Efficient 9?-hydroxy-4-androstene-3,17-dione production by engineered Bacillus subtilis co-expressing Mycobacterium neoaurum 3-ketosteroid 9?-hydroxylase and B. subtilis glucose 1-dehydrogenase with NADH regeneration

9?-Hydroxylated steroids are important precursors in the synthesis of steroidal hormone pharmaceuticals, which have been attracted increasing attention (Donova and Egorova 2012; Donova 2007; Fernandes et al. 2003). Highly specific reactions are required to produce functionalized compounds with therapeutic use and commercial value. Due to the high region- and stereo-selectivity of the reactions, together with the mild conditions required, the high yield biological production process, which are more environmentally friendly than their chemical synthesis counterparts, has been developed (Fernandes et al. 2003). Microbial fermentation has been wildly used to accumulate some important steroids intermediates, such as 4-androstene-3,17-dione (AD), androst-1,4-diene-3,17-dione (ADD) and 9?-hydroxy-4-AD (9OHAD) (Zhang et al. 2013; Shao et al. 2015a; Yuan et al. 2015). The 3-ketosteroid 9?-hydroxylase (KSH) and 3-ketosteroid-?¹-dehydrogenase (KSDD) are key enzymes in the process of microbial steroids degradation. KSH catalyzes the 9?-hydroxylation reaction of AD/ADD to 9OHAD/9?-hydroxy-4-ADD (9OHADD), whereas KSDD catalyzes the reaction of ?¹-dehydrogenation of AD/9OHAD to ADD/9OHADD. In this process, however, 9OHADD could subsequently form 3-hydroxy-9,10-secoandrost-1,3,5(10)-triene-9,17-dione (3HSA) by B-ring cleavage, spontaneously (Martin 1977; Kieslich 1985) (Fig. 1). Therefore, KSH combined with KSDD lead to the opening of the B-ring of steroid degradation.

Bioconversion of AD to 9ODAD and their degradation pathway. AD 4-androstene-3,17-dione, ADD androst-1,4-diene-3,17-dione, 9OHAD 9?-hydroxy-4-AD, 9OHADD 9?-hydroxy-4-ADD, KSH 3-ketosteroid 9?-hydroxylase, KSDD 3-ketosteroid-?¹-dehydrogenase, 3HSA 3-hydroxy-9,10-secoandrost-1,3,5(10)-triene-9,17-dione

The activity of KSH has been found in various bacterial genera, such as Mycobacterium (Wovcha et al. 1978; Brzostek et al. 2005; Van der Geize et al. 2007), Nocardia (Strijewski 1982), Rhodococcus (Van der Geize et al. 2002; Petrusma et al. 2009; Datcheva et al. 1989) and Arthrobacter (Dutta et al. 1992). Heterologous expression of ksh and characterization of KSH have been reported, and the conserved sequences analysis demonstrated that KSH is a Rieske monooxygenase. It belongs to class IA monooxygenase, including a terminal oxygenase (KshA) and a ferredoxin reductase (KshB) (Petrusma et al. 2009; Capyk et al. 2009; Arnell et al. 2007). It has been certified that KshA and KshB are essential for KSH activity by gene deletion studies of kshA and kshB (Andor et al. 2006). There were some reports about microbial fermentation from phytosterols to 9OHAD. However, due to low enzyme activities of steroids degradation pathway, they took long fermentation durations (about 120–144 h). For example, it has been reported that the mutant Mycobacterium sp. 2–4 M can be used to produce 9OHAD as a major product from sitosterol, with a 50 % molar yield (Donova et al. 2005). By using the resting Rhodococcus sp. cells to transform AD to 9OHAD, the substrate conversion ratio reached to about 85 % (Angelova et al. 1996). Generally, it is difficult to accumulate 9OHAD using fermentation method despite bacteria that can degrade steroids, because 9OHAD could be ?¹-dehydrogenated to 9OHADD and then spontaneously form 3HSA in these strains. Since the present of KSDD isoenzymes prevented the accumulation of intermediates (Van der Geize et al. 2000), deletion of all ksdd genes and overexpression of kshA resulted in accumulation of 9OHAD in Mycobacterium neoaurum (about 6.78–7.33 g L^?1). However, the fermentation duration was more than 150 h (Yao et al. 2014). Thus, the strains which could accumulate 9OHAD, might be lack of KSDD or deficiency in KSDD (Seidel and Horhold 1992). Hence, double-stage fermentation method was developed to produce 9OHAD. The first step was the side-chain cleavage of sterols to form AD by one strain, and then the second step was 9?-hydroxylation of AD accomplished by another strain (Seidel and Horhold 1992).

Our laboratory has been devoted to using microorganisms to produce steroids intermediates with non-pollution and non-toxic biological technology (Shao et al. 2015a, 2016a). For example, we have co-expressed human 17?-hydroxysteroid dehydrogenase type 3 (17?-HSD3) and Saccharomyces cerevisiae glucose 6-phosphate dehydrogenase (G6PDH) to construct the NADPH regeneration system for efficient testosterone (TS) production form AD (Shao et al. 2016b). The M. neoaurum JC-12 (CCTCC No. M208135), capable of producing AD and ADD from phytosterol or cholesterol by fermentation method, was isolated with phytosterol as the sole carbon source from soil (Zhang et al. 2013). Genes of steroids degradation pathway from M. neoaurum JC-12 had been heterologous over-expressed to construct bioconversion system for steroids intermediates production. For example, cholesterol oxidase gene (choM) had been over-expressed in Bacillus subtilis for bioconversion of cholesterol to 4-cholesten-3-one (Shao et al. 2015b). 3-ketosteroid-?¹-dehydrogenase (ksdd) gene had been over-expressed in B. subtilis for bioconversion of AD to ADD (Zhang et al. 2013). The previous work indicated that B. subtilis might be a preferred host for bioconversion of steroids intermediates as compared with M. neoaurum strains. Hence, in this study, we cloned kshA and kshB gene from M. neoaurum JC-12 and first heterologously co-expressed them in B. subtilis 168. For efficiently bioconversion of AD to 9OHAD, glucose 1-dehydrogenase (GDH, EC 1.1.1.47, encoded by gdh gene) was co-expressed with KSH to construct a NADH regeneration system (Additional file 1: Fig. S1). The intracellular NADH concentration and the whole-cell bioconversion capability of recombinant B. subtilis were detected. This work provided a new reference for 9OHAD production.