Analysis of the membrane proteome of ciprofloxacin-resistant macrophages by stable isotope labeling with amino acids in cell culture (SILAC)

Abstract

Overexpression of multidrug transporters is a well-established mechanism of resistance to chemotherapy, but other changes may be co-selected upon exposure to drugs that contribute to resistance. Using a model of J774 macrophages made resistant to the fluoroquinolone antibiotic ciprofloxacin and comparing it with the wild-type parent cell line, we performed a quantitative proteomic analysis using the stable isotope labeling with amino acids in cell culture technology coupled with liquid chromatography electrospray ionization Fourier transform tandem mass spectrometry (LC-ESI-FT-MS/MS) on 2 samples enriched in membrane proteins (fractions F1 and F2 collected from discontinuous sucrose gradient). Nine hundred proteins were identified with at least 3 unique peptides in these 2 pooled fractions among which 61 (F1) and 69 (F2) showed a significantly modified abundance among the 2 cell lines. The multidrug resistance associated protein Abcc4, known as the ciprofloxacin efflux transporter in these cells, was the most upregulated, together with Dnajc3, a protein encoded by a gene located downstream of Abcc4. The other modulated proteins are involved in transport functions, cell adhesion and cytoskeleton organization, immune response, signal transduction, and metabolism. This indicates that the antibiotic ciprofloxacin is able to trigger a pleiotropic adaptative response in macrophages that includes the overexpression of its efflux transporter.

Figure 1. Plasma membrane enrichment strategy.

Left: Illustration of the discontinuous sucrose gradient. Sucrose solutions densities (see materials and methods) and the resulting interphases (F1 to F5) are depicted before (thin lines) and after centrifugation (irregular lines). Right: Expression of a plasma membrane protein MRP1 and prohibitin, a mitochondrial marker in sucrose interphases. MRP1 and prohibitin were detected by Western blot in each of the interphases of MDCKII-MRP1 cells (top) and wild-type J774 macrophages (bottom). Anti-MRP1 (1∶2000) and anti-prohibitin (1∶1000) antibodies were followed by the appropriate anti-IgG HRP-labeled antibodies.

Figure 2. Overview of the SILAC experimental approach.

Left: WT-J774 and CIP-R cells were cultured in media containing ¹³C₆-Lys/¹³C₆-Arg and ¹²C₆-Lys/¹²C₆-Arg, respectively. Following isolation of membrane samples for each strain independently, equal amounts from each matching fraction were pooled (F1_wt+F1_CIP and F2_wt+F2_CIP) and then separated by SDS-PAGE. Right: Lanes of pooled sample fractions were first cut in sections (A-K for pooled F1 and A-J for pooled F2) and further sliced in thinner, numbered gel plugs which were processed for Trypsine in-gel digestion as described in the experimental procedures. For pooled F1 gel line, section A includes gel plugs from 1 to 8, B 9 to 14, C 15 to 20, C 21 to 24, E 25 to 29, F 33 to 35, G 36 to 41, H 42 to 45, I 46 to 52, J 53 to 60 and K 61 to 65. For sample pooled F2 gel line, section A includes bands 1 to 6, B 7 to 11, C 12 to 18, D 19 to 25, E 26 to 32, F 33 to 39, G 40 to 47, H 48 to 56, I 57 to 65 and J 66 to 72.

Figure 3. Classification of the identified proteins.

Distribution charts of cellular localization (left) and biological function (right) of all proteins identified with high significance in both pooled F1 and F2 fractions. Proteins with multiple cellular localizations or biological functions are assigned to the one it is best known. Transporter proteins are involved in both transmembrane and intracellular transport.

Figure 4. Sample quantization and classification of modulated proteins.

Top: Histograms showing distribution of log₂ protein Light/Heavy ratios after normalization toward the median for proteins identified by 3 or more tryptic peptides from pooled F1 and F2 samples, respectively. After normalization the distributions fitted to a Gaussian curve indicating that protein ratios have a median of 0 on the log2 scale as expected for a 1∶1 mixture. Bottom: Distribution chart of the biological function of proteins that show a 95% significant differential abundance ratio in pooled fraction F1 or F2.

Figure 5. MS/MS spectrum of one of the Mrp4 peptides.

The MS/MS spectrum is showing a CID fragmentation profile which was measured in the LTQ ion trap mass analyzer. The double charged precursor ion (m/z 914.4136) was previously detected by the high-resolution FT-ICR mass analyzer. Matched y- and b-ions are indicated above and below the amino acid sequence given in the one letter code.

Figure 6. Validation by Western blot of differentially expressed proteins.

Western blots of membrane proteins prepared from wild-type (WT) and ciprofloxacin-resistant (CIP-R) J774 macrophages. Gels were loaded with the indicated amounts of proteins. Top: Detection of Dnajc3 with a rabbit anti-mouse Dnajc3 polyclonal antiserum (1∶500) and prohibitin with an anti-prohibitin polyclonal antibody (1∶500) followed by the appropriate anti-IgG HRP-labeled antibodies. Bottom: Revelation of amyloid precursor protein App with a rabbit antiserum (1∶2000) and of prohibitin as loading control, followed by the appropriate anti-IgG HRP-labeled antibodies (1∶5000). For Tlr-7 detection, gels were loaded with the amounts of protein indicated and revealed with anti-Tlr7 antibody (1∶75), followed by the appropriate anti-IgG HRP-labeled antibody (1∶250).

Figure 7. ER stress response evaluation.

Immunoblot of Grp78 and actin as loading control in whole cell lysates from cells growing in standard conditions (line 1, WT cells in control medium; line 2, CIP-R cells in medium added by 0.2 mM ciprofloxacin), or in WT cells exposed for 24 h to 0.2 mM ciprofloxacin (line 3), or in cells exposed for 24 h to 0.5 µg/ml tunicamycin (TM; line 4, WT cells; line 5, CIP-R cells). The anti-Grp78 is a rabbit antiserum used at 1∶500 dilution. The hatched square highlights the standard conditions of culture for each cell line.