Phleomycin D1

Multinuclear Yeast Magnusiomyces (Dipodascus, Endomyces) magnusii is a Promising Isobutanol Producer

Olena O. Kurylenko, Justyna Ruchala, Kostyantyn V. Dmytruk, Charles A. Abbas, and Andriy A. Sibirny*

Abstract

Higher alcohol isobutanol is a promising liquid fuel. During alcoholic fermentation, Saccharomyces cerevisiae produces only trace amounts of isobutanol. Screening the collection of nonconventional yeasts show that Magnusiomyces magnusii accumulates 440 mg of isobutanol per L in rich YPD medium. Here, the transformation protocol for M. magnusii is adapted based on the use of the dominant markers conferring resistance to nourseothricin or zeocin; the strong constitutive promoter TEF1 is cloned and a reporter system based on LAC4 gene from Kluyveromyces lactis coding for 𝜷-galactosidase is constructed. In order to increase isobutanol production in M. magnusii, the heterologous gene ILV2 from S. cerevisiae is expressed in M. magnusii under control of the TEF1 promoter. The best stabilized transformants produce 620 mg of isobutanol per L in YPD medium and 760 mg L−1 in the medium with 2-oxoisovalerate. This suggests that M. magnusii is a promising organism for further development of a robust isobutanol producer. small due to the high cost of handling and processing of this feedstock and the lack of strains that effectively ferment all sugars released during lignocellulose hydrolysis that are resistant to the inhibitors produced during pretreatment.[1–3]
In spite of the current uses of ethanol as a biofuel, it has some serious drawbacks. The production of anhydrous ethanol requires a separate dehydration step following azeotropic distillation. Absolute ethanol can readily absorbs moisture from the air thereby causing pipe corrosion during transportation as well as corrosion of car engine and gas station pumps. Prevention of this requires the use of more expensive alloys in pipes and pumps and the use of additives to prevent moisture in ethanol

Keywords
acetolactate synthase, isobutanol, Magnusiomyces (Dipodascus Endomyces) magnusii, yeasts

1. Introduction

Large scale worldwide commercial production of the liquid biofuels, bioethanol, and biodiesel from renewable feedstocks, has led to a significant decrease in greenhouse gas emissions. In 2018, 120 billion liters of bioethanol was produced with over 100 billion liters used as fuel (https://knect365.com/energy/article/c 07f7fba-48fa-464f-9f21-12f913fc67f7/world-ethanol-productionto-expand-steadily-in-2019). Most of this ethanol was produced from conventional feedstocks (starch-derived glucose and sucrose) and is known as first generation ethanol. The production of second generation ethanol from lignocellulose is relatively blended fuels in impacting combustion performance of car engines. Ethanol contains relatively low energy density relative to gasoline and can be blended with gasoline at 15% in standard car engines but cannot be used in blends with diesel fuel. Higher alcohols, like butanol and isobutanol, have several advantages as biofuels when compared to ethanol. They are not hygroscopic, contain much higher energy density, similar to that of gasoline, and can be mixed with gasoline and diesel fuel in any ratio as well as used in jet fuels.[4–6] Higher alcohols also have other uses in chemical synthesis, and as solvents as well as additives to perfumes (http://www.biobut anol.com/Butanol-Isomers-isobutanol,-n-butanol,-tert-butanol. html).
Butanol (n-butanol) is produced by many species of Clostridia, in particular, by C. acetobutylicum and some engineered bacterial species.[7,8] There are also strains of genetically engineered Sacccharomyces cerevisiae[9–11] and Yarrowia lipolytica engineered that can produce butanol and isobutanol.[12] The highest published concentration of n-butanol produced by an engineered strain of S. cerevisiae has been reported at around 250 mg of butanol per L in the medium without feeding with butanol precursors, and when precursors are used is approaches 2.4 g of butanol per L.[9,11] Isobutanol is present as a normal component of fusel alcohols produced by S. cerevisiae during alcoholic fermentation from catabolism of the amino acid, valine. Unfortunately, yeast accumulates very low amounts of isobutanol, near 1000 times less than that of ethanol. During last decade several publications reported the construction of more active isobutanol-producing microorganisms. Metabolic engineering was successful in the construction of efficient isobutanol producing Escherichia coli,[13–15] Bacillus subtilis,[16] Corynebacterium glutamicum,[17,18] but less efficient in Zymomonas mobilis.[19] There are also several works on increased isobutanol production in S. cerevisiae that involved genes in valine biosynthesis from pyruvate like ILV2, ILV5, ILV3 (of native mitochondrial or cytosolic localization of the products) and those on the overexpression of aldehyde dehydrogenase ARO10 that led to the accumulation of up to 600 mg of isobutanol per L.[10,20–22] Deletion of BAT1 gene coding for mitochondrial branched chain amino acid transaminase, the elimination of valine and isoleucine biosynthetic pathways, the overexpression of genes coding for mitochondrial pyruvate carrier, or the heterologous expression of bacterial phosphosoenolpyruvate carboxylase also led to an increase in isobutanol production.[23–25] Substantial increase in isobutanol production was achieved by the elimination of the alternative pathway for pyruvate utilization by the deletion of LPD1 gene of pyruvate dehydrogenase complex in combination with the overexpression of genes coding for enzymes of transhydrogenase shunt (pyruvate carboxylase, malate dehydrogenase, and malic enzyme). These engineered strains accumulated 1.62 g of isobutanol per L.[26] The activation of isobutanol biosynthetic pathway together with elimination of the competing pathways led to accumulation of 2.09 g of isobutanol per L.[27]
Up till now, yeast diversity was not used widely to search for more active isobutanol producers though availability of such yeast species cannot be excluded. There are limited reports on isobutanol production by non-Saccharomyces yeasts. The wild type strain of the methylotrophic yeast Pichia pastoris (Komagataella phaffii) synthesizes negligible amounts of isobutanol. Overexpression of the endogenous enzymes of valine biosynthesis and Ehrlich pathway led to an accumulation of 2.22 g of isobutanol per L which is the highest titer described for yeast.[28] In the current work, we report that the wild strain of the multinuclear yeast Magnusiomyces (Diplodascus, Endomyces) magnusii accumulates significant amounts of isobutanol (0.44 g L−1). We also communicate on the development of the basic methods of molecular research for this species and construction of the advanced isobutanol producer which accumulates 0.62 g isobutanol per L and 0.76 g L−1 in the medium with 2-oxoisovalerate. We also found that the wild type strains of M. magnusii accumulates even higher amounts of n-butanol (up to 0.82 g L−1) during incubation with 2-oxovalerate.

2. Results

2.1. Isobutanol Synthesis by Selected Nonconventional Yeast Strains

We searched our collection of yeast strains for the production of ethanol and other higher alcohols. In total, 39 strains representing 22 genera were tested. The list of these strains is presented in Table S2, Supporting Information. Most of tested strains accumulated small or negligible amounts of higher alcohols. Only two of the 38 tested strains, accumulated significant amounts of higher alcohols, mostly isobutanol: M. magnusii and Saccharomycopsis fibuligera. S. fibuligera accumulated 76 mg of isobutanol per L whereas M. magnusii accumulated 440 mg of isobutanol per L. For comparison, one of the tested strains of S. cerevisiae (B3) accumulated under the same condition only 41 mg of isobutanol per L whereas, the other industrial ethanol producing strain, PE2, did not show even traces of isobutanol. In the case cells were cultivated in synthetic YNB medium without yeast extract, the amount of isobutanol produced were lower (up to 170 mg L−1), however the addition of 0.1% yeast extract slightly increased isobutanol production (up to 380 mg L−1). No accumulation of n-butanol during cultivation of M. magnusii in YPD medium was observed.

2.2. Development of Efficient Transformation Protocols for M. magnusii

M. magnusii is a unique yeast by forming large multinuclear cells. Due to this feature, M. magnusii is a favorite organism for cytological and biochemical (including experiments on isolated mitochondria) studies.[29–32] Molecular studies of M. magnusii are at the preliminary stage and genome has not been sequenced. Multinuclear nature of the cells makes isolation of recessive auxotrophic mutants practically impossible. Mitochondrial DNA of M. magnusii was isolated and characterized.[33] Electrophoretic DNA separation of M. magnusii indicate the presence of 13 DNA molecules with a total genome size of 38 Mb.[34] Using a dominant marker for zeocin resistance, a method of genetic transformation was developed.[35] Homologs of ADE2 and URA3 genes were isolated from the gene library of M. magnusii.[36] Very little is known on the biotechnological potential of M. magnusii. Earlier published works demonstrated that due to high fumarase activity, this yeast is an efficient L-malate producer from fumarate.[37,38]
After our initial observation that the wild type strain of M. magnusii produces a high concentration of isobutanol which compares favorably to other yeast as well as that produced by engineered yeast species, it was decided to construct more robust strains of this yeast for the production of isobutanol. It is known that isobutanol is synthesized in S. cerevisiae in combination of valine biosynthesis and Ehrlich pathway[39] and the overexpression of valine biosynthesis enzymes of mitochondrial or cytosolic localization enhanced isobutanol production.[20,21,40] The same approach was used to improve isobutanol synthesis in M. magnusii. As its genome was not sequenced, it was decided to perform the heterologous expression of S. cerevisiae native ILV2 gene (coding for mitochondrially located acetolactate synthase) and that without mitochondrial targeting signal (coding for presumably cytosolic enzyme) in M. magnusii with subsequent study of the production of isobutanol in the obtained transformants.
Unfortunately, poor genetic studies, multinuclear cells and lack of genome sequence hampered this goal. Therefore, the development of new basic molecular tools for M. magnusii was required. The previously described system for genetic transformation of M. magnusii was based on the plasmid carrying endogenous autonomously replicating sequence and the dominant selective marker, the gene conferring resistance to zeocin. Several attempts to obtain stable integrative transformants in M. magnusii were unsuccessful.[35] To improve the transformation efficiency, the protocol for transformation by electroporation described earlier was modified. The cells harvested in exponential phase were treated with LiAc/TE buffer for 30 min, then washed with water and resuspended in 1 m sorbitol. After electroporation (electric pulse of 5 kV cm−1 at 1000 Ω, 25 µF settings in Gene Pulser apparatus), cells were incubated in a YPD medium containing 1 m sorbitol for at least 5 h at 30 °C. The plasmid pUC57-LAC4-ZeoR containing gene ZeoR conferring resistance to zeocin (see Section 2.3) was AhdI-linearized and used for transformation. The concentrations of zeocin 0.8 g L−1 and 1 g L−1 were used for the selection of transformants. After transformation using plasmid pUC57-LAC4-zeo around ten colonies (transformation frequency was 3 transformants per µg of DNA) were observed on each YPD plate supplemented with 0.8 g L−1 and 1 g L−1 of zeocin. It was also found that M. magnusii is sensitive to low concentrations of nourseothricin (10 µg mL−1). However, to apply this selection marker, transformation protocol was further modified. 10 µg of DNA were used for each sample and cells after electroporation were incubated in the medium YPD with 1 m sorbitol for at least 6 h at 30 °C. After 4 days of incubation, transformants were selected on a YPD medium supplemented with 10 mg L−1 or 20 mg L−1 of nourseothricin.

2.3. Heterologous Expression of S. cerevisiae ILV2 Genes in M. magnusii

The first enzyme of isobutanol synthesis de novo is acetolactate synthase, encoded by ILV2 gene. The shortened version of ILV2 gene of S. cerevisiae without mitochondrial targeting signal under control of ADHI promoter of S. cerevisiae was integrated into the genome of M. magnusii. For this, the plasmid pUC57-ADHIILV2-natNT2[41] was SacI-linearized and used for transformation of M. magnusii. The transformants obtained were resistant to the selective agent after the transfer of cells to fresh YPD medium supplemented with 20 mg L−1 of nourseothricin, however, the presence of the corresponding plasmid in the genome of transformants was not definitely confirmed. Diagnostic PCR using primers Ko469 and Ko470 for natNT2 gene resulted in a weak positive signal. Obtained transformants were tested for their isobutanol production and compared to the parental M. magnusii strain. The efficiency of isobutanol production was analyzed under condition of full aeration (220 rpm) as well as under fermentation condition (120 rpm) using a YPD medium (see Section 2.4). Four recombinant strains were characterized by slightly increased isobutanol production on the first day of cultivation under semi-aerobic conditions when compared to the parental strain. The highest amount of isobutanol was synthetized by strain Em/ILV2_31 reaching 0.54 g L−1 demonstrating a 17% increase in isobutanol production as compared to wild type. In addition, almost all transformants produced slightly higher amount of isobutanol on the second day of cultivation (Figure S3, Supporting Information). It is interesting to note that all recombinant strains were characterized by higher glycerol production as compared to the parental strain. The observed phenomenon may be a result of the shortened ILV2 gene overexpression, as was shown previously in S. cerevisiae. We showed that overexpression of the shortened ILV2 gene resulted in decreasing the cellular content of pyruvate. This apparently led to NADH reduction of dihydroxyacetone phosphate (instead of acetaldehyde produced from pyruvate) to glycerol-3-phosphate which further dephosphorylated to glycerol.[41]

2.4. Cloning Strong Constitutive Promoter TEF1 M. magnusii

We speculated that further increase in isobutanol synthesis could be achieved in the case gene ILV2 by the overexpression of this gene under the control of a strong M. magnusii promoter.
To overexpress genes of interest, it is important to isolate a strong native promoter from M. magnusii, for example the TEF1 promoter of the gene responsible for translation of the elongation factor. Fortunately, the TEF1 coding sequence (ORF) is known (GenBank: JQ699093.1).[42] The isolation of the TEF1 promoter of M. magnusii was performed by inverse PCR. First, the primers Ko753 and Ko755 were designed for isolation of the M. magnusii TEF1 ORF. A corresponding fragment of expected size (894 bp) was amplified using these primers with total DNA of M. magnusii VKMY-1072asatemplate.TheinverseprimersKo757andKo759 complementary to M. magnusii TEF1 ORF region were designed. A range of restriction endonucleases were used to choose the appropriate one, which is located not far from a start codon of the TEF1 gene of M. magnusii. Genomic DNA of M. magnusii was digested with each of these restriction endonucleases, self-ligated, and the resulting DNA samples used as templates for PCR with the inverse primers Ko757 and Ko759. The ≈1kb fragment amplified in the inverse PCR, was fragment of M. magnusii genomic DNAdigestedwithSacIthatwasusedasatemplate.Theobtained fragment was cleaned up and sequenced. Sequencing of the obtained M. magnusii genomic fragment revealed 902 bp upstream from the start codon of the TEF1 gene. In the sequenced region, a second ATG start codon at position −300 bp was found. Efficient gene expression depends on recognition of specific promoter sequences by transcriptional regulatory proteins. These consensus sequences play an important role in promoter efficiency and determine its strength. In order to investigate further the necessary length of the promoter that can support the correct expression level for the target gene in M. magnusii and evaluate which ATG codon is responsible for the initiation of expression, two versions of TEF1 gene promoter were amplified by PCR (599 bp and 902 bp).
The study of gene expression was facilitated by using a reporter gene assay. The gene encoding 𝛽-galactosidase is one of the most popular reporter gene that has been used successfully for many organisms. A reporter system which includes the K. lactis LAC4 gene encoding 𝛽-galactosidase that was successfully applied in an earlier study of promoter activities in S. cerevisiae, Ogataea polymorpha, and Candida famata was used in this study.[43]
For comparison of the strength of M. magnusii TEF1 gene promoter with strong constitutive promoters of S. cerevisiae, plasmidsharboringreportergene LAC4 K. lactis drivenby S. cerevisiae TEF1 or ADH1 gene promoters were constructed (see Section 2.3 and Figure S1, Supporting Information). Subsequently, M. magnusii was transformed with these plasmids and transformants wereassayedfor𝛽-galactosidaseactivity.DataofTable1showthat short version of M. magnusii TEF1 promoter (pEmTEF1599) was the most efficient one for LAC4 gene expression. Similarly, the heterologous S. cerevisiae TEF1 promoter (pScTEF1) was found of comparable strength whereas the native M. magnusii TEF1 promoter (long version) and S. cerevisiae ADH1 promoters were very weak.

2.5. Heterologous Expression of S. cerevisiae ILV2 and ARO10 Genes in M. magnusii under Control of Promoter TEF1

As positive effect on isobutanol expression in S. cerevisiae had both overexpression of native and shortened versions of ILV2 gene (without mitochondrial targeting signal),[20,21,44] we decided to overexpress both versions of S. cerevisiae ILV2 gene under control of TEF1 promoter of M. magnusii. As overexpression of ARO10 gene coding for phenylpyruvate decarboxylase (in fact, broad specificity 2-ketoacid decarboxylase), the first specific enzyme of Ehrlich pathway is also useful for isobutanol overproduction in S. cerevisiae,[45] we decided to overexpress also S. cerevisiae ARO10 gene under control of M. magnusii TEF1 promoter in combinations with the overexpression of the native or truncated ILV2 gene. The schemes of the constructed plasmids are presented in Figure S2, Supporting Information.
In total, several tens of transformants of each type were analyzed for specific activity of acetolactate synthase and isobutanol production after their stabilization. Stabilization is the critical moment of handling M. magnusii transformants due to multinuclear character of cells. This was achieved by several rounds of transfer into selective and nonselective media. The results indicated a 25–60% increase in specific activities of the Ilv2 in M. magnusii transformants overexpressing native or shortened ILV2 gene of S. cerevisiae (Table 2), which suggests an efficient expression of heterologous ILV2 gene in M. magnusii. This conclusion is directly supported by qRT-PCR data on enhanced expression of S. cerevisiae ILV2 in M. magnusii transformants (Figure 1). The expression level of S. cerevisiae ILV2 in transformants of M. magnusii was close to that determined in the wild type strain of S. cerevisiae. This suggests that apparently each nucleus of M. magnusii transformants harbored the heterologous S. cerevisiae ILV2 gene. In the case only one (of 8 in average) nuclei received heterologous gene, gene expression would be near 12% of that in S. cerevisiae etc.
A good correlation between the increase in isobutanol production, activity of acetolactate synthase, and expression level of ILV2 gene was observed. It was found that in each case of transformants, there are those with elevated isobutanol production (Table 3). The best transformants had an increase in isobutanol accumulation of 38–53%. Positive effects of overexpression of native and truncated ILV2 genes were comparable. No positive effect of ARO10 overexpression was observed. The difference in biomass accumulation between tested recombinant strains and the wild type strain during fermentation was not observed (data not shown).

2.6. Isolation of the Advanced Isobutanol Producer Using Acetolactate Inhibitor Bispyribac

As indicated earlier, isobutanol synthesis starts by acetolactate synthase, which is also the first enzyme in the valine and isoleucine biosynthetic pathway. To the best of our knowledge, the increase in acetolactate synthase activity was achieved before only by the overexpression of ILV2 gene. Previous section also shows the success of this approach. Here we describe antimetabolite selection for enhancing activity of acetolactate synthase. It has been known for many years that acetolactate synthase is specifically inhibited by 2,6bis(4,6-dimethoxypyrimidin-2-yl)oxy)benzoate, which is commonly known by its trade name bispyribac and used as wide spectrum high effective herbicide.[46] There are many publications on bispyribac-resistant plants, which indicate target site and nontarget site for herbicide tolerance mutations.[47] Target-site mutations as a rule lead to modifications of acetolactate synthase protein.[48] However, bispyribac has not been used for the selection of the resistant mutants in yeast. We propose that although the majority of M. magnusii bispyribac-resistant mutants will possess mutations in acetolactate synthase protein, some resistant mutants could evolve due to regulatory mutations which lead to an increase in the quantity of the normal enzyme. If some determined bispyribac concentration totally inhibits acetolactate synthase activity, this inhibition could be only partial in the case of cells that synthesize more enzyme molecules.
The minimal inhibiting concentration of bispyribac in YNB medium for M. magnusii was established as 55 mm for cells that were plated onto different concentrations of herbicide. The resistant colonies were isolated and tested for isobutanol synthesis. The colonies resistant to sodium bispyribac were re-streaked on fresh YNB plates containing corresponding concentrations of this reagent and tested in fermentation experiments. More than 40 resistant strains were selected and evaluated for the isobutanol production during glucose fermentation. The initial fermentation screening of selected strains was performed in tested tubes with the highest isobutanol producing colonies further tested in shake flasks fermentations. One of the bispyribac-resistant strains produced the highest amount of synthetized isobutanol reaching 0.51 g L−1. This strain had a 13% increase in isobutanol accumulation when compared to the initial strain (Table 3). In this case, we found a correlation between increase in isobutanol accumulation and acetolactate synthase activity of bispyribacresistant mutant. As genome sequence of M. magnusii ILV2 gene is not known, for now we could not study expression level of this gene in the mutant.

2.7. Effects of Glycine, 2-Oxovalerate and 2-Oxoisovalerate on Production of Higher Alcohols

The synthesis of isobutanol and butanol in S. cerevisiae was stimulated when glycine, 2-oxoisovalerate, or 2-oxovalerate were added to the fermentation media.[11,49] In this study, the isobutanol production of M. magnusii was also tested in the medium with glycine. Yeast cells were grown in minimal medium with yeast extract (0.1%) using glycine (15 g L−1) instead of ammonium sulfate as nitrogen source. The isobutanol production was stimulated by glycine both in parental M. magnusii and strain overexpressing native ILV2 gene by 46% and 61%, respectively (Table 3).
The more pronounced effect on isobutanol production in the recombinant strain overexpressing ILV2 gene was observed after supplementation of YPD medium with 2-oxoisovalerate (0.2%), whichreached0.76gL−1 (Figure2B).Therecombinantstrainhad no difference from parental strain by glucose consumption and ethanol production (Figure 2C,D), but produced higher amounts of glycerol (Figure 2A,B). It is interesting to note that though the wild type strain of M. magnusii did not accumulate n-butanol in YPD medium, the addition of 2-oxovalerate resulted in butanol accumulation. After 48 h of incubation with 2-oxovalerate, the wild type strain of M. magnusii accumulated large amounts of nbutanol which reached from 0.62 to 0.82 g L−1.

3. Discussion

Multinuclear yeast M. magnusii is an interesting organism for cytological studies due to large cell volume.[30–32] It also has some biotechnological interest.[37,38] In this work, we showed that M. magnusii is a very promising producer of higher alcohol isobutanol which is accumulated in large quantities (more than 0.4 g L−1 or 10–20 times more than that in S. cerevisiae). However, further development of this yeast system is prevented by the poor development of the basic molecular methods and lack of genome sequence. Here, we developed advanced transformation protocol, cloned strong constitutive promoter TEF1 of M. magnusii and created the reporter system to test promoter strength. These methods permitted to express in M. magnusii heterologous S. cerevisiae ILV2 gene coding for acetolactate synthase which led to advanced isobutanol producer accumulated more than 0.7 g of isobutanol per L. Improvement of isobutanol synthesis was also achieved by selection for resistance to bispyribac, the inhibitor of acetolactate synthase. It is interesting to note that feeding M. magnusii with 2-oxovalerate led to an accumulation of high amounts of n-butanol which exceeded 0.8 g L−1. All these data clearly show that M. magnusii has great potential as producer of higher alcohols isobutanol and n-butanol. Further progress in this direction will be possible after sequencing and annotation of M. magnusii genome and adaptation several modern tools including CRISPR-Cas9 genome editing.

4. Experimental Section

Strains, Media, Cultivation Conditions: M. magnusii cells were grown on YPD (10 g L−1 yeast extract, 10 g L−1 peptone, 20 g L−1 glucose) or mineral medium (6.7 g L−1 YNB without amino acids, 20 g L−1 of glucose) at 30 °С. Alcoholic fermentation of yeast strains was fulfilled by cultivation in liquid YPD or mineral medium at oxygen-limited conditions at 30 °C. For this, the cells of the tested strains were pre-grown for 24 h in YPD medium under aerobic conditions (220 rpm) at 30 °C. Then the harvested cells were inoculated into 20 mL of YPD medium or YNB with 0.1% of yeast extract supplemented with 8% glucose to the final OD 1.5 mg of dry weight per mL. Fermentation was carried out at the temperature of 30 °C under semiaerobic conditions (120 rpm). Samples were taken daily during 2 or 3 days of fermentation and analyzed for production of ethanol and higher alcohols using HPLC.
The E. coli DH5𝛼 strain (Φ80dlacZΔM15, recA1, endA1, gyrA96, thi-1, hsdR17(rK−, mK+), supE44, relA1, deoR, Δ(lacZYA-argF)U169) was used as a host for plasmid propagation. Strain DH5𝛼 was grown at 37 °C in LB medium as described previously. Transformed E. coli cells were maintained on a medium containing 100 mg L−1 of ampicillin.
Molecular-Biology Techniques: Standard cloning techniques were carried out as described.[50] Genomic DNA of M. magnusii was isolated using the Wizard Genomic DNA Purification Kit (Promega, Madison, WI, USA). Restriction endonucleases and DNA ligase (Fermentas, Vilnius, Lithuania) were used according to the manufacturer specifications. Plasmid isolation from E. coli was performed with the Wizard Plus SV Minipreps DNA Purification System (Promega, Madison, WI, USA). DNA fragments were separated on a 0.8% agarose (Fisher Scientific, Fair Lawn, NJ, USA) gel. Isolation of fragments from the gel was carried out with a DNA gel extraction kit (Millipore, Bedford, MA, USA). PCR-amplification of the fragments of interest was done with Phusion High-Fidelity DNA Polymerase (Thermo Scientific, USA) according to the manufacturer specification. PCRs were performed in GeneAmp PCR System 9700 thermocycler (Applied Biosystems, Foster City, CA, USA).
Construction of Reporter Cassette for Evaluation of Different Promoters: The gene ble (1131 bp) conferring resistance to zeocine (ZeoR) was amplified using primers Ko354 and Ko355 and pPICZB plasmid as a template, digested with EcoRI and inserted into the corresponding site of the plasmid pUC57. The resulted plasmid was named pUC57-ZeoR. The LAC4 gene of Kluyveromyces lactis was amplified using primers Ko381 and Ko382 (Table S1, Supporting Information) and the plasmid pLAC4[43] harboring the mentioned gene as a template. The resulting DNA fragment with total size of 3292 bp was XbaI/SacI digested and cloned to the pUC57-ZeoR/XbaI/SacI linearized plasmid, resulting in the recombinant construct pUC57-LAC4-ZeoR (Figure S1, Supporting Information). This plasmid was used as an initial for subcloning of tested promoters.
DNA fragments bearing promoter of M. magnusii TEF1 gene (599 bp and 902 bp) were amplified using primers OK92/OK93 and OK92/OK94, respectively.
For comparison of the strength of M. magnusii TEF1 gene promoter with strong constitutive promoters of S. cerevisiae, plasmids harboring reporter gene LAC4 K. lactis driven by S. cerevisiae TEF1 or ADH1 gene promoters were constructed. Promoter of gene TEF1 (599 bp) of S. cerevisiae was amplified using primers OK95 and OK96. Promoter of gene ADH1 (807 bp) of S. cerevisiae was amplified using primers OK97 and OK98. Genomic DNA of S. cerevisiae BY4742 strain was used as a template for above mentioned PCR amplifications.
DNA fragments bearing promoter of M. magnusii TEF1 gene (599 bp and 902 bp), as well as promoters of S. cerevisiae TEF1 and ADH1 genes were XbaI/XhoI digested and ligated with pUC57-LAC4-ZeoR/XbaI/XhoI linearized plasmid resulting in the recombinant constructs pEmTEF1599, pEmTEF1902, pScTEF1, pScADH1 (Figure S1, Supporting Information). Construction of Plasmids for Overexpression of S. cerevisiae ILV2 and ARO10 Genes Driven by the Homologous TEF1 Promoter in M. magnusii: DNA fragment bearing promoter of M. magnusii TEF1 gene (599 bp) was amplified using primers OK105 and OK106 from previously constructed plasmid pEmTEF1599. The ORF of S. cerevisiae ILV2 gene with own terminator (2082 bp) was amplified using primers OK107 and Ko575 and plasmid pUC57-ADHI-ILV2-natNT2 as a template. Both fragments were fused by overlap PCR using primers OK105 and Ko575. Obtained DNA fragment (2699 bp) was HindIII/PstI digested and ligated with pUC57ZeoR/HindIII/PstI linearized plasmid resulting in the recombinant construct pUC57-prTEF1_ILV2-ZeoR (Figure S2, Supporting Information).
The ORF of S. cerevisiae ARO10 gene with own terminator (2113 bp) was amplified using primers OK110 and OK111 and genomic DNA of S. cerevisiae BY4742 as a template. DNA fragment bearing promoter of E. magnusii TEF1 gene (599 bp) was amplified using primers OK108 and OK109 from previously constructed plasmid pEmTEF1599. Both fragments were fused by overlap PCR using primers OK108 and OK111. Obtained DNA fragment (2730 bp) was BamHI/KpnI digested and ligated with pUC57ZeoR/BamHI/KpnI linearized plasmid resulting in the recombinant construct pUC57-prTEF1_ARO10-ZeoR (Figure S2, Supporting Information).
For simultaneous overexpression of ILV2 and ARO10 genes a BamHI/KpnI-restriction fragment containing prTEF1_ARO10 was isolated from the plasmid prTEF1_ARO10-ZeoR and cloned into the BamHI/KpnIlinearized vector pUC57-prTEF1_ILV2-ZeoR. The resulting plasmid was named pUC57-prTEF1_ILV2-prTEF1_ARO10-ZeoR (Figure S2, Supporting Information).
Quantitative Real Time PCR: The expression of the ILV2 gene was analyzed by real time PCR. Total RNA was extracted using the GeneMATRIX Universal RNA Purification Kit with DNAseI (EURx Ltd, Gdansk, Poland) according to the manufacturer’s instructions. Samples for the reaction were taken on the second day of 30 °C fermentation. The purity and quantity of samples were determined using Epoch Spectrophotometer System and BioTek Take3 Volume Plate, subsequently diluted in RNAse free water. Real time amplification reactions were performed in 96 well plates using SYBR Green detection chemistry and run in triplicate using 96-wells plates with the 7500 Fast Real Time PCR System (Applied Biosystems). Normalized amount of RNA (500 ng per reaction) and 0.4 µm of each of the two gene-specific primer pairs, and ROX reference passive dye according to the manufacturer’s instructions were used in a total reaction volume of 25 µL (SG OneStep qRT-PCR kit, EURx Ltd, Gdansk, Poland). The amplification was performed with the following cycling profile: reverse transcription step at 50 °C for 30 min; initial denaturation at 95 °C for 3 min at preparation step; followed by 45 cycles of 15 s at 94 °C and 30 s at 60 °C. Melting curve analysis was performed to verify the specificity and identity of PCR products from 65 to 95 °C in the software of real time cycler. The amplification after 35—45 cycles, gave abundance of PCR product indicating saturation phase. The increase for the amplicon relative to the control sample was normalized to an internal control gene TEF1 and calculated according to the comparative Ct (ΔΔCt) method. All data points were analyzed in triplicate. Sequences of the genes ILV2 and TEF1 were taken from S. cerevisiae genome database (https://www.yeastgenome.org/).
Biochemical Methods: The enzyme activity was measured directly after the preparation of cell-free extracts. Protein concentration was determined with Folin reagent. To confirm the proper expression of heterologous ILV2 gene in selected M. magnusii recombinant strains, the activity of Ilv2 (acetolactate synthase) was measured. The cells harvested from YPD medium after 24 h of incubation were used for enzyme activity assay. The specific activity of acetolactate synthase was measured in cell free extracts using colorimetric assay as described elsewhere.[41]
All assay experiments were repeated at least twice.
Analyses: The biomass was determined turbidimetrically with a Helios Gamma spectrophotometer (OD, 600 nm; cuvette, 10 mm) with gravimetric calibration. Concentrations of isobutanol and butanol from fermentation in medium broth were analyzed by HPLC (PerkinElmer, Series 2000, USA) with an Aminex HPX-87H ion-exchange column (Bio-Rad, Hercules, USA). A mobile phase of 4 mm H2SO4 was used at a flow rate 0.6 mL min−1 and the column temperature was 35 °C. Experiments were performed at least twice.
Selection of Isobutanol Overproducing Mutants: For the selection of Phleomycin D1 isobutanol overproducing mutants, 100 µL of a cell suspension of E. magnusii wild type strain VKM Y-1072 (OD600 0.3; 0.5 and 1.0 per mL cells) was plated on YNB solid medium supplemented with toxic concentrations of bispyribac sodium (55 mm) or valine (30 mm) and incubated for 7–10 days. Toxic concentrations of these agents were defined as the amount of the selective agent in YNB solid medium that restrict growth of VKM Y1072 strain plating 100 µL of a cell suspension with an OD 1.0 per plate. Single yeast colonies resistant to 55 mm of bispyribac sodium or 30 mm of valine were picked up and re-streaked on fresh YNB plates containing the corresponding concentrations of the agent and used in fermentation experiments. For both type of mutants, more than 40 selected strains were selected and evaluated for isobutanol production in fermentation with glucose as the primary carbon source. The initial fermentation screening of selected strains was performed in test tubes. Then strains producing the highest level of isobutanol were further tested in shake flasks.

References

[1] C. Schubert, Nat. Biotechnol. 2006, 24, 777.
[2] A. K. Gombert, A. J. A. van Maris, Curr. Opin. Biotechnol. 2015, 33, 81.
[3] M. L. A. Jansen, J. M. Bracher, I. Papapetridis, M. D. Verhoeven, H. de Bruijn, P. P. de Waal, A. J. A. van Maris, P. Klaassen, J. T. Pronk, FEMS Yeast Res. 2017, 17, fox044.
[4] D. Antoni, V. V. Zverlov, W. H. Schwarz, Appl. Microbiol. Biotechnol. 2007, 77, 23.
[5] N. Savage, Nature 2011, 474, S9.
[6] J. Hewitt, K. S. Lokare, Biofuels Dig. 2016.
[7] B. Schiel-Bengelsdorf, J. Montoya, S. Linder, P. Dürre, Environ. Technol. 2013, 34, 1691.
[8] C. Xue, F. Liu, M. Xu, I. C. Tang, J. Zhao, F. Bai, S. T. Yang, Bioresour. Technol. 2016, 219, 158.
[9] T. Si, Y. Luo, H. Xiao, H. Zhao, Metab. Eng. 2014, 22, 60.
[10] W. C. Generoso, V. Schadeweg, M. Oreb, E. Boles, Curr. Opin. Biotechnol. 2015, 33, 1.
[11] R. Swidah, O. Ogunlabi, C. M. Grant, M. P. Ashe, Appl. Microbiol. Biotechnol. 2018, 102, 9857.
[12] A. Yu, Y. Zhao, Y. Pang, Z. Hu, C. Zhang, D. Xiao, M. W. Chang, S. S. J. Leong, Microb. Cell Fact. 2018, 17, 166.
[13] S. Atsumi, T. Y. Wu, E. M. Eckl, S. D. Hawkins, T. Buelter, J. C. Liao, Appl. Microbiol. Biotechnol. 2010, 85, 651.
[14] S. Atsumi, J. C. Liao, Curr. Opin. Biotechnol. 2008, 19, 414.
[15] C. T. Chen, J. C. Liao, FEMS Microbiol. Lett. 2016, 363, fnw020.
[16] S. Li, J. Wen, X. Jia, Appl. Microbiol. Biotechnol. 2011, 91, 577.
[17] K. M. Smith, K. M. Cho, J. C. Liao, Appl. Microbiol. Biotechnol. 2010, 87, 1045.
[18] S. Yamamoto, M. Suda, S. Niimi, M. Inui, H. Yukawa, Biotechnol. Bioeng. 2013, 110, 2938.
[19] M. He, B. Wu, H. Qin, Z. Ruan, F. R. Tan, J. L. Wang, Z. X. Shui, L. C. Dai, Q. L. Zhu, K. Pan, X. Y. Tang, W. G. Wang, Q. C. Hu, Biotechnol. Biofuels 2014, 7, 101.
[20] D. Brat, C. Weber, W. Lorenzen, H. B. Bode, E. Boles, Biofuels 2012, 5, 65.
[21] J. L. Avalos, G. R. Fink, G. Stephanopoulos, Nat. Biotechnol. 2013, 31, 335.
[22] S. Park, J. Hahn, Sci. Rep. 2019, 9, 3996.
[23] K. Ida, J. Ishii, F. Matsuda, T. Kondo, A. Kondo, Microb. Cell Fact. 2015, 14, 62.
[24] S. H. Park, S. Kim, J. S. Hahn, Appl. Microbiol. Biotechnol. 2016, 100, 7591.
[25] S. K. Hammer, J. L. Avalos, Metab. Eng. 2017, 44, 302.
[26] F. Matsuda, J. Ishii, T. Kondo, K. Ida, H. Tezuka, A. Kondo, Microb. Cell Fact. 2013, 12, 119.
[27] J. Wess, M. Brinek, M. Boles, Biotechnol. Biofuels 2019, 12, 173.
[28] W. Siripong, P. Wolf, T. P. Kusumoputri, J. J. Downes, S. Tanapongpipat, W. Runguphan, Biotechnol. Biofuels 2018, 11, 1.
[29] J. Slavík, A. Kotyk, Biochim. Biophys. Acta – – Bioenerg. 1984, 766, 679.
[30] Y. I. Deryabina, E. P. Isakova, E. I. Shurubor, R. A. Zvyagilskaya, Biochemistry (Moscow) 2004, 69, 1025.
[31] J. Hasek, P. Trachtulcová, S. D. Kohlwein, E. Streiblová, Folia Microbiol. 2003, 48, 177.
[32] A. G. Rogov, A. P. Ovchenkova, T. N. Goleva, I. I. Kireev, R. A. Zvyagilskaya, Anal. Biochem. 2018, 552, 24.
[33] P. Griac, J. Nosek, Curr. Genet. 1993, 23, 549.
[34] D. Filipp, P. Filipp, J. Nosek, M. Hladká, Curr. Genet. 1995, 29, 81. [35] L. Adamíková, P. Griac, L. Tomáska, J. Nosek, Yeast 1998, 14, 805.
[36] M. Kucej, P. Kosa, L. Adamíková, J. Nosek, Curr. Genet. 2002, 41, 20.
[37] M. Rosenberg, H. Miková, L. Kristofíková, Lett. Appl. Microbiol. 1999, 29, 221.
[38] H. Hronská, S. Micháliková, M. Rosenberg, J. Food Nutr. Res. 2017, 56, 219.
[39] L. A. Hazelwood, J. M. Daran, A. J. A. van Maris, J. T. Pronk, J. R. Dickinson, Appl. Environ. Microbiol. 2008, 74, 2259.
[40] X. Chen, K. F. Nielsen, I. Borodina, M. C. Kielland-Brandt, K. Karhumaa, Biotechnol. Biofuels 2011, 4, 21.
[41] L. Murashchenko, C. Abbas, K. Dmytruk, A. Sibirny, Yeast 2016, 33, 463.
[42] C. P. Kurtzman, C. J. Robnett, FEMS Yeast Res. 2013, 13, 23.
[43] O. P. Ishchuk, K. V. Dmytruk, O. V. Rohulya, A. Y. Voronovsky, C. A. Abbas, A. A. Sibirny, Enzyme Microb. Technol. 2008, 42, 208.
[44] T. Kondo, H. Tezuka, J. Ishii, F. Matsuda, C. Ogino, A. Kondo, J. Biotechnol. 2012, 159, 32.
[45] W. H. Lee, S. O. Seo, Y. H. Bae, H. Nan, Y. S. Jin, J. H. Seo, Bioprocess Biosyst. Eng. 2012, 35, 1467.
[46] A. J. Fischer, D. E. Bayer, M. D. Carriere, C. M. Ateh, K. O. Yim, Pestic. Biochem. Physiol. 2000, 68, 156.
[47] S. B. Powles, Q. Yu, Annu. Rev. Plant Biol., 2010, 61, 317.
[48] K. Kawai, K. Kaku, N. Izawa, T. Shimizu, Y. Tanaka, J. Pestic. Sci. 2007, 32, 89.
[49] P. Branduardi, V. Longo, N. M. Berterame, G. Rossi, D. Porro, Biotechnol. Biofuels 2013, 6, 68.
[50] J. Sambrook, E.F. Fritsch,T. Maniatis, Molecular Cloning: A Laboratory Manual, 2nd ed., Cold Spring Harbor Laboratory, New York 2001.