Summary Although the function of the mammalian pancreas hinges on complex interactions of distinct cell types, gene expression profiles have primarily been described with bulk mixtures. Here we implemented a droplet-based, single-cell RNA-seq method to determine the transcriptomes of over 12,000 individual pancreatic cells from four human donors and two mouse strains. Cells could be divided into 15 clusters that matched previously characterized cell types: all endocrine cell types, including rare epsilon-cells; exocrine cell types; vascular cells; Schwann cells; quiescent and activated stellate cells; and four types of immune cells. We detected subpopulations of ductal cells with distinct expression profiles and validated their existence with immuno-histochemistry stains. Moreover, among human beta- cells, we detected heterogeneity in the regulation of genes relating to functional maturation and levels of ER stress. Finally, we deconvolved bulk gene expression samples using the single-cell data to detect disease-associated differential expression. Our dataset provides a resource for the discovery of novel cell type-specific transcription factors, signaling receptors, and medically relevant genes.

Many extracellular proteins with diverse functions contain domains similar to epidermal growth factor (EGF), a number of which have a consensus Asp/Asn, Asp/Asn, Asp*/Asn*, Tyr/Phe (where the asterisk denotes a beta-hydroxylated residue). These include the coagulation factors IX and X, proteins with two EGF-like domains, the first of which contains the consensus residues. The first EGF-like domain of human factor IX contains a calcium-binding site, which is believed to be responsible for one of the high-affinity sites detected in this protein. Similar results have been obtained for bovine factor X. We have now used protein engineering and 1H-NMR techniques to investigate the importance of individual consensus residues for ligand binding. Measurement of a calcium-dependent Tyr 69 shift in the isolated first EGF-like domain from human factor IX demonstrates that Asp 47, Asp 49, and Asp 64 are directly involved in this binding. Gln 50, whose importance has previously been overlooked, is also involved in this binding. Two mutations in this domain, Asp 47----Glu, and Asp 64----Asn, present in patients with haemophilia B, reduce calcium binding to the domain greater than 4-fold and greater than 1,000-fold, respectively. Furthermore, the defective calcium binding of Asn 64 can be partially rescued by the compensatory mutation Gln 50----Glu. This latter mutation, when introduced singly more than doubles the affinity of the domain for calcium. This study thus defines residues involved in a new type of calcium-binding site and provides strong circumstantial evidence for calcium-binding motifs in many extracellular proteins, including the developmentally important proteins of Drosophila, notch, delta and crumbs.

The dumpy gene encodes a gigantic extracellular molecule that we predict to be a membrane-anchored fibre of almost a micrometer in length. Insertion and crosslinking of this fibre within the cuticle may provide a strong anchor for the underlying tissue, allowing it to maintain mechanical tension at sites under stress. This would explain its contribution to tissue morphogenesis through its regulation of mechanical properties.

Bipolar disorder (BP) is a severe and common psychiatric disorder characterized by extreme mood swings. Family, twin and adoption studies strongly support a genetic component. The mode of inheritance is complex and likely involves multiple, as yet unidentified genes. To identify susceptibility loci, we conducted a genome-wide scan with 343 microsatellite markers in one of the largest, well-characterized pedigree samples assembled to date (373 individuals in 40 pedigrees). To increase power to detect linkage, scan statistics were used to examine the logarithm of odds (lod) scores based on evidence at adjacent chromosomal loci. This analysis yielded significant evidence of linkage (genome-wide Po0.05) for markers on 2p13-16. Standard linkage analysis was also supportive of linkage to 2p13-16 (lod ¼ 3.20), and identified several other interesting regions: 4q31 (lod ¼ 3.16), 7q34 (lod ¼ 2.78), 8q13 (lod ¼ 2.06), 9q31 (lod ¼ 2.07), 10q24 (lod ¼ 2.79), 13q32 (lod ¼ 2.2), 14q21 (lod ¼ 2.36) and 17q11-12 (lod ¼ 2.75). In this systematic, large-scale study, we identified novel putative loci for BP (on 2p13-16, 8q13 and 14q21) and found support for previously proposed loci (on 4q31, 7q34, 9q31, 10q21-24, 13q32 and 17q11-12). Two of the regions implicated in our study, 2p13-14 and 13q32, have also been linked to schizophrenia, suggesting that the two disorders may have susceptibility genes in common.