Introduction

The Spathaspora clade contains several d-xylose-fermenting yeast species isolated from rotting wood or wood-boring insects (Lopes et al. 2016; Cadete and Rosa 2017). Spathaspora was proposed by Nguyen et al. (2006) to accommodate the teleomorphic species Spathaspora passalidarum, isolated from a passalid beetle in Louisiana, USA, and the anamorphic species Candida jeffriesii, isolated from a beetle in Chiriqui, Panama. The genus is composed of a core group that is well supported by phylogenetic analyses of D1/D2 LSU sequences and includes Sp. passalidarum, the type species of the genus, C. jeffriesii, Candida materiae, Sp. arborariae, Sp. brasiliensis, Sp. girioi, and Sp. suhii (Barbosa et al. 2009; Cadete et al. 2009, 2013; Daniel et al. 2014; Lopes et al. 2016; Cadete and Rosa 2017). The species Sp. roraimanensis shares the unique ascospore morphology of the genus and is closely related to Sp. gorwiae, Sp. hagerdaliae and Sp. xylofermentans, in which ascospore production have not been observed. Lopes et al. (2016), using phylogenomic analyses, showed that Sp. gorwiae and Sp. hagerdaliae form a subclade within the Spathaspora clade. Spathaspora allomyrinae was described by Wang et al. (2016) from three strains isolated from the gut of Allomyrina dichotoma beetles in China. This species forms two elongated ascospores similar to those of the Spathaspora clade. However, its phylogenetic position within this clade is less clear (Wang et al. 2016; Lopes et al. 2016). Morais et al. (2017) described the species Sp. boniae based on two strains producing asci containing elongate ascospores with curved ends typical of the genus Spathaspora. That species was isolated from rotting wood in Brazil. A phylogenomic analysis placed Sp. boniae in an early-emerging position relative to the larger Candida albicans/Lodderomyces clade. Based on these results, the authors suggested that the genus Spathaspora is, in fact, paraphyletic. Cadete and Rosa (2017) suggested that whole genome sequencing of all Spathaspora species and those of related genera, combined with the discovery of new species of the clade will help to elucidate the phylogeny of this genus.

Yeasts capable of metabolizing d-xylose, the second most abundant sugar in lignocellulosic based-feedstocks, may produce ethanol or xylitol as main value-added products (Cadete et al. 2017; Cadete and Rosa 2017). This biotechnological trait has encouraged efforts to search for novel d-xylose-fermenting yeasts, leading to the description of new species endowed with this ability. Some of the most promising yeasts showing these properties are members of the Spathaspora clade, including Sp. passalidarum and Sp. arborariae, which has stimulated studies on yeasts of this genus (Cadete and Rosa 2017).

During an exploration of yeasts associated with rotting wood in an Atlantic Rainforest area in the municipality of Piracicaba, São Paulo state, Brazil, many d-xylose-fermenting strains were isolated. Analyses of ITS-5.8S and D1/D2 sequences of the large subunit rRNA gene showed that two of these strains represent a new species belonging to the Spathaspora clade. The strains did not produce ascospores when examined individually or mixed in pair. We propose the name Spathaspora piracicabensis sp. nov. for this species. We also characterized ethanol production from d-xylose by the two strains.

Materials and methods

Yeast isolation and identification

The yeasts were isolated from samples of rotting wood collected from an ornamental tree (purple quaresmeira, Tibouchina granulosa) in an area (22º42′33″S and 47º37′58″W) of Atlantic Rainforest in the state of São Paulo, Brazil. The area is located on the campus of the Escola Superior de Agricultura Luiz de Queiroz (ESALQ, Universidade de São Paulo), in the municipality of Piracicaba. The predominant vegetation is characterized as an Atlantic Rainforest biome. Twelve rotting wood samples were collected from decayed branches (15–20 cm length) of the tree T. granulosa in March 2013. The samples were stored in sterile plastic bags and transported under refrigeration to the laboratory over a period of no more than 24 h. Approximately 3 g of each sample were placed separately in 250 mL Erlenmeyer flasks with 50 mL sterile YNB-xylose medium (yeast nitrogen base 0.67%, d-xylose 0.5%, chloramphenicol 0.02% and tetracycline 0.02%, pH 5.0), as described previously (Cadete et al. 2017). The flasks were incubated at 25 °C in an orbital shaker (New Brunswick™ Innova® 40/40R, USA) at 150 rpm for 5 days. When growth was detected, 0.5 mL of each culture was transferred to flat-bottomed tubes containing 5 mL sterile YEPX broth (yeast extract 1%, peptone 1%, d-xylose 2%, chloramphenicol 0.02% and tetracycline 0.02%, pH 5.0). The tubes were incubated in an orbital shaker at 150 rpm, 25 °C during 48 h. After growth detection in YEPX broth, serial decimal dilutions were performed in sterile distilled water to obtain cell concentrations of 2–4 × 102 cells mL−1. An aliquot of 100 µL of the appropriated dilution was seeded on YEPX agar and the plates were incubated at 28 °C for 5 days, until yeast colonies developed.

The different yeast morphotypes were purified by repeated plating on YEPX agar, and preserved at − 80 °C. Yeasts showing potential for ethanol production from d-xylose in the screening tests described below were morphologically and physiologically characterized using the methods described in Kurtzman et al. (2011). Sporulation was investigated using 1% glucose, 0.01% yeast extract (GY), yeast carbon base plus 0.01% ammonium sulfate (YCBAS), dilute (1:9 and 1:19) V8, Fowell’s acetate and YM agars at 15 and 25 °C for up to 4 weeks.

Species identification was performed by the analysis of the D1/D2 variable domains of the 26S ribosomal RNA gene and the ITS-5.8S region of the large subunit of the rRNA gene, as described previously (White et al. 1990; O’Donnell 1993; Lachance et al. 1999). The amplified DNA was concentrated, cleaned and sequenced in an ABI 3130 Genetic Analyzer automated sequencing system (Life Technologies, USA) using BigDye v3.1 and POP7 polymer. The sequences were assembled, edited and aligned with the program MEGA6 (Tamura et al. 2013). The phylogenetic trees were constructed based on LSU rRNA gene D1/D2 domains and ITS-5.8S sequences by Maximum Likelihood analysis of 510 and 379 aligned positions, respectively, using the Tamura-Nei substitution model with a gamma rate distribution and invariant sites. Bootstrap values were determined from 1000 pseudoreplicates.

Screening of d-xylose-fermenting yeasts

To assess the ability to ferment d-xylose, the yeasts obtained were subjected to fermentation assays carried out on complex medium containing d-xylose as sole carbon source. Spathaspora arborariae (UFMG-CM-Y352T = CBS 11463T), Sp. passalidarum (UFMG-CM-Y470), Scheffersomyces parashehatae (UFMG-CM-Y507) and Sc. stipitis NRRL Y-7124T were used as positive controls for D-xylose fermentation (Cadete et al. 2012, 2017).

Yeast inocula were prepared by adding one loopfull of a recent yeast growth culture (24–48 h) in test tubes containing 3 mL of YEPD broth (yeast extract 1%, bacteriological peptone 1%, dextrose 2%) incubated at 28 °C for 48 h without shaking. After growth, 100 µL of each culture were re-inoculated in test tubes containing 3 mL of YEPD medium and incubated at 28 °C for 36 h. Screening of d-xylose-fermenting yeasts were performed by adding 100 µL of the inoculum (OD600nm = 0.1) in 35 mL flat bottomed tubes containing 6 mL of YEPX medium (d-xylose 2%). The tubes were incubated at 28 °C on an orbital shaker at 150 rpm. Samples (1 mL) were taken regularly at ranges of 12 h to quantify the cell biomass, xylose and metabolite concentration. After this first screening step, a second d-xylose fermentation assay with selected strains was performed by adding 200 µL of the inoculum with (OD600nm = 0.1) in 35 mL flat bottomed tubes containing 6 mL of YEPX medium (d-xylose 2%). The tubes were incubated and sampled as described above.

Cell concentration was determined in spectrophotometer (Micronal B-380, Brazil) by optical density at 600 nm. Afterwards, samples were centrifuged and the resulting supernatants were diluted and filtered using a 0.2 µm cellulose acetate filter (Sartorius, Germany). Levels of d-xylose, xylitol, and ethanol were determined by a CG480C high performance liquid chromatography (HPLC) system with a Bio-Rad Aminex HPX-87H column (Hercules, USA) and a 410 GC refractive index detector (Scientific Instruments CG LTDA, Brazil). The mobile phase was 5 mM H2SO4 at a flow rate of 0.6 mL min−1, 65 °C, and a sample injection of 5 µL (Basso et al. 2008).

Results and discussion

Yeast isolation and screening

A total of 83 yeast strains were obtained from the 12 rotting wood samples collected.

After the screening test and performing a second D-xylose fermentation assay (d-xylose 2%), the strains UFMG-CM-Y5867T (= ESALQ-I54T) and UFMG-CM-Y6112 (= ESALQ-I38) showed the most substantial ethanol production from d-xylose, with titers of 3.37 and 3.48 g L−1 ethanol, respectively, in 24 h. These results were higher to those observed at 24 h of fermentation for Sc. stipitis NRRL Y-7124 (2.47 g L−1) and Sp. arborariae UFMG-CM-Y352 (3.03 gL−1) but lower than the ethanol titers obtained by Sp. passalidarum UFMG-CM-Y470 (4.63 g L−1) and Sc. parashehatae UFMG-CM-Y507 (4.58 g L−1). Xylitol production in 24 h for UFMG-CM-Y5867T and UFMG-CM-Y6112 was equivalent to 1.32 and 1.08 g L−1, respectively, whereas for the remaining yeasts it varied from zero to 0.43 g L−1 xylitol (data not showed). In this preliminary d-xylose culture assay, it was possible to observe that UFMG-CM-Y5867T and UFMG-CM-Y6112 were able to produce higher ethanol than xylitol titers. However, other d-xylose fermentation assays are necessary to verify the real and potential biotechnological application of these strains with regard to the production of bioethanol from lignocellulosic sugars.

Species delineation and phylogenetic placement

The analyses of the sequences of the ITS-5.8S and the D1/D2 sequences of LSU rRNA of strains UFMG-CM-Y5867T and UFMG-CM-Y6112 showed that these yeasts represent a novel species of the clade Spathaspora, and is phylogeneticly related to Sp. brasiliensis, C. materiae and Sp. girioi (Fig. 1). Sequences of the ITS-5.8S and D1/D2 domains of both isolates were identical. The novel species differs by six substitutions in D1/D2 domains from Sp. brasiliensis and ten substitutions from C. materiae and Sp. girioi. In the ITS-5.8S region, both strains of the novel species differ by 24 substitutions and 13 indels from Sp. brasiliensis, six substitutions and three indels from C. materiae, and 29 substitutions and 16 indels from Sp. girioi.

Fig. 1
figure 1

Phylogenetic placement of Spathaspora piracicabensis sp. nov. based on LSU rRNA gene D1/D2 domains sequences. The tree was constructed by maximum likelihood analysis of 510 aligned positions using the Tamura-Nei substitution model with a gamma rate distribution and invariant sites. Bootstraps were determined from 1000 pseudoreplicates and values above 50% are indicated in the tree. Bar, 0.05 substitutions per site

Full size image

Phylogenetic analysis of the ITS-5.8S region confirmed that the new species is distinct from closely related species (Supplementary Fig. 1). Despite the description of new species such as Sp. boniae improved our knowledge regarding Spathaspora phylogeny by suggesting it as a paraphyletic genus (Morais et al. 2017), the new species here described was phylogenetically placed between known Spathaspora spp. using both genetic markers (sequences of ITS-5.8S and D1/D2-26S rRNA), not contributing for clarifying the phylogenetic boundaries of the genus Spathaspora. Therefore, the description of new species is still needed for a better comprehension of the complex phylogeny of this genus.

Both strains were examined individually, or mixed in pairs on several sporulation media, but asci or signs of conjugation were not seen. The name Spathaspora piracicabensis sp. nov. is proposed to accommodate this species. The mention forma asexualis (f.a.) was added in the title as a reminder that a sexual state is not known (Lachance 2012).

Description of Spathaspora piracicabensis sp. nov. C.S. Varize, R.M. Cadete, L.D. Lopes, R.M. Christofoleti-Furlan, M.A. Lachance, L.C. Basso and C.A. Rosa sp. nov. Mycobank number MB 822320

Etymology: The specific epithet pi.ra.ci.ca.ben’sis N.L. nom. adj., piracicabensis, refers to the Brazilian city of Piracicaba, where this yeast was found.

On YM agar, after 7 days at 25 °C, the cells are spherical to ovoid (3–4 × 3–4 μm). Budding is multilateral (Fig. 2). Colonies are white, smooth, glistening and convex and pseudohyphae are present. Asci or signs of conjugation were not seen on sporulation media. Fermentation of d-glucose, maltose, trehalose and d-galactose are positive. Xylose fermentation is negative using Durham tubes, but ethanol is produced from xylose when determined by HPLC. d-glucose, d-galactose, l-sorbose (variable), maltose, sucrose, cellobiose, trehalose, melezitose, d-xylose, d-ribose, ethanol, erythritol, ribitol, d-mannitol, d-glucitol, salicin, succinate, hexadecane (variable, slow and weak), xylitol, ethylacetate and N-acetyl-d-glucosamine are assimilated. No growth occurs on lactose, melibiose, raffinose, inulin, soluble starch, l-arabinose, d-arabinose, l-rhamnose, glycerol, galactitol, dl-lactate, citrate, myo-inositol, methanol, acetone, isopropanol and d-gluconate. Assimilation of nitrogen compounds is positive for lysine and negative for nitrate and nitrite. Growth in amino-acid free medium is positive. Growth at 37 °C is positive, and at 40 °C is negative. No growth is observed on 50% (w/w) glucose, 1% acetic acid or 10% (w/v) sodium chloride/5% (w/v) glucose. Growth on 0.01% cycloheximide is negative. Starch-like compounds are not produced. Acid production is negative.

Fig. 2
figure 2

Budding yeast cells and pseudohyphae of Spathaspora piracicabensis sp. nov. UFMG-CM-Y5867T on YM agar after 7 days at 25 °C

Full size image

The differentiation of the novel species from Sp. brasiliensis (its near neighbour, Fig. 1) is based on the assimilation of DL-lactate, which is negative for the novel species and positive for Sp. brasiliensis, and the assimilation of d-ribose, which is positive for Sp. piracicabensis and negative for Sp. brasiliensis. Spathaspora piracicabensis differs from C. materiae based on the assimilation of ethylacetate and ethanol, which are both positive for Sp. piracicabensis and negative for C. materiae, and the assimilation of glycerol, which is negative for the novel species and positive for C. materiae. Lastly, Sp. piracicabensis differs from Sp. girioi based on the assimilation of dl-lactate and citrate, which are both negative for the novel species and positive for Sp. girioi, and the assimilation of d-ribose and ethylacetate, which are both positive for Sp. piracicabensis and negative for Sp. girioi.

The known habitat of the new species is rotting wood collected in Atlantic Rain forest in Brazil. The type strain UFMG-CM-Y5867T was isolated from rotting wood of an ornamental tree (purple quaresmeira, Tibouchina granulosa) in an area of Atlantic Rainforest biome, in Piracicaba, São Paulo state, Brazil. It has been deposited in the Collection of Microorganisms and Cells of Federal University of Minas Gerais (Coleção de Micro-organismos e Células da Universidade Federal de Minas Gerais, UFMG), Belo Horizonte, Minas Gerais, Brazil, as strain UFMG-CM-Y5867T, and it is permanently preserved in a metabolically inactive state. An ex-type culture has been deposited in the collection of the Westerdijk Fungal Biodiversity Institute, Utrecht, the Netherlands, as strain CBS 15,054. The MycoBank number is MB 822,320. The Genbank accession numbers for the ITS-5.8S and D1/D2-26S rRNA partial gene sequences of strains UFMG-CM-Y5867T (ESALQ-I54) and UFMG-CM-Y6112 (ESALQ-I38) are KR864907/KR864906 and KX528211/KX528210, respectively.