Skip to main page content
U.S. flag
See More

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2016 Oct 26;3(4):346-360.e4.
doi: 10.1016/j.cels.2016.08.011. Epub 2016 Sep 22.

A Single-Cell Transcriptomic Map of the Human and Mouse Pancreas Reveals Inter- and Intra-cell Population Structure

Maayan Baron  1 , Adrian Veres  2 , Samuel L Wolock  3 , Aubrey L Faust  4 , Renaud Gaujoux  5 , Amedeo Vetere  6 , Jennifer Hyoje Ryu  4 , Bridget K Wagner  6 , Shai S Shen-Orr  5 , Allon M Klein  7 , Douglas A Melton  8 , Itai Yanai  9
Affiliations

A Single-Cell Transcriptomic Map of the Human and Mouse Pancreas Reveals Inter- and Intra-cell Population Structure

Maayan Baron et al. Cell Syst. .

Abstract

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.

PubMed Disclaimer

Figures

Figure 1
Figure 1. A Transcriptomic Map of the Human and Mouse Pancreas
(A) Single-cell RNA-seq was carried out on human and mouse pancreatic islets using the inDrop microfluidics system to generate data that allow for quantification of transcript abundance across cells and genes. (B and C) Heatmap of all cells clustered by recursive hierarchical clustering (STAR Methods), showing selected marker genes for every population of human (B) and mouse (C). The top bar indicates assigned cluster identity (legend shown in F). The bottom bar indicates the donor of origin. (D and E) tSNE plot of cells from donor 1 based on the expression of highly variable genes for human (D) and mouse strain C57BL/6 (E). The detected clusters are indicated by different colors as shown in (F). (F) Schematic of the pancreatic islet and the cellular neighborhood along with the identified cell types and their respective markers.
Figure 2
Figure 2. The Endocrine Transcriptome Is Readily Distinguished from the Other Cell Types and across Donors
(A) Plot comparing the average expression (log10, transcripts per million [tpm]) of donor 1 beta and acinar cells. Genes with high differential expression are noted. (B) Dendrogram showing relationships among the cell types in human (left) and mouse (right). The dendrogram was computed using hierarchical clustering with average linkage on the log10 tpm values of the highly variable genes. (C) Same as (A) for the ductal cells of donors 1 and 3. (D) Heatmap indicating correlations on the averaged profiles among donors for all cell types (STAR Methods). (E) Same as (A) for human beta cells of donor 1 and mouse beta cells of mouse 1. (F) Heatmap indicating Pearson's correlations on the averaged profiles among common cell types for human and mouse.
Figure 3
Figure 3. Endocrine Transcriptomes Reveal Novel Expression Patterns of Key Genes
(A) Violin plots for expression of UCN3, FFAR4, LEPR, IAPP, and DLK1 across alpha, beta, gamma, and delta cell types for all human donors (left) and two mice (right). The percent number indicates the fraction of cells with detectable expression of the gene. (B) FFAR4 (GPR120) is expressed in delta and beta cells of human islet cells. C-peptide and Somatostatin mark beta and delta cells, respectively. Note the co-positive yellow cells in both merged images showing FFAR4 expression in both beta and delta cells. (C) Heatmap showing gene expression of transcription factors across the human (left) and mouse (right) endocrine cell types.
Figure 4
Figure 4. Multiple Modes of Pancreatic Stellate Cell Activation and Existence of Pancreatic Adult Neural Crest Stem Cells
(A) Illustration of pancreatic stellate cells within the pancreas, indicating their typical periacinar localization and activation from a quiescent state. (B) Hierarchical clustering of both human and mouse stellate cells on the basis of genes differentially expressed within human stellate cells and their mouse homologs. Genes highlighting the distinct clusters are displayed in the heatmap. Three groups of cells are indicated: quiescent, standard activated, and immune-activated. Most activated mouse stellate cells follow the pattern of standard activation. The three bottom rows indicate species and donor identities. (C) Plot comparing the average expression (tpm) of the two distinct populations of human activated stellate cells reveals genes involved in immune signaling and secretion of the extracellular matrix, as indicated by annotated genes. Annotated genes are differentially expressed (fold change > 2 and above variation expected from Poisson sampling) and indicative of different biological functions.
Figure 5
Figure 5. Subpopulations of Ductal Cells in the Human Pancreas
(A) Differential expression of MUC1, CFTR, TFF1, and CD44 in PC space defined by the ductal cells. Bottom: moving averages for each gene computed based upon equidistant ranges across PC1. (B) Heatmaps showing genes that are differentially expressed across PC1 for each of the three donors. Heatmaps including all genes names can be found in Figure S6. (C) Schematic of the location of terminal and centroacinar ductal cells. (D) Immunohistochemistry stains of human pancreata for CFTR and MUC1. Note their spatial isolation among the ductal cells. (E) Differential expression of Muc1, Cftr, Tff2, and Plat in PC space defined by the mouse ductal cells. Bottom: moving averages for each gene computed based upon equidistant ranges across PC1.
Figure 6
Figure 6. Heterogeneity of Beta Cells Reveals the Unfolded Protein Response
(A) PCA on beta cells colored by the transcript numbers. Beta cells were filtered for having at least 3,000 detected transcripts. (B) Moving average of HSPA5, MAFA, HERPUD1, DDIT3, and UCN3 sorted by PC1 score. Bottom: PCA plots colored by the expression of the same genes. (C) Heatmap showing expression of genes that contribute more to PC1 than expected. The profiles were computed as the moving average of 30 bins across PC1. Heatmaps including all genes names can be found in Figure S6.
Figure 7
Figure 7. BSeq-SC Uses Single-Cell RNA-Seq to Deconvolve Bulk Heterogeneous Tissue Data and Decouples Disease-Associated Differential Expression from Cell Type Proportion-Associated Differences
(A) Schematic of the BSeq-SC analysis. Single-cell RNA-seq information is used to deconvolve bulk pancreatic islet RNA-seq samples to estimate the cell type proportion of key cell types. Statistical deconvolution is used to leverage the variation in cell type frequencies between individuals to estimate average cell type-specific expression in diabetic versus healthy individuals and to compute cell type-specific differential expression. (B) Proportion differences of pancreatic cell types between normal and diabetic samples. (C) The majority of genes identified as differentially expressed between normal and diabetic in bulk samples are not significantly different following statistical adjustment to cell type proportions derived from the deconvolution. Genes that were not significant in the un-adjusted bulk analysis but were found to be significant after adjustment are shown in red. (D) Cell type-specific effect size in alpha (blue) and beta cells (purple) for the top ten significant genes between hyper- and normoglycemic groups. The ER stress genes UCN3 and MAFA were upregulated and downregulated in alpha and beta cells, respectively (see the complete list in Figure S9D).
See this image and copyright information in PMC

Comment in

References

    1. Benner C, van der Meulen T, Cacéres E, Tigyi K, Donaldson CJ, Huising MO. The transcriptional landscape of mouse beta cells compared to human beta cells reveals notable species differences in long non-coding RNA and protein-coding gene expression. BMC Genomics. 2014;15:620. - PMC - PubMed
    1. Blodgett DM, Nowosielska A, Afik S, Pechhold S, Cura AJ, Kennedy NJ, Kim S, Kucukural A, Davis RJ, Kent SC, et al. Novel Observations From Next-Generation RNA Sequencing of Highly Purified Human Adult and Fetal Islet Cell Subsets. Diabetes. 2015;64:3172–3181. - PMC - PubMed
    1. Blum B, Hrvatin SSŠ, Schuetz C, Bonal C, Rezania A, Melton DA. Functional beta-cell maturation is marked by an increased glucose threshold and by expression of urocortin 3. Nat Biotechnol. 2012;30:261–264. - PMC - PubMed
    1. Bolger AM, Lohse M, Usadel B. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics. 2014;30:2114–2120. - PMC - PubMed
    1. Bonal C, Herrera PL. Genes controlling pancreas ontogeny. Int J Dev Biol. 2008;52:823–835. - PubMed

Publication types

Substances