Lipids and insulin regulate mitochondrial‐derived peptide (MOTS‐c) in PCOS and healthy subjects

Polycystic ovarian syndrome (PCOS) is a heterogeneous endocrine disorder associated with mitochondrial dysfunction and insulin resistance (IR). MOTS‐c, a mitochondrial peptide, promotes insulin sensitivity (IS) through activating AKT and AMPK‐dependent pathways. The current study was designed to examine the response of MOTS‐c to lipids (intralipid) followed by insulin in PCOS and healthy subjects.


| INTRODUC TI ON
Polycystic ovarian syndrome (PCOS) is a complex heterogeneous endocrine disorder affecting 5%-20% of women in the reproductive age group and is one of the most common causes of infertility and metabolic disorder in these women. 1 In Qatar, the frequency of PCOS among women aged between 18 and 30 years is estimated to be 12% according to original NIH criteria for its diagnosis. 2 PCOS is a chronic proinflammatory state associated with obesity, diabetes and dyslipidemia. 1 Women with PCOS commonly have symptoms of oligomenorrhoea, polycystic ovaries, hirsutism, subfertility and insulin resistance (IR). 3 PCOS women have elevated levels of cardiovascular disease risk (CVD) factors including type 2 diabetes (T2DM), carotid intima-media wall thickness and platelet dysfunction. 4 Weight loss improves many clinical features of PCOS, including menses regularity and fertility, 5 and reduces CVD risk factors and IR 6 Cellular mechanisms leading to IR in PCOS are not clearly understood. Adipose tissue dysfunction and dysregulated expression of adipokines have been implicated in the pathophysiology of PCOS. Adipose tissue may communicate with the brain, ovaries and uterus through adipokines, to regulate reproductive functions and metabolic features of women with PCOS. 7 Over-production of proinflammatory adipokines such as TNFα, hyperstimulation of GLUT4 glucose transporter 8 and reduced expression of beneficial adipokines such as adiponectin have been observed in PCOS. 9 Mitochondrial transcriptome data analysis provided initial evidence for the presence of small RNAs and polypeptides originating from mtDNA 10 that possess significant biological activity. 11,12 Humanin was the first mitochondria-derived peptide discovered from the 16s ribosomal region of mtDNA. 11 Recently, another mitochondrial peptide originating from the 12S ribosomal region of mtDNA was identified and named MOTS-c (mitochondrial open reading frame of the 12S rRNA type-c). As expected, due to its mitochondrial origin, MOTS-c is expressed in key metabolic organs such as heart, skeletal muscle, testes, liver and brain. 12 The expression of MOTS-c is downregulated following calorie restriction in skeletal muscle in mice. 12 Studies in rodents demonstrated that small mitochondrial peptides promote mitochondrial metabolism, regulate critical processes such as ageing, inflammation and reversed IR [12][13][14][15] MOTS-c enhanced glucose utilization, promoted IS and restored metabolic homeostasis through activation of AMPK-dependent mechanisms in skeletal muscle and protected rodents against IR induced by obesity and ageing. 12 Some of the biological responses mediated by MOTS-c were due to its ability to upregulate cellular accumulation of 5-aminoimidazole-4-carboxamide ribonucleotide (AICAR), a potent activator of the AMP-activated protein kinase (AMPK) pathway and mitochondrial metabolism (24). MOTS-c treatment suppressed bone loss following ovariectomy via activation of an AMPK-dependent pathway, suggesting the importance of MOTS-c in bone metabolism and osteoporosis. 16 The role of MOTS-c in human pathophysiological conditions such as PCOS that is associated with multiple metabolic abnormalities including hyperinsulinemia, IR, obesity and hyperlipidemia has not been studied previously; therefore, the present study was designed to examine the functional effects of acute intravenous administration of lipids, insulin and a combination of both before and after 8 weeks of a moderate exercise intervention on the circulatory levels of MOTS-c in control and PCOS subjects. Pharmacia) with a hyperinsulinemic-euglycaemic clamp in the last 2 hours to determine IS (M-value). 18 Following this procedure, all the participants underwent supervised moderate intensity (60% maximum oxygen consumption) exercise, 3 hours weekly for 8 weeks. During the second phase of this study, the phase 1 protocol was repeated in all the subjects. Normal control women had the initial clamp in the first week of their menstrual cycle, whilst PCOS women were clamped after 6 weeks amenorrhoea.

| Insulin clamps
After fasted blood samples were taken, either normal saline 1.5 mL/min or 20% intralipid 1.5 mL/min, along with unfractionated heparin sodium 0.3 unit/kg/min, was infused for 5 hours. At 180 minutes, a 2-hours hyperinsulinemic-euglycaemic clamp was started using intravenous soluble insulin (Humulin S; Eli Lilly and Co) at a rate of 80 mU/m 2 surface area/min for the first 20 minutes, followed by a constant rate of 40 mU/m 2 surface area/minute for the remaining 100 minutes. Plasma glucose was clamped at 5.0 mmol/L with a variable infusion rate of 20% dextrose, adjusted relative to arterialized blood glucose measurements undertaken every 5 minutes. Endogenous glucose production was more than 90% suppressed by an acute rise of insulin with the primed insulin infusion. The rate of insulin-stimulated glucose disposal (mg/ kg/min) (M-value), a measure of IS, was calculated from the mean of the five 20-minute periods from 20 to 120 minutes during the clamp using the Defronzo method. 19 Overall study scheme is shown in Figure 1

| Biochemical measurements
Plasma MOTS-c concentrations were measured using a commercially available ELISA kit (Peninsula Laboratories International, Inc) according to manufacturer's recommended protocol, with an intraassay variation of less than 10%. Serum insulin was assayed using a competitive chemiluminescent immunoassay (Euro/DPC). Plasma glucose was measured using a Synchron LX20 analyzer (Beckman-Coulter). Triglycerides (TG) and total cholesterol were measured by a Synchron LX20 analyzer (Beckman-Coulter). The free androgen index (FAI) was calculated by dividing the total testosterone by SHBG, and then multiplying by 100. Serum testosterone (nmol/L) was assessed by high-performance liquid chromatography linked to tandem mass spectrometry (Waters Corporation), and SHBG (nmol/L) levels were measured by immunometric assay with fluorescence detection on the DPC Immulite 2000 analyzer (Euro/DPC). FSH (IU/L) was measured by an Architect analyser (Abbott laboratories); TCH (mmol/L), TG (mmol/L) and HDL (mmol/L) were measured using a Synchron LX20 analyzer (Beckman-Coulter); and LDL (mmol/L) was calculated using the Friedewald equation. Plasma glucose was measured using a Synchron LX20 analyzer (Beckman-Coulter). NEFA was measured using an enzymatic colorimetric method (Wako NEFA-H2) on a Konelab20 auto-analyzer with a coefficient of variation of 1.4%.
All the above measurements were done according to manufacturer's recommended protocol. analysis was performed to evaluate significant differences within the variable. A P value < 0.05 (two-tailed) was considered for statistical significance. In post hoc analysis, comparing multiple groups is as follows: a: baseline to saline comparison; b: saline to saline + insulin comparison; and c: baseline to saline + insulin comparison; *P < 0.05, **P < 0.01 and ***P < 0.001. Spearman Rank correlation was used to evaluate significant associations between variables and MOTS-c.

| Statistical analyses
The SPSS 22.0 statistical package was used for the study analysis, and Graphpad prism 5 software was used for the data analysis.
F I G U R E 1 Effect of acute administration of intralipid and insulin on plasma MOTS-c before and after exercise in control and PCOS subjects. A, Saline and insulin before exercise; B, intralipid and insulin before exercise; C, saline and insulin after exercise; D, intralipid and insulin after exercise. ***P < 0.0001; **P < 0.001; *P < 0.05

| Baseline characteristics of control and PCOS subjects
Anthropometric measurements for both PCOS and healthy control subjects are shown in Table 1. There was no significant difference in height, weight or BMI between the controls and PCOS subjects at the time of recruitment into the study.
The mean age (23.4 ± 5.9 in controls and 28.8 ± 6.6 in PCOS, [P < 0.05]), waist (78.4 ± 11.9 in controls and 91.4 ± 14.9 in PCOS, [P < 0.05]) and hip size (97.5 ± 13 in controls and 109.8 ± 12.6 in PCOS, [P < 0.05]) were significantly higher in the PCOS group compared with the healthy volunteers. However, the waist-tohip ratio was not significantly different between the two groups.
Eight weeks of exercise did not make any significant alterations to BMI, waist, hip or waist-to-hip ratio in either the healthy or the PCOS subjects (Table 1).
Basic hormonal measurements such as LH, FSH, oestradiol, prolactin, TSH, DHEAS, androstenedione, lipid profiles (including TCH, TG, HDL and LDL), creatinine, creatine kinase, platelets, FPG, 2 hour PG and HbA1c levels were similar between PCOS and healthy subjects. However, blood parameters that are a characteristic feature of PCOS subjects, such as the androgen profile, were elevated in patients with PCOS as assessed by the FAI (2.05 ± 1.9 in controls and 6.12 ± 3.1 in PCOS, [P < 0.001]). The elevated FAI was due to a significant decrease in SHBG in the PCOS subjects (70.42 ± 28.9 in controls and 31 ± 20.3 in PCOS, [P < 0.001]) although total testosterone was not significantly different between the 2 groups (Table 1). ALT (16.75 ± 5.1 in controls and 26.25 ± 13.5 in PCOS, [P < 0.05]) was elevated in PCOS compared with healthy subjects and may reflect fatty liver in the PCOS subjects.

| Acute insulin administration does not affect plasma MOTS-c before exercise intervention
In the first phase of the study, we tested the effects of saline and acute insulin administration on blood metabolic and biochemical parameters in the study subjects (Table 2). Plasma MOTS-c levels were not affected by the saline or insulin infusion in either the PCOS or control subjects. Saline infusion did not alter TCH, TG, NEFA and MOTS-c levels either in control or PCOS subjects.
Administration of insulin infusion along with saline during the last two hours of the euglycaemic clamp procedure suppressed both TG (b *** ) and NEFA (b *** ) in healthy volunteers, but only NEFA (b *** ) was reduced in PCOS subjects. One-way ANOVA comparing mean differences among baseline, saline and the combination of saline + insulin showed significant changes in TG (P < 0.001 in controls vs NS changes in PCOS) and NEFA (P < 0.001 in controls vs P < 0.001 in PCOS subjects). Mean IS index M-value was significantly higher in controls (4.96 ± 2.0) compared with PCOS subjects (3.26 ± 0.8, Table 2).

| Acute intralipid infusion increases MOTS-c levels in both control and PCOS subjects
Intralipid infusion altered many of the blood biochemical parameters except TCH in both groups (Table 3). Descriptive statistics assessing the mean differences between each group and the overall group comparison showed that the lipid infusion significantly increased plasma MOTS-c (a ** , P < 0.05 in controls vs a ** , P < 0.01 in PCOS subjects) and also elevated TG (a *** , P < 0.001 in controls vs a *** , P < 0.001 in PCOS subjects) and NEFA (a *** , P < 0.001 in controls vs a *** , P < 0.001 in PCOS subjects). To assess the effects of supraphysiological insulin whilst maintaining normal plasma glucose at  Table 2.
The percentage changes in plasma MOTS-c following intralipid and insulin before exercise intervention is depicted in Figure 1B. Baseline levels were adjusted to 100%. Intralipid elevated plasma MOTS-c from 100% to 232 ± 124% in control subjects and in PCOS subjects to 349 ± 206%, respectively. Administration of insulin suppressed intralipid-induced MOTS-c significantly in both groups, to 165 ± 97% and 183 ± 177% in control and PCOS subjects, respectively. Thus, insulin suppressed the lipid-induced increase in MOTS-c levels.

| Exercise intervention followed by acute insulin infusion does not affect plasma MOTS-c
Regular exercise has many beneficial and positive effects in PCOS women. Blood biochemical measurements and plasma MOTS-c were measured in response to saline and hyperinsulinemic-euglycaemic clamp in our study subjects following 8 weeks of supervised exercise intervention. Exercise intervention did not show any improvement in the IS index M-value in either group (Table 4) compared with the sensitivity index M-value prior to exercise intervention ( Table 2).
Infusion of insulin following saline did not have any significant effects on circulating TG and MOTS-c; however, overall group analysis showed significant changes in TCH (P < 0.001) and NEFA (P < 0.001) in healthy subjects. In PCOS subjects, the effects were more profound after exercise intervention compared with controls.
In PCOS subjects, insulin infusion suppressed circulating NEFA (b *** , P < 0.001) as assessed by post hoc analysis comparison between each group and overall group comparison by ANOVA analysis (Table 4).

| Acute intralipid infusion enhances, and insulin suppresses, intralipid-induced plasma MOTS-c following exercise intervention
In the second phase of the study, we compared the effects of lipid       VO 2 max of 60% in each of the participating subjects. We observed a small but insignificant increase in baseline plasma MOTS-c in both healthy and PCOS subjects following exercise.
The overall group ANOVA analysis revealed the effect of intralipid and intralipid + insulin insults following exercise on blood biochemical parameters, such as TCH, TG and NEFA, were similar to that found prior to exercise training in healthy subjects. The net change in plasma MOTS-c in the lipid and the intralipid + insulin treatment group was significantly elevated following exercise in healthy subjects (P < 0.01) compared with (P < 0.05) before exercise measurements in the same subjects (Table 3 compared to Table 5).
However, this differed to what was found in the PCOS subjects where the combination of intralipid + insulin administration lowered MOTS-c (P < 0.05, Table 5) after exercise compared with observed changes (P < 0.01, Table 3 and 212 ± 231% in control and PCOS subjects, respectively.

| Correlation of MOTS-c with covariates
In all participants (n = 10 controls and n = 12 PCOS subjects), Spearman Rank analysis showed that, prior to exercise, plasma MOTS-c negatively correlated with plasma glucose (PG) at baseline (0 minute) only in the PCOS group (r = −0.773; P = 0.01).  (Table S1).

| D ISCUSS I ON
In this study, the role of FFA in the regulation of MOTS-c was evaluated and showed that MOTS-c was increased in both PCOS and controls by the intralipid infusion, but that the MOTS-c levels did not fall following the infusion of insulin that reduced NEFA and TG levels.
The hyperinsulinemic-euglycaemic clamp technique is a well-accepted model for studying in vivo IR, skeletal muscle IS and glucose metabolism in human subjects. 19 Whilst there are studies in vitro   and in animal models, human studies are scant and this is the first to report the response of MOTS-c in such an interventional study.
Intralipid administration elevates circulating FFA levels that have deleterious effects in humans, 20 promotes proinflammatory cytokine TNFα secretion, interferes with glucose metabolism, obstructs glucose oxidation and reduces glucose disposal rate. 21 We hypothesized that intralipid and insulin may differentially regulate IR, resulting in altered metabolic processes and altered glucose metabolism that may influence plasma MOTS-c expression. Regular moderate exercise reduces diabetes risk by more than 50% in subjects at risk of developing diabetes. 28 For individuals with diabetes, exercise makes it easier to control blood glucose via a reduction in skeletal muscle IR and an improvement in IS, leading to a reduction in long-term diabetes-related comorbidities. 29 Obese PCOS women have an increased risk of metabolic syndrome with elevated cardiovascular risk factors. 30 Lifestyle modification and weight reduction enhance sex hormone-binding globulin (SHBG) levels, reduce testosterone, restore ovulation and improve pregnancy rates. 31 Increased physical activity and weight loss improve the biochemical and phenotypic features of PCOS, thus reducing metabolic complications and thereby improving quality of life. [32][33][34] Endurance exercise intervention did not reduce weight or alter plasma MOTS-c significantly in either group; however, enhanced glucose utilization, through an improvement in IS, 35 and a reduction in IR with lower plasma glucose levels were seen following the hyperinsulinemic-euglycaemic clamp in PCOS subjects (Table 2 and   Table 4). Increasing physical activity is a therapeutic option for reducing abnormalities associated with metabolic syndrome. 36  weight gain. 38 From the above study in mice, it could be speculated that the rise in plasma MOTS-c observed following intralipid administration could be a defence mechanism kicking in as a compensatory response to counteract the deleterious effects caused by the rise in NEFA and triglycerides. Furthermore, the changes in IR associated with MOTS-c may be indirectly due to alterations in FFA levels.
The main strengths of this study are this is the first in vivo study to show the relationship between circulating lipids and MOTS-c in humans and that gold standard methodology for this intensive interventional study and well-supervised exercise intervention were employed. However, the limitations were that this was a small group of women with PCOS, though all fulfilled all three of the diagnostic criteria for its diagnosis, thereby reducing heterogeneity; however, the groups were not well matched for BMI and age. For the reduction of MOT-c following insulin during the lipid infusion that did not differ in the control subjects, the study was underpowered to determine the significance of this finding that could represent a type 2 statistical error, given that the reduction of MOT-c in the PCOS group was significant. The diagnostic criteria used to define the PCOS group would likely not have altered the results here, as IS was reported to be no difference between the original NIH criteria and Rotterdam criteria. 39 In addition, the findings may not be generalized to other populations; however, due to the intense nature of the study protocol requiring multiple visits, and the frequent blood sampling needed to stabilize plasma glucose at steady state levels during clamp conditions, it is not feasible to perform such studies in a large cohort of subjects. 19 For this reason, the results must be interpreted cautiously, and thus further studies in large cohorts of subjects are warranted to confirm our findings.
In conclusion, this is the first study to show lipid increases cir- Clinical trial registration Number: ISRCTN42448814.

ACK N OWLED G EM ENTS
The authors would like to thank Qatar Metabolic Institute, Medical Research Center, iTRI, Hamad Medical Corporation, Doha, Qatar.
Weill Cornell Medicine Qatar, Qatar Foundation, Doha, Qatar and Qatar National Library for the support.

CO N FLI C T O F I NTE R E S T
The authors have no conflict of interest to declare.

DATA AVA I L A B I L I T Y
The data that support the findings of this study are available from the corresponding author upon reasonable request.