InteractiveFly: GeneBrief

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

BIOLOGICAL OVERVIEW

In the absence of mutations in any of the mammalian Homer genes, their in vivo roles remain unknown. A single gene has been identified in Drosophila encoding a protein homologous to the mammalian Homer proteins (Kato, 1998; Xiao, 1998 and Diagana, 2002). Drosophila Homer is enriched in the nervous system, where it is localized to the endoplasmic reticulum (ER) and targeted to dendritic processes. Genetic evidence is provided that homer is required for the function of the neural networks controlling locomotor activity and behavioral plasticity (Diagana, 2002).

Mammalian Homer proteins have been proposed to play a role in synaptogenesis, synapse function, receptor trafficking, and axon pathfinding. The Drosophila gene homer, the single Homer-related gene in fly, has been isolated and characterized. Using anti-Homer antibody it has been shown that Homer is expressed in a broad range of tissues but is highly enriched in the CNS. Similar to its mammalian counterpart, Drosophila Homer localizes to the dendrites and the endoplasmic reticulum (ER). This subcellular distribution is dependent on an intact Enabled/Vasp homology 1 domain, suggesting that Homer must bind to one or more of its partners for proper localization. Flies homozygous for a mutation in homer are viable and show coordinated locomotion, suggesting that Homer is not essential for basic neurotransmission. However, homer mutants display defects in behavioral plasticity and the control of locomotor activity. These results argue that in the CNS, Homer-related proteins operate in the ER and in dendrites to regulate the development and function of neural networks underlying locomotor control and behavioral plasticity (Diagana, 2002).

Proteins of the Homer family have been implicated in synaptogenesis, signal transduction, receptor trafficking, and axon pathfinding (Xiao, 2000; Foa, 2001; Thomas, 2002). In mammals, three independent genes (Homer-1, -2, and -3) encode at least six Homer proteins through differential splicing (Kato, 1998; Xiao, 1998). Members of the Homer family are expressed in various tissues but appear to be enriched in the CNS, where they have partially overlapping domains of expression (Brakeman, 1997; Xiao, 1998). Homer proteins are bipartite, consisting of an N-terminal Enabled/Vasp homology 1 (EVH1) domain and a C-terminal coiled-coil (CC) domain that mediates self-association (Brakeman, 1997; Kato, 1998; Tu, 1998; Xiao, 1998; Tadokoro, 1999). The EVH1 domain binds to group I metabotropic glutamate receptors (mGluRs) and their downstream effectors, the inositol-triphosphate receptor (InsP3R), by interacting with a proline-rich motif (PPxxF) found in these proteins (Brakeman, 1997; Tu, 1998, 1999; Diagana, 2002 and references therein).

One can envision a model in which Homer proteins, via their ability to self-associate, modulate group I mGluR function by mediating the formation of a multimolecular complex required for local and fast increase of Ca²⁺ concentration during mGluR activation (Xiao, 2000). Supporting this notion is the finding that Homer proteins regulate the intracellular trafficking of mGluRs (Roche, 1999; Ango, 2000). Further modulation is provided by Homer 1a, one of the proteins encoded by the Homer 1 gene, that consists of the EVH1 domain without the CC domain required for multimerization. Homer 1a is upregulated during synaptic activity (Brakeman, 1997; Xiao, 1998) and is capable of attenuating mGluR-evoked intracellular calcium release in vitro, presumably by disruption of a putative mGluR-Homer-InsP3R multimeric complex (Tu, 1998; Diagana, 2002 and references therein).

Homer proteins also bind to Shank/ProSAP, a postsynaptic protein that is part of a complex including the NMDA-type glutamate receptors. Recently, Shank has been implicated in the regulation of dendritic spine morphology and synaptic function, and this regulation is dependent on Shank binding to Homer (Sala, 2001). Therefore, Homer-related proteins may be part of a large multimolecular complex that modulates the structural and functional plasticity of glutamatergic synapses (Naisbitt, 1999; Sala, 2001; Diagana, 2002 and references therein).

It has been proposed that in vertebrates, Homer-related proteins might regulate synaptic plasticity possibly underlying learning and memory via the modulation of mGluR function. To test for a possible role of Homer in behavioral plasticity, the performance of homer^R102 mutant males was evaluated in a courtship conditioning assay, an associative learning paradigm in Drosophila. Courtship is a plastic behavior that can be conditioned by previous experience. Male flies display a complex and robust courtship behavior toward a female in response to olfactory, visual, and tactile cues. After exposure to a nonreceptive mated female, wild-type males will repress their level of courtship. This repression is sustained during subsequent exposure to a receptive, virgin female. Courtship repression is thus a conditioned behavior and is thought to depend on the association of positive stimuli with an aversive chemosensory signal from the mated female (Diagana, 2002).

In the courtship conditioning assay, individual males were placed with a nonreceptive mated female in a conditioning chamber for 1 hr. The mated female was then removed, and after 2 min, each male was individually tested for levels of courtship with an anesthetized virgin female. The amount of courtship displayed by these 'trained' males was monitored over a 10 min period and compared with courtship levels of 'naive' males that had been manipulated identically in the conditioning protocol, except without the nonreceptive mated female. In the assay, no defects were detected in the sequence or the length of the different steps of male courtship behavior of homer mutants, and thus homer is not essential for the execution of the courtship behavior (Diagana, 2002).

homer mutants display defects in behavioral plasticity. homer⁺ control males show a statistically significant reduction of courtship after training, as do homer^R102/homer⁺ heterozygotes. In contrast, homer^R102 trained males show no significant reduction of courtship. In addition, although there is no significant difference in the amount of courtship displayed by naive homer^R102 and naive homer⁺ flies, the trained homer^R102 mutant males show a significantly higher level of courtship when compared with the trained homer⁺ males. Collectively, these results demonstrate that homer^R102 mutants show behavioral plasticity deficits and fail to form and/or retain the conditioning by the nonreceptive mated female (Diagana, 2002).

To determine whether homer^R102 mutants show defects in the acquisition of conditioning during training, the amount of courtship displayed was monitored by tested males over the first and last 10 min of the conditioning period. Both homer^R102 mutant flies and homer⁺ controls show a statistically significant reduction of courtship level after conditioning by the mated female. homer^R102 mutant males show higher initial and final courtship levels when compared with homer⁺ controls. Median values for initial and final courtship levels are, respectively, 331 and 133 sec for homer^R102 mutants and 160 and 12 sec for the homer⁺ controls. These data demonstrate that homer^R102 mutants do suppress courtship behavior after conditioning by the mated female but show higher levels of initial and final courtship (Diagana, 2002).

homer mutant flies do not show olfactory defects but display deficits in the control of locomotor activity. The conditioned repression of male courtship is dependent on the perception of an aversive chemosensory cue secreted by the nonreceptive mated female. Thus, defects in olfaction could be an explanation for the poor conditioning of homer^R102 mutants in the courtship conditioning assay. To evaluate the olfactory competence of homer^R102 adult flies, the chemosensory jump assay was used. When suddenly exposed to chemical vapors, a wild-type fly exhibits an escape response consisting of a jump. The chemosensory jump response (CJR) of homer mutant flies to two different chemicals was tested. The CJR to propionic acid of homer^R102 mutant flies is similar to that of homer⁺ control flies, and the CJR of homer^R102 mutant flies to benzaldehyde is actually higher than that of homer⁺ control flies, demonstrating that homer^R102 mutant flies are not severely defective in olfaction. However, homer mutants do display deficits in their control of locomotor activity. homer^R102 mutant flies show a higher level of spontaneous locomotor activity when compared with homer⁺ homozygous flies, a feature that is consistent with homer^R102 flies showing higher courtship levels (Diagana, 2002).

The colocalization of Homer and Synaptotagmin (see Drosophila Syntaxin) in the dorsal-most region of the neuropil suggests that Homer is localized to regions containing a concentration of synapses. By expressing an epitope-tagged Homer, evidence has been provided that the Drosophila Homer is targeted to dendrites, similar to the targeting described for vertebrate Homer proteins. Within neuronal cell bodies, the Drosophila Homer colocalizes with a marker for the ER. In transfected cells, Homer 1b has similarly been shown to be localized in the ER compartment (Roche, 1999), a subcellular localization thought to be functionally relevant because Homer-related proteins are capable of binding the ER-resident receptor InsP3R (Tu, 1998). Thus, the evolutionary conservation of the subcellular localization of Homer-related proteins suggests that their functional roles might also have been conserved (Diagana, 2002 and references therein).

Mammalian Homer binds to several proteins, including Shank, Group I mGluRs, and the InsP3 receptor (Tu, 1998; Xiao, 1998; Tu, 1999). Homer specifically binds to a short proline-rich motif, PPxxF, present in each of these proteins (Tu, 1998). In support of a functional role for this binding is the finding that mutations in the PPxxF motif of mGluR5 that abolish the binding to Homer in vitro also abolish the ER retention of mGluR5 in cells cotransfected with mGluR5 and Homer (Roche, 1999). In addition, mutation of the Homer binding site of mammalian Shank disrupts the Shank-dependent targeting of Homer 1b to dendritic spines (Sala, 2001). Evidence is provided that Drosophila Homer likely requires binding to at least one of its putative partners to be properly localized. Mutation of amino acids within the EVH1 domain predicted to be required for PPxxF binding abolishes the localization of Homer to the ER and its targeting to dendrites. Similar mislocalization has been described in cell culture for deletion of the EVH1 domain of Homer 2a (Shiraishi, 1999). At present, it is not known which protein regulates Drosophila Homer subcellular localization. Given the finding that the EVH1 domain of Homer binds Drosophila Shank, it is possible that Shank might have a function similar to what has been implicated in vertebrates (Sala, 2001). Further genetic and biochemical studies will be required to address this question (Diagana, 2002).

In terms of candidate receptors in Drosophila to which Homer might bind, there are several putative mGluRs in the genome sequence database. Of these, only one, DmGluRA, has been characterized, and it has been pharmacologically classified as a group II mGluR (Parmentier, 1996). Consistent with this classification, the cytosolic domain of DmGluRA contains no Homer-binding motif and fails to bind Homer in yeast two-hybrid assay. In vertebrates, two ER-resident proteins, the ryanodine receptor (RyR) and the InsP3R, both control intracellular Ca²⁺ stores and contain Homer-binding motifs in their cytosolically disposed N termini (Tu, 1998). Moreover, the InsP3R coimmunoprecipitates with Homer proteins (Tu, 1998). There is a single InsP3R and a single RyR in Drosophila, both of which contain putative Homer-binding motifs. It will be of interest to determine whether Homer binds in vivo to either of these receptors (Diagana, 2002).

The finding that homer mutant flies are able to walk, fly, and exhibit an escape response to visual stimuli suggests that Homer has no essential role in vision or the performance of basic motor skills. However, homer mutants are hyperactive for both spontaneous locomotion and courtship behavior, implicating Homer in the control of locomotor activity. The homer mutants also exhibit deficits in behavioral plasticity, as assayed in a courtship conditioning paradigm. In this type of assay, the Drosophila memory mutant amnesiac shows specific defects in the retention of courtship conditioning. Similarly, homer mutants are capable of suppressing courtship behavior toward a nonreceptive mated female, but this suppression is not retained when subsequently tested with the virgin female. At present the possibility cannot be ruled out that the defective control of locomotor activity interferes somehow with the formation and retention of the conditioning. Addressing this will require the genetic separation of homer function in the control of locomotor activity and behavioral plasticity (Diagana, 2002).

In contrast to recent results suggesting a function for Homer-related proteins in axon pathfinding (Foa, 2001), no obvious pathfinding defects were detected in the CNS of homer mutant embryos. It is still possible, however, that Drosophila Homer could play a developmental role during synaptogenesis and that loss of Homer function results in structural defects undetectable at the light microscopy level used in this study. An alternative, but not mutually exclusive, possibility is that Drosophila Homer functions in the modulation of neuronal circuits by regulating synaptic plasticity, perhaps through the modulation of mGluR signaling. Loss-of-function of group I mGluRs in mice causes deficits in spatial learning and locomotor control, and although the basic synaptic physiology in these mutant animals is unaffected, aspects of synaptic plasticity are impaired. Vertebrate Shank and Homer have also been implicated in the regulation of the structure and function of the synaptic junction (Sala, 2001). Evidence for a physical interaction between Drosophila Homer and Shank raises the possibility of a similar synaptic function for these proteins in flies (Diagana, 2002).

REGULATION

Protein Interactions

Through the EVH1 domain, mammalian Homer-related proteins interact in vivo with at least three different group of proteins, Shank/ProSAP, Group I mGluRs, and the InsP3R (Brakeman, 1997; Tu, 1998; Naisbitt, 1999; Tu, 1999). To determine whether such interactions might exist in Drosophila, a yeast two-hybrid bait was constructed consisting of the Homer EVH1 domain fused to the DNA-binding domain of the GAL4 transcription factor. A mutated Homer EVH1 bait (mutEVH) was constructed by introducing several amino acid substitutions in the EVH1 domain. Noteworthy is the substitution of the second glycine in the conserved GLGF motif for a histidine. The crystal structure of the vertebrate Homer-1 EVH1 domain bound to its binding-peptide has shown that any residue other than glycine at this position (position 89 of Homer-1) would be likely to sterically interfere with binding of Homer to its partners (Diagana, 2000).

From a yeast two-hybrid screen of a Drosophila embryonic library using the Homer EVH1 domain as bait, the single fly ortholog of the vertebrate Shank proteins was isolated. Like the vertebrate Shank proteins, Drosophila Shank contains the Homer-binding consensus motif PPxxF near its C terminus. The Homer EVH1 domain shows interaction with the C terminus of Drosophila Shank in the yeast two-hybrid assay. In contrast to the wild-type Homer EVH1 domain, mutEVH fails to interact with Shank C terminus, consistent with the introduced EVH1 mutations abolishing binding. As expected, neither the Homer EVH1 domain nor mutEVH showed any interaction with the C terminus of the characterized Drosophila Group II DmGluRA (Parmentier, 1996), which like vertebrate Group II mGluRs lacks any Homer binding site consensus sequence. These results argue that the ability of Homer to bind Shank is dependent on an intact EVH1 domain and further suggest that at least one of the protein-protein interactions mediated by the EVH1 domain of Homer-related proteins has been evolutionarily conserved (Diagana, 2000).

Homer colocalizes with the synaptic marker Synaptotagmin (see Drosophila Syntaxin) in the dorsal-most region of the neuropil. The regional colocalization of Homer and Synaptotagmin dorsally in the VNC neuropil suggests that Homer is concentrated in synapse-rich regions. Because little or no Homer was detected in the axons of motor neurons and sensory neurons, it seemed possible that Homer might be localized to dendrites. To investigate this possibility an epitope-tagged version of Homer (Homer-myc) was created by fusing six c-myc epitopes to the C-terminal end of the protein. Homer-myc-green fluorescent protein (GFP) was also created, in which six myc epitopes plus the GFP were similarly fused to the C terminus. The two epitope-tagged versions of Homer gave identical results in all the described experiments. The GAL4/UAS transactivation system was used. Homer-myc and Homer-myc-GFP were cloned downstream of the UAS regulatory sequences in the pUASt transformation vector, and multiple transformant lines were generated for each. To express Homer-myc in single neurons, the RCC-GAL4 line, which stochastically expresses GAL4 in a small subset of identified neurons, including aCC, pCC, and RP2, was used. When expressed in the aCC neuron using the RCC GAL4 driver, Homer-myc staining is detected primarily in the ipsilateral and contralateral dendritic arborizations, with little or no staining seen in the axon. This result indicates that Homer is primarily targeted to dendritic processes and thus might function postsynaptically (Diagana, 2002).

To determine the importance of the EVH1 domain for Homer subcellular localization, mutEVH-Homer-myc was created by introducing in the EVH1 domain the same amino acid substitutions that disrupt binding of the Homer EVH1 domain to Shank. When expressed in the aCC neuron using the RCC-GAL4 driver, mutEVH-Homer-myc fails to label the dendrites and instead remains in the cell body. When expressed in the PNS using the pan-neuronal C155 GAL4 driver, Homer-myc has a punctate distribution in many neuronal cell bodies, similar to the pattern characteristic of endogenous Homer. Double-immunofluorescence staining has confirmed that in these cells, Homer-myc colocalizes with the ER compartment marker BiP. In contrast to Homer-myc, mutEVH-Homer-myc fails to show the characteristic punctate ER distribution in PNS neurons and instead is evenly distributed throughout the cytoplasm. Taken together, these results demonstrate that the EVH1 domain of Homer is required for its subcellular localization and further suggests that this localization may be mediated by one of the binding partners of Homer (Diagana, 2002).

Roles of Bifocal, Homer, and F-actin in anchoring Oskar to the posterior cortex of Drosophila oocytes

Transport, translation, and anchoring of osk mRNA and proteins are essential for posterior patterning of Drosophila embryos. Homer and Bifocal act redundantly to promote posterior anchoring of the osk gene products. Disruption of actin microfilaments, which causes delocalization of Bifocal but not Homer from the oocyte cortex, severely disrupts anchoring of osk gene products only when Homer (not Bifocal) is absent. The data suggest that two processes, one requiring Bifocal and an intact F-actin cytoskeleton and a second requiring Homer but independent of intact F-actin, may act redundantly to mediate posterior anchoring of the osk gene products (Babu, 2004).

Both Bif and Hom show asymmetric localization at the apical cortex of embryonic neuroblasts, indicating that these F-actin binding proteins may be involved in neuroblast asymmetric divisions. However, animals lacking both the maternal and zygotic components of either gene are fertile, viable, and show no obvious defects in embryonic CNS development. This prompted construction of double mutants of bif and hom. However, although double homozygous mutant females are viable, they show defects in oogenesis, with the vast majority of the eggs produced remaining unfertilized, as judged by the lack of staining in eggs using an antibody directed against the sperm tail. In the few fertilized embryos that do undergo development, the numbers of Vasa positive germ cells are drastically reduced, suggesting possible defects in the function or localization of posterior determinants during oogenesis (Babu, 2004).

Analyses of bif and hom single mutants as well as double mutant oocytes indicate that the two genes act in a redundant manner for the correct anchoring of posteriorly but not anteriorly localized molecules. In stage 10 oocytes, posterior group molecules, including oskar (osk) RNA, the two isoforms of Osk proteins, Staufen (Stau), and a fusion protein in which ß-galactosidase has been fused to the N-terminal extension of the long form of the Osk protein (referred to as Osk-ßGal, used as a marker for the long form of the Osk protein, which has been shown to have a role in the posterior cortical maintenance of Osk), are all localized as tight posterior cortical crescents in wild-type oocytes. In most double mutant oocytes, these molecules, when detectable, are present largely at the posterior region; however in contrast to wild-type oocytes, they show a diffuse distribution that extends into regions of the posterior cytoplasm distinctly interior to the posterior cortex. In about 30% of the cases, Osk or Stau protein cannot be detected. The defects seen in the oocytes of double mutants are essentially absent in the single mutant oocytes. These findings indicate that whereas bif and hom are individually dispensable, together they are required for the localization of the posterior components of the oocytes. These defects in localization are specific for the posterior group molecules, because the anterior/dorsal localization of Gurken and anterior localization of bicoid RNA are unaffected (Babu, 2004).

Not surprisingly, staining with anti-Bif and anti-Hom antibodies indicates that both proteins are expressed in the oocyte. Bif localizes to the oocyte cortex in a manner very similar to that seen for F-actin. The staining seen with the anti-Hom antibody is highly punctated and, although localization is cortically enriched, Hom is also present in the cytoplasm. The cortical staining seen in wild-type oocytes (and nurse cells) is absent in mutant oocytes stained with the corresponding antibodies, confirming the specificity of both antibodies and that the mutant alleles do not produce detectable amounts of protein. Since hom^LL17 is a complete deletion of the coding region and bif^R47 removes a significant portion of the coding region, they are both likely to be null alleles (Babu, 2004).

Several observations indicate that the defect in posterior localization of the osk gene products is due to defective anchoring and not transport or translation of osk RNA. Osk RNA and Osk proteins as well as Stau are localized normally in stage 9 double mutant oocytes. Consistent with this, both the F-actin and microtubule cytoskeletons in the double mutants are indistinguishable from those in the wild-type oocytes. Not only does the polarity of the microtubules appear normal, as assayed using a kinesin heavy chain (khc) ß-Gal marker, the cytoskeleton-dependent cytoplasmic streaming is also absent in stage 9 oocytes and occurs normally in stage 10 double mutant oocytes, as is seen with wild-type oocytes. Taken together, these observations indicate that the double mutant oocytes retain, at a gross level, normal cytoskeletal structure. They can transport osk mRNA to the posterior cortex, and translate it appropriately, but do not maintain the posterior anchoring of the osk gene products (Babu, 2004).

An intact F-actin cytoskeleton is thought to be required for asymmetric protein localization in several contexts. In the oocyte, loss or reduction of the actin binding proteins moesin and tropomyosin have been shown to affect the posterior anchoring of Osk in the oocyte and the embryo, respectively. To assess the requirement for an intact microfilament cytoskeleton for the anchoring of osk RNA and proteins in the oocyte, the localization of these molecules was examined in wild-type oocytes treated with an actin depolymerizing drug, Latrunculin A (Lat A). Following treatment with 20 µm Lat A, cortical F-actin in the oocytes was largely undetectable with phalloidin, yet, unexpectedly, both Osk proteins (short and Osk-ßGal) and osk RNA remain localized to the posterior cortex of the great majority of wild-type oocytes from stage 9-10B. This was seen even when the oocytes were overstained for the osk RNA: in around 17% of the Lat A-treated wild-type oocytes, mild defects in anchoring are observed. osk RNA and proteins show a diffuse localization at the posterior cortex. However, in no cases were they seen concentrated in the cytoplasm or had they become delocalized or undetectable (as seen in the hom/bif double mutant oocytes) as would be expected if an intact F-actin cytoskeleton were to be an absolute requirement for the normal anchoring of Osk. These mild effects on protein localization in the oocyte are in distinct contrast to those seen in embryonic neuroblasts where severe and high-penetrance defects in asymmetric protein localization are observed following disruption of microfilaments. These observations suggest that the role of microfilaments in the anchoring of proteins to the cortex may differ in the different cellular contexts (Babu, 2004).

Additional experiments were performed to ascertain whether the mild effects on Osk posterior anchoring following disruption of microfilaments are peculiar to Lat A treatment. In fact, the posterior cortical localization of Osk remained in the great majority of oocytes even after treatment with cytochalasin D (CD), Lat A followed by CD, CD followed by Lat A, and up to 100 µM Lat A. These results are consistent with previous reports of CD disruption of F-actin, for example. However, they indicate that an intact F-actin cytoskeleton, although required for the normal posterior anchoring of Osk in a small proportion of oocytes, is probably not the only factor involved in normal anchoring of Osk to the posterior cortex. There are at least two possible explanations for these observations. First, a small amount of residual F-actin might remain even after sequential treatment with Lat A and CD, which is sufficient to anchor Osk normally in a small fraction of the drug-treated oocytes. Alternatively, there may be other factors besides an intact F-actin cytoskeleton, which can, in parallel, contribute toward the posterior anchoring of osk RNA and proteins (Babu, 2004).

The requirement for intact microfilaments on Hom and Bif localization was assessed. Both Bif and Hom localize to the cortex (and in the case of Hom also the cytoplasm) of wild-type oocytes. Following depolymerization of F-actin with Lat A, such that cortical F-actin becomes undetectable in the oocyte, Hom localization appears unchanged from the wild-type pattern in all oocytes, but Bif becomes highly diffuse with essentially no detectable enrichment at the cortex. This appears to be an effect on Bif localization and not stability, because the levels of the protein are not reduced as judged by Western blot analysis. Treating oocytes with colchicine, which disrupts the microtubules, did not affect either Hom or Bif localization in the oocyte. These findings raise the possibility that bif function might be dependent on intact F-actin; however, Hom localization is Lat A-insensitive, suggesting that its function in the oocyte may not require intact microfilaments. However, the possibility cannot be excluded that Lat A treatment allows for the retention of a small amount of the F-actin cytoskeleton, and this is stabilized in some way by Hom (Babu, 2004).

These data raise the possibility that there might be two processes, one that is microfilament-dependent and requires Bif, and another which is not dependent on intact microfilaments and requires Hom. Either process is sufficient to anchor the osk gene products to the posterior cortex of the great majority of the oocytes. One prediction of this hypothesis is that hom should be necessary to anchor posterior components in the absence of intact F-actin. Indeed, when hom single mutant oocytes were treated with Lat A, a large amount of cytoplasmic Osk was found at stage 9 and 10 near the posterior pole, and there was a large reduction in the Osk-ßGal signal. This could indicate that the loss of the longer Osk isoform may be the primary defect seen in Lat A-treated hom mutants; this longer Osk isoform is known to be essential for osk RNA and protein anchoring. The defects induced by depolymerizing F-actin in hom oocytes are similar to but more severe than those seen in bif;hom double mutant oocytes. This is probably due to the fact that F-actin disruption also leads to premature streaming in stage 9 and enhanced streaming in stage 10 oocytes, thus accentuating the effects of the loss of Osk anchoring at the posterior cortex (Babu, 2004).

The above results demonstrate that disruption of F-actin in the absence of hom function disrupts anchoring of the osk gene products. Similar results were obtained when hom mutants were treated with just CD or treated successively with Lat A and CD or vice versa. CD does not cause loss of F-actin as seen with phalloidin staining, and causes changes in the cortical F-actin as well as causing some of the F-actin to be seen in the cytoplasm. This latter effect is not seen with Lat A. This could be attributed to the difference in the mechanism of action between CD and Lat A. However, despite this difference between CD and Lat A, the effects of these drugs singly or in combination on Osk posterior anchoring are similar, causing mild defects in Osk posterior anchoring in only one-fifth of the treated wild-type oocytes and severe defects in the great majority of treated hom oocytes (Babu, 2004).

A second prediction is that disruption of microfilaments in the absence of bif should not affect anchoring of Osk. Indeed, most wild-type and bif single mutant oocytes treated with Lat A show largely wild-type anchoring of the Osk gene products, similar to Lat A treatment of wild-type oocytes. These results are consistent with the notion that Bif functions in an F-actin-dependent manner in maintaining Osk to the posterior of the oocyte (Babu, 2004).

If there are two independent mechanisms that act redundantly for normal Osk anchoring at the posterior cortex, then it would follow that the bif;hom double mutants in the absence of an intact F-actin cytoskeleton would show a phenotype similar to that of hom single mutants treated with Lat A, and not a more severe phenotype. This is indeed the case that is observed on testing osk RNA and Osk proteins in Lat A-treated double mutant oocytes (Babu, 2004).

In light of the finding that Hom posterior cortical localization remains unchanged following F-actin disruption, its ability to localize Osk to the posterior may be because it forms a complex with Osk. Co-immunoprecipitation experiments, using Drosophila ovarian extracts, indicate that Hom and Osk form a complex in vivo. Further, the stability of this complex is not dependent on an intact F-actin cytoskeleton (Babu, 2004).

The maintenance of Osk at the posterior of the oocyte may be mediated by two distinct mechanisms, either of which is sufficient, at least for the great majority of the oocytes. One mechanism does not require an intact F-actin cytoskeleton, and Hom seems to be an important player in this process. Hom can complex with Osk, and the stability of this complex is not dependent on an intact F-actin cytoskeleton. The second mechanism requires an intact F-actin cytoskeleton. Bif seems to be required for this mechanism. Overexpression of Bif can promote actin polymerization in cultured cells. Since it is also known that F-actin forms a complex with Bif in Drosophila embryonic lysates and that Bif binds directly to F-actin filaments in vitro, it is possible that Bif acts to stabilize actin filaments. In this scenario its absence may cause subtle changes in the F-actin cytoskeleton that may affect its capacity to anchor molecules at the cortex when hom is absent. In contrast, hom can function and is required to anchor Osk in the absence of an intact F-actin cytoskeleton or in the absence of bif function. Only when both mechanisms are disrupted in the oocyte, either through the simultaneous disruption of both hom and bif, or when F-actin is disrupted in the absence of hom, do the osk gene products fail to remain anchored to the posterior cortex (Babu, 2004).

Recent studies showed that Drosophila moesin is essential to link the cortical F-actin to the oocyte cell membrane. When moesin function is compromised, cortical F-actin can detach from the cell membrane and "fall" into the oocyte cytoplasm, and this results in the mislocalization of Osk. The effects of loss of moesin function on the localization of both Bif and Hom were examined; in these mutant oocytes, where the cortical F-actin detaches from the mutant oocyte cell membrane, components of both of the proposed anchoring pathways, Hom and Bif, also detach. These observations are consistent both with the Osk mislocalization phenotype seen in moesin mutant oocytes and with the model. It will be interesting to identify additional molecules involved in these separate pathways and to elucidate the mechanisms that are required to localize Hom to the posterior cortex of the oocyte in the absence of an intact actin cytoskeleton (Babu, 2004).

Mg²⁺ block of Drosophila NMDA receptors is required for long-term memory formation and CREB-dependent gene expression

NMDA receptor (NMDAR) channels allow Ca²⁺ influx only during correlated activation of both pre- and postsynaptic cells; a Mg²⁺ block mechanism suppresses NMDAR activity when the postsynaptic cell is inactive. Although the importance of NMDARs in associative learning and long-term memory (LTM) formation has been demonstrated, the role of Mg²⁺ block in these processes remains unclear. Using transgenic flies expressing NMDARs defective for Mg²⁺ block, it was found that Mg²⁺ block mutants are defective for LTM formation but not associative learning. It was demonstrated that LTM-dependent increases in expression of synaptic genes, including homer, staufen, and activin, are abolished in flies expressing Mg²⁺ block defective NMDARs. Furthermore, it was shown that genetic and pharmacological reduction of Mg²⁺ block significantly increases expression of a CREB repressor isoform. These results suggest that Mg²⁺ block of NMDARs functions to suppress basal expression of a CREB repressor, thus permitting CREB-dependent gene expression upon LTM induction (Miyashia, 2012).

Although the mechanism through which Mg²⁺ block restricts NMDAR activity is well known, the cellular and behavioral functions of Mg²⁺ block have not been extensively studied. In this study, transgenic flies expressing dNR1^N63IQ to show that Mg²⁺ block is important for formation of LTM. Previous studies of hypomorphic mutants have shown that NMDARs are required for both learning and LTM. In contrast, our Mg²⁺ block mutants do not have learning defects. This suggests that although Ca²⁺ influx through NMDARs is important for learning, inhibition of influx during uncorrelated activity is not. Notably, elav/dNR1^N63IQ flies have slightly enhanced learning. Consistent with this result, NMDAR-dependent induction of hippocampal LTP is enhanced in the absence of external Mg²⁺. In the current studies, Mg²⁺-block-defective dNR1 was overexpressed in an otherwise wildtype background, so it cannot be definitively concluded that Mg²⁺ block is dispensable for learning. However, electrophysiology experiments indicate that Mg²⁺ block is abolished in the flies at physiological potentials. Furthermore, it was demonstrated that expression of Mg²⁺-block-defective dNR1 rescues learning defects in dNR1 hypomorphs, consistent with a model in which Mg²⁺ block is not required for learning. Interestingly, the dNR1^N63IQ transgene does not rescue the semilethality of dNR1 hypomorphs, suggesting that Mg²⁺ block has an essential biological function unrelated to learning (Miyashia, 2012).

The results suggest that Mg²⁺-block-dependent suppression of NMDAR activity and Ca²⁺ influx at the resting state is critical for LTM formation. Supporting this idea, chronic reduction of NMDAR-mediated Ca²⁺ influx at the resting state has been shown to enhance long-term synaptic plasticity. Extending these results, it was found that Mg²⁺ block is required for CREB-dependent gene expression during LTM formation. A CREB-dependent increase in staufen expression upon spaced training is essential for LTM formation, and this study shows that Mg²⁺ block is required for this increase. Two other genes, activin and homer, were identified that are expressed upon LTM induction in a CREB-dependent manner. It is proposed that all three genes are maintained in an LTM-inducible state by Mg²⁺-block-dependent inhibition of CREB repressor, and it was shown that the amount of increase in expression of dCREB2-b in Mg²⁺ block mutants correlates with the ability of dCREB2-b to suppress LTM. The 4-fold increase in dCREB2-b protein in Mg²⁺ block mutant flies is comparable to the increase in dCREB2-b in heat-shocked hs-dCREB2-b flies showing equivalent defects in LTM (Miyashia, 2012).

Next the homer gene was further characterized and it was determined to be required specifically for LTM but not for learning or ARM. It was determined that spaced training increases HOMER expression in several brain regions, including the antennal lobes, lateral protocerebrum, protocerebral bridge, and calyx of the MBs. This increase does not occur in the absence of Mg²⁺ block. Significantly, when Mg²⁺ block is abolished by dNR1^N63IQ expression, specifically in the MBs, increased Homer expression is suppressed in the MBs but not in other regions, including the protocerebral bridge, indicating that Mg²⁺ block regulates CREB repressor and LTM-associated gene expressions in a cell autonomous manner (Miyashia, 2012).

Electrophyisiological experiments demonstrate that 20 mM Mg²⁺ is sufficient to block Drosophila NMDAR currents at the resting potential (-80 mV). Although this concentration is higher than the concentrations needed to block mammalian NMDARs, the Mg²⁺ concentration in Drosophila hemolymph has been shown by various groups to be between 20 and 33 mM, which is correspondingly higher than the Mg²⁺ concentration reported in mammalian plasma. In mammals, Mg²⁺ concentration is higher in cerebrospinal fluid than in plasma, further suggesting that the 20 mM Mg²⁺ concentration used in this study is likely to be within the physiologically relevant range (Miyashia, 2012).

An N/Q substitution at the Mg²⁺ block site of mammalian NR1 disrupts Mg²⁺ block and reduces Ca²⁺ permeability, while a W/L substitution in the TM2 domain of NR2B disrupts Mg²⁺ block and increases Mg²⁺ permeability. This raises the possibility that Mg²⁺-block-independent changes in channel kinetics and Mg²⁺ permeability may be responsible for the effects observed in the dNR1^N63IQ-expressing flies. While this possiblity cannot be completely ruled out, increases were observed in dCREB-2b protein in wild-type neurons in Mg²⁺-free conditions, indicating that disruption of Mg²⁺ block, rather than changes in other channel properties, causes increased CREB repressor expression and decreased expression of LTM-associated genes (Miyashia, 2012).

A chronic elevation in extracellular Mg²⁺ enhances Mg²⁺ block of NMDARs, leading to upregulation of NMDAR activity and potentiation of NMDA-induced responses at positive membrane potentials (during correlated activity) (Slutsky, 2010). This raised the possibility that the Mg²⁺ block mutations may cause a downregulation of NMDAR-dependent signaling and decreased NMDA-induced responses at positive membrane potentials. Since this study recorded NMDA-induced responses from various sizes of cells, it was not possible to directly compare amplitudes of NMDA-induced responses between elav/dNR1^wt cells and elav/dNR1^N63IQ cells. However, training-dependent increases in ERK activity, required for CREB activation, occurred normally in both elav/dNR1^wt cells and elav/dNR1^N63IQ cells, while it was significantly suppressed in dNR1 hypomorphs. These results suggest that the Mg²⁺ block mutations do not alter NMDA-induced responses at positive membrane potentials (Miyashia, 2012).

Similar to dNR1 Mg²⁺ block mutants, dNR1 hypomorphic mutants also have defects in CREB-dependent gene expression upon LTM formation. However, dNR1 hypomorphs and Mg²⁺ block mutants are likely to have opposing effects on Ca²⁺ influx. While hypomorphic dNR1 mutants should have decreased Ca²⁺ influx during spaced training because of a reduction in the number of dNMDARs, elav/dNR1^N63IQ flies are unlikely to have this effect. Conversely, while elav/ dNR1^N63IQ flies should have increased Ca²⁺ influx during the resting state when uncorrelated activity is likely to occur, dNR1 hypomorphs should not. Supporting a model in which dNR1 hypomorphs and Mg²⁺ block mutants inhibit LTM-dependent gene expression through different mechanisms, it was shown that Mg²⁺ block mutants increase basal expression of dCREB2-b repressor while NMDAR hypomorphs do not. Conversely, the data indicating that NMDAR hypomorphs are defective for training dependent increases in ERK activity, while elav/dNR1^N63IQ flies are not. These data fit a model in which there may be two equally important requirements for NMDARs in regulating LTM-dependent transcription. First, during correlated, LTM-inducing stimulation, a large Ca²⁺ influx through channels, including NMDARs, may be required to activate kinases, including ERK, necessary to activate CREB. dNR1 hypomorphs are defective for this process. However, a second and equally important requirement for NMDARs may be to inhibit low amounts of Ca²⁺ influx during uncorrelated activity to maintain the intracellular environment in a state conducive to CREB-dependent transcription. Mg²⁺ block is required for this process (Miyashia, 2012).

Although it is unclear what types of uncorrelated activity are suppressed by Mg²⁺ block, one type may be spontaneous, action potential (AP)-independent, single vesicle release events (referred to as 'minis'). Supporting this idea, an increase in dCREB2-b was observed in cultured wild-type brains in Mg²⁺-free medium in the presence of TTX, which suppresses AP-dependent vesicle releases but does not affect minis. In addition, a significant increase was observed in cytosolic Ca²⁺, [Ca²⁺]i, in response to 1 mM NMDA in the presence of extracellular Mg²⁺ in neurons from elav/dNR1^N63IQ pupae. In neurons from transgenic control and wild-type pupae, which have an intact Mg²⁺ block mechanism, 1 mM NMDA does not cause Ca²⁺ influx and membrane depolarization. The concentration of glutamate released by minis is on the order of 1 mM at the synaptic cleft, suggesting that an increase in frequency of mini-induced Ca²⁺ influx due to decreased Mg²⁺ block may contribute to the increase in dCREB2-b in elav/ dNR1^N63IQ flies (Miyashia, 2012).

Correlated, AP-mediated NMDAR activity has been proposed to facilitate dCREB2-dependent gene expression by increasing activity of a dCREB2 activator. The present study suggests that, conversely, Mg²⁺ block functions to inhibit uncorrelated activity, including mini-dependent Ca²⁺ influx through NMDARs, which would otherwise cause increased dCREB2-b expression and decreased LTM. Other studies have also suggested opposing roles of AP-mediated transmitter release and minis. For activity-dependent dendritic protein synthesis, local protein synthesis is stimulated by AP-mediated activity and inhibited by mini activity. In the case of NMDARs, the opposing role of low Ca²⁺ influx in inhibiting CREB activity must be suppressed by Mg²⁺ block for proper LTM formation (Miyashia, 2012).

DEVELOPMENTAL BIOLOGY

Embryonic

Whole-mount in situ hybridization was performed using of a full-length DIG-labeled homer antisense probe. A widespread, low-level expression of homer was detected during early embryogenesis. Beginning at stage 12 the levels of homer mRNA increase in the developing CNS and PNS, such that by late stage 16, homer RNA is enriched in the nervous system. Close examination of in situ hybridization performed on dissected embryos shows that within the CNS and PNS of stage 16 embryos, most if not all neurons express homer mRNA (Diagana, 2002).

A polyclonal antibody was raised against Homer that recognizes a 47 kDa protein on a Western blot of protein extracts from adult fly heads. This size is in agreement with the protein predicted from the sequence of homer cDNAs. Two results confirm the specificity of the anti-Homer antibody: (1) no signal can be detected in protein extracts of flies homozygous for a homer loss-of-function allele; (2) the antibody specifically recognizes a truncated version of Homer lacking the C terminus (HomerDeltaC) in protein extracts of flies that express HomerDeltaC in the CNS from a transgene (Diagana, 2002).

Using the anti-Homer antibody, immunofluorescence staining of Drosophila embryos was performed. In wild-type embryos, high levels of Homer expression are detected in the CNS and PNS, with a low level of expression detectable in other tissues, including the epidermis, the gut, and the somatic muscles. As expected, all staining was found to be abolished in embryos homozygous for the homer^R102 mutation. Homer expression in the nervous system is maintained throughout development into adulthood, where in the brain it is expressed at high levels within the lamina and the medulla neuropil of the optic lobes and at lower levels in the central brain neuropil (Diagana, 2002).

Within the embryonic ventral nerve cord (VNC), the majority of staining is concentrated in the neuropil regions of each neuromere. Confocal analysis of immunostainings (using anti-Homer and antibodies to HRP) that recognize a neuronal surface epitope present on all axons and dendrites, reveals that Homer is highly enriched in the dorsal-most region of the neuropil. To examine whether this enrichment might represent localization of Homer to synaptic regions, triple labeling for anti-HRP, Homer, and the synaptic protein Synaptotagmin was performed. Synaptotagmin is similarly localized to the dorsal region of the neuropil and Homer extensively colocalizes with it. These results suggest that in Drosophila, Homer is targeted to synapses, similar to the localization of vertebrate Homer proteins (Diagana, 2002 and references therein).

A striking feature of Homer expression is the punctate pattern seen in neuronal cell bodies, a pattern reminiscent of the staining of Golgi stacks. By performing double immunostaining with anti-Homer and antibodies that label the Golgi, Homer labeling was found to partially overlap with the Golgi staining but does not fully colocalize with it. This suggested that Homer is localized in the closely juxtaposed ER. To label the ER compartment, a monoclonal antibody was generated against the rat ER-resident protein that recognizes a Drosophila protein retained in the ER. In double immunostainings anti-Homer staining was found to overlap extensively with the ER marker anti-BiP, arguing for Homer localization in the ER compartment (Diagana, 2002).

EVOLUTIONARY HOMOLOGS

Identification of mammalian Homer genes

Spatial localization and clustering of membrane proteins is critical to neuronal development and synaptic plasticity. Recent studies have identified a family of proteins, the PDZ proteins, that contain modular PDZ domains and interact with synaptic ionotropic glutamate receptors and ion channels. PDZ proteins are thought to have a role in defining the cellular distribution of the proteins that interact with them. A novel dendritic protein, Homer, contains a single, PDZ-like domain and binds specifically to the carboxy terminus of phosphoinositide-linked metabotropic glutamate receptors. Homer is highly divergent from known PDZ proteins and seems to represent a novel family. The Homer gene is also distinct from members of the PDZ family in that its expression is regulated as an immediate early gene and is dynamically responsive to physiological synaptic activity, particularly during cortical development. This dynamic transcriptional control suggests that Homer mediates a novel cellular mechanism that regulates metabotropic glutamate signalling (Brakeman, 1997).

Vesl-1S (186 amino acids, also called Homer) is a protein containing EVH1- and PDZ-like domains whose expression in the hippocampus is regulated during long term potentiation (LTP), one form of synaptic plasticity thought to underlie memory formation. This study reports additional members of the Vesl/Homer family of proteins, Vesl-1L and Vesl-2. Vesl-1L (366 amino acids), a splicing variant of Vesl-1S, shares N-terminal 175 amino acids with Vesl-1S and contains additional amino acids at the C terminus. Vesl-2 (354 amino acids) is highly related to Vesl-1L in that both contain EVH1- and PDZ-like domains at the N terminus (86% conservation) and an MCC (mutated in colorectal cancer)-like domain and a leucine zipper at the C terminus. In contrast to vesl-1S, no changes are observed in the levels of vesl-1L and vesl-2 mRNAs during dentate gyrus LTP. All these proteins interact with metabotropic glutamate receptors (mGluR1 and mGluR5) as well as several hippocampal proteins in vitro. Vesl-1L and Vesl-2, but not Vesl-1S, interact with each other through the C-terminal portion that is absent in Vesl-1S. Vesl-1L and Vesl-2 may mediate clustering of mGluRs at synaptic junctions. It is proposed that Vesl-1S may be involved in the structural changes that occur at metabotropic glutamatergic synapses during the maintenance phase of LTP by modulating the redistribution of synaptic components (Kato, 1998).

Homer is a neuronal immediate early gene (IEG) that is enriched at excitatory synapses and binds group 1 metabotropic glutamate receptors (mGluRs). A family of Homer-related proteins is derived from three distinct genes. Like Homer IEG (now termed Homer 1a), all new members bind group 1 mGluRs. In contrast to Homer 1a, new members are constitutively expressed and encode a C-terminal coiled-coil (CC) domain that mediates self-multimerization. CC-Homers form natural complexes that cross-link mGluRs and are enriched at the postsynaptic density. Homer 1a does not multimerize and blocks the association of mGluRs with CC-Homer complexes. These observations support a model in which the dynamic expression of Homer 1a competes with constitutively expressed CC-Homers to modify synaptic mGluR properties (Xiao, 1998).

A developmentally regulated Homer/Vesl isoform, Cupidin (Homer 2a/Vesl-2Delta11), was isolated from postnatal mouse cerebellum using a fluorescent differential display strategy. The strongest expression of Cupidin is detected in the cerebellar granule cells at approximately postnatal day 7. Cupidin is enriched in the postsynaptic density fraction, and its immunoreactivity is concentrated at glomeruli of the inner granular layer when active synaptogenesis occurs. Cupidin protein can be divided into two functional domains: the N-terminal portion that is highly conserved among Homer/Vesl family proteins, and the C-terminal portion that consists of a putative coiled-coil structure, including several leucine zipper motifs. The N-terminal fragment of Cupidin, which is able to associate with metabotropic glutamate receptor 1 (mGluR1), also interacts with F-actin in vitro. In keeping with this, F-actin immunocytochemically colocalizes with Cupidin in cultured cerebellar granule cells, and a Cupidin-mGluR1-actin complex was immunoprecipitated from crude cerebellar lysates using an anti-Cupidin antibody. The C-terminal portion of Cupidin binds to Cdc42, a member of Rho family small GTPases, in a GTP-dependent manner in vitro, and Cupidin functionally interacts with activated-Cdc42 in a heterologous expression system. Together, these findings indicate that Cupidin may serve as a postsynaptic scaffold protein that links mGluR signaling with actin cytoskeleton and Rho family proteins, perhaps during the dynamic phase of morphological changes that occur during synapse formation in cerebellar granule cells (Shiraishi, 1999).

Alternative Homer transcripts

Three Homer genes regulate the activity of metabotropic glutamate receptors mGluR1a and mGluR5 and their coupling to releasable intracellular Ca2+ pools and ion channels. Only the Homer 1 gene evolved bimodal expression of constitutive (Homer 1b and c) and immediate early gene (IEG) products (Homer 1a and Ania 3). The IEG forms compete functionally with the constitutive Homer proteins. The complex expression of the Homer 1 gene, unique for IEGs, focused the attention of this study on gene organization. In contrast to most IEGs, which have genes that are <5 kb, the Homer 1 gene was found to span approximately 100 kb. The constitutive Homer 1b/c forms are encoded by exons 1-10, whereas the IEG forms are encoded by exons 1-5 and parts of intron 5. RNase protection has demonstrated a >10-fold activity-dependent increase in mRNA levels exclusively for the IEG forms. Moreover, fluorescent in situ hybridization has documented that new primary Homer 1 transcripts are induced in neuronal nuclei within a few minutes after seizure, typical of IEGs, and that Homer 1b-specific exons are excluded from the activity-induced transcripts. Thus, at the resting state of the neurons, the entire gene is constitutively transcribed at low levels to yield Homer 1b/c transcripts. Neuronal activity sharply increases the rate of transcription initiation, with most transcripts now ending within the central intron. These coordinate transcriptional events rapidly convert a constitutive gene to an IEG and regulate the expression of functionally different Homer 1 proteins (Bottai, 2000).

Homer structure

Homer EVH1 (Ena/VASP Homology 1) domains interact with proline-rich motifs in the cytoplasmic regions of group 1 metabotropic glutamate receptors (mGluRs), inositol-1,4,5-trisphosphate receptors (IP3Rs), and Shank proteins. The crystal structure of the Homer EVH1 domain complexed with a peptide from mGluR (TPPSPF) has been determined. In contrast to other EVH1 domains, the bound mGluR ligand assumes an unusual conformation in which the side chains of the Ser-Pro tandem are oriented away from the Homer surface, and the Phe forms a unique contact. This unusual binding mode rationalizes conserved features of both Homer and Homer ligands that are not shared by other EVH1 domains. Site-directed mutagenesis confirms the importance of specific Homer residues for ligand binding. These results establish a molecular basis for understanding the biological properties of Homer-ligand complexes (Beneken, 2000).

PSD-Zip45 (also named Homer 1c/Vesl-1L) is a synaptic scaffolding protein that interacts with neurotransmitter receptors and other scaffolding proteins to target them into post-synaptic density (PSD), a specialized protein complex at the synaptic junction. Binding of the PSD-Zip45 to the receptors and scaffolding proteins results in colocalization and clustering of its binding partners in PSD. It has an Ena/VASP homology 1 (EVH1) domain in the N terminus for receptor binding, two leucine zipper motifs in the C terminus for clustering, and a linking region whose function is unclear despite the high level of conservation within the Homer 1 family. The X-ray crystallographic analysis of the largest fragment of residues 1-163, including an EVH1 domain reported in this study, demonstrates that the EVH1 domain contains an alpha-helix longer than that of the previous models, and that the linking part included in the conserved region of Homer 1 (CRH1) of the PSD-Zip45 interacts with the EVH1 domain of the neighbor CRH1 molecule in the crystal. The results suggest that the EVH1 domain recognizes the PPXXF motif found in the binding partners, and the SPLTP sequence (P-motif) in the linking region of the CRH1. The two types of binding are partly overlapped in the EVH1 domain, implying a mechanism to regulate multimerization of Homer 1 family proteins (Irie, 2002).

Homer interacts with metabotropic glutamate receptors

The molecular basis for glutamate receptor trafficking to the plasma membrane is not understood. Homer 1b (H1b), a constitutively expressed splice form of the immediate early gene product Homer (now termed Homer 1a) regulates the trafficking and surface expression of group I metabotropic glutamate receptors. H1b inhibits surface expression of the metabotropic glutamate receptor mGluR5 in heterologous cells, causing mGluR5 to be retained in the endoplasmic reticulum (ER). In contrast, mGluR5 alone or mGluR5 coexpressed with Homer 1a successfully travels through the secretory pathway to the plasma membrane. In addition, point mutations that disrupt mGluR5 binding to H1b eliminate ER retention of mGluR5, demonstrating that H1b affects metabotropic receptor localization via a direct protein-protein interaction. Electron microscopic analysis reveals that the group I metabotropic receptor mGluR1alpha is significantly enriched in the ER of Purkinje cells, suggesting that a similar mechanism may exist in vivo. Because H1b is found in dendritic spines of neurons, local retention of metabotropic receptors within dendritic ER provides a potential mechanism for regulating synapse-specific expression of group I metabotropic glutamate receptors (Roche, 1999).

Several scaffold proteins for neurotransmitter receptors have been identified as candidates for receptor targeting. However, the molecular mechanism underlying such receptor clustering and targeting to postsynaptic specializations remains unknown. PSD-Zip45 (also named Homer 1c/vesl-1L) consists of the NH(2) terminus containing the enabled/VASP homology 1 domain and the COOH terminus containing the leucine zipper. It has been demonstrated immunohistochemically that metabotropic glutamate receptor 1alpha (mGluR1alpha) and PSD-Zip45/Homer 1c are colocalized to synapses in the cerebellar molecular layer but not in the hippocampus. In cultured hippocampal neurons, PSD-Zip45/Homer1c and N-methyl-D-aspartate receptors are preferentially colocalized to dendritic spines. Cotransfection of mGluR1alpha or mGluR5 and PSD-Zip45/Homer 1c into COS-7 cells results in mGluR clustering induced by PSD-Zip45/Homer 1c. An in vitro multimerization assay shows that the extreme COOH-terminal leucine zipper is involved in self-multimerization of PSD-Zip45/Homer 1c. A clustering assay of mGluRs in COS-7 cells also reveals a critical role of this leucine-zipper motif of PSD-Zip45/Homer 1c in mGluR clustering. These results suggest that the leucine zipper of subsynaptic scaffold protein is a candidate motif involved in neurotransmitter receptor clustering at the central synapse (Tadokoro, 1999).

The physiological actions of neurotransmitter receptors are intimately linked to their proper neuronal compartment localization. The effect of the metabotropic glutamate receptor (mGluR)-interacting proteins, Homer1a, b, and c, were examined in the targeting of mGluR5 in neurons. mGluR5 is exclusively localized in cell bodies when transfected alone in cultured cerebellar granule cells. In contrast, mGluR5 is found also in dendrites when coexpressed with Homer1b or Homer1c, and in both dendrites and axons when cotransfected with Homer1a. In dendrites, cotransfected mGluR5 and Homer1b/c form clusters that colocalize with the synaptic marker synaptophysin. Interestingly when transfected alone, the Homer proteins are also translocated to neurites but do not form such clusters. Depolarization of the neurons with a mixture of ionotropic glutamate receptor agonists, NMDA and kainate, or potassium channel blockers, tetraethylammonium and 4-aminopyridine, induces transient expression of endogenous Homer1a and persistent neuritic localization of transfected mGluR5 even long after degradation of Homer1a. These results suggest that Homer1a/b/c proteins are involved in the targeting of mGluR5 to dendritic synaptic sites and/or axons and that this effect can be regulated by neuronal activity. Because the activity-dependent effect of endogenous Homer1a is also long-lasting, the axonal targeting of mGluR5 by this protein is likely to play an important role in synaptic plasticity (Ango, 2000).

The metabotropic glutamate (mGlu) receptors are a family of receptors involved in the tuning of fast excitatory synaptic transmission in the brain. Experiments performed in heterologous expression systems suggest that cell surface expression of group I mGlu receptors is controlled by their auxiliary protein, Homer. However, whether or not this also applies to neurons has been controversial. In cultured cerebellar granule cells, the group I mGlu receptor subtype, mGlu5, transfected alone is functionally expressed at the surface of these neurons. Transfected Homer1b causes intracellular retention and clustering of this receptor at synaptic sites. Recombinant Homer1a alone does not affect cell surface expression of the receptor, but in neurons transfected with Homer1b, excitation-induced expression of native Homer1a reverses the intracellular retention of mGlu5 receptors, resulting in the receptor trafficking to synaptic membranes. Transfected Homer1a also increases the latency and amplitude of the mGlu5 receptor Ca2+ response. These results indicate that Homer1 proteins regulate synaptic cycling and Ca2+ signaling of mGlu5 receptors, in response to neuronal activity (Ango, 2002).

Homer and the Shank scaffold protein

Shank is a family of postsynaptic proteins that function as part of the NMDA receptor-associated PSD-95 complex. Shank proteins also bind to Homer. Homer proteins form multivalent complexes that bind proline-rich motifs in group 1 metabotropic glutamate receptors and inositol trisphosphate receptors, thereby coupling these receptors in a signaling complex. A single Homer-binding site is identified in Shank, and Shank and Homer coimmunoprecipitate from brain and colocalize at postsynaptic densities. Moreover, Shank clusters mGluR5 in heterologous cells in the presence of Homer and mediates the coclustering of Homer with PSD-95/GKAP. Thus, Shank may cross-link Homer and PSD-95 complexes in the PSD and play a role in the signaling mechanisms of both mGluRs and NMDA receptors (Tu, 1999).

The Shank family of proteins interacts with NMDA receptor and metabotropic glutamate receptor complexes in the postsynaptic density (PSD). Targeted to the PSD by a PDZ-dependent mechanism, Shank promotes the maturation of dendritic spines and the enlargement of spine heads via its ability to recruit Homer to postsynaptic sites. Shank and Homer cooperate to induce accumulation of IP3 receptors in dendritic spines and formation of putative multisynapse spines. In addition, postsynaptic expression of Shank enhances presynaptic function, as measured by increased minifrequency and FM4-64 uptake. These data suggest a central role for the Shank scaffold in the structural and functional organization of the dendritic spine and synaptic junction (Sala, 2001).

Homer proteins interact with other scaffold proteins

A novel rat gene, tanc (GenBank Accession No. AB098072), has been cloned that encodes a protein containing three tetratricopeptide repeats (TPRs), ten ankyrin repeats and a coiled-coil region, and is possibly a rat homolog of Drosophila rolling pebbles. The tanc gene is expressed widely in the adult rat brain. Subcellular distribution, immunohistochemical study of the brain and immunocytochemical studies of cultured neuronal cells indicate the postsynaptic localization of TANC protein of 200 kDa. Pull-down experiments have shown that TANC protein binds PSD-95, SAP97, and Homer via its C-terminal PDZ-binding motif, -ESNV, and fodrin via both its ankyrin repeats and the TPRs together with the coiled-coil domain. TANC also binds the alpha subunit of Ca2+/calmodulin-dependent protein kinase II. An immunoprecipitation study shows TANC association with various postsynaptic proteins, including guanylate kinase-associated protein (GKAP), alpha-internexin, and N-methyl-D-aspartate (NMDA)-type glutamate receptor 2B and AMPA-type glutamate receptor (GluR1) subunits. These results suggest that TANC protein may work as a postsynaptic scaffold component by forming a multiprotein complex with various postsynaptic density proteins (Suzuki, 2004).

Signaling downstream of Homer proteins

Group I metabotropic glutamate receptors (mGluRs) activate PI turnover and thereby trigger intracellular calcium release. mGluRs form natural complexes with members of a family of Homer-related synaptic proteins. Homer proteins form a physical tether linking mGluRs with the inositol trisphosphate receptors (IP3R). A novel proline-rich 'Homer ligand' (PPXXFr) is identified in group 1 mGluRs and IP3R, and these receptors coimmunoprecipitate as a complex with Homer from brain. Expression of the IEG form of Homer, which lacks the ability to cross-link, modulates mGluR-induced intracellular calcium release. These studies identify a novel mechanism in calcium signaling and provide evidence that an IEG, whose expression is driven by synaptic activity, can directly modify a specific synaptic function (Tu, 1998).

Group I metabotropic glutamate receptors (mGluR1 and 5) couple to intracellular calcium pools by a family of proteins, termed Homer, that cross-link the receptor to inositol trisphosphate receptors. mGluRs also couple to membrane ion channels via G-proteins. The role of Homer proteins in channel modulation was investigated by expressing mGluRs and various forms of Homer in rat superior cervical ganglion (SCG) sympathetic neurons by intranuclear cDNA injection. Expression of cross-linking-capable forms of Homer (Homer 1b, 1c, 2, and 3, termed long forms) occludes group I mGluR-mediated N-type calcium and M-type potassium current modulation. This effect is specific for group I mGluRs. mGluR2 (group II)-mediated inhibition of N-channels was unaltered. Long forms of Homer decrease modulation of N- and M-type currents but do not selectively block distinct G-protein pathways. Short forms of Homer, which cannot self-multimerize (Homer 1a and a Homer 2 C-terminal deletion), do not alter mGluR-ion channel coupling. When coexpressed with long forms of Homer, short forms restore the mGluR1a-mediated calcium current modulation in an apparent dose-dependent manner. Homer 2b induces cell surface clusters of mGluR5 in SCG neurons. Conversely, a uniform distribution was observed when mGluR5 was expressed alone or with Homer short forms. These studies indicate that long and short forms of Homer compete for binding to mGluRs and regulate their coupling to ion channels. In vivo, the immediate early Homer 1a is anticipated to enhance ion channel modulation and to disrupt coupling to releasable intracellular calcium pools. Thus, Homer may regulate the magnitude and predominate signaling output of group I mGluRs (Kammermeier, 2000).

Homer proteins form an adapter system that regulates coupling of group 1 metabotropic glutamate receptors (mGluR) with intracellular inositol trisphosphate receptors (IP3R), and is modified by neuronal activity. Homer proteins also physically associate with ryanodine receptors type 1 (RyR1) and regulate gating responses to Ca2+, depolarization, and caffeine. In contrast to the prevailing notion of Homer function, Homer1c (long form) and Homer1-EVH1 (short form) evoke similar changes in RyR activity. The EVH1 domain mediates these actions of Homer, and is selectively blocked by a peptide that mimics the Homer ligand. 1B5 dyspedic myotubes expressing RyR1 with a point mutation of a putative Homer-binding domain, exhibit significantly reduced (~33%) amplitude in their responses to K+ depolarization compared to cells expressing wild-type protein. These results reveal that in addition to its known role as an adapter protein, Homer is a direct modulator of Ca2+ release gain. Homer is the first example of an 'adapter' that also modifies signaling properties of its target protein. The present work reveals a novel mechanism by which Homer directly modulates the function of its target protein RyR1 and E-C coupling in skeletal myotubes. This form of regulation may be important in other cell types that express Homer and RyR1 (Feng, 2002).

Homers are scaffolding proteins that bind G protein-coupled receptors (GPCRs), inositol 1,4,5-triphosphate (IP3) receptors (IP3Rs), ryanodine receptors, and TRP channels. However, their role in Ca2+ signaling in vivo is not known. Characterization of Ca2+ signaling in pancreatic acinar cells from Homer2-/- and Homer3-/- mice show that Homer 3 has no discernible role in Ca2+ signaling in these cells. In contrast, Homer 2 tunes intensity of Ca2+ signaling by GPCRs to regulate the frequency of [Ca2+]i oscillations. Thus, deletion of Homer 2 increases stimulus intensity by increasing the potency for agonists acting on various GPCRs to activate PLCbeta and evoke Ca2+ release and oscillations. This is not due to aberrant localization of IP3Rs in cellular microdomains or IP3R channel activity. Rather, deletion of Homer 2 reduces the effectiveness of exogenous regulators of G proteins signaling proteins (RGS) to inhibit Ca2+ signaling in vivo. Moreover, Homer 2 preferentially binds to PLCbeta in pancreatic acini and brain extracts and stimulates GAP activity of RGS4 and of PLCbeta in an in vitro reconstitution system, with minimal effect on PLCbeta-mediated PIP2 hydrolysis. These findings describe a novel, unexpected function of Homer proteins, demonstrate that RGS proteins and PLCbeta GAP activities are regulated functions, and provide a molecular mechanism for tuning signal intensity generated by GPCRs and, thus, the characteristics of [Ca2+]i oscillations (Shin, 2003).

Homer proteins and axon guidance

Homer proteins are members of a family of multidomain cytosolic proteins that have been postulated to serve as scaffold proteins that affect responses to extracellular signals by regulating protein-protein interactions. Whether Homer proteins are involved in axon pathfinding in vivo was tested by expressing both wild-type and mutant isoforms of Homer in Xenopus optic tectal neurons. Time-lapse imaging has demonstrated that interfering with the ability of endogenous Homer to form protein-protein interactions results in axon pathfinding errors at stereotypical choice points. These data demonstrate a function for scaffold proteins such as Homer in axon guidance. Homer may facilitate signal transduction from cell-surface receptors to intracellular proteins that govern the establishment of axon trajectories (Foa, 2001).

Degradation of Homer

The vesl-1S/homer-1a gene is up-regulated during seizure and long term potentiation. Other members of the Vesl family, Vesl-1L, -2, and -3, are constitutively expressed in the brain. The regulatory mechanisms governing the expression level of Vesl-1S protein, either an exogenously introduced one in COS7 or human embryonic kidney 293T cells or an endogenous one in rat brain neurons, were examined in cultures. In both cases, application of proteasome inhibitors increases the amount of Vesl-1S protein but not that of Vesl-1L, -2, or -3 protein. Deletion analyses revealed that the C-terminal 11-amino acid region is responsible for the proteolysis of Vesl-1S by proteasomes. Application of proteasome inhibitors promotes ubiquitination of Vesl-1S protein but not that of the Vesl-1S deletion mutant, which evades proteasome-mediated degradation. These results indicate that ubiquitin-proteasome systems are involved in the regulation of the expression level of Vesl-1S protein (Ageta, 2001).

Homer proteins regulate sensitivity to cocaine

Drug addiction involves complex interactions between pharmacology and learning in genetically susceptible individuals. Members of the Homer gene family are regulated by acute and chronic cocaine administration. Deletion of Homer1 or Homer2 in mice causes the same increase in sensitivity to cocaine-induced locomotion, conditioned reward, and augmented extracellular glutamate in nucleus accumbens as that elicited by withdrawal from repeated cocaine administration. Moreover, adeno-associated virus-mediated restoration of Homer2 in the accumbens of Homer2 KO mice reverses the cocaine-sensitized phenotype. Further analysis of Homer2 KO mice revealed extensive additional behavioral and neurochemical similarities to cocaine-sensitized animals, including accelerated acquisition of cocaine self-administration and altered regulation of glutamate by metabotropic glutamate receptors and cystine/glutamate exchange. These data show that Homer deletion mimics the behavioral and neurochemical phenotype produced by repeated cocaine administration and implicate Homer in regulating addiction to cocaine (Szumlinski, 2004).

The effect of deleting other genes on cocaine-induced behaviors has been examined. Deleting RGS9-2 or 5HT1B produces augmented behavioral responses to cocaine that are akin to those elicited by Homer1 or Homer2 deletion, while ß-arrestin and DARP-32 deletions are without effect and mGluR5 deletion results in mice that are apparently insensitive to the behavioral effects of cocaine. The sensitization that is associated with the reduced capacity of mGluR1/5 stimulation to induce extracellular glutamate in Homer2 KO mice would seem to contradict the lack of cocaine responsiveness produced by mGluR5 deletion. However, the increase in extracellular glutamate by DHPG is mediated by mGluR1, not mGluR5. In contrast with Homer, RGS9-2 is elevated by repeated cocaine administration and is postulated to serve a compensatory rather than facilitatory role in regulating cocaine behaviors. Deletion of Homer2 elicits effects that appear to be selective for cocaine behaviors and apparently do not impact natural reward, aggression, anxiety, or learning and memory, while 5HT1B KO mice show elevated levels of aggression. The detailed neurochemical adaptations that are describe in the Homer2 KO mice have not been reported in these other genetic models, but the convergence of Homer2 focuses attention on G protein signaling. A role for altered G protein signaling is further implicated by the similar elevation in AGS3 produced by Homer2 deletion and withdrawal from repeated cocaine administration and that inactivation of Gialpha by elevated AGS3 sensitizes cocaine-induced behaviors and increases in extracellular glutamate in the accumbens (Szumlinski, 2004 and references therein).

While it can be anticipated that additional genetic models may be discovered that mimic or block behaviors associated with cocaine addiction, the striking concordant neurochemical phenotype between Homer2 deletion and withdrawal from chronic cocaine treatment indicates that Homer is a particularly good candidate to play a central role in cocaine addiction. Moreover, Homer gene products are regulated by acute and repeated cocaine administration, and Homer1a is dynamically responsive to environmental cues and stress. Thus, not only does Homer provide an important window to understand cocaine-induced neuroplasticity and addiction, but also to study the molecular basis of the important link between environmental stress and cocaine addiction (Szumlinski, 2004 and references therein).

REFERENCES

Search PubMed for articles about Drosophila homer

Ageta, H., et al. (2001). Regulation of the level of Vesl-1S/Homer-1a proteins by ubiquitin-proteasome proteolytic systems. J. Biol. Chem. 276(19): 15893-7. 11278836

Ango, F., et al. (2000). Dendritic and axonal targeting of type 5 metabotropic glutamate receptor is regulated by Homer1 proteins and neuronal excitation. J Neurosci 20: 8710-8716. 11102477

Ango, F., et al. (2002).Homer-dependent cell surface expression of metabotropic glutamate receptor type 5 in neurons. Mol. Cell. Neurosci. 20(2): 323-9. 12093163

Babu, K., Cai, Y., Bahri, S. Yang, X. and Chia, W. (2004). Roles of Bifocal, Homer, and F-actin in anchoring Oskar to the posterior cortex of Drosophila oocytes. Genes Dev. 18: 138-143. 14752008

Beneken, J., et al. (2000). Structure of the Homer EVH1 domain-peptide complex reveals a new twist in polyproline recognition. Neuron 26(1): 143-54. 10798399

Bottai, D., et al. (2000). Synaptic activity-induced conversion of intronic to exonic sequence in Homer 1 immediate early gene expression. J. Neurosci. 22(1): 167-75. 11756499

Brakeman, P. R., et al. (1997). Homer: a protein that selectively binds metabotropic glutamate receptors. Nature 386: 284-288. 9069287

Diagana, T. T., (2002). Mutation of Drosophila homer disrupts control of locomotor activity and behavioral plasticity J. Neurosci. 22(2): 428-436. 11784787

Feng, W., et al. (2002). Homer regulates gain of ryanodine receptor type 1 channel complex. J. Biol. Chem. 12223488

Foa, L., et al. (2001). The scaffold protein, Homer 1b/c, regulates axon pathfinding in the central nervous system in vivo. Nat. Neurosci. 4: 499-506. 11319558

Irie, K., et al. (2002). Crystal structure of the Homer 1 family conserved region reveals the interaction between the EVH1 domain and own proline-rich motif. J. Mol. Biol. 318(4): 1117-26. 12054806

Kammermeier, P. J., et al. (2000). Homer proteins regulate coupling of group I metabotropic glutamate receptors to N-type calcium and M-type potassium channels. J. Neurosci. 20(19): 7238-45. 11007880

Kato, A., et al. (1998). Novel members of the Vesl/Homer family of PDZ proteins that bind metabotropic glutamate receptors. J. Biol. Chem. 273: 23969-223975. 9727012

Miyashita, T., Oda, Y., Horiuchi, J., Yin, J. C., Morimoto, T. and Saitoe, M. (2012). Mg²⁺ block of Drosophila NMDA receptors is required for long-term memory formation and CREB-dependent gene expression. Neuron 74(5): 887-98. PubMed Citation: 22681692

Naisbitt, S., et al. (1999). Shank, a novel family of postsynaptic density proteins that binds to the NMDA receptor/PSD-95/GKAP complex and cortactin. Neuron 23: 569-582. 10433268

Parmentier, M. L., Pin, J. P., Bockaert, J. and Grau, Y. (1996). Cloning and functional expression of a Drosophila metabotropic glutamate receptor expressed in the embryonic CNS. J. Neurosci. 16: 6687-6694. 8824309

Roche, K. W., et al. (1999). Homer 1b regulates the trafficking of group I metabotropic glutamate receptors. J Biol Chem 274: 25953-25957. 10464340

Sala, C., et al. (2001). Regulation of dendritic spine morphology and synaptic function by Shank and Homer. Neuron 31: 115-130. 11498055

Shin, D. M., et al. (2003). Homer 2 tunes G protein-coupled receptors stimulus intensity by regulating RGS proteins and PLCbeta GAP activities. J. Cell Biol. 162(2): 293-303. 12860966

Shiraishi, Y., et al. (1999). Cupidin, an isoform of Homer/Vesl, interacts with the actin cytoskeleton and activated rho family small GTPases and is expressed in developing mouse cerebellar granule cells. J. Neurosci. 19: 8389-8400. 10493740

Slutsky, I., Abumaria, N., Wu, L.J., Huang, C., Zhang, L., Li, B., Zhao, X., Govindarajan, A., Zhao, M.G., Zhuo, M., et al. (2010). Enhancement of learning and memory by elevating brain magnesium. Neuron. 65: 165-177. PubMed Citation: 20152124

Suzuki, T., et al. (2004). A novel scaffold protein, TANC, possibly a rat homolog of Drosophila Rolling pebbles (Rols), forms a multiprotein complex with various postsynaptic density proteins. Eur. J. Neurosci. 21(2): 339-50. 15673434

Szumlinski, K. K., et al. (2004). Homer proteins regulate sensitivity to cocaine. Neuron 43(3): 401-13. 15294147

Tadokoro, S., et al. (1999). Involvement of unique leucine-zipper motif of PSD-Zip45 (Homer 1c/vesl- 1L) in group 1 metabotropic glutamate receptor clustering. Proc. Natl. Acad. Sci. 96: 13801-13806. 10570153

Thomas, U. (2002). Modulation of synaptic signalling complexes by Homer proteins. J. Neurochem. 81(3):407-13. 12065649

Tu, J. C., et al. (1998). Homer binds a novel proline-rich motif and links group 1 metabotropic glutamate receptors with IP3 receptors. Neuron 21: 717-726. 9808459

Tu, J. C., et al. (1999). Coupling of mGluR/Homer and PSD-95 complexes by the Shank family of postsynaptic density proteins. Neuron 23: 583-592. 10433269

Xiao, B., et al. (1998). Homer regulates the association of group 1 metabotropic glutamate receptors with multivalent complexes of Homer-related, synaptic proteins. Neuron 21: 707-716. 9808458

Xiao, B., Tu, J. C. and Worley, P. F. (2000). Homer: a link between neural activity, glutamate receptor function. Curr. Opin. Neurobiol. 10: 370-374. 10851183

date revised: 5 March 2013

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