Production of (2S)-sakuranetin from (2S)-naringenin in Escherichia coli by strengthening methylation process and cell resistance

(2S)-Sakuranetin is a 7-O-methylflavonoid that has anticancer, antiviral, and antimicrobial activities. Methylation process is involved in biosynthesizing (2S)-sakuranetin from (2S)-naringenin, in which S-adenosylmethionine (SAM) serves as the methyl donor. In this study, after methyl donor and substrate inhibition were identified as limiting factors for (2S)-sakuranetin biosynthesis, an efficient (2S)-sakuranetin-producing strain was constructed by enhancing methyl donor supply and cell tolerance to (2S)-naringenin. Firstly, PfOMT3 from Perilla frutescens was selected as the optimal flavonoid 7-O-methyltransferase (F7-OMT) for the conversion of (2S)-naringenin to (2S)-sakuranetin. Then, the methylation process was upregulated by regulating pyridoxal 5′-phosphate (PLP) content, key enzymes in methionine synthesis pathway, and the availability of ATP. Furthermore, genes that can enhance cell resistance to (2S)-naringenin were identified from molecular chaperones and sRNAs. Finally, by optimizing the fermentation process, 681.44 mg/L of (2S)-sakuranetin was obtained in 250-mL shake flasks. The titer of (2S)-sakuranetin reached 2642.38 mg/L in a 5-L bioreactor, which is the highest titer ever reported. This work demonstrates the importance of cofactor PLP in methylation process, and provides insights to biosynthesize other O-methylated flavonoids efficiently in E. coli.

(2S)-Sakuranetin is produced in plants in response to ambient pressures. For example, ultraviolet light or pest infection could activate 7-Omethyltransferase  in Oryza sativa to synthesize (2S)-sakuranetin [11]. Chemical synthesis and biosynthesis have been used for the synthesis of (2S)-sakuranetin. In a previous study, (2S)-sakuranetin was synthesized by chemically utilizing the regioselective deacetylation of naringenin triacetate [12]. However, this method is cumbersome and requires harsh reaction conditions. With the development of metabolic engineering and genetic engineering, (2S)-sakuranetin could be produced from (2S)-naringenin by flavonoid 7-O-methyltransferases (F7-OMTs) catalysis in E. coli. In a previous report, 40 mg/L of (2S)-sakuranetin was obtained by constructing a de novo biosynthesis pathway in E. coli [13]. 79 mg/L of (2S)-sakuranetin was produced in a 2.5-L bioreactor by using an E. coli-E. coli modular co-culture system [14].
F7-OMTs are S-adenosylmethionine (SAM)-dependent flavonoid methyltransferases that can transfer the methyl group from SAM to the 7-OH of flavonoids [6]. The limited supply of methyl donor in E. coli can restrict the accumulation of (2S)-sakuranetin. Previous studies focused on balancing the biosynthesis of (2S)-naringenin during the biosynthetic Peer review under responsibility of KeAi Communications Co., Ltd.
process of (2S)-sakuranetin, and there are no studies on strengthening the methylation of (2S)-naringenin to improve the production of (2S)-sakuranetin. In addition, flavonoids can interact with DNA, inhibit DNA gyrase [15] and damage the cytoplasmic membrane structure [16], which would interfere with the function of E. coli. Considering that microbes have developed certain mechanisms to adapt to environmental stress, various genetic strategies can be used to alleviate cell pressure, such as overexpression of molecular chaperones to refold the denatured proteins [17] and small regulatory RNAs (sRNAs) to alleviate stress [18,19]. Overexpression of molecular chaperones or sRNAs related to E. coli stress response might enhance cell tolerance to flavonoids so as to relieve substrate inhibition.
In this study, methylation process and substrate tolerance were enhanced to obtain (2S)-sakuranetin from (2S)-naringenin efficiently. Firstly, the optimal F-7OMT was selected. Then key enzymes (MetA, CysE) and cofactors (NADPH, ATP) essential for SAM biosynthesis were regulated to promote the methylation reaction. Moreover, the availability of an indispensable cofactor in the transsulfuration pathway, pyridoxal 5 ′ -phosphate (PLP), was enhanced to increase the production of (2S)-sakuranetin. Additionally, sRNA RpoS was identified to be effective in enhancing cell tolerance to (2S)-naringenin. Finally, via the combinatorial expression of genes that were beneficial for (2S)-sakuranetin accumulation, and the optimization of fermentation process, 681.44 mg/L of (2S)-sakuranetin was obtained in shake-flask scale. The titer of (2S)-sakuranetin reached 2642.38 mg/L when scaled up in a 5-L bioreactor.

Strains, plasmids, and genes
The strains used in this study are listed in Table 1. E. coli JM109 was used for plasmid propagation. E. coli BL21(DE3) was used for protein expression. The genes (OsNOMT, GenBank accession no. AB692949.1; SaOMT2, GenBank accession no. BAC70093; PfOMT3, GenBank accession no. MT909556) were synthesized by Sangon Biotech (Shanghai, China). The pET-28a (+), pACYCDuet-1 and pCDFDuet-1 expression vectors were obtained from Novagen (Darmstadt, Germany). Highfidelity Phusion DNA polymerase from Vazyme (Nanjing, China) was used for amplification. Seamless cloning kit purchased from Sangon Biotech (Shanghai) was used for plasmid assembly. The process for plasmids construction and primers are listed in the Supplementary File and Table S1, respectively.

Growth media and shake flask culture condition
The Luria-Bertani (LB) medium was used for culturing seed liquid. The Terrific Broth (TB) medium was used for (2S)-sakuranetin accumulation. Strains were cultured at 37 • C for 10 h in LB medium. Then a 2% (v/v) of LB strain culture was inoculated into the TB medium with shaking at 220 rpm. Isopropyl beta-D-thiogalactoside (IPTG) was added to the medium at a final concentration of 0.1 mM when OD 600 reached 0.8-1, at which time the temperature was shifted from 37 • C to 25 • C. 350 mg/L of (2S)-naringenin was added after adding IPTG for 3 h and 13 h to a final concentration of 700 mg/L. The total fermentation time in shake-flask scale was 26 h.
This study

Fermentation condition in bioreactor
Bioreactor fermentation was conducted in a 5-L glass bioreactor (T&J Bioengineering, Shanghai, China) containing 2.5 L of TB medium. The (2S)-naringenin was dissolved in methanol to a concentration of 50 g/L. The optimal strain was selected for scale-up culture in a 5-L bioreactor. Colonies from the culture plate were inoculated into LB liquid medium and incubated at 37 • C with shaking at 220 rpm for 10 h. Then, 4% (v/v) inoculum of seed culture was inoculated into a 5 L bioreactor containing 2.5 L of the TB medium. In the early stage, the fermentation temperature was kept at 37 • C and the rotating speed was 400 rpm. IPTG was added at a final concentration of 0.1 mM when the OD 600 reached 6, at which time the temperature was shifted to 25 • C and the stirring speed was changed to the pattern of depending on dissolved oxygen content (40%). (2S)-Naringenin was fed in batches at 6.5 h, 12.5 h, 18.5 h, 21.5 h, 41.5 h, and 48 h to a final concentration of 4 g/L. 500 g/L of glycerol was fed at a speed of 8 mL/h after substrate was added, and 6 mM Mg 2+ was added at 6.5 h. The pH was maintained at 7 ± 0.1 with 6 M NaOH during the fermentation.

Analytical methods
To quantitatively analyze the titer of (2S)-sakuranetin, a 500 μL fermentation sample was mixed with 500 μL methanol. Then, the supernatant was filtered using a 0.22-μm organic phase filter membrane after 12000g centrifugation for 5 min. The samples were detected by HPLC (Shimadzu Corporation, Japan) equipped with a Thermo Fisher C18 column (4.6 mm × 250 mm, 5 μm) at 290 nm. The mobile phase was as follows: A phase was water containing 0.1% TFA, and B phase was acetonitrile containing 0.1% TFA. The flow rate was 1 mL/min. The following procedure was used: 0-10 min, 10-60% B; 10-20 min, 40-80% B; 20-25 min, 80-10% B. The determination of the intracellular PLP level followed the procedure described by Cabo et al. [20].  2.5. Spot assays to assess E. coli substrate tolerance E. coli only harboring PfOMT3 and E. coli harboring cell tolerance genes were cultured at 37 • C in TB medium with shaking at 220 rpm. IPTG was added to the medium at a final concentration of 0.1 mM when OD 600 reached 0.8. After 10 h post-induction at 25 • C, seeds were diluted by gradient (10 − 2 , 10 − 3 , 10 − 4 , 10 − 5 ), and 2 μL of seeds dilution were spotted on solid LB plates containing (2S)-naringenin (200 mg/L, 300 mg/L or 400 mg/L). These LB plates were then incubated for 12 h at 37 • C.

Regulation of methionine and ATP contents to improve (2S)sakuranetin production
Methyl donor SAM is involved in the conversion of (2S)-naringenin to (2S)-sakuranetin (Fig. 1). Exogenously supply of methionine can enter into cells and improve the intracellular SAM availability [23]. To test the importance of methyl donor in the biosynthesis of (2S)-sakuranetin, 250 mg/L (2S)-naringenin and different concentrations of methionine were added after cells were induced for 10 h at 25 • C (Fig. 3A). When 1 g/L methionine was added, 129.21 mg/L of (2S)-sakuranetin was obtained after reaction for 10 h, which was 36.17% higher than that of the strain 7-FOMT-3. In order to enhance methionine production of E. coli, metA (encoding homoserine succinyltransferase) and cysE (encoding L-serine O-acetyltransferase) were overexpressed (strain NS01). (2S)-Sakuranetin titer of strain NS01 reached 142.59 mg/L without the addition of methionine, which was 50.26% higher than the strain 7-FOMT-3. When metK (encoding SAM synthetase) was overexpressed in strain NS01, 135.26 mg/L of (2S)-sakuranetin was obtained, a slight decrease compared with the strain NS01.
ATP is consumed in the conversion of methionine to SAM, and is an important cofactor in methionine biosynthesis pathway. ydaO, which can dynamically regulate ATP accumulation in bacteria, was heterologously expressed in strain NS01 and NS02, resulting in strain NS03 and NS04, respectively; the (2S)-sakuranetin titer increased by 62.38% and 124.45%, respectively, compared with the strain 7-FOMT-3 (Fig. 3B). NADPH is another important cofactor in the methionine biosynthesis pathway. The NADH kinase-encoding gene POS5 from S. cerevisiae was heterogeneously expressed in strain 7-FOMT-3 (resulting in strain NS14), improving the titer of (2S)-sakuranetin to 132.58 mg/L, which was 43.35% higher than that of the strain 7-FOMT-3 (Fig. 3C). These results suggested that the SAM pool is a limiting factor for the conversion of (2S)-naringenin to (2S)-sakuranetin. Regulating the levels of methionine and ATP are essential for (2S)-sakuranetin production.

Effects of combinatorial expression of methylation-enhancing genes on (2S)-sakuranetin production
Genes that can upregulate methylation reaction were expressed by combination. To avoid the negative effect of high concentration of (2S)naringenin and obtain a higher titer of (2S)-sakuranetin, (2S)-naringenin was added in batches (final concentration was 500 mg/L, 600 mg/L, and 700 mg/L, respectively), in which the half quantity of (2S)naringenin was added after adding IPTG for 1 h, and another 50% of (2S)-naringenin was added after 10 h. The result showed that strain NS17 was the best to accumulate (2S)-sakuranetin, and 451.47 mg/L of (2S)-sakuranetin was obtained when added 500 mg/L (2S)-naringenin (Fig. 5). When POS5 was expressed with PfOMT3 (strain NS14), this could promote the production of (2S)-sakuranetin, but (2S)-sakuranetin titer decreased when POS5 was co-expressed with genes that enhanced the methylation reaction. Strain NS18 (expressing POS5 in NS17) accumulated only 191.56 mg/L of (2S)-sakuranetin when added 500 mg/L (2S)-naringenin, resulting in a 57.57% decrease compared with the strain NS17. Based on the (2S)-sakuranetin titer, genes combination in strain NS17 was the best for strengthening the methylation process.

Fermentation process optimization to improve (2S)-sakuranetin production
The production of (2S)-sakuranetin was further improved when substrate tolerance enhancement genes rpoS, secB and methylation enhancement genes were co-expressed. The result showed strain NS34 accumulated the highest titer of (2S)-sakuranetin; 545.98 mg/L of (2S)sakuranetin was obtained when 600 mg/L (2S)-naringenin was added (Fig. 7A), which was 95.43% higher than the optimal strain improving cell substrate tolerance alone, and was 131.71% higher than the optimal strain enhancing methylation process alone.
To determine the optimal substrate addition time point, (2S)-naringenin was first supplemented after IPTG was added for 1 h, 3 h, 5 h, 7 h and 9 h (final concentration was 700 mg/L), respectively. The result showed that the optimal substrate addition time was 3 h after the addition of IPTG (Fig. 7B). In addition, considering that Mg 2+ can activate the leaving group (ATP and giving phosphate) in the biosynthesis of SAM [43] and that SAM synthetase requires Mg 2+ , different concentrations of Mg 2+ were added. 681.44 mg/L of (2S)-sakuranetin was obtained when added 6 mM Mg 2+ (Fig. 7C). Strain NS34 was used for scale-up culturing in a 5-L bioreactor, and 2642.38 mg/L of (2S)-sakuranetin was obtained at 70 h from 4 g/L (2S)-naringenin (Fig. 7D).

Discussion
Previous studies have revealed that E. coli has the ability to express SAM-dependent F7-OMTs. In this study, PfOMT3 was selected as the optimal F7-OMT to convert (2S)-naringenin to (2S)-sakuranetin. Given that exogenously supplied of methionine could improve (2S)-sakuranetin titer and that the conversion efficiency decreased with increasing concentration of (2S)-naringenin, methyl donor and cell tolerance to substrate were considered as limiting factors for (2S)-sakuranetin biosynthesis. Experiments were performed to upregulate methionine, cofactor ATP and PLP content to enhance methylation process, and to select the best molecular chaperone, sRNA or ribosomal subunits to improve strain tolerance to (2S)-naringenin. No previous studies have focused on enhancing the methylation process and improving the bacterial resistance to achieve the efficient synthesis of (2S)-sakuranetin.
The methyl donor SAM, biosynthesized from methionine and ATP by SAM synthetase (encoded by metK in E. coli), plays an important role in methylation process. As the direct precursor of SAM, methionine content was enhanced by upregulating key enzymes MetA [44] and CysE [45] in methionine biosynthesis pathway, and the production of (2S)-sakuranetin increased by 50.26% compared with strain 7-FOMT-3. (2S)-Sakuranetin titer decreased when metK was overexpressed, which may due to the overproduction of metK would consume large amount of ATP and cause a methionine deficiency [46], resulting in the imbalance of metabolic flux.
Methylation proceess was further upregulated by regulating ATP and NADPH contents, which are important cofactors for SAM biosynthesis. Previous studies proved that ydaO from B. subtilis can dynamically regulate ATP content in E. coli [47,48]. The accumulation of (2S)-sakuranetin increased by 162.38% in the ydaO-expressing strain compared with strain 7-FOMT-3. POS5 is known to be involved in NADPH regeneration in S. cerevisiae and can increase the intracellular NADPH availability in E. coli [49]. (2S)-Sakuranetin titer increased when POS5 was expressed in strain 7-FOMT-3. However, the titer of (2S)-sakuranetin decreased when POS5 was co-expressed with metA, cysE, ydaO, and the PLP biosynthesis genes. Considering that the disturbance of NADPH pool would trigger the redistribution of metabolic flux in E. coli [50], it was speculated that the redox and metabolic flux imbalance caused the decrease of (2S)-sakuranetin titer. In addition, the extreme metabolic burden of overexpression too many foreign genes may also result in the decline of (2S)-sakuranetin production.
Enzymes involved in the transsulfuration pathway utilize PLP as cofactor [51], whether the increase of PLP content could enhance the synthesis of (2S)-sakuranetin was explored. RFC4 and RPS18B are known can increase intracellular pyridoxine and pyridoxamine concentrations [25], and pyridoxine and pyridoxamine can be converted into PLP [52]. Given that genes involved in the transsulfuration pathway have a high homology between E. coli and S. cerevisiae [53], RPS18B and RFC4 from S. cerevisiae were heterologously expressed in E. coli. The (2S)-sakuranetin titer increased by 180.72% when SNZ3, RPS18B, and RFC4 were co-expressed compared with the strain 7-FOMT-3. This result provides a valuable insight to biosynthesize other methylated products in E. coli, and demonstrates the significance of PLP in methylation process.
Flavonoids are known to have antimicrobial properties; they can inhibit DNA gyrase [15], alter the cell membrane permeability, and change cellular morphology [54], which would affect the metabolism and function of cells. In addition, (2S)-sakuranetin titer decreased with increasing concentrations of (2S)-naringenin. Therefore, substrate resistance was considered as another limiting factor. Genes that can enhance cell tolerance to (2S)-naringenin were identified in this study. Strains expressing rpsQ His3 1 Pro , rpoS, and secB showed better performance on (2S)-sakuranetin production and cell growth. The spot assay reflected more intuitively that RpsQ His3 1 Pro , RpoS, and SecB protein could improve the (2S)-naringenin tolerance of host cells. This result provides an idea to alleviate cell stress in biosynthesizing other natural products with antibacterial effect in E. coli, and the selected genes may also be used to relieve cells pressure in the conversion of other flavonoids.
In conclusion, an efficient (2S)-sakuranetin-producing strain was constructed by F7-OMTs selection, methylation process enhancement, cell resistance improvement and fermentation process optimization. The titer of (2S)-sakuranetin reached 2642.38 mg/L in a 5-L bioreactor, which is the highest titer reported to date. Given that flavonoid Omethyltransferases (FOMTs) have substrate specificity [55], other valuable methylated flavonoids might be obtained efficiently based on selecting specific FOMTs and adopting strategy of this study.

Declaration of competing interest
The authors declare that they do not have any financial or commercial conflict of interest in connection with the work submitted. Development Program of China (2019YFA0904800), and the National Natural Science Foundation of China (21908078).