The present invention relates to genetically-modified bacteria and the use of such bacteria in the bioconversion of steroidal substrates into steroidal compounds of Interest. The genetically-modified bacteria may be from the genera Rhodococcus or Mycobacterium.

Steroids are a large and diverse class of organic compounds, with many essential functions in eukaryotic organisms. For example, naturally occurring steroids are involved in maintaining cell membrane fluidity, controlling functions of the male and female reproductive systems and modulating Inflammation.

As signalling through steroid controlled pathways is important In a wide variety of processes, the ability to modulate these pathways using synthetically produced steroid drugs means they are an Important class of pharmaceuticals. For example, corticosteroids are used as anti-inflammatories for the treatment of conditions such as asthma and rheumatoid arthritis, synthetic steroid hormones are widely used as hormonal contraceptives and anabolic steroids can be used to increase muscle mass and athletic performance. The synthesis of steroids for use as pharmaceuticals involves either semi-synthesis from natural sterol precursors or total synthesis from simpler organic molecules. Semi- synthesis from sterol precursors such as cholesterol often involves the use of bacteria. The advantages of using bacteria to carry out these bioconversions are that the synthesis Involves less steps and the reactions performed by the enzymes are stereospecific, resulting in the production of the desired isomers without the need for protection and deprotection used In traditional chemical synthesis. The products of bacterial bioconversions can then be used as pharmaceuticals or as precursors for further chemical modification to produce the compound of interest. Steroids naturally occur in both plant, animal and fungal species, and are produced by certain species of bacteria. Despite them only occurring naturally in only a few bacterial species, several bacterial species are able to metabolise sterol compounds as growth substrates. Examples of bacteria that can degrade sterol compounds include those from the genera Rhodococcus and Mycobacterium.

The bacterial sterol metabolism pathway involves progressive oxidation of the sterol side- chain, and breakdown of tire polycyclic ring system. The pathway of sterol side-chain degradation in Rhodococcus has been previously investigated using mutant strains (Wilbrink etal, 2011. Applied and Environmental Microbiology, 77(13):4455-4464) and an overview of the cholesterol catabolic pathway is shown in Figure 1. It has now been found that bacterial species may be used for steroid compound production by genetic modification to block the degradation pathway prior to breakdown of the polycyclic ring system and at various points in side-chain oxidation to allow accumulation of the steroidal compounds of interest in order to improve the yields obtained.

In a first aspect, the invention provides a genetically-modified bacterium blocked in the steroid metabolism pathway prior to degradation of the polycydic steroid ring system, wherein the bacterium is disrupted in the steroid side-chain degradation pathway, and wherein the bacterium converts a steroidal substrate into a steroidal product of interest.

By "steroid” or "steroidal" compounds we include the meaning of a class of natural or synthetic organic compounds derived from the steroid core structure represented below (with lUPAC-approved ring lettering and atom numbering):

Steroidal compounds generally comprise four fused rings (three six-member cyclohexane rings (rings A, B and C above) and one five-member cyclopentane ring (ring D above)) but vary by the functional groups attached to that four-ring core and by the oxidation state of the rings. For example, sterols are a sub-group of steroidal compounds where one of the defining features is the presence of a hydroxy group (OH) at position 3 or the structure shown above. The structure formed by the atoms labelled 20 to 27 (including positions 24 ¹ and 24 ² ) in the above diagram is referred to as the steroid side-chain. Non-limiting examples of steroids include: sterols, 3-oxo-4-cholenic add, 3-oxo-chola-4,22-dlen-24-oic acid, 3-oxo-7-hydroxy-4-cholenlc acid, 3-oxo-9-hydroxy-4-cholenic add, 3-oxo-7,9- dihydroxy-4-cholenic acid, 3-oxo-23,24-bisnor-4-cholene-22-olc add (4-BNC), 3-oxo- 23,24-bisnor-1 ,4-choladiene-22-oic add (1 , 4-BNC), 4-androstene-3,17-dione (AD), 1 ,4- androstadiene-3, 17-dione (ADD), sex steroids (e.g. progesterone, testosterone, estradiol), corticosteroids (e.g. cortisol), neurosterolds (e.g. DHEA and allopregnanolone), and secosteroids (e.g. ergocalciferol, cholecalciferol, and calcitriol). Non-limiting examples of steroidal compounds are also shown in Figure 4. By“disrupted in the steroid side-chain degradation pathway" we indude the meaning of a bacterium in which the normal degradation of the steroid side-chain Is impaired. Normally, degradation of the steroid side-chain involves the initial cycle of side-chain activation followed by three successive cydes of b-oxidation (i.e. first second, and third cydes of 13- oxidation). In an unimpaired side-chain degradation pathway, the final product of the side- chain degradation steps is usually 4-androstene-3,17-dione (AD). Thus, a bacterium disrupted in the steroid side-chain degradation pathway will accumulate steroidal products that are upstream of the production of AD. The suggested side-chain degradation pathways of the sterols cholesterol and b-sitosterol are shown in Figure 2 and Figure 3 respectively (Wilbrink, 2011. Microbial sterol side chain degradation in Actinobacteria. Thesis).

By“polycyclic steroid ring system" we include the meaning of the ABCD system of rings found in the core steroidal structure shown above in the definition of steroidal. In some embodiments, the disruption In the steroid side-chain degradation pathway occurs after the first cycle of b-oxidation.

By“first cycle of b-oxidation” we indude the meaning of the first cyde of b-oxidation in the steroid side-chain degradation pathway (Wlpperman et a/, 2014. Ortt. Rev. Biochem. Mol. Biol., 49(4):269-293). Specifically, the first cyde of ^-oxidation is the process immediately following the side-chain activation cyde step, resulting in the shortening of the side-chain and the production of a C24 steroidal compound.

In some embodiments, the steroidal substrate may be a sterol substrate. In certain embodiments, the sterol substrate may comprise: 7-oxo-cholesterol or 7-hydroxy-p-cholesterol;

7-oxo-phytosterol; or a combination thereof.

By“sterol" we include the meaning of molecules belonging to a class of lipids which are a sub-group of steroids with a hydroxyl group at the 3-position of the A-ring. Sterols have the general structure:

Sterols may also be referred to as steroid alcohols, and occur naturally in plants (phytosterols), animals (zoosterols), and fungi, and can be also produced by some bacteria. Non-limiting examples of sterols include: b-sltosterol, T-oco-b-sitosterol, 7- hydroxy-p-sitosterol, cholesterol, 7-oxo-cholesterol, 7-hydroxy-fJ-cholesterol, campesterol, stigmasterol, fucosterol, 7-oxo-phytosterol, adosterol, atheronals, avenasterol, azacosterol, cerevisterol, colestolone, cycloartenol, 7-dehydrocholesterol, 5- dehydroeplsterol, 7-dehydrositosterol, 20a,22R-dihydroxycholesterol, dinosterol, epibrassicasterol, episterol, ergosterol, ergosterol peroxide, fecosterol, fucosterol, fUngisterol, ganoderiol, ganodermadiol, 7a-hydroxycholesterol, 22R-hydroxycholesterol, 27-hydroxycholesterol, inotodlol, lanosterol, lathosterol, lichesterol, lucldadlol, lumisterol, oxycholesterol, oxysterol, parkeol, spinasterol, trametenolic add, and zymosterol. Nonlimiting examples of sterols are also shown In Figure 5.

In some embodiments, the steroidal product of Interest comprises an intact polycyclic ring system.

By“Intact polycyclic ring system" we Include the meaning of a steroidal molecule in which the ABCD ring system of the core steroid structure Is still present, i.e. the ABCD ring system has not undergone degradation and/or oxidation such that any of the rings have been opened or removed.

In some embodiments, the steroidal product of interest Is a steroidal compound with a side-chain having a backbone of five carbons.

By“backbone” we include the meaning of the longest consecutive chain of carbon atoms in the steroid side-chain being five carbon atoms in length. Generally, the five carbons in the backbone are those at positions 20, 21 , 22, 23, and 24, as shown In the diagram of the steroid core structure In the definition of the term“steroidal" above. in certain embodiments, the steroidal product of interest may be:



wherein R can be hydroxyl or oxo;

3-0X0-23, 24-bisnor-4-cholene-22-oic acid (4-BNC);

3-0X0-23, 24-bisnor-1,4-choladiene-22-oic add (1,4-BNC); or variants thereof.

In other preferred embodiments, the steroidal product of interest may be

3-oxo-4-cholenic add;

Chola 4,22-dien-24-oic acid, 3-oxo (CAS 59648-73-6, or CAS 82637-22-7 for pure E isomer);

3-oxo-7-hydroxy-4-choienic add;

wherein R can be hydroxyl or oxo; or variants thereof.

In some embodiments, the genetically-modified bacterium may be of the Actinobacteria class or the Gammaprotaobactaria class.

In certain embodiments, a genetically modified bacterium of the Actinobacteria class may be a Rhodococcus species, a Mycobacterium species, a Nocardia species, a Corynebacterium species, or an Arthrobacter species.

Where the bacterium is of a Rhodococcus species, the Rhodococcus species may be Rhodococcus rhodochrous, Rhodococcus arythropolis, Rhodococcus jostii, Rhodococcus ruber, preferably Rhodococcus rhodochrous.

Where the bacterium Is of a Mycobacterium species, the Mycobacterium species may be Mycobacterium neoaurum, Mycobacterium smegmatis, Mycobacterium tuberculosis, or Mycobacterium fortuitum, preferably Mycobacterium neoaurum.

Where the bacterium is of a Nocardla species, the Nocardia species may be Nocardia restrictus, Nocardia coralline, or Nocardia opaca.

Where the bacterium is of a Arthrobacter species, the Arthrobacter species may be Arthrobacter simplex. in some embodiments, the genetically-modified bacterium comprises one or more genetic modifications. In certain embodiments, the genetic modification of the genetically-modified bacterium may comprise inactivation of the genes: kshA1 (SEQ ID NO: 1 ), kshA2 (SEQ ID NO: 2), kshA3 (SEQ ID NO: 3), kshA4 (SEQ ID NO: 4), kshAS (SEQ ID NO: 5), or homologs thereof. By“genetic modification” we include the meaning of an artificial alteration or addition to the genetic material present in an organism. For example, a genetic modification may be a directed deletion of a gene or genomic region, a directed mutagenesis of a gene or genomic region (e.g. a point mutation), the addition of a gene or genetic material to the genome of the organism (e.g. an integration), or, In the case of bacteria, the transformation of such ceils with plasmid.

By“homolog" we Include the meaning of a second gene or polypeptide that has a similar biological function to a first gene or polypeptide and may also have a degree of sequence similarity to the first gene or polypeptide. A homologous gene may encode a polypeptide that exhibits a degree of sequence similarity to a polypeptide encoded by the corresponding first gene. For example, a homolog may be a similar gene in a different species derived from a common ancestral gene (ortholog), or a homolog may be a second similar gene within the genome of a single species that Is derived from a gene duplication event (paralog). A homologous gene or polypeptide may have a nucleotide or amino acid sequence that varies from the nucleotide or amino add sequence of the first gene or polypeptide, but still maintains functional characteristics assodated with the first gene or polypeptide (e.g. In the case where a homologous polypeptide is an enzyme, the homologous polypeptide catalyses the same reaction as the first polypeptide). The variations that can occur in a nucleotide or amino add sequence of a homolog may be demonstrated by nudeotide or amino add differences in the overall sequence or by deletions, substitutions, insertions, inversions or additions of nudeotides or amino adds in said sequence.

In some embodiments, the genetic modification further comprises re-introduction of a wild type copy of the kshAS gene comprising SEQ ID NO: 5, or a homolog thereof.

In other embodiments, the genetic modification comprises inactivation of the genes: kshA1 (SEQ ID NO: 1), kshA2 (SEQ ID NO: 2), kshA3 (SEQ ID NO: 3), and kshA4 (SEQ ID NO: 4), or homologs thereof.

In some embodiments, the genetic modification of the genetically-modified bacterium further comprises inactivation of the genes: kstD1{SEQ ID NO: 6), kstD2 (SEQ ID NO: 7), and kstD3 (SEQ ID NO: 8), or homologs thereof. In some embodiments, the genetic modification comprises inactivation of one or more of the genes: fadE34 (SEQ ID NO: 9; SEQ ID NO: 12), tadE34#2 (SEQ ID NO: 10), or homologs thereof. In other preferred embodiments, the genetic modification of the genetically-modified bacterium further comprises inactivation of the gene: fadE26 (SEQ ID NO: 11), or homologs thereof.

In some embodiments, where the genetic modification comprises a gene inactivation, the gene activation is by gene deletion.

By "gene deletion" we indude the meaning of removal of all or substantially all of a gene or genomic region from the genome of an organism, such that the functional polypeptide produces) encoded by that gene or genomic region is no longer produced by the organism.

In certain embodiments, the homolog has a nudeotide sequence with at least 50% sequence identity with the nudeotide sequence of a first gene. In other embodiments, the homolog has a nudeotide sequence that has a sequence Identity of at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% with the nudeotide sequence of a first gene.

In some embodiments, the homolog encodes a polypeptide that has an amino add sequence with at least 50% sequence identity with the amino add sequence of a first polypeptide. The homolog encodes a polypeptide that has an amino acid sequence Identity of at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 91 %, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99%. By "sequence identity" we include the meaning of the extent to which two nucleotide or amino add sequences are similar, measured in terms of a percentage Identity. Optimal alignment is determined by comparing two optimally aligned sequences over a comparison window, where the fragment of the nudeotide or amino add sequence in the comparison window may comprise additions or deletions (e.g. gaps or overhangs) as compared to the reference sequence (which does not comprise additions or deletions) for optimal alignment of the two sequences. The percentage is calculated by determining the number of positions at which the identical nudeotide base or amino add residue occurs in both sequences to yield the number of matched positions, dividing the number of matched positions by the total number of positions in tire window of comparison and multiplying the result by 100 to yield the percentage of sequence identity. Optimal alignment of sequences for comparison may be conducted by the local homology algorithm of Smith and Waterman Add. APL Math. 2:482 (1981), by the homology alignment algorithm of Needleman and Wunsch J. Mol. Biol. 48:443 (1970), by the search for similarity method of Pearson and Lipman Proc. Natl. Acad. Sci. (USA) 85: 2444 (1988), by computerized implementations of these algorithms (GAP, BESTFIT, BLAST, PASTA, and TFASTA in the Wisconsin Genetics Software Package, Genetics Computer Group (GCG), 575 Science Dr., Madison, Wl), or by Inspection. Given that two sequences have been Identified for comparison, GAP and BESTFIT are preferably employed to determine their optimal alignment Typically, the default values of 5.00 for gap weight and 0.30 for gap weight length are used. In certain embodiments, the genetically-modified Rhodococcus rhodochrous bacterium may be of strain: LM9 (Accession No. NCIMB 43058), LM19 (Accession No. NCIMB 43059), or LM33 (Accession No. NCIMB 43080).

In certain embodiments, the genetically-modified Mycobacterium neoaurum bacterium may be of strain: NRRL B-3805 Mnao-bSadE34 (Accession No. NCIMB 43057).

In a second aspect, the invention provides a genetically-modified bacterium according to the first aspect for use in the conversion of a steroidal substrate Into a steroidal compound of interest.

In a third aspect, the invention provides a method of converting a steroidal substrate Into a steroidal product of Interest, comprising the steps of:

(a) inoculating culture medium with genetically-modified bacteria according to the first or second aspect and growing the bacterial culture until a target ODeoo is reached;

(b) adding a steroidal substrate to the bacterial culture when the target ODeoo is reached;

(c) culturing the bacterial culture so that the steroidal substrate is converted to tiie steroidal product of interest; and,

(d) extracting and/or purifying the steroidal product of interest from the bacterial culture. By "culture medium" we include the meaning of a solid, liquid, or semi-solid medium designed to support the growth of microorganisms or cells. In some embodiments, the culture medium may be Luria-Bertani (LB) medium (10g/L tryptone; 5g/L yeast extract; 10g/L NaCI) or minimal medium (4.65g/L K2HPO4; 1.5g/L NahbPOt-hfeO; 3g/L NH ₄ CI; 1g/L MgS04.7H20; 1mlZL Vishniac trace element solution).

In certain embodiments, in step (a) of the method the bacterial culture may be grown to a target ODeoo of at least 0.25, at least 0.5, at least 0.75, at least 1.0, at least 1.5, at least 2.0, at least 2.5, at least 3.0, at least 3.5, at least 4.0, at least 4.1 , at least 4.2, at least 4.3, at least 4.4, at least 4.5, at least 4.6, at least 4.7, at least 4.8, at least 4.9, or at least 5.0. Preferably, the target ODeoo may be at least 1.0, more preferably at least 4.0, yet more preferably at least 4.5, most preferably at least 5.0.

In some embodiments of the method, the steroidal substrate may be a sterol substrate. In certain embodiments, the sterol substrate may comprise:

In some embodiments of the method, the steroidal product of interest may comprise an intact polycyclic ring system. In certain embodiments, the steroidal product of interest may be a steroidal compound with a side-chain having a backbone of five carbons.

Chola 4,22-dlen-24-oic acid, 3-oxo (CAS 59648-73-6, or CAS 82637-22-7 for pure E isomer);

3-oxo-7-hydroxy-4-cholenic add;

3

wherein R can be hydroxyl or oxo;

3-0X0-23, 24-bisnor-4-cholene-22-oic acid (4-BNC);

3-oxo-23,24-bisnor-1 ,4-choladiene-22-oic acid (1 , 4-BNC); or variants thereof. In some preferred embodiments, the steroidal product of interest may be:

In some embodiments, in step (b) of the method, the steroidal substrate may be added at a concentration of at least 0.1 mM, at least 0.2mM, at least 0.3mM, at least 0.4mM, at least 0.5mM, at least 0.6mM, at least 0.7mM, at least 0.8mM, at least 0.9mM, at least 1.0mM, at least 1.1mM, at least 1.2mM, at least 1.3mM, at least 1.4mM, at least 1.5mM, at least 1.6mM, at least 1.7mM, at least 1.8mM, at least 1.9mM, or at least 2.0mM. Preferably, the steroidal substrate may be added at a concentration of at least 1 mM, more preferably at least 1.5mM, most preferably at least 2.0mM.

In some embodiments, In step (b) of the method a cydodextrin may be added to the culture medium.

By“cydodextrin" we Include the meaning of a compound made up of sugar molecules bound together in a ring, where the ring Is composed of 5 or more a-D-glucopyranoside units linked 1®4. Non-limiting examples of cyclodextrins Include: a-cydodextrin, b- cydodextrin, g-cydodextrin, methyl-p-cyclodextrin, and 2-OH-propyl-{3-cydodextrin. in certain embodiments, the cydodextrin may be a b-cydodextrin or a y-cydodextrin. Where the cydodextrin is a b-cyclodextrin, It may be a methyl-|3-cydodextrin or a 2-OH- propyl-(3-cydodextrin.

In some embodiments, the cydodextrin Is added at a concentration of 1 mM to 50mM, 2mM to 45mM, 3mM to 40mM, 4mM to 35mM, 5mM to 30mM, 6mM to 29mM, 7mM to 28mM, 8mM to 27mM, 9mM to 26mM, 10mM to 25mM, 11mM to 24mM, 12mM to 23mM, 13mM to 22mM, 14mM to 21mM, 15mM to 21mM, 16mM to 20mM, 17mM to 19mM, 1mM to 18mM. Preferably, the cydodextrin may be added at a concentration of 1mM to 25mM, more preferably 5mM to 25mM.

In other embodiments, the cydodextrin is added at a concentration of at least 1mM, at least 2mM, at least 3mM, at least 4mM, at least 5mM, at least 6mM, at least 7mM, at least 8mM, at least 9mM, at least 10mM, at least 11 mM, at least 12mM, at least 13mM, at least

14mM, at least 15mM, at least 16mM, at least 17mM, at least 18mM, at least 19mM, at least 20mM, at least 21 mM, at least 22mM, at least 23mM, at least 24mM, at least 25mM, at least 30mM, at least 35mM, at least 40mM, at least 45mM, or at least 50mM. Preferably the cydodextrin is added at a concentration of at least 1 mM, preferably at least 5mM, more preferably at least 12.5mM, most preferably at least 25mM.

In some embodiments, in step (b) of the method an organic solvent may be added to the culture medium. By "organic solvent” we Indude the meaning of a carbon-based solvent capable of dissolving other substances. Non-limiting examples of organic solvents Indude: ethanol, dimethylformamide (DMF), acetone, methanol, Isopropanol, dimethyl sulfoxide (DMSO), and toluene.

In certain embodiments, the organic solvent may be ethanol, dimethylformamide (DMF), or acetone. Preferably, the organic solvent may be ethanol.

In some embodiments, the organic solvent Is added the culture medium at a volume/volume (v/v) concentration of 1% to 20%, 2% to 19%, 3%, to 18%, 4% to 17%, 5% to 16%, 6% to 15%, 7% to 14%, 8%, to 13%, 9% to 12%, 10 % to 11%. Preferably, the organic solvent may be added at a volume/volume (v/v) concentration of 5% to 20%, more preferably 5% to 15%.

In some embodiments, toe organic solvent is added to toe culture medium at a volume/volume (v/v) concentration of at least 1%, at least 2%, at least 3%, at least 4%, at least 5%, at least 6%, at least 7%, at least 8%, at least 9%, at least 10%, at least 11%, at least 12%, at least 13%, at least 14%, at least 15%, at least 16%, at least 17%, at least 18%, at least 19%, or at least 20%. Preferably, toe organic solvent may be added at a volume/volume (v/v) concentration of at least 1%. More preferably, toe organic solvent may be added at a volume/volume (v/v) concentration of at least 5%.

In some embodiments, in step (b) of the method a cyclodextrin and an organic solvent are added to the culture medium.

In certain embodiments, where a cyclodextrin and an organic solvent are added to the culture medium, the cyclodextrin is added at a concentration of 1mM to 25mM, 2mM to 24mM, 3mM to 23mM, 4mM to 22mM, 5mM to 21mM, 6mM to 20mM, 7mM to 19mM, 8mM to 18mM, 9mM to 17mM, 10mM to 16mM, 11mM to 15mM, 12mM to 14mM, 1mM to 13mM, and the organic solvent is added at a volume/volume (v/v) concentration of 1% to 20%, 2% to 19%, 3%, to 18%, 4% to 17%, 5% to 16%, 6% to 15%, 7% to 14%, 8%, to 13%, 9% to 12%, 10%to 11%. Preferably, the cydodextrin may be added at concentration of 1mM to 25mM and toe organic solvent may be added at a volume/volume (v/v) concentration of 1% to 10%. More preferably, toe cydodextrin may be added at concentration of 1 mM to 10mM and toe organic solvent may be added at a volume/volume (v/v) concentration of 1% to 10%. Yet more preferably, toe cydodextrin may be added at concentration of 1 mM to 5mM and toe organic solvent may be added at a volume/volume (v/v) concentration of 1% to 5%. Most preferably, toe cydodextrin may be added at concentration of 5mM and the organic solvent may be added at a volume/volume (v/v) concentration of 5%.

In other embodiments, where a cyclodextrin and an organic solvent are added to the culture medium, the cyclodextrin is added at a concentration of at least 1 mM, at least 2mM, at least 3mM, at least 4mM, at least 5mM, at least 6mM, at least 7mM, at least 8mM, at least 9mM, at least 10mM, at least 11mM, at least 12mM, at least 13mM, at least 14mM, at least 15mM, at least 16mM, at least 17mM, at least 18mM, at least 19mM, at least 20mM, at least 21 mM, at least 22mM, at least 23mM, at least 24mM, at least 25mM, and the organic solvent is added at a volume/volume (v/v) concentration of at least 1 %, at least

2%, at least 3%, at least 4%, at least 5%, at least 6%, at least 7%, at least 8%, at least 9%, at least 10%, at least 11%, at least 12%, at least 13%, at least 14%, at least 15%, at least 16%, at least 17%, at least 18%, at least 19%, or at least 20%. Preferably, the cyclodextrin may be added at concentration of at least 1mM and the organic solvent may be added at a volume/volume (v/v) concentration of at least 1%. More preferably, the cyclodextrin may be added at concentration of at least 5mM and the organic solvent may be added at a volume/volume (v/v) concentration of 5%. in a fourth aspect, the Invention provides a steroidal product of interest produced by the method of the third aspect.

In a fifth aspect, the invention provides a kit for converting a steroidal substrate into a steroidal product of interest wherein the kit comprises:

(a) a genetically-modified bacterium according to the first aspect; and,

(b) Instructions for using the kit

The kit may further comprise a steroidal substrate. In some embodiments, the steroidal substrate may be a sterol substrate. In certain embodiments, the sterol substrate comprises:

7-oxo-phytosterol; or a combination thereof. in some embodiments, the kit may further comprise a cyclodextrin such as a b-cyclodextrin or a Y-cyclodextrin. Preferably, the cydodextrin is a b-cyclodextrin, more preferably a methyl-p-cyclodextrin or a 2-OH-propyi-p-cyclodextrin. In some embodiments, the kit may further comprise an organic solvent. In certain embodiments, the organic solvent is ethanol, dimethylformamide (DMF), or acetone, preferably ethanol.

The listing or discussion of an apparently prior-published document in this specification should not necessarily be taken as an acknowledgement that the document is part of the state of the art or is common general knowledge. The deposits referred to in this disclosure (Accession Nos. NCIMB 43057, NCIMB 43058,

NCIMB 43059, and NCIMB 43060) were deposited at the National Collection of Industrial, Food and Marine Bacteria, Ferguson Building, Craibstone Estate, Bucksbum, Aberdeen, AB21 9YA, UK by Cambrex Kariskoga AB on 29 May 2018. The present Invention will now be described In more detail with reference to the following non-limiting figures and examples.

DESCRIPTION OF THE FIGURES FIGURE 1. Overview of cholesterol catabolic pathway.

FIGURE 2. Overview of cholesterol side-chain degradation pathway.

FIGURE 3. Overview of b-srtosterol side-chain degradation pathway.

FIGURE 4. Examples of steroidal compounds

FIGURE 5. Examples of steroidal substrates. FIGURE 6. Total Ion chromatogram obtained by LC-MS for LM3 cultured when cholesterol Is the starling substrate. Peaks at 7.67 minutes and 8.25 minutes indicate accumulation of 4-androstene-3,17-dione (AD) and 3-oxo-23,24-bisnor-4-cholene-22-oic acid (4-BNC) respectively. NL: Normalisation Level = base peak intensity. FIGURE 7. Product Ion mass spectra obtained by LC-MS for LM9 when cholesterol Is the starting substrate. (A) Peak at Peak at m/z of 345.24 (positive mode) corresponds to 4-BNC being accumulated when cholesterol is the starting substrate. (B) Peak at m/z of 373.27 (positive mode) corresponds to 3-oxo-4-cholenic acid being accumulated when cholesterol Is toe starting substrate. NL: Normalisation Level = base peak Intensity.

FIGURE 8. Product Ion mass spectra obtained by LC-MS for LM9 when cholesterol, p-sttosterol, or 7-oxo-sterol Is the starting substrate. (A, top) Peak at m/z of 389.27 (positive mode) corresponds to production of 3-oxo-7-hydroxy-4-cholenlc acid when 7- oxo-sterol is toe starting substrate. (B, middle) Peak at m/z of 373.27 (positive mode) corresponds to production of 3-oxo-4-cholenlc add when b-sitosterol Is toe starting substrate. (C, bottom) Peak at m/z of 373.27 (positive mode) corresponds to production of 3-oxo-4-cholenic acid when cholesterol Is toe starting substrate. NL: Normalisation Level = base peak Intensity.

FIGURE 9. Extracted Ion chromatograms obtained by LC-MS for LM19 and LM9 when cholesterol or b-sitosterol Is the starting substrate. (A) Strain = LM9; Substrate = Cholesterol. Peak at 9.70 minutes corresponds to production of 3-oxo-4-cholenic acid by LM9. (B) Strain = LM19; Substrate = Cholesterol. Peak at 8.07 minutes corresponds to production of 3-oxo-9-OH-4-cholenic acid by LM19. (C) Strain = LM9; Substrate = b- sitosterol. Peak at 9.68 minutes corresponds to production of 3-oxo-4-cholenic acid by LM9. (D) Strain = LM19; Substrate = b-sitosterol. Peak at 8.09 minutes corresponds to production of 3-oxo-9-OH-4-cholenlc acid by LM19. NL Normalisation Level « base peak intensity.

FIGURE 10. Product Ion mass spectra obtained by LC-MS confirming Identity of peaks produced by LM9 and LM19 when cholesterol or b-sltosterol Is the starting substrate. (A) Strain = LM19; Substrate = Cholesterol or b-sitosterol. Peak at m/z of approximately 389.27 (positive mode) corresponds to production of 3-oxo-9-OH-4- cholenic add by LM19. (B) Strain = LM9; Substrate = Cholesterol or p-sitosterol. Peak at m/z of 373.27 (positive mode) corresponds to production of 3-oxo-4-cholenic acid by LM9. NL: Normalisation Level = base peak Intensity.

FIGURE 11. Product Ion mass spectra obtained by LC-MS for LM19 when 7-oxo- sterol Is the starting substrate. Peak at m/z of 405.26 (positive mode) corresponds to production of 3-oxo-7,9-dihydroxy-4-cholenic add by LM19. NL: Normalisation Level = base peak Intensity.

FIGURE 12. HPLC analysis comparing the steroidal products produced by LM9 and LM33 when b-sitosterol Is the starting substrate and the culture medium is supplemented with methyl-p-cyclodextrins. The upper line on the graph represents the HPLC trace for the steroidal compounds produced by LM9 and the lower line represents the HPLC trace for the steroidal compounds produced by LM33. FIGURE 13. HPLC analysis comparing the activity of LM9 and LM33 towards 3-oxo- 4-cholenic acid as the starting substrate and the culture medium Is supplemented with methyl-p-cyclodextrlns. The upper line on the graph represents the HPLC trace for the steroidal compounds produced by LM9 (T= 72h) and the lower line represents the HPLC trace for the steroidal compounds produced by LM33 (T =72h).

FIGURE 14. Product Ion mass spectrum obtained by LC-MS for LM9 when b- sltosterol Is the starting substrate and the culture medium Is supplemented with 2- OH-propyl-p-cyclodextrlns. Peak at m/z of 373.27 (positive mode) corresponds to production of 3-oxo-4-cholenlc add by LM9. NL: Normalization Level = base peak intensity.

FIGURE 15. HPLC analysis of steroidal compounds produced by LM9 b-sitosterol Is the starting substrate and the culture medium Is supplemented with 2-OH-propyl- b-cyclodextrins. (A) LM9 products at T = 24h; (B) LM9 products at T = 48h; (C) LM9 products at T = 72h; (D) 3-oxo-4-cholenic add standard (0.025mg/mL).

FIGURE 15. Extracted Ion chromatograms obtained by LC-MS for LM9 when 7- oxosterols Is the starting substrate and the culture medium Is supplemented with 2- OH-propyl-p-cyclodextrlns. (A) LM9 products In the presence of 2-OH-propyi-{5- cyclodextrins (T = 48h). Peak at 7.74 minutes corresponds to production of 3-oxo-7- hydroxy-4-cholenic add. (B) LM9 products In the absence of 2-OH-propyi-p-cyclodextrins (T = 48h) Peak at 7.76 minutes corresponds to production of 3-oxo-7-hydroxy-4-cholenlc acid. (C) LM9 products in the presence of 2-OH-propyf-p-cydodextrins but no substrate (T = 48h). NL: Normalization Level = base peak intensity.

FIGURE 17. HPLC analysis of steroidal compounds produced by My¥bacterium neoaurum NRRL B-3805 (parent strain) and MnaoAfadE34 when cholesterol Is the starting substrate. The upper line on the graph represents the HPLC trace for the steroidal compounds produced by the parent strain (T =72h) and the lower line represents the HPLC trace for the steroidal compounds produced by MneoAfadE34. FIGURE 18. HPLC analysis of steroidal compounds produced by Mycobacterium neoaurum NRRL B-3805 (parent strain) and MneoAfadE34 when p-sltosterol is the starting substrate. The upper line on the graph represents the HPLC trace for the steroidal compounds produced by MneoAfadE34 (T =72h) and the lower line represents the HPLC trace for the steroidal compounds produced by the parent strain

FIGURE 19. HPLC analysis of steroidal compounds produced by Mycobacterium neoaurum NRRL B-3805 (parent strain) and MneoA†adE34 when 7-oxo-sterols are the starting substrate. The upper line on the graph represents the HPLC trace for the steroidal compounds produced by MneoAfadE34 (T =72h) and the lower line represents the HPLC trace for the steroidal compounds produced by the parent strain.

FIGURE 20. HPLC analysis of steroidal compounds produced by MneoAfadE34 when phytosterol mix (Aturex 90) Is the starting substrate and the culture medium is supplemented with methyl-P-cyclodextrlns. From bottom to top the traces shown correspond to the steroidal compounds produced by MneoAfadE34 at T = Oh, 24h, 48h, 72h, 96h, and 168h respectively.

FIGURE 21. HPLC analysis of steroidal compounds produced by MneoAfadE34 when 3-oxo-4-cholenlc acid is the starting substrate and the culture medium Is supplemented with methyl-p-cyciodextrlns. From bottom to top the traces shown correspond to the steroidal compounds produced by MneoAfadE34 at T = Oh, 24h, 48h, 72h, 96h, and 168h respectively. FIGURE 22. NMR analysis of steroidal compounds produced by LM33 after fermentation with phytosterol compounds in tile presence of hydroxypropyl-b- cyclodextrln. (A) The ¹ H-spectrum obtained from the product of the fermentation; (B) Magnified view of the spectrum of Figure 22A showing peaks in the region 0.65 to 2.55 ppm only; (C) The ¹³ C-spectrum obtained from the product of the fermentation; (D) Magnified view of the spectrum of Figure 22C showing peaks In the region 11 to 58 ppm only. Both the ¹ H-spectrum and the ¹³ C-spectrum indicate the presence of 3-oxo-4- cholenic add in the culture; (E) Data parameters used to obtain the ¹ H-spectrum shown in Figures 22A and 22B; (F) Data parameters used to obtain the ¹³ C-spectrum shown in Figures 22C and 22D.

Examples Example 1 - Construction of strains

Materials and methods

Construction of RG41 strain

RG41 was originally constructed from the parent strain RG32 which was made by unmarked gene deletion of five homologs of 3-ketosteroid-9a-hydroxylase (kshA1-5) as reported by (Wilbrinkef a/, 2011. Appl Environ Microbiol., 77(13): 4455-4464).

RG32 was used as parent strain for the construction of ft rhodochrous strain RG41 by deletion of 3 homologs of 3-ketosteroid-A1 -dehydrogenase (kstDs) as detailed below.

The construction of a mutagenic plasmid for kstD3 unmarked deletion was performed as follows. A genomic library of ft rhodochrous DSM43269 was obtained as explained in (Petrusma ef a/, 2009. Appl Environ Microbiol., 75^6): 5300-5307), which was used for isolation of a done (pKSH800; Wilbrink et at, 2011) carrying kshA3 and also kstD3. A 4 kb EcoRI fragment of pKSHSOO was ligated Into EcoRI-dlgested pZErO-2.1, which was subsequently digested with BglWEcoR\. Next, a 2.7 kb Bgl\\/EcoR\ fragment was ligated into BamH l/EcoRl-d igested pK18mobsacB, which was then digested with EcoRV/Afri/l and finally self-llgated, rendering the plasmid pKSH841 for kstD3 gene deletion In ft rhodochrous RG32 strain => RG32MsfD3 = strain RG35 (Appendix C).

The construction of a mutagenic plasmid for kstD1 unmarked deletion was performed as follows. Specific kstD1 primers (kstD1-F and kstD1-R, Appendix D) were used for the amplification of a 2.4 kb PCR product that was ligated into EcoRV-digested pBluescript, which was then digested with Stu\ISty\, blunt-ended by Klenow and self-ligated. Then, the construct was digested with BamHI/H/ndlll and, finally, a 1.3 kb SamHl/H/ndlll fragment was ligated Into BamHI/H/ndlll-dlgested pK18mobsacB, rendering the plasmid pKSH852 for kstD1 gene deletion in RG35 => RG32MsfD 1£J<stD3 = strain RG36 (Table Appendix

C).

The construction of a mutagenic plasmid for kstD2 unmarked deletion was performed as follows. Chromosomal DMA of ft rhodochrous RG36 was isolated using a genomic DMA isolation kit (Sigma-Aldrich), digested by Xho\, and ligated Into Xhol-d igested pZErO-2.1. Transformation of E. coll DH5a with the ligation mixture generated a genomic library of approximately 12,000 transformants. A done carrying the kstD2 gene (pKSD321) was identified by means of PCR using specific kstD2 primers (kstD2-F and kstD2-R, Appendix D) and Isolated from the genomic library of strain RG36. Then, pKSD321 was digested with Xmn\, self-ligated and subsequently digested with Sma\IXho\. Finally, a 2.2 kb Sma\/Xho\ was ligated into Smal/SaAdigested pK18mobsacB, rendering the plasmid pKSD326 for the kstD2 gene deletion in RG36 => RG32MsfD 1AkstD2AkstD3 = strain RG41 (Appendix C).

Mutagenic plasmids were transferred to Escherichia coll S17-1 by transformation and subsequently mobilized to the corresponding ft rhodochrous strain by conjugation as described previously (van der Gelze et al, 2001. FEMS Microbiol. Lett., 205(2): 197-202). All mutants were verified by PCR using specific primers (Appendix D) to confirm deletion of the target gene(s).

Therefore, strain RG41 is a kshA null + AkstD1AkstD2AkstD3 mutant (8-fold mutant), which was then used as parent strain for the construction of deletion mutants in genes involved in side-chain degradation of steroids.

Construction of deletion mutation strains

The single mutant strains LM3 (D fadE34#1), LM15 (D fadE34#2) were constructed by deletion of fadE34tt1 or fadE34#2 from the parent strain RG41 (kshA null + AkstDI + LkstD2+ AkstD3).

Unmarked in frame gene deletion mutants were constructed using the sacB counterselection marker (van der Geize et a/, 2001). PCR amplification of the upstream and downstream flanking regions of the target genes was performed from wild-type ft rhodochrous DSM43269 template using the primers listed in Appendix D. The obtained 1.5 kb PCR products (called UP and DOWN) were cloned together Into pK18mobsacB vector, yielding pk18 JacO4-UP+D0WN and pk18_/adE34#2-UP+DOWN constructs. pDEL-fedAe, previously constructed by Wilbrink et a/., 2011, was used for the deletion of fadA6. Mutagenic plasmids were transferred to Escherichia coli S17-1 by transformation and subsequently mobilized to the corresponding ft. rhodochrous strain by conjugation as described previously (van der Geize et al, 2001 ). All mutants were verified by PCR using specific primers (Appendix D) to confirm deletion of the target gene(s). LM3 and LM15 single mutant strains were constructed by deletion of fadE34 or fadE34#2, respectively, using RG41 as parent strain. Example 2 - Bloconverslons using strains LM3 (AfadE34#1) and LM15 (AfadE34#2)

Materials and methods Mutant strains were inoculated in 100ml Lurla-Bertanl (LB) medium and incubated at 30°C and 200rpm for 48 hours. When foe ODeoonm=5, foe LB preculture was divided Into 10ml cultures and foe starting sterol substrate added at 2mM (dissolved in acetone to 4% final concentration).

The time of addition of foe starting sterol substrate was treated as T=0 hours. Cultures were incubated at 30 ^e C and 200rpm for several days. 250mI aliquots were taken from foe culture at 0 hours, 24 hours, 48 hours, and 72 hours, and frozen at -20 ^e C until needed. Samples were prepared for HPLC and/or LC-MS analysis by thawing at room temperature and adding 1ml MeOH before briefly vortexing and centrifuging at 4°C and 12,000rpm for 10-15 minutes. The supernatants were then filtered (0.2pm filter size) and analysed by HPLC and/or LC-MS. HPLC was performed using a Kinetex C18 column (250x4.8mm, particle size 5 pm). A mobile phase of 80% MeOH and 0.1% formic acid was used at a flow rate of 1ml/min and a column temperature of 35°C. 20mI of sample was injected. A 30-minute detection time was used, and steroidal compounds were detected at 254nm. Quantification of foe steroidal products produced was achieved by construction of a calibration line of peak areas measured from a known standard. This was used to calculate the amount of product produced in g/l, followed by back calculation of the percentage yield.

LC-MS analysis was carried out using an Accella1250™ HPLC system coupled with foe benchtop ESI-MS Orbitrap Exactive ¹ * ¹ (Thermo Fisher Scientific, San Jose, CA). A sample of 5 pi was injected into a Reversed Phase C18 column (Shim Pack Shimadzu XR-ODS 3x75 mm) operating at 40°C and flow rate 0.6 ml/min. Analysis was performed using a gradient from 2% to 95% of acetonitrile:water (adding 0.1% formic acid) as follows: 2 min 2% acetronitrile, 8 minutes gradient from 2% to 95% acetonitrile, 4 min 95% acetonitrile. The column fluent was directed to foe ESI-MS Orbitrap operating at foe scan range (m/z 80 - 1600 Da) switching positive / negative modes. Voltage parameters for positive mode were: 4.2 kV spray, 57.5 V capillary and 95 V tube lens. Voltage parameters for negative mode were: 3kV spray, -25V capillary and -75V tube lens. Capillary temperature 325 ^e C, sheath gas flow 70, aiodliary gas off. Thermo XCalibur™ processing software was used for foe data analysis. All foe products reported in this work were detected in foe positive mode (M+H*).

Result s The total ion chromatogram obtained by LC-MS for the LM3 strain shows an accumulation of AD and 4-BNC from the starting cholesterol substrate (Figure 6), Indicating there is no blockage of side-chain degradation in the LM3 single mutant strain. The same result was obtained for the LM15 single mutant strain (data not shown).

Example 3 - Bioconversions using LM9 (hfadE34#1IAfadE34#Z)

Materials and methods

The same culture conditions, sample preparation techniques and HPLG/LC-MS protocol were used as outlined in Example 2 above.

Results

The total ion chromatogram obtained by LC-MS for the LM9 strain (Figure 7, product ion mass spectra shown) using cholesterol as toe starting substrate, shows an accumulation of both 4-BNC (peak at m/z of 345.24, positive mode) (figure 7A, top) and 3-oxo-4- cholenic acid (peak at m/z of 373.27, positive mode) (Figure 7B, bottom). Extracted ion chromatograms, produced by extracting data for toe mass to charge ratio (m/z) of toe compound of interest, show that 3-oxo-4-cholenic add is produced by LM9 when cholesterol (figure 8C, bottom trace, peak at m/z of 373.27) or b-sitosterol (Figure 8B, middle trace, peak at m/z of 373.27) is toe starting substrate, and that 3-oxo-7-hydroxy-4- cholenic add is produced when 7-oxo-sterol is toe starting substrate (Figure 8A, top trace, peak at m/z of 389.27). These results indicate that there is some blockage of side-chain degradation in toe LM9 strain. Example 4 - Bioconversions using strain LM19 (hfadE34#1IAfadE34#2 complemented with kshAS)

Materials and methods

Construction of LM19 strain

A wild-type copy of toe kshA5 gene and its flanking regions was amplified by PCR using the primers kshA5-complem-F and kshA5-complem-R (Appendix D). The obtained PCR product of 2.2 kb was deaned-up, restricted with SamHI/H/ndlll and subsequently ligated into pk18mobsacB, yielding toe construct pki 8+/rsM 5-complementation. This construct was transformed into £. coll S17-1 and transferred to strain LM9 by conjugation. The resulting complemented mutant LM19, in which toe deleted copy of kshA5 was replaced by toe wild-type one, was obtained following toe same conjugation protocol used for toe construction of toe mutant strains, as described in van der Geize et al, 2001. Bloconvereions with LM19

As described above, kshA5 and its flanking regions was reintroduced into strain LM9 to produce strain LM19, in which hydroxylase activity Is restored to produce variant compounds with a 9-hydroxyl group. The expected compounds accumulated were 3-oxo-

9-OH-4-cholenic add (from b-sitosterol and cholesterol) and 3-oxo-7,9-dihydroxy-4- cholenlc add (from 7-oxo-sterols).

The same culture conditions, sample preparation techniques and HPLC/LC-MS protocol were used as outlined in Example 2 above.

Results

Comparison of the extracted Ion chromatograms produced for LM9 and LM19 strains shows that 3-oxo-9-OH-4-cholenic acid (peak at 8.07-8.09 minutes) is produced by LM19 only when the starting sterol is cholesterol or b-sitosterol (Figure 9A and 9C respectively) and 3-oxo-4-cholenic add (peak at 9.68-9.70 minutes) is produced by LM9 only when the starting sterol is cholesterol or b-sitosterol (Figure 9B and 9D respectively). Those peaks were confirmed as 3-oxo-9-OH-4-cholenic add (peak at m/z of approximately 389.27, positive mode) is produced by LM19 when the starting sterol Is cholesterol or b-sitosterol (Rgure 10A) and 3-oxo-4-cholenic add (peak at m/z of 373.27, positive mode) Is produced by LM9 when the starting sterol is cholesterol or b-sitosterol (Figure 10B).

When the starting sterol Is 7-oxo-sterol the expected product Is 3-oxo-7,9-dihydroxy-4- cholenic add. The extracted ion chromatogram for LM19 in figure 11 has a peak corresponding to 3-oxo-7,9-dihydroxy-4-cholenic add (peak at m/z of 405.26, positive mode). However, this peak is of lower intensity than those produced for LM19 in Figure 10. In overview, these results Indicate the successful use of LM19 in the production of variant steroidal compounds with a 9-hydroxy group. Example 5 - Bloconvereions using strain LM33 (hfadE34#1/AfadE34#2lhfadE26)

Materials and methods

An additional mutant strain AfedE34#7/AfadE34#2/A/adE26 (LM33) was produced by deletion of fadE26 from the LM9 strain. FadE26 is involved in the first cyde of b-oxidation (figures 2 and 3) and may also use 3-oxo-4-cholenic add as a substrate (Yang ef a/, 2015.

ACS Infect Dis., 1(2):110-125), thereby limiting its accumulation. Thus, it was thought that deletion of fadE26 might lead to a reduction in unwanted oxidation of 3-oxo-4-cholenic add.

The same culture conditions, sample preparation techniques and HPLC/LC-MS protocol were used as outlined in Example 2 above.

Results

A comparison of the bioconversion of b-sitosterol by the LM9 and LM33 strains in the presence of 25mM methyl-p-cyclodextrins (MGDs) (see Example 7 below), shows that the major peak in the HPLC trace for the LM33 sample is 3-oxo-4-cholenic add and the peaks corresponding to AD and 4-BNC are much smaller, while the converse is observed in the HPLC trace for LM9 (Figure 12). This indicates that the additional deletion of fadE26 in LM33 enables the further accumulation of 3-oxo-4-cholenic add, suggesting that unwanted oxidation of 3-oxo-4-cholenlc add Is reduced.

Furthermore, a comparison of the activity of the LM9 and LM33 strains towards 3-oxo-4- cholenic add as the starting substrate in the presence of 25mM methyl-fJ-cyclodextrins (MCDs) shows that the major peak in the HPLC trace for the LM33 sample remains as 3- oxo-4-cholenlc add and peaks corresponding to AD and 4-BNC are very small. In contrast, in the HPLC trace for LM9 (Figure 13) the peak for 3-oxo-4-cholenlc add is decreased and the peaks for AD and 4-BNC are much more prominent. This indicates that in LM9 the concentration of 3-oxo-4-cholenic acid decreases with time as AD and 4- BNC are formed but in LM33, where fadE26 is also deleted, the conversion of 3-oxo-4- cholenlc add to AD and 4-BNC is significantly reduced. Those results therefore suggest that unwanted oxidation of 3-oxo-4-cholenic acid Is reduced in LM33.

Example 6 - Bioconversions using LM9 in a culture medium supplemented with 2- OH-propyl-p-cyclodextrlns Materials and methods

The addition of 2-OH-propyl-(3-cyclodextrins to the culture medium was attempted to improve the solubility of the hydrophobic sterol starting compounds.

The LM9 strain was cultured as described in Example 2 until the ODeoonm=5 after approximately 48 hours. The culture was centrifuged at room temperature and 4,500rpm for 15-20 minutes. The cells were resuspended in the same volume of minimal medium (K2HP0 ₄ (4.65g/l), NaH ₂ P0* H ₂ 0 (1.5g/l), NH ₄ CI (3g/l), MgS0 ₄ -7H ₂ 0 (1g/l), and Vishniac trace element solution (1 ml/l)). This was divided into 10ml cultures and 25m M 2-OH- propyi-p-cyclodextrins, 25mM NaHCOa and 2mM sterols were added In powder form.

The same sample preparation techniques and HPLCZLC-MS protocol were used as outlined in Example 2 above.

Results

The extracted ion chromatogram obtained by LC-MS of the LM9 strain using b-sttosterol as the starting substrate shows that 3-oxo-4-cholenic acid (peak at m/z of 373.27, positive mode) is produced by LM9 in the presence of 2-OH-propy1-|3-cyclodextrins (Figure 14). In order to quantify the amount of 3-oxo-4-choienic acid produced HPLC analysis was performed (Figure 15), with a yield of 11.64% observed in the sample taken at the 72-hour time point (Table 1 below).

Similar experiments were performed using 7-oxo-sterols as the starting substrate, and the extracted ion chromatograms show the production of 3-oxo-7-hydroxy-4-cholenlc add at T = 48h (Figure 16). Comparison of the LC-MS spectra in the presence and absence of 2-OH-propy1-|3-cyclodextrins (Figure 16A and 16B) reveals a more intense base peak (evidenced by the NL values on the traces in Figure 16) in the presence of 2-OH-propyi- b-cyclodextrlns, indicating a higher yield of 3-oxo-7-hydroxy-4-cholenlc add in those cultures. However, due to the lack of an available standard for HPLC quantification, there is no available data on obtainable percentage yields.

Equivalent experiments were carried out in which the culture was not supplemented with NaHCOa (data not shown). In those experiments there was no significant difference from the results shown in Figures 14, 15, and 16 and presented in Table 1, thereby indicating that the presence of NaHCOs is not required to produce a positive effect on yield in cultures supplemented with 2-OH-propyl-|3-cydodextrins.

Example 7 - Bioconversions using LM9 and LM33 In a culture medium supplemented with methyl-p-cyclodextrins

Materials and methods

The addition of methyi-p-cyclodextrins to the culture medium was attempted to further improve the solubility of the hydrophobic sterol starting compounds.

The LM9 strain was cultured as described In Example 2 until the ODeoonm=5 after approximately 48 hours. The culture was centrifuged, and the cells resuspended in the same volume of minimal medium, as described In Example 6. This was divided Into 10ml cultures and 25mM methyl-p-cyclodextrins and 2mM sterols were added in powder form.

In an attempt to further maximise the yield of 3-oxo-4-cholenic add, methyH3-cyclodextrins were added to the LM33 strain (see Figure 12). The LM33 strain was cultured in LB medium as described in Example 2 until the ODeoonm=5 after approximately 48 hours. Then, the preculture was divided into 10ml cultures and 25mM methyl-|3-cyclodextrins and 2mM sterols were added in powder form. Alternatively, the culture was centrifuged, and the cells resuspended In the same volume of minimal medium. This was divided into 10ml cultures and 25mM methyHJ-cydodextrins and 2mM sterols were added in powder form.

The same sample preparation techniques and HPLC/LC-MS protocol were used as outlined In Example 2 above.

Results

Table 2 below summarises the maximum percentage yields of 3-oxo-4-cholenic add obtained by HPLC analysis for LM9 in the presence of methyl-p-cydodextrins using b- sitosterol as the starting substrate and compares those yields to the yields obtained in toe presence of 2-OH-propyl-fJ-cyclodextrins (see Example 6 above). The overall result indicates that yields are higher in toe presence of metoyl-p-cydodextrins.

Quantification of the amount of product produced by LM9 in the presence of methyl-b- cyclodextrins (25mM) was carried out using HPLC analysis, b-sitosterol was the starting substrate and the analysed sample was collected at the 72-hour timepolnt. The percentage yields were calculated as outlined In Example 2 above and are presented in Table 3 below.

Similarly, Table 4 below compares bioconversions In LM9 in the presence of methyl-b- cyclodextrins using 7-oxo-sterols as the starting substrate. Due to the lack of available standard for 3-oxo-7-hydroxy-4-cholenic acid, peak areas obtained by HPLC are compared rather than expressed as a percentage yield. However, the results still demonstrate that larger peak areas are achieved In the presence of methyl-p-cyclodextrins compared with 2-OH-propyl-p-cydodextrins.

pp y ( )

Table 5 below summarises the percentage yields of 3-oxo-4-cholenic acid obtained by HPLC analysis for LM33 using both cholesterol and b-sitosterol as starting substrates and culturing In both LB and minimal medium In the presence of methyl-fi-cydodextrins. Comparing the data in Table 3 above and Table 5 below shows that culturing LM33 In the presence of methyl-p-cyclodextrlns results in the highest percentage yield of 3-oxo-4- cholenic acid when b-sitosterol is the starting substrate.

T

m

Example 8 - Bioconversions using LM33 In culture medium supplemented with organic solvents and cyclodextrins

Materials and methods

The LM33 strain was cultured as described In Example 1 until the ODeoonm=5 after appro)dmately 48 hours. The culture was centrifuged at 4,500rpm at room temperature for 15-20mins. The cells were resuspended In the same volume of minimal medium and the culture divided Into 10ml cultures. 2mM b-sltosterol was added dissolved in ethanol (5% or 10% final volume/volume concentration) and different amounts of methy!-b- cyclodextrins (5mM, 12.5mM, or 25mM) were added in powder form. The same sample preparation techniques and HPLC/LC-MS protocol were used as outlined in Example 2 above.

Results

HPLC data for all concentrations of ethanol and methyl-p-cydodextrins was processed as described in Example 2 to obtain the percentage yields of 3-oxo-4-cholenic add displayed

In Table 6 below. Overall, the use of 5% ethanol and 5mM methyl-P-cydodextrins In combination results In the highest percentage yield.

Example 9 - Bioconversions using Mycobacterium neoaurum NRRL B-3805 AfadE34 (MneoAfadE34)

Materials and methods

The Mycobacterium neoaurum NRRL B-3805 AfadE34 strain was produced by Introducing a deletion of fadE34 into the parent strain NRRL B-3805 (Marsheck at al, 1972. Applied Microbiology, 3(1 ):72-77), with the aim of preventing the oxidation of 3-oxo-4-cholenic add. This followed the same strategy described in Example 1, using the parent strain NRRL B- 3805 template and the primers listed in Appendix D, pk18_fadE34_Mneo-U P+DOWN plasmid was constructed. This mutagenic plasmid was transferred to NRRL B-3805 strain by electroporation (2.5kV, 25pF, 600Q). The mutant strain was verified by PCR using specific primers (Appendix D) to confirm deletion of fadE34.

MneoAfadE34 precultures were grown to an ODeoonm=5 (~72h at 37°C). The culture was centrifuged, and the cells suspended in the same volume of minimal medium. 2mM of the starting steroid substrate was added in powder form.

The same sample preparation techniques and HPLC/LC-MS protocol were used as outlined in Example 2 above. Results

The HPLC traces of Figure 17, figure 18 and Figure 19 compare the compounds produced by the Mneo- parent strain and MneoMadE34 strain when cholesterol, b-stosteroi and 7- oxosterols are the respective starting substrates. In the case of cholesterol (Figure 17) and b-sitosterol (Figure 18), the MneotJadE34 strain accumulates higher levels of 3-oxo-

4-cholenic acid and lower levels of AD and ADD than the Mneo- parent strain. These results indicate that the MneohfadE34 strain is blocked In side-chain oxidation at the 3- oxo-4-cholenic acid step. The Mneo parent strain NRRL B-3805 was described as lacking KSH and KstD, however, it was observed that there is also a peak that corresponds to production of 3-oxo-1 ,4-choladlenoic acid, Indicating that MneotJadE34 (and therefore the parent strain NRRL B-3805) may have residual KstD activity.

When 7-oxosterols are tire starting substrate (Figure 19), the traces obtained for the Mneo parent strain NRRL B-3805 and MneoLfedE34 are very similar, indicating that 7-OH compounds are not able to be accumulated.

Example 10 - Bloconverslons using Mycobacterium neoaurum NRRL B-3805 LfadE34 (MneotfadE34) In culture medium supplemented with methyl-b- cyclodextrlns

Materials and methods

The same strains and culture conditions were used as outlined In Example 9 above, and 25mM methyl-p-cyclodextrins was added in powder form. 2mM phytosterol mix (Aturex 90) or 3-oxo-4-cholenic acid were added as the starting compounds. The same sample preparation techniques and HPLCZLC-MS protocol were used as outlined In Example 2 above.

Results

Mneo-MadE34 accumulates a possible peak of 3-oxo-1 ,4-choladlenoic add when those cells are cultured In minimal medium in the presence of methyl-f3-cydodextrins and phytosterol mix Is the starting substrate (Figure 20).

When 3-oxo-4-cholenlc add Is the starting substrate, there is no accumulation of 3-oxo- 1 ,4-cholenlc add (Figure 21), Indicating it is likely that its production Is not activated by the presence of 3-oxo-4-cholenlc acid. Example 11 - Bioconversions using LMI33 in culture medium supplemented with hydroxy-propyl-B-Cyclodextrln

Materials and methods

The bioconversion was carried out with growing cells i.e. with bioconversion reagents added to the reactor at the beginning of the fermentation. A pre-culture was prepared as follows:

1) 50 mL LB medium was added to 100 m L conical flask;

2) 200 pL R. rhodochrous LM33 was inoculated from glycerol stock;

3) The culture was incubated at 400 RPM on an orbital shaker for 48 hours at 30 ^e C;

4) 1% (5 mL) of OD 5 culture was Inoculated into the bioreactor.

The bioreactor was loaded with (final concentrations in brackets): Tryptone (10 g/L); Yeast Extract (10 g/L); NaCL (0.5 g/L); Antifbam DR204 (0.015 g/L); Hydroxy-propyl·^· Cyclodextrin (23.3 g/L); a premade mixture of Phytosterols AS-7 (70 g/L) and Tween-80 (17.5 g/L); and water. The mixture was autoclaved in the reactor at 121°C for 3 minutes.

i

The bioconversion was run at 30 ^* C, pH 7.0 with aeration from surface at 200 mL/min and d02 set point at 40%.

The initial growth lasted for less than 12 hours as judged from oxygen consumption and a slight CO2 production. After 48 hours from the start of the experiment there was no CO2 production and the bioconversion was reinoculated from a fresh pre-culture, at an inoculation rate of 10% (50 mL). After the second inoculation, there was also an initial oxygen consumption phase, which lasted for 6 hours, again followed by a reduction in oxygen consumption. However, after that reduction the culture recovered and started consuming oxygen and producing C<¾ again.

Formation of 3-oxo-4-cholenic acid was detected at the 112th hour. The experiment was concluded after 160 hours, at which point the product concentration had reached 6.09 mM.

The biomass and unreacted phytosterols were separated by first increasing the pH of the culture solution to pH 10 by the addition of 2M NaOH, followed by centrifugation at 4700 g for 10 minutes at 4 *C, affording a clear solution (453 g) containing the product. From this solution, 360g (pH=7.2) was extracted with 4x100 ml MTBE, then adjusted to pH-2.1 with diluted HCI and extracted again with 2x100 ml toluene. The majority of 3-oxo-4-cholenic add was detected in MTBE extracts and a minority in toluene extracts. The extracts were evaporated to dryness and pooled by dissolving in MTBE. The solution was washed with diluted HCI and concentrated on rotavap. From the obtained residue, 3-oxo-4-cholenic acid was predpitated by overnight stirring. The identity of 3-oxo-4-cholenic add was confirmed by NMR (Figure 22).

Results

The spectra of Figure 22 confirm the Identity of the isolated produti as 3-oxo-4-cholenic add. Figure 22(A) and (B) depict the ¹ H-spectrum and Figure 22(C) and (D) depict the ¹³ C-spectrum obtained from the products. The labelling of the peaks corresponds to the functional groups depicted in the formula of 3-oxo-4-cholenic add shown in Figure 22(A) and (C).