Steroid Metabolism in Thermophilic Actinobacterium Saccharopolyspora hirsuta VKM Ac-666T

The application of thermophilic microorganisms opens new prospects in steroid biotechnology, but little is known to date on steroid catabolism by thermophilic strains. The thermophilic strain Saccharopolyspora hirsuta VKM Ac-666T has been shown to convert various steroids and to fully degrade cholesterol. Cholest-4-en-3-one, cholesta-1,4-dien-3-one, 26-hydroxycholest-4-en-3-one, 3-oxo-cholest-4-en-26-oic acid, 3-oxo-cholesta-1,4-dien-26-oic acid, 26-hydroxycholesterol, 3β-hydroxy-cholest-5-en-26-oic acid were identified as intermediates in cholesterol oxidation. The structures were confirmed by 1H and 13C-NMR analyses. Aliphatic side chain hydroxylation at C26 and the A-ring modification at C3, which are putatively catalyzed by cytochrome P450 monooxygenase CYP125 and cholesterol oxidase, respectively, occur simultaneously in the strain and are followed by cascade reactions of aliphatic sidechain degradation and steroid core destruction via the known 9(10)-seco-pathway. The genes putatively related to the sterol and bile acid degradation pathways form three major clusters in the S. hirsuta genome. The sets of the genes include the orthologs of those involved in steroid catabolism in Mycobacterium tuberculosis H37Rv and Rhodococcus jostii RHA1 and related actinobacteria. Bioinformatics analysis of 52 publicly available genomes of thermophilic bacteria revealed only seven candidate strains that possess the key genes related to the 9(10)-seco pathway of steroid degradation, thus demonstrating that the ability to degrade steroids is not widespread among thermophilic bacteria.


Introduction
Steroids are abundant biomolecules in various environments and growth substrates for diverse bacteria. Sterols (e.g., cholesterol, ergosterol, and phytosterols) are steroid 3βalcohols with an alkyl side chain consisting of 8-10 carbon atoms. Structurally, bile acids differ from sterols by cis-A/B-ring juncture, α-orientation of hydroxyl at C3, a saturated steroid core, and a C5 acyl side chain. Due to the unique lipophilic/amphiphilic properties, steroidal compounds play vital functions in all living organisms. Annually large amounts of sterols, bile acids, and other steroids enter into the environment via the decay of biomass or excretion by humans and animals and as industrial wastes of steroid production plants.

Mass-Spectrometry (MS), 1 H-and 13 C-Nuclear Magnetic Resonance Spectroscopy ( 1 H-and 13 C-NMR Spectroscopy)
MS spectra of compounds II, III, and IV were recorded on a tandem mass spectrometer LCQ Advantage MAX (Thermo Finnigan, Waltham, MA, USA) in the positive ion [M + H] + mode at an evaporator temperature of 350 • C and capillary temperature of 170 • C. MS/MS spectra were obtained using normalized collision energy (Normolized Collision Energy TM ) ranging from 20% to 40%. Data were collected and processed using the Xcalibur software. HRMS experiments for compounds V, VI, VII, and VIII were performed with an Orbitrap Elite mass spectrometer (Thermo Fisher Scientific GmbH, Bremen, Germany) with an ESI source. 1 H-and 13 C-NMR spectra were recorded at 400 and 100. 6 MHz, respectively, with a Bruker Avance 400 spectrometer. Chemical shifts were measured relative to the solvent signal. Only characteristic signals are given in 1 H-NMR of steroids.

Genome Analysis
Annotation of the genome was carried out using NCBI PGAP [28], RAST (http://rast. nmpdr.org/, accessed on 10 September 2019) [29,30] and KAAS (https://www.genome.jp/ tools/kaas/, accessed on 10 September 2019) [31]. Orthologous and paralogous relations between genes of the S. hirsuta VKM Ac-666 T , Mycobacterium tuberculosis H37Rv and Rhodococcus jostii RHA1 genomes were found using OrthoFinder 2.5.1 [32,33] with inflation parameter 1.5. A BLAST search [34] against non-redundant protein sequences (NCBI database) was used as an additional tool to confirm the predetermined enzyme function. Reciprocal BLAST was used in several cases to search for the genes that correspond to the known steroid catabolism genes one-to-one.

Phylogenetic Analysis
A phylogenetic dendrogram showing the relationships of KstD homologs was constructed by the maximum likelihood algorithm in MEGA7 [35]; the sequences were aligned with MUSCLE. Default parameters were used in all cases.

BLAST Search for Steroid Catabolism Genes
Search for the key genes of the steroid catabolic 9,10-seco-pathway (kstD, kshA, and kshB) was carried out against several dozen available genomes of thermophilic strains, using the BLAST+ program [36]. The protein sequences of KstD (NP_218054.1), KshA (NP_218043.1), and KshB (NP_218088.1) of M. tuberculosis H37Rv were used as reference ones. A list of bacteria to be screened (Supplementary Table S1) was compiled on the basis of the literature data [37] on thermophilic and thermotolerant actinobacteria with known complete genome sequences or annotated contigs and available sources on other known thermophilic bacteria of diverse phylogenetic positions.
The genomes of Geobacillus kaustophilus and Parageobacillus thermoglucosidasius strains capable of performing some modifications of steroid compounds were screened for the steroid catabolism genes (Supplementary Table S2) using the BLAST+ program [36].
No other steroids without a lateral chain (C 19 -steroids) or a partially oxidized side chain (C 22 -or C 24 -steroids) were detected among the intermediates. Based on the structures and the time courses of the steroids detected, the following scheme was proposed for cholesterol bioconversion with S. hirsuta VKM Ac-666 T (Figure 3).
No other steroids without a lateral chain (C19-steroids) or a partially oxidized side chain (C22or C24-steroids) were detected among the intermediates. Based on the structures and the time courses of the steroids detected, the following scheme was proposed for cholesterol bioconversion with S. hirsuta VKM Ac-666 T (Figure 3).

High-Performance Liquid Chromatography (HPLC), Mass-Spectrometry (MS), 1 H-and 13 C-Nuclear Magnetic Resonance
Spectroscopy (                 Among the lithocholic acid bioconversion intermediates, the compounds with both the unmodified A-ring structure and the 3-keto-4-ene moiety were found (Supplementary Figure S31A,B).
In Ac-666 T , clusters 2 and 3 ( Figure 4, Supplementary Tables S3 and S4) contain candidate genes related to the cholate degradation pathway, namely, orthologs of the kshA and kshB subunit genes; two orthologs of kstDs: kstD2 and kstD1; the A/B-ring opening operon hsaEGF and orthologs of hsaD3 and hsaB3; the ksdI steroid delta-isomerase gene; kstR3 for a predicted transcriptional regulator; and orthologs of the casACEHI genes, which determine degradation of the cholate side chain. Cluster 1 (Figure 4, Supplementary Tables S3 and S4) contains candidate genes related to a sterol side chain degradation pathway, A/B-ring oxidation, and the Mce4 system (operon mceABCDEF and the genes coding for two permease subunits YrbEa and YrbEb). In total, four mce loci (F1721_29585-F1721_29620, F1721_32550-F1721_32585, F1721_10830-F1721_10865, F1721_13950-F1721_13915) were found in S. hirsuta. The choD, choE, and fadD3 genes, presumably encoding cholesterol oxidases and HIP-CoA synthetase, respectively, were found out of the clusters in Ac-666 T (Figure 4, Supplementary Tables S3 and S4).
In Ac-666 T , clusters 2 and 3 ( Figure 4, Supplementary Tables S3 and S4) contain candidate genes related to the cholate degradation pathway, namely, orthologs of the kshA and kshB subunit genes; two orthologs of kstDs: kstD2 and kstD1; the A/B-ring opening operon hsaEGF and orthologs of hsaD3 and hsaB3; the ksdI steroid delta-isomerase gene; kstR3 for a predicted transcriptional regulator; and orthologs of the casACEHI genes, which determine degradation of the cholate side chain. Figure 5 shows the scheme proposed for cholesterol bioconversion with the participation of the candidate genes of S. hirsuta VKM Ac-666 T .
Microorganisms 2021, 9, x FOR PEER REVIEW 10 of 19 Figure 5 shows the scheme proposed for cholesterol bioconversion with the participation of the candidate genes of S. hirsuta VKM Ac-666 T .

BLAST Search for the Key Enzymes of Steroid Catabolism in 52 Thermophilic/ Thermotolerant Strains
The key steroid catabolism enzymes KstD, KshA, and KshB of M. tuberculosis H37Rv were used as reference enzymes in a BLAST search carried out against several dozen publicly available genomes of thermophilic bacteria of different phylogenetic positions (Supplementary Table S1).

Discussion
Several thermophilic bacterial species have been reported to carry out distinct structural modifications of steroids [19][20][21]47], while sterol degradation by thermophilic microorganisms has not been studied so far. As shown in this research, thermophilic S. hirsuta transformed cholesterol (Figure 1). The cholesterol degradation pathway was predicted (Figure 4) based on the time courses of the intermediates (Figure 1) and bioinformatics analysis (Figure 4). The set and the order of the genes putatively involved in steroid catabolism in S. hirsuta are similar to the clusters described for M. tuberculosis H37Rv and R. jostii RHA1 [5] (Figure 4).
ChOs are most likely involved in 3β-ol-5-ene-moiety modification in S. hirsuta since no candidate genes coding for 3β-HSDs were found in Ac-666 T [25]. Two candidate cho genes, choD F1721_14655 and choE F1721_09795, were revealed in Ac-666 T . Similar to other cho in actinobacteria [51], both genes are out of the steroid catabolism clusters.
The phylogenetic analysis of acyl-CoA synthetases revealed four different types of acyl-CoA synthetases from R. jostii RHA1 and M. tuberculosis H37Rv, which are specific to the chain length of steroids [54]. FadD19 from M. tuberculosis H37Rv activates cholesterol metabolites with the C8-side chain, whilst FadD17 from H37Rv acts in the case of the C5-or longer side chains; and CasG from R. jostii RHA1, in the case of the cholate C5-side chain. Metabolites with the C3-side chain are activated by the steroid-22-oyl-CoA synthetase CasI during cholate oxidation by R. jostii RHA1 [54]. Orthologs of fadD19 (F1721_32635), fadD17 (F1721_32615), casG (F1721_02405), and casI (F1721_28770), which encode acyl-coenzyme A synthases, were revealed in S. hirsuta (Supplementary Tables S3 and S4). Probably, the presence of the homologous genes encoding various acyl-coenzyme A synthases in Ac-666 T contributes to the adaptation of the thermophilic microorganism in nature.
As shown for R. rhodochrous RG32, decomposition of the sterol C24-branched side chain is mediated by aldol lyases encoded by ltp3 and ltp4 [55]. The candidate genes ltp3 (F1721_32665) and ltp4 (F1721_32660) putatively involved in degrading sterols with branched side chains were identified in S. hirsuta (Supplementary Tables S3 and S4).
The role of thiolase FadA5 in the last cycle of cholesterol side chain β-oxidation has been demonstrated for M. tuberculosis H37Rv [58]. Orthologous fadA5 (F1721_32685) is present in S. hirsuta (Supplementary Tables S3 and S4).
The phylogenetic dendrogram with the KstD homologs demonstrates that KstD2 from S. hirsuta is in close identity with KstD2 from N. simplex (= Pimelobacter simplex) (AIY19529.1) ( Figure 6). KstD from M. tuberculosis is in the same clade with KstD3 from S. hirsuta, while KstD1 from S. hirsuta is more similar to the corresponding enzymes of N. simplex ( Figure 6). 9α-Hydroxylation is carried out by 3-ketosteroid 9α-hydroxylase KshAB, which consists of an oxygenase component (KshA) and a reductase component (KshB) [66]. Five different paralogous genes have been reported to encode the KshA subunits in Mycolicibacterium fortuitum VKM Ac-1817D (=Mycobacterium sp. VKM Ac-1817D) [61], thus providing for 9α-hydroxylation of steroid metabolites at various stages of sitosterol catabolism [67]. Several KshAs with different substrate specificities have similarly been found in R. rhodochrous DSM 43269: KshA1 was shown to participate only in the cholic acid catabolism, while KshA5 could hydroxylate several substrates [68]. Two kshA orthologs (F1721_32745 and F1721_00725) and two kshB orthologs (F1721_32755 and F1721_00735) were revealed in S. hirsuta (Figure 4, Supplementary Tables S3 and S4). Most likely, these two KshABs might differ on their substrate specificity in Ac-666 T .
It should be noted that no C 19 -steroid intermediates, such as androstenedione, androstadienedione, testosterone, or 1(2)-dehydrotestosterone, were detected during the cholesterol transformation by S. hirsuta. This could be explained either by their rapid degradation to concentrations below the detection level, or by disruption of the A/B-rings in intermediates with a preserved side chain. For instance, 9,10-seco-steroid intermediates with partially degraded side chains form during bile acid transformation with Rhodococcus strains, evidencing that side chain degradation and B-ring opening occur simultaneously [69,70]. 9α-Hydroxylation is carried out by 3-ketosteroid 9α-hydroxylase KshAB, which consists of an oxygenase component (KshA) and a reductase component (KshB) [66]. Five different paralogous genes have been reported to encode the KshA subunits in Mycolicibacterium fortuitum VKM Ac-1817D (=Mycobacterium sp. VKM Ac-1817D) [61], thus providing for 9α-hydroxylation of steroid metabolites at various stages of sitosterol catabolism [67]. Several KshAs with different substrate specificities have similarly been found in R. rhodochrous DSM 43269: KshA1 was shown to participate only in the cholic acid catabolism, while KshA5 could hydroxylate several substrates [68]. Two kshA orthologs (F1721_32745 and F1721_00725) and two kshB orthologs (F1721_32755 and F1721_00735) were revealed in S. hirsuta (Figure 4, Supplementary Tables S3 and S4). Most likely, these two KshABs might differ on their substrate specificity in Ac-666 T .
It should be noted that no C19-steroid intermediates, such as androstenedione, androstadienedione, testosterone, or 1(2)-dehydrotestosterone, were detected during the cholesterol transformation by S. hirsuta. This could be explained either by their rapid degradation to concentrations below the detection level, or by disruption of the A/B-rings in intermediates with a preserved side chain. For instance, 9,10-seco-steroid intermediates with partially degraded side chains form during bile acid transformation with Rhodococcus strains, evidencing that side chain degradation and B-ring opening occur simultaneously [69,70].
Degradation of the C/D-rings begins with the action of FadD3, whose physiological role has been studied in M. tuberculosis [41]. Unlike in M. tuberculosis H37Rv and R. jostii RHA1, in which fadD3 encoding HIP-CoA synthetase lies in the corresponding cluster, the ortholog of fadD3 is out of the clusters in S. hirsuta (Figure 4). IpdE1(FadE30) and IpdE2 (FadE33) of M. tuberculosis have been shown to form a complex that catalyze the dehydrogenation of 5-OH-HIP-CoA to 5-OH-HIPE-CoA [44].

Search for the Key Genes of Steroid Catabolism in the Genomes of Thermophilic/ Thermotolerant Bacteria
In order to find out whether steroid degraders are widespread among thermophilic bacteria, a BLAST search for the kstD and kshAB key genes of the steroid catabolic 9,10-secopathway was performed using 52 publicly available genomes of thermophilic/ thermotolerant strains (Supplementary Table S1). Only seven actinobacterial strains were identified as putative steroid degraders (Supplementary Table S5). The other thermophilic/thermotolerant strains do not contain enzymes similar to KstD and KshAB of M. tuberculosis H37Rv by more than 35% and, most likely, do not degrade steroids.
The thermophilic G. kaustophilus and P. thermoglucosidasius strains have been reported to provide separate reactions of steroid modification [19,20]. The BLAST search for more than 20 steroid catabolism enzymes (Supplementary Table S2) in these bacteria discovered the putative proteins that are 47% and 45% similar to the reference FadA5, respectively, and the P. thermoglucosidasius enzymes that are similar to HsaF and HsaE of M. tuberculosis H37Rv by 48% and 41%, respectively (Supplementary Table S6). FadA5 is known additionally to be involved in fatty acid β-oxidation; thus, the corresponding proteins of G. kaustophilus and P. thermoglucosidasius may not be intended for steroid catabolism. HsaEF participate in oxidation of the hydroxydiene derivative of hexanoic acid, meaning that similar enzymes do not necessarily participate in the catabolism of steroid compounds.

Conclusions
The thermophilic strain Saccharopolyspora hirsuta VKM Ac-666 T is capable of transforming various steroids [23,24]. As confirmed in this study, the strain efficiently transforms cholesterol and 26-alcohols with both 3β-ol-5-ene and 3-keto-4-ene A-ring structures being key intermediates. The genes related to sterol metabolism and cholic acid catabolism were for the first time identified in the genome of this thermophilic strain. The organization of the steroid catabolism genes is generally similar to that in other actinobacteria, with some differences related to individual genes and their grouping. Future transcriptomic and proteomic studies are of significance for a clearer understanding of the peculiarities of steroid catabolism in thermophilic actinobacteria.
The presence of key enzymes responsible for steroid core disruption was identified only in seven of 52 thermophilic bacteria of various phylogenetic positions, thus suggesting that steroid-degrading activity is not common in the thermophilic species.
The results contribute to the knowledge on the diversity of microbial steroid degraders and the features of steroid catabolism by thermophilic actinobacteria and could be useful for application in pharmaceutical and environmental steroid biotechnology.

Conflicts of Interest:
The authors declare no conflict of interest in this work.