Systems-level analysis of 1,6-hexanediol stress response and tolerance engineering in Escherichia coli

Abstract

1,6-Hexanediol (1,6-HDO) is an important chemical platform widely used in polymer and pharmaceutical industries, while its toxicity remains a major bottleneck limiting efficient microbial production. This study demonstrates that 1,6-HDO exhibited strong inhibitory effects on Escherichia coli, reducing the maximum OD₆₀₀ to approximately 50% of the control level under 10 g/L stress. Transcriptomic analysis was performed under 1,6-HDO stress, identifying 779 differentially expressed genes enriched in pathways related to energy metabolism, sulfur metabolism and amino acid metabolism. To systematically identify functional targets associated with tolerance, genome-wide screening using the ASKA overexpression library identified 93 tolerance-related genes that improved E. coli growth under 1,6-HDO stress by 0.06- to 3.17-fold. Notably, multiple targets were functionally associated with energy and amino acid metabolism, consistent with the transcriptomic analysis. Furthermore, co-expression of rumA and yhbO showed the greatest improvement in growth, reaching a biomass of 9.43 under 1,6-HDO stress, which was 2.61-fold higher than that of the control strain. Collectively, this study systematically elucidates the global stress response and identifies key determinants contributing to tolerance, providing valuable targets for developing robust microbial cell factories for efficient 1,6-HDO bioproduction.

1. Introduction

1,6-Hexanediol (1,6-HDO), a linear diol widely used in the production of polyesters, polyurethanes, and coatings, has attracted increasing industrial interest due to its broad applications and growing market demand [1]. Recently, microbial biosynthesis of 1,6-HDO has emerged as a sustainable alternative to traditional chemical synthesis, and several studies have demonstrated the feasibility of its production in Escherichia coli [1,2]. However, accumulation of 1,6-HDO exerts inherent toxicity on the host strain, thereby impairing cell robustness and limiting production efficiency.

Alcohol-based chemicals generally impair microbial growth through multiple stress responses, including membrane perturbation, osmotic imbalance, oxidative stress, and intracellular toxicity accumulation [[3], [4], [5]]. Similar toxicity patterns have also been observed for structurally related short-chain diols, such as 1,2-propanediol and 2,3-butanediol [6]. To improve tolerance against chemical stress, current engineering strategies mainly rely on local genetic modifications (e.g., membrane、efflux and global regulator engineering), as well as adaptive laboratory evolution. In the case of 1,6-HDO, reducing intracellular accumulation through export engineering has also been proposed as a potential strategy [1]. However, chemical stress responses are typically governed by complex regulatory and metabolic networks, making it difficult for local engineering approaches alone to systematically identify effective tolerance determinants.

In recent years, researchers have widely adopted transcriptomic methods to systematically characterize bacterial responses to various alcohols, such as ethanol [7], isobutanol [8], and isopropanol [9]. Meanwhile, genome-wide functional screening strategies based on overexpression, knockout, or CRISPR interference (CRISPRi) libraries have emerged as powerful tools for identifying tolerance-associated genes and pathways. For example, genome-wide overexpression in Corynebacterium glutamicum identified novel targets associated with osmotic tolerance and l-lysine production, revealing the underlying mechanisms [10], while screening of E. coli Keio knockout library revealed key genes involved in tolerance to trans-cinnamic acid [11]. In addition, genome-scale CRISPRi screening has enabled systematic identification of stress-responsive genes under antibiotic pressure [12]. Despite these advances, systematic studies on tolerance to 1,6-HDO remain limited, and a genome-wide understanding of its toxicity mechanism, transcriptional response, and key tolerance determinants is still lacking.

In this study, we systematically investigated the cellular response of E. coli to 1,6-HDO stress by combining transcriptomic analysis with genome-wide overexpression library screening. RNA-seq analysis revealed key stress-responsive pathways associated with energy metabolism, sulfur metabolism, and amino acid biosynthesis, providing a systems-level understanding of the cellular response to 1,6-HDO stress. Building on this mechanistic insight, genome-wide screening using the ASKA overexpression library identified 93 tolerance-related genes. Individual overexpression of these genes conferred varying degrees of stress tolerance, with some genes (e.g., genes pfkA, yfbU) showing the most pronounced protective effects. Combinatorial expression of tolerance-associated genes markedly improved cellular resistance to 1,6-HDO stress, with the best-performing combination achieving a 2.61-fold increase in biomass over the control strain, demonstrating the effectiveness of the engineering strategy. Overall, this study provides systems-level insights into the cellular response and tolerance mechanisms of E. coli under 1,6-HDO stress and offers valuable genetic targets for engineering robust microbial cell factories.

2. Materials and methods

2.1. Strains, plasmids, and culture conditions

The strain used for transcriptomic analysis and genome-wide overexpression screening was E. coli KA-12 (AG1) (Table S1). The ASKA (A Complete Set of E. coli K-12 ORF Archive) library, which consists of 4123 individual strains, each harboring a pCA24N-based plasmid carrying an open reading frame (ORF) from E. coli K-12 W3110 under the control of an IPTG-inducible T5/lac promoter [13], was used for tolerance screening. The empty pCA24 N vector served as the control. Regarding the culture media, LB medium (used for strain activation and seed culture) contained 10 g/L tryptone, 5 g/L yeast extract, and 5 g/L NaCl; for solid medium, 20 g/L agar was added. M9 minimal medium (used for growth inhibition assays and tolerance screening) contained 17.91 g/L Na₂HPO₄, 2.72 g/L KH₂PO₄, 1.07 g/L NH₄Cl, and 0.58 g/L NaCl, supplemented with 0.4% (w/v) glucose. Rich medium (used for combinatorial tolerance screening) contained 20 g/L glucose, 10 g/L (NH₄)₂SO₄, 5 g/L peptone, 23.13 g/L 3-morpholinopropanesulfonic acid (MOPS) sodium salt, 1.6 g/L MgSO₄·7H₂O, 0.5 g/L KCl, 2 g/L yeast extract, 0.3 g/L CaCO₃, and trace elements including 0.08 g/L ZnSO₄·7H₂O, 0.3 g/L methionine, 0.07 g/L CuSO₄, 0.01 g/L nicotinamide, 0.03 g/L MnSO₄·H₂O, 0.3 g/L threonine, 30 μg/L biotin, 0.06 g/L vitamin B₁, and 0.03 g/L FeSO₄·7H₂O. The pH was adjusted to 7.0 using HCl or NH₄OH. Medium containing 1,6-HDO was prepared by adding filter-sterilized 1,6-HDO to the above M9 and rich medium to desired concentrations (5, 10, 15, 20 g/L).

2.2. Growth inhibition curve and survival rate determination of E. coli under 1,6-HDO stress

The overnight seed culture of E. coli KA-12 (AG1) was inoculated into media containing different concentrations of 1,6-HDO (0, 5, 10, 15, 20 g/L) at an initial OD₆₀₀ of 0.1. Cultures were grown in 250 mL shake flasks containing 50 mL of medium and incubated at 37°C with shaking at 200 rpm for 24 h. Samples were taken every 4 h, and the optical density at 600 nm (OD₆₀₀) was measured using a spectrophotometer to construct growth curves. For cell viability determination, aliquots of the cultures after 12 h of growth were serially diluted to 10²-, 10⁴-, and 10⁶-fold with sterile water. Then, 100 μL of each dilution was spread onto LB agar plates and incubated at 37°C for 24 h, after which the colonies were counted. Viability (%) was calculated as (the number of colonies on 1,6-HDO-containing plates/the number of colonies on control plates) × 100%.

2.3. Transcriptome sequencing and data analysis

E. coli KA-12 (AG1) was cultured in M9 medium containing 15 g/L 1,6-HDO for 18 h, using the same culture conditions without 1,6-HDO as a control. Cells were harvested by centrifugation at 4000 rpm for 10 min at 4°C, washed twice with pre-chilled PBS, snap-frozen in liquid nitrogen, and stored at −80°C. The samples were sent to Shanghai OE Biotech Co., Ltd. for RNA extraction, rRNA depletion, cDNA synthesis, library construction, and Illumina sequencing. After quality control, the sequencing data were aligned to the E. coli K-12 (AG1) reference genome. Differentially expressed genes (DEGs) were identified using DESeq2 with the criteria |log₂ FC| ≥ 1 and adjusted P < 0.05.

2.4. Determination of intracellular ATP levels

E. coli KA-12 (AG1) was cultured in M9 medium containing a series of concentrations of 1,6-HDO for 18 h. The cells were harvested by centrifugation at 4°C and 4000 rpm for 10 min, then washed twice with pre-cooled PBS. The cells were resuspended to an OD₆₀₀ of approximately 1, following centrifugation, the pellet was resuspended in 1 mL of extraction buffer. The cells were disrupted by sonication in an ice bath (30% amplitude, 2 s on and 1 s off pulses, total 1 min). The lysate was centrifuged at 4°C and 10,000 rpm for 10 min, and the supernatant was mixed with 500 μL of chloroform by vortexing. After another centrifugation for 3 min, the upper aqueous phase was collected. ATP concentration was determined using an ATP assay kit (Solarbio) at a wavelength of 340 nm, and the ATP concentration was calculated based on a standard curve.

2.5. Determination of intracellular reactive oxygen species (ROS) levels

Intracellular ROS levels were measured using a Reactive Oxygen Species Assay Kit (Beyotime, DCFH-DA method). E. coli KA-12 (AG1) was cultured in M9 medium containing a series of concentrations of 1,6-HDO for 18 h. The cells were harvested by centrifugation at 4°C and 4000 rpm for 10 min, washed twice with pre-cooled PBS, and resuspended to an OD₆₀₀ of 1.0. Then, 1 μL of DCFH-DA (final concentration 10 μM) was added, and the mixture was incubated in a shaking incubator at 37°C for 1 h in the dark. After incubation, the cells were washed twice with PBS to remove extracellular probe, and resuspended in 1 mL of PBS. Fluorescence intensity was measured using a fluorescence spectrophotometer (excitation wavelength: 488 nm, emission wavelength: 525 nm). Results were expressed as relative fluorescence units per OD₆₀₀.

2.6. Screening of 1,6-HDO tolerance genes using the ASKA library

For library activation, 180 μL of LB medium (containing 34 μg/mL chloramphenicol) was added to each well of a 96-well shallow-well plate, and 20 μL of each culture was inoculated, followed by incubation at 37°C in a constant-temperature shaker at 200 rpm for 8 h. For primary screening of tolerance genes, the activated ASKA library cultures were transferred at a 10% (v/v) inoculum into a new 96-well plate containing 180 μL per well of M9 medium supplemented with 15 g/L 1,6-HDO, 34 μg/mL chloramphenicol, and 0.5 mmol/L IPTG, and incubated at 37°C and 200 rpm for 18 h. The OD₆₀₀ was measured using a microplate reader, with the empty vector pCA24 N strain as a control, and clones showing significantly higher OD₆₀₀ compared to the control were selected as positive candidates. For plasmid extraction and sequencing, positive clones were streaked for isolation on LB agar plates containing 34 μg/mL chloramphenicol; single colonies were picked and inoculated into 5 mL of LB broth (containing 34 μg/mL chloramphenicol) and cultured for 12 h. Plasmids were extracted using the Tiangen Plasmid Mini Kit and sent to a sequencing company for insert fragment sequencing to identify the overexpressed genes. For shake-flask re-screening, the candidate genes confirmed by sequencing were retransformed into E. coli KA-12 (AG1), and the transformants were cultured in 250 mL shake flasks containing 50 mL of M9 medium (supplemented with 15 g/L 1,6-HDO, 34 μg/mL chloramphenicol, and 0.5 mmol/L IPTG) for 18 h, after which the OD₆₀₀ was measured to further validate the tolerance phenotype.

2.7. Construction of combinatorial genes

Using plasmids from the ASKA library as templates, specific primers were designed to amplify the target genes (e.g., yhbO, rumA, etc.). Meanwhile, the linearized vector fragment was amplified from the pCA24 N empty plasmid using primers pCA24 N F/R. To insert the second gene into the pCA24 N plasmid already containing the first gene, a homologous recombination kit (ClonExpress II One Step Cloning Kit, Vazyme, Nanjing, China) was used. The ligation product was transformed into competent cells, and the transformants were plated on chloramphenicol-containing agar plates. Single colonies were selected and verified by colony PCR and sequencing, yielding dual-gene co-expression strains. The plasmid harboring the two genes, confirmed by sequencing, was re-transformed into E. coli KA-12 (AG1) and E. coli NT1003. The transformants were inoculated into 250 mL shake flasks containing 50 mL of rich medium supplemented with 15 g/L 1,6-HDO, 34 μg/mL chloramphenicol, and 0.5 mmol/L IPTG. After 18 h of cultivation, the OD₆₀₀ was measured to further verify the tolerance phenotype.

2.8. Statistical analysis

Bioinformatics analysis of DEGs: DEGs were subjected to enrichment analysis using the Gene Ontology (GO) database (http://www.geneontology.org/) and the Kyoto Encyclopedia of Genes and Genomes (KEGG) database (http://www.genome.jp/kegg/pathway.html) to identify their biological functions and molecular pathways.

3. Results and discussion

3.1. Inhibitory effect of 1,6-HDO on cells

To evaluate the inhibitory effect of 1,6-HDO on E. coli KA-12 (AG1), cells were cultivated under different 1,6-HDO concentrations (0–20 g/L) and growth performance was monitored. 1,6-HDO exhibited a clear concentration-dependent inhibitory effect on cell growth (Fig. 1A). At 5 g/L, cell growth was only slightly affected, with a maximum biomass (OD_{600, max}) of 2.75 (85.97% of the control). However, increasing the concentration to 10 g/L and 15 g/L markedly reduced OD_{600, max} to 1.82 and 1.33, respectively, accompanied by decreased survival rates of 88.15% and 46.96% (Fig. 1B). At 20 g/L, cell growth was almost completely inhibited. In addition, the maximum specific growth rate (μ_max) gradually decreased with increasing 1,6-HDO concentrations, from 0.47 h⁻¹ in the control to 0.38 at 5 g/L and 0.36 h⁻¹ at 10–15 g/L, respectively. Moreover, the period during which μ_max was reached shifted from 0 to 4 h in untreated cells to 4–8 h under 1,6-HDO exposure, indicating that 1,6-HDO not only reduced the growth rate but also prolonged the lag phase of E. coli (Table S2). Similar concentration-dependent inhibitory effects have also been reported for other alcohols [14]. Notably, previous studies have reported that engineered E. coli can achieve 1,6-HDO titers up to 14.5 g/L [15]. This suggests that industrially relevant production levels are already approaching concentrations that strongly inhibit cell growth. Therefore, product toxicity represents a major constraint for further improvement of 1,6-HDO biosynthesis. Alleviating product inhibition will be essential for achieving higher production levels.

Fig. 1.

Effects of various concentrations of 1,6-HDO on the growth and viability of E. coli KA-12 (AG1). Growth curves (A) and survival rates (B) of E. coli KA-12 (AG1) exposed to different concentrations of 1,6-HDO. Data are presented as mean ± SD from three independent biological replicates, *p < 0.05, **p < 0.01 vs. control group.

3.2. Transcriptomic analysis of the response to 1,6-HDO stress

The effects of 1,6-HDO supplementation on E. coli transcriptome were analyzed by RNA-seq. Differential expression analysis identified 394 up-regulated and 385 down-regulated DEGs in the 1,6-HDO group compared to the control (Fig. 2A). KEGG pathway enrichment analysis revealed that DEGs were significantly involved in flagellar assembly, glyoxylate metabolism, and amino acid metabolism (Fig. 2B), consistent with previous reports showing differential regulation of amino acid metabolism under ethanol stress [16].

Fig. 2.

Transcriptomic data analysis of E. coli under 1,6-HDO stress. (A) Volcano plot of DEGs under 1,6-HDO stress. (B) KEGG enrichment analysis of DEGs. (C) The classification analysis of GO pathways.

To identify regulatory pathways involved in the response to 1,6-HDO, GO functional annotation analysis classified the DEGs into 21 biological process terms, 5 cellular component terms, and 10 molecular function terms (Fig. 2C). The enriched terms could be broadly grouped into three major functional categories. First, multiple terms related to energy metabolism, including the TCA cycle, glyoxylate pathway, and ATP-associated activities, were significantly enriched, suggesting extensive metabolic reprogramming under stress conditions. Second, several terms associated with bacterial flagella, secretion systems, pore complexes, and ion transport were enriched, indicating substantial remodeling of cellular structures and membrane-associated functions. Third, enrichment of DNA binding, unfolded protein binding, and ATP-dependent protein folding chaperones pointed to activation of protein quality control and stress adaptation mechanisms. Similar functional responses have been reported during microbial adaptation to ethanol and other alcohol stresses [7,9,[17], [18], [19], [20]]. Collectively, these findings indicate that cells respond to 1,6-HDO stress by coordinated regulation of energy-dependent processes, cellular structural adaptation, and proteostasis-related functions.

3.3. Functional analysis of DEGs

3.3.1. Energy metabolism

Cellular energy metabolism plays a central role in microbial adaptation to environmental stress [21]. Under 1,6-HDO stress, transcriptomic analysis revealed up-regulation of TCA cycle and glyoxylate shunt genes, including genes sucC, sucD, acnA, aceA, and aceB, which increased by 1.44-, 1.83-, 2.11-, 2.32-, and 2.99-fold, respectively (Fig. 3A). Similar enhancement of energy metabolism has been widely reported during adaptation to alcohol and organic solvent stresses [22]. Meanwhile, genes associated with ATP-consuming molecular functions were downregulated to reduce other cellular energy consumption. Consistent with these transcriptional changes, intracellular ATP levels increased from 0.79 μmol to 1.56 μmol as the 1,6-HDO concentration increased from 0 to 15 g/L (Fig. 3B). Together, these results suggest that 1,6-HDO stress is accompanied by metabolic reprogramming centered on the TCA cycle and glyoxylate shunt, which may contribute to maintaining cellular energy homeostasis during stress adaptation [23,24].

Fig. 3.

Schematic illustration and cellular responses of E. coli KA-12 (AG1) under 1,6-HDO inhibition. (A) Schematic diagram of the inhibitory mechanism of 1,6-HDO. (B) Intracellular ATP levels. (C) Intracellular ROS levels. (D) Effect of exogenous amino acids on cell growth. Data are presented as mean ± SD from three independent biological replicates, *p < 0.05 vs. control group.

3.3.2. Sulfur metabolism

Transcriptomic analysis revealed significant enrichment of sulfur-related GO terms, including cellular response to sulfur starvation, sulfur compound metabolic process, and iron-sulfur cluster assembly (Fig. 2C). Specifically, the transcriptional levels of ssuA, ssuC, tauA, tauB, and tauC were up-regulated by 2.70-, 2.04-, 1.90-, 2.75-, and 3.96-fold, respectively (Fig. 3A). In E. coli, the ssuABC and tauABC operons encode transport systems for alkanesulfonates and taurine [25], which contribute to sulfur assimilation and the biosynthesis of sulfur-containing compounds involved in oxidative stress defense [26]. To further evaluate oxidative stress under 1,6-HDO exposure, intracellular ROS levels were measured. As shown in Fig. 3C, ROS levels remained relatively stable from 0 to 10 g/L 1,6-HDO, but increased significantly to 2.99-fold of the control at 15 g/L. These results indicated that up-regulation of sulfur metabolism genes may help maintain ROS homeostasis under moderate stress, whereas excessive ROS accumulation occurs at higher 1,6-HDO concentrations. Similar ROS-mediated toxicity has also been reported for other alcohol stresses [27,28].

3.3.3. Amino acid metabolism

In contrast, genes involved in arginine, valine, leucine, and isoleucine were significantly down-regulated under 1,6-HDO stress. Specifically, the argABCDEGH, ilvBDEI, and leuBC operons showed transcriptional decreases of approximately 1- to 3-fold (Fig. 3A). Similar repression of amino acid biosynthesis pathways has been reported in E. coli under n-butanol and other environmental stresses [4,29]. To evaluate whether amino acid limitation contributes to growth inhibition, arginine, valine, leucine, and glutamate were individually supplemented under 15 g/L 1,6-HDO stress. However, none of the supplements significantly improve cell growth, providing only marginal alleviation of growth inhibition (Fig. 3D). These results suggest that the effects of 1,6-HDO cannot be readily compensated by supplementation of single or a limited number of amino acids and may instead reflect broader metabolic perturbations induced by stress.

Together, 1,6-HDO stress triggers a coordinated stress response in E. coli primarily involving oxidative stress, metabolic reprogramming, and motility regulation. Specifically, 1,6-HDO stress was associated with activation of sulfur metabolism and increased intracellular ROS levels, suggesting that oxidative stress is an important component of its toxicity (Fig. 3A and C). Concurrently, cells exhibit activation of the TCA cycle and glyoxylate shunt together with elevated intracellular ATP levels (Fig. 3B), suggesting that 1,6-HDO stress induces extensive metabolic reprogramming and alters cellular energy metabolism. In parallel, multiple amino acid biosynthesis pathways were down-regulated, suggesting perturbation of amino acid metabolism under stress conditions. In addition, flagellar assembly and chemotaxis-related genes were up-regulated, implying a potential motility-based adaptive strategy in response to adverse environmental conditions.

3.4. Screening and functional analysis of a tolerance gene element library

3.4.1. Screening of the tolerance gene element library

To identify genes associated with 1,6-HDO tolerance, the ASKA overexpression library was screened under 15 g/L 1,6-HDO (Fig. 4A). Strains exhibiting OD_{600, max} values above that of the control strain (red dashed line) were considered to show improved 1,6-HDO tolerance (Fig. 4B). Among 4123 strains, 93 genes were identified to enhance cell growth under 15 g/L 1,6-HDO stress (Fig. 4C). The ten genes with the most prominent growth promoting effects were pfkA, yfbU, sodA, iadA, paaF, yieK, ndk, elaB, gcvR, and astD. These genes can be broadly classified into several functional groups, including central carbon and energy metabolism (pfkA, ndk, astD, paaF), oxidative stress defense (sodA, yieK, elaB), macromolecular maintenance and repair (iadA, yfbU), and transcriptional regulation (gcvR). These genes fall into functional categories broadly matching the transcriptomic profile, with gcvR overexpression also proving beneficial, marking it as a candidate for engineering.

Fig. 4.

Screening and validation of tolerance-conferring gene elements in E. coli KA-12 (AG1) under 1,6-HDO stress. (A) Schematic workflow of ASKA whole-genome library screening under 15 g/L 1,6-HDO stress. (B) Tolerance gene elements obtained after screening, with strains above the red line selected as candidate tolerance gene elements. (C) Validation of the tolerance performance of candidate tolerance gene elements via shake-flask rescreening. (D) Venn diagram of overlapping genes between the ASKA tolerance library (93 genes) and the transcriptomic DEG dataset (779 genes). Data are presented as mean ± SD from three independent biological replicates. Statistical analyses were performed for all groups, most groups showed significant differences compared with the control (*p < 0.05, **p < 0.01, ***p < 0.001), whereas only a few groups were not statistically different (p > 0.05).

To further investigate the relationship between the ASKA screening results and the transcriptional response to 1,6-HDO, we performed an overlap analysis between the 93 tolerance-conferring genes and the 779 DEGs identified from the transcriptome. A total of 17 overlapping genes (Fig. 4D), and their functional classifications are summarized in Table S4. These genes were also mainly associated with oxidative stress response, energy and carbon metabolism, amino acid metabolism, and iron homeostasis, consistent with the major functional categories identified in both the transcriptomic and screening analyses. Notably, the overlapping genes included both stress-induced genes and genes that were transcriptionally repressed under stress but whose overexpression still enhanced tolerance. This observation suggests that beneficial tolerance functions are not always activated during the native response to 1,6-HDO stress, indicating additional opportunities for strain improvement beyond the endogenous stress response.

3.4.2. Functional clustering analysis of tolerance gene elements

To elucidate the cellular tolerance mechanisms toward 1,6-HDO, functional clustering analysis was performed on 93 identified tolerance-related genes. These genes are mainly associated with metabolism, transport systems, transcriptional regulation, DNA/RNA repair, and protein synthesis (Fig. 5).

Fig. 5.

Schematic diagram of the major metabolic regulation mediated by tolerance gene elements under 1,6-HDO stress.

Among them, metabolism-related genes represented the largest category and were therefore analyzed in detail (Table 1) [[30], [31], [32]]. Overexpression of aceA, sucB, atpA, pfkA, purA, fruK, and dgoK improved cell growth under 1,6-HDO stress by 0.66- to 3.14-fold compared with the control strain (Fig. 4C). These genes are mainly involved in central carbon metabolism and energy generation, including the glyoxylate shunt (aceA), TCA cycle (sucB), ATP synthesis (atpA), and carbohydrate utilization (pfkA, fruK, and dgoK) [30,31,[33], [34], [35], [36], [37]]. Similar enhancement of central metabolism has also been reported to improve bacterial tolerance to alcohol stresses [38]. Together, these results indicate that reinforcement of energy metabolism contributes substantially to adaptation under 1,6-HDO stress.

Table 1.

Functional clustering of tolerance gene elements.

Functional category	Subcategory	Tolerance gene elements
Metabolism	Energy metabolism	atpA, purA, pfkA, aceA, sucB, eda
	Amino acid metabolism	astD, glnB, argC, argT, hisS
	Carbohydrate metabolism	fruK, dgoK
Transport	Amino acid transport	livJ
	Ion transport	nikA, nikR
Transcriptional regulation	Transcriptional regulators	nikR, gcvR, envR
DNA/RNA	DNA/RNA synthesis and repair	hchA, gyrB, nrdD, recT, rumA, ndk
Protein	Protein synthesis and repair	yhbo, yajL, ppiA

Amino acid metabolism-related genes represented another important category among the identified tolerance elements. Overexpression of astD, glnB, argC, and livJ improved growth under 1,6-HDO stress by 1.40- to 2.18-fold compared with the control strain (Fig. 4C). These genes are involved in arginine metabolism (astD and argC), amino acid utilization (glnB), and branched-chain amino acids transport (livJ) [[39], [40], [41]]. Notably, although exogenous amino acid supplementation showed limited effects on growth recovery (Fig. 3D), reinforcing intracellular amino acid metabolism through genetic regulation significantly improved tolerance. This suggests that regulation of intracellular amino acid metabolism may play a more important role than simple amino acid supplementation under 1,6-HDO stress. Similar enhancement of solvent tolerance through modulation of amino acid metabolism has also been reported in E. coli [42].

Consistent with the transcriptomic observation of altered metal ion transport, overexpression of nikA and nikR, which are involved in nickel ion uptake and regulation through the NikABC system [[43], [44], [45], [46]], significantly improved growth under 1,6-HDO stress by 0.68- and 2.30-fold, respectively (Fig. 4C). Similarly, enhanced expression of metal homeostasis-related genes has been reported to improve bacterial tolerance to organic solvent stresses [47]. These findings suggest that regulation of metal ion homeostasis may contribute to 1,6-HDO adaptation.

Several tolerance genes associated with DNA/RNA maintenance and protein quality control, including hchA, gyrB, nrdD, recT, rumA, and ndk improved cell growth under 1,6-HDO stress by 1.00- to 2.35-fold when overexpressed (Fig. 4C). These genes are mainly involved in genome maintenance, RNA modification, as well as protein folding and quality control [[48], [49], [50], [51], [52], [53]], suggesting that maintenance of macromolecular integrity may contribute to cellular adaptation under stress conditions. Similar roles of DNA/RNA maintenance and protein quality control systems have been reported in bacterial adaptation to environmental stresses [54,55].

Integration of transcriptomic profiling and tolerance gene screening suggests that 1,6-HDO tolerance in E. coli is mediated by a network involving metabolic adjustment, redox regulation, and macromolecular maintenance. The identified tolerance determinants were mainly distributed in these functional modules. In response, enhanced activity of central carbon metabolism, including the TCA cycle and glyoxylate shunt, together with the enrichment of energy metabolism-related tolerance genes, suggests that metabolic adaptation and energy supply may contribute to maintaining cellular growth. Meanwhile, sulfur metabolism-associated responses and oxidative stress-related genes indicate a potential role of redox regulation in mitigating 1,6-HDO-induced stress. Furthermore, DNA/RNA maintenance and protein quality control-related genes identified from the screening also suggest the importance of macromolecular maintenance during 1,6-HDO stress.

3.5. Effects of tolerance gene element combinations on strain tolerance

To further enhance 1,6-HDO tolerance, gene combinations were constructed based on the identified single-gene hits and different functional categories. Among the 93 tolerance genes identified in E. coli KA-12 (AG1), genes that also showed positive effects in an independent screening performed in E. coli NT1003 were selected for combinatorial analysis. A total of 11 dual-gene co-expression strains were constructed based on two design strategies: functional complementation (nikR-yieK and maoP-cof) and pathway synergy (rumA-yhbO, hchA-dicC, gcvR-pflD, livJ-fucI, iadA-ybdZ, fucI-bglA, bglA-ybdZ, yebB-agaS, and ybdZ-fucI). However, most of the combination strains barely grew in the stress medium, suggesting that dual-gene co-expression may impose an additional metabolic burden on the cells, leading to severely compromised cell growth performance. Only rumA and yhbO combination showed positive effects, reaching a biomass of 9.43 in the 1,6-HDO stress medium, which was 2.61-fold that of the control strain (Fig. 6A). The biomass of this dual-gene co-expression strain was 1.12-fold that of the strain overexpressing rumA alone, and 1.23-fold that of the strain overexpressing yhbO alone (Fig. 6B). RumA and YhbO are associated with translational fidelity and protein quality control, respectively. Previous studies have shown that enhancing function and protein quality control systems can significantly improve bacterial tolerance to oxidative and solvent stresses by reducing stress-induced translational errors and protein damage [56]. YhbO has been reported as a general stress-responsive protein in E. coli, contributing to cellular protection against multiple environmental stresses [57]. Meanwhile, oxidative stress can impair ribosome function, and RumA [58], as an rRNA methyltransferase involved in ribosome modification, may contribute to maintaining ribosome stability and translational fidelity under adverse conditions. Therefore, the enhanced tolerance conferred by RumA-YhbO co-expression may result from complementary protection of protein homeostasis at both translational and post-translational levels under 1,6-HDO stress [59,60].

Fig. 6.

Screening of recombinant strains harboring combinations of tolerance gene elements and comparison of their tolerance performance with single gene elements. (A) Screening of recombinant strains harboring combinations of tolerance gene elements. (B) Comparison of tolerance performance between combinations and single tolerance gene elements. Data are presented as mean ± SD from three independent biological replicates, *p < 0.05, **p < 0.01 vs. the control group, in Fig. 6A, all experimental groups showed **p < 0.01 vs. the control group.

4. Conclusion

This study systematically characterized the cellular response and tolerance determinants of E. coli under 1,6-HDO stress through transcriptomic analysis and genome-wide functional screening. Transcriptomic results revealed that 1,6-HDO stress induced significant metabolic reprogramming, characterized by activation of the TCA cycle, glyoxylate cycle, and sulfur metabolism, accompanied by increased ATP and ROS levels, while amino acid synthesis and metabolism were suppressed. And ASKA library screening identified tolerance-conferring genes involved in largely overlapping functional categories, supporting the relevance of these pathways in stress adaptation.

Integration of transcriptomic and screening datasets further revealed that stress-responsive pathways and tolerance determinants partially overlap but are not completely identical. This highlights that some beneficial functions are not fully activated during the native stress response and therefore represent potential targets for rational tolerance engineering. Furthermore, the successful enhancement of tolerance through RumA-YhbO co-expression suggests that coordinated protection of translation and protein homeostasis represents an additional strategy for improving cellular robustness under solvent stress. Overall, this work provides both mechanistic insights and a practical strategy for engineering robust 1,6-HDO-producing strains.

CRediT authorship contribution statement

Xinyi Zhang: Data curation, Methodology, Writing – original draft. Lisha Qin: Data curation, Methodology, Writing – original draft. Hongxu Li: Data curation. Xin Wang: Methodology, Resources, Writing – review & editing. Kequan Chen: Resources, Supervision.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following is the Supplementary data to this article: