InteractiveFly: GeneBrief

almondex: Biological Overview | References

TM2 domain containing (TM2D) proteins are conserved in metazoans and encoded by three separate genes in each model organism species that has been sequenced. Rare variants in TM2D3 are associated with Alzheimer's disease (AD) and its fly ortholog almondex is required for embryonic Notch signaling. However, the functions of this gene family remain elusive. This study knocked-out all three TM2D genes (almondex, CG11103/amaretto, CG10795/biscotti) in Drosophila and found that they share the same maternal-effect neurogenic defect. Triple null animals are not phenotypically worse than single nulls, suggesting these genes function together. Overexpression of the most conserved region of the TM2D proteins acts as a potent inhibitor of Notch signaling at the gamma-secretase cleavage step. Lastly, Almondex is detected in the brain and its loss causes shortened lifespan accompanied by progressive motor and electrophysiological defects. The functional links between all three TM2D genes are likely to be evolutionarily conserved, suggesting that this entire gene family may be involved in AD (Salazar, 2021).

Alzheimer's disease (AD) is the most common neurodegenerative disease affecting the aging population and accounts for the large majority of age-related cases of dementia. AD is pathologically characterized by histological signs of neurodegeneration that are accompanied by formation of extracellular plaques and intra-neuronal tangles. Numerous studies have identified genetic factors that contribute to AD risk and pathogenesis. Rare hereditary forms of AD are caused by dominant pathogenic variants in APP (Amyloid Precursor Protein), PSEN1 (Presenilin 1) or PSEN2 (Presenilin 2). These three genes have been extensively studied using variety of experimental systems, and the resultant knowledge has led to greater understanding of how they contribute to the formation of extracellular plaques found in both familial and sporadic AD brains. PSEN1 and PSEN2 are paralogous genes that encode the catalytic subunit of the γ-secretase, a membrane-bound intramembrane protease complex. γ-secretase substrates include many type-I transmembrane proteins including APP as well as Notch receptors that play various roles in development and physiology (Salazar, 2021).

While studies of genes that cause familial AD have been critical in providing a framework to study pathogenic mechanisms of this disorder, pathogenic variants in APP and PSEN1/2 are responsible for only a small fraction of AD cases. Familial AD can be distinguished from more common forms of AD because most patients with pathogenic APP or PSEN1/2 variants develop AD before the age of 65 [early-onset AD (EOAD)]. The majority (>95%) of AD cases are late-onset (LOAD, develops after 65 years of age), sporadic and idiopathic in nature. In these patients, it is thought that multiple genetic and environmental factors collaborate to cause damage to the nervous system that converges on a pathway that is affected by APP and PSEN1/2. To reveal common genetic factors with relatively small effect sizes, multiple genome-wide association studies (GWAS) have been performed and identified over 40 loci throughout the genome that confer increase risk to developing AD. The most notable risk-factors are variant alleles in Apolipoprotein E (APOE). Although the precise molecular mechanism by which different alleles of APOE increase or decrease the risk of AD has been extensively debated, a number of studies have proposed that this gene is involved in the clearance of toxic Aβ peptides. A recent meta-analysis has also identified ADAM10 (encoding a β-secretase enzyme that cleaves APP and Notch) as an AD associated locus, suggesting that genes involved in familial EOAD and sporadic LOAD may converge on the same molecular pathway. Functional studies of these and other newly identified risk factors for AD are critical to fully understand the etiology of this complicated disease that lack effective treatments or preventions (Salazar, 2021).

A rare missense variant (rs139709573, NP_510883.2:p.P155L) in TM2D3 (TM2 domain containing 3) is significantly associated with increased risk of developing LOAD through an exome-wide association study in collaboration with the CHARGE (Cohorts for Heart and Aging Research in Genomic Epidemiology) consortium. This variant was also associated with earlier age-at-onset that corresponds to up to 10 years of difference with a hazard ratio of 5.3 after adjusting for the common ε4 allele of APOE. Although the function of this gene in vertebrates was unknown and this missense variant was not predicted to be pathogenic based on multiple variant pathogenicity prediction algorithms including SIFT, PolyPhen and CADD, it was experimentally demonstrated that p.P155L has deleterious consequences on TM2D3 function based on an assay established using Drosophila embryos. The Drosophila ortholog of TM2D3, almondex (amx), was initially identified based on an X-linked female sterile mutant allele (amx¹) generated through random mutagenesis. Although homozygous or hemizygous (over a deficiency) amx¹ mutant females and hemizygous (over Y chromosome) males are viable with no morphological phenotypes, all embryos laid by amx¹ hemi/homozygous mothers exhibit severe developmental abnormalities including expansion of the nervous system at the expense of the epidermis (Shannon, 1973; Lehmann, 1983). This 'neurogenic' phenotype results when Notch signaling mediated lateral inhibition is disrupted during cell-fate decisions in the developing ectoderm . By taking advantage of this scorable phenotype, this study showed that the maternal-effect neurogenic phenotype of amx¹ hemizygous females can be significantly suppressed by introducing the reference human TM2D3 expressed under the regulatory elements of fly amx, but TM2D3p.P155L expressed in the same manner fails to do so (Jakobsdottir, 2016). This showed that the function of TM2D3 is evolutionarily conserved between flies and humans, and the molecular function of TM2D3 that is relevant to LOAD may also be related to Notch signaling. More recently, another rare missense variant (p.P69L) in this gene has been reported in a proband that fit the diagnostic criteria of EOAD or frontotemporal dementia, indicating that other TM2D3 variants may be involved in dementia beyond LOAD (Salazar, 2021).

TM2D3 is one of three highly conserved TM2 domain containing (TM2D) proteins encoded in the human genome. The two other TM2 domain-containing proteins, TM2D1 and TM2D2, share a similar protein domain structure with TM2D3, and each protein is encoded by a highly conserved orthologous gene in Drosophila that has not been functionally characterized. All TM2D proteins have a predicted N-terminal signal sequence and two transmembrane domains that are connected through a short intracellular loop that are found close to the C-terminus. Within this loop, there is an evolutionarily conserved DRF (aspartate-arginine-phenylalanine) motif, a sequence found in some G-protein coupled receptors that mediates their conformational change upon ligand binding. The extracellular region between the signal sequence and first transmembrane domain is divergent in different species as well as among the three TM2D containing proteins. In contrast, the sequences of the two transmembrane domains as well as the intracellular loop is highly conserved throughout evolution as well as between the three TM2 domain containing proteins. The three proteins also have short C-terminal extracellular tails that are evolutionarily conserved within orthologs but vary among the three proteins (e.g. TM2D1 has a slightly longer C'-tail than TM2D2 and TM2D3). The molecular functions of these conserved and non-conserved domains of TM2D proteins are unknown (Salazar, 2021).

Amx has been proposed to function at the γ-secretase cleavage step of Notch activation based on a genetic epistasis experiment performed previously (Michellod, 2008). Notch signaling activation is initiated by the binding of the Notch receptor to its ligands (Delta or Serrate in Drosophila). This induces a conformational change of Notch to reveal a cleavage site that is recognized by ADAM10 [encoded by kuzbanian (kuz) in Drosophila]]. Notch receptor that has undergone ADAM10 cleavage (S2 cleavage) is referred to as NEXT (Notch extracellular truncation) and becomes a substrate for γ-secretase. NEXT that is cleaved by γ-secretase (S3 cleavage) releases its intracellular domain (NICD), which then translocates to the nucleus and regulates transcription of downstream target genes. To determine how amx regulates Notch signaling, Michellod and Randsholt (2008) attempted to suppress the embryonic neurogenic phenotype of embryos produced from amx¹ mutant mothers by zygotically overexpressing different forms of Notch using a heat-shock promoter. While NICD was able to weakly suppress the neurogenic defect, NEXT was not able to do so, suggesting that Amx somehow modulates the cleavage step that involves the γ-secretase complex. However, because the phenotypic suppression observed by NICD in this study was very mild and since the authors used Notch transgenes that were inserted into different regions of the genome (thus NICD and NEXT are likely to be expressed at different levels and cannot be directly compared), additional data is required to fully support this conclusion (Salazar, 2021).

In this study, clean null alleles of all three Drosophila TM2D genes were generated using CRISPR/Cas9-mediated homology directed repair (HDR), and their functions assessed in vivo. Surprisingly, CG10795 (TM2D1) and CG11103 (TM2D2) knockout flies are phenotypically indistinguishable from amx (TM2D3) null animals, displaying severe maternal-effect neurogenic phenotypes. Double- and triple-knockout animals were assessed to determine whether these three genes have redundant functions in other Notch signaling dependent contexts during development. The triple-knockout of all TM2D genes did not exhibit any obvious morphological phenotypes but shared the same maternal-effect neurogenic phenotype similar to the single null mutants, suggesting these three genes function together. Further evidence is provided that Amx functions on γ-secretase to modulate Notch signaling in vivo, and a previously unknown role of this gene in the maintenance of neural function in adults was uncovered (Salazar, 2021).

This study functionally characterized TM2D genes through gene knockout and over-expression strategies in Drosophila to gain biological knowledge on this understudied but evolutionarily conserved gene family that has been implicated in AD. It was first shown that the knockout allele of amx (Drosophila homolog of TM2D3) generated by CRISPR is phenotypically indistingusiable from the classic amx¹ allele and displays female sterility and a maternal-effect neurogenic defect. Recently, it was reported that this allele also shows a maternal-effect inductive signaling defect to specify the mesoectoderm during embryogenesis which is another Notch-dependent event (Das, 2020), demonstrating that amx is maternally required for multiple Notch signaling dependent processes during embryogenesis. In addition, the first knockout alleles of amrt (Drosophila ortholog of TM2D2) and bisc (Drosophila ortholog of TM2D1) were generated ,and each null allele that was phenotypically documented mimics the loss of amx. Furthermore, it was revealed that the triple knockout of all three TM2D genes in Drosophila show identical maternal-effect neurogenic phenotypes without exhibiting other obvious Notch signaling-related developmental defects. Moreover, although the over-expression of the full-length Amx do not cause any scorable defects, it was serendipitously found that expression of a truncated form of Amx that lacks the majority of the extracellular domain can strongly inhibit Notch signaling in the developing wing imaginal disc, a tissue in which all three fly TM2D genes are expressed endogenously. Through genetic epistatic experiments using newly generated UAS-Notch transgenic lines, this inhibitory effect was mapped to the γ-secretase cleavage step of Notch activation. Subsequently, it was found that amx null animals have a shortened lifespan, a phenotype that can be rescued by reintroduction of Amx in neurons. This shortening of lifespan phenotype was also seen in amrt and bisc null flies, suggesting that these three genes may also function together in the aging brain. Finally, through assement of climibing behavior and electrophysiological recordings of the giant fiber system, amx null flies were shown to have age-dependent decline in neural function. In summary, it was demonstrated that all three TM2D genes play critical roles in embryonic Notch signaling to inhibit the epithelial-to-neuron cell fate transformation as maternal-effect genes, and that amx is required for neuronal maintenance in the adult nervous system, a function that may be related to the role of human TM2D3 in AD (Salazar, 2021).

Within commonly used genetic model organisms, TM2D genes are found in multicellular animals (both in invertebrates and in vertebrates) but are absent in yeasts (e.g. Saccharomyces cerevisiae, Schizosaccharomyces pombe) and plants (e.g. Arabidopsis thaliana), suggesting that this family of genes arose early in the metazoan lineage. In humans and flies, there are three TM2 domain-containing genes (TM2D1, TM2D2, TM2D3 in Homo sapiens, and three in Drosophilam, bisc, amrt, amx, each corresponding to a single gene in the other species. Interestingly, this 1:1 ortholog relationship is also seen in mouse (Tm2d1, Tm2d2, Tm2d3), frog (Xenopus tropicalis: tm2d1, tmd2d, tm2d3) zebrafish (Danio rerio: tm2d1, tm2d2, tm2d3) and worm (Caenorhabditis elegans: Y66D12A.21, C02F5.13, C41D11.9). In general, most genes have more paralogous genes in humans compared to flies (for example, one Drosophila Notch gene corresponding to four NOTCH genes in human) as vertebrates underwent two rounds of whole-genome duplication (WGD) events during evolution. Furthermore, teleosts including zebrafish underwent an extra round of WGD, leading to formation of extra duplicates in 25% of all genes (e.g. one NOTCH1 gene in human corresponds to notch1a and notch1b in zebrafish). Hence it is interesting that each of the three TM2D genes remained as single copy genes in various species despite whole genome level evolutional changes, suggesting that there may have been some selective pressure to keep the dosage of these genes consistent and balanced during evolution (Salazar, 2021).

Although the in vivo functions of TM2D1 and TM2D2 have not been studied extensively in any organism, several lines of studies performed in cultured cells suggest that these genes may also play a role in AD pathogenesis. Through an yeast-two hybrid screen to identify proteins that bind to Aβ42, TM2D1 was identified, and this protein was referred to as BBP (beta-amyloid binding protein). TM2D1 was also shown to interact with Aβ40, a non-amyloidogenic form of Aβ, and there is preliminary data that it also binds to APP. The interaction between Aβ peptides and TM2D1 was shown to require the extracellular domain as well as a portion of the first transmembrane domain of TM2D1. Because overexpression of TM2D1 in a human neuroblastoma cell line (SH-SY5Y) increased the sensitivity of these cells to cell death caused by incubation with aggregated Aβ and since the DRF motif was found to be required for this activity, the authors of this original study proposed that TM2D1 may function as a transmembrane receptor that mediates Aβ-toxicity. However, a follow-up study from another group refuted this hypothesis by providing data that TM2D1 is not coupled to G proteins using a heterologous expression system in Xenopus oocytes. Therefore, although the physical link between TM2D1 and Aβ42 is intriguing, the significance of this interaction and its role in AD pathogenesis has been obscure (Salazar, 2021).

Surprisingly, loss of bisc/TM2D1 and amrt/TM2D2 were phenotypically indistinguishable from the loss of amx/TM2D3 in Drosophila. The zygotic loss of each gene did not exhibit any strong developmental defects into adulthood, despite their relatively ubiquitous expression pattern according to large transcriptome datasets Furthermore, the loss of either amrt or bisc caused a reduction of lifespan, again mimicking the loss of amx. More interestingly, the triple null mutants are viable and do not exhibit any morphological defect, suggesting that these genes are not required zygotically during development. In contrast, maternal loss of any single TM2D gene causes a strong embryonic neurogenic defect, which is also seen in embryos laid by triple knockout animals. Neurogenic defect is a classical phenotype in Drosophila that was originally reported in the mid-1930s, and the study of mutants that show this phenotype led to the establishment of the core Notch signaling pathway in the late 1980s and early 1990s. Although the study of neurogenic phenotypes and genes has a long history, this phenotype is a very rare defect that has so far been associated with only 19 genes according to FlyBase, prior to this work. Seven genes show this defect as zygotic mutants [aqz, bib, Dl, E(spl)m8-HLH, mam, N and neur], seven genes are zygotically-required essential genes with large maternal contributions (hence the need to generate maternal-zygotic mutants by generating germline clones to reveal the embryonic neurogenic defect) [Gmd, Gmer, gro, Nct, O-fut1, Psn and Su(H)], one gene has only been investigated by RNAi (Par-1) and four genes including amx are non-essential genes that show maternal-effect neurogenic defects (amx, brn, egh, pcx). Hence, this study has revealed two new genes that are evolutionarily closely linked to amx in this Notch signaling related process (Salazar, 2021).

The similarity of sequences and phenotypes caused by loss of amx, amrt and bisc suggests that the proteins encoded by these genes may function together. The lack of additive or synergistic phenotypes in the double and triple null mutant flies also support this idea. Interestingly, high-throughput proteomics experiments based on co-immunoprecipitation mass spectrometry (co-IP/MS) from human cells have detected physical interactions between TM2D1-TM2D3 and TM2D2-TM2D3, suggesting these proteins may form a protein complex. Further biochemical studies will be required to clarify the functional relationship between the three TM2D proteins. Two additional mammalian datasets further support the hypothesis that these three proteins functions together. First, all three TM2D genes were identified through a large scale cell-based CRISPR-based screen to identify novel regulators of phagocytosis. Individual knockout of TM2D genes in a myeloid cell line was sufficient to cause a similar phagocytic defect based on the parameters the authors screened for (e.g. substrate size, materials to be engulfed). Although the authors of this study did not generate double or triple knockout cell lines to determine whether there were any additive or synergistic effects when multiple TM2D genes were knocked out, this suggests that these three genes may function together in phagocytosis. The authors further note that because these genes are broadly expressed in diverse cell types beyond phagocytic cells in the nervous system, they may play other roles in the brain, consistent with the finding that amx is required neuronally to maintain a normal lifespan. It would be interesting to explore whether age-dependent phenotypes seen in TM2D null mutants can also be suppressed through glial specific rescue experiments to understand the in vivo significance of this group's in vitro findings. Second, preliminary phenotypic data from the International Mouse Phenotyping Consortium indicates that single knockout of mice of Tm2d1, Tm2d2 and Tm2d3 are all recessive embryonic lethal prior to E18.5. Although detailed characterization of these mice will be required and further generation of a triple knockout line is desired, the shared embryonic lethality may indicate that these three genes potentially function together in an essential developmental paradigm during embryogenesis in mice (Salazar, 2021).

Attempts to unravel the function of Amx through overexpression of the full-length protein was uninformative since this manipulation did not cause any scorable phenotype. However, it was serendipitously found that a truncated form of Amx, which only contains the most conserved region of TM2D family proteins, has the capacity to strongly inhibit Notch signaling during wing and notum development. These results were surprising because no wing or bristle defects were seen in the triple TM2D gene family knockout flies, even though all three genes are endogenously expressed in the wing imaginal disc. Through epistasis experiments using a set of new UAS-Notch transgenic lines, AmxΔECD to inhibit Notch signaling at the S3 cleavage step which is mediated by the γ-secretase complex. This data is consistent with earlier epistasis experiments performed on amx¹ in the context of embryonic neurogenesis, further supporting the idea that amx may regulate γ-secretase function in vivo. It was further determined that over-expression of AmxΔECD in the wing imaginal disc causes an accumulation of Notch protein within the cell and at the cell membrane. This phenotype is similar to what is seen upon knockdown of Psn in the wing imaginal disc, which is consistent with earlier findings showing Notch accumulation at the cell membrane in neuroblasts of Psn mutants. In summary, this study showed that ectopic over-expression of a portion of Amx that is conserved among all TM2D proteins causes a strong Notch signaling defect by disrupting a process that realates to γ-secretase, providing additional links between Amx/TM2D3 and AD (Salazar, 2021).

By aging the amx null male flies that are visibly indistinguishable from the control flies (amx null flies with genomic rescue constructs), it was found that loss of amx causes a significant decrease in lifespan, a phenotype that was observed in amrt and bisc null flies. By generating a functional genomic rescue transgene in which Amx is tagged with an epitope tag, this protein was found to be expressed in the adult brain. The short lifespan defect of amx null flies was effectively rescued by reintroduction of Amx in neurons, demostrating the importance of this gene in this cell type. In addition to the lifespan defect, amx null flies also exhibited an age-dependent climing defect, suggesting that their neural function declines with age. By further performing electrophysiological recordings of the giant fiber system, which is a model circuit that is frequently used in neurological and neurodegenerative research in Drosophila, it was found that there is indeed an age-dependent decline in the integrity of this circuit. Through this assay, it was observed that the DLM branch of the giant fiber system begins to show failures earlier than the TTM branch. The DLM is activated by giant fiber neurons that chemically synapse onto PSI (peripherally synapsing interneuron) neurons through cholinergic synapses, which in turn chemically synapse onto motor neurons (DLMn which are glutamatergic) through cholinergic connections. The TTM, in contrast, is activated by giant fiber neurons that electrically synapse onto motor neurons (TTMn which are glutamatergic) through gap junctions, causing a more rapid response. Considering the difference in the sensitivity of the two branches, cholinergic neurons/synapses may be more sensitive to the loss of amx, a neuronal/synaptic subtype that is severely affected in AD in an age-dependent manner (Salazar, 2021).

How does amx maintain neuronal function in aged animals and is this molecular function related to AD? One potential molecular mechanism is through the regulation of γ-secretase in the adult brain. By knocking down subunits of the γ-secretase complex, Psn and Nct, specifically in adult neurons, it was shown that reduction of γ-secretase function decreases lifespan, which was associated with climbing defects as well as histological signs of neurodegeneration. The requirement of γ-secretase components in neuronal integrity has also been reported in mice, suggesting this is an evolutionarily conserved phenomenon. Interestingly, the role of the γ-secretase complex in neuronal maintenance is unlikely to be due to defects in Notch signaling because neurodegeneration has not been observed upon conditional removal of Notch activity in post-developmental brains in flies and in mice. While the precise function of γ-secretase in neuronal maintenance is still unknown, several possibilities including its role in regulating mitochondrial morphology and calcium homeostasis has been proposed based on studies in C. elegans and mice. Investigating whether Amx does indeed regulate γ-secretase in adult neurons and whether it impacts the aforementioned processes will likely facilitate understanding on how this gene regulates neuronal health. Furthermore, considering that TM2D3 and other TM2D genes have been proposed to function in phagocytic cells, and because phagocytosis process plays many roles beyond engulfment of toxic Aβ molecules in the nervous system, Amx may also be playing a role in engulfing unwanted materials that are harmful for the adult brain. For example, loss of the phagocytic receptor Draper in glia cells causes age-dependent neurodegeneration that is accompanied by accumulation of non-engulfed apoptotic neurons throughout the fly brain. Interestingly, a recent study has shown that over-expression of phagocytic receptors can also promote neurodegeneration, indicating the level of phagocytic activity needs to be tightly controlled in vivo. Further studies of amxΔ mutants (as well as amxΔ amrtΔ biscΔ triple mutants) in the context of phagocytosis will likely reveal the precise molecular function of Amx and other TM2D proteins in this process (Salazar, 2021).

Finally, could there be any molecular link between the role of TM2D genes in Notch signaling (proposed based on experiments in Drosophila) and phagocytosis (revealed based on mammalian cell culture based studies), or are they two independent molecular functions of the same proteins? All TM2D proteins have two transmembrane domains connected by a short intracellular loop, making them an integral membrane protein. By tagging the amx genomic rescue construct with a 3xHA tag that does not influence the function of Amx, it was observed that 3xHA::Amx is localized to the plasma membrane as well as intracellular puncta, which likely reflects intracellular vesicles. Interestingly in embryos laid by amxΔ mutant females, a mild and transient but significant alteration was observed in Notch distribution during early embryogenesis. Moreover, a strong accumulation of Notch was observed when AmxΔECD was overexpressed in the developing wing primordium. These data indicate that Amx may affect protein trafficking, which in turn may impact the processing of Notch by the γ-secretase complex. Indeed, Notch signaling is highly regulated by vesicle trafficking and alterations in exocytosis, endocytosis, recycling and degradation all impact the signaling outcome. In fact, multiple studies have proposed that γ-secretase cleavage occurs most effectively in acidified endocytic vesicles. Hence, while amx may be specifically required for the proper assembly or function of the γ-secretase complex, it may alternatively be necessary to bring Notch and other substrates to the proper subcellular location for proteolytic cleavages to occur efficiently. The subcellular localization differences observed between the punctate AmxΔECD, which causes a dramatic Notch signaling defect accompanied by mistrafficking of Notch, and the membranous AmxFL, which does not have this effect, may further support the idea that Amx function as a trafficking module. Similar to Notch signaling, phagocytosis also requires coordination of many cellular trafficking events to expand the plasma membrane to form a phagophore, internalize the particle of interest to generate a phagosome, and fuse the phagosome to lysosomes to degrade its content. By studying the role of TM2D genes and proteins in embryonic Notch signaling, phagocytosis and age-dependent neuronal maintenance, the precise molecular and cellular function of this evolutionarily conserved understudied protein family will likely be understood, which will likely lead to further understanding of molecular pathogenesis of AD and other human diseases. Considering the phenotypic similarities of amrt and bisc to amx in Drosophila embryonic neurogenesis, the similarities between TM2D1-3 in human cells in the context of phagocytosis, and the similarities of Tm2d1-3 knockout mice in the context of embryogenesis, it is proposed that studies of rare genetic variants, epigenetic regulators or proteomic changes in other TM2D genes may reveal novel risk factors or biomarkers in epidemiologic study of AD and other forms of dementia (Salazar, 2021).

Maternal almondex, a neurogenic gene, is required for proper subcellular Notch distribution in early Drosophila embryogenesis

Notch signaling plays crucial roles in the control of cell fate and physiology through local cell-cell interactions. Drosophila almondex, which encodes an evolutionarily conserved double-pass transmembrane protein, was identified in the 1970s as a maternal-effect gene that regulates Notch signaling in certain contexts, but its mechanistic function remains obscure. This study examined the role of almondex in Notch signaling during early Drosophila embryogenesis. In addition to being required for lateral inhibition in the neuroectoderm, almondex was also found to be partially required for Notch signaling-dependent single-minded expression in the mesectoderm. Furthermore, it was found that almondex is required for proper subcellular Notch receptor distribution in the neuroectoderm, specifically during mid-stage 5 development. The absence of maternal almondex during this critical window of time caused Notch to accumulate abnormally in cells in a mesh-like pattern. This phenotype did not include any obvious change in subcellular Delta ligand distribution, suggesting that it does not result from a general vesicular-trafficking defect. Considering that dynamic Notch trafficking regulates signal output to fit the specific context, it is speculated that almondex may facilitate Notch activation by regulating intracellular Notch receptor distribution during early embryogenesis (Das, 2020).

Analysis of amx^N, the first clean null allele of amx, clearly showed that maternal amx is required for Notch signaling during early embryogenesis; this is consistent with previous findings for the amx¹ and amxm alleles. A study using the amxm allele reported that amx is also required zygotically for imaginal disc development (Michellod et al., 2003). However, as with amx¹ zygotic mutants, amxN zygotic mutant flies had no morphological phenotype. Considering that amx^m, is a complex allele with the potential to affect other genes, the zygotic phenotypes reported for the amx^m, allele are likely due to defects in genes other than amx (Das, 2020).

amx^mz embryo phenotypes were examined; amx is required not only for neuroectoderm specification during mid-embryogenesis but also for mesectoderm specification during early embryogenesis. Notch signaling is required for sim expression in mesectodermal cells; in various mutants of core Notch signaling pathway genes, including Notch and Suppressor of Hairless, sim expression is restricted to a very few cells. Interestingly, sim expression was reduced only partially in amx^mz embryos, although these embryos exhibited a strong neurogenic phenotype with full penetration. These results show a partial requirement of amx for activating Notch signaling in mesectodermal cells. This is reminiscent of pcx, another maternal neurogenic gene with a strong neuroblast-segregation defect, since pcx is partially required for activating sim during mesectoderm specification. One interpretation of this differential contribution is that amx and pcx are absolutely required in neuroectodermal cells, but only partially required for mesectoderm specification. However, the molecular mechanisms that would account for such differences remain elusive (Das, 2020).

This analysis demonstrates that the subcellular distribution of the Notch receptor, but not the Dl ligand, depends on a function of maternal amx during early cellularization— specifically in mid-stage 5, when the structure of the apical plasma membrane with its microvilli changes drastically. Immunostaining did not show any noticeable structural changes in F-Actin or the apical plasma membrane in the absence of amx function, and the ER, Golgi, and endosome morphology also appeared unaltered. Thus, amx appears to have a specific role in regulating Notch trafficking during this stage (Das, 2020).

In amx^mz embryos at mid-stage 5, Notch accumulated abnormally in a mesh-like subcellular structure but without any marked increase in colocalization with markers for the ER, Golgi apparatus, or endosomal compartments. Thus, it is clear that Notch mislocalizes in the absence of amx function, but it is not clear which intracellular compartment(s) it localizes to. It is important to note that Notch mislocalization in amx^mz embryos is not caused by aberrant Notch signaling, since Notch localizes normally in pcx^mz embryos at mid-stage 5. It is tempting to speculate, however, that Notch mislocalization might contribute to defects in Notch signaling in the absence of amx function (Das, 2020).

Vesicular Notch trafficking can activate or inhibit Notch signaling according to the context. Importantly, endocytic Notch trafficking facilitates S3 Notch cleavage, likely because γ-secretase cleavage is a pH-dependent protease that is more active in acidic environments, such as the late endosome. Considering that epistasis analysis of amx and full-length and activated forms of Notch placed amx in the S3 cleavage step of signal transduction, it would be interesting to further explore the relationship between γ-secretase-mediated S3 cleavage and defective Notch trafficking in the absence of amx function. It would also be interesting to investigate exactly how Amx is involved in Notch trafficking through live-imaging analyses of Notch intracellular transportation (Das, 2020).

Implication of the Drosophila beta-amyloid peptide binding-like protein AMX in Notch signaling during early neurogenesis

Lateral inhibition provides a mechanism to regulate neuroblast specification during early neurogenesis in Drososphila melanogaster embryos. This mechanism is mediated by the highly conserved Notch pathway. Defective lateral inhibition results in CNS hyperplasia at the expense of ectoderm development, hence genes causing this defect are called neurogenic. D. melanogaster almondex (amx) is a maternal neurogenic gene, crucially required for embryonic lateral inhibition. Genetic interaction studies previously revealed amx as a positive Notch pathway partner in several processes, acting potentially upstream of Notch. This study shows that embryonic overexpression of Notch intracellular domain partially rescues maternal lack of amx, suggesting a role for AMX at the level of Notch processing. The molecular data reveal that amx is ubiquitously expressed and encodes a conserved putative transmembrane protein, composed of several domains that are differently required for amx function in the fly. Sequence comparisons identify AMX as a Drosophila Beta-amyloid peptide Binding Protein (BBP) family member, a BBP-like protein or dBLP. Based on these data, the potential molecular function of AMX in early neurogenesis is discussed (Michellod, 2008).

A role for the Drosophila neurogenic genes in mesoderm differentiation

The neurogenic genes genes also control mesoderm development. Embryonic cells that express nautilus are overproduced in each of seven neurogenic mutants (Notch, Delta, Enhancer of split, big brain, mastermind, neuralized, and almondex), at the apparent expense of neighboring, nonexpressing mesodermal cells. The mesodermal defect does not appear to be a simple consequence of associated neural hypertrophy, suggesting that the neurogenic genes may function similarly and independently in establishing cell fates in both ectoderm and mesoderm. Altered patterns of beta 3-tubulin and myosin heavy chain gene expression in the mutants indicate a role for the neurogenic genes in development of most visceral and somatic muscles (Corbin, 1991).

Search PubMed for articles aboutn Drosophila Almondex

Corbin, V., Michelson, A. M., Abmayr, S. M., Neel, V., Alcamo, E., Maniatis, T., Young, M. W. (1991). A role for the Drosophila neurogenic genes in mesoderm differentiation. Cell, 67(2):311-323 PubMed ID: 1913825

Das, P., Salazar, J. L., Li-Kroeger, D., Yamamoto, S., Nakamura, M., Sasamura, T., et al. (2020). Maternal almondex, a neurogenic gene, is required for proper subcellular Notch distribution in early Drosophila embryogenesis. Dev Growth Differ. 62: 80-93. PubMed ID: 31782145

Jakobsdottir J, van der Lee SJ, Bis JC, Chouraki V, Li-Kroeger D, Yamamoto S, et al. Rare Functional Variant in TM2D3 is Associated with Late-Onset Alzheimer's Disease. Haines JL, editor. PLOS Genet. 2016;12: e1006327. PubMed ID: 27764101

Lehmann R, Jimenez F, Dietrich U, Campos-Ortega JA. On the phenotype and development of mutants of early neurogenesis in Drosophila melanogaster. Wilhelm Roux's Arch Dev Biol. 1983;192: 62–74. PubMed ID: 28305500

Michellod MA, Randsholt NB. Implication of the Drosophila beta-amyloid peptide binding-like protein AMX in Notch signaling during early neurogenesis. Brain Res Bull. 2008;75: 305–309. PubMed ID: 18331889

Masuda, W., Yamakawa, T., Ajima, R., Miyake, K., Umemiya, T., Azuma, K., Tamaru, J. I., Kiso, M., Das, P., Saga, Y., Matsuno, K., Kitagawa, M. (2023). TM2D3, a mammalian homologue of Drosophila neurogenic gene product Almondex, regulates surface presentation of Notch receptors. Sci Rep, 13(1):20913 PubMed ID: 38016980

Salazar, J. L., Yang, S. A., Lin, Y. Q., Li-Kroeger, D., Marcogliese, P. C., Deal, S. L., Neely, G. G., Yamamoto, S. (2021). TM2D genes regulate Notch signaling and neuronal function in Drosophila. PLoS Genet, 17(12):e1009962 PubMed ID: 34905536

Shannon MP. The development of eggs produced by the female-sterile mutant almondex of Drosophila melanogaster. J Exp Zool. 1973;183: 383–400. PubMed ID: 4633255

date revised: 20 September 2024

Home page: The Interactive Fly © 2024 Thomas Brody, Ph.D.