Epigenomic-based identification of major cell identity regulators within heterogeneous cell populations [ChIP-seq]

Cellular heterogeneity within embryonic and adult tissues is involved in multiple biological and pathological processes. Here we present a simple and broadly applicable epigenomic strategy that allows the functional dissection of cellular heterogeneity. By integrating H3K27me3 Chip-seq and RNA-seq data, we demonstrate that the presence of broad H3K27me3 domains at transcriptionally active genes reflects the heterogeneous expression of major cell identity regulators. Using dorsoventral patterning of the spinal neural tube as a model, the proposed approach successfully identifies the majority (~90%) of previously known dorsoventral patterning transcription factors with high sensitivity and precision. Moreover, poorly characterized patterning regulators can be similarly predicted, as shown for ZNF488, which confers p1/p2 neural progenitor identity. Finally, we show that, as our strategy is based on universal chromatin features, it can be also used to functionally dissect cellular heterogeneity within various organisms and tissues, thus illustrating its potential applicability to a broad range of biological and pathological contexts. Overall design: Genome-wide binding profiles for various histone modifications were generated by ChIP-seq using spinal neural tube sections isolated at the brachial levels of stage HH14 or HH19 chicken embryos. For each ChIP-seq experiment, spinal neural tube sections of approximately 25-30 different embryos were pooled.