Pathogenesis of Candida albicans infections in the alternative chorio-allantoic membrane chicken embryo model resembles systemic murine infections

Abstract

Alternative models of microbial infections are increasingly used to screen virulence determinants of pathogens. In this study, we investigated the pathogenesis of Candida albicans and C. glabrata infections in chicken embryos infected via the chorio-allantoic membrane (CAM) and analyzed the virulence of deletion mutants. The developing immune system of the host significantly influenced susceptibility: With increasing age, embryos became more resistant and mounted a more balanced immune response, characterized by lower induction of proinflammatory cytokines and increased transcription of regulatory cytokines, suggesting that immunopathology contributes to pathogenesis. While many aspects of the chicken embryo response resembled murine infections, we also observed significant differences: In contrast to systemic infections in mice, IL-10 had a beneficial effect in chicken embryos. IL-22 and IL-17A were only upregulated after the peak mortality in the chicken embryo model occurred; thus, the role of the Th17 response in this model remains unclear. Abscess formation occurs frequently in murine models, whereas the avian response was dominated by granuloma formation. Pathogenicity of the majority of 15 tested C. albicans deletion strains was comparable to the virulence in mouse models and reduced virulence was associated with significantly lower transcription of proinflammatory cytokines. However, fungal burden did not correlate with virulence and for few mutants like bcr1Δ and tec1Δ different outcomes in survival compared to murine infections were observed. C. albicans strains locked in the yeast stage disseminated significantly more often from the CAM into the embryo, supporting the hypothesis that the yeast morphology is responsible for dissemination in systemic infections. These data suggest that the pathogenesis of C. albicans infections in the chicken embryo model resembles systemic murine infections but also differs in some aspects. Despite its limitations, it presents a useful alternative tool to pre-screen C. albicans strains to select strains for subsequent testing in murine models.

Figure 1. Characterization of the course of infection in chicken embryos infected with C. albicans SC5314 at different developmental days.

(A) Mortality after infection on the CAM (Kaplan-Meyer curve). N = 20 per group per experiment, two independent experiments. Significant mortality (compared to age-matched PBS control, log rank test) was only observed for embryos infected on developmental day (DD) 8 (P<0.01). The logrank test for trend was significant for comparison of infected groups (P<0.0001). (B) Comparison of fungal burden in CAM in embryos infected at different developmental days (DD) as mean and SD, n = 10. (C) Frequency of positive isolation of C. albicans from liver (n = 10 per DD and time point). Dark: positve isolation; white: no fungi isolated. (D and E) Histology of embryos infected on developmental day 10. Periodic acid-Schiff stain (fungal elements: pink). (D) Hyphae invading into and penetrating the full thickness of the CAM 24 h after infection. Arrow: blood vessel penetrated by C. albicans. (E) Histology of macroscopically visible plaque 3 days after infection. Arrows indicate fungal cells.

Figure 2. Age-dependent mortality after application of 100 µg LPS on the CAM.

Survival is shown as Kaplan-Meyer curve, n = 20 per group per experiment, two independent experiments. Significant mortality (compared to age-matched PBS control, log rank test) was only observed for embryos infected on developmental day (DD) 8 (P<0.001) and DD10 (P<0.01). The log rank test for trend was significant for comparison of LPS groups (P<0.005).

Figure 3. Influence of embryonic age on the transcriptional levels of cytokines after LPS application and infection with C. albicans SC5314.

Embryos were either challenged with 100 µg LPS (A) or infected with 10⁵ cfu C. albicans on the CAM (B). N = 5 per challenge and time point, data is shown as mean and SD. DD: developmental day. Asterisks indicate statistically significant differences (P<0.05; 2-way ANOVA and Bonferroni post test).

Figure 4. Addition of IL10 reduces mortality after C. albicans infection.

On developmental day 10, embryos were treated with 100 ng recombinant chicken IL10 (IL10) in PBS/BSA or PBS/BSA alone ( ) and infected with 10⁷ cfu C. albicans SC5314. Survival is shown as Kaplan-Meyer curve, n = 20 per group, two independent experiments. Treatment had a significant effect on survival (P<0.05, log rank test).

Figure 5. Characterization of the course of infection in chicken embryos infected with C. glabrata ATCC2001.

(A) Mortality after infection on the CAM (Kaplan-Meyer curve). N = 20 per group per experiment, two independent experiments. No significant mortality (compared to age-matched PBS control, log rank test) was observed, independent of the developmental day (DD) at infection. (B) Comparison of fungal burden in CAM, n = 10 per group, mean and SD. (C) Frequency of positive isolation in livers of embryos infected at different DD as mean and SD. Dark: positve isolation; white: no fungi isolated. (D) Representative histology of C. glabrata infected CAM. (E) Comparison of cytokine transcription in embryos infected with either C. albicans SC5314 (white) or C. glabrata ATCC2001 (grey) on DD10. N = 5 per time point, data is shown as mean and SD. Asterisks indicate statistically significant differences (P<0.05; 2-way ANOVA and Bonferroni post test).

Figure 6. Fungal burden and dissemination frequencies of C. albicans mutants in infected embryos.

Embryos were infected on developmental day 10 with 10⁵ cfu C. albicans SC5314. (A to C) Fungal burden in the CAM (n = 10 per strain and time point). (D to F) Dissemination frequencies based on positive isolation from the liver (n = 10 per strain and time point). Dark: positive isolation; white: no fungi isolated. (A, D) Comparison of SC5314, CAI4+pCIP10 and BWP17. (B, E) CAI4-derived mutants with altered fungal burden and/or dissemination frequencies. (C, F) BWP-derived mutants with altered fungal burden and/or dissemination frequencies.

Figure 7. Fungal morphology during infection of the CAM.

Histological sections stained with PAS 24 h (except H: 72 h) after infection with 10⁵ cfu. 63× magnification.

Figure 8. Cytokine transcription in the CAM after infection with C. albicans mutants.

N = 5 per time point, data is shown as mean and SD. Asterisks indicate statistically significant differences (P<0.05; 2-way ANOVA and Bonferroni post test). (A) CAI4-derived mutants with altered cytokine transcription profiles. (B) BWP-derived mutants with altered cytokine transcription profiles.