Sequence-specific binding of Swi6 to the dg-dh-specific siRNAs in vitro.

A Schematic representation of the dg and dh regions in the centromere, mating type (cenH) and the sub-telomeric genes. tlh1 and tlh2. Right panel, a schematic showing siRNAs being generated from the dg-dh sequences.

B List of si RNA sequences, reported earlier (Reinhardt & Bartel, 2002), that were synthesized in vitro.

C EMSA assay showing sequence specific binding of Swi6 to the ss D-For RNA and its competition by excess col RNA sequences B, D, E and K. Arrows indicate the complexes I, II and II between Swi6 and ‘D-For’ RNA.

D The mutant Swi63K→3A shows weaker binding to the D-For RNA. EMSA was performed as in (C).

E Estimation of equilibrium binding constant (Kd) for the binding of Swi6 to D-For RNA. The data of protein concentration of Swi6 along X axis and the fraction of siRNA bound was plotted as described (Heffler et al, 2012).

F Schematic depiction of the domain structure of Swi6 showing the chromo-, chromo-hinge and chromo-shadow domains. Alignment of the sequences of the hinge domain between HP1α, HP1β, HP1γ and Swi6, showing conservation of the lysine triplet located at residues 242-244.

G EMSA showing that the RNA binding could be ascribed primarily to the CD and CD-hinge but not the CSD.

Localization of Swi6-bound siRNAs to dg-dh regions is abrogated in swi63K→3A mutant.

A Overall distribution of normalized small-RNA accumulation at centromeres, telomeres and mating-type locus in WT and swi63K→3A mutant.

B Chromosomal track of RIP-seq fold change (log2 swi63K→3A /WT ratio) at and around centromere of chr-I at 5kb (top) and 200bp (bottom) resolutions, left telomere of chr-I (200bp resolution) and the mating locus (200bp resolution). Locus annotations are given at the bottom of the plot. Plots for other centromeres and telomeres are given in the Supplementary Fig10.

C Distributions of normalized RIP-seq signals in swi6+ and swi63K→3A strains for regions containing the sequence motif(s) known to interact with Swi6. The p-values were calculated using two-tailed Mann-Whitney U tests.

Abrogation of silencing in swi63K→3A mutant accompanies Swi6 delocalization at the outer repeat otr1R and imr1L regions of cenI and his3-telo.

A Diagrammatic representation of the genotype of the strain. Centromere of each fission yeast chromosome comprises of a central core (cnt), immediately flanked on both sides by inverted repeats (imr); imr on either side are flanked by outer repeat regions (otr). Reporter genes ura4, ade6 and his3 have been inserted into the imr, otr of centromere and telomeric repeats, respectively of chrl. The strain has a swi6Δ mutation and carries internal deletions in native ura4 locus, denoted as ura4DS/E, and in ade6, denoted as ade6DN/N, which act as euchromatic controls in ChlP assay.

B-D Derepression of the otrIR::ade6 locus in swi63K→3A mutant, showing pink/white colonies on adenine limiting plates (B).

C ChIP assay showing reduction of Me2-K9-H3 and Swi6 at the ade6 locus in swi63K→3A mutant. Ratio of the heterochromatic ade6 and euchromatic ade6DN/N locus is shown

D Quantitation of data shown in (C).

E Spotting assay showing inability of swi63K→3A to restore silencing at the imrl::ura4 locus in swi6Δ mutant, as indicated by growth on plates lacking uracil and lack of growth on FOA plates.

F ChIP assay showing delocalization of Swi6 from the imrl::ura4 locus in the swi63K→3A mutant, where ura4 represents the heterochromatin and ura4DS/E the euchromatin.

G Quantitation of the data showing enrichment of Swi6 at ura4 versus ura4DS/E in (F).H-K Derepression of the subtelomerichis3 locus in swi63K→3A mutant.

H gives a schematic representation of the his3 gene inserted distal to the telomeric repeats, denoted as his3-telo.

I Spotting assay showing the enhanced growth of the strain having his3-telo reporter in presence of the swi6 mutation on plates lacking histidine.

J ChIP assay reduced localization of mutant Swi63K→3A at the his3-telo locus as compared to wt. act1 was used as a control.

K Quantitation of data shown in (J).

Role of Swi6 binding to the dg-dh sequences as RNA-DNA hybrids in silencing.

A Swi6 shows strong affinity towards the RNA-DNA hybrid. EMSA assay was performed using the hybrid between the D-For RNA and its complementary DNA sequence and purified GST-Swi6 or GST-Swi63K→3A. GST was used as control.

B Data in (A) was plotted to determine the equilibrium binding constant, Kd.

C Delocalization of GFP-Swi6 upon overexpression of rnh1 but not rnh201. Fluorescence micrographs showing the effect of overexpressed nmt1-rnh1 and nmt1-rnh201 on the subcellular localization of Swi6 with respect to the three heterochromatic loci, mat, cen and tel as distinct green foci. The strain had the native copy of swi6 gene chromosomally tagged with GFP. Nuclei were stained with DAPI. Strains expressing GFP-tagged Swi6 was transformed with vector alone, rnh1 or rnh201 under the control of nmt1 promoter. Transformed cells were grown in the presence (+) or absence (-) of the repressor, thiamine, visualized under confocal microscope and photographed.

D Microscopic data shown in (C) was quantified as distinct green foci pertaining to GFP-Swi6 subcellular localizing to the heterochromatic loci. Quantitation of the localization of GFP-Swi6 as three, two or one spots.

E-H Overexpression of rnh1 causes loss of silencing at the otr1R::ade6 locus.

E Schematic representation of the centromere I having an ade6 reporter gene inserted into the otr1R.

F The effect of expression of empty vector alone (nmt1), nmt1-GFP-swi6 or nmt1-GFP- swi63K→3A on the expression of the ade6 reporter. Colors of the colonies indicated the state of the ade6reporter. Red colonies indicate suppression of ade6repression while pint/white colonies indicated repression. The data are represented in terms of percent white colonies.

G Plot showing the per cent colonies giving pink/white appearance in different transformants, as shown in (F).

H Delocalization of Swi6 from the otr1R::ade6 locus in the transformants shown in (F). Derepression of the ade6 reporter at otr1R repeat as a result of overexpression of rnh1 and rnh201, as observed in (G) was biochemically verified through Swi6 enrichment at the locus through ChIP assay.

Swi6 binds to and protects the RNA-DNA hybrid at the dh region in vitro.

A ChIP assay was performed to quantitate the dh regions enriched with anti-RNA-DNA hybrid antibody in the indicated strains.

B The RNA-DNA hybrid shows preferential binding to the Me2-K9-H3. EMSA was performed to detect binding of Swi6 to the RNA-DNA hybrid (D-For RNA/DNA duplex) in presence of unmethylated (M0), Lys9 dimethylated (M2) and Lys9 trimethylated H3 (M3) peptides. No binding of the hybrid was observed with any of the peptides. To be able to observe the supershift (indicated by arrowheads) in presence of the peptides, electrophoresis had to be performed longer to run out the labeled hybrid, which also led to a diffuse signal.

C si RNA generation is required for Swi6 enrichment at dh region. ChIP assay was performed to quantitate the dh region associated with Swi6 in wt versus cid14Δ mutant.

Speculative model about the role of nucleic acid binding function of Swi6/HP1 in initiating RNAi dependent heterochromatin assembly in Fission Yeast.

Based on data described here, Swi6/HP1 binds with moderate affinity to the single stranded siRNA mapping to the dg-dh repeats. This helps it to be recruited to chromatin at the stage in the RNAi pathway where the RITS complex is already tethered through the nc transcript. This stage is sensitive to RNase H. The recruitment of Swi6-ss-siRNA complex is facilitated by the occurrence of the complementary DNA strand in single stranded form. The high affinity binding of Swi6 to the RNA-DNA hybrid helps to retain Swi6 and makes its localization RNase H-resistant. Subsequent recruitment of the CLRC complex helps to methylate the histones and promote formation of a more stable ternary complex with Swi6-HP1.