The WT and engineered Syn7942 were grown at 37°C in BG11 liquid medium (pH 7.5) or on agar plates under a light intensity of approximately 100 μmol photons m⁻² s⁻¹ in an HNY-211B Illuminating incubator Shaker of 200 rpm in 100 ml flask (Honour, Tianjin, China). Twenty-five milligrams per liter of chloramphenicol (Solarbio, Beijing, China) was added during the culture of engineering strains. Different salt concentrations of media used in the culture process were obtained by adding a suitable amount of 3 M to 0 M NaCl BG11 medium. Cell density was measured at 750 nm (OD₇₅₀) by an ELx808 Absorbance Microplate Reader (BioTek, VT, United States). E. coli TOP10 was grown in LB liquid medium or LB agar plates, with 50 mg/l of chloramphenicol or spectinomycin for screening and maintaining the stability of engineered bacteria.

The strains used in this study are listed in Table 1. Among them, E. coli TOP10 was used for plasmid construction. pSI-ect vector with a chloramphenicol-resistant cassette was constructed for express ectoine production cassette based on laboratory plasmid (Li et al., 2018). To knock out the sucrose production pathway and express the ectoine synthesis pathway after the knockout of the sucrose production pathway, pSPS-ect was constructed by replacing the upstream and downstream homologous arms based on pSI-ect, and pSPS was built by deleting an ectoine expression cassette based on pSPS-ect. Primers and plasmids used for this study are listed in Supplementary Table S1. All primers were synthesized by Azenta (Suzhou, China). The genes ectA, ectB, and ectC were synthesized by Azenta based on the Halomonas elongata after codon optimization. The sequences are shown in Supplementary Table S2. The template plasmids were purified by FastPure Plasmid Mini Kit (Vazyme Biotech, Nanjing, China). The target fragments were amplified by Phanta Super-Fidelity DNA Polymerase (Vazyme Biotech, Nanjing, China) and purified by FastPure Gel DNA Extraction Mini Kit (Vazyme Biotech, Nanjing, China). Then, fragments were ligated by ClonExpress MultiS One Step Cloning Kit (Vazyme Biotech, Nanjing, China) or Golden Gate (Thermo Fisher Scientific Inc., CA, United States). All constructs were verified by PCR and Sanger sequencing. After the constructed plasmids were extracted and cut through NdeI (Thermo Fisher Scientific Inc., CA, United States), and were transformed into Syn7942 by the method of natural transformation.

One milligram liter of cultures on the 7th day was taken out and placed at −80°C. When preparing the samples, the cultures were taken out from −80°C, placed at 70°C for 2 h, centrifuged at 13,000 rpm. And the supernatant was taken for ectoine concentration determination. Based on the previous study (Ning et al., 2016), the ectoine was detected using high-performance liquid chromatography (Agilent 1,260 Series HPLC, Agilent Technologies, Santa Clara, CA, United States), and the concentration of ectoine in the samples was determined using the ectoine standard (Sigma-Aldrich, Shanghai, China). The chromatographic column was Ultimate AQ-C18, 5 μm 4.6*250 mm (Welch, China), and the mobile phase condition was ultrapure water of 0.6 ml/min. The ectoine was monitored using a UV detector at a wavelength of 210 nm.

WT and Syn7942/Δsps-ect were cultured in BG11 medium with 0 mM as control-1 and control-2 and 300 mM NaCl as experimental-1 and experimental-2, respectively. On the fourth day, samples were collected and sent to Azenta for transcriptome sequencing and data analysis. Each sample had three biological replicates. The log₂(fold-change) > 1.5 and the value of p <0.05 were set as the threshold for DEGs identification. KEGG enrichment analysis was done using TBtools (Version v1.098769; Chen C. et al., 2020). There was still a small number of reads for the Synpcc7942_0808 gene in Syn7942/Δsps-ect. In order to exclude data interference, the number of reads for the Synpcc7942_0808 gene in Syn7942/Δsps-ect was ignored.

The transcriptome data has been uploaded to GEO database (GSE222067).

The gene cluster ectABC was codon-optimized for Syn7942 and then has been chemically synthesized and introduced into the NSI site of Syn7942 afterwards. The expression is controlled by the strong promoter P_cpc560 (Figure 1). This engineered strain was named Syn7942/NSI-ect. To evaluate the salt tolerance of Syn7942/NSI-ect, WT was selected as the control strain, and the growth curves of Syn7942/NSI-ect and WT were measured in BG11 medium with addition of 0, 100, 200, 300, 400, and 500 mM NaCl, respectively. Under the growth condition of 0 mM NaCl, the OD₇₅₀ of Syn7942/NSI-ect was decreased by 19% compared with that of WT on the 7th day, which may be caused by the competition between ectoine synthesis pathway and endogenous amino acid metabolism pathway for L-aspartate-4-semialdehyde. And the growth of Syn7942/NSI-ect and WT was almost the same under 100 and 200 mM NaCl conditions, while the OD₇₅₀ of Syn7942/NSI-ect on the 7th day was increased by 20% than that of WT under 300 mM NaCl and increased by 80% under the condition of 400 mM NaCl (Figure 2A; Supplementary Figures S1A,B), demonstrating the improved salt tolerance of strain Syn7942/NSI-ect. Under 500 mM NaCl condition, the OD₇₅₀ of Syn7942/NSI-ect on the 7th day could reach 0.362 while WT could not survive at all. This suggested that the salt tolerance of Syn7942 could be improved by the introduction of ectoine biosynthesis pathway.

The production of ectoine in Syn7942/NSI-ect under different salt concentrations on the 7th day was then determined using the external standard method. The standard curve was shown in Supplementary Figure S2. As shown in Figure 2B, under the condition of 0 mM NaCl, the yield of ectoine was 49.8 mg/l in Syn7942/NSI-ect. Moreover, under 100 mM NaCl, the yield of ectoine was significantly increased, reaching 98.9 mg/l. It has also been reported that ectoine production increased with the concentration of NaCl in the environment (Ning et al., 2016; Yu et al., 2022), and this phenomenon also occurred in the synthesis of glucosylglycerol and sucrose (Song et al., 2016; Cui et al., 2021). With the further increase of salt concentration, the yield of ectoine decreased compared with under 100 mM NaCl, most probably due to the worse growth caused by salt stress.

In previous studies, the salt environment was reported to be favorable for the synthesis of ectoine (Ning et al., 2016). Here, our results showed that the synthesis of ectoine could significantly improve the salt tolerance of Syn7942/NSI-ect, and the amount of ectoine synthesis was responsive to the high-salt environment.

The accumulation of compatible solutes is an essential way for microorganisms to resist the high-salt environment, and the only compatible solute in Syn7942 is sucrose (Klähn and Hagemann, 2011). The loss of the sucrose synthesis pathway led to the generation of a salt-sensitive strain, which indicated that sucrose was the only compatible solute. However, sucrose is not considered an ideal osmocompatible substance against salt stress (Klähn and Hagemann, 2011). In a study of adaptations to salt tolerance in Synechocystis sp. PCC 6803, the accumulation of glucosylglycerol was much higher than sucrose, and the contribution of glucosylglycerol was much higher than that of sucrose during the salt adaptation process; therefore, in natural selection, sucrose was more critical as an energy storage substance rather than playing a role in salt tolerance (Klähn et al., 2021). As a compatible solute retained by natural evolution in halophilic bacteria (Göller et al., 1998), ectoine was believed to have unique advantages in salt tolerance. Here, to clarify the role of ectoine and exclude the influence of the native compatible solute sucrose in salt stress adaption in Syn7942, the sucrose production pathway of Syn7942 was knocked out to obtain strain Syn7942/Δsps. The growth of Syn7942/Δsps was measured under the conditions of 0, 100, 200, 300, 400, and 500 mM NaCl (Supplementary Figure S1C), and the results showed that the OD₇₅₀ of Syn7942/Δsps on the 7th day was decreased to 62% under the 100 mM NaCl condition, 36% under the 200 mM NaCl condition compared with WT, and the strain could not survive under the 300 mM NaCl condition (Figure 3A).

Next, the ectoine synthesis module was introduced into Syn7942 instead of sucrose synthesis to obtain strain Syn7942/Δsps-ect; and the salt-adaptive ability was then investigated for growth at 0, 100, 200, 300, 400, and 500 mM NaCl, respectively (Supplementary Figure S1D). As shown in Figures 3B,C, compared with Syn7942/Δsps, the introduction of the ectoine synthesis module successfully recovered the salt tolerance of Syn7942/Δsps to WT level under 100, 200 and 300 mM NaCl conditions. Ectoine production was also determined in Syn7942/Δsps-ect grown under different salt concentrations. As shown in Figure 3D, the yield of ectoine on the 7th day was 53 mg/l under 0 mM NaCl, the similar level as that of Syn7942/NSI-ect. The yields of ectoine in Syn7942/Δsps-ect grown under 100, 200, and 300 mM NaCl conditions were 115.2 mg/L, 112.5 mg/L, and 72.9 mg/L, respectively, which were slightly improved compared with the yields in Syn7942/NSI-ect under the same conditions, probably due to that an additional part of ectoine need to be synthesized to make up for the lack of sucrose under the condition of salt stress. In addition, in Syn7942/Δsps-ect the lack of sucrose accumulation saved a part of carbon source for ectoine synthesis.

To explore the working mechanism of ectoine in salt stress adaptation in Syn7942/Δsps-ect, comparative transcriptomic analysis was performed in Syn7942/Δsps-ect and WT, under 0 and 300 mM NaCl conditions, respectively. After data quality control, more than 16 million reads per sample were mapped to the reference genome, and the detected genes in each sample covered more than 94% of all the predicted genes (Supplementary Table S3).

As shows in Figure 4, under 0 mM NaCl condition, only 11 DEGs were identified between Syn7942/Δsps-ect and WT. This indicated that under the condition of 0 mM NaCl, ectoine synthesis did not have significant effect on the overall metabolism.