Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 43 (20), e132

Biological Chromodynamics: A General Method for Measuring Protein Occupancy Across the Genome by Calibrating ChIP-seq

Affiliations

Biological Chromodynamics: A General Method for Measuring Protein Occupancy Across the Genome by Calibrating ChIP-seq

Bin Hu et al. Nucleic Acids Res.

Abstract

Sequencing DNA fragments associated with proteins following in vivo cross-linking with formaldehyde (known as ChIP-seq) has been used extensively to describe the distribution of proteins across genomes. It is not widely appreciated that this method merely estimates a protein's distribution and cannot reveal changes in occupancy between samples. To do this, we tagged with the same epitope orthologous proteins in Saccharomyces cerevisiae and Candida glabrata, whose sequences have diverged to a degree that most DNA fragments longer than 50 bp are unique to just one species. By mixing defined numbers of C. glabrata cells (the calibration genome) with S. cerevisiae samples (the experimental genomes) prior to chromatin fragmentation and immunoprecipitation, it is possible to derive a quantitative measure of occupancy (the occupancy ratio - OR) that enables a comparison of occupancies not only within but also between genomes. We demonstrate for the first time that this 'internal standard' calibration method satisfies the sine qua non for quantifying ChIP-seq profiles, namely linearity over a wide range. Crucially, by employing functional tagged proteins, our calibration process describes a method that distinguishes genuine association within ChIP-seq profiles from background noise. Our method is applicable to any protein, not merely highly conserved ones, and obviates the need for the time consuming, expensive, and technically demanding quantification of ChIP using qPCR, which can only be performed on individual loci. As we demonstrate for the first time in this paper, calibrated ChIP-seq represents a major step towards documenting the quantitative distributions of proteins along chromosomes in different cell states, which we term biological chromodynamics.

Figures

Figure 1.
Figure 1.
ChIP-seq profiles are unaffected by reference cells. Crude extracts prepared from exponentially grown S. cerevisiae cells (K14601, MATa, Scc1PK9::KanMX), C. glabrata (K23308, MATa, Scc1PK9::NatMX), or a mixture were processed for ChIP-seq. (A) All sequences from whole cell extracts (W) were aligned to the experimental (S.cerevisiae) and reference genomes (C. glabrata) or both. The numbers in the left (red) or right (blue) circles indicate the percentage of reads that align uniquely to S. cerevisiae or C. glabrata genomes. Those in the intersection indicate the percentage of reads that align to both. ChIP-seq distributions of SacCer_Scc1 on chromosome I (B) or rDNA region (C), aligning either all reads or only those unique to S. cerevisiae, from pure S. cerevisiae or mixed cultures respectively. The Y-axis indicates the numbers of reads covering every base pair and the X-axis indicates position of every base pair adopted from SGD (http://www.yeastgenome.org).
Figure 2.
Figure 2.
Occupancy ratio (OR) is unaffected by the amount of reference cells. ChIP-seq experiments were carried out from exponentially growing S. cerevisiae cells mixed with exponentially growing C. glabrata cells in the indicated ratios. Only unique reads to either genome were used for alignment. (A) Correlation between percentages of reads aligning to S. cerevisiae genome from IP and whole cell extract (W) samples. (B) ORs of mixtures with the indicated ratios. (C) ChIP-seq distributions of SacCer_Scc1 on chromosome I from mixtures with different ratios of S. cerevisiae and C. glabrata cells. The Y-axis indicates the numbers of reads covering every base pair and the X-axis indicates position of every base pair adopting from SGD (http://www.yeastgenome.org).
Figure 3.
Figure 3.
Using calibrated ChIP-seq to measure cohesin's cell cycle dependent association with the S. cerevisiae genome. Exponential phase S. cerevisiae cells growing at 25°C were arrested in G1 by treatment with α-factor pheromone for 150 min. and then transferred by filtration to fresh medium lacking the pheromone. This triggered their synchronous passage through the cell cycle. At the indicated time points, samples were removed from the synchronous S.cerevisiae culture to measure DNA content by FACS (A), the fraction of cells with buds (budding index) (B), the fraction of binucleate cells (B), and OR values for each time point (C). In the case of the latter, S.cerevisiae cell samples were first fixed with formaldehyde and only subsequently mixed with pre-fixed C. glabrata cells, a protocol that made it simpler to sample continuously for up to 120 min. from the S. cerevisiae synchronous culture. Occupancy ratio (OR) values for PK-tagged Scc1 were calculated for each time point using reads unique to S. cerevisiae and C. glabrata cells in IP and whole cell extract (W) aliquots, as described in Materials and Methods. The amount of CEN6 DNA within IP samples was also measured using qPCR and the values compared to estimates for the equivalent locus derived from the calibrated ChIP-seq profiles (D) (see also Figure 5). Both sets of values are plotted on an arbitrary linear scale designed so that the ‘areas’ under each curve were identical. Note that CEN6 occupancy rises and falls more rapidly than OR. Also shown in D are CEN6 values calculated from uncalibrated data.
Figure 4.
Figure 4.
Conventional ChIP-seq profiles of cohesin (Scc1) during synchronous S. cerevisiae cell cycles. The ChIP-seq profile of Scc1 along chromosome 1 from each time point of the experiment described in Figure 3 shown on the basis of reads_per_million. Note that under these circumstances, which is the conventional way of presenting ChIP-seq data, the area under the curve (for the whole genome) will be identical between each sample. The Y-axis indicates the numbers of reads covering every base pair and the X-axis indicates position of every base pair adopting from SGD (http://www.yeastgenome.org).
Figure 5.
Figure 5.
Cohesin chromodynamics: calibrated ChIP-seq profiles of Scc1 during synchronous S.cerevisiae cell cycles. The calibrated ChIP-seq profiles of Scc1 were calculated by multiplying the conventional ChIP-seq signals (Figure 4) with the OR values of each time point (Figure 3C). Unlike conventional ChIP-seq, the calibrated profiles reveal the changes in cohesin's occupancy of all sites within the genome as cells progress through their synchronous cell cycles. The Y-axis indicates the numbers of reads covering every base pair and the X-axis indicates position of every base pair adopting from SGD (http://www.yeastgenome.org).
Figure 6.
Figure 6.
Calibrated ChIP-seq unlike conventional ChIP-seq distinguishes signals from noise. (A) Calibrated ChIP-seq profiles for chromosome 1 in pheromone arrested cells (0) and 60 min after release (60). (B) Calibrated ChIP-seq profiles for chromosome 1 obtained using the PK-specific antibody from exponentially growing S. cerevisiae cells with or without PK tagged Scc1 (K14601 or K699). To calibrate each sample, they were mixed with exponentially growing C. glabrata cells prior to fixation. (C) Calibrated ChIP-seq profiles of PK tagged Scc1 in the presence or absence of Scc2 activity. Wild type (K14601) and ts scc2–45 cells (K22390) growing exponentially at 25°C were uniformly arrested in early G1 by incubation with α factor pheromone for 2.5 hour. The arrested S.cerevisiae cells were released into a new cell cycle by transferring to YPD media containing nocodazole at the restrictive temperature 37°C. After 60 min, by which time most cells had completed DNA replication, cells were fixed with formaldehyde and mixed with separately fixed C. glabrata cells before processing samples for ChIP-seq. The Y-axis indicates the numbers of reads covering every base pair and the X-axis indicates position of every base pair adopting from SGD (http://www.yeastgenome.org).
Figure 7.
Figure 7.
Calibrated ChIP-seq profiles of wild type, acetylation-mimicked mutant and ATPase hydrolysis mutant Smc3. The yeast cells ectopically expressing PK-tagged wild type Smc3 (K17407), acetylation-mimicked mutant smc3_K112Q, K113Q (K22703) or ATPase hydrolysis defective mutant smc3_E1155Q (K17409) were exponentially grown at 25°C and mixed with C. glabrata cells. Calibrated ChIP-seq was performed against PK epitope. (A) The calibrated ChIP-seq profiles of Chr 1 are shown with a full scale of Y-axis and the position of CEN1 is indicated. (B) The enlarged DNA association profiles of WT and indicated mutant Smc3 are shown with a smaller scale of Y-axis. The detailed distribution of WT and mutant Smc3 at pericentromere (around 30 kb of flanking region on either side of CEN1) is shown underneath. The Y-axis indicates the numbers of reads covering every base pair and the X-axis indicates position of every base pair adopting from SGD (http://www.yeastgenome.org). Note that the E11155Q peak associated with CEN1 has been truncated in (B).
Figure 8.
Figure 8.
Cohesin dynamics around the centromere. The number of reads per base pair (Y-axis) in the 15kb regions either side of the CDEIII element for each chromosome (X-axis) were averaged and calibrated to reveal the average cohesin distribution around the centromere. (A) The profiles of the average cohesin distribution throughout the cell cycle from the data in Figure 3. (B) The average cohesin distribution in cycling Smc3 E1155Q ATP hydrolysis mutant cells compared to WT cells.
Figure 9.
Figure 9.
Pericentric cohesin viewed by calibrated ChIP-seq and cell images. (A) Cohesin's peri-centric distribution in G2/M phase cells. The number of reads per base pair (Y-axis) in the 30 kb regions either side of the CDEIII element for each chromosome (X-axis) were averaged and calibrated to reveal the average cohesin distribution around the centromere. The data was extracted from ChIP-seq of cells at 60 min. after released from G1-arrest in Figure 5. The ‘5 kb’ peaks are marked by red boxes and centromere-proximal cohesin marked by blue box. (B) High-resolution image of cohesin in G2/M phase. Tetraploid cells containing 64 chromosomes with endogenous Scc1 tagged with EGFP (K18719) were exponentially grown in YPD at 25°C and fixed with 2% of formaldehyde. The cells were observed under high-resolution OMX microscope. The lateral view is shown on the top panel and the transverse view on the bottom one. (C) Bi-orientation of sister kinetochores and peri-centric cohesion mediated by ‘5 kb’ peaks (red boxes) in G2/M phase cells. Sixty-four such bi-oriented chromosomes will be clustered around pole to pole microtubules, generating the cohesin barrels shown in (B). Chromosome axes – the black fibre. Kinetochores – mauve balls. Centromere-proximal cohesin - blue box. Pole to pole microtubules – green cylinders.

Similar articles

See all similar articles

Cited by 43 articles

See all "Cited by" articles

References

    1. Barski A., Cuddapah S., Cui K., Roh T.-Y., Schones D.E., Wang Z., Wei G., Chepelev I., Zhao K. High-resolution profiling of histone methylations in the human genome. Cell. 2007;129:823–837. - PubMed
    1. Park P.J. ChIP-seq: advantages and challenges of a maturing technology. Nat. Rev. Genet. 2009;10:669–680. - PMC - PubMed
    1. Teytelman L., Thurtle D.M., Rine J., van Oudenaarden A. Highly expressed loci are vulnerable to misleading ChIP localization of multiple unrelated proteins. Proc. Natl. Acad. Sci. U.S.A. 2013;110:18602–18607. - PMC - PubMed
    1. Bonhoure N., Bounova G., Bernasconi D., Praz V., Lammers F., Canella D., Willis I.M., Herr W., Hernandez N., Delorenzi M., et al. Quantifying ChIP-seq data: a spiking method providing an internal reference for sample-to-sample normalization. Genome Res. 2014;24:1157–1168. - PMC - PubMed
    1. Orlando D.A., Chen M.W., Brown V.E., Solanki S., Choi Y.J., Olson E.R., Fritz C.C., Bradner J.E., Guenther M.G. Quantitative ChIP-Seq normalization reveals global modulation of the epigenome. Cell Rep. 2014;9:1163–1170. - PubMed

Publication types

MeSH terms

Feedback