Introduction

Ectoine, a compatible solute accumulated by halophilic microorganisms, plays important roles in improving cellular resistance to high temperature, radiation, drought, and high osmotic pressure [1, 2]. Ectoine is widely used as functional components in food, cosmetic and biologics, and it has an annual demand of approximately 15,000 tons and a market price of ∼$1,000/kg [1,2,3].

In recent years, microbial production of ectoine has emerged as a promising alternative to traditional chemical synthesis, offering an economically feasible and environmental friendly approach to large-scale production [2]. The ectoine biosynthetic pathways have been intensively researched in halophilic microorganisms [4, 5]. Two major strategies of ectoine biosynthesis are proposed, including the production of ectoine by halophilic bacteria and heterologous production of ectoine by Escherichia coli and Corynebacterium glutamicum [6,7,8,9,10,11]. Moreover, two genome-editing approaches have been developed in halophilic bacteria, including CRISPR/Cas9 and suicide plasmid/counterselection [12,13,14]. Unsterile production of high value-added products such as polyhydroxyalkanoate (PHA) and ectoine by halophilic bacteria-based cell factories has been achieved by utilizing anti-microbial contamination of halophilic bacteria [9, 15], which highlights the potential of halophilic bacteria as chassis cells for the next-generation industrial biotechnology (NGIB) [16]. In previous studies, engineered Halomonas hydrothermalis Y2 and H. bluephagenesis TD01 could produce 3.13 and 6.3 g/L ectoine in shake-flask fermentation using glucose as the carbon source, respectively [8, 9]. Since the vast majority of ectoine producers utilize cereal crops derived carbohydrates as the carbon sources, the large-scale production of ectoine is restricted by high substrate cost and limited cereal resource.

Currently, the biosynthesis of bioplastics, biofuels and platform chemicals by using sustainable and cost-effective carbon sources such as lignocellulose and methane has become a research hotspot [17,18,19,20,21,22]. Lignocellulosic biomass (LCB) with abundant and stable sources has a global annual production of more than 1.5 trillion tons [23, 24]. Glucose and xylose are two main monosaccharides derived from the enzymatic hydrolysis of cellulose and hemicellulose components of lignocellulose [22]. Most ectoine-producing strains are unable to utilize xylose, thus, high-value bioconversion of xylose has become a critical challenge in the production of ectoine from LCB. Tanimura et al. [17] constructed the △ectD mutant strain of H. elongata, with the ectoine yield of 53.53 mg/g fresh cell weight using glucose and xylose as the co-carbon sources. Recently, Methylotuvimicrobium alcaliphilum 20Z was metabolically engineered to simultaneously utilize methane, glucose and xylose to produce ectoine by introducing glucose and xylose metabolism modules from E. coli and Zymmonas mobilis, and the finally constructed strain 20ZXG/ΔectR1 grown on methane, glucose and xylose reached the ectoine yield of 37.93 ± 3.27 mg/g dry cell weight [22].

In this study, a halophilic bacterium H. cupida J9, which was previously isolated by our lab from high-salt wastewater [25], was demonstrated to be capable of producing ectoine from either glucose or xylose. Moreover, the salt-tolerant mechanism of H. cupida J9 was elucidated by genome and transcriptome analysis. Furthermore, the engineered bacterium J9U-P8EC was constructed by inserting the strong promoter P8 in the upstream of the ectABC and ppc genes, which resulted in an increased production of ectoine from either xylose or a glucose-xylose mixture. Finally, the capacity of J9U-P8EC to utilize corn straw hydrolysate for ectoine production was confirmed in an open fermentation process.

Materials and methods

Strains and culture conditions

E. coli strains were cultivated in Luria-Bertani (LB) medium [26] at 37 °C. H. cupida J9U (J9 △upp) and its mutants were cultured in high-salt LB medium containing extra 50, 90 and 150 g/L NaCl (LB60, LB100 and LB150) and mineral salt medium [22] supplemented with 30 g/L glucose (MMG) at 37 °C. The salt concentration of the fermentation medium was 60 g/L, which is the optimal for the growth of H. cupida J9. Mineral salt medium plus 3 g/L urea (MM3) was used as the optimized medium. For ectoine production, H. cupida strains were used in MM3 supplemented with 30 g/L glucose (MMG3), MM3 supplemented with 30 g/L xylose (MMX3), and MM3 supplemented with 20 g/L xylose and 10 g/L glucose (MMXG3) at 37 °C and 200 rpm for 60 h. The preparation of lignocellulose hydrolysate medium MML3 is described in Sect. 2.5. When required, the media were supplemented with 50–100 µg/mL kanamycin (Kan) or 100 µg/mL 5-fluorouracil (5-FU).

RNA-seq and RT-qPCR analysis

After J9U was cultivated in LB60, LB100 and LB150 at 37 °C and 200 rpm for 12 h, 5 mL of the culture broths were centrifuged to collect the mid-log phase cells. Total RNA samples were extracted from the cells and purified according to previous protocols [27]. Quality control of RNA samples was performed by electrophoresis on an Agilent 2100 Bioanalyser. RNA samples with high purity and integrity were used to construct sequencing library. Libraries were sequenced on an Illumina Novaseq 6000 by Shanghai Majorbio Bio-pharm Technology Co., Ltd. Differentially expressed genes (DEGs) in J9U cells grown in media containing different salt concentrations were determined using the DESeq2 package. Statistical enrichment of gene ontology (GO) was performed using Goatools (https://github.com/tanghaibao/GOatools) and Fisher’s exact test.

mRNA extracted from J9U cells was reverse transcribed to cDNA using a HiScript II Q RT SuperMix (Vazyme). Real-time qPCR was performed using cDNA and a ChamQ Universal SYBR qPCR Master Mix (Vazyme) on a real-time qPCR system (Applied Biosystems). The 2−ΔΔCt method [28] was used to calculate the relative transcription levels of the ppc, lysC, asd, ectA, ectB, ectC, ectD and doeA genes in J9U using the 16 S rRNA gene as the internal reference. Primers for RT-qPCR are listed in Table S1.

Construction of H. Cupida mutants

In this study, a genome-editing approach based on suicide plasmid/counterselection described in Wang et al. [14] was used to construct H. cupida mutants. Firstly, the P8 promoter from Pseudomonas putida KT2440 (P8KT) [29] plus ribosome binding site was spliced with two homologous arms by overlap PCR [30], and the fusion fragment was ligated into the upp-containing pKJU (derived from a suicide plasmid pK18mobsacB [31], to generate pKJU-P8-ectABC and pKJU-P8-ppc. Subsequently, the promoter knock-in vectors were introduced into a 5-FU resistant mutant J9U (J9 Δupp) [14] by conjugative transfer using E. coli S17-1 as the donor strain. The upp gene encodes uracil phosphoribosyltransferase, which catalyzes the conversion of 5-fluorouracil (5-Fu) to 5-fluorodeoxyuracil monophosphate, thereby inhibiting cell growth. The single- and double-crossover mutants were screened by the Kan and 5-FU resistance on LB60 agar plates supplemented with 100 µg/mL Kan and 5-FU, respectively, to obtain the promoter knock-in mutants J9U-P8E and J9U-P8EC. To obtain the ectABC knockout mutant J9U-dE, the construction of ectABC knockout vector, the introduction of knockout vectors into J9U, and the screening of the single- and double-crossover mutants were performed as described in the construction of promoter knock-in mutants. Finally, the successful construction of J9U-dE, J9U-P8E and J9U-P8EC was confirmed by DNA sequencing. The strains, plasmids, and primers used for mutant construction are listed in Table 1 and S1.

Table 1 Strains and plasmids used in this study
Full size table

Shake-flask fermentation for ectoine production

A single colony of J9U and its mutants were cultivated overnight in a tube containing 5 mL of LB60 at 37 °C and 200 rpm. Then, the primary seed culture was transferred to a 500 mL flask containing 100 mL of LB60 at a 1% (v/v) ratio and cultured at 37 °C and 200 rpm for 12 h. Subsequently, the secondary seed culture was inoculated into a 500 mL flask containing 100 mL of ectoine fermentation media (MMG3, MMX3 and MMXG3) (Sect. 2.1.) at a 5% (v/v) ratio and fermented at 37 °C and 200 rpm for 60 h. For ectoine extraction, the fermentation broth (1 mL) was transferred to a 1.5 mL freezing tube with 95% zirconia beads, and then the cells were thoroughly lysed by centrifugation at 2,400 g for 90 s using a high-efficiency tissue cell-destroyer 1000 (Xinzongke Biotech Co. Ltd.). Next, the mixture was transferred to a 1.5 mL Eppendorf tube and centrifuged at 4 °C and 13,500 g for 10 min. The supernatant was collected and passed through a 0.22 μm organic pore filter. To determine the ectoine content, samples were subjected to LC/MS analysis (Sect. 2.6.). To determine cell dry weight (CDW), the fermentation broth was centrifuged at 4 °C and 13,500 g for 10 min to collect the cells, and then the cells were lyophilized in a vacuum freeze dryer for 24 h and weighed. Growth and sugar consumption by J9U-P8EC, J9U-P8E and J9U were measured in ectoine fermentation media using the methods described in Sect. 2.6.

Ectoine production from corn straw hydrolysate

Corn straw powder was dried at 60 °C and pretreated with 2% H2SO4 at 100 °C for 2 h in a 10:1 liquid-to-solid ratio. Then, enzymatic hydrolysis of the hydrolysate was performed with 30 FPU of Cellic CTec3 HS (Novozymes) at 50 °C and pH 5.0 for 36 h (0.5 ml of enzyme for every 100 ml of hydrolysed solution). To adsorb toxic substances (including phenolic compounds, furfural, HMF, and acetic acid) from the non-saccharide compounds [32], 0.5% (w/v) activated carbon (Aladdin) was added to the hydrolysate, and the suspension was then incubated at 50 °C and 150 rpm for 60 min. Finally, the insoluble matter was removed by centrifuging the suspension at 9,800 g for 10 min, to obtain the straw hydrolysate with a xylose to glucose ratio of 3:1 and a total sugar concentration of 30 g/L. To use straw hydrolysate as the carbon source for ectoine production, the MML3 medium was prepared by dissolving all components of MM3 in the straw hydrolysate. J9U-P8EC or J9U was inoculated into a 500 mL flask containing 100 mL of MML3 and fermented at 37ºC and 200 rpm for 60 h. Growth and sugar consumption by J9U-P8EC and J9U were measured in MML3 using the methods described in Sect. 2.6. The ectoine titre and CDW at the end of fermentation were determined as described in Sect. 2.4. and 2.6.

Analytical methods

The ectoine content was determined using Agilent Q-TOF LC/MS equipped with an InfinityLab Poroshell 120 SB-Aq column. Acetonitrile and water (2:98, v/v) was used as the mobile phase at a flow rate of 0.5 mL/min. The column temperature was maintained at 30 °C. The UV detection wavelength was set at 220 nm. Cell density (OD600 nm) of J9U and its mutants grown in media was monitored using a UV-1900i spectrometer (Shimadzu). A SBA-40D biosensor equipped with glucose and xylose oxidase electrodes (Shandong Academy of Sciences) was used to detect the concentration of glucose and xylose in the fermentation broth.

Results and discussion

Elucidating the salt-tolerant mechanism of H. Cupida J9

Halophilic bacteria combat high-salt environments by accumulating compatible solutes and synthesize ectoine as the major low molecular weight compatible solutes [4, 6]. Ectoine biosynthesis pathway is highly conserved among halophilic microorganisms and has been intensively studied [4, 6, 8, 9]. In this study, whole-genome sequencing and functional annotation of H. cupida J9 suggests that the complete ectoine biosynthesis pathway is present in the genome of H. cupida J9. This pathway mainly consists of aspartate kinase (lysC), aspartate semialdehyde dehydrogenase (asd), 2,4-diaminobutyryltransferase (ectA), 2,4-diaminobutyric acid transferase (ectB), and ectoine synthase (ectC) (Fig. S1). The sequencing results reveal that single-copy ectA, ectB and ectC genes form the ectABC cluster within a shared operon in the genome H. cupida J9. The genome sequence of H. cupida J9 was deposited in the GenBank database under accession no. ON045848.

To elucidate the role of ectoine in improving salt tolerance of H. cupida J9, the mutant lacking ectABC (J9U-dE) were constructed. The growth inhibition was observed with ectABC knockout mutant under high salt conditions. The growth rate and the final cell density of LB60-grown J9U-dE were lower than that of J9U (Fig. 1A). The ectABC knockout mutant J9U-dE lost the capability to synthesize ectoine and the growth of J9U-dE was significantly inhibited by high salt concentrations (Fig. 1B and C), which clearly indicated that the accumulation of ectoine by H. cupida J9 played crucial roles in improving salt tolerance of H. cupida J9. Production of ectoine by H. cupida J9 was verified by LC-MS analysis of fermentation products. The retention time of a chromatographic peak in the HPLC chromatograms of the fermentation products corresponded to that of the authentic standard of ectoine, and the compound corresponding to the peak had the same mass spectrogram compared to the authentic standard of ectoine (Fig. S2). In addition, J9U accumulated 0.26, 0.40 and 0.68 g/L ectoine, respectively, when grown in LB60, LB100 and LB150 (Fig. 1D). Ectoine accumulation by J9U increased with the increase of salt concentration, suggesting that salt stress may stimulate the synthesis of ectoine in J9U.

Fig. 1
figure 1

(A, B and C) Growth curves of J9U, J9U-dE and J9U-dB (D) Ectoine accumulation by J9U under salt stress. These strains were cultured at 37 °C and 200 rpm in LB60, LB100 and LB150. The data represent the mean ± standard deviation of triplicate measurements

Full size image

Correlation of the transcription of ectoine metabolism modules with salt stress

Three transcriptome samples, named H6 (6% NaCl), H10 (10% NaCl) and H15 (15% NaCl), were used to assess the DEGs in J9U under the stimulation of different salt concentrations. As shown by the inter-group difference analysis, 551 DEGs including 192 upregulated and 359 downregulated genes were found by a comparison of H6 and H10. 1126 DEGs including 669 upregulated and 457 downregulated genes were found by a comparison of H10 and H15. 1310 DEGs comprising 700 upregulated and 610 downregulated genes were found by a comparison of H6 and H15 (Fig. S3A). Venn diagram analysis indicated that the numbers of unique genes in three comparative groups were 113, 270 and 355, respectively, with 147 genes being common across three comparative groups (Fig. S3B).

A total of 1341 DEGs were annotated using the GO database. Among them, 692, 547 and 102 DEGs were associated with molecular function, cellular component and biological process, respectively (Fig. 2A). Under salt stress, nearly all genes related to ectoine synthesis displayed an upregulation trend, notably with significantly elevated transcription levels of the ectB, ectC and ectD genes. Within the PPO-node, the transcription levels of ppc (encoding phosphoenolpyruvate carboxylase) and pyk (encoding pyruvate kinase) genes were also elevated. Additionally, those genes associated with ectoine metabolism displayed varying degrees of upregulation (doeA, doeB and hom) or downregulation (doeC and doeD) (Fig. 2B). Furthermore, the DEGs associated with ectoine synthesis under salt stress were validated by RT-qPCR. We found that the transcription levels of ectoine synthesis related genes (ectB, ectC and ectD) were gradually elevated with increasing salt concentration (Fig. 2C). Overally, the transcription of ectoine metabolism modules is more active in H. cupida J9 under salt stress.

Fig. 2
figure 2

Transcriptome analysis of H. cupida J9 and RT-qPCR analysis of ectoine synthesis related genes under salt stress. (A) GO annotation of differentially expressed genes in comparative transcriptome analysis of three samples under salt stress. (B) Inter-group differential expression levels of ectoine synthesis related genes as shown by comparative transcriptome analysis of three samples under salt stress. (C) RT-qPCR assays for measuring the transcription levels of ectoine synthesis related genes under salt stress. The transcription levels of the endogenous genes in LB60 were set as 1. Detailed procedures for RNA-seq and RT-qPCR analysis are described in Sect. 2.2

Full size image

Genome analysis reveals the pathway for the biosynthesis of ectoine from xylose

In this study, we have identified enzymes involved in the Weimberg pathway (WBG pathway) for xylose utilization in the J9U genome, including xylose dehydrogenase (xylB), xylose dehydratase (xylD), and 2,5-diketo-D-gluconic acid reductase (aldH). This pathway operates without carbon loss and feeds directly into the central metabolism’s TCA cycle via the end product 2-oxoglutarate. Based on this, a pathway for ectoine biosynthesis from xylose in H. cupida J9 was proposed (Fig. 3), which proceeds through xylonate, 2-dehydro-3-deoxy-D-xylonate, 2,5-dioxopentanoate, 2-oxoglutarate, oxaloacetate, aspartate, and L-aspartate-β-semialdehyde, and eventually enters the ectoine biosynthesis module [4, 5]. Fermentation results demonstrated that MMX3-grown H. cupida J9 efficiently utilized xylose as the sole carbon source to synthesize ectoine (Fig. S7).

Fig. 3
figure 3

Proposed pathway for the production of ectoine from xylose and glucose in H. cupida J9 and enhanced production of ectoine by overexpressing the ectABC and ppc genes in H. cupida J9. Enzymes: xylB, xylose dehydrogenase; xylD, xylose dehydratase; aldH, 2,5-dioxopentanoate dehydrogenase; ppc, phosphoenolpyruvate carboxylase; lysC, aspartate kinase; asd, aspartate semialdehyde dehydrogenase; ectB, 2,4-diaminobutyric acid transferase; ectA, 2,4-diaminobutyryltransferase; ectC, ectoine synthase; ectD, ectoine hydroxylase; doeA, ectoine hydrolase; doeB, N2-acetyl-L-2,4-diaminobutanoate deacetylase; doeC, aspartate-semialdehyde dehydrogenase; doeD, L-2,4-diaminobutyrate transaminase

Full size image

Furthermore, the J9U genome contains both the WBG xylose metabolism pathway and the EMP glucose metabolism pathway (Fig. 3), indicating its capacity to utilize mixed carbon sources for ectoine synthesis. Fermentation results in Sect. 3.5 and 3.6 indicated that both J9U exhibited highly efficient mixed sugar metabolism, making it optimal halophilic platform for developing engineered strains capable of co-utilizing xylose and glucose (Fig. 4).

Fig. 4
figure 4

Cell biomass and ectoine production of MMXG3-grown J9U, J9U-P8E and J9U-P8EC. (A, B and C) Cell growth and total sugar consumption of J9U, J9U-P8E and J9U-P8EC. (D) CDW and the titre and productivity of ectoine obtained by fermentation with J9U, J9U-P8E and J9U-P8EC. These strains were cultured in MMXG3 at 37 °C and 200 rpm for 60 h. The data represent the mean ± standard deviation of triplicate measurements. ** and **** indicate P < 0.01 and P < 0.0001, respectively

Full size image

Optimization of ectoine fermentation media

The majority of ectoine synthesized by J9U is secreted extracellularly (Fig. S4A). Since a substantial nitrogen supply is required for the efficient production of ectoine by microbes, it is necessary to add additional nitrogen sources to optimize the C/N ratio in fermentation media [16, 33]. The inorganic nitrogen source (NH4)2SO4 in MMG has a low nitrogen content (21%) and lowers the medium pH by the formation of acidic substances, which is unfavorable to the growth of H. cupida strains. In contrast, urea with higher nitrogen content (46%) is an efficient organic nitrogen source that can be rapidly utilized by microorganisms [34]. In contrast to (NH4)2SO4, intracellular urea is metabolized into NH4⁺ and the weak acid ions CO3²⁻, the latter of which results in a gradual decrease in the pH of the surrounding environment [35].

J9U-P8E was used in MMG supplemented with 1, 3, 5, 7–9 g/L urea. Fermentation with MMG plus 5 g/L urea obtained the highest ectoine titre (4.23 g/L), followed by MMG plus 3 g/L urea (Fig. S4B). However, the decrease in the ectoine yield was observed in fermentation with MMG plus 7–9 g/L urea. Considering the ectoine titre and the substrate cost, 3 g/L urea was added as the nitrogen source to fermentation media.

Furthermore, at the end of fermentation, the final pH of the fermentation broth was measured. We found that the final pH of the fermentation broth decreased less significantly (initial pH was 9.0) as the urea concentration increased (Table S2). The addition of 3 and 5 g/L urea as the nitrogen source maintained an optimal pH range during fermentation, which is favorable for strain growth and ectoine synthesis. The addition of 3 g/L urea as a nitrogen source maintains an optimal pH range (5.56 ∼ 7.53) during fermentation, which is favorable to strain growth and ectoine synthesis. The pH of the fermentation broths MMX3, MMXG3, and MML3 with 3 g/L urea remained stable between 6.2 and 7.04, thereby confirming the aforementioned observation (Table S2).

Metabolic pathway engineering enhances ectoine biosynthesis

In previous studies, the overexpression of the phaC operon, the itu operon, and the srfA operon with strong promoters can enhance the biosynthesis of PHA, iturin A, and surfactin in P. putida KT2440 and Bacillus amyloliquefaciens LL3 [36,37,38]. In this study, the strong promoter P8KT was inserted into upstream of the ectABC cluster to enhance the biosynthetic pathway from L-aspartate-β-semialdehyde to ectoine (Figs. 3 and 5A and C). The resulting mutant J9U-P8E accumulated 4.24 g/L ectoine at 60 h in MMG3, with a 38.56% increase compared to J9U (Fig. 5E).

Fig. 5
figure 5

Colony PCR results, the suicide plasmid maps, and ectoine production of MMG3- and MMX3-grown J9U, J9U-P8E and J9U-P8EC. (A) PCR confirmation of the construction of P8KT + RBS insertion mutant J9U-P8E. Lanes: M, Marker Ш; 1, J9U; 2, J9U-P8E. (B) PCR confirmation of the construction of P8KT + RBS insertion mutant J9U-P8EC. Lanes: M, Marker Ш; 1, J9U; 2, J9U-P8EC. (C) The suicide plasmid map of pKJU-P8-ectABC. (D) The suicide plasmid map of pKJU-P8-ppc. All specific PCR primers are listed in Supplementary file 1: Table S1. (E) CDW and the titre and productivity of ectoine obtained by fermentation with J9U, J9U-P8E and J9U-P8EC in MMG3 medium. All strains were cultured at 37 °C and 200 rpm for 60 h. (F) CDW and the titre and productivity of ectoine obtained by fermentation with J9U, J9U-P8E and J9U-P8EC in MMX3 medium. All strains were cultured at 37 °C and 200 rpm for 60 h. The data represent the mean ± standard deviation of triplicate measurements. ** and *** indicate P < 0.01 and P < 0.001, respectively

Full size image

Oxaloacetate is a crucial precursor in the biosynthesis of ectoine [4, 11]. In H. cupida J9, phosphoenolpyruvate carboxylase (encoded by the ppc gene) is responsible for the conversion of phosphoenolpyruvate to oxaloacetate. In this study, the strong promoter P8KT was inserted into upstream of the ppc gene in the genome of J9U-P8E, to generate J9U-P8EC (Figs. 3 and 5B and D). When grown in MMG3 for 60 h, J9U-P8EC, J9U-P8E and J9U showed the similar growth trend and glucose consumption trend and the final CDW of the three strains was also similar at 60 h (Fig. S5). The results from RT-qPCR showed that the transcription levels of the ppc, ectA, ectB and ectC genes in J9U-P8EC were elevated by 2- to 60-fold compared to those detected in J9U (Fig. S6A). Ectoine accumulation by J9U-P8EC was monitored in a 72 h fermentation period, indicating that the highest ectoine titre was obtained at 60 h (Fig. S6B). J9U-P8EC accumulated 5.06 g/L ectoine at 60 h in MMG3, with a 22.82% increase compared to J9U-P8E (Fig. 5E). In summary, ectoine biosynthesis can be enhanced by enhancing the ectoine biosynthesis module and the intracellular supply of the precursor oxaloacetate in J9U-P8EC.

In previous studies, P. putida and Halomonas sp. were engineered to metabolize xylose for PHA production [14, 39, 40]. In MMX3 medium, J9U-P8EC achieved an ectoine titre of 4.12 g/L at 60 h (∼ 68.67 mg/L·h), a 2.03-fold increase compared to J9U (Fig. 5F). The growth trend and xylose consumption rate of the three strains grown in MMX3 were similar in a 60 h fermentation period and the final CDW of J9U was higher than those of J9U-P8E and J9U-P8EC at 60 h (Fig. 5F and S7). Compared to previous studies using H. elongata and M. alcaliphilum with the ectoine production of 53.53 mg/g fresh cell weight and 37.93 mg/g CDW [17, 22], 664.52 mg ectoine/g CDW produced by J9U-P8EC is the highest ectoine production obtained using xylose as the sole carbon source so far (Table 2). All ectoine titre and productivity for J9U and its mutant strains can be found in Table 3.

Table 2 Cell dry weight, ectoine titre, and productivity of three different chassis cells from xylose, glucose and lignocellulosic hydrolysate
Full size table
Table 3 The ectoine titre, productivity and conversion efficiency of J9U and its recombinant strains from different carbon sources
Full size table

Biosynthesis of ectoine from a glucose-xylose mixture and corn straw hydrolysate

To evaluate the capacity of J9U, J9U-P8E and J9U-P8EC to synthesize ectoine from a glucose-xylose mixture, in this study, the three strains were cultured in MMXG3 for 60 h. The growth trend and sugar consumption rate of the three strains grown in MMXG3 were similar in a 60 h fermentation period (Fig. 4). Notably, the ectoine productivity and sugar conversion efficiency of J9U-P8EC were 142.50 mg/L·h and 0.32 g/g, respectively, with a 3.23- and 3.57-fold increase compared to that of J9U (Fig. 4D). These results highlight the capacity of H. cupida J9 to utilize glucose and xylose as the co-carbon sources for ectoine biosynthesis.

Halophilic bacteria can use seawater instead of freshwater as their water source, and their alkaline and high-salt environment eliminates the need for sterilization procedures, thereby reducing the complexity and cost of downstream processing [16]. The use of lignocellulose-rich feedstocks for producing high-value chemicals either fully utilize agricultural waste or reduces the substrate cost of microbial fermentation [19]. In this study, we explored the feasibility of H. cupida J9 in synthesizing ectoine from non-sterilized lignocellulosic biomass. Both J9U and J9U-P8EC were cultured in MML3 for 60 h to evaluate ectoine productivity. Both strains exhibited the similar growth rates in an open fermentation process (Fig. 6A). The ectoine productivity and sugar conversion efficiency of J9U-P8EC reached 21.67 mg/L·h and 0.06 g/g, respectively, with a 1.55-fold and 2.00-fold increase compared to that of J9U (Fig. 6B; Table 3). These results highlight the potential of J9U-P8EC to synthesize ectoine from lignocellulosic biomass in an open fermentation system.

Fig. 6
figure 6

Cell biomass and ectoine production of MML3-grown J9U and J9U-P8EC. (A) Cell growth and total sugar consumption of J9U and J9U-P8EC. (B) CDW and the titre and productivity of ectoine obtained by fermentation with J9U and J9U-P8EC. These strains were cultured in MML3 (xylose: glucose = 3: 1, total 30 g/L) at 37 °C and 200 rpm for 60 h in an open fermentation process. The data represent the mean ± standard deviation of triplicate measurements. *** indicates P < 0.001

Full size image

Conclusions

Under salt stress, the transcription levels of ectoine biosynthesis genes are upregulated in H. cupida J9 and ectoine accumulation is enhanced. Most ectoine molecules are secreted into the culture medium. The constructed J9U-P8EC has the highest productivity of 142.50 mg/L·h obtained by xylose fermentation so far. Moreover, the capacity of J9U-P8EC to synthesize ectoine from either a glucose-xylose mixture or corn straw hydrolysate highlights the potential of this strain to utilize lignocellulose-rich feedstocks for ectoine production. Utilization of inexpensive substrates by J9U-P8EC for open production of ectoine makes J9U-P8EC an ideal producer for large-scale production of ectoine. In the future, the ectoine yield will be further improved by enhancing the tolerance of J9U-P8EC to inhibitors in lignocellulose hydrolysate and optimizing the fermentation process in a bioreactor.