Identification of the efflux transporter of the fluoroquinolone antibiotic ciprofloxacin in murine macrophages: studies with ciprofloxacin-resistant cells

Abstract

Ciprofloxacin, the most widely used totally synthetic antibiotic, is subject to active efflux mediated by a MRP-like transporter in wild-type murine J774 macrophages. To identify the transporter among the seven potential Mrps, we used cells made resistant to ciprofloxacin obtained by long-term exposure to increasing drug concentrations (these cells show less ciprofloxacin accumulation and provide a protected niche for ciprofloxacin-sensitive intracellular Listeria monocytogenes). In the present paper, we first show that ciprofloxacin-resistant cells display a faster efflux of ciprofloxacin which is inhibited by gemfibrozil (an unspecific MRP inhibitor). Elacridar, at a concentration known to inhibit P-glycoprotein and breast cancer resistance protein (BCRP), only slightly increased ciprofloxacin accumulation, with no difference between resistant and wild-type cells. Analysis at the mRNA (real-time PCR) and protein (Western blotting) levels revealed an overexpression of Mrp2 and Mrp4. Mrp4 transcripts, however, were overwhelmingly predominant (45% [wild-type cells] to 95% [ciprofloxacin-resistant cells] of all Mrp transcripts tested [Mrp1 to Mrp7]). Silencing of Mrp2 and Mrp4 with specific small interfering RNAs showed that only Mrp4 is involved in ciprofloxacin transport in both ciprofloxacin-resistant and wild-type cells. The study therefore identifies Mrp4 as the most likely transporter of ciprofloxacin in murine macrophages but leaves open a possible common upregulation mechanism for both Mrp4 and Mrp2 upon chronic exposure of eukaryotic cells to this widely used antibiotic.

FIG. 1.

Influence of gemfibrozil on the accumulation and efflux of ciprofloxacin in wild-type and ciprofloxacin-resistant J774 macrophages. Data are the means of three independent measurements ± standard deviations (SD) (when not visible, the SD bar is smaller than the symbol). (A) Cells were incubated with 50 μM ciprofloxacin for 2 h at 37°C in the presence of gemfibrozil at concentrations ranging from 0 to 1 mM. Regression parameters (nonlinear, sigmoidal dose-response curve [Hill's coefficient = 1]) for wild-type and ciprofloxacin-resistant cells were as follows: R² = 0.975 and 0.956, respectively; and EC₅₀ (μM [95% confidence interval]; shown by the open and closed triangles on the abscissa) = 58.1 (30.1 to 112.4) and 338.7 (142.6 to 804.4), respectively. (B) Cells were incubated for 2 h at 37°C with 50 μM ciprofloxacin alone for wild-type cells and with 50 μM ciprofloxacin plus 200 μM gemfibrozil for ciprofloxacin-resistant cells (to reach cell contents allowing for sufficiently accurate measurements during efflux from both cell types; actual initial values, 338 ± 23 and 474 ± 44 ng/mg protein for wild-type and ciprofloxacin-resistant cells, respectively), transferred to ciprofloxacin-free medium in the absence (left) or presence of 1 mM gemfibrozil (right), and reincubated for up to 30 min at 37°C (for the sake of clarity, the graphs show only the data recorded during the first 5 min, but all data points were used for analysis). Results are expressed as percentages of the ciprofloxacin cell content observed before transfer to ciprofloxacin-free medium. Data were best fitted to a two-phase exponential decay function, with the first phase covering more than 70% of the analyzed time span. Regression parameters for wild-type and ciprofloxacin-resistant cells in the absence of gemfibrozil and for wild-type and ciprofloxacin-resistant cells in the presence of gemfibrozil were as follows: R² = 0.975, 0.998, 0.990, and 0.945, respectively; and initial half-lives (minutes [95% confidence interval]) = 1.75 (1.31 to 2.64), 0.09 (0.07 to 0.11), 4.49 (3.71 to 5.68), and 4.35 (2.90 to 8.68), respectively (the second phase is not visible on the graphs, except for ciprofloxacin-resistant cells in the absence of gemfibrozil, for which the first phase is sufficiently rapid).

FIG. 2.

Quantification of mRNA transcripts of Mrps 1 to 7 in ciprofloxacin-resistant (RS) and revertant (Rev) J774 macrophages in comparison with wild-type cells (WT). (Top) Increase in expression over that in WT cells (set arbitrarily at 1 [dotted line]). Values of all samples were normalized using Ywhaz and Rpl13A as housekeeping genes. Ratios shown are means ± SD (n = 3). Statistical analysis was done by one-way analysis of variance with the Dunnett multiple-comparison test. ***, P < 0.001; **, P < 0.01; *, P < 0.05 (compared to values in wild-type cells). (Bottom) mRNA transcripts (copy number determined by real-time PCR) expressed as percentages of the total number of Mrp transcripts detected, starting from 1 μg purified total RNA for each cell type. Data were obtained from the mean value for the number of copies calculated for each Mrp in each cell type. Note that the boxes corresponding to Mrp2 and Mrp6 are not visible because the values are smaller than the corresponding surrounding lines.

FIG. 3.

Western blots of proteins prepared from wild-type (WT), ciprofloxacin-resistant (RS), and revertant (Rev) J774 macrophages. Gels were loaded with the amounts of protein indicated. (Top) Whole-cell lysates, with revelation with anti-Mrp4 (upper row; 1:1,000) or anti-actin (lower row; 1:600) antibodies, followed by anti-immunoglobulin G (anti-IgG) HRP-labeled antibodies (1:1,500). (Bottom) Membrane proteins, with revelation with anti-Mrp4 antibody (upper row; 1:2,000) or anti-Mrp2 antibody (lower row; 1:200), followed by anti-IgG HRP-labeled antibody (1:1,500). Note that enriched membrane samples did not contain actin.

FIG. 4.

Confocal microscopy of permeabilized J774 macrophages. WT, wild type; RS, ciprofloxacin resistant. (A) Cells stained with rhodamine-labeled phalloidin (to label actin [red channel]) and monoclonal rat anti-Mrp4 antibody (followed by fluorescein isothiocyanate-labeled anti-rat IgG antibodies [green channel]). (B) Cells stained with rhodamine-labeled phalloidin (to label actin [red channel]) and polyclonal rabbit anti-Mrp2 antibodies (followed by fluorescein isothiocyanate-labeled anti-rabbit IgG antibodies [green channel]). (C) Cells stained with monoclonal rat anti-Mrp4 antibody (followed by Texas Red-labeled polyclonal anti-rat IgG antibodies [red channel]) and polyclonal rabbit anti-Mrp2 antibodies (followed by fluorescein isothiocyanate-labeled anti-rabbit IgG antibodies [green channel]). Note that there is no staining for actin in this panel. r, red channel only; g, green channel only; m, merged images.

FIG. 5.

Effect of siRNAs on expression of Mrp4 and accumulation of ciprofloxacin in ciprofloxacin-resistant J774 macrophages. Cells were either (i) transfected with a specific anti-Mrp4 siRNA (targeting exon 31 [284556] or exon 8 [284555]) or with a nontargeting siRNA (neg. contr.), each at 25 nM, or (ii) left untreated (none). (A) Mrp4 mRNA level determined by real-time PCR and expressed as a percentage of the level observed in untransfected cells (for comparison, the graph also shows the value observed in wild-type J774 macrophages [WT]). Data are from a typical experiment (with determinations made in duplicate). (B) Western blot analysis of whole-cell extracts of the corresponding cells. The gel was loaded with the same amount of protein for each sample (5 μg/well). (Upper row) Gel was revealed with anti-Mrp4 antibody (1:1,000) followed by the appropriate anti-IgG HRP-labeled antibody (1:666). (Lower row) Gel was revealed with anti-actin antibody (1:1,000) followed by the appropriate anti-IgG HRP-labeled antibody (1:1,500). (C) Accumulation of ciprofloxacin, expressed as the percent increase over that in untransfected cells. Cells were incubated with 50 μM ciprofloxacin for 2 h at 37°C. Data are the means of two independent experiments with measurements made in triplicate.

FIG. 6.

Effect of siRNAs on cell content of Mrp2 and accumulation of ciprofloxacin in ciprofloxacin-resistant J774 macrophages. Cells were either (i) transfected with a specific anti-Mrp2 siRNA (targeting exon 10 [161716] or exon 28 [71770]) or with a nontargeting siRNA (neg. contr.), each at 25 nM, or (ii) left untreated (none). (A) Western blot analysis of whole-cell extracts of the corresponding cells. The gel was loaded with the same amount of protein for each sample (45 μg/well). (Upper row) Gel was revealed with anti-Mrp2 antibody (1:200) followed by the appropriate anti-IgG HRP-labeled antibody (1:1,500). (Lower row) Gel was revealed with anti-actin antibody (1:1,000) followed by the appropriate anti-IgG HRP-labeled antibody (1:1,500). (B) Accumulation of ciprofloxacin, expressed as the percent increase over that in untransfected cells. Cells were incubated with 50 μM ciprofloxacin for 2 h at 37°C.