DNA methylation is a fundamental epigenetic mark that governs gene expression and chromatin organization, thus providing a window into cellular identity and developmental processes1. Current datasets typically include only a fraction of methylation sites and are often based either on cell lines that underwent massive changes in culture or on tissues containing unspecified mixtures of cells2–5. Here we describe a human methylome atlas, based on deep whole-genome bisulfite sequencing, allowing fragment-level analysis across thousands of unique markers for 39 cell types sorted from 205 healthy tissue samples. Replicates of the same cell type are more than 99.5% identical, demonstrating the robustness of cell identity programmes to environmental perturbation. Unsupervised clustering of the atlas recapitulates key elements of tissue ontogeny and identifies methylation patterns retained since embryonic development. Loci uniquely unmethylated in an individual cell type often reside in transcriptional enhancers and contain DNA binding sites for tissue-specific transcriptional regulators. Uniquely hypermethylated loci are rare and are enriched for CpG islands, Polycomb targets and CTCF binding sites, suggesting a new role in shaping cell-type-specific chromatin looping. The atlas provides an essential resource for study of gene regulation and disease-associated genetic variants, and a wealth of potential tissue-specific biomarkers for use in liquid biopsies.

We identify the alpha4 subunit of integrin as a predominant integrin expressed by neural crest cells in both avian and murine embryos. Using degenerate primers, we obtained a PCR fragment of the chick integrin alpha4 subunit that was subsequently used to clone the full-length subunit with a predicted amino acid sequence 60% identical to human and mouse alpha4 subunits. In situ hybridization demonstrates that chick integrin alpha4 mRNA is expressed at high levels by migrating neural crest cells and neural crest-derived ganglia at both cranial and trunk levels. An antibody against the murine alpha4 subunit revealed similar distribution patterns in mouse to chick. In addition to neural crest cells, the integrin alpha4 subunit was later observed on the muscle masses of the limb, the apical ectodermal ridge, and the developing liver. To examine the functional role of the integrin alpha4 subunit in neural crest cell migration, we used an explant preparation that allows visualization of neural crest cells in their normal environment with or without perturbing reagents. In the presence of a blocking antibody against the mouse integrin alpha4 subunit, there was a profound abrogation of neural crest cell migration at trunk and hindbrain levels. Both the numbers of migrating neural crest cells and the total distance traversed were markedly reduced. Similarly, avian embryos injected with synthetic peptides that contain the integrin alpha4 binding site in fibronectin displayed abnormal neural crest cell migration. Our results suggest that the integrin alpha4 subunit is important for normal neural crest cell migration and may be one of the primary alpha subunits used for neural crest cell migration in vivo. Furthermore, the integrin alpha4 subunit represents a useful neural crest marker in the mouse.

Circulating tumor cells (CTC) mediate metastatic spread of many solid tumors and enumeration of CTCs is currently used as a prognostic indicator of survival in metastatic prostate cancer patients. Some evidence suggests that it is possible to derive additional information about tumors from expression analysis of CTCs, but the technical difficulty of isolating and analyzing individual CTCs has limited progress in this area. To assess the ability of a new generation of MagSweeper to isolate intact CTCs for downstream analysis, we performed mRNA-Seq on single CTCs isolated from the blood of patients with metastatic prostate cancer and on single prostate cancer cell line LNCaP cells spiked into the blood of healthy donors. We found that the MagSweeper effectively isolated CTCs with a capture efficiency that matched the CellSearch platform. However, unlike CellSearch, the MagSweeper facilitates isolation of individual live CTCs without contaminating leukocytes. Importantly, mRNA-Seq analysis showed that the MagSweeper isolation process did not have a discernible impact on the transcriptional profile of single LNCaPs isolated from spiked human blood, suggesting that any perturbations caused by the MagSweeper process on the transcriptional signature of isolated cells are modest. Although the RNA from patient CTCs showed signs of significant degradation, consistent with reports of short half-lives and apoptosis amongst CTCs, transcriptional signatures of prostate tissue and of cancer were readily detectable with single CTC mRNA-Seq. These results demonstrate that the MagSweeper provides access to intact CTCs and that these CTCs can potentially supply clinically relevant information.