Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2015 Feb 15;26(4):659-73.
doi: 10.1091/mbc.E14-10-1463. Epub 2014 Dec 24.

ADAM10 Controls Collagen Signaling and Cell Migration on Collagen by Shedding the Ectodomain of Discoidin Domain Receptor 1 (DDR1)

Affiliations
Free PMC article

ADAM10 Controls Collagen Signaling and Cell Migration on Collagen by Shedding the Ectodomain of Discoidin Domain Receptor 1 (DDR1)

Yasuyuki Shitomi et al. Mol Biol Cell. .
Free PMC article

Abstract

Discoidin domain receptor 1 (DDR1) is a receptor tyrosine kinase that binds and transmits signals from various collagens in epithelial cells. However, how DDR1-dependent signaling is regulated has not been understood. Here we report that collagen binding induces ADAM10-dependent ectodomain shedding of DDR1. DDR1 shedding is not a result of an activation of its signaling pathway, since DDR1 mutants defective in signaling were shed in an efficient manner. DDR1 and ADAM10 were found to be in a complex on the cell surface, but shedding did not occur unless collagen bound to DDR1. Using a shedding-resistant DDR1 mutant, we found that ADAM10-dependent DDR1 shedding regulates the half-life of collagen-induced phosphorylation of the receptor. Our data also revealed that ADAM10 plays an important role in regulating DDR1-mediated cell adhesion to achieve efficient cell migration on collagen matrices.

Figures

FIGURE 1:
FIGURE 1:
Collagen-induced DDR1 ectodomain shedding. (A) HEK293 cells were transiently transfected with empty vector (Mock) or N-terminally FLAG-tagged DDR1 (DDR1-NF) and treated with 100 μg/ml collagen I for 24 h. Conditioned media and cell lysates were analyzed by Western blotting using anti–DDR1 ectodomain (Med), anti–DDR1 C-terminus (Cell), and anti-actin (actin) antibodies. CTF, C-terminal fragment. (B) A431 and MCF-7 cells were incubated for 24 h in the presence or absence of collagen I. Conditioned media and cell lysates were analyzed as in A. Arrows point to detected protein bands for shed DDR1, full-length DDR1, CTF, and actin.
FIGURE 2:
FIGURE 2:
The sheddase responsible for collagen-induced DDR1 shedding is TIMP-insensitive. (A) HEK293 cells were transiently transfected with DDR1-NF and treated with collagen I for 24 h in the presence or absence of 10 μM GM6001, 10 μM Mst, 10 μM E-64, 10 μM pepstatin A, 100 μM AEBSF, PI cocktail, 1 μM γ-secretase inhibitor X (L-685, 458), or vehicle control (DMSO). Conditioned media and cell lysates were subjected to Western blotting using anti–DDR1 ectodomain (Med), anti–DDR1 C-terminus (Cell), or anti-actin antibodies. CTF, C-terminal fragment. (B, C) HEK293 cells transiently transfected with N-terminally FLAG-tagged DDR1 (B) and A431 (C) cells were treated for 24 h with serum-free DMEM containing 100 μg/ml collagen I in the presence or absence of TIMP-3 at a concentration of 100, 200, or 300 nM. Conditioned media and cell lysates were analyzed by Western blotting using anti–DDR1 ectodomain (Med), anti–DDR1 C-terminus (Cell), anti-FLAG (TIMP-3, Med), or anti-actin antibodies. The band intensity of shed DDR1 was analyzed with Phoretix and standardized by the intensity of shed DDR1 in collagen-treated medium. Data shown at the top represent the means ± SEM. N = 3 for each condition. ***p < 0.005; ****p < 0.0001; NS, not significant (p > 0.05; one-way analysis of variance [ANOVA]) compared with collagen-treated medium.
FIGURE 3:
FIGURE 3:
ADAM10 is the sheddase responsible for collagen-induced DDR1 ectodomain shedding. (A) A431 cells were transfected with siRNA for ADAM8, ADAM9, ADAM10, ADAM17, or ADAM19 or with nontargeting (NT) siRNA. After 48 h, cells were treated with collagen I (100 μg/ml) for a further 24 h. We confirmed that the level of mRNA for each enzyme was reduced by 60−99% after 72 h (right, RT-PCR). Mst (10 μM) was used as a positive control for inhibition of DDR1 shedding. Conditioned media and cell lysates were analyzed by Western blotting using anti–DDR1 ectodomain (Med), anti–DDR1 C-terminus (Cell), or anti-actin antibodies. Note that ADAM10 knockdown resulted in inhibition of endogenous DDR1 shedding. CTF, C-terminal fragment. (B) A431 cells were transfected with siRNA for MT1-MMP (si-MT1) or si-NT. Cells were then treated with collagen I for 24 h. Conditioned media and cell lysates were analyzed by Western blotting using anti-DDR1, anti-MMP14 (EP1264Y) (MT1), and anti-actin antibodies. (C) A431 cells were treated with 1 μM IM, 25 ng/ml PMA, or DMSO vehicle control (0.001%) for 1 h. Conditioned media and cell lysates were analyzed as in A. A431 cells were also treated with 1 μM IM for 1 h in the presence or absence of 10 μM Mst (right). (D) A431 cells were transiently transfected with ADAM10 or mock vector and treated with or without collagen I for 24 h. Conditioned media and cell lysates were subjected to Western blotting using anti–DDR1 ectodomain (Med), anti–DDR1 C-terminus (Cell), anti-ADAM10 (A10), and anti-actin antibodies.
FIGURE 4:
FIGURE 4:
Coimmunoprecipitation of DDR1 with ADAM10. (A) HEK293 cells were cotransfected with DDR1-NF, wild-type ADAM10, or ADAM10ΔMP in combination as indicated and subjected to immunoprecipitation with anti-FLAG affinity beads. Bound materials were analyzed by Western blotting using anti–DDR1 ectodomain or anti-ADAM10 antibodies. Asterisks indicate the protein that was pulled down. MP, metalloproteinase domain. (B) HEK293 cells were transiently cotransfected with FLAG-tagged ADAM10 (ADAM10-F), ADAM10ΔMP-F, or DDR1 in combination as indicated. Cell lysates were immunoprecipitated with anti-FLAG antibody, followed by Western blotting with anti-DDR1 ectodomain or anti-ADAM10 antibodies. DDR1 was coimmunoprecipitated with ADAM10-F or ADAM10ΔMP-F (FLAG-immunoprecipitation, top). Asterisks indicate the protein that was pulled down. (C) Coimmunoprecipitation of endogenous DDR1 and ADAM10. A431 cells were subjected to cell surface biotinylation before cell lysis. Cell lysates were subjected to two-step affinity precipitation using antixDDR1 ectodomain or anti-ADAM10 ectodomain conjugated to protein G–coated Dynabeads followed by streptavidin beads. Coimmunoprecipitated DDR1 and ADAM10 bound to streptavidin beads were visualized by Western blotting using anti-DDR1 C-terminus or anti-ADAM10 antibodies. Control sample was incubated with protein G–Dynabeads without antibodies. Active, active form of ADAM10; Pro, proform. Blank lanes that were cropped out of the blot are indicated by black lines.
FIGURE 5:
FIGURE 5:
DDR1 is in a complex with ADAM10 on the cell surface. (A) HEK293 cells stably expressing DDR1-NF were subjected to PLA using anti–DDR1 ectodomain and anti-ADAM10 ectodomain antibodies. The PLA signal is shown in red. Nuclei were stained with DAPI (blue). Scale bars, 200 and 50 μm. (B) A431 cells were stimulated with collagen I for 24 h in the presence or absence of 10 μM Mst and subjected to PLA as in A. Scale bar, 200 μm.
FIGURE 6:
FIGURE 6:
Collagen binding, but not DDR1 phosphorylation, is required for DDR1 shedding. (A) Schematic representation of mutant DDR1 constructs used in the experiments. DD, discoidin-homology domain; ∆C, cytoplasmic domain-deleted; DLD, discoidin-like domain; FLAG, FLAG tag (DYKDDDDK); HA, HA tag (YPYDVPDYA); JM, juxtamembrane region; KD, kinase dead; S, signal peptide; TM, transmembrane domain; TKD, tyrosine kinase domain. (B) HEK293 cells expressing DDR1-WT (WT), DDR1-KD (KD), or DDR1∆C (∆C) were treated with collagen I for 24 h. Conditioned media and cell lysates were analyzed by Western blotting using anti-DDR1 ectodomain (Med and Cell), anti–phosphotyrosine 4G10 (PY), or anti-actin. Asterisk indicates tyrosine-phosphorylated proteins other than DDR1. (C) HEK293 cells expressing DDR1-WT were treated with collagen I in the presence or absence of 50 nM dasatinib or vehicle control (DMSO) for 24 h. Conditioned media and cell lysates were analyzed by Western blotting using anti–DDR1 ectodomain (Med), antixDDR1 C-terminus (Cell), anti-PY, or anti-actin. (D) HEK293 cells were transiently transfected as indicated and treated with collagen for 24 h. Conditioned media and cell lysates were subjected to Western blotting using anti-DDR1 ectodomain and anti-actin antibodies. For the phosphotyrosine blot, cells treated with collagen for 1 h were used. The relative intensities of shed DDR1 are shown at the bottom of the top panel. (E) HEK293 cells were transfected with N-terminally FLAG-tagged DDR1 mutants as indicated and treated with collagen for 24 h. Conditioned media and cell lysates were analyzed by Western blotting as in D.
FIGURE 7:
FIGURE 7:
Engineering shedding-resistant DDR1 mutants. (A) Schematic representation of mutant DDR1 constructs used in the experiments. Arrow indicates the cleavage site identified. DD, discoidin-homology domain; DLD, discoidin-like domain; JM, juxtamembrane region; S, signal peptide; TKD, tyrosine kinase domain; TM, transmembrane domain. (B) HEK293 cells were transfected with expression plasmids for DDR1 mutants as indicated. Cells were treated with collagen for 24 h (DDR1 and Actin) or 1 h (PY). Conditioned media and cell lysates were subjected to Western blotting using anti-DDR1 (DDR1), anti-actin, and anti-PY antibodies. The band intensity of shed DDR1 was analyzed with Phoretix software, and fold changes were normalized to actin and DDR1 in cell lysates and sample volumes. The normalized shedding was calculated as described in Materials and Methods. Data are shown at the right and represent the means (%) ± SEM. n = 3 for each condition. *p < 0.01; **p < 0.0001; NS, not significant (p > 0.05; two-tailed Student's t test) compared with WT, collagen treated. Statistical analyses were performed with Prism, version 6 (GraphPad, La Jolla, CA). (C) Cleavage sites of DDR1 mutants were determined by N-terminal sequencing. Arrows indicate the identified cleavage sites. (D) HEK293 cells were transfected as indicated. Cells were treated with collagen for 24 h and analyzed as in B. The relative intensities of DDR1 shed forms are shown in the bottom of the top panel. (E) HEK293 cells stably expressing DDR1 mutants as indicated were transfected with siRNAs for ADAM10 (si-A10) or with nontargeting siRNA (si-NT). After 48 h, cells were treated with collagen I (100 μg/ml) for a further 24 h. Conditioned media and cell lysates were analyzed by Western blotting with antixDDR1 ectodomain (DDR1), anti-ADAM10 (A10), or anti-actin antibodies. The band intensity of shed DDR1 mutants was analyzed with Phoretix, and relative intensities were standardized to WT or ∆C treated with collagen. Data are shown at the bottom of the top panel. Active, active form of ADAM10; Pro, proform of ADAM10.
FIGURE 8:
FIGURE 8:
Reduced DDR1 shedding is associated with its sustained phosphorylation. (A) HEK293 stably expressing FLAG-tagged wild type (WT) or -6xD mutant DDR1 (6xD) were treated with collagen (20 μg/ml) for 1 h, excess collagen was then washed out, and cells were incubated further as indicated. RIPA lysates were collected at the indicated times and immunoprecipitated with anti-FLAG beads, followed by Western blotting with anti-DDR1 (Cell) and anti-PY antibodies. Shed DDR1 at each time point was analyzed by anti-DDR1 antibody (Med). Representative data from two independent experiments. (B) Relative intensities of phosphorylated DDR1 bands from A (PY). The intensities were standardized by the intensity of phosphorylated DDR1 in 1-h collagen-treated sample for each construct. (C) Relative band intensities of shed DDR1 in A (DDR1, Med). Relative intensities were normalized to the band intensity of shed DDR1-6xD at 0-h incubation time.
FIGURE 9:
FIGURE 9:
DDR1 ectodomain shedding is required for efficient cell migration on a collagen matrix. (A) A431 cells transfected with siRNAs for DDR1 (si-D1), ADAM10 (si-A10), or both DDR1 and ADAM10 (si-A10+si-D1) were subjected to wound-closure assay on a collagen matrix. Control cells were also treated with 50 μM Mst to inhibit DDR1 shedding. si-NT, nontargeting siRNA. Scale bar, 300 μm. Wound-healing edges of cells were traced and indicated with dashed lines. (B) Relative cell migration of each treatment in A. The data are presented as mean ± SEM (n = 6). **p < 0.01; ***p < 0.005 (one-way ANOVA) compared with each si-NT. p < 0.05 is considered statistically significant. Statistical analyses were performed with Prism, version 6 (GraphPad). (C) Knockdown levels of each protein were confirmed by Western blotting with anti-DDR1, anti-ADAM10, and anti-actin antibodies. Pro, proform of ADAMs; Active, active form of ADAMs. (D) A431 cells stably expressing DDR1-wild type (WT), -6xD, ΔC, and ΔC-6xD were subjected to wound-closure assay on a collagen matrix. Representative images from each treatment. Wound-healing edges of cells were traced and are indicated with dashed lines. (E) Data from D were analyzed as in B (n = 16). *p < 0.05; **p < 0.01; ****p < 0.0001 (two-tailed Student's t test). (F) A431 cells stably expressing DDR1 mutants were subjected to Western blotting to confirm the expression levels of each mutant. All constructs contain an N-terminal HA tag and a C-terminal FLAG tag. Asterisk indicates nonspecific band.

Similar articles

See all similar articles

Cited by 13 articles

See all "Cited by" articles

References

    1. Abdulhussein R, McFadden C, Fuentes-Prior P, Vogel WF. Exploring the collagen-binding site of the DDR1 tyrosine kinase receptor. J Biol Chem. 2004;279:31462–31470. - PubMed
    1. Ahonen M, Poukkula M, Baker AH, Kashiwagi M, Nagase H, Eriksson JE, Kahari VM. Tissue inhibitor of metalloproteinases-3 induces apoptosis in melanoma cells by stabilization of death receptors. Oncogene. 2003;22:2121–2134. - PubMed
    1. Amour A, Knight CG, English WR, Webster A, Slocombe PM, Knauper V, Docherty AJ, Becherer JD, Blobel CP, Murphy G. The enzymatic activity of ADAM8 and ADAM9 is not regulated by TIMPs. FEBS Lett. 2002;524:154–158. - PubMed
    1. Amour A, Knight CG, Webster A, Slocombe PM, Stephens PE, Knauper V, Docherty AJ, Murphy G. The in vitro activity of ADAM-10 is inhibited by TIMP-1 and TIMP-3. FEBS Lett. 2000;473:275–279. - PubMed
    1. Barker KT, Martindale JE, Mitchell PJ, Kamalati T, Page MJ, Phippard DJ, Dale TC, Gusterson BA, Crompton MR. Expression patterns of the novel receptor-like tyrosine kinase, DDR, in human breast tumours. Oncogene. 1995;10:569–575. - PubMed

Publication types

MeSH terms

LinkOut - more resources

Feedback