Skip to main page content
U.S. flag

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
. 2021 Nov 2;37(5):109919.
doi: 10.1016/j.celrep.2021.109919.

Single-cell analysis of the human pancreas in type 2 diabetes using multi-spectral imaging mass cytometry

Affiliations

Single-cell analysis of the human pancreas in type 2 diabetes using multi-spectral imaging mass cytometry

Minghui Wu et al. Cell Rep. .

Abstract

Type 2 diabetes mellitus (T2D) is a chronic age-related disorder characterized by hyperglycemia due to the failure of pancreatic beta cells to compensate for increased insulin demand. Despite decades of research, the pathogenic mechanisms underlying T2D remain poorly defined. Here, we use imaging mass cytometry (IMC) with a panel of 34 antibodies to simultaneously quantify markers of pancreatic exocrine, islet, and immune cells and stromal components. We analyze over 2 million cells from 16 pancreata obtained from donors with T2D and 13 pancreata from age-similar non-diabetic controls. In the T2D pancreata, we observe significant alterations in islet architecture, endocrine cell composition, and immune cell constituents. Thus, both HLA-DR-positive CD8 T cells and macrophages are enriched intra-islet in the T2D pancreas. These efforts demonstrate the utility of IMC for investigating complex events at the cellular level in order to provide insights into the pathophysiology of T2D.

Keywords: histopathology; human pancreas; imaging mass cytometry; immunolabeling; multiplexed imaging; spatial information; type 2 diabetes.

PubMed Disclaimer

Conflict of interest statement

Declaration of interests The authors declare no competing interests.

Figures

Figure 1.
Figure 1.. Overview of the imaging mass cytometry study of the human pancreas
(A) Imaging mass cytometry workflow. (B and C) Representative multiple-channel image overlays of non-diabetic (ND) (B) and type 2 diabetic (T2D) (C) pancreas. From left to right, displayed channels are as follows: on the left image, C-peptide (green), glucagon (red), somatostatin (blue), PP (white), and Ghrelin (cyan); in the middle image, collagen (green), CD31 (red) and CD99 (blue); on the right image, CD68 (green), CD3 (red), and CD99 (blue). (D) Heatmap of mean expression values of each protein in each cell type. Input data are mean level of each protein in each cell type. Color indicates Z- score. The x-axis labels are the proteins, and the yaxis label are the cell types. (E and F) The proportion of each cell type in ND (E) and T2D (F). See also Tables S1 and S2 and Figure S1.
Figure 2.
Figure 2.. Altered alpha and beta cell as well as macrophage distribution in the T2D pancreas
(A) Boxplots showing the density of beta cells in the human pancreas, either for the whole organ or separately for the head, body, or tail. Each dot represents the mean density of each donor. The black line inside of the boxplot indicates the median density, and the whisker shows ±1.53 interquartile range. ND, n = 13; T2D, n = 16. *p < 0.05, **p < 0.01. Mann-Whitney U test. (B) Boxplots showing the density of alpha cells in the human pancreas, analogous to (A) *p < 0.05, **p < 0.01. Mann-Whitney U test. (C) Boxplots of the alpha cell/beta cell ratio in the ND and T2D pancreas, either for the whole organ or separately for the head, body, or tail. Each dot represents the mean ratio of each donor. The black line inside of the boxplot indicates the median ratio, and the whisker shows ±1.53 interquartile range. ND, n = 13; T2D, n = 16. **p < 0.05, **p < 0.01, ***p < 0.001. Mann-Whitney U test. (D) Beta cell density is inversely proportional to type 1 collagen deposition in the T2D pancreas (n = 16). Islet collagen is measured by the percentage of expanded islet area (islet region + within 50 mm from islet boundary) with positive collagen signal. Statistical significance was tested by linear regression t test. Each dot demonstrates one donor. (E) Representative image of CD68+ macrophages with or without HLA-DR expression in the pancreas from ND (above) and T2D (below) organ donor. Islets outlined by CD99 (blue). (F) Boxplots showing the total macrophage density intra-islet, peri-islet, or in the exocrine pancreas. Data are presented for the whole pancreas and separately for head, body, and tail of the organ. Each dot represents the mean density of each donor. The black line inside of the boxplot indicates the median density, and the whisker shows ±1.53× interquartile range. ND, n = 13; T2D, n = 16. *p < 0.05, **p < 0.01. Mann-Whitney U test. (G) Boxplots showing HLA-DRhigh macrophage density intra-islet, peri-islet, or in the exocrine pancreas. Data are presented for the whole pancreas and separately for head, body, and tail of the organ. Each dot represents the mean density of each donor. The black line inside of the boxplot indicates the median density, and the whisker shows ±1.53 interquartile range. ND, n = 13; T2D, n = 16. **p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Mann-Whitney U test. Figures S2 and S3.
Figure 3.
Figure 3.. Intra-islet CD8 T and CD8/HLA-DRhigh T cell density are increased in T2D
(A) CD8 T cell and HLA class II expression in the pancreas of ND (left) and T2D (right) organ donors. CD68 (green), HLA-DR (red), collagen (blue), and CD99 (white). The zoomed-in image is 1.5-fold of the original image. (B) Boxplots showing total CD8 T cell density intra-islet, peri-islet, or in the exocrine pancreas. Data are presented for the whole pancreas and separately for head,body, and tail of the organ. Each dot represents the mean density of each donor. The black line inside of the boxplot indicates the median density, and the whisker shows ±1.5× interquartile range. ND, n = 13; T2D, n = 16. *p < 0.05. Mann-Whitney U test. (C) Boxplots showing CD8/HLA-DRhigh CD8 T cell density intra-islet, peri-islet, or in the exocrine pancreas. Data are presented for the whole pancreas and separately for head, body, and tail of the organ. Each dot represents the mean density of each donor. The black line inside of the boxplot indicates the median density, and the whisker shows ± 1.5× interquartile range. ND, n = 13; T2D, n = 16. *p < 0.05, **p < 0.01. Mann-Whitney U test. See also Figure S5.
Figure 4.
Figure 4.. Neighborhood analysis
(A) Cell type annotation of the 14 major cell types determined after image segmentation was projected back to assemble the tissue map shown on the right. Note the close match to the original IMC image shown on the left, confirming the cell type annotation pipeline. The zoomed-in image is 2-fold of the original image. (B) Principle of neighborhood analysis. For any cell “A,” termed query cell, the cell types of the adjacent cells (neighboring cells, “B”) were counted and summarized as a frequency matrix. Enrichment scores were calculated by comparing the observed frequency to the expected frequency obtained from randomizing cell labels for 100 rounds. Enrichment scores were then compared for statistical significance between control (n = 13) and T2D (n = 16) by permutation test. (C) Heatmap displaying cell-cell interaction frequencies between control and donors with T2D. Statically significant results with Benjamini-Hochberg adjusted a p value of <0.05, and differences in enrichment scores of >0.1 are outlined by black boxes. Teal color indicates interactions that are more frequent in ND, and red indicates those that are more abundant in the patients with T2D. The red box highlights the finding that the presence of CD8 T cells near beta cells is more likely in the T2D pancreas, whereas the blue box emphasizes the point that macrophage/beta cell interactions are more likely in T2D than in the ND pancreas. See also Figure S4.

References

    1. Ahlqvist E, Prasad RB, and Groop L (2020). Subtypes of Type 2 Diabetes Determined From Clinical Parameters. Diabetes 69, 2086–2093. - PubMed
    1. Atkinson MA (2014). Pancreatic biopsies in type 1 diabetes: revisiting the myth of Pandora’s box. Diabetologia 57, 656–659. - PMC - PubMed
    1. Atkinson MA, Campbell-Thompson M, Kusmartseva I, and Kaestner KH (2020). Organisation of the human pancreas in health and in diabetes. Diabetologia 63, 1966–1973. - PMC - PubMed
    1. Berg S, Kutra D, Kroeger T, Straehle CN, Kausler BX, Haubold C, Schiegg M, Ales J, Beier T, Rudy M, et al. (2019). ilastik: interactive machine learning for (bio)image analysis. Nat. Methods 16, 1226–1232. - PubMed
    1. Bilekova S, Sachs S, and Lickert H (2020). Pharmacological Targeting of Endoplasmic Reticulum Stress in Pancreatic Beta Cells. Trends Pharmacol. Sci 42, 85–95, . - PubMed

Publication types

MeSH terms