Development of Zn²⁺-controlled expression system for lactic acid bacteria and its application in engineered probiotics

Abstract

Lactococcus lactis and Streptococcus thermophilus are considered as ideal chassis of engineered probiotics, while food-grade genetic tools are limited in those strains. Here, a Zn²⁺-controlled gene expression (ZICE) system was identified in the genome of S. thermophilus CGMCC7.179, including a transcriptional regulator sczA_st and a promoter region of cation transporter czcD (P_czcDst). Specific binding of the SczA_st to the palindromic sequences in P_czcDst was demonstrated by EMSA analysis, suggesting the regulation role of SczA_st on P_czcDst. To evaluate their possibility to control gene expression in vivo, the sczA_st-P_czcDst was employed to drive the expression of green fluorescence protein (GFP) gene in L. lactis NZ9000 and S. thermophilus CGMCC7.179, respectively. Both of the transformants could express GFP under Zn²⁺ induction, while no fluorescence without Zn²⁺ addition. For optimal conditions, Zn²⁺ was used at a final concentration of 0.8 mM in L. lactis and 0.16 mM in S. thermophilus at OD₆₀₀ close to 0.4, and omitting yeast extract powder in the medium unexpectedly improved GFP expression level by 2.2-fold. With the help of the ZICE system, engineered L. lactis and S. thermophilus strains were constructed to secret cytokine interleukin-10 (IL-10) with immunogenicity, and the IL-10 content in the supernatant of the engineered L. lactis was 59.37 % of that under the nisin controlled expression system. This study provided a tightly controlled expression system by the food-grade inducer Zn²⁺, having potential in development of engineered probiotics.

1. Introduction

Lactococcus lactis and Streptococcus thermophilus are worldwide industrial workhorses used for dairy fermentation [1]. They have relatively simple carbon metabolism and high glycolytic flux to produce enough lactic acid for rapid acidification of raw milk, and also acetoin, diacetyl, acetaldehyde as aromatic end-products [2,3]. Due to the long history of safe use in food production, these two strains have ‘‘Generally Recognized as Safe’’ (GRAS) status in USA and a Qualified Presumption of Safety (QPS) status in the European Union [4,5]. In addition, L. lactis and S. thermophilus have potential as probiotics, as demonstrated by various health effects, including transient survival, moderate adherence, anti-inflammatory and immunoregulation in the gastrointestinal tract [6,7]. Thanks to the GRAS status and the known probiotic effects, L. lactis and S. thermophilus are considered as ideally biotechnological chassis to deliver therapeutic proteins in the gastrointestinal tract [2,5,8]. Compared with oral administration or intramuscular injection, delivery of therapeutic proteins into intestinal tract by engineered probiotics are thought to preserve activity to a greater extent and reduce dose-dependent side effects [9,10]. In this context, food grade gene expression systems with tight-control properties are critical tools when achieving this biotechnological application.

Nowadays, the nisin controlled expression (NICE) system is the only commercial food grade expression system for lactic acid bacteria (LAB). The NICE system consists of the inducible promoter P_nisA and the two-component signal transduction components nisR/K belonging to the bacteriocin nisin biosynthesis gene cluster [11]. When a targeted gene is placed behind the P_nisA, efficient gene expression can be induced by the addition of sub-inhibitory amounts of nisin to the culture medium [12]. A lot of merits have been proposed for the NICE system, including the food-grade inducer, high level gene expression as well as wide range hosts (L. lactis, S. thermophilus and Lactobacillus), making this system the most widely used controllable expression system in LAB. However, there remains limitations, such as background expression, the sensitivity of nisin to proteinase and also its cost. With the explaining of molecular mechanism of sugar utilization, material transport and stress response, the relevant genetic elements were adopted to construct expression systems controlled by lactose, xylose, metal ions, temperature or acid in L. lactis [[13], [14], [15], [16]] as well as S. thermophilus [[17], [18], [19], [20], [21]]. Thereinto, the genetic components of metal ions-controlled expression system are relatively simple, and metal ions as inducer are easy to prepare and manageable conditions. What's more, it is possible to avoid transcriptional leakage by using metal ions-controlled expression systems. However, the food grade expression systems under metal ions control are still limited for LAB.

Maintenance of bacterial intracellular homeostasis of metal ions is dependent on metal ion uptake and efflux systems. Researches in pathogenic Streptococcus pneumoniae established a Zn²⁺ efflux system, which consisted of a cation diffusion facilitator CzcD and the adjacent TetR family regulator SczA that activates expression of the czcD gene in the presence of Zn²⁺ [22]. Food grade S. thermophilus has a close phylogenetic relationship with S. pneumoniae [23,24]. The possibility that the similar Zn²⁺ efflux system existing in S. thermophilus is yet to be verified. In this work, putative genetic elements for Zn²⁺ efflux system were found in the genome of S. thermophilus CGMCC7.179, and cloned to develop a new rigorous gene expression system controlled by Zn²⁺ in L. lactis and S. thermophilus. The influences of various conditions on the system were evaluated to obtain the optimal gene expression level. Finally, the system was applied in the expression of IL-10 with immunogenicity by the host cells L. lactis NZ9000 and S. thermophilus CGMCC7.179.

2. Materials and methods

2.1. Bacterial strains and growth conditions

Strains and plasmids used in this work are summarized in Table S1. Escherichia coli was cultured aerobically in Luria Bertani broth at 37 °C. L. lactis NZ9000 [25] and its derivatives were cultivated in M17 broth (Oxoid, Basingstoke, United Kingdom) containing 0.5 % glucose (GM17) at 30 °C under static conditions. S. thermophilus CGMCC7.179 and its derivatives were cultivated in M17 broth containing 1.0 % lactose (LM17) at 42 °C under static conditions. If necessary, chloramphenicol (Sangon, China) was used at 5 μg/mL for L. lactis, 5 μg/mL for S. thermophilus and 10 μg/mL for E. coli, ampicillin (Sangon, China) 100 μg/mL for E. coli. Nisin (Sigma, USA) was added to a final concentration of 10 ng/mL.

2.2. Sequence analysis

The SczA and P_czcD from S. thermophilus CGMCC7.179 were named as SczA_st and P_czcDst. Amino acid sequence of the SczA_st as well as DNA sequence of the P_czcDst were aligned with those from S. pyogenes MGAS2221, S. agalactiae A9 and S. pneumoniae R6, respectively. Multiple-sequence alignments were performed using Clustal W and ESPript 3.0. Protein structure of the SczA_st was modeled by Alphafold2. The promoter prediction was conducted in the website of Softberry (http://www.softberry.com/).

2.3. Plasmid and recombinant strain construction

The primers used in this work are listed in Table S2. The coding sequence of IL-10 fused with the signal peptide Usp45 was synthesized as stated in Table S3. DNA synthesis and sequencing were performed by RuiBotech Biological Biotechnology., Ltd (China). All molecular manipulations were carried out as standard techniques [26]. Plasmid DNA and genomic DNA were extracted according to the instructions of the Plasmid Mini Kit (Omega) and TIANamp Bacteria DNA kit (TIANGEN).

DNA fragment of the sczA_st was PCR amplified from the genome of S. thermophilus CGMCC7.179 with primers SczAF/SczAR, and the product was subcloned into the corresponding sites of the pET-15b. The resultant plasmid pETSczA was transformed into chemically competent E. coli BL21 cells, generating the recombinant strain E. coli/pETSczA.

The P_czcDst-SczA_st fragment was PCR amplified from the genome of CGMCC7.179 with primers SF/PR, and inserted into the compatible sites of the plasmid pNZ8148 [27], generating the basic ZICE vector pNZST. To express gfp gene, the DNA fragment of gfp was PCR amplified from pGFP [28] using primers GF/GR, and inserted into the NcoI/KpnI sites of pNZST and pNZ8148, generating the recombinant plasmids pNZST-gfp and pNZ-gfp, respectively. To deliver human IL-10, DNA fragment of the usp45-IL10 was PCR amplified from the synthetic template with primers IF/IR, and inserted into pNZST and pNZ8148, generating the recombinant plasmid pNZST-IL and pNZ-IL, respectively. The recombinant plasmids were electroporated into the host L. lactis NZ9000 or S. thermophilus CGMCC7.179 to obtain the corresponding recombinant strains according to the previous methods [29,30].

2.4. Purification of the SczA_st protein and gel mobility shift assay (EMSA)

The E. coli/pETSczA was grown in LB broth until OD₆₀₀ reached 0.4. Isopropyl-β-d-thiogalactopyranoside (IPTG) was added to a final concentration of 0.1 mM, and the temperature was lowered to 16 °C for the induction of SczA_st overexpression. After 16 h of induction, the cells were harvested and the SczA_st was purified by the HisTrap FF column (GE Healthcare) according to the manufacturer's instructions.

The gel mobility shift assay was performed as previous description with some modification [31]. 133-bp DNA fragment of the promoter P_czcDst was PCR amplified from the genome of CGMCC7.179 with primers PF/PR2. DNA fragments containing 1 - 62 bp (F1) and 63 - 133 bp (F2) of the P_czcDst were PCR amplified from the plasmid pNZST-gfp using primers M1F/M1R and M2F/M2R, respectively. DNA fragments of F3 and F4 with site mutations in F1 and F2 were PCR amplified from the plasmid pNZST-gfp using primers M3F/M1R and M2F/M3R. The lactate dehydrogenase promoter, used as control, was PCR amplified from the genome of CGMCC7.179 using primers L1F/L1R. 100 ng DNA probes were added to 20 μL EMSA reaction mixtures containing the purified SczA_st protein and binding buffer at a concentration of 0, 25, 50, 75 or 100 ng per reaction. Immediately after incubation for 30 min at room temperature, samples were loaded onto a 6 % nondenaturing polyacrylamide gel. Gel electrophoresis was performed at 100 V for 60 min in 0.5 × Tris-acetate-EDTA (TAE) buffer. The binding complex was visualized by ethidium bromide staining.

2.5. Determination of relative fluorescence unit (RFU)

The recombinant L. lactis and S. thermophilus strains were cultivated overnight at the conditions as mentioned above. The overnight culture was diluted 100-fold in 5 mL of fresh medium. To find proper induction conditions for L. lactis, the concentration of inducer ZnSO₄ was added in 0.2 mM increment from 0 to 1.0 mM, and inductions were initiated at the time of inoculation or OD₆₀₀ reaching 0.2, 0.4, 0.6, 0.8 or 1.0. For S. thermophilus, ZnSO₄ was added from 0 to 0.24 mM. After induction for 4 h or more at 18 °C, the cells were harvested, washed twice and resuspended in phosphate-buffer saline (PBS; 137 mM NaCl, 2.7 mM KCl, 10 mM Na₂HPO₄, 2 mM KH₂PO₄, pH 7.4). Whole cell fluorescence intensity was measured on LS-50B spectrofluorometer (Perkin-Elmer, USA) by excitation at 488 nm and emission at 511 nm. The relative fluorescence unit (RFU) was calculated as fluorescence intensity/OD₆₀₀.

2.6. Western blotting

The recombinant L. lactis strains NZ-ZICEIL and NZ-NICEIL, and recombinant S. thermophilus strain ST-ZICEIL were cultivated and induced for protein expression at the proper Zn²⁺ conditions. Then, the extracellular proteins of each culture were collected as previous method [32]. Protein samples were subjected to SDS-PAGE, and the IL-10 protein was detected by western blot using an anti-IL-10 monoclonal antibody (Proteintech).

2.7. Statistical analysis

Experimental data were reported as the mean ± standard deviation. Statistical significance between treatment and control conditions was assessed by unpaired 2-tailed Student's t-tests. P < 0.05 were considered statistically significant.

3. Results and discussion

3.1. Putative Zn²⁺ efflux system in S. thermophilus

Genome-wide search of S. thermophilus CGMCC7.179, focused on cation diffusion systems, led to the identification of a putative Zn²⁺ efflux system which was composed of two genes transcribed oppositely, a TetR family regulator SczA_st and a cation diffusion facilitator CzcD_st (Fig. 1A). Further search revealed that the Zn²⁺ efflux system was highly conserved and widely equipped in other S. thermophilus strains, such as CNRZ1066, LMG 18311, JIM8232, ND03, MN-ZLW-002, ASCC1275 and S9. Besides, the genetic organization of the putative Zn²⁺ efflux systems from S. thermophilus was similar to that of pathogenic S. pneumoniae R6 [22].

Fig. 1.

Prediction of putative Zn²⁺efflux system from S. thermophilus CGMCC7.179. (A) Genetic organization of the putative Zn²⁺ efflux system from S. thermophilus CGMCC7.179 and that from S. pneumoniae R6. (B) Multiple sequence alignment of the SczA from S. pyogenes MGAS2221 (Spyo), S. agalactiae A9 (Saga), S. thermophilus CGMCC7.179 (Sther), S. pneumoniae R6 (Spneu). The alignment was generated using ClustalW, and the figure was prepared using ESPript. Secondary structural elements of S. agalactiae SczA are shown as α-helices (coils; α1-α10) and residue numbering across the top refers to the S. agalactiae SczA sequence. The blue stars indicate the conserved basic residues R6 and K10. (C) Multiple sequence alignment of the 133 bp sequence of intergenic region between the sczA_st and czcD_st (Sther) with those from S. pyogenes MGAS2221 (Spyo), S. agalactiae A9 (Saga) and S. pneumoniae R6 (Spneu). Motif1 and motif2 are the predicted binding sites for the SczA regulator in Streptococcus. The red stars indicate the palindromic sequences in motif1 and motif2. RBS, ribosome binding site.

Sequence alignment showed that the amino acid sequence of the SczA_st had 74 % identity with that from S. pneumoniae R6, and conserved amino acid residues R6 and K10 existed in the N-terminal region of SczA_st (Fig. 1B). The dibasic residues could enhance DNA binding through contact with the phosphate backbone, which were thought to be vital to transcriptional regulator function [22]. Protein structure of the SczA_st modeled by Alphafold2 predicted an N-terminal DNA-binding domain and a variable C-terminal regulatory domain (Figure S1), suggesting that the SczA_st might act as a transcriptional regulator. The 133 bp sequence of intergenic region between the sczA_st and czcD_st was subjected to promoter prediction, and the promoter elements (-35 region, −10 region and ribosome binding site) were found (Fig. 1C). And two palindromic sequences (motif1 and motif2) were located in the intergenic region, which have been proved to act as the SczA binding sites in S. pneumoniae R6 [22,33]. Therefore, the 133 bp sequence (renamed as P_czcDst) might contain the operator sequence of the czcD_st, that was transcriptional regulated by the SczA_st.

3.2. In vitro binding of the SczA_st to the P_czcDst

To test specific binding of the SczA_st to the P_czcDst, the SczA_st with N-terminal His₆-tag was purified after overexpressed in E. coli. As shown in Fig. 2A, the protein band about 25 kDa was observed, corresponding to the theoretical molecular weight (MW) of SczA_st (179 aa, 21.5 kDa) plus the N-terminal His₆-tag. Using the purified SczA_st protein, EMSA analysis exhibited that the shifts of P_czcDst were gradually retarded along with the increase of SczA_st protein, so were the two truncations P_czcDst-1 lacking motif2 and P_czcDst-2 lacking motif1 (Fig. 2B). These results indicated the specific binding between the SczA_st and the individual DNA fragment, and binding sites existing in both of the two truncations. Moreover, binding of the SczA_st to P_czcDst or P_czcDst-1 resulted in complexes of higher MW, while to P_czcDst-2 yielded the complexes of low MW. This difference might result from the monomeric SczA_st binding to the P_czcDst-2 (containing motif2) while the oligomeric SczA_st binding to the P_czcDst or P_czcDst-1 (containing motif1), as results reported in S. pneumoniae [22]. When the base mutations were introduced into the palindromic sequences of motif1 and moitf2, the interaction of the SczA_st with P_czcDst-1 were greatly weakened, with P_czcDst-2 were abolished, implying the palindromic sequences were crucial for the interaction (Fig. 2C). The shifts of P_czcDst-3 showed a bit of retardation with the increase of SczA_st protein, as the seven bases at the 5’ end of motif1 could act as another binding sites for SczA_st [22]. Therefore, the in vitro EMSA analysis demonstrated that the SczA_st specifically bond to the palindromic sequences in P_czcDst, suggesting a regulation role of SczA_st on P_czcDst.

Fig. 2.

In vitro interaction of the SczA_stwith the P_czcDst. (A) SDS-PAGE of the SczA_st protein (indicated by arrow) purified by HisTrap FF column. (B) EMSA analysis of the binding between the SczA_st and the P_czcDst or its truncations P_czcDst-1, P_czcDst-2. (C) EMSA analysis the SczA_st binding to the P_czcDst-3 and P_czcDst-4 which contained site mutations in the palindromic sequences of motif1 and motif2. The red bars indicate mutant bases. (D) The lactate dehydrogenase promoter was used as a negative control.

3.3. Controlled gfp expression at optimal Zn²⁺ conditions in L. lactis

To find out whether the SczA_st could regulate the expression of gene downstream the P_czcDst in vivo, the sczA_st-P_czcDst was used to drive the gfp expression in L. lactis. As shown in Fig. 3A, the relative fluorescence unit (RFU) was almost same in the strain NZ-ZICEgfp as that in the control strain NZ-ZICE when they were cultured in GM17 broth during 0 h to 6 h, suggesting the gfp was not expressed. When 0.2 mM to 1.0 mM ZnSO₄ was added at the time of inoculation, the RFU in the NZ-ZICEgfp exhibited obvious increase, and the highest value was obtained by 1.0 mM ZnSO₄ for 6 h induction. These results indicated that Zn²⁺ addition could induce the expression of gene under the P_czcDst in L. lactis. The possible mechanism would be proposed that the SczA_st bound to the motif1 but disaggregated from the motif 2 of the P_czcDst by Zn²⁺ addition to generate transcriptional activation of gene downstream P_czcDst, as suggest in S. pneumoniae [22,33].

Fig. 3.

Zn²⁺controlled gfp expression in L. lactis. (A) GFP expression levels of L. lactis NZ-ZICE and L. lactis NZ-ZICEgfp when Zn²⁺ of different final concentrations were added at the time of inoculation. (B) Growth curves of L. lactis NZ-ZICE and L. lactis NZ-ZICEgfp when Zn²⁺ of different final concentrations were added at the time of inoculation. (C) GFP expression levels of L. lactis NZ-ZICEgfp induced by 0.8 mM Zn²⁺ added at the time of inoculation (OD600 ≈ 0) or OD₆₀₀ close to 0.2, 0.4, 0.6, 0.8 or 1.0.

When ZnSO₄ was added at the time of inoculation, growth of the NZ-ZICEgfp was clearly affected by 1.0 mM ZnSO₄, and a lesser extent by 0.6 mM and 0.8 mM (Fig. 3B), as excess of metal ions can be deleterious to the cell [34]. To relieve this adverse impact, ZnSO₄ at a final concentration of 0.8 mM was added to the culture when OD₆₀₀ reached 0.2, 0.4, 0.6, 0.8 or 1.0. The results showed that the growth inhibition was gradually removed with the retardation of Zn²⁺ addition (Figure S2A), and GFP expression level increased by 42.8 % when Zn²⁺ was added at OD₆₀₀ = 0.4 compared with that at the time of inoculation (Fig. 3C).

Using the transcriptional regulator SczA_st and promoter P_czcDst, a Zn²⁺ controlled expression (ZICE) system was constructed in L. lactis. At the optimal Zn²⁺ condition, the expression strength of ZICE was about 45 % of that produced with the NICE system (Figure S2B). Although like this, zinc salt as an inducer could be much cheaper than nisin used for the NICE system [35]. Notably, the control by the ZICE system was rigorous, so it would be favorable to produce proteins toxic to the host without leakage problem.

3.4. Feasibility of the ZICE system in S. thermophilus

To test feasibility of the ZICE system in S. thermophilus, the plasmids pNZST and pNZSTgfp were introduced into the host strain CGMCC7.179, yielding the ST-ZICE (control) and ST-ZICEgfp. As shown in Fig. 4, the two recombinants had similar RFU under 0 mM Zn²⁺ conditions, indicating no leakage occurred in the ST-ZICEgfp. When Zn²⁺ from 0.08 to 0.2 mM was added to the culture of strain ST-ZICEgfp at OD₆₀₀ = 0.4, the RFU increased with the increase of Zn²⁺ levels. Considering that 0.2 mM Zn²⁺ severely inhibited the growth of ST-ZICEgfp, 0.16 mM Zn²⁺ was used to induce gene expression in S. thermophilus. Compared with L. lactis, S. thermophilus showed better sensitivity to Zn²⁺, as 0.16 mM Zn²⁺ induced ST-ZICEgfp and 0.8 mM Zn²⁺ induced NZ-ZICEgfp resulted in similar GFP expression levels, indicating that S. thermophilus showed better response to Zn²⁺. We supposed that S. thermophilus with Zn²⁺ efflux system in the chromosome would enhance the cell sensitivity to Zn²⁺.

Fig. 4.

Zn²⁺controlled gfp expression in S. thermophilus.S. thermophilus ST-ZICEgfp was induced by Zn²⁺ at different final concentrations when OD₆₀₀ close to 0.4. Squares indicate OD₆₀₀.

The adaptability of the ZICE system was also tested in Lactobacillus casei BL23, but it did not work (data not shown). Comparison with the NICE system adapted in L. lactis and Lb. casei, the ZICE system could be applied in L. lactis and S. thermophilus.

3.5. Effects of nitrogen and carbon sources on gene expression level under the ZICE system

To detect effects of nitrogen source on gene expression level under the ZICE system, soya peptone, tryptone, beef extract and yeast extract powder were omitted from the GM17 broth respectively. After the strain NZ-ZICEgfp was cultivated in the broth omitting each of the nitrogen source to OD₆₀₀ close to 0.4, Zn²⁺ at a final concentration of 0.8 mM was added to the culture, and the cell densities as well as RFU were measured. As shown in Fig. 5, the absence of the nitrogen sources resulted in lower biomass after 4 h induction, ranging from 37.5 % (tryptone absent) to 71.8 % (beef extract absent) of the normal GM17 broth. The absence of yeast extract powder elevated the RFU by 2.2-fold compared with the normal GM17 broth. The yeast extract powder plays important roles in the bacterial growth and intracellular metabolism. Unexpectedly, omitting of yeast extract powder enhanced the GFP expression level of the NZ-ZICEgfp induced by Zn²⁺, agreeing with the previous report that nitrogen sources impacted the promoter P_srfA activity in LAB [36]. The reason for this effect needs further experiments.

Fig. 5.

The effects of nitrogen source on GFP expression level under the ZICE system.L. lactis NZ-ZICEgfp was induced by Zn²⁺ at a final concentration of 0.8 mM when OD₆₀₀ close to 0.4. Squares indicate OD₆₀₀.

The effects of carbon sources on gene expression level under ZICE system were also tested by cultivating the NZ-ZICEgfp in the medium containing glucose, fructose, sucrose, lactose and maltose as the sole carbon source. However, the four carbon sources had no obvious effect on GPF expression levels (data not shown).

3.6. IL-10 delivery of engineered LAB under ZICE system

L. lactis is a well-known host for bioproduction of fine chemicals and food ingredients [2]. Besides, recent studies have also focused on the construction of S. thermophilus recombinant strains to improve healthy and functional characteristics of food [5]. Considering their abilities to survive the intestinal passage and to be metabolically active in gastrointestinal tract, L. lactis and S. thermophilus are promising chassis for delivering various biological molecules to the gastrointestinal tract [8]. Here, we applied the ZICE system to direct the delivery of cytokine interleukin-10 (IL-10) by L. lactis and S. thermophilus (Fig. 6A). Western blot using anti-IL-10 monoclonal antibody showed that the bands corresponding to IL-10 could be detected in the supernatant of L. lactis NZ-ZICEIL at 4 h postinduction by Zn²⁺, as well as in the supernatants of L. lactis NZ-NICEIL at 4 h postinduction by nisin (Fig. 6B). Densitometry analysis of the bands indicated that IL-10 content induced by the ZICE system was 59.73 % of that induced by the NICE system. For the recombinant S. thermophilus ST-ZICEIL, IL-10 was detected in the culture supernatant, of which the IL-10 content was 57.47 % of that in the supernatant of L. lactis NZ-ZICEIL (Fig. 6C). These results confirmed that the ZICE system could drive the production of active proteins in LAB hosts. Compared with the widely used NICE system, the ZICE system developed here showed a lower expression level of IL-10. However, nisin, inducer of the NICE system, is sensitive to the enzymolysis of trypsin commonly distributed in the intestinal juice [37]. Zn²⁺, a kind of inorganic ion, would be more stable and suitable for controlling gene expression in the intestinal environment, which might be provided by dietary supplement or rectal injection.

Fig. 6.

IL-10 delivery by engineered LAB under the ZICE system. (A) Schematic representation of the expression cassette of IL-10 under the ZICE system as well as the NICE system. (B) Western blot and densitometric analysis of IL-10 in the 50-fold concentrated extracellular fractions of L. lactis NZ-ZICEIL and L. lactis NZ-NICEIL. (C) Western blot and densitometric analysis of IL-10 in the 100-fold concentrated extracellular fractions of S. thermophilus ST-ZICEIL and L. lactis NZ-ZICEIL. *P < 0.05. In panels (B) and (C), 1, 2, and 3 represent three independent experiments.

4. Conclusion

In this work, a new Zn²⁺ controlled expression (ZICE) system was developed for L. lactis and S. thermophilus. The ZICE system had a great deal to recommend it, since it avoided transcriptional leakage, originated from a food-grade strain and used a cheap and stable inducer. Therefore, the system facilitated to generate recombinant probiotic strains with IL-10 delivery which was thought to cope with intestinal inflammation. Further works are needed such as the improvement of gene expression level controlled by the ZICE system as well as application of the system to produce heterologous proteins in different environments such as intestinal environment.

CRediT authorship contribution statement

Xiaoning Xu: Investigation, Methodology, Visualization. Lingwen Zhang: Investigation, Methodology. Yue Cui: Investigation, Methodology. Jian Kong: Supervision, Resources, Writing – review & editing. Tingting Guo: Conceptualization, Supervision, Funding acquisition, Writing – original draft, Writing – review & editing.

Declaration of competing interest

The authors have declared no conflict of interest.

Appendix A. Supplementary data

The following is the Supplementary data to this article: