Luận án tiến sĩ: A New Genetic Pathway Mediating Multidrug Resistance in the Yeast Saccharomyces Cerevisiae

In the yeast Saccharomyces cerevisiae, one of the main mechanisms causing drug resistance is the loss of function of ABC transporters that are responsible for effluxing the drug from the

NOTE TO USERS

This reproduction is the best copy available

®

UMI

THE CATHOLIC UNIVERSITY OF AMERICA

A New Genetic Pathway Mediating Multidrug Resistance in

the Yeast Saccharomyces Cerevisiae

A DISSERTATION

Submitted to the Faculty of the Department of Biology School of Arts and Science

Of The Catholic University of America

In Partial Fulfillment of the Requirements

For the Degree Doctor of Philosophy

By Anne E Fleckenstein

Washington, D.C

2006

INFORMATION TO USERS

The quality of this reproduction is dependent upon the quality of the copy submitted Broken or indistinct print, colored or poor quality illustrations and photographs, print bleed-through, substandard margins, and improper alignment can adversely affect reproduction

In the unlikely event that the author did not send a complete manuscript

and there are missing pages, these will be noted Also, if unauthorized

copyright material had to be removed, a note will indicate the deletion

®

UMI

UMI Microform 3198179 Copyright 2006 by ProQuest Information and Learning Company

All rights reserved This microform edition is protected against unauthorized copying under Title 17, United States Code

ProQuest Information and Learning Company

300 North Zeeb Road P.O Box 1346

Ann Arbor, Ml 48106-1346

A New Genetic Pathway Mediating Multidrug Resistance in

the Yeast Saccharomyces cerevisiae

Anne E Fleckenstein

Director: John Golin, Ph.D

Multiple drug resistance is quickly becoming an obstacle to the treatment of disease Bacteria, parasitic protazoa, yeast and mammalian cancer cells develop mutations that render them resistant to a wide variety of structurally and chemically different compounds

In the yeast Saccharomyces cerevisiae, one of the main mechanisms causing drug resistance is the loss of function of ABC transporters that are responsible for effluxing the drug from the cell The main ABC transporter responsible for efflux of many different drugs is PdrSp However, earlier work (Fleckenstein et.al 1999, Shallom and Golin, 1996) shows that this is not the only pathway mediating

resistance to these drugs The global regulator Sin4p and the transcription factor YRR/ operate in a PDR5-independent pathway to confer drug resistance to the cell

While wild type YRR/ is not required for multidrug resistance, a gain-of- function mutation in this gene can restore resistance in a previously drug

hypersensitive Apdr5 mutant This resistance requires both Sin4p, a member of the RNA Polymerase Mediator complex and Snf5p, a component of the chromatin remodeling complex SWI/SNF

Disruption of these genes can cause profound drug hypersensitivity that cannot be explained by the changes in PDR3S transcription or function observed Furthermore,

accumulation that is seen in a ApdrS mutant The SNF5, SIN4 and YRRI genes define

a new, major pathway required for mediating multidrug resistance

This dissertation by Anne E Fleckenstein fulfills the dissertation requirement for the

doctoral degree in Cellular and Microbial Biology approved by John Golin, Ph.D., as Director, and by James Greene, Ph.D and Pamela Tuma, Ph.D as Readers

john G nm, Ph.D Director

lem ZL Pamela Tuma, Ph.D Reader

il

Introduction

Yeast as a Model System for Studying Multidrug Resistance

Drug Resistance in Yeast

Mechanism of Yeast Multidrug Resistance

RNA polymerase II holoenzyme and subcomplexes

Materials and Methods

Complementation testing Tetrad analysis

Minimum Inhibitory Concentration (MIC)

Measurement of rhodamine 6G efflux

Efflux Assay for [°H]-tritylimidazole

4 Use of the mTn3::LEU2 transposon to create mutants

5 Complementation test of RR mutants and Asin4 mutant

6 Complementation test of RR3 and AsnfS mutant

7 Minimum Inhibitory Concentration of isogenic single and double

Asin4 and AsnfS mutants

8 RT-PCR of SIN4 and ACT] transcripts in YRR/ mutants

9 RT-PCR of possible targets of SNF'5 in YRRI-2 and AsnfS mutants

10 Pdr5 transcript in AsnfS mutant

11 PDR5 transcript with varying concentrations of RNA in WT and Asnf5

strains

12A A representative histogram plot of rhodamine 6G accumulation in

AsnfS and Apdr5 mutants

12B A representative histogram plot of rhodamine 6G accumulation in

Asin4 and Apdr5 mutants

13 The accumulation of [*H] tritylimidazole does not saturate

14 The accumulation of tritylimidazole is not energy dependent

15 PDRS transcription in wild type and Aspr20 strains

16 A representative histogram plot of rhodamine 6G accumulation in

Aspt20 and ApdrS strains

Minimum Inhibitory Concentration of YRR/ mutants (uM)

Minimum Inhibitory Concentration of Asin¢Asnf5 Strains

RT-PCR of SIN4 in YRR/ mutants

Minimum Inhibitory Concentration of SC4741 background (uM)

Band Densities of Potential Targets of SnfŠp

Levels of PDRS promoter activation in Asnf5 Mutant and Wild Type

10 RT-PCR of PDRS in SNFS and Asnf5 Strains

11 Rhodamine 6G accumulation in SC mutants

12 H]-tritylimidazole Accumulation Assay

Acknowledgements

My success in graduate school has truly been a team effort; | could never have done this alone While there are too many people to list separately, 1 must thank the following

people:

My parents, James and Marilynn Fleckenstein for their unwavering support (both

emotional and financial),

Rosemary and Jim Carey for always providing a place to escape the lab,

Dr Joshua Shallom and Sr Susan Cronin, Ph.D for teaching me the ropes of the Golin

lab,

Dr Dottie Hutter, Ms Sherry Supernavage, Ms Donia Palomo, Mrs Leanne Hanson,

Dr Michael Mitchell and Lt Col Lynette Hamilton, Ph.D whose friendship, support and willingness to join me at bars will always make me remember graduate school fondly,

The faculty and students of the Biology Department, And especially Dr John Golin, whose genuine enthusiasm for research and teaching kept

me going more times than I can count!

VI

Multiple drug resistance is quickly becoming an obstacle to the treatment of disease Drugs that are used for the treatment of infectious disease as well as those required for cancer chemotherapy are no longer effective because the organisms or cell that they target have developed a way to combat the drug Many bacteria have become resistant to commonly used drugs such as penicillin and other antibiotics (Lewis e?.al., 1994; Nikaido et.al., 1996) New drugs are being developed, but organisms can become resistant to those quickly The CDC reports that the first bacteria resistant to the drug fluorquinolone appeared after that drug had only been in use for only 6 months

(www.cde.gov) The emergence of drug resistant tuberculosis is caused by the

incomplete treatment or lack of treatment for infection by a drug susceptible strain

(Cohen and Murray 2004) and is a growing threat in many developing countries

(www.cdc.gov)

Not only are bacterial infections becoming more difficult to treat, but cancer cells

can become resistant to chemotherapeutics as well (Bradley et.al., 1988; Endicott et.al

1989) Itis estimated that 40% of operable cancers and 80% of inoperable cancers develop drug resistance (www.cdc.gov) Mutations in the gene p53 are not only

associated with the development of cancer, but Zhan et.al (2005) demonstrated that deletion of p53 is also associated with the development of multidrug resistance Broad based resistance is usually due to overexpression of members of the ATP-binding cassette (ABC) superfamily of membrane transporters (Gottesmann et.al 1995) In breast cancer, the ABC transporter BCRP (breast cancer resistance protein) is required for resistance to

2 mitoxantrone and anthracycline (Sarkadi et.al 2004) Over-expression of this protein causes hyper-resistance to anti-tumor agents and reduces the accumulation of the drugs This over-expression of BCRP also enhances the efflux of rhodamine 123 (Doyle and Ross 2003) Multidrug resistance in human cancers is primarily due to two proteins P- glycoprotein (P-gp) encoded by the MDRI (multidrug resistance 1) gene and the

multidrug resistance associated protein, MRP1 (Gottesman, Fojo and Bates 2002; Cole

et.al 1992; Chen et.al 1986) There are 30 ABC transporters in yeast and 47 in humans The general structure of the ABC transporter consists of four domains: two hydrophobic transmembrane regions and two soluble ATP-binding cassettes The hydrophobic

domains are thought to form a pore through which the substrate is moved These

domains are not well conserved among the different transporters and may be the

determinants of substrate specificity The nucleotide binding domain is the most

conserved part of the protein There are several consensus motifs including the Walker A and Walker B motifs and a region called the signature region Although other proteins that bind ATP contain the Walker A and B motifs, the presence of the signature region defines an ABC transporter ABC transporters use the energy of ATP hydrolysis to

efflux or import substrates (Gottesman et.al 1995, Horio, Golttesman and Pastan 1998)

and have many functions in cells besides multidrug resistance For example, the protein Ste6 (sterile 6) is responsible for transporting the steroid mating hormone across the cell membrane (www.yeastgenome.org)

Mutations in proteins other than transporters can alter drug resistance For

example, loss of the chromatin remodeling complex SWI/SNF (mating type

switch/sucrose non-fermenting) may render tumor cells resistant to the lethal effects of

cis-platin (Strobeck et.al 2002) Not only can mutations in this complex render a cancer

cell resistant to a substrate, but the lack of this complex can cause cancer In humans, the SWI/SNF ATPase subunits, BRG1 and BRM, are lost in a subset of human cancer cell

lines and human primary cancers (Reisman et.al 2003) These proteins, the human

homologs of yeast SNF2 (sucrose non-fermenting 2), are tumor suppressors Another

member of the human SWI/SNF complex is associated with the development of tumors

in soft tissue and the nervous system INI1 (integrase interactor 1) is the human homolog

of the yeast SNFS5 and was first identified as a protein required to bind the Human

Immunodeficiency Virus (HIV) protein integrase and stimulate its DNA joining activity

(Kalpana et.al 1994) Homozygous loss of function mutations in mice result in death

between embryonic 3.5 and 5.5 days Fifteen percent of the mice that were heterozygous for the JNI/ mutation developed a loss of heterozygocity at this locus, which resulted in undifferentiated or poorly differentiated sarcomas (Guidi et.al 2001) Versteege et.al (1998) found that many rod-shaped tumors in humans are associated with mutations in INI] This mutation causes aggressively malignant tumors that often arise in children less than two years old (Kufe et.al 2003) Because the only treatment is surgery and

relapses generally occur after six months, the current survival rate for these cancers is

less than 10% Analysis of the sequence found a frameshift or nonsense mutation that caused a truncation of the INI1 protein In addition, the truncation of one allele was

associated with the loss of the other allele

4 Cancers caused by mutation in JN// are probably so aggressive because IN//, in addition to being part of a complex that modifies histone, also plays a role in the

regulation of the cell cycle Zhang et.al (2002), found that expression of INI1/hSNF5 causes G,-G, arrest in a histone deacetylase (HDAC) dependent manner INI1 is directly recruited to the cycling D1 promoter If INI1 is not present, cyclin D1 expression is not repressed and the cell continues to progress through the cell cycle Clearly more research

is required to understand the roles these chromatin remodeling proteins are playing in the progression of cancer Our studies suggest the role may be in a regulator of multidrug resistance

Yeast as a Model System for Studying Multidrug Resistance

Yeast are an excellent model system for studying multidrug resistance

Saccharomyces cerevisiae is a eukaryote that has all the major organelles found in higher eukaryotes The genome is completely sequenced and encodes nearly 6,000 genes

(Goffeau et.al., 1996) Yeast grow quickly with a doubling time of approximately 2 hours minutes and most importantly for genetic studies, the organism can exist as either haploid

or diploid There are two mating types (a and a) In haploid cells, recessive mutations can be screened easily and the two mating types can be crossed to form a diploid The diploid will undergo meiosis under nitrogen starvation creating an ascus containing a tetrad of spores Analysis of these spores can yield information about recombination between two genes and other genetic interactions, thus providing mechanistic clues to the proteins required for the development of multidrug resistance

Early studies by Linnae (1968), Cohen and Eaton (1979) and Lucchini et.al (1979) identified a set of mutants that conferred resistance to chemically different

inhibitors such as chloramphenicol, tetracycline and erythromycin These drugs interfere with mitochondrial protein synthesis, but the mutations causing the resistance are in

nuclear genes Rank et.al (1975) found that reduced ['*C]-chloramphenicol

accumulation by resistant cells was the result of a single nuclear mutation Saunders and Rank (1982) evaluated all these mutants and complementation testing showed that they all bore mutations in the same gene This gene was named PDR (Pleiotropic Drug

Resistance 1) The 3.5 kb RNA codes for a zinc-finger transcription factor that is

classified as a general positive regulator of permeability genes (Goffeau et.al 1996) Disruption of the PDR/ gene in the resistant phenotype restores wild type drug sensitivity while the loss of this gene in the wild type causes hypersensitivity

Since the discovery of PDRJ, other genes in this network mediating multidrug resistance have been discovered Falco and Dumas (1984) isolated mutants resistant to sulfometuron methyl (SM), a strong herbicide that inhibits growth on minimal medium Some of these mutations were in PDR/ while others were in a gene designated PDR2

(also termed YRR/J) In 1986, Subik ef.al isolated a mutant resistant to mucidin, a drug

that inhibits the mitochondrial transport system The gene responsible for this resistance, PDR3, produces a protein that like Pdrlp, contains a DNA-binding domain consisting of

a zinc-finger motif

is highly homologous to PDR5 (Servos et.al., 1993), but the protein is involved in

mediating resistance to 4-Nitroquinolone Oxide and other unrelated compounds YORI (Yeast Oligomycin Resistance) shares a sequence similarity to mammalian multidrug resistance associated protein, the yeast cadmium resistance protein, and the cystic fibrosis transmembrane conductance regulator (Katzmann et.al 1995), The protein is responsible for some oligomycin resistance in yeast Pdr5p, Yorlp, and Snq2p share some substrates (Decottignies et.al 2001) However, only PdrŠp transports cycloheximide,

tritylimidizaole and rhodime 6G, three substrates used extensively in our studies

Sequencing of the yeast genome revealed the existence of 30 members of the ABC- transporter superfamily Although some of these such as Pdr15p appear to be drug transporters, none have the same substrate specificity as Pdr5p

In 1994, Leonard, Rathod and Golin, showed that a pdr5::Tn5 mutant exhibits

reduced chlormaphenicol efflux Similar results were obtained later with rhodamine 6G (Kolazckowxki et.al 1996) Thus, pdr5::Tn5 loss of function mutation is hypersensitive

to many xenobiotic compounds because of decreased efflux In the pdr5A mutant, the substrate is effluxed 4-6 fold less than in cells that have a functional PDR5p PDR5p is important for resistance to drugs other than chloamphenicol A study by Meyers et.al (1992) showed that the disruption of PDR5 causes sensitivity to cycloheximide, SM, and the ergosterol biosynthesis inhibitor, clotrimazole Genetic analysis of a pdr5::Tn5, PDRI-3 mutant indicates that the observed hyperresitance to cycloheximide requires a functional PdrS5p Northern analysis shows that the 5.2 kb PDRS transcript is over-

expressed in PDR/ (resistant) strains, is missing in the Tn5-insertion mutant and reduced

ina pdrlA This shows that PDR3S is a target of PDR1p and is positively regulated by it

PDR3 is homologous to PDR/ and Katzmann et.al (1994) showed a functional overlap between the two genes In addition to regulating PDR5 transcription, Pdr1p activates transcription of PDR3 Both of these genes regulate transcription from PDRS5 as shown by Katzmann et.al (1994) In this study, a plasmid containing the PDR5 promoter fused to a B-galactosidase gene was used to measure the levels of PDR5S transcription in the mutant A double disruption of PDRI and PDR3 markedly decreases PDR5

transcription The drug sensitive phenotype of this double disruption is very similar to that of a Apdr5 mutant (Katzmann et.al 1994)

8 PDRSp and the genes that regulate it are not the only genes that play a role in multidrug resistance in yeast Shallom and Golin (1996) isolated a series of second site revertants of a pdr5::URA3 deletion mutant These mutations that caused drug resistance

in a previously hypersensitive Apdr5 strain mapped to three clusters Two of these, Revertant 4 and Revertant 11, are in the PDR2 (YRR1/) gene This indicates that there must be a pathway mediating multidrug resistance that is independent of PDR5 Furthermore, mutations in SJN4 (Switch Independent) are multidrug hypersensitive (Fleckenstein, Shallom and Golin 1999)

SIN4 codes for a protein that is a component of Mediator that is associated with RNA polymerase II (Song and Carlson 1998, Bjorklund and Kim 1996) It is involved in the positive and negative regulation of transcription (Jiang and Stillman, 1992; Chen et.al., 1993; Piruat et.al., 1997) A loss of function mutation in this gene causes significant hypersensitivity to a many of the same inhibitors as Apdr5, but the two do not interact A B-galactosidase assay shows no decrease of PDR35 transcription in a Asin4 mutant The double Apdr5, Asin4 mutant is more hypersensitive than either single mutant, indicating that the two genes operate in independent pathways that are essential for a basal level of drug resistance (Fleckenstein, Shallom and Golin 1999) Since Sin4p

is known as a global regulator that affects the negative and positive transcription of many genes through chromatin remodeling (Jiang and Stillman, 1995), a loss of function mutation causes many different phenotypes including temperature sensitivity, less superhelical coiling of plasmid DNA and expression of genes lacking a upstream

activating sequence (Jiang and Stillman, 1995; Macatee et.al., 1997; Chen et.al., 1993)

A Asin4 mutation suppresses some mutations such as those in Anhp6, Agcn5, and Aswi6 that make up chromatin remodeling complexes but not Aswi5 or Aswi2 (Yu et.al.,2000) The Asin4 may relax the RNA polymerase holoenzyme’s specificity so that it can activate transcription in the absence of factors that are normally required The Sin4p part of Mediator may function as an activator checkpoint to verify the presence of activator at the promoter before the enzyme can begin transcription (Yu et.al 2000) In addition to Mediator, other complexes that work with RNA polymerase II are known to be required for multidrug resistance in yeast and other eukaryotic systems

RNA polymerase II holoenzyme and subcomplexes

An understanding of how RNA polymerase II and its accessory complexes

regulate gene transcription is required to determine the role these complexes may play in multidrug resistance RNA polymerase IJ is the protein complex that is responsible for the transcription of genes into mRNA The enzyme complex has affinity for DNA and can bind, but it cannot recognize any specific promoters Jn vitro experiments show that there is a low level of transcription when only the DNA and the RNA polymerase II are combined (Myer and Young 1998) Alone, RNA polymerase II may be sufficient for the transcription of low levels of unregulated genes Those genes that are not always active

or that require greater levels of transcription also require modulator elements that

enhance transcription At any promoter, the transcriptional machinery can include the Mediator complex, Srb10 CDK complex, SWI/SNF complex and SAGA (Spt-Ada-Gen5 acetyltransferase) (Bjorklund et.al 1999; Meyer and Young, 1998)

10 The core of yeast RNA polymerase II has 12 subunits that have different

functions (Cramer et.al 2000) It comprises over half the total mass of the Pol II and is composed of elements common to all RNA Polymerases (Gnatt et.al 2001), RPB3,

RPB10, RPB11, RPB12 and regions of RPB1i and RPB2 The cleft of the core is at the

center of the enzyme and contains the active site This cleft is where incoming DNA enters from one side The highly conserved domain at the C-terminus (CTD) of the largest subunit (RPB1) contains multiple repeats (25-52) of consensus sequence

YSPTSPS and phosphorylation of Serine 2 and Serine5 are substrates for several kinases that have a role in the regulation of gene expression (Myer and Young, 1995; Bently 2002) The phosphorylation and dephosphorylation of the CTD is regulated and is part of the transcription cycle This phosphorylation allows the coupling of transcription and mRNA processing CTD appears to mold itself to its binding partner differently

depending upon the protein being bound and its phosyphorylation state (Fabrega et.al 2003) This may be a way for CTD to bind to many different factors

At the initiation of transcription the gene specific regulatory factors bind near the initiation site of the promoter These regulatory factors can act either indirectly by

recruiting factors that modify chromatin or directly by interacting with the transcriptional machinery to recruit it to the core promoter; the minimum sequence needed for basal

transcription (Ptashne and Gann 2002; Cosma 2002) After initiation of transcription in

vitro, many of the general transcription factors remain at the promoter in the scaffold complex (Yudkovsky, Ranish and Hahn 2000) The purpose of the scaffold may be to mark genes that have been transcribed and bypass the need for recruitment for further

rounds of transcription

General transcription factors are non-gene-specific accessory proteins that help in general transcription Transcription Factors (TF) IID and ITH act with the RNA

polymerase core enzyme to make up the RNA polymerase II holoenzyme Three

components are required for transcription to be initiated in eukaryotes TFIID is

composed of the TATA box binding protein (TBP) and approximately 14 TBP-associated factors (TAFs) nearly all of which are conserved throughout evolution (Albright and Tijan 2000) Yeast have only one form of TFIID but D melanogaster and humans have various forms that are cell type and development specific (Chen et.al 1994, Thut et.al

1995, Wasserman and Sauer 2001)

The core promoters recognized by RNA Polymerase II can contain several

sequence elements: TATA which binds the TBP of Pol II, BRE (TFIB-recognition element, Inr (Initiation element) and DPE (downstream promoter element) Most

promoters contain one or more of these elements (Smale and Kadonaga 2003)

Only 30% of mRNA genes in Drosophila melanogaster contain recognizable TATA elements (Ohler et.al 2002) In D melanogaster and human promoters many of these TATA-less promoters have some combination of Inr and DPE (Smale and

Kadonaga 2003) While there are other sequence elements that can be recognized by Pol

II mutations in TATA box show that if the sequence is present in a promoter, it is

generally necessary If the TATA box is mutated from the consensus site transcription is severely decreased (Wobbe and Struh! 1990) Biochemical studies of the HIS4

(Histidine) promoter (Ranish, Yudkovsky and Hahn 1999) show that mutation of the

12

TATA sequence to a GC rich sequence allows recruitment of the transcriptional

machinery to the promoter at a reduced level, but does not allow initiation of

transcription If TBP interacts with TATA-less promoters, it must do so by a different mechanism A mutation in TBP that inhibits binding to TATA-containing promoters has

no effect on binding to Inr-containing promoters (Martinez et.al 1995)

The full TFIID is not required at all promoters In yeast, TAF-independent promoters recruit TBP but not the TAFs upon gene activation (Kuras et.al 2000; Li, Bhauymik and Green 2002) At some TFI[D-independent promoters, a TAF containing complex such as SAGA, which is thought to contain Sin4p, may functionally replace TFIID In yeast, TFIID and SAGA may functionally overlap at many genes (Lee et.al

complexes such as SAGA and SWI/SNF, which contain Sin4p and Snf5p respectively and are known to be required for multidrug resistance Enhancer elements are regulatory DNA sequences to which proteins can bind that influence the rate of transcription of a gene Enhancers can be several thousand base pairs away from the gene that is regulated (Alberts et.al 2002) Enhancers are proteins that influence gene expression by

introducing bends and other distortions into the DNA (Thomas and Travers 2001) This bending may form loops in the chromatin that bring together proteins that are bound to distant sequences of DNA which may then act to activate transcription, The most

common activator proteins use leucine zippers, helix-loop-helix, homeodomains, helix- turn-helix proteins and zinc fingers These domains are rich in the positively charged amino acids and lysine and can therefore interact with the negatively charged DNA The contact of the protein with the enhancer region of the DNA can recruit other proteins to the area that will remodel the nucleosome and initiate transcription Coactivators and corepressors are protein factors that do not bind DNA, but rather contain a recognition site for domains present on a DNA binding transcription factor (Alberts et.al 2002) These proteins bind to these other proteins and may then act to either recruit other

proteins to the site or by carrying a modifying factor such as a kinase or acetyltransferase (Merika and Thanos 2001) The transcription of most genes requires a specific

combination of transcription factors and coactivators or corepressors

The interaction of coactivators and corepressors with RNA polymerase II

depends on an ancillary factor, mediator Mediator associates with the enzyme through the non-phosphorylated CTD and stimulates its phosphorylation by TFIH This complex forms a physical connection between activators and Pol II which suggest that rather than

an activator having a direct interaction with Pol II, the signal is transduced by Mediator (Thompson et.al 1993; Kim et.al 1994) It is required for both basal and activated transcription (Myers e7.al 1998; Lee and Young 2000), but the interaction between Mediator and RNA polymerase II is transient and occurs only during transcription

14

initiation (Svejstrup et.al.1997) Mediator is required because it interacts with the UAS

in the absence of a TATA box, a key core promoter element, whereas RNA polymerase II cannot (Kuras et.al 2003) This complex is found in all eukaryotes, but the subunits that comprise it are the least conserved of the transcriptional machinery The proteins found in the yeast Mediator are homologous to those found in Drosophila melanogaster and humans (Boube et.al 2002) and many mediator subunits may be targets of regulatory

factors (Malik and Roeder 2000; Boube et.al 2002) In mammalian cells, there are

several different versions of Mediator and the composition of the complex varies

depending upon cell type There is evidence that each cell-specific complex responds to different, although possibly overlapping, activators (Kingston 1999; Malik and Roeder

2000)

The proteins that comprise Mediator were first isolated in screens that suppressed mutations or truncations in the RNA polymerase II CTD These were named SRBs for suppressors of RPB mutation (Thomson et.al 1993) The function of Mediator is to transmit the activating effect from DNA binding transcription factors to the actual RNA polymerase II transcriptional machinery (Naar et.al 2001) Mediator is only associated with remodeled chromatin that has been loosened from the histones (Kuras et.al., 2003) and has three biochemical activities It functions to stimulate basal transcription, it enables the stimulation of transcription by activators and it stimulates CTD

phosphorylation (Sakurai and Fukasawa, 1998) This phosphorylation of CTD may result in promoter clearance by Pol II and help recruit the capping and splicing machinery required for elongationof the transcript (Orphnides and Reinberg 2002) Mediator is

composed of 16 proteins consisting of the Srb proteins (-2,-4,-5,-6,-7), the Med proteins (-1,-2,-4,-6,-7,-8), Gal 11, Sin4, Rgr1, Rox3, and Pgd1 Mutations in the Srb or Med proteins result in inviability but those in the tail region are not essential Those proteins

present in the tail of Mediator (Gal 11, Sin4, Pgd1) were identified as a result of their

influence on repression of certain genes (Carlson, 1997) Mutations in these genes cause

an increase in expression of genes whose promoters lack Upstream Activating Sequences (UAS) Giang and Stillman, 1992; Chen ef.al.,1993) Loss of function mutations in SIN4 and GALI] have many of the same phenotypes (www.yeastgenome.org) Both of these proteins are in the tail region of Mediator and Gallp is bound to the rest of Mediator through its interaction with Sin4p A loss of function Asin4 mutant lacks the entire tail region of Mediator (Boube et.al 2002)

Many of the complexes that either activate or repress transcription do so by altering histone acetylation or remodeling the chromatin DNA is tightly coiled and wrapped around histones (Kornberg, 1974) The nucleosome consists of 146 bp of DNA wound around the histone protein octamer The core particle of this octomer consists of a

pair of each histones H3, H4, H2A and H2B The histone proteins are arranged so that

the N-terminal tails project out from the core and contain many positively charged amino acids such as Lysine and Arginine These positively charged amino acids have an

affinity for the negatively charged DNA The histones may be modified by N-

acetylation, methylation or phosphorylation The degree of interaction between the negatively charged DNA and the histones is dependent upon the level of histone

modification

16 Histone acetyltransferase (HAT) activity is associated with many proteins HAT proteins contain a region called the bromodomain which consists of a four-helix bundle that acts as a binding site for acetyllysine (Jacobson et.al 2000) The 1.8 MDa SAGA complex contains the histone acetyltransferase, Gcn5 When an activator such as

Mediator binds to the nucleosome in the presence of SAGA, the level of histone

acetylation increases (Bjorklund et.al 1999) and the DNA is made more accessible to RNA Polymerase II and various transcription factors

ATP-dependent chromatin remodeling complexes such as SWI/SNF, of which SnfŠp is a component, use ATP to remodel nucleosomes to facilitate transcription

(Phelan et.al., 1999) The yeast SWI and SNF genes were originally identified in screens for genes required for the expression of HO (Homothalism) gene (Stern et.al 1984) and SUC2 (Neigeborn and Carlson 1984) The SWI/SNF is a 2 MDa complex composed of

11 proteins that regulates the transcription of a subset of genes (Geng et.al 2001) It is required for transcriptional initiation (Gregory et.al 1999; Hirschorn et.al., 1992) and associates directly with target promoters (Dimovi et.al., 1999) Mutations in SWI/SNF decrease transcription and alter chromatin structures at the promoters (Sternberg et.al

1987, Kruger and Herskowitz 1991) The role of this complex in chromatin remodeling

is indicated by the suppression of defects in SWI/SNF by histone mutations (Kruger et.al

1995) If the histones are not formed correctly, the absence of a functional SWI/SNF no

longer hinders transcription The action of SWI/SNF is not presently clear, but the following effects on the chromatin have been observed When SWI/SNF is present, the total length of DNA per nueleosome decreases, the susceptibility of DNA to nucleases

increases, the histone protein octomer slides along the DNA but is not itself disrupted and

there is a change in the superhelical twist of the DNA (Flaus and Owen-Hughes 2001) Biggar and Crabtree (1999) have shown that SWI/SNF is not only required in the

initiation of transcription, but also in the elongation of the transcript This indicates that the chromatin must continue to be remodeled after the transcriptional machinery has

bound to the promoter of the gene Schnitzler et.al (2001) used atomic force microscopy

to directly image the human SWI/SNF remodeled genes They observed that remodeling

of mononucleosomes by hSWI/SNF resulted in a dimmer of mononucleosomes in which DNA was more loosely bound that in non-remodeled nucleosomes The histone

octomers were not destroyed, but were arranged in clusters with long stretches of bare DNA between the clusters Not only do the nucleosomes slide along the DNA, but there

is also a structural alteration of the nucleosomes that results in less affinity for the DNA

Mutations in the genes that code for the proteins that comprise SWI/SNF cause pleiotropic changes in the cell phenotype implying that SWI/SNF has a global role in regulation However, microarray analysis (Sudarsanam et.a/ 2000) shows that the

SWI/SNF dependent genes are scattered throughout the genome and share no common motifs Even though mutations in this complex have many effects, only 5% of genes in Saccharomyces cerevisiae are directly regulated by SWI/SNF The limited recruitment

of SWI/SNF may be due to two factors SWI/SNF is not very abundant in yeast The average haploid cell contains only 100-200 complexes (Coté et.al 1994; Cairns et.al 1999) This is compared with an estimated 3,000 molecules of RNA Polymerase II per cell (Wilson et.al 1996) SWI/SNF is approximately 10 fold less abundant than the

18 related chromatin remodeling complex RSC (Cairns et.al 1996) The second factor for the paucity of genes that require show great differences in transcription in the absence of SWI/SNF is the redundant function with histone acetyltransferase Gcn5 (Biggar and

Crabtree 1999; Pollard and Peterson 1997; Roberts and Winston 1997) A loss of

function mutation in GCN5 has a small effect on global transcription, similar to a Asnf2 mutation (Holstege et.al 1998) Presumably, a double Asnf2Agcn5 would have a greater effect on transcription

The manner in which SWI/SNF regulates gene transcription is still unclear Humans have multiple SWI/SNF remodeling complexes (Muchardt and Yaniv 1999;

Kingston and Narlikar 1999) and complexes purified from different tissues display

significant subunit heterogeneity (Armstrong et.al 1997; O’Neill et.al 1999; Wang et.al 1996) This difference in complex composition may reflect the cell specific

differences in transcription that are required in different tissue types and may reflect a level of regulatory control SWI/SNF complexes in yeast, D melanogaster and humans have been shown to contain either actin or actin-related proteins (Papoulas et.al 1998; Cairns et.al 1998; Zhao et.al 1998) The actin may be required for SWI/SNF to

associate with the nuclear matrix but not for the complex’s catalytic activity (Zhao et.al 1998)

SWI/SNF-dependent genes have no common motifs (Sudarsanam et.al 2000) so

it is unlikely that control of transcription is through direct binding if the complex to the DNA Recent data suggest that DNA-binding regulatory proteins recruit SWI/SNF to specific promoters Yudkovsky et.al (1999) showed that SWI/SNF can be isolated from

yeast nuclear extract by a DNA-bound activator Chromatin Immunoprecipitation (ChIP) shows that SWI/SNF directly interacts with the transcriptional activator Swi5 (Neely et.al 1999) for transcription from the HO promoter SWI/SNF has also been shown to

directly associate with the activation domains of the activators Gcen4, Hap4, Gal4-AH and VP16 (Neely et.al 1999; Natarajan et.al 1999; Yudkovsky et.al 1999) Alternatively,

SWI/SNF may interact with general transcriptional machinery Studies have shown that there is an association of SWI/SNF with the Pol II holoenzyme in yeast and humans (Cho

et.al 1998; Neish et.al 1998; Wilson et.al 1996) However, Hirschorn et.al (1992)

showed that in vivo nucleosome remodeling at the SUC2 promoter occurs in the absence

of a TATA box and presumably, a TATA-binding protein (TBP) recruitment

The interaction of these various complexes in transcription are best understood in the activation of the HO (homothalism) gene responsible for switching mating type in yeast (See Figure land Yu et.al 2000) Chromatin structure plays role in the regulation of HO

transcription and expression is altered in mutations of SWI/SNF, SAGA histone

acetyltransferase complex and the Sin3/Rpd3 deacetylase complex Cosma et.al (1999) showed that the activation of HO transcription involves the ordered recruitment of

transcription factors The Swi5p enters the nucleus and binds to the HO promoter Swi5 recruits SWI/SNF which then recruits SAGA Both of these must be bound in order for SBF (Swi4p and Swi6p) binding which is required for the direct activation of HO

mutation in SIN4 has a phenotype similar to that of a Aho mutation which may be due to

a defect in SAGA required for HO transcription Sin4p has recently been shown to be a

member of SAGA (Yu et.al 2000)

20

Since transcription of most genes in Saccharomyces cervisiae is a complex

activity requiring various gene-specific transcription factors as well as more general activators and chromatin remodeling complexes, it is not surprising that mutations in this process may have an impact on multidrug resistance Sin4p is a component of both the mediator that is required for RNA polymerase II to recognize and bind to various

promoters and the SAGA complex that remodels chromatin with the histone

acetyltransferase Gcn5p (Jiang and Stillman, 1995; Yu et.al 2000) A loss of function

mutation in SJN4 causes drug hypersensitivity and is in a different genetic pathway than

PDRS (Fleckenstein, Shallom and Golin, 1999), Additionally, the YRR/-2 gain of

function revertant is resistant even though it bears a deletion of PDR5 Both YRRI-2 and SIN4 are in PDR5- independent pathways Do they function in the same pathway

mediating multidrug resistance? As shown in the transcription of HO, SAGA is recruited

to the promoter by the SWI/SNF complex Loss of function mutations in SJN4 can suppress swi/snf mutations and it has been shown that Sin4p and Snf2p, the ATP-

dependent enzyme in SWI/SNF complex, act in the same genetic pathway for activation

of CTS/ and HIS4 activation, but in different pathways for the expression of the Ty/ delta

promoter (Moss and Laybourn, 2000; Roberts and Winston, 1997; Jiang and Stillman,

1995) The work described in this study shows that a loss of function mutation in one of the components of this complex, Snf5p, is required for mediating PdrSp-independent multidrug resistance, although it has only a small effect on PDR5 expression

Figure 1 A model for the transcription of the yeast HO gene (Bhoite et.al 2001).

Materials and Methods

Overview of Experiment

The purpose of the experiments performed in this study was to identify and characterize new genes required for mediating multidrug resistance in Saccharomyces cerevisiae Previous work (Fleckenstein et.al 1999) showed that the global transcription factor Sin4p is required for providing a basal level of drug resistance that operates

independently of Pdr5p Genetic crosses of Asin4 and gain-of-function YRR/ mutations that are drug resistant even in the absence of a functional PDR5 showed that this

resistance requires a functional Sin4p To identify further genes required in this pathway drug sensitive mutants were constructed using transposon mutagenesis (mTn3::LEU2 transposon) Any drug sensitive mutants were mated to a known Asin4 mutant to

determine which of these new mutants were in SJN4 and which were in new genes A new gene required for mediating multidrug resistance was isolated and identified using vectorette PCR RT-PCR was used to determine which genes have altered transcription

in this drug sensitive mutant and transport assays (FACS, Tritylimidazole efflux) to determine the process by which the protein mutated conferred drug resistance

22

Strains and media

All yeast and bacterial strains are described in Table 1 Where appropriate, strains containing plasmids were grown in selective medium

Most experiments using yeast require growth of the cells on selective media All components for the media were obtained from US Biological and dissolved in reverse- osmosis water (RO-H,O) YPD media was made by combining 10 g of yeast extract, 20

g of peptone, 20 g of dextrose, and 20 g of agar, and per liter YPG media contains the same components except the 20 grams of dextrose were replaced by 20 ml of glycerol

SD (synthetic dextrose) media was made by combining 20 grams of dextrose, 20 grams

of agar, 7 grams of yeast nitrogen bases without amino acids and 10 ml of 1% solutions

of the appropriate amino acids Sporulation media was composed of 2.5 grams of yeast extract, 20.0 grams of potassium acetate, | gram of dextrose and 20 grams of agar After the components were mixed, acetic acid or potassium hydroxide was used to adjust the

pH to 7.0 LB (Luria-Bertani Broth) media was made by mixing 10 grams of tryptone, 5 grams of yeast extract, 10 grams of sodium chloride and 20 grams of agar All media was autoclaved at 121°C and 15 psi for 30 minutes Cells were grown in YPD broth for most experimental procedures For transport assays, cells were cultured in SD media

All compounds were purchased from Sigma-Aldrich (St Louis MO) and

dissolved in DMSO unless otherwise indicated Media to which inhibitors were to be added were cooled to 50°C after autoclaving Ampicillin was dissolved in sterile water to make a stock solution of 50 mg/ml This was diluted in the LB media to a final

concentration of 50 ug/ml Cycloheximide powder was dissolved in

24

DY150 MATa PDRS SIN4 YRRI(PDR2) ura3 | David Stillman

leu2 trp] ade2 can]

DY1704 Isogemc to DY 150, but sin4::URA3 David Stillman DYK2.1 MATa pdrS::URA3 SIN4 YRRI leu2 Katzmann et al

ura3 his3 trp lys2 (1994)

REV4 Isogenic to DYK2.1, but YRR1-2 Shallom and

(PDR2-2) Golin (1996)

REV11 Isogenic to DYK2.1, but YRRI-3 Shallom and

(PDR2-3) Golin (1996) JG499-24A Pdr5::Tn5 SIN4 YRRI-3 Shallom and

Golin (1996)

RR2 Isogenic to REV4, but This study

sin4::Tn3::LEU2 RR3 Isogenic to REV4, but snf5 This study SC4741 Mata PDRS SIN4 YRRI SNF5 ura3 Invitrogen

leu2 his3 metl5

SC2909 Isogemc to SC4741, but pẩrŠ5::genf Invitrogen

SC1976 [sogenrc to SC4741, but sin4:: genf Invitrogen SC7175 Isogenic to SC4741, but snfS-:gent Invitrogen SC1586 Isogenic to SC4741 but snf2::gent’ Invitrogen SC1790 Isogenic to SC4741 but sng2::gent’ Invitrogen SC7390 Isogenic to SC4741 but spt20::gent’ Invitrogen SEY6210 Isogenic to DYK2.1, but PDR5 Katzmann et al (1994)

Table 1 Yeast and Bacterial Strains used in this study The designation of

gene:: marker gene indicates that the gene has been replaced with the marker gene

sterile water to make stock solutions of either 10 mg/ml or 1 mg/ml This was diluted in YPD media to final concentrations ranging from 0.01 wg/ml- 0.5 ug/ml Clotrimazole was used to make the same concentrations of stock solutions and then diluted in YPD media to final concentrations ranging from 0.05 ug/ml-5.0 ug/ml Anisomycin was dissolved in DMSO to make a stock solution of 10 mg/ml and diluted in YPD media to final concentrations ranging from 0.05 ug/ml-20.0 ug/ml Geneticin was added directly

to cooled, sterile YPD media to a concentration of 200 mg/L 5-FOA (5-fluoroorotic

acid) was added to SD-ura media to a final concentration of 0.1%

General Techniques

The reporter plasmid used in this study is a shuttle vector that can be grown in both E.coli and S cerevisiae Growth in E.coli allows for rapid amplification of the plasmid and the collection of greater amounts of DNA

Purification of Plasmid DNA from Bacteria (Qiagen Midi-Prep)

Plasmid DNA was prepared using a Qiagen Midi Prep Kit The bacterial culture was grown overnight in 200 m! LB broth to which ampicillin (SOmg/ml) had been added

at 37°C with shaking The bacterial cells were harvested by centrifugation at 6000g for

15 minutes at 4°C and the supernatant was discarded The cell pellet was resuspended in

10 ml of resuspension buffer (P1) and then 10 ml of lysis buffer (P2) was added After mixing, the tube was incubated at room temperature for 5 minutes Ten milliliters of

26 chilled neutralization buffer (P3) was added and the mixture was incubated on ice for 20 minutes The tubes were centrifuged at 20,000g for 30 minutes at 4°C The supernatant was removed and re-centrifuged at 20,000g for 15 minutes at 4°C The supernatant was applied to a resin column that had been equilibrated with 10 ml of equilibration buffer (QBT) and entered the resin column by gravity flow After all of the supernatant had passed through the resin column, 60 ml of washing buffer is applied The DNA was then eluted with 15 ml of Buffer QF and precipitated with 10.5 ml of room-temperature isopropanol This was mixed and centrifuged at 14,000g for 30 minutes at 4°C and the supernatant was removed The DNA pellets were washed with 5 ml of 70% ethanol and centrifuged at 14,000g for 10 minutes The supernatant was carefully removed and the DNA pellet was resuspended in 35-50 yl sterile distilled water DNA concentration was determined by spectrophotometric analysis using the formula:

DNA concentration (ug/ml)= Az¿; X dilution factor X 50

in 20 ml of YPD broth and grown at 30°C with shaking until it reached exponential

phase The cells were collected by centrifugation at 3000g for 5 minutes and the

supernatant was discarded The cell pellet was resuspended in 20 ml of sterile double distilled water and centrifuged at 3000g for 5 minutes The water was discarded and the cell pellet was resuspended in 1 ml of sterile water and transferred to a 1.5 ml

microcentrifuge tube The cells were pelleted again in the microcentrifuge at top speed

for 30 seconds The supernatant was discarded and the pellet resuspended in 350 ul of sterile water Fifty microliters of this cell suspension was placed in a new 1.5 ml

microcentrifuge tube and the cells pelleted at top speed for 15 seconds The supernatant was removed and the following was added to the cell pellet: 240 ul of PEG, 36 ul 1M Lithium acetate, 25 ul single-stranded DNA, 1-2 ug of transforming DNA and sterile water to make the final volume of the reaction mixture 351 ul The microcentrifuge tube was vortexed for at least one minute to fully resuspend the cell pellet and then placed at 42°C for 40 minutes The cells were pelletted at top speed for 30 seconds and the

supermatant was discarded The cell pellet was resuspended in 1 ml of water and 200 ul was plated onto the appropriate medium The plates were incubated at 30°C for 2-4 days

Electroporation of E.coli

Electrocompetent cells were prepared by growing cells to an O D of 0.60 and then chilling on ice for 30 minutes The cells were pelleted by centrifugation at 4,000 g (4°C) and resuspended in 1L of sterile, cold water After washing, the cells were again centrifuged at 4,000 g (4°C) and resuspended in 500 ml of sterile, cold water The cells

28 were repelleted, resuspended in 100 ml of sterile, cold water and recentrifuged The final cell pellet was resuspended in a final volume of 6 ml sterile, cold 10% glycerol 40ul of the cell suspension was placed in a pre-chilled 3 ml cuvette along with 1 ul of DNA (50-100 ng) A field strength of 13.8 kV/cm was applied for 5-6 msec using a BTX electrocell manipulator 600 (Genetronics, San Diego, CA) Following electroporation, 1ml of LB was added to the cell suspension and incubated at 37°C with shaking for 1 hour The suspension was then plated onto LB plates containing amplicillin and

incubated overnight at 37°C

B-galactosidase Assay

PDRS 1s one of the main genes studied in the Golin laboratory It is a major ABC transporter whose expression is regulated by different genes The yeast global regulator, SIN4, operates independently of Pdr5p and therefore does not regulate PDR5

transcription Any new genes isolate in this study were tested to determine if they could

be in this Sin4p, Pdr5p-independent pathway

One of the ways in which PDR3S transcription was measured with the reporter plasmid A865 that contains 865 bp of PDR5 upstream sequence fused to the E.coli LacZ gene The yeast strains that have been transformed with the plasmid A865 were grown in

2 ml of selective medium overnight at 30°C with shaking In the morning, they with diluted 1:10 and grown until they reach exponential phase 10° cells were spun in a

clinical centrifuge at speed 6 for 1 min to pellet the cells The cells were resuspended in

1 ml of Z buffer (60mM Na,HPO, 40mM NaH,PO, 10mM KCI 1mM MgSo, 1%

B-mercaptoethanol) As a control, a tube of Z buffer alone was treated throughout the experiment To each tube, 20 ul of 0.1% SDS and 30 ul of chloroform was added One hundred micrograms of glass beads was added to each tube which were then vortexed for

40 seconds to lyse the cells The cell suspension was warmed in a 30°C incubator for 30 minutes before 200 ul of 4 mg/ml ONPG was added to the tubes The tubes were placed

at 30°C until the yellow reaction product of the reaction was observed The time was noted and 500 ul of 1M Na,HCO, was added to stop the reaction The tubes were spun

in a clinical centrifuge at speed 6 for Imin to pellet the glass beads and cellular debris The supernatant was poured into cuvettes and the absorbance read at 420 nm The Miller Units of the reaction were determined using the following equation:

Ayo X 1000 AgooX ml of cells X time (min)

Genetic Analyses

Complementation testing

Complementation testing is used to determine if a new mutation is in a previously identified gene The unknown mutant is mated to a strain that contains a known mutation and the resulting diploids are tested For example, when testing new mutltidrug

resistance genes, the diploids are grown on medium containing the drug cycloheximide The growth of the diploid is compared to the growth of the drug sensitive haploids used

in the mating If the diploid can grow on the drug medium where the haploid cannot, this indicates that the two mutations supplied by the strains complement each other The

30

diploid has at least one wild type copy of all the genes required for drug resistance and the two strains bear mutations in different genes Conversely, if the diploid is as drug sensitive as the haploid mutants, this indicates that the diploid lacks one a wild type copy

of one of the genes required for drug resistance The two strains bear mutations in the

produced by Meiosis Once the ascus is dissolved each of these spores can be separated and grow individually

Following mating, diploids were streaked to sporulation medium and the plate was incubated for at least five days at 30°C The cell walls surrounding the tetrad were dissolved with glusulase and the spores teased apart using a glass needle The individual spores were germinated on YPD plates at 30°C The resulting colonies were transferred

to anew YPD plate for further testing