Candida boidinii isolates from olive curation water: a promising platform for methanol-based biomanufacturing

Abstract

Methanol is a promising feedstock for biomanufacturing, but the efficiency of methanol-based bioprocesses is limited by the low rate of methanol utilization pathways and methanol toxicity. Yeast diversity is an attractive biological resource to develop efficient bioprocesses since any effort with strain improvement is more deserving if applied to innate robust strains with relevant catabolic and biosynthetic potential. The present study is in line with such rational and describes the isolation and molecular identification of seven isolates of the methylotrophic species Candida boidinii from waters derived from the traditional curation of olives, in different years, and from contaminated superficial soil near fuel stations. The yeast microbiota from those habitats was also characterized. The four C. boidinii isolates obtained from the curation of olives’ water exhibited significantly higher maximum specific growth rates (range 0.15–0.19 h⁻¹), compared with the three isolates obtained from the fuel contaminated soils (range 0.05–0.06 h⁻¹) when grown on methanol as the sole C-source (1% (v/v), in shake flasks, at 30°C). The isolates exhibit significant robustness towards methanol toxicity that increases as the cultivation temperature decreases from 30°C to 25°C. The better methanol-based growth performance exhibited by C. boidinii isolates from olives´ soaking waters could not be essentially attributed to higher methanol tolerance. These methanol-efficient catabolizing isolates are proposed as a promising platform to develop methanol-based bioprocesses.

Introduction

Among the one-carbon (C1) compounds derived from carbon dioxide (CO₂), methanol is a promising feedstock for biomanufacturing for the transition from a petroleum-based economy to a sustainable bioeconomy (Frazão and Walther 2020; Cotton et al. 2020; Sarp et al. 2021; Zhang et al. 2022a; Mitic et al. 2023). Since it is liquid, methanol is easy to store, and transport compared to gaseous C1 compounds (Cotton et al. 2020). Methanol does not compete with food crops and land being an attractive feedstock in bioprocesses due to its relative low price, renewability, and high availability (Zhang et al. 2018, 2022a; Cotton et al. 2020; Zhan et al. 2021; Lee and Sarwar 2022). Methanol can be produced in high quantities from low-quality coal and from CO₂ by photocatalytic or electrical reduction, thus contributing to a nearly zero CO₂ footprint in methanol-based biomanufacturing (Liu et al. 2018a; Rumayor et al. 2019; Fabarius et al. 2021; Sahoo et al. 2021; Kim et al. 2022). Moreover, compared with sugar-based feedstocks, the higher degree of reduction of methanol is beneficial for the production of reduced bioproducts with commercial value (e.g. alcohols, fatty acids, and carboxylic acids) (Sarwar and Lee 2023). Native methylotrophic microorganisms can utilize methanol as the only source of carbon and energy. Although they are ideal hosts for methanol bioconversion and have been examined and engineered to produce value-added bioproducts from methanol, the bioprocess efficiency is restricted by the rate of methanol utilization (Fan et al. 2018; Wang et al. 2023b). This efficiency is limited by the low rate of methanol utilization pathways and by its toxicity. Methanol toxicity (Mota et al. 2021) may limit the productivity of methanol-based bioprocesses (Wang et al. 2020b; Zhan et al. 2021). The toxic effects of this liposoluble molecule are not only due to methanol itself, but also to formaldehyde resulting from methanol bioconversion (Wakayama et al. 2016). Formaldehyde is a DNA-protein crosslinking agent and a branch point in methylotrophic yeasts between the dissimilatory pathway which leads to the generation of CO₂ and the assimilatory pathway, responsible for biomass production (the Xylulose Monophosphate Pathway (XuMP) (Chistoserdova and Kalyuzhnaya 2018; Lee and Sarwar 2022).

Described methylotrophic yeasts belong to a limited number of genera: Candida, Pichia and some genera that have been more recently separated from Pichia, including Komagataella, Kuraishia, and Ogataea (Yurimoto et al. 2011; Yurimoto and Sakai 2019). The most relevant representatives of methylotrophic yeasts are Candida boidinii, Ogataea polymorpha (syn. Hansenula polymorpha or Pichia angusta), and Komagataella phaffii (syn. Pichia pastoris, Ogataea methanolica or Pichia methanolica) (Hartner and Glieder 2006; Yurimoto and Sakai 2019). Different strategies towards the optimization of methanol catabolism in yeasts have been explored, involving the design of synthetic methylotrophic hosts or the genetic engineering of native methylotrophs (Cai et al. 2022a; Singh et al. 2022; Gan et al. 2023). However, the genetic manipulation of synthetic or native methylotrophic yeasts depends on the availability of efficient genetic engineering tools.

The exploitation of yeast diversity for the identification of natural isolates from different environments is a promising complementary alternative to develop efficient methanol-based bioprocesses. Indeed, any effort with yeast strain improvement is more deserving if applied to innate robust strains with relevant catabolic and biosynthetic potential. The present study is in line with such rational and describes the isolation of efficient methylotrophic yeast strains from two different habitats. They were molecularly identified as belonging to the Candida boidinii species and obtained from traditional curation of olives soaking waters or superficial soil contaminated with fuels near two different fuel stations. This isolation process was first designed with the objective of characterizing the yeast microbiota of these samples and to select yeast strains with potential to efficiently utilize glucose and xylose, the two major sugars present in lignocellulosic hydrolysates. The species C. boidinii belongs to the Ascomycota phylum, being phylogenetically related to the Ogataea clade. C. boidinii isolates have been obtained from natural environments (soil, seawater, sap fluxes of many sugar-rich tree species) samples associated with human activities (e.g. wine or olive fermentations) (Leventdurur et al. 2016; Arous et al. 2017; Oliveira et al. 2017). Genome annotated sequences from different C. boidinii strains are currently available (Borelli et al. 2016; Camiolo et al. 2017; Krassowski et al. 2018; Shen et al. 2018) being important ant to guide further studies. For all these reasons, the Candida boidinii isolates examined in this study are promising hosts for methanol-based bioprocesses.

Materials and methods

Isolation of yeasts

The Candida boidinii strains used in this study (indicated in Table 1) were obtained during this work from soaking waters resulting from the traditional curation of olives from Mértola region, Portugal, in different years (IST 350, 473, 509, and 605), or from the superficial layer of soil, collected near two different fuel stations (IST 592, 599, and 600). During the isolation processes, additional yeast isolates from other species were obtained from the samples examined. The isolates C. boidinii IST 350 and IST 605 were obtained using ten-fold serial dilutions, in 0.85% (w/v) NaCl, of the original sample of water from the curation of olives. These diluted samples were plated (100 µl) in YPD agar [10 g/L yeast extract (Difco), 20 g/L peptone (Difco), 20 g/L d-glucose (Merck), with 20 g/L agar (NZYTech)]. C. boidinii isolates IST 473, 509, 592, 599 and 600, were obtained based on three cycles of culture enrichment (Zaky et al. 2016). Approximately 0.5 mL of the original sample of the soaking waters from the curation of olives (isolates IST 473, and 509) or, in the case of C. boidinii isolates IST 592, 599 and 600, one gram of each soil sample, were introduced in 50 mL of growth medium containing 3 g/L malt extract (Sigma-Aldrich), 3 g/L yeast extract (Difco), 5 g/L peptone (Difco), 1 g/L (NH₄)₂SO₄ (Panreac), 0.25 g/L KH₂PO (Panreac, pH 5.0) supplemented with 30 g/L of glucose (Scharlau) and 30 g/L of xylose (Sigma-Aldrich), as carbon sources. Those growth media were supplemented with chloramphenicol (100 µg/mL) and incubated at 30°C with orbital agitation (150 rpm) for 48 h (first enrichment). Then, 1 mL of the first enrichment cultures was added to 49 mL of the same medium for a second enrichment step. The third enrichment consists of a differential enrichment performed as previously described but using either 60 g/L glucose (IST 592, 599 and 600) or 60 g/L xylose (IST 473 and 509). These media included 3 g/L yeast extract (Difco), 5 g/L peptone (Difco), 1 g/L (NH₄)₂SO₄ (Panreac), 0.25 g/L KH₂PO₄ (Panreac), and 20 g/L agar (NZYtech), supplemented with either 60 g/L glucose or 60 g/L xylose (Zaky et al. 2016) as carbon sources, both supplemented with chloramphenicol (100 µg/mL). Plates were incubated at 30°C for 48 h. Isolated colonies were streaked onto fresh YPD agar plates that were incubated under the same conditions to confirm the purity of cultures. Yeast isolates were maintained at 4ºC until DNA extraction for molecular identification. For long-term storage, the isolates are preserved at −80°C in YPD medium containing 15% (v/v) glycerol.

Table 1 Yeast isolates obtained from different samples with the correspondent ID, date of sample collection, and isolation methodology

Molecular identification of yeast isolates

For the molecular identification of the isolates obtained, genomic DNA was extracted using the phenol: chloroform: isoamyl alcohol method (Hoffman 1997) and used as template for the amplification of the D1/D2 domain sequence of the 26S ribosomal DNA (rDNA) and the internal transcribed spacer (ITS) region of rDNA. Polymerase Chain Reaction (PCR) was performed using the primer pairs NL-1 (5’-GCATATCAATAAGCGGAGGAAAAG-3’) and NL-4 (5’-GGTCCGTGTTTCAAGACGG-3’), and ITS1 (5’-TCCGTAGGTGAACCTGCGG-3’) and ITS4 (5’-TCCTCCGCTTATTGATATGC-3’), respectively, considered effective for the taxonomic identification of yeasts (Kurtzman and Robnett 1998). The PCR protocol included a denaturation step at 98°C for 30 s, followed by 35 cycles of denaturation at 98⁰C for 10 s, annealing at 52⁰C for 20 s, and extension at 72°C for 30 s. A final extension was performed at 72°C for 10 min. To amplify the DNA fragments, the Phusion™ High-Fidelity DNA Polymerase (Thermofisher Scientific) was used. The amplicons from D1/D2 and ITS regions were purified using NZYGel pure (NZYTech) and Sanger-sequencing (Stabvida) using the corresponding primer. The identification was performed by comparing their D1/D2 and ITS sequences with those deposited in GenBank using the BLAST algorithm from the National Center for Biotechnology Information (NCBI) (http://www.ncbi.nlm.nih.gov/blast). The consensus sequences from D1/D2 and ITS rDNA regions were deposited in the GenBank under the accession numbers listed in Table 1. All the relevant strains proposed in this study were deposited into the culture collection Micoteca da Universidade do Minho (MUM, WDCM 816) and the deposit numbers are available in Table 1.

Characterization of C. boidinii isolates growth using methanol as the sole carbon source

The preliminary characterization of the ability of the seven C. boidinii isolates obtained to catabolize methanol as the sole carbon source, mid-exponential cells pre-grown in YPD medium were used to inoculate a medium with methanol as the sole carbon source. This medium was prepared with 6.4 g/L Yeast Nitrogen Base (YNB, from Difco) at pH 6.0, supplemented with 1% (v/v) of methanol. A standardized initial OD_600nm of 0.20 ± 0.05 was used. Growth was performed at 30°C with orbital agitation at 250 rpm. Growth curves were followed based on the increase of culture OD_600nm for 90 h.

The effect in the growth curve and methanol utilization of increasing the initial methanol concentrations was examined under identical conditions for isolate IST 350 using 1% (v/v), 2% (v/v), 3% (v/v) or 4% (v/v) methanol (VWR), at two incubation temperatures: 25°C and 30°C.

The concentration of methanol present during C. boidinii IST 350 cultivation was determined by HPLC (Hitachi LaChrom Elite, Tokyo, Japan), using a column Aminex HPX- 87 H (Bio-Rad, Hercules, CA, USA) coupled with a refractive index detector. Culture samples were centrifuged (9700× g, 3 min) and 100 µl of the supernatant was pipetted into high-performance liquid chromatography (HPLC) vials and diluted with 900 µl of 50 mM H₂SO₄. Ten microliters of each sample were robotically loaded on the column and eluted with 5 mM H₂SO₄ as mobile phase at a flow rate of 0.6 mL/min for 30 min. The column and refractive index detector temperature was set at 50ºC, respectively. The concentration of methanol was calculated using a calibration curve.

Maximum specific growth and methanol consumption rates were calculated by least-square fitting to the linear parts of semi-logarithmic (optical density at 600 nm and methanol concentration, respectively) plots versus time.

Methanol- and salt-induced stress susceptibility assays

The tolerance to methanol or to salt of the seven Candida boidinii isolates was tested by spot assays. Other isolates obtained from the curation of olives’ water, Pichia kluyveri IST 472 and Pichia membranifaciens IST 493 (Table 1) were also tested for salt tolerance. The highly salt-tolerant species Debaryomyces hansenii IST 375 (MUM 24.14) was also used as a positive control. Yeast suspensions were prepared from mid-exponential cell cultures grown in YPD medium. Cells were harvested, washed, and re-suspended in sterile water to a standardized optical density (OD_600nm) value of 1.0. Serial 10-fold dilutions (10⁰ to 10⁴) of these cell suspensions were spotted (4 µL) onto the corresponding solid medium to evaluate salt or methanol tolerance. To assess methanol tolerance, yeast cell suspensions, prepared as described before, were spotted on YPD-agar medium supplemented with 8% (v/v) or 10% (v/v) methanol and plates were incubated at 25°C or 30°C for 48 h. Regarding methanol assimilation efficiency as the sole carbon source (C-source), at increasing methanol concentrations, yeast cell suspensions were spotted on YNB-agar medium supplemented with 1% (v/v) or 6% (v/v) methanol. This spot growth test allowed the assessment of both methanol assimilation efficacy and methanol tolerance. Growth on YNB-agar plates with methanol as C-source was recorded after 90 h of incubation at 25°C or 30°C. To assess salt susceptibility, YPD-agar medium was supplemented with sodium chloride (1.0 M NaCl, Panreac) and yeast cell suspensions were spotted in this medium and plates incubated at 30°C for 72 h. The results shown are representative of equivalent results obtained from at least two independent experiments, following several tests for the selection of the methanol concentrations to be tested.

Results

Isolation and molecular identification of yeast microbiota

The yeast microbiota from the water resulting from traditional curation of olives, collected in different years, and from the superficial layer of soil contaminated with fuels near two different fuel stations, was characterized using the isolation procedures summarized in Table 1. Only ascomycetous yeasts were isolated, including five different genera: Candida, Pichia, Wickerhamomyces, Torulaspora, and Meyerozyma. The microbiota of the curation of olives’ water included the species Candida boidinii, Pichia kluyveri, Pichia membranifaciens, Candida tropicalis, and Wickerhamomyces anomalus (Table 1) whose isolation from olives has already been reported (Heperkan 2013). From the fuel station superficial soil samples, isolates of C. boidinii, Candida albicans, Meyerozyma caribbica, Meyerozyma guilliermondii, Torulaspora delbrueckii and Torulaspora quercuum (Table 1), were obtained. The accession numbers of the D1/D2 and ITS consensus sequences of these isolates, molecularly identified at the species level, were deposited at the GenBank NCBI database (Table 1).

Among the yeast species isolated from the two different habitats, only C. boidinii is described as methylotrophic (Lachance et al. 2011). Four isolates, molecularly identified as C. boidinii, were obtained from the waters resulting from traditional curation of olives, collected in different years: IST 350 in 2017, IST 473 and 509 in 2019, and IST 605 in 2020 (Table 1). Three other C. boidinii isolates were obtained from the superficial layer of soil contaminated with fuels, collected near fuel station 1 (IST 592) and fuel station 2 (IST 599, and IST 600) (Table 1). C. boidinii isolates IST 350 and IST 605 were isolated directly from the original sample whereas C. boidinii IST 473, 509, 592, 599 and 600 isolates were obtained after three cycles of enrichment. The third enrichment step was performed in glucose- (IST 473, 592, 599, 600) or xylose-supplemented (IST 509) medium.

Growth efficiency of C. boidinii isolates on methanol as the sole carbon source

The performance of the C. boidinii isolates obtained was compared concerning methanol utilization as the sole C-source. The concentration selected for the first comparative analysis was 1% (v/v) methanol. Although low to be used as the sole C-source, this concentration limits methanol toxicity. The ability to grow on methanol as the sole carbon source was confirmed for the seven C. boidinii isolates tested (Fig. 1). Remarkably, all the four isolates obtained from the curation of olives´ water in the period 2017–2020 exhibited significantly higher values (around 3-fold) for the maximum specific growth rates (range 0.15–0.19 h⁻¹), compared with all the three isolates obtained from soil near two different fuel stations (range 0.05–0.06 h⁻¹).

Fig. 1

Growth curves of Candida boidinii isolates obtained using 1% (v/v) of methanol as the sole carbon source. The C. boidinii isolates obtained in this work from waters from traditional curation of olives in 2017 (IST 350), 2019 (IST 473 and 509) and 2020 (IST 605) or from the superficial layer of soil contaminated near fuel station 1 (IST 592) or station 2 (IST 599 and IST 600) were grown in YNB medium at pH 6.0 supplemented with 1% (v/v) methanol, at 30ºC, in shake flasks with orbital agitation (250 rpm). Growth was followed during 90 h based on culture optical density at 600 nm (OD_600nm)

For one of the efficient methanol catabolizing isolates obtained from olives curation water, isolate IST 350, the growth curves and methanol utilization profiles were obtained during growth in YNB with increasing initial concentrations of methanol (range 1–4% (v/v)), and at the temperatures 25°C and 30°C. The objective of reducing the cultivation temperature to 25°C was to assess the expected decrease of methanol toxicity given that the optimal and maximum temperatures for growth decrease as the concentration of an inhibitory compound increases (Van Uden 1985; Godinho et al. 2021). For the lower concentration tested (1% (v/v)), methanol was fully catabolized in the first 40 h of cultivation while for higher concentrations of 2% and 3% (v/v) of methanol the cultivation time for complete methanol utilization increased, not surpassing 70 h for 3% (v/v) (Fig. 2). The effect of the temperatures tested on the growth curve was marked for an initial concentration of 4% (v/v). At this high concentration, methanol was not fully consumed at both temperatures, but a clearly higher utilization rate was registered at the lower temperature, 25°C. For lower initial methanol concentrations, a slight improvement of methanol bioconversion at 25ºC was only registered for initial concentrations above 1–2% (v/v), consistent with the balance between the effect of the optimum temperature for growth and methanol tolerance of C. boidinii IST 350.

Fig. 2

Growth curves and methanol assimilation by C. boidinii IST 350 using increasing initial methanol concentrations. Growth curves (black circles) and methanol concentration profiles (grey squares) during cultivation of C. boidinii IST 350 in YNB medium supplemented with increasing methanol concentrations of 1% (v/v) (a), 2% (v/v) (b), 3% (v/v) (c), or 4% (v/v) (d), with orbital agitation (250 rpm) at 30 °C. Results are from a representative shake flask experiment of other independent experiments leading to similar results

For a closer look on the effect that increasing initial concentrations of methanol had on growth parameters at the incubation temperatures 25°C and 30°C, these values were estimated based on the growth curves shown in Fig. 2(Fig. 3). Incubation at the lower temperature (25°C) apparently led to shorter lag phase duration, this being particularly evident at higher methanol concentrations (Fig. 3, (a)). Regarding the maximum specific growth rate at 25°C for the lower concentration of methanol, the value was below the value calculated at 30ºC and was not affected by the increase of initial methanol concentration up to 2% (v/v) but, for higher concentrations, this value decreased; however, those values were, for all the concentrations tested, above those calculated for growth at 30ºC (Fig. 3, (b)). The maximum specific rate of methanol consumption also suffered a higher methanol inhibition at 30ºC for concentrations above 2% (v/v) (Fig. 3, (c)). The estimated maximum yeast biomass produced, associated to maximum culture OD_600nm, increased twice at 25°C as the concentration of methanol increased from 1 to 2% (v/v) methanol (2.1-fold at 25°C, compared with 1.4-fold at 30°C), consistent with the undetectable inhibitory effect of methanol observed at 25°C, for concentrations below 2% (v/v) (Fig. 3, (d)). The differential of the maximum biomass produced at 25°C compared with 30°C increased with the initial methanol concentration, for concentrations above 2% (v/v) with the biomass concentrations attained at 30°C also being below the concentrations attained at 25°C. At the initial concentration of methanol of 4% (v/v), the biomass concentration produced after around 90 h of cultivation, at both temperatures, suffered a marked decrease because this carbon source was not fully catabolized, with a stronger impact at 30ºC (Fig. 3, (d)).

Fig. 3

Effect of increasing initial methanol concentrations on the growth parameters of C. boidinii IST 350 incubated in shake flasks at 25°C or 30°C. Effect of initial methanol concentration on the duration of the lag phase (a), the maximum specific growth rate (b) and the maximum specific methanol consumption rate (c), and the final biomass produced, associated to culture OD_600nm (d), of C. boidinii IST 350, estimated based on the growth curves shown in Fig. 2. Open symbols are used for cultivations at 25°C and closed symbols at 30 °C. The methanol concentrations present at the end of cultivation are also displayed in the final biomass production panel (squares in (d)))

Methanol and salt susceptibility assays of C. boidinii isolates

To understand if the higher growth efficiency of the group of C. boidinii isolates from the soaking waters from the curation of olives (IST 350, 473, 509, and 605), compared with the group of isolates from superficial layer of soil contaminated with fuels (IST 592, 599, and IST 600), is due to differences in tolerance to methanol or to a more efficient methanol catabolism, methanol susceptibility of the isolates was compared by spot assays in YPD agar plates supplemented with increasing concentrations of methanol (Fig. 4, (a)). The methanol concentrations that had to be used in these spot assays were much higher than those used in liquid medium, possibly because a rich solid medium was used and methanol evaporation during incubation is expected to be higher, even though the growth phenotypes were registered after 48 h of incubation to avoid an exacerbated evaporation. Interestingly, at 1% (v/v) methanol, the less efficient methanol- catabolizing isolates obtained from soil contaminated with biofuels were found to be more tolerant to methanol than some of the isolates obtained from water from olives’ curation (Fig. 4, (a)). Remarkably, the isolate IST 350, selected to be examined for growth on increasing concentrations of methanol in shake flasks, exhibits the lower methanol tolerance, indicating that the performance of other isolates from olive curation waters may be better than the registered in Fig. 2 results.

The more efficient growth on methanol of the group of isolates from the water from olives’ curation, compared with the other group of isolates, was confirmed by spot assays in YNB medium with methanol as the sole C-source, especially for 1% (v/v) of methanol while such higher efficiency was not so clear for the more toxic methanol concentration of 8% (v/v) (Fig. 4, (b)). Photographs were taken after 90 h of incubation due to slower growth when YNB medium with methanol as the only C-source was used (Fig. 4, (a)).

Regarding salt tolerance, other isolates obtained from the curation of olives’ water, Pichia kluyveri IST 472 and Pichia membranifaciens IST 493 were included in the analysis as well as a strain of the highly salt tolerant species Debaryomyces hansenii, as a positive control. As observed for methanol tolerance, the C. boidinii isolates obtained from salted water obtained from traditional curation of olives (IST 350, 473, 509, and 605), except for isolate IST 605, were not more tolerant to salt compared with the soil isolates (IST 592, 599, and 600) after 72 h of incubation at 30°C, (Fig. 4, (c)). P. kluyveri IST 472 was found to be highly susceptible to saline stress since its growth in YPD-agar medium supplemented with 1.0 M NaCl growth was abrogated whereas P. membranifaciens IST 493 isolated from the same salted habitat was highly salt tolerant, similar to D. hansenii IST 375. All together, these results appear to suggest that, in general, highly tolerant yeasts are not rigidly selected in the salted habitat of the soaking waters from olives curation.

Fig. 4

Susceptibility to methanol- or saline- induced stresses, by spot assays, of the seven C. boidinii isolates. Spot growth of C. boidinii isolates obtained from water from the curation of olives (in bold) or from soil near fuel stations (underlined) in YPD-agar (a) or in YNB-agar (b) supplemented with increasing concentrations of methanol, used as toxicant (a) or as the sole carbon source (b), or in YPD-agar supplemented with a high salt (NaCl) concentration (c). For comparative analysis, the highly salt-tolerant yeast Debaryomyces hansenii IST 375 was used as positive control. Salt susceptibility of Pichia kluyveri IST 472 and Pichia membranifaciens IST 493, obtained from the same samples of soaking water from olives’ curation, were also tested. Registration of phenotypes was performed after incubation for 48 h (a) or 90 h (b) at 25 °C or 30 °C or for 72 h of growth at 30 °C (c)

Discussion

The rich yeast diversity in different ecological environments is a natural resource to be explored for the development of a bio-based economy. Therefore, it is instrumental to obtain strains with diverse useful physiological characteristics such as an efficient catabolism of diverse C- sources, specialized biosynthetic ability to produce value-added bioproducts, and tolerance to a wide range of stresses of industrial relevance. The selection of yeast phenotypes highly suited to specific bioprocesses can be the basis for subsequent effective generation of additional genetic improvement through classic or site-specific mutagenesis, adaptive laboratory evolution, metabolic engineering, and synthetic biology design, depending on the available genetic tools and information for the specific yeast species. Here we propose a group of yeast strains, particularly those isolated from olive curation soaking waters, for methanol-based biomanufacturing. Methanol is a promising feedstock considered as a next-generation feedstock (Zhang et al. 2022a) but using methanol as sole C-source is challenging due to the inefficiency of methylotrophic microorganisms in its assimilation rate into biomass and bioproducts (Zhang et al. 2022a, b). Consistent with the characterization of Candida boidinii as methylotrophic (Hartner and Glieder 2006; Yurimoto et al. 2011), all the seven C. boidinii isolates obtained were capable to grow with methanol as the sole carbon and energy source. Remarkably, all the strains isolated from the traditional curation of olives soaking waters, in different years, showed more promising results for highly efficient biomass production from moderate concentration of methanol, compared to those isolated from the superficial layer of soil contaminated with fuels.

The soaking waters resulting from the curation of olives are unique habitats due its low concentration of sugar and relatively high concentrations of oil, phenolic compounds and salt (Arroyo-López et al. 2008; Mujdeci and Ozbas 2021). Together with P. kluyveri, P. membranifaciens and W. anomalus (Arroyo-López et al. 2008; Perpetuini et al. 2020), C. boidinii was reported to be one of the predominant yeasts in olive fermentation, having a strong lipase activity that contributes to the final organoleptic characteristics of this fruit by changing the olives’ free fatty acids composition (Bonatsou et al. 2018; Perpetuini et al. 2020). Interestingly, among all the yeast species of the isolates obtained in this study from the curation of olives´ water, C. boidinii, Candida tropicalis, Meyerozyma guilliermondii and Wickerhamomyces anomalus, are described in the literature as oleaginous (Ramírez-Castrillón et al. 2017; Fabricio et al. 2019; Abeln and Chuck 2021; Mota et al. 2022).

The isolate C. boidinii IST 350 obtained from those soaking waters exhibit maximum specific growth rates when grown in YNB with methanol as the sole carbon source in shake flasks above those reported for methylotrophic yeasts (Brinkmann et al. 1990; Tomàs-Gamisans et al. 2018; Wefelmeier et al. 2023) and comparable with some methylotrophic bacteria (Šmejkalová et al. 2010; Simões et al. 2022). The apparent lower methanol tolerance exhibited by this isolate, compared with others isolated from the same habitat, in different years, suggests that there is still space to improve methanol-based growth by exploring other bioresources gathered in this work, especially under optimized growth conditions in bioreactors. Initial methanol concentrations in the range 2−3% (v/v) led to the best results for this strain in the range of cultivation temperatures 25–30°C being more favorable at 25°C likely due to reduced methanol toxicity (Sá-Correia and Van Uden 1983; Van Uden 1985). The use of the lower temperature tested (25°C) favors growth and methanol bioconversion compared with 30ºC, and the effect is more evident with the increase of the initial methanol concentration. This effect includes the reduction of the duration of methanol-induced lag phase and the increase of biomass yield and maximum specific growth and methanol consumption rates, emphasizing the described effect of lipophilic compounds in decreasing the optimal and maximum temperature of growth in yeasts (Sá‐Correia and Van Uden 1983; Van Uden 1985). In fact, the major bottleneck in methanol-based biomanufacturing is the limited methanol assimilation also related with methanol susceptibility of the majority of the methylotrophic microorganisms (Wang et al. 2023b).

The robustness towards methanol of the C. boidinii isolates obtained from the different habitats examined was found to be significant and the much better performance of those isolated from olives curation waters, compared with those isolated from soil contaminated with fuels, could not be related with a higher methanol tolerance but, apparently, with a higher assimilation capacity, at least for moderate methanol concentrations. The methylotrophic yeast C. boidinii has been found associated with plant leaves (Kawaguchi et al. 2011) and in other pectin-rich sources, including fruits and wines (Nakagawa et al. 2000). C. boidinii exhibits a pectinolytic activity, being able to degrade pectin at the methyl ester moiety using extracellular pectin methylesterases and to utilize this methanol and D-galacturonic acid, the main component of pectin, as C-sources (Nakagawa et al. 2000). Among the yeast species isolated from the curation of olives´ water, C. boidinii is the only methylotrophic species (Kurtzman 2011a, b; Lachance et al. 2011). Since the methanol concentrations found in these environments are relatively low (Sánchez et al. 2000; Montaño et al. 2003), it is likely that a high methanol tolerance is not a dominant trait among the methylotrophic yeast isolates present while methanol-based growth efficiency should be to enhance the assimilation of pectic substances as well as competitiveness and adaptation to this specific habitat (Stratilová et al. 1998). However, methanol assimilation demands higher maintenance energy compared to growth on glucose (Tomàs-Gamisans et al. 2018; Guo et al. 2023), partly due to the low energy efficiency of the Xylulose monophosphate (XuMP) pathway. In this pathway, one molecule of ATP is consumed for each pyruvate molecule produced while in glycolysis one molecule of ATP is generated per pyruvate molecule (Guo et al. 2023). Furthermore, a significant portion of the assimilated carbon is oxidized to carbon dioxide in the dissimilatory branch of methanol metabolism (Jordà et al. 2014). Although methylotrophs are ideal hosts for methanol bioconversion and have been examined and engineered to produce value-added bioproducts from methanol, the bioprocess efficiency is restricted by the rate of methanol utilization (Fan et al. 2018; Wang et al. 2023b). This efficiency is limited by the low rate of methanol utilization pathways and by methanol toxicity. Attempts to improve methanol catabolism in the methylotrophic yeast Pichia pastoris (now Komagataella phaffii) included the rewiring of central metabolism for efficient production of free fatty acids from methanol (Cai et al. 2022c) or the reduction of alcohol oxidase activity and addition of sodium citrate (Liu et al. 2023), previously described to improve glutathione synthesis (Gao et al. 2022), and the stimulation of the activity of isocitrate dehydrogenase and strengthening of the tricarboxylic acid cycle (Chen et al. 2016). The supplementation of growth media with yeast extract, amino acids or vitamins was also found to be beneficial to cell growth and metabolism, compensating the cytotoxicity and low energy content of methanol(Zhang et al. 2022a) and the limited supply of amino acids, nucleotide sugars and acetyl-CoA that are the precursors for the biosynthetic machinery (Cai et al. 2021; Zhang et al. 2022a; Wegat et al. 2022; Wang et al. 2023a).

C. boidinii is considered an oleaginous yeast suitable for lipid production for biodiesel either derived from glycerol (Papanikolaou et al. 2017) or pumpkin peel hydrolysate (Demiray et al. 2022) substrate. However, although under non-optimized conditions, preliminary results suggest that methanol-based lipid production by the isolates obtained in this work does not allow lipid contents above 10% (v/v). Nevertheless, the engineering of nativ methylotrophs to produce increased lipid concentration has potential, as is the case of other bioproducts. Interesting examples include the production of malic acid (Guo et al. 2021; Wefelmeier et al. 2024), lactic acid (Wefelmeier et al. 2023), isoprene and acetone (Wefelmeier et al. 2024), β-alanine (Miao et al. 2021), lovastatin (Liu et al. 2018b), long-chain α-alkenes (Cai et al. 2022b). The availability of genome annotated sequences from different C. boidinii strains(Borelli et al. 2016; Camiolo et al. 2017; Krassowski et al. 2018; Shen et al. 2018) pave the way for the rational genetic manipulation of the C. boidinii isolates obtained in this work as robust and efficient hosts for methanol biomanufacturing processes. If the necessary efficient genome engineering tools are not available or can be easily developed, the adaptive laboratory evolution (ALE) strategy is a promising alternative (Sandberg et al. 2019; Arora et al. 2020; Wang et al. 2023b). There are examples of the successful use of ALE to obtain evolved methanol tolerant strains, as in the case of the non-methylotrophic oleaginous yeast Rhodotorula toruloides (Fernandes et al. 2023), or to increase methanol assimilation in non-native methylotroph bacteria, such as Corynebacterium glutamicum (Wang et al. 2020a) or Escherichia coli (Bennett et al. 2021), or even to stimulate methanol catabolism in S. cerevisiae (Espinosa et al. 2020). Concerning methylotrophs, the genetic engineering of the native methylotrophic yeast Ogataea polymorpha, combining adaptive laboratory evolution, rational metabolic engineering and multi-omics analyses was a successful example of an efficient production of free fatty acids using methanol as the sole carbon source (Zhang et al. 2022b). These examples emphasize the relatively unexplored potential of optimization of native methylotrophs as robust cell factories and the results obtained in this study highlight the potential of the C. boidini isolates obtained towards the development of efficient methanol conversion bioprocesses.