Engineered stable, antibiotic-free, high-level protein expression in the probiotic chassis Escherichia coli Nissle 1917

The application of engineered live biotherapeutic products (LBPs) to secrete small molecules, peptides, or proteins to benefit a human or animal host, relies on heterologous protein expression. Key challenges in this area include expressing protein in a targeted location, the use of antibiotic-free platforms, and expressing recombinant proteins at titers capable of the desired therapeutic effect. In this study, we sought to engineer the promising candidate probiotic chassis Escherichia coli Nissle 1917 (EcN) as an in-situ drug delivery platform. Despite its long history of safe human use and general probiotic characteristics, wild-type EcN is not optimal for routine protein expression. In this work, we present several approaches to improve protein production in this host. First, we enable stable antibiotic-free protein expression system via native cryptic plasmids. Next, we integrate the T7 RNA polymerase for high-level protein expression. Finally, we knock out OmpT protease activity, enabling expression levels comparable to the industry standard E. coli BL21 (DE3). To demonstrate its application, the above system was adapted to express antimicrobial peptide microcin L (MccL) from EcN, which can potentially reduce gut related pathogens and enhance fitness of the probiotic in the competitive niche of the gut. Overall, this study establishes an antibiotic free and high-level protein expression platform in EcN, expandable for in situ delivery of therapeutic proteins. 1. INTRODUCTION In recent years, protein-based therapeutics have revolutionized modern medicine for the treatment of various diseases, including but not limited to monoclonal antibodies, vaccines, cytokines, enzymes, growth factors and antimicrobial peptides (AMP) (Leader et al., 2008). These proteins are predominantly produced using recombinant DNA technology, where the gene of interest (GOI) is cloned into a plasmid, transferred into a host for expression. The development of efficient and scalable production of above molecules is limited by various factors such as identifying a suitable expression host, site specific drug delivery, dosage optimization, presence of antibiotic resistance genes and expression at high enough titers for economical downstream separation. To address these challenges, there is growing interest in developing bacterial therapies for targeted drug delivery, i.e., engineered live biotherapeutic products (eLBPs) (Murali and Mansell, 2024). Often based on bacteria with a long history of human use, these live biotherapeutics are generally regarded as safe, can survive harsh gut conditions and can be used for in situ drug delivery targeting the gastrointestinal tract. Escherichia coli Nissle 1917 (EcN) is a well-established probiotic strain with decades of safe documented use in humans. It is licensed as Mutaflor and marketed in Europe and Canada for the treatment of chronic constipation and ulcerative colitis. EcN possess various health benefits such as anti-inflammatory effects, immunomodulation and inhibition of pathogens (Sonnenborn, 2016). The wide availability of E. coli genetic tools and its compatibility with EcN make it a promising probiotic strain for protein expression. Several reports suggest the use of engineered EcN for the expression of therapeutic proteins, peptides, and antigens for vaccine delivery (Lynch et al., 2022), with one engineered strain reaching a Phase 2 clinical trial (Vockley et al., 2023). In most cases, an extrachromosomal expression vector, e.g. a plasmid, facilitates the production of a target protein. However, many plasmids require antibiotic resistance genes for long-term maintenance, which restricts their application as bacterial therapy in vivo . Previously, Kan et al. sought to mitigate the above challenges using EcN’s native cryptic plasmid pMUT1 and pMUT2 as expression vectors for protein production (Kan et al., 2021). We also tested the stability of gene expression in EcN from engineered cryptic plasmids over many generations without antibiotic selection and developed a strain cured of native plasmids using CRISPR-Cas9 (EcNc) (Zainuddin et al., 2019). The prevalence of antibiotics poses a major challenge in the evolution of antibiotic-resistant organisms, and broad-spectrum antibiotics can disrupt the microbiome composition causing dysbiosis (Patangia et al., 2022). As an application to expand our protein production system, we chose the secretion of antimicrobial peptides (AMPs). Engineering probiotics to express AMPs serves as an efficient and scalable system for in situ delivery (Cesaro et al., 2023). Current and past efforts have attempted to engineer EcN for the expression of AMPs against various pathogenic infections. Examples include the secretion of human beta-defensin to target infection from Salmonella enterica and Listeria monocytogenes (Seo et al., 2012) and enterocin A, enterocin B and Hiracin JM79 to target vancomycin-resistance Enterococcus pathogen (Geldart et al., 2018). Microcins (Mcc), including MccB, MccC, MccJ, MccL, MccV, and others, inhibit the growth of competing bacteria. Among them, we were interested in exploring the potential of MccL as it exhibits strong antibacterial activity against Enterobacteriaceae, including Salmonella enterica (Geldart et al., 2016) . We sought to build on this work by engineering EcN for even higher titers of MccL secretion and production. Inspired by BL21(DE3), the industry gold standard for protein expression, several works have adapted the RNA polymerase from T7 phage (T7RNAP) for use in EcN either by co-expression of T7RNAP on a plasmid (Ba et al., 2024) or integration into the genome (Effendi and Ng, 2022). We integrated the T7RNAP gene under the control of the LacUV5 promoter into the EcNc genome, which led to more stable expression than under plasmid co-expression of the polymerase. Co-expression of the LacI repressor enhanced the overall number of cells expressing the target protein as well. We also found that after several generations, while both pMUT1 and pMUT2 cryptic plasmids were maintained in a large majority of cells, engineered pMUT2 had higher stability when compared with pMUT1. In addition, we sought to increase expression by mimicking the protease knockout present in the BL21 expression strain, namely the OmpT outer membrane protease. Collectively this work establishes a protein expression platform leveraging EcN’s native plasmids to express GOI, building engineered EcN-V2 as a promising probiotic model strain for antimicrobial peptide secretion. 2. MATERIALS AND METHODS 2.1 Bacterial strains, plasmids, and culture conditions. E. coli strains and plasmid used in this study are listed in table 1. E. coli DH5α (cloning host), E. coli BL21 (protein expression host), EcN (wild type probiotics) and EcNc (native plasmid cured) as a probiotic host are predominantly investigated in this study. E. coli strains were cultured in Luria-Bertani broth (BD Difco™ LB Franklin Lakes, NJ, USA) at 37 –ͦ C or 30 –ͦ C and incubated at 250 rpm for 24 hours. Kanamycin (50 µg/mL), Chloramphenicol (25 µg/mL), Ampicillin (100 µg/mL), spectinomycin (50 µg/mL) and Isopropyl-β-d-thiogalactoside (IPTG) (0, 0.1, 1 mM) working concentrations were supplemented to the culture broth when needed. 2.2 Antibiotics free protein expression using cryptic plasmids The schematic representation of antibiotics free protein expression platform is shown in figure1. The pMUT1 and pMUT2 cryptic plasmids were isolated from overnight wild-type EcN cultures using QIAprep® Spin Miniprep Kit (QIAGEN GmbH, Hilden, Germany) according to the manufacture’s instruction. Cryptic plasmids were linearized by PCR primers listed in supplementary table1. The gene of interest, mScarlet, was constructed under the constitutive promoter J23100. The cat antibiotics resistance gene flanked by frt regions was cloned into pMUT1/2 plasmids using Gibson assembly as shown in supplementary figure 3a. The engineered plasmid was transformed into EcNc via electroporation. EcNc cells were subcultured and harvested at an OD 600 0.5. 2 mL of cells was collected, centrifuged at 10000 rpm for 1 min, the supernatant was discarded, and 10% ice-cold glycerol was used to wash the cells. The final cell pellet is concentrated to 50 uL and 1uL of plasmid was added (100 ng/uL). The electrocompetent EcNc cells with plasmid were transformed into 1mm sterile electroporation cuvette (pre chilled in ice) and electroporated at 1800V using Bio-Rad GenePulser Xcell TM (Bio-Rad Laboratories, Hercules, CA, USA.). After electroporation, immediately 950 uL of SOC media was added and incubated at 250 rpm for 1 hours in 37 –ͦ C. 50 uL of cells were plated in LB agar with respective antibiotics and incubated overnight. After pMUT1/2 transformation, pCP20 was also co-transformed to remove antibiotic cassette via flp recombinase expression (Cherepanov and Wackernagel, 1995). Finally, the cells were cultured at 42 –ͦ C to cure the pCP20 plasmid as represented in figure 1a. Cured clones were replica plated on kanamycin, ampicillin, and LB agar plates to confirm removal of the KanR cassette and pCP20 plasmid. 2.3 Genome engineering of EcN for protein expression The EcNc genome was engineered to develop T7 RNAP based protein expression system. To generate EcN-V1, EcNc was first transformed with pKD46 (helper plasmid with λ red recombination genes) as show in Figure 2a (Datsenko and Wanner, 2000). For genomic insertion, the mgsA locus in the EcNc genome was targeted. Linear double standard DNA template was constructed with T7 RNA polymerase gene amplified from E. coli BL21 DE3. To facilitate the double cross, the insert template had 500 base pair of mgsA homology region flanking to the right and left side of T7RNAP (Datsenko and Wanner, 2000). Further, to improve the efficiency of protein expression EcN-V2 was developed from EcN-V1. Similar to above procedure lacI was also inserted downstream to T7 RNAP. CRISPR cas9 system was used to delete OmpT protease from the EcN-V1 genome. Two plasmids, pCas (#62225) and pTargetF (#62226) were purchased from Addgene (Watertown, MA, USA). The N20 targeting the mgsA locus was constructed in pTargetF using Gibson assembly. Initially, the pCas plasmid was transformed into EcN-V1 and glycerol stocked at -80 –ͦ C. Later, the electrocompetent EcN-V1 cells containing pCas were transformed with pTargetF_OmpT and linear double standard template consisting of 1000 base pairs (500 base pair upstream and downstream of ompT gene) as the homology region (Jiang et al., 2015). Kanamycin-resistant recombinants were isolated and the knockout of ompT was confirmed by colony PCR and whole-genome sequencing (Plasmidsaurus, Louisville, KY). 2.4 Flow cytometry sample preparation All E. coli (BL21, EcN, EcN-V1 and EcN-V2) strains were initially grown overnight in 5ml LB broth. 1% inoculum from overnight cultures was sub cultured and incubated at 37 ºC until the optical density (OD₆₀₀) reached 0.5–0.6. IPTG was then added to the culture to a final concentration of 1mM. After induction, cells were incubated overnight for 24 hours at 250 rpm in 37 –ͦ C. Flow cytometry samples were prepared by diluting 10 µL of cells in 490 µL of PBS buffer. Samples were then analyzed at the Iowa State Flow Cytometry Facility using a BD FACS Canto Plus RUO (San Jose, CA) flow cytometer equipped with a 488 nm, 20 mW solid-state blue laser and a 505 nm long-pass and 525/50 nm band-pass filter set. Statistical analysis of mean fluorescence was performed using an unpaired, two-tailed Student’s t-test and p-values were reported. 2.5 Anti-microbial secretion from EcN-V2 The gene encoding microcin MccL and the cvaAB transporter genes were amplified from pMPES-mccL plasmid (Geldart et al., 2016) and cloned in pET28a vector under T7 promoter. EcN-V2 was transformed with pET28a:mccL (plasmid map shown in supplementary figure 3c) according to the protocol in section 2.2. LB agar plates (1.5 % agar) were supplemented with a final working concentration 50 µg/mL of kanamycin and 0.1 mM of IPTG. E. coli DH5α and EcN was transformed with empty pET28a vector. Other strains, BL21 and EcN-V2 transformed with pET28a:mccL was cultured overnight for 24 hours. 1% inoculum was subculture, and the cell were harvested at OD₆₀₀ 0.5. 50 µL of DH5α cells with empty pET28 vector was evenly spread in LB agar plates, dried for 5 minutes in laminar biosafety hood. Spots of 5 µL of EcN with empty pET28a, BL21 and EcN-V2 with pET28a:mccL plasmid were placed on the LB agar plate. After drying, the LB plates were incubated at 37 –ͦ C for 24 hours and the width of the halo was measured. 3. RESULTS & DISCUSSION We aimed to determine whether the proteins of interest could be stably expressed in EcN without the need for antibiotic selection. To monitor protein expression and plasmid stability, EcN’s native plasmids pMUT1 and pMUT2 were engineered to express the fluorescent protein reporter mScarlet. Two different plasmids, pMUT1-mScarlet- cat and pMUT2-mScarlet- cat were developed to test their stability for protein expression. The mScarlet gene in the cryptic plasmid was cloned under the control of strong constitutive promoter pJ23100 along with the chloramphenicol acetyltransferase ( cat ) flanked with frt sites for use in cloning. The plasmids were successfully transformed into EcNc and bright pink color colonies were observed. The cat gene was then cured from the plasmids in vivo using pCP20 to express the FLP recombinase enzyme, which recognized the frt sites as shown in Figure 1a. We observed complete removal of cat gene from all colonies plated on respective antibiotic agar plates, a day after the transformation, indicating high efficiency of FLP recombinase as reported in previous literatures (Datsenko and Wanner, 2000). After the standard three-day process of pCP20 recombination (transformation, growth at 30ºC, followed by curing via growth at 42 ºC), over which we estimate more than 100 generations (cell doublings), the cells were serially diluted, and 40 colonies were streaked in LB agar plate with no antibiotics (Figure 1b). All 40 colonies transformed with pMUT2-mScarlet in EcNc were observed to be pink in color, representing stable expression of mScarlet without selection (Fig 1b). 37 out of 40 colonies transformed with pMUT1-mScarlet remained fluorescent. In addition, we also observed differential mScarlet expression in colonies whose intensity had colony-to-colony variation ranging from slight pink to intense pink. Taken together, we found that while both plasmids were stable after many generations, pMUT2 had better plasmid maintenance and more consistent expression than pMUT1. This is in agreement with previous reports that the presence of a relB-relE toxin-antitoxin system on pMUT2 promotes plasmid stability (Kan et al., 2021). Inspired by the industry standard E. coli BL21 T7 protein expression platform, we aimed at engineering EcNc for improved heterologous protein expression. Based on our observation in Figure 1b, we chose pMUT2 as a stable vector for protein expression. We first cloned the T7 RNA polymerase in pMUT2 under the LacUV5 promoter (as in BL21) and transformed the plasmid into EcNc that with cryptic plasmids removed via CRISPR-Cas9 as previously demonstrated by our lab (Zainuddin et al., 2019). EcNc-pM2-T7 was further transformed with pET28:GFP to test the efficiency of protein expression. The result observed in Supplementary Figure 1 shows a low level of expression, likely due to high leaky expression of T7RNAP from multiple copies of the gene in pMUT2 or metabolic burden. We hypothesize that leaky T7RNAP expression leads to increased cell stress by high-level transcription of the target GFP mRNA, creating a selective pressure for cells to either shed the T7RNAP or GFP-encoding plasmids. To overcome the above limitation, the pMUT1 vector was cloned with the lacI gene and transformed into EcNc-pM2-T7. Increasing the production of LacI has been shown to control the leakiness of T7 RNA Polymerase (Matthews and Nichols, 1998). EcNc co-transformed with pM2-T7 and pM1-LacI had 2.5-fold higher protein expression when compared to EcNc-pM2-T7, but still much lower levels than BL21. To reduce leakiness by further reducing the copy number, the T7RNAP cassette from BL21 was amplified and integrated into the EcNc genome to increase gene stability via the standard method of Datsenko and Wanner as shown in Figure 2a. The mgsA gene was identified as a safe site for genomic insertion, as its mutant variant displayed a growth profile similar to the wild type across different carbon source, highlighting no effects on cellular fitness (Tötemeyer et al., 1998). The wild type EcN transformed with pET28:GFP did not exhibit detectable fluorescence. In contrast, the engineered strain with T7RNAP (EcN-V1) integrated in the genome displayed a green fluorescent signal, indicating successful GFP expression, as shown in Figure 2c, 20-fold higher than the wild-type control. However, the level of GFP observed was lower than that of the BL21 strain. We hypothesized that the low GFP expression levels were due to the high susceptibility of T7RNAP to proteolytic cleavage (Grodberg and Dunn, 1988). Previous literature showed that the OmpT protease can limit T7RNAP efficiency and deleting the protease from the E. coli K12 genome helped increase protein expression (Waegeman et al., 2013). In this study, CRISPR-Cas9-assisted recombineering was used to delete the OmpT protease from EcN (Jiang et al., 2015). The final strain, named EcN-V2, showed 3.3-fold improvement in GFP expression when compared to EcN-V1 as shown in Figure 2c. In addition, the flow cytometry data of E. coli cells expressing GFP was also analyzed. On comparing the percentage of GFP expressing cell between BL21 and EcN-V1, we found only 27% of cells in EcN-V1 to be GFP positive, possibly indicating cell stress or shedding of the pET28 plasmid or genome-level mutations in the promoter or gene encoding T7RNAP. Literature reports suggest that high activity of T7RNAP can have a negative effect on plasmid stability, protein expression and cell growth (Vethanayagan and Flower, 2005). The lacUV5 promoter was reported to be leaky and may produce T7RNAP when not induced, the overproduction of which was linked to metabolic burden of the cell (Snoeck et al., 2024). Further, this characteristic was observed when EcN-V1 was induced at 1mM IPTG or above as this level of induction had a negative effect on cell growth (data not shown). To overcome these challenges, the lacI gene was integrated into the genome. Figure 2d shows that introducing lacI in the EcN genome increases the GFP-positive cells from 27% in EcN-V1 to 91% in EcN-V2. Preventing the leakiness of T7RNAP aids in reducing metabolic burden and increases plasmid stability. Hence, integrating T7RNAP, deleting OmpT proteases, and integrating the LacI repressor into the genome helped us in building EcN-V2 as a robust system for protein expression. This agrees with previous literature in which OmpT has been shown to cleave T7 RNA polymerase, limiting protein production (Redenti et al., 2024). Finally, EcN-V2 was adapted to express another heterologous target, the antimicrobial peptide microcin L, which has broad activity against several bacteria including pathogens coli ) that might compete with EcN for a niche in the gut (Geldart et al., 2016). The pET28a:mccL plasmid was constructed to express the antimicrobial peptide along with secretory transporter CvaB and accessory protein CvaA under the control of T7 promoter as shown in supplementary figure 3c. The plasmid was transformed in BL21 (positive control), EcN (negative control) and EcN-V2 to test the efficiency of killing via a halo assay. The zone of inhibition from EcN-V2-pET28a:mccL was 16.7 mm in diameter shows comparable results with industry standard BL21, whose diameter was 16.8 mm (Figure 3a). Antimicrobial activity from BL21-pET28a:mccL and EcN-V2-pET28a:mccL was observed in both the plates with and without IPTG. As leakiness was still present (observed a halo without induction), this platform could also be further expanded for inducer-free expression, when needed, as we observed strong antimicrobial activity of MccL in Figure 3b. Our work demonstrates a robust, antibiotic free system in EcN by leveraging native plasmids for stable gene expression. Integrating T7RNAP into the EcN genome enables protein expression comparable to the industry standard BL21. Overall, our study establishes the potential of EcN as a bioengineered chassis for antimicrobial production, which could be further expanded for therapeutic protein, vaccine and monoclonal antibodies. ACKNOWLEDGMENTS This work was funded by a grant from the National Institutes of Health (R35GM143074) to TJM. HSZ was partially supported by the Ministry of Education of Malaysia. TJM is additionally partially supported by the Karen and Denny Vaughn Faculty Fellowship. Figures were created using Biorender under license. The pAW-mScarlet plasmid was a gift from Dr. Laura Jarboe. The pET28:GFP plasmid was a gift from Dr. Matthew Bennett. We also thank Dr. Shawn Rigby of the Iowa State Flow Cytometry facility for supporting fluorescence experiments. Finally, this work is dedicated to the memory of Dr. Yiannis Kaznessis, who provided mentorship and guidance in this area along with the pMPES-mccL vector.

Ba F, Zhang Y, Ji X, Liu WQ, Ling S, Li J. 2024. Expanding the toolbox of probiotic Escherichia coli Nissle 1917 for synthetic biology. Biotechnol J 19 :2300327. Cesaro A, Lin S, Pardi N, de la Fuente-Nunez C. 2023. Advanced delivery systems for peptide antibiotics. Adv Drug Deliv Rev 196 :114733. Cherepanov PP, Wackernagel W. 1995. Gene disruption in Escherichia coli: TcR and KmR cassettes with the option of Flp-catalyzed excision of the antibiotic-resistance determinant. Gene 158 :9–14. Datsenko KA, Wanner BL. 2000b. One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products. Proc Natl Acad Sci U S A 97 :6640–6645. Effendi SSW, Ng IS. 2022. Reprogramming T7RNA Polymerase in Escherichia coli Nissle 1917 under Specific Lac Operon for Efficient p-Coumaric Acid Production. ACS Synth Biol 11 :3471–348. Geldart K, Forkus B, McChesney E, McCue M, Kaznessis YN. 2016. pMPES: A Modular Peptide Expression System for the Delivery of Antimicrobial Peptides to the Site of Gastrointestinal Infections Using Probiotics. Pharmaceuticals 2016, Vol. 9, Page 60 9 :60. Geldart KG, Kommineni S, Forbes M, Hayward M, Dunny GM, Salzman NH, Kaznessis YN. 2018. Engineered E. coli Nissle 1917 for the reduction of vancomycin-resistant Enterococcus in the intestinal tract. Bioeng Transl Med 3 :197–208. Grodberg J, Dunn JJ. 1988b. ompT encodes the Escherichia coli outer membrane protease that cleaves T7 RNA polymerase during purification. J Bacteriol 170 :1245–1253. Jiang Y, Chen B, Duan C, Sun B, Yang J, Yang S. 2015. Multigene editing in the Escherichia coli genome via the CRISPR-Cas9 system. Appl Environ Microbiol 81 :2506–2514. Kan A, Gelfat I, Emani S, Praveschotinunt P, Joshi NS. 2021a. Plasmid Vectors for in Vivo Selection-Free Use with the Probiotic E. coli Nissle 1917. ACS Synth. Biol. 10 :94–106. Leader B, Baca QJ, Golan DE. 2008. Protein therapeutics: a summary and pharmacological classification. Nature Reviews Drug Discovery 2007 7:1 7 :21–39. Lynch JP, Goers L, Lesser CF. 2022. Emerging strategies for engineering Escherichia coli Nissle 1917-based therapeutics. Trends Pharmacol Sci 43 :772–786. Matthews KS, Nichols JC. 1998. Lactose repressor protein: functional properties and structure. Prog Nucleic Acid Res Mol Biol 58 :127–150. Murali SK, Mansell TJ. 2024. Next generation probiotics: Engineering live biotherapeutics. Biotechnol Adv 72 :108336. Patangia D V., Anthony Ryan C, Dempsey E, Paul Ross R, Stanton C. 2022. Impact of antibiotics on the human microbiome and consequences for host health. Microbiologyopen 11 :e1260. Redenti A, Im J, Redenti B, Li F, Rouanne M, Sheng Z, Sun W, Gurbatri CR, Huang S, Komaranchath M, Jang Y, Hahn J, Ballister ER, Vincent RL, Vardoshivilli A, Danino T, Arpaia N. 2024. Probiotic neoantigen delivery vectors for precision cancer immunotherapy. Nature 635 :453. Seo EJ, Weibel S, Wehkamp J, Oelschlaeger TA. 2012. Construction of recombinant E. coli Nissle 1917 (EcN) strains for the expression and secretion of defensins. Int J Med Microbiol 302 :276–287. Snoeck S, Guidi C, De Mey M. 2024. “Metabolic burden” explained: stress symptoms and its related responses induced by (over)expression of (heterologous) proteins in Escherichia coli. Microbial Cell Factories 2024 23:1 23 :1–19. Sonnenborn U. 2016. Escherichia coli strain Nissle 1917-from bench to bedside and back: history of a special Escherichia coli strain with probiotic properties. FEMS Microbiol. Lett. 363 . Tötemeyer S, Booth NA, Nichols WW, Dunbar B, Booth IR. 1998. From famine to feast: the role of methylglyoxal production in Escherichia coli. Mol Microbiol 27 :553–562. Vethanayagan JGG, Flower AM. 2005. Decreased gene expression from T7 promoters may be due to impaired production of active T7 RNA polymerase. Microb Cell Fact 4 :3. Vockley J, Sondheimer N, Puurunen M, Diaz GA, Ginevic I, Grange DK, Harding C, Northrup H, Phillips JA, Searle S, Thomas JA, Zori R, Denney WS, Ernst SL, Humphreys K, McWhorter N, Kurtz C, Brennan AM. 2023. Efficacy and safety of a synthetic biotic for treatment of phenylketonuria: a phase 2 clinical trial. Nature Metabolism 2023 5:10 5 :1685–1690. Waegeman H, De Lausnay S, Beauprez J, Maertens J, De Mey M, Soetaert W. 2013. Increasing recombinant protein production in Escherichia coli K12 through metabolic engineering. N Biotechnol 30 :255–261. Zainuddin HS, Bai Y, Mansell TJ. 2019a. CRISPR-based curing and analysis of metabolic burden of cryptic plasmids in Escherichia coli Nissle 1917. Eng. Life Sci. 19 :478–485 GRAPHICAL ABSTRACT Figure 1: a) Schematic representation of antibiotic-free protein expression platform using EcN cryptic plasmids b) EcNc pMUT1/2-mScarlet colonies plated on LB media with respective antibiotics to confirm the removal of both cat gene and pCP20 vector. Non-fluorescent colonies are indicated with an asterisk (*). Figure 2: Schematic representation of a) genomic insertion of T7 RNA polymerase in EcNc at mgsA locus by recombineering. b) Deletion of ompT from EcNc. c) Fluorescence measurements by flow cytometry of various EcNc strains. Data reported are mean values of green fluorescence by cells (averaged over three replicates, error bars represent standard deviation). (d) GFP positive cells (%) from each strain (average of three replicates, error bars represent standard deviation) as compared to non-fluorescent negative control (EcN wild-type) and positive control (BL21). Statistical comparisons were performed using student t-tests, two-tailed (single asterisk represent p < 0.05, double asterisk p < 0.01 and triple asterisk p < 0.001). Figure 3: Antimicrobial activity of engineered EcN expressing MccL. Agar diffusion assay representing the zone of inhibition on DH5α (lawn) cells by engineered EcN-pET28a (negative control), BL21-pET28a:MccL (positive control), and EcN_V2-pET28a:MccL under a) non-inducing (0 mM IPTG) and b) inducing (0.1 mM IPTG) conditions. (c) Quantitation representation of anti-microbial activity in terms of halo diameter (mm). Statistical comparisons were performed using student t-tests, two-tailed (single asterisk represent p < 0.05, double asterisk p < 0.01 and triple asterisk p < 0.001). Table 1. Strains/Plasmids used in this study | E. coli Nissle 1917 (EcN) | E. coli O6:K5:H1 | (Zainuddin et al., 2019) | | E. coli Nissle 1917-cured (EcNc) | E. coli O6:K5:H1 with pMUT1 and pMUT2 plasmid cured. | (Zainuddin et al., 2019) | | E. coli NEB 5-alpha (DH5α) | fhuA2 Δ(argF-lacZ)U169 phoA glnV44 Φ80 Δ(lacZ)M15 gyrA96 recA1 relA1 endA1 thi-1 hsdR17 | New England Biolabs | | E. coli BL21 DE3 (BL21) | fhuA2 [lon] ompT gal (λ DE3) [dcm] ∆hsdS λ DE3 = λ sBamHIo ∆EcoRI-B int::(lacI::PlacUV5::T7 gene1) i21 ∆nin5 | New England Biolabs | | pMUT1-mScarlet- cat | pMUT1 with mScarlet expression under P J23100 , cat genes in between FRT recognition sites, Cm R | This work | | pMUT2-mScarlet- cat | pMUT2 with mScarlet expression under P J23100 , cat genes in between FRT recognition sites, Cm R | This work | | pCP20 | Contains FLP recombinase, Amp R | This work | | pKD46 | Contains exo, beta and gam genes under arabinose promoter for λ red homologous recombination, Amp R | This work | | pCas | Cas9 expression, λ red recombination genes under arabinose expression, Kan R | (Jiang et al., 2015) | | pTargetF- OmpT | N20 gRNA targeting OmpT gene in EcN, Kan R | This work | | EcNc-pM2-T7 | EcNc with pMUT2-cat-T7RNAP plasmid | This work | | EcNc-pM2-T7 + pM1- lacI | EcNc with pMUT2-cat-T7RNAP plasmid and pMUT1-spec- lacI plasmid. | This work | | EcN-V1 | EcNc ∆ mgsA ::(P LacUV5 ::T7RNAP ) | This work | | EcN-V2 | EcNc ∆ mgsA ::(P LacUV5 ::T7RNAP :: lacI ) ∆ ompT | This work | | pET28a-GFP | pET28a plasmid with gfp gene under T7 promoter, Kan R | This work | | BL21-pET28a:mccL | BL21 transformed with pET28a:mccL, Kan R | This work | | EcN-V2-pET28a:mccL | EcN-V2 transformed with pET28a:mccL, Kan R | This work | Information & Authors Information Version history Peer review timeline Published Biotechnology and Bioengineering Version of Record24 Dec 2025Published Copyright This work is licensed under a Non Exclusive No Reuse License.

Authors Metrics & Citations Metrics Article Usage 894views 387downloads Citations Download citation Halimatun Sakdiah Zainuddin, Sanjeeva Kumar Murali, Thomas Mansell. Engineered stable, antibiotic-free, high-level protein expression in the probiotic chassis Escherichia coli Nissle 1917. Authorea. 25 May 2025. DOI: https://doi.org/10.22541/au.174816201.11324415/v1 DOI: https://doi.org/10.22541/au.174816201.11324415/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu.