Skip to main content

Genotypic and phenotypic diversity among Komagataella species reveals a hidden pathway for xylose utilization

Abstract

Background

The yeast genus Komagataella currently consists of seven methylotrophic species isolated from tree environments. Well-characterized strains of K. phaffii and K. pastoris are important hosts for biotechnological applications, but the potential of other species from the genus remains largely unexplored. In this study, we characterized 25 natural isolates from all seven described Komagataella species to identify interesting traits and provide a comprehensive overview of the genotypic and phenotypic diversity available within this genus.

Results

Growth tests on different carbon sources and in the presence of stressors at two different temperatures allowed us to identify strains with differences in tolerance to high pH, high temperature, and growth on xylose. As Komagataella species are generally not considered xylose-utilizing yeasts, xylose assimilation was characterized in detail. Growth assays, enzyme activity measurements and 13C labeling confirmed the ability of K. phaffii to utilize D-xylose via the oxidoreductase pathway. In addition, we performed long-read whole-genome sequencing to generate genome assemblies of all Komagataella species type strains and additional K. phaffii and K. pastoris isolates for comparative analysis. All sequenced genomes have a similar size and share 83–99% average sequence identity. Genome structure analysis showed that K. pastoris and K. ulmi share the same rearrangements in difference to K. phaffii, while the genome structure of K. kurtzmanii is similar to K. phaffii. The genomes of the other, more distant species showed a larger number of structural differences. Moreover, we used the newly assembled genomes to identify putative orthologs of important xylose-related genes in the different Komagataella species.

Conclusions

By characterizing the phenotypes of 25 natural Komagataella isolates, we could identify strains with improved growth on different relevant carbon sources and stress conditions. Our data on the phenotypic and genotypic diversity will provide the basis for the use of so-far neglected Komagataella strains with interesting characteristics and the elucidation of the genetic determinants of improved growth and stress tolerance for targeted strain improvement.

Introduction

Yeasts are unicellular fungi naturally occurring in a wide variety of different habitats. More than 1500 species with high phylogenetic diversity are known so far. Several of ascomycete yeast species are very well studied, of which many are used in the food industry and for biotechnological applications. However, most isolated species have not been thoroughly characterized so far. Analysis of genome sequences and phenotypic traits of these yeasts will help to understand the evolution of yeast species and provide a great resource for industrial strain development.

The yeast genus Komagataella consists of seven methylotrophic species classified according to the divergence of marker gene sequences [1]. All the available strains were isolated from tree environments in Europe and North America. So far, only strains of two Komagataella species, K. phaffii and K. pastoris (both formerly classified as Pichia pastoris), have been studied in detail. Originally investigated for the production of single-cell protein from methanol, strains of K. phaffii and K. pastoris are now mostly used for recombinant protein production. Complete genome assemblies and gene annotations have been published for K. phaffii CBS 7435, its histidine auxotrophic derivative GS115, an AOX1 deletion (mutS) strain, and the K. pastoris type strain CBS 704 [2,3,4,5,6]. More recently, the genome sequencing data of six additional natural K. phaffii isolates, including the CBS 2612 type strain, have become available and up to 44,000 single nucleotide polymorphisms (SNPs) could be identified between some of the strains [7, 8]. Also, several transcriptomics and proteomics studies under different industrially relevant conditions have been performed [9], which enables targeted genome engineering for improved recombinant protein production and metabolic engineering. However, there is little information available on strains of other Komagataella species.

As methylotrophic yeasts, Komagataella species can utilize methanol as a carbon and energy source, and they show fast growth to high cell densities on carbon sources like glucose, glycerol, ethanol, and methanol under aerobic conditions. Usually, cells are grown at temperatures of 25–30 °C, but growth between 20 and 37 °C is possible. K. phaffii strains can tolerate neutral to acidic pH down to pH 3 [10], while the pH is usually kept at 5–6 during fermentation processes.

Komagataella species are generally not known to assimilate the pentose sugar d-xylose, although one study reported slow growth of non-engineered K. phaffii GS115 cells [11]. Also, the formation of xylitol from xylose without visible cell growth has been described for K. pastoris [12]. Xylose is highly abundant in lignocellulosic biomass (around 30% of total carbohydrate monomers), which makes it a promising renewable resource for biotechnological processes. Xylose can be utilized by different species of bacteria, archaea, yeast and filamentous fungi. Identification of robust xylose utilizing species and genetic engineering of xylose-fermenting yeasts is highly relevant for industrial ethanol and chemical production [13,14,15]. In efficient xylose-utilizing yeast species like Scheffersomyces stipitis and Sugiyamaella lignohabitans, xylose assimilation takes place via an oxidoreductase pathway. In the first step, xylose is reduced to xylitol by xylose reductase (XR), which requires NADPH or NADH as cofactor [16]. Subsequently, xylitol is converted to xylulose by xylitol dehydrogenase (XDH) using NAD+ [17]. As a result, xylose reductases preferentially using NADPH as cofactor can cause cofactor imbalance and carbon loss through excess xylitol formation [18]. In the next step, xylulose is converted to xylulose-5-phosphate by xylulokinase (XK) using ATP and is further metabolized via the pentose phosphate pathway [13, 19]. Compared to other carbon sources, xylose is fermented by only few described yeast species. However, many more species have been shown to grow on xylose as carbon source or have putative xylose assimilation pathway genes [20, 21].

We present a comparative characterization of the phenotypic and genotypic diversity of 25 natural isolates from all seven known Komagataella species. Differences in cell growth were assessed on different carbon sources and in the presence of stressors at two different temperatures. In addition, we performed a detailed characterization of xylose assimilation, including 13C labeling. To assess the genotypic diversity within the genus, the genomes of the type strains of K. pseudopastoris, K. populi, K. kurtzmanii, K. ulmi and K. mondaviorum, as well as additional K. phaffii and K. pastoris strains were sequenced on the PacBio platform and used for comparative analysis of genome structure and sequence diversity.

Results

Genome sequencing of all Komagataella species

We generated whole-genome sequencing data for all seven species of the genus Komagataella using the Pacific Biosciences (PacBio) technology resulting in 7.9 × 106 subreads (43.8 Gbp) with read- lengths between ~ 1 kbp and 20 kbp (Additional file 1: Figure S1). For five species (K. kurtzmanii, K. mondaviorum, K. populi, K. pseudopastoris, and K. ulmi) the type strains were sequenced. Genome assemblies of the remaining two species, K. pastoris type strain CBS 704 and K. phaffii CBS 7435, which was recently shown to be highly similar to the type strain K. phaffii CBS 2612, were already available [4, 5, 7, 8]. Therefore, we selected additional strains of these species for sequencing. In total, eight strains were sequenced and assembled (two strains of K. pastoris, one strain for each of the other species). The quality of our assemblies was comparable to the two available assemblies with an average N50 size of 2.34 Mbp (Table 1).

Table 1 Assembly metrics for eight Komagataella strains assembled using PacBio whole-genome sequencing reads before removal of mitochondrial sequences

Phylogeny based on whole-genome sequencing data

The phylogenetic relationship of the seven Komagataella species was determined based on pairwise comparisons of a representative selection of subsequences from each genome assembly (resulting in “Mash” distances, Additional file 1: Table S1). In total, eleven genomes were included in the analysis, i.e. eight genomes assembled by us (see above), the two publicly available genomes of K. phaffii and K. pastoris, and the genome of Citeromyces matritensis [21] as outgroup. In the phylogenetic tree based on Mash distances the three K. pastoris strains clustered together, as well as the two K. phaffii strains (Fig. 1). The tree showed the relationship of K. ulmi to K. pastoris and of K. kurtzmanii to K. phaffii in two subtrees that were placed as sister groups, and the relationship of the species K. mondaviorum, K. populi, and K. pseudopastoris grouping together as a separate clade. The tree topology confirmed the findings of a previous study based on selected marker genes [1].

Fig. 1
figure 1

Phylogenetic tree of Komagataella species. The tree was calculated based on pairwise genomic distances between the assembled genomes as determined by Mash [22]. The genome sequence of Citeromyces matritensis NRRL Y-2407 (NCBI id ASM370516v1) was used as outgroup

Divergence in the genome structure of Komagataella species

Previous genome comparisons between K. phaffii and K. pastoris revealed genomic rearrangements [5, 23] which can be summarised as one exchange between chromosomes 1 and 3 and another one between chromosomes 2 and 4. We compared our genome assemblies with K. phaffii CBS 7435 as reference and detected the same pattern in K. ulmi (genome structure similar to K. pastoris) and K. kurtzmanii (genome structure similar to K. phaffii) consistent with the placement in the phylogenetic tree (Fig. 2A–E). The method to determine genomic distances for tree calculation takes short subsequences into account rather than large genomic regions, so that the divergence between the two clades was demonstrated both on the sequence level and in the genome structure.

Fig. 2
figure 2

Regions of similarity between the reference strain K. phaffii CBS 7435 (NCBI assembly id ASM170808v1) and K. ulmi CBS 12361 (A), K. pastoris DSMZ 70877 (B), K. pastoris CBS 9178 (C), K. kurtzmanii CBS 12817 (D), K. phaffii UWOPS 03-328y3 (E), K. pseudopastoris CBS 9187 (F), K. populi CBS 12362 (G) and K. mondaviorum CBS 15017 (H) based on whole-genome comparisons. Colors refer to the four chromosomes (chr1: orange, chr2: blue, chr3: green, chr4: purple), contigs appear in their initial order and orientation of the assemblies. Only matches spanning at least 10 kbp are shown and non-matching contigs were removed

Based on these genome comparisons and available chromosome information [5] we inferred the contig order and orientation for the five genome assemblies including chromosomal assignment (Table 2). The reference assemblies used for comparison consisted of four sequences (manually connected contigs, see [5]) corresponding to four chromosomes.

Table 2 Chromosome assignment, order, and orientation of contigs based on whole-genome comparisons to K. phaffii CBS 7435 and available chromosome information [5]

Genome comparisons of the more distant species K. pseudopastoris, K. populi, and K. mondaviorum to K. phaffii as reference showed a larger number of structural differences (Fig. 2F–H), and a direct assignment of contigs to K. phaffii chromosomes was not possible. However, when comparing K. pseudopastoris to K. populi, there appeared no major structural difference (Fig. 3A), and K. mondaviorum compared to K. populi showed fewer rearrangements than compared to K. phaffii (Fig. 3B). Although the genomic distance between K. mondaviorum and K. populi was smaller than the distance between the clusters of K. phaffii and K. pastoris, there was much more genomic rearrangement between K. mondaviorum and K. populi than between K. phaffii and K. pastoris. Each contig of K. populi showed similarity to several different contigs of K. mondaviorum, e.g., parts of K. populi contig2 matched in four different contigs of K. mondaviorum. According to the position in the sequence-based tree, the differences between these species were more obvious on the structural level than on the sequence level.

Fig. 3
figure 3

Whole-genome comparison of K. pseudopastoris CBS 9187 (A) and K. mondaviorium CBS 15017 (B) against K. populi CBS 12362. The order and coloring of K. populi contigs were assigned after inspecting the comparisons with K. phaffii (remains preliminary due to heavy rearrangements). Contigs of K. pseudopastoris and K. mondaviorum were ordered and oriented manually to achieve congruence with K. populi as far as possible. ctg contig, K. pop K. populi, rev reverse

Sequence divergence between Komagataella species

To assess the genomic divergence on the sequence level we used K. phaffii CBS 7435 as reference sequence for comparison with each of the eight assembled genomes (Table 3). The divergence based on sequence alignments resulted in three groups in terms of similarity with the closest to the reference being K. phaffii UWOPS 03-328y3 and K. kurtzmanii CBS12817, the second closest group represented by the K. pastoris strains and K. ulmi CBS 12361, and the most distant group comprising K. mondaviorum CBS 15017, K. populi CBS 12362, and K. pseudopastoris CBS 9187.

Table 3 Average sequence identity in matching regions ≥ 1000 bp between each of the listed genomes and the chromosomes of K. phaffii CBS 7435

Phenotypic characterization of 25 Komagataella strains

For characterization of the phenotypic diversity within the 25 Komagataella strains, spotting assays on different carbon sources and in the presence of high salt concentration, oxidative stress, and high and low pH conditions were performed. Colony formation and density were evaluated after 2 and 7 days of incubation at optimal (30 °C) or increased growth temperature (37 °C). Figure 4 shows the results of the growth tests. Overall, a rather homogenous growth behavior was observed under most conditions tested. As expected, at 30 °C all strains could grow well on glucose and glycerol as carbon sources. Growth was slightly reduced on plates with methanol, but to a similar level for all strains except K. populi CBS 12362, which showed a better growth on methanol. Growth on ethanol was similar to methanol for most strains. Only K. phaffii CBS 7435, K. pastoris CBS 9178, the two K. pseudopastoris strains, and K. populi CBS 12362 showed superior growth compared to the other strains tested. Interestingly, although Komagataella species are generally described as non-xylose utilizing yeasts, slow growth on xylose as the sole carbon source was observed for all the strains. However, the growth was much weaker than for the other carbon sources tested. Compared to all other strains tested, growth on xylose was slightly better for K. pastoris CBS 704 and K. populi CBS 12362 (Fig. 4B).

Fig. 4
figure 4

A Summary of spotting assays at 30 and 37 °C. Cell growth was evaluated after 2 and 7 days of incubation according to the following criteria: 0 = no growth (white), 1–5 rating depending on the dilution down to which growth was observed, 6 = large colonies in every dilution (dark blue). All plates with inhibitors contained glucose as carbon source. B Growth on xylose as the only carbon source. Plate after incubation at 30 °C for 3 days. C Growth at pH 9. Plates after incubation at 30 °C for 7 days

Cell growth was strongly inhibited by 1 mol L−1 NaCl after 2 days of incubation at 30 °C, but most strains were able to adapt to the high salt concentration within 7 days of incubation. Also, the medium with pH 9 had a strong inhibitory effect after 2 days, which both K. pseudopastoris strains and K. kurtzmanii CBS 12817 could completely overcome until day 7 (Fig. 4C). As expected, all strains showed a high tolerance against acidic pH, where small differences in colony growth could only be observed after 2 days of incubation. Also, all strains were able to grow in the presence of 2 g L−1 acetic acid, with K. populi CBS 12362 showing the fastest growth after 2 days at 30 °C. Resistance to oxidative stress was tested by incubation on plates with 0.5 and 1 mmol L−1 hydrogen peroxide, which had an inhibitory effect after 2 days, which was no longer visible after 7 days, likely due to evaporation of the inhibitor. In this experiment, K. phaffii UWOPS 85-263.1 and K mondaviorum CBS 15017 stood out as being more tolerant to 0.5 mmol L−1 hydrogen peroxide.

Additionally, all spotting assays were also performed at 37 °C (Fig. 4A). As expected, cell growth was generally worse at elevated temperature. When comparing cell growth of the different strains relative to each other, strains showing reduced growth at 30 °C also showed reduced growth at 37 °C on the different carbon sources as well as on the plates with an inhibitory environment. Only K. populi CBS 12362 grew to relatively high densities on the different carbon sources. As described previously, K. kurtzmanii and K. mondaviorum were not able to grow at 37 °C under any of the conditions tested [1, 24]. Similar to 30 °C, nearly every strain, except K. pastoris CBS 704, both K. pseudopastoris strains, K. ulmi CBS 12361 and K. populi 12362, was resistant to low pH. Likewise, only K. pastoris CBS 704, K. pseudopastoris CBS 9187 and the K. ulmi strains showed reduced growth in the presence of 0.5 mmol L1 hydrogen peroxide at 37 °C.

K. populi CBS 12362 shows faster growth on xylose than the other 24 strains

Xylose is a highly abundant substrate for biotechnological processes. Engineering of efficient xylose utilization in Komagataella phaffii has previously been achieved by the introduction of the xylose isomerase pathway [11]. To our knowledge, however, growth of non-engineered K. phaffii cells solely on xylose was only reported for experiments performed in complex medium [11]. Based on the previous spotting assay results, strains of three different Komagataella species were chosen for a more detailed investigation of xylose utilization. Growth of K. phaffii X-33 as industrially relevant strain from the K. phaffii CBS 7435 background, K. pastoris CBS 704 and K. populi CBS 12362 was analyzed in minimal medium containing 20 g L−1 xylose in shake flasks with an initial OD600 of 12 (circa 3.5 g L−1 yeast dry mass) (Fig. 5).

Fig. 5
figure 5

Growth, xylose consumption and xylitol production of K. phaffii X-33 (A), K. pastoris CBS 704 (B), and K. populi CBS 12362 (C) in YNB with 2% xylose (continuous lines with filled symbols). Dashed lines with empty symbols indicate samples from the control cultures in YNB with no addition of carbon source. Data represent the average of three biological replicates with standard deviation

No cell growth and metabolite formation were observed in the control conditions without a carbon source. K. phaffii X-33 was the best xylose consumer, using up an average of 95.70% ± 3.2 of the available xylose for nearly one doubling (1.85 ± 0.08-fold) within 10 days of incubation. K. pastoris CBS 704 had the slowest growth and xylose uptake in comparison to the other two strains (1.64 ± 0.13-fold and 47.1% ± 13.2 of the xylose, respectively). K. populi CBS 12362 showed the fastest growth and doubled the number of cells (2.04 ± 0.09-fold), consuming 61.4% ± 3.0 of the available xylose in the process. Xylitol was detected as a by-product with a final average of 0.51 ± 0.16 g L−1 in all three strains, with maximum production between 96 and 192 h of cultivation. K. pastoris CBS 704 was distinctively the best xylitol producer reaching 1.16 ± 0.54 g L−1 after 192 h of cultivation. Other metabolites like ethanol, acetic acid, and glycerol could not be detected as fermentation products.

To identify potential reasons for the observed differences in xylose utilization between K. phaffii X-33 and K. populi CBS 12362, the sequences of putative xylose pathway genes and their transcript levels were analyzed. Orthologs of xylose reductase, xylitol dehydrogenase, and xylulokinase have previously been identified in K. phaffii [11, 20]. We used the protein sequences of xylose reductase (length of 318 amino acids), xylitol dehydrogenase (363 amino acids) and xylulokinase (623 amino acids) from the xylose utilizing yeast S. stipitis to identify orthologous sequences in the newly sequences Komagataella species. According to sequence homology we inferred the putative xylose reductase gene GRE3 (PP7435_Chr3-0488, PAS_chr3_0744) encoding a 319 amino acid protein (320 amino acids in K. ulmi) in all our Komagataella assemblies with around 68% sequence identity to S. stipitis xylose reductase (Additional file 1: Table S2, Figure S2A). Interestingly, the deletion of GRE3 in K. phaffii CBS 7435 did not abolish but only reduce cell growth and spot density on xylose plates (Fig. 4B), indicating the presence of at least one other enzyme capable of catalyzing the xylose reductase reaction. Indeed, homology-based search identified the NADPH-dependent aldo–keto reductases Ypr1, PP7435_Chr2-0714 and PP7435_Chr4-0551 as potential candidates, sharing 29–37% sequence identity with S. stipitis xylose reductase (Additional file 1: Table S2, Figure S3). Based on sequence homology, SOR1 (PP7435_Chr1-0597, PAS_chr1-1_0490) was identified as putative xylitol dehydrogenase gene in all Komagataella assemblies. It encodes a 348 amino acid protein (409 amino acids in K. pastoris) with 54–55% sequence identity to the S. stipitis protein (Additional file 1: Table S3, Figure S2B). The inferred xylulokinase orthologs Xks1 (PP7435_Chr1_0598, PAS_chr1-1_0280) have a size of 605, 607 or 617 amino acids in the different Komagataella species and share 49–50% sequence identity with the S. stipitis protein (Additional file 1: Table S4, Figure S2C). Interestingly, the SOR1 and XKS1 genes were found next to each other in all the Komagataella assemblies analyzed. The impact of sequence variation between the different Komagataella enzymes on xylose utilization remains to be investigated.

Transcript levels of the putative xylose pathway genes in the presence of glucose or xylose as a carbon source were analyzed by quantitative PCR. For this, three clones of each strain were cultivated in YPD and then transferred to minimal medium with 20 g L−1 glucose or xylose. Samples for RNA extraction were taken after 6 h in minimal medium.

Figure 6A shows the relative transcript levels of the putative xylose pathway genes in K. phaffii X-33 and K. populi CBS 12362. When comparing the transcript levels of the two different strains in xylose medium, transcript levels of GRE3 and XKS1 were higher in K. populi CBS 12362 than in K. phaffii X-33, while SOR1 levels were higher in K. phaffii X-33. Expression of all genes, except K. phaffii XKS1, was upregulated at least 4.8-fold in xylose medium when compared to the expression of the same gene in glucose medium. The absence of induction of XKS1 expression in K. phaffii X-33 could be a reason for the slower growth of this strain on xylose.

Fig. 6
figure 6

A Relative transcript levels of putative xylose pathway genes in K. phaffii X-33 and K. populi CBS 12362. Transcript levels are given as fold change compared to the average SOR1 transcript level in K. phaffii X-33 grown in minimal medium with glucose. Error bars represent the standard deviation of three biological replicates shown as individual data points. B Specific enzyme activities of cell-free extracts prepared from cells grown in minimal medium with xylose. Cultures of the xylose utilizing yeast S. lignohabitans grown in the same medium were used as a control. Values are given as U mg−1 total protein. Errors represent the standard deviation of three biological replicates. Nd activity not detectable

Additionally, enzyme activities of xylose reductase and xylitol dehydrogenase were measured in cell-free extracts of cultures grown in YNB with 20 g L−1 xylose. Cell-free extracts of the xylose-utilizing yeast S. lignohabitans CBS 10342 grown under the same conditions were used as a positive control (Fig. 6B). As expected from the slow growth on xylose, the measured specific activities were very low for both reactions and around 100 times lower than for the positive control S. lignohabitans. Also, no significant differences in specific activity between K. phaffii X-33 and K. populi CBS 12362 could be detected. In many yeasts the reduction of xylose depends on NADPH as the preferred cofactor, however, efficient NADH-dependent xylose conversion has also been described for some species [25, 26]. For K. phaffii X-33 and K. populi CBS 12362 cell-free extracts, xylose reductase activity could only be determined with NADPH, indicating that the activity with NADH is either much lower or absent in these strains.

Xylose assimilation in K. phaffii X-33—13C labeling

To show the incorporation of carbon from xylose into cellular metabolites and yeast biomass, an isotope labeling experiment with 1-13C xylose was performed. After a preculture in YP medium with glycerol followed by an overnight culture in xylose medium for adaption, K. phaffii X-33 cultures were grown in triplicates in minimal medium with 12C or 1-13C xylose as carbon source for 10 days before sampling and quenching of biomass for metabolite analysis. The total 13C content of the biomass was measured by Elemental Analysis—Isotope Ratio Mass Spectrometry (EA-IRMS). For the samples grown in medium with 1-13C xylose, an average total 13C content of 9% was measured, which is slightly higher than the expected value of approximately 5% calculated from the biomass increase during the cultivation. As expected, the total 13C content of the samples grown with 12C xylose was around 1.1%, which represents the exact natural 13C abundance [27].

The incorporation of 13C from xylose into the metabolites of the pentose phosphate pathway and glycolysis was confirmed by gas chromatography (GC)—time of flight mass spectrometry (TOFMS) mass spectrometry isotopologue distribution analysis. An overview of the core metabolism including the putative xylose assimilation reactions is shown in Fig. 7A. The labeling pattern of relevant metabolites confirmed the incorporation of 13C from xylose into intermediates of the non-oxidative pentose phosphate pathway (Fig. 7B, Additional file 1: Table S5) as well as glucose, glucose-6-phosphate, 3-phosphoglycerate, and 2-phosphoglycerate as intermediates of glycolysis. Hardly any 13C label could be detected in 6-phosphoglycerate, showing that the carbon is entering the pentose phosphate pathway at the level of pentoses, while the oxidative part of the pathway is less active when cells are grown on xylose. As expected, a very low 13C content was measured for metabolites from control cultures grown on standard 12C xylose. For comparison, metabolic fluxes were also calculated using a previously published stoichiometric model of the central carbon metabolism with the addition of the xylose utilization reactions (Fig. 7A, Additional file 2) [28]. The prediction showed no metabolic flux through the oxidative pentose phosphate pathway, which corresponds to the results obtained from the 1-13C labeling experiment, where very little 13C was detected in 6-phosphoglycerate. A deviation from the model was only observed for glucose-6-phosphate and glucose, where the label was clearly detected in the metabolite analysis.

Fig. 7
figure 7

Xylose assimilation in K. phaffii X-33—13C labeling and modeling of metabolic fluxes. A Putative xylose utilization pathway. The isotopologue distribution of all colored metabolites was measured by mass spectrometry. The colors indicate the pathway to which the metabolites mainly belong (orange—glycolysis, green—oxidative pentose phosphate pathway, blue—non-oxidative pentose phosphate pathway). Absolute specific flux rates per carbon atom (mmol g−1 h −1) with 95% confidence intervals were calculated by 13C metabolic flux analysis. Values of each reaction refer to the total number of carbon atoms involved in the corresponding reaction. For reversible fluxes, only net fluxes are shown. B Relative abundance of 13C labeled isotopologues in selected metabolites after cultivation on 12C or 1-13C D-xylose

Discussion

This study provides an overview of the genetic and phenotypic diversity of Komagataella strains of all seven described species available from public strain collections. Genome sequencing of eight different strains by PacBio whole genome sequencing resulted in high-quality genome assemblies, including the first published whole genome sequences of strains belonging to K. kurtzmanii, K. mondaviorum, and K. ulmi. Phylogenetic analysis based on a representative set of subsequences overall confirmed the phylogenetic relationship between the species as determined by marker gene sequences [1]. However, the species K. mondaviorum, K. populi, and K. pseudopastoris showed a larger distance to the subtrees of K. phaffii and K. pastoris than in the analysis based on marker genes. The newly sequenced K. phaffii and K. kurtzmanii, and K. pastoris and K. ulmi strains, respectively, share the global differences in genome structure previously described for K. phaffii and K. pastoris [5]. The small number of chromosomal rearrangements between these four species allowed the assignment of chromosome number and orientation based on the published reference genomes. While the number of rearrangements in the other three species was too high for chromosome assignment based on the K. phaffii reference genome, pairwise comparison of the genome structure showed no large rearrangements between K. pseudopastoris and K. populi, while the extent of rearrangement between K. populi and K. mondaviorum was found to be higher than between K. phaffii and K. pastoris. Overall, the phylogenetic relationship of the species as shown in Fig. 1 is consistent over the different types of genome comparisons performed within this study.

The phenotypic diversity within the genus was analyzed by growth assays on different carbon sources and different stress conditions. It has previously been suggested that different species of Komagataella cannot be distinguished based on growth tests alone [1]. Accordingly, the observed growth was very similar for most of the conditions tested, especially for common carbon sources like glucose, glycerol, and methanol. Only on ethanol plates, some strains stood out as better growers at 30 °C. Clear differences could also be observed in tolerance to high pH and oxidative stress conditions. For all the strains growth was reduced at 37 °C, but only K. kurtzmanii CBS 12817 and K. mondaviorum CBS 15017 were unable to grow at this temperature in any of the conditions tested, as previously described for growth on glucose [1, 24]. K. populi CBS 12362 stood out as one of the fastest growing strains on all carbon sources and in most stress conditions tested. It will be interesting to investigate if these phenotypic characteristics are strain or species specific as soon as there are more isolates available.

The most intriguing observation from the growth assays was the slow but significant cell growth on minimal medium with xylose as the sole carbon source. Generally, Komagataella species are not known to utilize xylose and so far, growth on xylose has only been characterized for one K. phaffii strain in complex medium or after the introduction of a heterologous xylose isomerase pathway [11]. Detailed analysis of cell growth in liquid minimal medium showed long doubling times of around 10 days. Within this time, all three strains consumed xylose, with K. phaffii X-33 consuming more than 95% of the xylose in the medium while K. populi CBS12362 showed slightly faster growth with lower xylose consumption. The formation of xylitol as a byproduct was similar in all three strains and indicated the assimilation of xylose through an oxidoreductase pathway as described for other xylose consuming yeasts like S. stipitis [13, 29]. Using homology-based search, orthologs of xylose reductase, xylitol dehydrogenase and xylulokinase could be identified in all Komagataella species. Interestingly, many yeast species not known to utilize xylose have one or more copies of xylose pathway genes in their genomes, indicating that efficient utilization of xylose also depends on other factors like gene regulation and enzyme activity [20]. In K. phaffii X-33 and K. populi CBS 12362, the expression of putative xylose pathway genes, except for K. phaffii X-33 XKS1, was upregulated in xylose medium. Also, xylose reductase and xylitol dehydrogenase activities were measurable in cell free extracts, with activities around 100 times lower than in the xylose-utilizing yeast S. lignohabitans [30]. Upregulation of GRE3 and XKS1 gene expression and enzyme activity in the presence of xylose were both slightly higher in K. populi CBS 12362, which might explain the faster growth. Labeling experiments and metabolite analysis with 13C xylose showed high levels of 13C incorporation into metabolites of the non-oxidative pentose phosphate pathway and glycolysis, confirming the utilization of xylose via the oxidoreductase pathway. Further experiments are required to clearly determine the regulation of xylose utilization in Komagataella. Also, it remains to be tested if the overexpression of the native xylose utilization pathway genes can improve the growth to allow production of recombinant proteins or metabolites with xylose as carbon source.

Overall, our data on the phenotypic and genotypic diversity will provide the basis for use of so-far neglected Komagataella isolates with interesting characteristics and elucidation of the genetic determinants of improved growth and stress tolerance for targeted strain improvement.

Materials and methods

Yeast strains

All Komagataella isolates used in this study are listed in Table 4. All strains are available from public culture collections and were obtained from the Westerdijk Fungal Biodiversity Institute (CBS-KNAW, Netherlands), the German Collection of Microorganisms and Cell Cultures (DSMZ, Germany), and the University of Western Ontario, Department of Biology Yeast Culture Collection (UWOPS, Canada). The K. phaffii CBS 7435 gre3Δ strain was generated using a GoldenPiCS CRISPR/Cas9 vector with a gRNA targeting GRE3 (PP7435_Chr3-0448) and a homology template for complete removal of the coding sequence [31, 32]. The deletion of GRE3 was confirmed by PCR.

Table 4 List of strains used in this study

Genome sequencing

The strains selected for whole-genome sequencing are indicated in Table 4. Cells were grown in YPD medium (yeast extract 1%, peptone 2%, glucose 2%) overnight and genomic DNA was extracted using the Genomic-tip 100/G kit (Qiagen) according to the manufacturer’s protocol. Sequencing libraries were prepared with the SMRTbell Express Template Prep Kit 2.0 together with the Barcoded Overhang Adapter Kits 8A/8B (PacBio) and sequenced on a PacBio Sequel RSII instrument. The genomes were assembled with the HGAP4-pipeline included in PacBio SMRTLink software v8.0 using standard parameters [34]. Library preparation, sequencing, and genome assembly were performed by the Next Generation Sequencing Facility at Vienna BioCenter Core Facilities (VBCF), a member of the Vienna BioCenter (VBC), Austria. Sequencing reads and genome assemblies are available from NCBI database under the BioProject accession PRJNA770850. Mitochondrial sequences as identified through the NCBI Contamination Screen were removed from the assemblies.

Comparative genome analysis

As reference genomes, we used the assemblies of K. phaffii CBS 7435 (NCBI identifier ASM170808v1) and K. pastoris CBS 704 (ASM170810v1). Genome comparisons were performed using minimap2 v2.12-r827 [35] with default parameters. Matches with mapping quality > 0 and spanning ≥ 1 kbp were kept for further analysis. Circular plots were generated using Circos v0.69-8 [36] based on the minimap2 comparisons whereby only matches ≥ 10 kbp were displayed. Contigs without a match were removed. To assess the sequence divergence, the eight assembled genomes were matched against K. phaffii CBS 7435 as reference using nucmer v3.1 with parameter -g 1000, delta-filter with parameters -l 100 -o 0 -q, and show-coords with parameters -qldcoHT from the mummer suite [37]. Only matches of length 1 kbp or larger were evaluated. For calculation of % sequence identity per chromosome the sequence identity in each match region was multiplied by the match length, and the sum for all match regions within one chromosome was divided by the total match length per chromosome. Orthologous Komagataella genes of XP_001385181.1 (XR), XP_001386982.1 (XDH), and XP_001387325.2 (XKS) from Scheffersomyces stipitis were identified by searching the eight assemblies using tblastn v2.12.0 [38] followed by manual selection of start- and stop-codons. The coding sequences of the identified genes were translated using standard genetic code and used as input for multiple sequence alignments with Clustal Omega [39].

The genome of Citeromyces matritensis NRRL Y-2407 (NCBI id ASM370516v1) was used as outgroup in the phylogenetic analysis. Pairwise distances were determined between all genome assemblies employing the dist function implemented in Mash [22]. The distance-based phylogenetic tree was calculated using the Fitch algorithm [40] included in PHYLIP [41] with ten times jumbling (jumble seed 23893) and global rearrangements. For tree display, we used the Newick utilities toolkit [42]. Custom scripting and analyses were performed using Perl 5 and bash. The read length distribution (Additional file 1: Figure S1) was plotted with R v3.6.0 [43].

Spotting assays

For spotting assays, fresh colonies from glycerol stocks were resuspended in PBS or grown in YPD medium (yeast extract 1%, peptone 2%, glucose 2%) overnight. Cells were washed twice and diluted in PBS to OD600 0.3 in a 96-well plate. Using a multichannel pipette, five series of 1:10 dilutions were done (up to 1:105). Starting from the highest dilution, 4 µL of each dilution were pipetted onto YNB agar plates (YNB without amino acids and ammonium sulfate 3.4 g L−1, ammonium sulfate 10 g L−1, potassium phosphate buffer 0.1 mol L−1 pH 6, biotin 0.4 mg L−1, agar–agar 2%) supplemented with different carbon sources and/or inhibitors. To test growth on different carbon sources, plates were supplemented with 2% glucose, glycerol, methanol, or xylose. For inhibitor tests, YNB agar with 2% glucose was used. The final concentrations of inhibitors were: NaCl 1 mol L−1, H2O2 0.5 and 1 mmol L−1, acetic acid 2 g L−1. Plates were incubated at 30 or 37 °C for 168 h. Pictures were taken every 24 h on a scanner EPSON® perfection V750 PRO.

Growth assays in shake flask

Fresh colonies from glycerol stocks were used to inoculate 50 mL YPG (yeast extract 1%, peptone 2%, glycerol 2%) and incubated at 25 °C and 180 rpm overnight. Cells from the glycerol pre-cultures were washed twice and used to inoculate 25 mL of YNB medium (YNB without amino acids and ammonium sulfate 3.4 g L−1, ammonium sulfate 10 g L−1, potassium phosphate buffer 0.1 mol L−1 pH 6, biotin 0.4 mg L−1) supplemented with 2% xylose or no carbon source at a starting OD600 of 12. Cultures were incubated at 25 °C and 180 rpm for 10 days (240 h). A mixture of 50 U mL−1 penicillin and 50 µg mL−1 streptomycin (Gibco) were added to the medium to prevent contamination. Cell growth was monitored by measuring OD600 and the cell dry mass was determined from the pre-culture and at the end of the cultivation. Xylose consumption and metabolite profiles were monitored by HPLC. For the 13C labeling experiment, cells from pre-cultures in YPG were incubated in 30 mL YNB with 2% xylose overnight and then used to inoculate 10 mL YNB with 2% 12C or 1-13C D-xylose (Sigma-Aldrich) at a starting OD600 of 9. Cells were grown at 25 °C and 180 rpm for ten days and samples for OD600, cell dry mass, and HPLC analysis were taken at the beginning and end of the cultivation. All growth assays were performed in triplicates.

Transcript level analysis

For transcript level analysis, three clones per strain were cultivated in YPD medium (yeast extract 1%, peptone 2%, glucose 2%) in shake flasks overnight. Cells were washed and used to inoculate YNB medium (YNB without amino acids and ammonium sulfate 3.4 g L−1, ammonium sulfate 10 g L−1, potassium phosphate buffer 0.1 mol L−1 pH 6, biotin 0.4 mg L−1) with 2% glucose or xylose as carbon source. After 6 h of cultivation at 25 °C, cells were harvested and RNA was extracted according to the TRI reagent (Sigma-Aldrich) protocol, followed by DNase treatment with the Ambion DNA-free kit (Invitrogen) and cDNA synthesis with oligo(dT)23 primers (NEB) and the Biozym cDNA synthesis kit. RNA integrity was analyzed by agarose gel electrophoresis. Quantitative PCR was performed on a Rotor-Gene Q instrument (Qiagen) using the Blue S´Green qPCR kit (Biozym). Transcript levels were normalized to ACT1 expression and relative expression levels were calculated using the average expression of SOR1 in K. phaffii X-33 grown in YNB glucose as reference.

Enzyme activity assays

Fresh colonies grown in YPD (yeast extract 1%, peptone 2%, glucose 2%) overnight were harvested, washed, and used to inoculate YNB media (YNB without amino acids and ammonium sulfate 3.4 g L−1, ammonium sulfate 10 g L−1, potassium phosphate buffer 0.1 mol L−1 pH 6, biotin 0.4 mg L−1) supplemented with 2% xylose at a starting OD600 of 10. After 24 h of incubation at 25 °C, cells were harvested and stored at − 20 °C. For protein extraction, approximately 108 cells per sample were washed with PBS and suspended in cell lysis buffer (HEPES 20 mmol L−1, NaCl 420 mmol L−1, MgCl2*6H2O 1.5 mmol L−1, pH 8) with protease inhibitor (SIGMAFAST protease inhibitor cocktail tablets, Sigma-Aldrich). Cell lysis was performed by mechanical cell disruption using glass beads. Cell lysates were centrifuged (19,000g, 4 °C, 30 min) and the supernatants were stored on ice. The total protein content of the cell-free extracts was determined using a standard Bradford assay with BSA as a standard (Quick Start Bradford 1 × Dye Reagent, BioRad).

The enzyme activity assays were performed at room temperature in a total volume of 250 µL in 96-well plates as described previously [44]. For xylose reductase activity, the reaction mixture contained 20 µL cell-free extract in triethanolamine buffer (triethanolamine 100 mmol L−1, pH 7), 0.2 mmol L−1 NADH or NADPH, and 350 mmol L−1 xylose. For the xylitol dehydrogenase reaction, 0.5 mmol L−1 NAD+ and 300 mmol L−1 xylitol were used. Fifteen min after the addition of the cofactor, the reactions were started by the addition of the substrate and monitored by measuring the change in absorption at 340 nm. Specific activities were calculated in units per mg of total protein (U mg−1). One enzyme unit was defined as the amount of enzyme required to oxidize or reduce 1 μmol of cofactor per minute.

13C content of biomass

For 13C content determination in the yeast biomass, samples were taken at the end of the cultivation and washed twice with HCl 0.1 mol L−1 and ultrapure water. Cell pellets of approximately 10 mg of dry biomass were stored at – 20 °C until analysis. The 13C/12C content of the yeast biomass was analyzed by Imprint Analytics GmbH (Austria) by elemental analysis with isotope ratio mass spectrometry (EA-IRMS).

GC-TOFMS isotopologue distribution analysis of intracellular metabolites

Sampling and quenching of biomass samples for metabolome analysis were performed as described previously [45]. For the determination of the 13C labeling patterns of the metabolites under investigation the method of Mairinger et al. [46] was developed further. Briefly, aliquots of 180 µL (corresponding to 1.7 µg CDW) were dried in 1.5 mL chromatography vials after the addition of 20 µL of a solution of ethoxyamine hydrochloride in pyridine (c = 20 mg mL−1 pyridine) for protection of the keto and aldehyde carbonyl groups. The dried samples were placed on a cooled rack (7 °C) of the Gerstel MPS2 GC-(Q)TOFMS derivatization robot and autosampler equipped with trays, syringes, and heated agitators for automated two-step derivatization and just in time online injection. Derivatization was performed as follows: ethoximation using 18 µL of O-ethylhydroxylamine hydrochloride in water free pyridine (c = 19 mg mL−1) (Sigma Aldrich) at 40 °C for 90 min, followed by trimethylsilylation using 42 µL of N-methyl-N-(trimethylsilyl) trifluoroacetamide + 1% trimethylchlorosilane (Thermo Scientific) at 40 °C for 50 min. One µL aliquots of the derivatized samples were injected into an Agilent split/splitless injector with an Ultra Inert straight splitless liner in splitless mode (280 °C, septum purge flow 3 mL min−1). The following column setup was used for gas chromatographic separation on an Agilent 7890B GC (Agilent Technologies Inc.): (1) a non-polar deactivated pre-column (3 m × 0.25 µm i.d., Phenomenex), followed by a backflush unit (purged ultimate union, Agilent Technologies Inc.) for connection with (2), the analytical column (Optima® 1 MS Accent (100% dimethylpolysiloxane, 60 m × 250 μm × 0.25 μm, Macherey Nagel), which was connected via a second purged ultimate union T-piece to (3), a non-polar deactivated restrictor column (3 m × 0.18 µm i.d., Phenomenex). Helium 5.0 served as carrier gas and a constant flow of 1.1 mL min−1 for the pre-column, 1.2 mL min−1 for the analytical column and 1.4 mL min−1 for the post column was used. In contrast to Mairinger et al. [46] the following GC-temperature program with a lower initial ramp was employed (total run time 33.2 min): 70 °C for 1 min, 15 °C min−1–190 °C, 5 °C min−1–225 °C, 3 °C min−1–260 °C, 20 °C min−1–310 °C, hold for 3 min. The transfer line temperature was set to 280 °C.

An Agilent Technologies 7200B Q-TOF mass spectrometer (Agilent Technologies Inc.) was operated in positive chemical ionization mode using methane as reagent gas employing the settings described previously [47]. Retention times of the metabolites under investigation, mass/charge ratios evaluated for 13C isotopologue distribution analysis, and the respective extraction windows are listed in Additional file 1: Table S6.

Peak areas were integrated in Mass Hunter Workstation, Quantitative Analysis (Version 10.1, Agilent Technologies Inc.). In order to obtain the isotopologue distribution of the carbon backbone, integrated peak areas were then corrected for naturally distributed heavy stable isotopes. For that, results were processed using the ICT correction toolbox [48], which corrects for H, N, O, Si, and S isotopes of the derivatized molecule as well as the C isotopes stemming from derivatization (i.e., all C atoms apart from backbone C atoms). Quality control was performed using a cell extract of K. phaffii fed with 50:50 12CH3OH: 13CH3OH (Isotopic Solution, Austria), leading to a carbon isotopologue distribution following the Pascal’s Triangle [49]. The relative abundance RAi of each isotopologue i was calculated based on the ICT corrected Peas Areas Ai according to Eq. 1 (n = number of carbon atoms in metabolite backbone).

$${\mathrm{RA}}_{\mathrm{i}} \left[\mathrm{\%}\right]=\frac{{\mathrm{A}}_{\mathrm{i}}}{{\sum }_{\mathrm{i}=0}^{\mathrm{n}}{\mathrm{A}}_{\mathrm{i}}}$$
(1)

Modeling of metabolic fluxes

The stoichiometric model (Additional file 2) used for the calculation of intracellular fluxes was based on a previously published model of K. phaffii central carbon metabolism [28]. For xylose utilization via the oxidoreductase pathway and transport of xylose and xylitol 5 reactions were added (Additional file 2, R25, R48-50, R59). OpenFLUX with standard settings was used for flux calculation and a Monte Carlo algorithm was applied for sensitivity analysis to calculate the confidence intervals (95%) [50]. Average values from GC-TOFMS isotopologue distribution analysis with corresponding standard deviations were used as input data for the flux calculations. All calculations were done with MATLAB (R2019ba, The MathWorks, Inc., Natick, Massachusetts, USA).

Availability of data and materials

The datasets supporting the conclusions of this article are included within the article and its additional files. Whole genome sequencing reads and genome assemblies are available from NCBI database under BioProject accession PRJNA770850.

References

  1. Naumov GI, Naumova ES, Boundy-Mills KL. Description of Komagataella mondaviorum sp. nov., a new sibling species of Komagataella (Pichia) pastoris. Antonie Van Leeuwenhoek. 2018;111:1197–207.

    Article  CAS  PubMed  Google Scholar 

  2. De Schutter K, Lin YC, Tiels P, Van Hecke A, Glinka S, Weber-Lehmann J, et al. Genome sequence of the recombinant protein production host Pichia pastoris. Nat Biotechnol. 2009;27:561–6.

    Article  PubMed  CAS  Google Scholar 

  3. Küberl A, Schneider J, Thallinger GG, Anderl I, Wibberg D, Hajek T, et al. High-quality genome sequence of Pichia pastoris CBS7435. J Biotechnol. 2011;154:312–20.

    Article  PubMed  CAS  Google Scholar 

  4. Sturmberger L, Chappell T, Geier M, Krainer F, Day KJ, Vide U, et al. Refined Pichia pastoris reference genome sequence. J Biotechnol. 2016;235:121–31.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  5. Love KR, Shah KA, Whittaker CA, Wu J, Bartlett MC, Ma D, et al. Comparative genomics and transcriptomics of Pichia pastoris. BMC Genom. 2016;17:550.

    Article  CAS  Google Scholar 

  6. Valli M, Tatto NE, Peymann A, Gruber C, Landes N, Ekker H, et al. Curation of the genome annotation of Pichia pastoris (Komagataella phaffii) CBS7435 from gene level to protein function. FEMS Yeast Res. 2016;16(6):fow051.

    Article  PubMed  CAS  Google Scholar 

  7. Braun-Galleani S, Dias JA, Coughlan AY, Ryan AP, Byrne KP, Wolfe KH. Genomic diversity and meiotic recombination among isolates of the biotech yeast Komagataella phaffii (Pichia pastoris). Microb Cell Fact. 2019;18(1):211.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  8. Brady JR, Whittaker CA, Tan MC, Kristensen DL, Ma D, Dalvie NC, et al. Comparative genome-scale analysis of Pichia pastoris variants informs selection of an optimal base strain. Biotechnol Bioeng. 2020;117:543–55.

    Article  CAS  PubMed  Google Scholar 

  9. Zahrl RJ, Peña DA, Mattanovich D, Gasser B. Systems biotechnology for protein production in Pichia pastoris. FEMS Yeast Res. 2017;17(7):fox068.

    Article  CAS  Google Scholar 

  10. Cregg JM, Cereghino JL, Shi J, Higgins DR. Recombinant protein expression in Pichia pastoris. Mol Biotechnol Mol Biotechnol. 2000;16(1):23–52.

    Article  CAS  PubMed  Google Scholar 

  11. Li P, Sun H, Chen Z, Li Y, Zhu T. Construction of efficient xylose utilizing Pichia pastoris for industrial enzyme production. Microb Cell Fact. 2015;14:22.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  12. Araújo D, Freitas F, Sevrin C, Grandfils C, Reis MAM. Co-production of chitin-glucan complex and xylitol by Komagataella pastoris using glucose and xylose mixtures as carbon source. Carbohydr Polym. 2017;166:24–30.

    Article  PubMed  CAS  Google Scholar 

  13. Zhao Z, Xian M, Liu M, Zhao G. Biochemical routes for uptake and conversion of xylose by microorganisms. Biotechnol Biofuels. 2020;13:21.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  14. Sun L, Jin YS. Xylose Assimilation for the efficient production of biofuels and chemicals by engineered Saccharomyces cerevisiae. Biotechnol J. 2021;16(4):e2000142.

    Article  PubMed  CAS  Google Scholar 

  15. Veras HCT, Parachin NS, Almeida JRM. Comparative assessment of fermentative capacity of different xylose-consuming yeasts. Microb Cell Fact. 2017;16:153.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  16. Amore R, Kötter P, Küster C, Ciriacy M, Hollenberg CP. Cloning and expression in Saccharomyces cerevisiae of the NAD(P)H-dependent xylose reductase-encoding gene (XYL1) from the xylose-assimilating yeast Pichia stipitis. Gene. 1991;109:89–97.

    Article  CAS  PubMed  Google Scholar 

  17. Kötter P, Amore R, Hollenberg CP, Ciriacy M. Isolation and characterization of the Pichia stipitis xylitol dehydrogenase gene, XYL2, and construction of a xylose-utilizing Saccharomyces cerevisiae transformant. Curr Genet. 1990;18:493–500.

    Article  PubMed  Google Scholar 

  18. Almeida JRM, Bertilsson M, Hahn-Hägerdal B, Lidén G, Gorwa-Grauslund MF. Carbon fluxes of xylose-consuming Saccharomyces cerevisiae strains are affected differently by NADH and NADPH usage in HMF reduction. Appl Microbiol Biotechnol. 2009;84:751–61.

    Article  CAS  PubMed  Google Scholar 

  19. Jin YS, Jones S, Shi NQ, Jeffries TW. Molecular cloning of XYL3 (D-xylulokinase) from Pichia stipitis and characterization of its physiological function. Appl Environ Microbiol. 2002;68:1232–9.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  20. Riley R, Haridas S, Wolfe KH, Lopes MR, Hittinger CT, Göker M, et al. Comparative genomics of biotechnologically important yeasts. Proc Natl Acad Sci USA. 2016;113:9882–7.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  21. Shen XX, Opulente DA, Kominek J, Zhou X, Steenwyk JL, Buh KV, et al. Tempo and mode of genome evolution in the budding yeast subphylum. Cell. 2018;175:1533-1545.e20.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  22. Ondov BD, Treangen TJ, Melsted P, Mallonee AB, Bergman NH, Koren S, et al. Mash: fast genome and metagenome distance estimation using MinHash. Genome Biol. 2016;17(1):132.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  23. Ohi H, Okazaki N, Uno S, Miura M, Hiramatsu R. Chromosomal DNA patterns and gene stability of Pichia pastoris. Yeast. 1998;14:895–903.

    Article  CAS  PubMed  Google Scholar 

  24. Naumov GI, Naumova ES, Tyurin OV, Kozlov DG. Komagataella kurtzmanii sp. nov., a new sibling species of Komagataella (Pichia) pastoris based on multigene sequence analysis. Antonie Van Leeuwenhoek. 2013;104:339–47.

    Article  CAS  PubMed  Google Scholar 

  25. Lee J-K, Koo B-S, Kim S-Y. Cloning and characterization of the xyl1 gene, encoding an NADH-Preferring xylose reductase from Candida parapsilosis, and its functional expression in Candida tropicalis. Appl Environ Microbiol. 2003;69:6179.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  26. Hou X. Anaerobic xylose fermentation by Spathaspora passalidarum. Appl Microbiol Biotechnol. 2012;94:205–14.

    Article  CAS  PubMed  Google Scholar 

  27. Meija J, Coplen TB, Berglund M, Brand WA, De Bièvre P, Gröning M, et al. Isotopic compositions of the elements 2013 (IUPAC Technical Report). Pure Appl Chem. 2016;88(3):293–306.

    CAS  Google Scholar 

  28. Baumann K, Carnicer M, Dragosits M, Graf AB, Stadlmann J, Jouhten P, et al. A multi-level study of recombinant Pichia pastoris in different oxygen conditions. BMC Syst Biol. 2010;4:141.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  29. Jagtap SS, Rao CV. Microbial conversion of xylose into useful bioproducts. Appl Microbiol Biotechnol. 2018;102(21):9015–36.

    Article  CAS  PubMed  Google Scholar 

  30. Bellasio M, Mattanovich D, Sauer M, Marx H. Organic acids from lignocellulose: Candida lignohabitans as a new microbial cell factory. J Ind Microbiol Biotechnol. 2015;42:681–91.

    Article  CAS  PubMed  Google Scholar 

  31. Prielhofer R, Barrero JJ, Steuer S, Gassler T, Zahrl R, Baumann K, et al. GoldenPiCS: A Golden Gate-derived modular cloning system for applied synthetic biology in the yeast Pichia pastoris. BMC Syst Biol. 2017;11(1):123.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  32. Gassler T, Heistinger L, Mattanovich D, Gasser B, Prielhofer R. CRISPR/Cas9-mediated homology-directed genome editing in Pichia pastoris. Methods Mol Biol. 2019;1923:211–25.

    Article  CAS  PubMed  Google Scholar 

  33. Guilliermond A. Zygosaccharomyces pastori, nouvelle espèce de levures à copulation hétérogamique. Bull la Société Mycol Fr. 1920;36:203–11.

    Google Scholar 

  34. Chin CS, Alexander DH, Marks P, Klammer AA, Drake J, Heiner C, et al. Nonhybrid, finished microbial genome assemblies from long-read SMRT sequencing data. Nat Methods. 2013;10:563–9.

    Article  CAS  PubMed  Google Scholar 

  35. Li H. Minimap2: pairwise alignment for nucleotide sequences. Bioinformatics. 2018;34:3094–100.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  36. Krzywinski M, Schein J, Birol I, Connors J, Gascoyne R, Horsman D, et al. Circos: an information aesthetic for comparative genomics. Genome Res. 2009;19:1639–45.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  37. Kurtz S, Phillippy A, Delcher AL, Smoot M, Shumway M, Antonescu C, et al. Versatile and open software for comparing large genomes. Genome Biol. 2004;5(2):R12.

    Article  PubMed  PubMed Central  Google Scholar 

  38. Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. Basic local alignment search tool. J Mol Biol. 1990;215(3):403–10.

    Article  CAS  PubMed  Google Scholar 

  39. Sievers F, Wilm A, Dineen D, Gibson TJ, Karplus K, Li W, et al. Fast, scalable generation of high-quality protein multiple sequence alignments using Clustal Omega. Mol Syst Biol. 2011;7:539.

    Article  PubMed  PubMed Central  Google Scholar 

  40. Fitch WM. Toward defining the course of evolution: minimum change for a specific tree topology. Syst Biol. 1971;20:406–16.

    Article  Google Scholar 

  41. Felsenstein J. PHYLIP—phylogeny inference package (version 3.2). Cladistics. 1989;5:164–6.

    Google Scholar 

  42. Junier T, Zdobnov EM. The Newick utilities: high-throughput phylogenetic tree processing in the UNIX shell. Bioinformatics. 2010;26:1669–70.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  43. R Core Team. R: a language and environment for statistical computing. Vienna: R Found Statistical Computing; 2019.

    Google Scholar 

  44. Trichez D, Steindorff AS, Soares CEVF, Formighieri EF, Almeida JRM. Physiological and comparative genomic analysis of new isolated yeasts Spathaspora sp. JA1 and Meyerozyma caribbica JA9 reveal insights into xylitol production. FEMS Yeast Res. 2019;19(4):foz034.

    Article  PubMed  CAS  Google Scholar 

  45. Russmayer H, Troyer C, Neubauer S, Steiger MG, Gasser B, Hann S, et al. Metabolomics sampling of Pichia pastoris revisited: rapid filtration prevents metabolite loss during quenching. FEMS Yeast Res. 2015;15(6):fov049.

    Article  PubMed  CAS  Google Scholar 

  46. Mairinger T, Steiger M, Nocon J, Mattanovich D, Koellensperger G, Hann S. Gas chromatography-quadrupole time-of-flight mass spectrometry-based determination of isotopologue and tandem mass isotopomer fractions of primary metabolites for 13C-metabolic flux analysis. Anal Chem. 2015;87:11792–802.

    Article  CAS  PubMed  Google Scholar 

  47. Chu DB, Troyer C, Mairinger T, Ortmayr K, Neubauer S, Koellensperger G, et al. Isotopologue analysis of sugar phosphates in yeast cell extracts by gas chromatography chemical ionization time-of-flight mass spectrometry. Anal Bioanal Chem. 2015;407:2865–75.

    Article  CAS  PubMed  Google Scholar 

  48. Jungreuthmayer C, Neubauer S, Mairinger T, Zanghellini J, Hann SICT. Isotope correction toolbox. Bioinformatics. 2016;32:154–6.

    CAS  PubMed  Google Scholar 

  49. Millard P, Massou S, Portais JC, Létisse F. Isotopic studies of metabolic systems by mass spectrometry: using pascal’s triangle to produce biological standards with fully controlled labeling patterns. Anal Chem. 2014;86:10288–95.

    Article  CAS  PubMed  Google Scholar 

  50. Quek LE, Wittmann C, Nielsen LK, Krömer JO. OpenFLUX: efficient modelling software for 13C-based metabolic flux analysis. Microb Cell Fact. 2009;8:25.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

Download references

Acknowledgements

We are grateful to Marc-Andre Lachance for providing the Komagataella strains from the University of Western Ontario, Department of Biology Yeast Culture Collection (UWOPS) and for critical reading of the manuscript. Also, we thank Viktoria Kowarz for her help with transcript level analysis, Philipp Tondl for his support with mass spectrometry measurements and Hannes Russmayer for help with metabolite extraction and HPLC analysis.

Funding

LH and DK were funded by the Christian Doppler Research Association (Christian Doppler Laboratory for Innovative Immunotherapeutics) and Merck KGaA. BGP was supported by an Ernst Mach Worldwide grant (ICM-2019-13299) by the Austrian Agency for Education and Internationalization (OeAD) and a scholarship from Coordination for The Improvement of Higher Education (CAPES—Brasil). ÖA was supported by the Austrian Science Fund FWF (grant M2891) and by BMK, BMDW, SFG, Standortagentur Tirol, Government of Lower Austria und Vienna Business Agency in the framework of COMET—Competence Centers for Excellent Technologies managed by the Austrian Research Promotion Agency FFG. This project was supported by EQ-BOKU VIBT GmbH and the BOKU Core Facility Mass Spectrometry.

Author information

Authors and Affiliations

Authors

Contributions

LH and DM planned and designed the study. LH, BGP and DK conducted the growth experiments and enzyme assays and analyzed the data. JCD performed the comparative genome data analysis. CT and TSM developed the mass spectrometry based analytical techniques for the determination of 13C isotopologue distributions, measured and evaluated the data. ÖA performed the metabolic flux calculations. LH, BGP and JCD wrote the manuscript. All authors read and approved the final version of the manuscript.

Corresponding author

Correspondence to Lina Heistinger.

Ethics declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1: Figure S1.

Length distribution of PacBio subreads. Figure S2. Multiple sequence alignments of xylose pathway enzymes. Figure S3. Multiple sequence alignments of the K. phaffii CBS 7435 Gre3 homologs. Table S1. Pairwise distances between genomes of different Komagataella species. Table S2. Orthologs of S. stipitis xylose reductase (XR) in different Komagataella species. Table S3. Orthologs of S. stipitis xylitol dehydrogenase (XDH) in different Komagataella species. Table S4. Orthologs of S. stipitis xylulokinase (XKS) in different Komagataella species. Table S5. Relative abundance of isotopologues of selected metabolites after cultivation on 12C or 1-13C D-xylose. Table S6. List of metabolites and corresponding GC retention times (RT), chemical ionization adducts/fragments evaluated for 13C isotopologue distribution analysis.

Additional file 2:

Stoichiometric model of the central carbon metabolism plus xylose utilization reactions.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Heistinger, L., Dohm, J.C., Paes, B.G. et al. Genotypic and phenotypic diversity among Komagataella species reveals a hidden pathway for xylose utilization. Microb Cell Fact 21, 70 (2022). https://doi.org/10.1186/s12934-022-01796-3

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s12934-022-01796-3

Keywords