|
|
||||||||
|
FJ
EXPRESS SUMMARY ARTICLE The Full-length version of this article is also available, published online March 28, 2001 as doi:10.1096/fj.00-0626fje. |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||


* Department of Hematology and Oncology,
** Department of Cardiology and Pneumology, Georg-August-University, 37075 Göttingen, Germany;
Department of Internal Medicine III, University Hospital Grosshadern, Ludwig-Maximilians University, 81377 München, Germany; and
Department of Extracellular Matrix Signaling, Georg-Speyer-Haus Institute for Biomedical Research, Johann Wolfgang von Goethe University, 60596 Frankfurt, Germany
2Correspondence: Department of Hematology and Oncology, Georg-August-University, Robert-Koch Str. 40, 37075 Göttingen, Germany. E-mail: falves{at}gwdg.de
SPECIFIC AIMS
The tyrosine kinase receptor DDR1 is activated by various types of collagen and consists of three isoformsDDR1a, b, and cas a result of alternative splicing in the cytoplasmic region. To search for the presence of additional splice variants of DDR1 within the juxtamembrane region, we applied a polymerase chain reaction (PCR) -based approach in conjunction with primers localized within the transmembrane region (exon 9) and the tyrosine kinase domain (exon 13).
PRINCIPAL FINDINGS
1. Identification of two novel kinase-deficient isoforms
All seven different human colon carcinoma cell lines analyzed
express DDR1a as demonstrated by an amplification product of 513 bp.
All cell lines also revealed a major amplification product of 624 bp,
most likely corresponding to DDR1b mRNA, except for Colo 206F, where
only a faint band was detected. The detection of additional bands with
sizes of
750, 600, 420, and 270 bp suggested the presence of
further isoforms (Fig. 1
). The complete PCR product of the cell line Colo 206F was further
analyzed by cloning and sequencing that confirmed the presence of DDR1a
and DDR1b. Compared to DDR1a, DDR1b has an insertion of 37 amino acids
in the juxtamembrane region encoded by exon 11. Furthermore, two novel
sequences of shorter sizes were isolated, which we named DDR1d and
DDR1e. Both are produced by alternative splicing in the juxtamembrane
region using different acceptor sites. The DDR1d amplification product
was 268 bp long and was most abundant in the PCR of Colo 206F and SW
480 (Fig. 1)
. DDR1d not only lacks the 111 bp corresponding to exon 11,
but also 245 bp corresponding to exon 12. Whereas the deletion of exon
11 retains the reading frame, the deletion of exon 12 results in a
frame shift mutation. By joining exon 10 with exon 13, a new reading
frame is generated. Translation of this reading frame in exon 13
resulted in the amino acid sequence GAPV, followed by a stop codon
(Fig. 2
). The translation product of the DDR1d cDNA is predicted to be a
membrane-anchored receptor missing the kinase and carboxyl-terminal
tail region. The protein consists of 508 amino acids with a calculated
molecular mass of 56 kDa. The second novel sequence, named DDR1e, was
186 bp long but was not detected as major product of the PCR reaction
(Fig. 1)
. Sequence comparison showed that the first half of exon 10
(exon 10a, 82 bp) is missing in addition to exons 11 and 12 (Fig. 2)
.
By joining exon 9 with the second half of exon 10 (exon 10b, 84 bp), a
shift in the open reading frame is generated. Joining exon 10b with
exon 13 reverts the sequence back into the previous reading frame. The
second half of exon 10 in DDR1e translates to the new sequence
VLESHPRTRSPGLVGIRPTPLPVSPMAL, followed by the unaltered sequence
coded by exon 13. Upon careful examination of the 5' intron/exon
boundary sequence of exon 10, it was apparent there were two possible
splice acceptor sites. The use of the distal 5' acceptor site in the
sequence gcc cag GTC CTA (nucleotides 16151627) resulted in the
deletion of exon 10a and the production of a cDNA of 84 bp (exon 10b)
with disruption of the open reading frame (Fig. 2)
. The deletion of
exon 10a, exon 11, and exon 12 resulted in an altered sequence of 28
amino acids in the juxtamembrane region and loss of the ATP binding
site GEGQFG (amino acids 617623), producing a receptor molecule
lacking intrinsic tyrosine kinase activity with 767 amino acids and a
predicted molecular mass of 86 kDa.
|
|
2. RNA expression of DDR1 isoforms in human colon carcinoma cell
lines
Northern blot analysis using a cDNA fragment coding
for the extracellular domain as a probe, which is common in all
isoforms, revealed a 4.3 kb DDR1 mRNA signal in both cell lines used,
Colo 206F and SW 480 (data not shown). With a cDNA probe corresponding
to exon 12, no signal was seen in Colo 206F but was observed in SW 480.
This correlates with the mRNA content detected by PCR amplification,
where Colo 206F cells had high amounts of DDR1d but very little
product for DDR1b and reduced mRNA amplification of DDR1a (Fig. 1)
. In
contrast, SW 480 cells showed high amounts of DDR1a and DDR1b as well
as DDR1d, which are reflected by the mRNA expression signals seen with
the two different probes. The presence of mRNA for both isoforms was
confirmed by nucleic acid sequence-based amplification (NASBA)
analysis.
3. Expression of DDR1 isoforms in human colon carcinoma
cell lines
The existence of DDR1d and DDR1e in vivo is supported by
Western blot analysis (Fig. 3A
). A 68 kDa protein representing the glycosylated DDR1d
receptor was seen only when an antibody against the extracellular
domain was used (
DDR1 N), but not when the carboxyl-terminal
antibody was applied (
DDR1 C). A protein of the apparent molecular
mass of 95 kDa was seen when applying
DDR1 N and
DDR1 C, representing most likely DDR1e (see Fig. 3A
, B
, e.g.). Most of the colon carcinoma cell lines tested
coexpress the kinase-deficient isoform DDR1d, together with kinase
active DDR1a/b, with an apparent molecular mass of 125 kDa in
comparable amounts. Furthermore, we performed an immunoprecipitation
with a monoclonal antibody raised against the extracellular domain of
DDR1, followed by Western blot analysis. As shown in Fig. 3C
, no expression of DDR1a and DDR1b could be observed in
Colo 206F, whereas in Colo 678 and HCT 116, a prominent band of 125 kDa
was detected with both antibodies
DDR1 N and
DDR1 C. The
ß-subunit of the receptor could not be detected by the
DDR1 C,
because the cell lysates were precipitated with an amino-terminal
antibody. In all cell lines, most prominently in Colo 206F cells, a
band with an apparent molecular mass of 68 kDa was detected by
DDR1
N (Fig. 3C
, right panel) that represents the glycosylated
DDR1d receptor. In Colo 206F and HCT 116, a protein of
95 kDa
was detected after immunoblotting with both antibodies, most likely
corresponding to DDR1e. In Colo 678, a notable larger band with
98 kDa was seen with the
DDR1 N, but not with the
DDR1 C, indicating there is potentially an even higher
diversity of DDR1 isoforms in this cell line. The absence
of DDR1e in Colo 678 cells is supported by the fact that we
failed to detect DDR1e transcripts by NASBA (data not shown).
|
4. Evidence for a nondominant negative mechanism of DDR1d
signal modulation
Since Colo 206F cells showed a substantial amount of
DDR1d protein expression in comparison to the full-length
receptors DDR1a and DDR1b, we studied whether the kinase-deficient
isoform DDR1d functions as dominant negative receptor. We could
show by immunoprecipitation of DDR1b from cells stimulated with rat
type I collagen for 3 h that tyrosine phosphorylation of DDR1b
took place in response to collagen in cells expressing high amounts of
DDR1d (data not shown). To further understand the role of the DDR1d
isoform during kinase activation, a full-length DDR1d cDNA was
constructed by a PCR-based approach and transiently overexpressed in
human embryonic kidney fibroblasts 293 cells. Compared with
DDR1a-transfected cells, a prominent band of 68 kDa was detected in
DDR1d-transfected cells only with the
DDR1 N antibody (data not
shown). To study signaling, we examined tyrosine phosphorylation in
response to rat type I collagen. In cells overexpressing DDR1d, no
tyrosine phosphorylation was observed in response to collagen.
Cotransfection of DDR1a together with increasing amounts of DDR1d
receptor showed that receptor phosphorylation in response to rat
collagen was not efficiently blocked in the presence of the tyrosine
kinase-deficient DDR1d (data not shown).
CONCLUSIONS
Two novel isoforms of DDR1DDR1d and DDR1ewere identified from human colon carcinoma Colo 206F cells. Both new isoforms are predicted to be membrane anchored but kinase-deficient receptors. The alternative splicing event takes place in the juxtamembrane region, which contains sequence motives essential for the interaction with cellular substrates and regulatory proteins. Through a frameshift in exon 13, DDR1d displays the sequence motif GAPV at the carboxyl terminus. This motif can function as a potential binding site for the growing family of proteins containing one or several PDZ (PSD-95/DLG/ZO-1) domains. Based on their structure, receptors with mutated or deleted kinase domain have been proposed to act as suppressors of full-length, enzymatic active receptors by forming heterodimers and blocking signaling in a dominant negative manner. Two lines of evidence suggest that at least DDR1d behaves differently: 1) in Colo 206F, DDR1b is tyrosine phosphorylated upon collagen stimulation despite strong expression of DDR1d, and 2) in transfected 293 cells, a fourfold excess of DDR1d did not efficiently block DDR1 tyrosine phosphorylation in response to collagen. Therefore, our study argues against a role of DDR1d as dominant negative inhibitor of DDR1 signaling. If DDR1d (and presumably DDR1e as well) does not influence collagen-mediated DDR1 signaling, other roles of the truncated receptors might be postulated in analogy to the kinase-deficient TrkB and TrkC receptors such as a role in cell adhesion, having its own signaling abilities, or sequestering and presenting collagen as ligand to the DDR1 full-length receptor. The challenge for future work will be to define the mechanisms that regulate the expression pattern of the novel DDR1 isoformsfor example, during embryogenesis and tumor progressionand to analyze the function and signaling events of DDR1 isoforms by overexpression of different isoforms in different cellular environment.
FOOTNOTES
1 To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.00-0626fje ; to cite this article, use FASEB J. (March 28, 2001) 10.1096/fj.00-0626fje ![]()
This article has been cited by other articles:
![]() |
C. Franco, G. Hou, P. J. Ahmad, E. Y.K. Fu, L. Koh, W. F. Vogel, and M. P. Bendeck Discoidin Domain Receptor 1 (Ddr1) Deletion Decreases Atherosclerosis by Accelerating Matrix Accumulation and Reducing Inflammation in Low-Density Lipoprotein Receptor-Deficient Mice Circ. Res., May 23, 2008; 102(10): 1202 - 1211. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Abdulhussein, D. H. H. Koo, and W. F. Vogel Identification of Disulfide-linked Dimers of the Receptor Tyrosine Kinase DDR1 J. Biol. Chem., May 2, 2008; 283(18): 12026 - 12033. [Abstract] [Full Text] [PDF] |
||||
![]() |
C.-Z. Wang, H.-W. Su, Y.-C. Hsu, M.-R. Shen, and M.-J. Tang A Discoidin Domain Receptor 1/SHP-2 Signaling Complex Inhibits {alpha}2beta1-Integrin-mediated Signal Transducers and Activators of Transcription 1/3 Activation and Cell Migration Mol. Biol. Cell, June 1, 2006; 17(6): 2839 - 2852. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. Matsuyama, M. Watanabe, Y. Shirahama, H. Mitsuyama, I. Higashimoto, M. Osame, and K. Arimura Discoidin Domain Receptor 1 Contributes to the Survival of Lung Fibroblast in Idiopathic Pulmonary Fibrosis Am. J. Pathol., March 1, 2006; 168(3): 866 - 877. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. Matsuyama, H. Mitsuyama, M. Watanabe, Y. Shirahama, I. Higashimoto, M. Osame, and K. Arimura Involvement of Discoidin Domain Receptor 1 in the Deterioration of Pulmonary Sarcoidosis Am. J. Respir. Cell Mol. Biol., December 1, 2005; 33(6): 565 - 573. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. Matsuyama, M. Watanabe, Y. Shirahama, K.-i. Oonakahara, I. Higashimoto, T. Yoshimura, M. Osame, and K. Arimura Activation of Discoidin Domain Receptor 1 on CD14-Positive Bronchoalveolar Lavage Fluid Cells Induces Chemokine Production in Idiopathic Pulmonary Fibrosis J. Immunol., May 15, 2005; 174(10): 6490 - 6498. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. Matsuyama, L. Wang, W. L. Farrar, M. Faure, and T. Yoshimura Activation of Discoidin Domain Receptor 1 Isoform b with Collagen Up-Regulates Chemokine Production in Human Macrophages: Role of p38 Mitogen-Activated Protein Kinase and NF-{kappa}B J. Immunol., February 15, 2004; 172(4): 2332 - 2340. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. Matsuyama, M. Faure, and T. Yoshimura Activation of Discoidin Domain Receptor 1 Facilitates the Maturation of Human Monocyte-Derived Dendritic Cells Through the TNF Receptor Associated Factor 6/TGF-{beta}-Activated Protein Kinase 1 Binding Protein 1{beta}/p38{alpha} Mitogen-Activated Protein Kinase Signaling Cascade J. Immunol., October 1, 2003; 171(7): 3520 - 3532. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. A. Curat and W. F. Vogel Discoidin Domain Receptor 1 Controls Growth and Adhesion of Mesangial Cells J. Am. Soc. Nephrol., November 1, 2002; 13(11): 2648 - 2656. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. A. Curat, M. Eck, X. Dervillez, and W. F. Vogel Mapping of Epitopes in Discoidin Domain Receptor 1 Critical for Collagen Binding J. Biol. Chem., November 30, 2001; 276(49): 45952 - 45958. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |