PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
Clin Pharmacol Ther. Author manuscript; available in PMC 2014 February 1.
Published in final edited form as:
PMCID: PMC3671611
NIHMSID: NIHMS468755

The effect of novel promoter variants in MATE1 and MATE2 on the pharmacokinetics and pharmacodynamics of metformin

Abstract

Interindividual variation in response to metformin, first-line therapy for type 2 diabetes, is substantial. Given that transporters are determinants of metformin pharmacokinetics, we examined the effects of promoter variants in both multidrug and toxin extrusion protein 1 (MATE1) (g.−66T→C, rs2252281) and MATE2 (g.−130G→A, rs12943590) on variation in metformin disposition and response. The pharmacokinetics and glucose-lowering effects of metformin were assessed in healthy volunteers (n = 57) receiving metformin. The renal and secretory clearances of metformin were higher (22% and 26%, respectively) in carriers of variant MATE2 who were also MATE1 reference (P < 0.05). Both MATE genotypes were associated with altered post-metformin glucose tolerance, with variant carriers of MATE1 and MATE2 having an enhanced (P < 0.01) and reduced (P < 0.05) response, respectively. Consistent with these results, patients with diabetes (n = 145) carrying the MATE1 variant showed enhanced metformin response. These findings suggest that promoter variants of MATE1 and MATE2 are important determinants of metformin disposition and response in healthy volunteers and diabetic patients.

Keywords: metformin, MATE1, MATE2, genetic polymorphism, pharmacokinetics, pharmacodynamics, glucose, healthy volunteers, type II diabetic patients, HbA1c

INTRODUCTION

As the first line therapy for the treatment of type II diabetes, metformin is the most frequently prescribed anti-diabetic drug [1]. Although controversial, studies suggest that metformin’s pharmacological action is related to its activation of AMPK, which reduces hepatic glucose production, enhances glucose uptake in hepatic cells and peripheral tissues, decreases absorption of glucose from the gastrointestinal tract and increases insulin sensitivity in peripheral tissues [2, 3].

The pharmacokinetics of metformin have been studied extensively in both healthy volunteers and type II diabetic patients. About 50% of an oral dose is absorbed [4, 5] into the blood and rapidly distributed to various tissues. Metformin is not bound to plasma proteins [6] and is eliminated into urine as unchanged drug [4, 7]. The renal clearance of metformin is much greater than glomerular filtration rate, suggesting a significant contribution of tubular secretion to its elimination. Considerable inter-individual variability in the renal clearance of metformin has been observed in healthy volunteers (150–700 mL/min) [7] which includes a strong genetic component [8, 9]. In addition to pharmacokinetic inter-individual variability, the response to metformin varies substantially, with more than 30% of patients receiving metformin monotherapy classified as non-responders [10].

Metformin relies on facilitated transport for delivery to the liver, kidney and peripheral tissues. Indeed, previous studies demonstrate that membrane transporters contribute to the inter-individual variability in the pharmacokinetics and pharmacodynamics of metformin [1115]. Metformin is transported primarily by the organic cation transporters (OCTs), particularly OCT1 and OCT2, and multi-drug and toxin extrusion proteins (MATEs), namely MATE1 and MATE2. MATE1 and OCT1 have been implicated as determinants of metformin response primarily due to their tissue distribution at major sites of metformin action. MATE1 is highly expressed in the kidney and liver (Supplemental Figure 1) with lower expression in skeletal muscle and adipose tissue [16], whereas, OCT1 is predominantly expressed in the liver. Previous reports from our laboratory and others have shown that OCT1 is the major determinant of metformin uptake into hepatocytes and polymorphisms of OCT1 are associated with reduced response to metformin in both healthy volunteers [11, 12] and diabetic patients [17, 18]. Interestingly, MATE1 and OCT1 have been shown to mediate transcellular transport of metformin in vitro [19, 20] and to effect metformin response in diabetic patients [14]. Other studies indicate MATE1 polymorphisms alone affect metformin response in diabetic patients [21].

In addition to effects on pharmacodynamics, transporters play a major role in metformin renal elimination [22]. In particular, OCT2 mediates the entry of metformin into the renal tubular cells, whereas MATE1 and MATE2 contribute to the efflux of metformin into the urine [19, 20] (Figure 1). Previous studies have shown that a nonsynonymous variant in OCT2 (A270S, rs316019) alters the renal clearance of metformin in healthy volunteers [15, 23, 24]. In addition, renal clearance and tissue distribution of metformin is altered in Mate1(−/−), but not Mate1(+/−) mice [25, 26]. However, the effect of genetic variation in MATE1 and MATE2 on the pharmacokinetics of metformin in humans remains unclear.

Figure 1
Representative cartoon of metformin transporters in the hepatocyte of the liver and nephron of the kidney

Until recently, clinical studies focused on the effects of promoter variants on drug disposition and response have been less well studied than coding and intronic region polymorphisms. Recently, our laboratory has shown that a common promoter variant, MATE2 (g.−130G>A, rs12943590), increases luciferase activity in vitro and associates with reduced response to metformin in diabetic patients [27]. In addition, another common promoter variant, MATE1 (g.−66T>C, rs2252281) exhibited reduced luciferase activity in reporter assays in vitro and was shown to associate with reduced-expression of MATE1 mRNA transcripts in the kidney [28]. In the current study, we hypothesized that these two promoter variants are determinants of metformin renal clearance and anti-diabetic response in healthy volunteers and diabetic patients. Because the two variants may have opposing effects, we also considered gene-gene interactions in our association studies. Our data demonstrate that in the absence of the MATE1 promoter variant, the MATE2 promoter variant is associated with an increased renal clearance of metformin. Interestingly, both variants associate with the glucose-lowering effects of metformin in healthy volunteers and in diabetic patents, but in opposite directions.

RESULTS

Healthy male and female Asian (n=18), African American, (n=33) and Caucasian (n=6) volunteers were genotyped for MATE1 (g.−66T>C, rs2252281), MATE2 (g.−130G>A, rs12943590), OCT1 (420del, rs72552763) and OCT2 (A270S, rs316019). All alleles were in Hardy-Weinberg equilibrium. The pharmacokinetics and pharmacodynamics of metformin were evaluated in these volunteers after oral dosing of the drug (1850 mg in total). The study design and characteristics of the type II diabetic patients have been reported previously [27]. Demographic characteristics for the healthy volunteers and a subset of patients are shown in Supplemental Table S1 and S2. Below we first discuss our analysis of the association of the MATE1 variant with the pharmacokinetics and pharmacodynamics of metformin in healthy volunteers and diabetic patients, followed by an analysis focused on the MATE2 promoter variant. For both promoter variants, we first analyzed of the effect of either variant alone, and then adjusted for each of the additional transporter variants (see Methods).

The MATE1 promoter variant, g.−66T>C, has no effect on the pharmacokinetics of metformin in healthy volunteers

The pharmacokinetic parameters obtained in the present study are similar to those previous reported in healthy volunteers [11, 15, 2932].

The MATE1 g.−66T>C genotype had no significant effect on the pharmacokinetics of metformin (reference n=32, variant n=25), even after adjusting for creatinine clearance (CLCR). The pharmacokinetics of metformin remained similar even after exclusion of volunteers carrying the OCT1 or OCT2 polymorphisms (Supplemental Table S3).

The MATE1 promoter variant, g.−66T>C, is associated with a greater response to metformin in healthy volunteers

Before metformin dosing, the area under the curve (AUC) of glucose (mean±SD, reference, 359±56 mg/dL/h; variant 352±77 mg/dL/h) and insulin (reference, 129±83 mU/L/h; variant 141±102 mU/L/h) were similar between MATE1 genotypes. After metformin administration, volunteers who were homozygous for the variant MATE1 allele had significantly lower glucose AUC (greater response) after the oral glucose tolerance test (OGTT) than those volunteers carrying at least one reference allele (reference, 309±39 mg/dL/h; variant, 250±37 mg/dL/h; P=0.002; Figure 3a-b). The association of the MATE1 allele with glucose AUC persisted in subsequent analysis of volunteers who were also homozygous for the reference OCT1 (MATE1 reference, n=43, 308±40 mg/dL/h; MATE1 variant, n=5, 242±41 mg/dL/h; P=0.005) or OCT2 (MATE1 reference, n=41, 306±41 mg/dL/h; MATE1 variant, n=4, 262±12 mg/dL/h; P =0.03) polymorphisms. We were unable to detect a significant effect of the MATE1 variant in healthy volunteers after removal of individuals with the MATE2 g.−130G>A variant because of a reduction in sample size, which resulted in a substantial loss of power. Insulin AUC (reference 124±74 mU/L/h; variant 109±68 mU/L/h) and concentrations 2 hours after glucose administration (reference 41±32 mU/L; variant 27±18 mU/L/h) were similar for both MATE1 reference and variant volunteers.

Figure 3
The MATE1 promoter variant (g.−66T>C) is associated with different response to metformin in healthy volunteers and type II diabetic patients

The MATE1 promoter variant, g.−66T>C, is associated with a greater response to metformin in type II diabetes mellitus patients

The effect of the MATE1 promoter variant on the response to metformin (relative change in HbA1c) was examined in diabetic patients from a previously described [27] cohort of Caucasian (n=185) and African American (n=64) patients receiving metformin monotherapy. Alone, the MATE1 promoter variant was not associated with the relative change in HbA1c (P>0.6). However, in our secondary analysis, in which we examined the effect of the MATE1 promoter variant together with other transporter variants, we obtained the following results. The MATE1 promoter variant was not associated with response to metformin after removal of patients who are carriers of OCT2 A270S or MATE2 g.−130G>A. In contrast, when patients carrying one or more OCT1 reduced-function variants were removed from the analysis, the MATE1 variant allele had a significant effect on response (Figure 3c, Table 1). That is, Caucasian and African American patients homozygous for the MATE1 variant allele had a significantly larger relative change in HbA1c levels (i.e. greater response to metformin) than patients carrying at least one reference MATE1 allele (Figure 3c, Table 1, P=0.01).

Table 1
Association analyses of MATE1 g.−66T>C with metformin response (relative change in HbA1c) in patients with type II diabetes.

The MATE2 promoter variant, g.−130G>A, is associated with increased renal clearance

With the exception of half-life, the pharmacokinetics of metformin were similar for individuals with the MATE2 reference (n=27) and those with the variant allele (n=30, Figure 4a and Table 2), even after adjustment for CLCR. The elimination half-life was longer in volunteers carrying at least one MATE2 (g.−130G>A) variant allele compared to those with the reference MATE2 genotype. This association remained even after adjustment for gender, body mass index (BMI) and MATE1 (g.−66T>C, rs2252281) genotype (P=0.02).

Figure 4
The effect of MATE2 (g.−130G>A) on the pharmacokinetics of metformin in 57 healthy volunteers
Table 2
Summary of the metformin pharmacokinetic parameters in healthy volunteers with known MATE1, OCT1 and OCT2 genotype and homozygous for the MATE2 reference allele (−130G/G) and heterozygous or homozygous for the MATE2 variant allele (−130G/A ...

In the secondary analysis, we removed volunteers carrying at least one minor allele of the MATE1 polymorphism. In individuals who were homozygous for the reference MATE1 (n=32), the MATE2 variant was associated with lower plasma levels between 2 and 5 hours after metformin administration (P<0.05, Figure 4b). Although most of the pharmacokinetic parameters remained similar between the MATE2 variant groups, the CLR and CLSR of metformin were significantly higher in volunteers carrying at least one MATE2 variant compared to those homozygous for reference MATE2 (P<0.05, Table 2, Figure 4).

The MATE2 promoter variant, −130G>A, is associated with reduced metformin response in healthy volunteers

After metformin dosing, the glucose AUC was higher for volunteers homozygous for the MATE2 variant allele (333±37 mg/dL/h) as compared with those carrying at least one reference MATE2 allele (295±44 mg/dL/h; P=0.02; Figure 5a-b), while no effect of the MATE2 variant was observed in baseline glucose AUC before metformin treatment. Removing individuals with variants in MATE1 (MATE2 reference, 301±38 mg/dL/h; MATE2 variant, 345±36 mg/dL/h; P =0.02), OCT1 (MATE2 reference, 298±45 mg/dL/h; MATE2 variant, 333±37 mg/dL/h; P=0.04) or OCT2 (MATE2 reference, 299±41 mg/dL/h; MATE2 variant, 342±30 mg/dL/h; P =0.02) resulted in no change in the significance level of the glucose AUC between reference and variant MATE2 genotypes during metformin treatment.

Figure 5
MATE2 genetic variants are associated with different response to metformin in healthy volunteers and type II diabetic patients

Using linear regression, we determined if sex, age, BMI, fasting glucose, fasting insulin, MATE1 genotype, MATE2 genotype and metformin exposure (AUC) predicted the glucose AUC during metformin treatment. The MATE1 (P=0.02) and MATE2 genotypes (P=0.02) were the only significant predictors of metformin response, with each genotype alone explaining 7% of the variability in response to metformin. When both MATE1 and MATE2 genotype were included in a multiple linear regression model, 15% of the variance in metformin response was explained (P=0.005). The OCT2 genetic polymorphism has been previously associated with the renal clearance of metformin and nephrotoxicity of cisplatin [15, 33]. In our study, the OCT2 genotype (rs316019) alone had no effect on variation in the pharmacokinetics or pharmacodynamics of metformin.

DISCUSSION

Previous reports on genetic variants of MATE1, MATE2, OCT1 and/or OCT2 on metformin pharmacokinetics [17] and/or pharmacodynamics [11, 12, 17, 21, 26, 34] have focused on synonymous or nonsynonymous single nucleotide polymorphisms (SNPs) within non-regulatory regions of these genes. In this study, we determined the effects of two promoter variants of MATE1 (rs2252281) and MATE2 (rs12943590), discovered previously in our laboratory [27, 28], on the pharmacokinetics and pharmacological response to metformin in healthy volunteers and on the glycemic response to metformin in type II diabetic patients. In our primary analysis, we considered each variant separately. Given the co-localization of MATE1 and OCT1 in hepatocytes, and MATE1, MATE2 and OCT2 in proximal tubule cells (Figure 1), polymorphisms affecting the accumulation and/or elimination of metformin in these tissues, could confound the effects of the individual MATE1 and MATE2 promoter variants on metformin disposition and response. Therefore, whenever possible, we performed secondary analyses excluding individuals who had OCT1 (rs72552763), OCT2 (rs316019) or the other MATE genotypes that could potentially confound the measured parameters. Our major findings include: (a) The MATE1 reduced-expression promoter variant is associated with increased response to metformin in healthy volunteers and type II diabetic patients who were homozygous for the OCT1 reference allele. (b) The MATE2 enhanced-expression promoter variant is associated with reduced response to metformin in healthy volunteers. (c) The CLR and CLSR of metformin were significantly greater in volunteers carrying the promoter variant of MATE2 compared to those homozygous for reference MATE2 in a subset of healthy individuals who were homozygous for the MATE1 reference allele. To our knowledge, this is the first study to show that functional promoter polymorphisms in MATE1 and MATE2 contribute to variation in the response and disposition of metformin in both healthy volunteers and diabetic patients.

Although the MATE1 polymorphism (g.−66T>C) did not influence metformin disposition, healthy volunteers who were homozygous for the MATE1 variant allele had a greater glucose-lowering response to metformin (Figure 3a-b). Similarly, diabetic patients without reduced-function alleles of OCT1 who were also homozygous for the MATE1 variant allele had a 15% greater relative reduction in Hb1Ac compared to patients carrying the MATE1 reference allele (Figure 3c, Table 1). The finding that a MATE1 variant affects the pharmacodynamics, but not the pharmacokinetics of metformin, underscores the importance of transporters in tissue specific drug distribution. MATE1 is highly expressed in both the liver and kidney [16] (Supplemental Figure S1), with the liver being a major site of pharmacologic action and the kidney being predominantly a site of metformin elimination (although effects on glucose may also occur). In the kidney, MATE1 is redundantly co-expressed with MATE2 and MATE2-K, and therefore a variant in MATE1 would need to have a large effect size to have a measurable effect on the renal elimination of the drug. In contrast, in the liver, MATE1 appears to be the sole metformin transporter expressed on the bile canalicular membrane; therefore a genetic variant in MATE1 may have a more measurable effect on the pharmacodynamics compared to the pharmacokinetics of metformin. Mechanistically, our results are consistent with MATE1 acting as a proton gradient-driven efflux pump in tissues of pharmacodynamic importance, such as the liver [35, 36]. The reduced-expression promoter polymorphisms of MATE1 would presumably result in reduced transporter expression levels, leading to reduced efflux and correspondingly higher tissue levels of metformin. The higher tissue levels of metformin are predicted to associate with a greater pharmacologic response. Interestingly, when diabetic patients carrying the OCT1 reduced-function variant alleles were included in the analyses, we did not observe an effect of the MATE1 promoter variant on the response to metformin. It is possible that the reduced-function OCT1 variant masked the effects of the MATE1 variant (Figure 1). Particularly in the liver, the variant allele of OCT1 would result in lower drug levels and thus oppose the effects of the reduced-expression variant of MATE1. Our results also suggest that the MATE1 variant has a noticeable effect in patients with type II diabetes. For example, a typical diabetic patient with a baseline HbA1c of 8% (Supplemental Table S2) receiving metformin monotherapy (1000 mg/day) and reference for the OCT1 genotype would have their HbA1c decreased by an additional 1.2% if they were a carrier of the MATE1 variant allele instead of the reference MATE1. The magnitude of this effect is large and of clinical significance, given that, on average, metformin monotherapy lowers HbA1c by 1.12% within the first year of therapy [37, 38]. If our results are replicated in other cohorts, genotyping for the MATE1 polymorphism as a basis for personalizing metformin hypoglycemic therapy should be considered.

The effect of genetic variants in MATEs on metformin pharmacokinetics and pharmacologic action also suggests the potential for clinical significant drug-drug interactions (DDIs) that may "phenocopy" the effects of genetic variants. In fact, in an in vitro cell system (Chinese Hamster Ovary), inhibitors of human MATEs were identified from a diverse set of drug classes (e.g., pyrimethamine, baclofen, ketoconazole, naloxone, propranolol) [39]. In addition, recently Kusuhara et al. [40] demonstrated that pyrimethamine, an anti-protozoal drug and inhibitor of MATEs, reduced metformin renal clearance in human volunteers. This example highlights the importance of follow-up clinical studies to elucidate the clinical consequences of any in vitro DDI identified and how such a DDI may phenocopy the reduced-expression variant of MATE1.

Recent studies in our laboratory have shown that a common MATE2 promoter variant (g.−130G>A) associates with reduced response to metformin in diabetic patients [27]. In the present study we hypothesized that this variant is associated with reduced response to metformin as a result of a pharmacokinetic mechanism. Surprisingly, in healthy volunteers, MATE2 g.−130G>A did not affect the CLR and CLSR of metformin (Figure 4 and Table 2). Because MATE1 and MATE2 are expressed on the apical membrane of the proximal tubule and are likely to work together to mediate metformin renal elimination (Figure 1), we hypothesized that the effect of the MATE2 variant may be masked by the opposing effect of the MATE1 variant. Indeed, when we removed the individuals with the reduced-expression promoter variant of MATE1 g.−66T>C, we observed that the CLR and CLSR of metformin were significantly greater in individuals carrying the MATE2 variant allele compared to the reference allele (Figure 4). Consistent with this finding, the mean metformin exposure (AUC) was lower and the mean amount of metformin excreted in urine was higher in volunteers with the variant MATE2 allele, although the difference was not statistically significant, possibly because of variability in the bioavailability of metformin among our participants (Table 2). Interestingly, metformin concentrations were significantly lower in volunteers carrying the variant MATE2 at time points after the Cmax. Importantly, we observed a longer half-life in individuals with the MATE2 variant alleles (P=0.03). As metformin exhibits apparent flip-flop kinetics [41], early time points reflect elimination, whereas, later time points predominantly reflect drug absorption. The lower plasma levels after the Cmax and the longer half-life of the drug in individuals homozygous for the variant MATE2 allele are consistent with a reduced rate and extent of absorption of metformin. These results suggest that expression polymorphisms of MATE2 may alter the absorption of metformin from the gastrointestinal tract, as well as, its secretion (clearance) from then kidney. Although currently not implicated in the absorption of metformin, MATE2, is expressed in the small intestine (Supplemental Figure S1) and could contribute to the interindividual variation in the bioavailability of metformin. Current immunohistochemistry data suggests that MATE2 is expressed at a moderate level in the different sections of the gastrointestinal tract (e.g. duodenum, stomach, small intestine and colon) [42]. Further studies are required to confirm the localization and determine the function of MATE2 in the gastrointestinal tract and delineate the impact of MATE2 on the absorption of metformin.

The current study in healthy volunteers is consistent with previous studies in diabetic patients [27] that associate the MATE2 promoter variant with a reduced response to metformin, and provides evidence that this effect may be the result of a pharmacokinetic mechanism. That is, plasma levels of metformin were significantly lower in the volunteers homozygous for the variant MATE2 allele (Figure 4), throughout the OGTT (2–5 hours after metformin administration), potentially decreasing glucose uptake after the OGTT. Alternatively, by modulating metformin levels in a target tissue, the MATE2 variant may directly affect the pharmacodynamics of metformin. In particular, MATE2 is predominantly expressed in the kidney (Supplemental Figure S1) [16] an organ increasingly recognized to play a significant role in both systemic glucose production (20–25%) and glucose utilization (10%) in the fasting state [4346]. In type II diabetes, renal gluconeogenesis and glucose uptake increases [47]. Thus, individuals homozygous for the MATE2 enhanced-expression variant would achieve lower levels of metformin in the kidney, possibly translating into a reduced pharmacologic effect. A more comprehensive understanding of the peripheral targets of metformin is needed to completely interpret these results.

In conclusion, this study demonstrates that promoter variants of MATE1 and MATE2 contribute to the glycemic response to metformin in healthy volunteers and in patients with type II diabetes. Further, the study provides evidence that MATE1 and MATE2 work in concert in the kidney to mediate metformin renal elimination and that genetic variants of MATEs and OCTs should be considered together when ascertaining the genetic determinants of renal elimination of metformin. Finally, the results of our study suggest an important role of MATE2 in the pharmacokinetics and pharmacodynamics of metformin.

METHODS

Healthy human volunteers

The Committee on Human Research at the University of California, San Francisco (IRB 10-03087 and 10-02578) approved this study. Healthy male and female volunteers were recruited directly from the Study of Pharmacogenetics in Ethnically Diverse Populations (IRB 10-03167) and participants were enrolled only after provided informed consent was provided. To be eligible for this study, volunteers were older than 18 years of age and not taking any medications other than vitamins and/or oral contraceptives. Screening included a comprehensive medical history, physical examination, and laboratory studies (complete blood count, electrolytes, BUN and creatinine, albumen, and liver enzymes). Volunteers with elevated liver enzymes, anemia, elevated creatinine concentrations or a positive pregnancy test were excluded.

Genotyping

MATE2 (g.−130G>A, rs12943590), OCT1 coding variants (R61C [rs12208357], G401S [rs34130495], 420del [rs72552763] and G465R [rs34059508]) and OCT2 (A270S, rs316019) were genotyped by a TaqMan assay using standard procedures. MATE1 (g.−66T>C, rs2252281) was genotyped by PCR amplification followed by sequencing of the promoter region. The OCT1 coding variants (R61C, G401S and G465R) were only genotyped in the type II diabetic cohort.

Clinical study procedures

Once enrolled, volunteers were advised to maintain stable activity levels seven days before commencing the study. After the initial 3-day carbohydrate controlled diet (200–250 g/day), volunteers were admitted to the General Clinical Research Center (GCRC) at San Francisco General Hospital and remained there for the duration of the study (72 h). After an overnight fast (10 h), each underwent a 3 h oral glucose tolerance test (OGTT, 75 g; day 1). Volunteers were dosed with 1000 mg metformin (Major Pharmaceuticals, Livonia, MI) in the evening of study day 1 followed by a dose of 850 mg early on the second study day (day 2). A second OGTT was administered 2 h after metformin administration on study day 2. Standardized meals were provided on both study days after completion of the OGTT. Following the metformin dose, volunteers were asked to drink 8 oz of water every 2 h to maintain urine flow and pH.

Timed blood samples were collected after the first (0, 0.5, 1, 2 and 11 h) and second (0, 0.5, 1, 1.5, 2, 2.25, 2.5, 3.75, 3, 3.5, 4, 6, 8, 10, 12 and 24 h) metformin dose, respectively, for the determination of plasma metformin concentrations. For metformin pharmacodynamics (glucose/insulin concentrations) blood samples were collected at 0, 15, 30, 45, 60, 90, 120, 180 min after glucose administration. An additional blood sample was collected at 12 h after the second dose of metformin to determine serum creatinine. Urine samples were collected during the following time intervals: 0–12 h after the first dose of metformin and 0–2, 2–4, 4–6, 6–8, 8–12 and 12–24 h after the second dose of metformin.

Analytical methods

Metformin concentrations in plasma and urine were assayed by a validated liquid chromatography-tandem mass spectrometry method [11]. The quantification limit was 4 mg/L for plasma and 40 ng/mL for urine. Both the intra-day and inter-day coefficients of analysis variation were less than 5%. Glucose concentrations in plasma and creatinine concentrations in plasma and urine were determined using standard colorimetric assays. Insulin concentrations in plasma were determined using an immunoassay (Mercodia, NC) following the manufacturer’s instructions.

Clinical pharmacokinetics

The concentration-time profile of metformin was evaluated by non-compartmental analysis (WinNonlin 4.1, Pharsight Corporation, Mountain View, CA). The pharmacokinetics of metformin from both plasma and urine were calculated after the second dose as described previously [11].

Type II diabetic patients

Diabetic patients of Caucasian or African American ethnicity were recruited into a multi-center retrospective study described previously [27]. Briefly, all patients were prescribed metformin monotherapy as their initial hypoglycemic medication, had HbA1c levels measured both before and after commencement of metformin treatment and a medication possession ratio greater than 80%. The Institutional Review Boards of the Marshfield Clinic Research Foundation, Kaiser Permanente South East, Georgia and UCSF approved this study and informed consent was obtained. The review process for BioVU has been previously described [48].

Statistical Analysis

Data are presented as mean ± standard deviation unless indicated otherwise. Unpaired and paired nonparametric Student’s t-tests were used to analyze the differences in metformin pharmacokinetic and pharmacodynamic parameters, respectively, for each genotype using GraphPad Prism 4.0 (GraphPad Software, SD). A statistically significant result was defined when P<0.05. The 95% confidence intervals for the relative change in HbA1c were calculated by a nonparametric bootstrap method using the R software package (www.r-project.org, version 2.12.0). Linear regression and multivariate analyses were carried out using the R software.

All analyses were conducted firstly, by assessing the effect of each variant alone and secondly, after the exclusion of individuals carrying confounding genotypes on the pharmacokinetics and pharmacodynamics of metformin. A recessive genetic model was used, unless the analyses were underpowered, whereby a dominant genetic model was used. In the secondary analyses we removed individuals who were carrying at least one minor allele for MATE1 (g.−66T>C), MATE2 (g.−130G>A), a reduced-function OCT1 coding variant (R61C, G401S, 420del and G465R) or OCT2 coding variant (A270S) that could confound the measured parameters. This secondary analysis was conducted because the MATE1 (g.−66T>C) variant opposes the effects of the MATE2 (g.−130G>A) variant (i.e. the MATE1 and MATE2 polymorphisms result in reduced and enhanced expression, respectively, in reporter assays). In addition, coding variants in OCT1 and OCT2, are known to effect metformin pharmacodynamics [11, 12, 17, 18] and pharmacokinetics [15, 23, 24], respectively, and could confound the effects of MATE variants on the disposition and response to metformin.

Figure 2
The effect of MATE1 (g.−66T>C) on the pharmacokinetics of metformin in 57 healthy volunteers

STUDY HIGHLIGHTS

What is the current knowledge on the topic?

Previous studies have focused on the effect of less common nonsynonymous and/or intronic variants in membrane transporters on metformin pharmacokinetics and response.

What question this study addressed?

In this study, we investigated whether two common promoter variants in multidrug and toxic extrusion 1 (MATE1) and 2 (MATE2) affect the pharmacokinetics and pharmacodynamics of metformin in healthy volunteers and in patients with type II diabetes.

What this study adds to our knowledge?

This study provides evidence that common functional promoter variants in MATE1 and MATE2 have a significant impact on the pharmacokinetics and pharmacodynamics of metformin.

How this might change clinical pharmacology and therapeutics?

This study adds to the growing body of literature demonstrating that genetic variants in membrane transporters significantly influence the pharmacokinetics and response of metformin. In future, genotyping of MATE1 and MATE2 may be used to inform metformin therapy.

Supplementary Material

Supplementary Figure 1

Supplemental Figure S1. The expression profile of MATE1 and MATE2/2K in various human tissues:

Total RNA isolated from various human tissues (BioChain, Hayward, CA) were reverse transcribed and mRNA levels of MATE1 and MATE2 were determined by real-time PCR [49] using commercially available primer-probe sets (Life Technologies, Grand Island, NY). The delta cycle threshold (CT) values for all the genes in each sample were calculated by subtracting the mean CT values for three housekeeping genes (GAPDH, β-Actin, and PGK-1) from the CT for each target gene. The relative quantity of each gene was then determined by calculating the 2-ΔCT value. Data represent the mean and standard deviation from two experiments.

Supplementary Table 1

Supplementary Table 2

Supplementary Table 3

ACKNOWLEDGEMENTS

This project was supported by NIH/NCRR UCSF-CTSI Grant Number UL1 RR024131 and NIH (GM61390). Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the NIH. This work was also made possible in part by core services provided by the General Clinical Research Center (GCRC) at San Francisco General Hospital (SFGH), funded by the National Center for Research Resources (NIH MO1-RR00083-44). We thank Hector Vizoso and all the staff at GCRC at SFGH for their professionalism and enthusiastic assistance in our clinical study. We thank Yong Huang at the UCSF Drugs Services Unit for access to their bioanalytical facilities and Chav Doherty from UCSF SFGH Clinical Laboratory for their excellent service. The authors would also like to acknowledge the following funding sources for S.L.S. (Allergan), K.M.M. (NIH Training Grant T32 GM007175, FDA-CDER/ORISE) and A.D. (NIH R25 CA112355).

Footnotes

CONFLICT OF INTEREST

The authors declare no conflict of interest

AUTHOR CONTRIBUTIONS

SLS: Development of study protocol, coordinating and conducting study, bioanalytical analysis, data analysis, manuscript preparation

KMM: Bioanalytical analysis, manuscript preparation

SWY: Data analysis, manuscript preparation

RAC: Development of study protocol, co-ordination of study, medical assistance

LX: Manuscript preparation

AD: Development of study protocol

AHR: Contributed clinical data

DMR: Contributed clinical data

RAW: Contributed clinical data

CM: Contributed clinical data

RLD: Contributed clinical data

CMB: Medical assistance

KMG: Development of study protocol, manuscript preparation

.

This work was presented, in part, at the 113th Annual Meeting of the American Society for Clinical Pharmacology and Therapeutics, National Harbor, MD (March 2012).

REFERENCES

1. Alexander GC, Sehgal NL, Moloney RM, Stafford RS. National trends in treatment of type 2 diabetes mellitus: 1994–2007. Arch Intern Med. 2008;168:2088–2094. [PMC free article] [PubMed]
2. Zhou G, Myers R, Li Y, Chen Y, Shen X, Fenyk-Melody J, et al. Role of AMP-activated protein kinase in mechanism of metformin action. J Clin Invest. 2001;108:1167–1174. [PMC free article] [PubMed]
3. Cusi K, Consoli A, DeFronzo RA. Metabolic effects of metformin on glucose and lactate metabolism in noninsulin-dependent diabetes mellitus. J Clin Endocrinol Metab. 1996;81:4059–4067. [PubMed]
4. Pentikainen PJ, Neuvonen PJ, Penttila A. Pharmacokinetics of metformin after intravenous and oral administration to man. Eur J Clin Pharmacol. 1979;16:195–202. [PubMed]
5. Tucker GT, Casey C, Phillips PJ, Connor H, Ward JD, Woods HF. Metformin kinetics in healthy subjects and in patients with diabetes mellitus. Br J Clin Pharmacol. 1981;12:235–246. [PubMed]
6. Sirtori CR, Franceschini G, Galli-Kienle M, Cighetti G, Galli G, Bondioli A, et al. Disposition of metformin (N,N-dimethylbiguanide) in man. Clin Pharmacol Ther. 1978;24:683–693. [PubMed]
7. Graham GG, Punt J, Arora M, Day RO, Doogue MP, Duong JK, et al. Clinical pharmacokinetics of metformin. Clin Pharmacokinet. 2011;50:81–98. [PubMed]
8. Yin OQ, Tomlinson B, Chow MS. Variability in renal clearance of substrates for renal transporters in chinese subjects. J Clin Pharmacol. 2006;46:157–163. [PubMed]
9. Leabman MK, Giacomini KM. Estimating the contribution of genes and environment to variation in renal drug clearance. Pharmacogenetics. 2003;13:581–584. [PubMed]
10. Cook MN, Girman CJ, Stein PP, Alexander CM. Initial monotherapy with either metformin or sulphonylureas often fails to achieve or maintain current glycaemic goals in patients with Type 2 diabetes in UK primary care. Diabet Med. 2007;24:350–358. [PubMed]
11. Shu Y, Brown C, Castro RA, Shi RJ, Lin ET, Owen RP, et al. Effect of genetic variation in the organic cation transporter 1, OCT1, on metformin pharmacokinetics. Clin Pharmacol Ther. 2008;83:273–280. [PMC free article] [PubMed]
12. Shu Y, Sheardown SA, Brown C, Owen RP, Zhang S, Castro RA, et al. Effect of genetic variation in the organic cation transporter 1 (OCT1) on metformin action. J Clin Invest. 2007;117:1422–1431. [PMC free article] [PubMed]
13. Tzvetkov MV, Vormfelde SV, Balen D, Meineke I, Schmidt T, Sehrt D, et al. The effects of genetic polymorphisms in the organic cation transporters OCT1, OCT2, and OCT3 on the renal clearance of metformin. Clin Pharmacol Ther. 2009;86:299–306. [PubMed]
14. Becker ML, Visser LE, van Schaik RH, Hofman A, Uitterlinden AG, Stricker BH. Interaction between polymorphisms in the OCT1 and MATE1 transporter and metformin response. Pharmacogenet Genomics. 2010;20:38–44. [PubMed]
15. Chen Y, Li S, Brown C, Cheatham S, Castro RA, Leabman MK, et al. Effect of genetic variation in the organic cation transporter 2 on the renal elimination of metformin. Pharmacogenet Genomics. 2009;19:497–504. [PMC free article] [PubMed]
16. Masuda S, Terada T, Yonezawa A, Tanihara Y, Kishimoto K, Katsura T, et al. Identification and functional characterization of a new human kidney-specific H+/organic cation antiporter, kidney-specific multidrug and toxin extrusion 2. J Am Soc Nephrol. 2006;17:2127–2135. [PubMed]
17. Christensen MM, Brasch-Andersen C, Green H, Nielsen F, Damkier P, Beck-Nielsen H, et al. The pharmacogenetics of metformin and its impact on plasma metformin steady-state levels and glycosylated hemoglobin A1c. Pharmacogenet Genomics. 2011;21:837–850. [PubMed]
18. Becker ML, Visser LE, van Schaik RH, Hofman A, Uitterlinden AG, Stricker BH. Genetic variation in the organic cation transporter 1 is associated with metformin response in patients with diabetes mellitus. Pharmacogenomics J. 2009;9:242–247. [PubMed]
19. Konig J, Zolk O, Singer K, Hoffmann C, Fromm MF. Double-transfected MDCK cells expressing human OCT1/MATE1 or OCT2/MATE1: determinants of uptake and transcellular translocation of organic cations. Br J Pharmacol. 2011;163:546–555. [PubMed]
20. Sato T, Masuda S, Yonezawa A, Tanihara Y, Katsura T, Inui K. Transcellular transport of organic cations in double-transfected MDCK cells expressing human organic cation transporters hOCT1/hMATE1 and hOCT2/hMATE1. Biochem Pharmacol. 2008;76:894–903. [PubMed]
21. Becker ML, Visser LE, van Schaik RH, Hofman A, Uitterlinden AG, Stricker BH. Genetic variation in the multidrug and toxin extrusion 1 transporter protein influences the glucose-lowering effect of metformin in patients with diabetes: a preliminary study. Diabetes. 2009;58:745–749. [PMC free article] [PubMed]
22. Meyer zu Schwabedissen HE, Verstuyft C, Kroemer HK, Becquemont L, Kim RB. Human multidrug and toxin extrusion 1 (MATE1/SLC47A1) transporter: functional characterization, interaction with OCT2 (SLC22A2), and single nucleotide polymorphisms. Am J Physiol Renal Physiol. 2010;298:F997–F1005. [PubMed]
23. Wang ZJ, Yin OQ, Tomlinson B, Chow MS. OCT2 polymorphisms and in-vivo renal functional consequence: studies with metformin and cimetidine. Pharmacogenet Genomics. 2008;18:637–645. [PubMed]
24. Song IS, Shin HJ, Shim EJ, Jung IS, Kim WY, Shon JH, et al. Genetic variants of the organic cation transporter 2 influence the disposition of metformin. Clin Pharmacol Ther. 2008;84:559–562. [PubMed]
25. Tsuda M, Terada T, Mizuno T, Katsura T, Shimakura J, Inui K. Targeted disruption of the multidrug and toxin extrusion 1 (mate1) gene in mice reduces renal secretion of metformin. Mol Pharmacol. 2009;75:1280–1286. [PubMed]
26. Toyama K, Yonezawa A, Tsuda M, Masuda S, Yano I, Terada T, et al. Heterozygous variants of multidrug and toxin extrusions (MATE1 and MATE2-K) have little influence on the disposition of metformin in diabetic patients. Pharmacogenet Genomics. 2010;20:135–138. [PubMed]
27. Choi JH, Yee SW, Ramirez AH, Morrissey KM, Jang GH, Joski PJ, et al. A common 5'-UTR variant in MATE2-K is associated with poor response to metformin. Clin Pharmacol Ther. 2011;90:674–684. [PMC free article] [PubMed]
28. Ha Choi J, Wah Yee S, Kim MJ, Nguyen L, Ho Lee J, Kang JO, et al. Identification and characterization of novel polymorphisms in the basal promoter of the human transporter, MATE1. Pharmacogenet Genomics. 2009;19:770–780. [PMC free article] [PubMed]
29. Timmins P, Donahue S, Meeker J, Marathe P. Steady-state pharmacokinetics of a novel extended-release metformin formulation. Clin Pharmacokinet. 2005;44:721–729. [PubMed]
30. Rao N, Chou T, Ventura D, Abramowitz W. Investigation of the pharmacokinetic and pharmacodynamic interactions between memantine and glyburide/metformin in healthy young subjects: a single-center, multiple-dose, open-label study. Clin Ther. 2005;27:1596–1606. [PubMed]
31. Sambol NC, Chiang J, O'Conner M, Liu CY, Lin ET, Goodman AM, et al. Pharmacokinetics and pharmacodynamics of metformin in healthy subjects and patients with noninsulin-dependent diabetes mellitus. J Clin Pharmacol. 1996;36:1012–1021. [PubMed]
32. Sambol NC, Chiang J, Lin ET, Goodman AM, Liu CY, Benet LZ, et al. Kidney function and age are both predictors of pharmacokinetics of metformin. J Clin Pharmacol. 1995;35:1094–1102. [PubMed]
33. Filipski KK, Mathijssen RH, Mikkelsen TS, Schinkel AH, Sparreboom A. Contribution of organic cation transporter 2 (OCT2) to cisplatin-induced nephrotoxicity. Clin Pharmacol Ther. 2009;86:396–402. [PMC free article] [PubMed]
34. Jablonski KA, McAteer JB, de Bakker PI, Franks PW, Pollin TI, Hanson RL, et al. Common variants in 40 genes assessed for diabetes incidence and response to metformin and lifestyle intervention in the diabetes prevention program. Diabetes. 2010;59:2672–2681. [PMC free article] [PubMed]
35. Otsuka M, Matsumoto T, Morimoto R, Arioka S, Omote H, Moriyama Y. A human transporter protein that mediates the final excretion step for toxic organic cations. Proc Natl Acad Sci U S A. 2005;102:17923–17928. [PubMed]
36. Tanihara Y, Masuda S, Sato T, Katsura T, Ogawa O, Inui K. Substrate specificity of MATE1 and MATE2-K, human multidrug and toxin extrusions/H(+)-organic cation antiporters. Biochem Pharmacol. 2007;74:359–371. [PubMed]
37. Little RR, Rohlfing CL. Analytical goals for HbA1c: are HbA1c results good enough for optimal use? J Diabetes. 2011;3:3–6. [PubMed]
38. Hirst JA, Farmer AJ, Ali R, Roberts NW, Stevens RJ. Quantifying the effect of metformin treatment and dose on glycemic control. Diabetes Care. 2012;35:446–454. [PMC free article] [PubMed]
39. Astorga B, Ekins S, Morales M, Wright SH. Molecular determinants of ligand selectivity for the human multidrug and toxin extruder proteins MATE1 and MATE2-K. J Pharmacol Exp Ther. 2012;341:743–755. [PubMed]
40. Kusuhara H, Ito S, Kumagai Y, Jiang M, Shiroshita T, Moriyama Y, et al. Effects of a MATE protein inhibitor, pyrimethamine, on the renal elimination of metformin at oral microdose and at therapeutic dose in healthy subjects. Clin Pharmacol Ther. 2011;89:837–844. [PubMed]
41. Scheen AJ. Clinical pharmacokinetics of metformin. Clin Pharmacokinet. 1996;30:359–371. [PubMed]
42. Uhlen M, Oksvold P, Fagerberg L, Lundberg E, Jonasson K, Forsberg M, et al. Towards a knowledge-based Human Protein Atlas. Nat Biotechnol. 2010;28:1248–1250. [PubMed]
43. McGuinness OP, Fugiwara T, Murrell S, Bracy D, Neal D, O'Connor D, et al. Impact of chronic stress hormone infusion on hepatic carbohydrate metabolism in the conscious dog. Am J Physiol. 1993;265:E314–E322. [PubMed]
44. Gerich JE. Role of the kidney in normal glucose homeostasis and in the hyperglycaemia of diabetes mellitus: therapeutic implications. Diabet Med. 2010;27:136–142. [PMC free article] [PubMed]
45. Gerich JE. Physiology of glucose homeostasis. Diabetes Obes Metab. 2000;2:345–350. [PubMed]
46. Meyer C, Dostou JM, Welle SL, Gerich JE. Role of human liver, kidney, and skeletal muscle in postprandial glucose homeostasis. Am J Physiol Endocrinol Metab. 2002;282:E419–E427. [PubMed]
47. Meyer C, Woerle HJ, Dostou JM, Welle SL, Gerich JE. Abnormal renal, hepatic, and muscle glucose metabolism following glucose ingestion in type 2 diabetes. Am J Physiol Endocrinol Metab. 2004;287:E1049–E156. [PubMed]
48. Roden DM, Pulley JM, Basford MA, Bernard GR, Clayton EW, Balser JR, et al. Development of a large-scale de-identified DNA biobank to enable personalized medicine. Clin Pharmacol Ther. 2008;84:362–369. [PMC free article] [PubMed]
49. Chen L, Pawlikowski B, Schlessinger A, More SS, Stryke D, Johns SJ, et al. Role of organic cation transporter 3 (SLC22A3) and its missense variants in the pharmacologic action of metformin. Pharmacogenet Genomics. 2010;20:687–699. [PMC free article] [PubMed]