The desA and desB genes from Clostridium scindens ATCC 35704 encode steroid-17,20-desmolase

Clostridium scindens is a gut microbe capable of removing the side-chain of cortisol, forming 11 β -hydroxyandrostenedione. A cortisol-inducible operon ( desABCD ) was previously identified in C. scindens ATCC 35704 by RNA-Seq. The desC gene was shown to encode a cortisol 20 α -hydroxysteroid dehydrogenase (20 α -HSDH). The desD encodes a protein annotated as a member of the major facilitator family, predicted to function as a cortisol transporter. The desA and desB genes are annotated as N-terminal and C-terminal transketolases, respectively. We hypothesized that the DesAB forms a complex and has steroid-17,20-desmolase activity. We cloned the desA and desB genes from C. scindens ATCC 35704 in pETDuet for overexpression in E. coli . The purified recombinant DesAB was determined to be a 142 ± 5.4 kDa heterotetramer. We developed an enzyme-linked continuous spectrophotometric assay to quantify steroid-17,20-desmolase. This was achieved by coupling DesAB-dependent formation of 11 β -hydroxyandrostenedione with the NADPH-dependent reduction of the steroid 17-keto group by a recombinant 17 β -HSDH from the filamentous fungus, Cochliobolus lunatus . The pH optimum for the coupled assay was 7.0 and kinetic constants using cortisol as substrate were K m of 4.96 ± 0.57 µ M and k cat of 0.87 ± 0.076 min -1 . Substrate-specificity studies revealed that rDesAB recognized substrates regardless of 11 β -hydroxylation, but had an absolute requirement for 17,21-dihydroxy 20-ketosteroids. clustering gut mobile phase, 2 formic acid and formic (B); total mobile mobile phase mobile phase next until 35%. next


Introduction
Clinical studies in the 1950s reported urinary excretion of 17-ketosteroids following rectal infusion of cortisol for the treatment of ulcerative colitis. Concomitant neomycin treatment precluded urinary 17-ketosteroid excretion, implicating gut bacteria in the conversion of a C 21 glucocorticoid to a C 19 androstane/androgenic metabolite (1,2). Later, it was shown that rat and human fecal bacteria were capable of this cortisol side-chain cleavage reaction, which became known as steroid-17,20-desmolase (3,4). Serial dilutions of human and rat fecal samples revealed that ~ 10 6 bacterial gram -1 of wet-weight express steroid-17,20-desmolase activity (5). Subsequently, an organism was isolated in pure culture that expressed steroid-17,20-desmolase and 20α-hydroxysteroid dehydrogenase (20α-HSDH) activities, and was named Clostridium scindens, whose species epithet means "to cut", a reference to separation of the side-chain from the steroid D-ring (6).
Cortisol-inducible steroid-17,20-desmolase was partially purified, and 20α-HSDH was purified to apparent electrophoretic homogeneity (SDS-PAGE) by traditional chromatographic separation (7,8). The application of genome-wide transcriptomics (RNA-Seq) identified a polycistronic cortisol-inducible operon (desABCD) which included a gene encoding 20α-HSDH (desC) (9). The desA and desB genes are annotated as Nterminal and C-terminal transketolases, respectively. Transketolase catalyzes the TPPdependent transfer of glycol aldehyde to an aldose acceptor. Comparison of the transketolation reaction to that of steroid-17,20-desmolase suggests an analogous reaction, and that desAB encodes a novel steroid transketolase (9). C. cadavaris and Butyricicoccus desmolans were shown previously to generate 20β-dihydrocortisol as well by guest, on July 19, 2018 www.jlr.org on LB agar plates supplemented with ampicillin (100 µg/mL) and chloramphenicol (50 µg/mL). After 16 hours, five isolated colonies were used to inoculate 10 mL of fresh LB medium supplemented with antibiotics and grown at 37°C for 6 hours with vigorous aeration. The pre-cultures were then added to fresh LB medium (1 L) supplemented with the same antibiotics at the same concentrations and grown with vigorous aeration at 37°C. At OD 600 of 0.3, IPTG was added to each culture at a final concentration of 0.1 mM and the temperature was decreased to 16°C. Following 16 hours of culturing, cells were pelleted by centrifugation (4,000 x g, 30 min, 4°C) and re-suspended in 30 mL of binding buffer (20 mM Tris-HCl, 150 mM NaCl, 20% glycerol, and 10 mM 2-mercaptoethanol at pH 7.9). The cell suspension was subjected to four passages through an EmulsiFlex C-3 cell homogenizer (Avestin, Ottawa, Canada), and the cell lysate was clarified by centrifugation at 20,000 x g for 30 min at 4°C.
The recombinant steroid-17,20-desmolase was then purified using TALON® Metal Affinity Resin (Clontech Laboratories, Mountain View, CA) as per the manufacturer's 7 protocol. The recombinant protein was eluted using an elution buffer composed of 20 mM Tris-HCl, 150 mM NaCl, 10% glycerol, 10 mM 2-mercaptoethanol, and 250 mM imidazole at pH 7.9. The recombinant 17β-HSDH was then purified using Strep-Tactin® resin (IBA GmbH, Goettingen, Germany) as per manufacturing protocol. The recombinant protein was eluted using an elution buffer composed of 20 mM Tris-HCl, 150 mM NaCl, 10% glycerol, 10 mM 2-mercaptoethanol, and 2.5 mM desthiobiotin at pH 7.0. The purity of the proteins were assessed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), and protein bands were visualized by staining with Coomassie Brilliant Blue G-250 Dye. The protein concentrations were calculated based on the computed molecular mass and extinction coefficient. The observed subunit mass for each protein was calculated by migration distance of purified protein to standard proteins using ImageJ (https://imagej.nih.gov/ij/docs/faqs.html).

Gel filtration chromatography
Gel filtration chromatography was carried out on a Superose-6 10/300 analytical column (GE Healthcare, Piscataway, NJ) attached to an ÄKTAxpress chromatography system (GE Healthcare, Piscataway, NJ) at 4°C. The eluted recombinant steroid-17,20desmolase from the metal affinity resin was concentrated and loaded on to the column with a buffer composed of 50 mM Tris-Cl, 150 mM NaCl, 10% glycerol, and 10 mM 2mercaptoethanol at pH 7.0. The native molecular mass of the protein was calculated by peak retention volume of protein to retention volume of standard proteins. total run time including equilibration, 41 min. The initial mobile phase composition was 70% mobile phase A and 30% mobile phase B. The percentage of mobile phase B was changed linearly over the next 5 min until 35%. Over the next 25 min, the percentage was increased to 98% linearly. After that the percentage was maintained for 5 min, the mobile phase composition was allowed to return to the initial conditions and allowed to equilibrate for 5 min. The injection volume was 10 µL. The DAD detector was used at a wavelength of 254 nm. Peak retention times and peak areas of samples were compared with standard steroid molecules.
The mass spectrometer (LCMS-IT-TOF) was operated with an electrospray ionization (ESI) source in positive mode. The nebulizer gas pressure was set at 150 kPa with the source temperature of 200 °C and the gas flow at 1.5 L min -1 . The detector voltage was 1.65 kV. High-purity nitrogen gas was used as collision cell gas. The raw chromatogram and mass spectrogram data were processed with the LC solution Workstation software (Shimadzu).

Continuous enzyme-coupled spectrophotometric assay
Steroid-17,20-desmolase activity was measured aerobically at 25°C by monitoring the conversion of NADPH to NADP + 340 nM (ε=6,220 M -1 .cm -1 ) reflecting DesABcatalyzed conversion of cortisol to 11β-hydroxyandrostenedione followed by NADPHdependent 17β-HSDH conversion of 11β-hydroxyandrostenedione to a 11β-testosterone. substrates lacking a 17-hydroxyl, or those substrates whose 17-keto products had low specificity for recombinant C. lunatus 17β-HSDH (rCL17HSDH), was to run the basic assay without rCL17HSDH and NADPH. Reactions were terminated with 1N HCl and reaction products extracted and separated by LC/MS as described above. A standard curve of peak area vs. concentration of the reaction products was generated to quantify reaction rates.

Cloning, overexpression of desAB, and purification of recombinant DesAB
The desA gene encodes a 296 amino acid protein (CLOSCI_00899) with predicted The steroid-17,20-desmolase assay previously utilized is discontinuous, timeconsuming, and expensive. We therefore sought another approach to quantify steroid-17,20-desmolase activity using a continuous enzyme-coupled spectrophotometric assay. Suspecting that an aldose-acceptor may be utilized in the reaction, glyceraldehyde-3phosphate and xylulose-5-phosphate were tested at 50 µM, 100 µM, and 500 µM; however, we did not observe increased enzymatic activity.
We tested substrate-specificity using a combination of methods, depending on whether steroids were 17-hydroxylated. The enzyme-coupled continuous spectrophotometric assay was relegated to 17-hydroxylated substrates. Those reaction products from substrates lacking 17-hydroxy were characterized by LC-MS. Furthermore, we tested NADPH-dependent rCL17HSDH activity against the expected C-19 steroid-17,20desmolase products to determine whether we could make a direct comparison between substrates whose products differed structurally. We determined that Ring-A reduced products (50 µM concentration), such as etiocholanolone and 11β-hydroxyetiocholanolone, are not substrates for rCL17HSDH at, resulting in 8.06 ± 2.65% and 4.73 ± 0.68% activity, respectively, relative to 11β-hydroxyandrostenedione (three technical replicates). These substrates were therefore tested for activity using rDesAB with separation of steroids by LC/MS.

Discussion
In the current study, we report that the desA and desB genes from C. scindens lymphophilum express 20β-HSDH (11) in addition to steroid-17,20-desmolase. Oxidationreduction of the C20 oxygen is predicted to function as a regulatory "switch", interconverting cortisol between DesAB substrate and non-substrate forms. The desD gene is co-expressed with desABC (9) and encodes a predicted 51.2 kDa transport protein in the major facilitator superfamily. DesD is predicted to transport cortisol from the gut lumen into the bacterial cytoplasm. RNA-Seq analysis also identified a cortisol-inducible ABC-type multidrug transport system predicted to export 11βhydroxyandrostenedione (9).
The conversion of cortisol to 11β-hydroxyandrostenedione by C. scindens is an important step in a multi-species gut microbial endocrine pathway that results in the formation of 11β-hydroxyandrogens (9). Previous studies reveal a number of cortisol metabolites following incubation with mixed human or rat fecal bacteria included 3α,11β,17β-trihydroxy-5β-androstane, 3α,11β,17β-trihydroxy-5α-androstane, demonstrating the ability of gut microbes to form 5α-, 17β-reduced metabolites (3,4). Some of the taxa responsible for these biotransformations have been identified. C.
Extrapolations from radiometric studies in non-human primates (24) and data from humans (25)  downstream 5α-, 17β-reduced metabolites, are now thought to play an important role in activation of androgen receptor (27,28). The field of microbial endocrinology has emerged and recognizes the need to characterize androgen production by C. scindens [29][30][31][32][33][34]. Many host immune cells express androgen-receptor [35,36], and it is possible that C. scindens affects immune function through generation of androgens. Androgens directly affect bacterial growth and virulence [37][38][39]. Intriguingly, steroids have been shown to influence C. difficile germination [40]. 11β-hydroxy-androgens are also inhibitors of the host enzyme 11β-HSD2 [41,42], but in kidney and vascular smooth muscle may be pro-hypertensive [42]. Inhibition of 11β-HSD2 in colonocytes may be protective against colorectal cancer [43]. Prostate tissue contains 11β-hydroxyandrogens despite inhibition of host enzymes involved in synthesizing androgens (chemical castration) [27,28]. The role of 11β-OHAD formation by the gut microbiota in androgen accumulation in the prostate is unknown; however, we discovered that bacteria isolated from urine also express the bacterial steroid-17,20-desmolase pathway [11], and the majority of cortisol excretion from the body is via the urine [24,25]. Of note, we recently reported identification of the steroid-17,20-desmolase gene cluster in the genomes of urinary bacterial isolates and confirmed steroid-17,20-desmolase activity in an isolate of Propionimicrobium lymphophylum [11]. Verification that desA and desB genes encode steroid-17,20-desmolase is an important step towards the application of nucleic acidbased approaches such as quantitative polymerase chain, and metagenomic sequencing in determining whether correlations exist between levels/expression of desAB and host phenotypes and disease states. www.jlr.org

Downloaded from
As gut metagenomic sequencing becomes less costly, and more widespread, assigning function to particular nucleic acid and deduced amino acid sequences will be important for hypothesis generation, and testing. The desA and desB genes are a clear case in point. The desABCD operon is annotated as participating in carbohydrate metabolism, due largely to the desA and desB gene product homology to N-terminal and C-terminal TPP-dependent transketolases, respectively (Figure 1). Early culture-based studies established that steroid-17,20-desmolase activity is induced by cortisol (6)(7)(8).
Utilization of RNA-Seq to identify differentially expressed genes, coupled with biochemical characterization of the recombinant gene products was necessary to determine the nucleic acid sequences encoding this bacteria steroid biotransforming pathway (9).
A recent report describes side-chain cleavage of cholesterol by Mycobacterium tuberculosis and revealed a thiolase-derived enzyme, a steroid-aldolase, involved in microbial steroid-degradation (44) (Figure 7). It is therefore possible that additional sidechain cleaving gut microbial members will be identified that are capable of aldolase cleavage of host and dietary steroids in the gut. This is in contrast to eukaryotic cells which typically utilize oxygen-dependent P450 monooxygenases (45).
The low turnover number is within the range of previously reported prokaryotic enzymes involved in secondary metabolism (46), and may indicate that selection pressure may be less stringent for optimization of enzyme rate for functions that may not contribute greatly to organismal fitness. Studies are now ongoing to understand important amino acid residues involved in substrate-binding to steroid-17,20-desmolase, and catalysis, as well as the identity and fate of the side-chain. Development of a novel enzyme-coupled continuous spectrophotometric assay to quantify steroid-17,20-desmolase activity against cortisol is expected to hasten results from these efforts.
Current limitations of the assay with respect to androgen specificity of rCL17HSDH may be addressed in the future either through protein engineering, or by identifying naturally occurring 17α-or 17β-HSDHs. In addition, future studies will be required to determine the identity of the side-chain and whether an aldose acceptor is utilized in vivo. A detailed understanding of the steroid-17,20-desmolase will be important in uncovering how anaerobic bacteria co-opted pentose-phosphate pathway enzymes for steroid side-chain cleavage.      c.