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: 17/2/292 02-0430fjev1    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 GUTWEIN, P. Articles by ALTEVOGT, D. P. Search for Related Content PubMed PubMed Citation Articles by GUTWEIN, P. Articles by ALTEVOGT, D. P.

ADAM10-mediated cleavage of L1 adhesion molecule at the cell surface and in released membrane vesicles ¹

PAUL GUTWEIN², SABINE MECHTERSHEIMER², SVENJA RIEDLE², ALEXANDER STOECK, DANIELA GAST, SAFWAN JOUMAA, HANSWALTER ZENTGRAF^*, MINA FOGEL and D. PETER ALTEVOGT³

Tumor Immunology Programme, G0100 and
^* Applied Tumor Virology, F0900 German Cancer Research Center, D-69120 Heidelberg, Germany; and
Department of Pathology, Kaplan Hospital, Rehovot, Israel

³Correspondence: Tumor Immunology Programme, G0100, German Cancer Research Center, Im Neuenheimer Feld 280, D-69120 Heidelberg, Germany.

SPECIFIC AIMS

L1, an important molecule for cell migration of neural and tumor cells, is released by membrane-proximal cleavage and soluble L1 promotes cell migration. Since previous studies have implicated a role of ADAM10 in L1 release, we examined the localization of L1 and ADAM10 and the mode of cleavage.

PRINCIPAL FINDINGS

1. ADAM10 mediates cleavage of the L1 ectodomain
We reinvestigated the cleavage of L1 in HEK293 cells, which do not express L1 but are positive for ADAM10. Expression of human L1 in these cells allowed detection of soluble L1–200 in the medium. Overexpression of ADAM10 in combination with L1 enhanced the release of soluble L1. In contrast, HEK293 cells cotransfected with L1 and dominant-negative ADAM10 showed a reduction of L1 shedding as detected by a decrease in soluble L1–200 and L1–32, which represents the membrane retained cleavage fragment.

2. Subcellular localization of L1 and ADAM10
We studied in more detail the localization of L1 and ADAM10 using an antibody to the extracellular part. Confocal fluorescence microscopy showed that ADAM10 staining was predominant in the Golgi and colocalized with the Golgi marker GM-130. L1 was mostly present at the cell surface but ADAM10 and L1 colocalized in the Golgi.

3. L1 cleavage at the cell surface and in the Golgi/TGN
The major cellular residence of ADAM10 may not necessarily be the site of enzymatic activity. We investigated the cellular localization of L1 cleavage using a well-defined sucrose density fractionation method. Consistent with the immunofluorescence data, ADAM10 was predominantly expressed in Golgi/TGN-enriched fractions where most of the mature ADAM10 was present. Mature ADAM10 was detected by cell surface iodination at the cell surface and in the ER/PM-enriched fractions in the bottom of the gradient. The L1–32 cleavage fragment colocalized with mature ADAM10 in the Golgi/TGN but was also detectable in ER/PM-enriched fractions. These biochemical results supported the notion that L1 ectodomain cleavage occurs in the Golgi as well as at the plasma membrane.

4. PMA and MCD treatment augment L1 cleavage by distinct mechanisms
Cholesterol efflux by MCD treatment caused L1 release in a time- and dose-dependent manner (Fig. 1 A). Compared to PMA, the MCD-induced L1 release was less efficiently blocked by Ro 31–9790. We concluded that supernatants from MCD-treated cells contained L1 in a nonsoluble form. A refined analysis allowed the detection of L1–220 in addition to L1–200 (Fig. 1B ). Similar observations were made in supernatants from pervanadate-treated cells. Sphingomyelinase treatment, just like PMA, resulted in the generation of only soluble L1–200.

View larger version (24K): [in this window] [in a new window]   Figure 1. MCD treatment induces the release of vesicles and enhances L1 cleavage. A) AR cells were treated with MCD in serum-free medium. Supernatants were analyzed for the presence of L1 using mAb UJ127.11 and ECL detection. B) AR cells were treated with sphingomyelinase (SM), PMA, MCD, or pervanadate. Supernatants were TCA precipitated and analyzed with mAb UJ 127.11, followed by ECL detection. C) AR cells were treated with MCD or PMA in the absence or presence of Ro inhibitor. Cell-free supernatants were centrifuged at 100000 g. Pellets were dissolved in sample buffer and supernatants were TCA precipitated. All samples were analyzed for the presence of L1 using pcyt L1 and ECL detection. D) The same blot was probed with mAb UJ127.11 to the ectodomain of L1. E) The stripped blot was reprobed with anti ADAM10.

5. MCD induces the release of membrane vesicles containing L1 and ADAM10
MCD treatment can cause membrane vesicles formation. Indeed, MCD treatment strongly induced the release of vesicles that appeared in the 100000 g pellet and contained full-length L1–220 as it was detectable with pcytL1 (Fig. 1C , lane 3). Vesicle release was also observed in cells kept in suspension at 37°C (lane 1). In contrast, PMA treatment caused little vesicle formation (compare lanes 2 and 5). MCD-induced vesicles contained the cleavage fragment L1–32 (lane 3). In the presence of Ro 31–9790, L1–32 disappeared and only full-length L1 was detectable (lane 4). The 100000 g supernatant showed only signals with the mAb against the ectodomain of L1 (Fig. 1D , lanes 7–12). This suggested that L1 was without a cytoplasmic tail and was truly soluble. The 68 kDa form of ADAM10 was present in MCD-induced vesicles but was undetectable in PMA-treated samples (Fig. 1E ).

6. Cleavage of L1 in membrane vesicles
The presence of L1 in MCD induced vesicles was confirmed by immunoelectron microscopy. Cleavage of L1 could proceed in isolated vesicles as evidenced by an fivefold increase in the amount of the L1–32 in a time-dependent fashion. This was paralleled by a similar increase of soluble L1–200. Treatment of cells with the HMG-CoA reductase inhibitor lovastatin induced the formation of vesicles positive for L1–32 and the active form of ADAM10. Concomitantly, the 100000 g supernatant contained increased amounts of soluble L1–200.

7. L1 in microvesicles can trigger cell migration
We reported before that soluble L1 can trigger haptotactic cell migration of CHO cells on various substrates. To demonstrate that L1 in vesicles could fulfill a similar biological function, CHO cells were exposed to L1-positive vesicles obtained from AR or OVM tumor cells. Intact vesicles or vesicle supernatants were able to induce haptotactic migration in an L1-dependent fashion, suggesting that vesicles were functionally active.

8. The RSLE exon is involved in vesicle recruitment of L1
Biochemical analysis indicated that MCD treatment of AR cells augmented the release of vesicles 0.1.8- to 3-fold over spontaneous release. We noted that HEK293 cells showed a higher level of spontaneous vesicle formation. Spontaneous and MCD-induced vesicles were positive for ADAM10.

After transfection of HEK293 cells with L1, cell lysate and released vesicles became positive for L1–220 and the cleavage product L1–32. This suggested that ADAM10 had constitutive access to vesicles and that L1 was recruited when expressed by the cells. L1–85 was present in the cell lysate but was missing in released vesicles. L1–85 cleavage is mediated by plasmin at the cell surface, suggesting that vesicles were not derived from the cell surface (see below).

L1 occurs in different isoforms generated by alternative splicing. In neural L1, exon 27 is expressed resulting in the insertion of four amino acids (RSLE) in the cytoplasmic part. The RSLE motif facilitates interaction with the µ2 chain of the clathrin adapter AP-2. We analyzed whether L1 and L1_RSLE had access to vesicles. When transfected, both forms of L1–220 were detected at similar levels in the lysate. However, in secreted vesicles the level of L1–220 was much different and L1–32 was absent in vesicles from L1_RSLE-transfected cells. This was not due to a different amount of vesicles as ADAM10 was equally detected.

Similar results were obtained in CHO cells stably expressing L1 or L1_RSLE. The amount of soluble L1–200 in the conditioned medium after removal of vesicles was higher in cells expressing the full-length L1-form vs. the L1_RSLE form. This supported the notion that cleavage in vesicles could generate soluble L1, but was not the only pathway for its production.

9. MCD-vesicles are not derived from the plasma membrane
We finally investigated the origin of MCD released vesicles. Cell surface biotinylation experiments indicated that the bulk of soluble L1 released by MCD was not derived from the cell surface, as it carried no biotin label. In contrast, soluble L1 released by PMA was strongly biotinylated, suggesting its origin directly from the cell surface.

Given the colocalization of L1 and ADAM10 in the Golgi, we analyzed Golgi morphology after treatment. Staining for GM-130 was unchanged in PMA-treated cells, but in pervanadate and MCD-treated cells the Golgi morphology appeared partially dispersed in vesicular structures. We investigated released vesicles for the presence of coated vesicle markers using Western blot analysis. Spontaneously released vesicles from HEK293 were negative for - and ß-adaptin but positive for clathrin heavy chain and -adaptin. The latter marker proteins are involved in clathrin-coated vesicle transport from the Golgi to endosome.

CONCLUSIONS

We present evidence for two pathways of L1 cleavage depending on ADAM10 localization (Fig. 2 ): 1) release of soluble L1–200 directly from the cell surface caused by treatment of cells with PMA or sphingomyelinase; 2) cleavage of L1 in membrane vesicles containing ADAM10 and L1. The release of such vesicles can be enhanced by MCD or pervanadate treatment. The vesicle cleavage pathway is operative in cells expressing full-length L1, such as neural cells and certain tumor cells.

View larger version (17K): [in this window] [in a new window]   Figure 2. Model for L1 cleavage at the cell surface or in released membrane vesicles.

We observed enhanced L1 release into the medium after cholesterol efflux by treatment of cells with MCD. Careful analysis revealed the presence of L1 in vesicles that contained the active form of ADAM10. MCD released L1 was not derived from the cell surface. Cleavage of L1 could proceed in isolated vesicles. These findings demonstrated that ADAM10 is not only a membrane-localized sheddase but can act as a vesicle-based protease.

The most likely explanation of our results is that ADAM10- and L1-containing vesicles were derived from the Golgi consistent with the biochemical evidence that L1 cleavage takes place in the Golgi. It has been shown that MCD treatment and changes in the cholesterol level affect Golgi morphology and can cause a partial vesiculation of Golgi structures. Our results on pervanadate and MCD-treated tumor cells using GM-130 and ADAM10 confirm the published data. Moreover, released vesicles contained the clathrin heavy chain and the adaptor protein -adaptin, which play a role in the vesicular transport from the Golgi.

Is the difference between L1 cleavage at the cell surface or in vesicles only a mechanistic one? The important role of tumor-derived vesicles is increasingly recognized. Tumor cells frequently release vesicles containing proteolytic activity like matrix-degrading MMPs, urokinase plasminogen activator, ß1 integrins, and an array of other proteins. It is believed that the consequence of shedding from cancer cells includes effects on the ability of the cells to infiltrate, metastasize and alter their microenvironment. We demonstrate here that L1 containing vesicles can trigger haptotactic cell migration and that only the neural form of L1 is recruited and cleaved in vesicles. Human AR tumor cells are of non-neural origin, yet could release and cleave L1 in vesicles. We found RSLE in tumor cell lines and fresh tumor specimens of ovarian cancers that frequently overexpress L1. The concentration of proteolytic and migration promoting activity in released vesicles could promote the ability of tumor cells to overcome the barrier of basal membranes.

The finding that ADAM10 is a vesicle-based protease has important implications beyond L1 cleavage. ADAM10 was proposed to be involved in the development of arthritis. ADAM10 protein primarily in the membrane-bound form was present in human articular cartilage and was up-regulated in osteoarthritis. It is possible that the ADAM10-mediated cleavage of APP involves a similar vesicle mechanism as described here for L1.

FOOTNOTES

¹ To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.02-0430fje; to cite this article, use FASEB J. (December 3, 2002) 10.1096/fj.02-0430fje

² These authors contributed equally to this publication.

This article has been cited by other articles:

				D. Finas, M. Huszar, A. Agic, S. Dogan, H. Kiefel, S. Riedle, D. Gast, R. Marcovich, F. Noack, P. Altevogt, et al. L1 cell adhesion molecule (L1CAM) as a pathogenetic factor in endometriosis Hum. Reprod., May 1, 2008; 23(5): 1053 - 1062. [Abstract] [Full Text] [PDF]

				N. Gavert, M. Sheffer, S. Raveh, S. Spaderna, M. Shtutman, T. Brabletz, F. Barany, P. Paty, D. Notterman, E. Domany, et al. Expression of L1-CAM and ADAM10 in Human Colon Cancer Cells Induces Metastasis Cancer Res., August 15, 2007; 67(16): 7703 - 7712. [Abstract] [Full Text] [PDF]

				E. Marcello, F. Gardoni, D. Mauceri, S. Romorini, A. Jeromin, R. Epis, B. Borroni, F. Cattabeni, C. Sala, A. Padovani, et al. Synapse-Associated Protein-97 Mediates {alpha}-Secretase ADAM10 Trafficking and Promotes Its Activity J. Neurosci., February 14, 2007; 27(7): 1682 - 1691. [Abstract] [Full Text] [PDF]

				M. L. Rodrigues, L. Nimrichter, D. L. Oliveira, S. Frases, K. Miranda, O. Zaragoza, M. Alvarez, A. Nakouzi, M. Feldmesser, and A. Casadevall Vesicular Polysaccharide Export in Cryptococcus neoformans Is a Eukaryotic Solution to the Problem of Fungal Trans-Cell Wall Transport Eukaryot. Cell, January 1, 2007; 6(1): 48 - 59. [Abstract] [Full Text] [PDF]

				K. Dassler, M. Zydek, K. Wandzik, M. Kaup, and H. Fuchs Release of the Soluble Transferrin Receptor Is Directly Regulated by Binding of Its Ligand Ferritransferrin J. Biol. Chem., February 10, 2006; 281(6): 3297 - 3304. [Abstract] [Full Text] [PDF]

				M. J.E. Arlt, I. Novak-Hofer, D. Gast, V. Gschwend, G. Moldenhauer, J. Grunberg, M. Honer, P. A. Schubiger, P. Altevogt, and A. Kruger Efficient Inhibition of Intra-Peritoneal Tumor Growth and Dissemination of Human Ovarian Carcinoma Cells in Nude Mice by Anti-L1-Cell Adhesion Molecule Monoclonal Antibody Treatment Cancer Res., January 15, 2006; 66(2): 936 - 943. [Abstract] [Full Text] [PDF]

				T. Maretzky, M. Schulte, A. Ludwig, S. Rose-John, C. Blobel, D. Hartmann, P. Altevogt, P. Saftig, and K. Reiss L1 Is Sequentially Processed by Two Differently Activated Metalloproteases and Presenilin/{gamma}-Secretase and Regulates Neural Cell Adhesion, Cell Migration, and Neurite Outgrowth Mol. Cell. Biol., October 15, 2005; 25(20): 9040 - 9053. [Abstract] [Full Text] [PDF]

				F. Bianco, E. Pravettoni, A. Colombo, U. Schenk, T. Moller, M. Matteoli, and C. Verderio Astrocyte-Derived ATP Induces Vesicle Shedding and IL-1{beta} Release from Microglia J. Immunol., June 1, 2005; 174(11): 7268 - 7277. [Abstract] [Full Text] [PDF]

				P. Gutwein, A. Stoeck, S. Riedle, D. Gast, S. Runz, T. P. Condon, A. Marme, M.-C. Phong, O. Linderkamp, A. Skorokhod, et al. Cleavage of L1 in Exosomes and Apoptotic Membrane Vesicles Released from Ovarian Carcinoma Cells Clin. Cancer Res., April 1, 2005; 11(7): 2492 - 2501. [Abstract] [Full Text] [PDF]

				N. Gavert, M. Conacci-Sorrell, D. Gast, A. Schneider, P. Altevogt, T. Brabletz, and A. Ben-Ze'ev L1, a novel target of {beta}-catenin signaling, transforms cells and is expressed at the invasive front of colon cancers J. Cell Biol., February 14, 2005; 168(4): 633 - 642. [Abstract] [Full Text] [PDF]

				K. R. Santhamma, R. Sadhukhan, M. Kinter, S. Chattopadhyay, B. McCue, and I. Sen Role of Tyrosine Phosphorylation in the Regulation of Cleavage Secretion of Angiotensin-converting Enzyme J. Biol. Chem., September 17, 2004; 279(38): 40227 - 40236. [Abstract] [Full Text] [PDF]

				A. R. Walmsley, G. McCombie, U. Neumann, D. Marcellin, R. Hillenbrand, A. K. Mir, and S. Frentzel Zinc metalloproteinase-mediated cleavage of the human Nogo-66 receptor J. Cell Sci., September 1, 2004; 117(19): 4591 - 4602. [Abstract] [Full Text] [PDF]

				B. von Tresckow, K.-J. Kallen, E. P. von Strandmann, P. Borchmann, H. Lange, A. Engert, and H. P. Hansen Depletion of Cellular Cholesterol and Lipid Rafts Increases Shedding of CD30 J. Immunol., April 1, 2004; 172(7): 4324 - 4331. [Abstract] [Full Text] [PDF]

				P. J. Gough, K. J. Garton, P. T. Wille, M. Rychlewski, P. J. Dempsey, and E. W. Raines A Disintegrin and Metalloproteinase 10-Mediated Cleavage and Shedding Regulates the Cell Surface Expression of CXC Chemokine Ligand 16 J. Immunol., March 15, 2004; 172(6): 3678 - 3685. [Abstract] [Full Text] [PDF]

				D. R. McCulloch, P. Akl, H. Samaratunga, A. C. Herington, and D. M. Odorico Expression of the Disintegrin Metalloprotease, ADAM-10, in Prostate Cancer and Its Regulation by Dihydrotestosterone, Insulin-Like Growth Factor I, and Epidermal Growth Factor in the Prostate Cancer Cell Model LNCaP Clin. Cancer Res., January 1, 2004; 10(1): 314 - 323. [Abstract] [Full Text] [PDF]

This Article Full Text (PDF) All Versions of this Article: 17/2/292 02-0430fjev1    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 GUTWEIN, P. Articles by ALTEVOGT, D. P. Search for Related Content PubMed PubMed Citation Articles by GUTWEIN, P. Articles by ALTEVOGT, D. P.