Cellular uptake of proMMP-2:TIMP-2 complexes by the endocytic receptor megalin/LRP-2

Abstract

Matrix metalloproteinases (MMPs) are regulated at multiple transcriptional and post-transcriptional levels, among which receptor-mediated endocytic clearance. We previously showed that low-density lipoprotein receptor-related protein-1 (LRP-1) mediates the clearance of a complex between the zymogen form of MMP-2 (proMMP-2) and tissue inhibitor of metalloproteinases, TIMP-2, in HT1080 human fibrosarcoma cells. Here we show that, in BN16 rat yolk sac cells, proMMP-2:TIMP-2 complex is endocytosed through a distinct LRP member, megalin/LRP-2. Addition of receptor-associated protein (RAP), a natural LRP antagonist, caused accumulation of endogenous proMMP-2 and TIMP-2 in conditioned media. Incubation with RAP also inhibited membrane binding and cellular uptake of exogenous iodinated proMMP-2:TIMP-2. Moreover, antibodies against megalin/LRP-2, but not against LRP-1, inhibited binding of proMMP-2:TIMP-2 to BN16 cell surface. BIAcore analysis confirmed direct interaction between the complex and megalin/LRP-2. Conditional renal invalidation of megalin/LRP-2 in mice resulted in accumulation of proMMP-2 and TIMP-2 in their urine, highlighting the physiological relevance of the binding. We conclude that megalin/LRP-2 can efficiently mediate cell-surface binding and endocytosis of proMMP-2:TIMP-2 complex. Therefore megalin/LRP-2 can be considered as a new actor in regulation of MMP-2 activity, an enzyme crucially involved in many pathological processes.

Figure 1

Expression of functional megalin/LRP-2 but not LRP-1 in BN16 cells. (a) Expression of megalin/LRP-2 and LRP-1 was evaluated by western blot analysis in BN16 cell extracts (50 μg protein). As positive control, receptor expression was determined in tissue extracts (50 μg protein) from rat liver. Images were cropped for presentation. Additional data without cropping are presented in Supplementary Fig. 1. (b) Analysis of media conditioned by BN16 cells in the presence of the LRP competitor, RAP. BN16 cells were cultured for 24 h in serum-free DMEM in the absence or presence of 1 μM RAP. Conditioned media were then collected, and total cell protein was measured using bicinchoninic acid microassay. Top panel, gelatin zymogram of medium conditioned by the equivalent of 5 μg of cell protein. Bottom panel, reverse gelatin zymogram of medium conditioned by the equivalent of 10 μg of cell protein. Representative results of three independent experiments. Reference (ref.) corresponds to medium conditioned by mouse calvarium, as previously reported. Values under the gels indicate the fold-increase by comparison with the first non-treated sample (*). Band area and intensity were measured by using IMAGEJ image analysis software.

Figure 2

Time-course of cell-surface binding of ¹²⁵I-proMMP-2:TIMP-2 complex. BN16 cells were incubated with 10 nM ¹²⁵I-proMMP-2:TIMP-2 complex at 4 °C for the indicated intervals to allow surface binding, in the absence (open circles) or presence of 1 μM RAP (closed circles), then washed and treated with pronase^® for detachment. Surface-bound ¹²⁵I-proMMP-2:TIMP-2 complex was defined as pronase^®-sensitive radioactivity. Values are means ± S.D. of three dishes. This experiment was performed twice with similar results.

Figure 3

Blocking of megalin/LRP-2 but not LRP-1 inhibits membrane binding of proMMP-2:TIMP-2 complex. BN16 cells were incubated with 10 nM ¹²⁵I-proMMP-2:TIMP-2 complex at 4 °C for 2 h to allow surface binding, in the absence or presence of 1 μM RAP, anti-megalin/LRP-2 IgG (100 μg/ml), anti-LRP-1 IgG (100 μg/ml) or non-immune IgG (100 μg/ml). After 2 h, surface-bound tracer was measured as in Fig. 2. Values are means ± S.D. of three dishes. **p < 0.001; NS, not significant, vs control, using Student’s t test.

Figure 4

Uptake and degradation of ¹²⁵I- proMMP-2:TIMP-2 complex by rat yolk sac BN16 cells. BN16 cells were incubated with 10 nM ¹²⁵I-proMMP-2:TIMP-2 complex at 4 °C for 2 h to allow surface binding, in the absence (open circles) or presence of 1 μM RAP (closed circles). Some of the cultures were pre-treated with 100 μM chloroquine for 1 h at 37 °C (open triangles). After washing, cells were further incubated with fresh medium pre-warmed at 37 °C, without or with RAP or chloroquine as above. At the indicated times, the amounts of ligand (a) remaining surface-bound (pronase^®-sensitive radioactivity), (b) internalized (pronase^®-resistant radioactivity) and (c) degraded (trichloroacetic acid-soluble radioactivity in conditioned medium) were measured. Values are means ± S.D. of three dishes. This experiment was performed twice with similar results.

Figure 5

Invalidation of megalin/LRP-2 induces accumulation of proMMP-2 and TIMP-2 in urines. Twenty-four-hour urines were collected from 7-month-old Meg ^lox/lox ;Wnt4-Cre ⁺ mice (cKO) or their wild-type littermates (WT) and samples were analyzed, without concentration, by direct (32 µl/lane; top panel) or reverse (16 µl/lane; bottom panel) zymography. Reference (ref.) corresponds to medium conditioned by mouse calvarium for direct zymography (as in Fig. 1) or 20 ng recombinant TIMP-2 for reverse zymography.