Insight into Different Stages of Steroid Degradation in Thermophilic Saccharopolyspora hirsuta VKM Ac-666T Strain

Steroids are abundant molecules in nature, and various microorganisms evolved to utilize steroids. Thermophilic actinobacteria play an important role in such processes. However, very few thermophiles have so far been reported capable of degrading or modifying natural sterols. Recently, genes putatively involved in the sterol catabolic pathway have been revealed in the moderately thermophilic actinobacterium Saccharopolyspora hirsuta VKM Ac-666T, but peculiarities of strain activity toward sterols are still poorly understood. S. hirsuta catalyzed cholesterol bioconversion at a rate significantly inferior to that observed for mesophilic actinobacteria (mycobacteria and rhodococci). Several genes related to different stages of steroid catabolism increased their expression in response to cholesterol as was shown by transcriptomic studies and verified by RT–qPCR. Sequential activation of genes related to the initial step of cholesterol side chain oxidation (cyp125) and later steps of steroid core degradation (kstD3, kshA, ipdF, and fadE30) was demonstrated for the first time. The activation correlates with a low cholesterol conversion rate and intermediate accumulation by the strain. The transcriptomic analyses revealed that the genes involved in sterol catabolism are linked functionally, but not transcriptionally. The results contribute to the knowledge on steroid catabolism in thermophilic actinobacteria and could be used at the engineering of microbial catalysts.

The thermophilic actinobacterial genera Amycolatopsis, Microbispora, Micromonospora, Thermobifida, Thermobispora, Thermomonospora, and others have been reported to thrive on decaying organic matter [3] and are known to be beneficial in the composting process, which is characterized by their synergistic action along with bacteria and fungi. A predominance of thermotolerant actinobacteria (Streptomyces albus and Streptomyces griseus) is generally observed in thermobiotic conditions generated by preceding bacteria [4]. Moderately thermophilic actinobacteria of the Saccharopolyspora genus have been reported to play a

Results
In our previous work, the temperature dependence of bacterial growth was evaluated and a temperature of 45 • C was chosen as optimal for S. hirsuta VKM Ac-666 cultivation and cholesterol bioconversion [15]. In this study, (i) cholesterol was evaluated as the only growth substrate for the strain and (ii) culture growth and cholesterol bioconversion were evaluated in a rich medium. Due to filamentous growth, OD was difficult to use as a growth indicator, and a gravimetric method and colony forming units (CFU) calculation were used instead.

Results
In our previous work, the temperature dependence of bacterial growth was evaluated and a temperature of 45 °C was chosen as optimal for S. hirsuta VKM Ac-666 cultivation and cholesterol bioconversion [15]. In this study, (i) cholesterol was evaluated as the only growth substrate for the strain and (ii) culture growth and cholesterol bioconversion were evaluated in a rich medium. Due to filamentous growth, OD was difficult to use as a growth indicator, and a gravimetric method and colony forming units (CFU) calculation were used instead.

Cholesterol as a Growth Substrate
When cholesterol (1) was used as the only growth substrate, 26-hydroxycholesterol (2), 3β-hydroxycholest-5-en-26-oic acid (3), cholest-4-en-3-one (4), 26-hydroxycholest-4en-3-one (5), 3-oxo-cholest-4-en-26-oic acid (6), 3-oxo-cholesta-1,4-dien-26-oic acid (7), and cholesta-1,4-dien-3-one (8) were detected in the medium ( Figure 1A-C). However, poor culture growth was observed on cholesterol as the sole source of carbon and energy, as evidenced by both biomass (dry cell weight, DCW) measurements and CFU counts (Figure 2A,B).  Interestingly, some growth retardation was observed after the addition of cholesterol to the medium. After 120 h of cultivation, the biomass remained the same in both variants for the next 240 h (Figure 2A). The CFU data were consistent with the results of biomass determination by dry weight ( Figure 2B). Interestingly, some growth retardation was observed after the addition of cholesterol to the medium. After 120 h of cultivation, the biomass remained the same in both variants for the next 240 h (Figure 2A). The CFU data were consistent with the results of biomass determination by dry weight ( Figure 2B).
The S. hirsuta strain slowly utilized cholesterol in a mineral medium. The cholesterol conversion degree was more than 80%, and ∼9 mol% cholesterol remained unconverted after 336 h of transformation ( Figure 1C). Most likely, cholesterol was only partially used as a source of carbon, since another part (∼20%) of it was converted into derivatives with a preserved side chain and steroid core (compounds 2-8 and others) after 336 h ( Figure  1C).
The strain utilized cholesterol in the rich medium ∼2 times faster than in the mineral medium. Comparisons of the bioconversion rates in mineral and rich media resulted in the conclusion that bioconversion patterns were similar, but the rates were different (Figure 1C vs. Figure 3C).
The steroid derivatives formed from cholesterol during its conversion by S. hirsuta via the action of the enzymes whose genes could be inducible by cholesterol. The S. hirsuta strain slowly utilized cholesterol in a mineral medium. The cholesterol conversion degree was more than 80%, and~9 mol% cholesterol remained unconverted after 336 h of transformation ( Figure 1C). Most likely, cholesterol was only partially used as a source of carbon, since another part (~20%) of it was converted into derivatives with a preserved side chain and steroid core (compounds 2-8 and others) after 336 h ( Figure 1C).
The strain utilized cholesterol in the rich medium~2 times faster than in the mineral medium. Comparisons of the bioconversion rates in mineral and rich media resulted in the conclusion that bioconversion patterns were similar, but the rates were different ( Figure 1C vs. Figure 3C).
The steroid derivatives formed from cholesterol during its conversion by S. hirsuta via the action of the enzymes whose genes could be inducible by cholesterol.
Analysis of the S. hirsuta growth in a rich medium in the absence (control) and presence of cholesterol revealed a pattern ( Figure 4A) that was analogous to that in the mineral medium ( Figure 2A). The CFU count data was similar to the results of biomass determination by dry weight (Figure 4A,B).
The results showed insignificant influence of cholesterol on the strain growth in the mineral or rich medium. Analysis of the S. hirsuta growth in a rich medium in the absence (control) and presence of cholesterol revealed a pattern ( Figure 4A) that was analogous to that in the mineral medium ( Figure 2A). The CFU count data was similar to the results of biomass determination by dry weight (Figure 4A,B).   Analysis of the S. hirsuta growth in a rich medium in the absence (control) and presence of cholesterol revealed a pattern ( Figure 4A) that was analogous to that in the mineral medium ( Figure 2A). The CFU count data was similar to the results of biomass determination by dry weight ( Figure 4A,B).

Quantitative Reverse Transcription-PCR (RT-qPCR)
In this study, basal expression and transcriptional responses to cholesterol of five candidate genes from one large sterol catabolism gene cluster of the S. hirsuta genome ( Figure 5) were investigated depending on the time of strain growth and cholesterol exposure. Expression of the candidate sterol catabolism genes was studied in the context of cholesterol decomposition and the growth phase of the strain, which might help to understand the mechanisms of sterol degradation and ecological adaption of the moderately thermophilic S. hirsuta strain.
CFU count in the dynamics of incubation S. hirsuta in the rich medium in the presence of cholesterol (1.5 mM) (squares) or without steroids (circles). Arrows indicate the addition of cholesterol.
The results showed insignificant influence of cholesterol on the strain growth in the mineral or rich medium.

Quantitative Reverse Transcription-PCR (RT-qPCR)
In this study, basal expression and transcriptional responses to cholesterol of five candidate genes from one large sterol catabolism gene cluster of the S. hirsuta genome ( Figure 5) were investigated depending on the time of strain growth and cholesterol exposure. Expression of the candidate sterol catabolism genes was studied in the context of cholesterol decomposition and the growth phase of the strain, which might help to understand the mechanisms of sterol degradation and ecological adaption of the moderately thermophilic S. hirsuta strain. genes coding for C/D-ring degradation are in purple; genes coding for transport systems are in blue; and regulatory elements are in yellow. Five genes of interest are marked with light violet frames. The two transcriptional repressor genes kstR and kstR2 are marked with yellow frames (adopted from [15]).
The following genes were selected for the RT-qPCR expression analysis: cyp125, which is associated with the initial step in cholesterol side chain degradation [22,23]; kstD3 and kshA, which encode enzymes responsible for key degradation steps of the steroid core [24,25]; and fadE30 and ipdF, which are associated with the C/D ring degradation pathway [26][27][28]. Cyp125 is responsible for C26-hydroxylation of the terminal methyl group of cholesterol/cholestenone; KstDs and KshAs are the key enzymes for the ring B opening in the so-called 9(10)-seco pathway; IpdF is responsible for reduction of the C5-oxo-group to hydroxyl of the KstR2 inducer molecule HIP-CoA [1]; and FadE30 is involved in dehydrogenation of 5-OH-HIP-CoA [29] ( Figure 5). The 16S rRNA gene was used as a reference gene.
First of all, in order to determine whether the genes of interest (cyp125, kstD3, kshA, ipdF, and fadE30) function in the conditions under study, we analyzed the contents of their mRNAs in S. hirsuta cell samples taken at 0, 24, 72, and 120 h of incubation with cholesterol in the rich medium. The mRNAs of all five genes were detected in all samples, and their largest amounts were observed in the 72 h sample ( Figure 6). Subsequently, changes in mRNA amount were evaluated only for 24-and 72-h samples in the rich medium. genes coding for C/D-ring degradation are in purple; genes coding for transport systems are in blue; and regulatory elements are in yellow. Five genes of interest are marked with light violet frames. The two transcriptional repressor genes kstR and kstR2 are marked with yellow frames (adopted from [15]).
The following genes were selected for the RT-qPCR expression analysis: cyp125, which is associated with the initial step in cholesterol side chain degradation [22,23]; kstD3 and kshA, which encode enzymes responsible for key degradation steps of the steroid core [24,25]; and fadE30 and ipdF, which are associated with the C/D ring degradation pathway [26][27][28]. Cyp125 is responsible for C26-hydroxylation of the terminal methyl group of cholesterol/cholestenone; KstDs and KshAs are the key enzymes for the ring B opening in the so-called 9(10)-seco pathway; IpdF is responsible for reduction of the C5-oxo-group to hydroxyl of the KstR2 inducer molecule HIP-CoA [1]; and FadE30 is involved in dehydrogenation of 5-OH-HIP-CoA [29] ( Figure 5). The 16S rRNA gene was used as a reference gene.
First of all, in order to determine whether the genes of interest (cyp125, kstD3, kshA, ipdF, and fadE30) function in the conditions under study, we analyzed the contents of their mRNAs in S. hirsuta cell samples taken at 0, 24, 72, and 120 h of incubation with cholesterol in the rich medium. The mRNAs of all five genes were detected in all samples, and their largest amounts were observed in the 72 h sample ( Figure 6). Subsequently, changes in mRNA amount were evaluated only for 24-and 72-h samples in the rich medium.
As follows from the data presented for the cyp125 gene in Figures 6 and 7, the relative mRNA amount increased 4.12 and 43.9 times in S. hirsuta cells cultivated in the presence of cholesterol for 24 h and 72 h, respectively.
For the genes kstD3, kshA, ipdF, and fadE30, no change in mRNA content was observed in S. hirsuta cells cultivated in the presence of cholesterol for 24 h. However, the amounts of their mRNAs increased statistically significantly after 72 h cholesterol exposure (Figure 7).
Upregulation levels of gene expression with cholesterol induction varied greatly among the genes of interest. Thus, there was only a 10.6-fold increase in expression of cyp125 from 24 h to 72 h of cholesterol exposure. However, there were a 21.3-fold increase in kshA and 64.2-, 82-, and 169.3-fold increases in kstD3, ipdF, and fadE30 expression, respectively, from 24 h to 72 h of cholesterol exposure. Based on these data, it can be assumed that the cyp125 gene is more sensitive to cholesterol exposure during the initial 24 h than the other four genes of interest. Expression of kshA, kstD3, ipdF, and fadE30 is greater after 72 h of cholesterol exposure.      Interestingly, the relative mRNA amount decreased for all genes under study throughout the subsequent cholesterol exposure from 72 to 120 h, but not to the same extent ( Figure 6). The lowest relative mRNA amount corresponded to ipdF, and the greatest, to kstD.

High-Throughput mRNA Sequencing
An accumulation of two main cholesterol derivatives, 26-hydroxycholesterol and cholest-4-en-3-one, after 24 h of strain incubation with cholesterol in the rich medium ( Figure 3C) indicated that the S. hirsuta enzymes that simultaneously and independently carry out 26-hydroxylation and transformation of the 3β-OH-5-ene-into the 3-oxo-4-enefragment of the cholesterol molecule, respectively, could be induced by cholesterol or one or more of its derivatives. It is known that cholesterol degradation in actinobacteria is regulated by transcriptional repressors of the TetR type, KstR and KstR2 [30,31]. Since 3-oxocholest-4-en-26-oic acid and its derivatives may regulate the transcription factor KstR [32] in actinobacteria, the finding of the structurally similar compounds 26-hydroxycholesterol and cholest-4-en-3-one among cholesterol degradation intermediates of the actinobacterial thermophile allowed us to suggest that comparative transcriptome profiling at 24 h of cholesterol exposure would shed light on the induction process in S. hirsuta. Given that oxidative degradation of cholesterol proceeds rapidly in some mesophilic actinobacteria and that, in contrast, a rather long lag period is observed in the process of cholesterol oxidation by the thermophilic S. hirsuta strain, we planned to record a clear picture of the sequential induction of genes related to the sterol degradation pathway at its early steps.
Based on these considerations, we decided to conduct whole-transcriptome analysis of S. hirsuta cells grown in the presence or absence of cholesterol for 24 h in the rich medium. The total number of reads, the percentage of reads mapping to the rRNA genes, and links to the SRA repository in NCBI for experimental variants are summarized in Table 1. To identify the genes upregulated in response to cholesterol, differential gene expression was calculated in pairs of control and corresponding cholesterol-exposed samples. The total number of genes that increased in expression is shown in Table 2.  Table S1) [15].
Activities of other enzymes are most likely not associated with degradation of sterols. They could code for housekeeping proteins involved in regulating a variety of biological processes. Possible examples include transport and various signaling (F1721_03575); lipid export to the outer membrane through the inner membrane using ATP (F1721_08445); export of lipid components through the cell membrane (F1721_04135); pleiotropic regulation of carbon catabolite repression (F1721_06445); transport coordination of genes expression necessary for adaptation (F1721_08455); hydrolysis of a wide range of substrates bearing amide or ester functional groups (F1721_08470); control of gene expression by regulating the initiation or extension of transcription (F1721_08460); the ATP-binding cassette (ABC) transporters (F1721_08450); mediation of protein-protein interactions and assembly of multiprotein complexes (F1721_15540); hydrolysis of a wide range of substrates, such as beta-lactams, DNA or RNA (F1721_15545); 1 4-condensation between 5-carbon isoprene units (F1721_22945); methylation of proteins, small molecules, lipids, and nucleic acids using SAM as a cofactor (F1721_27890); metal ions resistance (F1721_28715); and a role of carrier proteins (F1721_06460).
The F1721_15555 protein could be mentioned as potentially involved in metabolism of steroids (Table 2). A 2.9-fold increase was observed in expression of the F1721_32655 gene in response to cholesterol. Other actinobacteria bearing steroid catabolism gene clusters have F1721_32655 homologs: Rv3521 (in M. tuberculosis H37Rv) and RHA1_RS22865 (in R. jostii RHA1) [15], which are close to the aldol lyase gene ltp4 in the corresponding genomes. The products encoded by the Rv3521 and RHA1_RS22865 genes are known to be hypothetical conserved proteins. The product of the S. hirsuta F1721_32655 gene potentially could bind to DNA, like KstR.
Known microbial sterol degradation begins with C26-hydroxylation of the sterol alkyl side chain and/or modification of the A ring of the steroid core by converting 3β-hydroxy-5-ene-into a 3-oxo-4-ene-structure [1,42]. These two processes can occur simultaneously. Indeed, independent and simultaneous conversion of cholesterol into C26-hydroxylated 3βhydroxy-5-ene-and 3-oxo-4-ene-corresponding derivatives has been clarified previously in the case of S. hirsuta [15].
As shown in this study, cholesterol is a poor growth substrate for S. hirsuta. The strain transforms cholesterol to form corresponding derivatives with the preserved side chain, such as cholest-4-en-3-one or 26-hydroxylated steroids. Much higher strain growth and cholesterol conversion/degradation rates were observed when a rich medium was used instead of a mineral one.
The fact that cholesterol bioconversion proceeds rapidly in mesophilic actinobacteria and slowly in a thermophilic actinobacterial strain prompted us to study the transcriptomic profiles of S. hirsuta cells grown with and without cholesterol in the GSMY medium at an early stage of sterol bioconversion.
The apparently longer duration of cholesterol bioconversion by S. hirsuta cells compared to mesophilic actinobacteria makes it possible to identify the early intermediates (or products) of cholesterol bioconversion. Surprisingly, whole transcriptome profiling revealed that, with the exception of cyp125, the vast majority of the genes previously described as cholesterol inducible did not increase their expression in response to cholesterol.
The choE gene (F1721_09795) is outside the sterol catabolism clusters in the S. hirsuta genome [15] and was not upregulated in response to cholesterol, though expression of its N. simplex counterpart (KR76_09550) [49,50] has been reported to increase 10-fold and 13-fold in response to phytosterol and cholesterol, respectively [37,47]. Notably, the choE gene is outside the steroid catabolism clusters and possesses a putative KstR binding site in N. simplex [51]. It is also of interest that the important gene coding for 3β-HSD (Rv1106c) does not appear to be induced by cholesterol in M. tuberculosis H37Rv [43,52]. It is possible that the formation of 3-keto-4-ene steroids by the thermophilic S. hirsuta strain is due to high basal expression of another important enzyme, cholesterol oxidase encoded by choE (F1721_09795), because no gene coding for 3β-HSD has been revealed in its genome.
Despite the recorded formation of cholest-4-en-3-one and 26-hydroxycholesterol from cholesterol, no increase in the expression of kstR, choE, and or mce4 genes was detected in response to cholesterol exposure of S. hirsuta cells for 24 h. Probably, the mce4 genes could not be induced at the first stage of the metabolic response in S. hirsuta, but further incubation would increase the expression of the transporter genes.
It is possible to conclude that, in the absence of the induction of the genes mce4 and kstR upon exposure to cholesterol for 24 h, the increase in expression of the cyp125 gene is an important adaptive mechanism in the sterol-decomposing moderately thermophilic S. hirsuta strain.
Analysis of the gene arrangement in the S. hirsuta genome revealed that cyp125 is not a part of the fadA5 operon, based on the fact that cyp125 and fadA5 are on the opposite DNA strands ("head to head"). The cyp125 gene could have its own promoter and, possibly, its own regulation. Thus, the genes of the sterol catabolism cluster are linked functionally, but not transcriptionally in the S. hirsuta genome.
In this work, activities of genes related to different stages of sterol catabolism in the thermophilic S. hirsuta strain was studied for the first time by RT-qPCR during the process of cholesterol bioconversion. Differential expression of five genes of interest: cyp125 (F1721_32680), kstD3 (F1721_32740), kshA (F1721_32745), ipdF (F1721_33710), and fadE30 (F1721_33715) was observed in the stationary phase of cell growth in the GSMY medium following 72 h of cholesterol exposure.
The kstD, kshA, and kshB genes encode the corresponding KstD, KshA, and KshB enzymes involved in aerobic degradation of the steroid core in the so-called 9(10)-seco pathway, resulting in the formation of the chemically unsTable 1,4-diene-9α-hydroxy compound [1,6] (Figure 9). The location of kstD3 (F1721_32740) and kshA (F1721_32745) side by side on the same DNA strand in the S. hirsuta genome ( Figure 5) and the efficient mRNA synthesis in response to cholesterol make it possible to assume that the genes could be linked transcriptionally.
IpdF is an acyl-CoA dehydrogenase encoded by the ipdF gene. It is believed that the main function of this enzyme is reduction of the keto group to hydroxyl at C5 of HIP-CoA [1]. The FadE30 enzyme has been implicated in dehydrogenation of 5-OH-HIP-CoA and thus plays a significant role in steroid catabolism at the level of degradation of ethylhexahydroindanone propionate [29].
Expression of ipdF (F1721_33710) and fadE30 (F1721_33715) in S. hirsuta cells was very high in response to cholesterol on evidence of RT-qPCR. Moreover, these genes are located one after the other on the same DNA strand ( Figure 5).
Overall, the expression pattern of the sterol catabolism genes in the thermophilic S. hirsuta strain is similar to known examples of sterol catabolic gene expression in mesophilic actinobacteria. A significant increase in expression of the kstD3 (F1721_32740) and kshA (F1721_32745) genes implicates the 9(10)-seco pathway in cholesterol degradation in S. hirsuta cells. Further destruction of the steroid core and the C and D rings is most likely due to the participation of the products of the ipdF (F1721_33710) and fadE30 (F1721_33715) genes, which also showed a high level of expression in response to cholesterol. Past precedents for bioconversion of sterols have also confirmed that ring B opening can occur in the corresponding steroid intermediates, both containing and not containing an aliphatic side chain [53,54].

Microorganism, Cultivation, and Cholesterol Conversion
The strain of Saccharopolyspora hirsuta VKM Ac-666 T was obtained from the All-Russian Collection of Microorganisms (VKM).

Computational Analyses
Read quality was controlled using FastQC [56]. Adapter sequences and low quality regions in raw reads were removed using Trimmomatic program version 0.39 [57]. A mapping of prepared reads to the S. hirsuta VKM Ac-666 T genome obtained previously (accession number DDBJ/ENA/GenBank VWPH01000000) [14] was carried out using the Bowtie2 program version 2.4.5 [58]; mapped reads were counted using the featureCounts program version 1.6.4 [59]. The DESeq package version 1.34.1 was used to evaluate the differential gene expression [60]. A gene was considered to have changed the expression level if padj was >0.05.

Preparation of Total RNA for RT-qPCR
A S. hirsuta culture was grown in the GSMY medium as described in 4.2 Microorganism, cultivation, and cholesterol conversion. Aliquots (1 mL) of cultures grown with and without cholesterol were withdrawn at 0, 24, 72, and 120 h of cholesterol addition and centrifuged at 7000 g, 4 • C for 30 min. Total RNA was isolated from 40 mg of raw biomass pretreated with 5% phenol in ethyl alcohol. The biomass was resuspended in 400 µL of the TE buffer (pH 8.0) supplemented with 20 mg/mL lysozyme (AppliChem GmbH, Darmstadt, Germany) and 350 µg/mL proteinase K (Thermo Fisher Scientific, Waltham, MA, USA), kept at 37 • C for 10 min, and then combined with 50 mg of glass beads (~100 mesh). After four cycles of shaking at 3000 rpm for 2 min and incubation for 10 min at 37 • C, a lysis solution from the Aurum total RNA mini kit (Bio-Rad, Irvine, CA, USA) supplemented with polyvinylpyrrolidone 40,000 (AppliChem GmbH, Darmstadt, Germany) was added to a final concentration of 6%. The cells were kept at room temperature with continuous stirring for 30 min and then centrifuged at 13,400 rpm for 10 min. The resulting supernatant was mixed with 70% isopropanol as recommended by the manufacturer of the Aurum total RNA mini kit (Bio-Rad, Irvine, CA, USA). Further isolation of total RNA was carried out in accordance with the manufacturer's protocol. The RNA concentration was determined spectrophotometrically. The quality of preparations was assessed by agarose gel electrophoresis.

RT-qPCR
Specific primers were constructed using the Primer-BLAST tool (Table 3). Table 3. Genes of S. hirsuta whose expression was increased in response to induction with cholesterol for 24 h.

Gene
Primer Sequence Amplicon Size E R2  (2)(3)(4). For each pair of primers, the amplification efficiency was determined from the slope of the log-linear portion of the calibration curve (using a series of ten-fold dilutions of the template DNA). The reaction specificity was confirmed by agarose gel electrophoresis.

Genes of interest
Total RNA preparations were treated with DNase I (New England Biolabs, Ipswich, MA, USA). A RevertAid RT Reverse Transcription kit (Thermo Fisher Scientific, Clontech Laboratories, Inc., Mountainview, CA, USA) and random hexamer primers were used to synthesize the first cDNA strand. The total RNA preparation (100 ng) was added to the reaction. Genomic DNA (gDNA) contamination was quantified (1/Eˆ(Cp«-RT»-Cp«+RT»)) and did not exceed 3%. According to BestKeeper version 1 [61], the most stable was the 16S rRNA gene (SD 0.12, CV 1.7%).

RT-qPCR Data Analysis
Data to construct the heat map were obtained using the average of two biological replicates. Experiments to determine the relative mRNA amount (differential gene expression) were performed in six biological replicates. The relative mRNA amount was determined using the Pfaffl formula [61]. RStudio Desktop 2022.02.1 + 461 was used for statistical data analyses. The Shapiro-Wilk test was used to check the normality of the data. Data comparison was performed using Student's unpaired t-test.

Growth Estimations
Cultures of S. hirsuta were grown in the GSMY or mineral medium as described in 4.2 Microorganism, cultivation, and cholesterol conversion. The S. hirsuta biomass grown in the presence or absence of cholesterol in the GSMY or mineral medium was assayed by CFU and dry weight. For a CFU count, serial dilutions were plated on Petri dishes with the agar GSMY medium and CFUs were subsequently counted after 48 h of growth. For dry weight measurements, samples (45 mL) of the cultivation broth were centrifuged at 5600× g for 40 min and the cakes were washed twice with 45 mL of 10% (w/v) aqueous MCD to remove cholesterol and then twice with 45 mL of distilled water. The washed cells were dried at 85 • C [62]. The growth experiments were carried out in three replicates.
Retention times (

Conclusions
The thermophilic actinobacterium S. hirsuta utilizes cholesterol as the only carbon and energy source with a very low rate. Using a complex organic medium is favorable for cholesterol oxidation by the strain. The overall pattern of C26 steroid degradation observed during cholesterol conversion by the organism was similar to that reported for mesophilic actinobacteria. The genes involved in different stages of steroid degradation 9(10)-seco-pathway were upregulated sequentially in response to cholesterol in S. hirsuta and were not transcriptionally linked. To the best of our knowledge, this is the first report on the molecular mechanism of steroid degradation in thermophilic actinobacteria.

Conflicts of Interest:
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.