BIOLOGICAL OVERVIEW

In the absence of a functional phyl gene product, ommatidia are usually missing three photoreceptor cells: two outer cells with large rhabdomeres (the light-sensitive apparatus of the photoreceptor), and one inner cell with a small rhabdomere. These photoreceptors correspond to R1, R6, and R7, the last three photoreceptors to be recruited into the ommatidia of the developing Drosophila eye. These three cells are born during the second of two waves of cell division that generate the population of cells from which the eye is assembled.

The first wave occurs as a broad band ahead of the morphogenetic furrow and gives rise to the five cells of the precluster: R8, R2, R5, R3 and R4. All the remaining cells (R1, R6 and R7) are generated in the second mitotic wave, which occurs as a tight band several rows behind the furrow. The shortage of photoreceptors in phyl clones is not a consequence of a failure of the second mitotic wave divisions since nonphotoreceptor cells within phyl clones are a product of cell division, suggesting that mitosis occurs normally in the absence of phyl. Since there are extra cone cells in phyl ommatidia containing only five photoreceptors, and the presumptive R7 photoreceptor can adopt a cone cell fate in the absence of a neuronal differentiation pathway, it is likely that failed R1, R6 and R7 cells all adopt cone cell fates (Chang, 1995).

What pathway(s) and what gene(s) target phyl (either directly or indirectly) during the process of eye morphogenesis? Both Pointed and Jun related antigen are known to regulatephyllopod transcription. Prior to its targeting of phyllopod, Pointed is itself a target of the ras pathway and Jun related antigen. R7 photoreceptor fate in the Drosophila eye is induced by the activation of the Sevenless receptor tyrosine kinase and the RAS/MAP kinase signal transduction pathway. Jun interacts with Pointed to promote R7 formation. This interaction is cooperative when both proteins are targeted to the same promoter and is antagonized by Yan, a negative regulator of R7 development. It has been suggested that Jun and Pointed act on common target genes to promote neuronal differentiation in the Drosophila eye, and that phyllopod is likely to be their common target (Treier, 1995). Sevenless may not be the only receptor tyrosine kinase that indirectly targets phyllopod. Sevenless is inactive in the R1 and R6 cells that are known to express phyllopod. Another activator of the Ras pathway, the Epidermal growth factor receptor (Egf-r), could function to indirectly activate phyllopod transcription in R1 and R6 cells, since Egf-r is active in this cell pair (Freeman, 1996).

In turn, what does Phyllopod target in photoreceptor differentiation? Tramtrack (Ttk) has been identified as a direct target of Phyl. Ttk expression represses neuronal fate determination in the eye. Phyllopod acts to antagonize this repression by means of an interaction that also requires Seven in Absentia (Sina). Sina, Phyl and Ttk physically interact: Sina interacts physically with a ubiquitin conjugating enzyme, UdcD1, an enzyme that puts a molecular tag (ubiquitin) on enzymes targeted for proteolytic destruction (Treier, 1992). Both Sina and Phyl bind to the N-terminal domain of Ttk. Loss of Phyl or Sina causes accumulation of Ttk in photoreceptor cells, and Ttk does not accumulate in cone cells if both Phyl and Sina are present (Tang, 1997 and Li, 1997).

It thus appears that a major function of Ras pathway signaling in photoreceptor cells, operating through Pointed and Jun, is the transcriptional activation of phyllopod. Phyllopod and Sina in turn interact with Tramtrack to form a molecular complex that targets Ttk to destruction, assuring a neural cell fate for R1, R6 and R7. As Ttk acts downstream on Notch (Guo, 1996), it is possible that the regulation of Ttk protein stability by Phyl, which is also required for normal PNS developent (Chang, 1995), may represent a point of downstream integration of signals from the Ras/MapK and Notch signaling pathways.

Regulated proteolysis has already been implicated in eye morphogenesis. sina has been found to genetically interact with fat facets (Carthew, 1994), a gene that encodes a ubiquitin dependent protease (Huang, 1996). On the basis of their protein structures, neither Sina nor Phyl resemble known ubiquitination factors. They may represent a novel class of molecules in this pathway. Sina contains a RING domain that is shared among a diverse group of cell regulatory proteins. One of these, PML, has been found to physically interact in vitro and in mammalian cells with PIC1, a ubiquitin-like protein of unknown function (Boddy, 1996). The mechanism utilized by Phyl and Sina to destabilize regulatory proteins such as Ttk may be conserved for the control of cell differentiation and other processes in vertebrates. Ttk possesses a BTB domain, implicated in interaction with Sina (Li, 1997). BTB domain-containing proteins are found in many vertebrates.

phyllopod is a target gene of proneural proteins in Drosophila external sensory organ development

Proneural basic helix-loop-helix (bHLH) proteins initiate neurogenesis in both vertebrates and invertebrates. The Drosophila Achaete (Ac) and Scute (Sc) proteins are among the first identified members of the large bHLH proneural protein family. phyllopod (phyl), encoding an ubiquitin ligase adaptor, is required for ac- and sc-dependent external sensory (ES) organ development. Expression of phyl is directly activated by Ac and Sc. Forced expression of phyl rescues ES organ formation in ac and sc double mutants. phyl and senseless, encoding a Zn-finger transcriptional factor, depend on each other in ES organ development. These results provide the first example that bHLH proneural proteins promote neurogenesis through regulation of protein degradation (Pi, 2004).

In phyl²-null mutant clones, adult ES organs are absent, and this defect is caused by a failure in SOP specification. In phyl²/phyl⁴ hypomorphic mutants, most ES organs are also absent, and expression of two SOP markers, ase-lacZ and the A101 enhancer trap line, are strongly compromised. However, Sens is expressed in single, selected SOPs at 12-14 h after puparium formation (APF), suggesting a defect in SOP differentiation, but not in SOP selection in phyl hypomorphic mutants (Pi, 2004).

Ac expression, which is initially in proneural clusters and restricted in SOPs at 12-14 APF in wild type, was examined. However, in phyl²/phyl⁴ mutants, Ac expression is not only detected in SOPs, but also weakly in SOP-neighboring cells. Ac expression in SOP-neighboring cells is later diminished at 16-18 APF. This result suggests that lateral inhibition is partially affected. To test this, E(spl)m8-lacZ was used as a reporter to monitor Notch signaling. Although E(spl)m8-lacZ is strongly expressed in a proneural pattern in wild type, the expression is abolished in phyl²/phyl⁴ mutants, suggesting that activation of the Notch pathway in the SOP-neighboring cells is compromised in phyl mutants (Pi, 2004).

In wild-type ES organ development, Sens staining appears in two SOP-daughter cells at 16-18 h APF and in four daughter cells at 24-28 h APF. In phyl²/phyl⁴ mutants, Sens is still maintained mostly in single cells even at 24-28 h APF. In wild-type animals, SOPs express elevated levels of the cell-cycle regulator Cyclin E (CycE). In phyl²/phyl⁴ mutants, SOPs fail to express a higher level of CycE, suggesting a failure in cell cycle progression. The SOPs and SOP daughter cells of ES organs express cut, a selector gene in the determination of ES organ identity. In phyl²/phyl⁴ mutants when SOP differentiation has been arrested, Cut expression is absent. Taken together, these data indicate that Phyl is required for gene expression in SOP differentiation and lateral inhibition, for SOP cell cycle progression and for ES organ identity (Pi, 2004).

Ac and Sc are bHLH transcriptional activators, and Ac/Da and Sc/Da heterodimers bind specifically to the E boxes CAG(G/C)TG with high affinity and CACGTG with low affinity. Within the 4.1-kb phyl promoter region, there are four such E boxes (E1-E3, CAGCTG; E4, CACGTG). Three phyl reporter genes were constructed by fusing 4.1-, 3.4-, and 2.2-kb promoter regions of phyl to GFP, and all three reporters show similar expression patterns with difference in the GFP signal intensities (the 4.1-kb promoter being the strongest and 2.2-kb being the weakest). For example, the 3.4-kb region is sufficient to drive GFP expression in embryonic SOPs, SOPs of the late third-instar larval wing and leg discs, and SOPs in early pupal nota. These phyl-GFP reporter genes are also expressed in the proneural clusters at earlier stages in both wing discs and pupal nota (Pi, 2004).

To test whether these promoter regions are sufficient for phyl in vivo function, phyl^4.1-ORF and phyl^3.4-ORF rescue constructs were made by fusing the 4.1- and 3.4-kb promoter regions, respectively, to the phyl ORF. The phyl¹/phyl² mutants die at late embryonic or first-instar larval stages. However, both phyl^4.1-ORF and phyl^3.4-ORF are sufficient to rescue the viability of phyl¹/phyl² animals to the adult stage, with well developed ES organs on the notum. The inabilities to fully rescue the viability and ES organ number of phyl¹/phyl² are caused by insufficient expression levels of the transgenes, as suggested by the fact that two copies of phyl^3.4-ORF further improve the viability of the phyl¹/phyl² mutants to 77% and increase the bristle number to 110 ± 7. Hypomorphic phyl⁴/phyl²²⁴⁵ mutants, which display a greatly reduced number of ES organs on the notum, are completely rescued by two copies of phyl^3.4-ORF. Therefore, all of these results show that both 4.1- and 3.4-kb regions of the phyl promoter contain sufficient temporal and spatial information in regulating phyl expression (Pi, 2004).

Whether activity of the 3.4-kb promoter region is regulated by ac and sc was tested. sc^10-1 is a compound mutation in which both ac and sc are inactivated. Expressions of phyl^3.4-GFP in sc^10-1 wing discs and pupal nota are abolished. In contrast, when sc is misexpressed by dpp-GAL4 at the anterior/posterior boundary of the wing disc, phyl^3.4-GFP is strongly activated in this region. Similar results are also observed for phyl^4.1-GFP. Therefore, these results clearly show that proneural genes ac and sc are necessary and sufficient to activate phyl promoter activity (Pi, 2004).

To test whether Ac and Sc directly regulate phyl expression, all four E boxes in the 3.4-kb promoter region were mutated to make the phyl^3.4DeltaE-GFP. The expression of phyl^3.4DeltaE-GFP in the SOPs of ES organs in late third-instar wing and leg discs and in pupal nota is strongly reduced when compared to the expression of phyl^3.4-GFP. When the GFP intensity was quantified in the anterior wing margin SOPs, E box mutations in the 3.4-kb promoter region contribute to a 50% reduction. In contrast, the expression level of phyl^3.4Delta E-GFP in the SOPs of chordotonal (CH) organs promoted by the proneural gene ato is comparable to that of phyl^3.4-GFP. These results indicate that the phyl promoter is activated by Ac and Sc through these four E boxes. To test the in vivo significance of the four E boxes, the rescue abilities were compared between phyl^3.4-ORF and phyl^3.4DeltaE-ORF. Although phyl^3.4-ORF can rescue phyl¹/phyl² to the adult stage, phyl^3.4DeltaE-ORF-rescued animals only survive to the third-instar larval stage. The abilities of phyl^3.4DeltaE-ORF to rescue the viability and the notal ES organ of phyl⁴/phyl²²⁴⁵ mutants are strongly reduced to 36 ± 11% and 67 ± 12, respectively. Many of the rescued ES organs show abnormal configuration such as double hair/double socket, which is a phenotype frequently observed in hypomorphic phyl mutants. Therefore, these results suggest that these four E boxes are required for full phyl promoter activity in SOPs (Pi, 2004).

In sc^10-1 flies, phyl expression is diminished and ES organ development is disrupted. It was asked whether forced expression of phyl can functionally substitute for the absence of ac and sc activities. Misexpression of phyl by Eq-GAL4 in sc^10-1 flies efficiently rescues ES organ formation, to a level similar to that rescued by misexpression of the proneural gene sc. The rescued ES organs by phyl are arranged in a pattern similar to that of the wild-type flies; the ES organs are aligned in rows and well separated. SOP-specific expressions of neu-LacZ (A101), ase-LacZ, Sens, and Cut, as well as expression of E(spl)m8-LacZ, are restored. As a comparison, sens, whose expression also depends on ac and sc was misexpressed by Eq-GAL4; sens poorly rescues sc^10-1 in ES organ formation, although sens is more effective than phyl and sc in inducing ES organs in wild-type background. Therefore, these results suggest that phyl is able to execute the developmental program of ES organs in the absence of proneural genes ac and sc.

Ac and Sc activate the bHLH gene ase in SOPs to promote SOP differentiation. Misexpression of ase or another bHLH gene lethal of scute (l'sc) is capable of generating ES organs independent of ac and sc. Whether phyl can rescue ES organ formation in the absence of all four bHLH genes, ac, sc, ase, and l'sc, in sc^B57 mutant clones, was tested. Although, in a control experiment, misexpression of sc can rescue the ES organ formation in sc^B57 mutant clones, misexpression of phyl fails to rescue. From this result, it is inferred that phyl requires ase (and/or l'sc) in inducing ES organ formation (Pi, 2004).

The promoter analysis suggests that phyl expression in SOPs might be activated by factors other than Ac and Sc. Within the 4.1-kb promoter region, eight putative Sens-binding sites (AAATCA, S box) were identified, with three sites distributed within the 3.4-kb proximal region and five sites in a cluster located in a very distal region. Whether Sens plays a role in phyl activation in SOPs was tested, using phyl^4.1-GFP as a reporter. At 10-12 h APF, phyl^4.1-GFP is expressed in dorsoventral stripes along the notum in a pattern analogous to early Ac and Sc expression patterns. At 15 h APF, phyl^4.1-GFP expression is restricted in SOPs. In sens^E2-null clones, phyl^4.1-GFP is expressed in dorsoventral stripes, and this expression is quickly restricted to single SOPs at 16 h APF, identical to that in wild-type tissue. At 20 h APF, when wild-type SOPs have divided to two daughter cells, phyl^4.1-GFP expression in sens^E2 clones is still maintained in single SOPs, and mostly in two cells at 23 h APF when wild-type cells are in GFP-positive clusters containing three or four cells. Therefore, these results suggest that, in the absence of sens activity, SOP development is delayed, but phyl^4.1-GFP expression is minimally affected (Pi, 2004).

To determine the contribution of Sens binding sites to phyl expression, the 3.4-kb phyl promoter region (whose expression pattern is analogous to the 4.1-kb promoter in both wild-type and sens mutant background) was tested. The phyl^3.4DeltaS-GFP reporter with all three S boxes mutated expresses little difference in the GFP pattern and intensity when compared to phyl^3.4-GFP. However, the reporter with mutations in all four E boxes and three S boxes (phyl^3.4DeltaES-GFP) enhances GFP intensity by 20% when compared to phyl^3.4DeltaE-GFP with mutations only in four E boxes. This 20% increase in GFP intensity reflects an increase in the phyl activity in vivo because phyl^3.4DeltaES-ORF shows stronger abilities than phyl^3.4E-ORF in rescuing both the viability and the ES organ number of phyl⁴/phyl²²⁴⁵ flies. Therefore, these data suggest that these S boxes play a negative role in regulation of phyl activity (Pi, 2004).

To test whether phyl regulates sens expression, Sens protein expression was examined in phyl mutants. In phyl²-null clones, Sens expression was almost diminished in all stages examined, including the single-SOP stage, the two-cell stage and the four-cell stage, suggesting that phyl is required for Sens expression in ES organ development (Pi, 2004).

To analyze the functional relationship between phyl and sens further, rescue experiments were performed. Misexpression of sens by Eq-GAL4 fails to induce ES organ formation in phyl² mutant clones. Similarly, ES organ formation induced by phyl misexpression is blocked in sens^E2 mutant clones. This result suggests that although Sens expression depends on phyl activity, Sens and Phyl function in parallel to promote ES organ development (Pi, 2004).

It is concluded that phyl is a non-bHLH gene that can functionally substitute for proneural bHLH genes to execute neural developmental program. This ability of phyl is also manifested from the analysis of phyl loss-of-function phenotypes: sens and ase, required for SOP differentiation, are inactivated, and in addition, neuralized (A101 insertion locus), implicated in the activation and E(spl)-m8 in the transduction of the Notch pathway, is not expressed. Furthermore, SOP cell division, a prerequisite step to generate distinct daughter cells for constructing a complete ES organ, is blocked in phyl mutants. This defect likely reflects a role for phyl in controlling cell cycle progression, because CycE expression in SOPs maintains at a basal level. Therefore, although SOPs have been selected from proneural clusters in phyl hypomorphs, they are associated with several defects as described (Pi, 2004).

Studies of proneural genes have shown that ac and sc promote ES organ identity, whereas ato promotes CH organ identity. cut is the selector gene to specify the ES organ identity; in its absence ES organs are transformed into CH organs and misexpression of cut transforms CH organs into ES organs. The absence of Cut expression in phyl mutants suggests that specification of ES organ identity may be through a regulation of cut expression by Phyl. Although phyl is expressed in SOPs for both ES and CH organs, it was found that, in phyl²/phyl⁴ and phyl¹/phyl⁴ mutants, A101 expression in leg CH organ precursors remained normal. Also, misexpression of phyl fails to rescue ato mutants in CH organ formation. These results suggest that phyl mediates functions of ac and sc only in ES organ development (Pi, 2004).

One well characterized function of Phyl is to bring the Ttk protein to the ubiquitin-protein ligase Sina for degradation. During SOP development, phyl is expressed in SOPs, and Ttk is expressed ubiquitously except in the SOPs and the proneural clusters. Genetic studies among phyl, sina, and ttk suggest that phyl and sina promote ES organ development by antagonizing ttk activity. Ttk contains a BTB/POZ domain and functions as a transcriptional repressor. Therefore, degradation of Ttk can lead to the derepression of SOP-specific genes. These studies suggest that degradation of a general transcriptional repressor plays a crucial role in regulating gene expression in different aspects of neural precursor differentiation (Pi, 2004).

GENE STRUCTURE

Exons - 2

PROTEIN STRUCTURE

Amino Acids - 400

Structural Domains

Phyllopod is hydrophobic and shows no significant homolgy to any protein sequences in available databases. The protein can be divided into three domains: a neutral N-terminal domain of 110 amino acids, a highly basic central domain of 152 residues, and an acidic domain (rich in glutamate) of 103 amino acids extending until just before the C-terminus. Within the N-terminal domain are two CxxC motifs that might contribute to a divalent metal-binding domain. The predicted sequence contains a potential bipartite nuclear localization signal at residues 193-194 and 208-210 (Dickson, 1995 and Chang, 1995))

date revised:  26 August 97

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.