Fig 2: MRIP enables DDR1 cluster formation and growth. (A) Cells were cultured on fibrillar collagen films or on fibronectin for 8 h. An antibody against the extracellular N-terminal region of DDR1 (AF2396, R&D) was used to stain non-permeabilized preparations (top, red) followed by phalloidin to stain F-actin (bottom, green). Quantification of total cluster area, (B), number, (C), and single cluster area, (D), of DDR1 adhesions on collagen or on fibronectin using Fiji. Scale bar, 20 µm. (E,F) GD25 WT or WTMRIP-/- cells were transfected with mouse DDR1b-YFP (full-length kinase-active isoform), or mouse DDR1bK653A-YFP and cultured on fibrillar collagen or on fibronectin as indicated. (G–I) Quantification of total cluster area, number, and single cluster area of DDR1b-YFP or DDR1bK653A-YFP adhesions on collagen or on fibronectin as indicated using Fiji. Data are reported as mean ± SD, n = 3. * p < 0.05, ** p < 0.005, *** p < 0.0005, **** p <0.0001. Scale bar, 20 µm.

Fig 3: MRIP enables the feedforward loop that links DDR1 clustering and activation. (A) GD25 WT or WTMRIP-/- were transfected either with DDR1b-YFP or DDR1bk653A-YFP and cultured on fibrillar collagen or fibronectin as indicated. For fluorescence recovery after photobleaching measurements, regions of interest (ROIs: 1 μm2 area indicated with arrow) in (C) containing DDR1 clusters were bleached with argon laser at 488 nm. Fluorescence in ROIs was measured before bleaching and for 150 s after bleaching. (B) Typical normalized FRAP curves for each condition are displayed. (C) Images for each condition before bleaching (i: i-pre in B), after bleach (ii: ii-bleach in B), at intermediate recovery point (iii: iii-mid in B), and at the end of record fluorescent recovery (iv: iv-end in B). The FRAP-half-life recovery, (D), and FRAP mobile fractions, (E), for each condition were then calculated from the normalized curves. (F) DDR1 clustering was detected by co-transfecting GD25 WT or WTMRIP-/- cells with DDR1b-CFP (donor) and DDR1b-YFP (acceptor) probe as FRET pair. To access the contribution of DDR1 activation for cluster growth we co-transfected WT cells with the DDR1-CFP and DDR1K653A-YFP (YFPK653A) FRET probe. The cells were co-transfected with the indicated FRET pairs and cultured on collagen or fibronectin. (G) FRET efficiency was calculated from donor and acceptor pre- and post-bleach images (arrow indicates bleach ROI) with the acceptor-photobleaching-FRET mode in Leica SP8 confocal microscope. (H) Plot represents the mean FRET efficiency for interactions of indicated FRET pairs in cells in the indicated conditions. Data are reported as mean ± SD, n = 3. ++, ** p < 0.005, *** p < 0.0005, **** p < 0.0001.

Fig 4: MRIP ablation enhances DDR1-NMIIA association but disrupts the myosin contraction cycle. (A) Collagen bead-bound fractions were prepared from OE and OEMRIP-/- cells, which were immunoblotted for DDR1, NMIIA, and vimentin. Relative amounts of DDR1 and NMIIA in collagen- bound, (B) and unbound (C) fractions were quantified by densitometry of immunoblots, n = 3. (D), Whole-cell lysates of cells cultured on fibrillar collagen-coated tissue culture plates were immunoblotted for phospho-MLCSer19 (#3675, CS), MLC #3672S, CS), and ß-actin. (E) Quantification of pMLCSer19/MLC by densitometry of immunoblots, n = 3. Cells were cultured on collagen for 3 h and co-immunostained for DDR1 (C-6: sc-374618) and pMLCSer19 (#3671S, CS), (F), or for pMLC Ser19 and F-actin, (G–I). Pearson coefficients were obtained by quantification of the fluorescent images using the colocalization2 plug-in in Fiji. n = 3, scale bar, 10 µm. (J) Cells were plated on collagen for 3 h, active GTP-RhoA was accessed by pull-down assay and measured by densitometry of GTP-RhoA/Total-RhoA on immunoblots, n = 3. (K) Temporal regulation of MRIP dependent myosin contractility. MRIP binds MLCP and RhoA and targets them to myosin enabling dephosphorylation of the MLC and consequently myosin contraction cycle. (L) DDR1 (C-6: sc-374618) or NMIIA (BT-564) was co-immunoprecipitated from cell lysates of cells cultured on fibrillar collagen. Immunoprecipitates were then immunoblotted for NMIIA or DDR1 respectively. (M) DDR1 (C-6: sc-374618) was immunoprecipitated from cell lysates of cells cultured on fibrillar collagen for indicated time points and immunoblotted for NMIIA. The immunoprecipitates were immunoblotted for NMIIA and DDR1, respectively. Quantification of NMIIA bound to DDR1 by densitometry of immunoblots, n = 2. Data are reported as mean ± SD for H and I or as mean ± SEM, n = 3. * p < 0.05, **** p < 0.0001.

Fig 5: MRIP deletion inhibits cell migration on collagen. (A) Immunoblots for evaluation of indicated protein expression levels in whole cell lysates of NIH 3T3 WT, GD25- WT, OE, OEMRIP-/-, and WTMRIP-/- cells. The plot on the right represents the ratio of indicated proteins to ß-actin levels for the in vitro cell model. (B) Wound-healing assay of confluent monolayers of WT, OE, OEMRIP-/-, and WTMRIP-/- cells cultured on collagen and record at 0, 4, 8, and 24 h post-wounding (B, 0 and 24 h) and quantified in the plot on the right. (C) By Western blot we found an approximately 25% restoration of MRIP expression in OEMRIP-/- cells stably transfected with mouse MRIP variant 1 (OEMRIP-/-+MRIP). The indicated cell lysates were also immunoblotted for DDR1 and for ß-actin. (D) Wound-healing assay of confluent monolayers of OE, OEMRIP-/-, and OEMRIP-/-+MRIP cells cultured on collagen. Wound closure was recorded every hour for 24 h after wounding. Quantification is shown in the plot on the right. Data are reported as mean ± SD, n = 3, * p < 0.05, ++ p < 0.005, +++ p < 0.0005, **** p<0.0001. Scale bar, 10 µm.