Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Metabolic engineering of Bacillus subtilis toward the efficient and stable production of C30-carotenoids

Metabolic engineering of Bacillus subtilis toward the efficient and stable production of... Commercial carotenoid production is dominated by chemical synthesis and plant extraction, both of which are unsustainable and can be detrimental to the environment. A promising alternative for the mass production of carotenoids from both an ecological and commercial perspective is microbial synthesis. To date, C carotenoid production in Bacillus subtilis has been achieved using plasmid systems for the overexpression of biosynthetic enzymes. In the present study, we employed a clustered regularly interspaced short palindromic repeat-Cas9 (CRISPR-Cas9) system to develop an efficient, safe, and stable C carotenoid-producing B. subtilis strain, devoid of plasmids and antibiotic selection markers. To this end, the expression levels of crtM (dehydrosqualene synthase) and crtN (dehydrosqualene desaturase) genes from Staphylococcus aureus were upregulated by the insertion of three gene copies into the chromosome of B. subtilis. Subsequently, the supply of the C carotenoid precursor farnesyl diphosphate (FPP), which is the substrate for CrtMN enzymes, was enhanced by expressing chromosomally integrated Bacillus megaterium-derived farnesyl diphosphate synthase (FPPS), a key enzyme in the FPP pathway, and abolishing the expression of farnesyl diphosphate phosphatase (YisP), an enzyme responsible for the undesired conversion of FPP to farnesol. The consecutive combination of these features resulted in a stepwise increased production of C carotenoids. For the first time, a B. subtilis strain that can endogenously produce C 30 30 carotenoids has been constructed, which we anticipate will serve as a chassis for further metabolic engineering and fermentation optimization aimed at developing a commercial scale bioproduction process. Key points • Overexpression of chromosomally integrated crtMN genes improved C carotenoid production • Overexpression of FPPS and branch pathway attenuation further enhanced C carotenoid yield • A stable plasmid-less, marker-less C carotenoid-producing B. subtilis strain was constructed Keywords B. subtilis, C carotenoids, CRISPR-Cas9, Metabolic engineering † Faculty of Pharmacy and Food Science Technology, Department Oriana Filluelo and Jordi Ferrando have contributed equally to this of Biology, Healthcare and the Environment, Microbiology Section, work. University of Barcelona, Avinguda Joan XXIII, 27-31, Barcelona *Correspondence: 08028, Spain Pere Picart perepicart@ub.edu © The Author(s) 2023. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. Filluelo et al. AMB Express (2023) 13:38 Page 2 of 11 Introduction CrtM (dehydrosqualene synthase), which catalyzes the Terpenoids (also known as isoprenoids) constitute one of head-to-head condensation of two molecules of FPP to the largest and structurally most diverse groups of natu- dehydrosqualene. The enzyme CrtN (dehydrosqualene ral products with diverse biological functions (Zhang desaturase) then converts dehydrosqualene to the yel- and Hong 2020). An economically important class of low C carotenoid, DNP, a relatively unstable compound terpenoids are the carotenoids, which are ubiquitous that can suffer further oxidation by CrtMN to yield DLP. lipid-soluble pigments responsible for the red, yellow, The action of these two enzymes probably constitutes the and orange colors of plants, algae, fungi, and bacteria most common route of C carotenoid biosynthesis in (Cardoso et al. 2017). Although commercial carotenoid bacteria. Notably, these yellow pigments have attracted production is dominated by chemical synthesis and plant interest from the pharmaceutical industry owing to their extraction, these processes are not sustainable or ecologi- powerful antioxidant activities (Yoshida et al. 2009), as cal. Carotenoids are chemically synthesized under harsh well as their role as immunomodulators, significantly conditions, generating byproducts and hazardous waste, enhancing the immune system (Jing et al. 2017, 2019; Liu whereas sourcing carotenoids from plant extracts is gen- et al. 2016, 2017). Consequently, microbial cell engineer- erally dependent on the seasons and geographic areas, ing approaches aimed at improving C carotenoid yields which cannot always be standardized (Siziya et al. 2022). are required to achieve industrial-scale production. Therefore, microbial production is emerging as one of To date, the metabolic engineering of B. subtilis toward the most promising safe and environmentally friendly enhanced C carotenoid production has focused on options to satisfy the fast-growing demands for carot- using two-plasmid systems comprising pHY_crtMN enoids (Siziya et al. 2022). (Yoshida et al. 2009), mediating crtMN gene overexpres- B. subtilis is generally recognized as safe (GRAS), has sion under tetracycline selection, and xylose-inducible a high growth rate, and is easy to genetically manipulate pHCMC04G (Xue et al. 2015), mediating stable over- and cultivate, with a wide substrate range (Earl, 2008; expression of all MEP pathway enzymes under non- Schallmey et al., 2004). In addition, it is one of the high- selection conditions (Abdallah et al. 2020). However, est producer of isoprene (the smallest terpenoid) among two-plasmid systems may impose a metabolic burden eubacteria, thus constituting an ideal microbial host for on the host cells, leading to lower growth rates and use as a terpenoid cell factory (Kuzma et al. 1995; Wagner increased productivity costs (Wu et al. 2016). Another et al. 2000; Julsing et al. 2007; Moser and Pichler 2019; drawback is the high-cost of the inducer compounds Guan et al. 2015). This bacterium is able to initiate ter - and, more importantly, the requirement for antibiotic penoid biosynthesis from simple carbon sources through usage, which is restricted by governmental regulations the methylerythritol 4-phosphate (MEP) pathway, a route and can thus hinder the establishment of a commercially with eight enzymatic reactions leading to the synthesis viable industry. On the other hand, very little work has of isopentenyl diphosphate (IPP; C5) and dimethylallyl been done to explore the effects of modulating crtMN diphosphate (DMAPP; C5), the universal precursors of all gene expression and other competing branch pathways terpenoids (Guan et al. 2015). The consecutive condensa - (which can limit FPP availability) on C carotenoid pro- tion of IPP and DMAPP is catalyzed by prenyl diphos- duction, leaving room for improvement. In this study, phate synthase (IspA) to produce starting precursors for we initially compared the expression levels of plasmid- the synthesis of different classes of terpenoids: geranyl based and chromosomally integrated crtMN genes, and diphosphate (GPP; C10), a monoterpenoid precursor; then implemented CRISPR-Cas9-based metabolic engi- farnesyl diphosphate (FPP; C15) for the production of neering strategies to achieve an efficient C carotenoid- sesquiterpenoids, triterpenoids and C -carotenoids, and producing strain of B. subtilis, a bacterium that naturally geranylgeranyl diphosphate (GGPP; C20), the precur- produces yellow pigments (Fig.  1). u Th s, with the aim of sor of diterpenoids and carotenoids (Moser and Pichler increasing the supply of the carotenoid precursor FPP, we 2019). Most carotenoids contain a 40-carbon backbone planned (i) to introduce a chromosomally integrated copy (C carotenoids), including β-carotene, lycopene and of FPPS (farnesyl diphosphate synthase), and (ii) to abol- astaxanthin, whereas those with 30-carbon backbones ish the activity of a competing branch pathway that uses (C carotenoids), such as 4,4’-diaponeurosporene (DNP) FPP. With this approach, it was envisaged that we could and 4,4’- diapolycopene (DLP), are synthesized by a lim- construct a stable and efficient C carotenoid-produc- ited group of bacteria, including Staphylococcus aureus ing B. subtilis strain that was plasmid- and marker-free, (Marshall and Wilmoth 1981), and Heliobacteria spp. an attribute of paramount importance for its potential (Takaichi et al. 1997). Genes responsible for C carot- development into a commercially viable bioprocess. enoid biosynthesis in S. aureus have been characterized (Pelz et al. 2005; Wieland et al. 1994).) The first dedi - cated enzyme in the C carotenoid synthetic pathway is 30 Filluelo et al. AMB Express (2023) 13:38 Page 3 of 11 Table 1 Bacterial strains and plasmids used in this study Strain Genotype or description Source/Reference E. coli NEB® turbo F’ proA + B + lacIq Laboratory stock ∆lacZM15 / fhuA2 ∆(lac- proAB) glnV galK16 galE15 R(zgb-210::Tn10)TetS B. megaterium DSM Source of fpps gene DSM B. subtilis 168 KO7-S ΔnprE ΔaprE Δepr Δmpr ΔnprB BGSC Δvpr Δbpr ΔsigF BsMN0 B. subtilis KO7-S strain harbor- This study ing the plasmid pHY_crtMN BsMN1 Contains one crtMN gene- Ferrando et al., copy integrated into the 2023 genome BsMN2 Contains two crtMN gene- Ferrando et al., copies integrated into the 2023 genome BsMN3 Contains three crtMN gene- Ferrando et al., copies integrated into the 2023 genome BsMN4 B. subtilis KO7-S strain This study harboring the plasmid pBS0E_crtMN BsMN5 BsMN3 strain with an fpps This study gene from B. megaterium replacing the sigX gene BsMN6 BsMN5 strain with a trun- This study cated copy of the yisP gene Plasmid Description Source/Reference pHY_crtMN Plasmid pHY300PLK contain- Yoshida et al. 2009 ing the crtMN operon from Fig. 1 Metabolic pathways associated with terpenoid biosynthesis in Staphylococcus aureus B. subtilis and engineering strategies for the production of yellow C 30 pBS0E Plasmid containing the Popp et al. 2017 carotenoids 4,4’-diaponeurosporene and 4,4’-diapolycopene (C path- 30 xylose-inducing promoter way). Foreign genes are marked in red. Yellow arrows outlined in black xylose-repressor system indicate the reactions reinforced by chromosomic overexpression of the pJOE8999 PmanP-cas9, pUC, pE194ts, Altenbuchner fpps gene (farnesyl diphosphate synthase) from B. megaterium DSM 319, kanr 2016 crtM (squalene desaturase) and crtN (dehydrosqualene desaturase or di- pJOE8999_VG_MN Plasmid used to replace the Ferrando et al., apophytoene desaturase) genes from S. aureus, and deletion of the yisP spoVG locus for the crtMN 2023 (farnesyl diphosphate phosphatase) gene, yielding the C carotenoid genes pigments 4,4’-diaponeurosporene and 4,4’-diapolycopene. Enzymes in pBS0E_crtMN Plasmid containing crtMN This study the MEP (Methylerythritol 4-phosphate) pathway: 1-deoxy-D-xylulose- genes under the control of 5-phosphate synthase (Dxs); 1-deoxy-D-xylulose-5-phosphate reduc- xylose-inducing promoter toisomerase or 2-C-methyl-D-erythritol 4-phosphate synthase (Dxr, also pJOE8999. Plasmid used to replace the This study known as IspC); 2-C-methyl-D-erythritol 4-phosphate cytidylyltrans- sg_sigX_fpps sigX locus for the fpps gene ferase (IspD); 4-(cytidine 5′-diphospho)-2-C-methyl-D-erythritol kinase (IspE); 2-C-methyl-D-erythritol 2,4-cyclodiphosphate synthase (IspF); pJOE8999.sg_ΔyisP Plasmid used to delete the This study (E)-4- hydroxy-3-methylbut-2-enyl-diphosphate synthase (IspG); 4-hy- yisP gene droxy-3- methylbut-2-enyl diphosphate reductase (IspH); and isopentenyl- diphosphate delta-isomerase (Idi). Geranyltransferase (IspA, also known as with seven inactivated protease genes, was used as a host YqiD) refers to the B. subtilis gene responsible for the supply of GPP (geranyl diphosphate) and FPP (farnesyl diphosphate) precursors strain for C carotenoid production. DNA isolation and. manipulations were carried out using standard proto- Materials and methods cols. The bacterial strains employed in this research are Bacterial strains listed in Table 1. The E. coli NEB® turbo strain (New England Biolabs) was used as the host strain for routine molecular cloning and Medium and culture conditions plasmid construction operations, and B. subtilis KO7-S E. coli strains were cultured in Luria-Bertani (LB) (Bacillus Genetic Stock Center), an asporogenous strain medium at 37 ºC, while B. subtilis KO7-S strains were grown in Tryptic Soy Broth (TSB) (17 g/l tryptone, 3 g/l Filluelo et al. AMB Express (2023) 13:38 Page 4 of 11 soytone, 2.5  g/l dextrose, 5.0  g/l NaCl, 2.5  g/l K HPO ) TS2F and TS2R targeting the yisP gene were synthe- 2 4 or Bacillus subtilis 1 (BS1) medium, typically used in sized and ligated to the vector, thus obtaining plasmid industrial fermentation (Wenzel et al. 2011). The BS1 pJOE8999.g_yisP. A 1.6 kb repair template, containing the medium contained standard salts (in g/l: 2 (NH4)2SO4; 800-bp upstream region and 800-bp downstream region 18.3 K2HPO4·3H2O; 6 KH2PO4; 1 Na+-citrate; 0.2 of the yisP gene, was PCR-amplified using the B. subtilis MgSO4·7H2O), trace metals (in mg/l: 120 FeSO4·7H2O; KO7-S genome as a template. Primer sets P5F/P5R and 30 MnSO4·H2O; 12 CuSO4·5H2O; 12 ZnCl2) and was P6F/P6R were used to amplify each fragment and fused supplemented with 12  g sucrose/l and 18  g soybean together by overlapping PCR. The repair template was meal/l (Sigma Aldrich). All strains were incubated at 37 further digested with Sfi I for ligation with pJOE8999.g_ ºC on a rotatory shaker at 200 rpm. When necessary, the yisP to obtain the editing plasmid pJOE8999.sg_ΔyisP, growth media were supplemented with antibiotics at the which was used to delete the yisP gene. following concentrations: 30 µg/ml kanamycin for E. coli, and 6 µg/ml kanamycin or 10 µg/ml tetracycline or 2 µg/ Transformation and plasmid curing ml erythromycin for B. subtilis. To induce the CRISPR- The well-established plasmids (1  µg) were then trans - Cas9 system in B. subtilis cells, 0.5% D-mannose was formed to B. subtilis KO7-S according to the standard added. methods described by Yasbin and coworkers (Yasbin et al. 1975). For the CRISPR-Cas9-induced genome editing, Plasmid construction and primers the resulting transformants were passaged three times The plasmids used in this study are listed in Table  1 and on LB agar plates (without any antibiotics) at 50  °C for the primers in Table S1. For the insertion of crtMN genes 24  h to cure the plasmid. The colonies were confirmed into the pBS0E vector (Popp et al. 2017), the pHY_crtMN as cured of the editing plasmid by streaking them onto plasmid (Yoshida et al. 2009) was used as a template to LB agar plates containing kanamycin or no antibiotics; amplify crtMN genes using primers P1F/P1R. The result - plasmid cured colonies fail to grow at 37 °C. To confirm ing DNA amplicon was treated with EcoRI and SpeI and whether the desired insertion or deletion in the genome cloned into the replicative plasmid pBS0E for the con- of B. subtilis had been performed, a colony PCR was con- struction of the xylose-inducible pBS0E_crtMN vector. ducted to amplify the target fragments from the bacterial CRISPR-Cas9-mediated genome editing in B. subtilis chromosome and validated by further Sanger sequencing. was performed using the pJOE8999 vector as the paren- tal plasmid, according to a previously described method Extraction of carotenoids from B. subtilis (Altenbuchner 2016). Carotenoids were extracted from the engineered B. subti- lis cells according to the literature (Xue et al. 2015) with Chromosomal integration of the fpps gene some modifications. Briefly, recombinant strains were To generate the sigX gene replacement by the fpps gene, inoculated in 50 ml TSB at an optical density (OD ) of oligonucleotides for 20 pb gRNA (TS1F and TS1R) were 0.05 and cultured for 24 h at 37 ºC (250 rpm). In the case synthesized and ligated to BsaI-digested pJOE8999. of xylose-inducing experiments, 1% xylose was added at sigX-targeting gRNA containing pJOE8999 was named an OD of 0.6, and strains were then cultured for an pJOE8999.g_sigX. A repair template for fpps integration additional 24 h in the same conditions. Samples were col- into the sigX gene was constructed in vitro by overlap lected by centrifugation at 8000 g for 15 min and washed extension PCR of three fragments as follows: the 800- with 1 ml TE buffer (10 mM Tris/HCl, 1 mM EDTA, bp upstream flanking genomic region of sigX (P2F/P2R pH 8.0). The cells were resuspended in 500 µl TE buffer. primers) followed by the fpps gene (P3F/P3R primers) To extract the carotenoids, cell suspensions were lysed and the 800-bp downstream flanking genomic region of with 25  µl of 100  mg/ml lysozyme, followed by incuba- sigX (P4F/P4R primers). Homologous arms were ampli- tion for 15  min at 37  °C. The cell lysate was then trans - fied using the B. subtilis KO7-S chromosome as a tem- ferred into a glass tube, covered in aluminum foil to avoid plate, while the fpps gene was amplified using genomic light exposure, and centrifuged for 20 min at 2100 g. The DNA from B. megaterium DSM 319. The fused fragment supernatant was removed, and 1 ml acetone was added was digested with Sfi I and then ligated into pJOE8999.g_ to the pellets. These were vortexed for 4 min, heated for sigX, which had also been digested with Sfi I to obtain the 2  min in a water-bath at 55ºC, and then vortexed again editing plasmid pJOE8999.g_sigX_fpps, used for FPPS for 2  min. After centrifugation at 2300  g for 15  min, the overexpression. supernatants were collected and transferred to a new glass tube. The acetone extraction was repeated four Deletion of the yisP gene times. Next, the acetone extracts were evaporated, and To generate the yisP knockout mutant, a procedure simi- the remaining carotenoids were dissolved in 100  µl ace- lar to the one described above was performed. Primers tone and collected in HPLC vials, prior to their analysis Filluelo et al. AMB Express (2023) 13:38 Page 5 of 11 Table 2 Comparison of dry cell weight, titer and yield of C carotenoids produced by engineered B. subtilis strains and relative increase compared to the control strain Strain DCW (g/L Titer Ca- Yield Ca- Relative culture) rotenoids rotenoids increase (mg/L (mg/g culture) DCW) BsMN0 1.36 ± 0.09 1.01 ± 0.08 0.74 ± 0.07 1 BsMN1 1.23 ± 0.13 2.40 ± 0.13 1.95 ± 0.12 2.64 BsMN2 1.35 ± 0.05 2.96 ± 0.07 2.19 ± 0.08 2.96 BsMN3 1.43 ± 0.06 3.30 ± 0.11 2.31 ± 0.16 3.12 BsMN4 1.87 ± 0.16 4.22 ± 0.23 2.26 ± 0.32 3.05 BsMN5 1.32 ± 0.09 4.49 ± 0.19 3.39 ± 0.33 4.58 BsMN6 1.47 ± 0.08 6.51 ± 0.12 4.42 ± 0.19 5.97 c d BsMN6 2.98 ± 0.14 9.11 ± 0.36 3.20 ± 0.24 NA The total amount of carotenoids was measured in triplicate (± standard deviation) Fig. 2 Quantitative analysis of C carotenoids produced by engineered B. The relative increase is calculated as the amount of carotenoids produced in subtilis strains. Samples were analyzed by HPLC after C carotenoid extrac- the engineered B. subtilis strain divided by the amount of carotenoids produced tion with acetone. Quantification of each C carotenoid was performed in the control strain (BsMNO) harboring the pHYCrtMN vector comparing peak areas with the standard reference curve, and then nor- Strain cultured in BS1 medium malized to the dry cell mass of each culture. The amount of DNP is indi- Not applicable cated in orange and the amount of DLP in yellow. The experiments were performed in triplicate using an HPLC system. Cell dry weight was determined by pelleting and drying a fraction of the culture. to an OD of 0.05 in TSB and grown in shake flasks at 220 rpm and 37 ºC for 24 h. Then, samples were taken to HPLC analysis of carotenoids quantify both the DCW and the total amounts of DNP Carotenoid extracts were analyzed with a Shimadzu and DLP by HPLC. The latter were calculated as mg/g HPLC system equipped with a Gemini® NX-C18 col- DCW to allow comparison between the strains. The umn (5 μm, 110 Å, 250 × 4.60 mm) and a UV/VIS detec- parental B. subtilis strain (BsMN0) containing only the tor at 25  °C. The mobile phase consisted of acetonitrile pHY_crtMN plasmid was used as a control. and water (85:15%) at a flow rate of 2 ml/min. DNP After 24  h of growth, all engineered B. subtilis strains and DLP were identified from their absorption spec - had an OD of 7–8, with DCW values of 1.23–1.47 g/L, tra and quantified by comparing their peak areas using showing a slight increase in DCW as the crtMN gene an standard calibration curve prepared with known copy number increased (Table  2). HPLC chromatogram amounts of β-carotene (quantified by absorbance), then analysis revealed two major peaks at 450  nm, which multiplying by the molar extinction coefficient (ε) of eluted at 2.4 and 2.8  min, with absorption spectra for β-carotene (138,900  M − 1  cm − 1 at 450  nm) (Britton et each peak identical to those of DLP and DNP, respec- al., 2004), and dividing by the ε value for the carotenoid tively (Fig. S1) (Takaichi 2000; Takaichi et al. 1997). As in question (147,000  M − 1  cm − 1 at 440  nm for DNP, the two peaks were present in the chromatograms of all 185,000 M − 1  cm − 1 at 470 nm for DLP) (Furubayashi et samples, both compounds were calculated individually al. 2014). Production weights of carotenoids were then as well as together as total carotenoids, with the results normalized to the dry cell weight (DCW) of each culture. provided in Fig.  2; Table  2. Surprisingly, the BsMN1 strain harboring a single copy of crtMN genes produced Results a titer of 2.40 ± 0.13  mg/L carotenoids with a yield of Dependence on thecrtMNgene copy number in 1.95 ± 0.12  mg/g DCW, which was already more than a C carotenoid production 2-fold increase in total carotenoid production compared A set of plasmid-less, marker-free B. subtilis strains to strain BsMN0 containing the pHY_crtMN plasmid harboring one (BsMN1), two (BsMN2) or three copies (0.74 ± 0.12  mg/g DCW). We observed that DCW and (BsMN3) of crtMN genes in their chromosomes under carotenoid yield slightly increased with increasing crtMN the control of the constitutive spoVG promoter were copy number and the highest titer of 3.30 ± 0.11  mg/L previously constructed by our research group, but not carotenoids was achieved in BsMN3, with a yield of characterized (Ferrando et al. 2023). Therefore, to inves - 2.31 ± 0.16  mg/g DCW, which constituted a 3.12-fold tigate the effect of multiple crtMN gene copy expression increase in carotenoid production compared to BsMN0 on the intracellular accumulation of C carotenoids, cells (Fig.  2 and Table  2). The yield obtained in BsMN0 was of a stationary overnight culture in TSB were diluted comparable with previously reported values (Xue et al., Filluelo et al. AMB Express (2023) 13:38 Page 6 of 11 2015; Abdallah et al. 2020), which demonstrates the feasi- medium. Based on these results, we surmised that heter- bility and robustness of the comparative studies. ologous expression of FPPS in B. subtilis is beneficial for The low carotenoid yield obtained in BsMN0 sug - the construction of a high-yielding C carotenoid-pro- gested that crtMN genes are poorly expressed through ducing strain. the pHY_crtMN plasmid. To test this hypothesis, we cloned the crtMN genes in the xylose-inducible medium Branch pathway engineering to increase C carotenoid copy number pBS0E plasmid (Popp et al. 2017), which production is particularly useful for overcoming bottlenecks in pro- To provide enough FPP for C carotenoid biosynthe- tein overproduction generated by limited expression of sis, it is crucial to attenuate branch pathways that use targeted genes (Toymentseva et al. 2012). The B. subti- this precursor as the starting material. In the biosyn- lis strain bearing the pBS0E_crtMN plasmid (BsMN4) thesis of farnesol lipids, each FPP molecule is converted showed a higher cell growth compared to BsMN0 – to farnesol by the action of farnesyl diphosphate phos- BsMN3 strains, with a DCW of 1.87  g/L, probably due phatase (YisP) (Fig.  1); therefore, this branch pathway to the addition of an extra carbon source (D-xylose was selected as a candidate for engineering. Plasmid inducer) to the media. As expected, BsMN4 exhibited a pJOE8999_ΔyisP was constructed to knock out a 770- notable increase in carotenoid yield (3.05-fold) compared bp fragment of yisP in strain BsMN5 and inactivate the to BsMN0, demonstrating a higher expression of crtMN function of YisP, thus blocking the synthesis of farnesol genes through this plasmid (Table  2). More importantly, in the newly generated strain BsMN6 (Fig. 3b and c). Dis- the yield obtained for strain BsMN4 was similar to that of ruption of the yisP gene in resulting transformants was BsMN3, indicating that plasmid-bearing and multicopy confirmed by PCR amplification, as previously (Fig.  3e), strains had a comparable performance. and further verified by sequencing. The positive clone was cured from the plasmid and subjected to fermenta- Optimization of the C carotenoid biosynthetic pathway tion for 24 h to measure the production of DLP and DNP. In the C carotenoid metabolic pathway in B. subtilis, Again, BsMN6 growth was similar to the parental strain farnesyl diphosphate synthase (IspA) converts the uni- BsMN5, indicating that yisP disruption in BsMN6 did not versal terpenoid precursors DMAPP and IPP to FPP, affect cell growth. However, C carotenoid production in which is the substrate for CrtMN enzymes in C carot- strain BsMN6 was significantly enhanced, being 130.4% enoid biosynthesis (Fig.  1). In order to further improve relative to BsMN5 after fermentation (Fig.  2; Table  2). the production of C carotenoids, we aimed to increase Overall, combining the simultaneous overexpression of the FPP supply, as studies report that enhanced FPP farnesyl diphosphate synthase, dehydrosqualene syn- availability drives metabolic flux toward their synthesis thase, and dehydrosqualene desaturase encoded by fpps, (Xue et al. 2015; Abdallah et al. 2020; Song et al. 2021). crtM and crtN, respectively, and the disruption of the This has been achieved previously by introducing either yisP gene positively affected C carotenoid production in an extra copy of ispA to release the theoretical bottleneck strain BsMN6, which was up to 6-fold higher compared within the metabolic pathway or an improved variant of to the control strain BsMN0 (Fig. 2; Table 2). the enzyme with enhanced catalytic properties (Zhao et al. 2013). In the present study, farnesyl diphosphate Stability of BsMN6 in C carotenoid production and its synthase (encoded by the fpps gene) from B. megate- cultivation in industrial fermentation medium rium DSM 319, which is an active highly specific enzyme The stability of C carotenoid production in strain exclusively yielding FPP (Hartz et al., 2018), was overex- BsMN6 without antibiotic selection was tested. An over- pressed to enhance the FPP pool. To this end, plasmid night culture of BsMN6 in TSB was diluted 1:1000 in pJOE8999.sigX_fpps was constructed for the replace- the same medium. The cells were grown in shake flasks ment of the sigX gene of BsMN3 (codifying for sigma at 37  °C to the stationary phase and diluted again 1000- factor SigX) with the fpps gene, setting the expression fold. This was repeated five times and in the last transfer, of the encoded FPPS under the control of a strong sigX when the stationary phase was reached, the strain was promoter (Song et al. 2016), and strain BsMN5 was gen- cultured again in TSB and the C carotenoid yield was erated (Fig.  3a). The insertion of the fpps gene in cured determined. As shown in Fig.  4a, BsMN6 produced sim- transformant cells was confirmed by diagnostic PCR ilar levels of C carotenoids for at least 50 generations (Fig.  3d) and further Sanger sequencing. Fermenta- (every round of growth to stationary phase corresponds tion studies revealed a remarkable 46.8% increase in the to about ten generations without antibiotic supplemen- production of C carotenoids compared with BsMN3 tation, calculated by dividing the length of the exponen- (Fig.  2and Table  2). Additionally, BsMN5 grew at a simi- tial growth phase (about 300  min) by the doubling time lar rate to the parental strain BsMN3, indicating that the of BsMN6 (approximately 30  min) in TSB medium), overexpression of FPPS did not affect cell growth in TSB Filluelo et al. AMB Express (2023) 13:38 Page 7 of 11 Fig. 3 Engineering of the genome-integrated farnesyl diphosphate synthase (FPPS) and disruption of farnesyl diphosphate phosphatase (YisP) in B. subtilis. (a) pJOE8999.g_sigX_fpps was designed to allow the replacement of the sigX gene from B. subtilis by the fpps gene from B. megaterium under the control of a strong promoter P . (b) pJOE8999.g_ΔyisP was constructed for the disruption of the yisP gene from B. subtilis (c) Upon transformation, the sigX resulting B. subtilis strain harboring both genomic modifications along with three gene-copies of the crtMN genes under the control of the constitutive promoter P was designated as BsMN6. (d) Confirmation of the sigX gene replacement by fpps in the BsMN5 strain. Lane 1 corresponds to an ampli- sigX fication band of 2.5 kb using primers P2F/P4R to verify fpps integration at the sigX locus site in BsMN5. Lane 2 corresponds to an amplification band of 2.25 kb using the same primers in recipient strain BsMN3. M corresponds to the molecular marker weight. (e) Confirmation of the yisP gene disruption in strain BsMN6. Lane 1 corresponds to an amplification band of 1.75 kb using primers P5F/P6R to verify yisP deletion in BsMN6. Lane 2 corresponds to an amplification band of 2.5 kb using the same primers in recipient strain BsMN5. M corresponds to the molecular marker weight demonstrating that BsMN6 achieved a high yield of C bacterial feed (Wenzel et al. 2011). To this end, BsMN6 carotenoids with stable productivity. was cultured for 24 h in TSB and BS1 media before ana- To date, recombinant production of C carotenoids in lyzing DCW and C carotenoid production. As shown 30 30 B. subtilis has been exclusively tested by culturing engi- in Fig.  4b and c; Table  2, BsMN6 was able to double neered strains in TSB medium at the shake flask level the cell biomass concentration when grown in BS1 (Yoshida et al. 2009; Xue et al. 2015; Abdallah et al. 2020). medium (2.98 ± 0.14  g/L culture) compared to the same However, TSB is a nutritious medium designed to sup- strain growing in TSB medium (1.47 ± 0.08  g/L culture). port the growth of a wide variety of microorganisms, and Although the yield of C carotenoids obtained in TSB inappropriate for B. subtilis fermentation on an indus- (4.42 ± 0.19  mg/g DCW) was higher compared to BS1 trial scale due to its high cost. We therefore decided to medium (3.20 ± 0.24  mg/g DCW), the titer of C carot- investigate the capacity of strain BsMN6 to accumu- enoids obtained in the latter was 40% higher than the titer late C carotenoids in BS1, a commonly used industrial obtained in TSB, reaching a value of 9.11 ± 0.36 mg/L C 30 30 Filluelo et al. AMB Express (2023) 13:38 Page 8 of 11 Fig. 4 Stability and C carotenoid production in strain BsMN6. (a) C carotenoid production in the BsMN6 strain diluted 1000-fold and grown to the 30 30 stationary phase, repeated 5 times, without antibiotics in TSB media. (b) Relative DCW and (c) relative titers of C carotenoids produced by strain BsMN6 cultured in TB and BS1 media, after 24 h of fermentation. The error bars represent the average ± standard deviation of three biological replicates carotenoids. This indicates that BS1 medium can stimu - maintain the cloned genes by genome integration, thus late cell growth and had a significantly positive effect on ensuring high stability in the absence of antibiotic selec- the C carotenoid titer in comparison with TSB. tion pressure. Nevertheless, the main drawback of this approach is that the resulting strains have a low gene dos- Discussion age unless multiple gene copies are integrated into the The market demand for carotenoids is continuing to genome (Yomantas et al. 2011; Huang et al. 2017; Wang grow due to their antioxidant, anti-inflammatory, and et al. 2004), until reaching expression levels comparable anticancer properties. In particular, the biotechnological to those of cells carrying multiple copies of a recombi- production of carotenoids to replace artificial pigments is nant plasmid. Our study clearly shows that the low copy rapidly gaining interest, despite technological, economic, number pHY_crtMN plasmid (5–15 per cell), a deriva- and legislative limitations. E. coli and B. subtilis strains tive of pHY_300PLK (Ishiwa and Shibahara 1985), is an have been engineered to accumulate C carotenoids unfavorable vector for maximizing crtMN gene expres- utilizing suitable expression vectors for relevant crtMN sion. We hypothesize that the reason for the low expres- genes, the overexpression of MEP pathway enzymes, and sion achieved is that crtMN genes are the second and the concomitant use of antibiotic drugs and plasmids. third genes transcribed from the promoter of the tetra- However, the current trend in industrial bioprocesses is cycline resistant gene (Isamu Maeda personal commu- to circumvent the use of antibiotic selection markers by nication). Within an operon, the expression of a gene at developing marker-free production systems due to con- the first position is expected to be higher compared to cerns derived from the massive overuse of antibiotics. In the gene at the second position, which in turn should be many areas of biotechnology, restrictions on antibiotic more expressed than a gene at the third position (Lim et usage have been imposed by regulatory authorities (Min- al. 2011). In contrast, C carotenoid production in cells gon et al., 2015). In the present work, we constructed a carrying multiple copies of the xylose-inducible medium plasmid-less, marker-free strain of B. subtilis, a bacterium copy number pBS0E_crtMN plasmid (15–25 per cell) that can naturally produce C carotenoids in the absence was significantly improved; more importantly, its per - of any inducer or antibiotic compound. Optimization formance was comparable to the plasmid-less strain steps involving crtMN gene dosage and an enhanced harboring three crtMN gene copies in the chromosome. supply of the precursor FPP were carried out using the Presumably, when these conditions occur, increasing CRISPR-Cas9 system, resulting in the generation of an the copy number no longer enhances expression levels efficient, safe, and stable C carotenoid-producing B. (Widner et al. 2000) and the potential bottlenecks in C 30 30 subtilis strain. carotenoid production rely on the expression of other Reliance on the use of plasmids and antibiotic selection rate-limiting enzymes in the biosynthetic pathway. Nota- markers constitutes a major limiting factor for the imple- bly, the insertion of three crtMN gene copies into the B. mentation of an optimal B. subtilis chassis able to execute subtilis chromosome debottlenecked an unexplored rate- the functions needed for efficient C carotenoid produc- limiting step in the C carotenoid biosynthetic route 30 30 tion. To bypass this limitation, an interesting option is to and at the same time alleviated the need for antibiotic Filluelo et al. AMB Express (2023) 13:38 Page 9 of 11 selection for plasmid maintenance. Moreover, its stabil- B. subtilis reported to date (Abdallah et al. 2020). In the ity and potential ecological safety suggests that the engi- present study, the combination of chromosomal overex- neered B. subtilis strain has great promise as an efficient pression of farnesyl diphosphate synthase, dehydrosqua- C carotenoid cell factory with practical application in lene synthase and dehydrosqualene desaturase encoded industrial settings (García-Moyano et al. 2020; Su et al. by fpps, crtM and crtN, respectively, with the simul- 2020). taneous disruption of the yisP gene, resulted in a titer To further improve the B. subtilis carotenoid produc- of 9.11  mg/L C carotenoids, and a yield of 4.42  mg/g tion capacity, we focused on modulating some of the DCW. Although the C carotenoid accumulation is simi- well-recognized regulatory elements that tightly control lar to that achieved in E. coli strains and lower (4.7-fold) the metabolic flux to C carotenoid biosynthesis from than in B. subtilis overexpressing the eight enzymes of the the universal precursors DMAPP and IPP. Specifically, MEP pathway, it should be noted that we only focused on our aim was to enhance the FPP pool and also amelio- improving the last three steps downstream of the MEP rate its consumption by removing the competing path- pathway. Consequently, one could expect that combining way yielding farnesol. The first attempt to overexpress both strategies would serve to obtain a superior produc- the fpps gene from B. megaterium resulted in a significant tive strain. Additionally, we demonstrated that routinely improvement (1.46-fold) of C carotenoid production. used industrial bacterial feed (antibiotic- and xylose- This result is in accordance with a previous study that inducer-free) may provide a cost-effective bioprocess for achieved 1.36-fold higher carotenoid yields by introduc- the industrial production of C carotenoids. In a nut- ing an extra copy of the homologous fpps gene from B. shell, taking advantage of its inherent capacity to synthe- subtilis (ispA) (Xue et al. 2015). The additional expres - size C carotenoids, we have developed a plasmid-less, sion of the fpps gene from Saccharomyces cerevisiae marker-free, B. subtilis strain that can serve as a stepping also increased the supply of the precursor FPP (Song et stone for further genetic engineering and fermentation al. 2021). We therefore conclude that the heterologous process optimization targeted at a sustainable and effi - expression of FPPS from B. megaterium increased C30 cient production of C carotenoids. carotenoid biosynthesis in B. subtilis, similarly to the val- ues obtained when an extra copy of the native IspA was Supplementary Information The online version contains supplementary material available at https://doi. overexpressed (Xue et al. 2015). It has also been reported org/10.1186/s13568-023-01542-x. that attenuation of a competing FPP-consuming path- way toward C55 heptaprenyl diphosphate contributed Supplementary Material 1 to a 1.15-fold increase in terpenoid synthesis (Song et al. 2021). Accordingly, we assumed that abolishing non- Acknowledgements essential expression of yisP, the only phosphatase that We thank Dr. Isamu Maeda for providing us with the plasmid pHY_crtMN. catalyzes the conversion of FPP to farnesol, would also Author contribution lead to less FPP consumption in this competing path- PP designed research. OF and JF conducted experiments. PP analyzed the way, and the resulting extra FPP could be used by CrtMN data. OF, JF and PP wrote the manuscript. All authors read and approved the manuscript. enzymes to increase C carotenoid yield. In the ΔyisP mutant, known to exhibit no FPP phosphatase activity Funding information (Feng et al. 2014), excess FPP was distributed to increase This work was supported by the Pla de Doctorats Industrials del Departament de Recerca i Universitats de la Generalitat de Catalunya and Gestió d’ Ajuts the carotenoid yield in the engineered strain 1.39-fold Universitaris de Recerca for grant number 2021 DI 77. (Fig.  2; Table  2). Thus, for the first time, the role of yisP knockout in an increased accumulation of C carot- 30 Data availability The datasets generated during and/or analyzed during the current study are enoids in B. subtilis was demonstrated. available from the corresponding author on reasonable request. Cell engineering techniques have been previously used to improve C carotenoid productivity in E. coli and B. Declarations subtilis. E. coli strains were engineered to accumulate C carotenoids, with production levels ranging from 30 Conflict of interest The authors declare no financial or commercial conflict of interest. 0.5  mg/ gDCW to 10.8  mg/L (Chae et al. 2010; Kim et al. 2010, 2022; Takemura et al. 2021). B. subtilis has also Ethical statement been engineered using two-plasmid systems comprising This article does not describe any studies with human participants or animals performed by any of the authors. pHY_crtMN (Yoshida et al. 2009), mediating crtMN gene overexpression, and xylose-inducible pHCMC04G (Xue Received: 5 February 2023 / Accepted: 5 April 2023 et al. 2015), mediating stable overexpression of all MEP pathway enzymes. In total, the yield of C carotenoids achieved was 21  mg/g DCW, the highest production in Filluelo et al. AMB Express (2023) 13:38 Page 10 of 11 References Liu H, Xu W, Chang X, Qin T, Yin Y, Yang Q (2016) 4,4’-diaponeurosporene, a C Abdallah II, Xue D, Pramastya H, van Merkerk R, Setroikromo R, Quax WJ (2020) A carotenoid, effectively activates dendritic cells via CD36 and NF-kappaB regulated synthetic operon facilitates stable overexpression of multigene ter- signaling in a ROS independent manner. Oncotarget 7(27):40978–40991. penoid pathway in Bacillus subtilis. J Ind Microbiol Biotechnol 47(2):243–249. https://doi.org/10.18632/oncotarget.9800 https://doi.org/10.1007/s10295-019-02257-4 Liu H, Xu W, Yu Q, Yang Q (2017) 4,4’-Diaponeurosporene-producing Bacillus subtilis Altenbuchner J (2016) Editing of the Bacillus subtilis genome by the CRISPR-Cas9 increased mouse resistance against Salmonella typhimurium infection in a System. Appl Environ Microbiol 15(17):5421–5427. https://doi.org/10.1128/ CD36-Dependent manner. Front Immunol 8:483. https://doi.org/10.3389/ AEM.01453-16 fimmu.2017.00483 Britton G, Liaaen-Jensen S, Pfander H (2004) Carotenoids Handbook. Birkhäuser Marshall JH, GJ Wilmoth (1981) Pigments of Staphylococcus aureus, a series of Verlag triterpenoid carotenoids. J Bacteriol 147:900–913. https://doi.org/10.1128/ Cardoso LAC, Karp SG, Vendruscolo F, Kanno KYF, Zoz LIC, Carvalho JC (2017) jb.147.3.900-913.1981 Biotechnological Production of Carotenoids and their applications in Mignon C, Sodoyer R, Werle B (2015) Antibiotic-free selection in biotherapeu- Food and Pharmaceutical Products. Carotenoids. InTech. doi:https://doi. tics: now and forever. Pathogens 4(2):157–181. https://doi.org/10.3390/ org/10.5772/67725 pathogens4020157 Chae HS, Kim K-H, Kim SC, Lee PC (2010) Strain-dependent carotenoid produc- Moser S, Pichler H (2019) Identifying and engineering the ideal microbial terpe- tions in metabolically engineered Escherichia coli. Appl Biochem Biotechnol noid production host. Appl Microbiol Biotechnol 103(14):5501–5516. https:// 162(8):2333–2344. https://doi.org/10.1007/s12010-010-9006-0 doi.org/10.1007/s00253-019-09892-y Earl AM, Losick R, Kolter R (2008) Ecology and genomics of Bacillus subtilis. Trends Pelz A, Wieland KP, Putzbach K, Hentschel P, Albert K, Götz F (2005) Structure and Microbiol 16(6):269–275. https://doi.org/10.1016/j.tim.2008.03.004 biosynthesis of staphyloxanthin from Staphylococcus aureus. J Biol Chem Feng X, Hu Y, Zheng Y, Zhu W, Li K, Huang CH, Ko TP, Ren F, Chan HC, Nega M, 280:32493–32498. https://doi.org/10.1074/jbc.M505070200 Bogue S, López D, Kolter R, Götz F, Guo RT, Oldfield E (2014) Structural and Popp PF, Dotzler M, Radeck J, Bartels J, Mascher T (2017) The Bacillus BioBrick Box functional analysis of Bacillus subtilis YisP reveals a role of its product in 2.0: expanding the genetic toolbox for the standardized work with Bacillus biofilm production. Chem Biol 21(11):1557–1563. https://doi.org/10.1016/j. subtilis. Sci Rep 7(1):15058. https://doi.org/10.1038/s41598-017-15107-z chembiol.2014.08.018 Schallmey M, Singh A, Ward OP (2004) Developments in the use of Bacillus species Ferrando J, Filluelo O, Zeigler DR, Picart P (2023) Barriers to simultaneous multilo- for industrial production. Can J Microbiol 50:1–17. https://doi.org/10.1139/ cus integration in Bacillus subtilis tumble down: development of a straightfor- w03-076 ward screening method for the colorimetric detection of one-step multiple Siziya IN, Hwang CY, Seo MJ (2022) Antioxidant potential and capacity of micro- gene insertion using the CRISPR-Cas9 system. Microb Cell Fact 22:21. https:// organism-sourced C Carotenoids-A review. Antioxid (Basel) 11(10):1963. doi.org/10.1186/s12934-023-02032-2 https://doi.org/10.3390/antiox11101963 Furubayashi M, Li L, Katabami A, Saito K, Umeno D (2014) Construction of carot- Song Y, Nikoloff JM, Fu G, Chen J, Li Q, Xie N, Zheng P, Sun J, Zhang D (2016) enoid biosynthetic pathways using squalene synthase. FEBS Lett 588(3):436– Promoter screening from Bacillus subtilis in various Conditions Hunting for 442. https://doi.org/10.1016/j.febslet.2013.12.003 Synthetic Biology and Industrial Applications. PLoS ONE 11(7):e0158447. García-Moyano A, Larsen Ø, Gaykawad S, Christakou E, Boccadoro C, Puntervoll P, https://doi.org/10.1371/journal.pone.0158447 Bjerga GEK (2020) Fragment Exchange plasmid tools for CRISPR/Cas9-Medi- Song Y, He S, Abdallah II, Jopkiewicz A, Setroikromo R, van Merkerk R, Tepper PG, ated gene integration and protease production in Bacillus subtilis. Appl Envi- Quax WJ (2021) Engineering of multiple modules to improve Amorphadi- ron Microbiol 87(1):e02090–e02020. https://doi.org/10.1128/AEM.02090-20 ene production in Bacillus subtilis using CRISPR-Cas9. J Agric Food Chem Guan Z, Xue D, Abdallah II, Dijkshoorn L, Setroikromo R, Lv G, Quax WJ (2015) Meta- 69(16):4785–4794. https://doi.org/10.1021/acs.jafc.1c00498 bolic engineering of Bacillus subtilis for terpenoid production. Appl Microbiol Su Y, Liu C, Fang H, Zhang D (2020) Bacillus subtilis: a universal cell factory for indus- Biotechnol 99(22):9395–9406. https://doi.org/10.1007/s00253-015-6950-1 try, agriculture, biomaterials and medicine. Microb Cell Fact 19(1):173. https:// Hartz P, Milhim M, Trenkamp S, Bernhardt R, Hannemann F (2018) Characterization doi.org/10.1186/s12934-020-01436-8 and engineering of a carotenoid biosynthesis operon from Bacillus megate- Takaichi S (2000) Characterization of carotenes in a combination of a C(18) rium. Metab Eng 49:47–58. https://doi.org/10.1016/j.ymben.2018.07.017 HPLC column with isocratic elution and absorption spectra with a Huang K, Zhang T, Jiang B, Yan X, Mu W, Miao M (2017) Overproduction of Rum- photodiode-array detector. Photosynth Res 65(1):93. https://doi.org/10.102 meliibacillus pycnus arginase with multi-copy insertion of the argr-pyc 23 3/A:1006445503030 cassette into the Bacillus subtilis chromosome. Appl Microbiol Biotechnol Takaichi S, Inoue K, Akaike M, Kobayashi M, Oh-oka H, Madigan MY (1997) The 101:6039–6048. https://doi.org/10.1007/s00253-017-8355-9 major carotenoid in all known species of heliobacteria is the C carotenoid Ishiwa H, Shibahara H (1985) New shuttle vectors for Esch- 4,4’-diaponeurosporene, not neurosporene. Arch Microbiol 168:277–281. erichia coli and Bacillus subtilis. Jpn J Genet 60:235–243. https://doi. https://doi.org/10.1007/s002030050499 org/10.1016/0378-1119(84)90161-6 Takemura M, Takagi C, Aikawa M, Araki K, Choi SK, Itaya M, Shindo K, Misawa N Jing Y, Liu H, Xu W, Yang Q (2017) Amelioration of the DSS-induced colitis in mice (2021) Heterologous production of novel and rare C -carotenoids using Pla- bypretreatment with 4,4′-diaponeurosporene-producing Bacillus subtilis. Exp nococcus carotenoid biosynthesis genes. Microb Cell Fact 20(1):194. https:// Ther Med 14(6):6069–6073. https://doi.org/10.3892/etm.2017.5282 doi.org/10.1186/s12934-021-01683-3 Jing Y, Liu H, Xu W, Yang Q (2019) 4,4′-Diaponeurosporene-producing Bacillus subti- Toymentseva AA, Schrecke K, Sharipova MR, Mascher T (2012) The LIKE lis promotes the development of the mucosal immune system of the piglet system, a novel protein expression toolbox for Bacillus subtilis based gut. Anat Rec (Hoboken) 302:1800–1807. https://doi.org/10.1002/ar.24102 on the liaI promoter. Microb. Cell Fact 11:143. doi: https://doi. Julsing MK, Rijpkema M, Woerdenbag HJ, Quax WJ, Kayser O (2007) Functional org/10.1186/1475-2859-11-143 analysis of genes involved in the biosynthesis of isoprene in Bacillus subtilis. Wagner WP, Helmig D, Fall R (2000) Isoprene biosynthesis in Bacillus subtilis via Appl Microbiol Biotechnol 75(6):1377–1384. https://doi.org/10.1007/ the methylerythritol phosphate pathway. J Nat Prod 63:37–40. https://doi. s00253-007-0953-5 org/10.1021/np990286p Kim J, Kong MK, Lee SY, Lee PC (2010) Carbon sources-dependent carotenoid Wang J-J, Rojanatavorn K, Shih JCH (2004) Increased production of Bacillus production in metabolically engineered Escherichia coli. World J Microbiol keratinase by chromosomal integration of multiple copies of the kerA gene. Biotechnol 26(12):2231–2239. https://doi.org/10.1007/s11274-010-0408-5 Biotechnol Bioeng 87:459–464. https://doi.org/10.1002/bit.20145 Kim M, Jung DH, Hwang CY, Siziya IN, Park YS, Seo MJ (2022) 4,4’-Diaponeu- Wenzel M, Müller A, Siemann-Herzberg M, Altenbuchner J (2011) Self-inducible rosporene production as C carotenoid with antioxidant activity in Bacillus subtilis expression system for reliable and inexpensive protein recombinant Escherichia coli. Appl Biochem Biotechnol Sep 6. https://doi. production by high-cell-density fermentation. Appl Environ Microbiol org/10.1007/s12010-022-04147-5 77(18):6419–6425. https://doi.org/10.1128/AEM.05219-11 Kuzma J, Nemecek-Marshall M, Pollock WH, Fall R (1995) Bacteria produce the Widner B, Thomas M, Sternberg D, Lammon D, Behr R, Sloma A (2000) Develop- volatile hydrocarbon isoprene. Curr Microbiol 30(2):97–103. https://doi. ment of marker-free strains of Bacillus subtilis capable of secreting high org/10.1007/BF00294190 levels of industrial enzymes. J Ind Microbiol Biotech 25:204–212. https://doi. Lim HN, Lee Y, Hussein R (2011) Fundamental relationship between operon orga- org/10.1038/sj.jim.7000051 nization and gene expression. Proc Natl Acad Sci U S A 108(26):10626–10631. Wieland B, Feil C, Gloria-Maercker E, Thumm G, Lechner M, Bravo JM, Poralla K, https://doi.org/10.1073/pnas.1105692108 Goëtz F (1994) Genetic and biochemical analyses of the biosynthesis of the Filluelo et al. AMB Express (2023) 13:38 Page 11 of 11 yellow carotenoid 4,40 -diaponeurosporene of Staphylococcus aureus. J Bac- mpr gene into the Bacillus subtilis chromosome. Microb Cell Fact 10:64–73. teriol 176:7719–7726. https://doi.org/10.1016/S0021-9258(19)39076-3 https://doi.org/10.1186/1475-2859-10-64 Wu G, Yan Q, Jones JA, Tang YJ, Fong SS, Koffas MAG (2016) Metabolic burden: Yoshida K, Ueda S, Maeda I (2009) Carotenoid production in Bacillus subtilis Cornerstones in Synthetic Biology and Metabolic Engineering Appli- achieved by metabolic engineering. Biotechnol Lett 31(11):1789–1793. cations. Trends Biotechnol 34(8):652–664. https://doi.org/10.1016/j. https://doi.org/10.1007/s10529-009-0082-6 tibtech.2016.02.010 Zhang C, Hong K (2020) Production of terpenoids by Synthetic Biology Xue D, Abdallah II, de Haan IE, Sibbald MJ, Quax WJ (2015) Enhanced C carot- Approaches. Front Bioeng Biotechnol 8:347. https://doi.org/10.3389/ enoid production in Bacillus subtilis by systematic overexpression of MEP fbioe.2020.00347 pathway genes. Appl Microbiol Biotechnol 99(14):5907–5915. https://doi. Zhao J, Li Q, Sun T, Zhu X, Xu H, Tang J, Zhang X, Ma Y (2013) Engineering central org/10.1007/s00253-015-6531-3 metabolic modules of Escherichia coli for improving beta-carotene produc- Yasbin RE, Wilson GA, Young FE (1975) Transformation and transfection in lyso- tion. Metab Eng 17:42–50. https://doi.org/10.1016/j.ymben.2013.02.002 genic strains of Bacillus subtilis: evidence for selective induction of prophage in competent cells. J Bacteriol 121:296–304. https://doi.org/10.1128/ Publisher’s Note jb.121.1.296-304.1975 Springer Nature remains neutral with regard to jurisdictional claims in Yomantas YA, Abalakina EG, Golubeva LI, Gorbacheva LY, Mashko SV (2011) published maps and institutional affiliations. Overproduction of Bacillus amyloliquefaciens extracellular glutamylendopep- tidase as a result of ectopic multi-copy insertion of an efficiently expressed http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png AMB Express Springer Journals

Metabolic engineering of Bacillus subtilis toward the efficient and stable production of C30-carotenoids

Filluelo, Oriana; Ferrando, Jordi; Picart, Pere
AMB Express , Volume 13 (1) – Apr 29, 2023
Download PDF
11 pages

Loading next page...
 
/lp/springer-journals/metabolic-engineering-of-bacillus-subtilis-toward-the-efficient-and-JMi9i8suBa

References (53)

Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2023
eISSN
2191-0855
DOI
10.1186/s13568-023-01542-x
Publisher site
See Article on Publisher Site

Abstract

Commercial carotenoid production is dominated by chemical synthesis and plant extraction, both of which are unsustainable and can be detrimental to the environment. A promising alternative for the mass production of carotenoids from both an ecological and commercial perspective is microbial synthesis. To date, C carotenoid production in Bacillus subtilis has been achieved using plasmid systems for the overexpression of biosynthetic enzymes. In the present study, we employed a clustered regularly interspaced short palindromic repeat-Cas9 (CRISPR-Cas9) system to develop an efficient, safe, and stable C carotenoid-producing B. subtilis strain, devoid of plasmids and antibiotic selection markers. To this end, the expression levels of crtM (dehydrosqualene synthase) and crtN (dehydrosqualene desaturase) genes from Staphylococcus aureus were upregulated by the insertion of three gene copies into the chromosome of B. subtilis. Subsequently, the supply of the C carotenoid precursor farnesyl diphosphate (FPP), which is the substrate for CrtMN enzymes, was enhanced by expressing chromosomally integrated Bacillus megaterium-derived farnesyl diphosphate synthase (FPPS), a key enzyme in the FPP pathway, and abolishing the expression of farnesyl diphosphate phosphatase (YisP), an enzyme responsible for the undesired conversion of FPP to farnesol. The consecutive combination of these features resulted in a stepwise increased production of C carotenoids. For the first time, a B. subtilis strain that can endogenously produce C 30 30 carotenoids has been constructed, which we anticipate will serve as a chassis for further metabolic engineering and fermentation optimization aimed at developing a commercial scale bioproduction process. Key points • Overexpression of chromosomally integrated crtMN genes improved C carotenoid production • Overexpression of FPPS and branch pathway attenuation further enhanced C carotenoid yield • A stable plasmid-less, marker-less C carotenoid-producing B. subtilis strain was constructed Keywords B. subtilis, C carotenoids, CRISPR-Cas9, Metabolic engineering † Faculty of Pharmacy and Food Science Technology, Department Oriana Filluelo and Jordi Ferrando have contributed equally to this of Biology, Healthcare and the Environment, Microbiology Section, work. University of Barcelona, Avinguda Joan XXIII, 27-31, Barcelona *Correspondence: 08028, Spain Pere Picart perepicart@ub.edu © The Author(s) 2023. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. Filluelo et al. AMB Express (2023) 13:38 Page 2 of 11 Introduction CrtM (dehydrosqualene synthase), which catalyzes the Terpenoids (also known as isoprenoids) constitute one of head-to-head condensation of two molecules of FPP to the largest and structurally most diverse groups of natu- dehydrosqualene. The enzyme CrtN (dehydrosqualene ral products with diverse biological functions (Zhang desaturase) then converts dehydrosqualene to the yel- and Hong 2020). An economically important class of low C carotenoid, DNP, a relatively unstable compound terpenoids are the carotenoids, which are ubiquitous that can suffer further oxidation by CrtMN to yield DLP. lipid-soluble pigments responsible for the red, yellow, The action of these two enzymes probably constitutes the and orange colors of plants, algae, fungi, and bacteria most common route of C carotenoid biosynthesis in (Cardoso et al. 2017). Although commercial carotenoid bacteria. Notably, these yellow pigments have attracted production is dominated by chemical synthesis and plant interest from the pharmaceutical industry owing to their extraction, these processes are not sustainable or ecologi- powerful antioxidant activities (Yoshida et al. 2009), as cal. Carotenoids are chemically synthesized under harsh well as their role as immunomodulators, significantly conditions, generating byproducts and hazardous waste, enhancing the immune system (Jing et al. 2017, 2019; Liu whereas sourcing carotenoids from plant extracts is gen- et al. 2016, 2017). Consequently, microbial cell engineer- erally dependent on the seasons and geographic areas, ing approaches aimed at improving C carotenoid yields which cannot always be standardized (Siziya et al. 2022). are required to achieve industrial-scale production. Therefore, microbial production is emerging as one of To date, the metabolic engineering of B. subtilis toward the most promising safe and environmentally friendly enhanced C carotenoid production has focused on options to satisfy the fast-growing demands for carot- using two-plasmid systems comprising pHY_crtMN enoids (Siziya et al. 2022). (Yoshida et al. 2009), mediating crtMN gene overexpres- B. subtilis is generally recognized as safe (GRAS), has sion under tetracycline selection, and xylose-inducible a high growth rate, and is easy to genetically manipulate pHCMC04G (Xue et al. 2015), mediating stable over- and cultivate, with a wide substrate range (Earl, 2008; expression of all MEP pathway enzymes under non- Schallmey et al., 2004). In addition, it is one of the high- selection conditions (Abdallah et al. 2020). However, est producer of isoprene (the smallest terpenoid) among two-plasmid systems may impose a metabolic burden eubacteria, thus constituting an ideal microbial host for on the host cells, leading to lower growth rates and use as a terpenoid cell factory (Kuzma et al. 1995; Wagner increased productivity costs (Wu et al. 2016). Another et al. 2000; Julsing et al. 2007; Moser and Pichler 2019; drawback is the high-cost of the inducer compounds Guan et al. 2015). This bacterium is able to initiate ter - and, more importantly, the requirement for antibiotic penoid biosynthesis from simple carbon sources through usage, which is restricted by governmental regulations the methylerythritol 4-phosphate (MEP) pathway, a route and can thus hinder the establishment of a commercially with eight enzymatic reactions leading to the synthesis viable industry. On the other hand, very little work has of isopentenyl diphosphate (IPP; C5) and dimethylallyl been done to explore the effects of modulating crtMN diphosphate (DMAPP; C5), the universal precursors of all gene expression and other competing branch pathways terpenoids (Guan et al. 2015). The consecutive condensa - (which can limit FPP availability) on C carotenoid pro- tion of IPP and DMAPP is catalyzed by prenyl diphos- duction, leaving room for improvement. In this study, phate synthase (IspA) to produce starting precursors for we initially compared the expression levels of plasmid- the synthesis of different classes of terpenoids: geranyl based and chromosomally integrated crtMN genes, and diphosphate (GPP; C10), a monoterpenoid precursor; then implemented CRISPR-Cas9-based metabolic engi- farnesyl diphosphate (FPP; C15) for the production of neering strategies to achieve an efficient C carotenoid- sesquiterpenoids, triterpenoids and C -carotenoids, and producing strain of B. subtilis, a bacterium that naturally geranylgeranyl diphosphate (GGPP; C20), the precur- produces yellow pigments (Fig.  1). u Th s, with the aim of sor of diterpenoids and carotenoids (Moser and Pichler increasing the supply of the carotenoid precursor FPP, we 2019). Most carotenoids contain a 40-carbon backbone planned (i) to introduce a chromosomally integrated copy (C carotenoids), including β-carotene, lycopene and of FPPS (farnesyl diphosphate synthase), and (ii) to abol- astaxanthin, whereas those with 30-carbon backbones ish the activity of a competing branch pathway that uses (C carotenoids), such as 4,4’-diaponeurosporene (DNP) FPP. With this approach, it was envisaged that we could and 4,4’- diapolycopene (DLP), are synthesized by a lim- construct a stable and efficient C carotenoid-produc- ited group of bacteria, including Staphylococcus aureus ing B. subtilis strain that was plasmid- and marker-free, (Marshall and Wilmoth 1981), and Heliobacteria spp. an attribute of paramount importance for its potential (Takaichi et al. 1997). Genes responsible for C carot- development into a commercially viable bioprocess. enoid biosynthesis in S. aureus have been characterized (Pelz et al. 2005; Wieland et al. 1994).) The first dedi - cated enzyme in the C carotenoid synthetic pathway is 30 Filluelo et al. AMB Express (2023) 13:38 Page 3 of 11 Table 1 Bacterial strains and plasmids used in this study Strain Genotype or description Source/Reference E. coli NEB® turbo F’ proA + B + lacIq Laboratory stock ∆lacZM15 / fhuA2 ∆(lac- proAB) glnV galK16 galE15 R(zgb-210::Tn10)TetS B. megaterium DSM Source of fpps gene DSM B. subtilis 168 KO7-S ΔnprE ΔaprE Δepr Δmpr ΔnprB BGSC Δvpr Δbpr ΔsigF BsMN0 B. subtilis KO7-S strain harbor- This study ing the plasmid pHY_crtMN BsMN1 Contains one crtMN gene- Ferrando et al., copy integrated into the 2023 genome BsMN2 Contains two crtMN gene- Ferrando et al., copies integrated into the 2023 genome BsMN3 Contains three crtMN gene- Ferrando et al., copies integrated into the 2023 genome BsMN4 B. subtilis KO7-S strain This study harboring the plasmid pBS0E_crtMN BsMN5 BsMN3 strain with an fpps This study gene from B. megaterium replacing the sigX gene BsMN6 BsMN5 strain with a trun- This study cated copy of the yisP gene Plasmid Description Source/Reference pHY_crtMN Plasmid pHY300PLK contain- Yoshida et al. 2009 ing the crtMN operon from Fig. 1 Metabolic pathways associated with terpenoid biosynthesis in Staphylococcus aureus B. subtilis and engineering strategies for the production of yellow C 30 pBS0E Plasmid containing the Popp et al. 2017 carotenoids 4,4’-diaponeurosporene and 4,4’-diapolycopene (C path- 30 xylose-inducing promoter way). Foreign genes are marked in red. Yellow arrows outlined in black xylose-repressor system indicate the reactions reinforced by chromosomic overexpression of the pJOE8999 PmanP-cas9, pUC, pE194ts, Altenbuchner fpps gene (farnesyl diphosphate synthase) from B. megaterium DSM 319, kanr 2016 crtM (squalene desaturase) and crtN (dehydrosqualene desaturase or di- pJOE8999_VG_MN Plasmid used to replace the Ferrando et al., apophytoene desaturase) genes from S. aureus, and deletion of the yisP spoVG locus for the crtMN 2023 (farnesyl diphosphate phosphatase) gene, yielding the C carotenoid genes pigments 4,4’-diaponeurosporene and 4,4’-diapolycopene. Enzymes in pBS0E_crtMN Plasmid containing crtMN This study the MEP (Methylerythritol 4-phosphate) pathway: 1-deoxy-D-xylulose- genes under the control of 5-phosphate synthase (Dxs); 1-deoxy-D-xylulose-5-phosphate reduc- xylose-inducing promoter toisomerase or 2-C-methyl-D-erythritol 4-phosphate synthase (Dxr, also pJOE8999. Plasmid used to replace the This study known as IspC); 2-C-methyl-D-erythritol 4-phosphate cytidylyltrans- sg_sigX_fpps sigX locus for the fpps gene ferase (IspD); 4-(cytidine 5′-diphospho)-2-C-methyl-D-erythritol kinase (IspE); 2-C-methyl-D-erythritol 2,4-cyclodiphosphate synthase (IspF); pJOE8999.sg_ΔyisP Plasmid used to delete the This study (E)-4- hydroxy-3-methylbut-2-enyl-diphosphate synthase (IspG); 4-hy- yisP gene droxy-3- methylbut-2-enyl diphosphate reductase (IspH); and isopentenyl- diphosphate delta-isomerase (Idi). Geranyltransferase (IspA, also known as with seven inactivated protease genes, was used as a host YqiD) refers to the B. subtilis gene responsible for the supply of GPP (geranyl diphosphate) and FPP (farnesyl diphosphate) precursors strain for C carotenoid production. DNA isolation and. manipulations were carried out using standard proto- Materials and methods cols. The bacterial strains employed in this research are Bacterial strains listed in Table 1. The E. coli NEB® turbo strain (New England Biolabs) was used as the host strain for routine molecular cloning and Medium and culture conditions plasmid construction operations, and B. subtilis KO7-S E. coli strains were cultured in Luria-Bertani (LB) (Bacillus Genetic Stock Center), an asporogenous strain medium at 37 ºC, while B. subtilis KO7-S strains were grown in Tryptic Soy Broth (TSB) (17 g/l tryptone, 3 g/l Filluelo et al. AMB Express (2023) 13:38 Page 4 of 11 soytone, 2.5  g/l dextrose, 5.0  g/l NaCl, 2.5  g/l K HPO ) TS2F and TS2R targeting the yisP gene were synthe- 2 4 or Bacillus subtilis 1 (BS1) medium, typically used in sized and ligated to the vector, thus obtaining plasmid industrial fermentation (Wenzel et al. 2011). The BS1 pJOE8999.g_yisP. A 1.6 kb repair template, containing the medium contained standard salts (in g/l: 2 (NH4)2SO4; 800-bp upstream region and 800-bp downstream region 18.3 K2HPO4·3H2O; 6 KH2PO4; 1 Na+-citrate; 0.2 of the yisP gene, was PCR-amplified using the B. subtilis MgSO4·7H2O), trace metals (in mg/l: 120 FeSO4·7H2O; KO7-S genome as a template. Primer sets P5F/P5R and 30 MnSO4·H2O; 12 CuSO4·5H2O; 12 ZnCl2) and was P6F/P6R were used to amplify each fragment and fused supplemented with 12  g sucrose/l and 18  g soybean together by overlapping PCR. The repair template was meal/l (Sigma Aldrich). All strains were incubated at 37 further digested with Sfi I for ligation with pJOE8999.g_ ºC on a rotatory shaker at 200 rpm. When necessary, the yisP to obtain the editing plasmid pJOE8999.sg_ΔyisP, growth media were supplemented with antibiotics at the which was used to delete the yisP gene. following concentrations: 30 µg/ml kanamycin for E. coli, and 6 µg/ml kanamycin or 10 µg/ml tetracycline or 2 µg/ Transformation and plasmid curing ml erythromycin for B. subtilis. To induce the CRISPR- The well-established plasmids (1  µg) were then trans - Cas9 system in B. subtilis cells, 0.5% D-mannose was formed to B. subtilis KO7-S according to the standard added. methods described by Yasbin and coworkers (Yasbin et al. 1975). For the CRISPR-Cas9-induced genome editing, Plasmid construction and primers the resulting transformants were passaged three times The plasmids used in this study are listed in Table  1 and on LB agar plates (without any antibiotics) at 50  °C for the primers in Table S1. For the insertion of crtMN genes 24  h to cure the plasmid. The colonies were confirmed into the pBS0E vector (Popp et al. 2017), the pHY_crtMN as cured of the editing plasmid by streaking them onto plasmid (Yoshida et al. 2009) was used as a template to LB agar plates containing kanamycin or no antibiotics; amplify crtMN genes using primers P1F/P1R. The result - plasmid cured colonies fail to grow at 37 °C. To confirm ing DNA amplicon was treated with EcoRI and SpeI and whether the desired insertion or deletion in the genome cloned into the replicative plasmid pBS0E for the con- of B. subtilis had been performed, a colony PCR was con- struction of the xylose-inducible pBS0E_crtMN vector. ducted to amplify the target fragments from the bacterial CRISPR-Cas9-mediated genome editing in B. subtilis chromosome and validated by further Sanger sequencing. was performed using the pJOE8999 vector as the paren- tal plasmid, according to a previously described method Extraction of carotenoids from B. subtilis (Altenbuchner 2016). Carotenoids were extracted from the engineered B. subti- lis cells according to the literature (Xue et al. 2015) with Chromosomal integration of the fpps gene some modifications. Briefly, recombinant strains were To generate the sigX gene replacement by the fpps gene, inoculated in 50 ml TSB at an optical density (OD ) of oligonucleotides for 20 pb gRNA (TS1F and TS1R) were 0.05 and cultured for 24 h at 37 ºC (250 rpm). In the case synthesized and ligated to BsaI-digested pJOE8999. of xylose-inducing experiments, 1% xylose was added at sigX-targeting gRNA containing pJOE8999 was named an OD of 0.6, and strains were then cultured for an pJOE8999.g_sigX. A repair template for fpps integration additional 24 h in the same conditions. Samples were col- into the sigX gene was constructed in vitro by overlap lected by centrifugation at 8000 g for 15 min and washed extension PCR of three fragments as follows: the 800- with 1 ml TE buffer (10 mM Tris/HCl, 1 mM EDTA, bp upstream flanking genomic region of sigX (P2F/P2R pH 8.0). The cells were resuspended in 500 µl TE buffer. primers) followed by the fpps gene (P3F/P3R primers) To extract the carotenoids, cell suspensions were lysed and the 800-bp downstream flanking genomic region of with 25  µl of 100  mg/ml lysozyme, followed by incuba- sigX (P4F/P4R primers). Homologous arms were ampli- tion for 15  min at 37  °C. The cell lysate was then trans - fied using the B. subtilis KO7-S chromosome as a tem- ferred into a glass tube, covered in aluminum foil to avoid plate, while the fpps gene was amplified using genomic light exposure, and centrifuged for 20 min at 2100 g. The DNA from B. megaterium DSM 319. The fused fragment supernatant was removed, and 1 ml acetone was added was digested with Sfi I and then ligated into pJOE8999.g_ to the pellets. These were vortexed for 4 min, heated for sigX, which had also been digested with Sfi I to obtain the 2  min in a water-bath at 55ºC, and then vortexed again editing plasmid pJOE8999.g_sigX_fpps, used for FPPS for 2  min. After centrifugation at 2300  g for 15  min, the overexpression. supernatants were collected and transferred to a new glass tube. The acetone extraction was repeated four Deletion of the yisP gene times. Next, the acetone extracts were evaporated, and To generate the yisP knockout mutant, a procedure simi- the remaining carotenoids were dissolved in 100  µl ace- lar to the one described above was performed. Primers tone and collected in HPLC vials, prior to their analysis Filluelo et al. AMB Express (2023) 13:38 Page 5 of 11 Table 2 Comparison of dry cell weight, titer and yield of C carotenoids produced by engineered B. subtilis strains and relative increase compared to the control strain Strain DCW (g/L Titer Ca- Yield Ca- Relative culture) rotenoids rotenoids increase (mg/L (mg/g culture) DCW) BsMN0 1.36 ± 0.09 1.01 ± 0.08 0.74 ± 0.07 1 BsMN1 1.23 ± 0.13 2.40 ± 0.13 1.95 ± 0.12 2.64 BsMN2 1.35 ± 0.05 2.96 ± 0.07 2.19 ± 0.08 2.96 BsMN3 1.43 ± 0.06 3.30 ± 0.11 2.31 ± 0.16 3.12 BsMN4 1.87 ± 0.16 4.22 ± 0.23 2.26 ± 0.32 3.05 BsMN5 1.32 ± 0.09 4.49 ± 0.19 3.39 ± 0.33 4.58 BsMN6 1.47 ± 0.08 6.51 ± 0.12 4.42 ± 0.19 5.97 c d BsMN6 2.98 ± 0.14 9.11 ± 0.36 3.20 ± 0.24 NA The total amount of carotenoids was measured in triplicate (± standard deviation) Fig. 2 Quantitative analysis of C carotenoids produced by engineered B. The relative increase is calculated as the amount of carotenoids produced in subtilis strains. Samples were analyzed by HPLC after C carotenoid extrac- the engineered B. subtilis strain divided by the amount of carotenoids produced tion with acetone. Quantification of each C carotenoid was performed in the control strain (BsMNO) harboring the pHYCrtMN vector comparing peak areas with the standard reference curve, and then nor- Strain cultured in BS1 medium malized to the dry cell mass of each culture. The amount of DNP is indi- Not applicable cated in orange and the amount of DLP in yellow. The experiments were performed in triplicate using an HPLC system. Cell dry weight was determined by pelleting and drying a fraction of the culture. to an OD of 0.05 in TSB and grown in shake flasks at 220 rpm and 37 ºC for 24 h. Then, samples were taken to HPLC analysis of carotenoids quantify both the DCW and the total amounts of DNP Carotenoid extracts were analyzed with a Shimadzu and DLP by HPLC. The latter were calculated as mg/g HPLC system equipped with a Gemini® NX-C18 col- DCW to allow comparison between the strains. The umn (5 μm, 110 Å, 250 × 4.60 mm) and a UV/VIS detec- parental B. subtilis strain (BsMN0) containing only the tor at 25  °C. The mobile phase consisted of acetonitrile pHY_crtMN plasmid was used as a control. and water (85:15%) at a flow rate of 2 ml/min. DNP After 24  h of growth, all engineered B. subtilis strains and DLP were identified from their absorption spec - had an OD of 7–8, with DCW values of 1.23–1.47 g/L, tra and quantified by comparing their peak areas using showing a slight increase in DCW as the crtMN gene an standard calibration curve prepared with known copy number increased (Table  2). HPLC chromatogram amounts of β-carotene (quantified by absorbance), then analysis revealed two major peaks at 450  nm, which multiplying by the molar extinction coefficient (ε) of eluted at 2.4 and 2.8  min, with absorption spectra for β-carotene (138,900  M − 1  cm − 1 at 450  nm) (Britton et each peak identical to those of DLP and DNP, respec- al., 2004), and dividing by the ε value for the carotenoid tively (Fig. S1) (Takaichi 2000; Takaichi et al. 1997). As in question (147,000  M − 1  cm − 1 at 440  nm for DNP, the two peaks were present in the chromatograms of all 185,000 M − 1  cm − 1 at 470 nm for DLP) (Furubayashi et samples, both compounds were calculated individually al. 2014). Production weights of carotenoids were then as well as together as total carotenoids, with the results normalized to the dry cell weight (DCW) of each culture. provided in Fig.  2; Table  2. Surprisingly, the BsMN1 strain harboring a single copy of crtMN genes produced Results a titer of 2.40 ± 0.13  mg/L carotenoids with a yield of Dependence on thecrtMNgene copy number in 1.95 ± 0.12  mg/g DCW, which was already more than a C carotenoid production 2-fold increase in total carotenoid production compared A set of plasmid-less, marker-free B. subtilis strains to strain BsMN0 containing the pHY_crtMN plasmid harboring one (BsMN1), two (BsMN2) or three copies (0.74 ± 0.12  mg/g DCW). We observed that DCW and (BsMN3) of crtMN genes in their chromosomes under carotenoid yield slightly increased with increasing crtMN the control of the constitutive spoVG promoter were copy number and the highest titer of 3.30 ± 0.11  mg/L previously constructed by our research group, but not carotenoids was achieved in BsMN3, with a yield of characterized (Ferrando et al. 2023). Therefore, to inves - 2.31 ± 0.16  mg/g DCW, which constituted a 3.12-fold tigate the effect of multiple crtMN gene copy expression increase in carotenoid production compared to BsMN0 on the intracellular accumulation of C carotenoids, cells (Fig.  2 and Table  2). The yield obtained in BsMN0 was of a stationary overnight culture in TSB were diluted comparable with previously reported values (Xue et al., Filluelo et al. AMB Express (2023) 13:38 Page 6 of 11 2015; Abdallah et al. 2020), which demonstrates the feasi- medium. Based on these results, we surmised that heter- bility and robustness of the comparative studies. ologous expression of FPPS in B. subtilis is beneficial for The low carotenoid yield obtained in BsMN0 sug - the construction of a high-yielding C carotenoid-pro- gested that crtMN genes are poorly expressed through ducing strain. the pHY_crtMN plasmid. To test this hypothesis, we cloned the crtMN genes in the xylose-inducible medium Branch pathway engineering to increase C carotenoid copy number pBS0E plasmid (Popp et al. 2017), which production is particularly useful for overcoming bottlenecks in pro- To provide enough FPP for C carotenoid biosynthe- tein overproduction generated by limited expression of sis, it is crucial to attenuate branch pathways that use targeted genes (Toymentseva et al. 2012). The B. subti- this precursor as the starting material. In the biosyn- lis strain bearing the pBS0E_crtMN plasmid (BsMN4) thesis of farnesol lipids, each FPP molecule is converted showed a higher cell growth compared to BsMN0 – to farnesol by the action of farnesyl diphosphate phos- BsMN3 strains, with a DCW of 1.87  g/L, probably due phatase (YisP) (Fig.  1); therefore, this branch pathway to the addition of an extra carbon source (D-xylose was selected as a candidate for engineering. Plasmid inducer) to the media. As expected, BsMN4 exhibited a pJOE8999_ΔyisP was constructed to knock out a 770- notable increase in carotenoid yield (3.05-fold) compared bp fragment of yisP in strain BsMN5 and inactivate the to BsMN0, demonstrating a higher expression of crtMN function of YisP, thus blocking the synthesis of farnesol genes through this plasmid (Table  2). More importantly, in the newly generated strain BsMN6 (Fig. 3b and c). Dis- the yield obtained for strain BsMN4 was similar to that of ruption of the yisP gene in resulting transformants was BsMN3, indicating that plasmid-bearing and multicopy confirmed by PCR amplification, as previously (Fig.  3e), strains had a comparable performance. and further verified by sequencing. The positive clone was cured from the plasmid and subjected to fermenta- Optimization of the C carotenoid biosynthetic pathway tion for 24 h to measure the production of DLP and DNP. In the C carotenoid metabolic pathway in B. subtilis, Again, BsMN6 growth was similar to the parental strain farnesyl diphosphate synthase (IspA) converts the uni- BsMN5, indicating that yisP disruption in BsMN6 did not versal terpenoid precursors DMAPP and IPP to FPP, affect cell growth. However, C carotenoid production in which is the substrate for CrtMN enzymes in C carot- strain BsMN6 was significantly enhanced, being 130.4% enoid biosynthesis (Fig.  1). In order to further improve relative to BsMN5 after fermentation (Fig.  2; Table  2). the production of C carotenoids, we aimed to increase Overall, combining the simultaneous overexpression of the FPP supply, as studies report that enhanced FPP farnesyl diphosphate synthase, dehydrosqualene syn- availability drives metabolic flux toward their synthesis thase, and dehydrosqualene desaturase encoded by fpps, (Xue et al. 2015; Abdallah et al. 2020; Song et al. 2021). crtM and crtN, respectively, and the disruption of the This has been achieved previously by introducing either yisP gene positively affected C carotenoid production in an extra copy of ispA to release the theoretical bottleneck strain BsMN6, which was up to 6-fold higher compared within the metabolic pathway or an improved variant of to the control strain BsMN0 (Fig. 2; Table 2). the enzyme with enhanced catalytic properties (Zhao et al. 2013). In the present study, farnesyl diphosphate Stability of BsMN6 in C carotenoid production and its synthase (encoded by the fpps gene) from B. megate- cultivation in industrial fermentation medium rium DSM 319, which is an active highly specific enzyme The stability of C carotenoid production in strain exclusively yielding FPP (Hartz et al., 2018), was overex- BsMN6 without antibiotic selection was tested. An over- pressed to enhance the FPP pool. To this end, plasmid night culture of BsMN6 in TSB was diluted 1:1000 in pJOE8999.sigX_fpps was constructed for the replace- the same medium. The cells were grown in shake flasks ment of the sigX gene of BsMN3 (codifying for sigma at 37  °C to the stationary phase and diluted again 1000- factor SigX) with the fpps gene, setting the expression fold. This was repeated five times and in the last transfer, of the encoded FPPS under the control of a strong sigX when the stationary phase was reached, the strain was promoter (Song et al. 2016), and strain BsMN5 was gen- cultured again in TSB and the C carotenoid yield was erated (Fig.  3a). The insertion of the fpps gene in cured determined. As shown in Fig.  4a, BsMN6 produced sim- transformant cells was confirmed by diagnostic PCR ilar levels of C carotenoids for at least 50 generations (Fig.  3d) and further Sanger sequencing. Fermenta- (every round of growth to stationary phase corresponds tion studies revealed a remarkable 46.8% increase in the to about ten generations without antibiotic supplemen- production of C carotenoids compared with BsMN3 tation, calculated by dividing the length of the exponen- (Fig.  2and Table  2). Additionally, BsMN5 grew at a simi- tial growth phase (about 300  min) by the doubling time lar rate to the parental strain BsMN3, indicating that the of BsMN6 (approximately 30  min) in TSB medium), overexpression of FPPS did not affect cell growth in TSB Filluelo et al. AMB Express (2023) 13:38 Page 7 of 11 Fig. 3 Engineering of the genome-integrated farnesyl diphosphate synthase (FPPS) and disruption of farnesyl diphosphate phosphatase (YisP) in B. subtilis. (a) pJOE8999.g_sigX_fpps was designed to allow the replacement of the sigX gene from B. subtilis by the fpps gene from B. megaterium under the control of a strong promoter P . (b) pJOE8999.g_ΔyisP was constructed for the disruption of the yisP gene from B. subtilis (c) Upon transformation, the sigX resulting B. subtilis strain harboring both genomic modifications along with three gene-copies of the crtMN genes under the control of the constitutive promoter P was designated as BsMN6. (d) Confirmation of the sigX gene replacement by fpps in the BsMN5 strain. Lane 1 corresponds to an ampli- sigX fication band of 2.5 kb using primers P2F/P4R to verify fpps integration at the sigX locus site in BsMN5. Lane 2 corresponds to an amplification band of 2.25 kb using the same primers in recipient strain BsMN3. M corresponds to the molecular marker weight. (e) Confirmation of the yisP gene disruption in strain BsMN6. Lane 1 corresponds to an amplification band of 1.75 kb using primers P5F/P6R to verify yisP deletion in BsMN6. Lane 2 corresponds to an amplification band of 2.5 kb using the same primers in recipient strain BsMN5. M corresponds to the molecular marker weight demonstrating that BsMN6 achieved a high yield of C bacterial feed (Wenzel et al. 2011). To this end, BsMN6 carotenoids with stable productivity. was cultured for 24 h in TSB and BS1 media before ana- To date, recombinant production of C carotenoids in lyzing DCW and C carotenoid production. As shown 30 30 B. subtilis has been exclusively tested by culturing engi- in Fig.  4b and c; Table  2, BsMN6 was able to double neered strains in TSB medium at the shake flask level the cell biomass concentration when grown in BS1 (Yoshida et al. 2009; Xue et al. 2015; Abdallah et al. 2020). medium (2.98 ± 0.14  g/L culture) compared to the same However, TSB is a nutritious medium designed to sup- strain growing in TSB medium (1.47 ± 0.08  g/L culture). port the growth of a wide variety of microorganisms, and Although the yield of C carotenoids obtained in TSB inappropriate for B. subtilis fermentation on an indus- (4.42 ± 0.19  mg/g DCW) was higher compared to BS1 trial scale due to its high cost. We therefore decided to medium (3.20 ± 0.24  mg/g DCW), the titer of C carot- investigate the capacity of strain BsMN6 to accumu- enoids obtained in the latter was 40% higher than the titer late C carotenoids in BS1, a commonly used industrial obtained in TSB, reaching a value of 9.11 ± 0.36 mg/L C 30 30 Filluelo et al. AMB Express (2023) 13:38 Page 8 of 11 Fig. 4 Stability and C carotenoid production in strain BsMN6. (a) C carotenoid production in the BsMN6 strain diluted 1000-fold and grown to the 30 30 stationary phase, repeated 5 times, without antibiotics in TSB media. (b) Relative DCW and (c) relative titers of C carotenoids produced by strain BsMN6 cultured in TB and BS1 media, after 24 h of fermentation. The error bars represent the average ± standard deviation of three biological replicates carotenoids. This indicates that BS1 medium can stimu - maintain the cloned genes by genome integration, thus late cell growth and had a significantly positive effect on ensuring high stability in the absence of antibiotic selec- the C carotenoid titer in comparison with TSB. tion pressure. Nevertheless, the main drawback of this approach is that the resulting strains have a low gene dos- Discussion age unless multiple gene copies are integrated into the The market demand for carotenoids is continuing to genome (Yomantas et al. 2011; Huang et al. 2017; Wang grow due to their antioxidant, anti-inflammatory, and et al. 2004), until reaching expression levels comparable anticancer properties. In particular, the biotechnological to those of cells carrying multiple copies of a recombi- production of carotenoids to replace artificial pigments is nant plasmid. Our study clearly shows that the low copy rapidly gaining interest, despite technological, economic, number pHY_crtMN plasmid (5–15 per cell), a deriva- and legislative limitations. E. coli and B. subtilis strains tive of pHY_300PLK (Ishiwa and Shibahara 1985), is an have been engineered to accumulate C carotenoids unfavorable vector for maximizing crtMN gene expres- utilizing suitable expression vectors for relevant crtMN sion. We hypothesize that the reason for the low expres- genes, the overexpression of MEP pathway enzymes, and sion achieved is that crtMN genes are the second and the concomitant use of antibiotic drugs and plasmids. third genes transcribed from the promoter of the tetra- However, the current trend in industrial bioprocesses is cycline resistant gene (Isamu Maeda personal commu- to circumvent the use of antibiotic selection markers by nication). Within an operon, the expression of a gene at developing marker-free production systems due to con- the first position is expected to be higher compared to cerns derived from the massive overuse of antibiotics. In the gene at the second position, which in turn should be many areas of biotechnology, restrictions on antibiotic more expressed than a gene at the third position (Lim et usage have been imposed by regulatory authorities (Min- al. 2011). In contrast, C carotenoid production in cells gon et al., 2015). In the present work, we constructed a carrying multiple copies of the xylose-inducible medium plasmid-less, marker-free strain of B. subtilis, a bacterium copy number pBS0E_crtMN plasmid (15–25 per cell) that can naturally produce C carotenoids in the absence was significantly improved; more importantly, its per - of any inducer or antibiotic compound. Optimization formance was comparable to the plasmid-less strain steps involving crtMN gene dosage and an enhanced harboring three crtMN gene copies in the chromosome. supply of the precursor FPP were carried out using the Presumably, when these conditions occur, increasing CRISPR-Cas9 system, resulting in the generation of an the copy number no longer enhances expression levels efficient, safe, and stable C carotenoid-producing B. (Widner et al. 2000) and the potential bottlenecks in C 30 30 subtilis strain. carotenoid production rely on the expression of other Reliance on the use of plasmids and antibiotic selection rate-limiting enzymes in the biosynthetic pathway. Nota- markers constitutes a major limiting factor for the imple- bly, the insertion of three crtMN gene copies into the B. mentation of an optimal B. subtilis chassis able to execute subtilis chromosome debottlenecked an unexplored rate- the functions needed for efficient C carotenoid produc- limiting step in the C carotenoid biosynthetic route 30 30 tion. To bypass this limitation, an interesting option is to and at the same time alleviated the need for antibiotic Filluelo et al. AMB Express (2023) 13:38 Page 9 of 11 selection for plasmid maintenance. Moreover, its stabil- B. subtilis reported to date (Abdallah et al. 2020). In the ity and potential ecological safety suggests that the engi- present study, the combination of chromosomal overex- neered B. subtilis strain has great promise as an efficient pression of farnesyl diphosphate synthase, dehydrosqua- C carotenoid cell factory with practical application in lene synthase and dehydrosqualene desaturase encoded industrial settings (García-Moyano et al. 2020; Su et al. by fpps, crtM and crtN, respectively, with the simul- 2020). taneous disruption of the yisP gene, resulted in a titer To further improve the B. subtilis carotenoid produc- of 9.11  mg/L C carotenoids, and a yield of 4.42  mg/g tion capacity, we focused on modulating some of the DCW. Although the C carotenoid accumulation is simi- well-recognized regulatory elements that tightly control lar to that achieved in E. coli strains and lower (4.7-fold) the metabolic flux to C carotenoid biosynthesis from than in B. subtilis overexpressing the eight enzymes of the the universal precursors DMAPP and IPP. Specifically, MEP pathway, it should be noted that we only focused on our aim was to enhance the FPP pool and also amelio- improving the last three steps downstream of the MEP rate its consumption by removing the competing path- pathway. Consequently, one could expect that combining way yielding farnesol. The first attempt to overexpress both strategies would serve to obtain a superior produc- the fpps gene from B. megaterium resulted in a significant tive strain. Additionally, we demonstrated that routinely improvement (1.46-fold) of C carotenoid production. used industrial bacterial feed (antibiotic- and xylose- This result is in accordance with a previous study that inducer-free) may provide a cost-effective bioprocess for achieved 1.36-fold higher carotenoid yields by introduc- the industrial production of C carotenoids. In a nut- ing an extra copy of the homologous fpps gene from B. shell, taking advantage of its inherent capacity to synthe- subtilis (ispA) (Xue et al. 2015). The additional expres - size C carotenoids, we have developed a plasmid-less, sion of the fpps gene from Saccharomyces cerevisiae marker-free, B. subtilis strain that can serve as a stepping also increased the supply of the precursor FPP (Song et stone for further genetic engineering and fermentation al. 2021). We therefore conclude that the heterologous process optimization targeted at a sustainable and effi - expression of FPPS from B. megaterium increased C30 cient production of C carotenoids. carotenoid biosynthesis in B. subtilis, similarly to the val- ues obtained when an extra copy of the native IspA was Supplementary Information The online version contains supplementary material available at https://doi. overexpressed (Xue et al. 2015). It has also been reported org/10.1186/s13568-023-01542-x. that attenuation of a competing FPP-consuming path- way toward C55 heptaprenyl diphosphate contributed Supplementary Material 1 to a 1.15-fold increase in terpenoid synthesis (Song et al. 2021). Accordingly, we assumed that abolishing non- Acknowledgements essential expression of yisP, the only phosphatase that We thank Dr. Isamu Maeda for providing us with the plasmid pHY_crtMN. catalyzes the conversion of FPP to farnesol, would also Author contribution lead to less FPP consumption in this competing path- PP designed research. OF and JF conducted experiments. PP analyzed the way, and the resulting extra FPP could be used by CrtMN data. OF, JF and PP wrote the manuscript. All authors read and approved the manuscript. enzymes to increase C carotenoid yield. In the ΔyisP mutant, known to exhibit no FPP phosphatase activity Funding information (Feng et al. 2014), excess FPP was distributed to increase This work was supported by the Pla de Doctorats Industrials del Departament de Recerca i Universitats de la Generalitat de Catalunya and Gestió d’ Ajuts the carotenoid yield in the engineered strain 1.39-fold Universitaris de Recerca for grant number 2021 DI 77. (Fig.  2; Table  2). Thus, for the first time, the role of yisP knockout in an increased accumulation of C carot- 30 Data availability The datasets generated during and/or analyzed during the current study are enoids in B. subtilis was demonstrated. available from the corresponding author on reasonable request. Cell engineering techniques have been previously used to improve C carotenoid productivity in E. coli and B. Declarations subtilis. E. coli strains were engineered to accumulate C carotenoids, with production levels ranging from 30 Conflict of interest The authors declare no financial or commercial conflict of interest. 0.5  mg/ gDCW to 10.8  mg/L (Chae et al. 2010; Kim et al. 2010, 2022; Takemura et al. 2021). B. subtilis has also Ethical statement been engineered using two-plasmid systems comprising This article does not describe any studies with human participants or animals performed by any of the authors. pHY_crtMN (Yoshida et al. 2009), mediating crtMN gene overexpression, and xylose-inducible pHCMC04G (Xue Received: 5 February 2023 / Accepted: 5 April 2023 et al. 2015), mediating stable overexpression of all MEP pathway enzymes. In total, the yield of C carotenoids achieved was 21  mg/g DCW, the highest production in Filluelo et al. AMB Express (2023) 13:38 Page 10 of 11 References Liu H, Xu W, Chang X, Qin T, Yin Y, Yang Q (2016) 4,4’-diaponeurosporene, a C Abdallah II, Xue D, Pramastya H, van Merkerk R, Setroikromo R, Quax WJ (2020) A carotenoid, effectively activates dendritic cells via CD36 and NF-kappaB regulated synthetic operon facilitates stable overexpression of multigene ter- signaling in a ROS independent manner. Oncotarget 7(27):40978–40991. penoid pathway in Bacillus subtilis. J Ind Microbiol Biotechnol 47(2):243–249. https://doi.org/10.18632/oncotarget.9800 https://doi.org/10.1007/s10295-019-02257-4 Liu H, Xu W, Yu Q, Yang Q (2017) 4,4’-Diaponeurosporene-producing Bacillus subtilis Altenbuchner J (2016) Editing of the Bacillus subtilis genome by the CRISPR-Cas9 increased mouse resistance against Salmonella typhimurium infection in a System. Appl Environ Microbiol 15(17):5421–5427. https://doi.org/10.1128/ CD36-Dependent manner. Front Immunol 8:483. https://doi.org/10.3389/ AEM.01453-16 fimmu.2017.00483 Britton G, Liaaen-Jensen S, Pfander H (2004) Carotenoids Handbook. Birkhäuser Marshall JH, GJ Wilmoth (1981) Pigments of Staphylococcus aureus, a series of Verlag triterpenoid carotenoids. J Bacteriol 147:900–913. https://doi.org/10.1128/ Cardoso LAC, Karp SG, Vendruscolo F, Kanno KYF, Zoz LIC, Carvalho JC (2017) jb.147.3.900-913.1981 Biotechnological Production of Carotenoids and their applications in Mignon C, Sodoyer R, Werle B (2015) Antibiotic-free selection in biotherapeu- Food and Pharmaceutical Products. Carotenoids. InTech. doi:https://doi. tics: now and forever. Pathogens 4(2):157–181. https://doi.org/10.3390/ org/10.5772/67725 pathogens4020157 Chae HS, Kim K-H, Kim SC, Lee PC (2010) Strain-dependent carotenoid produc- Moser S, Pichler H (2019) Identifying and engineering the ideal microbial terpe- tions in metabolically engineered Escherichia coli. Appl Biochem Biotechnol noid production host. Appl Microbiol Biotechnol 103(14):5501–5516. https:// 162(8):2333–2344. https://doi.org/10.1007/s12010-010-9006-0 doi.org/10.1007/s00253-019-09892-y Earl AM, Losick R, Kolter R (2008) Ecology and genomics of Bacillus subtilis. Trends Pelz A, Wieland KP, Putzbach K, Hentschel P, Albert K, Götz F (2005) Structure and Microbiol 16(6):269–275. https://doi.org/10.1016/j.tim.2008.03.004 biosynthesis of staphyloxanthin from Staphylococcus aureus. J Biol Chem Feng X, Hu Y, Zheng Y, Zhu W, Li K, Huang CH, Ko TP, Ren F, Chan HC, Nega M, 280:32493–32498. https://doi.org/10.1074/jbc.M505070200 Bogue S, López D, Kolter R, Götz F, Guo RT, Oldfield E (2014) Structural and Popp PF, Dotzler M, Radeck J, Bartels J, Mascher T (2017) The Bacillus BioBrick Box functional analysis of Bacillus subtilis YisP reveals a role of its product in 2.0: expanding the genetic toolbox for the standardized work with Bacillus biofilm production. Chem Biol 21(11):1557–1563. https://doi.org/10.1016/j. subtilis. Sci Rep 7(1):15058. https://doi.org/10.1038/s41598-017-15107-z chembiol.2014.08.018 Schallmey M, Singh A, Ward OP (2004) Developments in the use of Bacillus species Ferrando J, Filluelo O, Zeigler DR, Picart P (2023) Barriers to simultaneous multilo- for industrial production. Can J Microbiol 50:1–17. https://doi.org/10.1139/ cus integration in Bacillus subtilis tumble down: development of a straightfor- w03-076 ward screening method for the colorimetric detection of one-step multiple Siziya IN, Hwang CY, Seo MJ (2022) Antioxidant potential and capacity of micro- gene insertion using the CRISPR-Cas9 system. Microb Cell Fact 22:21. https:// organism-sourced C Carotenoids-A review. Antioxid (Basel) 11(10):1963. doi.org/10.1186/s12934-023-02032-2 https://doi.org/10.3390/antiox11101963 Furubayashi M, Li L, Katabami A, Saito K, Umeno D (2014) Construction of carot- Song Y, Nikoloff JM, Fu G, Chen J, Li Q, Xie N, Zheng P, Sun J, Zhang D (2016) enoid biosynthetic pathways using squalene synthase. FEBS Lett 588(3):436– Promoter screening from Bacillus subtilis in various Conditions Hunting for 442. https://doi.org/10.1016/j.febslet.2013.12.003 Synthetic Biology and Industrial Applications. PLoS ONE 11(7):e0158447. García-Moyano A, Larsen Ø, Gaykawad S, Christakou E, Boccadoro C, Puntervoll P, https://doi.org/10.1371/journal.pone.0158447 Bjerga GEK (2020) Fragment Exchange plasmid tools for CRISPR/Cas9-Medi- Song Y, He S, Abdallah II, Jopkiewicz A, Setroikromo R, van Merkerk R, Tepper PG, ated gene integration and protease production in Bacillus subtilis. Appl Envi- Quax WJ (2021) Engineering of multiple modules to improve Amorphadi- ron Microbiol 87(1):e02090–e02020. https://doi.org/10.1128/AEM.02090-20 ene production in Bacillus subtilis using CRISPR-Cas9. J Agric Food Chem Guan Z, Xue D, Abdallah II, Dijkshoorn L, Setroikromo R, Lv G, Quax WJ (2015) Meta- 69(16):4785–4794. https://doi.org/10.1021/acs.jafc.1c00498 bolic engineering of Bacillus subtilis for terpenoid production. Appl Microbiol Su Y, Liu C, Fang H, Zhang D (2020) Bacillus subtilis: a universal cell factory for indus- Biotechnol 99(22):9395–9406. https://doi.org/10.1007/s00253-015-6950-1 try, agriculture, biomaterials and medicine. Microb Cell Fact 19(1):173. https:// Hartz P, Milhim M, Trenkamp S, Bernhardt R, Hannemann F (2018) Characterization doi.org/10.1186/s12934-020-01436-8 and engineering of a carotenoid biosynthesis operon from Bacillus megate- Takaichi S (2000) Characterization of carotenes in a combination of a C(18) rium. Metab Eng 49:47–58. https://doi.org/10.1016/j.ymben.2018.07.017 HPLC column with isocratic elution and absorption spectra with a Huang K, Zhang T, Jiang B, Yan X, Mu W, Miao M (2017) Overproduction of Rum- photodiode-array detector. Photosynth Res 65(1):93. https://doi.org/10.102 meliibacillus pycnus arginase with multi-copy insertion of the argr-pyc 23 3/A:1006445503030 cassette into the Bacillus subtilis chromosome. Appl Microbiol Biotechnol Takaichi S, Inoue K, Akaike M, Kobayashi M, Oh-oka H, Madigan MY (1997) The 101:6039–6048. https://doi.org/10.1007/s00253-017-8355-9 major carotenoid in all known species of heliobacteria is the C carotenoid Ishiwa H, Shibahara H (1985) New shuttle vectors for Esch- 4,4’-diaponeurosporene, not neurosporene. Arch Microbiol 168:277–281. erichia coli and Bacillus subtilis. Jpn J Genet 60:235–243. https://doi. https://doi.org/10.1007/s002030050499 org/10.1016/0378-1119(84)90161-6 Takemura M, Takagi C, Aikawa M, Araki K, Choi SK, Itaya M, Shindo K, Misawa N Jing Y, Liu H, Xu W, Yang Q (2017) Amelioration of the DSS-induced colitis in mice (2021) Heterologous production of novel and rare C -carotenoids using Pla- bypretreatment with 4,4′-diaponeurosporene-producing Bacillus subtilis. Exp nococcus carotenoid biosynthesis genes. Microb Cell Fact 20(1):194. https:// Ther Med 14(6):6069–6073. https://doi.org/10.3892/etm.2017.5282 doi.org/10.1186/s12934-021-01683-3 Jing Y, Liu H, Xu W, Yang Q (2019) 4,4′-Diaponeurosporene-producing Bacillus subti- Toymentseva AA, Schrecke K, Sharipova MR, Mascher T (2012) The LIKE lis promotes the development of the mucosal immune system of the piglet system, a novel protein expression toolbox for Bacillus subtilis based gut. Anat Rec (Hoboken) 302:1800–1807. https://doi.org/10.1002/ar.24102 on the liaI promoter. Microb. Cell Fact 11:143. doi: https://doi. Julsing MK, Rijpkema M, Woerdenbag HJ, Quax WJ, Kayser O (2007) Functional org/10.1186/1475-2859-11-143 analysis of genes involved in the biosynthesis of isoprene in Bacillus subtilis. Wagner WP, Helmig D, Fall R (2000) Isoprene biosynthesis in Bacillus subtilis via Appl Microbiol Biotechnol 75(6):1377–1384. https://doi.org/10.1007/ the methylerythritol phosphate pathway. J Nat Prod 63:37–40. https://doi. s00253-007-0953-5 org/10.1021/np990286p Kim J, Kong MK, Lee SY, Lee PC (2010) Carbon sources-dependent carotenoid Wang J-J, Rojanatavorn K, Shih JCH (2004) Increased production of Bacillus production in metabolically engineered Escherichia coli. World J Microbiol keratinase by chromosomal integration of multiple copies of the kerA gene. Biotechnol 26(12):2231–2239. https://doi.org/10.1007/s11274-010-0408-5 Biotechnol Bioeng 87:459–464. https://doi.org/10.1002/bit.20145 Kim M, Jung DH, Hwang CY, Siziya IN, Park YS, Seo MJ (2022) 4,4’-Diaponeu- Wenzel M, Müller A, Siemann-Herzberg M, Altenbuchner J (2011) Self-inducible rosporene production as C carotenoid with antioxidant activity in Bacillus subtilis expression system for reliable and inexpensive protein recombinant Escherichia coli. Appl Biochem Biotechnol Sep 6. https://doi. production by high-cell-density fermentation. Appl Environ Microbiol org/10.1007/s12010-022-04147-5 77(18):6419–6425. https://doi.org/10.1128/AEM.05219-11 Kuzma J, Nemecek-Marshall M, Pollock WH, Fall R (1995) Bacteria produce the Widner B, Thomas M, Sternberg D, Lammon D, Behr R, Sloma A (2000) Develop- volatile hydrocarbon isoprene. Curr Microbiol 30(2):97–103. https://doi. ment of marker-free strains of Bacillus subtilis capable of secreting high org/10.1007/BF00294190 levels of industrial enzymes. J Ind Microbiol Biotech 25:204–212. https://doi. Lim HN, Lee Y, Hussein R (2011) Fundamental relationship between operon orga- org/10.1038/sj.jim.7000051 nization and gene expression. Proc Natl Acad Sci U S A 108(26):10626–10631. Wieland B, Feil C, Gloria-Maercker E, Thumm G, Lechner M, Bravo JM, Poralla K, https://doi.org/10.1073/pnas.1105692108 Goëtz F (1994) Genetic and biochemical analyses of the biosynthesis of the Filluelo et al. AMB Express (2023) 13:38 Page 11 of 11 yellow carotenoid 4,40 -diaponeurosporene of Staphylococcus aureus. J Bac- mpr gene into the Bacillus subtilis chromosome. Microb Cell Fact 10:64–73. teriol 176:7719–7726. https://doi.org/10.1016/S0021-9258(19)39076-3 https://doi.org/10.1186/1475-2859-10-64 Wu G, Yan Q, Jones JA, Tang YJ, Fong SS, Koffas MAG (2016) Metabolic burden: Yoshida K, Ueda S, Maeda I (2009) Carotenoid production in Bacillus subtilis Cornerstones in Synthetic Biology and Metabolic Engineering Appli- achieved by metabolic engineering. Biotechnol Lett 31(11):1789–1793. cations. Trends Biotechnol 34(8):652–664. https://doi.org/10.1016/j. https://doi.org/10.1007/s10529-009-0082-6 tibtech.2016.02.010 Zhang C, Hong K (2020) Production of terpenoids by Synthetic Biology Xue D, Abdallah II, de Haan IE, Sibbald MJ, Quax WJ (2015) Enhanced C carot- Approaches. Front Bioeng Biotechnol 8:347. https://doi.org/10.3389/ enoid production in Bacillus subtilis by systematic overexpression of MEP fbioe.2020.00347 pathway genes. Appl Microbiol Biotechnol 99(14):5907–5915. https://doi. Zhao J, Li Q, Sun T, Zhu X, Xu H, Tang J, Zhang X, Ma Y (2013) Engineering central org/10.1007/s00253-015-6531-3 metabolic modules of Escherichia coli for improving beta-carotene produc- Yasbin RE, Wilson GA, Young FE (1975) Transformation and transfection in lyso- tion. Metab Eng 17:42–50. https://doi.org/10.1016/j.ymben.2013.02.002 genic strains of Bacillus subtilis: evidence for selective induction of prophage in competent cells. J Bacteriol 121:296–304. https://doi.org/10.1128/ Publisher’s Note jb.121.1.296-304.1975 Springer Nature remains neutral with regard to jurisdictional claims in Yomantas YA, Abalakina EG, Golubeva LI, Gorbacheva LY, Mashko SV (2011) published maps and institutional affiliations. Overproduction of Bacillus amyloliquefaciens extracellular glutamylendopep- tidase as a result of ectopic multi-copy insertion of an efficiently expressed

Journal

AMB Express – Springer Journals

Published: Apr 29, 2023

Keywords: B. subtilis; C30 carotenoids; CRISPR-Cas9; Metabolic engineering

There are no references for this article.