candida dubliniensis: phylogeny and putative virulence factors

Microbiology (1 998), 144,829-838 Printed in Great Britain

Candida dubhiensis: phylogeny and putative virulence factors

Gregor D. Gilfillan,' Derek J. Su l l i~an,~t~ Ken Haynest4 Tanya Parkin~on,~ David C. Coleman2t3 and Neil A. R. Gow'

Author for correspondence: Neil A. R. Gow. Tel: + 44 1224 273179. Fax: +44 1224 273144. e-mail: [email protected]

1 Department o f Molecular and Cell Biology, University of Aberdeen, Institute of Medical Sciences, Foresterhi I I, Aberdeen AB25 ZZD, UK

* Department of Oral Medicine and Oral Pathology, School of Dental Science, University of Dublin, Trinity College, Dublin 2, Republic of Ireland

3 The Moyne Institute of Preventive Medicine, Department of Microbiology, University of Dublin, Trinity College, Dublin 2, Republic of Ireland

Diseases and Bacteriology, Royal Postgraduate Medical School, Hammersmith Hospital, Du Cane Road, London, UK

5 Department o f Discovery Biology, Pfizer Central Research, Sandwich, Kent CT13 9NJ, UK

4 Department of Infectious

Candida dubhiensis is a recently identified species which is implicated in oral candidosis in HIV-infected and AIDS patients. The species shares many phenotypic characteristics with, and is phylogenetically closely related to, Candida albicans. In this study the phylogenetic relationship between these two species was investigated and a comparison of putative virulence factors was performed. Four isolates of C. dubhiensis from different clinical sources were chosen for comparison with two reference C. albicans strains. First, the distinct phylogenetic position of C. dubhiensis was further established by the comparison of the sequence of its small rRNA subunit with representative Candida species. The C. dubhiensis isolates formed true unconstricted hyphae under most induction conditions tested but failed to produce true hyphae when induced using N-acetylglucosamine. Oral C. dubhiensis isolates were more adherent to human buccal epithelial cells than the reference C. albicans isolates when grown in glucose and equally adherent when grown in galactose. The C. dubhiensis isolates were sensitive to f luconazole, itraconazole, ketoconazole and amphotericin B. Homologues of seven tested C. albicans secretory aspartyl proteinase (SAP) genes were detected in C. dubhiensis by Southern analysis. In vivo virulence assays using a systemic mouse model suggest that C. dubhiensis is marginally less virulent than C. albicans. These data further confirm the distinct phenotypic and genotypic nature of C. dubhiensis and suggest that this species may be particularly adapted to colonization of the oral cavity.

Keywords : aspartic proteinase, Candida dubliniensis, HIV and AIDS, rRNA gene sequences, virulence

INTRODUCTION

Numerous reports have described a significant increase in the incidence and diagnosis of opportunistic and systemic candidosis during the last decade (Hazen, 1995 ; Sullivan et al., 1996; Coleman et al., 1997a). The reasons for this increase are many, but one of the most important contributory factors has been the increase in numbers of immunocompromised individuals, particularly those

Abbreviations: BEC, buccal epithelial cell; SAP, secretory aspartyl proteinase.

The EMBL accession number for the sequence reported in this paper is x99399.

infected with HIV. Candida albicans is the most pathogenic species of the genus Candida and is the species most frequently associated with candidosis (Odds et al., 1990). However, recently the proportion of infections diagnosed as being caused by this species has decreased relative to the increasing incidence of infections apparently caused by other, less pathogenic species such as C. tropicalis, C. glabrata and C. krusei (Powderly, 1992; Hazen, 1995; Pfaller, 1996; Sullivan et al., 1996; Coleman et al., 1997b). The reasons for this epidemiological shift are not clear ; however, the decreased susceptibility of these species relative to C. albicans to commonly used antifungal agents such as fluconazole has been implicated as being at least partially responsible (Powderly, 1992; Chavanet et al., 1994).

~~ ~

0002-2202 0 1998 SGM 829

G. D. GILFILLAN a n d OTHERS

In addition to the reports of the increased detection of non-C. albicans species in cases of human infection, there have been several reports describing the emergence of what were originally described as 'atypical C. albicans' isolates (Mahrous et al., 1990; Schmid et al., 1992; Sullivan et al., 1993; McCullough et al., 1994, 1995; Boerlin et al., 1995; Le Guennec et al., 1995; Anthony et al., 1995). The isolates described by Sullivan et al. (1993), McCullough et al. (1994,1995) and Boerlin et al. (1995) were subjected to further detailed analysis (Sullivan et al., 1995, 1997). These isolates can be distinguished from conventional C. albicans isolates by a number of phenotypic and genetic characteristics. They produce germ tubes and chlamydospores, although the latter are produced abundantly and in unusual formations. They have distinct carbohydrate assimilation profiles, grow poorly or not at all at 42 "C and yield dark green colonies on primary isolation from clinical specimens on the discriminatory chromogenic medium CHROMagar Candida (Coleman et al., 1997b; Sullivan et al., 1997; Sullivan & Coleman, 1998). Furthermore, the atypical isolates fail to express /?- glucosidase activity whereas isolates of C. albicans do (Boerlin et al., 1995; Sullivan et a/., 1997; Sullivan & Coleman, 1998). The atypical isolates have a distinctive genomic organization when fingerprinted with the C. albicans-specific semi-repetitive probe 27A and oligo- nucleotides homologous to microsatellite sequences (Sullivan et al., 1995, 1997). Random Amplified Poly- morphic DNA (RAPD) and karyotype analysis also confirmed the distinct genetic nature of the atypical isolates. These data, combined with phylogenetic analysis based on the sequence of a portion of the large rRNA subunit gene, implied that these isolates represent a novel species of Candida, for which the name C. dubliniensis has been proposed (Sullivan et al., 1995). Further evidence that C. dubliniensis is distinct from C. albicans has since been provided by the discovery of sequence differences in non-functional stem-loops of group I introns present in the large rRNA subunit of both species (Boucher et al., 1996). Isolates of C. dubliniensis have mainly been recovered from the oral cavities of HIV-infected individuals (Coleman et al., 1997b). However, a small number of isolates have been recovered from the oral cavities and the vaginas of normal healthy individuals (Sullivan et al., 1995 ; Moran et al., 1997). Although most isolates to date have been recovered in Ireland, they have also been recovered from patients throughout Europe, North and South America and Australia (Sullivan et al., 1997). Recently, it has been demonstrated that although most isolates of C. dubliniensis are sensitive to the commonly used antifungal agent fluconazole, a significant proportion of isolates are resistant to the drug (Moran et al., 1997). Furthermore, some isolates of C. dubliniensis have been shown to become resistant on exposure to subinhibitory doses of fluconazole, suggesting that this may be the reason why this species has emerged in recent years (Moran et al., 1997). Because C. dubliniensis and C. albicans are phylo-

genetically closely related and share several phenotypic characteristics, further studies are required to elucidate the phenotypic and genetic relationships between these two species. The aims of the present study were twofold. First, to further investigate the phylogenetic position of C. dubliniensis in relation to other Candida species by comparing the sequences of the small rRNA subunit (small rRNA) genes from representative species. Com- parative sequence analysis of small rRNA genes has been used most extensively in phylogenetic studies with a wide variety of microbial species (Olsen & Woese, 1993). Second, to compare C . dubliniensis and C. albicans isolates with respect to potential virulence factors, such as adhesion, hyphal production and the possession of genes encoding secretory aspartyl proteinase (SAP) homologues.

METHODS

Yeast strains. Representative C. dubliniensis and control C. albicans isolates used in this study are shown in Table 1. The yeasts were grown routinely on Potato Dextrose Agar (PDA, Oxoid) at p H 5.5 and Yeast Peptone Dextrose (YPD) agar and broth [YPD (g 1-l) : yeast extract (Oxoid), 10; peptone (Oxoid), 20 ; glucose, 20 ; pH 5-5 ; 20 g agar (Oxoid) was added to the broth to make the solid medium] for 48 h at 37 "C. The identity of C. dubliniensis isolates was confirmed by APILAB ID32C (bioMerieux) carbohydrate assimilation assays, analysis of growth at 42°C and by DNA fingerprinting and karyotype analysis, according to the methods described by Sullivan et al. (1995). Chemicals, enzymes and radioisotopes. Analar-grade or molecular biology-grade chemicals were purchased from BDH, Boehringer Mannheim, Fisons and Sigma. Restriction endonucleases and Klenow fragment DNA polymerase were purchased from Boehringer Mannheim or Promega and were used as directed by the manufacturers. [ c ~ - ~ ~ P ] ~ C T P (3000 Ci mmol-l; 111 TBq mmol-l) used in random priming of DNA probes was purchased from Amersham. MIC antifungal assays. MIC assays were conducted for fluconazole, itraconazole, ketoconazole and amphotericin B using the broth dilution method (Pfaller et al., 1990). Hyphal formation. Hyphal formation assays were performed on cells grown to stationary phase in the defined medium of Lee et al. (1975) supplemented with 400 mM arginine, 0.001 YO (w/v) biotin and trace metals (0.2 mM ZnSO,, 0.25 mM CuSO,, 1 mM FeCl,, 1 mM MgCl,, 1 mM CaCl,). Cells were grown at either 25 or 37°C in orbital incubators (New Brunswick Scientific) set at 180 r.p.m. The pH/temperature shift induction of hyphae (Buffo et al., 1984) was performed by inoculating 25 ml supplemented Lee's medium, pH 6.5,37 "C with 0.5 ml of cells grown to stationary phase in supplemented Lee's medium, pH 4.5,25 "C (Buffo et a/., 1984). N-Acetyl-D- glucosamine (GlcNAc) induction of hyphal formation (Mattia et al., 1982) was performed by resuspending cells that had been grown to stationary phase in supplemented Lee's medium, pH 4.5,37 "C, washed and starved in distilled water for 24 h, into 25 ml basal salt medium [0*5% (w/v) (NH,),SO,, 0.02 YO (w/v) MgSO,, 0.5 '/O (w/v) NaCl, 0.001 '/O

biotin] with GlcNAc added to 4 mM, to a cell density of 2 x lo6 cells ml-l at 37 "C. Serum induction of hyphal formation (Gow & Gooday, 1982) was performed by resuspending cells that had been grown to stationary phase in supplemented Lee's medium, pH 4.5, 37 "C, and washed in

830

Candida dubliniensis virulence

Table 1. C. albicans and C. dubliniensis strains used

Species Strain Origin Reference

C. albicans 3153 Systemic isolate Odds (1974) C. albicans SC5314 Systemic isolate Gillum et al. (1984) C. dubliniensis CD36 Irish oral isolate from HIV+ patient Sullivan et al. (1995) C. dubliniensis CD41 Irish oral isolate from HIV- healthy subject Sullivan et al. (1995) C. dubliniensis CD57 Irish vaginal isolate from HIV- patient Moran et al. (1997) C. dubliniensis CD43 Irish oral isolate from HIV- intravenous drug user Sullivan et al. (1997)

distilled water, into water containing 10 '/o (v/v) newborn-calf serum (Sigma) to a cell density of 2 x lo6 cells ml-l at 37 "C. Adherence assays. Adherence assays were conducted on yeast cells grown for exactly 24 h in medium containing 0.67% (w/v) yeast nitrogen base without amino acids (Difco) and either 50 mM glucose or 500 mM galactose (Douglas et al., 1981). Yeast cells (1 x lo8 ml-l) and human buccal epithelial cells (BECs; 1 x lo5 ml-l) were incubated together for 45 min at 37 "C, then recovered on polycarbonate filters (pore size 12 pm; Costar) and Gram-stained. Adherence assay results are presented as the mean number of yeasts adhering to 100 BECs, with samples being taken in triplicates of 100 BECs and on 3 consecutive days. Isolation of plasmid and genomic DNA. Genomic DNA was isolated from spheroplasts of Candida isolates by the method described by Rose et al. (1990). Plasmid DNA was isolated from Escherichia coli DH5a by the alkali lysis method described by Kraft et al. (1988). Cloning and sequencing of the small rRNA gene from C. dubliniensis strain CD36. The gene encoding the C. dubliniensis small rRNA was amplified using primers specific for the conserved sequences at the 3' and 5' ends of eukaryotic small rRNA genes (primers A and B, Medlin et al., 1988). The amplification reaction was performed in a Perkin Elmer 9600 thermal DNA cycler in a 100 pl volume containing 10 mM Tris/HCl, pH 8.4, 50 mM KCl, 2 mM MgCl,, 200 mM each of dATP, dTTP, dGTP and dCTP, 20 mM each of primer A and B and 2.5 U Taq polymerase. The cycling conditions used were 95 "C for 2 min, 37 "C for 2 min and 72 "C for 6 min for 30 cycles. Following amplification, a 90 pl aliquot of the amplified PCR mix was added to 210 p1 ddH,O and purified on a Microcon 100 column (Amicon) as recommended by the manufacturers. The purified amplimer was cloned into T- tailed pGEM3Zf and subsequently sequenced using the PRISM Ready Reaction DyeDeoxy Terminator Cycle Sequencing Kit (Applied Biosystems) . The sequencing primers used were 395-343R, 536-519R, 705-689R and 1242-1255R (Barns et al., 1991) and the M13 forward and reverse primers. All sequencing was performed on a 373A STRETCH auto- mated sequencer (Applied Biosystems) and the results analysed using the computer program SEQED. Comparison of small rRNA gene nucleotide sequences from various species was performed using multiple sequence alignments generated using the program CLUSTAL w. The sequences compared included those from C. dubliniensis (this study; accession no. X99399), C. albicans (Hendriks et al., 1989; X53497), C. tropicalis (Hendriks et al., 1991; M55527), C. glabrata (Wong & Clark- Walker, 1990; X51831), C. lusitaniae (Hendriks et al., 1991 ; M55526), C. krusei (Hendriks et al., 1991; M5552S) and Saccharomyces cerevisiae (Rubtsov et al., 1980 ; M27607). These data were then used to construct a genetic distance matrix incorporating corrections for multiple base changes

according to the method of Jukes & Cantor (1969). A phylogenetic tree was produced from this matrix using the neighbour-joining method of Saitou & Nei (1987) and bootstrap values for the tree were generated according to Felsenstein (1985). PFGE. Yeast chromosomes were prepared in agarose plugs as described by Vazquez et al. (1991) and separated in 1.4% (w/v) agarose gels using the CHEF-Mapper PFGE system (Bio-Rad) under conditions described previously (Sullivan et al., 1995). Plasmids and SAP probe DNA. DNA probes for use in hybridization experiments were generated from C. albicans SAP genes 2 , 2, 3, 5 and 7. The SAPl probe was a 1.35 kb NotI-XhoI fragment of the SAPl ORF from plasmid PAS1 (Hube et al., 1994); the SAP2 probe was a 0.65 kb BamHI-Hind111 fragment of the SAP2 ORF from plasmid pCA2 (Monod et al., 1994) ; the SAP3 probe was a 1 kb EcoRI fragment of the SAP3 ORF from plasmid pCA3 (Monod et al., 1994) ; the SAP5 probe (referred to later as SAP4-6 due to the cross-hybridization of these three highly homologous genes) was a 0.6 kb BglII fragment of the SAP5 ORF from plasmid pCA5 (Monod et al., 1994) and the SAP7 probe was a 1 kb EcoRI-HincII fragment of the SAP7 ORF from plasmid pCA7 (Monod et al., 1994). After plasmid isolation, restriction digestion and fragment separation by agarose gel electrophoresis, the desired probe fragments were purified from the agarose by the Prep-A-Gene kit (Bio-Rad) in preparation for random priming. Southern transfer and hybridization. Restriction digestion, agarose (1 %) gel electrophoresis in l x TAE buffer (40 mM Tris/acetate, 1 mM EDTA) and Southern transfer using a vacuum manifold (Pharmacia) were completed as described by Sambrook et al. (1989). Nylon membrane filters were then rinsed with 5x SSC ( l x SSC is 015 M sodium chloride, 0.015 M sodium citrate) and prehybridized at 42 "C for at least 5 h in 40% (v/v) formamide, 5x SSC, 5x Denhardt's reagent (1 x Denhardt's is 0.2 g 1-1 each of Ficoll, polyvinylpyrrolidone and BSA), 0.5% (w/v) SDS, 0.01 M EDTA, 0.025 M Na,PO, and 100 mg denatured sheared herring sperm DNA ml-'. Denatured radiolabelled probes were added and incubated with fresh prehybridization solution for 12-18 h. Membranes were then washed twice for 15 min with each of 2x SSCP ( l x SSCP is 015 M NaCl, 15 mM sodium citrate, 5 mM Na,HPO,, 7 mM NaH,PO,), 0.1 '/o SDS at 42 "C (very low stringency), 0 . 5 ~ SSCP, 0.1% SDS at 55 "C (low stringency) and 0 . 2 ~ SSCP, 0.1 '/o SDS at 60 "C (high stringency). PFGE gels were Southern blotted as described by Lee et al. (1991) and hybridized to the probes under the high stringency conditions described above. In vivo virulence tests. Candida cells grown overnight on Sabouraud's dextrose agar were washed off in saline and inoculated at two inoculum densities into the tail vein of CR

83 1

G. D. G I L F I L L A N a n d OTHERS

CD1 female mice (mean weight 22g) and the animals were monitored for 8 d.

RESULTS

Comparison of the nucleotide sequences of the small rRNA genes of C. dubhiensis and other Candida species

Comparison of the nucleotide sequence of the V3 region of the large ribosomal subunit genes from a variety of species was previously used as evidence for the des- ignation of C. dubtiniensis as a taxonomically distinct species (Sullivan et al., 1995). Previously, comparisons of the nucleotide sequences of the entire small rRNA genes have been used most extensively for phylogenetic analyses with a wide range of microbial species. To confirm this phylogenetic position of C. dubliniensis we cloned the entire small rRNA gene from the type strain of C. dubliniensis (CD36), determined its complete nucleotide sequence and compared it to the corresponding sequences previously obtained from other representative yeast species.

PCR amplification of the C. dubliniensis small rRNA gene with primers previously used in the amplification of small rRNA sequences from other Candida species (Medlin et al., 1988) yielded a single product of approximately 1800 bp. This product was then cloned into T-tailed pGEM3Zf and sequenced using standard protocols. The amplified sequence was found to be 1791 bp. When this sequence was compared with the corresponding sequence in C. albicans (accession no. X53497) it was found to differ at 25 bp (corresponding to 1.4 % sequence divergence). An evolutionary distance matrix (Table 2) incorporating corrections for multiple base changes was generated by the method of Jukes & Cantor (1969) using the small rRNA genes of C. dubliniensis (X99399), C. albicans (X53497), C. tropicalis (M55527), C . glabrata (X51831), C. lusitaniae (M55526), C. krusei (M5.5528) and Saccharomyces cerevisiae (M27607). It is interesting to note that the C. tropicalis and C. albicans sequences differed by approximately 1.4%; this difference is similar to the sequence difference found between C. albicans and C. dubliniensis (Table 2). An evolutionary tree was gen-

C. albicans

C. dubliniensis C. tropicalis

C. glabrata

S. cerevisiae

I C. krusei I 1 % , Fig- 1. An unrooted phylogenetic neighbour-joining tree generated from the alignment of the sequences encoding the small rRNA genes from C. dubliniensis CD36 and other Candida species listed in Table 2. The scale bar represents a 1 % difference in nucleotide sequence. Numbers at each node were generated by bootstrap analysis and represent the percentage of times the arrangement occurred in 1000 randomly generated trees.

erated from these data using the neighbour-joining method of Saitou & Nei (1987) (Fig. 1). It is evident from this tree, and from the bootstrap values also obtained, that C. dubliniensis is phylogenetically distinct from the other Candida species, including C. albicans.

Susceptibility to antif ungals

The four C. dubliniensis isolates and C. albicans 3153 were assayed for their susceptibility to the antifungal drugs fluconazole, ketoconazole, itraconazole and amphotericin B using the broth dilution method (Pfaller et al., 1990). All strains were susceptible to the antifungals tested, with only minor variations in susceptibility between strains (Table 3).

Hyphal formation

The relative ability of reference C. albicans and C. dubliniensis isolates to produce true non-constricted hyphae under a variety of induction conditions was assayed. Hyphal growth was induced using serum (Gow & Gooday, 1982), N-acetyl-D-glucosamine (GlcNAc ; Mattia et al., 1982) and pH/temperature shift (Buff0 et al., 1984). All results are presented as the formation of

Table 2. Genetic distance matrix based on comparison of sequences of the small rRNA gene

C . du. C . al. C . tr. C . g l . S . ce. C . Zu. C . kr. 1 C . dubliniensis -

C . albicans 1.4 - C. tropicalis 1.9 1.4 - C. glabrata 5.4 4.8 4.5 - S. cerevisiae 5.5 4-6 4.4 2.2 - C. lusitaniae 7.6 7.3 7.0 7.8 8.2 -

C . krusei 8.9 8.4 7.8 8-0 8.5 10.1

832


Table 3. MIC of selected antifungals

1 Strain MIC (pg ml-')

Fluconazole Itraconazole Ketoconazole Amphotericin B

C. albicans 3153 0.39 0.003 0.003 0.098 C. dubliniensis CD36 0.39 0.003 0.003 0.049 C. dubliniensis CD41 0.39 0.006 0.003 0.012 C. dubliniensis CD57 0.78 0-003 0.003 0.024 C. dubhiensis CD43 0 19 0.003 0003 0012

1 2 3 4 5 6 1 2 3 4 5 Time (h)

Fig. 2. Hyphal formation by C. albicans strains 3153 (m) and SC5314 (0) and C. dubhiensis strains CD36 (0) and CD57 (A) under different induction conditions. (a) Newborn-calf serum (lo%), (b) pH/temperature shift and (c) GlcNAc (4 mM). Similar results to those obtained with C. dubhiensis isolates CD36 were also obtained with C. dubhiensis isolates CD41 and CD43. A value of 100% represents germ tube outgrowth from all cells. Error bars indicate SD (n = 3).

hyphae over a time course after exposure to hypha- inducing conditions and are shown in Fig. 2. C. dubliniensis isolates CD41 and CD43 produced results similar to that shown for CD36 (data not shown).

The four C. dubliniensis isolates tested produced hyphae less rapidly than control C. albicans isolates under the in vitro induction conditions described above. This effect was seen under all conditions tested, with the exception of hyphal induction of CD57 in serum medium, but was most pronounced when hyphae were induced in medium containing GlcNAc, where all C. dubliniensis isolates failed to produce hyphae. In GlcNAc induction medium, however, CD57 displayed a low level of hyphal induction. In this case hyphae were formed during the 24 h period of starvation in water - part of the published protocol to generate inocula for germ tube induction experiments. The inability of C. dubliniensis to form hyphae in GlcNAc was not due to an incapacity to utilize GlcNAc as a carbon source, as N-acetylglucosamine assimilation has been demonstrated for this species (Sullivan et al., 1995).

Adherence to human buccal epithelial cells

The adherence of the four C. dubliniensis isolates was compared to that of the two control C. albicans strains (Fig. 3) using the adherence of yeast form Candida cells to human BECs as described by Douglas et al. (1981). The Candida strains were grown in both glucose- and galactose-based media. Adherence was studied in both control strains of C. albicans and with C. dubliniensis CD36 and CD57, and the experiments repeated on 3 consecutive days. Adherence values were similar on each of the three days. C. albicans 3153 showed adherence values similar to that reported previously (Douglas et al., 1981) and similar to that of the C. albicans clinical isolate SC.5314, the parental strain of genetically marked strains used in most Candida gene disruption experiments (Fonzi & Irwin, 1993). Adherence of all these strains was approximately five times higher in cells grown in galactose medium over those grown in the glucose medium. Adherence of C. dubliniensis CD57 was similar to that of the C. albicans isolates. C. dubliniensis CD36 showed no significant difference in

833


T

T I

" GI ucose G a I actose

, . . . . . . . . . . . . . . . . . . . , . . , . . . . . . . , . , . . . . . . . . . . . . . . . . . . . . . . , . . . . . . . . . . . . . . . . . . . . , . , . . . . . . . . , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I. . . . . . . . . . . . . Fig. 3. Adherence of C. albicans strains 3153 (0) and SC5314 (m) and C. dubliniensis strains CD36 (El) and CD57 (H) to human BEC. Data shown are mean values from nine replicates, three replicates per day on three consecutive days. Results were analysed by one-way analysis of variance and calculation of minimum significant differences using Student's t distribution. Error bars indicate SD (n=9).

adherence compared to the C. albicans isolates when grown in galactose medium, but was significantly more adherent ( P < 0.01) when grown in glucose medium. Adherence of two other C. dubliniensis oral isolates, CD41 and CD43, to BECs was examined for 1 d only. Both exhibited increased adherence in the presence of glucose equivalent to that determined for CD36.

Homologues to C. albicans SAP genes

SAP activity of strains now identified as C. dubliniensis has been reported previously as significantly greater than that of reference C. albicans isolates (McCullough et al., 1995). We therefore chose to examine C. dubliniensis for the presence of homologues of the known C. albicans SAP genes, using probes derived from seven members of the SAP multigene family. The ORFs of C. albicans SAP4, 5 and 6 are 90% identical (Monod et al., 1994) and cannot readily be distinguished by hybridization to probes derived from any of their sequences. A SAP5 sequence was therefore chosen to be used as probe for all three of the above genes and is taken to indicate the presence of one or more members of the SAP4-6 gene family.

Four C. dubliniensis isolates were compared with the control C. albicans strain 3153 using PFGE with chromosome blotting and probe hybridization (Fig. 4a). Each probe hybridized to a single chromosome. The same chromosome was recognized in all strains of the same species.

In addition to the PFGE results, DNA from C. albicans 3153 and each of the four C. dubliniensis strains was digested with EcoRI and BglII, separated by agarose gel electrophoresis and examined by Southern analysis. Fig. 4(b) shows the results obtained when such a filter was probed with SAP2. In all cases the restriction fragments for C. dubliniensis strains differed to those from C. albicans 3153 (data not shown). C. dubliniensis strains CD36, CD41 and CD57 gave identical restriction frag-

Figrn 4. Southern hybridization analysis of C. dubliniensis isolates using the C. albicans genes SAPl, 2, 3, 5 and 7 as probes. (a) Electrophoretic karyotypes of C. albicans 3153 (lane 1) and C. dubliniensis CD36, 57,43 and 41 (lanes 2-5, respectively). The chromosome-sized bands to which the C. albicans SAP2 and 7 genes hybridized are indicated with open circles, the bands to which the C. albicans SAP3 gene hybridized are indicated with triangles and the bands to which the C. albicans SAP7 and 5 genes hybridized are indicated with squares. Chromosomes from 5. cerevisiae (Bio-Rad), used as molecular size markers, are on the left of the gel. (b) Southern analysis. EcoRI- and Bglll-digested genomic DNA was probed with C. albicans SAP2. Lanes: 1 and 2, C. albicans 31 53 DNA; 3 and 4, C. dubliniensis CD36; 5 and 6, CD57; 7 and 8, CD43; 9 and 10, CD41. In all cases the left-hand lane contains the EcoRl digest and the right-hand lane the BgllI digest. Bacteriophage 2 DNA digested with Hindlll was used as size reference markers (lane M).

834


Table 4. Hybridization of C. albicans SAP gene probes to restriction-endonuclease- digested C. dubliniensis genomic DNA

SAP gene probe Digest C. dubliniensis bands (kb) hybridizing to probe"

CD36, CD41 & CD57 CD43

1

2

3

4-6

7

EcoRI BglII

EcoRI BglII

EcoRI BglII

EcoRI BglII

EcoRI BglII

4.3, (3.8) (105), (41), (0.5)

(3.8), 3.3 (951, 1.7 4-3, (3-8) 9.6, (2.1) 3.8, (3.3) (W, (2.2)

1.5 7 0

'' Numbers in parentheses indicate weak hybridization. Identical hybridization patterns were seen between CD36, CD41 and CD57. CD43 produced a different pattern for all but SAP4-6.

Table 5. In vivo virulence in mouse systemic candidosis model

Mice used were CR CD1 SPF females injected into the tail vein with the doses stated and followed for 8 d post-infection. For the inoculum size of 2 x lo6, there was no statistical difference in the survival times of C. albicans 3153 and C. dubliniensis CD57. The other C. dubliniensis strains were significantly less virulent than C. albicans (P<O.l). For the inoculum size of 1 x lo7, CD41 was significantly less virulent than all of the other strains (P<O.O5), but there was no statistically significant difference between any of the other strains.

Strain ~

Inoculum size Mean survival No. of survivors at day 8 (cells per mouse) time (d)

C. albicans 3153

C. dubliniensis CD36




1 x 107

1 x 107

1 x 107

1 x 107

1 x 107

2 x lo6

2 x lo6

2 x lo6

2 x lo6

2 x lo6

1.66 475 0.5

> 8-25 > 6.0 > 8.5

0-5 > 9.0

3.0 > 6.25

0/3*

014 014

113

013

3/4

314 0/4 3/4

2/4 ~~ ~

"All mice died by day one.

ment sizes (Table 4). Strain CD43 gave different restriction patterns to the other three strains for S A P l , SAP2 and SAP3, but identical hybridizing fragments for SAP4-6 and SAP7 (Table 4). The C. dubliniensis genome therefore has homologues of all C. albicans SAP genes that were examined and polymorphisms may exist at certain SAP loci in some C. dubliniensis strains.

ln vivo virulence tests

The relative virulence of C. albicans and C. dubliniensis was assayed using a mouse model of systemic candidosis (Table 5 ) . At the higher inoculation density of lo' cells per mouse, there was a wide variation in mean survival

times amongst the mice infected with C. dubliniensis. At a lower inoculum density of 2 x lo6 cells per mouse, mice survival times were 1.5-2 times greater for infections by C. dubliniensis strains compared to those caused by C. albicans 3153. Therefore, in this model infection C. dubliniensis had lower in vivo virulence than C. albicans.

DISCUSSION

Based on preliminary phylogenetic evidence, C. dubliniensis is the species most closely related to C. albicans. It is therefore not surprising that the two species share many phenotypic traits, such as the ability

835


to form true germ tubes, adhere to epithelial surfaces, secrete a range of aspartic proteinases and form chlamydospores. C. albicans is the Candida species most frequently recovered from patients presenting with candidosis ; consequently it is generally accepted that it possesses significantly greater pathogenic potential than other members of the genus. In this study, we have further established the phylogenetic relationship between C. dubliniensis and other Candida species, including C. albicans, by comparing the nucleotide sequences of their small rRNA genes. In addition, we compared C. albicans and C. dubliniensis with respect to a number of parameters which have been postulated to play a role in the virulence of C . albicans.

The small rRNA gene of the C. dubliniensis type strain, CD36, was amplified using PCR primers specific for conserved sequences at the 5’ and 3’ ends of eukaryotic small rRNA genes. When the sequence of this 1791 bp product was compared to that previously described for the C. albicans small rRNA gene, the two sequences were found to differ by 1-4 % , the same level of difference found between C. albicans and C. tropicalis. In addition, when an evolutionary tree was generated by comparing the C. dubliniensis small rRNA gene sequence with the corresponding sequences from a variety of other Candida species it was evident that C. dubliniensis formed a separate branch, thus confirming earlier data generated from comparisons of the V3 region of the large rRNA subunit genes (Sullivan et al., 1995; Kurtzman & Robnett, 1997).

One of the putative virulence factors of C. albicans is its ability to form true unconstricted germ tubes. The clinical implications of this dimorphic behaviour are not clear as both yeast and hyphal forms have been observed in infected tissue. However, the hyphal form of C. albicans is thought to be more adherent than the yeast form and hyphae are more likely to be able to penetrate tissue (Cutler, 1991). The hyphal induction experiments described in this study suggest that the kinetics of hyphal formation were slower in C. dubliniensis compared to C. albicans. Nevertheless, C. dubliniensis formed unconstricted germ tubes that would pass the ‘germ tube test’ used to differentiate C. albicans from other non-C. albicans species. None of the strains of C. dubliniensis tested were able to form hyphae in GlcNAc. This raises the possibility that C. dubliniensis may carry a mutation in the pathway controlling hyphal induction in response to GlcNAc. However, the inability to form hyphae on exposure to GlcNAc is also a feature of some C. albicans isolates (Mattia et al., 1982; Gow et al., 1994). This suggests that C. dubliniensis may be as potent as C. albicans at penetrating the epithelial layers of host tissues and may relate to the primary association of C. dubliniensis with superficial infections of the oral mucosa (Coleman et al., 1997a).

The finding that oral C . dubliniensis isolates adhered to BECs to a greater extent than reference C. albicans isolates when grown in the presence of glucose may also be relevant to the observation that C. dubliniensis is

found almost exclusively in the oral cavity (Coleman et al., 1997a). The increased adherence of what were originally described as atypical C. albicans isolates, now known to be C. dubliniensis (Sullivan et al., 1995,1997), has been described before (McCullough et al., 1995). Interestingly, the vaginal C. dubliniensis isolate CD57 showed no signs of increased adherence when grown in glucose-containing medium. It has not yet been established whether this increased adherence in relation to growth in glucose is a feature common to oral-adapted strains of Candida, or is peculiar to C. dubliniensis. Diabetes, antibiotic treatment and steroid therapy increase a patients level of salivary glucose - and all three conditions predispose a patient to candidosis (McCourtie & Douglas, 1981). It is possible that the increased adhesion of CD36 and other C . dubliniensis oral isolates described by McCullough et al. (1995) compared with C. albicans may be due to the elevated expression of receptors, such as the glucose-responsive C. albicuns analogue to the mammalian neutrophil integrin protein, CDllb/CDl8 (Gustafson et al., 1991). Expression of the C. albicans integrin analogue is elevated in the presence of 20 mM glucose and can be reversed if cells are treated with mAbs to the integrin analogue (Gustafson et al., 1991). However, the con- centration of sugars employed in this standard adhesion assay protocol are likely greater than those found in vivo and so the resulting data must be interpreted cautiously. SAPS are believed to be involved in the attachment and tissue invasion of Candida cells during infection (Borg & Ruchel, 1988) and correlations between proteolytic activity and virulence of Candida strains have been reported (Ruchel, 1992). T o date at least seven SAP genes (SAPI-7) have been cloned and sequenced in C. albicans and the presence of homologous multigene families in other Candida species has also been suggested (Monod et al., 1994). The seven SAP genes of C. albicans that were examined here are expressed differentially under a variety of conditions influenced by the medium composition, growth phase, morphological form and strain of the organism (Hube et al., 1994). Isolates originally identified as atypical C. albicans, later confirmed as C. dubliniensis strains, were demonstrated to be highly proteolytic, producing greater amounts of proteinase than reference C . albicans strains (McCullough et al., 1995). In this study we showed that all the C. dubliniensis isolates examined possessed homologues to each of the seven C. ulbicans SAP genes tested. Whether the products of these genes have the same activities and pathological roles or are produced under the same environmental conditions as their C. albicans homologues has yet to be determined. The majority of C. dubliniensis isolates have been recovered from HIV-positive patients with oral candidosis and the oral cavities of healthy individuals, suggesting that C. dubliniensis may be adapted specifically to the oral cavity. However, in the absence of an established oral candidosis animal model, we used a systemic mouse model to investigate the relative virulence of C. dubliniensis and C. albicans strains. Under

836


these experimental conditions, the four C. dubliniensis isolates indicated that the new species may be less virulent than C. albicans.

Moran et al. (1997) recovered clinical C. dubliniensis isolates resistant to fluconazole and found that resistance could rapidly be developed on culture in vitro. Never- theless, the majority of C. dubliniensis isolates are susceptible to fluconazole, as were the four isolates examined here (MIC range 0-19-0-78 pg ml-l). Like those isolates examined previously, these isolates were also susceptible to itraconazole, ketoconazole and amphotericin B. Clearly, C. dubliniensis and C. albicans are closely related and share many properties. The decreased ability of C. dubliniensis to form hyphae and its increased levels of proteinase production may help to provide an explanation as to why C. dubliniensis is almost exclusively associated with oral candidosis. The precise roles of these and other potential virulence factors, such as the acquisition of antifungal resistance have yet to be addressed.

ACKNOWLEDGEMENTS

Work in the laboratory of Professor N. A. R. Gow was supported by grant nos. 044275/2/95Z and 039643/2/93/ JRSfRF from The Wellcome Trust. A personal Royal Society Leverhulme Trust Senior Fellowship to N. A. R. G. is acknowledged. Work in the laboratory of Dr D. C. Coleman was supported by grant no. 047204 from The Wellcome Trust and grant no. 41/96 from the Irish Health Research Board. We thank Dr B. Hube for provision of plasmids carrying the SAP genes and Bernie Lewis for help with the animal studies.

REFERENCES

Anthony, R. M., Midgley, J., Sweet, 5. P. & Howell, S. A. (1995). Multiple strains of Candida albicans in the oral cavity of HIV positive and HIV negative patients. Micro6 EcoZ Health Dis 8,

Barns, 5. M., Lane, D. J., Sogin, M. L., Bibeau, C. & Weisburg, W. G. (1 991). Evolutionary relationships among pathogenic Candida species and relatives. J Bacteriol 173, 2250-2255. Boerlin, P., Boerlin-Petzold, F., Durussel, C., Addo, M., Pagani, J.-L., Chave, J.-P. & Bille, J. (1995). Cluster of atypical Candida albicans isolates in a group of human immunodeficiency virus- positive drug users. ] Clin Microbiol33, 1129-1135. Borg, M. & Ruchel, R. (1988). Expression of extracellular acid proteinase by proteolytic Candida spp. during experimental infection of oral mucosa. Infect Imrnun 56, 626-631. Boucher, H., Mercure, S., Montplaisir, 5. & Lemay, G. (1996). A novel group I intron in Candida dubliniensis is homologous to a Candida albicans intron. Gene 180, 189-196. Buffo, J., Herman, M. A. & 5011, D. R. (1984). A characterisation of pH-regulated dimorphism in Candida albicans. Mycopathologia

Chavanet, P., Lopez, J., Grappin, M., Bonnin, A., Duong, M., Waldner, A., Buisson, M., Camerlynck, P. & Portier, H. (1994). Cross-sectional study of the susceptibility of Candida isolates to antifungal drugs and in vitro-in vivo correlation in HIV-infected patients. AIDS 8, 945-950.

23-30.

85, 21-30.

Coleman, D. C., Sullivan, D. J., Bennett, D. E., Moran, G. P., Barry, H. 1. & Shanley, D. B. (1997a). Candidiasis: the emergence of a novel species, Candida dubliniensis. AIDS 11, 557-567. Coleman, D., Sullivan, D., Haynes, K., Henman, M., Shanley, D., Bennett, D., Moran, G., McCreary, C., O’Neill, L. & Harrington, B. (1 997b). Molecular and phenotypic analysis of Candida dubliniensis : a recently identified species linked with oral candidosis in HIV-infected and AIDS patients. Oral Dis 3, suppl.

Cutler, 1. E. (1991). Putative virulence factors of Candida albicans. Annu Rev Microbiol45, 187-218. Douglas, J. L., Houston, J. G. & McCourtie, J. (1981). Adherence of Calzdida albicans to human buccal epithelial cells after growth on different carbon sources. FEMS Microbiol Lett 12, 241-243. Felsenstein, J. (1985). Confidence limits on phylogenies : an approach using the bootstrap. Evolution 39, 783-791. Fonzi, W. A. & Irwin, M. Y. (1993). Isogenic strain construction and gene mapping in Candida albicans. Genetics 134, 717-728. Gillum, A. M., Tsay, E. Y. H. & Kirsch, D. R. (1984). Isolation of the Candida albicans gene for orotidine-5’-phosphate decarboxylase by complementation of S. cerevisiae ura3 and E. coli pyrF mutations. Mol Gen Genet 198, 179-182. COW, N. A. R. & Gooday, G. W. (1982). Vacuolation, branch production and linear growth of germ tubes of Candida albicans. ] Gen Microbiol 128, 2195-2198. Cow, N. A. R., Robbins, P. W., Lester, J. W., Brown, A. J. P., Fonzi, W. A., Chapman, T. & Kinsman, 0. 5. (1994). A hyphal-specific chitin synthase gene (CHS2) is not essential for growth, dimorphism, or virulence of Candida albicans. Proc Nut1 Acad Sci USA 91, 6216-6220. Gustafson, K. S., Vercellotti, G. M., Bendel, C. M. & Hostetter, M. K. (1991). Molecular mimicry in Candida albicans. ] Clin Invest 87, 1896-1902. Hazen, K. C. (1995). New and emerging yeast pathogens. Clin Microbiol Rev 8, 462-478. Hendriks, L., Goris, A., Neefs, J., Van de Peer, Y., Hennerbert, G. & De Wachter, R. (1989). The nucleotide sequence of the small ribosomal subunit RNA of the yeast Candida albicans and the evolutionary position of the fungi among the eukaryotes. Syst Appl Microbiol 12, 223-229. Hendriks, L., Goris, A., Van de Peer, Y., Neefs, I.-M., Vancanneyt, M., Kersters, K., Hennebert, G. L. & De Wachter, R. (1991). Phylogenetic analysis of five medically important Candida species as deduced on the basis of small ribosomal subunit RNA sequences, ] Gen Microbiol 137, 1223-1230. Hube, B., Monod, M., Schofield, D. A., Brown, A. 1. P. & Cow, N. A. R. (1994). Expression of seven members of the gene family encoding secretory aspartyl proteinases in Candida albicans. Mol Microbiol 14, 87-99. Jukes, T. H. & Cantor, C. R. (1969). Evolution of protein molecules. In Mammalian Protein Metabolism, vol. 3, pp. 21-132. Edited by H. N. Munro. New York: Academic Press. Kraft, R., Tardiff, J., Ktauter, K. S. & Leinwand, L. A. (1988). Using mini-prep plasmid DNA for sequencing double stranded templates with Sequenase@. BioTechniqtkes 6, 544-547. Kurtzman, C. P. & Robnett, C. J. (1997). Identification of clinically important ascomycetous yeasts based on nucleotide divergence in the 5‘ end of the large-subunit (26s) ribosomal DNA gene. J Clin Microbiol 35, 1216-1223. Lee, H., Birren, B. & Lai, E. (1991). Ultraviolet nicking of large DNA molecules from pulsed-field gels for Southern transfer and hybridisation. Anal Biochem 199, 29-34.

1 , S96-Sl01.

837

G. D. GILFILLAN and OTHERS

Lee, K. L., Buckley, H. R. & Campbell, H. R. (1975). An amino acid liquid synthetic medium for development of mycelial and yeast forms of Candida albicans. Sabouraudia 13, 148-153.

Le Guennec, R., Reynes, J., Mallie, M., Pujol, C., Janbon, F. & Bastide, J.-M. (I 995). Fluconazole- and itraconazole-resistant Candida albicans strains from AIDS patients : multilocus enzyme electrophoresis analysis and antifungal susceptibilities. J Clin Microbiol33, 2732-2737.

McCourtie, 1. & Douglas, 1. (1981). Relationship between cell surface composition of Candida albicans and adherence to acrylic after growth on different carbon sources. Znfect Zmmun 32, 1234-1241.

McCullough, M. J., Ross, B. C, Dwyer, B. D. & Reade, P. C. (1994). Genotype and phenotype of oral Candida albicans from patients infected with the human immunodeficiency virus. Microbiology

McCullough, M., Ross, B. & Reade, P. (1995). Characterization of genetically distinct subgroup of Candida albicans strains isolated from oral cavities of patients infected with human immunodeficiency virus. J Clin Microbiol 33, 696-700.

Mahrous, M., Lott, T. J., Meyer, 5. A., Sawant, A. D. & Ahearn, D. G. (1990). Electrophoretic karyotyping of typical and atypical Candida albicans. J Clin Microbiol28, 876-881.

Mattia, E., Carruba, G., Angiolella, L. & Cassone, A. (1982). Induction of germ tube formation by N-acetyl-D-glucosamine in Candida albicans : uptake of inducer and germinative response. J Bacteriol 152, 555-562.

Medlin, L., Elwood, H. J., Stickel, 5. & Sogin, M. L. (1988). The characterisation of enzymatically amplified eukaryotic 16s-like rRNA coding regions. Gene 71,491499.

Monod, M., Togni, G., Hube, B. & Sanglard, D. (1994). Multiplicity of genes encoding secreted aspartic proteinases in Candida species. Mol Microbiol 13, 357-368.

Moran, G. P., Sullivan, D. J., Henman, M. C., McCreary, C. E., Harrington, B. J., Shanley, D. B. & Coleman, D. C. (1997). Anti- fungal drug susceptibilities of oral Candida dubliniensis isolates from human immunodeficiency virus (H1V)-infected and non- HIV-infected subjects and generation of stable fluconazole- resistant derivatives in vitro. Antimicrob Agents Chemother 41, 617-623.

Odds, F. C. (1974). The effect of growth medium on filament production in Candida albicans. Sabouraudia 12, 112-119.

Odds, F. C., Schmid, J. & Soll, D. R. (1990). Epidemiology of Candida infections in AIDS. Mycoses in AIDS Patients, pp. 67-73. Edited by H. Van den Bossche, D. W. R. MacKenzie, G. Cauwenbergh, J. Van Cutsem, E. Drouhet & B. Dupont. New York: Plenum Press. Olsen, G. J. & Woese, C. R. (1993). Ribosomal RNA: a key to phylogeny. FASEB J 7, 113-123.

Pfaller, M. A. (1996). Nosocomial candidiasis : emerging species, reservoirs, and modes of transmission. Clin Znfect Dis 22 (S2),

Pfaller, M. A., Rinaldi, M. G., Galgiani, 1. N., Bartlett, M. S., Body, B. A., Espinel-lngroff, A., Fromtling, R. A., Hall, G. S., Hughes, C. E., Odds, F. C. & Sugar, A. M. (1990). Collaborative investi- gation of variables in susceptibility testing of yeasts. Anti- micro6 Agents Chemother 39, 1-8.

140, 1195-1202.

S89-S94.

Powderly, W. G. (1992). Mucosal candidiasis caused by non- albicans species of Candida in HIV-positive patients. A D S 6, 604-605.

Rose, M. D., Winston, F. & Hieter, P. (1990). Methods in Yeast Genetics - a Laboratory Course Manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory. Rubtsov, P. M., Musakhanov, M. M., Zakharyev, V. M., Krayev, A. S., Skryabin, K. G. & Bayev, A. A. (1980). The structure of the yeast ribosomal RNA genes. I. The complete nucleotide sequence of the 18s ribosomal RNA gene from Saccharomyces cerevisiae. Nucleic Acids Res 8, 5779-5794.

RUchel, R. (1992). Proteinase. In N e w Strategies in Fungal Disease, pp. 17-31. Edited by J. E. Bennet, R. J. Hay & P. K. Peterson. Edinburgh : Churchill Livingstone. Saitou, N. & Nei, M. (1987). The neighbor-joining method for reconstructing phylogenetic trees. Mol Biol Evol4, 406-425.

Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory. Schmid, J., Odds, F., Wiselka, M. J., Nicholson, K. G. & 5011, D. R. (1992). Genetic similarity and maintenance of Candida albicans strains from a group of AIDS patients, demonstrated by DNA fingerprinting. J Clin MicrobioZ30, 935-941.

Sullivan, D. & Coleman, D. (1998). Candida dubliniensis: characteristics and identification. J Clin Microbiol 36, 329-334.

Sullivan, D., Bennett, D., Henman, M., Harwood, P., Flint, S., Mulcahy, F., Shanley, D. & Coleman, D. (1993). Oligonucleotide fingerprinting of isolates of Candida species other than C . albicans and of atypical Candida species from human immunodeficiency virus-positive and AIDS patients. J Clin Microbiol31,2124-2133.

Sullivan, D. J., Westerneng, T. J., Haynes, K. A., Bennet, 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,

Sullivan, D. J., Henman, M. C., Moran, G. P., ONeill, L., Bennett, D. E., Shanley, D. B. & Coleman, D. C. (1996). Molecular genetic approaches to identification, epidemiology and taxonomy of non- albicans Candida species. J Med Microbiol44, 399408.

Sullivan, D. J., Haynes, K. A., Bille, J., Boerlin, P., Rodero, L., Lloyd, S., Henman, M. & Coleman, D. C. (1997). Widespread geographic distribution of oral Candida dubliniensis strains in human immunodeficiency virus-infected individuals. J Clin Microbiol35, 960-964.

Vasquez, 1. A., Beckley, A., Sobel, J. D. & Zervo, M. 1. (1991). Comparison of restriction enzyme analysis and pulsed-field gradient gel electrophoresis as typing systems for Candida albicans. 3 Clin Microbiol29, 962-967.

Wong, 0. C. & Clark-Walker, G. D. (1990). Sequence of the gene for the cytoplasmic ribosomal RNA small subunit from Candida (Torulopsis) glabrata. Nucleic Acids Res 18, 1888.

1507-1521.

Received 6 October 1997; revised 22 December 1997; accepted 5 January 1998.

838

candida dubliniensis: phylogeny and putative virulence factors

Documents