Cx36 makes channels coupling human pancreatic beta-cells, and correlates with insulin expression

Abstract

Previous studies have documented that the insulin-producing beta-cells of laboratory rodents are coupled by gap junction channels made solely of the connexin36 (Cx36) protein, and have shown that loss of this protein desynchronizes beta-cells, leading to secretory defects reminiscent of those observed in type 2 diabetes. Since human islets differ in several respects from those of laboratory rodents, we have now screened human pancreas, and islets isolated thereof, for expression of a variety of connexin genes, tested whether the cognate proteins form functional channels for islet cell exchanges, and assessed whether this expression changes with beta-cell function in islets of control and type 2 diabetics. Here, we show that (i) different connexin isoforms are differentially distributed in the exocrine and endocrine parts of the human pancreas; (ii) human islets express at the transcript level different connexin isoforms; (iii) the membrane of beta-cells harbors detectable levels of gap junctions made of Cx36; (iv) this protein is concentrated in lipid raft domains of the beta-cell membrane where it forms gap junctions; (v) Cx36 channels allow for the preferential exchange of cationic molecules between human beta-cells; (vi) the levels of Cx36 mRNA correlated with the expression of the insulin gene in the islets of both control and type 2 diabetics. The data show that Cx36 is a native protein of human pancreatic islets, which mediates the coupling of the insulin-producing beta-cells, and contributes to control beta-cell function by modulating gene expression.

Figure 1.

Different connexin transcripts are expressed in the adult human pancreas and its endocrine islets. Reverse transcriptase-polymerase chain reaction revealed the expression of Cx26, Cx31.9, Cx32, Cx36, Cx37, Cx43 and Cx45 mRNAs in intact adult human pancreas, and of Cx30.3, Cx31, Cx31.1, Cx31.9, Cx36, Cx37, Cx43 and Cx45 mRNAs in the endocrine islets, which were isolated from the gland. In all the experiments, samples of water and of human liver, brain, skin and heart served as negative and positive controls, respectively. Amplification of the cyclophillin mRNA provided an internal standard.

Figure 2.

The major pancreatic connexins are differentially expressed in the exocrine and endocrine portions of the human gland (A–C). Western blots detected Cx26 (A) in extracts of total pancreas but not of purified islets. A faint 32 kDa band, corresponding to Cx32, was also detected in total pancreas, but not isolated islets (B). Cx43 (C) were found in both total pancreas and islet extracts. (D) In contrast, antibodies detected Cx36 in extracts of purified islets but not total pancreas. More Cx36 was detected in highly purified (90%) than in poorly purified islets (70%). Twenty micrograms of protein were loaded per lane. Immunolabeling of α-smooth muscle actin served as loading reference. Extracts of human liver and heart, as well as of the insulin-producing MIN6 cells served as positive and negative controls, depending on the connexin isoform.

Figure 3.

Cx36 is expressed by pancreatic β-cells. (A) Reverse transcriptase-polymerase chain reaction (RT–PCR) amplification of total RNA showed expression of Cx36 in FACS-purified and green fluorescence protein (GFP)-expressing human β-cells, and of both Cx36 and Cx43 in the GFP-negative non-β-cells. Neither fraction expressed Cx26, Cx32 and Cx45. Because of the high sensitivity of the non-quantitative RT–PCR approach, some insulin transcript was detected in both cell fractions, owing to the unavoidable contamination of the non-β-cell fraction by a limited number of small size β-cells. (B) Immunoperoxidase staining of a section of a control human pancreas showed that several cells containing abundant insulin (left panel) also expressed significantly lower levels of Cx36, whereas islet cells lacking detectable immunolabeling for insulin (solid arrows) as well as cells of islet vessels (open arrow) lacked Cx36 (middle panel). No islet cell was stained after incubation with non-immune IgGs (right panel). Bar, 50 µm. (C) Immunofluorescence showed that most Cx36 was distributed as small, discrete spots (green; middle panel) all along the membrane of the insulin-containing β-cells (red; left panel). The right panel is the merge of the left and middle panels showing the location of Cx36 in β-cells (yellow). Bar, 30 µm.

Figure 4.

β-Cells form Cx36 gap junctions in lipid rafts of their membrane. (A) Western blots detected Cx36 in extracts of both purified islet cell membrane (m) and total isolated islets (T), but not in extracts of total pancreas. Forty micrograms of protein were loaded per lane. Immunolabeling of α-smooth muscle actin served as loading reference. Extracts of wild type (RIN2A) and Cx36-transfected RIN2A insulin-producing cells (RIN36) served as negative and positive controls, respectively. (B) Discontinuous sucrose density gradient centrifugation of a Triton X-100 lysate of human islets showed that most Cx36 (middle panel) was recovered in fractions of 4–7, which were enriched in lipid rafts as judged by the presence of the GM1 ganglioside (top panel). The bottom panel shows the western blot with an antibody against Cx36. Fraction 2 = 35% sucrose; fraction 8 = 5% sucrose. (C) Freeze-fracture electron microscopy showed that β-cell membranes contained typical gap junction plaques (arrow heads) made of connexin oligomers, that in most cases were found tightly associated to tight junctions (arrows). The presence of connexin particles on the P fracture face of one membrane (left panel) and of corresponding pits on the E fracture face of the adjacent membrane (right panel) demonstrates the trans-membrane disposition of the connexin channels. Bar, 120 nm.

Figure 5.

Human β-cells are coupled by Cx36 channels. (A) Microinjection of Lucifer Yellow into individual cells of freshly isolated human islets resulted in the diffusion of the tracer from the injected cell into some of its neighbors (yellow; left panel). A much larger intercellular diffusion was observed when another β-cell of the very same islet was microinjected with ethidium bromide (red; right panel) Bar, 25 µm. (B) Quantitative analysis showed that the average extent of β-cell coupling revealed by the small and positively charged ethidium bromide (solid bar) was three- to four-fold larger than that revealed by the larger size and negatively charged Lucifer Yellow (open bar). ***P < 0.001 versus ethidium bromide diffusion. Values within columns are number of microinjections.

Figure 6.

The expression of Cx36 in human β-cells parallels that of the insulin gene. Quantitative PCR amplification of total RNA extracted from isolated islets of normoglycemic (open symbols) and type 2 diabetics (solid symbols) revealed a quantitatively variable expression of Cx36 in all samples. The expression levels of the connexin were significantly correlated (r² = 0.834, P < 0.001, n = 21) with those of the insulin gene. Cx36 and insulin mRNAs were expressed relative to the levels of cyclophillin.