Up-regulation of CD147 and matrix metalloproteinase-2, -9 induced by P-glycoprotein substrates in multidrug resistant breast cancer cells

Abstract

Treatment of animals bearing multidrug resistant (MDR) tumor cells with P-glycoprotein (P-gp) substrates could worsen host survival. It is assumed that this is due to increased tumor metastasis. To clarify the mechanism(s) underlying this observation, the MDR human breast cancer cell line, MCF-7/AdrR, and its sensitive parental line, MCF-7, was treated with various concentrations of P-gp substrate drugs (vincristine, paclitoxel, adriamycin) and a P-gp non-substrate drug (bleomycin) in serum-free media. Increased production of CD147, and matrix metalloproteinases (MMP)-2, -9 was observed only in MDR cancer cells exposed to P-gp substrates, as determined using real-time polymerase chain reaction, western blotting and zymography. Correspondingly, P-gp substrates significantly enhanced the in vitro invasion abilities of MCF-7/Adr cells. It was also found that the drug-induced promotion of CD147, and MMP-2, -9 was consistent with increased expression of epidermal growth factor receptor (EGFR) and that inhibition of either EGFR or P-gp activity could significantly interrupt the downstream effects, and so inhibit in vitro invasion abilities motivated by P-gp substrates. These results imply that treatment of MDR tumors with P-gp substrates could adversely affect therapeutic outcomes through modulating the production of CD147, MMP-2, -9, and EGFR, and suggest that this effect may be initiated by the transporter function of P-gp.

Figure 1

Production of CD147 and matrix metalloproteinase (MMP)‐2, ‐9 in MCF‐7 and MCF‐7/Adr cells when treated with various drugs. After cells were treated with drugs for 24 h, total RNA were extracted, and lysates were electrophoresed on 8% sodium dodecyl sulfate (SDS)‐polyacrylamide gel or on gelatin containing 8% SDS‐polyacrylamide gel. Endogenous CD147 and MMP‐2, ‐9 were measured using real‐time polymerase chain reaction, western blot and zymography, respectively. (a) CD147 and MMP‐2, ‐9 mRNA, protein and MMP‐2, ‐9 activities in MCF‐7, MCF‐7/Adr cells exposed to bleomycin (0.1 peak plasma concentration [PPC], 1.0 PPC, 10.0 PPC). (b) CD147 and MMP‐2, ‐9 mRNA, protein and MMP‐2, ‐9 activities in MCF‐7, MCF‐7/Adr cells exposed to vincristine (0.1 PPC, 1.0 PPC, 10.0 PPC). (c) CD147 and MMP‐2, ‐9 mRNA, protein and MMP‐2, ‐9 activities in MCF‐7, MCF‐7/Adr cells exposed to paclitaxel. (d) CD147 and MMP‐2, ‐9 mRNA, protein and MMP‐2, ‐9 activities in MCF‐7, MCF‐7/Adr cells exposed to adriamycin (0.1 PPC, 1.0 PPC, 10.0 PPC). ADM, adriamycin; BLM, bleomycin; Toxel, paclitoxel; VCR, vincristine. Bar graphs represent mean ± SEM of three independent experiments. Results are representative of three similar experiments. **P < 0.05 versus vehicle group.

Figure 1

Production of CD147 and matrix metalloproteinase (MMP)‐2, ‐9 in MCF‐7 and MCF‐7/Adr cells when treated with various drugs. After cells were treated with drugs for 24 h, total RNA were extracted, and lysates were electrophoresed on 8% sodium dodecyl sulfate (SDS)‐polyacrylamide gel or on gelatin containing 8% SDS‐polyacrylamide gel. Endogenous CD147 and MMP‐2, ‐9 were measured using real‐time polymerase chain reaction, western blot and zymography, respectively. (a) CD147 and MMP‐2, ‐9 mRNA, protein and MMP‐2, ‐9 activities in MCF‐7, MCF‐7/Adr cells exposed to bleomycin (0.1 peak plasma concentration [PPC], 1.0 PPC, 10.0 PPC). (b) CD147 and MMP‐2, ‐9 mRNA, protein and MMP‐2, ‐9 activities in MCF‐7, MCF‐7/Adr cells exposed to vincristine (0.1 PPC, 1.0 PPC, 10.0 PPC). (c) CD147 and MMP‐2, ‐9 mRNA, protein and MMP‐2, ‐9 activities in MCF‐7, MCF‐7/Adr cells exposed to paclitaxel. (d) CD147 and MMP‐2, ‐9 mRNA, protein and MMP‐2, ‐9 activities in MCF‐7, MCF‐7/Adr cells exposed to adriamycin (0.1 PPC, 1.0 PPC, 10.0 PPC). ADM, adriamycin; BLM, bleomycin; Toxel, paclitoxel; VCR, vincristine. Bar graphs represent mean ± SEM of three independent experiments. Results are representative of three similar experiments. **P < 0.05 versus vehicle group.

Figure 2

Effects of P‐glycoprotein (P‐gp) substrates, the P‐gp non‐substrate and inhibition of P‐gp function on Matrigel invasion of MCF‐7 and MCF‐7/Adr cells. Matrigel invasion was evaluated using trans‐well chambers. Bar graphs represent mean ± SEM of three independent experiments. *P < 0.05 versus vehicle group.

Figure 3

Epidermal growth factor receptor (EGFR) was involved in up‐regulation of CD147 and MMP‐2, ‐9 in MCF‐7/Adr cells treated with P‐glycoprotein (P‐gp) substrates. (a) Production of EGFR mRNA and protein in MCF‐7, MCF‐7/Adr cells treated with P‐gp substrates and a non‐P‐gp substrate. (b) Inhibition of EGFR activity interrupted the promotion of CD147 and MMP‐2, ‐9 induced by P‐gp substrates in MCF‐7/Adr cells. MCF‐7/Adr cells were starved for 24 h and preincubated with 250 nM AG1478, an EGFR specific inhibitor, for 40 min prior to treatment with various doses of P‐gp substrates (0.1 peak plasma concentration [PPC], 1.0 PPC, 10.0 PPC). Production of CD147 and MMP‐2, ‐9 was then determined using western blot and zymography. ADM, adriamycin; BLM, bleomycin; Toxel, paclitoxel; VCR, vincristine. Bar graphs represent mean ± SEM of three independent experiments. Results are representative of three similar experiments **P < 0.05 versus vehicle group.

Figure 4

Effects of P‐glycoprotein (P‐gp) neutralizing antibody on the production of epidermal growth factor receptor (EGFR), CD147 and MMP‐2, ‐9 in MCF‐7/Adr cells treated with P‐gp substrates. MCF‐7/Adr cells were exposed to various doses of P‐gp substrates (0.1 peak plasma concentration [PPC], 1.0 PPC, 10.0 PPC), respectively, with 20 µg/mL of the anti‐P‐gp monoclonal antibody for 3 h before adding P‐gp substrates. EGFR, CD147 and MMP‐2, ‐9 production was then assessed using western blot and zymography. ADM, adriamycin; Toxel, paclitoxel; VCR, vincristine. Results are representative of three similar experiments.