Characterization of Clinical and Environmental Isolates of Cryptococcus neoformans / Cryptococcus gattii Complex Maintained in Yeast Culture Collection in São Paulo , Brazil

Objective: As isolates of Cryptococcus are frequently kept collection stocks in institutions, sometimes without proper characterization, we sought to determine the genotype profiles, protease and phospholipase activities “in vitro” and the susceptibility testing for azoles and amphotericin B. Methodology: 84 isolates from several regions of Brazil (40 samples from clinical origin and 44 isolates from environmental origin) were maintained at the microorganism’s bank of the Biomedical Science Institute (ICB-USP) of the São Paulo University, in São Paulo, Brazil. This isolates was submitted fungal strains determination, DNA extraction and purification, determination of genotype by URA5 gene RFLP of CGB-positive isolates, protease and phospholipase activity and susceptibility to antifungals. Results: Of six CGB positive isolates tested by RFLP-PCR, only four presented a genomic profile consistent C. gattii species How to cite this paper: Cardoso, P.H.M., Da Silva, B.C.M., Dos Santos, J.I., Dos Santos, R.L.O., Leire-Jr, D.P., Auler, M.E., Ruiz, L.S., Da Silva, E.G., Moreira, D., Domaneschi, C., Baroni, F.A., Melhem, M.S.C., Martins, M.A. and Paula, C.R. (2018) Characterization of Clinical and Environmental Isolates of Cryptococcus neoformans/Cryptococcus gattii Complex Maintained in Yeast Culture Collection in São Paulo, Brazil. Open Journal of Epidemiology, 8, 76-92. https://doi.org/10.4236/ojepi.2018.82007 Received: April 8, 2018 Accepted: May 22, 2018 Published: May 25, 2018 P. H. M. Cardoso et al. DOI: 10.4236/ojepi.2018.82007 77 Open Journal of Epidemiology (VGII), while two other were C. neoformans (VNI and VNIII), indicating the existence of canavanine-resistante C. neoformans isolates in the culture collections. The clinical isolates secreted higher levels of phospholipase and environmental isolates but no differences were observed for the protease levels. Almost all isolates were sensible to azoles and amphotericin B. Conclusion: We point out in this research the existence of C. neoformans strains resistant to canavanine and intrinsic characteristic of C. gatti. These results demonstrate the importance to perform a detailed characterization of isolates kept in culture collections.


Introduction
Basidiomycetous yeasts of the genus Cryptococcus comprise seventy species, some of which are associated with invasive infections in humans and animals, such as Cryptococcus neoformans and Cryptococcus gattii which belong to the Cryptococcus neoformans/Cryptococcus gattii complex [1] [2] [3].C. neoformans comprises two varieties named variety grubii (serotype A) and variety neoformans (serotype D), while C. gattii has two serotypes (B and C).They have distinct, though occasionally overlapping, ecological niches, with C. neoformans often been isolated from pigeon (Columba livia) nests and droppings, and C. gattii often been isolated from decaying trees and soil [4] [5] [6] [7].
Considering that the biological characteristics of many C. neoformans complex stocks in Brazil are not generally tested in detail, and that some isolates are recorded with outdated taxonomic nomenclatures or even misidentified, they need to be constantly reevaluated.
pase activities and the susceptibility testing to azoles and amphotericin B, in clinical and environmental isolates of the C. neoformans/C.gattii complex, which are kept stored at the microorganism's bank of the Instituto de Ciências Biomédicas (ICB-USP) of the Universidade de São Paulo (USP), in São Paulo, Brazil.

Fungal Strains
This study was done on 84 isolates from several regions of Brazil.Of them, 40 samples were isolated from either cerebrospinal fluid or blood samples of human and animal cryptococcosis cases and 44 samples were isolated from environment bird droppings or the leaves of Eucalyptus.They were formerly identified as C. neoformans and were maintained in chloramphenicol-supplemented Sabouraud dextrose agar slants plus mineral oil, at the microorganism's bank at the Laboratory of Pathogenic Yeasts of the ICB-USP.Standard strains ICB 110-C.neoformans var.grubii, ICB134-C.neoformans var.neoformans, ICB 162-C.gattii and ICB 12A-Candida albicans were used in all tests.The isolates were identified by their urease activity, growth in L-canavanine glycine bromothymol blue (CGB) medium and assimilation of nitrate [3].The institutional review board of ICB-USP committee approved this study (Protocol 367/2010).
The mixtures were briefly vortexed and incubated at 56˚C for 2 hours or until the cell lysis was completed.DNA was extracted by phenol/chloroform/isoamyl alcohol (25:24:1) method and precipitated with isopropanol [29].The DNA pellet was washed with 70% ethanol, centrifuged for 10 min at 10,000 g and air-dried.The DNA was suspended in 200 μL of sterile ultra-pure water containing RNase 20 μg/mL, kept at 4˚C overnight and stored at −20˚C.DNA concentrations were determined in O.D. at 260 nm, while DNA purity was determined by the ratio of O.D. at 260 and 280 nm.For PCR amplification one μL of each DNA sample (85 -90 ng) was used as DNA template [29] [30].

Protease and Phospholipase Activity
Protease activity was performed in triplicate according to Rüchel et al. ( 1982) [31].Briefly, a culture medium dish containing sterile bovine albumin fraction V (Sigma-Aldrich, St. Louis, MO, USA) was distributed into Petri dishes.Each isolate was inoculated on the central surface of the plate and they were incubated at 32˚C in 5% CO 2 for 15 days and examined every two days until the 15th day.A translucent halo formed around the yeast colony indicated protease activity.Proteolytic activity (Pz value) was calculated as the ratio of the diameter of the colony (Dc) and to the diameter of the colony plus that of the degradation zone diameter (Dzd) (Pz = Dc/Dc + Dzd).Therefore, Pz = 1.00 indicated no activity (negative); Pz < 1.00 and > 0.64 indicated median activity (positive) and Pz < 0.64 indicated a strong activity (strongly positive) for protease activity.
Phospholipase activity was performed in triplicate by the egg yolk agar test [32].Briefly, the isolates were previously cultured during 48 hours Sabouraud Dextrose Agar Supplemented with chloramphenicol (Difco Laboratories Inc., Detroit, MI, USA) and then inoculated on the central surface of the phospholipase test plate and were incubated at 32˚C in 5% CO 2 for 15 days.An opaque halo of precipitation around the colony was indicative of phospholipase activity.The plates were examined for phospholipase activity every two days up to the 15th day.Phospholipase activity (Pz value) was calculated as the ratio of the diameter of the colony (Dc) and to the diameter of the colony plus that of the precipitation zone diameter (Pzd) (Pz = Dc/Dc + Pzd).Therefore, Pz = 1.00 indicated no activity (negative) strain; Pz < 1.00 and > 0.64 indicated median activity (positive) and Pz < 0.64 indicated a strong activity (strongly positive) for phospholipase activity.
Etest strips were applied to each Petri dish, and the plates were incubated at 35˚C in 5% CO 2 .The strip MICs were read at 48 and 72 hours of incubation.
The azole minimum inhibitory concentration (MIC) was read at the lowest concentration at which the border of the elliptical inhibition zone intercepted the scale on the strip.Any growth, such as microcolonies, throughout a discernible inhibition ellipse was ignored [34].The amphotericin B MICs were read at the point at which the zone of almost complete inhibition intersected the strip.All susceptibility tests were repeated twice.

Results
The isolates of clinical and environmental origin were re-identified and were consistent with the colonial and cellular morphology and physiological characteristics of C. neoformans complex, as proposed by Kurtzman et al. ( 2011) [3].The isolates were also cultured on CGB medium, for differentiation of C. neoformans and C. gattii.Of the 40 isolates of clinical origin, 36 (90.0%) were identified as C. neoformans and four (10.0%) as C. gattii.On the other hand, of the 44 isolates of environment origin, 42 (95.5%)were identified as C. neoformans and two isolates (4.5%) as C. gattii.

RFLP of URA5 Gene of CGB-Positive Isolates
Of the six isolates formerly identified as C. gattii four were classified as genotype VGII (serotype B and C) according to Figure 1, but two were identified as C. neoformans, one as genotype VNI (serotype A, var.grubii) and one as genotype VNIII (hybrid serotype AD) (Figure 2).Open Journal of Epidemiology

Protease and Phospholipase Activity
As shown in Table 1, of the 40 isolates of clinical origin nine (22.5%) were strongly positive for protease, 19 (47.5%) were positive and 12 (30.0%)were negative for protease.On the other hand, of the 44 isolates of environmental origin 20 (45.5%) were strongly positive for protease, 13 (29.5%)were positive for protease and 11 (25.0%) were negative for protease.No statistically differences were observed in the groups for protease.
Of the 44 isolates of environmental origin 24 (54.5%) were strongly positive and 20 (45.5%) were positive for phospholipase.The difference between the groups was statistically significant for phospholipase (P = 0.0003).

Susceptibility to Antifungals
Fluconazole-In isolates of clinical origin, the MIC values ranged from 0.0940 to 4.000 μg/mL, with a mean value of 0.8091 μg/mL.In isolates of environment origin, the values ranged from 0.0320 to 8.000 μg/mL, with a mean value of 0.6200 μg/mL.Itraconazole-In isolates of clinical origin, the MIC values ranged from 0.0080 to 0.2500 μg/mL, with a mean value of 0.0661 μg/mL.In isolates of environment origin, the values ranged from 0.0080 to 0.5000 μg/mL, with a mean value of 0.0388 μg/mL.Voriconazole-In isolates of clinical origin, the MIC values ranged from 0.0020 to 0.0320 μg/mL, with a mean value of 0.0065 μg/mL.In isolates of environment origin, the values ranged from 0.0020 to 0.0640 μg/mL, with a mean value of 0.0057 μg/mL.
Posaconazole-In isolates of clinical origin, the MIC values ranged from 0.0020 to 0.0940 μg/mL, with a mean value of 0.0214 μg/mL.In isolates of

Discussion
In this study all isolates of C. Traditionally, C. neoformans has been grouped in three varieties: C. neoformans var.gattii (serotype B and C), C. neoformans var.grubii (serotype A) and C. neoformans var.neoformans (serotype D and AD) [37].However, the gattii variety has been renamed as C. gattii [3] [38].According this nomenclature, all our strains of C. neoformans were retested from the biochemical tests and the six strains that were positive in the CGB medium were then presumptively reclassified as C. gattii [3].
In relation to the methods used in the phenotypic characterization, Klein et al.
(2009) [8] analyzed 102 strains of yeast to verify urease activity, melanin production and glycine assimilation in the CGB medium as well as sequencing.The 17 C. gattii strains tested were positive in the CGB medium, the 54 C. neoformans strains were negative in the CGB medium, one of which had weak staining activity in the medium, six of 20 other Cryptococcus species were also positive in the CGB medium, but negative for melanin production.This demonstrates that there is a limitation in the exclusive use of CGB culture for identification of Cryptococcus species and suggests that all isolates, clinical or environmental should be tested by genomic methods such as PCR-RFLP to confirm their real taxonomic identity [8].Indeed, in our study we observed that two strains of C. neoformans displayed urease activity which is not a common phenomenon.However, there are been some reports on urease activity by well characterized C. neoformans isolates [40] [41] [42].The possible explanation is that these strains suffered some kind of mutations and became canavanine-resistant [40] [41] [42].
There are several published works that describe the importance of different extracellular enzymes, such as urease, protease, phospholipase and phenoloxidase, among others, that allow the yeast to alter pH in the phagolysosome, degrades proteins of importance for the immune response of the host, favoring invasion [23] [43].In our study, most of the Cryptococcus strains, both of clinical origin and environmental origin, presented high protease values, although some of them, in the two groups, were not observed to produce significant amounts of this enzyme.This demonstrates the difficulty of correlating proteolytic activity observed by in vitro experiments with possible in vivo effects, for example, determining their actual contribution to disease development.
When the production of phospholipase was evaluated, it was observed that most of the isolates of clinical origin showed high production of the enzyme (92.5% were strongly positive and 2.5% positive), at a higher frequency than that observed in isolates of environmental origin (54.5% were strongly positive and 45.5% positive).Vidotto et al. (1996) [44] tested 23 isolates of C. neoformans for phospholipase production and found positivity in 95.7% (22) isolates, with a Pz ranged between 0.271 and 0.949.
According to the authors, there was a correlation between the phospholipase production and the size of the capsule in the isolated from AIDS patients.However, in this study, it was not possible to establish a relationship between capsular thickness and phospholipase production.In addition, genetic disruption studies have revealed the importance of the phospholipase B gene in yeast virulence in animal models, observing a decrease in the activity of this enzyme when the gene is altered [45] [46].Thus, phospholipase activity would be a more useful phenotypic marker to determine and evaluate the potential degree of virulence among isolates from clinical and environmental isolates of C. neoformans complex [45] [46] [47].
Extracellular phospholipases are a heterogeneous group of enzymes produced by several bacteria and fungi and have been implicated in their pathogenicity because they can cause damage to host cell membranes that result in the destabilization of membranes, cell lysis and release of lipid second messengers [48] [49] [50].Studies of Candida albicans demonstrated that phospholipase activity was correlated with mucosal invasion [51] and increased mortality in mice [52].Also, production of phospholipases A, B, C and D by Aspergillus fumigatus has been reported [52] [53].
In relation to ecological aspects of Cryptococcus several studies suggest that species of the C. neoformans/C.gattii complex are interrelated epidemiologically and ecologically [5] [54]- [59].C. neoformans is found in most cryptococcal infections worldwide, mainly in patients underlying immunocompromised condi-tion as AIDS patients and presents the pigeon droppings and soil of as the main reservoir.In our study, the majority of isolates of environmental origin used came from pigeon droppings and other birds, where the most frequent species is C. neoformans [6] [59] [60] [61] [62].Although C. neoformans and C. gattii occupy similar ecological niches, C. gattii is mainly associated with deteriorated wood, plant remains and soil.In particular, eucalyptus trees represent the most common substrate for C. gattii [62]- [67], and we could confirm this, as our samples isolated from Eucalyptus leaves belong to the C. gattii species.
Cryptococcal infections in humans usually are treated with amphotericin B, associated or not with 5-flucytosine, almost followed by maintenance therapy with some azole antifungal, mainly fluconazole.High rates of cryptococcosis recidiva suggest a potential emergence of antifungal resistance, clinical or otherwise, among C. neoformans isolates [68] [69] [70].In view of this situation, it is necessary to routinely determine the sensitivity profile of C. neoformans isolates to different antifungal agents and, in addition, to allow an evaluation of the efficacy of the treatments used and the evolution of the disease.
In this present study, we used disk diffusion tests, the commercial Etest kit that has been described as a good method for routine laboratory use, proving to have a good alternative for in vitro sensitivity tests for azoles [71] [72] [73].This method allows the determination of MICs, and has shown excellent benefits, such as easy execution and quick results [74].It also demonstrates good correlation with the standard method by CLSI [35] which is more labor-intensive and a long reading time.Although CLSI methods are not considered ideal for testing C. neoformans [75] [76], broth microdilution and agar diffusion method have been applied worldwide to study the antifungal sensitivities of C. neoformans isolates.
Both clinical and environmental isolates were highly sensible to the antifungals amphotericin B, posaconazole and voriconazole.Several studies have reported that amphotericin B as well as new triazoles (voriconazole and posaconazole) has potent activity against C. neoformans, where more than 99% of the isolates show MICs ≤ 1.0000 μg/mL, regardless of geographic origin of isolates [77] [78] [79], suggering that this new triazoles may bealternative treatements when similar isolates are finded in clinical practice [79] [80].
A limitation of this study was that we did not genotype all samples nor could test the samples for their susceptibility to antifungals by the CLSI protocol, due to financial constraints.On the other hand, we were able to better characterize beyond a simple taxonomic identification the isolates of the Cryptococcus kept as stock at microrganism's bank of the University of São Paulo.We also could demonstrate the occurrence of variations on the samples such as more phospholipase activity in samples collected from clinical cases than in samples from environment, the presence of two canavanine-resistant C. neoformans strains and none of the samples were resistant to any antifungals tested.Therefore, the importance of maintaining a well characterized collection of Cryptococcus is that one can be better assess their characteristics as new laboratory tools become available.

Conclusion
We point out in this research the existence of C. neoformans strains resistant to canavanine and intrinsic characteristic of C. gatti.The other important results obtained were the higher levels of phospholipase by clinical strains than environmental strains but no differences were observed for the protease levels.Selected phenotypic characteristics of C. neoformans and C. gatti complex must be compared to molecular studies to perform a correct diagnosis to the species.
neoformans were retested in relation to biochemical tests and the six isolates, four from clinical origin and two from environmental origin, which were positive in CGB-positive, were submitted to molecular typing by RFLP-PCR.Only four presented a genomic profile consistent with the C. gattii species (VGII), while two other were, in fact, C. neoformans (VNI and VNIII).

Table 1 .
Production of protease and phospholipase by the isolates.