Induction of SAP7 Correlates with Virulence in an Intravenous Infection Model of Candidiasis but Not in a Vaginal Infection Model in Mice

The secretions of hydrolytic enzymes, specifically, the 10 secreted aspartic proteinases Sap1p through Sap10p have been comprehensively studied as key virulence determinants of Candida albicans (1, 2, 13, 14, 17, 18, 24). Previous studies utilizing reverse transcription-PCR and gene deficient mutant strains have highlighted the importance of certain SAP genes in relation to various models of infection (12). Thereby, individual members of the SAP gene family might have their own roles in the infectious process because environmental conditions have been shown to affect production of Sap proteins differentially, suggesting that Sap proteins may contribute to the development of active Candida infection in different tissues and at certain stages of infection (7, 9-11, 27).

Sap7p shares at most 27% similarity with other Sap proteins (18), and investigation of SAP7 should focus on its genetic characteristics in vivo, because in vitro induction of SAP7 mRNA expression has not yet been detected. SAP7 transcript was detected in 60% of oral candidiasis patients as opposed to 25% of Candida carriers by means of reverse transcription-PCR (19). Felk et al. found that SAP7 transcript was not induced in response to an intraperitoneal model of C. albicans infection that examines invasion of parenchymal organs (4). Vaginal samples examined for the presence of SAP7 transcript showed that strains collected from patients with vaginal candidiasis induced SAP7 significantly compared to samples collected from carriers (20). Furthermore, a study examining infection of reconstituted human vaginal epithelium found induction of SAP7 transcript after 24 h of infection (25).

Due to our own results showing that SAP7 is up-regulated in response to vaginal candidiasis in mice, we postulated that SAP7 plays a role in the pathogenesis of candidiasis. Therefore, we generated a homozygous Δsap7/Δsap7 mutant, which we examined in vitro for growth, stress response, and morphogenetic development as well as in vivo for invasion, dissemination, and survival in host tissue.

Genetically engineered C. albicans strains created and used in this study are summarized in Table 1. C. albicans strains were cultured on agar plates with complete supplement medium without uridine or in liquid medium, respectively (28). Nucleic acid techniques and microscopic analysis were performed as described previously (26). The vaginal, intravenous, intraperitoneal, and pulmonary models of C. albicans infection were performed as described previously (3, 5, 14, 15, 26). Survival curves were generated according to the Kaplan-Meier method by using the PRISM program (GraphPad Software) and compared using the log rank test (16). Total RNA from the lavage samples containing mucus, epithelial cells, and C. albicans isolates was isolated by hot acid phenol extraction. A FastStart DNA Master SYBR green I kit (Roche, Mannheim, Germany) was used for real-time PCR analysis using the LightCycler system (Roche, Mannheim, Germany). Oligonucleotide primers SAP7.01 (5′-GAAATGCAAAGAGTATTAGAGTTATTAC-3′) and SAP7.02 (5′-GAATGATTTGGTTTACATCATCTTCAACTG-3′) were used for detection of SAP7 cDNA. Primers EFB1.03 (5′-AACGAATTCTTGGCTGACA-3′) and EFB1.04 (5′-GCGGCTGGGGCTTTACC-3′) were used to amplify EFB1 that encodes elongation factor-1β. Copy numbers were calculated, and values represented are normalized (number of copies of SAP7 cDNA/number of copies of EFB1 cDNA). Sequential homologous recombination (6) was used to delete both SAP7 alleles in C. albicans strain CAI-4 to generate the homozygous Δura3/Δura3 Δsap7/Δsap7 mutant BNT16. This strain was then transformed with pVEC (21) to make BNT18 [Δsap7/Δsap7(pVEC)] or plasmid p348 (pVEC-SAP7) containing the SAP7 wild-type gene to generate the revertant strain BNT19 [Δsap7/Δsap7(pVEC-SAP7)]. Appropriate genotypes and presence or absence of SAP7 expression of BNT16, BNT18, and BNT19 were confirmed by PCR and Southern blot analysis in vitro and in vivo.

We infected mice intravaginally with the SC5314 wild-type strain and performed lavages on days 1, 7, and 17. SAP7 levels were detectable by real-time PCR after overnight growth in complete supplement medium without uridine (inocula) at very low levels (between 6.0 × 10⁻⁶ and 1.6 × 10⁻⁵ copies SAP7/copy EFB1) but rose at least 30-fold after 24 h of infection (Fig. 1). By day 7 postinfection, levels had risen over 1,000-fold and remained high through day 17 of infection. This suggested that induction of SAP7 occurred as a consequence of murine vaginal infection and is consistent with results seen with vaginal samples from patients.

The deletion of SAP7 had no effect on cellular replication, colony phenotype, anaerobic growth, resistance to hydrogen peroxide, or the ability of C. albicans to form hyphae at 37°C in medium containing 10% fetal calf serum (data not shown). Furthermore, deletion of SAP7 had no effect on the number of C. albicans CFU recovered from the vaginal canal, as illustrated in Fig. 2A. Vaginal fungal burdens of mice infected with either the Δsap7/Δsap7 mutant or the control increased after 24 h but then dropped and remained stable (between [5.3 ± 3.3] × 10³ and [1.3 ± 0.7] × 10⁴) throughout the 28 days of observation.

In the intravenous model of systemic candidiasis (15, 23, 26), injection of the control BNT19 resulted in the death of 80% of the mice by day 12 and 100% by day 24 postinfection (Fig. 2B). In contrast, 60% of mice survived up to 56 days after infection with the Δsap7/Δsap7 mutant BNT18, and the remaining animals showed no clinical signs of disease (P < 0.02). Although previous work had suggested a role of proteinases in a pulmonary model of candidiasis (3), no significant differences between mutant and control strains during short-term pulmonary infection of immunosuppressed mice were observed (data not shown). Moreover, invasiveness of the Δsap7/Δsap7 mutant was only slightly but not significantly reduced as measured by organ fungal burdens after intravenous infection.

SAP7 has provided difficulty for researchers due to the futile attempts to identify conditions that induce SAP7 in vitro. By use of sensitive techniques, trace amounts can be found in overnight cultures; however, no tested conditions seem to significantly up-regulate SAP7 transcription (11, 22). However, the observation that SAP7 is strictly an in vivo factor signifies that it may be required during the pathogenesis of candidiasis. In our vaginal model of murine candidiasis, SAP7 expression in SC5314 rose after one day of infection and remained highly induced compared to inoculum throughout the 17 days of observation. This confirms findings from previous studies, including those involving patient samples, mouse models, and the examination of C. albicans by using an in vitro reconstituted human vaginal epithelium model of disease (19, 20, 22, 25). Surprisingly, in the vaginal model in which SAP7 had been induced, no significant differences between the SAP7-deficient strain and the control organisms were observed over 28 days of infection. Although the observed up-regulation of SAP7 is a consequence of this type of superficial infection (20, 25), it could well be that C. albicans expresses at least one or even several redundant genes which compensate for the SAP7 defect, and therefore, the Δsap7/Δsap7 mutant does not display an obvious phenotype in virulence in this model of candidiasis. It underscores the fact that induction or lack thereof is not a predictor of involvement in a process. Genes induced under a set of conditions may not be essential to the process being observed. The increase in the survival time of mice after intravenous infection with the Δsap7/Δsap7 mutant was very clear, while the number of fungal CFU recovered from the kidneys indicated only a trend towards a reduced invasiveness. The reasons for this type of selective virulence attenuation of the Δsap7/Δsap7 mutant only after intravenous infection may be due to niche specificity which has been reported for other members of the SAP gene family (27, 28).

• Open in new tab
• Download powerpoint

FIG. 1.

Real-time PCR analysis of vaginal lavage samples after C. albicans infection demonstrates induction of SAP7. Levels of SAP7 cDNA in pooled lavage samples from vaginally infected mice (n = 3) after days 1, 7, and 17 postinfection. Each sample was examined in triplicate for the presence of SAP7 and EFB1 (housekeeping gene) cDNA. Levels of SAP7 were normalized to levels of EFB1, and the ratio (Log₁₀) is represented in the graph. Tissue, prior to infection, was negative for SAP7. Results of two independent experiments are shown in parallel (white bars, data from experiment 1; black bars, data from experiment 2; error bars, standard deviations).

• Open in new tab
• Download powerpoint

FIG. 2.

The Δsap7/Δsap7 mutant is not attenuated after vaginal infection. BALB/c mice were challenged intravaginally (A) or intravenously (B) with the Δsap7/Δsap7 mutant strain BNT18 (open squares) or the control strain BNT19 (solid squares). (A) Fungal burdens (log₁₀ CFU) in vaginal lavage samples from mice after inoculation with 5.0 × 10⁴C. albicans cells (n = 5). Data are representative of three independent experiments. (B) Survival curves of mice (n = 5) injected with 5 × 10⁵C. albicans cells. Graph is representative of five independent experiments.

• Copyright © 2005 American Society for Microbiology