Embed Size (px)
Citation preview
GENE EXPRESSION STUDIES IN CANDIDA ALBICANS
by
PRIYA UPPULURI, M.S.
A DISSERTATION
IN
MEDICAL MICROBIOLOGY
Submitted to the Graduate Faculty of
Texas Tech University Health Sciences Center in Partial Fulfillment of the Requirements
for the Degree of
DOCTOR OF PHILOSOPHY
Advisory Committee
LaJean Chaffin (Chairperson) Abdul Hamood Daniel Hardy
Michael San Francisco Brandt Schneider
Accepted
Roderick Nairn, Ph.D. Dean of the Graduate School of Biomedical Sciences
Texas Tech University Health Sciences Center
December, 2006
Copyright 2006, Priya Uppuluri
ii
ACKNOWLEDGEMENTS
Having reached this significant milestone of completing my doctorate degree,
there are many people I would like to thank. The help and support of these people helped
make my 5 year journey uneventful and enjoyable.
I begin by thanking my father Dr. Autar K. Miskeen, Ph.D (Microbiology). In my
life there has been no one as loving, encouraging and humorous as him. I have learnt to
appreciate the beauty, necessity, and application of medical microbiology from him. For
all your endless love, blessings, prayers, financial and emotional support, and instilling
the ‘I – know – I – can’ optimism in me dear papa, I dedicate this dissertation to you.
My lonely existence in an alien land found a soul mate, and I changed my name.
Nagesh Uppuluri, my husband – an epitome of an ideal man any woman could ask for.
His involvement with every aspect of my life makes my every day so much easier. You
had a hand (and a brain) in many steps during my PhD, and I thank you for being my
rudder so my ship sailed to the shore.
As an international student, what I always missed the most was my mothers
cooking, cleaning and washing! – activities I took for granted all my life. I thank you
mummy for your loving ways and instilling in me, an independent streak.
I would next like to thank my brother Puneet who always lifted my spirits on the
phone, and assured me that ‘they were not having too much fun without me’. Yes! That
did make me feel better bro.
I never knew parents-in-law could be an excellent asset to life, until I met mine. It
makes me happy to no end, to know of their pride for me having achieved my goal. I am
grateful for their kind words, total support, unconditional love and their enthusiasm for
everything that I do.
Then there is my extended family, my cousins, my aunts and uncles and my grand
parents. I am what I am because of you all. You molded my personality, and my life is
blessed because I have you.
iii
I consider myself extremely lucky to have had Dr. LaJean Chaffin as my mentor
for my doctorate research. Four years of her guidance taught me commitment, good work
ethics, and maintenance of high standards in my research work. Her hands off, no-
pressure approach helped my imagination fly and design my own experiments. Her lab
has been bountiful for me both financially and career-wise, and I thank her for her
thoughtful nature, patient disposure and her presence for whenever I needed her. I pray I
get a boss just like her in whatever job I do in the future.
Next, I would like to thank my excellent committee members, Dr Daniel Hardy,
Dr. Brandt Schneider, Dr. Michael Sanfrancisco and Dr. Abdul Hamood. They never
were any hassle to handle, always gave effective and intelligent suggestions for my
research, let me use their labs for equipments and software, and let me graduate on time!!
I would especially like to thank Dr. Abdul Hamood. A 10 month rotation in his lab taught
me the basics of molecular biology techniques. His hardworking nature and passion for
science was infectious, and I hope I never recover!
My acknowledgements will not be complete if I do not mention my lab members,
Dr. Bhaskarjyoti Sarmah, Dr. Palani Perumal, Satish Mekala and Dr. Gabriel Nkwanyuo.
I thank you for letting me get involved in your projects, for effective discussions and for
letting me borrow your tip boxes and eppendorf tubes! You guys certainly were a joy to
work with.
Finally, I thank the faculty members of my department, who made the decision to
accept me as a graduate student; who were always warm, kind and helpful. I also thank
my friends and fellow graduate students Ganesh Shankarling, Janet Dertein, Shyla
Narasimhachar, Dr. Nancy Carty, Dr. Andy Schaber, Dr. Jennifer Gaines, Dr. Revathi
Govind, Colby Layton, Matt Fogle, Uma Thippeswmy, Arkadi – all of whom had a large
part to play in the smooth running of my projects.
My life is under construction, and I thank my architect for being kind and
generous.
iv
TABLE OF CONTENTS ACKNOWLEDGEMENTS ii ABSTRACT viii LIST OF TABLES x LIST OF FIGURES xi CHAPTER I. INTRODUCTION Candida albicans the pathogen 1 C. albicans polymorphism 2 C. albicans biofilm 3 C. albicans growth and stationary phase 4
C. albicans growth and quorum sensing 6
Conclusions 7
II. ANALYSIS OF RNAs OF VARIOUS SIZES FROM STATIONARY
PHASE PLANKTONIC YEAST CELLS OF CANDIDA ALBICANS Abstract 9 Introduction 9 Methods 11 Results 14 Isolation of RNA 14 Bias in mRNA extraction 17 Discussion 20
v
III. DEFINING CANDIDA ALBICANS STATIONARY PHASE BY
CELLULAR AND DNA REPLICATION, GENE EXPRESSION
AND REGULATION Abstract 25 Introduction 26 Methods 28 Results and Discussion 31 Growth profiles of C. albicans cells during exponential through 11 days 31 Monitoring diauxic shift in C. albicans 34 Measuring the ethanol content in the C. albicans growth medium 34 mRNA profile over the time course of growth 35 Defining C. albicans growth phases 36 Overview of changes in global gene expression at different phases in the growth curve 37 Validation of microarray gene expression by RT-RTPCR 38 Alteration of gene expression during growth and stationary phase 41 Cluster I: C. albicans growth and proliferation regulating genes Cluster III: DNA repair, stress resistance and aging Cluster III: Gluconeogenesis and antagonists of TOR pathway Clusters IV and V: RNAses and proteases Clusters IV and V: Mannoproteins and other cell wall proteins Clusters IV and V: Drug resistance and virulence genes Screening of C. albicans transcription factor and cell wall mutants 49
vi
IV. CANDIDA ALBICANS SNO1 AND SNZ1 EXPRESSED IN STATIONARY
PHASE PLANKTONIC YEAST CELLS AND BASE OF BIOFILM
Abstract 54
Introduction 55
Methods 56
Results 61
Viability of stationary phase organisms 61
Expression of SNZ1 and SNO1 during planktonic growth 62
Biofilm formation and gene expression 62
Protein localization of Snz1p-YFP and Sno1p-YFP in planktonic cells 64
Protein expression in planktonic cells 66
Protein expression in biofilm organisms 67
Discussion 68
V. EFFECT OF FARNESOL AND CONDITIONED MEDIUM ON CANDIDA
ALBICANS GENE EXPRESSION AND YEAST GROWTH
Abstract 73
Introduction 73
Methods 75
Results and Discussion 78
Effect of farnesol and CM on germ tube induction 78
Alteration of gene expression in response to farnesol and CM 79
vii
Activities and pathways affected by farnesol and CM addition 81
Gene expression in the presence of CM 82
Gene expression in the presence of 40 µM farnesol 83
Effect of farnesol on C. albicans growth 86
Rescue from farnesol mediated delay in yeast growth resumption 90
VI. CONCLUDING REMARKS 94
REFERENCES 100
viii
ABSTRACT
Candida albicans is part of the normal flora of the human oral, gastrointestinal,
vaginal and cutaneous surfaces. However, in the compromised host the organism can
cause infection of those surfaces as well as systemic disease. C. albicans can also form
biofilms on host surfaces as well as abiotic device surfaces such as dentures and
catheters. Phenotypic drug resistance of C. albicans biofilms poses a therapeutic
dilemma. Stationary phase C. albicans cells are phenotypically more resistant to
antifungals. Identifying if cells in a biofilm reach stationary phase could give some
insight into the mechanism of biofilm resistance. To test this possibility we first
characterized the C. albicans stationary phase and established criteria by which stationary
phase could be defined.
Planktonic stationary phase cells in vitro are known to survive for long periods of
time in media composed of metabolites excreted by the cells during growth. This
conditioned medium also contains quorum sensing molecules that confer various
properties to the fungus. However, the global effect on gene expression of either the
conditioned medium or any of its individual quorum sensing molecule is not well studied.
We studied the mechanism by which the conditioned medium and a quorum sensing
molecule affected C. albicans biology.
To study C. albicans stationary phase, we used a variety of descriptive techniques
and cDNA microarray technology. We have defined for the first time, the different
growth phases of C. albicans and determined the genes and processes important for entry
into stationary phase. We have also identified genes important for the survival of cells in
stationary phase. Additionally, by establishing an improved extraction protocol that
yields RNA of all classes and sizes we have overcome the difficulty associated with
extracting RNA from stationary phase cells. Using stationary phase gene markers we
demonstrated that even after prolonged incubation, only 40% of the founder cells of a C.
albicans biofilm reached stationary phase. The results of this study will expand our
ix
existing knowledge of C. albicans stationary phase, and serve as a foundation for more
systematic and unbiased studies in C. albicans research.
x
LIST OF TABLES
2.1 Primers used for PCR
3.1 Primers used for PCR
3.2. Correlation between microarray gene expression and reduction in viability of the
mutant strains
4.1 Primers used for PCR
5.1 Primers used for PCR
5.2. Genes differentially expressed in the presence of farnesol or CM compared to un-
supplemented medium
5.3 Differences in cell sizes of farnesol treated and untreated cells
xi
LIST OF FIGURES
1.1 Schematic representation of different morphologies of C. albicans
1.2 Scanning electron microscopy images of C. albicans biofilm
1.3 Schematic representation of growth phases in C. albicans
2.1 Analysis of RNA from different growth forms
2.2 Analysis of RNA from stationary phase planktonic yeast cells
2.3 Size dependant extraction of mRNA
3.1 Growth curve of cells from culture grown in YPD for extended periods
3.2 Analysis of budding, DNA profiling and cell sizing of cells grown in YPD for
extended periods
3.3 Measurement of glucose and ethanol in the growth medium
3.4 Reduction in C. albicans mRNA abundance
3.5 Cluster analysis of genes differentially expressed in different growth phases of C.
albicans
3.6 RT-RT PCR verification of microarray expressed genes
3.7 Metabolic reprogramming inferred from changes in gene expression during
diauxic shift
3.8 Screening of C. albicans transcription factor and cell wall mutants by drop plate
method
4.1 Schematic representation of construction of the fluorescent construct,
recombination into C. albicans genomic DNA, and verification
4.2 Viability of cells from culture grown in YNB for extended periods.
4.3 Expression of SNZ1 and SNO1
4.4 Expression of Snz1p–YFP by yeast cells and pseudohyphae
4.5 Localization of Sno1p–YFP in yeast cells
4.6 Expression of Snz1p–YFP and Sno1p–YFP during progression into and exit from
the stationary phase
4.7 Expression of Sno1p–YFP in different layers of a 6-day-old biofilm
xii
5.1 Venn diagram of the number of upregulated genes that are unique to, or common
between the three conditions, Farnesol (F), CM (C) and control medium (M)
5.2 RT-RTPCR verification of genes differentially expressed in Farnesol (F) group
and CM (C) group, obtained by microarray analysis
5.3 Effect of farnesol on growth retardation of C. albicans cells
5.4 Flow cytometry analysis of cells grown for 8 hours in untreated YNB medium
(control), in YNB with 300 µM farnesol and in YNB with 300 µM farnesol and
50 µM OAG
5.5 Differential expression of genes involved in phosphatidylinositol type signaling
pathway
1
CHAPTER I
INTRODUCTION
“C. albicans has an identity crisis; it thinks it’s a part of the human body”
- Carol Kumamoto, Professor, Tufts University
Candida albicans the pathogen
Classification
Kingdom: Fungi
Phylum: Ascomycota
Class: Saccharomycetes
Genus: Candida
Species: C. albicans
Candida albicans, a diploid asexual fungus is a part of the normal flora of the human
oral, gastrointestinal, vaginal and cutaneous surfaces. In healthy individuals, C. albicans
normally does not cause disease. However, when the balance of the normal flora is
altered, during antibiotic or hormonal therapy, or in conditions when the skin is exposed
to moisture for prolonged periods of time, C. albicans can cause painful cutaneous or
subcutaneous infections such as, vaginitis, oral thrush, diaper rash, conjunctivitis, or
infections of the nail and rectum. In immunocompromised individuals, such as
immunosuppressed patients undergoing cancer chemotherapy, C. albicans can be
responsible for life threatening diseases only when it enters the blood stream. It is then
capable of affecting almost any part of the body and causing hepatosplenic abscesses,
myocarditis, central nervous system or pulmonary infections. C. albicans infections are a
major public health concern. In the USA, Candida is the fourth most common cause of
nosocomial infections, with annual Medicare costs reported to exceed one billion dollars.
2
Also in the USA alone, there are approximately 10,000 deaths a year due to Candida
infections (Sudbery, Gow et al. 2004). The advances of modern medicine have led to
larger populations of compromised patients susceptible to candidiasis, increasing the
importance of C. albicans as a pathogen and providing impetus for the detailed study of
C. albicans biology.
C. albicans polymorphism
A striking feature of C. albicans biology is its ability to grow in a variety of
morphological forms. Unicellular budding yeast can reversibly switch to form true
hyphae with parallel-sided walls. In between these two extremes, the fungus can exhibit
another growth form termed as pseudohyphae, in which the daughter bud elongates but
fails to separate, and remains attached to the mother cell (Fig 1.1).
Figure 1.1. Schematic representation of different morphologies of C. albicans.
The ability to switch between yeast, hyphal and pseudohyphal morphologies is often
considered to be necessary for virulence. Both hyphae and pseudohyphae are invasive
(i.e. they invade the agar substratum when they grow in the laboratory). It is speculated
that this property could promote tissue penetration during the early stages of infection,
3
whereas the yeast form might be more suited for dissemination in the bloodstream. The
filamentous forms may also be important for colonization of organs, such as the kidney
(Gow, Brown et al. 2002). Additionally, morphogenesis plays a pivotal role in C.
albicans biofilm development.
C. albicans biofilms
Biofilms: Structured microbial communities in which the cells bind tightly to a
surface and become embedded in a matrix of extracellular polymeric substances
produced by these cells.
C. albicans biofilms are structurally well organized communities of yeast,
pseudohyphal, and hyphal cells, enclosed in an extracellular matrix comprising
polysaccharide and protein (Figure 1.2A, B). Dental plaque is a natural example of a
biofilm formed by C. albicans along with other oral bacteria. Candida albicans can also
populate, penetrate and form a biofilm on indwelling medical devices such as dental
implants, catheters, heart valves, ocular lenses, artificial joints, and central nervous
system shunts (Donlan 2001; Douglas 2003).
A B
Figure 1.2. Scanning electron microscopy images of C. albicans biofilm: Hyphae (A) and
one yeast cell (B) covered with extracellular matrix.
Approximately 10% of the infections linked vascular/urinary catheters and heart valves
are due to Candida species and 40 % of patients with intravenous catheters develop acute
4
fungaemia (Kumamoto and Vinces 2005). Finally, a biofilm provides C. albicans
protection against some of the major antifungal drugs such as fluconazole, nystatin,
amphotericin B, and chlorhexidine (Kuhn and Ghannoum 2004). Recent data indicate
that resistance is phase-specific and multifactorial, involving efflux pumps and sterol
synthesis (at early and mature biofilm phases, respectively) (Kuhn and Ghannoum 2004).
Another explanation for resistance to antifungals could be attributed to the structural
complexity of the biofilm that may create a gradient of environmental conditions in which
the C. albicans cells enter distinct physiological states. One such state may be equivalent
to stationary phase. C. albicans cells in stationary phase adhere better to both biotic and
abiotic surfaces; and proper adherence is the first step to biofilm formation. Stationary
phase is also the prime reason for phenotypic drug resistance in planktonic C. albicans
cells. Hence, investigating the C. albicans growth phase in a biofilm may help understand
the properties of the cells in the biofilm.
C. albicans growth and stationary phase
On inoculation into fresh medium in vitro, C. albicans undergoes four major
growth phases (Figure 1.3), 1. Lag phase, a phase where the yeast cells sense their
environment, and take time to adapt to it before doubling, 2. Logarithmic phase, which
immediately follows the lag phase, where cells start growing exponentially and actively
metabolize nutrients, 3. Stationary phase, when the cells exhaust nutrients and stop
multiplying. The cells in this phase can survive for long periods of time without
additional nutrients, while completely retaining their capacity to bud if and when
inoculated into fresh medium, 4. Death – Aging, as well as accumulation of toxic
metabolites in the medium, finally pushes the cells into apoptosis, or death.
5
Figure 1.3: Schematic representation of growth phases in C. albicans: lag phase (A),
exponential phase (B), stationary phase (C) and death (D).
Most C. albicans research has been carried out with exponentially growing cells,
and stationary phase has been poorly studied in this yeast. In fact, not even the timing of
entry into stationary phase has been clearly defined. In some studies, an overnight, 24h or
48h grown C. albicans culture is considered to be in stationary phase (Masuoka and
Hazen 1999; Westwater, Balish et al. 2005; Zhao, Daniels et al. 2005) while other studies
report stationary phase to start much later in culture, i.e. between 3d and 8d (Cassone,
Kerridge et al. 1979; Dudani and Prasad 1985; Lyons and White 2000). Lack of study of
the C. albicans stationary phase is surprising, given the fact that important properties are
acquired by the yeast in this phase. First, only stationary phase cells can generate an
extensive production of true hyphae in C. albicans, an important tool for invasion
(Westwater, Balish et al. 2005). Secondly, stationary phase Candida albicans cells adhere
better both in vitro (to polystyrene and acrylic) (McCourtie and Douglas 1981), and in
vivo, to all major organs of mice compared to exponential phase cells (Cutler, Brawner et
al. 1990; Granger, Flenniken et al. 2005). Also the cell walls of stationary phase C.
albicans cells become 60% thicker and less porous than cells from any other phase – the
main cause for phenotypic drug resistance. Proper phenotypic characterization of C.
albicans stationary phase is extremely important to understand correctly, the basic
biology of this pathogenic yeast. Also, studying this phase at the molecular level, e.g.
B
D
A
C
6
discovering the genes involved in the entry and maintenance of stationary phase, will
subsequently help unravel the mechanism which gives C. albicans important properties
(antifungal resistance, virulence, immunogenicity, stress resistance etc). Interference with
any of these properties at the genetic level could help in the attenuation of C. albicans
virulence.
C. albicans growth and quorum sensing
By the time C. albicans cells reach stationary phase, they have excreted in the
medium metabolites, some of which may have signaling properties, also known as
quorum sensing molecules. Such medium has been referred to as conditioned medium.
Conditioned medium is known both to stimulate and inhibit germ tube formation (Chen,
Fujita et al. 2004; Lopez-Ribot 2005). Farnesol and tyrosol, two quorum sensing
molecules purified from conditioned medium are known to mediate such sometimes
contradictory activities. While tyrosol induces hyphae in permissive conditions and
reduces lag phase in C. albicans (Chen, Fujita et al. 2004), farnesol prevents yeast cell to
hyphal transition in similar conditions and in turn inhibits biofilm formation (Lopez-
Ribot 2005; Nickerson, Atkin et al. 2006). Thus, the metabolites in the CM have various
effects on cells depending upon the environmental conditions and suggesting a complex
cell response. The intensity of the quorum sensing effect probably is the highest during
the stationary phase, when concentration of metabolites in the medium is greatest.
Indeed, conditioned medium recovered from stationary phase cells has been shown to
protect C. albicans against oxidative stress. This resistance was mediated partially due to
the presence of farnesol in the conditioned medium (Westwater, Balish et al. 2005).
Not much is known about what other processes conditioned medium, or its
component farnesol effect in C. albicans. Also, the effect of these treatments at the
transcriptional level is not yet studied. Identification of the genes or processes affected by
these molecules may help reveal the incomplete information behind their mode of action,
specially relating to their effect on C. albicans morphology. A part of this thesis was to
7
study the global gene expression profiling of C. albicans cells treated with farnesol as
well as the conditioned medium.
Conclusions
Most of the time C. albicans survives peacefully as a commensal in the healthy
immunocompetent host, rarely, if ever, causing any infections. Only in conditions that
perturb the normal flora or in immunocompromised hosts do these yeast cells cause
morbidity and/or severe systemic diseases. So, as a commensal, what is the growth state
of C. albicans in the human body? The answer to this question is not yet known. Perhaps,
in organs such as the gut, due to anaerobic conditions and competition for nutrients with
hundreds of different species of bacteria, C. albicans survives in stationary phase; while
in the oral cavity, due to the constant influx of nutrients, it never reaches the stationary
phase. It could be possible that in many parts of the body, C. albicans cells probably exist
in conditions akin to those of long-term stationary-phase cultures, in which expression of
a wide variety of stress-response genes and alternative metabolic pathways are essential
for survival. Also, the growth state of C. albicans in a biofilm is not known. Early in our
studies of biofilm, we posed the question “Are all the C. albicans cells in a biofilm in the
same physiological state?” To answer this question we hypothesized that some cells
within the biofilm reach a physiological state equivalent to stationary phase in planktonic
organisms. To test this hypothesis we first had to obtain additional characterization of C.
albicans stationary phase and establish a criterion by which stationary phase could be
identified.
The major findings of my Ph.D. study are as follows,
Chapter 2. We found that large molecular weight RNA (both ribosomal and mRNA)
could not be extracted by conventional methods of RNA extraction from stationary phase
C. albicans cells. We optimized a new method of RNA extraction that can yield all
classes and sizes of RNA, especially from late stationary phase cells. We stress that this
method for RNA extraction will improve the quality of research pertaining to C. albicans
stationary phase (Uppuluri, Perumal et al. 2006).
8
Chapter 3. Using a set of descriptive methodologies, such as monitoring growth and
DNA profiles and measurement of carbohydrate and ethanol concentrations in medium,
we delineated the different C. albicans growth phases, specially relating to post-diauxic
and stationary phase. We hope that giving a definite structure to C. albicans stationary
phase physiologically, will reduce the confusion that exists regarding the exact timing of
entry into stationary phase.
Using cDNA microarray technology we monitored the global gene expression
profiles of C. albicans in exponential, diauxic and stationary phase. Notable differences
in gene expression observed between the three growth phases emphasize that diauxic
shift and stationary phase are two distinctly different stages of growth, and should not be
interchangeably used, as done often when studying C. albicans stationary phase.
By screening C. albicans transcription factor and cell wall deletion mutants, we
identified genes important for entry (carbohydrate metabolism, cell wall maintenance)
and maintenance (mitochondrion maintenance, unknown genes) of C. albicans stationary
phase. Additionally, after comparing the microarray results and the mutant screening data
we learned that steady state expression of many genes throughout the growth curve is
important for survival in stationary phase.
Chapter 4. Using fluorescent constructs of two stationary phase genes, we found that only
40% of the bottom-most layer of adhered cells in a C. albicans biofilm reached stationary
phase (Uppuluri, Sarmah et al. 2006). We conclude that biofilm mediated drug resistance
may not be a consequence of presence of stationary phase cells in the biofilm.
Chapter 5. Using cDNA microarrays we studied the differential gene expression of C.
albicans when treated with conditioned medium and a quorum sensing molecule,
farnesol. From this study, we identified pathways that these compounds affect to mediate
their inhibitory effect on C. albicans dimorphic switching. Additionally we found that
farnesol could increase lag phase in C. albicans and force cells to enter a stationary
phase-like unbudded phenotype, which could be relieved by adding a protein kinase C
activator oleyl acetyl glycerol.
9
CHAPTER II
ANALYSIS OF RNAs OF VARIOUS SIZES FROM STATIONARY
PHASE PLANKTONIC YEAST CELLS OF CANDIDA ALBICANS
Abstract
We initiated a comparison of Candida albicans stationary phase gene expression
with other growth states. The widely used hot acid phenol method (HAP) for RNA
extraction did not extract rRNA from late stationary phase cells. The RNA from growing
yeast cells, hyphae and biofilm, was biased towards small sized RNA. The 2:1 ratio
between the two large rRNA bands was rarely obtained. Real time reverse transcriptase
PCR (RT-RTPCR) was used to determine mRNA extraction by several methods for
OXR1, IRA2, RAD50, PNC1, CHS2, having 300 bp to 8 kb coding regions and ACT1,
EFB1, and TDH3, sometimes used as internal standards. Only smaller sized cDNAs were
amplified from some extracts. Crushing cells with glass beads in liquid nitrogen before
RNA extraction by hot phenol method (CGB) yielded an unbiased distribution for rRNA
and mRNA as verified by RT-RTPCR. With the CGB method the large mRNAs,
RAD50, IRA2 and OXR1, were present throughout stationary phase while the CSH2
transcript increased. The ACT1, EFB1 and TDH3 transcripts decreased in stationary
phase, making them unsuitable for standardization. The CGB method yielded high
quality RNA from the various growth conditions and permitted the comparison of
stationary phase transcripts with those of other conditions.
Introduction
Study of the stationary phase in C. albicans is still in its nascent stages. It is,
however, known that there are changes in the ultrastructure of the cell wall of C. albicans
when it enters the stationary phase (Cassone, Kerridge et al. 1979). The yeast cell wall
becomes significantly thicker and less porous than exponential phase cells (Werner-
Washburne, Braun et al. 1993; Mukherjee, Chandra et al. 2003) and confers the property
of phenotypic drug resistance (Gale, Johnson et al. 1980; Suci and Tyler 2003). We are
10
interested in exploring the gene expression pattern in stationary phase C. albicans and
how stationary gene expression relates to expression of the same gene(s) of organisms
under different growth conditions. Isolation of high quality RNA reflecting in vivo
transcriptional profiles of cells in each of the growth conditions is crucial for accurate and
meaningful results. The thick stationary phase cell wall may also be a deterrent for the
extraction of RNA that reflects in vivo profiles.
In this study we used several methods to extract RNA from C. albicans cells in
stationary phase and other physiologic and morphologic states. The hot acid phenol
method (HAP) with or without glass bead vortexing did not yield larger rRNAs from late
stationary phase cells. Further, in other conditions where RNA was extracted, the RNA
was biased toward smaller RNAs (ribosomal and mRNA). In contrast, a method that has
been used for Saccharomyces cerevisiae to extract RNA-protein complexes with intact
RNA (Schultz 1999; Lopez de Heredia and Jansen 2004) involving grinding cells with
glass beads in liquid nitrogen prior to RNA extraction was modified and successfully
applied to C. albicans cells from all cultures. This method yielded an unbiased
representation of RNA populations.
This methodology was used to examine expression of 8 genes during stationary
phase. We found that expression of several large genes could be detected for at least 11
days and in the case of one small gene we observed that expression increased at three
days and then persisted. On the other hand, the three small genes that have sometimes
been used as internal standards for mRNA comparisons were found to decrease during
stationary phase making them unsuitable for standardization across growth conditions
that include stationary phase.
Methods
Organism and culture conditions. C. albicans strain SC5314 was maintained on YPD
(yeast extract 1% w/v, peptone 2% w/v, dextrose 2% w/v) agar plates and transferred to
YNB (yeast nitrogen base medium with amino acids, Difco Laboratories, Detroit,
Michigan) with 50 mM glucose for suspension culture with shaking (180 rpm).
11
Planktonic C. albicans cells were grown at 250C for 1-11 days and collected by
centrifugation. Filament formation was induced by resuspending early stationary phase
(about 2x108 cells/ml) yeast cells at 1x107 cells/ml in the same fresh medium at 370C for
90-120 m with shaking at 180 rpm. Germ tube formation was greater than 90%. Biofilm
was formed by resuspending 250C late exponentially grown cells at 5x107 cells/ml for
incubation at 370C with 9x2x0.1 cm polymethylmethacrylate strips (prepared by Dr.
Thomas McKinney, Baylor College of Dentistry, Dallas, Texas) for 2 h. The strips,
placed in a 50 ml syringe barrel were washed with YNB to remove non-adhered cells and
fresh YNB medium was flowed through the syringe at 50ml/h for 48 h at 370C. Sterile
air was supplied into the medium at 1L/h. The yeast and hyphal cells of the biofilm were
scraped from the support and collected by centrifugation.
RNA extraction. RNA was extracted by the HAP method (Kohrer and Domdey 1991;
Ausubel, Brent et al. 2002). For some experiments, stationary phase yeast cells were first
resuspended in 400 µl SAB buffer (50 mM sodium acetate, 10 mM EDTA, pH 5.2), 400
µl phenol:chloroform:isoamyl alcohol (25:24:1) with 100 µl glass beads and vortexed
vigorously three times for 30 seconds with intervals on ice (Fuge, Braun et al. 1994).
Alternatively, cells were placed in a Mini-BeadBeater (Biospec Products, Bartlesville,
OK) for three cycles of 45 seconds at 30 second intervals. Broken and intact cells were
determined microscopically with a minimum of 1000 cells counted.
Another method employing grinding frozen S. cerevisiae cells before extraction
was modified (Schultz 1999; Lopez de Heredia and Jansen 2004). Cells were harvested
by centrifugation and the pellet was flash frozen with liquid nitrogen and maintained at -
800C. A mortar and pestle were chilled with liquid nitrogen and the cell pellet along with
an equal volume of glass beads was added. Liquid nitrogen was added as needed to
maintain the frozen state of the organism. The mixture was ground until the glass beads
formed a fine powder and then grinding was continued for 5 minutes. For 50-100 µl cell
pellet (approximately 50-100 mg wet weight cells), a mixture 600 �l phenol (pH 4.2), 450
�l SAB buffer, 150 �l chloroform, and 45 �l 20% sodium dodecyl sulfate (SDS) was
12
added to a 2 ml centrifuge tube. The ground cell mixture was transferred with a chilled
spatula to the tube and vortexed vigorously for 30 seconds. The mixture was incubated at
65oC for 20 minutes with vortexing after every 5 minutes and centrifuged (12,000 x g) for
10 m at room temperature. The upper aqueous layer, approximately 500 �l, was re-
extracted with an equal volume of phenol. The aqueous phase was extracted once with
24:1 mix of chlorofom:isoamyl alcohol. The aqueous phase RNA was mixed with 0.1
volume of 3 M sodium acetate, pH 5.2, and 2.5 volumes of 100% ethanol and incubated
at -80°C for 15 minutes. The precipitate was collected by centrifugation (12,000 x g for
15 minutes at 4°C) and the pellet washed with 70% ethanol. The pellet was collected by
centrifugation (12,000 x g for 5 minutes at 4°C), air dried and dissolved in diethyl
pyrocarbonanate (DEPC)-treated water. Aqueous solutions were treated with DEPC and
RNase-free plastic ware was used during isolation.
RNA analysis RNA (5-10 µg) was separated by electrophoresis on 1% agarose
containing 2% formaldehyde, stained with ethidium bromide by standard methods and
visualized (Imgemaster; Amersham Pharmacia Biotech, Piscataway, NJ). Densitometric
measurements were obtained. Peak area calculations were generated using ImageJ
software (Version 1.32e).
Real time reverse transcriptase PCR. Primers were designed for 8 genes ACT1, CHS2,
EFB1, IRA2, OXR1, PNC1, RAD50 and TDH3 and are listed in Table 2.1.
13
Table 2.1 Primers used for PCR. Forward (F) and Reverse (R) primers were used to
amplify a region of the orf from cDNA and the whole orf from DNA or to confirm
absence of DNA from RNA preparation, EFB1 (DNA).
Parameters for primer design are set according to the recommendations of
Applied Biosystems (Foster City, CA). Briefly, the primer sizes were between 20-25b in
length, with Tm of each primer at 58oC. The amplicons were between 90 – 110bp in size.
Total RNA was DNase treated (Ambion, Austin, TX) and purified with the RNeasy kit
(Qiagen Inc., Valencia, CA). The absence of DNA was confirmed with EFB1 (Maneu,
Martinez et al. 2000). cDNA was made using the SuperScript III First Strand Synthesis
System (Invitrogen, Carlsbad, CA) and 1 µg total RNA template. cDNA was diluted 1:4
Gene (systematic name)
Primer Sequence (5’-3’)
ACT1 ACT1F TCATGATGGAGTTGAAAGTGGTTT (orf19.8001) ACT1R AGAGCTCCAGAAGCTTTGTTC ACT1fullF ATGGATGGACCAGATTCGT ACT1fullR TCAAGTTATCACTATTGG CHS2 CHS2F TAATAAATTCCGCAATACGCCTAAC (orf19.737) CHSR TAGTGGCACACATTCTCTTTCATTTT CHS2fullF ATGTTTATATTTTCTTGTTTCA CHS2fullR TCAATGATTATTATAAAAATGGCGGAT EFB1 EFB1F ACGAATTCTTGGCTGACAAATCA (orf19.3838) EFB1R TCATCTTCTTCAACAGCAGCTTGT EFB1fullF ATGAGTGACAAAGAAGATTTAAA EFB1fullR TTACAATTTTTGCATAGCAGC IRA2 IRA2F CCTTGATACAAAGTCGAGCTTAGGA (orf19.5219) IRA2R TAGGAGCTGTTGGCCAGGTATT IRA2fullF AATGGAGCAGAAGAGTTATTGTCGGACATT IRA2fullR TTAATCCTCCAATTTCGACCCACTGAT OXR1 OXR1F TCGTCACATTCTAGTGTTTCTAGTCTG (orf19.243) OXR1R TAGTAATCGATGATGAGTTGATTCTT OXR1fullF ATGTCATTTCTTTTTAGAAGATCT OXR1fullR TTACTCAAAAGTACCTATT PNC1 PNC1F AACTTGACCCGAAAACGAATCA (orf19.6684) PNC1R AGCTCCCTTGGTGCCTTGTAC PNC1fullF ATGAAGAAAACAGCATTAATAGT PNC1fullR TCAATTCAGTATGATGTACCCCCA RAD50 RAD50F CAGGGACATTGCCTCCAAAT (orf19.1648) RAD50R CAGTTACAGCAGTTCGAGAGCTTAAG RAD50fullF ATGATCCATATTGAAAAACTATTTT RAD50fullR TTAGCCTTGAATTCTACCAAT TDH3 TDH3F AGGACTGGAGAGGTGGTAGAACTG (orf19.6814) TDH3R AATAACCTTACCAACGGCTTTAGC TDH3fullF ATGGCTATTAAAATTGGTATTAA TDH3fullR TCAAGCAGAAGCTTTAGCAACGT EFB1 (DNA) EF1-BF ACGAATTCTTGGCTGACAAATCA (orf19.3838) EF1-BR TCATCTTCTTCAACAGCAGCTTGT
14
with RNase-free water. Analysis of transcript was carried out in 25 µl using SYBR Green
PCR Master Mix (Applied Biosystems) in ABI Prism 7700 Sequence Detection System
(Applied Biosystems) for 40 cycles (thermal cycling conditions: Initial steps of 50o C, 2
min and 95o C, 10 min; and then, 40 cycles of 95o C,15 sec; 60o C, 1 min). The Ct values
for each experiment were recorded. To quantify transcripts, a standard curve was
constructed using DNA of each gene as standard. For this, genomic DNA was isolated
from C. albicans SC5314 strain following standard protocol (Adams, Gottschling et al.
1997) and each ORF was PCR amplified using gene specific primers that amplify the
complete ORF. PCR products were separated in a 1.0% agarose gel, DNA eluted from
gel and quantitated spectrophotometrically. RT-RTPCR reaction for each gene was set
up using serial 10 fold dilutions of the amplified ORF as the DNA template. Three
independent biological and technical replicates were used for normalization. All
replicates gave significantly similar amplification values as analyzed using ANOVA
(p<0.05).
Results
Isolation of RNA. RNA was extracted from planktonic yeast cells, hyphae and biofilm by
the commonly used HAP method. The 25S and 18S rRNAs are processed from a
precursor and the larger 25S species should be more intense than the smaller 18S species.
Analysis of the RNA showed intact rRNA but a biased distribution of the extracted RNA
molecules, as evident from the relative intensities of visible ribosomal RNA bands (Fig.
2.1A) and the areas of the corresponding peaks using a densitometer scan of the gel (Fig.
2.1B).
We also applied the method to stationary phase organisms. In C. albicans
stationary phase is reached at about day 5 with changes beginning day 3 (Uppuluri et al,
manuscript in preparation). Intact RNA was isolated from cells grown for 24 hours and 3
day old early stationary phase cells, although as observed previously there was a bias
towards extraction of small RNAs. Ribosomal RNA was not observed in extracts from 5,
7, and 11 day stationary phase cells using the HAP method (Fig 2.2 A-C). We
15
considered that releasing RNA from the cells along with extraction would provide intact
RNA of all classes. HAP with glass beads vortexing was used with 5, 7, and 11 day old
cells without improvement in RNA extraction (Fig 2.2 A-C).
We tried another more vigorous method of cell breakage. When stationary phase
cells (5 days or older) were homogenized with glass beads in a bead beater, the cell
breakage was poor with approximately 50% of the cells broken. The RNA extracted still
showed a bias towards smaller RNAs (Fig 2.2 F, G).
A different method of breakage has been successfully applied to frozen S.
cerevisiae cells to obtain RNA-protein complexes (Schultz 1999; Lopez de Heredia and
Jansen 2004). We applied a variant of these methods (called crushed glass beads, [CGB])
to extraction of RNA from the various physiological and morphological states.
Organisms were maintained frozen after collection and during grinding with glass beads
before addition of a phenol buffer to separate RNA and protein. Using the CGB method,
good quality RNA was extracted from all the three conditions of growth – yeast, hyphae
and biofilm. Moreover, the intensity of the large 25S rRNA band was greater than the
smaller 18S moiety (Fig 2.1 B, D). Unlike the HAP method, with the CGB method two
rRNA bands and one band comprising 5S rRNA and tRNA were observed when RNA
was extracted from the late stationary phase cells (days 5 - 11) (Fig 2.2 D, E ). The 25S
rRNA band was also more intense than the 18S rRNA band in these extracts from
stationary phase cells.
Others have speculated that an increase in RNases that co-purify with the
stationary phase RNA could result in the loss of RNA by degradation (Werner-
Washburne, Braun et al. 1993). To assess this possibility, RNA was extracted using
RNAwiz, a reagent known to contain chaotropic denaturants that inhibit RNases and
stabilize RNA. When RNA was extracted using the RNAwiz reagent without bead
beating, the two large ribosomal bands were not observed (data not shown). Only the
third (5S rRNA and tRNA) band was obtained. When bead beating was added to the
protocol, rRNA was obtained but the bias towards the smaller species was still observed.
16
Figure 2.1 - Analysis of RNA from different growth forms. Samples of RNA extracted
from yeast (Y), biofilm (B), and hyphae (H) organisms by the HAP method (A, B) and
the CGB method (C, D) were separated by electrophoresis. An image (A, C) and
densitometer scan (B, D) were obtained. Scans B and D are representative scans for A
and C respectively. A bias was observed towards extraction of small molecular weight
RNA by using the HAP method. Also the 2:1 ratio between the two large ribosomal
bands was not obtained. These shortcomings were eliminated by using the CGB method.
17
Figure 2.2 – Analysis of RNA from stationary phase planktonic yeast cells. RNA from
different growth phases, 1 day and 3 through 11 days was extracted by the HAP beads
vortexing (A), by the CGB (D), and by HAP with bead beating (F) method and was
separated by electrophoresis. An image (A, D and F) and representative densitometer
scan for HAP extracted RNA from 1d (label 1) and 3 day (B), 5 to 11 days (C), as well
as, CGB extracted RNA (E) and HAP with bead beating method (G) for all the time
points were obtained. The crushed glass beads method was found to be a better method of
RNA extraction than the HAP method.
Bias in mRNA extraction. Since mRNAs can be both larger and smaller than rRNAs, we
determined whether the bias towards smaller RNAs extended to mRNAs. RNA was
extracted from 1 day old yeast cells as well as from the stationary phase (3, 5, 7, and 11
day old) cells, by both the HAP method as well as the CGB method. RT-RTPCR was
performed using primers for 5 genes having coding regions ranging from approximately,
300bp to 8Kb. These genes were CHS2 (339bp), PNC1 (360bp), OXR1 (1.038kb),
RAD50 (4kb) and IRA2 (7.9kb). We also included three genes that are used in C. albicans
or other systems for internal reference: ACT1 (372bp), EFB1 (693bp), and TDH3 (335bp)
(glyceraldehyde phosphate dehydrogenase).
RT-RTPCR results revealed that only the small molecular weight cDNAs, PNC1
and CHS2 were amplified from all the time points when extracted by the HAP method
18
(Fig 2.3B). CHS2 increased expression between 1 and 3 days and remained elevated
through 11 days. The medium to large sized cDNAs (OXR1, IRA2 and RAD50) could be
amplified from RNA extracted from 1, 3 and 5 day old cells and not from the 7 and 11
day old yeast cells (Fig 2.3 A). However, all the 5 cDNAs were efficiently amplified
from cDNA obtained from cells at all growth stages by using RNA extracted by the CGB
method (Fig 2.3 A,B). Although improved extraction was most dramatic for the large
RNAs from late stationary phase cells, extraction of the small RNAs also improved with
CGB extraction. With the CGB method it was apparent that expression of RAD50, IRA2
and OXR1 had only a small decline in expression during stationary phase rather than a
large decline and cessation. Additionally, we found that all the three housekeeping
genes, ACT1, EFB1 and TDH3, due to their small coding sequence sizes, could be
efficiently extracted by both the RNA extraction methods. However, regardless of the
method used for RNA extraction, all the three showed significantly altered gene
expression in stationary phase (Fig 2.3 C). Again, the CGB method did improve the
extraction of mRNA of the house keeping genes when compared to the HAP method.
19
Figure 2.3 – Size dependant extraction of mRNA. A. Comparison of the ability of the
two RNA extraction methods, HAP and CGB, in recovery of medium sized message,
OXR1 to large sized messages, IRA2 and RAD50. B. Comparison of the recovery of
small messages PNC1 and CHS2 by the two methods. C. Comparison of the two RNA
extraction methods for the recovery of 3 housekeeping genes, TDH3, ACT1 and EFB1.
Copies are shown for 5 ng mRNA.
20
Discussion
In the past decade, microarray technology that analyzes the relative abundance
profile of mRNA molecules expressed in response to given treatments has become a
major tool for high-throughput comprehensive analysis of gene expression. Whether
mRNA populations are assessed for a single condition or compared between conditions,
it is essential that the mRNA used be representative of the in vivo population. Therefore,
serious considerations must be given to the application of the proper RNA extraction
procedure to minimize errors (Baldi and Hatfield 2002). Fungal cells, including the
yeasts S. cerevisiae and C. albicans, are surrounded by a rigid cell wall that varies in
thickness and composition depending upon growth conditions and that may be a barrier
to extraction of cellular contents. Treating cells with a reducing agent, e.g. dithiothreitol,
followed by a zymoylase or lyticase treatment can effectively remove the cell wall barrier
and generate intact spheroplasts which are readily lysed (Hudspeth, Shumard et al. 1980).
However, such approach is not appropriate because of possible alteration of the
transcriptional profile of cells during the required incubation at 37°C for one hour as has
been confirmed in S. cerevisiae (Hauser, Vingron et al. 1998). In S. cerevisiae, CLN3,
BCK2, and CDC28, also exhibit significant rapid induction responses (within 10 minutes)
upon transferring cells from a minimal medium to glucose containing medium
(Newcomb, Diderich et al. 2003). In C. albicans, 2-fold induction has been reported
within 10 minutes for heat shock proteins HSP12, HSP70, HSP78, and HSP104 as a
result of exposing cells to rapid temperature shifts from 23 to 37°C (Enjalbert, Nantel et
al. 2003). Similarly, hyperosmotic shocks and oxidative stress resulted in the rapid
transient induction of several other genes (Enjalbert, Nantel et al. 2003). Such responses
in gene expression to rapid changes in the environment present a strong argument against
washing cells much less treatment to remove the cell wall prior to RNA extraction. The
mechanical disruption of cell walls using glass beads and a homogenizer may also alter
gene expression due to fluctuations in temperature during the process and change in
nutrients.
21
One of the commonly used protocols to extract RNA from yeast cells is the HAP
method combined with various methods for cell breakage (McEntee and Hudson 1989;
Kohrer and Domdey 1991; Collart and Oliviero 1995; Manna, Massardo et al. 1996;
Burk, Dawson et al. 2000; Rivas, Vizcaino et al. 2001; Ausubel, Brent et al. 2002). In
this study, the commonly used HAP method with or without glass bead vortexing,
resulted in a biased distribution of the extracted RNA molecules, as evident from the
relative intensities of visible ribosomal RNA bands (Fig 2.1A) and the areas of the
corresponding peaks using a densitometer scan of the gel (Fig. 2.1B). The HAP method
was not useful in obtaining rRNAs from the late stationary phase cells (days 5 to 11) (Fig
2.2A). Even when coupled with homogenizing with glass beads the biased distribution
remained in the late stationary phase RNA. Such a bias in extraction may have
contributed to the report that the largest class of putative mRNA (3’ polyadenylated
RNA) is reduced in stationary phase (Sogin and Saunders 1980). A similar bias in the
distribution of rRNA bands from S. cerevisiae was obtained by the method described by
Rivas, et al (Rivas, Vizcaino et al. 2001) and freeze-thaw cycles before hot phenol
extraction (Manna, Massardo et al. 1996).
Werner–Washburne, et al (Werner-Washburne, Braun et al. 1993) hypothesized
that the increase in RNases that co–purify with the stationary phase mRNA through
phenol chloroform extraction could contribute significantly to the loss of poly(A) RNA.
We found that RNases may not be the sole reason for the observed bias in the RNA
extracted by the HAP method with or without bead beating as extraction in the presence
of RNase denaturing agents did not improve the extraction and profile of RNAs.
Recently, Lopéz de Heredia and Jansen (Lopez de Heredia and Jansen 2004) also
reported that glass bead milling and lysis by French Press lead to degraded mRNAs even
in the presence of RNase inhibitors. A possible interpretation of the observed biased
distribution is that the cell wall capsules act as a sieve through which the permeability of
larger RNA molecules is limited and small RNA molecules move more freely.
The size bias was not seen when cells were ruptured using the crushed glass beads
method as described above. The 25S and 18S rRNA bands showed the approximate ratio
22
of 2:1, when resolved on denaturing gels, as expected in eukaryotic cells (Fig. 2.1C, D
and 2D, E). Our RT-RTPCR results showed that large message (>1kb) could not be
recovered from late stationary phase cells (5-11 day old) by using the HAP method for
RNA extraction (Fig 2.3A). As for 5S and tRNA, small message could be efficiently
extracted by the HAP method. However, when the cell walls were disrupted by using the
CGB method, the large molecular weight mRNAs (as large as 8 kb) were extracted. (Fig
2.3C). These observations suggest that quantitative analyses of yeast RNA populations
particularly in stationary phase cells and even in growing cells may have under
representation of large RNAs, if the RNA has been extracted by HAP (with or without
glass bead vortexing). About 200 S. cerevisiae ORFs and 180 C. albicans ORFs have
very large coding regions that would be most affected by biased extraction (Hong,
Balakrishnan et al.; Arnaud, Costanzo et al. 2005). In C. albicans, of the 180 large
ORFs, more than half are not annotated. Among the annotated genes are ALS2 (a
member of an adhesin family implicated in biofilm formation), INT1 (a protein
implicated in adherence and found in the septin ring), POL2 (DNA polymerase induced
by interaction with macrophage), GSC1 (a subunit of Beta 1,3-glucan synthase), MEC1
(cell cycle checkpoint protein) and KEM1 (exoribonuclease required for hyphal growth
and biofilm formation), that have ORF sizes greater than 4Kb. Without the CGB or
similar effective method the determination of abundance of large and small mRNAs
would not represent the cellular abundance and expression of genes with large mRNA
would be missed in stationary phase.
This analysis demonstrated that the extracted mRNA is suitable for analysis by
RT-RTPCR and microarray analysis (unpublished observations). Previously Schultz
(Schultz 1999) and Lopéz de Heredia and Jansen (Lopez de Heredia and Jansen 2004)
reported that grinding liquid nitrogen frozen cells under liquid nitrogen yielded high
quality extracts with large mRNAs and associated proteins. Hauser, et al (Hauser,
Vingron et al. 1998) also found that liquid nitrogen frozen S. cerevisiae cells maintained
frozen while beating with tungsten carbide beads gave the best quality RNA for
microarray analysis compared to enzymatic lysis or beating with glass beads followed by
23
warmed phenol extraction. Although our results suggest that the preferred method would
yield an unbiased RNA size distribution, Hauser et al (Hauser, Vingron et al. 1998) did
not report size distribution of RNA extracted by the various methods. Further, we wanted
to know if the bias in RNA extraction was the reason for the drastic reduction in the
levels of housekeeping genes reported in stationary phase. C. albicans ACT1 mRNA is
reported to decrease drastically in stationary phase as determined by Northern blot
analysis (Delbruck and Ernst 1993; Swoboda, Bertram et al. 1994). A 100 fold reduction
in the levels of ACT1 transcript in the S. cerevisiae stationary phase has been reported
earlier using RT-RTPCR and Northern blot analysis (Wenzel, Teunissen et al. 1995;
Monje-Casas, Michan et al. 2004). But we found that, regardless of the method of RNA
extraction used, the commonly used housekeeping gene standard, ACT1, showed reduced
levels of mRNA in the stationary phase. We also showed that the mRNA of two more
genes which are frequently used as internal standards for relative quantitation of
transcript levels, EFB1 and TDH3, were reduced significantly in the stationary phase and
therefore none of these three is suitable for standardization of stationary phase analyses.
In a microarray analysis of response to glucose starvation, as occurs in stationary phase,
within one hour the abundance of EFB1 and TDH3 decreased (Lorenz, Bender et al.
2004).
In contrast to the decrease in expression of these three genes, the expression of
CHS2 increased between day 1 and day 3 and remained elevated through 11 days (Fig
2.3B). While Chs3p is responsible for most of the chitin synthesis in yeast and hyphal
forms, the Chs2p contributes to increased synthesis in hyphal forms (Munro, Schofield et
al. 1998). Insoluble glucan (residual glucan+chitin) is greatest in early stationary phase
cells compared to other growth stages or forms (Sullivan, Yin et al. 1983). Chitin of
stationary phase cells may contribute to reduced drug susceptibility since treatment of
cells with chitinase partially reduced the phenotypic amphotericin B resistance of
stationary phase cells (Gale, Ingram et al. 1980). These observations raise the possibility
that increased expression of CHS2 may contribute to the cell wall changes that develop in
stationary phase cells and that likely contribute to the failure to extract large RNAs from
24
stationary phase yeast cells. The CGB extraction method revealed that the three genes,
IRA2, RAD50, OXR1, with large coding regions continued to be expressed through day
11.
In summary, the extraction of RNA from C. albicans cells frozen and ground with
glass beads reduced bias against large RNAs. In addition, RNAs were extracted from late
stationary phase cells, suggesting that this or a similar method may be essential for
analysis of RNAs from such cells. Analysis of the selected genes showed that genes
continue to be expressed during stationary phase with patterns of unchanged, increased or
decreased expression after active growth. Decreased expression of several genes
frequently used for internal calibration showed that they were not suitable for stationary
phase studies. Because the method of extraction affected the RNA profile, this method or
a similar method should be considered for applications requiring proportional
representation of RNA populations.
25
CHAPTER III
DEFINING CANDIDA ALBICANS STATIONARY PHASE BY CELLULAR
AND DNA REPLICATION, GENE EXPRESSION AND REGULATION
Abstract
Stationary phase Candida albicans yeast cells harbor properties of better adherence,
virulence and elevated drug resistance. Ironically, C. albicans stationary phase is not well
characterized in vitro either physiologically, or molecularly. Candida albicans yeast cells
were grown in rich medium with 2% glucose. Based on growth and DNA profiles of
cells, and by measurement of glucose and ethanol in the medium, we categorized C.
albicans growth curve into three distinct phases – exponential/diauxic, post-diauxic and
stationary phase. We found that, compared to exponential phase cells, mRNA content
was less abundant in post-diauxic and even less in stationary phase C. albicans cells.
Further analysis of the C. albicans transcriptome with oligonucleotide-based microarrays
revealed that although the overall mRNA content had decreased, transcripts of many
genes increased in post-diauxic as well as stationary phase. Genes involved in process
such as, gluconeogenesis, stress resistance, adhesion, DNA repair and aging were
upregulated at and beyond post-diauxic phase. Many C. albicans genes associated with
virulence, drug resistance and cell wall biosynthesis were upregulated only at stationary
phase. By screening 108 C. albicans transcription factor and cell wall mutants we could
identified 17 genes essential for either entry or survival in stationary phase at 30oC.
26
Introduction
C. albicans is a part of the normal flora of human oral, gastrointestinal, vaginal and
cutaneous surfaces. In immunocompromised patients the organism can cause infection of
the surfaces that it colonizes as well as causes systemic diseases. Additionally, C.
albicans can develop a biofilm on a large range of implanted devices as well as on some
host surfaces (Kumamoto and Vinces 2005; Mukherjee, Zhou et al. 2005). Biofilm cells
are notorious for being resistant to antifungal agents, thus making biofilm related
infections hard to treat. Under in vitro conditions, when nutrients are abundant and
conditions are favorable for growth, organisms grow exponentially. However in their
natural habitat, rarely do they encounter conditions that permit long periods of
exponential growth. In fact, many pathogenic organisms including C. albicans regularly
encounter environments of ‘feast and famine’, especially in the human host, e.g. within
the oral cavity, with respect to dietary sugars (Finkel 2006; Thurnheer, van der Ploeg et
al. 2006); or in the human gut, where C. albicans faces intense competition for nutrients
with hundreds of co–commensal prokaryotic species, thus leading to potential
compromise in functioning at the peak of its metabolic capacity. Nutrient starvation
induces cessation of growth and entrance into stationary phase, that allows
microorganisms, especially yeasts to maintain viability for several days (Werner-
Washburne, Braun et al. 1993; Finkel 2006; Uppuluri, Sarmah et al. 2006). Stationary
phase is an advantageous growth state for many organisms. In pathogenic bacteria such
as Mycobacterium tuberculosis, Escherichia coli, Streptococcus mutans, Salmonella
typhimurium, planktonic stationary phase cells are more tolerant to various stresses, and
are more resistant to antimicrobial drugs when compared to exponential phase cells
(Herbert, Paramasivan et al. 1996; McLeod and Spector 1996; Svensater, Bjornsson et al.
2001; Finkel 2006). Additionally, stationary phase is partly responsible for resistance of
Klebsiella pneumoniae and Pseudomonas aeruginosa biofilm cells to antibiotics
(Spoering and Lewis 2001; Anderl, Zahller et al. 2003).
Like bacteria, stationary phase C. albicans cells have many unique properties that
have proven favorable for the organism. Pathogenesis of C. albicans largely depends on
27
adherence to the tissues they colonize; and hyphae are an important virulence tool to help
C. albicans penetrate and invade the adhered tissue. C. albicans stationary phase cells
show better adherence to tissues of almost all organs in mice, when compared to
exponential phase cells (King, Lee et al. 1980; Cutler, Brawner et al. 1990). Also, only
stationary phase cells can generate an extensive production of true hyphae in C. albicans
(Westwater, Balish et al. 2005). Not only can stationary phase be an advantageous growth
state for C. albicans virulence, it can also cause C. albicans to be many fold more
resistant to almost all classes of antifungal drugs (Cassone, Kerridge et al. 1979; Gale,
Johnson et al. 1980; Beggs 1984). C. albicans cells in a biofilm have a similar or even
higher level of antifungal drug resistance (Nobile and Mitchell 2006). We have recently
reported that ~ 40% of the founder cells of a C. albicans biofilm reach stationary phase
(Uppuluri, Sarmah et al. 2006). However, unlike bacterial biofilms (Spoering and Lewis
2001; Anderl, Zahller et al. 2003), a direct relationship between C. albicans stationary
phase and biofilm drug resistance has not yet been shown.
Despite the fact that significant properties are acquired by C. albicans in
stationary phase, it is surprising that not even the timing of entry into stationary phase is
clearly defined. In some studies, an overnight, 24h or 48h grown C. albicans culture is
considered to be in stationary phase (Cutler, Brawner et al. 1990; Masuoka and Hazen
1999; Westwater, Balish et al. 2005; Zhao, Daniels et al. 2005), while other studies report
stationary phase to start much later in culture, i.e. between 3d and 8d (Cassone, Kerridge
et al. 1979; Dudani and Prasad 1985; Cutler, Brawner et al. 1990; Lyons and White 2000;
Song, Harry et al. 2004). Additionally, molecular research pertaining to C. albicans
stationary phase is in its nascent stages. Only a handful of genes and processes playing a
role in C. albicans stationary phase have been studied (Postlethwait and Sundstrom 1995;
Bertram, Swoboda et al. 1996; Sarthy, McGonigal et al. 1997; Lamarre, LeMay et al.
2001; Zaragoza, de Virgilio et al. 2002; Moreno, Pedreno et al. 2003; Galan, Casanova et
al. 2004; Bates, MacCallum et al. 2005; Granger, Flenniken et al. 2005; Roman, Nombela
et al. 2005; Uppuluri, Sarmah et al. 2006). On the other hand, stationary phase has been
extremely well characterized in the budding yeast S. cerevisiae (Werner-Washburne,
28
Braun et al. 1993; Gray, Petsko et al. 2004; Martinez, Roy et al. 2004). Studies in S.
cerevisiae routinely report 7d old cells as stationary phase cells and molecular techniques
such as microarrays have helped identify genes and biological processes necessary for
entry and maintenance of S. cerevisiae stationary phase (Martinez et al., 2004; Aragon,
Quinones et al. 2006; Swinnen, Wanke et al. 2006).
In the present study, we have characterized C. albicans stationary phase by
studying the pattern of growth and DNA profile of planktonic yeast cells in stationary
phase compared to exponential phase. Further, using cDNA microarrays, we have
explored the genomic expression patterns in the yeast cell as it progresses into the
stationary phase. Finally, by screening deletion mutants, we have identified genes
important for entry and maintenance of C. albicans stationary phase.
Methods
Cells harvesting and RNA preparation C. albicans strain SC5314 was maintained on
YPD (yeast extract 1% w/v, peptone 2% w/v, dextrose 2% w/v) agar plates and
transferred to YPD suspension culture with shaking (180 rpm) at room temperature (RT).
Exponentially growing C. albicans cells were subcultured in fresh YPD medium and
incubated at 30o C. Cells were recovered after various time points, exponential phase, 3d,
5d, 7d and 11d for RNA extraction. Total RNA was isolated using the standard hot acid
phenol method following grinding frozen cells using a mortar and pestle in liquid
nitrogen (Chapter II). The RNA preparation was DNAse treated and the absence of DNA
contamination was confirmed with the housekeeping gene EFB1 (Maneu, Martinez et al.
2000). RNA quality and quantity were determined as described (Uppuluri, Sarmah et al.
2006). Cell viability was determined as colony forming units (cfu) by plating replicates of
dilutions of planktonic cells prepared in sterile water on YPD plates and incubating at
37oC for 24 to 48 hours. Colonies were enumerated manually and the average
determined. Particles in a suspension culture were determined by use of a hemacytomer
and OD (optical density) measurement was determined at 600 nm. Cell size (for 1x106
29
cells/ml) and in some cases cell density were measured using a Z - series Coulter counter
(Beckman Coulter, Fullerton, CA).
Determining C. albicans diauxic shift C. albicans exponentially growing cells were
inoculated at a concentration of 1x105cells/ml into YPD with 2% glucose, and incubated
at 30oC. Aliquots of culture were recovered every hour beginning 13h to 27h, centrifuged
and filtered to remove yeast cells. Glucose concentration in the cell free media was
measured using the QuantiChrom Glucose assay kit (Bioassay systems, Hayward, CA).
Determining levels of extracellular ethanol in C. albicans growth medium Aliquots of
cultures were recovered at 0h (immediately after C. albicans inoculation), 2h, 6h, every
hour from 16h to 29h, 45h, 48h and 55h after inoculation. The cultures were centrifuged
and the supernatant retained. The levels of ethanol present in these cell-free supernatants
were determined using commercially available kits (Boehringer Mannheim/R-Biopharm)
according to manufacturer’s instructions.
Analysis of cellular DNA by fluorescence flow cytometry Aliquots (500 µl) of cells (1-
2x107 cells/ml) were sonicated for 5 seconds and fixed by incubating at 4oC, overnight in
1.5 ml of 95% ethanol. The cells were washed with 50 mM sodium citrate pH 7.0, and
resuspended in the same buffer. The cells were sonicated for 2 s, treated with 25 µl 10
mg/ml RNase A, and then with 25 µl of 20 mg/ml Proteinase K and incubated for 1 h at
50oC for both treatments. Finally, 1 ml of propidium iodide (PI) was added at a final
concentration of 16 µg/ml and the samples were stored at 4oC. A total of 1x105 PI
stained cells were analyzed with a Beckman Coulter Epics XL flow cytometer (Beckman
Coulter, Fullerton, CA). The results were analyzed with Expo V2 Analysis software
(Beckman Coulter).
Transcriptional analysis cDNA was synthesized with 10 µg total RNA using Oligo-(dT)20
primer, 10-mM dNTP (includes AA-dUTP) mix and SuperScript III RT (Invitrogen,
Carlsbad, CA). The cDNA was labeled with Cy3 NHS ester (Amersham, Piscataway, NJ)
30
and purified using the cDNA labeling and purification module (Invitrogen). Labeled
cDNA was estimated spectrophotometrically at 550 and 650nm. Corning Ultra gap II
slides were printed with 70 mer oligonucleotides (QIAGEN Inc, Valencia, CA) by the
Microarray Research Facility of the Oklahoma Medical Research Foundation. Labeled
cDNA was hybridized on to the blocked (with 3.5 ml 0.2 M sodium borate of pH 8.0,
31.5 ml of 1-methyl-2-pyrrolidone and 0.5 gm succinic anhydride) microarray slides at
42oC for 6 hours. Slides were washed at high stringency at 56oC (twice for 10 minutes
each, using 2X SSC, 0.1% SDS, twice for 10 minutes using 0.1X SSC, 0.1% SDS, and
three times for 5 minutes each using 0.1X SSC). Intensity of the hybridized signal was
determined by Axon Genepix scanner and Genepix Pro 5.0 microarray image analysis
software (Axon Instruments, Inc., Aberdeenshire, Scotland). Standard quality control
parameters applied to slides included median signal-to-background >3, Mean of median
background <500 and median signal-to-noise >10 and features with saturated pixels
<0.1%. Triplicate independent cultures were analyzed for each condition. Microarray
analyses (e.g. data normalizations, ANOVA and cluster analysis) were performed using
the Genespring V 7.2 microarray analysis software (Agilent Technologies, Palo Alto,
CA) (P�0.05). To identify the primary biological processes, molecular functions, and
cellular components associated with the different clusters, we used MAPPFinder
(GenMAPP version 2.0) to connect the gene expression data to the Gene Ontology (GO)
hierarchy. This program computed a statistically weighed score (Z score) that ranked GO
terms by their relative amounts of gene expression changes. For this analysis, S.
cerevisiae orthologs of C. albicans genes were used.
Real time RT-PCR (RT-RTPCR) The amount of mRNA in the total RNA was quantified
with the Poly (A) mRNA Detection System kit. (Promega, Madison, WI). cDNA was
synthesized from known amounts of mRNA, and equal amounts of cDNA were used as
starting template for RT-RTPCR. The detailed protocol for RT-RTPCR analysis is
described by us elsewhere (Uppuluri, Sarmah et al. 2006). Primers designed for 4 genes
NET1, MSH5, PHO80 and SNF5 are listed in Table 3.1
31
Table 3.1 Primers used for PCR. Forward (F) and Reverse (R) primers were used to
amplify a region of the orf from cDNA.
Screening of C. albicans mutants A total of 83 transcription factor and 22 cell wall C.
albicans mutants were screened for their ability to survive in stationary phase at 30oC.
These mutants were obtained from Dr. Aaron Mitchell, Columbia University.
Exponential phase YPD grown cells were subcultured into fresh YPD medium and
incubated at 30o C. Cells were assayed for growth defects at 10h, 1d, 2d, 3d, 4d, 6d, 8d
and 11d after inoculation. Serial 10–fold dilutions with cell densities ranging from 5x106
to 5x103 cells/ml were prepared for mutant and wild type cells for drop culture screening.
Five microliters of each diluted sample were spotted onto a YPD plate and incubated at
30o C. Each spot was checked for growth after 2 days. Moderate growth defects were
identified by slow growth, reduction in colony size, or marginal reduction in colony
counts; severe defects by major loss in viability. Colony characteristics of the mutant
were compared with that of the parental strain spotted on the same plate. Viability counts
for the mutants were determined as described above.
Results and Discussion
Growth profiles of C. albicans cells during exponential through 11 days. To understand
stationary phase, we first monitored C. albicans growth pattern. We recovered aliquots of
YPD growing C. albicans cells at various time points (24h to 11d). At every time point,
we measured the OD600, counted cell numbers and performed viable counts. The cells
remained viable throughout the time course of growth (Fig 3.1). The cell numbers and
OD600 revealed that the cells grew rapidly until 24h, after which they grew slowly for
Gene (systematic name)
Primer Sequence (5’-3’)
NET1 NET1F TCATGATGGAGTTGAAAGTGGTTT NET1R AGAGCTCCAGAAGCTTTGTTC MSH5 MSH5F ATGGATGGACCAGATTCGT MSH5R TCAAGTTATCACTATTGG PHO80 PHO80F TAATAAATTCCGCAATACGCCTAAC PHO80R TAGTGGCACACATTCTCTTTCATTTT SNF5 SNF5F ATGTTTATATTTTCTTGTTTCA SNF5R TCAATGATTATTATAAAAATGGCGGAT
32
another 4 days. Beyond 5 days, the growth leveled off and there was no significant
increase in cell number (p<0.05) (Fig 3.1). Cell sizing study showed that the 5 hour old
cells were significantly larger in size when compared to all other time points (Fig 3.2). As
the cells grew further, the cell size decreased even more. However, there was no
significant difference in cell size between any of the other early time points (8h, 12h, 24h
and 3d). The cell size at day 5 and the subsequent time points reduced significantly from
the cell size of the earlier time points. We have previously shown that slower growing
cells are smaller in size than more rapidly growing cells (Chaffin 1984).
Since we saw that the cells were growing slowly for as long as 5 days, we
investigated the growth profiles of C. albicans further. We first counted the proportion of
budding versus non-budding cells at different times in the growth curve, and then
performed FACS analysis to study the DNA content of these cells. We found that the 5h
and 8h old cultures had >80% cells in the budding form. As expected, for these diploid
organisms, a similar proportion of cells were in the budding 4n state and the un-budding
2n state (Fig. 3.2). At 12h, 61% of the cells were budding, and the DNA profile of these
cells showed a sharp shift from the 4n state to 2n state, as seen by the overlap between
the two states at this time point. After 24h, almost half of the cells in the culture were still
budding and ~ 40% of the cells were in the 4n state. About one third of the cells were still
budding at 3d; but by 11d, ~90% of the cells had stopped budding. The FACS profile also
showed a similar trend (Fig. 3.2). These results reinforced the findings from the growth
curve measurements that C. albicans cells grew slowly at least for a few days beyond 24
hours, before the budding and DNA replication stopped, in turn affecting the cell counts.
33
Fig. 3.1 Growth curve of cells from culture grown in YPD for extended periods. Cell
counts, viability and turbidity were measured daily for 10 days. = p<0.01. The cells
grew rapidly until 24 hours and then slowly for another 4 days. After 5 days there was no
increase in cell numbers.
* *
34
Fig. 3.2 Analysis of budding, DNA profiling and cell sizing of cells grown in YPD for
extended periods. Budding vs. non-budding patterns of cells were counted, DNA content
was analyzed by flow cytometry and cell size was determined using a coulter counter.
The cell sizes are given as numbers within each box.
Monitoring diauxic shift in C. albicans As S. cerevisiae yeast cells proliferate, they
preferentially utilize glucose as a source of energy. When glucose is exhausted from the
growth medium, the yeast cells undergo a shift in metabolism from fermentation to
respiration on ethanol. This shift, termed as diauxic shift, accompanies major changes in
gene expression due to metabolic readjustments (Werner-Washburne, Braun et al. 1993;
DeRisi, Iyer et al. 1997). Diauxic shift in C. albicans with respect to glucose has not been
studied. Since stationary phase in yeasts follows diauxic shift (Werner-Washburne, Braun
et al. 1993), we wanted to know if this shift occurred during C. albicans growth. For this
we measured the glucose and ethanol concentration in the growth medium. Our results
showed that glucose concentration in the medium remained fairly constant between 13
and 15 hours of growth, after which the levels dropped drastically (Fig 3.3). By 21 hours,
the glucose concentration was only 3mM and an hour later, C. albicans completely
exhausted the remaining glucose in the growth medium. Interestingly, only a couple of
hours before glucose exhaustion, there was a sudden increase in the level of ethanol in the
8h
12h
24h
3d
5d
7d
11d
Non – budding
Budding
61 49
35 24 13
2n 4n
43.65 + 1.7 42.71 + 0.6 41.6 + 0.4
40.27 + 0.2 37.85 + 1.25 37.17 + 0.06
5h
FL
Cel
l num
ber
Per
cent
bud
s
50.67 + 4.48
89 81
17
37.39 + 1.4
35
medium. The ethanol levels remained elevated for eight hours after glucose exhaustion,
but dropped drastically after 28 hours. It is reported that, in conditions of glucose
abundance, C. albicans prefers to grow aerobically and metabolize glucose by respiration
rather than fermentation, unlike S. cerevisiae (Ihmels, Bergmann et al. 2005). We
speculate that upon glucose exhaustion, C. albicans switches metabolism from respiration
to fermentation. The ethanol may arise from fermentation of the last of the glucose or
from an unidentified carbon source. After the complete metabolism of glucose, this
ethanol is metabolized for gluconeogenesis.
..
Fig 3.3 Measurement of glucose and ethanol in the growth medium. Supernatants were
collected from C. albicans planktonic cultures and used as samples to determine the
ethanol and glucose levels using specific assay kits, and their levels expressed as g/dL
and mM, respectively. A couple of hours before glucose exhaustion; ethanol is detected
in the medium for at least 8 hours.
mRNA profile over the time course of growth The budding, DNA and cell sizing
profiles suggested that most of the C. albicans cells stopped growing beyond 5 days. We
36
questioned if prolonged incubation of C. albicans in YPD medium would have a
reduction in mRNA quantity, as observed in S. cerevisiae (Sogin and Saunders 1980;
Werner-Washburne, Braun et al. 1993; Aragon, Quinones et al. 2006). We first extracted
total RNA from C. albicans cells grown for 24h, 3d, 5d, 7d and 11d. For this we used an
RNA extraction method recently reported by us (see material and methods), that yields
good quality RNA even from late stationary phase cells. Next, we estimated the mRNA
content within the total RNA. Indeed, the mRNA content in C. albicans decreased ~1.6
fold from 24h to 3d. As the cells were incubated further, the mRNA content kept
gradually decreasing. Between 5d and 11d, there was a 2 – 4 fold reduction in the mRNA
content compared to the exponentially growing cells (Fig. 3.4). This indicated that a
decline in overall transcription occurs as cells approach and enter stationary phase.
Fig. 3.4 Reduction in C. albicans mRNA abundance. The amount of C. albicans mRNA
decreased beyond diauxic shift.
Defining C. albicans growth The growth curve and the timing of diauxic shift revealed
that the C. albicans cells grew rapidly until glucose was exhausted in the medium, after
which they grew slowly for another 4 days. Budding and DNA profiles also showed that
beyond 24 hours to 3 days more than one third of the cells were still budding and
replicating their DNA. However, beyond 5 days, greater than three fourths of the cells in
culture had stopped budding or replicating their DNA. Based on these results, we
10
20
30
40
50
60
70
1 3 5 7 11
mR
NA
(pg)
Days
37
characterized C. albicans growth into 3 distinct phases. First, the exponential or diauxic
phase in which C. albicans cells grow actively and utilize glucose as the sole source of
carbon. Second, the post-diauxic phase, that lasts for ~ 4 days after diauxic shift. At this
stage C. albicans cells grow relatively slowly, metabolizing other carbon compounds for
growth. Immediately following the post-diauxic phase is the stationary phase (beyond 5
days), during which there is no net increase in cell number.
A few studies have also reported this typical pattern of growth in C. albicans.
Some of these studies considered 48h to 3 day old C. albicans cells as post-diauxic phase
cells and 8 day old cells as stationary phase cells (Lyons and White 2000; Song, Harry et
al. 2004; Vinces, Haas et al. 2006).
Overview of changes in global gene expression at different phases in the growth curve.
After characterizing the various growth phases of C. albicans, we investigated C.
albicans stationary phase at a molecular level. Although the overall quantity of mRNA
decreased beginning diauxic shift, not all C. albicans transcripts decrease in abundance in
stationary phase. Few of them such as, those encoded by the superoxide dismutase SOD3,
the cell wall integrity protein CWT1, the ATP binding cassette transporter CDR1 and two
other C. albicans stationary phase genes SNO1 and SNZ1, have been found to accumulate
at high levels in stationary phase compared to the exponential phase (Lyons and White
2000; Lamarre, LeMay et al. 2001; Moreno, Pedreno et al. 2003; Uppuluri, Sarmah et al.
2006). However, apart from these handful of genes, not many having a role in C. albicans
stationary phase have been identified. We used oligonucleotide microarray analysis as a
tool to understand better, the regulation of C. albicans stationary phase at a genomic
level. We picked five time points for analysis which corresponded to C. albicans
exponential phase (12h), post-diauxic phase (3d) and stationary phase (5d, 7d and 11d). A
comparison of the Log2 gene expression ratios between all the time points revealed 1130
genes differentially expressed > 2 fold, (p > 0.01) between all the time points. Of these
genes, 395 were of unknown functions. K – Means clustering of the differentially
expressed genes revealed 5 major patterns of gene expression (regulated > 2 fold) over
38
the different time points. Of the 5 clusters, Cluster I included 298 genes upregulated only
in the exponential phase compared to the other time points. Cluster II comprised of 166
genes having similar expression patterns between exponential phase and 3 d time point,
and were expressed higher than the other time points. Of the other three remaining
clusters, the first cluster (cluster III) comprised of 229 genes that, when compared to the
exponential phase were upregulated at day 3 and remained elevated until day 11. The
second cluster (cluster IV) included 148 genes that showed earliest upregulation at day 5
and the final cluster, (cluster V) included 64 genes showing a major upregulation late in
the time course, at day 11. The remaining genes showed random patterns of expression
and could not be included as a part of any cluster.
To identify the primary biological processes, molecular functions, and cellular
components associated with the different clusters, we used MAPPFinder (GenMAPP
version 2.0) to connect the gene expression data to the Gene Ontology (GO) hierarchy.
This program computed a statistically weighed score (Z score) that ranked GO terms by
their relative amounts of gene expression changes. The five clusters and the biological
processes governing them are illustrated in Figure 3.5.
Validation of microarray gene expression by RT-RTPCR
We used RT-RTPCR to support the differential gene expression observed with
global transcriptional analysis at all the time points. We picked eleven genes for this
analysis. The RT-RTPCR expression of seven genes (PNC1, IRA2, RAD50, CHS2,
TDH3, ACT1 and EFB1), at all time points has been recently reported by us (Uppuluri,
Perumal et al. 2006). The results for these genes by RT-RTPCR were similar to the
results obtained by microarray analysis. For example, microarray analysis revealed that
compared to exponential phase, the transcript level of PNC1 had a transient decrease at 3
days, but increased at, and remained elevated beyond 5 days. On the other hand, the
mRNA of CHS2 was first upregulated at 5 days and remained upregulated throughout the
time course. Also, mRNA abundance of the house keeping genes, TDH3, ACT1, and
EFB1 reduced many fold during the post-diauxic phase and stationary phase, when
39
compared to exponential phase cells. We saw an identical pattern of expression in our
RT-RTPCR results (Chapter II).
For the present study, we analyzed four more genes (MSH5, SNF5, PHO80 and
NET1) for the verification of microarray gene expression, using the similar technique. We
found that, for these genes too, the RT-RTPCR corroborated the differential expression
observed by microarray analysis (Fig. 3.6). Due to the unsuitability of the housekeeping
genes for normalization, we used the same mRNA quantity as starting material instead of
the commonly used total RNA (for both studies), for performing RT-RTPCR and also
calculated absolute transcript numbers for every gene.
40
Fig. 3.5 Cluster analysis of genes differentially expressed in different growth phases of
C. albicans. Based on the similarities between the patterns of expression, all the
differentially expressed genes were categorized into 5 different clusters. Cluster I
contained genes upregulated only in exponential phase denoted as E (A), Cluster II
included genes having highest expression at E and 3d (B), Genes in Cluster III were first
upregulated at 3d, and remained elevated until 11d (C), Cluster IV and V contained
genes upregulated late in the time course, at 5d and 11d respectively (D, E)
41
Fig. 3.6 RT-RT PCR verification of microarray expressed genes. Four genes
differentially expressed in microarrays identified by different signal intensity (log2) (A),
were verified by using RT-RTPCR (B). Absolute counts of the mRNA are represented as
log10 values in the RT-RTPCR results.
Alteration of gene expression during growth and stationary phase.
Cluster I: C. albicans growth and proliferation regulating genes A total of 335 genes were
upregulated at 12 hours, in exponentially growing cells (cluster I) when compared to any
other time point. All the biological processes that these genes represented were effected
significantly (p > 0.05), and showed a z – score of at least 2.6. As expected from
exponentially growing cells (cluster I), one of the major upregulated processes (compared
to all other time points) included the cyclin dependant kinases CDC28, CDC1 and the
other protein kinases, CKA1 and KIN3, all of which are known to be important for
progression of the cell cycle (Hartwell, Culotti et al. 1970; Sherlock, Bahman et al. 1994;
Bachewich, Nantel et al. 2005; Bruno and Mitchell 2005) (Fig 3.5A). In yeasts, the G
proteins, RAS1 and RAS2 are activated when nutrients are abundant and in turn activate
the adenylate cyclase gene CYR1, leading to an increase in intracellular cAMP and in turn
promoting bud growth and cell polarity (Gray, Petsko et al. 2004). These three genes, and
some more genes involved in the Rho GTPase mediated signal transduction, e.g. BEM2,
BNR1, RSR1 (Yaar, Mevarech et al. 1997; Weeks and Spiegelman 2003; Bassilana,
Hopkins et al. 2005; Li, Wang et al. 2005) showed highest levels of expression in the
exponential phase time point. At 12 hours, glucose is abundant in the medium and is the
42
primary source of carbon. As a result, most of the genes of the glycolysis pathway, and
glucose transport were positively regulated during the exponential phase time point.
Consistent with the budding and DNA profiling which revealed that 80% of the cells
were still budding and replicating their DNA during exponential phase, we found genes
involved in C. albicans DNA replication, mitosis and cell wall maintenance as primary
processes upregulated in the exponential phase time point. Another large category of
genes showing the highest expression at exponential phase were the ribosomal biogenesis
or the rRNA processing genes. Interestingly, many mitochondrial ribosomal genes were
found upregulated in this cluster, indicating that C. albicans indeed prefers aerobic
respiration for growth when growing exponentially. Upregulation of three mitochondrial
ribosomal genes during the active growth phase has also been previously reported
(Ihmels, Bergmann et al. 2005). However, the mitochondrial genes were not all
upregulated in exponential phase. Many genes either did not change throughout the time
course of 11 days of growth, or were upregulated late in the time course. Few of these
genes are discussed later in this study.
Cluster II We found many more genes related to DNA replication and mitosis being
upregulated at levels similar to exponential phase, even in the post diauxic phase (3d time
point; cluster II) (Fig 3.5B). This gene expression result supported our growth and DNA
profiling results that C. albicans cells were still growing in the post – diauxic shift phase,
albeit slowly. However, further analysis of the microarray data revealed that the genes
expressed during the post-diauxic phase also shared properties with the genes expressed
in stationary phase. Many processes were upregulated first at this time point (3d), and
remained elevated until 11 days (cluster III).
Cluster III: DNA repair, stress resistance and aging Although the 3d time point shared
many common highly expressed genes with the exponential phase time point, many
genes were also upregulated greater than the exponential phase at this time point.
Grouped under cluster III, these were 244 genes, the transcript levels of which were first
43
upregulated beginning 3 days and kept gradually increasing till 11d. While the 3d old
cells were still replicating their DNA and growing slowly, these cells also transcribed
genes coding for different mechanisms of DNA repair. C. albicans upregulates the repair
genes in the post-diauxic phase (a point in time just before stationary phase), to combat
the extensive DNA damage that is expected during stationary phase (Gray, Petsko et al.
2004). We found a few stress resistance genes being upregulated at 3d. One of these
genes was SOD6, a superoxide dismutase known to provide resistance against oxidative
stress in C. albicans (Martchenko et al., 2004). Many orthologs of S. cerevisiae stress
resistance genes, having unknown functions in C. albicans were also included in Cluster
III (Fig 3.5C). Between growth and stationary phase, C. albicans cells excrete
metabolites into the medium that have quorum sensing functions; and the conditioned
medium recovered from stationary phase cells is known to protect cells from oxidative
stress (Westwater, Balish et al. 2005). In fact, stationary phase yeast cells are by
themselves more resistant to various stresses compared to exponential phase cells
(Aragon, Quinones et al. 2006).
An interesting category of genes upregulated during the diauxic phase were those
involved in yeast cell aging. As yeast cells age, they accumulate in their nucleus,
extrachromosomal rDNA circles (ERC) that dilute the DNA replication machinery and
lead to senescence and cell death. Silencing of the rDNA locus, thus preventing ERC
formation is carried out by histone deacetylases thus promoting longevity in yeast cells
(Vijg and Suh 2005). Overexpression of the C. albicans chromatin silencing genes first in
diauxic phase and its gradual increase in stationary phase is in concert with the well
known fact that stationary phase advances replicative aging in yeast cells (Ashrafi,
Sinclair et al. 1999)
Cluster III: Gluconeogenesis and antagonists of TOR pathway In Cluster III, we found
upregulation of genes whose transcription levels are reported to be elevated when
engulfed by macrophages. In glucose limiting conditions such as the post-diauxic shift,
yeast cells utilize alternative sources of organic molecules for metabolism, and undergo
44
the process of gluconeogenesis (Gray, Petsko et al. 2004). Also, when phagocytosed, C.
albicans cells elicit a similar response, characterized by the upregulation of the
gluconeogenesis and beta-oxidation pathway (Lorenz, Bender et al. 2004). Consequently,
we also saw increase in the transcript levels of most genes involved in gluconeogenesis,
the beta-oxidation pathway and the glyoxylate pathway during C. albicans post-diauxic
phase. Since beta oxidation occurs in the peroxisome (Gurvitz, Hiltunen et al. 2001),
many genes having functions in peroxisome organization and biogenesis were also
similarly upregulated (Fig 3.5C). Corresponding to this increase, a decrease in the
glycolytic pathway genes was also observed (Fig 3.7).
45
Fig 3.7 Metabolic reprogramming inferred from changes in gene expression during
diauxic shift. Only key metabolic intermediates are identified. The names of genes
upregulated in diauxic shift are in shown in boxes, while those downregulated are circled.
The magnitude of induction or repression is indicated for these genes. The direction of
the arrows connecting reversible enzymatic steps indicate the direction of the flow of
metabolic intermediates.
46
The gene encoding the kinase TOR1 was found downregulated while two
transcription factors, GLN3 and GAT1 were upregulated at 3 days and was categorized
into cluster III. The later two genes are normally induced by a shift from good to poor
nitrogen or carbon nutrient sources (Cruz, Goldstein et al. 2001; Gray, Petsko et al.
2004). The two genes are also considered to act antagonistically to the TOR pathway –
the global nutrient sensing pathway of most pathogenic fungi that promotes cell
proliferation in nutrient rich condition. Upregulation of the two genes and the
downregulation of TOR1 indicated that the TOR pathway was inhibited beyond diauxic
shift. Inhibition of TOR function also activates the mitochondrial signaling pathway
genes and genes involved in aerobic respiration (Gray, Petsko et al. 2004). One gene
from each of the two processes, RTG3 and HAP1, were over expressed by the 3 day old
cells, and remained upregulated throughout the next 7 days. Also upregulated beyond 3
days were genes involved in the repair and maintenance of the mitochondrial genome,
mitochondrial ribosome biogenesis, and mitochondrial electron transport. This highlights
the fact that mitochondrial function is essential for cells in post-diauxic and stationary
phase. Earlier in this study we had shown that C. albicans transiently switched
metabolism from respiration to fermentation. Upregulation of mitochondrial genes
beyond 3 days probably means that aerobic respiration is the prominent way by which C.
albicans survives in the stationary phase.
Clusters IV and V: RNAses and proteases From cluster III, we found that many genes
and processes important for stationary phase were first upregulated at post-diauxic phase.
However, further analysis of the microarray results revealed that genes upregulated first
in stationary phase (at and beyond 5 days), functioned in a few processes unique from
that seen in exponential or post-diauxic phase cells (Fig 3.5D, E).
In this study, we earlier found that, C. albicans diauxic shift led to a modest
decrease in mRNA quantity that became greater in the stationary phase. Although we did
find a couple of genes coding for ribonucleases (RNAse) at 3d, mRNA degradation was
the major process upregulated beginning at 5d, and remaining elevated until 11d (cluster
47
IV). A few more genes involved in this process were also upregulated late into stationary
phase, at 11d (cluster V). Not only were the RNAse genes upregulated, but so were some
genes coding for proteases. Given the observation that the overall transcriptional as well
as translational machinery is down regulated in S. cerevisiae stationary phase (Sogin and
Saunders 1980; Werner-Washburne, Braun et al. 1993; Kuzj, Medberry et al. 1998;
Aragon, Quinones et al. 2006), finding these genes was not surprising.
Clusters IV and V: Mannoproteins and other cell wall proteins In our gene expression
data we found the ATPase gene PMRI, upregulated greater than 2 fold at 5d. Also
upregulated were genes involved in the secretory pathway (SEC21, SEC59, COD2, SLY1,
GEA2 and RUD3). For the maintenance of viability in stationary phase, trafficking
between ER to Golgi is known in both C. albicans and S. cerevisiae to be an essential
process (Werner-Washburne, Braun et al. 1993; Bates, MacCallum et al. 2005). The
Golgi P-type ATPase, PMR1 plays an important role in this process by transporting
divalent cations to the Golgi, in turn activating mannosyltransferases and finally effecting
glycosylation of mannoproteins (Bates et al., 2005). Corresponding to this function, we
found genes belonging to the highly studied C. albicans PMT (PMT2, PMT4) and MNT
(MNT4) family of mannosyltransferases (Bates, MacCallum et al. 2005), as well as genes
coding for cell wall mannoprotein biosynthesis (MNN2, MNN4, ALG2, ALG5, MIT1),
upregulated beyond 5d. The mannoproteins play important roles in C. albicans adhesion,
antigenicity and modulation of the host immune responses (Bates, MacCallum et al.
2005).
Besides genes involved in mannoprotein biosynthesis, our microarray
results revealed that, C. albicans beta-1,6-glucan biosynthesis genes, KRE5, KRE9, as
well as chitin biosynthesis and distribution genes, YEA4, GNT1, BNI4 and CHS2 had
increased mRNA content in the stationary phase C. albicans cells (beyond 5d), compared
to any other time points – exponential or post diauxic. These two components of C.
albicans cell wall, beta-1,6-glucan and chitin are responsible for the unbalanced cell wall
growth in stationary phase, leading to phenotypic drug resistance (Cassone, Kerridge et
48
al. 1979; Gale, Johnson et al. 1980; Beggs 1984; Cassone 1986). The increase in these
two components is known to cause significant changes in the ultrastructure of the C.
albicans cell wall. Cassone et al., 1979, reported that the C. albicans cell wall becomes ~
65% thicker after 6 days of growth compared to the exponential phase cells. Also there is
no difference in the thickness of the cell wall between exponential phase and 3 day old C.
albicans cells (Cassone, Kerridge et al. 1979). This supports our defined framework of
the different C. albicans growth phases which proposes that stationary phase begins at
day 5, while 3d old cells are still in the post-diauxic phase. We have earlier indicated that
such a thick cell wall may also prove to be a major barrier for extraction of large
molecular weight RNA from stationary phase cells (Uppuluri, Perumal et al. 2006).
Clusters IV and V: Drug resistance and virulence genes Change in phenotype is a major,
but not the only mechanism by which C. albicans can protect itself from antifungal drugs.
C. albicans can express many genes coding for different multidrug efflux pumps, thus
conferring resistance to the very potent azole family of fungicides (White, Marr et al.
1998; Lyons and White 2000). Besides other genes involved in processes leading to a
high basal resistance towards antifungal drugs such as PMT4 and orf19.6382 (Gaur et al.,
2005; Prill et al., 2005), we found the C. albicans drug efflux pump gene CDR1
upregulated first at the post diauxic phase and remaining elevated throughout the
stationary phase. Lyons and White, 2000, have also reported an identical expression
pattern and suggested that this efflux pump perhaps helped in the riddance of toxins that
accumulate during stationary phase (Lyons and White 2000; Gaur, Choudhury et al.
2005; Prill, Klinkert et al. 2005).
If diauxic shift (3d) was the phase when C. albicans had improved adherence
properties, the yeast expressed many virulence genes by stationary phase (5d). Secretory
aspartyl proteases (SAP) are one of the major virulence factors of C. albicans (Tavanti,
Pardini et al. 2004; Andes, Lepak et al. 2005). Our present gene expression studies
showed that although most of the C. albicans SAP genes had similar levels of
transcription throughout the time course, some of these had a small but significant
49
increase in the stationary phase (~ 1.5 fold). However, a few other putative C. albicans
virulence genes such as those induced during infection in murine kidney, were
upregulated more robustly at or beyond 5d. Just like post-diauxic phase, mitochondria
associated genes were also upregulated in stationary phase. Screening mutants of known
S. cerevisiae stationary phase genes suggested that mitochondrial function was critical for
the entry into stationary phase in that organism (Martinez, Roy et al. 2004).
Screening of C. albicans transcription factor and cell wall mutants Unlike S. cerevisiae,
whole genome mutants for C. albicans are not yet available. However, strains bearing
mutations in many (but not all) C. albicans transcription factors as well as cell wall
associated genes are readily available. To determine whether any of these regulatory
genes might be essential for stationary phase, we screened 83 putative C. albicans
transcription factor and 22 cell wall genes. All 105 mutant strains grew as well as the
wild type strain for the first 3 days. Between 4 and 11 days, 34 and 17 mutants showed
moderate and severe growth defects respectively (Fig 3.8). Of the strains showing severe
growth defects, 2 were cell wall mutants (WSC1 and SUN41) while 15 were transcription
factor mutants. A total of 7 mutants (including one cell wall mutant, WSC1) showed a
major growth defect at 4d compared to the wild type cells (Table 3.2). These strains bore
defects in genes important for various functions such as gluconeogenesis (GAL4, RMD2),
lysine biosynthesis (LYS142), biofilm formation (BCR1) and maintenance of cell wall
integrity (WSC1) (Fig 3.8A). The remaining two genes (orf19.1694 and STP4) had
unknown functions. Ten strains showed fatal growth defects later in the time course - 6d
(3 mutants), 8d (5 mutants) and 11d (2 mutants). At 11d, even the control showed defects
characterized by slow growth that manifested into pinpoint colonies (Fig 3.8C). However,
the mutant strains exhibited more severe growth defects. The strains showing fatal
growth defects in stationary phase (6d to 11d) carried deletions in genes coding for C.
albicans cell wall proteins (SUN41), for proteins involved in C. albicans steroid
biosynthesis (ZCF17) (Fig 3.8A,B) and proteolytic processing (STP1). Five mutants
harbored defects in genes having unknown functions in C. albicans, but were orthologs to
50
S. cerevisiae genes that had various functions. These genes were, orf19.2315 (ScRTG3)
(Fig 3.8B) - a gene involved in S. cerevisiae mitochondrial signaling pathway and
conferring longevity, orf19.173 (ScAZF1) - a mitochondrial genome maintenance gene,
CaZCF34 (ScHAP1) - a gene important for aerobic respiration, orf19.1589 (ScRRN7) - a
gene involved in rDNA transcription by Pol I, and CaUME7 - a gene having unknown
functions in C. albicans, but orthologous to ScUME6 that is important for meiosis and
sporulation in S. cerevisiae (Fig 3.8C). Lastly, we identified 2 mutants that had mutations
in genes having unknown functions in both C. albicans and their S. cerevisiae orthologs –
SEF1 and ZCF28.
To complement the drop plate mutant screening results, we performed viable
counts for each of the 17 mutants showing severe growth defects. We found that the day
of severe defect by drop plate screening, and the day at which the viable (colony) counts
of the individual mutants reduced drastically, was essentially identical for 14 mutants.
For the remaining 3 mutants (RRN7, AZF1 and SEF1) the reduction in viable counts
showed a delay of 2 days. Since we screened the mutants every other day and not
everyday, it could be possible that delay was shorter than 2 days.
Ten out of the 17 stationary phase defective mutant genes were differentially
expressed in our microarray analysis, while the remaining 7 genes showed a steady level
of expression throughout (Table 3.2). Two genes falling into the later category were the
S. cerevisiae orthologs of C. albicans transcription factors involved in gluconeogenesis
and upregulated in S. cerevisiae post-diauxic phase - RMD5 and GAL4. GAL4 has a
completely different function in C. albicans (regulation of glycolysis and TCA cycle
genes) than S. cerevisiae (galactose metabolism) (Martchenko, Levitin et al. 2006). Also
function of RMD5 in C. albicans is not yet known. It has been recently discovered that
functions of many transcription factors, in C. albicans are rewired compared to S.
cerevisiae (Ihmels, Bergmann et al. 2005). Hence, although the two C. albicans
transcription factors may not be involved in gluconeogenesis, a steady expression state of
these two genes is probably essential for C. albicans growth in the post diauxic phase.
51
Mutation in one other gene AZF1, having unknown function in C. albicans, but
involved in mitochondrial genome maintenance in S. cerevisiae, lead to severe
compromise in viability after 6 days of growth. A similar growth defect was observed in
S. cerevisiae. Our microarray results revealed that the level of AZF1 expression was
similar both in glucose abundant (exponential phase) and in glucose deficient post-
diauxic and stationary phase. This was in accordance with the protein expression pattern
of S. cerevisiae Azf1 protein (Slattery, Liko et al. 2006). This re-emphasizes the
observation that steady state expression levels of some genes are essential for C. albicans
survival in stationary phase. On the other hand, two more mitochondrial associated genes
RTG3 and HAP1 showed microarray gene expression-specific growth defect at 6 days.
The mutant screening results and differential gene expression patterns revealed that just
like in S. cerevisiae, the mitochondrion played a critical role, both in the C. albicans post
diauxic phase and stationary phase.
However, for a majority of mutant strains (10/17), the day of severe growth defect
in the individual mutant strains correlated well with the day when the gene showed the
greatest change in microarray expression (Table 3.2).
52
Figure 3.8 Screening of C. albicans transcription factor and cell wall mutants by drop
plate method. Mutant strains were examined for survival during various phases of growth
at 30oC.
2x104 2x103 2x102 2x101
Control
Day 4
Moderate defect ZCF17
Severe defect WSC1
Control
Severe defect ZCF17
Severe defect RTG3
Day 8
Day 11
Control
Severe defect UME7
A B
C
53
Table 3.2. Correlation between microarray gene expression and reduction in viability of
the mutant strains. For ten genes, the day of severe growth defect in the individual mutant
strains corresponded with the day when the gene showed the greatest change in
microarray expression. For the remaining seven there was no significant change (NC) in
the microarray Log2 values at any of the time points.
54
CHAPTER IV
CANDIDA ALBICANS SNO1 AND SNZ1 EXPRESSED IN STATIONARY
PHASE PLANKTONIC YEAST CELLS AND BASE OF BIOFILM
Abstract
The Candida albicans orthologs of the most studied Saccharomyces cerevisiae
stationary-phase genes, SNO1 and SNZ1, were used to test the hypothesis that, within a
biofilm, some cells reach stationary phase within continuously fed, as well as static, C.
albicans biofilms grown on dental acrylic. The authors first studied the expression
patterns of these two genes in planktonic growth conditions. Using real-time RT-PCR
(RT-RTPCR), increased peak expression of both SNZ1 and SNO1 was observed at 5 and
6 days, respectively, in C. albicans grown in suspension culture. SNZ1–yellow
fluorescent protein (YFP) and SNO1–YFP were constructed to study expression at the
cellular level and protein localization in C. albicans. Snz1p–YFP and Sno1p–YFP
localized to the cytoplasm with maximum expression (>90 %) at 5 and 6 days,
respectively, in planktonic conditions. When yeast growth was reinitiated, loss of
fluorescence began immediately. Germ tubes and hyphae were non-fluorescent.
Pseudohyphae began appearing at 9 days in planktonic yeast culture and expressed each
protein by 11 days; however, the cells budding from pseudohyphae were not fluorescent.
Biofilm was formed in vitro under either static or continuously fed conditions. Increased
expression of the two genes was shown by RT-RTPCR, beginning by day 3 and
increasing through to day 15 (continuously fed biofilm). Only the bottommost layer of
acrylic-adhered cells in the biofilm showed 25 and 40 % fluorescence at 6 and 15 days,
respectively. These observations suggest that only a few cells in C. albicans biofilms
express genes associated with the planktonic stationary phase and that these are found at
the bottom of the biofilm adhered to the surface.
55
Introduction
Candida albicans is capable of forming a biofilm on mucocutaneous surfaces, as
well as on medical devices such as dentures and catheters (reviewed by Douglas, 2003;
Kumamoto & Vinces, 2005; Mukherjee et al., 2005). Sixty-five percent of edentulous
individuals suffer with denture stomatitis and 40 % of patients with intravenous catheters
develop acute fungaemia due to the growth of C. albicans biofilms on the associated
biomedical devices. Biofilms also show enhanced resistance to antifungal drugs.
In vitro, biofilms have been shown to form on catheter, polymethylmethacrylate
(denture acrylic) and polystyrene surfaces (Douglas, 2003; Kumamoto & Vinces, 2005;
Mukherjee et al., 2005). C. albicans forms a biofilm in three distinct developmental
stages. The bottommost layer of adhered yeast cells act as founder cells, anchoring the
developing biofilm to the substrate. The middle layer is composed of hyphae and
pseudohyphae, and the topmost part of the biofilm consists mostly of a thicker and open
hyphal layer and more extracellular matrix (ECM) (Baillie & Douglas, 1999; Chandra et
al., 2001; Ramage et al., 2001). After 48 h, these biofilms range in thickness from 25 to
>450 µm and are metabolically active communities of cells interspersed with ramifying
water channels. The structural complexity of the biofilm may create a gradient of
environmental conditions in which the C. albicans cells enter distinct physiological states.
One such state may be equivalent to that of stationary-phase planktonic yeast cells, and,
in particular, the founding yeast cells at the surface of the substratum may cease growing.
The stationary phase and the genes involved in its progress and maintenance have not yet
been well characterized in C. albicans, although several genes have been reported to
show increased expression as active growth slows (Lamarre et al., 2001; Moreno et al.,
2003). In this study, we investigated the expression patterns of the C. albicans orthologs
of the two most studied stationary-phase genes in Saccharomyces cerevisiae, SNO1 and
SNZ1. We first determined whether the expression pattern of these genes, monitoring
both RNA and protein, was associated with the stationary phase of planktonic yeast cells,
hyphae and pseudohyphae. We then used these two genes as indicators of the stationary
phase to study the physiological state of the cells in a biofilm.
56
Methods
Organisms and growth conditions. C. albicans SC5314 or CAI4 (obtained from William
Fonzi, Georgetown University Medical Center) was maintained on YPD medium (1 %
yeast extract, 2 % peptone, 2 % glucose)-containing 2 % agar plates. Planktonic yeast cell
cultures were grown in yeast nitrogen base (YNB) medium with amino acids (Difco
Laboratories) containing 2 % glucose at room temperature on a gyratory shaker
(180 r.p.m.). Hyphae were induced by inoculating 1x107 yeast cells ml–1 from 24 h culture
into YNB medium at 37 °C and incubated with shaking for 2 h or 2 days. For stationary-
phase studies, seven flasks (one for each time point) containing YNB broth (200 ml) were
inoculated with 5x105 yeast cells ml–1 from an overnight culture in YNB. The cells were
harvested by centrifugation at 4 °C and maintained at –80 °C before RNA isolation. Cell
viability was determined as c.f.u. by plating replicate dilutions of planktonic cells
prepared in sterile water on YPD plates and incubating at 37 °C for 24 h. Colonies were
enumerated manually and the mean determined. Particles in suspension culture were
determined by use of a haemocytometer and by OD595 measurement. Denture acrylic
(polymethylmethacrylate) pieces (1.0 cm2 square or 90x20x1.5 mm) prepared by Dr
Thomas McKinney (Baylor College of Dentistry, Dallas, TX) were used to support the
biofilm formation in two model systems. Pieces of acrylic were placed in disposable
polystyrene dishes (35x10 mm). A suspension of yeast cells (4 ml) grown to a density of
1x107 cells ml–1 in YNB was added to the dish and incubated for 2 h at 37 °C without
shaking. The liquid was gently aspirated; 4.0 ml fresh medium was added and incubated
for 6 days. Alternatively, the strips were placed in a 50 ml syringe barrel with a yeast
suspension and then washed with YNB to remove non-adhered cells, before starting YNB
medium flow through the syringe at 50 ml h–1 at 37 °C. Sterile air was supplied into the
medium at 1 l h–1. Acrylic pieces were removed, washed gently by dipping in PBS
(10 mM phosphate buffer, 2.7 mM KCl and 137 mM NaCl, pH 7.4). For microscopy, the
top matrix layer, mostly consisting of hyphae, was collected by very gently dipping and
shaking the washed biofilm in PBS. The middle layer, composed of yeast, pseudohyphae
and some hyphae, was collected using sterile forceps, while the bottommost acrylic-
57
adhered layer, composed exclusively of yeast cells and germ tubes, was collected by
scraping the thoroughly washed acrylic using a scalpel and ice-cold water. Similar
distribution of the three forms of C. albicans was observed from at least four independent
biofilms. Differences in expression were determined by ANOVA (P 0.05). The viability of
the recovered cells from three biofilms was determined at 20 days, as described above.
RNA extraction Total RNA was isolated using the standard hot acid phenol method
following grinding frozen cells using a mortar and pestle in liquid nitrogen (Uppuluri et
al., 2006a; Chapter II). The RNA preparation was DNAse treated and the absence of
DNA contamination was confirmed with the housekeeping gene EFB1 (Maneu, Martinez
et al. 2000). RNA quality and quantity were determined as described (Uppuluri, Sarmah
et al. 2006).
Real time RT-PCR (RT-RTPCR) The amount of mRNA in the total RNA was quantified
with the Poly (A) mRNA Detection System kit. (Promega, Madison, WI). cDNA was
synthesized from known amounts of mRNA, and equal amounts of cDNA were used as
starting template for RT-RTPCR. The detailed protocol for RT-RTPCR analysis is
described by us elsewhere (Chapter II).
Construction of strains expressing fluorescent fusion protein C. albicans transformations
were carried out using the Alkali-Cation Yeast kit (Qbiogene). Genomic DNA from C.
albicans CAI4 strain was obtained by standard methods (Adams et al., 1997). For
construction of the yellow fluorescent protein (YFP) fusion protein, the method of
Gerami-Nejad et al. (2001) was used (Schematic Fig 4.1). Briefly, PCR was performed
using primers with 5' ends corresponding to the SNZ1 or SNO1 target gene sequences and
3' ends that directed amplification of the YFP gene along with the selectable marker
URA3. The primers used are listed in Table 4.1.
58
Table 4.1 Primers used in this study. Primers were used for obtaining full-length ORFs
(full) and short sequences, and for verification (v) of the gene–YFP constructs. The
direction of primers is indicated as forward (F) or reverse (R).
Primer Sequence (5’-3’) FOR RT-RTPCR SNO1 F TCAAACCCGGACGAATATGC SNO1 R TCTCCGCCAGGAATAACCAA SNZ1 F CAATTGGGATGTGATGGTGTTT SNZ1 R TTGTAGTGAGTGGTAGCGTTGACA SNO full F GTCTGATGAAAGTTCAACTTC SNO full R CTGTCTGTATTTCTTTTGTG SNZ full F CAACAACCTTTGTAAATAACCAAC SNZ full R CATAGATATATATACAAGGTTTC FOR SNO1-YFP AND SNZ1-YFP CONSTRUCTION
SNO1-YFP F CTTGATGAGTTTGTGATAAAGAAACTGCAACAATATATTGATAGAATAATAGGTGG
TGGTTCTAAAGGTGAAGAATTATT SNO1-YFP R CAACTGTGATTTAGTACTCTCTCTCTACTACTTACTTACTTCCTATACACACAAGATC
TAGAAGGACCACCTTTGATTG SNZ1-YFP F ATTGCCATTGATTCAATTAAAGAAGAAGAGAAATTGGAAAAAAGAGGCTGGGGTGG
TGGTTCTAAAGGTGAAGAATTATT SNZ1-YFP R TTATGTCCACAAAATCATTGTTTACTCCTCCATACAACAGAAATCAACTATCCATATC
TAGAAGGACCACCTTTGATTG FOR VERIFICATION
SNOYFP Fv CCAGAGCTAGCTGAGGATTA SNZYFP Fv TATTCAACTGATTTGGGTGAATTGAT ADH1 Rv CACAGTGGATCCAGACAATG
PCR was performed with 100 ng pYFP-URA3 (cassettes obtained from Cheryl Gale,
University of Minnesota) as the template, 0.6 µM each primer, 3.5 mM MgCl2, 5 µl
10xPCR buffer, 0.4 mM each dNTP, and 2 U Expand High Fidelity Polymerase (Roche
Applied Science). The 50 µl reactions were run for 94 °C for 4 min, then for 25 cycles at
94 °C for 1 min, 60 °C for 1 min and 72 °C for 3 min, followed by 72 °C for 10 min. The
products from 10 reactions were pooled, precipitated with ethanol, resuspended in 50 µl
water, and used to transform C. albicans CAI4 and URA3 recombinants selected in YNB
without uridine. Identification of transformants carrying the integrated cassette was
performed by PCR on genomic DNA with a primer that annealed within the
59
transformation module and a second primer annealing to the 3’ region located outside the
module.
Plasmid name
pYFP – URA3
PCR
Forward Primer (FP): 80bp ( 20bp GFP + 60bp SNO1/SNZ1 structural gene overhang)
Reverse Primer (RP): 79bp ( 23bp URA3 + 56bp SNO1/SNZ1 structural gene overhang)
PCR Product (used for transformation)
Continued…..
YFP ADH1 ter URA3 SNO1/SNZ1 SNO1/SNZ1
FP
RP SNO1/SNZ1
SNO1/SNZ1
60
Transformation
Homologus recombination
PCR product integrated into the C. albicans genomic DNA
P
Verification
P
Fig 4.1 Schematic representation of construction of the fluorescent construct,
recombination into C. albicans genomic DNA, and verification. P = Promoter
YFP ADH1 ter URA3 SNO1/SNZ1 SNO1/SNZ1
Genomic DNA; SNO1/SNZ1 ORF
YFP ADH1 ter URA3 SNO1/SNZ1 SNO1/SNZ1
YFP ADH1 ter URA3 SNO1/SNZ1 SNO1/SNZ1
FP
RP
900bp
Marker verification
61
Fluorescence microscopy. For fluorescence microscopy, cells were used without
fixation. YFP-tagged proteins were visualized in live cells with an Olympus IX71
microscope with appropriate filters. Images were captured and documented using a
Photometrics Cool Snap HQ digital camera and analyzed with Meta Morph software. For
localization, a bright-field image and a fluorescent image were first pseudo-coloured
green and red, respectively. The resultant images were then merged.
Results
Viability of stationary phase organisms. In S. cerevisiae, the stationary phase has been
associated with cultures that are at least 5 days old (Braun et al., 1996; Radonjic et al.,
2005). Since most studies with C. albicans terminate experiments after 24–48 h of
growth, we wanted to determine whether cells remain viable in culture for an extended
period. We examined the ability of organisms cultured in YNB for 2 weeks to carry out
cell division by determining the c.f.u. in the culture during the stationary phase (Fig. 4.2).
Cells did not lose viability for 10 days. At 2 weeks, >60 % of the cells were viable, after
which there was a progressive decline in the number of cells forming colonies. On day 9,
there was appearance of pseudohyphae in an appreciable fraction of cells ( 30 %).
FIG.4.2. Viability of cells from culture grown in YNB for extended periods. Viability
was analysed at different times of growth by cell counts (open circles) using a
haemocytometer and colony formation (solid circles) on YPD plates.
62
Expression of SNZ1 and SNO1 during planktonic growth To initiate our study of the
stationary phase of C. albicans, we used RT-RTPCR to examine the expression of the
two genes, SNZ1 and SNO1, predicted to be indicative of the stationary phase. In S.
cerevisiae there are three pairs of SNZ and SNO genes. The two genes of each pair are
adjacent and divergently transcribed (Balakrishnan et al., 2002). A single SNZ and SNO
gene pair was found in the annotated C. albicans genome (Arnaud et al., 2005) and a
(Altschul et al., 1997) BLAST search did not reveal other unannotated pairs. SNZ1
(Orf19.2947) and SNO1 (Orf19.2948) were adjacent and divergently transcribed, with a
sequence of 1.5 kb separating the translation initiation sites. Transcripts of the two genes
were quantified at intervals during progression into the stationary phase (Fig. 4.3A).
Although a small peak in expression was noted on day 2, the greatest expression for SNZ1
was reached on day 5, after which the transcription declined. There was a 67-fold
increase in expression between day 1 and day 5. For SNO1, there was decreased
expression up to day 4, followed by increased expression that peaked on day 6, after
which expression declined. The increased expression at day 6 was 10.5-fold higher
compared to that on day 1. In addition to a 1 day difference in peak expression, SNZ1
transcript (5.6x106–3.8x108) was more abundant than SNO1 transcript (4.2x105–4.2x106)
at all intervals measured, and the increase between day 1 and the peak of expression was
greater for SNZ1 (67-fold) compared to SNO1 (10.5-fold). The expression of these genes
began to increase after the cell numbers stopped increasing and the pattern was consistent
with that of genes with preferential expression in the stationary phase.
Biofilm formation and gene expression. C. albicans has been studied most frequently as
planktonic cells in suspension culture. However, the organism can grow in both in vivo-
and in vitro-produced biofilms. In biofilms, both yeast cells and hyphae are observed in
an ECM-covered community (Baillie & Douglas, 1999, 2000; Chandra et al., 2001). To
determine if some organisms in a biofilm reach a physiologic state in which the SNZ1 and
SNO1 markers for stationary phase are expressed, we examined biofilms formed under
two conditions. Biofilms were formed on pieces of acrylic in YNB from a yeast-cell
63
inoculum under static conditions, i.e. maintained without shaking for the duration of the
experiment. There was an 1.5- and fourfold increase in SNZ1 and SNO1 expression,
respectively, between day 1 and day 6 (Fig. 4.3B). However, overall expression of both
genes in biofilm conditions at day 6 was at least 25 times less than the maximum that was
recorded under planktonic conditions.
The reduced expression in the biofilm compared to planktonic cells could be
attributable to the heterogeneous nature of the biofilm, which contains hyphal,
pseudohyphal and yeast organisms, or to the presence of growing organisms responsible
for the release of primarily yeast cells from the biofilm. To address these possibilities, we
used a different system for biofilm formation. Biofilms were formed under flow
conditions that replenished medium and permitted the biofilm to be maintained for
15 days. To answer the question of whether the bottommost layer of the biofilm formed
from founder yeast cells reaches the stationary phase earlier than the rest of the biofilm,
we separated two layers of the biofilm. We collected the bottommost adhered layer and
the upper layers of the biofilm separately to monitor gene expression of the two
stationary-phase genes, again, at various time points (Fig. 4.3C, D). We found that, in
flow conditions, the level of expression of both the genes in the upper layers of the
biofilm decreased over 15 days. When gene expression changes were monitored in the
bottommost adhered cells of the biofilm, a different pattern of expression was revealed.
Expression of both the genes was observed on day 1 and increased over the 15 days.
64
FIG. 4.3. Expression of SNZ1 and SNO1. RT-RTPCR analysis of SNZ1 and SNO1
transcripts was determined and is shown as copy number. Expression patterns of SNZ1
(open triangles) and SNO1 (solid triangles) in planktonic conditions (A), in static biofilm
conditions (B), in upper layers (C) and in bottommost adhered cells (D) of the biofilm
formed under flow conditions, are shown.
Protein localization of Snz1p–YFP and Sno1p–YFP in planktonic cells. We used YFP
cassettes to tag SNZ1 and SNO1 in C. albicans, and observed the localization and
expression of the two encoded proteins under different growth conditions and
morphologies. Fluorescence was observed for both proteins in yeast cells (Fig. 4.4A, B).
The proteins were then localized within the cells. When bright-field and fluorescent
images were compared visually, fluorescence could be easily localized within the
cytoplasm (Fig. 4.4A, B). The two images were pseudocoloured green and red using the
Meta Morph software and then merged (Fig. 4.5C). This method confirmed that the two
proteins localized to the cytoplasm.
65
Fig. 4.4. Expression of Snz1p–YFP by yeast cells and pseudohyphae. Yeast cells from a
5-day culture (A, B) and pseudohyphae from an 11-day culture (C, D) were examined for
organisms exhibiting fluorescence using bright-field (A, C) and fluorescence (B, D)
imaging. Arrows indicate non-fluorescent buds being formed from fluorescent
pseudohyphae. Sno1p–YFP showed a similar expression pattern and is not included in
the figure. Bar, 5 µm.
66
Fig. 4.5. Localization of Sno1p–YFP in yeast cells. Sno1p–YFP yeast cells were
examined using bright-field (A) and fluorescence imaging (B). The two images were
pseudocoloured green and red, respectively, and merged using the Meta Morph software
for localization (C). Thin arrows indicate non-fluorescent vacuoles and block arrows
indicate non-fluorescent cell wall. Snz1p–YFP showed a similar localization pattern and
is not included in the figure. Bar, 30 µm.
Protein expression in planktonic cells Planktonic yeast cells began expressing both
proteins after 3 days in culture (Fig. 4.6A). About 25 % of cells expressed Snz1p–YFP on
day 4 and >90 % expressed the protein on day 5. On day 6, there was a greater than
threefold decrease in the fluorescent cells. Expression of Sno1p–YFP began a day later
than Snz1p–YFP. About 40 % of cells expressed Sno1p–YFP on day 5, while >90 %
expressed it on day 6. The expression of Sno1p–YFP did not decrease in the subsequent
7 days (data not shown). We next examined the fate of fluorescence when yeast cells
resumed growth. Fluorescent cells, expressing either gene when inoculated into fresh
medium, lost fluorescence within 24 h (Fig. 4.6B).
67
Fig. 4.6. Expression of Snz1p–YFP (solid squares) and Sno1p–YFP (solid diamonds)
during progression into and exit from the stationary phase. Cultures of each strain were
grown in YNB and the proportion of fluorescent cells was determined daily for 7 days
(A). Five-day-old cultures of strains expressing either Snz1p–YFP or Sno1p–YFP were
diluted and resuspended in fresh medium and the proportion of fluorescent cells was
determined at various intervals (B).
Protein expression in biofilm organisms. We first examined expression of both Snz1p–
YFP and Sno1p–YFP in a static model of biofilm formed on acrylic placed in the well of
a polystyrene plate. On day 6, there were more fluorescent cells (P 0.01) in the
bottommost layer of adhered cells (25 %) than in the upper biofilm layer (11 %). Biofilms
were formed in the second model system under flow conditions in which medium was
continuously replenished (Fig. 4.7A–C). No fluorescence was observed in the uppermost
0
20
40
60
80
100
1 2 3 4 5 6 7 Days
A
0
20
40
60
80
100
120
0 3 7 11 15 18 21 24
Hours
Fluo
resc
ent c
ells
(%)
B
Fluo
resc
ent c
ells
(%)
68
layer which mainly contained hyphae, even though hyphae had been present in the
biofilm from the first day. A few fluorescent organisms were observed in the middle
layer, which had mixed morphologies. As in the static biofilm formation, 25 % of the
bottommost adhered cells were fluorescent at day 6; the additional days of growth in the
flow system showed that the number of fluorescent cells increased to 40 % on day 15.
The bottommost layer of adhered yeast cells recovered from the biofilm retained 88 %
viability (1x107 out of 1.2x107 cells), even up to 20 days.
Fig. 4.7. Expression of Sno1p–YFP in different layers of a 6-day-old biofilm. Three
layers, bottommost adhered (A), middle (B) and top (C) were separated from a
continuously fed biofilm and examined by bright-field (top row) and fluorescent (bottom
row) microscopy. Snz1p–YFP showed similar expression patterns and is not included in
the figure. Bar, 5 µm. Approximately 40% of the bottom-most adhered layer of yeast
cells, 11% of the middle layer and 1% of the top-most hyphal layer showed fluorescence.
69
Discussion
C. albicans forms a structurally complex biofilm. A mature 36–48 h-old biofilm formed
in YNB medium contains the three major morphological forms of C. albicans: yeast,
hyphae and pseudohyphae (Baillie & Douglas, 1999; Chandra et al., 2001; Ramage et al.,
2001). This 450 nm-thick mature biofilm is also interspersed with water channels and is
sheltered by an ECM. Thus, such a varied, closely packed community of cells may lead to
a gradient of environmental conditions within the biofilm, in which the C. albicans cells
may enter distinct physiological states. The goal of this study was to determine if C.
albicans cells in a biofilm reach a physiologically similar state to that of planktonic
stationary-phase C. albicans cells. However, the stationary phase in C. albicans has not
been characterized, and our first steps were to confirm that cells remained viable in
planktonic culture after the increase in cell number ceased (Fig. 4.2), and to identify a
marker for the planktonic C. albicans stationary phase. Drawing on some paradigms
represented by S. cerevisiae, we initiated a study to verify the expression patterns of the
C. albicans orthologs of the two most studied stationary-phase genes in S. cerevisiae,
SNO1 and SNZ1. In S. cerevisiae, there are three pairs of SNO and SNZ genes. The genes
of each pair are adjacent and divergently transcribed (Balakrishnan et al., 2002). The pairs
are coordinately regulated with both the SNO2–SNZ2 and SNO3–SNZ3 pairs which are
expressed prior to diauxic shift and the stationary phase (Arnaud et al., 2005). Only a
single pair of SNO–SNZ genes was found in C. albicans, and they were adjacent and
divergently transcribed.
We found that, in planktonic-grown C. albicans, the expression of SNZ1 and
SNO1 appeared during entry into the stationary phase, peaking several days later
(Fig. 4.3A). Expression of SNZ1 peaked on day 5, 1 day before SNO1 peak expression,
and the level of SNZ1 expression and the magnitude of increase were greater than those of
SNO1. This paralleled the observations in S. cerevisiae for SNZ1 and SNO1 (Braun et al.,
1996). However, in a mutant strain of S. cerevisiae, in which SNO1–SNZ1 is the only pair
of genes present, the genes are expressed prior to diauxic shift (Braun et al., 1996). Based
on this analogy, we might have expected the C. albicans SNO1 and SNZ1 expression to
70
parallel that of the S. cerevisiae mutant strain. This was not observed. Thus, it would
seem that the function of SNZ and SNO genes prior to the stationary phase is dispensable
in C. albicans. Snz1p–YFP expression was detected at 3 days (Fig. 4.6), perhaps
reflecting the transient increase in transcript level seen on day 2 (Fig. 4.3A). The peak
protein expression was observed on day 5, coincident with peak transcript level
(Fig. 4.4A, B). This suggests that the increase in transcript level is derived from an
increase in most cells of the population rather than in only a few cells. Sno1p–YFP
expression began increasing on day 2 (Fig. 4.6A), at the same time that transcription level
showed a small decrease. Maximum expression was reached on day 6, coincident with the
peak transcription level. Unlike Snz1–YFP, Sno1p–YFP continued to be observed in
cells, even though the transcription level began to decrease on day 7. The greater stability
of Sno1p–YFP compared to that of Snz1p–YFP may explain the increase in the number of
fluorescent cells at low transcript levels, as the protein accumulates and the fluorescent
cells persist even when the peak transcript level declines. However, when stationary-
phase planktonic yeast cells resumed growth, the number of fluorescent cells began
decreasing immediately, such that only a few fluorescent cells were detected in the
growing culture (Fig. 4.6B). The fluorescent cells decreased at a similar rate for both
proteins, suggesting that, when the cell resumed growth, the proteins expressed for the
stationary phase were lost. When protein expression was examined in hyphae, no
fluorescence was observed (data not shown). Subapical compartments are arrested in G1
phase, are extensively vacuolated, and have very little cytoplasm (Barelle et al., 2003).
Two possibilities for the lack of fluorescence in hyphae are that the G1-arrested, non-
growing state of subapical hyphal cells is different from that of the G1 stationary-phase
yeast cells, or that the expression in the small amount of cytoplasm of these subapical
cells is below the level of detection. When pseudohyphae were observed in planktonic
yeast cultures, they were fluorescent, but daughter buds were not. Since the buds were
growing, this is consistent with the loss of fluorescence when cells resume growth, and
also suggests that the partition of cytoplasm between parent and daughter cells did
include the same level of stationary-phase protein found in the mother cell.
71
As expression of these two genes is a marker for stationary-phase planktonic yeast
cells, we used the two genes to determine if cells within a biofilm reach a physiological
state in which these genes are expressed. Biofilms were formed on denture acrylic under
static conditions and under conditions of continuous medium flow. When formed under
static conditions, expression increased over the 6 days of observation. As in planktonic
yeast cells, SNO1 was expressed at a higher level than SNZ1 (Fig. 4.3B). However, both
genes were expressed at 4 % of the maximum expression of planktonic yeast cells.
Under conditions of medium replenishment, in which the biofilm could be observed for
>2 weeks, expression was determined in the upper layers of the biofilm and in the cells
adhered to the substrate (Fig. 4.3C, D). Expression from the upper-level biofilm
organisms decreased by day 6 and remained at lower levels. In contrast, expression in the
adhered cells increased and was about 100-fold higher than that in the upper layers and 5–
11-fold higher than that in the static biofilm. At 15 days, levels of SNZ1 and SNO1 in the
adhered cells were only five and 2.4 fold less than their peak expression levels in 5 and 6
day old planktonic cells, respectively. These adhered cells are likely to be founder cells of
the biofilm and therefore older, and may show less proliferation than cells at the
periphery of the biofilm.
When protein expression was examined in cells from different portions of the
biofilm, the organisms at the top were almost exclusively hyphal and non-fluorescent, as
seen in planktonic cultures (Fig. 4.7). Most of the fluorescent cells were found adhered to
the substrate. The number of fluorescent cells increased between days 3 and 6, as did
gene expression (Fig.4. 3). The 15 day continuously fed biofilm, with 40 % of the
adhered yeast cells showing fluorescence and a 2.5–5-fold reduction in gene expression
level for the cell population, suggests that the level of expression in the fluorescent cells
may be similar to that of planktonic, 5–6-day stationary-phase yeast cells. Although
hyphae were present in the biofilm from day 1, no fluorescence was observed and, as with
planktonic hyphae, this may have arisen from a difference in the G1 state of subapical
compartments, or an inability to detect fluorescence in the reduced cytoplasm of these
cells.
72
The proteins were localized in the cytoplasm (Fig. 4.5). In a genome-wide S.
cerevisiae study, Snz1p could not be localized by green fluorescent protein (GFP) fusion,
due to low GFP expression signals or to other technical difficulties, while a low-level
cytoplasmic fluorescence was noted for Sno1p (Huh et al., 2003). The greater level of
fluorescence in this study may reflect a higher level of expression of these proteins in C.
albicans, or the successful tagging of the protein.
In summary, C. albicans has a single pair of SNZ and SNO genes that was
expressed in the stationary phase of planktonic yeast cells but not in hyphae. Proteins
were localized in the cytoplasm and >90 % of 5 and 6 day stationary-phase yeast cells
expressed the proteins. Expression of these genes was less in biofilms, whether formed
under static or medium-flow conditions. Expression of the genes increased during biofilm
formation and was primarily associated with founder yeast cells adhered to the substrate.
This finding suggests that some cells at the base of a biofilm either were in the stationary
phase or had reached a physiological state in which genes associated with the stationary-
phase planktonic yeast cells were expressed.
73
CHAPTER V
EFFECT OF FARNESOL AND CONDITIONED MEDIUM ON CANDIDA
ALBICANS GENE EXPRESSION AND YEAST GROWTH
Abstract
During C. albicans yeast cell growth to early stationary phase, metabolites
accumulate in the medium, including the quorum sensing molecule farnesol. Both
farnesol and 75% conditioned medium (CM) inhibited germ tube formation while >100
�M farnesol delayed resumption of yeast cell growth. Transcriptional analysis of yeast
cells resuspended in fresh medium with or without 40 �M farnesol or in 75% CM under
germ tube induction conditions revealed differential expression of 406 genes. Farnesol
upregulated genes were involved in lipid metabolism, mitochondria and peroxisome
maintenance and nucleic acid transport. Besides hyphal genes, the downregulated genes
encoded GTPase activators, proteins involved in mitosis, DNA replication and adherence.
Genes involved in nuclear division and microtubule organization were upregulated, while
those coding for ribosomal proteins were downregulated in presence of 75% CM.
Farnesol mediated delay in resumption of yeast growth was relieved by addition of a
diacylglycerol analogue, implicating phosphatidylinositol signaling in the delay.
Diacylglycerol is an activator of protein kinase C (PKC) in mammalian cells; however,
fungal PKCs are not responsive and this was confirmed with a C. albicans PCK1 mutant.
Introduction
C. albicans is a commensal of the human oral, gastrointestinal, vaginal and
cutaneous surfaces. However, when the balance of the normal flora is altered, during
antibiotic or hormonal therapy, or in conditions when the skin is exposed to moisture for
prolonged periods of time, C. albicans can cause painful cutaneous or subcutaneous
infections such as, vaginitis, oral thrush, diaper rash, conjunctivitis, or infections of the
nail and rectum. In immunocompromised individuals, such as immunosuppressed
patients undergoing cancer chemotherapy, C. albicans can be responsible for life
74
threatening diseases only when it enters the blood stream. It is then capable of affecting
almost any part of the body and causing hepatosplenic abscesses, myocarditis, central
nervous system or pulmonary infections. Additionally, C. albicans can form biofilms on
host surfaces as well as abiotic device surfaces such as dentures and catheters
(Kumamoto and Vinces 2005; Mukherjee, Zhou et al. 2005). Catheter-related infections
due to Candida albicans biofilms are a leading cause of fungal nosocomial bloodstream
infection. Biofilm cells are resistant to most antifungal agents making biofilm related
infections hard to treat.
In vitro, planktonic C. albicans yeast cells in suspension culture, after having
reached a certain concentration, stop growing and enter stationary phase. During growth
to stationary phase, cells release metabolites into the medium, including molecules that
may have a quorum sensing function (Hornby, Jensen et al. 2001; Chen, Fujita et al.
2004). In an in vitro C. albicans biofilm, some yeast cells in the bottom-most adhered
layers of the mature biofilm reach stationary phase (Uppuluri, Sarmah et al. 2006).
Mature biofilms also produce quorum sensing molecules (Ramage, Saville et al. 2002).
The culture supernatant or CM may be recovered by filter sterilization (Hornby, Jensen et
al. 2001; Ramage, Saville et al. 2002; Chen, Fujita et al. 2004; Westwater, Balish et al.
2005). CM has been shown to abolish lag phase, stimulate germ tube formation, inhibit
germ tube formation, and protect against oxidative stress (Hornby, Jensen et al. 2001;
Chen, Fujita et al. 2004; Westwater, Balish et al. 2005). Farnesol and tyrosol have been
purified from conditioned medium and shown to mediate such sometimes contradictory
activities. Tyrosol accelerates the appearance of germ tubes in germ tube permissive
conditions and reduces lag phase in C. albicans (Chen, Fujita et al. 2004). The effects of
farnesol, another component has recently been reviewed (Nickerson, Atkin et al. 2006).
Farnesol prevents yeast cell to germ tube transition and in turn inhibits biofilm formation
(Hornby, Jensen et al. 2001; Ramage, Saville et al. 2002; Hornby, Kebaara et al. 2003).
Microarray analysis (complete genome) comparing farnesol treated and untreated
planktonic C. albicans cells in germ tube permissive conditions, reported a number of
genes involved in processes such as hyphal formation, mitosis, fatty acid metabolism,
75
stress resistance and DNA damage (Enjalbert and Whiteway 2005). Microarray analysis
(3102 Orfs) for the effect of farnesol on C. albicans biofilms revealed that many similar
biological processes were affected (Cao, Cao et al. 2005). Farnesol contributes to the
protection from oxidative stress (Westwater, Balish et al. 2005). While a higher
concentration of farnesol (100 �M) is reported to have a delaying effect on C. albicans
resumption of yeast growth in glucose salts medium (Kim, Kim et al. 2002). At even
higher concentrations, no effect was observed when cultures were observed after many
hours of growth (Hornby, Jensen et al. 2001; Ramage, Saville et al. 2002). Thus, the
metabolites in the CM have various effects on cells depending upon the environmental
conditions and suggesting a complex cell response.
Using Yeast Nitrogen Base (YNB) medium we confirmed that 75% CM and
farnesol (40 �M) prevented C. albicans germ tube formation. However, since CM
contains a metabolite mixture, the cellular response may differ in some aspects from that
of farnesol alone. To determine if there was a difference in gene expression when germ
tubes were induced in the presence of CM or farnesol or medium alone, we used
oligonucleotide microarrays. A number of genes were found to be differentially
expressed between the three conditions, including genes involved in hyphal formation,
DNA replication and mitosis, ribosome biogenesis and phosphatidylinositol type
signaling. As phosphatidylinositol signaling has been implicated in control of cell wall
integrity through protein kinase C (PKC), we examined the effect of farnesol and PKC
activators on yeast cell growth and on a PKC1 deletion mutant.
Methods
Organism and growth conditions. C. albicans strain SC5314 was maintained on YPD
(yeast extract 1% w/v, peptone 2% w/v, dextrose 2% w/v) agar plates and transferred to
YNB (Yeast Nitrogen Base medium with amino acids, Difco Laboratories, Detroit, MI)
with 50 mM glucose for suspension culture with shaking (180 rpm) at room temperature
(RT). The cells were recovered after 24 hours; subcultured in YNB medium and
incubated overnight at RT. CM was prepared by centrifugation followed by sterile
76
filtration to remove cells. Different dilutions of CM were prepared by mixing appropriate
quantities of conditioned medium with fresh YNB medium. E,E-farnesol obtained as a
3.7 M solution (Sigma Chemical Co. St. Louis, MO) was diluted in 100% (vol/vol)
methanol to obtain a 40 mM stock solution. Germ tubes were induced by inoculating
yeast cells (1x106 cells/ml) from the 24 h culture into YNB medium, 75% CM and into
YNB medium containing 40µM farnesol. The cultures were incubated at 370C with
shaking for 2 h. To study the effect of farnesol on yeast cell growth, cells were grown for
24 hours at RT in YNB medium containing 10 µM, 25 µM, 40 µM, 100 µM and 300 µM
farnesol. In agreement with other reports (Romandini, Bonotto et al. 1994; Yazdanyar,
Essmann et al. 2001; Ramage, Saville et al. 2002), methanol alone at 0.75%
(concentration at 300 µM farnesol) had no effect on growth (data not shown). The pH of
some cultures was determined by measurement of cell free culture medium. In some
experiments, farnesol was added after the cells had grown to a concentration of 1x107
cells/ml. C. albicans strains bearing a mutation in the protein kinase C gene, PKC1, was
obtained from Dr. Aaron Mitchell, Columbia University. The parent C. albicans strain
DAY185 and the PKC1 mutant strain were grown in YPD containing 1M sorbitol and
subjected to different concentrations of farnesol and incubated as above. Diacylglycerol
cell permeable analog, 1-oleoyl-2-acetyl-sn-glycerol (OAG) (Sigma Chemical Co.) was
dissolved in DMSO to obtain a stock concentration of 2.5 mM. OAG (25 µM) was added
to 12 h old and 48 h old cultures containing different concentrations of farnesol. The
proportion of germ tubes was determined microscopically by counting at least 200
organisms using a haemocytometer. Cell size (for at least 1x106 cells/ml) and in some
cases cell density were measured using a Z series Coulter counter (Beckman Coulter,
Fullerton, CA). Protein kinase C activator phorbol 12-myristate 13-acetate (PMA) was
obtained as a 10mM stock solution from Dr. Thomas Pressley, Texas Tech University
Health Sciences Center. PMA (0.5 nM to 50 µM) were added to cultures containing 100
µM and 300 µM farnesol and the cell number counted using a haemocytometer.
77
Analysis of cellular DNA by fluorescence flow cytometry. Cells for flow cytometry were
prepared as described elsewhere (Chapter III).
RNA extraction Total RNA was isolated using the standard hot acid phenol method
following grinding frozen cells using a mortar and pestle in liquid nitrogen (Uppuluri et
al., 2006a; Chapter II). The RNA preparation was DNAse treated and the absence of
DNA contamination was confirmed with the housekeeping gene EFB1 (Maneu, Martinez
et al. 2000). RNA quality and quantity were determined as described (Uppuluri, Sarmah
et al. 2006).
Transcriptional analysis: The protocol for transcriptional analysis is described in detail in
Chapter III.
Real time RT-PCR (RT-RTPCR). The amount of mRNA in the total RNA was
quantitated with the Poly(A) mRNA Detection System kit. (Promega, Madison, WI).
cDNA was synthesized from known amounts of mRNA, and equal amounts of cDNA
were used as starting template for RT-RTPCR. Analysis of transcript was carried out
using SYBR Green PCR Master Mix (Applied Biosystems, Foster City, CA) in ABI
Prism 7700 Sequence Detection System (Applied Biosystems). Each reaction was set up
in triplicate in 25.0 µl volume with 1.0 µl of cDNA for 40 cycles (thermal cycling
conditions: initial steps of 50o C, 2 min and 95o C, 10 min; and then, 40 cycles of 95o
C,15 sec; 60o C, 1 min). The primers are given in Table 5.1. Relative gene expression was
quantified using the CT method (Zhao et al., 2005). The target genes were normalized
to total mRNA. The fold change was calculated for each sample using the equation,
. Each RNA replication was treated separately and the results were averaged after
the calculation of each PCR run.
78
Table 5.1. Primers for analysis of selected genes by RT-RTPCR. The direction of
primers is indicated as forward (F) or reverse (R).
Gene (systematic name)
Primer Sequence (5’-3’)
BEM3 BEM3F GTATGCAGTTATCACCAACTA (orf19.2771) BEM3R AGAGCTCCAGAAGCTTTGTTC HST2 CHS2F TAATAAATTCCGCAATACGCCTAAC (orf19.2580) CHSR TAGTGGCACACATTCTCTTTCATTTT RDS1 EFB1F ACGAATTCTTGGCTGACAAATCA (orf19.4767) EFB1R TCATCTTCTTCAACAGCAGCTTGT ADR1 IRA2F CCTTGATACAAAGTCGAGCTTAGGA (orf19.2752) IRA2R TAGGAGCTGTTGGCCAGGTATT SSA2 OXR1F TCGTCACATTCTAGTGTTTCTAGTCTG (orf19.1065) OXR1R TAGTAATCGATGATGAGTTGATTCTT HST7 PNC1F AACTTGACCCGAAAACGAATCA (orf19.469) PNC1R AGCTCCCTTGGTGCCTTGTAC MCA1 RAD50F CAGGGACATTGCCTCCAAAT (orf19.5995) RAD50R CAGTTACAGCAGTTCGAGAGCTTAAG COX11 TDH3F AGGACTGGAGAGGTGGTAGAACTG (orf19.1416) TDH3R AATAACCTTACCAACGGCTTTAGC EFB1 (test for DNA contamination) (orf19.3838)
EFB1F EFB1R
ACGAATTCTTGGCTGACAAATCA TCATCTTCTTCAACAGCAGCTTGT
Results and Discussion
Effect of farnesol and CM on germ tube induction. Farnesol and CM have previously
been reported to inhibit germ tube formation in growth supporting and non-supporting
conditions (Hornby, Jensen et al. 2001; Ramage, Saville et al. 2002; Hornby, Kebaara et
al. 2003). We confirmed that this effect extended to C. albicans planktonic yeast cells
grown in YNB medium. When germ tubes were induced in YNB at 37oC, >90% of the
cells formed germ tubes in 2 h. The addition of 40 µM farnesol inhibited >90% germ
tube formation. The effect of 25%, 50% and 75% CM was examined and 75% CM
completely suppressed germ tube formation. Farnesol is produced in YNB during growth
of yeast cells in suspension culture (Hornby, Jensen et al. 2001) and is presumably the
major effector compound among the CM metabolites suppressing germ tube formation.
However, as noted previously other metabolites may also effect yeast cell responses
under different conditions (Chen, Fujita et al. 2004), and metabolites in addition to
farnesol may have some effect on the response in normally germ tube inducing
conditions. Hornby et al (Hornby, Jensen et al. 2001) reported with a different medium
that about 71% conditioned or spent medium completely inhibited germ tube induction
79
under non-growing conditions while about 75% inhibition was attained with 75% CM
under conditions supporting growth. Whether the differences between the effect of CM
on germ tube induction in this study and that previously reported (Hornby, Jensen et al.
2001) is significant and reflects the medium used is unknown.
Alteration of gene expression in response to farnesol and CM. Since hyphal formation
involves a change in gene expression as shown in numerous studies including global
transcription approaches (Murad, d'Enfert et al. 2001; Murad, Leng et al. 2001; Nantel,
Dignard et al. 2002; Garcia-Sanchez, Mavor et al. 2005; Kadosh and Johnson 2005;
Singh, Sinha et al. 2005), we considered that repression of hyphal formation would also
alter the global transcript profile. Microarray analysis was used to analyze difference in
expression when C. albicans was grown in fresh medium with or without 40 �m farnesol
or 75% CM under germ tube permissible conditions for 2 h. There were 406 genes
differentially regulated between the three groups (ANOVA, p<0.05) (Fig 5.1). Of these
genes 100 were of unknown functions. When compared to medium alone, farnesol
addition resulted in more than twice as many unique changes (136 genes upregulated, 99
genes downregulated) as did CM (47 genes upregulated, 42 genes downregulated). In
both conditions 31 genes were upregulated and 51 downregulated compared to YNB
only. Farnesol is expected to accumulate in medium to 10-50 µM (Hornby, Jensen et al.
2001). Farnesol in CM may differ from cultures with exogenously added farnesol and
the effect of concentration dependent differences is not known. In addition, CM is more
complex and contains other metabolites that may affect the cell. A Venn diagram of the
number of upregulated genes that are unique to or common between the three conditions
is shown in Fig. 5.1.
To support the differential expression observed with global transcriptional
analysis, we picked eight genes for analysis by RT-RTPCR (Fig. 5.2). We found that the
gene expression pattern obtained by RT-RTPCR corroborated the differential expression
observed by microarray analysis. However, for some genes, the magnitude of fold change
was higher in RT-RTPCR than microarray analysis. This could reflect the greater
sensitivity of RT-RTPCR.
80
Figure 5.1. Venn diagram of the number of upregulated genes that are unique to, or
common between the three conditions, Farnesol (F), CM (C) and control medium (M).
Figure 5.2. RT-RTPCR verification of genes differentially expressed in Farnesol (F)
group and CM (C) group, obtained by microarray analysis. Results are shown as fold
change (x-axis) compared to the control group. ** p < 0.01, * p< 0.05. Table on the
right shows microarray fold change of the individual genes in the F and C group
compared to the control group (M). + No significant difference between the test and the
control.
81
Activities and pathways affected by farnesol and CM addition. The differentially
regulated genes were analyzed for cellular processes and pathways affected by the
addition of farnesol and CM. There was some overlap in both upregulated (31) and
down regulated (51) genes between the conditions (Table 5.2). Since germ tube formation
was inhibited in both cases, we expected to find overlap in reduced expression of genes
associated with hyphal formation. Hyphal genes involved in the MAP (mitogen-activated
protein) kinase pathway, RAS1, HST7 and GAP1 (Whiteway 2000) were downregulated
greater than 2 fold. This is in agreement with a previous study that showed similar down
regulation during farnesol mediated inhibition of germ tube formation and suggested that
farnesol acts through suppression of MAP kinase signaling (Sato, Watanabe et al. 2004).
In our study, other downregulated hyphal genes were the adenylate cyclase associated
gene CAP1, the ribosomal gene RPS26A that is regulated by other hyphal genes NRG1
and TUP1 and the succinate metabolism gene SDH4. Hyphal formation genes involved in
the cAMP-EFG1 pathway (Whiteway 2000) such as EFG1, HYR1 and HWP1 were not
differentially expressed under any condition. Lack of differential expression of these
hyphal genes in germ tube inhibiting conditions such as treatment with farnesol has been
previously reported (Sato, Watanabe et al. 2004). However, a recent report by Enjalbert
et. al. (Enjalbert and Whiteway 2005) found these genes upregulated in cells treated with
farnesol. They also found a few other hyphal genes upregulated that showed no change in
gene expression in our study. This could be due to differences in the growth medium used
in the different studies. While Enjalbert et al REFused a rich medium (YPD) for their
study, we used a synthetic defined medium (YNB) for our studies.
Besides hyphal genes, genes involved in C. albicans biofilm formation were
significantly downregulated; while those involved in osmotic, heat or oxidative stress
resistance, aging and nutrient sensing and regulation of apoptosis were significantly
upregulated in the presence of both farnesol and CM (Table 5.2). C. albicans CM and to
some extent farnesol, are known to induce antioxidant genes and thus protect yeast cells
from oxidative stress (Westwater, Balish et al. 2005). Stress resistance genes are known
to be upregulated when germ tubes are induced in the presence of farnesol in YPD
82
medium (Enjalbert and Whiteway 2005). The apoptotic effect of farnesol has been shown
in Aspergillus nidulans as well as mammalian cells (Voziyan, Haug et al. 1995;
Semighini, Hornby et al. 2006)
When the cells were recovered for microarray analysis, the pH of all the three
growth medium was acidic (conditioned medium – 2.3, farnesol – 2.3 and control
medium – 4). All of these pH values were below or at pH 4, the acidic pH frequently
used to demonstrate pH-dependent regulation (Davis, Edwards et al. 2000; Davis, Wilson
et al. 2000). Hence, we consider that the gene expression changes that were recorded
were not a consequence of difference in the pH of the media. In addition we did not find
any of the pH regulated genes being differentially expressed in our microarray results.
Gene expression in the presence of CM. A total of 89 genes were differentially expressed
exclusively in the 75% CM (Fig 5.1). Expression of a few genes encoding DNA
replication machinery and those involved in cell division have been previously found to
be upregulated in the presence of tyrosol (a component of CM) (Chen, Fujita et al. 2004).
This upregulation was the reason behind CM mediated abolishment of C. albicans lag
phase (Chen, Fujita et al. 2004). Our microarray study identified several additional genes
involved in similar functions of DNA replication, microtubule organization and nuclear
division, when cells were grown in the presence of 75% CM (Table 5.2). Other genes
upregulated in the presence of this condition had functions in negative regulation of
transcription, mRNA catabolism, filamentous growth regulation, histone deacetylation
and stationary phase.
The largest category of genes downregulated in the presence of 75% CM was the
ribosomal protein genes (Table 5.2). Global transcriptional analysis of C. albicans yeast
to hyphae transition revealed that protein synthesis genes including genes coding for
ribosomal proteins were a major category of upregulated genes (Singh, Sinha et al. 2005).
Hence, downregulation of the ribosomal proteins in the presence of CM is congruent with
the inhibition of germ tubes by that treatment. We also found the histone deacetylase
genes being upregulated in the presence of CM. Upregulation of histone deacetylases
83
(Perrod, Cockell et al. 2001; Lamming, Latorre-Esteves et al. 2005) and downregulation
of ribosomal genes are characteristics of aging and/or stationary phase cells of S.
cerevisiae (Motizuki and Tsurugi 1992). CM with its reduced nutrients and no glucose as
shown in Chapter III could possibly trigger some stationary phase genes in C. albicans.
Gene expression in the presence of 40 µM farnesol. A total of 136 genes were
upregulated and 99 genes downregulated under germ tube permissive conditions in the
presence of 40 µM farnesol. The genes upregulated had functions in ion transport, nucleic
acid transport and lipid metabolism. Two genes involved in peroxisomal matrix
organization (ADR1, orf19.164) were upregulated in farnesol conditions The entire
pathway for farnesol biosynthesis is located in the peroxisome, and farnesol has to be
transported out of the peroxisome for further metabolism (Westfall, Aboushadi et al.
1997; Aboushadi, Engfelt et al. 1999). These processes have been shown earlier to be
affected by the presence of farnesol (Cao, Cao et al. 2005; Enjalbert and Whiteway
2005). Another organelle that was most affected in response to farnesol was the
mitochondrion. We found upregulation of genes that coded for mitochondrial cytochome
oxidase C assembly proteins (COX7, COX11), mitochondrial ATP synthases (ATP7, and
ATP10), a mitochondrial maintenance protein (AMI3), a mitochondrial ion transporter
(LPE10), a mitochondrial oxidative stress resistance protein (FMP27) and a
mitochondrial phosphate carrier protein (PIC2) upregulated in the presence of farnesol.
In S. cerevisiae and Aspergillus nidulans, mitochondria protect cells against oxidative
stress and mediation of apoptosis after treatment with farnesol (Machida, Tanaka et al.
1998; Machida, Tanaka et al. 1999). We found the drug resistant gene – RDS1,
upregulated in farnesol condition. Cao et. al had found a different drug resistance gene
FCR1 upregulated in farnesol treated C. albicans biofilm cells (Cao, Cao et al. 2005). On
the other hand, Enjalbert et. al (Enjalbert and Whiteway 2005) found CDR1 and CDR2
upregulated when planktonic C. albicans cells were treated with 30 µM farnesol.
Although the experimental design of our study was similar to this, we used a different
growth medium for cell growth, a different temperature for growing the inoculum for
84
germ tube induction and a higher concentration of farnesol for our study. These could be
a few reasons for the difference in gene expression in the above two similar studies.
A large category of genes differentially regulated in farnesol conditions were
those involved in mitosis, cell proliferation and DNA replication. The genes falling in
these categories were significantly downregulated in the presence of farnesol. These
genes played a role in reorganization of chromosomes during cell division, G1/S
transition or G2/M transition of the mitotic cycle, bud site selection, GTPase activation,
DNA strand elongation and DNA damage recognition during mitosis (Table. 2). It is
reported that under germ tube permissive conditions, in rich medium, C. albicans down
regulates genes involved in the G2/M phase of the mitotic cycle and in cell wall
organization and biogenesis, after a hour of farnesol treatment (Enjalbert and Whiteway
2005). Growth inhibition due to interference in DNA replication was also observed when
S. cerevisiae cells were treated with farnesol (Machida, Tanaka et al. 1999). Other genes
downregulated in the presence of farnesol were those having functions in hyphal
induction (YHB5, RAS2, HST7, GAP1, CAP1, RPS26A, SDH4, URA2, SEC23), those that
were regulated by GCN4 or GCN2 in response to amino acid starvation (LEU42, LYS4)
(Tournu, Tripathi et al. 2005) or those regulated in the presence of hemoglobin (HBR1
and the above mentioned three genes). Also downregulated were genes involved in
adhesion (ALS3), heat shock protein HSP90 and a HSP70 chaperone gene SSA2.
Downregulation of some of these genes were also reported when C. albicans biofilm cells
were treated with farnesol (Cao, Cao et al. 2005).
85
Table 5.2. Genes differentially expressed in the presence of farnesol or CM compared to
unsupplemented medium
GENES AND FUNCTIONS FARNESOL (Z – score) CONDITIONED MEDIUM (Z – score)
Upregulated Nuclear division (4. 5) GAC1, SPO7, RDH54, SRC1, MAD1, AMN1 Beta – 1,6 glucan metabolism (2.9) KRE6, KRE9 Clathrin binding (2.9) APM1, YAP180 Microtubule based preocess (2.9) TUB4 – microtubule nucleation CIN1 – beta tubulin folding ASE1, KIP3, RRD2 – mitotic spindle assembly N – methyltransferase activity (2.9) SWD2, ABD1 Other up regulated genes SKI2 - mRNA catabolism SSL2 – negative regulator of transcription FGR29, SPT6 – filamentous growth regulation Orf 19.1676 – stationary phase gene Downregulated Ribosomes and ribosome biogenesis (2.9) RPS26A, RPS16A, RPS12, RPS15, YST1, DBP7 Other down regulated genes ARP2 – Membrane growth Orf 19.2394, orf19.5877 - Alcohol metabolism CDC68, PSH1 – RNA polymeraseII elongation factors CONDITIONED MEDIUM AND FARNESOL
Upregulated Nucleic acid transport (2. 53) AIR1, CRM1 – RNA export from nucleus NUP146, NUP60 – Nuclear pore maintenance Peroxisome organization and biogenesis (3.0) ADR1 – regulator of genes involved in ethanol, glycerol and fatty acid utilization Orf 19.164 – Peroxisome organization Other upregulated genes PHO87, PHO88 – Inorganic phosphate transport FET3, CNH1 - ion transport COX7, COX11, ATP7, ATP10, AMI3, LPE10, FMP27, orf19.1395 – Mitochondria associated genes orf 19.4767 – Drug resistance gene, xenobiotic stress resistance ASK10, CCP1 – response to oxidative stress Downregulated Mitosis (3.7) SMC4, SRC1 – organization of chromosomes PTC2, HPC2, PIN4 - G1/S, G2/M transition BUD27, EXO70 – Bud site selection RAS2,GLO3, GEA2, BEM3, SEC23 – GTPase activators SLD2, POL1, POL30 – DNA strand elongation RDH54, RAD14 – DNA repair during mitosis Actin filament organization (3.0) CCT5, CAP1, SAC6, SCD5, BEM3, ACT1 Phospholipid metabolism (3.0) PIS1 – Phosphotidyl inositol synthase INO1 – myo inositol VPS15 – phosphorylation of PI 4 kinase BST1, GIT2, LSB6 – metabolism of GPI anchors PLB1, PLB3 – Phospholipase B Other downregulated genes URA2, SEC23, ALS3/8 – filamentation and host adherence Orf19.4525, QCR6 – Ubiquinone oxidoreductase SSA2, HSP90 – Heat shock proteins YHB3, HBR1, LEU42, LYS4 – Hemoglobin regulated ATF1, RHR2, GPI7 - Lipid metabolism
Upregulated genes RPD2, BNA6, DCS1 – Aging and nutrient sensing CTA8, RRD2, SLN1– stress resistance MCA1 - apoptosis Downregulated genes RAS2, HST7, GAP1, CAP1, RPS26A, SDH4 - Hyphal genes ILV3, MET15 – Biofilm formation
86
Effect of farnesol on C. albicans growth. Microarray results revealed that genes involved
in cell growth or mitosis were down regulated when early stationary phase cells were
treated with 40 µM farnesol. We questioned to what extent such changes were reflected
in cell growth. We inoculated YNB medium with exponentially growing (12 h) and 48 h
old C. albicans to a final concentration of 1x106 cells/ml. We added different
concentrations (10 µM, 25 µM, 40 µM, 100 µM and 300 µM) of farnesol to separate
flasks and incubated at 30oC for yeast cell growth. For exponential phase cells, farnesol
(10 µM and 25 µM) had no effect on the growth of the C. albicans cells. However, 40
µM farnesol significantly (ANOVA, p<0.01) reduced the growth rate of C. albicans
when compared with control cells with no farnesol (2.5 fold at 5 h to 35 fold at 24 h). In
fact, the cells did not grow at all when treated with >100 µM farnesol throughout 24 h
(Fig 5.3). Viable counts of these treated cultured revealed that farnesol had a fatal effect
on the cells (Data not shown).
Interestingly, a very different effect was seen when 48 h old culture was treated
with high concentrations of farnesol. These older cells showed a much higher level of
resistance to farnesol. Only the two highest concentrations of farnesol - 100 µM and 300
µM, had an effect on growth. However, even this effect was not as pronounced as seen in
exponential phase cells (Fig 5.3). Although there was a reduction in the rate of growth,
there was no reduction in cell viability when 48h old cultures were treated with farnesol
(Data not shown). Flow cytometry analysis was performed to monitor the DNA content
of C. albicans cells in 8 h old cultures. As expected for diploid organisms control cells
showed >70% of cells in the budding 4n state and most of the remainder in the 2n state
(Fig. 5.4). When cells were treated with 300 µM farnesol, only about 10% of the cells
were seen in the 4n state, the rest being in the 2n state (Fig. 5.4). Microscopic
examination confirmed that most cells were in the unbudded state (Fig. 5.4). In support of
this notion, our study as noted above revealed many genes involved in mitosis and DNA
replication being downregulated by farnesol under germ tube permissive conditions.
(Table 5.2).
87
Farnesol is an amphiphilic molecule and has been shown to solubilize in model
membranes (Funari, Prades et al. 2005; Rowat, Keller et al. 2005). In mammalian cells it
can effect membrane ion channels (Roullet, Luft et al. 1997; Bringmann, Skatchkov et al.
2000) and in Staphylococcus aureus it inhibits biofilm formation and compromises cell
membrane integrity (Jabra-Rizk, Meiller et al. 2006). It is known that �-mannosides
exposed on mannoproteins and/or phospholipomannan are increased in cells that are
approaching stationary phase rather than exponential phase cells (Martinez-Esparza,
Sarazin et al. 2006). Also, older cells have a thicker, more non-porous cell wall than
exponential phase cells that is known to be a major factor for phenotypic drug resistance
in the C. albicans cells (Beggs 1984; Beggs 1989). We speculate that this difference in
the cell wall could be a reason why 48 h old cells are more resistant to farnesol compared
to the exponential phase cells.
88
Figure 5.3. Effect of farnesol and OAG on C. albicans growth. Exponentially growing
(A) and 48 h old (B) C. albicans yeast cells were grown in the presence of different
concentrations of farnesol (F) in YNB medium. Farnesol retarded the growth rate of the
C. albicans cells. The farnesol treated exponential (C) and 48 h old cells (D) were treated
with 25 µM OAG. This treatment rescued the farnesol mediated growth defect. The
average values of three independent experiments are plotted for each time point.
89
Figure 5.4. Flow cytometry analysis of cells grown for 8 hours in untreated YNB
medium (control), in YNB with 300 µM farnesol and in YNB with 300 µM farnesol and
50 µM OAG. Left X-axis represents cell number counts and Y- axis is the fluorescence
intensity. Right axis indicates budding.
We have previously shown that slower growing cells are smaller in size than more
rapidly growing cells (Chaffin 1984). To ascertain if farnesol mediated reduction in cell
division had any effect on cell size, we determined cell size using a Coulter counter. Cell
sizing results revealed that the cells that were treated with 300 µM farnesol were
significantly smaller in size when compared to the control cells at both 8 h and 18 h in
culture (Table 5.3.).
Table 5.3. Differences in sizes of cells grown in YNB with 300 µM farnesol and in YNB
with 300 µM farnesol and 50 µM OAG at 8 and 18 hours. All the values were
significantly different from each other (ANOVA p< 0.05). Control is fresh untreated
YNB.
Cell size ( fL)
Time (hours)
Control 300 µM Farnesol 300 µM Farnesol + 25 µM OAG
8 52.08 + 3 43.27 + 1.5 54.52 + 1.5
18 57.30 + 1.5 47.98 + 1.7 53.19 + 2.3
90
Rescue from farnesol mediated delay in yeast growth resumption. Farnesol is known to
retard growth in S. cerevisiae as well as in the leukemia cell line CEM-C1, albeit at low
concentrations, 25 µM and 22 µM respectively (Voziyan, Haug et al. 1995; Machida,
Tanaka et al. 1999). This growth retardation was found due to farnesol interference with
either the phosphatidylinositol–type signaling or the biosynthesis of phosphatidylcholine.
The microarray results from our present study also revealed that, with the exception of
phospholipase C, PLC1, many of the genes involved in the phosphatidylinositol pathway
(PI pathway) were downregulated greater than two fold in cells treated with farnesol
(Table 5.2; Fig 5.5). This was not observed in a similar study by Enjalbert et. al.
(Enjalbert and Whiteway 2005). In their conditions, barring the phosphatidylinositol
synthase gene PIS1, the rest of the PI pathway was unchanged. Interference in the
pathway reduces the intracellular concentration of inositol–1,4,5-triphosphate (IP3),
phosphatidyl–3,4,5–triphosphate (PIP3) and diacylglycerol (DAG) that mediate signal
transduction (Flanagan, Schnieders et al. 1993; Carman and Kersting 2004). DAG is a
physiological activator of protein kinase C (PKC) in mammalian cells (Voziyan, Haug et
al. 1995) and PKC is important for normal budding and viability in C. albicans
(Paravicini, Mendoza et al. 1996). In S. cerevisiae and mammalian cells, addition of a
membrane permeable DAG analogue rescued the farnesol mediated growth arrest
(Voziyan, Haug et al. 1995; Machida, Tanaka et al. 1999). Since the concentration of
farnesol that inhibited growth of C. albicans was at least four times greater than required
to inhibit S. cerevisiae, we questioned if adding DAG to the growth medium would
rescue the farnesol mediated growth arrest in the C. albicans cells. Indeed, when 25 µM
of the membrane permeable DAG analogue OAG was added to exponential and 48 h old
cultures treated with different concentrations of farnesol, there was resumption of growth
(Fig. 5.3b). OAG by itself did not have any effect on C. albicans cell growth. By 24 h,
there was a 12 and 3 fold increase in number of exponential cells treated with 100 and
300 µM farnesol respectively. However, this increase only became evident after 12 - 15 h
of growth (Fig 5.3). We speculate that OAG protected some cells from farnesol mediated
killing, eventually leading to increase in cell number. In the case of 48 h old cells
91
however, there was no lag in growth as seen when the cells were treated with farnesol.
The cells treated with even the highest concentration of farnesol grew at the same rate as
the control cells. There was also a 2-6 fold increase in cell number when these farnesol
treated cells were grown in the presence of OAG. In S. cerevisiae OAG at 10 µM and 20
µM partially and completely reversed the farnesol effect respectively (Voziyan, Haug et
al. 1995; Machida, Tanaka et al. 1999). Since, OAG was found to reverse the farnesol
mediated growth delay, we questioned if it could also reverse the farnesol mediated
inhibition of germ tube formation. However that was not observed (Data not shown).
OAG could not induce germ tubes from farnesol arrested cells. This suggests that the
farnesol may act on a different pathway to mediate C. albicans hyphal suppression.
The observation that farnesol arrested the C. albicans cell growth, and DAG, a
mammalian PKC activator could reverse this growth defect, suggested that farnesol may
mediate its effect through PKC in C. albicans. If this is true then C. albicans lacking
PKC should be inert to the effect of farnesol. However this was not observed. C. albicans
devoid of the gene encoding PKC (PKC1), showed similar growth defects as the wild
type C. albicans when treated with farnesol. This indicated that the farnesol mediated
growth defect may not be via PKC in C. albicans. To confirm that farnesol does not
affect C. albicans PKC, we added different concentrations of another known mammalian
PKC activator, PMA to farnesol treated C. albicans cells. We found that unlike DAG,
PMA did not reverse the farnesol mediated growth defect. Mammalian as well as S.
cerevisiae Pkc1p contain two cystine – rich domains C1A and C1B. In mammalian cells,
these C1s are analogues of DAG and have the ability to bind phorbol esters (Schmitz and
Heinisch 2003). However sequence alignments of the yeast C1 repeats showed that
Pkc1p does not bind DAG and belongs to a class of atypical C1 domains. The same is
true for C. albicans Pkc1p.
Both germ tube formation and yeast cell growth require replication capacity. We
found a striking difference between the farnesol concentration needed to inhibit germ
tube formation and that required to inhibit resumption of yeast cell proliferation.
Farnesol was recently reported to inhibit germ tube formation at 1 µM (Mosel, Dumitru
92
et al. 2005) while inhibition of yeast proliferation is observed at 100-fold or higher
concentrations of farnesol (this study, (Kim, Kim et al. 2002). In S. cerevisiae yeast cell
proliferation is inhibited at 22 µM farnesol (Machida, Tanaka et al. 1999) which is the
same range for inhibition of C. albicans germ tube formation. On the other hand the
concentration of OAG that relieves farnesol imposed delay in yeast cell growth
resumption is similar in both yeasts ((Voziyan, Haug et al. 1995; Machida, Tanaka et al.
1999) This suggests that the key step in C. albicans growth delay is resistant to farnesol
compared S. cerevisiae while the downstream process mediated by phosphatidyl inositol
type signaling is similar. In addition, to signaling DAG or OAG is lipid soluble and may
locate to membranes where any associated membrane change may counter that of lipid
soluble farnesol. In C. albicans the differences in concentration to arrest germ tube
formation and yeast cell proliferation may be a survival advantage to allow yeast growth
to continue even if germ tube formation is inhibited.
93
Figure 5.5. Differential expression of genes involved in phosphatidylinositol type
signaling pathway. Expression values (log2 scale) are shown in bold next to the genes.
94
CHAPTER VI
CONCLUDING REMARKS
Candida species are ubiquitous commensal yeasts that usually reside as part of an
individual's normal mucosal (oral cavity, gastrointestinal tract and the vagina) microflora
and can be detected in up to 71% of the healthy population (Naglik, Albrecht et al. 2004).
However, if the balance of the normal flora is disrupted or the immune defenses are
compromised, Candida species can invade mucosal surfaces and cause disease
manifestations. Immunocompromised individuals such as AIDS patients or intensive care
patients, experience some forms of superficial mucosal candidosis, most commonly
thrush. Also, severely compromised individuals develop systemic infections where
mortality rates are high. In addition, nearly three quarters of all healthy women
experience at least one vaginal yeast infection, and about 5% endure recurrent bouts of
disease (Naglik, Albrecht et al. 2004). C. albicans can cause invasive infections by
producing hyphae or pseudohyphae. Candida albicans is capable of forming a biofilm on
mucocutaneous surfaces, as well as on medical devices such as dentures and catheters
(reviewed by Douglas, 2003; Kumamoto & Vinces, 2005; Mukherjee et al., 2005). Sixty-
five percent of edentulous individuals suffer with denture stomatitis and 40 % of patients
with intravenous catheters develop acute fungaemia due to the growth of C. albicans
biofilms on the associated biomedical devices. The growth phase of C. albicans cells in a
biofilm is not known. We hypothesized that some cells within the biofilm reach a
physiological state equivalent to stationary phase in planktonic organisms. To test this
hypothesis we first had to obtain additional characterization of C. albicans stationary
phase and establish a criterion by which stationary phase could be identified.
Based on growth profiles and carbohydrate measurements, we defined the timing
of entry of C. albicans into stationary phase. The growth profile studies revealed some
interesting information about the biology of C. albicans. We found that in glucose rich
conditions, C. albicans, unlike S. cerevisiae did not ferment glucose but metabolized it by
oxidative fermentation. At approximately 20h (two hours before complete glucose
exhaustion), C. albicans switches metabolism and utilizes the last of the glucose by
95
fermentation, thus producing ethanol in the medium. Fermentation takes place up to at
least eight hours after glucose exhaustion, indicating that during this time C. albicans
probably ferments an alternative carbon source. This was the diauxic shift. About 10
hours after glucose exhaustion, C. albicans switches metabolism to respiration. These
observations indicate that oxidative phosphorylation is the default mode of metabolism in
C. albicans. It is not known why under certain conditions C. albicans switches
metabolism from respiration to fermentation, such as during glucose exhaustion. Another
example where C. albicans does so during growth is at the lag phase. It has been reported
that when C. albicans is inoculated in fresh medium, it ferments glucose for at least the
first 5 hours, but switches metabolism to oxidative phosphorylation as it enters the
exponential phase. It appears as though C. albicans prefers fermentation either during
sudden availability of glucose as in the lag phase, or during exhaustion of glucose at 22 h.
One reason why C. albicans behaves differently from S. cerevisiae could be
because C. albicans genome is rewired. Although the genome of both organisms is
significantly homologus, many individual genes have completely different functions in C.
albicans. For example, both the yeasts have the transcription factor GAL4. However,
while Sc.GAL4 codes for regulation of galactose utilization, Ca.GAL4 is involved in
glycolysis and TCA cycle. Another example is the gene ROX1, a suppressor of hypoxic
genes in S. cerevisiae. The homolog of the same gene in C. albicans is a negative
regulator of hyphal genes, and plays no role in anaerobic growth of C. albicans. Thus C.
albicans and S. cerevisiae regulate the same processes by different regulatory circuits.
Also because C. albicans is a commensal of the human body, and has evolved differently
from S. cerevisiae, it could grow differently and respond to environment in a manner
biochemically different from S. cerevisiae.
Using global gene expression analysis and mutant screening studies, we
identified processes important for stationary phase and genes essential for survival in this
phase. We found that the post-diauxic phase cells acquire many but not all characteristics
of stationary phase cells. Stationary phase cells over expressed genes involved in the
thickening of cell wall, production of ribonucleases and proteases and genes involved in
96
protein trafficking. By screening transcription factor mutants, we found that
mitochondrial function and cell wall organization were two processes essential for
viability of cells in stationary phase. Thus we could draw a distinction between cells
found in the post-diauxic shift phase of a culture (often mistaken to be stationary phase
cells) and the cells that are actually in the stationary phase.
One of the important parts of the C. albicans stationary phase study was
standardizing a new protocol for extraction of good quality RNA from all phases of
growth, called the crushed glass beads method. We found that by the traditional methods
of RNA extraction, large molecular weight RNA could not be extracted. To extract RNA
of all classes from stationary phase cells it was essential to grind frozen cells using a
mortar and pestle with glass beads in liquid nitrogen. This assured complete disruption of
the thick stationary phase cell walls of the C. albicans cells and yielded all sizes of RNA.
We speculate that the stationary phase cell wall acts as a sieve to large molecular weight
RNA while the small RNA can move out easily. By using the crushed glass beads
method, we also showed that the commonly used house keeping genes such as ACT1,
TDH3 or EFB1 could not be used for normalizing gene expression data relating to
stationary phase. This was because in stationary phase the transcript levels of these genes
reduced drastically. The results obtained by techniques that measure transcript levels such
as Real time PCR, Northern blots, reverse transcriptase PCR, need to be normalized with
standard housekeeping genes - genes whose transcript levels do not change in all
conditions tested. Finding such reliable genes for the purpose of normalization is the need
of the hour. Earlier studies in C. albicans may have used a less efficient method for RNA
extraction, that may have affected interpretation of the observations. Crushed glass beads
method should be considered for applications requiring proportional representation of
RNA populations.
Studying C. albicans stationary phase will aid in understanding many other aspects of C.
albicans biology. For example, these results could give us an insight to how C. albicans
cells survive along with other bacteria in mixed species biofilm setting. In one of our
ongoing studies, microarray analyses revealed that C. albicans growth genes such as
97
CDC28 and RAS1 were downregulated several fold when C. albicans formed a biofilm
along with the bacteria Rothia dentocariosa on a substrate used for manufacturing voice
prosthesis. These genes are also known to be downregulated in stationary phase
conditions. In this mixed species biofilm, C. albicans could persist in stationary phase,
given the fact that R. dentocariosa multiply faster than the yeast and hence potentially
use up most of the nutrients in the medium. This in vitro mixed species biofilm could be
considered as a miniature model of the real picture in the human body e.g. the gut, where
C. albicans has to compete with hundreds of co-commensal bacteria for space and
nutrients. It is tempting to hypothesize that at least in this organ; a subpopulation of C.
albicans may prefer to persist in stationary phase – a phase which helps long term
survival of organisms in nutrient limiting conditions. Some species of bacteria too are
hypothesized to exist in stationary phase in the gut (Finkel 2006). It would be interesting
to peruse this hypothesis by inoculating C. albicans into the long-term GI tract
colonization mouse model and monitoring both phenotypic as well as gene expression
changes suggestive of stationary phase in C. albicans recovered from the gut.
Consistent with the same idea, we tagged two C. albicans stationary phase genes
– SNO1 and SNZ1 with YFP and monitored fluorescence in C. albicans cells both in
planktonic as well as biofilm condition. We found that even after prolonged incubation,
only 40% of the founder cells of the biofilm (bottom-most substrate adhered cells)
fluoresced, indicating stationary phase, while the rest of the biofilm did not. Hence, we
concluded that major part of the in vitro biofilm is made up of non-stationary phase C.
albicans cells. The presence of stationary phase cells in the bottom-most layer of the
biofilm also indicates that these cells sustained for long periods of time on the substrate
and could serve as a firm base for attaching the body of the biofilm to the surface. It
would be interesting to know the growth phase of the biofilm cells in vivo. To answer this
question we are collaborating with Dr. Dave Andes, University of Wisconsin, who has
established an in vivo (rat) central venous catheter biofilm model. We intend to inoculate
the SNZ1-YFP strain of C. albicans into this model to know if the C. albicans cells in an
in vivo biofilm reach stationary phase.
98
From our studies with the fluorescent strains we found two interesting
observations. In the biofilm, most of the fluorescent cells in the bottom-most layer were
found as separate clusters adhered to different parts of the substrate. This probably means
that there are gradients of environmental conditions in the structurally complex biofilm;
one such caused by nutrient limitation in some parts of biofilms. It is known that biofilms
are structurally complex and are interspersed by ramifying water channels. It could be
possible that cells close to those channels never reach stationary phase, while cells in
certain parts within the biofilm away from the channels reach stationary phase. It would
be interesting to investigate further, what other environments are created in the biofilm,
e.g. differences in pH, oxygen content, etc, in the different layers or niches in the biofilm.
This could be done by using microprobes or monitoring environmentally sensitive gene
transcripts. This knowledge could help us better understand the biofilm entity and may
give us clues for getting rid of the biofilm.
We localized fluorescence in the cell to the cytoplasm. On monitoring
fluorescence, we also found that though both C. albicans yeast and pseudohyphae
fluoresced having reached stationary phase, no fluorescence could ever be observed in C.
albicans true hyphae. The absence of fluorescence from hyphae could be explained by
the fact that hyphae contain very little cytoplasmic material. Alternatively, it could also
be possible that hyphae do not reach stationary phase like the other forms of C. albicans.
During growth to stationary phase, cells release metabolites into the medium,
including molecules that may have a quorum sensing function (Hornby, Jensen et al.
2001; Chen, Fujita et al. 2004). In an in vitro C. albicans biofilm, some yeast cells in the
bottom-most adhered layers of the mature biofilm reach stationary phase (Uppuluri,
Sarmah et al. 2006). Mature biofilms also produce quorum sensing molecules (Ramage,
Saville et al. 2002). One such molecule – farnesol, has been extensively studied in C.
albicans, since it is known to inhibit yeast to hyphae transition and in turn prevent
biofilm formation. Having found the same effect in our studies, we wanted to know what
C. albicans genes and processes were affected by farnesol in planktonic conditions.
Using microarray analysis we found that many hyphal genes were down regulated.
99
Strikingly, the genes necessary for promoting growth and proliferation were the largest
class of downregulated genes. Having found these results we questioned if farnesol might
have a negative effect on growth of C. albicans cells. Indeed we found that there was a
growth retardation when cells were treated >40 µM farnesol. Interestingly, the
exponential phase cells (12 h old) were more susceptible to the effects of farnesol than
cells grown for a longer time (48 h, 3 d old); and high concentrations of farnesol (100 µM
and 300 µM) proved fatal to the exponential phase cells. On the other hand, the older
cells were resistant to killing by farnesol and also had a lesser impact of farnesol on
growth. By 48 h or 3 d, the cells in culture would be in contact with a much higher
concentration of farnesol in the growth medium, than the 12 h old cells. Hence, it could
be possible that such older cells are more immune to high concentrations of farnesol than
exponential phase cells, hence resistant to its effect.
Our microarray results showed that the phosphatidylinositol pathway that
produces DAG, was one of the major pathways downregulated in cells treated with
farnesol. When an analog of diacylglycerol (DAG) – Oleyl acetyl glycerol (OAG) was
added into the medium, the farnesol mediated growth defect was reversed. However, we
found that this effect was not due to activation of protein kinase C, the regular target of
DAG in mammalian cells (activation of which promotes growth). The target of DAG in
C. albicans is not known. It would be interesting to discover what DAG targets in the cell
to reverse the farnesol mediated growth defect. By identifying the target of DAG, we can
in effect identify a potential target of farnesol in the cell.
Our identification of genes and processes regulated during diauxic phase, entry
and maintenance of stationary phase could have significant implications in understanding
the biology of C. albicans. In the long term, the insights gained by this study could lead
to the development of treatment strategies based on the growth state of the cells. In the
short term, the results of this study will expand our existing knowledge of C. albicans
stationary phase, and serve as a foundation for more systematic and un-biased studies in
this area of C. albicans research.
100
REFERENCES
Aboushadi, N., W. H. Engfelt, et al. (1999). "Role of peroxisomes in isoprenoid biosynthesis." J Histochem Cytochem 47(9): 1127-32. Adams, A., D. E. Gottschling, et al. (1997). Methods in Yeast Genetics: A Cold Spring Harbor Laboratory Course Manual. Cold Spring Harbor, New York, Cold Spring Harbor Laboratory Press. Anderl, J. N., J. Zahller, et al. (2003). "Role of nutrient limitation and stationary-phase existence in Klebsiella pneumoniae biofilm resistance to ampicillin and ciprofloxacin." Antimicrob Agents Chemother 47(4): 1251-6. Andes, D., A. Lepak, et al. (2005). "A simple approach for estimating gene expression in Candida albicans directly from a systemic infection site." J Infect Dis 192(5): 893-900. Aragon, A. D., G. A. Quinones, et al. (2006). "Release of extraction-resistant mRNA in stationary phase Saccharomyces cerevisiae produces a massive increase in transcript abundance in response to stress." Genome Biol 7(2): R9. Arnaud, M. B., M. C. Costanzo, et al. (2005). "The Candida Genome Database (CGD), a community resource for Candida albicans gene and protein information." Nucleic Acids Res 33(Database issue): D358-63. Ashrafi, K., D. Sinclair, et al. (1999). "Passage through stationary phase advances replicative aging in Saccharomyces cerevisiae." Proc Natl Acad Sci U S A 96(16): 9100-5. Ausubel, F. M., B. Brent, et al. (2002). Short Protocols in Molecular Biology. New York City, New York, John Wiley & Sons, Inc. Bachewich, C., A. Nantel, et al. (2005). "Cell cycle arrest during S or M phase generates polarized growth via distinct signals in Candida albicans." Mol Microbiol 57(4): 942-59. Baldi, P. and G. W. Hatfield (2002). DNA Microarrays and Gene Expression: From Experiments to Data Analysis and Modeling. Cambridge, United Kingdom, Cambridge University Press. Bassilana, M., J. Hopkins, et al. (2005). "Regulation of the Cdc42/Cdc24 GTPase module during Candida albicans hyphal growth." Eukaryot Cell 4(3): 588-603.
101
Bates, S., D. M. MacCallum, et al. (2005). "Candida albicans Pmr1p, a secretory pathway P-type Ca2+/Mn2+-ATPase, is required for glycosylation and virulence." J Biol Chem 280(24): 23408-15. Beggs, W. H. (1984). "Growth phase in relation to ketoconazole and miconazole susceptibilities of Candida albicans." Antimicrob Agents Chemother 25(3): 316-8. Bertram, G., R. K. Swoboda, et al. (1996). "Structure and regulation of the Candida albicans ADH1 gene encoding an immunogenic alcohol dehydrogenase." Yeast 12(2): 115-27. Bringmann, A., S. N. Skatchkov, et al. (2000). "Farnesol modulates membrane currents in human retinal glial cells." J Neurosci Res 62(3): 396-402. Bruno, V. M. and A. P. Mitchell (2005). "Regulation of azole drug susceptibility by Candida albicans protein kinase CK2." Mol Microbiol 56(2): 559-73. Burk, D., D. Dawson, et al. (2000). Methods in Yeast Genetics: A Cold Spring Harbor Laboratory Course Manual. Cold Spring Harbor, New York, Cold Spring Harbor Laboratory Press. Cao, Y. Y., Y. B. Cao, et al. (2005). "cDNA microarray analysis of differential gene expression in Candida albicans biofilm exposed to farnesol." Antimicrob Agents Chemother 49(2): 584-9. Carman, G. M. and M. C. Kersting (2004). "Phospholipid synthesis in yeast: regulation by phosphorylation." Biochem Cell Biol 82(1): 62-70. Cassone, A. (1986). "Cell wall of pathogenic yeasts and implications for antimycotic therapy." Drugs Exp Clin Res 12(6-7): 635-43. Cassone, A., D. Kerridge, et al. (1979). "Ultrastructural changes in the cell wall of Candida albicans following cessation of growth and their possible relationship to the development of polyene resistance." J Gen Microbiol 110(2): 339-49. Cassone, A., D. Kerridge, et al. (1979). "Ultrastructural changes in the cell wall of Candida albicans following cessation of growth and their possible relationship to the development of polyene resistance." J Gen Microbiol 110: 339-349. Chaffin, W. L. (1984). "The relationship between yeast cell size and cell division in Candida albicans." Can J Microbiol 30(2): 192-203. Chen, H., M. Fujita, et al. (2004). "Tyrosol is a quorum-sensing molecule in Candida albicans." Proc Natl Acad Sci U S A 101(14): 5048-52.
102
Collart, M. S. and S. Oliviero (1995). Preparation of Yeast RNA. Short Protocols in Molecular Biology. New York City, New York, John Wiley & Son, Inc.: 13-26. Cruz, M. C., A. L. Goldstein, et al. (2001). "Rapamycin and less immunosuppressive analogs are toxic to Candida albicans and Cryptococcus neoformans via FKBP12-dependent inhibition of TOR." Antimicrob Agents Chemother 45(11): 3162-70. Cutler, J. E., D. L. Brawner, et al. (1990). "Characteristics of Candida albicans adherence to mouse tissues." Infect Immun 58(6): 1902-8. Delbruck, S. and J. F. Ernst (1993). "Morphogenesis-independent regulation of actin transcript levels in the pathogenic yeast Candida albicans." Mol Microbiol 10(4): 859-66. DeRisi, J. L., V. R. Iyer, et al. (1997). "Exploring the metabolic and genetic control of gene expression on a genomic scale." Science 278(5338): 680-6. Donlan, R. M. (2001). "Biofilms and device-associated infections." Emerg Infect Dis 7(2): 277-81. Douglas, L. J. (2003). "Candida biofilms and their role in infection." Trends Microbiol 11(1): 30-6. Dudani, A. K. and R. Prasad (1985). "Differences in amino acid transport and phospholipid contents during the cell cycle of Candida albicans." Folia Microbiol (Praha) 30(6): 493-500. Enjalbert, B., A. Nantel, et al. (2003). "Stress-induced gene expression in Candida albicans: absence of a general stress response." Mol Biol Cell 14(4): 1460-7. Finkel, S. E. (2006). "Long-term survival during stationary phase: evolution and the GASP phenotype." Nat Rev Microbiol 4(2): 113-20. Flanagan, C. A., E. A. Schnieders, et al. (1993). "Phosphatidylinositol 4-kinase: gene structure and requirement for yeast cell viability." Science 262(5138): 1444-8. Fuge, E. K., E. L. Braun, et al. (1994). "Protein synthesis in long-term stationary-phase cultures of Saccharomyces cerevisiae." J Bacteriol 176(18): 5802-13. Funari, S. S., J. Prades, et al. (2005). "Farnesol and geranylgeraniol modulate the structural properties of phosphatidylethanolamine model membranes." Mol Membr Biol 22(4): 303-11.
103
Galan, A., M. Casanova, et al. (2004). "The Candida albicans pH-regulated KER1 gene encodes a lysine/glutamic-acid-rich plasma-membrane protein that is involved in cell aggregation." Microbiology 150(Pt 8): 2641-51. Gale, E. F., J. Ingram, et al. (1980). "Reduction of amphotericin resistance in stationary phase cultures of Candida albicans by treatment with enzymes." J Gen Microbiol 117(2): 383-91. Gale, E. F., A. M. Johnson, et al. (1980). "Phenotypic resistance to miconazle and amphotericin B in Candida albicans." J Gen Microbiol 117: 535-538. Gale, E. F., A. M. Johnson, et al. (1980). "Phenotypic resistance to miconazole and amphotericin B in Candida albicans." J Gen Microbiol 117(2): 535-8. Garcia-Sanchez, S., A. L. Mavor, et al. (2005). "Global roles of Ssn6 in Tup1- and Nrg1-dependent gene regulation in the fungal pathogen, Candida albicans." Mol Biol Cell 16(6): 2913-25. Gaur, M., D. Choudhury, et al. (2005). "Complete inventory of ABC proteins in human pathogenic yeast, Candida albicans." J Mol Microbiol Biotechnol 9(1): 3-15. Gow, N. A., A. J. Brown, et al. (2002). "Fungal morphogenesis and host invasion." Curr Opin Microbiol 5(4): 366-71. Granger, B. L., M. L. Flenniken, et al. (2005). "Yeast wall protein 1 of Candida albicans." Microbiology 151(Pt 5): 1631-44. Gray, J. V., G. A. Petsko, et al. (2004). ""Sleeping beauty": quiescence in Saccharomyces cerevisiae." Microbiol Mol Biol Rev 68(2): 187-206. Gurvitz, A., J. K. Hiltunen, et al. (2001). "Saccharomyces cerevisiae Adr1p governs fatty acid beta-oxidation and peroxisome proliferation by regulating POX1 and PEX11." J Biol Chem 276(34): 31825-30. Hartwell, L. H., J. Culotti, et al. (1970). "Genetic control of the cell-division cycle in yeast. I. Detection of mutants." Proc Natl Acad Sci U S A 66(2): 352-9. Hauser, N. C., M. Vingron, et al. (1998). "Transcriptional profiling on all open reading frames of Saccharomyces cerevisiae." Yeast 14(13): 1209-21. Herbert, D., C. N. Paramasivan, et al. (1996). "Bactericidal action of ofloxacin, sulbactam-ampicillin, rifampin, and isoniazid on logarithmic- and stationary-phase cultures of Mycobacterium tuberculosis." Antimicrob Agents Chemother 40(10): 2296-9.
104
Hirooka, K., Y. Yamamoto, et al. (2005). "Improved production of isoamyl acetate by a sake yeast mutant resistant to an isoprenoid analog and its dependence on alcohol acetyltransferase activity, but not on isoamyl alcohol production." J Biosci Bioeng 99(2): 125-9. Hong, E. L., R. Balakrishnan, et al. Saccharomyces Genome Database. Hornby, J. M., E. C. Jensen, et al. (2001). "Quorum sensing in the dimorphic fungus Candida albicans is mediated by farnesol." Appl Environ Microbiol 67(7): 2982-92. Hornby, J. M., B. W. Kebaara, et al. (2003). "Farnesol biosynthesis in Candida albicans: cellular response to sterol inhibition by zaragozic acid B." Antimicrob Agents Chemother 47(7): 2366-9. Hudspeth, M. E., D. S. Shumard, et al. (1980). "Rapid purification of yeast mitochondrial DNA in high yield." Biochim Biophys Acta 610(2): 221-8. Ihmels, J., S. Bergmann, et al. (2005). "Rewiring of the yeast transcriptional network through the evolution of motif usage." Science 309(5736): 938-40. Jabra-Rizk, M. A., T. F. Meiller, et al. (2006). "Effect of farnesol on Staphylococcus aureus biofilm formation and antimicrobial susceptibility." Antimicrob Agents Chemother 50(4): 1463-9. Kadosh, D. and A. D. Johnson (2005). "Induction of the Candida albicans filamentous growth program by relief of transcriptional repression: a genome-wide analysis." Mol Biol Cell 16(6): 2903-12. Kim, S., E. Kim, et al. (2002). "Evaluation of morphogenic regulatory activity of farnesoic acid and its derivatives against Candida albicans dimorphism." Bioorg Med Chem Lett 12(6): 895-8. King, R. D., J. C. Lee, et al. (1980). "Adherence of Candida albicans and other Candida species to mucosal epithelial cells." Infect Immun 27(2): 667-74. Kohrer, K. and H. Domdey (1991). "Preparation of high molecular weight RNA." Methods Enzymol 194: 398-405. Kuhn, D. M. and M. A. Ghannoum (2004). "Candida biofilms: antifungal resistance and emerging therapeutic options." Curr Opin Investig Drugs 5(2): 186-97. Kumamoto, C. A. and M. D. Vinces (2005). "Alternative Candida albicans Lifestyles: Growth on Surfaces." Annu Rev Microbiol.
105
Kumamoto, C. A. and M. D. Vinces (2005). "Alternative Candida albicans lifestyles: growth on surfaces." Annu Rev Microbiol 59: 113-33. Kuzj, A. E., P. S. Medberry, et al. (1998). "Stationary phase, amino acid limitation and recovery from stationary phase modulate the stability and translation of chloramphenicol acetyltransferase mRNA and total mRNA in Escherichia coli." Microbiology 144 (Pt 3): 739-50. Lamarre, C., J. D. LeMay, et al. (2001). "Candida albicans expresses an unusual cytoplasmic manganese-containing superoxide dismutase (SOD3 gene product) upon the entry and during the stationary phase." J Biol Chem 276(47): 43784-91. Lamming, D. W., M. Latorre-Esteves, et al. (2005). "HST2 mediates SIR2-independent life-span extension by calorie restriction." Science 309(5742): 1861-4. Li, C. R., Y. M. Wang, et al. (2005). "The formin family protein CaBni1p has a role in cell polarity control during both yeast and hyphal growth in Candida albicans." J Cell Sci 118(Pt 12): 2637-48. Lopez-Ribot, J. L. (2005). "Candida albicans biofilms: more than filamentation." Curr Biol 15(12): R453-5. Lopez de Heredia, M. and R. P. Jansen (2004). "RNA integrity as a quality indicator during the first steps of RNP purifications: a comparison of yeast lysis methods." BMC Biochem 5: 14. Lorenz, M. C., J. A. Bender, et al. (2004). "Transcriptional response of Candida albicans upon internalization by macrophages." Eukaryot Cell 3(5): 1076-87. Lyons, C. N. and T. C. White (2000). "Transcriptional analyses of antifungal drug resistance in Candida albicans." Antimicrob Agents Chemother 44(9): 2296-303. Machida, K., T. Tanaka, et al. (1998). "Farnesol-induced generation of reactive oxygen species via indirect inhibition of the mitochondrial electron transport chain in the yeast Saccharomyces cerevisiae." J Bacteriol 180(17): 4460-5. Machida, K., T. Tanaka, et al. (1999). "Farnesol-induced growth inhibition in Saccharomyces cerevisiae by a cell cycle mechanism." Microbiology 145 (Pt 2): 293-9. Maneu, V., J. P. Martinez, et al. (2000). "Identification of Candida albicans clinical isolates by PCR amplification of an EFB1 gene fragment containing an intron-interrupted open reading frame." Med Mycol 38(2): 123-6.
106
Manna, F., D. R. Massardo, et al. (1996). "A simple and inexpensive method for RNA extraction from yeasts." Trends Genet 12(9): 337-8. Martchenko, M., A. Levitin, et al. (2006). "Analysis of the Candida albicans galactose regulon, and the role of the Gal4 protein." 8th ASM Conference on Candida and Candidiasis Abstract B26: 41-42. Martinez, M. J., S. Roy, et al. (2004). "Genomic analysis of stationary-phase and exit in Saccharomyces cerevisiae: gene expression and identification of novel essential genes." Mol Biol Cell 15(12): 5295-305. Masuoka, J. and K. C. Hazen (1999). "Differences in the acid-labile component of Candida albicans mannan from hydrophobic and hydrophilic yeast cells." Glycobiology 9(11): 1281-6. McCourtie, J. and L. J. Douglas (1981). "Relationship between cell surface composition of Candida albicans and adherence to acrylic after growth on different carbon sources." Infect Immun 32(3): 1234-41. McEntee, C. M. and A. P. Hudson (1989). "Preparation of RNA from unspheroplasted yeast cells (Saccharomyces cerevisiae)." Anal Biochem 176(2): 303-6. McLeod, G. I. and M. P. Spector (1996). "Starvation- and Stationary-phase-induced resistance to the antimicrobial peptide polymyxin B in Salmonella typhimurium is RpoS (sigma(S)) independent and occurs through both phoP-dependent and -independent pathways." J Bacteriol 178(13): 3683-8. Monje-Casas, F., C. Michan, et al. (2004). "Absolute transcript levels of thioredoxin- and glutathione-dependent redox systems in Saccharomyces cerevisiae: response to stress and modulation with growth." Biochem J 383(Pt 1): 139-47. Moreno, I., Y. Pedreno, et al. (2003). "Characterization of a Candida albicans gene encoding a putative transcriptional factor required for cell wall integrity." FEMS Microbiol Lett 226(1): 159-67. Mosel, D. D., R. Dumitru, et al. (2005). "Farnesol concentrations required to block germ tube formation in Candida albicans in the presence and absence of serum." Appl Environ Microbiol 71(8): 4938-40. Motizuki, M. and K. Tsurugi (1992). "The effect of aging on protein synthesis in the yeast Saccharomyces cerevisiae." Mech Ageing Dev 64(3): 235-45. Mukherjee, P. K., J. Chandra, et al. (2003). "Mechanism of fluconazole resistance in Candida albicans biofilms: phase-specific role of efflux pumps and membrane sterols." Infect Immun 71(8): 4333-40.
107
Mukherjee, P. K., G. Zhou, et al. (2005). "Candida biofilm: a well-designed protected environment." Med Mycol 43(3): 191-208. Munro, C. A., D. A. Schofield, et al. (1998). "Regulation of chitin synthesis during dimorphic growth of Candida albicans." Microbiology 144 (Pt 2): 391-401. Murad, A. M., C. d'Enfert, et al. (2001). "Transcript profiling in Candida albicans reveals new cellular functions for the transcriptional repressors CaTup1, CaMig1 and CaNrg1." Mol Microbiol 42(4): 981-93. Murad, A. M., P. Leng, et al. (2001). "NRG1 represses yeast-hypha morphogenesis and hypha-specific gene expression in Candida albicans." Embo J 20(17): 4742-52. Naglik, J., A. Albrecht, et al. (2004). "Candida albicans proteinases and host/pathogen interactions." Cell Microbiol 6(10): 915-26. Nantel, A., D. Dignard, et al. (2002). "Transcription profiling of Candida albicans cells undergoing the yeast-to-hyphal transition." Mol Biol Cell 13(10): 3452-65. Newcomb, L. L., J. A. Diderich, et al. (2003). "Glucose regulation of Saccharomyces cerevisiae cell cycle genes." Eukaryot Cell 2(1): 143-9. Nickerson, K. W., A. L. Atkin, et al. (2006). "Quorum sensing in dimorphic fungi: farnesol and beyond." Appl Environ Microbiol 72(6): 3805-13. Nobile, C. J. and A. P. Mitchell (2006). "Genetics and genomics of Candida albicans biofilm formation." Cell Microbiol. Paravicini, G., A. Mendoza, et al. (1996). "The Candida albicans PKC1 gene encodes a protein kinase C homolog necessary for cellular integrity but not dimorphism." Yeast 12(8): 741-56. Perrod, S., M. M. Cockell, et al. (2001). "A cytosolic NAD-dependent deacetylase, Hst2p, can modulate nucleolar and telomeric silencing in yeast." Embo J 20(1-2): 197-209. Postlethwait, P. and P. Sundstrom (1995). "Genetic organization and mRNA expression of enolase genes of Candida albicans." J Bacteriol 177(7): 1772-9. Prill, S. K., B. Klinkert, et al. (2005). "PMT family of Candida albicans: five protein mannosyltransferase isoforms affect growth, morphogenesis and antifungal resistance." Mol Microbiol 55(2): 546-60.
108
Ramage, G., S. P. Saville, et al. (2002). "Inhibition of Candida albicans biofilm formation by farnesol, a quorum-sensing molecule." Appl Environ Microbiol 68(11): 5459-63. Rivas, R., N. Vizcaino, et al. (2001). "An effective, rapid and simple method for total RNA extraction from bacteria and yeast." J Microbiol Methods 47(1): 59-63. Roman, E., C. Nombela, et al. (2005). "The Sho1 adaptor protein links oxidative stress to morphogenesis and cell wall biosynthesis in the fungal pathogen Candida albicans." Mol Cell Biol 25(23): 10611-27. Romandini, P., C. Bonotto, et al. (1994). "Superoxide dismutase, catalase and cell dimorphism in Candida albicans cells exposed to methanol and different temperatures." Comp Biochem Physiol Pharmacol Toxicol Endocrinol 108(1): 53-7. Roullet, J. B., U. C. Luft, et al. (1997). "Farnesol inhibits L-type Ca2+ channels in vascular smooth muscle cells." J Biol Chem 272(51): 32240-6. Rowat, A. C., D. Keller, et al. (2005). "Effects of farnesol on the physical properties of DMPC membranes." Biochim Biophys Acta 1713(1): 29-39. Sarthy, A. V., T. McGonigal, et al. (1997). "Phenotype in Candida albicans of a disruption of the BGL2 gene encoding a 1,3-beta-glucosyltransferase." Microbiology 143 (Pt 2): 367-76. Sato, T., T. Watanabe, et al. (2004). "Farnesol, a morphogenetic autoregulatory substance in the dimorphic fungus Candida albicans, inhibits hyphae growth through suppression of a mitogen-activated protein kinase cascade." Biol Pharm Bull 27(5): 751-2. Schultz, M. C. (1999). "Chromatin assembly in yeast cell-free extracts." Methods 17(2): 161-72. Semighini, C. P., J. M. Hornby, et al. (2006). "Farnesol-induced apoptosis in Aspergillus nidulans reveals a possible mechanism for antagonistic interactions between fungi." Mol Microbiol 59(3): 753-64. Sherlock, G., A. M. Bahman, et al. (1994). "Molecular cloning and analysis of CDC28 and cyclin homologues from the human fungal pathogen Candida albicans." Mol Gen Genet 245(6): 716-23. Singh, V., I. Sinha, et al. (2005). "Global analysis of altered gene expression during morphogenesis of Candida albicans in vitro." Biochem Biophys Res Commun 334(4): 1149-58.
109
Slattery, M. G., D. Liko, et al. (2006). "The function and properties of the Azf1 transcriptional regulator change with growth conditions in Saccharomyces cerevisiae." Eukaryot Cell 5(2): 313-20. Sogin, S. J. and C. A. Saunders (1980). "Fluctuation in polyadenylate size and content in exponential- and stationary-phase cells of Saccharomyces cerevisiae." J Bacteriol 144(1): 74-81. Song, J. L., J. B. Harry, et al. (2004). "The Candida albicans lanosterol 14-alpha-demethylase (ERG11) gene promoter is maximally induced after prolonged growth with antifungal drugs." Antimicrob Agents Chemother 48(4): 1136-44. Spoering, A. L. and K. Lewis (2001). "Biofilms and planktonic cells of Pseudomonas aeruginosa have similar resistance to killing by antimicrobials." J Bacteriol 183(23): 6746-51. Suci, P. A. and B. J. Tyler (2003). "A method for discrimination of subpopulations of Candida albicans biofilm cells that exhibit relative levels of phenotypic resistance to chlorhexidine." J Microbiol Methods 53(3): 313-25. Sudbery, P., N. Gow, et al. (2004). "The distinct morphogenic states of Candida albicans." Trends Microbiol 12(7): 317-24. Sullivan, P. A., C. Y. Yin, et al. (1983). "An analysis of the metabolism and cell wall composition of Candida albicans during germ-tube formation." Can J Microbiol 29(11): 1514-25. Svensater, G., O. Bjornsson, et al. (2001). "Effect of carbon starvation and proteolytic activity on stationary-phase acid tolerance of Streptococcus mutans." Microbiology 147(Pt 11): 2971-9. Swinnen, E., V. Wanke, et al. (2006). "Rim15 and the crossroads of nutrient signalling pathways in Saccharomyces cerevisiae." Cell Div 1: 3. Swinnen, E., V. Wanke, et al. (2006). "Rim15 and the crossroads of nutrient signalling pathways in Saccharomyces cerevisiae." Cell Div 1(1): 3. Swoboda, R. K., G. Bertram, et al. (1994). "Fluctuations in glycolytic mRNA levels during morphogenesis in Candida albicans reflect underlying changes in growth and are not a response to cellular dimorphism." Mol Microbiol 13(4): 663-72. Tavanti, A., G. Pardini, et al. (2004). "Differential expression of secretory aspartyl proteinase genes (SAP1-10) in oral Candida albicans isolates with distinct karyotypes." J Clin Microbiol 42(10): 4726-34.
110
Thurnheer, T., J. R. van der Ploeg, et al. (2006). "Effects of Streptococcus mutans gtfC deficiency on mixed oral biofilms in vitro." Caries Res 40(2): 163-71. Tournu, H., G. Tripathi, et al. (2005). "Global role of the protein kinase Gcn2 in the human pathogen Candida albicans." Eukaryot Cell 4(10): 1687-96. Uppuluri, P., P. Perumal, et al. (2006). "Analysis of RNAs of various sizes from stationary phase planktonic yeast cells of Candida albicans." FEMS Yeast Research manuscript in press. Uppuluri, P., B. Sarmah, et al. (2006). "Candida albicans SNO1 and SNZ1 expressed in stationary-phase planktonic yeast cells and base of biofilm." Microbiology 152(Pt 7): 2031-8. Uppuluri, P., B. Sarmah, et al. (2006). "Candida albicans SNO1 and SNZ1 expressed in stationary phase planktonic yeast cells and base of biofilm." Microbiology in press. Vijg, J. and Y. Suh (2005). "Genetics of longevity and aging." Annu Rev Med 56: 193-212. Vinces, M. D., C. Haas, et al. (2006). "Expression of the Candida albicans morphogenesis regulator gene CZF1 and its regulation by Efg1p and Czf1p." Eukaryot Cell 5(5): 825-35. Voziyan, P. A., J. S. Haug, et al. (1995). "Mechanism of farnesol cytotoxicity: further evidence for the role of PKC-dependent signal transduction in farnesol-induced apoptotic cell death." Biochem Biophys Res Commun 212(2): 479-86. Weeks, G. and G. B. Spiegelman (2003). "Roles played by Ras subfamily proteins in the cell and developmental biology of microorganisms." Cell Signal 15(10): 901-9. Wenzel, T. J., A. W. Teunissen, et al. (1995). "PDA1 mRNA: a standard for quantitation of mRNA in Saccharomyces cerevisiae superior to ACT1 mRNA." Nucleic Acids Res 23(5): 883-4. Werner-Washburne, M., E. Braun, et al. (1993). "Stationary phase in the yeast Saccharomyces cerevisiae." Microbiol Rev 57(2): 383-401. Werner-Washburne, M., E. L. Braun, et al. (1996). "Stationary phase in Saccharomyces cerevisiae." Mol Microbiol 19(6): 1159-66. Westfall, D., N. Aboushadi, et al. (1997). "Metabolism of farnesol: phosphorylation of farnesol by rat liver microsomal and peroxisomal fractions." Biochem Biophys Res Commun 230(3): 562-8.
111
Westwater, C., E. Balish, et al. (2005). "Candida albicans-conditioned medium protects yeast cells from oxidative stress: a possible link between quorum sensing and oxidative stress resistance." Eukaryot Cell 4(10): 1654-61. White, T. C., K. A. Marr, et al. (1998). "Clinical, cellular, and molecular factors that contribute to antifungal drug resistance." Clin Microbiol Rev 11(2): 382-402. Whiteway, M. (2000). "Transcriptional control of cell type and morphogenesis in Candida albicans." Curr Opin Microbiol 3(6): 582-8. Yaar, L., M. Mevarech, et al. (1997). "A Candida albicans RAS-related gene (CaRSR1) is involved in budding, cell morphogenesis and hypha development." Microbiology 143 (Pt 9): 3033-44. Yazdanyar, A., M. Essmann, et al. (2001). "Genistein effects on growth and cell cycle of Candida albicans." J Biomed Sci 8(2): 153-9. Zaragoza, O., C. de Virgilio, et al. (2002). "Disruption in Candida albicans of the TPS2 gene encoding trehalose-6-phosphate phosphatase affects cell integrity and decreases infectivity." Microbiology 148(Pt 5): 1281-90. Zhao, R., K. J. Daniels, et al. (2005). "Unique aspects of gene expression during Candida albicans mating and possible G(1) dependency." Eukaryot Cell 4(7): 1175-90.
