Regulation of DCC localization by HTZ-1/H2A.Z and DPY-30 does not correlate with H3K4 methylation levels

Abstract

Dosage compensation is a specialized form of gene regulation that balances sex-chromosome linked gene expression between the sexes. In C. elegans, dosage compensation is achieved by the activity of the dosage compensation complex (DCC). The DCC binds along both X chromosomes in hermaphrodites to down-regulate gene expression by half, limiting X-linked gene products to levels produced in XO males. Sequence motifs enriched on the X chromosome play an important role in targeting the DCC to the X. However, these motifs are not strictly X-specific and therefore other factors, such as the chromatin environment of the X chromosome, are likely to aid in DCC targeting. Previously, we found that loss of HTZ-1 results in partial disruption of dosage compensation localization to the X chromosomes. We wanted to know whether other chromatin components coordinated with HTZ-1 to regulate DCC localization. One candidate is DPY-30, a protein known to play a role in DCC localization. DPY-30 homologs in yeast, flies, and mammals are highly conserved members of histone H3 lysine 4 (H3K4) methyltransferase Set1/MLL complexes. Therefore, we investigated the hypothesis that the dosage compensation function of DPY-30 involves H3K4 methylation. We found that in dpy-30 animals the DCC fails to stably bind chromatin. Interestingly, of all the C. elegans homologs of Set1/MLL complex subunits, only DPY-30 is required for stable DCC binding to chromatin. Additionally, loss of H3K4 methylation does not enhance DCC mislocalization in htz-1 animals. We conclude that DPY-30 and HTZ-1 have unique functions in DCC localization, both of which are largely independent of H3K4 methylation.

Figure 1. H3K4 is differentially methylated on dosage compensated X chromosomes.

A. IF of the DCC (red) and H3K4me1, H3K4me2, H3K4me3 or HTZ-1 in wildtype, or dosage compensation mutants sdc-1(e415) and dpy-21(e428) shows slightly higher levels of H3K4me1 and lower levels of both H3K4me3 and HTZ-1 on dosage compensated X chromosomes in adult hermaphrodite intestinal nuclei. DCC is marked by a DPY-27-specific antibody (with H3K4me2, H3K4me3 and HTZ-1 co-stain) or a CAPG-1-specific antibody (with H3K4me1 co-stain). DAPI (grayscale) stains DNA. Scale bar equals 5 µm. B. Quantification of H3K4 methylation and HTZ-1 occupancy on the dosage compensated X chromosomes by IF in wildtype and dosage compensation mutants. The average intensity of H3K4 methylation or HTZ-1 signal was recorded for the nucleus and for the DCC-marked X chromosome regions. The ratio of the average intensity on X versus the average intensity for the entire nucleus was calculated and the average values are shown. Error bars indicate the standard deviation. Star indicates significant difference from wildtype by t-test analysis (p = 6×10⁻³).

Figure 2. DCC binding to or stability on chromatin is compromised in dpy-30 animals.

A. DCC localization by DPY-27 IF (green) combined with X-paint FISH (red) with and without detergent extraction of nucleoplasmic contents in wildtype, dpy-30(y228), and htz-1(tm2469) hermaphrodite intestinal nuclei shows that the diffuse DCC localization observed in dpy-30 animals can be extracted with detergent treatment while the mislocalized DCC in htz-1 cannot. B. Condensin I^DC components mislocalize in dpy-30(y228) hermaphrodite animals. DPY-27, DPY-28, CAPG-1, and MIX-1 (green) all show the same diffuse, nuclear pattern in hermaphrodite adult intestinal nuclei. DPY-27 localization in wildtype (WT) is shown as a control. The nuclear envelope is marked with a nuclear pore complex antibody (red) as a staining control. In all panels DAPI is shown in grayscale and scale bar equals 5 µm.

Figure 3. Depletion of Set1/MLL subunits by RNAi leads to reduction in H3K4me3 levels.

H3K4me3 (green) and DPY-27 (red) IF in control RNAi and Set1/MLL depletion animals shows a reduced H3K4me3 signal upon Set1/MLL RNAi in adult intestinal nuclei. Note that only DPY-30 depletion leads to a diffuse DCC localization phenotype. In all panels DAPI is shown in grayscale and scale bar equals 5 µm.

Figure 4. DCC localization to the X chromosomes is disrupted in dpy-30 depleted and mutant animals but not in other Set1/MLL depletion and mutant animals.

DPY-27 (green) and X-Paint FISH (red) in control and Set1/MLL depletion (A) and mutant (B) animals. In all panels DAPI is shown in grayscale and scale bar equals 5 µm.

Figure 5. Severe loss of H3K4 methylation by Set1/MLL depletion in set-2(ok952) hermaphrodites.

Hermaphrodite intestinal nuclei co-stained for H3K4me3 (green) and DPY-27 (red) in Set1/MLL complex depletion in set-2 mutant animals demonstrates that increased loss of H3K4 methylation does not correlate with DCC mislocalization. In all panels DAPI is shown in grayscale and scale bar equals 5 µm.

Figure 6. Reduced H3K4 methylation by Set1/MLL depletion does not enhance DCC mislocalization in htz-1(tm2469).

DPY-27 (green) and X-Paint FISH (red) colocalization analysis upon Set1/MLL depletion in htz-1(tm2469)m+z- mutant animals indicates that reduced H3K4 methylation does not further disrupt DCC localization. In all panels DAPI is shown in grayscale and scale bar equals 5 µm. B. Quantification of colocalization by Pearson's correlation coefficient R indicating the degree of overlap between the DCC and X-Paint masks. R measurements were binned every 0.1 from 0 to 1 (where 0 indicates no detectable overlap and 1 indicates perfect colocalization) and percent of total sample per bin is shown. See Figure S7 for statistical analysis.

Figure 7. H3K4 methylation compexes do not function in dosage compensation.

Rescue of sensitized male animals in which the xol-1 mutation has ectopically turned on dosage compensation is used as a measure of genetic function of dosage compensation. Depletion of DCC subunit CAPG-1 by RNAi is able to disrupt ectopic dosage compensation, rescuing about 40% of expected male progeny. The only Set1/MLL complex subunit depletion with significant male rescue is dpy-30. Star indicates p<0.05 by T-test (p = 6.83×10⁻⁵ for dpy-30 and p = 1.03×10⁻⁹ for capg-1).