Dosage compensation is an essential process that equalizes X-linked gene expression between sexes that differ in their number of X chromosomes. In C. elegans, XX hermaphrodites implement dosage compensation by halving the level of transcripts from each of their two X chromosomes, to achieve the same level as from the single X of XO males. For this down-regulation of X expression to occur, a well-characterized complex of proteins must localize specifically to the hermaphrodite X chromosomes, but not to the autosomes. We are interested in understanding how this protein complex distinguishes X from the autosomes. In mammalian dosage compensation, the random inactivation of one female X is initiated at a small, cis-acting locus on the X chromosome, the X Inactivation Center (XIC). XIC nucleates the assembly of the inactivation machinery, which then spreads into adjacent chromatin and eventually the entire X chromosome. If XIC is translocated to an autosome, adjacent autosomal sequences are inactivated. To determine if a similar X-recognition mechanism works in C. elegans, and to identify cis-acting X-recognition factors, we took the following approaches. First, embryos carrying large duplications of either the right (mnDp10, about 40% of X) or left (yDp14, about 22% of X) end of X were stained with antibodies against DPY-27, a dosage compensation protein that is normally localized to the X chromosomes. Both mnDp10 and yDp14 are attached to the left end of chromosome I. Through 3D reconstruction and volume rendering of confocal images, we asked: (1) Is there an increase in the number DPY-27-staining bodies per nucleus in duplication-bearing strains? (2) If so, is there a new class of DPY-27 staining bodies in duplication-bearing strains, and how large are the bodies compared to a normal X? (3) Does total DPY-27 staining occupy a larger portion of the nuclear volume in duplication-bearing strains? A total of 3172 N2 nuclei (22 embryos), 3130 mnDp10 nuclei (23 embryos), and 1352 yDp14 nuclei (6 embryos) were observed over three independent experiments. Embryos containing yDp14, a duplication that might be too small for detection by this method, were indistinguishable from wild-type. However, in mnDp10 strains, we found an increase in the number of DPY-27-staining bodies per nucleus, a new class of DPY-27 staining bodies, and an increase in total DPY-27 staining. This data strongly suggests that the dosage compensation proteins can recognize and localize to mnDp10. The new DPY-27 staining bodies observed in the mnDp10 strain occupy a small volume compared to the X-chromosome or chromosome I, indicating that the binding of the dosage compensation machinery does not spread far onto the autosome, if at all. Our second approach to identify cis-acting X-recognition elements is based on the hypothesis that sex-determination gene
her-1 (V) is regulated directly by the sdc gene products (See abstract by Dawes et al.). If the SDC proteins act directly on
her-1, we reasoned that dosage compensation components may be recruited to the promoter by the SDC proteins, since SDCs also function in very young embryos to allow the assembly of the dosage compensation machinery on X. If so, the
her-1 promoter may contain cis-acting elements that share characteristics with X-recognition elements. By utilizing a lacI-GFP/lacO system for identifying extrachromosomal arrays in situ, we found that the dosage compensation proteins DPY-26, DPY-27, and MIX-1 can localize to three contiguous, non-overlapping 1 kb pieces of the
her-1 genomic region. These data indicate that the SDC proteins are able to recruit the dosage compensation proteins to a location away from X. These 1 kb segments are being narrowed further, and complementary experiments with GFP-tagged arrays that contain X-linked cosmids are underway.