Induction of Protective Immunity to Cryptococcal Infection in Mice by a Heat-Killed, Chitosan-Deficient Strain of Cryptococcus neoformans

Abstract

Cryptococcus neoformans is a major opportunistic fungal pathogen that causes fatal meningoencephalitis in immunocompromised individuals and is responsible for a large proportion of AIDS-related deaths. The fungal cell wall is an essential organelle which undergoes constant modification during various stages of growth and is critical for fungal pathogenesis. One critical component of the fungal cell wall is chitin, which in C. neoformans is predominantly deacetylated to chitosan. We previously reported that three chitin deacetylase (CDA) genes have to be deleted to generate a chitosan-deficient C. neoformans strain. This cda1Δ2Δ3Δ strain was avirulent in mice, as it was rapidly cleared from the lungs of infected mice. Here, we report that clearance of the cda1Δ2Δ3Δ strain was associated with sharply spiked concentrations of proinflammatory molecules that are known to be critical mediators of the orchestration of a protective Th1-type adaptive immune response. This was followed by the selective enrichment of the Th1-type T cell population in the cda1Δ2Δ3Δ strain-infected mouse lung. Importantly, this response resulted in the development of robust protective immunity to a subsequent lethal challenge with a virulent wild-type C. neoformans strain. Moreover, protective immunity was also induced in mice vaccinated with heat-killed cda1Δ2Δ3Δ cells and was effective in multiple mouse strains. The results presented here provide a strong framework to develop the cda1Δ2Δ3Δ strain as a potential vaccine candidate for C. neoformans infection.

Importance: The most commonly used anticryptococcal therapies include amphotericin B, 5-fluorocytosine, and fluconazole alone or in combination. Major drawbacks of these treatment options are their limited efficacy, poor availability in limited resource areas, and potential toxicity. The development of antifungal vaccines and immune-based therapeutic interventions is promising and an attractive alternative to chemotherapeutics. Currently, there are no fungal vaccines in clinical use. This is the first report of a C. neoformans deletion strain with an avirulent phenotype in mice exhibiting protective immunity when used as a vaccine after heat inactivation, although other strains that overexpress fungal or murine proteins have recently been shown to induce a protective response. The data presented here demonstrate the potential for developing the avirulent cda1Δ2Δ3Δ strain into a vaccine-based therapy to treat C. neoformans infection.

FIG 1

The time required for the host to completely eradicate chitosan-deficient cda1Δ2Δ3Δ cells in the lung depends on the size of the yeast inoculum. CBA/J mice were inoculated with various doses (equivalent to 10⁵, 10⁶, or 10⁷ CFU) of either strain KN99α or cda1Δ2Δ3Δ mutant cells through intranasal inoculation. At 1, 3, and 7 days p.i., lungs were harvested, homogenized, and serially diluted for CFU enumeration by plating on YPD. Fungal burden is expressed as CFU per lung. The detection limit of the assay was <10 CFU/lung. Error bars indicate standard errors of the means for three mice per treatment group.

FIG 2

The kinetic profiles of mouse pulmonary cytokines were dependent on the inoculum load of C. neoformans cells and the presence or absence of chitosan in the cell wall. (A) Representative cytokine profiles in the lungs at various times after infecting the mice with various doses (equivalent to 10⁵, 10⁶, or 10⁷ CFU) of either KN99α or cda1Δ2Δ3Δ cells. Pulmonary cytokines that showed significantly different levels at day 3 p.i. in mice infected with 10⁷ CFU of KN99α or the cda1Δ2Δ3Δ mutant, in two independent experiments with three mice per treatment group, are displayed. Error bars indicate standard errors of the means. (B) Data from panel A are expressed as fold induction in protein levels of cytokines on day 3 p.i. in the lungs of the mice infected with various doses of either the cda1Δ2Δ3Δ mutant or KN99α compared to their levels in the PBS-inoculated control mouse lungs (n = 3). ****, P < 0.0001 by two-way ANOVA and Bonferroni’s multiple comparisons test. For IL2p70, **, P = 0.0053, and *, P = 0.0193; for CCL2, **, P = 0.002; for CXCL3, ***, P = 0.0006, and *, P = 0.026; for IL-4, **, P = 0.0027, and *, P = 0.05; and for IL-10, ***, P = 0.0009. All data are presented as mean values ± standard errors of the means (SEM). (C, D) Selective high-level induction of Th1-associated cytokines was observed only in cda1Δ2Δ3Δ-infected mouse lungs (C), while their upregulation was not as robust in KN99α-infected mouse lungs (D). The intensity of upregulation of Th2-associated cytokines was not as dramatic as that of the Th1 cytokines. The cytokine levels for each group were normalized to the results for the PBS control group.

FIG 2

The kinetic profiles of mouse pulmonary cytokines were dependent on the inoculum load of C. neoformans cells and the presence or absence of chitosan in the cell wall. (A) Representative cytokine profiles in the lungs at various times after infecting the mice with various doses (equivalent to 10⁵, 10⁶, or 10⁷ CFU) of either KN99α or cda1Δ2Δ3Δ cells. Pulmonary cytokines that showed significantly different levels at day 3 p.i. in mice infected with 10⁷ CFU of KN99α or the cda1Δ2Δ3Δ mutant, in two independent experiments with three mice per treatment group, are displayed. Error bars indicate standard errors of the means. (B) Data from panel A are expressed as fold induction in protein levels of cytokines on day 3 p.i. in the lungs of the mice infected with various doses of either the cda1Δ2Δ3Δ mutant or KN99α compared to their levels in the PBS-inoculated control mouse lungs (n = 3). ****, P < 0.0001 by two-way ANOVA and Bonferroni’s multiple comparisons test. For IL2p70, **, P = 0.0053, and *, P = 0.0193; for CCL2, **, P = 0.002; for CXCL3, ***, P = 0.0006, and *, P = 0.026; for IL-4, **, P = 0.0027, and *, P = 0.05; and for IL-10, ***, P = 0.0009. All data are presented as mean values ± standard errors of the means (SEM). (C, D) Selective high-level induction of Th1-associated cytokines was observed only in cda1Δ2Δ3Δ-infected mouse lungs (C), while their upregulation was not as robust in KN99α-infected mouse lungs (D). The intensity of upregulation of Th2-associated cytokines was not as dramatic as that of the Th1 cytokines. The cytokine levels for each group were normalized to the results for the PBS control group.

FIG 3

Pulmonary leukocyte analysis revealed increased recruitment of Th1-type CD4⁺ T cells in the lungs of mice infected with the chitosan-deficient cda1Δ2Δ3Δ C. neoformans cells. (A, B) Mice inoculated with 10⁷ CFU of the cda1Δ2Δ3Δ strain exhibited significant enrichment of total leukocytes (A) and neutrophils (B) in the lungs on day 3 p.i. compared to their levels in the lungs of mice inoculated with either 10⁷ CFU of KN99α or 50 µl of PBS. (C) The populations of CD4⁺ Th1 cells on days 10 and 21 p.i. were significantly higher in the cda1Δ2Δ3Δ mutant-infected mouse lungs than in the lungs of either PBS- or KN99α-inoculated animals. (D) The frequencies of Th2-type CD4⁺ leukocytes on day 10 p.i. were significantly higher in the KN99α-infected animals than in either PBS- or cda1Δ2Δ3Δ-inoculated animals. (E) Th1 cells induced during infection with cda1Δ2Δ3Δ cells showed increased expression of CD44 on their surface on day 10 p.i., and this was maintained during immune homeostasis (day 21 p.i.). The gating strategy for total leukocytes, neutrophils, Th1 and Th2 cells, and CD44 expression on Th1 cells per whole lung was based upon these leukocyte populations being CD45⁺, CD45⁺ CD24⁺ CD11c⁻ Siglec F-LY6G⁺, CD45⁺ CD3⁺ CD4⁺ CD183⁺ CD196⁻, CD45⁺ CD3⁺ CD4⁺ CD183⁻ CD196⁻, and CD45⁺ CD3⁺ CD4⁺ CD183⁺ CD196⁻ CD44⁺, respectively. The total number of leukocytes was calculated by multiplying the frequency of the leukocyte by the total number of cells determined from the single-cell mouse lung suspensions. Data are presented as the mean results ± standard deviations for three mice per group from one biological experiment. Means were compared among groups within each day using one-way ANOVA followed by the Bonferroni multiple-comparison test. *, P < 0.05; **, P < 0.01.

FIG 4

Vaccination of CBA/J mice with 10⁷ CFU of live cda1Δ2Δ3Δ cells conferred robust protective immunity to subsequent infection with wild-type KN99α C. neoformans cells. Mice were inoculated with 10⁷ CFU of cda1Δ2Δ3Δ cells through inhalation. PBS-inoculated mice served as control. Animals were left for 40 days to resolve the infection. Subsequently, both groups of mice were challenged with 10⁵ CFU of KN99α cells. Virulence was recorded as mortality of mice. Mice that lost 25% of starting body weight were considered to be moribund and were sacrificed. The percentage of mice that survived was plotted against the day p.i. Each survival curve is the average of three independent experiments that had five mice per experimental group.

FIG 5

Heat-killed (HK) cda1Δ2Δ3Δ cells of C. neoformans induced strong protective immunity to a subsequent challenge with wild-type KN99α infection. (A) Mice were immunized with 10⁷ CFU of HK cells of either the wild-type KN99α or the cda1Δ2Δ3Δ strain. Control mice were inoculated with PBS. After 40 days, mice were challenged with 10⁵ CFU of wild-type KN99α cells. Survival of the animals was recorded as described above. The data shown are the average results from four experiments with five mice per experimental group. (B) Pulmonary fungal burden analysis of the mice that were vaccinated initially with 10⁷ HK cda1Δ2Δ3Δ cells and later challenged with 50,000 virulent KN99α cells. At 70 days p.i., lungs were harvested and fungal CFU were enumerated. (C) Mice were vaccinated with various doses (10⁵, 10⁶, or 10⁷ CFU) of HK cells of either the wild-type KN99α or the cda1Δ2Δ3Δ strain. After 40 days, they were challenged with 50,000 wild-type KN99α cells. Survival of the mice was monitored as described above.

FIG 6

Vaccination of mice with HK cells of cda1Δ2Δ3Δ C. neoformans conferred partial protection to subsequent challenge with a lethal dose of a wild-type strain of C. gattii. A group of 10 mice (CBA/J) was subjected to initial vaccination with 10⁷ HK cda1Δ2Δ3Δ cells and later challenged with 10⁵ CFU of either C. gattii R265 or C. gattii WM276. Survival of the mice was monitored as described above. PBS-inoculated mice served as control.

FIG 7

Protective immunity induced by HK cda1Δ2Δ3Δ cells was effective in different inbred mouse strains. (A) A group of 10 C57BL/6 mice were vaccinated with 10⁷ CFU of live cda1Δ2Δ3Δ cells and later challenged with 10⁵ CFU of KN99α cells as described above. Survival was monitored, and the percentage of survival was plotted against the day postchallenge. (B to D) Groups of five mice each of strains 129 (B), A/J (C), and BALB/c (D) were vaccinated with 10⁷ HK cda1Δ2Δ3Δ cells. At 40 days postvaccination, mice were challenged with 50,000 wild-type KN99α cells. Survival was monitored as described above. (E) At 80 days postchallenge, lungs from the surviving mice from the experiment whose results are shown in panels B, C, and D were harvested and subjected to lung fungal burden analysis.