All animal experiments were approved by Regierungspräsidium Karlsruhe. Informed consent was obtained from patients.
Expression analysis of the Candida albicans lipase gene family during experimental infections and in patient samples1
Article first published online: 9 JAN 2006
FEMS Yeast Research
Volume 4, Issue 4-5, pages 401–408, January 2004
How to Cite
Stehr, F., Felk, A., Gácser, A., Kretschmar, M., Mähnß, B., Neuber, K., Hube, B. and Schäfer, W. (2004), Expression analysis of the Candida albicans lipase gene family during experimental infections and in patient samples. FEMS Yeast Research, 4: 401–408. doi: 10.1016/S1567-1356(03)00205-8
- Issue published online: 9 JAN 2006
- Article first published online: 9 JAN 2006
- Received 1 July 2003, Revised 12 September 2003, Accepted 18 September 2003
- Candida albicans;
- Human pathogen;
- Virulence factor;
Secreted lipases of Candida albicans are encoded by a gene family with at least 10 members (LIP1–LIP10). The expression pattern of this multigene family was investigated using reverse transcription polymerase chain reaction in experimental infections and in samples of patients suffering from oral candidosis. The findings illustrate that individual lipase genes are differentially regulated in a mouse model of systemic candidosis with some members showing sustained expression and others being transiently expressed or even silent. The lipase gene expression profile depended on the stage of infection rather than on the organ localization. This temporal regulation of lipase gene expression was also detected in an experimental model of oral candidosis. Furthermore, the expression of candidal lipase genes in human specimens is shown for the first time.
The human pathogen Candida albicans is the fourth leading cause of nosocomial bloodstream infections . Although this pathogen has been associated with a high mortality rate, C. albicans is described as a harmless commensal of the normal human microflora residing on both skin and mucosal surfaces. The shift from a non-pathogenic inhabitant into a life-threatening pathogen is triggered by predisposing host factors or by iatrogenic factors including antibiotic treatment, vascular catheters , parenteral nutrition, as well as immunosuppressants used for cancer therapy and organ transplantation. The polymorphic yeast C. albicans is not only able to develop resistance to antifungal drugs [3,4], it also has the capability of adjusting to various environments, making antifungal treatment difficult. Furthermore, many antifungal drugs can cause severe side effects after prolonged treatment. Therefore, it is necessary to identify new drug targets. The virulence factors of C. albicans are currently the focus of intense study to investigate their usefulness as potential drug targets [5,6]. This complex organism expresses a variety of putative virulence factors, which are differentially regulated and act in concert during infection. Putative drug targets include proteins necessary for adhesion, for example, the cell surface adhesion proteins encoded by the ALS gene family . Another virulence trait of C. albicans is its ability to change its morphology between a yeast and a hyphal or pseudohyphal growth form. This phenomenon is accompanied and/or caused by the expression of different subsets of genes . The transition from the yeast to the hyphal form allows the pathogen not only to invade tissue, but also to escape from immunosurveillance. Lastly, degrading hydrolases, secreted actively by C. albicans, are able to destroy different host tissues, and have been suggested as potential drug targets. The secreted aspartic proteinases are encoded by a large gene family (SAP1–10) , and form a complex network of different virulence factors due to a high complexity within the gene family. Individual members of this gene family are differentially regulated and expressed at different stages and in different clinical forms of candidosis in vivo . Surprisingly, even the two alleles of a single SAP gene, SAP2, can be differentially regulated . We have recently reported the cloning and in vitro characterization of another gene family of secreted hydrolases of C. albicans. This gene family consists of at least 10 members (LIP1–10), which encode putative lipases. This lipase gene family is differentially regulated at the mRNA level in several culture media. By applying reverse transcription polymerase chain reaction (RT-PCR), we have been able to show that some LIPs are constitutively expressed, while others are induced under different environmental conditions . Although secreted lipolytic enzymes of pathogenic microorganisms have been discussed as virulence factors, predominantly during bacterial infections , little is known about the involvement of lipases in fungal virulence. To elucidate the expression profile of the extracellular lipases of C. albicans during pathogenesis, we examined the expression of the LIP gene family during different experimental infections and in specimens of patients with oral candidosis.
2Materials and methods
2.1Yeast strains and culture conditions
The C. albicans clinical isolate SC5314 (wild-type)  and clinical isolates from patients suffering from oral candidosis were grown in YPG medium containing 1% (w/v) yeast extract, 1% (w/v) Bacto peptone, and 2% (w/v) glucose. Cultures were grown at 37°C in an orbital shaker at 180 rpm. The wild-type strain was applied in different infection models. All infection models were performed with cells which were harvested from the second subculture.
2.2Mouse infection and tissue preparation
BALB/c mice were infected i.p. with 1×108 SC5314 cells, as a model for systemic C. albicans infection . Cells were grown in Sabouraud broth medium and washed three times with phosphate-buffered saline (PBS) before inoculation. Four hours and 72 h after infection mice were killed by cervical dislocation. Kidneys and liver were aseptically removed and immediately shock-frozen in liquid nitrogen. The organs were stored at −70°C until RNA isolation.
2.3Experimental infection and histology of reconstituted human oral epithelium model (RHE)
Human epithelium without stratum corneum for the in vitro model of oral candidosis  was supplied by Skinethic Laboratory (Nice, France). The culture was grown without antibiotics and antimycotics. Pre-cultivated C. albicans cells, which were grown at 25°C, were washed and resuspended in PBS. The RHE was inoculated with 2×106 wild-type cells in 50 μl PBS. Inoculated cultures were then incubated for 12, 36 or 48 h in an atmosphere of 5% CO2 at 37°C. As a control, one culture was incubated in 50 μl PBS alone. After the respective time of incubation, samples were divided. One half was shock-frozen and stored at −70°C until use for RNA isolation, while the remaining half was fixed in 10% formalin and embedded in paraffin. Four-micrometer vertical sections were stained with hematoxylin-eosin and viewed at 400× magnification (done by Skinethic Laboratory). Triplicate infection experiments were performed.
2.4Clinical sample collection
For expression studies, clinical specimens were obtained from patients suffering from oral candidosis. These patients were not treated with antimycotic drugs at the time of sampling. The patients were not tested for HIV infection. One specimen was collected from the saliva and was directly transferred into 500 μl of RNAPure (PeqLab, Erlangen, Germany) and stored in liquid nitrogen until RNA isolation for expression studies. A second sample from the same patient was taken to confirm that C. albicans was the causative pathogen. This specimen was streaked on a Sabouraud glucose plate containing 4% (w/v) glucose, 1% (w/v) peptone, and 1.5% (w/v) agar. To identify C. albicans as the causative agent, colonies were transferred onto rice agar plates (2% (w/v) rice extract, 2% (w/v) agar) and checked for chlamydospore formation. To differentiate C. albicans from Candida dubliniensis we isolated genomic DNA from chlamydospore-positive colonies and used the DNA as template in a published PCR method .
In addition to experimental infections, we investigated the LIP expression profile in patient samples. To confirm that all LIP primers (Table 1) matched the corresponding LIP genes and their transcripts from each isolated clinical strain and to rule out any polymorphisms due to strain differences, genomic DNA from these strains was isolated according to the method described in the Breeden lab (http://www.fhcrc.org/labs/breeden/Methods/genomic_DNAprep.html). Briefly, a 5-ml overnight culture from the clinical isolate was pelleted and washed. Then the cells were vortexed with glass beads in a lysis buffer for 2 min. The supernatant was recovered and 275 μl of 7 M ammonium acetate (pH 7) was added. After a 5-min incubation at 65°C, the sample was left on ice for another 5 min. Then 500 μl chloroform was added and the sample was centrifuged for 2 min at 21 000×g. The aqueous phase containing the DNA was precipitated with 1 ml isopropanol. The pellet was washed with 70% (v/v) ethanol, air-dried, and resuspended in 50 μl H2O containing 10 μg ml−1 RNase A (Roche, Basel, Switzerland). The quality and quantity of the isolated DNA was determined on a 1% TBE gel .
|Primer||Sequence||Annealing temperature (°C)||RT-PCR product size (bp)|
2.6RNA isolation and RT-PCR
RT-PCRs were performed to generate reproducible transcription profiles of the 10 LIP genes during different infections. For total RNA isolation, 500 μl RNAPure and 300 μl of acid-washed glass beads were added to the infected tissue and vortexed for 20 min. After the recovery of the liquid phase, the samples were centrifuged at 21 000×g for 5 min at 4°C to remove cell debris; 250 μl of chloroform was added to the supernatant. After 15 s of vortexing and a 10-min incubation at room temperature, the samples were centrifuged again at 21 000×g for 10 min at 4°C. Total RNA was precipitated from the aqueous fraction with isopropanol overnight. The samples were centrifuged at 21 000×g for 30 min at 4°C, the RNA pellets were washed with 70% ethanol and dissolved in diethylpyrocarbonate-treated water. Total RNA concentrations were quantified using a spectrophotometer. Equal amounts of RNA were used for RT-PCR.
Approximately 1 μg or 10 μg (when working with organs) of total RNA from each sample were treated with DNase I (Invitrogen, Paisley, UK) according to the manufacturer's instructions. The digest was incubated overnight to ensure that all contaminating genomic DNA was eliminated. Then cDNA was synthesized with Superscript II reverse transcriptase and oligo(dT)-primers (Invitrogen) following the manufacturer's instructions.
2.7EFB1 and LIP PCR
For an internal mRNA control and for detection of genomic DNA, we used primers specific for the EFB1 gene of C. albicans, which contains an intron of 365 bp . Absence of genomic DNA was verified by a single intronless 554-bp-long PCR product of EFB1. The EFB1 PCR cycling was performed as follows: an initial cycle with 3 min denaturation at 95°C, 3 min of annealing at 55°C, and 3 min of extension at 72°C, followed by 35 cycles of 30 s at 95°C, 30 s at 55°C, and 2 min at 72°C. To confirm that similar concentrations of cDNA were achieved, signals of EFB1 PCR were compared.
For LIP PCRs, one PCR reaction was used for all 10 LIP primer sets, which differed only in the annealing temperature. This temperature has been optimized for each primer set using a gradient PCR cycler (Table 1). The samples were subjected to 30 cycles of denaturation for 30 s at 95°C, annealing for 30 s at 55–63°C, and extension for 1 min at 72°C, followed by 72°C for 10 min.
3.1Temporal expression of the lipase gene family during systemic liver infection in mice
BALB/c mice were infected i.p. with the clinical C. albicans wild-type strain SC5314. To investigate the temporal change in lipase gene expression, mice were killed either at an early stage (4 h) or at a later stage of infection (3 days). The liver was removed immediately and expression of the 10 lipase genes in the infecting Candida cells was analyzed using RT-PCR (Fig. 1). By using primers that span part of the EFB1 gene including the intron, it was shown that the RT-PCR was successful. There was no genomic DNA contamination and according to the signals of the amplification products equal amounts of RNA were used for RT-PCR (data not shown). Four hours after infecting mice i.p. fungal wild-type cells adhered to and started to invade the liver . In later phases of the infection (72 h), fungi became encapsulated by inflammatory cells and granulomas and microabscesses were formed. After reaching the bloodstream Candida cells were systemically distributed and infection of the kidneys followed. During the early stage of infection, LIP4–9 were expressed in more than 50% of all tested organs with transcripts of LIP5 and LIP8 being predominant throughout the whole time. No expression of LIP10 was detected at any time point, although this gene is transcribed in vitro in bovine serum albumin medium (data not shown). The results illustrated in Fig. 1 show that in the late stage of the infection process LIP3, LIP4, LIP6, and LIP9 were found to be expressed in a lower percentage of animal samples than in the early stage. There was a two-fold reduction of tested specimens expressing detectable transcripts of LIP3 and LIP4, and a six-fold decrease in the number of samples expressing LIP9. LIP1 and LIP5 were detected in a slightly higher percentage of analyzed samples at the later stage of infection. Interestingly, mRNA of LIP2 was only detectable in the late phase of infection. Expression of LIP8 was found in all samples at both time points.
3.2Comparison of lipase gene expression during systemic liver and kidney infection in mice
To analyze if the expression of the lipase gene family depends on the localization of the infection, we compared the lipase gene expression profiles of Candida cells infecting kidney and liver 3 days after infection (Fig. 2). The expression profile was generated with LIP-specific primers. An EFB1 PCR verified the absence of genomic DNA contamination by detecting a fragment of the spliced mRNA of this housekeeping gene. Fungal cells infecting kidney and liver showed similar lipase gene expression profiles 72 h after inoculation. After 3 days of infection, transcripts of LIP5, LIP6, and LIP8 were predominant. LIP1–3 and LIP9 were expressed in less than 30% of the specimens. There was one exception comparing the profiles of Candida cells infecting kidney and liver, namely the expression of LIP2. This gene was only expressed in a few samples of liver, but the expression was absent in all kidney samples tested.
3.3Expression profile of the lipase gene family during an experimental infection of mucosal epithelium
In addition to the model of systemic candidosis, we also tested the expression of the lipase gene family during an experimental infection of artificial mucosa. The RHE model of oral candidosis was used for this purpose. Histological examinations and expression analysis were performed at three different time points after inoculation with C. albicans in order to examine the temporal regulation of mRNA expression of the 10 known lipase genes during an infection of the RHE model. One part of the artificial skin was fixed for histological studies to follow the progression of infection (Fig. 3), while the remaining epithelial tissue was used for RNA isolation. Transcription of the lipase genes was analyzed by RT-PCR. No amplification was observed for EFB1 or LIP1–10 from uninfected RHE tissue. At 12 h post infection (p.i.) wild-type yeast and hyphal cells of C. albicans caused edema and detachment (acantholysis) of the upper cell layers. Predominantly, hyphal cells were detected at 36 h p.i. At that time point, the tissue of all suprabasal layers was disintegrated and spongiosis and partial necrosis were detectable. As infection progressed, severe mucosal erosion accompanied with vacuolation appeared. EFB1 transcripts were reproducibly detected from all C. albicans-infected RHE samples and at every time point tested (Fig. 4). The cDNA origin of the templates and the absence of genomic DNA were demonstrated by amplification of a 554-bp fragment of the intron-containing gene EFB1. Signals of LIP3 were barely detectable after 2 days of infection while LIP7 was constitutively expressed throughout the duration of the infection. The transcription profile of the lipase gene family is summarized in Table 2. Six lipase genes were constitutively expressed (LIP1 and LIP4–LIP8). In accordance with the results of the mouse model, expression of LIP2 was detectable only at a later stage (36 h) of the infection process and transcripts of LIP9 were only detectable after 36 h p.i. Transcripts of LIP10 were barely detectable at the end of infection.
|LIP||Time after infection|
|12 h||36 h||48 h|
3.4Lipase gene expression in samples of patients with oral candidosis
To investigate whether LIP genes are expressed during human oral candidosis, eight patient samples of saliva were used to analyze the transcriptional profile of the LIP gene family. All patient samples contained only C. albicans which was confirmed by PCR . RT-PCR analysis showed that lipase genes of C. albicans are expressed and differentially regulated during human oral candidosis. Transcripts of different lipase genes were detected in eight clinical samples (Fig. 5). While no expression of LIP10 was detectable in any specimen, LIP1–9 were found to be expressed in at least one saliva sample. Transcripts of LIP1–3, LIP7, and LIP9 were found in one sample and of LIP6 in two samples. Expression of the lipase genes LIP4, LIP5, and LIP8 was detected in at least 50% of the specimens.
We have recently reported that some members of the lipase gene family of C. albicans are constitutively expressed, while others are induced, using different in vitro culture conditions . The present study describes a detailed sequential analysis of the differential and temporal expression of this gene family in different infection models and in patient samples. The yield of fungal mRNA from patient specimens is very low, making conventional expression protocols such as Northern analysis impossible. We used RT-PCR as a highly sensitive and reproducible method to investigate transcription profiles in samples with small amounts of fungal cells. In contrast to Northern blots, RT-PCR allows the discrimination of transcripts from closely related genes such as the members of the LIP gene family.
We analyzed the expression of the LIP gene family in an established mouse model of systemic infection. Systemic infection with Candida cells is the most severe form of candidosis, which is correlated with a high mortality rate. Using RT-PCR, we obtained reproducible data showing that the lipase genes LIP5 and LIP8 were predominantly expressed in cells infiltrating liver and kidney, while no transcripts of LIP10 were detectable. When comparing the early stage to the late stage of systemic mouse infection in the liver, we noticed marked differences in the LIP expression profiles of Candida cells. This clearly shows that the expression profile of the lipase gene family depends on the stage of infection. The group of LIPs that are down-regulated during the progression of infection might encode enzymes that are involved in the early processes of adhesion of fungal cells to liver tissue, which is a prerequisite for subsequent organ invasion. The group of lipases that are expressed during the late stage of infection may help C. albicans to persist and proliferate in the host organ by providing carbon sources like free fatty acids and glycerol. We did not detect any obvious differences in expression profiles of cells infecting liver and kidney, although the kidneys were invaded later than the liver. This is consistent with previous studies dealing with the SAP gene family . Staib et al. have shown that C. albicans cells infecting kidney in a model of systemic mouse infection had a SAP expression pattern similar to cells invading the liver. Our data revealed that members of the lipase gene family are expressed and differentially regulated during systemic mouse infection. The expression status of the fungal cells depends on the stage of infection rather than on the organ localization.
C. albicans can both colonize and penetrate any body surface. For example, it can adhere and proliferate on mucosa of the oral cavity and on dentures. Adhesion itself is an important initial step in pathogenesis of oral Candida infections. It has been demonstrated that there is a direct relationship between adhesion capacity and the ability to cause disease . Besides protein–protein and protein–carbohydrate interactions, the hydrophobicity is believed to provide adherence function. In this scenario Candida lipases might contribute to adhesion by releasing hydrophobic free fatty acids. It has been shown for the black yeast Hortaea werneckii, the causative agent of tinea nigra, that its cell surface is very hydrophobic. This led to the postulation that the lipolytic activity of the fungus increased hydrophobic interactions by liberating free fatty acids , thus enabling H. werneckii to adhere to host tissue. If candidal lipases fulfill the same function during oral candidosis, the lipase genes need to be expressed during this pathogenic interaction.
To ascertain a possible role of the lipase gene family in the colonization of human epithelial tissue, we conducted experiments following two strategies. First, we investigated the infection of artificial oral mucosa (RHE). This model has the advantage of being highly reproducible compared to patient samples. Furthermore, it is possible to perform histological examinations and to analyze temporal changes in gene expression at any stage of infection. Second, we analyzed specimens of patients suffering from oral candidosis. This is, to our knowledge, the first time that the expression pattern of the lipase gene family of a human pathogenic fungus was investigated in human subjects. Using the RHE model, a number of studies have been performed to analyze the SAP gene family [16,24]. Our data clearly demonstrate that at least eight LIP genes are expressed during the RHE infection process, some lipase genes showing sustained expression and others being weakly expressed at a later stage of infection. It should be pointed out that expression of LIP2 and LIP9 was induced only in the later stages of the artificial skin infection. For LIP2 this temporal regulation also resembles the picture of the systemic liver infection, where it was found only in a few samples of liver 3 days after infection. In contrast to the temporal expression profile obtained from the liver infection in mice, LIP9 was induced at the late stage of RHE infection. These findings indicate that lipase genes are differentially regulated depending on the type of infection when comparing systemic and oral candidosis. Besides the differential and temporal regulation in experimental models, we were able to show the expression of the LIP gene family in specimens of patients suffering from oral candidosis. Once more, LIP5 and LIP8 were expressed predominantly, in 50% and 63% of the specimens, respectively. Both genes are expressed under all tested in vitro  and in vivo conditions indicating their importance for the pathogen. LIP4 transcripts were found in four out of eight patient samples. This LIP gene was also expressed during the infection of the RHE model, but it only showed marked expression at the early stage of systemic liver infection. Therefore, we propose that LIP4 preferentially plays a role during superficial infections. This is in agreement with results recently obtained analyzing phenotypic switching in C. albicans. Lan et al. used microarrays to show that LIP4 is regulated during phenotypic switching in C. albicans. LIP4 expression was up-regulated in opaque cells as compared to white cells, which is consistent with the propensity of opaque cells to be more virulent in a cutaneous mouse model than in a systemic mouse model .
Taken together these differential expression patterns suggest that individual LIP genes, possibly encoding Lip isoenzymes which differ in biochemical properties, might have their own special role. It seems reasonable that C. albicans expresses the lipase or lipases according to the demands of the surrounding environment. The availability of related but non-identical and differentially induced forms of lipases might improve the adaptation of C. albicans to different nutritional sources in changing environments. Besides nutritional aspects, these lipases might have additional functions. As we have shown recently, most LIP genes are expressed in media lacking lipid substrates . For another human pathogen, Staphylococcus epidermidis, it has been shown that one of its lipases (GehD) is bi-functional, acting both as a lipase and as a collagen-binding protein . Such a specialized property might support bacterial growth and colonization on both skin and artificial surfaces. Besides the possibility of each lipase having additional functions apart from cleaving lipids, there are a large number of other virulence factors, which we have mentioned earlier, that might interact indirectly with these secreted lipases. This might cause synergistic effects enabling the fungus to colonize its host. For example, it might be possible that Lips lower the pH of the microniche by releasing free fatty acids and thereby optimize conditions for other enzymes such as the secreted proteinases, which are also differentially expressed during systemic candidosis in mice  and during infection of RHE [16,24]. Another possible synergistic effect may be caused by simultaneous expression of candidal phospholipases [28,29]. In Pseudomonas aeruginosa it has been shown that an extracellular lipase and a phospholipase C had a combined action on the increase in 12-hydroxyeicosatetraenoic acid from human platelets and leukotriene B4 from neutrophils . These reactions might lead to disorganized immune reactions, which in turn could initiate tissue damage and stimulate inflammatory processes. Extracellular lipases can also directly impair different defense systems of the host. It was shown that granulocytes incubated with purified lipase of Staphylococcus aureus had a diminished phagocytotic killing of bacteria . Furthermore, Long and colleagues have proposed that staphylococcal lipase in conjunction with fatty-acid-modifying enzyme inactivates bacteriocidal lipids and enhances staphylococcal survival within abscesses .
Our findings illustrate the flexibility of the pathogenic fungus to adapt and change its gene expression profile in response to environmental signals. This trait might enhance the ability of C. albicans to survive and infect different body sites while confronted with different defense systems including variable immune responses. The present study provides evidence that the lipase gene family of C. albicans is expressed during systemic mouse infection, artificial skin infection, and human oral candidosis. Induction depended on the progress of infection. Some LIPs were expressed throughout the infectious process, whereas others were expressed transiently. The expression of the LIP gene family depended on the type of infection.
The further analysis of the lipase gene family will reveal new insights into how this pathogen colonizes and infects the human host successfully.
We thank Siegfried Salomon for helpful discussions and Kimberly Kotovic for critical reading of the manuscript. This work was supported by a grant from the Boehringer Ingelheim Fonds awarded to F.S.
- Sambrook, J., Fritsch, E.R. and Maniatis, T. (1998) Molecular Cloning: A Laboratory Manual, 2nd edn. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.