ATM encodes a serine/threonine protein kinase ... OCT1, OCT2 and ATM genes in 676 women with PCOS and 90 control ..... Diastolic blood pressure (mmHg).
Research Article Metformin resistance alleles in polycystic ovary syndrome: pattern and association with glucose metabolism Insulin-sensitizer treatment with metformin is common in polycystic ovary syndrome (PCOS). OCT alleles were investigated in PCOS patients to identify genetic ‘bad responders’ and ‘nonresponders’ to metformin including their possible effects on glucose metabolism without treatment. We genotyped eight SNPs in OCT1, OCT2 and ATM genes in 676 women with PCOS and 90 control women, we also measured oral glucose tolerance tests prior to treatment. Nonfunctional alleles were present in 29.8% and low-functional alleles in 57.9% of our PCOS cohort. OCT variants were significantly associated with elevated baseline and glucose-induced C-peptide levels in PCOS. Metformin bad responders or nonresponders based on OCT genotypes might be relevant in clinical practice – their modulation of metformin pharmacokinetics and pharmacodynamics and metformin-independent glucose effects remain to be elucidated. Original submitted 7 June 2013; Revision submitted 28 October 2013 Keywords: ATM n C-peptide n metformin resistance n organic cation transporters n polycystic ovary syndrome n SNPs
Polycystic ovary syndrome (PCOS) is one of the most common hormonal disorders in women of reproductive age with a prevalence ranging from 5 to 15% [1] . PCOS is associated with infertility, obesity and insulin resistance. A widely used oral agent in the treatment of insulin resistance in PCOS patients is the biguanide metformin. This insulin-sensitizer has been used for a long time, it reduces glucose absorption in the gastrointestinal tract and hepatic gluconeogenesis by inhibiting the mitochondrial respiratory-chain complex [2,3] , and increases glucose uptake in muscle and liver cells [4] . The mechanism of action of metformin remains controversial since an AMPK dependent as well as an independent mode of action have been shown [5,6] . In addition to its actions on glucose metabolism, metformin is able to suppress lipogenesis and to increase the synthesis of sex hormone-binding globulin (SHBG) in the liver [7] . In clinical use, metformin improves menstrual frequency, ovulation, conception and live-birth rates, thus increasing its value in the treatment of PCOS [8,9] . Accumulating evidence shows a great variability in the clinical response to metformin, which is mediated mainly by genetic factors: recent genome-wide association studies identified rs11212617 as a modifier of metformin response [6,10] , a SNP mapping to a locus on chromosome 11q22 containing seven genes including the ATM gene. The SNP rs11212617 is associated with
the lowering of HbA1c in diabetic patients as well as being associated with other effects [10] . ATM encodes a serine/threonine protein kinase that regulates AMPK activity by phosphorylation and therefore plays a key role in cellular energy homeostasis [11] . ATM might therefore be involved in the AMPK-dependent regulation of energy homeostasis by metformin. In vitro studies identified metformin as a substrate of the OCT family of transporters [12,13] . The liver-specific OCT1 and the kidneyexpressed OCT2 modulate the pharmacokinetic and pharmacodynamic profile of metformin [14,15] . OCT1 mediates the entry of metformin into hepatocytes where it exerts its actions by unknown pathways. A variety of SNPs in OCT1 and OCT2 have been identified, which influence hepatic intake or renal clearance of metformin in vivo and in vitro [16–23] . According to the literature, there is a relationship between glucose levels, expression and activity of OCT proteins, since in vitro experiments demonstrated a reduced expression and function of OCT proteins in Caco-2 cells cultured in high glucose containing medium [24] . This relationship was also seen in vivo, since in diabetic animals the expression of OCT proteins was significantly reduced [25–27] . Furthermore, in vitro experiments showed an increase in OCT1 expression in HepG2 cells after insulin treatment [28] . These data suggest an influence of OCT proteins as well as of their genetic variants on
10.2217/PGS.13.223 © 2014 Future Medicine Ltd
Pharmacogenomics (2014) 15(3), 305–317
Natascha Schweighofer*1, Elisabeth Lerchbaum1, Olivia Trummer1, Verena Schwetz1, Thomas Pieber1 & Barbara Obermayer-Pietsch1 Division of Endocrinology & Metabolism, Department of Internal Medicine, Medical University of Graz, Auenbruggerplatz 15, 8036 Graz, Austria *Author for correspondence: Tel.: +43 316 385 14290 Fax: +43 316 385 12032 natascha.schweighofer@ medunigraz.at
1
part of
ISSN 1462-2416
305
Research Article
Schweighofer, Lerchbaum, Trummer, Schwetz, Pieber & Obermayer-Pietsch
glucose metabolism. Therefore, we investigated the association of these SNPs with parameters of glucose metabolism prior to metformin treatment. We aimed to investigate the frequency of the metformin-response-modulating SNPs R61C, G465R, S14F, S189L, G220V, P160L and M408V in the OCT1 gene, A270S in the OCT2 gene and the intronic SNP rs11212617 in the ATM gene in a large cohort of Caucasian women with PCOS and controls. We intended to examine the number of patients belonging genetically to the group of ‘nonresponders’ or to the group of ‘bad responders’ to define the necessity of a clinical follow-up and a potential change to metformin dosing regimens. Furthermore, we aimed to explore the association of these SNPs with the parameters of oral glucose tolerance tests (OGTTs) in women with PCOS independently of metformin use to detect their possible influence on glucose metabolism.
Material & methods Subjects Women with PCOS (n = 880), were routinely referred to our outpatient clinic for PCOS evaluation from 2006 to 2012. Two out of the following three criteria were required for the diagnosis according to the 2003 Rotterdam criteria [29] : oligoovulation and/or anovulation, clinical and/ or biochemical signs of hyperandrogenism and polycystic ovaries. Oligoovulation and anovulation were defined by the presence of oligo menorrhea and amenorrhea. Hyperandrogenism was defined by the clinical presence of hirsutism, acne or alopecia and/or elevated androgen levels. Polycystic ovarian morphology was examined by ultrasound. Disorders with a similar clinical presentation (hyperprolactinemia, congenital adrenal hyperplasia, Cushing’s syndrome and androgen-secreting tumors) were excluded prior to the study by specific laboratory analyses (cortisol, corticotropin, 17aOH-progesterone, dehydroepiandrosterone sulfate and thyreotropin). Out of the original number of PCOS patients, 676 women with PCOS (aged 27 ± 7 years [mean ± standard deviation (SD)]) did not take any medication known to affect carbohydrate metabolism (including metformin) or endocrine drugs for at least 3 months before being included in the study. The control cohort consisted of 439 healthy women without clinical or laboratory evidence of PCOS, who were invited for metabolic testing during a routine thyroid evaluation in our outpatient clinic from 2009 to 2012. None of them had endocrine or autoimmune disorders. Out 306
Pharmacogenomics (2014) 15(3)
of this number of controls, 95 control women (aged 38 ± 12 years [mean ± SD]) did not take any medication known to affect carbohydrate metabolism (including metformin) or endocrine drugs for at least 3 months before being included in the study. When we performed subgrouping of women with PCOS and controls according to BMI, we found 239 lean women with PCOS and 43 lean controls with a BMI
