Identification of Pathogen Genomic Differences That Impact Human Immune Response and Disease during Cryptococcus neoformans Infection

Abstract

Patient outcomes during infection are due to a complex interplay between the quality of medical care, host immunity factors, and the infecting pathogen’s characteristics. To probe the influence of pathogen genotype on human survival, immune response, and other parameters of disease, we examined Cryptococcus neoformans isolates collected during the Cryptococcal Optimal Antiretroviral Therapy (ART) Timing (COAT) Trial in Uganda. We measured human participants’ survival, meningitis disease parameters, immunologic phenotypes, and pathogen in vitro growth characteristics. We compared those clinical data to whole-genome sequences from 38 C. neoformans isolates of the most frequently observed sequence type (ST), ST93, in our Ugandan participant population and to sequences from an additional 18 strains of 9 other sequence types representing the known genetic diversity within the Ugandan Cryptococcus clinical isolates. We focused our analyses on 652 polymorphisms that were variable among the ST93 genomes, were not in centromeres or extreme telomeres, and were predicted to have a fitness effect. Logistic regression and principal component analysis identified 40 candidate Cryptococcus genes and 3 hypothetical RNAs associated with human survival, immunologic response, or clinical parameters. We infected mice with 17 available KN99α gene deletion strains for these candidate genes and found that 35% (6/17) directly influenced murine survival. Four of the six gene deletions that impacted murine survival were novel. Such bedside-to-bench translational research identifies important candidate genes for future studies on virulence-associated traits in human Cryptococcus infections. IMPORTANCE Even with the best available care, mortality rates in cryptococcal meningitis range from 20% to 60%. Disease is often due to infection by the fungus Cryptococcus neoformans and involves a complex interaction between the human host and the fungal pathogen. Although previous studies have suggested genetic differences in the pathogen impact human disease, it has proven quite difficult to identify the specific C. neoformans genes that impact the outcome of the human infection. Here, we take advantage of a Ugandan patient cohort infected with closely related C. neoformans strains to examine the role of pathogen genetic variants on several human disease characteristics. Using a pathogen whole-genome sequencing approach, we showed that 40 C. neoformans genes are associated with human disease. Surprisingly, many of these genes are specific to Cryptococcus and have unknown functions. We also show deletion of some of these genes alters disease in a mouse model of infection, confirming their role in disease. These findings are particularly important because they are the first to identify C. neoformans genes associated with human cryptococcal meningitis and lay the foundation for future studies that may lead to new treatment strategies aimed at reducing patient mortality.

Cryptococcal Immune Reconstitution Inflammatory Syndrome: From Blood and Cerebrospinal Fluid Biomarkers to Treatment Approaches

“…The genomic differences in clinical isolates may underlay differential drug susceptibility and virulence of Cryptococcal spps ., which play important roles in the severity of CM and CM-IRIS [ 32 ]. Genetic make-up allowed some Cryptococcal spps .…”

Is It Time To Kill the Survival Curve? A Case for Disease Progression Factors in Microbial Pathogenesis and Host Defense Research

“…Finally, an important recent approach to identify DPFs involves the collection and characterization of within-host isolates of a given microbe via next-generation sequencing (37)(38)(39)(40). These studies are increasingly identifying within-host mutations that provide fitness benefits to the microbe and are strong DPF candidates.…”

The interplay of phenotype and genotype inCryptococcus neoformansdisease

“…VNB is more diverse, was initially discovered in Botswana, and patients infected with these strains tend to have poor clinical outcomes. VNI causes the most cases of cryptococcosis and has three evolutionarily distinct subpopulations; ST93 and ST5 are in separate subpopulations [ 21 , 54 , 55 ].…”

“…The next WGS study that investigated the link between C. neoformans genotype and patient phenotype was published in 2019 and specifically explored differences between ST93 isolates causing changes to patient outcome. Gerstein et al performed WGS on clinical isolates from patients in Uganda, where ST93 is the most frequently identified sequence type [ 55 ]. They identified non-synonymous INDELs and SNPs that were specific to ST93 and not located in the extreme telomeric or centromeric regions.…”

“…The GWAS identified 145 variants associated with 40 genes, with only two previously known to be virulence determinants. Gerstein et al identified that four of these genes were novel virulence factors by testing deletion strains in a mouse inhalation model [ 55 ]. Perhaps most intriguing, considering the evolutionary distance between ST5 and ST93 ( Figure 1 B), is that four genes ( CNAG_ 05185, CNAG_05987, CNAG_02475 , and CNAG_05185 ) were identical between the two studies and an additional three were only a single gene off ( CNAG_07704, CNAG_04921, CNAG_01240 versus CNAG_07703, CNAG_04922, CNAG_01241 for Day et al and Gerstein et al, respectively).…”