a pseudogene and recently as a truncated gene (at5g40942) by the
Arabidopsis genome sequence project (http://www.Arabidopsis.org). The R2B from the
Columbia (Col) accession, which was used for genome sequencing, possesses a 2 bp
29/7162/7167/71630/71660/716
Recombinants
SUPERMAN
BACs
Cosmid
900010600
11400
1200013200kb
GAPA
GL1
MXE2MQP15
MOJ10
0 22314553
60
 73 kb
2 11 022
Recombinants
G
D
TSO2/R2b
Predicted genes
Physical postion
Markers
2
Figure 2-4. Schematic diagram of map-based cloning of the TSO2 gene.
63
deletion resulting in a frameshift and a subsequent stop (Fig. 2-5).  In contrast, R2B from
Ler and Wassilewskija (WS) accessions do not harbor the 2 bp deletion and differ from
Col in 13 additional nucleotides, five of which change amino acids (Table 2-1).
As shown in Fig.2-5, the gene structure of R2B is similar to TSO2. While TSO2
has one intron, R2B has no introns. Since R2B and TSO2 share 82% identity to each other
at the nucleotide level, it suggests that they might be derived from gene duplication. In
contrast, R2A only shares 70% identity to TSO2 at amino acid level and almost no
similarity at nucleotide level. R2A has a distinct gene structure with eight introns,
indicating that R2A has a different origin from TSO2 and R2B.
Nucleotide change (Ler-Col)
a
Position
b
Nature of the mutation
A-T 393 silent
T-C 411 silent
TG-deletion in Col418-419frameshift
G-A 438 silent
C-A 573 silent
C-A 639 silent
C-T 661 Leu-Phe
C-T 726 silent
C-T 736 Pro-Ser
C-G 824 Ala-Gly
G-C 842 Ser-Thr
A-G 879 silent
A-G 931 Ile-Val
Table 2-1. Polymorphisms in the R2B gene between Col and Ler
a
No polymorphism is found between Ler and WS
b
The start codon ATG of the Ler R2B gene is designated +1.
64
All RNR proteins are localized in cytoplasm with exception of human p53R2.
None of the three Arabidopsis R2 proteins has a nuclear localization signal. Intriguingly,
TSO2 and R2B proteins have a potential PEST sequence (MPSMPEEPLLTPTPD) at the
N-terminus, which acts as a protein degradation signal. In addition, three E2F binding
sites (consensus sequence TTTG/CG/CCGC) are found in the promoter of the TSO2 gene
(Table 2-2). There is one E2F binding site in R2B promoter but not in R2A. E2Fs are
transcription factors that are essential for G1/S transition (see Chapter one), regulating a
large number of genes such as cyclin A, cyclin E, pRb, p34
cdc2
, DNA polymerase ?,
ORC1, PCNA, DHFR, TK, RNR etc (Chaboute et al., 2000, 2002; Lavia and Jansen-
Durr, 1999). Interestingly, all the E2F binding sites shown in Table 2-2 are located within
?700 in the promoters of RNR genes, consistent with a genome-wide survey that the
ATG
1
TGA
340
r2a-1 W
193
?stop
r2b-c  frameshift ? 142-333
ATGTAG
1 333
ATG
TAA
1 332
tso2-1,4 D
49 
?
 
N
tso2-3 R
97
  C
tso2-5 ?30-332
tso2-2 G
170
? S
R2a
R2b
TSO2
tso2-6 ?232-332
    Figure 2-5. Gene structures of the three R2 genes in Arabidopsis. Exons are
shown as rectangles. The lines indicate introns and UTRs.Triangles indicate the
positions of T-DNA insertion.
R2A
R2B
65
majority of E2F binding sites are located within 1 kb promoter region starting from the
ATG (Ramirez-Parra et al., 2003)
Genetic interaction among three R2 genes
In order to further dissect the functions of the three R2 genes, we isolated an
EMS-induced r2a mutation (r2a-1) in the Col accession using a reverse genetics method
known as TILLING (McCallum et al., 2000). r2a-1 is a nonsense mutation that deletes
one third of the protein (Fig. 2-5).  In the Col background, r2a-1 is actually a r2a-1 r2b-1
double mutant, and r2a-1 r2b-1 plants are phenotypically wild type (Fig. 2-6A).
Aided by PCR-based markers specific to each mutant allele, the F2 progeny of a
cross between tso2-1 (Ler) and r2a-1 r2b-1 (Col) was analyzed.  tso2-1 r2a-1 and tso2-1
r2b-1 double mutants were identified, which exhibited a much stronger phenotype than
either parent.  First, tso2-1 r2a-1 double mutant seedlings did not develop beyond the two
RNR genePosition (ATG as +1)Binding site
TSO2 -307 CCGCCAAA
-329 CCGGCAAA
-469 TTTGGCGC
R2B -666 TTTGGCGC
R1 -228 CCGGGAAA
-232 TTTCCCGG
-251 GCGGGAAA
Rice R2A -197 GCGCGAAA
-262 TTTCCCGG
-314 CCGCGAAA
Rice R2B -168 TTTGGCGG
Tobacco RNR1b-177 GCGGGAAA
Tobacco R2 -294 GCGGCAAA
-355 TTTCCCGC
Table 2-2. E2F binding sites in the promoters of various RNR genes
Note: cannonical binding sites are shown in red. 
66
to four-leaf stage (Fig. 2-6C).  Their SAM appeared arrested (Fig. 2-6D), and their leaves
exhibited massively disorganized surfaces  (Fig. 2-6E, F).  Second, tso2-1 r2b-1 double
mutants were embryo lethal.  PCR-based genotyping detected tso2-1 r2b-1 double
mutants only among aborted seeds, but not among viable plants from tso2-1/tso2-1; r2b-
1/+ parents (Table 2-3). Finally, tso2-1/tso2-1 plants heterozygous for r2a-1, r2b-1, or
r2a-1 r2b-1 were viable but exhibited a much stronger phenotype than tso2-1/tso2-1 as
shown by their smaller statures, more severely reduced fertility, more frequent stem
fasciation, and larger white sectors (Fig. 2-6G-I). These results indicate that R2A and R2B
become essential in the tso2-1 background.
A
BC D
E F G H
I
    Figure 2-6. Genetic interaction among the three R2 genes. (a) r2a-1 r2b-1
seedling.  (B) A tso2-1 seedling. White tissues are forming in the emerging
leaves.  (C) typical tso2-1 r2a-1 double mutant seedling, which will not
develop further.  (D) The SAM of a tso2-1 r2a-1 seedling that is flat and
appears arrested.  (E) A wild-type leaf.  (F) A tso2-1 r2a-1 leaf showing
severely disrupted leaf surface.  (G) 6-week old tso2-1/tso2-1 plants singly or
doubly heterozygous for r2a-1 or/and r2b-1.  (1) tso2-1/tso2-1; r2a-1/+; r2b-
1/+. (2) tso2-1/tso2-1; r2a-1/+. (3) tso2-1/tso2-1; r2b-1/+. (4) tso2-1/tso2-1.
(H) tso2-1 siliques showing small white sectors in the green carpel valve.  (I)
tso2-1/tso2-1; r2b-1/+ siliques showing larger white sectors and smaller
siliques. Bar: 100 ?m in D; 1mm in E, F.
67
tso2-1 is sensitive to hydroxyurea (HU)
 Hydorxyurea (HU) has been shown to be the specific inhibitor of the R2 subunit
(acting as the radical scavenger) and yeast rnr mutants are hypersensitive to HU
treatment. Does it hold true for r2 mutants in Arabidopsis? To this end, 5-day-old
r2b genotyper2b-1/r2b-1r2b-1/++/+Total
Expected9 18 9 36
Observed0 22 14 36
Green0 13 5 18
White2 33 3 38
b
: 36 tso2-1/tso2-1 plants were PCR-genotyped for the r2b locus.
c
: 18 normal (green) and 38 abnormal (white) seeds were genotyped.
No. of Plants
b
No. of seeds
c
Table 2-3. Genotyping among the progeny of tso2-1/tso2-1 r2b-1/+ parent
a
a
: Only the genotype at R2B is shown as all plants are tso2-1/tso2-1.
Ler
Ler
tso2-1
tso2-1
HU (mM)00.20.5125
    Figure 2-7. tso2-1 is hypersensitive to HU treatment. Wild type (Ler) and
tso2-1 plants were grown in duplicate in the same plate and the experiment
was repeated three times with similar results.
68
seedlings of r2 mutants (tso2-1, r2a r2b, r2b) and wild type (Ler) were exposed to
different concentrations of HU and the sensitivity was assayed one week later. As shown
in Fig. 2-7, tso2-1 is sensitive to HU treatment at a concentration 10-fold lower than the
concentration wild type is sensitive to. Wild type seedlings are still viable in the presence
of 2 mM HU while tso2-1 seedlings cannot survive under 0.2 mM HU. However, neither
r2b single mutant (Col) nor r2a r2b double mutants showed increased sensitivity to HU
(data not shown), suggesting that RNR activity in these mutants is similar to wild type
and TSO2 alone is sufficient to confer wild type-like resistance to HU.
dNTP pool size is reduced in tso2-1 mutants
Although the TSO2 gene encodes the small subunit of RNR, it is based on its
sequence homolog from other species. Are the developmental defects indeed caused by a
defective R2? Could the other R2 genes (R2A and R2B) maintain the RNR activity to near
wild type levels? Two established methods are currently used to measure RNR activity:
one utilizes radioactively labeled substrate (normally CDP) and monitors product (dCDP)
formation. The other monitors NADPH consumption. We measured the dNTP pool sizes
directly in different r2 mutants using DNA polymerase assay with inflorescence tissues
since impaired RNR activity will lead to reduction of dNTP levels.
Whereas wild type (Ler), r2b and r2a r2b double mutants showed similar dNTP
levels (data not shown), tso2-1 mutants contain significantly reduced dNTP pools (Fig. 2-
8). In all eukaryotic cells, dTTP is always the most abundant, and dGTP the least. The
relative levels of dATP and dCTP vary in different species (Mathews and Ji, 1992). Since
69
dCTP is reduced to only 35% of wild type, and dGTP drops to 60% of wild type, this
different degree of reduction could lead to an imbalanced dNTP pool in tso2-1. The
reduction of dNTP pool sizes in tso2-1 mutants, but not in r2b and r2a r2b mutants, is
consistent with the observation that only tso2-1 is hypersensitive to HU treatment and
exhibits developmental defects.
The mRNA expression patterns of TSO2, R2A and R2B
As mentioned earlier, the first four rosette leaves of tso2-1 do not exhibit any
visible defects. Could this be due to the tissue specific or developmental stage-specific
expression of TSO2 or other R2 genes? TSO2, R2A and R2B transcripts were examined in
different Arabidopsis tissues (Fig. 2-9A).  TSO2 transcripts were found in roots, rosette
and cauline leaves, stems and flowers.  While R2A transcripts were not detected in roots
they were detected in rosette and cauline leaves, stems and flowers.  In contrast, R2B
dATPdCTPdGTPdTTP
0.0%
10.0%
20.0%
30.0%
40.0%
50.0%
60.0%
70.0%
80.0%
90.0%
100.0%
1
Relative dNTP level 
Ler
tso2-1
    Figure 2-8.  Relative dNTP level in Ler and tso2-1 mutants. The level of dTTP
(which is the most abundant in eukaryotic cells) in Ler is designated as 100%, and
the level of other three dNTPs was adjusted accordingly. The measurement was
repeated three times with different batches of tissues.The level was normalized to
an internal control, ADH activity (see Methods and Materials)
70
transcripts were only detected by Southern blots of RT-PCR products and were present in
all tissues tested (Fig. 2-9A in R2b.S panel).  Thus, all three genes, in most cases, are
expressed widely with R2B expression at a much lower level than TSO2 and R2A.
RNR genes are cell cycle-regulated at the transcriptional level in most species. To
examine TSO2 transcription during the cell cycle, the ?-glucuronidase (GUS) reporter
gene was fused to the TSO2 1.2kb promoter.  Transgenic seedlings harboring the
pTSO2::GUS were treated with aphidicolin or colchicine, which arrest cells at S and M
phase, respectively.  GUS expression was detected in more cells and at higher levels
when the seedlings were arrested at the S phase (Fig. 2-9C).  In contrast, GUS expression
    Figure 2-9. mRNA expression of TSO2, R2A and R2B.  (A) Expression of
TSO2, R2A and R2B mRNA. RT: root; RL: rosette leaves; CL: cauline leaves;
ST: stems; and FL: flowers.  The RT-PCR products of R2B were only detected
by Southern hybridization with a R2B-specific probe shown in lane R2B.S.
Actin2 is the loading control. (B) pTSO2::GUS expression in a 3-day old wild
type root. (C) pTSO2::GUS expression in a 3-day old root treated with
aphidicolin, (D)  expression in a 3-day old root treated with
colchicine. (E) in situ hybridization showing TSO2 mRNA expression in wild-
type SAM. (F) in situ hybridization sporadic TSO2 expression in
developing ovules (Ov).  (G) pTSO2:GUS reporter gene expression in transgenic
plants showing intense GUS activity in primary and lateral root meristems
(arrows).  (H) A 1-week old seedling showing GUS activity in shoot apex, young
leaves, and vasculatures.
71
was dramatically reduced when the seedlings were arrested at the M phase (Fig. 2-9D).
Hence, TSO2 transcription occurs predominantly in the S-phase.  
TSO2 mRNA distribution during plant development was examined by in situ
hybridization and by pTSO2::GUS.  The sporadic rather than uniform pattern of TSO2
mRNA in developing floral tissues (Fig. 2-9E, F) is characteristic of cell cycle phase-
specific expression.   In pTSO2::GUS transgenic plants, GUS expression is
predominantly present in tissues and organs with active cell division activities, including
young root tips, young leaves, and developing vasculatures (Fig. 2-9G, H).  The TSO2
expression pattern is consistent with its role in dNTP biosynthesis during DNA
replication.
DNA damage induction of the RNR genes
Yeast RNR genes can be induced by various DNA damage agents. In contrast,
mammalian RNR genes are specifically induced only by UV-C but not HU or methyl
methanesulfonate (MMS). Do plant RNR genes show a similar response to DNA damage
induction to animals or yeast? We treated Col cell suspensions with HU, MMS and 4-
nitroquinoline-1-oxide (4-NQO) for 2, 12 and 24 hrs. Total RNA was extracted at each
time spot and RT-PCR was performed with a small number of cycles to avoid saturation.
PCR products were Southern blotted and hybridized with gene-specific primers (see
Materials and Methods). As shown in Fig. 2-10, all four RNR genes can be induced by
HU. TSO2 is induced about 20-fold, R2B 13-fold, R1 5-fold and R2A 2-fold. However,
both R2A and R1 genes are not induced by MMS and 4-NQO whereas TSO2 and R2b
were slightly induced (about 2-fold).
72
0
5
10
15
20
25
22448
Time after HU treatment (hr)
Relative mRNA level
TSO2
R2a
R2b
R1
0
0.5
1
1.5
2
2.5
22448
Time after MMS treatment (hr)
Relative mRNA level
TSO2
R2a
R2b
R1
0
0.5
1
1.5
2
2.5
22448
Time after 4-NQO treatment (hr)
Relative 
mRNA 
level
TSO2
R2a
R2b
R1
480224
HUMMS4-NQO
022402244848
TSO2
R2a
R2b
R1
ACTIN2
A
B C
D
 Figure 2-10. DNA damage induction of RNR genes. (A) semi-quantitative RT-PCR.
RT-PCR was performed with 12-20 cycles. The PCR products were Southern blotted
and hybridized with gene-specific probes. (B-D) Normalization of mRNA expression
level under HU treatment (B), MMS treatment (C) and 4-NQO treatment (D). The
mRNA levels of each RNR gene at different time points were compared with the
level without treatment, which is defined as 1.
73
tso2-1 exhibits defects in cell cycle progression
Since yeast rnr mutants display a cdc phenotype, we tested if cell cycle
progression is affected in tso2-1 mutants. A cyclin B1-GUS chimeric protein driven by
the Cyclin B1;1 promoter (pcycB1::GUS) was introduced into tso2-1 mutants.  The
reporter expression is detected only in the late G2/early M phases of the cell cycle in root
tips (Colon-Carmona et al., 1999).  While the GUS expression was normally not
detectable in the above ground tissues of wild type plants (Fig. 2-11A, B), GUS was
expressed strongly and in a sporadic fashion in tso2-1 SAM and developing leaf and
floral organs (Fig. 2-11C, D).  The difference in pcycB1::GUS expression between wild
type and tso2-1 pcycB1::GUS plants was reproducible, even when the wild type and tso2-
1 plants were siblings harboring identical pcycB1::GUS transgenes and when wild type
and tso2-1 sibling plants showed similar GUS expression in roots.  Therefore, tso2-1
 Figure 2-11.  Cell cycle arrest in
tso2-1 mutants indicated by
pcycB1::GUS expression. (A)
 expression is not
detectable in wild type SAM.
(B) pcycB1::GUS expression is
not detectable in wild type
flowers showing sepals (s) and
stamens (st). (C) pcycB1:: GUS
is highly expressed in patches of
cells in the tso2-1 SAM and
emerging leaf primordia,
suggesting an increased number
of cells at late G2 or early M
phase. (D) pcycB1::GUS
expression is found in cells of
tso2-1 floral organs including
sepals (s), and stamens (st).
74
above ground tissues may possess a significant number of cells arrested at the M/G2-
phases of the cell cycle, which correlated with abnormal development in these tissues.
tso2-1 seedlings show increased sensitivity to UV-C
Since RNR also provides dNTPs for DNA damage repair, the reduced dNTP pool
sizes in tso2-1 might lead to hypersensitivity to DNA damage treatment, which is the
case for yeast rnr mutants. Therefore, we tested the sensitivity of tso2-1 to different DNA
damaging agents including MMS (alkalating agent), Mitomycin C (MMC, cross-linking
agent) and 4-NQO (UV-mimetic agent). With the same assay used for HU sensitivity
(Fig. 2-7), we did not observe any difference in the sensitivity to MMS, MMC and 4-
NQO among Ler , Col, r2a r2b and tso2-1 (data not shown). Filatov et al. (1996) found
that mammalian RNR genes could be induced by UV-C but not by HU or MMS. They
argued that mammalian RNR might be induced only by DNA damage requiring
nucleotide excision repair. Although 4-NQO is a UV-mimetic agent, published studies
have shown that different repair pathways might be used for DNA damage caused by 4-
NQO and UV-C. Therefore, we tested the sensitivity of wild type (Ler), tso2-1, r2b-1,
and r2a-1 r2b-1 mutants to UV-C.  tso2-1 exhibited an increased sensitivity to UV-C
(Fig. 2-12A, B) while r2b-1 or r2a-1 r2b-1 did not (data not shown). The similar level of
UV-C sensitivity of r2a-1 r2b-1 double mutants to wild type (Ler) indicates that TSO2
alone is sufficient to provide wild type level of protection against UV-C.  The embryonic
and seedling lethality prevented us from testing tso2-1 r2b-1 or tso2-1 r2a-1 double
mutants, which may show a more significant sensitivity to UV-C than tso2-1 single
mutants.
75
A
B
C
    Figure 2-12.  Increased sensitivity to UV-C and epigenetic misregulation in
r2 mutants. (A) An increased sensitivity of tso2-1 to UV-C as shown by a
larger number of dead seedlings at both 1500 J/m
2
 and 2000 J/m
2 
doses. (B)
A graph illustrating the survival rates of wild type (Ler) and tso2-1 seedlings
under increasing UV-C dosages.  (C) 25 cycles of RT-PCR were used to
detect TSI transcripts.  Lanes 1, 2, 3, and 4 correspond to WT (Ler), tso2-1,
r2a-1 r2b-1, and tso2-1 r2a-1, respectively.
76
tso2-1 r2a-1 double mutants exhibit defects in epigenetic regulation,
accumulate DNA damage, and undergo programmed cell death (PCD)
The phenotype of tso2-1 associated with meristem fasciation and increased
sensitivity to DNA-damaging agents resembles several Arabidopsis mutants defective in
DNA/chromatin replication and assembly, including mutants of BRU1, FAS1, FAS2,
AtCAP-E1, AtCAP-E2, and AtMRE11 (Bundock and Hooykaas, 2002; Kaya et al., 2001;
Siddiqui et al., 2003; Takeda et al., 2004).  All these mutants were found to release
transcriptional gene silencing (TGS) (Takeda et al., 2004) at the pericentromeric repeats
called Transcriptional Silent Information (TSI) (Steimer et al., 2000).  We found that
tso2-1 r2a-1 double mutant seedlings also released silencing at TSI but tso2-1 single
mutant seedlings did not (Fig. 2-12C).  We also tested tso2-1 single mutants at later
developmental stages. There is no release of TSI silencing in tso2-1 inflorescence tissues
(data not shown).
Are reduced dNTP levels in tso2-1 sufficient to impede DNA replication fork
progression and induce DNA strand breaks?  Using comet assay (single cell gel (SCG)
electrophoresis) (Angelis et al., 1999), we measured DNA damage levels in the
seedlings of tso2 single and double mutants.  While tso2-1 or r2a-1 r2b-1 double mutants
did not exhibit increased DNA damage, tso2-1 r2a-1 mutants exhibited significantly
increased DNA damage (Fig. 2-13A). Consistent with the comet assay, only tso2-1 r2a-1
seedlings were found to induce the expression of molecular markers associated with
DNA strand breaks (Fig. 2-13B).  These markers included Poly (ADP-ribose)
polymerase-1 (AtPARP1), AtPARP2, and AtRAD51 (Doucet-Chabeaud et al., 2001;
Doutriaux et al., 1998).
77
0
50
100
150
200
250
300
350
1
DNA damage (arbitrary units)
Lertso2r2br2a r2b
tso2 r2atso2 r2a
2-week-old5-week-old
B
A
    Figure 2-13.  Increased DNA damage in tso2-1 r2a-1 seedlings. (A) Results
from the comet assay indicating the relative amount of DNA damage in different
genotypes.  While tso2-1, r2b-1, or r2a-1 r2b-1 mutants exhibited similar levels
of DNA damage to wild type (Ler), tso2-1 r2a-1 double mutants exhibited an
increased level of DNA damage even at 2-week age.  The DNA damage level is
increased further in 5-week old tso2-1 r2a-1 double mutants. The extent of DNA
damage in each nucleus was indicated by 0, 1, 2, 3, or 4 units; examples of 0, 1,
2, and 3 are shown in the inset. The DNA damage units per genotype were
derived by summing up the units from 100 nuclei. (B) 27-30 cycles of RT-PCR
were used to detect the induction of AtPARP1, AtPARP2 and AtRAD51.  Lanes 1,
2, 3, and 4 correspond to WT (Ler), tso2-1, r2a-1 r2b-1, and tso2-1 r2a-1,
respectively.
78
The tso2 r2a double mutants do not survive beyond the 2 ~ 4-leaf stage. Very
often we found that the leaves of the double mutants turn yellowish, show symptoms of
chlorosis and eventually die. These symptoms resemble senescence and hypersensitive
response (HR)-mediated programmed cell death. In animals, genomic instability often
induces programmed cell death (PCD) in a p53-dependent fashion (Chernova Olga et al.,
1995; Vogelstein et al., 2000).  We tested if the accumulating DNA damage in tso2-1
r2a-1 seedlings observed above could lead to PCD using histochemical and molecular
markers.  Trypan blue (TB) was used to stain dead cells (Rate et al., 1999). Large patches
of TB-stained cells were observed in tso2-1 r2a-1 double mutant leaves, but not in wild
type nor tso2-1 (Fig. 2-14A).  Further, a high level of H
2
O
2 
and a large number of callose
depositions were detected in the leaves of tso2-1 r2a-1 double mutants (Fig. 2-14A).
Both H
2
O
2 
production and callose deposition are indicators of plant cells undergoing
hypersensitive PCD during incompatible plant-pathogen interactions (Brodersen et al.,
2002; Dietrich et al., 1994).  Additionally, tso2-1 r2a-1 seedlings expressed molecular
markers associated with hypersensitive PCD in plants (Brodersen et al., 2002) including
SAG13, Peroxidase C (PRXc), Glutathionine-S-Transferase (GST), Pathogenesis-Related
1 (PR1) and EDS1 (Fig. 2-14B).  However, SAG12, a cysteine protease specific to
senescence-induced PCD (Pontier et al., 1999), was not detected under the RT- PCR
condition used (Fig. 2-14B). Overall, tso2 r2a-mediated PCD appears similar to
hypersensitive PCD in plants.
79
tso2-1 is an recessive antimorphic allele
Since tso2-1,2,3, and 4 alleles are all missense mutations, it is not known if they
are loss-of-function alleles or antimorphic. A null mutation in TSO2 will help distinguish
the above alternatives. By sequence-indexed database search, I identified a T-DNA
insertion in TSO2. This allele was named tso2-5, in which the T-DNA is inserted 80 bp
downstream of the start codon ATG and causes a 10-bp deletion at the insertion site (Fig.
2-5), resulting in a frame-shift and truncation. Therefore, even though TSO2 mRNA level
is only reduced 2-fold (data not shown) in tso2-5 mutants, tso2-5 is very likely a null
allele. Surprisingly, tso2-5 does not exhibit any visible phenotype (Fig. 2-15A).
    Figure 2-14.  Expression of cell death markers in tso2-1 r2a-1 mutants. (A) Leaves of wild
type and tso2-1 r2a-1 double mutants were examined for programmed cell death using three
different histochemical markers: trypan blue staining, H
2
O
2
 production and callose
deposition.  Dark blue patches stained by trypan blue, reddish-brown deposits (reaction
products between 3,3-diaminobenzidine and H
2
O
2
), and callose deposition revealed by aniline
blue staining indicated programmed cell death in tso2-1 r2a-1 mutants.  (B) 25 cycles of RT-
PCR were used to detect different molecular markers associated with PCD.  SAG13, EDS1,
PR1, GST11, and PRXcb were induced to various levels in tso2-1 r2a-1 mutants.  GST6
remained the same in both wild type and tso2-1 r2a-1.  SAG12 and PR5 were not induced.
However, SAG 12 could be detected using 30 cycles of PCR (data not shown).
80
Since tso2-5 putative null allele displays no phenotype, the missense tso2 alleles
are likely antimorphic by either titrating out R1 or by heterodimerizing with and
inactivating R2A or R2B. Since tso2-1 is recessive, it suggests that tso2-1 dosage may be
important to cause a phenotype. To confirm that, we first crossed tso2-1 into tso2-5 to
generate a tso2-1/tso2-5 transheterozgyote. tso2-1/tso2-5 showed wild type phenotype
(Fig. 2-15D). We crossed tso2-5 with r2a r2b double mutants to generate triple mutants.
tso2-5 r2a r2b triple mutants are embryonic lethal because this triple mutant could not be
isolated from the progeny of wild type-like tso2-5/+; r2a/r2a; r2b/r2b plants. tso2-5 r2b
double mutants still display no phenotype. However, tso2-5/tso2-5; r2a/+; r2b/+ and
tso2-5/tso2-5; r2a/+ both are tso2-1-like. This indicates that two copies of the poisonous
tso2-1 have a similar effect as removing 100% TSO2 as well as 50 % of R2A and R2B.
These genetic data suggest that tso2-1 is an antimorphic allele.
 Another surprising observation is that tso2-5/tso2-5; r2a-1/r2a-1 is more severe
than tso2-1 single mutants and is similar to tso2-1/tso2-1; r2a-1/r2a-1, which are
seedling lethal (Fig. 2-16). In some extreme cases, tso2-5 r2a double mutants only
tso2-5tso2-5/tso2-1
tso2-1
Ler
Figure 2-15. tso2-5 nul
alel display a wild
type henotpe. (A)
wild type (Ler). (B)
tso2-1. (C) tso2-5. (D)
tso2-1/tso2-5.
B
C
A
D
81
develop one fully expanded cotyledon, and the development of the other cotyledon is
retarded (Fig. 2-16E). If tso2-1 interferes with both R2A and R2B, tso2-1 r2a double
mutants should display a more severe phenotype than tso2-5 r2a. This suggests that the
poisonous tso2-1 protein may only interfere with R2A, but not R2B.
Discussion
Our molecular genetic studies of tso2 led to the identification of mutations in all
three R2 genes in the Arabidopsis genome.  The analyses of single, double and triple r2
mutants demonstrated for the first time that normal dNTP pool and RNR function is
critical for the plant response to mutagens and for proper plant development. Our data is
supported by a recent transcriptome profiling study showing a 10-fold induction of TSO2
transcripts upon treatment with bleomycin plus mitomycin C (Chen et al., 2003).
Interestingly, while all four tso2 missense alleles exhibited morphological defects, single
r2a-1, r2b-1 and double r2a-1 r2b-1 mutants did not.  We propose several alternative but
nonexclusive explanations.  First, TSO2 may play a more prominent role than R2A and
R2B.  Second, R2A and R2B may function only under certain environmental or
A B C
D E F
Figure 2-16. tso2-1
r2a and tso2-5 r2a
exhibit similar
phenotype. (A) r2a.
(B-C) tso2-1 . (D-
F) tso2-5 r2a.
82
physiological conditions. Third, the four tso2 missense alleles may be recessive
antimorphic alleles. Recessive antimorphism is possible if the mutant TSO2 proteins
interact with R1 or other R2 subunits, inactivating the RNR enzyme complex altogether,
leading to a more severe phenotype than the single knock-out mutants. This antimorphic
nature of tso2-1 missense mutation is supported by the identification of tso2-5 and
genetic analysis between tso2-1 and tso2-5.
Differential regulation of three R2 genes
Although all three R2 genes are expressed in almost all tested tissues with
different abundance (Fig. 2-9A), there might be differential regulation for each of them.
First, no or very low levels of R2A transcript are observed in roots. The expression level
of R2B in cauline leaves is much lower than that in other tissues (Fig. 2-9A). Second, the
GUS staining pattern and in situ hybridization indicated that the TSO2 gene is cell cycle-
regulated, with a maximal expression levels in S-phase. Whether R2A and R2B are cell
cycle-regulated remains to be tested. Finally, the existence of E2F binding sites in the
promoters of the TSO2 and R2B but not R2A genes suggests that they are regulated by
different mechanisms. The consensus E2F binding site is TTT(C/G)(C/G)CGC in both
human and plants. The TSO2 gene contains one canonical (TTTGGCGC) and two
atypcial E2F binding sites (TTTGGCGG and TTTGCCGG in opposite direction) whereas
the R2B gene has only one canonical E2F binding site (TTTGGCGC). The only potential
E2F binding site (TTTCGCCG) in the R2A gene has CC at position 6 and 7, the two
positions that are intolerant for any change for E2F binding (Ouellette et al., 1992,
Ramirez-Parra et al., 2003). Thus, R2A is unlikely to be regulated by E2F.
83
Chamovitz and his colleagues found that Arabidopsis R2A but not TSO2,
physically interacts with CSN7/FUS5, one subunit of the COP9 signalosome (CSN)
complex (Levanon et al., 2003). This suggests that R2A might be regulated at the protein
level, rather than at mRNA level, resembling the mammalian R2 protein (Chabes et al.,
2003). TSO2 and R2A also differ in their ability to rescue yeast rnr mutants (Levanon et
al., 2003).
Intriguingly, cell cycle regulation of the human R1 gene is controlled by the YY-1
transcription factor (Johansson et al., 1998). In contrast, the human R2 gene contains a
CCAAT box (recognized by NF-Y) and an adjacent E2F4 binding site (TCTCCCGC) for
its cell cycle-regulation. It is noteworthy that E2F4 acts as a repressor (Chabes et al.,
2003). Transcriptional regulation of human R1 and R2 genes by different transcription
factors might allow the cells to fine-tune its RNR activity in response to different signals.
It is possible that other transcription factors in addition to E2F/DP are involved in
regulating plant RNR genes.
Induction of RNR genes by DNA damage
In yeast, the control of dNTP pool sizes has evolved as an important survival
mechanism (Zhao et al., 2003). Not surprisingly, RNR genes are the targets of the
surveillance machinery in response to different stresses. As a consequence, rnr mutants
are hypersensitive to various DNA damaging agents including MMS, 4-NQO, and UV. In
contrast, mammalian RNR genes seem to respond only to specific DNA damages. The
expression of mouse R2 gene cannot be induced by MMS or HU. However, the mouse R1
84
gene can be induced up to 3-fold, and the R2 gene 10-fold, by UV in a dose-dependent
manner (Filatove et al., 1996). They argued that MMS causes base damage by alkylation,
which is thought to be repaired by a short patch repair mechanism or dealkylation that
requires very little if any dNTPs (Filatove et al., 1996). Only upon UV irradiation, which
produces bulky lesions in DNA, are RNR genes induced to provide dNTPs needed for the
nucleotide excision repair pathway.
Our data on the induction of Arabidopsis RNR genes by DNA damage suggest
that plants might respond to DNA damage in a similar manner as animals, at least in
terms of induction of RNR genes. Like mammalian RNR genes, Arabidopsis R1 and R2A
genes are not responsive to MMS and 4-NQO induction and the TSO2 and R2B genes are
only slightly induced. Consistent with this, neither r2a r2b nor tso2-1 plants are
hypersensitive to MMS or 4-NQO treatments.  However, unlike animals, Arabidopsis
RNR genes can be induced by HU with the TSO2 gene increased by 20-fold but R2A by
only 2-fold. We have not tested the induction of RNR genes by UV treatment. We
showed that tso2-1 mutants are hypersensitive to UV-C (Fig. 2-12), suggesting that RNR
activity in tso2-1 is suboptimal for repairing bulky lesions caused by UV-C. Culligan et
al. (2004) reported that the Arabidopsis R1 gene is not induced by either HU or UV-B
treatment. The discrepancy could be due to different tissues used (they used whole
seedlings and we used cell suspensions) or different methods (they used Northern blot
and we used a more sensitive semi-quantitative RT-PCR). In addition, they used 1 mM
HU while we used 20 mM HU.
85
Using comet assay, Menke et al. (2001) found that over a wide range of
concentrations and treatment times, various DNA damaging agents including MMS and
the radiomimetic bleomycin, could cause DNA damage. However, combinations of
bleomycin and mitomycin C exert a strong induction of DNA damage-responsive genes,
such as Rad17, Rad51, and poly (ADP-ribose) polymerase (PARP-2) (Chen et al., 2003).
Under their conditions (1-1.5 ?g/ml bleomycin and 50-75 ?M mitomycin C), TSO2 was
induced 10-fold. This suggests that plant RNR genes might respond only to certain types
of DNA damage. In the case of tso2-1 mutants, the presence of R2A and R2B may be able
to provide enough dNTPs for small DNA damage caused by MMS and 4-NQO.
The induction of RNR genes by HU suggests that, unlike in animals, there might
be a feedback regulation of RNR genes in response to depletion of the dNTP pools in
plants. This is because HU, the specific inhibitor of the small subunit of RNR, will
abolish RNR activity and hence deplete the dNTP pools. In the yeast rnr2-314 mutant,
the damage response pathway is partially constitutive, as indicated by a 10-fold increase
in the RNR2 promoter activity compared with the wild type (Elledge and Davis, 1989).
This is consistent with our data that DNA damage accumulates and DNA repair genes are
up-regulated in tso2 r2a double mutants (Fig. 2-13).
Decreased dNTP levels affect organelle replication
One unique feature of tso2 mutants is the formation of white sectors on green
organs.  We observed a lack of chloroplasts in the photosynthetic mesophyll cells,
unusual air spaces beneath the epidermis, and abnormally small mesophyll cells in tso2
mutant tissues (Fig. 2-3E and F). In yeast, lower RNR activity or reduced dNTP level was
86
shown to increase the formation of mitochondrial DNA-deficient cells (petite cells) (Zhao
et al., 2001; Zhao et al., 1998), suggesting that organelle DNA replication is highly
sensitive to reduced dNTP levels. Therefore, stochastic depletions of chloroplasts and/or
mitochondria may underlie the white sectors and small cells in tso2 mutants.
In animals, the salvage pathway is involved in providing the dNTPs for
mitochondrial genome replication (see Chapter one). Several human disorders such as
mitochondrial DNA-depletion syndromes (MDS) are caused by mutations in the dNTP
salvage pathway. The white sectors and small cells in tso2 strongly suggest that the de
novo biosynthesis of dNTP is required for both chloroplast and mitochondrial DNA
replication. The development of white sectors occurring in the fifth leaf and later-arising
organs of tso2 could result from a gradual dilution of dNTP, which may be highly
enriched in embryos/seedlings via salvage pathways (Reichard, 1988; Saada et al., 2001).
Interestingly, Arabidopsis thymidine kinase 1 (TK1) seems to be a cytoplasmic protein
whereas thymidine kinase 2 (TK2) is localized in mitochondria. This is similar to human
where TK2 is also a mitochondrial protein. Deoxyguanisine kinase (dGK), however, is a
chloroplast protein in Arabidopsis. It seems that both de novo and salvage pathways are
involved in the replication of all three genomes in plants.
Fasciation is an indirect consequence of epigenetic mis-regulation in tso2
mutants
 Fasciation is manifested by distorted phyllotaxy, broadening and flattening and, in
extreme cases, bifurcation of the stem. Several mutants that have an extreme fasciation
phenotype, such as clv1, clv2 and clv3 mutants (Clark et al., 1993; Kayes and Clark,
87
1998; Fletcher et al., 1999), have been well studied in Arabidopsis. In these mutants, the
fasciation is manifested as enlarged shoot apical meristem, sometimes 1000-times larger
than wild type (Clark et al., 1997). Another class of fasciation mutants exhibits
bifurcation of the stem. In tso1-1 mutants, fasciation results from extensive bifurcation or
multiple splits of the inflorescence meristem (Song et al, 2001). Like tso1, tso2 mutants
belong to this second class of fasciation mutants. In tso2 r2a/+ and tso2 r2a/+ r2b/+
plants, meristem bifurcation is more frequently observed than in tso2-1 single mutants.
Mutants that are defective in either DNA/chromatin replication or assembly also exhibit
this second class of fasication. These mutants include bru1, fas1, fas2, AtCAP-E1,
AtCAP-E2, top1?, top1? and mre11 (see Chapter one; Bundock and Hooykaas, 2002;
Kaya et al., 2001; Siddiqui et al., 2003; Takeda et al., 2004). TOP1? and TOP1? encode
a type I DNA toposiomerase, promoting the relaxation of supercoiled DNA by
introducing a transient double-strand breaks (DSB) and acting in a number of different
DNA metabolisms such as DNA replication, transcription, repair, and chromatin
compaction (Takahashi et al., 2002). FAS1 and FAS2 encode the two subunits of
chromatin assembly factor with a function in assembling nucleosomes onto nascent DNA
(Kaya et al., 2000). The MRE11 gene encodes a component of the MRX complex with
roles in repairing DSB and maintaing telomere length (Bundock and Hooykass, 2002).
The BRU1 gene encodes a novel nuclear protein (Takeda et al., 2004).
Takeda et al. (2004) found that bru1, fas1, fas2, mre11, AtCAP-E1-/- and
AtCAPE2-/- are hypersensitive to DNA damage and can release transcriptional gene
silencing (TGS). This prompted them to propose that defects in the S-phase DNA-
88
damage checkpoint, or chromatin/DNA replication, can cause epigenetic instability of
newly replicated chromatin. As a result, release of gene silencing at crucial regulatory
gene loci such as WUS will lead to fasciation (Mayer et al., 1998; Takeda et al., 2004).
We observed the release of silencing at the pericentromeric repeats in tso2 r2a double
mutants (Fig. 2-13C), suggesting that epigenetic mis-regulation of WUSCHEL and
AGAMOUS may also underlie the fasciated SAM and homeotic transformation of floral
organs observed in tso2-1 mutants.
In Drosophila, a tesmin/TSO1 homolog, Mip120 is a member of Myb-containing
complex involved in site-specific DNA replication (Beall et al., 2002). It was shown that,
Mip120 and other Myb-interacting proteins are members of a repressive complex to
ensure a silenced transcriptional and/or replicative state. Mutations of each of the
members will lead to unscheduled DNA replication (Beall et al., 2004). The increased
DNA ploidy in tso1 meristem cells suggests that TSO1 might function in a similar way.
Therefore, it is very likely that the fasciation in tso1 mutants is also caused by defective
DNA replication. The link between chromatin/DNA replication and maintenance of gene
silencing might also exist in other organisms. Cell polarity is disrupted in C. elegans in
div mutants that are defective in DNA polymerases (Encalada et al, 2000), which is likely
caused by ectopic expression of certain developmental regulator genes.
Since rnr mutants of yeast or mammals have not been reported to have epigenetic
defects, it remains to be seen if release of TGS could underlie growth abnormalities in rnr
mutants of fungi or animals. Alternatively, defects in cytokinesis (shown in tso1) and cell
89
division (shown in tso2) may interfere with the CLAVATA/WUSCHEL signaling pathway
so that multiple stem cell organizing centers (specified by WUSCHEL) are initiated.
Cell division and development
Could the phenotype of tso2 be attributed solely to reduced dNTP pool sizes and
defective DNA repair? Alternatively, could RNR function in some unknown capacity?
Downes et al (2000) reported that mammalian S-phase checkpoint integrity depends on
transformation status and purine dexoyribonucleosides. They found that purine
deoxyribonucleosides could prevent the caffeine-induced activation of mitotic p34
cdc2
kinase. They speculated that normal high S-phase intracellular levels of purine
deoxyribonucleotides could act as an intracellular signal for S-phase, and could retard
progression towards mitosis. This explains why RNR is involved in oncogenic
transformation more so than RNR being a simple metabolic enzyme.
tso2-1 is a recessive antimorphic allele
According to Muller?s definition (1932), if a mutant allele behaves identically to a
deficiency or deletion, it is a null. If an allele leads to more severe phenotypes than a
deficiency, it is an antimorph.  A very common example of antimorph is the dominant-
negative effect. A dominant-negative allele normally produces a truncated or inactive
protein that interferes with its own wild type protein. The interference could be carried
out by: (1) competing with the wild type copy for the substrate or binding partner; (2) in
the case of homodimer or multimer formation, titrating out the wild type copy; therefore,
leading to a dominant effect.
90
Several recessive antimorphic alleles were reported in Drosophila (Marenda et al.
2003). However, the underlying molecular mechanism remains unclear. As RNR is a
tetrameric complex and R2 consists of a small gene family in Arabidopsis, it is not
surprising that the null or loss-of-function alleles of all three R2 genes exhibited no
phenotype due to functional redundancy, and the missense mutation tso2-1 exhibits
developmental defects due to its antimorphic nature. The mutant tso2-1,2,3,4 proteins can
be catalytically inactive but retain the ability to bind to other R2 or R1 subunits.
Alternatively, the mutant proteins lose their ability to bind one subunit but retain the
ability to bind other subunit. In either case, the mutant protein can compete the R1 or R2
subunits with other functional RNR subunits.
tso2-5 is a putative null allele as the T-DNA is inserted in the very beginning of
the TSO2 gene and causes a 10-bp deletion at the insertion site. The resulting protein
could be either truncated (containing the first 30 aa) or frame-shifted. r2a-1 is caused by
a premature stop codon, but retains two-thirds of the protein. The R2A mRNA level is
reduced in r2a-1 due to nonsense-mediated decay (data not shown), suggesting that r2a-1
might be a null allele. The R2B pseudogene in Col ecotype will make a truncated protein
(four-tenth of the wild type), indicating that r2b is likely a null allele.
The ultimate proof of tso2-1 as an antimorphic allele is to introduce the tso2-1
mutant transgene into wild type (Ler) or tso2-5 plants to see if the transgene can confer
mutant phenotype in a wild type background, and to see if it is dosage-dependent (to
91
explain why tso2-1 is recessive). The tso2-1 gene could be expressed under a strong
constitutive 35S promoter. p35S::TSO2-1 could cause a tso2-1-like phenotype in both Ler
and tso2-1 plants. In addition, further experiments need to be done in confirm that tso2-5
is a null. Since TSO2 transcripts are only reduced 2-fold in tso2-5, it might suggest that
the T-DNA could be spliced out or a cryptic promoter in the T-DNA sequence can still
drive the transcription of TSO2 from downstream of the T-DNA insertion in tso2-5. First,
Northern analysis could be performed to reveal the real size of the TSO2 transcript in
tso2-5. Second, 5? RACE could be conducted with a primer downstream of the T-DNA
insertion site to unveil the sequence of the TSO2 transcript, which can determine if
frame-shift does occur in tso2-5. The last and indisputable evidence could be harvested
from Western analysis to detect if any TSO2 protein is made in tso2-5. It is note worthy
that although tso2-5 alone displays no visible phenotype, tso2-5/tso2-5 r2a/+ exhibits a
tso2-1 phenotype. In addition, tso2-5 r2a and tso2-1 r2a show a similar phenotype.
Moreover, while tso2-5 r2b still fail to show any phenotype, tso2-1 r2b is embryonic
lethal. These genetic results are strongly against that tso2-5 might be a weak and tso2-1 a
strong loss-of-function mutant.
tso2-mediated PCD
Senescence and the hypersensitive reponse (HR)-mediated cell death are the two
major types of programmed cell death (PCD) in plants. As a defense mechanism, cells
undergo PCD to limit the spread of the invaders (HR) or to recycle the energy and
resources (senescence). In animals, genomic instability often leads to p53 dependent or
p53-independent apoptosis (Chernova et al., 1995; Vogelstein et al., 2000), so that cells
92
with extensive DNA damage and mutations are eliminated, in order to avoid the passage
of defective genetic information to the next generation.
While genomic instability-caused apoptosis has been well studied in animals, it is
rarely reported in plants and the mechanism remains unknown. Since plants do not have a
p53 homolog, it is likely that plants evolved a different signal transduction pathway in
response to genomic instability. Both PCD and DNA damage are not observed in tso2-1
single mutants, but both were observed in tso2 r2a double mutants, suggesting a causal
correlation between DNA damage and PCD.
Several histochemical and molecular markers were used to detect PCD in tso2 r2a
double mutants. SAG13, a widely recognized PCD marker, is induced in tso2 r2a double
mutants. EDS and PR-1, which are up-regulated in HR-mediated PCD, also accumulate
in the double mutants (Fig. 2-15B). SAG12 is a senescence-specific marker, the
expression of which is strictly dependent on age (Noh and Amasino, 1999). We detected
a low level of SAG12 in tso2 r2a (with longer cycles of PCR), suggesting that a
senescence process might also be turned on. It is possible that PCDs, triggered by
different signals in plants, share common downstream effectors. By examining if PCD
occurs in other plant mutants defective in genome integrity, one may pinpoint the
DNA/cellular lesions required to switch on PCD. Although there is no p53 homolog in
plants, our data suggest that plants also possess a mechanism to induce cell cycle arrest
and eliminate damaged cells via PCD.  The viable tso2 mutants will facilitate genetic
screens for mutations in p53-like genes in higher plants.
93
Chapter three
Isolation of Suppressors of tso2-1
Abstract
To further understand the regulation of RNR genes and to identify novel
components of plant checkpoint pathways, a genetic screen of suppressors of tso2-1 was
performed. A preliminary screen from only 4,000 M2 seeds led to the isolation of 18
independent suppressor mutations that can partially or completely suppress the phenotype
of tso2-1. Two of these 18 suppressors display a novel phenotype and are dominant.
Complementation tests showed that the 18 suppressors fall into at least four different
complementation groups. The potential pathways that these suppressor mutations may
define are discussed.
Introduction
It is widely known that due to genetic redundancy, knockout mutations in a gene
may not lead to any visible or detectable phenotype. Arabidopsis is a superb model
organism for genetics, as the large number of T-DNA insertion and transposon-tagged
lines makes it possible to knockout almost every gene in the genome. If a gene belongs to
a small gene family, each individual gene can be knocked-out or knocked-down and
double or triple mutants can be constructed to reveal their functions. For instance, the
Arabidopsis genome has three SEPPLATA (SEP) genes. Single mutants of sep1, sep2 or
sep3 or double mutants of any two combinations do not exhibit any visible phenotype.
The triple mutant sep1 sep2 sep3, however, produces flowers consisting of all sepals,
94
revealing the functions of these three SEP genes in petal, stamen and carpel development
(Pelaz et al. 2000).
Modifier screens have been very powerful to isolate genes whose functions are
hidden or masked by their redundant genes. This is based on the principal that functions
of a gene become more crucial in certain genetic backgrounds, where the functions of its
redundant members are reduced.  A good example is the isolation of several C-class
floral genes in Arabidopsis. For a long time, AGAMOUS (AG) was the only C-class gene
identified.  Additional C?class genes, HUA1 and HUA2, were only isolated much later in
a modifier screen in the background harboring a weak ag-4 allele. Single mutants of hua1
or hua2 and even the hua1 hua2 double mutants do not display any floral defects on their
own but can dramatically enhance the weak ag-4 phenotype (Chen and Meyerowitz,
1999). Such ?modifier? screens are widely practiced in other genetic organisms such as
Drosophila and C.elegans.
Ribonucleotide reductase (RNR) is an essential enzyme that catalyzes the rate-
limiting step for the production of the deoxyribonucleotides required for DNA synthesis.
Therefore, the RNR genes have been tightly regulated at different levels. For example,
RNR genes can be transcriptionally regulated by Crt1 repressors, post-transcriptionally by
Cird13, post-translationally regulated by Sml1 inhibition protein or by cytoplasm-to-
nucleus translocation (For detail, see Chapter one). Although a large amount of
information is available on the regulatory mechanisms of RNR in yeast and animals, little
is known about how RNR expression and activity are regulated in plants. As shown in
95
Chapter two, unlike yeast, expression of Arabidopsis RNR genes are not induced, or
slightly induced, by certain types of DNA damaging agents such as MMS and 4-NQO (at
least under the conditions we used). Since mammalian RNR genes cannot be induced by
MMS, it suggests that similar mechanisms of regulation of RNR in response to DNA
damage might be conserved between animals and plants. However, the TSO2 gene can be
induced up to 10-fold by a combination of bleomycin and mitomycin C whereas human
RNR genes can be induced 3~10-fold by UV treatment. The expression level of R2B is
very low compared to TSO2 and R2A (Fig. 2-10A). Is R2B normally repressed by general
co-repressors? Is there a Sml1 type inhibitory protein of RNR in Arabidopsis? Mutations
that inactivate negative regulators of RNR may suppress tso2-1 phenotypes. Taking
advantage of the fertile phenotype of tso2-1 and the fact that the dNTP level is below the
threshold for normal development in tso2-1, creating a sensitive genetic background, a
modifier screen will be highly informative.
In yeast, a genetic screen aimed at identifying genes that regulate RNR3 turned
out to be a big success. Several important DNA damage and replication checkpoint genes
were isolated and novel regulation mechanisms were revealed (Zhou and Elledge, 1992).
In yeast, the mRNA level of RNR3 is almost undetectable under normal growth
conditions but can be induced up to 100-fold by DNA damaging agents. Taking
advantage of this feature, Zhou and Elledge designed a genetic screen to isolate mutants
that can express high levels of RNR3 in the absence of DNA damaging agents. More than
202 independent crt (constitutive RNR3 transcription) mutants were obtained and fell into
9 complementation groups. Five of the CRT genes have been identified as previously
96
cloned genes: TUP1, SSN6, RNR1, RNR2, and POL1/CDC17. crt4, which encodes the
TUP1, has 58 alleles and crt1, which encodes a DNA binding protein, has 131 alleles.
CRT1 acts by recruiting the general repressors TUP1 and SSN6 to the promoters of
damage-inducible genes including RNR. The discovery that POL1/CDC17 is one of the
CRT genes illustrated for the first time that directly blocking DNA replication can
provide a signal to induce the DNA damage response (Zhou and Elledge, 1992).
Due to the essential role of RNR, rnr null alleles are embryonic lethal in
Drosophila and mice, preventing us from understanding the functions of RNR in later
developmental stages in higher eukaryotes. tso2-1 is the first viable rnr mutant that
exhibits developmental defects in higher eukaryotes. Genetic screen for suppressors or
enhancers of tso2-1 will provide a tremendous opportunity to uncover genes involved in
the regulation of RNR or DNA replication. There are at least eight different mechanisms
by which a mutated gene could suppress the tso2-1 phenotype.  First, mutations in genes
encoding transcriptional regulators of the TSO2 gene can be identified. Mutations in this
type of regulatory genes will reduce or abolish tso2-1 transcription, leading to a reduction
of the antimorphic tso2-1 product. As described in Chapter two, the tso2-5 null allele and
the tso2-1/tso2-5 transheterozygote both show a wild type phenotype. Hence, reduced
tso2-1 may be equal to tso2-1/tso2-5 or tso2-5/tso2-5. Second, a nonsense mutation in the
TSO2 ORF can act as the intragenic suppressor. Such suppression will eliminate the
antimorphic tso2-1 protein product. Third, mutations in negative regulatory genes of R1,
like the budding yeast SML1 (see Chapter one), may release the inhibitory effect on the
R1 subunit and enhance the overall RNR activity. Fourth, gain of function mutations in
97
genes involved in the RNR salvage pathway, or recessive mutations in genes that
negatively regulate the salvage pathway, may result in more dNTP production, thus
compensating for the tso2-1 effect. Fifth, gain-of-function mutations in R1 or other R2
genes (R2a and R2b) may result in higher RNR activity so that the tso2-1 effect will be
compensated or compromised. In both yeast and human, mutations in the allosteric
activity site of R1 that eliminate the dATP feedback inhibition results in two times higher
dNTP production (Chabes et al., 2003). Sixth, mutations in genes involved in
posttranslational regulation of RNR may either stabilize the R1 or other R2 subunit, or
promote the degradation of the TSO2 protein. As a consequence, the effect of tso2-1 can
be reduced, eliminated or compromised. Seventh, mutations in genes involved in DNA
replication, such as DNA polymerase may somehow more efficiently utilize the low
dNTPs in tso2-1 to reach normal DNA replication, thus compromising the effect of tso2-
1. Although unlikely, it was found that overexpression of R1 could rescue a mutant
phenotype caused by defective mitochondrial DNA polymerase MIP1 gene in yeast
(Lecrenier and Foury, 1995). Eighth, mutations in genes involved in DNA replication
checkpoints or a sensor of dNTPs, might trigger a feedback pathway that more effectively
induces the expression of other RNR genes. Mutations in any one of these eight pathways
alone might not display a visible phenotype but may be recovered in tso2-1 background.
This chapter describes the isolation of suppressors of tso2-1. Eighteen suppressors
of tso2-1, named shuai (sua) (means handsome in Chinese), were screened from only
4,000 M2 seeds. Complementation tests indicate that they fall into at least four
complementation groups. Cloning of these sua genes in the future will not only enhance
98
our understanding of the regulation of RNR genes in plants, but also shed light on how
the size of dNTPs pool is maintained and how cell division is integrated in the context of
development in plants.
Results
The suppressor screen was designed to isolate mutations that could suppress the
phenotype of tso2-1, either partially or completely. If the suppression is complete, then
wild type-like plants are expected to emerge in the M2 generation. Caution has been
exercised in the screen as tso2-1 plants exhibit a certain degree of variation in phenotypic
severity from batches to batches and under different growth conditions.  To this end, all
the potential suppressors were grown to the M3 generation and the phenotype of each line
was confirmed. For those lines that continue to show the suppressor phenotype, PCR-
based genotyping was conducted to make sure they have tso2-1 homozygous mutation in
the background in order to eliminate wild type contaminants. The majority of the lines
were also sequenced in the TSO2 ORF to look for any potential intragenic suppressors.
The sequencing results were always consistent with the PCR-based genotyping using the
dCAPS marker, i.e. detecting original tso2-1. None of them were intragenic suppressors.
Since the suppressors can suppress the tso2 phenotype, we name them shuai (sua) (means
handsome in Chinese).
99
Forty-one wild type-like plants, or plants displaying very subtle tso2-1
phenotypes, were initially isolated from screening about 4, 000 M2 plants (from 50 M1
families). Seeds were collected from these 41 lines and grown alongside tso2-1 plants for
phenotypic comparison.  Three lines exhibited a tso2-1 phenotype and were eliminated.
Thirteen lines segregated in a 3:1 ratio of wild type and tso2 phenotypes in the M3, and
genotyping/sequencing analyses showed that none of the wild type-plants were tso2-1
homozygous, suggesting that they were likely derived from cross-pollination between
tso2-1 and wild type. Seven lines are wild type contaminants as they were wild type at
the TSO2 locus. The remaining18 suppressors were kept for further analysis. These 18
lines all showed a wild type phenotype but were homozygous for tso2-1 mutation (Fig. 3-
1). Since they were isolated from different M1 families, they were very likely
independent lines. Later on, one of the 18 lines, DS6 consistently exhibited a very subtle
tso2-1 phenotype (shown in Fig. 3-1H) where only the center of the inflorescence became
whitish in later developmental stages, suggesting that DS6 only partially suppresses tso2-
1.
    Figure 3-1. Representative suppressor lines of tso2-1. (A) wild type (Ler).
(B) tso2-1.(C-H) suppressor lines DS2, DS8, DS11, DS12, DS15 and DS6.
Note that the center of the inflorescence is whitish in DS6, indicating it
partially suppresses tso2-1.
Lertso2-1
A B C D
E F GH
100
In order to determine whether these 18 lines were allelic to each other, crosses
were made among these lines. To reduce the workload, the suppressor lines were
randomly divided into four groups. Lines in group A were crossed to lines in Group B
and Lines in group C were crossed to lines in Group D. Results from the
complementation tests are shown in Table 3-1 and Table 3-2.
Assuming each line is recessive, if the two lines complement each other (shown
as ?+? in the table), the F1 generation will display a tso2-1 phenotype as they can no
longer suppress the tso2-1 phenotype. Thus, they belong to different complementation
groups. If the two lines do not complement each other (shown as ?X? in the table), they
belong to the same complementation group and the F1 generation will still exhibit a wild
DS12DS15DS17DS19DS21
DS2+ + + + +
DS6+ + + + +
DS8+ + X + X
DS11+ + + + +
Table 3-1. Complementation test of 9 suppressor lines
Note: "+" complementated; "X" not complementated.
DS38DS41DS53
DS31X N/A +
DS33+ partialX
DS35+ partialX
DS42+ partialX
Table 3-2. Complementation test of 7 suppressor lines
Note: "+" complementated; "X" not complementated
 "N/A" no data.
101
type phenotype because the tso2-1 phenotype is still suppressed. As shown in Table 3-1,
DS8, DS17 and DS21 belong to the same complementation group and they are allelic to
each other. While DS 2, DS 6 and DS11 can complement with each of DS12, DS15 and
DS19, these 9 lines fall into at least two other complementation groups. Crosses among
DS2, DS6 and DS1as well as among DS12, DS15 and DS19 are in progress.
As shown in Table 3-2, DS33, DS42 and DS53 cannot complement each other
and hence, they are allelic to each other. DS31 and DS38 belong to the same
complementation group. Interestingly, the F1 plants from the crosses between DS41 and
DS33, DS41 and DS35, or DS41 and DS42 exhibit an intermediate phenotype (Fig. 3-2
C-F), that is, the phenotype is much weaker than the tso2-1 plants grown side by side.
Some of the floral organs still have white sectors but the leaves are normal. The
intermediate phenotype could be due to the environmental variations or due to DS41
being semi-dominant.
A C
D E
F
B
    Figure 3-2. DS41 might be semi-dominant. (A) Wild type. (B) tso2-1.
(C) F1 plants from the cross between DS41 and DS33. (D) F1 plants from
the cross between DS41 and DS35. (E-F) F1 plants from the cross
between DS41 and DS42.
102
 To segregate the suppressor mutations away from tso2-1, each suppressor line
was crossed into wild type (Ler) and each suppressor mutation will be recovered in the
F2. It will be difficult to identify suppressor lines that do not display any visible
phenotype in the absence of tso2-1. In this case, the suppressors will always be kept in
tso2-1 background and the double mutants (suppressors in tso2-1 background) will be
used for map-based cloning. 
Although the majority of the suppressor lines exhibited a wild type phenotype,
two suppressors showed a novel phenotype in addition to their ability to suppress the
tso2-1 phenotype. In the progeny of the self-pollinated DS101 line (which we named
tso2-1sua1-1/+ tso2-1
sua1-1 tso2-1sua1-1 tso2-1
A
C
B
D
E F
Figure 3-3. Phenotype
of sua1-1 suppressor
(DS101) in tso2-1
background. (A) tso2-1;
(B)  sua1-1/+ tso2-1.
Note that the down-
dripping inflorescence;
(C) sua1-1 tso2-1. The
plant is dwarf and the
inflorescence is smaller
consisting of a few
number of flowers; (D)
sua1-1 tso2-1. Note
leaves with chlorotic
lesions characteristic of
cell death. (E) A single
flower of wild type
(Ler) (left) and sua1-1/+ tso2-1 (right). (F) A silique of wild type
(bottom) and sua1-1/+ tso2-1 (top).
103
sua1-1), one quarter of the plants showed a tso2-1 phenotype. One quarter of the progeny
(presumably sua1/sua1 tso2-1/tso2-1) were dwarf and completely sterile with compact
inflorescences (Fig. 3-3C).  The rosette leaves of this dwarf plant displayed chlorotic
lesions, a characteristic of cell death (Fig. 3-3D). The other half of the progeny (sua1/+
tso2-1/tso2-1) showed a compact inflorescence and the flowers were smaller, with
downward pointing siliques, only one-third the length of the wild type siliques (Fig. 3-
3B, E, F). The segregation pattern indicated that DS101 is a dominant mutant. Another
suppressor, DS102, behaves similarly to DS101 (data not shown) and is likely allelic to
DS101. Complementation tests between them are in progress. In the F1 progeny of a
cross between sua1-1/+ tso2-1 and wild type, none of the plants exhibited this novel
phenotype. That is, sua1-1/+ tso2-1/+ plants are wild type in phenotype, suggesting that
sua1-1 in a wild type background is recessive.
In summary, 15 recessive, one semi-dominant, and two dominant suppressors of
tso2-1 were isolated from a screen of 4,000 M2 seeds. 18 suppressors fall into 3-10
complementation groups. We preliminarily assigned them into six complementation
groups (Table 3-3). Crosses between lines in Group 1, 2, 3 and lines in Group 4, 5, 6 are
Alleles in each groupRecessive or dominant
Group 1DS8, DS17, DS21recesive
Group 2
a
DS2, DS6, DS11recesive
Group 3
a
DS12, DS15, DS19recesive
Group 4DS41 semi-domiant?
Group 5DS31, DS38recesive
Group 6DS33, DS35, DS42, DS53recesive
Group 7
a
DS101, DS102dominant
Table 3-3. Potential complementation groups of the 18 suppressors
 
a
: Allelism has not been verified by complementation test among the group members
104
in progress. Crosses among the lines within Group 2, Group 3, and Group 7 are also in
progress.
As mentioned in Chapter two, tso2-1 is likely an antimorphic allele. A null allele,
tso2-5 can suppress the tso2-1 phenotype as tso2-1/tso2-5 exhibited a wild type
phenotype. This suggests that a cis-mutation, either in the promoter of tso2-1 that
abolishs the expression of tso2-1, or in the tso2-1 ORF that produces a truncated tso2-1
protein, which removes the poisonous effect of tso2-1, will suppress tso2-1. Obviously,
these intragenic suppressors will be dominant. Nevertheless, we sequenced the tso2-1
ORF in all 15 recessive suppressors and one dominant suppressor, and found no cis-
mutations in any of these suppressor lines.
Discussion
At first glance, the number of suppressors is unusually high from the preliminary
screen of about 4,000 M2 seeds even after the number of real suppressors is reduced from
41 to 18. Considering that 202 crt mutants were isolated from the small genome of yeast,
it is likely that the number of Arabidopsis genes involved in the regulation of RNR is
large, suggesting that the regulation of RNR in plants is highly complex.
It is surprising that none of the 18 suppressor lines is an intragenic suppressor.
Based on the sequencing results, no nonsense mutations were found in the TSO2 ORF in
any of the 18 suppressor lines. There are two possible explanations. First, mutations in
many genes in at least eight different pathways can suppress the tso2-1 phenotype and
105
nonsense mutations in the TSO2 gene are not among the most common ways to suppress
tso2-1. Again, since the intragenic suppressors are expected to be dominant, such
dominant suppressors might have been eliminated at the M1 generation because of their
wild type phenotype, which could be mistaken for contaminants. A more focused screen
in the M1 generation might be necessary to identify such mutants. Alternatively, tso2-1
might not be an antimorphic allele. This is not supported,however, by the fact that the
tso2-5 null allele and tso2-5/tso2-1 transheterozygotes are wild type. In addition, even if
tso2-1 were not an antimorphic allele, tso2-5 like mutations could still suppress the tso2-1
phenotype and; hence, act as intragenic suppressors.
Although backcrosses between the suppressors and tso2-1 were not conducted,
the complementation tests suggest that 16 of the 18 suppressors are recessive. Since all
16 suppressor lines were grown to the M3 generation and none segregated a tso2-1
phenotype in the M3, they are all homozygous for the suppressor mutations. If any one of
the suppressor lines is dominant, then the F1 progeny of the cross between the dominant
homozygous suppressor and a recessive suppressor should all show the suppressed
phenotype and no tso2-1 phenotypes. In contrast, if all the F1 progeny of the cross
between two suppressors showed a tso2-1 phenotype, then both suppressor lines are
recessive. A detailed genetic analysis of the complementation test is shown in Fig. 3-4.
The mutations in the suppressor lines could be quickly divided into two classes.
One class is a cis-mutation. They are either nonsense mutations in the TSO2 ORF,
producing a truncated protein, or point mutations or deletions in the TSO2 promoter,
106
reducing or abolishing TSO2 transcription. In both cases, the poisonous effect of tso2-1 is
removed. As a result, tso2-1
sup (null)
/tso2-1 is wild type in phenotype and should be
dominant over tso2-1. The two dominant suppressor lines (DS101 and DS102) are
unlikely cis-mutations as they exhibited novel phenotypes such as downward pointing
siliques, compact inflorescences, and dwarfism. It is possible that these dominant
suppressors encode transcription factors that control the expression of TSO2 as well as
tso2-1DS1
tso2-1DS1
tso2-1DS2
tso2-1DS2
?
tso2-1DS1
tso2-1+
DS2
+
?
tso2-1DS1
tso2-1
DS2
DS1 and DS2
are recessive
but not allelic
?
DS1 and DS2 are
allelic, either
dominant or recessive
15/16
1/163/16
13/16
Both DS1 and DS2
are dominant but
not allelic
Either DS1 or DS2
is dominant but not
allelic
tso2-1DS1
tso2-1+
DS2
+
F1
F2
  Figure 3-4. Diagram of complementation tesst of suppressor lines.  If the two
lines can complement each other, the F1 plants will exhibit tso2-1 phenotype,
suggesting the two lines are recessive and not allelic to each other.
107
other target genes responsible for the novel phenotypes. Alternatively, the dominant
suppressors (DS101 and DS102) could be involved in protein degradation pathways;
mutations of which may increase the levels of R1, compensating for tso2-1 poisonous
effect. This may also increase the stability of other proteins, leading to novel phenotypes.
Trans-mutation is the second class of suppressors. The majority of this class
should be recessive. This type of mutation could define global repressors (like yeast Tup1
and Ssn6), RNR specific repressors (like Crt1 in yeast) or RNR inhibitors (like yeast
Sml1). They could also define genes in RNR mRNA stability (like Cid13 in yeast), genes
in DNA replication checkpoints, DNA polymerase, or the salvage pathway of dNTP
biosynthesis. The first thing we can do before mapping and cloning of these suppressor
genes is to determine the mRNA levels of RNR in the suppressors by RT-PCR. It will
help us to distinguish among these different pathways in which the suppressors might be
involved.
108
Chapter four
Concluding Remarks
Cell division and development
Although the basic events of the cell cycle have been well studied in yeast and
animals, how cell division is regulated in the context of development is not clear. Two
opposing views have been proposed to address the function of cell division in plant
growth, which are probably also applicable to animals (Kaplan, 1992). The ?organismal
theory? regards cells as merely compartments of developmental space that is defined at a
higher level by genes specifying the morphology. Therefore, the final body plan could be
altered by mechanisms that do not directly affect cell-cycle parameters. This theory also
implies that the stereotyped cell division patterns become insignificant. The opposite ?cell
theory? considers cells as fundamental building blocks of the organism, and thus, the
spatial and temporal control of cell proliferation is essential for the pattern formation.
While supports for each theory are accumulating (see Chapter one), there are
increasing observations suggesting that the two theories are not mutually exclusive, and
development could be controlled at both levels. For example, constitutive expression of
the cell cycle inhibitory protein KRP2 in Arabidopsis reduces cell division rates and
forms leaves with fewer, but larger cells (De Veylder et al., 2001; Wang et al., 2000),
indicating an overall organismal control. However, the leaves of KRP2-overexpressing
plants also become narrow and serrated, and grow at much lower rates, indicating a
cellular mechanism.
109
Overexpressing E2Fa induces extra cell divisions, leading to enlarged cotyledons.
However, when its dimerization partner DPa is also overproduced, massive over-
proliferation occurs in most organs, resulting in distorted growth (De Veylder et al.,
2002). These data suggest that cell cycle regulation is not just a simple speeding up or
slowing down of cell production, which translates in faster or slower growth. Growth and
development is governed by a variety of signaling (including light, temperature, internal
positional and developmental cues) and growth substances (such as hormones and
nutrients), which can create an interaction between cellular and whole-organ behavior
and allow cells to compensate for impaired cell division by increasing cell expansion and
vice versa.
Mutants with aberrant (or incomplete) cell division tend to show stronger
developmental defects than those with reduced or increased cell division rates. It is
probably because these kinds of mutants also disrupt cell-to-cell communication, through
which a general machinery can manage cell division locally. As mentioned in Chapter
one, tso1 mutants cause defects in both cell division and cytokinesis. tso1 exhibits
enlarged and fasicated inflorescence meristems and callus-like flowers consisting of
misshapen sepals only. Recent discovery of Drosophila p120, a TSO1 homolog, being a
member of Myb-containing repressive complex to ensure a silenced transcriptional and/or
replicative state (Beall et al., 2002), suggests a similar role of TSO1 in DNA replication.
110
The similar fasication phenotype in tso2 and other mutants such as fas1, bru1 and
mre11, indicates a link between DNA/chromatin replication or DNA damage checkpoint
and epigenetic stability. That is, mis-regulation of the shoot apical meristem regulatory
gene WUS likely underlies the fasication phenotype shared by these mutants. However,
unlike fas1, bru1 and mre11 mutants, in which morphological defects are largely limited
to meristem fasciation, both tso1 and tso2 exhibit additional developmental defects that
could not be explained by ectopic expression of certain developmental genes. For
example, tso1-1 flowers could not develop any petals, stamens and carples. The sepals
are in irregular shape and very often the number of sepals is increased from 4 to 6 (Liu et
al., 1997). In tso2-1 mutants, sepal epidermal cells are disorganized and often outgrown
to form protrusions from the flat surface. One explanation is that the delayed or aberrant
cell division could disrupt the local cell-to-cell communication and hence, the positional
information is changed. An upstream machinery that controls cell division and cell
growth in plant development by assessing planes and rates of cell division and directing
the division of neighboring cells accordingly, may fail to adjust to this change, leading to
improper development.
Compensatory proliferation
The outgrowth of cells in tso2-1 sepal epidermis and the uneven thickness of
rosette leaves could also be due to over-proliferation in certain regions. This over-
proliferation is more obvious in the leaves of tso2 r2a double mutants. Then, how to
explain that, a lower dNTP pool caused by impaired RNR activity in tso2-1 could lead to
over proliferation? The striking similarity of tso2 r2a double mutants and Drosophila
111
hippo mutants suggests that there might also be a compensatory proliferation induced by
certain mechanisms in plants (Wu et al., 2003).
Genetic and molecular analyses showed that Hippo is important for transcriptional
repression of cyclin E and DIAP1, an inhibitor of apoptosis. Therefore, in hippo mutants,
cell division is promoted whereas cell death is inhibited, leading to over proliferation in
eyes (Wu et al., 2003). The phenotype is in contrast to overexpression of growth
stimulating genes Myc or E2F in Drosophila, which results in extra cell growth or
division, but has little or no effect on body size (Neufeld et al., 1998). Normally, cell
division and cell death are coupled so that excessive cell growth and proliferation could
    Figure 4-1. Eyes of Drosophila (left panel) and leaves of Arabidopsis
(right panel). (A)The head of a normal fruitfly, showing its perfectly
organized eyes; (B) The head of hippo mutant, vastly overgrown
tissues resulted from increased cell proliferation and decreased cell death;
(C) A leaf of wild type, showing flat smooth surface; (D) A leaf of tso2 r2a
double mutants, showing disorganized and overgrown cells. (A-B) Pictures
are reproduced from Wu et al., 2003.
A
B
C
D
112
be compensated for by extra apoptosis. Conversely, cell death could also lead to
compensatory proliferation (Milan et al., 1997).
Recently, Huh et al (2004) reported that stimulation of cell death signaling in the
wing disc, even in the absence of cell death itself, resulted in increased proliferation.
They proposed that the Dronc gene, as an apoptosis inducer, plays two distinct roles. In
its primary role, Dronc transduces death signals in the cells in which it is activated.
Meanwhile, it promotes these cells to produce a signal that drives the proliferation of
nearby cells. Similarly in tso2 r2a double mutants, a signal might be produced in those
cells defective in DNA replication or delayed in cell division, to inform surrounding cells
that a particular developmental event has not been completed and/or provides instructions
on what to do next. It is also possible that differentiation of surrounding cells is delayed
and hence, their cell fates might be changed due to altered positional information.
The shoot meristem of tso2 r2a double mutants often enlarged and developed a
tumor-like structure. Surprisingly, our preliminary in situ data showed that expression of
both STM and ANT genes are largely reduced in the shoot meristem of tso2 r2a double
mutants (data not shown). STM is a shoot meristem regulator gene that is specifically
expressed in the shoot meristem whereas ANT is an organogenesis marker that is only
expressed in developing organ primordia. This result may suggest that cells in the
enlarged and disorganized shoot meristem of tso2 r2a double mutants fail to gain their
differentiation status or their cell fates are unspecified. It is also possible that these cells
113
are dying, and hence the transcription machinery is shut down so that neither STM nor
ANT can be expressed.
In hippo mutants, it was suggested that the mutant cells still retain the ability to
differentiate, but become incompetent at recognizing or responding to signals that would
normally promote differentiation. The resembling phenotype shown in tso2 suggests the
existence of a similar mechanism in plants. Further studies of tso2 and tso1, and the
isolation of both tso2 and tso1 suppressors will provide important insights into the
question of how cells acquire, integrate, and utilize positional information on the status of
their environment to control growth and differentiation.
114
Chapter five
Materials and Methods
Plant growth, mutant strains, and materials
Plants were grown under 16 hour light/8 hour dark cycle at 20
o
C.  The tso2-1, -2,
-3, and -4 alleles were isolated in two separate EMS mutagenesis screens (Liu and
Meyerowitz 1995; Levin, Fletcher et al. 1998), and all four alleles were generated in the
Ler background. . tso2-5 was a T-DNA insertion line (217A06) in WS background
purchased from INRA-Versailles (http://flagdb-genoplante-
info.infobiogen.fr/projects/fst/DocsIntro/introCollection.html). tso2-6 was a T-DNA
insertion line (Salk_140203) in Col background obtained from Arabidopsis Biological
Resource Center (ABRC). The method of TILLING (McCallum, Comai et al. 2000) was
used to isolate r2a-1.   Specifically, a primer pair 5? TTTGCTGTGAGGCTGGTCGCTT
TT 3? and 5? CTTCCAGATTCGATGGCGGATTCA3? was used to amplify ~ 1 kb R2A
genomic DNA from EMS-mutagenized Col M2 plants.  HPLC-based detection of DNA
heteroduplex led to the identification of r2a-1.  A primer pair, 5? CGATTAAATCTTCTT
CACCGGA 3? and 5? GGCTCCAATCCTTTTTGGAT 3? was used to PCR-amplify the
entire R2B genomic DNA from three different Arabidopsis accessions (Col, Ler, and WS)
and 15 polymorphisms were found between Col and Ler including a 2-bp insertion that
gives rises to a functional R2B gene in Ler and WS. pcycB1:GUS plant seeds were
provided by Dr. Peter Doerner (Colon-Carmona et al., 1999).  TSO2 and R2A cDNAs,
RZL13g10F (AV546418) and SQ022c10F (AV555679), were provided by the Kazusa
DNA Research Institute.
115
Genetic analyses
tso2-1 (Ler) was crossed with r2a-1 r2b-1 (Col).  The F1 heterozygous progeny
for all three mutations are wild type.  dCAPS and CAPS markers (see Appendix Table
A1) were used to screen tso2-like F2 plants at R2A and R2B loci to identify plants of
various genotypes. PCR-based genotyping was performed for individual seeds from tso2-
1/tso2-1; r2b-1/+ parents.  tso2-1 r2b-1 double mutants were only detected among
aborted seeds (Table 2-3). The r2a-1/r2a-1; tso2-1/+ plants, which are phenotypically
normal, were identified from the same cross described above by PCR-based genotyping
of normal-looking plants in the F2.  tso2-1 r2a-1 double mutants were obtained from the
self progeny of  tso2-1/+; r2a-1/r2a-1plants.  tso2-5 was also crossed with r2a r2b
double mutants. The F1 heterozygous progeny for all three mutations is wild type. tso2-5
r2b was identified from the F2 generation. tso2-5/tso2-5 r2a/+ plants display a tso2-1-
like phenotype. tso2-5 r2a double mutants were identified from the self progeny of tso2-
5/tso2-5 r2a/+ plants.
Microscopic and histological analysis
For scanning electron microscopy (SEM), samples were fixed, coated, dissected
and photographed as described previously (Bowman et al, 1989). Images were directly
captured with the semicaps software and the AMRAY 1000A scanning electron
microscope.  Whole mount floral photomicrographs were taken through a Zeiss Stemi
SV6 dissecting microscope.  Slides containing longitudinal sections of inflorescences
from in situ hybridization experiments were examined and photographed under a NIKON
116
ECL1PSE E600W microscope with Nomarski optics equipped with a digital still camera
DXM1200.
For thin tissue sections, leaves, inflorescences and siliques were fixed for 3 hr at 4
?C in 4% glutaraldehyde, 50 mM NaPO4, pH 7.0, then rinsed in 50 mM NaPO4, pH 7.0
overnight.  Some siliques were cut on both sides of the septum so that individual green or
white valves were fixed separately. After through ethanol series, the tissues were
embedded in JB-4 according to manufacturer?s instruction (Electron Microscopy
Sciences, catalog 14270-000). Two ?m sections were stained in 0.1% Toluidine Blue in
0.1% sodium borate.
Map-based cloning of TSO2
tso2-1 (in Ler background) was crossed to sup-2 (in Col background). A total of
358 F
2 
tso2 plants were assayed individually for linkage to various CAPS and dCAPS
markers (Appendix Table A1), which mapped TSO2 to a 20-kb region within the P1
clone, MOJ10.  A cosmid library was constructed from MOJ10 using the binary vector
pCLD04541. The procedure is essentially the same as in Bent et al. (1994).  Two
overlapping cosmids, D and G, were transformed into tso2-1 mutants using the floral dip
method.  All eight transgenic plants harboring Cosmid D and none of the five transgenic
plants harboring cosmid G are wild-type in phenotype, suggesting that cosmid D, not G
contains the TSO2 gene.  Cosmid D showed 100% co-segregation with the rescued
phenotype in T2 transgenic plants.  Sequencing of the four candidate genes At3g27040,
At3g27050, At3g27060 and At3g27070 in cosmid D only identified a mutation in
At3g27060 in tso2-1, suggesting that At3g27060 is TSO2.
117
Molecular analyses of TSO2, R2A and R2B
p35S::R2B was constructed by PCR-amplifying the ORF using Ler genomic
DNA as a template.  The PCR fragment was subsequently inserted into the pBI121 vector
(Clonetech).  The R2B primer pairs are 5? GCTCTAGACGATTAAATCTTCTTCACCG
GA 3? and 5? CGGGATCCGGCTCCAATCCTTTTTGGAT 3?.  Xba I and BamH I sites
were introduced into 5? and 3? primers respectively.   For the pTSO2::GUS reporter
construct, a 2.1 kb BamH I/EcoR I fragment containing the GUS ORF and a 3? Nos was
excised from pBI121 and inserted to the BamH I/EcoR I site of pBIN20 (Hennegan and
Danna, 2000) to create the pBIN20GUS vector.  PCR-amplification of 1.2 kb promoter
sequence of TSO2 was conducted with following forward primer pair: 5? GCTCTAGAA
TAAGGCCCTGTTCGTTTCC3? and 5?CGGGATCCGAATCTGTCTG GGGTTGGT
G 3?.  PCR products were digested with BamH I/Xba I and inserted into BamH I/Xba I-
digested pBIN20GUS vector.  All PCR was performed with high fidelity DNA
Polymerase Pwo or Tgo (Roche Molecular Biochemicals).  GUS staining was according
to Parcy et al., (1998).  GUS stained tissues were embedded in paraplast and sectioned
according to Parcy et al.
For semi-quantitative RT-PCR, total RNA was isolated with Tri-Reagent (Sigma)
from one-week old roots and seedlings, two-week-old rosette leaves, four-week-old
cauline leaves, stems and inflorescences.  Two ?g of total RNAs were used to synthesize
cDNA with oligo (dT) and Superscript II reverse transcriptase (Invitrogen).  PCR was
conducted with 1 ?l of diluted RT reaction at 94?C, 20 sec; 57 ?C, 20 sec; 72 ?C, 40 sec
118
for 20-25 cycles. The primers for TSO2, R2A and R2B were listed in the Appendix,
Table A2. The PCR products were Southern blotted and hybridized with a R2B specific
probe.
To assay the induction of cell death-related markers, DNA damage-inducible
genes and transcriptional silence information (TSI), similar RNA isolation, RT-reaction,
20-25 cycles of PCR were used.  Primers for each of the genes are listed in the Appendix,
Table A2.
In situ hybridization was essentially the same as previously described (Liu et al.,
2000) except that the RNA probes were not hydrolyzed. The TSO2 EST clone
(RZL13g10F) was linearized with Hind III and served as a template for the TSO2
antisense probe using a T7 promoter.
DNA damage induction of RNR expression
Cell line T87 was a gift from Dr. Rebecca Stevens (Mariconti et al., 2002) and
was subcultured weekly in B5 Gamborg?s medium (Sigma), pH 5.8, supplemented with
3% sucrose and 1 ?M NAA at 25?C in darkness. Four-day-old cell suspensions were
treated with 20 mM HU, 0.01% MMS and 2 ?g/ml 4-NQO, respectively. At different
time points, cells were harvested and quickly frozen in liquid nitrogen before isolation of
total RNA. Total RNA was isolated with Tri-Reagent (Sigma) and 2 ?g total RNA was
used for cDNA synthesis primed by oligo (dT)
18
. RT-PCR was performed with 11 cycles
for TSO2, 13 cycles for R2A, R1 and Actin2, 17 cycles for R2B. PCR products were
119
Southern blotted and hybridized with gene-specific probes (developed from 3?UTR
region).
Measurement of dNTP pool levels
dNTPs pool was measured by a polymerase-based assay (Roy, Beuneu et al.
1999). Inflorescences with unopened flower buds were harvested in liquid nitrogen,
ground to fine powder, weighed, and extracted with 60% cold methanol by vigorous
vortex. The extracts were heated at 95
o
C for 5 minutes and centrifuged at 17,000g for 15
minutes. The supernatants were dried in a Speedvac, resuspended in sterile distilled
water, and stored at ?20
o
C.  1 ?l of each sample was used for the polymerase assay.
Commercial dNTPs were tested in parallel to establish a linear calibration curve.  As the
internal control, a portion of the ground tissues was assayed for alcohol dehydrogenase
(ADH) activity (Russell, Wong et al. 1990).  The ADH activity per gram sample was then
used to normalize the dNTP levels for each gram of ground tissues.
Comet assay
A cometAssay kit from Trevigen (Gaithersburg, MD) was used with minor
modifications. One to two week-old seedlings were chopped with a razor in a Petri dish
that was kept on ice and contained 500 ?l 1 ? PBS plus 20 mM EDTA.  The resulting
mixture was filtered through a 60 ?m nylon mesh.  40 ?l nuclei were mixed with 400 ?l
1% low-melting agarose (pre-warmed at 37?C) and placed onto Trevigen pre-coated
slides. After incubating in Lysis Solution (2.5 M NaCl, 100 mM EDTA pH 10, 10 mM
Tris, 1% Sodium Lauryl Sarcosinate and 1% Triton X-100) for 1 hour at 4?C, the nuclei
120
on slides were unwound in Alkaline Solution (0.3 N NaOH, 5 mM EDTA) for 40 minutes
and neutralized 2-3 times in 1 ? TBE for 5 minutes.  The slides were run at 1 V/cM for
10 minutes in 1 ? TBE and then dipped in 70% ethanol for 5 minutes. After air dry, the
slides were stained with a 1:10,000 dilution of SYBR Green.  Anti-fade solution (1% p-
Phenylenediamine dihydrochloride in 0.1 ? PBS and 90% glycerol) was added to slides
that were examined by epifluorescence microscopy.  The percentage of DNA in each
comet tail (T DNA%) was evaluated by the Comet Score software
(http://www.autocomet.com) and assigned with a number (0, 1, 2, 3, or 4) with a higher
number corresponding to a higher T DNA% (Collins, Dusinska et al. 1997). DNA
damage units for each genotype were derived by averaging the data from four slides.  Onr
hundred comets were scored per slide.
UV-C sensitivity assay
Ten-day-old seedlings were irradiated by UV-C at a distance of 5 centimeter with
different dosages using a Stratalinker 1800 (Stratagene).  Seedlings were then grown
under F40GO gold fluorescent lights for 5 days to avoid photoreactivation. Plants were
returned to normal lighting for 1-2 weeks before scoring survival rate and photography.
DNA damage sensitivity assay
Assays of hydroxyurea (HU), mitomycin C (MMC), and methyl methanesulfonate
(MMS) sensitivity are according to Masson et al (1997). Briefly, 5-day-old seedlings
were transferred into 24-well plates, each well containing 1 ml of 1 ? MS medium
(Sigma) with various concentrations of HU, MMC or MMS. Plates were placed under
normal growth conditions for one to two-weeks before photographs were taken.
121
Cell death assays
For trypan blue assay (Rate, Cuenca et al. 1999), 10-day old rosette leaves were
boiled in lactophenol containing 10 mg of trypan blue for 1 minute, cleared in alcoholic
lactophneol (95% ethanol:lactophenol at 2:1) for 2 minutes, washed in 50% ethanol and
stored in water.  For aniline blue staining of callose (Rate, Cuenca et al. 1999), 10-day
old rosette leaves were boiled for 2 minutes in alcoholic lactophenol, rinsed in 50%
ethanol and then in water. Cleared and rinsed leaves were stained for 1 hour at room
teperature in a solution of 0.05% aniline blue in 0.15 M K
2
HPO4. Stained leaves were
examined under ultraviolet epifluorescence.  DAB-uptake method (Thordal-Christensen,
Zhang et al. 1997) was used to detect H
2
O
2
. Seedlings were placed in 1 mg/ml 3,3?-
diaminobenzidine (DAB) in 10 mM ascorbic acid for 2 hours. The seedlings were then
boiled in 96% ethanol for 10 minutes and stored in 96% ethanol. H
2
O
2
 production is
visualized as a reddish-brown coloration.
Mutagenesis of tso2-1 seeds
About 5,000 tso2-1 seeds were treated with 0.1% Tween-20 for 15 min and then
with 0.3% EMS over night. After the EMS was removed and seeds were rinsed with
water for 4 hrs, the seeds were resuspended in 50 ml 0.1% agar and dispersed into 50
planting pots. They were cold treated at 4?C for 3 days and then placed under 16 hr light
at 20?C. Seeds were pooled from M1 plants of each pot so that 50 M1 families were
collected. About 80 seeds from each of the M1 families were sown and screened for
suppressors of tso2-1. Those plants that displayed a wild type, or a very subtle tso2-1
phenotype, were scored as potential suppressors and seeds were collected from individual
lines.
122
Complementation test
The M3 potential suppressor lines were grown to flowering to confirm that they
consistently showed a wild type phenotype. After the phenotype was confirmed, their
genotype was verified with the dCAPS marker for tso2-1. During the first batch of
screening, 9 suppressor lines were isolated and randomly divided into 2 groups. Lines in
Group 1 were crossed with lines in Group 2 and each line was also backcrossed with wild
type (Ler). During the second round of screening, 9 more suppressor lines were obtained.
Two of them are dominant with novel phenotypes. The rest of 7 lines were again
randomly divided into two groups and crosses were made between these two groups. The
phenotypes of F1 plants from those crosses were recorded. If all the F1 plants displayed a
tso2-1 phenotype, then the two lines used for that cross were scored as ? complemented?,
that is, they belong to different complementation groups. If all the F1 plants exhibited a
wild type phenotype, then the two lines used for that cross were scored as ?not
complemented?, that is, they belong to the same complementation group and they are
allelic to each other.
123
Appendix
Marker nameRestrictionRestriction fragments Restriction fragmentsOligonucleotide sequence
 enzymefor Ler (bp)for Col (bp)
MQP15Hind III1128 871, 257Left: ATCTCCAATCTCAAATCTCC
Right: ATGTAGCACTAAGAAACCAGCC
MXE2Hinf I551, 322, 63614,  322Left: CGATAGCATAGATGTTTCAAGC
Righte: AAGGAAATACAACGCTTAGTCG
MMG15Sac I705,613,134839, 613Left: CGGCAGTGGTGTTGATTTCACG
Right: TACAACAAAACACAAAGCCC
MOJ10-0KBstU I1488 1244, 244Left: AGTAAGCCAGCGAAAAAGAAGG
Right: CACGACCGTTGTTCTTGAATGG
MOJ10-31K99 113Left: GGTTTAGCTTTAGACTCGGTTT
Right: AGGTCTAAGTTTTGAAATTTTATCAT
MOJ10-45KEcoR I611,  565, 111676,  611Left: CTGTGAAATCAGCCCCTCAT
Right: CGTTCGATTGGATGGAACTT
MOJ10-53K Taq I774, 473, 51774, 524Left: TCAGACCTCGGCACAGATAA
Right: CTGGTCTTGGTTCCATTCTAAA
MOJ10-73K Dde I652
a
, 567, 151612, 567, 151Left: GAGGCAACTTCACGACGAATGG
Right: TGGTGAGTCCATAACACAATCC
Marker nameRestrictionRestriciton fragmentsRestriciton fragmentsOligonucleotide sequence
 enzymefor Ler (bp)for mutant (bp)
tso2-1Mbo II62 90 Left: AAGAAGTAGATCTATCAGAA
Right: AGAAAGCGAGGACGTGTTTG
tso2-2Kpn I599, 5211120Left: AGCATCAAAACAAACGCTCT 
Right: AATCCCTTCAAAGATGCCTTC 
tso2-3Hha I
b
532, 300, 145 852, 145Left: same as above
Right: AACCTTAGAGAAGCATC
r2a-1PflM I785, 154939Left: TTTGCTGTGAGGCTGGTCGCTTTT 
Right: CTTCCAGATTCGATGGCGGATTCA 
r2b-CHae III600, 6001200Left: CGATTAAATCTTCTTCACCGGA
Right: GGCTCCAATCCTTTTTGGAT
a
There is about 40-bp insertion in Ler sequence.
b
Hha I produces three more smaller fragments (71,63,10) in both Ler and tso2-3.
Table A1. CAPS and dCAPS markers used in this study
124
GenesForward primerReverse primer
TSO2CATCTCTTTCGTGCCATTGATCGCAGACGATTGATTTCAC
R2AACTGATGTGCAACAGTGGGATCCTTCTCCTTCGAGTCCTTT
R2BCGTTTGAAAGCGATTGTTTGTGATCATCATTAAACGCAGC
R1 AGGAACTCCAAGGCCTCAATCCCTGTGTTCTTCCTTTCCA
Actin2GTTGGGATGAACCAGAAGGACTTACAATTTCCCGCTCTTC
EDS1GGAGCAGTCGTAATCTTCGCCCAAATGTCACACAACGAGG
GST6TCAAGCTCGTCTTCCCTTGTGGTTGCCTTGACTTTCTTGC
SAG12CCAAACTAAAATGTCGCCGTTTGACTCAGTTGTCAAGCCG
PR-5TCGTCTTCATCACAAGTGGCTCGGATCCATGATCCTAAGC
PRXcbAGTTAAGGTCGGACCCTCGTCTCTCCTTCCCAAAGGAACC
SAG13CTCTCGTGACCAACGAGTGATCATTTGCTTCTCCAACACG
GST11GCAGGAATCAAAGTTTTCGGGGAGCCAAGGGAGACAAGTT
PR-1GTGCTCTTGTTCTTCCCTCGACTTTGGCACATCCGAGTCT
TSIGAACTCATGGATACCCTAAAATACCTCTACCCTTTGCATTCATGAATC
Rad51GGCCATGTACATTGATGCTGCCAGCAAAAAGAGCTGAACC
PARP-1CAAATTCAAGAGCAGGCACACATGGCATCGGTAAAGTCCT
PARP-2AACGAGGCTTGAGGAACTGACGTATATCCCGAATGCGTCT
Table A2. Gene-specific rimers used for RT-PCR
125
Bibliography
Angelis,, K.J., Dusinska, M. and Collins, A.R. (1999) Single cell gel electrophoresis:
detection of DNA damage at different levels of sensitivity. Electrophoresis, 20,
2133-2138.
Angus, S. P., Wheeler, L. J., Ranmal, S. A., Zhang, X., Markey, M. P., Mathews, C.
K. and Knudsen, E. S. (2002). Retinoblastoma tumor suppressor targets dNTP
metabolism to regulate DNA replication. J Biol Chem 277, 44376-84.
Aono, N., Sutani, T., Tomonaga, T., Mochida, S. and Yanagida, M. (2002). Cnd2 has
dual roles in mitotic condensation and interphase. Nature 417, 197-202.
The Arabidopsis Genome Initiative. (2000). Analysis of the genome sequence of the
flowering plant Arabidopsis thaliana. Nature. 408, 796-815.
Beall, E. L., Bell, M., Georlette, D. and Botchan, M. R. (2004). Dm-myb mutant
lethality in Drosophila is dependent upon mip130: positive and negative
regulation of DNA replication. Genes Dev 18, 1667-80.
Beall, E. L., Manak, J. R., Zhou, S., Bell, M., Lipsick, J. S. and Botchan, M. R.
(2002). Role for a Drosophila Myb-containing protein complex in site-specific
DNA replication. Nature 420, 833-7
Blakley, R. L. and Barker, H. A. (1964). Cobamide stimulation of the reduction of
ribotides to deoxyribotides in Lactobacillus leichmannii. Biochem Biophys Res
Commun 16, 391-7.
Blilou, I., Frugier, F., Folmer, S., Serralbo, O., Willemsen, V., Wolkenfelt, H., Eloy,
N. B., Ferreira, P. C., Weisbeek, P. and Scheres, B. (2002). The Arabidopsis
HOBBIT gene encodes a CDC27 homolog that links the plant cell cycle to
progression of cell differentiation. Genes Dev 16, 2566-75.
Bowman, J.L., Smyth, D.R. and Meyerowitz. E.M. (1989). Genes directing flower
development in Arabidopsis. Plant Cell 1: 37-52.
Broadhvest, J., Baker, S. C. and Gasser, C. S. (2000). SHORT INTEGUMENTS 2
promotes growth during Arabidopsis reproductive development. Genetics 155, 899-907.
Bundock, P. and Hooykaas, P. (2002). Severe developmental defects, hypersensitivity
to DNA-damaging agents, and lengthened telomeres in Arabidopsis MRE11
mutants. Plant Cell 14, 2451-62.
Capron, A., Serralbo, O., Fulop, K., Frugier, F., Parmentier, Y., Dong, A.,
Lecureuil, A., Guerche, P., Kondorosi, E., Scheres, B. et al. (2003). The
126
Arabidopsis anaphase-promoting complex or cyclosome: molecular and genetic
characterization of the APC2 subunit. Plant Cell 15, 2370-82.
Carles, C. C. and Fletcher, J. C. (2003). Shoot apical meristem maintenance: the art of
a dynamic balance. Trends Plant Sci 8, 394-401.
Chabes, A., Domkin, V., Larsson, G., Liu, A., Graslund, A., Wijmenga, S. and
Thelander, L. (2000). Yeast ribonucleotide reductase has a heterodimeric iron-
radical-containing subunit. Proc Natl Acad Sci U S A 97, 2474-9.
Chabes, A., Domkin, V. and Thelander, L. (1999). Yeast Sml1, a protein inhibitor of
ribonucleotide reductase. J Biol Chem 274, 36679-83.
Chabes, A., Georgieva, B., Domkin, V., Zhao, X., Rothstein, R. and Thelander, L.
(2003a). Survival of DNA damage in yeast directly depends on increased dNTP
levels allowed by relaxed feedback inhibition of ribonucleotide reductase. Cell
112, 391-401.
Chabes, A. and Thelander, L. (2000). Controlled protein degradation regulates
ribonucleotide reductase activity in proliferating mammalian cells during the
normal cell cycle and in response to DNA damage and replication blocks. J Biol
Chem 275, 17747-53.
Chabes, A. L., Bjorklund, S. and Thelander, L. (2004). S Phase-specific transcription
of the mouse ribonucleotide reductase R2 gene requires both a proximal
repressive E2F-binding site and an upstream promoter activating region. J Biol
Chem 279, 10796-807.
Chabes, A. L., Pfleger, C. M., Kirschner, M. W. and Thelander, L. (2003b). Mouse
ribonucleotide reductase R2 protein: a new target for anaphase-promoting
complex-Cdh1-mediated proteolysis. Proc Natl Acad Sci U S A 100, 3925-9.
Chaboute, M. E., Clement, B. and Philipps, G. (2002). S phase and meristem-specific
expression of the tobacco RNR1b gene is mediated by an E2F element located in
the 5' leader sequence. J Biol Chem 277, 17845-51.
Chaboute, M. E., Clement, B., Sekine, M., Philipps, G. and Chaubet-Gigot, N.
(2000). Cell cycle regulation of the tobacco ribonucleotide reductase small
subunit gene is mediated by E2F-like elements. Plant Cell 12, 1987-2000.
Chaboute, M. E., Combettes, B., Clement, B., Gigot, C. and Philipps, G. (1998).
Molecular characterization of tobacco ribonucleotide reductase RNR1 and RNR2
cDNAs and cell cycle-regulated expression in synchronized plant cells. Plant Mol
Biol 38, 797-806.
127
Chellappan, S. P., Hiebert, S., Mudryj, M., Horowitz, J. M. and Nevins, J. R. (1991).
The E2F transcription factor is a cellular target for the RB protein. Cell 65, 1053-
61.
Chen, I. P., Haehnel, U., Altschmied, L., Schubert, I. and Puchta, H. (2003). The
transcriptional response of Arabidopsis to genotoxic stress - a high-density colony
array study (HDCA). Plant J 35, 771-86.
Chen, X. and Meyerowitz, E. M. (1999). HUA1 and HUA2 are two members of the
floral homeotic AGAMOUS pathway. Mol Cell 3, 349-60.
Chernova, O. B., Chernov, M. V., Agarwal, M. L., Taylor, W. R. and Stark, G. R.
(1995). The role of p53 in regulating genomic stability when DNA and RNA
synthesis are inhibited. Trends Biochem Sci 20, 431-4.
Clark, S. E., Jacobsen, S. E., Levin, J. Z. and Meyerowitz, E. M. (1996). The
CLAVATA and SHOOT MERISTEMLESS loci competitively regulate meristem
activity in Arabidopsis. Development 122, 1567-75.
Clark, S. E., Running, M. P. and Meyerowitz, E. M. (1993). CLAVATA1, a regulator
of meristem and flower development in Arabidopsis. Development 119, 397-418.
Clark, S. E., Running, M. P. and Meyerowitz, E. M. (1995). CLAVATA3 is a specific
regulator of shoot and floral meristem development affecting the same processes
as CLAVATA1. Development 121, 2057-2067
Clark, S. E., Williams, R. W. and Meyerowitz, E. M. (1997). The CLAVATA1 gene
encodes a putative receptor kinase that controls shoot and floral meristem size in
Arabidopsis. Cell 89, 575-85.
Climent, I., Sjoberg, B. M. and Huang, C. Y. (1992). Site-directed mutagenesis and
deletion of the carboxyl terminus of Escherichia coli ribonucleotide reductase
protein R2. Effects on catalytic activity and subunit interaction. Biochemistry 31,
4801-7.
Cockcroft, C. E., den Boer, B. G., Healy, J. M. and Murray, J. A. (2000). Cyclin D
control of growth rate in plants. Nature 405, 575-9.
Culligan, K., Tissier, A. and Britt, A. (2004). ATR regulates a G2-phase cell-cycle
checkpoint in Arabidopsis thaliana. Plant Cell 16, 1091-104.
D'Amours, D. and Jackson, S. P. (2002). The Mre11 complex: at the crossroads of dna
repair and checkpoint signalling. Nat Rev Mol Cell Biol 3, 317-27.
Danon, A., Delorme, V., Mailhac, N., and Gallois, P. (2000). Plant programmed cell
death: a common way to die. Plant Physiol Biochem 38, 647-655.
128
De Veylder, L., Beeckman, T., Beemster, G. T., de Almeida Engler, J., Ormenese, S.,
Maes, S., Naudts, M., Van Der Schueren, E., Jacqmard, A., Engler, G. et al.
(2002). Control of proliferation, endoreduplication and differentiation by the
Arabidopsis E2Fa-DPa transcription factor. Embo J 21, 1360-8.
De Veylder, L., T. Beeckman, G. T. Beemster, L. Krols, F. Terras, I. Landrieu, E.
van der Schueren, S. Maes, M. Naudts and Inze, D. (2001). Functional analysis
of cyclin-dependent kinase inhibitors of Arabidopsis. Plant Cell 13, 1653-68.
De Veylder, L., Beemster, G. T., Beeckman, T. and Inze, D. (2001). CKS1At
overexpression in Arabidopsis thaliana inhibits growth by reducing meristem size
and inhibiting cell-cycle progression. Plant J 25, 617-26.
den Boer, B. G. and Murray, J. A. (2000). Control of plant growth and development
through manipulation of cell-cycle genes. Curr Opin Biotechnol 11, 138-45.
Desany, B. A., Alcasabas, A. A., Bachant, J. B. and Elledge, S. J. (1998). Recovery
from DNA replicational stress is the essential function of the S-phase checkpoint
pathway. Genes Dev 12, 2956-70.
Dewitte, W. and Murray, J. A. (2003). The plant cell cycle. Annu Rev Plant Biol 54,
235-64.
Dewitte, W., Riou-Khamlichi, C., Scofield, S., Healy, J. M., Jacqmard, A., Kilby, N.
J. and Murray, J. A. (2003). Altered cell cycle distribution, hyperplasia, and
inhibited differentiation in Arabidopsis caused by the D-type cyclin CYCD3.
Plant Cell 15, 79-92.
Di Leonardo, A., Linke, S. P., Yin, Y. and Wahl, G. M. (1993). Cell cycle regulation
of gene amplification. Cold Spring Harb Symp Quant Biol 58, 655-67.
Doerner, P., Jorgensen, J. E., You, R., Steppuhn, J. and Lamb, C. (1996). Control of
root growth and development by cyclin expression. Nature 380, 520-3.
Domkin, V., Thelander, L. and Chabes, A. (2002). Yeast DNA damage-inducible Rnr3
has a very low catalytic activity strongly stimulated after the formation of a cross-
talking Rnr1/Rnr3 complex. J Biol Chem 277, 18574-8.
Downes, C. S., Bachrati, C. Z., Devlin, S. J., Tommasino, M., Cutts, T. J., Watson, J.
V., Rasko, I. and Johnson, R. T. (2000). Mammalian S-phase checkpoint
integrity is dependent on transformation status and purine deoxyribonucleosides. J
Cell Sci 113 ( Pt 6), 1089-96.
Durocher, D., Henckel, J., Fersht, A. R. and Jackson, S. P. (1999). The FHA domain
is a modular phosphopeptide recognition motif. Mol Cell 4, 387-94.
129
Ebel, C., Mariconti, L. and Gruissem, W. (2004). Plant retinoblastoma homologues
control nuclear proliferation in the female gametophyte. Nature 429, 776-80.
Ehrenberg, A. and Reichard, P. (1972). Electron spin resonance of the iron-containing
protein B2 from ribonucleotide reductase. J Biol Chem 247, 3485-8.
Eklund, H., Uhlin, U., Farnegardh, M., Logan, D. T. and Nordlund, P. (2001).
Structure and function of the radical enzyme ribonucleotide reductase. Prog
Biophys Mol Biol 77, 177-268.
Elledge, S. J. (1996). Cell cycle checkpoints: preventing an identity crisis. Science 274,
1664-72.
Elledge, S. J. and Davis, R. W. (1989). DNA damage induction of ribonucleotide
reductase. Mol Cell Biol 9, 4932-40.
Elledge, S. J. and Davis, R. W. (1990). Two genes differentially regulated in the cell
cycle and by DNA-damaging agents encode alternative regulatory subunits of
ribonucleotide reductase. Genes Dev 4, 740-51.
Elledge, S. J., Zhou, Z. and Allen, J. B. (1992). Ribonucleotide reductase: regulation,
regulation, regulation. Trends Biochem Sci 17, 119-23.
Elledge, S. J., Zhou, Z., Allen, J. B. and Navas, T. A. (1993). DNA damage and cell
cycle regulation of ribonucleotide reductase. Bioessays 15, 333-9.
Encalada, S. E., Martin, P. R., Phillips, J. B., Lyczak, R., Hamill, D. R., Swan, K. A.
and Bowerman, B. (2000). DNA replication defects delay cell division and
disrupt cell polarity in early Caenorhabditis elegans embryos. Dev Biol 228, 225-
38.
Fan, H., Huang, A., Villegas, C. and Wright, J. A. (1997). The R1 component of
mammalian ribonucleotide reductase has malignancy-suppressing activity as
demonstrated by gene transfer experiments. Proc Natl Acad Sci U S A 94, 13181-
6.
Fan, H., Villegas, C. and Wright, J. A. (1996). Ribonucleotide reductase R2 component
is a novel malignancy determinant that cooperates with activated oncogenes to
determine transformation and malignant potential. Proc Natl Acad Sci U S A 93,
14036-40.
Filatov, D., Bjorklund, S., Johansson, E. and Thelander, L. (1996). Induction of the
mouse ribonucleotide reductase R1 and R2 genes in response to DNA damage by
UV light. J Biol Chem 271, 23698-704.
130
Fletcher, J. C., Brand, U., Running, M. P., Simon, R. and Meyerowitz, E. M. (1999).
Signaling of cell fate decisions by CLAVATA3 in Arabidopsis shoot meristems.
Science 283, 1911-4.
Fontecave, M., Eliasson, R. and Reichard, P. (1989). Oxygen-sensitive ribonucleoside
triphosphate reductase is present in anaerobic Escherichia coli. Proc Natl Acad
Sci U S A 86, 2147-51.
Girling, R., Bandara, L. R., Zamanian, M., Sorensen, T. S., Xu, F. H. and La
Thangue, N. B. (1993). DRTFI/E2F transcription factor: an integrator of cell-
cycle events with the transcriptional apparatus. Biochem Soc Trans 21, 939-42.
Gisel, A., Barella, S., Hempel, F. D. and Zambryski, P. C. (1999). Temporal and
spatial regulation of symplastic trafficking during development in Arabidopsis
thaliana apices. Development 126, 1879-89.
Grandjean, O., Vernoux, T., Laufs, P., Belcram, K., Mizukami, Y. and Traas, J.
(2004). In vivo analysis of cell division, cell growth, and differentiation at the
shoot apical meristem in Arabidopsis. Plant Cell 16, 74-87.
Greenberg, J. T. (1996). Programmed cell death: a way of life for plants. Proc Natl
Acad Sci U S A 93, 12094-7.
Haecker, A. and Laux, T. (2001). Cell-cell signaling in the shoot meristem. Curr Opin
Plant Biol 4, 441-6.
Hemerly, A., Engler Jde, A., Bergounioux, C., Van Montagu, M., Engler, G., Inze,
D. and Ferreira, P. (1995). Dominant negative mutants of the Cdc2 kinase
uncouple cell division from iterative plant development. Embo J 14, 3925-36.
Hemerly, A. S., Ferreira, P., de Almeida Engler, J., Van Montagu, M., Engler, G.
and Inze, D. (1993). cdc2a expression in Arabidopsis is linked with competence
for cell division. Plant Cell 5, 1711-23.
Hemerly, A.S., Ferreira, P.C.G., Van Monague, M., and Inze, D. (1999). Cell cycle
     control and plant morphogenesis: is there an essential link? BioEssays 21, 29-37.
Hemerly, A. S., Ferreira, P. C., Van Montagu, M., Engler, G. and Inze, D. (2000).
Cell division events are essential for embryo patterning and morphogenesis:
studies on dominant-negative cdc2aAt mutants of arabidopsis. Plant J 23, 123-30.
Hennegan, K.P., and Danna, K.J. (1998). pBIN20: an improved binary vector for
Agrobacterium-mediated transformation. Plant Mol Biol Rep 16, 129-131.
131
Huang, A., Fan, H., Taylor, W. R. and Wright, J. A. (1997). Ribonucleotide reductase
R2 gene expression and changes in drug sensitivity and genome stability. Cancer
Res 57, 4876-81.
Huang, M. and Elledge, S. J. (1997). Identification of RNR4, encoding a second
essential small subunit of ribonucleotide reductase in Saccharomyces cerevisiae.
Mol Cell Biol 17, 6105-13.
Huang, M., Zhou, Z. and Elledge, S. J. (1998). The DNA replication and damage
checkpoint pathways induce transcription by inhibition of the Crt1 repressor. Cell
94, 595-605.
Huh, J. R., M. Guo and Hay, B.A. (2004). Compensatory proliferation induced by cell
death in the Drosophila wing disc requires activity of the apical cell death caspase
Dronc in a nonapoptotic role. Curr Biol 14, 1262-6
Jeong, S., Trotochaud, A. E. and Clark, S. E. (1999). The Arabidopsis CLAVATA2
gene encodes a receptor-like protein required for the stability of the CLAVATA1
receptor-like kinase. Plant Cell 11, 1925-34.
Johansson, E., Hjortsberg, K. and Thelander, L. (1998). Two YY-1-binding proximal
elements regulate the promoter strength of the TATA-less mouse ribonucleotide
reductase R1 gene. J Biol Chem 273, 29816-21.
Jordan, A. and Reichard, P. (1998). Ribonucleotide reductases. Annu Rev Biochem 67,
71-98.
Kaya, H., Shibahara, K. I., Taoka, K. I., Iwabuchi, M., Stillman, B. and Araki, T.
(2001). FASCIATA genes for chromatin assembly factor-1 in arabidopsis
maintain the cellular organization of apical meristems. Cell 104, 131-42.
Kayes, J. M. and Clark, S. E. (1998). CLAVATA2, a regulator of meristem and organ
development in Arabidopsis. Development 125, 3843-51.
Kimura, T., Takeda, S., Sagiya, Y., Gotoh, M., Nakamura, Y. and Arakawa, H.
(2003). Impaired function of p53R2 in Rrm2b-null mice causes severe renal
failure through attenuation of dNTP pools. Nat Genet 34, 440-5.
King, D. S. and Reichard, P. (1995). Mass spectrometric determination of the radical
scission site in the anaerobic ribonucleotide reductase of Escherichia coli.
Biochem Biophys Res Commun 206, 731-5.
Kolberg, M., Strand, K. R., Graff, P. and Andersson, K. K. (2004). Structure,
function, and mechanism of ribonucleotide reductases. Biochim Biophys Acta
1699, 1-34.
132
Kosugi, S. and Ohashi, Y. (2003). Constitutive E2F expression in tobacco plants
exhibits altered cell cycle control and morphological change in a cell type-specific
manner. Plant Physiol 132, 2012-22.
Kuo, M. T., Vyas, R. C., Jiang, L. X. and Hittelman, W. N. (1994). Chromosome
breakage at a major fragile site associated with P-glycoprotein gene amplification
in multidrug-resistant CHO cells. Mol Cell Biol 14, 5202-11.
Lam, E., N. Kato and Lawton, M. (2001). Programmed cell death, mitochondria and
the plant hypersensitive response. Nature 411, 848-53.
Laufs, P., Grandjean, O., Jonak, C., Kieu, K. and Traas, J. (1998). Cellular
parameters of the shoot apical meristem in Arabidopsis. Plant Cell 10, 1375-90.
Laux, T., Mayer, K. F., Berger, J. and Jurgens, G. (1996). The WUSCHEL gene is
required for shoot and floral meristem integrity in Arabidopsis. Development 122,
87-96.
Lazar, G., Zhang, H. and Goodman, H. M. (1993). The origin of the bifunctional
dihydrofolate reductase-thymidylate synthase isogenes of Arabidopsis thaliana.
Plant J 3, 657-68.
Lecrenier, N. and Foury, F. (1995). Overexpression of the RNR1 gene rescues
Saccharomyces cerevisiae mutants in the mitochondrial DNA polymerase-
encoding MIP1 gene. Mol Gen Genet 249, 1-7.
Lee, Y., Vassilakos, A., Feng, N., Lam, V., Xie, H., Wang, M., Jin, H., Xiong, K., Liu,
C., Wright, J. et al. (2003). GTI-2040, an antisense agent targeting the small
subunit component (R2) of human ribonucleotide reductase, shows potent
antitumor activity against a variety of tumors. Cancer Res 63, 2802-11.
Lenhard, M., Jurgens, G. and Laux, T. (2002). The WUSCHEL and
SHOOTMERISTEMLESS genes fulfil complementary roles in Arabidopsis shoot
meristem regulation. Development 129, 3195-206.
Lenhard, M. and Laux, T. (1999). Shoot meristem formation and maintenance. Curr
Opin Plant Biol 2, 44-50.
Lenhard, M. and Laux, T. (2003). Stem cell homeostasis in the Arabidopsis shoot
meristem is regulated by intercellular movement of CLAVATA3 and its
sequestration by CLAVATA1. Development 130, 3163-73.
Leyser, H.M.O., and Furner, I.J. 1992. Characterisation of three shoot apical meristem
mutants of Arabidopsis thaliana. Development 116, 397?403
133
Lin, Z. P., Belcourt, M. F., Cory, J. G. and Sartorelli, A. C. (2004). Stable suppression
of the R2 subunit of ribonucleotide reductase by R2-targeted short interference
RNA sensitizes p53(-/-) HCT-116 colon cancer cells to DNA-damaging agents
and ribonucleotide reductase inhibitors. J Biol Chem 279, 27030-8.
Lincker, F., Philipps, G. and Chaboute, M. E. (2004). UV-C response of the
ribonucleotide reductase large subunit involves both E2F-mediated gene
transcriptional regulation and protein subcellular relocalization in tobacco cells.
Nucleic Acids Res 32, 1430-8.
Liu, C., Powell, K. A., Mundt, K., Wu, L., Carr, A. M. and Caspari, T. (2003).
Cop9/signalosome subunits and Pcu4 regulate ribonucleotide reductase by both
checkpoint-dependent and -independent mechanisms. Genes Dev 17, 1130-40.
Liu, Z., Running, M. P. and Meyerowitz, E. M. (1997). TSO1 functions in cell division
during Arabidopsis flower development. Development 124, 665-72.
Lowndes, N. F. and Murguia, J. R. (2000). Sensing and responding to DNA damage.
Curr Opin Genet Dev 10, 17-25.
Lukas, J., Petersen, B. O., Holm, K., Bartek, J. and Helin, K. (1996). Deregulated
expression of E2F family members induces S-phase entry and overcomes
p16INK4A-mediated growth suppression. Mol Cell Biol 16, 1047-57.
Ma, C., Martin, S., Trask, B. and Hamlin, J. L. (1993). Sister chromatid fusion
initiates amplification of the dihydrofolate reductase gene in Chinese hamster
cells. Genes Dev 7, 605-20.
Mandel, H., Szargel, R., Labay, V., Elpeleg, O., Saada, A., Shalata, A., Anbinder, Y.,
Berkowitz, D., Hartman, C., Barak, M. et al. (2001). The deoxyguanosine
kinase gene is mutated in individuals with depleted hepatocerebral mitochondrial
DNA. Nat Genet 29, 337-41.
Marenda, D. R., Zraly, C. B., Feng, Y., Egan, S. and Dingwall, A. K. (2003). The
Drosophila SNR1 (SNF5/INI1) subunit directs essential developmental functions
of the Brahma chromatin remodeling complex. Mol Cell Biol 23, 289-305.
Maser, R. S., Mirzoeva, O. K., Wells, J., Olivares, H., Williams, B. R., Zinkel, R. A.,
Farnham, P. J. and Petrini, J. H. (2001). Mre11 complex and DNA replication:
linkage to E2F and sites of DNA synthesis. Mol Cell Biol 21, 6006-16.
Mathews, C. K. and Ji, J. (1992). DNA precursor asymmetries, replication fidelity, and
variable genome evolution. Bioessays 14, 295-301.
134
Mayer, K. F., Schoof, H., Haecker, A., Lenhard, M., Jurgens, G. and Laux, T.
(1998). Role of WUSCHEL in regulating stem cell fate in the Arabidopsis shoot
meristem. Cell 95, 805-15.
Menke, M., Chen, I., Angelis, K. J. and Schubert, I. (2001). DNA damage and repair
in Arabidopsis thaliana as measured by the comet assay after treatment with
different classes of genotoxins. Mutat Res 493, 87-93.
Meyerowitz, E. M. (1996). Plant development: local control, global patterning. Curr
Opin Genet Dev 6, 475-9.
Milan, M., S. Campuzano and Garcia-Bellido, A.  (1997). Developmental parameters
of cell death in the wing disc of Drosophila. Proc Natl Acad Sci U S A 94, 5691-6
Mironov, V. V., De Veylder, L., Van Montagu, M. and Inze, D. (1999). Cyclin-
dependent kinases and cell division in plants- the nexus. Plant Cell 11, 509-22.
Moussian, B., Schoof, H., Haecker, A., Jurgens, G. and Laux, T. (1998). Role of the
ZWILLE gene in the regulation of central shoot meristem cell fate during
Arabidopsis embryogenesis. Embo J 17, 1799-809.
Muller, H. J. (1932). Further studies on the nature and causes of gene mutations. Proc.
6th Int. Cong. Genetics, Brooklyn Botanic Gardens USA 1, 213-255.
Muller, H., Bracken, A. P., Vernell, R., Moroni, M. C., Christians, F., Grassilli, E.,
Prosperini, E., Vigo, E., Oliner, J. D. and Helin, K. (2001). E2Fs regulate the
expression of genes involved in differentiation, development, proliferation, and
apoptosis. Genes Dev 15, 267-85.
Nasmyth, K. (1996). Viewpoint: putting the cell cycle in order. Science 274, 1643-5.
Navas, T. A., Sanchez, Y. and Elledge, S. J. (1996). RAD9 and DNA polymerase
epsilon form parallel sensory branches for transducing the DNA damage
checkpoint signal in Saccharomyces cerevisiae. Genes Dev 10, 2632-43.
Neufeld, T. P., A. F. de la Cruz, L. A. Johnston and Edgar, B. A.  (1998).
Coordination of growth and cell division in the Drosophila wing. Cell 93, 1183-
93.
Nevins, J. R. (1992). E2F: a link between the Rb tumor suppressor protein and viral
oncoproteins. Science 258, 424-9.
Noh, Y. S. and Amasino, R. M. (1999). Identification of a promoter region responsible
for the senescence-specific expression of SAG12. Plant Mol Biol 41, 181-94.
135
Oliver, F. J., Collins, M. K. and Lopez-Rivas, A. (1996a). dNTP pools imbalance as a
signal to initiate apoptosis. Experientia 52, 995-1000.
Ollagnier, S., Mulliez, E., Gaillard, J., Eliasson, R., Fontecave, M. and Reichard, P.
(1996). The anaerobic Escherichia coli ribonucleotide reductase. Subunit structure
and iron sulfur center. J Biol Chem 271, 9410-6.
Ouellette, M. M., Chen, J., Wright, W. E. and Shay, J. W. (1992). Complexes
containing the retinoblastoma gene product recognize different DNA motifs
related to the E2F binding site. Oncogene 7, 1075-81.
Pardee, A. B. (1989). G1 events and regulation of cell proliferation. Science 246, 603-8.
Pelaz, S., Ditta, G. S., Baumann, E., Wisman, E. and Yanofsky, M. F. (2000). B and
C floral organ identity functions require SEPALLATA MADS-box genes. Nature
405, 200-3.
Pellicioli, A., Lucca, C., Liberi, G., Marini, F., Lopes, M., Plevani, P., Romano, A.,
Di Fiore, P. P. and Foiani, M. (1999). Activation of Rad53 kinase in response to
DNA damage and its effect in modulating phosphorylation of the lagging strand
DNA polymerase. Embo J 18, 6561-72.
Pennell, R. I. and Lamb, C.  (1997). Programmed Cell Death in Plants. Plant Cell 9,
1157-1168.
Philipps, G., Clement, B. and Gigot, C. (1995). Molecular characterization and cell
cycle-regulated expression of a cDNA clone from Arabidopsis thaliana
homologous to the small subunit of ribonucleotide reductase. FEBS Lett 358, 67-
70.
Porceddu, A., Stals, H., Reichheld, J. P., Segers, G., De Veylder, L., Barroco, R. P.,
Casteels, P., Van Montagu, M., Inze, D. and Mironov, V. (2001). A plant-
specific cyclin-dependent kinase is involved in the control of G2/M progression in
plants. J Biol Chem 276, 36354-60.
Ramirez-Parra, E., Frundt, C. and Gutierrez, C. (2003). A genome-wide
identification of E2F-regulated genes in Arabidopsis. Plant J 33, 801-11.
Reichard, P. (1988). Interactions between deoxyribonucleotide and DNA synthesis.
Annu Rev Biochem 57, 349-74.
Ren, B., Cam, H., Takahashi, Y., Volkert, T., Terragni, J., Young, R. A. and
Dynlacht, B. D. (2002). E2F integrates cell cycle progression with DNA repair,
replication, and G(2)/M checkpoints. Genes Dev 16, 245-56.
136
Reynolds, J. O., Eisses, J. F. and Sylvester, A. W. (1998). Balancing division and
expansion during maize leaf morphogenesis: analysis of the mutant, warty-1.
Development 125, 259-68.
Rinne, P. L. and van der Schoot, C. (1998). Symplasmic fields in the tunica of the
shoot apical meristem coordinate morphogenetic events. Development 125, 1477-
85.
Riou-Khamlichi, C., Huntley, R., Jacqmard, A. and Murray, J. A. (1999). Cytokinin
activation of Arabidopsis cell division through a D-type cyclin. Science 283,
1541-4.
Riou-Khamlichi, C., Menges, M., Healy, J. M. and Murray, J. A. (2000). Sugar
control of the plant cell cycle: differential regulation of Arabidopsis D-type cyclin
gene expression. Mol Cell Biol 20, 4513-21.
Roy, B., Beuneu, C., Roux, P., Buc, H., Lemaire, G. and Lepoivre, M. (1999).
Simultaneous determination of pyrimidine or purine deoxyribonucleoside
triphosphates using a polymerase assay. Anal Biochem 269, 403-9.
Saada, A., Shaag, A., Mandel, H., Nevo, Y., Eriksson, S. and Elpeleg, O. (2001).
Mutant mitochondrial thymidine kinase in mitochondrial DNA depletion
myopathy. Nat Genet 29, 342-4.
Saitoh, S., Chabes, A., McDonald, W. H., Thelander, L., Yates, J. R. and Russell, P.
(2002). Cid13 is a cytoplasmic poly(A) polymerase that regulates ribonucleotide
reductase mRNA. Cell 109, 563-73.
Sancar, A., Lindsey-Boltz, L. A., Unsal-Kaccmaz, K. and Linn, S. (2004). Molecular
mechanisms of mammalian DNA repair and the DNA damage checkpoints. Annu
Rev Biochem 73, 39-85.
Sanchez, Y., Desany, B. A., Jones, W. J., Liu, Q., Wang, B. and Elledge, S. J. (1996).
Regulation of RAD53 by the ATM-like kinases MEC1 and TEL1 in yeast cell
cycle checkpoint pathways. Science 271, 357-60.
Scheres, B. and Heidstra, R. (1999). Digging out roots: pattern formation, cell division,
and morphogenesis in plants. Curr Top Dev Biol 45, 207-47.
Schoof, H., Lenhard, M., Haecker, A., Mayer, K. F., Jurgens, G. and Laux, T.
(2000). The stem cell population of Arabidopsis shoot meristems in maintained by
a regulatory loop between the CLAVATA and WUSCHEL genes. Cell 100, 635-
44.
137
Shao, J., Zhou, B., Zhu, L., Qiu, W., Yuan, Y. C., Xi, B. and Yen, Y. (2004). In vitro
characterization of enzymatic properties and inhibition of the p53R2 subunit of
human ribonucleotide reductase. Cancer Res 64, 1-6.
Shechter, D., Costanzo, V. and Gautier, J. (2004). ATR and ATM regulate the timing
of DNA replication origin firing. Nat Cell Biol 6, 648-55.
Shen, W. H. (2002). The plant E2F-Rb pathway and epigenetic control. Trends Plant Sci
7, 505-11.
Siddiqui, N. U., Stronghill, P. E., Dengler, R. E., Hasenkampf, C. A. and Riggs, C. D.
(2003). Mutations in Arabidopsis condensin genes disrupt embryogenesis,
meristem organization and segregation of homologous chromosomes during
meiosis. Development 130, 3283-95.
Sjoberg, B. M., Reichard, P., Graslund, A. and Ehrenberg, A. (1978). The tyrosine
free radical in ribonucleotide reductase from Escherichia coli. J Biol Chem 253,
6863-5.
Smith, L. G., Hake, S. and Sylvester, A. W. (1996). The tangled-1 mutation alters cell
division orientations throughout maize leaf development without altering leaf
shape. Development 122, 481-9.
Song, J. Y., Leung, T., Ehler, L. K., Wang, C. and Liu, Z. (2000). Regulation of
meristem organization and cell division by TSO1, an Arabidopsis gene with
cysteine-rich repeats. Development 127, 2207-17.
Soni, R., Carmichael, J. P., Shah, Z. H. and Murray, J. A. (1995). A family of cyclin
D homologs from plants differentially controlled by growth regulators and
containing the conserved retinoblastoma protein interaction motif. Plant Cell 7,
85-103.
Steimer, A., Amedeo, P., Afsar, K., Fransz, P., Scheid, O. M. and Paszkowski, J.
(2000). Endogenous targets of transcriptional gene silencing in Arabidopsis. Plant
Cell 12, 1165-78.
Steves, T.A. and Sussex, I.M. (1989). Patterns in Plant Development, 2nd. ed.,
Cambridge Univ. Press, Cambridge, 388 pp.
Stewart, R. N. and Dermen, H. (1970). Determination of number and mitotic
activity of shoot apical initial cells by analysis of mericlinal chimeras. Am.
J. Bot. 57, 816-826
Takahashi, T., Matsuhara, S., Abe, M. and Komeda, Y. (2002). Disruption of a DNA
topoisomerase I gene affects morphogenesis in Arabidopsis. Plant Cell 14, 2085-
93.
138
Takeda, S., Tadele, Z., Hofmann, I., Probst, A. V., Angelis, K. J., Kaya, H., Araki,
T., Mengiste, T., Scheid, O. M., Shibahara, K. et al. (2004). BRU1, a novel link
between responses to DNA damage and epigenetic gene silencing in Arabidopsis.
Genes Dev 18, 782-93.
Tanaka, H., Arakawa, H., Yamaguchi, T., Shiraishi, K., Fukuda, S., Matsui, K.,
Takei, Y. and Nakamura, Y. (2000). A ribonucleotide reductase gene involved
in a p53-dependent cell-cycle checkpoint for DNA damage. Nature 404, 42-9.
Trimarchi, J. M. and Lees, J. A. (2002). Sibling rivalry in the E2F family. Nat Rev Mol
Cell Biol 3, 11-20.
Trotochaud, A. E., Hao, T., Wu, G., Yang, Z. and Clark, S. E. (1999). The
CLAVATA1 receptor-like kinase requires CLAVATA3 for its assembly into a
signaling complex that includes KAPP and a Rho-related protein. Plant Cell 11,
393-406.
Trotochaud, A. E., Jeong, S. and Clark, S. E. (2000). CLAVATA3, a multimeric
ligand for the CLAVATA1 receptor-kinase. Science 289, 613-7.
Uhlin, U. and Eklund, H. (1994). Structure of ribonucleotide reductase protein R1.
Nature 370, 533-9.
Vandepoele, K., Raes, J., De Veylder, L., Rouze, P., Rombauts, S. and Inze, D.
(2002). Genome-wide analysis of core cell cycle genes in Arabidopsis. Plant Cell
14, 903-16.
Vialard, J. E., Gilbert, C. S., Green, C. M. and Lowndes, N. F. (1998). The budding
yeast Rad9 checkpoint protein is subjected to Mec1/Tel1-dependent
hyperphosphorylation and interacts with Rad53 after DNA damage. Embo J 17,
5679-88.
Vlieghe, K., Vuylsteke, M., Florquin, K., Rombauts, S., Maes, S., Ormenese, S., Van
Hummelen, P., Van de Peer, Y., Inze, D. and De Veylder, L. (2003).
Microarray analysis of E2Fa-DPa-overexpressing plants uncovers a cross-talking
genetic network between DNA replication and nitrogen assimilation. J Cell Sci
116, 4249-59.
Voegtli, W. C., Ge, J., Perlstein, D. L., Stubbe, J. and Rosenzweig, A. C. (2001).
Structure of the yeast ribonucleotide reductase Y2Y4 heterodimer. Proc Natl
Acad Sci U S A 98, 10073-8.
Vogelstein, B., Lane, D. and Levine, A. J. (2000). Surfing the p53 network. Nature 408,
307-10.
139
Waites, R. and Hudson, A. (1995). phantastica: a gene required for dorsoventrality in
leaves of Antirrhinum majus. Development 121, 2143-2154.
Wang, H., Y. Zhou, S. Gilmer, S. Whitwill and Fowke, L. C.  (2000). Expression of
the plant cyclin-dependent kinase inhibitor ICK1 affects cell division, plant
growth and morphology. Plant J 24, 613-23.
Wang, P. J., Chabes, A., Casagrande, R., Tian, X. C., Thelander, L. and Huffaker,
T. C. (1997). Rnr4p, a novel ribonucleotide reductase small-subunit protein. Mol
Cell Biol 17, 6114-21.
Weinberg, R. A. (1992a). The integration of molecular genetics into cancer
management. Cancer 70, 1653-8.
Wells, J., Graveel, C. R., Bartley, S. M., Madore, S. J. and Farnham, P. J. (2002).
The identification of E2F1-specific target genes. Proc Natl Acad Sci U S A 99,
3890-5.
Windle, B., Draper, B. W., Yin, Y. X., O'Gorman, S. and Wahl, G. M. (1991). A
central role for chromosome breakage in gene amplification, deletion formation,
and amplicon integration. Genes Dev 5, 160-74.
Wu, L., Timmers, C., Maiti, B., Saavedra, H. I., Sang, L., Chong, G. T., Nuckolls, F.,
Giangrande, P., Wright, F. A., Field, S. J. et al. (2001). The E2F1-3
transcription factors are essential for cellular proliferation. Nature 414, 457-62.
Wu, S., J. Huang, J. Dong and Pan, D.  (2003). hippo encodes a Ste-20 family protein
kinase that restricts cell proliferation and promotes apoptosis in conjunction with
salvador and warts. Cell 114, 445-56.
Xue, L., Zhou, B., Liu, X., Qiu, W., Jin, Z. and Yen, Y. (2003). Wild-type p53
regulates human ribonucleotide reductase by protein-protein interaction with
p53R2 as well as hRRM2 subunits. Cancer Res 63, 980-6.
Yamaguchi, T., Matsuda, K., Sagiya, Y., Iwadate, M., Fujino, M. A., Nakamura, Y.
and Arakawa, H. (2001). p53R2-dependent pathway for DNA synthesis in a
p53-regulated cell cycle checkpoint. Cancer Res 61, 8256-62.
Yao, R., Zhang, Z., An, X., Bucci, B., Perlstein, D. L., Stubbe, J. and Huang, M.
(2003). Subcellular localization of yeast ribonucleotide reductase regulated by the
DNA replication and damage checkpoint pathways. Proc Natl Acad Sci U S A
100, 6628-33.
Zhao, X., Chabes, A., Domkin, V., Thelander, L. and Rothstein, R. (2001). The
ribonucleotide reductase inhibitor Sml1 is a new target of the Mec1/Rad53 kinase
cascade during growth and in response to DNA damage. Embo J 20, 3544-53.
140
Zhao, X., Georgieva, B., Chabes, A., Domkin, V., Ippel, J. H., Schleucher, J.,
Wijmenga, S., Thelander, L. and Rothstein, R. (2000). Mutational and
structural analyses of the ribonucleotide reductase inhibitor Sml1 define its Rnr1
interaction domain whose inactivation allows suppression of mec1 and rad53
lethality. Mol Cell Biol 20, 9076-83.
Zhao, X., Muller, E. G. and Rothstein, R. (1998). A suppressor of two essential
checkpoint genes identifies a novel protein that negatively affects dNTP pools.
Mol Cell 2, 329-40.
Zhao, X. and Rothstein, R. (2002). The Dun1 checkpoint kinase phosphorylates and
regulates the ribonucleotide reductase inhibitor Sml1. Proc Natl Acad Sci U S A
99, 3746-51.
Zhou, B., Liu, X., Mo, X., Xue, L., Darwish, D., Qiu, W., Shih, J., Hwu, E. B., Luh,
F. and Yen, Y. (2003). The human ribonucleotide reductase subunit hRRM2
complements p53R2 in response to UV-induced DNA repair in cells with mutant
p53. Cancer Res 63, 6583-94.
Zhou, B. B. and Elledge, S. J. (2000). The DNA damage response: putting checkpoints
in perspective. Nature 408, 433-9.
Zhou, Z. and Elledge, S. J. (1992). Isolation of crt mutants constitutive for transcription
of the DNA damage inducible gene RNR3 in Saccharomyces cerevisiae. Genetics
131, 851-66.
Zhou, Z. and Elledge, S. J. (1993). DUN1 encodes a protein kinase that controls the
DNA damage response in yeast. Cell 75, 1119-27.