Role of organic cation transporter 3 (SLC22A3) and its missense variants in the pharmacologic action of metformin

Abstract

Objectives: The goals of this study were to determine the role of organic cation transporter 3 (OCT3) in the pharmacological action of metformin and to identify and functionally characterize genetic variants of OCT3 with respect to the uptake of metformin and monoamines.

Methods: For pharmacological studies, we evaluated metformin-induced activation of AMP-activated protein kinase, a molecular target of metformin. We used quantitative PCR and immunostaining to localize the transporter and isotopic uptake studies in cells transfected with OCT3 and its nonsynonymous genetic variants for functional analyses.

Results: Quantitative PCR and immunostaining showed that OCT3 was expressed high on the plasma membrane of skeletal muscle and liver, target tissues for metformin action. Both the OCT inhibitor, cimetidine, and OCT3-specific short hairpin RNA significantly reduced the activating effect of metformin on AMP-activated protein kinase. To identify genetic variants in OCT3, we used recent data from the 1000 Genomes and the Pharmacogenomics of Membrane Transporters projects. Six novel missense variants were identified. In functional assays, using various monoamines and metformin, three variants, T44M (c.131C>T), T400I (c.1199C>T) and V423F (c.1267G>T) showed altered substrate specificity. Notably, in cells expressing T400I and V423F, the uptakes of metformin and catecholamines were significantly reduced, but the uptakes of metformin, 1-methyl-4-phenylpyridinium and histamine by T44M were significantly increased more than 50%. Structural modeling suggested that these two variants may be located in the pore lining (T400) or proximal (V423) membrane-spanning helixes.

Conclusion: Our study suggests that OCT3 plays a role in the therapeutic action of metformin and that genetic variants of OCT3 may modulate metformin and catecholamine action.

Figure 1. Tissue distribution of OCT3 and its paralogs

A. mRNA expression levels of OCT1, OCT2 and OCT3 in various tissues. OCT1-3 mRNA expression was normalized to hGAPDH expression for each tissue. Data are shown as mean ± SD from a single experiment with triplicate wells and plotted as log. U.D. (undetermined). B. Immunofluorescence of human skeletal muscle, heart, kidney and liver sections probed with rabbit anti-OCT3 at 1:100 and detected with 2^nd antibody Alexa Fluor® 488 (green). Nuclei were stained with DAPI (blue). Scale bar: 50 µm. C. OCT3 antibody specificity tested in HEK-Mock, HEK-OCT1-3.

Figure 1. Tissue distribution of OCT3 and its paralogs

A. mRNA expression levels of OCT1, OCT2 and OCT3 in various tissues. OCT1-3 mRNA expression was normalized to hGAPDH expression for each tissue. Data are shown as mean ± SD from a single experiment with triplicate wells and plotted as log. U.D. (undetermined). B. Immunofluorescence of human skeletal muscle, heart, kidney and liver sections probed with rabbit anti-OCT3 at 1:100 and detected with 2^nd antibody Alexa Fluor® 488 (green). Nuclei were stained with DAPI (blue). Scale bar: 50 µm. C. OCT3 antibody specificity tested in HEK-Mock, HEK-OCT1-3.

Figure 2. Metformin activation of AMPK in skeletal muscle cells

A. Western blot assay of AMPK activation mediated by OCT3 in human primary skeletal muscle cells. Metformin (1.5 mM) stimulated AMPK α2 (T-AMPK) Thr172 phosphorylation (P-AMPK) in human adult primary skeletal muscle cell and that the OCT inhibitor, cimetidine (1 mM) and shRNA, inhibited its action on AMPK phosphorylation. pGFP-V-RS: empty vector without shRNA. B, Densitometry of immunoblotting assays. The bar chart represents the intensity of each band (P-AMPK) normalized to the total AMPK based on 3 independent experiments. C, Validation of shRNA knock-down of mRNA expression level of OCT3 in human primary skeletal muscle cells by RT-PCR. D. TaqMan quantitative PCR analysis of OCT3 mRNA expression level in shRNA or empty vector treated human primary skeletal muscle cells.

Figure 3. Uptake of various endogenous amines and MPP+ in HEK293 cells stably expressing OCT3 and 6 missense variants

A. The time course of uptake of monoamines and MPP+. B. Substrate selectivity of genetic variants of OCT3 on model substrate and monoamines. Data are shown as mean ± SD from 3 repeated experiments and each experiment was performed in triplicate. OCT3-reference is used as control set at 100% for each substrate. **p<0.01 versus OCT3 – reference.

Figure 3. Uptake of various endogenous amines and MPP+ in HEK293 cells stably expressing OCT3 and 6 missense variants

A. The time course of uptake of monoamines and MPP+. B. Substrate selectivity of genetic variants of OCT3 on model substrate and monoamines. Data are shown as mean ± SD from 3 repeated experiments and each experiment was performed in triplicate. OCT3-reference is used as control set at 100% for each substrate. **p<0.01 versus OCT3 – reference.

Figure 4. Metformin uptake and kinetics in HEK293 cells expressing OCT3 and its genetic variants

A, Metformin uptake in HEK293 cells expressing reference OCT3 and 6 missense variants. OCT3-reference is used as control for each substrate. ***p< 0.001 versus OCT3–reference. B. Metformin kinetics in HEK293 cells expressing reference OCT3 and 6 missense variants. Michaelis-Menten parameters were calculated and are shown in Table 3B.

Figure 5. Subcellular localization and expression quantification of GFP-tagged various OCT3

A. Subcellular localization of GFP tagged OCT3 reference and 4 nonsynonymous variants (GFP-T44M, GFP-A116S, GFP-T400I and GFP-V423F). Various OCT3 GFP fusion constructs were stably expressed in HEK-293 cells and visualized by fluorescent microscopy. The nucleus is stained with DAPI (blue). The plasma membrane was stained using AlexaFluor 594-labeled wheat germ agglutinin (red) for co-localization staining. GFP-OCT3 protein is shown in green. B. Western blotting assay of GFP-Mock, GFP-tagged OCT3-reference and 4 variants above. Mouse monoclonal anti- Na+/K+ ATPase served as loading control. The GFP-OCT3 position is around 100 kD shifted from the 60 kD position of OCT3 itself. The bar chart represents the intensity of each band normalized to the Na+/K+ ATPase band relative to the GFP-OCT3-reference based on 3 independent assays.

Figure 6. Structural analysis of OCT3 genetic variants, T400 and V423F

A. Multiple sequence alignments of OCT3 in various species and with its orthologs, human OCT1 and OCT2. The proteins in each species are known as OCT3 or predicted organic cation transporters. T400 and V423F are highlighted in the frame. The color scheme in the protein sequence is based on the amino acid residues and the score under the yellow bar indicates their conservation rate at their position with conserved residues labeled as *. B. Chemical structures of endogenous monoamine substrates of OCT3 are illustrated here. The common shared moiety, ethylamine is shown as blue and the unique hydroxyl group in the ethylamine of norepinephrine and epinephrine is colored as yellow. The catecholamine specific hydroxyl group in the phenyl ring is indicated as red.