Comparative genomics using Candida albicans DNA microarrays reveals absence and divergence of virulence-associated genes in Candida dubliniensis

Gary Moran1, Cheryl Stokes1, Sascha Thewes2, Bernhard Hube2, David C. Coleman1 and Derek Sullivan1

1 Microbiology Research Unit, Department of Oral Surgery, Oral Medicine and Oral Pathology, School of Dental Science, University of Dublin, Trinity College, Dublin 2, Republic of Ireland
2 Robert Koch Institut, Nordufer 20, D-13353 Berlin, Germany

Correspondence
Derek Sullivan
derek.sullivan{at}dental.tcd.ie


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Candida dubliniensis is a pathogenic yeast species closely related to Candida albicans. However, it is less frequently associated with human disease and displays reduced virulence in animal models of infection. Here comparative genomic hybridization was used in order to assess why C. dubliniensis is apparently less virulent than C. albicans. In these experiments the genomes of the two species were compared by co-hybridizing C. albicans microarrays with fluorescently labelled C. albicans and C. dubliniensis genomic DNA. C. dubliniensis genomic DNA was found to hybridize reproducibly to 95·6 % of C. albicans gene-specific sequences, indicating a significant degree of nucleotide sequence homology (>60 %) in these sequences. The remaining 4·4 % of sequences (representing 247 genes) gave C. albicans/C. dubliniensis normalized fluorescent signal ratios that indicated significant sequence divergence (<60 % homology) or absence in C. dubliniensis. Sequence divergence was identified in several genes (confirmed by Southern blot analysis and sequence analysis of PCR products) with putative virulence functions, including the gene encoding the hypha-specific human transglutaminase substrate Hwp1p. Poor hybridization of C. dubliniensis genomic DNA to the array sequences for the secreted aspartyl proteinase-encoding gene SAP5 also led to the finding that SAP5 was absent in C. dubliniensis and that this species possesses only one gene homologous to SAP4 and SAP6 of C. albicans. In addition, divergence and absence of sequences in several gene families was identified, including a family of HYR1-like GPI-anchored proteins, a family of genes homologous to a putative transcriptional activator (CTA2) and several ALS genes. This study has confirmed the close relatedness of C. albicans and C. dubliniensis and has identified a subset of unique C. albicans genes that may contribute to the increased prevalence and virulence of this species.


Abbreviations: CGH, comparative genomic hybridization; GPI, glycosylphosphatidylinositol

The GenBank/EMBL/DDBJ accession numbers for the sequences reported in this paper are AJ634664AJ634672, AJ634675AJ634676 and AJ634382.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Candida dubliniensis is associated with oral candidosis in the HIV-infected population in which this species was first identified in 1995 (Jabra-Rizk et al., 2001; Sullivan et al., 1995, 2004). Recently, several studies have also identified C. dubliniensis as a cause of oral disease in diabetic and cancer patients (Sebti et al., 2001; Willis et al., 2000). However, the closely related species Candida albicans appears to be more successful than C. dubliniensis as a commensal of the human oral cavity in healthy individuals, as determined by standard oral swab sampling methods (Coleman et al., 1997). In addition, the incidence of C. dubliniensis isolation from blood cultures is extremely low compared to C. albicans (Kibbler et al., 2003; Pfaller & Dieker, 2004). In a recent study of Candida species recovered from blood cultures in six sentinel hospitals in England and Wales between 1997 and 1999, C. dubliniensis was isolated from only 2 % of samples compared to 65 % for C. albicans. Two studies have also demonstrated that C. dubliniensis is less virulent than C. albicans in a murine model of systemic candidosis (Gilfillan et al., 1998; Vilela et al., 2002). The reason for the apparent difference in virulence between the two species is unknown as they are phenotypically very similar and seem to share many of the traits traditionally associated with virulence in C. albicans. In particular both species have the ability to form true hyphae, to adhere to human epithelium and to produce secreted aspartyl proteinases (Gilfillan et al., 1998; Hannula et al., 2000; Vilela et al., 2002). However, C. dubliniensis does not form hyphae as rapidly as C. albicans in response to shifts in pH/temperature or when incubated in serum (Gilfillan et al., 1998). In contrast, when cultured on Staib agar or Pal's agar C. dubliniensis forms abundant hyphae, pseudohyphae and chlamydospores, whereas C. albicans remains in the yeast phase (Al Mosaid et al., 2001, 2003). C. dubliniensis also seems to be more sensitive to environmental stress, such as elevated temperature and NaCl concentration (Alves et al., 2002; Pinjon et al., 1998).

Comparative genomic hybridization (CGH) studies with DNA microarrays provide a rapid and cost-effective method for obtaining informative data about uncharacterized genomes and have been used extensively to compare gene content in prokaryotic and eukaryotic micro-organisms (Daran-Lapujade et al., 2003; Dong et al., 2001; Murray et al., 2001). The completion of the C. albicans genome project and the availability of C. albicans DNA microarrays now enables genomes of different strains of C. albicans and closely related species, such as C. dubliniensis, to be compared. In the present study, CGH was performed between C. albicans and C. dubliniensis using C. albicans DNA microarrays in order to identify genomic differences that might account for the difference in virulence between C. albicans and C. dubliniensis. This approach was deemed feasible as all C. dubliniensis genes analysed to date share more than 90 % identity at the nucleotide sequence level with the orthologous C. albicans genes. Total genomic DNA from C. albicans and C. dubliniensis was co-hybridized to C. albicans DNA microarrays and the relative hybridization efficiency of C. dubliniensis and C. albicans DNA to each gene-specific spot was compared. This approach allowed us to identify the presence of thousands of C. albicans-homologous genes in C. dubliniensis without the need for sequence analysis and has guided us towards genes which are highly divergent or even absent from C. dubliniensis. We anticipate that this collection of C. albicans-specific sequences may contain genes that contribute to the observed differences in virulence and epidemiology between these two organisms.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Candida strains and culture conditions.
C. albicans strain SC5314 was used as a control in all CGH experiments using Eurogentec C. albicans DNA microarrays. C. dubliniensis strains used in this study included the C. dubliniensis type strain CD36 (American Type Culture Collection reference MYA-178, British National Collection of Pathogenic Fungi reference NCPF3949), which is a representative of C. dubliniensis Cd25 fingerprint group I (genotype 1), and C. dubliniensis strain CD514, a strain representative of Cd25 fingerprint group II (genotype 3) (Gee et al., 2002). Strains were routinely grown on potato dextrose agar (PDA; Oxoid) medium, pH 5·6, at 37 °C. For liquid culture, cells were grown in yeast extract-peptone-dextrose (YEPD) broth, also at 37 °C (Gallagher et al., 1992).

Chemicals, enzymes and radioisotopes.
All chemicals used were of molecular biology grade and were purchased from Sigma-Aldrich. Molecular biology enzymes and kits were purchased from Promega or New England Biolabs unless otherwise indicated. Cy5 and Cy3 dUTP were purchased from Amersham Biosciences Europe. Supplies of [{alpha}-32P]dATP (6000 Ci mmol–1, 220 TBq mmol–1) were purchased from NEN Life Sciences.

DNA microarrays.
C. albicans DNA microarrays used in this study were constructed by Eurogentec based on the Galar Fungail consortium's annotation of the C. albicans SC5314 genome sequence in the CandidaDB (http://www.pasteur.fr/Galar_Fungail/CandidaDB/). This annotation was produced based on the genome sequence released by the Stanford Genome Technology Center. Each glass slide microarray contained sequences corresponding to 6039 ORFs (98 % of annotated genes) that were approximately 300 bp in length and spotted in duplicate.

Genomic DNA preparation.
High-molecular-mass total genomic DNA was recovered from Candida strains by organic extraction following digestion of the cell wall with Zymolyase 20T (Seikagu) and proteinase K treatment (Roche Diagnostics) as described by Gallagher et al. (1992).

Genomic DNA labelling and microarray hybridization.
For DNA labelling experiments with Cy5 dUTP and Cy3 dUTP, total genomic DNA (2 µg) was first fragmented by either restriction endonuclease digestion or sonication. For restriction endonuclease digests, two 1 µg aliquots of DNA were separately digested with Tru1I or RsaI (Fermentas). These digests were then heat inactivated, extracted once with a mixture of phenol/chloroform/isoamyl alcohol (25 : 24 : 1, by vol.) and ethanol precipitated. The two separate aliquots of digested DNA were combined to give a mixture of Tru1I and RsaI fragments (50–4000 bp) for labelling. Alternatively, separate DNA samples were prepared for labelling by sonication using a Sonoplus HD70 sonicator (Bandelin Electronic) at 75 % power for 30 cycles producing DNA fragments ranging from 500 to 5000 bp.

Each labelling reaction was carried out with 2 µg of either sheared or digested genomic DNA using the RadPrime random priming labelling system (Invitrogen) incorporating Cy5 dUTP or Cy3 dUTP into C. albicans SC5314 or C. dubliniensis DNA fragments. After labelling, reaction products were purified with Nucleospin PCR clean up columns (Macherey-Nagel) and concentrated to a final volume of <5 µl with a Microcon YM-30 column (Millipore). Cy5-labelled and Cy3-labelled reactions were mixed together in DIG EasyHyb buffer (Roche Diagnostics) to a final volume of 60 µl for hybridization. The mixture was denatured at 98 °C for 5 min then chilled on ice. Microarray slides (Eurogentec) were placed in a hybridization chamber (Corning), covered with a glass LifterSlip (Erie Scientific Company) and the labelling reaction was carefully applied at the edges of the slide. The chamber was sealed and incubated in the dark at 42 °C for 16–18 h. Slides were washed at high stringency at room temperature as follows: (i) 5 min in 1x SSC, 0·03 % (w/v) SDS, (ii) 5 min in 0·2x SSC and (iii) 5 min in 0·05x SSC. Following washing, slides were dried thoroughly by centrifugation at low speed for 5 min in a 50 ml disposable plastic tube (Greiner Bio-One) and scanned immediately.

Each of the hybridizations performed using digested DNA and sheared DNA was carried out on two separate occasions. A third sheared DNA hybridization was performed with the Cy3 and Cy5 dyes swapped. One additional hybridization was performed between sheared Cy5-labelled C. albicans DNA and sheared Cy3-labelled C. dubliniensis CD514 DNA.

Data analysis.
DNA microarray slides were scanned with the GenePix 4000B scanner (Axon Instruments). Data were extracted from scanned images using the GenePix Pro 4.1.1.4 software package (Axon Instruments). Data normalization and subsequent analysis was carried out with the GeneSpring 6.1 software package (Silicon Genetics). Hybridization data from each DNA ‘spot’ on the slide were only included for analysis if the control (C. albicans) channel signal was above local background plus two standard deviations. Signal intensities in both channels were background corrected. Measurements were normalized across the whole chip by dividing each measurement by the median of all measurements taken for that chip. A normalized fluorescence ratio value was determined for each spot by dividing the C. albicans control channel normalized signal by the C. dubliniensis normalized signal values. The log2 value of each ratio was determined and the log2 ratios of duplicate spots were averaged. The significance of normalized ratios of <1 was determined in replicate experiments using Student's t test. The raw data from all of the experiments has been submitted in a MIAME-compliant format to the ArrayExpress database at the European Bioinformatics Institute (accession code E-MEXP-99).

The relationship between log2 ratio values and nucleotide sequence homology was determined by linear regression analysis using Prism 4.0 (GraphPad Software). For this analysis, nucleotide sequence homology between the array printed C. albicans sequences (~300 bp) and the corresponding region of available homologous C. dubliniensis sequences was determined using DNA Strider 3.1 software. Sequences used in this analysis included the available C. dubliniensis gene sequences from GenBank and PCR-amplified sequences described here (Table 1). Sequences were included for analysis only when a minimum of 100 bp of uninterrupted sequence could be aligned. Gaps of over 50 bp in length were excluded from homology calculations. On the basis of this analysis, we chose genes with normalized ratios of <0·5 (P<0·05) in replicate dye-swap experiments with sheared genomic DNA for further study.


View this table:
[in this window]
[in a new window]
 
Table 1. Percentage nucleotide sequence homology of C. dubliniensis genes to corresponding C. albicans Eurogentec microarray sequences

 
Larger sequence alignments (>500 bp) described in the results were carried out using the CLUSTAL_W software package (Higgins & Sharp, 1988).

Southern hybridization.
Southern hybridization analysis was carried out as described previously (Moran et al., 1998; Southern, 1975) using DNA sequences labelled with [{alpha}-32P]dATP by random primer labelling (Prime-a-Gene system; Promega) or using DIG-labelled probes incorporating DIG-11-dUTP (Roche Diagnostics) during PCR amplification as described by the manufacturers. In order to identify homologues of individual ORFs in C. dubliniensis, the entire C. albicans ORF was used in labelling reactions. In the case of gene families (e.g. CTA2, IFA, IFF), probes were designed by identifying the most conserved region of the ORFs in CLUSTAL_W alignments. These gene family probes represented >35 % of the entire ORF and constituted a different region of the ORF to that present on the microarray. The post-hybridization washes were performed at low stringency (60 °C with 0·5x SSC, 0·1 %, w/v, SDS) unless otherwise indicated.

PCR amplification of C. dubliniensis genome sequences.
PCR amplification from the C. dubliniensis genomic DNA template was carried out as described previously (Moran et al., 1998, 2002). Oligonucleotide primers used in this study were synthesized by Sigma-Genosys (Table 2). PCR-amplified DNA fragments were sequenced where indicated using the dideoxy chain-termination method by Lark Technologies.


View this table:
[in this window]
[in a new window]
 
Table 2. Sequences of oligonucleotide primers used in this study

 

   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Comparative genomic microarray hybridization
To label Candida chromosomal DNA efficiently by random priming with Cy3 or Cy5 dUTP, it was first necessary to fragment the chromosomal DNA. Two DNA fragmentation methods, sonication and restriction endonuclease digestion, were compared in order to determine if either labelling method introduced artifacts into the microarray results. In replicate experiments involving C. albicans SC5314 genomic DNA labelled with Cy3 dUTP following restriction endonuclease or sonication, over 98 % of spots gave fluorescent signals that were two standard deviations above background. Using the same criteria, C. dubliniensis CD36 genomic DNA prepared by either sonication or restriction endonuclease digestion hybridized to at least 95 % of gene-specific spots in replicate hybridizations. The dataset chosen for analysis here was generated with sonicated genomic DNA from C. albicans and C. dubliniensis labelled with either Cy5 dUTP or Cy3 dUTP in dye-swapped replicate experiments. In these hybridizations, 5931 duplicate spots (98 %) exhibited signals in the C. albicans channel two standard deviations above background and were included for analysis.

Relative hybridization efficiency of co-hybridized C. albicans SC5314 and C. dubliniensis CD36 genomic DNA to the microarrays was assessed by determining a normalized ratio of C. albicans and C. dubliniensis signal intensities at each spot included for analysis (Cy5/Cy3 normalized ratios). In dye-swapped replicate experiments, 96 % of these normalized ratios differed by less than twofold.

In order to investigate whether a relationship existed between the strength of hybridization of C. dubliniensis DNA to the array and the degree of nucleotide homology between the corresponding C. albicans and C. dubliniensis sequences, we plotted log2 ratio values versus percentage nucleotide sequence homology. We determined the nucleotide sequence homology between the C. albicans probe sequences present on the array and the corresponding sequences of 11 C. dubliniensis genes sequences available in GenBank (Table 1). We also attempted to PCR amplify sequences from C. dubliniensis using C. albicans-specific oligonucleotide primers (Table 2) corresponding to 35 genes which hybridized poorly with C. albicans genomic DNA (normalized ratios ranging from 0·17 to 0·53) on the microarray. Under low-stringency conditions, 10 of these PCR primer sets yielded PCR amplification products with sequences homologous to the corresponding C. albicans gene (nucleotide sequence homology range 59–80 %, Table 1). We plotted the percentage nucleotide sequence homology between the 11 GenBank and 10 PCR-amplified sequences and their C. albicans homologues versus the log2 ratio values from the 21 DNA spots representing these genes on the array (Fig. 1). Linear regression analysis was used to generate a best-fit-line from the dataset, which demonstrated a relationship between nucleotide sequence homology and log2 ratio (r2=0·80, P<0·0001). The 12 most homologous nucleotide sequences (80–98·8 % identity), including housekeeping genes such as ACT1, URA3 and ERG11, all had normalized ratio values of >0·5. The remaining nine genes formed a second group with intermediate sequence homologies of 59–79 %. These nine sequences possessed normalized ratios ranging from 0·17 to 0·46. Based on this analysis we categorized genes into three homology groups based on normalized ratio values: (I) a high-homology group (normalized ratio >0·5; >80 % nucleotide sequence homology), (II) a medium-homology group (normalized ratio 0·25–0·5; 60–80 % nucleotide sequence homology) and (III) sequences that possessed low homology or were possibly absent in C. dubliniensis (normalized ratio<0·25; <60 % nucleotide sequence homology).



View larger version (11K):
[in this window]
[in a new window]
 
Fig. 1. Standard curve used to determine the relationship between percentage nucleotide sequence homology of C. albicans SC5314 and C. dubliniensis CD36 sequences and normalized fluorescence ratio. GenBank sequences for 11 C. dubliniensis genes of known homology and 10 novel PCR-amplified sequences were included. The mean of the log2 ratio values (Table 1) for each gene was plotted against percentage nucleotide sequence homology. Linear regression analysis was used to predict the best-fitting line. The r2 value was calculated using Prism 4.0.

 
In total, 677 sequence-specific spots exhibited normalized ratios below 0·5 (P<0·05) in replicate dye-swap array experiments with sheared genomic DNA. From this group, 418 sequences were classified as likely to possess intermediate nucleotide sequence homology (60–80 %). The remaining 259 sequences (representing 4·4 % of the spots analysed) gave normalized ratios of <0·25 and were predicted to possess low nucleotide sequence homology (<60 %) or were possibly absent in C. dubliniensis.

Categorization of divergent genes
We decided to examine the group of sequences predicted to possess low nucleotide sequence homology in more detail as these genes were most likely to be functionally different or possibly even absent in C. dubliniensis. This group of 259 sequences was found to correspond to 247 genes (Table 3) since 12 genes were found to be represented on the array by two different duplicate spots. However, 134 (54 %) of these were hypothetical genes with no homology to genes of known function. Within this group, 43 genes could be classified as conserved hypothetical genes due to significant homology to other hypothetical genes in GenBank or CandidaDB (Table 4). Twenty-one sequences (8·5 %) were found to have homology to genes encoding C. albicans retrotransposon elements, including transposases and reverse transcriptases. Two sequences identified, IPF17727 and HOK, are contained in the C. albicans-specific fingerprinting probes 27A and Ca3 (Soll, 2000). IPF17727 is a region of the C. albicans RPS region and the HOK gene is found within the C. albicans-specific fingerprinting probe Ca3 that has previously been shown to hybridize very poorly to C. dubliniensis genomic DNA (Sullivan et al., 1995).


View this table:
[in this window]
[in a new window]
 
Table 3. Functional categories of C. albicans genes predicted to be of low nucleotide sequence homology or absent in C. dubliniensis (normalized ratio<0·25)

 

View this table:
[in this window]
[in a new window]
 
Table 4. C. albicans SC5314 genes predicted to be of low homology (<60 % nucleotide sequence identity) or absent in C. dubliniensis CD36

 
(i) Putative transcriptional regulators.
A group of 19 genes were identified that possessed homology to putative transcriptional regulators. Eight of these regulators had strong homology to genes encoding transcriptional activators in Saccharomyces cerevisiae with zinc-finger DNA-binding motifs. A further eight corresponded to a family of genes encoding proteins with homology to a putative C. albicans transcriptional activator, CTA2 (Kaiser et al., 1999). All eight CTA2-like genes included for analysis exhibited normalized ratios of <0·25. A CLUSTAL_W-generated alignment of the nucleotide sequences of CTA21, CTA22, CTA25 and CTA26 from C. albicans revealed that these sequences were 89 % identical. A PCR primer pair (CTA2F/CTA26R, Table 2) homologous to conserved sequences in these ORFs did not yield amplimers from C. dubliniensis genomic DNA template at a primer annealing temperature of 50 °C. A CTA2 probe was PCR amplified from C. albicans template DNA using the CTA2F/CTA2R primer pair. This probe which corresponds to >90 % of the CTA2 ORF was at least 90 % homologous to six C. albicans CTA2-like sequences annotated in the CandidaDB. In Southern hybridization experiments this probe revealed multiple hybridizing fragments in EcoRI- or HindIII-digested C. albicans genomic DNA (Fig. 2). Hybridization of the same sequence to C. dubliniensis CD36 genomic DNA digested with EcoRI or HindIII at low stringency did not reveal any hybridizing fragments (Fig. 2). In addition, this CTA2 probe did not hybridize to genomic DNA from seven additional epidemiologically unrelated C. dubliniensis isolates in Southern hybridization experiments (data not shown). These findings are in agreement with the array data that members of this gene family are significantly divergent (i.e. share low nucleotide sequence homology) or are absent in C. dubliniensis.



View larger version (44K):
[in this window]
[in a new window]
 
Fig. 2. Southern hybridization analysis of C. albicans and C. dubliniensis DNA with a DIG-11-dUTP-labelled probe homologous to nucleotides +1 to +781 of CTA26. The blot contains C. dubliniensis CD36 genomic DNA and C. albicans SC5314 genomic DNA digested with EcoRI and HindIII, as indicated. Molecular size markers are indicated on the left. Washes were performed at reduced stringency (60 °C in 0·5x SSC).

 
(ii) Putative membrane transporters.
Eight genes encoding proteins with strong homology to transporters of nutrients or other small molecules were also found to hybridize poorly with C. dubliniensis genomic DNA. These included the genes encoding the oligopeptide transporter OPT1, the choline transporters HNM3 and HNM4, the uracil permease FUR4 and the allantoin permease DAL52. The absence of homologous sequences for this group of genes was confirmed in C. dubliniensis by Southern hybridization analysis with probes corresponding to the complete C. albicans ORF (Fig. 3).



View larger version (65K):
[in this window]
[in a new window]
 
Fig. 3. Southern hybridization analysis of C. albicans and C. dubliniensis DNA with [{alpha}-32P]dATP-labelled probes corresponding to the complete ORF sequences of the C. albicans genes OPT1, FUR4, HNM3 and HNM4. The ORFs were amplified from C. albicans genomic DNA with the primer sets OPTA/B, FUR4A/B, HNM3A/B and HNM4A/B (Table 2). Each blot contains EcoRI-digested genomic DNA from C. albicans SC5314, C. dubliniensis CD36 and C. dubliniensis CD514. Molecular size markers are indicated on the left. Washes were performed at reduced stringency (60 °C in 0·5x SSC).

 
(iii) A leucine-rich repeat family of proteins.
A large family of genes encoding proteins with leucine-rich repeats (termed the IFA family, Pasteur CandidaDB) in C. albicans was also identified. Of 26 IFA sequences included for analysis, 18 gave normalized ratios of <0·25 following hybridization of C. dubliniensis genomic DNA. Four IFA sequences gave intermediate ratios (0·25–0·5) and four sequences (IFA3, IFA13, IFA20 and IFA21) gave ratios above 0·5. A CLUSTAL W-generated alignment of four IFA sequences with ratios of <0·25 (IFA1, IFA2, IFA4 and IFA5) revealed that the 3' half of these ORFs exhibited the highest levels of conservation. In order to detect sequences homologous to IFA genes in C. dubliniensis we amplified a conserved region of 690 bp representing 34 % of the IFA1 ORF using the primer pair IFA1F/IFA1R (Table 2). Southern hybridization experiments with this conserved C. albicans sequence revealed the presence of a single hybridizing 1·5 kb EcoRI fragment in CD36 and in seven additional epidemiologically unrelated C. dubliniensis isolates (data not shown). We applied the IFA1F/IFA1R primer pair to CD36 template DNA in low-stringency PCR reactions (50 °C annealing temperature) and successfully amplified a 728 bp fragment with 70 % homology to the C. albicans IFA8 ORF, in keeping with its ratio value between 0·25 and 0·5.

(iv) Genes encoding GPI-anchored proteins.
Genes encoding GPI-anchored proteins, including the hypha-specific protein Hyr1 (Bailey et al., 1996), were also identified in our CGH analysis. Two sequences, representing the 3' end and an internal fragment of the HYR1 gene, were present on the array. Both sequences yielded normalized ratios of <0·3, and no homologous sequence could be identified in C. dubliniensis by low-stringency Southern blot analysis of C. dubliniensis CD36 genomic DNA with PCR-amplified C. albicans HYR1 gene sequences (nucleotides +95 to +1183, representing 40 % of the ORF, chosen because the domain bears most homology to other GPI-anchored protein-encoding genes in C. albicans) (Fig. 4a). This was confirmed in Southern hybridization experiments with seven additional epidemiologically unrelated C. dubliniensis isolates (data not shown).



View larger version (105K):
[in this window]
[in a new window]
 
Fig. 4. Southern hybridization analysis of C. albicans and C. dubliniensis genomic DNA with sequences corresponding to highly conserved regions of C. albicans GPI-anchored protein encoding genes. Panel (a) was hybridized with an [{alpha}-32P]dATP-labelled probe corresponding to nucleotides +95 to +1183 of the C. albicans HYR1 gene. Panel (b) was hybridized with an [{alpha}-32P]dATP-labelled probe corresponding to nucleotides +1 to +844 of IFF1. Each blot contains C. dubliniensis CD36 genomic DNA and C. albicans SC5314 genomic DNA digested with EcoRI and HindIII, as indicated. Molecular size markers are indicated on the left. Washes were performed at low stringency (60 °C in 0·5x SSC).

 
Several genes encoding Hyr1-related proteins (termed the IFF gene family, Pasteur CandidaDB) were also present on the array. Of the nine IFF sequences included for analysis (IFF2 to IFF11) only IFF10.5 and IFF11 gave ratios above 0·5. Sequence alignments of C. albicans IFF genes generated with CLUSTAL_W identified IFF1 as the most likely ancestral gene based on its homology to other members, with the most conserved sequences present in the 5' region. As sequences homologous to this region of IFF1 were not included on the Eurogentec array, we amplified 844 bp at the 5' end of IFF1 (~50 % of the ORF) from C. albicans SC5314 with the primer pair IFF1F/IFF1R (Table 2) and hybridized it to C. dubliniensis CD36 genomic DNA to reveal a single hybridizing band (Fig. 4b). We could detect this hybridizing fragment in the genomic DNA of seven additional unrelated C. dubliniensis isolates in Southern hybridization experiments. We used the IFF1F/IFF1R primer pair in PCR amplification reactions using C. dubliniensis CD36 template DNA and successfully generated a 750 bp amplimer with 86 % homology to C. albicans IFF1.

Other sequences with characterized gene products or functions that could be inferred from homology searches included genes involved in biotin synthesis (BIO3, BIO4) and several unrelated genes encoding metabolic enzymes (Table 4).

Putative virulence factors
The dataset of genes with normalized ratios of <0·5 was searched to identify genes which had previously been associated with C. albicans virulence.

(i) Genes encoding putative adhesins.
Of the eight sequences with homology to members of the ALS gene family of GPI-anchored proteins (encoding putative adhesins) included for analysis, all gave normalized ratios of <0·5, with sequences specific for ALS1, ALS5, ALS6 and ALS7 yielding normalized ratios of <0·25 (Hoyer, 2001). Spots homologous to ALS2, ALS3 and ALS9 were excluded from our analysis due to poor hybridization with C. albicans genomic DNA.

We also investigated whether sequences encoding another group of GPI-anchored proteins were present, namely HWP1, which encodes a hyphal adhesin, and related sequences, RBT1 and IPF14331 (Braun et al., 2000; Staab et al., 1999). HWP1 has been associated with virulence in C. albicans by mediating adhesion to epithelial cells (Staab et al., 1999). HWP1 and RBT1 both yielded normalized ratios of <0·5 (0·48 and 0·37, respectively) in all experiments with C. dubliniensis CD36 genomic DNA. In order to identify a homologue of HWP1, a primer set designed based on the C. albicans HWP1 sequence (HWP1F/HWP1R, Table 2) was used to amplify a 1·3 kb region of C. dubliniensis genomic DNA. The putative 5' upstream region was also amplified with primers designed based on the sequence of the corresponding C. albicans region (APL6F and HWPR2, Table 2). An ORF of 1266 bp with homology to C. albicans HWP1 was identified in these sequences. However, the ORF shared only 49 % identity with the nucleotide sequence of C. albicans HWP1 due to the presence of several large deletions within the coding sequence (GenBank accession no. AJ632273). The overlapping 5' region amplified from C. dubliniensis contained upstream sequences homologous to the C. albicans APL6 gene. This synteny between HWP1 and APL6 is conserved in C. albicans, and provides further evidence that this gene is a C. dubliniensis HWP1 homologue. However, the predicted protein encoded by the C. dubliniensis gene was 421 amino acids in length, 213 residues shorter than the C. albicans protein. The first 50 residues of each protein were highly homologous, both containing the KR signature of the KEX2 cleavage site (Fig. 5a). However, the remainder of the N-terminal half of CdHwp1p contained several large deletions compared to the C. albicans protein, including most of the region rich in proline, glutamine and aspartate residues (Fig. 5a) (Sundstrom, 2002). Two of these deletions (of 89 bp and 119 bp) spanned the region homologous to the microarray probe and were likely to be responsible for the low signal detected with C. dubliniensis genomic DNA. Further deletions were found in the serine-threonine-rich region, although the {omega}-site for GPI-anchor addition was conserved.



View larger version (16K):
[in this window]
[in a new window]
 
Fig. 5. (a) Diagram illustrating regions of homology to C. albicans CaHwp1p and the extent of deletions in the predicted CdHwp1p protein sequence. The upper rectangular box represents the CaHwp1p protein and shows the position of the KEX2 cleavage site (arrow), the recombinant rHwp1p domain (shaded area) shown to possess transglutaminase substrate activity (Sundstrom, 2002), the serine-threonine-rich region (Ser-Thr rich) and the carboxy-terminal {omega}-site. The lower boxes represent the homologous regions of the predicted C. dubliniensis CdHwp1p protein. The numbers in parentheses indicate the positions of the homologous C. dubliniensis protein domains relative to the corresponding C. albicans amino acid residues. (b, c) Southern hybridization analysis of C. albicans SC5314 and C. dubliniensis CD36 genomic DNA with sequences corresponding to (b) HWP1 and (c) RBT1. DNA in (b) was hybridized with an [{alpha}-32P]dATP-labelled probe corresponding to the entire C. dubliniensis HWP1 ORF. DNA in (c) was hybridized with an [{alpha}-32P]dATP-labelled probe corresponding to nucleotides +694 to +1410 of RBT1 amplified from C. albicans genomic DNA. Lanes 1 and 2 in both panels contain EcoRI-digested genomic DNA from C. albicans and C. dubliniensis, respectively. Molecular size markers are indicated on the left. Washes were performed at reduced stringency (60 °C in 0·5x SSC).

 
Similarly, a C. dubliniensis sequence 513 bp in length was PCR-amplified using the primers RBTF2/RBTR2 designed based on the C. albicans RBT1 sequence (Table 2). This sequence was 45 % identical to base pairs +580 to +1410 of the C. albicans RBT1 ORF, the region encoding the internal serine-threonine-rich domain of the C. albicans protein. Like the CdHWP1 gene, this sequence also contained deletions (of 52 bp and 82 bp), when aligned to the homologous C. albicans sequence, that were responsible for the poor hybridization of C. dubliniensis DNA to the CaRBT1 array spots.

Southern blot analysis was performed to determine whether single or multiple HWP1 and RBT1 homologues could be detected in C. dubliniensis. Hybridization of the C. dubliniensis HWP1 amplified sequence to EcoRI-digested C. albicans genomic DNA revealed a single hybridizing fragment of 4 kb (Fig. 5b). In C. dubliniensis genomic DNA, a strongly hybridizing fragment of 9 kb was detected and a second, weakly hybridizing fragment of 5 kb in EcoRI-digested DNA (Fig. 5b). This second fragment was identical in size to the fragment detected in Southern blots of C. dubliniensis DNA with sequences corresponding to CaRBT1 (Fig. 5c), indicating that this second hybridizing fragment was likely to correspond to CdRBT1.

(ii) Secreted aspartyl proteinases.
Sequences homologous to the 10 C. albicans secreted aspartyl proteinase (SAP)-encoding genes (SAP1 to SAP10) were included on the arrays. All of the SAP genes with the exception of SAP4, SAP5 and SAP6 gave normalized ratios of >0·6. SAP5 gave a mean ratio of 0·38 (among genes with intermediate homology) in all C. dubliniensis CD36 experiments. We probed the C. dubliniensis genome for homologues of SAP4–6 using PCR primers homologous to conserved regions of these genes (SAP4-6F/SAP4-6R, Table 2). Amplification using C. dubliniensis genomic DNA as the template yielded a PCR product of 750 bp that shared 86 % identity with SAP4 and SAP6. Using an inverse PCR strategy (primers InvSAPF/InvSAPR, Table 2) we amplified flanking sequences from C. dubliniensis genomic DNA to obtain the complete ORF (GenBank accession no. AJ634382). This C. dubliniensis gene was found to lie upstream of the C. dubliniensis homologue of the C. albicans SAP1 gene and was equally homologous to SAP4 and SAP6 (~85 %). This ORF was designated CdSAP4 as the synteny at this locus with SAP1 is identical to that at the SAP4 locus in C. albicans. We used the entire C. dubliniensis SAP4 gene as a probe in Southern blots with C. albicans and C. dubliniensis genomic DNA in order to identify SAP5 and SAP6 homologues in C. dubliniensis. The SAP4–6 genes in C. albicans are highly homologous (89 % nucleotide sequence identity), so we anticipated that the C. dubliniensis SAP4 gene should hybridize strongly to any C. dubliniensis SAP5 or SAP6 homologues. Indeed, the C. dubliniensis SAP4 gene hybridized to four separate KpnI fragments in C. albicans genomic DNA. Three of these fragments corresponded in length to the predicted sizes of KpnI fragments expected to contain homologues of SAP4, SAP5 and SAP6 based on the C. albicans genome sequences at the Stanford Genome Technology Center. The SAP4-hybridizing fragment was less intense than the SAP5- and SAP6- containing fragments. However, a fourth hybridizing fragment of similar intensity at 7 kb was also detected and could contain a second SAP4 allele due to an RFLP (Fig. 6). The C. dubliniensis SAP4 gene was hybridized to C. dubliniensis genomic DNA digested with KpnI, HindIII and several restriction endonucleases that do not cleave within the CdSAP4 ORF (BglII, SpeI, SalI, XbaI); this revealed only one significantly hybridizing band in C. dubliniensis genomic DNA (Fig. 6). Furthermore, hybridization of the C. albicans SAP5 and SAP6 genes to C. dubliniensis DNA resulted in hybridization to the same restriction fragment harbouring the C. dubliniensis SAP4 gene (data not shown). These findings were confirmed by Southern blot analysis on eight epidemiologically unrelated isolates of C. albicans and C. dubliniensis (data not shown).



View larger version (106K):
[in this window]
[in a new window]
 
Fig. 6. Southern hybridization analysis of C. albicans and C. dubliniensis genomic DNA with sequences corresponding to the C. dubliniensis CdSAP4 gene. The blot was hybridized with an [{alpha}-32P]dATP-labelled probe of the entire CdSAP4 ORF. The blot contains genomic DNA from C. albicans SC5314 digested with KpnI and genomic DNA from C. dubliniensis CD36 digested with KpnI, BglII, HindIII, SpeI, SalI or XbaI, as indicated. The markers on the left side of the panel indicate the predicted positions of the SAP4, SAP5 and SAP6 genes in C. albicans SC5314. Molecular size markers are indicated on the right. Washes were performed at reduced stringency (60 °C in 0·5x SSC).

 
Hybridization of a second C. dubliniensis strain to microarrays
In order to confirm the above data obtained with C. dubliniensis CD36 and to investigate the levels of intraspecies variation between unrelated C. dubliniensis strains, we hybridized genomic DNA from a second C. dubliniensis isolate, CD514, to these arrays. We chose this strain as it has been shown to be genetically unrelated to C. dubliniensis CD36 based on the DNA fingerprint pattern obtained with the C. dubliniensis fingerprint probe Cd25. Sheared genomic DNA from C. dubliniensis CD514 was labelled with Cy3 and co-hybridized with Cy5-labelled DNA from C. albicans SC5314. We compared the dataset from CD514 with that generated from CD36 in order to identify genes unique to each strain. Only three additional genes were discovered that hybridized significantly to CD514 DNA (ratio>0·59) and that were deemed absent in CD36 (normalized ratio<0·2, P<0·035). These genes were IPF4450 and IPF17652.3, with homology to an integrase and a reverse transcriptase, respectively, and the oligopeptide transporter encoding gene OPT1. The presence of the OPT1 sequence in CD514 and its absence in CD36 was confirmed by Southern hybridization with the C. albicans OPT1 sequence (Fig. 2). Conversely, only one sequence, encoding GIT1 (glycerophosphoinositol transporter), was identified which failed to hybridize with CD514 DNA (ratio 0·112, P value 0·008) and gave significant signals with CD36 DNA (ratio 1·1).


   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Phylogenetic analysis of rRNA sequences has confirmed that C. dubliniensis and C. albicans are the two most closely related Candida species of clinical importance in humans (Sullivan et al., 1995, 2004). However, whereas C. albicans is the most significant yeast pathogen, responsible for superficial and deep-seated infections, C. dubliniensis is of lesser clinical importance in mucosal infections in non-HIV-infected patients, and is relatively insignificant in the case of bloodstream infection (Kibbler et al., 2003; Meis et al., 2000). This apparent lower virulence of C. dubliniensis is also evident in data from animal model infection studies. However, the exact reasons why C. albicans is more virulent are not clear. In this study we have utilized recently available C. albicans whole-genome DNA microarrays to investigate and identify genomic differences between C. albicans and C. dubliniensis that could account, at least in part, for the enhanced virulence potential of C. albicans relative to C. dubliniensis and for the differences in epidemiology between the two species.

The findings presented in this study obtained by CGH reinforce the phylogenetic data that originally implied the close relatedness of the two organisms (Gilfillan et al., 1998; Sullivan et al., 1995). Our data show that only 4·4 % of C. albicans sequences analysed in our studies (normalized ratio<0·25) were likely to be absent or highly divergent (<60 % homologous at the nucleotide sequence level) in C. dubliniensis. That the vast majority of C. albicans genes are highly conserved in C. dubliniensis indicates that the two species have probably only diverged relatively recently and thus are likely to inhabit similar environments in the human body. Thus only a small subset of C. albicans genes seem to be unique to this species and could possibly be important contributory factors to the greater success of C. albicans as a commensal on the human mucosal epithelium and as a pathogen in compromised hosts. Surprisingly, some spots exhibited reproducibly higher fluorescent signals with C. dubliniensis genomic DNA. There are several possible reasons for these findings, including cross-hybridization of paralogous C. dubliniensis sequences to the same spot, increases in gene copy number or differences in ploidy.

Of the 247 C. albicans genes identified that were predicted to have less than 60 % homology at the nucleotide sequence level or even possibly be absent in C. dubliniensis, 134 were hypothetical genes of unknown function (Table 3). However, 43 of these hypothetical ORFs were conserved, with homology to genes in S. cerevisiae, Aspergillus nidulans or paralogous sequences in the C. albicans genome. Of the 113 genes with a confirmed or hypothetical function, few were identified that corresponded to housekeeping genes involved in central metabolism, cell structure, or molecular biosynthesis. However, several transporter-encoding genes involved in nutrient uptake were identified as being absent, including two of the four genes encoding choline permeases in C. albicans (HNM3, HNM4), a uracil permease (FUR4) and an allantoin permease (DAL52). The only confirmed intrastrain difference between the two Cd25 fingerprint group C. dubliniensis strains analysed here was the presence of sequences homologous to the oligopeptide transporter OPT1 (Lubkowitz et al., 1997) in the Cd25 fingerprint group II strain CD514 which were absent in the fingerprint group I isolate CD36. It is not known whether the absence of these genes could affect the ability of C. dubliniensis to grow relative to C. albicans in vivo, since, for example, our data suggest other genes encoding choline (HNM2) and allantoin permeases (DAL51) are likely to be present in C. dubliniensis. C. dubliniensis also seems to lack sequences involved in the biosynthesis of biotin (BIO3 encoding DAPA aminotransferase and BIO4 encoding dethiobiotin synthetase). Although biotin is required for growth, C. albicans and C. dubliniensis probably acquire sufficient biotin from exogenous sources in the oral cavity, most likely from commensal bacteria (Phalip et al., 1999).

Twenty-one sequences corresponded to genes present in retrotransposons of C. dubliniensis, indicating that since their divergence the genomes of the two species may have acquired different mobile genetic elements.

Seventeen sequences homologous to genes encoding various GPI-anchored proteins were identified in our analysis as being absent or of low homology in C. dubliniensis by a combination of array hybridization data, PCR analysis and Southern blot analysis (Sundstrom, 2002). Poor C. dubliniensis hybridization signals were detected from sequences homologous to the C. albicans hyphal specific HYR1 gene (mean ratio 0·13), and no homologous gene was identified in C. dubliniensis following Southern hybridization with conserved C. albicans HYR1 sequences (Bailey et al., 1996). Sequences corresponding to several HYR1-related GPI-anchored proteins in C. albicans also exhibited poor hybridization signals with C. dubliniensis genomic DNA (IFF family genes). Although specific functions have not been assigned to proteins encoded by these genes, their likely location on the cell surface indicates possible roles in maintaining cell wall integrity, environmental signalling or adhesion to host surfaces.

Interestingly, subsequent analysis (Southern blotting) with sequences homologous to the C. albicans IFF1 gene (absent from Eurogentec microarrays) identified a homologous gene in C. dubliniensis for which sequences were later identified by PCR. These data suggest that at least one IFF-like gene is present in the C. dubliniensis genome. This may represent an ancestral IFF-related gene; however, additional IFF-related genes may be present in the C. dubliniensis genome but may be difficult to detect by CGH as they could have diverged more extensively than essential housekeeping genes with greater sequence-based constraints on protein function. A similar conclusion could be reached with regard to sequences homologous to genes encoding proteins of the {alpha}-agglutinin-like ALS family of adhesins. The ALS probes on the Eurogentec arrays used in this study consist of sequences from the 3' region of the ORFs, which within the ALS family are the least conserved regions (Hoyer et al., 2001). By Southern hybridization analysis Hoyer and co-workers noted that the 3' regions of the C. albicans ALS genes are poorly conserved in C. dubliniensis. We observed low hybridization ratios (<0·25) for several members of this family including ALS1, ALS5, ALS6 and ALS7. However, Hoyer and co-workers identified partial 5' nucleotide sequences for three ALS homologues in C. dubliniensis (ALSD1, ALSD2 and ALSD3). Their study demonstrated that ALSD1 is closely related to ALS6 and ALSD3 is closely related to ALS4. The present study confirms their findings that the 3' regions of the ALS genes are poorly conserved in C. dubliniensis, but does not provide further evidence for the existence of other C. dubliniensis ALS homologues. Since the microarray DNA spots correspond to 300–400 bp regions of each gene, our data reflect differences present in these regions only.

As the CGH data obtained for the ALS gene family demonstrate, data indicating the absence or divergence of a particular gene require confirmation as these regions may encompass non-conserved regions of the gene. Conversely, there may be divergent regions in many genes that remain undetected as they lie outside the regions compared here. Similarly, minor genetic differences (e.g. point mutations) and differences in non-translated regions that cannot be detected using these methods could also influence virulence and epidemiology. In addition, phenotype can also be influenced by post-transcriptional events unrelated to DNA sequence.

Low hybridization ratios were also observed for the HWP1 gene and the related sequences RBT1 and IPF14331 (Braun & Johnson, 1997; Staab et al., 1999; Sundstrom, 2002). The HWP1-encoded protein has been identified as a hypha-specific substrate for host transglutaminases involved in covalent adhesion to host cells. However, the functions of the other two gene products are as yet uncharacterized. In this study we identified the C. dubliniensis HWP1 homologue. The C. dubliniensis HWP1 gene hybridized poorly to the C. albicans array HWP1 sequences due to the presence of large deletions in the C. dubliniensis ORF. The predicted translated protein encoded by CdHWP1 contains several large deletions compared to the C. albicans protein (Fig. 5a). These deletions lie within the N-terminal glutamine- and proline-rich repeat domain containing the transglutaminase substrate activity and the internal serine- and threonine-rich domain. It will be of interest to determine whether the transglutaminase substrate activity of the C. dubliniensis homologue is affected by the presence of deletions in glutamine-rich regions of the N-terminus. A defect in the ability of C. dubliniensis to form stable attachments to oral epithelium may partly explain its reduced prevalence in the oral cavities of healthy individuals and patients with oral disease.

One of the most intensely studied virulence attributes of C. albicans is the ability to secrete aspartyl proteinases, encoded by 10 separate genes (Naglik et al., 2003). Sequences from all 10 SAP genes were present on the array. Sequences from only one of these genes, SAP5, gave consistently low signals from C. dubliniensis hybridizing DNA. SAP5 is a member of the SAP4–6 subfamily of proteinases, which are all highly homologous at the nucleotide sequence level and preferentially expressed by hyphae (Hube et al., 1994). In our efforts to determine if SAP5 was present in C. dubliniensis, we identified a gene that is most homologous to SAP4 and SAP6, which we designated CdSAP4 as the ORF was located upstream of CdSAP1, identical to the synteny observed in C. albicans (Miyasaki et al., 1994). Southern hybridization analysis with this CdSAP4 sequence revealed that it could hybridize to multiple fragments of C. albicans restriction-endonuclease-digested DNA corresponding to sequences of SAP4, SAP5 and SAP6. Such cross-hybridization is likely to be responsible for the strong signal detected from spots representing SAP6 on the C. albicans microarray. However, the CdSAP4 sequence consistently hybridized to only one single band (between ~2 and 10 kb) in Southern hybridization experiments with C. dubliniensis genomic DNA. These data indicate that only one gene with strong homology to the SAP4–6 subfamily exists in C. dubliniensis. Attempts to identify the corresponding genomic loci of putative SAP5 and SAP6 genes in C. dubliniensis by low-stringency PCR were unsuccessful (data not shown).

Together, the SAP4–6 subfamily has been shown to play an important role in the establishment of C. albicans systemic infections in mice and SAP6 has been shown to be the most important gene within this family in the establishment of murine intraperitoneal infection (Felk et al., 2002; Sanglard et al., 1997). As C. dubliniensis only possesses one gene with homology to SAP4–6, this could partly explain why C. dubliniensis is less able than C. albicans to establish systemic infection. In vivo virulence studies and epidemiological data support this hypothesis, since C. dubliniensis is less virulent than C. albicans in a murine systemic model of infection and the incidence of recovery of this organism from human blood cultures is extremely low (Gilfillan et al., 1998; Kibbler et al., 2003). The absence of these hypha-specific proteinases in C. dubliniensis may affect the ability of its hyphae to penetrate host tissues, acquire nutrients or evade killing by macrophages. We are currently testing the role of C. dubliniensis Sap proteins in infection models.

Differences in gene regulation have not been explored in any great detail in C. dubliniensis to date. The array CGH data indicates the presence of genes homologous to many of the transcription factors involved in regulating hypha formation in C. albicans (e.g. EFG1, CPH1, TUP1). However, poor hybridization signals were detected from several other genes encoding putative transcriptional regulators that could affect regulatory circuits in C. dubliniensis. Eight of these sequences had homology to genes encoding proteins with zinc-finger DNA-binding motifs in S. cerevisiae. Another group of genes with homology to the putative C. albicans transcriptional activator encoding gene CTA2 were also identified. CTA2 (GenBank ID AJ006637)was identified by Kaiser et al. (1999) in a one-hybrid screen in S. cerevisiae for C. albicans proteins with transcriptional activating properties. A family of possibly up to 10 genes with strong homology to CTA2 has been identified in C. albicans. Twelve sequences homologous to these genes were included in our analysis and all gave normalized ratios of <0·25. Southern hybridization also failed to identify any sequences with significant homology to these genes in C. dubliniensis. Although the function of these proteins has yet to be confirmed, the absence or divergence of a large family of transcriptional activators in C. dubliniensis could have major implications for the growth and virulence of this fungus.

In the present study, C. albicans DNA microarrays facilitated a whole-genome comparison between C. albicans and its close relative C. dubliniensis. Our experiments have revealed the absence and divergence of several genes and gene families in C. dubliniensis. These include putative virulence factors and many genes specific for or preferentially expressed in the hyphal phase, such as SAP5, SAP6, HWP1 and HYR1. C. dubliniensis is generally less efficient than C. albicans at forming hyphae in response to serum and the absence of these hypha-regulated genes may also indicate that C. dubliniensis hyphae are less specialized as virulence-promoting structures (Gilfillan et al., 1998). We have endeavoured to confirm the absence or divergence of genes directly involved in virulence (e.g. HWP1, SAP5); however, conclusive confirmation of this data will have to await the completion of the C. dubliniensis genome sequencing project which is currently under way (http://www.sanger.ac.uk/Projects/C_dubliniensis/). When complete this project will facilitate more thorough comparative genomic analysis and will allow the identification of C. dubliniensis-specific genes that could be responsible for phenotypic differences. Until then, this dataset represents a framework for further investigations into genetic and phenotypic differences between C. albicans and C. dubliniensis.


   ACKNOWLEDGEMENTS
 
This study was supported by the Microbiology Research Unit, Dublin Dental School and Hospital.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Al Mosaid, A., Sullivan, D., Salkin, I. F., Shanley, D. & Coleman, D. C. (2001). Differentiation of Candida dubliniensis from Candida albicans on Staib agar and caffeic acid-ferric citrate agar. J Clin Microbiol 39, 323–327.[Abstract/Free Full Text]

Al Mosaid, A., Sullivan, D. J. & Coleman, D. C. (2003). Differentiation of Candida dubliniensis from Candida albicans on Pal's agar. J Clin Microbiol 41, 4787–4789.[Abstract/Free Full Text]

Alves, S. H., Milan, E. P., de Laet Sant'Ana, P., Oliveira, L. O., Santurio, J. M. & Colombo, A. L. (2002). Hypertonic Sabouraud broth as a simple and powerful test for Candida dubliniensis screening. Diagn Microbiol Infect Dis 43, 85–86.[CrossRef][Medline]

Bailey, D. A., Feldman, P. J. F., Bovey, M., Gow, N. A. R. & Brown, A. J. P. (1996). The Candida albicans HYR1 gene, which is activated in response to hyphal development, belongs to a gene family encoding yeast cell wall proteins. J Bacteriol 178, 5353–5360.[Abstract]

Braun, B. R. & Johnson, A. D. (1997). Control of filament formation in Candida albicans by the transcriptional repressor TUP1. Science 277, 105–109.[Abstract/Free Full Text]

Braun, B. R., Head, W. S., Wang, M. X. & Johnson, A. D. (2000). Identification and characterization of TUP1-regulated genes in Candida albicans. Genetics 156, 31–44.[Abstract/Free Full Text]

Coleman, D. C., Sullivan, D. J., Bennett, D. E., Moran, G. P., Barry, H. J. & Shanley, D. B. (1997). Candidiasis: the emergence of a novel species, Candida dubliniensis. AIDS 11, 557–567.[CrossRef][Medline]

Daran-Lapujade, P., Daran, J. M., Kotter, P., Petit, T., Piper, M. D. & Pronk, J. T. (2003). Comparative genotyping of the Saccharomyces cerevisiae laboratory strains S288C and CEN.PK113-7D using oligonucleotide microarrays. FEMS Yeast Res 4, 259–269.[CrossRef][Medline]

Dong, Y., Glasner, J. D., Blattner, F. R. & Triplett, E. W. (2001). Genomic interspecies microarray hybridization: rapid discovery of three thousand genes in the maize endophyte Klebsiella pneumoniae 342, by microarray hybridization with Escherichia coli K-12 open reading frames. Appl Environ Microbiol 67, 1911–1921.[Abstract/Free Full Text]

Donnelly, S. A., Sullivan, D. J., Shanley, D. B. & Coleman, D. C. (1999). Phylogenetic analysis and rapid identification of Candida dubliniensis based on analysis of ACT1 intron and exon sequences. Microbiology 145, 1871–1882.[Medline]

Felk, A., Kretschmar, M., Albrecht, A. & 7 other authors (2002). Candida albicans hyphal formation and the expression of the Efg1-regulated proteinases Sap4 to Sap6 are required for the invasion of parenchymal organs. Infect Immun 70, 3689–3700.[Abstract/Free Full Text]

Gallagher, P. J., Bennett, D. E., Henman, M. C., Russell, R. J., Flint, S. R., Shanley, D. B. & Coleman, D. C. (1992). Reduced azole susceptibility of Candida albicans from HIV-positive patients and a derivative exhibiting colony morphology variation. J Gen Microbiol 138, 1901–1911.[Medline]

Gee, S. F., Joly, S., Soll, D. R., Meis, J. F., Verweij, P. E., Polacheck, I., Sullivan, D. J. & Coleman, D. C. (2002). Identification of four distinct genotypes of Candida dubliniensis and detection of microevolution in vitro and in vivo. J Clin Microbiol 40, 556–574.[Abstract/Free Full Text]

Gilfillan, G. D., Sullivan, D. J., Haynes, K., Parkinson, T., Coleman, D. C. & Gow, N. A. R. (1998). Candida dubliniensis: phylogeny and putative virulence factors. Microbiology 144, 829–838.[Medline]

Hannula, J., Saarela, M., Dogan, B., Paatsama, J., Koukila-Kahkola, P., Pirinen, S., Alakomi, H. L., Perheentupa, J. & Asikainen, S. (2000). Comparison of virulence factors of oral Candida dubliniensis and Candida albicans isolates in healthy people and patients with chronic candidosis. Oral Microbiol Immunol 15, 238–244.[CrossRef][Medline]

Higgins, D. G. & Sharp, P. M. (1988). CLUSTAL: a package for performing multiple sequence alignment on a microcomputer. Gene 73, 237–244.[CrossRef][Medline]

Hoyer, L. L. (2001). The ALS gene family of Candida albicans. Trends Microbiol 9, 176–180.[CrossRef][Medline]

Hoyer, L. L., Fundyga, R., Hecht, J. E., Kapteyn, J. C., Klis, F. M. & Arnold, J. (2001). Characterization of agglutinin-like sequence genes from non-albicans Candida and phylogenetic analysis of the ALS family. Genetics 157, 1555–1567.[Abstract/Free Full Text]

Hube, B., Monod, M., Schofield, D. A., Brown, A. J. P. & Gow, N. A. R. (1994). Expression of seven members of the gene family encoding secretory aspartyl proteinases in Candida albicans. Mol Microbiol 14, 87–99.[Medline]

Jabra-Rizk, M. A., Ferreira, S. M., Sabet, M., Falkler, W. A., Merz, W. G. & Meiller, T. F. (2001). Recovery of Candida dubliniensis and other yeasts from human immunodeficiency virus-associated periodontal lesions. J Clin Microbiol 39, 4520–4522.[Abstract/Free Full Text]

Kaiser, B., Munder, T., Saluz, H.-P., Kunkel, W. & Eck, R. (1999). Identification of a gene encoding the pyruvate decarboxylase gene regulator CaPdc2p from Candida albicans. Yeast 15, 585–591.[CrossRef][Medline]

Kibbler, C. C., Seaton, S., Barnes, R. A., Gransden, W. R., Holliman, R. E., Johnson, E. M., Perry, J. D., Sullivan, D. J. & Wilson, J. A. (2003). Management and outcome of bloodstream infections due to Candida species in England and Wales. J Hosp Infect 54, 18–24.[CrossRef][Medline]

Kurzai, O., Heinz, W. J., Sullivan, D. J., Coleman, D. C., Frosch, M. & Muhlschlegel, F. A. (1999). Rapid PCR test for discriminating between Candida albicans and Candida dubliniensis isolates using primers derived from the pH-regulated PHR1 and PHR2 genes of C. albicans. J Clin Microbiol 37, 1587–1590.[Abstract/Free Full Text]

Lubkowitz, M. A., Hauser, L., Breslav, M., Naider, F. & Becker, J. M. (1997). An oligopeptide transporter gene from Candida albicans. Microbiology 143, 387–396.[Medline]

Meis, J. F. G. M., Lunel, F. M. V., Verweij, P. E. & Voss, A. (2000). One-year prevalence of Candida dubliniensis in a Dutch university hospital. J Clin Microbiol 38, 3139–3140.[Free Full Text]

Miyasaki, S. H., White, T. C. & Agabian, N. (1994). A fourth secreted aspartyl proteinase gene (SAP4) and a CARE2 repetitive element are located upstream of the SAP1 gene in Candida albicans. J Bacteriol 176, 1702–1710.[Abstract]

Moran, G. P., Sanglard, D., Donnelly, S. M., Shanley, D. B., Sullivan, D. J. & Coleman, D. C. (1998). Identification and expression of multidrug transporters responsible for fluconazole resistance in Candida dubliniensis. Antimicrob Agents Chemother 42, 1819–1830.[Abstract/Free Full Text]

Moran, G., Sullivan, D., Morrschhauser, J. & Coleman, D. (2002). The Candida dubliniensis CdCDR1 gene is not essential for fluconazole resistance. Antimicrob Agents Chemother 46, 2829–2841.[Abstract/Free Full Text]

Murray, A. E., Lies, D., Li, G., Nealson, K., Zhou, J. & Tiedje, J. M. (2001). DNA/DNA hybridization to microarrays reveals gene-specific differences between closely related microbial genomes. Proc Natl Acad Sci U S A 98, 9853–9858.[Abstract/Free Full Text]

Naglik, J. R., Challacombe, S. J. & Hube, B. (2003). Candida albicans secreted aspartyl proteinases in virulence and pathogenesis. Microbiol Mol Biol Rev 67, 400–428.[Abstract/Free Full Text]

Perea, S., Lopez-Ribot, J. L., Wickes, B. L., Kirkpatrick, W. R., Dib, O. P., Bachmann, S. P., Keller, S. M., Martinez, M. & Patterson, T. F. (2002). Molecular mechanisms of fluconazole resistance in Candida dubliniensis isolates from human immunodeficiency virus-infected patients with oropharyngeal candidiasis. Antimicrob Agents Chemother 46, 1695–1703.[Abstract/Free Full Text]

Pfaller, M. A. & Diekema, D. J. (2004). Twelve years of fluconazole in clinical practice: global trends in species distribution and fluconazole susceptibility of bloodstream isolates of Candida. Clin Microbiol Infect 10 Suppl 1, 11–23.[CrossRef][Medline]

Phalip, V., Kuhn, I., Lemoine, Y. & Jeltsch, J.-M. (1999). Characterization of the biotin biosynthesis pathway in Saccharomyces cerevisiae and evidence for a cluster containing BIO5, a novel gene involved in vitamer uptake. Gene 232, 43–51.[CrossRef][Medline]

Pinjon, E., Sullivan, D., Salkin, I., Shanley, D. & Coleman, D. (1998). Simple, inexpensive, reliable method for differentiation of Candida dubliniensis from Candida albicans. J Clin Microbiol 36, 2093–2095.[Abstract/Free Full Text]

Pinjon, E., Moran, G. P., Jackson, C. J., Kelly, S. L., Sanglard, D., Coleman, D. C. & Sullivan, D. J. (2003). Molecular mechanisms of itraconazole resistance in Candida dubliniensis. Antimicrob Agents Chemother 47, 2424–2437.[Abstract/Free Full Text]

Sanglard, D., Hube, B., Monod, M., Odds, F. C. & Gow, N. A. R. (1997). A triple deletion of the secreted aspartyl proteinase genes SAP4, SAP5 and SAP6 of Candida albicans causes attenuated virulence. Infect Immun 65, 3539–3546.[Abstract]

Sebti, A., Kiehn, T. E., Perlin, D., Chaturvedi, V., Wong, M., Doney, A., Park, S. & Sepkowitz, K. A. (2001). Candida dubliniensis at a cancer center. Clin Infect Dis 32, 1034–1038.[CrossRef][Medline]

Soll, D. R. (2000). The ins and outs of DNA fingerprinting the infectious fungi. Clin Microbiol Rev 13, 332–370.[Abstract/Free Full Text]

Southern, E. (1975). Detection of specific sequences among DNA fragments separated by gel electrophoresis. J Mol Biol 98, 503–517.[Medline]

Staab, J. F., Bradway, S. D., Fidel, P. L. & Sundstrom, P. (1999). Adhesive and mammalian transglutaminase substrate properties of Candida albicans HWP1. Science 283, 1535–1538.[Abstract/Free Full Text]

Staib, P., Moran, G. P., Sullivan, D. J., Coleman, D. C. & Morschhauser, J. (2001). Isogenic strain construction and gene targeting in Candida dubliniensis. J Bacteriol 183, 2859–2865.[Abstract/Free Full Text]

Sullivan, D. J., Westerneng, T. J., Haynes, K. A., Bennett, D. E. & Coleman, D. C. (1995). Candida dubliniensis sp. nov.: phenotypic and molecular characterization of a novel species associated with oral candidosis in HIV-infected individuals. Microbiology 141, 1507–1521.[Medline]

Sullivan, D. J., Moran, G. P., Pinjon, E., Al-Mosaid, A., Stokes, C., Vaughan, C. & Coleman, D. C. (2004). Comparison of epidemiology, drug resistance mechanisms, and virulence of Candida dubliniensis and Candida albicans. FEMS Yeast Res 4, 369–376.[CrossRef][Medline]

Sundstrom, P. (2002). Adhesion in Candida spp. Cell Microbiol 4, 461–469.[CrossRef][Medline]

Vilela, M. M., Kamei, K., Sano, A., Tanaka, R., Uno, J., Takahashi, I., Ito, J., Yarita, K. & Miyaji, M. (2002). Pathogenicity and virulence of Candida dubliniensis: comparison with C. albicans. Med Mycol 40, 249–257.[Medline]

Willis, A. M., Coulter, W. A., Sullivan, D. J., Coleman, D. C., Hayes, J. R., Bell, P. M. & Lamey, P. J. (2000). Isolation of C. dubliniensis from insulin-using diabetes mellitus patients. J Oral Pathol Med 29, 86–90.[CrossRef][Medline]

Received 7 April 2004; revised 22 July 2004; accepted 27 July 2004.



This Article
Abstract
Full Text (PDF)
Alert me when this article is cited
Alert me if a correction is posted
Citation Map
Services
Email this article to a friend
Similar articles in this journal
Similar articles in PubMed
Alert me to new issues of the journal
Download to citation manager
Google Scholar
Articles by Moran, G.
Articles by Sullivan, D.
Articles citing this Article
PubMed
PubMed Citation
Articles by Moran, G.
Articles by Sullivan, D.
Agricola
Articles by Moran, G.
Articles by Sullivan, D.


HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
INT J SYST EVOL MICROBIOL MICROBIOLOGY J GEN VIROL
J MED MICROBIOL ALL SGM JOURNALS
Copyright © 2004 Society for General Microbiology.