A NASP (N1/N2)-Related Protein, Sim3, Binds CENP-A and Is Required for Its Deposition at Fission Yeast Centromeres

Summary A defining feature of centromeres is the presence of the histone H3 variant CENP-ACnp1. It is not known how CENP-ACnp1 is specifically delivered to, and assembled into, centromeric chromatin. Through a screen for factors involved in kinetochore integrity in fission yeast, we identified Sim3. Sim3 is homologous to known histone binding proteins NASPHuman and N1/N2Xenopus and aligns with Hif1S. cerevisiae, defining the SHNi-TPR family. Sim3 is distributed throughout the nucleoplasm, yet it associates with CENP-ACnp1 and also binds H3. Cells defective in Sim3 function have reduced levels of CENP-ACnp1 at centromeres (and increased H3) and display chromosome segregation defects. Sim3 is required to allow newly synthesized CENP-ACnp1 to accumulate at centromeres in S and G2 phase-arrested cells in a replication-independent mechanism. We propose that one function of Sim3 is to act as an escort that hands off CENP-ACnp1 to chromatin assembly factors, allowing its incorporation into centromeric chromatin.

Levels of myc-CENP-A Cnp1 protein are similar in wild type and sim3 mutants grown at 25 o C and 36 o C (6 hr). Western using anti-myc antibodies to detect myc-tagged CENP-A Cnp1 protein (expressed from its endogenous promoter), or anti-tubulin as a loading control.

Figure S4
Levels of histone H3 protein are similar in wild type and sim3 mutants grown at 25 o C and 36 o C (6 hr). Western analysis using anti-histone H3 C terminal antibody to detect levels of histone H3 in wild type and sim3 mutants at 25 o C and 36 o C (6 hr), or anti-tubulin as a loading control.

Figure S5
sim3 mutants are partially rescued by increased expression of histone H4 overexpression but antagonised by additional H3. Serial dilution assay of wild type and sim3 cells expressing additional CENP-A Cnp1 , H3 or H4. Light pink colonies on phloxine B indicate healthy growth; dark pink indicates accumulation of dead cells.

Figure S7
Top: GST-Sim3 or GST alone was incubated with recombinant histone H3-H4-H2A-H2B octamers, H3-H4 tetramers or free core histones from Calf Thymus (Upstate). GST-Sim3 has an affinity for all four histone regardless of the source. BSA was added as a competitor in binding assays with free core histone (right panel). Upper histone band is H3, then H2B, then H2A and lowest is H4. GST-Sim3 may enrich for H3 over H2A and H2B in binding assays with free histones (right panel). Bottom: Two Hybrid assay in S. cerevisiae with strains expressing CENP-A Cnp1 , H3 or Swi6 fused to the Gal4 DNA Binding domain (GBD) or GDB alone in PAS2-1 and therefore targeted to the Gal4 sites residing adjacent HIS2 and ADE2 genes in the tester strain. Strains also express Sim3, H3, or Swi2 fused to the Gal4 activation domain (ACT) or ACT alone in pACT2. Interactions are indicated by increased growth on plates lacking Histidine (-His) and containing 20 mM 3-AT or on plates lacking adenine (-Ade). Two hybrid plasmids were selected on plates lacking tryptophan and leucine (-Trp -Leu). The Swi6-Swi2 interaction serves as a positive control.

Figure S8
Newly induced GFP-CENP-A Cnp1 spots are localised at centromeres. inv-GFP-CENP-A Cnp1 was induced in wild type cells for 1 hour at 25 o C and cells were fixed and co-stained with anti-GFP (green), anti-Sad1 (red) to mark the SPB and DNA was stained with DAPI (blue). Strain FY8481.

Figure S9
Localisation of newly induced HA tagged histone H3 in wild type and sim3 mutants at 25 o C. Wild type and sim3 mutants containing the pinv-H3HA plasmid (Choi et al., 2005) were shifted from repressed to inducing conditions 1 hour prior to fixation and immunolocalisation with anti-HA antibody (green) and staining of DNA (red). The percentage of cells in which H3-HA localised to chromosomal DNA was determined (n=200). Scale bar, 10 µm. Culture of fission yeast followed standard protocols (Moreno et al., 1991).

Yeast strain construction
A list of S. pombe strains used in this study is shown in Table S2. Strain FY8481 was constructed as follows: the inv1 promoter was isolated by PCR as a 2.3 kb A diploid with the sim3 + ORF replaced with the KanMX cassette (obtained from Bioneer) was spoulated and tetrads dissected. Slow growing sim3∆::KanMX (sim3∆) haploids were isolated and further characterised.

Plasmids and primers
A list of the primers used in this study is shown in Table S2. Blots were developed using ECL reagents (Amersham Biosciences).

Cytology
The following antibodies were used: sheep anti-CENP-A Cnp1 antiserum 1:300 to

Northern Blotting
For Northern blotting, 800 bp GFP and 1 kb adh1 probes were amplified from genomic DNA, were synthesised using High Prime (Roche) and were hybridised overnight at 55 o C.

Expression Profiling
cDNA expression profiling was carried out according to (Xue et al., 2004). We used the S. pombe ORF spotted microarrays containing 5029 ORF probes (Eurogentec, Belgium). Gene expression in logarithmically growing cultures of sim3 mutant cells was compared to that in wt controls to establish lists of genes affected by sim3-143 and sim3-205. 'GeneSpring' software (Agilent) was used for all the data analysis.
The 'Lowess' (per spot per chip) intensity-dependent normalization, which corrects nonlinear rates for dye incorporation, was used. The 'Gene dB' product descriptions for the affected genes were downloaded from www.genedb.org/genedb/pombe/ and linked to the gene names using Microsoft Access (See Table S1). The Gene Expression omnibus (GEO) submission series for the sim3 microarray data is GSE7560 at http://www.ncbi.nlm.nih.gov/geo/.