*For correspondence. E-mail firstname.lastname@example.org; Tel. (+33) 1 30 81 54 52; Fax (+33) 1 30 81 54 57.
Complete glycosylphosphatidylinositol anchors are required in Candida albicans for full morphogenesis, virulence and resistance to macrophages
Article first published online: 7 MAY 2002
Volume 44, Issue 3, pages 841–853, May 2002
How to Cite
Richard, M., Ibata-Ombetta, S., Dromer, F., Bordon-Pallier, F., Jouault, T. and Gaillardin, C. (2002), Complete glycosylphosphatidylinositol anchors are required in Candida albicans for full morphogenesis, virulence and resistance to macrophages. Molecular Microbiology, 44: 841–853. doi: 10.1046/j.1365-2958.2002.02926.x
- Issue published online: 7 MAY 2002
- Article first published online: 7 MAY 2002
Glycosylphosphatidylinositol (GPI)-anchored proteins are involved in cell wall integrity and cell–cell interactions. We disrupted the Candida albicans homologue of the Saccharomyces cerevisiae GPI7/LAS21 gene, which encodes a GPI anchor-modifying activity. In the mutant and on solid media, the yeast-to-hyphae transition was blocked, whereas chlamydospore formation was enhanced. However, the morphogenetic switch was normal in liquid medium. Abnormal budding patterns, cytokinesis and cell shape were observed in both liquid and solid media. The cell wall structure was also modified in the mutants, as shown by hypersensitivity to Calcofluor white. In vitro and in vivo assays revealed that the mutant interacted with its host in a modified way, resulting in reduced virulence in mice and reduced survival in the gastrointestinal environment of mice. The mitogen-activated protein (MAP) kinase pathway of macrophages was downregulated by the wild-type cells but not by the ΔCagpi7 null strains. In agreement with this abnormal behaviour, mutant cells were more sensitive to the lytic action of macrophages. Our results indicate that a functional GPI anchor is required for full hyphal formation in C. albicans, and that perturbation of the GPI biosynthesis results in hypersensitivity to host defences.
Candida albicans is a ubiquitous commensal and a major opportunistic human pathogen that causes superficial infections such as oropharyngeal candidiasis, which is a frequent complication in patients with human immunodeficiency virus (HIV) infections. In other clinical conditions, such as neutropenia, C. albicans can invade host tissues and cause severe, often fatal, disseminated infections (Odds, 1988). Owing to the associated morbidity and mortality, prophylactic or prolonged antifungal treatments have been used in patients at risk of developing candidaemia, such as patients undergoing chemotherapy or patients with recurrent superficial infections such as HIV. This has led to the emergence of isolates with decreased susceptibility to the commercially available antifungal agents, especially azoles (Fridkin and Jarvis, 1996).
In C. albicans, the dimorphic conversion between the yeast and hyphal forms is thought to be critical for pathogenesis, as mutations that block the transition to either form attenuate virulence (Braun and Johnson, 1997; Lo et al., 1997). Several studies have revealed some of the genetic components involved in the control of the morphogenetic switch (Kohler and Fink, 1996; Sharkey et al., 1999; Young et al., 2000). Unlike most laboratories that work on pathways initially described in Saccharomyces cerevisiae, we used Yarrowia lipolytica as a model for morphogenesis. This non-conventional yeast also undergoes complete yeast-to-hyphae transition like C. albicans and can be genetically manipulated (Barth and Gaillardin, 1997; Dominguez et al., 2000). We screened a series of recessive Fil− mutants that were unable to display a morphogenetic switch (Richard et al., 2001) and found a GPI7 homologue, a member of the glycosylphosphatidylinositol (GPI) gene family, which is involved in the biosynthesis of GPI anchors (Benghezal et al., 1995).
GPI anchoring is a eukaryotic mechanism for attaching proteins to the cell surface. Proteins destined to be GPI anchored have conserved features, an N-terminal signal sequence for localization to the endoplasmic reticulum (ER) and a C-terminal signal sequence for attachment of the GPI anchor. Shortly after protein synthesis in the ER, the preformed GPI anchor replaces the C-terminal transmembrane region. The core GPI anchor consists of a lipid group (which acts as a membrane anchor), myoinositol, glucosamine, several mannose groups and a phosphoethanolamine group, which ultimately connects the GPI anchor to the protein via an amide bond (Tiede et al., 1999). The number of mannose groups and the position of side-chains on the GPI anchors vary widely between species.
In mammalian cells, over 100 cell surface proteins of various sizes and functions are putative GPI-anchored proteins (Ohishi et al., 2000). The first step in GPI anchor synthesis in mammals involves four proteins, Pig-A, Pig-H, Pig-C, hGpi1 and probably Pig-P, which act as a protein complex (Watanabe et al., 1998; 2000). GPI anchoring is not essential in mammals at a cellular level, because several GPI-deficient cell lines have been established (Ohishi et al., 2000). However, an acquired GPI-anchoring deficiency in haematopoietic stem cells causes paroxysmal nocturnal haemoglobinuria (Takeda et al., 1993), a rare but serious human disease. In S. cerevisiae, DNA sequencing studies and Von Heijne algorithm studies identified 58 potential GPI-anchored proteins (Caro et al., 1997). GPI anchor synthesis is essential in S. cerevisiae because any pairwise combinations of GPI1, GPI2 (PIG-C homologue) and GPI3 (similar to PIG-A) deletions are unviable (Leidich et al., 1995). Gpi8 and Gaa1 are also essential in yeast and are thought to be catalytic components of a transamidase complex that cleave the GPI attachment signal and mediate the attachment of GPI anchor to proteins (Ohishi et al., 2000). Structural studies have shown that S. cerevisiae GPI anchors possess four mannose groups with a phosphoethanolamine linked on the first three groups (Tiede et al., 1999). Recent studies have identified genes and proteins involved in the addition of each of these phosphoethanolamine: Mcd4 adds a phosphoethanolamine on the first mannose, Gpi7 on the second and Gpi13 on the third one that links to the proteins (Benachour et al., 1999; Flury et al., 2000; Taron et al., 2000). The deletion of GPI7 in S. cerevisiae leads to Calcofluor white hypersensitivity and mating deficiency. In Y. lipolytica, a ΔYlgpi7 null mutant is also hypersensitive to Calcofluor white and has a clear defect in morphogenesis in all conditions tested (Richard et al., 2001).
Given these results in Y. lipolytica, we wondered whether a GPI7 homologue existed in C. albicans and if it was involved in morphogenesis. We report the prelimi-nary characterization of CaGPI7 in C. albicans, a GPI7 homologue. This is the first gene involved in the biosynthesis of GPI anchors to be studied in C. albicans. We present evidence suggesting that Gpi7 is required for normal morphogenesis, full virulence, survival in its host’s gastrointestinal tract and resistance to macrophage lysis.
GPI7 homologue in C. albicans
Gpi7 of S. cerevisiae participates in the synthesis of the core structure of the GPI anchors (Benachour et al., 1999; Toh-e and Oguchi, 1999). Our work on morphogenesis in Y. lipolytica (Richard et al., 2001) highlighted the potential role of Gpi7 in the morphogenesis of this yeast. This prompted us to determine whether a GPI7 homologue existed in C. albicans and whether it had a similar function. The putative CaGPI7 homologue was sequenced (see Experimental procedures). This gene encodes a predicted protein of 892 amino acids showing 42.5% identity to Gpi7 of S. cerevisiae and 44% identity to YlGpi7 of Y. lipolytica. The predicted protein possesses 8–10 transmembrane domains and four putative glycosylation sites (CaGPI7 accession no. AF348498).
Chromosomal deletion of CaGPI7 of C. albicans
To investigate the function of CaGpi7 in C. albicans further, we used the Ura-blaster technique (Fonzi and Irwin, 1993) to obtain ΔCagpi7 null mutants. We replaced a 1460 bp fragment of CaGPI7 with a hisG-CaURA3-hisG cassette that includes most of the conserved region of Gpi7 (Fig. 1A). After the first round of transformation, Ura+ transformants were selected on SC–Uri minimal medium, and several isolates were tested by polymerase chain reaction (PCR) to confirm the replacement. Three primers were used simultaneously for a multiplex PCR amplification (Fig. 1A, Table 2). These hybridized to the promoter sequence of CaGPI7 (outGPIdis), to the sequence of CaURA3 (CaURA-up) and to the deleted sequence (XhoIup) in the mutant. In the case of non-homologous recombination, PCR amplification resulted in a single 1580 bp fragment corresponding to the un-disrupted wild-type locus. If homologous recombination occurred, two fragments were expected, one corresponding to the undisrupted wild-type allele and a 2 kb fragment corresponding to the ectopic disrupted allele (data not shown). This generated two independent Ura+ transformants (MLR2 and MLR4), which were used as parental strains to generate two independent Ura− segregants using 5-fluoroorotic acid (5FOA) selection (MLR2A and MLR4A). Multiplex PCR amplification was also used to confirm the deletion event (see Experi-mental procedures), using one of the Ura+ transformants as a positive control.
|Primer||Sequence (5′ 3′)|
In a second round of transformation, the remaining CaGPI7 allele was disrupted in MLR2A and MLR4A to generate the Ura+ strains MLR2A42 and MLR4A2. These two mutants were tested by Southern blotting to confirm that both alleles had been disrupted. Genomic DNA was digested with EcoRI and NheI to yield a 5.3 kb fragment containing the wild-type allele, a 4.9 kb fragment containing the disrupted allele in which hisG remains alone, a 3.9 kb and a 2.8 kb fragment containing the disrupted allele in which the hisG-CaURA3-hisG cassette remains completely (Fig. 1B).
Two independent transformants, MLR2A42 and MLR4A2, were included in all experiments to ensure that the phenotypic traits observed resulted solely from the CaGPI7 mutation and not from unrelated mutations that may have occurred during the construction of ΔCagpi7 null strains. Moreover, a revertant strain (MLR2-16) has also been constructed by reintroduction of a wild-type copy of CaGPI7 (see Experimental procedures).
To confirm that CaGpi7 is involved in the biosynthesis of GPI anchors, we undertook a preliminary characterization of the GPI precursors produced in the different strains. Recent studies on S. cerevisiae showed that M4, which is a precursor of GPI anchors lacking a side-chain linked to the second mannose group, accumulated in the GPI7 mutant (Benachour et al., 1999). As shown in Fig. 2, ΔCagpi7 exhibited a similar pattern of precursor accumulation, but no biochemical analysis has been done in order to characterize each spot. These results suggest that, as in S. cerevisiae, the deletion of CaGPI7 modifies the synthesis of the GPI anchors and leads to precursor accumulation.
ΔCagpi7 cells exhibit marked sensibility to Calcofluor white and to high osmolarity
To confirm previous data obtained with S. cerevisiae (Benachour et al., 1999) or Y. lipolytica (Richard et al., 2001), which indicated that GPI7 mutants are hypersensitive to Calcofluor white, ΔCagpi7 mutants were grown in the presence of increasing concentrations of Calcofluor white (Fig. 3A). Identical results were obtained after 24 h at either 30°C or 37°C; ΔCagpi7 was found to be sensitive to Calcofluor white, whereas the revertant and the wild type resist, suggesting a cell wall structural defect in the mutant.
It has been reported that some S. cerevisiae mutants defective in cell wall assembly were affected in components of the HOG pathway that responds to hyperosmotic stresses and contributes to cell wall modelling and integrity by influencing the expression of genes encoding cell wall-modifying enzymes (Jiang et al., 1995; Alonso-Monge et al., 2001). ΔCagpi7 mutants were grown in the presence of increasing concentrations of NaCl and were found to be sensitive to NaCl after 24 h at 37°C (Fig. 3B).
ΔCagpi7 cells exhibit no thermosensibility and no hypersensitivity to several antifungal drugs
In order to investigate extensively the modification in-duced by the deletion of Cagpi7, we tested whether the growth of the mutant was affected at high temperature or by antifungal drugs. Dot test experiments on YPD plates at 39°C and 42°C showed no differences between the mutant, the revertant or the wild type (data not shown). Minimum inhibitory concentration (MIC) experiments were done on the two independent mutants (MLR2A42 and MLR4A2) and the wild type using four different drugs: fluconazole, itraconazole, amphotericin B and 5-fluorocytosine. We have obtained the following results for all the strains tested: MIC fluconazole 0.25 μg ml−1; MIC itraconazole 0.06 μg ml−1; MIC amphotericin B 0.5 μg ml−1; and MIC 5-fluorocytosine 0.0125 μg ml−1. These results suggest that deletion of Cagpi7 has no consequence on resistance mechanisms or on behaviour in high-temperature stress.
CaGPI7 is required for hyphal formation on solid media
To study the role of CaGPI7 in the yeast-to-hyphae transition, the ΔCagpi7 null mutant and the revertant strains were grown in media that induce the morphological switch: solid Spider and serum media at 37°C for 5 and 2 days respectively. Wild-type (SC5314) and the revertant (MLR2-16) strains developed filaments that emerged from the edge of the colonies on solid Spider and serum media. In contrast, the ΔCagpi7 null strains showed suppressed hyphal and pseudohyphal formation on Spider media after 5 days and only formed smooth colonies (Fig. 4). Additionally, when grown on agar with 5% serum, the ΔCagpi7 null mutants showed slight pseudohyphal development, whereas wild-type colonies showed clear hyphae development (Fig. 4, bottom row).
As the yeast-to-hyphae switch may be affected by the physical state of the media (solid or liquid) (Kohler and Fink, 1996; Leberer et al., 1996), we analysed the ability of the ΔCagpi7 null mutants to form hyphae in liquid media. In liquid Lee’s and 5% serum–salt base media, the ΔCagpi7 null strains developed hyphae similarly to the wild type, and no difference in the pattern and timing of germ tube formation was observed between ΔCagpi7 and SC5314 (data not shown).
However, a proportion of ΔCagpi7 cells aggregated, and cells with modified shapes were observed in both liquid and solid media. The proportion of abnormal cells differed dramatically between liquid (85%) and solid cultures (15%). Transmission electron microscopy confirmed these preliminary observations and showed that multiple and random buds were present on numerous cells, as well as peanut- or pear-shaped cells (data not shown). This suggests a separation defect and/or a modification of the cell wall integrity and that there is a relationship between functional GPI anchors and normal cytokinesis, related to physical constraints of the medium.
To determine whether CaGPI7 is required for chlamydospore formation, chlamydospore induction medium was inoculated with the following strains: SC5314, MLR2A42 and MLR2-16 (see Experimental procedures), covered with coverslips and incubated at 25°C in the dark. After 5 days, aberrant chlamydospore formation was observed in the mutant cells. Unlike in the wild type or in the revertant, in which chlamydospores were terminal cells on hyphae, in the mutant strains, chlamydospores were on highly branched pseudomycelium (Fig. 5). Moreover, the number of chlamydospores per colony increased dramatically in the ΔCagpi7 homozygous strain, and they were randomly located along the cell chains.
Virulence of ΔCagpi7 mutants
To determine whether Cagpi7 is required for virulence, we investigated the ability of ΔCagpi7 null strains to establish infection in a murine model of haematogenously disseminated candidiasis. Before animal studies, we evaluated growth rates that may also affect the virulence of the mutants. The growth rate of the ΔCagpi7 null strains or the revertant did not differ significantly from the wild type (data not shown) at 37°C in liquid yeast nitrogen base supplemented with 5 g l−1 ammonium sulphate or in liquid yeast peptone dextrose. Mice infected with 1 × 105ΔCagpi7 cells were not killed (Fig. 6A), whereas 25–35% of those infected with 1 × 106ΔCagpi7 cells were killed (Fig. 6B). However, both doses killed mice infected with wild-type SC5314, and the weaker 1 × 105 cells per mouse kills the mice infected with the revertant. This indicates that CaGPI7 is required for full pathogenesis of C. albicans.
The kidneys of mice infected with ΔCagpi7 cells contained both yeasts and hyphae (data not shown). This suggests that the reduction in virulence was not primarily caused by complete morphogenesis blocking in vivo.
Murine models of survival in the gastrointestinal tract
To determine whether the capacity of ΔCagpi7 null mutants to grow and to survive in the gastrointestinal environment was affected, we used 7-day-old outbred mice that were colonized with ΔCagpi7. Colonization was monitored as described in Experimental procedures by counting yeast colony-forming units (cfu) released in faecal pellets 5–14 days after a single feeding with 1 × 108 cells. The number of wild-type cfu in faecal pellets was four to five times higher than that obtained with the Δcagpi7 null strains (Fig. 7), indicating that a fully functional CaGpi7 is required for Candida cells to survive in the gastrointestinal environment.
Fungicidal effect of macrophage on mutant cells
Macrophages were incubated with yeast cells and, after washing, ingested yeasts were recovered from lysed macrophages. Cell viability was estimated by comparing cell counts and cfus 24 h after plating on Sabouraud’s medium at 37°C. The following results were obtained after 60 min incubation of yeasts with the macrophages, expressed as a percentage of survival after triplicate experiments: 40% (± 2.1) for SC5314, 32% (± 1.4) for MLR2A42, 41% (± 1.5) for MLR2-16 and 1% (± 2.2) for SU1. Mutant cells were more sensitive to macrophage killing, showing a significant decrease in survival compared with wild-type cells, but remained more resistant to macrophages than S. cerevisiae SU1 cells. The revertant strain exhibited resistance comparable with the wild-type strain, indicating that increased killing reflected deletion of CaGPI7. This suggests that CaGPI7 is required for the full resistance of C. albicans to lysis by macrophages.
Phagocytosis of mutant cells by macrophages does not cause downregulation of Erk1/2 phosphorylation
Recent studies (Ibata-Ombetta et al., 2001) have revealed that, in the macrophages, downregulation of Erk1/2 phosphorylation is part of the C. albicans strategy to escape from macrophage microbicidal activity (Hii et al., 1999). Indeed, extracellular signal-regulated kinases (Erk1/2) are central targets for one of the pathways that is most frequently modified, the mitogen-activated protein (MAP) kinases pathway, which has a major role in the response to pathogens (Ireton and Cossart, 1998). We compared the patterns of MAPK phosphorylation in J774 cells after incubation with SC5314, ΔCagpi7, S. cerevisiae SU1 and revertant cells to determine whether there are any differences in the phosphorylation of Erk1/2 in macrophages exposed to the different strains. Cell lysates were prepared after different incubation times with the yeasts and were immunostained with monoclonal antibody specific for phosphorylated Erk1/2. After 15 min, phosphorylation of both Erk1 and Erk2 was two to three times lower in the macrophages that had been incubated with wild-type cells than in those that had been incubated with mutant cells or SU1 cells (Fig. 8A). After 60 min, both Erk1 and Erk2 were dephosphorylated in all cases. A strong pulse of Erk1/2 activation was thus observed with ΔCagpi7 cells and with SU1 cells, whereas this response was downregulated in virulent SC5314 or revertant cells (Fig. 8B). This suggests that mutant cells do not possess or do not present the proteins involved in the modulation of the MAP kinase pathway, and thus behave like the non-pathogenic S. cerevisiae under these conditions.
The morphological switch from yeast to hyphae is one of the most important biological features that enables C. albicans to colonize, invade and survive in host tissues during infection. Although Y. lipolytica is rarely an opportunistic human pathogen (Shin et al., 2000), it displays a true hyphal transition unlike S. cerevisiae. We recently isolated six genes involved in morphogenesis in Y. lipolytica (Richard et al., 2001). These included YlGPI7, a GPI7 homologue from S. cerevisiae. As no GPI anchor synthesis genes have been described in C. albicans, we aimed to determine whether Gpi7 existed in C. albicans, whether it was involved in morphogenesis and whether other characteristics of ΔCagpi7 mutant cells were preserved. Disruption of CaGPI7 led to Calcofluor white hypersensitivity, which suggests cell wall fragility. This was also observed in Δgpi7 mutants of Y. lipolytica and S. cerevisiae. Microscopic observation of C. albicans mutant cells revealed the presence of cell aggregates indicating a defect in septation, a feature not observed in other yeast models. Some GPI-anchored proteins, such as Gas1 or its homologue in C. albicans, Phr1 (Popolo et al., 1993; Caro et al., 1997; Popolo and Vai, 1998), are involved in cell wall assembly and in cytokinesis; therefore, these results are consistent with a potential role for CaGpi7 in anchoring normal and functional GPI-anchored proteins.
Furthermore, mutant C. albicans cells had defective morphogenesis on solid inducing media but grew normally in liquid inducing media. Different phenotypes on solid and liquid media have also been observed in strains harbouring mutations in CPH1 filamentation pathway genes (Kohler and Fink, 1996; Leberer et al., 1996; Csank et al., 1998). Thus, we can hypothesize that the disruption of CaGPI7 may affect some proteins such as receptors, which act as sensors at the beginning of the CPH1 filamentation pathway, but we have no direct evidence for that yet. Moreover, chlamydospore formation was completely modified in mutant cells. This abnormal behaviour suggests that differentiation signals are not properly detected or executed in this genetic background. We hypothesize that extracellular sensors of these inducing conditions may be absent or modified at the cell surface, mislocalized in the cell wall layers or weakly linked to the cell wall. Alternatively, normal morphogenesis in liquid media may occur because physical constraints are lower in liquid media and because the sensors involved in the signal pathway are different or more expressed.
Moreover, unlike the wild type, the ΔCagpi7 strain failed to downregulate phosphorylation of Erk1/2 after phagocytosis by macrophages. Thus, in these conditions, the mutant strains behave like the non-pathogenic yeast S. cerevisiae, which is extremely sensitive to phagocytosis by macrophages (Ibata-Ombetta et al., 2001). Interestingly, in mice, we observed a clear reduction in virulence in the model of disseminated candidiasis and a lower resistance to the gastrointestinal tract environment. These observations, together with the increased sensitivity of mutant cells to phagocytosis by the macrophages’ lytic action, suggest that the disruption of CaGPI7 is responsible for the absence of some GPI-anchored proteins involved in cell–signal or cell–host interaction.
It is becoming increasingly obvious that numerous pathogens, including yeasts, bacteria or viruses, evade host defence mechanisms by disrupting important host functions. The mechanisms usually described involve modulation of several components of the host defence pathways, such as the disruption of the immune re-sponse, i.e. neutrophil migration or inflammatory media-tor production (Scherle et al., 1998; Hii et al., 1999), or interference with microbicidal activity of phagocytotic cells (Procyk et al., 1999). Several studies in prokaryotic or eukaryotic pathogens have suggested that the deregulation of the macrophage response is one mechanism used to escape host defences (Ruckdeschel et al., 1997; Hii et al., 1999; Nandan et al., 1999). Until now, no mechanism has been completely elucidated, but modulation of the phosphorylation of Erk1/2 by activation of the phosphatases or modulation of other MAP kinases is often involved.
Our study emphasizes the pleiotropic action of the disruption of CaGPI7 in a fungal pathogen, and shows that strains that are disrupted for both alleles of CaGPI7 become sensitive to lysis by macrophages and are unable to downregulate the activation of the Erk pathway. This suggests that GPI-anchored proteins participate in this regulation, although neither their nature nor their targets are known. Investigation of the role of GPI-anchored proteins and the way they are linked to the cell surface should further our understanding of host–pathogen interactions and possibly lead to the development of new antifungal drugs.
Strains and microbial techniques
The C. albicans strains used in this study are listed in Table 1. All strains were grown routinely on YPD (yeast extract– peptone–dextrose), Sabouraud dextrose (Difco) or SC–Uri medium (0.67% bacto-yeast nutrient base without amino acids, 2% glucose, 0.2% amino acids drop-out mix – uridine) at 30°C. For certain experiments, strains were grown on solid or liquid media at either 30°C or 37°C. Filamentation on solid 2% agar medium was induced using either Spider medium (1% nutrient broth, 1% mannitol, 0.2% K2HPO4) or serum medium (2% agar, 5% horse serum; Sigma) at 30°C or 37°C. Filamentation in liquid medium was induced using Lee’s medium (Lee et al., 1975) and 5% serum–salt base medium (0.45% NaCl, 0.335% yeast nitrogen base without amino acids, 2.5 mM N-acetylglucosamine) (Sonneborn et al., 2000). For chlamydospore induction, cells were grown on 17 g l−1 Bacto corn meal agar (Difco) supplemented with 0.33% Tween 80 (Sigma). Escherichia coli DH5α was grown in LB medium supplemented with 100 μg ml−1 ampicillin as required.
|SU1||Brewing yeast||S. cerevisiae wild type||Sendid et al. (1996)|
|W303-1B||S. cerevisiae||MATαade2-1 can1-100 ura3-1 leu2-3,112 trp1-1 his3-11,15||Benachour et al. (1999)|
|FBY15||W303-1B||gpi7-1||Benachour et al. (1999)|
|FBY185||W303-1B||gpi7::KanMX4||Benachour et al. (1999)|
|SC5314||Clinical isolate||C. albicans wild type||Gillum et al. (1984)|
|CaI4||SC5314||ura3::imm434/ura3::imm434||Fonzi and Irwin (1993)|
Standard molecular genetic techniques were used (Sambrook et al., 1989). Restriction enzymes and polymerases were supplied by Gibco BRL or New England Biolabs. Genomic DNA from yeast transformants was prepared as described previously (Querol et al., 1992). Southern blots were generated with EcoRI- and NheI-digested DNA separated on a 0.8% agarose gel and transferred onto nylon Hybond N+ membranes (Amersham Pharmacia Biotech). Probes were labelled using the ECL direct nucleic acid labelling and detection system (Amersham Pharmacia Biotech). A Perkin-Elmer thermal cycler 9600 was used for PCRs. Sequencing was carried out in the Applied Biosystems 373 DNA sequencer. The Genetics Computer Group (GCG) package was used for sequence analysis.
Sequence of CaGPI7
We used the sequence database from Stanford (http://candida.stanford.edu/ybc.html) and found a 1.2 kb sequence matching S. cerevisiae GPI7. With these data, we designed four primers (XhoIup, XhoIdw, RPCRgpiUP and RPCRgpiDW; see Table 2) for reverse PCR. SC5314 genomic DNA was digested with XhoI, ligated and amplified with RPCRgpiUP/XhoIdw and RPCRgpiDW/XhoIup yielding a 5.2 kb fragment and a >11 kb fragment. The sequence of part of these PCR fragments was used to complete the sequence of the whole CaGPI7 open reading frame (ORF).
Disruption of CaGPI7
To obtain homozygous ΔCagpi7 mutants, we constructed pMR7, which carries a cassette that deletes most of the CaGPI7 ORF as follows: two oligonucleotides, GPIup and GPIdw (Table 2), were used to amplify a 2.5 kb fragment of SC5314. The PCR fragment was ligated into pGEM-T easy (Promega) to form pMR6. The 4.1 kb DNA fragment containing the cassette hisG-CaURA3-hisG was extracted from pMB7 by digestion with HindIII and BglII and ligated into pMR6 opened by HindIII and BglII to yield pMR7. pMR7 was digested with NotI to release the 5.2 kb disruption cassette used to transform CaI4. Approximately 5 μg of DNA was used for each transformation via the lithium acetate method (Sanglard et al., 1996). Transformed (Ura+) cells were selected on SC–Uri medium, and spontaneous Ura− derivatives from four Ura+ independent clones were selected on 5FOA medium (0.67% bacto-yeast nitrogen base, 0.2% amino acids drop-out mix – uracil, 2% glucose, 1 mg ml−1 FOA and 50 μg ml−1 uridine). Two independent Ura+ transformants were then used to delete the second allele of CaGPI7 in a second round with the same cassette. Disruptions were confirmed by Southern blot analysis, using the 2.5 kb fragment amplified by GPIup and GPIdw as probe.
Construction of the revertant by reintroduction of a CaGPI7 wild copy
A wild copy of CaGPI7 was reintroduced in MLR2A42a (ΔCagpi7, Ura−) by co-transformation with a PCR fragment beginning 600 bp upstream of the starting codon of CaGPI7 and ending 100 bp after the stop codon and the CaGPI7 Ura-blaster cassette. We resorted to this technique because cloning of a full-length, unrearranged copy of CaGPI7 in E. coli appeared to be impossible. The integration occurred at the same locus and was verified by sequencing. The sequences confirmed that a complete wild copy of CaGPI7 had been integrated in the locus. The new strain MLR2-16 was thus heterozygous for CaGPI7 and CaURA3. The growth rate of this strain was comparable with that of the wild type.
Analysis of abnormal [2-3H]-myoinositol-labelled lipids from ΔCagpi7
Cells were labelled with [2-3H]-myoinositol ([2-3H]-Ins) as described previously (Benghezal et al., 1995). Briefly, cells were pregrown overnight in SD-based medium lacking inositol (SD–Ins) to an OD600 of 2.5. Exponentially growing cells were harvested, suspended in 250 μl of SD–Ins and labelled with 25 μCi of [2-3H]-Ins for 40 min. The cells were subsequently diluted with 750 μl of fresh SD–Ins and incubated for a further 80 min. Lipid extractions and lipid desalting were carried out as described previously (Reggiori et al., 1997). Desalted lipid extracts were analysed by thin-layer chromatography (0.2-mm-thick silica gel 60 plates; Merck) using chloroform–methanol–water (10:10:3) as solvent. For detection, plates were sprayed with EN3HANCE (Nen Life Science Products) and exposed to film for 1 week (Kodak) at –80°C.
For microscopic observations, cells were streaked out on different media, serum and Spider media, and grown at 37°C for 2 or 5 days respectively. A binocular microscope, WILD M3C (Henbrugg), with a magnification of 16–40 or a LABORLUX microscope (Leitz) with a magnification of 100–320 was used for observation.
Mouse model of haematogenously disseminated candidiasis
The C. albicans strains used in these experiments included a wild-type control (SC5314), the revertant and two independent Ura+ strains deleted for both alleles of CaGPI7 (MLR2A42 and MLR4A2). All strains were grown for 3 days on Sabouraud–dextrose agar with 500 μg ml−1 chloramphenicol at 37°C, harvested and suspended to a density of 5 × 105 and 5 × 106 cells ml−1 in physiological saline. For each concentration, 200 μl of cell suspension was injected intravenously in the lateral tail vein of each mouse, and each C. albicans strain was inoculated into 10 BALB/c mice (Charles Rivers). Survival was monitored daily until day 43 after inoculation. Surviving mice were then killed. The right kidney was removed under aseptic conditions, and a portion was placed in 2 ml of physiological saline and homogenized. An aliquot of this homogenate was used to observe the morphology of yeast cells in kidneys after infection by a previously described extraction method (Odds et al., 2000). Briefly, 1 ml of each homogenate was treated with 1 ml of KOH (60% w/v) for 2 h at 30°C, 100 r.p.m., centrifuged for 5 min at 1500 g and the supernatant removed. The pellets were washed three times in distilled water and suspended in 100 μl of physiological saline for microscopy observation of C. albicans cell morphology.
Mouse model of gastrointestinal tract colonization
We used the model designed by Cole et al. (1989) to compare the virulence of the strains. Briefly, litters of 10–15 infant Swiss mice (CD1 (ICR) BR; suckling, 3–4 days old) were obtained from Charles River. Six-day-old infant mice were separated from their dams for 3 h, inoculated and placed back with their dams 1 h after the gavage to ensure the complete ingestion of the inoculum before normal feeding. We used the same C. albicans strains in this experi-ment as in the intravenous model. The inoculum was prepared as described previously in the intravenous model but at a concentration of 2 × 109 cells ml−1. The viability of the yeasts was determined retrospectively by duplicate counts on Sabouraud–dextrose agar. The suspension (50 μl) was administered by gavage of 50 μl through the flexible (Teflon) part of a sterile intravenous catheter (Insyte 24 GA, 0.7 × 19 mm, B.D.) and a 1 ml syringe.
The degree of gut colonization was recorded by counting the cfu per dropping for each infant mouse. To obtain a suspension for counting of cfus, the first dropping obtained each day was crushed in 100 μl of sterile distilled water, and the suspension was entirely plated onto Sabouraud–dextrose agar containing 500 μg ml−1 chloramphenicol. After incubation at 35°C for 24 h, cfu were counted.
Co-culture of yeast cells with mammalian cells
The mouse macrophage-like cell line J774 (ECACC 85011428) was derived from a tumour taken from a female BALB/c mouse. Adherent J774 cells were precultured at 37°C in 5% CO2 in Dulbecco’s modified Eagle medium (DMEM; Sigma-Aldrich) supplemented with 10% heat-inactivated fetal calf serum (FCS; Valbiotech), 5 mM L-glutamine, 100 μg ml−1 streptomycin and 50 μg of penicillin. The J774 cells were gently scraped with a rubber policeman and placed into 12-well culture dishes at a concentration of 1 × 106 cells per well. After 18 h, the adherent cells were washed with culture medium and incubated with 20 yeasts per J774 cell in the same medium as that used for the preculture. Yeasts cells were prepared from an overnight culture on YPD plates and resuspended in PBS.
After incubation for 15 and 60 min, the cultures were washed with DMEM to remove unbound yeast cells and prepared for either biochemical analysis or fungicidal assays. For Erk phosphorylation measurements, after incubation, the cells were washed with 1 ml of ice-cold PBS supplemented with 1 mM Na3VO4 and 10 mM NaF (Sigma-Aldrich). The cultures were extracted with 500 μl of boiling twice-concentrated electrophoresis sample buffer (1x CEBS: 125 mM Tris-HCl, pH 6.8, 2% SDS, 5% glycerol, 1%β-mercaptoethanol and bromophenol blue). Lysates were collected and clarified by centrifugation for 10 min at 12 000 g at 4°C.
Extracted proteins were separated on 10% SDS–PAGE before blotting onto a nitrocellulose membrane (Protran; Schleicher and Schuell) for 2 h at 200 mA in a semi-dry transfer system. After staining with 0.1% Ponceau S in 5% acetic acid to confirm that equal quantities of proteins had been loaded and transferred, the membrane was blocked by incubation with TNT (10 mM Tris, 100 mM NaCl, 0.1% Tween) containing 5% bovine serum albumin (BSA) for 1 h at 20°C. Membranes were probed with the phosphospecific anti-bodies (diluted 1:1000) in TNT–1% BSA overnight at 4°C. After washing several times, the membranes were incubated for 1 h at 20°C in a 1:2000 dilution of horseradish peroxidase (HRPO)-conjugated anti-rabbit IgG in TNT–BSA. After washing, the membrane was incubated with ECL detection reagents (SuperSignal chemiluminescent substrate; Pierce) and exposed to ECL hyperfilm. MAPK phosphospecific antibodies (Erk1/2) were purchased from New England Biolabs. HRPO-conjugated anti-rabbit IgG was obtained from Zymed Laboratories.
After co-culture (see above), macrophage cells were washed in DMEM, recultured for a further 60 min and then lysed by the addition of 500 μl of sterile water. Yeast cells were harvested by centrifugation and suspended in PBS. Approximately 100 cells were then plated on Sabouraud medium and grown for 24 h at 37°C. Survival was determined by compari-son of the number of cfu.
We thank Alistair J. P. Brown for generously providing pMB7, strains SC5314 and CaI4, and Andreas Conzelmann for donating the S. cerevisiae strains, W303-1B, FBY15 and FBY185, and for giving us access to the complete protocol for [2-3H]-Ins labelling. We are grateful to Daniel Poulain for continuous discussion and critical reading of the manuscript. We also thank Andrée Lepingle and Annie Auger for help in sequence acquisition, and Luce Improvisi for precious technical help. This work was supported by EC grant QLK2 CT 2000-00795. Françoise Dromer, Thierry Jouault and Claude Gaillardin were supported by a grant from the Programme de Recherche Fondamentale en Microbiolgie et Maladies Infectieuses et Parasitaires, and Mathias Richard received a CNRS-Aventis grant.
- 2001) Hyperosmotic stress response and regulation of cell wall integrity in Saccharomyces cerevisiae share common functional aspects. Mol Microbiol 41: 717–730.DOI: 10.1046/j.1365-2958.2001.02549.x , , , , , and (
- 1997) Physiology and genetics of the dimorphic fungus Yarrowia lipolytica. FEMS Microbiol Rev 19: 219–237. , and (
- 1999) Deletion of GPI7, a yeast gene required for addition of a side chain to the glycosylphosphatidylinositol (GPI) core structure, affects GPI protein transport, remodeling, and cell wall integrity. J Biol Chem 274: 15251–15261. , , , , , et al. (
- 1995) Identification of six complementation classes involved in the biosynthesis of glycosylphosphatidylinositol anchors in Saccharomyces cerevisiae. J Cell Biol 130: 1333–1344. , , and (
- 1997) Control of filament formation in Candida albicans by the transcriptional repressor TUP1 (see comments). Science 277: 105–109. , and (
- 1997) In silico identification of glycosyl-phosphatidylinositol-anchored plasma-membrane and cell wall proteins of Saccharomyces cerevisiae. Yeast 13: 1477–1489. , , , , , and (
- 1989) Gastrointestinal and systemic candidosis in immunocompromised mice. J Med Vet Mycol 27: 363–380. , , , and (
- 1998) Roles of the Candida albicans mitogen-activated protein kinase homolog, Cek1p, in hyphal development and systemic candidiasis. Infect Immun 66: 2713–2721. , , , , , et al. (
- 2000) Yarrowia lipolytica: an organism amenable to genetic manipulation as a model for analyzing dimorphism in fungi. In Dimorphism in Pathogenic and Apathogenic Yeasts, Ernst, J.F. & Schmidt, A.(eds). Basel, Switzerland: Karger, pp. 151–172. , , and (
- 2000) YLL031c belongs to a novel family of membrane proteins involved in the transfer of ethanolaminephosphate onto the core structure of glycosylphosphatidylinositol anchors in yeast. J Biol Chem 275: 24458–24465. , , and (
- 1993) Isogenic strain construction and gene mapping in Candida albicans. Genetics 134: 717–728. , and (
- 1996) Epidemiology of nosocomial fungal infections. Clin Microbiol Rev 9: 499–511. , and (
- 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. , , and (
- 1999) Role of the extracellular signal-regulated protein kinase cascade in human neutrophil killing of Staphylococcus aureus and Candida albicans and in migration. Infect Immun 67: 1297–1302. , , , , and (
- 2001) Role of the extracellular signal-regulated protein kinase cascade in macrophage killing of Candida albicans (in process citation). J Leuk Biol 70: 149–154. , , , and (
- 1998) Interaction of invasive bacteria with host signaling pathways. Curr Opin Cell Biol 10: 276–283. , and (
- 1995) Regulation of cell wall beta-glucan assembly: PTC1 negatively affects PBS2 action in a pathway that includes modulation of EXG1 transcription. Mol Gen Genet 248: 260–269. , , , , and (
- 1996) Candida albicans strains heterozygous and homozygous for mutations in mitogen-activated protein kinase signaling components have defects in hyphal development. Proc Natl Acad Sci USA 93: 13223–13228. , and (
- 1996) Signal transduction through homologs of the Ste20p and Ste7p protein kinases can trigger hyphal formation in the pathogenic fungus Candida albicans. Proc Natl Acad Sci USA 93: 13217– 13222. , , , , , et al. (
- 1975) An amino acid liquid synthetic medium for development of mycelial and yeasts form of Candida albicans. Sabouraudia 13: 148–153. , , and (
- 1995) Temperature-sensitive yeast Gpi anchoring mutants gpi2 and gpi3 are defective in the synthesis of N-acetylglucosaminyl phosphatidylinositol. Cloning of the GPI2 gene. J Biol Chem 270: 13029–13035. , , , , , et al. (
- 1997) Nonfilamentous C. albicans mutants are avirulent. Cell 90: 939–949. , , , , , and (
- 1999) Activation of phosphotyrosine phosphatase activity attenuates mitogen- activated protein kinase signaling and inhibits c-FOS and nitric oxide synthase expression in macrophages infected with Leishmania donovani. Infect Immun 67: 4055–4063. , , and (
- 1988) Candida and Candidosis. London: Baillere Tindall. (
- 2000) Survival in experimental Candida albicans infections depends on inoculum growth conditions as well as animal host. Microbiology 146: 1881–1889. , , and (
- 2000) Gaa1p and Gpi8p are components of a glycosylphosphatidylinositol (GPI) transamidase that mediates attachment of Gpi to proteins. Mol Biol Cell 11: 1523–1533. , , , , , and (
- 1998) Defects in assembly of the extracellular matrix are responsible for altered morphogenesis of a Candida albicans phr1 mutant. J Bacteriol 180: 163–166. , and (
- 1993) Physiological analysis of mutants indicates involvement of the Saccharomyces cerevisiae GPI-anchored protein gp115 in morphogenesis and cell separation. J Bacteriol 175: 1879–1885. , , , , , et al. (
- 1999) Salmonella typhimurium and lipopolysaccharide stimulate extracellularly regulated kinase activation in macro-phages by a mechanism involving phosphatidylinositol 3-kinase and phospholipase Das novel intermediates. Infect Immun 67: 1011–1017. , , , and (
- 1992) A comparative study of different methods of yeast strain characterization. Syst Appl Microbiol 15: 439–446. , , and (
- 1997) Lipid remodeling leads to the introduction and exchange of defined ceramides on Gpi proteins in the ER and Golgi of Saccharomyces cerevisiae. EMBO J 16: 3506–3518. , , and (
- 2001) Tagging morphogenetic genes by insertional mutagenesis in the yeast Yarrowia lipolytica. J Bacteriol 183: 3098–3107. , , , , and (
- 1997) Yersinia enterocolitica promotes deactivation of macrophage mitogen-activated protein kinases extracellular signal-regulated kinase-1/2, 38, and c-Jun NH2-terminal kinase. Correlation with its inhibitory effect on tumor necrosis factor-alpha production. J Biol Chem 272: 15920–15927. , , , , , et al. (
- 1989) Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press. , , and (
- 1996) Susceptibilities of Candida albicans multidrug transporter mutants to various antifungal agents and other metabolic inhibitors. Antimicrob Agents Chemother 40: 2300–2305. , , , and (
- 1998) Inhibition of MAP kinase kinase prevents cytokine and prostaglandin E2 production in lipopolysaccharide-stimulated monocytes. J Immunol 161: 5681–5686. , , , , , et al. (
- 1996) Specific antibody response to oligomannosidic epitopes in Crohn’s disease. Clin Diagn Lab Immunol 3: 219–226. , , , , , et al. (
- 1999) HWP1 functions in the morphological development of Candida albicans downstream of EFG1, TUP1, and RBF1. J Bacteriol 181: 5273–5279. , , , , and (
- 2000) Nosocomial cluster of Candida lipolytica fungemia in pediatric patients. Eur J Clin Microbiol Infect Dis 19: 344–349. , , , , , et al. (
- 2000) Protein kinase A encoded by TPK2 regulates dimorphism of Candida albicans. Mol Microbiol 35: 386–396.DOI: 10.1046/j.1365-2958.2000.01705.x , , , , , and (
- 1993) Deficiency of the Gpi anchor caused by a somatic mutation of the PIG-A gene in paroxysmal nocturnal hemoglobinuria. Cell 73: 703–711. , , , , , et al. (
- 2000) Glycosylphosphatidylinositol biosynthesis defects in Gpi11p- and Gpi13p-deficient yeast suggest a branched pathway and implicate Gpi13p in phosphoethanolamine transfer to the third mannose. Mol Biol Cell 11: 1611– 1630. , , , and (
- 1999) Biosynthesis of glycosylphosphatidylinositols in mammals and unicellular microbes. Biol Chem 380: 503–523. , , , , and (
- 1999) Las21 participates in extracellular/cell surface phenomena in Saccharomyces cerevisiae. Genes Genet Syst 74: 241–256. , and (
- 1998) The first step of glycosylphosphatidylinositol biosynthesis is mediated by a complex of PIG-A, PIG-H, PIG-C and GPI1. EMBO J 17: 877–885. , , , , , et al. (
- 2000) Initial enzyme for glycosylphosphatidylinositol biosynthesis requires PIG-P and is regulated by DPM2. EMBO J 19: 4402–4411. , , , , , et al. (
- 2000) A STE12 homolog is required for mating but dispensable for filamentation in Candida lusitaniae (in process citation). Genetics 155: 17–29. , , and (