Chromatin Central: towards the comparative proteome by accurate mapping of the yeast proteomic environment

High resolution mapping of the proteomic environment and proteomic hyperlinks in fission and budding yeast reveals that divergent hyperlinks are due to gene duplications.


Figure S2
Images of Coomassie stained gels produced in immunoprecipitation experiments that deciphered the Chromatin Central environment in S.cerevisiae. Corresponding baits are indicated on the top of each lane. Proteins were identified by mass spectrometry as described in the Materials and Methods section, although for presentation clarity, only bands containing proteins relevant for Chromatin Central are annotated. Full list of proteins identified in each of the experiments is provided in Table S2 in Additional data file 1 IPs were performed in sequential rounds, as explained in the text. The series started with IP of ESA1-TAP and then 5 candidate subunits of plausible ESA1-containing complex (termed NuA4) were, in turn, tagged and immunopurified in Round 1. Immunoisolation of YNG2-TAP and EPL1-TAP did not bring new interactors, compared to IP of ESA1-TAP and validated the composition of the complex. At the same time, IPs of YAF9-TAP and SWC4-TAP produced new interactors that were not associated with ESA1 directly. They were subjected to tagging and immunoisolation in rounds 2 and 4. Baits for the new round of IPs were selected in the similar manner, as presented in the figure.
Gel images of immunoisolations of DOT6-TAP, ASA3-TAP and TEL2-TAP (Rounds 3 and 5, respectively) are not presented since no bands were detectable by Coomassie staining. These gels were cut in ca.30 slices that were separately digested with trypsin and analyzed by LC-MS/MS. Gel image of the immunoisolation of SNT2-TAP (Round 5) is presented in Figure 5A

Figure S3
A YBR095C/Rxt2 family alignment  KKEP  Table 1) at which point the two proteins show a reciprocal best hit relationship. All Rxt2 orthologues share an N-and C-terminal region of sequence similarity with a long central gap that lacks significant sequence similarity.

B: Multiple sequence alignment of fungal Dep1 family members
Budding and fission yeast orthologues of Dep1 detected each other with significant Evalues at the 2 nd PSI-BLAST iteration (Supplemental Table 1) and show a reciprocal best hit relationship. Following a short region of similarity in the N-terminus of Dep1 family members, the proteins contain a central region without significant similarity that is followed by a highly conserved C-terminal part. In the S. pombe sequence, an SDS3 domain is detected in the C-terminal conserved region.

C: Multiple sequence alignment of fungal Swc3 family members
S. cerevisiae Swc3 did not pick up the S. pombe orthologue in PSI-BLAST searches due to profile drifting that result from extended low-complexity regions present in Swc3 protein sequences. The S. pombe SPAC4H3.02c protein however detected S.
cerevisiae Swc3 the 2 nd PSI-BLAST iteration as the best hit in S. cerevisiae with an Evalue of 3E-05 (Supplemental Table 1 Table 1). The conserved region of this protein family is limited to the very C-terminal region of the proteins.

E: Multiple sequence alignment of fungal Ies4 family members
S. cerevisiae Ies4 did not detect the predicted S. pombe SPAC23G3.04 due to profile drifting induced by extended low-complexity regions of Ies4 family members.
SPAC23G3.04 did however pick up the orthologue of S. cerevisiae Ies4 as its best hit.
K. lactis Ies4 also showed a reciprocal best hit relationship with both, the S. cerevisiae and S. pombe Ies4 proteins. Even though the conservation of Ies4 members is very remote, they share a central region of similarity with a distinct pattern of charged and aromatic amino acids.

F: Multiple sequence alignment of fungal Asa1 family members
S. cerevisiae Asa1 detected S. pombe SPAC1006.02 as the best hit in fission yeast in the 2 nd PSI-BLAST iteration with an E-value of 4E-28. The S. pombe protein SPAC1006.02 does not detect Asa1 as the best hit, which is mostly due to repetitive alignments with alternative WD40 repeat proteins from budding yeast that result in lower E-values. It shows however a best reciprocal hit relationship with for example the A. fumigatus orthologue of Asa1. Identification of Iec6p -a new protein with unknown sequence in S.pombeusing mass spectrometry and similarity searching. A: MS spectrum acquired from intensive Coomassie band running at ca 20kDa on Sp_Rvb1p-TAP IP gel (see corresponding gel pic.); trypsin autolysis products are marked with "T". 5 peptide precursors were fragmented from the sample; however MASCOT search against a complete protein database with uninterpreted MS/MS data did not produce any confident hits. Fragmentation spectra of five peptide precursors (indicated with arrows) were manually interpreted, then all sequence suggestions were merged in one string and searched against protein database and S.pombe genome using MS BLAST similarity searching program (peptide query is shown in the insert). When no confident hits were found in a protein DB, four out of five peptides were matched exactly to a 132 aa-long non-coding region of the chromosome I. B: Iec5p final sequence. The new protein was cloned and all five peptides suggested by mass spectrometry were found in the final sequence (underlined in bold). C. IEC5 gene contains two introns.