Skip Navigation


Human Molecular Genetics Advance Access originally published online on June 12, 2007
Human Molecular Genetics 2007 16(14):1742-1751; doi:10.1093/hmg/ddm122
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Supplementary Data
Right arrow All Versions of this Article:
16/14/1742    most recent
ddm122v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (5)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Muranen, T.
Right arrow Articles by Carpén, O.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Muranen, T.
Right arrow Articles by Carpén, O.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2007. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

The tumor suppressor merlin interacts with microtubules and modulates Schwann cell microtubule cytoskeleton

Taru Muranen1,*, Mikaela Grönholm1, Aurelie Lampin2, Dominique Lallemand2, Fang Zhao1, Marco Giovannini2 and Olli Carpén1,3

1 Program of Neuroscience, Department of Pathology, University of Helsinki, Biomedicum Helsinki C511, PL 63, 0014, Finland, 2 Inserm U674, Fondation Jean Dausset-CEPH et Institut Universitaire d'Hématologie, 75010 Paris, France and 3 Department of Pathology, University of Turku and Turku University Central Hospital, 20520 Turku, Finland

* To whom correspondence should be addressed. Tel: +358 9 19125651; Fax: +358 9 4717194; Email: taru.muranen{at}helsinki.fi

Received March 9, 2007; Accepted April 30, 2007


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 REFERENCES
 
The lack of neurofibromatosis 2 tumor suppressor protein merlin leads to the formation of nervous system tumors, specifically schwannomas and meningiomas. Merlin is considered to act as a tumor suppressor at the cell membrane, where it links transmembrane receptors to the actin cytoskeleton. Several tumor suppressors interact with another component of the cytoskeleton, the microtubules, in a regulated manner and control their dynamics. In this work, we identify merlin as a novel microtubule-organizing protein. We identify two tubulin-binding sites in merlin, one residing at the N-terminal FERM-domain and another at the C-terminal domain. Merlin's intramolecular association and phosphorylation of serine 518 regulate the interaction between merlin and tubulin. Analysis of cultured glioma cells indicates colocalization between merlin and microtubules especially during cell division. In primary mouse Schwann cells only minor colocalization at the cell periphery of interphase cells is seen. However, these cells drastically change their microtubule organization upon loss of merlin indicating a functional association of the proteins. Both in vitro assays and in vivo studies in Schwann cells indicate that merlin promotes tubulin polymerization. The results show that merlin plays a key role in the regulation of the Schwann cell microtubule cytoskeleton and suggest a mechanism by which loss of merlin leads to cytoskeletal defects observed in human schwannomas.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 REFERENCES
 
Inactivation of the neurofibromatosis 2 (NF2) tumor suppressor gene leads to the development of multiple benign tumors of the nervous system, particularly meningiomas and schwannomas (1). The NF2 gene encodes a 595-amino acid protein merlin (schwannomin), which is related to the ezrin–radixin–moesin (ERM) protein family. Merlin and ERM proteins are located primarily underneath the cell membrane where they anchor transmembrane proteins to the actin cytoskeleton (2,3). They form homo- and heterotypic interactions (4) which in turn regulate their binding to other proteins (5–7). Head-to-tail binding leads to a closed conformation of the ERM proteins (4); for merlin, the closed form is thought to act as a tumor suppressor, whereas the open protein is unable to regulate growth. Phosphorylation of a C-terminal serine (S518) by p21-activated kinase (PAK) or cAMP-dependent protein kinase A (PKA) weakens merlin's self-association and is believed to inactivate the growth-suppressing activity of merlin (8,9). However, the functional regulation of merlin is still not completely understood.

Merlin is also involved in receptor recycling and endocytosis. It inhibits platelet-derived growth factor receptor degradation (10) and binds hepatocyte growth factor-regulated tyrosine kinase substrate (HRS) (11), which is known to regulate receptor tyrosine kinase trafficking to the degradation pathway (12). Recently, merlin was shown to regulate EGF receptor recycling and turnover in Drosophila (13). Thus, increasing evidence positions merlin at the membrane where it can bind membrane receptors and regulate their expression and localization.

A common feature of many tumor suppressors is their ability to interact with microtubules and regulate microtubule stability (14,15). Microtubules, themselves, are also known to regulate cell growth. They, for instance, control cell division and regulate endocytosis and recycling of growth factor receptors (16). We set out to study merlin–tubulin interaction when we noticed that endogenous merlin colocalizes with tubulin in mitotic structures of U251 glioma cells (17). Although merlin has previously been suggested to bind tubulin in vitro (3,18), no further evidence has been provided to support the binding. Here, we have studied the association between merlin and tubulin, the regulation of the interaction, the effects of merlin on microtubule dynamics and the consequences of loss of merlin on microtubules in Schwann cells. Our results show that merlin plays an important role in regulating microtubule cytoskeleton of mouse primary Schwann cells.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 REFERENCES
 
Merlin associates with microtubules
We have previously shown that in synchronized U251 glioma cells merlin localizes to mitotic structures (17,19). This led us to analyze the distribution of endogenous merlin and tubulin at various cell cycle stages in U251 glioma cells (Fig. 1A–H). We noticed parallel accumulation of merlin and tubulin around the nucleus in interphase cells before mitosis (A). During mitosis, merlin colocalized with microtubules at the mitotic spindles, although some diffuse merlin staining was also noticed in the cytoplasm (B and C). During cytokinesis especially the midbody demonstrated high degree of colocalization between merlin and tubulin (D and E). The colocalization was lost at early G1, when merlin accumulated in the nucleus (F). The staining patterns remained separate during the entire G1 and S phase (G). At G2, a partial colocalization was seen (H). We also studied cultured primary mouse Schwann cells, whose purity was verified by a Schwann cell-specific marker p75 (Supplementary Material, Fig. S1). In interphase cells merlin had a predominantly submembraneous localization, whereas tubulin was localized at the cytoplasm. Areas of potential colocalization were identified from confocal sections using image analysis software. The analysis revealed occasional sub-plasmalemmal regions and (I and J) and cell extensions (K and L) (colocalized points shown in white in Fig. 1J, L), in which codistribution was seen. In mitotic Schwann cells staining was mostly diffuse.


Figure 1
View larger version (59K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 1. Distribution of merlin and microtubules in U251 glioma and primary mouse Schwann cells. (A–H) U251 glioma cells were synchronized and stained for merlin (green, A19) and tubulin (red, {alpha}-tubulin). The cells were fixed at different points of the cell cycle, the cell cycle phase was verified by FACS analysis and the localization of the two proteins was analyzed by confocal microscopy. Merlin and tubulin start to colocalize already in interphase cells when cells are approaching mitosis (A). During mitosis merlin localizes to mitotic spindles (B, C) and during cytokinesis at the midbody (D, E and insets). After mitosis the two proteins have separate staining patterns (F, G). At late G2, some colocalization can be seen again (H). (I–L) In the primary mouse Schwann cells merlin (green, A19) shows a predominantly submembraneous distribution, whereas microtubules (red, ß-tubulin) are distributed within the cells body. Analysis of confocal sections with Image J analysis software demonstrates occasional submembraneous regions, where the two proteins colocalize (white areas in J and L). Scale bar 10 µm.

 
Merlin binds polymerized microtubules in vitro
We then wanted to identify the tubulin-binding regions in merlin. We mapped the interaction sites with a tubulin pull-down assay using various purified glutathione S-transferase (GST)-merlin constructs. Tubulin was incubated with different GST-constructs after which polymerized tubulin was pulled down by centrifugation and quantified. Both the N-terminal FERM- (amino acids 1–314) and the C-terminal (amino acids 492–595) domain were recovered from the tubulin pellet, whereas the {alpha}-helical region (amino acids 314–492) of merlin or GST alone (quantification not shown) did not bind tubulin (Fig. 2B, C). We further mapped the N-terminal binding site to amino acids 1–100 and the C-terminal binding site to residues 492–547 (Fig. 2D). As a positive control we used microtubule-associated protein 2 (MAP2), a well-characterized tubulin interacting protein. As an additional control, we tested the N-terminus of ezrin, a FERM-domain protein homologous to merlin. Similar to merlin, the N-terminus of ezrin could be found in the tubulin pellet (Fig. 2D).


Figure 2
View larger version (45K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 2. Merlin binds polymerized tubulin in vitro. (A) A schematic drawing on merlin's domains. (B) Tubulin binding sites in merlin were mapped by a tubulin pull-down assay, in which recombinant GST-merlin was pulled down with polymerized microtubules by ultracentrifugation. Merlin N (amino acids 1–314), {alpha}-helical merlin (amino acids 314–492) and merlin C (amino acids 492–595) constructs were used in the assay. Total (t), supernatant (s) and pellet (p) fractions with or without tubulin were run into SDS–PAGE, blotted either with merlin (A19 or C18) or GST-antibody and analyzed. (C) The amount of merlin in the pellet fraction was quantified by western blots by Typhoon Imager and compared to the total amount of merlin. (D) The binding sites in merlin were further mapped with merlin constructs 1–100 and 492–547. GST alone, MAP2 and ezrin N (1–309) were used as controls in the pull-down assay. The tubulin blots can be found from Supplementary Material, Figure S2. (E) In vitro translated 35S methionine labeled merlin was produced and used for in vitro tubulin pull-down experiments. Merlin was pulled down with polymerized microtubules and the amount of merlin in the pellet was analyzed and compared with total amount of merlin (%). Merlin isoforms 1 and 2 and C-terminally deleted merlin constructs associated with microtubules. A patient mutation construct, merlin 1–547, did not associate with microtubules equally well. The graph represents the average of five experiments.

 
In addition, we produced in vitro-translated methionine-labeled merlin constructs and pulled them down with polymerized tubulin. With this assay we could see an interaction between tubulin and merlin isoforms I and II (Fig. 2E). C-terminal deletion constructs 1–569 and 1–587 bound tubulin similar to isoform I. As truncating mutations around amino acid 547 cause NF2 (http://neurosurgery.mgh.harvard.edu/NFclinic/NFresearch.htm), we tested whether merlin 1–547 would show altered binding to tubulin. Indeed, this construct demonstrated reduced binding in comparison with isoform I. However, a further truncation of 10 residues returned the binding activity (Fig. 2E) showing the complexity of the binding interphase.

Intramolecular association and phosphorylation of merlin regulate its binding to microtubules
Merlin undergoes conformational regulation and can form intramolecular association with its N- and C-terminal domains (6). This intramolecular association can mask many of the binding sites in merlin. Therefore, we studied whether intramolecular binding affects merlin–tubulin binding. Purified C-terminal GST-merlin was incubated together with increasing amounts of N-terminal merlin in order to saturate the C-terminal binding site. When C-terminus was saturated with the N-terminus it could no longer bind tubulin (Fig. 3A). GST alone incubated with the C-terminus was used as a negative control.


Figure 3
View larger version (27K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 3. Intramolecular association and phosphorylation of S518 in merlin inhibits its binding to tubulin. (A) In vitro tubulin pull-down assay was performed with purified GST-merlin constructs. C-terminal merlin (amino acids 492–595) was incubated with increasing amounts of N-terminal merlin (1–314). Total (t), supernatant (s) and pellet (p) samples were analyzed by western blot with an antibody recognizing N- or C-terminal merlin (A19 or C18). When the C-terminus was saturated with the N-terminus, it could no longer bind tubulin. As a control, the C-terminus was incubated with plain GST. (B) The C-terminus of GST-merlin (wild-type or with S518A mutant) was phosphorylated in vitro with PKA and tubulin pull-down assay was performed. The amount of merlin in pellet was analyzed and compared with the total amount of merlin (%). PKA activity decreased markedly (P = 0.006) the binding of wild-type merlin to tubulin, but did not affect binding of the S518A mutant. The graph shows average ± SD of six experiments.

 
The C-terminal S518 of merlin can be phosphorylated by PAKs and cAMP-dependent PKA (8,9,20). This phosphorylation is thought to affect merlin's growth-regulating activity (21–23). We studied whether the phosphorylation plays a role in merlin–tubulin binding. We treated the C-terminus (492–595) of wild-type merlin or non-phosphorylatable merlin S518A mutant in vitro with PKA and performed the tubulin pull-down assay (Fig. 3B). Phosphorylation was verified by 32P ATP labeling (not shown). Significantly lower amount of merlin was pulled down after PKA treatment indicating that the phosphorylation of the serine 518 decreases the in vitro binding of merlin to tubulin.

Merlin enhances microtubule polymerization in vitro
Many tumor suppressor proteins have been reported to affect tubulin polymerization (14) and therefore we studied whether merlin has an effect on tubulin polymerization. Purified tubulin was polymerized in vitro and the polymerization was monitored at OD350. Microtubule-associated proteins (MAPs) and Taxol were used as positive controls and GST alone as a zero control. Nocodazole was used to control tubulin depolymerization (not shown). When recombinant merlin was added to the polymerization reaction, the rate of tubulin polymerization was increased (Fig. 4). Merlin N-, {alpha}- or C alone were not able to induce tubulin polymerization (not shown) and merlin 1–547, which has reduced affinity for tubulin, was also unable to induce tubulin polymerization.


Figure 4
View larger version (19K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 4. Merlin enhances tubulin polymerization in vitro. Purified tubulin from bovine brain was polymerized in vitro at +37°C and the polymerization was monitored at OD350. Merlin GST-fusion proteins were incubated together with tubulin. Taxol and MAP2 were used as positive controls for tubulin polymerization, tubulin alone and tubulin together with GST were used as zero controls. When full-length merlin (WT) was added to the polymerization reaction, tubulin polymerized more rapidly. Merlin 1–547 was not able to induce tubulin polymerization. The graph shows an average of three experiments.

 
Merlin affects the structure of Schwann cell microtubule cytoskeleton
As schwannomas, the hallmark of Neurofibromatosis 2, originate from Schwann cells, we studied merlin's effect on microtubule cytoskeleton in this cell type. We used primary mouse Schwann cells lacking merlin (Nf2flox2/flox2, after addition of Cre-recombinase leading to genotype Nf2{Delta}2/{Delta}2, from now on marked as Nf2–/–, see also discussion) (24) and re-introduced wild-type merlin via adenovirus infection (Nf2 add-back). The microtubule cytoskeleton of the Nf2–/– mouse Schwann cells differed markedly from cells re-expressing merlin via adeno-infection (Nf2 add-back). In the Nf2 add-back cells, microtubules were assembled as dense cables reaching from one end of the cell to the other and the cells displayed a normal spindle-like structure (Fig. 5A, D). Instead, the microtubules of the Nf2–/– cells were disorganized, and the overall cell shape was more spread (Fig. 5B, D). Similar morphological differences have also been described in cultured human Schwann cells versus schwannoma cells (25).


Figure 5
View larger version (79K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 5. Microtubule cytoskeleton is altered in NF2–/– Schwann cells. Primary mouse Schwann cells with merlin null background (NF2{Delta}exon2, here marked as –/– (B) and the same cells with adeno-infected wild-type merlin (MWT) (A) were stained for merlin (green, A19) and ß-tubulin (red) and imaged by fluorescence microscope and the images were deconvoluted. Microtubule organization is different in merlin –/– cells and in cells expressing merlin. Magnification x630. (C) Triton X-100 lysates made from the –/–, +/+ and add-back (MWT) cells. Protein amounts were analyzed by Bradford assay and equal protein amounts used. As an additional loading control, the membranes were stained with Ponceau red. The lysates were probed for merlin (A-19), {alpha}-, ß- and acetylated-tubulin by western blot. Total lysates of the –/–, +/+ and merlin add-back cells (MWT) were solubilized in 6 M Urea buffer and probed for ß-tubulin and merlin (A-19). (D) A high-magnification image of tubulin in MWT and –/– cells. Scale bar 10 µm.

 
The amount of polymerized versus unpolymerized (i.e. soluble) tubulin was analyzed from Schwann cells extracted with Triton X-100. More soluble {alpha}- and ß-tubulin were present in Nf2–/– cells than in wild-type (Nf2flox2/flox2, marked as Nf2+/+) or add-back cells (MWT) (Fig. 5C). The amount of soluble tubulin was associated with the expression level of merlin; increasing merlin expression resulting in reduced soluble tubulin. The total amount of tubulin in Nf2–/– and the add-back Schwann cells were evaluated after solubilization in 6 M urea buffer. When the urea-lysates were analyzed in SDS–PAGE the add-back (MWT) and the Nf2+/+ cells appeared to contain at least equal amount of tubulin as the Nf2–/– cells. From this data we concluded that the ratio of polymerized/unpolymerized tubulin differs between Nf2–/– and add-back cells; i.e. in Nf2–/– Schwann cells the proportion of free tubulin ({alpha}/ß) is higher than in the wild-type cells.

Tubulin polymerization dynamics differ in –/– and Nf2 add-back Schwann cells
The difference in the integrity of the tubulin cytoskeleton in –/– and add-back Schwann cells led us to test whether a difference in the polymerization rates of tubulin is also seen in vivo. We treated the Schwann cells with Nocodazole, which sequesters tubulin monomers and thereby depolymerizes microtubules. Incubation with Nocodazole reversed the normal morphology of Schwann cells and the cells obtained more fibroblast-like morphology (Fig. 6A, B; Supplementary video 1 and 2, for add-back and –/– cells, respectively). After 5 min of Nocodazole block release both cell types began to regain their normal shape by loosing the flattened morphology. The add-back Schwann cells began to obtain their normal morphology (i.e. long extensions) 2 h after the block release, whereas the –/– cells had only unorganized microtubules at this time-point (Fig. 6C, D; 120 min). After 4 h of block release both cell types had their microtubules re-grown. In addition, the add-back cells had regained their spindle-like shape (Fig. 6C, D; 240 min). This implies that the normal Schwann cell morphology is dependent on intact microtubule cytoskeleton and that merlin enhances microtubule organization in Schwann cells.


Figure 6
View larger version (121K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 6. Microtubule recovery is faster in merlin-expressing primary Schwann cells. (A, B) Primary mouse Schwann cells lacking merlin (–/–) (B) or with adeno-infected merlin add-back (MWT) (A) were treated with Nocodazole to depolymerize microtubules. The block was released and cells were either imaged with DIC filter for 4 h once every 30 s (Supplementary videos 1 and 2, 200x magnification) or still images were captured from the indicated time points. Alternatively add-back MWT (C) and –/– (D) cells were fixed at indicated time points after which cells were stained with ß-tubulin and a Schwann cell marker p75 and imaged with fluorescence microscope (magnification 630x) after which the images were deconvoluted. Nuclei were stained with Hoechst (blue). (E) For FRAP experiments primary Schwann cells (–/– and add-back) were transfected with EGFP-tubulin and 24 h after transfection the cells were imaged for FRAP with confocal microscope. Pre-bleach images were obtained and the region of interest (equal size region of interest was used with all the cells) was bleached. Post-bleach images were acquired at 2 s intervals. From the data, background was subtracted and normalized similarly for all the samples. Eight cells of each construct (–/– and add-back/MWT) were imaged and the averaged data was used for data analysis. Cells that were expressing merlin had a faster recovery rate of tubulin than the –/– cells (P < 0.001).

 
In addition, we used fluorescence recovery after photobleaching (FRAP) technology to study, whether tubulin dynamics were altered in primary Schwann cells lacking merlin. Schwann cells were transfected with EGFP-tubulin and 24 h after transfection the cells were analyzed by FRAP. Cells expressing merlin had a 25% faster tubulin recovery rate than the Nf2–/– cells (Fig. 6E) (P < 0.001). The data suggest that merlin is able to enhance microtubule dynamics of Schwann cells in vivo.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 REFERENCES
 
In this study we describe a novel function for merlin in the regulation of microtubule organization and dynamics. We show that merlin directly binds tubulin and regulates microtubule dynamics in vitro and in vivo. Our results show that merlin contains two tubulin-binding sites, one in the FERM-domain and another in the C-terminal domain. The existence of two separate binding sites in merlin indicates that merlin could bind either laterally to the sides of the microtubules or cross-link individual microtubules to each other, this way bundling the microtubules to thicker entities. In addition, the entire protein is needed to induce tubulin polymerization, as shorter merlin fragments were not able to induce polymerization. Interestingly, the truncating patient mutation 1–547 showed reduced tubulin binding and did not induce polymerization, although it harbors the two identified tubulin binding sites. A further 10 residue deletion returned the binding activity indicating that mere loss of the binding site does not explain the result. It is possible that the 1–547 mutation leads to an altered folding of the protein and thereby prevents the interaction with tubulin.

An intramolecular association and phosphorylation of S518 in merlin can regulate its binding to microtubules. The binding of several other proteins to microtubules is regulated by phosphorylation. For example a major phosphoprotein in the brain, MAP2, is phosphorylated by PKA and upon phosphorylation dissociates from microtubules (26). Both MAP2 and merlin function as PKA-anchoring proteins (AKAPs) (27,28) and therefore merlin might have a similar function as MAP2, targeting PKA to microtubules. It is of interest that both kinases, PKA and PAK, known to regulate merlin phosphorylation, localize to centrosomes at mitosis and affect microtubule formation, in analogy with merlin (8,9,20,29,30). Previous results have suggested that phosphorylation of merlin results in the conformational opening of the molecule allowing it to interact with other proteins (23). However, in light of our study (and our unpublished observations from other interaction studies) it seems that the regulation is more complex than previously thought. We hypothesize that merlin can be ‘open’ but non-phosphorylated at S518 since in our results only the open but S518 non-phosphorylated merlin binds tubulin.

To study the role of merlin on microtubule dynamics in vivo, we used cells from genetically engineered mouse, in which wild-type Nf2 gene has been replaced with a NF2flox2/flox2 variant. Addition of Cre-recombinase to the cell cultures leads to deletion of exon 2 (Nf2{Delta}exon2/{Delta}exon2 genotype) and to complete loss of merlin (Fig. 5). Overexpression of a {Delta}exon 2 mutant in cultured cells is known to disturb cell adhesion (31). However, in our model, replacement of the wild-type allele with the Nf2{Delta}exon2 leads to a greatly reduced merlin expression, when compared with the wild-type Nf2 gene as shown in mouse tissues (24). Indeed, Cre-recombinase treated MEF cells from the same mice have previously been used as a model of genetically engineered Nf2–/– cells (32).

Our data suggest that merlin plays an important role in regulating Schwann cell microtubule cytoskeleton. Loss of merlin in Schwann cells results in a change of the spindle-like morphology and alteration in microtubule organization. Also the ratio of polymerized versus unpolymerized tubulin is affected by merlin, i.e. Schwann cells that lack merlin have more unpolymerized tubulin than cells expressing merlin. The Nf2–/– cells contain more acetylated tubulin, which is regarded as a marker of older and more stable structures. This might imply that tubulin polymerization/turn-over is not as efficient in the Nf2–/– cells as in merlin-expressing cells. The results from the Nocodazole-repolymerization experiment confirm these data. Our FRAP data further supports this finding as tubulin turnover appears to be higher in merlin-expressing Schwann cells. Unfortunately, in our experimental FRAP set-up, it was not possible to monitor the dynamics of individual microtubules due to the thickness of the cells and to the high amount of free tubulin dimers. Therefore, by FRAP we have only analyzed tubulin recovery as a tubulin pool and not as individual microtubule polymers. The difference might have been even greater if we were able to study individual microtubule polymers. It also seems that microtubules are a major contributor to the Schwann cell morphology, since Schwann cell morphology resembled that of fibroblasts after depolymerization of microtubules.

Microtubules regulate endocytic pathways (16,33). Recent papers have linked merlin to endocytic events and receptor recycling (10–13). It has been shown that cells use a specific sorting mechanism of fast and slow maturation of the endocytic vesicles. Certain receptors, such as EGFR, are recycled through a dynamic pool of early endosomes that are highly mobile on microtubules and that mature rapidly towards the late endosomes. This rapid movement and maturation is completely dependent on microtubules; if microtubules are disrupted, the dynamic vesicles will non-selectively join all early endosomes and are less-efficiently degraded (34). Two papers have localized merlin to early endosomes (11,13). Thus, merlin may function as a linker at the plasma membrane, helping receptor-containing endocytosed vesicles to attach to microtubules. If cells lack merlin, this receptor recycling might be slower, and the receptors would be cleared from the membrane less efficiently. We propose that merlin plays a dual role at the membrane of Schwann cells, partly by transferring early endocytic vesicles to rapidly growing microtubules with the help of its interaction partner HRS and partly by increasing the microtubule polymerization rate, thus enhancing the vesicle maturation process.

In the past years several tumor suppressor proteins, including p53, APC, VHL and BRCA1 have been shown to bind and regulate microtubules (14). In line with merlin, also APC and VHL show only limited colocalization with microtubules, but yet affect their dynamics in vivo (35,36). Here, we show that also merlin binds microtubules, regulates their polymerization and has an important role in the establishment of normal Schwann cell morphology. Schwannoma cells from NF2 patients display altered morphology with long, multiple extensions (37). Based on our study this altered morphology may be associated with a disturbed microtubule cytoskeleton due to the lack of merlin, thus linking the interplay of merlin and tubulin to normal Schwann cell development and possibly to schwannoma formation.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
SUPPLEMENTARY MATERIAL
 REFERENCES
 
Cell lines and antibodies
U251 glioma cells were maintained in Dulbecco's Minimum Essential Medium (MEM), supplemented with 10% fetal calf serum (FCS) (PromoCell, Heidelberg, Germany), 1% L-glutamine and 50 µg/ml gentamycin (Invitrogen). Cells were fixed in 3.5% paraformaldehyde, pH 7.5. Primary Schwann cells were isolated and cultured as previously described (38) from NF2 knock-out mice with both alleles of the NF2 exon 2 loxed, thus producing after Cre recombination the NF2{Delta}2/{Delta}2 genotype (24). The cells were used either at passage two or three. Anti-merlin polyclonal rabbit antibodies A-19 sc-331 (epitope 2-21), C-18 sc-332 (Santa Cruz Biotechnology, Santa Cruz, CA, USA), 1398 NF2 (39), anti-schwannomin (40) and mouse mAb KF10 (39) were used. TO-PRO 3-iodide probe (Invitrogen, Molecular Probes) was used for DNA staining. Monoclonal anti-{alpha}-, -ß- and -acetylated-tubulin antibodies (Sigma–Aldrich) were used to detect tubulin. Anti-GST polyclonal goat antibody (GE Healthcare) was used to detect GST proteins. p75 antibody (41) was used as a Schwann cell marker. Alexa 488-, 568-, 594- and 633-conjugated goat anti-mouse and goat anti-rabbit antibodies (Invitrogen, Molecular Probes) were used as secondary antibodies in immunofluorescence and HRP-conjugated rabbit anti-mouse and swine anti-rabbit and anti-goat (Santa Cruz) secondary antibodies (DAKO A/S, Glostrup, Denmark) in western blot analysis.

Plasmids and protein expression constructs
The following merlin constructs: isoform I (amino acids 1–595), isoform II and isoform I variants S518A, S518D, 1-314, 1-537 and 1-547) in pcDNA3 (Invitrogen) were used for in vitro translation. Luciferase cDNA was used as a negative control. For expression of recombinant GST-merlin fusion proteins, the following constructs were used: merlin N-terminus (amino acids 1–100 and 1–309), merlin a-helical domain (amino acids 314–477), C-terminus (amino acids 492–595) and C-terminus with S518A mutation as described previously (9), merlin 1–537 and merlin 1–547.

Adenovirus construct were made using Stratagene's AdEasy system. Full-length merlin was cloned into the adeno vector and the virus was produced according to manufacturer's instructions in 293A cells. The infection titer was optimized to have 100% efficiency and verified by immunofluorescence staining. Schwann cells lacking merlin were infected with adenoviruses 3 days before the experiments were conducted. EGFP-tubulin construct was obtained from Invitrogen, and transfection to primary Schwann cells was done with Lipofectamine PLUS (Invitrogen).

In vitro translation and quantification of labeled proteins
Merlin pcDNA3 plasmids (Invitrogen) were used as a template for a T7-coupled rabbit reticulocyte transcription–translation system (Promega, Madison, WI, USA) in the presence of 35S-methionine. Five microlitres of 50 µl reaction were run in to SDS–PAGE, gel was dried and exposed to film to determine the size and amount of the labeled protein. For quantification the gel was also exposed to PhosphoImager low energy-plate (GE Healthcare), read by TyphoonImager 9400 (GE Healthcare) and analyzed by ImageQuantTL2003 software (GE Healthcare).

GST-fusion protein production
GST-fusion proteins were expressed in Escherichia coli DH5{alpha} and purified following standard protocol. Fusion protein was eluted from the Glutathione Sepharose beads (GE Healthcare) by 5 mM reduced glutathione in 50 mM Tris–HCl pH 8.0, over night at +4°C. Before some tubulin pull-down assays different amounts of N- and C-terminus were incubated in 50 mM Tris–HCl, 150 mM NaCl (pH 8.0) for 30 min at room temperature. The eluted fusion proteins for the PKA assay were dialyzed against 20 mM Tris–HCl, 10 mM MgCl2, pH 7.4 at +4°C o/n. Wild-type merlin was produced in Sf9 insect cell line. Merlin with an N-terminal GST-tag was cloned into a baculovirus transfer vector pAcG2T (BD Biosciences) and then produced with the BaculoGOLD system (BD Biosciences). Sf9 cells were used to produce recombinant merlin according to user's manual (Invitrogen). Full-length merlin was purified after 1 day of infection following standard GST-purification protocol. The proteins were used immediately after purification.

In vitro phosphorylation
GST-proteins were dialyzed against the reaction buffer and run into SDS–PAGE, protein amounts estimated and equal amounts of each construct were used in the in vitro phosphorylation assay. Total volume of the reaction was 40 µl including PKA reaction buffer (20 mM Tris–HCl, 10 mM MgCl2, pH 7.4), 200 mM ATP and purified human or bovine catalytic subunit of PKA (Sigma–Aldrich). Reaction was incubated 30 min at +30°C and was stopped by adding 20 mM PKA inhibitor H89 (Sigma–Aldrich).

Tubulin pull-down assay
Tubulin pull-down was performed with 35S-labeled in vitro translated protein, eluted GST-fusion protein or with in vitro PKA phosphorylated eluted GST-fusion protein. Equal protein amounts were used in all experiments. Purified bovine tubulin (Cytoskeleton) (70 µg) was added to tubulin polymerization buffer (80 mM PIPES, 0.5 mM MgCl, 1 mM ethylene glycol-bis, 1 mM GTP, pH 6.9, 10% glycerol, 10 mM Taxol) to a final volume of 200 µl and the microtubules were allowed to polymerize at +37°C for 30 min in the presence of different merlin constructs. After this 50 µl were removed and labeled as ‘total’ fraction and the remaining 150 µl were centrifuged at 11 503 g for 30 min at +30°C to collect the polymerized microtubules. Supernatant was removed and the pellet was resuspended in 150 µl of polymerization buffer. Twelve microlitres of each fraction were analyzed on SDS–PAGE.

Western blot
Primary Schwann cells were lysed in ice-cold ELB buffer (150 mM NaCl, 50 mM Hepes pH 7.4, 5 mM EDTA, 0.5% NP40 and complete protease inhibitor cocktail tablet, Roche) or in mild Triton-X buffer (50 mM Tris pH 7.4, 150 mM NaCl, 1 mM EDTA, 1% Triton-X 100, 1 mM PMSF) including protease inhibitors (Complete), cells were incubated on ice for 15 min and centrifuged at full speed in +4°C for 15 min. Alternatively, cells were scraped in ice-cold PBS, centrifuged briefly and the cell pellet was suspended into Urea-buffer (50 mM Tris, 6 M Urea, pH 7.4). The cells were lysed for 15 min and briefly sonicated. The protein amounts were analyzed either by the Bradford assay or by Coomassie staining. Equal amounts of proteins were run into gel transferred into nitrocellulose membrane and blotted with antibodies. As an additional loading control, loading was verified after western blot by staining the filters with Ponceau red. A representative band of the whole filter of approximately 150 kD was chosen as the loading band for Figure 5C.

Immunofluorescence, laser scanning confocal microscopy, live cell imaging and fluorescence recovery after photobleaching
U251 and Schwann cells were fixed in 3.5% PFA for 10 min, washed in PBS and permeabilized for 5 min in 0.1% Triton X-100/PBS and blocked at 5% BSA in PBS. Cells were stained with merlin and tubulin antibodies (diluted 1:100 and 1:200, respectively) followed by secondary antibodies or TO-PRO 3-iodide or DAPI nuclear stain. Double stainings were performed sequentially. Coverslips were mounted in DABCO (Sigma) and Mowiol (Calbiochem) or Vectashield (Vector Laboratories, Burlingame, CA, USA). Cells were examined by confocal microscope: U251 cells with Leica SP2 equipped with Ar and Kr lasers (Leica Microsystems, Heerbrugg, Switzerland) and Schwann cells with Zeiss LSM 510 META (Fig. 1) using the sequential scanning mode. Schwann cell images (Figs 5 and 6D) were acquired by immunofluorescence microscope (Zeiss Axiophot equipped with AxioCam cooled CCD-camera, Carl Zeiss, Esslingen, Germany).

Live cell imaging was performed in pre-warmed microscope chambers at +37°C with 20 mM Hepes as a buffering agent in the medium. Cells were plated on LabTek borosilicate no. 1.5 imaging chambers (Nunc, Naperville, IL, USA) and imaged either with Olympus inverted IX81 microscope (supplemented with 5% CO2, for the Nocodazole experiment) and CellR program or with confocal Zeiss LSM 510 META microscope. For videos, cells were imaged once every 30 s for 4 h (exposure 122 ms). FRAP analysis was performed with the Zeiss confocal microscope, pre-bleach images were obtained after which lasers were turned to full power and region of interest was bleached 50 times, the same settings were used for all FRAP experiments including the region of interest. Data analysis was performed in Zeiss META and in Microsoft Excel.

Statistical and image analysis
All statistical analyses were performed in Excel with Student's t-test using two-tailed distribution and image analysis was performed with Image J 1.36b. Images were processed with Adobe Photoshop, or with Image J 1.36b. Background was subtracted from DIC images with FFT bandpass filter and with Pseudo flat-field filter in Image J. Wide-field immunofluorescence images (Figs 5A, B and 6C, D) were deconvoluted with Huygens Deconvolution software (Scientific Volume Imaging), 30 iterations for each channel.

In vitro tubulin polymerization and other experiments
In vitro tubulin polymerization was performed on UV-permeable 96-well plates. Tubulin was purified from bovine brain as previously described (42) and recombinant merlin was produced in Sf9 insect cells as described in the GST-fusion protein section. Nocodazole 20 µM, Taxol 20 µM (Sigma) and MAP's (purified from bovine brain) were used as controls in the polymerization reaction. The polymerization buffer (80 mM PIPES, 0.5 mM MgCl, 1 mM Ethylene glycol-bis, 1 mM GTP, pH 6.9, 10% glycerol) was mixed with tubulin (25 µM) and merlin constructs (different amounts) were added on ice. At the beginning of the experiment the 96-well plate was transferred to pre-warmed 96-well plate reader that measured the OD at 350 nm every 5 min at +37°C for 60 min. The Nocodazole experiment was performed on primary Schwann cells with or without the re-expression of merlin via adeno-infection. Cells were incubated with 8 µM Nocodazole overnight after which the block was released and the cells were fixed at indicated time points. The cells were stained for tubulin and merlin.


   SUPPLEMENTARY MATERIAL
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
REFERENCES
 
Supplementary Material is available at HMG Online.


   ACKNOWLEDGEMENTS
 
We would like to thank Niclas Setterblad at the Imaging Department of the Institut Universitaire d'Hématologie IFR105 for valuable help in confocal microscopy and FRAP experiments. The imaging department is supported by grants from the Conseil Regional d'Ile-de-France and the Ministère de la Recherche. In addition we would like to thank all the personnel in Dr Giovannini's laboratory for the help with Schwann cell experiments. We also thank M. Hukkanen and M. Liljeström from Molecular Imaging Unit in Biomedicum Helsinki for their help with the image analysis, H. Ahola for her skillful technical assistance and L. Heiska for her comments and critical reading of the manuscript. This study was supported by the grants of Department of Defense DAMD17-00-0550 and W81XWH-05-1-0469, the Finnish Cancer Organizations, Helsinki Graduate School of Biotechnology and Molecular Biology, French embassy in Helsinki, Centre for International Mobility (CIMO), Emil Aaltonen Foundation, Ida Montin Foundation, Maire Taponen Foundation, Svenska kulturfonden and Biomedicum Helsinki Foundation.

Conflict of Interest statement. None declared.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 REFERENCES
 

  1. Gutmann D.H. Molecular insights into neurofibromatosis 2. Neurobiol. Dis. (1997) 3:247–261.[CrossRef][Web of Science][Medline]

  2. den Bakker M.A., Riegman P.H., Suurmeijer A.P., Vissers C.J., Sainio M., Carpen O., Zwarthoff E.C. Evidence for a cytoskeleton attachment domain at the N-terminus of the NF2 protein. J. Neurosci. Res. (2000) 62:764–771.[CrossRef][Web of Science][Medline]

  3. Xu H.M., Gutmann D.H. Merlin differentially associates with the microtubule and actin cytoskeleton. J. Neurosci. Res. (1998) 51:403–415.[CrossRef][Web of Science][Medline]

  4. Bretscher A., Chambers D., Nguyen R., Reczek D. ERM-merlin and EBP50 protein families in plasma membrane organization and function. Annu. Rev. Cell Dev. Biol. (2000) 16:113–143.[CrossRef][Web of Science][Medline]

  5. Gonzalez-Agosti C., Wiederhold T., Herndon M.E., Gusella J., Ramesh V. Interdomain interaction of merlin isoforms and its influence on intermolecular binding to NHE-RF. J. Biol. Chem. (1999) 274:34438–34442.[Abstract/Free Full Text]

  6. Grönholm M., Sainio M., Zhao F., Heiska L., Vaheri A., Carpen O. Homotypic and heterotypic interaction of the neurofibromatosis 2 tumor suppressor protein merlin and the ERM protein ezrin. J. Cell. Sci. (1999) 112:895–904.[Abstract]

  7. Nguyen R., Reczek D., Bretscher A. Hierarchy of merlin and ezrin N- and C-terminal domain interactions in homo- and heterotypic associations and their relationship to binding of scaffolding proteins EBP50 and E3KARP. J. Biol. Chem. (2001) 276:7621–7629.[Abstract/Free Full Text]

  8. Shaw R.J., Paez J.G., Curto M., Yaktine A., Pruitt W.M., Saotome I., O'Bryan J.P., Gupta V., Ratner N., Der C.J., et al. The Nf2 tumor suppressor, merlin, functions in rac-dependent signaling. Dev. Cell (2001) 1:63–72.[CrossRef][Web of Science][Medline]

  9. Alfthan K., Heiska L., Grönholm M., Renkema G.H., Carpen O. Cyclic AMP-dependent protein kinase phosphorylates merlin at serine 518 independently of P21-activated kinase and promotes merlin-ezrin heterodimerization. J. Biol. Chem. (2004) 279:18559–18566.[Abstract/Free Full Text]

  10. Fraenzer J.T., Pan H., Minimo L. Jr, Smith G.M., Knauer D., Hung G. Overexpression of the NF2 gene inhibits schwannoma cell proliferation through promoting PDGFR degradation. Int. J. Oncol. (2003) 23:1493–1500.[Web of Science][Medline]

  11. Scoles D.R., Huynh D.P., Chen M.S., Burke S.P., Gutmann D.H., Pulst S.M. The neurofibromatosis 2 tumor suppressor protein interacts with hepatocyte growth factor-regulated tyrosine kinase substrate. Hum. Mol. Genet. (2000) 9:1567–1574.[Abstract/Free Full Text]

  12. Lloyd T.E., Atkinson R., Wu M.N., Zhou Y., Pennetta G., Bellen H.J. Hrs regulates endosome membrane invagination and tyrosine kinase receptor signaling in Drosophila. Cell (2002) 108:261–269.[CrossRef][Web of Science][Medline]

  13. Maitra S., Kulikauskas R.M., Gavilan H., Fehon R.G. The tumor suppressors merlin and expanded function cooperatively to modulate receptor endocytosis and signaling. Curr. Biol. (2006) 16:702–709.[CrossRef][Web of Science][Medline]

  14. Fisk H.A., Mattison C.P., Winey M. Centrosomes and tumour suppressors. Curr. Opin. Cell Biol. (2002) 14:700–705.[CrossRef][Web of Science][Medline]

  15. Hergovich A., Lisztwan J., Barry R., Ballschmieter P., Krek W. Regulation of microtubule stability by the von hippel-lindau tumour suppressor protein pVHL. Nat. Cell Biol. (2003) 5:64–70.[CrossRef][Web of Science][Medline]

  16. Murray J.W., Wolkoff A.W. Roles of the cytoskeleton and motor proteins in endocytic sorting. Adv. Drug Deliv. Rev. (2003) 55:1385–1403.[CrossRef][Web of Science][Medline]

  17. Muranen T., Grönholm M., Renkema G.H., Carpen O. Cell cycle-dependent nucleocytoplasmic shuttling of the neurofibromatosis 2 tumour suppressor merlin. Oncogene (2005) 24:1150–1158.[CrossRef][Web of Science][Medline]

  18. Stokowski R.P., Cox D.R. Functional analysis of the neurofibromatosis type 2 protein by means of disease-causing point mutations. Am. J. Hum. Genet. (2000) 66:873–891.[CrossRef][Web of Science][Medline]

  19. Grönholm M., Muranen T., Toby G.G., Utermark T., Hanemann C.O., Golemis E.A., Carpen O. A functional association between merlin and HEI10, a cell cycle regulator. Oncogene (2006) 25:4389–4398.[CrossRef][Web of Science][Medline]

  20. Kissil J.L., Johnson K.C., Eckman M.S., Jacks T. Merlin phosphorylation by p21-activated kinase 2 and effects of phosphorylation on merlin localization. J. Biol. Chem. (2002) 277:10394–10399.[Abstract/Free Full Text]

  21. Morrison H., Sherman L.S., Legg J., Banine F., Isacke C., Haipek C.A., Gutmann D.H., Ponta H., Herrlich P. The NF2 tumor suppressor gene product, merlin, mediates contact inhibition of growth through interactions with CD44. Genes Dev. (2001) 15:968–980.[Abstract/Free Full Text]

  22. Xiao G.H., Beeser A., Chernoff J., Testa J.R. p21-activated kinase links Rac/Cdc42 signaling to merlin. J. Biol. Chem. (2002) 277:883–886.[Abstract/Free Full Text]

  23. Rong R., Surace E.I., Haipek C.A., Gutmann D.H., Ye K. Serine 518 phosphorylation modulates merlin intramolecular association and binding to critical effectors important for NF2 growth suppression. Oncogene (2004) 23:8447–8454.[CrossRef][Web of Science][Medline]

  24. Giovannini M., Robanus-Maandag E., van der Valk M., Niwa-Kawakita M., Abramowski V., Goutebroze L., Woodruff J.M., Berns A., Thomas G. Conditional biallelic Nf2 mutation in the mouse promotes manifestations of human neurofibromatosis type 2. Gene Dev. (2000) 14:1617–1630.[Abstract/Free Full Text]

  25. Utemark T., Schubert S.J., Hanemann C.O. Rearrangements of the intermediate filament GFAP in primary human schwannoma cells. Neurobiol. Dis. (2005) 19:1–9.[Medline]

  26. Ozer R.S., Halpain S. Phosphorylation-dependent localization of microtubule-associated protein MAP2c to the actin cytoskeleton. Mol. Biol. Cell (2000) 11:3573–3587.[Abstract/Free Full Text]

  27. Rubino H.M., Dammerman M., Shafit-Zagardo B., Erlichman J. Localization and characterization of the binding site for the regulatory subunit of type II cAMP-dependent protein kinase on MAP2. Neuron (1989) 3:631–638.[CrossRef][Web of Science][Medline]

  28. Grönholm M., Vossebein L., Carlson C.R., Kuja-Panula J., Teesalu T., Alfthan K., Vaheri A., Rauvala H., Herberg F.W., Tasken K., et al. Merlin links to the cAMP neuronal signaling pathway by anchoring the RIbeta subunit of protein kinase A. J. Biol. Chem. (2003) 278:41167–41172.[Abstract/Free Full Text]

  29. Lamb N.J., Cavadore J.C., Labbe J.C., Maurer R.A., Fernandez A. Inhibition of cAMP-dependent protein kinase plays a key role in the induction of mitosis and nuclear envelope breakdown in mammalian cells. EMBO J. (1991) 10:1523–1533.[Web of Science][Medline]

  30. Banerjee M., Worth D., Prowse D.M., Nikolic M. Pak1 phosphorylation on t212 affects microtubules in cells undergoing mitosis. Curr. Biol. (2002) 12:1233–1239.[CrossRef][Web of Science][Medline]

  31. Koga H., Araki N., Takeshima H., Nishi T., Hirota T., Kimura Y., Nakao M., Saya H. Impairment of cell adhesion by expression of the mutant neurofibromatosis type 2 (NF2) genes which lack exons in the ERM-homology domain. Oncogene (1998) 20:801–810.

  32. Lallemand D., Curto M., Saotome I., Giovannini M., McClatchey A.I. NF2 deficiency promotes tumorigenesis and metastasis by destabilizing adherens junctions. Genes Dev. (2003) 17:1090–1100.[Abstract/Free Full Text]

  33. Petiot A., Faure J., Stenmark H., Gruenberg J. PI3P signaling regulates receptor sorting but not transport in the endosomal pathway. J. Cell Biol. (2003) 162:971–979.[Abstract/Free Full Text]

  34. Lakadamyali M., Rust M.J., Zhuang X. Ligands for clathrin-mediated endocytosis are differentially sorted into distinct populations of early endosomes. Cell (2006) 124:997–1009.[CrossRef][Web of Science][Medline]

  35. Kita K., Wittmann T., Nathke I.S., Waterman-Storer C.M. Adenomatous polyposis coli on microtubule plus ends in cell extensions can promote microtubule net growth with or without EB1. Mol. Biol. Cell (2006) 17:2331–2345.[Abstract/Free Full Text]

  36. Lolkema M.P., Mehra N., Jorna A.S., van Beest M., Giles R.H., Voest E.E. The von Hippel-Lindau tumor suppressor protein influences microtubule dynamics at the cell periphery. Exp. Cell Res. (2004) 301:139–146.[CrossRef][Web of Science][Medline]

  37. Rosenbaum C., Kluwe L., Mautner V.F., Friedrich R.E., Muller H.W., Hanemann C.O. Isolation and characterization of schwann cells from neurofibromatosis type 2 patients. Neurobiol. Dis. (1998) 5:55–64.[CrossRef][Web of Science][Medline]

  38. Manent J., Oguievetskaia K., Bayer J., Ratner N., Giovannini M. Magnetic cell sorting for enriching schwann cells from adult mouse peripheral nerves. J. Neurosci. Methods (2003) 123:167–173.[CrossRef][Web of Science][Medline]

  39. den Bakker M.A., Riegman P.H., Hekman R.A., Boersma W., Janssen P.J., van der Kwast T.H., Zwarthoff E.C. The product of the NF2 tumour suppressor gene localizes near the plasma membrane and is highly expressed in muscle cells. Oncogene (1995) 10:757–763.[Web of Science][Medline]

  40. Lutchman M., Rouleau G.A. The neurofibromatosis type 2 gene product, schwannomin, suppresses growth of NIH 3T3 cells. Cancer Res. (1995) 55:2270–2274.[Abstract/Free Full Text]

  41. Huber L.J., Chao M.V. Mesenchymal and neuronal cell expression of the p75 neurotrophin receptor gene occur by different mechanisms. Dev. Biol. (1995) 167:227–238.[CrossRef][Web of Science][Medline]

  42. Hyman A.A., Drexel D., Kellog D., Salser S., Sawin K., Steffen P., Wordeman L., Mitchison T.J. Preparation of modified tubulins. Meth. Enzymol. (1991) 196:478–486.[Web of Science][Medline]


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 BrainHome page
C. O. Hanemann
Magic but treatable? Tumours due to loss of Merlin
Brain, March 1, 2008; 131(3): 606 - 615.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Supplementary Data
Right arrow All Versions of this Article:
16/14/1742    most recent
ddm122v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (5)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Muranen, T.
Right arrow Articles by Carpén, O.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Muranen, T.
Right arrow Articles by Carpén, O.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?