Brain Advance Access first published online on October 16, 2007
This version published online on February 14, 2008
Brain, doi:10.1093/brain/awm249
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Review Article |
Magic but treatable? Tumours due to loss of Merlin
Clinical Neurobiology, Peninsula Medical School, The John Bull Building, Tamar Science Park, Research Way, Plymouth, PL6 8BU, UK
Correspondence to:
C. O. Hanemann, Clinical Neurobiology, Peninsula Medical School, The John Bull Building, Tamar Science Park, Research Way, Plymouth, PL6 8BU, UK. E-mail: oliver.hanemann{at}pms.ac.uk
 |
Summary |
|---|
 Top Summary IntroductionMaterials and MethodsConclusionsReferences |
|---|
Alterations in the NF2 gene coding for merlin cause all tumours that occur in patients suffering from neurofibromatosis type 2, all spontaneous schwannomas and the majority of meningiomas. Thus merlin's tumours are quite frequent and also numerous when inherited as part of meurofibromatosis type 2. Tumours caused by mutations in the NF2 gene are benign and thus do not respond to classical chemotherapy. Surgery and radiosurgery are only local therapies and the patients frequently require multiple treatments. This highlights the medical need to understand how merlin loss results in tumourigenesis and the need to find new systemic therapies. The benign, and therefore genetically stable and homogenous character of the tumours allows establishment of meaningful tumour models. This brings about the rather unique opportunity to both analyse the consequences of the gene defect and identify new therapeutic targets. In this review, I will first describe the phenotypes associated with ‘merlin’ mutations and consider differential diagnosis, in particular Schwannomatosis, for which a gene defect has been described recently. Existing therapeutic options, surgery and radiosurgery, including new data on the latter will be reviewed. Finally, I will discuss how loss of merlin leads to tumourigenesis in order to understand the rationale for emerging new therapeutic targets.
Key Words: schwannoma; meningioma; ependymoma; neurofibromatosis 2; merlin
Abbreviations: HRS, hepatocyte growth factor-regulated tyrosine kinase substrate; Merlin, Moesin-ezrin-radixin-like protein; NF2, neurofibromatosis 2; PAK, P21-activated kinase
Received July 12, 2007. Revised August 22, 2007. Accepted September 17, 2007.
|
Introduction |
|---|
|
TopSummaryIntroductionMaterials and MethodsConclusionsReferences |
|---|
Mutations in the NF2 gene coding for merlin are the cause all tumours associated with autosomal dominant inherited neurofibromatosis 2 (NF2) namely schwannomas, meningiomas, retinal hamartomas and ependymomas. Both, NF2, and the related inherited disease Schwannomatosis are characterized by a huge total tumour load. While some schwannomas in Schwannomatosis seem to be caused by mutations in the INI1/SMARCB1 gene, most are caused by a mutation in the NF2 gene as in NF2. Mutations in the NF2 gene are also responsible for 50–60% of spontaneous meningiomas, a small proportion of spontaneous ependymomas and probably all spontaneous schwannomas. Tumours caused by NF2 mutations are thus frequent and genetically well described, thus it is relevant to try to understand their biology. Merlin (Moesin-ezrin-radixin-like protein), also called schwannomin, is a 69 kDa protein encoded by the NF2 gene (Rouleau, 1993
; Trofatter, 1993). ‘Merlin's’ tumours are actually caused by alterations in both gene copies and therefore loss of merlin function. I will discuss the clinical presentation and treatment of the most frequent tumours caused by merlin loss: schwannomas and meningiomas. As only a subset of spontaneous ependymomas are caused by merlin loss, and as ependymomas are also less frequent than schwannomas and meningiomas in NF2, I will discuss ependymomas only briefly. For NF2, and for Schwannomatosis, there are specific issues due to the multiplicity of these tumours which I will discuss separately. This is of particular importance as distinguishing NF2 from Schwannomatosis clinically can sometimes be difficult. Finally, I will try to paint a picture of the common underlying pathogenesis of all merlin deficient tumours nurturing the hope of useful biomarkers, and for a molecular targeted, systemic therapy.
|
Materials and Methods |
|---|
|
TopSummaryIntroductionMaterials and MethodsConclusionsReferences |
|---|
Meningiomas
It is estimated that up to 20–30% of all brain tumours are meningiomas, and the frequency in autopsies is 1.4% (Louis, 2002
). The majority of meningiomas are spontaneous and there is a link between their development and exposure to ionizing radiation, however there are rare cases of familial occurrence. Meningiomas come in different histopathological types and are grouped by the WHO into grades I-III with 90% being benign.
Mutations in both alleles of the NF2 gene, leading to loss of merlin, are the most common gene defects in meningiomas. Total 50–60% of all spontaneous meningiomas and all NF2 associated meningiomas have mutations in the NF2 gene. In some of the remaining 40–50% there is evidence for additional mechanisms of merlin inactivation, for example calpain-mediated proteolysis of merlin or aberrant methylation in the 5' region of the NF2 gene (Kimura, 1998; Lomas, 2005). ‘Merlin’ mutations are generally assumed to be an initial gene defect in meningiomas and when tumours progress to WHO grade II and III they acquire additional genetic defects. Thus there seems to be no correlation between ‘merlin’ mutation and meningioma grade. ‘Merlin’ mutations tend to be found less, however, if the meningioma is of a meningeothelial or secretory histopathological type no. (Riemenschneider, 2006; Simon, 2007).
Symptoms caused by meningiomas depend on localization and are beyond the scope of this review, they may however be asymptomatic for quite some time reviewed in (Whittle, 2004).
Total 5–15% of patients presenting with meningioma exhibit multiple meningiomas and are of particular clinical interest. Multiple meningiomas can be explained by intradural or subarachnoidal spread, or by a genetic predisposition to develop meningiomas. It is important to point out, that patients with two or more meningiomas and any NF associated sign e.g. juvenile lens opacity raises, could be diagnosed with presumptive or probable NF2 due to the diagnostic criteria of NF2 (Table 1) which has implications for genetic counselling. A population based study in Finland showed that 20% of patients with multiple meningiomas had NF2 (Antinheimo, 2000) Evans et al. (2005) reported that 8% of NF2 patients present with meningioma before developing schwannoma, however in children with multiple meningiomas the percentage that the develop additional schwannoma seems to be as high as 20%. Children should therefore be screened and followed up for other manifestations of NF2. In non-familial multiple meningiomas Heinrich et al. (2003) found that three out of seven patients had NF2 mutations in the tumour but not in the blood. The actual mutation rate in these patients may however be higher as sensitivity of mutation detection is around 90% even if tumour tissue is used. Correct diagnosis and genetic counselling can be further complicated by the fact that that adults with multiple meningiomas can be genetically mosaic for ‘merlin’ mutations. In those cases there is a slight risk of the offspring developing NF2. Identical NF2 gene mutations in different meningiomas from the same patient would support the mosaic diagnosis. It is clear however that there are patients with multiple meningiomas but without merlin loss at all and, in summary, adults with multiple non-familial meningiomas and no other signs of NF2 are usually not considered to have a high risk of NF2.
|
Meningeoangiomatosis is rare and histologically different from multiple meningiomas as it seems to involve the cortex and has more vascular involvement. However
15% of these tumours have an association with NF2 (Omeis, 2006).
Diagnosis and treatment of meningiomas caused by ‘merlin’ mutations is not different from standard treatment of meningiomas. Antinheimo et al. (1997) found more mitotic figures in NF2-associated meningiomas than in sporadic meningiomas. MR and surgery are gold standard in enlarging or symptomatic tumours, radiotherapy is recommended in malignant and unresectable or those resected incompletely (Goldsmith, 2006). Control rates with stereotactic radiosurgery are at 60–93% (Chin, 2003). Again patients with multiple meningiomas or NF2 are important to differentiate as one should be a little more reluctant with radiotherapy in patients with multiple meningiomas because one has to take into account that these patients can have a genetic predisposition to develop meningiomas and radiation therapy may evoke the second hit resulting in an increased rate of secondary malignancies. However, benign meningiomas caused by merlin loss, singular or multiple, sporadic or familial, have potentially new molecular targets in common with other tumours caused by merlin loss which is discussed at the end of the review.
Schwannomas
All schwannomas are caused by loss of merlin expression due to alterations in the NF2 gene. Schwannomas are encapsulated tumours of pure Schwann cells that do not invade the nerve. Schwannomas are quite frequent and occur as part of the hereditary tumour diseases NF2 and Schwannomatosis as well as spontaneously. It is extremely rare for a schwannoma to transform and become malignant. Schwannomas occur in different locations. Total 3–4% of all autopsies show cranial nerve schwannomas (Schneider, 1983), the vestibular nerve being the most frequent involved cranial nerve resulting tinnitus, hearing loss and verigo A UK survey revealed that 1 in 1000 will be diagnosed with VS in their lifetime (Evans, 2005). In addition to the vestibular nerve, excepting the trigeminal nerve, other cranial nerves are rarely affected. In a Finnish study (Antinheimo, 2000) 3% of the patients examined had multiple schwannomas in association with NF2, and 2% had Schwannomatosis. Schwannomas usually only cause subtle symptoms, depending on their location, and most do not cause pain (except in Schwannomatosis). Schwannomas occur on the spinal nerve roots and along peripheral nerves, with an estimated annual occurrence of 1 in 4000 (Antinheimo, 2000). Of patients with spinal schwannomas one study reported that 76% had an isolated tumour and 13% had NF2 (Evans, 1992). When occurring sporadically, spinal schwannomas are usually only diagnosed once they become symptomatic. Their presenting features are related to their location and can include muscle weakness, sensory abnormalities, occasionally localized or radicular pain and changes in bladder function. In NF2, spinal schwannomas are detected more frequently because of now routine screening with spinal MRI. Spinal tumour-related symptoms are reported by 26–33% of individuals with NF2
On peripheral nerves, schwannomas occur as fusiform swellings usually separate from the overlying skin. In NF2, tumours may also occur intracutaneously as a plaque-like lesion that is thicker and darker than the overlying skin and often more hairy (Evans, 1992). In Schwannomatosis these plaque lesions do not occur. The incidence of peripheral tumours is almost certainly underestimated, as in NF2 close to 70% of individuals have some evidence of peripheral involvement (Evans, 1992; Sperfeld, 2002), and Antinheimo et al. (2000) reported a 1 in 2000 lifetime risk of developing cutaneous schwannomas.
For reliable detection of vestibular schwannomas thin (3 mm) and overlapping MRI sections at the brainstem are now standard. The usefulness of audiological follow-up for VS has been questioned (Masuda, 2004). Modern imaging as a means to follow patients has come into sharper focus, despite the fact that, speech and pure tone audiometry are necessary to detect clinically relevant deterioration of hearing (Mautner, 2002; Slattery, 2004). A study using two-dimensional morphometric MRI analysis of vestibular schwannoma, showed a pattern of slow tumour growth which declines with patient age. However, there was some variability in the two-dimensional measurements indicating that three-dimensional volumetric analysis may be a better tool for future studies of this type (Herwadker, 2005).
As schwannomas are benign tumours that respond poorly to classical chemotherapeutic regimes, surgery and increasingly radiosurgery are the current standard therapies. A number of surgical approaches can be employed and hearing preservation is possible in selected cases. Depending on extent off preoperative symptoms, tumour size, location, surgical approach and experience of the centre, hearing loss and facial nerve involvement can be as high as 50% resp. 40% function (Samii, 1997; Ho, 2002). An alternative to surgery is stereotactic radiation treatment; local tumour control rate seems to be over 90% and importantly hearing loss and involvement of facial nerves seems to be below 5% (Rowe, 2003; Combs, 2005).
In NF2, results are slightly worse with lower hearing preservation reduced to around 68% and tumour control around 80% (Rowe, 2003; Mathieu, 2007). However, longer follow-up periods are needed on patients who have had radiosurgery to compare with surgery. These follow-up studies should also carefully analyse the possibility of new tumours and malignant transformation especially in NF2 patients after radiotherapy (Evans, 2006). Another treatment approach is to ‘watch and wait’. As schwannomas are very slow-growing tumours many patients, especially older ones never require treatment. This also calls into question the apparent ‘tumour control’ of radiation treatment in some patients. Resection of non-vestibular cranial nerve schwannomas can result in significant side effects. In general, non-vestibular schwannomas in NF2 do not grow inexorably and can be left alone.
Very few studies have addressed the management of spinal schwannomas. The clinical course of patients harbouring multiple spinal tumours is varied, with some tumours growing slowly and remaining asymptomatic, whilst others exhibit rapid growth with resultant deterioration in neurological status. Larger tumour size at presentation may be a guide to the likelihood of future progression (Dow, 2005). Thus there is variation in management of these tumours, with some authors advocating surgery for cord compression in the absence of symptoms, whilst others advocate watchful waiting except when tumours show rapid growth or symptomatic progression (Kim, 1989; Klekamp, 1998).
Ependymomas
Ependymomas account for 2–5% of all intracranial tumours and are thus the least frequent of the tumours discussed here. Approximately 29–38% of ependymomas show loss of merlin expression (Gutmann, 1997; Lamszus, 2001; Rajaram, 2005). Loss of merlin expression is probably more common in spinal ependymomas and does not significantly vary with tumour grade (Rajaram, 2005). Clear cell ependymoma, a rare histopathological variant, shows merlin loss less frequently than other ependymoma types (Fouladi, 2003). It seems clear that there are significant number of ependymomas that are caused by other gene defects and good candidates are other protein 4.1 family members (Rajaram, 2005). A recent multi-centre French study demonstrated that supratentorial ependymomas seem to be of higher grade than infratentorial. Complete resection being the gold standard, the authors also showed that in incompletely resected ependymomas of low-grade adjuvant radiotherapy prolonged the progression free survival (Metellus, 2007).
NF2
NF2 is caused by mutations in the NF2 gene on chromosome 22 and is inherited in an autosomal dominant manner. Fifty percent of cases however arise from de novo mutations, and thus no family history is apparent. NF2 has a birth incidence between 1 in 25 000 and 1 in 30 000 (Evans, 2005) and is characterized by the development of schwannomas, as well as meningiomas, ependymomas, posterior subcapsular lenticular opacities and retinal harmartomas. Schwannomas are the most frequent tumours, however 50–60% of NF2 patients will also develop meningiomas and 6% ependymomas and spinal MRI also detects spinal tumours in 67–90% of patients (Mautner, 1995; Dow, 2005).
The clinical presentation of schwannomas and meningiomas has been described previously. Importantly, these tumours are numerous in NF2. There is evidence for histological differences between NF2 related and spontaneous schwannomas. Both the multiplicity and the possibly different histology influence treatment decisions. An additional clinical feature of NF2 is the occurrence of neuropathies (Evans, 1999; Sperfeld, 2002). It is possible that tumourlets around peripheral nerves at exits in bony foramina may contribute to asymmetric peripheral neuropathy in some patients. However, haploinsufficiency of the NF2 gene might cause NF2-related polyneuropathies (Hanemann, 2007).
The type of constitutional mutation in the NF2 gene influences the disease severity with nonsense and frameshift mutations that cause protein truncation confer the worst phenotype (Parry, 1996; Ruttledge, 1996; Baser, 2004). A major implication for severity is genetic mosaicism. Total 25–30% of de novo patients are mosaic (Kluwe, 2003; Mohyuddin, 2003). In more than half of these patients the underlying mutation can only be detected by analysing multiple tumours from the same person. Although mosaic disease is relatively mild and the chance of transmission is small if the mutation is not detected in blood lymphocytes, affected offspring will contain the mutation in all their cells and be more severely affected.
Since NF2 patients also have a higher tumour recurrence rate, they need to have MRI scans of the cerebellopontine angle on a regular basis (Evans, 2005). NF2 patients also need additional MRI scans of the spinal canal initially to define the tumour load and further scans once they become symptomatic. Whole body MRI is sometimes used to estimate total tumour load in neurofibromatosis and Schwannomatosis patients, with many investigators using fat suppressed STIR sequences to detect schwannomas on whole body MRI. Individuals at 50% risk should be screened for tumours clinically and with MRI of head starting around 10–14 years, depending on the severity of the disease in the family (Evans, 2005). It seems that in children ocular manifestations are more frequent than in adults, which needs to be kept in mind when examining families (Evans, 1999). Genetic testing for mutations detects about 60% of mutations in isolated cases and 90% in families with more than one affected patient. Testing at risk individuals at around 10–14 years of age can identify the majority of non-affected offspring. Analysis of tumour material will enhance the detection rate in isolated patients by detecting the mosaic mutation (Mohyuddin, 2003). In the absence of a mutation test MRI screening of at risk patients should probably be continued until the age of 30, particularly in the later onset, mildly affected families. After 30 years, this could probably be reduced to screening once every three years for the presence of VS.
The important difference to merlin's spontaneous tumours is that they are numerous in NF2 implying difficult therapeutic decisions. The mere presence of a tumour however does not indicate it needs to be removed. Thus the decision when and which tumour to treat is an important issue in NF2 and one should try to maintain quality of life and balance that with multiple surgeries. This again underlines the need for a systemic treatment. Outcomes in NF2 are usually worse than for sporadic VS. The life span in NF2 is substantially shortened, mostly caused by meningioma disease (Baser, 2002), although many NF2 patients actually die due to swallowing problems (Slattery, WH, personal communication).
As mentioned when discussing schwannomas, radiotherapy including radiosurgery is probably less successful in NF2 than in spontaneous schwannomas. Malignant transformation occurs in a small but significant minority and tumour control rates may diminish to 60% (Baser, 2000; Rowe, 2003). This may be due to the fact that in NF2 the tumours may arise at multiple sites at the cerebellopontine angle. Thus peripheral tumour parts may receive submaximal radiation dose.
NF2 patients should be managed in specialty centres (Evans, 2005) with a permanent staff of a neurosurgeon, otolaryngologist, neurologist, geneticist, nurse and audiologist. Patients who are deaf can be offered brain stem implant (Colletti, 2006), although hearing function lags behind cochlear implants. Nonetheless, if the cochlear nerve is intact patients can be rehabilitated with a cochlear implant (Evans, 2005).
Schwannomatosis
Recently, a clinically and molecularly distinct form of hereditary disease with multiple schwannomas, Schwannaomatosis, has been described. It is important to differentiate Schwannomatosis from NF2. Schwannomatosis is usually sporadic, but families exhibiting autosomal dominant transmission do exist (Evans, 1997; MacCollin, 2005). Schwannomatosis does usually not shorten the lifespan, but quality of life can be affected by the tumour-induced pain. Patients suffering from schwannomatosis have multiple schwannomas, but do not have bilateral VS that are characteristic of NF2. Occasionally they develop meningiomas. Ependymomas and ocular abnormalities have, however, not been described in schwannomatosis. Schwannomatosis is characterized by the development of multiple peripheral nerve and spinal nerve root schwannomas but absence of plaque like intracutaneous lesions found in NF2 (clinical criteria are shown in Table 2). Unlike in NF2, motor symptoms seem to be less frequent in schwannomatosis. The frequency of schwannomatosis is not clear but was suggested to be nearly equivalent to NF2 (Antinheimo, 2000; MacCollin, 2005), however according to the Finnish study the incidence for schwannomatosis was estimated 1 in 134 000 (Antinheimo, 2000). Genetically schwannomatosis seems to be heterogeneous. A Dutch group recently found a mutation in the INI1/SAMRCB1 gene in the germ line and in the tumours of schwannomatosis patients (Hulsebos, 2007) that presumably can be found in 20% of schwannomatosis patients (Evans DG, personal communication). Some schwannomatosis patients exhibit NF2 mutations in their tumours along with loss of the wild-type allele. Although it is usually possible to distinguish schwannomatosis from NF2, mosaic NF2 can often present with clinical features consistent with schwannomatosis criteria (Mohyuddin, 2003; Murray, 2006), thus there still may be clinical overlap between schwannomatosis and neurofibromatosis. It has hence been suggested to diagnose schwannomatosis only when patients do not meet diagnostic criteria for NF2 and have no constitutional NF2 mutation (Baser, 2006; Murray, 2006). Patients with an inherited schwannomatosis disease should also have spinal axis MRI.
|
In summary, both NF2 and schwannomatosis are characterized by early onset and by the occurrence of multiple tumours. The early tumour burden and tumour load are the main therapeutic dilemmas.
The current treatment of surgery and radiosurgery are, prima vista, aiming at a single specific tumour. In patients suffering from a hereditary disease it is often difficult to decide whether to treat the tumours and, if this course is chosen, which tumour should be treated first. Furthermore, tumours in NF2 patients seem to have a reduced tumour control rate with both therapies, at least when comparing NF2 schwannomas to spontaneous schwannomas. In order to understand the rationale for emerging new therapeutic targets we need to discuss how merlin functions normally and how merlin loss may be involved in tumourigenesis. New therapeutic targets in merlin's tumours are certain growth factor receptors, the Rac/PAK signalling pathway, the Pi3K/Akt signalling pathway and the Ras/Raf/Mek signalling pathway. Luckily, some of these targets have been investigated in other cancers and therefore drugs targeting these pathways already passed phase I trials successfully. I will explain what each of these targets is and why they are promising for merlin's tumours.
The neurobiology of merlin
‘Merlin’ shows close homology to the family of ERM (ezrin, radixin, moesin) proteins, which act as membrane cytoskeletal linkers; highest homology is found in the N-terminal globular FERM domain. Merlin has also been demonstrated to act as a tumour suppressor (Gautreau, 2002). There are different splice forms of ‘merlin’, however the most prominent are isoforms 1 and 2 which differ only in the very C-terminal end. Isoform 1 consists of exons 1–15 and exon 17 coding for 595AA while isoform 2 is comprised of exons 1–16 resulting in a 590AA protein. It is thought that isoform 1 can adopt a closed conformation on dephosphorylation and then acts as a tumour suppressor, although this is not finally settled yet (Sherman, 1997).
Although merlin is expressed in all tissues affected in NF2, it is also expressed in unaffected tissues. This is, however, a frequently encountered problem in human genetics, and may be caused by other proteins taking up merlin's role in tissue not affected by merlin loss or by differential expression of relevant merlin interaction partners. More specifically, merlin is expressed in a variety of tissues during embryonic development, (Akhmametyeva, 2006), in adulthood expression has been demonstrated in the retina (Chan, 2002), lens (Claudio, 1997), testis, ovary, adrenal gland and neuronal tissue (Gutmann, 1995). There are conflicting data on expression in other tissues, however, expression in neuronal tissue includes both Schwann cells as well as neurons (Gronholm, 2005).
On a subcellular level merlin is primarily found in actin-rich cellular protrusions like the leading edge of the cell (lamellipodia and membrane ruffles) (Gonzalez-Agosti, 1996), and in cell–cell and cell–matrix contact sites co-localizing with focal adhesion and focal complex proteins (Fernandez-Valle, 2002; Lallemand, 2003). Interestingly, a Finnish group has shown that merlin also shuttles to the nucleus in a cell cycle-dependent manner (Muranen, 2005). Many merlin interaction partners have been identified in the last couple of years including transmembrane proteins, scaffolding/adapter proteins, signalling molecules/kinases, cytoskeletal proteins, various other proteins and proteins of yet unknown function [Kissil J, Merlin, UCSD Nature-Molecular-Pages (2006), doi:10.1038/mp.a001631.01; Hanemann CO in Monograph in Human Genetics, editor D. Kaufmann in press]. Until now only some of these merlin interaction partners have been examined in cells or tissues that are actually affected by ‘merlin’ mutations in humans, but nevertheless these interactions provide important insight into merlins function. For some of the interaction partners it has been shown or speculated that merlin exerts its role as a tumour suppressor through the interaction with binding partners. Importantly, some interaction partners are components of multi-molecular signalling pathways and thus part of a bigger picture discussed below.
Different in vitro models using a variety of cell lines and mouse models have been instrumental in understanding how merlin loss leads to tumourigenesis (Huynh, 1996; Sherman, 1997; Giovannini, 2000; McClatchey, 2000; Kalamarides, 2002). As the cell lines used have, however, different underlying mutations, one has to be careful to transfer results one to one to explain phenotypes of cells where the primary mutation is merlin loss. Some of the mouse models, which used a conditional knockout technique applying the flox technology reflect parts of the human disease (Giovannini, 2000; Kalamarides, 2002) others show an interesting but different phenotype (McClatchey, 1997, 1998). One explanation as to why mouse models have a slightly different phenotype from that found in human, is that some merlin interaction partners identified in different rodent cells and cell lines are not found in human Schwann cells. Another reason for the differing phenotypes between mice and men might be that the timing of the mutation in the human disease is different from that in the animal models. As it has been shown on many occasions that there is no merlin expression in tumours caused by ‘merlin’ mutations, it is generally assumed that ‘merlin’ loss or inactivation is the starting point. Mathematical modelling (Woods, 2003) and the fact that conditional knockout mice develop schwannomas late in life suggest a third hit. However, so far, no additional genetic or epigenetic events have been found in NF2 patients in tumours caused by merlin loss.
By comparing primary human schwannoma cells with normal human Schwann cells as an in vitro model, it was shown that merlin-deficient human cells show slightly increased proliferation and increased cell spreading (Pelton, 1998; Rosenbaum, 1998) reversible on the reintroduction of ‘merlin’ (Schulze, 2002), increased adhesion to extracellular matrix (Utermark, 2003), slightly decreased apoptosis (Utermark, 2005) and altered cytoskeleton (Pelton, 1998; Utermark, 2005; Flaiz, 2007). Figure 1 shows two spreading and ruffling schwannoma cells (NF2–/–) compared to an elongated slim normal Schwann cell (NF2+/+).
|
So what does all this mean in regard to possible disease mechanism?
It is impossible to mention all hypotheses on how merlin loss could lead to tumourigenesis, thus only those which have accumulated evidence through multiple publications and are of potential future therapeutic relevance will be discussed here.
Before discussing individual theories one has to consider that many theories are based on in vitro comparison of NF2–/– and NF2+/+ cells in confluent as well as subconfluent conditions, whichever reflect the in vivo situation best. In confluent cultures of NF2–/cells, ‘merlin’ expression is increased and these cells show loss of contact inhibition (Shaw, 1998; Rosenbaum, 2000; Lallemand, 2003). In subconfluent cultures NF2–/– cells, including primary human schwannoma cells, show slightly increased proliferation and increased adhesion to the extracellular matrix (Pelton, 1998; Rosenbaum, 1998; Utermark, 2003).
Regarding adhesion it is interesting to note that merlin binds to adhesion molecules such as ß1 integrin and layillin and to molecules which are part of the focal adhesion complexes such as paxillin and focal adhesion kinase (Fernandez-Valle, 2002; James, 2004; Bono, 2005). Thus it makes sense that merlin-deficient cells show deregulated adhesion to extracellular matrix, as shown in schwannoma cells (Utermark, 2003) This might explain why schwannoma cells seem not to properly myelinate axons but instead build pseudomesaxon around the extracellular matrix already described in the older electron microscopic literature (Dickersin, 1987). Additionally, merlin seems to be directly involved in cytoskeletal organization that is highly relevant in myelination. Not only does it bind actin and tubulin (Muranen, 2007) but it also inhibits the actin nucleation promotion factor N-WASP, thereby regulating actin polymerisation (Manchanda, 2005). This implies that merlin loss would lead to more actin-rich cellular protrusions, which has been clearly demonstrated in human schwannoma cells (Flaiz, 2007). The first pathway which was thought to be of potential interest as a therapeutic target is thus a signalling pathway involved in adhesion, cytoskeleton regulation and in merlin phosphorylation. P21-activated kinase (PAK), downstream of the small GTPases Rac and Cdc42, has been show to phosphorylate merlin (Sherman, 1997; Xiao, 2002; Kissil, 2003; Alfthan, 2004). In a negative feedback–loop merlin inhibits Rac and PAK activation (Shaw, 2001; Kaempchen, 2003; Kissil, 2003; Hirokawa, 2004), and inhibits Rac recruitment to the membrane (Okada, 2005). Thus when merlin is lost, Rac is activated and recruited to the membrane where it is possibly further activated by integrins, which have been shown to be overexpressed in schwannomas both in vitro and in sections (Utermark, 2003). Rac at the membrane exerts its function in cytoskeletal and adhesive structures. Supporting this hypothesis merlin loss leads to activated Rac at the membrane in human schwannoma cells (Kaempchen, 2003) and importantly this GTPase activation is non-localized and long-lasting (Nakai, 2006; Flaiz, 2007). One could therefore postulate that NF2–/–schwannoma cells are non-polarized, in contrast to the highly polarized normal Schwann cells (see also Fig. 1), and thus fail to myelinate the axon. Rac/PAK pathway inhibitors have been successfully tried in in vitro models and reversed pathological adhesion, membrane extensions and proliferation of schwannoma cells (Pelton, 1998; Hirokawa, 2004; and unpublished data), however, no pharmacological Rac1 or PAK inhibitor is yet available to the author's knowledge.
Looking at contact inhibition, NF2–/– cells show impaired cell–cell contact due to destabilised adherens junctions (Lallemand, 2003; and unpublished data). Merlin's role in stabilizing adherens junctions might involve the PDZ protein called erbin, found to bind indirectly to merlin (Rangwala, 2005). In confluent cells dephosphorylated merlin also binds the transmembrane hyaluronan receptor CD44 and acts as a growth inhibitor (Morrison, 2001). Furthermore, merlin inhibits contact inhibition by suppressing Rac recruitment to the membrane at confluency (Okada, 2005). CD44 and intergrins are also believed to act as co-receptors for growth factor tyrosine kinase receptors, further prompting the study of the signalling pathways downstream of most of these receptors, particularly the Ras–Raf–Mek–Erk and the PI3–Kinase–Akt pathway, both well known in oncogenesis (Fraenzer, 2003; Lim, 2003; Rong, 2004; Chadee, 2006; Lim, 2006; Morrison, 2007). Merlin has been shown to inhibit the Ras–Raf–Mek–Erk pathway at different levels (Lim, 2006) and merlin inhibits Ras and Rac activation upstream of the Raf–Mek–Erk after growth factor stimulation (Morrison, 2007). It is interesting that merlin is inhibiting Ras as this points out a parallel to NF1, where the gene mutated is a Ras inactivator (Ras-Gap).
In the context of growth factor signalling an interesting finding is that merlin interacts with hepatocyte growth factor-regulated tyrosine kinase substrate (HRS). HRS is a regulator of tyrosine kinase trafficking to the degradation pathway and an inhibitor of the STAT pathway. Merlin requires HRS to inhibit the STAT pathway and acts as a growth suppressor via HRS (Scoles, 2002, 2005). Thus there is accumulating evidence that merlin seem to interact with pathways downstream of growth factors.
An attractive, although yet unproven, hypothesis is that merlin is involved in the degradation of growth factors. This is supported by the fact that merlin binds ebp50 (Murthy, 1998) and erbin (Rangwala, 2005) both shown to be relevant in growth factor distribution. There is now increasing evidence that one finds overexpression of certain growth factor receptor receptors like PDGF-R, Erbb2 and Erbb3 in merlin's tumours.
|
Conclusions |
|---|
|
TopSummaryIntroductionMaterials and MethodsConclusionsReferences |
|---|
In summary, there is accumulating evidence that merlin has an important role in the coordination of two relevant and interdependent processes which are cellular adhesion and growth factor receptor response. The role of ‘merlin’ in these different pathways is roughly summarized in Fig. 2.
|
It is thus perfectly reasonable that the above mentioned pathways are targeted in merlin's tumours with compounds like herceptin, inhibitors of the Ras/Raf/Mek [e.g. Sorafenib (Bayer), PD325901 (Pfizer)] and Pi3K-Akt [e.g. OSU3013 (AstraZeneca), Rapamycin)] pathways or a PDGF-R inhibitor (e.g. Sorafenib) all of which have passed Phase I trials in other diseases.
Interestingly, an Akt inhibitor and a Ras/Raf/Mek/Erk pathway inhibitor have recently successfully been tried in a disease models for schwannoma where they clearly reduce tumour cell growth. Experiments with growth factor receptors antagonists are underway. By carefully reviewing the literature on meningiomas from the time before ‘merlin’ mutations could be detected, and therefore meningiomas without ‘merlin’ mutation were included, one finds that meningioma cells growth can be inhibited by antagonizing the IGF1 pathway (McCutcheon, 2001) and by PDGFR antagonist (Todo, 1996). Thus I believe it will not be long before some of those drugs will be in phase II trials in order to treat merlin's tumours. As these tumours are genetically defined and as there is a medical need. Merlin's tumours should be a prime target of such new drugs. However, as it seems from the experimental data that there may be more than one pathway involved in tumourigenesis following merlin loss it is likely that combination therapy will be needed to treat these tumours efficiently.
|
Acknowledgements |
|---|
I thank Peninsula Medical School and all the patients for the support of our research and John Zajicek for critical reading of the manuscript. To the comprehensive scope of this review it was impossible to mention all papers related to the topic, of which there were many. I apologize to those authors whose work has not been cited. Funding to pay the Open Access publication charges for this article was provided by The Fritz Thyssen Stiftung.
|
References |
|---|
|
TopSummaryIntroductionMaterials and MethodsConclusionsReferences |
|---|
Akhmametyeva EM, Mihaylova MM, Luo H, Kharzai S, Welling DB, Chang LS. Regulation of the neurofibromatosis 2 gene promoter expression during embryonic development. Dev Dyn (2006) 235:2771–85.[CrossRef][ISI][Medline]
Alfthan K, Heiska L, Gronholm M, Renkema GH, 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) 279:18559–66.
Antinheimo J, Haapasalo H, Haltia M, et al. Proliferation potential and histological features in neurofibromatosis 2-associated and sporadic meningiomas. J Neurosurg (1997) 87:610–4.[ISI][Medline]
Antinheimo J, Sankila R, Carpen O, Pukkala E, Sainio M, Jaeaeskelaeinen J. Population-based analysis of sporadic and type 2 neurofibromatosis-associated meningiomas and schwannomas (see comments). Neurology (2000) 54:71–6.
Baser ME, Evans DG, Jackler RK, Sujansky E, Rubenstein A. Neurofibromatosis 2, radiosurgery and malignant nervous system tumours (letter). Br J Cancer (2000) 82(998).
Baser ME, Friedman JM, Aeschliman D, et al. Predictors of the risk of mortality in neurofibromatosis 2. Am J Hum Genet (2002) 71:715–23.[CrossRef][ISI][Medline]
Baser ME, Friedman JM, Evans DG. Increasing the specificity of diagnostic criteria for schwannomatosis. Neurology (2006) 66:730–2.
Baser ME, Kuramoto L, Joe H, et al. Genotype-phenotype correlations for nervous system tumors in neurofibromatosis 2: a population-based study. Am J Hum Genet (2004) 75:231–9.[CrossRef][ISI][Medline]
Bono P, Cordero E, Johnson K, et al. Layilin, a cell surface hyaluronan receptor, interacts with merlin and radixin. Exp Cell Res (2005) 308:177–87.[CrossRef][ISI][Medline]
Chadee DN, Xu D, Hung G, et al. Mixed-lineage kinase 3 regulates B-Raf through maintenance of the B-Raf/Raf-1 complex and inhibition by the NF2 tumor suppressor protein. Proc Natl Acad Sci USA (2006) 103:4463–8.
Chan CC, Koch CA, Kaiser-Kupfer MI, et al. Loss of heterozygosity for the NF2 gene in retinal and optic nerve lesions of patients with neurofibromatosis 2. J Pathol (2002) 198:14–20.[CrossRef][ISI][Medline]
Chin LS, Szerlip NJ, Regine WF. Stereotactic radiosurgery for meningiomas. Neurosurg Focus (2003) 14(e6).
Claudio JO, Veneziale RW, Menko AS, Rouleau GA. Expression of schwannomin in lens and Schwann cells. NeuroReport (1997) 8:2025–30.[ISI][Medline]
Colletti V. Auditory outcomes in tumor vs. nontumor patients fitted with auditory brainstem implants. Adv Otorhinolaryngol (2006) 64:167–85.[Medline]
Combs SE, Volk S, Schulz-Ertner D, Huber PE, Thilmann C, Debus J. Management of acoustic neuromas with fractionated stereotactic radiotherapy (FSRT): long-term results in 106 patients treated in a single institution. Int J Radiat Oncol Biol Phys (2005) 63:75–81.[CrossRef][ISI][Medline]
Dickersin GR. The electron microscopic spectrum of nerve sheath tumors. Ultrastruct Pathol (1987) 11:103–46.[ISI][Medline]
Dow G, Biggs N, Evans G, Gillespie J, Ramsden R, King A. Spinal tumors in neurofibromatosis type 2. Is emerging knowledge of genotype predictive of natural history? J Neurosurg Spine (2005) 2:574–9.[ISI][Medline]
Evans DG, Baser ME, O’Reilly B, et al. Management of the patient and family with neurofibromatosis 2: a consensus conference statement. Br J Neurosurg (2005a) 19:5–12.[CrossRef][ISI][Medline]
Evans DG, Birch JM, Ramsden RT. Paediatric presentation of type 2 neurofibromatosis. Archives of Disease in Childhood (1999) 81:496–9.
Evans DG, Birch JM, Ramsden RT, Sharif S, Baser ME. Malignant transformation and new primary tumours after therapeutic radiation for benign disease: substantial risks in certain tumour prone syndromes. J Med Genet (2006) 43:289–94.
Evans DG, Huson SM, Donnai D, et al. A clinical study of type 2 neurofibromatosis. Q J Med (1992) 84:603–18.[ISI][Medline]
Evans DG, Mason S, Huson SM, Ponder M, Harding AE, Strachan T. Spinal and cutaneous schwannomatosis is a variant form of type 2 neurofibromatosis: a clinical and molecular study. J Neurol (1997) 4:361–6.
Evans DG, Moran A, King A, Saeed S, Gurusinghe N, Ramsden R. Incidence of vestibular schwannoma and neurofibromatosis 2 in the North West of England over a 10-year period: higher incidence than previously thought. Otol Neurotol (2005b) 26:93–7.[CrossRef][ISI][Medline]
Evans DG, Watson C, King A, Wallace AJ, Baser ME. Multiple meningiomas: differential involvement of the NF2 gene in children and adults. J Med Genet (2005c) 42:45–8.
Fernandez-Valle C, Tang Y, Ricard J, et al. Paxillin binds schwannomin and regulates its density-dependent localization and effect on cell morphology. Nat Genet (2002) 31:354–62.[ISI][Medline]
Flaiz C, Kaempchen K, Matthies C, Hanemann CO. Actin-rich protrusions and nonlocalized GTPase activation in Merlin-deficient schwannomas. J Neuropathol Exp Neurol (2007) 66:608–16.[ISI][Medline]
Fouladi M, Helton K, Dalton J, et al. Clear cell ependymoma: a clinicopathologic and radiographic analysis of 10 patients. Cancer (2003) 98:2232–44.[CrossRef][ISI][Medline]
Fraenzer JT, Pan H, Minimo L. Jr, Smith GM, Knauer D, Hung G. Overexpression of the NF2 gene inhibits schwannoma cell proliferation through promoting PDGFR degradation. Int J Oncol (2003) 23:1493–500.[ISI][Medline]
Gautreau A, Louvard D, Arpin M. ERM proteins and NF2 tumor suppressor: the Yin and Yang of cortical actin organization and cell growth signaling. Curr Opin Cell Biol (2002) 14:104–9.[CrossRef][ISI][Medline]
Giovannini M, Robanus-Maandag E, vander V, et al. Conditional biallelic Nf2 mutation in the mouse promotes manifestations of human neurofibromatosis type 2. Genes Dev (2000) 14:1617–30.
Goldsmith B, McDermott MW. Meningioma. Neurosurg Clin N Am (2006) 17:111–20. vi.[CrossRef][ISI][Medline]
Gonzalez-Agosti C, Xu L, Pinney D, et al. The merlin tumor suppressor localizes preferentially in membrane ruffles. Oncogene (1996) 13:1239–47.[ISI][Medline]
Gronholm M, Teesalu T, Tyynela J, et al. Characterization of the NF2 protein merlin and the ERM protein ezrin in human, rat, and mouse central nervous system. Mol Cell Neurosci (2005) 28:683–93.[CrossRef][ISI][Medline]
Gutmann DH, Giordano MJ, Fishback AS, Guha A. Loss of merlin expression in sporadic meningiomas, ependymomas and schwannomas. Neurology (1997) 49:267–70.
Gutmann DH, Wright DE, Geist RT, Snider WD. Expression of the neurofibromatosis 2 (NF2) gene isoforms during rat embryonic development. Hum Mol Genet (1995) 4:471–8.
Hanemann CO, Diebold R, Kaufmann D. Role of NF2 Haploinsufficiency in NF2-associated Polyneuropathy. Brain Pathol (2007) doi: 10.1111/j.1750-3639.2007.00086.x.
Heinrich B, Hartmann C, Stemmer-Rachamimov AO, Louis DN, MacCollin M. Multiple meningiomas: investigating the molecular basis of sporadic and familial forms. Int J Cancer (2003) 103:483–8.[CrossRef][ISI][Medline]
Herwadker A, Vokurka EA, Evans DG, Ramsden RT, Jackson A. Size and growth rate of sporadic vestibular schwannoma: predictive value of information available at presentation. Otol Neurotol (2005) 26:86–92.[CrossRef][ISI][Medline]
Hirokawa Y, Tikoo A, Huynh J, et al. A clue to the therapy of neurofibromatosis type 2: NF2/merlin is a PAK1 inhibitor. Cancer J (2004) 10:20–26.[ISI][Medline]
Ho SY, Kveton JF. Acoustic neuroma. Assessment and management. Otolaryngol Clin North Am (2002) 35:393–404. viii.[CrossRef][ISI][Medline]
Hulsebos TJ, Plomp AS, Wolterman RA, Robanus-Maandag EC, Baas F, Wesseling P. Germline Mutation of INI1/SMARCB1 in Familial Schwannomatosis. Am J Hum Genet (2007) 80:805–810.[CrossRef][ISI][Medline]
Huynh MD, Pulst SM. Neurofibromatosis 2 antisense oligonucleotides induce reversible inhibition of schwannomin synthesis and cell adhesion in STS26T and T98G cells. Oncogene (1996) 13:73–84.[ISI][Medline]
James MF, Beauchamp RL, Manchanda N, Kazlauskas A, Ramesh V. A NHERF binding site links the betaPDGFR to the cytoskeleton and regulates cell spreading and migration. J Cell Sci (2004) 117:2951–2961.
Kaempchen K, Mielke K, Utermark T, Langmesser S, Hanemann CO. Upregulation of the Rac1/JNK signaling pathway in primary human schwannoma cells. Hum Mol Genet (2003) 12:1211–1221.
Kalamarides M, Niwa-Kawakita M, Leblois H, et al. Nf2 gene inactivation in arachnoidal cells is rate-limiting for meningioma development in the mouse. Genes Dev (2002) 16:1060–5.
Kim P, Ebersold MJ, Onofrio BM, Quast LM. Surgery of spinal nerve schwannoma. Risk of neurological deficit after resection of involved root. J Neurosurg (1989) 71:810–4.[ISI][Medline]
Kimura Y, Koga H, Araki N, et al. The involvement of calpain-dependent proteolysis of the tumor suppressor NF2 (merlin) in schwannomas and meningiomas. Nat Med (1998) 4:915–22.[CrossRef][ISI][Medline]
Kissil JL, Wilker EW, Johnson KC, Eckman MS, Yaffe MB, Jacks T. Merlin, the product of the Nf2 tumor suppressor gene, is an inhibitor of the p21-activated kinase, Pak1. Mol Cell (2003) 12:841–9.[CrossRef][ISI][Medline]
Klekamp J, Samii M. Surgery of spinal nerve sheath tumors with special reference to neurofibromatosis. Neurosurgery (1998) 42:279–89.[CrossRef][ISI][Medline]
Kluwe L, Mautner V, Heinrich B, et al. Molecular study of frequency of mosaicism in neurofibromatosis 2 patients with bilateral vestibular schwannomas. J Med Genet (2003) 40:109–14.
Lallemand D, Curto M, Saotome I, Giovannini M, McClatchey AI. NF2 deficiency promotes tumorigenesis and metastasis by destabilizing adherens junctions. Genes Dev (2003) 1090–100.
Lamszus K, Lachenmayer L, Heinemann U, et al. Molecular genetic alterations on chromosomes 11 and 22 in ependymomas. Int J Cancer (2001) 91:803–8.[CrossRef][ISI][Medline]
Lim JY, Kim H, Jeun SS, Kang SG, Lee KJ. Merlin inhibits growth hormone-regulated Raf-ERKs pathways by binding to Grb2 protein. Biochem Biophys Res Commun (2006) 340:1151–7.[CrossRef][ISI][Medline]
Lim JY, Kim H, Kim YH, et al. Merlin suppresses the SRE-dependent transcription by inhibiting the activation of Ras-ERK pathway. Biochem Biophys Res Commun (2003) 302:238–45.[CrossRef][ISI][Medline]
Lomas J, Bello MJ, Arjona D, et al. Genetic and epigenetic alteration of the NF2 gene in sporadic meningiomas. Genes Chromosomes Cancer (2005) 42:314–9.[CrossRef][ISI][Medline]
Louis DN, Pomeroy SL, Cairncross JG. Focus on central nervous system neoplasia. Cancer Cell (2002) 1:125–8.[CrossRef][ISI][Medline]
MacCollin M, Chiocca EA, Evans DG, et al. Diagnostic criteria for schwannomatosis. Neurology (2005) 64:1838–45.
Manchanda N, Lyubimova A, Ho HY, et al. The NF2 tumor suppressor Merlin and the ERM proteins interact with N-WASP and regulate its actin polymerization function. J Biol Chem (2005) 280:12517–22.
Masuda A, Fisher LM, Oppenheimer ML, Iqbal Z, Slattery WH. Hearing changes after diagnosis in neurofibromatosis type 2. Otol Neurotol (2004) 25:150–4.[CrossRef][ISI][Medline]
Mathieu D, Kondziolka D, Flickinger JC, et al. Stereotactic radiosurgery for vestibular schwannomas in patients with neurofibromatosis type 2: an analysis of tumor control, complications, and hearing preservation rates. Neurosurgery (2007) 60:460–8.[ISI][Medline]
Mautner VF, Baser ME, Thakkar SD, Feigen UM, Friedman JM, Kluwe L. Vestibular schwannoma growth in patients with neurofibromatosis Type 2: a longitudinal study. J Neurosurg (2002) 96:223–8.[ISI][Medline]
Mautner VF, Tatagiba M, Lindenau M, et al. Spinal tumors in patients with neurofibromatosis type 2: MR imaging study of frequency, multiplicity, and variety. AJR Am J Roentgenol (1995) 165:951–5.
McClatchey AI. Neurofibromatosis type II: mouse models reveal broad roles in tumorigenesis and metastasis. Mol Med Today (2000) 6:252–3.[CrossRef][ISI][Medline]
McClatchey AI, Saotome I, Mercer K, et al. Mice heterozygous for a mutation at the Nf2 tumor suppressor locus develop a range of highly metastatic tumors. Genes Dev (1998) 12:1121–33.
McClatchey AI, Saotome I, Ramesh V, Gusella JF, Jacks T. The Nf2 tumor suppressor gene product is essential for extraembryonic development immediately prior to gastrulation. Genes Dev (1997) 11:1253–65.
McCutcheon IE, Flyvbjerg A, Hill H, et al. Antitumor activity of the growth hormone receptor antagonist pegvisomant against human meningiomas in nude mice. J Neurosurg (2001) 94:487–492.[ISI][Medline]
Metellus P, Barrie M, Figarella-Branger D, et al. Multicentric French study on adult intracranial ependymomas: prognostic factors analysis and therapeutic considerations from a cohort of 152 patients. Brain (2007) 130:1338–49.
Mohyuddin A, Baser ME, Watson C, et al. Somatic mosaicism in neurofibromatosis 2: prevalence and risk of disease transmission to offspring. J Med Genet (2003) 40:459–63.
Morrison H, Sherman LS, Legg J, et al. The NF2 tumor suppressor gene product, merlin, mediates contact inhibition of growth through interactions with CD44. Genes Dev (2001) 15:968–980.
Morrison H, Sperka T, Manent J, Giovannini M, Ponta H, Herrlich P. Merlin/neurofibromatosis type 2 suppresses growth by inhibiting the activation of Ras and Rac. Cancer Res (2007) 67:520–7.
Muranen T, Gronholm M, Lampin A, et al. The tumor suppressor merlin interacts with microtubules and modulates Schwann cell microtubule cytoskeleton. Hum Mol Genet (2007) 16:1742–51.
Muranen T, Gronholm M, Renkema GH, Carpen O. Cell cycle-dependent nucleocytoplasmic shuttling of the neurofibromatosis 2 tumour suppressor merlin. Oncogene (2005) 24:1150–58.[CrossRef][ISI][Medline]
Murray AJ, Hughes TA, Neal JW, Howard E, Evans DG, Harper PS. A case of multiple cutaneous schwannomas; schwannomatosis or neurofibromatosis type 2? J Neurol Neurosurg Psychiatry (2006) 77:269–71.
Murthy A, Gonzalez-Agosti C, Cordero E, et al. NHE-RF, a regulatory cofactor for Na(+)-H+ exchange, is a common interactor for merlin and ERM (MERM) proteins. J Biol Chem (1998) 273:1273–6.
Nakai Y, Zheng Y, MacCollin M, Ratner N. Temporal control of Rac in Schwann cell-axon interaction is disrupted in NF2-mutant schwannoma cells. J Neurosci (2006) 26:3390–95.
Okada T, Lopez-Lago M, Giancotti FG. Merlin/NF-2 mediates contact inhibition of growth by suppressing recruitment of Rac to the plasma membrane. J Cell Biol (2005) 171:361–71.
Omeis I, Hillard VH, Braun A, Benzil DL, Murali R, Harter DH. Meningioangiomatosis associated with neurofibromatosis: report of 2 cases in a single family and review of the literature. Surg Neurol (2006) 65:595–603.[CrossRef][ISI][Medline]
Parry DM, MacCollin MM, Kaiser-Kupfer MI, et al. Germ-line mutations in the neurofibromatosis 2 gene: correlations with disease severity and retinal abnormalities. Am J Hum Genet (1996) 59:529–539.[ISI][Medline]
Pelton PD, Sherman LS, Rizvi TA, et al. Ruffling membrane, stress fiber, cell spreading and proliferation abnormalities in human Schwannoma cells. Oncogene (1998) 17:2195–209.[CrossRef][ISI][Medline]
Rajaram V, Gutmann DH, Prasad SK, Mansur DB, Perry A. Alterations of protein 4.1 family members in ependymomas: a study of 84 cases. Mod Pathol (2005) 18:991–7.[CrossRef][ISI][Medline]
Rangwala R, Banine F, Borg JP, Sherman LS. Erbin regulates mitogen-activated protein (MAP) kinase activation and MAP kinase-dependent interactions between Merlin and adherens junction protein complexes in Schwann cells. J Biol Chem (2005) 280:11790–7.
Riemenschneider MJ, Perry A, Reifenberger G. Histological classification and molecular genetics of meningiomas. Lancet Neurol (2006) 5:1045–1054.[CrossRef][ISI][Medline]
Rong R, Tang X, Gutmann DH, Ye K. Neurofibromatosis 2 (NF2) tumor suppressor merlin inhibits phosphatidylinositol 3-kinase through binding to PIKE-L. Proc Natl Acad Sci USA (2004) 101:18200–5.
Rosenbaum C, Kamleiter M, Grafe P, et al. Enhanced proliferation and potassium conductance of Schwann cells isolated from NF2 schwannomas can be reduced by quinidine. Neurobiol Dis (2000) 7:483–91.[CrossRef][ISI][Medline]
Rosenbaum C, Kluwe L, Mautner VF, Friedrich RE, Mueller HW, Hanemann CO. Isolation and characterization of Schwann cells from neurofibromatosis type 2 patients. Neurobiol Dis (1998) 5:55–64.[CrossRef][ISI][Medline]
Rouleau GA, Merel P, Lutchman M, et al. Alteration in a new gene encoding a putative membrane-organizing protein causes neuro-fibromatosis type 2 (see comments). Nature (1993) 363:515–21.[CrossRef][Medline]
Rowe JG, Radatz MW, Walton L, Soanes T, Rodgers J, Kemeny AA. Clinical experience with gamma knife stereotactic radiosurgery in the management of vestibular schwannomas secondary to type 2 neurofibromatosis. J Neurol Neurosurg Psychiatry (2003) 74:1288–93.
Ruttledge MH, Andermann AA, Phelan CM, et al. Type of mutation in the neurofibromatosis type 2 gene (NF2) frequently determines severity of disease. American Journal of Human Genetics (1996) 59:331–342.