A total of 84 soil samples were collected from 14 administrative regions. At each sampling site, triplicates were obtained and placed in sterilized polyethylene bags. Then, 1ml of each soil suspension (1:10 w/v dilutions) was cultured on the selective medium Dichloran Rose-Bengal Chloramphenicol Agar (DRBC) (Becton & Dickinson). The DRBC plates were stored at 25°C and 37°C, and observed every two days. When typical features of Scedosporium species were observed, the colonies were subcultured in 2% malt extract agar (MEA), potato dextrose agar (PDA) and Sabouraud agar (SA), at 25°C and 37±1°C in the dark. The strains studied in the present work were isolated from soil collected in urban areas from several Chilean cities: one from Arica (North of Chile); two from Valparaiso (central region of Chile) and the last from Puerto Varas (South of Chile).

Morphological features were determined on PDA at 25±1°C (7 days). Cultures were deposited at the Chilean Fungal Collection (ChFC), Mycology Unit, University of Chile as S. dehoogii ChFC 148, 151, 153, and 161. Microscopic features were studied in lactophenol stained with cotton blue (Merck) using Optika B1000Ph microscope (Optika, Italy). Microphotographs were taken with Optikam B9 camera (Optika, Italy).

Fungal DNA was extracted directly from pure 7 day-old cultures by means of the E.Z.N.A® Fungal DNA Mini Kit (Omega Biotek Store, USA). PCR was conducted with a 3Prime Thermal Cycler (Techne, UK) using ITS1 and ITS4 primers for ITS1-5.8 S-ITS2 region36; Cal1 and Cal2 primers for the calmodulin gene,28 and Tub2F and Tub2R primers for one region within the β-tubulin gene.8 A 25μl-mixture containing 17μl of BioMix™ Red (New England Biolabs), 10pM of the primers, and 10ng DNA was amplified following a protocol of initial denaturation at 94°C for 5min, 35 cycles of denaturation at 94°C for 1min, annealing at 52°C for 1min for ITS region, and 55°C for CAL and TUB; extension at 72°C for 2min, and final extension at 72°C for 7min. The PCR product was checked on 1% agarose gel. Purification was performed with FavorPrep™ Gel/PCR purification Mini Kit (Favorgene, Taiwan) following the manufacturer guidelines. Sequencing was done on Macrogen (Macrogen, Korea). An NCBI Blastn search for each locus region was performed.2 Sequences from type and reference strains were gotten from GenBank database. The sequences were aligned in MUSCLE software11 followed by manual adjustments with a text editor.

Phylogenetic analyses were performed using Maximum Parsimony (MP) method in PAUP 4.0b10 software. For MP inference the most-parsimonious tree (MPT) was obtained after one hundred heuristic searches with random sequence addition and tree bisection-reconnection branch-swapping algorithms, collapsing zero-length branches and saving all minimal-length trees (MulTrees). Support for internal branches was assessed using a heuristic parsimony search of 1000 bootstrapped data sets. The combined datasets of the ITS, CAL, and TUB were tested for incongruence with the partition homogeneity test and implemented in PAUP 4.0b10 software. Lomentospora prolificans FMR 7294 and Petriellopsis africana CBS 311.72 were used as outgroup.

In vitro susceptibility testing was performed using broth microdilution for filamentous fungi according to CLSI document M38-A2.5 The MICs and MECs (minimal effective concentration) were read at 48 and 72h. Although the clinical breakpoints for Scedosporium species have not been developed, the following suggested cut-off breakpoints were used: susceptible (S), ≤1μg/ml; intermediate (I), 2μg/ml; and resistant (R), ≥4μg/ml (CLSI, M38 A-2). Pure active powders, of known potency, of amphotericin B (AMB) (Sigma–Aldrich), voriconazole (VRC) (Sigma–Aldrich), itraconazole (ITC) (Sigma–Aldrich), posaconazole (PSC) (Sigma–Aldrich), caspofungin (CSF) (Sigma–Aldrich), micafungin (MCF) (Astellas Pharma), anidulafungin (ANF) (Pfizer), α-pinene (α-P) (partner laboratory), geraniol (GER) (partner laboratory), clioquinol (CLIO) (Sigma–Aldrich), and 5,7-dibromo-8-hydroxyquinoline (5–7C) (Sigma–Aldrich) were used. All tests were performed in duplicates. Candida parapsilosis ATCC 22019 and Paecilomyces variotii ATCC MYA 3630 were included as quality controls.

Several genera and species were isolated from soil samples (i.e., Aspergillus, Cunninghamella). In the same way, other Scedosporium and related species such as S. boydii, S. apiospermum, and L. prolificans were isolated on DRBC plates (unpublished data).

In the first approach, all Scedosporium isolates were identified based on the morphological and physiological features. The final identification for each species recovered from DRBC plates was obtained by molecular analyses of tubulin gene sequences.

In general, the four strains studied displayed the typical features described by Gilgado et al.15 The colonies on PDA reached 28mm diameter after 7 days at 25±0.2°C, they were dense, cottony to lanose, grayish-white to white, and with a whitish irregular margin; the reverse was yellowish or light-brown. Solitary conidiogenous cells were subhyaline smooth-walled, mainly cylindrical, 10–54μm length by 1–1.4μm width, and produced pale brown, obovoid or ellipsoidal 5.9–6.2μm×3.6–4.4μm conidia. Sessile conidia were subhyaline, thin-walled, mostly obovate (5.9−)7.3(−7.8)×(3.5−)3.8(−4.0)μm. Hyphae, 1.2–2.5μm, were hyaline to subhyaline (Fig. 1). A teleomorph was not observed even after 60 days of incubation.

Fig. 1.

(A–D) Scedosporium dehoogii, ChFC 151. (A) Colony growing on PDA after 14 days at 25°C. (B–D) Conidiogenous cells and conidia. (E, F) Conidia.

(0.28MB).

The growth rates were similar in all media tested. The minimum, maximum and optimal temperatures were 15°C, 40°C and 36±1°C respectively.

The identification of all the strains studied as S. dehoogii was feasible by means of the three loci used in the present study. The partition homogeneity test demonstrated that the three loci sequence data sets were congruent (p=0.05) and could therefore be combined. A fragment of 1698bp was obtained in the concatenated analysis. The phylogeny inference was analyzed as separate (data not shown) and combined dataset.

The tree topologies of concatenated loci by MP were highly similar to those observed in the individual tree for each gene tested. High clade support values were obtained in the concatenated analysis, being mainly over 90%. A total of 90 MPT were obtained from heuristic search from concatenated dataset (Fig. 2). The trees had a consistency index of 0.81, a retention index of 0.89, and a homoplasy index of 0.18. Analyses of the combined partitions allowed the identification and recognition of 8 species of Scedosporium (Fig. 2). In the same way, the separation of L. prolificans from the Scedosporium clade was observed. The four strains isolated from Chilean soil samples were placed at the S. dehoogii clade, with the reference and type strains sequences.

Fig. 2.

One of the 90 most-parsimonious trees obtained from heuristic searches based on the analysis of the combined data set (ITS+CAL+TUB). Bootstrap support values of 100% are indicated at the nodes. CI, consistency index; RI, retention index; HI, homoplasy index. Chilean strains are indicated with boldface type. T=type strains.

(0.53MB).

The sequences generated in this study were deposited in the GenBank database under the following accession numbers: KT600015 (ChFC148 ITS), KT600016 (ChFC151 ITS), KT600017 (ChFC153 ITS), KT600018 (ChFC161 ITS), KT820715 (ChFC148 TUB), KT820716 (ChFC151 TUB), KT820717 (ChFC153 TUB), KT820718 (ChFC161 TUB), KT820719 (ChFC148 CAL), KT820720 (ChFC151 CAL), KT820721 (ChFC153 CAL), KT820722 (ChFC161 CAL).

The results of the antifungal susceptibility tests for the Scedosporium strains are shown in Table 1. Among the classical antifungal agents, voriconazole, itraconazole and posaconazole were the most active. Echinocandins were the less active against all the strains. Among the putative new antifungal molecules, geraniol and α-pinene were not active, obtaining high MIC values. On the other hand, clioquinol, and 5,7-dibromo-8-hydroxyquinoline showed promising results with low MICs values.