ABSTRACT
Steroid drug precursors, including C19 and C22 steroids, are crucial to steroid drug synthesis and development. However, C22 steroids are less developed due to the intricacy of the steroid metabolic pathway. In this study, a C22 steroid drug precursor, 9-hydroxy-3-oxo-4,17-pregadiene-20-carboxylic acid methyl ester (9-OH-PDCE), was successfully obtained from Mycolicibacterium neoaurum by 3-ketosteroid-Δ1-dehydrogenase and enoyl-CoA hydratase ChsH deficiency. The production of 9-OH-PDCE was improved by the overexpression of 17β-hydroxysteroid dehydrogenase Hsd4A and acyl-CoA dehydrogenase ChsE1-ChsE2 to reduce the accumulation of by-products. The purity of 9-OH-PDCE in fermentation broth was improved from 71.7% to 89.7%. Hence, the molar yield of 9-OH-PDCE was improved from 66.7% to 86.7%, with a yield of 0.78 g/L. Furthermore, enoyl-CoA hydratase ChsH1-ChsH2 was identified to form an indispensable complex in Mycolicibacterium neoaurum DSM 44704.
IMPORTANCE C22 steroids are valuable precursors for steroid drug synthesis, but the development of C22 steroids remains unsatisfactory. This study presented a strategy for the one-step bioconversion of phytosterols to a C22 steroid drug precursor, 9-hydroxy-3-oxo-4,17-pregadiene-20-carboxylic acid methyl ester (9-OH-PDCE), by 3-ketosteroid-Δ1-dehydrogenase and enoyl-CoA hydratase deficiency with overexpression of 17β-hydroxysteroid dehydrogenase acyl-CoA dehydrogenase in Mycolicibacterium. The function of the enoyl-CoA hydratase ChsH in vivo was revealed. Construction of the novel C22 steroid drug precursor producer provided more potential for steroid drug synthesis, and the characterization of the function of ChsH and the transformation of steroids further revealed the steroid metabolic pathway.
KEYWORDS: phytosterols, Mycolicibacterium, bioconversion, enoyl-CoA hydratase, 9-hydroxy-3-oxo-4, 17-pregadiene-20-carboxylic acid methyl ester
INTRODUCTION
Steroid drugs are widely used in human disease prevention and pharmaceutical therapy, including repression of virus duplication, anti-inflammation, alleviation of asthma and allergies, and regulation of hormones levels and blood pressure (1). The majority of steroid drugs can be chemically synthesized from C19 steroid intermediates (2). Furthermore, C22 steroid intermediates are also worthy sterol metabolites and showed better suitability for synthesizing progestational and adrenocortical hormones compared with C19 steroids (2). In the steroid drug industry, C19 steroids are the most widely used steroid drug precursors (1). In contrast, the production of C22 steroids is rarely explored, and efficient industrial strains which produce them have yet to be developed (3, 4). Therefore, developing novel C22 steroids will fulfill the needs of the steroid drug industry.
Phytosterol bioconversion by microorganisms is a vital method of steroid drug precursor production (4, 5). Phytosterol is an economical and readily available natural plant sterol (6) which exists in food industry waste, including deodorizer distillate in the vegetable oil refining process, sugar mud, corn flour, and filter cake in the sugarcane industry (7–10). Compared to traditional chemical synthesis, phytosterol bioconversion for C19 and C22 steroids shows evident advantages of being shorter, more environmentally friendly, and more cost-effective (1, 11, 12).
Current studies have focused on bioconversion of phytosterol by actinomycetes, especially mycolicibacteria (13). Actinobacteria, especially Mycolicbacterium, are reported to be capable of sterol bioconversion (14). Many mycolicibacteria species have been industrialized to produce various steroid intermediates (4, 11, 15, 16). Many C19 steroids have been successfully obtained during phytosterol bioconversion through engineered mycolicibacteria (17–19). However, the intricate steroid structure and unknown sterol biodegradation pathway enhance the difficulty of developing C22 steroids and their producing strains. Optimization of current industrial processes requires a better understanding of steroid metabolism (14). Recently, concerning the steroid metabolic pathway in mycolicibacteria, the 17-hydroxysteroid dehydrogenase/3,22-dioxo-25,26-bisnorchol-4-ene-24-oyl CoA dehydrogenase (Hsd4A) was found to be the key enzyme in manipulating the sterol metabolic flux to accumulate either C19 or C22 steroid products (Fig. 1A) (3). The sterol metabolite 3-oxo-4,17-pregnadiene-20-carboxyl-CoA (PDC-CoA) is a C22 steroid involved in the C19 steroid pathway. The enoyl-CoA hydratase ChsH1-ChsH2 was identified in the igr operon of M. tuberculosis H37Rv to catalyze the hydration of PDC-CoA in vitro. Structural and enzymatic analysis of ChsH1-ChsH2 from M. tuberculosis H37Rv revealed that ChsH1-ChsH2 forms an α2β2 heterotetrametric structure to hydrate the substrate. Although ChsH1-ChsH2 prefers to bind substrates such as steroid enoyl-CoA, it can also hydrate aliphatic enoyl-CoAs such as octenoyl-CoA and decanoyl-CoA, indicating its broad substrate specificity (20). However, the functions of ChsH1-ChsH2 in vivo have yet to be explored. Theoretically, disturbance of enoyl-CoA hydration may lead to the accumulation of new C22-steroid metabolites in vivo (Fig. 1A). A C22 steroid, 3-oxo-4-pregene-20-carboxylic acid methyl ester (PECE), was once identified as a by-product in the industrial strain Mycolicibacterium neoaurum NRRL 3805 for 4-androstene-3,17-dione (AD) production (21), which implied the possibility of accumulating novel C22-steroid metabolites. In this study, the enoyl-CoA hydration genes from Mycolicibacterium neoaurum DSM 44074 were identified and the enzymes were characterized. The ChsH1-ChsH2 complex is indispensable for function. Moreover, ChsH1-ChsH2 was determined to hydrate 9-hydroxy-3-oxo-4,17-pregadiene-20-carboxylic acid (9-OH-PDCA) into 9,17-dihydroxy-3-oxo-4,17-pregadiene-20-carboxylic acid (9,17-OH-PDCA). A C22 steroid metabolite, 9-OH-PDCE, was obtained in an engineered Mycolicibacterium by the loss of chsH2. Additionally, the productivity of 9-OH-PDCE was further enhanced by overexpressing the key genes hsd4A, chsE1, and chsE2. As a result, this study elucidated the ChsH1-ChsH2 complex in vivo and constructed a 9-OH-PDCE-producing strain from phytosterol bioconversion with a considerable molar yield, which was achieved by a ChsH deficiency.
FIG 1.
Schematic diagram of phytosterol side chain degradation in mycolicibacteria. (A) Schematic profiles of 9-hydroxy-3-oxo-4,17-pregadiene-20-carboxylic acid methyl ester (9-OH-PDCE) production. ChoMs, cholesterol oxidases; Cyps, cytochrome P450 proteins; ChsEs, acyl-CoA dehydrogenases; Ltps, lipid transferases; FadA5, thiolase; Kshs, 3-ketosteroid-9-alpha-hydroxylases; PDC-CoA, 3-oxo-4,17-pregnadiene-20-carboxyl-CoA; 9-OH-ADD, 9α-hydroxyandrostene-1,4-diene-3,17-dione. (B) Localization of igr operon in the genome of M. neoaurum DSM 44074.
RESULTS
Characterization of ChsH from Mycolicibacterium neoaurum DSM 44074.
The enzyme ChsH1-ChsH2 in Mycobacterium tuberculosis H37Rv was determined to catalyze the hydration of 3-oxo-4,17-pregnadiene-20-carboxyl-CoA (PDC-CoA), resulting in 17-hydroxy-3-oxo-4-pregnene-20-carboxyl-CoA (17-OH-PEC-CoA). The encoding genes chsH1 and chsH2 are located in the intracellular growth operon (igr). Through genome sequencing, an igr operon identified in M. neoaurum DSM 44074 was identified to have the same structure as that of M. tuberculosis H37Rv (Fig. 1B). The two components of hydratase were predicted and named ChsH1 and ChsH2. The ChsH1 and ChsH2 in Mycolicibacterium neoaurum DSM 44074 show 75.0% and 72.0% amino acid identities with those in M. tuberculosis H37Rv, respectively. Three-dimensional structures were predicted using the state-of-art protein structure prediction program RoseTTAFold (see Materials and Methods). As shown in Fig. 2A, ChsH1 and ChsH2 constituted a heterodimer to form an active site in their interface. The dimerized ChsH1/ChsH2 enzymes exhibited a similar conformation to that of ChsH1-ChsH2 from M. tuberculosis H37Rv (PDB: 4WNB, Fig. 2B), indicating that the ChsH1/ChsH2 complex is likely a cadmium-type hydratase (20). The hydration activity analysis was carried out as described in Materials and Methods. As shown in Fig. 2C, ChsH2 was soluble with His6 tag, while ChsH1 was expressed as a soluble protein with glutathione S-transferase (GST)-tag. Both His6-ChsH2 and GST-ChsH1 proteins were readily purified to a single band, which was shown on SDS-PAGE gel (Fig. 2C). GST pulldown experiments confirmed that ChsH1 and ChsH2 hydratase formed a complex (Fig. 2D). First, the monomer of ChsH1 and ChsH2 was assayed to verify whether ChsH1 or ChsH2 alone could hydrate the substrate 9-hydroxy-3-oxo-4,17-pregadiene-20-carboxylic acid (9-OH-PDCA). However, no transformation of 9-OH-PDCA was observed with ChsH1 or ChsH2. Furthermore, the purified ChsH1 and ChsH2 proteins were used to construct complexes for activity assays by high-performance liquid chromatography (HPLC) and thin-layer chromatography (TLC) using 9-OH-PDCA as the substrate. As shown in Fig. 2E, these two enzymes could significantly catalyze the hydration of 9-OH-PDCA to 9,17-OH-PDCA.
FIG 2.
Structural characterization and enzymatic analysis of ChsH1-ChsH2. (A) Structural analysis of the homology model of ChsH1 and ChsH2. One ChsH1 chain and one ChsH2 chain fold into a heterodimer. (B) ChsH1/H2 complexes have a similar conformation to the model template (PDB: 4WNB); (C) SDS-PAGE analysis of purified ChsH1 and ChsH2 proteins. M, molecular weight marker; 1, His6-ChsH2; 2, GST-ChsH1. (D) Glutathione S-transferase (GST)-pulldown analysis of ChsH1/2 complex. (E) High-performance liquid chromatography (HPLC) analysis of -hydroxy-3-oxo-4,17-pregadiene-20-carboxylic acid (9-OH-PDCA) products catalyzed by hydratase ChsH1/ChsH2 complex in vitro. Row 1, reaction; row 2, 9-OH-PDCA (substrate); row 3, 9,17-OH-PDCA (product which was purified by thin-layer chromatography [TLC]). Inset: TLC analysis of 9-OH-PDCA catalyzed by the hydratase ChsH1/ChsH2 complex in vitro. Column 4, partially purified 9-OH-PDCA (substrate); column 5, crude 9-OH-PDCA sample; column 6, reaction in 1% Tween 80 for 22 h; column 7, reaction in the absence of Tween 80 for 22 h. Spots for 9,17-OH-PDCA are marked with arrows.
Accumulation of 9-hydroxy-steroids by knocking out kstDs.
It has been reported that 9-hydroxy-C22-steroids can be easily converted into C21 corticosteroids, such as 9-hydroxy-progesterone, which can be readily used for steroid drug synthesis (22). The bioconversion of phytosterol involves degradation of the C17-side chain and steroid rings (23). Thus, the steroid skeleton degradation caused by 3-ketosteroid-9a-hydroxylase (KSH) and 3-ketosteroid-Δ1-dehydrogenase (KSTD) must be blocked in order for 9-hydroxy C22 steroids to accumulate (Fig. 1A) (23–25). Three kstDs were knocked out to accumulate 9α-hydroxy derivatives by Δ1-dehydrogenation deficiency in M. neoaurum DSM 44074, resulting in the ΔkstD strain (Fig. 3A). The growth curve revealed that ΔkstD showed a suppressed growth rate compared with M. neoaurum DSM 44704, indicating that KSTDs had a significant impact on the strains cultured with phytosterol and glucose (Fig. 3B) (24). Thus, ΔkstD was cultivated in a fermentation medium containing 1 g/L phytosterol. As shown in Fig. 4A, both 9-hydroxyandrost-4-en-3-one (9-OH-AD) (peak a) and 9,21-dihydroxy-20-methylpregn-4-en-3-one (9-OH-4-HP) (peak b) were identified in the phytosterol metabolites, and their yields reached 0.66 g/L and 0.019 g/L, respectively (Fig. 4B and C). The production of 9-OH-AD and 9-OH-4-HP from ΔkstD mutant indicated that the steroid nucleus degradation was blocked due to KSTD deficiency.
FIG 3.
Effects of KstD1, KstD2, KstD3, and ChsH2 on the cell growth of Mycolicibacterium neoaurum DSM 44074. (A) Deleted sequences kstD1, kstD2, kstD3, and chsH2. The PAM (protospacer adjacent motif) was selected as TTTN in the CRISPR-Cpf1 system. (B) Cell growth curve for strains in fermentation medium with 1 g/L phytosterol and 20 g/L glucose.
FIG 4.
Phenotype analysis of M. neoaurum DSM 44074 and ΔkstD. (A) HPLC analysis of M. neoaurum DSM 44074 and ΔkstD cultured with 1 g/L phytosterol. Peak a, 9-hydroxyandrost-4-en-3-one (9-OH-AD); peak b, 9,21-dihydroxy-20-methylpregn-4-en-3-one (9-OH-4-HP). (B) Time course of 9-OH-AD accumulation by M. neoaurum DSM 44074 and ΔkstD. (C) Time course of 9-OH-4-HP accumulation by M. neoaurum DSM 44074 and ΔkstD.
Accumulation of 9-OH-PDCE by knocking out chsH2.
The enoyl-CoA hydratase subunit ChsH2 was disturbed by deleting chsH2 in ΔkstD to determine the function of ChsH1-ChsH2 in sterol side chain degradation in vivo and accumulate C22 steroids (Fig. 3A), resulting in the ΔkstDΔchsH2 strain. Afterward, chsH2 was complemented in ΔkstDΔchsH2, resulting in the complement strain ΔkstDΔchsH2+chsH to verify the function of ChsH2. As shown in Fig. 3B, the growth of ΔkstDΔchsH2 was comparable to that of the ΔkstD strain. However, once chsH2 was complemented, the growth rate of the ΔkstDΔchsH2+chsH2 strain was recovered to a similar level to the ΔkstD strain. However, both ΔkstD and ΔkstDΔchsH2+chsH2 still grew slower than M. neoaurum DSM 444074 (Fig. 3B). All these growth discrepancies indicate that both KSTDs and ChsH greatly impact the bacterial growth cultured with phytosterol and glucose.
Thence, ΔkstDΔchsH2 and ΔkstDΔchsH2+chsH2 were cultured in a fermentation medium with 1 g/L phytosterol. As shown in Fig. 5A, no 9-OH-AD was obtained in the ΔkstDΔchsH2 strain incubation, indicating that ChsH activity was eliminated due to the loss of chsH2, and the abrogation of enoyl-CoA hydration activity blocked the C19 pathway involved in sterol side chain degradation. However, 9-OH-4-HP was still identified in the phytosterol metabolites (Fig. 5A, peak d). Four new products were identified and characterized as compounds b, c, e, and f in ΔkstDΔchsH2 incubation (Fig. 5A). The intermediates were separated and analyzed by liquid chromatography mass spectrometry (LC-MS) and were identified as 9-hydroxy-3-oxo-4,17-pregadiene-20-carboxylic acid (9-OH-PDCA, Fig. 5A, peak b), 9-hydroxy-3-oxo-4-pregene-20-carboxylic acid (9-OH-PECA, Fig. 5A, peak c), 9-hydroxy-3-oxo-4,17-pregadiene-20-carboxylic acid methyl ester (9-OH-PDCE, Fig. 5A, peak e) and 9-hydroxy-3-oxo-4-pregene-20-carboxylic acid methyl ester (9-OH-PECE, Fig. 5A, peak f), respectively. The main product obtained in ΔkstDΔchsH2 incubation was 9-OH-PDCE with a purity of 71.7%. The purities of the by-products 9-OH-PDCA, 9-OH-PECA, 9-OH-4-HP, and 9-OH-PECE were 1.3%, 1.2%, 6.8%, and 16.4%, respectively (Table 1).
FIG 5.
Phenotypic analysis of ΔkstDΔchsH2 and ΔkstDΔchsH2+chsH2. (A) HPLC chromatogram of ΔkstDΔchsH2 and ΔkstDΔchsH2+chsH2 cultured with 1 g/L phytosterol. Peak a, 9-OH-AD; peak b, 9-OH-PDCA; peak c, 9-OH-PECA (9-hydroxy-3-oxo-4-pregene-20-carboxylic acid); peak d, 9-OH-4-HP; peak e, 9-OH-PDCE; peak f, 9-OH-PECE (9-hydroxy-3-oxo-4-pregene-20-carboxylic acid methyl ester). (B) Time course of 9-OH-AD accumulation by ΔkstDΔchsH2 and ΔkstDΔchsH2+chsH2; (C) Time course of 9-OH-PDCA accumulation. (D) Time course of 9-OH-PECA accumulation. (E) Time course of 9-OH-PDCE accumulation. (F) Time course of 9-OH-PECE accumulation. (G) Time course of 9-OH-4-HP accumulation.
TABLE 1.
Relative purity of the steroid intermediates obtained in Mycolicibacterium neoaurum DSM 44074, ΔkstD, ΔkstDΔchsH2, and ΔkstDΔchsH2+chsH2 fed with 1 g/L phytosterola
| Strain | Steroid intermediate relative purity (%) |
||||||
|---|---|---|---|---|---|---|---|
| 9-OH-AD | 9-OH-PDCA | 9-OH-PECA | 9-OH-4-HP | 9-OH-PDCE | 9-OH-PECE | Others | |
| M. neoaurum DSM 44074 | u.d. | u.d. | u.d. | u.d. | u.d. | u.d. | u.d. |
| ΔkstD | 90.4 ± 2.1 | u.d. | u.d. | 5.6 ± 0.9 | u.d. | u.d. | 4.0 ± 1.5 |
| ΔkstDΔchsH2 | u.d. | 1.3 ± 0.8 | 1.2 ± 0.4 | 6.8 ± 1.1 | 71.7 ± 1.2 | 16.4 ± 0.4 | 2.5 ± 1.1 |
| ΔkstDΔchsH2+chsH2 | 88.1 ± 2.6 | u.d. | u.d. | 5.1 ± 0.4 | u.d. | u.d. | 6.8 ± 1.4 |
9-OH-PDCA, 9-hydroxy-3-oxo-4,17-pregadiene-20-carboxylic acid; 9-OH-PECA, 9-hydroxy-3-oxo-4-pregene-20-carboxylic acid; 9-OH-PDCE, 9-hydroxy-3-oxo-4,17-pregadiene-20-carboxylic acid methyl ester; 9-OH-PECE, 9-hydroxy-3-oxo-4-pregene-20-carboxylic acid methyl ester; 9-OH-AD, 9-hydroxyandrost-4-en-3-one; 9-OH-4-HP, 9,21-dihydroxy-20-methylpregn-4-en-3-one; u.d., undetectable.
No detectable 9-OH-AD was obtained from the fermentation broth of the ΔkstDΔchsH2 strain (Fig. 5B). The yields of both 9-OH-PECA and 9-OH-PDCA were 0.02 g/L (Fig. 5C and D). The main product from ΔkstDΔchsH2 fermentation broth was 9-OH-PDCE, with a yield of 0.6 g/L (Fig. 5E). The molar yield of 9-OH-PDCE was 66.7%. Another new C22 steroid intermediate obtained in ΔkstDΔchsH2 fermentation broth was 9-OH-PECE, with a yield of 0.18 g/L (Fig. 5F). Moreover, 9-OH-4-HP still accumulated in ΔkstDΔchsH2 fermentation, with a yield of 0.08 g/L (Fig. 5E).
As expected, the complement of ChsH2 fully recovered the ΔkstDΔchsH2 phenotype (Fig. 5A). With 1 g/L phytosterol fed, the ΔkstDΔchsH2+chsH2 strain accumulated 0.67 g/L 9-OH-AD along with 0.08 g/L 9-OH-4-HP, which is concident with the strain ΔkstD (Fig. 5A and G).
Enhancement of the 9-OH-PDCE accumulation by overexpressing hsd4A, chsE1, and chsE2.
The molar yield of 9-OH-PDCE from ΔkstDΔchsH2 was undesirable because of the accumulation of by-products. The 17β-hydroxysteroid dehydrogenase Hsd4A was previously reported to be responsible for the accumulation of 9-OH-4-HP (4, 26). Thus, a ΔkstDΔchsH2+hsd4A strain was created by overexpressing hsd4A to improve the production of 9-OH-PDCE and inhibit the accumulation of 9-OH-4-HP. As shown in Fig. 6A, the accumulation of 9-OH-4-HP from ΔkstDΔchsH2+hsd4A significantly decreased. The yield was 0.04 g/L in the ΔkstDΔchsH2+hsd4A strain incubation (Fig. 6B), which is 50% lower than that of ΔkstDΔchsH2. The molar yield decreased to 4.18%, compared to 9.32% for ΔkstDΔchsH2. In addition, 9-OH-PDCA, 9-OH-PECA, and 9-OH-PECE production were not influenced by hsd4A overexpression, with yields of 0.01, 0.02, and 0.17 g/L, respectively (Fig. 6C to E). In contrast, the yield of 9-OH-PDCE improved to 0.67 g/L from 0.60 g/L, an increase of 10.8% (Fig. 6F).
FIG 6.
Phenotypic analysis ofΔkstDΔchsH2, ΔkstDΔchsH2+hsd4A, ΔkstDΔchsH2+chsE1+chsE2 and ΔkstDΔchsH2+hsd4A+chsE1+chsE2. (A) HPLC chromatogram of the strains cultured with 1 g/L phytosterol. Peak a, 9-OH-PDCA; peak b, 9-OH-PECA; peak c, 9-OH-4-HP; peak d, 9-OH-PDCE; peak e, 9-OH-PECE. (B) Time course of 9-OH-4-HP accumulation. (C) Time course of 9-OH-PDCA accumulation. (D) Time course of 9-OH-PECA accumulation. (E) Time course of 9-OH-PECE accumulation. (F) Time course of 9-OH-PDCE accumulation.
9-OH-PECE was another by-product characterized in ΔkstDΔchsH2 with phytosterol incubation. The structural difference between 9-OH-PECE and 9-OH-PDCE is that C17 is reductive in 9-OH-PECE. It has been verified that acyl-CoA dehydrogenase ChsE1-ChsE2 from M. tuberculosis H37Rv catalyzes the dehydrogenation of 3-oxo-17-pregnene-20-carboxyl-CoA (PEC-CoA) to 3-oxo-4,17-pregnadiene-20-carboxyl-CoA (PDC-CoA) in vitro (27). The encoding genes, chsE1 and chsE2, are located in the igr operon upstream of chsH2. Considering that ChsE1-ChsE2 can dehydrogenate PEC-CoA at C-17, it may also catalyze the dehydrogenation of 9-OH-PECE to 9-OH-PDCE. Hence, chsE1 and chsE2 from M. neoaurum DSM 44074 were overexpressed in ΔkstDΔchsH2, resulting in the ΔkstDΔchsH2+chsE1+chsE2 strain. As shown in Fig. 6A, the production of 9-OH-PECE was distinctly suppressed. The yield of 9-OH-PECE was 0.03 g/L in ΔkstDΔchsH2+chsE1+chsE2 incubation (Fig. 6E). The yield of 9-OH-PECE decreased by 83.4% due to chsE1 and chsE2 expression, and other by-products were not influenced by chsE1 and chsE2 overexpression (Fig. 6B to D); while the yield of 9-OH-PDCE from ΔkstDΔchsH2+chsE1+chsE2 improved to 0.72 g/L, 20% higher than that from ΔkstDΔchsH2 (Fig. 6F). The molar yield of 9-OH-PDCE from ΔkstDΔchsH2+chsE1+chsE2 reached 80.0%. This suggests that the overexpression of chsE1 and chsE2 could enhance the transformation of 9-OH-PECE to 9-OH-PDCE, resulting in improved 9-OH-PDCE production.
Following these observations, hsd4A, chsE1 and chsE2 were overexpressed together in ΔkstDΔchsH2, resulting in the ΔkstDΔchsH2+hsd4A+chsE1+chsE2 strain. The yield of 9-OH-4-HP was 0.04 g/L in the ΔkstDΔchsH2+hsd4A+chsE1+chsE2 incubation (Fig. 6B), a decrease of 50% compared with that of ΔkstDΔchsH2. The production of 9-OH-PECE decreased by 88.9% compared with that of ΔkstDΔchsH2, with a yield of 0.02 g/L (Fig. 6E). The yields of 9-OH-PDCA and 9-OH-PECA were only 0.02 g/L (Fig. 6C and D), and the production of 9-OH-PDCA and 9-OH-PECA was not influenced by the overexpression of hsd4A, chsE1 and chsE2 (Table 2). The production of 9-OH-PDCE improved to 0.78 g/L, which is 30% higher than that of the strain ΔkstDΔchsH2. The molar yield of 9-OH-PDCE reached 86.7%.
TABLE 2.
C22 steroid metabolite yields of ΔkstDΔchsH2, ΔkstDΔchsH2+hsd4A, ΔkstDΔchsH2+chsE1+chsE2, and ΔkstDΔchsH2+hsd4A+chsE1+chsE2 fed with 1 g/L phytosterola
| Strain | Yield (g/L) |
||||
|---|---|---|---|---|---|
| 9-OH-PDCA | 9-OH-PECA | 9-OH-4-HP | 9-OH-PDCE | 9-OH-PECE | |
| ΔkstDΔchsH2 | 0.02 ± 0.01 | 0.02 ± 0.01 | 0.08 ± 0.01 | 0.6 ± 0.03 | 0.18 ± 0.01 |
| ΔkstDΔchsH2+hsd4A | 0.01 ± 0.01 | 0.02 ± 0.01 | 0.04 ± 0.01 | 0.67 ± 0.02 | 0.17 ± 0.01 |
| ΔkstDΔchsH2+chsE1+chsE2 | 0.02 ± 0.01 | 0.01 ± 0.01 | 0.08 ± 0.01 | 0.72 ± 0.02 | 0.03 ± 0.01 |
| ΔkstDΔchsH2+hsd4A+chsE1+chsE2 | 0.02 ± 0.01 | 0.02 ± 0.01 | 0.04 ± 0.01 | 0.78 ± 0.04 | 0.02 ± 0.01 |
9-OH-PDCA, 9-hydroxy-3-oxo-4,17-pregadiene-20-carboxylic acid; 9-OH-PECA, 9-hydroxy-3-oxo-4-pregene-20-carboxylic acid; 9-OH-4-HP, 9,21-dihydroxy-20-methylpregn-4-en-3-one; 9-OH-PDCE, 9-hydroxy-3-oxo-4,17-pregadiene-20-carboxylic acid methyl ester; 9-OH-PECE, 9-hydroxy-3-oxo-4-pregene-20-carboxylic acid methyl ester.
DISCUSSION
The enoyl-CoA hydratase ChsH1-ChsH2 was identified from M. neoaurum DSM 44074 for the first time. This ChsH1-ChsH2 forms a similar α2β2 heterotetramers with that in M. tuberculosis H37Rv. The monomer of ChsH1 and ChsH2 was confirmed to be unable to hydrate the substrate, while the ChsH1-ChsH2 complex is indispensable for enoyl-CoA hydration. Moreover, besides steroid enoyl-CoA, the acid form of 9-OH-PDCA was identified to be the substrate of ChsH1-ChsH2, indicating that ChsH1-ChsH2 has a broader spectrum of this substrate. One study reported that ChsB1, which was reported to be a 3-hydroxyl acyl-CoA dehydrogenase, can also use 17β-hydroxyandrost-4-en-3-one as a substrate, which contains no CoA structure (28). 9-OH-PDCA and 9-OH-PDC-CoA have the same steroid nucleus and similar side chains. However, the C17 side chain of 9-OH-PDCA is smaller than that of 9-OH-PDC-CoA, which implies that the substrate-binding pocket of ChsH could also adopt 9-OH-PDCA as substate.
C22 steroid intermediates were successfully accumulated in M. neoaurum with ChsH deficiency. However, it is worth noting that 3-oxo-4-pregnene-9-hydroxy-20-carboxyl-CoA (9-OH-PEC-CoA) and 3-oxo-4,17-pregnadiene-9-hydroxy-20-carboxyl-CoA (9-OH-PDC-CoA) were not identified in the ChsH-deficiency strain with phytosterol incubation. Instead, the acid and acid ester forms were obtained at C17 of 9-OH-PEC-CoA and 9-OH-PDC-CoA. These steroid esters have been identified as by-products in other wild-type or engineered strains (29–31). It is possible that the degradation of the phytosterol side chain is similar to the β-oxidation of fatty acid (32) and that fatty-CoA could turn into a fatty acid in vivo. Fatty acids could subsequently convert into fatty acid esters (26). However, the metabolic mechanism of these steroid esters’ production in Mycolicibacterium is unclear. It is assumed that the metabolic mechanism of the transformation of 9-OH-PDCA to 9-OH-PDCE (or 9-OH-PECA to 9-OH-PECE) might be the action of a methyl transfer reaction, which are likely catalyzed by carboxyl methyltransferases (CMTs). Similar results have been reported in other mutant strains (30, 31) However, the genes encoding CMTs are still unknown (31).
Moreover, we observed that the accumulation of both 9-OH-PDCA and 9-OH-PECA increased within 72 h and then decreased (Fig. 5C and D). These two compounds took up only about 2.5% of the final steroid intermediates in total. In contrast, their ester forms, i.e., 9-OH-PDCE and 9-OH-PECE, continuously increased to the maximum (Fig. 5E and F). These observations implied that 9-OH-PEC-CoA and 9-OH-PDC-CoA were first transformed into 9-OH-PECA and 9-OH-PDCA, and then 9-OH-PECA and 9-OH-PDCA were subsequently transformed into 9-OH-PECE and 9-OH-PDCE.
The initial molar yield of 9-OH-PDCE was unsatisfactory due to the existence of by-products such as 9-OH-4-HP. Hsd4A was reported to manipulate the metabolic flux of the phytosterol side chain to switch between the C19 steroid pathway and the C22 steroid pathway (3, 4). Overexpression of hsd4A could enhance the production of C19 steroids. Similar results have been reported to demonstrate that the enhancement of Hsd4A could reduce the accumulation of C22 steroids and enhance the production of C19 steroids (33). Our findings demonstrated that the overexpression of hsd4A could inhibit the accumulation of the by-product 9-OH-4-HP and increase the 9-OH-PDCE yield.
The accumulation of another by-product, 9-OH-PECE, was suppressed by the overexpression of ChsE1-ChsE2. Surprisingly, the transformation of 9-OH-PECA to 9-OH-PDCA was not significantly influenced by the enhancement of ChsE1-ChsE2 activity (Table 2). On one hand, PEC-CoA has been verified to be the substrate of ChsE1-ChsE2 in vitro (29). According to our observation, the overexpression of ChsE1-ChsE2 improved the transformation of 9-OH-PECE to 9-OH-PDCE, indicating that ChsE1-CHsE2 could also dehydrogenate 9-OH-PECE (or PECE). Whether ChsE1-ChsE2 can dehydrogenate 9-OH-PECA to 9-OH-PDCA remains to be determined in vitro. On the other hand, given that ChsE1-ChsE2 can catalyze the dehydrogenation of 9-OH-PECA, there is competition between the transformations of 9-OH-PECA to 9-OH-PDCA and 9-OH-PECA to 9-OH-PECE. During phytosterol bioconversion by the ChsH-deficient strains, the yield of 9-OH-PECA was significantly higher than that of 9-OH-PDCA before 120 h (Fig. 6C and D). However, during the whole period of phytosterol bioconversion, the yield of 9-OH-PECE was 3.08% to 30.7% that of of 9-OH-PDCE. These observations indicate that 9-OH-PECA prefers to be transformed into 9-OH-PECE instead of 9-OH-PDCA. Thus, 9-OH-PECE is dehydrogenated by ChsE1-ChsE2, resulting in 9-OH-PDCE. Therefore, the overexpression of chsE1-chsE2 did not influence the ratio of 9-OH-PECA to 9-OH-PDCA.
The purity of 9-OH-PDCE improved to 89.7% in ΔkstDΔchsH2+hsd4A+chsE1+chsE2, compared to 71.7% in ΔkstDΔchsH2. The production of the by-products 9-OH-4-HP and 9-OH-PECE was inhibited by overexpressing the corresponding genes (Table 2). However, small amounts of 9-OH-4-HP and 9-OH-PECE still occurred during ΔkstDΔchsH2+hsd4A+chsE1+chsE2 incubation, suggesting that the enzymatic activities of Hsd4A and ChsE1-ChsE2 were insufficient to fully suppress the production of these by-products. Nevertheless, the strain ΔkstDΔchsH2+hsd4A+chsE1+chsE2 was able to accumulate 9-OH-PDCE with considerable molar yield and purity.
One previous study reported a 9-OH-PDCE-producing strain which was achieved by the loss of the thiolase Ltp2 in Mycolicibacterium (NwIB-IΔltp2) (34). The yield of 9-OH-PDCE was less than 0.6 g/L in NwIB-IΔltp2. However, in our study, ΔkstDΔchsH2 acquired a higher 9-OH-PDCE yield of 0.6 g/L. The purity of 9-OH-PDCE was undesirable in NwIB-IΔltp2-VIII. The ratio of 9-OH-PDCE to 9-OH-PDCA was 7.8:1. However, the ratio of 9-OH-PDCE to 9-OH-PDCA in ΔkstDΔchsH2+hsd4A+chsE1+chsE2 was 39:1. 9-OH-PDCA was rarely accumulated in our work. In the early stage of phytosterol biodegradation by ΔkstDΔchsH2+hsd4A+chsE1+chsE2, 9-OH-PDCA was the main steroid intermediate. However, 9-OH-PDCA was transformed into 9-OH-PDCE subsequently. Here, these observations were characterized for the first time. In addition, we noticed that medium composition significantly influenced the ratio of 9-OH-PDCE to 9-OH-PDCA. When ΔkstDΔchsH2 was cultured in medium which used an organic nitrogen source (such as corn steep powder, yeast extraction or tryptone), the ratio of 9-OH-PDCA increased to 24.0% (Fig. S1). It has been reported that different nitrogen sources could affect pH, cell growth, and intracellular redox levels, resulting in the metabolic flux of phytosterol bioconversion (35). The ratio of AD/4-HP increased from 2.1 to 5.5 in M. neoaurum R10 when the bioconversion was carried out in medium which used NaNO3 as nitrogen source. Thus, the transformation of 9-OH-PDCA to 9-OH-PDCE might be influenced by the composition of the culture medium. In addition, phytosterols were emulsified with hydroxypropyl-β-cyclodextrin (HP-β-CD) to promote phytosterol bioconversion in this study. Interestingly, the ratio of phytosterols and HP-β-CD could also influence the ratio of 9-OH-PDCA in the final products. The ratio of phytosterols and HP-β-CD was 1:1.5 (vol/vol) in this study. However, when the ratio of phytosterols to HP-β-CD increased to 1:3 (vol/vol), the ratio of 9-OH-PDCA increased to 13.6% in the final products (Fig. S1). HP-β-CD was reported to influence bioconversion efficiency and the activity of certain proteins (36). Thus, to explain this phenomenon, a transcriptome study was performed to characterize the differential gene expression when phytosterol bioconversion was carried out with ratios of phytosterols to HP-β-CD of 1:1.5 and 1:3. As shown in Fig. S2, the expression of three putative CMT was influenced due to the loss of chsH2 and increase in HP-β-CD. The knockout of chsH2 induced the upregulation of genes 2403, 3834, and 5201. However, the increase in HP-β-CD inhibited the expression of these three genes, which may explain why less 9-OH-PDCA was transformed into 9-OH-PDCE when the ratio of phytosterols to HP-β-CD was 1:3. The accumulation of 9-OH-PDCA might be also induced by the hydrolysis of 9-OH-PDCE, which can be catalyzed by esterase (37). Gene 3077 was annotated to encode an esterase. As shown in Fig. S2, the expression of gene 3077 was upregulated due to the increase in HP-β-CD, which may lead to more hydrolysis of 9-OH-PDCE to 9-OH-PDCA, resulting in increased 9-OH-PDCA in the final products. In conclusion, the mechanisms of the formation of 9-OH-PDCA and 9-OH-PDCE should be studied further.
In this study, the hydratase ChsH1-ChsH2 was identified for the first time in M. neoaurum DSM 44074. The ChsH1-ChsH2 complex was verified to be indispensable for function. A 9-OH-PDCE-producing strain was constructed by knockout of chsH2 and overexpression of hsd4A, chsE1, and chsE2 in a KSTD-deficient strain of M. neoaurum DSM 44704. The titer of 9-OH-PDCE increased to 0.78 g/L from 0.6 g/L, with a molar yield of 86.7% and a purity of 89.7%. The construction of a 9-OH-PDCE-producing strain provides more possibilities for the development of new steroid drugs.
MATERIALS AND METHODS
Strains, plasmids, primers, and culture conditions.
The strains, plasmids, and primers used in this study are listed in Table 3. Mycolicibacterium neoaurum DSM 44074, obtained from the German Collection of Microorganisms and Cell Cultures GmbH (Deutsche Sammlung von Mikroorganismenund Zellkulturen [DSMZ], Germany), was selected as the original strain. Plasmids were constructed and preserved in Escherichia coli DH5α. E. coli DH5α was cultured in Luria-Bertani (LB) medium consisting of 1% NaCl, 1% tryptone, and 0.5% yeast extract under the conditions of 37°C, 200 rpm, and an initial pH of 7.0. MYD medium was used for mycolicibacteria cell cultivation (4). The bioconversion of phytosterols by Mycobacterium was carried out in fermentation medium consisting of 20 g/L glucose, 12 g/L (NH4)2HPO4, 0.5 g/L MgSO4, 0.5 g/L NaNO3, 3 g/L critic acid, 0.05 g/L ammonium ferric citrate, and 0.2% Tween 80 (vol/vol). The bioconversion was carried out under the conditions of 30°C, 200 rpm, and an initial pH of 7.5.
TABLE 3.
Strains, plasmids, and primers used in this studya
| Primer, plasmid, or strain | Sequence or descriptionb | Source or purpose |
|---|---|---|
| Primers | ||
| kstD1-F | GTTCTACATGACTGAACAGG | kstD1 amplification |
| kstD1-R | TCAGGCCTTTCCAGCGAGAT | |
| kstD2-F | GTGACCGACCAGAAGAACAT | kstD2 amplification |
| kstD2-R | GGCGTGGTGAGCCGCGATAT | |
| kstD3-F | AGTTCGATGTCATCGTCGCC | kstD3 amplification |
| kstD3-R | TCATTCCGCTGAGCTCTGTG | |
| PCR-chsH1-F | AAGACCGGCAACGGGGTGCG | chsH1 and chsH2 amplification |
| PCR-chsH2-R | TGGGTGGCCGGGTGGTCTTG | |
| chsH1-F | GGAATTCAGTCTGACCCTGACCGAC | construction of ChsH1 expression vector |
| chsH1-R | CGGAATTCTCACTTGTCCCCCAGGGT | |
| chsH2-F | GAAGATCTAGCGAACTGCAGGCGGGTATCG | construction of ChsH2 expression vector |
| chsH2-R | GGAATTCTCATGCTTGCGGCTCCCATGC | |
| P38Mu-Hsd4A-F | TAAGAAGGAGATATACATATGAACGACAACCCGATCGACC | construction of P38Mu-Hsd4A |
| P38Mu-Hsd4A-R | GATGAATTCGGATCCTCAAGAGCCCATGAGCTCGG | |
| P38Mu-Hsd4A-F2 | ATTAAGAAGGAGATATACATATGAACGACAACCCGATCGA | construction of P38Mu-Hsd4A-ChsE1-ChsE2 |
| P38Mu-Hsd4A-R2 | CCGAGCTCATGGGCTCTTGACGGAGGAATCACTTCGCACC | |
| P38Mu-ChsE1-F | ATTAAGAAGGAGATATACATATGGACTTCACGCCGAAGCC | construction of P38Mu-ChsE1-ChsE2 |
| P38Mu-ChsE1-R | TGCCGAGGGTGACCCGGTGAGGATCCGAATTCATCGATAA | |
| Plasmids | ||
| PSBY1 | Derived from pMV261, contains FnCpf1 C. glutamicum codon-optimized; KanR | 42 |
| PCR-Hyg | Plasmid for sgRNA production | 43 |
| Pam-kstD1 | PCR-Hyg containing kstD1 spacer | This study |
| Pam-kstD2 | PCR-Hyg containing kstD2 spacer | This study |
| Pam-kstD3 | PCR-Hyg containing kstD3 spacer | This study |
| Pam-chsH2 | PCR-Hyg containing chsH2 spacer | This study |
| pGEX-4T-1 | ChsH1 expression vector with GST-tag | |
| pSZD | ChsH2 expression vector with His6 tag | |
| pMV306 | Mycolicibacteria integrative vector without promoter, KanR | This study |
| P38Mu | pMV306 with Psmyc promoter, KanR | This study |
| P38Mu-Hsd4A | Recombinant P38Mu for expression of Hsd4A | This study |
| P38Mu-ChsE1-ChsE2 | Recombinant P38Mu for expression of ChsE1-ChsE2 | |
| P38Mu-Hsd4A-ChsE1-ChsE2 | Recombinant P38Mu for expression of Hsd4A and ChsE1-ChsE2 | |
| Strains | ||
| M. neoaurum DSM 44704 | Wild-type strain, sterol consumer with no detectable intermediates | |
| ΔkstD | KSTD deficiency strain of M. neoaurum DSM 44704 | This study |
| ΔkstDΔchsH2 | KSTD and ChsH deficiency strain of M. neoaurum DSM 44704 | This study |
| ΔkstDΔchsH2+chsH2 | ChsH-complemented strain of ΔkstDΔchsH2 | This study |
| ΔkstDΔchsH2+hsd4A | hsd4A-overexpressed strain of strain of ΔkstDΔchsH2 | This study |
| ΔkstDΔchsH2+chsE1+chsE2 | chsE1- and chsE2-overexpressed strain of ΔkstDΔchsH2 | This study |
| ΔkstDΔchsH2+hsd4A+chsE1+chsE2 | hsd4A-, chsE1-, and chsE2 overexpressed strain of ΔkstDΔchsH2 | This study |
M. neoaurum, Mycolicibacterium neoaurum; C. glutamicum, Corynebacterium glutamicum; sgRNA, single guide RNA; GST, glutathione S-transferase; Hsd4A, 17-hydroxysteroid dehydrogenase/3,22-dioxo-25,26-bisnorchol-4-ene-24-oyl CoA dehydrogenase; KSTD, 3-ketosteroid-Δ1-dehydrogenase.
The sequence underline indicates restriction sites.
Predicted structures of the ChsH1/ChsH2 complexes.
The structural models of the ChsH1/ChsH2 complexes were predicted with RoseTTAFold using online resources and the protocol described by Dubry et al. (38, 39). The ChsH1 and ChsH2 enzyme sequences were submitted as the query sequences through the RoseTTAFold open access resource (https://robetta.bakerlab.org/) to conduct the RoseTTAFold prediction, and the first of multiple solutions was used as the template for the enzyme structures. The RoseTTAFold prediction utilized a three-track neural network to generate the ChsH1/ChsH2 models based on their protein sequences. The prediction also contained side chain information.
Expression, purification, and assay of hydration activity of ChsH1-ChsH2.
To obtain soluble ChsH1 and ChsH2 enzymes, pGEX-4T-1 and pSZD vectors containing GST-tag and His6 tag, respectively, were used to express the ChsH1 and ChsH2 enzymes in GST-tagged ChsH1 and His6-tagged ChsH2 fusion proteins, respectively (40).
To express ChsH1, chsH1 was amplified and inserted into the EcoRI site of pGEX-4T-1 vector. Similarly, ChsH2 was amplified and inserted into the BglII and EcoRI sites of pGEX-4T-1 and pSZD vector. E. coli BL21(DE3) cells were used for protein expression. The expression and purification of ChsH1 and/or ChsH2 correspond to the methods of Wang et al. (41).
The hydration activity of ChsH1-ChsH2 was assayed in 50 mM Tris-HCl buffer (pH 8.0). The substrate 9-OH-PDCA (45 nM) was dissolved in 2% methanol. The hydration of ChsH1-ChsH2 was initiated by adding partially purified ChsH1-ChsH2 at 30°C. Product formation was quantified using the formula ε263 nm = 6,700 M−1 cm−1, which corresponds to the α- and β-unsaturation of the enoyl-CoA substrate. The hydration was evaluated by analyzing 9-OH-PDCA and 9,17-OH-PDCA concentrations by HPLC using a Waters 2945 UV detector with a C18 reversed-phase ODS analytical column (150 mm × 4.6 mm, 5 μm) under a detection wavelength of 254 nm.
Construction of the engineered M. neoaurum.
Based on previous reports, the deletion of kstD1, kstD2, kstD3 and chsH2 was carried out using a CRISPR system (4). The plasmid p38Mu (pMV306hsp with the Psmyc promoter) was used as a basic skeleton for complementation of chsH2 and overexpression of hsd4A, chsE1, and chsE2. The target genes were amplificated from M. neoaurum DSM 44074 and inserted into the NdeI or HindIII site of p38Mu. The resulting plasmids were transfected into mycolicibacterium-competent cells and the positive colony harboring the target plasmid was verified by PCR and sequencing.
Bioconversion and analysis.
Fermentation medium with added phytosterol was used for bioconversion by original and mutant mycolicibacterium. The phytosterol concentration was 1 g/L. A mycolicibacteria monocolony was incubated in 10 mL MYD medium for 24 h. Subsequently, 3 mL of the cell broth was used as a seed and incubated into 30 mL fermentation medium for 168 h of cultivation. The cell broth was sampled every 24 h with three replicates. Ethyl acetate was mixed into the sample at a ratio of 3:1 (vol/vol) to dissolve the steroid intermediates. Then, the solvent was evaporated, and the residue was redissolved in methanol. The concentration of steroid intermediates was analyzed by HPLC with a ZORBAX Eclipse XDB-C18 column, and the detection wavelength was 254 nm. The mobile phase was consisted of 75% methanol and 25% water.
Data availability.
The genome assembly of Mycolicibacterium neoaurum DSM 44074 has been deposited into the National Center for Biotechnology Information (NCBI) database under the accession number GenBank GCA_000724065.1. The gene accession numbers for chsH1 and chsH2 are OL504963 and OL504964, respectively.
ACKNOWLEDGMENTS
This work is supported by the National Key R&D Program of China (no. 2017YFE0112700) and the Zhang Baoguo Expert Workstation Project of Lijiang City in 2021.
We are grateful to Yu Jiang (CAS Center for Excellence in Molecular Plant Sciences Institute of Plant Physiology and Ecology, Chinese Academy of Sciences, Shanghai, China) and Yichen Sun (Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China) for helping us set up the CRISPR-Cpf1 system.
Footnotes
Supplemental material is available online only.
Contributor Information
Baoguo Zhang, Email: zhangbg@sari.ac.cn.
Pablo Ivan Nikel, Novo Nordisk Foundation Center for Biosustainability.
REFERENCES
- 1.Donova MV, Egorova OV. 2012. Microbial steroid transformations: current state and prospects. Appl Microbiol Biotechnol 94:1423–1447. 10.1007/s00253-012-4078-0. [DOI] [PubMed] [Google Scholar]
- 2.Zhao A, Zhang X, Li Y, Wang Z, Lv Y, Liu J, Alam MA, Xiong W, Xu J. 2021. Mycolicibacterium cell factory for the production of steroid-based drug intermediates. Biotechnol Adv 53:107860. 10.1016/j.biotechadv.2021.107860. [DOI] [PubMed] [Google Scholar]
- 3.Xu L-Q, Liu Y-J, Yao K, Liu H-H, Tao X-Y, Wang F-Q, Wei D-Z. 2016. Unraveling and engineering the production of 23,24-bisnorcholenic steroids in sterol metabolism. Sci Rep 6:21928. 10.1038/srep21928. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Yuan C-Y, Ma Z-G, Zhang J-X, Liu X-C, Du G-L, Sun J-S, Shi J-P, Zhang B-G. 2021. Production of 9,21-dihydroxy-20-methyl-pregna-4-en-3-one from phytosterols in Mycobacterium neoaurum by modifying multiple genes and improving the intracellular environment. Microb Cell Fact 20:229. 10.1186/s12934-021-01717-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Kreit J. 2017. Microbial catabolism of sterols: focus on the enzymes that transform the sterol 3 beta-hydroxy-5-en into 3-keto-4-en. FEMS Microbiology Lett 364:fnx007. 10.1093/femsle/fnx007. [DOI] [PubMed] [Google Scholar]
- 6.Tang R, Ren X, Xia M, Shen Y, Tu L, Luo J, Zhang Q, Wang Y, Ji P, Wang M. 2021. Efficient one-step biocatalytic multienzyme cascade strategy for direct conversion of phytosterol to C-17-hydroxylated steroids. Appl Environ Microbiol 87:e0032121. 10.1128/AEM.00321-21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Liu M, Xiong L-B, Tao X, Liu Q-H, Wang F-Q, Wei D-Z. 2018. Metabolic adaptation of Mycobacterium neoaurum ATCC 25795 in the catabolism of sterols for producing important steroid intermediates. J Agric Food Chem 66:12141–12150. 10.1021/acs.jafc.8b04777. [DOI] [PubMed] [Google Scholar]
- 8.Fernandes P, Cabral JM. 2007. Phytosterols: applications and recovery methods. Bioresour Technol 98:2335–2350. 10.1016/j.biortech.2006.10.006. [DOI] [PubMed] [Google Scholar]
- 9.Yang H, Yan F, Wu D, Huo M, Li J, Cao Y, Jiang Y. 2010. Recovery of phytosterols from waste residue of soybean oil deodorizer distillate. Bioresour Technol 101:1471–1476. 10.1016/j.biortech.2009.09.019. [DOI] [PubMed] [Google Scholar]
- 10.Tang R, Shen Y, Xia M, Tu L, Luo J, Geng Y, Gao T, Zhou H, Zhao Y, Wang M. 2019. A highly efficient step-wise biotransformation strategy for direct conversion of phytosterol to boldenone. Bioresour Technol 283:242–250. 10.1016/j.biortech.2019.03.058. [DOI] [PubMed] [Google Scholar]
- 11.Zhou X, Zhang Y, Shen Y, Zhang X, Zan Z, Xia M, Luo J, Wang M. 2020. Efficient repeated batch production of androstenedione using untreated cane molasses by Mycobacterium neoaurum driven by ATP futile cycle. Bioresour Technol 309:123307. 10.1016/j.biortech.2020.123307. [DOI] [PubMed] [Google Scholar]
- 12.Zhou X, Zhang Y, Shen Y, Zhang X, Xu S, Shang Z, Xia M, Wang M. 2019. Efficient production of androstenedione by repeated batch fermentation in waste cooking oil media through regulating NAD+/NADH ratio and strengthening cell vitality of Mycobacterium neoaurum. Bioresour Technol 279:209–217. 10.1016/j.biortech.2019.01.144. [DOI] [PubMed] [Google Scholar]
- 13.Zhou X, Zhang Y, Shen Y, Zhang X, Zhang Z, Xu S, Luo J, Xia M, Wang M. 2019. Economical production of androstenedione and 9α-hydroxyandrostenedione using untreated cane molasses by recombinant mycobacteria. Bioresour Technol 290:121750. 10.1016/j.biortech.2019.121750. [DOI] [PubMed] [Google Scholar]
- 14.Fernández-Cabezón L, Galán B, García JL. 2018. New insights on steroid biotechnology. Front Microbiol 9:958. 10.3389/fmicb.2018.00958. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Zhang Y, Zhou X, Yao Y, Xu Q, Shi H, Wang K, Feng W, Shen Y. 2021. Coexpression of VHb and MceG genes in Mycobacterium sp. strain LZ2 enhances androstenone production via immobilized repeated batch fermentation. Bioresour Technol 342:125965. 10.1016/j.biortech.2021.125965. [DOI] [PubMed] [Google Scholar]
- 16.Shao M, Zhao Y, Liu Y, Yang T, Xu M, Zhang X, Rao Z. 2019. Intracellular environment improvement of Mycobacterium neoaurum for enhancing androst-1,4-diene-3,17-dione production by manipulating NADH and reactive oxygen species levels. Molecules 24:3841. 10.3390/molecules24213841. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Wei W, Fan S-Y, Wang F-Q, Wei D-Z. 2014. Accumulation of androstadiene-dione by overexpression of heterologous 3-ketosteroid Δ(1)-dehydrogenase in Mycobacterium neoaurum NwIB-01. World J Microbiol Biotechnol 30:1947–1954. 10.1007/s11274-014-1614-3. [DOI] [PubMed] [Google Scholar]
- 18.Yuan J, Chen G, Cheng S, Ge F, Qiong W, [Li W, Li J. 2015. Accumulation of 9α-hydroxy-4-androstene-3,17-dione by co-expressing kshA and kshB encoding component of 3-ketosteroid-9α-hydroxylase in Mycobacterium sp. NRRL B-3805. Sheng Wu Gong Cheng Xue Bao 31:523–533. [PubMed] [Google Scholar]
- 19.Su L, Shen Y, Gao T, Luo J, Wang M. 2017. Improvement of AD biosynthesis response to enhanced oxygen transfer by oxygen vectors in Mycobacterium neoaurum TCCC 11979. Appl Biochem Biotechnol 182:1564–1574. 10.1007/s12010-017-2418-3. [DOI] [PubMed] [Google Scholar]
- 20.Yang M, Guja KE, Thomas ST, Garcia-Diaz M, Sampson NS. 2014. A distinct MaoC-like enoyl-CoA hydratase architecture mediates cholesterol catabolism in Mycobacterium tuberculosis. ACS Chem Biol 9:2632–2645. 10.1021/cb500232h. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Sripalakit P, Wichai U, Saraphanchotiwitthaya A. 2006. Biotransformation of various natural sterols to androstenones by Mycobacterium sp and some steroid-converting, microbial strains. J Mol Catal B-Enzym 41:49–54. 10.1016/j.molcatb.2006.04.007. [DOI] [Google Scholar]
- 22.Toro A, Ambrus G. 1990. Oxidative decarboxylation of 17(20)-dehydro-23,24-dinorcholanoic acids. Tetrahedron Lett 31:3475–3476. 10.1016/S0040-4039(00)97426-4. [DOI] [Google Scholar]
- 23.Giorgi V, Menendez P, Garcia-Carnelli C. 2019. Microbial transformation of cholesterol: reactions and practical aspects—an update. World J Microbiol Biotechnol 35:131. 10.1007/s11274-019-2708-8. [DOI] [PubMed] [Google Scholar]
- 24.Yao K, Xu L-Q, Wang F-Q, Wei D-Z. 2014. Characterization and engineering of 3-ketosteroid-Δ1-dehydrogenase and 3-ketosteroid-9α-hydroxylase in Mycobacterium neoaurum ATCC 25795 to produce 9α-hydroxy-4-androstene-3,17-dione through the catabolism of sterols. Metab Eng 24:181–191. 10.1016/j.ymben.2014.05.005. [DOI] [PubMed] [Google Scholar]
- 25.Wei W, Wang F-Q, Fan S-Y, Wei D-Z. 2010. Inactivation and augmentation of the primary 3-ketosteroid-{delta}1-dehydrogenase in Mycobacterium neoaurum NwIB-01: biotransformation of soybean phytosterols to 4-androstene-3,17-dione or 1,4-androstadiene-3,17-dione. Appl Environ Microbiol 76:4578–4582. 10.1128/AEM.00448-10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Deng X, Chen L, Hei M, Liu T, Feng Y, Yang G-Y. 2020. Structure-guided reshaping of the acyl binding pocket of 'TesA thioesterase enhances octanoic acid production in E. coli. Metab Eng 61:24–32. 10.1016/j.ymben.2020.04.010. [DOI] [PubMed] [Google Scholar]
- 27.Thomas ST, Sampson NS. 2013. Mycobacterium tuberculosis utilizes a unique heterotetrameric structure for dehydrogenation of the cholesterol side chain. Biochemistry 52:2895–2904. 10.1021/bi4002979. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Yuan T, Werman JM, Yin X, Yang M, Garcia-Diaz M, Sampson NS. 2021. Enzymatic β-oxidation of the cholesterol side chain in Mycobacterium tuberculosis bifurcates stereospecifically at hydration of 3-oxo-cholest-4,22-dien-24-oyl-CoA. ACS Infect Dis 7:1739–1751. 10.1021/acsinfecdis.1c00069. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Thomas ST, VanderVen BC, Sherman DR, Russell DG, Sampson NS. 2011. Pathway profiling in Mycobacterium tuberculosis: elucidation of cholesterol-derived catabolite and enzymes that catalyze its metabolism. J Biol Chem 286:43668–43678. 10.1074/jbc.M111.313643. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Peng H, Wang Y, Jiang K, Chen X, Zhang W, Zhang Y, Deng Z, Qu X. 2021. A dual role reductase from phytosterols catabolism enables the efficient production of valuable steroid precursors. Angew Chem Int Ed Engl 60:5414–5420. 10.1002/anie.202015462. [DOI] [PubMed] [Google Scholar]
- 31.Li X, Chen X, Wang Y, Yao P, Zhang R, Feng J, Wu Q, Zhu D, Ma Y. 2018. New product identification in the sterol metabolism by an industrial strain Mycobacterium neoaurum NRRL B-3805. Steroids 132:40–45. 10.1016/j.steroids.2018.02.001. [DOI] [PubMed] [Google Scholar]
- 32.Yang M, Lu R, Guja KE, Wipperman MF, St Clair JR, Bonds AC, Garcia-Diaz M, Sampson NS. 2015. Unraveling cholesterol catabolism in Mycobacterium tuberculosis: ChsE4-ChsE5 α2β2 acyl-CoA dehydrogenase initiates β-oxidation of 3-oxo-cholest-4-en-26-oyl CoA. ACS Infect Dis 1:110–125. 10.1021/id500033m. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Sun H, Yang J, He K, Wang Y-P, Song H. 2021. Enhancing production of 9α-hydroxy-androst-4-ene-3,17-dione (9-OHAD) from phytosterols by metabolic pathway engineering of mycobacteria. Chem Eng Sci 230:116195. 10.1016/j.ces.2020.116195. [DOI] [Google Scholar]
- 34.Liu Y-J, Ji W-t, Song L, Tao X-Y, Zhao M, Gao B, Meng H, Wang F-Q, Wei D-Z. 2022. Transformation of phytosterols into pregnatetraenedione by a combined microbial and chemical process. Green Chem 24:3759–3771. 10.1039/D1GC04819H. [DOI] [Google Scholar]
- 35.Wang X, Chen R, Wu Y, Wang D, Wei D. 2020. Nitrate metabolism decreases the steroidal alcohol byproduct compared with ammonium in biotransformation of phytosterol to androstenedione by Mycobacterium neoaurum. Appl Biochem Biotechnol 190:1553–1560. 10.1007/s12010-019-03175-y. [DOI] [PubMed] [Google Scholar]
- 36.Donova MV, Nikolayeva VM, Dovbnya DV, Gulevskaya SA, Suzina NE. 2007. Methyl-beta-cyclodextrin alters growth, activity and cell envelope features of sterol-transforming mycobacteria. Microbiology (Reading) 153:1981–1992. 10.1099/mic.0.2006/001636-0. [DOI] [PubMed] [Google Scholar]
- 37.Vaquero ME, Barriuso J, Martínez MJ, Prieto A. 2016. Properties, structure, and applications of microbial sterol esterases. Appl Microbiol Biotechnol 100:2047–2061. 10.1007/s00253-015-7258-x. [DOI] [PubMed] [Google Scholar]
- 38.Baek M, DiMaio F, Anishchenko I, Dauparas J, Ovchinnikov S, Lee GR, Wang J, Cong Q, Kinch LN, Schaeffer RD, Millán C, Park H, Adams C, Glassman CR, DeGiovanni A, Pereira JH, Rodrigues AV, van Dijk AA, Ebrecht AC, Opperman DJ, Sagmeister T, Buhlheller C, Pavkov-Keller T, Rathinaswamy MK, Dalwadi U, Yip CK, Burke JE, Garcia KC, Grishin NV, Adams PD, Read RJ, Baker D. 2021. Accurate prediction of protein structures and interactions using a three-track neural network. Science 373:871–876. 10.1126/science.abj8754. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Dubey S, Majumder P, Penmatsa A, Sardesai AA. 2021. Topological analyses of the l-lysine exporter LysO reveal a critical role for a conserved pair of intramembrane solvent-exposed acidic residues. J Biol Chem 297:101168. 10.1016/j.jbc.2021.101168. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Peng F, Cheng X, Wang H, Song S, Chen T, Li X, He Y, Huang Y, Liu S, Yang F, Su Z. 2019. Structure-based reconstruction of a Mycobacterium hypothetical protein into an active Δ5–3-ketosteroid isomerase. Biochim Biophys Acta Proteins Proteom 1867:821–830. 10.1016/j.bbapap.2019.06.008. [DOI] [PubMed] [Google Scholar]
- 41.Wang H, Song S, Peng F, Yang F, Chen T, Li X, Cheng X, He Y, Huang Y, Su Z. 2020. Whole-genome and enzymatic analyses of an androstenedione-producing Mycobacterium strain with residual phytosterol-degrading pathways. Microb Cell Fact 19:187. 10.1186/s12934-020-01442-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Sun B, Yang J, Yang S, Ye RD, Chen D, Jiang Y. 2018. A CRISPR-Cpf1-assisted non-homologous end joining genome editing system of Mycobacterium smegmatis. Biotechnol J 13:e1700588. 10.1002/biot.201700588. [DOI] [PubMed] [Google Scholar]
- 43.Yan MY, Yan HQ, Ren GX, Zhao JP, Guo XP, Sun YC. 2017. CRISPR-Cas12a-Assisted Recombineering in Bacteria. Appl Environ Microbiol 83:e00947-17. 10.1128/AEM.00947-17. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Fig. S1 and S2. Download aem.01303-22-s0001.pdf, PDF file, 0.2 MB (237.2KB, pdf)
Data Availability Statement
The genome assembly of Mycolicibacterium neoaurum DSM 44074 has been deposited into the National Center for Biotechnology Information (NCBI) database under the accession number GenBank GCA_000724065.1. The gene accession numbers for chsH1 and chsH2 are OL504963 and OL504964, respectively.