It seems that loss or gain of cato function causes neuronal defects. Like cato, embryos mutant for pros show neuronal defects, although the basis of these defects is not known. It was asked whether cato expression depends on pros function. Strikingly, loss of pros function results in the ectopic appearance of cato transcription within the ganglion mother cells (GMCs) and neurons of the developing CNS. This correlates with the wild-type expression of Pros in GMCs, indicating that Pros is normally a transcriptional repressor of cato in the CNS. There are indications that cato derepression begins in neuroblasts, even though pros is not thought to function until GMC formation: although Pros protein is present in the neuroblasts, it becomes localized only in the nucleus in the GMCs. Interestingly, although pros is apparently recessive, a weak derepression of cato is also observed in heterozygote pros/+ embryos. In this case cato expression is observed most clearly as nuclear sites of nascent transcription (Golding, 2000b).
In situ hybridization to mRNA in whole-mount embryos reveals that cato is expressed dynamically during neurogenesis and is confined to the developing PNS. The manner of expression is very different from that of proneural genes: it is initiated in SOPs after their formation. Expression continues in the division products of the precursors (after proneural genes are normally switched off) and the pattern broadly resembles that of the PNS enhancer trap line A37. Expression is switched off as cells begin to undergo terminal differentiation. cato expression appears to be activated earliest in chordotonal SOPs. Thus the initial pattern of cato expression in stage 10/11 strongly marks the appearance of chordotonal SOPs, and in this respect the pattern distinctly resembles the later expression of ato as it becomes refined from proneural clusters to the SOPs. Double-labelling experiments using antibodies to Ato in conjunction with a Cato RNA probe confirm that the onset of cato expression correlates with the refinement of Ato expression to isolated SOPs. Later, after ato is switched off, cato remains expressed in the division products of the SOPs (Golding, 2000b).
In PNS cells other than the chordotonal SOPs, cato expression appears delayed relative to SOP formation. This is most clearly seen when the first external sense organ SOP (the A cell) and chordotonal SOP (the P cell) are formed in each abdominal segment during embryonic stage 10. cato is initially activated only in the P cell, appearing later in the A cell during stage 11. Although subsequent SOP formation is dynamic and individual SOPs are difficult to follow, this delay seems to hold true for other external sense organ cells too. Thus, colabelling experiments show that cato expression in external sense organ precursors appears later than either Cut (a specific marker of these SOPs) or Asense (a marker of all neural precursors). Interestingly, the difference in timing of expression between cato and Ase appears reversed in chordotonal SOP formation. Thus, unlike cato, Ase expression appears relatively late in most chordotonal SOPs with the exception of the P cell and shows little overlap with Ato (Golding, 2000b).
Expression of cato appears to continue in the progeny of many or all SOPs during late stage 11 and early stage 12 and at this point resembles the pattern of ase. At this stage cato expression is too complex to discern the identity of individual cells. Nevertheless, during stage 12/13, Cato mRNA appears to become restricted to one of the daughter cells of each sense organ before being finally switched off. Double-labelling experiments with Cut and Pros antibodies suggest that this cell is the neuron. By the end of stage 13, cato expression is largely extinguished, and there is little apparent overlap with markers of neuronal differentiation, such as Elav (Golding, 2000b).
The temporal difference in cato activation between different SOP subtypes suggests that there may be separable functions for cato. It also suggests that different regulatory processes are at play. In particular, the overlap in Ato and cato expression in chordotonal SOPs suggests that cato may be directly activated by Ato. However, no evidence has been found for this. Although most chordotonal expression of cato is indeed absent in ato mutant embryos, this could merely reflect the failure of chordotonal SOP formation that occurs in this mutant. Indeed, cato expression is still detectable in the P cell chordotonal precursor, whose formation is not completely dependent on ato function. Therefore, at least in the P cell, factors other than Ato are capable of activating catoís expression (Golding, 2000b).
Expression in the larval imaginal discs mirrors the embryonic expression by being PNS specific and confined to developing SOPs and their progeny. In the wing disc, for instance, the pattern of cato expression broadly resembles that of the SOP enhancer trap marker, A101. Like ase, it persists through at least one round of cell division in the case of the external sense organ precursors. As in the embryo, Cato RNA appears strongly in chordotonal SOPs soon after their formation in the wing, antenna, and leg discs, whereas it is strongly delayed in external sense organ cells. Significantly, cato is not expressed in developing photoreceptors in the eye disc. In the leg disc, cato expression is more transient in the precursors of the femoral chordotonal organ than elsewhere; its expression here is switched off abruptly at the time when Ato itself is abruptly downregulated (Golding, 2000b).
The expression pattern suggests that cato supplies specific neural information to the determined SOPs or their progeny to ensure their proper development. There are as yet no point mutations of cato, and so to obtain an indication of the gene's function, embryos deficient for cato were examined. Analysis of PNS cell markers, including couch potato (cpo), showed that cato-deficient embryos [Df(2R)Jp7/Df(2R)Jp8] retain a grossly normal number of sense organs. Thus, consistent with its nonproneural expression pattern, cato has no role in SOP selection. In addition, specific markers of sense organ cells (including Elav for the neurons and Pros for the glial/sheath cells) show no gross defects in cell identity within sense organs, suggesting that cato has no role in assigning cell fates during or after asymmetric cell division (Golding, 2000b).
Thus cato cannot be functioning as a proneural gene. Markers of neuronal morphology, however, reveal neuronal disruption. Mab22C10 in particular visualizes the sensory neuronal cell bodies and processes, and this antibody reveals widespread defects in neuronal morphology. Such alteration is subtle and most easily observed in the chordotonal neurons, which have a highly specialized dendritic apparatus closely associated with a characteristic structure (the scolopale) formed by the sheath cell. In cato-deficient embryos, the dendrites of chordotonal neurons are consistently malformed, appearing longer and often thicker than their wild-type equivalents. The dendrite tips occasionally appear enlarged around a nonstaining hole at the point where the dendrite enters the scolopale. Antibodies that mark the scolopale, such as Ab1188, show that this structure is still formed, but is disorganized. The arrangement of the neurones itself is also disorganized, with their characteristic dendritic orientations often awry. Although specific cato mutations will be required for confirmation, the deficiency phenotype suggests a role for cato in neural differentiation (Golding, 2000b).
The single exception to the observation that cato deficiency does not affect neuronal number was a duplication of v'ch1 chordotonal neurons. However, v'ch1 duplication appears not to be a cato-dependent effect, since it has not been replicated in dsRNA interference experiments, even though a preliminary analysis suggests that neuronal morphology is affected (Golding, 2000b).
Goulding, S. E., zur Lage, P. and Jarman, A. P. (2000a). amos, a proneural gene for Drosophila olfactory sense organs that is regulated by lozenge. Neuron 25: 69-78.
Goulding, S. E., White, N. M. and Jarman, A. P. (2000b). cato encodes a basic helix-loop-helix transcription factor implicated in the correct differentiation of Drosophila sense organs. Dev. Biol. 221: 120-131.
Kanekar, S., Perron, M., Dorsky, R., Harris, W. A., Jan, L. Y., Jan, Y. N. and Vetter, M. L. (1997). Xath5 participates in a network of bHLH genes in the developing Xenopus retina. Neuron 19: 981-994.
Ledent, V., Gaillard, F., Gautier, P., Ghysen, A. and Dambly-Chaudiere, C. (1998). Expression and function of tap in the gustatory and olfactory organs of Drosophila. Int. J. Dev. Biol. 42(2): 163-170.
date revised: 20 June 2000
Home page: The Interactive Fly © 1997 Thomas B. Brody, Ph.D.
The Interactive Fly resides on the
Society for Developmental Biology's Web server.