Characterization of a highly conserved histone related protein, Ydl156w, and its functional associations using quantitative proteomic analyses

Abstract

A significant challenge in biology is to functionally annotate novel and uncharacterized proteins. Several approaches are available for deducing the function of proteins in silico based upon sequence homology and physical or genetic interaction, yet this approach is limited to proteins with well-characterized domains, paralogs and/or orthologs in other species, as well as on the availability of suitable large-scale data sets. Here, we present a quantitative proteomics approach extending the protein network of core histones H2A, H2B, H3, and H4 in Saccharomyces cerevisiae, among which a novel associated protein, the previously uncharacterized Ydl156w, was identified. In order to predict the role of Ydl156w, we designed and applied integrative bioinformatics, quantitative proteomics and biochemistry approaches aiming to infer its function. Reciprocal analysis of Ydl156w protein interactions demonstrated a strong association with all four histones and also to proteins strongly associated with histones including Rim1, Rfa2 and 3, Yku70, and Yku80. Through a subsequent combination of the focused quantitative proteomics experiments with available large-scale genetic interaction data and Gene Ontology functional associations, we provided sufficient evidence to associate Ydl156w with multiple processes including chromatin remodeling, transcription and DNA repair/replication. To gain deeper insights into the role of Ydl156w in histone biology we investigated the effect of the genetic deletion of ydl156w on H4 associated proteins, which lead to a dramatic decrease in the association of H4 with RNA polymerase III proteins. The implication of a role for Ydl156w in RNA Polymerase III mediated transcription was consequently verified by RNA-Seq experiments. Finally, using these approaches we generated a refined network of Ydl156w-associated proteins.

Fig. 1.

The identification of histone associated proteins. A, Venn diagram illustrating the shared nature of the histone core. Subunits common to all four baits are indicated by a black box. B, The relative abundance of the prey proteins in the core histones represented by dNSAF. Average dNSAF values for 25 preys that were shared by all histones and scored highly in the first singular vector of the SVD analysis. The blue color intensity is indicative of the dNSAF value according to the scale shown at the right side of the heat map. High dNSAFs are represented by the light blue whereas low dNSAF values correspond to dark blue. C, Relative protein abundance determined for the core histones as obtained from the H4 (n = 4 biological replicates), H3 (n = 4 biological replicates), H2A (n = 4 biological replicates), and H2B (n = 5 biological replicates) TAP purifications. Data in (C) is expressed as average dNSAF values + standard deviations.

Fig. 2.

dNSAF analyses of the histones and Ydl156w. A, Silver stained gel for the Ydl156w-TAP purification. The histone proteins along with Ydl156w are indicated on the right side of the gel. B, Relative protein abundance for the core histones and Ydl156w obtained from the Ydl156w-TAP (n = 6). C, The relative abundance of additional prey proteins (that are also identified in the four histone purifications) in the Ydl156w-TAP (n = 6 biological replicates), represented by dNSAF. Data in (B) and (C) are expressed as average dNSAF values + standard deviations.

Fig. 3.

Detected proteins in the Ydl156w-TAP and their functional classifications. A, Gene Ontology classes (“biological process”) significantly overrepresented (in percent, as measured by GOstat (20)) in the Ydl156w purifications. All six biological replicates were used for GO analysis, however, for visualization purposes we only represented the top 20 overrepresented (by percentage) categories of “biological processes.” B, An interaction network that depicts the Ydl156w associated proteins identified by MudPIT. Proteins identified in at least three biological replicates are depicted in (B) and used to create a network of Ydl156w associations using Cytoscape (56, 57).

Fig. 4.

Ydl156w exists in high molecular weight complexes. The yeast protein Ydl156w was purified from S. cerevisiae and fractionated by size exclusion chromatography on a superose 6 HR 10/30 column. A, Fractions were collected and aliquots were TCA precipitated, separated using SDS-PAGE and transferred to nitrocellulose membranes for Western blotting. The TAP-tagged Ydl156w was detected with polyclonal anti-TAP IgG. B, MudPIT analyses were carried out on superpose six fractions and dNSAF values of the 17 proteins identified in fraction 5, as well as their relative abundance levels in fraction 23 are shown.

Fig. 5.

Classification of detected proteins in H4-TAP in comparison with H4-TAP in ydl156w deletion strain. A, The relative abundance of proteins of the RNA Polymerase III complex obtained from H4-TAP (n = 4 biological replicates) and H4-TAP in a ydl156w deletion strain (n = 3 biological replicates). B, Relative protein abundance determined for the major H4 interactions in the ydl156w mutant strain. Data in (A) and (B) are expressed as average dNSAF values + standard deviations. C, The top 20 overrepresented classes (“biological process”; in percent) as determined by GOstat. Gray color corresponds to the overrepresented classes in “biological process” in the H4-TAP whereas the red color illustrates the enriched categories in “biological processes” of the H4-TAP in the ydl156wΔ.

Fig. 6.

RNA-Seq analysis of gene expression in a ydl156w deletion strain. A, The results of comparing gene expression between wild type and ydl156w deletion cells are shown in the MA plot. The y axis represents the log₂ ratio of expression values for ydl156w/wt, whereas the x axis represents the overall magnitude of the signal coming from both samples, calculated by the expression log₂(sqrt(wt*ydl156w)). Values highlighted in color represent genes with an adjusted p value of less than 0.05. The number of genes enriched in ydl156w or wild-type by this criteria is indicated. The dash gray line represents a twofold difference in expression up or down. B, 10 most significant functional classes as determined by GOstat analysis in the group of up- and down-regulated transcripts. The x axis depicts the percentage relative to all transcripts of the functional class. C, Results of the hypergeometric test comparing the RNA-Seq data to the RNA Polymerase III ChIP-chip data performed by Roberts et al. (24). The p values indicating the statistical significance of enrichment of RNA Polymerase III occupied transcripts in the RNA-Seq data are reported for different average percentiles.