Introduction

Cyphellophora was introduced by de Vries (1962) with C. laciniata as type species. The genus is characterised by having a melanized thallus with phialidic conidiogenesis, phialide openings being inserted directly on hyphae or occasionally on flask-shaped conidiogenous cells, and producing small clusters of olivaceous, septate, mostly curved conidia. The species Cyphellophora laciniata, as well as the later described C. pluriseptata (de Vries et al. 1986) were reported as agents of superficial skin or nail infections in humans. Later, several new species, i.e. C. vermispora, C. taiwanensis, C. indica, C. guyanesis and C. eucalypti, isolated from plant materials, were added to the genus (Walz and de Hoog 1987; Matsushima 1987; Jacob and Bhat 2000; Decock et al. 2003; Cheewangkoon et al. 2009).

Several researchers noted relationships of Cyphellophora with Pseudomicrodochium and Phialophora (Decock et al. 2003; de Hoog et al. 2009, 2011). Pseudomicrodochium is defined by P. aciculare which has hyaline thallus and white to pale salmon colonies (Sutton 1975). Species with dark colonies, such as P. fusarioides and P. suttonii share phialidic conidiogenesis, relatively simple conidiogenous cells and septate conidia with Cyphellophora. Taxonomic affinities between the genera were also inferred from analysis of partial nuclear ribosomal DNA SSU (18S) sequence data (Decock et al. 2003). Cyphellophora laciniata, C. vermispora, and C. guyanensis formed a well-supported clade with P. fusarioides and P. suttonii. The authors therefore concluded that the Cyphellophora and Pseudomicrodochium species were congeneric. Pseudomicrodochium fusarioides and P. suttonii were transferred to Cyphellophora (Decock et al. 2003).

SSU phylogeny (de Hoog et al. 2011) also demonstrated Cyphellophora to be related to some species of Phialophora, even though species of the latter genus produced single-celled phialoconidia rather than multiseptate ones. Phialophora europaea and P. reptans clustered with C. laciniata at 94 % bootstrap support, forming a ‘europaea-clade’ within the order Chaetothyriales (de Hoog et al. 2011) containing species predominantly causing superficial infections in humans. However, the type species of Phialophora, P. verrucosa, was positioned outside this lineage, located in a ‘carrionii-clade’ in the same order (de Hoog et al. 2011) containing primarily agents of human chromoblastomycosis. While the backbone of the Chaetothyriales remained largely unresolved, it was evident that P. europaea, P. reptans and P. verrucosa were not members of the same clade.

Members of the europaea-clade containing Cyphellophora and some Phialophora species are frequently recovered from clinical samples, and their prevalence may have been underestimated. In a comprehensive investigation of the frequency of black yeast-like fungi in dermatological specimens in Denmark, P. europaea, Knufia epidermidis (Li et al. 2008; Tsuneda et al. 2011), and Exophiala lecanii-corni were encountered repeatedly (Saunte et al. 2012). In 108 cultures of black yeasts from this study, Phialophora europaea accounted for 26.9 %. The clinical manifestations of P. europaea infections vary between hyperkeratosis (de Hoog et al. 2000), and asymptomatic colonization (Eckhard et al. 2007; Saunte et al. 2012). The same holds true for C. laciniata, C. pluriseptata and C. suttonii (de Vries 1962; de Vries et al. 1986; Sutton et al. 1991). Recently, many of these species were found in showers and sinks in bathrooms, and in washing machines (Lian and de Hoog 2010; Hamada and Abe 2010). Black yeast-like organisms were found to compose a consistent part of the flora of indoor wet cells. Lian and de Hoog (2010) supposed that human skin softened during bathing might be more prone to infection by these fungi. However, no satisfactory agreement on the infection process has been achieved.

The taxonomy of Cyphellophora and its relatives has not been sufficiently studied. In the present article, the genus and its relatives in Phialophora are analysed in a mutilocus study using the internal transcribed spacer region (ITS), the DNA dependent RNA polymerase II largest subunit (RPB1), the partial beta tubulin gene (BT2), and the nuclear large subunit rDNA gene (LSU). These data were compared with morphology and physiology.

Materials and methods

Fungal strains

Strains used in this study were obtained from the Centraalbureau voor Schimmelcultures Fungal Biodiversity Centre (CBS) (Table 1), which included all available ex-type strains of described species, some having been sent upon request, and supplemented with fresh isolates. Stock cultures were maintained on slants of oatmeal agar (OA, Difco) and potato dextrose agar (PDA, Difco). All cultures in this study are maintained in the culture collection of the CBS (Utrecht, the Netherlands). Another 71 isolates of C. eucalypti, P. europaea and P. reptans which are also maintained in the CBS are not listed here; a summary of all strains analysed with their sources of isolation is provided in Fig. 1. Taxonomic information for new species was deposited in MycoBank (www.MycoBank.org).

Table 1 Strains used in phylogenetic analysis of Cyphellophora and relatives
Full size table
Fig. 1
figure 1

Sources information of Cyphellophora and related Phialophora species maintained in the CBS. The total number of isolates included was 101. The number on top of each column indicated the numbers of sources

Full size image

Morphology

Strains were cultured on OA and PDA and incubated at 24 °C in the dark for three weeks. Microscopic observations were based on slide culture techniques using PDA and OA because these media readily induce sporulation and suppress growth of aerial hyphae. Agar blocks of ~1 cm2 were placed on a sterile object glass supported by a V-shaped glass bar and inoculated at the four sides. The block was subsequently covered with a sterile cover slip (~2 cm2). Growth was allowed in a closed glass Petri dish; the bottom was covered with sterile paper filter soaked with 5 ml sterile water to ovoid drying of the culture. The chambers were incubated at room temperature for 7, 14, 21 or 28 d. Slides were made in lactic acid or lactophenol cotton blue. Permanent slides were sealed with polyvinyl alcohol. Micrographs were taken using a Nikon Eclipse 80i microscope and DS Camera Head DS-Fi1/DS-5 m/DS-2Mv/DS-2MBW using NIS-Element freeware package (Nikon Europe, Badhoevedorp, the Netherlands). Dimensions were determined with the Nikon Eclipse 80i measurement module on slides in lactic acid and the mean and standard deviation were calculated from measurements of 50 conidia (Crous et al. 2009a).

Physiology

Cardinal growth temperatures of strains were determined on 2 % malt extract agar (MEA, Difco). Plates were incubated in the dark for 3 weeks at temperatures ranging at 3 °C intervals from 21–36 °C. In addition, growth was recorded at 37 and 40 °C. In order to evaluate whether high temperature was fungicidal or not, all cultures were returned back to 25 °C and incubated for one week more. Experiments consisted of three simultaneous replicates for each isolate; averages of three measurements were calculated. ABTS-agar medium (Kaluskar et al. 1999) containing 0.1 %, 0.3 % and 0.5 % ABTS [2, 2-azino-bis (3-ethylbenzothiazoline-6-sulphonic acid) diammonium salt] was used for assay laccase activity. Each plate was inoculated with one agar plug (8 mm diam) obtained from the edge of an actively growing mycelium of each test fungus and incubated at 24 °C in the dark for 7 d. A positive laccase reaction was indicated by a green zone around the colony. Candida albicans SC 5314 served as negative control, and Cryptococcus neoformans CBS 7229 as positive control. The width of the zone was considered to be directly related to the amount of extracellular laccase produced. Urease activity and cycloheximide sensitivity tests were also investigated with the appropriate test media. The entire procedures were repeated once.

DNA extraction

Approximately 1 cm2 mycelium was transferred to a 2 ml Eppendorf tube containing 490 μl 10 % cetyltrimethylammonium bromide (CTAB) buffer and 6–10 acid washed glass beads (diam 1.5–2 mm, Sigma). 10 μl proteinase K was added and vortexed for 10 min, then the mixture was incubated at 60 °C for 60 min. Subsequently 500 μl chloroform : isoamylalcohol (24 : 1) was added to the solution and was shortly mixed by shaking, incubated for 30 min on ice water and centrifuged for 10 min at 14,000 r.p.m. The supernatant was transferred to a new tube with 225 μl 5 M NH4-acetate, incubated on ice water for 30 min and centrifuged again for 10 min at 14,000 r.p.m. The supernatant was transferred to another Eppendorf tube with 0.55 vol isopropanol mixed carefully by flipping and spun for 5 min at 14,000 r.p.m. The pellet was washed with ice cold 70 % ethanol and after drying at room temperature it was re-suspended in 50 μl TE buffer (de Hoog et al. 1999). For some strains, DNA was prepared with Ultra Clean Microbial DNA Isolation Kit (MoBio, Carlsbad, U.S.A.) according to the manufacturer’s instructions. DNA quality was verified by NanoDrop®ND-1000 Spectrophotometer using ND-1000 v. 3.3.0 software (Coleman Technologies, Wilmington, DE, U.S.A.). DNA extracts were stored at −20 °C prior to use.

PCR amplification, cloning and sequencing

Four genes were amplified: the internal transcribed spacer region (ITS), the DNA dependent RNA polymerase II largest subunit (RPB1), the partial β-tubulin gene (BT2), and the nuclear large subunit rDNA gene (LSU). Primers used for amplification and sequencing were shown in Table 2. PCR reactions were performed on a GeneAmp PCR System 9700 (Applied Biosystems, Foster City, CA, U.S.A.) in 25 μl volumes containing 10 ng of template DNA, 2.5 μl reaction buffer (0.1 M Tris-HCl, pH 8.0, 0.5 M KCl, 15 mM MgCl2, 0.1 % gelatin, 1 % Triton X-100), 0.1 mM of each dNTP and 1.0 U Taq DNA polymerase (ITK Diagnostics, Leiden, The Netherlands). Amplifications for ITS, BT2 and LSU were performed as follows: 95 °C for 5 min, followed by 35 cycles consisting of 95 °C for 45 s, 52 °C for 45 s and 60 °C for 1 min and post elongation step at 72 °C for 7 min. PCR amplifications for RPB1 were performed at 5 cycles of 30 s denaturation at 94 °C, followed by primer annealing for 30 s at 55 °C, and extension for 1 min at 72 °C; followed by 5 cycles with an annealing temperature at 53 °C and 30 cycles at 51 °C, finalized with an extension for final 10 min at 72 °C. Amplicons were purified using GFX PCR DNA and gel band purification kit (GE Healthcare, Buckinghamshire, U.K.). The resulting PCR products of three C. suttonii-like species were cloned into JM109 competent cells with the pGEM-T Vector System II cloning kits (Promega, Leiden, the Netherlands) following the protocols. Cycle sequencing was performed as follows: 95 °C for 1 min, followed by 30 cycles of 95 °C for 10 s, 50 °C for 5 s and 60 °C for 2 min. Reactions were purified with Sephadex G-50 fine (GE Healthcare, Uppsala, Sweden) and sequencing was done on an ABI 3730XL automatic sequencer (Applied Biosystems). Sequence data obtained in this study were adjusted using the SeqMan module (DNAStar, Madison, WI, U.S.A.).

Table 2 Primers used in this study for amplification and sequencing
Full size table

Alignment and phylogenetic reconstruction

To assess the phylogenetic position of Cyphellophora, phylogenetic analyses of LSU was performed for 98 taxa representing the order Chaetothyriales. The alignments provided by Gueidan et al. (2008) and de Hoog et al. (2011) were taken as a basis for this study. Alignment was performed automatically and adjusted iteratively by hand with BioNumerics v. 4.61 (Applied Maths, Kortrijk, Belgium). DnaSP v. 5.01 (Librado and Rozas 2009) and MEGA v. 5.05 (Tamura et al. 2011) were used, respectively, to estimate DNA polymorphisms within and among species and to estimate DNA divergence between species Phylogenetic analyses were conducted in MEGA v. 5.05 (Tamura et al. 2011), PhyML v.3.0 (Guindon and Gascuel 2003) and MrBayes v.3.1.2 (Ronquist and Huelsenbeck 2003).To address the phylogenetic relationships among taxa, Maximum Likelihood (ML), Neighbour Joining (NJ), Maximum Parsimony (MP) and Bayesian Inference (BI) were used. The most appropriate model of DNA substitution was selected by MEGA v. 5.05 (Tamura et al. 2011) (Table 3). Nodal support was assessed by bootstrap analysis from 100 replicates. Bootstrap values equal or higher than 70 % were considered significant. The ML tree was obtained using the PhyML program (Guindon and Gascuel, 2003) with the General Time Reversible (GTR) model by the Akaike information criterion (AIC) using the MEGA v. 5.05 Models test (Tamura et al. 2011). Eight rate categories were taken into account in ML estimates of base frequencies. The NJ tree was derived from the TN93+G model, which corresponds to the Tamura-Nei model with a rate variation among sites modulated by the gamma parameter. The MP tree was constructed with PAUP 4.0b 10 (Swofford 2002) under constraint of simple addition of sequences and tree bisection and reconnection (TBR) branch swapping. All characters were unordered and of equal weight and gaps were treated as missing data. Branches of zero length were collapsed and all multiple, equally parsimonious trees were saved. Bayesian analysis based on Markov Monte Carlo chain approach was run as implemented using MrBayes v.3.1.2 (Ronquist and Huelsenbeck 2003). Using the model identified by Modeltest and flat priors, Four MCMC chains were run for 1 × 106 generations, and trees were sampled every 100 generations. The first 1000 trees were discarded as part of the burn-in period. Posterior probabilities for the Bayesian approach were determined by calculating a 50 % majority rule consensus tree from the remaining 9000 trees. Sequences CBS 101157 and AF 050277 representing the LSU gene of Sarcinomyces petricola and Phaeococcomyces catenatus, respectively, were used to root the LSU tree (de Hoog et al. 2011).

Table 3 Details of each analysis of the data sets used for generating Figs. 2, 3 and 4
Full size table

Thirty strains were selected, and their aligned sequences of four genes, LSU, ITS, RPB1 and BT2, were concatenated by FASTA alignment joiner of Fabox (http://www.birc.au.dk/fabox/ ). Topological conflicts were evaluated visually and by using the Partition Homogeneity Test (PHT) implemented in PAUP v. 4.0b10 (Swofford 2002). Multilocus trees were constructed with neighbour-joining and with 500 bootstrap replicates using MEGA v. 5.05 software (Tamura et al. 2011). Bootstrap values equal or higher than 80 % were considered significant. Cladophialophora immunda, CBS 834.96 was taken as outgroup. The phylogenetic trees were edited using Tree View v. 1.6.6 (Page 1996).

Results

Molecular phylogeny

Irrespective of the algorithm (NJ, MP, ML and BI) employed, all phylogenetic reconstructions showed similar congruent topologies that agreed on clustering and branching patterns. Since analyses produced almost identical results not changing any of the interpretations, only the phylogenetic NJ consensus trees are displayed. As shown in Fig. 2, a general tree was constructed of a large set of representatives of the order Chaetothyriales using LSU data. The LSU alignment had base frequences f (A) = 0.250, f (T) = 0.244, f (C) = 0.214, f (G) = 0.292. The NJ tree of LSU was calculated with MEGA v. 5.05 with Tamura-Nei and Gamma distribution. The ancestral lineages, with 96 % bootstrap support, contained prevalently rock-inhabiting fungi. The derived part of the tree matching with the family Herpotrichiellaceae, at rather large distance and containing the genera Capronia, Cladophialophora, Cyphellophora, Exophiala, Fonsecaea, and Phialophora had 99 % bootstrap support.

Fig. 2
figure 2

Phylogeny of a representative selection of species in Chaetothyriales, obtained from a NJ analysis based on LSU rDNA sequences. Bootstrap support was calculated from 100 replicates; values ≥70 % are shown above/below the branches; supported branches were drawn in bold. Coloured boxes represent species complexes taken from de Hoog et al. (2011) and Voglmayr et al. (2011). (T) means ex-type strain of the species or one of its synonyms

Full size image

The Herpotrichiellaceae clade contained 34 bootstrap-supported branches, indicated in bold in Fig. 2. The backbone of the tree was poorly resolved. When comparing the SSU groups previously recognized in the Herpotrichiellaceae by de Hoog et al. (2011), four highly supported clades were found in both, namely the bantiana-, carrionii-, and europaea-clades and a clade centred around rock-inhibiting fungi, while the dermatitidis-, jeanselmei- and salmonis-clades (de Hoog et al. 2011) had a slightly different topology. Three clades of ant-associated fungi introduced by Voglmayr et al. (2011) remained separate with high bootstrap support in our LSU phylogeny. Most of the clades were morphologically heterogeneous, each containing different anamorph genera. One rock-inhabiting fungus and one ant-associated fungus were paraphyletic to the europaea-clade, which was also observed in SSU phylogenies (Gueidan et al. 2008; Voglmayr et al. 2011).

The majority of described Cyphellophora species (C. laciniata, C. vermispora, C. eucalypti, C. pluriseptata, C. suttonii, and C. fusarioides) clustered in the europaea-clade (Fig. 2), together with Phialophora europaea, P. reptans and P. oxyspora. This clade was analysed in more detail using multilocus data composed of LSU, ITS and RPB1 (Fig. 3) and BT2 sequences (Fig. 4). Cyphellophora hylomeconis and C. eugeniae, two plant-associated species, were excluded since they were shown to be phylogenetically distinct from the members of the europaea-clade. Unfortunately, there was no culture available of the species C. indica and C. taiwanensis, impeding any sound phylogenetic inference of the relationships with Cyphellophora or other genera of Chaetothyriales. Cladophialophora immunda CBS 834.96 was selected as outgroup for both trees (Figs. 3 and 4).

Fig. 3
figure 3

Neighbour-joining tree based on combined data set of partial LSU, ITS and RPB1 sequences showing the relationship among members of Cyphellophora and related Phialophora species. Bootstrap support was calculated from 500 replicates; values ≥80 % are shown above/below the branches in bold. Five major groups, group A–E, of the europaea-clade are shown in boxes. General sources information are marked in small color boxes: red, clinical; green, environmental. (T) means ex-type strain of the species or one of its synonyms

Full size image
Fig. 4
figure 4

Consensus trees of Cyphellophora and relatives together with clones of the three strains (CBS 449.91, CBS 112.94 and CBS 284.85) in group A based on BT2, constructed with MEGA5 and 500 bootstrap replicates; values ≥80 % are shown above/below the branches in bold. Cladophialophora immunda, CBS 834.96 was taken as outgroup. (T) means ex-type strain of the species or one of its synonyms

Full size image

PHT based on heuristic search of four genes (LSU, ITS, BT2 and RPB1), revealed BT2 conflict with the other three genes (P = 0.01), while no conflict of signal was observed in LSU-ITS-RPB1 combination (P = 0.29). The LSU-ITS-RPB1 matrix contained 530 parsimony-informative characters of 2081 total characters. In the combined LSU-ITS-RPB1 tree eleven terminal branches were statistically supported with bootstrap values >80 % (Fig. 3). Groups B, D and E were clearly resolved at high bootstrap values, including P. reptans, P. oxyspora and P. europaea, respectively. A number of clusters were united as group A by a high bootstrap value (100 %), including C. laciniata, C. vermispora, C. suttonii and C. fusarioides; its nearest neighbor was group B with 100 % bootstrap support, which was clearly beyond the species borderline. Group A contained six resolved subclades (A1–A6). Group C, recognized at 100 % bootstrap support, consisted of four subclades (C1–C4). Groups A4–6 as well as group C2 and C3 were single isolates (CBS 449.91, CBS 112.94, CBS 284.85, CBS 285.85 and CBS 235.93, respectively). The topology of BT2 tree was congruent with the combined LSU-ITS-RPB1 tree. Repeated cloning for all genes was performed of the three phylogenetically isolated strains (CBS 449.91, CBS 112.94 and CBS 284.85) in group A to exclude possible intra-strain diversity. Clones matched in all cases with direct sequencing results, and each of the clouds of clones and PCR-products had 100 % bootstrap support. Data for BT2 are shown in Fig. 4. Dxy, the average number of nucleotide substitutions per site between two populations, and Da (the number of net nucleotide substitutions per site between populations), are measures of nucleotide diversity. The range of Dxy among groups A4–6 was 0.178–0.265, and Da was 0.138–0.260. Dxy between group A4 and the other two groups was relatively large. The nucleotide diversity for all 45 BT2 sequences analysed was 0.183, while this was 0.171 in groups A4–6. Of the 412 positions analysed, 5 haplotypes were identified in group A4, 3 for group A5 and 2 for group A6, respectively.

Morphology

Members of group A shared (1−) 3–5 (−8)-septate conidia produced in bundles from poorly differentiated conidiogeneous cells with unclear phialidic collarettes, sometimes resembling annellated zones. Degrees of conidiation varied greatly. Members of groups A2, A4 and A6 had slightly curved to sickle-shaped conidia. In groups A4 and A6 the conidia were mostly produced from sessile conidiogenous loci on intercalary cells, whereas strains in group A2 produced distinct, lateral or terminal phialides with swollen basal parts. Group B was characterised by sessile collarettes and one-celled conidia which may become larger germinating cells in a later stage of development; torulose hyphae were often present in the centre of the culture. Members of group C had sickle-shaped, (1−) 3 (−5)-septate conidia with darkened basal scars produced from clear phialides with dark collarettes. Strains in group C4 had longer and more slender, multiseptate conidia. Group C3 (CBS 235.93) had differently shaped phialides, some reduced to sessile collarettes, with heads of ellipsoidal, one-celled conidia having darkened basal scars, and was considered to represent a new species. Members of group D were clearly different by light brownish, straight, tapering phialides with distinct, tubular or funnel-shaped collarettes producing one-celled, bactriform conidia in chains; 1-septate conidia were occasionally observed. Group E had variously shaped phialides or adelophialides with heads of spherical to broadly ellipsoidal, one-celled conidia. Collarettes were small but clearly distinct.

Physiology

The cardinal growth temperature test showed that all cultures obtained in this study had their optimal development at 21–27 °C, with growth abilities ranging between 21–36 °C (Fig. 5). No growth was observed at 40 °C. For the type strains of C. plurisepta, C. suttonii, P. europaea and P. oxyspora, the optimum growth temperature on MEA was 27 °C, with a maximum of 33 °C, respectively. However, most species in group A and C are unable to grow at 33 °C and above (Fig. 5). After 3 weeks, most cultures incubated above 33 °C showed no growth when returned back to 25 °C, demonstrating that high temperature was fungicidal. Cultures of P. oxyspora at 36 °C and 37 °C started to grow again.

Fig. 5
figure 5

Colony diameters of Cyphellophora and relatives at various temperatures ranging from 21 to 40 °C, measured after 3 weeks on 2 % MEA. (T) means ex-type strain of the species or one of its synonyms

Full size image

Totally 13 strains of Cyphellophora and relatives were cultured on ABTS-agar medium. Three concentrations of ABTS-agar, viz. 0.1 %, 0.3 % and 0.5 %, respectively, revealed that isolates were found to secrete laccase, the colourless ABTS-agar medium turning green due to the oxidation of ABTS to ABTS-azine in the presence of laccase (Fig. 6). The positive reactions were unambiguous even after 3 d of incubation on 0.1 % ABTS-agar medium (data not shown), which proved to be the best indicator for evaluating the production of laccase and fungal biomass. The higher concentration of ABTS in the medium, the stronger and faster the reaction of the green zone was seen, but the growth of the colony was inhibited (Fig. 6). All strains were found to be positive for laccase, though with different halo sizes (Table 4). With the same concentrations, potential biomass production, as measured from growth on standard medium (MEA) was not correlated with intensity of the reaction, represented by the size of the halo. All tested isolates were urease positive and were inhibited by 0.1 % cycloheximide.

Fig. 6
figure 6

Laccase activity assay. Green zones around the colonies were measured from 7-day-old cultures at 25 °C on ABTS-agar medium. a–c. Growth of isolate Cyphellophora laciniata, CBS 174.79 on 0.1 %, 0.3 % and 0.5 % ABTS-agar medium. d–f. Growth of isolate Cyphellophora pluriseptata, CBS 286.85 on 0.1 %, 0.3 % and 0.5 % ABTS-agar medium

Full size image
Table 4 Laccase activity and other physiological tests of Cyphellophora and Phialophora species in the europaea-clade
Full size table

Discussion

Based on the sequences of ITS, RPB1, BT2 and LSU genes (Fig. 2, 3 and 4), five major groups of the europaea-clade could be distinguished, which were supported by high bootstrap values, i.e. group A, which contained the strains of C. laciniata, C. vermispora, C. suttonii, C. pauciseptata nob. and C. fusarioides; group B, which contained the strains of P. reptans; group C, which contained the strains of C. pluriseptata, P. ambigua nob., C. guyanesis and C. eucalypti; group D, which contained the strains of P. oxyspora; and group E, which contained the strains of P. europaea. Groups A and C were morphologically variable, while the others were morphologically and phylogenetically consistent. The topology of these two variable groups was similar with each marker, but species delimitation remained difficult, especially to the C. suttonii-like isolates CBS 449.91, CBS 112.94 and CBS 284.85 (group A4–6), which were located at single branches in each marker. Intraspecies and even intra-individual variation of the ribosomal operon has been identified for some fungal species (O’Donnell and Cigelnik 1997, Pawlowska and Taylor 2004). To understand the possible existence of intragenomic variation in these three C. suttonii-like species, we PCR-amplified and cloned the ITS, RPB1, BT2 and LSU regions. The results showed that each strain had nearly uniform clones for each marker, and clones formed populations separate from neighbouring species. It was therefore concluded that the three C. suttonii-like species represented separate entities. In order to take taxonomic decisions, more strains of these ambiguous groups are needed.

The genus Cyphellophora is morphologically characterised by forming intercalary or terminal phialides bearing flaring, thin or outstanding collarettes; in some cases the collarettes are more deeply pigmented than the body of the phialides. This resembles the conidiogenous cells of Phialophora, but the former have larger, fusiform to sigmoid, multiseptate conidia, whereas Phialophora has one-celled conidia. In accordance with current practice in black yeast taxonomy we maintain anamorph genera as ecotypes, which are reflected in their morphology. The genus Phialophora is phylogenetically defined to the extent that species with with phialophora-type morphology but unrelated to the order Chaetothyriales are assigned to other genera (Gams 2000). Within the order, however, clades are too unstable to serve as a basis for generic taxonomy, and therefore morphology is maintained as a defining criterion. The type species of Phialophora, P. verrucosa, is member of the carrionii-clade, whereas that of Cyphellophora, C. laciniata, is located in the europaea-clade (de Hoog et al. 2011). In summary, Phialophora is reserved for phialidic species in Chaetothyriales with one-celled conidia, whereas conidia of Cyphellophora are multiseptate and curved.

In the present study, we investigated the genetic diversity and species delimitation of Cyphellophora. The phylogenetic relationship of Cyphellophora and Phialophora, and their relationships with the other black yeast-like fungi were inferred from a NJ analysis based on partial LSU sequence data of 98 taxa. In main traits our results are in agreement with published phylogenies of Chaetothyriales (e.g. de Hoog et al. 2011; Haase et al. 1999; Voglmayr et al. 2011). Compared to SSU data (de Hoog et al. 2011; Haase et al. 1999), the backbone of the tree is still poorly resolved, and a number of groups is difficult to delimit. In contrast, the europaea-clade is clearly recognizable in all studies. Most members of this clade cause superficial infections in humans. This type of infection otherwise is relatively uncommon in the Chaetothyriales. It has been found in Cladophialophora boppii (Badali et al. 2008; Brasch et al. 2011; Saunte et al. 2012) [the species has also been reported in a subcutaneous infection (Pereira et al. 2010) but this probably concerned a Neoscytalidium species] in the carrionii-clade, in some Cladophialophora species in the bantiana-clade (Badali et al. 2009), while also waterborne Exophiala species were repeatedly derived from human skin (de Hoog et al. 2011). Also in the ancestral clade there are some superficial species, such as Knufia epidermidis (Li et al. 2008; Tsuneda et al. 2011). Most members of Chaetothyriales cause cutaneous, subcutaneous (mycetoma and chromoblastomycosis), systemic or disseminated infections.

The plant-associated Cyphellophora species, C. eugeniea (Crous et al. 2007) and C. hylomeconis (Crous et al. 2009b) are unrelated with the europaea-clade. The europaea-clade contains some plant-inhabiting species, but true pathology has not been proven. Cyphellophora guyanesis was isolated from leaf spots of Eucalyptus (as C. eucalypti) (Cheewangkoon et al. 2009; Decock et al. 2003), but also from dead plant debris. One strain CBS 235.93 was found in paraphyletic position to C. guyanesis, but was isolated from human nail without precise clinical information. Morphologically, this clinical strain was entirely different from the plant inhabiting C. guyanesis, having Phialophora-type conidia. Hence we regard this as a new species, which is described below. Cyphellophora vermispora was originally isolated from roots of grasses, but the same species was also found on human skin (Fig. 1). Possibly, this Cyphellophora species resides on skin and nail as a saprobe derived from soil particles. At the other end of the spectrum, Phialophora europaea was almost exclusively recovered from clinical samples, mostly from skin and nails (Fig. 1). The only environmental strains of P. europaea were encountered in bathrooms and washing machines (Lian and de Hoog 2010, Hamada and Abe 2010). It seems that indoor wet cells are preferred environmental niches of P. europaea, where the species lives as an oligotrophic fungus with low competitive ability.

Eckhard et al. (2007) showed relatively high numbers of black yeasts, among which was P. europaea, co-occurring with dermatophytes affecting the skin of diabetic patients. The group of diabetic patients investigated used public swimming facilities and was involved in intensive sport activities relatively often. Black fungi and dermatophytes are likely to contaminate human skin that is softened during bathing. It is hypothesized that especially diabetic patients are vulnerable because they have high glucose concentrations in skin and low blood circulation in the extremities. This may hold true for more species in the europaea-clade, such as C. laciniata.

Other putative virulence factors for black yeast-like fungi are the ability to produce melanin pigments and to tolerate extreme temperatures. The production of melanin was shown to be involved in pathogenicity of many fungi including Chaetothyriales, because of its role in protection against oxygen radicals and sustained the growth of fungi at 37 °C infection temperature (Jacobson 2000, Liu et al. 2004). In many organisms, laccase is associated with melanin production. In the present study, all strains of Cyphellophora and relatives had a positive reaction for laccase when using ABTS as laccase secretion indicator. Tolerance of human body temperature is an essential requirement for pathogenicity. Species able to grow at 40 °C may cause systemic infections; those able to grow at 37 °C may cause chromoblastomycosis or mycetoma, while those unable to grow at 37 °C may cause superficial or mild cutaneous infections (Badali et al. 2008). Our results showed that Cyphellophora and relatives had their optimal development at 21–27 °C, and that part of them grew well at epidermal temperatures below 37 °C, which agreed with the clinical observation that members of the europaea-clade mainly cause superficial infections. Judging from the above, species of the europaea-clade are likely to have a low degree of virulence. Surprisingly, P. suttonii, CBS 449.91 was derived from an infection of a dog’s ear (Ajello et al. 1980), while CBS 112.94 and CBS 130291 were isolated from human sputum and bronchial washing, respectively (Sutton et al. 1991). Well-documented clinical cases to develop a balanced view on the clinical significance of members of the europaea-clade are insufficiently available.

Taxonomy of Cyphellophora and Phialophora species in the europaea-clade

Cyphellophora fusarioides (B. Sutton & C.K. Campbell) DeCock, in DeCock, Delgado, Buchet & Seng, Antonie van Leeuwenhoek 84: 213. 2003–MycoBank MB487997. Fig. 7.

Fig. 7
figure 7

Cyphellophora fusarioides (CBS 130291). a–b. Colony on OA, obverse and reverse. c–d. Flask-shaped phialides with distinctive collarettes. e–k. Intercalary conidiogenous cells with sessile collarettes and septate conidia. l. Hyphae. m. Slightly curved, septate conidia. Scale bars = 5 μm

Full size image

≡ Pseudomicrochium fusarioides B. Sutton & C.K. Campbell, in Sutton, Campbell & Goldschmied-Reouven, Mycopathologia 114: 160. 1991.

Typus

Israel, from bronchial lavage fluid of human patient, ex-type culture CBS 130291=MUCL 44033=IHEM 6073=IMI 339361=NCPF 2733.

Description of CBS 130291 after one month incubation on OA, 24 °C: Colonies growing slowly, olivaceous black in the centre, pale olivaceous grey towards the sharply defined, continuous margin. Reverse olivaceous black. No diffusible pigment produced. Hyphae hyaline to slightly pigmented, septate, 1.5–2.5 μm wide. Distinct conidiophores absent. Phialides intercalary, with conspicuous, subhyaline to pale brown, flaring collarettes of 2.5–3.5 μm deep. Conidia hyaline, l (−2)-septate, (3.5–) 5.0–6.0(–8.5) × 1.0–1.5 μm, smooth-walled, straight to slightly fusiform, adhering in small bundles. Chlamydospores absent. Teleomorph unknown. Cardinal temperatures: minimum 21 °C, optimum 24 °C, maximum 33 °C.

Notes

Cyphellophora fusarioides resembles C. suttonii in having curved or attenuated, multiseptate conidia. The species are similar in the absence of discrete conidiogenous cells: in both species the conidiogenous cells are reduced to undifferentiated, intercalary compartments with the conidiogenous loci inserted on lateral protuberances (adelophialides). CBS 130291 was isolated from the bronchial washings of a human suspected to have a mycotic pulmonary infection, but the pathology remained unproven (Sutton et al. 1991). The species was originally described in Pseudomicrodochium, a genus typified by P. aciculare. This and the remaining species of Pseudomicrodochium (P. lauri, P. cylindicum and P. candidum) all have septate, cylindrical conidia with more obtuse apices. The main morphological difference between Cyphellophora and Pseudomicrodochium is found in the melanization of the thallus, Pseudomicrodochium species being hyaline. Their phylogenetic position has as yet not been determined.

Cyphellophora guyanensis Decock & G. Delgado, in Decock, Delgado-Rodríguez, Buchet & Seng, Antonie van Leeuwenhoek 84: 210. 2003–MycoBank MB488295. Fig. 8.

Fig. 8
figure 8

Cyphellophora guyanensis (CBS 124764). a–b. Colony on OA, obverse and reverse. c–d. Intercalary conidiogenous cells with sessile collarettes. e–i. Phialides with swollen basal parts and slightly darkened collarettes. j. Flexuose hyphae. k–l. Fusiform, septate conidia. Scale bars = 5 μm

Full size image

= Cyphellophora eucalypti Cheewangkoon & Crous, Persoonia 23: 63. 2009b.

Typus

French Guiana, from dead leaf of angiosperm, ex-type culture MUCL 43737.

Description of CBS 124764 after one month incubation on OA, 24 °C: Colonies growing moderately rapidly, pale grey-brown in the centre, covered with loose, cottony mycelium. Margin entire, flat, dark brown. Reverse olivaceous black. No diffusible pigment produced. Hyphae thin- to slightly thick-walled, smooth-walled, loosely septate, hyaline to pale brown, 1.5–2.5 μm wide. Conidiophores absent or reduced to a short cell supporting the conidiogenous cells. Phialides discrete, ampulliform to flask-shaped, mostly arising at right angles from undifferentiated hyphae, occasionally terminal, widest in the lower third or in the middle, subhyaline to pale olivaceous brown, thin-walled, each with a conspicuous funnel-shaped collarette which is sometimes darker than the body of the phialides. Conidia hyaline to pale brown, (4.0–) 8.0–11(–15) × 1.5–2.0 μm, smooth-walled, nearly straight to slightly sigmoid, with somewhat truncate bases and rounded apices, aggregating in bundles. Chlamydospores absent. Teleomorph unknown. Cardinal temperatures: minimum 21 °C, optimum 27 °C, maximum 33 °C.

Notes

Cyphellophora guyanensis, MUCL 47737 was isolated from a dead leaf of an undetermined angiosperm, while Cyphellophora eucalypti, CBS 124764 was isolated from leaf spots of Eucalyptus. Both species share dark colonies and large, flared collarettes on well-developed phialides, and septate conidia. In the ITS phylogenetic tree (data not shown), Cyphellophora guyanensis, MUCL 47737 and C. eucalypti, CBS 124764 are joined in a very well-supported subclade (bootstrap 93 %); no live material of C. guyanensis was available for study. Cyphellophora eucalypti is here reduced to synonymy of C. guyanensis.

Cyphellophora laciniata G.A. de Vries, Mycopath. Mycol. Appl. 16: 47. 1962–MycoBank MB329520. Fig. 9.

Fig. 9
figure 9

Cyphellophora laciniata (CBS 190.61). a–b. Colony on OA, obverse and reverse. c–f. Degenerate hypha-like conidia. g–i. Curved, sickle-shaped, septate conidia. Scale bars = 5 μm

Full size image

Typus

Basel, Switzerland, from skin scales of human patient, ex-type culture CBS 190.61=ATCC 14166=MUCL 9569.

Description modified according to de Vries (1962): Colonies restricted, woolly-velvety in various shades of grey. Reverse olivaceous black. No diffusible pigment produced. Fertile hyphae pale brown, 1.5–3.0 μm wide, sometimes with constrictions at the septa. Phialides intercalary, sometimes on short side branches, each with a short, lateral or terminal collarette. Conidiophores absent. Conidia curved, brown, smooth-walled, 1–3-septate, 11–25 × 2.0–5.0 μm, adhering in small bundles. Chlamydospores absent. Teleomorph unknown. Cardinal temperatures: minimum 21 °C, optimum 21 °C, maximum 27 °C.

Notes

The authentic culture of de Vries (1962), CBS 190.61 was analysed, but this strain had become degenerate. At observation on OA, conidia were hypha-like without proper production process from phialidic collarettes (Fig. 9). The most striking morphological character of this species should be the curved, sickle-shaped, septate conidia produced in bundles from sessile phialide openings. Phylogenetically C. laciniata is close to C. vermispora, the latter having longer, multiseptate conidia.

Cyphellophora pauciseptata P. Feng & de Hoog, sp. nov.–MycoBank MB564874. Fig. 10.

Fig. 10
figure 10

Cyphellophora pauciseptata (CBS 284.85). a–b. Colony on OA, obverse and reverse. c–f. Conidial apparatus with septate conidia. g–j. Intercalary conidiogenous cells with sessile collarettes. k–m. Hypha-like conidia. n. Hyphal coil. o–p. Curved, septate conidia. Scale bars = 5 μm

Full size image

Typus

Roermond, the Netherlands, from human hand skin, holotype CBS-H 20941, ex-type culture CBS 284.85.

Etymology

The name refers to the thin conidial septation which is difficult to ascertain with light microscopy.

Description of CBS 284.85 after one month incubation on OA, 24 °C: Colonies restricted, olivaceous black. Reverse olivaceous black. No diffusible pigment produced. Hyphae smooth- and moderately thin-walled, 0.8–1.5 μm wide. Phialides intercalary, collarettes sessile on undifferentiated hyphae, short cylindrical, up to 3.0 μm long. Conidia hyaline, smooth- and thin-walled, mostly curved, (12–) 14–16(–20) × 0.8–1.0 μm, with several very thin septa. Chlamydospores absent. Teleomorph unknown. Cardinal temperatures: minimum 21 °C, optimum 21 °C, maximum 27 °C.

Notes

Morphologically, C. pauciseptata is similar to C. vermispora, both species having long, septate conidia, but it lacks flask-shape phialides, conidia being produced on sessile collarettes. Originally, strain CBS 284.85 was identified as C. suttonii (ex-type strain CBS 449.91), which is morphologically similar, but the two species differ 7.8 % and 23.9 % in ITS and RPB1 sequences, respectively.

Cyphellophora pluriseptata G.A. de Vries, Elders & Luykx, Antonie van Leeuwenhoek 52: 141. 1986.–MycoBank MB103253. Fig. 11.

Fig. 11
figure 11

Cyphellophora pluriseptata (CBS 286.85). a–b. Colony on OA, obverse and reverse. c–e. Conidial apparatus with conidia in head, phialides with swollen basal parts and slightly darkened collarettes. f. Erect phialides with conidia. g. Flexulose hyphae. h. Cylindrical to fusiform conidia with rounded ends and conspicuous scars. Scale bars = 5 μm

Full size image

Typus

Roermond, the Netherlands, from foot skin of human patient, ex-type culture CBS 286.85=ATCC 62672.

Description of CBS 286.85 after one month incubation on OA, 24 °C: Colonies growing slowly, olivaceous black in the centre, with velvety, dark grey defined edge. Reverse olivaceous black. No diffusible pigment produced. Fertile hyphae pale brown, 1.5–2.5 μm wide. Phialides usually terminal or lateral, occasionally intercalary, pale brown, up to 24.0 μm long, 1.0–3.0 μm wide, basal part swollen, with a conspicuous collarette at the tip. Conidia pale brown, (1.0–) 3.5–5.5(–8.0) × 1.0–2.5 μm, cylindrical to fusiform, with rounded ends, and truncate at the base. Chlamydospores absent. Teleomorph unknown. Cardinal temperatures: minimum 21 °C, optimum 24 °C, maximum 33 °C.

Notes

Cyphellophora pluriseptata differs from C. laciniata in its more differentiated phialides which are only rarely reduced to sessile collarettes, and its generally straight conidia with more cross septation. There is considerable sequence diversity, the species differing 14.2 % in ITS. The species is also close to C. guyanensis, but this has longer, multiseptate conidia, mostly aggregating in bundles and takes a clearly distinct position in all trees.

CBS 285.85 (Fig. 12) differs from Cyphellophora pluriseptata in its tapering phialides and generally straight, generally one-septa conidia; whether or not multiseptate conidia were produced in a later stage of development, as in C. pluriseptata (Fig. 11), could not be ascertained. In the ITS-RPB1 tree, the strain clustered with C. pluriseptata with 96 % bootstrap support, but differed significantly at 5.1 % and 13.5 % ITS and RPB1 sequence diversity, respectively, which would indicate a separate species status for CBS 285.85. However, the source of isolation of strains CBS 284.85, 285.85 and 286.85 is remarkably similar: a single hospital in the city of Roermond, the Netherlands, during the same year but from skin and nails of different patients (de Vries et al. 1986). Three species have been encountered in a single location, and were never found again except once C. pluriseptata in Germany (Table 1). This sheds doubt on the suitability of used genetic markers to differentiate species of Cyphellophora, for which reason we are reluctant to introduce a novel species for CBS 285.85. Considering the polymorphism in Cyphellophora and the difficulties to delimit species in this group, we temporarily refer to CBS 285.85 as Cyphellophora cf. pluriseptata.

Fig. 12
figure 12

Cyphellophora cf. pluriseptata (CBS 285.85). a–b. Colony on OA, obverse and reverse. c–d. Phialides with slightly darkened collarettes. e–j. Erect, tapering phialides with conidia. k–l. Hyphal coils with cylindrical conidia. m. Septate conidia with rounded ends and visible scars. Scale bars = 5 μm

Full size image

Cyphellophora suttonii (Ajello, A.A. Padhye & M. Payne) DeCock, in Decock, Delgado, Buchet & Seng, Antonie van Leeuwenhoek 84: 213. 2003–MycoBank MB487354. Fig. 13.

Fig. 13
figure 13

Cyphellophora suttonii (CBS 449.91). a–b. Colony on OA, obverse and reverse. c. Intercalary conidiogenous cell with single conidium. d. Lateral conidiogenous cell with bundled conidia. e–i. Conidial apparatus with single to clustered, septate conidia. j–k. Anastomosing conidia. l–m. Straight to falcate conidia. Scale bars = 5 μm

Full size image

≡ Pseudomicrodochium suttonii Ajello, A.A. Padhye & M. Payne, Mycotaxon 12: 133. 1980.

Typus

U.S.A., from subcutaneous lesion of dog ear, ex-type culture CBS 449.91=CDC B-2963=IMI 233463.

Description of CBS 449.91 after one month incubation on OA, 24 °C: Colonies restricted, grey to olivaceous black, somewhat moist. Reverse olivaceous black. No diffusible pigment produced. Hyphae 1.5–2.0 μm wide, brownish. Conidiophores absent. Conidiogenous cells intercalary or lateral, then often without basal septum, cylindrical, with an indistinct collarette (adelophialides), producing subsequent conidia in more or less sympodial order. Conidia brown, smooth-walled, straight to falcate, acicular, (10–) 15–20(–30) × 1.0–1.2 μm, 3–8-septate. Chlamydospores absent. Teleomorph unknown. Cardinal temperatures: minimum 21 °C, optimum 27 °C, maximum 33 °C.

Notes

Cyphellophora suttonii was originally isolated from a phaeohyphomycosis of the ear of a dog. A human case of C. suttonii infection was in a female patient receiving long-term prednisone therapy for sarcoidosis and presented with an ulcerative weeping lesions on her right leg (Perfect and Schell 1996). The culture from biopsied tissue was not sequenced and was not available for study. Nunes et al. (1999) reported an isolate from Brazilian soil, but the identity of that isolate was not confirmed either.

Cyphellophora vermispora A. Walz & de Hoog, Antonie van Leeuwenhoek 53: 143. 1987–MycoBank MB133484. Fig. 14.

Fig. 14
figure 14

Cyphellophora vermispora (CBS 228.86). a–b. Colony on OA, obverse and reverse. c–g. Ampulliform to flask-shaped phialides with single or bundled conidia. h–j. Conidial bundles on conidiogenous cells. k. Hyphal coil with conidiogenous cell. l. Slightly curved conidia. Scale bars = 5 μm

Full size image

Typus

Monheim, Germany, from wheat field, ex-type culture CBS 228.86.

Description of CBS 228.86 after one month incubation on OA, 24 °C: Colonies restricted, olivaceous black, with mouse grey aerial mycelium in the outer zone; margin straight and sharp. Reverse olivaceous black. No diffusible pigment produced. Hyphae smooth- and moderately thin-walled, 1.5–2.0 μm wide. Phialides discrete, pale olivaceous brown, arising at right angles from undifferentiated hyphae, solitary or in small groups, hyaline, thin-walled, ampulliform to flask-shaped, 5.5–6.5 μm long, widest at or below the middle; collarettes unclear, hyaline, in a later stage becoming annellide-like; collarettes occasionally sessile on intercalary hyphal cells. Conidia hyaline, smooth- and thin-walled, vermiform, slightly to clearly curved, (6.5–) 8.0–11(–13) × 1.0–1.5 μm, with several very thin septa. Chlamydospores absent. Teleomorph unknown. Cardinal temperatures: minimum 21 °C, optimum 24 °C, maximum 30 °C.

Notes

The conidia in our cultures of ex-type strain CBS 228.86 are significantly shorter and have less septa than described in the original publication of Walz and de Hoog (1987). The species is easily distinguished morphologically from other Cyphellophora species by its pronounced, flask-shaped phialides. Its nearest phylogenetic neighbor is C. laciniata, at 0.91 % ITS divergence.

Phialophora ambigua P. Feng & de Hoog, sp. nov.–MycoBank MB564880. Fig. 15.

Fig. 15
figure 15

Phialophora ambigua (CBS 235.93). a–b. Colony on OA, obverse and reverse. c–f. Sessile, slightly darkened collarettes. g–k. Different types of conidial apparatus with sessile collarettes and conidial heads. l–m. Irregular, septate hyphae. n. One-celled conidia with truncate bases. Scale bars = 5 μm

Full size image

Typus

the Netherlands, from human toenail, holotype CBS-H 20942, ex-type culture CBS 235.93.

Etymology

The name refers to the morphology of the fungus combining typical features of P. europaea, P. reptans, and P. oxyspora.

Description of CBS 235.93 after one month incubation on OA, 24 °C: Colonies restricted, olivaceous black. Reverse olivaceous black. No diffusible pigment produced. Hyphae slightly pigmented, irregularly septate, 0.8–1.5 μm wide. Phialides intercalary or terminal, (3.0−) 3.5–6.0 (−6.5) × 1.0–1.5 μm, flask-shaped to elongate. Collarettes very short, flaring, somewhat darkened, sometimes sessile on undifferentiated hyphae, producing conidia in heads. Conidia subhyaline, smooth-walled, 1.0–2.0 × 0.8–1.0 μm, obovoidal, with a broadly truncate base. Chlamydospores absent. Teleomorph unknown. Cardinal temperatures: minimum 21 °C, optimum 21 °C, maximum 30 °C.

Notes

On the basis of sequence data, CBS 235.93 was first identified as C. guyanesis (formerly C. eucalypti). It is the only strain representing P. ambigua, and was isolated from a clinical specimen. Phylogenetically, this strain deviated from C. guyanesis at 3.35 % ITS diversity, and this is supported by large differences in morphology. Its flask-shaped phialides are similar to those of P. europaea, the sessile collarettes resemble those of P. reptans, and the truncate, obovoidal conidia are similar to those of P. oxyspora.

Phialophora europaea de Hoog, Mayser & Haase, Mycoses 43: 414. 2000–MycoBank MB482897. Fig. 16.

Fig. 16
figure 16

Phialophora europaea (CBS 101466). a–b. Colony on OA, obverse and reverse. c. Spirally twisted hyphae with flask-shaped phialides. d. Intercally or terminal, cylindrical phialides. e–i. Different types of conidial apparatus with short collarettes. j, k. Tolurose hyphae and conidia. l. Conidia. Scale bars = 5 μm

Full size image

Typus

Dordrecht, the Netherlands, from foot of human patient, coll. H.M.E. Frénay, ex-type culture CBS 101466.

Description of CBS 101466 after one month incubation on OA, 24 °C: Colonies developing slowly, with olivaceous black, somewhat slimy centre, olivaceouse grey towards the flat marginal zone. Reverse olivaceous black. No diffusible pigment produced. Hyphae pale olivaceous brown, 1.5–2.0 μm wide, regularly septate about every 10–20 μm. Fertile hyphae bearing phialides either directly, or with one to two on slightly swollen subtending cells of about 2.0–3.0 μm wide. Phialides flask-shaped to elongate, somewhat narrowed towards the tip, often with a nearly cylindrical apical portion, 6.0–9.0 μm long. Collarettes very short, flaring, somewhat darkened, producing conidia in fragile chains or in heads. Conidia subhyaline (sub) spherical, 1.0–2.5 × 1.0–1.5 μm. Chlamydospores absent. Teleomorph unknown. Cardinal temperatures: minimum 21 °C, optimum 24 °C, maximum 33 °C.

Notes

Isolates of Phialophora europaea originate almost exclusively from mild cutaneous infections of humans, mostly from skin and nails. Recently, P. europaea was found co-occurring relatively often with dermatophytes affecting the skin of diabetic patients who used public swimming facilities and were involved in intensive sport activities (Eckhard et al. 2007). The only environmental strains of P. europaea were encountered in indoor wet cells, such as bathrooms (Lian and de Hoog 2010) and washing machines (Hamada and Abe 2010).

Phialophora oxyspora W. Gams, in Gams & Holubová-Jechová, Stud. Mycol. 13: 64. 1976.–MycoBank MB319853. Fig. 17.

Fig. 17
figure 17

Phialophora oxyspora (CBS 698.73). a–b. Colony on OA, obverse and reverse. c–d. Flask-shaped phialides with collarettes and conidial heads. e–g. Hyphal coils with flask-shaped phialides. h. Conidial chains. i. Germinating conidia. Scale bars = 5 μm

Full size image

Typus

Sri Lanka, Peradeniya Botanic Garden, from decaying leaf of Clerodendron monahassa, ex-type culture CBS 698.73=UAMH 4935.

Description of CBS 698.73 after one month incubation on OA, 24 °C: Colonies moderately expanding, olivaceouse black, with slimy centre and olivaceouse grey, flat margin. Reverse olivaceous black. No diffusible pigment produced. Hyphae slightly pigmented, 1.2–2.0 μm wide, regularly septate about every 10–20 μm. Conidiophores mostly absent; phialides rarely with an additional septum. Phialides brown, acicular, (7.5−) 9.0–16 (−20) × 1.0–2.0 μm, narrowed towards the tip, often with a nearly cylindrical apical portion. Collarettes somewhat divergent, slightly darker than the phialide body. Conidia forming long chains, with a broadly truncate base and a long, pointed or rarely rounded tip, slightly pigmented, smooth-walled, 1.0–3.5 × 0.8–1.5 μm. Chlamydospores absent. Teleomorph unknown. Cardinal temperatures: minimum 21 °C, optimum 27 °C, maximum 33 °C.

Notes

Phylogenetically P. oxyspora is clearly recognizable within the europaea-clade (Fig. 3). Morphologically the chalara-like phialides with long conidial chains (Fig. 17h) are exceptional in the black yeast-like fungi. Two of the three strains known to date (Table 1) originated from human cutaneous samples.

Phialophora reptans de Hoog, in de Hoog, Weenink & Gerrits van den Ende, Stud. Mycol. 43: 117. 1999–MycoBank MB460051. Fig. 18.

Fig. 18
figure 18

Phialophora reptans (CBS 113.85). a–b. Colony on OA, obverse and reverse. c. Torulose hypha, conidium and swollen cell. d–g. Hyphae with conidial heads. h–m. Hyphae with sessile collarettes and phialides. n, o. Torulose hyphae. p. Conidia with inflated germinating cells. Scale bars = 5 μm

Full size image

Typus

Uppsala, Sweden, from food stuffs, ex-type culture CBS 113.85.

Description of CBS 113.85 after one month incubation on OA, 24 °C: Colonies restricted, olivaceous black, flat, with slightly velutinous centre. Reverse olivaceous black. No diffusible pigment produced. Torulose hyphae present at colony centre. Expanding hyphae 1.5–2.5 μm wide, sometimes slightly inflated, smooth-walled, subhyaline to olivaceous brown, regularly septate every 10–20 μm. Phialidies intercalary, sometimes on terminal cells; rarely lateral. Collarettes mostly sessile on undifferentiated cells, conspicuous, slightly darker than the rest of the phialide, narrowly funnel-shaped to almost cylindrical, up to 2.5 μm long. Conidia subhyaline, smooth-walled, obovoidal, 1.0–3.0 × 1.0–1.8 μm, finally inflating to ellipsoidal germinating cells measuring about (2.0−) 3.5–4.0 (−6.0) × 1.5–4.0 μm, sometimes with sessile collarettes. Chlamydospores absent. Teleomorph unknown. Cardinal temperatures: minimum 21 °C, optimum 21 °C, maximum 27 °C.

Notes

Phialophora reptans was isolated from bark, soil water, human nails, and skin scrapings (de Hoog et al. 1999; Table 1). The limited expansion growth, presence of germinating cells and non-catenate conidia all indicate a relationship with the Herpotrichiellaceae.