Binding of SU(VAR)3-9 Partially Depends on SETDB1 in the Chromosomes of Drosophila melanogaster

Abstract

H3K9 methylation is known to play a critical role in gene silencing. This modification is established and maintained by several enzymes, but relationships between them are not fully understood. In the present study, we decipher the interplay between two Drosophila H3K9-specific histone methyltransferases, SU(VAR)3-9 and SETDB1. We asked whether SETDB1 is required for targeting of SU(VAR)3-9. Using DamID-seq, we obtained SU(VAR)3-9 binding profiles for the chromosomes from larval salivary glands and germline cells from adult females, and compared profiles between the wild type and SETDB1-mutant backgrounds. Our analyses indicate that the vast majority of single copy genes in euchromatin are targeted by SU(VAR)3-9 only in the presence of SETDB1, whereas SU(VAR)3-9 binding at repeated sequences in heterochromatin is largely SETDB1-independent. Interestingly, piRNA clusters 42AB and 38C in salivary gland chromosomes bind SU(VAR)3-9 regardless of SETDB1, whereas binding to the same regions in the germline cells is SETDB1-dependent. In addition, we compared SU(VAR)3-9 profiles in female germline cells at different developmental stages (germarium cells in juvenile ovaries and mature nurse cells). It turned out that SU(VAR)3-9 binding is influenced both by the presence of SETDB1, as well as by the differentiation stage.

Keywords: Drosophila; SETDB1; SU(VAR)3-9; chromosome 4; female germline; heterochromatin; piRNA clusters; salivary glands.

Figure 1

(a) Venn-diagrams indicating the numbers of SU(VAR)3-9 targets in the chromosomes from salivary glands and germline of wild type (wt) and egg mutants (upper row); distribution of SU(VAR)3-9 targets showing distinct response to egg mutation across euchromatin, heterochromatin, and the chromosome 4 (bottom row). (b) Changes in H3K9me2 binding in mutants (egg or Su(var)3-9) in comparison with wild type across the thee groups of genes, as well as genome-wide. Gene counts in each group are shown in brackets. Box-plot diagram is based on the original data on H3K9me2 distribution in the salivary gland chromosomes [35]. Chromosome-wide H3K9me2 enrichment across 30 bp intervals was calculated as log₂(IP/input) and median values for each gene were defined. Box plots show the range of values obtained by subtracting the median gene enrichment values in wild type animals from those in mutants.

Figure 2

SU(VAR)3-9 profiles in the sls gene in salivary glands (SG) and germline of whole juvenile ovaries (OV) from wild type (wt) and egg mutants. Data for SG wt are taken from [28], other profiles are obtained in this study. Significance of binding of SU(VAR)3-9-Dam (values above X axis) or Dam (values below X axis) as a −log₁₀ of p-value is plotted on the Y axis. Black horizontal lines denote threshold p-value corresponding to the FDR <5%. Genomic coordinates on the X axis and intron/exon structure of the sls gene correspond to the BDGP Release 6.

Figure 3

SU(VAR)3-9 profiles for the pericentric heterochromatin of the X chromosome (het) in the germline cells of the wild type (OV wt) and egg mutant females (OV egg). Single copy heterochromatin-resident genes featuring SETDB1-dependent SU(VAR)3-9 binding are highlighted by the grey frame. The density of grey and black vertical lines on the LINE/LTR panel visualizes the density of various repeated DNA sequences. Axis labels are the same as in Figure 2.

Figure 4

The border between euchromatin (eu) and heterochromatin (het) in the chromosome arms 2R and 3L in the salivary glands (SG) and germline cells of the ovary (OV) from wild type (wt) and egg mutants. Data for SG wt are taken from [28], the rest of the profiles are obtained in this study. Axis labels are the same as in Figure 2. Arrows point to the shift between the positions of the border between thegenotypes.

Figure 5

SU(VAR)3-9 binding profiles across select genes depending on the developmental stage of the ovary and the genotype: germline cells of juvenile ovaries of wild type females (OV wt) and egg mutants (OV egg) or mature nurse cells of wild type (NC wt). Data for NC wt are taken from [28], the rest of the profiles are obtained in this study. Axis labels are the same as in Figure 2.

Figure 6

Venn-diagrams showing the numbers of genes that are associated with SU(VAR)3-9 in the ovary in a differentiation-associated manner (upper part); sub-grouping of these gene targets depending on the egg effects (bottom part). NC and OV—nurse cells and germline cells of juvenile ovaries, respectively(wild type in both cases). “Common wt egg”—gene targets shared between wild type and egg profiles. “Lost in egg”—genes that are associated with SU(VAR)3-9 only in the presence of SETDB1.

Figure 7

SU(VAR)3-9 distribution (colored profiles) across the four piRNA clusters in salivary glands (SG; a,b) and germline cells of the juvenile ovaries (OV; f,g) from wild type (wt) and egg mutants, as well as H3K9me2 distribution (black profiles) in the salivary glands from wild type, egg, and Su(var)3-9 mutants (c–e). Data for SU(VAR)3-9 SG wt are taken from [28]. H3K9me2 profiles have been retrieved from the raw data reported by [35], the rest of the profiles are obtained in the present study. Grey frames in figures (b,g) denote novel SU(VAR)3-9 peaks in the piRNA cluster 20A, observed in egg mutants. Y axis shows significance of SU(VAR)3-9 binding as a −log₁₀ scale of p-value (colored profiles) or log₂(IP/input) value for H3K9me2 (black profiles). Genomic coordinates on the X axis correspond to the BDGP Release 6.