Systematic comparison of monoclonal versus polyclonal antibodies for mapping histone modifications by ChIP-seq

Abstract

Background: The robustness of ChIP-seq datasets is highly dependent upon the antibodies used. Currently, polyclonal antibodies are the standard despite several limitations: They are non-renewable, vary in performance between lots and need to be validated with each new lot. In contrast, monoclonal antibody lots are renewable and provide consistent performance. To increase ChIP-seq standardization, we investigated whether monoclonal antibodies could replace polyclonal antibodies. We compared monoclonal antibodies that target five key histone modifications (H3K4me1, H3K4me3, H3K9me3, H3K27ac and H3K27me3) to their polyclonal counterparts in both human and mouse cells.

Results: Overall performance was highly similar for four monoclonal/polyclonal pairs, including when we used two distinct lots of the same monoclonal antibody. In contrast, the binding patterns for H3K27ac differed substantially between polyclonal and monoclonal antibodies. However, this was most likely due to the distinct immunogen used rather than the clonality of the antibody.

Conclusions: Altogether, we found that monoclonal antibodies as a class perform equivalently to polyclonal antibodies for the detection of histone post-translational modifications in both human and mouse. Accordingly, we recommend the use of monoclonal antibodies in ChIP-seq experiments.

Keywords: Antibodies; ChIP-seq; Methods; Monoclonal; Polyclonal.

Fig. 1

Read coverage across the genome. Images of tiled data files (TDFs) generated by the IGV browser [34, 35] displaying the density tracks of reads aligned across the genome. The tracks show the correspondence in read coverage in monoclonal and polyclonal antibodies over representative genomic loci. a Chromosome 7: 44,829,782–44,930,648 (about 100 kb), shows the read coverage of histone modifications associated with “active chromatin” (H3K4me1, H3K4me3 and H3K27ac). The correspondence of read coverage of datasets for two major histone modifications associated with repression: b H3K27me3 [chromosome 22:19,492,023–19,849,594 (about 350 kb)] and c H3K9me3 [chromosome 19: 51,746,058–53,362,194 (about 1.6 Mb)]

Fig. 2

a Saturation curve showing the number of bases called as being in peaks as a function of sequencing depth. The final dataset of the merged technical replicates was randomly downsampled to 20 different read depths, and peaks were called in each dataset using HOMER. b Distribution of the canonical ENCODE regions of the genomic bases identified as being in peaks. Note that distribution of bases called in both the monoclonal and polyclonal antibody differs from the distribution of bases called by only one antibody with fewer bases in their expected regions. c Left bases of the genome that were designated as peaks were identified as being in the expected canonical ENCODE region versus other regions. Only genomic bases annotated in the ENCODE segmentation tracks for K562 are included in this calculation. Right Venn diagrams displaying the overlap of peak calls in the monoclonal and polyclonal antibodies. The bases of the genome are identified as being in peaks by the monoclonal (red), polyclonal (blue) or both (purple) antibodies

Fig. 3

Reads in peaks mapping to canonical chromatin regions of the genome as defined by the ENCODE mappings. This plot displays the percentage of reads that map to each canonical genome region. The canonical genome regions were defined by the combined ENCODE mapping and are abbreviated as follows: CTCF-enriched elements (CTCF), promoter flanking regions (PF), transcription start sites (TSS), transcribed regions (T), enhancers (E), weak enhancers (WE) and repressed regions (R). Only reads that were located at regions identified as peaks were used for this plot. For each peak dataset the reads were normalized by insert size

Fig. 4

Correlation between monoclonal and polyclonal antibodies across the genome. Scatter plots (Loglog) presenting counts of reads per bin in non-overlapping 2000-bp windows tiled throughout the genome in replicates of the monoclonal antibody (left; gray), the polyclonal antibody (right; gray) and polyclonal versus monoclonal (center; blue). The H3K27me3 data (a), show that the reproducibility is nearly indistinguishable from the reproducibility of data derived from technical replicates using the same antibody, while the H3K27ac data (b) show divergence between polyclonal and monoclonal antibodies

Fig. 5

Variability in H3K27ac patterns is dependent on the immunogen. a Scatter plots where each point represents the count of reads aligning to a non-overlapping, variably sized region as annotated in the chromatin regions determined by ENCODE mapping of the genome. Values are summed for the replicates of monoclonal and polyclonal H3K27ac antibodies. The red line (on the left and right plots) represents slope = 1. b H3K27ac antibodies in HeLa cells. R ² is indicated for all points, TSS and enhancer regions

Fig. 6

Correlation between two monoclonal lots across the genome. Scatter plots (Loglog) presenting counts of reads per bin in non-overlapping 2000-bp windows tiled throughout the genome comparing either technical or lot replicates from ChIP-seq done with H3K4me3 monoclonal antibody in K562, GM12878 and mES