The Magnesium Concentration in Yeast Extracts Is a Major Determinant Affecting Ethanol Fermentation Performance of Zymomonas mobilis

Zymomonas mobilis is a model ethanologenic bacterium for diverse biochemical production. Rich medium (RM) is a complex medium that is routinely used to cultivate Z. mobilis, which contains carbon sources such as glucose, nitrogen sources such as yeast extract (YE), and KH2PO4. Glucose consumption and cell growth of Z. mobilis is usually coupled during ethanol fermentation. However, sometimes glucose was not consumed during the exponential growth phase, and it took extended time for cells to consume glucose and produce ethanol, which eventually reduced the ethanol productivity. In this study, the effects of different nitrogen sources, as well as the supplementation of an additional nitrogen source into RM and minimal medium (MM), on cell growth and glucose consumption of Z. mobilis were investigated to understand the uncoupled cell growth and glucose consumption. Our results indicated that nitrogen sources such as YE from different companies affected cell growth, glucose utilization, and ethanol production. We also quantified the concentrations of major ion elements in different nitrogen sources using the quantitative analytic approach of Inductively Coupled Plasma Optical Emission Spectroscopy (ICP-OES), and demonstrated that magnesium ion in the media affected cell growth, glucose consumption, and ethanol production. The effect of magnesium on gene expression was further investigated using RNA-Seq transcriptomics. Our results indicated that the lack of Mg2+ triggered stress responses, and the expression of genes involved in energy metabolism was reduced. Our work thus demonstrated that Mg2+concentration in nitrogen sources is essential for vigorous cell growth and ethanol fermentation, and the difference of Mg2+concentration in different YE is one of the major factors affecting the coupled cell growth, glucose consumption and ethanol fermentation in Z. mobilis. We also revealed that genes responsive for Mg2+ deficiency in the medium were majorly related to stress responses and energy conservation. The importance of magnesium on cell growth and ethanol fermentation suggests that metal ions should become one of the parameters for monitoring the quality of commercial nitrogen sources and optimizing microbial culture medium.

Zymomonas mobilis is a model ethanologenic bacterium for diverse biochemical production. Rich medium (RM) is a complex medium that is routinely used to cultivate Z. mobilis, which contains carbon sources such as glucose, nitrogen sources such as yeast extract (YE), and KH 2 PO 4 . Glucose consumption and cell growth of Z. mobilis is usually coupled during ethanol fermentation. However, sometimes glucose was not consumed during the exponential growth phase, and it took extended time for cells to consume glucose and produce ethanol, which eventually reduced the ethanol productivity. In this study, the effects of different nitrogen sources, as well as the supplementation of an additional nitrogen source into RM and minimal medium (MM), on cell growth and glucose consumption of Z. mobilis were investigated to understand the uncoupled cell growth and glucose consumption. Our results indicated that nitrogen sources such as YE from different companies affected cell growth, glucose utilization, and ethanol production. We also quantified the concentrations of major ion elements in different nitrogen sources using the quantitative analytic approach of Inductively Coupled Plasma Optical Emission Spectroscopy (ICP-OES), and demonstrated that magnesium ion in the media affected cell growth, glucose consumption, and ethanol production. The effect of magnesium on gene expression was further investigated using RNA-Seq transcriptomics. Our results indicated that the lack of Mg 2+ triggered stress responses, and the expression of genes involved in energy metabolism was reduced. Our work thus demonstrated that Mg 2+ concentration in nitrogen sources is essential for vigorous cell growth and ethanol fermentation, and the difference of Mg 2+ concentration in different YE is one of the major factors affecting the coupled cell growth, glucose consumption and ethanol fermentation in Z. mobilis. We also revealed that genes responsive for Mg 2+ deficiency in the medium were majorly related to stress INTRODUCTION trace nutrients, vitamin B complex and other important growth factors for various microorganisms (Zarei et al., 2016).
Despite the fact that manufacturers produce different YE to have the same composition of total nitrogen content and free amino acid nitrogen, YE produced by different companies has different trace components such as growth factors and metal ions due to the differences in their production processes. This then affects the microbial cell growth because vitamins and metal ions (e.g., Mg 2+ , Cu 2+ , Zn 2+ , and Fe 2+ ) are cofactors of enzymes involved in various metabolic activities. For example, Kosaka et al. (2020) investigated the effect of different metal ions on the thermotolerance of Z. mobilis TISTR548 and related mutants, and demonstrated that the addition of Mg 2+ and K + reduced intracellular reactive oxygen species (ROS) accumulation at critical high temperature (CHT) with an increase of CHT by 1 • C, which is probably due to the stabilization of both outer and inner membranes as well as the maintenance of homeostasis for cellular metabolism by the addition of Mg 2+ and K + . In addition, the effect of supplementation of zinc on the ethanol fermentation performance of the self-flocculating yeast in the continuous ethanol fermentation was studied (Zhao et al., 2009;Xue et al., 2010;Ismail et al., 2014), and the roles of zinc and zinc containing proteins in yeast metabolism and cellular stress responses were summarized (Zhao and Bai, 2012).
With all nutrients needed for cell growth in the culture medium including a nitrogen source from YE, cell growth of Z. mobilis ZM4 is generally expected to be coupled with its glucose consumption and ethanol production. However, results and literature reports in a few studies from different research groups indicated that cell growth of Z. mobilis ZM4 and its glucose consumption were uncoupled, glucose was not completely consumed for ethanol production when cells reached stationary phase (Jones and Doelle, 1991;He et al., 2012;Yang et al., 2014a;Xia et al., 2018;Duan et al., 2019). This uncoupling phenomenon between growth and fermentation performance has also been reported in other microorganisms, including yeast, resulting in a prolonged fermentation time and decreased ethanol productivity (Cot et al., 2007;Pagliardini et al., 2013). For example, ethanol production of yeast cells was reported to be related to the length of the uncoupling phase during the batch fermentation process (Pagliardini et al., 2013).
A previous study in E. coli demonstrated that the inadequate amount of magnesium in the rich medium of buffered tryptone broth supplemented with glucose (TB7-glucose) is the major determinant leading to the uncoupled cell growth and glucose consumption (Christensen et al., 2017). They further demonstrated that the supplementation of sufficient magnesium can increase cell growth yield in other peptide-based media when carbon is in excess. In addition, their study also showed that magnesium can increase cell growth of multiple E. coli strains and other bacterial species such as Bacillus subtilis and Vibrio fischeri (Christensen et al., 2017). Recently, Lozano Terol et al. (2019) studied protein overexpression for E. coli BL21 and five derived mutants in M9 minimal medium and TB7 complex medium, and discovered a similar phenomenon that carbon source of glucose or glycerol was not consumed during the exponential growth phase in TB7 complex medium but not M9.
The uncoupling of growth and fermentation in the late growth phase could be beneficial for ethanol fermentation since carbon source will be channeled into ethanol production instead of biomass buildup especially for yeast ethanol production. However, the uncoupling of growth and glucose consumption at early growth phase is unfavorable since there will be no sufficient cell biomass for efficient sugar fermentation to ethanol, which is mostly due to stressful physical and chemical growth conditions, such as extreme temperature, extreme pH, and toxic compounds in the medium. Limited nutrient sources of carbon, nitrogen and metal ions used for fermentation were also reported to affect cell growth and fermentation performance (Jones and Doelle, 1991;Cot et al., 2007;Pagliardini et al., 2013;. Although the effect of diverse nitrogen sources on cell growth and fermentation performance has been reported in various microorganisms, such as the ethanologenic yeast S. cerevisiae and lactate-producer Sporolactobacillus inulinus (Martinez-Moreno et al., 2012;Kemsawasd et al., 2015;Klotz et al., 2017), metal ions within nitrogen sources and their impact on cell growth and fermentation have not been investigated extensively.
To understand the uncoupling phenomenon of cell growth and glucose consumption for both efficient carbon utilization and maximum ethanol productivity, we investigated the effects of different nitrogen sources on cell growth, glucose consumption and ethanol production of Z. mobilis in this study to help achieve the goal of optimal titer, rate, and yield for economic bioethanol production using Z. mobilis. Our results demonstrated that glucose consumption and ethanol production can be coupled with cell growth by changing the nitrogen sources in the rich medium, and the concentration of metal ions such as Mg 2+ in the nitrogen source is a major factor affecting cell growth and ethanol production.

Effect of Supplementation of Nitrogen Sources Into RM
To understand the uncoupling of cell growth, glucose consumption and ethanol production in Z. mobilis, we evaluated factors in the medium affecting cell growth. The recipe of RM for Z. mobilis is relatively simple with carbon sources such as glucose, phosphate sources of KH 2 PO 4 , and YE as the source of nitrogen, minerals, and vitamins. Thus, 10 mM NH 4 Cl was added into RM using YE from the company OXOID (RM OXOID ) to increase inorganic nitrogen content. However, extra NH 4 Cl supplemented into RM OXOID did not reduce the time for glucose utilization and ethanol production of ZM4 (data not shown).
The effect of supplementing different organic nitrogen sources such as peptone and tryptone were then tested. Our results showed that the supplementation of peptone did not reduce the time of glucose utilization. It still took more than 24 h for cells to consume all glucose in the media with 20 g/L peptone supplemented into RM OXOID . The final biomass in terms of OD 600 value did increase slightly in correspondence to the increase of peptone added (Figure 1).
Subsequently, different concentrations of tryptone were added in RM OXOID in a separate experiment. The results showed that the addition of extra tryptone could significantly increase both cell biomass and glucose consumption of ZM4 (Figure 2). The addition of more than 5 g/L tryptone in RM OXOID not only enhanced both cell growth and glucose utilization of ZM4, but also coupled the growth and ethanol fermentation of ZM4. The final highest OD 600 value of ZM4 in RM OXOID with 5 g/L tryptone increased from 1.8 to 4.7, and all glucose was completely consumed within 12 h. Z. mobilis in medium without tryptone only utilized half the glucose at the same time point of 12-h post-inoculation (Figure 2).
The positive effect of adding tryptone into RM OXOID was further evaluated by comparing cell growth along with glucose utilization and ethanol production of ZM4 in both RM OXOID and RM OXOID + 5T at different temperatures of 30, 36, and 40 • C. The results demonstrated that the supplementation of tryptone increased the growth and fermentation performance of ZM4 at each of these different temperatures, including the specific growth rate (µ), ethanol yield (Y p/s ), and productivity (Q P ) ( Table 1).
It also increased cell growth, glucose consumption rate, and ethanol productivity of ZM4 at higher temperatures of 36 and 40 • C than those at the normal temperature of 30 • C in RM OXOID supplemented with 5 g/L tryptone, although ethanol yields were similar among these conditions (Table 1 and Figure 3). Compared with a normal temperature of 30 • C, the time that ZM4 utilized all glucose at 36 and 40 • C in RM OXOID with extra tryptone was reduced by about 3 h, but 1 day was not sufficient for ZM4 to completely consume glucose at any of the aforementioned temperatures in RM OXOID (Figure 3). ZM4 took less time to completely utilize glucose with maximum ethanol production achieved in RM OXOID + 5T at 36 and 40 • C (9 h) than at 30 • C (12 h) with a corresponding higher growth rate and ethanol productivity, which was increased from 1.75 ± 0.01 at 30 • C to 2.19 ± 0.22 and 2.23 ± 0.11 at 36 and 40 • C, respectively (Table 1 and Figure 3).

Effect of Exchanging YE From OXOID With Different Brand Ones in RM
As briefly mentioned above, YE distributed by different companies are produced with different processes which may lead to different amounts of total nitrogen and trace elements such as growth factors and metal ions. Besides adding extra nitrogen sources into the RM OXOID as discussed above (Figures 1-3), we also tested the effect of YE from different companies, including those from Becton Dickinson (YE BD ) and Sangon Biotech Co., Ltd. (YE SG , Shanghai, China). In addition, another industrial FIGURE 1 | Cell growth and glucose (Glu) consumption of ZM4 in rich medium (RM) using yeast extract from OXOID company (RM OXOID ) with extra peptone (P) supplemented at concentrations of 0.1, 1, 10, and 20 g/L.
TABLE 1 | Glucose consumption (Y s ), ethanol yield (Y p/s ), theoretical ethanol yield (%Y p/s ), ethanol productivity (Q P ), and specific growth rate (µµ) of ZM4 in RM OXOID and RM OXOID supplemented with 5 g/L tryptone at different temperatures within 24 h.  nitrogen source of corn steep liquid (CSL) from Macklin was evaluated at the same time as the two yeast-based nitrogen sources discussed above. All media used in this research are listed in Tables 5, 6. YE OXOID in RM OXOID was replaced with different nitrogen sources including peptone (P), tryptone (T), CSL, YE BD , and YE SG . Cell growth, glucose utilization, and ethanol production of ZM4 cultured in these media were then examined. Our results showed ZM4 performed differently in these media (Figure 4). The biomass of ZM4 in RM(T) was about two times higher than that in RM OXOID , while the biomass of ZM4 in RM(P) was the lowest (Figure 4B). Tryptone appeared to be better than YE OXOID and peptone for cell growth, which was consistent with the aforementioned experiments of supplementation of nitrogen source into RM OXOID (Figures 1, 2).
Except for the fact that the specific growth rates of ZM4 in RM with 1% of peptone or CSL as the sole nitrogen source were lower, which can be seen from the slopes of the lines in the exponential phase, the specific growth rates of ZM4 in other RM were almost the same ( Figure 4A). Compared with YE OXOID as the sole nitrogen source, YE BD , YE SG , and 5% CSL all increased the final biomass of ZM4 (Figure 4B), reduced the time of glucose consumption ( Figure 4C), and coupled cell growth and ethanol production. The biomass of ZM4 in RM BD was more than two times of that in RM OXOID and the time for glucose consumption in RM BD was about two-thirds shorter than that in RM OXOID (Figures 4B,C). YE SG , though cheaper than YE BD , was almost just as good. In the future, YE SG could be used in largescale fermentation for ethanol production, and the appropriate nitrogen source can be chosen as needed based on the results above (Figures 1-4).

Effect of Exchanging (NH 4 ) 2 SO 4 in MM With Different Organic Nitrogen Sources
Although the RM recipe is relatively simple containing only three ingredients as discussed above, the composition of organic nitrogen sources such as YE is still complicated containing nitrogen, vitamins, metal ions, etc. Because the only nitrogen source in minimal medium (MM) is the inorganic nitrogen source (NH 4 ) 2 SO 4 , it was selected to further examine the effect of different organic nitrogen sources on ZM4. The representative organic nitrogen sources used to replace (NH 4 ) 2 SO 4 in MM included YE BD , YE OXOID , and 5%CSL.
Our results demonstrated that the replacement of inorganic nitrogen source (NH 4 ) 2 SO 4 with organic nitrogen source enhanced cell growth, glucose utilization, and ethanol productivity ( Figure 5). Interestingly, ZM4 had similar growth and glucose consumption rates in two MM media with either YE BD or YE OXOID as the sole nitrogen source (Figure 5). These results differed from those when YE BD or YE OXOID was used as the sole nitrogen source in RM where ZM4 grew better in RM with YE BD than in RM with YE OXOID as the nitrogen source (Figure 4). Therefore, some ingredients must exist in MM which RM lacks when YE OXOID is used as the nitrogen source. This difference is contingent on compositions of NaCl, Na 2 MoO 4 , and MgSO 4 which differ in the ingredients of the two media (Tables 5, 6).
Since magnesium ion is a cofactor of diverse enzymes involved in various metabolic activities, we investigated the effect of removing the magnesium ion from both MM BD and MM OXOID to test the growth and fermentation performance of ZM4. The lack of Mg 2+ in MM BD had no effect on the growth of ZM4, but the biomass decreased in MM OXOID lacking Mg 2+ (Figure 6). This result indicates that the concentration of Mg 2+ in YE OXOID is different from that in YE BD , and Mg 2+ plays important roles in FIGURE 5 | Cell growth, glucose (Glu) consumption and ethanol (Eth) production of ZM4 in MM containing 20 g/L glucose with different nitrogen sources.
FIGURE 6 | Cell growth, glucose (Glu) consumption and ethanol (Eth) production of ZM4 in MM containing 20 g/L glucose with original inorganic nitrogen sources replaced by YE BD and YE OXOID , respectively, as well as in above two media with Mg 2+ removed. Magnesium ion was tested twice while other ions were tested only once. Multiple comparisons of magnesium concentrations in different nitrogen sources were conducted with 1% YE OXOID as a control, **represents a significant difference (P < 0.01). cell growth of Z. mobilis, and Mg 2+ was likely one of the factors affecting the coupling of cell growth, glucose consumption and ethanol production.

Determination of Concentrations of Ions in Different Nitrogen Sources and the Impact of the Mg 2+ on Fermentation Performance
In order to verify this hypothesis, the concentrations of several major ion elements in different organic nitrogen sources were then measured by ICP-OES. The result showed that the concentrations of these ion elements (Fe, K, Mg, P, S) were different among different nitrogen sources with peptone and tryptone containing fewer of these ions ( Table 2). This may be one reason why cell growth and fermentation performance of Z. mobilis in media using tryptone and peptone as the sole nitrogen source was not as efficient as in other media using YE (Figure 4). At the same concentration of nitrogen sources, the Mg 2+ in the YE from OXOID was 8.3 mg/L less than that from Becton Dickinson. Mg 2+ is the component with the major difference between these two nitrogen sources of YE from OXOID and BD companies, and therefore could be the reason that cell growth and ethanol fermentation of Z. mobilis in RM OXOID was not as efficient as in RM BD ( Table 2). Since 0.5 g/L MgSO 4 ·7H 2 O containing 49 mg/L Mg 2+ was provided in MM which was sufficient for cell growth (Tables 2, 4), the growth difference of Z. mobilis in MM OXOID and MM BD was therefore not as obvious as that in RM OXOID and RM BD (Figures 4, 5). The concentration of Mg 2+ in peptone was the lowest, which could also be one reason that ZM4 grew poorly in RM(P) compared to other nitrogen sources (Figure 4). These results further confirm the importance of sufficient Mg 2+ concentration in media for optimal cell growth, glucose consumption and ethanol fermentation. The result that the concentration of Mg 2+ in tryptone was higher than that in YE OXOID was consistent with the results of the nitrogen source supplement experiment (Kosaka et al., 2020), and provided a conjecture that perhaps the concentration of magnesium ion in nitrogen sources plays an important role in promoting temperature tolerance and cell growth (Figures 2, 3).
To verify the impact of Mg 2+ on cell growth, different concentrations of Mg 2+ were then added into the RM OXOID and MM OXOID without MgSO 4 (MM OXOID MgSO 4 ), and the growth of ZM4 in these media was measured by the Bioscreen C (Figure 7). Our results indicated that in both RM OXOID and MM OXOID MgSO 4 , even a small amount of Mg 2+ at 4 mg/L could boost cell growth significantly.
Cell growth, glucose utilization, and ethanol production were further investigated using the shake flask experiment with OD 600 values, glucose and ethanol concentrations measured (Figure 8). Consistent with the result of Bioscreen C (Figure 7), the supplementation of at least 8 mg/L of Mg 2+ in RM OXOID or FIGURE 8 | Cell growth, glucose (Glu) consumption, and ethanol (Eth) production of ZM4 in RM OXOID , RM OXOID + 8Mg, RM BD , and RM SG . At least two independent experiments were carried out with similar results. Values are the mean of one representative experiment with two or more technical replicates. Error bars represent standard deviations.
FIGURE 9 | Volcano plots of significantly differentially expressed genes of Z. mobilis cultured in RM OXOID and RM BD (RM OXOID /RM BD , (A), RM OXOID and RM OXOID + 8Mg (RM OXOID /RM OXOID + 8Mg, (B), as well as RM BD and RM OXOID + 8Mg (RM BD /RM OXOID + 8Mg, (C). X-axis is the log 2 -based ratios between two conditions examined, and Y-axis is the -log 10 (P-value) of the difference. The dots above the horizontal red dash line indicate genes significantly differentially expressed, and the vertical red dash line indicate genes significantly differentially expressed with ratio greater than 2 (log 2 -based ratio greater than 1). Gene name with red and blue color font indicates up-regulated and down-regulated ones, respectively. Gene names with bold font indicate common ones between different comparisons of RM OXOID /RM BD , RM OXOID /RM OXOID + 8Mg, and RM BD /RM OXOID + 8Mg. MM OXOID MgSO 4 allowed ZM4 to grow as well as in RM BD or MM BD MgSO 4 . At the same time, cell growth, glucose consumption, and ethanol production of ZM4 in RM OXOID with 8 mg/L Mg 2+ added (RM OXOID + 8Mg) were coupled as in RM BD (Figure 8). This suggests that Mg 2+ is crucial for cell Ratio is the log 2 -based expression difference between ZM4 cultured in RM OXOID and RM BD (RM OXOID /RM BD ). growth, glucose consumption and fermentation performance of Z. mobilis, and a minimum concentration of at least 8 mg/L is needed for optimal cell growth and ethanol fermentation.

Effects of Mg 2+ on Gene Expression of Z. mobilis
After confirmation of the impact of Mg 2+ on cell growth and ethanol fermentation of Z. mobilis (Figure 8), next-generation sequencing (NGS)-based RNA-Seq transcriptomics was further applied to identify genes responsive for the uncoupling of cell growth, glucose consumption and ethanol fermentation due to the difference of Mg 2+ concentration in the medium. Despite that cell growth, glucose consumption and ethanol production exhibited apparent differences (Figure 8), RNA-Seq results showed that only a few genes were differentially expressed. There were seven and four genes down-regulated when Z. mobilis was cultured in RM OXOID and RM BD (RM OXOID /RM BD , Figure 9A and Table 3) as well as cultured in RM OXOID and RM OXOID + 8Mg (RM OXOID /RM OXOID + 8Mg, Figure 9B and Table 4), respectively. Although genes up-regulated in RM OXOID were more than those down-regulated (Figure 9 and Tables 3, 4), there were only 11 and 10 up-regulated genes identified when Z. mobilis was cultured in RM OXOID and RM BD (RM OXOID /RM BD , Figure 9A and Table 3) as well as cultured in
Genes related to stress responses were up-regulated while genes related to protein secretion were down-regulated in RM OXOID medium compared with in RM OXOID + 8Mg or RM BD medium (Figure 9 and Tables 3, 4). For example, genes encoding catalase (ZMO0918), general stress protein CsbD (ZMO0740), NADH dehydrogenase (ZMO1113), and succinatesemialdehyde dehydrogenase SSADH (ZMO1754) were upregulated when RM OXOID was used compared with those in RM OXOID + 8Mg or RM BD (Figure 9 and Tables 3, 4). These results indicate that Mg 2+ , a key element of cofactor, is essential for vigorous cell growth, and the lack of Mg 2+ will trigger energy-consuming stress responses while slowing down energy-consuming metabolism with genes encoding levansucrase (ZMO0374) and secretion-related protein (ZMO0934) being down-regulated (Figure 9 and Tables 3, 4).
Considering that the supplementation of 8 mg/L Mg 2+ into RM OXOID (RM OXOID + 8Mg) could restore the coupling of cell growth, glucose consumption and ethanol fermentation of Z. mobilis (Figure 8) and that there was only one gene upregulated (ZMO2034) and one down-regulated (ZMO1522) when Z. mobilis grew in RM BD compared with RM OXOID + 8Mg (Figure 9C), the transcriptomics study further confirmed our hypothesis that the difference of Mg 2+ concentrations in different nitrogen sources is one of the determinants affecting the coupling of cell growth, glucose consumption and ethanol fermentation in Z. mobilis.

CONCLUSION
The effects of nitrogen sources on cell growth, glucose consumption, and ethanol fermentation performance of Z. mobilis ZM4 were investigated to understand the uncoupling of cell growth, glucose consumption and ethanol fermentation of ZM4 in this study. Through the supplementation and replacement of inorganic or organic nitrogen sources in both RM and MM, we found that YE BD , YE SG , or 5% CSL were better than YE OXOID . We also quantified the ion elements in different nitrogen sources using ICP-OES, and demonstrated that the difference of magnesium ion in YE is one of the major factors affecting cell growth and ethanol fermentation. We further verified the impact of Mg 2+ on cell growth of ZM4 by supplementing various concentrations of Mg 2+ into the medium, and used the RNA-Seq transcriptomics approach to identify genes responsive for the uncoupling of cell growth, glucose consumption and ethanol fermentation when the medium lacked Mg 2+ . These findings can be used as a reference for the selection and/or modification of nitrogen sources using Z. mobilis ZM4. The concentrations of ion elements in nitrogen sources affecting cell growth and fermentation performance can also be used as a parameter for optimizing and monitoring the components of nitrogen sources for efficient cell growth and fermentation using other microorganisms.
The recipes of different rich medium (RM) and minimal medium (MM) with different nitrogen sources used in this work were listed in Tables 5, 6, respectively.

Growth Test by Bioscreen C
The Bioscreen C automatic growth curve analyzer has the functions of culturing cells and measuring cell turbidity. It is similar to the plate reader, but can continuously measure the turbidity of cells, while maintaining a constant temperature and rotation speed to maintain normal cell growth.
The seed culture of Z. mobilis was centrifuged to remove RM OXOID . Cells were resuspended with test medium. Bioscreen C assays were carried out as described previously (Franden et al., 2009;Yang et al., 2010Yang et al., , 2014aYang et al., ,b, 2016bYang S. et al., 2018) except that cells were inoculated into Bioscreen C wells containing a total volume of 200 µL test medium at an initial OD 600 value of 0.05 and incubated without shaking at 30 • C. Triplicate were used for each condition, and turbidity measurements (OD 600 ) were taken every 15 min till cells grew into stationary phase. At least two independent experiments were carried out with similar results. Values are the mean of one representative experiment with two or more technical replicates. Error bars represent standard deviations.

Flask Fermentation and Analytical Analysis
The seed culture of Z. mobilis was used to inoculate the shake flask containing 80% of test medium with a sealing gas permeable membrane sealed at an initial OD 600 of 0.1, and cultured at 30 • C, 100 rpm. At least two replicates were used for each condition.
The OD 600 values of the bacterial culture was measured by UV-visible spectrophotometer UV-1800 (AoYi Instrument Co., Ltd., Shanghai, China) every 3 h. At the same time, 1-mL culture was centrifuged at 12,000 rpm for 1 min to obtain the supernatant for measuring the glucose and ethanol concentrations in the culture. Biosensor analyzer M-100 (Sieman Technology Co., Ltd., Shenzhen, China) was used for quick assessment of the concentrations of glucose and ethanol. The supernatant was also filtered through a 0.45 µm filter before applying on a Shimadzu LC-2030 high pressure liquid chromatography (HPLC) with refractive index detector (RID). Bio-Rad Aminex HPX-87H (300 × 7.8 mm) column was used to separate the fermentation products, and 0.005 M H 2 SO 4 was used as the mobile phase at a flow rate of 0.5 mL/min. Temperatures of detector and column were 40 and 60 • C, respectively. The measurement of glucose and ethanol by HPLC and Biosensor analyzer M-100 was compared, and the results showed that the measurement data of the two instruments on the same sample were close (data not shown). At least two independent experiments were carried out with similar results. Values are the mean of one representative experiment with two or more technical replicates. Error bars represent standard deviations.
One percent (w/v) of different organic nitrogen sources were prepared in ddH 2 O and then filter-sterilized. The concentrations of different ions in these samples were then measured by Inductively Coupled Plasma Optical Emission Spectroscopy (ICP-OES, Wuhan SouSepad Testing Technology Co., Ltd., Wuhan, China).

Ethanol Production Calculations
Cell growth was monitored by optical density spectrophotometrically at 600 nm. The time points when cells reached stationary phase and when glucose was completely consumed with the maximum ethanol produced were recorded. The concentrations of glucose and ethanol determined by HPLC or Biosensor analyzer M-100 were then used for the calculation of the ethanol yield (Y p/s ), theoretical ethanol yield (%Y p/s ), ethanol productivity (Q P ), and specific growth rate (µ) using the formula below: Y p/s = maximum ethanol (g)/consumed glucose (g); %Y p/s = (ethanol(g)/(consumed sugar(g)*0.511))*100; Q P = maximum ethanol titer (g/L)/time to reach the maximum ethanol. µ = ln((OD 600 at t 2 )/(OD 600 at t 1 ))/(t 2 −t 1 ), t 1 and t 2 are the time points in the log phase.

Transcriptomic Analysis
The transcriptomics study followed previous work (Franden et al., 2009;Zhao et al., 2009;Martinez-Moreno et al., 2012;Kemsawasd et al., 2015;Klotz et al., 2017;Yang S. et al., 2018;Lozano Terol et al., 2019). Briefly, cell culture samples were collected during the exponential phase with the OD 600 values around 0.6-0.8. RNA-Seq was performed using paired-end sequencing technology according to standard Illumina protocols (Genewiz Inc., Suzhou, China). The quality of RNA-Seq fastq data was evaluated using FastQC software (Babraham Bioinformatics, United States). Data passing the quality control were imported into CLC Genomics Workbench (Ver. 11.0) for RNA-Seq analysis to get the RPKM value (reads mapping to the genome per kilobase of transcript per million reads sequenced) of each gene with the reference genome. Genome sequence of Z. mobilis was used as the reference for RPKM calculation (Yang et al., 2010;Martinez-Moreno et al., 2012). The RPKM value of each gene was then imported into JMP Genomics (Ver. 9.0); data normalization and statistical analysis were conducted to identify differentially expressed genes when three different media were used. Duplicate samples were used for each condition. RNA-Seq raw data were deposited at Sequence Read Archive (SRA) database with the Bioproject accession number of PRJNA601020.

DATA AVAILABILITY STATEMENT
The RNA-Seq raw data was deposited at Sequence Read Archive (SRA) database with the BioProject accession number PRJNA601020.