BIOLOGICAL OVERVIEW

In developmental biology it often happensthat a gene product can serve multiple functions, and be involved at different developmental stages as well. Because of the long history of research into Drosophila developmental biology, genes may receive multiple names, corresponding to their multiple functions. For example, the genes ventral veins lacking (aka: ventral veinless) and drifter were both cloned in 1995. They turn out to be two names for the same gene, a gene previously called Cf1, identified as a POU-domain gene, one that binds to a neuron-specific regulatory element C, of the dopa decarboxylase gene. (Johnson, 1990). Two other POU-homeodomain transcription factors are found in Drosophila, PDM-1 and PDM-2.

drifter is the name given to vvl based on vvl's tracheal phenotype. drifter mutants display severe tracheal defects and defects in ventral midline glia migration. The glial defects result in defective commissure formation resulting from defects in axon pathfinding. Drifter is expressed in midline glia and in tracheal cells (Anderson, 1995).

Defects in tracheal cell migration are similar to those seen in breathless and pointed mutations. breathless is a homolog of the vertebrate FGF receptor tyrosine kinase; pointed is an ETS domain transcription factor. breathless functions upstream of Ras and Raf.The tracheal and glial cell defects seem attributable to defective cell migration. It is therefore likely that genes affecting migration are regulated by Drifter. It would be reasonable to look at the ras/raf pathway as a candidate for ventral veins lacking regulation.

Slow border cells, a gene implicated in migration of follicle cells is unlikely to regulate breathless expression during embryogenesis because SLBO expression in the tracheal system does not begin until long after breathless expression (Rorth, 1992). Drifter, may enhance btl expression in tracheal cells. Drifter protein is expressed in tracheal cells near the time that btl expression initiates: the dfr mutant phenotype is similar to btl; and dfr expression is not altered in btl mutants (Anderson, 1995). Thus it is possible, even likely, that dfr regulates btl expression. Preliminary experments suggest that dfr is not expressed in the border cells. One interpretation then, is that DFR may regulate btl in the embryo in much the same way that SLBO does in the ovary (Murphy, 1995).

ventral veins lacking mutations evince a variable phenotype consisting of the absence of proximal stretches of specific wing veins. It appears that mutant clones in dorsal veins do not affect ventral veins, and vice versa. vvl is expressed in both dorsal and ventral regions of the presumptive wing blade and wing base coinciding with the sites of future veins. A similar expression profile has been noted for rhomboid, a gene central to the induction of wing veins. Since rhomboidis involved in enhancing responses to the EGF receptor, which triggers the ras/raf pathway, this pathway is again implicated in vvl regulation (de Celis, 1995).

What are the targets of vvl and its cognate drifter? Are they the same or do they have significant differences? To what extent do they overlap? Perhaps dopa decarboylase is a common element in both functions. Knowing the array of targets could result in some generalization as to why the same transcription factor has a role in such different developmental pathways.

bHLH-PAS proteins represent a class of transcription factors involved in diverse biological activities. Previous experiments have demonstratedthat the PAS domain confers target specificity. This suggests an association between the PAS domain and additional DNA-binding proteins, Such an association is essential for the induction of specific target genes. A candidate for interactionwith PAS domain protein Trachealess (Trh) is Drifter/Ventral veinless. A dual requirement for Trh and Drifter has been identified for the autoregulation ofTrh and Drifter expression. Furthermore, ectopic expression of both Trh and Dfr (but not each one alone) triggers trh autoregulation in several embryonic tissues. A direct interaction between Drifter and Trh proteins, mediated by the PAS domain of Trh and the POU domain of Drifter, has been demonstrated (Zelzer, 2000).

Transcription of the trh gene is autoregulated, thus maintaining its expression throughout tracheal development,after the initial cues that determine the position of thetracheal placodes have disappeared. However, severalexperimental results suggest that the Trh/ARNT heterodimer is not sufficient for autoregulation of the trh gene. (1) Examination of Trh-Sim chimeras demonstrates that target gene specificity is determined by the PAS domain, possiblythrough interactions with other proteins.(2) Ubiquitous Trh can induce ectopic trh expressionoccasionally, at stage 11, but only at the position of trachealpits in segments that do not normally form tracheal pits, suggesting that additionalprotein(s) expressed in this pattern need to cooperate withTrh. A candidate protein that may interact with Trh is thePOU-domain protein Drifter/Ventral veinless (Dfr). Thisprotein was previously shown to participate in trachealmorphogenesis. Initially, dfr is expressed in the ten tracheal placodes,as well as in the position of placodes in segments that normally do not produce tracheal pits. dfr mutations show a reducedexpression of tracheal-specific genes such as breathless(btl), and accordingly exhibit migration defects that arereminiscent of the btl phenotype (Zelzer, 2000 and references).

An important feature of both Trh and Dfr expression istheir capacity to be autoregulated. Once the exogenous cues that direct expressionof these genes in the tracheal placodes diminish, expressionis maintained by autoregulation. Since the trh and dfr genes themselves can be regarded as targets for Trh or Dfr, respectively, a test was performed to see whether autoregulation of each of the two genes requires both Trhand Dfr. Two phases of Trh expression have been defined; at stage 12, expression induced by exogenous cues is diminished and autoregulation ensues. Staining for the Trh protein in dfr mutant embryos has demonstrated that the initial phase of Trh expression in the placodes is normal. However, starting at stage 12 thelevels of Trh are reduced, and are almost undetectableby stage 15. Failure of the cells in the tracheal pits of dfr mutant embryos to express Trh is not due to the death of these cells. Previous examination of thetracheal pits of dfr mutant embryos has shown that the cells are viable and capable of secreting tracheal lumen material, regardless of their failure to migrate properly. It can be concluded that Dfr is required for the autoregulation, and hence the maintenance of trh expression (Zelzer, 2000).

In the case of Dfr, a distinct 514 bp fragment has beendefined as the dfr-autoregulatory element, which begins todrive Dfr expression at stage 11/12. This fragment also confers expression in the oenocytes. In trh mutant embryos, lacZ expression driven by this fragment in the oenocytes is retained, but completely abolished inthe trachea. Again, the absence of expression inthe tracheal placodes, which fail to invaginate in the trh mutant background, is not due to death of these cells. Staining of trh mutant embryos with anti-Dfr antibodies or with a probe detecting dfr RNA, has revealed the early, Trh-independent phase of expression up to stage 11. The uninvaginated placode cells in trh mutant embryos are thus intact, but fail to express the dfr autoregulation reporter. These experiments demonstrate that Trh and Dfr are required simultaneously for the autoregulation of Trh and Dfr themselves (Zelzer, 2000).

Rho (Rhomboid) functions as a regulator for processing the EGF receptor ligand Spitz, and is expressed aembryonic stage 9/10 in the midline glial cells, as well asin cells positioned at the center of the tracheal placodes. The parallelexpression of rho in the tissues where Sim and Trh arefunctional, suggests that it may be a transcriptional targetof these two bHLH-PAS proteins. In trh mutant embryos, expression of rho in the tracheal placodes is abolished. Similarly, in sim mutantembryos, expression of rho in the midline is eliminated. To determine ifrho expression is regulated by direct binding of Sim andTrh, a 762 bp fragment of the rho 50 regulatoryregion was dissected: this is sufficient for midline and tracheal expression. The sequence of this fragment contains four sites with theSim/Trh (ST) binding consensus. Similar sites havepreviously been shown by in vivo and in vitro analysis torepresent the binding sites for Sim/ARNT or Trh/ARNTheterodimers. The 762 bp rho regulatory region was further dissected,and the capacity of smaller fragments to induce midline or tracheal expression in embryos was followed.The following conclusions were reached: Sim/Trh bindingsites STc and STd are neither sufficient nornecessary for tracheal or midline expression.In contrast, Sim/Trh binding sites STa and STb are essentialfor midline and tracheal expression. Distinct cis elements appear to be required to promote midline vs. tracheal expression (Zelzer, 2000).

The paradigm that Trh or Dfr alone are not sufficient toinduce their target genes or autoregulation, broadens thescope of activities of the two proteins. Trh is required notonly for the induction of tracheal fates, but also for patterning the salivary ducts and posterior spiracles. It is possible that in these tissues, Trhassociates with other proteins and induces a different setof tissue-specific target genes. Similarly, Dfr is alsoexpressed in the midline cells. Dfr is not necessary for theinduction of Sim-target genes, as can be deduced from thenormal expression of rhomboid in the midline of dfr-mutantembryos. However, Dfr could be functioning in conjunctionwith other midline proteins such as the Sox-domain protein Dichaete (Zelzer, 2000).

GENE STRUCTURE

The use of alternative polyadenylation sites produces two VVL mRNA transcripts(Anderson, 1995).

Exons - one

Bases in 3' UTR - 1588

PROTEIN STRUCTURE

Structural Domains

The POU domain protein includes both a divergent homeodomain and an additional POU-specific domain that function together as a bipartite DNA-binding domain (Klemm 1994).

The 75 amino acid POU-specific (POUs) domain and a 60 amino acid carboxy-terminal homeo (POUh) domain are joined by a hypervariable linker segment that can vary from 15 to 56 amino acids in length in different POU domain proteins. Thus the POU domain is not a single structural domain; indeed, the POUs and POUh segments form separate structurally independent domains. The POUs and POUh domains are, however, always found together and have therefore coevolved. Both POUs and POUh domains contain helix-turn-helix motifs. The POUs-domain structure is very similar to that of lambda and 434 bacteriophage proteins, but there are significant differences in the length of the first alpha helix, and the "turn" connecting the two HTH alpha helices is also longer. Both POUs and POUh bind DNA, and the length of the linker regulates the efficacy of binding various DNA sequence motifs, especially because POUs and POUh DNA binding sites have different spacings in different promoter elements (Herr, 1995).

date revised: 20 Feb 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.