All soil samples were collected near the Yangzi Petrochemical Plant (Nanjing, China), and the resultant bacterial isolates were deposited into the laboratory’s collection. Standard monosaccharides and trifluoroacetic acid (TFA) were purchased from Sigma Chemicals Co., Ltd. (St. Louis, United States). All other chemicals and reagents were of reagent grade and purchased from the China National Pharmaceutical Group.

The soil samples were suspended at 20% (w/v) in sterilized water to isolate the EPS-producing bacteria, and then the suspension was filtered to remove the soil. The filtrate was added to the Nutrient Broth Medium (Toupu Biol-engineering Co., Ltd., Zhaoyuan, China) at a ratio of 1: 9 and incubated at 30°C for 48 h. The culture was then diluted (10⁻¹ to 10⁻⁶) with sterilized water. From each dilution, 200 μL was plated onto solid nutrient agar plates and incubated at 30°C for 48 h. Single colonies were selected for Gram staining and microscope (XS-212-202, Japan) observation and inoculated into the fermentation medium. The fermentation medium (3.0 g/L beef extract, 10.0 g/L peptone, 2.0 g/L yeast extract, 5.0 g/L NaCl, pH 7.4–7.6) was used to produce EPSs with shaking (250 rpm) for 72 h at 30°C. The strain with the highest EPSs yield was used as the test strain, and subsequent genetic identification of this strain based on 16S rDNA sequencing was performed by Paiseno Biological Co., LTD (Shanghai, China). The observed bacterial growth index was evaluated using dry cell weight (DCW) and EPSs yield. The DCW was computed from a curve of OD₆₀₀ with respect to dry weight. An OD₆₀₀ of 1.0 represented 400 mg dry weight/L. The EPSs yield was defined as EPSs production per unit of DCW.

Crude EPSs were isolated from the fermentation medium using improved methods reported by Sun et al. (2021). The culture broth was diluted with ultra-pure water, mixed with diatomite, and filtered by vacuum filtration. The filtrate was concentrated to 1/4 of the initial volume. Later, the concentrated supernatant was deproteinated by washing thrice with Sevag reagent (isoamyl alcohol: chloroform = 1:4, v/v). Before collection, the crude EPSs were precipitated with ethanol (1:5, v/v) at 4°C overnight. The EPSs yield was evaluated after lyophilization. EPSs purification was performed according to the previously described method (Sun et al., 2021). The crude EPSs were dissolved in deionized water, centrifuged (15,000 × g, 30 min), and the supernatant was applied to a DEAE-cellulose DE-32 column (16 mm × 40 cm) (Shanghai yuanye Bio-Technology Co Ltd., China) equilibrated with 0.1 M PBS buffer (pH 7.0). The column was sequentially eluted with 200 mL deionized water and 200 mL NaCl solutions at different concentrations of 0.3, 0.5, 0.7, 1.0 mol/L and a flow rate of 1 mL/min, respectively. The eluents (10 mL/tube) were monitored by the phenol-sulfuric acid method. The fractions obtained by 0.5 mol/L and 0.7 mol/L NaCl were collected and further purified by gel permeation chromatography on a Sephadex G-25 column (16 mm × 40 cm) using ultra-pure water as an eluent at a flow rate of 0.2 mL/min. The fractions were collected, concentrated and lyophilized, and then the pure polysaccharide was obtained.

The UV–Vis spectra, FTIR spectroscopy, average molecular weight, chemical composition, and uronic acid analysis of EPSs were performed according to the previously reported methods (Sun et al., 2021). The surface charge of a polysaccharide was determined using Zeta potential measurement according to the method reported by Varma and Kumar (2018). EPSs derivatization was performed following a previously reported method (Wang et al., 2018), and the operating conditions for monosaccharide composition analysis were set as mentioned in a previous method (Tayebeh et al., 2019). Monosaccharides were identified by comparing them with standard glucose, galactose, xylose, and N-acetylglucosamine.

Whole-cell synthesis of cadaveric amines was performed in a 1 L fermenter (Infors, Switzerland) with a conversion fluid volume of 0.3 L. The transformation broth contained 50 g/L L-lysine and 3 g DCW/L of engineered Escherichia coli DFC1001 (Guo et al., 2020) with or without EPSs. Approximately 100–300 mg of EPSs were added per gram of DCW E. coli DFC1001. Biotransformation was performed at 37°C, 300 rpm, and the pH was adjusted to 7.0 by the automatic addition of 2 M HCl. After 1 h of biotransformation, the surviving bacteria in the culture solution were calculated by plating on Luria-Bertani (LB) agar plates containing 50 μg/mL streptomycin (Guo et al., 2020), followed by serial dilution. The survival rate of engineered Escherichia coli DFC1001 was defined as the percentage of viable cells after and before whole-cell synthesis. The cadaverine yield was calculated by dividing the cadaverine concentration by the initial concentration of the substrate L-lysine (50 g/L). The culture solution was centrifuged at 4000 x g for 5 min and the concentration of cadaverine was determined by high-performance liquid chromatography. Cadaverine was separated using a YMC Carotenoid C30 column (4.6 × 250 mm, 5 μm). The values of flow rate, injection volume, and column temperature were set at 0.8 mL/min, 10 μL, and 35°C, respectively. 0.5% trifluoroacetic acid and 5% acetonitrile were used as mobile phases A and B, respectively.

Single-factor tests were employed to investigate the effects of fermentation conditions on bacterial growth and EPSs synthesis. The strains were inoculated in the fermentation medium and cultured in a shaker incubator. The range of fermentation conditions (influencing factors and levels) were selected as follows: pH (5.02, 5.86, 7.13, 8.19, 8.61), rotational speed (0, 50, 100, 150, 200, 250 rpm), temperature (10, 18, 24, 30, 37°C). Then, the preliminary range of synthetic variables, namely pH (X₁), rotational speed (X₂), and temperature (X₃), was determined based on single factor experiments. A three-factor-three level Box–Behnken design was employed using design export software to determine which combination of synthesis variables would yield the largest amount of dry cell weight (DCW) (Y₁) and EPSs field (Y₂).

About 54 single colonies with the EPSs producing ability were isolated from the soil samples. These 54 strains were cultured and fermented separately using the same medium and growth conditions. EPSs production by different strains was measured, as shown in Table 1. The strain numbered NJ 1023 showed the highest EPSs yield.

The 16S rDNA nucleotide sequence of strain NJ1023 was analyzed by BLAST software. Strain NJ1023 had 99.79% 16S rDNA similarity with its closest relative E. aerogenes B19. Therefore, strain NJ1023 was identified as an E. aerogenes and was submitted to the China Center for Type Culture Collection (CCTCC) (Wuhan, China) with an accession number of CCTCC M 20211243.

The crude EPSs were extracted from E. aerogenes NJ1023 by vacuum filtration, deproteinization, ethanol precipitation, and lyophilization. The dry sample was then separated by DEAE-cellulose DE-32, and one fraction was obtained (Figure 1A). The fraction eluted with 0.3 M NaCl solutions was further purified with a Sephadex G-25 size-exclusion chromatography column. The major elution fraction (Figure 1B) was collected. The EPSs produced by E. aerogenes NJ1023 contained approximately 92.44% total sugar, 0.41% protein, 3.40% glycuronic acid, 2.13% P, and 0.85% S. The absorption peaks were not found at 260 nm, indicating an absence of nucleic acids. A novel molecular weight of 2.7 × 10³ Da was calculated using known MW standards. Complete hydrolysis of the EPSs, followed by gas chromatography mass spectrometry (GC–MS) demonstrated that the polysaccharide was composed of D-xylose, D-glucose, D-galactose, and N-acetylglucosamine with a molar ratio of 0.27: 4.52: 1.74: 0.2. The molecular weight and monosaccharide composition of these EPSs significantly differed from the EPSs produced by other strains of the same genus. EPSs produced by other strains of Enterobacter spp. had a high molecular weight of approximately 1.0 × 10⁶ Da, and were mainly composed of fucose, galactose, and glucose (Isobe et al., 2001; Wang et al., 2013).

In the infrared wavelengths, the EPSs had the characteristics of polysaccharides (Figure 2). The sample showed the absorption peaks characteristic of polysaccharides at 3233, 2967, 1,650, 1,234, and 1,082 cm⁻¹. The peaks near 3,233 cm⁻¹ were attributed to the stretching vibration of O-H. Low intensity banding at 2967 cm⁻¹ was attributed to the C-H stretching and bending. The peak observed at 1650 cm⁻¹ was attributed to the C=O stretch of amide. The absorption band at 1234 cm⁻¹ was attributed to the sugar structure of O-H in the carboxyl group, indicating the presence of glycuronic acid in the polysaccharide (Guo et al., 2022).

Cadaverine, a common polyamine with important physiological functions, plays a crucial role in maintaining cell morphology and physiological functions (Igarashi and Kashiwagi, 2018). However, excessive polyamines can be toxic to cells, causing oxidative damage (Murray et al., 2018). Polysaccharides are often used as antioxidants to significantly reduce oxidation and protect cells (Li et al., 2020). This study aimed to investigate the role of EPSs from E. aerogenes NJ1023 in response to cadaverine toxicity in E. coli DFC1001 cells. As shown in Table 2, the addition of extracellular polysaccharides increased the conversion of lysine to cadaverine, which could be attributed to the increased survival rate of the engineered E. coli cells. Compared to the control without EPSs addition, the addition of 300 mg/g EPSs increased the survival rate of the strain by 57.27%.