TRANSCRIPTIONAL RESPONSE OF O6 - METHYLGUANINE
METHYLTRANSFERASE DEFICIENT YEAST TO METHYL-NNITRO-N-NITROSOGUANIDINE (MNNG)
by
Anoop Rao
Submitted to the Biological Engineering Division,
School of Engineering, Massachusetts Institute of Technology
in partial fulfillment of the requirements for the degree of
Master of Science in Biological Engineering
MASSACHUSETrS INSTrnJTE
OF TECHNOLOGY
at the
FEB 1 2004
© 2004, Massachusetts Institute of Technology
-- . ·.u-.
February 2004
_I__L
LIBRARIES
__I_
·I_
Signature of the Author _
Biological Engineering Division
/
February 5, 2004
Approved by -
...
Leona Samson
/3
//7
Professor
Thesis supervisor
Certified by
-
--- ''-'
('G/1/
Alan J Grodzinsky
Professor
Chairman of Biological Engineering Graduate Committee
1
ARCHIVES
___
TRANSCRIPTIONAL RESPONSE OF O6 - METHYLGUANINE
METHYLTRANSFERASE DEFICIENT YEAST TO METHYL-N-NITRO-NNITROSOGUANIDINE (INNG)
by
Anoop Rao
Submitted to the Biological Engineering Division on February 5, 2004, in partial fulfillment of the requirements
for the Degree of Master of Science in Biological Engineering.
Abstract
Damage to DNA can occur by means of endogenous biochemical processes or exogenous chemicals
such as alkylating agents. If left unrepaired, alkylated bases, most notably, 06 Methylguanine (O 6MeG) can be
mutagenic and cytotoxic to the cell. Luckily, DNA methyltransferase (encoded by the gene MGT1 in yeast),
repairs this damage. By using transcriptional profiling as a tool, an attempt to elucidate the role of MGTI has
been made. First, the basal expression profile of the mgtl was established. Then, the response of wild-type
(WI) yeast and yeast lacking MGT1 (mgti) to the alkylating agent, MNNG was studied using exponentially
growing WT and mgti cultures which were exposed to 30btg/ml of MNNG for 10 to 60 minutes.
Basal expression profile of yeast lacking MGTi showed up-regulation of RETV7, a gene implicated in
spontaneous mutagenesis. Response to MNNG was invoked immediately and was dramatic and widespread
involving 30% of the genome in both WT and mgt/. Cell-cycle checkpoints, damage signal amplifiers, DNA
repair genes (nucleotide excision repair, photoreactive repair, mismatch repair) and chromatin remodeling
genes were induced. Genes involved in maintaining mitochondrial structure and mitochondrial genome were
also induced. Intriguingly, RPN4, a key regulator of proteasomal system was found to be repressed.
Environmental stress response genes were culled out to examine the effects of MNNG on WT and mgtl,
more carefully.
Temporal gene expression profiles in WT and mgtl were informative in delineating differences in the
distinct responses mounted by WT and mg/tl. The magnitude of response in mgt/ is more profound than in
WT. The differences in the dynamic trends between the two suggest that mgt/ initiates a coordinated response
involving repression of transcription factors and subsequently, induction of RNA processing (35% of genes
incrementally induced) and kinases involved in protein phosphorylation. In the W\T, the response was
restricted to a transient repression of fundamental biochemical processes. Interestingly, a gene whose
repression is known to mimic apoptosis was found to be repressed in the ~WT. The overwhelming induction
of ribosomal protein synthesis genes in both WT and mgtl in response to MNNG is an unexpected result that
could signify a successful recovery following wide-spread cellular damage.
Thesis supervisor: Leona D Samson
Title: Professor
2
Acknowledgements
I thank Prof. Leona Samson for her guidance, advice and suggestions during the course of
the experiments and data analysis. I also thank her for providing unflinching financial support. I
extend a special word of thanks to Dr Rebecca Fry for offering excellent advice with the
experiments, data analysis and her words of encouragement.
Brad Hogan was superb with his
technical assistance in running the first set of arrays. Merny, from the BioPolymer center, provided
timely help in scanning the arrays. Sean and Sanchita provided excellent support via the
BioMicrocenter. Other members of the Samson Lab added to the hustle.
The experiments were possible because of funding from the NIH and the excellent
infrastructure that was provided by the Biological Engineering Division, BioMicro Center,
BioPolymer Center and Massachusetts Institute of Technology.
3
~~
"
-
I-~~~~~----
I---~~~~-
_
Table of Contents
Title page ................................ ................................................................................................... 1
Abstract......................................................................................................................................2
Acknowledgem ents ...................................................................................................................
3
Table of Contents ...................................................................................................................... 4
List of Figures ...........................................................................................................................
5
List of Tables ............................................................................................................................. 6
Chapter 1: Introduction ............................................................................................................. 7
Mechanism of DNA Damage .......................................................................................... 22
Overview of DNA Repair m echanism s ............................................................................ 26
Damage caused by alkylating agents ............................................................................... 28
Spontaneous mutations .................................................................................................... 28
Saccharomyces cerevisiae as a model system .................................................................. 28
Chapter 2: Expression Profiling .............................................................................................. 22
Materials and Methods ......................................
22
Data Analysis .................................................................................................................... 26
Results .......
.......................................................................................................................
28
Effect of mgtl on transcriptional profile .......................................................................... 35
Effect of MNNG on transcriptional profile ..................................................................... 40
Discussion ............................................................................................................................... 49
Summary..................................................................................................................................64
Bibliography ....................................................................................................................
66
Appendix of Tables .................................................................................................................
76
Appendix of Protocols .................................................................................................................
4
----
I-----'-
List of Figures
Figure 1: Unchecked DNA damage may lead to cancer ........................................................................
7
Figure 2: The nitrogenous bases; Purines (A,G) and Pyrimidines (C,T,U) ............................................... 9
Figure 3: Causes and consequences of DNA damage ........................................................................
11
Figure 4: Outcomes of DNA damage ........................................................................
12
Figure 5: Alkylating agent, M NNG ........................................................................
13
Figure 6: O 6 Mehyltguanine methyltransferase accepts the methyl group at its Cys residue .............. 13
Figure 7: Overview of DNA repair mechanisms ........................................................................
13
Figure 8: Sites within the nitrogenous bases susceptible to alkylation damage ...................................... 16
Figure 9: Steps in cRNA preparation and hybridization to GeneChip .................................................... 23
Figure 10: MNNG gradient plate assay for Wild-type BY4741 (WIT) and BY4741 mgt/i A................27
Figure 11: MNNG-induced killing in wild-type (WT) and MTase deficient yeast (mgt). .................... 28
Figure 13: Box plots of intensities before (A) and after (B) RMA quantile normalization ................. 32
Figure 14: Expression ratio plot for yeast methylguanine methyltransferase mutant (mgtl) ............... 36
Figure 15: Principal component analysis of the experimental groups .................................................... 37
Figure 16: Heat map of the expression ratio from wild-type (WT) and Mtase mutant (mgt) ............. 38
Figure 17: Genes responsive to treatment with MNNG in the wild-type (WT) yeast .......................... 39
Figure 18: Gene expression responsiveness for some functional categories .......................................... 40
Figure 19: Venn-diagram of gene expression responsiveness in wild-type (WT) and methylguanine
methyltransferase deficient yeast (mgl) upon treatment with MNNG .................................
41
Figure 20: Venn-diagram of genes that are incrementally induced or repressed upon increasing
length of exposure to M NNG .........................................................................
46
5
---
'---il-----
List of Tables
Table 1: Experimental design of oligonucleotide expression study ........................................................ 22
Table 2: Parameters of post-hybridization quality control ..............................................................
31
Table 3: R2 values of WT (3A) and mgtl (3B) replicates ..............................................................
33
Table 4: A subset of the genes that are up-regulated in basal mgtl ......................................................... 76
Table 5: A subset of the genes that are down-regulated in basal mgtl.....................................78
Table 6: Subset of genes that are specifically induced in WT upon MNNG treatment ....................... 79
Table 7: A subset of the genes that are induced specifically in mgtl ........................................................
82
Table 8: A subset of the genes that are repressed specifically in WT.
84
.....................................
Table 9: A subset of the genes that are repressed specifically in mgtl .....................................................
86
Table 10: Genes that are induced in WT and mngtl upon MNNG treatment ......................................... 88
Table 11: Genes repressed in both WT and mgtl upon MNNG treatment .......................................... 98
Table 12: Genes involved in DNA replication and repair ................................................................
Table 13: Genes that are incrementally induced in both WT and
107
g/tl ................................................. 109
Table 14 Genes that are incrementally induced specifically in WT .
.....................................
111
Table 15: Genes that are incrementally induced specifically in mgtl ...................................................... 113
Table 16: Genes that incrementally repressed in both WT and mgtl ..................................................... 116
Table 17: Genes that are incrementally repressed specifically in mgtl ................................................... 117
Table 18 Genes that are incrementally repressed specifically in WT .
.....................................
120
6
Chapter 1: Introduction
BACKGROUND
Genetic information in any cell is chemically stored in the form of deoxyribonucleic
acid (DNA) and it essentially comprises a sequence of repeating nucleotides. Nucleotides in
turn consist of a pentose sugar, a nitrogenous base and a variable number of phosphate
groups stacked and aligned in an orderly fashion. Maintaining the physical and chemical
integrity of the DNA structure is vital for its function. Unfortunately, however, errors can be
introduced during replication, recombination and even repair. They can also be introduced via
damage due to physical and chemical agents. Eventually, if errors remain uncorrected, it may
lead to instability of the chemical structure and modification of the molecular structure. Such
an alteration classifies as DNA damage and this may sometimes preclude the semiconservative replication of DNA, lead to cell cycle arrest and often, cell death. Another
ominous outcome of unchecked DNA damage is the accumulation of mutations that can lead
to cancer. (Figure 1)
APOPTOSIS
/1
I
DNA damage
! '.
Cell cycle
Mutations
CANCER
T
DNA repair
arrest
Figure 1: Unchecked DNA damage may lead to cancer.
7
_
I_
I___
The mechanism of damage to DNA
Spontaneous alterations
The nitrogenous bases, purines and pyrimidines, occasionally undergo a spontaneous
alteration in their chemistry. The bases can lose their exocyclic amino group and undergo
deamination. This modification increases the propensity for anomalous pairing of bases and
an inappropriately incorporated base can introduce a transition or transversion mutation. In
addition, instability of base pairing may also lead to replication arrest. Purines and pyrimidines
can also be spontaneously hydrolyzed and lost. (Lindahl, 1993). Abasic sites that are produced
can also lead to mutations during replication. (Loeb, 1986)
Unlike the fleeting deamination and hydrolysis reactions, oxidative damage to DNA is more
elaborate and once initiated, results in a chain reaction. The cause of damage can be
exogenous or endogenous. Exogenous sources of oxidative DNA damage include radiation,
near UV light at 320 to 380nm and several drugs (Friedberg, 1995). Endogenously, redox
reactions which ubiquitously occur in cells, are the major source of reactive oxygen species.
Notable among them are the by-products of aerobic mitochondrial respiration. Singlet oxygen
and hydrogen peroxide inflict damage via formation of hydroxyl radicals through metal
catalyzed reactions. The intermediates and by-products of such reaction are independently
capable of inflicting more damage on intracellular macromolecules. Fortunately, there are
several cellular defense mechanisms help to mitigate the effect of these reactive oxygen
species (ROS). These include antioxidant enzymes such as superoxide dismutase, glutathione
peroxidase and other scavengers, Vitamin C and a tocopherol. Potentially unfavorable cellular
responses triggered by the presence of ROS may include the inhibition of cell cycle
progression, the initiation of apoptosis and the activation of a degradative process to replace
macromolecules. Oxygen radicals are also known to induce chromosome breaks. In general,
irrespective of the source and mechanism of alteration or damage, the most ominous outcome
is typically a mismatch of base pairs during DNA synthesis. The most important oxidative
base adduct, 8-oxoguanine can mispair and be mutagenic (Lindahl., 1993, Friedberg, 2002).
8
1_··1___·_··111_____
NH,
NH.X
0il
.
N
.H.
NH-,
'
H
Ade
(A
Guanine
Guanine
(G)
Cytosine
(C)
0
H
Thymine
(1)
Figure 2: The nitrogenous bases; Purines (A,G) and Pyrimidines (C,T,U)
Environmental damage
Physical agents - onizing Radiation: Apart from spontaneous alterations and damage to
DNA, physical and chemical agents in the environment inflict a substantial amount of
damage. Exposure to ionizing radiation as a therapeutic, diagnostic or occupational hazard
induces a variety of lesions by direct damage. Radiolysis of water generates reactive oxygen
species that damage cellular macromolecules. Glutathione, a radioprotector, can counteract
the damage at several tiers of radical production. Damage to bases, sugar moieties and direct
induction of strand breaks can occur. Strand breaks are a special problem since mere DNA
ligation may not be sufficient to repair the lesion. (Burrows, 1998). Ultraviolet (UV) radiation
induces covalent linkage of adjacent pyrimidines producing cyclobutane pyrimidine dimers
(CPD). These lesions distort the helix and lead to extensive bending of DNA, albeit variably.
A less frequent lesion is the pyrimidine-pyrimidone (6-4) photoproduct that, like CPD's,
distorts the helix (Ravanat, 2001). These physical distortions may result in an obligatory arrest
of replication.
Chemical agents: Environmental exposure to chemicals included as food agents
(Sugimura., 2002), inhalation of polluted air and ingestion of contaminated water are by far
9
the most common modes of encountering DNA damaging agents. Occupational hazards and
therapeutic intervention, most notably by anti-cancer agents are responsible for most of the
DNA damage in humans. The diverse class of chemicals known to cause DNA damage
includes psoralens, benzo[a]pyrene, aflatoxins and nitroquinolones and alkylating agents.
Historically, there has been an interest in examining the potential for food additives to be
potent carcinogens. Most of the evidence has relied on demonstrating that electrophilic
metabolites of the parent compound can forms DNA or protein adducts. Their metabolism is
dependent on an inducible system of membrane proteins called the cytochrome P-450 system.
Apart from this P-450 system, there are other enzymes that conjugate compounds to make
them more water soluble and permit easy elimination from the system. They include
acetyltransferases, glucoronyl transferases, adenosylating enzymes and methylating enzymes.
In principle, the cell has several mechanisms of dealing with DNA damage but it is
unfortunate, however, that metabolites can themselves be more harmful than the parent
compounds.
Means of response to DNA damage
In essence, physical and chemical damage to DNA is a universal phenomenon across
living systems. Damage by physical and chemical agents can lead to arrest of replication and
transcription. Fortunately, there are sub-cellular systems that operate in a coordinated way to
sense and respond to damage. Broadly, they can be classified as DNA repair mechanisms
(specific set of events to eliminate the primary lesion) and DNA damage checkpoint
mechanisms - accessory events that stall the cell cycle and permit the specific DNA repair
mechanisms to act.
10
DNA damage
Replication stress
Signals
Sensors
Transducers
Figure 3: Causes and consequences of DNA damage.
An elaborate, overlapping set of enzymes, proteins and mechanisms deal with these
deleterious lesions. Cells respond to DNA damage by delaying cell cycle progression and by
increasing the expression of a few genes involved in the repair and tolerance of DNA damage
(Friedberg et al,1995). Surveillance mechanisms in eukaryotic cells monitor and regulate the
cell cycle and its progress. The cell-cycle checkpoints are activated by one or more signals and
ultimately results in the inhibition of cell cycle progression. The checkpoint mechanism first
detects damaged DNA and then generates a signal that arrests cells in the GI/S or G 2/M
phase of the cell cycle. It also slows down S phase (DNA synthesis). This mechanism is
thought to prevent the replication of damaged templates and the segregation of broken
chromosomes.
Consequences of unrepairedlesions
Despite the orchestrated response, however, the damage may not be mitigated. If
lesions are not repaired, they would pose a problem by being mutagenic or lethal. In response
to DNA damage the cell has four major routes of responses. Cell-cycle arrest provides the
crucial time for repairing the damage. If the damage load is too profound for the cell to
handle, it may undergo apoptosis (programmed cell death) to avoid the propagation of highly
defective cells. The lesions may be fixed by DNA repair pathways or alternatively, the
11
-
-----
unrepaired lesions may generate sequence changes in the genome to be passed on as a
mutation.
]
Figure 4: Outcomes of DNA damage
Overview of repair mechanisms
Photoreactivationbyphotolyases
UV light exposure generates 2 major classes of stable DNA lesions - cyclobutane
pyrimidine dimers (CPDs) and 6-4 photoproducts (6-4 PD). Unless repaired, these lesions
may lead to blockage of transcription, mutations, cell death and cancer. CPD's and 6-4 PD's
are removed by two pathways: (i) nucleotide excision repair (NER); and (ii)photoreactivation.
Some plants, bacteria, and yeast possess a photolyase that preferentially reverses the CPD's in
the non-transcribed strand of active genes. DNA photolyases catalyze the light-dependent
repair of pyrimidine dimers in DNA. (Carell, 2001). Photolyases bind tightly to CPDs and, on
excitation by 340-400-nm light, catalyse the cleavage of the cyclobutane linkage between the
adjacent pyrimidines and restore the monomeric bases without cutting the phosphodiester
backbone of DNA.
In yeast, the Phrl gene that codes for photolyase has been shown to be upregulated by several
DNA-damaging agents such as UV radiation, 4NQO,
IMMNS and MNNG (Sebastian et al,
1990). If Phrl is unable to resolve cyclobutane linkages then it will also try to enhance the
NER of CPDs. The photolyase enzyme functions as a model fr proteins that interact with
sites of DNA damage and have the potential to facilitate DNA-damage recognition by repair
pathways. (Sebastian and Sancar, 1991)
Methylguanine methyltransferase
The methylguanine methyltransferase
reverts O6-methylguanine to guanine by
transferring the methyl group from DNA to a reactive cysteine group of the protein in an
irreversible reaction. This covalent attachment of the alkyl group to the cysteine residue
12
inactivates the enzyme. Alkylguanine methyl transferase is a suicide enzyme. The mechanism
for this reaction is indicated in Figure 7.
Figure 5: Alkylating agent, MNNG
Cys-- 3 -, M31'1
Cys-SH
inac:tive
active
N)
\
. methyltransferase
/
!
H 2N
N
0 6 -Methylguanine
N
H2 5
R
nucleotide
R
Guanine nucleotide
Figure 6: 0 6Mehyltguanine methyltransferase accepts the methyl group at its cysteine
residue
F
P1
Figure 7: Overview
of I DNA
Figure 7: Overview of DNA repair mechanisms
Excision Repair
In contrast to direct repair there is cleavage of the sugar phosphate backbone in excision
repair.
13
81___
_
I
_____
Base Excision Repair
Damage to DNA from deamination, oxidation and alkylation is mainly repaired by
BER. In base excision repair, the DNA bases that are altered by small chemical modifications
are replaced through the excision of only the damaged nucleotide (short patch BER) or
through the removal of 2-13 nucleotides containing the damaged nucleotide (long-patch
BER). DNA glycosylases initiate BER by excising damaged bases from DNA and generating
abasic sites.
Nucleotide Excision Repair (NER)
In cases where the alteration involves the addition of large chemical additions or
cross-links, the DNA bases are excised using the nucleotide excision repair where a short,
single-stranded segment containing the damage is removed. NER helps in repair of bulky base
adducts
formed
by
UV
radiation,
various
environmental
mutagens,
and
certain
chemotherapeutic agents. In NER. (Wood, 1997)
Mismatch Repair (MMR).
An important replication-associated correction function is provided by the postreplicative mismatch repair system. Base-base mismatches or loops of extra bases if left
unrepaired, will generate point or frameshift mutations respectively. NMisincorporation of noncomplementary bases by DNA polymerases is a major source of the occurrence of
promutagenic base-pairing errors during DNA replication or repair. MMR is conserved from
bacteria to humans. It identifies and corrects mispaired bases and 1-3-nucleotide loops that
result from DNA polymerase errors during replication.
Double strand break (DSB) repair
Double strand breaks are rare and two independent pathways handle them;
homologous recombination (HR) and non-homologous end joining (NHEJ). Homologous
recombination uses extensive homology to code DNA and maintain accuracy. Nonhomologous end joining involves a coordinated rejoining of the broken ends and uses no or
extremely limited regions of homology as a template for repair. Consequently, this process is
inaccurate and the deletions of a few nucleotides are introduced at the site of the DSB. HR
14
and NHEJ are important in all eukaryotes and HR is more important in rapidly dividing cells
and NHEJ is more important in quiescent or terminally differentiated cells. HR is important
for meiosis or the repair of inter-strand cross-links, while NHEJ is required for joining of
DNA fragments while generating the diversity of the immune system.
Mechanism of damage by alkylation
Alkylating agents are electrophilic compounds with affinity for nucleophilic centers in
organic macromolecules. They are probably the broadest class of chemicals that have the
potential to cause
profound
damage to DNA. Alkylating agents
are classified as
monofunctional or bifunctional depending on the number of reactive groups, and therefore,
the ability to react with multiple sites within DNA. Alkylating agents attack nitrogen and
oxygen at various sites within nitrogenous bases with different reactivity's. Apart from these
veritable hot spots within nitrogenous bases, alkylating agents can react with oxygen in the
phosphodiester linkage to form a phosphotriester.
15
1 111
1
11_
111_1____
+
H
N-H--------------- H-
dN
N
\ \
dR
dR
°-------- H-N\
tO
H
cytosine
H 3C
N
guanine
O--------H
3N-H -----N--99 2
dR
dR
,
adenine
thymine
Figure 8: Sites within the nitrogenous bases susceptible to alkylation
damage.
By virtue of its ability to reach several nucleophilic sites within the nitrogenous bases, inter
and intra-strand cross links can occur as a consequence of exposure to bifunctional alkylating
agents. The covalent link sustains this anomaly and prevents strand separation (if there is an
inter-strand crosslink) leading to a complete block of replication and transcription.
Damage to DNA bases
Alkylating agents are structurally diverse group of chemicals that cause a wide range of
biological effects including cell death, mutation and cancer. DNA damaged by these agents
contains
widely
amounts
different
of
12
alkylated
purines/pyrimidines
and
two
16
I_ __
_ _
_________
phosphotriester isomers. They are used in anticancer therapy and are also found in cigarette
smoke.
Lesions caused by a/lkylation
Attack in 0 6 position of guanine leads to the formation of the adduct, O6
Methylguanine (06 MeG). This is a relatively minor lesion compared to 04Methylthymine
(0 4 MeT), but potentially the most deleterious lesion if left unrepaired in the system. Other
potentially harmful lesions include 3 Methyl Adenine (3MeA). If the 06 MeG lesion remains
unrepaired, then it permits G-A transition mutation following 2 rounds of replication. This
happens in both eukaryotes and prokaryotes. Recombination and cell death that may ensue
but both need a functional MMR system.
In E.coli, it has been shown that the miscoding alkylation adducts on the template
strand would lead to anomalous base pairs upon replication. Provocation of mismatch repair
by such lesions would result in a futile turnover of the newly synthesized strand because the
offending adduct is not removed from the template DNA, a process that could lead to cell
death. Luckily, the 06 methylguanine MTase protein is able to counter this effect by
irreversibly and covalently binding to the methyl group and plucking it off the base. Since the
cysteine which is methylated is not regenerated at all, the capacity for repair of O6 methylguanine is limited by the number of molecules of the MTase available within the cell.
Alkyltransferases across systems
Methyltransferase belongs to a class of proteins, the alkyltransferases. There are close
to a 100 alkyltransferases but the structure of only 3 three family members: the Ada-C protein
from Escherichia coli (Moore et al 1994), the human alkyltransferase (hAGT) (Daniels, 2000),
and Pyrococcus kodakaraensis (Hashimoto 1999) are known. The protein is non-enzymatic in
nature and therefore the protection by alkyltransferase depends on the regulation of its
synthesis and degradation.
Structure andfunction of Yeast methyltranserase (IGT1)
Repair of 0 6-MeG in yeast extracts was shown to be performed by a 25-kilodalton
protein Methyl transfer was accompanied by the formation of S-methylcysteine. The S.
cerevisiae MGTI codes for a 188 amino acid protein. About half of the MGT1 protein has
17
alll
_ I I
I
homology with four bacterial MTases and also the human DNA MTase. (Xiaoand Samson
1992)
Exponentially growing yeast cultures have about 150 molecules of MITase in each cell. The
yeast MTase has a half-life of about 4 min at 37°C. Synthesis of the yeast DNA MTase is not
inducible by sublethal exposures to alkylating agent. The substrates for yeast MTase include
O6 MeG and O 4MeT. Unlike this, the human MITase is very specific for O 6MeG.
Spontaneous mutations
Mutations are defined as spontaneous when they arise in cells that are not actively exposed to
exogenous, xenobiotic mutagens. Spontaneous mutations occur due to either uncorrected DNA
replication errors, or endogenous metabolites that cause lesions on DNA. Oxidative damage and
alkylation damage are the 2 major sources of endogenous DNA damage. It results as a consequence of
cellular metabolism and failure to correct this damage due to genetic defects results in significantly
increased spontaneous mutation rates. Spontaneous mutations have been studied earlier using several
systems including MGTI deletions in yeast. MGTI deleted mutants were shown to have an increased
spontaneous mutation rate suggesting an endogenous source of DNA methylation damage. (Xiao and
Samson, 1993).
Alkylation due to endogenousprocesses
S-adenosyl methionine (SAM) a cellular methylase co-factor has a reactive methyl group and is
responsible for enzymatic methylation of DNA, RNA and proteins. Under physiological conditions,
6 MeG
SAM has been shown to non-enzymatically methylate DNA to form 3-Methyl adenine and 0O
(Rydberg and Lindahl 1982).
Endogenous processes might include the 'aberrant' methylation of
guanine by S-adenosylmethionine and the endogenous nitrosation of compounds containing primary
amino groups and their subsequent breakdown to methylating species (Sedgwick, 1997). nnitrosoglycocholic acid has been shown to be able to methylate DNA in vitro and in ivo (Shuker and
Margison, 1997).
18
-L--~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
----
Saccharomyces cerevisiae as a model system to study genome-wide expression.
The baker's yeast S. cerezvisiae is an informative model organism in traditional genetic
studies. It also presents an ideal model genome for large-scale functional analysis. Relative to
other eukaryotes, S. cerevisiae has a compact genome. Approximately, 70% of its total (nonribosomal DNA) genetic complement
chromosomes, the 12-megabase (b)
is protein-coding sequence.
Encompassing
16
yeast genome is predicted to encode about 6,200 genes,
with 1 gene per 2 kb of genomic sequence. (Goffeau, 1996).S. cerevisiae is an informative
predictor of human gene function; nearly 50% of human genes implicated in heritable diseases
have yeast homologues. (Bassett, 1996, 1997. Venter. 2001).
Since its development in the mid-1990s (Schena, 1995, Chee,. et al 1996), the DNA
microarray has emerged as the pre-eminent tool for functional genomics. The ability to
analyse thousands of DNA samples simultaneously by hybridization-based assay has provided
a popular method for analysing the relative levels of mRNA transcripts on a genome-wide
scale. Typically, DNA microarrays have been used to identify genes, the expression of which
is either induced or repressed during specific cellular responses. For example, DeRisi (DeRisi
et al. 1997) used DNA microarrays to monitor relative changes in mRNA levels during the
shift from anaerobic fermentation to aerobic respiration in yeast.
ficroarrays have also been
used to identify genes differentially expressed during sporulation (Chu et al 1998), as well as
genes periodically expressed during the cell cycle (Cho. et al 1998, Spellman, 1998). Jelinsky
and Samson (1999) used oligonucleotide arrays to identify over 400 genes that are either
induced
or repressed in response
to the DNA-damaging,
alkylating
agent methyl
methanesulphonate (MMS). These and other microarray-based studies have identified genes
that putatively function in common regulatory pathways; such pathways are also being
delineated by transcriptional profiling of strains mutated for key regulatory components.
Affymetrix has used the genomic sequence of the budding yeast Saccharomyces cerevisiae to
design and synthesize high-density oligonucleotide arrays for monitoring the expression levels
of nearly all yeast genes. This direct and highly parallel approach involves the hybridization of
total mRNA populations to a set of four arrays that contain a total of more than 260,000
specifically chosen oligonucleotides synthesized in situ using light-directed combinatorial
chemistry.
19
----
Chapter 2: Expression Profiling
Transcriptional response of Saccharomyces cerevisiae wild type and DNA
methyltransferase mutants.
Materials and Methods
Yeast strains andgrowth conditions
To study the genome-wide transcriptional response of Saccharomyces cerevisiae upon
exposure to MNNG two strains were obtained from Research Genetics, Carlsbad, CA. The
wild-type (WT) BY4741 (MAT a his3A1 leu2AO metl5AO ura3A0O) and the methylguanine
methyltransferase lacking strain, BY4741 mgtl A (mgtl). The mgtl A strain was originally
created using a PCR based gene deletion strategy (Baudin et al., 1993 and Wach et al., 1994).
This gene deletion is viable since MGT is a non-essential gene. The cells were grown and
maintained on YPD (10 g yeast extract, 20 g peptone, 20 g dextrose, 20 g agar/liter)
containing 2 00g/ml of G418 (Geneticin, Sigma Chemicals).
N-Methyl-N'-Nitro-N-Nitrosoguanidine
A 1% stock solution of DNA damaging agent, N-Methyl-N'-Nitro-N-Nitrosoguanidine
(MNNG) was prepared and stored in amber tubes away from light. This stock was used for all
the experiments. First, the phenotypes of the strains were established using a gradient plate
assay. Briefly, equal number of BY4741 and BY4741 mtl A cells were plated in YPD+G418
plates which had MNNG in concentrations of 0
g/ml, 5 ig/ml, 10 jig/ml, 15 jtg/ml, 20
jg/ml and 25 jtg/ml. The agar was allowed to settle at an angle in the plates. This permitted
variable exposure of the cells to a fixed concentration of MNNG. Experiments were
performed in duplicate. Colony formation was observed after 3 days of growth at 30°C.
20
---
Growth Curve
Single colonies of yeast were picked from YPD+G418 plates to inoculate 5 ml of YPD
culture in a test-tube that was rotated overnight at 250 rpm at 300 C.
v
100 1
of each strain was
inoculated in 150 ml of YPD+G418 rotating at 300 rpm, at 300 C. OD and cell counts were
taken over time to follow growth in cell number. Growth curves were plotted for BY4741 and
BY4741 mgt/ A.
MNNGC-induced cell killing
The difference in the MNNG-induced cell killing between BY4741 and BY4741 mgt!
was studied. One colony of wild-type (WT) and MTase lacking mutant (mgtl), was picked and
grown in 2 separate test-tubes with 5ml YPD (with G418), overnight. Then,
v
100 1
was
transferred into 150 ml of media (with G418) in a 250 ml flask. This was grown for 12 hours
at 30 C after which 10 ml of culture was transferred into four 15ml tubes. To this 150 v1 of
1% MNNG stock was added to establish the final concentration of NINNG at 30tg/ml. After
incubating them for 0, 20, 40 and 60 minutes, the test-tubes were centrifuged. The
supernatant was discarded and the cells were re-suspended in 10 ml of distilled water. Over a
series of dilutions, the re-suspended cells were plated onto YPD agar plates with G418.
Colonies were counted after 3 days of incubation at 30 C.
Cell preparation for Microarray Analysis
Single colonies of WT and mgtl were picked from YPD plates to inoculate 5 ml of
YPD culture. They were incubated at 250 rpm at 300 C, overnight. 100Vd of each strain was
inoculated in 150 ml of YPD+G418 media in a flask rotating at 300 rpm maintained at 300 C.
The cells were grown to mid-log phase (OD = 0.8). The cultures were split into 3 volumes of
21
19··-···P··llll··11111111·--·---
-I-
,
50 ml before being exposed to MNNG (30 itg/ml) for a variable length of time as shown in
Table 1. The control samples were mock-treated with same amount of double distilled water.
(DDW). After exposure for the appropriate length of time, the cells were spun down in 50 ml
tubes centrifuged at 8000 rpm. The cells were snap-frozen and stored at -80°C.
Table 1: Experimental design of oligonucleotide expression study.
The number of BY4741 and BY4741 mgt1 A samples that were treated with double distilled
water (DDW) or MNNG.
Control
Treated with 30 ig/ml MNNG
(DDW)
10min
WT
-3
mgtl
20min
3
3
:
.
30mrain
4min
50 min
3
3
3
3
3
3
3
3
60 min
.
3
3
Total RNA preparation
Total RNA was extracted from the frozen cells using the enzymatic lysis protocol
(Qiagen RNEasy Mini Protocol - Standard Version) as detailed in the Appendix of Protocols.
Briefly, the cells were incubated for 20-30 minutes (with gentle shaking, every 5 minutes) with
2 ml of buffer made from Sorbitol (1M), EDTA (0.1M)
mercaptoethanol (0.1%) and 50 U
of lyticase (Sigma) per 1x107 cells. The cells were then centrifuged to pellet the spheroplasts.
In a series of steps the cell wall was lysed and the lysate was made to pass across a silica-gel
membrane to trap the RNA. Finally, RNAse free water was used to elute the RNA out before
estimating the concentration using a spectrophotometer. 260/280 absorbance readings were
measured for total RNA. A ratio of 260/280 ratios between 1.8 to 2.1 was considered
acceptable. If the ratio was below 1.8 (indicates possible contamination) or above 2.1
22
---
-~-~"~l""l""---s"-p--
-------
--
`
(indicates presence of degraded RNA truncated cRNA transcripts, and/or excess free
nucleotides), the total-RNA process was repeated.
Pre-tybrizdiationqualitcontrol
Forty-two samples of total RNA from the different experimental groups (Table 1) was
isolated and stored at -20°C. The samples were also tested using the Agilent 2100 Bioanalyser
system. This permits rapid visualization of RNA sample quality and quantity. A rRNA ratio of
28S/18S close to 2 implies minimal degradation of RNA, a prerequisite for efficient reverse
transcription, cDNA synthesis and in-vitro transcription and to ensure the highest quality
RNA hybridization to the gene expression microarrays. The steps for cRNA synthesis from
total RNA is illustrated in Figure 9 and details are included in the Appendix.
RNA Isolation
First strand cDNA synthesis
Second strand cDNA synthesis
Biotin labeled cRNA synthesis via IVT
Fragmenting the cRNA for target preparation
Hvbridization of cRNA to target arrav
Probe array wash and SAPE staining
Probe Array analysis
Figure 9: Steps in cRNA preparation and hybridization to GeneChip.
23
--------------
GeneChip® hybridizations and Image analysis.
Fragmented cRNA samples were hybridized to GeneChip® arrays containing the
complete yeast genome for a total of 42 arrays (YG-S98 arrays, Affymetrix, CA). The
GeneChip® Yeast Genome S98 Array contains probe sets for approximately 6,400 S. cerevisiae
(S288C strain) genes identified in the Saccharomyces Genome Database (December 1998).
This array also contains approximately 600 additional probe sets representing putative open
reading frames (ORFs) identified by SAGE analysis, mitochondrial proteins, TY proteins,
plasmids, and a small number of ORFs for strains other than S288C.
Scanning was carried out at the MIT Biopolymers Laboratory after hybridizing fragmented
cRNA at a concentration of 0.05 ug/xl to GeneChip(s in 2001d of Affy buffer (100 mM
MES, 1 M NaCl, 20 mM EDTA, 0.01% Tween 20) with GeneChip
eukaryotic hybridization
controls (GeneChip® Eukaryotic Hybridization Controls Kit, Affymetrix, CA) in the
presence of 0.1 mg/ml herring sperm DNA and 0.5 mg/ml acetylated BSA at 40 °C for 16 h
with constant rotation. Arrays were rinsed after hybridization with 200 p1 of stringent wash
buffer (100 mM MES, 0.1 M NaCl, 0.01% Tween 20) followed by a non-stringent wash
(6XSSPE, 0.01% Tween 20). 20XSSPE had the following composition (3 M NaCl, 0.2 M
NaH2PO4, 0.02 M EDTA). Staining was done with 2 ug/ml streptavidin-phycoerytherin and
1 mg/ml acetylated BSA in 6xSSPE-T. Arrays were scanned by a HP G2500A GeneArray
scanner.
24
-- ------
-·-------
I-
Data analysis
A total of 42 hybridizations were performed and the scanned images from Micorarray
Suite 5.0 were stored for computational analysis that was performed using Spotfire, MS Excel
and S-plus Array analyzer.
DataNormaligation
The .cel files generated after scanning using the Affymetrix suite were used for this
analysis. All the 42 .cel files from the control and the experimental groups were analyzed
simultaneously.
The variation between high-density oligonucleotide arrays was reduced by normalizing the
data. The Robust Multichip Average (RNLN) algorithm was used to adjust the background and
perform quantile normalization. The expression results for each ORF/gene were represented
as logarithm (to base 2) of the expression value. The software package RMRLExpress 0.2 alpha
1 version for Windows was used for this purpose. RNLAExpress combines the 16-20 probe
pair intensities for a given gene to define a measure of expression that represents the amount
of the corresponding mRNA species. The normalization takes only perfect matches into
account and the mismatch probe cells are not used for calculation
the signal
intensity/measure of expression.
Analysis of RMA otputfiles
The output of the RLNExpress, in the form of log2 values of expression, was exported to MS
Excel. For further analysis, average expression value from 3 biological replicates (Table 1), for
each time point was computed. Average expression values derived from all the arrays were
compared with the average expression of untreated WT to get an expression ratio (ER). This
25
--------L-i----------
blanket comparison to compute ER ensures that all the data are compared to an unambiguous
baseline. The comparisons were represented as log 2 of expression ratio's (log 2ER).
Post-normalizationcut-off
A log2 ER for a gene/ORF > 1 indicates average fold change induction factor of 2 for that
particular gene/ORF. Analogously, a log 2 ER < -1 indicates an average fold change repression
factor of 2. If log2 ER's were between 1 and -1, they were classified as not significant (NS).
For the purpose of visualization, the log2 ER's across all treatment time-points for both WT
and mgtl were exported and visualized in Spotfire's functional genomics module.
26
- - - ---- ---
l
_·_l__ _-~
___l
II~
RESULTS
MNNG gradient plate
The phenotype of the WT and gtl strains, in response to MNNG, was ascertained
using gradient plates (Figure 10). Upon exposure to increasing concentrations of MNNG,
fewer colonies of mgtl survived when compared to WT.
WT BY4741
Ogg/ml MNNG
BY4741 mgtl A
WT BY4741
5 gtg/ml MNNG
BY4741 mgtl A
WT BY4741
10 gg/ml MNNG
BY4741 mgtl A
WT BY4741
15 gg/ml MNNG
BY4741 mgtl A
E
7
7
T
=
7
. TA
X
DE. ..
WT BY4741
20 gg/ml MNNG
I
~~
~ ~ ~ ~ ~~
BY4741 mgtl A
__.~~~~~~~~~..~~~~~~~.
._
I~~~~~~~~~~~~
WT BY4741
25 g/ml MNNG
BY4741 mgtl A
MNNG Exposure
Figure 10: MNNG gradient plate assay for Wild-type BY4741 (WT) and
BY4741 mgtl A.
The agar was allowed to settle at an angle in the plates. This permitted variable exposure
of the cells to a fixed concentration of MNNG. The same numbers of cells were
inoculated on the surface of agar plates with increasing concentrations of MNNG (527
IIC.-I_··---·II·P---··-··-·-^
C
-_
I··))I)·IIIIIII__Il---LI
-YIII_-·-
---__
25gg/ml). The cells were grown at 30°C for 3 days before being observed and
photographed. The increased sensitivity of mgtl to killing by MNNG can be attributed to
the lack of MTase.
MNNG-induced cell killing
MNNG-induced cell killing in cultures was compared between WT and mgt/.
Exposure to MNNG at 30 gg/ml cause significantly more killing in mgtl than in the WT. The
lack of MTase and the inability to repair alkylation induced repair in the mgtl imparts this
difference.
100
10
-o--0
20
40
BY4741
BY4741mgtl
60
Length of Exposure (in minutes) to 30 gg/ml MNNG
Figure 11: MNNG-induced killing in wild-type (WT) and MTase
deficient yeast (mgtl).
WT and mgtl cells were picked and grown separately in 5ml YPD (wvith G418), overnight.
Then, 10 0 dl was transferred into 150 ml of media (with G418) in a 250 ml flask. After growth
28
-·--·l--LII
111
_11
--.-
_---I·_
lli___
for 12 hours at 30 C, 10 ml of culture was transferred to four 15ml tubes. To each, 1501 of
1% MNNG stock was added to establish the final concentration of MNNG at 301g/ml. After
incubating them for 0, 20, 40 and 60 minutes, the test-tubes were centrifuged. The
supernatant was discarded and the cells were re-suspended in 10 ml of distilled water. Over a
series of dilutions, the re-suspended cells were plated onto YPD agar plates with G418.
Colonies were counted after 3 days of incubation at 30 C. The colony count, for each
duration of exposure (0, 20, 40 or 60 minutes), was compared to the colony count at time
point 0. This was expressed as the percent survival (% survival) for that duration. The %
survival plot for increasing length of exposure to MINNG, for WT and mgtl, is shown in
Figure 11. mgtl was more sensitive to MNNG than WT. In addition, increasing the length of
exposure to MNNG killed more mgtl cells than WT.
Total RNA extraction from exponentially growing cells.
WT or mgtl cells were grown to mid-log phase (OD = 0.8) as described earlier. The
cultures were split into 3 volumes of 50 ml before being exposed to MNNG (30 tg/ml) for a
variable length of time as shown in Table 1. The control samples were mock-treated with
same amount of double distilled water. After exposure for the appropriate length of time, the
cells were spun down in 50 ml tubes centrifuged at 8000 rpm. The cells were snap-frozen and
stored at -80°C before total RNA extraction using the Qiagen RNEasy protocol.
Pre hybridization quality control of total RNA samples
A few of the total RNA samples were tested for their quality using the Agilent
Bioanalyzer system. Lane 1 in figure 12 shows the ladder. Lane 2 in Figure 12 depicts a sample
that had a good 28s:18s ratio (1.84).
29
"PQB--------------
.. a,_ 28s
:'~'
Figure 12: Pre-hybridization quality control gel of total RNA
sample. This sample was from exponentially growing wild-type
(WT) cells treated with 30jtg/ml of NMNNG for 40 minutes.
18s
Ladder Sample RNA
The total RNA samples from the different experimental groups were tested using the Agilent
2100 Bioanalyzer system. This system permits rapid visualization and quality control of the
RNA sample. The rRNA ratio (28S/18S) was between 1.61-1.87 in the tested samples. Quality
of RNA is a prerequisite for efficient in vitro-transcription reaction.
Post-hybridization quality control.
After hybridization to the target Affymetrix YGS98 array, the quality of the arrays was
judged by the following factors; percent present calls, presence of spiked control cRNA,
background values and noise. All these quality control standards were met satisfactorily
(summarized in Table 2). The hybridization efficiency was judged by the percentage of absent
and present calls. On an average, the present calls were > 74% before normalization.
Hybridization controls, BioB, bioC, and bioD represent genes in the biotin synthesis pathway of
E. coli. Cre is the recombinase gene from P1 bacteriophage. The GeneChip Eukaryotic
Hybridization Control Kit contains 20x Eukaryotic Hybridization controls composed of a
mixture of biotin-labeled cRNA transcripts of bioB, boC, bioD, and cre, prepared in staggered
concentrations (1.5 pM, 5 pM, 25 pM, and 100 pM for bioB, bioC, bioD, and cre, respectively).
30
---
·
*111141(··11---------
LI_
The 20x Eukaryotic hybridization controls are spiked into the hybridization cocktail,
independent of RNA sample preparation, and are thus used to evaluate sample hybridization
efficiency to gene expression arrays. BioB is at the level of assay sensitivity and should be
called "Present" at least 50% of the time. BioC, bioD, and cre should always be called "Present"
with increasing Signal values, reflecting their relative concentrations. The 20x Eukaryotic
Hybridization Controls can be used to indirectly assess RNA sample quality among replicates.
The overall intensity for a degraded RNA sample, or a sample that has not been properly
amplified and labeled, will be lower when compared to a normal replicate sample.
Controls that were spiked-in were detected as expected (BioB-90%, BioC, D, Cre -100%).
Their relative intensities were also in accordance with expectations (BioB < BioC < BioD <
Cre). Ideally, the BioB control cRNA is spiked in at the detection threshold (1.5 pM) and
should receive 'present' detection call in approximately 50 percent of all samples. BioB was
present in 90% of the samples. Average background values ideally range from 20 to 100 for
arrays scanned with GeneArray® Arrays. In our data, the average background value was
about 60. Noise is a measure of the pixel-to-pixel variation of probe cells on a GeneChip
array. The two main factors that contribute to noise - electrical noise of the GeneArray
Scanner and sample quality. The datasets had an average noise of 1.4.
Parameters
Percent present calls
Spiked controls
Background value
Noise (Q)
Outcome
> 74%
As expected, BioB < BioC < BioD` <''Cre''
60 (Normal range 20-100)
1.4
Table 2: Parameters of post-hybridization quality control.
Four parameters; percent present calls, presence of spike-in controls in the appropriate
order, low background value and the presence of noise were assessed and found to be
favorable.
31
,_
.
.-.-.
RMA normalization
The data from 42 arrays were normalized using RLA method of normalization
(Bolstad, 2003). The benefit of RMA normalization is depicted in Figure 13. Probe-set
intensities from six arrays (three replicates each, from 2 experimental groups) are shown
before (Figure 13A) and after (Figure 13B) normalization. After normalization, the variation
between the arrays was minimized dramatically. S-plus's Array analyzer module was used to
create and compare these plots directly from the Affymetrix .cel files obtained upon scanning.
Expression Summaries
it
I
Lye:<
I
·
--
'"'-
I
I
I
CEL WT10.3
CEL
mti10 3.CELWT101 CELWT10.2
,Nt10.1CELmgt10.2.CEL
A
r
i
t- -
II
I
t
I
tlO
WT10.3.CEL
CEL ItO.2.CELmgI10
3 CELWT1 1.CELWT10.2.CEL
B
Figure 13: Box plots of intensities before (A) and after (B) RMA quantile
normalization.
The intensities for six sample Affymetrix .cel files (3 WT and 3 mgtl) are plotted on log2 scale
before (A) and after RLMA quantile normalization. The Affymetrix .cel file intensities were
imported to S-plus Arrayanalyzer module and RNLN quantile normalization was performed.
The normalization of the intensities reduces the variation between the samples and thus helps
in making comparisons between disparate sets of oligonucleotide arrays.
32
1"'""1------^---------
Variability between the replicates
The variability between the replicates was assessed by computing the R2 value between
the log2 of RNMA-normalized intensities (Table 3).
Table 3B
Table 3A
WT1
WT2
WT3
WT1
1
0.94
0.96
WTZ2'
0.94
1
0.96
0.95
WT3
I
... ,
mgt
mgt2
mgt3
mgtl
1
0.97
0.94
0.95
mngt2
0.97
1
0.93
1
mgt3
0.94
0.93
1
-
Table 3: R2 values of WT (3A) and mgtl (3B) replicates
R2 value for the intensity plots for comparing the replicate arrays from the wild-type (WT) (3A)
and the methylguanine methyltransferase mutant (mgtl) (3B). The samples used here included
triplicate untreated WT and untreated mgtl arrays.
33
Computation of log2 Expression Ratio
Mean gene expression profiles from all experimental groups were compared to mean
expression from untreated WT. This uniform denominator allows comparisons to be made between
the experimental groups. The numerator was either mean expression from triplicate arrays at a given
time point or mean expression across all the arrays in WT or mgtl (18 each, representing 3 for each
of the 6 time-points). To delineate genes of importance, the log 2 expression ratio (log2ER) was used
to classify genes. A log2 ER > 1 indicates average fold change induction factor of 2 for that
particular gene. Analogously, a log2 ER < -1 indicates an average fold change repression factor of 2.
If log2 ER's were between 1 and -1, they were classified as not significant (NS). The log2 ER's across
all treatment time-points for both WT and mgtl were exported and visualized in Spotfire's functional
genomics module. The expression ratios were the basis of making comparisons between the
experimental groups in this study.
Comparison with the phenotypic database
We also studied the expression profile in the context of the phenotypic database
(htp: /genomicphe not-yping.mit.edu) which includes information on sensitivity of 4800 yeast gene
deletion strains to MMS. These deletions strains are of those genes that are nonessential. It was
shown earlier that several of these genes are important for cellular recovery after mutagen exposure.
In addition, it was observed that transcriptional responsiveness to these mutagens was not predictive
of contribution of a gene to the recovery from the damage.
Basal gene expression profile in methylguanine methyltransferase deficient mutant (mgt).
The basal gene expression profile of a gene in the O6 MeG methylguanine methyltransferase
deficient mutant (mgtl) was assessed by computing a ratio of mean expression in mgtl (for that gene
34
---------'
from triplicate arrays) to mean expression in WT for the same gene (from triplicate arrays). If the
expression ratio (ER) for any gene/ORF was > 2, then that gene was classified as being up-regulated
and if the ratio was < 0.5, then it was classified as down-regulated. A plot of the gene expression
ratio's for mgtl versus WT for the entire gene population is shown in Fig 14. In the examination of
mgti expression profile, 148/9275 (1.6%) genes were found to be up-regulated (maximum fold
change was 14.8x) and 92 genes (<10)
were found to be down-regulated (maximum fold change
was 95x).
Genes up-regulated in basal mgtl
A subset of these genes that were up-regulated in basal mgtl expression, listed by function, is
shown in Table 4A (Appendix of Tables). Genes that were up-regulated but can otherwise be a part
of the environmental stress response are indicated in table 4B. Notable among the genes upregulated in the basal mgtl are those involved in detoxification and drug transport (YER185W,
YOR378W, YHL047C, YELO65W), amino-acid biosynthesis (BATi, LEU) and transport (BAP3,
ALP). Other important genes included metabolism (CHAl) genes, genes involved in cell cycle
control (BATi, BUBi, CKA1), and transcription (CKA, TFG1, GAI80, GCN5, CBP2, RNA15).
Cell cytoskeleton and mitochondrial biogenesis (SMP2, DCGi, GICI, Q0183) and RET/7, a subunit
of DNA polymerase-zeta (Pol-zeta) were also induced.
Genes down-regulated in basal mgtl
A subset of these genes that were down-regulated in basal mgt/i has been ordered by function
and is shown in Table 5A. Sixty-five of them had no known function and have not been shown.
Genes that were down-regulated but is otherwise a part of the environmental stress response are
indicated in table 5B. Among the genes that are repressed were those involved in amino acid
35
---
***3---·--···--·----(l-L-
(MET14, CYS3, and BASi) and carbohydrate metabolism (SUC2, MIG2, HXTI, HXT3 and HXT4)
and transcription (TFSl,
CPR6 and MIG2).
3
2
0)
E
I
o
.2
0
o.
x
1:
(D -2
0)
J
-3
I
0
2000
4000
I
8000
6000
Gene Population
Figure 14: Expression ratio plot for yeast methylguanine methyltransferase
mutant (mgtl).
Expression values (triplicate) of mgtl were compared to wild-type (WT) yeast
expression and log 2 transformed. Points above the y-axis grid line 1 indicate genes
that are up-regulated in mgtl. Points below the y-axis grid line of -1 indicate genes
that are down-regulated in mgtl.
36
··-W··II(-·II--a·C-·a·-·--·------"---
---------
Principal Component Analysis of experimental groups
The data from different experimental groups were also analyzed using Principal component
analysis (PCA) in Matlab v6.0. Three untreated WT arrays 3 untreated mgtl datasets were compared
with 6 datasets each from WT and mgtI treated with MINNG. Each WT and mgtI dataset included
the mean expression from triplicate arrays. The PCA plot (Figure 15) shows 4 distinct clusters of
data representing the untreated WT, untreated mgtl, treated WT and treated mgtl.
10
c. WT+ MNNG
..
a5~~~~.
o
0
0
2
O)- a.WT
...
d. mgMNNG
- -
'
50
- -
.3.
i
-
b.
-85
- -
--
Figure 15: Principal component analysis of the experimental groups.
Three wild-type (WT) arrays (a) and 3 methylguanine methyltransferase mutants (mgt/) (b) were
compared with 6 datasets (each representing the mean from triplicate arrays) from WT treated with
alkylating agent, MNNG (c) and 6 datasets (each representing the mean from triplicate arrays) from
mg/l treated with alkylating agent, MNNG.
37
--------------------
------
Effect of MNNG on transcriptional profile - Cluster Analysis
The log2 expression ratio's (ER's) were calculated for different experimental groups and exported to
Spotfire. Hierarchical clustering (using Wards Method) of log2ER's was used to generate a qualitative
picture of the effect of MNNG on yeast. This heat map was generated using the data from 39 arrays
(3 arrays were used as baseline for comparison).
WT
mgtl
A
A
t - A I t 1I
0
10 20 30 40 50 60 10 20 30 40 50 60
Duration of exposure (in minutes) to MNNG
Duration of exposure (in minutes) to MNNG
_i
> 2x Repression
-~
1
> 2x
Induction
Figure 16: Heat map of the expression ratio from wild-type (WT) and Mtase mutant
(mgtl). Log2 expression ratio (log 2ER) was calculated for 9335 probe-sets by dividing
the mean expression (from triplicates) for each probe-set by the mean expression for
the same probe-set in the untreated WT (WTO). The first column of data represents
the basal expression in mgt/. The other columns of data represent the mean
expression value of triplicate arrays where yeast strains, either WT or mgtl, that was
exposed to MNNG for variable length of time (10-60 minutes). This indicates that
the genome is very responsive to the treatment with MNNG.
38
--911·-----1--------------L---------
--
Temporal effects of gene categories
The temporal effects of INNG on the yeast strains on a genome-wide scale were examined
using the average expression profile for the treated WT (Figure 17A) and mgt! (Figure 17B). About
1200-1400 genes are induced or repressed upon treatment with MNNG in both WT and mgtl.
Figure 17A
Figure 17B
1600
1600
Represnduced
E[ Repressed
1400
H
v
L_
~
1400
l"
1200
1200
aD
o)
) 1000
D1000
2 800
6 800
E 600
z
400
E 600
z
400
o
200
0
-l
--
-,
mgtO WT10 WT20 WT30 WT40 WT50 WT60
Experimental groups
200
0
Fh;
mgtO mgt1O mgt20 mgt30 mgt40 mgt50 mgt60
Experimental groups
Figure 17A: Genes responsive to treatment with MNNG in the wild-type (WT) yeast.
The number of genes that are either induced or repressed in the WT yeast upon exposure to
MNNG for varying lengths of time (10-60 minutes, WT10 through WT60) is shown. The untreated
mgtl (mgt/O) serves as a comparison.
Figure 17B: Genes responsive to treatment with MNNG in the methylguanine
methyltransferase (mgtl) yeast.
The number of genes that are either induced or repressed in the methylguanine methyltransferase
mutant (mgti) yeast, upon exposure to MNNG for varying lengths of time (10-60 minutes, mgtl0
through mgt60) is shown. The untreated mgtl (mg/O) category serves as a comparison.
39
r
35
[ WT Induced
* mgt Induced
El WT Repressed
3 mgt Repressed
a, 30
(4
C
4) 25
0)
4-
O 20
.4
Q.
15
M
C
a, 10
I-
4)
a.
5
-1I
(1
-r
0t
-.
z
w
A:
OM
f
O
_
WJ
UO
X gr Oz
Gn Ca
0
Lj
O
(/>
W
U
Gene Categories
Figure 18: Gene expression responsiveness for some functional categories.
Average gene expression ratios for each gene within a particular functional category were compared
across the experimental groups. The percentages of genes within a particular category, that are
induced or repressed upon treatment with MNNG are indicated.
The expression profiles from entire categories of genes were examined to study the effect of
MNNG on them and in particular if the gene expression was different between WT and mgtl. Mean
fold induction and fold repression for each ORF/gene, across all the WT and mgtl arrays was
calculated. In each category, the percentage of genes that were responsive upon MNNG treatment is
shown in Figure 18. To calculate this, data from 18 WT and 18 mgtl arrays was used.
40
----
lli
----
-'"
WT
WfT
mgt
mgtH
Induced
Repressed
Figure 19: Venn-diagram of gene expression responsiveness in wild-type (WT) and
methylguanine methyltransferase deficient yeast (mgtl) upon treatment with MNNG.
Genes that had a fold change (FC) > 2 that were either induced or repressed in WT and/or mgtl are
represented using the Venn-diagram. This representation allows examination of effects that are
unique to WT or mgtl or common both, upon exposure to NINNG. Gene expression response,
induction or repression, that is specific to WT and mgtl can therefore be distinguished from a
response that is found in both WT and mngtl. The upper-middle panel in the Venn diagram
represents genes that are induced in both WT and mgtl (977 genes) as a common response to
MNNG exposure. Alternatively, genes that are induced only in MNNG treated WT (225), or only in
MNNG treated mgtl (274) are represented by the non-overlapping segments of the 2 circles. The
lower-middle panel in the Venn diagram represents genes that are repressed in both WT and mgtl
(1039 genes) as a common response to MNNG exposure. Alternatively, genes that are repressed
only in MNNG treated WT (145), or only in MNNG treated magtl (191) are represented by the nonoverlapping segments of the 2 circles in the lower panel.
Genes induced specifically in WT upon treatment with MNNG
The upper-left panel in the Venn diagram (Figure 19) illustrates the response that can be
attributed to WT strain upon exposure to MNNG. Upon MNNG-treatment, a total of 225 ORF's
were specifically induced only in the WT. Since the only difference between WT and mgtl is the lack
of MTase because of the deletion, it could be postulated that these genes are induced in the WT
because of MNNG's effect in the presence of NTase. A total of 127 genes had a known function
(listed in Tables 6A and 6B). Table 6B includes genes that are a part of the ESR. Table 6A includes
41
17 (17%) genes that are essential (highlighted in red). Interestingly 12 (12%) of the induced genes
were involved in protein biosynthesis. Three DNA repair genes UNG/, SIR2 and RAD52 were also
7 , were found to be induced. Table 7B includes
induced. Two mitochondrial genes, RIM and ERT1
30 genes that are a part of the ESR. This included 12 genes (40%) that are otherwise essential in
yeast (highlighted in red).
Genes that are induced specifically in mgtl upon treatment with MNNG
There were 274 genes in this category. 103 of them had a known function and are listed in
Table 7A and 7B. Table 7A includes 87 known genes that are induced in the mgtl upon MNNG
treatment. Of them, 20 (22%) were essential genes. Notable among them were DNA repair genes
PRI2 and CCE1 and several mitochondrial associated genes (MRSI1 TIM8, CCE1, COQ3, DH2
and DICI). In contrast to WT, only 3 protein synthesis genes were found to be specifically induced
in mgtl. Interestingly, SWI6, a substrate of Rad53 in the G(1)/S DNA damage checkpoint was
activated. The homothallic switching (HO) endonuclease, which creates a site-specific double-strand
break (DSB) in the genome at the mating-type (LAT) locus, was also induced. Table 7B includes 16
genes with known function that are induced in mgtl and are a part of the ESR. Nine (56%) were
found to be essential.
Genes repressed in WT upon MNNG treatment
A total of 145 genes were repressed exclusively in WT. Table 8A lists 70 genes with known
function that are repressed exclusively in WT upon MINNG treatment. This included 7 essential
genes (10%). Six genes (SUM1, DOT6, ESCI, SW2, NGG and SET3) involved in chromatin
silencing and histone modification were repressed. Two DNA repair genes (RAD16 and RAD28)
42
--
--------i-
were repressed. Table 8B lists 10 ESR genes that are repressed in WT. Only 1 gene was found to be
essential.
Genes repressed in mgtl upon MNNG treatment
A total of 191 genes were repressed exclusively in mgt/. Table 9A lists 70 genes with known
function that are repressed exclusively in mgtl upon MNNG treatment. This includes 14 essential
genes (20%). Three genes (STNF 1, SPT10, TBFI) involved in chromatin remodeling were repressed.
CKSi, a cyclin-dependent kinase regulatory subunit, was also repressed. Table 9B lists 3 ESR genes
that are repressed in mgtl.
Genes induced upon treatment with MNNG in both WT and mgtl
In contrast to a damage induced response exclusive to WT or mngt, there were several genes
that were induced upon treatment with MNNG in both WT and mgtl. This overlapping response
indicated in the upper-middle panel of Figure 19 included 977 genes that were induced in both WT
and mgtl. Of them, 127 genes were included as ESR genes. The function of 282 genes was not
known. The remaining 568 genes, whose function was known and those that were not a part of the
ESR are listed by function in Table 10A. 121 (21%) of these are essential genes. Briefly, the genes
that were induced upon treatment with MNNG, and were not a part of ESR, included those
involved in maintaining cellular structure and function. Primarily, these included genes involved in
cell wall organization, ergosterol biosynthesis, amino acid metabolism, mitochondrial organization
and biogenesis. A few other genes of interest included chromatin silencing genes (APC5, ISWVI,
ORC3, MRC1, ORC5, NNTi, SWD1, SWD3) genes involved in DNA damage response (HUG1,
DUN, PCL2), DNA recombination (CDC9), DNA repair (HAI/i, MSHI, RAD18, RHCo8,
POLl), DNA replication (POL5, RNR3, RR4, RNR/), DNA topological change (TOF1) and
43
I
DNA unwinding (CDC46, HFM1, MCM2). Specific DNA repair genes included those involved in
nucleotide excision repair (RFA , RFA2, RFA3, CDC2, POL30 and DPB2). Interestingly, 68 genes
(11%) of the genes induced in both WT and mgtl were involved in protein synthesis. Table 10B lists
127 ESR genes that are induced upon treatment with MNNG in both WT and mtH. Interestingly, 57
(44%) of these are essential genes. Eleven genes are involved in ubiquitin-dependent protein
catabolism.
Genes repressed upon treatment with MNNG in both WT and mgtl
A total of 1039 genes were repressed in both WT and mngtl. Of them, 545 had a known
function and are shown in Tables 11A and 11B. Table 11A includes genes that were repressed in
both WT and mgtl and are not a part of the ESR. These included 75 genes (13%) that were essential.
Notable among them were genes involved in cell wall organization, fatty acid metabolism, G1 /S cell
cycle transition genes, mRNA splicing, methionine biosynthesis and mitochondrial organization. The
largest category of genes that were affected was involved in transcription and its regulation. About
43 genes (7%) of the genes belonged to this category. Interestingly several DNA repair genes were
repressed. These included genes involved in DNA recombination and repair (HEX3, SLX8, IXR3 1,
NSE) DNA replication (RRM3, TAHI 1,RTM4) DSB repair (YKU80, SIR4, LRP1, FYV6) nucleotide
excision repair (TFB3, DPBI1, RAD4). Table 11B includes 60 ESR genes with known function that
are repressed in both WT and mgtl upon treatment with MINNG.
DNA damage response and repair genes
The DNA damage response and repair genes were of additional interest and were therefore
examined as a separate class. The mean expression profile of the DNA damage response and repair
genes is indicative of DNA repair pathways are likely to be activated in WT and mgtl in response to
44
-i-----
MNNG. A total of 133 genes were pooled into this category based on Affymetrix annotations that
are derived from the SGD annotations. Twenty five genes were induced (Table 12A) and 20 genes
were repressed (Table 12B). More than half the induced genes are essential or are sensitive upon
deletion (Table 12A). 80 genes did not have an appreciable fold change to be classified as induction
or repression. A total of 8 of the DNA repair and replication genes were classified as ESR genes
(Table 12C).
Temporal effects of individual genes
To further elucidate the differences between WT and mgt, 1 the temporal profiles of
individual genes were examined. Mean expression profiles at 6 time points for WT (3 arrays per time
point) and 6 time points for mgtI (3 arrays per time point) were used. The temporal profile of a gene
indicates a change in its mean expression upon increasing length of exposure to MNNG. The
change was judged by 2 methods; a) the slope of the expression and b) the net fold change. The net
fold-change can be defined as the ratio of expression at the
1 0 th
60 th
minute (Exp6 0) to expression at the
minute (Expl0 ) after exposure to MNNG. All genes that had an Exp 60 /Exp l0 ratio >2 were
classified as induced and Exp 60 /Exp
0Oratio
< 0.5 were classified as repressed. The Venn diagram in
Figure 20 summarizes the differences and similarities in the responses between WT and mgt/. The
upper panel represents genes that are incrementally induced starting at the first time of exposure
(10 th minute). The lower panel represents genes that are repressed over time, starting at the first time
point. There are unique and shared responses by the WT and mgtl to MNNG.
45
1____·_1_1_·
1_1_
LI_
Figure 20: Venn-diagram of genes that are incrementally induced or repressed
upon increasing length of exposure to MNNG.
The net fold-change (FC) over time, was calculated from ratio of mean expression at
the
6 0t'
minute to mean expression at the 10th minute. A cut-off of FC>2 (for
induction) and FC<0.5 (for repression) was used to select genes that have been
included in this representation.
Genes induced over time
When all such profiles were examined, the expression of 60 genes was found to have an
incremental induction over time in the WT. Similarly, 210 genes in mgtl were found to increase upon
increasing the length of exposure to MNNG. The incremental response of 39 genes was common to
WT and mngtl. Induced genes that were known to have a function are represented in Table 13. In the
WT, only 2 (out of 24 genes with known function, 8o) were essential. In contrast, 34 (out of 79
genes with known function, 43/'0) were found to be essential genes.In the WT, DNA damage
effectors HUGI and RNIR3 were incrementally induced with increasing length of exposure to
MNNG. In contrast to WT (61), there were more genes in mgtl (210) that were responsive. These
46
-I------------·------
-
-
-
--------- ---
genes are involved in ribosomal RNA processing, mRNA processing and transcription were found
to be incrementally induced. Several other ESR genes were also found to be a part of pre-rRNA
processing and ribosomal protein synthesis. Overall, there were about 28 genes (35%) involved in
mRNA, rRNA, tRNA and ribosomal function, that were incrementally induced in mgt/ upon
exposure to MNNG over time. 20 (out of 52 genes) fitted a profile of a dramatic initial repression
(after 10 minutes of exposure to MNNG) followed by a steady increment towards the basal levels.
In contrast, 7 genes were consistently induced. The reflex response by 20 genes is likely to represent
a response that follows the perturbation. The 7 genes that are incrementally induced over and above
basal levels are likely to represent processes that are integrated with damage response and recovery.
Genes repressed over time
Ninety genes were repressed exclusively in WT. In contrast 118 genes were repressed in
mgtl. Only 35 genes were seemed to be repressed in both WT and mgt1/. Repressed genes that were
known to have a function are represented in Table 14. The corresponding sets of genes that belong
to ESR are also listed.
In the WT, only 1 gene (out of 18) was found to be essential. Notable among them was the
homothallic switching (HO) endonuclease, which creates a site-specific double-strand break (DSB)
in the genome at the mating-type (LAT)
locus. This gene was listed earlier as a gene that was
induced in mgt/ but not in WT. Closer examination of the profile revealed that there is an
instantaneous induction of HO after the first exposure to MNNG in both WT and mg/tl.
While this
induction is sustained in the mgti (hence the mean log, ER>2), there is a decline in the induction
over time in the WT (therefore, explaining its classification as "repression").
Genes that were repressed in mgt/ includes 6 genes (out of 47) that are essential. Notable among
them are genes involved in protein folding, methionine metabolism and translation. 18 (out of 47
47
'-11111
1-------------·--I---···-----------
genes) fitted a profile of a dramatic initial induction (after 10 minutes of exposure to MNNG)
followed by a decline. In contrast, 13 genes were consistently repressed beyond the basal levels. The
reflex response by 18 genes is likely to represent a response that follows the perturbation. The 13
genes that are incrementally repressed below basal levels are likely to represent processes that are
integrated with damage response and recovery. The differential response between WT and mgtl can
provide valuable insight into the differences in the mechanism of response to MNNG in the
presence and absence of MTase.
48
A
DISCUSSION
Several authors have studied the genome-wide transcriptional effects of chemical and
physical agents on yeast using microarrays. The yeast transcriptome response to MMS Jelinsky and
Samson 1999) and MNNG (Jelinsky et al 2000) have yielded valuable data on how the yeast adapts
to these alkylating agents. Specifically, these studies explored the transcriptional response of S.
cerevisiae to a wide range of chemical and physical damaging agents in an attempt to delineate the
response of each ORF to these agents. Agarwal et al 2003, studied the genome-wide effects of
antifungal agents on yeast in an attempt to characterize their mechanism of action. Several other
microarray-based studies have examined the effect of single gene deletions on yeast transcriptome.
For example, Ohkuni et al 2003, studied the genome-wide expression in the Deltanapl cells in order
to study the transcriptional control of NAPI, a nucleosomal assembly protein. Fry et al 2003,
studied the effect of SGS/ deletion on transcriptional profile in yeast because of its homology with
human genes involved in Werner and Blooms syndrome. Gasch et al 2001, studied basal expression
profile in MECI, DUNI and CRTI deficient yeast and their transcriptional changes in response to
MMS. The study uncovered the role of MECI, the human ATR homolog in yeast. It was also
concluded that MEC1 was an integral part of controlling the environmental stress response. A
literature survey yielded no genome-wide expression profile studies where deletion strains of DNA
repair genes were used. The current study is the first report of the transcriptional responsiveness
where a known DNA repair deficient (06 methylguanine methyltransferase deficient strain, mgtl) has
been studied. While undertaking this study, the goals were two-fold. The first was to study the
effects of deletion of mgtl on the yeast transcriptome. The second was to study the effect of
alkylation induced transcriptional changes over time in the WT and mgtl and to examine the
differences between them.
49
Ir*ll
-*II------·-----·---
-F-----?l-
----·l----·-a*l---
'
·---·Il-------"I--------
-·
MNNG specific response genes enriched by culling out genes involved in environmental
stress response (ESR).
In response to environmental perturbations, S. cerezisiae cells elicit rapid transcriptional
reprogramming involving both activation and repression of gene expression. Some of these
transcriptional changes represent responses that are common to chemical and physical stresses.
Removing these ESR genes from the observed response, will help to enrich the set of genes that are
specific to MNNG treatment and/or presence or absence of MTase. This was achieved by
comparing the expression profile in the current study with ESR dataset from Gasch et al 2000 to
identify ESR genes that might confound interpretation of the data. In the second tier of comparison,
phenotypic
sensitivity
information
of
yeast
deletion
strains
from
(http/ noenomicphenotyping.mit.edu) was incorporated along with expression.
Deletion of MGT1
induces dramatic basal transcription changes, activates cell cycle
checkpoints, transcription factors and a gene involved in spontaneous mutagenesis.
MGT is involved in direct repair of 06 Methylguanine and O 4 MIethylthymine lesions. Genes
that are induced upon deletion of MGT1 are likely to be involved in a) a direct interaction with
MGT1 or b) a downstream effect of a lack of AIGT1. A direct interaction can follow from the
argument that MGT1 normally, represses these genes and removing MGTI induces them.
Alternatively, lack of MGT/ can lead to increased spontaneous mutations and DNA damage and
genes induced might be a part of the downstream cellular processes that are involved in handling
this damage. It was interesting to note the up-regulation of 148 genes upon the deletion of one
single gene (MGT1). The deletion appears to up-regulate a gene involved in spontaneous
mutagenesis and several genes that are transcription factors and 2 others that are involved in cell
cycle control. The 7-fold up-regulation of REIT'7, a subunit of DNA polymerase-zeta (Pol-zeta, Pol
50
-----
) is important in the context of spontaneous mutations. The other subunit of pol-zeta, is REV3
(which was not up-regulated). RE1/7 is the processivity factor for REV3 and complex together, to
get involved in translesion (TLS) synthesis, a mechanism that probably helps cells cope with DNA
lesions that have escaped the efficient DNA repair systems. TLS is invoked when there is a
replication blocking lesion that the normal polymerases are not able to copy past. O 6Methylguanine
(O 6MeG) is a mutagenic lesion and is not considered as a replication blocking lesion. It is likely,
therefore, that the endogenous lesion leading to the up-regulation of RETV7, is probably not because
of 06 MeG. At the same time that TLS helps to copy past the lesion, it has potentially mutagenic
consequences making it responsible for the majority of spontaneous mutations (Friedberg 1995).
Increase in spontaneous mutations in mgtl has been observed earlier (Xiao and Samson 1992). The
same study found that wild-type mutation rate was restored when the mg/t mutant was transformed
with a functional MGTI. The seven-fold induction of RET'7 in this context may suggest a
downstream effect of the deletion of MGT1 rather than via 06 MeG. Cell cycle control genes, BUB/
and CKA1 were up-regulated. BUBi is a protein kinase and serves as a mitotic spindle checkpoint.
CKA
is the alpha unit of protein kinase CIK2 and is known to be involved in DNA damage
response and cell-cycle control. Among the transcription factors that were up-regulated GCN5 has
two tiers of significance. First, Gcn5p plays a role in controlling the expression of 5 % of the yeast
genome (Holstege et al, 1998). Secondly, GCNS, a histone acetyl transferase allows efficient access
of the repair machinery to chromosomal DNA damage either indirectly via influencing transcription
or directly via modifying chromatin structure. Gcn5 functions before or during the DNA repair
process. An earlier report suggested that Gcn5 is recruited upstream of the damaged area by a
hitherto unknown DNA damage sensor (Teng et al, 2002). Overall, it appears as though the deletion
of MGT1 is leads to increased DNA damage good reason why cell cycle checkpoints are upregulated. Simultaneous up-regulation of a key component of the translesion synthesis suggests the
51
----
role of error-prone damage tolerance mechanisms in response to possible replication blocking
lesions. Up-regulation of GCNS, a gene that aids DNA repair and controls expression of 5% of the
yeast genome is indeed remarkable.
Cellular response to MNNG in WT and mgtl
About 30% of the yeast genome (as represented on the Affymetrix GeneChip MArray YGS98)
was responsive to treatment with MNNG (Figure 16). As the heat map in Figure 16 indicates,
response to most of the damage that was inflicted by MNNG was initiated in the first 10 minutes of
exposure. Thereafter, the total number of responsive genes did not change dramatically. Evaluation
of the genome-wide response across the length of time might lead us to miss changes in a subset of
genes that might be instrumental in understanding the response to MNNG. Therefore, in order to
dissect the transcriptional response further, genes belonging to several functional sub-categories
were examined. This, however, did not yield any substantial change in the responsiveness in gene
expression over time, to MNNG (Figure 17A and 17B).
MNNG induced damage activates cell-cycle checkpoint cascade, DNA damage signal
amplifiers and downstream effectors
As the cells respond to an adverse condition such as exposure to MNNG, several cellular
responses are mounted by WT and mgtl. The Venn diagram (Figure 19) indicates that a majority
(977 genes, 66%) of this response was common to both WT and mgtl. The response that is shared
by WT and the gtl is indicative of cellular processes that are common to both strains in response
to MNNG. Among the shared response are genes that serve as a part of S-phase checkpoint. The
checkpoint regulatory mechanism has an important role in maintaining the integrity of the genome
and results in a temporary cessation of DNA replication. Eukaryotic cells activate checkpoint
52
--P
g---·,
IIC
,
._
,
.___
.__
pathways that arrest cell cycle progression and induce the expression of genes that are required for
DNA repair. This checkpoint machinery consists of proteins that recognize DNA damage and
initiate the signaling response. The identification of the damage also needs to be amplified in order
to recruit other mediators of DNA damage response.
MRC! and TOF! are DNA damage signal
amplifiers. TOFI and MRC1 were induced in both WT and mgtl. Upon damage to DNA, TOF1 gets
activated and forms a part of a replication-pausing complex. TOFI, located at the arrested forks
activates checkpoint cascades, leading to repair of the damaged DNA. Recently, it was demonstrated
that Tofl and Mrcl interact directly with the damaged DNA (atou
et al 2003). It has also been
postulated that Toflp links Meclp with Rad53p (Foss, 2001). This is an interesting finding in the
context that MEC! and RAD53 is an indispensable component of DNA damage response. Rad53
and Mecl are protein kinases required for DNA replication and recovery from DNA damage in S.
cerevisiae. DNA damage during S phase slows down the rates of replication fork elongation (Tercero
and Diffley, 2001) and triggers a Rad53/Mecl-dependent block. As a result, DNA damage leads to
an abrupt decrease in DNA synthesis (Paulovich and Hartwell, 1995). In addition, Mecl and Rad53
are required to prevent DNA damage-induced collapse of replication forks (Tercero and Diffley
2001), via their ability to phosphorylate replication and repair proteins at stalled replication forks.
The essential function of Mecl and Rad53 in S. cererisiae is to promote deoxyribonucleotide
triphosphate (dNTP) production during S phase to coincide with DNA replication. This is achieved
via phosphorylation and subsequent degradation of Smll (Zhao et al., 2001), an inhibitor of
ribonucleotide reductase (RN\R). Ribonucleotide reductase (RNR) catalyzes the rate limiting step in
the production of deoxyribonucleotides needed for DNA synthesis. Its synthesis is tightly regulated
at the level of transcription. It is cell-cycle regulated and provides a metabolic state that facilitates
DNA replicational repair processes. Dunl, a protein kinase, controls inducibility of RNR, 2 and 3
53
-`""""""~~"~~~""I~~I--
in response to DNA damage and replication blocks. RNR genes in yeast form a regulon that is
coordinately regulated by protein phosphorylation in response to DNA damage.
In our dataset, the log2ER for MEC! and RAD53 was not induced more than 2 fold. It is
likely that this is because they are kinases and hence present transiently. DUN and HUGI are DNA
damage response genes down-stream of MECI and RAD53 and were induced. HUG (hydroxyurea
and UV and gamma radiation induced) is a component of the MVEC/-mediated checkpoint response
to DNA damage and leads to replication arrest. The HUGI gene was identified as a component of
the DNA-damage checkpoint response using deletion and overexpression mutants of S. cerevisiae
(KIaplun 2000). DNA damage-specific induction of HUG/ is independent of the cell cycle stage.
HUG! induction also increased with increasing exposure to MNNG in both WT and mgtl. Its
induction response to MNNG is therefore consistent with its role in DNA damage response.
MNNG induced damage activates chromatin silencing.
Mecl is the central transducer of these stress-response signals (Zhou and Elledge 2000).
Both Rad53 and Mecl are key proteins involved in the response to replication blocks and they act
together with a novel regulator of Rad53, Mrcl. The DNA damage response pathway has been
linked to the control of chromatin organization. In response to DNA damage, certain proteins that
are normally relocalize to silence telomeric chromatin (Martin et al. 1999, Mills et al. 1999). This
relocation is dependent on Mecl (Craven and Petes 2000). In the current data, 6 genes involved in
chromatin silencing are induced and 3 of them are also known to be essential. This chromatinmediated maintenance of transcriptional inactivation is in accordance with expectations.
54
-3·1··^-·1----·-·-·--·-···------- 111111(---·-·-----L-l·ll·
--- ·--- ·
MNNG induced damage activates genes involved in DNA replication and repair.
DNA repair mechanisms are by far the most important components of the cellular response
that gets induced upon damage to DNA by MNNG. DNA polymerase alpha (POLl) is an essential
gene required for initiation of replication and lagging-strand synthesis. It was also found to be a part
of the ESR. The MCM2 is a part of the Mcm2-NIcm7 protein complex that forms a DNA helicase
that unwinds the DNA ahead of the replication fork (Labib and Diffley, 2001). Additionally, all the
essential subunits of replication protein A (RPA) were induced. RPA is a single-stranded DNA
binding protein (SSB) involved in DNA replication, recombination and repair (Kim C et al 1992). It
has been recently shown that RPA facilitates telomerase action (Schramke, 2004). Proliferating cell
nuclear antigen (PCNA), encoded by the POL30 gene, is essential for DNA replication (in
association with RFC) and DNA repair. PCNA is a ring-shaped DNA polymerase accessory protein
that can encircle duplex DNA. PCNA interacts with Pol eta to permit efficient lesion bypass.
Another notable gene that was induced in response to MNNG was H,4MI. It is known that
overexpression of the yeast HAM1 gene prevents 6-N-hydroxylaminopurine mutagenesis in E. coli
(Kozmin et al 1998) suggesting that it might play a protective role in MNNG induced damage.
HAM/ controls 6-N-hydroxylaminopurine (HAP) sensitivity and mutagenesis in S. cerevisiae. The
HAMVl protein protects the cell from HAP, either on the level of deoxynucleoside triphosphate or
the DNA level by a yet unidentified set of reactions (Noskov, 1996). It was intriguing to note that
H1AMl deletion phenotype was not sensitive to MNIMS
Repair of MNNG induced mitochondrial DNA damage
CDC9 gene encodes a DNA ligase protein that is targeted to both the nucleus and the
mitochondria and this yeast rely upon a single DNA ligase, Cdc9p, to carry out mitochondrial DNA
replication and recovery from both spontaneous and induced mitochondrial DNA damage
55
-
__.
...__
.___..;
(Donahue 2001). MSH1 is a DNA-binding protein in yeast mitochondria that recognizes nucleotide
mismatches in DNA and plays a role in mitochondrial mutation avoidance. MSH1 protein is targeted
to the mitochondria where its mitochondrial-targeting sequence is removed (Chi and Kolodner,
1994). Taken together, the induction of CDC9 and MSH1 appears to be a part of a program to
repair damage to mitochondrial DNA.
MNNG induced damage activates ubiquitin mediated protein catabolism
Twenty genes involved in protein ubiquitination and ubiquitin mediated protein catabolism
were induced (RPTI, DOA41, RPN2, RPNI, PRE9, SCL1, RPN9, UFD2, RPN7, RPT5, PRE8).
Among them, DOA1 is thought to encode a regulatory component of the proteasome pathway,
which involves ubiquitin (Ub)-dependent protein degradation (Ghislain, 1996).
MNNG induced damage activates protein synthesis genes
The yeast ribosomal proteins (RPs) of are encoded by more than 100 genes. These are
among the most transcriptionally active genes in the yeast genome. It consumes a prodigious amount
of the cell's resources and, consequently, is tightly regulated. Interestingly, 68 genes (11%) of the
genes induced in both WT and mgtl were involved in ribosomal protein synthesis. This is in contrast
to earlier observations where protein synthesis genes were found to be repressed upon exposure to
MMS Jelinsky et al 1999) and osmotic stress (Rep, 2000). Notable among these include RAP1, a
multifunctional transcription factor that has a BRCT domain. The BRCT domain is found
predominantly in proteins involved in cell cycle checkpoint functions responsive to DNA damage.
RAP! is essential for cell viability and can function as either an activator or a repressor of
transcription, depending upon the context of its binding site. RAP! was incidentally found to be
56
-- · · 1-·1111·-----·-
----
repressed and it is likely that repression of RAPI is responsible for induction of the 68 ribosomal
protein synthesis genes.
In an earlier study MAG, 3-methyladenine DNA glycosylase, an integral component of the base
excision repair pathway, was shown to be induced upon damage to DNA by MMS. It was also
shown that MAGi and MGTI have a common upstream regulatory sequence. In this dataset,
MAG was not found to be induced more than 2-fold.
Ubiquitin-proteasome regulator, RPN4, is repressed upon MNNG treatment.
Intracellular proteolysis in yeast occurs mainly via the ubiquitin-proteasome system.
Expression of this system is under the control of the transcription factor, Rpn4p (Mannhaupt G et
al, 1999). It has been shown earlier that alkylating agent IMMS resulted in activation of genes that are
involved in ubiquitin and 26S proteasome-dependent protein degradation. Rpn4p is a major
transcription regulator that acts by binding to proteasome-associated control element (PACE) with a
unique upstream activating sequence (5'-GGTGGCAAVt-3'). This binding either stimulated or
inhibited transcription. In this dataset, however, RPN4 was found to be repressed 4.5 times.
Genes involved in non-homologous end joining (NHEJ), recombination and some elements
of nucleotide excision repair are repressed in MNNG exposed yeast cells.
Among the genes repressed in both WT and agtil are those involved in DSB repair via
NHEJ. (YKU80, SIR4, LRP/, FYT6). The components of the non-homologous end joining
(NHEJ) repair pathway were repressed 2-3.5 fold. Additionally, few other genes involved in
recombination and replication were also repressed. HEX3 and SLX8 have DNA binding activity
and are implicated in recombination repair. SLX8 is required to resolve recombination intermediates
that arise in response to DNA damage. The RAD4 gene of yeast is required DNA binding and for
57
.
.~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~-
the incision of damaged DNA during nucleotide excision repair (NER) (Owsianik 2002). TFB2 is a
transcription/repair factor (TFIIH) subunit that is a transcription initiation factor required for NER.
Finally, two discrete set of genes, the "WT-specific MNNG-damage signature" and "mgtlspecific MNNG-damage signature" was created. These include uniquely up-regulated in the WT or
mgtl upon exposure to MNNG. This signature may help define new candidates for involvement in
cellular responses to MNNG in a wild-type and upon O6 methylguanine methyltransferase deletion.
MNNG induces genes involved in mRNA turnover and protein synthesis in WT
Since the only difference between WT and mgtl is the lack of MTase, it could be postulated
that the set of transcripts are specifically induced in WT are because of MNNG's effect in the WT
MTase background. Expression of 7 genes involved in mRNA splicing (MSL1), cleavage (PTIl) and
catabolism (NMD4, PUB1), specifically in the WT is indicative of more mRNA turnover. There
were 3 others that were classified as a part of ESR. Accordingly, there were 18 protein synthesis
genes (6 were ESR genes) that were induced. It is likely that the overwhelming induction of protein
synthesis genes is representative of a recovery response after the initial insult. More protein synthesis
genes are induced in the WT as opposed to mgt, probably because it is able to recover faster (than
the mngtl). Mitochondrial DNA damage response was exemplified by induction of RIM1. This has
single stranded binding (SSB) activity and is involved in mitochondrial genome maintenance. RIMI
forms an essential component of the yeast mtDNA replication apparatus (Van Dyck, 1992). ERVl
gene is essential for cell viability and for the biogenesis of functional mitochondria.
Repression of chromatin remodeling genes following MNNG induced damage
While the genes that are specifically induced in the WT upon NINNG damage are involved
in fundamental metabolic processes, the genes that are repressed include 5 chromatin remodeling
58
_C___l_____q___l_____
and histone modification genes (of 18). In addition, RPI1, a repressor of the ras-cAMIP pathway and
UBP10, a deubiquitinating enzyme are repressed. Loss of UBPIO function is known to lead to partial
impairment of silencing at telomeres. A study of ubpl0 deletion revealed that it mimicked oxidative
damage by intracellular accumulation of reactive oxygen species and eventually leading to DNA
fragmentation and phosphatidylserine externalization, which happen to be the 2 markers of
apoptosis (Orlandi, 2003).
MNNG induces mitochondrial damage in mgtl
It appears as though the lack of NITase in magt1 results in increased damage to mitochondria.
This was reflected in the responsiveness of several mitochondrial proteins that were induced
specifically in azgt1. These genes are involved in fundamental biochemical processes in the
mitochondria. MRS1 1 and 71718 are protein transporters in the mitochondria. CCEI is involved in
DNA recombination and is also present in the inner-mitochondrial membrane. C0Q3 is involved in
ubiquinone biosynthesis, and is a component of the inner mitochondrial membrane. IDH2 is
involved in the TCA cycle and also localizes to the inner mitochondrial membrane. DICI is involved
in dicarboxylic transport across the mitochondrial membrane. There were only 3 protein synthesis
genes
MNNG induced damage activates multifunctional transcription factor SWI6, in mgtl
SWIF6 is a transcription factor involved in controlling genes involved in cell wall biogenesis
and architecture. It is also a key component of G1/S checkpoint. When a cell detects damaged
DNA, Rad53 checkpoint kinase activity is dramatically increased, which ultimately leads to changes
in DNA replication, repair, and cell division. S7IFq6, a substrate of Rad53 in the G(1)/S DNA
damage checkpoint was activated in mgt indicating that there is more DNA damage in mgt/. SIF716
59
enhances the expression level of the recombination genes in meiosis in a dosage-dependent manner,
which results in an effect on the frequency of meiotic recombination (Leem 1998). Another gene
involved in recombination is CCE1.
Ho endonuclease introduces a site-specific double strand break (DSB) in the mating type (MA4I)
gene of yeast and its expression is tightly regulated. This endonuclease is known to be induced for a
short duration and quickly degraded via the ubiquitin-26 S proteasome system (Kaplun, L et al
2000). Taken together, it appears as though mgtl has a propensity to undergo more mitochondrial
damage and DNA recombination repair.
An exaggerated response of mgtl-specific MNNG-damage was limited to genes that seem to be
involved in remodeling of the cell cytoskeleton, translation and signal transduction activity.
Role of environmental stress response genes
The environmental stress response, ESR is a stereotypical pattern of changes in the
expression of approximately 900 genes evoked by a large variety of environmental stresses, including
heat shock, osmotic shock, DTI, nitrogen starvation, and peroxide. Many of the genes in this
program are induced in response to stressful environments and therefore may play a critical role of
maintaining internal homeostasis. As expected, the ESR was rapidly initiated in wild-type and mgtl
cells responding to MNNG, and it was sustained through the entire course of the experiment. From
Gasch et al 2000, a total of 95 microarray hybridization experiments were used to deduce the
environmental stress response genes that were found to be responsive in this study. Approximately
48% of the genes found in the ESR were also induced/repressed upon treatment with MNNG.
Approximately 78% of these ESR genes were common to the response by WT and mgtl. This
overlap is very suggestive of cellular responses that are common to processes that help the cell
survive and achieve internal homeostasis.
60
IC
____
In an earlier study Gasch et al 2000 reported the presence of a DNA damage signature
cluster comprising nine genes including the ribonucleotide reductase subunits RNR2 and RNR4, the
DNA-damage repair genes RADS1 and RAD54, the DNA-damage activated kinase DUN, the
DNA-damage-inducible mitochondrial nuclease DIN7, PLM2, which has homology to the forkhead associated-domain found in several transcription factors and kinases, and two uncharacterized
ORFs (YER004W and YBR070C). A total of 4 genes from the 9 from their cluster (DUNI, PLM2
RNR4, and the ORF YBRO70C) were found to be induced in the current study.
Temporal response to MNNG.
A comparison of temporal expression profiles of genes incrementally induced or repressed
in WT and mgtl revealed interesting trends. Overall, there were more genes that showed induction
or repression in the mgtl (210 induced, 118 repressed) than WT (61 induced, 90 repressed). There
were a few genes that showed trends (39 induced, 35 repressed) that were shared between WT and
mgtl. This result indicates that the perturbation in the mgtl is more profound than in the WT.
In the WT there is a reflex repression of several fundamental biochemical processes when
the cells are first exposed to MNNG. These processes include glucose metabolism, lipid signaling
pathways, fatty acid metabolism, electron transport system and the glyoxylate cycle. The repression
is transient and is eased as the cell tries to recover from the perturbation. Almost simultaneously,
there is a reflex induction of processes involved in maintaining the cell wall structure and function.
Their induction wanes over time. Other genes that follow this pattern are involved in
phosphatidylethanolamine and serine biosynthesis and threonine catabolism. Other interesting genes
in this category included, TLC1 which encodes the RNA subunit of telomerase (Singer, 1994) and
HO endonuclease. The components of the DNA damage response pathway are known to degrade
HO endonuclease via the ubiquitin 26s proteasome system (Kaplun L). This could explain the
61
_________I___C__·___1____11_
waning of "induction" over time. These responses are likely to be related to processes that are
involved in damage recovery. A few genes were consistently induced after the initial exposure to
MNNG. These responses need to be distinguished from the ones that are observed above. These
responses are sustained as long as the cells are exposed to MNNG. Therefore, these are likely to be
critical and related to processes that are directly related to the damage caused by MNNG or its
downstream processes. To exemplify, GTT2 codes for a glutathione transferase and its deletion
strain is also found to be very sensitive to MMS. In addition to its role as an antioxidant, glutathione
has several physiological functions, such as detoxification of various cytotoxic compounds, acting as
a co-factor for enzymes, protection of proteins' SH groups. Upon invasion by xenobiotics,
glutathione-S-conjugates
are formed by glutathione S-transferase and the conjugates or their
degraded compounds are exported from the cytoplasm by some transporters. In S. cerevisiae, two
glutathione S-transferase genes (GYTI
and GTT2) have been identified (Choi 1998). and
glutathione-S-conjugates are transported into the vacuole by the YCF1 gene product, which is an
ATP-binding-cassette transporter on the vacuolar membrane (Li, 1996). Induction of GSH synthesis
in yeast has been shown to protect the mitochondrial DNA (mtDNA) from oxidative damage (IKeiichi, 2001).
The damage by MNNG is more profound in mgtl than WT. Unlike the WT, the initial
exposure to MNNG represses genes that are more likely to be transcriptional factors. After the
initial repression these genes were incrementally induced and tended to approach WT levels. By
virtue of affecting transcription factors, their influence on the expression profile is more
pronounced than what is apparent. Other genes in this category included ones involved in RNA
processing and kinases involved in protein phosphorylation. It was very intriguing to find RAD28 in
this category of genes. It is a homolog of the Cockayne syndrome A (CSA) gene. CSA patients
exhibit severe developmental and neurological abnormalities. In contrast genes that were induced as
62
-9-.··11--·----------·
·
·-
a reflex response and waned over time included those involved in copper and lipid transport, glucan,
sulfur and methionine metabolism. An interesting class of genes here were 5 genes involved in
translation initiation, elongation and regulation. The temporal responsiveness in mgt/ was striking on
several other counts. About 35% of the induced genes were involved in ribosomal function and 43%
of all the genes induced were found to be essential.
Overall, the propensity for damage to macromolecules and cellular processes is much more
in the mgt/ than the WT. Given that mgt
is not able to repair O6 Methylguanine and
04Methylthymine, it can be postulated that the profound damage in mgti could be because of these
lesions itself or due to downstream processes. The repression of transcription factors followed by
the induction of ribosomal and RNA components is an interesting finding that suggests that the
recovery after damage is coordinated at the transcriptional level. This is unlike the WT that can
repair the O 6MeG lesion to a greater extent than the mgtl. In the WT, the reprogramming involves a
transient repression of genes restricted to fundamental metabolic processes indicating that the
damage is limited compared to mgtl.
63
."·*IC-"PIC·II.*·---·-sl-·l·1C-----···-
Summary
In conclusion, the transcriptional changes precipitated by deleting MTase in yeast are indicative of
DNA damage induction, cell cycle checkpoint activation and eventually, damage tolerance via
REV7. This finding of error-prone translesion bypass polymerase activity correlates well with an
earlier result where increased spontaneous mutagenesis was demonstrated in MTase deficient yeast.
The effect of alkylating agent MNNG on yeast is dramatic and about 30%o of the genome is instantly
responsive. The initial insult with MNNG is rapidly followed by a repression of major metabolic
processes but an induction of genes involved in maintaining cell wall structure and function. Other
processes that try to maintain homeostasis after the initial insult with MNNG set in early and are
sustained. A reflex response orchestrated by cell-cycle checkpoints serve to stall the cell-cycle and
provide sufficient time to repair the DNA. This was evidenced by induction of DNA damage
sensors, signal amplifiers and effectors. Nucleotide excision repair genes were the predominant class
of repair proteins induced.
While most of the response to DNA damage is shared by the wild-type and mgt1, a fraction of the
genes respond differentially and they include individual components of DNA repair systems. It
appears as though the lack of MTase in mgt/ leads to increased damage in mitochondria and a
program that increases the transcription of genes pre-mRNA processing, mRNA splicing and
ribosomal biogenesis. Damage due to alkylation does not limit itself to genetic material in the
nucleus. It affects organelles such as mitochondria which appear to be very sensitive.
In the WT treated with MNNG, double-strand break repair was induced along with uracil DNA
glycosylase (UNG1). There was more protein synthesis and transport across the subcellular
organelles. In contrast, in mg/tl, there was more mismatch repair (ISH2), and mitochondrial repair
genes. The induction, over increasing length of exposure to MINNG, of 30 genes involved in pre-
64
-~"""""---II-"---I
I-----
RNA processing, mRNA splicing and ribosome maintenance could be attributed to the induction of
SWI16, a transcriptional co-activator.
Culling out environmental stress responsive genes (ESR genes) from the current study permitted a
study of gene expression attributes specific to WT or magtl. Five out of the 9 genes identified as the
DNA damage cluster by Gasch et al 2000 were found to be induced upon MNNG exposure. These
genes included DUN, PLM2, RNR4 and the ORF of yet unknown function, YBR070C.
It appears that the MNNG induced damage is not limited to the DNA alone. Other macromolecular
processes are affected considerably. An equal dose of MNNG imparts more damage in mgtl than the
WT. While the O6 Methylguanine lesion is successfully repaired by the MTase in WT, lack of MTase
in mgtl is not able to do so. With more O6 Methylguanine in the genome, the mgtl cells are killed
faster than WT. The transcriptional changes that accompany imply that the cellular processes in mgtI
sustain more damage. Hence the transcriptional responsiveness is more elaborate than in the WT.
As a rule, the fundamental metabolic processes (glucose metabolism, amino acid metabolism and
fatty acid metabolism) are transiently repressed in order to cope up with this stress and cell wall
synthesis genes are induced in both WT and mgtl.
65
-""-~"-"1-""11------i
Bibliography
Agarwal AK, Rogers PD, Baerson SR, Jacob MR, Barker KS, Cleary JD, Walker LA, Nagle DG,
Clark AM. Genome-wide expression profiling of the response to polyene, pyrimidine, azole, and
echinocandin antifungal agents in Saccharomyces cerevisiae. J Biol Chem. 2003 Sep 12;278(37):349985015.
Bassett, D. E. et al. Genome cross-referencing and XREFdb: implications for the identification and
analysis of genes mutated in human disease. Nature Genet. 15, 339-344 (1997).
Bassett, D. E., Boguski, M. S. & Hieter, P. Yeast genes and human disease. Nature 379, 589-590
(1996).
Baudin, A., Ozier-Kalogeropoulos, O., Denouel, A., Lacroute, F. & Cullin, C. A simple and efficient
method for direct gene deletion in Saccharomyces cerevisiae. NucleicAcids Res. 21, 3329-3330 (1993).
Begley TJ, Rosenbach AS, Ideker T, Samson LD. Damage recovery pathways in Saccharomyces
cerevisiae revealed by genomic phenotyping and interactome mapping. 1Mol Cancer Res. 2002
Dec;1 (2):103-12.
Bhatia P,
Verhage RA, Brouwer J, Friedberg EC. Molecular cloning and characterization of
Saccharomyces cerevisiae RAD28, the yeast homolog of the human Cockayne syndrome A (CSA)
gene. J Bacteriol. 1996 Oct;178(20):5977-88.
Bolstad, B.M., Irizarry R. A., Astrand, M., and Speed, T.P. (2003), A Comparison of Normalization
Methods for High Density Oligonucleotide Array Data Based on Bias and Variance. Bioinformatics
19(2):185-193.
66
-^*"*C·aD1···i··s·-··l--·---
Gasch AP, Spellman PT, JKao CM, Carmel-Harel O, Eisen MB, Storz G, Botstein D, Brown PO.
Genomic expression programs in the response of yeast cells to environmental changes. Mo! Biol Cell.
2000 Dec;11(12):4241-57.
Burrows C. J., J. G. Muller, Oxidative Nucleobase Modifications Leading to Strand Scission.
Chem Rev. 1998 May 7;98(3):1109-1152.
Carell T, Burgdorf LT, KIundu LM, Cichon M. The mechanism of action of DNA photolyases. Curr
Opin Chem Biol. 2001 Oct;5(5):491-8
Chee, M. et al. Accessing genetic information with high-density DNA arrays. Science 274, 610-614
(1996).
Chi NW, Kolodner RD. The effect of DNA mismatches on the ATPase activity of MSH1, a protein
in yeast mitochondria that recognizes DNA mismatches. J Biol Chem. 1994 Nov 25;269(47):29993-7.
Cho, R. J. et al. A genome-wide transcriptional analysis of the mitotic cell cycle. Mol. Cell 2, 65-73
(1998).
Choi, J. H., Lou, W. and Vancura, A. (1998) A novel membrane-bound glutathione S-transferase
functions in the stationary phase of the yeast Saccharomyces cerevisiae. J. Biol. Chem. 273, 2991529922
Chu, S. et al. The transcriptional program of sporulation in budding yeast. Science 282, 699-705
(1998).
Craven RJ, Petes TD. Involvement of the checkpoint protein Meclp in silencing of gene expression
at telomeres in Saccharomyces cerevisiae. Mool CellBiol. 2000 Apr;20(7):2378-84.
67
-·"I---------------
--
"I--
I
Daniels DS, Mol CD, Arvai AS, Kanugula S, Pegg AE, Tainer JA. Active and alkylated human AGT
structures: a novel zinc site, inhibitor and extrahelical base binding. EiMIBO J. 2000 Apr 3;19(7):171930.
Daniels DS, Tainer JA. Conserved structural motifs governing the stoichiometric repair of alkylated
DNA by 0(6)-alkylguanine-DNA alkyltransferase. Mutat Res. 2000 Aug 30;460(3-4):151-63.
DeRisi, J. L., Iyer, V. R. & Brown, P. O. Exploring the metabolic and genetic control of gene
expression on a genomic scale. Science 278, 680-686 (1997).
Donahue SL, Corner BE, Bordone L, Campbell C. Mitochondrial DNA ligase function in
Saccharomyces cerevisiae. Nucleic Acids Res. 2001 Apr 1;29(7):1582-9.
Estruch F. Stress-controlled transcription factors, stress-induced genes and stress tolerance in
budding yeast. FEMS Microbiol Rev. 2000 Oct;24(4):469-86.
Foss EJ. Toflp regulates DNA damage responses during S phase in Saccharomyces cerevisiae.
Genetics. 2001 Feb;157(2):567-77.
Friedberg, E. C., Wagner, R., and Radman, M. Specialized DNA Polymerases, Cellular Survival and
the Genesis of Mutations (2002) Science 296: 1627-1630.
Friedberg, E.C., Walker, G.C., and Siede, W. (1995) DINA Repair and Mutagenesis, American Society
of Microbiology, Washington, DC.
Fry RC, Sambandan TG, Rha C. DNA damage and stress transcripts in Saccharomyces cerevisiae
mutant sgsl. MechAgeing Dev. 2003 Jul;124(7):839-46.
Gasch, A. P., Huang, M., Metzner, S., Botstein, D., Elledge, S. J., & Brown, P. 0. (2001). Mol. Biol.
Cell 12, 2987
68
-91------
--
Gasch, A.P., Spellman, P.T., Kao, C.M., Carmel-Harel, O., Eisen, M.B., Storz, G., Botstein, D.,
Brown, P.O. (2000). Genomic expression programs in the response of yeast cells to environmental
changes. Mol. Biol. Cell 11(12):4241-4257.,
Ghislain M, Dohmen RJ, Levy F, Varshavsky A. Cdc48p interacts with Ufd3p, a WD repeat protein
required for ubiquitin-mediated proteolysis in Saccharomyces cerevisiae.
EMBOJ. 1996 Sep 16;15(18):4884-99.
Goffeau, A. et al. Life with 6000 genes. Science 274, 546, 563-567 (1996).
Hashimoto, H., Inoue, T., Nishioka, M., Fujiwara, S., Takagi, M., Imanaka, T., and Kai, Y. (1999) J.
Mol. Biol. 292, 707-716. J Biol Chem. 1997 Sep 19,-272(38):23465-8
Jacob S, Praz F. DNA mismatch repair defects: role in colorectal carcinogenesis. Biochimie. 2002
Jan;84(1):27-47
Jamieson ER, Lippard SJ. Structure, Recognition, and Processing of Cisplatin-DNA Adducts. Chem
Rev. 1999 Sep 8;99(9):2467-98.
Jelinsky, S. A. & Samson, L. D. Global response of Saccharomyces cerevisiae to an alkylating agent.
Proc. Natl/Acad. Sci. USA 96, 1486-1491 (1999).
Kaplun, L., Ivantsiv, Y., Kornitzer, D., and Raveh, D. (2000) Functions of the DNA damage
response pathway target Ho endonuclease of yeast for degradation via the ubiquitin-26S proteasome
system Proc. Natl. Acad. Sci. U. S. A. 97, 10077-10082
Karran, P., and Marinus, M. G. (1982). Mlismatch correction at 06-methylguanine residues in E. coli
DNA. Nature. 1982 Apr 29;296(5860):868-9.
69
_ll/D
__LIIIIIIIII____------
Katou Y, Kanoh Y, Bando M, Noguchi H, Tanaka H, Ashikari T, Sugimoto K, Shirahige K. S-phase
checkpoint proteins Tofl and Mrcl form a stable replication-pausing complex. Nature. 2003 Aug
28;424(6952): 1078-83.
Kei-ichi S, Atsuki K, Shingo I and Yoshiharu I Role of glutathione in heat-shock-induced cell death
of Saccharomyces cerevisiae Biochemn. J. (2000) 352 (71-78)
Kim C, Snyder RO, Wold MS. Binding properties of replication protein A from human and yeast
cells. Mol CellBiol. 1992 Jul;12(7):3050-9.
Kozmin SG, Leroy P, Pavlov YI.
Overexpression of the yeast HAM1 gene prevents 6-N-
hydroxylaminopurine mutagenesis in Escherichia coli. Acta Biochim Pol. 1998;45(3):645-52.
Labib K, Diffley JF. Curr Opin Genet Dev. 2001 Feb; 11(1): 64-70. Is the MCM2-7 complex the
eukaryotic DNA replication fork helicase?
Lawley PD, Phillips DH. Related Articles, Links
DNA adducts from chemotherapeutic agents.
Mutat Res. 1996 Aug 17;355(1-2):13-40.
Leem SH, Chung CN, Sunwoo Y, Araki H.
Meiotic role of SWI6 in Saccharomyces cerevisiae.
Nucleic Acids Res. 1998 Jul 1;26(13):3154-8.
Li, Z. S., Szczypka, M., Lu, Y. P., Thiele, D. J. and Rea, P. A. (1996) The yeast cadmium factor
protein (YCF1) is a vacuolar glutathione S-conjugate pump. J. Biol. Chem. 271, 6509-6517
Lindahl T and R. D. Wood Quality Control by DNA Repair. Sence, Dec 3, 1999; 286(5446): 1897 1905.
Lindahl T, Sedgwick B, Sekiguchi M, Nakabeppu Y Regulation and expression of the adaptive
response to alkylating agents. Annu Rev Biochem. 1988;57:133-57.
70
-·llsllP-·lsasa3-p-·-
Lindahl T.
Instability and decay of the primary structure of DNA. Nature. 1993 Apr
22;362(6422):709-15.
Loeb L. A, B. D. Preston, Mutagenesis by apurinic/apyrimidinic sites. Annu. Rev. Genet. 1986, 20,
201
Mannhaupt G, Schnall R, Karpov V, Vetter I, Feldmann H. Rpn4p acts as a transcription factor by
binding to PACE, a nonamer box found upstream of 26S proteasomal and other genes in yeast.
FEBS Left. 1999 Apr 30;450(1-2):27-34.
Martin SG, Laroche T, Suka N, Grunstein M, Gasser SM. Relocalization of telomeric Ku and SIR
proteins in response to DNA strand breaks in yeast. Cell. 1999 May 28;97(5):621-33.
Mills ID, Ferguson DO, Alt FW. The role of DNA breaks in genomic instability and tumorigenesis.
Immzanol Rev. 2003 Aug; 194:77-95.
Moore,M.H., GulbusJ.M., Dodson,E.J., Demple,B. and Moody,P.C.E. (1994) Crystal structure of a
suicidal DNA repair protein: the Ada 06-methylguanine-DNA methyltransferase from E.coli.
EMBOJ., 13, 1495-1501
Modrich P Mechanisms and biological effects of mismatch repair. Annu Rev Genet. 1991;25:229-53.
Noskov VN, Staak K, Shcherbakova PV, Kozmin SG, Negishi K, Ono BC, Hayatsu H, HAM1, the
gene controlling 6-N-hydroxylaminopurine sensitivity and mutagenesis in the yeast Saccharomyces
cerevisiae. Yeast. 1996 Jan;12(1):17-29.
Ohkuni K, Shirahige K, IKikuchi A. Genome-wide expression analysis of NAP1 in Saccharomyces
cerevisiae. Biochem Biophys Res Conmzun. 2003 Jun 20;306(1):5-9.
Orlandi I, Bettiga M, Alberghina L, Vai M Transcriptional profiling of ubpl0 null mutant reveals
subtelomeric gene expression and insurgence of oxidative stress response. J Biol Chem. 2003 Nov 17
71
Paulovich AG, Hartwell LH. A checkpoint regulates the rate of progression through S phase in S.
cerevisiae in response to DNA damage. Cell. 1995 Sep 8;82(5):841-7.
Pavlov YI. HAM1, the gene controlling 6-N-hydroxylaminopurine sensitivity and mutagenesis in the
yeast Saccharomyces cerevisiae. Yeast 1996;12:17-29.
Pegg AE. Repair of 0(6)-alklylguanine by alkyltransferases. Mutatl Res. 2000 Apr;462(2-3):83-100.
Ravanat JL, Douki T, Cadet J. Direct and indirect effects of UV radiation on DNA and its
components. J Photochem PhotobiolB. 2001 Oct;63(1-3):88-102.
Rep, M., Reiser, V., Gartner, U., Thevelein, J.M., Hohmann, S., Ammerer, G., and Ruis, H. (1999).
Osmotic stress-induced gene expression in Saccharomyces cerevisiae requires Msnlp and the novel
nuclear factor Hotlp. Mol. Cell. Biol. 19, 5474-5485
Rydberg B, Lindahl T. Nonenzymatic methylation of DNA by the intracellular methyl group donor
S-adenosyl-L-methionine is a potentially mutagenic reaction. EMIBO J. 1982;1 (2):211-6.
Schena, M., Shalon, D., Davis, R. W. & Brown, P. O. Quantitative monitoring of gene expression
patterns with a complementary DNA microarray. Science 270, 467-470 (1995).
Schramke V, Luciano P, Brevet V, Guillot S, Corda Y, Longhese MP, Gilson E, Geli V. RPA
regulates telomerase action by providing Estlp access to chromosome ends.
Nat Genet. 2004 Jan;36(1):46-54.
Sebastian J, IKraus B, Sancar GB. Expression of the yeast PHR1 gene is induced by DNA-damaging
agents. Mol Cell Biol. 1990 Sep;10(9):4630-7.
Sebastian J, Sancar GB. A damage-responsive DNA binding protein regulates transcription of the
yeast DNA repair gene PHR1. Proc Natl Acad Sci U S A. 1991 Dec 15;88(24):11251-5.
72
Sedgwick, 1997. Carcinogenesis. 1997 Aug;18(8):1561-7. Nitrosated peptides and polyamines as
endogenous mutagens in 06-alkylguanine-DNA alkyltransferase deficient cells.
Seeberg E, Eide L, Bjoras M. The base excision repair pathway. Trends Biochem Sci. 1995
Oct;20(10):391-7.
Shuker, D. E. G. & Margison, G. P. (1997) Nitrosated glycine derivatives as a potential source of
06-methylguanine in DNA. Cancer Res 57:366-369
Singer, M. S., and D. E. Gottschling. 1994. TLC1: template RNA component of Saccharomyces
cerevisiae telomerase. Science 266:404-409
Spellman, P. T. et al. Comprehensive identification of cell cycle-regulated genes of the yeast
Saccharomyces cerevisiae by microarray hybridization. AIol. Biol. Cell 9, 3273-3297 (1998).
Sugimura T, Food and cancer, Toxicology, Volumes 181-182, 27 December 2002, 17-21
Swanson, Rebecca L., Natalie J. Morey, Paul W. Doetsch, and Sue Jinks-Robertson. 1999.
Overlapping specificities of base excision repair, nucleotide excision repair, recombination, and
translesion synthesis pathways for DNA base damage in Saccharomjyces cerevisiae. Mol. Cell. Biol.
19:2929-2935
Teng Y, Yu Y, Waters R. The Saccharomyces cereVisiae histone acetyltransferase Gcn5 has a role in
the photoreactivation and nucleotide excision repair of UV-induced cyclobutane pyrimidine dimers
in the MFA2 gene. JMol Biol. 2002 Feb 22;316(3):489-99.
Tercero JA, Diffley JF. Regulation of DNA replication fork progression through damaged DNA by
the Mecl/Rad53 checkpoint. Nature. 2001 Aug 2;412(6846):553-7.
73
Torres-Ramos, Carlos A., Robert E. Johnson, Louise Prakash, and Satya Prakash. 2000. Evidence
for the involvement of nucleotide excision repair in the removal of abasic sites in yeast. Mol. Cell.
Biol. 20:3522-3528
Van Dyck E, Foury F, Stillman B, Brill SJ. A single-stranded DNA binding protein required for
mitochondrial DNA replication in S. cerevisiae is homologous to E. coli SSB. EMBO J. 1992
Sep;11 (9):3421-30.
Venter, J. C. et al. The sequence of the human genome. Science 291, 1304-1351 (2001).
Wach A, Brachat A, Pohlmann R, Philippsen P. New heterologous modules for classical or PCRbased gene disruptions in Saccharomyces cerevisiae. Yeast. 1994 Dec;10(13):1793-808.
Ward, J. H. Hierarchical grouping to optimize an objective function. J. Am. Stat. Assoc. 58, 236-244
(1963).
Wibley, J. E. , Pegg, A. E, P. C. Moody, Crystal structure of the human 06-alkylguanine-DNA
alkyltransferase NucleicAcids Res. 2000, 28, 39
Winzeler, E. A. et al. Functional characterization of the S. cerevisiae genome by gene deletion and
parallel analysis. Science 285, 901-906 (1999).
Xiao W, L Samson. In vivo Evidence for Endogeneous DNA Alkylation Damage as a Source of
Spontaneous Mutation in Eukaryotic Cells PNAS, Vol 90, 2117-2121, 1993
Xiao W, Derfler B, Chen J, Samson L. Primary sequence and biological functions of a
Saccharomyces cerevisiae 06-methylguanine/04-methylthymine
DNA repair methyltransferase
gene. EMBO J. 1991 Aug;10(8)
74
_Illb_
_
1I
Xiao W, Treena Fontanie, Sonya Bawa and Lester Kohalmi REV3 is required for spontaneous but
not methylation damage-induced mutagenesis of Saccharomyces cerevisiae cells lacking 06methylguanine DNA methyltransferase Mutation Research Vol 431, 16 1999, Pages 155-165
Xiao W, Samson L, The Saccharomyces cerevisiae MGT1 DNA repair methyltransferase gene: its
promoter and entire coding sequence, regulation and in vivo biological functions. Nucleic Acids Res.
20 (1992), pp. 3599-3606
Yang W, Rogozin IB, Koonin EV. Yeast POL5 is an evolutionarily conserved regulator of rDNA
transcription unrelated to any known DNA polymerases. Cell Cycle. 2003 Mar-Apr;2(2):120-2.
Zhou BB, Elledge SJ. The DNA damage
response: putting checkpoints
in perspective.
Nature. 2000 Nov 23;408(6811):433-9.
Zhao X, Chabes A, Domkin V, Thelander L, Rothstein R. The ribonucleotide reductase inhibitor
Smll is a new target of the Mecl/Rad53 kinase cascade during growth and in response to DNA
damage.
75
o EO
~0
0
o
t
u
o
0
0o
(0
(D
N
o
0
0
lo
O
LUw
c/)
rn ,=,
e
Q,~~~~~~~~~~~~~~~
oo
c,)
Ci
N C,
1
.
.0
cLL~~
, o IC.2 .
a
M
a)
Eoo aC '.0 (U C
2
0.0
Ci)Q
o
(U
0
-0
0
:, a)
-
0 0
0
)
g
o.
o
0
o..
(
U
E EnE
a
0
:
E.2.2.2
,,
,,
E .0.0
Q.0
0...
)
22o
00. (U
0
0
0
0
-!C
<
o~~~~~~C
(U
(U
0~
(U
_o0..
0... 0
0
0
0
rr,
a.
t-
.
r_ mi
Q
~f
C,
, .
V
co
-
VL
(U
~~ ~XX~
Lc <(
Z , I-IT
~-
i
0
:D
(U .<
Y C.
2- 7
5)
E
t
0Lmm
- 0.
O
)
~0
(N
o, 'D 0O Ir
w In
go
o,
~
Q,
t-
Cc
D m
._a
1:
L
-onwm
00 >
a
m Z5 N (
_C5OOZZCOX:
000
LU
0
a
c
j
o.
L 0LU
(N
0
0
ao
C;
X
o
W
Q
O Q~H~
C~Owm
z
0
>0
rooo
C.)
~
J
4 )
_ (1.;,
'i
(C
a,,
D I
(.0
4)
rPy
(I.,0
1 N(0(0
o
o (
o
~~~~ ~· ~ ~~~~~~~~~~L
2222222Z£
:
D
o
)
I
c
O
s:O
11
m
oo
o
"
CD,--.
a~~~9'
.0
-.
N
.
o'
-~
N N IN
C*
g m
m
o
mo~~o
g
o a
m o m m c,
a~ ~
r-
C)
Q
Lo
o~~~~~~~~~~~
c,,
~
<d O*
r
*-
_
c~d,
Ci)
C
4..U' U( (00( t N - 0
E
,=
tI
((
t
(N
(
(0
=
0
I>N
(
I
(0
<
0t 0(It
t
a,
(0
0(0 N t-)
t - 0
0
0o
0-
o o
m
(0
0
(
o l IDL
I0
C,
C.
-,
0
~
~
V)
r~~~~
0
: C't t 0 '5:.
A
d~~~-
Ž
E
00¢.
0)
0
w)
(A
5% 3t
C)
I-
U..
o
WO
945scC=mC
o
I-
=
3
a'" rS=~21
x
r
Vo
B
2
c
.)
(0(0
<
Qw
-.aWrsgaasLI4ar-.--------------I--
.
p
0
LU
<
(N
0
;
(C.
((E
-2
(E
.2
l(00
-
w
.0
(00
w
0000(00
(U
0
CD 0
C-
IT
(N
C)
(Urif
D
nco
0o
T- 2
E
a)
jC) c,
t
>
Iw
(N
a-
-
C-
-
(N-
)
LE0
)D0
0
o00 UCO~
C)
Z
0
co~~~~~~~~~~~~~~~~~~~a
U
C,
2.
0-DC)
~~cl~~~ ~ ~ ~ ~ ~~~~~v
(0(0
=tar
Ir
E
Ec2
c o '(
a2_ o
E CD
-
(0~~
lD.0LLWn
E0
E
(N
0
o
0
X
0
01
~~~~~~~~~~
m
(D n
4-
0
.
(
(U
C.
e
.
C)(
0
LU~~~~~~~~::
-0
It
Ch
' in
0
L.
ot w
m
.
-E
0
I.
1F
CO
lo
(A
(N
ID
CD
2gE
q)
C)
a
(
cCI
lu
!5
1*
e.
o~~~~~~o~~~
E
a
z z
LLz 0!9
cL
ww
Z
gj
C
CS n
~
U
C L
W)
9:9
CA
CD
N
u
-
w
o
·clg
w
oQ
LL
->->
0
2
____
N-
IC
o
(D O
o
1
0
0
C
C
0
au.
N
q
N
0 C N
c
.5
co
U)
CO
a
C C
N
CO
()
C
r
C)I
.i
C.)
0
C
a,
U,
10
E
(,
U)
o~. BN
o
E
E
0
C
0
E
OO
U)
cn
12
0
U)
0
rr
0
ot
04
.t0
U2
z
co
CO
d
ci
~ i~
i
>.N
4-·
>-
C
.
>
a
C
LL
-
Co
i
. N)
I) .
.0.
11 c0
co
C)
v)
C
cr- x
ry
O
z,
i
v
co
w
L
U)
ao U)
O
N
O
O
O
li
.
(i
0
*
t -t
0
ll
-o
O
o)
)tUc i
C,
0
i-
,, 0;
0
0
0
C
¢
E.
O
o
C)
=
O
U
o
I'
0°
o
C
~O
X<
I
1
a
i
-0
.
0
a:
N
0
)E
c
-
U
o
CC
v,
4*.
4-4
,
B
.0
i0
C
5c)
U.
ce
-r .5 rC
¢)
N>- N-)
0
O I
C
D
~
0)
>-(0
>)
-
N n
2C)->
ct
'5QY
N
.M
----
0
O
---- ·----·-- -
._
U..
0~i
0
'
0
c.J
0
0
I
Cq
C
C
C
0
0
0oi
C
C
0
.W
',I
L
0
0
U
0
N
I(N
(
(
(N
ll
(Nq (N
0
0
IN
0
0
0
(
(
N
N
(
U,
(/L4.,
0
(N
(
(N
0
o
E
E
2
~
E
2
,
o
o
C
2
~5
~~~~E
c
'
0
-
U
DI
E'
I
a
2
·0 E ]~
D
c
0
E
T
ti
a,
U)
C~oo
E
.o 0
~
Z
oo
,. cuEE
o
U)
:2
zC
.U)
E
2
ii
U)a)
1U)U
0
-2
U)
U)
0
0)
-o. u
._
0_
2
0
'
-/)
-2)
<
444
U)
.)
H
UtU
U
2. co
~ofl =~
= ._O
=¢b 0
o ~
[Lu o o o c 62).).
0
Li.2600U)
*0
~
a)(D
=0
C,,,
~.
2Z o T~22
~o LL.
2 o
~
!.u =o
F
u.
~
o ', >u
o
>
U
U)
S (Lz Ž =
<,.uoE
<
_O
2E
.2
~~~~~cZZ
o
_
o
U)
)
)>
IDa'
U)E
m -S2
o
0r
EO N
c~~~lg LL EL
ID
LL 'D 0 I
~ C c)
~
Li
~
'
w
--
z
CL
a.101"~~n
LL
a
0_o
-(N
U)~
0
N
1
0
U
10~~~
· ~~~~~~~~~~~~~
(N
-t
00
-=
N.
U
U)~~~~~~~~~~~~C
-
o
o
o
U
-(
co
.. =..co)o~
o
U)
_
i
o
U
U.
U!
00u
E .,_
Ln
0
v
'
S
: .oo
X
Y
J
Q
1,,2
<,
'
00
(N
LiU
U
o 00
L,
dl
U
0
000
2.)<
4
0
-..
Z~
0(N
C-
NL (N
C.,
C
N
O
C
Lic>
(N.
N0 N
N
0)C:
c(NC,)
U)
(
U)
c
C
-C
2
W"o
0U') '-C
.. Q . U)U)
l
ku° 3°
1 . I? o oo
C
C
4
q 11-.C~.4
olC
00
C,
~~~~~~~~~~
>-
2
0
0~~~~~~i
Z
,.E.T.-E
U)
U)
0
5U)
..
<
0-
0
.
.
E0
UU
a
2
=).
<
aU)-
m
aU)
U)
>.0
U
-U
o:
0-).2
a E- 0
C
,-o
-
0
E
5
U
U)
c 0
o~ 75
o0
WW
CCE
in
a)
U)0
U)
U)
U
U
0 .0
.0
0~
U)C
22
IU~
U
2 CU
0 02 02 ii
1~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
0
U)- 00
f
,
0
000
2!a
)
~
U)
O..<
<
(<
<)
C,
0
>
0-
0
zCO
0
~
0
0
<(N
~0n~~~
C,
I
t
I>
I
C,,
z
~
~
'bX 2 o 0
.)L>
xw~~~~~~~~~~~~~~~~~C
0
a)
-
>
CO
w0
(NmI-
c
t
C,
Cl)C
()
C
U)
100
I
10
a, 9
CO
1o
o
LL
z z
o
D
0
0
I
(9 r
9
i
----- -I
------0-·---r
I
t-
O0
0
Cc ) o
oo
Lo
o c
L
-
od ,
I®0
g)
0
100 iii10 lili
0
10
IV) 0 ill; lli! 10
~~~~~~~~~~~~~~~1
~.!
:
,
,U
L
~o
~
0
0~~~~~
0
10 10 10
ci
0
t
c0i ~ 109(g
L
o
a)
g
o
-o r~~~~~~~~~~
c:
CL
EE- 0 >0
all)0
Co
a,. o
cl
m
w
o
o
-
C o_
o Co
t
ID
CL
m0
·--
-------
SE
T(V
V
0
<
m
.
m
-0
N
cl
~
~L ~
2o
(N
o
, .srn
cL
o a,,ru
o pm 0Em
~ ~- ~U
~D ~~~<n~0·c
'-
------
g '-
; 20-aJS
F0
0
o
N((
'o'-c
a
O~o
O
0
t,0
w
a-
o0
C
Q'(a,
7E
-0
_uo 0 aQl
w
~o
-------
(V
0)
a'
-0
Z
,p
u
5
--.
'
o
a
po5
w
E
'ac):
oE
:z
a.)
a)
.2~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
<
c)
.o
a-<
Q_
Q.
cr
co
-
-
xQO
Oa
c
B
-C~ '
zo
-
a02 m ZZ
') ' C ')
c o~a
I00q
U
cn
U
C)
0
4
..
0
I
Li,
U.
0
L
0
L
LLww
Ld r
I-
0
CD
Cl)
~~~~~~~~
~ ~ ~~
0
CD
U.
0
o_
E~~~~
S
o 3~CSI"--~
r
o
U -o N
.~U)C.
,.o
0
C0.
=
8
.14 5, -m
tW
-~o/ C/
-F
-n
o
zU)
0
-0
CC
84
0
C
~
0
>
>
1U)
~,~<C~~p~
0
2
q'5
0
0
).
D
t
5
0
0
0
C
U0
'.
)0
.0
-.
(
0
>> ID0'
.(
'U
0
E
0
ID.
0
0
0~
*
C>
.. . 0. . -0c~~~~~~~~~~~~~~~~~~~~~~~~o
C' · c
~-~ ~~~~~~~~~
Cl
,) N 0'" 0
S
0
- -.0 E
....
0 'U- M
~
~ 0 >>~~~mEg: ' o ~~~~~
E
0
.
1
So
o =S.EE
~
'U>o
U) > >_CU00U
>
>2 _>
.~> >.U~ ~ .uo~0.> >o~ >~
0°0>.
> > ~<
0 >.
....
O...
) O ' 0
Z
_0g
a
L
C
m
C
U)
(
2
C
.2
0
)
(
1C,
00-4 o.
°
=
C)
,1 U~5
.)u.. (,
Žo02
U)
C.)0
0
U
U
CD
.~~~~~
V)~
U
C o -~J
o wo
Z
Z
0
7
Z~~C0o
C
y
C
0-0
O
'<
a~. x
U) 0
0.
a-
C/
>
0,
2.
0
0
.iC)>
U
U)Cl)
C)
Cl)
0
-
0
':0
)
U)
C)
C4
W
-;
H
0
0
-a
:S
,42 UUC
~'
0C
=
<
CS
Cl4
V~C)
IU,
U
0
0-.0
- l)
0
0
0
N
N
Cl
O
--N
.0
-4
Cx,
> 0Al
.Q
F
C)~~~~~~~~
:
m
3
.C:M
:_4<
c0 o .:
C o
¢
A
¢
C,
24,N
C
E~~~~~
2
-~
U
M
-
<
o2 269.~
~ -g~~~~
0
o~~~~
0
~
.
~
0~
6O
0
(U
~~~0
~~~08 20~~~~
U M
0a,
C
U
(U
cu
U)
V0
-
0
C
0
_
0.'
CE
U.
aa
0
U
~~~~~~~~~~~~.
. .
05
02-
U
a~
O
~
0
C
.
U)
,U
U>
~~~~~~~
g ..
C
02
CO, 0
i
)
w
0
0.
°U
-.
8~~~
(U
M*50.2.0
o
E)
C)vP.,S
0ol
=
ct
Cl)
~~~~C
0
C
C.
.2.
.s1
-"
~~~ ~
2~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~*
.2
.0
ol~0
C,)
(0
5
010~~~~~0
'Um
:
X
0
C- Nt >-)T
/)
O
F-U(Cl)
>
U)
~~
~l >>- > -0
>~~m0~I~
ID')
IFu-
-
>.
->->
-
>0
- >-
.C)
<.0(0U.
)
IL
CD
o
02
U)Oj
~
0M
~
-
IO
l
Cl
C~Cj
li
O
N
.0
.
>-
"')
00
) 10
ooo
ON o
0
!
(
to
0
0
Cg
0)
o
0 0)
-
C)
o
to
o
0
C)o
0
0-
0
0
0
to
0
0
LO
O
CO
c
0
CO
0
N
to
C
0
.2
-O
O
-O
0
C_
O
D
0
0
to
v
75
0
,
0
Z
n
an
IL
o
>
0
L
C |O
|
0
1
o-
Z
Q, Q
S:
-1 N a
1°
u
|.N
t
0
to
-
C)
,
(c
to
to
O
C2
1E
10
I
1.
1
D 10
Itlo
Il1
(
N-
1~- IU)
·09
0
L
N
10
N
C')
to
=
0
=f
I
C
-
a
C
I
o
I
C0
o
C
0
U,)
C)
I
c o
00
aN10
10
°
I
52)
-'I
")
-)
I
N
-
1
M
?7,:N
I
i
o
L,
o
IY
ni
03
E §
-
I
|
>
0
2
.
cc
|
I
I0
o
o
,.2
a
o
.
.2
r
|B|
,Cu
0
0
o
i,
?
.jj
co
CO
2.
co
8
-P
o_
o
i3
ro
U.
>
-
fl' , VI~~.
~~~~~~~p
.. 2
L
1
1
1
0
CO
LO
(0
It
Q-
1
E
0.0.~r:
co
z
OC7C
-
I 1- = VB IB
I
|.I
,,,
-t
10
(
-
o
I
(
U)
-
CI~~~~~~~C
1
-
E
0
|
0 0
o
to
.
co O
E aZ
U)CDo
-
=E
C
_
oo2to
0)
0
-
CN
O
II0
0):
l
.0
,CIO
00
co
CD
0
EJC)
(-i
E
1
1
1
,I0
:t
U')
00
U4
UN 0
In
(D
I
U.
~ ~
0
U
~
0
U)
UN (0(
~
E
00
(N
ID
U
O
U'
(0
LU
LU
LU
U')
C')
000,0
-,
U'
-
g
2 0.
.0
_
O
E
~
0
U)
.0.0
,U
2
0
0
C
0 .
~~~
~ ~ ~ ~ -~ ~
~2m
0
E
0
.,%
E
~
Ea
01
U)CCCC
) 0T
)-000
C,
00
0
-0
0
U
L
oo
0,
3
Z0
(N
0 U)
N
- 2
U
00
)
U
)N
(
N
(N
0(
U)
(
-
0
'
<U
0
0-
U
-
0
,
-a
U
(
0
.
U
0
C
U
LL
z
*-4
CD
0
~I IC ~
C)C
CD
C,
z
a.
~C'~1
~
'
pO'.~
-'
(N) LU)
- ,
(-
00
(
,
0
0)0)0W
E
F-~
0
LU
LU
It
0
U
a.
0
U
coZo~O
~-
'
0-
0
-
0
(
'
)(
r
~~~~~~
C=N
oI-
~0 'I
'
U'
0U)0
OuNL
z
-
-
C
-j
'
co
U Cc>-o
a)
0
C0.
3 U
U
0)
N -
U
o~~ a.U)U)
ry
E
£~~~~2
2 U),
0
mU
o~~~~ U"~~
0
0.~~~~~-6C
w
E )
-
0
) ))CO
C
U
)00
_
-
)
C0
3
00330,
oU.o
L
E
LU2
~~
~8
C
~~
<
o3 ~~~~~.
C
U'J~~~~~~~~~~~~~g
~
~
~
0-O (0~~~O0~~~0OH
W
CDCJOIOLU
~~~~~~
0
0
aE
0
O
00
.0
0
N
E
C
C
N2
2~~
N
~~~~~~~~~~~~~~~~~~~~02
LU
0
L
0
0.0)a)
0(
E~
E
0
U)
0
00
0
0
Z
0
W
D
)
t
m
z
-0CU)0
1
0
,
=
33.0
U)
L.
,
CC
0
l
U)
C
D
4
~~~
~~~~~~~~
~~~~~
0)
0
lo
_.
2
0
U)
0
U
Z
02.~
0
EIE
8.2~~~~
U)U)~~~~~~~~~
.0~~~~~~
a.
U)
C
o
D
2
0U
2 0
o~~~~
2
W
M
oi
U)
U)
U)
02 22.
-E2C C.) 0.
C)0
0
3
-
-
V
E
-ZE 2
0
~~C
0
.2
2
(C
.2 3
8
2
U,
0)
'U
U)
UN '-(0~~~
~~c
O~~~~~~~~~~~O
0
.
~~~~~~~
2
H
3-
U))0
o2.0
'.
-E
~ ~~0.0C-U)
U
'U
(0
(NUNE
a
o
2UN
0z
C
(0~~~~~~~~~~~~~~~~~a
U
co c
10~~~
co
)
0
(NO
j(
C')
'
(N
0
U
(0
) )0
0
- .0
2
co'
C'
0)0.
0
z
o
:
0
-'
U
3~~~~~~~.
.0*
0>.
o0LUU
U)
0C
.
30
.0
0 <
U)(0
0
(0
(
(0
Z
U)
~
'
(0
0
UN (
-
~ ~~~~00.U)~o ~~~~~~~~~~~~~~~~~Ea;0"
0
C') II
0
-
-
-
U')
- - ------
U')
0
0
0))
N
00
N
1)
U
NC0,
U
'U
)
.
0
0)
U)
IN
CD
U)
N
'
0
C,~
I)
00
>
.
Z
C
Lo
1D I
)
(1)-
0
o)
.4
Cr
cj
cN
I-0
C
0
C
'3
0
z
z z z
U-
C)
O
3LL
w
0
U-
LU
CD
3
S
CU)
lu
o
3
N
-1
-W
-)
c
(U
<
s
0
LI
cl
0
0
0
0
O
N
0
UJ
0 (N
Co
(0
't
0
0
t-
U)t
U,
U
4
V
o io ioIc
C0
· .
0
0
N
)
oi
It
ct
m)
0
.
C5
W;
E 0C0
E
0
w
cn
O0-
C C'iniNi
rA,
Q
tv
4-o
99
z,
c
wJ
n
C
o
LL
a
0
mr
EE
0.
._
C0)0 C
0
CO
U)
o
Q
.D
2
L
0
O
)
D
-D
E
z
0
O
C2
C
O
r
0
U)
-
O0
(N
CO
B.o
0
CN
)
9r
00
Z
z
z
>
LL
DQ
(D I)
oo
_
0
-------
2
z z
C/)
C/
C
)
.2
(m
F-
c
C
o
lr-
~
1
~0C
-J
Z
D-
U.
C
I-
-I,
C
Z
-m
0
00oo
(-CU
C0
)
(IN
"C
C
C
C
)
C
C
L
c
C
C)
N
C
C
-.
i
C.
ILUL
L
Q4To
U
Up
CO4W
CCC
C
l
N
(
u
(
C
(
CO
(
C
C
N
CO
cLU 2
(
CO
-.
CCC'0o
C
N
(0
.O '
. (Ni
(N
.
(No(
.
.
H.,
a
¢Ub.
Ct
*
2)4-
m~r-
2e
)
.0
E
a
Cu
a
ID
CC
U
o
0
;:
V.2_9 =
z
co
E
'o
c9.2
'a
'
C
C
-a -a
E~~~~~~o I ff
*~C '~ CC
"I -a
cl
~ 75 _
C
o
2
C
LL2 :t'o
E
<<
<<
4E) 2 o
-
O
Lc
(N
-<
<
)
CCCC
0
0
N
CO
C
CC
~o~o~~ ~ ~
a)
-
~~
N
0
a)I
20
0
ID
a.
<
o
<0zC)
)
0
a
a
~
o
U~~~
=..
05
.
2
-
U
Ucc
C
a
a
a
Z
<
N)
c
?T
2C)
C)
_
'<
C
CC
2
I
)
f
-
y<
C
2
o
L
UC
(l
Cu
2
CC
Q
a
<
(f
N
-
ar
w
o
~~ ~
N
(.
a
~ o
-
C
o
22
ID I
z
2
z
Co
w <
oLc o
C
C
C
0
>~~~~~~~~~~
C
C
C
Cl
0
CD
L
C
i
CDc,
C, ,,
C~~~~~~~~~~
,
c,
c,
i iC
L.
,,
q~
q~
co
o
C.) ~~~~
i
(N~ (NoL
ci
c
)-
(
C
C
O
N
N
L
)
z-4
.o
=
~0
.C2(
2
.0
Cu
a,
Cu
a
E
Eo~~~~~~~~~!
.2
a
0 Cu CuN)
aao o V a CL
C2CV
Cu
>
Cu
Cu
Cu
Cu
Cu
Cu
C
C<
U C)
-m
~0
-
,
C)
o
E
C
C
.>,
0
u
2
.0~~~c0
cu
Cu
C.C.00C)
Cr'
2.
>,=
0
<
Cu
C
CO
-:
(N
0
C
u
0-
0
Co
0
CC
275
05
Cu 000 o
o
C
-
ECC
Cu
00
=
o
C
C
2
(D Cu
..
0
CO
L
a
<
<
-
.C
0.0
O
LU~~C
co
-
(
CO
~~~~~~~~~~~
o ~~~~~~~~~
(0CCC
C)
ou
C
*E
c>.
EL( -:
cL
0
0
u
Cu
'
40>--
Q
2
00002~aaoo~.
He
to
U
~q ID~~~~~~~~~
2l
O
(N
u-<
C
C
Q
z
ZI
L CO
w
y
L
C,
: - (N
__
_
L
D
00
>
0
LLL
LUL O
0
C
O,
C
O0
O
0
0
LL
-- h
C~ C
CO
0
-(0(0
C)
O
, C - C_
n
(
_
N
U,
LZ
z
zZg
U,
o
5
,U
QI
_
(9O
(
Q
0
0
C
._
=
z
4U,
co:4
ELLCJ
oo
0)
oC
o
0
0
O
O
(NC -
O
- C
L
'icJ O
N C-
O
Co
laU,
-O
~-O
iU)~0
ut
0
C
N
c)
(3)
OD
o9O
0)
9
C:
ci
1=
U,
L.
0.
,
uu
tC,
0
2
O0 E
E
o o o E
=o
)0 .
E 00 E
-.
.
a
0o
.2
. .
.2
0
0
U]
a)
a
0
0
0)
aa
r:
+.
aa
o o
. .
>0
E2
,0
a)
e
E
o -E0 c
a
>
C
0
C
C
0
a
C
c
(U
_
4n
Wo
.0
(A
C43
UIr
z
W
CQ
.0
M,
i*l
-
o
._ 0
U
_ ____.- -
8T
I0
rA
E
o2
Xo
o
E
In
W
CO
0
(\j
0
00
w
r-
00)IcN10
M
N
W
0)
10.
(
0
0
0
1
ll~
Cq
1~1~7
10 11
I
N
N
01
(010
-
LU
0
1
~
(
N
II
U!
N10
2
I
010
0
"I
`N1"I
.
N
C'
N
C!
IN
(C
NN 0
lq
0
*
IN
(
0
(0
I1
)10
II
:
~-~~~~~~~~~~~~~~:
Eo o
0
IN
I) I
0IIN
0
0
1
IN
0
In
~~~
00
00
o
-
C.
d,.
.0
-
0
0
.0
C)
0
'~
10
9
'V
'~
Ce~
N t"
IL
t.4-
-C~
>->-
n
~ ef
10
1010
C,, i)
V
4-
V ' -coI
I
IC0);
CQ
C,
O
0
I
In
10
In
IN
m -C N
IN
§-~~~~~~~~~~~~~
,~ =03!
~ o~o
n
In
I
.
n
I
C,~
In
In
V
N-
V
I
0
00
IN
IN
10
N-
N
IN
.2 ~.2 ..
In
In
0
n0 ~
O~~~~~n
n
I
O
I
In
5
)
.
-
.I
0.D
InCE
~
5In
..
n.
.)0
C0
V
0
N
- N
V
co
n
~
n~
)
)
co
~~0In~
-
0
0
.0
o
~
)
a,~'I
0
0
~~~-'~
"~~~~~~~~~~~~~~~~~~~~~~~~~~~~
0
10
N 00D1
N~
0 0
O~~°-°C
J
J-
0
Z0=~CD-iw00I0~0
N
C,
LI'
~ ;i=
-C>
-
,
o~
~.
0
C
~
C,
) -
)
0
a
0o
C
V
0~0'N
0
a)
In~~~~ 0
C)
I
V
N
10
IN
!2
-
o
;;
<4l
'0
'
.I
-In
0
'>
0
00
w
>>
:
>0)10
>~
CO ~
10
00
1 0~3lf0~~~
-00
0
0
01 I01
10N
0
-~IO00.JI~000-N
E,
I
00
0
10
°n
-
5
C) C)
I n~~~~~~~~~~~~~n~~
n . 0
3
M55
0
I
InCCCCE)
EEO O~~~~~~~~~~~~~~~~ D~~~~n
o~~~~~~
t- o
'~ .>-
U.~
0
10-T
0
00
'I N
5
.
I
I
<
10ID
10
-
.
m~~~~~~~~~~~~~~~~~~
00
IN
0 IN
0)10(0
IN IN
§)
10
00
IN
NN
In
I
'
0
>
.
>' 0) >~~~~~
In
((n
0
IN
1
0
1
In
Q50~~~~~~~~~a
o..
.
-
2
"T
>,>,'
~~~~~~~ E2
>~~~
0
2S2
t
IN
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~
5
SE E
0)
0
1 0)N
0
INT 'T
1
-
0 0
22
Ž
0
n)
IN
1
N
1011010IN
In
In
o~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~.
>-Ž
11)1a)
0
-
>
~
~'
~'
In -
In
000
N(N
IN"I(N
-(
() ))
'In
101 M
N
IN 100
q
IN IN 1
10
nl
:9
I
,3
.0
In
10
I N
NI
1
00
0)~
N-0
I N 11
N10
~
ON.
1 0
"10
I N
,
0)
1
10
IN
0
0)
"I
10 I
1D0
10
"( IN
N
0000
N- 0
Io
IN IN~ 100
C>!
1
I N
.
~
n1
(N
V
IN
1
10
IN
In
I
1101
N
0,
CO
0Z
(5
- '-0
.
~
~
~~~~~~~~~~~~.-4
co
4-4
P ~
,C)
o o o~~~~~ a)0)
(b~~~~~~~~~~~~~~~
0
-E
Z
0
0
-2~0
NNan
'
I-
a .~ X B(
o
O !- 4 b
V
000
O
10100)
In
00
-
=
'-
5
SEI
C)
0
2
IEnE
C)
0o
Ž
2
7:5
0
0
0
I-c
0
00
0
a
I
0)
5
n
lo)
V
m <
5
5
5
)
In
-4
X
QoeU0o
U**
5.
0E
101010:
w.0
In
~~~~~~~~~~'
°
< <
UO~ .,-
r
ES
E~~~~~~~~~~v~
0 C C)0 C) l
0 2'
.2
2
o
--
Ma:Sa°Qm
V
4.4%
D0 E
0
0
.
.9
50
^
'5
4-
.0
0
0E
E
°
C .
I
~
0
U0 0
~
0KJ
C)
C)
0U
C
V
.
I~-
>
>- 0
>i~~~
¢'0,
93
.J0
V
)
INZ10C
o~i
>
I
>
Lto
0>-
~
j0>'
>-
>--
zz
o
-
XC
0>-
Ox
oWLL LLJ
LU
fn 0ON
0 0
b
O
!
sNNN
in
1.L
I
C"
LL
CIO0)
0
1NX1
IE
,
I
-
I
o
I
0N
int
-
'
I
I
Y 'I~
B
o
m 0
in
On
,
0O
L
I
I
in
1
0
0
0
'0
0
-0
0
,'
0o
Co| 0 0
2
E
0
'o
V0
V
!
t
U
N5
in
<0
S
I
B IB
IC
B
L
N
0
00
NO
N
0
0
- N0000
~C.C
CnN 0 0CIJ N
LLI
-N
CD0(0
in
-
N
U,
m
N
NC
N
N
-
C
(0Z
i
'C
in
0
0
'-
in
O
0-?"
00~
C)
0
N
0
N
C,,
In
V,
I
V,
-
0
V
I
I
0
C
'0
W
N In
o
U
U
0
c.
C,
0
0
C)
Sin
0
LU
0
0
In
n
(D
in
00
I
~ ~ 0 ~'2'~ ~ ~ ~ ~~~~
*
_
_
0)
0)
~
In
I n
C
C
2
…===oI.I52
co m2
Z
n
o
~
n
0
N
0-i
1I,
2
o
0
0
n
0
I
-~Ci
N
n
.,
C,
10
'0
UL
In
In
L
)1
jW
0m 0
Z
Q
N
i
V
1
0cso09o
0
o
lo I I I
i0
0
0
0
In
In
I
C)W
.
I
0
F
i
LUU
oo o
II
0
C)
NO
IC
0
C0
0
NO
"
0-N
N
C
N
N
NO0
C
0
0
0
c
NNNO
In
C'!N
of
W
<
M
V
O
In
"I
0
0
0
0
0:
C~~ C
C
0
i. C
L~~~~~~~~~~~~~
CC
C
0
I
C
.i
Zn
5CCI
C)0
n
'
I
CCC',0
C,>
~
L
ni C
N6
a)
~
a.
,
NOWC
-- i
00
n
C,
-
0
>0 >
N-IO N
I
i
0
-
(D'
anIin
00
ON-N-
10
0)
N
Z'
I
0
0
2
-L'
-~~~
W
~
n~C
a
C1CCC
In
CZ
~
0C
-
'0
n
O
-
0
N
0
In
C,
00)
W U
In
N
0
---
CC
n'
C,
0.0
j
0
~~~~~~~~~~~~~~~~~~~~~~~~~~~~0
on
2
'Z.
Z
0
N
-
C
|
.
C, 0 I
0
C,
~
C
|
Ir
INi
U
<
N
n
C,
0)00)0
C
?00,
0,0
.2 .2 .2 .
00
NOO >
N
00
i C')
N
0
0
0
0 I
C,
"
5C
o
I
.- .|W
c~
|'l L |LL:|!2 |o
cxi
F- °
0
N
In
N
NI
o
I
C
".0
~
Indoor
i, .
I
InM In
.'
.E
00
NO
m NO C>N
.
I
In
|o
N C NI
NOOa.
>
0
C,00c" j
-N
In
I
E2E2E
5In
o
C
CC
oi
0
C1 1~ 1 1 I II I I
C-iC
CW. Nk
2 2 .2
ZO
ZZ
C
V
LU
"
LU
(1
1
-~~~
C
0
'20
C, C,
in
C
V
C
N
E
~
V
C
I
N0
IC)
10
N (iN
.2
CO
U
0
NC!
N
N
in
00)0
in000)0
iC
01
ON
-
005.22.
5in 5in
C C C C
In
N-
I
C"9 |N
Nz
C
Io I0o I0o IoC I0o
C
iN
|N
T-
0
C C 0
C, C,C,
C
.
No
EN
In In
In
NT
U. C,
I
W|IO
UL!L
i
Z0)0
NO
0
N
S0a,
a,
C0
C,C,
. C
W.
0
0'o0,I(
Z,N-
- 00
00 C > -I>
N
T
4) ID
O
N
!
N
!
O
100
LLI I
I
0C)0
I
0
0
LJ-
LLLU
C,
I
LU
0
1 1 IC
I
C
t ,jN\Nt
0h
0
o~~~~~~~~~~~~~~~~
U
0
- c> e.0 ,.d
lJ 0..0
Ciu
I,
C
C
L!L
n'
1
0
I
N
,0
N
11 1 I
B |B
2
2
, C
00
C ,
41o
C,
11
1
0
.L8
i
0i
0
uO6 c , 'c0. o0
1
I
, C
-
L
10
C)N NC-IN0
i
c-
0
-l
Ni.
0
0LJ
0
is NSNN
tn0.
-
0
o
B
'
Ni
N
I
-I .U.
0)
0
NCI
In In
g,
0
IxC°l
oO c,,
0i
I
LLI
NO
1 1, 1 1
=,
0
i.
n
N
>
I t I lo i nooo Io
I
ZB
i
O
-
U.
1'~~~~~~C
' '0
Z
I
I
0
|
ID
.00)0
I I
Z oO
Z 2_
0l
I
<:
00
i09 0NI; CNO
N 01'
N II) 0N
i
o, I
=0
0~
0
o
ONII CI N
0~~~~~~~~~~~~~~~~~~
in
in01
0
-NO 0
1 1:
I
n
N
LLL
U
O
0 0
N 0-
0
0
... ....
C-
Ci) I.
0)
Z Z
I~ x
C0
0
0C"
N
a
II
I
I
I0-
~ ~~~
~2:
~Z ~Z0
2B
(0
0
0
0
I In
C)))0 In
F-:O' ¼
o4Ci UJ !o : c~i
Ii ,, ,q: LO
C, Ct
C
I0
I .
I
·
Ir
LO
i
N
1 1 1 1t 'IT
In
I~~~~
O-N
CI
I
L ()~oz
.
O00
-
N
C
O
0in00~~~-
Of
0
C0(
.
Z) I
>>0
0I
00in
Z
0
I
)~
>H I<-
i0
0 0
(0
(N
0
Co
N'
C
(0)
U)
0
0000
0
"I(0C
0
Co
0)
. Co o(0
0(0
o
L
0
r-
I.
C
co
00
(
' cN
!coc(N
0IN (0
(N
Co Co00(N
00 (0Co0 0 0'
(
N
j
0.
0
00
"
N
(-N
(N
LO
0 ( '0
:
(
NI
u
0(0 It'C0I
'0
0 N (0N (0
0
N
C N o'
000
00
'.
(N
'
7
-'0 0o
ID0)(NJ0
NC't
C
(o
o
(
co
0
0)00
0
N
) (
0
o
0)
0
(N,,
O
(0
l
0
I.(
Co
NiCoO
(N
N
(00
.:
0
LLi
.
.
00 .
0
0
0)
(C (C
0) )
0
E
... .
E
2~~~20
(C
'
00000
.~
C').0.0
I'll
o5
005
0
0
5~
5
(C
-~
-~
C,
002
>,.
o:
3- -=
M
0
0
m
00-r5
Z0
. ., 0
0
a
0~~~~~~~~~~~~~~~~~~~~~~~~~~~,O
) Co
0C)
(
5
C,)
(0
2
~E
(C
(C
222
0
.-- E
0
0
.
.
2 020
0
(C0
o
N
0
(C. (C
0
C
(C
.0
0
-( :Z t
c,-
c
0. 0
o .
(C
.
~ ~~~In
(0 N
Ni'~ "ID
'~
LI
.1
0
(0 o~
(N>0I
(N
Co
(0C
In
Co
=
a,(00I
(N 0(00
,-(00)0)0
IN
0 C,,
(NI00- CoI
Co, I.,
-
(N
C
.3
CNO (
.
N
..
N
00)0
N
(01!
"IN
0
'-
CR -
0)0
0
0
N
~5 .
'..'N,..N
N
U(O(
'-
0
3
00)
I0
00(N00000NO00
aC
_0
N
0.0
o
C
C
..... 4
o
(C
(
E0
75
(
C
N.--
0
..
(Ca2.
(C
0
22'o
)o
0
_
0
C
00-
.
.2.2 C.
CC~~
- (C
(
~ ~ ~~ ~~~~~~~t
(/t~
0
m
o
o.
(C
..3
0 (C
.2 (C
(C (C
.,
>, i>
.2
o
S?
I
0
N0
.(
E
.9
o
(C.0
0.
0C
OOO~g~~Ca)
~.
~. ~C>,
~. ~
.3.
.3.3.300)0
>,
o
(
0.
z
co.c
o ?
~~~~~~~~~
(C 00
0 ~
0
00
0 ~
~
~ ~- E ~ 7
~ ~5..
(C
.3~ (C ~
(C
(
0.0.5
LL
S
0
Co. N
co
N
C
.000)0
0
Co
o (0'0)(000 .C0 ( N ((o N (
o~(
( (NN Co (N (
~~~~~~~~0
.. .
.
0.
P.
(N
(C
ID
I
0
Co
(N
ImE
(C (C
(C _(C
.20.
EI.I
0.
(C 0
0U
.2
t
3(C.
0. 3
0
0
(C
00~~~~~~~~~~~~~~~r
0
0
0. CC
(
(
.
(C
0000
0 C-:,
0 0 0
0
-
I
'0(0
000
>
_
.
0CCO
o
Co
0>
.3
C(C
E0.0.0.
0
.C
.
0 '.-N
' (N N (
(C-
0
C
c
c
20
(
(
0
~~~~
Z
00(CZ.a0
Q.U
u .~~~w
0
(CCO
0'N'
(
CC C
0.3C
000
0
0(N C:
N0
CO
C~~~~~~oW<WOCCJ0.O0.CDDCD0)CD.
NZ
(N
Co
>_>0
0
~
U-22
C>~~~~~~~~~~~~~~~~~~~~~0.0
C
0
5
00 .,U 00
04 LOWU.ICOCOW C)
(N
o (Nr(,.4 C
0>,
-
t20
0.
'
S
(
C
.
(CE
(O
(o
.
C*0
.3.3.3.3.3.0.0
0 0 0
C- 0
.2(C .3
.3
0
c(C
I::IT
C C E~~(
>
0.0..
0
a
0>.
(0
o(C m N-0
w,
N(
(CJ0
I,. >I>(N -0
>
Co .-_
>
(0N
>
o-o
o aflfl
0..0)
U30(C(
`..
-
LCC<0CD.3
(00 (10cO
0
> (N LU>
s(0
ICLI_
o
C
0> (No>
I_
I
1
I
I
-"·LiDllllli··CllllIlsl*ll·CI---------I
0
0
N
Et
1
I
O
i(.~
O
0)
N
St
O0
00
90U
0
02
(N(NC@O(NC
s
c
c
c4
c
c) c
C)
C-CO
|O 0
CO( o,
B~ tq
I
-
N
I0. I 0
I.J(N
(0
j
E
j
00
-
j
ID
N
(N
C
(0
COO
I0
COO
CO CO)
0
m
0
0
0
0
Ns
S:It
1(01
00
OLLIL
.44
o
.
0o
E
t
o-
~
CC)N
E
'a 0 >
s
(~(N
N
.
0
C)
C
CC
> 0.
0.0.0
j)j
II
CC COO
CO
IL
j
.~
~~~c
0
04
.4
o
~ C->
2 LB
(oooo
0z
0 0 .
0,0.
-
CO
C
Ci
>
0..
0a
(N
ID,
0
0.
I--
2
c'
(NO
o-o
c
(
(C
-. o;
.
. o
c:
c'
'
'
C
I-N
~
- o -
o,
-
C-
0)
clc c ~ o
IX
XOC
CO
c1
c
c1
I
--
S-
-
'I
"I
N
N
t
.2
)
N
i
c
l..J
C
(
|W
B0(
0.
0 CO
(N~
0
0
5CO
L
0
n
(C(D0
0-
O
0
O
1
N
C
C
0
2
2
2
2
2
2
'1
II a
cIc CIc1
.
a)
Q(
Ic
..
QC
ID Q
(N-O
(
(N
c
.
N
C(N
B)
!
(N
.
.
,
o 0..0
"0(
0)
>
I
(
C-C
SS
Q
oS
G
n
(Dl
N
(N
(CN
C-
(
C
'
<
]
o(_ )
ww0' 0
NN 1
N
0
O.
nN
0
CO
-(
0
2
2
.
N
n
o
O
<
00 1
C
C
(0
0
(0(0(00• 0.
1cl
0
(0(0.J~C
(0 *1
$
oL
-cL
l
ac
1
(0
0
00
.
0)
(N|
(D0
0ZCq
C C
.
.
.
>- >
>> > >
9.
<<<<(2<<(2.
0
.
N
0
.
2.
0
(N
(0(
CON(c
.
I0
C-C))gg
CO C
CO
>
.
2~ CO(0
o ~C-----~
~
.
2(0(
(((
~
2C
,- a.
2,,
OL
ooc~ao-D
I
.
Q
) I a)
-
2
o
K
Ic
loa:
.
~.
.
c
-o - 1oo 1 o 1w
10 10
<o(101
C
(NO
I(
1'1oNL
00
0.0
(02NO 1
oNIX
(
.
a
0
32: 1
0)
~
I
I o Lolo 1t
1 3EoN|Xol
-'
w m- Clm oIno
OO
IcC
C)
C
C
CC ao5.
.C. -:(C
1o
(C
(C
,w.
0
0
0
.
t
N N I
CO1) CD N"1
OI
c
0)0
1<
1 ( 1o <0
00.>0 >- i>>- >>0.0.0.0.0.0.0.0........
>- > >>-
0 c
c6
m,~ mooEOI
-
o o1J
(
m
<
0
.
Oxm
|
c~c~
-a-
,; .O
2.
(
(N 0)
.
.((C
a, .a) .a, .a) .aQw
.0.0.0.0.0.0.0.0.0.0......
- 1 (N 1
mo
iS
CO !,L
IN
CN
O
cC
COCO
00
0.
C0
(N
0
0
CIX
C0 0
O
O
6maam166
0)
2
CC-
IN
c)
~~~~~
0 00.a.
0.'
CBB
OO
CO
9
co
0
0
L5, 0.
.0.0.0
cs
0
0 1a.E
0
_
.5(00
(NN cc
(No
- o,0
I <<<<
J
C- 000
(N
-
c
N
C
N
I
cIN CO
COO
0
E
7 II oo-l ol 1o "3m -'t It
~ ~~ ~ ~~~L'
.9,
0~
o ~ ~(C
a
_
**
0af
LI.
0
=u
-1
B
(N
0D
0
. 0..
co
0~,
E
WN
0.
C
0.l
NW
I
E
N
N
.
- IjCO|0I
c
C
00)00
(N
mm
s G
,
~~~~~~~~~~~lo
|~
6
0
00
-I -<<coco<
-
(N
O
N
c'
6om
N
0
0
Co0
NO
cCO( c' cl
)
m
) I
N CC
N
00)
.
.o.2
(C
) . C.QC
QC ID .Q. GI .Q .QQ
.0.0.0.0..
c~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~c
~~)C~JCC
~ ~ -1 ~ COXt ~~-0- ~ 1)I ~XI
CO - N (0 0
I I }CO
W oL -.
j 1 X
,
0C
N
(
-t
.jL.
CO
N
00
N
N0
N
c
r-.0
- IC, C ICN<I w<<(0
I
.0
0 0
-a
0
0OCONN((CQ
NE No N ,E
0
IN
N
ccc
@~~~~~~~~~~~~~~~~~I
IL
N
2
) a2
.0.0.0.0..
C
a,
(N
0
c: c IR
oI. o (
2
o
C) C)CL)
0 .0.0.0..
(CQ ).
0
'I
N C
00
cLc
Q
0
CO N N
N?
0)
C-
j-j
C-
,-
-D.
- -(0(
c
00
2(
i
o
x
Ni)
r4
~
C'.-
Ij
, ,o0 oI
.
q
0 00(N
N0
(N
ID
0
d
6
,6
co
00000
(
In
,
0
0(0
N
,l
.q-
ol
.0
-
(N
-
IN
I~ -10
o'>
u'>
r
u->
.q,q -
(N
(
m c (,N N
N
,-)
0. 10
Ncj
(00 m
mN t~
Ni N('C4
(0 C'
IDI
ID I)
W
(0N
Ci
'
(0q
0~
i jC' NC'
0N
cai.
I00000
j
N-
!
0
(0
c- c4
cq
i 0 j11mN
(0
(N
o
o
In
0
(o
q
IN)C
-
N
0. N
C'
(0
(N
C'(N
-
)
0(0-C
I'(N
r
I-
I0
C
c
i)
C'i
C
I'
N
j
(N(NO (N
i-o
0o 0 0
aC,
ID I a) I
C,
a
I
v.
CL
FD
-
a
rz
a
ED
N
a
Ja
I
<
<
N-
C 00.-CD
l 1
cr0
~
'O
~ ~C'>(
~[N~4CN>N ~<
>0
~
I
0
0
(
-
-
--
0
EE>
»
n
=
C/[
0( 00
>C'-)
0
D
0.
uJ
a
0.0>,0
0
W
Z:W
4
W>
0
a
W
U
0
00
/
a
I---)4 0
M
0
'=O 0
-
(
E0
CE
IL
----
0
i(
-
(Nj
(0
Iu
Nj
0~
o.
(C-i.
0
0
0aa
n
C
0
4
t
-.- -->
I7CC0
~
04(0CC
r0
'
(00
,
,
E
00
,,
(N
d)I)(N
4).
N
(0n
(N
'>'>
a
0
"
(0(
a
a,
,
a
_
0
O
(N0,4)
m
(N.-
>00
0.a
M
~
E
EE
')
(m
0
Q)
N
<
<(0(0(0(0(0
<(0
(0
(0
r
7
· ~
0
>.0>0
v
4)4)4I4)4)4)4
C'>
I0
ci
M
0C
N
(0
(0
D
-
N
(
(
cn
co
0
0
00
0
0
0
c
O
O
0
000
-
I-
I
0
0W
0
C
N
0 0
o
CO~ CO~1`
(N
CO
Co
b~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~,
~~,
,-
-~~
CO
COO
N
COO
0
- 00(N~~~~ll
(N
CO
N
CO CO"I(NO
) 0
)
O
.
N
-0,
N
N
cl,
CO
0
C-
(
(
N
0O(NCO
W~CN0
C
CO N-
0
IN
( N
CO
C
0"* (N
CO
N
0 1
CO, N
C
(
1O
(N
(N
N-(0
N
)
-
CO
CO
C)
N
IN
C
C
(N
EuCO
LL
(N
~~~
~ ~ ~
C
00~
CE
oo
EC
0o
o
o
a
U..
C
E~~~~~~~
C
a)
C)I
~
.a
-
aC
o j
c- >-
c>
(D
~r
~
CO'
CO
CO
C
m
(N
a,
E)
0
E
0
o
o EEE~~~~~~~~~~~~~
a
a
C
C2
C)
C)
o
t_
o
C
.j
----
--------
--------
E
o
0C:O
o2
~
-
O-O-
CC
IN
.-
E0
o
00
f~~~~~~~~~~
-
2
C
CO C
I
-
C
C
ZZ,
CC
U
C
(N
z
, I
-J
Z
100
.
>
(
>-
Ka
0
0
CI
0m
C o22222222o
0
e&
g
u~~~~~~~
)
c,
0
N
,- .~o
E
a
C C,)
2C C
m
0
,oc
o
o
I
I----
(N o
Wof_
-
22
>-oo
0
N-~ CO~ (N
-
-- Ea,-
C)
C)
m maaczaaS~~~~~~~~~~~
cl)~~~~
c~r~rF-Ycr
C
N
'-
> -
(N
oo
E
a0.a.C)C
.2 022"0*
C
,-
022
2
E
ECE
~
w
I-
O
0
ct
-
~0~~
-
I,-
)
N
t
oac~.2
CO
C
>
o
-}
CC
x
C)
C
xcl
Z5a
t
=
) C
j
r
00 s
0
of
m
Ca
I
-
a> - 0
co
CO CO
0
'IT
CO I
'IT
"I0
0
.9
U-
IN
~u.~
N
m
0)
N
DON
m
-- m
((
-)w
COr
040Q.
N
()
0
Co
(00
"
OCOC)
(0
a
o
N
N
0)0
0C00
C0
0
0
0
0
N"IN
N
CO
0
0
N0
I
04
O
0
00o
w
0
-040N
CO
-1
Iq
N
U)0m
1N
0))
(w
N
(004
-N
o
a.
0
1~
-- 0
(
IN
I
cl
0*
)
-
CO 0
0
2
E
c5
0>
(
2 (
oo
E
0
2(
U,~~~~~~~~~~~~~~~~~~~~~~~~C
COO
(0
4
0(
0
0(
(
'!
<
5 (U (U
0
.2 2 .
15 Z
(2
(2
>
-0
C)
0
(
0
(0
2
... ...0
222
0
.
0
(
>¢
0.0.0.0.
-~~(0
0
0
0
0
0
m
a
<a.
w
CO
~
'
a.
IC
a
a
0
0
0
(0
~~
00
U-
-
0
(0
T
N
0(0040040-000400
0
(0
(2
(2
(0
(2
-T
N
SI
.
N
2(.0
.
.
(00
(0 (0
(2 (2
(
0
E
00
4)
U
0Ui 4
F-
>-C
)-
) U
xN
0
4 a4
)
(c 0
-0~~~~~ll~ '
.20~
(2~~
>
a,
N
'~~~~o
Nr-
N>0)
EE
>
-00
0
>c
0I
>
m
'~
cl
0
>
U
.
>
o)
0c
2
E
z
>
<0CC
4
Z
'
Z
).
N0
looz
0
-
)z
4
04
0aC
)
0
CO
>
U
U)(
<
004'
-
N0(0
40
.
c
z
0
CO
-
0
(
E$
0
C)
.0
(0
0
0
co
.'
o~~~~~~
C I
(2
.41 (2
C
a
(D
*0.
~~~
~~~~
o-0. (2
~~~~
0
5) ~~
~~~
>
(2
0
04
U2(
-
z
.
0
2U
C4
cL
4
n
00)N
U
.
IC)
40(0
0
)C4
c'0
.
2
(2
~~'
c)~~~~..oW o o" . .
2
~ ~ ~~
2
~~ .5'
0
1
-
'>
CO0
>N C04
>
Cl
0
04
; >-j
N
-
o ro
(00)
0004
n
C
)
CO(0COCONC
0
0
0
(W
o
(
(
m
0
400
0(0
<
<
Zz
-
<
Z
Z
CO
<
0
z
04
0
a
<
w
N
a0-
0
N
04L
1
LUJ0
~
ujWO
0
L
(N
(0
(0
LU (N
0)
(NO
(N
LU
t*'N' L L LU
) N.(N LU
it
0
V
Lu
OWLI
IN
(N1 (l0
(0
LU N.
LU
LU
I
LU
C-U
U
(
(0~~~~~~~~~~'
LU
LU
(N
LU
(C
co't (0•,~~~~~~ID
U 'it
i)
L
it'
(N
C
0
0
'
E
a
Li
cz
z
co(0
>.
<
LUl C(N N
.
L
(
(N
O
N.
N
N
(UN N
'
C0
4
i
2
O
Z
<
LZO<
0
i
0
IU
LU
: )JW 'I' oL
NC
LU
C')
L
2
02
U
U
iL
U
U
N
L
WOOI
oL
LU
(N
LU
LU
C
(CN
C
C
0
i ooi
0
a, a-
0t
.:
0
0
CN (Ni
0
N
LU
(N
N
i0
.
C
.
(C,
i'
70
C
it'
LU
it C
.0.
itoit
io .0.
C
0.
C
C
.
.
.
.
.
a0~.
<
H
0
Qw0..00000.1.
.
U6
LU
L~
0
.2~I i0
*C
.
'..0
0
'
C
1
C), C)
Lo
0
0
00
LU
(N
'i
(0
it
N
(
N
(N
U
(N
i
i
N
L
.
0i
2(0.
2
LU L
(
LUL
4_
Z
~ itoit
.
.
itoit
0. 0
it
0
.
.
.
U
L
UL .000 UL
CL
O
C.0 .00
.
.
.
it
LUL2UL2U22F2
.<<<
(2
~~ ~ ~~~~~~~~~~~~~~~~~~~~~~
0 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~ ~ ~ ~ ~ ~ ~ ~ ~~~~~C
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~
0~~~~~~~~~~~~~~~~~~~~~0C)
,U LU, LU
AI!A!!SUeS
(
(N
'IT
L
i
~,~~~~~~~~~~~~~~~~~~~~c
it
0Ž1
- itoito
N~~i
LU
-> ~-
t
i
L
.2aC
i
i
0
I-
aM
L
L
z
E C ~~
v
<-<
CZ
z~ 0 z
<LCCI(CL0>
LU
IDi
it
'a
0o
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~0
- i
t 1- I
LU
0
ID
la
0
LU
UL I
~~~~ ~ ~ ~ ~ ~ ~~o
it
CD
'a
-
r
i
N.L
LU
i
CLU
00
'q
',b
.U
C0
0
C"
LU CD (N
(N
(N
C)
C)
(0
'-
L
U)
LU
CO
C
LU
N
I
LU
'
LU
LUJ lU-
't
(
LU
©
'it
3114uW
:
o
I(0
'it
-
(
L
(N
(NO
(N
, (0
(N
LU
(0
'C
'it
(0
t
0(
LU
N
~
19~~~~~~(
Iq 9
LC)
. '
0l
C)
-- LU
( (N
0m 'i(
i it
it
I
-
z
E
t
it~~~c
0._
.U( C
U
LU
'
L
=2
i
<
5
itm
it
~2
0
C C 0 0
'~
231020.
-0
2020c
2
it
L'C
Cl-
f
v
it
(CC C
(0
co. co
2
0C_
" (N
C"o
ID
._'(N(N
o1
4-
"
0
. .2
C
(
0
4
~..x~C
E
=
o
'
(N
. ,', 0
~1
qI
4
4'~
'
~
i0
E
2
.2
.±
'0
22
it
'-
0
z
it
x
i.
Q
it1
it
it
z
z
i0
i
t
itr
it
o
'
i
'z
C
i'2
0
oSL
C
Li
C
-2
0
it
,
>>CC
U)
4
20
it
0
O
LU-
i it
>0
<
0
It
"'
<<z<
>
~ ~
o
(<00
2
~
~
"' 0
w
~
'
Qoj
~
a.ZZx<
-
-
(,. 9
z
(
-
't
zN
cO oo
0
0
~ > --~ ~> >1~
Cc0.0
c-H>'
0
'-'
(2
(
LU 0
0>-
.
U]
Jo
00
cD0 tJJo
0
0
0LJb
0
CL
C:O LJ
U
(0(0
(N
C'-
CL
(N
(0(I
(I
C
( (
D0
CO
(
-
Co
.
(NC
C
(.i
CC
C, C,..,
('
0(0(
(0
i-,ID(
! L
CO 0-C
,
N CL
C
C'-
"I
0I
0
LJ
N
C
ii
1
i
0
0N
(0-0(0(0(
(
i ci
0 1U
1L
Li
00
C0
o
, CoL
N-t-
CO (N (N
C'( (0
C
C-.0
N
O
i
IN
C-
,
0
,
bN
I0,
C-
N
N
(0(I
CN
C- (N
(NL CL
0
o
Ec
0
(0
>• - .00 >
.5
*0
ID CI 0 ID
0
-
020
>
- -5
75
<<
0
o)000o
0
(0
0.
(N
(0
(0
0
0-Z
0
CO
(0
(
'O
1~~~~
ll~~
ol
m 11
I 91~~
. 11 1-1 ll~~
I
ID
0
I.
m
i
J
L-
0
N
~
co
llz~~~
o
~
0
-
(
o
"
I
)
0
(0
(N C
OD
(0
~0.
I I
C
(N
(0Z
m
(0(0
m
O(
0
00
-
T
oi
(0
O
N
2
C I
(0
2
<
-
-IOC
Wi
3:
>
w
0
'
-I
0
LiiCCii
;0
CLJ
.1
O
<
Wu
a)~.<
. CD
0
-
IL
Li
LL N
D
111
kj, U
LJ.I
N...
CZ
-
PCli
I
z
.j0
0
--
l
ii
N O
C)~
L'
0
iii0
0
(
(DO>
O z
1C L(t
C-i
C
<(0
R
DI<<
co-I- LC 0.a11
L
0 C
CON
(
C
(NO~~OO 0-0
(N
(i.0
.
'
C
CO
C)a-
-C
<~
oz
0
0
(Ni
0
(
CCCD(
(L)0
C- (0
- ( (0(0
(
2
5<0
.0
0
C0
w
m
00
~ ~ ~ ~ ~ ~ ~~~M
(
C
-(0(0CL'0
LCd
.0
0 oo(o0(0(0(0~
0
00(0
C-- (0
o
1
•00..00.
•O
(NO
C(D-J0
0
0"''~~~M0.E
(0CL
~0
O.
:
5
<II
E
0
_
CCCO0.W
U.
0.00•
SC(000C>Si
~ ~ ~ ~ ~ ~
0
r c0
0~~
~~~~~~~~~~~~~~~
Q
0~
g
2E
0o
0
CL
*1~~~~~~
(.0
0l
ii
(0
N
C
,
Ic
:
E
ID
(0
(
CD
>- ->>.P_
. . . -.
(
2
2
(0
(
-0
(0
2
0 0 0 0 00
0
. 0- (
. (N
0. -2
a.
0
CL
(N
0.0
0
0
p
-
(
(
2
(0
2
0
0
0~
-2
2. 0
(N a.
0(
C,
CL
'-
2
2
0.
0.
(Nc
20
D
0. N0
-f
"-C
(N
0
0
~~>
i 0
U:
ID
>-
-
(0
0
L
0
CI
a
(N
(
o
0E (
0
0
(0111
0(0(000
~o
ID2a
o~~~~~~~
0
00
>
7
0(0
(0
(0
--
•
0
>~
0
22
00
IOll
;W
3
00!
d
oo
O
o
c6
u
un
n
UflJ
-.
F
B
o
E
o
C,
'2
.-
z
u)
~'
,
0
0)U) C)
U)U)~~
0
0
B
~U)
0 )
~
oE<
U
>..0 (U
0
~~~~
-
U)
0)U)
U
.
0
)
0
0
U) U 0
U) U
0
C
0
)
C,
(()
U
)
0
2
2
IC
C)
o
,,-,
.
-O
C,
0
E
o
x0
o
<-
~
U)
U
)
C,
E
~
(U
(U .0.0
L
a
0
<
U)
C)
o
o
0
)
U)
0
)N
al
.D
a,
0
)a.
)T-
..
0>1
O
0)
C) 0,
C,)
.I
8
0
0
U)
0
0
_
8.
IL)
0
'
.
x2
UE)
'L
T
V
ID
)
o
..
0
)
.
)N
)E)
m w
.L
I
N' L
.
C, )
C
N-
C)
C)
w
w
-
CO)
.
0)
'
(r
m z
o
U)
Uo
)
cx
)
I
L
I
0
U
|N°
N
N
.
...
0
0
N
N
.
.
2
.
) U ) D
U
C)
C
)
C
U
(L0) ,0)
o
U)
C>
a
-0 co
r:n,,U)
.
) U
L
C-O
)
0),-
U)
00
-~ xO
V
I
) '~U ) U:
C
C)
C
U)
a--, -,:
C) L) it
0
.NT
2
<
c')
.LI T
0)
..
.
.
) .El
U U)
(U
(U U12)E
U)
21
2E
212
0
0
0
0
0
0
N
N
N
N
N
N
N
N
a
)
CC
U)
N
U
N
1
N6 U)
.l .D U)l .D ).
0) a,
0
0 ) a,
E
0I
U)
' .
L'
2U
:
0
U)
'T
0
eM
2m co m QIoI
..
)~
C
I
a
U)
.2
~~~C
C
C,
-
,
I
0
E
c.
o =
I
r
2
~.
1.
0l
t.
'
~
)
,-
D
o
2.
C
co6a,
-
oq
o,
2 2.
~
0
o0o
) -
U
)
C)C)
o
)D.
CU
O~~
~.
-'0
O
~
o
U
U
0
§0,
C)
' HI0 H E
°
-N
.CDu..
LL. ,
0
C)
U
<)
)
t
)
o CU
a,)
0
o
)
0d~~~0)U)U)0)U)~~U)0)
-'0)0
U) U) -
-
0
< < 2< 2
.
o
N
ta
E
22
~~
UGE)
C'S
o
0
o
U)
~~0
'E
o
cli
0
o
aE
z2
~ls
6
)
E
OIT::)
~
oo
-
~2
c-t .6
0)j
o)
z
N
3~~~~~~~~~~~~~~~~~~~~~~~~~~
U
)
0
U
)
)
_
o0oo00o
N
0-,
U
o0
x
00
~0'0
A;!^!!suOS
6
i
4-D
5
V
)
Nd 0)s
0)NU
U
).
(U
N U)
)
-
U)
U)
0)
U
C
U)
U)O
t.C
2
r
X
r,)b~
CC)
~~~ ¢54S
~Q ~ ~~~~~~~~-2
-~
<
~~~~~C)
ax
N
U
S4
:O
N
)
t
Z7(
-(
-
~~~~~ "~- ~0
U
o
i -
2)~~~~~~~-
C)
(
-
C
__
M E
C<. <
5
.
N
Ni N
o
._ ._ oc
_ _
D
CZ
If)
.
D C 21
o
O
2)
N
o
)
E E
m
U
C)C ) C)
C)
C
C
o
C,
U)
o
.0
o
a
0
o
)
C)
C)
cla--C
C
o002~2-
U)
Ua
N-
U)C
Oo
o
~~~~
~ ~~~c
CCx
-
C...
..
) .CO~~~~~~0 0
I-
0C,
...... In.JZ-C00w
N
>-~~c
U)
E
~
._
22a,._
a).2 a)
.2
CI0)
~
cz
:
<
14ZC,)rn-QC.
C
o
E
B, E-00 --.E a
0
U)
4-
E E
O
o
a,
U) 222
U)
U
O
O,
C:
Nst
<|
IC
C
~~~~ui
All
.0
3:1
N
L
C)
I
'-
N
0
(
CON
NJ1N~N
NCn
N
~
IM
CU
.~
.~
CU h
CU CUr
.~
.~
~- 0
0j
~
UO~)
~~ ~
~
~ LL
CY
~ 0M
0
o
CO
0
li
T
000
0o.
h0
C
~ ~
00
.$
,-m
-J~0
CO
0
0
c
N
C
(0
CO(0C-
0
g
C
~
~
bI
o
Im~~~~~~~.
C
oi
~~r.o)
00
0i
~
CC
0 IcO
m
dLi LLI CO
*0
~
aE
OO C
CC
I
CO
:CO
0
N
C'
C
6 c-
c-
C
N
.,
o
.
~
_-
X ).C
76 c.
5 ??0
5
~ ~
)~C
EEEL
a8
V
.
Z
E
.
1
0
0
0o
0
~
0
C
O
10
NE
N~
(0C'-(0(0
ON
000 a)0
)
CLC N
Y
)w
04
oB
uj
0
O~0
cJ
01
IoZ~
,, v
wO~
0
0
0
CD 0
>
O'
>
-Cmm
r
>
N
O
l
-.
U
0
U
M 0
Q_) 'D
>
E
o
CC CW
-----
0
Z
(
M
,-
V
1PllllOi·(··ll·lllIll·P··
000
b
0
z 0 o o
M
~
~"
I~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~.
Z
<D
r
UJ
I)
?2
-
_
~
Im`5j Im
'5 '5
Oi
I
0)
un-
C
0
..
Ci
0
~~~~~
unj~~C
.
LU
L(
N
CU
w0
0
C
C
C,~~~~~~~~C
<
w
EE
-oO
0
<
Al!
(N
N
0
30 d
0
0
(0
0
000
NOLU
t- ( N
N-
LU ,
0W C
0 ).
00
0)
',
L
'o) IT
J
ZoC~
(N
C~
10
0::t~~~~
)N
(N
~
C-'i
0cC
(N
C~
LU
C'j
N
"i
(
N
E
,
Uflf
un.1
~
2
N
~
-5
a)
L
0
0
a-S
N
21
N 00N
2 E
N2 ID
M
N
~
IC
I
CC~fl,0000O
E-
E
E
2N
L
100
0
N
~2
EE-
0
fl
.2
-
N~
N
t
.
ID
N~E
0
C
C::
a
ON
I
E
o4
ZL<
a-fOLKW0
0
N
N4
(2
LL.
(
N
O
o
0
*C
U
o 0 o.
.2
0
*
C3
0
N
N
0
Ci
i
N
a
((3
(3
L
3
N
-
N
0
N4
21
2
0
0
0
C
2
~ ~
0
C
ID
0C
0
C
C(NoCD
-5
5
i
-~
5
o
m0
~2
2
It
CN
0
(
C
L
LU
LU
a.
a
0
0
(
c0-
I
I
~
0-R'
5
C:~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
C
C
C
0
0
C>
C
C
o
C
.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~'
~ O-,
ISZ
0
-i
C)
0
C0N
0)
2
~~~~~~~~~~~~~~~~~~~~~~-0
)
~~0
0
N
LU
o
Ci
N
~~)
0
C
0
N
~~
1°
Lu
NN
.
2
0,
N
~
o
E2
.2
(0
E22
I
LO (0
LULU
_
~~~~~ o DQ-,~
ui
0-1
Cm
-
ID
.
C
(3)
~~
2'2
N-N-D
a
ue~~~
CCN-
~8 ~ ~~
~2
EIDC
N
D
0m
N
C
NaN
N
0
EEEE-
0
0
(
C
(3
10
E-
0
C
N-
~
0
0
~~~~
m~~~~~~~~~~
0l
0
(
C
0)
0
(0
W0
0
o
c
0
-
'N
3:1
IM~~~~
0
N
0 c00
U-
Z<0
0
(0
N-
fr
rnm
00r
00
0
(0
-,co
0
C
m
o
z
U
All
1)
r~-
d0
0)
LU
C
LU
0
0C
31
0
3=1
9
oq
U.)
'IT
of. L
0
L U
L
LU
")
LUr
LU
~~aq)LU
.
'-
~~
,
,.
o)
L
01
>
0
E
E0)
>0
E cm a,
>0
C
oo
ao3
U)
LL
CD
-
LU
'
`U
LU
C2
0
a~~~~~~~~~V
LU w~
LU
Ci
0
CC')
C
50
a)'
L0
IU
9:
'-
~
0)
LUL LU
9
)
a
N
w
U)
2
aU)
- U =
.2.2
a.
a)
01
LU
U
L
LU
'IT
C'
,
Z-.
a---
CR
-,
a-.o
0t
%
U
0a'
'
0 a
ID~~'
.
U
.
0 0~~~~~~~~~~~~~~~~~~~~~~~~~~~~~5
I
0Ž0
0
0
a.
.
0
v5
L.~.U
a
E~
U
V
M
L0
0U)0U
2UL 2
Z~.
-I
>.
U
C-
z
LU
L
Lo'5oO
L
')LU
, -
I
N
N-ILU
- 0)
C)
dJ co
o
C)) CL)
o U)
- D-Ua)
<
U)
tO
~
-j
0N
'."C
0
d
LU)-
'LUN--ONI?
'-
U)
N
Lz
C
N
-C)
LU, LU a)L
,
0i0
< -)
m
.0 , 0 o
U
0>
a
0a
')I
U)
UOEV'
L0
a
~
J
C!N
LU4
C')
0>0
0
0a
a)
U)J U.)
C )
a
L.
C,~<D
w~LU
0o
a)
:
-LU)
C,.
U)
0
'5uJ
U2 U)
a ) N~
C,
0oo0
U
c
0)
a)
,a
a
COLC?
LU
L
0
'5
a.
>->0>0
U)<>0
00
-I".
O
a
a
a,
O)
EU
LU
LU'50
a
=
-
a
0F
C
a.>-
'5050U
U
<
a
-
0-j-
LU
<
,
<
a
0)
.<
a<C
6
a
_C
__c
y
w
r-a
(L
I 0
a
m
Ia)~~~~~~~~~~~~~I
LU
NI
.DLI..
0
0
(L
2a
000
-
<
C
ui
0
I
LU
.N)
NC)'C?C
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~U
0
(D
U)
Y"
a *2
oL
o
0a)
N 00a)
LU
LU
5
,wO
0
C')LU
,
N
L '-., L
0
U
(-)U)
N
o
02
a
c
0>Ua.
-
CL
0
o
o.2
LU
0)
,
L
CU)
_
U
c0
5a
U
.2
C
0
a. U.
CO o.
"C.z 0
.
U)
U)
>
L~..
0
C)
~
a
a
>a.
mCL -,
a)
2
a
Oa0
0N-
~
a
a.
LU
CD
.m
<
0
311M~~~~~~~~~~Ca3Ima)~
LU C'1
uoiloun1
0
-
N
,
0
cli
N
0
LU
0
~
>0
C
aIJ
w
.
ueo
CO ~ EL 0
N-
o3dP5 W i
c,.I a o
co
cc
.
0
)
E E aE
4: C>0
~
~
C
~ 0)E0)E0)E0)E0)E0)ELUO~~~~L
UeE)
.
I'
LU
2'
UU
U
E~~~ a.
'5
E
5'5
>-'
)
IDUCL
L
LU
IU
U
U0
E
CaVO
C
L
a
LU
L CO )
U
~~~~~~~
~
m~~o Co
'5'5
U
'
N C'"I
~~ ~ ~ ~ ~ ~~.
',
5
0
>2
'
=~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
=
LC.
:)E
LUC
i:
0
C
~ ~~~~~
.
C,
'
2 2
' 2
02
ou
la
0
U~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~0
0, *a
a
.
'.
5...U
U)
0 _ 0_
0
).E
'5o'
'
O
)
U 0
,-
0
LU~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~'
LU LU
M.
'
CI4
~ ~~~~
ar
~~~~
aa
0
., of
L:
U
-
u~)
~
0
L~')
C) LU
U
o~~~z
L
CN)L
U
L
C
~~~~
C
U
L
C)
L
L
)
~~~ ~~~~~~~~~~
CL0
LU
0
' C')
CU
01
Cf)
-.
C)
a
COL
L
C)
2
L)
-
'wL
)
.
0)
C)
N-
a)
Q
ULUC'
U)
~0
Ap
0
0(
N
-'
0
0
0
0
1
0
0
0
~
0
0(
00
3)3)
)3)
lo
0(
A!1!S
.
0o
(0 C)
o
.
AIJIS~~~~~~~~~~~~~~~~~I
(0
-'t3)
)(0C)3)C)3)3
.LM
o
C)
(0C)3)3
-
~
3 3
Uflf~~~~~~~~~~
~0 ~~~~~~~3'~~~~
.
0
=
_~r~
0
E,=
=
'-
'-=
0'-' 0
0
0
0
0
0
w
0
m
.)
~
0
C) N m.
e
(0
t')~~~O(0
0
C')
(NC'~O(N
20 2
75 l
Z
Q
U))
3C
0
U)
a.
->
>
o
N
N-
~5*5
0 0
.l)
30
U)
U) U)0.3. 0
Z
m,
..
I-
0
--
I0
'
0
~
00~
0U)
.30
U) U
.. 0.
~
3
m
)
~ ~
03
I)
0.0
o
`~
~
20.
p0
'
~
->
>-
C)
0.
m
Z 5'a 0 - 0
co
.5
.
.
N
w
C'e
)
.
-.
3
3
2N
0
0
ID a)
. 0 .
-s
a5
U
a)
0.
'.
0o
!!
,
0
0
C)
C)
.3
C)
U)
~
0
.
) a)3
U)-S
~~
~E
)
.30
C
U)
Z
->->->
0'
->->
ro
E
00
C,
.
a
0
)
a->
~3
3
o.
)0.(
0Q.N
-
-(
(03)
o0
.
C 33 -.
CL
)
0
0
C)
U)
(0
0
')
0
...
0,,~,c
E3.3
E
)U
(0
U)2
0
U).
00o
,~
0o
0
c
3
C
C~~~~~~~~~~
U)
(0
.
>.>>>0(0-)-)
'
, 0.0
m)c0
Z
~
~
o0
~
~0.0.~ ~ ~ ~
0
U
0
=~ =j
N-
a
0 '- o
>
~;S:-;
m m
0.I
.3.30.3.3.3.
.
'))
C
C,
0O
U)~
00
3
0
CL
o~~~~~~
~- ~----…
0_
r
~
0
0
0
0
00
==
U)
3
Z
U .
0
o
EE
1:05~
0000
o
;
w
~
00
'N
:c
C')N-U)C')03)(')3)00
'E
co
O
O
U)a o0
.
3
.0
0~
00
- c->
>->-->--`
o Eoo
3~~~
00
N
E'
=5'
2
7
0
..
(0
O(0.
.)Q0
U)C'
.0
~ ~~~~~~~~~~
o
~-
UO!13aUflIs;-s
'
3)~~~~~~~e
~'--~~
o
un. 1
')43
_
C!-
3)3)
......
C')
3)(0
~~~~~~~~~~~~~~~~~~~555555~C~C~C~c
C)
~1
N-
'0(0
o
>>>->
00.00005O~~
CD_
_
=
l ')
..
0-
=
.
Z.
1'0.
-
0
Q.
_
lo
3 )
W
->
'C
11
-
00
Lo
0
I1
-
Ai.S
c
(0
-
{
0o
W
LL
Lu
0
0
0
0
0
0, C
0(t)
'
UC 0
I
0
R')
^p,{s
~~ ~.-~
·
JBW~~
~
~~D
0 w0
1Lo'o O
m
c
(
0
C
I?
CoC-
m
c
I
C,
Co-?
w
11
.
*
C~~~~
1a,
1
m
('
NO
C')
.
.
C'.
.
.
C
C'
C,
0)
'D
C,
C,~~~~~~~~~~~'(-C
....
0
CO
~~~~~~~~~~
0)i
.
-a
C')
(
-0)
C
.
T
.
.
6
C')4
IM
..Z
U
z
b
olp
un.1
~
75
.
-0
(
0
c
o
Uflj
.'2
o
0
0
0
(V0
0
0
-
2
2OC
0
0
0
00
00
0
0
0
0
-
0
.
a
0
Cu
C.
2
(V (3
>o
>o
0
.
.
0
0
0
-
0"
0a
0
-oC.C
v
V
0a 0c
000
..
u
2
'
2
2
g5
0 C.E0--
-6
C.
'- 1
2~2
2.2
0~m
o
0
0o (V
2
- -
0
00-0
0V.
0
00
.2
~
-0~
,75
0
c
a.
o.~~~~~~~~~~~~~~~
00
0
0
1
4
z
0
2
0
0
..
0
0
2
--
zC
'25
a.
a.
0
0
0
~~~~~~~~~~~~~0
2
0
0.
r-
0
00
00 00 0 00 00 0
~
2
0
'
a.
EL
-
(
0
0
0
a-=
0o 0
2
-
0E
5,
00)0)0-2222
0
>,
0
0
0
L
0
0
'c'
0
0
(27 ( -F(V21
oC0= 0
0
0
E
o
o
_-
5
b2
.
,
.
.
a
..
000cc
00
0
.
CC
0
.
ID
. CO'
C
(CC
_
ue~
~
'
:
C
n
L
o
0)
.0
z~o~-,o-,~a
Al{!sue9
~
uj
CD
>00
2
C
CO
--
0
0
o '),~ ~
co
0
0) ~
CD ('co ~ C') C ~CcSo
cJcc4~
c
AIIAIIISUOS
')
0
C") C-r C'
C
' ' ' ' ) C)C C) (
"I~
a.
3IJ.N~
cl
C)
C
)
0
''~1 C')
.00
(0
0
CC
C)
C"
(0)
,
C'S0
0
')(0
0)
'=
0) 00)0 ' - = '0 ' 000
0
2) 2
.2'
0(
00
,
0
aa(
-0
co0o
0(0
(V
Eo
V
0(V
(V
(V
0
> m,,,~ ~_~.~
0 0 ~~ 9292
0
0
> > w
2o
,a
8,
0
0
C,
)
0)
0X
C,
0
2
8
V
0
w
ama.
8.
0
,. 0
2
2
2
C~a
~~!
w w a
X
w w
2
0-
0-
2
--
o
)
.
C)
C
C'
-(i
0
c,(Cq ~
C') (0
C,1
c4C-~
O(
.
a a
r
aD '0
D
a. .
Ca a
a .0
>. >.o >' . >. >'
~~
c
~
0 0
0
0
0
~
0-
2
>-
a
w
C
00
2
u~C-
Z
0=
a- a
>- C9~-C
)1
0o_
<
-
0
0
:
a
=
oIo...
o ' (V
-0
(
L
C,
o
o
(
'0
.0
'-00
0'
z
w0
i-
o
(0a
0
0
'1
C,)
a
C')
C~
001~~(
0C'c
0
V
(
)
E
>0CD
a
<
C)
: C')
0
0o
2
0
C
V
(
C
0)
<
' 2~a
'
2
2
C0
C/D
2 o)
> '
aC-
.
o
.~~~2/
a
0
.0
0
>.
C)
.0
2
c
.0
'
CCD
0
>O
oo
0V
0 (V0 (V
,0 V..
o-0
V'(
>-
Q.>22a.
0
C)
0'
o
0
00000
.o o o ''"
W '
O)
(V (V (V (-,(V (V (.
202 020 02 0
a C
2
(C
0
2
0>),OX
C,
LL
,
'V:
e
I
ed
0
L
'~
uoiloun::l
fl
CSC')
C') 1
-
ooe4{~
C
m C,~~~
.0(0
C)
'
'
0)0
C'-C
r
.C
0
00
.
C (C
Cc
C') c'C
C'
0C
(0
~0) (0
~
2a220
C'
0
( C'i
~ C
,t
t ~co
D~:,4C
C4t0,
~ ~~~~~~~
LO~'o a)
CN~
~ I~~~~;
~
.0 C'~~
q
N C,4NC,
-.4 Zo
C~~~
C~~
C~ I mo, C,
C"C
C)
............
co
1.
0~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~,
- .0 ~
0- C'
'0')1
C)
C
(0
-
><
.0
<
1
>
>
<~
>- co,<>>
>
a
C
O~~~~~~CO
:C C'
(0
JZ0JCO-)<0ID0O2LL-Z~~~~~20QC0)0)%C2~~~~~[..J
)C,
~~
1'
.
0'
CD
II
ID
1N
O
r-00 0D
C 1
C
C
Oo C
Co
Lu
W
L
-
N-
Co
u
N-
C')
oN
N~
N
L-J
Li
U0
I
Co
0
0
0
0
Cm
I-
.0
UN
CO
N-
CO
Co
CO
Co
1-
COI
NO a
w
I-
u
-`1
O
N
LO
CO
Co
CO
LI
a
a 0
-
0
0
12
0
I
O
~
~
C
J.M
01IL 000
Uflj
C
CUCU~~~lC
0
0
0
0
0
a>.>.>.
C
CU
0
0
0oCUCUCU
.TL
0
E
e~
~
>
000
'D
0000E
E
0
E
T00
0o
0
0
0
0
"
~
E
.
L
NO
a
2
Co
0
a aj
o(9a~~~~ai~0mCD~~a
>-
*
0
.
0
E
o
a'
.o
0Co
2
>a
0:
0.
"-
2
Q
a~
C
>O- --
Co
0)
02
a6
(9
-
,
>
N~
>-
Co
00
0
~
o
I-
io
/ooo
H
2. ~ I2
0 ,2~~~~
~
0)00/ O
~
C
U
l
I
'I
1
00N
~
0
CUCUC
0
I-
0
~~
~~
- N>
jaz~0
E
0/
m
o
0
CU
C,
-
-
mm
C
-
EO
a Co
E
Q-
-
C
E
-->- 0z
U
Co
Lu
Lu
OW
0
0?
U
0
0
I
0_
0
0
CU
C.9.C
CC
0.0>.
-
C
0-q~
,
OO
a
N-
-
IJ
l
NL
co3
L
L i L L
o
Co
-
o
CU
>.~~~~~~~~~
. m .9~~
L
CUI
0
n
0
O
O
0
CU
0
L
0
0L
E
0
9
CUm10
~
·
C
C
L
C
U9o
CU.
E)
.
>
U
CC
0
CU -.
I
E
0
y
y
0
)c
C CUC CUCUI0
0/ON
C
9oC
) 0
C C
.0
.
U
COv
-O-0
0
U
CU00
0
C
CU
CU:)
.C
CU
0
Co
E
C
..
0
-
~ Oa10
Co 0
)
-
a
CU
2
CUECU~
U 0 V:aC 1
U
0
oWWO
<
C
0
CU It
90CU)
N
N>- 0 ' O ,
O
aCO
zoo
L(9
N
CU
0
I',UC))J
C
O
C
CU:
u~~~~~~~~~~~~~~~~~~~~~~~~~~~m
o-, nCO
0000
IlL N
Lu~~~0
U) L
- LU 0-
CU z
Co'
1
O
Co
CU
Cl
OC
.
CoO)L
0
U
U)
--
--
~
CU
C,
(,I
o
o
N
~
>a-
o
0.CI
O
z
- -
2
N
C
zLCz
111)
>2
N
N-
C
.0~~~~~~~~~~~~~~~~~~~~~~~~~
00
000
>O u
-
C
~
0
N2
0
.
CDo
0
2
00
.
I
C
~
CU
-O~
9CU9
LL
N00
U
0)
L0'
0)
w~~~~~~~~~~~~~~~~~~~~
9.0
0
0. 0
CUV
9
'IT
O
~ ~
~ ~ ~~~C,
c 9.2
)
~ Zqa
UCIm
C1
C
~~~~~~U
N <
000
N
C.O
,-
CI
LLJ0 I
I',
o
o
M>
~~~~E
~1
-
0C C?
o
?o
0=
0 000.
0
N0
oO
C,)
Z
1
U
EEEa
0o
1
(.i
. .
. m
xLi
C
O
-I
io
/L
(
owo:
0
O
O-
2lCUC2CUL
00
COC,
~~~ 9
a.
LO I
0
~
Co
00,-
000
CU
U)
~
Co
0 CULu
000000
bo
u8E)
~
o
T>-0C . . .
>>~~~~~)~~>->~~>2
UO!~~~Uflj
o~0
C
._
C
CU
-
0
a
aa
0', · 0 ,-
~
0
)
0 U
.c
a
'0
- ......... -a-E-
~~~~~~~~~~~~~
ua
CUE
Lco
2S
)- -
' '_
0
E
.0-
0
°
'2
0
75o
"
0>-
o
IL
1
-a
0
000
---
=
cGl ub
Li...
CU
0
a
'2
ORO
9
0
0
0
Oo~C1
00
-
C
)
LLEECONC
.
No
,C
a) D -0-a-a,
99
U 0U
C
)
C
C
0/
O 9
) oO
a
CO
o
0
0
U
0.
.O N
-N
- I= - u
Z0
C
0
0
C
N
AIlIS
Al!
3d
o
o
o
·
m)S10
3:1
~
N
o
.
I
-N
o
~
o
o
O
o
0oo
L
o~
o
oo
10(0
N0
1
N
N
_
(0
(0
N
o
o
o
o
o
~
~
tDNNOOOO
O
- 0 (N
N 000
D N(
(N
_
o
N
N
N
o
0o
o
o
N00
0(
NN
(-
o
o00
0N
M N N
o o0o
·
.
st
lo
Nt (N
0
O
00
0
N
o
0
o
:
(NO
ON
_0
o
.
o
()!
Da
10
(N
1o
o
(N
N-
r
a
a
(0|0
(N
O(
N(0(0NNnIN
)I
U
II I I I
j~
unlj
u
=z<
z= <
<
<~~~~(
'<
L
_
'It
B
B
B
z
z<=z
am
10
< z.z
z
(N
0
0
zz zz
1
0N
T
/{1}!^!1!sues
ue
~~~~~~~~~~~~~~.=_.
I
A
1n1t
N
Ujq..
~~ ~
oL
I(
3JJ.
Ct-CCCi
:~~~~~~~~~L
g OV)( 1c
CU
C
(N
CU(N
C
0N
0
0(0
C
a~
_gB
0
1
N-N-N-N
0
0
U-
C,-)
"iC
uo.i0loun0000(
.)
1
(
,0
0(0
I0
1
1.
aA)3
(N
___
C-
z
0>
zz zz
(
N
0
(
>
N-
ON
i
0
C
Z
,.
.1
N
1
WA~
</ WN
)O N
U
s
!
I
,
z
0
0
Z
0
m
|N
L
m
-
(
z
ea'Nw
C
0
(
01 N
co
(N (-y
1
z
zz
U>-
0(l
Z0
(N
N
N
C-
NC(N
(0 I
ON
0
N
(N
xO
101O
()
8
N
0 3Z
N<
(N
o
c~
L
O
N
WO Z
,
Z
0
Li(0
I
N
(NO
(N
(N
o
(N-
CO
Z
o
1
,
~t"
iN
R o-,
' N4 N-6 o C
' '
|N 101
(N1
I
{N 1NN (N0
F-(N ,i N
ar
C'
N
D
CO
N
|
(
I2Ymo
z
Q
eq
|2
z
10N
C tN
(1)
U
C<
C~~
o
o
00
N
1°
N
-z
LL
0
.
1o
o
(
z
(M
LLL
.
ooCN
w
0>;w
aL -(DD
0
|
D
00:1
000
0
a 0 o
1N a 00C
CU0,)
(0
N
t °CC--CU .0
C? 1° 1?
CUC
ilzv1
Bn
0
zz
~
1q
Z
sIN
(o
~ ~~o
~~~0
0143
zz z<=
N
' N
i-
-
C
C~
(
N
Co
C~C~C~
j
-
A!!sues
0
0
0
00
0
0
0
0
0
00
0
0
0
0
U)
0
0
0
0
j0
I'
3jt0iuC
3p6(0( ..
' o
6 ,
(00)
:
0
.
NOu,
CO
~ ~ci ~
Zo
I
N
Ncq
o t
CO co
c6' 'i
N '6
0o
,oq
N
C)
c
li
oo0
o CO
q ni
cq ~
¢)
.._
.)
uoijounl8
¢)
EE~C
U)
hi
0
-0-
L
2
0
)
0
I3
I
I,
i
1
r
$U)
CO
Coo
CL C
>)
(n
0
0
c
2)
_,
CO
aO
o
¢)
n
<
cz
2)
~IAI
~_l(UUS
.
o
00
uo,unj
~1
(O 0
0
U)
o
o
o0 CD0
o:
)) -I
~0 0
C:o
U)
i
n
n
0U<
U)
o
U)
O
U
1
*0
J.M
>
D
0, ) 0
_ C:
a.
r a,
nr~~~~~~~~~~
-
zzzz:z
0
0
zzzzz
~~~~~~~c
~
4
jo-
o
oio
,d
ci
0
eue ~o o
,
o
Ajjlj!UOS
[~~~
.
.
C') 1 CO
c,
a
o
r('C
o
,
;
CO
('0
0
C0
od
,
'-
W
)NCD
,
CO
0
C
CO
C O
D
c
s
-0
c.
0
0
-E
2)
CC)
M
:i~IOCooo
o ('4N 1
O
0.:)~
0,
,
C, oPE
llq
C,'o
0I-('40)
JO
C3
C3Q
(t
C)
e
(I,Q ,<
,0
X
CO
N
0.0
N
a
I N,
N N;
0CO 00-tC'
C69( 01
O~ Ci
ClC
COZ>H
8
s
('4'-* (0W)
CD
a a
rLL
~
(OW CO
a,
0)
ID
i
C
o
Z
a,
A-U
2)
rI
2)
'A
2)
C.)
o
,6
0
o
.)
-0
m
Co
2)
2)
)
0
)
.0
rI
ji2)
Eu
W
UI
;W4
~m 0 m
v,
-~
~
--
ooo<
A1!R~ua
,,
)'
Com
-
N
'N
w
C:
CO
CO
CO
uD
~ - ~iDI ~
C:C .2E
00
<j
.C,, CE
o,
33
d
0=
U)U
CO
E,
U)>
0
) )
uo~~launj
II
I
Co
2)
>'>-- >
C
~
~N
(-
_
O C
~
-
~
U)
~o o
0
-t
CO
U
U)~
U)
0
0
~~2 ~
,N
LL N0.
0
COO
O
0,I_
-
C'
(
-
CO
C? 0
iT
('(
N
('0
C
C,
cL
C,
Li-
m
C
C:
(1))
I,
C
00)
'C:
1
V
.0
C
C)
V~a
U
ISd
c)'E
C,
LL
0.
E=
U)
C:
m
L
f
C
U)
CO
L"N
75
U
75
C
0
0
i:
CD C,~~m,~~
'a
)
0-
yw
o
A-
g
>
N
->
- > >- ->- >'i'
>
->->-
0
2)
-·
E
:
0
x
C
0 0>.U
,,
U[)0
C
0a2
rry
II
1
0(
~
WC
In~~0o
2,
. 0
1 7
-
'.
-
2)
C.)
tnm
0.
o
C,
2i >-W
.9 a
U0
C E
c
>
) ID
c
0))CQV,.
I
~
--
,
U)a
0,
6,
~ ~~o ~ ~
am
L
c( (
E)
.
o
Cf
~ ~ ~ ~~L E
)
C
~E
~ y ~l ~ 2
E
.~ 2 )O 2 ,
C:
CU) 0~~
E
0.2Q OZI"~
Oacu
5 ~2
tr)d ~~~~ca~N4Q
~It C, .0m
C,
~, m 2 oO~
eue~~~~~~~
0
~
~
~
~
~=
~
o o mm o
U)~
.C:
2)
r0i
4-1
3316ur0
C
m
N
C.)
$:
2)
C (0 (4
N (4(
N
->
0C
'
>
(
C
O
0
D
W)
:
FO
C
0
z
C
I6
>
0)-
I
->
I
I
->
1'-
t"'i
-e O,-~
*-
aa)
(U
.~=
4
C<4
_i..5_,
4-3
.-
-;-i
C~ C)i ;
Ei
-i
-^
i
t
O
G)
A
L
im_
C_C
-t
C)
C)O-
ct
Ci
CO(N
CZ O
N
tCO CD Co
N00 I
(NI
ocO
_
-
CN
(N
C)
C)
J
O
D
C_
CD
-iC
)
(NC-
CO
(N
CO)
ICN
O
_
CO
_
cl
_
N
C)w
N
0)
C)
( C
C'i
Ci
4
0
0
'IT
N
-
C)
(
X)
NI
3
Qi
(4)
OC
1
CN!
CD C0
C
-
flLO 0)
C'-~ CD -
C-
N
o')
c,.r
M
r
-
C)
(N
N
CO
cP
-
°U
C1
c,5 4
'CN
U
0
)
CD
LO
U
m-0 CD
i
(N
)
-,
NOD
C
-
O
4
O
c
O
0 (NJ (
C
C)
NNtn<^
c-i
2
C)
O
C
7
C)
O
C) °?
cli
A oQ V
(C)
>
Cl
-e
i
.
.C's
0)
o
NJ
'i
C)
f
C)
5(C)
JCO
Si,
C
-ct
V
*JJCX
LU
Lu
LLi
li
CO
CO Li' C)
LUi C)
0-~~~~~~~~~~~~~~~~~~~~~~~0
C)
)
',U
LU
C
)
LU ILU
LU
Lu i
.i
~
t~
~
C
0)fj
,<~s
C
.° o
C
ao
-C
o
t
C
C
<
v ._
5'
c
2
2
2
C
a
C ,
C
C
'
Ct
x-E
C
C
~
Cv
C
:,I
)
-
7C
acVQQWuth'
2
C
<
cf
wzz C
2
C
C-
C'
LOC
CN
C
C
C) CD
W1:
)
_
C
U)
C)
.l
CD
~~~~~~~~~,
r.)
IC
C
W
co
-
---------
ry
~
to
C
UD
C)
0
C
x
*C
0
f-
C
C
_
~~~o
C
I
a
~
e
-----
C
U-
y
ofC
C
0
0
CD
co
C
-Vi
C
PC
(D
[~
Z-j
)-
(D
PN
(N)
-·-·---------------------
C-0i
-C)
N
C)
CO
1''-
0
C)
C
-
ID
_j~~~~~~~a
-
<
C)
O
~C
N,
CD
U-C
.
)
_j C)
C)-
x x
C
ry
o-
)
cD
,0
co
<(
M
0
CC)
C,
C
a. cc
CD
'lCCCCL)ON
0
CO
CI
vcov
C
x
~
~
0
C
C~
CN~~~~
C
C
o
Cb-DmC~
1/)
<w<mzOzz00
0
C
1-I--------------
t
C
m
sci .2
c
C)Z
C
<
@
Zi
r
-
c->a
C~~~~~~~~~~~~~~~
IwL~~~~~~~~~~
.j
0)
~~~~~~~~~~~~~~~~~~~~ci
fD
C
C
O B)
C
C
>
>
'
I o,
,I
~
0
1
O 0:. -
L
U
-T
I
( N
LO
(
1
t1
1
-
N
1
N
1
O
1
0
,
(N
i
Ct
O
fi
I
CD
N
-
L
(
(-
,
(D
.1L
t
LO
CD
-tXNNN
7
9
t
|NC-
00
cq
'T
u.) lo
Cq co
C
c
In
C
___j
0
L
OIN CO
1
0
O l
0)
(N
In
-t
CO (N
(NO
In
CO
N
cn
L)
1
CO
)
N
N
t
C!
0X
0N
(N
r,
CO
O)
.(1 CO
)
(N
.t
O
u:
-e0
C
·
0) ~
(N
n
O
C
0-
0
C
(N
o
(1
c
N
a)
IN
~T
Oj
c~
CD
C'i
co
N
Lo
0
o.
*0h
Ct2
a1
4'
0
(a
U,
CO
a
In
v
.
0
OO tO
O
N3 OL
_~~~~~~~
0
O
o
__
o
. -4
0
0
CL4
0,
~~ ~ ~ ~ ~ ~
(n
o -z
O
o
o
o
0
oLoI
_
o
-f( I?
Z
"t
ZI w
0
q
I
Z4
W
w
. )=:zz zz
.2
z
z
z
z
zO c o
z
. Z -,
o
,
__ _ _
_
_
_
_
_
arC
a-
c
a)
-0
m
~_______
w
0nw0<
~O
o
X
o
2c
0
>~-rwC
>-<
oWQ
-D
o
o
o:(
> >- >-~
B)L-rr
Bo
- cu 0
-
>-
2
m
>-
>
w
-
w
m
~
~
.5 3-
~
>
- > >u>-~QJ-JWo
zz
n-
a) 6
0
a a)
1.
CD
a)
N
I
IX
-
.,------------.-iI---I-------
0.
w
w
Z
C)
.2
z
0
12
2Q
Iz
co
o
Iz
1)
C9
00
-
200
~0
w .0
.E
Iz
a)
0
(O
U)
2
0
(0 co
CD
-J
-j
L>
m
02
.R
(D
Iz
E .(N
0C
E,
(D 0
o
Z6
(12
CD
CD
LL
ix
°
a)
l5
c
11
LU 0z
m
I
cl
Co
a)
0
,o
6
.4
.)'t B
_
m
co
E
.2
(4
ma ma m cz
m
O
O
.
D
()zEz
_~~~~~
i_
0
0.
0.
:(4*
_
r_
0)
l(n
Ca
C
;
.
0
w
12
_
14
(z
I
5'
o
;
-,~=a Lq C'?
0r
(12
Cs
4
I'
c
(0
0
0
tl i
C;I !J
l O)NOLU
00,,
0~~~~~~~~~~~~~~~~~~~l
_, _ _~_ _
Lo
co
mc
fi
It
z
O
IN
lo
0
UD )
0
>-
z
1L
1>
0
0
>-11
Ilo
Q
(0
0J
09
0)
N-
00)
ol0
6
6
-~~~~~~~~~~c"
0
(0
I
0
0)
c~~~~~~~~~~~~'
,0 (a
0
4X
It
AV
4s&W
6'
WV
0w
.
~
-
(0
(0
It
I(0
0)
~0
0 3 cn
(00((0
100
1
0)
'0
N
C,~~~
- '(N
(0J.~
-
0
(0
N
0
00
(0
~~~~~~~~~(ci
llo
OLI6W6AV
BV
L;Bu
(0
I0(
N(4
(:0
(00)
0
-A
(0
-
0
0)
~~~~~~~~~~~~~~~~~~~~~~~~~~NN'IT
C~~~~~~~~~~~~~~,
t ,t
N
0
o
(0
I
(N
C,(0
0 0(
C'(
N
(0
6M
(
(0
(
0
-11
N-
0(
1~
(0
(0
(
I
(0
-
(N
N
-I
6
(0
(~N
0
(0
(0
,
I 0
(
(0
1
(0
0
(((
a',
,,,
0(
0
(
0((
(N,(0
0
-
® N(0
-
N
(0
0
00N
I-
( 0 ( 0 -
(0
(
-
(0(00
0)
0
0
(4
0
,0I:0 ((
a(0,
0)N
~
OC;BW BAV
(0o'0
(N
(
N:(6
(
6
O k,.L
10
(0
(0co
W§AV
0
0
(0
0)
-
Lj
(N
N-
(
.
(00
(0
(0
(
(0
(0
(0
(~~~~~~~~~~~~~N
-
N-
0
(0
(
)
(0~~~~~~~~~~~~~~~~~~~~~~~~C,
(0
0
N 0
0
N
)
N-
(0
I
N-.(00.(~~~~~~~~N
UD
(0
a 0 00
(0
N((
0
-
(N
m
(0
a
w00.6 w
(0
(N
0( N C
0
N
(0(
(N
LO
(N
(0
Nl-
(N
(0
N
(0
(
-
c')
-o1
0D
09
N
J.M
(0 (
(0
(0
(NW (
(N
0
N-~~~~~~~~~ (0
(0(0N
(
-
(
C
N-
(0
(N
IN
(0
-
(0
(0
(0(0
-
(0
(N
-
.i 5
O
JMuAV
O~.L~6~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
(0
(0
0)
0
(
m
(N
(
(N
N-
N-
(0
In
0
0
In
N-
0)
0
N0
N
(0
(
0
NN-
,,V
(0)
0
(0
(0
0
OtJ.LM AV
(
a,
O54_ U
<
_a
zU
X
O
·
(0
-
N 0
(
(0
(0
C
0
(0
N
(N
(0
-D
-
(
N
I
(
0
U
o
00
co
X
.0
N-(
0
-
-
(
-
(0
0
~~~~~
~~~~~~~~~--
(0(0
A~~IAI~~IS~
S oU
. E-=
EE
~~
0~
-
0
E
E
.2
-
(0D0 --
0
0
E~
U) co
(
(0(0
0
.0
.) I
0
m
0
0
.l
zt
0a 0
)
C,0<n2
0
:-
>
0
E0
c
E 0
~
(No
0
-J-
00
0
(0
o
(N
N
0
0E
E -c
Z
2
~z0
U)U
a
C
0
U
E
U
_
'F
E 0~
0
0
L
0
~z
:__
o
-
(((
0
-_
C
E
0
E
0
0
n
tm
U)
cu
E
-
0)
-
O
-
co
0
(
00
(No
0 -
>-
C
>
Ol.16mtO91
09
6
i,.: O
W 6 AV N
i
N N
(6
tN
(
.- ON-
O
Ot lW
1.4
Ocn
AV
-
L
0
c'
OC 6W 6AV
LD
i.
r'-
(6 i:
6r
OZ ;6W BAV
(
0
CO
0
N
0
C
0
o
-
IO)
IC n
oC)
co
.4
0
c,
60" -6
09.M AV
N
O c0
N
®
N
(N
m
CO
)
..
ss
0
09SIM 6 AV
z
Ot'.JLM 6AV6
0
(NMO
-
(O
C')
I
10
c,
0
c(
c
m
'I=
OZ.LM BAV
(6 6,
V'
-N0
-
(_
0)
0
<
<
;.4
Qd
c,
U~
;.4
O
I
0d
coo
0
0
co
(A,
CUL0 U
w
;4
en,
F-4
E
....
I
-
.- >0- >.
__
¢-
o
0
o
t
n
rC
c
N
N
1
N
c
-t
O
N
N
N
N
N N
'N
-
-_o_
_
O
C)
U'O (0
C
~(N
N
N
t
N
N
N
N
LO
\N
IN N
"
N
C)
aI
O
)
)
O
N
~N
N
O
CO
(N
tQ
(
o
t|
(N
.
0
CO
N
(
LN
C\C!
(
Co
0DN
C)
(
IN
00
LN
04
LO)
4
(o
(
NO
0
,r-
o
_(
oa)
)
CD
c
cO Co
)
C\
N
(O
2
'I0
0
I- '
Ic.O O.
N
r ?cO
- a t
N
(N
I- o
N
OCON'
'-CO
I-'-
~
c
X
O
0m- 1
CtO
-d
CO
N
NO
CO
0
L
-
- N-CO
-
(NC)
oC)
C0
2
CC
O
0)
.
(
N
'
I)
tO
-
r
C
C
'
M
tN(NN
t
o
&t
I
o
-, 0
(D
C
C
t)O
s_-
O
O
S
N
C
o
cOO
-
0
(N
Ct
O
N O
O
C
'ON n
-N
N-
0
N -(N(
o
N
o
0
CO
ON(N r N
O
N
NO N
CJ
O
N
N
(S
CO
C)
N
)
0
C
CD
O
C
N
'T
CO)
-c
-
C02N
tC)
o
m
ON
oN'
-.
.
O
C
o
OtOo
N
C,
)
)0
tN
CO
(9
COW
N
0o
1
COt
C
-
CO LO
COW
(N
00)
O
OC
)
O
O
Dw
N .s IT I
C,
.
T,
.0
i'
.c
IT
m
I
N
wNmmm
"t
m
T I
"T
IE
~Z
.
._
,
,,:-
0
~EC
-
C.)U
A-'
I-
EC o
-'
0
a)
0
N O.
Z 0
OI
0
0
0i
o, =
Oc
CL
C,,
.> E
>
B
B
0
w
o2
) >
o
0
0>-(
C)A
N0
0
0
>-
0
)
0)I
0-
E- 20.
00
tE
C
-- 0
7
Z
Z
Z
(N-Co
'
-<
0
0
:
>'>O>-
C'
7- I> >->- -
C
)I
t
---------
C
N
~ 0
cr
~
_
6
a:
C)C)
· ,.z
z
CO
-
a)C
Eo
A-'~ tO
0
C)
(,
f
'.~
a)
Q
C
C)Pi
D (NW
CO7)
(N
C)CA-1 C).
ACe
c)
I )
N
O)
'
-
CN
CO4
N ON .
CO
I--
CO
CN
c
'
5o
r
0)
N
c)
CI-
9. C)C
B~0
,
wo,
N
I
o
t
-
(N
O
C)
D'(0
CO
N
T
O
LOI-
11 N- I_ N
(N
0)
Co
0)
~~1b
Ni
N
N
o
Co O
)
O 1 0, 2
a
5~
Q)
C)
NC
1CO
-
OO)stNrq
CO
-
o
b
.) ,5~
2=
P
-OIt CD
(a
N
1
-~,r,--
t
o
N
o
m:
w
O
N
P-CO
CO
0
NO
NO co
I_C
CO
--
0)0
1
*_l
om
(
C N-
.
C)
0
o
O
C.)
-
· ,~
~
O
0-
Co
O)
(n
o
o
n
C_
O
2
O
C
OC
L
t
o
a O a;
N
O?
4
N
---
-·.-----ssl
-
--
5-
0
I-l
N
u,
Co
o
0)
C,
~o
r
N
Ct
CO
co
N
co
0)
(N 0
co
0) Co
0
co
"t
(D co
0
Co
m
co
0
CO
CO
co
Co
0
m u, 3
N
C,
0
co
C)
to
CO
m
G
CJ
N
')
co
D
0
N
N
co
I--
0
>1-o
co
Co co
N r
C)
co
o
0
Z 0
'o
$ 3,o
'
0
N
0
Co
N
o
0 0
.
cyo
0)
N
Co
t
co
0)
co
CO
a,
N
Co
-E
5
I-Q
a2
3
C
6
C)
0.
ti
0)
C.) w E0
c
0
Ita.
10
r
N
Co
a,
n
.9
-S
(U
J
0
nr
C
n
-a
0)
0o
V
o
C%
c
C
O
.O
0Q
0
,
N
C
a,
0
0
4
C
w
,I
r-~
C
O
rr
C.,
o
o
a
a)
70
E
nr
:u
m
n
z
JI
(U
Q
ao
i
-
Co
y-
D
r
N
14
D
D
y
:t
;l
I.
-i
L
ask
~~~~' (0
0
¢
.(-o
D
~w
E
.0
z
E
(0
o
0
"N
't IT (N
0,
0
<
-
E
LO~
~ ~~~~~~~~~~,,
c
(N
CD
-ON
CD
c
._ a0
·
CD
~~~~-.
m
-
-
N-
Nl-C
m
N
CD
CD
C
0
LO
LO
C
f- N~~~~~~~~~~CD N--
CD~
CD
g
3
2)
DUIN (N
CD y,
CD N
0~~~C
CD
N
CD
,,~~~~~~~~~~~(
co
I
CD
CD
-
r'-'~ CD
CD
CD
0
0
CD
.-
CD
C
(,
o '
CD DC
00
C ON-N-
N
OC r (
N-D
-
CD
~~(N
NCO
O
-
NC
D
CCD
(NIDTC
CDl
D
N
(N
-
CD N-s
1
N
(N
C
0
00
(N
(N
L
C
IO
(N
C
,-
D
CO
D
NC
(
D
D 0
c 0.1 O
DUO
C
CD
D
C
N
cD
CD N
rCD
-
6
(N
r-
!;t
CD
CD
C
(
Nl-
I
C
N
co
00
C
N
N
C
CD
N,- C
CD
C
CD0
N
D
.-_
U'
N-Dco
C
0
NC
'I
-'-CD
CoC
N
- ci
(N
D
I'l
~
LO
Lt
(N
N
LD
O
N-
N-t
CD
,
CD
T
C
u'
.,D'(
,l
co '
C co
CD
co
I-)
-
C
CD
4CDi
0
CO
fl-C
LO
DO
'IT
DCO4
Lo CD to
O
N-
(N
m
LOC
N-
-O
-
CD (N
CD~~~~~L
CD (N
CD
N C
CD -
LO
C
N
'I
0,rC
(N(COC
C 1N
'CD
....
"
~0
~
V-
0(
C
CD
C
C
e
(ND
C
D
C
-
N
C
CO
'1)c,
(
IDCDC
N
C
N-CD
0)000)
'
0
0
0
N-CLD
C00
C
.,.~, a
U)~~~~~~~~~~~~~D
~~~~~~~~~~~~~~~
~~~~~~~~~o
U)
~~~~~
.2
C
CDo I0
a
eo0C
C
(NO
D DCD
o
~ ( C0
o0 'to
._
eOC
0
(N
(N
CO-
oo '__
,V)LlCB o
CS*
E
=
-Q
>o
= .5C .0
-t
-
co
N
C
CD
.2
=
(D 'm
LOCDN
N'
(D
N
CDD
_a.D
._
q
CO
C
L
0
(N
u)C
CO C' l
CD
CO
CO
N
CD
000
X
C
N
L
;z
CD
cD
o0<
- N-C
2
-D(
~ d
~ o..0
Vj
C CC
0N-C 'ITD
C"
c
D
'IN'- D
N CD
CD 0
0
C- N
- N-
D
0
CDCDO
- CD O
C~~) CD
0 0
·
N
N
DtC
DUNMCD CD
CD 0CD CD
C.)
,m*ni
. . -.
CoCD
CD
CD
00CO
0C
( N
0)~~~~~~~~~~~~~~~~~~~~~~~~'
0
*',~
g
-,
Co D~
o~
(0
Jo
c
C
DQ
-U*
CU
C
OCD
(
0
2
0~~~~~~~~~~~2.2O
I
)
C 0C~~~~~~~~~~~~~C
o
C .
C7C
C-
C
C.2
C
00
C
,
I.)
:>-
*t
a
~~De
E
C
CD
0
E.
0
3D
5t
C
C*
~0C~~~~)
~ ID~~~a
i)
ID
CU
CU Cl)
CCUC~~~2)
C
2 g
O
O .C
E*
C
-(
-
E~
E3
C
.C
.
c
M
E
.
0
~~>C
C>.-
-C
U
0
.2
C)I
CDC 0
0.0O
E
I
..
l)
-
IDV
0
0
~
<)
<i
<I
Ci)'
CL~~~~~~~~~~~C
~-CD
~~~~
~~~~~~~~~~CD
C
0
0
>, (
o
=Ao
U)
>~~~~~>..
<
0
w
C
C
O
C
D
r
In
~ C.C.
0-.
z
z
0)
E0
f
IDCOID
s
VC
-
C
0
v
H 0S,
;
Oq
CDr CD
(O
0
,
~~~O
Q~~~~~~~LQ~~~~~~O~o
w
N
coo
Q'QN',
~
D
_M
~
a
00
O.)
O
o
,
OZ
)0
co
(O
'm'(
N
(N (0
0
(N
0
a
0
0I
(N
CC~ (~
f-
0
(0()
C)
i
.
0
a)(
(N
C0
0
(JN
0
I-
0
CIN (N
N
(N1 (N
C'N
Nl
N:
N
CN (N
(N
.
(0)
E
0
E
IC,0()000
N -00
0 m)
C)
O
(N
E
m
N(
co (0
-t
0
M)
E
NO
(~m q0
rl0
LO 0
)
0
0
(0A
(
ON
UN)
'0
0
o
0q
0
~
(0
0
)
(N
0
IC)
0
.0
Iq
N I0) -(
0
0
0
00
0
0
0
co
N
0
0
'0
C'?
(N LO0 N
0
0
f-
a
a)
)
0
0
co
0
C
C'
0(
0
0
N
0
-
a
0
,
0
-
a)
F,
0D
0 (O (00
"0
t
-Cl 0
)
0co'0
(NO
(00
(N
N
<-
E0(0
0
o
;
(N
0
-
(
)0
00
0000
,-
U')
(NO
(N
~ 00
(N CI
N
~~~~~~~~~-
-( -
(N
(0
4
0
-I
(N
0
'
0
LO 0
a)rC)
co
0~ N
(N
'
'-N
0
(0
00
(0-M
0
~~~~~~a) ~~-
00
N
r
0
N
(
(N
IT I(0~
(N
-o
(0
0
c
CC 0
(
0
a)
No
)
(0
(
co
a)
(N
(0
'CON C
) 0
0
(
C)
(0
0
0-
N
a
N
r-(
N
(
(
0
0
-M
(
0(
)(
"-000
"a
0cC
0
N
)
00
0
( m( (
0
a)
0 Z~ 0
0C
a)
C'n
0
0
-
C
a)
(N~~~~c 0
'-(0 c
0
0,
(0
N0
0
al )
No o
r- NO
0
0
-
0
(N
a)
0
co
0
0
N
0
0(0000
- 'D0
(o
'C'
0
0
00
t
0
) 0
NO
c No
(00
(N
0
00
(0
0)
0
0) 0
O~~~~~~~~~~~~~~~~~~~~~~O
0
N
0
0 (
0
(N
0
0
(N
0
a)
-
-0
'
(
0
N
'--
0
(0 0'(N
0
0
(NO
0
N
(0
(N
C0
0
..
I
-
-
0
N
0'IT0
0
.
00
0
0
'-
(NO
(Nt0(0.(
-
00
c' o
0(0If,
N
0.
(00
0 (D
0
(0
'0
Ia) 0(0
00
E
I<
0
(
N
(N
0
0)
0
I
N
0
0
a)
0
O
NT
0
0
C
E
0o )
E
0~~~~~~~~~~~~~
~
~ ~
c
~
>,
27~~~~~~~~~~~-
0
0
0.
0.o
0.
_
-
aD
0
>
~
o
>>
>CC'
2
o
--~_
m
(I)~~~~~~~~~~~~~~~~~~~~(
2.. a C' 2 0
Q . '0
2
0
!CE.
E
a.~~~~~~~~~~~~~~~~I
a.
20
w
>. o~~~~~~~~~~~~~~~~0
>~ z.
2I-LT 7Wo
0
75
z<E 0=°C). °.=-E
6 a.
R
--
CY
5L
0
..
<
2b
In
2zl
~
~
2a
,
.>=
a)
2
0..
-
-
0:
0
-.
a.<a.
.
,z~
a.
0)
-0
C
0
~
(N
-
0
CO
O
0,z
a
0c
C)
(D 0
0IO
I-
a
0
r-o
-C
IN
N
C>C~~~2
(9
I
M
0
(
t.
0
t
00
a
0
0
Cl)~
9
0
0
No(0
(
C-J0
w
oL (
z
o
(-
~~CNN- ~~~i~~ (NJ
~ NL
N
Nj
c-~~~J
o
N
CN
0)
j
co
z4
-,i
(N
N
N
N
'-
N
-
N0
NN
cN
N
-N
C)
N
NN
N
N
N
N
,J
CN
q
cN
0
N
CC
N
4
~D
Cq CN N
-N
NO
Ci
0
N
N
O0
Nl
N
c ( O N
qN C
N
N
-4-CON
C0 N
"J
04
N
1-
9
'0:
N
N
N
4
E
10)
E
0)00)
CO N
CO
CC
N
'D
VNO) N
> JO
7
N
N
)
r-.
'--
-
-'
rto
4-NCLr
N
'q
Nr
N'
4--
i-
m
L
4-
N
N
E
NN
N
(NJ
N
l
N'- N
N
CoN
N
CN
4-N
N
N
-<-
04
NE
>
-
N
',N
NO
CC!
4)~~~~~L
N
N
N
- N
N
N
co
N
~-l ~
N-N
N
4N,
N
oN 4-6
4_
C-
I
6
66
'-N4-41,
NNOCONM0)
N
N
N
4 -'
- N
4~~~~~~-
- j NIf
N
R
-
N-
C(4 - N
-r
N
N
N
-
O
N
J
N
4N' N 4 - N
N 0)
D
4
,'
N
C
-
N
N
N'
'
4-
N
I-:
CC!
CC!
N
vN'J
N
O
l
-0)
N_
C
-
,
N
N
.
C ON
N"
N
N
N
-C
4N
mC O m-
J
N
N
0
N-
co
N
0)
CD
N
N
NlN
N
co
co0cN
-
JO
NN - -~- N CC!7-N
N N4-0)N
N
t
( N~~~~~~~~~~~~~~~~~~~~UCOr
C
C
4-i
N
( N ,I N
4-N 4
N- N
(
N
N
N
aN-N It
C
t4N
-,-
OCO
~- 4-
N
r-N
0)-
'-
-
4-N t~
N-
4-
N
4--N
N
Nl - C co
NC m-4
I-
N'
N
D
)
-0)
N
OR N-N
4N
N
N CN N
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~2,
N
C -coN
(44N
N
N
~N 4-NN
N
- NN N 4 - N
4 ~~~
0N NN4 .N~~N
N N(~~~
~~~'-
0
co-
Nl
N:
m
IQN
N
CD0c
O
0
N
N
0
-
-
<-
7E
O0_ 0
N
N
(
C
N
CON (NjJ
l
Nc
N
N~ NO N
-
--
-
-
-l
4-
O
Jl
N
-
C4C
N
JO
N
IN
N
CO
to-'N
CO
4
C
CO6
CO
J
0)
N
N
Nl -Nt N
0), N
N
N
JO- CD
N
N
C
N
co0
0
N
It
r-N-NO
0
0
0
-
CD
NN
C-
'"I
'C!
E
N
NCON
'-N N ' N N U(NJC
4-N
N-
~~
NN
N~
~~~ ~~N-
x
LN
4-N4-N N
~
4-JON
mN. N
N
4.4
0
'4
~ ~ ~ ~
-
C) r75)~~~~
E
-
(-,~
cl~
.o
,M
`l
~
N.
L9
,
'N
I'll
Ca
co
'
C~
C~=
>
0
E
.o_
Co
4o
t
0
CP
>
I¢l)
-4 )
' 4 E) 4 )44 )) 4 )
-o
u
~N
CDC
0
C)
'
0C
2
-
02
(
N
D
;
4
~~~~~~~~~
Ca0
Ca
N
Na
)
C >-
N
0
.
15
M
c.')~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~I
C
i~C
N
a
)C
C
,-~
C 0 0C
a>)
ZE
o~~~0
L)
o
Z0
CC!
a ONa
a,
4-CON-N
a
~< a, ~~~~~~~
C)
Z~~~~~~~~~~
H
N-H
H
N
0
~> H~N 0 ~~HH NH ~~~~0
0~
~ ~ ~~~~>
0
E
Q,
P
>-
.-
o
=-
Na
0 <~CC <4)
:
U)
a)
0
C<<
a0
a
?
1u
0x
~~
2
LO
(D S Ca~~~~
o
N
N
2N
0
-J
C)~~~~
-N
N CO
0
0
~~~~~~~~~~~~~~~~~~N
C
0
N
,u
CO
Z4)4
N
c
N m
M
It
Q~~~~OQ~~~ (Z(~~~~~
OC~~~~~~
0
0
W
I
0
o-
Q
)
(4
-J
.J
0
N
Ca:
Ca
0
o
a,
<
N
(44-JO
N
N(.JJQ~~~~~~~~~
r 0•C ~~Mcof')
~~~~~~~~
..I
Z ( 9t o
N
0 )0
>ICC-)~
Ca >-
Q~~~~~~~~~~~~~U
O6(.)
N04
N
4
CDE
0~~~~~~t' 0
o
.D
a,
a,
0
C
4~~~~~~~~
a)
0)
0
)
Co
I-
O
JO
'0
V
Ca C
,
0
a
O
--
N-.
-
0
<a
m
0
c
.0
a~
0
cN
a
c
0
CD0-
Q
aD
N
0I
C
C
CO
0
0
OUM091M 6
)
a)
1
N.
C5
a'
to
.
0
'
0
N.
C)
-
u
)0
C
C
C)
C')
C)
C
co
C')
b.
o
~~6oo6 6 o0 6 oO
~~~~~ ~9~~ ~~~~~~~~
o o 9t ;I 5 L0uJ u
0
Ol16WlM/O
4
N
OS
6
AVtV
WA
u
'
N
N
p c
DO
I_ O)
i
o
)
b,,.
O
)o
^V
OS¢1§w
X
N-)OiD(
) L
0Cl6W AV
N
o
O 16w AV
c~o
0
O
C'
tO
C-4
O
N)- q(
CA
f6
OZ JOW AV
4-a.cnL
C.)
No O
0
0)
N
I.
T(N IT)
t
to
N
_
O
.
tO
N )
-
to
N
o
to
N.
C' -
U') 0o
(D
(N
0
NN)
to
.
0916w 6V
t
-
O
N
I)
0
N
a~~~'to
0
10
O O O
0) O~~~~~~~~1
0
t
I_
o
ooo
t
t
O
)
)t
o;
t
=
'0.
Ol/091M
w
O
C4..t
UUcf
( 4C4
O4- "Or
0
9J,
AV
N
N
m.
tt
. Nom
.
O t.L MA V
o')
(No!22o
Z
0)
g.
(
o
L'> 'I0
O~1W AV
Io t
L~~o
N
^
6w 6
ox~~~~~~~~~~~~~~~~~
No
to
o
036w 6AV
t
,I:
" o
u
s
.
t
o
O) (N
-0o
o tO
0,)
O
s
0
)
10 lo o
f
O
17
O
oN
C '. ) 0
n
'IT
O
V:
(X)t
0
a,
N 1:<to
>
.C C')
0) N
o> 0)
o> to
') N.
0)
00
0
0
6t
CD
o @ss
w X o
'D
to
09LM AV
C')
~~~co
AV
OgLM\
C.-
AV
o
09
0a
4s-s
'
Ao
0
l o
w
o
o
OM
',.
OZ~
6AV
011MAV
O.LM AV
NC')
0O S^ lt
c,
N (N
-
0)
'T
0tN.1to
.) o
O
V:
0)
o
o
"'
O.LMAV ^
Cu
*4
0
40
0
C 0)O
N.
r--
L LMAVN
M
I'
o
') t
C)
D
6 a) t ot o
o tL.
NCo
)
(N
(aF
(
0_
LM^AV
g
nu2 N
O uN f
lt 1
(N
utLMA^t
to
o
r
_)
0
~~~~
~
Oa)(0 t
l
'6
~~~
j ol o
)
t.
o
)
n
N. too
N.
CO
Aj!A!!SUGS
(0
C.
o.
N
C)
to
O
C')
"o
r.
0
o
v
co
Co
~.-)
*C~~
0Z^t 1M
N
O)
C4
k'¢
-
AI!A!I!SUeS
a
..C..)
IC$
4.)
44
Co
c)
C'
2:
so
N.
N
~~~
o
')
O) I_
N
.
.
a)
a)
C
O
)
o0
_2_
O
N
, .
.(0
O
.
. a)D
1X)
E
c
_
X
f -)
N stON
0
a )C O )
0
>CO.
OIMAo
a)
)
N
co
..
C.)
0
_
()
`
ri~
!o
j ffis ^N s O
)N
:
C'-
_
_
l
_
~,~~~~~~~~~~~~~~~~a
_
uoplunj
N
Nc
a)
.
o
__
1D
e
a
o
E
a
u
N
O
N
U4
)
O
0
-'-.
,_
<o
:
0 -
-
eue
J
o
a
L
o )a_ 8
s
=
Z . Zz
-
_
=2)
t
0o
4o
2e
0
N N
~~~~~-j
No
l
_
'E~a
e
to
I.)
Cu
Cv
.
z
bu0
~
to
N/)
Y
0_______^______
N.
sjJ
(
0
,o i-
O (
CI QQQ
-l
)
I-
cC
-
a
r
.
0
-J
L
-
__________
41)
o
C
V CI
C''
d
©
(0
(N
0
.2-E
146 WI9iru
aN
0916W BAV
N0
,
0)
(0
0)
0
C)
0
C>
01
r's
u
16W
} 6
0 w£
V
O-PILU
osC
-o
5
s:
~~
AV
JB
0Z
)B6U
100
(N(0
(N
(0'(
.
OL BW AV
0~~ ~ ~~
0;
j
uoi:loun..I ·
6W
(0
_
a,m1
c
010
0
a
(0
6
6
-
N
_O
N
:n
-
N-
(0
0_
(N
O
0
(
(N
0)
0
(N
-
(0
(0
(0
(0
0
-
-
(0
D
O
(0
)
(N
_
O
(0
-
N-
(
(0
-
-I
0
C
(
(0
N0
(
-
0 (
(1
00
N0
(
N
IN(0
(
(0
O
O0
9
( 0
(
0
0
(N
66O
O
N
I0
6
(-
-
N-C
'IT
(N
-
(0
0 0)`
D
(0
(
0
-< (
(0N
( 0
N
-
(0
0
9N
_
(
0
-
0
NI
0)
N0
(
6
.
(0"I0
to
(
)
6
C
I~
-:
C,
0~~~~~~~~~~~~~~~~~~~~~~c
(0
(0
(
N
-
c~
(0
(0
o
A.
_
OO
_
C)d0.o
0 (
CE
(0
-C-
M (0
.C
00)
O.C
C
C)
C)
c:
0
a.~
_
_
N
N
0)
0
N
(0
)0
(N
0o
O
,
(
0
_
(
0
E)
=
0
*
C)
0.
O
o
o
_
.~~~~~I0
C
(0
c
N
cc
0
°
-
I 2
~~~~~~~~~~~~Q2C
E~~
~~~~~
E
9
~~~~~~~~ Z2
~~~~~~~~~~~~>1
c-,~
1(0.-o')
4.
(0
0
N
(
(N
)
(0
0
_
o_
i~~~~~~~~~~
Co
5
-
1(0
2~
4.4
ooo
(0
0~
0
UOIP~~~~~~flI
O C
0
o
,
4
.66
i6
c, 10
~ CD~6 Nl
0~:~§w~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~v~~
~o2
~ ~ ~~~~~~~~~~~~~~0
__.___
1-<
CO
N-
'1
o!a
.......
._s
a
N
0
(
O16W BAV
J
'C3
O
0
0 (
'IT
(0
N
(0(0
(0
0
0)
CD
cN o
-
(
(0
C,
(DON
N
O)
(N
CO
CO
Iq
-
AV
0
0=
0
*
X
OCs,
*X)
OC
CI=
}~~~~~-
3 0
OWCIB
AV
m
~~~N
o
At
( 0 ( 0(N 0 ( 0
'I0
1
10
0
0
(N
(N
(0
0
N
0
o
- N
Q:
(0
(0
(
0.
C
0.
CC
C
(0
mZ0
C
0
.--
(D
~~~~~~~~~~~~~
x,
~~
C_
C
0.
m
C~
CC~~~~~~~~~~~~
C)~~~~~~~~~~~~~~~~<
.=l
0
0.
.3~~~~~~~~~~E>Of 0.
.
)
0
C 0 )
a1
NO
0-
t:
0(~
E
)
CC
)
(0Q
o
._
4.s1u
P 0
e3'N
Z <co
m~~
C
co co
EL
>
<C:'C:
U
_
euao
0
U
6
0
0j>
>
>(0
0(0(0(0
.
Zma.
(
I
Q"~~~~~~~~~~~
I
C)
f,
L
10
>.
I
Cl
C)
>.
(.
C)
(-)C
M
C.
C
Z
0
_(J 0
~i.
0
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~Z
>- )CO
LL
0
~
C
C
coC
)
IN
_)
O
6 O
O
6C
0BW/094W
0
o
") 0
OS
}6^tt s ccN Ot0N
0}6Wt
t~u^
-
0
C' N C:
C
NCG
(
O
C
O
0
4BW A
'U'C)
0
.
Co CU
U)
C
N
CU
C'
q
)
ot
CU
OC
JBW AV
OZ
JBW
Av
0
40W
A
~~~0 ,
C')
C')
O
U)
'C)
C)
O)
U)
U)
I
"I1
0
0
0N NU
C0N U)
U)
NU)
0
C
D0
~ oN ~
_C)
O
~
o~
e ~ ~ ¢
"'t
i
C
C"l
n
C'
0
)
)
Cu
CU
C)
C
CU
xt
0
N
C
U)
ID
CU
U)
C
C"
U
CCCOC'
CC
C')
CU
I
U)
C
U
I
U'
I
U)
N
C')
U
U)
CO
U
oO
d
Ul
"I
C'
'
0)-N
U
U
0)
U
0
0
C
O
U
u'
C'
c'-
NU)
U)
U)
w
0)9
)
Ui'1 U)
C'
9
C'
C'- C'-C')
CU
U)
4
CU
CU
U
C
:o-
CU
s
r_
_'
C0t'
U)
O
C'
U)
-
1.
0)
C)
N'
C
0o
U C
C
N 1~ 0U'°
U)
d
.
.
U
O~~~~6W6AV
~~~~~~~~~~~~~~~O
0
C U) '
't
C
U)
U
"Ck
W A'O
Oz~~~U
I
N4 C'
6 N
O
0
')
)
, 4
O~~~~C
:-
: UN
N
O
dC
CO0
O
C)
0
~ ¢ 2, i ~ ,~ ~ ~
09
C
o
0
G
O
_
0
6
CI
D
'U
It
/O91§
l}
O
-
C
C
o C d
d
C)
0
)
0
o
11
C
U)
U
0
D0
C)l
N
U
0,
N'
D I_
N
Nl
"
~
U)
0
0
U
IU
U
U)
U)
0)
G
"
0)
U)
U1
0)
')
-"1
_)
0
O
)
U
U)
snG
C'-
0)
°
U)
U
C
C'U)
U
C'
C'
)U
C
C
N
0
U
''~
Cl
)
.
0)
r
0'T
C'
)
U)
r
-
C-0 1
-
I
I
I
U '
C'
.
.
-
"-
I
U'N
N
'CU
-:
C'
'U
0
04
C'
N
U'
i
U)
0)I
U)
0
CO'CU
6
C_
s Ov10
CoC
U
I)
'IO
0
C
-
.
I
o
I
I
f)
0
'4
-
I
.
-- i
o oo
o
o0
C')
U
U
CN 1
U)
U
)
-D
C
CO
CU
C ) ')
UZ)
U)
o
0)
CU CO
N
r-C- 0
I'
U)
C')
-
I
I
_
*
o
o
E
z
uoilounj
C
C')
w
~
~~~~~~~~~~~~~~~~~~~U'UCU0)m,,
COCU C-U®
.
- -o
ooo ~
AIi!'!!SUeS
A1!A!IJsueS
0
)
t
CUe
U'
G)
B
-s
') U)
o
s
C'.
U
0
O. C
cn
a,
"I _
0
-m
C
C'
U)
'
0)
ID
9
'
0
)CU
-t
C, 0 t ) a,
U
O
O
00
0 0r l _ G
ID
0
O
c0
N 0 O
U)
N
<
ID
_
2~~~~~~~~~~~~~~~~'
C
>
Z
o
)
>C
"- '
o ,
o
>uo
a.
0C
-
'5
C
C
0.
-
<U
)
VU
~
C)
0
.,o 0.
0
co
..
C
cC
01
a. -
0)
.
~~~~~~~>
>)oo.cr.
'
C)
CU
C)
CU C')
C
C
0)
a- U)
z
U
E
C
.
"
C
0
C 0
00
0
'C
C
U)
>
~.
CU
0
CU
C
0
U
.:
CID
0
C
~z
.
E
f,~~~~~ ~
U
.
COC
0
C
C
Yo
0
C
>
.
E
0
>-
C)
0_
.
CU
0
0~~~~~~~~~)
C -aCCC
z
C
C)
C
C/Cm )2 2C 2 2
U
0
-N
0.'2
20 C '2
C
0.
0.
0.
0)~~~~~~~~~~~~'.
c
',,
'*-L2
L
'
0-
0)
C
ID ID s3.Z
)'
-~~ ~-~;~ .2
, ~.
C
'-
C:~~~~~~~~~~~~_
o
c0C
)
0
0
CCO
CU)
C
C
.
,,. ,, z
C)
z
.<I
00.
C)CUZO
3.
'
-
>-
(L)0
CU)
,
z
"'I O
z-L-
>-->
°1
z
z
n
z
. ,-C_>
.>
>_>
-,
>9.
0
OU)
CUf
OL16W/O9W0
L
lD
01,16w/91,6w
0916W 6AV
~l lol~6
o
~ lTo o
_
o-7
o
U^! t
N
t
,
o
4
0
a C ID
O,
6
09 1 W AV
X~,
'
U
D
c O
o
~(Nai
to
1to
-¢)
OusC
(N
(O
tof
o
t
o
o t
C
O
-
o
COtC
Ct
-I_
(
N
._
.
(
0
N
0
d
It. t
s
Qto to
tO
©
OoooooiAV
0Z 6W AV
N-
t
mo O
m
4
._
__.,
_. t
oi
N-no.-
I
Lo
0
z_
(N
_ _
-
la
*o o
uu
>1
uoijo)unj
5
E
o
o
o
Ia
I
o
11;
o
..
uu
a
CA
W
,is
0=
0::1
az
T
0-
P
0
20'0
U) z
-
O
aa
4-4
Or
w
WI
C4
1*
~
--
I0
0
7j)
0
u
c~
ct
*cd
o
~0
Q
0
OUM091m
.o
0
(0
C-
-
0
AV
,.A
5~~~~~~~~~
c-I
0SMAV
c)
(00
)
09 (
C)
C
M
A
_I
.
O
~
(2
NN-O
(0
o
o
W
(
oi
0
I
o
o
(0
1
0
0
16
~
0
O)
0
0
N 6
(N
0t
C)
0
6
(
(
a
N0
0
0
Nf-,~~~~~~~~~~~1
a; O16
MA
0
6
-~~~~~~~~~~~~~~~~~~~~~~~"
-I
lo
U:1t
v~~~~~~~~~~~~~~~~~~~~I
OS.LM^V
0
oi
CD
C)
O
0
W
o
0c
(0
(
(
i
4
S
1
1t
o
(O
NN
r-
((
lp
o
(0
-
(N
I
n
(0
N
CF>
(0
1t
Vw
6
6
(
0(
N
N
0
C)
c
oo
N
(0
0
N-s
(0 c2
0C
N(0
cj
4
N
4:C) 4~~~~
_ _
-
--
_
.,..4~~~~~~~~
~ ~ ~ ~~ ~ ~
~~~~~~~ct
_
OJ MAV
r-
i~~~~~~~~~~it
°n
C)
*
.o
'±IVVt
Q d%
0U
uo p o
)
N--
O
1:0
_
c
(0
It
(N
-n0
"
c),
co
_
)
-0
Cu4
t_-
(0
~
(0
cl9
N
U4(0
)(1(0
~2
(0
0
0)
.
o
(0
llD
-J
1-
'V
q
C~
:
(,
aN
Iq
C¢)
rl-
C.,
0 )OO
N0
ol
N)
N
-I
-(
0)
'O
ae
MtOt
0e
C,
0(,0
(0
N)
0)
0
(0
1O~
(N
O~
-
0
ll
c
) C
-
o0
!2
C->;
C.) o (0
co
C) o
CJ
C'
2
V
O_
c
N
(0
(0
C)
R
N-
(0
(-O0)
OG.LM^V
AV
cJ
O,.LM AV
S
O.LM AV
N
CN
I
7
0D
0
4)
1
co
.
(0
6
C)O
N V::J
N
- 0)
C
0~ 100(0(00e
C
o
(0t(N
.C
-m M OZ
AV
(0
CD
O0.LMO9.M
(0
N0
)
,,< 6
0
O O
I)
(0
"C
O
6O
O~
0
Cu
(4
OS.LM
^ AV
C)
0
Cu
-o
0
OC)
0I
O
't
V
0N-
(N
(0
(
( N(0
O
0 O
Cu
I.
V
0
(N
O
00)>
N-
(0
0
0
C _0
(0
C
-
N- (0
s
(N N-
N
0
W5
(h
Bo o
c2
-- 5 o _B3NCBNN
~~~~~~~~~~~~E
2oo2~
io 'aSU
uoq~~~~~~~un~C"-I
Ev
a,
N-~~~~
(0
C
·
C)
m
C
cd~VC
0
0
0
c
!)
O
¢l
X
C
S
~ ~ l~
-
C~kI
O 2
I
Z
llJ
~~
z
V
V
0
_
2
AK4!A4!SUGS
UO! .flA ^V
c DI
-2
2
0X
U
T
eueo
C
.
-
O00O(V
00
0~~~~
~
~
'a 0 C: a)
a.
0C
U
o1.oo
C)
C%
.0
0
C,)~~~C
cO O l
Q2(w2 I
C
Cu
U14
rJ
Cd
*
w
~-
0
I-~t
O~
_
._
C%
x9
_e
Cu
C-
~z 0~ ~0 ~
o
C-)
~ D)C)OO..c
CZn,
(1)
'
-,,bJ
C
x
0
C
LL0
0
fl)
(V
(
,
,
-D-..~-Z
:
(V
C
~~~~~~~~~~~~
C
w
Ed
(V
(C,
-o-,-
Cu~~~~~~~~~~~~~
H
,'
.
0~....
>-
o
-
Appendix: Gene Expression Analysis Protocols
1
Total RNA synthesis
i-
.
,
,.
.
,
.I
,
i
RNeasy Mini Protocol for Isolation of Total RNA from Yeast
Enzymatic Lysis Protocol - standard version
Use an appropriate number of yeast cells
Important notes before starting
* For RNA isolation from yeast, cells should be harvested in log-phase growth. Use only
freshly harvested cells for the enzymatic lysis protocol.
* Prepare Buffer Y1
Buffer Y1
1 M sorbitol
0.1 M EDTA, pHl 7.4
Just before use, add:
0.1% -mercaptoethanol
50 U lyticase/zymolase per 1 x 107 cells
Depending on the yeast strain and enzyme used, the incubation time, enzyme concentration,
and composition of Buffer Y1 may vary. Please adhere to the guidelines of the enzyme
supplier.
Important
*· -Mercaptoethanol (-ME) must be added to Buffer RLT before use. -ME is toxic;
dispense in a fume hood and wear appropriate protective clothing. Add 10 ,l I -ME per
1 ml Buffer RLT. Buffer RLT is stable for 1 month after addition of -ME.
* Buffer RPE is supplied as a concentrate. Before using for the first time, add 4
volumes of ethanol (96-100%), as indicated on the bottle, to obtain a working solution.
* Buffer RLT may form a precipitate upon storage. If necessary, redissolve by warming, and
then place at room temperature. Buffer RLT and Buffer RW'1 contain a guanidine salt and are
therefore not compatible with disinfecting reagents containing bleach. Guanidine is an irritant.
Take appropriate safety measures and wear gloves when handling.
* After enzymatic lysis, all steps of the RNeasy protocol should be performed at room
temperature. During the procedure, work quickly.
* After harvesting the cells, all centrifugation steps should be performed at 20-25 0 C in a
standard microcentrifuge. Ensure that the centrifuge does not cool below 20C.
1. Harvest yeast cells in a 12 ml or 15 ml centrifuge tube by centrifuging at 1000 x gfor 5
min at 4°C. (Do not use more than 5 x 107 yeast cells.) Decant supernatant, and
carefully remove remaining media by aspiration. After centrifuging, heat the centrifuge
to 20-250 C if the same centrifuge is to be used in the following centrifugation steps of
the protocol.
Incomplete removal of the supernatant will affect digestion of the cell wall in step 2.
Note: Freshly harvested cells must be used.
2. Resuspend cells in 2 ml freshly prepared Buffer Y1 containing lyticase or zymolase.
Incubate for 10-30 min at 30 0 C with gentle shaking to generate spheroplasts.
Spheroplasts must be handled gently.
Depending on the yeast strain used, the incubation time, amount of enzyme and composition
of Buffer Y1 may vary. For best results, follow the guidelines of lyticase/zymolase supplier.
--II--·-----·II
*r*eCIP--i-·-------·--i
--
11"-
------------------------
Appendix: Gene ExpressionAnalysis Protocols
2
Complete spheroplasting is essential for efficient lysis. Note: Freshly harvested cells must be
used for preparation of spheroplasts.
3. Centrifuge for 5 min at 300 x g to pellet spheroplasts. Carefully remove and discard
the supernatant.
Note: Incomplete removal of the supernatant will inhibit lysis and dilute the lysate,
affecting the conditions for binding of RNA to the RNeasy silica-gel membrane. Both
effects may reduce RNA yield.
4. Add 350 jil Buffer RLT to lyse spheroplasts, and vortex vigorously. If insoluble
material is visible, centrifuge for 2 min in a microcentrifuge at maximum speed, and
use only the supernatant in subsequent steps.
Note: Ensure that 3-ME is added to Buffer RLT before use (see "Important notes before
starting").
5. Add 1 volume (usually 350 Al) of 70% ethanol to the homogenized lysate, and mix
thoroughly by pipetting. Do not centrifuge. Continue immediately with step 6.
A precipitate may form after the addition of ethanol, but this will not affect the RNeasy
procedure.
6. Apply the sample (usually 700 Atany precipitate that may have formed, to an RNeasy
mini column placed in a 2 ml collection tube (supplied). Close the tube gently, and
centrifuge for 15 s at 8000 x g (10,000 rpm). Discard the flowthrough.
Reuse the collection tube in step 7.
If the volume exceeds 700 , load aliquots successively onto the RNeasy column, and
centrifuge as above. Discard the flow-through after each centrifugation step.*
7. Add 700 jl Buffer RW1 to the RNeasy column. Close the tube gently, and centrifuge
for 15 s at -1 8000 x g ( 10,000 rpm) to wash the column. Discard the flow-through and
collection tube.*
8. Transfer the RNeasy column into a new 2 ml collection tube (supplied). Pipet 500 Al
Buffer RPE onto the RNeasy column. Close the tube gently, and centrifuge for 15 s at
L_ 8000 x g (Ci 10,000 rpm) to wash the column. Discard the flow-through.
Reuse the collection tube in step 9.
Note: Buffer RPE is supplied as a concentrate. Ensure that ethanol is added to Buffer RPE
before use
9. Add another 500 pl Buffer RPE to the RNeasy column. Close the tube gently, and
centrifuge for 2 min at 1 8000 x g ( 10,000 rpm) to dry the RNeasy silica-gel
membrane. Continue directly with step 10, or, to eliminate any chance of possible
Buffer RPE carryover, continue first with step 9a.
It is important to dry the RNeasy silica-gel membrane since residual ethanol may interfere with
downstream reactions. This centrifugation ensures that no ethanol is carried over during
elution.
Note: Following the centrifugation, remove the RNeasy mini column from the collection tube
carefully so the column does not contact the flow-through as this will result in carryover of
ethanol.
9a. Optional: Place the RNeasy column in a new 2 ml collection tube (not supplied),
and discard the old collection tube with the flow-through. Centrifuge in a
microcentrifuge at full speed for 1 min.
3
Appendix: Gene Expression Analysis Protocols
* Flow-through contains Buzfer RLT or Buzr RWV and is therefore not compatible with bleach.
10. To elute, transfer the RNeasy column to a new 1.5 ml collection tube (supplied).
Pipet 30-50 pl RNase-free water directly onto the RNeasy silica-gel membrane. Close
the tube gently, and centrifuge for 1 min at - 8000 x g ( 10,000 rpm) to elute.
11. If the expected RNA yield is >30 jig, repeat the elution step (step 10) as described
with a second volume of RNase-free water. Elute into the same collection tube.
To obtain a higher total RNA concentration, this second elution step may be performed by
using the first eluate (from step 10). The yield vwill be 15-30%/o less than the yield obtained
using a second volume of RNase-free water, but the final concentration will be higher.
1 st
Strand Synthesis
Ill
~~~~~~~~~~ ~ ~ ~ ~ ~
Amount of RNA
* Using a 1.5 ml centrifuge tube, mix reagents according to the following Table.
STEP 1
Reagent
Volume (l)
DEPC-water
RNA (1.0 tg/pl)
T7 Primer (100 pmol/bl)
x
Y
I
I
TOTAL
*
*
*
12
Incubate @ 70'C in a water bath for 10 minutes
Spin briefly and place on ice till ready to proceed.
When incubation is done, to the same tube, add reagents according to the Table below.
STEP 2
Reagent
*
*
*
2 nd
Volume (l)
Tube from step 1
5X 1 strand cDNA buffer
0.1 DTT
10 mMNdNTP mix
12
4
2
1
TOTAL
19
Mix well with a pipette (carefully, in and out) and incubate at 42 C, 2 minutes.
Add variable amount of SS Reverse Transcriptase (final reaction volume =20 pl)
Mix well with a pipette and incubate at 42 'C, 1 hour.
Strand Synthesis
Add the following components to the 1 t strand tube from above:
2nd STRAND
Reagent
1t Strand Reaction
DEPC-Water
5X 2 ; strand cDNA buffer
10 mmM dATP, dCTP, dGTP, dTP
DNA Ligase (10 U/,u)
DNA Polymerase (10 /,ul)
RNase H (2 U/d )1
TOTAL
_~
...~
__
~.--
--
~
1~~_
IIXTURE
Volume (l)
20
91
30
3
1
4
1
150
Appendix: Gene Expression Analysis Protocols
*
*
*
*
Transfer this total amount to a PCR tube and place in the thermocycler at 16°C for 2 hours.
Add 2 ud of T4 DNA Polymerase (10 U/[dl) and incubate at 16°C for 5 minutes.
Add 10 [tl of 0.5 I EDTA. (The total added volume is -162 tu).
Proceed to cDNA cleanup or store the reaction at -20o'C.
i-
5
Appendix: Gene Expression Analysis Protocols
cDNA Clean-up
*
·
·
Pellet the PLG light in the green tube and set aside.
Transfer the 2nd strand cDNA solution from previous step back to a 1.5 ml tube.
Add to the 2nd strand cDNA tube (equal volume, assuming 150 dl recovery):
The total volume is now approximately 300 l
·
*
*
Vortex the 2d strand cDNA tube and transfer the entire amount to the pelleted PLG-light
tube.
Centrifuge at 14,000 rpm for 2 minutes. Transfer the aqueous (top) layer to a new tube.
Add the following to the aqueous layer (assuming recovery of 150 d):
Glycogen (5 mg/ml)
NH 4OAc (7.5 M)
EtOH (100 %)
*
*
*
*
1 1
75 d1
562 l
Vortex and immediately centrifuge for 20 minutes at 14,000 rpm, RT.
Remove supernatant and wash pellet twice with 500 tl of 80 % EtOH.
After the last wash, centrifuge and eliminate EtOlH using a micropipettor. Centrifuge again
and eliminate EtOH remnants with micropipettor.
Air dry the pellet (5 min at 370C and 5 min RT) and resuspend in 12 dul
DEPC water
ote: Comlete drying is rve
imor/ant. Te resence of ethanolwillinhibit the IT/T reacio4
Biotin Labeling by In Vitro Transcription Reaction (IVT)
I
I
IIl
I
,
Ill;I
,I
.
Since we started with 5-8 4tg of total RNA, the Affymetrix manual recommends using 10 l out of the
12 tl cDNA solution to setup a 40 pl IVT reaction (see chart in the manual). Using reagents from the
ENZO kit, add the following components to a 1.5 ml centrifuge tube.
IVT Reaction Setup
Reagent
cDNA template
DEPC-Water
10X HY reaction buffer
10 biotin labeled NTP
10X DTT'
10X RNase inhibitor
20X T7 RNA polymerase
TOTAL
9-
I-------
Volume (i)
10
12
4
4
4
4
2
40
Appendix. Gene Expression Analysis Protocols
*
*
*
*
6
Mix the reagents and centrifuge briefly
Incubate @ 37oC for 5 hours in an oven to avoid evaporation.
Make sure to mix with a pipette every 30-45 minutes.
Store @ -20oC or go to the clean up step
-
I
Appendix: Gene Expression Analysis Protocols
IVT Clean-up
RNeasy RNA Cleanup Procedure
Notes:
-B&ffr RLT ma),form a prediitate cpon storage. If this happens, varm to to redissobe.
-Add 1O/l oJf -IME to 1 ml of Bffer RIT beJire se (stabklfor I month).
-Baffer RPE is protided as a concentrate. Before usingfor the first time, add 4 aolumes of ethanol (96-100%) as indicated on the bottle to obtain a
working solution.
-All entrifiugationsare done at room temperature.
Work in groups of two: each group will process halfof the sample for RNA cleanup
1.
Measure the volume that you now have in your IVT tube (should be -40 jtl).
2.
Split the sample in half and continue with the cleanup procedure on half of the sample. Adjust the
volume of your portion (-20 41) to 100 tli using RNase-free water.
3.
Add 350 dtl
Buffer RIT to the sample, and mix thoroughly.
Note: Ensure that _ -IE is addedto B/ffer RLT before ase.
4.
Add 250[d ethanol (96-100%) to the sample, and mix well by pipetting. Do not centrifuge.
5.
Apply sample (now 700 Ld)to an RNeasy mini spin column sitting in a collection tube. Centrifuge for
15 seconds at 12,000 rpm.
6.
Discard flow-through and collection tube.
7.
Transfer RNeasy column into a new 2-ml collection tube. Pipet 5 00 p1 of wash Buffer RPE onto the
RNeasy column, and centrifuge for 15 sec at 12,000 rpm to wash.
Ensare that ethanol is added to Baffer RPE before zese.
8.
Discard flow-through and reuse the collection tube in the following step.
9.
Pipet 500 ll of wash Buffer RPE onto the RNeasy column. Centrifuge for 2 min at 14,000 rpm to dry
the RNeasy membrane.
10. Following the spin, remove the RNeasy column from the collection tube carefully so that the column
does not contact the flow-through as this will result in carryover of ethanol.
It is important to d')the RNea' membrane since residelethanol maj intefere ith stbeqeent reacrions. Thefolloig spin ensures that no ethanol is
carried over daurig elation.
11. Place the RNeasy spin column in a new 1.5-ml collection tube, and discard the old collection tube with
the filtrate. Centrifuge at 14,000 rpm for 1 min.
12. Transfer the RNeasy column into a new 1.5-ml collection tube, and pipet 30 d of RNase-free water
directly onto the RNeasy membrane. Centrifuge for 1 min at 12,000 rpm to elute.
A second elution step can beperformed using another30-0 ,ulRt ase-fiee mater. This might improe yield.
of water and use this for a spec reading using the Biophotometer
13. Dilute 1 Ill of the reaction into 99 1d
(Eppendorf).
14. Store the rest at -20°C or proceed with the next step.
-I
8
Appendix: Gene Expression Analysis Protocols
Fragmentation
*
Use the spec reading to determine what volume will give you 20 ,ug for fragmentation.
Apply the convention that 1 OD at 260 nm equals 40gg/ mL R.NA.
1.
2.
3.
*
Check the OD at 260 nm and 280 nm to determine the sample concentration and purity.
Maintain the A26,/A2s) ratio close to 2.0 for pure RNA (ratios between 1.9 and 2.1 are acceptable).
In our test run, we obtained a post- IT cleanup concentration of 1.356 ag/Fl (32 l), for a total of 43.4
means that we need 14.7 1 of IYT cleaned up product for the Fragmentation reaction.
pg.
This
Set up the reaction. The volumes shown in the Table below are from our example run done before the
workshop. Your volumes might vary...
STEP 1
Reagent
Volume (l)
RNA (20 ug)
DEPC-Water
5X Fragmentation Buffer
TOTAL
*
*
14.7
17.3
8
40
Mix well, and incubate at 94)C in a water bath for 35 minutes.
Place on ice or at -20oC.
Adjusted cRNA calculation
*
Calculate the adjusted reaction yield from the Cleanup step of the JYT reaction. This is done using the
following formula:
Adjusted cRNA yield= RNA,,- (total RNA,)(y)
RNA = amount of cRtNA measured after IVT (gg)
Total RNA, = starting amount of total RINA (g)
Y = fraction of cDNA reaction used in ITT
In our example:
RNAm = 43.4[tg
Total RNA = 7 .0 g
Y = 10/12 (fraction of cleaned up cDNA used in the IVT reaction)
Therefore, our adjusted cRNA yield: 37.6 lag[i.e., 43.4 - 7(10/12)]
The volume was 30 sal(see under Fragmentation)
Hence the adjusted concentration previous Fragmentation is 1.25 plg/al
Since we used 14.7 ±1of the prefragmented cR NA in the Fragmentation step, the adjusted amount was
18.4 ig [i.e., 14.7 l X (1.25 g/ld)].
9
Appendix: Gene Expression Analysis Protocols
Prepare the Target Hybridization Reaction
The Affymetrix manual recommends using 15 tg of sample in the target hybridization. Use the adjusted
concentration to calculate the volume of sample needed for 15 4g.
=
Amount of adjusted cRNA used in the Fragmentation reaction 18.4 Itg
The volume of the Fragmentation reaction was 40 I
In order to get 15 atg from the Fragmentation reaction we need: 15 jag X (40 .l/18.4 jag)=32.6 1al
Note: One group of two willperform the Test3 Array Hybridizaton, the correspondingpartnergroup will
perform the Species Array Hybridization.
Step 1: Test3Array Hybridization
*
*
*
*
*
Set the Test3 chip at RT before setting up the reaction.
Prepare 85 il of 1X HIybridization buffer. Add 80 1tl of this solution to the Test3 Array Chip (Mini
format) to wet it.
Put it in the oven at 450C for at least 10 minutes at 40-50 rpm.
Ieat the Eukaryotic H[-ybridization Controls to 65oC for 5 minutes prior to adding it to the reaction
below.
Set up the Target hybridization reaction according to the following Table:
Reagent
H-YBRIDIZATION COCKTAIL
Volume (l)
Fragmented cRNA (15 jtg)
Control oligonucleotide B2
20X Eukaryotic Hybridization Controls
32.6
5
15
Herring Sperm DNA (10 mg/ml )
*
3
Acetylated BSA (50 mg/ml)
2X Hyb. Buffer
Water
3
150
81.4
TOTAL
300
After setting up the reaction, remove 100 d1from the Hybridization cocktail solution (save the rest of
the solution at -20"C) and process it as follows:
HIeat at 99C for 5 minutes
Heat at 45"C for 5 minutes
Centrifuge at 14,000 rpm for 5 minute to clarify the cocktail
*
*
Remove the 1X Flyb solution from the 'Test3 Chip' and then add 80 d1 of the Hyb. Cocktail.
Incubate at 45'"C at 60 RPM for 16 hours.
Step 2: Species Array Hybridization
*
*
*
*
*
Set the Species Array chip at RT before setting up the reaction.
Prepare 210 ldof 1X Hybridization buffer. Add 200 J.dof this solution to the Species Array Chip
(Standard format) to wet it.
Put it in the oven at 45°C for at least 10 minutes at 40-50 rpm.
Feat the Eukaryotic Flybridization Controls to 650C for 5 min prior to adding it to the reaction below.
Set up the Target hybridization reaction according to the following Table:
-
I_.._..
10
Appendix: Gene Expression Analysis Protocols
Reagent
*
-nYBRIDIZATION COCKTAIL
Volume (l)
Fragmented cRNA (15 g)
Control oligonucleotide B2
20X Eukaryotic Hybridization Controls
Herring Sperm DNA (10 mg/ml)
Acetylated BSA (50 mg/ml)
2X Hyb. Buffer
Water
32.6
5
15
3
3
150
81.4
TOTAL
300
After setting up the reaction, remove 250 dtlfrom the 1-ylbridization cocktail solution (save the rest of
the solution at -200C) and process it as follows:
Heat at 99oC for 5 minutes
Heat at 45C for 5 minutes
Centrifuge at 14,000 rpm for 5 minute to clarify the cocktail
*
*
Remove the 1X I-lyb solution from the 'Species Array Chip' and then add 200 tdlof the E-yb Cocktail.
Incubate at 450C at 60 RPMI for 16 hours.
Washing, Staining and Scanning Probe Arrays
*
*
·
Shortly before the 16 hour incubation is done, follow the fluidics station setup according to
Chapter 4, Section 3.
For the Test3 Array, prepare the SAPE stain solution under 'Vashing and Staining Procedure
1: Single Stain' (Chapter 4, Section 4). Streptavidin Phycoerythrin (SAPE) stocks should be
stored in amber tubes at 40C. Remove SAPE stocks from refrigerator and mix well before
preparing stain solution. Do not freeze concentrated SAPE or diluted SAPE stain solution.
Always prepare the SAPE stain solution immediately before use.
For 600 jpL of SAPE Stain solution:
300 ,tL 2X Stain Buffer
270 uL water
24 4uL of 50 mg/mL acetylated BSA (final concentration of 2
mg/mL)
6 btL of 1 mg/mL streptavidin phycoerythrin (SAPE) (final
concentration of 10 [g/mL)
For the Species Array (Human U95A), prepare the SAPE stain solution and Antibody solution
under 'Washing and Staining Procedure 2: Antibody Amplification' (Chapter 4, Section 4).
For 1200 jgL SAPEStain solution:
600 ,uL of 2X Stain Buffer
540 L of water
48 uL of 50 mg/mL acetylated BSA (final concentration of 2
[tg/4tL)
12 gtL of 1 mg/mL SAPE (final concentration of 10 ptg/mL)
Mix well and divide into two aliquots of 600 L each to be used for stains 1 and 3
respectively.
For 660 jpl of Antibody Solution:
300 uL of 2X Stain Buffer
266.4 i~Lof water
___
11
Appendix: Gene Expression Analysis Protocols
24 tL of 50 mg/mL acetylated BSA (final concentration of 2
mg/mL)
6.0 4L of 10 mg/mL normal goat IgG (final concentration of 0.1
mg/ml,)
3.6 [tL of 0.5 mg/mL biotinylated antibody (final concentration of
3 ptg/mL)
*
After 16 hours, remove the cocktail and save it (-20°C). Do not let the chip dry out.
Immediately add 100 tl of the non-stringent wash buffer to the chip.
Plastic cartri¢
Probe array
glass substra
GeneChip ® Probe Array
-clC*es"··Y"-·------c---ra·-r^-------
Appendix
12
RNEasy Cleanup Procedure
Affymetrix recommends an additional cleanup step for the RNA sample before using in a microarray
experiment. This is done using the RNeasy reagents as follows (procedure was adapted from user's
instructions in the kit):
Notes: Do not exceed the RNTA binding capaciy (100 lg) of the RNeasy mini spin columns.
Buffer RLT mayform aprecipitateupon storage. If this happens, warm tup to redissolve.
Add 10 ul ofJ -ME to
ml of Blffer RLT before use (stablefor month).
Buffer RPE is provided as a concentrate. Before using for the first time, add 4 volumes of ethanol (9 6-100o%) as
indicated on the bottle to obtain a working solution.
All centrifugations are done at room temperature.
1.
Buffer RLT to the sample,
Adjust sample to a volume of 100 tl with RNase-free water, add 350 u1d
and mix thoroughly.
2.
Note: Ensure that 3 -ME is added to Buffer RLT before use.
3.
Add 350 pd Buffer RLT to the sample, and mix thoroughly.
4.
Note: Ensure that O -ME is added to Buffer RLT before use.
5.
Add 250~d ethanol (96-100%) to the lysate, and mix well by pipetting. Do not centrifuge.
6.
Apply sample (now 700 i) to an RNeasy mini spin column sitting in a collection tube. Centrifuge
for 15 seconds at 12,000 rpm.
7.
Discard flow-through and collection tube.
8.
Transfer RNeasy column into a new 2-mi collection tube. Pipet 5001 of wash Buffer RPE onto the
RNeasy column, and centrifuge for 15 sec at 12,000 rpm to wash.
9.
Ensure that ethanol is added to Buffer RPE before use.
10. Discard flow-through and reuse the collection tube in the following step.
11. Pipet 500 l of wash Buffer RPE onto the RNeasy column. Centrifuge for 2 min at 14,000 rpm to
dry the RNeasy membrane.
12. Following the spin, remove the RNeasy column from the collection tube carefully so that the column
does not contact the flow-through as this will result in carryover of ethanol.
13. It is important to dry the RNeasy membrane since residualethanol ma inteere with subsequent reactions. The
following spin ensures that no ethanolis carriedover during elution.
14. Place the RNeasy spin column in a new 2-ml collection tube, and discard the old collection tube with
the filtrate. Centrifuge at 14,000 rpm for 1 min.
15. Transfer the RNeasy column into a new 1.5-ml collection tube, and pipet 30 1lof RNase-free water
directly onto the RNeasy membrane. Centrifuge for 1 min at 12,000 rpm to elute.
16. A second elution step can be performed using another 30-50 l RNase-free rwater. This might improveyield.
17. Dilute 1 dtlof the reaction into 99 ,ul of water and use this for a spec reading using the
Biophotometer (Eppendorf).
Appendix
13
18. Store the rest at -80C or proceed with the next step.
2g. Spectrophotometric analysis
1. To determine the concentration and purity of the RNA solution, transfer 2 tl of your RNA solution
into an RNase-free tube containing 98 ,ul of DEPC-water. The lab instructors will measure the
A260/A28 0. Pure RNA will give a ratio of approximately 2.0.
1.0 A260= 40 ptg/ml RNA
Dilution factor in spectrophotometric cuvette= 50
RNA solution conc. (g/ml)=
"srr·-·-----r·rr-------
(A2 60 )( 100)(40ug/ml)
I