InteractiveFly: GeneBrief

Mats: Biological Overview | Regulation | Developmental Biology | Effects of Mutation | Evolutionary Homologs | References

BIOLOGICAL OVERVIEW

Appropriate cell number and organ size in a multicellular organism are determined by coordinated cell growth, proliferation, and apoptosis. Disruption of these processes can cause cancer. Recent studies have identified the Large tumor suppressor (Lats)/Warts (Wts) protein kinase as a key component of a pathway that controls the coordination between cell proliferation and apoptosis. Growth inhibitory functions are described for a Mob superfamily protein, termed Mats (Mob as tumor suppressor), in Drosophila. Loss of Mats function results in increased cell proliferation, defective apoptosis, and induction of tissue overgrowth. Mats and Wts function in a common pathway. Mats physically associates with Wts to stimulate the catalytic activity of the Wts kinase. A human Mats ortholog (Mats1) can rescue the lethality associated with loss of Mats function in Drosophila. Since Mats1 is mutated in human tumors, Mats-mediated growth inhibition and tumor suppression is likely conserved in humans (Lai, 2005).

Individual Mob family proteins also interact with Tricornered (Trc), the Drosophila Ndr (Nuclear Dbf2-related) serine/threonine protein kinase that is required for the normal morphogenesis of a variety of polarized outgrowths including epidermal hairs, bristles, arista laterals, and dendrites. In yeast the Trc homolog Cbk1 needs to bind Mob2 to activate the RAM pathway. Genetic and biochemical data is provided that Drosophila Trc interacts with and is activated by Drosophila Dmob proteins, specifically Mats and Dmob2 (FlyBase terms the gene Dmob2 Mob1). Evidence is also provided that Drosophila Mob proteins interact with the related Warts/Lats kinase, which functions as a tumor suppressor in flies and mammals. In trc mutants the overall pattern of denticles is partly disorganized and many denticles are split. Split denticles are infrequent in wild-type larvae. The denticle pattern of mats mutant larvae is also disorganized and contains many split denticles. Interestingly, the overgrowth tumor phenotype that results from mutations in Dmob1 (mats) is only seen in genetic mosaics and not when the entire animal is mutant. Unlike in yeast, in Drosophila individual Mob proteins interact with multiple kinases and individual NDR family kinases interact with multiple Mob proteins; in particular, Mats interacts physically and genetically with trc and mats phenotype resembles that of trc. Notably, trc:mats double mutant larvae do not have a more severe phenotype than the single mutants. This lack of additivity argues that trc and mats function in a common pathway during denticle development. These observations also suggest that mats functions with both Trc and Wts (He, 2005b).

During normal development of multicellular organisms, appropriate cell number and organ size are determined by coordinated cell growth, cell proliferation, and apoptosis. Disruption or malfunction of these processes can cause diseases such as cancer. Using model organisms such as Drosophila melanogaster, genetic studies have helped identify novel molecules and pathways that are critical for regulating these processes. In particular, a pathway that involves Hippo (Hpo), Salvador (Sav)/Shar-pei, and Large tumor suppressor (Lats)/Warts (Wts) proteins has been shown to play a crucial role in tissue growth and cell number control (reviewed by Hay, 2003: Rothenberg, 2003; Ryoo, 2003; Lai, 2005 and references therein).

A critical role of hpo pathway in the linkage of cell proliferation and apoptosis was first elucidated through genetic studies, because hpo mutations result in increased tissue growth and impaired apoptosis. Like hpo, clones of sav or wts mutant cells acquire growth advantage compared to their wild-type neighboring cells and display reduced apoptosis. The hpo gene encodes a protein kinase highly related to mammalian Mst1 and Mst2 proteins. Hpo associates with and phosphorylates Sav scaffold protein, and association with Sav promotes Wts phosphorylation by Hpo. The Hpo-Sav-Wts pathway has been shown to regulate cell proliferation by targeting key cell cycle regulators such as Cyclin E; Cyclin E expression was elevated in the absence of hpo, sav, or wts function. Moreover, Hpo directly phosphorylates an apoptosis inhibitor DIAP1 and may regulate DIAP1 levels through degradation. Hpo may also negatively regulate diap1 at the transcriptional level. Thus, the Hpo-Sav-Lats pathway functions to coordinate cell proliferation and cell death by regulating the levels of key molecules required for cell cycle and apoptosis control. As an important component of this newly discovered pathway, wts encodes a serine/threonine protein kinase related to the NDR and Dbf2 kinases (reviewed by Tamaskovic, 2003). In particular, the putative kinase activity of Wts has been shown to be required to inhibit cell proliferation and induce apoptosis. However, it is not clear how the catalytic activity of Wts protein kinase can be directly regulated for growth inhibition and apoptosis promotion (Lai, 2005 and references therein).

The Drosophila compound eye was used to address how cell proliferation and apoptosis are coordinately regulated for the determination of cell number and organ size during development. Drosophila eye development has been extensively studied, which greatly facilitates functional analysis of genes and pathways involved in fundamental biological processes such as cell proliferation and apoptosis. This study describes tumor suppressor functions for a Mob superfamily protein Mats. Since the first Mob gene mob1 (Mps one binder 1) was identified in yeast (Luca, 1998), over 130 members in the Mob supergene family have been found in all major kingdoms ranging from protists to animals. However, functions of most mob genes remain poorly understood. Loss of mats function leads to increased cell proliferation and dramatic tumor growth in Drosophila. The results also suggest that mats is required to facilitate apoptosis during early eye development. Importantly, mats synergistically interacts with wts and appears to function with wts in a common pathway. Mats associates with Wts and functions as an activating subunit of the Wts protein kinase. It was found that a human Mats ortholog, Mats1, can functionally replace Drosophila Mats to suppress tumor and lethal phenotypes induced by mats mutations, and Mats1 loss-of-function mutations appear to occur in both human and mouse tumor cells. These results suggest that Mats-mediated growth inhibition and tumor suppression may be conserved in mammals such as humans (Lai, 2005).

As a unique group of the Mob superfamily, Mats orthologs exist in both plants and animals. Since Mats proteins are highly conserved, their function may be conserved across species. In support of this, human Mats1 was found to functionally substitute for mats in Drosophila. Importantly, loss-of-function mutations in Mats1 have been identified in a human skin cancer and a mouse breast tumor, suggesting that mammalian Mats genes may indeed act as tumor suppressors. Further molecular analysis of mammalian Mats genes from tumor tissues will be needed to test this hypothesis. On the basis of these data, it is speculated that all mats genes from animals and plants may negatively regulate cell number and tissue growth by restricting cell proliferation and promoting apoptosis (Lai, 2005).

Tumor suppressors normally act as inhibitors of cell proliferation or activators of apoptosis and use a variety of mechanisms in tissue growth suppression. This work provides evidence that mats functions to restrict cell proliferation and promote apoptosis in Drosophila. In this regard, functions of mats are similar to those of hpo, sav, and wts. Like hpo, sav, and wts, mats negatively regulates expression of CycE and DIAP1, two key regulators involved in cell cycle or apoptosis control. However, the overgrowth phenotypes of mats mutants appear to be stronger than those of hpo, sav, and wts and therefore cannot be explained simply by increased expression of Cyclin E and loss of apoptosis. It is suspected that mats might use other mechanisms to regulate cell number and organ size. For instance, Mats may negatively regulate cell cycle regulators such as Cdc25 protein phosphatase that are required for the G2-M transition. Since yeast Mob1 is able to form a complex with Mps1 (Mono polar spindle 1) kinase, Mats may also play a role in the spindle assembly checkpoint by acting together with Mps1. Mps1 has been previously shown to be involved in the spindle assembly checkpoint in yeast, and Mps1 is also implicated in this process in vertebrate cells. Involvement of Mats in the spindle assembly checkpoint would help explain the dramatic overgrowth phenotypes of mats mutants. Clearly, further investigations are needed to test these hypotheses (Lai, 2005 and references therein).

Consistent with a model that Mats functions as a critical component of the Hpo-Sav-Wts pathway, the data show that Mats associates with Wts to form a protein complex. Supporting this, crystal structure analysis of human Mats1/Mob1A reveals that several evolutionarily conserved acidic residues are exposed on the surface to provide a strong electrostatic potential for mediating protein-protein interactions (Stavridi, 2003). Based on this finding, Mats binding regions are expected to be basic and indeed such regions do exist in Wts family proteins. It remains to be addressed as to how exactly Mats interacts with Wts and whether the Mats-Wts complex can be associated with Hpo and Sav. Excitingly, it was found that Mats functions as an activating subunit to stimulate Wts kinase activity. In this way, Wts activation can be effectively controlled by the availability of Mats protein through differential distribution of Mats in different tissues, cells, or subcellular locations. With Mats acting as an activator of Wts kinase, the relationship between Mats and Wts mimics that of Cyclin and Cyclin-dependent kinases, which are essential for cell cycle control (Lai, 2005).

How does Mats association lead to Wts activation? In a model, association with Mats may allow Wts to undergo an allosteric conformational change critical for Wts activation or to simply relieve an autoinhibition of Wts. Interestingly, the N-terminal region of Wts was shown to be able to associate with its C-terminal kinase domain through intramolecular binding, and this interaction may be inhibitory for the Wts kinase activity. Thus, association with Mats may activate Wts by disrupting this intramolecular binding within Wts. In the case of human Ndr kinase, an autoinhibitory sequence has been identified and binding of the hMats1/hMob1A protein induces a release of this autoinhibition (Bichsel, 2004). In another model, Mats association may allow the Mats-Wts complex to recruit additional coactivators or to prevent coinhibitors from being recruited in order for Wts to be activated. Clearly, any model of Wts activation would have to consider the effect of Wts phosphorylation. (1)Wts has been shown to be phosphorylated in a cell cycle-dependent manner. Because Wts kinase activity can be increased through treatment of phosphatase inhibitors, phosphorylation appears to be critical for Wts kinase activity. (2) The Drosophila homolog of C-terminal Src kinase (dCsk) gene has been shown to genetically interact with wts to inhibit cell proliferation, and dCsk phosphorylates Wts in vitro. (3) Human Wts2 is a phosphorylation target of Aurora-A kinase, and this phosphorylation plays a role in regulating centrosomal localization of hWts2. (4) Hpo can directly target Wts for phosphorylation, and this event is facilitated by Sav. At present, it is unclear how Mats may affect Wts phosphorylation by Hpo or how Mats-Wts complex may be regulated by Hpo through phosphorylation. In yeast, Mob1 is essential for the phosphorylation of Dbf2 kinase by an upstream kinase Cdc15. Further studies on Wts phosphorylation are expected to provide a better understanding of how Wts is regulated (Lai, 2005).

While functions of most Mob superfamily proteins are still poorly understood, this work on Mats supports that a common feature of Mats proteins is to function as coactivators of protein kinases such as Wts. Identification and functional studies of Mats have revealed a mechanism for the control of Wts tumor suppressor activity. Because Mats-mediated growth inhibition and tumor suppression appear to be evolutionarily conserved, it extends the understanding of tissue growth and cell number control during development and tumorigenesis and raises the possibility that Mats-dependent growth inhibition may have important implications for the understanding and treatment of human cancers (Lai, 2005).

Trc, Dmob, and Fry Function in a Common Signaling Pathway

Previous genetic data pointed out the importance of tricornered (trc) and furry (fry), encoding a large conserved protein with multiple isoforms, for the morphogenesis of polarized cellular extension. Based on homology to yeast regulatory pathways involving homologs of trc, furry and mob, it seemed likely that one or more of the Drosophila mob genes would function along with trc and fry. Evidence was found supporting this hypothesis but the results are complicated by both pleiotropy and redundancy. This was illustrated most clearly in experiments with mats. Mutations in mats displayed phenotypes that were typical of both trc (split denticles and multiple hair cells) and of wts/lats (tumors, bulged cells, advanced hair differentiation). Mats can also interact with both Trc and Wts as detected using the yeast two-hybrid system. These observations stand in contrast to the situation in yeast, in which individual mob genes show specificity for individual Ndr family members. Further evidence for redundancy comes from the gene dosage interactions seen between trc and the other Dmobs (He, 2005b).

Evidence for a direct physical interaction has been reported for Ndr and Mob family members from yeast, flies and mammals (Colman-Lerner, 2001; Mah, 2001; Weiss, 2002; Hou, 2003; Bichsel, 2004; Lai, 2005). Previous yeast two-hybrid experiments showed evidence for a physical interaction between Trc and Mats (Giot, 2003) and Mats and Wts (Lai, 2005). The results of this study extended these observations by showing a similar interaction between Trc and Dmob2 by both two hybrid and coimmunoprecipitation experiments and that Wts and Dmob2 interact in the two hybrid system. Residues known to be important for the interaction between yeast Mob1 and Dbf2 are also important for the interaction between Trc and Dmob2. The conservation of many of these residues in Dmob3 and Dmob4 suggests that these proteins will also interact with Trc. The genetic interactions seen between trc and Dmobs suggest that the binding of Dmobs to Trc is essential for in vivo function and activation of the protein. Consistent with this hypothesis it was found that Dmob2 and Trc colocalized to growing hairs in pupal wing cells (He, 2005b).

The observation that for two phenotypes (pupal wing cell cross section and time of hair initiation) mats and wts clone cells share a similar phenotype that is the opposite of trc is intriguing and needs to be reconciled with the positive gene dose interactions seen between mats and trc for the sensitized multiple wing hair cell assay and the similar denticle phenotype. Given this complexity it seems unlikely that a single simple mechanism is involved. Because Mats appears to function along with both Trc and Wts, some of the complexity may reflect interactions between these two kinase modules. Cells mutant for mats could have both modules inactive, although it is possible that the degree of possible mats redundancy might not be equivalent for the two modules. In principle these two modules could function in parallel or one could be upstream of the other. The observation that mats and wts clones have increased Fry accumulation in hairs is consistent with mats/wts being upstream of trc/fry/mats(mob). Because increased Fry accumulation in hairs is also seen in trc mutant cells, this hypothesis is also consistent with the positive gene dosage interactions. However, a different explanation is needed to explain the observation that with regard to cell size and the timing of hair initiation the mats/wts phenotype is opposite to that of trc/fry. If both modules are considered to be equally inactivated in a mats cell, then these latter observations suggest that trc/fry/mats could function antagonistically and upstream of mats/wts. In this situation a lack of trc function would result in increased wts function (and increased cell size and delayed hair formation). A lack of wts function would result in the reduced cross section and advanced hair morphogenesis. In a mats mutant a reduction is expected in both trc and wts function. This would result in a wts-like phenotype because a lack of trc inhibition of wts would be of no consequence in cells that already lack wts activity. However, this model does not explain the multiple hair cell interactions. Given the difficulties in any single model it is suggested that the interactions are context dependent and/or the two modules function entirely in parallel (He, 2005b).

The tumor phenotype of mats

The 'tumors' produced by mats clones were characterized by altered cell shape and proliferation. The altered cell shape could be seen in the cuticle of clone cells. The altered proliferation of clone cells was associated with them outcompeting neighboring cells so that, in wings in which recombination was induced at a high level using vg-Gal4 and UAS-flp, most cells in the wing were mutant and the wing was grossly larger than normal. These observations are very similar to those seen in wts/lats mutant clones. It is unclear how the altered cell shape and proliferation are related. Excess proliferation of a clone of cells that is surrounded by slower growing normal cells is expected to lead to compression of the faster growing clone cells and their immediate neighbors. This could be responsible for the decreased cross sectional area of mats and wts cells and their bulged apical surface. However, if compression is responsible for the change in cell shape, it would be expected that this change would smoothly spread into the surrounding wild-type cells and would be more severe in the center of clones than near their periphery. This does not appear to be the case although this issue deserves further study. The cells in entirely mutant discs also appeared to have a bulged shape, which further argues against excess growth-mediated compression being responsible for the cell shape changes (He, 2005b).

The tumors produced by mats clones and their ability to outcompete neighbors suggests the possibility that mats mutant cells grow and/or divide more rapidly than normal. Thus, it was surprising when it was found that discs in mats homozygous and hemizygous mutants are smaller than normal. This could be due to the mutant larvae being impaired in feeding, digestion, or absorption of nutrients. This could lead the disk cells to be effectively starved reducing their growth. Alternatively it could be due to the mutation resulting in a defect in the secretion of a growth factor. It remains possible, however, that the overgrowth requires the direct contact of mats mutant and wild-type cells (He, 2005b).

REGULATION

Protein Interactions

tricornered (trc) encodes the Drosophila Ndr protein kinase (Nuclear Dbf2 related; Geng, 2000). The Ndr kinases are members of a subfamily of serine/threonine kinases that includes Sax1 (Caenorhabditis elegans), Cbk1 (Saccharomyces cerevisiae), Dbf2 (S. cerevisiae), Warts/Lats (Drosophila), Orb6 (Schizosaccharomyces pombe) and Cot-1 (Neurospora), which regulate cell growth, cell division and cell morphology. In S. cerevisiae Cbk1 and Dbf2/Dbf20 play central roles in the RAM (regulation of AceII activity and cellular morphogenesis) and MEN (mitotic exit network) pathways. Mutants of Cbk1 or other RAM pathway genes including its binding partners Mob2 and Tao3 (Pag1) fail to activate the AceII transcription factor in daughter cells and result in rounder than normal cells due to a defect in axial growth of the bud. Little is known about the in vivo function of the two human Ndr genes but extensive study of their biochemical characteristics has been carried out for nearly 10 years (He, 2005b and references therein).

The function of trc and its partner furry (fry) is required for the development of epidermal hairs, sensory bristles, arista laterals, and dendrite arborization (da) sensory neuron dendrites. The morphogenesis of these cell extensions involves the regulated activation of both the actin and microtubule cytoskeletons. Mutations in trc and fry result in split and multipled hairs and laterals, split and deformed bristles and dendrites with extra branching and tiling defects (He, 2005b and references therein).

The single warts/large tumor suppressor (wts/lats) gene is the Drosophila kinase most closely related to trc (45% identical and 65% similar over 418 amino acids). Once again there are two wts homologues in mammals. There is no wts ortholog in yeast. wts was first identified in Drosophila as a tumor suppressor. Homozygous wts/lats mutant cells display defects in morphogenesis (such as deformed bristles and altered cuticle morphology) and extensive overgrowths (He, 2005b and references therein).

Ndr, like many kinases, is regulated by phosphorylation. The phosphorylation of the activation segment site Ser-281 and the hydrophobic motif site Thr-444 of Ndr increase Ndr kinase activity in vitro. Ser-281 phosphorylation is thought to be due to autophosphorylation, whereas Thr-444 is targeted by an as yet unidentified upstream kinase. These sites are also important regulatory sites for Trc function in the Drosophila epidermis and nervous system. The mutation of these sites in trc to alanine results in dominant negative proteins (He, 2005b and references therein).

Several Ndr family kinases have been shown to function with members of the Furry protein family, which consists of large conserved proteins that lack informative motifs. The first member of this family to be characterized was the Drosophila fry gene. Both genetic and biochemical experiments have shown that in flies trc and fry function in a common process, are present together in a complex and that Fry is required for Trc kinase activity. Mutations in both result in similar phenotypes in both the epidermis and sensory neurons. In addition, the subcellular localization/accumulation of Trc and Fry is interdependent in pupal wing cells. The subcellular localization of Cbk1 and Tao3 in S. cerevisiae and Orb6 and Mor2 in S. pombe has also been found to be interdependent, although the relationships differ in these systems (Tao3 and Mor2 are the Fry homologues in these systems) (He, 2005b and references therein).

The Trc and Furry family proteins appear to be conserved both in terms of sequence and function in a wide range of eukaryotes. This suggests that homologues of other members of the RAM pathway in S. cerevisiae will also play similar roles in higher eukaryotes. The Mob2 protein of yeast has been shown to bind to Cbk1 and be essential for Cbk1 kinase activity (Weiss, 2002). In vivo Mob2 is required along with Cbk1 for both mother/daughter separation after cytokinesis and the maintenance of polarized cell growth (Weiss, 2002). Furthermore, Mob2 and Cbk1 show interdependent localization (Nelson, 2003). A similar situation exists for the related Dbf2 kinase, which is a component of the mitotic exit network (MEN). Dbf2 binds to Mob1 (which is related to Mob2) and this complex is essential for activity (Mah, 2001). Similarly, S. pombe Mob2 interacts physically with the Orb6 protein kinase and is required for Orb6 function in the coordination of cell polarity with the cell cycle (Hou, 2003). Multicellular organisms possess multiple mob genes. Recently, it was shown that a basic sequence within the insert in the catalytic domain of Ndr has an autoinhibitory function and that Human Mob1 may stimulate Ndr activity by releasing the autoinhibitory effect of this sequence (Bichsel, 2004; Devroe, 2004; He, 2005b and references therein).

There are 4 Drosophila genes related to the yeast mob genes. Evidence for a two-hybrid interaction between CG13852 (Dmob1/mats) and Trc was described in a genome scale experiment; however; no evidence for such interactions were seen between Trc and any of the other Drosophila Mobs (Giot, 2003). Nor was there any indication in that paper that any of the Drosophila Mobs function with the related Warts/Lats kinase. Evidence is provided that Mob1 (Mats), interacts with both trc and wts and that at least one additional member of the Dmob gene family CG11711 (Dmob2: FlyBase confusingly terms this gene Mob1) can interact with trc and wts. While this paper was in revision, Lai (2005) reported that CG13852 interacted with and activated Wts. They named CG13852 mats, and the He (2005b) study follows their lead and uses that name (He, 2005b).

Evidence of Physical Interaction between Trc and Dmob2

The S. cerevisiae Cbk1 and Mob2 proteins are known to interact physically as do the S. pombe homologues (Orb6 and Mob2; Weiss, 2002; Hou, 2003). In addition the related Dbf2 and Mob1 proteins of S. cerevisiae also interact physically (Komarnitsky, 1998). Thus, it was expected that Trc and at least one of the Drosophila Mobs would interact physically. Indeed, Trc and Dmob1 (CG13582) were identified as interacting proteins in a genome scale two-hybrid screen (Giot, 2003). None of the other Dmobs were reported as being able to interact with Trc, and no Dmob was reported as interacting with Wts in that study (Giot, 2003), although a recent article (Lai, 2005) demonstrated an interaction between Wts and Mats (He, 2005b).

A yeast two-hybrid screen of a Drosophila cDNA library was performed using full-length trc cDNA as 'bait'. Most of the clear positive clones recovered contained fusions of segments of Dmob2 (FlyBase terms the Dmob2 gene Mob1) fused to the GAL4 activation domain. No clones were recovered of any of the other Dmobs. Perhaps they were not present in the library screened. To determine whether Mats and Trc interact in the yeast two-hybrid system, a cDNA clone for mats was obtained from the BDGP collection and it was subcloned into pGADT7. This plasmid was used and it was confirmed that Trc and Mats interact in the yeast-two-hybrid system. Thus, Trc appears to be able to interact with at least two different Mob family proteins in Drosophila. Similarly it was tested and confirmed that Wts is able to interact with both Mats and Dmob2 in the two hybrid system. Thus, no evidence was seen of specificity in the Drosophila NDR/Mob family interactions (He, 2005b).

To determine what portion of the Trc protein interacts with Dmobs, a set of plasmids was generated that contained trc C-terminal truncations, and they were assayed for an interaction with Dmob2 and Mats using the two-hybrid system. Similar but not identical results were obtained with these two mob family members. All of the Trc deleted proteins interacted strongly with Mats and the larger Trc proteins interacted strongly with Dmob2. However, Trc proteins that contained only amino acids 1-60 or 1-119 interact. The data for the binding of human Ndr1 to hMob1 indicates that important residues are found in the amino terminal region of Ndr1 (Bichsel, 2004). In hNdr, Tyr-31, Arg-41, Thr-74, and Arg-78 were found to be absolutely required for interaction, whereas the Lys-24, Arg-44, and Leu-79 mutants displayed reduced interaction (Bichsel, 2004). The corresponding residues in Trc are Tyr35, Arg45, Thr78, Arg81, Lys28, Arg48, and Leu82. Hence the data argue that for Mats binding to Trc requires only a subset of the residues needed in human Ndr1, whereas Dmob2 binding is enhanced by additional residues. The significance of these differences is not clear, but each of these examples is consistent with the amino terminal region of the NDR kinase family members being essential for the interaction with Mob family members (He, 2005b).

Because the protein kinase domain of Trc extends from residue 90-393, these results indicate that the kinase domain does not have to be intact for Trc to interact with Dmob2 or Mats. Mutations in the conserved regulatory phosphorylation sites, S292A+T453A did not interfere with the interaction. Thus it appears clear that the kinase activity of Trc is not important for its ability to bind to Dmob2. These results were similar to those seen (Komarnitsky, 1998) between kinase inactive Dbf2 and Mob1 (He, 2005b).

In S. cerevisiae Mob1 a number of sites have been identified as being important for the binding of Mob1 and Cbk1 (Luca, 1998). To test whether these sites are functionally conserved within the Dmob family, the sequence of the four Drosophila Mobs were aligned with yeast, human, and frog to identify the Dmob2 amino acids that correspond to the important sites in yeast Mob1 for interaction with Dbf2. Similar mutants in Dmob2 (RE70633) were generated to confirm the conservation of the Mob-Ndr interaction. Most of these mutations in Dmob2 disrupt the binding to Trc, consistent with the conclusion that the mechanism of interaction has been conserved. Most of the residues noted above are also conserved in Mats, Dmob3, and Dmob4, consistent with all Mob family members interacting with Ndr family members in the same manner (He, 2005b).

To assess whether these proteins were capable of associating in vivo in Drosophila cells, immunoprecipitation experiments were carried out in Drosophila S2 cells expressing both Trc and Dmob2. Trc was found in anti-Dmob2-8x HA, consistent with these two proteins interacting in vivo. As an additional test of these proteins interacting in vivo, the subcellular localization of Trc and Dmob2 protein was examined in wing cells. Trc distribution was examined with an anti-FLAG monoclonal antibody (Sigma) using UAS-trc^WT and UAS-trc^DN transgenes driven by ptc-GAL4. The proteins encoded by these transgenes carry an amino terminal FLAG epitope. It was found that the FLAG staining pattern of overexpressed Trc was the same as the endogenous Trc detected by anti-Trc antibody staining (He, 2005a). Dmob2 was localized in a similar way using a CFP tag because an anti-Dmob2 antibody was not available. Confocal microscopy demonstrated that before hair formation Trc is cytoplasmic and concentrated at the cell periphery. During hair outgrowth Trc accumulates in the hair, as is the case for the endogenous Trc (He, 2005a). Both before and after hair initiation Dmob-2 is localized similarly to Trc. The subcellular localization of these proteins is similar in flies that carry a single transgene or both UAS-trc and UAS-Dmob2 transgenes. It is concluded that Trc and Mob proteins can interact in vivo in Drosophila cells (He, 2005b).

Ten deficiencies from the 68C region were tested and the enhancing region was further mapped to a small interval (68C11-13) that contained CG11711 (Dmob2). Deficiencies from this region were able to similarly enhance the phenotypes that resulted from the directed expression of other dominant negative Trc proteins. The genome project annotation of Dmob2 suggests it is a complicated gene that encodes at least four variant mRNAs from exons that span >40 kb. These mRNAs encode four proteins with a common c-terminal segment but with different amino terminal regions. There are P insertions in a large intron of Dmob2, but these do not inactivate the gene to produce a mutant phenotype. Attempts to use imprecise excision to produce a deletion that would ensure that no Dmob2 protein could be made were not successful, because only small deletions were obtained that would eliminate one isoform. These did not produce a mutant phenotype. As an alternative approach transgenic flies were generated that carried UAS constructs that encoded either a tagged full-length Dmob2 (GH07469) protein or partial proteins aa 1-157 (Dmob2-N) and aa 148-354 (Dmob2-C) that might act as dominant negative proteins. The directed expression of the wild-type Dmob2 and Dmob2-N proteins by ap-GAL4 did not cause any notable visible phenotype. The interpretation of these results is limited by the fact that only one of 4 CG11711 isoforms was expressed. In contrast, overexpression of the common Dmob2-C protein segment resulted in a weak trc-like multiple hair cell phenotype and it also enhanced the dominant negative trc wing hair phenotype in a dose-sensitive way, consistent with Dmob2-C being a dominant negative and the normal function of Dmob2 being to activate Trc. In addition, overexpression of Dmob2-C caused an extra vein phenotype. This phenotype was enhanced by increasing the number of UAS-mob2c transgenes and it was also enhanced by heterozygosity for a deletion for CG11711. Thus, Dmob2-C acts as a dominant negative for this phenotype. A similar, but weaker vein phenotype was also seen when trc^DN was overexpressed (He, 2005b).

Mob as tumor suppressor is activated by Hippo kinase for growth inhibition in Drosophila

Tissue growth and organ size are determined by coordinated cell proliferation and apoptosis in development. Recent studies have demonstrated that Hippo (Hpo) signaling plays a crucial role in coordinating these processes by restricting cell proliferation and promoting apoptosis. Mob as tumor suppressor protein, Mats, functions as a key component of the Hpo signaling pathway. Mats associates with Hpo in a protein complex and is a target of the Hpo serine/threonine protein kinase. Mats phosphorylation by Hpo increases its affinity with Warts (Wts)/large tumor suppressor (Lats) serine/threonine protein kinase and ability to upregulate Wts catalytic activity to target downstream molecules such as Yorkie (Yki). Consistently, epistatic analysis suggests that mats acts downstream of hpo. Coexpression analysis indicated that Mats can indeed potentiate Hpo-mediated growth inhibition in vivo. These results support a model in which Mats is activated by Hpo through phosphorylation for growth inhibition, and this regulatory mechanism is conserved from flies to mammals (Wei, 2007).

Two protein kinases Hippo [Hpo and Warts (Wts)/large tumor suppressor (Lats)], and a scaffold protein Salvador (Sav)/Shar-pei, are key components of this pathway. Moreover, two FERM-domain proteins, Merlin (Mer) and Expanded (Ex), function upstream of Hpo, and Mob as tumor suppressor (Mats), associates with Wts to stimulate the catalytic activity of the Wts protein kinase. Recently, both putative receptor and ligand that function further upstream of, or in parallel with, Hpo signaling have been identified (Hariharan, 2006). A major signal output of this growth inhibitory pathway is to inactivate a transcription coactivator Yorkie (Yki) via phosphorylation by Wts kinase. In addition to Cyclin E and Drosophila inhibitor of apoptosis 1 (diap1), the bantam microRNA is also found to be a target of the Hpo pathway. Most components in this emerging signaling pathway are conserved from yeast to flies and humans, suggesting that this pathway plays a fundamental role in cellular regulation (Wei, 2007).

The function of Mob proteins has been better studied in yeast, Drosophila and mammalian cells, which revealed a conserved property of Mob proteins as a binding partner as well as a coactivator of protein kinases of the Ndr (nuclear Dbf2-related) family (Hergovich, 2006b). As stated above, Drosophila Mats/dMob1 is required for mediating Hpo signaling by regulating Wts kinase activity in growth inhibition and tumor suppression. All four Drosophila mob genes dMob1-4 genetically interact with trc (tricornered) (He, 2005a), the fly Ndr homolog important for maintaining integrity of epidermal outgrowths and regulating dentritic tiling and branching (Emoto, 2004; He, 2005b). In the budding yeast Saccharomyces cerevisiae, Mob1 binds to and activates Dbf2/Dbf20 protein kinases for controlling mitotic exit and cytokinesis (Komarnitsky, 1998; Lee, 2001; Mah, 2001). Similarly, Mob1 is required for the activation of Sid2, an Ndr family kinase in the fission yeast Schizosaccharomyces pombe essential for cytokinesis (Hou, 2000; Hou, 2004). In human, hLats1 preferentially interacts with hMob1/hMats, but not hMob2 protein, and appeared to be required for promoting mitotic exit (Bothos, 2005), as well as cytokinesis (Yang, 2004). Importantly, the function of Mob proteins has been highly conserved in evolution. For instance, the human Mob1A/Mats1 protein has been shown to act as a kinase activator and can rescue the lethality and tumor phenotypes ofDrosophila mats mutants (Lai, 2005; Wei, 2007 and references therein).

Structural analysis of a human Mob1 protein, Mob1A/Mats1, revealed several important features of Mob family proteins (Stavridi, 2003). One is that several highly conserved residues are responsible for generating an atypical Cys₂-His₂ zinc-binding site, which is predicted to contribute to the stability of the Mob protein. Another striking feature is that there is a flat surface rich in acidic residues on one side of the protein. This property provides the structural basis for a Mob protein to interact with its partner, such as Ndr family kinases through electrostatic forces. Indeed, a 65-amino-acid region rich in basic residues exists in the N-terminal side of the kinase domain of Ndr family kinases, and alterations in the basic residues can prevent the kinases from binding to Mob proteins (Bichsel, 2004; Bothos, 2005; Hergovich, 2006b). Finally, hMob1A adopts a globular structure involving residues throughout the polypeptide. Mob proteins are small and usually do not carry any other structural motifs other than the Mob domain (Wei, 2007).

Although previous studies suggest that Ndr family kinases can be activated by upstream regulators such as Cdc15, Hpo and Mst kinases via phosphorylation in yeast, flies or human cells, very little is known about how Mob is regulated. Studies carried out in yeast and mammalian cells suggested that Mob proteins may be regulated through phosphorylation. For instance, yeast Mob1 was shown to be essential for the phosphorylation of Dbf2 by an upstream protein kinase Cdc15 and Mob1 itself was also phosphorylated by Cdc15 (Mah, 2001). However, the functional significance of this modification has not been elucidated. Work on human Mob1A/Mats1 also suggested that phosphorylation might provide a mechanism for regulating hMob1A activity (Bichsel, 2004). This study has tested a hypothesis that Mats is directly activated by Hpo kinase to regulate Wts kinase activity for growth inhibition and tumor suppression. Using the Drosophila system, it was found that Mats can be complexed with Hpo and is a target of the Hpo protein kinase. Similarly, human Mats1 is also a target protein of mammalian Mst kinases. Mats phosphorylation by Hpo increases its affinity with Wts protein kinase and ability to increase Wts activity to target Yki. Moreover, epistatic analysis suggested that mats acts downstream of hpo. Genetic analysis indicated that Mats functions together with Hpo for mediating growth inhibition of developing organs. Therefore, the Mob as tumor suppressor protein, Mats, functions as a critical component of the Hpo signaling pathway. The results support a model in which Mats is activated by Hpo through phosphorylation for growth inhibition, and this regulatory mechanism is conserved from flies to mammals (Wei, 2007).

Recent studies have defined an emerging growth inhibitory pathway mediated by Fat, Mer/Ex, Hpo/Sav and Wts/Mats proteins in tissue growth and organ size control in Drosophila. Previous work has shown that Mats functions as a coactivator of the Wts protein kinase (Lai, 2005). This study has focused on addressing how Mats is activated to regulate Wts kinase activity. Fenetic analysis suggests that Mats acts downstream of Hpo and is a critical component of the Hpo signaling pathway. Moreover, evidence is provided that Hpo-mediated phosphorylation increases Mats's activity as a coactivator of the Wts protein kinase, and this regulatory mechanism is conserved from flies to humans. Therefore, Hpo-mediated phosphorylation of Mats significantly contributes to Wts activation. In a simple model, Hpo needs to directly phosphorylate Wts as well as Mats in order for Wts kinase to be fully activated. Although both Wts and Mats are activated by Hpo-mediated phosphorylation, further investigations are needed to address how Hpo phosphorylation and Mats binding are coordinated for Wts activation (Wei, 2007).

This report provides evidence that Mats is a target of Hpo/Mst protein kinases and Hpo/Mst-mediated phosphorylation positively regulates Mats protein's coactivator activity for Wts protein kinase. Importantly, it was found that Mats exists as a phosphoprotein in living cells, indicating that Mats phosphorylation occurs under physiological conditions. In addition to Hpo/Mst, Wts kinase has also been shown to target Mats for phosphorylation (Lai, 2005), although the physiological effect of this modification has not been elucidated. In S. cerevisiae, the founding member of the Mob superfamily Mob1 was found to be a phosphoprotein and a substrate for the Mps1 kinase. Mob1 is also phosphorylated by an upstream regulator Cdc15 kinase (Mah, 2001). However, the role of Cdc15 in Mob1 phosphorylation has not been revealed even though Mob1 is known to be required for Cdc15-mediated activation of its binding partner Dbf2 kinase. In mammalian cells, protein phosphatase 2A inhibition by OA treatment caused phosphorylation of a Mob family protein (Moreno, 2001). Moreover, OA-induced modification on hMob1 was shown to be critical for its binding to its partner Ndr kinase (Bichsel, 2004). Thus, phosphorylation appears to be a common mechanism for Mob regulation (Wei, 2007).

Consistent with the finding that Mats is activated by Hpo via phosphorylation for upregulating Wts kinase activity, epistatic analysis suggests that Mats is acting downstream of Hpo. This is the first case that Ste20 family protein kinase-mediated phosphorylation of Mob is critical for regulating the catalytic activity of Ndr family protein kinase such as Wts. At this point, it is not clear how Mob proteins function to activate Ndr family kinases. Based on the results from recent studies of human Mob1 and Ndr family kinases, a potential mechanism is that Ndr family kinase is rapidly recruited by hMob1 to the plasma membrane for activation (Hergovich, 2005; Hergovich, 2006a). It is speculate dthat Hpo phosphorylation might facilitate Mats to associate to the membrane through an unknown mechanism, which in turn recruits Wts to the membrane as evidenced by the observation that Hpo phosphorylated Mats has an increased affinity to Wts. Subsequently, Wts is activated by phosphorylations mediated by protein kinases such as Hpo. Mats as a target of Hpo kinase, is able to associate with Hpo in a protein complex. Since Hpo/Mst1 kinase was not present in the Mats/Wts protein complex (Lai, 2005), it appears that Mats simultaneously cannot associate with Hpo and Wts in the same protein complex (Wei, 2007).

In addition to the membrane recruitment model, the data also support an active and more direct role of Mats in upregulating Wts kinase. From in vitro kinase assays, it was found that Hpo-mediated phosphorylation increases the affinity between Mats and Wts, as well as the ability of Mats to activate Wts kinase activity in the absence of any membrane structures. The results support a model in which Mats binding likely causes a conformational change of Wts for Wts activation. In the case of human Ndr kinase, an autoinhibitory effect of hNdr can be released by hMob1 binding (Bichsel, 2004), which presumably induces a conformational change in hNdr for its activation. Finally, it was found that Mats increases the steady level of Wts protein, which contributes to the increase in Wts activity. Further investigation is needed to understand how Mats is able to stabilize and/or increase the production of Wts protein (Wei, 2007).

Previous work has shown that Mats negatively regulates tissue growth by binding to another tumor suppressor Wts and subsequently activating the catalytic activity of Wts kinase (Lai, 2005). Since loss of mats function leads to tissue overgrowth and tumor development, it suggests that Wts alone is not sufficient to inhibit tissue growth in the absence of Mats. Therefore, Mats is an indispensable component of the Hpo pathway, and Wts activation is dependent not only on Hpo-mediated phosphorylation, but also on Mats binding. Further studies are needed to understand how exactly Wts activation is coordinated by Hpo phosphorylation and Mats binding. This work has provide evidence that Mats activation can be mediated by Hpo phosphorylation (Wei, 2007).

The Hpo signaling pathway plays an important role in growth inhibition and tumor suppression in Drosophila, and this pathway appears to be also critical for tissue growth control and tumor suppression in mammals. For instance, mammalian NF2 tumor suppressor is a homolog of Drosophila Mer and Ex proteins, which are upstream regulators of the Hpo signaling pathway. Moreover, loss of Lats1 function in mouse causes soft tissue sarcomas and ovarian tumors. Recently, it was found that hMats1 can functionally replace fly Mats to suppress tumor development, and Mats1 is mutated in mammalian tumors (Lai, 2005). Thus, mechanisms for the control of Hpo signaling might be commonly used across species, and understanding such mechanisms should provide insights into tumor development in mammals. As shown in this report, one mechanism by which Hpo functions to control tissue growth is to target Mats for phosphorylation, and, consequently, Mats is activated to upregulate Wts kinase. Because mammalian Hpo orthologs, Mst kinases, regulates hMats1 in a similar manner, this mechanism is likely used in mammalian cells as well. Therefore, by understanding how Hpo/Mst kinases regulate Mats and Wts/Lats in normal as well as tumor cells, valuable insights will be gained into tissue growth inhibition and tumor suppression (Wei, 2007).

EFFECTS OF MUTATION

mats1 is required for both Trc and Wts/Lts functions in vivo

The directed expression of a dominant negative Trc protein provides a sensitized system for identifying interacting genes (He, 2005a). Deficiencies for each of the fly mob genes enhances the wing hair phenotype that results from driving expression of UAS-trc^T453A using either ap-GAL4 or ptc-gal4. The strongest enhancement is seen with deficiencies for mats and Dmob2. These results suggested the possibility that all 4 Dmobs can redundantly interact with Trc, although it is possible that the interactions could be indirect or due to other genes in deleted regions. It is worth noting that such interactions are not common. When the Drosophila deficiency collection was screened for enhancement or suppression of ap-GAL4 UAS-trc^T453A, <10% of the Dfs showed an interaction (He, 2005b).

To confirm that the genetic interaction between trc and Df (mats) was due to the reduction in mats dose, two independent alleles were used. One was the null allele described by Lai (2005) (mats^e235), which is deleted for almost the entire coding region, and the other was a lethal PiggyBac insertion allele of mats [PBac{RB}CG13852^e03077] (this allele is referred to as mats^PB). Because this later allele has not been well characterized, the insertion was determined to be lethal over a deficiency for the region (Df(3R)Exel6191), and it failed to complement the recessive lethality of mats^e235 consistent with the lethality being due to the PB insertion. This mutation could be reverted using a source of PiggyBac transposase. It was found that both mats alleles dominantly enhance the trc dominant negative phenotype and this enhancement is lost in the PB revertant. It was also found that over expression of mats from a UAS-mats transgene (Lai, 2005; driven by ptc-Gal4) partially suppresses the multiple hair cell phenotype that results from driving expression of Trc-DN using ptc-Gal4. These dose responses argue that Mats activates Trc. Interestingly, it was found that heterozygosity for a wts mutation also enhances the Trc dominant negative phenotype, although somewhat less strongly (He, 2005b).

Evidence was also obtained for mats functioning with trc and fry using simple loss of function mutations. Wild-type flies or flies heterozygous for either trc, fry, or mats appear normal and only rarely (on fewer than 5% of wings) is even a single multiple hair cell seen. Flies that were heterozygous for two of these genes showed a slightly higher frequency of wings with one or a couple of multiple hair cells (often ~10%) but the increase was not routinely significant. However, almost half of the wings from flies that are heterozygous for all three genes (e.g., fry² trc¹+/+ + mats^PB) show a weak multiple hair cell phenotype, a significant increase. This genetic interaction is further support for the hypothesis that trc, fry, and mats function together in regulating wing hair development. In this assay no equivalent interaction with wts^3-17 was seen (He, 2005b).

Previous studies established that trc also has a larval denticle phenotype (Geng, 2000). In trc mutants the overall pattern of denticles is partly disorganized and many denticles are split. Split denticles are infrequent in wild-type larvae. The denticle pattern of mats^PB/mats^PB homozygous larvae is also disorganized and contains many split denticles. The number of split denticles is similar in mats^PB/Df larvae, suggesting that for this phenotype, mats^PB is a strong, near phenotypic null allele. The mats^e235 also showed a similar denticle phenotype. The phenotype of mats^PB homozygotes is slightly less severe than that of trc^P/trc^P larvae. Notably, trc^P mats^PB/trc^P mats^PB double mutant larvae do not have a more severe phenotype than the single mutants. This lack of additivity argues that trc and mats function in a common pathway during denticle development (He, 2005b).

Both mats alleles are larval lethals with death typically in the second or early third instar. To examine the phenotype of mats in wing cells, mosaics were generated using FLP/FRT. mats clones on the wing, leg, thorax, and head display two types of phenotype. The most notable is indistinguishable from those produced by clones of wts, suggesting that mats also functions with wts as has been shown by Lai (2005). On the wing, small clones produced bulges that can be seen at low magnification with a stereomicroscope. In mounted wings individual cell outlines are visible in the cuticle and the cells appear to have a bulging apical surface. The hairs are located on an elevated pedestal, a phenotype that is indistinguishable from those seen in wts clones. The hairs were often broader than normal. Particularly in other body regions clones were abnormally pigmented (either darker or lighter than normal, and there were outgrowths of clone tissue. In highly abnormal wing clones, evidence of clustered and split hairs were often seen that were typical of trc mutant clones. Some multiple hair cells were seen in very abnormal wts clones but this phenotype appears less severe (e.g., number of hairs per cell) than that seen with mats or trc. These observations suggest that mats functions with both Trc and Wts (He, 2005b).

mats^PB and mats^e235 clones in pupal wings were examined. Mutant mats cells are able to outcompete their neighbors and end up comprising most of the wing when clones are induced early. As was expected from the morphology of clones in adult wings, the pupal clones produce bulges in the wing and individual cells also often appear bulged. Clone cells stain more brightly for F-actin. This is true both in developing hairs and in the general apical cortex. This phenotype is clear-cut enough that it could be used as a convenient marker of mats mutant cells. These phenotypes were seen with both mats alleles tested. A similar, increase in actin staining was seen in wts clones. A similar, but perhaps less severe increase in staining, is seen in trc clones (He, 2005a). In some, but not all mats clones, large numbers of multiple hair cells can be seen. At later stages a circular pedestal of actin staining could be seen surrounding the base of the hair in mutant cells but not in surrounding wild-type cells. At still later stages the wild-type cells also had a circular pedestal of actin staining, suggesting that the mutant cells might be developmentally more advanced. Consistent with this possibility, in many clones hair initiation and outgrowth appear to be advanced in mats mutant cells compared with neighbors. This is also the case for wts clones, but it is the opposite of trc clones, in which hair development is often delayed (He, 2005a). Cells in mats clones have a smaller cross section so that the array of hairs appears denser, which is also the opposite of what is seen in trc clones, in which there is an increase in cross-sectional area (He, 2005a). Once again the phenotype of the wts clone cells resembles that seen for mats cells. Thus, for several wing phenotypes mats mutant cells resembled wts and not trc cells. Indeed, the mats phenotype is the opposite of trc for both cell area and the timing of hair morphogenesis (He, 2005b).

It has been found that the accumulation of Fry in wing cells is subject to feedback control that is dependent on Trc activity. Hence, in a trc mutant, increased Fry accumulation is found (He, 2005b). Several of the observations described above suggest the hypothesis that mats functions along with trc and is important for Trc activation. From this it is predicted that Fry accumulation would also be elevated in Dmob1 clones. Increased levels of Fry immunostaining were found in Dmob1 clone cells; this is consistent with the hypothesis. This is also seen in wts mutant clones, although the increase appears less dramatic (He, 2005b).

Tumorous overgrowth phenotypes are a consequence of mutations in a number of Drosophila genes. In several cases, such as lethal giant discs overgrown imaginal discs are found in late third instar larvae. To determine whether that is also the case for mats, mats/Df mutant larvae were examined. These larvae grow slowly and after 5 d of growth, when wild-type larvae begin to pupate, mats/Df larvae are the size of early third instar larvae. These larvae routinely die without growing substantially larger. When 5.5-5-d-old mats larvae were dissected, not evidence of tumors or overgrowth of imaginal or other tissues was found. Rather, the imaginal discs were approximately the size of those seen in 4-d larvae. However, the mats homozygous discs did not appear normal, since they were abnormally shaped and more folded than normal discs of this size (He, 2005b).

In a number of experiments involving mats or wts, what appeared to be spontaneous tumors or clones was observed. This was seen most often in flies that also carried reduced doses of Dmob3 and Dmob4. It is thought that these overgrowths are due to spontaneous mitotic recombination, because when the flies were also mutant for trc or fry, evidence of trc or fry clones was seen. The trc and fry clones were seen less frequently. This could be due to these genes being located more proximally on the chromosome than mats or wts, but it might also be due to the competitive advantage of mats and wts clones, resulting in these clones being larger and easier to detect. The basis for these clones is unclear but suggests genomic instability in mats and/or wts mutants (He, 2005b).

Control of cell proliferation and apoptosis by Mob as tumor suppressor Mats

A spontaneous lethal mutation was identified in the Drosophila gene mob as tumor suppressor (mats). From mutant clones generated in mats mosaic flies, large tumors can be induced in many organs including the head, notum, eye, wing, leg, antenna, and halteres. Thus, mats appears to function as a general inhibitor of tissue growth. The tumor cells formed unpatterned tissue with many folds on the surface. Using green fluorescent protein (GFP) to positively label mutant cells, it was apparent that mutant cells overproliferated in comparison to wild-type cells. Mosaic larval eye discs with mats mutant clones were apparently larger and folded in many areas. Overproliferation of mats mutant cells could not be explained by change in cell size, because the size of the mutant cells is not significantly different from that of wild-type cells. To directly examine the cell proliferation phenotype, Bromdeoxyuridine (BrdU) was incubated with eye discs to identify cells in the S phase. In wild-type larval eye discs, BrdU incorporation is evident in cells anterior to the morphogenetic furrow (MF) and in the second mitotic wave (SMW), which is a narrow stripe of dividing cells a few rows posterior to the MF. Consistent with an inhibitory role of mats in regulating cell proliferation, BrdU incorporation was elevated in mats mutant clones in eye discs. This phenotype was more evident in mats clones located in the MF and SMW regions. Moreover, immunostaining with anti-phospho histone H3 (PH3) antibody was carried out to identify mitotic cells in eye discs. In wild-type, PH3-positive cells can be found in the anterior and SMW regions, but not in the MF. They were rarely detected in the posterior region. In mats mosaic eye discs, more mitoses were observed in mats clones compared to neighboring wild-type cells, and mitoses can even sometimes occur in the MF. Excess mitoses were also found in mats clones located in the posterior margin. On the basis of these observations, it is concluded that mats is required to restrict cell proliferation and loss of mats function allows cells to divide at a time when they should exit the cell cycle (Lai, 2005).

To characterize the molecular nature of mats gene, deficiency and meiotic mapping experiments were carried out, and mats was localized in the 94A region on the third chromosome. By using a P transposon-mediated site-specific recombination method, mats was further mapped to a 13 kb region at 94A12. Molecular analysis of a candidate gene in this region, designated CG13852, revealed that a 428 bp Roo transposon sequence was inserted behind codon 84 to cause a premature termination. This first mutant allele of CG13852 is named roo in this study. A second mutant allele of this gene, referred to as e235, was generated by mobilizing a P transposon EP(3)3303 inserted approximately 2 kb downstream of CG13852. Like roo, e235 causes homozygous lethality at early second larval stage and induces tumors in somatic clones of mosaic flies. Based on the larval lethal phenotype, e235 failed to complement roo as well as deficiency chromosomes with the 94A12 region deleted. Sequence analysis of e235 has indicated that the second and third exons of CG13852 were deleted. Although the first exon is still intact, it encodes only the first four amino acids of CG13852. Thus, e235 is a null allele of CG13852 (Lai, 2005).

CG13852 encodes a 219-amino acid polypeptide, which is approximately 25 kDa in size. Due to CG13852's apparent homology to the Mob superfamily, it is renamed here Mats (Mob as tumor suppressor). As a member of the Mob superfamily, Mats has no significant homology with any other previously characterized protein domains. To test if CG13852 can rescue defects induced by mats mutations, an in vivo assay was established to allow CG13852 transgene expression only in mutant cells by using the MARCM system. It was found that expression of a full-length CG13852 cDNA in mutant cells suppresses tumor growth and pupal lethality associated with the mosaic flies. Moreover, ubiquitous expression of UAS-CG13852 driven by arm-Gal4 rescues larval lethality of mats homozygous mutants. These results further demonstrated that CG13852 corresponds to mats (Lai, 2005).

Cyclin E, a key regulator for the G1-S transition, is normally expressed in the second mitotic wave (SMW) of larval eye discs. In mats mosaic eye discs, Cyclin E levels are elevated in mutant clones located in the morphogenetic furrow (MF) and SMW regions. Moderate upregulation of Cyclin A and Cyclin B expression is also observed. Thus, an important mechanism for mats to control cell proliferation is to negatively regulate expression of key cell cycle regulators such as Cyclins. Interestingly, Cyclin E expression in mutant cells immediately anterior to the MF is much less elevated than that immediately posterior to the MF, suggesting that mats may use a different mechanism to restrict cell proliferation in cells anterior to the MF. The cell proliferation defects observed in mats mutants are similar to those caused by sav, wts, and hpo mutations (Lai, 2005).

Whether mats plays a role in regulating cellular differentiation in the developing eye was investigated. In larval eye discs, photoreceptors (R) and cone cells in mats clones appear to be specified normally. However, they fail to fully differentiate to generate ommatidia in adult. Defective retinal differentiation occurs at least at mid-pupal stages. Thus, mats is required for cellular differentiation during eye development. In contrast, hpo, sav, and wts do not significantly affect differentiation of retinal cells during eye development (Lai, 2005).

Apoptosis provides an important mechanism for the control of cell number and organ size. To test if mats plays a role in cell death control, expression of DIAP1 in eye discs was examined. DIAP1 is a caspase inhibitor essential for cell survival. Through immunostaining of mats mosaic eye discs, it was found that the level of DIAP1 protein is increased in mats clones. To examine if mats regulates diap1 at the transcriptional level, an enhancer trap line th^j5C8 was used, in which a lacZ reporter gene is inserted in diap1 and expression pattern of diap1-lacZ reflect that of the endogenous diap1 gene. It was found that expression of diap1-lacZ was elevated. Thus, mats is required to negatively regulate DIAP1 expression. To directly test the idea that mats promotes apoptosis, mats mutant clones were induced in larval eye discs that overexpress an apoptosis-promoting gene head involution defective (hid) in all cells behind the MF. As expected, expression of hid in a wild-type background increased apoptosis to cause a reduced eye phenotype. Notably, removal of mats function blocks hid-induced cell death and significantly suppresses the small eye phenotype. In these same tissues, developmental cell death is observed in regions anterior to the MF, where expression of the hid transgene is not induced. In these cases, apoptosis occurs only in wild-type tissues but not in mats mutant clones. Thus, mats is also required for developmentally programmed apoptosis. All together, these findings support a model that mats is required to facilitate cell death, and loss of mats’ apoptosis-promoting activity may contribute to tumor development (Lai, 2005).

Through phylogenetic analysis, four major groups of Mob proteins within the Mob superfamily were found. CG13852/Mats and its orthologs from animals and plants form the Mats gene family. Mats proteins are highly conserved. For instance, fly Mats and human Mats1 (also named Mob1A) are 87% identical. Even plant Mats orthologs are over 64% identical to fly Mats. In comparison, fly Mats shares no more than 40%-50% sequence identity with all other non-Mats Mob proteins from species such as yeast, fly, and humans. Such high levels of sequence conservation suggest that function of Mats proteins is evolutionarily conserved. To test functional homology, human Mats1 was introduced into Drosophila; it can effectively suppress tumor growth and rescue pupal lethality of mats mosaic flies. Thus, the growth inhibitory function of Mats has been conserved from insects to humans (Lai, 2005).

To further test the hypothesis that mammalian Mats functions as a tumor suppressor, 89 human and 8 mouse tumor-derived Mats cDNA sequences reported as expressed sequence tags (EST) in GenBank were examined and 2 Mats1 ESTs with disruptions in the coding region were identified, which were subsequently verified by sequencing analysis. In the first case, three nucleotides were deleted in a Mats1 cDNA (hMats1^ΔS6/7) derived from a human skin melanoma, which caused deletion of the sixth (or seventh) codon for Ser. This mutation greatly destabilized hMats1 because no hMats1 protein was detectable from hMats1^ΔS6/7-transfected human 293T cells. Another Mats1 cDNA (mMats1^Δ6-216) derived from a mouse mammary gland carcinoma contains a 38 bp insertion immediately downstream of the fifth codon that causes a premature termination. As expected, no mMats1 protein product was detected. These findings are exciting as they support a model that Mats1 may function as a tumor suppressor in mammals (Lai, 2005).

Delineation of a Fat tumor suppressor pathway

Recent studies in Drosophila of the protocadherins Dachsous and Fat suggest that they act as ligand and receptor, respectively, for an intercellular signaling pathway that influences tissue polarity, growth and gene expression, but the basis for signaling downstream of Fat has remained unclear. This study characterizes functional relationships among Drosophila tumor suppressors and identifies the kinases Discs overgrown and Warts as components of a Fat signaling pathway. fat, discs overgrown and warts regulate a common set of downstream genes in multiple tissues. Genetic experiments position the action of discs overgrown (dco) upstream of the Fat pathway component dachs, whereas warts acts downstream of dachs. Warts protein coprecipitates with Dachs, and Warts protein levels are influenced by fat, dachs and discs overgrown in vivo, consistent with its placement as a downstream component of the pathway. The tumor suppressors Merlin, expanded (ex), hippo, salvador (sav) and mob as tumor suppressor (mats) also share multiple Fat pathway phenotypes but regulate Warts activity independently. These results functionally link what had been four disparate groups of Drosophila tumor suppressors, establish a basic framework for Fat signaling from receptor to transcription factor and implicate Warts as an integrator of multiple growth control signals (Cho, 2006).

Since Dachs is required for loss of Wts protein in fat mutants, and Dachs encodes a large Myosin protein, a model was considered in which Dachs acts as a scaffold to link Wts to proteins that promote Wts proteolysis, analogous to the roles of Costal2 in Hedgehog signaling, or APC in Wnt signaling. This model predicts that Dachs should be able to bind to Wts. To evaluate this possibility, tagged forms of Dachs and Wts were coexpressed in cultured cells and assayed for coimmunoprecipitation. These experiments identified a specific and reproducible interaction between Dachs and Wts (Cho, 2006).

Recent studies have identified the transcriptional coactivator Yorkie (Yki) as a downstream component of the Hippo pathway and a substrate of Wts kinase activity. Phosphorylation of Yki by Wts inactivates Yki, and overexpression of Yki phenocopies wts mutation. The determination that the Fat tumor suppressor pathway acts through modulation of Wts thus predicts that Yki should also be involved in Fat signaling. When the influence of Yki overexpression was examined on Fat target genes, expression of Wg in the proximal wing, Ser in the proximal leg and fj in the wing and eye were each upregulated by Yki overexpression, consistent with the inference that Fat tumor suppressor pathway signaling acts through Yki (Cho, 2006).

In order to identify additional components of the Fat tumor suppressor pathway, advantage was taken of the observation that loss of fat in clones of cells is associated with an induction of Wingless (Wg) expression in cells just proximal to the normal ring of Wg expression in the proximal wing, reflective of its role in distal-to-proximal wing signaling. It was reasoned that this influence on Wg expression could be used to screen other Drosophila tumor suppressors for their potential to contribute to Fat signaling. Analysis of mutant clones in the proximal wing identified dco, ex, mats, sav, hpo and wts as candidate components of the Fat tumor suppressor pathway. As for fat, mutation of each of these genes is associated with induction of Wg expression specifically in the proximal wing, whereas Wg expression is not affected in more distal or more proximal wing cells. Although Wg expression often seems slightly elevated within its normal domain, the effect of these mutations is most obvious in the broadening of the Wg expression ring. The induction of Wg expression does not seem to be a nonspecific consequence of the altered growth or cell affinity associated with these mutations, since Wg expression is unaffected by expression of the growth-promoting microRNA gene bantam or by expression of genes that alter cell affinity in the proximal wing (Cho, 2006).

dco encodes D. melanogaster casein kinase I delta/epsilon. The overgrowth phenotype that gave the gene its name is observed in allelic combinations that include a hypomorphic allele, dco³, and it is this allele that is associated with induction of Wg. Null mutations of dco actually result in an 'opposite' phenotype: discs fail to grow, and clones of cells mutant for null alleles fail to proliferate. This is likely to reflect requirements for dco in multiple, distinct processes, as casein kinase I proteins phosphorylate many different substrates, and dco has been implicated in circadian rhythms, Wnt signaling and Hedgehog signaling (Cho, 2006).

Mer and ex encode two structurally related FERM domain-containing proteins. ex was first identified as a Drosophila tumor suppressor, whereas Drosophila Mer was first identified based on its structural similarity to human Merlin. Mutation of Mer alone causes only mild effects on imaginal disc growth, but Mer and ex are partially redundant, and double mutants show more severe overgrowth phenotypes than either single mutant. Consistent with this, elevation of Wg expression was observed in ex mutant clones (7/10 proximal wing clones induced Wg) and not in Mer mutant clones (0/8 clones), whereas Mer ex double mutant clones showed even more severe effects on Wg than ex single mutant clones. Because of the partial redundancy between Mer and ex, when possible, focus was placed for subsequent analysis on Mer ex double mutant clones (Cho, 2006).

Wts, Mats, Sav and Hpo interact biochemically, show similar overgrowth phenotypes and regulate common target genes. Mats, Sav and Hpo are all thought to act by regulating the phosphorylation state and thereby the activity of Wts. Mutation of any one of these genes is associated with upregulation of Wg in the proximal wing. The effects of sav (47/84 clones in the proximal wing induced Wg) and hpo (23/31 clones) were weaker than those of mats (19/19 clones) and wts (92/97 clones), but this might result from differences in perdurance or allele strength. Because sav, hpo and mats all act through Wts, focus for most of the subsequent analysis was placed on wts (Cho, 2006).

The observation that mutation of dco, Mer, ex, mats, sav, hpo or wts all share the distinctive upregulation of Wg expression in the proximal wing observed in fat mutants suggests that the functions of these genes are closely linked. To further investigate this, the effects of these tumor suppressors were characterized on other transcriptional targets of Fat signaling. Expression of the Notch ligand Ser is upregulated unevenly within fat mutant cells in the proximal region of the leg disc. A very similar upregulation occurred in dco³, Mer ex, and wts mutant clones. fj is a target of Fat signaling in both wing and eye imaginal discs, and fj expression was also upregulated in dco³, Mer ex, or wts mutant clones. The observation that these genes share multiple transcriptional targets in different Drosophila tissues implies that they act together in a common process (Cho, 2006).

The hypothesis that Fat pathway genes and Hippo pathway genes are linked predicts that not only should Fat target genes be regulated by Hippo pathway genes, but Hippo pathway target genes should also be regulated by Fat pathway genes. The cell cycle regulator CycE and the inhibitor of apoptosis Diap1 (encoded by thread) have been widely used as diagnostic downstream targets to assign genes to the Hippo pathway. Notably, then, clones of cells mutant for fat showed upregulation of both Diap1 and CycE protein expression. Genes whose expression is upregulated within fat mutant cells (such as wg, Ser and fj) have been shown previously to be induced along the borders of cells expressing either fj or dachsous (ds), and Diap1 is also upregulated around the borders of ds- or fj-expressing clones. That thread is affected by fat at a transcriptional level was confirmed by examining a thread-lacZ enhancer trap line. The regulation of Diap1 by the Hippo pathway is thought to be responsible for a characteristic eye phenotype in which an excess of interommatidial cells results from their failure to undergo apoptosis; an increase was also observed in interommatidial cells in fat mutant clones. Upregulation of both Diap1 and CycE is also observed in Mer ex double mutant clones. In dco³ mutant clones, consistent upregulation was detected only for Diap1, and CycE was upregulated only weakly and inconsistently. dco³ also has weaker effects on Wg and fj expression; the weaker effects of dco³ could result from its hypomorphic nature. ex has recently been characterized as another Hippo pathway target, and an ex-lacZ enhancer trap that is upregulated in wts or Mer ex mutant clones is also upregulated in fat or dco³ mutant clones. Analysis of ex transcription by in situ hybridization also indicated that ex is regulated by fat. Altogether, this analysis of Hippo pathway targets further supports the conclusion that the functions of the Fat pathway, the Hippo pathway and the tumor suppressors Mer, ex and dco are linked (Cho, 2006).

Genetic epistasis experiments provide a critical framework for evaluating the functional relationships among genes that act in a common pathway. The relationships was evaluated between each of the tumor suppressors linked to the Fat pathway and dachs, using both wing disc growth and proximal Wg expression as phenotypic assays. dachs is the only previously identified downstream component of the Fat tumor suppressor pathway. It acts oppositely to fat and is epistatic to fat in terms of both growth and gene expression phenotypes (Cho, 2006).

dachs is also epistatic to dco³ for overall wing disc growth and for proximal Wg expression. The epistasis of dachs to dco³ implies that the overgrowth phenotype of dco³ is specifically related to its influence on Fat signaling, as opposed to participation of dco in other pathways. By contrast to the epistasis of dachs to dco³, both wts and ex are epistatic to dachs for disc overgrowth phenotypes, and wts and Mer ex are epistatic to dachs in their influence on proximal Wg expression. Together, these epistasis experiments suggest that dco acts upstream of dachs, whereas Mer ex and wts act downstream of dachs (Cho, 2006).

Because wts and Mer ex have similar phenotypes, their epistatic relationship cannot be determined using loss-of-function alleles. However, overexpression of ex inhibits growth and promotes apoptosis, which suggests that ex overexpression affects ex gain-of-function. Clones of cells overexpressing ex are normally composed of only a few cells, and over time most are lost, but coexpression with the baculovirus apoptosis inhibitor p35 enabled recovery of ex-expressing clones. These ex- and p35-expressing clones were associated with repression of proximal Wg expression during early- to mid-third instar, as has been described for dachs², consistent with ex overexpression acting as a gain-of-function allele in terms of its influence on Fat signaling. In epistasis experiments using overexpressed ex and mutation of wts, wts was epistatic; Wg was induced in the proximal wing. Additionally, when wts is mutant, coexpression with p35 was no longer needed to ensure the viability and growth of ex-expressing clones, indicating that wts is also epistatic to ex for growth and survival. Consistent with this conclusion, others have recently described phenotypic similarities between Mer ex and hpo pathway mutants and have reported that hpo is epistatic to Mer ex (Cho, 2006).

When Fat was overexpressed, a slight reduction was detected in Wg expression during early- to mid-third instar, suggesting that overexpression can result in a weak gain-of-function phenotype. Clones of cells overexpressing Fat but mutant for dco³ still showed reduced Wg levels, whereas clones of cells overexpressing Fat but mutant for warts showed increased Wg levels. Although experiments in which the epistatic mutation is not a null allele cannot be regarded as definitive, these results are consistent with the conclusion that wts acts downstream of fat and suggest that dco might act upstream of fat (Cho, 2006).

The epistasis results described above suggest an order of action for Fat tumor suppressor pathway genes in which dco acts upstream of fat, fat acts upstream of dachs, dachs acts upstream of Mer and ex, and Mer and ex act upstream of wts. However, the determination that one gene is epistatic to another does not prove that the epistatic gene is biochemically downstream, as it is also possible that they act in parallel but converge upon a common target. Thus, to better define the functional and hierarchical relationships among these genes, experiments were initiated to investigate the possibility that genetically upstream components influence the phosphorylation, stability or localization of genetically downstream (that is, epistatic) components. Focus in this study was placed on the most downstream of these components, Wts. As available antibodies did not specifically recognize Wts in imaginal discs, advantage was taken of the existence of functional, Myc-tagged Wts-expressing transgenes (Myc:Wts) to investigate potential influences of upstream Fat pathway genes on Wts protein. In wing imaginal discs, Myc:Wts staining outlines cells, suggesting that it is preferentially localized near the plasma membrane, and it was confirmed that expression of Myc:Wts under tub-Gal4 control can rescue wts mutation. Notably, mutation of fat results in a reduction of Myc:Wts staining. As Myc:Wts is expressed under the control of a heterologous promoter in these experiments, this must reflect a post-transcriptional influence on Wts protein. fat does not exert a general influence on the levels of Hippo pathway components; fat mutant clones had no detectable influence on the expression of hemagglutinin epitope-tagged Sav (HA:Sav) (Cho, 2006).

The decrease in Wts protein associated with mutation of fat contrasts with studies of the regulation of Wts activity by the Hippo pathway, which have identified changes in Wts activity due to changes in its phosphorylation state. To directly compare regulation of Wts by Fat with regulation of Wts by other upstream genes, Myc:Wts staining was examined in ex, sav and mats mutant clones. In each of these experiments, the levels and localization of Myc:Wts in mutant cells was indistinguishable from that in neighboring wild-type cells (Cho, 2006).

Since Myc:Wts appears preferentially localized near the plasma membrane, it was conceivable that the apparent decrease in staining reflected delocalization of Wts, rather than destabilization. To investigate this possibility, Wts levels were examined by protein blotting. Antisera against endogenous Wts recognized a band of the expected mobility in lysates of wing imaginal discs or cultured cells, and this band was enhanced when Wts was overexpressed. The intensity of this band was reproducibly diminished in fat or dco³ homozygous mutant animals but was not diminished in fat or dco³ heterozygotes or in ex mutants. Conversely, levels of Hpo, Sav, Mer or Mats were not noticeably affected by fat mutation (Cho, 2006).

The determination that Wts is affected by Fat, together with the genetic studies described above, place Wts within the Fat signaling pathway, as opposed to a parallel pathway that converges on common transcriptional targets. Indeed, given that even hypomorphic alleles of wts result in disc overgrowth, the evident reduction in Wts levels might suffice to explain the overgrowth of fat mutants. As a further test of this possibility, Wts levels were examined in fat dachs double mutants. As the influence of Fat on gene expression and growth is absolutely dependent upon Dachs, if Fat influences growth through modulation of Wts, its influence on Wts levels should be reversed by mutation of dachs. Examination of Myc:Wts staining in fat dachs clones and of Wts protein levels in fat dachs mutant discs confirmed this prediction (Cho, 2006).

Prior observations, including the influences of fat and ds on gene expression, and the ability of the Fat intracellular domain to rescue fat phenotypes, suggested that Fat functions as a signal-transducing receptor. By identifying kinases that act both upstream (Dco) and downstream (Wts) of the Fat effector Dachs and by linking Fat to the transcriptional coactivator Yki, these results have provided additional support for the conclusion that Fat functions as a component of a signaling pathway and have delineated core elements of this pathway from receptor to transcription factor. Fat activity is regulated, in ways yet to be defined, by Ds and Fj. The influences of Fat on gene expression, growth, and cell affinity, as well as on Wts stability, are completely dependent on Dachs, indicating that Dachs is a critical effector of Fat signaling. Since Dachs can associate with Wts or a Wts-containing complex, it is suggested that Dachs might act as a scaffold to assemble a Wts degradation complex. The observations that Fat, Ds and Fj modulate the subcellular localization of Dachs, that Wts is preferentially localized near the membrane and that Dachs accumulates at the membrane in the absence of Fat, suggest a simple model whereby Fat signaling regulates Wts stability by modulating the accumulation of Dachs at the membrane and thereby its access to Wts. The working model is that dco³ is defective in the phosphorylation of a substrate in the Fat pathway, but the recessive nature of dco³, the genetic epistasis experiments, and biochemical experiments argue that this substrate is not Wts, and further work is required to define the biochemical role of Dco in Fat signaling (Cho, 2006).

In addition to identifying core components of the Fat pathway, the results establish close functional links between the Fat pathway, the Hippo pathway and the FERM-domain tumor suppressors Mer and Ex. The common phenotypes observed among these tumor suppressors can be explained by their common ability to influence Wts. However, they seem to do this in distinct ways, acting in parallel pathways that converge on Wts rather than a single signal transduction pathway. The Fat pathway modulates levels of Wts, apparently by influencing Wts stability. By contrast, the Hippo pathway seems to regulate the activity of Wts by modulating its phosphorylation state. Thus, Wts seems to act as an integrator of distinct growth signals, which can be transmitted by both the Fat pathway and the Hippo pathway. It has been suggested that Mer and Ex also act through the Hippo pathway, although present experiments cannot exclude the possibility that Mer and Ex act in parallel to Hpo. Moreover, it should be noted that Mats might regulate Wts independently of Hpo and Sav and hence function within a distinct, parallel pathway. Although it is simplest to think of parallel pathways, there is also evidence for cross-talk. fj and ex are both components and targets of these pathways. Thus, they can be regarded as feedback targets within their respective pathways, but their regulation also constitutes a point of cross-talk between pathways. Another possible point of cross-talk is suggested by the observation that levels of Fat are elevated within Mer ex mutant clones. Although the potential for cross-talk complicates assessments of the relationships between tumor suppressors, the observations that fat, dco³ and dachs affect Warts protein levels in vivo, whereas ex, hippo, sav and mats do not, argues that there are at least two distinct pathways that converge on Warts. This conclusion is also consistent with the observations that ex, hippo, sav and mats can influence Wts phosphorylation in cultured cell assays, but Fat, Dachs and Dco do not (Cho, 2006).

Although the Fat and Hippo pathways converge on Wts, Hippo pathway mutants seem more severe. Thus, hpo, wts or mats mutant clones show a distinctive disorganization and outgrowth of epithelial tissues that is not observed in fat mutant clones, and they show a greater increase in interommatidial cells. This difference presumably accounts for the previous failure to recognize the tight functional link between Fat and Hippo signaling, and it can be explained by the finding that Wts levels are reduced but not completely absent in fat mutant cells. Thus, fat would be expected to resemble a hypomorphic allele of wts rather than a null allele, and consistent with this, a hypomorphic allele, wts^P2, results in strong overgrowth phenotypes. The effects of Yki overexpression on growth and target gene expression can be even stronger than those of fat or wts mutations, which suggests that Yki levels become limiting when upstream tumor suppressors are mutant (Cho, 2006).

fat encodes a protocadherin, which in the past has led to speculation that its influences on growth and cell affinity might result from Fat acting as a cell adhesion molecule. However, all of the effects of fat on growth and affinity require dachs, which is also required for the effects of fat on transcription. Additionally, targets of Fat signaling include genes that can influence growth and affinity; recent studies identified an influence of fat on E-cadherin expression, and as describe in this study, Fat influences CycE and Diap1 expression. Thus, one can account for the influence of fat on growth and affinity by its ability to regulate gene expression. fat interacts genetically with other signaling pathways, including EGFR and Wnt, and in some cells Fat signaling also influences the expression of ligands (such as Wg and Ser) for other signaling pathways. Regulation of these ligands contributes to fat overgrowth phenotypes, but since clonal analysis indicates that fat is autonomously required for growth control in most imaginal cells, the principal mechanism by which fat influences growth presumably involves the regulation of general targets (Cho, 2006).

Normal tissue growth and patterning depend on a relatively small number of highly conserved intercellular signaling pathways. The Fat pathway is essential for the normal regulation of growth and PCP in most or all of the external tissues of the fly and also participates in local cell fate decisions. In this regard, its importance to fly development can be considered comparable to that of other major signaling pathways. Although the biological roles and even the existence of a Fat pathway in mammals remain to be demonstrated, there is clear evidence that the mammalian Warts homologs Lats1 and Lats2 act as tumor suppressors and that a mammalian Yorkie homolog, YAP, can act as an oncogene. Moreover, other genes in the Drosophila Fat pathway have apparent structural homologs in mammals. Thus, it is likely that mammals also have a Fat tumor suppressor pathway that functions in growth control (Cho, 2006).

The mob as tumor suppressor gene is essential for early development and regulates tissue growth in Drosophila

Studies in Drosophila have defined a new growth inhibitory pathway mediated by Fat (Ft), Merlin (Mer), Expanded (Ex), Hippo (Hpo), Salvador (Sav)/Shar-pei, Warts (Wts)/Large tumor suppressor (Lats), and Mob as tumor suppressor (Mats), which are all evolutionarily conserved in vertebrate animals. The Mob family protein Mats functions as a coactivator of Wts kinase. This study shows that mats is essential for early development and is required for proper chromosomal segregation in developing embryos. Mats is expressed at low levels ubiquitously, which is consistent with the role of Mats as a general growth regulator. Like mammalian Mats, Drosophila Mats colocalizes with Wts/Lats kinase and cyclin E proteins at the centrosome. This raises the possibility that Mats may function together with Wts/Lats to regulate cyclin E activity in the centrosome for mitotic control. While Hpo/Wts signaling has been implicated in the control of cyclin E and diap1 expression, this study found that it also modulates the expression of cyclin A and cyclin B. Although mats depletion leads to aberrant mitoses, this does not seem to be due to compromised mitotic spindle checkpoint function (Shimizu, 2008).

Mats is essential for normal development; mats mutants stop their growth at the second instar larval stage and eventually die. In fact, this growth retardation phenotype facilitated identification of mats^roo and mats^e235 mutant larvae for DNA sequence analysis. Using mats^e235 allele and the P-element-induced allele mats^PB, it has been shown that mats homozygotes and hemizygotes grow slowly and their imaginal discs are much smaller than that of wild-type larvae at the same age. mats mutant cells in mosaic tissues acquire growth advantage likely through comparison and competition with neighboring wild-type cells. In contrast, the absence of wild-type cells in homozygous mats mutant animals renders no competitive growth advantage to mutant cells. The mechanism by which mats mutants acquire growth advantage in the context of mosaic tissue still needs to be investigated. mats mutant embryos missing both maternal and zygotic mats functions failed to hatch, indicating that mats is essential for embryonic development. By analyzing mitotic cells, it was found that maternally mats-depleted embryos show aberrant DNA segregation such that uneven amounts of DNA are segregated toward opposing centrosomes. However, this does not appear to be due to the compromised function of mitotic spindle checkpoint, since mats mutant tissue still accumulate M-phase cells in response to inhibition of mitotic spindle formation by colcemid treatment. Thus, mats is not required for mitotic spindle checkpoint, unlike mps1 (Shimizu, 2008).

Cyclin E is a critical cell cycle regulator. Through a Cdk2-dependent mechanism, cyclin E-Cdk2 plays a critical role in accelerating G₁-S transition in the cell cycle. As a general rule, cyclin E is tightly regulated during the cell cycle by Cdk2 and GSK-mediated phosphorylation and subsequent degradation. A nondegradable cyclin E mutant can cause extra rounds of DNA synthesis and polyploidy, and overexpression of cyclin E is frequently detected in tumor cells exhibiting polyploidy. Intriguingly, cyclin E is a centrosomal protein that functions to promote S-phase entry and DNA synthesis in a Cdk2-independent manner (Matsumoto, 2004). Loss of cyclin E expression in the centrosome inhibits DNA synthesis, whereas ectopic expression of cyclin E in the centrosome accelerates S-phase entry. Thus, the centrosome is an important subcellular organelle for cyclin E to regulate cell proliferation, and the level and activity of cyclin E in centrosomes must be tightly controlled. The fact that Mats and Wts colocalize with cyclin E at the centrosome raises the possibility that Mats may function together with Wts kinase to regulate cyclin E function in the centrosome for mitotic control. In support of this hypothesis, loss-of-function mutations in mats increase the levels of cyclin E protein and both gain- and loss-of-function mutant alleles of cyclin E modulate the eye phenotypes caused by Wts overexpression. Although Mats/Wts-mediated inhibition of cyclin E could occur through Yki to regulate cyclin E transcription, a direct control of cyclin E at the protein level would allow a rapid response to an upstream signal (Shimizu, 2008).

The fact that both Mats and Wts show a intracellular localization pattern very similar to that of their respective yeast relatives Mob1 and Dbf2 suggests that their function is conserved. This conservation may extend to mammals; human LATS1, LATS2, and MOB1A (MATS2) also localize at the centrosome. In addition, localization at the bud neck/midbody appears to be conserved in humans. Interestingly, such centrosomal localization of Mats and Wts does not seem to rely on Wts kinase activity as kinase-inactive Wts and Mats can be still localized at the centrosome. To examine whether endogenous Mats protein localizes at the centrosome, embryo immunostaining was done with Mats antibodies. As in larval tissues, expression of Mats protein in developing embryos does not exhibit any obvious pattern and Mats expression level is low and ubiquitous. Although centrosomal localization of endogenous Mats protein has not been shown, likely due to some technical problems, Mats (CG13852/Mob4) has been recently reported to be a centrosomal protein (Shimizu, 2008 and references therein).

Both loss- and gain-of-function analysis supports a model in which cyclin E and diap1 are critical downstream targets of Hpo/Wts signaling. Evidence in this report suggests that Hpo/Wts signaling may also target cyclin A and cyclin B. Consistent with this notion, elevated levels of cyclin B were found in ex mutant cells. In addition, wts has been shown to be required for a negative control of cyclin A but not cyclin B expression. In humans, LATS1 was shown to be a negative regulator of Cdc2/cyclin A and to function at the G₂/M-phase transition, while LATS2 affects cyclin E/Cdk2 activity and regulates G₁/S phase passage. Thus, the ability of Hpo/Wts signaling to target cyclin genes important for cell cycle progression appears to be evolutionarily conserved (Shimizu, 2008).

REFERENCES

Search PubMed for articles about Drosophila Mats

Bichsel, S. J., (2004). Mechanism of activation of NDR (Nuclear Dbf2-related) protein kinase by the hMOB1 protein. J. Biol. Chem. 279: 35228-35235. 15197186

Bothos, J., et al. (2005). Human LATS1 is a mitotic exit network kinase. Cancer Res 65: 6568-6575. PubMed Citation: 16061636

Cho, E., Feng, Y., Rauskolb, C., Maitra, S., Fehon, R. and Irvine, K. D. (2006). Delineation of a Fat tumor suppressor pathway. Nat. Genet. 38(10): 1142-50. 16980976

Colman-Lerner, A., Chin, T. E., and Brent, R. (2001). Yeast Cbk1 and Mob2 activate daughter-specific genetic programs to induce asymmetric cell fates. Cell 107: 739-750. 11747810

Devroe, E., Erdjument-Bromage, H., Tempst, P. and Silver, P.M. (2004). Human Mob proteins regulate the Ndr1 and Ndr2 serine-threonine kinases. J. Biol. Chem. 279: 24444-24451. 15067004

Emoto, K, et al. (2004). Control of dendritic branching and tiling by the Tricornered-kinase/Furry signaling pathway in Drosophila sensory neurons. Cell 119: 245-256. PubMed Citation: 16061636

Geng, W., He, B., Wang, M. and Adler, P. N. (2000). The tricornered gene, which is required for the integrity of epidermal cell extensions, encodes the Drosophila nuclear DBF2-related kinase. Genetics 156: 1817-1828. 11102376

Giot, L. et al. (2003). A protein interaction map of Drosophila melanogaster. Science 302: 1727-1736. 14605208

Hariharan, I. K. (2006). Growth regulation: a beginning for the hippo pathway. Curr. Biol. 16: R1037-1039. PubMed Citation: 17174912

Hay, B. A. and Guo, M. (2003). Coupling cell growth, proliferation, and death. Hippo weighs in. Dev. Cell 5: 361-363. 12967554

He, Y., Fang, X., Emoto, K., Jan, Y. N. and Adler, P. N. (2005a). The tricornered Ser/Thr protein kinase is regulated by phosphorylation and interacts with furry during Drosophila wing hair development. Mol. Biol. Cell 16(2): 689-700. 15591127

He, Y., Emoto, K., Fang, X., Ren, N., Tian, X., Jan, Y. N. and Adler, P. N. (2005b). Drosophila Mob family proteins interact with the related tricornered (Trc) and warts (Wts) kinases. Mol. Biol. Cell 16(9): 4139-52. 15975907

Hergovich, A., Bichsel, S. J. and Hemmings, B. A. (2005). Human NDR kinases are rapidly activated by MOB proteins through recruitment to the plasma membrane and phosphorylation. Mol. Cell. Biol. 25: 8259-8272. PubMed Citation: 16135814

Hergovich, A., Schmitz, D. and Hemmings, B. A. (2006a). The human tumour suppressor LATS1 is activated by human MOB1 at the membrane. Biochem. Biophys. Res. Commun. 345: 50-58. PubMed Citation: 16674920

Hergovich, A., Stegert, M. R., Schmitz, D. and Hemmings, B. A. (2006b). NDR kinases regulate essential cell processes from yeast to humans. Nat. Rev. Mol. Cell Biol. 7: 253-264. PubMed Citation: 16607288

Hou, M. C., Salek, J. and McCollum, D. (2000). Mob1p interacts with the Sid2p kinase and is required for cytokinesis in fission yeast. Curr. Biol. 10: 619-622. PubMed Citation: 10837231

Hou, M. C., Wiley, D. J., Verde, F., and McCollum, D. (2003). Mob2p interacts with the protein kinase Orb6p to promote coordination of cell polarity with cell cycle progression. J. Cell Sci. 116: 125-135. 12456722

Hou, M. C., Guertin, D. A. and McCollum, D. (2004). Initiation of cytokinesis is controlled through multiple modes of regulation of the Sid2p-Mob1p kinase complex. Mol. Cell. Biol. 24(8): 3262-76. 15060149

Komarnitsky, S. I., et al. (1998). DBF2 protein kinase binds to and acts through the cell cycle regulated MOBI protein. Mol. Cell. Biol. 18: 2100-2107. 9528782

Lai, Z.-C., et al. (2005). Control of cell proliferation and apoptosis by Mob as tumor suppressor Mats. Cell 12: 675-685. 15766530

Lee, S. E., et al. (2001). Order of function of the budding-yeast mitotic exit-network proteins Tem1, Cdc15, Mob1, Dbf2, and Cdc5. Curr. Biol. 11: 784-788. PubMed Citation: 11378390

Luca, F. C. and Winey, M. (1998). MOB1, an essential yeast gene required for completion of mitosis and maintenance of ploidy. Mol. Biol. Cell 9: 29-46. 9436989

Mah, A. S., Jang, J., and Deshaies, R. J. (2001). Protein kinase Cdc15 activates the Dbf2-Mob1 kinase complex. Proc. Natl. Acad. Sci. 98: 7325-7330. 11404483

Matsumoto, Y. and Maller, J. L. (2004). A centrosomal localization signal in cyclin E required for Cdk2-independent S phase entry. Science 306: 885-888. PubMed Citation: 15514162

Moreno, C. S., Lane, W. S., Pallas, D. C. (2001). A mammalian homolog of yeast MOB1 is both a member and a putative substrate of striatin family-protein phosphatase 2A complexes. J. Biol. Chem. 276: 24253-24260. PubMed Citation: 11319234

Nelson, B., et al. (2003). Ram: A conserved signaling network that regulates ace2p transcriptional activity and polarized morphogenesis. Mol. Biol. Cell 14: 3782-3803. 12972564

Ponchon, L., et al. (2004). NMR solution structure of Mob1, a mitotic exit network protein and its interaction with an NDR kinase peptide. J. Mol. Biol. 337(1): 167-82. 15001360

Rothenberg, M. E. and Jan, Y. N. (2003). Cell biology: the hippo hypothesis. Nature 425: 469-470. 14523431

Ryoo, H. D. and Steller, H. (2003). Hippo and its mission for growth control, Nat. Cell Biol. 5: 853-855. 14523394

Shimizu, T., Ho, L. L. and Lai, Z. C. (2008). The mob as tumor suppressor gene is essential for early development and regulates tissue growth in Drosophila. Genetics 178(2): 957-65. PubMed Citation: 18245354

Stavridi E. S., et al., (2003). Crystal structure of a human Mob1 protein: toward understanding Mob-regulated cell cycle pathways. Structure 11: 1163-1170. 12962634

Tamaskovic, R., Bichsel. S. J. and Hemmings, B. A. (2003). NDR family of AGC kinases -- essential regulators of the cell cycle and morphogenesis. FEBS Lett. 546: 73-80. 12829239

Weiss, E.L., Kurischko, C., Zhang, C., Shokat, K., Drubin, D.G. and Luca, F.C. (2002). The Saccharomyces cerevisiae Mob2p-Cbk1p kinase complex promotes polarized growth and act with the mitotic exit network to facilitate daughter cell-specific localization of Ace2p transcription factor. J. Cell Biol. 158: 885-900. 12196508

Wei, X., Shimizu, T. and Lai, Z. C. (2007). Mob as tumor suppressor is activated by Hippo kinase for growth inhibition in Drosophila. EMBO J. 26(7): 1772-81. PubMed Citation: 17347649

Yang, X., et al. (2004). LATS1 tumour suppressor affects cytokinesis by inhibiting LIMK1. Nat. Cell Biol. 6: 609-617. PubMed Citation: 15220930

date revised: 10 August 2009

Home page: The Interactive Fly © 2017 Thomas Brody, Ph.D.