EvoprintHD of senseless

BIOLOGICAL OVERVIEW

The Lyra mutation was first described by Jerry Coyne in 1935. Lyra causes recessive pupal lethality and adult heterozygous Lyra mutants exhibit a dominant loss of the anterior and posterior wing margins. Unlike many mutations that cause loss of wing tissue (e.g., scalloped, Beadex, cut, and apterous), Lyra wing discs do not exhibit increased necrotic or apoptotic cell death, nor do they show altered BrdU incorporation. However, during wing disc eversion, loss of the anterior and posterior wing margins is apparent. senseless (sens), a gene that is necessary and sufficient for peripheral nervous system (PNS) development (Nolo, 2000), is allelic to Lyra. Lyra/Sens is a nuclear protein with four Zn fingers that is expressed and required in the sensory organ precursors (SOPs) for proper proneural gene expression. Ectopic expression of Lyra in many ectodermal cells causes induction of PNS external sensory organ formation and is able to recreate an ectopic proneural field. Hence, Lyra is both necessary and sufficient for PNS development. Proneural genes activate Lyra expression (Nolo, 2000 and 2001).

Lyra alleles are neomorphic alleles of sens that cause ectopic expression of Senseless in the wing pouch. Similarly, overexpression of Senseless in the wing disc causes loss of wing margin tissue, thereby mimicking the Lyra phenotype. Lyra mutants display aberrant expression of Delta, Vestigial, Wingless, and Cut. As in Lyra mutants, overexpression of Senseless in some areas of the wing pouch also leads to loss of Wingless and Cut. Thus overexpression of Senseless causes a severe reduction in Notch signaling that in turn may lead to decreased transcription of several key genes required for wing development, leading to a failure in cell proliferation and loss of wing margin tissue (Nolo, 2001).

The onset of PNS development is marked by the expression of proneural genes of the achaete-scute complex (AS-C), atonal, or amos in clusters of ectodermal cells, termed the proneural clusters. The SOP is then selected from a small group of cells (the proneural field) within the proneural cluster that accumulates higher levels of proneural proteins than neighboring cells. Eventually, the SOP accumulates the highest levels of proneural proteins. The proneural basic helix loop helix (bHLH) proteins (Scute, Achaete, Asense, Atonal, and Amos) then implement the neuronal differentiation pathway as functional heterodimers with the Daughterless protein (Nolo, 2000 and references therein).

The upregulation of proneural gene expression in the SOP and the downregulation in adjacent ectodermal cells is mediated through a signaling pathway in which the genes of the Notch pathway, including Notch (N), Delta (Dl), Suppressor of Hairless [Su(H)], and the Enhancer of split complex genes [E(spl)-C], play a pivotal role. Loss-of-function mutations in these genes cause most or all cells of the proneural cluster to assume the SOP fate. Although it has been proposed that accumulation of proneural gene products in the SOP via Notch signaling is stochastic, the SOP always occupies a very similar position in the cluster. It has therefore been proposed that prepatterning factors predetermine a small subset of cells in the proneural cluster. One of these cells will become the SOP by accumulating more proneural protein than its neighbors and will produce more Delta protein. Although the upregulation of proneural gene expression is thought to be a prerequisite for sensory organ development, it is thought that several other unidentified proteins are required for the process (Nolo, 2000 and references therein).

Lyra/Senseless is a novel proneural-like protein. Severe loss-of-function mutations in sens abolish the further upregulation and maintenance of proneural gene expression in the SOPs. Interestingly, ectopic expression of Sens induces ectopic PNS organs, causing large tufts of bristles. Sens is both necessary and sufficient for PNS organ development but seems to require the activity of proneural proteins. sens encodes a transcription factor that enhances and maintains proneural gene expression in the SOP and is able to create a proneural cluster when ectopically expressed. Sens is therefore an essential component of the proneural Notch signaling pathway required for proper SOP differentiation (Nolo, 2000).

The embryonic PNS of Drosophila consists of three types of sensory transducers: the multiple dendritic (md) neurons; the external sensory organs (es), and the chordotonal organs. The latter two types of organs consist of four cells: a neuron, a glial cell, and two support cells. These originate from the SOPI that produces the SOPII a and b. Mutations in sens cause an extensive loss of many cells of the embryonic PNS, as revealed by staining with anti-Couch Potato, a marker that labels all nuclei of the PNS cells. Labeling with monoclonal antibody 22C10, a neural-specific antibody, shows that most neurons are absent. The few neurons that remain are mostly of the md type, but the md neurons that are lost are dependent on scute, atonal, and amos. Staining with anti-Prospero, a marker for PNS glial cells, reveals a very severe loss of glial cells. Staining with anti-Cut shows a significant loss of es support cells. In summary, these data show that all types of PNS cells in sens mutants are affected and that most cells are absent in mature stage 16 embryos (Nolo, 2000).

In the embryonic PNS nervous system, Sens is expressed at low levels in some cells surrounding the SOPIs and at high levels in SOPIs, SOPIIs, and differentiating progeny. In imaginal discs, Sens expression is first observed at low levels in ectodermal cells in proximity to many SOPs. This domain may correspond to the proneural field, which has been shown to accumulate higher levels of Achaete-Scute expression than other cells of the proneural cluster. However, levels of Sens protein expression are dramatically enhanced in the SOPIs. Sens is expressed prior to a typical SOP marker, A101 lacZ, but enhanced expression of Sens in the presumptive SOP is often coincident with A101 lacZ expression (Nolo, 2000).

The SOPI and SOPII cells are present in sens mutant embryos (Salzberg, 1994). Hence, there are two possible alternatives to explain the phenotype: the SOPII cells fail to divide, or the SOPII cells and/or their progeny are eliminated by apoptosis. Double labeling with terminal deoxynucleotidyl transferase (TUNEL) and 22C10 shows dying cells in the PNS of stage 16 sens embryos but not in wild-type control embryos. Similar data were obtained by in situ hybridization with grim, a marker for apoptotic cells in Drosophila. Since no alterations are observed in the expression of dacapo, a cyclin-dependent kinase inhibitor expressed in terminally dividing cells of the PNS, it is concluded that the severe loss of PNS cells in late embryos is due to apoptosis of SOPIIs and/or differentiating cells of the PNS (Nolo, 2000).

If Senseless is required for proneural expression, ectopic expression of Sens may induce proneural gene expression. Indeed, ectopic expression of Sens using the dpp-GAL4 driver causes robust expression of Sens in the expected wing stripe. This expression causes the formation of numerous bristles and sensilla campaniforma in the adult wing in proximity of the third wing vein where dpp is normally expressed. Similarly, ectopic expression of Sens in the leg disc causes many supernumerary bristles in the sternopleural area as well as in more distal portions of the leg. Ectopic bristles are observed with all UAS-sens reporters. Some UAS-sens transgenes driven by dpp-GAL4 cause very severe tufting in the adult notum, wings, and legs, and loss of tissues in other portions of imaginal discs, e.g., wing margins and distal leg structures. It is concluded that ectopic expression of Sens is sufficient to initiate ectopic external sensory organ development (Nolo, 2000).

To determine the molecular cascade underlying the formation of the extra external sensory organs, wing discs of UAS-sens; dpp-GAL4 larvae were stained with anti-Scute antibodies. Ectopic Sens expression causes ectopic activation of Scute and Asense. Hence, Sens is able to activate proneural gene expression. This provides a molecular basis for the generation of additional external sensory organs, since ectopic proneural gene expression has previously been shown to be sufficient to induce ectopic PNS organ formation (Nolo, 2000).

If Sens induces proneural gene expression and proneural genes are required for Sens production, a super-additive or synergistic interaction between sens and proneural genes may occur. Therefore, the weakest UAS-sens transgene (C1) was expressed in combination with an UAS-scute and an UAS-atonal transgene under the control of dpp-Gal4. Overexpression of Scute or Atonal alone causes a relatively mild phenotype with relatively few extra bristles. Scute expression induces Sens expression, but the expression levels of Sens are lower than those induced by dpp-Gal4; UAS-sens. Ectopic expression of Sens with the dpp-Gal4 driver causes a stronger phenotype when compared to ectopic expression of Scute or Atonal. However, simultaneous overexpression of Sens and Scute or Atonal causes severe tufting, including in many areas where Scute, Atonal, or Sens, when expressed individually, does not normally cause ectopic bristles. These areas do correspond to areas where the dpp-Gal4 driver has previously been shown to be expressed. It is therefore concluded that sens and the proneural genes can interact in a synergistic fashion (Nolo, 2000).

The data so far imply that Sens is dependent on proneural gene expression and that proneural gene expression can be induced by Sens. This raises another issue: can overexpression of Sens in the absence of proneural genes produce external sensory organs? Since sens expression depends on the expression of many proneural genes and since Sens is able to induce ectopic expression of several proneural proteins, removal of one or several proneural genes may not be fully effective, i.e., all proneural genes should be removed to test this hypothesis. Flies that are mutant for achaete and scute (Df(1)sc10-1) are almost devoid of micro- and macro-chaetae. Overexpression of Sens (C8) in Df(1)sc10-1 flies causes few ectopic bristles when compared to a wild-type background, suggesting that Sens requires the presence of proneural genes (Nolo, 2000).

To determine whether the proneural genes require Sens, Scute was overexpressed in the dorsal portion of the wing disc by driving UAS-scute with apterous-GAL4 and mutant clones were induced that lack sens in these discs. Flies that are UAS-scute; apterous-GAL4 exhibit numerous thoracic extra bristles. Six flies were obtained that have large clones, and in each fly a complete loss of all micro- and macro-chaetae in the clone was observed. These data clearly suggest that Sens is required for bristle development, even when Scute is overexpressed (Nolo, 2000).

These data indicate that ectopic expression of Sens is a more potent inducer of supernumerary PNS organs than is ectopic expression of proneural genes using the same drivers. For example, ectopic expression of Sens at the anterior-posterior border (the dpp domain between future wing veins 3 and 4) causes numerous bristles and sensilla campaniforma along the length of the wing blade, except in the most proximal portion, the wing hinge region. In some areas, the width of the field contains as many as five adjacent PNS organs. Sens is able to induce Scute in the ventral and the dorsal area of the wing pouch as well as in the dorsal portion of the disc. However, the area of the wing hinge region is much less sensitive to overexpression of Sens. This suggests that Sens does not have the same inductive capacity in all cells and also further supports the idea that Sens requires proneural activity to induce PNS organs (Nolo, 2000).

Induction of Sens does not only induce Scute and Asense. Sens expression using the dpp-GAL4 driver alters Delta expression. The domain that normally gives rise to the third wing vein, is altered in Sens-overexpressing discs. Overexpression of Sens induces Delta expression ectopically in the dpp domain, broadening and intensifying the endogenous Delta domain. In addition, a consistent reduction of expression in the fourth wing vein domain is observed. This ectopic Delta expression is likely to be mediated by Scute/Asense overexpression (Nolo, 2000 and references therein).

To determine the relationship between Sens expression and the proteins of the Enhancer of Split complex, wild-type discs were stained for both proteins. There is little overlap between the two nuclear proteins. Cells that express Sens are intermingled with E(spl) expressing cells, but the majority of cells that express Sens do not express E(spl). Similar observations were also made with E(spl)m8-lacZ and with E(spl)m4-lacZ. These data indicate that Sens expression in cells fated to develop into SOPs is concomitant with the presence of E(spl) proteins, but that elevation of Sens expression and cell enlargement during SOP specification accompanies a rapid removal of the E(spl) protein. These data are also in agreement with the proposition that E(spl) is a negative regulator of proneural gene expression and that its downregulation permits SOP development (Nolo, 2000 and references therein).

Ectopic expression of Sens may not only activate the proneural genes and Delta but may recreate an ectopic proneural field. Expression of several E(spl) proteins depends on the presence of the proneural genes. Therefore Sens was overexpressed using the dpp-GAL4 driver in E(spl)m8-lacZ and E(spl)m4-lacZ imaginal discs. Wild-type discs contain proneural clusters that express cytoplasmic ßgalactosidase [E(spl)] in which few cells are Sens positive. Overexpression of Sens causes a strong induction of ßgalactosidase staining associated with E(spl)m4-lacZ and E(spl)m8-lacZ. This induction is not restricted to cells in which Sens is expressed but can be detected in adjacent cells as well. This indicates that Sens can induce in a cell-nonautonomous fashion E(spl) expression, probably by activating Delta expression. A more detailed cellular analysis shows that when Sens expression is elevated in a particular cell, ßgalactosidase levels are consistently low or absent. It is inferred that ectopic Sens leads to expression of the essential components required to establish a proneural domain in some areas of the wing discs. This ability is most likely mediated by its ability to activate the proneural genes. The wing hinge region is, however, refractory to induction of Scute, Delta, and E(spl) upon overexpression of Sens (Nolo, 2000).

Since ectopic Sens is able to induce E(spl) expression and since elevated Sens levels are associated with low levels or absence of E(spl) protein during SOP specification in normal and ectopic conditions, it was of interest to enquire how ectopic expression of both proteins in the same cells would affect PNS organ development. Since overexpression of E(spl) causes a loss of external sensory organs, the component that is most downstream in the pathway should be epistatic to the more upstream component. The dorsal portion of the thorax of a dpp-GAL4; UAS-sens fly has extra bristles. Scutellar bristles are lost in dpp-GAL4; UAS-E(spl)m8 flies. Coexpression of both Sens and E(spl)m8 proteins always leads to a very strong reduction in supernumerary bristles in most areas, occasionally loss of bristles. Hence, ectopic E(spl), counteracting neurogenesis, is epistatic to ectopic Sens, stimulating neurogenesis, in the pathway that specifies the SOP (Nolo, 2000).

Although loss of sens affects all cells of the embryonic PNS, it affects multiple dendritic neurons less than the external and chordotonal sensory organs. The loss of PNS cells in embryos seems to be at least in part mediated by cell death. In embryos, Sens is therefore required for terminal differentiation and/or viability of most or all PNS cells. Hence, the phenotype associated with loss of Sens is similar but less severe than that associated with loss of proneural genes. This may be due to another homolog of sens (CG11243: probability e-46) that is expressed in some cells of the PNS and CNS (Nolo, 2000).

As shown above, ectopic expression of Sens causes many ectodermal imaginal cells to take the SOP fate. This implies that Sens is sufficient to induce external sensory organ formation. The data also indicate that Sens is downstream of proneural gene activity, as their presence is required for its transcription. Sens protein is not required for proneural gene expression in proneural clusters, but its presence in the SOPs is necessary and sufficient to enhance and maintain proneural expression to specify proper neuronal fate of SOPs. This model is supported by numerous observations. (1) Sens expression is dependent on proneural gene expression; (2) ectopic expression of Sens induces proneural gene expression; (3) ectopic proneural gene expression induces Sens expression; (4) Sens and the proneural genes scute and atonal interact in a synergistic fashion; (5) loss of Sens leads to a failure to enhance and maintain proneural protein expression in the SOPs; (6) the ability of Sens to recruit ectodermal cells into the SOP pathway is severely impaired in achaete and scute mutants. These observations also provide a molecular basis for the observations that ectopic expression of different proneural genes, including Atonal, leads to the production of a variety of es organs. They may all activate Sens, which in turn activates different proneural genes when the cells are not in their normal cellular context. However, the data also suggest that Sens must play another role that is required for the development of adult sensory organs, since overexpression of Scute in the absence of Sens is unable to promote es organ formation. This may be due to the inability of the progeny of the SOPs to survive or the inability to repress E(spl) expression (Nolo, 2000).

The ability of ectopic Sens to induce external sensory organ formation is most likely due to its ability to cause expression of many key players that are normally expressed in the proneural cluster. Proneural genes activate the transcription of the E(spl) genes and sens. The Sens protein may then act via two pathways in the SOP. (1) It may directly activate proneural gene expression participating in the initiation and/or maintenance of an autoregulatory loop. This mode of action is supported by the observation that ectopic Sens can induce proneural gene expression in the absence of endogenous proneural proteins or E(spl) proteins. In addition, the proneural genes contain consensus binding sites for the Sens protein, suggesting that the interaction may be direct. (2) Sens may first enhance and subsequently inhibit transcription of E(spl) genes. Expression of the genes of the E(spl) complex is clearly reduced in the SOPs to permit their specification. It is proposed that Sens also plays a role in this process by inhibiting transcription of the E(spl) genes in the SOPs. This in turn may allow further upregulation of proneural gene expresssion, followed by ectopic expression of E(spl) in neighboring cells that do not express Sens, suggesting that they receive a signal from the proneural proteins expressing cells. This signal is most likely Delta, since Delta expression is clearly upregulated in the cells that express Sens. Cells that do not express or express very low levels of Sens then accumulate more E(spl) than those that do express higher levels of Sens. Hence, the reduction of Notch signaling in the SOP may be strongly enhanced by the presence of Sens to help specify SOPs. Indeed, ectopic coexpression of E(spl)m8 and Sens dramatically reduces the action of Sens and in some areas of the notum creates a phenotype that is typically associated with overexpression of E(spl)m8 alone, i.e., loss of bristles. This suggests that E(spl)m8 acts downstream of Sens. In ectodermal cells, Sens normally does not play a role because none of these cells acquire enough proneural gene expression to activate Sens at a level that is sufficient to activate proneural gene expression above a required threshold. The latter statement is supported by the expression pattern of Sens, which is restricted to those cells that express the highest levels of proneural proteins and by the observation that robust levels of ectopic proneural gene expression must be attained to induce ectopic Sens expression. In summary, it is proposed that the function of Sens is to integrate proneural gene expression into the Notch signaling pathway to promote proper SOP development in the Drosophila PNS (Nolo, 2000).

GENE STRUCTURE

Transcript length - 6314 bases

Bases in 5' UTR - 175

Exons - 4

Bases in 3' UTR - 507

PROTEIN STRUCTURE

Amino Acids - 551

Structural Domains

The Lyra/Sens protein (Nolo, 2000), deduced from the sequence, contains a nuclear localization signal (PIVRKFK). The only domain that shows homology to known proteins is at the carboxy-terminal part of the protein between amino acids 410 and 518. This domain contains 4 Zn fingers of the 2Cys + 2His type that are homologous to Zn fingers found in the C. elegans Pag-3 protein (87% identity; Jia, 1997) and the vertebrate Gfi-1 protein (87% identity; Gilks, 1993). The Pag-3 protein has been shown to contain five Zn fingers and to affect neuronal lineages of motor neurons, interneurons, and touch neurons in C. elegans (Jia, 1996 and 1997). The Gfi-1 protein contains six Zn fingers, binds DNA, and is a transcriptional repressor in T lymphocytes (Grimes, 1996a; Zweidler-McKay, 1996). Loss of pag-3 causes lineage defects in worms (Jia , 1996), whereas activation of Gfi-1 in T lymphocytes causes interleukin-2-independent proliferation (Gilks, 1993) and T cell leukemia in mice (Schmidt, 1998b).

Given the sequence similarities, a test was performed to see whether Sens is able to bind the Gfi-1 consensus sequence (TAAATCAC core sequence; Zweidler-McKay, 1996). Full-length Sens is able to bind this oligonucleotide in an electromobility shift assay (EMSA) in the presence of poly(dI-dC). Sens protein that lacks the 4 C2H2 domain fails to bind, and cold nucleotides efficiently compete for binding of the full-length protein. Sens is unable to bind to a sequence with a single base mutation in the core (AAATGA). Thus Sens is a DNA binding protein with a similar specificity as Gfi-1 (Nolo, 2000).

The Gfi-1 protein has been shown to be a protooncogene (Gilks, 1993; Schmidt, 1998b) that functions as a transcriptional repressor (Grimes, 1996a). It has six Zn fingers, of which only the third, fourth, and fifth C2H2 domains are required for DNA binding and activity (Grimes, 1996a; Zweidler-McKay, 1996). Interestingly, these three domains correspond to the first, second, and third C2H2 domains of Sens. These three domains share 91% identity to the corresponding domains of Gfi-1, suggesting that both proteins have the same DNA binding specificity. This was confirmed by EMSA. Hence, by analogy to Gfi-1, Sens may act as a transcriptional repressor. However, Sens does not contain a SNAG repressor domain found in vertebrate Gfi-1 (Grimes, 1996a), suggesting that it may act differently from Gfi-1, i.e., as a repressor, an activator, or both. Interestingly, a sequence that fully matches the Gfi-1/Sens consensus sequence (TAAATCAC) is present once ~2 kb upstream of the scute and asense transcription start sites and about 700 bp upstream of achaete as well as upstream from E(spl)m8 (Nolo, 2000).

date revised: 28 February 2001

Home page: The Interactive Fly © 1995, 1996 Thomas B. Brody, Ph.D.

The Interactive Fly resides on the
Society for Developmental Biology's Web server.