We used 10°C as the experimental condition to match our previous study (Hao et al., 2024). As 25°C is the optimal growth temperature of strain RCBS9, it was used as the control condition to fully understand the low-temperature tolerance mechanisms.

Strain RCBS9 (which was preserved in our laboratory) was streaked on LB agar plates and placed in a 25°C incubator to activate it. The activated strain was incubated at 10 or 25°C in liquid LB medium or MSM medium (containing minimal nutrients) supplemented with 10 mg/L E2. The experiments in Sections 2.2–2.6 and 2.9 were conducted in both LB and MSM+E2 media. The experiments in Sections 2.7, 2.8, 2.10, and 2.11 were conducted only in LB medium. The medium composition was as described by Hao et al. (2024).

The change in cell membrane permeability of strain RCSB9 in response to a reduction in temperature was determined by two methods: crystal violet staining (Devi et al., 2013; Herndon et al., 2020) and conductivity measurements (Kong et al., 2024). The strain RCBS9 was cultured to the logarithmic growth phase at 25°C and then incubated at 10°C for 2, 24, 48, and 72 h. Samples were collected at each time point to be used in both experiments. For the crystal violet assays, strain RCBS9 samples were mixed with crystal violet (Aladdin Biochemical Technology Co. Ltd., Shanghai, China), incubated at 37°C for 10 min, and centrifuged (Eppendorf, Germany) at 13,500 rpm, and the optical density (OD) at 590 nm of the supernatant was measured using a microplate reader (Thermo Fisher, Swedish). For the conductivity assays, strain RCBS9 samples were centrifuged at 8,000 rpm for 5 min, and the conductivity of the supernatants was then measured using a conductivity meter (Shanghai Yidian Scientific Instrument Co., Ltd., China).

The ROS levels and DNA damage of strain RCBS9 at 10 and 25°C were quantified using a ROS detection kit (Beyotime Biotechnology) and an 8-OHdG ELISA kit (Sangon Biotech), respectively. The results were determined using a fluorescence microplate (Shimadzu, Japan) and measured using a microplate reader (Thermo Fisher, Swedish), respectively.

SOD activity in strain RCBS9 at 10 or 25°C was determined using a SOD assay kit (Nanjing Jiancheng Bioengineering Institute). Total protein in strain RCBS9 was quantified using a bicinchoninic acid (BCA) protein content assay kit (Beyotime Biotechnology). SOD activity was then calculated using the formula provided in the SOD assay kit. The results were determined using a measured using a microplate reader (Thermo Fisher, Swedish).

Carotenoid content was assessed as described previously (Flores et al., 2020; Liu et al., 2020). RCBS9 suspension was transferred to a brown centrifuge tube (protected from light), mixed with lysozyme, incubated at 37°C for 30 min to lyse the bacteria, mixed with acetone for 5 min, and centrifuged (Eppendorf, Germany). The values at OD480 nm of the supernatant was assessed using a microplate reader (Thermo Fisher, Swedish). The carotenoid content (C) was calculated according to the following formula (the extinction coefficient ε1cm1% was set at 1,600 g%⁻¹ cm⁻¹): C (ug/ml)=(A480×106)/(e1cm1% ×100).

Biofilm content was assessed as described previously (Yan and Xie, 2020). The cultured strain RCBS9 was placed in 12-well plates and incubated at 10 or 25°C for 48, 72, 96, or 120 h. The planktonic cells in the wells were aspirated and measured at OD600 nm using a microplate reader (Thermo Fisher, Swedish). Planktonic cells remaining in the wells were rinsed away with phosphate-buffered saline and the wells were air-dried to fix the adherent cells in the wells. Subsequently, 2 mL of 0.1 mg/mL crystal violet was added for 20 min to ensure complete attachment to the biofilm on the well surface. After rinsing away the excess dye and air-drying for fixation, 2 mL of 95% ethanol was added to solubilize the biofilm, and then the biofilm was quantified at OD590nm using a microplate reader (Thermo Fisher, Swedish). The relative biofilm content was determined based on the ratio of biofilm content to bacterial density.

Strain RCBS9 was cultured in LB medium at 10 or 25°C and processed according to the standard protocols of the MIDI/Hewlett-Packard Microbial Identification System (Keystone Laboratories, Edmonton, Canada; Song et al., 2008; Wang et al., 2022a). The fatty acid species and their proportions in the cell membrane were analyzed by gas chromatography-mass spectrometry (GC-MS) using an Agilent 7890B-5977B system (Agilent Technologies, Inc.).

Strain RCBS9 was cultured in LB medium at 10 or 25°C and stored at −80°C. The bacterial samples were then sequenced by Beijing Novozymes Technology Co. A probe was used to remove ribosomal RNA (rRNA) from the total RNA to enrich the mRNA. Standardization and statistical modeling were applied to calculate the p-values and then adjust for multiple testing (controlling the false discovery rate). Genes with fold change ≥2 and p_adj < 0.05 were classified as upregulated or downregulated differentially expressed genes (DEGs). ClusterProfiler software was used for Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses of the DEGs. String protein interaction database and cytoscape software were used for Protein-protein interaction analysis of DEGs from transcriptome analysis results.

The accuracy of the transcriptome data was verified by RT-qPCR (Zhou et al., 2022). The primer sequences (Supplementary Table 1) for selected genes were designed using SnapGene software. Samples of strain RCBS9 were prepared in LB and MSM+E2 media at 10 or 25°C, as previously described (Hao et al., 2023). Sample RNA was extracted using a bacteria total RNA isolation kit (Sangon Biotech). 16S RNA expression served as a calibration reference. Amplification was detected using a LightCycler^® 96 system. The RT-qPCR data were analyzed using the 2^−ΔΔCT method to determine fold changes in gene expression.

Primers for six target genes in strain RCBS9 related to survival at low temperature (identified in the transcriptome analysis), i.e., sHsps, DPS, GroEL, USP-1, Cu/Zn-SOD, and USP-2, were designed by Sangon Biotech (Shanghai; Supplementary Table 2). The genes were amplified by PCR and the PCR products were purified using a DiaSpin DNA Gel Extraction Kit (B110092-0100; Sangon Biotech) before being cloned into a pET-28a(+) plasmid. The constructed plasmids and empty plasmid were then transformed into E. coli BL21 (DE3) cells (Deng et al., 2014; Li R. et al., 2023).

Proteins in the abovementioned recombinant E. coli were analyzed by SDS-PAGE. Recombinant E. coli (BL21 pET-28a–target gene) and the control (BL21 pET-28a) were induced with identical concentrations of isopropyl β-D-thiogalactopyranoside (IPTG) at 16°C for 17 h to overexpress the fusion proteins (His–target protein) or His, respectively. Protein expression was then assessed by SDS-PAGE.

To monitor growth, after induction, the recombinant E. coli were resuspended in fresh LB medium, and the OD600 nm of the bacterial solution was adjusted to 1.0. The culture was then incubated at 10°C for 0, 4, 6, 8, and 24 h, with measurements of bacterial growth (OD600 nm) recorded at each time point (Xikeranmu et al., 2020).

Data were collated using Excel 2019 and analyzed (plotting graphs and conducting t-tests and one-way ANOVA) using GraphPad Prism 8.0. All experiments were performed in triplicate, and the results were expressed as means with standard deviations (ns, not statistically different; ^*P < 0.05, ^**P < 0.01, ^***P < 0.001, ^****P < 0.0001).

Although strain RCBS9 grew and degraded >70% of E2 at 10°C, its growth and degradation efficiency were significantly decreased at this low temperature compared to at 25°C (Hao et al., 2024). To elucidate the reasons for the decreases at low temperature, we measured the levels of ROS and 8-OHdG [a marker of ROS-induced DNA damage (Shan et al., 2020)] in the strain. The results showed that ROS and 8-OHdG were significantly higher at 10°C than 25°C (Figures 1A, B). At low temperatures, the increased oxygen solubility leads to ROS accumulation, causing oxidative damage to cellular biomolecules, with ROS levels serving as an indicator of bacterial stress (Irshad et al., 2021; Zou and Cheng, 2024). For instance, ROS levels have been used to assess the stress response to Fe₃O₄ nanoparticles in anaerobic ammonium oxidation microbial communities (Zhang et al., 2024). The experimental results showed that strain RCBS9 was subjected to oxidative stress at low temperature and the DNA was attacked by ROS, causing genetic damage and thus reducing the growth rate and E2 degradation capacity of strain RCBS9 at low temperature. Additionally, ROS and 8-OHdG levels were higher in MSM+E2 medium than in LB medium, likely due to the additional oxidative stress due to E2 (Chainy and Sahoo, 2020) and the nutrient-poor environment in MSM+E2 medium (Husain et al., 2022).

This study aimed to examine whether strain RCBS9 withstands low temperatures by producing SOD and carotenoids. The experimental results showed that Figures 2A, B, strain RCBS9 in LB medium had significantly higher SOD activity and carotenoid content at 10°C than at 25°C. Strain RCBS9 can scavenge ROS accumulated at low temperatures by increasing SOD activity and carotenoid content (Chen et al., 2023; Shi et al., 2022; Yu Z. et al., 2023), thus avoiding excessive damage or even death of the bacterium caused by high ROS levels and ensuring its growth and metabolism under low temperature conditions. Carotenoids also increase cell membrane stability (Chia et al., 2021), which is another assurance that strain RCBS9 can survive at low temperature. This phenomenon has also been observed in other cryotolerant bacteria (Flores et al., 2020; Li et al., 2022). In MSM+E2 medium, there was also higher SOD activity and carotenoid content at 10°C than at 25°C, but the differences were not significant (Figures 1C, D). The presence of E2 induces oxidative stress, which may alter intracellular SOD activity and carotenoid content, so additional low-temperature stress did not significantly modulate them. Taking the results of the two experiments together, under more severe conditions (low temperature and MSM+E2 medium), strain RCBS9 preferred to produce SOD rather than carotenoids to counteract low-temperature stress.

Low temperature affects the fluidity of the phospholipid bilayer, which is a part of the bacterial cell membrane, and membrane protein activity (Li et al., 2020). This affects cell membrane permeability and inhibits bacterial growth and reproduction. Therefore, the ability to maintain cell membrane function at low temperature is crucial for cold-tolerant bacteria. To examine the functional status of the cell membrane of strain RCBS9 under low-temperature stress, the cell membrane permeability was assessed.

First, crystal violet staining showed that the cell membrane permeability of strain RCBS9 in LB medium was significantly reduced after 2 h at 10°C compared to at 25°C, which was consistent with findings regarding other bacteria (Li et al., 2020; Wang et al., 2022a). The permeability gradually increased over time at 10°C, which may reflect a gradual recovery of the cell membrane function. In MSM+E2 medium, the cell membrane permeability was increased at after 2 h at 10°C compared to 25°C. We hypothesize that because E2 was the only carbon source, the cell membrane construction in strain RCBS9 was incomplete. This phenomenon has also been observed in other bacteria (Salton, 1967). The subsequent decrease in permeability over time may be due to strain RCBS9 restoring the cell membrane structure and function as it adapted to its environment.

Second, conductivity assays were carried out (in both LB and MSM+E2 media) during the bacterial growth plateau, to avoid errors due to increased bacterial counts. The results followed similar trends to those of the crystal violet assays. The only significant difference between the two experimental trends was that the conductivity of the bacterial fluid at 72 h in LB medium was significantly higher at 10°C than at 25°C, which may be due to the accumulation of metabolic by-products from prolonged incubation in a nutrient-rich environment.

Overall, the changes in cell membrane permeability of strain RCBS9 are a combination of the direct effects of low temperature on the cell membrane and the restoration of bacterial adaptive capacity over time. The recovery of cell membrane function over time is crucial for the survival of strain RCBS9, but the specific mechanisms have yet to be studied in depth.

A bacterial biofilm is a structured community of bacteria that adheres to a solid surface by secreting extracellular polymeric substances. These substances provide a physical protective barrier for the bacteria, as well as trapping nutrients while eliminating waste (Camba et al., 2024; Kuttel et al., 2017; Yan and Xie, 2020). We investigated whether the relative biofilm content of strain RCBS9 (defined as the ratio of biofilm content to the value at OD600 nm for bacterial solution) increased at low temperature. The relative biofilm content in LB medium was consistently higher at 10°C than at 25°C, suggesting that strain RCBS9 resists low temperature by producing biofilm. Notably, the relative biofilm content was significantly higher in MSM+E2 medium than LB medium. This phenomenon should indicate that, in nutrient-limitation stress (at 25°C in MSM), strain RCBS9 formed biofilms to capture carbon-E2 (Du et al., 2022). Additional low-temperature stress (at 10°C in MSM) did not increase the relative biofilm content compared to that at 25°C in MSM. Instead, the relative biofilm content decreased, which may be attributable to overwhelming stress (low-temperature stress plus nutrient-limitation stress; Figures 2C–E). In conclusion, the results demonstrated that biofilm production by strain RCBS9 is an effective adaptation strategy to low temperature and for capturing E2.

The aforementioned assays indicate that strain RCBS9 can undergo various adaptive changes to tolerate low temperature. Additionally, there was a degree of overlap in the strain's responses to low temperature, E2, and nutritional deficiency. When analyzing the strain's low-temperature tolerance mechanisms, the combination of multiple stress factors may lead to non-significant or even opposite-direction assay results. This could interfere with our comprehensive analysis of the low-temperature adaptive mechanisms of strain RCBS9. In order to gain a comprehensive understanding of the low-temperature response strategy of strain RCBS9 and to screen for effective low-temperature-tolerant protein, subsequent experiments were performed in LB medium with low temperature as the only variable to conduct in-depth fatty acid content analysis and transcriptome analysis.

Fatty acids are integral components of cell membranes, and alterations in the types and proportions impact the fluidity and permeability of cell membranes (Yang et al., 2020). To determine the mechanisms underlying the cell membrane permeability changes in strain RCBS9 at low temperature and the adaptive measures employed by strain RCBS9 to regulate the composition of its cell membranes at low temperature, we investigated the fatty acid types and proportions in strain RCBS9 in LB medium at low temperature. The number of fatty acid species decreased from 19 at 25°C to 14 at 10°C. The content and proportions of saturated fatty acids also decreased at 10°C compared to 25°C. However, the overall fatty acid proportion increased by 10.39%, and the content and proportions of unsaturated fatty acids strongly increased, particularly regarding methyl (E)-9-octadecenoate, which increased by a factor of 11 (Figure 3).

At low temperature, the changes in fatty acids in strain RCBS9 indicated increases in the synthesis of long-chain unsaturated fatty acids. Similar changes have been observed in several bacterial strains (Almeyda et al., 2020; Wang Y. et al., 2020). The ability of unsaturated fatty acids (compared to saturated fatty acids) to maintain superior cell membrane fluidity at low temperatures (Ballweg et al., 2020) is likely to contribute to the gradual and limited increase in the cell membrane permeability of strain RCBS9. Furthermore, studies have demonstrated increased unsaturated fatty acid content in Lactobacillus plantarum LIP-1 under acid stress and Listeria monocytogenes under nutrient stress (E et al., 2021; Wang et al., 2023), which represents an effective bacterial adaptation to harsh environments. The capacity of strain RCBS9 to undergo adaptive alterations in cell membrane composition, thereby restoring cell membrane permeability at low temperature, may represent a pivotal factor influencing its capacity to uptake carbon sources at low temperature.

To gain a better understanding into how strain RCBS9 adapts to low temperature at a molecular level, we conducted transcriptome sequencing. Figure 4 shows a correlation plot of gene expression and Figure 5 shows a principal component analysis plot of gene expression, which confirm the reliability of the transcriptome data.

The volcano plot in Figure 4C reveals that there were 2,012 upregulated and 1,926 downregulated genes at 10 vs. 25°C. GO annotation of the DEGs (Figure 5A) showed that genes related to organic acid metabolism and ribosomes exhibited the most significant differences. Most of the upregulated genes were associated with localized transport and energy production, whereas most of the downregulated genes were linked to biomolecule synthesis. KEGG annotation of the DEGs (Figure 5B) indicated that genes related to secondary metabolite synthesis, metabolic pathways, transporters, two-component systems, and ribosomes exhibited the most significant differences. Most of the upregulated genes were associated with ABC transporter proteins, amino acid metabolism, and fatty acid degradation, whereas most of the downregulated genes were associated with ribosomes. These results suggest that strain RCBS9 requires more energy and transporter protein production to survive at 10°C compared to 25°C. The strain also reduces unnecessary protein synthesis.

The following sections discuss the transcription of cold adaptation-related genes at 10 vs. 25°C to provide insights into the regulation of the various stress response systems (Bao et al., 2023; Kloska et al., 2020; Pátek et al., 2021; Yan and Xie, 2020) of strain RCBS9 at low temperature.

To verify the transcriptome results, RT-qPCR was performed to assess the transcription of strain RCBS9 in LB and MSM+E2 media. The RT-qPCR results of strain RCBS9 in the LB medium were completely consistent with the transcriptome results (Figure 6), which validated the transcriptome results. However, RT-qPCR results for two genes in MSM+E2 medium were inconsistent with the transcriptome results (Supplementary Figure 2). Combined with the results of the physiological assays described above, it can be assumed that the mechanisms of low-temperature tolerance of strain RCBS9 are similar under different conditions (LB medium vs. MSM+E2 medium), but they are appropriately adjusted to the specific environment.

Six target proteins related to survival at low temperature (according to the transcriptome results of strain RCBS9) were successfully expressed in recombinant E. coli BL21. The five SDS-PAGE lanes for these six proteins (from left to right) correspond to (1) marker, (2) induced control (BL21 PET28a), (3) uninduced control (BL21 PET28a), (4) induced BL21 pET-28a–target protein, and (5) uninduced BL21 pET-28a–target protein. The results demonstrated that all six proteins were successfully induced and in the expected positions (Figure 7).

The growth curves of each recombinant E. coli expressing the six target proteins were significantly different from that of the control (BL21-PET28a). The bacterial count of the control (BL21-PET28a) at 10°C was almost unchanged, with a value of about 1-1.1 at OD600 nm. However, the number of recombinant E. coli expressing the target proteins reached the peak at 4 h, and the value at OD600 nm of BL21-DPS, BL21-GroEL, and BL21-USP-2 reached about 1.4, and then began to decline. After 8 h, except for BL21-SHsps, the values of bacterial mass at OD600 nm were around 1.1 for all strains (Figure 8). Based on the results, we speculate that the induced target proteins improve low-temperature adaptation when they exist in large quantities in the recombinant E. coli. After 8 h, due to damage and degradation of the proteins, the recombinant E. coli could not produce a large amount of target proteins without additional IPTG induction, and the bacterial counts began to decline until they were comparable to those of the control (BL21-PET28a).

In addition to the protective function of the target proteins themselves, the target proteins may produce cascade reactions that increase other protective substances or improve the cellular structure for better survival at low temperature. More in-depth validation analyses of the reasons for the large decrease in the BL21-sHsps bacterial count at 6 h and the regulatory mechanisms underlying the low-temperature tolerance of strain RCBS9 are required. Thereafter, based on the current work, we will further investigate the mechanisms underlying the low-temperature tolerance of strain RCBS9.