Comprehensive summary of steroid metabolism in Comamonas testosteroni TA441: entire degradation process of basic four rings and removal of C12 hydroxyl group

ABSTRACT

Comamonas testosteroni is one of the representative aerobic steroid-degrading bacteria. We previously revealed the mechanism of steroidal A,B,C,D-ring degradation by C. testosteroni TA441. The corresponding genes are located in two clusters at both ends of a mega-cluster of steroid degradation genes. ORF7 and ORF6 are the only two genes in these clusters, whose function has not been determined. Here, we characterized ORF7 as encoding the dehydrase responsible for converting the C12β hydroxyl group to the C10(12) double bond on the C-ring (SteC), and ORF6 as encoding the hydrogenase responsible for converting the C10(12) double bond to a single bond (SteD). SteA and SteB, encoded just upstream of SteC and SteD, are in charge of oxidizing the C12α hydroxyl group to a ketone group and of reducing the latter to the C12β hydroxyl group, respectively. Therefore, the C12α hydroxyl group in steroids is removed with SteABCD via the C12 ketone and C12β hydroxyl groups. Given the functional characterization of ORF6 and ORF7, we disclose the entire pathway of steroidal A,B,C,D-ring breakdown by C. testosteroni TA441.

IMPORTANCE

Studies on bacterial steroid degradation were initiated more than 50 years ago, primarily to obtain materials for steroid drugs. Now, their implications for the environment and humans, especially in relation to the infection and the brain-gut-microbiota axis, are attracting increasing attention. Comamonas testosteroni TA441 is the leading model of bacterial aerobic steroid degradation with the ability to break down cholic acid, the main component of bile acids. Bile acids are known for their variety of physiological activities according to their substituent group(s). In this study, we identified and functionally characterized the genes for the removal of C12 hydroxyl groups and provided a comprehensive summary of the entire A,B,C,D-ring degradation pathway by C. testosteroni TA441 as the representable bacterial aerobic degradation process of the steroid core structure.

INTRODUCTION

Several actinobacteria and proteobacteria are known for their ability to degrade steroid compounds. Especially the underlying mechanism of Rhodococcus equi and Comamonas (formerly Pseudomonas) testosteroni has been studied extensively to obtain materials for the synthesis of steroidal drugs in the 1960s (1–5). These studies led to the identification of the main intermediates in the A- and B-ring degradation processes.

Nowadays, bacterial steroid degradation is reported from more genera of bacteria in both Actinobacteria and Proteobacteria, such as Pseudomonas, Sphingobium, Azoarcus, and Mycobacterium (6, 7), and steroid compounds and the bacteria that metabolize them are attracting attention for their impact on human health, especially in relation to the infection and the brain-gut-microbiota axis (8, 9). The mce4 operon, which encodes a cholesterol import system, is essential for the persistence of Mycobacterium tuberculosis in the lungs of chronically infected animals and for growth within interferon-gamma-activated macrophages (10). Cholesterol catabolism and its utilization by M. tuberculosis are important for the pathogen maintenance in the host (7). Primary bile acids, synthesized from cholesterol and secreted into the intestine, are converted by gut bacteria into secondary bile acids and other cholic acid derivatives. Bile acids are crucial for the digestion and absorption of dietary fats. They aid in emulsifying fats into smaller droplets, thereby increasing the surface area available for enzymes to break them down. Bile salts play essential roles in facilitating the digestion, transportation, and metabolism of nutrients within the gastrointestinal system. They act as signaling hormones for nutrients by activating specific nuclear receptors such as FXR, PXR, and Vitamin D, as well as G-protein-coupled receptors including TGR5, sphingosine-1 phosphate receptor 2 (S1PR2), and muscarinic receptors (11). Recent studies have demonstrated that bile acids influence the composition and function of the gut microbiota, the complex community of microorganisms residing in the gastrointestinal tract (9, 12, 13). Bile acids are also emerging as key signaling molecules within the brain-gut-microbiota axis, as direct reciprocal crosstalk between gut microbiota and bile acids has been reported (14). Bile acids also affect Alzheimer’s disease pathology; the disturbances in bile acid homeostasis in experimental and clinical samples of Alzheimer’s disease supported the association between bile acids and Alzheimer’s disease pathology as well as the significance of the brain-gut-microbiota axis (12). Comamonas testosteroni is an environmental bacterium unable to degrade cholesterol, but capable of utilizing cholic acid as a carbon and energy source, whereas some Comamonas species are emerging as important opportunistic pathogens. Cases of infections caused by Comamonas sp. are summarized in the review titled “The Emergence of the Genus Comamonas as Important Opportunistic Pathogens” by M.P. Ryan et al., published in 2022 (15). C. testosteroni has been identified in several cases of appendicitis (15), suggesting the possibility that these bacteria are present in the small intestine or around the junction of the large and small intestines and affect the intestinal microbiota because the cecum is located at the uppermost part of the large intestine.

Genetic studies on aerobic steroid degradation by C. testosteroni, now the leading bacterial model of this process, started around the year 1990. Thereafter, the enzymes catalyzing the early steps of steroidal degradation, 17β-dehydrogenase (16–19), 3α-dehydrogenase (20–24), 3-oxo-Δ5-steroid isomerase (25, 26), Δ1-dehydrogenase (27), Δ4-dehydrogenase (28), and the positive regulator (29), were identified. However, the degradation of steroidal A,B,C,D-rings remained unclear until we revealed the steroid degradation process in C. testosteroni TA441. TA441 degrades steroids through aromatization of the A-ring, along with cleavage of the B-ring, and subsequent cleavage and degradation of the aromatized A-ring. A similar process has been reported for other bacteria in both Actinobacteria and Proteobacteria (6, 7). Degradation of the C,D-ring and cleaved B-ring occurs via β-oxidation (Fig. 1; all the compound names and the numbers are summarized in Table S1 of supplemental materials) and a similar process has been reported for M. tuberculosis (30). According to the paper on the degradation of the C,D-ring and cleaved B-ring in M. tuberculosis (30), C. testosteroni CNB-2 also has a similar process. However, to the best of our knowledge, steroid degradation has not been reported for CNB-2. The genomic analysis of CNB-2 showed that it contains putative steroid-degrading genes that are almost identical to those of TA441 (31). Genes responsible for aromatic ring degradation and those mediating β-oxidation, respectively, form a cluster, which are located at the two ends of the 120 kb mega-cluster of steroid-degrading genes of TA441 (Fig. 1). The function of most genes in these two clusters has been elucidated, except for ORF6, ORF7, ORF25, and ORF26. The latter two are not required for steroid degradation (unpublished data: The gene-disrupted mutant of each gene exhibited robust growth and minimal accumulation of intermediate compounds, which were identical to those observed in TA441, when incubated with steroids), leaving only ORF6 and ORF7 to be characterized. Here, we describe the role of ORF6 and ORF7 in steroid degradation and provide a comprehensive summary of the processes guiding the degradation of steroidal A,B,C,D-rings in C. testosteroni TA441.

Fig 1.

Abbreviated steroid degradation pathway of Comamonas testosteroni TA441 indicating reactions and compounds described in this study. The mega-cluster of steroid degradation genes in C. testosteroni TA441 is shown below the pathway; the aromatic ring-degradation gene cluster (tesG to scdA) and the β-oxidation gene cluster (tesB to tesR) locate both ends of this 120kb-mega cluster. ORF7 and ORF6, indicated with an arrow and in the DNA region upstream of the β-oxidation gene cluster, are left unidentified. Compounds are as follows: 3α,7α,12α-trihydroxy-androstan-17-one (I); 3α,7α-dihydroxy-androstane-12,17-dione (II); 3α,7α,12β-trihydroxy-androstan-17-one (III); 7α,12β-dihydroxy-9,17-dioxo-1,2,3,4,10,19-hexanorandrostan-5-oic acid (IV); 9α-hydroxy-17-oxo-1,2,3,4,10,19-hexanorandrostan-5-oic acid (R1, R2 = H) (V); 9α-hydroxy-17-oxo-1,2,3,4,5,6,10,19-octanorandrostan-7-oic acid (R1 = H) (VI); 14-hydroxy-9-oxo-1,2,3,4,5,6,10,19-octanor-13,17-secoandrostane-7,17-dioic acid (R1 = H) (VII); 6-methyl-3,7-dioxo-decane-1,10-dioic acid-CoA ester (R1 = H) (VIII); 3-hydroxy-6-methyl-7-oxo-decane-1,10-dioic acid (R1 = H) (IX); 4-methyl-5-oxo-oct-3-ene-1,8-dioic acid (X), and 4-methyl-5-oxo-octane-1,8-dioic acid (XI). All the compound names and numbers are summarized in Table S1 of supplemental materials. 3α-DH: 3α-Hydroxy-dehydrogenase gene; ksi: 3-ketosteroid Δ4–5 isomerase gene.

RESULTS AND DISCUSSION

Identification of compounds in the culture of an ORF6-disrupted mutant incubated with cholic acid

C. testosteroni TA441 degrades steroidal A,B,C,D-rings with the pathway similar to aromatic compound degradation for A,B-ring cleavage followed by C,D, and cleaved B-ring degradation mainly by β-oxidation (cf. Fig. 5). The degradation genes are encoded in two clusters on both ends of the 120kb-mega cluster of steroid degradation genes: one consists of those for aromatic compound degradation and the other for β-oxidation (Fig. 1). ORF6 and ORF7 are the only two genes in these clusters whose function remains unknown. They are in the DNA region just upstream of the β-oxidation gene cluster, and downstream of the genes encoding SteA and SteB, which catalyze the conversion of the C12α hydroxyl to the C12β hydroxyl via a ketone group (Fig. 1) (32). Previously, steAB, ORF7, and ORF6 were shown to be involved in cholic acid degradation while their disruption had no influence on the growth of testosterone (32). ORF7-encoded enzyme and ORF6-encoded enzyme show 81% and 56% similarities to a 12-hydroxy steroid dehydratase Hsh1 and a steroid oxidoreductase Sor1 in Pseudomonas sp. strain Chol1, respectively (33). Other than these, a homology search indicated that the ORF7-encoded enzyme belonged to the nuclear transport factor 2 family, which includes Δ5–3-ketosteroid isomerases and LinA. LinA is a dehalogenase with hydrolase activity, which converts γ-hexachlorocyclohexane to γ-pentachlorocyclohexene (34). The ORF6-encoded enzyme was revealed to be an old yellow enzyme-like alkene reductase whose substrate was not clear. Here, gene-disrupted mutants of ORF7 and ORF6, denoted, respectively, as ORF7^- and ORF6^-, were incubated with cholic acid and its analogs, chenodeoxycholic acid, deoxycholic acid, and lithocholic acid, for 7 days and the cultures were analyzed by high-performance liquid chromatography (HPLC) with UV detection. Cholic acid harbors hydroxyl groups at positions C7 and C12. Chenodeoxycholic acid contains a hydroxyl group at position C7 and deoxycholic acid at position C12. Lithocholic acid does not have either of them (Fig. 2). Candidate intermediate compounds were detected only in the ORF6^- culture incubated with cholic acid (Fig. 2). The three compounds, denoted as 6a, 6b, and 6c (Fig. 2), were isolated and identified by fast atom bombardment mass spectrometry (FAB-MS) as having mass/charge ratios (m/z) 219, 219, and 217 (M+H⁺), respectively. Accordingly, their respective molecular formulae were deduced to be C₁₃H₁₄O₃, C₁₃H₁₄O₃, and C₁₃H₁₂O₃. Based on nuclear magnetic resonance (NMR) analysis, 6a, 6b, and 6c were identified as the lactone forms of 17-hydroxy-9-oxo-1,2,3,4,10,19-hexanorandrost-6,10 (12)-dien-5-oic acid (XII), 9,17-dioxo-1,2,3,4,10,19-hexanorandrost-6-en-5-oic acid (XIII), and 9,17-dioxo-1,2,3,4,10,19-hexanorandrost-6,10 (12)-dien-5-oic acid (XIV), respectively (all the compound names and the numbers are summarized in Table S1 of supplemental materials). The corresponding structures are presented in Fig. 2 and NMR data are listed in Table 1. These compounds had a ketone moiety at position C9 and, therefore, they corresponded to the intermediates found before β-oxidation of the cleaved B-ring (cf. Fig. 1, IV and V). No characteristic accumulations were detected in the culture medium of the ORF7^- in UV absorption.

Fig 2.

HPLC analysis of the culture of the ORF7^- and ORF6^- mutants incubated with cholic acid, deoxycholic acid, chenodeoxycholic acid, and lithocholic acid, and the compounds identified from the culture of ORF6^- mutant with cholic acid (6a–6c). These compounds (6a–6c) were lactones of 17-hydroxy-9-oxo-1,2,3,4,10,19-hexanorandrost-6,10-dien-5-oic acid (XII), 9,17-dioxo-1,2,3,4,10,19-hexanorandrost-6-en-5-oic acid (XIII), and 9,17-dioxo-1,2,3,4,10,19-hexanorandrost-6,10-dien-5-oic acid (XIV), respectively. In HPLC, the vertical axis indicates wavelength (nm), the horizontal axis indicates RT (min), and the UV absorbance of each compound is represented in contour.

TABLE 1.

NMR data of compounds accumulated by ORF6-disrupted mutant incubated with cholic acid^a

	Compound 6a (RT = 6.7 min) (in CDCl3)		Compound 6b (RT = 7.6 min) (in CDCl3)		Compound 6c (RT = 8.7 min) (in CDCl3)
No	13C-NMR [δ (ppm)]	1H-NMR [δ (ppm)], J (Hz)]	13C-NMR [δ (ppm)]	1H-NMR [δ (ppm)], J (Hz)]	13C-NMR [δ (ppm)]	1H-NMR [δ (ppm)], J (Hz)]
5	164.60		164.81		164.30
6	113.44	6.17 d (9.2)	113.66	6.20 d (9.2)	114.30	6.26 d (4.1)
7	145.13	7.41 d (9.2)	145.35	7.45 d (9.2)	144.46	7.56 d (9.2)
8	116.12		115.20		115.25
9	158.39		161.94		158.45
11	121.60	6.14 d (9.8)	26.40	2.75 m	122.69	6.24 d (4.6)
12	147.05	6.67 d (9.8)	28.18	1.72 ddd (13.3, 9.4, 9.2)	141.07	6.66 d (10.1)
13	47.92		49.00		51.10
14	43.48	2.83 (7.5, 5.1)	44.00	2.89	44.03	3.14 (13.1, 6.2)
15	21.07	1.94 m	21.51	1.85 m	19.43	2.06 m
		1.80 dddd (12.0, 12.0, 12.0, 6.0)		2.34 m		2.30 m
16	31.71	2.30 m	37.14	2.32 m	37.41	2.46 ddd (19.0, 9.5, 9.5)
		1.71 m		2.65 m		2.68 dd (16.7, 8.7)
17	75.68	4.09 dd (9.2, 7.8)	219.43		214.12
18	10.55	0.78 s	13.16	0.87 s	13.22	0.89 s

^a Abbreviations for NMR signals are as follows: s, singlet; d, doublet; m, multiplet.

Analysis of the culture of ORF7^- and ORF6^- mutants incubated with cholic acid and its analogs

Because HPLC could not detect compounds with weak UV absorption, ORF7^- and ORF6^- mutants were incubated individually with cholic acid, deoxycholic acid, chenodeoxycholic acid, and lithocholic acid for 7 days, and the respective cultures were then analyzed by reverse-phase liquid chromatography with tandem mass spectrometry (LC-MS/MS). Compounds with m/z 253 at retention time (RT) = 4.5 min and another with m/z 255 at RT = 4.3 min were detected in large amounts in the ORF7^- culture with cholic acid or deoxycholic acid (Fig. 3A1,2 and C1,2). Somewhat lower but substantial amounts of the same compound were detected in the ORF6^- culture (Fig. 3B1,2 and D1,2). These compounds were not detected in cultures incubated with chenodeoxycholic acid or lithocholic acid, which do not have a hydroxy group at position C12 (Fig. 3A3,4, B3,4, C3,4, and D3,4). Mass chromatograms of m/z 269, 271, 237, 235, 211, 227, 181, 197, 241, 259, 245, 243, and 201 of the ORF6^- culture with cholic acid (Fig. S1-1 in the supplemental material), ORF7^- culture with cholic acid (Fig. S1-2), ORF7^- culture with deoxycholic acid (Fig. S1-3), and ORF7^- culture with chenodeoxycholic acid (Fig. S1-4) with the mass spectra (Fig. S2 in the supplemental material) are presented in the supplemental material. Based on the peak with m/z 253 at RT = 4.5 min and its major fragments with m/z 253, 235, 191, 163, and 149 (Fig. S2A1), as well as the peak with m/z 255 at RT = 4.3 min and its major fragments with m/z 255, 237, and 193 (Fig. S2B1), the two peaks were assigned to 12β-hydroxy-9,17-dioxo-1,2,3,4,10,19-hexanorandrostan-5-oic acid (XV) and 12β,17-dihydroxy-9-oxo-1,2,3,4,10,19-hexanorandrostan-5-oic acid (XVI), respectively. Other characteristic peaks were observed in the mass chromatograms of m/z 235 (Fig. 3E1, 2), m/z 227 (Fig. 3E3, 4), m/z 241 (Fig. 3E5,6), and m/z 259 (Fig. 3E7,8). 9α-Hydroxy-17-oxo-1,2,3,4,10,19-hexanorandrosta-6,10 (12)-dien-5-oic acid (XVII), a compound with a C10(12) double bond, was detected as peaks with m/z 235 at RT = 6.55 and 6.85 min in the ORF6^- culture (Fig. 3E1), while it was absent in the ORF7^- culture (Fig. 3E2; Fig. S1F3,5 in the supplemental material). By contrast, the peak with m/z 227 at RT = 2.3 min showed accumulation in the ORF7^- culture (Fig. 3E3) but lower levels in the ORF6^- culture (Fig. 3E4). This compound presented major fragments with m/z 227, 209, 183, and 155 and was identified as 9α,12β-dihydroxy-17-oxo-1,2,3,4,5,6,10,19-octanorandrostan-7-oic acid (XVIII) (Fig. S2G1 in the supplemental material). All C-ring-containing intermediates detected in ORF7^- cultures incubated with cholic acid or deoxycholic acid harbored a hydroxyl group at C12 (Fig. 3; Fig. S1-2,3 in the supplemental material). Based on these findings and homology search results, the ORF7-encoded enzyme was presumed to be the dehydrase for the C12β hydroxyl group. ScdA is a coenzyme A (CoA)-transferase, which adds CoA to the C5 carboxylic group, thereby initiating β-oxidation. The main compound found to accumulate in a ScdA^- culture with cholic acid was 7α,12β-dihydroxy-9,17-dioxo-1,2,3,4,10,19-hexanorandrostan-5-oic acid (35). At the same time, none of the intermediates during β-oxidation of the cleaved B-ring identified in the previous studies contained the C12β hydroxyl group, indicating that the hydroxyl group is removed prior to β-oxidation. Present data indicate that C,D-ring cleavage proceeds less efficiently in the presence than in the absence of the C12β hydroxyl group.

Fig 3.

LC/MS/MS analysis of the culture of the ORF7^- and ORF6^- mutants incubated with cholic acid (CA), deoxycholic acid (DC), chenodeoxycholic acid (CDC), and lithocholic acid (LC). Chromatograms of m/z 253 (3A1-4, 3B1-4), m/z 255 (3C1-4, 3D1-4), m/z 235 (3E1, 3E2), m/z 227 (3E3, 3E4), m/z 241 (3E5, 3E6), and m/z 259 (3E7, 3E8) are shown. Two peaks with m/z of 241 at around RT = 5.7 min in ORF7^- culture (3E5) showed almost identical mass spectrums (Fig. S2J4 in the supplemental material). Closed arrows indicate the peak of the compound shown on the left side of the columns. For XIX and XXI, the carbons are given new numbers according to IUPAC numbering. Numbers in parentheses on XIX are based on the steroidal numbering. Compounds are as follows: 12β-hydroxy-9,17-dioxo-1,2,3,4,10,19-hexanorandrostan-5-oic acid (XV), 12β,17-dihydroxy-9-oxo-1,2,3,4,10,19-hexanorandrostan-5-oic acid (XVI), 9α-hydroxy-17-oxo-1,2,3,4,10,19-hexanorandrosta-6,10 (12)-dien-5-oic acid (XVII), 9α,12β-dihydroxy-17-oxo-1,2,3,4,5,6,10,19-octanorandrostan-7-oic acid (XVIII), 6-methyl-3,7-dioxo-dec-5-ene-1,10-dioic acid (XIX), and 5-hydroxy-6-methyl-3,7-dioxo-decane-1,10-dioic acid (XXI). The vertical axis indicates intensity (count/sec) and the horizontal axis indicates RT (min). Mass chromatograms of m/z 269, 271, 237, 235, 211, 227, 181, 197, 241, 259, 245, 243, and 201 are presented in Fig. S1-1 to S1-4. Fig. S1-1 shows ORF6^- culture with cholic acid. Fig. S1-2-3, and -4. show ORF7^- culture with cholic acid, deoxycholic acid, and chenodeoxycholic acid, respectively.

The mass spectrum of the large peak with m/z 241 detected at RT = 5.72 min (Fig. 3E5 and E6) exhibited an identical profile to that of 6-methyl-3,7-dioxo-dec-5-ene-1,10-dioic acid (XIX) in the accompanying paper. Note that we used IUPAC numbering for compounds without steroidal rings and steroidal numbering for those with at least one steroidal ring. This compound was initially identified as an intermediate produced after C-ring cleavage by ScdL1L2 in cultures incubated with steroids harboring a hydroxyl group at C12 (accompanying paper). XIX and 3-hydroxy-6-methyl-7-oxo-dec-5-ene-1,10-dioic acid (XX) (Fig. S1-1N,2N,3N in the supplemental material), whose main fragments possessed m/z values of 243, 199, 155, 153, 111, and 109 (Fig. S2M in the supplemental material), were the only intermediate compounds with a double bond at C5 (C12 of the C-ring), and which accumulated more in the ORF7^- than in the ORF6^- culture (Fig. 3E5,6; Fig. S1-2N,1N in the supplemental material). Therefore, the compounds originally present in the culture were hypothesized as 5-hydroxy-6-methyl-3,7-dioxo-decane-1,10-dioic acid (XXI) (Fig. 3E7) and 3,5-dihydroxy-6-methyl-7-oxo-decane-1,10-dioic acid (XXII) (Fig. S1-1N,2N,3N in the supplemental material), respectively. They were thought to become dehydrated during extraction and analysis in acidic conditions.

In ORF6^- cultures incubated with cholic acid and deoxycholic acid, compounds with a double bond at C10(12) and a single bond at C10(12) accompanied those with a C12 hydroxyl group (Fig. 3; Fig. S1 in the supplemental material). Intermediates harboring the C12 hydroxyl group were less abundant in ORF6^- cultures than in ORF7^- cultures, suggesting that the reaction catalyzed by the ORF6-encoded enzyme occurred after dehydration by the ORF7-encoded enzyme. Compounds with a double bond at C10(12) in the C-ring, including XII, XIV, XVII, and 9,17-dioxo-1,2,3,4,10,19-hexanorandrost-10 (12)-en-5-oic acid (XXIII), were detected only in the ORF6^- culture (Fig. 3E2; Fig. S1-1E in the supplemental material). Two peaks with m/z 235 at RT = 6.55 and 6.85 min showed almost identical mass spectra, characterized by major fragments with m/z 235, 217, 191, 173, 163, 135, 123, and 95 (Fig. S2E3,5 in the supplemental material) and were thought to be stereoisomers. The above data, together with homology search results, pointed to the ORF6-encoded enzyme being a reductase, which converted a double bond to a single bond at C10(12) in the C-ring. The presence of compounds with a single bond at C10(12) in ORF6^- cultures implied the existence of at least another reductase capable of acting on the double bond at C10(12), probably as a side reaction. A peak with m/z 227 at RT = 3.3 min was detected exclusively in the ORF6^- culture incubated with cholic acid or deoxycholic acid (Fig. 3E4), and was thought to be one of the key intermediates that could confirm the function of the ORF6-encoded enzyme. Unfortunately, we were unable to deduce the compound’s structure from the mass spectrum, which was characterized by a large fragment with m/z 227 and minor fragments with m/z 209, 183, 165, and 155 (Fig. S2G2 in the supplemental material).

Complementation of enzymes encoded by ORF7 and ORF6

To further investigate the function of ORF7- and ORF6-encoded enzymes, different combinations of mutant strains and plasmids were generated. These included ORF7^- carrying the broad-host-range plasmid pMFY42 (negative control), ORF7^- carrying a pMFY42-based plasmid encoding ORF7 (pMFYORF7), ORF6^- carrying pMFY42 (negative control), ORF6^- carrying a pMFY42-based plasmid encoding ORF6 (pMFYORF6), double ORF7^-ORF6^- mutant (ORF7^-6^-) carrying pMFY42 (negative control), ORF7^-6^- carrying pMFYORF7, and ORF7^-6^- carrying a pMFY42-based plasmid encoding ORF7 and ORF6 (pMFYORF76) (Tables 2–4). The mutants were cultivated with cholic acid for 7 days and the cultures were analyzed by LC-MS/MS. The mass chromatograms of m/z 253, 235, and 237 are shown in Fig. 4 and the mass chromatograms of m/z 227, 241, 197, and 201 are shown in Fig. S3 in the supplemental material. XV was less abundant in all complemented mutants, ORF7^- carrying pMFYORF7 (Fig. 4-1B), ORF6^- carrying pMFYORF6 (Fig. 4-1D), and ORF7^-6^- carrying pMFYORF76 (Fig. 4-1H). A similar, albeit less pronounced drop was observed for XV in ORF7^-6^- carrying pMFYORF7 (Fig. 4-1G). XVII and XXIII, both of which have a double bond at C10(12), were detected in the culture of ORF7^- carrying pMFYORF7 (Fig. 4-2B), ORF6^- carrying pMFY42 (Fig. 4-2C), ORF6^- carrying pMFYORF6 (Fig. 4-2D), and ORF7^-6^- carrying pMFYORF7 (Fig. 4-2G). Slightly lower levels of XVII and XXIII were detected in the culture of ORF6^- carrying pMFYORF6 compared to the one with ORF6^- carrying pMFY42. XIII, a compound with a single bond at C10(12), was detected in the cultures of the three complemented mutants (Fig. 4-2B, D, and H). The data for the ORF7^- culture incubated with chenodeoxycholic acid, which does not have a C12 hydroxyl group, are presented in Fig. 4-2E as an authentic for XIII. These results confirmed the ability of the ORF7-encoded enzyme to dehydrate the C12 hydroxyl group of XV, thereby generating XVII and XXIII. 9,17-Dioxo-1,2,3,4,10,19-hexanorandrostan-5-oic acid, (3aα-H-4α [3′-propionic acid]−7aβ-methylhexahydro-1,5-indanedione) (R1, R2 = H) (XXIV), which has a single bond at C10(12), was detected in the culture of ORF7^- carrying pMFYORF7 (Fig. 4-3B), ORF6^- carrying pMFY42 (Fig. 4-3C), ORF6^- carrying pMFYORF6 (Fig. 4-3D), and ORF7^-6^- carrying pMFYORF7 (Fig. 4-3G). The corresponding mass spectra are presented in Fig. S2D3 in the supplemental material. It is difficult to distinguish XXIV from 9α-hydroxy-17-oxo-1,2,3,4,10,19-hexanorandrost-6-en-5-oic acid (R1 = H) (XXV) based on the mass chromatogram, but both can be regarded as XV derivatives, with a single bond at C10(12) and without a hydroxyl group at C12. Less XV was detected in the culture of ORF7^-6^- carrying pMFYORF7 (Fig. 4-1G) than in the corresponding control carrying pMFY42 (Fig. 4-1F), although this reduction was less pronounced than the one observed with ORF7^- carrying pMFYORF7 (Fig. 4-1B). XV dropped to nearly undetectable levels in the culture of ORF7^-6^- carrying pMFYORF76 (Fig. 4-1H). These results implied that the C12β hydroxyl group was removed efficiently when both ORF7- and ORF6-encoded enzymes were expressed together.

TABLE 2.

Strains^a

Strains	Characteristics	Source or reference
TA441	Wild type	(36)b
ORF6-	ORF6: :Kmr mutant (BglII site) of TA441	(32)c
ORF7-	ORF7: :Kmr mutant (ApaI site) of TA441	(32)c
ORF7-6-	ORF7,6: :Kmr mutant (ApaI site to BglII site) of TA441	This work

^a Kmr: Km-resistance.

^b Arai et al. 1998. Microbiology 144: 2895–2903.

^c Horinouchi et al. 2008. J Bacteriol 190:5545–5554.

TABLE 3.

Plasmids^a

Plasmids	Characteristics	Source or reference
pUC19	Apr, lacZ	(54)b
pMFY42	Tcr, Kmr, RSF1010-based broad host range plasmid	(34)c
pMFYORF6	pMFY42 derivative carrying ORF6	This work
pMFYORF7	pMFY42 derivative carrying ORF7	This work
pMFYORF76	pMFY42 derivative carrying ORF7,6	This work
pMFYMhpRA	pUC19 derivative carrying mhpR and the promoter of mhp genes	[*]d
pMFYMhpORF76	pMFYMhpRA derivative carrying ORF7,6	This work

^a Kmr: Km-resistance.

^b Vieira, J., and Messing, J. 1987. Methods Enzymol. 153: 3-11.

^c Nagata Y. et al. 1993. J Bacteriol 175:6403-6410.

^d Horinouchi M, et al. (accompanying paper).

TABLE 4.

Primers

Primers	Sequences	Source or reference
Dra_ORF6	TTTAAATGAGTCAAGCGCTTTTCAC	This work
Dra_ORF6_T	TTTAAATCAGACGATCTGCACTTCTT	This work
Dra_ORF7	TTTAAATGAGTGAACCTGTGAATCA	This work
Dra_ORF7_T	TTTAAATTAGCGCACACGCATGCGCG	This work
ORF6_Km	AAGTGCAGATCGTCCCCGGGGTGGGCGAA	This work
ORF6_Km_R	TTCGCCCACCCCGGGGACGATCTGCACTT	This work
MhpRAPvuII_ORF7	GAGAATCTGGCCCAGATGAGTGAACCTGTG	This work
MFYPvuII_Kmr_T	GACAACGTCGAGCAGGTGGGCGAAGAACTC	[*]a

^a Horinouchi M, et al. (accompanying paper).

Fig 4.

Complementation experiments with ORF7^-, ORF6^-, and ORF7^-6^- mutants. The mass chromatograms of each mutant culture incubated with 0.1% cholic acid for 7 days are shown. Mutants are as follows: ORF7^- mutant carrying pMFY42 (ORF7^- with pMFY42) (negative control) (A), carrying pMFYORF7 (pMFY42 carrying ORF7) (B), ORF6^- mutant carrying pMFY42 (ORF6^- with pMFY42) (negative control) (C), carrying pMFYORF6 (ORF6^- with pMFYORF6) (D), ORF7^- mutant incubated with chenodeoxycholic acid (as an authentic for XIII) (E), ORF7^-6^- mutant carrying pMFY42 (ORF7^-6^- with pMFY42) (negative control) (F), carrying pMFYORF7 (G), and carrying pMFYORF76 (H). Mass chromatograms of m/z 253 (4-1), m/z 235 (4-2), and m/z 237 (4-3) are shown. Mass chromatograms of m/z 227, m/z 241, m/z 197, and m/z 201 are shown in Fig. S3 in the supplemental material. Compounds are as follows: 12β-hydroxy-9,17-dioxo-1,2,3,4,10,19-hexanorandrostan-5-oic acid (XV), 9α-hydroxy-17-oxo-1,2,3,4,10,19-hexanorandrosta-6,10 (12)-dien-5-oic acid (XVII), 9α,17-dioxo-1,2,3,4,10,19-hexanorandrost-10 (12)-en-5-oic acid (XXIII), 9α,17-dioxo-1,2,3,4,10,19-hexanorandrost-6-en-5-oic acid (XIII), 9,17-dioxo-1,2,3,4,10,19-hexanorandrostan-5-oic acid (XXIV) (R1, R2 = H), 9α-hydroxy-17-oxo-1,2,3,4,10,19-hexanorandrost-6-en-5-oic acid (XXV), and 17-hydroxy-9-oxo-1,2,3,4,10,19-hexanorandrost-10 (12)-en-5-oic acid (XXVI). The vertical axis indicates intensity (count/sec), and the horizontal axis indicates RT (min).

Expression of ORF7- and ORF6-encoded enzymes in the ORF7^-6^- mutant

To further confirm the function of ORF7- and ORF6-encoded enzymes, the pMFYMhpR plasmid was employed. pMFYMhpR is a pMFY42-based plasmid that harbors mhpR encoding the positive regulator of 3-(3-hydroxyphenyl)propionic acid (3HPP) degradation genes (mhp) (37) and the promoter region of mhp genes in TA441 (Fig. S4 above in the supplemental material) (accompanying paper). Genes cloned downstream of the mhp promoter are induced upon the addition of 3HPP. pMFYMhpRORF76, a pMFYMhpR-derivative carrying ORF7 and ORF6 downstream of the mhp promoter, was constructed and introduced into the ORF7^-6^- double mutant. ORF7^-6^- carrying pMFYMhpRORF76 and the corresponding negative control (ORF7^-6^- carrying pMFY42) were incubated with cholic acid for 7 days, after which 3HPP was added and the culture was analyzed every day for 3 days. Mass chromatograms of the cultures before and after incubation with 3HPP are reported in Fig. S4 in the supplemental material. We also analyzed the culture of ORF7^-6^- carrying pMFYMhpRORF76 incubated for 8 and 10 days in the absence of any 3HPP supplementation to exclude the possibility that a prolonged incubation altered the compounds. The mass chromatogram of the cultures was almost identical to the one obtained after 7 days (data not shown), thus excluding any effect of prolonged incubation time. Compounds harboring a 12β hydroxyl group, such as XV (Fig. S4 in the supplemental material A), XVIII (Fig. S4C), and XXI (detected as XIX in Fig. S4D), exhibited a large drop in amount, whereas XII, XVII, XXIII [compounds with a double bond at C10(12)], and XIII [a compound with a single bond at C10(12)] (Fig. S4B) exhibited an increase in 3HPP-induced ORF7^-6^- carrying pMFYMhpRORF76. These results confirmed that ORF7- and ORF6-encoded enzymes removed the C12β hydroxyl group via dehydration.

In conclusion, the enzymes encoded by steA, steB, ORF7, and ORF6 catalyze the dehydrogenation of the C12α hydroxyl group on the C-ring to a ketone, hydrogenation of the ketone to the C12β hydroxyl group (32), dehydration of the C12β hydroxyl group to produce a double bond at C10(12), and reduction of the double bond to a single bond to remove the C12α hydroxyl group, respectively. Accordingly, ORF7 and ORF6 were named steC and steD.

Overall steroid degradation in C. testosteroni: degradation of A,B,C,D-rings and removal of the hydroxyl group at C12

The function of ORF7 (steC) and ORF6 (steD) was revealed in this study and, therefore, the role of all genes in the two steroid degradation clusters on both ends of the mega-cluster responsible for steroidal A,B,C,D-ring breakdown was identified. Hereafter, we summarized the entire steroid degradation pathway in C. testosteroni TA441 (Fig. 5).

Fig 5.

Review of steroid degradation pathway in C. testosteroni TA441. The mega-cluster of steroid degradation genes in TA441 is shown below the degradation pathway; the aromatic ring-degradation gene cluster (tesG to scdA) and the β-oxidation gene cluster (steA to tesR) locate both ends of this 120kb-mega cluster. 3α-Hydroxy-dehydrogenase (3α-DH) gene and 3-ketosteroid Δ4–5 isomerase (ksi) gene are in the DNA region between the two clusters. Possible degradation genes for the side chain at C17 of cholic acid are in the DNA region indicated with stripes. Compounds in the box: the compound or the derivative was isolated and identified by NMR, etc. Compounds underlined: identified by mass spectrum and conversion experiments. Compounds are as follows: 3α,7α,12α-trihydroxy-17-oxo-androstan (I); 3α,7α-dihydroxy-12,17-oxo-androstan (II); 3α,7α,12β-trihydroxy-17-oxo-androstan (III); 7α,12β-dihydroxy-androst-4-ene-3,17-dione (XXVII); 7α,12β-dihydroxy-androst-1-ene-3,17-dione (XXVIII); androsta-1,4-diene-3,17-dione (ADD) (R1, R2 = H); 3-hydroxy-9,10-secoandrosta-1,3,5(10)-triene-9,17-dione (3-HSA) (R1, R2 = H) (XXIX); 3,4-dihydroxy-9,10-secoandrosta-1,3,5(10)-triene-9,17-dione (3,4-DHSA) (R1, R2 = H) (XXX); 4,5–9,10-diseco-3-hydroxy-5,9,17-trioxoandrosta-1(10),2-dien-4-oic acid (R1, R2 = H) (XXXI); (2Z,4Z)−2-hydroxyhexa-2,4-dienoic acid (XXXII); 9,17-dioxo-1,2,3,4,10,19-hexanorandrostan-5-oic acid (3aα-H-4α [3′-propionic acid]−7aβ-methylhexahydro-1,5-indanedione) (R1, R2 = H) (XXIV); 12β-hydroxy-9,17-dioxo-1,2,3,4,10,19-hexanorandrostan-5-oic acid (R1=βOH, R2 = H) (XV); 9,17-dioxo-1,2,3,4,10,19-hexanorandrost-10 (12)-en-5-oic acid (R2 = H) (XXXIII); 9,17-dioxo-1,2,3,4,10,19-hexanorandrost-6-en-5-oic acid (R1 = H) (XXXIV); 9α-hydroxy-17-oxo-1,2,3,4,10,19-hexanorandrostan-5-oic acid (R1 = H) (XXXV); 9α-hydroxy-17-oxo-1,2,3,4,10,19-hexanorandrost-6-en-5-oic acid (R1 = H) (XXV); 7β,9α-dihydroxy-17-oxo-1,2,3,4,10,19-hexanorandrostan-5-oic acid (R1 = H) (XXXVI); 9α-hydroxy-7,17-dioxo-1,2,3,4,10,19-hexanorandrostan-5-oic acid (R1 = H) (XXXVII); 9α-hydroxy-17-oxo-1,2,3,4,5,6,10,19-octanorandrostan-7-oic acid (R1 = H) (VI); 9,17-dioxo-1,2,3,4,5,6,10,19-octanorandrostan-7-oic acid (R1 = H) (XXXVIII); 9,17-dioxo-1,2,3,4,5,6,10,19-octanorandrost-8(14)-en-7-oic acid (R1 = H) (XXXIX); 17-dihydroxy-9-oxo-1,2,3,4,5,6,10,19-octanorandrost-8(14)-en-7-oic acid (R1 = H) (XL); 14-hydroxy-9-oxo-1,2,3,4,5,6,10,19-octanor-13,17-secoandrostane-7,17-dioic acid (R1 = H) (VII); 9,14-dihydroxy-1,2,3,4,5,6,10,19-octanor-13,17-secoandrostane-7,17-dioic acid (R1 = H) (XLII); 3-hydroxy-6-methyl-7-oxo-decane-1,10-dioic acid (VIII); 6-methyl-3,7-dioxo-decane-1,10-dioic acid (IX); 3,5-dihydroxy-6-methyl-7-oxo-decane-1,10-dioic acid (XXII); 5-hydroxy-6-methyl-3,7-dioxo-decane-1,10-dioic acid (XXI); 6-methyl-3,7-dioxo-dec-5-ene-1,10-dioic acid (XIX); 4-methyl-5-oxo-oct-3-ene-1,8-dioic acid (X); 4-methyl-5-oxo-octane-1,8-dioic acid (XI); 4-methyl-5-oxo-oct-2-ene-1,8-dioic acid (XLIII); and 3-hydroxy-4-methyl-5-oxo-octane-1,8-dioic acid (XLIV). Enzymes are as follows: SteA (dehydrogenase for 12α-OH to 12-ketone), SteB (hydrogenase for 12-ketone to 12β-OH), TesH (Δ1-dehydrogenase), TesI (Δ4-dehydrogenase), TesJ (ADD-hydroxylase at C9), TesA1A2 (3-HSA {XXIX}-hydroxylase at C4), TesB (meta-cleavage enzyme for 3.4-DHSA {XXX}), TesD (XXXI-hydrolase), TesE (XXXII-hydratase), TesF (aldolase), TesG (acetoaldehyde dehydrogenase), SteC [dehydradase for 12β-OH to produce a double at C10(12)], SteD (reductase for a double at C10(12) to a single bond), ScdA (CoA-transferase for XXIV), ScdG (hydrogenase primarily for 9-OH of VI-CoA ester), ScdC1C2 (Δ6-dehydrogenase for XXXV-CoA ester), ScdD (XXV-CoA ester hydratase), ScdE (XXXVI-CoA ester dehydrogenase at C7), ScdF (XXXVII-CoA ester thiolase/CoA-transferase), ScdK [Δ8(14)-dehydrogenase for XXXVIII-CoA ester], ScdY (XL-CoA ester hydratase), ScdL1L2 (putative CoA-transferase/isomerase necessary for C-ring cleavage of XLII-CoA ester), ScdM1M2 (XI-CoA ester dehydrogenase), and ScdN (XLIII-CoA ester hydratase). Genes for the C-, D-, and cleaved B-ring degradation are induced by the compounds with steroidal four rings but are not induced with indane (XXIV) and the derivatives (unpublished data).

Side chain degradation and conversion

Cholic acid degradation is initiated by the removal of the side chain at position C17. The major intermediates identified in a previous study (35) pointed to β-oxidation as the main degradation mechanism. The corresponding genes were thought to localize to the mega-cluster (striped regions in Fig. 5). After the side chain at C17 is removed, the C12α hydroxyl group on the C-ring (I) is converted to a ketone (II) by SteA, followed by hydrogenation of the ketone to the C12β hydroxyl group (III) by SteB (32). This inversion of stereochemistry is indispensable for subsequent B-ring cleavage.

A,B-ring cleavage

After dehydrogenation of the 3βhydroxyl group to a ketone and dehydrogenation of the A-ring by TesH (Δ1 dehydrogenase) and TesI (Δ4 dehydrogenase) to produce androsta-1,4-diene-3,17-dione or one of its derivatives, the addition of a hydroxyl group at position C9 by TesJ (formerly ORF17-encoded protein) leads to spontaneous cleavage of the B-ring. This is accompanied by aromatization of the A-ring to 3-hydroxy-9,10-secoandrosta-1,3,5(10)-triene-9,17-dione (R1, R2 = H) (XXIX) (38–40). A hydroxyl group is added at position C4 by TesA1A2 (XXX) (41) and the aromatized A-ring is cleaved by the meta-cleavage enzyme TesB (XXXI) (42), which is followed by TesD-mediated hydrolysis of XXXI into cleaved A-ring, (2Z,4Z)−2-hydroxyhexa-2,4-dienoic acid (XXXII), and C,D-ring with cleaved B-ring, XXIV (43, 44). This process is similar to the bacterial “meta-cleavage pathway” responsible for aromatic compound degradation such as biphenyl.

C,D-ring cleavage

In contrast to A,B-ring cleavage, C,D-ring cleavage proceeds through a series of β-oxidation cycles. After hydrolysis, CoA is added to XXIV at C5 by the CoA transferase ScdA (35). Removal of the C12β hydroxyl group and hydrogenation of the C9 ketone to a C9 hydroxyl occurs prior to the first β-oxidation cycle, which removes the cleaved B-ring. This observation is based on the accumulation of 7,12β-dihydroxy-9,17-oxo-1,2,3,4,10,19-hexanorandrostan-5-oic acid (R1 =βOH, R2 = OH) (XXIV) in ScdA^- incubated with cholic acid, whereas intermediates in the β-oxidation of the cleaved B-ring have a hydroxyl group at C9. Most intermediates in the first β-oxidation cycle do not possess a C12β hydroxyl group, but C,D-ring cleavage can proceed also in its presence, albeit less efficiently. Removal of the C12β hydroxyl group is a two-step reaction that involves dehydration by ScdC and hydrogenation by ScdD. Hydrogenation of the C9 ketone to the C9 hydroxyl group is indispensable to initiate β-oxidation, but the corresponding enzyme has not been identified. ScdG can potentially catalyze this reaction, yet the primary function of ScdG is the conversion of the C9 hydroxyl group to C9 ketone on 9α-hydroxy-17-oxo-1,2,3,4,5,6,10,19-octanorandrostan-7-oic acid (R1 = H) (VI)-CoA ester, a compound generated by the first β-oxidation (45). There may be one or more enzymes other than ScdG, which act on the C9 ketone/hydroxyl group. XXIV-CoA ester is converted to 9α-hydroxy-17-oxo-1,2,3,4,10,19-hexanorandrostan-5-oic acid (R1 = H) (XXXV)-CoA ester and then dehydrogenated by ScdC1C2 to XXV-CoA ester (46). When a hydroxyl group at C7 is present, XXV-CoA ester is produced in ScdC1C2^- cultures, suggesting a bypass route (presumably via dehydration) to produce XXV-CoA ester (Fig. 5: XXIV-CoA ester (R2 = OH) → XXXIV-CoA ester → XXV-CoA ester) (47). XXV-CoA ester undergoes β-oxidation to VI-CoA ester via ScdD hydratase, ScdE dehydrogenase, and ScdF CoA-transferase (48). Throughout the β-oxidation of the cleaved B-ring to the cleavage of the C-ring, compounds with “C9 ketone and a double bond at C8(14)” and “C9 hydroxyl group and a single bond at C8(14)” are major in most of the mutant cultures. Given that compounds with “C9 ketone and a single bond at C8(14)” and “C9 hydroxyl group and a double bond at C8(14)” have not been isolated except for XXIV and the derivatives, their CoA-esters are likely unstable and are converted to either one of the former two compounds. VI-CoA ester is dehydrogenated to 9,17-dioxo-1,2,3,4,5,6,10,19-octanorandrostan-7-oic acid (R1 = H) (XXXVIII)-CoA ester; however, this is a reversible reaction and XXXVIII-CoA ester is readily dehydrogenated by ScdK at C8(14) to yield 9,17-dioxo-1,2,3,4,5,6,10,19-octanorandrost-8(14)-en-7-oic acid (R1 = H) (XXXIX)-CoA ester (45). Next, a water molecule is added at position C17 to produce a geminal diol and the D-ring is cleaved at position C13(17) following the addition of another water molecule at C14 by ScdY (accompanying paper). The geminal diol, 17-dihydroxy-9-oxo-1,2,3,4,5,6,10,19-octanorandrost-8(14)-en-7-oic acid (R1 = H) (XL), was detected in the present study (accompanying paper), but the enzyme for the production is unclear. CoA-esters of 9-oxo-1,2,3,4,5,6,10,19-octanor-13,17-secoandrost-8(14)-ene-7,17-dioic acid (XLI), 9- hydroxy-1,2,3,4,5,6,10,19-octanor-13,17-secoandrosta-13-en-17-oic acid (XLV), and 13-hydroxy-9-oxo-1,2,3,4,5,6,10,19-octanor-13,17-secoandrost-8(14)-en-17-oic acid (XLVI) were at first proposed as intermediates generated along with the C-ring and cleaved D-ring based on analysis of ScdL1L2^- cultures (49). However, subsequent studies suggested XLI was produced from 14-hydroxy-9-oxo-1,2,3,4,5,6,10,19-octanor-13,17-secoandrostane-7,17-dioic acid (VII)-CoA ester during the isolation procedure (50) and the major product of ScdL1L2 was in fact 3-hydroxy-6-methyl-7-oxo-decane-1,10-dioic acid (R1 = H) (VIII)-CoA ester (accompanying paper). VIII and XLV suggested that the substrate of ScdL1L2 was 9,14-dihydroxy-1,2,3,4,5,6,10,19-octanor-13,17-secoandrostane-7,17-dioic acid (XLII)-CoA ester and the function of ScdL1L2 was the conversion of XLII-CoA ester to VIII-CoA ester. Given the similarity with steroid degradation by M. tuberculosis, whereby the C-ring of XLI-CoA ester is cleaved by the IpdAB hydrolase (a homolog of ScdL1L2) in the presence of FadA6 (a homolog of ScdF) to produce 6-methyl-3,7-dioxo-decane-1,10-dioic acid (R1 = H) (IX)-CoA ester (51), the details of the process in C. testosteroni and other genera of bacteria may require further elucidation. The ring cleavage product, VIII-CoA ester, undergoes the second β-oxidation, whereby the C3 hydroxyl group is dehydrogenated to IX-CoA ester, followed by the removal of two carbons by the ScdJ CoA-transferase (50). ScdF, and maybe other CoA-transferases, may contribute to this reaction because disruption of ScdJ did not completely stop the conversion of IX-CoA ester to 4-methyl-5-oxo-octane-1,8-dioic acid (XI)-CoA ester. XI-CoA ester undergoes the third β-oxidation involving the ScdM1M2 dehydrogenase, followed by the ScdN hydratase (52), which catalyzes the last reaction among those catalyzed by the enzymes encoded in these two clusters. When the initial steroid compound harbors a C12 hydroxyl group, a portion of it undergoes C,D-ring cleavage with the hydroxyl group, thereby yielding XXII-CoA ester, VIII derivative with a hydroxyl group at C5 (C12 on steroidal C-ring), and XI-CoA ester obtained via 4-methyl-5-oxo-oct-3-ene-1,8-dioic acid (X)-CoA ester (52). The low substrate specificity of degradation enzymes allows bacteria to utilize a variety of steroid compounds. C,D-ring cleavage process would be similar in bacterial estrogen degradation because XXIV is produced as an intermediate compound in estrone degradation by Rhodococcus sp. strain B50 (53).

Conclusion

This study revealed that the enzymes encoded by ORF7 (steC) and ORF6 (steD) are the dehydrase of the C12β hydroxyl group to produce a double bond at C10(12) and the reductase for the double bond to a single bond, respectively. Together with SteA and SteB, which convert the C12α hydroxyl group to the C12β hydroxyl group via the ketone group (32), they remove the C12α hydroxyl group of steroids. A study on TA441 so far has revealed most of the aerobic bacterial steroid degradation and the functions of the genes in the two major clusters at either end of the 120kb-mega-cluster of steroid degradation genes. Since these clusters were found, bacterial steroid degradation genes similar to these have been reported from many genera of bacteria in both Proteobacteria and Actinobacteria (6, 7), suggesting they are widespread in the environment. Further analysis of the distribution of these steroid-degrading bacteria is expected to clarify the degradation in the environment of plant- and microorganism-derived steroids such as saponins and glycoside steroids, steroid compounds derived from animal feces, and steroid compounds emitted by human activities. Some researchers are skeptical about the involvement of aerobic steroid degradation in the intestinal microbiota because no C. testosteroni or other aerobic steroid-degrading organisms have been detected in fecal samples. However, it has been reported that C. testosteroni is one of the members of the gut microbiome (54). C. testosteroni has been found in the cecum (15), suggesting the possibility that it is present near the end of the small intestine or at the junction of the small and large intestines, where there is more oxygen compared to the large intestine. The findings in TA441 will provide essential information to elucidate the effects of aerobic steroid degradation on the gut microbiota.

MATERIALS AND METHODS

Culture conditions

Mutant strains of C. testosteroni TA441 were grown at 30°C in a mixture of equal volumes of Lysogeny broth (LB) medium and C medium (a mineral medium for TA441) (42) with suitable carbon sources. This mixed media is used because the mutants accumulate more amount of intermediate compounds than with C medium or LB medium (unpublished data). Cholic acid and other steroids were added as filter-sterilized dimethyl sulfoxide (DMSO) solutions with a final concentration of 0.1% (wt/vol). 3HPP was purchased from Alfa Aesar (Lancashire, UK).

Construction of gene-disrupted mutants, plasmids, and mutants for complementation experiments

For the construction of ORF7^-6^- mutant (Table 2), pUC19 (55)-based plasmid carrying DNA region of TA441 with insertion of a kanamycin-resistance gene (Km^r) without a terminator between the ApaI site in ORF7 and BglII site in ORF6 (pUCORF76-Km^r) was used (Table 3). The plasmid was introduced into TA441 via electroporation. The mutants with the insertion by homologous recombination were selected on LB plates with kanamycin (36). Insertion of the Km^r was confirmed by southern hybridization and/or PCR amplification. DNA fragments containing ORF7, ORF6, and ORF76 were obtained by PCR amplification and introduced into a broad-host-range plasmid pMFY42 (34), which can be maintained in Pseudomonas spp. and several related species conferring tetracycline resistance to construct pMFYORF7 pMFYORF6, and pMFYORF76, respectively. Retention of the plasmids by the gene-disrupted mutants and the transformants was confirmed by PCR amplification. For PCR amplification, DNA polymerase KOD-plus ver. 2 (TOYOBO, Japan) was used. The primers are listed in Table 4.

Isolation and identification of compounds accumulated in ORF6^- culture

ORF6^- was grown in 500 mL of 1/2LB + C medium with 0.1% cholic acid and incubated at 30°C for about 3 days. 6b and 6c were isolated from the amount of 10 mg and 4.1 mg, respectively, from this culture. The amount of 6a was too small, so ORF6^- was grown again in a total of 1,000 mL of 1/2LB + C medium for 2 days, and 6 mg of 6a was isolated. After the incubation, the culture was extracted twice with the same volume of ethyl acetate. The ethyl acetate fraction was dried and concentrated, dissolved in a small amount of methanol. The compounds were separated by Waters 600 HPLC (Nihon Waters, Tokyo, Japan) with an Inertsil ODS-3 column (20 × 250 mm, GL Science) and a solvent, with the composition CH₃CN:CH₃OH:H₂O:TFA of 50:10:40:0.05, flow rate 1 mL/min, at 40°C and the fraction containing each compound was collected from the eluent. The fraction was dried and kept in the refrigerator and subjected to FAB-MS and NMR analyses.

General experimental procedures

FAB-MS (positive-ion mode) was recorded on a JEOL JMS-700 mass spectrometer (JEOL Ltd., Tokyo, Japan), using a glycerin matrix. 1D and 2D NMR spectra were recorded on a JNM-ECP500 or JNM-ECA600 spectrometer (JEOL Ltd, Tokyo, Japan). Tetramethylsilane at 0 ppm in CDCl₃ solution and residual proton signal at 2.49 ppm in DMSO-d6 solution were used as internal references for ¹H chemical shifts. ¹³C chemical shifts were obtained with reference to DMSO-d6 (39.5 ppm) or CDCl₃ (77.0 ppm) at 25°C.

HPLC analysis (Fig. 2)

After the addition of a double volume of methanol to the culture, the mixture was centrifuged, and the supernatant was directly injected into an HPLC. The HPLC (Alliance 2695 with UV detector and 996 photodiode array detector, Nihon Waters, Tokyo, Japan) equipped with an Inertsil ODS-3 column (4.6 × 250 mm, GL Sciences Inc., Tokyo, Japan) was used, and elution was carried out using a linear gradient from 20% solution A (CH₃CN:CH₃OH:TFA = 95:5:0.05) and 80% solution B (H₂O:CH₃OH:TFA = 95:5:0.05) to 65% solution A and 35% solution B over 10 min; this was maintained for 3 min and then changed to 20% solution A. The flow rate was 1.0 mL/min.

Reverse-phase liquid chromatography with tandem mass spectrometry

For LC/MS/MS analysis, 2 µL of the samples prepared the same way as those for HPLC/MS analysis was injected into the system. Agilent 1100 HPLC (Agilent, CA) with a mass spectrometer, 4000 QTRAP MS/MS system (AB SCIEX, Framingham, MA, USA) in negative ion mode, was used with L-column2 ODS (1.5 × 150 mm) Type L2-C 18.5 µm, 12 mm (GL Science, Tokyo, Japan) and elution was carried out using 90% solution A (H₂O:HCOOH = 100:0.1) and 10% acetonitrile for 1 min, followed by a linear gradient from 90% solution A and 10% acetonitrile to 20% solution A and 80% acetonitrile over 7 min, which was maintained for 2 min. The flow rate was 0.2 mL/min. The MS/MS conditions were as follows: ion source temperature, 450°C; spray needle voltage, −4.5 kV; sheath gas pressures, 60 units for gas 1 and 70 units for gas 2; and curtain gas flow, 15 units. The collision energy was 20 V. Ions were detected by the Q3 detector of the MS/MS.

DATA AVAILABILITY

The data that support the findings of this study are available from the corresponding author, M.H., upon reasonable request.

SUPPLEMENTAL MATERIAL

The following material is available online at https://doi.org/10.1128/aem.00143-23.

Supplemental file 1. aem.00143-23-s0001.pdf.

Fig. S1.

Supplemental file 2. aem.00143-23-s0002.pdf.

Fig. S2.

Supplemental file 3. aem.00143-23-s0003.pdf.

Fig. S3.

Supplemental file 4. aem.00143-23-s0004.pdf.

Fig. S4.

Supplemental file 5. aem.00143-23-s0005.pdf.

Table S1.

ASM does not own the copyrights to Supplemental Material that may be linked to, or accessed through, an article. The authors have granted ASM a non-exclusive, world-wide license to publish the Supplemental Material files. Please contact the corresponding author directly for reuse.