Soluble factors secreted by T cells promote β-cell proliferation

Abstract

Type 1 diabetes is characterized by infiltration of pancreatic islets with immune cells, leading to insulin deficiency. Although infiltrating immune cells are traditionally considered to negatively impact β-cells by promoting their death, their contribution to proliferation is not fully understood. Here we report that islets exhibiting insulitis also manifested proliferation of β-cells that positively correlated with the extent of lymphocyte infiltration. Adoptive transfer of diabetogenic CD4(+) and CD8(+) T cells, but not B cells, selectively promoted β-cell proliferation in vivo independent from the effects of blood glucose or circulating insulin or by modulating apoptosis. Complementary to our in vivo approach, coculture of diabetogenic CD4(+) and CD8(+) T cells with NOD.RAG1(-/-) islets in an in vitro transwell system led to a dose-dependent secretion of candidate cytokines/chemokines (interleukin-2 [IL-2], IL-6, IL-10, MIP-1α, and RANTES) that together enhanced β-cell proliferation. These data suggest that soluble factors secreted from T cells are potential therapeutic candidates to enhance β-cell proliferation in efforts to prevent and/or delay the onset of type 1 diabetes.

Figures

Figure 1

Adoptive transfer of diabetes stimulates β-cell proliferation in NOD.RAG1−/− recipients. A: Experimental strategy showing total splenocyte (DM or pre-DM) or depleted splenocyte (diabetic mice) transfer (1 × 107 cells) into NOD.RAG1−/− mice. BrdU (100 mg/kg/BW) was injected 3 weeks post‐transfer, and 6 h later, the pancreases were harvested for immunohistochemical analyses. B: Paraffin-embedded sections of pancreas from mice receiving DM or pre-DM splenocytes, costained with proliferation markers BrdU, pHH3, or Ki67 with insulin and DAPI. Scale bar, 50 µm. Arrows indicate proliferating β-cells (BrdU+/insulin+). Insets show magnified view of representative proliferating β-cells. Quantification of data shown in B for BrdU (C), pHH3 (D), and Ki67 (E) (n = 4–16 mice in each group). *P < 0.05; **P < 0.01; ***P < 0.001 (Student t test). F: TUNEL staining of paraffin-embedded sections of pancreatic tissues obtained from recipient mice receiving DM or pre-DM splenocytes for apoptosis detection. Scale bar, 100 µm. Arrows indicate TUNEL+/β-cell+ cells undergoing apoptosis. Inset shows a magnified representative image of TUNEL+ β-cell. Lower image represents positive control of TUNEL staining in rat tumor tissue. G: Quantification of data in F. n = 4–6 mice in each group. Data are expressed as means ± SEM. FACS, fluorescence-activated cell sorting; Spl, splenocyte.

Figure 2

Confirmation of proliferating β-cells. Two-dimensional (A) and three-dimensional (B) confocal microscopy view of pancreatic section derived from total diabetic NOD mouse splenocyte–injected animal. Dotted line in A represents the border between islet and immune cells. Magnified area highlighted from B shows three-dimensional imaging of insulin (red) (C), Ki67 (green) (D), and DAPI (blue) (E). Scale bar, 100 µm. Arrows indicate proliferating β-cells. F: Orthographic image of the same pancreatic section in A shows the horizontal and vertical view of a proliferating β-cell in a circle (ZEN-2009). G: Mouse spleen section stained as positive control for F4/80 (green), common macrophage marker, and Ki67 (red). H: Total diabetic splenocyte–injected NOD.RAG1−/− pancreatic section stained for F4/80 (green), insulin (blue), and Ki67 (red). Scale bar, 100 µm. Arrow indicates proliferating β-cells. Magnified view of proliferating (I) and nonproliferating (J) macrophages from H. K: Mouse spleen section stained as positive control for CD3 (red), T-cell marker, and Ki67 (green). L: Total diabetic splenocyte–injected NOD.RAG1−/− pancreatic section stained for CD3 (red), insulin (blue), and Ki67 (green). Scale bar, 100 µm. Magnified view of proliferating T cell (M) and β-cell (N) from L. O: Sorting of β- and non–β-cells from dispersed islets from mice after adoptive transfer of pre-DM or DM splenocytes by flow cytometry based on size (FS Lin) and autofluorescence (FL1). P: Sorted β-cells stained for CD45 and BrdU. Dot plot showing gated-out CD45+ cells (Q) and BrdU+ β-cells from pre-DM (R) or DM (S) splenocyte-transferred mice. T: Quantification of data in R and S. n = 3 each group. *Proliferating macrophages or T cells. **P < 0.01. Experiment was performed in triplicate. Data are expressed as means ± SEM. FSC-A, forward scatter detector A; SSC-A, side scatter detector A.

Figure 3

T-cell subsets play a major role in β-cell proliferation. A: Paraffin-embedded sections of pancreatic tissues derived from recipient mice receiving total diabetic or B-cell–depleted diabetic NOD mouse splenocytes, costained for the proliferation marker BrdU (green) with insulin (red) and DAPI (blue). Scale bar, 50 µm. Arrows indicate proliferating β-cells (BrdU+/insulin+). Insets show a magnified representative image of a proliferating β-cell. B: Quantification of proliferating β-cells in pancreatic sections obtained from mice receiving total (DM or pre-DM) or B‐cell–, CD4+-, CD8+-, and CD4+/CD8+–double-depleted diabetic NOD splenocytes. n = 4–6 mice each group. *P < 0.05; ***P < 0.001 (Student t test). Data are expressed as means ± SEM. Spl, splenocyte. (A high-quality color representation of this figure is available in the online issue.)

Figure 4

Pancreatic islet infiltration positively correlates with β-cell proliferation. A: Immunofluorescence (IF) and hematoxylin and eosin (H&E) staining of consecutive pancreatic sections harvested from NOD.RAG1−/− mice 3 weeks after receiving total (DM or pre-DM) or B-cell–, CD4+-, CD8+-, and CD4+/CD8+–double-depleted diabetic NOD splenocytes. Scale bar, 400 µm. Pancreatic islets are outlined with dotted lines for ease of comprehension. B: Pancreatic islets showing insulitis expressed as a percentage in the treated groups in A. C: Linear regression of islet infiltration and BrdU+ β-cells in pancreas sections harvested from NOD.RAG1−/− mice transferred with total diabetic splenocytes. Each square represents a mouse (n = 9) scored for insulitis in at least 20 islets (n = 9). r = 0.713; P = 0.004. D: Pancreatic sections harvested from STZ-induced diabetic NOD mice at day 3 and day 7, costained for the proliferation marker BrdU (green), with insulin (red) and DAPI (blue). Scale bar, 50 µm. Arrows indicate proliferating β-cells (BrdU+/insulin+). Insets show a magnified view of a representative proliferating β-cell. E: Quantification of β-cell proliferation (bars) and insulitis scores (red dots) in the pancreatic islets in STZ-injected NOD mice at 1–7 days post-STZ administration (n = 3 mice for each time point). Data are expressed as means ± SEM. dep, depleted.

Figure 5

β-cell proliferation is stimulated by infiltrating lymphocytes via soluble factors. A: Experimental strategy showing total splenocytes (DM or pre-DM) or depleted splenocytes (diabetic mice), cocultured with 5–6-week-old NOD.RAG1−/− mouse islets (150 islets/condition) for 48 h in 5 mmol/L glucose in the presence or absence of transwell inserts. B: Linear regression of insulitis score and BrdU+ β-cells in single pancreatic islets harvested from NOD.RAG1−/− mice transferred with total diabetic splenocytes. Each square represents a single islet out of 120 analyzed islets. r = 0.80; P < 0.001. C: Representing pancreatic sections derived from total splenocyte– or STZ-injected mice used for determining the ratio between infiltrating immune cells vs. insulin+ β-cells. Islets are indicated by dotted lines in the right panel. The area of infiltration is shown around the islet in the left panels. β-cell proliferation in agar-embedded NOD.RAG1−/− islets cocultured with total splenocytes from DM or pre-DM at a ratio of 1:1, 1:2, 1:5, or 1:10 or without splenocytes in the presence (D) or absence (E) of transwell inserts (n = 3–4). *P < 0.05; **P < 0.01 (Student t test). F: Quantification of proliferating β-cells in pancreatic islets cocultured with total (DM or pre-DM), negatively, or positively selected DM splenocytes at 1:1 or 1:10 ratio with transwell conditions (n = 3–6 for each condition). *P < 0.05; **P < 0.01; ***P < 0.001 (Student t test). Data are expressed as means ± SEM. ND, not detected; O/N, overnight; Spl, splenocyte.

Figure 6

Lymphocyte-secreted soluble factors that drive β-cell proliferation. Luminex assay results from culture medium obtained 48 h after coculturing NOD.RAG1−/− islets with NOD splenocytes (DM or pre-DM) at a ratio of 1:1, 1:2, 1:5, or 1:10 or only splenocytes at 10× in the presence or absence of transwell for IL-2 (A), IL-6 (B), IL-10 (C), MIP-1α (D), and RANTES (E) (n = 3–4 for each condition). *, §P < 0.05; **, §§P < 0.01; ***, §§§P < 0.001 (Student t test). *, diabetic vs. diabetic; §, diabetic vs. prediabetic. Data are expressed as means ± SEM.

Figure 7

Effect of soluble factors on β-cell proliferation is reversed by inhibitory/neutralizing antibody treatment in vitro. A: Detection of the cytokine/chemokine receptor subunit mRNAs by real-time PCR from sorted β-cells and splenocytes harvested from C57BL/6 mice. Tata-box-binding protein (TBP) was used as reference. B–F: Agar-embedded pancreatic islets from C57BL/BJ mouse treated in the absence (control) or presence of low-dose recombinant proteins with or without inhibitory/neutralizing molecules (as described in research design and methods) for 48 h (150 islets/condition, three to four replicates). Representative sections are shown. Islets were costained for the proliferation marker Ki67 (green) with insulin (red) and DAPI (blue). Arrows indicate proliferating β-cells (Ki67+/insulin+). Scale bar, 200 µm. Insets show a magnified image of a representative proliferating β-cell. G: Quantification of data in B–F (n = 3–4 in each group). *, #P < 0.05; **, ##P < 0.01; ***, ###P < 0.001 (Student t test). *, untreated vs. cytokine/chemokine or inhibitory/neutralizing antibody treated; #, cytokine/chemokine treated vs. inhibitory/neutralizing antibody treated. Data are expressed as means ± SEM. CCR1, C-C chemokine receptor type 1; GP130, glycoprotein 130; IL-2R, IL-2 receptor.

Figure 8

Lymphocyte-secreted soluble factors enhance β-cell proliferation in vitro. One hundred fifty agar-embedded islets harvested from C57BL/6 mice cocultured without (untreated control) (A) or with total DM splenocytes in the absence (treated control) (B) or presence of inhibitory/neutralizing molecule Ro 26-4550 (C), anti–IL-6 (D), anti–IL-10 (E), maraviroc 3.3 nmol/L (F), or maraviroc 5.2 nmol/L (G). Islets were costained for the proliferation marker Ki67 (green), insulin (red), and DAPI (blue). Arrows indicate proliferating β-cells (Ki67+/insulin+). Scale bar, 200 µm. Insets show magnified image of a representative proliferating β-cell. H: Quantification of proliferating β-cells in A–G (n = 3–4 in each group). *P < 0.05; **, ##P < 0.01; ***, ###P < 0.001 (Student t test). *, untreated vs. DM splenocyte treated; #, DM splenocyte treated vs. inhibitory/neutralizing antibody treated. Data are expressed as means ± SEM. Ins, insulin; Spl, splenocyte.