Efficacy of the Anti-Candida rAls3p-N or rAls1p-N Vaccines against Disseminated and Mucosal Candidiasis

Candida species are the third most common cause of nosocomial bloodstream infections [1]. Furthermore, disseminated candidiasis has an unacceptable attributable mortality of 40%–50%, even with modern antifungal therapy [2]. Mucosal candidal infections, including oropharyngeal and vaginal candidiasis, are also very common and cause considerable morbidity and cost [3]. Clearly, new strategies to prevent Candida infections are needed

We have developed an anticandidal vaccine based on an adhesin from C. albicans. Immunization with the recombinant N terminus of Als1p (rAls1p-N) markedly improved survival of immunocompetent or immunocompromised mice infected via the tail vein with C. albicans [4, 5]. The rAls1p-N vaccine also decreased the severity of oropharyngeal candidiasis in our steroid-treated mouse model [5]

The C. albicans ALS family is composed of at least 9 genes that encode adhesins with strikingly different adherence profiles [6]. Of the ALS family members, the ALS1 and ALS3 genes encode adhesins with the broadest array of substrate affinity. When compared with one another, Als1p mediates greater adherence to endothelial cells and gelatin but inferior adherence to epithelial cells [6]. Their differences in adherence qualities suggested that rAls3p-N might have different efficacy as a vaccine immunogen, compared with rAls1p-N. We therefore compared the efficacy of the rAls3p-N with that of the rAls1p-N vaccine in murine models of hematogenously disseminated, oropharyngeal, and vaginal candidiasis

MethodsC. albicans SC5314, a well-characterized clinical isolate that is highly virulent in animal models, was supplied by W. Fonzi (Georgetown University, Washington, DC). The organism was serially passaged 3 times in yeast peptone dextrose broth (Difco) before infection. Female BALB/c retired breeder mice (>6 months old) were obtained from the National Cancer Institute

rAls1p-N (aa 17–432 of Als1p) was produced in Saccharomyces cerevisiae as we described elsewhere [4, 5]. rAls3p-N, encompassing aa 17–432 of Als3p, was produced similarly. Briefly, a 1.3-kb fragment of ALS3 was synthesized by polymerase chain reaction (PCR) using pALS3 and the following 2 primers: 5′-CGGGATCCAAAAGACAATCACTGGTGTTTTCAAC-3′ and 5′-AAAACTGCAGTGGAACTTGTACAATGACAGTG-3′

The PCR fragment was cloned into pQE32 (Qiagen) after digestion with BamHI and PstI. The resulting plasmid (pQA-5) became a template for a second round of PCR with the primers 5′-CACGAGCTCTATGAGAGGATCTCACCA-3′ and 5′-CGCGAGCTCTATCAACAGGAGTCCAAG-3′, to generate a 6X-His–tagged ALS3 fragment. This second PCR fragment was digested with SacI and then cloned into pYEX-S1 (Clontech). The right-oriented clone (pULA3) was introduced into S. cerevisiae DY150 (Clontech) for expression, in accordance with the manufacturer’s instructions, using enriched complete medium

Mice were immunized by subcutaneous (sc) injection of 20 μg of rAls1p-N or rAls3p-N mixed with complete Freund’s adjuvant (CFA; Sigma-Aldrich) at day 0, boosted with another dose of the antigen with incomplete Freund’s adjuvant (Sigma-Aldrich) at day 21, and infected 2 weeks after the boost. Control mice received adjuvant alone on the same schedule

For survival studies, vaccinated mice were infected via the tail vein with the appropriate inoculum of C. albicans SC5314 blastospores. All procedures involving mice were approved by the institutional animal use and care committee, in accordance with the National Institutes of Health guidelines for animal housing and care

Vaccine experiments in our murine oropharyngeal candidiasis model were performed as described elsewhere [5, 7]. In brief, vaccinated mice were immunocompromised by treatment with cortisone acetate (225 mg/kg administered sc on days −1, 1, and 3 of infection). On the day of infection, the mice were anesthetized by intraperitoneal (ip) injection with xylazine and ketamine. Sublingual infection was achieved via calcium alginate urethral swabs saturated with C. albicans. After 5 days of infection, the tongue and hypoglossal tissue were excised for analysis of tissue fungal burden and histopathological analysis

The vaccine was also tested in the murine model of vaginal candidiasis [8, 9]. Vaccinated female BALB/c mice were treated with estradiol valerate (30 μg administered sc) dissolved in peanut oil (both from Sigma-Aldrich) on day −3 of infection, to induce pseudoestrus. On the day of infection, mice were sedated by ip administration of 100 mg/kg ketamine. Sedated mice were infected intravaginally with 1×10⁶ blastospores of C. albicans in 10 μL of Hank’s balanced salt solution. On day 3 after infection, vaginas and ∼1 cm of each uterine horn were dissected en block, homogenized, and quantitatively cultured

Two days before infection, mice were bled by nicking the tail vein. Serum antibody titers were determined by ELISA in 96-well plates, as we described elsewhere [4, 5]. Footpad swelling was determined as we described elsewhere [4]

For statistical analysis, the nonparametric log-rank test was used to determine differences in survival times of the mice. Antibody titers and footpad swelling were compared by use of the Mann-Whitney U test for unpaired comparisons, as appropriate. Correlations were calculated with the Spearman&amp;rank test. P<.05 was considered to be significant

ResultsMice vaccinated with rAls1p-N or rAls3p-N developed anti-Als1p antibody titers that were significantly greater than those of mice receiving CFA alone (median [75th, 25th quartiles] anti-Als1p log₁₀ antibody titer, 2.6 [3.3, 2.3] for CFA; 4.6 [4.7, 4.4] for rAls1p-N; and 4.7 [4.9, 4.4] for rAls3p-N; P<.05 for both vaccine groups vs. CFA alone). Mice vaccinated with rAls3p-N also developed anti-Als3p antibody titers that were significantly greater than those of mice receiving CFA alone (median [75th, 25th quartiles] anti-Als3p log₁₀ antibody titer, 2.6 [2.8, 2.6] for CFA vs. 4.7 [5, 4.7] for CFA + rAls3p-N; P=.002). However, anti-Als3p antibody titers in mice vaccinated with rAls1p-N were not significantly different from those in mice receiving CFA alone, and they were significantly lower than those in mice vaccinated with rAls3p-N (P=.02 for mice vaccinated with rAls3p-N vs. rAls1p-N)

The rAls1p-N and rAls3p-N vaccines resulted in similar delayed-type hypersensitivity responses in vivo, both of which were significantly greater than that induced by CFA alone (median [75th, 25th quartiles] footpad swelling [in millimeters], 0 [0, 0] for CFA; 0.4 [0.6, 0.3] for rAls1p-N; 0.2 [0.4, 0.1] for rAls3p-N; P<.008 for both vaccines vs. CFA alone)

Next, we compared the efficacy of vaccination with rAls3p-N versus rAls1p-N in the mouse model of hematogenously disseminated candidiasis. Both the rAls1p-N and rAls3p-N vaccines resulted in significant improvement in survival, compared with that in mice receiving adjuvant alone (figure 1A). Also, the efficacy of the 2 vaccines was similar (P=.6)

To determine whether anti-Alsp antibody titers or delayed-type hypersensitivity reactions correlated with survival in vaccinated mice subsequently infected with C. albicans vaccinated mice were bled and underwent footpad swelling tests 2 days before infection. Vaccinated mice that did not survive the infection nevertheless had a broad range of antibody titers, and many such mice had anti–rAls1p-N and anti–rAls3p-N antibody titers of ⩾1:50,000 (⩾4.5 log₁₀). As a result, antibody titers did not significantly correlate with survival (P=.06 and ρ=0.4, Spearman&amp;rank correlation test for anti-Als1p antibodies; P=.1 and ρ=0.4 for anti-Als3p antibodies). In contrast, the intensity of footpad swelling reactions did correlate with survival (P=.009 and ρ=0.6, Spearman&amp;rank correlation test) (figure 1B)

Because Als3p mediated superior adhesion to epithelial cells, compared with Als1p [6], we hypothesized that rAls3p-N might have unique efficacy in mucosal models of infection. We therefore compared the efficacy of rAls1p-N with that of rAls3p-N in our steroid-treated, oropharyngeal mouse model of infection and in a model of candidal vaginitis

In cortisone-treated mice with oropharyngeal candidiasis, the rAls1p-N vaccine mediated a strong trend toward reduced tongue fungal burden (P=.054) (figure 2A), whereas the rAls3p-N vaccine significantly reduced tongue fungal burden (P=.005) (figure 2A). By histopathological analysis, both rAls1p-N and rAls3p-N mediated marked reductions in the severity of tissue invasion and inflammatory response, compared with adjuvant alone (figure 2B). In mice receiving adjuvant alone, the surfaces of the tongues were denuded of epithelium and covered along their entire lengths with dense mats of extensively filamented fungus. In contrast, mice vaccinated with rAls1p-N or rAls3p-N had scattered, less-invasive lesions amid extensive areas of normally epithelialized tongue (figure 2B)

Similarly, in a nonimmunocompromised model of candidal vaginitis, the rAls3p-N vaccine, but not the rAls1p-N vaccine, mediated a significant reduction in vaginal fungal burden, compared with CFA alone (median [75th quartile, 25th quartile] [in log₁₀ cfu/g of vagina], 3.5 [3.8, 3.3] for CFA; 3.6 [4.1, 3.2] for rAls1p-N; 2.8 [3.2, 1.9] for rAls3p-N; P=.01 for rAls3p-N vs. CFA and P=.009 for rAls3p-N vs. rAls1p-N)

DiscussionWe have previously reported that the rAls1p-N vaccine protected mice against disseminated candidiasis and ameliorated oropharyngeal candidiasis and that the mechanism of protection was induction of type 1 (cell-mediated) immunity [4, 5]. Here we report that a vaccine based on rAls3p-N, which is 85% homologous to rAls1p-N at the amino acid level [10], was equally effective against disseminated candidiasis but was more effective than rAls1p-N against mucosal infection. The superiority of rAls3p-N was seen in both a steroid-treated model of oropharyngeal candidiasis and an immunocompetent model of candidal vaginitis

We have recently reported that our murine model of hematogenously disseminated candidiasis recapitulates the most severe form of disseminated candidiasis in humans, candidal septic shock [11]. In humans, candidal sepsis is associated with mortality rates of >55%, despite first-line antifungal therapy [12, 13]. In this context, the achievement of ⩾50% long-term survival in our murine model of candidal septic shock with no adjunctive antifungal therapy is encouraging and supports the continued preclinical development of these vaccine candidates

Antibody titers did not correlate with the protective effect of either vaccine during disseminated candidiasis, but the induction of delayed-type hypersensitivity in vivo did correlate with protection. These data are consistent with our prior study that demonstrated the mechanism of vaccine-induced protection was induction of type 1 (cell-mediated) immunity to the fungus [4]. Of interest was our finding that vaccination with either rAls1p-N or rAls3p-N induced similar titers of antibody against rAls1p-N but that rAls3p-N induced significantly higher titers of anti–rAls3p-N antibodies than did rAls1p-N. These data indicate that, despite their high degree of amino acid sequence homology, significant differences in protein structure exist between rAls1p-N and rAls3p-N, such that the humoral immune system can distinguish between them

We have previously reported that the rAls1p-N vaccine ameliorated the histopathological severity of murine oropharyngeal candidiasis [5]. We now report that the efficacy of rAls3p-N against oropharyngeal candidiasis was greater than that of rAls1p-N as assessed by colony counting, although the histopathological appearance was similar in mice vaccinated with either immunogen. However, we and others have previously reported that colony counting can underestimate the tissue fungal burden in the presence of extensively filamented fungi [14, 15], likely because tissue homogenization kills fungal filaments and, thus, artificially lowers colony counts. The present results are concordant, given that tissue fungal burden data appeared to markedly underestimate the effectiveness of both vaccines at reducing histopathological evidence of tongue invasion by filamentous fungi

In a previous study, we found that Als3p mediated more potent adhesion to epithelial cells than did Als1p [6]. Whether the differences in Als1p and Als3p epithelial cell adherence characteristics account for their different degrees of protection against mucosal candidiasis is not yet known. Regardless, the rAls1p-N and rAls3p-N vaccines were equally effective in protecting against hematogenously disseminated (i.e., endovascular) candidiasis

In summary, the anticandidal rAls3p-N vaccine induced equivalent cell-mediated but broader antibody-based responses than did the rAls1p-N vaccine. The immunogens resulted in an equivalent degree of protection against hematogenously disseminated candidiasis, but rAls3p-N mediated greater protection against both oropharyngeal and vaginal candidiasis. These data indicate that rAls3p-N is a promising candidate for further development as an anticandidal vaccine

• Received January 2, 2006.
• Accepted February 10, 2006.

• © 2006 by the Infectious Diseases Society of America