MDR1 synonymous polymorphisms alter transporter specificity and protein stability in a stable epithelial monolayer

Abstract

The drug efflux function of P-glycoprotein (P-gp) encoded by MDR1 can be influenced by genetic polymorphisms, including two synonymous changes in the coding region of MDR1. Here we report that the conformation of P-gp and its drug efflux activity can be altered by synonymous polymorphisms in stable epithelial monolayers expressing P-gp. Several cell lines with similar MDR1 DNA copy number were developed and termed LLC-MDR1-WT (expresses wild-type P-gp), LLC-MDR1-3H (expresses common haplotype P-gp), and LLC-MDR1-3HA (a mutant that carries a different valine codon in position 3435). These cell lines express similar levels of recombinant mRNA and protein. P-gp in each case is localized on the apical surface of polarized cells. However, the haplotype and its mutant P-gps fold differently from the wild-type, as determined by UIC2 antibody shift assays and limited proteolysis assays. Surface biotinylation experiments suggest that the non-wild-type P-gps have longer recycling times. Drug transport assays show that wild-type and haplotype P-gp respond differently to P-gp inhibitors that block efflux of rhodamine 123 or mitoxantrone. In addition, cytotoxicity assays show that the LLC-MDR1-3H cells are more resistant to mitoxantrone than the LLC-MDR1-WT cells after being treated with a P-gp inhibitor. Expression of polymorphic P-gp, however, does not affect the host cell's morphology, growth rate, or monolayer formation. Also, ATPase activity assays indicate that neither basal nor drug-stimulated ATPase activities are affected in the variant P-gps. Taken together, our findings indicate that "silent" polymorphisms significantly change P-gp function, which would be expected to affect interindividual drug disposition and response.

Figure 1

Characterization of LLC-MDR1 cell lines. A, Copy number determination by Southern blot analysis. Genomic DNA samples were run with a blank control and copy number standards. B, qRT-PCR analysis. The mean values are the average of 3 experiments and are plotted in the histogram. C, Recombinant P-gp expression by Western analysis. LLC-vector (lane1), LLC-MDR1-WT (lane 2), LLC-MDR1-3H (lane 3), LLC-MDR1-3HA (lane 4), KB-V1 (lane 5), and KB-8-5 (lane 6) are shown. The bar chart shows relative expression level of P-gp in each cell line calculated by densitometry analysis. D, Cellular morphology and localization of P-gp. Top: Cells visualized by light microscopy. Bottom: X–Z plane of the LLC-MDR1 cells. Cell monolayers were labeled with MRK16 (green, for P-gp), biotin-strepavdin-PE (red, for plasma membrane) and DAPI (blue, for nucleus). E, LLC-MDR1 cell lines have comparable growth rates. LLC-vector (black), LLC-MDR1-WT (green), LLC-MDR1-3H (red), LLC-MDR1-3HA (blue) cells are shown. Each point in the chart represents an average cell number from 3 samples.

Figure 2

Stably-expressed haplotype and mutant P-gps have altered protein conformations. A, FACS analysis of cells labeled with different anti-Pgp antibodies. Each histogram represents at least 2 independent experiments. B, Sensitivity of MDR1 wild-type (circle), haplotype (square), and mutant P-gp (triangle) to trypsin. Crude membranes prepared from the cell lines were treated with, from left to right, 0.000, 0.625, 1.250, 1.875, 2.500, 3.125, 3.750, 4.375, and 5.000 trypsin (μg). Charts and Western blots with trypsin (left) and with trypsin + 20 μM verapamil (right) are shown. The upper panel shows quantified data from Western blots shown below. Each point in the plot represents an average cell number from 3 independent experiments.

Figure 3

Mutant and haplotype P-gps remain longer on the cell surface. LLC-MDR1-WT (circles), LLC-MDR1-3H (squares) and LLC-MDR1-3HA (triangles) cells were labeled with biotin and MRK16 for FACS analysis. The percentage of biotin-labeled MRK16 positive cells remaining at each time point was plotted. Each point in the plot represents an average cell number from 3 independent experiments.

Figure 4

Variant P-gps exhibit different drug efflux functions in a substrate-dependent manner. Cells were incubated with rhodamine 123 for 20 min at 37°C, followed by 40 min efflux at 37°C and then FACS analysis. Assays were repeated at least 3 times.

Figure 5

P-gp transfected cell lines show differential sensitivity to cytotoxic drugs. LLC-vector (black), LLC-MDR1-WT (green), LLC-MDR1-3H (red), LLC-MDR1-3HA (blue) cells were tested with cytotoxicity assays (left panel) and drug efflux assays (right panel). For cytotoxicity assays, cells were incubated with increasing concentrations of drugs for 72 hrs. For drug accumulation-efflux assays, cells were incubated with mitoxantrone ± DCPQ for 30 min followed by 45 min efflux. For bodipy-vinblastine ± cyclosporine A, cells were incubated for 20 min followed by 30 min efflux. Assays were repeated at least 3 times. Each point in the cytotoxicity assay charts represents an average cell number from 3 independent experiments.

Figure 6

P-gp variant forms do not differ from wild-type in ATPase and photoaffinity labeling Assays. A, Substrate-stimulated ATPase activities using crude membranes from LLC-MDR1-WT (black) and LLC-MDR1-3H (gray) cells. B, [¹²⁵I]-IAAP Photoaffinity labeling of P-gp wild-type and haplotype in the absence (lane 1) and the presence of 10 μM tasigna (lane 2), 10 μM verapamil (lane 3), 10 μM tariquidar (lane 4), 10 μM vinblastine (lane 5), and 5 μM paclitaxel (lane 6). The scatter plot shows the incorporation of [¹²⁵I]-IAAP in LLC-MDR1-WT (circle) and LLC-MDR1-3H (triangle) membrane proteins with verapamil from 0 to 100 μM. Autoradiographs from representative experiments are shown. Similar results were obtained in two additional experiments.