Cross-reactivity between vaccine antigens from the chitin deacetylase protein family improves survival in a mouse model of cryptococcosis

Abstract

Meningitis due to the fungal pathogen Cryptococcus neoformans is estimated to cause nearly 200,000 deaths annually, mostly in resource-limited regions. We previously identified cryptococcal protein antigens which, when delivered in glucan particles, afford vaccine-mediated protection against an otherwise lethal infection. Many of these proteins exhibit significant homology to other similar cryptococcal proteins leading us to hypothesize that protection may be augmented by immunologic cross-reactivity to multiple members of a protein family. To examine the significance of protein cross-reactivity in vaccination, we utilized strains of Cryptococcus that are genetically deficient in select antigens, yet are still lethal in mice. Vaccination with a protein without homologs (e.g., Mep1 and Lhc1) protected against challenge with wild-type Cryptococcus, but not against a deletion strain lacking that protein. Contrastingly, vaccination with a single chitin deacetylase (Cda) protein protected against the corresponding deletion strain, presumably due to host recognition of one or more other family members still expressed in this strain. Vaccination with a single Cda protein induced cross-reactive antibody and interferon-gamma (IFNγ) immune responses to other Cda protein family members. Paradoxically, we saw no evidence of cross-protection within the carboxypeptidase family of proteins. Factors such as in vivo protein expression and the degree of homology across the family could inform the extent to which vaccine-mediated immunity is amplified. Together, these data suggest a role for prioritizing protein families in fungal vaccine design: increasing the number of immune targets generated by a single antigen may improve efficacy while diminishing the risk of vaccine-resistant strains arising from gene mutations.

Keywords: CD4 T cell; Cryptococcus; acquired immunodeficiency syndrome; cross-protection; fungal vaccine; protein family.

Figure 1

Infection with Cryptococcus strains deleted of protein used for vaccination. BALB/c mice received three subcutaneous GP-recombinant protein vaccines containing 10 μg antigen/dose before they were challenged with 2x10⁴ CFU of the indicated strain of Cryptococcus neoformans. Survival was measured until 70 DPI. For each group, data were combined from a minimum of two independent experiments. Vaccine antigens either had no cryptococcal homologs (A, B), belonged to the Cpd protein family (C), or the Cda protein family (D-F). C. neoformans KN99 was used as the wild-type strain (WT) and compared to single gene deletion strains corresponding to each studies’ vaccine protein. The number of mice per group is indicated by n=. Kaplan-Meier survival curves were compared using the Mantel-Cox, log-rank test. To adjust for multiple comparisons, significance was determined after applying the Bonferroni correction. ns, not significant; * significant at P ≤ 0.0125; ** significant at P ≤ 0.0025; *** significant at P ≤ 0.00025.

Figure 2

Vaccination with any of the Cda family proteins induces serum IgG reactive with at least one other family protein. BALB/c mice were vaccinated as in Figure 1 , and mice were euthanized for serum collection two weeks after the final vaccination. The reference gel shows the SDS-PAGE of purified recombinant proteins used in GP-vaccines. Each lane contains 2 μg of protein, and the gels were stained with Coomassie. To the right are the Western Blots against serum from mice treated with the GP-vaccines. Serum from mice was applied as a primary antibody with α-mouse IgG as the secondary. Serum was pooled from groups of n=4 for the assay. Mice vaccinated with GPs loaded with mouse serum albumin (MSA) served as a negative vaccine control, and recombinant ovalbumin (Ova) was used as a protein control. Blots are oriented such that the top of the gel is on the left, and molecular masses of the protein standards are denoted by the ticks on the bottom such that the left, middle, and right ticks for each blot represent a protein ladder of 50 kDa, 37 kDa, and 25 kDa, respectively. (A) Serum from mice vaccinated with the indicated recombinant protein. (B) Serum from mice vaccinated with the indicated synthesized Cda2 peptide. See Specht et al., 2022 (12) and Figure S3 for the peptide sequences. The Western blots have been rotated and cropped to show the relevant portions but not spliced or digitally manipulated.

Figure 3

Ex vivo stimulation of splenocytes from mice vaccinated with Cda proteins induce an IFNγ response to other protein family members. For Cda family proteins, BALB/c mice were vaccinated 3 times subcutaneously with 5 μg of antigen/dose adjuvanted with CAF01. Two weeks after the final vaccination, splenocytes were prepared from harvested spleens and 10⁶ cells/well left unstimulated (Unstim), or cultured with the indicated recombinant protein as denoted under the X axis. After 3 days of culture, the supernatants were analyzed for IFNγ by ELISA. Data are expressed as means ± the standard error of the mean. N=4 mice/group; each dot represents the average of two technical replicates for a single mouse. The dotted horizontal line represents the mean IFNγ production of cells stimulated with Ova. Recombinant Ova was expressed in E. coli and purified following the same protocol as recombinant cryptococcal proteins. Significance was determined by one-way ANOVA with Welch’s correction; comparisons were made to Ova. * significant at P ≤ 0.01; ** significant at P ≤ 0.001.

Figure 4

Loss of CD4⁺ T cells ablates serum IgG and ex vivo IFNγ production in the splenocytes of GP-Cda2 vaccinated mice. Mice were vaccinated as in Figure 2 . Two weeks after the third vaccination, the mice were euthanized and spleens and serum were collected for analysis as in Figure 3 and Figure 2 , respectively. (A–C) IFNγ production by the splenocytes of GP-Cda2 vaccinated mice following ex vivo stimulation. Error bars represent the standard error of the mean, with each sample graphed as the average of 2 technical replicates. Significance was determined by ordinary one-way ANOVA with Welch’s correction. ns, not significant; or *** significant at P ≤ 0.0001. (A) n=3 mice for C57BL/6 and n=5 mice for the MHCII-/-. (B) n=3 mice for C57BL/6 and n=4 for B2m. (C) Due to the small size of spleens from muMT mice, only 2.5x10⁵ splenocytes were stimulated per well. C57BL/6 n=3 and muMT n=5. (D) Western blots for IgG response in the pooled serum of GP-Cda2 vaccinated mice. Molecular masses of the protein standards are denoted by the ticks on the left such that the top, middle, and bottom ticks for each blot are 50 kDa, 37 kDa, and 25 kDa, respectively. The Western blots have been cropped to show the relevant portions but not spliced or digitally manipulated.

Figure 5

Vaccine antigen gene expression in vivo. Transcript rank is in accordance with the gene with the highest number of RNA seq reads being ranked “1”. Thus, the more abundantly transcribed genes are towards the top of the Y axis, while those with fewer transcripts are lower. RNA seq data are displayed for C. neoformans isolated from the CSF of human patients (n=33) (14, 34), the CSF of rabbits 1 day post intracisternal infection with C. neoformans clinical isolates (n=12) (34, 35), the lungs of monkeys 7 days after intratracheal infection with C. neoformans H99 (n=3) (36), and the lungs of C57BL/6 mice 7 days after intranasal infection with C. neoformans H99 (n=3) (36). Each dot represents an individual sample, except in the case of rabbit CSF, where each dot represents a pool of the CSF of 3 rabbits (n=12 pools of 3) (34, 35). Median values are denoted by the red horizontal bar. Transcripts from 6,967 genes were found in human CSF; 6,962 genes in rabbit CSF; and 6,975 genes in both monkey and mouse lungs (, –36).

Figure 6

Cross-reactivity and cross-protection generated by cryptococcal vaccination. (A) Cryptococcal antigens can be categorized by those that have no homologs within the genome and those belonging to a protein family. (B) Antigens which do not have any homologs are protected by the adaptive immune response to the respective recombinant protein vaccine only when the antigen is present in the challenge strain. (C) Vaccination with a single antigen that belongs to a protein family, such as the Cda family, can elicit protective adaptive immune responses to the vaccine antigen as well as other antigens within the family. This enables vaccine-mediated protection against deletion strains of Cryptococcus lacking the vaccine antigen.