Fig 2: SIM images reveal DDR1-immunostain double-walled structures for cells stimulated with collagen for 60 minutes at 37 °C. COS-7 cells transiently expressing DDR1 were stimulated with collagen I for 10 or 60 minutes at 37 °C or left unstimulated, then incubated on ice with mAb 7A9, before fixation and secondary Ab staining. 3D-SIM images were acquired using a Zeiss ELRYA microscope; images are from one of 15 slices. Lower panels show magnified views of the boxed areas in the upper panels. Arrowheads in A indicate example structures wider than 200 nm (full width at half maximum). Scale bars, 30 µm (upper panels) or 5 µm (magnified images). At least 10 cells were imaged for each condition.

Fig 3: Schematic diagram of wild-type and signalling-defective DDR1 mutants. The extracellular region consists of two globular domains, the N-terminal discoidin (DS) domain and the discoidin-like (DS-like) domain, followed by a highly flexible juxtamembrane (JM) region. The transmembrane (TM) region contains a dimerisation motif. The intracellular catalytic kinase domain is preceded by a large unstructured JM region. The collagen-binding trench in the DS domain is shown in red. Collagen binding to this site in wild-type DDR1 induces phosphorylation of cytoplasmic tyrosine residues in both the JM region and kinase domain (shown as yellow circles). None of the mutants are phosphorylated upon collagen incubation. DDR1-W53A has a mutation in the ligand binding pocket in the DS domain. DDR1-R32E and DDR1-L152E are signalling defective mutants with mutations in the ‘signal patch’ region in the base of the DS domain, near the DS-like domain. DDR1-TM1 is a mutant with impaired transmembrane helix association, and DDR1-K655A is a mutant with impaired catalytic function. The locations of mutations are indicated by red stars, and anti-DDR1 epitopes located in the DS-like domain are symbolised by blue and green ovals for wild-type DDR1.

Fig 4: Collagen I induces DDR1 redistribution on the cell surface. COS-7 cells transiently expressing DDR1 were stimulated with collagen as detailed below. (A) Cells were stimulated for the indicated times (in minutes) at 37 °C, then incubated on ice with mAb 7A9 against the DDR1 ectodomain, before fixation and secondary Ab staining. (B) Cells were stimulated for the indicated times at 37 °C. Staining was done as above. The graph shows mean ZC scores + SEM (N = 200–400 regions from 80–100 cells from 3 independent experiments). *p < 0.05; ****p < 0.0001 (one-way ANOVA, followed by Bonferroni post hoc test). Data are from a different set of experiments than those shown in panel A. (C,D) Cells were stimulated for the indicated times at 37 °C, then fixed and permeabilised, and immunostained for phospho-tyrosine 513 (pY-DDR1) and for DDR1. (C) Mean pY-DDR1 levels across three experiments: mean values were normalised within experiments and then mean values taken for each stimulation time. N is at least 30 for each stimulation time. (D) The proportion of cells expressing DDR1 with pY-DDR1 signal above background levels were manually counted for different stimulation times. Mean percentage values ± SEM (N = 55 for all stimulation times, from two independent experiments). (E) Cells were either stimulated with collagen I for 10 minutes at 37 °C or left unstimulated, then incubated on ice with mAb 7A9 against the DDR1 ectodomain (shown in green) and anti-collagen-I mAb (shown in magenta), before fixation, and secondary Ab staining. White boxes in left columns indicate corresponding areas shown at higher magnification in columns to the right. All cells were imaged using a widefield microscope. At least 30 cells were imaged for each condition. Scale bars, 30 µm or 10 µm (enlarged images).

Fig 5: Anti-DDR1 mAbs block phosphorylation of collagen-bound DDR1. The diagram at the top gives an overview of the experimental procedures. HEK293 cells transiently expressing wide-type DDR1 were first incubated with collagen I for 60 minutes on ice, in the presence (+) or absence (−) of the indicated anti-DDR1 mAbs (Phase 1). Following washes, cells were incubated for 30 minutes at 37 °C, in the absence (no mAb) or presence of the indicated mAbs (Phase 2). The sample shown in lane 1 was lysed immediately after the incubation with collagen on ice. Samples labelled 2 were replicate lysates from 3 different wells. Cell lysates were analysed by Western blot using a mAb against phosphorylated Tyr-513 (anti-pY). The blot was stripped and re-probed with rabbit anti-DDR1. The positions of molecular mass markers are indicated on the left in kDa. The bar chart shows the densitometry analysis of pY513 band intensities after normalization to total DDR1. Each value is a percentage of the sum of all the pY513/DDR1 signals on the blot. The graph shows mean band intensities + SEM (N = 3). ****p < 0.0001 (one-way ANOVA, followed by Dunnett's multiple comparison test).