Characterization and Application of the Sugar Transporter Zmo0293 from Zymomonas mobilis

Zymomonas mobilis is a natural ethanologen with many desirable characteristics, which makes it an ideal industrial microbial biocatalyst for the commercial production of desirable bioproducts. Sugar transporters are responsible for the import of substrate sugars and the conversion of ethanol and other products. Glucose-facilitated diffusion protein Glf is responsible for facilitating the diffusion of glucose uptake in Z. mobilis. However, another sugar transporter-encoded gene, ZMO0293, is poorly characterized. We employed gene deletion and heterologous expression mediated by the CRISPR/Cas method to investigate the role of ZMO0293. The results showed that deletion of the ZMO0293 gene slowed growth and reduced ethanol production and the activities of key enzymes involved in glucose metabolism in the presence of high concentrations of glucose. Moreover, ZMO0293 deletion caused different transcriptional changes in some genes of the Entner Doudoroff (ED) pathway in the ZM4-ΔZM0293 strain but not in ZM4 cells. The integrated expression of ZMO0293 restored the growth of the glucose uptake-defective strain Escherichia coli BL21(DE3)-ΔptsG. This study reveals the function of the ZMO0293 gene in Z. mobilis in response to high concentrations of glucose and provides a new biological part for synthetic biology.


Introduction
Molasses and waste starch from potato factories is a promising feedstock for biofuel production [1]. Glucose is the major component liberated from these wastes during hydrolysis. The efficient utilization of glucose by microorganisms is crucial for economic biofuel production. Zymomonas mobilis is a Gram-negative α-proteobacterium with attractive physiological properties for bioethanol production. It can rely on the unique Entner Doudoroff (ED) pathway to anaerobically metabolize glucose to produce ethanol. Additionally, this ethanologenic bacterium has some desirable industrial characteristics, such as a high ethanol titer, high specific glucose uptake rate, high ethanol tolerance of 16% (v/v), and a broad pH range (3.5-7.5, especially low pH), which makes it an excellent chassis for the production of biochemicals [2][3][4][5]. Currently, different CRISPR-Cas systems have been established in Z. mobilis that facilitate gene editing manipulations [6][7][8][9]. Moreover, significant progress has been made to expand substrate ranges and enhance robustness against lignocellulosic biomass hydrolysate inhibitors through systems biology and metabolic engineering in Z. mobilis. In addition, various biological parts of synthetic biology, such as promoters, sRNAs, 5 untranslated regions (5 UTRs), and functional genes, have been characterized [10][11][12][13]. Metabolic models and assembly strategies have also been established to guide and construct metabolic pathways, which makes Z. mobilis a promising microbial biocatalyst for economic biosynthesis to produce value-added products such as lactate, isobutanol, and polyhydroxybutyrate (PHB) [14][15][16].
Molecular transport is a key process for cellular growth and metabolism. The transport of glucose and other sugars is the first step of glycolysis and is also an important

ZMO02393 Deletion Attenuated Growth and Fermentation Ability of ZM4 under High Glucose Concentrations
Sugar transporters are responsible for the import of sugars and play an important role in producing valuable biofuels and chemicals. Structure prediction of Zmo0293 using AlphaFold illustrated that this transporter possesses twelve transmembrane regions with high confidence, which is similar to that of Glf (Supplementary Figure S1). To reveal the function of Zmo0293, we deleted this gene in ZM4 cells by the endogenous Type I-F CRISPR-Cas system (Supplementary Figure S2) [6]. The final OD of the ZM4-ΔZMO0293 mutant reached 4.0 and 6.0 in the presence of 20 and 100 g/L glucose, which is similar to Here, we aimed to analyze the function of the sugar transporter gene ZMO0293. Firstly, we obtained the ZMO0293 knockout mutant by the Type I-F CRISPR-Cas system and observed the influence on the growth rate, biomass, glucose consumption rate, and ethanol production rate. Secondly, qPCR was used to analyze the expression level of ZMO0293 and other ED pathway genes. Finally, the ZMO0293 gene was heterologously expressed to restore the growth of E. coli BL21(DE3)-∆ptsG. Our study was helpful for identifying the function of the sugar transporter gene ZMO0293 and provides a basis for understanding glucose uptake and sugar transporter engineering in Z. mobilis.

ZMO02393 Deletion Attenuated Growth and Fermentation Ability of ZM4 under High Glucose Concentrations
Sugar transporters are responsible for the import of sugars and play an important role in producing valuable biofuels and chemicals. Structure prediction of Zmo0293 using AlphaFold illustrated that this transporter possesses twelve transmembrane regions with high confidence, which is similar to that of Glf (Supplementary Figure S1). To reveal the function of Zmo0293, we deleted this gene in ZM4 cells by the endogenous Type I-F CRISPR-Cas system (Supplementary Figure S2) [6]. The final OD of the ZM4-∆ZMO0293 mutant reached 4.0 and 6.0 in the presence of 20 and 100 g/L glucose, which is similar to those of ZM4 (Figure 2A,B). However, this mutant showed an obvious growth defect in the presence of 240 g/L glucose. In particular, the final OD of ZM4-∆ZMO0293 was lower by 20.9% than that of the ZM4 control and had a longer lag phase of 12 h ( Figure 2C). the presence of 240 g/L glucose. In particular, the final OD of ZM4-ΔZMO0293 was lower by 20.9% than that of the ZM4 control and had a longer lag phase of 12 h ( Figure 2C). When metabolizing the same amount of glucose or fructose, ATP production through the ED pathway in ZM4 is half of that through the Embden-Meyerhof-Parnas (EMP) pathway by other microorganisms, such as S. cerevisiae. Therefore, the ED pathway has less biomass accumulation, and most of the carbon sources (>95%) are converted to ethanol under low glucose concentrations [7,29]. We evaluated the role of the ZMO0293 gene in ethanol fermentation. The ZM4-ΔZMO0293 mutant showed a slow glucose utilization rate and nearly the same ethanol titer in the presence of 20 g/L glucose compared to ZM4 (Figure 2D). The ZM4-ΔZMO0293 mutant exhibited a slow ethanol titer with almost the same glucose utilization rate in the presence of 100 g/L glucose ( Figure 2E). However, when 240 g/L glucose was added, the glucose utilization rate and ethanol titer of the ZM4-ΔZMO0293 mutant were markedly slower than those of ZM4 ( Figure 2F). After 96 h fermentation, the glucose consumption rate of ZM4-ΔZMO0293 was 2.3 g/L/h, 30.3% lower than that of ZM4 (3.3 g/L/h). Additionally, the ethanol rate of ZM4-ΔZMO0293 was 0.99 g/L/h, 34% lower than that of ZM4 (1.5 g/L/h).

Figure 2.
Effect of ZMO0293 deletion on growth and ethanol fermentation. The fresh inoculums of ZM4 and ZM4-ΔZMO0293 were transferred to 50 mL RMG medium with 1% inoculum and cultivated at 30 °C. The effect of the ZMO0293 gene on growth was evaluated by growth curve determination. (A-C) show the OD of ZM4 and ZM4-ΔZM0293 cultured in RMG medium supplemented with 20, 100, and 240 g/L glucose, respectively. The effect of the ZMO0293 gene on ethanol fermentation was evaluated by glucose consumption and ethanol production. (D-F) show the glucose and ethanol concentrations of ZM4 and ZM4-ΔZMO0293 grown in RMG medium containing 20, 100, and 240 g/L glucose, respectively. The samples were taken at set time intervals. Data are presented as the mean ± s.d. with three biologically independent samples.
Z. mobilis displays a long adaptation period under high concentrations of glucose or sucrose. Previous transcriptomic studies showed that ZM4 cells can regulate the transcription levels of genes associated with membrane channels and transporters, and metabolic pathways to adapt to high concentrations of glucose [28]. Our results showed that ZMO0293 deletion (ZM4-ΔZMO0293) can remarkably slow growth, glucose consumption rate, and ethanol production rate in the presence of a high glucose concentration (240 g/L), suggesting that the ZMO0293 gene may function under high glucose concentrations.
Compared to the ZM4, the ethanol titer of ZM4-ΔZMO0293 cells was reduced by 8.3% and 7.1% under glucose concentrations of 100 and 240 g/L, respectively ( Figure 2). When metabolizing the same amount of glucose or fructose, ATP production through the ED pathway in ZM4 is half of that through the Embden-Meyerhof-Parnas (EMP) pathway by other microorganisms, such as S. cerevisiae. Therefore, the ED pathway has less biomass accumulation, and most of the carbon sources (>95%) are converted to ethanol under low glucose concentrations [7,29]. We evaluated the role of the ZMO0293 gene in ethanol fermentation. The ZM4-∆ZMO0293 mutant showed a slow glucose utilization rate and nearly the same ethanol titer in the presence of 20 g/L glucose compared to ZM4 ( Figure 2D). The ZM4-∆ZMO0293 mutant exhibited a slow ethanol titer with almost the same glucose utilization rate in the presence of 100 g/L glucose ( Figure 2E). However, when 240 g/L glucose was added, the glucose utilization rate and ethanol titer of the ZM4-∆ZMO0293 mutant were markedly slower than those of ZM4 ( Figure 2F). After 96 h fermentation, the glucose consumption rate of ZM4-∆ZMO0293 was 2.3 g/L/h, 30.3% lower than that of ZM4 (3.3 g/L/h). Additionally, the ethanol rate of ZM4-∆ZMO0293 was 0.99 g/L/h, 34% lower than that of ZM4 (1.5 g/L/h).
Z. mobilis displays a long adaptation period under high concentrations of glucose or sucrose. Previous transcriptomic studies showed that ZM4 cells can regulate the transcription levels of genes associated with membrane channels and transporters, and metabolic pathways to adapt to high concentrations of glucose [28]. Our results showed that ZMO0293 deletion (ZM4-∆ZMO0293) can remarkably slow growth, glucose consumption rate, and ethanol production rate in the presence of a high glucose concentration (240 g/L), suggesting that the ZMO0293 gene may function under high glucose concentrations.
Compared to the ZM4, the ethanol titer of ZM4-∆ZMO0293 cells was reduced by 8.3% and 7.1% under glucose concentrations of 100 and 240 g/L, respectively ( Figure 2). These results suggested that ZMO0293 deletion might conditionally affect glucose metabolism and ethanol production, especially under high glucose concentrations, which is consistent with previous results [28]. In addition, it is difficult to delete the ZMO0366 (glf ) gene by the CRISPR/Cas9 system used in this study and other methods, such as RecET [30], indicating that this gene is probably an essential gene for ZM4 cells. In order to test the potential of Zmo0293, we constructed the complementary strain by plasmid expression in ZM4-∆ZMO0293 (Supplementary Figure S3). When the glucose content in the RMG medium reached 240 g/L, the complementary strain of ZM4-∆ZMO0293 mutant (ZM4-∆ZMO0293/pZM3pdc-ZMO0293) had a lag period of more than 30 h, which severely affected the normal growth. Therefore, the glucose concentration of 200 g/L was used to replace the 240 g/L. The results showed that the complementary strain of the ZM4-∆ZMO0293 mutant grew slowly and exhibited an obvious lag time ( Figure 3C). Additionally, the OD of the ZMO0293 and control ZMO0366 overexpressed strain (ZM4/pZM3pdc-ZMO0293 and ZM4/pZM3pdc-ZMO0366) was almost consistent with that of ZM4 and ZM4-∆ZMO0293 strains in the presence of 20 and 100 g/L glucose ( Figure 3A,B). Under 200 g/L glucose, the growth of the overexpressed strain of the ZMO0293 gene (ZM4/pZM3pdc-ZMO0293) was slower than that of ZM4 and the deletion strain (ZM4-∆ZMO0293), and the final OD was 9.2 % and 7.8% lower than that of the ZM4 and ZM4-∆ZMO0293 strains, respectively ( Figure 3C). It can be concluded that the overexpression effect of the ZMO0293 gene did not work as expected in low or high glucose concentrations. A possible reason is that plasmid overexpression of sugar transporter Zmo0293 needs correct folding, localization, and insertion into the phospholipid bilayer. The copy number of plasmid pZM3 is higher than that of gene in the genome, which possibly leads to excessive protein expression and further affects the protein maturation process in the cell. In addition, the burden of plasmid expression on growth is another possible reason.
ΔZMO0293 mutant grew slowly and exhibited an obvious lag time ( Figure 3C). Additionally, the OD of the ZMO0293 and control ZMO0366 overexpressed strain (ZM4/pZM3pdc-ZMO0293 and ZM4/pZM3pdc-ZMO0366) was almost consistent with that of ZM4 and ZM4-ΔZMO0293 strains in the presence of 20 and 100 g/L glucose ( Figure 3A,B). Under 200 g/L glucose, the growth of the overexpressed strain of the ZMO0293 gene (ZM4/pZM3pdc-ZMO0293) was slower than that of ZM4 and the deletion strain (ZM4-ΔZMO0293), and the final OD was 9.2 % and 7.8% lower than that of the ZM4 and ZM4-ΔZMO0293 strains, respectively ( Figure 3C). It can be concluded that the overexpression effect of the ZMO0293 gene did not work as expected in low or high glucose concentrations. A possible reason is that plasmid overexpression of sugar transporter Zmo0293 needs correct folding, localization, and insertion into the phospholipid bilayer. The copy number of plasmid pZM3 is higher than that of gene in the genome, which possibly leads to excessive protein expression and further affects the protein maturation process in the cell. In addition, the burden of plasmid expression on growth is another possible reason. The strain ZM4 represents WT, the strain ZM4-ΔZMO0293 for ZMO0293 deletion, the strain ZM4/pZM3pdc-ZMO0293 for the ZMO0293 overexpression in ZM4, the strain ZMO0366 ZM4/pZM3pdc-ZMO0366 for the ZMO0366 overexpression in ZM4, the strain ZM4-ΔZMO0293/pZM3pdc-ZMO0293 for the complementary strain of ZM4-ΔZMO0293 mutant.

ZMO0293 Deletion Influences Transcriptional Changes in Some Genes of the ED Pathway
The ED pathway is responsible for glucose metabolism. Four genes, glf (ZMO0366), zwf (ZMO0367), edd (ZMO0368), and glk (ZMO0369), constitute an operon that enables efficient glucose utilization [31]. glf is the main sugar transporter for glucose uptake in ZM4. To examine the influence of ZMO0293 deletion on the ED pathway, we determined the transcriptional level of the ED pathway by RT-qPCR. As shown in Table 1, after adding 240 g/L glucose, the relative expression level of ZMO0293 in ZM4 increased from 1.05 (T1) to 1.49 (T2) and 2.58 (T3), while the expression level of ZMO0366 decreased from 3.3 (T1) to −2 (T2) and −1.64 (T3). In the ZM4-ΔZM0293 strain, the expression level of Figure 3. Effect of ZMO0293 overexpression on growth. The growth curves of different strains were measured in RMG medium with 20 (A), 100 (B), and 200 (C) g/L glucose. The strain ZM4 represents WT, the strain ZM4-∆ZMO0293 for ZMO0293 deletion, the strain ZM4/pZM3pdc-ZMO0293 for the ZMO0293 overexpression in ZM4, the strain ZMO0366 ZM4/pZM3pdc-ZMO0366 for the ZMO0366 overexpression in ZM4, the strain ZM4-∆ZMO0293/pZM3pdc-ZMO0293 for the complementary strain of ZM4-∆ZMO0293 mutant. The glucose uptake mediated by sugar transporters is usually the first step for glucose metabolism. ED pathway enzymes account for approximately 50% of total cellular proteins, together with pyruvate decarboxylase (PDC) and alcohol dehydrogenase B (ADHB), which facilitate rapid ethanol production in Z. mobilis. The influence of ZMO0293 deletion on the ED pathway as measured by qPCR revealed that the inhibitory effect of ZMO0366 on transcription levels by 240 g/L glucose addition was less than that of ZM4 (Table 1). In addition, glucose metabolism is usually involved in multiple functional genes and can provide energy and materials for other reactions. The expression level of multiple genes was influenced by deleting the ZMO0293 gene, indicating that the sugar transporter Zmo0293 might be involved in central metabolism and the synthesis of the by-product lactic acid.

ZMO0293 Deletion Decreases the Activities of Key Enzymes Involved in Glucose Metabolism
GK and G6PDH are the limiting enzymes of glucose utilization through the ED pathway, and ADH is the key enzyme of ethanol production. To characterize the effect of ZMO0293 on the activities of glucose metabolism, three key enzymes (GK, G6PDH, and ADH) were selected, and their activities were measured in the presence of 240 g/L glucose. From Figure 4, the activities of these enzymes were similar in ZM4 and ZM4-∆ZMO0293 at 24-h fermentation. Nevertheless, these enzymatic activities of ZM4-∆ZMO0293 were gradually lower than those of ZM4. Maximal enzyme activities of GK, G6PDH, and ADH were reduced by 24.2% (60 h), 29.4% (60 h), and 31.4% (48 h), respectively.
It is noteworthy that almost all the enzymes of the ED pathway metabolize glucose near their maximal activities, and their activities are not regulated by feedback inhibition or other regulatory mechanisms [29]. The decreased enzymatic activities of GK, G6PDH, and ADH were also related to sugar transporter ZMO0293 deletion (Figure 4). One explanation is that high osmotic stress resulting from high glucose concentrations is harmful to cell metabolism, and ZMO0293 deletion is possibly detrimental to the absorption of glucose [32]. The molecular mechanism of the sugar transporter Zmo0293 and Glf in response to high glucose concentrations requires further elucidation in Z. mobilis. or other regulatory mechanisms [29]. The decreased enzymatic activities of GK, G6PDH, and ADH were also related to sugar transporter ZMO0293 deletion (Figure 4). One explanation is that high osmotic stress resulting from high glucose concentrations is harmful to cell metabolism, and ZMO0293 deletion is possibly detrimental to the absorption of glucose [32]. The molecular mechanism of the sugar transporter Zmo0293 and Glf in response to high glucose concentrations requires further elucidation in Z. mobilis. , and ADH (C) were determined at the set time points. Data are presented as the mean ± s.d. from three independent experiments. * represents a statistically significant difference of p < 0.05 by Student's t test; ** represents a statistically significant difference of p < 0.01; ns: no significant difference.

Integration Expression of ZMO0293 can Promote the Growth of E. coli Strains Deficient in Glucose Uptake
Sugar transporters are useful functional genes in accelerating sugar absorption, reducing intracellular PEP (phosphoenolpyruvate) consumption, and alleviating CCR [23,33]. E. coli uses the PTS to transport extracellular glucose into the cells [34]. To explore the application potential of the sugar transporter Zmo0293, we deleted the BL21(DE3) ptsG gene using CRISPR/Cas9 to obtain the BL21(DE3)-ΔptsG strain, which is deficient in glucose absorption ( Figure 5A). Chromosome integration of functional genes is more stable than plasmid expression [35]. Therefore, ZMO0293 and ZMO0366 genes under the weak E. coli gapAP1 promoter were integrated into the ykgH-betA neutral site of the genome of * represents a statistically significant difference of p < 0.05 by Student's t test; ** represents a statistically significant difference of p < 0.01; ns: no significant difference.

Integration Expression of ZMO0293 can Promote the Growth of E. coli Strains Deficient in Glucose Uptake
Sugar transporters are useful functional genes in accelerating sugar absorption, reducing intracellular PEP (phosphoenolpyruvate) consumption, and alleviating CCR [23,33]. E. coli uses the PTS to transport extracellular glucose into the cells [34]. To explore the application potential of the sugar transporter Zmo0293, we deleted the BL21(DE3) ptsG gene using CRISPR/Cas9 to obtain the BL21(DE3)-∆ptsG strain, which is deficient in glucose absorption ( Figure 5A). Chromosome integration of functional genes is more stable than plasmid expression [35]. Therefore, ZMO0293 and ZMO0366 genes under the weak E. coli gapAP1 promoter were integrated into the ykgH-betA neutral site of the genome of BL21(DE3)-∆ptsG, respectively ( Figure 5B,C). The medium type and cultivation volume were important factors affecting the function of gene expression. Here, we employed LB and M9 medium, together with flasks and 96-well plates, to evaluate the role of Zmo0293 and Glf in E. coli BL21(DE3)-∆ptsG. The results showed that the growth of the BL21(DE3)-∆ptsG strain was slower than that of the control BL21(DE3) strain in LB or M9 medium ( Figure 6). Furthermore, the integration expression of ZMO0293 and ZMO0366 can almost restore the growth of the BL21(DE3) ptsG deletion strain. We also found that the restoration of Zmo0293 is similar to that of Glf. In addition, the recovery performance of these two genes in the M9 medium was better than that in the LB medium. Subsequently, the effect of the integration expression of ZMO0293 in the BL21(DE3)-∆ptsG strain was further evaluated by a spot assay. The results also showed that the ZMO0293 gene can improve the cell growth of BL21(DE3)-∆ptsG in the M9 agar medium (Figure 7).
It has been demonstrated that glucose transport in recombinant E. coli is facilitated by the glucose transporter Glf from Z. mobilis [36]. The growth of BL21(DE3)-∆ptsG was promoted by the integrated expression of ZMO0293 driven by the weak promoter gapAP1, which is comparable to that of ZMO0366 (glf ). Medium type (LB and M9) can possibly affect the function of Zmo0293 and Glf in E. coli BL2(DE3)-∆ptsG, for which one explanation is that LB medium contains more rich nutrition than M9 (Figures 5 and 6). Enhanced growth ability will be beneficial for glucose utilization in E. coli BL2(DE3)-∆ptsG.
Mining sugar transporters can provide useful targets for genetic engineering and synthetic biology. For example, other studies proved that characterizing and engineering sugar uptake system can enhance the titer of milbemycins in Streptomyces bingchenggensis [37]. The expression of a xylose transporter xylE from E. coli could enhance the rate of xylose utilization in Z. mobilis at high xylose concentrations [38]. Nevertheless, the lack of a dedicated xylose transport system in recombinant Z. mobilis thus limits the capacity for dual xylose and glucose fermentation as well as the high xylose catabolic pathway. This bottleneck highlights the necessity to identify and engineer efficient xylose transport proteins in Z. mobilis. The absorbable carbohydrate spectrum and absorption kinetics of sugar transporter Zmo0293 need to be further studied. and M9 medium, together with flasks and 96-well plates, to evaluate the role of Zmo0293 and Glf in E. coli BL21(DE3)-ΔptsG. The results showed that the growth of the BL21(DE3)-ΔptsG strain was slower than that of the control BL21(DE3) strain in LB or M9 medium ( Figure 6). Furthermore, the integration expression of ZMO0293 and ZMO0366 can almost restore the growth of the BL21(DE3) ptsG deletion strain. We also found that the restoration of Zmo0293 is similar to that of Glf. In addition, the recovery performance of these two genes in the M9 medium was better than that in the LB medium. Subsequently, the effect of the integration expression of ZMO0293 in the BL21(DE3)-ΔptsG strain was further evaluated by a spot assay. The results also showed that the ZMO0293 gene can improve the cell growth of BL21(DE3)-ΔptsG in the M9 agar medium (Figure 7).     It has been demonstrated that glucose transport in recombinant E. coli is facilitated by the glucose transporter Glf from Z. mobilis [36]. The growth of BL21(DE3)-ΔptsG was promoted by the integrated expression of ZMO0293 driven by the weak promoter gapAP1, which is comparable to that of ZMO0366 (glf). Medium type (LB and M9) can possibly affect the function of Zmo0293 and Glf in E. coli BL2(DE3)-ΔptsG, for which one explanation is that LB medium contains more rich nutrition than M9 (Figures 5 and 6). Enhanced growth ability will be beneficial for glucose utilization in E. coli BL2(DE3)-ΔptsG.
Mining sugar transporters can provide useful targets for genetic engineering and synthetic biology. For example, other studies proved that characterizing and engineering sugar uptake system can enhance the titer of milbemycins in Streptomyces bingchenggensis The initial OD 600 of all strains was adjusted to 1.0, and these E. coli cells were serially diluted tenfold. All the strains were spotted on M9 agar plates and cultivated at 37 • C.
E. coli BL21(DE3) and its derivatives were cultured in Luria-Bertani medium (LB, 1% tryptone, 0.5% yeast extract, 1% NaCl, and pH 7.0) or M9 medium (2% glucose, 0.6% Na 2 HPO 4 , 0.3% KH 2 PO 4 , 0.05% NaCl, 0.1% NH 4  The ZMO0293 gene was deleted by a native CRISPR-Cas genome editing system [6]. Briefly, the editing plasmids were initially constructed with a spacer containing a 5 -CCC-3 PAM. The editing plasmid backbone was amplified by PCR. Oligonucleotides were annealed by first heating the reaction mixture to 95 • C for 5 min and subsequently cooling gradually to room temperature. The annealed spacer, the upstream and downstream homologous arms (HAs) of ZMO0293 with a length of 500-1000 bp, and the plasmid backbone were linked by seamless cloning (Vazyme, Nanjing, China), and then the reaction product was introduced into E. coli DH5α competent cells to generate the pMini-∆ZMO0293 editing plasmid. Correct plasmids were verified by colony PCR and Sanger sequencing. pZM3pdc-ZMO0366 and pZM3pdc-ZMO0366 were also constructed by seamless cloning using pZM3pdc as plasmid backbone [30]. All strains, plasmids, and primers used in this study are listed in Table 2 and Supplementary Tables S1 and S2.

Electroporation and Mutant Selection
The pMini-∆ZMO0293 plasmid was transformed into ZM4 competent cells (1 µg DNA with 50 µL competent cells) via electroporation using a Bio-Rad Gene Pulser (Bio-Rad, Hercules, CA, USA). The electroporation parameters were as follows: 0.2 cm electroporation cuvettes, 1.6 kV, 25 µF, and 200 Ω. The electroporated cells were immediately transferred to 1 mL RMG and recovered at 30 • C for 8-12 h. The cells were then spread on RMG agar plates containing appropriate antibiotics and incubated at 30 • C for 2-3 days to isolate single colonies. Colonies with correct plasmids were cultivated in an RMG medium containing appropriate antibiotics and verified by colony PCR. E. coli BL21(DE3) was used as the host for genomic manipulations. Gene deletion and integration expression were performed by the CRISPR/Cas9 system with two plasmids, one plasmid, pTargetF (Addgene number 62226), was responsible for target site recognition, and the other plasmid, pEcCas (Addgene number 73227), was used for target site cutting [39,40]. When needed, both spectinomycin and kanamycin were used at final concentrations of 50 µg/mL. L-arabinose and rhamnose were added at concentrations of 10 mM and 10 mM, respectively.

Plasmid Construction
The pTargetF plasmid used for deletion and integration of the target site was constructed by a pair of primers with the gRNA spacer sequence specific for each target and the flanked sequences homologous to the pTargetF backbone ( Table 2, Supplementary Table S1). The resultant PCR products were transformed into DH5α competent cells after Dpn I digestion and then ligated via homologous recombination to form pTargetF-ptsG and pTargetF-ykgH-betA.

Donor Construction
The upstream and downstream HAs with a length of 500-1000 bp were fused by overlap extension PCR to form a donor DNA for seamless deletion of the ptsG gene. The expression donor for ZMO0293 and ZMO0366 was constructed for chromosome integra-tion [41]. These two genes, ZMO0293 and ZMO0366, were driven by the promoters P gapAP1 and synthetic terminator T BBa_B1002 , respectively. The HA length at both ends was also 500-1000 bp.

Electroporation and Screening of Mutants
Briefly, 40 µL electrocompetent cells harboring the pEcCas plasmid, 400 ng donor DNA, and 100 ng gRNA plasmid were mixed for electroporation in a Gene Pulser ® II electroporator (Bio-Rad Laboratories, Hercules, CA, USA) using the following parameters: 1.0 mm cuvette gap, 1.8 kV, 200 Ω, and 25 µF [40]. After electroporation, the transformed cells were immediately resuspended in 1 mL of LB and allowed to recover at 37 • C for 1 h prior to plating on LB agar supplemented with kanamycin and spectinomycin. After overnight culture at 37 • C, the single colonies were randomly picked for colony PCR. The selected transformants were further confirmed by Sanger sequencing. The plasmids pEcCas9 and pTargetF could be cured when the strain was not subjected to further engineering.

Growth Curve Determination
Three single colonies of Z. mobilis were picked and cultivated in 5 mL RMG medium overnight at 30 • C without agitation. The cultures were transferred to 30 mL RMG medium at 1% inoculum. Three single colonies of E. coli were picked and cultivated in 5 mL of LB or M9 medium overnight at 37 • C and 220 rpm. The cultures were transferred to 30 mL LB or M9 medium in flasks or 250 µL LB or M9 medium in a 96-well plate for cultivation in a BioScreen Automated Microbial Growth Analyzer (Oy Growth Curves Ab Ltd., Helsinki, Finland) with 1% inoculum. Cell growth was monitored by measuring the absorbance at 600 nm (OD 600 ).

Real-time Quantitative PCR
ZM4 and ZM4-∆ZMO0293 strains were initially grown in RMG medium in the presence of 20 g/L glucose. When the OD 600 values reached 0.4, the samples were removed (T1) and immediately frozen in liquid nitrogen. Simultaneously, glucose was added to a final concentration of 240 g/L. After 20 min and 40 min of treatment, the samples served as T2 and T3, respectively. The samples without glucose addition were chosen as the control. The frozen samples were extracted to obtain RNA according to the manufacturer's protocol (Tiangen, Beijing, China).
Real-time quantitative PCR (qPCR) analysis was performed to evaluate ED pathway gene expression levels. The RNA samples were reverse-transcribed by PrimeScript ® RT Enzyme Mix I (Takara Biotechnology-Dalian Co., Ltd., Dalian, China). The sequences of the primer pairs are listed in Supplementary Table S3. The gene rrsA (ZMOr009), encoding the 16S ribosomal RNA, was used as a control gene. qPCR was performed using an iQ5 Real-time PCR System (Bio-Rad Laboratories, Inc., Hercules, CA, USA) with 2×TaKaRa SYBR Green Real-Time PCR Master Mix (Takara Biotechnology-Dalian Co., Ltd., Dalian, China). The data were calculated using the 2 −∆∆CT method [42].

Fermentation Analysis
RMG medium (50 mL) was added to a 250 mL Erlenmeyer flask. The final glucose concentration was adjusted to reach 20 g/L, 100 g/L, and 240 g/L. Fermentation was carried out without agitation at 30 • C with inoculation of 1% preculture. During the culture process, samples were taken every 4 h to determine the OD value. The glucose concentration in the fermentation broth was determined by the 3,5-dinitrosalicylic acid method [43]. The ethanol content was analyzed using an ethanol determination kit according to the manufacturer's instructions (Suzhou Grace Biotechnology Co., Ltd., Suzhou, China).

Determination of Enzyme Activities
ZM4 and ZM4-∆ZMO0293 were cultivated at 30 • C in an RMG medium containing 240 g/L glucose. The culture was centrifuged for 15 min at 12,000 rpm. The pellets were washed with a sterile physiological saline solution. The samples were pretreated according to the method by Algar and Scopes [29]. The activities of the enzymes glucokinase (GK), glucose-6-phosphate dehydrogenase (G6PDH), and alcohol dehydrogenase (ADH) were measured according to the manufacturer's instructions (Suzhou Grace Biotechnology Co., Ltd., Suzhou, China).
3.8. Spot Assay of Various E. coli BL21(DE3) Strains E. coli BL21(DE3) and its derivatives were cultured in an M9 medium at 37 • C for 24 h. The OD 600 of various cultures was adjusted to the same level, and 2.0 µL of the tenfold diluted suspensions of each strain was spotted on solid M9 medium. The plates were incubated at 37 • C for 2 days.

Conclusions
The deletion of the sugar transporter-encoding gene ZMO0293 can result in growth defect, low glucose consumption rate, and ethanol titer in Z. mobilis. It can also cause the low activities of key enzymes (GK, G6PDH, and ADH) for glucose metabolism and differential expression of some genes of the ED pathway. In addition, it can restore the growth of E. coli BL21(DE3)-∆ptsG deficient in glucose absorption. The results of this study are not only helpful for metabolic engineering regulation of glucose utilization but are also beneficial to alleviate or eliminate the CCR phenotype of recombinant strains in the future.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.

Conflicts of Interest:
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.