The expression pattern of the spin gene was analyzed by Northern blotting. A 3-kb transcript is observed throughout development, although the level of expression is very low in the embryonic and second-instar larval stages. The relative amount of each type of transcript was examined by RT-PCR methods. The type III and type IV transcripts are abundant, while types I and II are expressed at a moderate level and type V is very rare. No major differences were found during development between the sexes or between wild-type and spinP1 flies in this experiment. The spatial expression of the spin transcripts was examined by whole-mount in situ hybridization using an antisense RNA or cDNA probe that was able to detect all five types of transcript. The spin transcripts are detectable at the beginning of the germ band retraction (stage 12) in a subset of cells in the ventral nerve cord and the brain, and this expression pattern continued throughout development (Nakano, 2001).
In order to establish the identity of the Spin-expressing cells, double staining was performed using an anti-Repo antibody and an antisense probe to spin mRNA. Repo is a glia-specific homeobox protein expressed in all glial cells except for the midline glial and two segmental nerve root glial cells. More than 95% of Spin-expressing cells overlapped with Repo-expressing cells in the VNC and the brain. This expression pattern was confirmed by using spinP2, which carries a P-element with an enhancer trap reporter inserted in the middle of the first exon of the spin gene. The ß-Gal expression pattern observed in the embryos and the third-instar larval brain of the spinP2 heterozygotes was found to correspond well with spin expression as detected by in situ hybridization. The spin gene is expressed in the surface glial cells, which include the peripheral exit glia, the subperineural glia, and the channel glia of the nervous system. In addition, ß-Gal expression in spinP2/CyO flies in the larval and pupal stages is also observed in the trachea, gut, salivary glands, and ring gland (Nakano, 2001).
The expression of the spin gene is also observed in adult ovaries; expression is observed in the follicle cells in the manner of dorsal-ventral and anterior-posterior gradients but not in the nurse cells or the oocyte. This expression pattern has also been confirmed by in situ hybridization (Nakano, 2001).
spin is expressed in both motoneurons and muscle throughout the period of synaptic growth and development at the NMJ. The expression of spin was first assessed by embryonic in situ hybridization. Expression was observed throughout the CNS, including motoneurons. Weak expression was observed in embryonic muscle as well as other tissues. The larval expression of spin is of particular interest since this is the time period of synaptic growth. Unfortunately, RNA in situ analysis in the larval CNS and muscle is particularly difficult to interpret. Therefore, the spin expression pattern was determined in larvae by driving a GFP-tagged spin transgene (the same isoform used in the rescue experiments) with a spin promoter—GAL4 fusion (Nakano, 2001). When spin-GAL4 is used to drive expression of UAS-spin-GFP, Spin-GFP expression is observed throughout the larval CNS with pronounced expression in motoneurons. Strong expression is also observed in all body wall muscle as well as other tissues, including a subset of epithelial cells and the salivary glands. In these experiments, Spin-GFP localizes to a peri-nuclear region in both larval neurons and larval muscle. Spin-GFP fluorescence is also observed throughout the muscle and is observed in the nerves that include both sensory and motor axons. Because Spin-GFP puncta are present in both the nerve and underlying muscle, it is difficult to determine whether Spin-GFP puncta are present within the presynaptic nerve terminal in these experiments. These GFP-positive puncta are suggestive of a late endosomal localization pattern in nerve and muscle (Sweeney, 2002).
To assess the localization of the endogenous Spin protein, an antibody was raised against Spin using a combination of two peptides, one being an N-terminal peptide represented in all splice forms of the protein and the second being a peptide to a region between transmembrane domains 11 and 12 that is present in four of the five isoforms of spin. The antibody detects the same widely distributed punctate staining pattern that is observed when UAS-spin-GFP expression is driven by spin-GAL4. Spin immunoreactive puncta are observed in a peri-nuclear pattern in neurons and muscle, and these puncta are widely distributed throughout muscle fibers. In addition, Spin puncta are present within the presynaptic nerve terminal. The NMJ was co-stained with anti-Synapsin and anti-Spin. Three-dimensional optical reconstruction of individual synaptic boutons by confocal microscopy demonstrates that Spin immunoreactive puncta are present within the volume of the presynaptic Synapsin staining. To further demonstrate the presence of Spin in the presynaptic nerve terminal, animals expressing Spin-GFP presynaptically were fixed and stained and then costained with anti-Synapsin. Spin-GFP puncta are clearly present within the presynaptic nerve terminal (Sweeney, 2002).
The size and distribution of the Spin-positive puncta are suggestive of a late endosomal/lysosomal distribution. Colocalization of Spin with known endosomal and lysosomal markers was tested. Unfortunately, there is a paucity of endosomal/lysosomal markers in Drosophila. Therefore, the subcellular localization of Spin-GFP was examined in mammalian cells. Near perfect colocalization of Spin-GFP with LAMP-1 (a lysosomal marker) is found in HeLa cells, indicating that Spin is localized to the lysosome in these cells. The lysosomal localization of Spin-GFP is not influenced by where the GFP is located on the transgene. Unfortunately, anti-LAMP-1 does not recognize Drosophila lysosomes, and there are no other known lysosomal markers in the fly. However, Spin-GFP (driven by spin-GAL4) is specifically localized to a low pH compartment identified by the lysotracker vital dye. The perinuclear localization of this low pH compartment, and its scattered distribution radiating from the nucleus, is consistent with the localization of the late endosomal/lysosomal compartment in skeletal muscle as determined by electron microscopy. In these experiments, UAS-spin-GFP is driven by spin-GAL4, and the GFP localization in these animals closely matches the endogenous protein distribution detected with the antibody. The location of muscle nuclei in these experiments was determined by visualization with Nomarski optics. These data suggest that Spin is localized to a lysosomal compartment in vivo (Sweeney, 2002).
A battery of endosomal markers was tested for colocalization with Spin-GFP and anti-Spin. Partial colocalization of Spin was observed with anti-Hrs, particularly in the peri-nuclear region in neurons and muscle. It is notable that Hrs immunoreactivity in muscle is concentrated to the subsynaptic reticulum (SSR), a series of postsynaptic muscle folds of unknown function, whereas Spin does not localize to this site. Colocalization with anti-Rab5 (early endosome), anti-Hook (late endosome), and anti-Deep Orange (Dor; late endosome) was examined. Partial overlap is observed between Spin and both anti-Hook and anti-Dor. However, more frequently these markers reside in a vesicle-like compartment that appears immediately adjacent to the Spin vesicle-like compartment. No colocalization or juxtaposition of Spin with anti-Rab5 is observed. Taken together, these data support the conclusion that Spin localizes to a late endosomal/lysosomal compartment in Drosophila muscle. In motoneurons, the perinuclear localization is also consistent with a late endosomal/lysosomal localization. The identification of lysosomes at the synapse is more controversial, though lysosomes have been observed at newly formed and developing synapses and have been found to distribute down axons, being concentrated at the nodes of Ranvier. At the Drosophila NMJ, electron microscopy demonstrates that multivesicular bodies (MVB) are present within wild-type synaptic terminals at the NMJ. These MVB are sparsely distributed throughout synaptic boutons in a manner that is consistent with anti-Spin immunoreactivity. It is concluded, therefore, that Spin identifies a late endosomal/lysosomal compartment in muscle and within the presynaptic nerve terminal (Sweeney, 2002).
The localization of Spin to a late endosomal/lysosomal compartment, and the previous evidence that spin mutations are associated with accumulation of ceroid lipofuscin (Nakano, 2001) prompted an examination of the late endosomal architecture in nerve and muscle in spin mutations. Lysotracker was used to compare the late endosomal compartments in wild-type and spin mutant backgrounds. A dramatic expansion of a low pH compartment was observed in spin mutant muscle. The peri-nuclear localization and banding pattern of this low pH compartment in muscle is consistent with the localization of late endosomes/lysosomes in vertebrate skeletal muscle as determined by electron microscopy (Sweeney, 2002).
Importantly, a dramatic expansion of a low pH compartment is observed within the presynaptic nerve terminal. In these experiments, FITC-conjugated anti-HRP was used in combination with the lysotracker in a live staining protocol. FITC-anti-HRP efficiently stains extracellular epitopes on the presynaptic membrane during the time of lysotracker staining. This allows the visualization of lysotracker-positive compartments that are present within the presynaptic nerve terminal. At spin mutant NMJ, there are increased numbers of presynaptic lysotracker-positive puncta, and most notably, these puncta are significantly enlarged in size. At spin synapses, nearly all of the large boutons contain a large low pH compartment. Lysotracker staining of the wild-type synaptic terminals rarely identified such a compartment. However, on occasion these compartments were observed at one or two boutons within a wild-type synapse. Three-dimensional reconstruction of lysotracker-positive synaptic boutons demonstrates that the low pH compartments are present within the volume of the presynaptic bouton. Taken together these data demonstrate a dramatic expansion and alteration of the late endosomal compartment in both the presynaptic nerve-terminal and in muscle. Endosomal expansion has been observed in Hrs mutations in Drosophila embryos and mice. Giant endosome/lysosomes have also been observed in patients with Chediak-Higashi syndrome and in mutant beige mice. These structures are thought to arise from dysregulated homotypic fusion. It is hypothesized, therefore, that normal late endosomal function is perturbed in the spin mutant background (Sweeney, 2002).
To further investigate the expansion of the late endosomal system, the staining intensity of a variety of late endosomal markers in muscle was assessed, comparing wild-type with spin mutant muscle. Anti-Hrs, anti-Deep Orange, and anti-Cathepsin-L all have significantly elevated staining intensity in spin mutant muscle and presynaptic terminals compared to wild-type. Although the staining intensity is increased, the staining pattern is grossly normal for these proteins with the exception of anti-Dor. Anti-Dor staining is normally concentrated to the SSR at the postsynaptic side of the synapse. In spin mutant muscle anti-Dor is no longer strongly concentrated to the SSR but is now distributed throughout the muscle, giving the appearance that Dor is no longer localized to the SSR. Since Dor expression is elevated throughout the muscle, it is hypothesized that this reflects a redistribution of this protein throughout the endosomal system rather than a loss of Dor from the postsynaptic membranes. Finally, there is no change in the expression or distribution of the early endosomal protein Rab5, indicating that these changes are specific to the late endosomal compartment. These data further support the conclusion that the loss of spin causes an expansion and possibly a disruption of the late endosome compartment in Drosophila (Sweeney, 2002).
Mutations in the spinster gene are characterized by an extraordinarily strong rejection behavior of female flies in response to male courtship. Single male and female pairs were placed in a plastic syringe for 1 h. During this time the mating success was measured; it was found that, while 70% of the wild-type pairs copulated, only 4% of the spinP1 mutant females that were paired with wild-type males copulate under the same conditions. Females of a revertant line, obtained by P-element excision, exhibited essentially the same level of mating success as the wild-type females. This result demonstrates that the spinster courtship phenotype is caused by the P-element insertion, since excision is able to restore normal receptivity in females. The intensity of the male courtship can be quantified by the SAPI; this index represents the percentage of time spent by the male performing unilateral wing vibration during a 10-min observation period. The SAPI is almost the same for wild-type, spinP1, and revertant pairs, thus indicating that the females of these strains are able to elicit similar levels of courtship from the males. This means that the low mating success observed in spinP1 females cannot be accounted for by reduced attractiveness but rather that the low mating success may reflect the unwillingness of the spinP1 females to copulate (Nakano, 2001).
Indeed, the spinP1 females consistently display a number of rejection responses against the courting males; these included fending, kicking, flicking, curling, punching, and decamping; ovipositor extrusion (a normal rejection behaviour, in response to courting males or males that had previously performed courtship) is rarely seen. The pattern of rejection displayed by spinP1 females resembles that of immature wild-type virgin females rather than that of fertilized females, in that extrusion does not occur. However, the spinP1 females do exhibit kicking and curling behavior much more frequently than the wild-type females. In response to approaching males, the spinP1 females tend to raise their abdomens while spreading their vaginal plates. This spreading is unique to the spin mutant females and is also distinctly different from extrusion, in which the ovipositor protrudes from the female terminalia. Furthermore, the spinP1 female often rushes toward the courting male, pushing the male's head with her forelegs; this aggressive behavior is termed punching and is rare among wild-type females. A similarly pronounced refusal of suitors is observed in heteroallelic spinP1/spinP2 females. spinP1 male flies exhibit no obvious abnormality in their courtship behavior, while general locomotive activity is reduced in both sexes (12% reduction in females and 38% reduction in males 3 days old) (Nakano, 2001).
Apart from the regulation of female sexual behavior, the spin gene plays an additional vital role, since the partial loss-of-function mutation (spinP1) reduces viability and life span and the spinP2 mutation is lethal, yielding no adult flies. This reduction in viability caused by the spinP1 mutation is actually more extreme in males than in females (the viability of females is 22% and that of males is 11%), resulting in an uneven sex ratio (females/males, 2:1) of the emerged homozygous adults (Nakano, 2001).
During the tissue staining of spinP1 flies, the existence of autofluorescent material in the blue channel was observed. In the VNC, the autofluorescent material is observed mainly in the central region of the VNC, particularly in the abdominal ganglia of pupae and adults. The lethal allele, spinP2, shows an earlier onset of the accumulation of autofluorescent materials in the CNS (from larval stages). In the brain, this material is observed in the central brain and in the optic lobe. Part of the autofluorescence overlaps with spin gene expression; however, most does not. EM analysis of the VNC cells in spinP1 homozygotes and heterozygotes at the early pupal stage and 24 h after eclosion reveals that spin mutant samples exhibit cellular disorganization in that most of the spin mutant cells, including both the neurons and glial cells, contain multilamellate bodies and electron-dense lobulated granules; these structures are never observed in heterozygote cells. These aberrant structures are observed in the CNS from the early pupal stage and increase thereafter in spinP1 homozygotes. The early pupal VNC contains electron-dense lobulated granules, which appear to be precursors of the multilamellate bodies seen in the adult VNC. These structures are observed in both sexes, and no differences are observed between the sexes at the cellular level. Neither structure is found in gut or muscle cells from the spin mutant; in addition, the nucleus, mitochondria, and endoplasmic reticulum all appear to be normal in the spin mutant. The aberrant structures contained within spin mutant nerve cells are found to be very similar to lipofuscin, which is known to be induced by oxidative stress, some proteinase inhibitors, inherited lysosomal storage diseases, and the normal aging process (Nakano, 2001).
In order to identify the nature of these materials, biochemical analyses were performed. A large proportion of fluorescent pigments in tissues can be extracted by a chloroform-methanol solution and a sensitive fluorometric assay is available for the measurement of fluorescent lipid peroxidation products that have accumulated in the various tissues. Lipids were extracted from the heads of spin and wild-type flies and fluorescence spectra of the lipid extracts were then measured. The lipid extracts from the heads of spin flies have an excitation maximum at 368 nm and an emission maximum at 450 nm; these are characteristic of those observed with the lipofuscin pigments. The fluorescence intensity of the spin flies was 3.1 times higher than that of wild-type flies, suggesting that the accumulation of lipid-soluble lipofuscin-like substances was significant in spin flies. The accumulation of lipid peroxides was also examined using a thiobarbituric acid assay. The amounts of thiobarbituric acid-reactive substances in the homogenates and the lipid extracts from the heads of spin flies were found to be 30.2 ± 0.17 pmol/mg of head and 104.3 ± 43.6 pmol/mg of head respectively. These values are significantly higher than those observed with wild-type flies, namely, 21.5 ± 0.24 and 51.0 ± 6.3 pmol/mg of head, respectively. These results clearly indicate that the chemical nature of the lipofuscin-like pigments that accumulates in spin flies is quite similar to that reported for lipofuscin pigments in various mammalian tissues (Nakano, 2001).
spinP1 females not only exhibit very strong rejection behavior toward courting males, but they also rarely lay eggs. However, the distribution of motor nerve endings along the uterine muscles was found to be normal. In order to evaluate the possibility that spinP1 mutants are defective in egg production, ovarian development in spin flies was studied. At stage 12, nurse cells are found to dump cytoplasmic components into the oocyte, their nuclei accumulate at the anterior of the oocyte, and the actin bundles are well formed. At stage 14, the dorsal appendages are well formed and nurse cell nuclei have disappeared due to PCD in heterozygous flies. In spinP1 mutant flies, the dorsal appendages are again well formed at stage 14, but the nurse cell nuclei are still present and some oocytes are found to be degenerated; spinP1 mutant mature ovaries exhibit an accumulation of hundreds of nurse cell nuclei near the basal stalk. The same phenotype is also observed in the EP822 line. In contrast, spin mutant males produce normal offspring; therefore, male germ cells seem to develop normally in these flies (Nakano, 2001).
The Spin mRNA is alternatively spliced into five identified transcripts (types I-V) and these show different abilities to rescue survival and behavior deficits in spin mutants. In the behavioral-rescue experiment, the type III transgene was examined in spinP1/spinP1 flies, while in the lethality rescue experiment, the transgene was introduced into the spinP2/spinP2 flies. Type I cDNA can rescue both the behavioral and lethal phenotypes. Type V cDNA, which encodes half the protein of type I, is able to rescue the behavioral phenotype; however, it is not able to rescue the lethality phenotype. Type II, III, and IV cDNAs are unable to rescue the behavioral and lethal phenotypes (Nakano, 2001).
Given that the sole difference between the type I and type II transcripts is an alternative usage of exon 4 in type I and of exon 5 in type II, the exon 4 sequence is necessary for rescuing the behavioral and lethal phenotypes. These two exons both encode the same number of amino acids and share 56% homology, and although their sequences are different, the predicted topology does not change. Since the difference between type I and type III is located in the C-terminal region, the C terminus also seems to play an important role in Spin function. These data suggest that reduced amounts of Spin proteins, particularly the type I and type V products, induce spinP1 phenotypes (Nakano, 2001).
spinster (spin), which encodes a multipass transmembrane protein, has been identified in a genetic screen for genes that control synapse development. spin mutant synapses reveal a 200% increase in bouton number and a deficit in presynaptic release. spin is expressed in both nerve and muscle and is required both pre- and postsynaptically for normal synaptic growth. Spin has been localized to a late endosomal compartment and evidence is presented for altered endosomal/lysosomal function in spin mutants. Evidence is presented that synaptic overgrowth in spin is caused by enhanced/misregulated TGF-ß signaling. TGF-ß receptor mutants show dose-dependent suppression of synaptic overgrowth in spin. Furthermore, mutations in Dad, an inhibitory Smad, cause synapse overgrowth. A model is presented for synaptic growth control with implications for the etiology of lysosomal storage and neurodegenerative disease (Sweeney, 2002).
Mutation in spin was initially identified in a large-scale genetic screen for mutations involved in the regulation of synaptic structure and function. A single spinster mutation [EP(2)0822] referred to hereafter as spin1 was initially identified and subsequently four new P element alleles of spin were identified based on a screen of available databases. All of these alleles are predicted to be hypomorphic loss-of-functional mutations based on the P element insertion sites. In order to isolate a null allele of spin, a small deficiency (spinDelta2b was generated using the technique of male recombination from the spin1 P element. The spinDelta2b deficiency deletes approximately 5 kb of sequence including the entire first coding exon of spin, which is common to all known splice variants of this gene, as well as the majority of the first intron. A large deficiency was identified that uncovers the spin locus (Df(2)Jp4). Subsequent genetic and molecular data support the conclusion that spinDelta2b is a null mutation (Sweeney, 2002).
It was first determined that spin is necessary for viability, demonstrating that severe spin mutations cause lethality at the late pupal stage. The three P element mutations that reside within the spin transcript (spin1, spin4, and spin5) as well as the null allele, spinDelta2b, are all lethal at the pharate pupal stage. In addition, these three lethal P element insertions fail to complement the spinDelta2b and Df(2)Jp4 chromosomes with the same pupal lethal phase. RNA in situ experiments demonstrate that there is a significant maternal contribution of spin that could account for the late lethal phase. Alternatively, spin may have a particular requirement during pupal development, or may cause a progressive defect that is manifest as late pupal lethality (Sweeney, 2002).
An analysis of synaptic morphology at the third instar NMJ reveals a dramatic phenotype of synaptic overgrowth in all of the spin mutant combinations tested. The phenotype of synaptic overgrowth is highly penetrant, affecting every neuromuscular synapse examined, including muscles 6/7 and muscle 4 as well as muscles 12 and 13. Synaptic overgrowth at muscles 6/7 was quantified by counting synaptic boutons. Individual synaptic boutons within the NMJ were clearly identified by staining with the presynaptic marker anti-Synapsin. Bouton numbers were examined in wild-type as well as four genetic controls and these data were compared to quantification of seven different spin mutant combinations. Bouton numbers are increased by more than 200% in all of the strong loss-of-function mutant combinations. This remarkable synaptic overgrowth exceeds that observed in any known mutation in Drosophila, with the exception of the highwire mutation. In these experiments, bouton numbers were not normalized to muscle size as is commonly done to account for the presumed growth coupling between muscle size and presynaptic growth. spin mutant muscle fibers are slightly smaller than wild-type muscle fibers on average. Therefore, if coupling between muscle growth and presynaptic growth persists in the spin mutant background, then the increase in synapse size compared to wild-type has been significantly underestimated by not normalizing bouton counts to muscle volume (Sweeney, 2002).
There are several examples of mutations that change bouton number but do not change synapse area because altered bouton number is compensated by an opposing change in the size of individual boutons. By contrast, the dramatic increase in bouton number in spin does cause an expansion of total synaptic area. Synaptic span was measured and this parameter is increased by more than 200%. More importantly, the average bouton size is normal despite the observed 200% increase in bouton number. The two-dimensional area of individual synaptic boutons from wild-type and spin mutant synapses at muscles 6/7 were measured. The 2D bouton surface area of each individual bouton within a synapse was measured, and these numbers were then averaged across several synapses for each genotype, including several hundred boutons per genotype. Average bouton size in spin is not significantly different from wild-type. Thus, synaptic growth is enhanced at spin mutant synapses (Sweeney, 2002).
In order to determine whether spin is necessary in the neuron or the muscle for normal synaptic growth regulation, the spin mutant phenotype was rescued by overexpression of a spin cDNA (UAS-spin) using tissue-specific GAL4 drivers that express in either the nerve (elav-GAL4), the muscle (MHC-GAL4), or ubiquitously (tubulin-GAL4). The cDNA used for rescue experiments was annotated as type III by Nakano (2001) and is one of the two most abundantly expressed isoforms of the gene (Nakano, 2001). Overexpression of UAS-spin by the ubiquitous tubulin-GAL4 promoter does not alter synapse morphology, larval crawling behavior, or viability. Thus, there is no phenotype associated with Spin overexpression. However, when tubulin-GAL4 is used to overexpress UAS-spin in the spin mutant background, bouton numbers are completely rescued to wild-type. Overexpression of UAS-spin specifically on either the presynaptic (elav-GAL4) or postsynaptic (MHC-GAL4) side of the synapse achieves only partial rescue of the synaptic overgrowth phenotype. Thus, the data argue that spin is required on both sides of the synapse for normal synaptic growth regulation (Sweeney, 2002).
In order to determine whether synaptic overgrowth in spin is associated with altered synaptic function, synaptic transmission at muscles 6/7 was quantified in wild-type, spin, and rescue larvae. Despite synaptic overgrowth, quantal content is decreased in the spin mutant background by approximately 50%. There is no change in quantal size, resting potential, or muscle input resistance at spin mutant synapses. These data are consistent with a deficit in presynaptic transmitter release in the spin mutations (Sweeney, 2002).
Experiments were persued to rescue the deficit in presynaptic release with UAS-spin transgenes expressed either pre- or post-synaptically. Surprisingly, the rescue experiments demonstrate that spin is required either pre- or post-synaptically for normal presynaptic transmitter release. Expression of UAS-spin presynaptically rescues synaptic function to wild-type levels, as does expression of UAS-spin in muscle. These data are in contrast to results demonstrating that spin is required both pre- and post-synaptically for normal synaptic growth (overgrowth is only partially suppressed when spin is rescued on only one side of the synapse). One possible explanation is that the deficit in synaptic function is secondary to the dramatic synaptic overgrowth observed at spin mutant synapses. Partial rescue of synaptic overgrowth may enable normal synaptic function by bringing synaptic overgrowth below the threshold that normally causes synapse dysfunction. The alternative is that spin supplies some necessary component to functional synapse development that can be contributed from either the pre- or post-synaptic side of the synapse (Sweeney, 2002).
Importantly, the expression of spin specifically in either muscle or neurons also rescues the viability of the spin mutant animals to wild-type levels. Less than 1% of spin null mutants and only 23% of hypomorphic spin (spin4/spin5) animals are adult viable. However, when UAS-spin is expressed in either neurons (elav-GAL4 or D42-GAL4) or muscle (MHC-GAL4) adult viability in the spin4/spin5 animals is restored to 100% (Sweeney, 2002).
Protein trafficking decisions within the endosomal system can influence intracelluar signaling by specifying whether receptors are returned to the plasma membrane or targeted for degradation. Thus, a disruption of endosomal function could enhance growth factor signaling and lead to synaptic overgrowth. TGF-ß has been implicated in the regulation of synaptic growth at the Drosophila NMJ. Mutation in the type II TGF-ß receptor wishful thinking (wit) causes a significant decrease in bouton number. To test whether enhanced TGF-ß signaling causes synaptic overgrowth in the spin mutant background, a genetic analysis of TGF-ß signaling was persued at the wild-type and spin mutant synapse (Sweeney, 2002).
It was hypothesized that enhanced or unregulated growth factor signaling is the cause of overgrowth in spin, and therefore, whether enhanced TGF-ß signaling is sufficient to cause synaptic overgrowth was tested. If TGF-ß signaling is sufficient to cause enhanced synaptic growth, then a mutation in a negative regulator of TGF-ß signaling is predicted to cause an increase in bouton number. Daughters against DPP (Dad) encodes an inhibitory Smad that negatively regulates TGF-ß signaling in Drosophila and other systems. Synapse morphology was examined in a strong loss-of-function Dad mutation that is viable to third instar larvae. Dad mutant synapses reveal a dramatically altered morphology with increased numbers of clearly distinct, small boutons that sprout from what appears to be the normal synaptic process. This is a highly penetrant phenotype and is observed at muscles 6/7 and muscle 4. Quantification of total synaptic bouton number demonstrates a significant increase in bouton numbers that is nearly equivalent to that observed in the spin mutant. These data demonstrate that enhanced TGF-ß signaling can cause synaptic overgrowth (Sweeney, 2002).
To determine whether synaptic overgrowth in spin is caused by enhanced TGF-ß signaling, it was asked whether TGF-ß signaling is necessary for synaptic overgrowth in spin. The type II receptor mutation wit causes a severe decrease in bouton number at the NMJ. Type I TGF-ß receptors are known to function in concert with type II receptors, and the type I receptors tkv and sax participate in synaptic growth regulation in this system. Third instar larva mutant for sax or tkv have smaller neuromuscular synapses. This study confirms that there is a significant decrease in bouton number in wit, and that there is a similar decrease in bouton number in both tkv and sax. These receptors are shown to function in the larval motoneurons by demonstrating that pMAD staining in the cell bodies of larval motoneurons requires wit or sax. In this experiment, the larval CNS was costained with pMAD and anti-evenskipped, which labels a subset of motoneurons (Sweeney, 2002).
A genetic analysis of the TGF-ß receptor mutations wit, tkv, and sax in combination with spin demonstrates that TGF-ß signaling is necessary for synaptic overgrowth in spin. Heterozygous mutations in tkv, sax, and wit do not alter synaptic bouton numbers at the NMJ. Heterozygous mutations in tkv, sax, and wit suppress synaptic overgrowth when placed in the spin mutant background. Bouton numbers are significantly reduced in each case where a single copy of a receptor is mutated in combination with spin. Bouton numbers were quantified in each of the double mutant combinations of tkv, sax, or wit with spin. In each case, when both copies of a receptor were removed, synaptic overgrowth was suppressed in the spin mutant background further than when only a single copy of a receptor was mutated. These data demonstrate that TGF-ß receptor mutations suppress synaptic overgrowth in spin in a dose-dependent manner. Furthermore, since bouton numbers return to wild-type, or below wild-type levels, it demonstrates that TGF-ß signaling is necessary for synaptic overgrowth in spin. Taken together with the increase in bouton numbers seen in dad, these data support the conclusion that enhanced or misregulated TGF-ß signaling is a major determinant of synaptic overgrowth in spin. It is hypothesized that altered endosomal function due to loss of Spin causes enhanced TGF-ß signaling and subsequent synaptic overgrowth. Future experiments will be necessary to determine whether enhanced signaling is due to increased receptor number at the plasma membrane, or an inability to stop signaling within the late endosomal system (Sweeney, 2002).
How might endosomal malfunction be related to synaptic overgrowth? One possibility is that there is excessive or aberrant growth factor signaling from the endosome in spin mutations. It has recently become clear that signaling from activated plasma membrane receptors can continue in the endosomal compartment. Thus, the duration of any given signaling event could be controlled in the endosomal system, and a defect in this processing could dramatically prolong potent growth-related signals. Recent studies have demonstrated the importance of the late endosomal system for the regulation of intercellular signals such as wingless, Dpp, and epidermal growth factor. A variation of this model is based on well-known sorting decisions that determine whether receptors are delivered back to the plasma membrane or are degraded in the lysosome. If sorting decisions are altered in spin, it is possible that receptors are sorted back to the plasma membrane and cause enhanced growth factor signaling from the synaptic plasma membrane (Sweeney, 2002).
These models beg the question of whether synaptic overgrowth in spin is due to gross misregulation of synaptic signaling. The synaptic overgrowth observed in spin is extreme (>200% increase in bouton number) and far exceeds the effect of mutations in other synaptic signaling molecules which, in general, alter synaptic growth in the range of a 20%–50% change in bouton number. It is possible, therefore, that many different signaling molecules are altered simultaneously due to altered endosomal function in spin, thereby generating the observed synaptic overgrowth. However, if this were the case, one might predict that any disruption of the late endosomal system would cause synapse overgrowth. Previously, two synaptic endosomal proteins have also been examined for synaptic growth defects, Hook and Deep Orange. Hypomorphic mutations in these genes caused changes in synaptic growth of approximately 10%–25%, far less than that observed in spin. It is unlikely that the discrepancy between hook and dor versus spin is entirely due to differences in the severity of the mutations since even heterozygous spin larvae show an increase in synaptic growth that can exceed 30%, and hypomorphic mutations show synaptic growth that exceeds 100%. One interesting possibility is that the extraordinary overgrowth in spin mutants is caused by the misregulation of a specific signaling pathway that is not affected by Hook and Deep Orange-dependent protein trafficking (Sweeney, 2002).
Only two mutations have been identified that can cause synaptic expansion on a scale of 200% overgrowth: highwire and spin. highwire encodes a putative E3 ubiquitin ligase. Thus, both spin and highwire appear to function in protein trafficking or membrane sorting decisions, most likely in the endosomal/lysosomal system. In addition, two overexpression experiments have demonstrated this type of overgrowth; the overexpression of an ubiquitin hydrolase. These data suggest a logic for synaptic growth control. In this model, the pre- or post-synaptic release of a growth factor is the trigger for synaptic growth. However, the regulated release of growth factor is not instructive and may exceed the amount necessary for the precise growth of the synapse. Precision is achieved by sculpting excessive growth factor signaling in the endosomal/lysosomal system. Therefore, when the function of this system is perturbed, as in spin and highwire, dramatic synapse expansion ensues. An intriguing possibility is that the activity of the endosomal/lysosomal system could be modulated by intercellular signaling as well as intrinsic factors (Sweeney, 2002).
The spin phenotype is a candidate for a model of lysosomal storage disease. Strong spin mutations show a developmental defect in synaptic growth and function and are late pupal lethal. Hypomorphic, adult viable, mutations show reduced viability. Associated with these deficits is the accumulation of ceroid lipid pigment and GM2-ganglioside-like substance in neurons. In humans, there are a large number of lysosomal storage disorders that cause severe neurodegeneration, including Battens Disease, the most common childhood onset neurodegenerative disease. The parallels between spin and Battens Disease are extensive, including ceroid lipid accumulation and the presence of characteristic electron dense profiles within neurons. In addition, many known Batten Disease genes are lysosomal proteins. There are many ways to explain the cause of neurodegeneration with respect to endosomal/lysosomal malfunction, generally dealing with inappropriate protein trafficking. These data suggest that altered endosomal/lysosomal function, either pre- or post-synaptically, can have another important consequence -- the misregulation of potent and possibly diverse intercellular signaling systems (Sweeney, 2002).
Nakano, Y., et al. (2001). Mutations in the novel membrane protein Spinster interfere with programmed cell death and cause neural degeneration in Drosophila melanogaster. Mol. Cell. Biol. 21: 3775-3788. 11340170
Sweeney, S. T. and Davis, G. W. (2002). Unrestricted synaptic growth in spinster -- a late endosomal protein implicated in TGF-ß-mediated synaptic growth regulation. Neuron 36: 403-416. 12408844
date revised: 5 March 2003
Home page: The Interactive Fly © 1997 Thomas B. Brody, Ph.D.
The Interactive Fly resides on the
Society for Developmental Biology's Web server.