Optimization Condition for Ethanol Production from Sweet Sorghum Juice by Recombinant Zymomonas mobilis Overexpressing groESL Genes

Abstract

High-temperature ethanol fermentation (HTEF) using high-potential thermotolerant ethanologenic microorganisms is a promising platform for ethanol production in tropical or subtropical areas. This study aims to evaluate the ethanol production potential of recombinant Zymomonas mobilis R301 overexpressing groESL genes under normal and high-temperature conditions and the expression of genes involved in the heat shock response and ethanol production pathway during ethanol fermentation using sweet sorghum juice (SSJ) as feedstock. Growth characterization analysis revealed that the recombinant Z. mobilis R301 exhibited multi-stress tolerance toward heat, acetic acid, and furfural. Based on the statistical experimental design, the optimum conditions for ethanol production from SSJ by the recombinant R301 at 30 °C were a sugar concentration of 171.67 g/L, cell concentration of 9.42% (v/v), and yeast extract concentration of 10.89 g/L, while those at 40 °C were a sugar concentration of 199.48 g/L, yeast extract concentration of 10.88 g/L, MgSO₄ concentration of 1.05 g/L, and initial pH of 6.8. The maximum ethanol concentrations and productivities achieved in this study were 63.26 g/L and 1.17 g/L.h at 30 °C and 58.62 g/L and 1.22 g/L.h at 40 °C. The overexpression of the groES and groEL genes and upregulation of other heat shock-responsive genes at 40 °C enhanced cell growth, viability, and fermentation capacity of recombinant Z. mobilis R301 under heat stress. The current study demonstrated that recombinant Z. mobilis R301 exhibited high potential for ethanol production from SSJ or other sugar-based raw materials under high-temperature conditions.

1. Introduction

Global ethanol production was 27,290 million gallons in 2021, and the world’s largest ethanol producers are the United States and Brazil. Thailand is among the top seven global ethanol producers, producing approximately 380 million gallons annually, accounting for 1% of world production [1]. Most commercial ethanol production for biofuel used worldwide is currently based on sugar- and starch-based materials, such as sugarcane, molasses, cassava, and corn. Since these feedstocks are used directly as food for human needs and animal feed, alternative raw materials that avoid competition with those sectors are extensively studied. Several feedstocks have been proposed, and sweet sorghum (Sorghum bicolor L.) is among the most potent alternative energy crops for commercial biofuel ethanol production. In addition to having high biomass productivity, being well adapted to temperate climates, and requiring less water and fertilizer, it also contains high amounts of soluble and insoluble carbohydrates as well as certain essential minerals for microbial growth and metabolic activity comparable to those detected in sugarcane [2,3]. Importantly, sweet sorghum is not used directly for food and animal feed [4], making this resource a promising and ideal feedstock for bioethanol production.

Ethanol production in tropical and subtropical regions, such as Thailand, where the average temperatures are usually high throughout the year, is challenged not only by the costs of raw materials due to economic growth but also by the operating costs related to the cooling system employed for controlling temperature during the fermentation process. High-temperature ethanol fermentation (HTEF) has been proposed as a promising technology to address this problem [3,5,6,7]. In addition to having a high fermentation rate, reducing the risk of contamination by undesirable microorganisms, and lowering the operating cost of the cooling system, HTEF also offers the possible use of a simultaneous saccharification and fermentation (SSF) platform when coupled with a gas-stripping system for ethanol recovery [5]. One of the critical successes in ethanol production under high-temperature conditions is the utilization of high-potential thermotolerant microorganisms [5,6,8]. Several thermotolerant ethanologenic yeasts for HTEF have been reported, such as Kluyveromyces marxianus [8,9,10], Saccharomyces cerevisiae [3,5,6], Pichia kudriavzevii [11], Meyerozyma guilliermondii [12], and Saccharomycodes ludwigii [7]; however, very few reports have considered thermotolerant ethanologenic bacteria, specifically Zymomonas mobilis [13,14,15,16].

Z. mobilis is a Gram-negative facultative anaerobic bacterium with great promise as an ethanologenic bacteria. Compared to traditional yeast S. cerevisiae, Z. mobilis has several attractive features, such as a high sugar uptake rate, high ethanol productivity, and a high ethanol yield of up to 98% of the theoretical yield. It is also capable of withstanding pH fluctuations ranging from 3.8 to 7.5, high concentrations of sugar (up to 400 g/L) and ethanol (up to 160 g/L) [17,18], as well as acetic acid (up to 200 mM) and hydrogen peroxide (up to 0.4 mM) [14,15]. Moreover, it is widely recognized as safe (GRAS), making it applicable to various fields, including food, beverage, medical, pharmaceutical, fine chemical, and biofuel production [19]. However, like other ethanologenic microorganisms, heat stress remains a drawback that limits the application of Z. mobilis for HTEF, as it causes severe effects on bacterial growth, cell viability, metabolic activity, and ethanol fermentation. Thus, several attempts, including forward and reverse genetic approaches, such as chemical mutagenesis [15,20], adaptive laboratory evolution (ALE) [14,21], genome shuffling [22], and genome editing using the clustered regularly interspaced short palindromic repeats (CRISPR)-CRISPR-associated (CAS) system [23,24], have been made to create new strains exhibiting the desired characteristics, specifically heat stress, and other tolerance capabilities.

The groESL genes, the essential molecular chaperone, have been shown to play critical roles in various stress tolerances [25,26]. Based on the literature reviews, overexpression of these genes enhances the tolerance abilities of several microorganisms, such as heat and salinity stress in Anabaena sp. PCC7120 [27], solvent stress in Clostridium acetobutylicum [28,29], and toxic metabolite stress in Escherichia coli [17]. Recently, we successfully developed a recombinant Z. mobilis overexpressing groESL, designated Z. mobilis R301, which can withstand several stress conditions, including high temperatures, high sugar, and ethanol concentrations of up to 40 °C, 300 g/L, and 102.57 g/L, respectively [16]. This recombinant strain can also utilize sweet sorghum juice (SSJ) as a carbon source for ethanol production. However, the optimal conditions for ethanol production at high temperatures using SSJ as feedstock have not yet been elucidated. Several fermentation parameters, such as sugar concentration, microbial cell concentration, pH of the fermentation medium, and nitrogen supplementation, have been shown to influence ethanol production efficiencies [3,5,6,9,10]. Furthermore, several stress-responsive genes may also be stimulated during ethanol fermentation at high temperatures. Thus, this study aims to screen and optimize the fermentation parameters that positively affect ethanol production efficiency under normal growth temperature and high-temperature conditions by recombinant Z. mobilis overexpressing the groESL genes using SSJ as feedstock. Growth characterization of Z. mobilis under stress conditions and the molecular mechanism of the recombinant Z. mobilis strain in response to heat stress during ethanol fermentation by expression analysis of genes involved in the heat shock response and ethanol production pathway were also investigated using reverse-transcription quantitative real-time polymerase chain reaction (RT–qPCR). The current study demonstrated that under the optimum fermentation conditions, recombinant Z. mobilis R301 could produce ethanol concentrations of 63.26 and 58.62 g/L at 30 and 40 °C, respectively. The high ethanol fermentation efficiency of the recombinant strain at relatively high temperature may be associated with groESL genes’ overexpression and the upregulation of other heat shock-responsive genes.

2. Materials and Methods

2.1. Bacterial Strains, Culture Conditions, and Inoculum Preparation

Z. mobilis TISTR548, a wild-type strain, was purchased from the Thailand Institute of Scientific and Technological Research (TISTR). Z. mobilis R301, a recombinant strain overexpressing groESL genes, was constructed in our laboratory by Kaewchana et al. [16]. These strains were cultured in yeast extract peptone glucose (YPG) agar medium composed of 3 g/L yeast extract, 5 g/L peptone, 30 g/L glucose, and 15 g/L agar and maintained at 4 °C by subculturing monthly.

For inoculum preparation, a single colony of Z. mobilis wild-type or recombinant strain was grown in a 5 mL sterilized YPG medium and incubated at 30 °C, 100 rpm for 12 h (the optical density of the bacterial cells at 550 nm (OD₅₅₀) reached approximately 0.8–1.0). The resulting cells were transferred into 50 mL of a sterilized YPG medium with an initial cell density of 0.05 and subsequently incubated at 30 °C and 100 rpm. After 12 h of incubation, cells were used as a starter culture for further experiments.

2.2. Plant Material

The sweet sorghum cultivar KKU40 was kindly provided by the Faculty of Agriculture, Khon Kaen University, Thailand. This cultivar was developed by Assoc. Prof. Dr. Prasit Jaisil and the plant germplasm have been deposited at the Faculty of Agriculture, Khon Kaen University, Thailand, with the code “KKU40”. After extraction of the juice from sweet sorghum stems using a juice extractor, the sugar concentration in the fresh juice was concentrated to approximately 80 °Brix via an evaporation technique [3], and the resulting sweet sorghum syrup was kept at −20 °C before use at the Department of Biotechnology, Faculty of Technology, Khon Kaen University, with the code number KKUDB-SSJ-2020-01. All plant preparation methods followed the relevant guidelines in the method section.

2.3. Effect of Stress Conditions on Bacterial Growth

The effects of heat, acetic acid, and furfural stress on the growth of Z. mobilis wild-type and recombinant strains were investigated. For heat stress, starter cultures of wild-type and recombinant strains were transferred to a 100 mL sterilized YPG medium with an initial cell density (OD₅₅₀) of 0.05 and incubated at 30, 37, and 40 °C and 100 rpm for 24 h [30]. Acetic acid stress was imposed using a protocol by Liu et al. [31]. Briefly, starter cultures of Z. mobilis wild-type and recombinant strains were transferred to a 100 mL sterilized YPG medium containing glacial acetic acid (Sigma–Aldrich, St. Louis, MO, USA) at final concentrations of 0 (control), 150, and 200 mM with an initial cell density (OD₅₅₀) of 0.05. The cultures were incubated at 30 and 37 °C and 100 rpm for 24 h, and samples were collected and analyzed. For furfural stress, starter cultures of Z. mobilis wild-type and recombinant strains were transferred to a 100 mL sterilized YPG medium containing furfural (Sigma–Aldrich, St. Louis, MO, USA) at final concentrations of 0 (control), 10, and 20 mM with an initial cell density (OD₅₅₀) of 0.05. After incubation at 30 and 37 °C and 100 rpm for 24 h, the growth of bacterial cells was evaluated by monitoring OD₅₅₀ using a spectrophotometer (UV-1601, Shimadzu, Kyoto, Japan) and expressed as the mean ± standard deviation (SD) [32]. The differences in growth between Z. mobilis wild-type and recombinant strains were statistically compared using an independent t-test at a p value of 0.05.

2.4. Bacterial Cells Morphology Analysis

The morphology of Z. mobilis wild-type and recombinant cells under heat stress was determined by culturing cells in SSJ containing 160 g/L total sugars at 30 and 40 °C. Cells were separated at 18 h (mid-exponential growth phase) and 36 h (stationary growth phase) by centrifugation at 10,000 rpm and 4 °C for 5 min. After washing with sterile distilled water, the cell pellet was resuspended in a 0.05 M sodium phosphate buffer (pH 7.0) and fixed with 2% glutaraldehyde for 30 min. Dehydration of the cell pellets was performed using absolute ethanol at different concentrations ranging from 10 to 100% for 30 min at each concentration. The resulting cells were air-dried, sputter-coated with 1–2 nm gold, and monitored using field emission scanning electron microscopy (FESEM) (MIRA, TESCAN).

2.5. Screening and Optimization Conditions for Ethanol Production by Z. mobilis

Several factors or variables have been shown to influence ethanol fermentation efficiency under high-temperature conditions [3,5,6,9,10]. In this study, a statistical experimental model based on the Plackett and Burman design (PBD) was employed to screen for the most significant factors influencing ethanol production by Z. mobilis R301 at 30 and 40 °C, including sugar concentration (A), pH of the fermentation medium (B), initial cell concentration (C), yeast extract (D), diammonium phosphate (DAP) (E), and magnesium sulfate (MgSO₄) (F). Batch ethanol fermentation using a 250-mL Erlenmeyer flask containing 100 mL of SSJ (pH 5.0) was performed in triplicate based on the experimental run designed using the Design Expert 7.0 demo version (STAT EASE Inc., Minneapolis, MN, USA). The actual values of each variable (low and high levels) are summarized in

Table 1. Codes and actual values of the independent variables for the PBD experiment.

Code	Variable	Unit	Low Level (−1)	High Level (+1)
A	Sugar concentration	g/L	160	220
B	pH		5	7
C	Cell concentration	% (v/v)	5	10
D	Yeast extract	g/L	2	12
E	DAP	g/L	1.2	6.0
F	MgSO4	g/L	0.5	3.0

, and the ethanol concentration was used as the response variable in this study.

The fermentation factors that significantly affect ethanol production by Z. mobilis at 30 and 40 °C were selected and subjected to optimization experiments using response-surface methodology (RSM) based on central composite design (CCD). The statistically significant factors were estimated using analysis of variance (ANOVA), and the optimized conditions predicted based on the response-surface plot were validated using batch ethanol fermentation carried out in a 250 mL Erlenmeyer flask. During ethanol fermentation, samples were collected, and the ethanol concentration was determined using gas chromatography (GC).

2.6. Gene Expression Analysis Using RT–qPCR during Ethanol Fermentation

Total RNA was prepared by culturing Z. mobilis wild-type and recombinant cells under the optimum fermentation conditions selected based on the optimization experiment using SSJ as feedstock. After inoculation of the bacterial cells into a fermentation medium with an initial cell density (OD₅₅₀) of 0.05, the cultures were incubated at 30 and 40 °C and 100 rpm for 18 h (exponential growth phase) and 36 h (stationary growth phase). Cells were separated by centrifugation at 10,000 rpm and 4 °C for 5 min and washed twice with sterile distilled water. Total RNA was isolated from the resulting cells using a GF-1 total RNA extraction kit (Vivantis, Malaysia), and the quality and quantity of isolated RNA were determined using a BioDrop μLite (Denville Scientific Inc., South Plainfield, NJ, USA). RT–qPCR was performed in triplicate using the Applied Biosystems^TM QuantStudio^TM 5 Real-Time PCR System and the MeltDoctor^TM HRM Master mix (Applied Biosystems, Waltham, MA, USA). The RT–qPCR preparation and thermal cycling conditions followed the manufacturer’s instructions. The primer pairs used for RT–qPCR are listed in

Table 2. List of specific primers used for gene expression analysis by RT–qPCR.

No.	Gene	Gene Code	Primer Sequence (5′→3′)
1	Pyruvate decarboxylase	PDC–F	GAC TAC AAC CTC GTC CT
		PDC–R	CAG GGC ATG GGA GCA AT
2	Alcohol dehydrogenase I	ADH I–F	CAT GAA AGC AGC CGT CA
		ADH I–R	TAC ACC CGC GCA AGT GA
3	Alcohol dehydrogenase II	ADH II–F	GTC AAC GAA ATG GGC GA
		ADH II–R	GTG ACG GTC AAC AAT GG
4	Glyceraldehyde-3-phosphate dehydrogenase	G3PD–F	GCA AAG GTT AGC CAT TCC AA
		G3PD–R	GCA ATA CCA AAG CGG TTC AT
5	Small heat shock protein (Cochaperonin)	GroES–F	AGA AAA GAC AGC TGG CGG TA
		GroES–R	GAC CAT TTG CCG AAC AGA AC
6	Heat shock protein 60 (Chaperonin)	GroEL–F	ACC TTG AAC ATG CTG GGT TC
		GroEL–R	GTT GCA CCG CCA ACT TTA AT
7	Heat shock protein 70 (DnaK)	Hsp70–F	CAG CCG TTG AAT AGA CCT GC
		Hsp70–R	GAA CCC GGA TGA AGT CGT TG
8	Heat shock protein 100	Hsp100–F	CGA CCA TCC GAG GTT TCT AA
		Hsp100–R	GGA ACG ACG GCA CAA TAT CT
9	16S rDNA	16S rDNA–F	CAG CAC CTG TCT CTG ATC CA
		16S rDNA–R	GTT CGG AAT TAC TGG GCG TA

, and the 16S rRNA gene was used as an internal control. The expression of genes was calculated using the comparative critical threshold (2^−ΔΔCT) method and expressed as a relative expression level [33].

2.7. Analytical Methods

Growth of the Z. mobilis cells was measured as optical density (OD) at 550 nm using a spectrophotometer (UV-1601, Shimadzu, Kyoto, Japan). Total sugar was determined by the phenol–sulfuric acid method [34] using glucose as a standard. The ethanol concentration (P, g/L) was evaluated by GC (GC-14B, Shimadzu, Kyoto, Japan) using a polyethylene glycol (PEG-20 M) packed column and a flame ionization detector. N₂ was used as the carrier gas, and 2-propanol (10% v/v) was used as the internal standard. The volumetric ethanol productivity (Qp, g/L.h) and ethanol yield (Yp/s, g/g) were calculated as described by Nuanpeng et al. [5]. All the experiments, except the RT–qPCR analysis, were performed in duplicate and repeated twice, and the results were expressed as the mean ± SD. The mean differences between each treatment were analyzed by Duncan’s multiple-range test (DMRT) at a probability of p ≤ 0.05 using the SPSS program for Windows.

3. Results and Discussion

3.1. Effect of Stress Conditions on Growth of Z. mobilis Wild-Type and Recombinant Strains

During ethanol fermentation, heat stress due to a rising environmental temperature or heat generated from microbial metabolic activity caused adverse effects on cell growth, cell viability, and ethanol fermentation activity. Several ethanologenic microorganisms, including Z. mobilis, are sensitive to heat stress. This study evaluated and compared the growth performance of wild-type Z. mobilis TISTR548 and recombinant Z. mobilis R301 under heat stress, and the results are illustrated in Figure 1A. The growth of the wild-type and recombinant strains at 30 °C was not significantly different. However, the recombinant strain exhibited approximately 1.4- and 4.3-fold higher growth than the wild-type strain at 37 and 40 °C, respectively. Relatively low growth of the wild type was observed at 40 °C, in agreement with that reported by Sootsuwan et al. [30] and Kaewchana et al. [16]. The ability of recombinant Z. mobilis R301 to withstand heat stress was similar to that of the thermal-adapted strain Z. mobilis AD41 [14] and the ethyl methane sulfonate mutagenized strain Z. mobilis EMS-229 [15].

Lignocellulosic biomass is one of the most promising feedstocks for ethanol production since it does not impact food security. Before being used as feedstock for ethanol production, the biomass must be pretreated to convert the significant components, specifically cellulose and hemicellulose, into fermentable sugars. Among the different pretreatment methods, the chemical process using dilute acid is the most widely used. Although it provides several advantages, such as high efficiency in separating cellulose and hemicellulose with low operation cost, high yield of fermentable sugar, and applicability to a wide range of biomass, the chemical process also generates lignocellulosic byproducts known as lignocellulosic inhibitors apart from fermentable sugars. Weak acids and furan derivatives, specifically acetic acid and furfural, respectively, are the most predominant lignocellulosic inhibitors detected in dilute acid lignocellulosic hydrolysate, in which their concentrations vary depending on the type and concentration of biomass and the pretreatment conditions [35,36]. These two inhibitors can inhibit microbial growth, reduce cell viability, and disrupt metabolic activity, resulting in low ethanol fermentation efficiency, specifically at high concentrations [7,14,15,31]. In this study, the inhibitory effects of acetic acid and furfural at different concentrations on the growth of the wild-type and recombinant strains were evaluated, and the results are summarized in Figure 1. At 30 °C, the growth of Z. mobilis wild-type and recombinant strains was not significantly different when grown in a medium without acetic acid supplementation (control treatment). When acetic acid was added to the cultivation medium at 150 and 200 mM, the recombinant Z. mobilis R301 exhibited remarkably higher growth than the wild-type strain, approximately 1.7- and 13.6-fold more significant than the wild-type strain at 150 and 200 mM acetic acid, respectively (Figure 1B). Notably, wild-type growth was almost inhibited at 200 mM acetic acid at this temperature. Similar growth profiles of wild-type and recombinant Z. mobilis under acetic acid stress were also observed at 37 °C. The growth of the wild-type and recombinant strains was not significantly different when grown under control conditions without acetic acid stress. However, the recombinant strain exhibited approximately 3- and 20-fold higher growth than the wild type at 150 and 200 mM acetic acid, respectively. This finding indicated that the severity of the effect of acetic acid on the growth of Z. mobilis, specifically wild type, was more pronounced under a high-temperature condition than under a normal growth temperature, which was in line with that reported in Z. mobilis ZM481 (ATCC 31823) [31].

The inhibitory effect of furfural at different concentrations on the growth of Z. mobilis wild-type and recombinant strains at 30 and 37 °C is summarized in Figure 1C. In the control condition without furfural stress, the growth of the wild-type and recombinant strains was not significantly different at both temperatures tested. However, the recombinant strain exhibited remarkably higher growth than the wild type, specifically under stress conditions. The growth of recombinant Z. mobilis R301 at 30 °C was approximately 2.3- and 2.4-fold higher, and at 37 °C, it was approximately 3.4- and 20-fold higher than the wild type in the medium supplemented with 10 and 20 mM furfural, respectively. This finding demonstrated that under high-temperature conditions, furfural strongly inhibited the growth of Z. mobilis compared to the control temperature, in line with that observed under acetic acid stress.

Based on the current study, we speculated that the high growth performance of recombinant Z. mobilis R301 under heat, acetic acid, and furfural stress might be correlated with the overexpression of the groESL genes since the GroESL chaperone protein is associated with stress tolerance that can help regulate proper protein assembly and folding and degradation of denatured proteins under stress conditions [25,26,37]. The results in the current study are in agreement with those of Kaewchana et al. [16], who reported that overexpression of the groESL genes in Z. mobilis improved the ability of bacterial cells to withstand high concentrations of sugar and ethanol up to 300 g/L and 102.57 g/L, respectively.

3.2. Bacterial Cells Morphology of Z. mobilis Wild-Type and Recombinant Strains under Heat Stress

It has been previously reported that heat stress inhibits not only cell growth and viability but also microbial cell morphology, including the surface and size of the cells [14,15,32]. This study investigated the effect of heat stress at 40 °C on the cell morphology of wild-type Z. mobilis TISTR548 and recombinant Z. mobilis R301 in the mid-exponential and stationary growth phases using FESEM, and the results are illustrated in Figure 2 and

Table 3. The cell size of wild-type Z. mobilis TISTR548 and recombinant Z. mobilis R301 cultured under heat stress.

Temperature (°C)	Growth Phase	Cell Size (μm)
		Z. mobilis TISTR548	Z. mobilis R301
30	Mid-exponential	2.12 ± 0.33	2.14 ± 0.22
	Stationary	2.20 ± 0.33	2.55 ± 0.49
40	Mid-exponential	2.49 ± 0.32	2.60 ± 0.36
	Stationary	2.37 ± 0.34	2.71 ± 0.36

. At 30 and 40 °C, the cell surface morphology of the Z. mobilis wild-type and recombinant strains was similar at both stages of growth. However, the size of bacterial cells differed depending on the growth phase and incubation temperature. At 30 °C, wild-type and recombinant Z. mobilis cells were similar in size at the mid-exponential growth phase. However, the recombinant strain exhibited elongated cells compared with the wild-type strain at the stationary growth phase. At a high temperature of 40 °C, the recombinant Z. mobilis R301 displayed longer cells than the wild type at both mid-exponential and stationary growth phases. It was noteworthy that increasing the incubation temperature from 30 °C to 40 °C or shifting the bacterial growth from the mid-exponential to the stationary phase caused bacterial cell elongation, except in the wild-type strain incubated at 40 °C, in which shifting of the growth phase did not affect cell size. One possibility to explain the elongation of bacterial cells under high-temperature conditions and in the stationary growth phase is that heat and starvation due to the depletion of nutrients in the medium trigger other stressful conditions, specifically intracellular oxidative stress, which causes DNA damage and leads to the prohibition of cell division [32,38]. Since the mechanism-relating oxidative stress and cell division is complicated, involving several genes and proteins, metagenomic analysis of wild-type and recombinant strains under stress conditions may be needed to clarify this phenomenon.

3.3. Screening and Optimization Conditions for Ethanol Production by Z. mobilis

The ethanol production efficiency depends not only on the microbial species but also on environmental factors. Several fermentation variables have been shown to influence ethanol yield and productivity, specifically under high-temperature conditions [3,5,6,9,10]. Screening the main significant variables in ethanol production is one of the main steps before optimizing the conditions. Additionally, to minimize the number of experiments, statistical experimental models, such as CCD and Box–Behnken design (BBD), are widely used instead of modulating one factor at a time for optimization conditions. This study employed PBD and RSM based on CCD, the most widely used technique, for screening and optimizing conditions for ethanol production by recombinant Z. mobilis R301 at 30 and 40 °C using SSJ as feedstock. Six independent variables (Table 1) that may affect ethanol production by Z. mobilis R301 at 30 °C were first screened using PBD. Based on the PBD using the Design Expert program, 12 experimental runs were generated, and the response variable for each run, i.e., ethanol concentration at 54 h of fermentation, is presented in Supplementary Table S1.

The ethanol concentration produced by the recombinant Z. mobilis R301 varied depending on the fermentation conditions. The maximum ethanol concentration of 63.46 g/L was detected in run number 1 (160 g/L sugar concentration, pH 7.0, 10% (v/v) cell concentration, 12 g/L yeast extract, 1.2 g/L DAP, and 0.5 g/L MgSO₄), while the lowest ethanol concentration of 3.60 g/L was observed in run number 10 (220 g/L sugar concentration, pH 5.0, 5% (v/v) cell concentration, 2 g/L yeast extract, 1.2 g/L DAP, and 3.0 g/L MgSO₄), which was in line with the predicted values. Notably, a high sugar concentration affects ethanol production by Z. mobilis R301 negatively.

The reliability of the established model generated based on the PBD was tested using ANOVA at a p value of 0.05, and the results are shown in

Table 4. ANOVA of the PBD for screening the significant independent variables influencing ethanol production from SSJ at 30 °C by recombinant Z. mobilis R301.

Source	Sum Square	df	Mean Square	F Value	p Value Prob > F	Remark
Model	6091.79	6	1015.30	53.27	0.0002	Significant
A-sugar	5456.85	1	5456.85	286.30	<0.0001	Significant
B-pH	83.98	1	83.98	4.41	0.0899	Not significant
C-cell concentration	276.82	1	276.82	14.52	0.0125	Significant
D-yeast extract	195.09	1	195.09	10.24	0.0240	Significant
E-DAP	79.03	1	79.03	4.15	0.0973	Not significant
F-MgSO4	0.02	1	0.02	0.001	0.9777	Not significant
Residual	95.30	5	19.06
Cor. Total	6187.09	11
R-squared	0.9846
Adj. R-square	0.9661

. The results demonstrated that the p value of the established model was lower than 0.05, meaning that it was highly reliable and could be used to select the significant independent variables affecting ethanol production from SSJ by Z. mobilis R301 at 30 °C. Furthermore, a relatively high coefficient of determination (R² value of 0.9846 and adjusted R² value of 0.9661) implied that the established model (see on next page, Equation (1)) was highly accurate in explaining the variability in the response. Based on the ANOVA in Table 4, three independent variables, including sugar concentration (A), cell concentration (C), and yeast extract (D), were found to be the most significant variables (p value less than 0.05) affecting ethanol production by Z. mobilis R301 at 30 °C, while the other variables had no significant effect on ethanol production. Thus, these vital independent variables were selected for the optimization experiment.

RSM based on CCD was performed to identify the optimum levels of sugar, bacterial cell, and yeast extract concentrations for ethanol production from SSJ at 30 °C by the recombinant Z. mobilis R301. Twenty experimental runs based on the experimental design matrices and the response variable (ethanol concentration, g/L) are summarized in Supplementary Table S2. The actual ethanol concentrations detected from different combinations of each factor ranged from 4.86 to 62.52 g/L, in line with the predicted values. A quadratic polynomial regression model and a second-order polynomial Equation (1) for predicting the final ethanol concentration (P, g/L) as a function of sugar concentration (A), cell concentration (B), and yeast extract (C) concentration were developed using the actual response variables in Supplementary Table S2, and the resulting prediction Equation (1) was as follows:

P (g/L) = +50.24 − 15.21A + 10.17B + 6.37C + 4.01AB − 0.16AC + 1.46BC − 6.29A² − 3.96B² − 2.85C²

(1)

The statistical significance of the established model was evaluated using ANOVA at a p value of 0.05, and the results revealed that the model was highly significant (

Table 5. ANOVA of the CCD for ethanol production from SSJ at 30 °C by recombinant Z. mobilis R301.

Source	Sum Square	df	Mean Square	F Value	p Value Prob > F	Remark
Model	6059.22	9	673.25	88.42	0.0001	Significant
A-sugar	3159.88	1	3159.88	415.01	0.0001	Significant
B-cell concentration	1412.65	1	1412.65	185.53	0.0001	Significant
C-yeast extract	554.11	1	554.11	72.78	0.0001	Significant
AB	128.68	1	128.68	16.90	0.0021	Significant
AC	0.22	1	0.22	0.029	0.8696	Not significant
BC	17.16	1	17.16	2.25	0.1642	Not significant
A2	570.47	1	570.47	74.92	0.0001	Significant
B2	225.79	1	225.79	29.65	0.0003	Significant
C2	117.04	1	117.04	15.37	0.0029	Significant
Residual	76.14	10	7.61
Lack of fit	63.25	5	12.65	4.91	0.0528	Not significant
Pure Error	12.89	5	2.58
Cor. Total	6135.36	19
R-squared	0.9876
Adj. R-square	0.9764

). In addition to the linear terms of each independent variable tested, the quadratic terms of the sugar concentration (A²), cell concentration (B²), and yeast extract concentration (C²), as well as the interaction between sugar concentration and cell concentration (AB), were also statistically significant. A high R-squared value (0.9876) of the regression model and the statistically insignificant “lack of fit” of the model strongly implied that the established model was reliable and highly accurate (98.76%) in explaining the variability of the response values based on the description of Myers et al. [39].

Based on the experimental data in Equation (1), three-dimensional (3D) response surface plots of all significant independent variables for ethanol production from SSJ at 30 °C by Z. mobilis R301 were generated, and the results are illustrated in Figure 3. The results indicate that high bacterial cell (Figure 3A) or yeast extract concentrations (Figure 3B) slightly increase the ethanol concentration. However, the ethanol content remarkably decreased when sugar concentrations increased, possibly due to an adverse effect of high osmotic pressure on cell growth and fermentation activity [40,41]. In contrast, the ethanol concentration was increased if the concentrations of bacterial cells and yeast extract increased and the sugar concentration was set at the appropriate level (Figure 3C). This finding strongly indicates that ethanol production by Z. mobilis R301 at 30 °C was impacted by all selected variables, specifically sugar concentration.

The confirmatory experiment using the experimental data from the CCD and the quadratic polynomial equation was performed to verify the predicted optimum values. Based on the validated model, the optimum values of each significant independent variable were 171.67 g/L sugar concentration, 9.42% (v/v) cell concentration, and 10.89 g/L yeast extract. The maximum ethanol concentration from the confirmatory experiment was 63.26 g/L, which was relatively close to the predicted value (64.35 g/L). The ethanol productivity and yield detected in this study were 1.17 g/L.h and 0.45 g/g, respectively. This finding indicated that the established model in this study was reliable and reproducible and could be used to identify the optimal conditions for ethanol production from SSJ at 30 °C by recombinant Z. mobilis R301.

One of the aims of this study was to compare the ethanol production efficiency of recombinant Z. mobilis R301 under normal and HTEF conditions. Therefore, the optimization conditions for ethanol production from SSJ by Z. mobilis R301 at 40 °C were also determined in this study. Six fermentation parameters, the same as those determined at 30 °C (Table 1), were used as independent variables for screening by PBD, and ethanol concentration was used as the response variable. The PBD matrices comprising 12 experimental runs and the predicted and actual values of ethanol concentrations generated in this study are presented in Supplementary Table S3. The ethanol contents produced by recombinant Z. mobilis R301 under different fermentation conditions varied, ranging from 1.23 to 61.95 g/L, which aligned with the predicted values. Notably, the ethanol contents produced at a high temperature of 40 °C were slightly lower than those produced at 30 °C, suggesting that high temperature or heat stress negatively affected the fermentation activity of the recombinant Z. mobilis cells, similar to other ethanologenic microorganisms [3,5,6,7,10,11]. The fermentation conditions of experimental run number 1 (160 g/L sugar concentration, pH 7.0, 10% (v/v) cell concentration, 12 g/L yeast extract, 1.2 g/L DAP, and 0.5 g/L MgSO₄) and experimental run number 10 (220 g/L sugar concentration, pH 5.0, 5% (v/v) cell concentration, 2 g/L yeast extract, 1.2 g/L DAP, and 3.0 g/L MgSO₄) yielded the highest and lowest values of ethanol concentration, respectively, similar to those conditions observed in the screening experiments at 30 °C. Again, a high sugar concentration tended to negatively affect ethanol production by Z. mobilis R301 at 40 °C.

The effectiveness of the established model generated based on the PBD in screening the significant independent variables affecting ethanol production from SSJ at 40 °C by Z. mobilis R301 was tested by ANOVA at a p value of 0.05. The results in

Table 6. ANOVA of the PBD for screening the significant independent variables influencing ethanol production from SSJ at 40 °C by recombinant Z. mobilis R301.

Source	Sum Square	df	Mean Square	F Value	p Value Prob > F	Remark
Model	4396.72	6	732.79	1290.00	0.0065	Significant
A-sugar	652.54	1	652.54	11.48	0.0195	Significant
B-pH	1017.71	1	1017.71	17.91	0.0082	Significant
C-cell concentration	302.30	1	302.30	5.32	0.0692	Not significant
D-yeast extract	778.44	1	778.44	13.70	0.0140	Significant
E-DAP	91.91	1	91.91	1.62	0.2594	Not significant
F-MgSO4	1553.83	1	1553.83	27.34	0.0034	Significant
Residual	284.13	5	56.83
Cor. Total	4680.85	11
R-squared	0.9393
Adj. R-square	0.8665

indicated that the established model was highly reliable since the p value of the model was lower than 0.05. In addition, the model was also found to be highly accurate in explaining the variability of the response values since the coefficient of determination (R² value) was also high (0.9393). Four fermentation parameters, including sugar concentration (A), pH (B), yeast extract (D), and MgSO₄ (F), were the significant critical variables affecting ethanol production by Z. mobilis R301 at 40 °C, while the remaining variables had no significant effect on ethanol production based on the p value of each variable. Therefore, all four variables were selected for further experiments.

The optimum level of each significant independent variable influencing the ethanol production from SSJ at 40 °C was analyzed using the RSM in the CCD. Based on the CCD matrices using the ethanol concentration as a response variable, 30 experimental runs with a combination of each variable were generated, for which the codes and actual values for each variable are summarized in Supplementary Table S4. The ethanol contents derived from each experimental run align with the predicted values, ranging from 21.97 g/L to 59.89 g/L. Run numbers 25 (200 g/L sugar concentration, pH 7.5, 1 g/L MgSO₄, and 12 g/L yeast extract) and number 20 (140 g/L sugar concentration, pH 6.5, 2 g/L MgSO₄, and 8 g/L yeast extract) yielded the highest and lowest ethanol concentrations, respectively. The experimental data in Supplementary Table S4 were used to develop a quadratic polynomial regression model and a second-order polynomial Equation (2) for the prediction of ethanol concentration under the effects of sugar concentration (A), pH (B), MgSO₄ (C), and yeast extract (D). As a result, an Equation (2) for ethanol concentration prediction derived in this study was as follows:

P (g/L) = +51.69 + 6.67A + 1.75B + 1.17C − 4.20D − 0.55AB + 3.80AC − 1.53AD − 0.78BC − 1.78BD − 0.13CD − 4.53A² − 2.74B² − 3.94C² − 1.59D²

(2)

The statistical significance of the established model, the linear and quadratic terms, and the interaction between each significant variable was determined using ANOVA at a 95% confidence interval. As shown in

Table 7. ANOVA of the CCD for ethanol production from SSJ at 40 °C by recombinant Z. mobilis R301.

Source	Sum Square	df	Mean Square	F Value	p Value Prob > F	Remark
Model	2880.42	14	205.74	268.31	<0.0001	Significant
A-sugar	1068.40	1	1068.40	1393.31	<0.0001	Significant
B-pH	73.19	1	73.19	95.44	<0.0001	Significant
C-MgSO4	32.64	1	32.64	42.57	<0.0001	Significant
D-yeast extract	423.28	1	423.28	552.00	<0.0001	Significant
AB	4.81	1	4.81	6.27	0.0243	Significant
AC	230.51	1	230.51	300.61	<0.0001	Significant
AD	37.42	1	37.42	48.80	<0.0001	Significant
BC	9.69	1	9.69	12.63	0.0029	Significant
BD	50.73	1	50.73	66.16	<0.0001	Significant
CD	0.26	1	0.26	0.34	0.5708	Not significant
A2	561.85	1	561.85	732.71	<0.0001	Significant
B2	205.31	1	205.31	267.75	<0.0001	Significant
C2	426.26	1	426.26	555.89	<0.0001	Significant
D2	69.64	1	69.64	90.82	<0.0001	Significant
Residual	11.50	15	0.77
Lack of fit	10.23	10	1.02	4.02	0.0688	Not significant
Pure Error	1.27	5	0.25
Cor. Total	2891.92	29
R-squared	0.9960
Adj. R-square	0.9923

, the established model was highly significant because the p value was lower than 0.05, suggesting that this model is reliable and could be used to predict the optimum conditions for ethanol production from SSJ at 40 °C by recombinant Z. mobilis R301. The linear (A, B, C, and D) and quadratic terms (A², B², C², and D²) and the interaction of each of the selected significant variables, except the interaction between MgSO₄ (C) and yeast extract (D), were statistically significant, indicating that all of the selected variables significantly influence ethanol production at 40 °C. Based on the R-square value (0.9960) and a nonstatistically significant difference in the “lack of fit” of the model, the established model developed in this study was reliable, and the variability of the response values could be explained using the established model with high accuracy (99.60%).

The relationships between each selected significant independent variable influencing ethanol production from SSJ at 40 °C by Z. mobilis R301 were determined by 3D response surface plots using the data in Equation (2). As shown in Figure 4A, setting the MgSO₄ and yeast extract at the appropriate levels and increasing the sugar concentration and pH of the fermentation medium increased ethanol content. However, the ethanol concentration decreased at the highest sugar level (200 g/L) and pH (7.5). Similarly, the ethanol concentration was also increased when the concentrations of sugar and yeast extract increased, but not at the highest levels (Figure 4B). In contrast, a high sugar concentration and a relatively low level of MgSO₄ resulted in a high ethanol content (Figure 4C). When sugar and MgSO₄ concentrations were fixed, increasing the pH to approximately 7.0 and yeast extract concentration to approximately 9 g/L yielded the maximum ethanol content compared with other levels (Figure 4D). Considering the relationship between MgSO₄ and pH (Figure 4E) or MgSO₄ and yeast extract (Figure 4F) for ethanol production, when other factors were fixed, the ethanol concentration tended to increase in response to increasing the pH level or yeast extract concentration and lowering the MgSO₄ concentration.

Verification of the predicted optimum condition derived in this study was performed by conducting a confirmatory experiment using the experimental data from the CCD and the quadratic polynomial equation. The optimum fermentation conditions for ethanol production at 40 °C by Z. mobilis R301 using SSJ as feedstock were 199.48 g/L sugar concentration, 10.88 g/L yeast extract, 1.05 g/L MgSO₄, and a pH of 6.8. The ethanol concentration, productivity, and yield attained under the optimum conditions were 58.62 g/L, 1.22 g/L.h, and 0.43 g/g, respectively, in which the actual ethanol content detected in this study was very close to the predicted value (59.88 g/L). The results clearly demonstrated that the established model obtained in the current study was reliable and reproducible for predicting the optimum conditions for ethanol production from SSJ at a high temperature of 40 °C by the recombinant Z. mobilis R301.

Comparative analysis of ethanol production by recombinant Z. mobilis R301 using SSJ as feedstock with other Z. mobilis strains using different raw materials is summarized in

Table 8. Comparison of ethanol production by recombinant Z. mobilis R301 and the other Z. mobilis strains using different substrates.

Strain	Substrate	T (°C)	P (g/L)	Qp (g/L/h)	Yp/s (g/g)	Reference
Z. mobilis ATCC 29191	Molasses	30	55.60	1.16	0.34	Cazetta et al. [42]
Z. mobilis MTCC 92	Molasses	30	58.70	0.61	0.42	Behera et al. [43]
Z. mobilis TISTR548	Glucose	37	41.87	1.16	0.46	Sootsuwan et al. [30]
Z. mobilis ZM4	Corncob	30	54.42	0.91	N/A	Gu et al. [44]
Z. mobilis ZMT2	Manure hydrolysate	30	10.55	0.44	0.37	You et al. [45]
Z. mobilis R301	SSJ	30	63.26	1.17	0.45	This study
	SSJ	40	58.62	1.22	0.43	This study

. The ethanol concentration, productivity, and yield obtained in this study by Z. mobilis R301 were remarkably higher than those produced by Z. mobilis ATCC 29191 [42], MTCC 92 [43], ZM4 [44], and ZMT2 [45] at 30 °C. Interestingly, the recombinant Z. mobilis R301 produced comparatively higher ethanol concentration and productivity at 40 °C than that of Z. mobilis TISTR548 at 37 °C, suggesting that this recombinant strain is a good candidate for HTEF using SSJ as feedstock.

Considering the fermentation parameters affecting ethanol production in this study, sugar concentration and yeast extract are the most significant independent variables affecting ethanol production at both temperature conditions. In addition to sugar and yeast extract, the microbial cell concentration is more important for ethanol production at 30 °C, while MgSO₄ and pH play critical roles at 40 °C. Sugar concentration has been shown to affect the efficiency of ethanol production by microbial fermentation. In this study, increasing the sugar concentration in the fermentation medium tended to decrease the ethanol concentration under normal and high-temperature conditions, similar to the results reported by Nuanpeng et al. [5]. As demonstrated by Bai et al. [40] and Ozmichi and Kargi [41], high sugar concentrations cause high osmotic pressure, negatively impacting cell viability, morphology, and metabolic activity, resulting in reduced cell growth and ethanol fermentation efficiency. High levels of sugar, specifically more than 250 g/L, also caused a reduction in the substrate conversion rate, leading to a reduction in ethanol productivity and yield [5,41]. The current study shows that the optimum sugar concentrations were 171.67 g/L at 30 °C and 199.48 g/L at 40 °C, which was lower than those reported by Techaparin et al. [3] and Nuanpeng et al. [46] using the thermotolerant yeasts S. cerevisiae KKU-VN8 and S. cerevisiae DBKKU Y-53, respectively. It should be noted from the current study that approximately 30 and 62 g/L of sugar remained at the end of fermentation at 30 and 40 °C, respectively. Other fermentation techniques to improve the sugar utilization efficiency of the Z. mobilis recombinant strain should be investigated, such as fed-batch fermentation or continuous fermentation.

Nitrogen sources are among the most important compounds for microbial growth and metabolism. Several organic and inorganic nitrogen sources have been shown to affect the ethanol production performance of many ethanologenic microorganisms. Among the nitrogen sources, yeast extract has been recognized as a good organic nitrogen source that positively affects ethanol production, specifically under very high gravity and in high-temperature fermentation [6,47,48]. In addition to having high concentrations of nitrogen and vitamins, specifically vitamin B, yeast extract also contains some trace elements essential for microbial growth and enzyme activity [49]. During ethanol production, the availability of yeast extract to be used by the microbial cell depends on the microbial species, fermentation conditions, and feedstock used. In this study, yeast extract was considered an essential factor for ethanol production at 30 and 40 °C, and its optimum concentration was approximately 10.88–10.89 g/L, in line with that reported by Deesuth et al. [50].

The ethanol production efficiency can be improved by increasing the initial cell concentration of ethanologenic microbes. Several investigations revealed that increasing the microbial cell concentration enhanced the substrate consumption rate, ethanol productivity, and yield [3,5,6,9,46]. The current study also demonstrated that the concentration of Z. mobilis cells was one of the critical factors affecting ethanol production at 30 °C. At 40 °C, however, other factors, specifically MgSO₄ and pH, significantly impacted ethanol production compared to bacterial cell concentration. MgSO₄ is a crucial divalent cation involved in several physiological and biological processes, such as cell growth, division, and enzymatic activity. In yeast, MgSO₄ has been shown to play critical roles in cellular protection, stabilization, and recovery from stress [51,52]. Additionally, as demonstrated by Thanonkeo et al. [53], supplementing MgSO₄ into a culture medium improved cell growth and viability as well as ethanol production efficiency of Z. mobilis under heat at 40 °C and ethanol stress at 110.46 g/L. Similar to the current study, MgSO₄ is also an essential fermentation factor affecting ethanol production under high-temperature conditions, and the concentration of 1.05 g/L was considered the optimum level for ethanol production from SSJ by Z. mobilis R301 at 40 °C.

The ethanol production efficiency also depended on the pH of the fermentation medium. The activity of enzymes involved in the metabolic activity and ethanol production pathway is correlated with the pH level, i.e., lower and higher pH levels than the optimum value may inactivate enzyme activity, resulting in a reduction in cell growth and ethanol fermentation ability [54]. For ethanol production, the optimum pH varied depending on microbial species and fermentation conditions, such as temperature, aeration, and raw material used. In this study, the pH of the fermentation medium proved to be a critical factor influencing ethanol production at 40 °C by Z. mobilis R301, and the optimum pH value was 6.8, in line with the optimum value for bacterial growth (pH 5.0–7.0) [55].

3.4. Gene Expression Analysis Using RT–qPCR during Ethanol Fermentation

In response to heat, ethanol, oxidative, and other stresses, a large number of thermotolerant genes with classified into nine groups, including genes for metabolism, membrane stabilization, transport, DNA repair, tRNA modification, protein quality control, translation control, cell division, and transcriptional regulation, have been shown to be up- or downregulated in Z. mobilis [25,56]. Approximately 1.5% of Z. mobilis genomic genes are thermotolerant genes, which associate with fundamental activities of cells and cell survival at a critical high temperature. Although several studies have reported on the expression of genes involved in responses to stresses such as heat, ethanol, and acetic acid, less information is available regarding the expression of genes involved in the heat shock response and ethanol production pathway in wild-type Z. mobilis TISTR548 and recombinant Z. mobilis R301 overexpressing the groESL gene during HTEF using SSJ as feedstock. In this study, the expression of groESL and other genes involved in the synthesis of heat shock proteins (HSPs) and metabolism pathway for ethanol production in the wild-type and recombinant strains of Z. mobilis at 18 h (mid-exponential phase) and 36 h (stationary phase) after ethanol fermentation using SSJ as feedstock was evaluated by RT–qPCR. The results are summarized in

Table 9. The relative expression levels of the genes in recombinant Z. mobilis R301 during ethanol fermentation at 30 and 40 °C measured by RT–qPCR analysis.

Gene	Relative Expression Level (Fold) 1
	30 °C		40 °C
	Mid-Exponential	Stationary	Mid-Exponential	Stationary
pdc	0.98 ± 0.02 cB	1.02 ± 0.03 deB	3.10 ± 0.11 hA	3.09 ± 0.08 gA
adh I	1.18 ± 0.01 aC	0.96 ± 0.01 eD	4.08 ± 0.06 gA	3.70 ± 0.09 fB
adh II	0.87 ± 0.07 dD	1.14 ± 0.03 bC	5.08 ± 0.08 fB	6.59 ± 0.04 cA
g3pd	1.08 ± 0.02 bC	1.05 ± 0.01 cdC	5.46 ± 0.06 eA	4.07 ± 0.03 eB
groES	1.17 ± 0.04 aC	1.33 ± 0.04 aC	36.06 ± 0.06 aA	6.33 ± 0.13 dB
groEL	1.08 ± 0.03 bC	1.09 ± 0.01 bcC	20.04 ± 0.05 bA	14.09 ± 0.09 aB
hsp70	1.19 ± 0.03 aC	1.14 ± 0.03 bC	13.84 ± 0.04 cA	3.89 ± 0.08 efB
hsp100	1.06 ± 0.01 bC	1.12 ± 0.02 bcC	7.78 ± 0.03 dB	12.33 ± 0.06 bA

¹ The expression of the genes was compared with that of the wild-type strain. Means ± SDs followed by different lowercase letters within a column are statistically different at p < 0.05, while the values with different capital letters within a row are statistically different at p < 0.05 based on DMRT analysis.

. At 30 °C, the expression levels of all investigated genes in the mid-exponential phase were not different from those of the wild type; the relative expression levels ranged from 0.87- to 1.19-fold compared with the wild type. However, the expression levels of the genes in the stationary phase, specifically the groES gene, were slightly higher than those of the wild type, whereas the expression levels of other genes were almost the same. At 40 °C, all genes involved in HSPs synthesis and ethanol production pathway of the recombinant Z. mobilis R301 in the mid-exponential and stationary phases were upregulated compared to the wild type. Notably, the expression levels of the heat shock-responsive genes, including groES, groEL, hsp70, and hsp100, were statistically higher than those of the genes involved in the ethanol production pathway. Among the genes tested, the expression of groES and groEL genes was highest at the mid-exponential phase, approximately 36- and 20-fold higher than in the wild type. Compared to the others, the highest gene expression levels were detected for groEL, hsp100, and groES in the stationary phase. The highest expression of groES and groEL in recombinant Z. mobilis R301 at a high temperature of 40 °C was likely due to the large number of copies of groESL genes inserted, in line with the work of Kaewchana et al. [16], who previously reported increased GroESL protein production in this recombinant strain.

GroES, GroEL, and other HSPs, such as HSP70 and HSP100, are molecular chaperones that play crucial roles in the folding of nascent polypeptides or refolding of denatured proteins, preventing the aggregation of unfolded or partially folded polypeptides and degrading denatured proteins [57]. In addition to normal growth conditions, the expression levels of these molecular chaperones are also upregulated under heat, ethanol, osmotic, and oxidative stress [25,26,56]. Thus, overexpression of groES, groEL, and other heat shock-responsive genes may help improve cell growth and viability and maintain the normal cell structural integrity and metabolic activity of Z. mobilis R301 under high-temperature conditions and high concentrations of furfural and acetic acid, similar to those reported in recombinant Escherichia coli [58] and Riemerella anatipestifer [59]. Additionally, the current study also pointed out that overexpression of groES and groEL genes also improved the ethanol fermentation efficiency of recombinant Z. mobilis R301 at 40 °C. One possible explanation is that the GroESL proteins and other molecular chaperones may protect proteins, specifically, those enzymes involved in the ethanol production pathway, from denaturation or aggregation caused by heat stress [37].

4. Conclusions

This study demonstrates that recombinant Z. mobilis R301 overexpressing groESL exhibited multistress tolerance toward high temperatures of up to 40 °C and acetic acid and furfural at 200 and 20 mM concentrations, respectively. This Z. mobilis recombinant strain produced maximum ethanol concentrations, productivities, and yields of 63.26 g/L, 1.17 g/L.h, and 0.45 g/g at 30 °C, and 58.62 g/L, 1.22 g/L.h, and 0.43 g/g at 40 °C, respectively, under the optimum fermentation conditions using SSJ as feedstock. Since recombinant Z. mobilis R301 exhibited multi-stress tolerance toward lignocellulosic inhibitors, experiments on ethanol production under high-temperature conditions using lignocellulosic hydrolysate as feedstock may be necessary for potential applications in the future.