If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
To whom correspondence should be addressed: Dept. of Internal Medicine, Lab. of Molecular Biology 19-135S, Centre Hospitalier Universitaire Vaudois, Rue du Bugnon 46, CHUV-1011 Lausanne, Switzerland. Tel.: 41-21-314-09-26; Fax: 41-21-314-09-68
* This work was supported by grants from the Swiss National Science Foundation (31-68036.02 to J.-A. H., 32-66892.01 to G. W., 31-67023.01 to P. Ma., and 31-67788.02 to P. Me.), the Placide Nicod and Octave Botnar Foundations (to J. A. H.), the Juvenile Diabetes Foundation International (1-2001-555 to G. W., 1-2001-622 to P. Me.), the European Union (QLRJ-2001-01777 and QLG1–1999-00516 to P. Me.), the Fondation Romande pour la Recherche sur le Diabète, and the National Institutes of Health (DK-63443 to P. Me.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Connexin-36 (Cx36) is a gap junction protein expressed by the insulin-producing β-cells. We investigated the contribution of this protein in normal β-cell function by using a viral gene transfer approach to alter Cx36 content in the insulin-producing line of INS-1E cells and rat pancreatic islets. Transcripts for Cx43, Cx45, and Cx36 were detected by reverse transcriptase-PCR in freshly isolated pancreatic islets, whereas only a transcript for Cx36 was detected in INS-1E cells. After infection with a sense viral vector, which induced de novo Cx36 expression in the Cx-defective HeLa cells we used to control the transgene expression, Western blot, immunofluorescence, and freeze-fracture analysis showed a large increase of Cx36 within INS-1E cell membranes. In contrast, after infection with an antisense vector, Cx36 content was decreased by 80%. Glucose-induced insulin release and insulin content were decreased, whether infected INS-1E cells expressed Cx36 levels that were largely higher or lower than those observed in wild-type control cells. In both cases, basal insulin secretion was unaffected. Comparable observations on basal secretion and insulin content were made in freshly isolated rat pancreatic islets. The data indicate that large changes in Cx36 alter insulin content and, at least in INS-1E cells, also affect glucose-induced insulin release.
Gap junction channels provide enclosed conduits for direct exchange of signaling molecules, such as ions and metabolites, between cells. These conduits have pore diameters large enough to ensure passage of metabolites and signaling molecules with a mass as high as 1000 Da (
). Six Cx molecules, arranged so that their third transmembrane domain lines the channel lumen, compose the hemichannels, or connexons, that are contributed by each cell of a coupled pair.
Previous studies have shown that the presence of gap junctions between β-cells is required for proper control of insulin secretion. Thus, single β-cells, which are not coupled by Cx channels, exhibit poor insulin gene expression, release low amounts of the hormone, and barely increase these functions after stimulation (
). To study whether Cx36 may contribute to control insulin secretion and content, we first searched for a line of insulin-producing cells that retains glucose-induced insulin secretion and expresses Cx36. We used the recently developed INS-1E cell line (
). We then modified Cx36 content of INS-1E cells using an adenoviral gene transfer approach. Insulin content and glucose-induced insulin release were decreased in INS-1E cells, which expressed amounts of Cx36 largely higher or lower, respectively, than those observed in native INS-1E cells. These results were confirmed in isolated rat pancreatic islets with regard to insulin content.
MATERIALS AND METHODS
Cell Line and Rat Islet Isolation—Clonal INS-1E cells (
) were cultured in RPMI 1640 containing 11.1 mm glucose and supplemented with 5% fetal calf serum, 2 mml-glutamine, 1 mm sodium pyruvate, 10 mm HEPES, 50 μm β-mercaptoethanol, 110 units/ml penicillin, and 110 μg/ml streptomycin. This medium has long been preferred for cultures of pancreatic islets and insulin-producing cell lines derived thereof, because it preserved, better that other culture media, normal β-cell functions, including insulin content and secretion (
). INS-1E cells were kept at 37 °C in a humidified incubator, gassed with air and CO2 to maintain medium pH at 7.4, fed at 3-day intervals, and passed by trypsinization once per week.
Our institutional review committee for animal experiments approved all of the procedures for rat care, surgery, and euthanasia. Adult rats weighing 250–350 g were anesthetized by inhalation of 5% halothane (Arovet AG), sacrificed, and immediately used for pancreas sampling. Islets of Langerhans were isolated from the pancreas of male Wistar rats by collagenase digestion. After purification on a Ficoll gradient (Sigma), the isolated islets were washed twice in a phosphate-buffered saline (PBS), and the islets were cultured in RPMI 1640 containing 10% fetal calf serum, 10 mm HEPES, 1 mm sodium pyruvate, 2 mm glutamine, 110 units/ml penicillin, 110 μg/ml streptomycin, and 50 μm β-mercaptoethanol. For static experiments, batches of 10 islets were handpicked under a microscope.
RNA Isolation, Reverse Transcriptase (RT)-PCR, and Northern Blot Analysis—INS-1E cells were homogenized in Tripure Isolation Reagent (Roche Applied Science), and total RNA was extracted using the kit procedure. For RT-PCR, total RNA was treated 30 min in the presence of DNase I (DNA-free kit, Ambion, Cambridge, UK). One-μg aliquots of DNase-treated RNA were reverse-transcribed using the ImProm-II Reverse Transcription System (Promega), as described in the kit procedure. One-half of the reverse-transcribed products was used for PCR reaction in the presence of 20 ng of sense and antisense primers, using recombinant Taq DNA polymerase (Invitrogen). The products obtained after 30 PCR cycles (Biometra) were analyzed on an agarose gel. Negative controls included amplification of INS-1E samples that had not been reverse-transcribed. The following primers were used: 5′-CACAGCGATGGGGGAATGGA-3′ (sense) and 5′-TGCCCTTTCACACATAGGCA-3′ (antisense) to detect endogenous Cx36 mRNA, amplified as a 980-bp fragment; 5′-GCTCTAGACTTACTTCAATGGCTGCTCC-3′ (sense) and 5′-CGGAATTCAGGTCTGCTGCTGGCG-3′ (antisense) to detect Cx43 mRNA, amplified as a 321-bp fragment; 5′-CGGGTCTAGAATGAGTTGGAGCTTCCTGAC-3′ (sense) and 5′-CGCATGGCATAGGGTTTGCTC-3′ (antisense) to detect Cx45 mRNA, amplified as a 358-bp fragment; 5′-AGTCAGTGCTTCTGACACA-3′ (sense primer facing the intron sequence of viral DNA) and 5′-CTATATGGCTTCAGTGTCC-3′ (antisense primer in the rat Cx36 sequence) to detect adenoviral Cx36 mRNA of Ad-asCx36, amplified as a 550-bp fragment; and 5′-CGGAATTCCTGGTTCTGTCTCCTTACTGG-3′ (antisense primer for Cx36) and 5′-AGTCAGTGCTTCTGACACA-3′ to detect Ad-sCx36, amplified as a 730-bp fragment. As positive control, samples of DNA prepared by overnight digestion of purified viral particles were also amplified by PCR. The amplified DNA was visualized on agarose gels after ethidium bromide staining.
Generation of Recombinant Adenoviruses and Cell Infection—To modulate Cx36, we generated recombinant adenoviruses (
) in either a sense (Ad-sCx36) or antisense orientation (Ad-asCx36). The recombinant adenoviruses allowed for expression of either Cx36, green fluorescent protein (GFP), or Cx36 antisense RNA under control of the immediate, early cytomegalovirus promoter. The cDNAs were inserted into plasmid pXC15 (
), and the adenoviruses were then generated by cotransfecting 293 cells with this plasmid and plasmid pJM17, using a calcium phosphate procedure. Viruses were plaque-purified three times on HER911 cells (IntroGene, Leiden, The Netherlands). Stocks were further purified by two successive CsCl centrifugations, after which the virus band (1.5 ml) was collected and dialyzed at 4 °C against three changes (at least 200 volumes each) of 10 mm Tris-HCl buffer, pH 8.0, in a Slide-A-Lyzer (0.5–3.0-ml capacity), γ-irradiated 10K dialysis cassette (Pierce).
INS-1E cells were seeded at an initial density of 8 × 105 cells/well into 6-well dish plates (Falcon) and cultured for 48 h, at which time they were 80% confluent. For infection, cells were incubated for 60 min with different dilutions (multiplicity of infection, MOI = 5, 10, and 25) of a stock of the recombinant adenoviruses Ad-asCx36, Ad-sCx36, or Ad-GFP and then cultured in RPMI 1640 for 24 h before any experiment. Batches of 150–200 isolated rat pancreatic islets were also infected with either Ad-asCx36, Ad-sCx36, or Ad-GFP, according to the same protocol. Batches of 10 handpicked islets were used for secretion experiments.
Immunofluorescence—Peptides corresponding to amino acids 289–303 (peptide B: AKRKSVYEIRNKDLP) (
), 155–168 (peptide C: QNTETTSKETEPDC), and 114–127 of Cx36 (peptide D: LALDRDPAESIGGP) were conjugated to keyhole limpet hemocyanin and used to immunize rabbits (Eurogentec, Herstal, Belgium). Site-directed antibodies were affinity-purified by passing the sera diluted in PBS through HiTrap affinity columns (HiTrap NHS-activated Sepharose; Amersham Pharmacia Biotech AB) that had been coupled with the respective COOH-terminal tail peptides. The columns were washed sequentially with PBS, 0.5% PBS-Tween 20, and PBS. Antibody solutions were eluted with 100 mm glycine, pH 2.5, in a 1.5 m Tris-HCl buffer, pH 8.8, to reach neutrality.
For immunofluorescence labeling, cells were grown on glass coverslips, fixed for 3 min in –20 °C acetone, air-dried, rinsed in PBS, and incubated in a buffer containing 2% bovine serum albumin and 0.1% Triton X-100. Cells were then exposed for 20 h at room temperature in the presence of different rabbit polyclonal antibodies against rat Cx36, which were either mixed at a 1:1:1 ratio to increase the signal or which were used separately. In all cases, the sera were diluted 1:20. After repeated washing, cells were incubated a second time for 1 h at room temperature, using fluorescein-conjugated antibodies against rabbit immunoglobulins, diluted 1:250. After additional rinsing, cells were counterstained with Evans blue, coverslipped, and viewed on a microscope (Leica) fitted with filters for fluorescein detection. In these experiments, controls included use of wild-type (negative control) and Cx36-transfected positive control HeLa cells (a gift from Dr. K. Willecke).
Western Blotting—For preparation of total extracts, INS-1E cells and rat pancreatic islets were lysed by sonication in a buffer containing 5% SDS and 5 mm EDTA. For preparation of membrane proteins, INS-1E cells were homogenized in a 100 mm Tris-HCl buffer, pH 7.4, which was supplemented with 20 mm EDTA, 1 mg/ml pepstatin A, 1 mg/ml anti-pain (Merck), 1 mm benzamidine (Merck), 40 KIU/ml aprotinin (Bayer), 2 mm phenylmethylsulfonyl fluoride (Sigma), and 1 mm diisopropyl fluorophosphate (Aldrich). The homogenates were centrifuged at 3000 × g for 10 min. Supernatants were collected and centrifuged for 60 min at 100,000 × g and 4 °C. The membrane pellet was resuspended in 62.5 mm Tris-HCl buffer, pH 8.0, supplemented with 20% SDS, and 10 mm EDTA. Protein content was determined by the Bio-Rad DC protein assay reagent kit (Bio-Rad).
Aliquots of islets and cells were heated at 50 °C in loading buffer, fractionated by electrophoresis in a 12.5% polyacrylamide gel, and immunoblotted overnight onto Immobilon polyvinylidene difluoride membranes (Millipore) at a constant voltage of 20 V. Membranes were incubated for 1 h at room temperature in PBS containing 5% milk and 0.1% Tween 20 (blocking buffer) and then incubated overnight at 4 °C with an antiserum against either amino acids 289–303 (peptide B) (
) of rat Cx36, diluted 1:200, or tubulin (Sigma), diluted 1:2000 in blocking buffer. Antigen-antibody complexes were detected with the horseradish peroxidase Western blot detection system (Amersham Biosciences). In these experiments, controls included use of wild-type HeLa (negative control) and Cx36-transfected, positive control HeLa cells.
Electron Microscopy—Wild-type and infected INS-1E cells were fixed for 60 min at room temperature in 2.5% glutaraldehyde in a 0.1 m phosphate buffer (pH 7.4). For freeze-fracture, samples were infiltrated for 60 min in 30% phosphate-buffered glycerol and frozen in Freon 22 that had been cooled with liquid nitrogen. Cells were fractured and shadowed in a Balzers BAF 301 apparatus (High Vacuum Corp., Balzers, Lichtenstein). The replicas were washed in a sodium hypochlorite solution, rinsed in distilled water, mounted on Formvar- and carbon-coated copper grids, and examined in a Philips CM 10 (Philips Electron Optics, Mahwah, NJ) electron microscope.
Insulin Secretion—Wild-type and infected INS-1E cells were plated into 6-well plates at a density of 8 × 105/well and tested 2 days later at a confluence of about 80%. To this end, cells were maintained at 37 °C for 2 h in glucose-free RPMI 1640, washed, and preincubated in a glucose-free (0 mm) Krebs-Ringer-bicarbonate medium (pH 7.4) containing (in mm): 135 NaCl, 3.6 KCl, 5 NaHCO3, 0.5 NaH2PO4, 0.5 MgCl2, 1.5 CaCl2, 0.1% bovine serum albumin, and 10 HEPES (KRBH buffer). After a 30-min preincubation, cells were incubated for 30 min at 37 °C in the presence of either a basal (2.8 mm) or a stimulatory (16.7 mm) concentration of glucose. At the end of each incubation period, medium was collected and centrifuged for 10 min at 2,500 × g and 4 °C to remove cell debris. To measure protein and insulin content, INS-1E cells were extracted within the wells with 1 ml of 75% ethanol/0.2 m HCl. Supernatant and cell samples were immediately stored at –20 °C until insulin determination by radioimmunoassay, using rat insulin as standard, and as per the instructions of the kit (Linco Research, St. Charles, MO). Total protein content was measured by the DC protein assay kit (Bio-Rad). Insulin secretion was expressed as a percentage of cell insulin or protein content. Batches of 10 isolated islets were also tested for insulin secretion and content as described above.
Statistical Analysis—Data were expressed as mean ± S.E. Differences between means were assessed by Student's t test and/or analysis of variance. Statistical significance was defined at a value of: *, p < 0.05; **, p < 0.01; and ***, p < 0.001.
Adenoviruses Induce de Novo Cx36 Expression in Connexin-defective HeLa Cells—By immunofluorescence, we did not detect Cx36 in wild-type HeLa cells (data not shown), confirming that these cells are Cx-deficient (
). In contrast, the protein was detected in Cx36-transfected HeLa cells using the different Cx36 antibodies we tested in this study (Fig. 1A). After infection with Ad-sCx36, the same antibodies immunolabeled Cx36 at sites of membrane contacts between HeLa cells (Fig. 1E). The amount of staining increased with the MOI used in each infection experiment but, in all cases, had a pattern like that observed in Cx36-transfected cells (Fig. 1).
Expression of Cx36 and Gap Junctions Are Regulated by Gene Transfer in INS-1E Cells—RNAs coding for Cx43, Cx45, and Cx36 were detected by RT-PCR in freshly isolated pancreatic islets, whereas only a transcript for Cx36 was detectable in INS-1E cells (Fig. 2A). RT-PCR of RNA extracted from cells infected with recombinant adenoviruses showed the transcription of either a sense or an antisense Cx36 RNA 24 h after the infection of cells with Ad-sCx36 and Ad-asCx36, respectively (Fig. 2B). The sense Cx36 RNA and the cognate protein were already detectable 3 h after the infection, and their levels of expression increased during the first 24 h (Fig. 2, C and D). Cx36 was barely detectable in wild-type, non-infected INS-1E cells (Fig. 1B), and the levels of the protein were decreased after infection with two different doses (MOI = 10 and 25) of Ad-asCx36 (Fig. 1, C and D). In contrast, INS-1E cells infected with similar doses of Ad-sCx36 showed a drastic increase in Cx36 expression (Fig. 1, F–H).
Western blots of membrane extracts confirmed the absence of Cx36 in wild-type HeLa cells but immunolocated a protein of 36 kDa in HeLa cells transduced with three different doses (MOI = 5, 10, and 25) of Ad-sCx36 (Fig. 3A). The levels of this protein increased with the adenoviral dosage (Fig. 3A). Large levels of Cx36 were also observed in INS-1E cells infected with Ad-sCx36, contrasting with the low levels detected in both wild-type cells and cells infected with 10 MOI of the Ad-GFP vector (Fig. 3A). Western blots of total proteins (Fig. 3B) also showed a dose-dependent reduction of Cx36 content in INS-1E cells infected with 5, 10, or 25 MOI of Ad-asCx36 vector and, conversely, a dose-dependent increase in Cx36 in cells infected with the Ad-sCx36 vector (Fig. 3B). Quantitative assessment demonstrated that, relative to the levels of tubulin, those of Cx36 were reduced by 80% in cells infected with the Ad-asCx36 vector and increased 14-fold in cells infected with the Ad-sCx36 vector (Fig. 3C). Freeze-fracture electron microscopy confirmed that cells overexpressing Cx36 formed typical gap junction plaques that were not observed in wild-type, non-infected INS-1E cells (Fig. 4). Quantitative assessment showed that the number of membranes containing junctions increased 3–4-fold in cells infected with Ad-sCx36 compared with non-infected cells, whereas the number of gap junctions increased 6-fold (Table I).
Table IGap junction plaques increase between INS-1E cells after infection by Ad-sCx36
No. of cells scored
No. of gap junctions (plaques/linear)
No. of membranes with junctions/gap junctions only
Alterations in Cx36 Impair Insulin Content and Secretion of INS-1E Cells—No difference in protein content was observed between wild-type INS-1E cells and cells infected with the different adenoviral constructs (Fig. 5, A and E). In contrast, insulin content was reduced by 15–35% in Cx36-transduced (antisense and sense) cells, at the end of a 30-min incubation in the presence of either 2.8 mm (p < 0.05) or 16.7 mm (p < 0.01) glucose. This decrease was not observed in cells infected with Ad-GFP (Fig. 5, B and F).
Wild-type, non-infected cells and cells infected with Ad-GFP showed a 3–4-fold increase (p < 0.001) in insulin secretion when the glucose concentration of the incubation medium was raised from 2.8 to 16.7 mm (Fig. 5, C and G). Compared with these controls, cells infected with Ad-asCx36 showed a significant reduction in glucose-induced secretion, which depended on the viral dose. Thus, after exposure to 16.7 mm, Ad-asCx36-infected cells failed to significantly increase insulin secretion over the basal levels measured in the presence of 2.8 mm glucose (Fig. 5C). Overexpression of Cx36 in INS-1E cells infected with Ad-sCx36 also resulted in a 40–50% reduction in glucose-induced insulin secretion (Fig. 5G). When expressed as percentage of insulin content, the glucose (16.7 mm)-induced insulin secretion of INS-1E cells infected with either Ad-asCx36 or Ad-sCx36 was reduced by about 30% (Fig. 5, D and H).
Alterations in Cx36 Expression Affect Insulin Content of Rat Pancreatic Islets—Western blots of extracts of rat pancreatic islets revealed that, compared with the levels of tubulin, those of Cx36 were reduced up to 50% in islets infected with Ad-asCx36 (MOI = 20/islet). Conversely, these levels were increased 3–4-fold (p < 0.001) in islets infected with Ad-sCx36 (Fig. 6). Compared with the secretion observed in the presence of 2.8 mm glucose, control non-infected islets and islets infected with Ad-GFP showed a 2-fold increase (p < 0.001) in insulin release when exposed to 16.7 mm glucose (Fig. 7A). In contrast, islets infected with either Ad-asCx36 or Ad-sCx36 failed to exhibit glucose-induced insulin release (Fig. 7A). In the presence of 16.7 mm glucose, their secretion was 40–60% of the control value, depending on the adenovirus dose used (Ad-asCx36 or Ad-sCx36, Fig. 7A). However, when expressed as a percentage of insulin content, the glucose (16.7 mm)-induced insulin secretion of infected islets was similar to that of controls, whether an Ad-asCx36 or Ad-sCx36 vector had been used (Fig. 7B). In contrast, insulin content was significantly decreased in infected islets, representing 25 and 60% of control levels in islets infected with Ad-asCx36 and Ad-sCx36, respectively (Fig. 7C).
Previous studies on primary pancreatic islet cells and tumoral insulin-producing cells have indicated a relationship between Cx-dependent communication and secretory function, particularly with regard to the ability of cells to control insulin release in response to glucose (
) to transduce a Cx36 antisense cDNA, we have first reduced the expression of Cx36 and observed a marked decrease of insulin content in both INS-1E cells and native rat islets. This effect was not accounted for by a major decrease in the transcription of the insulin gene (data not shown) nor to the viral load itself, inasmuch as adenovirus coding for GFP had no functional consequence. These data indicate that large, although not complete, loss of the native levels of Cx36 is detrimental to proper function of tumoral as well as primary β-cells, particularly with regard to one of their main physiological functions, which is to synthesize and store insulin.
The reduced expression of Cx36 was also associated with a marked decrease in the glucose-induced insulin release of INS-1 cells, suggesting a further implication of the Cx in the multistep pathway that couples a glucose stimulation to insulin release. However, this effect was not clearly seen in isolated islets, the secretion of which was barely affected after infection with Ad-asCx36 vectors. In view of previous studies, which have documented a limited efficiency of adenoviral vectors for the infection of intact islets (
), it is possible that a larger effect may have been masked in these studies by a sizable proportion of β-cells that were not effectively transduced. Indeed, alterations in glucose-induced insulin release of primary β-cells have been reported recently in another model after the expression of Cx36 was sizably increased (
We have further demonstrated that INS-1E cells, which after transduction expressed amounts of Cx36 much higher than those of normal β-cells, also featured reduced insulin secretion. Thus, Cx36 shares the property of other, Cx-unrelated proteins, the overexpression of which in β-cells blunts glucose-stimulated insulin secretion (
). It remains to be elucidated by which mechanism Cx36 functions as a negative regulator of insulin biosynthesis and release and, hence, whether the Cx may affect the same steps of the secretory pathway that are targeted by either uncoupling proteins (
Tight control of insulin secretion is critical for glucose homeostasis. This control is exerted in part by circulating nutrients and particularly glucose. The metabolism of this sugar causes a rise in the ratio of ATP/ADP, which induces the closure of ATP-sensitive K+ channels. The resulting decrease in K+ conductance leads to membrane depolarization and to a subsequent opening of voltage-dependent Ca2+ channels, which ultimately triggers the exocytosis of insulin granules (
). Thus, it is plausible that a large overand underexpression of Cx36 may alter the synchronization of β-cells that is required for a normal control of insulin biosynthesis and release.
In conclusion, the data provide novel evidence for an implication of the gap junction protein Cx36 in the control of insulin content and secretion in both tumoral and primary β-cells. Specifically, the data demonstrate that proper storage of insulin requires that the levels of Cx36 be maintained within a 10-fold range of native values. They further suggest that larger changes in this protein may also alter the regulation of insulin release. In view of these findings, it will be important to assess whether Cx36 participates in the loss of glucose sensitivity, which β-cells exhibit in type 2 diabetes.
We are grateful to Thomas Tawadros for useful discussions.