Worm Breeder's Gazette 11(2): 24
These abstracts should not be cited in bibliographies. Material contained herein should be treated as personal communication and should be cited as such only with the consent of the author.
Starting with the mec-3-lacZ fusion described at the CSH C. elegans meeting, I have deleted mec-3 sequences to try to find sites involved in regulation of mec-3 expression. Most deletions either have no effect, abolish all Xgal staining, or reduce the overall frequency of staining cells without specifically eliminating staining in any particular subset of cells. (Recall that this lacZ fusion is expressed in the touch receptors ALM, AVM, PVM, and PLM, and the other neurons PVD and FLP.) However, a deletion between the two Xba sites (bp1880-2490 of Way & Chalfie, Cell 54, 5-16) within the mec-3 part of the mec-3-lacZ fusion causes -galactosidase expression in the sisters of some cells that normally express mec-3 . In several animals, I have seen the sisters of the PLMs staining with Xgal, to give animals that have four mec-3- expressing cells in the tail. The PVM sisters also sometimes stain. This additional staining is seen only in L1 animals, presumably because later synthesis of the mec-3-lacZ fusion depends on mec-3 itself, and mec-3 is not being synthesized in these cells. A similar deletion from bp 1970 to 2560 does not show this effect, suggesting that sequences between bp 1880 and 1970 are important for the normal repression of mec-3 in sisters of the cells that normally express this gene. The evidence that these cells are the sisters of the normal cells is as follows: the additional cells in the tail have an anterior and dorsal position relative to the PLMs, which is characteristic of the PLM sisters, the ALNs. These cells are in a cluster of neurons in the tail, and thus have not been unambiguously identified. The additional cells are present before the T divisions, so they are embryonically derived. SDQL, the PVM sister, can be identified as a staining cell because it only is seen in later L1 animals (after the Q divisions have taken place), and because it is a relatively isolated neuron. I have not seen staining of the ALM sister BDU or the PVD sister (which undergoes programmed cell death). The FLP sister AIZ stains occasionally in gravid adults even with the wild-type fusion, but with the Xba-deletion construction, AIZ expresses -galactosidase activity more often and at earlier stages of development. The expression of the mec-3-lacZ in sisters of the usual cells can be explained by two possible models. In a DNA-based model, the Xba deletion might remove a target for a repressor protein that would turn off mec-3 in sisters of mec-3-expressing cells. In a protein (or RNA)- based model, the fusion protein could initially be synthesized in the parent cell, then segregated into one of the two daughter cells. The Xba deletion would remove from the protein (or RNA) the signal sequences for sorting during asymmetric cell division. Intriguingly, bp 1880 to 1970 neatly bracket the first putative exon of Way and Chalfie (Cell 54, 5-16). (However, there is no evidence one way or the other that this putative exon is actually used. Also, the fact that -galactosidase activity is expressed at all from such a construction suggests that an internal start codon would have to be used.) Since mec-3 is important in specifying the fates of the touch receptors (and probably PVD and FLP), it may be that asymmetric expression of mec-3 is a cause (and not a passive consequence) of the asymmetry between the touch receptors and their sisters.