Xylose fermentation to ethanol by new Galactomyces geotrichum and Candida akabanensis strains

Isolation of xylose assimilator yeast

The fungi were isolated from samples of fruits and roots that included avocados (Persea americana), bananas (Musa balbisiana), potatos (Solanum tuberosum), beets (Beta vulgares esculenta), taro (Colocasia esculenta), passion fruit (Passiflora sp.), pepper (Capsicum annuun) and tomatos (Solanum lycopersicum). All these biomasses were obtained at local fairs and markets and at an advanced stage of maturity or early natural microbial decomposition. Sugarcane bagasse (Saccharum officinarum) and sweet sorghum bagasse (Sorghum bicolor L. Moench) were also used as sources of microorganism samples.

Microorganisms of interest were isolated from 5-g portions of the previously fragmented plant sample that were transferred to conical flasks containing 50 mL of YNBX medium (0.67% yeast nitrogen base and 3% D-xylose) with 0.02% chloramphenicol (Lorliam et al., 2013). These flasks were incubated at 28 °C for 120 h with stirring at 150 rpm in an orbital incubator (Nova Ética model 430) for prior enrichment of the population of fungi that assimilate D-xylose. Every 24 h, 100-µL aliquots of culture medium were collected, inoculated by spreading over YNBX containing 1.5% agar, and incubated at 28 °C for an additional 48 h. The colonies were isolated with the aid of a platinum loop, suspended in sterile water, inoculated in solid YMPD medium (0.3% yeast extract, 0.3% malt extract, 0.5% peptone, 1% glucose and 1.5% agar), and incubated for 48 h at 28 °C to confirm the purity of the colonies. The isolated and purified colonies were inoculated in liquid YMPD medium and incubated for 48 h at 28 °C. Sufficient glycerol was added to the medium to furnish a 10% solutions, 1-mL aliquots were transferred to cryogenic tubes, and the pure cultures were stored at −18 ± 1 °C for subsequent tests of fermentability, characterization and identification.

Assay of gas production from xylose as sole carbon source

The ability of the isolates to produce gas in the presence of xylose as the sole source of carbon was evaluated in test tubes with screw caps containing inverted Durhan tubes and the YNBX medium, according to the procedure described by Kurtzman (2011). The experiment was conducted at 28 °C in an orbital shaker at 120 rpm for 21 days with daily monitoring of gas production. The CBS6054 lineage of Scheffersomyces stipitis was used as a positive control.

Morphological and biochemical characterization of the selected fungi

Macroscopic and microscopic observations of the selected gas producing strains were performed after growth on solid YMPD medium at 28 °C for 48 h for the morphological characterization. In the macroscopic observations were analysed the texture, color, shape, type of surface, border and profile characteristics with aid of a stereoscope at 40× magnification. Microscopic examination was performed using 400× and 1,000× magnification with the aid of a trinocular microscope coupled to a 5.0-Mpixel digital camera to capture the images. The formation of pseudohyphae and true septated hyphae and the cell shapes were evaluated.

The reactivity of the colonies with Diazonium B Blue (DBB) was used to distinguish between ascomycetes and basidiomycetes (Hagler & Ahearn, 1981). The effect of temperature (28 °C, 30 °C, 35 °C, 37 °C, 40 °C and 42 °C) on the growth of the selected strains in YMPD medium was also observed for a period of up to 21 days. Biochemical assays of the assimilation of different nitrogen and carbon sources (cadaverine, creatinine, nitrate, nitrite, lysine, glucose, maltose, raffinose, trehalose, xylose, arabinose, fructose, sucrose, inulin, meso-erythrytol, methanol, xylitol, glycerol, starch, melibiose and galactose) were performed according to the procedure described by Kurtzman (2011). The technique of replica plating on solid media containing basal agar (0.67% YNB, 2% agar) and 0.2% of carbon or nitrogen source was used. The growth of colonies was observed for up to 48 h. The biochemical assay to verify the fermentation of sugars (glucose, xylose, sucrose, fructose, maltose, raffinose, galactose, melibiose and ribose) was also used to characterize the isolates utilizing the procedure described by Kurtzman (2011). To 2 mL of basal medium (4.5 g yeast extract, 7.5 g of peptone and 3 mg of bromothymol blue in 1 L of distilled water), previously autoclaved in test tubes with screw caps and containing inverted Durham tubes, was added 1 mL of 6% sugar solution, and the solution was inoculated with 100 µL of microbial suspension. The tubes were kept at 28 °C and monitored for 21 days for the production of gas. The ability of the selected strains to hydrolyze starch and to produce urease was also investigated according to the method described by Kurtzman (2011).

Molecular identification of selected fungi

Colonies grown on solid YMPD medium were transferred to centrifugal microtubes with the aid of a platinum loop, and the DNA was extracted according to the method described by Green & Sambrook (2012). Estimation of the amount and quality of extracted DNA was performed by electrophoresis in 1% agarose gel (w/v) followed by DNA band revelation with ethidium bromide. The molecular identification of the selected strains was performed by sequencing of rDNA regions using the NL1 (5′-GCATATCAATAAGCGGAGGAA-3′) and NL4 (5′-GGTCCGTGTTTCAAGACGG-3′) primers for amplification of the D1/D2 domain of the gene responsible for encoding the 26S rRNA region. The ITS1 (5′-TCCGTAGGTGAACCTGCGG-3′) and ITS4 (5′-TCCTCCGCTTATTGATATGC-3′) primers were employed for the amplification of the gene responsible for encoding the 5.8 S rRNA region. Amplification of regions of interest and sequencing of the PCR products were accomplished by the Macrogen Company (Rockville, MD, USA; http://www.macrogenusa.com). The sequences obtained were compared with sequences deposited in the GenBank nucleotide database (National Center for Biotechnology Information, NCBI, http://www.ncbi.nlm.nih.gov) using the BLAST program (basic local alignment search tool) (Zhang et al., 2000).

Assay of alcoholic fermentation

The isolates that tested positive for the production of gas from xylose were then evaluated for their capacity for the production of ethanol. The ethanol production assays were performed with the liquid medium described by Oliveira (2010), contained 20 g L⁻¹ of xylose, 1.25 g L⁻¹ of urea, 1.1 g L⁻¹ of KH₂PO₄, 2 g L⁻¹ of yeast extract and 40 mL L⁻¹ of micronutrient solution (12.5 g L⁻¹ MgSO₄.7H₂O, 1.25 g L⁻¹ CaCl₂, 2.5 g L⁻¹ citric acid, 10.9 g L⁻¹ FeSO₄.7H₂O, 0.19 g L⁻¹ MnSO₄, 0.3 g L⁻¹ ZnSO₄.7H₂O, 0.025 g L⁻¹ CuSO₄.5H₂O, 0.025 g L⁻¹ CoCl₂.6H₂O, 0.035 g L⁻¹ (NH₄)₆MoO₂₄.4H₂O, 0.05 g L⁻¹ H₃BO₃, 0.009 g L⁻¹ KI, and 0.0125 g L⁻¹ Al₂(SO₄)₃). The final pH was 5.0.

The frozen stock cultures were reactivated in YMPD solid medium and inoculated in 250-mL conical flasks containing 150 mL of previously described liquid medium, followed by incubation at 28 °C on an orbital shaker at 150 rpm until an optical density of one unit at 610 nm was reached. Subsequently, 20 mL of this culture containing thickened yeast was used as an inoculum for evaluation of ethanol production in the same medium. Fermentation experiments were conducted in conical 250-mL flasks with hydrophobic cotton plugs. The flasks contained 80 mL of medium, to which was added 20 mL of inoculum, were incubated at 28 °C on an orbital shaker at 150 rpm for 72 h. The fermentation process was monitored every 12 h by means of the determination of the concentration of reducing sugars, by a colorimetric method (Miller, 1959), and monitoring cell growth profile using a Neubauer chamber. The cell growth profile also was expressed in dry weigth by means of experimental correlation with cell suspension optical density reading at 610 nm.

At the end of each fermentation, the concentration of ethanol was determined using a spectrophotometric method (Isarankura-Na-Ayudhya et al., 2007). The production of ethanol was confirmed by HPLC, as described by Matos et al. (2018). The product yield by substrate (Y _P∕S, g g⁻¹) was calculated as the ratio between the amount of ethanol produced and the amount of reducing sugars consumed at end of fermentative process; the ethanol yield by cell growth (Y _P∕X, g g⁻¹) was calculated as the ratio between the amount of ethanol produced and the amount of cell biomass formed at end of fermentative process; the fermentation efficiency (Ef%) were calculated as ratio between Y _P∕S and theoretical alcoholic fermentative yield (0.511); the volumetric productivity (Q_P, g L⁻¹ h⁻¹), which is the ratio between the final concentration of ethanol and the fermentation time, also was acessed. The specific growth rate (µX, h⁻¹) was calculated as the number of cells produced in a defined time interval during the exponential growth phase using µ$X=\frac{\ln \left(X_1∕X_0\right)}{t_1-t_0}$, where x₁ and x₀ are the cellular concentrations (dry weigth) and t₁ and t₀ are the culture times (hours). The alcoholic fermentation tests and all the analytical determinations were performed in triplicate. Tukey’s test was performed at 0.05 p-level for comparison of means.

Isolation and selection of xylose fermenting fungi

Two hundred and two microbial colonies that were able to grow in solid medium with xylose as the sole carbon source were isolated from different ten plant biomasses (Table 1). However, only three microbial isolates, coded as UFVJM-R10, UFVJM-R150 and UFVJM-R131, which were obtained from taro, tomato and sugarcane bagasse samples, respectively, were able to produce gas in liquid medium containing xylose as the sole carbon source. This production occurred within 96 h of cultivation. Gas production was interpreted as evidence that these isolates could eventually achieve the desired alcoholic fermentation of xylose, which produces CO₂ as a coproduct. The confirmation of this expectation was evaluated with fermentative tests followed by determination of the production of ethanol.

Morphological characterization of isolated fungi strains

Photographs of the morphotypes of the D-xylose-fermenting strains cultured in solid YMPD medium at 28 °C for 48 h can be observed in Figs. 1 and 2. All the colonies had circular shapes and a lack of diffuse pigment (Fig. 1). The UFVJM-R10 and UFVJM-R150 strains had surfaces with concentric grooves, a border, filamentous growth and a radial aspect. A smooth profile and a central concavity in the form of a crater were also observed in the UFVJM-R10 and UFVJM-R150 strains. A creamy appearance, smooth surface and edges, flat profile and yellowish-white color were observed for the UFVJM-R131 strain.

The growth of the selected strains cultivated in solid YMPD medium at 28 °C was also evaluated (Table 2). All the selected strains had a minimum radial growth of 7 mm after 24 h of culture. The colonies of the UFVJM-R10 and UFVJM-R150 strains doubled in size in 48 h. Considering the time interval between 24 and 336 h, the size of the colonies of these same strains increased about eigth times, reaching up to 90 mm in diameter. The UFVJM-R131 strain increased comparatively slowly, reaching a maximum of 15 mm at the end of 336 h of cultivation.

As for the microscopic appearance of strains grown in YMPD at 28 °C for 48 h, cylindrical cells formed by true mycelium hyphae, positive germ tube and the presence of ascospores, chlamydospores and arthroconidia (Fig. 2) were observed for the the UFVJM-R10 and UFVJM-R150 strains. Globular and ovoid cells, pseudo-hyphae formation, and the presence of ascospores and blastoconidia were observed for the UFVJM-R131 strain.

The selected strains grew at 28 °C and 30 °C after 24 h of cultivation in YMPD medium, but no growth was observed at temperatures equal to or higher than 35 °C. This fact characterizes them as mesophilic. The DBB test realized with isolated colonies was negative for all the three selected strains, thereby indicating that they belong to the Ascomycetes group (Hagler & Ahearn, 1981). The results of the tests for starch hydrolysis and production of urease were negative for all three strains. These selected fungal strains possessed the ability to assimilate the pentoses D-xylose and L-arabinose (Table 3). With the exception of meso-erythritol and inulin, all other carbon sources tested were assimilated within 48 h. The UFVJM-R10 and UFVJM-R150 strains did not assimilate inulin. All the nitrogen sources tested (cadaverine, creatinine, nitrate, nitrite and L-lysine) were assimilated by the three strains (Table 3). The selected strains also exhibited the capacity to ferment glucose, fructose and xylose (Table 4). None of the strains was able to ferment D-ribose or melibiose. The UFVJM-R131 strain was the only one capable of fermenting sucrose, maltose, galactose and raffinose. The results of fermentation, when positive, were observed within 48 h of incubation.

Homology search of 5.8S rDNA and 26S rDNA regions

Nucleotide sequences of the PCR products obtained by amplifying the regions of the small and large subunit ribosomal RNA genes from selected xylose-fermenting strains were deposited on GenBank-NCBI (Table 5). The degree of identity in the NCBI nucleotide database was searched using the BLAST tool (Zhang et al., 2000). The UFVJM-R10 isolate presented 99% identity to strains of Geotrichum candidum and Galactomyces geotrichum when the D1/D2 region of the 26S rDNA amplified sequence was researched (GenBank Accession Number MF362099) (Table 5). A 100% identity with strains of Geotrichum candidum and Galactomyces geotrichum was observed for the UFVJM-R150 isolate using the ITS region amplified with ITS1/ITS4 primers (GenBank Accession Number MF360015) as a parameter for comparison to the GenBank. A 99% identity with the Galactomyces geotrichum and Geotrichum candidum species was observed for the same microbial isolate when the partial sequence of the D1/D2 region of the 26S rDNA amplified with NL1/NL4 primers (GenBank accession number MF371338) was researched. A 98% identity with Candida akabanensis species was observed for the UFVJM-R131 isolate when the ITS region amplified with ITS1/ITS4 primers (GenBank accession number KY325443) was used as a reference. A 99% identity with the same species was observed when the partial sequence of the D1/D2 region of the 26S rDNA gene amplified with the NL1/NL4 primers (GenBank accession number KY325444) was used as a reference in the GenBank.

Production of ethanol by selected isolates

The G. geotrichum UFVJM-R10, G. geotrichum UFVJM-R150 and C. akabanensis UFVJM-R131 strains were evaluated with regard to the consumption of xylose for microbial growth and for the production of ethanol. The three fungi strains consumed 100% of the xylose available in 60 h (Fig. 3). Considering only the growth curves expressed in dry weight, the G. geotrichum UFVJM-R10 and G. geotrichum UFVJM-R150 strains were shown to exhibit the same growth profile (Fig. 4). The steady state growth was achieved within 72 h after initiating the process (Fig. 4).

The yields resulting from the fermentation as a function of the substrate consumption (Y _P∕S) and cell growth (Y _P∕X), fermentation efficiency (Ef), volumetric productivity (Q_P), the production of ethanol and specific growth rate (µ) are presented in Table 6. The values of Y _P∕S ranged from 0.29 to 0.35, but they were not statistically different from one another. The ethanol production reached 5.12 g L⁻¹. However, there were no significant differences between the Q_P values or ethanol production for any of the strains. The specific growth rates observed for the C. akabanensis UFVJM-R131 strain (0.37 h⁻¹) was at least three times greater than those calculated for the other two strains (Table 6).