Citation
Identification and Characterization of Potential Virulence Factors Expressed by Candida albicans during Oropharyngeal Candidiasis in HIV-Infected Patients

Material Information

Title:
Identification and Characterization of Potential Virulence Factors Expressed by Candida albicans during Oropharyngeal Candidiasis in HIV-Infected Patients
Creator:
CHECKLEY, MARY ANN ( Author, Primary )
Copyright Date:
2008

Subjects

Subjects / Keywords:
Alleles ( jstor )
Candidiasis ( jstor )
Cell walls ( jstor )
Genes ( jstor )
Infections ( jstor )
Mice ( jstor )
Open reading frames ( jstor )
Triploidy ( jstor )
Virulence ( jstor )
Yeasts ( jstor )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Mary Ann Checkley. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
11/30/2005
Resource Identifier:
74459379 ( OCLC )

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

IDENTIFICATION AND CHARACTERIZATION OF POTENTIAL VIRULENCE FACTORS EXPRESSED BY CANDIDA ALBICANS DURING OROPHARYNGEAL CANDIDIASIS IN HIV-INFECTED PATIENTS By MARY ANN CHECKLEY A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2005

PAGE 2

Copyright 2005 by Mary Ann Checkley

PAGE 3

This document is dedicated to my parents, Celina and Jorge, for their unending efforts, encouragement, support, and love.

PAGE 4

ACKNOWLEDGMENTS I would like to acknowledge the unwavering support, encouragement, and guidance provided by my mentors Drs. Alfred Lewin and M. Hong Nguyen throughout my graduate career. Furthermore, I would also like to thank Dr. Neil Clancy for all his advice, support, and encouragement. I thank them for their genuine care and friendship in both professional and personal matters. I am thankful to the members of my supervisory committee, Drs. Henry Baker, Jeanine Brady, Paul Gulig, Jeffrey Hillman, Martin Handfield, and Ann Progulske-Fox, for their suggestions, advice, and constructive criticism. I thank Joyce Conners for her extreme efficiency in work and for taking care of me and all the other graduate students in this program. I would like to give special thanks to my family for their continued support and love, and for teaching me that I could achieve anything with commitment and hard work. Without their sacrifices I would have never achieved so much. My brother, William, has also provided invaluable support. I thank him for his love, care, advice, and eagerness to listen. I also give special thanks to my fianc, Ben Luttge, for his love, support, help, and patience. He has always been ready to listen about my project and has provided unyielding encouragement and motivation. Without my family and Ben’s constant support none of this would have come to fruition. Lastly, I would like to acknowledge past and present members from Drs. Lewin’s and Nguyen’s laboratory for their help, suggestions, and most importantly their friendship. I thank all the members of Dr. Nguyen’s and Clancy’s laboratories, especially iv

PAGE 5

Shaoji, Hassan, and Shuresh for their help with the mice experiments. Members of the Lewin lab have provided a home away from home. I thank our lab manager James Thomas for his professionalism, maintaining a pleasant laboratory environment, and his genuine friendship. I am grateful to Dr. Lynn Shaw, Dr. Patrick Whalen, and Robert Mino, for making me feel welcome when I joined the laboratory and for their technical advice. I would also like to acknowledge each of the present members of the Lewin laboratory, Alan, Marina, Lourdes, Jia, Jen, Verline, Fredric, Edgar, and all other members. They have all made my day-to-day experience in the lab worthwhile and enjoyable. I thank Brie Michael, a talented undergraduate student who worked with me in the final stages of my project. And finally, I would like to give special thanks to Lourdes and Jia for their friendship, care, and support both in and out of the lab. v

PAGE 6

TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES.............................................................................................................xi LIST OF FIGURES..........................................................................................................xii LIST OF ABBREVIATIONS..........................................................................................xiv ABSTRACT....................................................................................................................xvii 1 INTRODUCTION........................................................................................................1 Candida albicans..........................................................................................................1 Candidiasis....................................................................................................................2 Deep-seated and Disseminated Candidiasis (DC).................................................2 Mucocutaneous Candidiasis..................................................................................3 Oropharyngeal candidiasis (OPC)..................................................................3 Vaginal candidiasis........................................................................................5 C. albicans Virulence...................................................................................................6 Virulence Factors...................................................................................................6 Dimorphic transition......................................................................................6 Phenotypic switching.....................................................................................7 Adherence.......................................................................................................7 Secretion of proteases.....................................................................................8 Evasion of the immune system.......................................................................8 Tissue-Specific Virulence.....................................................................................9 In Vivo Expression of Virulence Factors....................................................................10 C. albicans Genes Identified using IVETs..........................................................11 Differential display RT-PCR techniques......................................................11 Antibody-based screening strategies............................................................11 Transcriptional profiling using microarrays.................................................13 In Vivo Induced Antigen Expression (IVIAT)....................................................14 IVIAT for C. albicans..................................................................................14 Library construction and sera used for IVIAT screening.............................16 Our Goals....................................................................................................................17 vi

PAGE 7

2 MATERIALS AND METHODS...............................................................................18 Strains and Growth Conditions...................................................................................18 Fungal Strains......................................................................................................18 Wild type and parent strains.........................................................................18 Mutants constructed.....................................................................................18 Bacterial Strains...................................................................................................20 Vectors.................................................................................................................20 DNA, RNA, and Protein Techniques.........................................................................21 DNA Techniques.................................................................................................21 Polymerase chain reaction (PCR)................................................................21 Cloning in bacteria.......................................................................................23 DNA sequencing..........................................................................................24 Yeast genomic DNA extraction and analysis...............................................24 Contour-clamped homogeneous electric field (CHEF)................................24 Southern blot................................................................................................25 RNA Techniques.................................................................................................26 Yeast RNA extraction and quantitation.......................................................26 Northern blot analysis..................................................................................27 Reverse transcriptase (RT)-PCR..................................................................28 Protein Techniques..............................................................................................28 Protein expression in E. coli and protein extraction.....................................28 Western blot.................................................................................................29 His-tag protein purification..........................................................................30 Generation of polyclonal antibodies to a fragment of IPF11959 protein.....30 Yeast protein extraction...............................................................................30 Enzyme linked immunosorbent assay (ELISA)...........................................31 Identification of C. albicans Proteins Expressed during OPC...................................31 In Vivo Induced Antigen Technology (IVIAT)...................................................31 Creation of a C. albicans genomic expression library in Escherichia coli..32 Sera collection and adsorption.....................................................................32 Library screening..........................................................................................33 Colony immunoblot......................................................................................33 Identification of potential open reading frames (ORFs)..............................34 Confirmation of the identified protein by the adsorbed sera........................35 Determination of In Vivo Expression of Genes Identified..................................36 RT-PCR with in vivo and in vitro grown C. albicans..................................36 Immunostaining of thrush sample with anti-IPF11959 polyclonal Ab........37 Determination of Identified Gene as a Virulence Factor............................................38 Creation of Isogenic Mutants..............................................................................39 Gene disruption............................................................................................39 Re-insertion of URA3 gene at its own locus................................................42 RE-insertion of One Copy of the Gene Back into its disrupted Locus...............42 Phenotypic Analysis of Mutants..........................................................................43 Growth curve................................................................................................43 Hyphae formation.........................................................................................43 Colony morphology and auxotrophies.........................................................43 vii

PAGE 8

Embedded agar phenotype...........................................................................44 Screening for cell wall defects.....................................................................44 Cell adherence assays...................................................................................45 Polymorphonuclear cells (PMNs) phagocytosis and killing assay..............47 Murine Models of Candidiasis............................................................................48 Disseminated candidiasis murine model......................................................49 Oropharyngeal candidiasis murine model....................................................50 Duplication Stability...................................................................................................51 In Vitro Conditions..............................................................................................51 Growth under non-stressful conditions........................................................51 Growth under stressful conditions................................................................51 Stability of the Triploid 1 KO Mutant Duplication During Growth in Mice......51 Southern Blot Analysis........................................................................................52 3 IDENTIFICATION OF CANDIDA ALBICANS ANTIGENS EXPRESSED DURING OROPHARYNGEAL CANDIDIASIS IN HIV-INFECTED PATIENTS.................................................................................................................53 Project Overview........................................................................................................53 Identification of C. albicans Antigenic Proteins Expressed During OPC Using IVIAT.....................................................................................................................54 IVIAT Steps.........................................................................................................54 Screening the C. albicans genomic expression library with the pre-adsorbed sera....................................................................................55 Identification of the ORFs responsible for the reactivity with the pre-adsorbed sera....................................................................................57 Identification of C. albicans Genes Encoding the Antigenic ORFs....................58 In Vivo Expression of Genes.......................................................................................62 Verification of RBF1 and IPF11959 mRNA Expression in Thrush Samples by RT-PCR.......................................................................................................62 Verification of IPF11959p expression within Thrush Samples by Immunostaining...............................................................................................63 IPF11959-His tagged protein fragment and anti-IPF11959 antibody..........64 Immunostaining of thrush samples..............................................................64 4 DETERMINATION IF GENES ENCODING ANTIGENIC PROTEINS EXPRESSED IN VIVO ARE IMPORTANT FOR CANDIDA ALBICANS VIRULENCE..............................................................................................................66 Step 1: Creating Isogenic Mutants..............................................................................66 Reasoning for Selecting ALG5 and IPF15632 for Pathogenesis Study..............67 Southern Blot Analysis for Identified Genes.......................................................67 Overview of the URA-Blaster Protocol..............................................................69 Disruption strategy for ALG5.......................................................................70 Disruption strategy for IPF15632................................................................71 Step 2: Analyzing the Mutants In Vitro......................................................................74 Step 3: Determining the Virulence of the Mutants.....................................................74 viii

PAGE 9

5 CHARACTERIZATION OF C. ALBICANS ALG5...................................................76 Isolation and Analysis of ALG5..................................................................................76 Isolation and Identification of ALG5..................................................................76 Sequence Analysis of ALG5................................................................................77 Creation of ALG5 Isogenic Mutants...........................................................................77 Disruption of ALG5.............................................................................................77 Re-insertion of URA3 at its Original Locus........................................................79 Phenotypic Analysis of Alg5 Null Mutant..................................................................80 Cell Wall Changes of alg5 Null Mutant..............................................................81 ALG5 KO phenotype in response to calcofluor white and SDS..................81 Susceptibility of ALG5 null mutant to aminoglycosides..............................82 Evaluation of ALG5 role in Pathogenesis...................................................................83 Cell Adherence Assays as Cell Models for OPC................................................83 Murine Model of DC...........................................................................................83 6 CHARACTERIZATION OF C. ALBICANS IPF15632.............................................85 Identification and Sequence Analysis of C. albicans IPF15632................................85 Isolation and Identification of IPF15632.............................................................85 Sequence Analysis of IPF15632..........................................................................85 Homologies and domains identified using BLAST.....................................85 Domains identified using SMART and PSORTII........................................86 Analysis of IPF15632 Alleles.............................................................................88 Identification of a full sequence for the so-called “false” orf6.961 allele....89 Analysis of IPF15632 on assembly 19........................................................92 Creation of IPF15632 Isogenic Null Mutant and Revertant......................................93 Disruption of Both Alleles of IPF15632.............................................................93 Re-insertion of URA3 Back into its Native Locus...............................................94 Creation of IPF15632 Revertant by Re-inserting One Copy of IPF15632.........95 Re-insertion of one copy of IPF15632 back at its own locus......................95 Confirmation that revertant is expressing IPF15632...................................96 Phenotypic Analysis of Diploid IPF15632 Null Mutant............................................98 Growth Rate and Hyphae Formation...................................................................98 Single Colony Growth Rate on Solid Media.......................................................99 Growth under embedded agar conditions.....................................................99 Growth on surface of solid media..............................................................101 IPF15632 Null Mutant has No Cell Wall Defect..............................................101 Evaluation of IPF15632 Role in Pathogenesis.........................................................104 Adherence to Primary Human Buccal Epithelial Cells.....................................104 IPF15632 Role in a Murine Model of DC........................................................105 Disseminated Candidiasis Tissue Burden Study...............................................106 Antigenicity of IPF15632p................................................................................108 Creation of a Strain Triploid for IPF15632..............................................................110 Phenotypic Analysis of triploid 1 KO IPF15632.....................................................111 Duplication on the Triploid 1 KO Mutants Caused Cell Wall Changes...........111 ix

PAGE 10

In Vitro Cell Adherence Assay..........................................................................115 Evaluation of Triploid 1 KO Strain Role in Pathogenesis........................................116 Stability of Duplication.............................................................................................117 7 DISCUSSION AND CONCLUSIONS....................................................................120 Identification of C. albicans Genes Expressed during OPC using IVIAT...............121 IVIAT Advantages and Limitations..................................................................121 RBF1 and CDC24: Known Virulence Factors Expressed During OPC............122 IPF11959 is Expressed by C. albicans Hyphae During OPC...........................123 Possible Reasons for Why Genes Identified by IVIAT are also Expressed In Vitro...........................................................................................................124 Identification of C. albicans Antigenic Intracellular Proteins with Sera from Patients with OPC..........................................................................................125 ALG5 and IPF15632..........................................................................................126 Characterization of C. albicans ALG5......................................................................126 ALG5 and its Role in C. albicans Cell Wall......................................................128 ALG5 Role in Glycosylation.............................................................................129 ALG5 Has No Role in Adherence to Human Epithelial Cells or in Pathogenicity of DC.......................................................................................130 Characterization of C. albicans IPF15632...............................................................132 IPF15632 and Cell-Surface Adhesion...............................................................133 IPF15632 Role in Pathogenesis........................................................................136 Triploid Strain at IPF15632 Locus...........................................................................137 Concluding Remarks................................................................................................139 LIST OF REFERENCES.................................................................................................140 BIOGRAPHICAL SKETCH...........................................................................................151 x

PAGE 11

LIST OF TABLES Table page 2-1. List of C. albicans wild type and parent strains......................................................18 2-2. List of ALG5 mutants..............................................................................................19 2-3. List of IPF15632 mutants.......................................................................................19 2-4. List of triploid IPF15632 mutants...........................................................................19 2-5. List of bacterial strains............................................................................................20 2-6. List of vectors used for cloning...............................................................................20 2-7. Primers used for IPF1532 mRNA amplifications by RT-PCR...............................28 2-8. List of genes and their primers for subcloning into pET30Ek/LIC.........................36 2-9. Primers used for analyzing gene expression in vivo and in vitro by RT-PCR........37 2-10. List of genes and their primers for cloning into pMB-7.........................................39 2-11. List of genes analyzed by Southern blot and conditions.........................................41 2-12. List of genes and primers for cloning into pMB7-1................................................42 3-1. Description of C. albicans genes identified by IVIAT screening...........................60 5-1. ALG5 recombinants.................................................................................................79 6-1. IPF15632 recombinants..........................................................................................95 6-2. IPF15632 DC fungal burden ...............................................................................108 6-3. Maintenance of duplication in the triploid 1 KO strain........................................119 xi

PAGE 12

LIST OF FIGURES Figure page 1-1. Oropharyngeal candidiasis........................................................................................4 2-1. Cloning vectors.......................................................................................................21 3-1. IVIAT steps............................................................................................................55 3-2. Immunoblots screened with pre-adsorbed sera from patients with OPC................57 3-3. Antigenicity of RBF1 and IPF11959 to sera from patients with OPC....................59 3-4. In vivo and in vitro expression of IPF11959 and RBF1..........................................63 3-5. Reactivity of mouse polyclonal antibody against IPF11959 protein fragment.......64 3-6. Expression of C. albicans IPF11959 within pseudomembranes from HIV-infected patients with OPC.............................................................................65 4-1. Southern blot analysis for genes identified using IVIAT........................................69 4-2. URA blaster protocol..............................................................................................71 4-3. Disruption strategy for ALG5..................................................................................72 4-4. Disruption strategy for IPF15632...........................................................................73 5-1. ALG5 ORF within contig10161. ALG5 is shown as a purple arrow.......................77 5-2. Southern blot with ALG5 disruption mutants..........................................................78 5-3. Re-insertion of URA3 into its native locus..............................................................80 5-4. Growth curve of alg5 mutants in YPD broth..........................................................81 5-5. Phenotypic analysis of alg5 null mutant with cell wall effectors...........................82 5-6. Alg5 null mutant susceptibility to G418..................................................................83 5-8. Murine model of DC for alg5 mutant.....................................................................84 xii

PAGE 13

6-1. Sequence analysis of IPF15632...............................................................................88 6-2. IPF15632 alleles in Assembly 6.............................................................................89 6-3. IPF15632 alleles in C. albicans clinical isolates....................................................91 6-4. Identification of a complete sequence for orf6.961 allele.......................................92 6-5. IPF15632 ORF within contig10177........................................................................93 6-6. Creation of isogenic mutants of IPF15632.............................................................95 6-7. Creation of IPF15632 revertant..............................................................................97 6-8. Growth curve for IPF15632 null mutant.................................................................98 6-9. IPF15632 hyphae length in M199..........................................................................99 6-10. Colony growth rate for IPF15632 null mutant in embedded agar conditions......100 6-11. Single colony growth rate on YPD agar for IPF15632 null mutant.....................102 6-12. Single colony growth on YPD agar for strains grown to stationary phase on YPD media at 37 C..............................................................................................103 6-13. Phenotypic analysis of IPF15632 null mutants with cell wall effectors...............104 6-14. IPF15632 mutant adherence to primary human buccal epithelial cells................105 6-15. DC murine model for IPF15632...........................................................................106 6-16. Tissue burden study for DC murine model of IPF15632 null mutant..................109 6-17. Triploid IPF15632 strain.......................................................................................112 6-18. Changes in the triploid mutants cell wall..............................................................114 6-19. Susceptibility to G418 by the triploid 1 KO mutant.............................................115 6-20. Triploid strain adherence defect to human epithelial and endothelial cells..........116 6-21. Murine models of candidiasis for triploid strains..................................................118 xiii

PAGE 14

LIST OF ABBREVIATIONS Ab antibody ABC avidin-biotin complex AIDS acquired immunodeficiency syndrome Amp ampicillin ATCC American Type Culture Collection BECs buccal epithelial cells BLAST Basic Local Alignment Search Tool bp base pair BSA bovine serum albumin CandidaDB Candida genome database CDC cell division cycle CI choloroform: isoamyl alcohol CFU colony forming unit CHEF contour-clamped homogeneous electric field contig contiguous map cpm counts per minute CR complement receptor CW calcofluor white DB database DC disseminated candidiasis DMSO dimethyl sulfoxide DNA deoxyribonucleic acid dATP deoxyadenosine triphosphate dNTPs deoxyribonucleotides DTT dithiothreitol ECL enhanced chemiluminescence EDTA ethylenediaminetetraacetic acid EF1 elongation factor 1 beta ELISA enzyme linked immunosorbent assay ER endoplasmic reticulum ExPASy Expert Protein Analysis System 5-FOA 5-fluoroorotic acid F1 fragment 1 F2 fragment 2 FaDu pharyngeal epithelial cell line FBS fetal bovine serum g grams G418 Geneticin GDP guanosine diphosphate xiv

PAGE 15

GI gastrointestinal HEPES N-(2-Hydroxyethyl)piperazine-N'(2-ethanesulfonic acid) HIV human immunodeficiency virus HT-29 colonic epithelial cell line HUH hisG-URA3-hisG cassette HWP hyphal wall protein ICR outbred mice Ig immunoglobulin IPF individual protein file IPTG isopropyl-beta-D-thiogalactopyranoside IVET in vivo expression technology IVIAT in vivo induced antigen technology J joules Kan kanamycin kb kilobase kDa kilodalton KO knockout LB Luria Bertani MAT mating type Mbp megabase pairs MEM minimal essential media NCBI National Center for Biotechnology Information NEB New England Biolabs OMP orotidine-5’-monophosphate decarboxylase OPC oropharyngeal candidiasis ORF open reading frame PAGE polyacrylamide gel electrophoresis PBS phosphate buffered saline PBS-T phosphate buffered saline Tween-20 PCI phenol: chloroform: isoamyl alcohol PCR polymerase chain reaction PLB phospholipase B PMNs polymorphonuclear cells PSORTII Protein sorting signals and localization sites in eukaryotic sequences PTH proline transporter helper RBF RPG box binding factor RE restriction endonuclease RNA ribonucleic acid RPG ribosomal protein genes RT reverse transcriptase SAP shrimp alkaline phosphatase SAPs secreted aspartyl proteinases SD Saboraud dextrose SDS sodium dodecyl sulfate SGD Saccharomyces genome database SGTC Stanford Genomics Technology Center xv

PAGE 16

SMART Simple Modular Architecture Research Tool spp. species SSC salinated sodium citrate ssDNA salmon sperm DNA Sub subclone TBE Tris borate EDTA TE Tris-EDTA Tris trishydroxymethylaminomethane UV ultraviolet light VVC vulvovaginal candidiasis wt wild type YPD yeast-peptone-dextrose media YPGal yeast-peptone-galactose YPGly yeast-peptone-glycerol xvi

PAGE 17

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy IDENTIFICATION AND CHARACTERIZATION OF POTENTIAL VIRULENCE FACTORS EXPRESSED BY CANDIDA ALBICANS DURING OROPHARYNGEAL CANDIDIASIS IN HIV-INFECTED PATIENTS By Mary Ann Checkley May 2005 Chair: Dr. Alfred Lewin Cochair: Dr. M. Hong Nguyen Major Department: Molecular Genetics and Microbiology Candida albicans is an important human fungal pathogen able to cause a wide variety of infections. We applied an antibody-based screening strategy, in vivo induced antigen technology (IVIAT), in order to identify potential C. albicans virulence factors expressed during oropharyngeal candidiasis (OPC) in HIV-infected patients. Pooled sera from 24 HIV-infected patients with OPC was depleted for antibodies against C. albicans proteins expressed in culture. This adsorbed sera was used to screen a genomic expression library consisting of 24 C. albicans clinical isolates. I identified five C. albicans antigenic proteins by screening the genomic expression library with the adsorbed sera. These proteins (RBF1, CDC24, IPF11959, ALG5, and IPF15632) were analyzed in this thesis. Genes encoding the antigenic proteins were further shown to be expressed within pseudomembranes from thrush samples. Both RBF1 and CDC24 are involved in the yeast-to-hyphae transition and have been previously identified as xvii

PAGE 18

virulence factors. PTH1 encodes a hypothetical protein with homology to S. cerevisiae YGR046w, an essential gene. ALG5 encodes a putative dolichol-phosphate glucosyltransferase involved in N-glycosylation of cell wall proteins. IPF15632 has no homology with any protein in GenBank or the S. cerevisiae database. I further characterized ALG5 and IPF15632 role in C. albicans pathogenesis. I found that ALG5 does not contribute to adherence to primary human buccal epithelial cells nor does it play a role in a murine model of DC, in spite of minor cell wall changes due to under glycosylation of cell wall proteins. However, mutants with disruption of IPF15632 produced small colonies on YPD solid media compared to the wild type and their virulence was attenuated in the murine model of DC. IPF15632 mutants were able to colonize mouse kidneys, spleens, and liver as well as the wild type 1 day post-infection. However, the mutant cells were cleared from the kidneys and spleens faster than the wild type 4 days post-infection, suggesting their elimination by the immune system. Furthermore, I created a triploid C. albicans strain that contains a duplication of all or part of chromosome 6. This strain has cell wall defects that interfere with virulence in a murine model of OPC but not of DC. xviii

PAGE 19

CHAPTER 1 INTRODUCTION Fungal infections have been known for centuries, but their clinical relevance has not been recognized until the last century. In the past three decades, there has been an increase in fungal infections, especially those caused by Candida, Aspergillus, and Cryptococcus neoformans. This increase is associated with recent advances in medicine that have prolonged the life expectancy and improved the lifestyle of individuals with debilitating diseases. However, these medical breakthroughs have also prolonged the survival of profoundly immunocompromised patients susceptible to microbial infections that could be fatal. (147, 156) Among the fungal pathogens, Candida is by far the most common cause of human infections. (41, 137, 180) Candida albicans The Candida genus belongs to the Ascomycetes phylum and has approximately 200 species (133). So far, 13 species are considered to be pathogenic, of which Candida albicans is the most pathogenic (1). C. albicans is a dimorphic organism since it can exist in either the yeast form or one of two filamentous forms, hyphae or pseudohyphae. The two filamentous forms differ in the way they are formed (12). A sexual cycle has recently been discovered for C. albicans, but its properties are not yet clearly defined (84, 164, 179). Most laboratory strains are unable to mate since they are heterozygous at the mating type (MAT) locus (100, 104). C. albicans has a diploid genome of 32 megabase pairs and eight homologous chromosomes, which range from about 1 to 3.2 kbp. The genome of the SC5314 strain of C. albicans used in this project has recently been 1

PAGE 20

2 sequenced. Approximately 6000 ORFs have been identified and 64% of these match with genes from Saccharomyces cerevisiae, a close relative of C. albicans (89). Candida is a member of the endogenous microbial flora of healthy individuals, found in the buccal cavity, gastrointestinal (GI) tract, and genitourinary tract. C. albicans is the most common species isolated from these sites. Invasive growth is normally suppressed by other microorganisms of the normal microbial flora and by an intact host immune system. When the host barrier or immune system are disrupted by one of many possible mechanisms, C. albicans is able to proliferate and invade the host, and possibly disseminate. This could occur, for example, by the use of broad spectrum antibiotics that disrupt the balance of the microbial flora, damage to the integument or mucosa by abdominal surgery, introduction of catheters or intervascular devices, or by having a congenital or an acquired immunodeficiency, such as AIDS. (156) Candidiasis C. albicans can cause a wide variety of infections, each of which is commonly referred to as candidiasis (17). Candidiasis can range from self-limiting cutaneous and mucocutaneous infections, such as on the skin (e.g. diaper rash), nail, or mucosal membranes of the mouth and vagina, to fatal deep-seated and disseminated infections involving various organs and tissue like the lung, kidney, reticuloendothelial system, and central nervous system. (133-135, 141) Deep-seated and Disseminated Candidiasis (DC) C. albicans is the fourth leading cause of nosocomial bloodstream infections. (65) Individuals at risk for DC are those with polymorphonuclear cell (PMNs) defects, such as patients with prolonged neutropenia due to bone marrow transplant or chemotherapy (particularly for leukemia). In neutropenic patients, DC starts either as a contamination of

PAGE 21

3 indwelling catheters or translocation of endogenous C. albicans from the GI tract. In these cases previous colonization by C. albicans is also a risk for candidemia (Candida in the blood). Once C. albicans has reached the bloodstream, virtually any organ or body site can be infected. Systemic infections are fatal if left untreated, and, even when patients are receiving the appropriate antifungal therapy, the mortality rate is as high as thirty-five percent (183). Furthermore, there are no specific diagnostic tests for the rapid detection of DC. Mucocutaneous Candidiasis The most common candidal diseases are the mucocutaneous infections, such as oral and vaginal candidiasis. Oropharyngeal candidiasis (OPC) C. albicans is a commensal organism of the mouth, where carriage rates in the general asymptomatic population range from 20% to 75%. However, oral candidiasis or OPC is the most common human fungal infection, especially in the extremes of life (early and late) because of reduced immunity. Lesions present are due to the conversion of the benign colonizer to the pathological overgrowth of C. albicans. (2) OPC can manifest itself as a variety of clinical lesions, such as acute pseudomembraneous candidiasis (oral thrush), erythematous and hyperplastic candidiasis, and angular chelitis. Thrush, the most common form of OPC, is readily diagnosed clinically and is characterized by the presence of white/yellow plaques on the surface of the buccal mucosa, palate, tongue, gum linings, and oropharynx. When the plaques are removed with a tongue blade they reveal a red bleeding surface. These lesions are also highly specific to Candida infection. The other forms of OPC are more

PAGE 22

4 difficult to diagnose and can often be confused with lesions due to the presence of other pathogens, medications, or other causes. (38) Clinical significance of OPC. Clinically, OPC reduces the quality of life because its painful lesions alter the sense of taste and impair chewing and swallowing, thus restricting the intake of food and medications, slowing recovery, and prolonging hospital stay. The white plaques are also aesthetically displeasing. (2) (See Fig. 1-1) Fig. 1-1. Oropharyngeal candidiasis. Picture of a baby with thrush, manifested by white plaques on the tongue and lips. Diseases and factors that contribute to OPC. Populations most susceptible to contracting OPC are individuals with congenital or acquired immunodeficiencies (particularly T cell deficiencies), individuals using broad-spectrum antibiotics, patients who have leukemia or are undergoing chemotherapy, neonates, diabetics, smokers, denture wearers, and patients who are terminally ill. OPC is also the most common infectious complication of patients with acquired immunodeficiency syndrome (AIDS), and it is well recognized as both an indicator of HIV infection and a predictor for AIDS progression. (38, 92, 148, 152)

PAGE 23

5 Other factors that predispose one to OPC are impaired salivary gland function (e.g. in individuals with Sjogren’s syndrome), inhaled steroids, dentures, and oral cancer. Saliva, by means of its secretion, not only removes organisms from the mucosa but also contains antimicrobial proteins, such as lactoferrin, sialoperoxidase, lysozyme, histidine-rich polypeptides, and specific anti-Candida antibodies that prevent overgrowth of Candida. Inhaled steroids can also suppress cellular immunity and phagocytosis locally, which is required to keep commensal organisms in check. Dentures predispose one to OPC due to the reduced flow of saliva under surfaces of the denture, the propensity for Candida to adhere to acrylic, and the general state of poor oral hygiene that often accompanies the use of dentures. Complications with treatment of OPC. The current methods to treat OPC include topical or systemic antifungal drugs, mainly amphotericin B and azoles. Despite the fact that these drugs ameliorate the disease they do not appear to eradicate colonization, thus patients tend to suffer from recurrent episodes of OPC. Treatment of OPC is further complicated by drug toxicity and the recent emergence of antifungal-resistant strains of C. albicans. (123, 150) Vaginal candidiasis C. albicans is the main cause of vulvovaginal candidiasis (VVC), which affects a significant number of women. It is known that up to 75% of all women will experience at least one episode of VVC in their life span. Predisposing factors are antibiotic and oral contraceptive usage, hormone replacement therapy, pregnancy, uncontrolled diabetes mellitus, immunosuppressive therapy, and possibly HIV infection. In a small subset of women (5-10%) recurrent episodes of VVC can occur and it is postulated to be due to an immune deficiency. (46, 103, 163)

PAGE 24

6 C. albicans Virulence In addition to the host immune status, expression of virulence factors by C. albicans also contributes to its transition from a commensal organism to an opportunistic pathogen. The fact that candidiasis typically occurs in an immunocompromised host does not diminish the importance of the organism’s ability to specifically express virulence factors during the disease process. In fact, gene deletions of specific virulence factors have resulted in attenuated virulence or avirulence of C. albicans in different animal models of candidiasis. Recently, several experts in the field have defined the term “virulence factor” as “a component of a pathogen that damages the host” (20, 21, 145) and as a “factors that interact directly with mammalian host cells” (136). C. albicans is not only a versatile organism that is able to persist at different anatomical sites but also a successful pathogen that adapts to a range of physiological pressures exerted at different sites within the host (17). Virulence Factors Pathogenesis of C. albicans has been attributed to the expression of a variety of virulence factors, which allow it to adhere to host tissues, survive and grow at different body sites, cause damage, evade host immunity, and spread to other tissues. (33, 70, 130) Dimorphic transition In order for C. albicans to become pathogenic it must undergo a dimorphic transition from the commensal yeast form to the more invasive filametous form. In the immunocompetent host the commensal yeast form is kept under control (i.e., in low numbers and at permitted body sites) by the immune system, while in the susceptible host, the filamentous forms are capable of invading tissue and expressing a unique set of antigens. (22) However, the presence of C. albicans in the filamentous form alone is not

PAGE 25

7 enough for virulence, since both forms appear to be required for full virulence in murine models of DC (120) . The transition itself as well as co-ordinate gene expression allow C. albicans to colonize the host and cause disease. Phenotypic switching Phenotypic switching refers to the ability of C. albicans to change phenotypically, usually observed experimentally by colony color and morphology. This phenomenon has also been observed in freshly isolated clinical strains, some of which appear to have higher frequencies of switching. The phenotypic switch most studied is the white-opaque switch in the WO-1 strain (162). In a cutaneous model, opaque cells are able to colonize skin better than white cells. However, in systemic animal models, opaque cells are less virulent than white cells.(95) Phenotypic switching contributes to C. albicans plasticity by providing changes in adherence properties, antigen expression, enzyme production, tissue affinity, and drug sensitivity (1). These properties could allow Candida to adapt to specific body sites and conceivably develop drug resistance and evade the immune system. Adherence C. albicans is able to adhere to a variety of tissues and inanimate surfaces. The ability to adhere to host tissues is crucial during the early stages of colonization and tissue invasion. For example, adherence of C. albicans to host buccal epithelial cells is one of the first essential events necessary for colonization of the oral mucosa and pathogenesis of OPC. Adherence to buccal epithelial cells prevents the yeast from being washed out by saliva or by the normal sloughing of epithelia cells.(180) Several proteins, such as Hwp1 (hyphal wall protein) (166, 177), Int1 (integrin-like protein) (53, 54) , and Mnt1 and Mnt2 (mannosyltransferase) (126), have been shown to be required for

PAGE 26

8 adhesion to different cells in vitro and for pathogenesis of both OPC and DC murine models. Secretion of proteases The most important extracellular hydrolytic proteinases secreted by C. albicans are the secreted aspartyl proteinases (SAPs), phospholipase B (PLB) enzymes, and lipases (128). These are grouped into large protein families, of which the SAPs family has been the most extensively studied. The SAPs family consists of 10 genes (SAP1-10), which are differentially expressed according to the site and stage of Candida infection (121, 127). Furthermore, SAPs are also differentially expressed by the yeast and hyphal forms (80). SAPs contribute to virulence in a variety of ways. They facilitate adherence to many host tissues and cell types by modifying target proteins or ligands on either the fungal surface or the host cells . For example, SAP2 can efficiently degrade laminin, fibronectin, and mucin, all of which are components of the extracellular matrix and epithelial cells. The production of SAPs by hyphae allow it to penetrate and invade tissues by digesting and distorting host cell membranes. SAP2 can hydrolyze many humoral host defense proteins, such as secretory IgA, thus allowing the yeast to evade the host immune system. (128) Phospholipases have also been implicated in host cell penetration. PLB1 null mutants were attenuated in a murine model of DC (98), and they were found to be less invasive in an oral-intragastric infant mouse-model (125). Evasion of the immune system C. albicans is able to express integrin-like proteins, iC3band C3d-binding proteins. Both iC3b and C3d are complement receptors (CR) (3, 91). iC3b is a ligand of the alternative complement fixation pathway, and C3d is a ligand of complement receptor type 2 (CR2) of B lymphocytes. C3d –binding proteins are located on the surface of

PAGE 27

9 hyphal cells. The CR-like proteins of C. albicans could provide an escape mechanism based on molecular mimicry of host proteins. (42, 62) C. albicans is also able to escape phagocytosis by macrophages (176). It has been shown with in vitro studies that once the yeast form of C. albicans is phagocytosed by cultured macrophages it differentiates into the filamentous hyphal form and by doing so it breaks out of the macrophage and continues to grow. (85) Thus, pathogenesis of C. albicans is not only dependent on the physiological abnormalities of the host immune system but also on the expression of virulence factors that take advantage of the specific deficiencies of the host immune response. (33) Tissue-Specific Virulence Each site of infection, such as the oral cavity, vagina, and blood, is a different niche and exerts unique environmental pressures. Because C. albicans is able to survive and proliferate in diverse tissues, specific virulence factors are likely to be important during different types and stages of infection. (50, 146) C. albicans possesses gene families that encode virulence factors. It is known that C. albicans virulence factors, such as the SAPs, are differentially expressed according to the environmental pressures exerted by particular anatomical sites. (11, 82, 167) Several studies have shown that members of the SAPs family are expressed during different stages and types of candidiasis. SAPs 1-3 have been demonstrated to be important for C. albicans infection of artificial oral mucosa as well as in an animal model of vaginal candidiasis, while only SAP2 was required for the initial stages of endothelial cell injury. On the other hand, SAPs 4-6 were found to be essential for DC in rodent models of intravenous (i.v.) infection and for organ invasion after an intraperitoneal (i.p.) infection. (1)

PAGE 28

10 Work by Bernardis et al. with PHR1 and PHR2, which are pH regulated genes involved in cell wall synthesis, have demonstrated that the site of infection provides the signals that regulate C. albicans gene expression and virulence. PHR1 expression is optimal under neutral conditions such as those in the blood (153); consequently the PHR1 null mutant is avirulent in a murine model of DC (34, 55). Conversely, PHR2 functions under acidic conditions such as those found in the vagina (124), and the PHR2 null mutant is avirulent in a rat model of VC (34). These stage and site-specific differences imply that certain virulence factors can confer a temporal and/or tissue-specific virulence. In Vivo Expression of Virulence Factors All organisms, including pathogens, are known to respond to environmental stimuli by regulating gene expression in order to optimize their survival and growth conditions. The identification and expression of many known virulence factors by pathogenic bacteria have been performed under laboratory conditions that try to mimic those conditions during infection. (108, 118) However, the milieu present under standard laboratory conditions cannot take into account the complex network of signals present during an infection within the host. For example, such studies cannot take into account how the infection environment alters cell density-sensing mechanisms. Furthermore, several in vitro expression studies after did not accurately predict a role for some bacterial virulence factors during in vivo conditions (106, 139). In the past two decades, the use of in vivo expression technologies (IVETs) has allowed the identification of genes expressed within infected animals, some of which are novel virulence factors. Moreover, some of the virulence factors expressed in vivo have included new targets for therapeutic and vaccine strategies. (131, 139, 144)

PAGE 29

11 C. albicans Genes Identified using IVETs Several IVETs have been adapted and developed lately to identify C. albicans genes expressed in vivo or ex vivo, such as differential display RT-PCR, antibody-based screening strategies, and transcriptional profiling using microarrays. Differential display RT-PCR techniques Prigneau et al. used differential display RT-PCR to identify C. albicans genes that are differentially expressed during phagocytosis by macrophages. They identified seven genes involved in metabolic processes. Among the genes identified, four encoded peroxisomal proteins with a role in the glyoxylate cycle. (140) Zhao et al. used differential display to compare C. albicans genes expressed during infection in the rat oral cavity (in vivo) with growth in culture (in vitro). They identified two fragments that were upregulated during rat oral infection. One of the fragments shared homology to a gene that belonged to the C. albicans ALS family, which encode cell surface glycoproteins that have a role in adherence to host surfaces. The other fragment belonged to a C. albicans gene whose function is unknown but has homology to a bacterial gene (Bacillus subtilis gidB) possibly involved in cell division. (186) Antibody-based screening strategies The first publication using an antibody screening method for C. albicans was by Swoboda et al.. They screened a hyphae library with sera pooled from five patients with oral or esophageal candidiasis and five healthy individuals. Their reasoning for using sera from healthy individuals is because most people have been exposed to C. albicans during their lifetime, and these individuals could possibly have protective antibodies against it. The researchers identified C. albicans glycolytic enzymes as immunogens, but not every patient’s sera recognized them as immunogens. (172)

PAGE 30

12 Several other laboratories have also screened cDNA libraries, created with RNA extracted from C. albicans hyphae form, with sera from patients with DC and neutropenic individuals with high levels of anti-C. albicans IgM antibodies (57). Also, cDNA libraries from yeast and/or germ tubes (yeast starting to form hyphae) alone, were screened using sera raised in animals against C. albicans cell wall components or cytoplasmic proteins (lysates). The purpose of these experiments was to identify antigenic proteins expressed by the particular morphological form of C. albicans. (4, 5, 44, 101, 158). In order to discover biomarkers for diagnosis and therapeutic monitoring of DC, Pitarch et al. used serum from patients with systemic candidiasis to identify immunogenic proteins. They identified 85 immunogenic proteins by immunoproteonomics, which used two-dimensional electrophoresis followed by Western blotting. Among the immunogenic proteins identified there were chaperones and heat shock proteins, metabolic enzymes (carbohydrate metabolism, glycolysis, fermentation, tricarboxylic acid cycle, and fatty acid metabolism), enzymes involved in energetic central metabolism, and translational apparatus proteins. They found that only four proteins (enolase, phosphoglycerate kinase, alcohol dehydrogenase, and pyruvate carboxylase) were also recognized by sera from healthy individuals. However, the reactivity from the sera of the latter group was considerably lower than that from patients with DC. Furthermore, they found that a specific pattern for the reactivity to the immunogenic proteins is detected in patients who responded to antifungal treatment, characterized by an increase in reactivity to enolase, compared to those who did not. (138)

PAGE 31

13 Transcriptional profiling using microarrays Fradin et al. used genomic arrays and cDNA subtraction methods to analyze the transcriptional profile of C. albicans exposed to heparinized fresh human blood for period of 60 minutes. They also compared C. albicans gene expression during incubation in blood and plasma (without host cells). The investigators observed differential expression of genes involved in the general stress and antioxidant response, glycolysis, glyoxylate cycle, and putative virulence properties. For example, during the first 10 minutes of exposure to blood, there was a high level of expression of C. albicans genes involved in protein synthesis, which probably allowed C. albicans to respond quickly to the conditions encountered in the blood. Transcription of genes encoding glycolytic enzymes was downregulated during the early stages. The authors speculated that this was due to the high levels of glucose in the blood or to internal stocks of carbohydrates accumulated during culture conditions prior to incubation with blood.. However, in the later stages (30-60 minutes) of exposure to blood, they observed an upregulation of both genes encoding enzymes involved in glycolysis and the glyoxylate cycle. These results suggested that two populations of cells existed: cells that were phagocytosed upregulated genes involved in the glyoxylate cycle during growth in phagolysosomes that lack glucose, and cells that had access to glucose in the blood. (51) Lorenz et al. studied the transcript profile of C. albicans after phagocytosis by mammalian macrophages (cultured cell line). During the early phase, inside macrophages, C. albicans upregulated genes involved in gluconeogenesis/glyoxylate pathways, the oxidative stress response, and morphogenesis, while genes involved in translation were repressed. During the late phase, concomitant with the yeast to hyphal switch and escape from macrophages, there was a restoration of the translational

PAGE 32

14 machinery, activation of glycolysis, and downregulation of stress response. (102) These results agree with similar findings in another study with C. albicans ingested by neutrophils (149). In Vivo Induced Antigen Expression (IVIAT) IVIAT was designed by J. Hillman, M. Handfield, A. Progulske-Fox, and J. Brady from the Department of Oral Biology at the University of Florida, to identify genes expressed by the oral pathogen Actinobacillus actinomycetemcomitans during the course of localized juvenile periodontitis (68). IVIAT uses antibodies from the sera of patients infected with a particular microbe to screen a genomic expression library made from the infecting microbe. (19, 68) IVIAT for C. albicans The laboratory of M. Hong Nguyen (Department of Infectious Diseases, Veterans’ Administration Hospital, Gainesville, Florida) has adapted an antibody based screening strategy, IVIAT, to identify antigens expressed by C. albicans during OPC of HIV-infected patients (27). Sera from HIV-infected patients with OPC. Asymptomatic colonization of the oral mucosa by C. albicans is a common finding in HIV-infected patients. Carriage rates are greater than 80% in HIV-infected individuals with CD4+ cell counts less than 500 cells per microliter. Over 90% of HIV-positive individuals will develop symptomatic OPC during the course of their disease. (48, 178, 181) The main host defenses against disseminated candidiasis are neutrophils and macrophages, which are able to contain the yeast in the superficial layer of the oral mucosa. Interestingly, individuals with HIV infection or AIDS are mainly susceptible to oral candidiasis but not to disseminated candidiasis, indicating that impaired cellular

PAGE 33

15 immunity is a risk factor for oral candidiasis. These patients maintain normal neutrophil function despite a decline in both macrophage function and the number of T-helper cells. The result is an overgrowth of colonizing yeast in the oral mucosa in individuals with low CD4+ cell counts. (86, 110, 147, 181) It has been demonstrated that HIV-infected patients colonized with C. albicans have systemic antibodies against this yeast. Compared to healthy controls, HIV-positive patients with oral candidiasis have been found to have higher levels of IgG against C. albicans in serum and saliva, as well as, higher levels of IgA1 and IgA2 subclasses in whole and parotid saliva. (24, 25, 30, 39, 184) Antibodies against specific C. albicans proteins, such as hsp90 and SAPs, have been identified in patients with HIV and OPC. Higher levels of IgA, IgM, and IgG against C. albicans Sap1 and Sap6 proteins were found in both serum and saliva of HIV-infected patients with oral candidiasis compared to those without oral candidiasis and an HIV-negative healthy control (39). Moreover, patients with AIDS and OPC, who had antibodies to an immunodominant 47 kDa subcomponent of C. albicans hsp90 did not develop DC. On the other hand, those patients with DC who did not have or seroconverted to anti-47 kDa antibodies eventually died. (114-116) These results provide evidence for a role of humoral immunity against disseminated candidiasis. Furthermore, they suggest the serum of patients with HIV-infection and OPC contains sufficient antibodies against C. albicans antigens to enable the identification of antigenic C. albicans proteins expressed during candidiasis. Heterogeneity in C. albicans clinical isolates. Usually individuals are colonized by a single strain at birth and the source for those who develop candidiasis is mostly

PAGE 34

16 endogenous. C. albicans primary mean of reproduction is clonal, but mating which, occurs in nature and not in the laboratory, can contribute to diversity to a lesser degree. It is known that C. albicans clinical isolates have considerable genomic heterogeneity and variability in genome size. (12, 105, 119). Clinical isolates have been found to differ in colony morphology (74) and these have been shown to be due to difference in gene expression (75, 185). There also appears to be a correlation between colony morphology, chromosome reorganization, and virulence in C. albicans clinical isolates (171). Therefore, it is important to study gene expression in different C. albicans clinical isolates, since these appear to differ in phenotype and probably virulence. Library construction and sera used for IVIAT screening Construction of C. albicans genomic expression library for IVIAT. The C. albicans genomic expression library was created from 24 C. albicans clinical isolates from patients with OPC. A large number of isolates was chosen to create the library in order to maximize the number of genes identified by IVIAT. (27) Sera used for screening the C. albicans library. The main tool in our approach to screen the C. albicans genomic expression library was sera collected from 24 HIV-infected individuals with active oral thrush (OPC). The clinical isolates used for creating the library were not obtained from these patients. Sera from a large number of HIV-infected individuals with OPC was collected to increase the population of diverse antibodies elicited by C. albicans during OPC and to compensate for the inherent heterogeneity in the immune response and any possible antibody deficiencies present in the individuals chosen. Furthermore, the 24 HIV-infected individuals with active oral thrush were probably infected with different isolates of C. albicans, which may each differ in the expression and sequence of some of their genes. Thus, this diverse antibody

PAGE 35

17 population should also assist in the identification of some genes expressed by particular C. albicans clinical isolates. Our Goals Long term goals of our laboratory include the identification of genes expressed during OPC and to determine their role during pathogenesis. Eventually, networks of gene expression will also be identified, since a single factor cannot account for Candida virulence. Rather, understanding a panel of virulence factors is required to elucidate the host-C. albicans interactions and pathogenesis. We hypothesize that certain genes expressed by C. albicans during an infection encode factors important for virulence. To explore this possibility my objectives were: 1) to identify genes expressed during OPC in HIV-infected patients and 2) To demonstrate that some of the genes identified are important for pathogenesis in murine models of candidiasis. The identification of genes important for virulence during candidiasis will further the understanding of how C. albicans can become such a successful opportunistic pathogen that is able to survive and cause a wide variety of diseases. We hope that by defining new virulence factors we can identify targets for novel therapeutic, vaccines, and diagnostic strategies.

PAGE 36

CHAPTER 2 MATERIALS AND METHODS Strains and Growth Conditions Fungal Strains All the C. albicans strains were normally grown in YPD (1% yeast extract, 1% bactopeptone, 2% D-glucose) medium at 30 C. Stocks were made with 40% glycerol final concentration and stored at -80 C. Wild type and parent strains The following C. albicans parent strains (Table 2-1) were kindly provided by Dr. Fonzi from Georgetown University. The table below provides their names, genotype, and a brief description. Table 2-1. List of C. albicans wild type and parent strains. Strain Genotype Description Origin SC5314 Wild type Strain sequenced and is used for most studies and disruptions Clinical isolate isolated from a patient with DC (58) CAF-2 URA3/ura3::imm434 Heterozygous for URA3, with one copy deleted Derived from SC5314 (49) CAI-4 Ura3::imm434/ ura3::imm434 Auxotroph, lacking URA3 Derived from CAF-2 (49) CAI-12 Ura3::/ura3::imm434 Heterozygous for URA3 with 1 copy re-inserted. This strain is used to compare with KO mutants. Derived from CAI-4 (49) Mutants constructed The following tables (2-2, 2-3, and 2-4) describes the mutants created for ALG5 and IPF15632 during the length of this thesis. Mutants were created in the background of CAI-4. 18

PAGE 37

19 Table 2-2. List of ALG5 mutants Strain name Genotype Strains used in studies 12 ALG5/alg5::hisG-URA3-hisG 12F1 ALG5/alg5::hisG 12F1 + U1 ALG5/alg5::hisG/URA3 Alg5 heterozygous mutant + URA3 12F1/62 alg5::hisG-URA3-hisG/alg5::hisG 12F1/62F4 alg5::hisG/alg5::hisG 12F1/62F4 + U2 alg5::hisG/alg5::hisG/URA3 Alg5 null mutant + URA3 Table 2-3. List of IPF15632 mutants Strain name Genotype Strain used in studies 6 (2003) IPF15632/IPF15632::hisG-URA3-hisG 6F7 IPF15632/IPF15632::hisG 6F7 + U2 IPF15632/IPF15632::hisG/URA3 IPF15632 heterozygous mutant + URA3 6F7/19 IPF15632::hisG-URA3-hisG/IPF15632::hisG 6F7/19F7 IPF15632::hisG/IPF15632::hisG 6F7/19F7 + U1 and U4 IPF15632::hisG/IPF15632::hisG/URA3 IPF15632 null mutants + URA3 Table 2-4. List of triploid IPF15632 mutants Strain name Genotype* for IPF15632 Strains used in studies 6 (2001) D allele A/allele A/allele b::hisG-URA3-hisG 6F2 D allele A/allele A/allele b::hisG 6F2 + U2 D allele A/allele A/allele b::hisG/URA3 1 KO + URA3 6F2/3 D allele A/allele a::hisG-URA3-hisG/allele b::hisG 6F2/3F12 D allele A/allele a::hisG/allele b::hisG 6F2/3F12 + U2 D allele A/allele a::hisG/allele b::hisG/URA3 2 KO + URA3 6F2/3F12/26 D allele a::hisG-URA3-hisG/allele a::hisG/allele b::hisG 6F2/3F12/26F6 Dallele a::hisG/allele a::hisG/allele b::hisG 6F2/3F12/26F6 + U1 Dallele a::hisG/allele a::hisG/allele b::hisG/URA3 3 KO (null mutant) + URA3 *All the strains in this table have a duplication (D) that was originated during the insertion of hisG-URA3-hisG cassette into the first allele of IPF15632 disrupted. The duplication has not yet been assigned to a region.

PAGE 38

20 Bacterial Strains The following Escherichia coli bacterial strains were obtained from Invitrogen. Strains were grown in Luria Bertani (LB) broth (1% Bacto-tryptone, 0.5% yeast extract, and 1% NaCl) at 37 C with the appropriate antibiotic according to the vector’s selection marker (ampicillin 100ug/mL; kanamycin 40 ug/mL). Stocks were made with 20% glycerol final concentration and stored at C. Table 2-5. List of bacterial strains Strain Genotype Use DH5 (E) F'phi80dlacZ delta(lacZYA-argF)U169 deoR recA1 endA1 hsdR17 (rk-, m k+) phoA supE44 lambda-thi-1 gyrA96 relA1/F' proAB+ lacIqZdeltaM15 Tn10(tetr) Use for plasmid transformation BL21 (DE3) E. coli B, F-, dcm, ompT, hsdS(rB-mB-), gall(DE3) Has the T7 RNA polymerase gene integrated in the bacterial chromosome, under control of the lacUV5 promoter, for protein expression with IPTG induction. Used for non-toxic protein expression Vectors The following list of vectors provides a description of their use. These vectors were the starting material for cloning. Vectors were stored in ddH2O at C. Table 2-6. List of vectors used for cloning Vector Use Provider pET30abc Inducible expression vector for protein production, for expression of protein in all 3 frames. Kan resistance. Novagen pET30EK/LIC Inducible expression vector, for in-frame cloning with a His tag epitope at the 3’ end. Kan resistance. Novagen PMB7 Shuttle vector containing HisG/URA3/HisG disruption cassette to disrupt genes in CAI-4 strain by homologous recombination. Amp resistance for selection in bacteria. Dr. Fonzi PMB7-1 Vector derived from pMB7, contains URA3/HisG, for re-inserting a gene into its disruption locus. Dr. Fonzi pURA3 Containing URA3 fragment for re-insertion at its own locus. Amp resistance for selection in bacteria. Dr. Fonzi

PAGE 39

21 pMB76597 bp URA3 AmpR HisG HisG XbaI X baI pMB7-15057 bp HisG URA3 AmpR A)B) pMB76597 bp URA3 AmpR HisG HisG XbaI X baI pMB76597 bp URA3 AmpR HisG HisG XbaI X baI pMB7-15057 bp HisG URA3 AmpR pMB7-15057 bp HisG URA3 AmpR A)B) KpnI PstI SacI SphI SalI KpnI PstI SacI XbaI SphI SalI KpnI PstI SacI SphI SalI KpnI PstI SacI SphI SalI KpnI PstI SacI XbaI SphI SalI KpnI PstI SacI XbaI SphI SalI Fig. 2-1. Cloning vectors. A) pMB7 B) pMB7-1. The HisG direct repeats are shown with green arrows and the selectable marker URA3 is shown with a red arrow. The yellow arrow represents the ampicillin resistance gene. Two multiple cloning sites are indicated with the list of restriction endonucleases. RE XbaI was used to removed one copy of the HisG direct repeat in order to create pMB7-1. DNA, RNA, and Protein Techniques DNA Techniques Polymerase chain reaction (PCR) Small fragment generation. Standard PCRs were used to produce short fragments (up to 2 kb) for either cloning or screening used Taq DNA polymerase (Promega or Sigma). PCRs were performed in a volume of 50 L and included 0.2 mM dNTPs (New England Biolabs (NEB)), 0.1 to 0.4 M of each of primer (forward and reverse), 1.5 to 3.5 mM MgCl2, template DNA (quantity was not standardized), PCR buffer (provided with the enzyme), and Taw DNA polymerase (0.05 units). To amplify PCR products from C. albicans genomic DNA, 0.1X bovine serum albumin (BSA) and 1% DMSO were added to the PCR mix. Standard PCR conditions included a denaturing step of 94 C for 5 minutes, followed by 25-35 cycles of 94 C for 30 seconds to 1 min, Ta C for 30 seconds to 1 min, 72 C for 30 seconds to 1 min, and finally an extension step of 72 C

PAGE 40

22 for 5-10 min (Ta is the annealing temperature which varied according to the primers used). These conditions varied according to the primers and product length. PCRs were done in a Perkin Elmer PCR machine. High fidelity large fragment generation. PCRs for generating sequences of C. albicans genes that were to be reintroduced into its genome used the Expand High Fidelity PCR System (Roche). The Expand High Fidelity PCR System is composed of an enzyme mix containing Taq DNA polymerase and Tgo DNA polymerase, which has proofreading activity. Colony PCR for screening C. albicans recombinants. Single colonies were resuspended in 20 L of dH2O and placed in a boiling water bath for 10 minutes. Resuspended cells were cooled on ice and briefly spun. PCR was done with the supernatant as indicated above for small fragment generation but instead 5 L of colony resuspension were used as template and BSA was added to the reaction. Purification of PCR product. PCR products that were cloned directly into TOPO pCR vector (Invitrogen) did not require any purification step. TOPO cloning was done using the TOPO Cloning Kit (Invitrogen). PCR products cloned into any other vector were purified using molecular weight cutoff filters (Ultrafree-MC, Millipore) or by gel elution using freeze-squeeze or the QIAEX kit (QIAGEN). Freeze-squeeze consisted of crushing the agarose gel containing PCR product with an equal volume of phenol (pH 8.0) and freezing it at C for 1 hour minimum. This was followed by centrifugation at 12,000 x g for 5 minutes, sequential extractions with

PAGE 41

23 equal volumes of PCI and CI, and precipitation (1/10 volume 5 M NH4Cl and 2 volumes ethanol, and placed at C for 1 hour or C for 20 min.). Cloning in bacteria Linearization and dephosphorylation of vectors. Starting vectors were linearized with the appropriate restriction enzymes (RE) (New England Biolabs (NEB), Promega, Sigma, Gibco) in the supplied buffer for 1 to 2 hours. Dephosphorylation of the 5’phosphate on the vector was carried out with shrimp alkaline phosphatase (SAP) (Promega) at 37 C for 1 to 2 hours. SAP was inactivated at 65 C for 15 minutes. Digested DNA was resolved by gel electrophoresis in an agarose gel with 1X Tris/Borate/EDTA (TBE) buffer containing 0.1 g/mL ethidium bromide. Bands of DNA were visualized and excised from the agarose using long wave UV. Ligation. Ligations were carried out in a volume of 10 to 20 L and included 200 ng linearized vector, insert, 0.1 units T4 DNA ligase enzyme (Promega), and 1X buffer (provided with enzyme). Insert to vector molar ratio used was either 1:2 or 1:3. Reactions were carried at 14 to 16 C overnight. Fast ligation reactions were carried out the same way but instead they used a 2X fast ligation buffer (Promega), which speeded up the ligation reaction to 5-15 minutes at room-temperature. Transformation. Most of E. coli DH5 and BL21 (DE3) transformations were done by electroporation (1.5 V, 5 mseconds, 0.25 F, 200 Ohms; 0.1 mm USA Scientific cuvette). Cells were allowed to recover in 1 mL LB for 1 hour at 37 C with gentle shaking, after which, the transformation was then spread on LB agar plates containing the appropriate antibiotic for selection of cells containing plasmids and incubated at 37 C overnight.

PAGE 42

24 Plasmid preparation. LB media containing the appropriate antibiotic was inoculated with individual bacterial clones picked from agar plates. Cells were incubated for 12 to 14 hours at 37 C with vigorous shaking. Plasmid were isolated using a Mini or Maxi Plasmid Prep Kit (QIAGEN, Sigma, or Eppendorf). Plasmids were analyzed by RE digest. DNA sequencing Sequencing of DNA (1 to 2 g) was carried out by the DNA Sequencing Core Facility, Interdisciplinary Center for Biotechnology Research (ICBR), at the University of Florida. Yeast genomic DNA extraction and analysis C. albicans was grown overnight in YPD at 30 C. Cell pellets were stored at -20 C overnight to speed up the DNA extraction process. Genomic DNA was extracted using the Wizard Genomic DNA Purification Kit (Promega), which used 0.6 mg lyticase per tube (20 to 30 minutes at 37 C with shaking at 1200 rpm). DNA was treated with RNase 20 minutes at 37 C. DNA was stored in ddH2O at -20 C. C. albicans DNA (approximately 2 to 5 g) was digested with RE; the enzymes used depended on the gene analyzed. The restriction fragments were resolved in a 0.7% agarose gel by gel electrophoresis in 1X Tris/Borate/EDTA (TBE) buffer. Contour-clamped homogeneous electric field (CHEF) C. albicans chromosome isolation and plug preparation. C. albicans strains grown in YPD at 30 C, were pelleted and washed with 2 mL of 50 mM EDTA. Each pellet was resuspended in 200 L of Solution 1 containing 5 mL of SCE (0.98 M Sorbitol, 1.47 g Sodium Citrate, and 10 mM EDTA), 0.85 M DTT, and treated with

PAGE 43

25 lyticase for 15 minutes at 37 C. Next, to make the plugs, 250 L of 1% low melt agarose (55 C) in 125 mM EDTA were added to each treated pellet. The resuspension was mixed gently and added to the wells in a plug mold using a Pasteur pipette. Then, plugs were allowed to solidify, after which they were removed from the mold and placed in a 6-well tissue culture plate. Plugs were incubated with 4 mL of Solution 2 containing 0.48 M EDTA, 10 mM Tris pH 8.0, and 52 mM DTT, for four hours at 37 C with gentle shaking. Afterwards, plugs were rinsed with Solution 3 (0.42 M EDTA, 0.94% N-lauroyl-sarcosine, 9.8 mM Tris pH 8.0, and 12 mg Proteinase K) and incubated with 4 mL of this solution at 50 C overnight. The next day, plugs were washed twice with SE buffer (0.98 M Sorbitol and 10 mM EDTA) and once with TE buffer. Each wash was carried out for 30 min with gentle rocking. Finally, plugs were stored in TE buffer at 4 C until they were used. Separation of C. albicans chromosomes. Plugs were placed inside the wells of a 1% agarose gel and resolved with 0.5X TBE buffer by pulse filed gel electrophoresis at 14 C. The gel was run with 120-480 seconds (linear ramp) at 3 V/cm with an angle of 106, for 50 hours. The gel was finally stained with ethidium bromide for visualization of chromosomes. Southern blot DNA in agarose gels was denatured with a solution containing 1.5 M NaCl and 0.5 M NaOH, and then neutralized in a solution containing 3 M NaCl, 0.5M Tris-HCl pH7.5. Each step, was carried out for 20 minutes with gentle shaking. DNA was transferred to N+ Hybond nylon membranes (Amersham Bioscience) using a vacuum blotter. Positions

PAGE 44

26 of the molecular weight standards were marked onto the membrane for reference. Membranes were crosslinked at 0.12 J with a UV crosslinker (wavelength 254 nm). Membranes were prehybridized in Hybridization solution (0.5 M pH 7.2 NaHPO4, 1 mM EDTA, 7% SDS, and 1% BSA) and with 5 L of denatured (10 mg/mL) salmon sperm (ss) DNA for 45 minutes to 1 hour at 65 C in a hybridization oven (Robbins Scientific). Membranes were hybridized with 0.1 to 0.2 million cpm/cm2 (in 5 mL used about 1-3 m cpm/mL) of denatured probe in Hybridization solution overnight at 65 C. The probe was prepared using a gel purified PCR fragment as template and alpha 32P-dATP by random priming using the NEBlot Kit (NEB), followed by probe purification using sephadex G-50 columns. After hybridization the membranes were washed with Washing solution (13.34 mM pH 7.2 NaHPO4, 1 mM EDTA, 5% SDS, and 0.5% BSA) until background and excess radioactivity were removed. Finally, the membranes were exposed to a phosphoimager screen and the screen was developed using Molecular Dynamics Storm 860 Phosphoimager. Quantitation of bands with the phosphoimager were done using the Molecular Dynamics Image Quant 5.0 program, which read the intensity of each band. RNA Techniques Yeast RNA extraction and quantitation Procedures were carried out with RNase free solutions. RNA was stored in ddH2O or 70% ethanol at C. RNA extraction from C. albicans cultures. C. albicans was grown in YPD at the desired temperature and to the OD600 required for maximal gene expression. RNA

PAGE 45

27 extraction was carried either using the Hot Phenol method or a RNA Extraction Kit (QIAGEN). The hot phenol method consisted of resuspending the cell pellet in AE buffer (50mM ammonium acetate, 20 mM EDTA, pH 5.3) with 1/10 volume of 10% SDS and vortexing for 30 seconds. RNA was extracted by adding to the resuspended cells a 1.2 volume of 65 C PC (phenol equilibrated in AE buffer:chlorofom in a 1:1 ratio), vortexing for 30 seconds, incubating for 5 minutes at 65 C (vortexing occasionally), and cooling down in a dry-ice ethanol bath for 10 seconds, followed by centrifugation at 2500 x g for 5 minutes. The organic phase was then discarded. The extraction process was repeated until very little material was found between the phases. A final extraction with PCI (phenol:chloroform:isoamyl alcohol in a 24:23:1 ration) was done and the RNA in the aqueous phase was precipitated (1/10 volume 5 M ammonium acetate (pH 5.3) and 2 volumes ethanol) and washed with 70% ethanol. RNA extraction from C. albicans in thrush samples. RNA was extracted from thawed thrush samples by Dr. Cheng using standard methods. (27) Quantitation of RNA. RNA was quantitated by determining the OD260. The quality of RNA was assessed by the OD260/OD280 ratio and gel electrophoresis. A good C. albicans RNA extraction was confirmed by visualization of both rRNA bands with a ratio of 1:2 and a tRNA band. Northern blot analysis RNA samples were prepared for electrophoresis by incubating 10 to 20 g RNA with sample loading buffer at 65 C for 10 minutes. Samples were briefly centrifuged and loaded onto 1.2% formaldehyde agarose. Following electrophoresis, the gel was

PAGE 46

28 transferred to N+ Hybond membrane (Amersham) using a vacuum blotter. After transfer, the membrane was cross-linked and prehybridized with ssDNA. Prehybridization and hybridization was carried out as described for Southern blots. Reverse transcriptase (RT)-PCR RT was carried out in a volume of 20 L using the Reverse Transcription System (Promega). This procedure required 1 g RNA. RT was done at 42 C using AMV RT enzyme and oligo(dT)15 primer provided with the kit or a gene specific primer. PCR was done with 1-4 L of cDNA (1/20 dilution of cDNA obtained from 1 g RNA) and Taq DNA polymerase as mentioned above. Table 2-7. Primers used for IPF1532 mRNA amplifications by RT-PCR Gene Forward and Reverse Primers Use IPF15632 For: TCATATAAGGACACCAGCTGCT Rev: GGAGATGGAGAAGTAGCCAAAT Null mutant and revertant in vitro mRNA expression Protein Techniques Protein expression in E. coli and protein extraction BL21 (DE3) clones containing gene fragments in pET30 vectors were grown to log phase (optical density at 600 nm [OD600] ~0.5) at 37 C on LB/Kan media. Each culture was divided into two halves, and protein expression was induced for one half of each culture with 1 mM IPTG for 1-3 hours at 37 C. Aliquots from both induced and non-induced cultures were taken every hour and pelleted. Cellular proteins were extracted by resuspending the pellet in 1/10 volume 1X phosphate buffered saline (PBS) and an equal volume of 2X SDS buffer (100 mM Tris pH 6.8, 4% SDS, 0.2% bromophenol blue, 20% glycerol, and 200 mM DTT), placing the samples in a boiling water bath for 5-8 min,

PAGE 47

29 followed by cooling on ice and a brief centrifugation. Denatured protein extracts were stored in the SDS buffer at C. Western blot Denatured proteins were separated according to molecular weight by SDS-polyacrylamide gel electrophoresis (PAGE) in a Tris/Glycine system. Proteins were resolved in SDS-polyacrylamide gels with a 4% stacking gel run at 25 mAmps and a 15% separating gel run at 50 mAmps. Appropriate negative controls, indicated in each experiment, were also run on the gels. To visualize proteins, gels were stained with Commassie blue. To detect specific proteins with antibodies, proteins on gels were transferred to nitrocellulose membranes by submerged electrophoretic transfer (48 mM Tris base, 39 mM glycine, 20% methanol, 0.037% SDS, pH 8.3) at 80 mAmps for 3 hours. The membranes were blocked with 5% non-fat dry skim milk in 1X PBS containing 0.5% Tween-20 (PBS-T) for 1 hour. All the antibodies (Ab) used were diluted in 1X PBS and the membranes were washed 4 times with excess 1X PBS-T between each incubation step. Membranes were incubated with primary Ab (1:5000 dilution adsorbed human sera from HIV-infected patients with OPC or 1:5000 dilution mouse anti-His tag Ab) for 1 hour. Protein expression was detected with a secondary Ab, (1:5000 dilution horseradish peroxidase [HRP] conjugated mouse anti-human immunoglobulins [Ig] G, A, and M [Cappel, ICN] or 1:5000 dilution HRP-conjugated rabbit anti-mouse IgG [ICN]), for 1 hour. Enhanced chemiluminescence (ECL) from Amersham Biosciences was used to increase sensitivity. All the incubations and washes were done at room temperature. Membranes were exposed with X-ray HyperFilm (Amersham Bioscience) and developed with an X-ray film processor (Kodak) in order to visualize the labeled proteins.

PAGE 48

30 His-tag protein purification Pellets of PTH subclone in pET30EK/BL21 (500 mL culture induced with IPTG for protein expression) were washed with 50 mL of 50 mM NaCl, 10 mM Tris (pH 7.4) and then pellets were stored at C. Pellets were resuspended in 10 mL of ice-cold 1X binding buffer (1 mM imidazole, 0.7 M NaCl, 50 mM HEPES (pH 7.6)). Cells were then French pressed twice at 18,000 psi. Lysates were spun at 35,000 rpm for 30 minutes at 4 C. PTH subclone protein fragment was purified by nickel (50 mM NiSO4) column chromatography and eluted with 40 mM to 400 mM imidazole, 0.7 M NaCl, 50 mM HEPES (pH 7.6). Fractions (6 mL) were resolved in 15% SDS-PAGE and analyzed by Commassie staining and Western blot using mouse anti-His tag antibody. Generation of polyclonal antibodies to a fragment of IPF11959 protein Polyclonal antibodies were raised in Balb/c mice against the PTH His-tag purified protein fragment. The protein was separated by gel electrophoresis and cut out of the Commassie blue stained gel. These gel fragments were used to inject mice intraperitoneally to raise the antibodies. Mice injections and antibody production was carried out at the University of Florida Hybridoma Core by their established protocol (94). Bleedings were performed after a week from each boost injection (3 boosts were given to each mouse) to test for antibody titers by Western blot and ELISA. IgG was then purified by column chromatography using protein A beads (BioRad). Yeast protein extraction C. albicans was grown at the desired temperature to either log phase or stationary phase. Cell pellets were resuspended in 1X PBS and 1X Laemmli buffer (100 mM Tris-HCl pH 6.8, 4% SDS, 20% glycerol, 0.2% bromophenol blue, and 1 mM DTT). Protein was extracted using different methods: vigorous shaking using a vortex for 10 minutes

PAGE 49

31 with glass beads in presence of 2X Laemmli buffer or French press (18,000 psi). Proteins were stored in 1X Laemmli buffer at C. Enzyme linked immunosorbent assay (ELISA) ELISA with proteins expressed in bacteria. His-tag purified protein (0.5 ug per well) was immobilized on 96-well plates in a sodium bicarbonate/sodium carbonate buffer pH 9.6 (45.3 mM NaHCO3, 18.2 mM Na2HCO3, and 0.02% NaN3). After incubation for 1 hr at 37 C, plates were washed with PBS containing 0.1% Tween-20. Wells were then blocked with 0.25% gelatin in PBS-T for 1 hr at 37 C and afterwards washed with PBS-T. Individual sera from patients with OPC or DC, or from healthy individuals (1:50 dilution in 1X PBS-T) were added to the wells and incubated for 1 hour at 37 C. Plates were then washed with PBS-T. To detect primary antibody binding, horseradish peroxidase (HRP)-conjugated goat anti-human (affinity purified) IgM, IgG, or IgA (1/500 dilution in PBS-T; TAGO) were added individually to each well and incubated for 1 hr at 37 C. After washing with PBS-T, the plates were developed at 37 C with 5 mg o-phenylenediamine (OPD) in 24 mM citric acid, 52 mM Na2HPO4, and 0.012% H2O2. The absorbance was read at 450 nm in a microtiter plate reader. The developing reaction was stopped with 1M phosphoric acid. Negative controls included no primary antibody and no primary and secondary antibody. ELISA with proteins extracted from C. albicans. Proteins extracted from C. albicans by French press were immobilized on a 96-well plate. (27) Identification of C. albicans Proteins Expressed during OPC In Vivo Induced Antigen Technology (IVIAT) For the purpose of this study the IVIAT protocol was adapted for C. albicans. IVIAT follows several steps (see Fig 3-1).

PAGE 50

32 Creation of a C. albicans genomic expression library in Escherichia coli C. albicans clinical isolates were collected by Dr. Nguyen and Dr. Clancy from thrush samples of 25 HIV-infected patients with OPC. The isolates were verified to be C. albicans using standard biochemical testing (API20C system; BioMerieux, France). The genomic expression library was constructed by Dr. Cheng. Genomic DNA from 24 C. albicans clinical isolates was isolated, pooled, and partially digested with Sau3A. RE digest fragments were separated by gel electrophoresis on a 0.7% agarose gel. Fragments from 0.5 to 1.5 kb were gel purified (QIAEX gel purification kit; QIAGEN) and ligated into pET30abc inducible expression vector, which was previously linearized with BamHI and dephosphorylated. Recombinant plasmids were used to transform E. coli DH5 to optimize plasmid production. Finally, plasmids were used to transform E. coli BL21 (DE3) to obtain maximal protein expression. Sera collection and adsorption Sera from 24 HIV-infected patients with active OPC (oral thrush) were collected by Dr. Nguyen and Dr. Clancy. These patients were not the same ones from which the C. albicans clinical isolates were collected from to construct the C. albicans genomic expression library. Serum from each individual patient was confirmed to have reactivity to C. albicans by using a modified ELISA with whole cells and French press cell lysates of an antigenically representative C. albicans clinical isolate (Ca172). Pooled sera for screening the library was prepared by members of Dr. Nguyen’s lab. Antibodies against C. albicans proteins expressed under standard laboratory conditions (YPD at 30 C) in vitro, were eliminated by repeatedly adsorbing the pooled sera against whole cells and French press lysates of Ca172 (denaturing and non-denaturing conditions). The efficiency of the adsorption was verified using ELISA with

PAGE 51

33 cell extracts of Ca172 immobilized to microtiter wells in a 96-well plate. Antibodies reactive with proteins expressed in vitro were considered eliminated when there was no reactivity detected with the ELISA at the lowest dilution tested (1:2). Library screening The C. albicans genomic expression library was diluted and spread on LB/Kan agar plates (250 to 500 CFU per plate) and incubated at 37 C for 12 hours. Using sterile velvet, plates were replicated onto LB/Kan plates containing 1 mM IPTG to induce protein expression of cloned fragments, for 4 hours at 37 C. Cytoplasmic proteins were exposed by partially lysing bacteria with chloroform vapors for 20 minutes. Proteins were then immobilized on nitrocellulose membranes (Gibco BRL) for 15 minutes at room-temperature. Non-specific sites on the membranes were blocked with 5% non-fat skim milk solution in phosphate buffered saline, pH 7.2, containing 0.5% Tween-20 (PBS-T) for 1 hour at room temperature or overnight. The membranes were probed with the adsorbed-pooled sera at a 1:5000 dilution in PBS-T for 1 hour at room temperature. Reactive clones were detected using peroxidase-conjugated goat anti-human IgG, M, and A (Cappel, ICN) in a 1:5000 dilution in PBS-T (1 hour at room temperature). Enhanced chemiluminescence kit (ECL, Amersham Biosciences) was used to increase the sensitivity of detection. Membranes were exposed to Hyperfilm (Amersham Biosciences) and developed. Colony immunoblot Reactive clones were isolated from the initial LB/Kan plates and streaked for single colony isolation. From each streaked clone, 10 single colonies were picked and resuspended in 20 L of LB/Kan broth. From each culture, 5 L were spotted onto 3

PAGE 52

34 LB/Kan agar plates (1 plate with IPTG and 2 without). A negative control (pET30b/BL21 [DE3] with no insert) and a positive control (pET30/BL21 [DE3] with HWP gene, a known C. albicans virulence factor that has been shown to be expressed in vivo (27, 166)) were also spotted on the plates. Protein expression on IPTG containing plates was induced for 4 hours at 37 C. Cytoplasmic proteins were exposed and immobilized on nitrocellulose membranes, which were then screened as previously described. This procedure was repeated until all single colonies isolated from the primary clone were reactive. Identification of potential open reading frames (ORFs) Sequencing and analysis of pET30 inserts. DNA inserts from plasmids purified from reactive clones were sequenced using pET30 primers (T7 promoter primer TAATACGACTCACTATAGGG, T7 terminator primer GCTAGTTATTGCTCAGCGG, 3.2 molar for each one; Novagen). Nucleotide and protein sequences for each clone were analyzed using the Expert Protein Analysis System (ExPASy) Proteomics Server tools ( http://au.expasy.org ), Swiss Institute of Bioinformatics. The sequences were surveyed for homology with other genes in the NCBI GenBank using Basic Local Alignment Search tools (BLAST) ( http://www.ncbi.nlm.nih.gov/BLAST/ ), Saccharomyces Genome Database (SGD) (Stanford University, CA: http://www.yeastgenome.org/ ), and the C. albicans genome databases (CandidaDB, Pasteur Institute, Fr: http://genolist.pasteur.fr/CandidaDB/ and Stanford Genome Technology Center (SGTC): http://sequence-www.stanford.edu/group/candida/ ). For the CandidaDB, “Sequence data for Candida albicans was obtained from the Stanford Genome Technology Center website at http://www

PAGE 53

35 sequence.stanford.edu/group/candida . Sequencing of Candida albicans was accomplished with the support of the NIDR and the Burroughs Wellcome Fund.” For the SGTC, “Nucleotide sequence data for Candida albicans were obtained from the Stanford Genome Technology Center website at http://www-sequence.stanford.edu/group/candida . Sequencing of C. albicans was accomplished with the support of the NIDR and the Burroughs Wellcome Fund. Information about coding sequences and proteins were obtained from CandidaDB available at http://www.pasteur.fr/Galar_Fungail/CandidaDB/ which has been developed by the Galar Fungail European Consortium (QLK2-2000-00795).” Protein sequences were analyzed with PSORTII (prediction of protein sorting signals and localization sites in eukaryotic amino acid sequences, http://www.psort.org/ (129) and domains were identified using SMART (Simple Modular Architecture Research Tools), http://smart.embl-heidelberg.de/smart/show_motifs.pl . Confirmation of the identified protein by the adsorbed sera Subcloning of ORF. Since more than one ORF could be in the same frame as the pET30 promoter, the ORF chosen had to be confirmed to encode the antigen reactive with the adsorbed sera. Primers were designed according to the sequence of the C. albicans gene or hypothetical ORF that matched with that of the insert’s ORF (See Table 2-8). Primer design took into account the codon usage in C. albicans and E. coli. Individual ORFs were amplified by PCR. PCR fragments were cloned in-frame into the expression vector pET30EK/LIC. The ligated plasmids were used to transform DH5. Plasmid from clones obtained were

PAGE 54

36 verified to contain the ORF by RE analysis and sequencing. Once the sequence was confirmed to be correct, the plasmids were used to transform strain BL21 (DE3). Table 2-8. List of genes and their primers for subcloning into pET30Ek/LIC Gene Name of subclone Forward and Reverse Primers Ma2a Sub (396 bp) For: GACGACGACAAGATG ATTACAAAATGTTATG Rev: GAGGAGAAGCCCGGTTA TCTTCTTCTAG PTH1 IFPTH (233 bp) For: GACGACGACAAGATG AAGACATCACAG Rev: GAGGAGAAGCCCGGT ATTATCAGCATGGAAC Mid-RBF1 (233 bp) For: GACGACGACAAGATG ATTGAGAATGAGATT Rev: GAGGAGAAGCCCGGTTA GGTGTAAGCAATTCT RBF1 5’-RBF1 (299 bp) For: GACGACGACAAGATG GCAGAATTACAAAGA Rev: GAGGAGAAGCCCGGTTA TTAGTATCAAATTCA Underlined bases represent the sequence of the pET30EK/LIC vector for independent ligation cloning. Colony immunoblot. The subclones in BL21 (DE3) were streaked for single colony isolation on LB/Kan agar plates. From each streaked clone, 10 single colonies were picked for colony immunoblot analysis. Western blot. Protein expressed by the subcloned ORF was verified to be reactive with the adsorbed sera by Western blot. Protein was extracted from the original clones and their sucblones. Protein fragments were resolved by PAGE in SDS-polyacrylamide gels (15% separating gel). Negative control of a vector without insert was included. Proteins bound on membranes were screened with the adsorbed sera at a 1:5000 dilution in 1X PBS and detected using goat anti-human IgG, M, and A conjugated with HRP and ECL. Determination of In Vivo Expression of Genes Identified RT-PCR with in vivo and in vitro grown C. albicans C. albicans RNA was extracted from thrush samples (in vivo) and from log phase cultures in YPD at 30 C (in vitro). RNA extraction was carried out as previously

PAGE 55

37 described and PCR used Taq DNA polymerase. Primers and conditions used for each gene are shown in the table below. All the PCR reactions were done using DMSO, BSA, and 2.5 mM MgCl2 and the conditions were the following: 94 C 10 min, 35 cycles (94 C 1 min, 50 C 1 min, and 72 C 1 min), and a final extension of 72 C for 10 min. Table 2-9. Primers used for analyzing gene expression in vivo and in vitro by RT-PCR Gene Forward and Reverse Primers Product length EF-1 For: ATTGAACGAATTCTTGGCTGAC Rev: CATCTTCTTCAACAGCAGCTTG 526 bp (cDNA) 891 bp (DNA) RBF1 For: CATGTCAGAGATGCGCTCACT Rev: ATGACCAGGTTCATCGGGTT 357 bp IPF11959 For: cggggtaccTCAAGGTACTAATGGGATGT Rev: gccgagctcTTCTTCAAGTGATATTGGTT 225 bp Non-capitalized bases represent the RE sequence and flanking bases for cloning into the pMB7 vector. PCR fragments were resolved by agarose gel electrophoresis. Quantitation of bands was done with ABI PRISM 7700 SDS instrument (Applied Bioscience) based on ethidium bromide staining. Concentrations for each band were estimated using a standard curve from serial dilutions of 0.5 g of 1 kb DNA molecular weight marker (NEB). Immunostaining of thrush sample with anti-IPF11959 polyclonal Ab Protein and polyclonal Ab production. IPF11959 protein fragment was purified from the Ma2aSub subclone by nickel column chromatography. Polyclonal Ab against His-tag purified IPF11959 protein fragment was raised and purified as described above. Polyclonal Ab reactivity against the IPF11959 protein fragment was verified by Western blot. Immuno-peroxidase staining of thrush samples. Thrush samples were collected from HIV-infected patients with OPC. Thrush pseudomembranes were smeared onto gelatin coated glass slides and ethanol heat fixed. Antigens were exposed by heating slides with 1.8 mM citric acid, 8.2 mM sodium citrate, pH 6 with sodium hydroxide,

PAGE 56

38 followed by cooling at room-temperature. Immunolocalization of antigen was accomplished using immunoperoxidase procedures. Slides were treated with 3% hydrogen peroxide for 30 minutes and then blocked with horse serum (1/5000 dilution in PBS). Slides were incubated with mouse polyclonal antibody raised against IPF11959 (1/250 dilution in PBS) at 4 C overnight. After washing with PBS, slides were incubated with biotinylated horse anti-mouse IgG (1/250 dilution in PBS, Vector Laboratories) for 30 minutes at room-temperature and washed again with PBS. Antigen was detected with avidin-biotin complex reagent (Vector ABC Kit, Vector Laboratories). After another wash with PBS, slides were exposed to diaminobenzidine tetrahydrochloride in an imidazole buffer containing 0.3% H2O2. Slides were then washed with dH20 and counterstained with hematoxylin. C. albicans cells (yeast and filamentous forms) and protein localization were visualized by light microscopy. Negative controls slides were treated in parallel and included thrush pseudomembranes with no primary antibody. Silver staining of thrush samples. Swab samples of thrush were stained with Gomori methamine silver stain by standard method in the clinical pathology laboratory at the VA Hospital in Gainesville, Florida. Determination of Identified Gene as a Virulence Factor To determine if a gene is a virulence factor, we followed three steps. First, we created isogenic mutants; second, we analyzed the phenotype of the mutant; and third, we tested the mutants’ virulence using murine models of candidiasis. Those genes whose disruption impaired important traits for virulence were given priority.

PAGE 57

39 Creation of Isogenic Mutants Gene disruption Construction of disruption vector. Fragments (approximately 200 bp) overlapping the 5’ and 3’ ends of the gene’s ORF to be disrupted were amplified by PCR. PCR primer pairs were designed using the C. albicans genome sequence available at the time and Vector NTI program. RE sites were added to each primer for directional cloning into pMB-7 vector. After each PCR fragment was purified, these were cut with the appropriate RE, according to the RE site added to each primer, and purified again using molecular weight cutoff spin filters. Therefore, two flanking fragments (F1 and F2) for each gene were created for ligation into pMB7. The list of primers for creating F1 and F2 fragments for ALG5 and IPF15632 are shown in Table 2-10. Table 2-10. List of genes and their primers for cloning into pMB-7 Gene fragment Forward and Reverse Primers Product length ALG5 F1 For: gcgaaactgcag CAGTCATAAACCTCGAAAAC Rev: acgcgtcgac CAAGATATTCAATGGCTTCA 225 bp ALG5 F2 For: cggggtacc GCATTAGTGTTTGTGTTTGG Rev: gccgagctc TGATCCATCAATTTCTTGCC 222 bp IPF15632 F1 For: acatgcatgc TGATCATTCACCAACAAACA Rev: acgcgtcgac ATTTTACGAGGTGAAGTTTT 230 bp IPF15632 UpF1 For: acatgcatgcTTCATCTTCCCACTCTTTTC Rev: acgcgtcgacATTAGGGAGTGAAAATTATC 232 bp IPF15632 F2 For: cggggtacc TGCTGCTAATGCTTCTTCCT Rev: gccgagctc TATTAAATGCACCACCTGGT 260 bp Non-capitalized bases represent the RE sequence and flanking bases for cloning into the pMB7 vector. The vector pMB7 was first linearized with SacI and Kpn I and then dephosphorylated with SAP. This allowed for F2 to be directionally ligated to the 3’ end of the cassette. After, the ligation was used to transform DH5 cells and transformants were analyzed by RE digest. Recombinant vectors that contained the F2 fragment were

PAGE 58

40 linearized with SphI and SalI and dephosphorylated with SAP. This time F1 was ligated directionally to the 5’ end of the cassette and used to transform DH5 cells. Transformants were analyzed by RE digest and sequencing. Prior to transformation, the disruption vector was linearized with SphI and SacI RE, precipitated, and resuspended in a small volume of ddH20 until ready to be transformed into C. albicans cells. URA3 blaster technique (49). Cells were grown in 100 mL YPD containing 0.4 mM uridine at 30 C until they reached an OD600 of 1.2. Cells were pelleted and resuspended in 25 mL of 5 mM lithium acetate, 10 mM DTT, and 1X TE (10 mM Tris-HCl pH7.5 and 1 mM EDTA). After 1 hour incubation at room temperature, cells were washed three times with ice-cold ddH20. A final wash was done in 1 M ice-cold Sorbitol. The pellet was resuspended in 0.5 mL of 1 M Sorbitol. To disrupt the fist allele in CAI-4 strain, about 2 to 10 g of the linearized fragment containing the disruption cassette were mixed with 50 L of cells (approximately 4 to 8 x 108 cells). To disrupt the second allele in the heterozygous mutant, a smaller concentration (approximately 1 g) of the linearized fragment was required. However, the amount of DNA used for transformation depended in the gene that was being disrupted. Cells were incubated with DNA for 5 minutes on ice. Transformation was done by electroporation (1 mm cuvette, 1.5 V). All of the transformation reactions, 200 L/plate, was spread immediately on Saboraud dextrose plates lacking uracil (SD-ura). Plates were incubated at 30 C for a few days until colonies appeared. Integration of the cassette at

PAGE 59

41 the correct locus was verified by Southern blot using an -32P-labeled probe specific for the ORF. URA+ recombinants were grown in YPD at 30 C overnight, after which 105 to 107 CFU were spread on 5-FOA plates (1.4% yeast nitrogen base (YNB), 0.2% 5-FOA, 0.005% uridine, and 4% dextrose). Urarecombinants were grown in YPD containing 0.1 mM uridine at 30 C overnight and their genomic DNA was analyzed by Southern blot. Recombinants that had one copy of the gene disrupted with one direct repeat of hisG were used to disrupt the next copy of the gene by transformation with the same disruption cassette. Screening for recombinant mutants by Southern blot. All the recombinants obtained from each selection were analyzed by Southern blot. The list of RE and probes used for analyzing the band pattern of IPF11959, ALG5, IPF15632 and CDC24, along with the expected size of the fragments on the blots are shown in Table 2-11. Table 2-11. List of genes analyzed by Southern blot and conditions Gene analyzed RE used Probe Band size expected or observed in wt HpaI/AvaI KO-F1 KOF1 hybridizes to a 1.079 kb fragment IPF11959 HpaI/AatII KO-F2 KOF2 hybridizes to a 3.5 kb fragment ALG5 HpaI/AatII ALG5-F2 ALG5-F2 hybridizes to a 1.5 kb fragment IPF15632 PvuII/SphI Mai-F2 MaiF2 hybridizes to the observed ~2.3 kb fragment and to a 3.5 kb. One band for each allele CDC24 HpaI CDC24-F1 CDC24-F1 hybridizes to a 2.7 kb fragment Approximately 470 bp within IPF15632 ORF was replaced by hisG. While for ALG5, approximately 375 bp were replaced by hisG.

PAGE 60

42 Re-insertion of URA3 gene at its own locus The vector pURA3 was cut with PstI/BglII RE and precipitated. The RE digest vector fragments were used to transform ura3mutants. Genomic DNA from recombinants was digested with EcoRI and analyzed by Southern blot. A 4.2 kb fragment from pURA cut PstI/BglII with was gel purified and labeled with -32P-dATP by random priming to be used as a probe. RE-insertion of One Copy of the Gene Back into its disrupted Locus The vector pMB7-1 was derived from pMB7 by removing an XbaI fragment, which contained the 3’ hisG of the cassette. Therefore, pMB7-1 contained the URA3 gene and once copy of HisG. The sequence to be replaced back into the genome was amplified by PCR (see table 2-12). Table 2-12. List of genes and primers for cloning into pMB7-1 Gene analyzed Forward and Reverse Primers Product length ALG5 For: gcgaaactgcag ATTTTTCAGTCATAAACCTC Rev: acgcgtcgac TTGTTTGATTGTATTGTGTG 1037 bp IPF15632 For: acatgcatgcTGATCATTCACCAACAAACA Rev: acgcgtcgac TTA GTTCAAATCTTCTTCAGAAATCAATTTTTGTTC CTTTTTCAGTAATGAATTAT Downstream sequence added: Down For: acgcgtcgac TGATGACCGTATATATATAT Down Rev: acgcgtcgac ACCCGTAACGAGCAAACTGG 1007 bp 203bp Non-capitalized bases represent the RE sequence (underlined) and flanking bases for cloning into the pMB7-1 vector. Bases in italics are the Myc tag sequence added to the 3’ end of IPF15632. Primers used for adding 203 bp of downstream sequence from IPF15632 ORF were amplified with Down forward and Down reverse primers Primers were designed using Vector NTI and the C. albicans genome database. RE sites were added to both primers for directional cloning of the PCR fragment into pMB7-1 linearized with SphI and SalI and dephosphorylated. Once the re-insertion vectors were

PAGE 61

43 created, these were linearized with SphI and SacI and precipitated. The linearized vector was used to transform the mutant strains with the according gene disrupted. Transformation and screening of recombinants were carried out as previously described. Phenotypic Analysis of Mutants Mutants were analyzed under a variety of conditions described below. The phenotypes observed for the mutants were compared with the wild type strain CAI-12 (see table 2-1). Growth curve Subcultures in YPD of C. albicans strains were started with an OD600 of 0.1. Cultures were grown at both 30 and 37 C. The OD600 was read every hour for a period of 12 hours or until cells had reached stationary phase using a spectrophotometer. Hyphae formation In solid media. For each strain, 105 CFU in 10 mL of PBS were spotted on different hyphae inducing media (YPD containing 5% fetal bovine serum [FBS]; RPMI; M199; Lee; Spider). Plates were incubated at 30 C or 37 C for 7 days. In liquid media. Cells from overnight cultures were resuspended in hyphae inducing liquid media (YPD containing 5% FBS or RPMI) and grown at 37 C. Cells were observed under a microscope at different time intervals (10 minutes to 2 hours). Colony morphology and auxotrophies For each strain, 105 CFU in 10 mL of PBS were spotted on YPD, YPGly containing 2% glycerol and YPGal containing 2% galactose agar plates. Plates were incubated at 30 C or 37 C for 7 days.

PAGE 62

44 Embedded agar phenotype Overnight grown strains were diluted in saline solution to give 50 to 100 CFU in 10 L, which were then added to warm YPD or YPGly agar medium and poured on plates to solidify. After the agar had solidified, the plates were incubated at 37 C overnight. Colony’s hyphae formation and growth under embedded conditions were recorded by photography (RT KE diagnostic camera) using an inverted microscope (Leica) with a 2.5X magnification lens. Screening for cell wall defects Strains were grown in YPD at 30 C overnight and washed with 1X phosphate-buffered saline (PBS) solution prior to being used in the experiments, unless otherwise noted. For each strain, CFUs were counted using a haemocytometer and diluted to the concentration indicated in each experiment. Calcofluor white and SDS. Different concentrations of each strain (105, 104, 103, and 102 CFUs, each in 5 mL) were spotted on YPD plates containing 50 mg/mL of Calcofluor white or 0.06% SDS. Plates were incubated at 30 C or 37 C for 2 days. Zymolyase. Strains grown to log phase in YPD medium at 37 C were washed with 1X PBS. Absorbance at 600 nm was adjusted to 0.5 in 1X PBS for each strain. Zymolyase-20T (0.1 mg/mL) was added to each strain suspension and incubated with shaking at 35 C for 2 hours. Absorbance at 600 nm was read every 10 min. A decrease in absorbance reflects lysis of yeast cells by Zymolyase. Aminoglycosides. Approximately 5000 CFU of each strain were streaked onto YPD plates containing either G418 (1.2mg/mL) or hygromycin B (100 or 200 mg/mL). Plates were incubated at 30 C or 37 C for 2 days.

PAGE 63

45 Cell adherence assays The pharyngeal epithelial cell line FaDu (human pharynx squamous cell carcinoma) and colon epithelial cell line HT-29 (human colon colorectal adenocarcinoma) were obtained from the American Type Culture Collection (ATCC HTB-43 and HTB-38 respectively). FaDu cell lines were routinely maintained in minimum essential medium (MEM) Eagle with 2 mM L-glutamine and Earle's BSS (1.5 g/L sodium bicarbonate, 0.1 mM non-essential amino acids, 1.0 mM sodium pyruvate, 90%; and 10% FBS) at 37 C with 5% CO2. HT-29 cell lines were routinely maintained in McCoy's 5a medium supplemented with 1.5 mM L-glutamine, 90%, and 10% FBS, at 37 C with 5% CO2. Primary human buccal epithelial cells (BECs). Human buccal epithelial cells (BECs) were collected from laboratory members by gently scraping the cheek mucosa with a cotton swab and dispensing the cells into 10 mL of PBS. The pooled BECs were washed four times and then diluted to 105 CFU/mL in PBS. Strains were diluted in PBS to 106 CFU/mL. To perform the adherence assay, 0.5 x 106 CFU of each strain were incubated with 0.5 x 105 BECs for 2 hours at 37 C with shaking. BECs alone were incubated in PBS as a negative control. After incubation, non-adherent yeast were removed by vacuum filtration through a 10 m pore size polycarbonate filter, which will only retain epithelial cells and yeast adhered to them. Each filter was washed 10 times with 1X PBS. The washed filters were then pressed gently onto glass slides. The slides were air-dried, heat fixed, and Gram stained. The number of yeast adhering to 100 epithelial cells was counted by light microscopy with a magnification of 40X. Each assay was carried out in duplicate three times on different days. Statistical analysis was

PAGE 64

46 performed using the PRISM program (GraphPad Software) using the Wilcoxon method. A P-value of <0.05 was considered significant. Each sample was done in triplicates. Pharyngeal epithelial cell line (FaDu). For each C. albicans strain, 104 CFU were incubated with a confluent FaDu pharyngeal epithelial cell line in Eagle’s media supplemented with 10% FBS (12-well plate) for 1.5 hours at 37 C with 5% CO2. Non-adherent yeast cells were removed by washing wells three times with Eagle’s media. FaDu cells were lysed with dH2O and removed by scraping wells. Wells were scraped and washed twice with dH2O; these were combined and the final volume was adjusted to 2 mL/well. Serial dilutions were spread on SD plates, which were incubated overnight at 37 C. Each assay was carried out in triplicate. Number of yeast adhered was calculated as the number of colonies recovered on SD plates. Statistical analysis was performed using the PRISM program (GraphPad Software) using the Mann-Whitney method. A P-value of <0.05 was considered significant. Each sample was done in triplicates. Colonic epithelial cell line (HT-29). For each C. albicans strain, 104 CFU were incubated with confluent HT-29 cell line in McCoy’s 5a media supplemented with 10% FBS (12-well plate) for 1.5 hours at 37 C with 5% CO2. Non-adherent yeast cells were removed by washing wells three times with McCoy’s 5a media. HT-29 cells were lysed with dH2O and removed by scraping wells. Wells were scraped and washed twice with dH2O, these were combined and the final volume was adjusted to 2 mL/well. Serial dilutions were spread on SD plates, which were incubated overnight at 37 C. Each assay was carried out in triplicate. Number of yeast adhered was calculated as the number of colonies recovered on SD plates. Statistical analysis was performed using the PRISM

PAGE 65

47 program (GraphPad Software) using the Mann-Whitney method. A P-value of <0.05 was considered significant. Each sample was done in triplicates. Polymorphonuclear cells (PMNs) phagocytosis and killing assay Yeast preparation. C. albicans mutant and wt (CAI-12) were grown in YPD at 30 C overnight. Cells were washed three times with ddH20 and diluted to a concentration of 105 CFU in MEM media. PMNs isolation. PMNs were isolated from heparinized blood by dextran sedimentation, this was followed by centrifugation at 1200 rpm in Ficoll-Hypaque. Red blood cells were removed by hypotonic lysis. We used a standard protocol for PMNs isolation. Phagocytosis assay. First, C. albicans cells (105 CFU/mL) were opsonized with 2.5% serum for 45 minutes. At the same time, PMNs were diluted to 5 x 106 cells/mL in MEM media, 100 L were placed in wells in a 96-well plate, and incubated at 37 C with 5% CO2 for 30 minutes. Non-adherent PMNs were removed and wells were gently rinsed twice with 200 L of MEM media. The opsonized C. albicans cells (104 CFU) were incubated with PMN adhered to wells at a 1:1 ratio in MEM media for 25 minutes at 37 C with 5% CO2. Non-phagocytosed yeast were removed by rinsing the wells twice with 300 L of MEM media. The washes were combined for each well and the volume was adjusted to 1 mL. Non-phagocytosed yeast were diluted and spread on SD plates. Plates were incubated at 37 C overnight and colonies were counted the next day. The percentage of phagocytosed yeast was calculated with the following equation: (1 – (number of yeast CFU obtained per well/ 104 CFU)) x 100. Each sample was carried out in triplicate in order to calculate standard deviation.

PAGE 66

48 Killing assay. After the wells were washed to remove non-adherent C. albicans cells, 200 L of MEM media and YPD (1:1 ratio) were added to each well and incubated at 37 C with 5% CO2 for 3 hours. Finally, the MEM/YPD solution was removed from the wells and each well was washed three times with 200 L of dH20 and by scraping with a wooden applicator. The washes were combined for each well and the volume was adjusted to 1 mL. Non-killed yeast were diluted and spread on SD plates. Plates were incubated at 37 C overnight and colonies were counted the next day. The percentage of killed yeast was calculated with the following equation: (1 – (number of non killed yeast CFU obtained per well/ number of non-phagocytosed yeast obtained for each corresponding well)) x 100. Each sample was carried out in triplicate in order to calculate standard deviation. Murine Models of Candidiasis Two murine models of candidiasis, DC and OPC, were used. The inocolum for both murine models was prepared in the same manner. Glycerol stocks of mutant strains and wild type CAI-12 were streaked for single colony isolation on YPD plates. One colony from each strain was used to spread an entire plate of YPD and grown at 30 C until the plate was confluent. Cell were scraped out of plates and used to heavily inoculate YPD media. Cultures were allowed to grow for 18 hours at 30 C. Cells were washed three times with sterile saline solution (1000 rpm for 5 min) and diluted in saline solution to the concentration indicated for each experiment. The concentration was verified by counting with a hemacytometer and by counting the number of viable CFU that grew on YPD plates after spreading a dilution of the inocolum. The number of mice infected is indicated for each experiment in the results section.

PAGE 67

49 Disseminated candidiasis murine model Infection. The DC murine model was carried out using the standard method. Seven week-old male ICR mice (Harlan-Sprague) were inoculated by intravenous (i.v.) injection of the lateral tail vein with 106 viable CFU of C. albicans strains in 0.2 mL of sterile saline solution. We used a lethal dose in which the wild type kills the mice in an average of 7 days. Mice were followed until they were moribund or for 30 days after which they were sacrificed. For analysis of the alg5 null mutant we used the non-standard inoculation dose of 105 viable CFU, while for the IPF15632 null mutant we used a dose of 2.5 x 105 viable CFU. Mice were housed under specific pathogen-free conditions. Survival curves were calculated according to the Kaplan-Meier method and statistics were calculated using the Newman Keuls analysis; a p-value of < 0.05 was considered significant. Tissue burden. For tissue burden experiments with the IPF15632 null mutant and CAI-12 strain, 9 mice were infected/group as described above. However, the mice were sacrificed at 22 hours and 4 days after i.v. inoculation. The liver, kidney, and spleen of each mouse were aseptically removed and weighed. Each organ was individually homogenized in saline solution containing Amp and Kan (3 mL, 2 mL, and 1 mL respectively) using glass tissue homogenizers. Homogenized organs were diluted with saline solution containing antibiotics and 100 L of each dilution was spread on SD plates containing Amp and Kan to prevent bacterial growth. Plates were incubated at 35 to 37 C overnight. The number of CFU/g of each organ was calculated from the number CFU present on the plates. Graphs were plotted in Excel program and statistical analysis

PAGE 68

50 were calculated using the Mann-Whitney method. A P-value of 0.05 was considered significant. Histopathology. For histopathology, a small section of each kidney was fixed immediately in formalin and embedded in paraffin. Thin sections were prepared and stained with Gomori methamine silver stain by standard method in the clinical pathology laboratory at the VA Hospital in Gainesville, Florida. Oropharyngeal candidiasis murine model Infection and tissue burden. Eight week-old male ICR mice (Harlan-Sprague) were immunosuppressed with 4 mg of cortisone acetate (sub-cutaneous injection) one day prior to infection. On the day of infection, mice were sedated with 2.4 mg of sodium pentobarbital by intraperitoneal injection. While sedated, mice (12 mice per strain) were inoculated orally with 108 viable CFU of C. albicans strains by means of a saturated cotton swab placed sublingually for 2 hours. Mice were kept immunosuppressed with cortisone on day 1 and 3 after the day of infection. In order to prevent development of bacterial infections, the immunosuppressed mice were given tetracycline hydrochloride (0.5 mg/mL) with their drinking water starting the day before infection. On day 6 post-infection, mice were sacrificed and the esophagus, tongue, and jaw muscles were aseptically removed. For each mouse, tissues were weighed and homogenized in 2 mL of PBS (with 60 ug/mL each of amikacin and pipericillin) using glass tissue homogenizers, diluted, and spread on SD plates containing antibiotics to prevent bacterial growth. Plates were incubated at 35 C. The number of CFU/g of tissues was calculated as the number of CFU recovered on SD plates per gram of tissue. Statistical analysis was performed

PAGE 69

51 using the PRISM program (GraphPad Software) and the Mann-Whitney method. A P-value of 0.05 was considered significant. Histopathology. For histopathology, a small section of each tongue was fixed immediately with formalin and embedded in paraffin. Thin sections were prepared and stained with Gomori methamine silver stain by standard method in the clinical pathology laboratory at the VA Hospital in Gainesville, Florida. Duplication Stability In Vitro Conditions Growth under non-stressful conditions The triploid IPF15632 one copy KO (6F2 + U2, see table 2-4) was streaked from the original glycerol stock onto a YPD plate and grown at 37 C. Cells were streaked from the plate and continuously passed 20 times on YPD plates at 37 C. After the 20th passage, 20 colonies were picked for analysis. Growth under stressful conditions The triploid IPF15632 one copy KO (6F2 + U2, see table 2-4) was streaked from the original glycerol stock onto a YPD plate containing 100 g/mL of Hygromycin B and grown at 37 C. Cells were streaked from the plate and continuously passed 20 times in YPD/Hygromycin B plates at 37 C. After the 20th passage, 20 colonies were picked for analysis. Stability of the Triploid 1 KO Mutant Duplication During Growth in Mice The triploid IPF15632 one copy KO (6F2 + U2, see table 2-4) was streaked from the original glycerol stock onto a YPD plate and grown at 37 C. The inocolum was prepared as mentioned above. Two mice were inoculated i.v. via the lateral tail vein with 105 CFU. Mice were sacrificed after 6 days post-infection. The liver, kidneys, and spleen

PAGE 70

52 were isolated under sterile conditions. Organs were homogenized separately in saline solution (3, 2, and 1 mL respectively) containing antibiotics. Homogenized tissues were spread on SD/Amp/Kan plates and grown at 30 C. From the plates, 50 colonies (25 from each mouse) belonging to kidneys homogenization and 2 colonies to spleen homogenization were picked for analysis. Southern Blot Analysis Colonies were grown in YPD media overnight at 30 C. Genomic DNA was isolated and cut with SphI/PvuII RE and analyzed by Southern blot using -32P labeled MaiF2 probe Intensity of each band was calculated using the Molecular Dynamics Image Quant 5.0 program.

PAGE 71

CHAPTER 3 IDENTIFICATION OF CANDIDA ALBICANS ANTIGENS EXPRESSED DURING OROPHARYNGEAL CANDIDIASIS IN HIV-INFECTED PATIENTS Project Overview One fundamental feature of microbial infections is that they are complex and dynamic processes. The site of infection undergoes constant changes from the moment the pathogen enters the human host and is recognized by it. It is technically challenging to determine and mimic all the complex and changing environmental signals that occur during infection. For these reasons it is unlikely that certain virulence factors could be identified by studying the pathogen under laboratory conditions. Therefore, to fully understand microbial pathogenesis, it is important to characterize microbial genes that are specifically activated or upregulated during host infection (i.e. growth in vivo). Over the past two decades, many methods developed to investigate gene expression and study pathogens during infection have allowed the identification of bacterial genes encoding novel virulence factors important for microbial survival within the host and disease progression. (69, 106, 107, 109) We hypothesized that selected antigens expressed by C. albicans during oropharyngeal candidiasis (OPC) are crucial for its pathogenicity. In order to identify proteins expressed by C. albicans during OPC we applied an antibody-based screening method called In Vivo Induced Antigen Technology (IVIAT). The first goals of this project were to identify five genes encoding antigenic proteins using IVIAT and to confirm that these proteins were expressed by C. albicans during OPC. The last goal, explained in the following chapters, was to investigate if the genes 53

PAGE 72

54 identified are important for virulence. The last goal was realized by disrupting the genes identified to encode the antigenic proteins in order to create isogenic mutants, analyzing the phenotype of the mutants created, and testing the mutants virulence using cell and murine models of candidiasis. Identification of C. albicans Antigenic Proteins Expressed During OPC Using IVIAT IVIAT Steps We adapted IVIAT to identify antigenic proteins expressed by C. albicans during OPC. IVIAT follows several steps shown in Fig. 3-1. First, a C. albicans genomic expression library was created in E. coli. The library consisted of 24 C. albicans clinical strains isolated from thrush samples of patients with OPC. Second, human sera from 24 HIV-infected patients with OPC were collected, pooled, and adsorbed against immobilized C. albicans proteins expressed during routine growth in the laboratory, in YPD medium at 30 C. The purpose of the pre-adsorption step was to enrich the sera for antibodies that will react with C. albicans antigens expressed mainly during OPC. The first two steps were performed by investigators in Dr. Nguyen’s laboratory. My project started at the third step, which was to screen the C. albicans genomic expression library with the pre-adsorbed human sera. Fourth, the ORFs responsible for the reactivity to the human sera were identified by isolating the plasmids from the clones reactive to the human sera, sequencing of the plasmid inserts containing C. albicans genomic DNA, and confirming the reactivity of the ORFs (Open Reading Frames) to the pre-adsorbed human sera by colony immunoblots and Western blots. In this final step, the C. albicans genes encoding the serum-reactive ORFs were identified.

PAGE 73

55 C. albicansGenomic DNApET30 a, b, c Ligate Replicate library on medium with IPTG Expression library in E. coli C. albicansgrown in vitro (laboratory) Probe expression libraryEliminate adsorbed antibodies Pooled serum from patients with HIV & OPC Purify plasmid and sequence cloned insert Identify C. albicansgene encoding the ORF reactive with adsorbed sera 1 3 2 4 4 C. albicansGenomic DNApET30 a, b, c Ligate Replicate library on medium with IPTG Expression library in E. coli C. albicansgrown in vitro (laboratory) Probe expression libraryEliminate adsorbed antibodies Pooled serum from patients with HIV & OPC Purify plasmid and sequence cloned insert Identify C. albicansgene encoding the ORF reactive with adsorbed sera 1 1 3 3 2 2 4 44 4 Fig. 3-1. IVIAT steps. There are four step s: 1) Creation of a C. albicans genomic expression library; 2) Adsorption of pooled sera collected from HIV-infected patients with OPC; 3) Screening the genomic expression library with the pre-adsorbed sera from patients with OPC; 4) Isolation of clones expressing the antigenic proteins and sequencing of inserts to identify which C. albicans gene encodes the reactive ORF. Screening the C. albicans genomic expression library with the pre-adsorbed sera Setting limits for library screening. To assure that each gene in the C. albicans genome included in our library was screened at least once, we have to take into account the 32 Mbp diploid genome size of C. albicans and the genomic inserts sizes of 0.5 to 1.5 kb. Following the equation N= (loge(1-P))/(loge(1-f)), where N represents the number of

PAGE 74

56 recombinants that need to be screened to achieve a 99% probability, P, that a clone will be found in a library, and f is the fraction of the genome in each recombinant, we would have to screen approximately 150,000 clones from the genomic expression library. However, my goal is not to completely screen the library and to identify all the genes expressed by C. albicans during OPC, but to identify a few genes in order to substantiate my hypothesis. Therefore, we determined that identification of five genes was optimal to allow the likelihood that at least some genes are involved in virulence without having to create an excess amount of mutants. Screening the library. From the first 6,000 colonies screened, I initially identified seven colonies reactive with the adsorbed sera. Colony immunoblots were prepared from densely plated 15 mL Petri dishes as described in Material and Methods (Chapter 2). To select the positive colony reactive with the sera, the X-ray film used to detect immuno reaction on the membrane was superimposed on its corresponding plate and colonies surrounding the positive dark spot were picked. False positive colonies were recognized by the presence of very dark spots which did not align with the colonies on the plate. An example of one membrane containing approximately 400 colonies screened with the adsorbed human sera is shown in Fig 3-2A. After each possible positive clone was streaked for single colony isolation, ten single colonies for each clone were picked and verified by colony immunoblot using the adsorbed human sera. Colony immunoblots included a negative control (clone containing pET30b vector alone) and a positive control (clone containing pET30 expressing Hwp1, a previously hyphal cell wall protein demonstrated to be expressed by hyphae during OPC and required for virulence (27, 166, 170, 177)). An example of a secondary membrane screened by colony immunoblot is

PAGE 75

57 shown in Fig 3-2B. Therefore, in order to isolate a single clone all 10 single colonies derived from it had to be positive by colony immunoblot screening. After several rounds of screening, only two of the 7 initial reactive colonies remained positive with the adsorbed sera. To achieve the pre-set goal for the identification of five C. albicans genes expressed in vivo, approximately 15,000 colonies were screened with the adsorbed human sera. A)B) (+) Ctrl (+) Ctrl (-) Ctrl A) A)B) (+) Ctrl (+) Ctrl (-) CtrlB) (+) Ctrl (+) Ctrl (-) Ctrl (+) Ctrl (+) Ctrl (-) Ctrl Fig 3-2. Immunoblots screened with pre-adsorbed sera from patients with OPC. A C. albicans genomic DNA expression library screened with the pre-adsorbed human sera from HIV-infected patients with OPC. A) A membrane representing the library screening. The black arrow points to a positive colony B) A membrane representing a colony immunoblot screening. The red arrows point to the spotted positive control (circled) and the blue arrow points to the spotted negative control (circled). Identification of the ORFs responsible for the reactivity with the pre-adsorbed sera Plasmids from the five positive clones were isolated and the cloned inserts were sequenced with flanking primers (T7 promoter and T7 terminator primers) within the pET30 vector. Potential ORFs from each insert sequence were identified using ExPASy Proteomics Server tools ( http://au.expasy.org ). A single ORF was identified for three out of the five clones (named Khoi, Lan, and Mai). The other two inserts, clones Ma1 and

PAGE 76

58 Ma2, contained two ORFs within them. To determine which of the ORFs for the latter two clones was responsible for the reactivity to the adsorbed human sera, I cloned individual ORFs within each insert in-frame in the expression vector pET30EK/LIC. These subclones were then screened individually by colony immunoblot to determine which encoded the serum-reactive antigen. I found that a single ORF from each insert (263 and 233 bp respectively) was reactive with the adsorbed human sera. Protein fragments expressed by the original clones and subcloned ORFs were also verified to be reactive with the adsorbed sera by Western blot. Western blots and Commassie stained SDS-polyacrylamide gels for the two representative subclones are shown in Fig 3-3. Identification of C. albicans Genes Encoding the Antigenic ORFs The ORF sequences responsible for the reactivity were analyzed by BLAST using the C. albicans genome databases (CandidaDB, Pasteur Institute and Stanford Genome Technology Center (SGTC)) and the Saccharomyces cerevisiae genome database (SGD, Stanford University, CA). Each of the five ORFs aligned within the sequence of a distinct gene in the C. albicans database. One of the genes, RBF1, had previously been studied and shown to be a virulence factor for C. albicans. Three of the genes identified, CDC24, ALG5, and IPF11959, had names and/or functions previously assigned either by homology to a S. cerevisiae gene or by an identified function. The fifth gene identified as IPF15632 by the CandidaDB encodes a putative soluble protein. (Table 3-1) RBF1. C. albicans RBF1 (ribosomal protein genes (RPG)-box binding factor) encodes a transcription factor that can bind RPG box DNA sequences. Disruption of RBF1 gene induces filamentous growth of C. albicans, indicating that RBF1 is involved in one of the signal transduction pathways for the yeast-hyphae transition. Moreover, rbf1 knockout mutants exhibit avirulence in a murine model of DC. (7, 87, 88)

PAGE 77

59 + 45 min+--M 3 hrs6.5 kDa14.4 kDaA) ++--M 45 min3 hrsB)C)6.5 kDa14.4 kDa20.1 kDa+ +--M 1 hr50 min D)50 min+ +-M1 hr + 45 min+--M 3 hrs6.5 kDa14.4 kDaA) + 45 min+--M 3 hrs6.5 kDa14.4 kDaA) ++--M 45 min3 hrs++--M 45 min3 hrsB)C)6.5 kDa14.4 kDa20.1 kDa+ +--M 1 hr50 min D)50 min+ +-M1 hr C)6.5 kDa14.4 kDa20.1 kDa+ +--M 1 hr50 min C)6.5 kDa14.4 kDa20.1 kDa+ +--M 1 hr50 min 6.5 kDa14.4 kDa20.1 kDa+ +--M 1 hr50 min + +--M 1 hr50 min+ +--M 1 hr50 min D)50 min+ +-M1 hr D)50 min+ +-M1 hr 50 min+ +-M1 hr50 min+ +-M1 hr Fig. 3-3. Antigenicity of RBF1 and IPF11959 to sera from patients with OPC. Commassie blue stained SDS-polyacrylamide gels and their corresponding Western blots with E. coli lysates of induced and non-induced ORF subcloned in the pET30EK/LIC vector. Western blots were screened with the pre-adsorbed human sera from HIV-infected patients with OPC. RBF1 protein fragment A) SDS polyacrylamide gel and B) Western blot. IPF11959 protein fragment C) SDS polyacrylamide gel and D) Western blot. M is the molecular weight marker. Western blots included as negative control E. coli lysates from subclones not induced with IPTG.

PAGE 78

60 Table 3-1. Description of C. albicans genes identified by IVIAT screening Clone Name & ORF Size Standard and Systematic Name* and Gene Size Function** Importance in Pathogenesis Ma1 (612 bp) RBF1 (1605 bp) (IPF27978.1; orf19.5558) Transcription factor that inhibits hyphal formation Null mutant has decreased pathogenicity in a murine model of DC. Khoi (858 bp) CDC24 (2535 bp) (IPF19625.1; orf19.3174) GTP/GDP exchange factor for CDC42p. Essential gene Disruption of one allele and expression of the other copy regulated by the MET3 promoter attenuated virulence in murine model of DC. Ma2 (761 bp) IPF11959 (1410 bp) (orf19.4232) Referred as PTH1 in the ATCC1001 strain***. Unknown Unknown Lan (972 bp) ALG5 (972 bp) (IPF9097.2; orf19.2837) Putative dolichyl-phosphate betaglucosyltransferase Unknown Mai (690 bp) IPF15632 (1188 bp) (orf19.3469) Unknown Unknown *The C. albicans genome annotation is based on Assemblies 6 and 19 of the C. albicans SC5314 genome sequences. The databases provide the genes standard names based on previous gene characterization or on a postulated functional homologue from S. cerevisiae. Genes that did not follow the mentioned criteria were designated systematic names either as Individual Protein File IPF in the CandidaDB or orf19 in the SGTC. **The functions of the genes were assigned on the basis of published data or by homology to proteins of known function. ***ATCC1001 is another C. albicans strain. CDC24. C. albicans CDC24 (cell division control 24) is a GTP/GDP exchange factor for CDC42p. In contrast to RBF1, disruption of one copy of CDC24 and expression of the other copy under the MET3 promoter blocks hyphal formation of C. albicans, pointing to a role in the formation of hyphae. CDC24 null mutants are not viable, but heterozygous mutants with ectopic expression of CDC24 display avirulence in a murine model of disseminated candidiasis. These results were determined by another group during the course of this project. (9, 10)

PAGE 79

61 IPF11959. When the clone Ma2 sequence was analyzed in 2000 it matched the PTH1 gene from the C. albicans strain ATCC1001, with the exception of the first 45 amino acids. The PTH1 name was described as proline transporter helper 1 protein by Prasad R. Y18210, CAB57448.1 EMBL/GenBank/DDBJ databases (submitted OCT-1999). Now with the finished C. albicans genome sequence, we found that PTH1 is the homolog of the predicted ORF IPF11959 in the C. albicans SC5314 strain. The genes differ only in their first 45 codons at the 5’ end, which in IPF11959 corresponds to a signal sequence recognized by SMART (Simple Modular Architecture Research Tools, http://smart.embl-heidelberg.de/smart/show_motifs.pl ). IPF11959 has regions with homology to those in S. cerevisiae YGR046w, which encodes an essential protein of unknown function and whose mRNA is targeted to the bud via a mRNA transport system (159). YGR046w has been speculated by two-hybrid assays to be involved in regulation of transcription, DNA-dependent regulation of protein biosynthesis, and mitochondrial matrix protein import. The role of C. albicans IPF11959 in pathogenesis has not yet been determined. ALG5. C. albicans ALG5 is a putative membrane dolichyl-phosphate glucosyltransferase involved in cell wall mannan biosynthesis in the endoplasmic reticulum. Its name and function were assigned by homology to the S. cerevisiae ALG5 protein (48.6% identity at the amino acid level) and by the presence of a glycosyl transferase 2 domain . The C. albicans ALG5 is a member of a large family group consisting of seven ALG genes. Neither its phenotype nor its role in pathogenesis has been elucidated yet.

PAGE 80

62 IPF15632. The Mai clone matched with the predicted ORF IPF15632, whose function is not known. IPF15632 has no homology with any gene in the S. cerevisiae genome database. The role of IPF15632 in C. albicans pathogenesis remains to be studied. In Vivo Expression of Genes Next, we determined if the genes identified by IVIAT were expressed and/or up-regulated during OPC compared to routine growth in the laboratory. We studied RBF1 and IPF11959 mRNA and protein expression in C. albicans SC5314 strain grown in YPD medium and in C. albicans isolates in thrush samples obtained from the oral cavity of HIV-infected patients. Verification of RBF1 and IPF11959 mRNA Expression in Thrush Samples by RT-PCR Transcript expression was determined by semi-quantitative RT-PCR. C. albicans RNA was extracted from thrush samples and from SC5314 overnight cultures grown in YPD at 37 C. After treating the RNA with DNaseI, its quality was confirmed by formaldehyde agarose gel electrophoresis. Equal concentrations of total RNA were used for transcription. The cDNA was normalized to an internal control, elongation factor 1 beta (EF1). EF1 has been shown to be constitutively expressed and an appropriate internal control with high levels of transcripts detected in different conditions and at different time points (154). Furthermore, PCR amplification with EF1 primers allowed us to verify that no genomic DNA contamination was present in the RNA samples. A 891 bp fragment was amplified from genomic DNA since it contains an intron of 365 bp size; otherwise, a 526 bp fragment was observed for the cDNA. All three genes (EF1, RBF1, and IPF11959) were found to be expressed in thrush samples as observed by the

PAGE 81

63 presence of the expected PCR bands sizes (526 bp, 357 bp, and 225 bp respectively) in agarose gels (See Fig 3-4). When normalizing the RBF1 cDNA PCR band to that of EF1 we found that the former was expressed in thrush samples at higher levels (about 2.5 times more) than when grown in liquid media. These results were consistent when repeated three times. However, expression of IPF11959 was not elevated after normalization to EF1. Furthermore, other C. albicans genes identified by our laboratory using IVIAT, including CDC24, were expressed at higher levels during OPC as compared to routine growth in liquid media (27). vitrovivovivovivovitrovitro EF1IPF11959RBF1M0.5 kb vitrovivovivovivovitrovitro EF1IPF11959RBF1M0.5 kb Fig 3-4. In vivo and in vitro expression of IPF11959 and RBF1. Agarose gel with RT-PCR samples run for EF1, RBF1 and IPF11959. Vivo represents RT-PCR from thrush samples and vitro represents RT-PCR from C. albicans SC5314 grown in YPD at 37 C. M represents the molecular weight marker. Verification of IPF11959p expression within Thrush Samples by Immunostaining IPF11959 protein expression during OPC was demonstrated by immuno-peroxidase staining of C. albicans in pseudomembranes of thrush samples.

PAGE 82

64 IPF11959-His tagged protein fragment and anti-IPF11959 antibody A fragment of IPF11959 protein (132 aa) fused to a His-tag at the 3’end was expressed using the pET30EK/LIC vector in the E. coli BL21(DE3) strain. The His-tag IPF11959 protein fragment was purified by nickel column chromatography and used to raise polyclonal antibodies in mice. The purified polyclonal mouse anti-IPF11959 IgG was confirmed to react against two IPF11959 protein fragments expressed in bacteria (Ma2aSub and IFPTH subclones) by Western blot. Reactivity to IFPTH subclones are shown in Fig 3-5. w/o IPTG His-tagged purified IPF11959IFPTH subclonelysateexpressing IPF11959E. colilysate(Neg. Ctrl)w/o IPTGw/ IPTGw/ IPTG w/o IPTG His-tagged purified IPF11959IFPTH subclonelysateexpressing IPF11959E. colilysate(Neg. Ctrl)w/o IPTGw/ IPTGw/ IPTG Fig 3-5. Reactivity of mouse polyclonal antibody against IPF11959 protein fragment. Western blot with IPF11959 subclone (IFPTH) lysate and His-tag purified IPF11959 protein fragment probed with polyclonal mouse anti-IPF11959 antibody. Immunostaining of thrush samples Thrush samples were collected and prepared for immunostaining on glass slides by heat fixation. The slides were incubated with the purified polyclonal anti-IPF11959 antibody and then stained with hematoxylin for visualization of C. albicans cells.

PAGE 83

65 IPF11959 protein was shown to be present in a beaded pattern along the cell wall of hyphae. IPF11959 protein was not detected in yeast forms in thrush samples (Fig3-6a). The presence of both yeast and hyphae forms were observed by silver staining of thrush samples obtained from the same patients (Fig3-6b). Negative controls included slides not incubated with primary antibody and slides with non-infected buccal mucosa. A)B) A)B) Fig 3-6. Expression of C. albicans IPF11959 within pseudomembranes from HIV-infected patients with OPC. A) Immuno-peroxidase staining of hyphae forms with mouse polyclonal anti-IPF11959. Red arrows point to stained IPF11959 protein. B) Silver staining of pseudomembranes from thrush sample. Red arrow points to hyphae and pseudohyphae forms and blue arrows point to yeast forms.

PAGE 84

CHAPTER 4 DETERMINATION IF GENES ENCODING ANTIGENIC PROTEINS EXPRESSED IN VIVO ARE IMPORTANT FOR CANDIDA ALBICANS VIRULENCE We sought to ascertain if the five genes encoding antigenic proteins during OPC were in fact virulence factors. To fulfill this goal, we followed a strategy that consisted of three steps. First, we disrupted the gene in the background of the C. albicans CAI-4 strain and created isogenic mutants. Second, we analyzed the phenotype of the mutants in vitro. Third, we tested the mutants’ virulence using cell and murine models of candidiasis. This chapter will go over this strategy and will narrow down the disruption process to two genes, ALG5 and IPF15632. Further characterization of these genes will be described in Chapters 5 and 6, respectively. Step 1: Creating Isogenic Mutants Prior to creating isogenic mutants in the uraCAI-4 strain, derived from SC5314, we had to determine which of the genes identified ought to be studied. From the five genes identified, I chose to concentrate the rest of my work on two genes, ALG5 and IPF15632, in order to accomplish my goals and prove my hypothesis within a reasonable amount of time. However, I established which would be the best restriction endonuclease for analyzing the restriction fragments pattern for the parent and corresponding mutants for the other genes by Southern blot. Next, we created the disruption vectors to disrupt ALG5 and IPF15632 using the URA-blaster protocol. This protocol was followed by re-insertion of the URA3 gene at its original locus to create isogenic null mutants. 66

PAGE 85

67 Reasoning for Selecting ALG5 and IPF15632 for Pathogenesis Study I chose to study C. albicans ALG5 because of its putative function in N-linked glycosylation of cell wall proteins, which was assigned by homology to S. cerevisiae ALG5. The cell wall along with the glycosylated cell wall proteins provide many important cellular function, such as protecting against mechanical stress, modulating the selective uptake of macromolecules or eliminating toxic substances, providing cell morphology, and participating in the initial interaction of C. albicans with the host tissues and immune system. For these reasons, we hypothesized that ALG5 might be important in C. albicans pathogenesis. I also chose to study IPF15632 because it has no homologs with any gene in the S. cerevisiae database and other microorganisms genes in GenBank. This placed IPF15632 as a putative and novel virulence factor. Southern Blot Analysis for Identified Genes CDC24, IPF11959, ALG5, and IPF15632 sequences, including the upstream and downstream sequences available for each gene in the C. albicans databases were examined using Vector NTI program (VNTI Suite 9.0, Informax Inc.). Genomic DNA was cut with the endonucleases chosen for corroborating the band size expected by sequence analysis with the one obtained by Southern blot. CDC24 and RBF1. SC5314 genomic DNA cut with different restriction endonucleases was analyzed by Southern blot. Membranes probed with an -32P labeled CDC24F2 fragment showed the presence of one band for both alleles as predicted by sequence analysis (See Fig. 4-1A). However, the disruption process for CDC24 was halted due to the publication of work on this gene by another group. Therefore, since

PAGE 86

68 RBF1 and CDC24 had been previously disrupted and identified as virulence factors by other laboratories, I did not continue to work on these genes. PTH1 (IPF11959). Southern blot analysis of genomic DNA cut with HpaI/AatII and hybridized with IPF11959F2 probe demonstrated 2 bands, a 1.67 kb band and a ~3.5 kb band (Fig 4-1B). The size of the 1.67 kb band was expected as shown from the DNA sequence analysis, but the other band could not be identified in the database. Further analysis with other restriction enzyme pairs for IPF11959 could not be continued at the time for the lack of sequence information. Nonetheless, the strategy for disrupting this gene was continued. However, Southern blot analysis of recombinants obtained from IPF11959 disruption did not matched the band pattern expected. For these reasons and the fact that IPF11959 was homologous to an essential gene from S. cerevisiae, I chose not to pursue work on this gene any further. ALG5. Southern blot with genomic DNA cut with two combinations of different restriction enzymes and hybridized with labeled ALG5-F2 probe confirmed the presence of one band as expected (Fig 4-1C). However, when the DNA was cut with HpaI/SacI the probe bound to two bands. We assumed that there was a polymorphism in the downstream sequence from ALG5, and that each band corresponded to one allele. IPF15632. Using genomic Southern analysis, DNA cut with PvuII/NsiI, which cut within the gene (at base 316) and just after the stop codon (at base +81) respectively, showed the presence of one band of approximately 1 kb. However, when DNA was cut with PvuII/SphI, where SphI cuts further downstream, the blot revealed 2 bands, a 3.5 kb band expected and an approximately 2.3 kb band not identified by sequence analysis (Fig 4-1D). When examining the sequences available for IPF15632 in the CandidaDB

PAGE 87

69 (Assembly 6) we found that there were two alleles (orf6.257 and orf6.961) present in two differently numbered contigs (contig6-1460 and contig6-1763). It appeared that each band in the Southern blots corresponded to one allele of IPF15632, the 3.5 kb band belonging to orf6.257 allele and the ~2.3 band corresponding to orf6.961 allele (see Fig4-1D). Similar patterns of two bands were also observed when performing Southern blots using different probes and restriction enzymes that cut further upstream and downstream. Expected size:3 kb2 kbCDC24 EcoRIHpaIHpaIAatII2.372.72.44A) 1.5 kbIPF119591 kb0.5 kb3 kb2 kbHpaIAatII1.67B)ALG5 HpaIAatIIAccIHpaISacI1.52.661.37C)IPF15632 PvuIINsiIPvuIISphI0.953.5D)Expected size:3 kb2 kbCDC24 EcoRIHpaIHpaIAatII2.372.72.44A)3 kb2 kbCDC24 EcoRIHpaIHpaIAatII2.372.72.44 EcoRIHpaIHpaIAatII2.372.72.44A) 1.5 kbIPF119591 kb0.5 kb3 kb2 kbHpaIAatII1.67B) 1.5 kbIPF119591 kb0.5 kb3 kb2 kbHpaIAatII1.67B)1.5 kbIPF119591 kb0.5 kb3 kb2 kbHpaIAatII1.67B)ALG5 HpaIAatIIAccIHpaISacI1.52.661.37C)ALG5 HpaIAatIIAccIHpaISacI1.52.661.37C)IPF15632 PvuIINsiIPvuIISphI0.953.5D)IPF15632 PvuIINsiIPvuIISphI0.953.5D) Fig 4-1. Southern blot analysis for genes identified using IVIAT. A) CDC24 hybridized with CDC24F2 ( bp), B) IPF11959 hybridized with KOF2 ( bp), C) ALG5 hybridized with ALG5-F2 (204 bp), D) IPF15632 hybridized with MaiF2 (242 bp). All the probes were labeled with -32P-dATP by random priming. Overview of the URA-Blaster Protocol C. albicans ALG5 and IPF15632 were disrupted using the URA-blaster protocol. This protocol consists of disrupting a gene in C. albicans using a standard hisG-URA3-hisG cassette in the genetic background of the ura3C. albicans CAI-4 strain. The hisG-URA3-hisG (HUH) cassette contains the C. albicans URA3 promoter and gene flanked by direct repeats of Salmonella typhimurium hisG. URA3, the selectable marker, encodes for orotidine-5’-monophosphate (OMP) decarboxylase and is required for candidal growth on media lacking uracil. The HUH cassette is integrated into the C. albicans genome by

PAGE 88

70 homologous recombination between the gene to be disrupted, for example YFG (Your Favorite Gene) in Fig. 4-2, and the two fragments of this gene, each cloned at one end of the hisG-URA3-hisG cassette (disruption cassette). The disruption cassette allows multiple disruptions to be done by repeated selection for and against URA3. When C. albicans Ura+ strains are grown on media containing 5-fluoroorotic acid (5-FOA), OMP decarboxylase converts 5-FOA to 5-fluorouracil, which is a toxic product. At some spontaneous frequency, the hisG direct repeats undergo intrachromosomal recombination and eliminate the URA3 gene, leaving behind a single copy of hisG at the disruption site. Growth on medium containing 5-FOA selects for cells in which this recombination event has occurred. Recycling of the URA3 marker to perform a second disruption allows for the creation of a double knockout mutant. Recombinants obtained at each step are analyzed by Southern blot. (See Figure 4-2) Disruption strategy for ALG5 C. albicans ALG5 is a putative membrane protein with one transmembrane spanning domain (codons 2-24; not shown in Fig. 4-3A). It has a putative N-terminal signal sequence located in the first 19 codons as depicted by a red square in Fig. 4-3A. There are also two intrinsic disordered regions (codons 40-52 and codons 309-323) shown with blue squares and a glycosyl transferase 2 domain (E value 3.10e-25) located between codons 67 through 245. The signal sequence and domains were identified using SMART (Simple Modular Architecture Research Tools). (See Fig. 4-3A) Two fragments within ALG5, F1 (203 bp long) covering codons 23 to 90 and F2 (204 bp long) covering codons 215 to 283, were each cloned at one end of the HisG-URA3-HisG cassette in the pMB7 vector as shown in Fig. 4-3B. By homologous

PAGE 89

71 Ura+Heterozygote recombinant hisG URA3 hisG F1 F2 YFG Selection for ura-heterozygote recombinant 5-FOA SelectionDisruption of YFG 2ndalleleby homologous recombination Repeat Transformation hisG F1 F2 YFG ura-Heterozygoterecombinant hisG URA3 hisG F1 F2 hisG F1 F2 Ura+/yfg C. albicansura-strain Disruption of YFG 1stallele by homologous recombinationYFG-HUH/pMB7 hisG URA3 hisG F1 F2 YFG YFG Transformation with linearized disruption cassetteUra+Heterozygote recombinant hisG URA3 hisG F1 F2 YFG Ura+Heterozygote recombinant Ura+Heterozygote recombinant hisG URA3 hisG F1 F2 YFG hisG URA3 hisG F1 F2 hisG URA3 hisG F1 F2 hisG URA3 hisG F1 F2 YFG YFG YFG Selection for ura-heterozygote recombinant 5-FOA SelectionSelection for ura-heterozygote recombinant 5-FOA SelectionDisruption of YFG 2ndalleleby homologous recombination Repeat TransformationDisruption of YFG 2ndalleleby homologous recombination Repeat Transformation hisG F1 F2 hisG F1 F2 YFG ura-Heterozygoterecombinant YFG YFG ura-Heterozygoterecombinant hisG URA3 hisG F1 F2 hisG F1 F2 Ura+/yfg hisG URA3 hisG F1 F2 hisG F1 F2 hisG URA3 hisG F1 F2 hisG URA3 hisG F1 F2 hisG URA3 hisG F1 F2 hisG F1 F2 hisG F1 F2 Ura+/yfg Ura+/yfg C. albicansura-strain Disruption of YFG 1stallele by homologous recombinationYFG-HUH/pMB7 hisG URA3 hisG F1 F2 YFG YFG Transformation with linearized disruption cassetteC. albicansura-strain C. albicansura-strain Disruption of YFG 1stallele by homologous recombinationYFG-HUH/pMB7 hisG URA3 hisG F1 F2 YFG YFG Transformation with linearized disruption cassette Disruption of YFG 1stallele by homologous recombinationYFG-HUH/pMB7 hisG URA3 hisG F1 F2 hisG URA3 hisG F1 F2 YFG YFG YFG YFG YFG YFG Transformation with linearized disruption cassette Fig 4-2. URA blaster protocol. This method describes how a gene in C. albicans, for example YFG (Your Favorite Gene), is disrupted using the hisG-URA3-hisG disruption cassette. F1 and F2 are fragments within YFG are cloned at the ends of HUH in pMB7 to create the disruption cassette. recombination the ALG5-HUH disruption cassette would remove approximately 375 bp within ALG5 including most of the glycosyl transferase 2 domain. Disruption strategy for IPF15632 C. albicans IPF15632 has a possible coiled-coil region between codons 17 and 43 as shown with a green square in Fig. 4-4B. We also identified a low complexity region (codons 79 to 95) depicted with a purple square and 4 intrinsically disordered regions

PAGE 90

72 A) ALG5HUH7004 bp URA3 AmpR HisG HisG ALG5F1 ALG5F2 Glycosyl transferase 2Domain F1F2B)A) ALG5HUH7004 bp URA3 AmpR HisG HisG ALG5F1 ALG5F2 ALG5HUH7004 bp URA3 AmpR HisG HisG ALG5F1 ALG5F2 Glycosyl transferase 2Domain F1F2 Glycosyl transferase 2Domain Glycosyl transferase 2Domain F1F2B) KpnI PstI SacI SalI KpnI PstI SacI SalI KpnI PstI SacI SalI Fig. 4-3. Disruption strategy for ALG5. A) Schematic diagram for ALG5 protein with putative domains analyzed by SMART. The red box represents a possible signal sequence, blue boxes represent intrinsic disordered regions, and the gray box represents the Glycosyl transferase 2 domain (E value 3.10e-25) The fragments F1 and F2 cloned into pMB7 vector are shown as black lines. B) ALG5-HUH disruption cassette in pMB7. (codons 120 to 200, 206 to 217, 255 to 318, and 380 to 395) as shown with blue squares in Fig. 4-4A. All these domains and regions were identified using SMART. Two fragments within IPF15632, F1 (210 bp long) expanding codons 73 to 143 and F2 (242 bp long) expanding codons 299 to 380, were each cloned at one end of the HisG-URA3-HisG cassette in pMB7 as shown in Fig. 4-4B. This disruption cassette, IPF15632-HUH, would remove approximately 470 bp within IPF15632. (See Fig. 4-4)

PAGE 91

73 A second disruption cassette was created in which the F1 fragment, UpF1 (from -230 bp to -19 bp; 212 bp long), was located upstream of both alleles. This cassette allowed the excision of approximately 910 bp of IPF15632 ORF. The UpF1 fragment is not shown in Fig. 4-4A but it is shown in Fig. 6-4 and 6-5 in Chapter 6. IPF15462HUH7044 bp URA3 AmpR HisG HisG MaiF1 MaiF2 B) F1F2A) IPF15462HUH7044 bp URA3 AmpR HisG HisG MaiF1 MaiF2 IPF15462HUH7044 bp URA3 AmpR HisG HisG MaiF1 MaiF2 B) F1F2 F1F2A) KpnI SacI SalI SphI KpnI SacI SalI SphI KpnI SacI SalI SphI Fig 4-4. Disruption strategy for IPF15632. A) Schematic diagram for IPF15632 protein with putative domains analyzed by SMART. The green box represents a coil-coil region, the purple box represents a region of low complexity, and the blue boxes represent intrinsic disordered regions. The fragments F1 and F2 cloned into pMB7 vector are shown as black lines. B) IPF15632-HUH disruption cassette in pMB7.

PAGE 92

74 Step 2: Analyzing the Mutants In Vitro After disrupting ALG5 and IPF15632, isogenic null mutants were analyzed under a variety of different conditions in vitro in order to determine if there was any gain or loss of function compared to wild type. First, I determined the growth rate in YPD at 30 and 37 C in batch cultures. If the mutant displayed significant growth defect under several conditions in vitro, I will not proceed to assay the mutants using murine models. Since, the growth defect rather than the disruption of the gene itself might attribute to the mutant’s inability to cause an infection in mice. Second, I determined the ability of the mutants to filament on hyphae-inducing liquid and solid media. This experiment is important because the C. albicans ability to undergo yeast-to-hyphae switching has been linked to virulence. Our objective was to identify a mutant phenotype that would be relevant to pathogenesis in murine models of candidiasis. Also, the phenotypes observed could pinpoint or clarify the function of the gene disrupted. All the mutants were compared with the CAI-12 strain, derived from CAI-4 after re-insertion of one copy of URA3 back at its original locus. Step 3: Determining the Virulence of the Mutants Several animal models have been developed to study C. albicans pathogenesis at the molecular and cellular level, including clinically relevant murine models for: systemic candidiasis (standard DC murine model); gastrointestinal (infant mouse model GI translocation); OPC: (cortisone immunosuppressed and transgenic mice expressing HIV-1); VC (estrogen model). There are models of transgenic and KO mice simulating specific host deficiencies. (35)

PAGE 93

75 We chose to determine our mutants’ virulence in the two main manifestations of candidiasis: disseminated and oropharyngeal candidiasis. We used the standard murine model of DC for determining if the alg5 or IPF15632 null mutants were still as virulent as the wild type strain (CAI-12). This murine model of DC consists of 106 CFU (colony forming units) injected into the lateral tail vein. This model has been used extensively to identify and determine C. albicans known virulence factors, since it is reproducible and technically simple. ICR mice injected with 105 to 106 CFU of the wild type strain usually succumb to the infection within 3 to 9 days. Candida cells are collected post mortem from the kidneys, liver, and spleen. In addition, we tested the alg5 and IPF15632 mutants using cellular models of OPC, in which adherence to epithelial cells, either primary human cells or human cell lines, was assayed. Adherence to epithelial cells was chosen due to the fact that this is the first critical step that C. albicans must perform in order to colonize and invade mucosal tissues. If a gene displayed reduced adherence to epithelial cells compared to the wild type, then we would test the respective mutants in a recently developed murine model of OPC (90). Given the fact that our screening strategy (IVIAT) uses sera from patients with OPC, we thought this murine model of OPC has potential relevance for studying the genes we identified. In this model, cortisone-immunosuppressed mice are infected orally with 108 CFU. Consistently, mice infected with the wild type strain became very ill beginning with wasting within 3-5 days post-infection, occasionally leading to death. Unfortunately, this model has considerable variation due to the method of infection (see Materials and Methods).

PAGE 94

CHAPTER 5 CHARACTERIZATION OF C. ALBICANS ALG5 Isolation and Analysis of ALG5 Isolation and Identification of ALG5 We used In Vivo Induced Antigen Technology (IVIAT) to identify C. albicans antigenic proteins expressed during oropharyngeal candidiasis (OPC) of HIV-infected patients. IVIAT uses sera from patients with OPC, previously depleted for antibodies against C. albicans proteins expressed in culture, to screen a C. albicans genomic expression library (27). One of the proteins identified was ALG5 in the CandidaDB (Institute Pasteur, Paris). ALG5, also referred as orf19.2837/orf19.10355 in Assembly 19, is annotated as a putative membrane UDP-glucose:dolichyl-phosphate glucosyltransferase involved in cell wall mannan biosynthesis. C. albicans ALG5 name and function were assigned by homology to the S. cerevisiae ALG5/YPL227c ortholog (3e-87; 47.4% identity). S. cerevisiae ALG5 is a transmembrane enzyme (334 aa) located in the endoplasmic reticulum (ER) and involved in N-linked glycosylation of proteins. This enzyme catalyzes the transfer of glucose from UDP-glucose to dolichol phosphate.(73) Furthermore, the ScALG5 null mutant is viable (56), thus suggesting that C. albicans ALG5 can be disrupted. In this chapter we describe C. albicans ALG5, its disruption, phenotypic characterization of alg5 null mutant, and its role on pathogenesis. We hypothesized that ALG5 would have a role in pathogenesis since this gene is involved in cell wall mannan biosynthesis and mannan is the first fungal component to come in contact with the host immune response. 76

PAGE 95

77 Sequence Analysis of ALG5 C. albicans ALG5 encodes a 324 amino-acid polypeptide with a predicted molecular mass of 36.7 kDa. Sequence analysis of ALG5 with SMART revealed a putative signal sequence, one transmembrane spanning domain, and a glycosyl transferase 2 domain (see Chapter 4). This type of glycosyl transferase 2 domain is found in a diverse family of glycosyltransferases involved in the transfer of sugar from UDP-glucose, UDP-N-acetyl-galactosamine, GDP-mannose or CDP-abequose, to various substrates such as cellulose, dolichol phosphate and teichoic acids (18, 79). According to the CandidaDB there are 21 C. albicans genes that have been directly annotated to N-linked glycosylation and 7 of these genes belong to the ALG family. C. albicans ALG5 is located in chromosome R and within contig10161 of Assembly 19. Analysis of contig10161 provided further information of genes adjacent to ALG5 and restriction enzymes that were best for Southern blot analysis (Fig. 5-1). We chose to analyze the recombinants obtained during the disruption process using HpaI and AatII (See Chapter 4, Fig. 4-1). F2 F1 ALG5 1.5 kb F2 F1 ALG5 1.5 kb AatII HpaI AatII HpaI Fig. 5-1. ALG5 ORF within contig10161. ALG5 is shown as a purple arrow. The size of the fragment to be analyzed by Southern blot is indicated by dotted lines between HpaI and AatII RE sites. Creation of ALG5 Isogenic Mutants Disruption of ALG5 To characterize ALG5 we constructed a null mutant in the background of the uraCAI-4 strain using the URA-blaster method developed by Fonzi and Irwin (49) and the

PAGE 96

78 ALG5-HUH disruption cassette (see Materials and Methods and Chapter 4). The first round of transformation to disrupt one allele of ALG5 yielded 54 colonies on SD-ura plates. From these 54 colonies, 19 were screened, and 4 clones (# 9, 10, 12, and 14) had one allele disrupted as observed by a shift in the 1.5 kb band to about 5.5 kb in Southern blots (Fig. 5-2). Disruption of the second allele of ALG5 in the uraheterozygote recombinants, clones 12F1 and 14F1, yielded approximately 100 and 250 colonies respectively after selection on SD-ura plates. About 120 colonies were screened by colony PCR and Southern blot and only 2 Ura+ recombinants (clones 58 and 62) had both alleles disrupted appropriately. The disruption of both alleles can be observed in Southern blots in Fig. 5-2. The genotypes for ALG5 mutants are indicated in Table 5-1. ALG5alg5::HUHalg5::HisG1 Copy KO2 Copy KO12F1/6212F1/62F4 HpaI and AatIIALG5F2-32P 1212F1 8 kb6 kb5 kb4 kb3 kb2 kb1.5 kb M ALG5alg5::HUHalg5::HisG1 Copy KO2 Copy KO12F1/6212F1/62F4 12F1/6212F1/62F4 HpaI and AatIIALG5F2-32P 1212F1 8 kb6 kb5 kb4 kb3 kb2 kb1.5 kb M1212F1 8 kb6 kb5 kb4 kb3 kb2 kb1.5 kb M 8 kb6 kb5 kb4 kb3 kb2 kb1.5 kb M Fig. 5-2. Southern blot with ALG5 disruption mutants. A 1 kb molecular weight marker (M) was included for each blot.

PAGE 97

79 Table 5-1. ALG5 recombinants Strain name Genotype 12 ALG5/alg5::hisG-URA3-hisG 12F1 ALG5/alg5::hisG 12F1 + U1 ALG5/alg5::hisG/URA3 12F1/62 alg5::hisG-URA3-hisG/alg5::hisG 12F1/62F4 alg5::hisG/alg5::hisG 12F1/62F4 + U2 alg5::hisG/alg5::hisG/URA3 The first number indicating the Ura+ clone with one allele disrupted is followed by the urarecombinant number selected with 5-FOA (F). U1 and U2 are the clones obtained after re-insertion of URA3 to its original locus. Re-insertion of URA3 at its Original Locus In order to construct isogenic mutants and prevent unwanted results due to the positional effects on URA3 expression (8, 97), we restored URA3 to its native locus. Restoration of URA3 was carried out for each urarecombinant by homologous recombination of a linear fragment containing the URA3 gene and its promoter flanked by lambda phage DNA ( imm434), which was used for disruption of URA3 in the CAI-4 strain (49) (Fig. 5-3A). ALG5 Ura+ recombinants were analyzed by Southern blot using an -32P-labeled probe specific for URA3 (Fig. 5-3B). The restriction fragment length polymorphism observed in the Southern blot 4.8 kb (middle) band for CAI-4 compared to SC5314 is due to the insertion of an EcoRI site along with imm434 DNA; this restriction enzyme is used for analyzing the recombinants (49). Re-insertion of URA3 is observed by the presence of a band at about 2.3 kb, this band is missing in the uraCAI-4 strain. ALG5 heterozygote and homozygote, Ura+ and ura-, recombinants attained (see Table 5-1) were analyzed subsequently.

PAGE 98

80 NheIBglIIStyIAvaIIEcoRVNheIBglIIEcoRVScaIEcoRI EcoRIHindIIIXbaIPstIEcoRI lmm434 StyIAvaIIXbaIPstI URA3CAI4ura3/ura3CAI-12URA3/ura3A)B) -32P-pURA3 + URA3 Copy KO EcoRISC12CAM NheIBglIIStyIAvaIIEcoRVNheIBglIIEcoRVScaIEcoRI EcoRIHindIIIXbaIPstIEcoRI lmm434 StyIAvaIIXbaIPstI URA3CAI4ura3/ura3CAI-12URA3/ura3 NheIBglIIStyIAvaIIEcoRVNheIBglIIEcoRVScaIEcoRI EcoRIHindIIIXbaIPstIEcoRI lmm434 StyIAvaIIXbaIPstI URA3 NheIBglIIStyIAvaIIEcoRVNheIBglIIEcoRVScaIEcoRI EcoRIHindIIIXbaIPstIEcoRI lmm434 StyIAvaIIXbaIPstI URA3CAI4ura3/ura3CAI-12URA3/ura3A)B) -32P-pURA3 + URA3 Copy KO EcoRISC12CAM -32P-pURA3 + URA3 Copy KO EcoRISC12CAM Fig. 5-3. Re-insertion of URA3 into its native locus. A) Schematic diagram for re-insertion of URA3. This diagram shows the re-insertion of URA3 for the CAI-12 strain. Re-insertion of URA3 for the uraalg5 mutants is carried the same way as shown for CAI-12 B) Southern blot of Ura+ recombinants for alg5 mutants. The SC5314 (SC) strain contains both copies of URA3, while in the CAI-4 (CA) strain both alleles of URA3 have been deleted with Imm434 DNA. ALG5 recombinants with one allele (1 copy KO) and both alleles (2 copy KO) disrupted have one copy of URA3 re-inserted back at its original locus. Phenotypic Analysis of Alg5 Null Mutant Null mutants for ALG5 were phenotypically analyzed and compared to the wild type CAI-12 strain. We did not observe any difference in the growth rate between mutants and wild type when grown in rich media (YPD) at both 30 and 37 C (Fig. 5-5). Neither did we observe any difference in the rate and shape of hyphae formation for the isogenic alg5 null mutant when compared to CAI-12.

PAGE 99

81 log OD600Time (hr)30C -1-0.500.51024681012 wild type alg5mutant 1 alg5mutant 2 -1-0.500.51024681037CTime (hr) 12 wild type alg5mutant 1 alg5mutant 2log OD600Time (hr)30C -1-0.500.51024681012 wild type alg5mutant 1 alg5mutant 2log OD600Time (hr)30C -1-0.500.51024681012 -1-0.500.51024681012 wild type alg5mutant 1 alg5mutant 2 wild type alg5mutant 1 alg5mutant 2 -1-0.500.51024681037CTime (hr) 12 wild type alg5mutant 1 alg5mutant 2 -1-0.500.510246810 12 -1-0.500.51024681037CTime (hr) 12 wild type alg5mutant 1 alg5mutant 2 wild type alg5mutant 1 alg5mutant 2 Fig. 5-4. Growth curve of alg5 mutants in YPD broth. Two different alg5 mutants were compared to the wild type. Cell Wall Changes of alg5 Null Mutant Since ALG5 is possibly involved in N-linked glycosylation of cell wall proteins, we screened the mutants with an array of cell wall effectors. ALG5 KO phenotype in response to calcofluor white and SDS We determined if the alg5 null mutant is susceptible to calcofluor white. Calcofluor white is a fluorescent dye that binds chitin and has pleiotropic effects on mutants with general cell wall defects. The alg5 null mutant was not able to grow as well as the wild type at a spotted density of 103 CFU (Fig. 5-6A) in presence of calcofluor white (50 g/mL). Another cell wall effector used to screen for cell wall changes is SDS, an ionic detergent. The alg5 null mutant was slightly resistant to 0.06% SDS at both 30 and 37 C when compared to the wild type CAI-12 strain. The null mutant’s resistance to SDS is observed by the presence of more colonies at a plating density of 104 CFU when compared to the wild type strain.

PAGE 100

82 CAI-12 (wt)Calcofluor White 105104103102 CFUalg5 + U2CAI-12alg5 + U2CFUCFU 0.06% SDS (30 C)105104103102 0.06% SDS (37 C)105104103102 CAI-12 (wt)Calcofluor White 105104103102 Calcofluor White 105104103102 CFUalg5 + U2CAI-12alg5 + U2CFUCFU 0.06% SDS (30 C)105104103102 0.06% SDS (30 C)105104103102 0.06% SDS (37 C)105104103102 0.06% SDS (37 C)105104103102 Fig. 5-5. Phenotypic analysis of alg5 null mutant with cell wall effectors. A) Calcofluor white (50 g/mL) in YPD plates. Plates were incubated for 3 days at 37 C. B) 0.06% SDS in YPD plates. Plates were incubated at 30 and 37 C for 3 days. CFUs for each strain were spotted as indicated in the figure. Susceptibility of ALG5 null mutant to aminoglycosides Susceptibility to aminoglycosides, such as G418 or Hygromycin B, has been reported for glycosylation deficient yeast (37). S. cerevisiae alg mutant have been shown to be susceptible to growth on Hygromycin B and G418 (37). Therefore, we determined if this was also the case for the C. albicans alg5 null mutant. We found that the alg5 mutant was more susceptible to G418 (1 mg/mL) than wt. This was scored by inhibition of growth of the null mutant relative to that of the wild type. (Fig. 5-7). Similar results were also observed when mutants were screened for Hygromycin B (200 ug/mL) susceptibility (data not shown).

PAGE 101

83 YPD (Control)G418 (1 mg/ml) alg5 + U2wt alg5 + U2wt A)B)YPD (Control)G418 (1 mg/ml) alg5 + U2wt alg5 + U2wt alg5 + U2wt alg5 + U2wt A)B) Fig. 5-6. Alg5 null mutant susceptibility to G418. A) Control plate with 3000 CFU streaked for each strain. B) YPD plate containing G418 (1 mg/mL) with 3000 CFU streaked for each strain. Plates were incubated for 3 days at 37 C. Evaluation of ALG5 role in Pathogenesis Cell Adherence Assays as Cell Models for OPC Glycosylated cell wall proteins are known to contribute to C. albicans adherence to host tissues (168, 169). In view that deletion of ALG5 lead to slight changes in the cell wall including underglycosylation of cell wall proteins, we measured the mutants’ adherence to primary human buccal epithelial cells. We chose an oral epithelial cell adherence assay as a model for OPC, since adherence is the first crucial step for establishing colonization and invading the oral mucosa. The results indicated that the heterozygous and null mutants were able to adhere (135% and 190% respectively) as well as the wild type (100%) to primary human buccal epithelial cells. Therefore, the cell wall defects observed for alg5 null mutant were not enough to decrease adherence to human buccal epithelial cells. However, ALG5 might have a role in adherence to other human epithelial cells. Murine Model of DC Finally, we evaluated the role of ALG5 during pathogenesis using the standard murine model of disseminated candidiasis (DC). We inoculated ICR mice intravenously

PAGE 102

84 with 105 CFU and followed the mice survival rate for a period of 20 days. We found no difference in mortality rate between the alg5 null mutant relative to the wild type CAI-12 (P=NS) (Fig. 5-8A). Time (Days)% Survival 02040608010002468101214161820 alg5 KO CAI-12Time (Days)% Survival 02040608010002468101214161820 02040608010002468101214161820 alg5 KO CAI-12 alg5 KO CAI-12 Fig. 5-8. Murine model of DC for alg5 mutant. A) Survival plot of mice injected i.v. with 105 CFU: alg5 null mutant + URA3 (blue) and wild type CAI-12 (yellow).

PAGE 103

CHAPTER 6 CHARACTERIZATION OF C. ALBICANS IPF15632 Identification and Sequence Analysis of C. albicans IPF15632 Isolation and Identification of IPF15632 We sought to identify genes expressed during oropharyngeal candidiasis (OPC) using In Vivo Antigen Expression Technology (IVIAT) in order to identify new potential virulence factors. Serum from HIV-infected patients with OPC was depleted for antibodies against C. albicans proteins expressed under laboratory conditions in YPD media. Using the adsorbed sera we screened a genomic expression library of C. albicans.(27) One of the proteins identified is IPF15632 in the CandidaDB (Institute Pasteur, Paris). IPF (individual protein file) is the systematic name given by the CandidaDB to genes not previously characterized and with no significant homology to any Saccharomyces cerevisiae gene. In this chapter I describe C. albicans IPF15632, its disruption, phenotypic characterization of the null mutant, and its role on pathogenesis. Because its function has not been previously characterized, I decided to pursue IPF15632. Sequence Analysis of IPF15632 Homologies and domains identified using BLAST IPF15632 is listed in the CandidaDB as weakly similar (~15% identical) to Saccharomyces cerevisiae MUC1/FLO11 gene, which encodes a cell wall protein involved in flocculation (cell-cell adhesion) and required for invasion and pseudohyphae formation in response to nitrogen starvation (52, 61, 96). However, we did not find a 85

PAGE 104

86 significant homology with any protein in the NCBI and S. cerevisiae databases with the exception of a 12 amino acid sequence (GADLLMYLATSP) that compromises codons 224 through 235 within the open reading frame. This short sequence was found to be highly conserved in S. cerevisiae Stb1p and other hypothetical proteins from different ascomycetes fungi, such as Neurospora crassa, Aspergillus nidulans, Magnaporthe grisea, and Gibberella zeae (Fig. 6-1A). This sequence appears to represent a consensus domain not yet studied. Stb1 (Sin three binding protein) is a trancriptional activator that interacts with Swi6 to regulate MBF (MCB-binding factor)-specific transcription at Start, the restriction point in yeast G1 phase that controls the rate of cell division. Swi6 is a component of the two heterodimeric transcription factors, MBF and SBF (SCB-binding factor) that are crucial for activation of transcription at Start. The MBF complex recognizes the MCB (Mlu1 cell cycle box) element and activates the S-phase cyclin genes and other DNA synthesis genes. The SBF complex binds the SCB (Swi4/6 cell cycle box) sequence element and activates G1 cyclins, HO endonuclease gene, Swe1 protein kinase, and many genes important for cell wall biosynthesis. Stb1 is thought to function in an alternate pathway from CLN3, a G1 cyclin, to activate G1-specific transcription. Studies by the Constanzo and Andrews laboratory at the University of Toronto are studying the mechanisms by which S. cerevisiae Stb1p activates SBFand MBF-dependent transcription. (31, 78) Domains identified using SMART and PSORTII IPF15632 encodes a putative 396 amino-acid soluble protein with a predicted molecular weight of 41.8 kDa. Further analysis of IPF15632 using SMART (Simple Modular Architecture Research Tools) exposed a coiled coil region (codons 17 through 43) as shown by a green square in Fig. 6-1B. Coiled coil structures are putative protein

PAGE 105

87 interaction domains found in some DNA-binding proteins, such as transcription factors (112). Within the coil-coil region, a bipartite nuclear localization (143) signal expanding codons 18 to 34 (orange square within the green one) was identified by PSORTII (Prediction of protein sorting signals and localization sites in eukaryotic amino acid sequences, http://www.psort.org/ (129)). A low complexity region depicted as a purple square and 4 intrinsically disordered regions presented as blue squares in Fig. 6-1B were also identified using SMART (see Chapter 4 for specific locations). Intrinsically disordered regions are nonstructural segments of protein that contain short linear motifs that are important for protein function, such as molecular recognition domains and protein folding inhibitors (40). IPF15632 contains several phosphorylation and N-glycosylation, and myristoylation sites within these disordered regions. Low complexity sequences within a protein are segments that contain few of the possible twenty amino acids, for example low complexity regions consist of repeats of single amino acids (161). In fact, IPF15632 contains repeats of glutamine and serine or proline, serine and alanine, and serine alone between codons 260 and 310 (not shown). Other three nuclear localization signals (classical type containing of 4 basic residues (77), each consisting of 6 amino acids, were discriminated by PSORTII and are located within the first intrinsical disordered region (codons 120 through 200). The unidentified conserved domain (GADLLMYLATSP) can be viewed as a yellow square in Fig. 6-1B. IPF15632 also has a possible ER membrane retention signal in the C-terminus identified by PSORTII. Results from PSORTII subprograms produced an 82.6% nuclear prediction for IPF15632. Therefore, the algorithms used in PSORTII predict, based on the protein sequence analysis, that there is an 82.6% chance this protein will localized to the nucleus.

PAGE 106

88 CaScNcAnMgGzConserved seqA)B) CaScNcAnMgGzConserved seq CaScNcAnMgGzConserved seqA)B) Fig. 6-1. Sequence analysis of IPF15632. A) Multialign showing the unidentified conserved domain (in red) in the following order: C. albicans (Ca) IPF15632 protein, 393 aa; S. cerevisiae (Sc) Stb1, 420 aa; N. crassa (Nc) EAA3034.1, 385 aa; A. nidulans (An) AN4881.2, 361 aa; M. grisea (Mg), 405 aa; and G. zeae (Gz) FG06368.1, 349 aa. B) Schematic protein sequence of IPF15632 with its domains, regions, and motifs identified by SMART and PSORTII. IPF15632 protein ORF is shown in light blue, coiled-coil region in green, low-complexity region in purple, intrinsically disordered regions in blue, nuclear localization signals in orange, and the unidentified conserved domain in yellow. Analysis of IPF15632 Alleles Analysis of IPF15632 in the CandidaDB (Institute Pasteur, Paris) in Assembly 6 revealed two alleles in the reference strain SC5314, orf6.257 and orf6.961 (see Fig. 6-2A), the latter being listed at the time as a false allele. Orf6.961 was considered a false allele due to incomplete sequencing data at the 5’ end of the ORF, which led to a presumed 195 bp N-terminal truncation. The false allele open reading frame (ORF) starts at the second methionine codon in the same frame as the full size allele ORF (1179 bp). Both alleles could be distinguished in Southern blots, a 3.5 kb band corresponding to orf6.257 and a ~2.3 kb band corresponding to orf6.961 (see Fig. 6-2B).

PAGE 107

89 F1F2 Unknown Sequence Allele BRegion of identity Contig6-1640Orf6.257 (1182 bp)Contig6-1763“False” orf6.961 (984 bp)Unknown Sequence F1F2Allele A 4 kb3 kb2 kballele Ballele A A)B) F1F2 Unknown Sequence Allele BRegion of identity Contig6-1640Orf6.257 (1182 bp)Contig6-1763“False” orf6.961 (984 bp)Unknown Sequence F1F2Allele A 4 kb3 kb2 kballele Ballele A F1F2 Unknown Sequence Allele B F1F2 Unknown Sequence Allele BRegion of identity Contig6-1640Orf6.257 (1182 bp)Contig6-1763“False” orf6.961 (984 bp)Unknown Sequence F1F2Allele A Unknown Sequence F1F2Allele A 4 kb3 kb2 kballele Ballele A 4 kb3 kb2 kballele Ballele A 4 kb3 kb2 kballele Ballele A A)B) PS P PS P PS PS P P Fig. 6-2. IPF15632 alleles in Assembly 6. A) Schematic diagram showing region of identity in brackets between both alleles within their corresponding contigs. Probes F1 and F2 are shown as blue rectangles on the schematic diagram of both alleles. The green rectangle represents regions (~500 bp) that were not yet sequenced. B) Southern blot with both alleles of IPF15632. Genomic DNA of SC5314 was cut with PvuII (P) and SphI (S) and hybridized with 32P-labebed F2 probe (RE sites and probe location are shown in Fig. 6-2A). Identification of a full sequence for the so-called “false” orf6.961 allele To assess the significance of the false allele, we determined if both alleles could be distinguished in different clinical isolates of C. albicans. We performed Southern blot analyses on 11 strains from patients with disseminated candidiasis (DC) and 9 from patients with OPC. As shown in Fig 6-2, Southern blot analysis with PvuII and SphI allowed us to discriminate between both alleles by size. We found that both alleles could be identified in 5 out of 11 strains from DC and 2 out of 9 from OPC. Of note, one strain of each type of infection showed only one band corresponding to the size of the so-called

PAGE 108

90 “false” allele in the SC5314 strain (Fig. 6-3A). We found no correlation between the pattern of alleles present in the C. albicans clinical isolates and the type of candidiasis (OPC or DC). These results indicated that the clinical isolates have restriction enzyme polymorphisms downstream of IPF15632. Analysis of IPF15632 mRNA expression in different clinical isolates. To determine if there was a difference in IPF15632 mRNA expression between clinical isolates with different allele patterns, we performed Northern blots on six clinical isolates. From the six C. albicans clinical isolates, three were obtained from patients with DC and the other three from patients with OPC, which, according to the Southern blot shown in Fig. 6-3A, contained either the one band or a two band pattern. We found that all six isolates had IPF15632 transcript (Fig. 6-3B). We expected to see that some strains would have a shorter transcript size due to the indicated truncation in one of SC5314 IPF15632 alleles. But we could not distinguish a change in the transcript size between the strains with the band size corresponding to orf6.257 or orf6.961. However, we did observe a difference in the relative level of IPF15632 mRNA expression between the isolates when compared to EF1 transcript. There was no correlation between the band pattern observed for the clinical isolates and the level of expression of IPF15632 mRNA. IPF15632 orf6.961 allele is not truncated. Since there was no difference in the clinical isolates IPF15632 transcript sizes by Northern blot, we amplified a 1.1 kb fragment of IPF15632 from genomic DNA of each of the six stains previously studied using the polymerase chain reaction (PCR). The PCR primers flanked almost all of the full size orf6.257 allele in SC5314 (Fig. 6-4A). All six isolates with different combinations of band pattern by Southern blot produced an identical PCR fragment that

PAGE 109

91 corresponded to a full size allele in SC5314 (Fig. 6-4B). Furthermore, Southern blots with CAI-12 genomic DNA cut with enzymes further upstream and downstream of IPF15632 and hybridized with UpF1 probe (shown in Fig. 6-4A) also revealed 2 bands on the Southern blot in Fig 6-4C. Each band in the Southern blot corresponded to one allele. A similar pattern was also observed when performing Southern blots with digested genomic DNA from SC5314 and the P162 probe (data not shown). These results suggest that the so-called “false” allele is not truncated in SC5314 nor do the clinical isolates have a shorter IPF15632 allele compared to SC5314 strain. 1920 18 1214131516171011BA 123456798A)EF1 IPF15632 “False” orf6.961BBBBOrf6.257 AAAA B) 1920 18 1214131516171011BA 123456798 1920 1920 18 1214131516171011 18 1214131516171011BA 123456798 123456798A)EF1 IPF15632 “False” orf6.961BBBBOrf6.257 AAAA B)EF1 IPF15632 “False” orf6.961BBBBOrf6.257 AAAA EF1 IPF15632 “False” orf6.961BBBBOrf6.257 AAAA B) Fig. 6-3. IPF15632 alleles in C. albicans clinical isolates. A) Southern blot of 20 clinical isolates. Strains numbered in red were isolated from patients with DC and those numbered in blue were isolated from patients with OPC. The reference strain SC5314 is number 19. B) Northern blot with IPF15632 and EF1 mRNA expression of 6 clinical isolates, 3 obtained from patients with DC and the other 3 from patients with DC. Within the group of the 3 isolates, these differed by containing either the two band pattern (corresponding to alleles A and B) or the one band pattern corresponding to allele A or allele B. Northern blots were normalized with total RNA. IPF15632 was identified using a 32P-labeled F2 probe shown in previous figures. EF1 transcript was identified using a 32P-labeled-specific for EF1 cDNA.

PAGE 110

92 Region of IdentitySequence Divergence 1.1 kb F1 F2 P162 UpF1 Orf6.257 1 kb1.5 kbABABABABBAB)C) 2 kb3 kb4 kb8 kb6 kballele Aallele BRegion of IdentitySequence Divergence 1.1 kb F1 F2 P162 UpF1 Orf6.257 Region of IdentitySequence Divergence 1.1 kb 1.1 kb F1 F2 P162 UpF1 Orf6.257 1 kb1.5 kbABABABABBA 1 kb1.5 kbABABABABBAB)C) 2 kb3 kb4 kb8 kb6 kballele Aallele B 2 kb3 kb4 kb8 kb6 kb 2 kb3 kb4 kb8 kb6 kb2 kb3 kb4 kb8 kb6 kballele Aallele B Fig. 6-4. Identification of a complete sequence for orf6.961 allele. A) Schematic diagram of full-sized orf6.257 of SC5314 strain. The diagram indicates four different probes shown with a blue line. Black arrows represent the location of the primers used for PCR shown in figure 6-4B. B) Agarose gel with PCR fragments of IPF15632 from the clinical isolates containing different band patters. Lane labeled with red are PCR fragments amplified from genomic DNA from DC isolates, while those in blue are from OPC isolates. The first lane shows the 1 kb and 1.5 kb bands for the molecular weight marker, lane 2 is the PCR fragment amplified from SC5314 strain, lanes 3-8 are the PCR fragments amplified from the 6 clinical isolates (the same ones used to detect IPF15632 transcript in Fig. 6-3B). C) Southern blot of CAI-12 genomic DNA cut with HindIII and NsiI and hybridized with 32P-labeled UpF1 probe, its location is indicated in Fig. 6-4A. Analysis of IPF15632 on assembly 19 Currently, with the release of the C. albicans SC5314 polymorphic genome sequence for Assembly 19, IPF15632 alleles are referred to as orf19.3469 (for orf6.257) and orf19.10973 (for orf6.961). The alleles are 99.3% identical and they are located on chromosome 6 within contig10177/20177 of Assembly 19. Analysis of contig10177 provided further information of the genes adjacent to IPF15632 and restriction enzymes that were best for Southern blot analysis. We chose to analyze the recombinants obtained during the disruption process using PvuII and SphI to discriminate which allele was being

PAGE 111

93 disrupted. Due to downstream restriction site polymorphisms with SphI, Southern blot analysis of genomic DNA cut with PvuII and SphI yield a 3.5 kb band for orf19.10973 (allele B) as shown in Fig. 6-3, and a ~2.3 kb band for orf19.3469 (allele A). F1 F2 IPF15632 NsiIPvuII SphI HindIII HindIII UpF1A)B)PvuII-SphI: 3.5 & ~2.2 Kb 5 kb4 kb3 kb2 kborf19.10973orf19.3469 3.5 kb F1 F2 IPF15632 NsiIPvuII SphI HindIII HindIII UpF1A)B)PvuII-SphI: 3.5 & ~2.2 Kb 5 kb4 kb3 kb2 kborf19.10973orf19.3469 B)PvuII-SphI: 3.5 & ~2.2 Kb 5 kb4 kb3 kb2 kborf19.10973orf19.3469 5 kb4 kb3 kb2 kborf19.10973orf19.3469 3.5 kb ALG11 SEC27 IPF15630 ALG11 SEC27 IPF15630 Fig. 6-5. IPF15632 ORF within contig10177. A) Schematic diagram of contig10177 with IPF15632 orf19.10973 (allele B) shown in turquoise. Orf19.3469 (allele A), located on contig 20177 (not shown), is identical to contig10177 for exception of RE site polymorphisms and divergence of short sequences. Fragments F1 and F2 (shown with a blue line under IPF15632 ORF) were cloned into the disruption cassette IPF15632-HUH (see Chapter 4). Genes adjacent to IPF15632 are shown in orange. B) Southern blot showing both alleles of IPF15632. SC5314 DNA cut with PvuII and SphI and probed with 32P-labeled F2 fragment. Creation of IPF15632 Isogenic Null Mutant and Revertant Disruption of Both Alleles of IPF15632 In order to characterize IPF15632 we constructed knockout mutants using the URA-blaster method developed by Fonzi and Irwin (49) (see Fig. 4-2). The IPF15632-hisG-URA3-hisG (HUH) disruption cassette (see Fig. 4-3 and Material and Methods) was used to disrupt both IPF15632 alleles in CAI4 urastrain, replacing approximately 470 bp of each ORF. After 2 rounds of transformations 9 recombinants were obtained and

PAGE 112

94 screened by Southern blot to verify proper integration. From these, 5 recombinants had allele B disrupted first, as noted by a shift from the 3.5 kb band to about an 8 kb band in Southern blots (Fig. 6-6A). Since allele B was preferentially disrupted when using the IPF15632-HUH disruption cassette, we created a different disruption cassette with an upstream (-230 to -19 bp) flanking fragment, UpF1 (see Fig. 6-5A). Nevertheless, allele B was still preferentially disrupted. Preference for recombination at allele B could be due to sequence variation at the F2 fragment used for homologous recombination. Disruption of the second allele of IPF15632 in the uraheterozygote recombinant, 6F7, yielded approximately 50 colonies on SD-ura plates. All the colonies were screened by Southern blot and only 1 Ura+ recombinant (clone 19) had both alleles disrupted appropriately, as observed by a shift for allele A ~2.3 kb band to about a 6 kb band. The disruption of both alleles can be observed in the Southern blot in Fig. 6-6A. Re-insertion of URA3 Back into its Native Locus Finally, to construct isogenic mutants and prevent unwanted phenotypes due to positional effects on URA3 expression (8, 97), each uramutant, 6F7 and 6F7/19F7, had URA3 restored to its native locus by homologous recombination (Fig. 6-6B, schematic diagram in Fig. 5-3). Re-insertion of URA3 is observed by the presence of the 2 kb band, which is absent in a urastrain, like in CAI-4 (see Fig. 5-3). The restriction fragment length polymorphism at the 4.8 kb band in SC5314 and CAI-12 (derived from CAI-4) was created during the construction of the CAI-4 (49), see Fig. 5-3. The genotypes for IPF15632 mutants are indicated in Table 6-1.

PAGE 113

95 1 Copy KO B 2 kb3 kb4 kb5 kb6 kb8 kb1.5 kb10 kbB::HUHB::hisG 66F7 U46F7/196F7/19F7 U16F7/19F7 U4CAI4 PvuII/SphI32P-F2BA)B)+URA3 2 kb3 kb4 kb5 kbSC12CAI12 Copy KOEcoRIM32P-pURA3 1 Copy KO B 2 kb3 kb4 kb5 kb6 kb8 kb1.5 kb10 kbB::HUHB::hisG 66F7 U46F7/196F7/19F7 U16F7/19F7 U4CAI4 PvuII/SphI32P-F2BA)B)+URA3 2 kb3 kb4 kb5 kbSC12CAI12 Copy KOEcoRIM32P-pURA3 2 Copy KO A A::HUHA::hisG A 2 Copy KO A A::HUHA::hisG A Fig. 6-6. Creation of isogenic mutants of IPF15632. A) Southern blot with ura+ recombinants obtained during the disruption process of IPF15632 and re-insertion of URA3. B) Southern blot of mutants, 6F7 and 6F7/19F7, with URA3 (U) re-inserted to its native locus. SC5314 (SC) is the parent strain and CAI-12 is the strain that had both URA3 genes replaced by imm434 DNA and then one copy of URA3 re-inserted back. Table 6-1. IPF15632 recombinants Strain name Genotype Strain studied 6 IPF15632/ipf15632::hisG-URA3-hisG 6F7 IPF15632/ipf15632::hisG 6F7 + U2 IPF15632/ipf15632::hisG/URA3 1 copy KO 6F7/19 ipf15632::hisG-URA3-hisG/ipf15632::hisG 6F7/19F7 ipf15632::hisG/ipf15632::hisG 6F7/19F7 + U1 or U4 ipf15632::hisG/ipf15632::hisG/URA3 2 copy KO Creation of IPF15632 Revertant by Re-inserting One Copy of IPF15632 Re-insertion of one copy of IPF15632 back at its own locus Since random mutations could be created during the disruption process, we reintegrated a wild type copy of IPF15632 back into its own locus to confirm that the observed phenotypes of the null mutant were due to the absence of IPF15632 expression

PAGE 114

96 alone. We chose to re-insert the gene back from within IPF15632 ORF (219 bp) at the F1 fragment to the downstream +181 bp (1150 kb PCR fragment). Re-insertion was carried out by homologous recombination between IPF15632 F1 fragment and the hisG remaining in each allele with those present in the re-insertion cassette. URA3 was used again as the selective marker. A Myc tag (33 aa), cloned in frame at the 3’ end of IPF15632 and followed by a stop codon was included in the re-insertion cassette for future localization studies. (Fig. 6-7A) IPF15632 ura+ revertants were analyzed by Southern blot. Those with one copy integrated at the allele A locus would show a shift from the ~3 kb band in the null mutant to a ~5.8 kb band, while those with integration at allele B locus would show a shift from the ~4.5 kb band in the null mutant to ~6.2 kb band (Fig. 6-7B). A revertant strain (clone 9), with one copy of the Myc-tagged IPF15632 integrated at the disrupted allele A locus, was used along with the IPF15632 null mutant and wild type strain in subsequent studies. Confirmation that revertant is expressing IPF15632 RT-PCR analysis with IPF15632 specific primers and total RNA extracted from the null mutant, revertant, and wild type was carried out to confirm expression of IPF15632 transcript in those strain that have the gene. Visualization of PCR fragments resolved by agarose gel electrophoresis confirmed that the null mutant was not producing intact IPF15632 mRNA, while the revertants and wild type (SC5314) were expressing IPF15632 transcripts (Fig. 6-7C). RT-PCR with EF1 primers verified that all three strains were expressing EF1 transcript, an internal control (data not shown).

PAGE 115

97 F2 UpF1PSF2 F1 URA3 hisG IPF15632Ura+ IPF15632heterozygote revertantura-IPF15632 null homozygote mutant UpF1 P S hisGF2 F1 URA3 hisG F1 F2 IPF15632re-insertion cassetteRe-insertion of 1 copy of IPF15632by homologous recombinationA)B)C) 19SC7M(-) revertants 10 kb8 kb6 kb5 kb4 kb3 kb2 kb19 revertants F2 UpF1PSF2 F1 URA3 hisG IPF15632Ura+ IPF15632heterozygote revertant F2 UpF1PSF2 F1 URA3 hisG IPF15632 F2 UpF1PSF2 F1 URA3 hisG IPF15632Ura+ IPF15632heterozygote revertantura-IPF15632 null homozygote mutant UpF1 P S hisGF2 F1 URA3 hisG F1 F2 IPF15632re-insertion cassetteRe-insertion of 1 copy of IPF15632by homologous recombinationura-IPF15632 null homozygote mutant UpF1 P S hisGF2 F1 URA3 hisG F1 F2 UpF1 P S hisGF2 F1 UpF1 P S hisGF2 F1 UpF1 P S hisGF2 F1 URA3 hisG F1 F2 URA3 hisG F1 F2 IPF15632re-insertion cassetteRe-insertion of 1 copy of IPF15632by homologous recombinationA)B)C) 19SC7M(-) revertants 19SC7M(-) revertants 10 kb8 kb6 kb5 kb4 kb3 kb2 kb19 revertants 10 kb8 kb6 kb5 kb4 kb3 kb2 kb 10 kb8 kb6 kb5 kb4 kb3 kb2 kb19 revertants IPF15630 IPF15630 IPF15630 IPF15630 Fig. 6-7. Creation of IPF15632 revertant. A) Schematic diagram illustrating re-insertion of one copy of IPF15632 back into its own locus in the urahomozygous IPF15632 null mutant. IPF15632 is shown with a turquoise arrow. F1 fragment used for homologous recombination is marked with a blue line. Myc tag is represented with a pink line after IPF15632 ORF. Downstream sequence for IPF15632, included in the re-insertion cassette, is shown with a gray square with hatched lines. Location of PvuII (P) and SphI (S) RE sites are with reference to contig10177 (allele A). B) Southern blot analysis of revertants. Genomic DNA was cut with PvuII and SphI. Lane 1: uranull mutant (6F7/19F7), Lane 2: 1 kb marker, lane 3: revertant #1, lane 4: revertant # 9. Revertants #1 and #9 have one copy of IPF15632 integrated at allele B and A, respectively. C) Agarose gel with IPF15632 RT-PCR fragment. Lane 1: 1 kb marker, lane 2: no DNA (neg. control), lane 3: SC5314, lane 4: null mutant, lanes 5-7: revertants.

PAGE 116

98 Phenotypic Analysis of Diploid IPF15632 Null Mutant Growth Rate and Hyphae Formation In order to determine the possible function of IPF15632, we analyzed the phenotypes of the IPF15632 isogenic null mutant (6F7/19F7+U4), the revertant (clone 9), and wild type (CAI-12) strain under different conditions. To test whether the IPF15632 null mutant had altered growth rate in rich media, strains were grown in YPD media at 30 C and 37 C. There was no obvious difference between all three strains growth rate at 30 C in YPD media. However, at 37 C the knockout mutant reached stationary phase at a slightly lower density when compared to the wild type or the revertant strain. (Fig. 6-8) These results were consistent when repeated three times. -100.510100200300400500600700 -0.5 Time (min)Log OD60030 C Revertant9Null mutantCAI-12 -1-0.500.510100200300400500600700 Revertant9Null mutantCAI-12 Time (min)37 C -100.510100200300400500600700 -0.5 Time (min)Log OD60030 C Revertant9Null mutantCAI-12 -100.510100200300400500600700 -0.5 Time (min)Log OD60030 C -100.510100200300400500600700 -0.5 -100.510100200300400500600700 -0.5 -100.510100200300400500600700 -0.5 -100.510100200300400500600700 -0.5 -100.5101002003004005006007000100200300400500600700 -0.5 Time (min)Log OD60030 C Revertant9Null mutantCAI-12 Revertant9Null mutantCAI-12 -1-0.500.510100200300400500600700 Revertant9Null mutantCAI-12 Time (min)37 C -1-0.500.510100200300400500600700 -1-0.500.510100200300400500600700 Revertant9Null mutantCAI-12 Revertant9Null mutantCAI-12 Time (min)37 C Fig. 6-8. Growth curve for IPF15632 null mutant. Strains: IPF15632 null mutant (red), revertant (blue), and wild type (yellow) were grown in YPD starting at an OD600 of 0.1 and followed until they reached stationary phase. Growth rate was done at 30 and 37 C. C. albicans mutants with defects in hyphae or pseudohyphae formation are usually found to be avirulent in murine models. Typically, these mutants can be screened for defects in hyphae formation when grown in media known to induce filamentation, for example media containing serum (12). Thus, we screened our KO mutants for hyphae formation in YPD liquid media containing 5% fetal bovine serum (FBS). We did not see

PAGE 117

99 any obvious difference in hyphae formation, rate or length, between the KO mutants and the wild type (data not shown). On the other hand, on different solid hyphae-inducing media, such as YP containing glycerol, RPMI, and M199, the filament length in the mutant colonies were slightly shorter than that of the wild type. Colonies’ filaments on solid media are characterized by a fuzzy orb around the colony. It should also be noted that the mutants wrinkled colony morphology differed from the organized flat colony morphology of the wild type on M199 media. (Fig. 6-9) The fact that we could see a difference in filamentation on solid media and not on liquid media is probably due to a weaker induction of filaments by solid media, which makes slight defects more visible on solid surfaces (43). CAI-12Null + U4Null + U1 CAI-12Null + U4Null + U1 Fig. 6-9. IPF15632 hyphae length in M199. Approximately 105 CFU in 5 L for each strain (wild type CAI-12, IPF15632 null mutant + URA3 clone 1 or IPF15632 null mutant + URA3 clone 4) were spotted on M199 solid media. Plates were allowed to grow at 37 C for 8 days after which plates were scanned. Single Colony Growth Rate on Solid Media Growth under embedded agar conditions Single colonies of IPF15632 mutants were tested for invasive growth when embedded in agar conditions. Single colony forming units were mixed with warm but not yet solidified media containing agar and incubated at 37 C. Null mutants showed a

PAGE 118

100 reduced colony size under embedded conditions in all the media tested, YPD, YPD containing 5% FBS, and YP containing 2% glycerol (YPGly) at 37 C. This difference was more obvious in YPGly, since colonies for all the strains grew in a radial form with long filaments extending outwards. In YPGly, the colony size for the null mutants was smaller than the wild type (see Fig. 6-10) and the revertant (not shown), as observed by the shorter filament length for the null mutant compared to the wild type. Representatives of the null mutant and the wild type (5 each) are shown in Fig. 6-10. Mutants did not appear to have an invasive filamentation defect when grown in presence of serum (data not shown); thus, the difference in colony diameter could be a reflection of slower growth rate rather than a defect on the cells’ ability to form filaments. However, after 4 days of incubation at 37 C it was harder to distinguish a difference in colony size between the null mutants and the wild type strain. Null mutant + U4CAI-12 (wt) Null mutant + U4CAI-12 (wt) Fig. 6-10. Colony growth rate for IPF15632 null mutant in embedded agar conditions. Approximately 50 CFU were mixed with warm YPGly media containing agar. Once solidified, plates were incubated at 37 C for 2 days. Pictures of various single colonies for each strain were taken with an inverted light microscope BRAND using a 2.5X BRAND lens.

PAGE 119

101 Growth on surface of solid media Since there appeared to be a slight difference in growth rate between the null mutants and the wild type or revertant strains when grown under embedded conditions at 37 C, we examined their growth rate of single colonies on YPD agar plates. Cells grown to log phase on YPD at 37 C were diluted, spread on YPD plates, and incubated at 30 and 37 C overnight. The null mutants colonies were consistently smaller than those of the wild type or the revertant at both temperatures, but the difference was more evident at 37 C (Fig. 6-11). For example, when we looked at the size of 5 individual colonies in a 25.6 cm2 inset (Fig. 6-11B) from the plates in Fig. 6-11A, we found that the null mutants colonies were on average approximately 25% smaller (~1.66 mm diameter) than the average colony size of the wild type and the revertant (~2.26 mm diameter) strains. The difference in size remained clear up to 3 days, after which it was harder to discern the difference by eye. Furthermore, the difference in colony size was observed for all three strain when grown to stationary phase at either 30 or 37 C. Strains were diluted to about 100 CFU in saline and spread onto YPD plates. The difference in colony size after overnight incubation at 37 C was even more pronounced (Fig. 6-12). IPF15632 Null Mutant has No Cell Wall Defect Since, IPF15632 is annotated in the CandidaDB as weakly similar to S. cerevisiae FLO11 gene, which encodes a GPI-anchored cell surface glycoprotein involved in cell-cell adherence (61), I decided to determine if disruption of IPF15632 would affect the integrity of the C. albicans cell wall. Wild type and mutants were compared for sensitivity to different cell wall effectors: calcofluor white and sodium dodecyl sulfate (SDS), both of which are used for identifying general cell wall defects. The null mutant

PAGE 120

102 Fig. 6-11. Single colony growth rate on YPD agar for IPF15632 null mutant. A) Cells grown for 3.5 hr to early log phase at 37 C were diluted and spread on YPD plates (100 mm2), which were incubated at 37 C overnight. An inset of 16 mm2 was drawn for each plate. B) Magnified inset for each plate on the panel above. A 6 mm yellow line was drawn along the diameter of one wt colony. This line was used as a reference to calculate the size of the rest of the colonies.

PAGE 121

103 CAI-12 (wt)Null + U4 CAI-12 (wt)Null + U4 Fig. 6-12. Single colony growth on YPD agar for strains grown to stationary phase on YPD media at 37 C. Stationary phase cells kept in saline for approximately 5 hours, diluted to about 70 CFU, and spread on YPD plates. Plates were incubated at 37 C overnight. was slightly more susceptible than wild type to calcofluor white at a density of 103 CFU but it was not more susceptible to SDS. (See Fig. 6-13). I also examined the mutants for defects in glycosylation of cell wall proteins when grown in presence of the aminoglycosides Hygromycin B (200 ug/mL) or G418

PAGE 122

104 (1.2 mg/mL). Mutants were not susceptible like the wild type to these aminoglycosides at either 30 or 37 C (data not shown). Calcofluor White Null +U1Null +U4CAI-120.06% SDS 105104103102 CFU 105104103102 CFUCalcofluor White Null +U1Null +U4CAI-120.06% SDS 105104103102 CFU 105104103102 CFU105104103102 CFU 105104103102 CFU 105104103102 CFU105104103102 CFU Fig. 6-13. Phenotypic analysis of IPF15632 null mutants with cell wall effectors. 5 L containing the indicated cell density on the figures, were spotted for each strain on YPD plates containing Calcofluor white 50 g/mL (left panel) and SDS 0.06% (right panel). Evaluation of IPF15632 Role in Pathogenesis Adherence to Primary Human Buccal Epithelial Cells I measured the adherence of the mutants to primary human buccal epithelial cells (BECs) as a cell model for OPC. We found that the mutants adhered as well as the wild type to BECs. A representative field of view with several epithelial cells and yeast bound to the cells is shown in Fig. 6-14. These results suggest that IPF15632 is not likely to have a role in maintaining cell wall integrity since the null mutants were not susceptible to diverse cell wall effectors nor did they have a defect in adherence to buccal epithelial cells. However, these results do not exclude that IPF15632 might have a role in adherence to other types host cells (e.g. colon epithelial cells or endothelial cells) or adherence between C. albicans cells.

PAGE 123

105 CAI-12 (wt)IPF15632null mutant CAI-12 (wt)IPF15632null mutant CAI-12 (wt)IPF15632null mutant Fig. 6-14. IPF15632 mutant adherence to primary human buccal epithelial cells. This figure shows some representative BECs with yeast bound to them. About 0.5 x 105 CFU of BECs were incubated with 0.5 x 106 CFU of either the wild type strain (left panel) or IPF15632 null mutant (right panel) in suspension for 2 hours at 37 C. Non-adherent yeast were removed by filtering them out and remaining yeast cells bound to BECs were gram+ stained. The nucleus of BECs is indicated with blue arrows and yeast cells are pointed with red arrows. The number of yeast cells bound to 100 BECs were counted using a light microscope with a 40X magnification lens. IPF15632 Role in a Murine Model of DC To evaluate the role of IPF15632 during pathogenesis, I used the murine model of disseminated candidiasis. Outbred ICR mice were inoculated intravenously with 2.5 x 105 CFU of the null mutant or wild type strain and monitored their survival for 43 days. As shown in Fig. 6-15, all the mice injected with the wild type CAI-12 strain died between day 3 and 9. Mice infected with the null mutant lived longer on average than those infected with the wild type (P < 0.01). From the eight mice infected for each group, 2 mice inoculated with the null mutant did not die during the length of this experiment. Therefore, the virulence of IPF15632 null mutants is attenuated in the murine model of disseminated candidiasis.

PAGE 124

106 020406080100024681012141618Time (Days)% Survival 20 mai + U4CAI-12 (wt) 020406080100024681012141618 20 020406080100024681012141618Time (Days)% Survival 20 mai + U4CAI-12 (wt) mai + U4CAI-12 (wt) Fig. 6-15. DC murine model for IPF15632. Mice (8 per group) were injected i.v. with 2.5 x 105 CFU in saline with either the null mutant (red) or the wild type CAI-12 strain (yellow). Survival of mice was followed for 43 days. Disseminated Candidiasis Tissue Burden Study Tissue fungal burden studies were performed to assess the severity of the infection caused by IPF15632 null mutants compared to the wild type strain. I analyzed the fungal burdens in kidneys, livers, and spleens of ICR mice (18 per group) inoculated i.v. via the lateral tail vein with 2.5 x 105 CFU of IPF15632 null mutants or CAI-12 strain. Mice (9 per group) were sacrificed on Day 1 (22 hours) and Day 4 (90 hours) post-infection. Liver, kidneys, and spleen were dissected from mice and homogenized with saline solution containing antibiotics to prevent bacterial growth. Dilutions for each homogenization were spread on SD plates containing antibiotics. The number of colony forming units (CFU) per gram of tissue was then calculated. The average log CFU/g for

PAGE 125

107 each organ and day are shown in Table 6-2 and the graphical representation can be seen in Fig. 6-16. We found no significant difference on day 1 between the null mutant and the wild type strain fungal burden in the kidneys (3.96 0.61 and 3.76 1.54 respectively, P = NS), liver (3.34 0.2 and 3.04 0.94 respectively, P = NS), and spleen (3.53 0.54 and 3.05 1.25 respectively, P = NS). One mouse inoculated with the wild type did not receive the full dose; this mouse was sacrificed on day 1. As shown in Fig. 6-16, the mouse that did not receive a full dose had fewer log CFU/g in all three organs analyzed. These data are indicated with a yellow square around the blue dot. Furthermore, when the data from this mouse were omitted, the difference between the two groups on Day 1 was still insignificant. These results imply that the null mutant was able to colonize the kidney, liver, and spleen as well as the wild type. Prior to sacrificing the mice on the fourth day post infection one mouse infected with the wild type died at day 3, thus leaving 8 mice instead of 9 to be analyzed in that group on day 4. Compared to day 1, the fungal burden in the kidneys on day 4 showed an increase for both the null mutant and wild type (average increase of 0.88 and 1.66 log CFU/g respectively). There were significantly fewer cells on day 4 in the kidneys of mice infected with null mutant (4.84 0.51) compared to wild type (5.42 0.31) with a P = 0.05. These results suggest that the null mutant did not proliferate as well as the wild type within mouse kidneys. The log CFU/g on day 4 for one mouse kidney infected with the wild type was not included in the statistical analysis due to the fact that there was apparently an error during the experimental procedure. On the other hand, the fungal burden in the liver from Day 1 to Day 4 decreased for both the wild type and the null mutant (average ~1 and 1.8 log CFU/g respectively).

PAGE 126

108 Even though fungal burden in the liver of mice infected with the null mutant (1.51 0.92 log CFU/g) compared to the wild type (2.00 0.57 log CFU/g) was not significantly different (P > 0.2), there was a trend of fewer null mutant cells with prolonged infection (Day 4). Furthermore, 8 mice out of 9 infected with the null mutant were able to clear the infection from the spleen. The remaining infected spleen still had fewer null mutant cells (1.85 log CFU/g) than those infected with the wild type strain (2.75 0.39 log CFU/g). Clearance of the null mutant from the spleen compared to the wild type on day 4 was significant (P < 0.001). These results suggest that the IPF15632 null mutant is cleared faster than the wild type from the spleen. In the kidney the null mutant is not able to proliferate as much as the wild type this is likely due to recognition and clearance of the mutant cells by the mouse immune system. The fact that tissue burdens were not different at day 1 argues against differences in growth rates in vivo. (See Fig. 6-16) Table 6-2. IPF15632 DC fungal burden Day 1 (log CFU/g of tissue) Day 4 (log CFU/g of tissue) Strains Kidney Liver Spleen Kidney Liver Spleen CAI-12 (wt) 3.76 1.54 3.04 0.94 3.05 1.25 5.42 0.31 2.00 0.57 2.75 0.39 Null mutant (2 Copy KO) 3.96 0.61 3.34 0.2 3.53 0.54 4.84 0.51 1.51 0.92 1.85a a Only 1 mouse out of eight had CFU in the spleen; the rest had cleared all of IPF15632 null mutant cells. Therefore, this number is not the average taken for the 9 mice. Log CFU/g values are followed by the calculated standard deviation. Antigenicity of IPF15632p Since, we identified IPF15632 as a protein reactive with sera from patients with OPC, we determined if patients with DC and healthy individuals had antibodies to IPF15632. We used a bacterial expression vector with a fragment of IPF15632 690 bp cloned in frame. IPF15632 protein fragment was expressed by bacteria upon induction

PAGE 127

109 0123456 Null mutantCAI-12 (wt)Kidney Day 4P = 0.05 01234 Null mutantCAI-12 (wt)Spleen Day 1P = NS 01234Null mutantCAI-12 (wt)Spleen Day 4P < 0.001 Liver Day 1Null mutantCAI-12 (wt) 01234 P = NS 01234 Null mutantCAI-12 (wt)Liver Day 4P > 0.2 0123456Null mutantCAI-12 (wt)Kidney Day 1P = NS log CFU/glog CFU/glog CFU/g 0123456 Null mutantCAI-12 (wt)Kidney Day 4P = 0.05 01234 Null mutantCAI-12 (wt)Spleen Day 1P = NS 01234Null mutantCAI-12 (wt)Spleen Day 4P < 0.001 Liver Day 1Null mutantCAI-12 (wt) 01234 P = NS 01234 Null mutantCAI-12 (wt)Liver Day 4P > 0.2 0123456Null mutantCAI-12 (wt)Kidney Day 1P = NS 0123456 Null mutantCAI-12 (wt)Kidney Day 4P = 0.05 01234560123456 Null mutantCAI-12 (wt)Kidney Day 4P = 0.05 01234 Null mutantCAI-12 (wt)Spleen Day 1P = NS 0123401234 Null mutantCAI-12 (wt)Spleen Day 1P = NS 01234Null mutantCAI-12 (wt)Spleen Day 4P < 0.001 0123401234Null mutantCAI-12 (wt)Spleen Day 4P < 0.001 Liver Day 1Null mutantCAI-12 (wt) 01234 P = NS Liver Day 1Null mutantCAI-12 (wt) 0123401234 P = NS 01234 Null mutantCAI-12 (wt)Liver Day 4P > 0.2 0123401234 Null mutantCAI-12 (wt)Liver Day 4P > 0.2 0123456Null mutantCAI-12 (wt)Kidney Day 1P = NS 0123456Null mutantCAI-12 (wt)Kidney Day 1P = NS log CFU/glog CFU/glog CFU/g Fig. 6-16. Tissue burden study for DC murine model of IPF15632 null mutant. The log CFU/g is shown on the Y-axis for Day 1 graphs (left) and Day 4 graphs (right). The yellow square for the wt at day 1 in all three graphs on the left represents the mouse that did not receive the full dose during the injection. The data point for one kidney at day 4, shown with a cross, was not included in the statistical analysis due to an apparent error during the experimental procedure. Organs with no CFU were given a numerical value of 1 to calculate the log CFU/g and to perform statistical analysis. The horizontal black line represents the median.

PAGE 128

110 with IPTG. By Western blot analysis we screened the reactivity of the expressed IPF15632p fragment with pooled sera from patients with OPC and DC (2). Both groups had antibodies to IPF15632 protein fragment at a dilution of 1/5000 in PBS. When we screened the Western blot with individual sera from three healthy individuals, we found that only 1 out of 3 had a slight reactivity to IPF15632p fragment with a dilution of sera of 1/5000 inn PBS. Thus, it appears that patients with OPC and DC, as well as, healthy individuals might have antibodies that recognize IPF15632p. Creation of a Strain Triploid for IPF15632 During the process of disrupting IPF15632 we unexpectedly created a strain that was triploid for this gene, presumably due to a gene duplication. This duplication probably occurred during the selection for ura+ transformants when the hisG-URA3-hisG (HUH) cassette was inserted at allele B of IPF15632. Gene duplication or chromosome duplication leading to trisomy has been reported to occur during disruptions of several other genes (26, 59), suggesting that this is not a rare event in C. albicans strains. Therefore, we had a strain with the IPF15632 allele B disrupted by the HUH cassette and two copies of allele A, one of which was created by the duplication (Fig 6-17A). We disrupted both copies of allele A sequentially. Thus, we constructed 1, 2, and 3 copy knockout (KO) mutants sequentially in the triploid strain. The 1 copy KO (1 KO) had allele B disrupted. The 2 copy KO (2 KO) had allele B plus one copy of allele A disrupted. The 3 copy KO (3 KO) is a null mutant for IPF15632. (Fig 6-17B) After disruption of IPF15632 in the triploid strain we restored URA3 to its native locus on chromosome 3 for all three uramutants (Fig. 6-17C). Re-insertion of URA3 back into its original locus would prevent unwanted phenotypes due to positional effects on URA3 expression. Northern blot analysis confirmed that no intact IPF15632 mRNA

PAGE 129

111 was present in total RNA extracted from the null 3 KO mutant as compared to the 1 and 2 KO (Fig. 6-17D). I verified the copy number of each allele by measuring the intensity of each of the two bands observed in Southern blots. The ratio of allele A to allele B remained 2:1 during the disruption of all three copies of IPF15632. Furthermore, the duplication that occurred in one arm of chromosome 6 was later confirmed by the presence of only one copy of the allele B in Southern blots. Separation of chromosomes by contour-clamped homogeneous electric field (CHEF) further indicated that the duplication occurred within chromosome 6, where IPF15632 is located. This is indicated by the presence of one band for all three mutants on agarose gels with C. albicans 8 chromosomes resolved (Fig. 6-17E) by pulse field electrophoresis. However, we do not know how many other alleles syntenic to IPF15632 allele A on chromosome 6 were duplicated. Phenotypic Analysis of triploid 1 KO IPF15632 Duplication on the Triploid 1 KO Mutants Caused Cell Wall Changes In spite of the duplication we found no difference between wild type CAI-12 strain and any of the knockout mutants in either growth rate (30 and 37 C) in YPD liquid media or hyphae formation in liquid or solid media. (Data not shown) However, I noticed that the 1 and 2 KO mutants had a smooth colony morphology while the 3 KO mutant and wild type had a wrinkled colony phenotype on YPGlycerol (Fig. 6-18A) and YPD plates grown at 37 C. In S. cerevisiae, activation of FLO genes, including FLO11, lead to flocculation, which is recognized by a wrinkled or ribbon-like colony morphology on agar plates (63). Thus, the 1 and 2 KO mutant smooth colony could be due to a defect in flocculation.

PAGE 130

112 R & 1456237 PPPP331122PE) 8 kb6 kb5 kb4 kb3 kb2 kb1.5 kb10 kbB::HUHB::hisG A wt1 KOMutants BB)A) Allele AF1F2 Allele AContig6-1640 with allele A F1F2 Contig6-1763 with allele B Allele B 1in.IPF15632mRNAEF1 mRNAC)D) + URA3 SC 1 2 3 CA Copy KO R & 1456237 PPPP331122PE) R & 1456237 PPPP331122P R & 1456237 R & 1456237 PPPP331122PPPPP331122PE) 8 kb6 kb5 kb4 kb3 kb2 kb1.5 kb10 kbB::HUHB::hisG A wt1 KOMutants B 8 kb6 kb5 kb4 kb3 kb2 kb1.5 kb10 kb 8 kb6 kb5 kb4 kb3 kb2 kb1.5 kb10 kbB::HUHB::hisG A wt1 KOMutants BB)A) Allele AF1F2 Allele AContig6-1640 with allele A F1F2 Contig6-1763 with allele B Allele B Allele AF1F2 Allele AContig6-1640 with allele A F1F2 Contig6-1763 with allele B Allele B F1F2 Contig6-1763 with allele B Allele B 1in.IPF15632mRNAEF1 mRNAC)D) + URA3 SC 1 2 3 CA Copy KO 1in.IPF15632mRNAEF1 mRNA 1in.IPF15632mRNAEF1 mRNAC)D) + URA3 SC 1 2 3 CA Copy KO + URA3 SC 1 2 3 CA Copy KO P S? PS P S? PS P S? S? PS PS A::HUHA::hisG 2 KO3 KO 23Cl A::HUHA::hisG 2 KO3 KO A::HUHA::hisG 2 KO3 KO 23Cl 23Cl 23Cl Fig. 6-17. Triploid IPF15632 strain. A) Schematic diagram picturing 2 copies of allele A and 1 copy of allele B. RE site polymorphism is shown for SphI (S). F1 and F2 are the fragments used for homologous recombination to insert the hisG-URA3-hisG cassette (green-red-green box) within allele B. B) Southern blot with all three mutants, each with one copy of IPF15632 disrupted sequentially in the triploid strain. C) Re-insertion of URA3 at its original locus for all 3 uramutants. D) Northern blot with IPF15632 transcript (upper) and EF1 transcript (bottom). Total RNA was extracted from the triploid 1, 2, and 3 KO mutants and from a clinical (Clin.) isolate. Northern blots were hybridized with gene specific -32P labeled probes. E) CHEF. Agarose gel stained with ethidium bromide to visualize C. albicans 8 chromosomes (left) and Southern blot of gel using -32P-F2 probe (right). Chromosomes were isolated for the CAI-12 and SC5314 parent (P) strains and from the triploid 1, 2, and 3 KO mutants.

PAGE 131

113 Since flocculation is cell-cell adherence, which in turn is a reflection of cell surface properties, we decided to determine if the triploid IPF15632 mutants’ cell wall had been compromised by the duplication. I compared sensitivity of the wild type and mutants to different cell wall effectors: calcofluor white (isofluor that binds chitin) and ionic detergent (SDS), and Zymolyase-100T. Both calcofluor white and SDS are used for identifying general cell wall defects. Zymolyase hydrolyses the -1,3 glucan bond and is used to screen for mutants with changes in the -1,3 glucan layer or in the external mannoprotein layer. I found that 1 and 2 KO were susceptible to the calcofluor white (50 mg/mL) but the 3 KO exhibited the same level of resistance as the wild type strain (Fig. 6-18B). Conversely, the 1 KO was slightly more resistant to SDS than the wild type or the 3 KO at 37 C (Fig, 6-18C). Similar to the results observed with SDS, the 1 KO was less sensitive to cell lysis by Zymolyase (0.1 mg/mL) compared to the wild type and 3 KO at 37 C (Fig. 6-18D). Furthermore, triploid 1 KO had defects in glycosylation of cell wall proteins, as observed by sensitivity to growth in presence of Hygromycin B (200 ug/mL) or G418 (1.2 mg/mL). Again the triploid 3 KO was as resistant as the wild type to aminoglycosides (Fig. 6-19). These results indicate that our triploid 1 KO mutant had a weakening of the cell wall. The phenotypic similarity between the 3 KO mutant and the wild type could be due to either random mutations created during the disruption of the third and last copy of IPF15632 or to the induction of a cell wall salvage mechanism to compensate for the changes in gene expression that are affecting cell wall integrity.

PAGE 132

114 wt1 KO3 KOColony morphology wt1 KO3 KOSDS (0.06%) wt1 KO2 KO3 KO105104103102 CFUCalcofluor whiteA600 (nm)Time (min) 00.10.20.30.40.5020406080100120140160180Zymolase: -Glucan Layerwt1 KO 3 KO B)C)A)D) wt1 KO3 KOColony morphology wt1 KO3 KO wt1 KO3 KOColony morphology wt1 KO3 KOSDS (0.06%) wt1 KO3 KOSDS (0.06%) wt1 KO2 KO3 KO105104103102 CFUCalcofluor white wt1 KO2 KO3 KO105104103102 CFUCalcofluor whiteA600 (nm)Time (min) 00.10.20.30.40.5020406080100120140160180Zymolase: -Glucan LayerA600 (nm)Time (min) 00.10.20.30.40.5020406080100120140160180A600 (nm)Time (min) 00.10.20.30.40.500.10.20.30.40.5020406080100120140160180Zymolase: -Glucan Layerwt1 KO 3 KO wt1 KO 3 KO wt1 KO 3 KO B)C)A)D) Fig. 6-18. Changes in the triploid mutants cell wall. A) Flocculation assay on YPGly agar. For each strain 105 CFU were spotted onto YPGly plates and incubated at 37 C for 1 week. B) Cell wall effector: calcofluor white (50 g/mL) in YPD plates. The cell density spotted on plates is indicated in the figure. Cells were allowed to grow at 37 C for 3 days. C) Cell wall effector: SDS (0.06%). Strains were spotted as indicated for calcofluor white. D) Cell wall effector: Zymolyase. Changes at the -glucan or mannoprotein layer are observed by the rate of cell lysis at 35 C in presence of 0.1 mg/mL of Zymolyase-100T. Absorbance was read at 600 nm every 10 minutes for a period of 3 hours.

PAGE 133

115 1 KO3 KO 2 KOwtwt1 KO3 KO 2 KOwtwt Fig. 6-19. Susceptibility to G418 by the triploid 1 KO mutant. Approximately 3000 CFU were streaked into YPD plate (control not shown) and YPD plates containing hygromycin B (200 ug/mL) or G418 (1.2 mg/mL). YPD + hygromycin B plate on the left panel includes four URA3 re-insertion clones in the triploid 1 KO mutant and one wild type. The YPD + hygromycin B plate on the right panel includes three URA3 re-insertion clones in the triploid 3 KO mutant. Plates were incubated at 37 C for 3 days. In Vitro Cell Adherence Assay Since the triploid 1 KO mutant appeared to have cell wall defects, I measured the mutants’ adherence to two different oral epithelial cells. Compared the wild type CAI-12 (100%), the triploid 1 KO mutant adhered less to primary human buccal epithelial cells (60%, P = 0.012), while the triploid 3 KO mutant did not (90%, P = NS) (Fig. 6-20A). In contrast, both 1 and 3 KO mutants adhered less to a pharyngeal epithelial cell line (FaDu) (23%, P < 0.001 and 56%, P < 0.01, respectively) when compared to wild type (100%) (Fig. 6-20B). To determine if the mutants also had impaired adherence to other mucosal epithelial cells, the same experiment was repeated using a colon epithelial cell line (HT-29). Again 1 and 3 KO adhered less to colon epithelial cells (43.6%, P = 0.0004 and

PAGE 134

116 46.5%, P = 0.006 respectively) when compared to wild type (100%) (Fig. 6-20C). These results suggest that the duplication could have led to changes in gene expression that reduced adherence to epithelial cells. 020406080100120% of yeast adhering to FaDuwt1 KO2 KO3 KOP = 0.001P = 0.01P = 0.001Pharyngeal Epithelial Cell lineB) 020406080100120% of yeast adhering to HT-29wt1 KO3 KO2 KOP = 0.0004P <0.0001P = 0.006Colonic Epithelial Cell lineC)% of yeast adhering to BECs 020406080100120wt1 KO2 KO3 KOP = 0.012Buccal Epithelial CellsA) 020406080100120% of yeast adhering to FaDuwt1 KO2 KO3 KOP = 0.001P = 0.01P = 0.001Pharyngeal Epithelial Cell lineB) 020406080100120% of yeast adhering to FaDuwt1 KO2 KO3 KOP = 0.001P = 0.01P = 0.001Pharyngeal Epithelial Cell line 020406080100120% of yeast adhering to FaDuwt1 KO2 KO3 KOP = 0.001P = 0.01P = 0.001Pharyngeal Epithelial Cell lineB) 020406080100120% of yeast adhering to HT-29wt1 KO3 KO2 KOP = 0.0004P <0.0001P = 0.006Colonic Epithelial Cell lineC) 020406080100120% of yeast adhering to HT-29wt1 KO3 KO2 KOP = 0.0004P <0.0001P = 0.006Colonic Epithelial Cell line 020406080100120% of yeast adhering to HT-29wt1 KO3 KO2 KOP = 0.0004P <0.0001P = 0.006Colonic Epithelial Cell lineC)% of yeast adhering to BECs 020406080100120wt1 KO2 KO3 KOP = 0.012Buccal Epithelial CellsA)% of yeast adhering to BECs 020406080100120wt1 KO2 KO3 KOP = 0.012Buccal Epithelial CellsA) Fig. 6-20. Triploid strain adherence defect to human epithelial and endothelial cells. A) Adherence assay to BECS. 106 CFU were incubated with 105 CFU of primary BECs. Individual single cells adhering to epithelial cells were counted after filtering out non-adherence yeast cells and gram+ staining. B). Adherence to pharyngeal epithelial cells. C) Adherence to colonic epithelial cells. Evaluation of Triploid 1 KO Strain Role in Pathogenesis To evaluate the role of the triploid strains during pathogenesis I used the standard murine model of disseminated candidiasis. ICR mice were inoculated intravenously with 106 CFU and no difference was detected in mortality rate between 1 or 3 KO relative to the wild type CAI-12 (P = NS) (Fig. 6-21A). Since PMNs are the first line of defense for

PAGE 135

117 disseminated candidiasis, I determined if the triploid mutant strains were susceptible to phagocytosis and killing by PMNs. There was no difference in either the level of phagocytosis or killing for the 1 or 3 KO compared to the wild type (Fig. 6-21B). Given that the 1 KO triploid strain had impaired adherence to epithelial cells, I studied its role in OPC pathogenesis. I used a murine model of OPC, in which 12 cortisone-treated mice per group were challenged orally with 108 CFU. There was a significant decrease in tissue burden (tongue, esophagus, and jaw muscles) for both triploid 1 and 3 KO mutants (3.01 and 4.18 log CFU/g respectively, p<0.0001) compared to the wild type (5.76 log CFU/g) (Fig. 6-21C). These results suggest that gene expression in this triploid mutant strain attenuated virulence in the OPC but not in the DC murine model. Stability of Duplication Recently, it has been shown that a trisomic strain of C. albicans had undergone changes in the copy number of chromosome 1 after 3 days of growth in kidneys in a haematogenously disseminated murine model. Also, trisomy at chromosome 2 has been found not to be stable. (26, 157, 157) Thus, I analyzed the stability of the duplication in the triploid 1 KO mutant by passage under stressful conditions. I chose growth in the presence of Hygromycin B (100 /mL) and growth in murine kidneys as stressful conditions while growth on rich YPD media as a non-stressful control. Since the duplication of our triploid 1 KO mutant could be recognized in Southern blots by the ratio of IPF15632 allele A to allele B (2:1), I measured the stability of the duplication in reference to the copy number of allele A to allele B. For each experiment, 20 single

PAGE 136

118 colonies were analyzed by Southern blot. After passage 1 (~20 generations) in YPD all 20 colonies obtained from the triploid 1 KO mutant maintained the duplication (band % of Cells Phagocytosed by PMNs 020406080100CAI-12Triploid 1 KOTriploid 3 KO 020406080100% of Cells Killed by PMNsCAI-12Triploid 1 KOTriploid 3 KOB)DC Murine ModelDays % survival 02040608010012013579111315171921 1 KO 3 KOwtA)log CFU/g0123456 wt1 KO3 KOP <0.0001P <0.0001OPC Murine ModelC)% of Cells Phagocytosed by PMNs 020406080100CAI-12Triploid 1 KOTriploid 3 KO 020406080100% of Cells Killed by PMNsCAI-12Triploid 1 KOTriploid 3 KOB)% of Cells Phagocytosed by PMNs 020406080100CAI-12Triploid 1 KOTriploid 3 KO% of Cells Phagocytosed by PMNs 020406080100CAI-12Triploid 1 KOTriploid 3 KO 020406080100 020406080100CAI-12Triploid 1 KOTriploid 3 KO 020406080100% of Cells Killed by PMNsCAI-12Triploid 1 KOTriploid 3 KO 020406080100 020406080100020406080100% of Cells Killed by PMNsCAI-12Triploid 1 KOTriploid 3 KOCAI-12Triploid 1 KOTriploid 3 KOB)DC Murine ModelDays % survival 02040608010012013579111315171921 1 KO 3 KOwtA)DC Murine ModelDays % survival 02040608010012013579111315171921 1 KO 3 KOwtDays % survival 02040608010012013579111315171921 1 KO 3 KOwt 1 KO 3 KOwtA)log CFU/g0123456 wt1 KO3 KOP <0.0001P <0.0001OPC Murine ModelC)log CFU/g0123456 wt1 KO3 KOP <0.0001P <0.00010123456 wt1 KO3 KOP <0.0001P <0.0001OPC Murine ModelC) Fig. 6-21. Murine models of candidiasis for triploid strains. A) DC murine model. ICR mice were inoculated via the lateral tail vein with 106 CFU. B) PMNs phagocytosis (left panel) and killing (right panel) of the wild type CAI-12, or the triploid 1 and 3 KO mutants. C) OPC murine model. Cortisone immunosuppressed mice were infected orally with 108 CFU. The bar graph represents the log CFU/g recovered from the tongue, esophagus, and jaw muscles from mice infected with the wild type CAI-12 strain, triploid 1 KO, or the triploid 3 KO mutant. intensity for allele B to allele A was 2:1). Also, after passage 1 in the presence of Hygromycin B, 90% of the colonies analyzed still had the duplication. Conversely, after passage 20 (~400 generations) in YPD and Hygromycin B, 45% and 95% respectively of the colonies analyzed lost the duplication at IPF15632 loci. Stability of the duplication in the 1 KO triploid was analyzed after growth in organs of ICR mice inoculated

PAGE 137

119 intravenously with 105 CFU. After six days post infection 20 colonies isolated from mice kidneys and 2 from spleen maintained the duplication. Therefore, the duplication in our 1 KO triploid strain remained stable after limited passage in rich media. However, the duplication at IPF15632 loci was lost after passage 20 when grown in rich media or when grown under stressful conditions, such as in presence of the aminoglycoside Hygromcyin B. Interestingly, growth in kidneys was not a stress that led to loss of duplication, this could be due to the possibility that the triploidy did not have detrimental effects during DC. The fact that the duplication was not stable after continuous passage on YPD or under stressful conditions when grown in presence of Hygromycin B indicates that C. albicans prefers to be in the diploid state. (See table 6-3) Table 6-3. Maintenance of duplication in the triploid 1 KO strain Passage Ratio YPD Hygromycin B Mouse Kidney and Spleen 1 2:1 100% 90% 100% 20 2:1 55% 5%

PAGE 138

CHAPTER 7 DISCUSSION AND CONCLUSIONS C. albicans is a commensal organism found in the oral cavity, GI tract, and vaginal tract of healthy individuals. Changes in the host environment determine if C. albicans remains as a commensal or if it proliferates and causes disease. This commensal state depends on the integrity of the host’s tissues and the immune system, when these defenses are breached, the opportunity is provided for C. albicans to proliferate resulting in invasive disease. C. albicans is a successful commensal and pathogen, able to persist in the host and cause a wide variety on infections ranging from superficial and mucosal to disseminated candidiasis. Nonetheless, impairment of host defense is not sufficient to result in candidiasis. The expression of virulence factors that allow C. albicans to adhere to, invade, and cause damage to host tissue are also required. In order for C. albicans, like other pathogens, to adapt to the dynamic and changing environment present during infection, it has to express genes necessary for growth and survival within the host. Thus, genes expressed by the pathogen in vivo reflect the response to the different environmental conditions encountered during infection, such as nutrients available, pH, and challenges by the host immune system. We sought to identify genes expressed by C. albicans during infection to gain insight into the mechanisms of C. albicans pathogenesis. We hypothesize that selected genes expressed during infection will encode factors that contribute to pathogenesis and survival within the host. To prove our hypothesis, we first identified C. albicans genes expressed during 120

PAGE 139

121 OPC in HIV-infected patients. We then characterized and determined the role of the genes identified during C. albicans pathogenesis using murine models of candidiasis. Identification of C. albicans Genes Expressed during OPC using IVIAT We identified genes expressed during infection using an antibody screening approach, IVIAT. This method allowed us to identify genes expressed during OPC by screening a C. albicans genomic expression library with sera from HIV-infected patients with active oral thrush. The sera had been previously enriched for antibodies that will recognize proteins expressed mainly during OPC with a pre-adsorption step against a single C. albicans clinical isolate grown in culture. This screening strategy allowed to identify previously known and also unrecognized virulence factors. (27, 28) IVIAT Advantages and Limitations Some of the advantages of IVIAT over other methods for identifying in vivo expressed genes is that it does not require the use of an animal model. Instead we detect C. albicans gene expressed in humans, thus, preventing confusing results due to gene expression within a different host. IVIAT is relatively simple since a genomic expression library created in E. coli is screened rather than creating a cDNA library, which could have a limited expression of some genes. However, as with other methods, IVIAT also has its drawbacks. This method only allows the identification of antigenic proteins that can be expressed in E. coli. Proteins that are not expressed in E. coli would not be identified, nor would proteins that are not antigenic or carbohydrates that are not detected by the system we used. Moreover, IVIAT screening does not take into account for post-translational modifications, such as glycosylation, phosphorylation, and ubiquitination, that occur within C. albicans, which could change the epitopes recognized by antibodies in human sera. Also, by the nature of

PAGE 140

122 C. albicans normal commensal relationship with the human host, the sera used for screening would not only identify proteins important during the pathogenic stage of C. albicans but also during its commensal stage. Therefore, genes expressed by C. albicans for survival or growth within the host will also be identified. Finally, due to the fact that we used sera from multiple individuals to identify genes expressed by the C. albicans library we could have potentially diluted those antibodies that existed in low concentrations. We might also not be able to recognize genes which are not highly expressed. In order to fully understand host-pathogen interactions it is important to use different in vivo expression methods. The results obtained from the combination of these different methods would provide further insight into the different aspects of C. albicans complex relationship with the human host during commensalism and pathogenesis. Nonetheless, our goal is not to identify every gene or virulence factor expressed in vivo. Rather, the screening method is a tool to identify selected novel virulence factors. RBF1 and CDC24: Known Virulence Factors Expressed During OPC Among the proteins I identified, RBF1 and CDC24 had been previously described to be required for C. albicans’ morphogenetic switch, a known virulence attribute. Disruption of RBF1 induced filamentous growth and Rbf1 null mutants were attenuated in a murine model of disseminated candidiasis (7, 87, 88). Heterozygous mutants expressing one copy of CDC24 under the MET3 promoter had impaired development of the filamentous form of C. albicans and were also less virulent in a murine model of DC candidiasis (9, 10). These results further confirm the importance that both C. albicans forms, yeast and hyphae, are required for pathogenesis. The yeast form is thought to allow for dissemination and spread of infection, and the filamentous form is required for

PAGE 141

123 penetration into host tissues (1). Furthermore, we confirmed by RT-PCR with RNA extracted from C. albicans within pseudomembranes of patients with thrush that both RBF1 and CDC24 are expressed at higher levels during OPC compared to routine growth in liquid media. Thus, these results corroborated our hypothesis that selected genes expressed during OPC are important for virulence, and validated the role of IVIAT in identifying virulence factors. IPF11959 is Expressed by C. albicans Hyphae During OPC I also identified three other proteins, IPF11959, ALG5, and IPF15632, whose role in pathogenesis had not been determined yet. The first protein identified in the screen, IPF11959, was encoded by a gene assigned the systematic name of PTH1, for proline transporter helper protein 1, in a different C. albicans strain (ATCC1001) from the one we used for gene analysis (SC5314). Presence of proline in culture media is known to promote hyphal formation in C. albicans (117). In addition, proline uptake transporters have been found to be required for Staphylococcus aureus survival in vivo (155). Therefore, expression of IPF11959 might allow for scavenging of amino acids found in limited amounts within the host. We verified by RT-PCR with RNA extracted from C. albicans in pseudomembranes recovered from patients with thrush that IPF11959 is also expressed during OPC. Moreover, we showed the expression of IPF11959p by direct visualization of the antigen along the walls of C. albicans hyphae in pseudomembranes from patients with OPC. Not only do these results confirm the expression of IPF11959 during OPC but they also suggest that the signal sequence and transmembrane domains identified by sequence analysis indeed have a function in localizing IPF11959 along the cell wall of C. albicans hyphae. Unfortunately, further characterization of IPF11959 was not possible at the time because of the unfinished sequencing of the C. albicans genome.

PAGE 142

124 However, the results provided so far for IPF119595 suggest that it could have a potential role during pathogenesis of C. albicans possibly by contributing to nutrient uptake by hyphal cells during infection. Possible Reasons for Why Genes Identified by IVIAT are also Expressed In Vitro Even though we screened the C. albicans genomic expression library with pooled sera from 24 HIV-infected patients with OPC, which was previously enriched for antibodies that recognize mainly proteins expressed in vivo, we still observed expression in vitro for some of our genes. One possibility for expression of genes identified in the cultured cells could be due to the fact that the sera from patients with OPC was pre-adsorbed only against one C. albicans clinical isolate (Ca172) grown in YPD at 37 C (27). Therefore, when we performed RT-PCR to compare the expression of the identified gene during OPC and routine growth in YPD, there are several factors that could lead to gene expression in vitro. First, different C. albicans clinical isolates were used for performing RT-PCR for in vivo and in vitro conditions. It is known that C. albicans clinical isolates have chromosome size and gene sequence variation (29, 119). It is also known that expression of genes as revealed by the presence of diverse colony morphologies, is an indication of variation in cell surface gene expression (74, 75, 185). Thus, it is possible that the expression of several of the genes identified differed between the C. albicans isolates used to study gene expression and the Ca172 isolate used for the pre-adsorption step. For example, down regulation of a particular gene in Ca172 when grown in vitro could lead to reduced protein expression, consequently all the antibodies against it would not be removed during the pre-adsorption step. Second, sequence variation among the same protein in different clinical isolates could lead to loss or gain of

PAGE 143

125 epitopes. Therefore, the adsorption of the pooled human sera against proteins from one clinical isolate will not eliminate all the antibodies that recognize the different proteins variants expressed in vitro due variation of epitopes present among the different isolates. For example, IPF11959 sequence differs at the N-terminus between the C. albicans ATC1001 strain and SC5314 strain. Second, the growth condition (YPD at 37 C) to prepare Ca172 for the pre-adsorption step differed from the growth condition of clinical isolates (YPD at 30 C) prior to RNA extraction. As a result, temperature differences could be another source of variation in gene expression too. Third, the library was constructed with 24 C. albicans clinical isolates that were different from the ones infecting the HIV-positive patients with OPC from whom the sera were obtained. Thus, variation between the clinical isolates used for protein expression from those to which antibodies were mounted during OPC could lead to differences in protein expression due to strain variation. Finally, we need to take into account that IVIAT selects for antigens that are preferentially expressed in the host. As mentioned, these might include antigens required for C. albicans colonization, as well as pathogenesis of candidiasis. Also, antigens preferentially expressed in the host and required for colonization or infection might be expressed to a lesser extent in culture. Indeed, we confirmed this for the known virulence factors RBF1 and CDC24. Identification of C. albicans Antigenic Intracellular Proteins with Sera from Patients with OPC Several of the proteins identified with IVIAT included antigenic intracellular proteins, such as the transcription factor RBF1. Identification of antigenic C. albicans intracellular proteins, such as hsp90 heat shock protein and enzymes involved in

PAGE 144

126 glycolysis and metabolism, has been shown in other studies too (172). There are several possible reasons for these observations. Antibodies could be mounted to C. albicans intracellular proteins due to their release by lysis or damage of cells by the immune system, leading to peptide uptake and processing by antigen presenting cells at the site of infection. Moreover, several proteins which have been predicted to be localized intracellulary in C. albicans have been found to localize to the cell wall (45). For example, hsp90 is not only localized to the cytosol, but it is also present in the cell wall (23). In addition, glycolytic enzymes, such as enolase (6), phosphoglycerate kinase (5), GAPDH (60), and alcohol dehydrogenase (93), have been shown to be cell wall bound. (23) Mechanisms by which these proteins are secreted or the roles they play on the cell wall are not yet completely understood. One hypothesis is that these intracellular proteins are secreted by substrate specific transporters, and one putative role for them is to adhere to host proteins (23). ALG5 and IPF15632 We sought to determine if ALG5 and IPF15632 contributed to Candida pathogenesis by creating isogenic mutants and analyzing their virulence using cell and murine models of candidiasis. If so, this would confirm that genes expressed during OPC and identified with antibodies encode previously unrecognized virulence factors. Characterization of C. albicans ALG5 ALG5 encodes a putative UDP-glucose: dolichol phosphate glucosyltransferase involved in cell wall mannan biosynthesis. The putative function of C. albicans ALG5 was assigned by homology (47.4%) to the S. cerevisiae ortholog. S. cerevisiae ALG5 provides glucose from UDP (uridine diphosphate)-glucose to the dolichol-linked

PAGE 145

127 oligosaccharide precursor, located at the ER membrane, prior to N-glycosylation of a protein (73). The three main macromolecules found in C. albicans and S. cerevisiae cell wall are chitin (unbranched polymers of -1,4 N-acetyl-D-glucosamine), -glucan (branched -1,3 and -1,6 glucose polymers), and mannan (polymers of mannose, with or without glucose, usually covalently bound to proteins) (14, 15, 23). Chitin and glucan provide the cell wall with its strength and rigidity to withstand damage to the cell by osmotic or other environmental attacks (23, 160). Mannan, found covalently bound to proteins and designated as mannoproteins (16), is deposited in the outer surface of the cell wall, thus covering it with a layer (160). The layer of mannoproteins provides internal regions of the cell wall with protection against proteases (13, 187). Biosynthesis of mannoprotein starts in the endoplasmic reticulum (ER) with further chain elongation in the Golgi, and then follows the secretory pathway for deposition on the outer surface of the cell wall. (15, 23, 76) Mannosylation of cell wall proteins can occur either by N (asparagine)-linked or O (serine/threonine)-linked glycosylation.(Lehle 95, in Glycoproteins; eds) Secreted proteins, as well, are modified by the attachment of short glycosyl chains (ref). S. cerevisiae ALG5 is involved in N-glycosylation at asparagine residues in the ER. Both of S. cerevisiae ALG5 and ALG6 are involved in the transfer of the first glucose unit to Man9GlcNAc2-P-P-dolichol, thus mutants for either gene accumulate Man9GlcNAc2 (83, 151). Since mannoproteins are located in the outer surface of the cell wall they mediate the initial interaction between C. albicans and the host. Adhesion of C. albicans is in part mediated by mannoproteins present in the cell wall. Recent studies have demonstrated

PAGE 146

128 that the glycosylated mannoproteins in the outer surface of the cell wall directly interact with host ligands (168). In fact, glycosylation defects due to the disruption of enzymes involved in glycosylation of mannan have been shown to impair C. albicans virulence (13, 165, 168, 174, 175, 175). Furthermore, mannoproteins have also been shown to be recognized during host invasion, and they are able to modulate the host immune response by inhibiting Candida-specific proliferation of normal human lymphocytes (47). Therefore, mannoproteins contribute to C. albicans pathogenesis by adhesion, immuno-surveillance, and immuno-modulation. Based on these reasons and that C. albicans ALG5 is probably involved in mannan biosynthesis, we hypothesized that ALG5 could be a potential virulence factor. ALG5 and its Role in C. albicans Cell Wall Our studies suggest that C. albicans ALG5 is associated with cell wall biosynthesis, since alg5 null mutant displayed a slight sensitivity to calcofluor white, a cell wall effector that binds chitin. Sensitivity to calcofluor white has been found as a pleitropic phenotype associated with yeast mutants that have general cell wall defects (64). However, the integrity of the C. albicans alg5 mutant’s cell wall was not completely compromised since we did not observe sensitivity to ionic detergent (0.06% SDS) but instead resistance. Thus, it is possible that the alg5 null mutant’s cell wall structure or composition was altered, leading to susceptibility to the chitin binding fluorochrome but resistance to SDS. Similar results of increased sensitivity to calcofluor white but decreased sensitivity to SDS were reported by Moreno et. al. for C. albicans cwt1 mutants, with disruption of a transcription factor required for cell wall integrity (122). The authors demonstrated that these results were due to changes in the concentration of

PAGE 147

129 cell wall components, in this case a diminishment of the -glucan fraction and an increase in the mannan fraction in the null mutant compared to their wild type. Thus, it is likely that alg5 mutants outermost cellular layer has incomplete or defective glycosylated mannoproteins and/or other glycoproteins present in the cell wall. These defective proteins are not preventing access of calcolfluor white to the innermost layer where chitin is located. However, possible changes in the composition or structure of the cell wall could allow the formation of unusual covalent bonds between glycoproteins and other cell wall proteins, thus making the cell more resistant to SDS. ALG5 Role in Glycosylation We verified that the C. albicans alg5 mutant is hypersensitive towards the aminoglycoside G418 and to Hygromycin B. In yeast, sensitivity to aminoglycosides correlates with defects in glycosylation (37). In fact, it has been previously reported that defects in all aspects of glycosylation lead to hypersensitivity to aminoglycosides and resistance to vanadate (37). Yeast mutants with glycosylation defects are usually screened by inhibition of growth with either Hygromycin B or Geneticin (G418), both of which are considered the most potent aminoglycoside inhibitors (37). Susceptibility to aminoglycosides has been described for other C. albicans proteins with a role in glycosylation, such as MNN9 (165), Pmt1 (174) and 6 (175), and GDPase (76). Therefore, we can speculate from the above results that C. albicans ALG5 is possibly involved in glycosylation of cell wall proteins, as is its S. cerevisiae homolog. Aminoglycosides are antibiotics that interfere with RNA-protein interactions and promote mistranslation of the genetic code in prokaryotes and eukaryotes (66, 67). In bacteria, aminoglycosides interact with the 16S ribosomal RNA and thus inhibit

PAGE 148

130 translation. In eukaryotes, they bind to a specific RNA motif and block splicing of group 1 introns. (37) The molecular mechanism of how defects in glycosylation lead to aminoglycoside susceptibility or permeability are not known. It is speculated that N-glycosylation of one or more cell wall or plasma membrane proteins are required for preventing easy import or allowing export of aminoglycosides, or that N-linked glycoproteins are involved in detoxification of aminoglycosides once inside the cell. (32, 37) ALG5 Has No Role in Adherence to Human Epithelial Cells or in Pathogenicity of DC Microbial adherence to host tissues is a crucial event in the establishment of colonization and pathogenesis. The cell wall of C. albicans is the first point of interaction between the fungus and the host (16) and thus mediates adherence. Mannoprotein, which compromise 30-40% of the cell wall and provide the cell surface with its properties, allows C. albicans to adhere to host cells, such as epithelial and endothelial cells (23). Adhesins in C. albicans are usually highly glycosylated cell wall proteins or polysaccharides (16, 169). Consequently, defects that lead to underglycosylation of cell wall proteins could affect adherence and pathogenesis of Candida. Indeed, it has been shown for C. albicans that defects in protein glycosylation affect the stability and function of the cellular surface and lead to loss of the adhesins binding sites (174). Since, C. albicans alg5 null mutants had defects in glycosylation of cell wall proteins we examined ALG5 role in adherence and later in virulence using a murine model of disseminated candidiasis. However, despite the observed cell wall changes, we found no difference in adherence to human buccal epithelial cells. Unlike other proteins involved in glycosylation, such as Pmt1 and 6, whose mutants were shown to adhere

PAGE 149

131 significantly less to different epithelial cells, alg5 mutants did not display any defect in adherence in vitro compared to the wild type. Adherence to in vitro epithelial or endothelial cells are ex vivo models used to mimic the first crucial step for establishment of colonization and OPC. However, we still need to investigate if alg5 null mutants have adherence defects to other human epithelial cells or if their virulence is attenuated in the murine model of OPC. Furthermore, alg5 null mutants were as virulent as the wild type when infecting mice intravenously with 105 CFU. Thus, the phenotypic alterations observed for alg5 mutant’s cell wall were not enough to affect growth rate, morphogenesis, adherence to human epithelial cells, or virulence in a murine model of DC. Biosynthesis of polysaccharides involves the action of many different glycosyltransferases. For instance, C. albicans has twenty-one genes that have been directly annotated to N-linked glycosylation and seven of these genes belong to the ALG family. In particular, ALG6 and ALG8, both of which are glucosyltransferases, are involved in the addition of three glucose residues in the final synthesis step of the oligosaccharide precursor. ALG6 shares the function with ALG5 in transferring the first glucose residue, while ALG8 transfers the second glucose residue. Therefore, a possible reason why we did not observe any defects in adherence or virulence in alg5 mutants could be due to compensation for the of ALG5 function by ALG6. It is possible that if either ALG6 or ALG8 are disrupted along with ALG5 we would see a more pronounced effect in glycosylation and consequently in adherence and virulence. Thus, there is a lot of redundancy in glycosylation of C. albicans cell wall proteins, and others enzymes may contribute to proper glycosylation of mannoproteins or other cell wall proteins.

PAGE 150

132 Characterization of C. albicans IPF15632 We identified IPF15632 as an antigenic protein that is reactive with sera from patients with OPC. IPF15632 has no homology with any gene in GenBank or in the S. cerevisiae database. Thus, being an apparently novel protein, we decided to characterize it in order to determine its role in C. albicans pathogenesis. In the CandidaDB IPF15632 is listed as being weakly similar to the S. cerevisiae Muc1/Flo11/STA4 gene (15% similarity at the amino acid level). The FLO genes are recognized for their role in the brewing industry, because their expression leads to agglutination and sedimentation of yeast. In the past two decades investigators have also started to analyze the implications of the FLO genes in the biology of S. cerevisiae. S. cerevisiae FLO genes encode proteins that confer adherence to agar, solid surfaces, and cell-cell adherence between yeast. All the Flo proteins confer distinct cell surface properties, which provide the yeast with cell surface variation. (61, 63, 142) The FLO gene family in S. cerevisiae has five members: FLO1, 5, 9, 10, and 11 (63). FLO11 encodes a cell surface glycoprotein called flocculin, which is involved in adhesion to agar and plastic surfaces, sliding motility, cell-cell adhesion, and pseudohyphal growth under nitrogen limiting conditions (52, 99, 142). However, epigenetic regulation of FLO11 has been implicated in the variation of filamentation often observed within a homogenous colony of S. cerevisiae. This occurs when only a subset of cells within a single colony are expressing FLO11 filament, presumably due to epigenetic modifications, while other cells do not (63). IPF15632 also contains a short amino acid sequence domain (GADLLMYLATSP), which is highly conserved in several ascomycetes. This domain has not yet been characterized. The only protein that is known to contain this motif and whose function is

PAGE 151

133 known is S. cerevisiae STB1. STB1 is a transcriptional regulator that is involved in the activation of heterodimeric transcription factors required for the initiation of transcription at the Start point in the G1 phase of the cell cycle, thus committing the cells to divide (78). IPF15632 and Cell-Surface Adhesion When examining the growth rate from isolated cells of the IPF15632 null mutants on the surface of YPD agar at 37 C, I found that the colonies from the mutants were smaller than those from the wild type. Consistently, mutant colonies were approximately 25% smaller than those of the wild type. Furthermore, we observed that single colonies of null mutants embedded in YPGlycerol agar, grown at 37 C, were not able to either invade or form long filaments as well as the isogenic wild type. I was later able to confirm that this phenotype was due to the deletion of IPF15632 and not due simply to random mutations, because a revertant strain, that had one copy of IPF15632 re-integrated at its own locus was able to form the same size colonies as the wild type. A recent study by Reynolds et al. (142) which looked at the ability of S. cerevisiae to form biofilm, demonstrated that FLO11 is required for attachment to plastic and mat formation. “Mat” is a colony-like complex multicellular structure composed of cells in the yeast form. This study shows that a S. cerevisiae flo11 grew smaller mats on semi-solid (0.3%) agar than strains expressing FLO11. Furthermore, they showed that expression of Flo11 on the surface of cells changes the mat morphology. Expression of Flo11 on the surface of a subset of cells within a colony provides cell-cell adhesion properties that allow the formation of the unusual but organized and complex colony structure. The authors suggested that the ability of Flo11’s to confer yeast cells with both

PAGE 152

134 cell-cell and cell-surface adhesion properties may in turn be due to the increased hydrophobicity of cells. Previous reports have mentioned that cells expressing FLO11 were actually more hydrophobic than those that do not (182). Thus, expression of FLO11 reduces the interaction of yeast cells with the aqueous surface by increasing cell-cell interaction which allows them to spread and form bigger colonies. Therefore, there are at least two possible mechanisms that would explain the phenotype of the IPF15632 null mutant. One possibility is that IPF15632 encodes a protein with a similar function to Flo11. In this case, IPF15632 mutants could be less hydrophobic and display higher frictional resistance to sliding on agar, which leads to smaller colonies compared to the wild type. Further evidence comes from the altered cell wall properties of IPF15632 null mutants. Mutants were slightly more sensitive to calcofluor white, which suggests changes in cell wall composition or structure. Also, colony morphology of the mutants on M199 agar was different from the wild type, the mutant displayed a wrinkled colony morphology while the wild type displayed a flat and organized colony morphology. Changes in colony morphology are usually a reflection of altered expression of surface proteins (111). One way to determine if the cell surface hydrophobicity of the IPF15632 null mutant has changed would be to measure how much Alcian blue dye the cells bind. Previous studies have shown that increased surface hydrophobicity of Candida albicans can be recognized by the binding of Alcian blue (113, 132). Studies have also linked increases in the hydrophobicity of C. albicans cell walls with virulence (33, 72). Cells that are more hydrophobic are able to adhere better to epithelial and endothelial cells as well as extracellular matrix and plastic devices, such as catheters (1, 71). For this reason,

PAGE 153

135 increased hydrophobicity has now been proposed as a virulence determinant (33). Another aspect that one could investigate is the size of colonies of each mutant on soft agar. The experiments we carried out for examining colony growth contained 2% agar, which is relatively rigid and contains less water at the surface. We might be able to observe an even more pronounced colony growth defect on media containing 0.3% agar if cell-cell adhesion can overcome adhesion to the aqueous component of the medium, as shown for S. cerevisiae flo11 mutants. A second possibility that would explain the phenotype of the IPF15632 null mutant arises from the putative domains identified in the IPF15632 protein sequence. IPF15632 contains a conserved motif also present in S. cerevisiae STB1, which is a transcription factor localized in the nucleus and is involved in the regulation of cell division. Since IPF15632 also contains this highly conserved motif as well as several putative nuclear localization signals, this sequence analysis suggests that it may also be a transcription factor that is possibly involved in the regulation of cell division or in the expression of FLO-like genes in C. albicans. The likelihood that IPF15632 may be a transcription factor is further supported by the presence of a coiled-coil domain at the N-terminus and glutamine repeats within the ORF, both of which are characteristic of yeast transcription factors. C. albicans IPF15632 may be acting as a transcription factor in the nucleus to regulate expression of FLO-like genes or other genes encoding cell wall proteins. This suggests that the phenotypes we observed were due to downregulation of other genes with a function in maintaining cell surface properties. We could determine if IPF15632 is localized to the nucleus by immunostaining of cells expressing Myc-tagged IPF15632

PAGE 154

136 protein. Furthermore, future transcriptional profiling studies of the mutants might allow the determination of which genes are upregulated or downregulated and are also responsible for the phenotypes observed. IPF15632 Role in Pathogenesis Either of the two functions proposed for IPF15632p, as described above, could lead to reduced pathogenicity of the IPF15632 null mutants, since colony growth is required for colonization and infection. We confirmed that IPF15632 null mutants were significantly less virulent than the wild type in a murine model of DC. To determine a possible mechanism for this attenuated virulence, I analyzed fungal burden in mouse kidney, liver, and spleen one and four days post infection with either wild type or IPF15632 mutant. We found that IPF15632 mutants were able to colonize all three organs as well as the wild type at one day post infection. However, the fungal burden for the mutant at day four was significantly lower in the kidney and spleen relative to wild type. In the mouse kidney, both mutant and wild type cells were able to grow, as shown by an increase of about 0.9 and 1.7 log CFU/g of kidney on day 4, respectively, relative to levels observed on day 1. Despite the increase in fungal burden by both strains, the average mutant CFU/g in the kidney was significantly lower than the wild type. This might reflect the fact that the mutant cells were not able to persist on the organs at day four, probably due to clearance by the immune system during infection. Clearance of cells also appears to be the contributing factor for the absence of IPF15632 mutant cells in the spleen, while wild type cells were still present. Clinical studies have shown that PMNs provide a major contribution to the defense of disseminated/deep-seated infections (36).

PAGE 155

137 Since expression of FLO genes in S. cerevisiae provides cells with distinct surface properties, deletion of FLO11 also causes changes in the cell surface. Similarly, deletion of IPF15632 could cause alterations in the cell wall of the mutant, which could make the yeast cells more susceptible to or more easily recognized by polymorphonuclear cells (PMNs) or macrophages. This would in turn lead to faster clearance of IPF15632 mutant cells from the kidney and spleen, accounting for fewer CFU present in organs by 4 days post infection. Follow up experiments will analyze whether IPF15632 mutants are more easily phagocytosed and/or killed by human PMNs. Results from these experiments would help determine if the rapid clearance of IPF15632 mutants by the immune system plays a dominant role in the attenuated virulence we have already observed. Finally, it will be interesting to determine the adherence of the mutant to endothelial and epithelial cells as a possible mechanisms for virulence. Future studies involve determining if IPF15632 also confers virulence in an OPC murine model, since this protein was identified using sera from patients with OPC. Triploid Strain at IPF15632 Locus We have shown that a triploid strain with deletion of one copy of IPF15632 was created by duplication of a segment of chromosome 6. This triploid 1 KO strain had general defects in the cell wall, thus affecting its integrity. Alteration of cell wall properties can cause defects in adhesion to host cells as we have shown that this triploid 1 KO strain had reduce adherence to primary human buccal epithelial cells and to human pharyngeal and colon epithelial cell lines. Furthermore, this strain was less virulent than a wild type strain in a murine model of oropharyngeal candidiasis, but it remained as virulent as the wild type in a murine model of disseminated candidiasis.

PAGE 156

138 Chromosomal rearrangements and duplications are known to contribute to the genomic diversity of C. albicans (173). For example, clinical isolates have been found to have different karyotypes and these display different phenotypes which could enhanced virulence. The strain we created with an unintentional chromosomal duplication is a further example of how duplication of certain chromosomes or segments might influence virulence. These results also provide further evidence that C. albicans virulence can be site-specific due to expression or repression of particular genes required for pathogenesis during disseminated candidiasis but not during OPC. Each site of infection, such as the oral cavity or the blood, is a different niche and exerts unique environmental pressures. Because C. albicans is able to survive and proliferate in diverse tissues, specific virulence factors are likely to be important during different types and stages of infection (81). These stage and site-specific differences imply that certain virulence factors can confer a temporal and/or tissue-specific virulence. We have shown that triploid 1 KO IPF15632 mutant cannot establish colonization in a murine model of OPC but it has a role in a murine model of DC. These results suggest that expression and/or repression of certain genes in our triploid strain provide site-specific virulence. To my knowledge, our triploid strain is the first C. albicans strain that is attenuated in a murine model of OPC but not in DC. Thus, our triploid mutants offers a tool with which to further study site-specific virulence. Further characterization of this triploid strain might elucidate the pathogenesis of C. albicans at the mucosal site, since the underlying reasons for attenuation of adherence and virulence in OPC is still unclear. Future studies involve determining if

PAGE 157

139 other genes apart from IPF15632 have been duplicated and studying gene expression of the triploid mutants during OPC. Concluding Remarks By screening a C. albicans genomic library with sera from HIV-infected patients with OPC previously depleted for antibodies that react with proteins expressed in vitro, we were able to identify both previously recognized (RBF1 and CDC24) as well as novel (IPF15632) virulence factors. These data support my hypothesis that certain genes expressed particularly during OPC may be important for pathogenesis of C. albicans. From the five genes I identified, two genes (ALG5 and IPF15632) were characterized and their role in pathogenicity was analyzed. ALG5 did not contribute directly to pathogenesis probably due to redundancy in N-glycosylation and compensation of ALG5 by other enzymes in the ALG family. On the other hand, IPF15632 contributes to disseminated candidiasis. IPF15632 could confer virulence by allowing C. albicans cells to either adhere and invade tissues or to prevent clearance by the immune system. Furthermore, I have created a C. albicans strain with a duplication of all or part of chromosome 6 that has alterations in the cell wall and interferes with virulence in OPC but not in DC.

PAGE 158

LIST OF REFERENCES Reference List 1. (2002) Candida and Candidiasis (ASM Press, Washington). 2. Akpan, A. & Morgan, R. (2002) Postgrad. Med. J. 78, 455-459. 3. Alaei, S., Larcher, C., Ebenbichler, C., Prodinger, W. M., Janatova, J. & Dierich, M. P. (1993) Infect. Immun. 61, 1395-1399. 4. Alloush, H. M., Lopez-Ribot, J. L. & Chaffin, W. L. (1996) J. Med. Vet. Mycol. 34, 91-97. 5. Alloush, H. M., Lopez-Ribot, J. L., Masten, B. J. & Chaffin, W. L. (1997) Microbiology 143 ( Pt 2), 321-330. 6. Angiolella, L., Facchin, M., Stringaro, A., Maras, B., Simonetti, N. & Cassone, A. (1996) J. Infect. Dis. 173, 684-690. 7. Aoki, Y., Ishii, N., Watanabe, M., Yoshihara, F. & Arisawa, M. (1998) Nippon Ishinkin. Gakkai Zasshi 39, 67-71. 8. Bain, J. M., Stubberfield, C. & Gow, N. A. (2001) FEMS Microbiol. Lett. 204, 323-328. 9. Bassilana, M., Blyth, J. & Arkowitz, R. A. (2003) Eukaryot. Cell 2, 9-18. 10. Bassilana, M., Hopkins, J. & Arkowitz, R. A. (2005) Eukaryot. Cell 4, 588-603. 11. Bein, M., Schaller, M. & Korting, H. C. (2002) Curr. Drug Targets. 3, 351-357. 12. Berman, J. & Sudbery, P. E. (2002) Nat. Rev. Genet. 3, 918-930. 13. Buurman, E. T., Westwater, C., Hube, B., Brown, A. J., Odds, F. C. & Gow, N. A. (1998) Proc. Natl. Acad. Sci. U. S. A 95, 7670-7675. 14. Cabib, E., Roberts, R. & Bowers, B. (1982) Annu. Rev. Biochem. 51, 763-793. 15. Cabib, E., Roh, D. H., Schmidt, M., Crotti, L. B. & Varma, A. (2001) J. Biol. Chem. 276, 19679-19682. 140

PAGE 159

141 16. Calderone, R. A. (1993) Trends Microbiol. 1, 55-58. 17. Calderone, R. A. & Fonzi, W. A. (2001) Trends Microbiol. 9, 327-335. 18. Campbell, J. A., Davies, G. J., Bulone, V. & Henrissat, B. (1997) Biochem. J. 326 ( Pt 3), 929-939. 19. Cao, S. L., Progulske-Fox, A., Hillman, J. D. & Handfield, M. (2004) FEMS Microbiol. Lett. 237, 97-103. 20. Casadevall, A. & Pirofski, L. (2001) J. Infect. Dis. 184, 337-344. 21. Casadevall, A. & Pirofski, L. A. (1999) Infect. Immun. 67, 3703-3713. 22. Cassone, A., De Bernardis, F., Ausiello, C. M., Gomez, M. J., Boccanera, M., La Valle, R. & Torosantucci, A. (1998) Res. Immunol. 149, 289-299. 23. Chaffin, W. L., Lopez-Ribot, J. L., Casanova, M., Gozalbo, D. & Martinez, J. P. (1998) Microbiol. Mol. Biol. Rev. 62, 130-180. 24. Challacombe, S. J. (1994) Oral Surg. Oral Med. Oral Pathol. 78, 202-210. 25. Challacombe, S. J. & Sweet, S. P. (1997) Oral Dis. 3 Suppl 1, S79-S84. 26. Chen, X., Magee, B. B., Dawson, D., Magee, P. T. & Kumamoto, C. A. (2004) Mol. Microbiol. 51, 551-565. 27. Cheng, S., Clancy, C. J., Checkley, M. A., Handfield, M., Hillman, J. D., Progulske-Fox, A., Lewin, A. S., Fidel, P. L. & Nguyen, M. H. (2003) Mol. Microbiol. 48, 1275-1288. 28. Cheng, S., Nguyen, M. H., Zhang, Z., Jia, H., Handfield, M. & Clancy, C. J. (2003) Infect. Immun. 71, 6101-6103. 29. Chibana, H., Beckerman, J. L. & Magee, P. T. (2000) Genome Res. 10, 1865-1877. 30. Coogan, M. M., Sweet, S. P. & Challacombe, S. J. (1994) Infect. Immun. 62, 892-896. 31. Costanzo, M., Schub, O. & Andrews, B. (2003) Mol. Cell Biol. 23, 5064-5077. 32. Cumming, D. A. (1991) Glycobiology 1, 115-130. 33. Cutler, J. E. (1991) Annu. Rev. Microbiol. 45, 187-218. 34. De Bernardis, F., Muhlschlegel, F. A., Cassone, A. & Fonzi, W. A. (1998) Infect. Immun. 66, 3317-3325.

PAGE 160

142 35. de Repentigny, L. (2004) Curr. Opin. Microbiol. 7, 324-329. 36. de Repentigny, L., Lewandowski, D. & Jolicoeur, P. (2004) Clin. Microbiol. Rev. 17, 729-59, table. 37. Dean, N. (1995) Proc. Natl. Acad. Sci. U. S. A 92, 1287-1291. 38. Dodd, C. L., Greenspan, D., Katz, M. H., Westenhouse, J. L., Feigal, D. W. & Greenspan, J. S. (1991) AIDS 5, 1339-1343. 39. Drobacheff, C., Millon, L., Monod, M., Piarroux, R., Robinet, E., Laurent, R. & Meillet, D. (2001) Clin. Chem. Lab Med. 39, 519-526. 40. Dunker, A. K., Lawson, J. D., Brown, C. J., Williams, R. M., Romero, P., Oh, J. S., Oldfield, C. J., Campen, A. M., Ratliff, C. M., Hipps, K. W. et al. (2001) J. Mol. Graph. Model. 19, 26-59. 41. Edmond, M. B., Wallace, S. E., McClish, D. K., Pfaller, M. A., Jones, R. N. & Wenzel, R. P. (1999) Clin. Infect. Dis. 29, 239-244. 42. Edwards, J. E., Jr., Gaither, T. A., O'Shea, J. J., Rotrosen, D., Lawley, T. J., Wright, S. A., Frank, M. M. & Green, I. (1986) J. Immunol. 137, 3577-3583. 43. Ernst, J. F. (2000) Contrib. Microbiol. 5, 98-111. 44. Eroles, P., Sentandreu, M., Elorza, M. V. & Sentandreu, R. (1995) FEMS Microbiol. Lett. 128, 95-100. 45. Eroles, P., Sentandreu, M., Elorza, M. V. & Sentandreu, R. (1997) Microbiology 143 ( Pt 2), 313-320. 46. Fidel, P. L., Jr. & Sobel, J. D. (1996) Clin. Microbiol. Rev. 9, 335-348. 47. Fischer, A., Ballet, J. J. & Griscelli, C. (1978) J. Clin. Invest 62, 1005-1013. 48. Fong, I. W., Laurel, M. & Burford-Mason, A. (1997) Clin. Invest Med. 20, 85-93. 49. Fonzi, W. A. & Irwin, M. Y. (1993) Genetics 134, 717-728. 50. Fradin, C. & Hube, B. (2003) Adv. Appl. Microbiol. 53, 271-290. 51. Fradin, C., Kretschmar, M., Nichterlein, T., Gaillardin, C., d'Enfert, C. & Hube, B. (2003) Mol. Microbiol. 47, 1523-1543. 52. Gagiano, M., van Dyk, D., Bauer, F. F., Lambrechts, M. G. & Pretorius, I. S. (1999) Mol. Microbiol. 31, 103-116. 53. Gale, C., Finkel, D., Tao, N., Meinke, M., McClellan, M., Olson, J., Kendrick, K. & Hostetter, M. (1996) Proc. Natl. Acad. Sci. U. S. A 93, 357-361.

PAGE 161

143 54. Gale, C. A., Bendel, C. M., McClellan, M., Hauser, M., Becker, J. M., Berman, J. & Hostetter, M. K. (1998) Science 279, 1355-1358. 55. Ghannoum, M. A., Spellberg, B., Saporito-Irwin, S. M. & Fonzi, W. A. (1995) Infect. Immun. 63, 4528-4530. 56. Giaever, G., Chu, A. M., Ni, L., Connelly, C., Riles, L., Veronneau, S., Dow, S., Lucau-Danila, A., Anderson, K., Andre, B. et al. (2002) Nature 418, 387-391. 57. Gil-Navarro, I., Gil, M. L., Casanova, M., O'Connor, J. E., Martinez, J. P. & Gozalbo, D. (1997) J. Bacteriol. 179, 4992-4999. 58. Gillum, A. M., Tsay, E. Y. & Kirsch, D. R. (1984) Mol. Gen. Genet. 198, 179-182. 59. Gow, N. A., Robbins, P. W., Lester, J. W., Brown, A. J., Fonzi, W. A., Chapman, T. & Kinsman, O. S. (1994) Proc. Natl. Acad. Sci. U. S. A 91, 6216-6220. 60. Gozalbo, D., Gil-Navarro, I., Azorin, I., Renau-Piqueras, J., Martinez, J. P. & Gil, M. L. (1998) Infect. Immun. 66, 2052-2059. 61. Guo, B., Styles, C. A., Feng, Q. & Fink, G. R. (2000) Proc. Natl. Acad. Sci. U. S. A 97, 12158-12163. 62. Gustafson, K. S., Vercellotti, G. M., Bendel, C. M. & Hostetter, M. K. (1991) J. Clin. Invest 87, 1896-1902. 63. Halme, A., Bumgarner, S., Styles, C. & Fink, G. R. (2004) Cell 116, 405-415. 64. Hampsey, M. (1997) Yeast 13, 1099-1133. 65. Han, Y., Ulrich, M. A. & Cutler, J. E. (1999) J. Infect. Dis. 179, 1477-1484. 66. Hancock, R. E. (1981) J. Antimicrob. Chemother. 8, 249-276. 67. Hancock, R. E. (1981) J. Antimicrob. Chemother. 8, 429-445. 68. Handfield, M., Brady, L. J., Progulske-Fox, A. & Hillman, J. D. (2000) Trends Microbiol. 8, 336-339. 69. Handfield, M. & Levesque, R. C. (1999) FEMS Microbiol. Rev. 23, 69-91. 70. Haynes, K. (2001) Trends Microbiol. 9, 591-596. 71. Hazen, K. C. (1989) Infect. Immun. 57, 1894-1900. 72. Hazen, K. C. (1989) Infect. Immun. 57, 1894-1900.

PAGE 162

144 73. Heesen, S., Lehle, L., Weissmann, A. & Aebi, M. (1994) Eur. J. Biochem. 224, 71-79. 74. Hellstein, J., Vawter-Hugart, H., Fotos, P., Schmid, J. & Soll, D. R. (1993) J. Clin. Microbiol. 31, 3190-3199. 75. Hellstein, J. H., Fotos, P. G., Law, S. S., Kovacevic, M. & Carriere, K. C. (1993) J. Oral Pathol. Med. 22, 312-319. 76. Herrero, A. B., Uccelletti, D., Hirschberg, C. B., Dominguez, A. & Abeijon, C. (2002) Eukaryot. Cell 1, 420-431. 77. Hicks, G. R. & Raikhel, N. V. (1995) Annu. Rev. Cell Dev. Biol. 11, 155-188. 78. Ho, Y., Costanzo, M., Moore, L., Kobayashi, R. & Andrews, B. J. (1999) Mol. Cell Biol. 19, 5267-5278. 79. Hubbard, S. C. & Ivatt, R. J. (1981) Annu. Rev. Biochem. 50, 555-583. 80. Hube, B. (1996) Curr. Top. Med. Mycol. 7, 55-69. 81. Hube, B. (2004) Curr. Opin. Microbiol. 7, 336-341. 82. Hube, B. & Naglik, J. (2001) Microbiology 147, 1997-2005. 83. Huffaker, T. C. & Robbins, P. W. (1983) Proc. Natl. Acad. Sci. U. S. A 80, 7466-7470. 84. Hull, C. M., Raisner, R. M. & Johnson, A. D. (2000) Science 289, 307-310. 85. Ibata-Ombetta, S., Jouault, T., Trinel, P. A. & Poulain, D. (2001) J. Leukoc. Biol. 70, 149-154. 86. Imam, N., Carpenter, C. C., Mayer, K. H., Fisher, A., Stein, M. & Danforth, S. B. (1990) Am. J. Med. 89, 142-146. 87. Ishii, N., Yamamoto, M., Lahm, H. W., Iizumi, S., Yoshihara, F., Nakayama, H., Arisawa, M. & Aoki, Y. (1997) Microbiology 143 ( Pt 2), 417-427. 88. Ishii, N., Yamamoto, M., Yoshihara, F., Arisawa, M. & Aoki, Y. (1997) Microbiology 143 ( Pt 2), 429-435. 89. Jones, T., Federspiel, N. A., Chibana, H., Dungan, J., Kalman, S., Magee, B. B., Newport, G., Thorstenson, Y. R., Agabian, N., Magee, P. T. et al. (2004) Proc. Natl. Acad. Sci. U. S. A 101, 7329-7334. 90. Kamai, Y., Kubota, M., Kamai, Y., Hosokawa, T., Fukuoka, T. & Filler, S. G. (2001) Antimicrob. Agents Chemother. 45, 3195-3197.

PAGE 163

145 91. Kanbe, T., Li, R. K., Wadsworth, E., Calderone, R. A. & Cutler, J. E. (1991) Infect. Immun. 59, 1832-1838. 92. Klein, R. S., Harris, C. A., Small, C. B., Moll, B., Lesser, M. & Friedland, G. H. (1984) N. Engl. J. Med. 311, 354-358. 93. Klotz, S. A., Pendrak, M. L. & Hein, R. C. (2001) Microbiology 147, 3159-3164. 94. Kurpisz, M., Gupta, S. K., Fulgham, D. L. & Alexander, N. J. (1988) J. Immunol. Methods 115, 195-198. 95. Kvaal, C., Lachke, S. A., Srikantha, T., Daniels, K., McCoy, J. & Soll, D. R. (1999) Infect. Immun. 67, 6652-6662. 96. Lambrechts, M. G., Bauer, F. F., Marmur, J. & Pretorius, I. S. (1996) Proc. Natl. Acad. Sci. U. S. A 93, 8419-8424. 97. Lay, J., Henry, L. K., Clifford, J., Koltin, Y., Bulawa, C. E. & Becker, J. M. (1998) Infect. Immun. 66, 5301-5306. 98. Leidich, S. D., Ibrahim, A. S., Fu, Y., Koul, A., Jessup, C., Vitullo, J., Fonzi, W., Mirbod, F., Nakashima, S., Nozawa, Y. et al. (1998) J. Biol. Chem. 273, 26078-26086. 99. Lo, W. S. & Dranginis, A. M. (1998) Mol. Biol. Cell 9, 161-171. 100. Lockhart, S. R., Pujol, C., Daniels, K. J., Miller, M. G., Johnson, A. D., Pfaller, M. A. & Soll, D. R. (2002) Genetics 162, 737-745. 101. Lopez-Ribot, J. L., Alloush, H. M., Masten, B. J. & Chaffin, W. L. (1996) Infect. Immun. 64, 3333-3340. 102. Lorenz, M. C., Bender, J. A. & Fink, G. R. (2004) Eukaryot. Cell 3, 1076-1087. 103. Maccato, M. L. & Kaufman, R. H. (1991) Curr. Opin. Obstet. Gynecol. 3, 849-852. 104. Magee, B. B. & Magee, P. T. (2000) Science 289, 310-313. 105. Magee, P. T., Bowdin, L. & Staudinger, J. (1992) J. Clin. Microbiol. 30, 2674-2679. 106. Mahan, M. J., Heithoff, D. M., Sinsheimer, R. L. & Low, D. A. (2000) Annu. Rev. Genet. 34, 139-164. 107. Mahan, M. J., Slauch, J. M., Hanna, P. C., Camilli, A., Tobias, J. W., Waldor, M. K. & Mekalanos, J. J. (1993) Infect. Agents Dis. 2, 263-268. 108. Mahan, M. J., Slauch, J. M. & Mekalanos, J. J. (1993) Science 259, 686-688.

PAGE 164

146 109. Mahan, M. J., Tobias, J. W., Slauch, J. M., Hanna, P. C., Collier, R. J. & Mekalanos, J. J. (1995) Proc. Natl. Acad. Sci. U. S. A 92, 669-673. 110. Marodi, L., Korchak, H. M. & Johnston, R. B., Jr. (1991) J. Immunol. 146, 2783-2789. 111. Martinez, J. P., Gil, M. L., Casanova, M., Lopez-Ribot, J. L., Garcia, D. L. & Sentandreu, R. (1990) J. Gen. Microbiol. 136 ( Pt 12), 2421-2432. 112. Mason, J. M. & Arndt, K. M. (2004) Chembiochem. 5, 170-176. 113. Masuoka, J. & Hazen, K. C. (1999) Glycobiology 9, 1281-1286. 114. Matthews, R., Burnie, J., Smith, D., Clark, I., Midgley, J., Conolly, M. & Gazzard, B. (1988) Lancet 2, 263-266. 115. Matthews, R., Hodgetts, S. & Burnie, J. (1995) J. Infect. Dis. 171, 1668-1671. 116. Matthews, R. C., Burnie, J. P. & Tabaqchali, S. (1987) J. Clin. Microbiol. 25, 230-237. 117. Mattia, E., Carruba, G., Angiolella, L. & Cassone, A. (1982) J. Bacteriol. 152, 555-562. 118. Mekalanos, J. J. (1992) J. Bacteriol. 174, 1-7. 119. Merz, W. G., Connelly, C. & Hieter, P. (1988) J. Clin. Microbiol. 26, 842-845. 120. Mitchell, A. P. (1998) Curr. Opin. Microbiol. 1, 687-692. 121. Monod, M., Togni, G., Hube, B. & Sanglard, D. (1994) Mol. Microbiol. 13, 357-368. 122. Moreno, I., Pedreno, Y., Maicas, S., Sentandreu, R., Herrero, E. & Valentin, E. (2003) FEMS Microbiol. Lett. 226, 159-167. 123. Morschhauser, J. (2002) Biochim. Biophys. Acta 1587, 240-248. 124. Muhlschlegel, F. A. & Fonzi, W. A. (1997) Mol. Cell Biol. 17, 5960-5967. 125. Mukherjee, P. K., Seshan, K. R., Leidich, S. D., Chandra, J., Cole, G. T. & Ghannoum, M. A. (2001) Microbiology 147, 2585-2597. 126. Munro, C. A., Bates, S., Buurman, E. T., Hughes, H. B., MacCallum, D. M., Bertram, G., Atrih, A., Ferguson, M. A., Bain, J. M., Brand, A. et al. (2005) J. Biol. Chem. 280, 1051-1060. 127. Naglik, J., Albrecht, A., Bader, O. & Hube, B. (2004) Cell Microbiol. 6, 915-926.

PAGE 165

147 128. Naglik, J. R., Challacombe, S. J. & Hube, B. (2003) Microbiol. Mol. Biol. Rev. 67, 400-28, table. 129. Nakai, K. & Horton, P. (1999) Trends Biochem. Sci. 24, 34-36. 130. Navarro-Garcia, F., Sanchez, M., Nombela, C. & Pla, J. (2001) FEMS Microbiol. Rev. 25, 245-268. 131. Nguyen, M. H., Cheng, S. & Clancy, C. J. (2004) Med. Mycol. 42, 293-304. 132. Odani, T., Shimma, Y., Wang, X. H. & Jigami, Y. (1997) FEBS Lett. 420, 186-190. 133. Odds, F. C. (1979) Candida and candidosis (Leicester University Press, Leicester). 134. Odds, F. C. (1987) Crit Rev. Microbiol. 15, 1-5. 135. Odds, F. C. (1994) J. Am. Acad. Dermatol. 31, S2-S5. 136. Odds, F. C., Gow, N. A. & Brown, A. J. (2001) Genome Biol. 2, REVIEWS1009. 137. Pfaller, M. A., Jones, R. N., Doern, G. V., Sader, H. S., Hollis, R. J. & Messer, S. A. (1998) J. Clin. Microbiol. 36, 1886-1889. 138. Pitarch, A., Abian, J., Carrascal, M., Sanchez, M., Nombela, C. & Gil, C. (2004) Proteomics. 4, 3084-3106. 139. Pragman, A. A. & Schlievert, P. M. (2004) FEMS Immunol. Med. Microbiol. 42, 147-154. 140. Prigneau, O., Porta, A., Poudrier, J. A., Colonna-Romano, S., Noel, T. & Maresca, B. (2003) Yeast 20, 723-730. 141. Randolph, S. (2002) RN. 65, 41-44. 142. Reynolds, T. B. & Fink, G. R. (2001) Science 291, 878-881. 143. Robbins, J., Dilworth, S. M., Laskey, R. A. & Dingwall, C. (1991) Cell 64, 615-623. 144. Rollins, S. M., Peppercorn, A., Hang, L., Hillman, J. D., Calderwood, S. B., Handfield, M. & Ryan, E. T. (2005) Cell Microbiol. 7, 1-9. 145. Romani, L., Bistoni, F. & Puccetti, P. (2002) Trends Microbiol. 10, 508-514. 146. Romani, L., Bistoni, F. & Puccetti, P. (2003) Curr. Opin. Microbiol. 6, 338-343. 147. Romani, L. & Howard, D. H. (1995) Curr. Opin. Immunol. 7, 517-523.

PAGE 166

148 148. Rothenberg, R., Woelfel, M., Stoneburner, R., Milberg, J., Parker, R. & Truman, B. (1987) N. Engl. J. Med. 317, 1297-1302. 149. Rubin-Bejerano, I., Fraser, I., Grisafi, P. & Fink, G. R. (2003) Proc. Natl. Acad. Sci. U. S. A 100, 11007-11012. 150. Ruhnke, M., Eigler, A., Tennagen, I., Geiseler, B., Engelmann, E. & Trautmann, M. (1994) J. Clin. Microbiol. 32, 2092-2098. 151. Runge, K. W., Huffaker, T. C. & Robbins, P. W. (1984) J. Biol. Chem. 259, 412-417. 152. Samaranayake, L. P. (1992) Oral Surg. Oral Med. Oral Pathol. 73, 171-180. 153. Saporito-Irwin, S. M., Birse, C. E., Sypherd, P. S. & Fonzi, W. A. (1995) Mol. Cell Biol. 15, 601-613. 154. Schaller, M., Schafer, W., Korting, H. C. & Hube, B. (1998) Mol. Microbiol. 29, 605-615. 155. Schwan, W. R., Wetzel, K. J., Gomez, T. S., Stiles, M. A., Beitlich, B. D. & Grunwald, S. (2004) Microbiology 150, 1055-1061. 156. Segal, E. (1987) Crit Rev. Microbiol. 14, 229-271. 157. Selmecki, A., Bergmann, S. & Berman, J. (2005) Mol. Microbiol. 55, 1553-1565. 158. Sentandreu, M., Elorza, M. V., Valentin, E., Sentandreu, R. & Gozalbo, D. (1995) J. Med. Vet. Mycol. 33, 105-111. 159. Shepard, K. A., Gerber, A. P., Jambhekar, A., Takizawa, P. A., Brown, P. O., Herschlag, D., DeRisi, J. L. & Vale, R. D. (2003) Proc. Natl. Acad. Sci. U. S. A 100, 11429-11434. 160. Shepherd, M. G. (1987) Crit Rev. Microbiol. 15, 7-25. 161. Sim, K. L. & Creamer, T. P. (2004) Proteins 54, 629-638. 162. Slutsky, B., Staebell, M., Anderson, J., Risen, L., Pfaller, M. & Soll, D. R. (1987) J. Bacteriol. 169, 189-197. 163. Sobel, J. D. (1989) Curr. Top. Med. Mycol. 3, 86-108. 164. Soll, D. R. (2004) Bioessays 26, 10-20. 165. Southard, S. B., Specht, C. A., Mishra, C., Chen-Weiner, J. & Robbins, P. W. (1999) J. Bacteriol. 181, 7439-7448.

PAGE 167

149 166. Staab, J. F., Bradway, S. D., Fidel, P. L. & Sundstrom, P. (1999) Science 283, 1535-1538. 167. Staib, P., Kretschmar, M., Nichterlein, T., Kohler, G., Michel, S., Hof, H., Hacker, J. & Morschhauser, J. (1999) Mol. Microbiol. 32, 533-546. 168. Sundstrom, P. (1999) Curr. Opin. Microbiol. 2, 353-357. 169. Sundstrom, P. (2002) Cell Microbiol. 4, 461-469. 170. Sundstrom, P., Balish, E. & Allen, C. M. (2002) J. Infect. Dis. 185, 521-530. 171. Suzuki, T., Kobayashi, I., Kanbe, T. & Tanaka, K. (1989) J. Gen. Microbiol. 135, 425-434. 172. Swoboda, R. K., Bertram, G., Hollander, H., Greenspan, D., Greenspan, J. S., Gow, N. A., Gooday, G. W. & Brown, A. J. (1993) Infect. Immun. 61, 4263-4271. 173. Tanaka, K. (1997) Nagoya J. Med. Sci. 60, 1-14. 174. Timpel, C., Strahl-Bolsinger, S., Ziegelbauer, K. & Ernst, J. F. (1998) J. Biol. Chem. 273, 20837-20846. 175. Timpel, C., Zink, S., Strahl-Bolsinger, S., Schroppel, K. & Ernst, J. (2000) J. Bacteriol. 182, 3063-3071. 176. Torosantucci, A., Romagnoli, G., Chiani, P., Stringaro, A., Crateri, P., Mariotti, S., Teloni, R., Arancia, G., Cassone, A. & Nisini, R. (2004) Infect. Immun. 72, 833-843. 177. Tsuchimori, N., Sharkey, L. L., Fonzi, W. A., French, S. W., Edwards, J. E., Jr. & Filler, S. G. (2000) Infect. Immun. 68, 1997-2002. 178. Tylenda, C. A., Larsen, J., Yeh, C. K., Lane, H. C. & Fox, P. C. (1989) J. Oral Pathol. Med. 18, 520-524. 179. Tzung, K. W., Williams, R. M., Scherer, S., Federspiel, N., Jones, T., Hansen, N., Bivolarevic, V., Huizar, L., Komp, C., Surzycki, R. et al. (2001) Proc. Natl. Acad. Sci. U. S. A 98, 3249-3253. 180. van Burik, J. A. & Magee, P. T. (2001) Annu. Rev. Microbiol. 55, 743-772. 181. Vargas, K. G. & Joly, S. (2002) J. Clin. Microbiol. 40, 341-350. 182. Verstrepen, K. J., Reynolds, T. B. & Fink, G. R. (2004) Nat. Rev. Microbiol. 2, 533-540. 183. Wenzel, R. P. (1995) Clin. Infect. Dis. 20, 1531-1534.

PAGE 168

150 184. Wray, D., Felix, D. H. & Cumming, C. G. (1990) Br. Dent. J. 168, 326-329. 185. Zhao, R., Lockhart, S. R., Daniels, K. & Soll, D. R. (2002) Eukaryot. Cell 1, 353-365. 186. Zhao, X. J., Newsome, J. T. & Cihlar, R. L. (1998) Microb. Pathog. 25, 121-129. 187. Zlotnik, H., Fernandez, M. P., Bowers, B. & Cabib, E. (1984) J. Bacteriol. 159, 1018-1026.

PAGE 169

BIOGRAPHICAL SKETCH Mary Ann Checkley was born in Lima, Peru, on March 19th, 1976 to parents Jorge and Celina Checkley. Mary Ann attended college at Ohio Wesleyan University and graduated with a Bachelor of Arts in zoology (genetics) and pre-med in 1998. During her summer vacations Mary Ann would volunteer to work in diverse laboratories of microbiology. She has worked for Dr. Robert Gilman at the Universidad Cayetano Heredia in Lima Peru with Vibrio cholerae, for Drs. Charles Sterling and Ynes Ortega in University of Arizona in Tucson with Cryptosporidium parvum and Cyclospora cayetanesis, and for Dr. Fred Koster in the University of New Mexico in Albuquerque with Hantaa virus. Work in these laboratories along with microbiology classes taken in college inspired her to pursue a Ph.D. degree in microbiology. She entered the University of Florida, College of Medicine Interdisciplinary Program in Biomedical Sciences in the fall of 1998. Mary Ann completed her doctoral dissertation in Microbiology and Immunology under the mentorship of Drs. M. Hong Nguyen and Alfred Lewin in the spring of 2005. Mary Ann plans to pursue a postdoctoral fellowship in infectious disease research. 151