Although schwannomin, the product of the neurofibromatosis type 2 gene, shares homology with three cytoskeleton-to-membrane protein linkers defining the ERM family, the mechanism by which it exerts a tumor suppressive activity remains elusive. Based on the knowledge of naturally occurring mutations, a functional study of schwannomin was initiated. Constructs encoding the two wild-type isoforms and nine mutant forms were transfected into HeLa cells. Transiently expressed wild-type isoforms were both observed underneath the plasma membrane. At this location they were detergent insoluble and redistributed by a cytochalasin D treatment, suggesting interaction with actin-based cytoskeletal structures. Proteins with single amino acid substitutions at positions 219 and 220 demonstrated identical properties. Three different truncated schwannomins, that are prototypic for most naturally occurring NF2 mutations, were affected neither in their location nor in their cytochalasin D sensitivity. However, they were revealed to be detergent soluble, indicating a relaxed interaction with the actin-based structures. An increased solubility was also observed for a mutant with a single amino acid substitution at position 360 in the C-terminal half of the protein. Mutant proteins with either a single amino acid deletion at position 118 or an 83 amino acid deletion within the N-terminal domain had lost the submembraneous localization and tended to accumulate in perinuclear patches that were unaffected by cytochalasin D treatment. A similar behavior was observed when the N-terminal domain was entirely deleted. Taken together these observations suggest that the N-terminal domain is the main determinant that localizes the protein at the membrane where it interacts weakly with actin-based cytoskeletal structures. The C-terminal domain potentiates this interaction. With rare exceptions, most naturally occurring mutant schwannomins that have lost their tumor suppressive activity are impaired in an interaction involving actin-based structures and are no longer firmly maintained at the membrane.
Neurofibromatosis type 2 (NF2) is an autosomal dominantly inherited disease which predisposes to the development of nervous system tumors, mainly bilateral vestibular schwannomas and to a lesser extent schwannomas affecting spinal nerve roots, meningiomas and ependymomas (1). The NF2 gene which when mutated causes this predisposition was identified in 1993 (2,3). Both alleles of the NF2 gene were observed mutated in tumors of NF2 patients (2-9) but also in sporadically occurring schwannomas and meningiomas (2,3,10,11), strongly suggesting that the NF2 gene has a tumor suppressive activity. In accord with this hypothesis, transfection of wild-type NF2 cDNA was shown to decrease the growth rate of NIH3T3 cells (12), and to revert their ras-induced malignant phenotype (13).
The NF2 gene product, designated either `schwannomin' (2) or `merlin' (3) is a 595 amino acid protein that exhibits homologies with ezrin, radixin and moesin, a group of three closely related proteins collectively termed the ERM family (14). This family is part of the band 4.1 superfamily, which also includes band 4.1 protein, talin and several tyrosine phosphatases (15). Some members of this superfamily have been shown to act as linkers between the plasma membrane and cytoskeletal structures. More specifically, ERM proteins reportedly colocalize with the cell surface proteins CD43, CD44, ICAM-2 (16-18) and possibly ICAM-1 (18). In addition, Hirao et al. (19) demonstrated that the N-terminal region of the ERM proteins interacts with CD44 andthat Rho may be involved in the regulation of this interaction. Furthermore, the 34 most C-terminal amino acids of ezrin and moesin, which are highly conserved among the three ERM proteins, have been shown to contain an actin-binding site (20,21).
The amino acid sequence of schwannomin reveals structural similarities with the ERM proteins: the N-terminal domain, that spans amino acids 1-314, shares 61-63% identity with the equivalent domain of each of the three ERM members. This domain has no obvious secondary structure. The C-terminal half of schwannomin (from amino acid 315 to 595) and those of the three ERM proteins are predicted to adopt mostly an alpha helical conformation. In the case of schwannomin, ezrin and radixin, this [alpha]-helix is interrupted by a proline rich sequence. Schwannomin has recently been reported to localize underneath the plasma membrane and to co-distribute with elements of the cytoskeleton in smooth muscle cells (22) and in Schwann cells (23). In fibroblasts and meningioma cells it has been shown to concentrate preferentially in membrane ruffles (24). Similar locations have been observed for members of the ERM family (14). The inhibition of NF2 gene expression in cultured cells by an antisense oligonucleotide strategy causes a decrease in cell adhesion that resembles that induced by the concomitant inhibition of the synthesis of all three ERM proteins (25). Taken together, these observations would suggest that schwannomin may share with the ERM family a role in the association of the cytoskeleton with the plasma membrane.
Schwannomin appears, however, significantly different from the ERM proteins. The latter proteins share 85% identity in their N-terminal domains, while the corresponding domain of schwannomin appears more distantly related. In contrast to the ERM proteins, schwannomin has two major isoforms which differ at their C-terminal end (26). Neither C-terminal ends of these isoforms demonstrate homology with the highly conserved 34 amino acid actin-binding domain identified at the C-terminus of the ERM proteins. At present, no direct experimental evidence supporting an interaction of schwannomin with cytoskeletal elements or characterized membrane components has yet been provided.
The NF2 gene, in contrast to the ERM-related genes, undergoes frequent biallelic inactivation in human tumor types, specifically in schwannomas, meningiomas and mesotheliomas (2,3,10,11,27, 28). Extensive mutational analyses have been performed on NF2 patients and on sporadic tumors. The majority of the mutations of the NF2 gene consists of nonsense, frameshift or splice-site mutations that result in the truncation of the gene product (2,3,5-11). However, a small number of mutations such as missense mutations, in frame interstitial deletions in exons or splice-site mutations causing exon skipping without frameshift, have been identified. We have taken advantage of the knowledge of such mutations to initiate the delineation of domains of interaction for schwannomin by examining the localization and detergent solubility of transiently expressed normal and mutant forms of schwannomin in intact HeLa cells or in HeLa cells with altered cytoskeleton.
In order to transiently express either isoforms of schwannomin, the corresponding cDNAs were placed under the control of the cytomegalovirus promoter of the pCB6 eukaryotic expression vector (Fig. 1). In both cases, two different constructs were generated that led either to the natural C-terminal end (Sch-Iso1 and Sch-Iso2) or to the linking of an epitope derived from the vesicular stomatitis virus glycoprotein G (VSV-G) (Sch-Iso1-Tag and Sch-Iso2-Tag). This procedure was followed in order to enable the detection of the plasmid-encoded proteins in transfected cells and their discrimination from endogenous proteins including ERM members and schwannomin.
A possible interaction of schwannomin with cytoskeletal elements was investigated by performing cell fractionation assays as described by Algrain et al. (30). It is established that this detergent extraction procedure preserves the cytoskeleton and maintains tightly linked proteins in the insoluble fraction. The detergent insolubility of such proteins can be visualized by immunofluorescence labeling after extraction or by immunoblot analysis of equal amounts of the extracted and the non-extracted fractions. HeLa cells were transfected with Sch-Iso1 or Sch-Iso1-Tag and detergent extracted. Immunostaining revealed neither a change in the distribution nor a major decrease in intracellular concentration of schwannomin isoform 1, suggesting that this protein stably interacts with cytoskeletal components. This hypothesis was confirmed when cell fractionation indicated that schwannomin was absent from the soluble fraction (Fig. 5). Double fluorescence labeling showed that schwannomin was not associated with actin stress fibers after extraction. The same conclusion was reached for schwannomin isoform 2 (Fig. 5A). In contrast, the truncated proteins encoded by Sch-T1, Sch-T2 and Sch-T3 were in part detergent extractable (Fig. 5B). Detergent extraction was also performed on cells transfected with constructs encoding missense mutant proteins. Fluorescence labeling with the anti-tag antibody revealed that the protein mutated at amino acid 360 was also detergent extractable. This observation was confirmed by immunoblotting (Fig. 5A). Immunofluorescence staining of cells expressing proteins Sch-[Delta]118, Sch-[Delta](39-121) or Sch-[Delta]Nter after detergent extraction showed that the staining pattern and intensity remained unaltered. Immunoblotting confirmed that these three mutant proteins were detergent insoluble (Fig. 5A).
To further characterize the interaction of schwannomin with cytoskeletal elements, cells overproducing schwannomin isoform 1 were treated with cytochalasin D, an agent that specifically disorganizes actin cytoskeletal structures. Double fluorescence labeling showed that full-length schwannomin co-distributed with the induced patches of actin (Fig. 6). Confocal microcopy analysis of the treated cells confirmed that schwannomin was relocalized in the periphery of the actin foci. Disruption of actin filaments by cytochalasin D led to a similar redistribution of the C-terminally truncated proteins encoded by constructs Sch-T1, Sch-T2 and Sch-T3 and of all missense mutant proteins. In contrast, cytochalasin D treatment of cells expressing Sch-[Delta]118, Sch-[Delta](39-121) or Sch-[Delta]Nter did not affect the perinuclear distribution of these mutant proteins.
Double staining of isoform 1 and of either tubulin, vimentin or keratins failed to reveal colocalization of schwannomin with microtubules or intermediate filaments. Similarly, Sch-[Delta]118 or Sch-[Delta]Nter were not found to colocalize with these cytoskeletal components. Treatment of cells with nocodazole, a microtubule-disrupting agent, did not modify the cellular distribution of schwannomin isoform 1, Sch-[Delta]118 or Sch-[Delta]Nter (data not shown).
ERM proteins have been shown to interact with the actin cytoskeleton (20,21). Owing to its homology with the ERM proteins, the colocalization of Schwannomin with cortical actin in several cell types was not unexpected (22-24). In order to document further the putative association of schwannomin with cytoskeletal components, transfected cells were extracted with Triton X-100, under conditions that preserve the cytoskeletal structure. Schwannomin was not extracted by this biochemical fractionation procedure. With the possible exception of smooth muscle cells (22), schwannomin was not found associated with actin stress fibers (23,24, this work). In contrast, exposure of the transiently transfected cells to cytochalasin D, which dissociates actin filaments, caused a marked change in the localization of schwannomin. Taken together these observations suggest that, in most cell types, schwannomin should interact with cortical F-actin either directly or indirectly. Interestingly, it has been shown that the 34 C-terminal amino acids of ezrin and moesin contain an F-actin-binding site (20,21). This region is highly conserved in all three ERM proteins but is markedly different in those observed in either isoforms of schwannomin. Thus even though schwannomin may share with the ERM proteins a similar localization, it may not have the same modes of interaction with the cytoskeleton.
Various mutant schwannomins were investigated for their interaction with the cytoskeleton. In contrast to the C-terminal domain of ezrin which colocalizes with F-actin (30), Sch-[Delta]Nter forms, at the periphery of the nucleus, non-vesicular aggregated structures which involve neither the actin nor the microtubule networks. This observation suggests that by itself the C-terminal domain of schwannomin does not form a stable interaction with the cytoskeleton. These aggregates may result from the formation of single or multicomponent structures or from local precipitation. As shown by the extraction experiments, these aggregates are stable even in the presence of non-ionic detergent. At least two subregions of this C-terminal domain may nonetheless contribute to the interaction of the intact protein with cytoskeletal components. Indeed, the removal of the last 70 amino acids or a single amino acid substitution at position 360 had the same phenotypic consequence: the mutant proteins remained in their normal submembranous location but exhibited a markedly increased detergent solubility. As shown by the cytochalasin D induced relocalization, these proteins maintained their interaction with the actin dependent structures. Such an interaction is not dependent on the presence of the polyproline rich region since identical properties were observed with a truncated protein that lacked this region. It may be suggested from these results that while the C-terminal domain of schwannomin is not sufficient to establish an interaction with cytoskeletal components by itself, it may contribute to stabilize such interaction in the intact schwannomin.
In vitro, the N-terminal domain of ERM proteins binds to the cytoplasmic region of CD44 (19). Although to date this interaction has not been reported for schwannomin, the homology of the N-terminal domain of schwannomin with that of the ERM proteins suggests that this domain may also be involved in binding to membrane protein(s). In accord with this hypothesis, mutations resulting in the deletion of a part or of the entire N-terminal domain of schwannomin were observed to cause a relocalization of the protein from the membrane to the periphery of the nucleus. More strikingly, a single deletion of one of two adjacent phenylalanines at position 118-119 had a similar effect, pointing to a critical amino acid doublet for the submembraneous localization. Interestingly, this doublet is conserved in mouse and Drosophila schwannomin and in the three ERM proteins.
Most mutations studied here are similar or identical to mutations that had occurred at the germline level in NF2 patients or at the somatic level in NF2 related tumors. They are therefore likely to impair the tumor suppressor activity of the encoded protein. With two exceptions, all mutant schwannomin studied here were affected in their ability to establish a stable interaction with actin based cytoskeletal structures. This was notable for the various C-terminally deleted proteins studied which may be taken as models for most naturally occurring mutant proteins. It is anticipated that decrease in the stability of this interaction would impair tumor suppressor activity. If this prediction were confirmed, this observation would provide the basis of a functional test to document the defective nature of new alleles observed in NF2 patients or in tumors.
However, demonstration by this assay of an unaltered interaction potential is not sufficient to infer that the tested protein has maintained its full tumor suppressor function. Indeed, two missense mutant proteins (Val219Met and Asn220Tyr) could not be distinguished from the wild-type protein in this test. Nonetheless it is likely that these two mutations are deleterious. The mutations, Val219Met and Asn220Tyr, were each independently reported twice, the former occurred at the somatic level (10,11), the latter at the germline level (4,8). In both cases the relevant amino acid positions are conserved in mouse and Drosophila schwannomin. Both positions also demonstrate high conservation with the corresponding positions of the ERM proteins. In one case, the Asn220Tyr mutation, the mutated allele was observed to cosegregate with the disease in the two corresponding unrelated NF2 families (4,8). Taken together, these observations strongly indicate that the region defined by the two mutations at adjacent positions (219 and 220) is likely to be critical for the tumor suppressive function of schwannomin. Further advances in the knowledge of the structure and function of schwannomin is mandatory to provide a detailed mechanism by which these mutations may impair tumor suppressor function.
HeLa cells were grown in Dulbecco's minimum essential medium (Gibco) supplemented with 10% fetal calf serum (Dutscher) and antibiotics, at 37°C, under a humidified 5% CO2 atmosphere.
The monoclonal antibody P5D4, raised against the 11-amino acid C-terminus of VSV-G, was purchased from Sigma. The affinity purified polyclonal antibody A19 (polyclonal anti-Nter), raised against a peptide corresponding to amino acids 2-21 of schwannomin, was purchased from TEBU. Monoclonal antibodies were raised against the C-terminal domain of schwannomin by injecting mice with a purified GST fusion protein corresponding to amino acids 469-595 of schwannomin. Screening of the antibodies was performed using a histidine-tagged DHFR fusion protein corresponding to the same domain of schwannomin. The monoclonal antibodies were purified on protein G columns. One of them, monoclonal 22, was used for the detection of the Sch-[Delta]Nter protein and for electron microscopy.
CTR433, a monoclonal antibody directed against the Golgi apparatus (31), was provided by M. Bornens. The monoclonal antibody OKT9, that reacts with the transferrin receptor (32) and the polyclonal antibody raised against the mannose 6-phosphate receptor (33) were supplied by A. Dautry and B. Hoflack, respectively, and were used as markers for endocytic compartments. The polyclonal antibody directed against the rough endoplasmic reticulum (34) was provided by D. Louvard. The anti-lamin B antibody (35) was supplied by J.-C. Courvalin. Monoclonal antibodies raised against the lysosomal protein lgp120 (Pharmingen), vimentin (DAKO), pancytokeratin (DAKO) and tubulin (Amersham) were also used. Rhodamine-coupled donkey anti-rabbit antibody and fluorescein-coupled sheep anti-mouse antibody were purchased from Amersham.
Construct encoding schwannomin isoform 1 tagged with VSV-G (Sch-Iso1-Tag). Site-directed mutagenesis of the HindIII site located at base pair 1487 was accomplished without modification of the coding properties of the sequence using the oligonucleotide (5'-GACATACCCAGCTTCAACC-3') and the Transformertm Site-Directed Mutagenesis Kit (Clontech). This allowed the use of the HindIII site located at base pair 864 of the schwannomin cDNA as a unique HindIII site. The cDNA encoding C-terminally tagged schwannomin was generated in two steps. First, the 5' region was amplified with a forward primer including a KpnI site (5'-GGGGTACCCCTGAGCCCCGCGCCATG-3') and a reverse primer overlapping the unique HindIII site at position 864 (5'-TTATTAACACGAAGCTTTGAGGAG-3'). Second, the 3' region was amplified with a forward primer overlapping the unique HindIII site at position 864 (5'-GTTTAACTCCTCAAAGCTTCGTG-3') and a reverse primer including the VSV-G tag sequence and a XbaI site (5'-GCTCTAGATTTACTTGCCCA-GCCGGTTCATCTCGATTCGGTGTATGGGCCTGGAGGGC-CCTCCCGGGGGAGCTCTTCAAAGAAGGCCAC-3'). These PCR products were inserted into pBluescript KS+ and the integrity of the amplified product was confirmed by sequencing. Finally, the entire tagged cDNA sequence was inserted into pCB6 eukaryotic expression vector to create the Sch-Iso1-Tag construct.
Constructs encoding the different tagged or untagged mutant cDNAs.Constructs encoding the various mutants of schwannomin were derived from Sch-Iso1 or Sch-Iso1-Tag by PCR using primers that overlap the mutation. Cloned PCR products were entirely resequenced to verify that no spurious mutation had been introduced during the PCR step. For the construct encoding the C-terminal domain of schwannomin isoform 1, the last 15 nucleotides of the 5'-untranslated region together with the three first codons were directly fused by PCR to position 752, corresponding to the codon of tyrosine 251.
Exponentially growing HeLa cells were seeded 24 h before transfection. DNA transfer was performed following the procedure described by Chen and Okayama (37). Briefly, cells were incubated overnight in the presence of calcium-phosphate DNA precipitate, washed with PBS, and fresh medium was then added. Cultures were analyzed for protein production after 24-30 h.
The transfected cells were washed with PBS and harvested by scraping directly into Laemmli sample buffer. Proteins were loaded onto a 10% polyacrylamide gel and separated under reducing conditions. Proteins were then electrotransferred onto nitrocellulose membrane (BioRad) and detected using the monoclonal anti-tag antibody or the polyclonal anti-Nter antibody. Detection was accomplished with horseradish peroxidase conjugated secondary antibodies (Amersham) and BM chemoluminescence blotting substrate (Boehringer Mannheim).
Transfected cells were fixed with 3% paraformaldehyde and permeabilized with 0.2% Triton X-100. Staining was performed in two sequential incubation steps. First, cells were incubated with the appropriate primary antibody depending upon the experiment. Then cells were incubated with the appropriate secondary fluorescent antibody, conjugated to either rhodamine or fluorescein.
For fluorescent double labeling of F-actin and schwannomin derived proteins, cells were incubated with a mixture of fluorescent phalloidin (conjugated to either rhodamine or fluorescein, 2 U per dish; Sigma) and the anti-tag antibody or the commercial anti-Nter antibody. Cells were mounted in Vectashield medium (Vector Laboratories) with DAPI (Sigma). Cells were examined with a Leica epifluorescence microscope or with a confocal Leica microscope.
Experiments were performed as described by Algrain et al. (30).
Indirect immunofluorescence of detergent-extracted HeLa cells. Cells on coverslips were washed with PBS, and dipped for 5 s in four different beakers each containing 30 ml of extraction buffer (50 mM MES, 3 mM EGTA, 5 mM MgCl2, 0.5% wt/vol Triton X-100, pH 6.4). Cells were then fixed with 3% paraformaldehyde and indirect immunofluorescence was performed.
Cell fractionation into detergent-soluble and detergent-insoluble material. HeLa transfected cells were extracted for 40 s with 300 µl of the extraction buffer described above. Detergent-soluble material was precipitated in 85% acetone for 4 h at -20°C and the pellet recovered after centrifugation. Detergent-insoluble material was scraped in PBS and then centrifuged. Pellets containing either the detergent-soluble or the detergent-insoluble material were resuspended in the same volume of Laemmli sample buffer. Equal amounts of each fraction were analyzed by immunoblotting.
Nocodazole treatment. Transfected cells were treated with 30 µM nocodazole (Sigma) for 5 min at 4°C, then transferred to 37°C for 30 min. After washing with PBS previously warmed to 37°C, cells were fixed with 3% paraformaldehyde at 37°C.
Transfected cells were fixed with 2% formaldehyde, 1% glutaraldehyde in 100 mM phosphate buffer pH 7.2. Cells were then embedded as a pellet in 7.5% gelatin. Ultra thin frozen sections were prepared according to Tokuyasu (38) on a RMC MT-7 cryoultramicrotome (Tucson). They were labeled with monoclonal 22 combined with anti-mouse IgG antibodies conjugated to 10 nm gold particles (Biocell, used at OD = 0.15). Labeled sections were positively-negatively stained according to Griffiths et al. (39) and observed on a Jeol 1210 electron microscope.
NF2, neurofibromatosis type 2; ERM, ezrin-radixin-moesin; GST, glutathione S-transferase; DHFR, dihydrofolate reductase; PCR, polymerase chain reaction; RT-PCR, reverse-transcriptase polymerase chain reaction; VSV-G, vesicular stomatitis virus glycoprotein G; PBS, phosphate buffered saline; F-actin, filamentous actin.
We wish to thank D. Louvard for his encouragement and P. Clark for numerous stimulating discussions and critical reading of the manuscript. We are grateful to P. Vicart for the generous gift of the anti-vimentin and anti-cytokeratin antibodies. We thank D. Louvard, A. Dautry, M. Bornens, B. Hoflack and J.-C. Courvalin for providing antibodies. We are grateful to S. Rempel and D. Lowy for providing SF1335 and RN22 cell lines, respectively. Thanks are also due to R. Hellio for his help in performing the confocal microscopy analysis. This work was supported by the Ligue Nationale Française contre le Cancer (Bureau National et Comité Départemental de l'Yonne), the Ministre de la Recherche et de l'Enseignement Supérieur, the Association pour la Recherche sur le Cancer, and the Fédération Nationale des Groupements des Entreprises Françaises dans la Lutte contre le Cancer.
Human Molecular Genetics Pages
Introdution
Results
Expression of wild-type and mutant schwannomin proteins
Subcellular distribution
Detergent extractability of schwannomin isoforms 1 and 2 and of mutant schwannomins
Cytochalasin D treatment
Search for interaction with other cytoskeletal components
Discussion
Materials And Methods
Cell culture
Antibodies
DNA constructs
Transient expression in HeLa cells
Immunoblotting
Indirect immunofluorescence
Detergent extraction
Cytoskeleton assays
Electron microscopy
Abbreviations
Acknowledgements
References
Schwannomin isoforms 1 and 2. Immunofluorescence staining with polyclonal anti-Nter antibody or monoclonal anti-tag of HeLa cells transiently transfected with Sch-Iso1 or Sch-Iso1-Tag indicated that schwannomin isoform 1 was predominantly distributed underneath the plasma membrane (Fig. 3A), a location that has been reported for schwannomin in COS cells (22). When double fluorescence staining with the anti-tag antibody and phalloidin was performed, schwannomin isoform 1 appeared concentrated in F-actin-rich membrane protrusions such as membrane ruffles, microspikes and filopodia (Fig. 3A and B). Schwannomin was not detected associated with actin stress fibers. Overproduction of schwannomin did not induce evident modification of actin-containing structures (Fig. 3B). Confocal microscopy analysis of transfected HeLa cells confirmed these findings and indicated a prominent distribution of schwannomin at the cortex. Cells transfected with a construct encoding schwannomin isoform 2 presented an identical distribution for the transiently expressed protein. The same submembranous localization was also observed in the pig kidney epithelial cell line LLC-PK1, in the monkey kidney fibroblast-like cell line CV-1 or in cell lines derived from a rat schwannoma (RN22, a kind gift from D. Lowy) or from a human meningioma (SF1335, kindly provided by S. Rempel) (data not shown).
Construct encoding wild-type schwannomin isoform 1 (Sch-Iso1).Human schwannomin cDNA containing the complete coding sequence flanked by 24 and 133 bp of the 5' and 3' untranslated regions, respectively, was inserted in the EcoRI site of the eukaryotic expression vector pCB6 (36). This insertion placed the schwannomin cDNA under the control of the cytomegalovirus promoter (construct Sch-Iso1).
Cytochalasin D treatment. Transfected cells were treated with 2.5 µM cytochalasin D (Sigma) for 20 min at 37°C. After washing with PBS, cells were fixed with 3% paraformaldehyde.
REFERENCES
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 6 Jan 1998
Copyright© Oxford University Press, 1998.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  J. T. Stickney, W. C. Bacon, M. Rojas, N. Ratner, and W. Ip Activation of the Tumor Suppressor Merlin Modulates Its Interaction with Lipid Rafts Cancer Res., April 15, 2004; 64(8): 2717 - 2724. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Kaempchen, K. Mielke, T. Utermark, S. Langmesser, and C. O. Hanemann Upregulation of the Rac1/JNK signaling pathway in primary human schwannoma cells Hum. Mol. Genet., June 1, 2003; 12(11): 1211 - 1221. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Gautreau, B. T. Fievet, E. Brault, C. Antony, A. Houdusse, D. Louvard, and M. Arpin Isolation and Characterization of an Aggresome Determinant in the NF2 Tumor Suppressor J. Biol. Chem., February 14, 2003; 278(8): 6235 - 6242. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Kressel and B. Schmucker Nucleocytoplasmic transfer of the NF2 tumor suppressor protein merlin is regulated by exon 2 and a CRM1-dependent nuclear export signal in exon 15 Hum. Mol. Genet., September 15, 2002; 11(19): 2269 - 2278. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Gautreau, J. Manent, B. Fievet, D. Louvard, M. Giovannini, and M. Arpin Mutant Products of the NF2 Tumor Suppressor Gene Are Degraded by the Ubiquitin-Proteasome Pathway J. Biol. Chem., August 23, 2002; 277(35): 31279 - 31282. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E Brault, A Gautreau, M Lamarine, I Callebaut, G Thomas, and L Goutebroze Normal membrane localization and actin association of the NF2 tumor suppressor protein are dependent on folding of its N-terminal domain J. Cell Sci., January 5, 2001; 114(10): 1901 - 1912. [Abstract] [PDF]
|
|
|
|
 |
|
 M. Giovannini, E. Robanus-Maandag, M. van der Valk, M. Niwa-Kawakita, V. Abramowski, L. Goutebroze, J. M. Woodruff, A. Berns, and G. Thomas Conditional biallelic Nf2 mutation in the mouse promotes manifestations of human neurofibromatosis type 2 Genes & Dev., July 1, 2000; 14(13): 1617 - 1630. [Abstract] [Full Text]
|
|
|
|
 |
|
 L. Goutebroze, E. Brault, C. Muchardt, J. Camonis, and G. Thomas Cloning and Characterization of SCHIP-1, a Novel Protein Interacting Specifically with Spliced Isoforms and Naturally Occurring Mutant NF2 Proteins Mol. Cell. Biol., March 1, 2000; 20(5): 1699 - 1712. [Abstract] [Full Text]
|
|
|
|
|
|
M. Giovannini, E. Robanus-Maandag, M. Niwa-Kawakita, M. van der Valk, J. M. Woodruff, L. Goutebroze, P. Mérel, A. Berns, and G. Thomas Schwann cell hyperplasia and tumors in transgenic mice expressing a naturally occurring mutant NF2 protein Genes & Dev., April 15, 1999; 13(8): 978 - 986. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||