Targeted disruption of organic cation transporter 3 attenuates the pharmacologic response to metformin

Abstract

Metformin, the most widely prescribed antidiabetic drug, requires transporters to enter tissues involved in its pharmacologic action, including liver, kidney, and peripheral tissues. Organic cation transporter 3 (OCT3, SLC22A3), expressed ubiquitously, transports metformin, but its in vivo role in metformin response is not known. Using Oct3 knockout mice, the role of the transporter in metformin pharmacokinetics and pharmacodynamics was determined. After an intravenous dose of metformin, a 2-fold decrease in the apparent volume of distribution and clearance was observed in knockout compared with wild-type mice (P < 0.001), indicating an important role of OCT3 in tissue distribution and elimination of the drug. After oral doses, a significantly lower bioavailability was observed in knockout compared with wild-type mice (0.27 versus 0.58, P < 0.001). Importantly, metformin's effect on the plasma glucose concentration-time curve was reduced in knockout compared with wild-type mice (12 versus 30% reduction, respectively, P < 0.05) along with its accumulation in skeletal muscle and adipose tissue (P < 0.05). Furthermore, the effect of metformin on phosphorylation of AMP activated protein kinase, and expression of glucose transporter type 4 was absent in the adipose tissue of Oct3(-/-) mice. Additional analysis revealed that an OCT3 3' untranslated region variant was associated with reduced activity in luciferase assays and reduced response to metformin in 57 healthy volunteers. These findings suggest that OCT3 plays an important role in the absorption and elimination of metformin and that the transporter is a critical determinant of metformin bioavailability, clearance, and pharmacologic action.

Fig. 1.

Characterization of the kinetics and tissue levels of mouse OCT3. (A) Overexpressing mouse OCT3 increases metformin uptake in HEK cells. Metformin uptake studies were conducted in HEK cells overexpressing mouse OCT3 (▪) or empty vector (⬤). Cells were incubated with increasing concentrations of metformin for 3 minutes. The uptake kinetic parameters (see Results) were calculated using the difference in accumulation between OCT3-overexpressing and empty vector cells. (B) Oct3 mRNA expression pattern was assessed in C57/B6J mice. The relative mRNA levels of Oct3 were determined by real-time polymerase chain reaction. The mRNA levels of Oct1, Oct2, and Oct3 were determined in the liver (C), kidney (D), skeletal muscle (E), and adipose tissue (F). Data represent mean ± S.D., n = 3 mice per group.

Fig. 2.

Oct3 deletion resulted in altered pharmacokinetics and tissue accumulation of metformin in mice. (A) The plasma metformin concentration-time profiles in wild-type (WT) and knockout (KO) mice after an intravenous dose (50 mg/kg metformin containing 0.2 mCi/kg [¹⁴C]metformin via tail vein). Significantly higher plasma concentrations were observed in knockout mice at 5, 10, 15, 30, 60 minutes after injection. (B–E) Tissue distribution of metformin in wild-type and knockout mice. Two hours after dosing, mice were killed and tissues were collected. Radioactivity in tissue homogenates were determined and converted to mass amounts. Tissue accumulation was significantly reduced in liver and adipose tissue of knockout mice. Metformin accumulation was higher in kidney of knockout mice. Data represent mean ± S.D., n = 4 mice per data points, *P < 0.05 wild-type versus knockout mice.

Fig. 3.

Oct3 deletion resulted in altered tissue accumulation of metformin after an oral dose. (A) The plasma metformin concentration-time profiles in wild-type (WT) and knockout (KO) mice after an oral dose (50 mg/kg metformin containing 0.2 mCi/kg [¹⁴C]metformin via oral gavage). Radioactivity in the plasma is determined and converted to mass amounts. (B–E) Tissue distribution of metformin in wild-type and knockout mice. Four hours after administration, mice were killed and tissues were collected. Radioactivity in tissue homogenates was determined and converted to mass amounts. Metformin accumulation was significantly reduced in liver, kidney, skeletal muscle, and adipose tissue of Oct3 knockout mice. Data represent mean ± S.D., n = 4 mice per data point, *P < 0.05, **P < 0.01, wild-type versus knockout mice.

Fig. 4.

Oct3 deletion resulted in reduced pharmacologic effects of metformin. The blood glucose concentration-time profiles of knockout (KO) mice treated with metformin or saline (A) and wild-type (WT) mice treated with metformin or saline (B). Mice were given metformin (150 mg/kg) or saline orally followed by glucose (3 g/kg) orally after 15 minutes. Data represent mean ± S.D., n = 4 mice per data point, *P < 0.05 metformin-treated group versus saline treated group. (B–E) Tissue distribution of metformin in wild-type and knockout mice. Thirty minutes after metformin (150 mg/kg, p.o.), mice were killed and tissues were collected. Radioactivity in tissue homogenates were determined and converted to mass amounts. Data represent mean ± S.D., n = 4 mice per data point, *P < 0.05, **P < 0.01, wild-type versus knockout mice. (G) Representative Western blots of adipose tissues from knockout and wild-type mice treated with 100 mg/kg, i.p, metformin or saline for 5 days.

Fig. 5.

Functional characteristics of an OCT3 3′UTR variant (rs2076828) and its association with metformin pharmacologic response in healthy volunteers. The OCT3 3′UTR variant (C>G) in healthy human volunteers is associated with changes in metformin pharmacodynamics. (A) Luciferase reporter assay of rs2076828. Luciferase activity was significantly lower in HCT-116, DU145, and A549 cell lines transfected with a reporter construct containing the minor allele (G) compared with the reference allele (C). Data represent mean ± S.D., n = 3 replicates per cell line, *P < 0.05 and **P < 0.01. (B) E-QTL analysis of rs2076828 in healthy human cohorts revealed lower mRNA level of OCT3 in adipose tissues of subjects with the minor allele (G), P < 0.01. (C) The change in glucose exposure after an oral glucose tolerance test (AUC) after metformin treatment in healthy volunteers. Change in glucose AUC was calculated as glucose AUC after metformin dosing minus glucose AUC before metformin dosing. Healthy volunteers with the minor allele had lower response to metformin (P < 0.05). The dot plot displays the mean ± S.D.