Introduction

Ectoine is a derivative of amino acid, 1,4,5,6-tetrahydro-2-methyl-4-pyrimidnecarboxylic acid, that is found intracellularly in a diverse range of organisms such as halophilic archaea, bacteria, fungi, and algae1,2,3,4. Its ability to enhance enzymatic activities, stabilize DNA structures, and protect proteins improves salt-stress resistance, making ectoine a valuable compound in various applications5,6,7,8. Additionally, ectoine acts as a free radical scavenger, and its anti-inflammatory properties have been utilized in biomedical applications, including skin trauma, allergic rhinitis, dry eye, and lung and intestinal diseases9,10,11,12,13,14. Therefore, due to its widespread application potential and market demand, ectoine is considered a promising high-value chemical. However, the chiral carbon atom structure of ectoine makes chemical synthesis challenging. Thus, microbial fermentation, which is efficient and cost-effective, is the primary large-scale source of ectoine15,16.

The metabolism of l-aspartate-4-semialdehyde, a forerunner substrate in the ectoine biosynthesis pathway, relies on enzymes encoded by the conserved gene clusters ectABC or ectABCD-ask17. The produced metabolite then undergoes a three-step enzymatic reaction to produce ectoine. In this pathway, the ectB-encoded diaminobutyric acid-2-oxoglutarate transaminase (EctB) first converts l-aspartate-β-galacturonide to l-2, 4-diaminobutyric acid (DABA). Second, DABA is converted to N-acetyl diaminobutyric acid (NADA) by the ectA-encoded l-2, 4-diaminobutyric acid aminotransferase. Finally, NADA is converted to ectoine by ectoine synthase, which is encoded by ectC18,19,20. Wild-type bacterial strains, which are typically obtained via direct screening surveys from natural environments, have a limited capacity for the intracellular accumulation of ectoine and cannot meet the needs of practical production. In addition, the high cost of producing ectoine from wild-type strains renders large-scale ectoine production difficult. Thus, ectoine overproduction has become an important area of research, and previous studies have identified excellent industrial strains, modified overproducing strains, constructed genetically engineered strains, and/or systematically constructed metabolically engineered strains that can be effectively used for ectoine overproduction21,22,23,24,25. Metabolically engineered strains with high ectoine production can be systematically generated using gene knockout technology targeting the ectoine biosynthesis metabolic pathway. For example, disrupting the glyoxalate cycle by knocking out its control gene iclR increases metabolic flow from the carbon source glucose to L-aspartic acid, generating strains with high ectoine yields (i.e., ectoine yields of 12–18 g ectoine L−1 after 6–8 h of fractionated fermentation)26. Similarly, knockout of the genes encoding the aspartate kinase/homoserine dehydrogenase thrA and the glyoxylate shunt transcriptional repressor gene iclR increases the oxaloacetate pool to produce strains with high ectoine yields (i.e., ectoine yields of 25.1 g ectoine L−1 with glucose fermentation under low salt conditions)27. It is equally possible to achieve high ectoine yields by knocking out key or nodal genes in the metabolic shunt pathways to disrupt competing pathways28.

The synthesis and accumulation of ectoine, a compatible solute, has been established as a significant process in various halophilic bacteria, including Halomonas. Ectoine is an amino acid derivative whose synthetic precursors are closely related to Asp and l-aspartate-4-semialdehyde. The latter is a crucial precursor in the biosynthesis of three distinct substances: ectoine, l-lysine, and glycine. The H. campaniensis strain XH26 is known to thrive in a wide range of salinities, from 0 to 3 mol L−1 NaCl, and produces multiple compatible solutes, exhibiting high ectoine yields29. Notably, hom-encoded homoserine dehydrogenase catalyzes the metabolic shunt of l-aspartate-4-semialdehyde to glycine, and knockout of hom is expected to increase ectoine synthesis effectively (Fig. 1). In this investigation, we employed a CRISPR/Cas9 system to knock out hom in the Halomonas campaniensis strain XH26, thus preventing the transformation of l-aspartate-4-semialdehyde into glycine, and examined the resulting impact on ectoine yields.

Figure 1
figure 1

Hypothesized effects of hom knock out on the ectoine biosynthesis pathway of Halomonas strain XH26. lysC: aspartate kinase; asd: aspartate-semialdehyde dehydrogenase; ectB: diaminobutyrate-2-oxoglutarate transaminase; ectA: L-2,4-diaminobutyric acid acetyltransferase; ectC: L-ectoine synthase; ectD: ectoine hydroxylase; hom: homoserine dehydrogenase; thrB: homoserine kinase; thrC: threonine synthase; ltaE: threonine aldolase. The dashed lines represent substances exported or imported by other metabolic pathways. The colored boxes correspond to metabolic pathways.

Full size image

Materials and methods

Bacterial strains, plasmids, and culture conditions

All bacterial strains and plasmids used in this study are shown in Table 1. Wild-type H. campaniensis strain XH26 (CCTCCM2019776) was isolated from the Xiaochaidan Salt Lake (Qinghai Province, China) and deposited in the China Center for Type Culture Collection (CCTCCM, Wuhan, China). E. coli DH5α and E. coli Stab13 cells were used as hosts for molecular cloning. The donor vector plasmid pMD19-T-L + R was used to seamlessly repair DNA sequences after knockout to prevent insertion and deletion mutations. The basal culture medium is composed of NaCl 50 g L−1 (adjustable), MgSO4·H2O 25 g L−1, KCl 55 g L−1, CaCl2 0.2 g L−1, sodium citrate 3 g L−1, sodium l-glutamate 6.5 g L−1, enzymatically hydrolysed casein 7.5 g L−1 and Yeast Extract 2 g L−1.

Table 1 Bacterial strains and plasmids used in this study.
Full size table

Genome analysis of ectoine metabolite pathway of H. campaniensis strain XH26

Based on the whole genome sequence of H. campaniensis strain XH26(NCBI no: CP071889)29, the coding sequences of the associated genes and transcriptional regulators involved in ectoine metabolism were mined and validated using BLAST searches against NCBI database. The ectoine metabolic pathway of H. campaniensis XH26 was evaluated using KEGG (Kyoto Encyclopedia of Genes and Genomes) database pathway analysis and salt-induced (0 M NaCl, 1.5 M NaCl, and 2.5 M NaCl) transcriptomic profiling30,31.

Construction of knockout strains

To completely knockout the entire hom gene, the target gene site was designed and sgRNA fragments were identified using the CRISPOR (http://crispor.tefor.net/) or CRISPRdirect (http://crispr.dbcls.jp) platforms. The sgRNA sequence (5′-GGTATCACGGGCAACATCCAGCAGG-3′) was synthesized and cloned into the pUC57 vector to create the recombinant plasmid pUC57-sgRNA32. Genomic DNA was extracted from H. campaniensis strain XH26 using the Bacteria Genomic DNA Extraction Kit (9763, Takara Bio Co., China). The overlapping RCR amplification was performed to construct the fulfilment donor vector (plasmid pMD19-T-L + R) and generate the defective gene fragment, using specific primers listed in Table 2 and genomic DNA as template. The donor vector, lacking 1347 bp in hom, was ligated into the pMD19-T and transferred into the strain E. coli DH5α33,34. The Wild-type strain XH26 was inoculated into liquid media formulated for strain activation. After activation, the cells were inoculated into 50 mL of liquid medium and incubated at 37 °C until reaching OD600 values of 0.5–0.8. Cells were subsequently incubated in an ice bath for 10 min and collected by centrifugation at 6000×g for 7 min at 4 °C. Collected cells were washed with pre-cooled MgCl2 and PEG, followed by a final resuspension in PEG35,36. The plasmids pMD19-T-L + R, pUC57-sgRNA, and pwtCas9-bacteria were extracted from E. coil DH5α/L + R-T, E. coli DH5α/pUC57-sgRNA, and E. coli Stab13/pwtCas9-bacteria cells, respectively, using a plasmid extraction kit (9760, Takara Bio Co., Ltd., China). The successfully transformed cells were incubated at 37 °C for 2 h. The pMD19-T-L + R, pUC57-sgRNA, and pwtCas9-bacteria plasmids were transformed into strain XH26 in batches and screened using a solid plate medium containing ampicillin and tetracycline (T8180, Solarbio Life Science Inc., Beijing, China). Monoclonal strains were selected and verified by PCR amplification and DNA sequencing.

Table 2 The primers used in this study.
Full size table

Genes related to ectoine synthesis in H. campaniensis strain XH26

In this study, total bacterial RNA was extracted from H. campaniensis strain XH26 using Trizol base lysis, under different salt gradient conditions [NS: no salt (0 mol L−1 NaCl), MS: medium salt (1.5 mol L−1 NaCl) and HS: high salt (2.5 mol L−1 NaCl)] and in triplicate. The purity of the extracted RNA was evaluated using a Nanodrop 2000 spectrophotometer (ND200, Thermo Fisher Scientific, USA), while RNA concentration was measured using a Qubit 2.0 Fluorometer (Q32867, invitrogen, USA). The cDNA was synthesized using a reverse transcription kit as per the manufacturer's instructions and the resulting reverse transcription products were stored at − 20 °C. Real-time quantitative PCR was performed to assess gene expression in response to different salt gradient conditions. The qRT-PCR primers were designed using the Primer 5.0 software and the primer sequences are shown in Table 2. Polymerase reactions were performed according to the qRT-PCR kit operation and run on a machine (Roche Light Cycler® 480 II, Roche, Switzerland) with run parameters of 95 °C for 3 min; 95 °C for 10 s, 65 °C for 20 s and 72 °C for 30 s for 40 cycles. The internal reference gene used was GADPH and each biological sample was analysed in triplicate. The relative expression levels of each gene were calculated using the 2−ΔΔCT method37.

Determination of ectoine and betaine yields

To determine ectoine and betaine production yields, strains XH26 and XH26/Δhom were cultivated in tubes with basal culture medium (with 5 g L−1 glucose supplementation) for seed preparation, by incubated at 35 °C, 150 rpm until OD600 = 0.6. With same biomass inoculation, these seed culture was inoculated to 500-mL baffled flask containing 250 mL of NaCl-gradient (0, 0.5, 1.0, 1.5, 2.0, 2.5, 3.0 mol L−1) basal culture medium supplemented with 5 g L−1 glucose. After shaking incubation at 35 °C for 48 h (ZQTY-90S, Zhichu Instrument CO., Ltd., China), cultures were sampled for measuring the cell growth and ectoine production. Three independent experiments were conducted to avoid artifacts caused by handling.

Determination of ectoine production from batch fermentation

The pre-cultured cells were inoculated into a 3 L microbial fermenter (MBF300ME, Tokyo RIKEN Co., Ltd, Japan) at 35 °C for batch fermentation. The yeast extract in the XH26 basal culture medium (with 5 g L−1 glucose supplementation) was then replaced by 0.5% aspartic acid as a nitrogen source. During fermentation, dissolved oxygen was controlled at 40% for 0–24 h and 20% for 24–72 h by means of an auto-adjusted agitation from 80 to 400 rpm, with the working volume of 1.8 L in the 3 L bioreactorand and a salinity control of 1.5 mol NaCl L−138,39,40,41. The pH of the fermentation condition of cells was controlled at pH 8.0 with addition of appropriate amounts of 0.1 N NaOH and 0.1 N HCl. Cofactors such as biotin (0.3 µg mL−1) can regulate the microbial cell metabolism, and thus promoted the metabolism of amino acids and enzyme synthesis42. Therefore, appropriate amounts of biotin were added to the fermentation process to increase the overall production of ectoine.

Analytical methods

Cell growth was monitored based on the optical density of the sample at the wavelength of 600 nm was measured with the UV/Visible spectrophotometer (SP-754, Shanghai Spectrum Instruments Co., Ltd., China). Ectoine and betaine concentrations in the extracts were determined using high-performance liquid chromatography (HPLC) with an ectoine standard curve as a reference43 as follows: First, 1 mL of each bacterial culture was centrifuged at 8000×g for 5 min and the supernatant was discarded. Then, 1 mL of ultrapure water was added to each tube, and the tube contents were ground for 5 min using a third-generation variable speed TGrinder (OSE-Y50, Tiangen Biotech Co., Ltd., China). After centrifugation at 8000×g for 5 min, the water phase (containing compatible solutes) was separated and filtered through a 0.22 µm filter membrane for HPLC analysis. Ectoine was identified using an Agilent Technologies 1260 Infinity HPLC instrument (USA) with a SeQuant ZIC-HILIC column (15.0 cm by 4.6 mm; 5 μm) (Sigma-Aldrich, USA). A solution of acetonitrile (1000161000, Sigma-Aldrich, USA) and pure water (v/v: 80/20) was used as the mobile phase, and the instrument was operated at a detection wavelength of 210 nm, a flow rate of 1.0 mL min−1, a column pressure of 3.486–4.761 MPa, a column temperature of 30 °C, and a sample volume of 10 µL. Betaine detection condition was as follows: a solution of acetonitrile and pure water (v/v: 85/15) was used as the mobile phase, detection wavelength of 195 nm, fow rate of 0.7 mL min−1, column pressure of 3.486–4.761 MPa, column temperature of 30 °C, and sample volume of 15 μL. A standard curve was generated using gradient-diluted compatible solute standards.

Results

Integrated reconfiguration of the ectoine metabolic pathway in H. campaniensis strain XH26

The synthesis and accumulation of compatible solute ectoine is a crucial adaptation mechanism in many halophilic bacteria, such as Halomonas, including H. elongata DSM 2581T, H. hydrothermalis Y230,44. Analysis of high-throughput sequencing data, including genomic, salt-excited transcriptomic, and proteomic data from H. campaniensis strain XH26, identified multiple genes associated with ectoine anabolism along various metabolic pathways. Asp, oxaloacetate and Glu were used as metabolic nodes to screen for differentially expressed genes or related transcription factors involved in metabolism pathways directly or indirectly associated with ectoine metabolic (Table 3). KEGG analysis revealed that ectoine biosynthesis is associated with Asp, aspartate hemiacetal metabolism, and these metabolic pathways are in turn linked to aspartate (Asn), glutamate (Glu), glutamine (Gln) and alanine (Ala). At the same time, the carbon/nitrogen metabolic flow is associated with the tricarboxylic acid cycle (α-ketoglutarate, succinic acid, ferredoxin, malic acid, oxaloacetate) and amino acid (Glu and Asp) transamination (Fig. 2)45,46,47. In the NS/MS comparison group, transcript expression of genes lysC, ectB, ectA, ectC, gltB, gltD, davT, hisD, alh-9, betA, acnB, pckA and gadA was up-regulated, while transcript expression of genes asd and gdhA was down-regulated. In the NS/HS and MS/HS comparator groups, the genes lysC, asd, ectB, ectC, gltB, gltD, davT, hisD, alh-9 and the gene betA were up-regulated and the genes ectA, acnB, pckA, gadA and gdhA were down-regulated. At the level of protein translation, there were no significant differences in the translation levels of most of the proteins, except for AcnB and EctB. The expression levels of AcnB and acnB followed the same trend, i.e. increasing and then decreasing with increasing salinity; the expression levels of EctB and ectB also followed the same trend (increasing with increasing salinity). Proteins act as executors of gene functions, and most gene protein translation levels in the ectoine synthesis pathway do not change significantly with increasing salinity, but the fact that ectoine production is increased is undeniable, and therefore increased protein EctB expression, may be the main cause of increased ectoine production. EctB is the key enzyme that catalyzes the change of l-aspartate-4-semialdehyde to l-2.4-Diamino-butanate, indicating that this pathway plays a key role in the synthesis of ectoine. Increasing the level of the enzyme in the synthetic pathway can effectively increase the yield of ectoine, and on this basis, increasing the concentration of the protein EctB catalytic substrate l-aspartate-4-semialdehyde in this pathway by knocking out hom can also further effectively increase the yield of ectoine.

Table 3 Key genes and transcriptional regulators associated with ectoine metabolism pathway.
Full size table
Figure 2
figure 2

Diagram of the ectoine metabolism-related pathway of H. campaniensis XH26 in this study. The ectoine metabolism-related pathway of H. campaniensis XH26 was reconstructed based on salt-induced transcriptome analysis.

Full size image

Transcript level changes of ectoine metabolite related genes

The biosynthesis of ectoine relies on the ectABC or ectABCD-ask gene clusters, which are evolutionarily conserved48,49. These gene clusters are present in the genome of H. campaniensis strain XH26, with the ectABC cluster located at 19,453,512–1,945,966 in the genome. To verify transcriptional changes in the ectoine synthesis genes identified in RNA-seq, qRT-PCR was conducted to measure mRNA expression levels. No significant differences in transcript levels were observed in the ectABC-linked gene cluster between strain XH26 and strain XH26/Δhom (Fig. 3). In the HS/NS and HS/MS comparison groups, transcription of ectB and ectC was significantly up-regulated while transcription of ectA was down-regulated (p < 0.05, |log2FC|≥ 1). In the MS/NS comparison group, the transcript levels of ectoine synthesis genes were up-regulated in ectA, ectB and ectC expression (p < 0.05, |log2FC|≥ 1). High salinity stress resulted in reduced ectoine synthesis that correlated with reduced transcript levels of ectA. We found that knockout of gene hom seems to increase only the substrate concentration in the enzymatic reaction of ectoine synthesis and had no effect on the enzyme content.

Figure 3
figure 3

Mapping of differential expression levels of genes related to ectoine synthesis. RNA-seq and qRT-PCR results showing the expression changes of ectoine synthesis under salt stress. For qRT-PCR, data are presented as the mean ± standard error (n > 3).

Full size image

Ectoine and betaine fermentation capacity of the genetically defective strains

To investigate the role of hom in ectoine production, the entire hom gene was completely knocked out in the constructed strain XH26/Δhom. The growth and ectoine accumulation of the wild-type strain XH26 and the defective strain XH26/Δhom were compared after 48 h of incubation at various salinities (Fig. 4). The results showed that the maximum growth of XH26/Δhom was higher than that of the wild-type strain, and the salinity tolerance associated with maximum growth was also altered. The wild-type strain exhibited maximum growth at a salinity of 1 mol NaCl L−1 (OD600 value of 1.86), while the defective strain exhibited maximum growth at a salinity of 1.5 mol NaCl L−1 (OD600 value of 1.96) (Fig. 4a). Ectoine accumulation trends were generally similar between the two strains, although ectoine production was higher in the defective strain than in the wild-type strain at all salinities. The highest yield of ectoine [351.13 mg (g CDW)−1 of isoflavones L−1 in shake flask fermentation] was obtained by XH26/Δhom at a salinity of 1.5 mol NaCl L−1, which was 47% higher than the wild-type strain (Fig. 4b). Halomonas species contain diverse compatible solutes and use various substances to counteract salt stress, with betaine functioning under low salt conditions. Betaine fermentation experiments showed that the betaine yields of XH26/Δhom were lower than those of the wild-type strain under low salt conditions. In particular, the betaine production of XH26/Δhom was 19.98 mg (g CDW)−1 at 1.5 mol L−1 NaCl, much lower than the 69.58 mg (g CDW)−1 of the wild-type strain (Fig. 4c). At a salinity of 3.0 mol NaCl L−1, the ectoine production of the wild-type strain was 0 and the ectoine production of the defective strain remained relatively high, while the betaine production of both was 0. Whether the deficiency of betaine is responsible for the relative boost in ectoine yields and the link between them needs to be explored in considerable depth. Together, our experimental results showed that hom is a nodal gene in the ectoine synthesis pathway. Knockout of hom blocked the use of the precursor substrate (l-aspartate-4-semialdehyde) in the glycine biosynthesis pathway, thereby allowing more l-aspartate-4-semialdehyde to be used for ectoine synthesis and leading to increased ectoine production, consistent with the results of anticipated experiments.

Figure 4
figure 4

Growth (a), ectoine accumulation (b) and betaine accumulation (c) of the wild-type strain XH26 and the defective strain XH26/Δhom under various salinity conditions. Cultures were grown in 500 mL baffled Erlenmeyer flasks, and each experiment was performed in triplicate.

Full size image

Batch fermentation of cells ectoine using bioreactor

During the fermentation process for ectoine production, a 0.5% concentration of aspartic acid was employed as a nitrogen source instead of yeast extract in the medium, supplemented with 5 g L−1 glucose. In addition, biotin was added to the fermentation medium to facilitate bacterial growth and ectoine production. The growth and ectoine production of wild type strain and XH26/Δhom were compared after 72 h of fermentation in batch (Fig. 5). The results revealed that both strains exhibited a significant increase in growth and ectoine yield, with the defective strain XH26/Δhom exhibiting higher growth and ectoine yield than the wild-type XH26 strain. The highest ectoine yield of 587.09 mg (g CDW)−1 was obtained by the defective strain XH26/Δhom at 56 h, which was higher than that of the wild type strain XH26 at 48 h. The highest ectoine yield of 385.03 mg (g CDW)−1 was obtained at 48 h by the wild-type strain XH26 (Fig. 5b). The maximum biomass (CDW) of XH26/Δhom was observed at 40 h (4.95 g L−1), while that of the wild-type strain was achieved at 48 h (4.64 g L−1) (Fig. 5a). Furthermore, the defective strain XH26/Δhom produced ectoine earlier than the wild-type strain, with ectoine already being produced by hour 4. Although the ectoine production of the defective strain was higher than that of the wild-type strain during fermentation, the trend of ectoine accumulation was essentially similar between the two strains.

Figure 5
figure 5

Growth (a) and ectoine accumulation (b) of the wild-type strain XH26 and the defective strain XH26/Δhom under batch fermentation. Each experiment was performed in duplicate.

Full size image

Discussion

In most ectoine-producing strains, multi-omic crosstalk analysis of biosynthetic initiation, response and network regulation of associated genes in compatible solute ectoine under salt-adapted conditions requires in-depth study50. In this context, the transcriptomic analysis of the salt adaptation mechanism of H. beimenensis was identified 16 possible related genes (spoT, prkA, mtnN2, rsbV, lon, smpB, rfbC, rfbP, tatB, acrR1 and acA) involved in salt adaptation, and verified that four genes (qrA, trkA2, nadA and gdhB) are involved in Na+ efflux, K+ uptake, H+ energy conversion and ectoine biosynthesis, respectively, with important regulatory roles51. In the H. campaniensis strain XH26, metabolic linkage analysis revealed that the main biosynthetic pathway of ectoine (lysC + asd + ectABC) is directly linked to Asp and aspartate semialdehyde metabolism, and the main metabolic streams are closely linked to the upstream Asn, Glu, Gln and Ala metabolic networks. Furthermore, the carbon/nitrogen metabolic flow of ectoine biosynthesis is closely related to the tricarboxylic acid cycle (succinic acid, ferredoxin and oxaloacetate), which may provide new insights for the optimization of the ectoine synthesis pathway and the design of metabolic pathway integration experiments (Glu → TCA → Asp → ectoine).

For the most part, the principles for constructing engineered strains with high ectoine yields are based on improving the utilisation of carbon/nitrogen sources48. The low utilization of carbon/nitrogen sources is primarily reflected in the accumulation of too many intermediate products, the coexistence of multiple products, and inefficient synthesis. The under-utilization of carbon sources, which limits ectoine production, is primarily addressed at present using three approaches: (i) Some studies have aimed to increase the input of carbon sources for ectoine synthesis. For example, ectoine synthesis is enhanced by the overexpression of sugar transporter proteins that increase glucose and xylose uptake49. (ii) Other studies have aimed to modify the ectoine synthetic pathway based on analyses of the ectoine metabolic pathway, thereby improving carbon source use. Primary pathway-related goals include increasing the activity of key enzymes in the pathway50, disruption of futile cycles 5151, removing competing amino acid synthesis pathways 5252, and blocking the ectoine catabolic pathway53. (iii) Finally, some studies have analyzed the effects of carbon supply on ectoine synthesis, aiming to determine how changes in carbon supply and the optimization of fermentation conditions might improve ectoine production and increase of ectoine demand54,55. Thus, with the defective strain XH26/Δhom constructed in this study, the synthetic precursor substrate metabolic shunt pathway was effectively removed, improving the availability of carbon/nitrogen sources for ectoine synthesis and increasing ectoine production by 1.47-fold relative to XH26.

In this study, a defective strain XH26/Δhom was constructed by knocking out the hom gene, which disrupted the shunt of the ectoine synthesis precursor substrate l-aspartate 4-semialdehyde, thereby improving the availability of carbon/nitrogen sources for ectoine synthesis56,57,58. As expected, ectoine fermentation experiments revealed that ectoine synthesis by the gene-deficient strain increased under saline culture conditions (0–3 mol NaCl L−1) compared to the wild-type strain. Ectoine yields increased by about 4.23-fold under low salinity (0.5 mol NaCl L−1), and ectoine yields was 351.13 mg (g CDW)−1 at a salinity of 1.5 mol NaCl L−1, much higher than the 239.18 mg (g CDW)−1 of the wild-type strain. In the 3-L bioreactor, optimising the batch fermentation parameters of nitrogen source, agitation and aeration. After 72 h fermentation, the maximum ectoine yields of XH26/Δhom was 587.09 mg (g CDW)−1. In addition, the growth of the defective strain also differed to that of the wild type, possible due to the knockout of the hom gene. Most notably, the salinity associated with optimal growth differed between the defective strain and the wild type. We suspect that the wild-type H. campaniensis strain XH26 produces less ectoine than is required to equilibrate cellular osmotic pressure. Therefore, the increase in the intracellular ectoine content of defective strain XH26/Δhom led to a concomitant increase in the salinity associated with optimal growth.

It is worth noting that glycine synthesis is closely related to the betaine metabolism. Knockout of hom blocks the synthesis of glycine from l-aspartate-4-semialdehyde, potentially reducing glycine levels in microorganisms and thereby decreasing betaine synthesis or increasing betaine catabolism. Betaine fermentation experiments revealed that betaine synthesis by the gene-deficient strain decreased under saline culture conditions (0–2.5 mol NaCl L−1) compared to the wild-type strain. Betaine yields was 19.98 mg g CDW)−1 at a salinity of 1.5 mol NaCl L−1, much lower than the 69.58 mg (g CDW)−1 of the wild-type strain. Halomonas species contain diverse compatible solutes and use various substances to counteract salt stress, some of which are antagonistic towards one another59,60. For example, betaine inhibits ectoine synthesis under low salinity conditions. Therefore, knocking out hom may also improve ectoine production by reducing betaine concentrations.