Genotyping Analysis of Cryptococcus deuterogattii and Correlation with Virulence Factors and Antifungal Susceptibility by the Clinical and Laboratory Standards Institute and the European Committee on Antifungal Susceptibility Testing Methods

Data about the relationship between their molecular types, virulence factors, clinical presentation, antifungal susceptibility profile, and outcome are still limited for Cryptococcus deuterogattii. This study aimed to evaluate the molecular and phenotypic characteristics of 24 C. deuterogattii isolates from the southeast region of Brazil. The molecular characterization was performed by multilocus sequence typing (MLST). The antifungal susceptibility profile was obtained according to CLSI-M27-A3 and EUCAST-EDef 7.1 methods. The virulence factors were evaluated using classic techniques. The isolates were divided into four populations. The molecular analysis suggests recombinant events in most of the groups evaluated. Resistance and susceptibility dose-dependent to fluconazole were evidenced in four isolates (16%) by EUCAST and in four isolates (16%) by CLSI methods. The agreement at ±two dilutions for both methods was 100% for itraconazole, ketoconazole, and voriconazole, 96% for amphotericin B, and 92% for fluconazole. Significant differences in virulence factor expression and antifungal susceptibility to itraconazole and amphotericin B were found. The mixed infection could be suggested by the presence of variable sequence types, differences in virulence factor production, and decreased antifungal susceptibility in two isolates from the same patient. The data presented herein corroborate previous reports about the molecular diversity of C. deuterogattii around the world.


Introduction
Cryptococcosis is a global fungal infection that occurs predominantly in immunocompromised hosts or apparently immune-competent individuals.It is caused by the Cryptococcus neoformans species complex (CNSC) and the Cryptococcus gattii species complex (CGSC), respectively [1].The clinical presentation, outcome, geographical distribution, host preference, ecological niche, and molecular profile are different in both species.results, and biological results are described in Table S1 and have already been published elsewhere [39].

Determination of Mating Types
Genomic DNA extraction and mating type PCR of these isolates were performed in accordance with previous report [40].

Minimum Spanning Tree, Multidimensional Scaling Plots, and Split Decomposition Analysis
The minimum spanning tree was generated using the goeBURST algorithm in the PHILOVIZ II software (http://www.phyloviz.net/wiki/,accessed on 15 April 2022) [43,44].This analysis was generated from the concatenated sequence regions to visualize the relatedness of geographic origin, production of virulence factors, and antifungal susceptibility profiles with their allelic profiles.
Classical Multidimensional Scaling plots were constructed using a similarity matrix by arranging samples in two-dimensional space according to their relative similarities in the software MLSTest v.1.0.1.23[45].In addition, the split decomposition analysis was performed in the software SplitsTree v. 4.13.1 [46] to evaluate the distribution of the main groups found by the other molecular techniques.

Nucleotide Diversity
The software DNAsp 5. 10 [47] was used to calculate the extent of the main markers of DNA polymorphisms.In addition, the presence of recombination was also checked in the software SplitsTree v. 4.13.1 [46].

Genetic Differentiation Based on Sequences of Populations
A hierarchical analysis of molecular variance (AMOVA) was performed in Arlequin 3.1 in order to examine the distribution of genetic variation and determine the extent of connectivity among populations based on allelic profiles.Statistical significance was determined by comparing the observed results with 10,000 permutated datasets established with a null hypothesis of no genetic differentiations among geographic populations within each analyzed dataset.The population differentiation test (Fst) was calculated from concatenated sequences of the seven MLST housekeeping genes and used to visualize the genetic distance among populations evaluated in the present study.
The melanin production was evaluated by direct visualization and spectrophotometry [55].The results were reported as optical densities (OD) at 480 nm and represented by the arithmetic mean absorbance values [56,57].The result was interpreted as follows: OD < 0.45, low activity; OD = 0.45 to 0.69, moderate activity; and OD > 0.69, high activity.The urease activity was evaluated by spectrophotometry in urea Christensen broth (Difco, Sparks, MD, USA) [58].The result was interpreted as follows: OD < 1.1, low activity; OD = 1.1 to 1.3, moderate activity; and OD > 1.3, high activity.
All experiments were performed in quadruplicate with at least two independent replications, and the results were described by the generated average.In all experiments, the reference strains C. gattii ATCC 24065 (serotype B) and Candida krusei ATCC 6258 were used as positive and negative controls, respectively.
The antifungal susceptibility tests were performed using the broth microdilution technique following the Clinical and Laboratory Standards Institute (CLSI) recommendations available in the documents M27-A3 [29], Supplement 4 of CLSI [30], and European Committee for Antimicrobial Susceptibility Testing-EUCAST-EDef 7.1 [31].The optical density (DO) values were recorded on the microplate reader (Benchmark Plus, Bio-Rad ®, , Alfred Nobel Drive, Hercules, CA, USA).

Statistical Analyses
Statistical analyses of phenotypic and molecular data were performed using Bioestate v. 5.0 (https://www.mamiraua.org.br/pt-br,accessed on 10 April 2021), MS Excel (Microsoft Corporation, Redmond, WA, USA), and GraphPad PRISM v. 6.0 (https: //www.graphpad.com,accessed on 10 April 2021).The normality of the data was evaluated using the D'Agostino Pearson test.The homogeneity of variances among groups was tested by Bartlett's test when the data presented a normal distribution.After these tests, all phenotypic and clinical data analyses were performed by non-parametric tests.The variables were evaluated by the Mann-Whitney test to compare two groups and the Kruskal-Wallis test to compare three or more groups, applying Dunn's post-test if necessary.The correlation between the two variables was evaluated through the Spearman test.To compare the genetic data with categorical variables, the Fisher's exact test or chi-square test was used.p-values less than 5% (p < 0.05) were considered statistically significant.
The sequence analysis of these isolates exhibited high variability (Hd = 1.00, π = 0.00144-0.03192,and k = 6.000-131.33)(Table 1).The most polymorphic was the Pop4 (Hd = 1.00, π = 0.03192, and k = 131.33),and the least polymorphic was the Pop1 (Hd = 1.00, π = 0.00144, and k = 6.000).In accordance with their origin, the most polymorphic group was composed of isolates from BH (Hd = 1.00, π = 0.02070, and k = 85.400), and the least polymorphic group was composed of isolates from the TM region (Hd = 1.00, π = 0.00317, and k = 13.179).The values of Fst revealed significant molecular differences among populations evaluated in the present study (Fst ranging from 0.023 to 0.599) (Table 2).The populations most closely related were Pop3 × Pop4 (Fst = 0.126), followed by Pop3 × Pop2 (Fst = 0.249), and the more distant were Pop1 × Pop3 (Fst = 0.599), followed by Pop1 × Pop2 (Fst = 0.493).In accordance with their clinical site, the most polymorphic group included those isolates recovered from cerebrospinal fluid (Hd = 1.00, π = 0.03192, and k = 14.267), and the least polymorphic group included isolates from the skin (Hd = 1.00, π = 0.00193, and k = 8.000).The neutrality tests (Tajima's D, Fu and Li's D, Fu and Li's F, and Fu's Fs) evidenced purifying selection or population expansion for most evaluated groups.The results obtained by the Watterson estimator (theta) method and by the PHI test suggest recombinant events in most of them (Table 1).The neutrality tests (Tajima's D, Fu and Li's D, Fu and Li's F, and Fu's Fs) evidenced purifying selection or population expansion for most evaluated groups.The results obtained by the Watterson estimator (theta) method and by the PHI test suggest recombinant events in most of them (Table 1).The main four groups are marked with five different colors.The isolates are described according to the sequence type number (ST), followed by your identification.(D) Classical Multidimensional Scaling Plot.The X and Y axes explain variabilities of 59.6% and 19.4%, respectively.Both axes explain 79% of the variability.(E) Split decomposition analysis applying the NeighborNet algorithm using the uncorrected-P parameter model and evidencing the diversity and branching ambiguities attributable to recombination events.The observation that isolates are linked to each other by multiple pathways and are forming an interconnected network rather than a single bifurcating tree is suggestive of recombination.The phi test for recombination implemented in the software SplitsTree showed significant evidence (p < 0.0001) for recombination.The STs belonging to the main clusters identified in the previous phylogenetic analysis were also separated using the split decomposition.The main clusters identified are highlighted as follows: Pop1: red; Pop2: green; Pop3: blue; and Pop4: yellow.
The isolates G8 (ST344) and G9 (ST345) were recovered from skin fragments of an apparently immunocompetent patient in two different cultures with a 22-day interval between them.These isolates showed differences in the GPD1 and SOD1 loci.In the GPD1 locus, the G8 isolate showed AT21, while the G9 isolate presented AT6.The nucleotide difference between these ATs was based on the insertion of guanine in AT21 at position 157 in relation to AT6.In the SOD1 locus, the G8 isolate showed AT14, while the G9 isolate presented AT58.These ATs exhibited several nucleotide differences, as described in Table 3.Moreover, the isolates also showed phenotypic differences in virulence factor production, and G9 presented a mildly decreasing antifungal susceptibility to fluconazole by CCLI and EUCAST methods (Table S1 and Figure 2).

Virulence Factors
The capacity for capsular synthesis was verified in all C. deuterogattii isolates.No differences in the production of melanin, proteases, yeast size, hemolytic activity, proteases, or phospholipases were found.A positive correlation between urease activity with hemolytic activity and gelatin hydrolysis with phospholipases was observed (Figure S1).A negative correlation between urease activity with phospholipase and urease activity with gelatin hydrolysis was identified (Figure S1).
In addition, most isolates of Pop4 were high producers of urea and high or medium producers of melanin.

The Minimum Spanning Tree
In order to infer patterns of evolutionary descent among clusters of related genotypes and to identify relations with virulence factors, antifungal susceptibility, clinical outcome, and geographic origin, the goeBURST analysis was applied (Figure 2).Two main clusters were identified.The first and minor are composed of the group founder (GF) GF353 and its descendants.The second and major are represented by GF338 and its descendants.All isolates from the TM region were grouped into GF338 as shown in Figure 2C.Regarding the production of virulence factors, the results show a correlation between gelatin hydrolysis absence and GF353 (Figure 2G).All isolates from GF345 presented high activity for phospholipases (Figure 2E) and urease (Figure 2D).Isolates from patients with poor outcomes were grouped in the GF344, whereas most of those who were grouped in the GF338 were cured (Figure 2F).Regarding antifungal susceptibility, most of the GF338 isolates were susceptible or wild-type to fluconazole by the CLSI method (Figure 2A), and the GF344 exhibited resistance to fluconazole by the EUCAST method (Figure 2B).

Discussion
The description of several outbreaks of cryptococcosis caused by the C. deuterogattii genotype in the last decades in some temperate areas of North America called the attention of the scientific community since it was formerly restricted to tropical and subtropical regions of the world [21,60].Due to this fact, the evaluation of several aspects of this molecular type became more relevant [2,15,61,62].
In this sense, the identification of potential correlations between molecular profiles with geographic origin or phenotypic characteristics such as antifungal susceptibility, pathogenicity, and virulence factors, among others, is pivotal.Currently, these correlations are weak, and some discrepancies and questions remain to be solved [63][64][65][66][67].
Data on the production of capsular polysaccharide and melanin, urease, and phospholipase activity, among others, partially contributed to separating the C. neoformans species complex (CNSC) from the CGSC ones [1,21,68,69].Although isolates from these species exhibit most of these features, it is unclear yet if different expression levels could have influenced the hypervirulence of C. deuterogattii isolates observed during the recent outbreaks [22].
In the present study, populations of C. deuterogattii isolates exhibited variable profiles regarding the production of virulence factors.The capsule size, urease production, and gelatine hydrolysis were more significant.In line with these findings, formerly reported studies described an association among STs and/or populations of Cryptococcus spp. with clinical and biological characteristics, such as patient outcome [63], male gender [66], immune response, capsule and melanin production [63], antifungal resistance [64,65], and genotype virulence [21].The latter was evaluated by a Galleria mellonella experimental model, where it was pointed out that virulence is related to the distinct characteristics of individual strains and is not specific to a particular molecular type of CGSC [67].
A significant difference in urease activity among the populations evaluated was found.Urease is an enzyme that acts in phagosomal acidification by hydrolyzing urea to generate ammonia [70].Its production has been considered pivotal during the process of central nervous system invasion [71,72].Additionally, recently it was reported that urease acts together with melanization, which is another important fungal virulence factor [73].
The identification of specific genotypes of Cryptococcus spp.and their correlations with patterns of antifungal susceptibility and virulence could be an important epidemiological tool to improve vigilance for the emergence of resistant and/or more virulent strains [38,74].
Herein, the isolates from Pop4 exhibited a higher production of virulence factors than the others.This fact could indicate a greater virulence of these genetically more dispersed isolates in relation to other populations.Of note, it is important to mention that these isolates were not grouped in a monophyletic way like the other populations, and they exhibited the greatest genetic variability.Despite the small number of isolates included, the preliminary data obtained could suggest a relationship between genotypic differences in C. deuterogattii and higher expression of virulence factors and how this could influence the patient's outcome.
Epidemiological surveillance of the antifungal resistance of C. deuterogattii isolates is pivotal in order to guide the therapy of patients with cryptococcosis caused by this genotype.Currently, validated antifungal breakpoints for Cryptococcus spp.are not available.Thus, the epidemiological cut-off values (ECVs) are used to aid in the detection of non-wild-type strains for antifungal susceptibility in a given fungal population [30,75].Then, the cut-off point of 8 µg/mL has already been used in previous reports as a cut-off point for SSD to fluconazole [29,59,76].All isolates herein evaluated presented a wild-type phenotype by the updated CLSI method, despite the fact that four presented SDD and four exhibited resistance by the EUCAST method using breakpoint interpretation for Candida species.
A decreased susceptibility among C. deuterogattii isolates with high MICs to azole drugs had already been observed by others elsewhere [35,36,[77][78][79][80][81].Several mechanisms of acquired resistance to fluconazole have been described.These include previous exposure [82], drug target alterations encoded by the gene ERG11 [83], subset populations presenting heterogeneous resistance that proliferate during treatment [83], the presence of group I introns in the mitochondrial large subunit rRNA gene (LSU) [27], and significantly higher expression of ABC transporters [28], among others.
The antifungal susceptibility difference among molecular types, STs, and/or populations had already been pointed out for the CNCS and CGCS complexes [64,65,84].Antifungal susceptibility differences among populations to ITZ by CSLI and to AMB by EUCAST were herein identified as well.This fact is relevant since it raises the question of whether patients infected with these isolates could have different clinical outcomes.
The meaning of Cryptococcus spp.mixed infections in the clinical picture and outcome context of patients with cryptococcosis is unknown, and it needs to be better evaluated, mainly for C. deuterogattii [85][86][87].In this report, C. deuterogattii isolates recovered from the same patient presented mixed infections (G8 and G9).They presented different STs (ST344 and ST345) and were recovered from the culture of two skin fragment samples collected at an interval of 22 days.The patient, a farmer who presented with primary cutaneous cryptococcosis after traumatism with eucalyptus tree logs, was apparently immunocompetent, and all clinical and laboratory exams were normal.It was cured after ten weeks of fluconazole.Despite his good outcome, these isolates presented differences related to the expression of virulence factors and antifungal susceptibility, which can raise the hypothesis that they represent a mixed infection and not a reinfection [88].
Data about mixed infections by Cryptococcus spp.are scarce, and this could be related to technical bias.Molecular studies usually select one single colony isolated from the culture of a unique anatomical site, which would make the identification of mixed infections unlikely [87,89].Mixed infections with pathogenic species of Cryptococcus have already been reported elsewhere, using different molecular tools and with frequencies ranging from 16.7% to 66% [64,85,87,90,91].The main fact supporting the occurrence of mixed infections could be the result of either co-inoculation or in vivo evolution (microevolution) [85].However, microevolution could not explain the simultaneous presence of A and D serotype isolates in the same patient, as formerly described [85,90].Despite the measures to avoid contamination in the routine mycology lab, cross-contamination of clinical samples cannot be ruled out.
Despite the low number of strains evaluated, the results herein described corroborate previous reports about the molecular diversity of the Brazilian VGII C. gattii/deuterogatti isolates and reinforce the occurrence of different virulence factors' expression levels and antifungal susceptibility patterns among them.These facts are relevant in the clinical and outcome contexts of patients with cryptococcosis caused by this genotype.

Figure 1 .
Figure 1.Phylogenetic analysis of 24 Cryptococcus deuterogattii isolates.Analysis performed by maximum likelihood (ML) (A), unweighted pair group method with arithmetic mean (UPGMA) (B), and neighbor-joining (NJ) (C) methods, using the concatenated data set of the seven MLST loci (CAP59, GPD1, LAC1, PLB1, SOD1, URA5, and the IGS1 region).The tree is drawn to scale, with branch lengths in the same units as those of the evolutionary distances used to infer the phylogenetic tree.

Figure 1 .
Figure 1.Phylogenetic analysis of 24 Cryptococcus deuterogattii isolates.Analysis performed by maximum likelihood (ML) (A), unweighted pair group method with arithmetic mean (UPGMA) (B), and neighbor-joining (NJ) (C) methods, using the concatenated data set of the seven MLST loci (CAP59, GPD1, LAC1, PLB1, SOD1, URA5, and the IGS1 region).The tree is drawn to scale, with branch lengths in the same units as those of the evolutionary distances used to infer the phylogenetic tree.The analysis involved 24 nucleotide sequences of isolates from the present study.There were a total of 4180 positions in the final dataset.Numbers at each branch indicate bootstrap values > 50% based on 1000 replicates by each of the three.Evolutionary analyses were conducted in MEGA 6.The main four groups are marked with five different colors.The isolates are described according to

Figure 2 .
Figure 2. Minimum spanning tree of the 24 studied Cryptococcus deuterogattii isolates using the goeBURST algorithm.The distribution of the STs was compared with susceptibility to fluconazole by CLSI (A) and EUCAST (B) methods.Procedence (C).Production of the main virulence factors and clinical data are as follows: urease (D), phospholipase (E), outcome (F), gelatin hydrolysis (G), hemolytic activity (H), and clinical source (I) and melanin production (J).The group founders (GF) (highlighted by a yellow line) found were GF338; GF353; GF345; and GF344.Two main clusters were identified.The first and minor are composed of the GF353 and its descendants.The second major is represented by GF338 and its descendants.Each circle represents a unique ST.The numbers presented in the dashed branches represent at least one and the maximum of five differences in alleles, respectively.TM region-Triângulo Mineiro region.

Figure 2 .
Figure 2. Minimum spanning tree of the 24 studied Cryptococcus deuterogattii isolates using the goeBURST algorithm.The distribution of the STs was compared with susceptibility to fluconazole by CLSI (A) and EUCAST (B) methods.Procedence (C).Production of the main virulence factors and clinical data are as follows: urease (D), phospholipase (E), outcome (F), gelatin hydrolysis (G), hemolytic activity (H), and clinical source (I) and melanin production (J).The group founders (GF) (highlighted by a yellow line) found were GF338; GF353; GF345; and GF344.Two main clusters were identified.The first and minor are composed of the GF353 and its descendants.The second major is represented by GF338 and its descendants.Each circle represents a unique ST.The numbers presented in the dashed branches represent at least one and the maximum of five differences in alleles, respectively.TM region-Triângulo Mineiro region.

Figure 3 .
Figure 3.Comparison of virulence factors among the four populations of Cryptococcus deuterogattii evaluated in the present study.The virulence factors evaluated are melanization (A), urease activity (B), hemolytic activity (C), phospholipases (D), gelatin hydrolysis (E), and proteases (F).Statistically significant differences (p < 0.05) are marked with (*), followed by the Kruskal-Wallis test and Dunn's test.The internal horizontal lines represent the median, the bars represent 25 ± 75% percentiles, and the horizontal lines represent the minimum and maximum.

Figure 3 .
Figure 3.Comparison of virulence factors among the four populations of Cryptococcus deuterogattii evaluated in the present study.The virulence factors evaluated are melanization (A), urease activity (B), hemolytic activity (C), phospholipases (D), gelatin hydrolysis (E), and proteases (F).Statistically significant differences (p < 0.05) are marked with (*), followed by the Kruskal-Wallis test and Dunn's test.The internal horizontal lines represent the median, the bars represent 25 ± 75% percentiles, and the horizontal lines represent the minimum and maximum.

Figure 4 .
Figure 4. Comparison of the antifungal susceptibility patterns (µg/mL) among 24 Cryptococcus deuterogattii evaluated in the present study via Eucast and CLSI methods.The drugs evaluated are amphotericin B via EUCAST method (A), amphotericin B via CLSI method (B), ketoconazole via EU-CAST method (C), ketoconazole via CLSI method (D), voriconazole via EUCAST method (E), voriconazole via CLSI method (F), itraconazole via EUCAST method (G), itraconazole via CLSI method (H), fluconazole via EUCAST method (I), and fluconazole via CLSI method (J).Statistically significant differences (p < 0.05) are marked with (*), followed by the Kruskal-Wallis test and Dunn's test.The internal horizontal lines represent the median; the bars are 25 ± 75% percentiles, and the horizontal lines are 10 ± 90% percentiles.CLSI: Clinical and Laboratory Standards Institute.EU-CAST: European Committee for Antimicrobial Susceptibility Testing.

Figure 4 .
Figure 4. Comparison of the antifungal susceptibility patterns (µg/mL) among 24 Cryptococcus deuterogattii evaluated in the present study via Eucast and CLSI methods.The drugs evaluated are amphotericin B via EUCAST method (A), amphotericin B via CLSI method (B), ketoconazole via EUCAST method (C), ketoconazole via CLSI method (D), voriconazole via EUCAST method (E), voriconazole via CLSI method (F), itraconazole via EUCAST method (G), itraconazole via CLSI method (H), fluconazole via EUCAST method (I), and fluconazole via CLSI method (J).Statistically significant differences (p < 0.05) are marked with (*), followed by the Kruskal-Wallis test and Dunn's test.The internal horizontal lines represent the median; the bars are 25 ± 75% percentiles, and the horizontal lines are 10 ± 90% percentiles.CLSI: Clinical and Laboratory Standards Institute.EUCAST: European Committee for Antimicrobial Susceptibility Testing.

Table 1 .
DNA polymorphisms according to populations, biological sources, and places where Cryptococcus deuterogattii isolates were recovered.
Legend: S-number of polymorphic sites; π-nucleotide diversity; K-average number of nucleotide differences; h-number of haplotypes; Hd-haplotype diversity; D-Tajima's D; FD-Fu and Li's D; FF-Fu and Li's F; FS-Fu's Fs; *-p < 0.05; Rm-minimum number of recombination events; Theta w-theta (per sequence) from S; Vtnr-variance of theta (no recombination); Vnfr-variance of theta (free recombination); θS-Watterson's estimate per sequence; #-four or more sequences are needed to compute the tests.The DNA polymorphism was evaluated, excluding sites with gaps.The repeated sequence types from different regions are not included in the total number.PHI-Pairwise Homoplasy Index; BH-city of Belo Horizonte; SP-city of São Paulo; TM-Triângulo Mineiro region; BAL-bronchoalveolar lavage; CSF-cerebrospinal fluid; @-there are too few informative characters to use the Phi Test; n-number of isolates.

Table 3 .
Differences in sequences between isolates G8 and G9 of Cryptococcus deuterogattii.