FIELD OF THE INVENTION

The invention relates to glyceryl tridecanoate and uses thereof in treating endocrine disorders. More particularly, glyceryl tridecanoate may be used to treat hyper-androgenism, insulin resistance and infertility caused by, for example, polycystic ovarian syndrome, prostate cancer; and may be used for aesthetic medical treatment for hirsutism and male pattern baldness.

BACKGROUND OF THE INVENTION

Polycystic ovarian syndrome (PCOS) is an endocrine disorder contributing to infertility in women. Hyperandrogenism, characterized by excessive production of Δ4 steroids such as androstenedione and testosterone, is a hallmark of PCOS and results in anovulation, oligomenorrhea, and infertility. The pathomechanism of hyperandrogenism is related to elevated LH levels observed in PCOS patients. It is long established that exposure of ovarian theca and granulosa cells to LH leads to increased levels of the intracellular second messenger cAMP, and stimulation of an ordered cascade of steroidogenic enzymes that produce Δ4 androgens from cholesterol.

Insulin resistance is a major pathophysiological feature of hyperandrogenic PCOS, occurring at a high incidence rate of 50% to 70% in women with PCOS. The compensatory hyperinsulinemia resulting from insulin resistance drives androgen production from theca cells. High insulin also decreases sex hormone-binding globulin (SHBG) thereby increasing the proportion of free bioactive androgens, while concomitantly raised androgen levels can reduce peripheral insulin sensitivity (Holmang A, et al., Am. J. Physiol. 259(4 Pt 1 ): E555- 560 (1990); Holmang A, et al., Am. J. Physiol. 262(6 Pt 1 ): E851-855 (1992); Rincon J, et al., Diabetes 45(5): 615-621 (1996); Holmang A, et al., Acta Physiol. Scand. 171(4): 427-438 (2001 ); Corbould A. J. Endocrinol. 192(3): 585-594 (2007)). Clinically, insulin-sensitising drugs such as the thiazolidinediones (TZDs) and metformin have been shown to reduce androgen levels and improve fertility in a proportion of PCOS women. However, the use of TZDs such as rosiglitazone has been contraindicated due to adverse effects including weight gain and edema, and severe side effects such as an increased risk of cardiovascular diseases. Moreover, metformin's action on hyperandrogenism in PCOS subjects is complex and not all women with PCOS respond to metformin with improved ovulation or decreased androgen levels (Barba M, et al., Clin. Endocrinol. (Oxf.) 70(5): 661-670 (2009); Pillai A, et al., J. Fam. Pract. 56(6): 444-453 (2007)). Although in vitro experiments suggests that metformin may inhibit androgen production by regulating steroidogenic enzymes such as HSD3B2 and CYP17A1 , a supraphysiological dose is required (Hirsch A, et al., Endocrinology 153(9): 4354-4366 (2012)). Thus, this effect does not appear to be primarily responsible for the lowering of circulating testosterone in women with PCOS. Rather, the overall improvement of peripheral insulin sensitivity leading to increased liver SHBG production may be the primary mechanism via which androgen bioavailability is reduced (Barba M, et al., Clin. Endocrinol. (Oxf.) 70(5): 661-670 (2009); Jensterle M, et al., J. Clin. Endocrinol. Metab. 99(8): E1476-1481 (2014)). The use of metformin in women with hyperandrogenic PCOS without insulin resistance remains controversial and its efficacy in reducing serum androgen levels in this group of patients has not been confirmed (Jayasena CN, Franks S. Nat. Rev. Endocrinol. 10(10): 624-636 (2014); Nawrocka J, Starczewski A. Gynecol. Endocrinol. Off. J. Int. Soc. Gynecol. Endocrinol. 23(4): 231-237 (2007); Hwang KR, et al., Clin. Exp. Reprod. Med. 40(2): 100-105 (2013)). For this reason, metformin is not recommended for the management of hyperandrogenism in PCOS patients without insulin resistance. Thus, there is a clinical need to identify new molecules with a safer profile to improve the current clinical management of PCOS.

SUMMARY OF THE INVENTION

In a preferred embodiment, the invention provides methods for inhibiting androgen biosynthesis, reducing insulin resistance and/or improving menstrual cycle regularity in a manner dependent or independent of androgen modulation. Glyceryl tridecanoate (1 ,3- di(decanoyloxy)propan-2-yl decanoate), also known by other names such as Tricaprin; Tridecanoin; Glycerol tricaprate; Glycerol tridecanoate; 1 ,2,3-Tridecanoylglycerol and Decanoin, is preferably used for the purpose of the invention.

In a preferred aspect of the invention there is provided a composition for the modulation of androgen levels and/or insulin levels and/or menstrual cycle regularity in a subject, the composition comprising glyceryl tridecanoate (1 ,3-di(decanoyloxy)propan-2-yl decanoate) or an active substituted variant thereof and a pharmaceutically acceptable carrier.

In a preferred embodiment, the composition is for modulation of androgen levels to manage hyperandrogenism. In another preferred embodiment, the composition is for the management of insulin resistance.

In another preferred embodiment, the composition is for the treatment of menstrual cycle irregularity. In another preferred embodiment, the composition is for the treatment of a disease associated with high testosterone levels selected from the group comprising polycystic ovarian syndrome (PCOS), prostate cancer, hirsutism and male pattern baldness.

Another aspect of the invention provides use of glyceryl tridecanoate (1 ,3- di(decanoyloxy)propan-2-yl decanoate) or an active substituted variant thereof for the manufacture of a medicament for the modulation of androgen levels and/or insulin levels and/or improve menstrual cycle regularity in a subject.

In a preferred embodiment, the modulation of androgen levels relates to the management of hyperandrogenism in a subject.

In another preferred embodiment, the medicament is for the modulation of insulin resistance in a subject, which modulation may be dependent or independent of androgen modulation.

In another preferred embodiment, the medicament is for improving menstrual cycle regularity in a subject in a manner dependent or independent of androgen modulation.

In a preferred embodiment, the use is to manufacture a medicament for the treatment of a disease associated with high testosterone selected from the group comprising polycystic ovarian syndrome (PCOS), prostate cancer, hirsutism and male pattern baldness.

In another preferred embodiment, the use is to manufacture a medicament for the modulation of androgen levels to improve the fertility of subjects with PCOS.

Another aspect of the invention provides a method of treatment of an androgen biosynthesis disorder in a subject, said method comprising administering to a human patient in need thereof a pharmacologically effective amount of a composition comprising glyceryl tridecanoate (1,3-di(decanoyloxy)propan-2-yl decanoate) or an active substituted variant thereof and a pharmaceutically acceptable carrier.

In a preferred embodiment, the androgen biosynthesis disorder causes hyperandrogenism.

In another preferred embodiment of the method, the disorder is selected from the group comprising polycystic ovarian syndrome (PCOS), prostate cancer, hirsutism and male pattern baldness. In another preferred embodiment of the method, administration of the composition is initiated upon diagnosis of polycystic ovarian syndrome (PCOS) or prostate cancer in said human patient.

In another preferred embodiment of the method, the modulation of androgen levels reduces serum androgen levels of subjects.

In another preferred embodiment of the method, the treatment reduces serum insulin levels in subjects with insulin resistance independent of its effects on androgen levels.

In a preferred embodiment of the testing method, the hyperandrogenism phenotype is selected from the group comprising polycystic ovarian syndrome (PCOS), prostate cancer, hirsutism and male pattern baldness.

As will be apparent from the following description, preferred embodiments of the present invention allow an optimal use of the isolated glyceryl tridecanoate formulations to take advantage of their ability to inhibit androgen production, lower serum testosterone and/or reduce insulin resistance. This and other related advantages will be apparent to skilled persons from the description below.

BRIEF DESCRIPTION OF THE FIGURES

Figures 1A, 1 B, 1C and 1 D show the effects of decanoic acid on androgen production in human steroidogenic NCI-H295R cells. The concentrations of DHEA, androstenedione, testosterone, and DHT in the culture media were assessed following treatments with increasing doses of DA in the presence of 8Br-cAMP (Fig. 1A), or in the absence (basal-state) of 8Br-cAMP (Fig. 1B). Metformin's effects are in striped bars. Data were mean ± SEM of at least 3 independent experiments and expressed as a percentage of 8Br-cAMP stimulated (Fig. 1A) or basal-state controls (Fig. 1 B). ** ^* P < 0.001 vs respective controls (dotted lines). (Fig. 1C) Dose-response effects of decanoic acid (DA) on the concentrations of androstenedione, testosterone, and DHT levels in the presence of 8Br- cAMP. (Fig. 1D) IC ₅₀ values of decanoic acid in inhibiting androstenedione, testosterone and dihydrotestosterone production in NCI-H295R cells. Concentration of DA at which the respective androgen concentration in the cell media is reduced to 50% of vehicle treated cells is defined as the IC ₅₀ . Figure 2 shows the effects of decanoic acid and metformin on HSD3B2 activity, gene, and protein expression in NCI-H295R cells. (A) HSD3B2 activity: Data were expressed as percentage change in androstenedione/DHEA ratios in the presence of 8Br-cAMP compared to controls (basal-state) (B). (C and D), HSD3B2 gene expression: Total RNA was extracted from cells following exposure to indicated treatments for 48h, and reverse transcribed to cDNA. Relative HSD3B2 gene expression was analyzed using real-time PCR and normalized to 18s rRNA. Metformin effects are in striped bars. Data were mean ± SEM of at least 3 independent experiments and expressed as percentages of (C) 8Br-cAMP-stimulated or (D) basal-state controls. * ^* *P < 0.001 vs respective controls (dotted lines). E and F, HSD3B2 protein expression following the indicated treatments were analysed by Western blotting, β-actin was used as a loading control. Representative blots are depicted.

Figure 3 shows the effects of DA and metformin on Nur77 and SF-1 binding to the HSD3B2 promoter in NCI-H295R cells. (A), Diagram of known binding sites of Nur77 (NBRE) and SF-1 (SF-1/LRH-1 ) on the HSD3B2 promoter. Arrows indicate positions of primers used in the ChIP assay. (B), ChIP assay of Nur77 and SF-1 binding to NBRE or SF-1/LRH-1 on the HSD3B2 promoter following DMSO (basal), 250 μΜ of DA, or 10 mM metformin treatment in the presence of 8Br-cAMP. DNA fragments bound to Nur77 or SF-1 were immunoprecipitated and analyzed by quantitative PCR. Rabbit IgG was used as a control. All data shown are mean ± SEM of at least 3 independent experiments. **P < 0.01 ; ***P < 0.001 vs. 8Br-cAMP stimulated controls

Figures 4A and 4B show the metabolism of glyceryl tridecanoate (GT) to decanoic acid (DA) in vivo. (Fig. 4A) Lipolysis of glyceryl tridecanoate to glycerol and decanoic acid by lipase. (Fig. 4B) Plasma concentrations of decanoic acid over time following a single dose administration of 1 g/kg GT via oral gavage to healthy adult rats. Results are expressed as mean ± SEM (n = 5).

Figure 5 shows the effects of glyceryl tridecanoate and metformin in control and PCOS rats. (A), Free serum testosterone and (B), total serum testosterone in rats before (week 0, white bars) and after treatment (week 5, black bars) with CMC (veh), DA, and metformin (met). (C), Representative IHC staining of HSD3B2 protein (brown) in adrenal glands from control and PCOS rats after 5 weeks of treatment. (D), HSD3B2 stain area was quantified using Image-J software, (n = 9, mean ± SEM) and expressed as fold change compared to Control-Veh. #P < 0.05; ##P < 0.01 ; ###P < 0.001 vs. Control-Veh. *P < 0.05; **P < 0.01 , ***P < 0.001 vs. PCOS-Veh. (E), Western blot analysis of HSD3B2 protein expression in representative adrenals and ovaries from three rats in each group after 5 weeks of treatment.

Figures 6A and 6B show the oestrous cycles of the rats before (Fig. 6A) and after (Fig. 6B) 5 weeks of treatment with CMC (veh), glyceryl tridecanoate (GT, 1g/kg/twice a day), or metformin (met, 350 mg/kg/day).) were charted. Three representative rats from each group (n = 9) are presented. P, pro-oestrus; E, oestrus; M, metoestrus; D, dioestrus. Figures 7A and 7B shows the metabolic consequences of glyceryl tridecanoate and metformin on the PCOS rat model. Fasting blood was assessed for insulin (Fig. 7A) and glucose (Fig. 7B) concentrations after 5 weeks of treatment with CMC (veh), glyceryl tridecanoate (GT, 1 g/kg/twice a day), or metformin (met, 350 mg/kg/day). C, HOMA-IR of rats from each group were calculated. Values are mean ± SEM. ###P < 0.001 vs. control- veh. *P < 0.05; **P < 0.01 vs. PCOS-veh.

DETAILED DESCRIPTION OF THE INVENTION

Bibliographic references mentioned in the present specification are for convenience listed in the form of a list of references and added at the end of the examples. The whole content of such bibliographic references is herein incorporated by reference.

Definitions

For convenience, certain terms employed in the specification, examples and appended claims are collected here.

The term "comprising" is herein defined to be that where the various components, ingredients, or steps, can be conjointly employed in practicing the present invention. Accordingly, the term "comprising" encompasses the more restrictive terms "consisting essentially of and "consisting of."

The term "isolated" is herein defined as a biological component (such as a nucleic acid, fatty acid, peptide or protein) that has been substantially separated, produced apart from, or purified away from other biological components in the cell of the organism in which the component naturally occurs, i.e., other chromosomal and extra-chromosomal DNA and RNA, fatty acids and proteins. Fatty acids which have been isolated thus include fatty acids purified by standard purification methods. The term also embraces fatty acids prepared by recombinant expression in a host cell as well as chemically synthesized fatty acids. The term "variant" as used herein, refers to a compound that is altered by one or more substitutions, but retains the ability to ameliorate the effects of hyperandrogenism, insulin resistance and/or menstrual cycle irregularity in a subject. The variant may have "conservative" changes, where one or more hydrogen atoms of a core structure have been replaced with a functional group like alkyl, hydroxy, or halogen yet retain similar structural or chemical properties.

The term "treatment", as used in the context of the invention refers to prophylactic, ameliorating, therapeutic or curative treatment. The term "subject" is herein defined as vertebrate, particularly mammal, more particularly human. For purposes of research, the subject may particularly be at least one animal model, e.g., a mouse, rat and the like. In particular, for treatment of hyperandrogenism and/or insulin resistance -linked diseases, the subject may be a human with PCOS.

The structure of glyceryl tridecanoate is shown below.

In a preferred aspect of the invention there is provided a composition for the modulation of androgen levels and/or insulin levels and/or menstrual cycle regularity in a subject, the composition comprising glyceryl tridecanoate (1 ,3-di(decanoyloxy)propan-2-yl decanoate) or an active substituted variant thereof and a pharmaceutically acceptable carrier.

In a preferred embodiment, the composition is for modulation of androgen levels to manage hyperandrogenism.

In another preferred embodiment, the modulation of insulin levels relates to the management of insulin resistance. In another preferred embodiment, the composition is for the treatment of menstrual cycle irregularity.

In another preferred embodiment, the composition is for the treatment of a disease associated with high testosterone levels selected from the group comprising polycystic ovarian syndrome (PCOS), prostate cancer, hirsutism and male pattern baldness.

Another aspect of the invention provides use of glyceryl tridecanoate (1 ,3- di(decanoyloxy)propan-2-yl decanoate) or an active substituted variant thereof for the manufacture of a medicament for the modulation of androgen levels and/or insulin levels and/or menstrual cycle regularity in a subject. In a preferred embodiment, the modulation of androgen levels relates to the management of hyperandrogenism in a subject.

In another preferred embodiment, the medicament is for the modulation of insulin resistance in a subject

In another preferred embodiment, the medicament is for improving menstrual cycle regularity in a subject in a manner dependent or independent of androgen modulation.

In a preferred embodiment, the use is to manufacture a medicament for the treatment of a disease associated with high testosterone selected from the group comprising polycystic ovarian syndrome (PCOS), prostate cancer, hirsutism and male pattern baldness.

In another preferred embodiment, the use is for the modulation of androgen levels to improve the fertility of subjects with PCOS.

Another aspect of the invention provides a method of treatment of an androgen biosynthesis disorder in a subject, said method comprising administering to a human patient in need thereof a pharmacologically effective amount of a composition comprising glyceryl tridecanoate (1 ,3-di(decanoyloxy)propan-2-yl decanoate) or an active substituted variant thereof and a pharmaceutically acceptable carrier.

In a preferred embodiment, the androgen biosynthesis disorder causes hyperandrogenism.

In another preferred embodiment of the method, the disorder is selected from the group comprising polycystic ovarian syndrome (PCOS), prostate cancer, hirsutism, male pattern baldness, insulin resistance and menstrual cycle irregularity. In another preferred embodiment of the method, administration of the composition is initiated upon diagnosis of polycystic ovarian syndrome (PCOS) or prostate cancer in said human patient.

In another preferred embodiment of the method, the modulation of androgen levels reduces serum androgen levels of subjects.

In another preferred embodiment of the method, the treatment reduces serum insulin levels in subjects with insulin resistance.

Since glyceryl tridecanoate is a compound obtained from natural food sources, a larger dose than may be required other compounds may be required for efficacy. One way to circumvent this problem is to deliver the required amounts in sachets, which can be mixed with drinks such as milk or water.

Having now generally described the invention, the same will be more readily understood through reference to the following examples which are provided by way of illustration, and are not intended to be limiting of the present invention.

A person skilled in the art will appreciate that the present invention may be practised without undue experimentation according to the method given herein. The methods, techniques and chemicals are as described in the references given or from protocols in standard biotechnology and pharmaceutical text books.

EXAMPLES

Standard molecular biology techniques known in the art and not specifically described were generally followed as described in Sambrook and Russel, Molecular Cloning: A Laboratory Manual, Cold Springs Harbor Laboratory, New York (2001 ).

Glyceryl tridecanoate is a naturally occurring triglyceride has been shown to improve insulin sensitivity and lipid profile in an in-vivo model of diabetes without the induction of weight gain or increase in adipogenesis.

NCI-H295R human adrenocortical cells were used to assess the effect of decanoic acid (fatty acid of glyceryl tridecanoate) on gene and protein expression of the steroidogenic enzymes. In addition, an in-vivo letrozole-induced PCOS rat model was used to assess if glycerol tridecanoate, like the other insulin sensitising agents, may also restore normal androgen production and alleviate the symptoms associated with PCOS and/or directly modulate androgen biosynthesis. Materials and Methods

8Br-cAMP was purchased from Santa Cruz Biotechnology (CA, USA). DA was obtained from Alfa Aesar (Massachusetts, USA). Metformin was purchased from Sigma (St. Louis, MO). All chemicals for LC-MS and GC-MS were obtained from Sigma, unless mentioned otherwise. Cell culture

The NCI-H295R human adrenocortical cell line (ATCC, CRL-2128; Manassas, USA) was chosen because it is a well-established model for studying steroidogenesis (Hirsch A, et al., Endocrinology 153(9): 4354-66 (2012)). Cells were cultured in DMEM/F-12 medium containing L-glutamine and 15 mM HEPES (GIBCO, UK), supplemented with 2.5% NU-I serum and 0.1% selenium/insulin/transferrin mix (Becton Dickinson, Switzerland). The serum-free medium contained only DMEM/F-12.

For steroid production, RNA, and protein extraction experiments, cells were plated at a density of 8 x 10 ⁵ cells/well in six-well plates. After subculturing for 48 hours, cells were treated with DMSO (vehicle), DA at the doses indicated, or 10 mM metformin in serum-free medium for 48 hours, either alone or in combination with 0.5 mM 8Br-cAMP. Conditioned culture media were collected and stored at -80°C for subsequent androgen analysis using LC-MS. Androgen concentration data was normalized to DNA content of the respective wells at the end of the experiment (reflecting the cell number).

Androgen measurements Dehydroepiandrosterone (DHEA), Androstenedione (A4), testosterone (T), and DHT in the cell culture media was determined by liquid chromatography-mass spectrometry (LC- MS). Briefly, 10 pL of d ₃ -T (10 nM in methanol) was added into 100 pL of cell culture media or calibration standards. 2mL of Methyl tert-butyl ether was then added to each tube and vortexed. After centrifugation for 3 minutes at 13,000 rpm, the upper organic layer was collected and evaporated under nitrogen gas in a 45°C water bath. For derivatisation, 300 pL of 0.1 M hydroxylamine chloride was added to the mixture. After incubation at 70°C for 15 minutes, the reaction mixture was centrifuged at 16,200 rpm at 4°C for 3 minutes. Finally, the samples were transferred into vials for LC-MS analysis.

LC/MSMS analysis was carried out on Agilent Technologies 6490 Triple Quadrupole combined with the 1290 Infinity LC system. Chromatographic and mass spectrometric data were recorded and processed by Agilent Technologies MassHunter Workstation Software LC/MS Data Acquisition for 6400 Series Triple Quadrupole (Version B.06.00 Build 60.6025.0). The separation was performed using a ZORBAX Eclipse Plus C18 column (Rapid Resolution HD 2.1 x 50 mm 1.8-Micron) coupled with a guard column at 40°C column temperature. Autosampler temperature was 8°C and injection volume set at 20 μΙ_. Chromatographic separation was carried out using gradient elution of mobile phases A: 0.1% formic acid, and B: acetonitrile delivered at a flow rate of 0.5 mL/min. Composition of B was increased from 20% to 40% within 30s, from 40% to 60% in 1 minute, and maintained for 1 minute. Thereafter, it was further increased to 100% within 30s and maintained for 1 minute. Finally, it was decreased to 20% within a minute and maintained for 2 minutes. Total run time was 7 minutes. During each run, from 0 to 0.8 minute, the inlet of the MS was switched to waste (bypassing the MS) to elute salts and polar impurities. From 0.8 to 5 minutes, the inlet was switched to the MS and the analytes were eluted and detected by the MS. From 5 to 7 minutes, the inlet was switched back to waste.

MS was carried out using electrospray ionization (ESI) under the positive mode. Multiple reactions monitoring (MRM) was applied for the quantification. Quantifier MRM transitions used were m/z 304.2 to 253.2 for DHEA-oxime, m/z 317.2 to 112.1 for A4-oxime, m/z 306.3 to 105.2 for DHT-oxime, m/z 304.2 to 124.0 for T-oxime, and m/z 307.2 to 124.0 for internal standard d ₃ -T-oxime. Qualifier MRM transition used were m/z 304.2 to 213.2 for DHEA-oxime, m/z 317.2 to 124.1 for A4-oxime, m/z 306.3 to 107.0 and m/z 347.3 to 306.0 for DHT-oxime, m/z 304.2 to 112.1 for T-oxime, and m/z 307.2 to 112.1 for d ₃ -T-oxime.

Real-time PCR (RT-PCR)

Total RNA was extracted using the RNeasy Mini Kit (Qiagen, Valencia, CA), and reverse transcribed to cDNA using the High Capacity cDNA Reverse Transcription Kit (Invitrogen, USA). Real-time PCR was performed with the StepOnePlus Real- Time PCR System using specific primers from TaqMan® Gene Expression Assays (Applied Biosystem, USA). Relative gene expression values were determined by the 2 ^{" Ct} method with 18S rRNA as the reference gene.

Western blotting

Cells or animal tissues were lysed with RIPA buffer (Thermoscientific, USA) containing a protease inhibitor cocktail (Roche, USA). 20 μg of protein lysate was resolved on a 10% SDS-PAGE gel and transferred onto a nitrocellulose membrane using the semidry transfer method. After blocking with 5% non-fat milk (Bio-Rad), the membrane was incubated with HSD3B2 (15516-1-AP, Proteintech) or β-actin (A5316, Sigma) primary antibodies overnight at 4°C. Secondary antibodies used were anti-rabbit or anti-mouse conjugated to horseradish peroxidase (Dako, USA). Protein bands were detected using Amersham ECL (GE Healthcare) exposed to CL-XPosure Films (Thermo scientific, USA), and developed using an X-Ray Developer (Carestream Health, USA).

Chromatin immunoprecipitation (ChIP)

For ChIP assays, NCI-H295R cells were grown to 80% confluence in a T-175 flask. The culture media was replaced with serum-free media for 48 hours before treatment. Cells were then treated with DMSO (basal), 0.5 mM 8Br-cAMP, 0.5 mM 8Br-cAMP and 250 μΜ DA, or 0.5 mM 8Br-cAMP and 10 mM metformin in serum-free media for 24 hours. The cells were processed for ChIP assay as described previously (Sun F, et al., Carcinogenesis (2015) doi:10.1093/carcin/bgv040). Briefly, formaldehyde (final concentration, 1%) was added for 10 minutes for cross-linking. Nuclei were isolated using Triton X-100 lysis buffer and sonicated for 15 cycles (1 cycle: 30s sonicate, 30s rest) using the Bioruptor Plus (Diagenode, Liege, Belgium). Precleared chromatin was immunoprecipitated with protein A-agarose and either 5 g of anti-Nur77 antibody (Santa Cruz Biotechnology), anti-SF-1 antibody (Millipore), or control rabbit IgG (Santa Cruz Biotechnology) overnight at 4 °C on a rotating wheel. Beads were then washed and DNA fragments were eluted with elution buffer (1% SDS, 0.1 M NaHC0 ₃ ). After reverse cross-linking at 65 °C overnight, DNA was quantified by qPCR using the following primers: forward primer 5'- CACACTGTGGCCTTAAGATTGG -3' (SEQ ID NO: 1 ), reverse primer 5'- TCCTCCAGAATCCCTTTTCTCA -3' (NBRE site of HSD3B2) (SEQ ID NO: 2), and forward primer 5'- TTCTGGAGGAGGAGGGAGC -3' (SEQ ID NO: 3), reverse primer 5'- GGAGGAAGGACTGGAGCTTT -3' (SF-1/LRH-1 site of HSD3B2) (SEQ ID NO: 4).

Pharmacokinetics

A single dose of 1g/kg of glyceryl tridecanoate (GT) (Tokyo Chemicals Industry, Japan) was administered to 5 Wistar rats by oral gavage. Blood samples were taken from the jugular vein at 0, 0.17, 0.5, 1.5, 3.5, 5, 6.5, 9, 12, and 24 hours. Plasma was collected and stored at -80°C until the analysis of DA concentration using GC-MS. Peak plasma concentration (C _max ) and time at which C _max was observed (T _max ) of DA were assessed using non-compartmental analysis with WinNonLin 6.2.1 (Pharsight, USA).

Determination of DA concentration in blood plasma

DA concentrations in the plasma samples were determined by gas chromatography- mass spectrometry (GC-MS) as reported previously with slight modifications (Malapaka RR, et al. The Journal of Biological Chemistry 287(1 ): 183-95 (2012)). 50 μΙ_ of internal standard, d ₃₃ -heptadecanoic acid (0.8 mg/ml in hexane), was added to 100 μΙ_ of blood plasma sample or calibration standard containing DA (40, 50, 60, 100, 150, 200, 400, 600, 800 μΜ) in a matrix of rat serum. 2 mL of the catalyst: methanol/acetyl chloride (20:1 v/v) was added to each sample. After sonication for 5 minutes, the sample was placed on a shaker for 10 minutes, followed by incubation at 70°C for 90 minutes. After cooling to room temperature, 5ml_ of 6% K ₂ C0 ₃ was added to the mixture to quench the esterification reaction. Next, 0.4 ml of hexane was added to the mixture and placed on a shaker for 1 minute. The upper organic layer containing the fatty acid methyl esters was collected and centrifuged at 14,000 rpm for 3 minutes. Finally, the samples were transferred into vials for GC-MS analysis.

GC-MS analysis of DA in rat serums was performed on an Agilent 6890NGC coupled to a 5975 inert XL mass selective detector (Agilent Technologies, Palo Alto, CA). An Omegawax-320 (Supelco, Bellefonte, PA) capillary column (ID: 30 m x 0.32 mm) was used for the separation of the methyl esters of the fatty acids. Purified helium gas (99.9999%, Soxal, Singapore) was used as the carrier gas at a constant column flow rate of 3 mL/min. Standards and samples were injected using an Agilent 7683B series autosampler. Injection volume was 2 μΙ_, and solvent delay was 3 minutes. Initial oven temperature was set at 120°C, held for 3 minutes, then increased to 180°C at 5°C/min and maintained for 2 minutes. This was followed by another increase to 230°C at 2°C/min, and maintained for 10 minutes. The front inlet and mass selective detector transfer line temperatures were kept at 280 and 300°C respectively. The mass selective detector was operated in selected ion monitoring mode with a dwell time of 100 milliseconds for the ions m/z 143 and m/z 155 for the methyl esters of DA and m/z 317 for d ₃₃ -heptadecanoic acid. The EMV voltage of the MS source was maintained at 200V. The DA concentrations of the samples were quantified using a 9- point calibration curve of peak area ratio (y-axis) for the analytes to internal standard d ₃₃ - heptadecanoic acid against the concentration ratio (x-axis) for the analytes to internal standard. Animal Study

The letrozole-induced PCOS rat model was used as it encompasses both the endocrine and metabolic abnormalities of PCOS, with clear manifestations of endogenous hyperandrogenism, estrous acyclicity, insulin resistance, and obesity (Maliqueo M, Endocrinology 154(1 ): 434-45 (2013)). At 3 weeks of age, female Wistar rats were randomly divided into 5 groups (n = 9 per group): Control-Veh, Control-DA, PCOS-Veh, PCOS-DA, PCOS-Met. Rats in the PCOS groups were subcutaneously implanted with a 90-days continuous-release pellet (Innovative Research of America, FL) containing 18 mg (daily dose, 200pg) of letrozole (Tokyo Chemicals Industry, Japan); controls were subcutaneously implanted with a placebo pellet. 5 weeks after pellet implantation, the rats were treated with vehicle, DA or metformin for 5 weeks. Rats in Control-Veh and PCOS-Veh groups were given 1% carboxymethyl cellulose (CMC) (4 mL/kg/d). Control-DA and PCOS-DA groups were administered glyceryl tridecanoate (2 g/kg/d), given as 1 g/kg twice a day. PCOS-Met group was given metformin (350 mg/kg/d). All interventions were given via oral gavage. The study was concluded at 13 weeks of age, and the rats were euthanized. Ovaries and adrenals were excised and snap frozen for protein analyses. The study was reviewed and approved by the Animal Ethics Committee at the National University of Singapore.

Vaginal smears

Vaginal smears were taken between 10:00-11 :00 hours each day, for 10 consecutive days before and after treatment. The tip of a moistened cotton bud swab was inserted approximately 1 cm into the vagina, at a 45° angle to the rat's body. With a rotation of the swab, the cells from the vaginal lumen and walls are gently removed, and transferred onto a glass slide. The stage of estrous cycle of each rat was determined by microscopic analysis of the predominant cell type in the vaginal smears (Marcondes FK, et al., Braz. J. Biol. 62(4A): 609-614 (2002)). Blood sampling and parameters

Tail blood was obtained after an overnight fast to assess testosterone, glucose, and insulin concentrations in the serum. In cycling rats, blood samples were obtained in the diestrous phase. Serum was collected and stored at -80°C. Serum testosterone and insulin concentration was measured using an ELISA kit according to the manufacturer's instructions (Testosterone: Enzo; Insulin: Crystal Chem). The intra- and inter-assay CV and sensitivity were: 7.2%, 9.8% and 5.34 pg/mL for testosterone; 2.5%, 8.3%, and 100 pg/ml for insulin. Glucose was assessed using the Cobas c111 Analyzer (Roche), and total CV was <4%. Homeostatic model of insulin resistance (HOMA-IR) = [fasting glucose (mmol/liter) x fasting insulin (mU/liter)]/22.5. Immunohistochemistrv (IHC)

After the rats were euthanized, left adrenal gland was excised, fixed, and paraffin embedded. 5 μΜ sections were mounted on a glass slide, and anti-HSD3B2 (15516-1 -AP, Proteintech) diluted at 1 :200 was used for IHC analysis. After paraffin embedding, specimens were serially sectioned and stained with hematoxylin and eosin. The primary antibody for HSD3B2 (15516-1 -AP, Proteintech) was used for IHC staining on formalin-fixed 5 m thick slides using a Bond Max autostainer (Leica Microsystems). For IHC quantification, representative adrenal gland sections from 9 rats in each group were captured using a bright field microscope (Nikon Instruments). The image analysis software, ImageJ (Maryland, USA), was then used to analyze the photographs of HSD3B2 staining and expressed as HSD3B2 stain area.

Statistical analyses

Results are expressed as mean ± SEM. Statistical analyses were performed using Prism 6 (GraphPad Software, Inc., San Diego, CA). Statistical differences were assessed by Student's t-test or one-way ANOVA followed by Tukey's post hoc analysis where appropriate. P < 0.05 was considered significant.

Results

DA increased DHEA, but inhibited androstenedione, testosterone, and DHT production in human steroidogenic cells

We examined the effects of DA on androgen production in the human NCI-H295R cell line, an established model of steroidogenesis (Hirsch A, et al., Endocrinology 153(9): 4354-66 (2012); Kempna P, et al., Mol Pharmacol 71 (3): 787-98 (2007)). To mimic the PCOS condition wherein chronically elevated LH stimulation is present (Maliqueo M, et al., Endocrinology 154(1 ): 434-45 (2013)), we performed experiments in the presence and absence of the 8Br-cAMP, simulating the effects of higher levels of the LH intracellular second messenger, cAMP. Conditioned culture media were analyzed for the androgens: DHEA, androstenedione, testosterone and DHT. Under 8Br-cAMP-stimulated conditions, DA increased DHEA concentration in a dose-dependent manner (Fig. 1A, top panel). In contrast, DA reduced androstenedione, testosterone and DHT at doses >100 μΜ (Fig. 1A). At the maximal dose of DA (250 μΜ), levels of androstenedione, testosterone and DHT were reduced by 44.3%, 57.2% and 66.2% compared to controls exposed to 8Br-cA P alone (P < 0.001 ). Similar effects were observed for metformin, wherein raised DHEA levels and decreased androstenedione, testosterone, and DHT levels in the culture media were observed (Fig. 1A, striped bars). In the absence of 8Br-cAMP stimulation (basal-state), DA and metformin did not affect the production of DHEA, androstenedione, and testosterone (Fig. 1 B). Under 8Br-cAMP-stimulated conditions, inhibition-response curves of DA indicate that IC ₅₀ for androstenedione, testosterone, and DHT production were 199.1 μΜ, 123.8 μΜ, and 92.5 μΜ respectively (Fig. 1 C and 1 D). The inhibitory effects on androstenedione, testosterone, and DHT production observed in 8Br-cAMP-stimulated, but not basal-state conditions, suggest the utility of DA for treating PCOS, a condition wherein increased LH/cAMP stimulation leads to excess androgen production. DA inhibited the bioactivity, gene and protein expression of HSD3B2

Production of androgens is dependent on the activity of an ordered cascade of steroidogenic enzymes. Increased DHEA coupled with reduced androstenedione suggested reduced activity of the HSD3B2 enzyme. Indeed DA dose-dependently suppressed androstenedione/DHEA ratios. DA at doses of 100 μΜ and 250 μΜ reduced androstenedione/DHEA ratios by 37% and 63% respectively compared to controls (Fig. 2A). Metformin also reduced androstenedione/DHEA ratios by 67% (Fig. 2A, striped bars). In comparison, DA and metformin did not change androstenedione/DHEA ratios in the basal state (Fig. 2B). Quantitative RT-PCR revealed a profound inhibition of 3 beta- hydroxysteroid dehydrogenase/delta(5)-delta(4) isomerase type II (HSD3B2) mRNA expression with DA. At the maximal dose, DA and metformin suppressed HSD3B2 gene expression by 98.6% and 94.7% compared to controls (P < 0.001 ) (Fig. 2C). In contrast, both DA and metformin did not influence mRNA expression of other steroidogenic enzymes (StAR, CYP1 1A1 , and CYP17A1 ) in the androgenic pathway (data not shown). DA did not alter HSD3B2 gene expression in the basal state (Fig. 2D). Western blot analyses revealed a dose-dependent inhibition of the 8Br-cAMP enhanced HSD3B2 protein expression by DA doses >100 μΜ (Fig. 2E). A similar effect was observed for metformin. Conversely in the absence of 8Br-cAMP stimulation, DA did not significantly alter HSD3B2 protein expression (Fig. 2F).

Effects of DA were PPARy-independent, but associated with reduction of Nur77 transcription factor binding to the HSD3B2 promoter

To investigate whether DA's effects on HSD3B2 expression is PPARv-dependent, we co-incubated DA with the PPARy antagonist, GW9662. Inhibitory effects of DA on HSD3B2 remained unchanged in the presence of PPARy antagonist (data not shown), indicating that inhibition of HSD3B2 gene and protein expression by DA was PPARy-independent. The HSD3B2 gene has been reported to be regulated by the transcription factors

Nur77 and SF-1 (Udhane S, et al., PloS One 8(7): e68691 (2013)), through their activity on the NBRE and SF-1/LRH-1 domains of the HSD3B2 promoter (Fig. 3A). To ascertain whether transcription factor recruitment to the HSD3B2 promoter is affected by DA, we performed ChIP assays followed by quantitative RT-PCR. 8Br-cAMP significantly increased the levels of Nur77 and SF-1 binding to the promoter region of HSD3B2 (Fig. 3B). However, cAMP-induced recruitment of Nur77 to the NBRE region was suppressed by DA and metformin treatment, indicating that their inhibitory action on HSD3B2 gene expression may be due to reduced recruitment of Nur77 to the HSD3B2 promoter (Fig. 3B, left panel). Isothermal titration calorimetry binding assays did not reveal any direct interaction between purified Nur77 ligand-binding domain with DA and metformin (data not shown). Notably, cAMP-induced SF-1 recruitment to the HSD3B2 promoter was not affected by either DA or metformin treatments (Fig. 3B, right panel).

DA reduced circulating serum testosterone in a PCOS rat model

We next investigated whether DA can inhibit androgen production in animal models of PCOS. DA in its native state can irritate the mucous membranes of the gastrointestinal tract, upper respiratory tract, eyes, and skin. However its triglyceride derivative, glyceryl tridecanoate (GT), which is found to be abundant in coconut oil is non-toxic (Elson CE. Crit Rev Food Sci Nutr 31 (1-2): 79-102 (1992)). When ingested orally, GT is broken down by pancreatic lipase resulting in the release and absorption of DA in the intestines (Fig. 4A)(Bach AC, Babayan VK. Am J Clin Nutr 36(5): 950-62 (1982)). Oral administration of GT to Wistar rats induced a rapid rise in DA concentration in the blood plasma (Fig. 4B). DA concentration in the plasma peaked at 723.5 ± 0.7 μΜ (C _max ), 4.1 ± 0.3 hours (T _max ) post administration. These plasma levels of DA were well within levels of DA shown to inhibit excess androgen production (Fig. 1 C). To profile the effects of DA and metformin in vivo, we used a letrozole-induced PCOS rat model. Letrozole, an aromatase inhibitor, increases circulating androgens and induces many features of PCOS, such as endogenous hyperandrogenism, acyclicity, and insulin resistance, when administered to female Wistar rats (Maliqueo M, et al., Endocrinology 154(1 ): 434-45 (2013)). 3 week old rats were subcutaneously implanted with continuous- release letrozole pellets (PCOS group), whilst control rats had placebo pellets. After 5 weeks of letrozole or placebo pellet implantation (treatment week 0), serum free testosterone (Fig. 5A, white bars) and total testosterone (Fig. 5B, white bars) were higher in the PCOS group compared to the control group (P < 0.01 ). Rats in the PCOS group were also heavier, acyclic (Figure. 6) and had higher levels of fasting insulin (Figure. 7) compared to controls. Subsequently, rats in both the control and PCOS groups were administered DA, metformin, or 1 % CMC (veh) by oral gavage for 5 weeks.

After 5 weeks of respective treatments (treatment week 5), blood was obtained for androgen analyses and tissues examined for HSD3B2 expression. When free testosterone was measured, both PCOS-DA and PCOS-Met groups had significantly lower serum free testosterone compared to the PCOS-Veh group (Fig. 5A, black bars). PCOS-DA, but not PCOS-Met group, had significantly lower serum total testosterone compared to the PCOS- Veh group (Fig. 5B, black bars). It is interesting to note that no changes in the serum levels of free or total testosterone levels were observed in the control group of rats given DA treatment (Fig. 5A, B; black bars). Notably HSD3B2 expression was also lower in the ovaries and adrenal glands of the PCOS rats treated with GT. DA treatment reduced HSD3B2 protein expression in the adrenals and ovaries of PCOS rats

As we have previously observed an inhibition of HSD3B2 protein expression by DA in adrenocortical NCI-H295R cells, we sought to determine the HSD3B2 protein expression in the rat adrenal gland. IHC staining (Fig. 5C) and quantification (Fig. 5D) for HSD3B2 protein expression were performed. Rats in the PCOS-Veh group had increased HSD3B2 protein expression compared to rats in the control group (P < 0.001 ) (Fig. 5C, D). Rats in the PCOS- DA group had significantly reduced HSD3B2 protein expression compared to the PCOS-Veh group (P < 0.01 ) (Fig. 5D). In contrast, there were no differences in HSD3B2 protein expression between PCOS-Met compared to PCOS-Veh group. Control-DA rats had similar HSD3B2 protein expression compared to Control-Veh rats. IHC findings were corroborated by western blot analyses, which demonstrated that only the PCOS-DA group had reduced HSD3B2 protein expression in both the adrenals and ovaries of PCOS rats (Fig. 5E).

DA treatment restored estrous cyclicity, lowered serum fasting insulin concentration and HOMA-IR, without toxicological effects

At baseline (week 0 of treatment), all rats in the control group (n=18) had a normal estrous cycle of 4-5 days, while the rats in the PCOS group (n=27) were acyclic (Fig. 6A). After 5 weeks of the respective treatments, all rats in the PCOS-Veh (n=9) group remained acyclic, while rats in the PCOS-DA (n=9) and PCOS-Met (n=9) groups recovered their 4-5 days estrous cyclicity (Fig. 6B). Rats in the Control-DA group (n=9) continued to have a normal estrous cycle, similar to Control-Veh (n=9) group of rats.

Since androgens and insulin resistance have been reported to be involved in the pathogenesis of PCOS, we investigated the metabolic consequences of DA on the PCOS rat model. PCOS-Veh group of rats had increased concentrations of fasting insulin and HOMA- IR compared to the control group, suggesting insulin resistance (P < 0.001 ) (Fig. 7A, C). After 5 weeks of treatment, both PCOS-DA and PCOS-Met groups of rats had significantly lower fasting insulin concentration and HOMA-IR compared to the PCOS-Veh group (Fig. 7A, C). Fasting glucose concentrations remained unchanged across all groups (Fig. 7B). Three weeks after pellet implantation, the PCOS group of rats gained more weight than the controls (P < 0.01 ; data not shown). It is interesting to note that body weight was not significantly different between Control-DA and Control-Veh groups throughout the study period. Similarly, PCOS-DA and PCOS-Met groups had comparable body weights with the PCOS-Veh group throughout the study. After five weeks of treatment, clinical chemistry, blood count, and hematocrit analysis revealed no toxicological effects in the hematological, hepatic and renal systems in rats treated with DA (data not shown). Discussion

To our knowledge this is the first report that DA, a ubiquitous fatty acid of dietary origin, can act as a potent inhibitor of androgen biosynthesis. Reduced Δ4 androgen production was associated with increased levels of the Δ5 precursor DHEA, implicating HSD3B2 as the target for DA. Indeed, DA profoundly repressed gene expression of HSD3B2, but not other enzymes such as StAR, CYP11A1 , and CYP17A1 in the androgenic pathway. ChIP studies indicate that repression of cAMP-stimulated gene and protein expression of HSD3B2 by DA may be associated with the inhibition of Nur77 recruitment to the HSD3B2 promoter. In a hyperandrogenic PCOS rat model, DA treatment suppressed serum total and free testosterone. Inhibition of androgen production by DA was associated with restoration of estrous cyclicity, and lowering of fasting serum insulin and HOMA-IR. Like the cellular model, reduction of serum testosterone by DA was associated with repression of HSD3B2 protein expression in the adrenal glands and ovaries of the PCOS rats. The ability of DA to reverse the endocrine and metabolic aberrations of PCOS was not associated with any discernable toxicological effects.

Physiological regulation of androgen production is controlled by LH and ACTH, both of which act through the second messenger cAMP to promote the transcription of steroidogenic enzymes, thus increasing androgen biosynthesis. In women with PCOS, LH levels are elevated with increase in pulse frequency and amplitude, thereby resulting in augmented ovarian androgen production (Ehrmann DA. N. Engl. J. Med. 352(12): 1223- 1236 (2005)). It is interesting to note that inhibitory effects of DA and metformin on HSD3B2 and androgen production were observed only in cells stimulated with 8Br-cAMP, but not in the unstimulated basal state, consistent with the increased LH/cAMP drive in the ovarian theca/granulosa cells of hyperandrogenic PCOS subjects (Hillier SG, et al., J. Clin. Endocrinol. Metab. 72(6): 1206-121 1 (1991 ); Yong EL, Baird DT, Hillier SG, Clin. Endocrinol. (Oxf.) 37(1 ): 51-58 (1992)), and suggestive of mechanistic action via the cAMP-stimulated pathway. Notably, gene expression of other enzymes (such as CYP17A1 ) in the androgenic pathway were not affected.

Our data reveal a novel mechanism where DA and metformin inhibit recruitment of the orphan nuclear receptor, Nur77, to the HSD3B2 promoter. Factors known to regulate expression of HSD3B2 include the nuclear receptors SF-1 and Nur77. Nur77 binding to the NBRE site (-130 bp) of the HSD3B2 promoter has been shown to be crucial for cAMP stimulation of the HSD3B2 gene expression, while SF-1 binding to the SF-1/LRH-1 site (-64 bp) was less essential (Udhane S, et al., PloS One 8(7): e68691 (2013)). Here we demonstrate that DA and metformin reduced Nur77 transcription factor recruitment to the HSD3B2 promoter, an effect that is likely to impair cAMP stimulation of HSD3B2 transcription. The mechanism(s) by which DA and metformin prevented recruitment of Nur77 to the HSD3B2 promoter remains elusive. Recent findings suggest that small molecules can bind to the ligand-binding domain of Nur77 to regulate its transactivational activity (Zhan Y, et al., Nat. Chem. Biol. 8(1 1 ): 897-904 (2012)). However, we were unable to detect binding of DA or metformin to the Nur77 ligand-binding domain using isothermal titration calorimetry. A challenge for the future is to define the molecular mechanism wherein DA and metformin inhibit binding of Nur77 to the HSD3B2 promoter.

In concordance with our in vitro data, treatment of hyperandrogenic PCOS rats with DA, delivered by oral administration of the precursor triglyceride GT, reduced serum testosterone and HSD3B2 expression to normal levels. We were pleasantly surprised that DA treatment did not alter serum testosterone levels in control rats with normal testosterone levels, an observation that parallels DA's lack of effect on androgen production and HSD3B2 expression in NCI-H295R cells in the unstimulated basal state. We utilized the letrozole- induced PCOS rat model as it closely recapitulates the human syndrome especially in terms of raised LH and testosterone in association with disrupted estrous cyclicity (Maliqueo M, et al., Endocrinology 154(1 ): 434-445 (2013)). While DA treatment significantly reduced both total and free serum testosterone in PCOS rats, metformin was observed to only reduce free testosterone levels. Hypoandrogenic effects of metformin on free, but not total, testosterone could be due to increased SHBG levels induced by metformin (Barba M, et al., Clin. Endocrinol. (Oxf.) 70(5): 661-670 (2009); Jensterle M, Kocjan T, Janez A, J. Clin. Endocrinol. Metab. 99(8): E1476-1481 (2014)), since increased binding of testosterone by SHBG will reduce the levels of free bioavailable testosterone. Nevertheless, beneficial effects of normalizing testosterone levels were evident from the improvement of estrous cyclicity in both DA and metformin treated PCOS rats.

Lowered serum total testosterone with DA treatment was associated with reduced HSD3B2 protein expression in the adrenal gland and ovaries of the PCOS rats, suggesting a causal suppressive effect on endogenous testosterone biosynthesis. HSD3B2 is the primary isoform found in the adrenals and gonads and is responsible for androgen biosynthesis in these glands. Increased expression of HSD3B2 protein is present after letrozole exposure (Zurvarra FM, et al., Reprod. Fertil. Dev. 21(7): 827-839 (2009)), and DA was able to reverse the HSD3B2 protein abnormality in the adrenal glands and ovaries of PCOS rats. Considering that increased expression and activity of HSD3B2 has been consistently found in hyperandrogenic PCOS women (Medeiros SF de, Barbosa JS, Yamamoto MMW, J. Obstet. Gynaecol. Res.41 (2): 254-263 (2015)), DA may have a beneficial therapeutic profile in this group of patients. As expected, metformin treatment of the PCOS rats lowered fasting insulin and HOMA-IR. The insulin sensitizing effects of metformin are well-established. In DHEA- induced PCOS rodent models, metformin treatment decreased serum insulin and HOMA-IR without any effect on body weight. Although a different model was used in our study, our findings that metformin treatment lowered fasting insulin and HOMA-IR of the letrozole- induced PCOS rats with no change in weight are consistent with these earlier reports. More remarkably, DA treatment also improved these metabolic parameters of the PCOS rats. In addition to its positive effects on endocrine parameters, DA treatment lowered fasting insulin and HOMA-IR to a similar extent as metformin. Although HOMA-IR is widely accepted as a surrogate measure of systemic insulin resistance, using the human constant of 22.5 to determine HOMA-IR for rodents may have its limitations. We have demonstrated previously that DA can improve glucose tolerance and lipid profile in a diabetic db/db mouse model, due to its binding and partial activation of PPARy. Reduction in fasting insulin and HOMA-IR observed in the letrozole-induced PCOS rat model could be attributed to the partial activation of the PPARy pathway by DA. Since hyperandrogenism and hyperinsulinemia are interlinked (Holmang A, et al., Am. J. Physiol. 259(4 Pt 1 ): E555-560 (1990); Holmang A, et al., Am. J. Physiol. 262(6 Pt 1 ): E851-855 (1992); Rincon J, et al., Diabetes 45(5): 615-621 (1996); Holmang A, et al., Acta Physiol. Scand. 171 (4): 427-438 (2001 ); Corbould A. J. Endocrinol. 192(3): 585-594 (2007)), reduction of serum testosterone by DA could also contribute to the lower fasting insulin levels observed in the PCOS rats. Considering the high prevalence of insulin resistance and hyperinsulinemia amongst women with hyperandrogenic PCOS, the dual effect of DA in ameliorating hyperandrogenism and reducing fasting insulin levels are promising.

Advantageously, the use of glyceryl tridecanoate to ameliorate hyperandrogenism and reduce fasting insulin levels will provide a better safety profile with reduced side-effects than the use of insulin-sensitising drugs such as the thiazolidinediones (TZDs) and metformin. These results indicate that inhibition of androgen production will be useful in the clinical management of hyperandrogenism, such as diseases associated with high testosterone, for example, polycystic ovarian syndrome, prostate cancer; and the aesthetic medicinal use for hirsutism (unwanted hair growth) and male pattern baldness (wanted hair growth . References

Bach AC, Babayan VK. Medium-chain triglycerides: an update. Am J Clin Nutr 36(5): 950-62 (1982).

Barba M, Schiinemann HJ, Sperati F, Akl EA, Musicco F, Guyatt G, Muti P. The effects of metformin on endogenous androgens and SHBG in women: a systematic review and meta-analysis. Clin. Endocrinol. (Oxf.) 2009; 70(5): 661-670.

Consensus on infertility treatment related to polycystic ovary syndrome. Hum. Reprod. 2008; 23(3): 462-477.

Corbould A. Chronic testosterone treatment induces selective insulin resistance in subcutaneous adipocytes of women. J. Endocrinol. 2007; 192(3): 585-594.

Elson CE. Tropical oils: nutritional and scientific issues. Crit Rev Food Sci Nutr 31(1 - 2): 79-102 (1992).

Ehrmann DA. Polycystic Ovary Syndrome. N. Engl. J. Med. 2005; 352(12): 1223-

1236.

Hirsch A, Hahn D, Kempna P, Hofer G, Nuoffer JM, Mullis PE, et al., Metformin

Inhibits Human Androgen Production by Regulating Steroidogenic Enzymes HSD3B2 and CYP17A1 and Complex I Activity of the Respiratory Chain. Endocrinology 153(9): 4354-66 (2012).

Hillier SG, Yong EL, lllingworth PJ, Baird DT, Schwall RH, Mason AJ. Effect of recombinant activin on androgen synthesis in cultured human thecal cells. J. Clin. Endocrinol. Metab. 1991 ; 72(6): 1206-121 1.

Holmang A, Svedberg J, Jennische E, Bjorntorp P. Effects of testosterone on muscle insulin sensitivity and morphology in female rats. Am. J. Physiol. 1990; 259(4 Pt 1 ): E555- 560.

Holmang A, Larsson BM, Brzezinska Z, Bjorntorp P. Effects of short-term testosterone exposure on insulin sensitivity of muscles in female rats. Am. J. Physiol. 1992; 262(6 Pt 1 ): E851-855.

Holmang A, Niklasson M, Rippe B, Lonnroth P. Insulin insensitivity and delayed transcapillary delivery of insulin in oophorectomized rats treated with testosterone. Acta Physiol. Scand. 2001 ; 171 (4): 427-438.

Hwang KR, Choi YM, Kim JJ, Chae SJ, Park KE, Jeon HW, Ku SY, Kim SH, Kim JG, Moon SY. Effects of insulin-sensitizing agents and insulin resistance in women with polycystic ovary syndrome. Clin. Exp. Reprod. Med. 2013; 40(2): 100-105.

Jayasena CN, Franks S. The management of patients with polycystic ovary syndrome. Nat. Rev. Endocrinol. 2014; 10(10): 624-636. Jensterle M, Kocjan T, Janez A. Phosphodiesterase 4 inhibition as a potential new therapeutic target in obese women with polycystic ovary syndrome. J. Clin. Endocrinol. Metab. 2014; 99(8): E1476-1481.

Kempna P, Hofer G, Mullis PE, Fluck CE. Pioglitazone inhibits androgen production in NCI-H295R cells by regulating gene expression of CYP17 and HSD3B2. Mol Pharmacol 71(3): 787-98 (2007).

Malapaka RR, Khoo S, Zhang J, Choi JH, Zhou XE, Xu Y, et al., Identification and mechanism of 10-carbon fatty acid as modulating ligand of peroxisome proliferator-activated receptors. The Journal of Biological Chemistry 287(1 ): 183-95 (2012).

Maliqueo M, Sun M, Johansson J, Benrick A, Labrie F, Svensson H, et al., Continuous administration of a P450 aromatase inhibitor induces polycystic ovary syndrome with a metabolic and endocrine phenotype in female rats at adult age. Endocrinology 154(1): 434-45 (2013).

Marcondes FK, Bianchi FJ, Tanno AP. Determination of the estrous cycle phases of rats: some helpful considerations. Braz. J. Biol. 62(4A): 609-614 (2002).

Medeiros SF de, Barbosa JS, Yamamoto MMW. Comparison of steroidogenic pathways among normoandrogenic and hyperandrogenic polycystic ovary syndrome patients and normal cycling women. J. Obstet. Gynaecol. Res. 2015; 41(2): 254-263.

Nawrocka J, Starczewski A. Effects of metformin treatment in women with polycystic ovary syndrome depends on insulin resistance. Gynecol. Endocrinol. Off. J. Int. Soc. Gynecol. Endocrinol. 2007; 23(4): 231-237.

Pillai A, Bang H, Green C. Metformin and glitazones: do they really help PCOS patients? J. Fam. Pract. 2007; 56(6): 444-453.

Sun F, Indran IR, Zhang ZW, Tan MHE, Li Y, Lim ZLR, Hua R, Yang C, Soon F-F, Li J, Xu HE, Cheung E, Yong E-L. A novel prostate cancer therapeutic strategy using icaritin activated arylhydrocarbon-receptor to co-target androgen receptor and its splice variants. Carcinogenesis 2015. doi: 10.1093/carcin/bgv040.

Udhane S, Kempna P, Hofer G, Mullis PE, Fluck CE. Differential regulation of human 3 -hydroxysteroid dehydrogenase type 2 for steroid hormone biosynthesis by starvation and cyclic AMP stimulation: studies in the human adrenal NCI-H295R cell model. PloS One 2013; 8(7): e68691.

Yong EL, Baird DT, Hillier SG. Mediation of gonadotrophin-stimulated growth and differentiation of human granulosa cells by adenosine-3',5'-monophosphate: one molecule, two messages. Clin. Endocrinol. (Oxf.) 1992; 37(1 ): 51-58.

Zhan Y, Chen Y, Zhang Q, Zhuang J, Tian M, Chen H, Zhang L, Zhang H, He J, Wang W, Wu R, Wang Y, Shi C, Yang K, Li A, Xin Y, Li TY, Yang JY, Zheng Z, Yu C, Lin S- C, Chang C, Huang P, Lin T, Wu Q. The orphan nuclear receptor Nur77 regulates LKB1 localization and activates AMPK. Nat. Chem. Biol. 2012; 8(11): 897-904.

Zurvarra FM, Salvetti NR, Mason Jl, Velazquez MML, Alfaro NS, Ortega HH. Disruption in the expression and immunolocalisation of steroid receptors and steroidogenic enzymes in letrozole-induced polycystic ovaries in rat. Reprod. Fertil. Dev. 2009; 21(7): 827- 839.