HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Full Text (PDF) All Versions of this Article: 16/2/234 01-0414fjev1    most recent Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by L’HÔTE, C. G. M. Articles by GANESAN, T. S. Search for Related Content PubMed PubMed Citation Articles by L’HÔTE, C. G. M. Articles by GANESAN, T. S.

Functional analysis of discoidin domain receptor 1: effect of adhesion on DDR1 phosphorylation ¹

ICRF Molecular Oncology Laboratories, Institute of Molecular Medicine, John Radcliffe Hospital, Headington, Oxford OX3 9DS, UK

²Correspondence: ICRF Molecular Oncology Laboratories, Institute of Molecular Medicine, John Radcliffe Hospital, Headington, Oxford OX3 9DS, UK. E-mail: ganesan{at}icrf.icnet.uk

SPECIFIC AIMS

Collagen, an essential constituent of the extracellular matrix, has been reported to activate discoidin domain receptors (DDR1 and DDR2). This raises the possibility that signaling by these receptors may be influenced by cell adhesion. The main aim of our study was to analyze the effect of cell adhesion on DDR1 activation by collagen in cancer cell lines and to investigate the signaling pathways.

PRINCIPAL FINDINGS

1. Variability of phosphorylation of DDR1 by collagen
We confirm the slow phosphorylation of DDR1 in three human epithelial breast cancer cell lines (T47D, ZR75, ZR11) and in one human epithelial colorectal cancer cell line (HCT-116) by soluble rat collagen I. However, collagen did not stimulate phosphorylation in all the cell lines: the ovarian cell line OAW42 or the breast cell lines MCF7, MD468, and SkBr3. This was not due to low levels of DDR1 expression in these cell lines or a mutation in the discoidin domain (excluded by sequencing).

2. Faster phosphorylation of DDR1 in suspension conditions compared with adherent status
To study the relation between DDR1 activation by collagen and adhesion we detached cells from the plate and stimulated them in suspension with collagen. Surprisingly, DDR1 phosphorylation was rapid (Fig. 1 A, B) and specifically induced by collagen I (Fig. 1C ). Replating cells on dishes coated with an attachment inhibitor (poly-HEMA) led to a similar result: DDR1 phosphorylation was detected after 20 min of stimulation (Fig. 1D ). We further analyzed the effect of adhesion, with a semiadherent cell line (Colo201) and a cell line growing in suspension (K562). DDR1 phosphorylation was not detected in adherent Colo201 cells, but was after 15 min of stimulation in the cells that were in suspension. Moreover, in K562 cells, DDR1 phosphorylation was rapid and peaked at 30 min (Fig. 2 ). Plating K562 cells on fibronectin- or polylysine-coated plates induced a delay in the activation of DDR1 by collagen.

View larger version (64K): [in this window] [in a new window]   Figure 1. Fast phosphorylation of DDR1 in detached cells. Serum-starved cells were washed three times with PBS, incubated with PBS and EDTA (3 mM), detached by pipetting, washed once in PBS, resuspended in DMEM with HEPES and acetic acid (0.05%) in tubes (A–C) or on HEMA-coated plates (D), then stimulated with or without rat collagen I (10 µg/ml). DDR1 was immunoprecipitated (with 6.4 or C.T.) and its phosphorylation detected on blot using an anti-phosphotyrosine antibody (4G10 or pY100). The blots were stripped and reprobed with an anti DDR1 antibody. A) The requirement of magnesium is shown by stimulating T47D cells with rat collagen I for 30 min in PBS with or without magnesium. A, C, D) T47D cells; D) 0i = no collagen and incubation time of 10 min; 0f = no collagen and incubation time of 330 min; B) HCT116 cells.

View larger version (58K): [in this window] [in a new window]   Figure 2. Fast phosphorylation of DDR1 in K562 cells. K562 cells were starved for 3 h, then resuspended in DMEM with HEPES and acetic acid (0.05%) and stimulated with human collagen V (Hu Coll V) or rat collagen I (Rat Coll I). T47D cells in adherent condition were also stimulated with rat collagen I on the same day as a control. DDR1 was immunoprecipitated using C.T. or 6.4 antibody. Tyrosine phosphorylation was detected with 4G10 antibody. The blot was stripped and reprobed with C.T. antibody.

To understand the mechanistic basis of the difference in the phosphorylation of DDR1 between adherent and suspension status of the cells, we have investigated two possibilities: the involvement of a phosphatase that would be active in adherent cells and the role of cell–cell contact rather than cell matrix interaction in this inhibition.

3. Role of phosphatases
Incubation of T47D cells with increasing concentration of pervanadate, a tyrosine phosphatase inhibitor, resulted in strong DDR1 phosphorylation even in the absence of ligand. However, there was no significant increase in DDR1 phosphorylation in the presence of collagen. It is possible that a phosphatase may inhibit DDR1 in the native state.

4. DDR1 slow phosphorylation kinetics is not dependent on cell–cell interactions
EDTA was used to disrupt cell–cell contacts without detaching the cells from the plates but no major effect was observed. DDR1 phosphorylation kinetics was comparable to those observed with untreated adherent T47D cells when simulated with collagen I.

5. Signaling: involvement of PI-3 kinase but absence of Erk activation in adherent or detached cells
MAP kinase pathway is commonly activated by receptor tyrosine kinases upon stimulation with cognate ligand. However, there was no activation of Erk1 or Erk2 detected in either adherent or detached T47D cells when stimulated with collagen I. To evaluate whether DDR1 signaled through the PI-3 kinase pathway, we performed coimmunoprecipitation experiments. Coimmunoprecipitation of p85 with DDR1 was detected in collagen-stimulated cells, but no correlation with DDR1 phosphorylation was observed. However, in the presence of pervanadate, an increase of p85 coimmunoprecipitation with DDR1 was observed that correlated with the increase in DDR1 phosphorylation. There was no difference in the levels of p85 that associated with DDR1 in the adherent or suspension conditions in T47D cells. Evaluation of the downstream kinase Akt1 upon stimulation with collagen did not show an appreciable increase in phosphorylation of Akt1 over time, although there was an increase from basal levels. As an alternative approach to evaluate the signaling properties of the catalytic domain of DDR1, we generated a chimeric TrkA/DDR1 receptor that was stably expressed in NIH3T3 cells. The chimeric receptor was expressed at the cell surface and dimerized upon nerve growth factor stimulation (NGF). Upon stimulation with NGF, there was no detectable phosphorylation of the chimeric receptor. Further, the chimeric receptor in an immune complex kinase assay could not phosphorylate the synthetic substrate poly (Glu⁸⁰:Tyr²⁰).

CONCLUSIONS

We have shown that in T47D and HCT116 cells there was rapid phosphorylation of DDR1 in cells placed in suspension compared to adherent status. This was an unexpected result, as this type of differential signaling has not been observed for other receptor tyrosine kinases. We were able to confirm this finding by studying other cell lines: a semiadherent cell line (colo201) and a cell line growing in suspension (K562). DDR1 phosphorylation is rapid in COLO201 or K562 cells in suspension, but we could not detect any DDR1 activation in adherent COLO201 upon collagen stimulation.

The possible mechanisms responsible for the effect of adhesion on DDR1 activation are discussed in this study and summarized in Fig. 3 . First, DDR1 may be involved in adhesion itself or bound to a matrix protein, and therefore less accessible to collagen in adherent cells. Argos is such an example, an inhibitory protein for the Drosophila EGF receptor. Second, the involvement of a phosphatase as suggested by the phosphorylation of DDR1 in the presence of pervanadate is also a possible negative regulatory mechanism. Third, there may be naturally occurring splice isoforms of the receptor (such as herstatin binding to c-erbB2) that act in a dominant negative manner in vivo inhibiting phosphorylation. Although some kinase-deficient isoforms of the DDR1 receptor have been identified, these do not inhibit phosphorylation of the receptor in vivo. Further research is required to identify the exact mechanism for the differential signaling by DDR1 upon stimulation by collagen.

View larger version (26K): [in this window] [in a new window]   Figure 3. Schematic diagram. In the adherent status, three mechanisms are proposed to explain the delayed DDR1 phosphorylation kinetics: 1) involvement of DDR1 in cell adhesion or bound to a specific inhibitory protein possibly in the extracellular matrix; 2) inhibition by a phosphatase dependent on cell adhesion; 3) inhibition by a splice variant of DDR1. In suspension, the disruption of cell matrix interaction rather than cell–cell contact releases the ‘inhibitory protein’, allowing DDR1 to dimerize more rapidly and become phosphorylated. DD: discoidin domain; KD: kinase domain.

Upon collagen stimulation of DDR1, our results suggest that the p85 subunit of PI-3 kinase binds to DDR1. This probably occurs through the SH2 domain of p85 binding to the ‘YELM’ motif present in the carboxyl-terminal tail of DDR1. There is probably further complexity in the regulation of DDR1 activation, as the chimeric receptor approach was unsuccessful. The relationship between DDR1 signaling and cellular context suggests cross-talk with adhesion molecules. This has been reported for the insulin receptor, EGF receptor, and c-kit being activated through interactions with integrin in the adherent state. The detachment of adherent cells over time results in a type of cell death called ‘anoikis’, and its suppression seems to be a hallmark of malignant transformation and metastases. It is tempting to speculate that the differential signaling in the case of DDR1 and its increased expression in cancer cells provide a link with metastases.

FOOTNOTES

¹ To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.01-0414fje; to cite this article, use FASEB J. (December 28, 2001) 10.1096/fj.01-0414fje

This article has been cited by other articles:

				B. Leitinger Molecular Analysis of Collagen Binding by the Human Discoidin Domain Receptors, DDR1 and DDR2. IDENTIFICATION OF COLLAGEN BINDING SITES IN DDR2 J. Biol. Chem., May 2, 2003; 278(19): 16761 - 16769. [Abstract] [Full Text] [PDF]

This Article Full Text (PDF) All Versions of this Article: 16/2/234 01-0414fjev1    most recent Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by L’HÔTE, C. G. M. Articles by GANESAN, T. S. Search for Related Content PubMed PubMed Citation Articles by L’HÔTE, C. G. M. Articles by GANESAN, T. S.