Effect of Evolutionary Adaptation on Metabolic Enzyme Activities of Thermotolerant Strain Kluyveromyces Marxianus NIRE-K1 and NIRE-K3 for Bioethanol Production

Evolutionary adaptation provides stability to the strains in the challenging environment. As extension of earlier study, the evolved strains Kluyveromyces marxianus NIRE-K1.1 and K. marxianus NIRE-K3.1 were subjected for secondary adaptation on minimal salt (MS) medium with the aim to enhance xylose utilization for ethanol production together with salt tolerance. Both the strains were adapted till saturated improvement in xylose uptake i.e., 54 generations on MS medium containing xylose. Xylose utilization increased from 14.21 to 45.80% and 10.55 to 45.31%, in evolved strains KmNIRE-K1.2 and KmNIRE-K3.2, respectively. Specic xylose reductase activity has also increased 2.04 and 3.36-folds in KmNIRE-K1.2 and KmNIRE-K3.2, respectively. Xylitol dehydrogenase activity was also increased by 2.82 and 1.35-folds in KmNIRE-K1.2 and KmNIRE-K3.2, respectively. Decrease in redox imbalance was observed in evolved strains, and hence there was a reduction in xylitol production during growth and fermentation. Xylose uptake rate increased by 2.53 and 1.5-folds in KmNIRE-K1.2 and KmNIRE-K3.2, respectively with 2.20 and 6.46-folds higher ethanol concentration, and 2.25 and 5.86-folds higher volumetric productivity, respectively. This study has demonstrated the role of evolutionary adaptation for developing robust yeast strains. KmNIRE-K1.2 and KmNIRE-K3.2 have shown enhanced ethanol production, enzyme activities and less by-product formation like xylitol during xylose metabolism.

uptake rate increased by 2.53 and 1.5-folds in KmNIRE-K1.2 and KmNIRE-K3.2, respectively with 2.20 and 6.46-folds higher ethanol concentration, and 2.25 and 5.86-folds higher volumetric productivity, respectively. This study has demonstrated the role of evolutionary adaptation for developing robust yeast strains. KmNIRE-K1.2 and KmNIRE-K3.2 have shown enhanced ethanol production, enzyme activities and less by-product formation like xylitol during xylose metabolism.
Background E cient conversion of lignocellulosic biomass to transportation fuels such as bioethanol is a major concern of biofuel researchers since last three decades for economic and environmental sustainability (Arora et. al., 2019;. Although, many process have been developed for cellulosic ethanol production but their economic viability is still unknown due to incomplete conversion of biomass to valuable products (Achinas et. al., 2019). Lignocellulosic biomass contains C6 and C5 sugars including glucose and xylose as major sugar i.e. 90% of total sugar present and simultaneous conversion of both sugars into ethanol is still a loophole (Rodrussamee et. al. 2018).
Simultaneous sacchari cation and fermentation (SSF) is one of the integrated approaches in lignocellulosic ethanol production has various advantages i.e., high productivity, avoids cellulases inhibition due to glucose accumulation, etc. However, the optimal temperatures of maximum cellulases activity and fermentation by conventional yeast Saccharomyces cerevisiae and other mesophilic yeasts are about 50 o C and 30 o C, respectively, and have wide difference to perform in a single step .
Nowadays thermotolerant ethanologenic yeast K. marxianus has attracted increasing attention due to its thermotolrance and broad range of substrate speci city including xylose (Sharma et. al., 2016;. It can grow up to 50 o C and produce ethanol while most of the yeasts cannot survive at this temperature . Various genetic engineering efforts have been applied on K. marxianus including heterologously expression of XYL1 and XYL2 genes in xylose utilization pathways from S. stipitis or XYL1 from Orpinomyces sp. Wang et. al., 2013). However, the ethanol yield and Growth study was carried out in a bioreactor of 3 L volume (NBS BioFlo-CelliGen 115) under aerobic condition with working volume of 2 L. Growth was performed for 24 h at 45 o C for native and adapted culture. The samples were collected at an interval of 2 h and analyzed for DCW, xylose concentration and extracellular metabolites. Fermentation was carried out in NBS BioFlo-CelliGen 115 (New Brunswick, USA) bioreactor of 3 L volume under controlled temperature of 45 o C and pH 5.5. Inoculum was prepared by growing the strains overnight in MSX medium at 45 o C. Cells were harvested by centrifugation at 5000 × g for 10 min and used as inoculum for fermentation in bioreactor with similar composition as the growth medium except xylose concentration of 30 g xylose l -1 . Bioreactor was loaded with 2 L salt medium and sterilized at 121 o C for 15 min. The initial cell mass concentration was kept at ~2.5 g cells l -1 .

Enzyme assay
The strains were grown in YEP medium containing 20 g xylose l -1 and/or 20 g glucose l -1 overnight at 45 o C to gain high cell mass. Cells were harvested at 3000 × g for 10 min at 4 o C in a centrifuge. Protein extraction was done by using Yeast Buster protein extraction reagent (Novagen, Germany) and procedure was followed according to manufacturer's instructions. Enzyme activity of xylose reductase (XR) and xylitol dehydrogenase (XDH) was determined according to Ikeuchi et al. (2000) using spectrophotometer at A 340 for KmNIRE-K1.1, KmNIRE-K1.2, KmNIRE-K3.1 and KmNIRE-K3.2 at variable temperatures ranging 30 to 50 o C. One unit (U) of XR activity is de ned as the amount of enzyme required to oxidize 1.0 μmole NADPH to NADP + for reduction of D-xylose to xylitol in one min; and one unit (U) of XDH activity are de ned as the amount of enzyme required to reduce 1.0 μmole NAD + to NADH for oxidation of xylitol to xylulose in one min. Enzyme activities were calculated as per the formulae given by Ikeuchi et al. (2000): Where, TV is total volume; V is volume of cell extract, ε is molar extinction coe cient (6.22 l mmol -1 for a path length of 1.0 cm); CF is dilution of cell extract.
Co-enzyme quanti cation Intracellular co-enzyme concentrations were determined using the cells harvested from YEPX medium containing 2% xylose as carbon source in log phase. Two milliliter sample was withdrawn and cell pellets were washed with phosphate buffer (pH 7.0) followed by extraction of NADH/NAD + and NADPH/NADP + by using NADH/NAD + and NADPH/NADP + extraction buffers (Sigma Aldrich, USA), respectively, as manufacturer's instructions. The co-enzyme quanti cation was done calorimetrically at following the manufacturer's instructions.

Real Time Polymerase Chain Reaction (RT-PCR)
RT-PCR was performed to determine relative expression of XYL1 and XYL2 genes in KmNIRE-K1.1, KmNIRE-K1.2, KmNIRE-K3.1 and KmNIRE-K3.2, respectively. These strains were cultivated in YEP medium containing 20 g l -1 xylose under aerobic condition at 45 o C. Total RNA was isolated by using Aurum total RNA kit (Biorad, USA) and puri ed according to manufacturer's instructions, followed by cDNA synthesised using iScript cDNA synthesis kit (Biorad, Canada). RT-PCR was performed in CFX connect real time system (Biorad, India) with iTaq universal syber green supermix (Biorad, Canada).
KmAtc1 encoding for actin was used as control. List of primers used in the gene expression analysis were reported in Table 1.

Analytical methods and kinetics
The concentration of xylose, ethanol, xylitol, glycerol and acetic acid were estimated using high-pressure liquid chromatography (HPLC) system (Agilent Technologies, USA) with HiPlex H column with 1mM H 2 SO 4 as the mobile phase and ow rate of 0.7 ml min -1 at 57 o C and detected by refractive index detector (RID) at 50 o C. Yeast cell concentration was measured through dry cell weight (DCW). Protein concentrations in cell extracts were estimated using Lowry's method. Growth and fermentation kinetic parameters were calculated using the formulae by Bailey and Ollis (Bailey et. al., 1986;Sharma et. al., 2016;

Results And Discussion
As reported in our previous studies, KmNIRE-K1.1 and KmNIRE-K3.1 were developed through primary adaptation of wild type strains (Sharma et. al., 2016;. The signi cant improvement was observed in adapted cultures, but still it was not signi cant in terms of xylose utilization and ethanol fermentation when grown in MSX. Hence, the primary adapted strains were further adapted in MSX medium for further improvement.
Evolutionary adaptation of KmNIRE-K1.1 and KmNIRE-K3.1 KmNIRE-K1.1 and KmNIRE-K3.1 were undergone secondary adaptation in MSX medium containing xylose as carbon source and improvement were monitored in the form of DCW and sugar concentrations in the broth after completion of each batch as shown in Fig. 1. There was almost 2-fold increase in xylose utilization in rst 10 batches in both the cultures. KmNIRE-K1.1 and KmNIRE-K3.1 utilized approximately 50% and 45% of total sugar after 50 batches. After 50 batches, no signi cant improvement was observed in cell growth for both the cultures and hence, the sub-culturing was stopped after 54 cycles. There was 3.5-fold increase in xylose utilization by KmNIRE-K1.1 (Fig. 1A) and renamed the adapted culture as KmNIRE-K1.2, while, there was 4.5-fold increase in xylose utilization by KmNIRE-K3.1 (Fig. 1B)  at 50 o C. XR activity in KmNIRE-K1.2 was about 1.34-fold higher in glucose/xylose mixture than xylose at 45 o C. The higher XR activity in cells grown in glucose/xylose mixture might be due to up regulation of XYL1 gene in the presence of glucose, which enhances xylose transport (Hamacher et. al., 2002).  . Similary, Dasgupta et al. (2016)  whereas, in KmNIRE-K3.1 it was 0.75. The decreased values of ratio of NAD + /NADH showed the better activity of XDH which converts more NAD + to NADH for production of xylulose from xylitol. Hence, the secondary adapted strains had the better XDH activity in terms of co-factor ratio. with NADPH preferring XR, NADP + preferring XDH and NADH dependent GLN1 and reported the increased co-enzyme concentration by 6.72-fold as compared to parent strain K. marxianus NBRC1777. Moreover, NADH/NAD + and NADPH/NADP + ratios were also higher i.e. 0.89 and 0.80 than K. marxianus NBRC1777 i.e. 0.28 and 0.69, respectively Zhang et al., 2016. In another study, Wei et al. (2013), developed FPS1 deleted strain S. cerevisiae SR8-fps1Δ to enhance ethanol production. SR8-fps1Δ showed 10 to 30 % enhanced ethanol yield together with increased NADP + /NADPH ratio as compared to parent strain S. cerevisiae SR8. However, NAD + /NADH ratio was similar as parent strain (Wei et al., 2013).
The researchers have reported various studies of relative expression of XYL1, XYL2 and other xylose metabolic genes. In another study, Qi et al. (2015), analysed gene expression of XYLA and XKS1 of evolutionary adapted recombinant S. cerevisiae SyBE003 and reported 0.53 and 3-folds relative expression, respectively, than that of SyBE002. Biswas et al. (2013) investigated the effect of different carbon sources on the expression level of XYL2 gene of Debaryomyces hansenii and reported 7.5-fold higher expression of XYL2 gene in xylose grown cells as compared glucose grown cells. Similarly, Yamasaki-Yashiki et al. (2014) also reported higher expression of XYL2 in Rhizomucor pusillus grown in xylose as compared to glucose. In another study, expression of XYL1 in R. pusillus NBRC 4578 was found to be higher in the cells grown in xylose and arabinose than that of grown in glucose (Komeda et al., 2015).

Comparative growth pro ling of primary and secondary adapted strains
Growth pro ling of primary and secondary adapted cells of KmNIRE-K1 and KmNIRE-K3 were performed in 3L Bio ow Bioreactor containing 2L MSX at 45 o C, 5.5 pH and 250 rpm. Growth pattern was drawn as shown in Fig. 4A 1 (8 h). The lag phase during microbial growth usually increases due to change in environment, which could be reduced through adaptation in the same media. Hence, reduction in lag phase after adaptation may be due to stabilization of regulatory mechanisms (Sharma et al., 2016;New et al., 2014). Variation in growth environment results slow metabolic activity, because transcription is not adapted for new changes (New et al., 2014).
However, after adaptation in new environment microorganisms show improved growth i.e. shorter lag phase (Sharma et al., 2016;New et al., 2014). Shorter lag phase was also found after adaptation with various inhibitors including furfural, HMF and acetic acid (Koppram et. al., 2012;Landaeta et. al., 2013;Slininger et. al., 2015). Engineered strains also showed shorter lag phase after evolutionary adaptation (Martín et. al., 2007, Koppram et. 2012. During growth in MSX medium xylose uptake rate increased by 2.53-fold in KmNIRE-K1.2 (0.48 ± 0.038 g g -1 h -1 ) as compared to KmNIRE-K1.1. Similarly, xylose uptake rate of KmNIRE-K3.2 was found to be 1.5fold higher than that of KmNIRE-K3.1 (Table 3). However, the cell mass yield was 1.38-fold lower in KmNIRE-K1.2 as compared to KmNIRE-K1.1. Similarly, the cell mass yield of KmNIRE-K3.2 was also slightly lower i.e., 0.30 ± 0.01 g g -1 as compared to KmNIRE-K3.1. The lower cell mass yield may be due to the formation of other metabolite i.e. ethanol under aerobic condition . However, the ethanol was not detected in the broth, which is supposed to escape of ethanol from vent due to air sparging. Sonderegger et al. (2003) developed S. cerevisiae TMB3001C1 and TMB3001C5 through adaptation with 460 generations and reported improved maximum growth rate as 0.64 ± 0.001 and 0.119 ± 0.001 h -1 , with xylose uptake rate of 0.13 ± 0.00 and 0.27 ± 0.02 g g -1 h -1 , respectively on 5 g xylose l -1 initial concentration. Similarly, Zhou et al. (2012) applied 3-stage evolutionary adaptation on engineered S. cerevisiae H131-A3 and reported improved speci c growth rate of 0.12 h -1 on xylose as sole source of carbon. Several improved features like enhanced fermentation rates, improved ethanol tolerance and yield have also been reported through adaptation. The current study shows that the evolutionary adaption in salt medium has proved to be a better approach for improvement of yeast strain in context to utilize xylose present in lignocellulosic biomass.
Aforesaid results indicate the role of secondary evolutionary engineering for enhancement of the xylose uptake, cell mass growth, and ethanol production and yield in MSX medium. Moreover, it has also enhanced the gene expression of metabolic gene required for xylose metabolism. Both the strains KmNIRE-K1.2 and KmNIRE-K3.2 showed better ratio of NAD + /NADP and NADPH/NADP + , which show the redox balance. However, further improvement is also suggested to reduce acetic acid and glycerol e ux during fermentation, which could further increase the ethanol production.

Conclusion
Two thermotolerant strains KmNIRE-K1.2 and KmNIRE-K3.2 were developed through evolutionary adaptation with faster xylose utilization capacity. Both the strains showed signi cant growth on xylose under aerobic condition along with better xylose uptake rate. Speci c activities of metabolic enzymes including XR and XDH drastically improved in both the strains. Further, co-enzyme quanti cation study also revealed the reduction in redox imbalance after the evolutionary adaptation. Gene expression of XYL1 and XYL2 also increased many folds during xylose metabolism. These improvements occurred due to some evolutionary mutation inside the genome. Higher activity of XR and XDH in adapted strain at 45 o C, will be fruitful for the SSF process and may increase the productivity of bioethanol. Therefore, further adaptation, optimization and modi cation at the genetic level will surely enhance bioethanol production and reduce by-product formation including acetic acid and glycerol which will be acceptable at the industrial point of view. However, the genomic characterization of the strains will be done in future together with possible intergeneric hybridization. Moreover, the method of strain development using evolutionary adaptation has been proven to useful for bioprocessing using robust non-conventional strains according to this study and opens the scope for integration of targeted and non-targeted genetic engineering methods.