Cryptococcus deuterogattii VGIIa Infection Associated with Travel to the Pacific Northwest Outbreak Region in an Anti-Granulocyte-Macrophage Colony-Stimulating Factor Autoantibody-Positive Patient in the United States

Mortality rates associated with C. gattii infections are estimated to be between 13% and 33%, depending on an individual’s predisposition, and C. gattii has caused at least 39 deaths in the PNW region. There have been four other international travel cases reported in patients from Europe and Asia with travel history to the PNW, but this report describes the first North American traveler who acquired C. deuterogattii infection presenting within the United States and the first case of a C. deuterogattii outbreak infection associated with anti-GM-CSF autoantibodies. Early and accurate diagnoses are important for disease prevention and treatment and for control of infectious diseases. Continual reporting of C. deuterogattii infections is necessary to raise awareness of the ongoing outbreak in the PNW and to alert travelers and physicians to the areas of endemicity with potential risks.

human and veterinary cases confirm that the outbreak is spreading geographically down the West Coast (22,23). C. deuterogattii is a threat not only to native inhabitants in the PNW and the surrounding areas (e.g., Washington and Oregon) but also to those who travel to the area. In fact, travel-related cases have been reported since the outbreak began (24)(25)(26)(27)(28), but the cases appear to represent international tourists from Europe and Asia visiting the PNW.
Here we present a case of travel-related cryptococcal central nervous system (CNS) disease in which an individual residing in North Carolina acquired a C. deuterogattii infection during travel to the PNW but did not begin to show symptoms until several years after returning home. Unique to this case is the extended incubation period of C. deuterogattii, which was three times longer than previously predicted latency periods, before causing disease. Additionally, the patient tested positive for anti-GM-CSF autoantibodies, a suggested risk factor for C. gattii infections (10,29). Although there have been reports of international travel to the PNW resulting in C. deuterogattii infections (25,26,28), to our knowledge this is the first reported case of a travel-associated cryptococcal CNS disease caused by C. deuterogattii from the Pacific Northwest presenting in the United States. Globalization continues to influence the emergence and resurgence of infectious diseases, and continual reporting of travel-related cases of C. deuterogattii is crucial to raise awareness of the ongoing fungal outbreak in the PNW and potential risks to travelers. Additionally, health providers treating domestic patients should be aware of their travel history, even in cases of remote travel years prior to presentation.

RESULTS
Clinical presentation and treatment. The patient was a 69-year-old man who presented with 3 weeks of clumsiness, dizziness, shuffling gait, memory trouble, headache, and worsening fine motor skills. He reported a cough and shortness of breath for 1 to 2 months that had resolved 3 months prior to presentation and had just returned from an intense bicycling trip that had lasted several days. Because of the new and progressive neurological symptoms, magnetic resonance imaging (MRI) of the brain with contrast was performed and showed multiple ring-enhancing lesions with associated edema and central necrosis (Fig. 1A). The largest lesion (3.7 by 3.1 cm) was in the right parietal lobe (Fig. 1A, white arrow), with a 3-mm leftward midline shift. Other abnormalities were present in the left inferior temporal lobe (2.2 by 1.8 cm), frontoparietal lobe (1.7 by 1.7 cm), and parietal lobe (2.3 by 2.0 cm), as well as along the midline dura (2.4 by 1.0 cm).
Given concern that the lesions represented metastases, computed tomography of the chest, abdomen, and pelvis was performed, revealing a 2-cm, spiculated right lower lobe nodule consistent in appearance with small cell lung cancer. The lung lesion was not amenable to sampling, so the left temporal lobe mass from the brain was partially resected for definitive diagnosis. Intraoperatively, a frozen section of the firm, graycolored mass revealed fungal organisms (Fig. 2B). Following a review of the patient's travel history that revealed a trip to Costa Rica 4 months prior to symptom onset where he visited a bird sanctuary, as well as a trip to Vancouver and Seattle 3 years prior lasting several weeks, patient serum was sent for cryptococcal antigen testing and returned positive results, with a titer of 1:1,280 ( Table 1). The patient was started on liposomal amphotericin B (AmB) at 4 mg/kg of body weight daily combined with 2,000 mg of 5-flucytosine (5-FC) four times daily as induction therapy for a presumed cryptococcoma. Additionally, an extended dexamethasone taper was prescribed for cerebral edema (4 mg twice daily for 6 days; the dose was then halved every 3 days for a total of an 18-day course).
A culture of the brain mass turned positive on day 3 of incubation. Matrix-assisted laser desorption-ionization time of flight (MALDI-TOF) analysis of the presumed fungal organism revealed a spectrum consistent with C. gattii. A lumbar puncture was performed immediately, and cerebral spinal fluid (CSF) was sent for cryptococcal antigen testing. This returned a positive (1:40) result, although culture of the CSF was negative (Table 1). Fourth-generation testing for HIV was negative, and the CD4 ϩ T cell count obtained once the patient had completed the dexamethasone taper was normal at 871. Hemoglobin A1c was also within the normal range. Anti-GM-CSF testing was performed on the patient's plasma and detected high titers of these autoantibodies; in addition, elevated levels of the immune system components interferon gamma (IFN-␥) and interleukin-17A (IL-17A) were noted ( Table 2).
The patient improved clinically over the next 4 weeks of induction therapy but then developed mild acute kidney injury (AKI) and cytopenia, prompting discontinuation of AmB and 5-FC treatment and transition to the renally dosed equivalent of 800 mg fluconazole daily. Two days later, a repeat MRI of the brain was done to assess response to induction treatment and revealed an increase in the size of the right parietal lesion and significant vasogenic edema (Fig. 1B). AmB treatment was then restarted and administered in addition to the fluconazole treatment, and a resection of the enlarging parietal lesion was performed. Pathology demonstrated a cryptococcoma with Grocott-Gomori's methenamine silver (GMS) stain positive for fungal elements morphologically consistent with Cryptococcus. No culture was performed at the time of the second resection. The patient received an additional 48 h of dexamethasone postoperatively, but as he was otherwise clinically stable, no further steroid treatment was prescribed.
At 2 weeks later, after improvement in the cytopenia results, fluconazole treatment was stopped and the patient was restarted on 5-FC with continuation of AmB. Over the following 2 weeks, the patient demonstrated continued slow improvement in symptoms, and brain MRI imaging showed stability of lesions (Fig. 1C). Therefore, he was again transitioned to consolidation therapy with a renally equivalent dose of 800 mg fluconazole daily. After 12 additional weeks of high-dose fluconazole, maintenance therapy was started with fluconazole 400 mg daily. His neurological symptoms gradually improved, with resolution of gait abnormalities and continued improvement in strength and coordination. As expected, the AKI and cytopenias resolved with cessation of the AmB and 5-FC. A follow-up brain MRI completed after approximately 10 months of treatment revealed significant improvement (Fig. 1D). Prolonged therapy with fluconazole was anticipated until achievement of complete resolution of the remaining intracranial lesions. A detailed time line of the patient presentation, resections, and antifungal treatment is shown in Fig. S1 in the supplemental material.
Phenotypic analysis placed the clinical isolate in the C. gattii species complex. To confirm the MALDI-TOF results that characterized the clinical isolate as C. gattii, the isolate was subjected to a number of phenotypic tests. The isolate was grown on media to test for growth (YPD); melanin production (niger seed); urease production (Christensen's agar); and growth and color development on medium containing canavanine, glycine, and bromothymol blue (CGB). The isolate produced dark pigment on niger seed (melanin) and pink coloration on Christensen's agar (urease); both results are consistent with all pathogenic Cryptococcus species (Fig. 3A). The isolate grew and produced blue pigment on CGB agar, which is unique to C. gattii species (Fig. 3A) (30). Cryptococcus produces a hydrophilic extracellular capsule as a means of evading the host immune system (31). To analyze capsule production, India ink staining and microscopic evaluation were performed after growth in CO 2 -independent medium for 72 h at 37°C. Using the C. deuterogattii R265 reference strain for comparison, the patient isolate produced similar capsules (Fig. 3B).
As previously mentioned, the majority of isolates characterized in C. gattii are of the MAT␣ mating type. To determine the mating type of the clinical isolate, it was cocultured with known C. deuterogattii VGIIa R265 MATa and MAT␣ isolates. The clinical isolate produced mating structures (hyphae, basidia, and spores) only when paired with a MATa partner, indicating that the isolate was of the ␣ mating type (Fig. 3C). No self-sporulation was observed when the patient isolate was incubated alone on mating media, providing evidence that the isolate was not an a-␣ diploid or self-fertile (Fig. S2).
Molecular analysis placed the clinical isolate in the VGIIa major clade. MLST analysis was performed to assign the clinical isolate to a species within the C. gattii species complex. MLST loci were PCR amplified with primers designed to amplify each sequence, the PCR products were purified and analyzed by Sanger sequencing, and the sequences were assembled in Sequencher using forward and reverse strands to ensure complete double-stranded coverage. Phylogenetic analysis was performed using representative sequences from each of the C. gattii molecular types: C. gattii (VGI), C. deuterogattii (VGII), C. bacillisporus (VGIII), and C. tetragattii (VGIV). The gene sequences of the clinical isolate clustered closely with the C. deuterogattii representative sequence, supported with high bootstrap values ( Fig. 4A; see also Fig. S3). BLAST analysis of the contigs from each locus showed 100% sequence identity to the same loci in VGIIa sequences already deposited in NCBI. Allele designations were assigned for each of the loci, resulting in a clear pattern of the VGIIa genotype (Fig. 4B). Taken together, the genotyping data assign this clinical isolate to the VGIIa major lineage of C. deuterogattii. Fluorescence-activated cell sorter (FACS) analysis was employed to determine that the isolate is haploid, containing a single copy of each chromosome (Fig. 4C).
A lineage of C. deuterogattii VGIIa-like isolates was previously described that displayed a hypermutator phenotype resulting from a loss-of-function mutation in the  MSH2 gene, encoding the MutS homolog required for mismatch DNA repair (32).
Hypermutator strains display increased resistance to the fluorinated orotic acid derivative 5-fluoroorotic acid (5-FOA), the immunosuppressant FK506, and the clinically relevant antifungal drug 5-FC, suggesting a mode of resistance to antifungal therapy (32). To determine whether the clinical isolate described here would exhibit the same phenotype, the isolate was exposed to 5-FC and the ability of the strain to produce 5-FC-resistant colonies was compared to the activity seen with the wild-type and hypermutator control strains. Unlike the de novo msh2Δ mutant, which produced a large number of 5-FC-resistant colonies, the clinical isolate did not produce any C. deuterogattii Infection in U.S. Anti-GM-CSF ϩ Traveler ® colonies on medium supplemented with 5-FC (Fig. 4D). This observation shows that the isolate displayed a largely wild-type level of intrinsic mutation and is not a member of the hypermutator lineage.
The clinical isolate was shown to share its origin with C. deuterogattii outbreak strains. The isolate was subjected to whole-genome sequencing, and the resulting sequence was used in phylogenetic analysis performed with other previously described C. deuterogattii outbreak strains from the PNW (15). The clinical isolate clustered with 10 highly clonal outbreak strains, including the VGIIa major reference strain R265 (Fig. 5A). In accord with the phenotypic data, it did not fall within the hypermutator lineage that includes NIH444, CBS7750, and ICB107 (32). Alignment of the clinical isolate to the R265 reference strain revealed a total of 32 single-nucleotide polymorphisms (SNPs) that differed between the two strains. Among these SNPs, 22 were shared with at least one other strain in the outbreak group whereas 10 SNPs were unique to the clinical isolate (Fig. 5B). Of the 10 private SNPs, 7 were intergenic or intronic mutations and 3 were missense mutations that resulted in an amino acid change in the protein coding sequence. These nonsynonymous amino acid substitutions were all in uncharacterized genes and likely had no impact on the patient's response to clinical treatment (Fig. 5C).

DISCUSSION
In this report, we document the first case of a previously healthy individual becoming infected with C. deuterogattii after travel from the United States to the PNW outbreak region. Three largely clonal groups of isolates of C. deuterogattii have been identified circulating within the PNW: VGIIa/major, which causes the majority of human and animal infections; VGIIb/minor, which causes fewer infections; and VGIIc/novel,  which thus far has been reported only from Oregon and not from Seattle or Canada (21). More recent studies have suggested that C. gattii originated in South America, where it is endemic, which makes the outbreak in the PNW exceptional, given the temperate climate (33,34). A 2018 study by Roe and colleagues estimated C. gattii was introduced to the PNW between 60 and 100 years ago, coinciding with the completion of the Panama Canal and the expansion of international shipping from South America (35). However, C. deuterogattii VGIIb more closely aligns with isolates from Australia, suggesting multiple possible routes, sources, and introductions. A report from 2009 by Byrnes and colleagues documented a case of a travel-acquired C. gattii (VGI) infection (36); since then, four international travel-related C. deuterogattii (VGII) cases in individuals from Denmark, The Netherlands, Germany, and Japan have been reported (25)(26)(27)(28). Overall, there has been a noticeable increase in the number of travel-related cases being reported. Combined with the C. deuterogattii infection described here, this continues to elevate the level of attention to cryptococcal infections.
Following clinical presentation, a fungal mass was removed from the brain of the patient and subjected to phenotypic and genotypic analyses. The patient fungal isolate was identified as C. deuterogattii by the use of microbiological techniques, and detailed MLST and whole-genome sequencing confirmed this finding. Further phylogenetic analysis showed that the patient isolate clusters closely with previously described PNW outbreak strains (15). Among the strains identified as part of the PNW outbreak, three were characterized as being hypermutators due to their ability to produce drugresistant colonies at a high frequency. Following exposure to the clinically relevant compound 5-FC, the patient isolate did not produce a large number of drug-resistant colonies, suggesting that it is not a member of this hypermutator lineage (32). Wholegenome sequencing revealed 32 SNPs differentiating the patient isolate from the R265 reference strain. Among those SNPs, 22 were also found in at least one outbreak strain, but the clinical isolate contained 10 unique SNPs. The private SNPs do not appear to have influenced the pathogenicity of the infecting strain or treatment outcome, as most are intergenic or resulted in amino acid changes in uncharacterized hypothetical proteins unlikely to play a significant role in disease (Fig. 5).
Taking the data together, the clustering of the patient sample with the 10 other highly clonal outbreak strains and the absence of a hypermutator phenotype provide strong evidence that the patient isolate originated within the PNW and that the patient had acquired the infection during his visit to the area 3 years prior, which laid dormant until symptoms began following the return of the patient to North Carolina. To our knowledge, this represents the longest incubation period for a human C. gattii infection to have been reported in the literature. However, there is currently limited evidence available to support an accurate estimate. Review of cases from the outbreak in the PNW suggests an incubation period of 2 to 11 months, with a median of 6 to 7 months (11). Although the patient also had a history of travel to a bird sanctuary in Costa Rica 4 months prior to the onset of symptoms, this seems less likely to have been the source of infection, as C. deuterogattii VGIIa infection is not known to be endemic in that area.
Cryptococcal meningitis in healthy individuals is somewhat rare (1 in 100,000), but the presentation of such cases suggests that normal hosts could harbor rare primary immunodeficiencies (37). Cryptococcal meningitis has been associated with a number of iatrogenic and inherited immunosuppressive conditions, including organ transplantation, hematopoietic malignancies, and steroid use. Furthermore, a number of defects in cell-mediated immunity have been identified as predisposing factors for cryptococcal meningitis. The most common immunodeficiency in non-HIV cryptococcal meningitis is idiopathic CD4 ϩ lymphopenia, a disease caused by various alleles, which is seen in over 25% of cases (37). Additionally, the presence of anti-GM-CSF and IFN-␥ autoantibodies has also been associated with cryptococcal disease (10,38). Monogenic mutations, such as those in the zinc transcription factor GATA2 gene, the transcriptional activator STAT3 gene, and the CD40 ligand CD40LG gene, are also associated with cryptococcosis in seronegative HIV patients (39-41). The patient described in this study did not have any documented iatrogenic causes of immunosuppression or history of immunodeficiency.
Hyper-IgM syndromes often present earlier in life, i.e., in infancy or childhood, manifesting as increased susceptibility to sinopulmonary and opportunistic infections; this patient did not have such a history. Whole-genome sequencing or exome sequencing of the patient could have determined if any of these underlying non-HIV-related predisposing factors played a role in the patient's acquisition of disease, but such analyses were not performed in this study. Additionally, there are data to suggest that total and memory B-cell levels are lower in HIV-uninfected patients with C. neoformans infections (42). However, only limited flow cytometry measuring T-cell but not B-cell populations was performed in this case.
Interestingly, analysis of the patient's blood did show high titers of anti-GM-CSF autoantibodies, which are known to cause acquired pulmonary alveolar proteinosis (PAP) through dysregulation of the function of macrophages, a cell type important for the control of cryptococcal infection. Specifically, these antibodies block phosphorylation of STAT5, which inhibits macrophage signaling (10,29). Biologically inhibitory anti-GM-CSF autoantibodies have been associated with cryptococcal meningitis in some otherwise immunocompetent patients and in those with idiopathic CD4 ϩ lymphopenia. However, it has yet to be confirmed whether they are induced by an unknown trigger, preceding and therefore increasing susceptibility to infection, or whether they are produced in response to infection. A notable absence of high-titer anti-GM-CSF autoantibodies has been found in the vast majority of healthy and immunodeficient control patients without cryptococcal disease reported in the literature thus far (29). In contrast, a recent study did find one healthy control who had anti-GM-CSF antibodies. His plasma inhibited STAT5 production but to a much lesser degree than is seen with those with cryptococcal infection and autoantibodies (10). Because screening for anti-GM-CSF autoantibodies is not routinely performed in healthy individuals, it is not possible to know whether the autoantibodies predated or followed the acquisition of C. deuterogattii by the patient described in this case. The patient was screened for PAP in the clinical evaluation, but a bronchoscopy was not performed as his symptoms did not warrant it (i.e., no progressive dyspenea on exertion, chronic cough, weight loss, or fatigue). As of 15 months after the initial diagnosis, he had not had any symptoms or computed tomography (CT) chest imaging results consistent with PAP.
Among HIV-uninfected patients, intracranial infection caused by C. gattii is associated with a delayed response to therapy and with a higher incidence of neurosurgical intervention than is associated with C. neoformans infection (43). In all patients, C. gattii infection is more likely to cause intracranial cryptococcomas, which often respond poorly to antifungal treatment. Guidelines recommend the same forms of induction, consolidation, and suppressive treatment for infections due to both cryptococcal species, with extension of induction therapy to 6 weeks in those with cryptococcomas. In addition, surgery should be considered for lesions Ն3 cm in diameter with mass effect or for those that fail to reduce in size after 4 weeks of therapy or if there is compression of vital structures (2010 Infectious Diseases Society of America [IDSA] guidelines). Use of adjunctive corticosteroids in the setting of mass effect or vasogenic edema is supported by moderate evidence, but it is of limited (level III) quality and is given weak recommendations. The patient described here showed initial clinical improvement in response to partial resection, steroids, and 4 weeks of induction therapy but demonstrated radiographic deterioration with enlargement of the lesions and surrounding edema. This could have been due to completion of the steroid taper (the last dose was administered 9 days prior to MRI results demonstrating increased sizes of lesions), incomplete response to induction therapy (especially in the setting of potential immunodeficiency), or an immune reconstitution inflammatory response-like syndrome such as has been reported during treatment of C. gattii infection (44). While pathology analysis did reveal fungal elements morphologically consistent with Cryptococcus, the GMS stain does not differentiate between live and dead organisms. Unfortunately, at the time of repeat resection, no culture of the lesion was done to determine if radiographic worsening was due to sterile inflammation or to persistent infection. Since the second neurosurgical procedure, the patient has continued to recover clinically and has also demonstrated significant radiological improvement. While he did receive 2 days of steroids postoperatively, it is unlikely that this would have resulted in the gradual, sustained improvement that ultimately followed. The slow progress suggests that although the second debulking surgery likely played a key role in his management, prolonged antifungal therapy was crucial to his treatment.
Disease caused by C. gattii can differ from that of C. neoformans based on host risks and clinical presentation. In this clinical case, a previously immunocompetent male presented with brain lesions, tested positive for anti-GM-CSF antibodies, and responded relatively well to treatment. Due to the extended (ϳ3-year) dormancy stage of this infecting strain, detailed travel history was key in determining the cause and geographical source of the brain lesion. The need for detailed travel history with heightened clinical suspicion is critical, as the hosts in which C. gattii presents often differ from those in which C. neoformans presents. This case represents another example of the presence of anti-GM-CSF autoantibodies in a host concurrent with C. gattii infection, further strengthening the argument for anti-GM-CSF autoantibodies as a risk factor for C. gattii infection. Increased awareness of this factor will enhance the identification of diseases by treating physicians. In turn, this will lead to treatments that are more appropriate, precise, and timely and, ultimately, to reduced morbidity and mortality.

MATERIALS AND METHODS
Magnetic resonance imaging (MRI). Imaging was performed using gadolinium contrast medium. Pathology. Intraoperative smear and frozen sectioning was performed at ϫ60 magnification using hematoxylin and eosin. Periodic acid-Schiff (PAS) staining and Grocott-Gomori's methenamine silver (GMS) staining were performed at ϫ20 and ϫ60 magnification using formalin-fixed, paraffin-embedded tissue.
Capsule staining. The R265 control strain and the SEC744 clinical isolate were inoculated into liquid YPD (1% yeast, 2% peptone, 2% dextrose) and grown overnight at 30°C with rotation. Cultures were diluted 1:10 into 5 ml of CO 2 -independent medium and incubated for 3 days at 37°C with rotation. A 500-l volume of cells was pelleted and resuspended in 50 l in sterile water. Following resuspension, a 1.5-l volume of cells was spotted onto a microscope slide along with 1 l of India ink. Capsule staining was visualized using bright-field microscopy on a Zeiss Axioskop 2 Plus microscope.
Mating assay. Using sterile toothpicks, the SEC744 clinical isolate and R265 MAT␣ and R265 MATa were patched together onto V8 (pH 5; 5% V8 juice, 0.05% KH 2 PO 4 , 4% agar) as shown in Fig. 2B. Mating plates were incubated in the dark at room temperature for 1 month, and mating patches were observed under a Nikon Eclipse E400 microscope at ϫ40 magnification.
Phylogenetic analysis. Genes were amplified under typical PCR conditions using primers listed in Table S1 in the supplemental material. PCR products were purified and sent for Sanger sequencing (Genewiz, Research Triangle Park, NC). Consensus sequences were built using Sequencher with forward and reverse strands. Gene phylogenies were built using Mega7 software with bootstrap values of 100 and the maximum likelihood method based on the Tamura-Nei model (45). Representative consensus sequences for VGI, VGII, VGIII, and VGIV were acquired from GenBank entries.
Multi-locus sequence typing (MLST). To determine the molecular subtype of the SEC744 clinical isolate, we aligned the consensus sequence from the phylogenetic analysis ( Fig. 4A; see also Fig. S3 in the supplemental material) to GenBank entries using Mega7 software. In cases in which a 100% homologous alignment was found, the allele was assigned with the already established allele designation numbers described by Fraser et al. (46).
DNA extraction and genome sequencing. Genomic DNA was extracted using cetyltrimethylammonium bromide (CTAB) buffer. A 25-ml culture of the SEC744 isolate was pelleted and lyophilized before pulverization with glass beads was performed. The dried cell debris was resuspended in 10 ml CTAB buffer and heated at 65°C for 10 min. An equal volume of chloroform was added to the mixture, and the solution was gently mixed. Following centrifugation, the aqueous phase was transferred to a new tube and DNA was precipitated using isopropanol. The DNA was pelleted, resuspended in dH 2 O, and treated with 20 g/ml of RNase A. Whole-genome sequencing was performed at the Duke University Sequencing and Genomic Technologies Shared Resource (Durham, North Carolina). Libraries were prepared using a Kapa Hyper Prep Kit, and the libraries were run on an Illumina Hi-Seq 4000 instrument to generate 150-bp paired-end reads.
SNP calling and phylogenetic networks analysis. The phylogenetic relationship of the SEC744 clinical isolate to other strains of the C. deuterogattii VGIIa subpopulation was inferred from a matrix containing the SNP calls for the selected set of samples. The genomic paired-end reads were mapped to the C. deuterogattii R265 reference genome (GenBank accession no. GCA_002954075.1) (47) using BWA-MEM short read aligner v.0.7.17-r1188 (48) with default settings. SNP discovery, variant evaluation, and further refinements were carried out with the Genome Analysis Toolkit (GATK) best practices pipeline (49) v.4.0.1.2, including the use of Picard tools to convert SAM to sorted BAM files, fix read groups (module: AddOrReplaceReadGroups; SORT_ORDERϭcoordinate), and mark duplicates. Variant sites were identified with the HaplotypeCaller from GATK using the haploid mode setting and were subsequently filtered using the VariantFiltration module with the following criteria: QD Ͻ 2.0 FS Ͼ 60.0 MQ Ͻ 40.0 MQRankSum Ͻ -12.5 ReadPosRankSum Ͻ -8.0 SOR Ͼ 4.0. Next, variants were parsed with VCFtools v0.1.16 (50) to include only high-quality SNPs, here defined as SNPs with a mean depth of Ͼ10 (-minDP 10) and a minimum quality of 30 (-minQ 30). The option -max-missing 1 was used to exclude sites with missing data in at least one sample. The resulting VCF files were parsed with a custom Perl script into a concatenated FASTA consisting of 526 sites and were then imported to SplitTree v.4.14.2 (51) to construct an unrooted split-network using the Neighbor-net method. The potential mutational effect of the 10 private variants found in the SEC744 clinical isolate was manually inspected by substituting the corresponding bases in the annotated genes of C. deuterogattii R265 with these variants.
Anti-granulocyte-macrophage colony-stimulating factor (GM-CSF) autoantibody testing. Anti-GM-CSF autoantibody testing was performed at the National Institute of Allergy and Infectious Diseases in Bethesda, MD. Briefly, the ability of the antibody to directly bind to interferon gamma (INF-␥), interleukin-17A (IL-17A), or anti-GM-CSF autoantibody was assessed and measured as a function of fluorescence intensity (52).
Data availability. The genomic reads generated here were deposited in the Sequence Read Archive under project accession number PRJNA493061.