A Systematic Screen for Dominant Second-Site Modifiers of Merlin/NF2 Phenotypes Reveals an Interaction With blistered/DSRF and scribbler

Corresponding author: Richard G. Fehon, B333 LSRC, Research Dr., Duke University, Durham, NC 27708-1000., rfehon{at}duke.edu (E-mail)

Communicating editor: K. ANDERSON

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Merlin, the Drosophila homologue of the human tumor suppressor gene Neurofibromatosis 2 (NF2), is required for the regulation of cell proliferation and differentiation. To better understand the cellular functions of the NF2 gene product, Merlin, recent work has concentrated on identifying proteins with which it interacts either physically or functionally. In this article, we describe genetic screens designed to isolate second-site modifiers of Merlin phenotypes from which we have identified five multiallelic complementation groups that modify both loss-of-function and dominant-negative Merlin phenotypes. Three of these groups, Group IIa/scribbler (also known as brakeless), Group IIc/blistered, and Group IId/net, are known genes, while two appear to be novel. In addition, two genes, Group IIa/scribbler and Group IIc/blistered, alter Merlin subcellular localization in epithelial and neuronal tissues, suggesting that they regulate Merlin trafficking or function. Furthermore, we show that mutations in scribbler and blistered display second-site noncomplementation with one another. These results suggest that Merlin, blistered, and scribbler function together in a common pathway to regulate Drosophila wing epithelial development.

NEUROFIBROMATOSIS type 2 (NF2) is a dominant autosomal disorder characterized by benign slow growing tumors associated with the glial cells of the eighth cranial nerve and other glial cells throughout the central nervous system (MARTUZA and ELDRIDGE 1988 ). The product of the NF2 tumor-suppressor gene is a protein called Merlin (ROULEAU et al. 1993 ; TROFATTER et al. 1993 ). Merlin is a novel member of the 4.1 superfamily and has the greatest similarity to the Ezrin, Radixin, and Moesin (ERM) proteins. ERM proteins are membrane/cytoskeletal adapters that link a variety of transmembrane proteins to the underlying actin cytoskeleton (ALGRAIN et al. 1993 ). ERM proteins have two functional domains: the N-terminal FERM domain (CHISHTI et al. 1998 ) interacts with transmembrane proteins, and the C-terminal domain binds filamentous actin. ERM proteins are believed to regulate processes such as signal transduction by organizing the plasma membrane into distinct functional domains (HELANDER et al. 1996 ). Consistent with this notion, ERM proteins have been shown to directly or indirectly regulate the localization and/or activity of several transmembrane proteins such as CD44, ICAM-2, and the ß-adrenergic receptor (TSUKITA et al. 1994 ; HELANDER et al. 1996 ; HEISKA et al. 1998 ). Interestingly, Merlin has been demonstrated to interact physically with CD44 (SAINIO et al. 1997 ) and a recent study indicates that this interaction may play an important role in growth regulation (HERRLICH et al. 2000 ).

The activity of ERM proteins is tightly regulated by binding PIP₂ and via a C-terminal phosphorylation event (MATSUI et al. 1999 ). Likewise, Merlin exists at the plasma membrane in at least two forms, transiting from an inactive to an active state by an unknown mechanism (LAJEUNESSE et al. 1998 ). Merlin expression level and phosphorylation are responsive to changes in cell adhesion, cell confluency, and growth factor stimulation, suggesting that Merlin activity is precisely regulated, perhaps through intercellular signaling mechanisms (SHAW et al. 1998 ). Merlin has been shown to form homotypic dimers and heterotypic dimers with other ERM proteins (GRONHOLM et al. 1999 ). In addition, it has been shown to bind actin filaments (XU and GUTTMAN 1998 ), ßII spectrin (SCOLES et al. 1998 ), and the EBP50/ Na⁺/H⁺ exchanger regulatory factor that also binds ERM proteins (RECZEK et al. 1997 ; MURTHY et al. 1998 ). Recently, another Drosophila protein 4.1 family member, expanded, was shown to interact genetically and physically with Merlin to regulate cellular proliferation and differentiation of imaginal disc tissue (MCCARTNEY et al. 2000 ). However, the significance of these interactions is unclear and the mechanisms by which Merlin function is regulated remain unknown.

To identify genes involved in Merlin function, we performed a genetic screen designed to identify dominant second-site modifiers of Drosophila Merlin phenotypes. Second-site modifier screens are powerful tools for dissecting pathways associated with specific cellular and developmental processes, and have been successful in identifying genes that interact functionally as well as physically with the target (SIMON et al. 1991 ; REBAY et al. 2000 ). Furthermore, the identification of genetic modifiers should be useful for understanding the NF2 disorder, as several studies have suggested a role for second-site genetic modifiers in the expressivity and penetrance of NF2-related phenotypes (MCCLATCHEY et al. 1998 ; BRUDER et al. 1999A , BRUDER et al. 1999B ). Through our genetic screens, we identified 29 modifying mutations out of 100,000 progeny derived from mutagenized parents, 23 of which fall into five multiallelic complementation groups. Three of the complementation groups contain new alleles of previously characterized loci: blistered, the Drosophila homologue of serum response factor (GUILLEMIN et al. 1996 ; MONTAGNE et al. 1996 ), the extra vein gene net, and a newly identified gene called scribbler (sbb) or brakeless, which encodes a novel nuclear protein of unknown function (RAO et al. 2000 ; SENTI et al. 2000 ; YANG et al. 2000 ). The remaining two multiallelic groups and the five single-allele groups are novel. Mutations in all five groups dominantly modified phenotypes caused by ectopic expression of a dominant negative Merlin allele and phenotypes associated with a recessive hypomorphic Merlin allele. In addition, alteration of Merlin subcellular localization was observed in tissues from larvae homozygous for two of the modifying loci. These results together with observations of second-site noncomplementation between some of the groups suggest that all function together with Merlin to regulate cell proliferation. Thus, characterization of all of these genes will provide further insight into mechanisms of Merlin function and the NF2 disorder.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Drosophila cultures and stocks used:
All Drosophila cultures were maintained on standard cornmeal, yeast, molasses, agar medium and all crosses were performed at 25° unless otherwise specified. Meiotic mapping of the complementation groups was based on lethality and used the following P-element insertion stocks: second chromosome mapping stocks, P{w^+mC = lacW}l(2)10424^k06801 at [2-18]/26B1-2; P{w^+mC = lacW}Pen^k14401 at [2-40.5]/31A; P{w^+mW.hs = GawB}ap^md544 at [2-55.2]/41F9-10; P{w^+mW.hs = GawB}559.1 at [2-59]/44D2-5; P{w^+mClacW}AA48 at [2-87]/56A1-2; and P{w^+mW.hs = GawB}Dll^md23 at [2-107]/60E1-2; third chromosome mapping stocks, P{w^+mC = lacW}l(3)L1170^L1170 at [3-0]/61C7-8; P{w^+mC = lacW}l(3)j2B9^j2B9 at [3-28]/67B4-5; P{w^+mC = lacW}Trl^s2325 at [3-36]/70F1-4; 5P{w^+mC = lacW}l(3)j1E6^j1E6 at [3-46]/82A3-5; and P{w^+mC = lacW}dco^j3B9 at [3-102]/100B2-4. The Gal4/UAS system (BRAND and PERRIMON 1993 ) was used to overexpress Mer^BB. A second chromosome insert of UASMer^BB (LAJEUNESSE et al. 1998 ) was recombined with either engrailed::Gal4 or apterous::Gal4 enhancer traps lines to generate the enBB and apBB chromosomes, respectively. Mer³ was previously described (FEHON et al. 1997 ; LAJEUNESSE et al. 1998 ; MCCARTNEY et al. 2000 ) All deficiencies were from the Bloomington deficiency kit collection (FlyBase).

Screen protocol:
Three-day-old white-eyed males from an isogenized second and third chromosome (marked with ebony, e⁴) stock were treated either with 25 mM EMS or 4000 rads of -ray radiation and mated to virgin enBB/CyO w⁺ females. In the F₁ generation, male flies having a modified phenotype were collected and backcrossed to enBB/CyO w⁺ females for two further generations to eliminate any somatic mosaics that arise during mutagenesis. In the F₃ generation mutations that still modified enBB phenotypes were mapped to a chromosome via the segregation of modification. Stocks were generated at this step by crossing mutations that mapped to the second chromosome to Sco/SM6a and those that mapped to the third to MRKS/TM6a. We established complementation groups by taking mutations that mapped to the same chromosome and testing for lethal noncomplementation.

Wing measurements:
All crosses for flies used in wing measurements were maintained in the same incubator at 25° and collected simultaneously to eliminate phenotypic variation due to environmental factors. Flies of the appropriate genotype were incubated in 70% ethanol for at least 24 hr. We have found that this makes the cuticles easier to manipulate. Wings were removed in a drop of water on a siliconized slide and mounted in a drop of Aquamount on a glass slide. Only wings that had been flattened during the mounting process were used for further analysis. Images were captured using a Zeiss (Thornwood, NY) Axioplan microscope equipped with a Sony DCX-760MD camera and imported into Adobe Photoshop. Using the free draw tool the area to be calculated was outlined, filled in, and analyzed using the MEASURE tool of NIH Image. For the enBB experiments, the area between wing vein II and the posterior margin was calculated, and for the Mer³ experiments the area of the entire wing was measured.

Sequencing and molecular characterization of brakeless alleles:
Homozygous third instar larvae from Group IIa/sbb²⁵⁶ and Group IIa/sbb³²⁴ were collected and their genomic DNA were extracted using a standard protocol (Berkeley Drosophila Genome Project). The genomic region containing the entire sbb coding sequence (8 kb) was amplified using intronic primers in four separate PCR reactions. These reaction products were sequenced using nested internal primers and the ABI-Prism Big Dye protocol. Sequences were then assembled and analyzed using the Sequencher program (Gene Codes Corporation, Ann Arbor, MI).

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Deficiency kit screen results:
To approximate the number of loci that modify Merlin phenotypes in a dose-sensitive fashion we initially screened a collection of deficiencies, the so-called "deficiency kit" (FLYBASE 1999 ), that represent 70% of the genome in haploids. Deficiencies were tested for dominant effects on phenotypes displayed by a hypomorphic allele of Merlin, Mer³, and by ectopic expression of a dominant-negative form of Merlin, Mer^BB (LAJEUNESSE et al. 1998 ; MCCARTNEY et al. 2000 ).

Mer³ flies are semiviable and display a variety of phenotypes in the head, eye, wings, and legs (MCCARTNEY et al. 2000 ). Wings from Mer³ hemizygous male flies are broadened and have a low penetrance of disruptions of the posterior cross vein (data not shown). In the head, Mer³ flies express slightly rough, smaller eyes (Fig 1B) and 10% have ectopic growths and vibrissae almost exclusively in the anterior ventral portion of the eyes. Twenty-three deficiencies were found to modify Mer³ phenotypes (Table 1). Two of these deficiencies were strong interactors, Df(2L)C144 and In(2R)bw^VDe2L. Mer³ flies heterozygous for Df(2L)C144 or In(2R)bw^VDe2L had head defects and small rough eyes (Fig 1, Fig C and Fig D). The chromosomal region encompassing Df(2L)C144 has been saturated for lethal mutations (LITTLETON and BELLEN 1994 ) but none of the lethal complementation groups uncovered by this deficiency modify the Mer³ phenotype (data not shown). This result suggests that more than a single mutation within this deficiency may be responsible for the dose-sensitive modification phenotype. Alternatively, the gene responsible for this effect is not mutable to lethality, or the Df(2L)C144 chromosome carries an independent mutation that interacts genetically with Mer³.

View larger version (167K): In this window In a new window Download PPT slide   Figure 1. Deficiencies that modify Mer3 phenotypes. (A) Scanning electron micrograph (SEM) of a wild-type adult male eye. (B) SEM of a Mer3/Y eye. Note the slightly smaller and rougher appearance compared to the wild-type eye. (C and D) Adult eyes from Mer3/Y; In(2R)bwVDe2L CyR/+ and Mer3/Y; Df(2L)C144/+ flies, respectively. In both cases, the presence of the deficiency results in an enhancement of Mer3 phenotypes including reduction in the size of the eye and formation of aberrant head cuticle, bristles, and outgrowths. Histological examination of sections taken through these eyes reveals very minor perturbation of ommatidial organization similar to those seen in Mer3/Y eyes alone (data not shown).

Expression of dominant-negative Mer^BB in the developing wing results in overproliferation of the wing blade (LAJEUNESSE et al. 1998 ). This dominant-negative form of Merlin has seven conserved amino acids within the FERM domain removed, and it interferes with the activation of wild-type Merlin (LAJEUNESSE et al. 1998 ). A chromosome carrying both the UAS::Mer^BB transgene and the engrailed::Gal4 driver (denoted enBB) displays a phenotype that is sensitive to gene dose (Fig 2), an essential feature for screens designed to identify extragenic dose-sensitive modifiers (SIMON et al. 1991 ; REBAY et al. 2000 ). Flies heterozygous for enBB have moderately overgrown posterior wing compartments with no disruption in venation (Fig 2B). Homozygosity for enBB results in enlargement of the posterior wing compartment with disruptions in venation, particularly along vein V and the posterior cross vein (Fig 2C). Flies homozygous for enBB also hold their wings out from the body axis (data not shown).

View larger version (61K): In this window In a new window Download PPT slide   Figure 2. The enBB phenotype is dose sensitive. (A) Wild-type female wing; (B) wing from female heterozygous for enBB/+. Note the folds in the posterior compartment (arrows) and the slight disruption of the anterior cross vein. (C) Wing from a female homozygous for enBB. Note the increase in posterior wing compartment in comparison to the heterozygous enBB flies in B, the disruption of the posterior cross vein, and the ectopic material along vein V and the loss of anterior cross vein. (D) Screen protocol: males carrying isogenized second and third chromosomes were treated with mutagen (either EMS or rays) and mated to virgin enBB/CyO w+ females. In the F1 generation, male flies having a modified phenotype were collected and backcrossed to enBB/CyO w+ females for two further generations to eliminate any somatic mosaics that arise during mutagenesis. Flies that displayed modification phenotypes in the F3 generation were established as stocks and the chromosomal location of the modifying mutation was mapped.

Using the deficiency kit, we identified 20 interacting deficiencies that enhance the phenotype caused by heterozygosity for the enBB chromosome (Table 1). The degree of enhancement ranged from slight (30–50% of wings showing ectopic vein material) to very strong (100% of wings showing ectopic vein material with some wing blistering). The strongest interacting deficiency, Df(2R)Px2, expressed a dominant extra vein phenotype in a wild-type Merlin background. However, heterozygosity for both Df(2R)Px2 and enBB resulted in a blistered wing phenotype not observed with either alone, suggesting the presence of an interaction. For all deficiencies except Df(3R)Antp17 and Df(3R)awd^krb, the observed interacting cytological regions were substantiated and defined by overlapping deficiencies that displayed the same enhancement phenotype. However, this does not exclude the possibility of multiple interacting genes within these regions or the presence of additional modifying loci on deficiency-bearing chromosomes.

In summary, between the two screens we identified eight deficiencies that modified both Mer³ and Mer^BB. However, neither analysis of lethal P-element insertions nor analysis of previously characterized genes uncovered by these deficiencies revealed the individual loci responsible for the modification of either Merlin phenotype (data not shown).

F₁ second-site modifier screen results:
As a complement to the deficiency kit screen, we performed a genetic screen looking for dose-sensitive modification of the phenotypes expressed by flies carrying the enBB chromosome. The design of this screen is shown in Fig 2D. To identify modifiers of the enBB phenotype, 100,000 F₁ male flies expressing the enBB transgene (75,000 from EMS-mutagenized flies and 25,000 from X-ray-mutagenized flies) and carrying potential modifiers were examined for ectopic venation along the vein V/posterior cross vein intersection and/or the presence of outheld wings. From this screen, we identified 29 enhancer mutations and no suppressor mutations. Twenty-three of the enhancer mutations fell into five allelic complementation groups on the basis of lethality (Table 2). Four of these complementation groups were on the second chromosome and one was on the third chromosome.

Examples of the modified enBB phenotypes that we observed in the screen are shown in Fig 3 (middle column). In all cases, an increase in ectopic venation at the vein V/posterior cross vein intersection was observed, although the amount of material varied between groups and within each group depending on allele strength. We kept only those mutations in which at least 50% of the wings expressed a modification of enBB phenotypes. Mutations that displayed dominant phenotypes in the absence of the enBB chromosome were discarded. To show that these mutations also affect overgrowth, we compared the area in the posterior compartment (between vein III and the posterior margin) of enBB/modifier wings to enBB/+ wings (Table 3). Several members of each complementation group were analyzed and in each case there was an increase in size of enBB wings with the presence of a modifying mutation when compared to outcrossed enBB wings.

View larger version (141K): In this window In a new window Download PPT slide   Figure 3. Examples of modifier wing phenotypes. (A) Wild-type wing. First column (D, G, J, M, and P), female wings heterozygous for mutations identified in the screen. Note that all mutations are recessive except for P, which has an abrupt vein V (arrow). Second column, modification of enBB heterozygous phenotype (B) by modifiers (E, H, K, N, and Q). Third column, modification of apBB (C) phenotypes showing that the modifications are not promoter specific. All wings are from female flies. The genotypes are as follows: (A) wild type; (B) enBB/+; (C) apBB/+; (D) Group IIc-bs242/+; (E) enBB/Group IIc-bs242; (F) apBB/Group IIc-bs242; (G) Group IIa256/+; (H) enBB/Group IIa256; (I) apBB/Group IIa256; (J) Group IId-net383/+; (K) enBB/Group IId-net383; (L) apBB/Group IId-net383; (M) Group IIb187/+; (N) enBB/Group IIb187; (O) apBB/Group IIb187; (P) Group IIIa239/+; (Q) enBB Group IIIa239; and (R) apBB/Group IIIa239.

View this table: In this window In a new window   Table 3. Enhancement of MerBB overproliferation wing phenotype

Genetic tests to identify relevant modifiers:
To further characterize these modifying mutations, several tests were designed to distinguish between mutations relevant for understanding the mechanisms of Merlin function and those mutations that are uninformative or misleading. For instance, mutations that affected the expression of the engrailed::GAL4 driver would indirectly influence the dominant Mer^BB phenotype and would be of little interest. To eliminate mutations that have dominant transcriptional effect on engrailed expression and to demonstrate the direct effect of a modifier on Mer^BB activity, we tested each candidate's ability to modify phenotypes displayed by apBB flies. In apBB flies, the apterous::Gal4 driver expresses UAS::Mer^BB at high levels throughout the dorsal surface of the developing wing blade (Fig 3C), resulting in phenotypes that are more severe than those expressed by enBB. Wings from apBB flies are outheld, overgrown, and have ectopic venation primarily along veins II and V (LAJEUNESSE et al. 1998 ). Group IIc and Group IIa mutations produced a significant blistered wing phenotype in combination with apterous Mer^BB (data not shown). Mutations in all five complementation groups display modification of the apBB phenotype (Fig 3, third column) indicating that the modification is due to an effect on Mer^BB activity and not the engrailed::Gal4 driver.

Modification of Mer³ phenotypes:
The second genetic test avoided overexpression of Mer^BB altogether and instead examined the ability of the interactor to modify phenotypes expressed by a hypomorphic Merlin mutant allele, Mer³. Mer³ hemizygous males are semiviable with visible phenotypes expressed in the wings, legs, and head already described in results (MCCARTNEY et al. 2000 ). Under normal conditions 50% of the expected number of Mer³ male flies eclose. Mutations in three complementation groups, Group IIa, Group IIb, and Group IIIa, dominantly enhanced Mer³ to lethality or reduced eclosion of Mer³ males to <1% (Table 2).

The other two groups, Group IIc and Group IId, also modified Mer³ phenotypes, but in a qualitatively different manner. The modification was restricted to the wing and we observed neither alteration of viability nor head or leg defects (Table 2). The Mer³ wing phenotype is characterized by increase in the size of the wing blade with mild disruptions in venation, particularly the posterior cross vein (Fig 4B, Table 4). Hemizygous Mer³ flies that are also heterozygous for a mutation in Group IIc had significantly smaller wings when compared to Mer³ wings alone, but had a significant increase in the number of posterior cross vein disruptions (Fig 4C, Table 4). Mutations in Group IId also significantly reduced the size of the wing. Unlike the Group IIc modification, however, Group IId dominantly reduced the disruptions in venation (Fig 4D, Table 4).

View larger version (88K): In this window In a new window Download PPT slide   Figure 4. Modification of the Mer3 wing phenotype by blistered and net. (A) Wing from a wild-type male fly; (B) wing from a Mer3 hemizygous male fly. Note the overall increase in size and slight disruption of the posterior cross vein (between the two arrows). (C) Wing from a Mer3 hemizygous male also heterozygous for Group IIc/bs242. Note the reduced size compared to Mer3 wing and disrupted posterior cross vein (arrowhead). (D) Wing from a Mer3 hemizygous male heterozygous for Group IId/net383. Note the reduced size and suppression of any defects in venation.

Disruption of Merlin subcellular localization:
As a further test of the relevance of the interacting complementation groups, we examined the subcellular localization of Merlin in cells that were homozygous for mutations in each group. In previous work, we showed that the proper subcellular distribution of Merlin is important for its function (LAJEUNESSE et al. 1998 ). Mutations in two of the modifying loci resulted in altered Merlin subcellular distribution. Within the cells of the imaginal epithelium, Merlin is found associated with the apical plasma membrane in the region of the adherens junction and throughout the apical cytoplasm associated with discrete punctate structures (MCCARTNEY and FEHON 1996 ). In Group IIa or Group IIc mutant backgrounds, Merlin is mislocalized to large vesicular bodies basal to the adherens junction (Fig 5). These vesicular bodies are not present in every cell within the imaginal epithelium and can be found to a greater extent within the central nervous system, particularly the ventral ganglion (data not shown). We have tested for the presence of actin, -spectrin, two adherens junctions components (Armadillo and Moesin), Notch, and the septate junction protein Coracle, and none co-localized with Merlin within these bodies (data not shown). The identity of these structures is currently unknown. Regardless, the observed alteration of the subcellular localization of Merlin is consistent with the idea that these modifiers function with Merlin to regulate proliferation and differentiation.

View larger version (85K): In this window In a new window Download PPT slide   Figure 5. Mislocalization of Merlin in Group IIa/scribbler genetic backgrounds. Optical cross section of the wing imaginal epithelia from a sbb324/sbb324 mutant third instar larva. The majority of Merlin (in red) is found in the apical regions of the cell (small arrowheads), as would be found in a wild-type genetic background. However, in some cells, basal aggregations of Merlin protein (large arrows) are seen. Asterisks mark the basal portions of the cells. The septate junction and cell membrane are labeled in green using anti-Coracle antibody.

Characterization of the complementation groups:
Group IIb: Group IIb mapped to the left arm of the second chromosome at 2-[33], between P{w^+mC = lacW}l(2)10424^k06801 at 2-[18]/26B1-2 and P{w^+mC = lacW}Pen^k14401 at 2-[36]/31A and has been placed in the cytological region 30D;31F on the basis of noncomplementation of a recessive wing phenotype with Df(2L)Mdh (Fig 6B). All Group IIb mutations complement all previously described mutations (see MATERIALS AND METHODS) that map to this region. All transallelic combinations of the three Group IIb alleles identified in this screen have an early larval lethality with no distinct phenotypes. The allele strengths of the three alleles based on interaction with enBB are as follows: IIb¹⁸⁷, IIb¹⁸² > IIb²⁰⁹.

View larger version (133K): In this window In a new window Download PPT slide   Figure 6. Wing phenotypes of Merlin modifiers. (A) Wild-type female wing. (B) Wing from IIb187/Df(2L)Mdh, which displays extra wing vein. (C) Wing from bs217/bs242 female fly. All intervein cells have been transformed into vein tissue giving the wing a tube-like appearance. (D) Wing from net383 homozygous female fly with disturbed normal vein patterning and extra veins. (E) Wing from IIa270/IIa324 escaper. Note ectopic wing vein material along veins II and V. (F) Wing from IIa270 +/ + bs242 transheterozygous female fly demonstrating second-site noncomplementation between two recessive mutations (compare to Fig 2D and Fig G). The broad arrowhead indicates a large blister in the wing. Note the alteration in the shape and size when compared to A and the presence of ectopic vein material (thin arrowhead).

Group IIc/blistered: Group IIc mapped to the end of the right arm of the second chromosome and failed to complement the lethality of Df(2R)Px2, a deletion of 60C6;60D9. Group IIc mutations were shown to be allelic to blistered (bs), a gene located within this interval, by failure to complement bs⁰³²⁶⁷, a null mutation. blistered encodes the Drosophila homologue of serum response factor (GUILLEMIN et al. 1996 ; MONTAGNE et al. 1996 ). Seven new alleles of blistered were identified. All are recessive and appear to be hypomorphs except for the X-ray allele, Group IIc/bs³⁶⁴. This allele is semidominant with ectopic vein material, which are characteristic properties of null or strong hypomorphic blistered alleles (FRISTROM et al. 1994 ). Moreover, all lethal allelic combinations showed the abbreviated larval tracheal phenotype, another characteristic blistered phenotype (data not shown; GUILLEMIN et al. 1996 ). The allelic series of blistered as determined by lethal period is identical to the allelic series established by the interaction with enBB with the order as follows: bs³⁶⁴ > bs²⁴² > bs²²¹ > bs²⁴⁶ > bs²⁵³ > bs²³⁷ > bs²¹⁷ > bs²¹¹. This result suggests that the interaction of blistered with Merlin is a direct function of gene dosage. Transheterozygous combinations of all blistered alleles (except for transallelic combinations of bs^{364, 242, 221}, which are larval lethal) produce adult escapers that have a tube wing phenotype due to conversion of all intervein tissue to vein (FRISTROM et al. 1994 ; MONTAGNE et al. 1996; Fig 6C).

Group IId/net: Deletion mapping localized Group IId to the tip of the left arm of the second chromosome. Complementation analysis using Group IId¹⁰⁷ suggested that it was a terminal deficiency at the tip of 2L, because it failed to complement Df(2L)PMF47c, Df(2L)net62, and three mutations located within these deficiencies: lethal (2) giant larvae, broad head, and net. Of these genes, only mutations in net modified Merlin phenotypes, suggesting that net is the modifying locus within this region. net belongs to the plexus phenotypic group within the "excess-of-vein" mutant class (DIAZ-BENJUMEA and GARCIA-BELLIDO 1990 ). Two new alleles of net were recovered. net³⁸³ is a viable X-ray allele that has the characteristic ectopic venation phenotype (Fig 6D).

Group IIIa: Group IIIa is a novel group that maps by meiotic recombination to the left arm of the third chromosome at 3-[26] between P{w^+mC = lacW}l(3)L1170^L1170 and P{w^+mC = lacW}l(3)j2B9^j2B9, falling roughly within cytological interval 66A;66D. Four alleles were identified. No deficiency was identified that uncovers this complementation group. The allele strength based on the lethality of heteroallelic combinations is as follows: IIIa²³⁹ > IIIa²⁷⁸ > IIIa³²⁰ > IIIa²⁰². The strongest allelic combination, Group IIIa²³⁹/Group IIIa²⁷⁸, displays late embryonic/early larval lethality with no distinct phenotypes. Weaker allelic combinations, such as Group IIIa³²⁰/Group IIIa²⁰² and Group IIIa²⁷⁸/Group IIIa²⁰², die as early pupae with little development past the white prepupa stage (data not shown). Group IIIa²³⁹ heterozygous flies have a weakly penetrant (28%) abrupt vein V phenotype, resulting in a gap between the margin and the end of the vein (Fig 3P, small arrow).

Group IIa/scribbler: Group IIa mutations mapped to 2-[83] between P{w + mc + lacW}AA48 and P{w^+mW.hs = GawB}559.1. A test of the available deficiencies in the 54D–55F region showed that Group IIa mutations mapped to cytological region 55C2;55F on the basis of failure to complement the lethality of Df(2R)PC4. In addition, two lethal P-element insertions in this cytological interval (l(2)04440 and l(2)k00702) failed to complement visible wing phenotypes of Group IIa mutations (data not shown). However, neither P-element mutation genetically interacted with either Mer^BB or Mer³.

The most severe Group IIa alleles were hemizygous pupal lethal, although rare escapers can be found with wing defects including reduced size and ectopic vein material (Fig 6E). On the basis of the hemizygous lethal period the following allelic series of Group IIa mutations was constructed: IIa²⁷⁰, IIa²⁵⁶, IIa¹⁵¹ > IIa³²⁴, IIa²⁵⁹. Two alleles, IIa⁹⁴ and IIa²¹⁶, are homozygous/hemizygous viable and express a wing phenotype in trans with the lethal Group IIa alleles and the deficiency that is similar to the wing phenotypes displayed by the rare escapers from the lethal Group IIa alleles. Both mutations were placed into Group IIa on the basis of meiotic mapping. There is no direct correlation of the allelic series based on lethal period with that based on interaction with Merlin phenotypes, suggesting that the Merlin interactions involve something other than simple loss of function.

During our investigation, three other laboratories identified the same P-element insertions (l(2)04440 and l(2)k00702) as mutations in a gene called scribbler or brakeless (RAO et al. 2000 ; SENTI et al. 2000 ; YANG et al. 2000 ). Two groups demonstrated that, in scribbler mutants, axons from photoreceptors R1–R6 failed to stop properly upon reaching their targets in the optic lobe of the pupal brain during Drosophila eye development (RAO et al. 2000 ; SENTI et al. 2000 ). A third group studying Drosophila foraging behavior named this gene scribbler because homozygous mutant larvae display aberrant crawling patterns (YANG et al. 2000 ). Although the sbb alleles isolated in our screen do not express either phenotype, Group IIa alleles (Group IIa²⁵⁶ and Group IIa³²⁴) fail to complement the lethality of a strongly hypomorphic scribbler allele sbb⁴, which correlates with a 436-bp deletion in the third exon of sbb (RAO et al. 2000 ). In addition, sequence analysis of the same two Group IIa alleles revealed nonsense mutations within the scribbler coding region (Fig 7). Together these results indicate that Group IIa corresponds to the scribbler gene, and is henceforth called by this name.

View larger version (15K): In this window In a new window Download PPT slide   Figure 7. Two of the Group IIA/sbb alleles identified in this screen are missense mutations that affect only the SBBB isoform. Group IIA/sbb324 correlates with a nonsense mutation at arginine 1608 and Group IIA/sbb256 with a nonsense mutation at arginine 1899. Both mutations may produce truncated SBBB proteins missing Region B, a novel conserved C-terminal domain.

Second-site noncomplementation between scribbler and blistered:
In the course of our complementation analysis, we observed second-site noncomplementation between the scribbler alleles identified in this screen and mutations in blistered. Such interactions are relatively uncommon and when observed are usually a good predictor of strong functional relationships between the interacting genes (SHEARN 1989 ; TRIPOULAS et al. 1996 ; HALSELL and KIEHART 1998 ). Interactions were observed using both those blistered alleles identified in our screen and the null sbb⁰³⁶²⁷ allele. Transheterozygous combinations of strong alleles of blistered and scribbler produced blistering and ectopic vein material in a significant fraction of the flies (Fig 6F). In contrast, the original scribbler P-element alleles (RAO et al. 2000 ; SENTI et al. 2000 ; YANG et al. 2000 ), which are likely hypomorphic, completely complement blistered phenotypes. However, the deficiency that uncovers sbb, Df(2R)PC4, showed weaker second-site noncomplementation with blistered mutations (data not shown), suggesting that the sbb alleles identified in our screen are distinct from null or strongly hypomorphic sbb alleles.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Merlin is the Drosophila homologue of the human NF2 gene and is required for the regulation of proliferation and differentiation of epithelial tissues. However, the mechanisms of Merlin function are unknown. To identify genes involved in Merlin's cellular functions we performed genetic screens for mutations that modify Merlin wing phenotypes caused by expression of a dominant negative form of Merlin, enBB, and a hypomorphic allele of Merlin, Mer³. In a screen of the deficiency kit, we identified 20 chromosomal regions that contain dose-sensitive modifiers of enBB phenotypes and 23 chromosomal regions that modify the Mer³ phenotype. Eight regions were identified in both screens, suggesting that they contain Merlin modifiers of particular interest. However, we were unable to identify the individual genes within the cytological regions that modify Merlin phenotypes. To complement the deficiency screen, we performed a dominant second-site modifier screen designed to identify dose-sensitive modifiers of Merlin wing phenotypes generated by enBB. In a screen of 100,000 progeny from mutagenized flies, we identified 29 recessive mutations that modify Merlin phenotypes. Twenty-three of the mutations fall into five complementation groups. Three of the complementation groups are new alleles of previously identified genes, sbb, bs, and net. Two groups, Group IIb and Group IIIa, are novel genes. Genetic tests suggest that mutations in all five complementation groups interact with Merlin to regulate proliferation or differentiation.

None of the modifying mutations were in the regions that were identified by the Mer³ deficiency kit screen, including one region (In(2R)bw^VDe2L) with a very strong genetic interaction. Moreover, no new alleles of expanded, another Drosophila 4.1 family member and a previously characterized Merlin modifier (MCCARTNEY et al. 2000 ), were identified, despite the fact that an existing amorphic expanded allele interacts with both Mer³ (MCCARTNEY et al. 2000 ) and enBB (data not shown). There are several explanations to account for these results. It is possible that the In(2R)bw^VDe2L deficiency (or the chromosome that carries it) contains additional mutations that singly do not exhibit an interaction above the phenotypic threshold used in this screen. Similarly, putative mutations in expanded may have fallen below the level of detection. Furthermore, mutations in other potential dose-sensitive modifiers of Merlin phenotypes may have generated dominant sterile or lethal interaction phenotypes with enBB and would not have been recoverable from this screen. Interestingly, we also found that while point mutations in the modifier groups displayed interactions with the Mer³ allele, in some cases deficiencies that uncovered these genes did not show similar interaction with Mer³ (Table 1). In these cases the observed genetic interactions may be allele specific (resulting from neomorphic, hypermorphic, or antimorphic mutations), although it seems likely that many, if not most, of the modifier mutations isolated are simple hypomorphic mutations. Regardless, this report illustrates that there are qualitative differences between a deficiency kit screen and a random mutagenesis screen for second-site modifiers. Although deficiency kit screens have been successful in identifying functionally related genes (HALSELL and KIEHART 1998 ), the ability to screen a range of mutations in a common genetic background can allow for more complete and less ambiguous identification of interacting loci.

Merlin clearly has a role in the regulation of proliferation and differentiation; however, the proteins and pathways that are involved with Merlin function remain unknown. Therefore, the intent of our genetic screens was to identify genes that functionally and/or physically interact with Merlin and thus define the molecular context in which Merlin functions. Of the five complementation groups identified in this screen, sbb and bs were characterized molecularly and at this point hold the most potential in understanding Merlin function. In addition, we showed that mutations in blistered and sbb disrupt the subcellular localization of Merlin and that both mutations exhibit strong second-site noncomplementation, suggesting an underlying functional relationship between sbb and bs gene products and Merlin.

In our screen we identified seven new alleles of sbb; allelism was based on noncomplementation with a null sbb allele and the presence of nonsense mutations in two sbb alleles identified. Null and strong hypomorphic mutations in scribbler result in aberrant axon guidance and behavioral phenotypes (RAO et al. 2000 ; SENTI et al. 2000 ; YANG et al. 2000 ). However, none of the sbb alleles identified in our screen display either of these phenotypes (data not shown). In addition, none of the previously identified P-element insertional mutations in sbb modify Merlin phenotypes, although the null sbb⁴ allele and Df(2R)PC4 do enhance Merlin phenotypes (data not shown and Table 1). These data suggest that sbb has two distinct functions, one in axon guidance of photoreceptor cells and the other in regulation of proliferation in epithelial cells, and that these functions are independent. Consistent with this model, previous studies showed that sbb encodes two novel proteins of unknown function, SBB-A and SBB-B (Fig 7; SENTI et al. 2000 ; YANG et al. 2000 ). Although it was shown that the two SBB isoforms are functionally redundant in axon guidance (SENTI et al. 2000 ), the presence of a zinc finger domain and a novel Region B in the larger SBB-B isoform suggests that it may have functions distinct from SBB-A. Sequence analysis indicates that two of the alleles we isolated as Merlin modifiers correlate with nonsense mutations that affect the SBB-B product but leave the BSKA product intact (the lesions in the other alleles have not yet been determined).

The identification of SBB-B mutations that specifically modify Merlin phenotypes but do not affect photoreceptor axon guidance supports a model where SBB-B has distinct functions in the proliferation and differentiation of wing tissue. Both SBB isoforms are reported to be nuclear proteins and the presence of a zinc finger in SBB-B suggests that it may be involved in transcriptional regulation. How SBB proteins interact with Merlin, a membrane-associated cytoplasmic protein, is unclear. The observation that Merlin subcellular localization is disrupted in sbb mutant cells makes this question particularly intriguing and suggests that sbb may play a role in a cellular pathway that regulates Merlin function. The identity of this pathway is currently unknown.

While sbb encodes novel proteins with unknown function, the bs gene product, also known as the Drosophila serum response factor (BS/DSRF), is a well-characterized transcription factor (AFFOLTER et al. 1994 ; GUILLEMIN et al. 1996 ). bs is required for formation of terminal tracheal branches and differentiation of the adult wing (FRISTROM et al. 1994 ; GUILLEMIN et al. 1996 ; MONTAGNE et al. 1996 ; ROCH et al. 1998 ). BS/DSRF activity, like that of its mammalian homologue, is regulated by the epidermal growth factor receptor (EGFR) signaling pathway (ROCH et al. 1998 ). During development of the wing imaginal disc, cells can adopt one of two fates; most cells form wing blade (intervein tissue), while a subset form the characteristic longitudinal veins. BS/DSRF is believed to promote the intervein cell fate—loss-of-function bs mutations result in wings in which all cells develop as vein tissue. Activity of the EGFR pathway is believed to promote the vein cell fate by downregulating BS/DSRF function in the vein primordia and promoting the expression of vein-specific genes. Thus interactions between the EGFR pathway and BS/DSRF play a crucial role in wing development.

The identification of bs as a dominant modifier of Merlin phenotypes suggests that Merlin, like Blistered, is involved in EGFR signaling. Specifically, the observation that bs mutations enhance Merlin dominant-negative and loss-of-function phenotypes suggests that Merlin may function antagonistically to EGFR pathway function (Fig 8). Although this hypothesis should be considered as tentative, several lines of evidence support this notion. First, developing wing cells that have lost both Merlin and expanded, which appear to function redundantly, produce abundant ectopic vein material adjacent to endogenous veins (MCCARTNEY et al. 2000 ). Second, net, which was also identified as a Merlin modifier, has been shown to modify phenotypes of components of EGFR signaling in the wing (STURTEVANT and BIER 1995 ; BIEHS et al. 1998 ). Third, a role for Merlin in negatively regulating EGFR function is consistent with the observation that Merlin mutations result in overproliferation phenotypes (LAJEUNESSE et al. 1998 ). Finally, a hypermorphic EGFR mutation called Ellipse enhances phenotypes expressed by dominant-negative and hypomorphic Merlin alleles (data not shown). However, despite these intriguing indications that Merlin may function to regulate EGFR pathway activity, it should be noted that Merlin does not interact genetically with several other known pathway members (Star, asteroid, and rhomboid), nor does it interact with hypomorphic EGFR mutations (data not shown). In addition, because other signaling pathways, including dpp, wingless, and Notch, are involved in vein specification (ROCH et al. 1998 ), it is possible that Merlin functions to regulate one or more of these either instead of or in addition to the EGFR pathway. In support of this notion, Merlin and expanded have both been shown to genetically interact with dpp (MCCARTNEY et al. 2000 ). Further experiments are required to determine the significance of these genetic interactions. Nonetheless, the identification of Merlin modifiers suggests testable hypotheses regarding Merlin cellular functions and opens new avenues for further investigation of the molecular basis of the NF2 disorder.

View larger version (13K): In this window In a new window Download PPT slide   Figure 8. Merlin may function antagonistically to EGFR pathway function. blistered expression is required for the formation of intervein regions within the wing and is negatively regulated by EGF signaling in presumptive vein tissue. Alteration in Merlin activity may hyperactivate EGF signaling in intervein regions, thus disrupting the differentiation of intervein regions and promoting the formation of ectopic veins.

	FOOTNOTES

¹ Present address: Department of Biology, University of North Carolina, Greensboro, NC 27402.
² Present address: Department of Biology, University of North Carolina, Chapel Hill, NC 27599.

	ACKNOWLEDGMENTS

We thank R. Lamb for the SEM of a wild-type Drosophila eye used in Fig 1. We thank Y. Rao for a stock of the sbb⁴ allele. We thank Marla Sokolowski and her colleagues for informative conversations and a preprint of her manuscript. This work was supported by National Institutes of Health grant NS34783 and DOD grant DAMD17-97-1-7345 to R. G. Fehon. D. LaJeunesse and B. McCartney were supported by Young Investigator Awards from the National Neurofibromatosis Foundation and D. LaJeunesse received a National Institutes of Health National Research Service Award.

Manuscript received January 11, 2001; Accepted for publication February 26, 2001.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

AFFOLTER, M., J. MONTAGNE, U. WALLDORF, J. GROPPE, and U. KLOTER et al., 1994  The Drosophila SRF homologue is expressed in a subset of tracheal cells and maps within a genomic region required for tracheal development. Development 120:743-753[Abstract].

ALGRAIN, M., O. TURUNEN, A. VAHERI, D. LOUVARD, and M. ARPIN, 1993  Ezrin contains cytoskeleton and membrane binding domains accounting for its proposed role as a membrane-cytoskeletal linker. J. Cell Biol. 120:129-139[Abstract/Free Full Text].

BIEHS, B., M. A. STURTEVANT, and E. BIER, 1998  Boundaries in the Drosophila wing imaginal disc organize vein-specific genetic programs. Development 125:4245-4257[Abstract].

BRAND, A. H. and N. PERRIMON, 1993  Targeted gene expression as a means of altering cell fates and generating dominant phenotypes. Development 118:401-415[Abstract].

BRUDER, C. E., K. ICHIMURA, O. TINGBY, K. HIRAKAWA, and A. KOMATSUZAKI et al., 1999a  A group of schwannomas with interstitial deletions on 22q located outside the NF2 locus shows no detectable mutations in the NF2 gene. Hum. Genet. 104:418-424[Medline].

BRUDER, C. E., K. ICHIMURA, E. BLENNOW, T. IKEUCHI, and T. YAMAGUCHI et al., 1999b  Severe phenotype of neurofibromatosis type 2 in a patient with a 7.4-MB constitutional deletion on chromosome 22: possible localization of a neurofibromatosis type 2 modifier gene? Genes Chromosomes Cancer 25(2):184-190[Medline].

CHISHTI, A. H., A. C. KIM, S. M. MARFARTIA, M. LUTCHMAN, and M. HANSPAL et al., 1998  The FERM domain: a unique module involved in the linkage of cytoplasmic proteins to the membrane. Trends Biochem. Sci. 23:281-282[Medline].

DIAZ-BENJUMEA, F. J. and A. GARCIA-BELLIDO, 1990  Genetic analysis of the wing vein pattern of Drosophila. Roux's Arch. Dev. Biol. 198:336-354.

FEHON, R. G., T. OREN, D. R. LAJEUNESSE, T. E. MELBY, and B. M. MCCARTNEY, 1997  Isolation of mutations in the Drosophila homologues of the human Neurofibromatosis 2 and yeast CDC42 genes using a simple and efficient reverse-genetic method. Genetics 146:245-252[Abstract].

FLYBASE,, 1999  The FlyBase database of the Drosophila genome projects and community literature. Nucleic Acids Res. 27:85-88[Abstract/Free Full Text].

FRISTROM, D. K., P. GOTWALS, S. EATON, T. B. KORNBERG, and M. A. STURTEVANT et al., 1994  blistered: a gene required for vein/intervein formation in wings of Drosophila. Development 120:2661-2671[Abstract/Free Full Text].

GRONHOLM, M., M. SAINIO, F. ZHAO, L. HEISKA, and A. VAHERI et al., 1999  Homotypic and heterotypic interaction of the Neurofibromatosis 2 tumor suppressor protein Merlin and the ERM protein ezrin. J. Cell Sci. 112:895-904[Abstract].

GUILLEMIN, K., J. GROPPE, K. DUCKER, R. TREISMAN, and E. HAFEN et al., 1996  The pruned gene encodes the Drosophila serum response factor and regulates cytoplasmic outgrowth during terminal branching of the tracheal system. Development 122:1353-1362[Abstract].

HALSELL, S. R. and D. P. KIEHART, 1998  Second-site noncomplementation identifies genomic regions required for Drosophila nonmuscle myosin function during morphogenesis. Genetics 148:1845-1863[Abstract/Free Full Text].

HEISKA, L., K. ALFTHAN, M. GRONHOLM, P. VILJA, and A. VAHERI et al., 1998  Association of ezrin with intercellular adhesion molecule-1 and -2 (ICAM-1 and ICAM-2). Regulation by phosphatidylinositol 4, 5-bisphosphate. J. Biol. Chem. 273:21893-21900[Abstract/Free Full Text].

HELANDER, T., O. CARPEN, O. TURUNEN, P. E. KOVANEN, and A. VAHERI et al., 1996  ICAM-2 redistributed by ezrin as a target for killer cells. Nature 382:265-268[Medline].

HERRLICH, P., H. MORRISON, J. SLEEMAN, V. ORIAN-ROUSSEAU, and H. KONIG et al., 2000  CD44 acts both as a growth- and invasiveness-promoting molecule and as a tumor-suppressing cofactor. Ann. NY Acad. Sci. 910:106-118[Abstract/Free Full Text].

LAJEUNESSE, D. R., B. M. MCCARTNEY, and R. G. FEHON, 1998  Structural analysis of Drosophila Merlin reveals functional domains important for growth control and subcellular localization. J. Cell Biol. 141:1589-1598[Abstract/Free Full Text].

LITTLETON, J. T. and H. J. BELLEN, 1994  Genetic and phenotypic analysis of thirteen essential genes in cytological interval 22F1–2; 23B1–2 reveals novel genes required for neural development in Drosophila. Genetics 138:111-123[Abstract].

MARTUZA, R. L. and R. ELDRIDGE, 1988  Neurofibromatosis 2 (bilateral acoustic Neurofibromatosis). N. Engl. J. Med. 318:684-688[Medline].

MATSUI, T., S. YONEMURA, and S. TSUKITA, S. TSUKITA, 1999  Activation of ERM proteins in vivo by Rho involves phosphatidyl-inositol 4-phosphate 5-kinase and not ROCK kinases. Curr. Biol. 9:1259-1262[Medline].

MCCARTNEY, B. M. and R. G. FEHON, 1996  Distinct cellular and subcellular patterns of expression imply distinct functions for the Drosophila homologues of moesin and the neurofibromatosis 2 tumor suppressor, Merlin. J. Cell Biol. 133:843-852[Abstract/Free Full Text].

MCCARTNEY, B. M., R. M. KULIKAUSKAS, D. R. LAJEUNESSE, and R. G. FEHON, 2000  The NF-2 homologue, Merlin, and the tumor suppressor expanded function together in Drosophila to regulate cell proliferation and differentiation. Development 127:1315-1324[Abstract].

MCCLATCHEY, A. I., I. SAOTOME, K. MERCER, D. CROWLEY, and J. F. GUSELLA et al., 1998  Mice heterozygous for a mutation at the NF2 tumor suppressor locus develop a range of highly metastatic tumors. Genes Dev. 12:1121-1133[Abstract/Free Full Text].

MONTAGNE, J., J. GROPPE, K. GUILLEMIN, M. A. KRASNOW, and W. J. GEHRING et al., 1996  The Drosophila Serum Response Factor gene is required for the formation of intervein tissue of the wing and is allelic to blistered. Development 122:2589-2597[Abstract].

MURTHY, A., C. GONZALEZ-AGOSTI, E. CORDERO, D. PINNEY, and C. CANDIA et al., 1998  NHE-RF, a regulatory cofactor for Na(+)-H(+) exchange, is a common interactor for Merlin and ERM (MERM) proteins. J. Biol. Chem. 273:1273-1276[Abstract/Free Full Text].

RAO, Y., P. PANG, W. RUAN, D. GUNNING, and S. L. ZIPURSKY, 2000  brakeless is required for photoreceptor growth-cone targeting in Drosophila. Proc. Natl. Acad. Sci. USA 97:5966-5971[Abstract/Free Full Text].

REBAY, I., F. CHEN, F. HSIAO, P. A. KOLODZIEJ, and B. H. KUANG et al., 2000  A genetic screen for novel components of the Ras/Mitogen-activated protein kinase signaling pathway that interact with the yan gene of Drosophila identifies split ends, a new RNA recognition motif-containing protein. Genetics 154:695-712[Abstract/Free Full Text].

RECZEK, D., M. BERRYMAN, and A. BRETSCHER, 1997  Identification of EBP50: a PDZ-containing phosphoprotein that associates with members of the ezrin-radixin-moesin family. J. Cell Biol. 139:169-179[Abstract/Free Full Text].

ROCH, F., A. BAONZA, E. MARTIN-BLANCO, and A. GARCIA-BELLIDO, 1998  Genetic interactions and cell behavior in blistered mutants during proliferation and differentiation of the Drosophila wing. Development 125:1823-1832[Abstract].

ROULEAU, G. A., P. MEREL, M. LUTCHMAN, M. SANSON, and J. ZUCMAN et al., 1993  Alteration in a new gene encoding a putative membrane organizing protein causes Neurofibromatosis type 2. Nature 363:515-521[Medline].

SAINIO, M., F. ZHAO, L. HEISKA, O. TURUNEN, and M. DEN BAKKER et al., 1997  Neurofibromatosis 2 tumor suppressor protein colocalizes with ezrin and CD44 and associates with actin-containing cytoskeleton. J. Cell Sci. 110:2249-2260[Abstract].

SCOLES, D. R., D. P. HUYNH, P. A. MORCOS, E. R. COULSELL, and N. G. ROBINSON et al., 1998  Neurofibromatosis 2 tumor suppressor schwannomin interacts with betaII-spectrin. Nat. Genet. 18:354-359[Medline].

SENTI, K. A., K. LELEMAN, F. EISENHABER, and B. J. DICKSON, 2000  brakeless is required for lamina targeting of R1–R6 axons in the Drosophila visual system. Development 127:2291-2301[Abstract].

SHAW, R. J., A. I. MCCLATCHEY, and T. JACKS, 1998  Regulation of the Neurofibromatosis type 2 tumor suppressor protein Merlin by adhesion and growth arrest stimuli. J. Biol. Chem. 273:7757-7764[Abstract/Free Full Text].

SHEARN, A., 1989  The ash-1, ash-2 and trithorax genes of Drosophila melanogaster are functionally related. Genetics 121:517-525[Abstract/Free Full Text].

SIMON, M. A., D. D. L. BOWTELL, G. S. DODSON, T. R. LAVERTY, and G. M. RUBIN<