Enhancement of Thermal Resistance by Metal Ions in Thermotolerant Zymomonas mobilis TISTR 548

The thermal resistance of fermenting microbes is a key characteristic of stable fermentation at high temperatures. Therefore, the effects of various metal ions on the growth of Zymomonas mobilis TISTR 548, a thermotolerant ethanologenic bacterium, at a critical high temperature (CHT) were examined. Addition of Mg2+ and K+ increased CHT by 1°C, but the effects of the addition of Mn2+, Ni2+, Co2+, Al3+, Fe3+, and Zn2+ on CHT were negligible. To understand the physiological functions associated with the addition of Mg2+ or K+, cell morphology, intracellular reactive oxygen species (ROS) level, and ethanol productivity were investigated at 39°C (i.e., above CHT). Cell elongation was repressed by Mg2+, but not by K+. Addition of both metals reduced intracellular ROS level, with only K+ showing the highest reduction strength, followed by both metals and only Mg2+. Additionally, ethanol productivity was recovered with the addition of both metals. Moreover, the addition of Mg2+ or K+ at a non-permissive temperature in 26 thermosensitive, single gene-disrupted mutants of Z. mobilis TISTR 548 revealed that several mutants showed metal ion-specific growth improvement. Remarkably, K+ repressed growth of two mutants. These results suggest that K+ and Mg2+ enhance cell growth at CHT via different mechanisms, which involve the maintenance of low intracellular ROS levels.


INTRODUCTION
Recently, bioethanol has gained attention as an alternative to fossil fuel because as a carbonneutral fuel, it can potentially delay the progress of global warming (Hahn-Hägerdal et al., 2006;Chisti, 2008). However, industrial scale bioethanol production requires a more costeffective process to be economically competitive. High-temperature fermentation (HTF; which enables fermentation at a temperature 5-10 • C higher than that used in the conventional process) may reduce (1) cooling cost, (2) running cost at the simultaneous saccharification and fermentation stage, and (3) contamination risks (Abdel-Banat et al., 2010;Kosaka et al., 2018). Bioethanol production by HTF requires high-efficiency ethanol production and thermotolerant microorganisms. Zymomonas mobilis, a Gram-negative, facultative, anaerobic bacterium, performs high-speed ethanol production (He et al., 2014) compared with the conventional ethanol producer Saccharomyces cerevisiae, whose cultivation temperature of >35 • C is not permissible for yeast growth (Aldiguier et al., 2004). Z. mobilis, which uses the Entner-Doudoroff pathway and an incomplete TCA cycle, is facultatively anaerobic and requires no oxygen for its growth; conversely, it assimilates glucose, fructose, and sucrose as the sole carbon sources (Panesar et al., 2006;He et al., 2014). We focused on Z. mobilis TISTR 548, one of the thermotolerant Z. mobilis strains that grew at 39 • C , and developed thermotolerant mutants by thermal adaptation enhancement of its critical high temperature (CHT), an upper limit for survival, up to 2 • C (Matsushita et al., 2016;Kosaka et al., 2019). We subsequently used this mutant strain with HTF using a model fermentation and distillation system to reveal the effectiveness of this method and bioethanol productivity by HTF with Z. mobilis (Murata et al., 2015).
Metal ions enhance the growth of ethanologenic microorganisms at CHT. Microorganisms require several ions as essential metals for the normal function and homeostasis of a wide range of cellular proteins (Reid et al., 2009), but these ions are toxic at high concentrations (Gadd, 1992). Among these ions, only Mg 2+ has been reported to enhance thermotolerance in Z. mobilis strains . Moreover, Mg 2+ helped recover thermosensitive mutants of Z. mobilis TISTR 548, in which genes for membrane stabilization or membrane formation were disrupted, suggesting that at CHT, Mg 2+ stabilizes membrane structure and protects cells from heat (Charoensuk et al., 2017). Mg 2+ also stabilizes the outer membrane (OM) structure, at least of lipopolysaccharide (LPS), of cells by divalent cation crossbridging (bridging action) in Gram-positive bacteria (Nikaido, 2003). Studies on several microorganisms, particularly Escherichia coli (Murata et al., 2011) and Lactobacillus strains (Yang et al., 2017), have revealed the thermotolerance-enhancing effect of Mg 2+ . However, although the enhancing effects of K + and Ca 2+ on S. cerevisiae (Lam et al., 2014) and lactic acid bacteria (Huang and Chen, 2013), respectively, have been reported, there is no report about the enhancement effects of these and other metals on Z. mobilis growth at CHT. This evidence suggests that the concentration of several metals in a fermentation medium is important for efficient HTF for bioethanol production. However, the effect of a wide range of metals on Z. mobilis TISTR 548 at CHT is yet to be investigated. Besides, the mechanism underlying the effects of these metals, such as Mg 2+ , on cell physiology at CHT remains unclear.

Bacterial Strains, Media, and Cultivation Conditions
The bacterial strains used in this study are listed in Table 1. To grow Z. mobilis, a preculture was prepared in 2 mL of YPD medium (0.3% yeast extract, 0.5% peptone, and 3% glucose) and incubated overnight at 30 • C. The overnight culture was subsequently inoculated into fresh YPD medium at an OD 550 of 0.05. Cultivation was performed under non-shaking (static) conditions.

Examination of the Effects of Various Materials on Cell Growth
To compare the effects of additional reagents, cells were subjected to two-step cultivation (Kosaka et al., 2019) at the same temperature to observe the effect of temperature or additional reagents. Two-step cultivation can simply determine the temperature-upper limit for the survival of cells because when the first culture is performed at a temperature just above a CHT, cells cannot grow in the second culture at the same temperature (Kosaka et al., 2019). In the first culture, the OD value of the culture increases even at a temperature over CHT because of cell elongation. The CHT of Z. mobilis TISTR 548 has been determined to be 38 • C by this method (Kosaka et al., 2019). Reagents were added to a medium at the desired condition before each inoculation. Briefly, the first cultivation was performed until the culture attained a late log phase at a temperature around a putative CHT; then, a portion of the first culture was transferred into a fresh medium at an OD 550 of 0.05 and cultured at the same temperature. All metals tested were obtained in the form of chloride salts.

Intracellular Reactive Oxygen Species Level
Zymomonas mobilis TISTR 548 cells were grown on YPD medium at 39 • C. At 12 h, 5 µM H 2 DCFDA was added to the first culture, and further cultivation was performed at 39 • C for 30 min. Then, cells were harvested by low-speed centrifugation and washed once with phosphate-buffered saline [130 mM NaCl, 10.8 mM Na 2 HPO 4 , 4.2 mM NaH 2 PO 4 (pH 7.2)]. The washed cells were disrupted by sonication for 30 min using an ultrasonic cell disruptor (Bioruptor; Cosmo Bio, Tokyo, Japan) and subjected to low-speed centrifugation. Supernatant fluorescence was measured using a microplate reader (POWERSCAN R HT; BioTek Instruments, Inc., Winooski, VT, United States). Protein concentration was determined using the Lowry method (Dulley and Grieve, 1975). The result obtained for intracellular reactive oxygen species (ROS) levels is expressed as fluorescence intensity per protein concentration, and the ratio of the number of cells grown in the presence of a metal ion to that of cells grown in its absence was estimated and expressed as percentage.

Effects of Metal Ions on Z. mobilis TISTR 548 Growth at CHT
To explore the metal ions that enhance Z. mobilis TISTR 548 growth at putative CHT, the growth levels with and without the addition of Ni 2+ , Zn 2+ , Fe 3+ , Al 3+ , Mn 2+ , Co 2+ , Mg 2+ , and K + were compared. The effect of the addition of metal ions was evaluated with two-step cultivation, wherein only viable and culturable cells grow, whereas dead or viable but non-culturable cells do not grow in fresh medium at the second cultivation (Kosaka et al., 2019). At 38 • C and 39 • C, the growth level under the conditions of 0.01 mM NiCl 2 , ZnCl 2 , FeCl 3 , AlCl 3 , MnCl 2 , and CoCl 2 was the same as that without the addition of metal ions (Figure 1). On the contrary, the addition of >0.1 mM NiCl 2 and CoCl 2 led to a lower growth level than no addition of metal ions at the first stage of cultivation (Figures 1A,B,K,L). Similarly, the growth level following the addition of ZnCl 2 and MnCl 2 was lower at 1 mM (Figures 1C,D,I,J). The growth trend did not change distinctly between 38 and 39 • C (Figure 1). Ten millimolar CaCl 2 or 10 mM NaCl suppressed growth in the second step of Z. mobilis TISTR 548 cultivation at 38 • C (data not shown).
On the other hand, when MgCl 2 and KCl were added to the medium, there was a 1 • higher growth than there was without adding metals even at 39 • C (Figures 2A-C). At 39.5 • C, there was negligible growth in the presence of MgCl 2 and KCl ( Figure 2D). The results suggested that the optimum concentrations of MgCl 2 and KCl for growth enhancement at 39 • C were 5 and 30 mM, respectively.
Physiological Effects of Mg 2+ and K + on Z. mobilis TISTR 548 at CHT Our previous report indicated that the cell length of Z. mobilis increased at CHT, and this increase reduced in thermotoleranceenhanced mutants (Kosaka et al., 2019). Indeed, cells grown at 39 • C had longer cells than those grown at 30 • C, which had granular shapes (Figures 3A,B). Cell morphology observed following the addition of MgCl 2 or KCl indicated that cell length increased, with the increase in cell length being relatively lower following the addition of Mg 2+ at 39 • C than without the addition of metal ions ( Figure 3C). On the other hand, the addition of KCl had no clear effect on cell length at 39 • C, with a predominance of longer filamentous cells ( Figure 3D). Cells cultured with both metals showed a mixture of granular and long filamentous shapes ( Figure 3E). Indeed, the median value of measured cell length at 30 • C, at 39 • C, with MgCl 2 at 39 • C, with KCl at 39 • C, or with both metals at 39 • C was 3.3, 7.6, 5.0, 7.2, or 6.1 µm, respectively ( Figure 3F). Ethanol productivity at 39 • C was also recovered to be close to the theoretical yield by adding MgCl 2 or KCl ( Figure 3G). Accumulation of intracellular ROS was observed in Z. mobilis TISTR 548 at CHT (Kosaka et al., 2019). Addition of MgCl 2 or KCl considerably reduced intracellular ROS levels at 39 • C, and the reduction strength was the highest for only K + , followed by that for both metals and then only Mg 2+ (Figure 3H). Effects of Mg 2+ and K + on Z. mobilis TISTR 548 Growth at CHT Previous results indicated that Mg 2+ and K + somehow affect the cell physiology of Z. mobilis TISTR 548 at CHT and reduce intracellular ROS levels but probably by different mechanisms. Several bacteria use glutathione as a reducing agent to maintain a strongly reducing environment in cells, and glutathione peroxidase is an ROS-scavenging enzyme (Cabiscol Català et al., 2000). We observed the effect of glutathione with MgCl 2 or KCl on cell growth when glutathione was added at several concentrations: 4 mM glutathione inhibited cell growth at 39 • C (data not shown) but 0.5 mM did not ( Figure 4A). An Mg 2+ plus glutathione effect was observed, but the effect was not considerably distinct from that observed following the addition of K + or both metals (Figures 4B-D). Next, an effective concentration of EDTA as a chelator of a divalent cation on cell growth at a CHT were explored, and then effects of metals under the presence of such a concentration of EDTA at a CHT were examined. When 0.05 mM EDTA was added to culture, cell growth was inhibited at 38 • C ( Figure 5A). MgCl 2 or KCl was subsequently added under the above condition, and the resulting effect was observed. The addition of Mg 2+ rescued EDTA inhibition at CHT (Figure 5B), but that of K + did not ( Figure 5C).

Effects of Mg 2+ and K + on the Growth of Thermosensitive Mutants
In a previous study, 26 thermosensitive single gene-disrupted mutants of Z. mobilis TISTR 548 were isolated (Charoensuk et al., 2017). To observe the effects of the addition of Mg 2+ and K + on these thermosensitive mutants, their growth upon the addition of each metal was examined. Addition of Mg 2+ significantly enhanced the growth of three mutants, whereas it inhibited the growth of one mutant (Table 2). Further, the addition of K +  a Thirty-six thermotolerant genes were identified, and they were classified into 10 groups (Charoensuk et al., 2017). b The growth of these mutants at 38, 39, and 39.5 • C compared with that of the parental strain on YPD medium. The symbols "+" represent the degree of cell growth of mutants at a critical high temperature compared with that of the parental strain, whereas "−" indicates no growth. c The ratio (%) was calculated from the OD 550 values for cells grown at 39.5 • C for 24 h with 5 mM MgCl 2 or 30 mM KCl divided by those cells grown without metals. Values represent means ± standard deviations of three replicates. Bold values indicate a ratio of >1.5 or <0.5 against the parental strain.
enhanced the growth of nine mutants, i.e., almost one-third of all thermosensitive mutants, but it inhibited the growth of one mutant (C13-36) at 39.5 • C ( Table 2). Remarkably, the addition of both Mg 2+ and K + enhanced the growth of only one mutant with gene encoding for phospholipase D; they had the opposite effect on the mutant TC13 ( Table 2). These results also suggest that Mg 2+ and K + affect the cell physiology of Z. mobilis TISTR 548 differently at CHT.

DISCUSSION
We observed the additional effects of Ni 2+ , Zn 2+ , Fe 3+ , Al 3+ , Mn 2+ , and Co 2+ on Z. mobilis TISTR 548 growth at high temperatures, but these metals showed only negative effects (Figure 1). Among these, the effects of Fe 3+ and Al 3+ were negligible under the tested conditions (Figures 1E-H). However, 1.0 mM Ni 2+ , Zn 2+ , Mn 2+ , and Co 2+ clearly inhibited Z. mobilis TISTR 548 growth (Figure 1). A previous report indicated that the addition of 0.35 mM Zn 2+ markedly inhibited ethanol productivity in Z. mobilis ZM4 probably by inhibiting metabolic enzymes (Liu et al., 2010). In the case of a different microorganism, i.e., S. cerevisiae, the quantity of Zn 2+ , Mg 2+ , and Mn 2+ required for effective fermentation was 0.01, 0.05, and 0.04 g/L, respectively (Deesuth et al., 2012), implying that 0.1 mM metals are usually required for growth, but excess concentrations can inhibit growth. This implies that only specific metals enhance the growth of specific microorganisms. Only Mg 2+ and K + enhanced Z. mobilis TISTR 548 growth and improved CHT from 38 to 39 • C (Figure 2A). At CHT, the intracellular molecular components of bacterial cells leaked (Haight and Morita, 1966;Allwood and Russell, 1967). In S. cerevisiae and probably other microorganisms, the addition of K + prevents ion leakage (Lam et al., 2014). K + channels are activated when tension in the lipid bilayer is increased (Iwamoto and Oiki, 2018), and Mg 2+ transporters are induced by heat treatment in Salmonella enterica (O'Connor et al., 2009). Regarding quantity, these two metal ions (K + and Mg 2+ ) are the principal and second highest ions in bacterial cells found at concentrations of 100-500 mM (Ballal et al., 2007) and ∼1 mM (Groisman et al., 2013), respectively. Therefore, the optimal concentrations of 30 mM K + and 5 mM Mg 2+ (Figures 2A,B) are probably related to their intracellular concentrations, further suggesting that similar ion conditions enhance cell metabolism by preventing ion leakage from cells or supporting ion transportation from the extracellular space. However, the effects of Mg 2+ and K + on the two common characteristics of bacteria, namely, cell elongation and ROS accumulation, observed in Z. mobilis TISTR 548 at CHT (Matsushita et al., 2016) were different. Cell length at CHT was suppressed by the addition of Mg 2+ but not by that of K + (Figure 3F). Although ROS accumulation reduced by the addition of both metals, the addition of K + showed a stronger effect than that of Mg 2+ (Figure 3H). The GSH results indicated that the additive effect of GSH was observed in both cases of Mg 2+ and K + (Figures 4B,C), suggesting that the growth enhancement effect of Mg 2+ or K + does not arise directly from the action of GSH added exogenously. In a Gram-negative bacterium, E. coli, GSH is important for periplasmic redox homeostasis (Pittman et al., 2005) and heterogeneous expression of glutathione reductase allows the microbe to be hydrogen peroxide tolerance (Kim et al., 2009). It is assumed that, in Z. mobilis TISTR 548, GSH keeps periplasmic redox homeostasis and/or somehow makes cells tolerate oxidative stress by its reducing power, but the major effects at CHT by Mg 2+ and K + are not likely the action by GSH. Moreover, EDTA treatment showed that K + did not complement the EDTA effect at CHT (Figure 5C). These results suggest that Mg 2+ and K + affect the cell physiology of Z. mobilis TISTR 548 at CHT using different mechanisms.
The effect of Mg 2+ on the cell physiology of Z. mobilis TISTR 548 at CHT has been described: Mg 2+ probably stabilizes membrane structure as proposed in E. coli (Charoensuk et al., 2017). Mg 2+ stabilizes OM (Nikaido and Vaara, 1985), particularly LPS, where Mg 2+ bridges lipid A (Nikaido, 2003). The present study results also showed that the addition of Mg 2+ repressed cell elongation at CHT ( Figure 3F) and restored the growth of the disrupted genes of ZZ6_0923, which encodes the cardiolipin biosynthesis protein (Table 2). However, Mg 2+ has been thought to stabilize proteins, enhance protein-nucleic acid interactions, mitigate oxidative stress, and act as a metabolic signal (O'Connor et al., 2009). Mg 2+ is required to maintain cell metabolism, DNA replication, transcription and translation, and DNA stabilization (Xu et al., 2018), and it plays a role in enzyme activations. For instance, Mg 2+ stabilizes pyruvate decarboxylase, an enzyme responsible for the decarboxylation of pyruvate in central metabolism, with thiamine diphosphate serving as a cofactor (Pohl et al., 1994). Besides, phosphoglycerate kinase uses Mg 2+ as a cofactor (Andreini et al., 2008). Addition of Mg 2+ reduced ROS accumulation at CHT (Figure 3H), and the growth of the disrupted Fe 2+ /Mn 2+ transporter (ZZ6_1289) recovered greatly ( Table 2). Therefore, maintaining an intracellular Mg 2+ concentration may enable heat tolerance either by ions or cytoplasmic Mg 2+ sensors, proteins, and RNAs. (Groisman et al., 2013).
K + , the most dominant intracellular cation, greatly contributes to pH homeostasis and turgor maintenance as well as bacterial osmotic adaptation, pH regulation, gene expression, and cell enzyme activation (Epstein, 2003;Ballal et al., 2007). Indeed, the addition of K + affected most cell physiology of Z. mobilis TISTR 548 at CHT (Table 2), e.g., the growth of 35% (9/26 strains) of mutants recovered greatly. These effects may contribute to reducing intracellular ROS levels ( Figure 3H). Between these, K + probably facilitates the functioning of periplasmic proteins in Z. mobilis TISTR 548 due to the growth recovery of disrupted tolQ, secD, tamB, and degP ( Table 2). The amount of intracellular K + directly affects membrane potential (Bakker and Mangerich, 1981), which is required for protein secretion to periplasm (Daniels et al., 1981). The membrane potential is hypothetically required for potassium transport from extracellular space to the cytoplasm via the membrane potentialdriven K + uptake system (Kup, ZMO1209, and ZZ6_0125) in Z. mobilis. Therefore, K + may also facilitate membrane potential maintenance in Z. mobilis TISTR 548.
Under CHT conditions (Figure 6A), the inner membrane fluidity increases to cause leakages of ions from cytoplasm and electrons from the respiratory chain, which lead to the accumulation of intracellular reactive oxygen species, resulting in damage of macromolecules of DNA, RNA, proteins and lipids, and thereby cells are elongated and unable to maintain intracellular homeostasis, causing cell death. However, by the addition of Mg 2+ (Figure 6B), the OM is stabilized by binding of Mg 2+ and the inner membrane is also stabilized, resulting in suppression of the leakage of intracellular ions as well as the leakage of electrons from the respiratory chain. On the other hand, by the addition of K + (Figure 6C), K + leakage is repressed to maintain homeostasis for cellular metabolism, by which intracellular ROS is reduced. Moreover, these observations suggest that Mg 2+ and K + exhibit diverse, rather than single, effects on Z. mobilis TISTR 548. Interestingly, when both Mg 2+ and K + exist in the medium at high concentrations, their crosstalk effects on cell physiology sometimes occur. These effects are partly specific to each ion; their additive effect on cell growth at CHT was observed, but that did not entail the whole sum of their effects ( Figure 2C). The thermotolerance acquisition mechanisms of Z. mobilis upon the addition of Mg 2+ and K + are more complex than the accumulated effects of their metals in accomplishing enhanced Z. mobilis growth at CHT.

CONCLUSION
Among various metals, only Mg 2+ and K + enhanced the thermotolerance of Z. mobilis TISTR 548. The primary effects of Mg 2+ and K + on the cell physiology of Z. mobilis TISTR 548 are largely different, but these metals reduce intracellular ROS accumulation. Based on the study results, several strategies for improving the CHT of Z. mobilis by membrane stabilization and intracellular metabolism maintenance can be expected. Further research is needed to reveal these mechanisms for improving its growth at CHT.

DATA AVAILABILITY STATEMENT
All datasets generated for this study are included in the article/supplementary material.