Given the high intrinsic antifungal resistance in Sporothrix, studies correlating antifungal susceptibility and genetic diversity among etiological agents of sporotrichosis are overdue. Here we used haplotype networks to estimate genetic diversity among clinical Sporothrix species. This approach has the strength of enabling the recognition of genetically deviating strains in a population, rather than randomly sampled frequent haplotypes with similar genetic background, as represented by the frequent haplotypes H1 and H8 in the S. brasiliensis and S. schenckii populations, respectively. However, haplotype-based strategies are dependent on the molecular marker used to estimate diversity, the set of strains comprising the database as well as the network methods . To date, the partial sequence of CAL and ITS region (ITS1/2 + 5.8 s) has been used to estimate genetic diversity and taxonomy of clinical Sporothrix species [2, 5, 10, 11]. Indeed, due to its highly discriminatory power and polymorphisms we choose these regions to give insights into the genetic diversity in sympatric populations of S. brasiliensis and S. schenckii s. str.
A genetically homogeneous population may be expected to have similar susceptibility profiles among the individuals, but increased fitness during population expansion can result in a diversity of susceptibility profiles. Alternatively, differential responses may emerge over time in ancestral genetically diverse species. Of the phylogenetically related species S. brasiliensis and S. schenckii s. str., the former has a lower genetic diversity [2, 3, 10, 13], and here we found that it also had a correspondingly lower variability of in vitro susceptibility. A larger degree of genetic diversity has been observed among S. schenckii s. str. isolates, which is accompanied by greater variation in antifungal profiles  virulence  and genomic organization . Moreover, the presence of median vectors (mv1-3, 5, 6) in the haplotype network (Figure 2) may indicate extinct or intermediate unsampled haplotypes within the S. schenckii s. str. populations, possibly suggesting the existence of even more variation. Conversely, the absence of median vectors in the S. brasiliensis population strongly suggests that our analysis detected most of the strain diversity circulating in the human epidemic, thus supporting the low variation of the MIC and MFC values in this species.
Itraconazole is the drug of choice for treating endemic mycoses caused by thermodimorphic fungi, including sporotrichosis. Our findings regarding the in vitro activities of the main antifungal drugs are in agreement with previous studies, indicating that ITC and PCZ were moderately effective against isolates of S. brasiliensis and S. schenckii s. str. [14, 32]. Despite the relatively high values found in all studies, several authors have demonstrated the efficacy of ITC in clinical outcome. Barros et al.  studied the clinical treatment of a large number of patients in Rio de Janeiro, and reported the success of ITC treatment (50–400 mg/day) in 94.6% patients (n = 610). Although this study did not molecularly identify the phylogenetic species involved, it was most likely S. brasiliensis, as this species is highly prevalent in Rio de Janeiro due to feline sporotrichosis outbreaks [1–3, 10]. Such an epidemiological profile could explain the favorable results for ITC. The relatively low MICs found for PCZ also suggest this molecule to be a promising drug in the treatment of sporotrichosis caused by S. schenckii s. str. or S. brasiliensis. Additionally, PCZ also showed low MICs against 2 out of 3 clinical isolates of the rare species S. mexicana, including CBS 132928 (MIC = 1 μg/mL) and CBS 132927 (MIC = 2 μg/mL). Only a few strains of S. mexicana have been described in the literature [1, 10], and the species has been reported as tolerant to most commercially available drugs [14, 33].
Resistance to amphotericin B has been reported in emerging pathogens, such as Aspergillus terreus, Candida lusitaniae, Fusarium spp., Scedosporium prolificans, and Trichosporon asahii [34, 35]. Here we found high MICs for AMB in five isolates of S. schenckii s. str. (Ss17, Ss22, Ss110, and Ss119 from H8, and isolate Ss51 from H12) with MIC and MFC values of >16 μg/mL. The multidrug resistant phenotype found for isolate Ss51 (H12; Figure 2B, D, and F) is particularly noteworthy in relation to its clinical origin, where it showed moderate virulence in a murine model . The high MICs found for the remaining compounds (FLC, VRC, CAS, and 5FC) are in agreement with previous reports [14, 32, 36].
Fungistatic drugs are capable of inhibiting the cell growth and reproduction of fungi without destroying them. Drug-sensitive pathogens may evolve resistance under the selective pressure imposed by continuous exposure to fungistatic drugs . The molecular mechanism that lies behind the recent emergence of drug-resistant phenotype among Sporothrix species is currently unknown. However, judging from other fungi, the increased and prolonged use of triazoles has raised concerns about resistant infections by Cryptococcus neoformans , Candida albicans  and Aspergillus fumigatus . Azole resistance may occur through a diversity of mechanisms including the upregulation of multidrug transporter genes that leads to enhanced efflux of azoles and therefore reduce drug accumulation ; multiple genetic alterations of the target enzyme that can affect the affinity of the enzyme and therefore prevents azole binding , and alteration of metabolism, usually sterol synthesis . More recently, genetic studies correlating antifungal resistance and alteration in chromosome copy number and genomics architecture has delivered an exciting view on the molecular mechanisms governing the increased fitness to the phenotype of azole-resistance [38, 44–46].
Candida albicans [39, 44, 45] and Cryptococcus neoformans [38, 46, 47] triazole-resistant isolates very often contain an abnormal number of chromosomes. Such genomic plasticity may offer additional copies of drug resistance genes leading chromosomal aneuploidies isolates to overcoming the drug effects [44, 45] and allowing rapid adaptive evolution . Although this phenomenon may occur spontaneously , drug-resistant aneuploidies arise very frequently within drug-exposed fungal populations and this may support our recent findings on chromosomal polymorphisms in Sporothrix species . The intra-specific polymorphisms in chromosome number and size in S. schenckii s. str. suggested that aneuploidy occur among clinical isolates, and it is tempting to hypothesize that this phenomenon could, in part, be responsible for the differences in drug profiles observed here. Therefore, the high karyotype diversity observed for S. schenckii s. str.  is reflected in the genetic diversity found in our haplotype network (Figure 2). However, testing for this pattern requires a larger number of isolates with dissimilar antifungal profiles as well as chromosomal polymorphism.
Fungicidal drugs may be defined as those that lead to a reduction of 99.9% of the initial inocula . It is difficult to make comparisons between published MFCs studies, since most were performed before the introduction of clinical species beyond S. schenckii s. str. to the genus Sporothrix. Two or more of these newer species were likely involved in the studies of McGinnis et al.  and Silveira et al. . In all clinical Sporothrix spp. evaluated, we found significant differences between the minimum concentrations needed to inhibit fungal cell growth and the concentration required for colony count reduction. In all studied species, most isolates were able to grow after 72 h in the presence of each drug at the maximum concentration tested, which is in agreement with results obtained by Trilles et al. . The moderately low MICs for triazoles imply a great resistance to killing among isolates of S. brasiliensis, the most pathogenic among Sporothrix species.
Spontaneous cure, as well as relapse, are common features of sporotrichosis. Our present data show that most antifungal agents had only a fungistatic effect against clinical isolates. It remains unclear whether there is a correlation between in vitro MIC/MFC data and clinical outcome in human sporotrichosis. Experimental murine models have shown the efficacy of posaconazole (5 mg/kg) for treating S. brasiliensis and S. schenckii s. str. infections . Voriconazole (40 mg/kg) has been demonstrated to only reduce fungal load in mice infected with S. schenckii s. str., and to have no activity against S. brasiliensis . No studies to date have linked the dissimilar phylogenetic species to in vitro and in vivo data. Our in vitro results for the highly tolerant strain CBS 133020 (=Ss265, haplotype 7), an isolate originated from a disseminated case of S. brasiliensis in an HIV patient , shown that despite ITC and AMB MICs of 1 and 4 μg/mL, respectively, and MFCs of >16 and 8 μg/mL, respectively, the patient showed positive clinical outcome after administration of intravenous AMB (including 10 days on L-AMB) , indicating a need to monitor MFCs values before and during treatment.