AGE-BSA down-regulates endothelial connexin43 gap junctions
© Wang et al; licensee BioMed Central Ltd. 2011
Received: 13 December 2010
Accepted: 16 May 2011
Published: 16 May 2011
Skip to main content
© Wang et al; licensee BioMed Central Ltd. 2011
Received: 13 December 2010
Accepted: 16 May 2011
Published: 16 May 2011
Advanced glycation end products generated in the circulation of diabetic patients were reported to affect the function of vascular wall. We examined the effects of advanced glycation end products-bovine serum albumin (AGE-BSA) on endothelial connexin43 (Cx43) expression and gap-junction communication.
In human aortic endothelial cells (HAEC) treated with a series concentrations of AGE-BSA (0-500 μg/ml) for 24 and 48 hours, Cx43 transcript and Cx43 protein were reduced in a dose dependent manner. In addition, gap-junction communication was reduced. To clarify the mechanisms underlying the down-regulation, MAPKs pathways in HAEC were examined. Both a MEK1 inhibitor (PD98059) and a p38 MAPK inhibitor (SB203580) significantly reversed the reductions of Cx43 mRNA and protein induced by AGE-BSA. Consistently, phosphorylation of ERK and p38 MAPK was enhanced in response to exposure to AGE-BSA. However, all reversions of down-regulated Cx43 by inhibitors did not restore the functional gap-junction communication.
AGE-BSA down-regulated Cx43 expression in HAEC, mainly through reduced Cx43 transcription, and the process involved activation of ERK and p38 MAPK.
Diabetes is known to accelerate the process of atherosclerosis, in which endothelial dysfunction plays a key role . Previous studies have shown that diabetes alters the expression of a variety of molecules involved in maintenance of endothelial function . Animal experiments also demonstrated that induction of diabetes, either alone or on top of hyperlipidemia, suppress the expression of endothelial gap junctions [3, 4]. Gap junctions are cell membrane channels made of paired hexamers of connexins, which allow exchange of ions and small signaling molecules between the cytoplasmic compartments of adjacent cells. In mammals endothelial cells mainly express connexin43 (Cx43), Cx40, and Cx37 , of which Cx43 is by far the predominant in the cultured endothelial cells [6, 7]. Several reports have shown that endothelial Cx43 gap junctions are down-regulated by factors causing endothelial dysfunction , such as aging , hypertension , and arsenic trioxide , the last of which was even reported to induce endothelial lesion. In contrast, other factors causing endothelial dysfunction, such as oscillatory shear stress, were shown to enhance Cx43 expression .
One key offending factor underlying the toxic effects of diabetes is glucose, the high level of which had been proved to affect the activities of endothelial cells in many aspects, including suppression of gap junctions [13, 14]. Since diabetes is associated with a more severe form of vascular disease, we suspected that molecules underlying the detrimental effects of diabetes other than glucose also affect endothelial gap junctions.
During chronic exposure to elevated blood glucose advanced glycation end products (AGE) are generated in the circulation. AGE are also versatile molecules and have been reported to possess multiple actions in the vascular wall, such as changes of release of cytokines, induction of expression of cell adhesion molecules, impairment of endothelial vasodilatation, and triggering of chronic inflammation . However, the effect of AGE on endothelial gap junctions remained unclear. To this end, we examined the expression of Cx43 as well as the gap-junction communication in human aortic endothelial cells (HAEC) treated with AGE and explore underlying mechanisms. A previous study had shown that the average serum AGE levels in diabetic patients ranged between 28.8 and 87.2 μg/ml and may reach 160-500 μg/ml in severe cases .
AGE-BSA was prepared by incubation of BSA (Fraction V, Sigma Chemical Co., St. Louis, MO, USA) at a concentration of 50 mg/ml with 0.5 M glucose in 10 mM phosphate-buffered saline (PBS) containing 0.5 mM EDTA, pH 7.4, at 37°C for 12 weeks. For a control group, the same concentration of BSA was incubated with PBS containing EDTA without glucose for 12 weeks. Free glucose was removed by extensive dialysis against PBS. The brown color of AGE-BSA showed the typical appearance of AGE. SDS-PAGE analysis showed the formation of AGE-BSA monomer and dimer with molecular weights approximately equaling to 78 and 157 kDa, respectively. However, unglycated BSA was about 69 kDa. The ratio of relative fluorescence intensities of AGE-BSA to unglycated BSA was approximately 64 folds in our preparations. Results were consistent with other's report . For a control group, BSA was processed as above procedures. BSA and AGE-BSA were stored at -70°C until use.
HAECs (Cascade Biologics) were seeded in 1% gelatin-coated plasticware and incubated at 37°C under a humidified 95% air and 5% CO2 atmosphere. Cells grown to confluence were dissociated with 3 ml of 0.25% trypsin-EDTA (GIBCO) at 37°C for 3 min. The suspension was diluted with 7 ml medium 200 supplemented with LSGS, centrifuged at 1200 rpm for 8 min, and resuspended in the culture medium. Cells were then replated in 35-mm Petri dishes (5 × 104 cells per cm2) and allowed to grow to confluence or seeded at the same density onto 12-mm glass coverslips coated with 1% gelatin (Sigma). Cells of passage 4 to 6 were used in the subsequent experiments. To evaluate the relative mRNA level of Cx37, Cx40, and Cx43, human umbilical vein endothelial cells (HUVEC, from PromoCell), maintained in endothelial cell growth medium (EGM, from PromoCell) were used as a reference. The confluent cells were treated with AGE-BSA at series concentrations of 25, 50, 100, 250, and 500 μg/ml for 24 and 48 hours. For a control group, cells were treated with BSA at series concentrations of 25, 50, 100, 250, and 500 μg/ml for 24 hours. For examination of activation of MAPK signaling pathways, cells were incubated with AGE-BSA at concentrations of 25, 50, 100, 250, and 500 μg/ml for 2 and 6 hours. For time course experiments, cells were incubated with AGE-BSA at 500 μg/ml for 0, 0.5, 1, 2, 4, and 6 hours. For drug treatments, cells were treated with 40 μM of PD98059 (Sigma), 15 μM of SB203580 (Sigma), and 10 μM of SP600125 (Sigma) for 30 min and then incubated with 500 μg/ml AGE-BSA for 24 hours.
Cells were collected in RIPA buffer containing 150 mM NaCl, 5 mM EDTA, 1% NP40, 2 mM PMSF, and 50 mM Tris-HCl, pH 7.4 or SB buffer containing 20% SDS, 0.1 M Tris-HCl, pH 6.8, and 10 mM EDTA followed by sonication for 30 sec. Thirty microgram of sample was loaded in each lane, resolved by 12% SDS-PAGE, and transferred onto a PVDF membrane (Amersham, UK). The membrane was incubated with the monoclonal anti-Cx43 antibody (1:1000; Chemicon), anti-Cx43 antibody (1:1000; BD Biosciences), anti-β actin antibody (1:2000; Chemicon), anti-ERK antibody (1:1000; Cell signaling), anti-p38 MAPK antibody (1:1000; Biosource), anti-JNK antibody (1:2000; Cell signaling), anti-phosphorylated ERK antibody (1:1000; Cell signaling), anti-phosphorylated p38 MAPK antibody (1:1000; Biosource), and anti-phosphorylated JNK antibody (1:1000; Cell signaling) at room temperature for 1 hour. After three washes with TBST (20 mM Tris pH 7.6, 150 mM NaCl, 0.1% Tween 20), a horseradish peroxidase-conjugated mouse anti-rabbit or donkey anti-mouse IgG (1:3000 in TBST plus 10% BSA) was added, and an enzyme-linked chemiluminescence system (ECL; Amersham, UK) was applied to check the bound antibody.
Total RNA was extracted using Trizol reagent (Invitrogen Life Technologies, Carlsbad, USA) according to the manufacturer's instructions. After phenol treatment and drying, RNA was dissolved in RNase-free water for determination of concentration using spectrophotometer. RNA quality was checked on agarose electrophoresis. For real-time PCR analysis, endothelial cDNA were amplified with primers specific for Cx43 (sense: 5'-CCT CTC GCC TAT GTC TCC TC-3') (antisense: 5'-GCT CAC TTG CTT GCT TGT TG-3'), Cx37 (sense: 5'-CGT AGA GCG TCA GAT GGC-3') (antisense: 5'-GCA CAC TGG CGA CAT AGG-3'), Cx40 (sense: 5'-TCA ATC CCT TCA GCA ATA ATA TG-3') (antisense: 5'-GTG ACC TGG TGA GAC TCC-3'), and β-actin (sense: 5'-TCC TGT GGC ATC CAC GAA ACT-3') (antisense: 5'-GAA GCA TTT GCG GTG GAC GAT-3') using iQTM SYBR Green Supermix reagent and detected with iQTM Single Color Real-Time PCR Detector System (all from Bio-Rad, California, USA). Relative mRNA levels were normalized with the corresponding levels of β-actin.
For Semi-quantitative RT-PCR, total RNA (2 μg) was reverse-transcribed into complementary DNA (cDNA) using the SuperScrip™ II reverse transcriptase and random primers (Invitrogen Life Technologies) according to the manufacturer's recommendations and diluted with DDW to a final volume of 20 μl. The optimal number of PCR cycles was determined for Cx43 (Cx-1: 5'-TCTGAGTGCCTGAACTTGC-3' and Cx-2: 5'-ACTGACAGCCACACCTTCC-3') and GADPH (GA-1: 5'-CATTGACCTCAACTACATGG-3' and GA-2: 5'-TTGCCCACAGCCTTGGCAGC-3') so that the amplification process was conducted during the exponential phase of amplification. Amplification of the GADPH gene transcript was used as the internal control to stringently control for any variability in RNA degradation and RT efficiency. Reactions were carried out in a 50 μl final reaction volume.
Confluent cells grown on coverslips were fixed with methanol at -20°C for 5 min. After blocking with 0.5% BSA, the cells were incubated with the anti-Cx43 antibody (1:1000; Chemicon) at 37°C for 1 hour, followed by incubation with a CY3 conjugated donkey anti-mouse antibody (Chemicon). The cells were then incubated with FITC-conjugated lectin Ulex europaeus agglutinin-1 (UEA-1; 10 μg/ml; Sigma) for 1 hour to confirm the identity of endothelial cells  or bisbenzamide (1 μg/ml; Sigma) for 15 min. Finally, the cells were mounted and examined using a Leica TCS SP confocal laser scanning microscope.
Gap-junction communication in HAECs treated with 250 and 500 μg/ml of AGE-BSA for 24 and 48 hours was evaluated by scrape loading assay. For drug experiments, HAEC were respectively pretreated with i) 40 μM of PD98059 and ii) 15 μM of SB203580 prior to addition of the AGE-BSA (500 μg/ml) for 24 hours. The medium of confluent cells were removed and rinsed with Hank's balanced salt solution (GIBCO). A 27-gauage needle was used to create multiple scrapes through the cell monolayer in the presence of PBS containing 0.5% rhodamine-dextran and 0.5% Lucifer yellow. After 3 min of incubation at room temperature, the culture was rinsed three times and then incubated for 5 min in medium 200 supplemented with LSGS to allow the loaded dye to transfer to adjoining cells. The cells were viewed and recorded using a fluorescence microscope.
Cells were incubated in medium 200 supplemented with LSGS containing 0.5 mg/ml MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide]. After 1 hour, the MTT solution was removed and dimethylsulfoxide (100 μl/well) was added.
Absorbances (550 nm) of the supernatant were read using a microplate spectrophotometer (Spectra Max 340; Molecular Devices, Sunnyvale, CA). Cell viability was expressed as the percentage of the absorbance values of treated cells to controls.
Densitometric scanning and analysis were performed on immunoblots and gel images using Imagemaster (Amersham Pharmacia Biotec, NJ, USA). Within each lane, total amounts of bands of Cx43 were divided by those of actins as loading controls. For RT-PCR, amounts of Cx43 were divided by those of GADPH. For the scrape loading assay, the area between the bilateral edges of lucifer yellow transfer and the scrape line was measured. The value of total amount of each sample was presented as mean (%) ± S.E. The significant differences are analyzed by t-test.
This study demonstrates that, in cultured HAEC, i) Cx43 is the predominant gap junction component protein, unlike HUVEC; ii) AGE-BSA reduced Cx43 transcript expression in a time-dependent manner; and iii) AGE-BSA reduced Cx43 protein expression and attenuated gap-junction intercellular communication in a dose-dependent manner; Although a substantial portion of HAEC died when the cells were treated with high dose of AGE-BSA (500 μg/ml) for more than 24 hours, the down-regulating effects of AGE-BSA on Cx43 expression should not be mainly attributed to a general reduction in protein expression during cell dying, since the expression level of actin protein along the reduction of Cx43 was relatively stable (see Figure 4). In addition, inhibition of gap-junction communication activity may cause cell death, as seen in retinal capillary cells, in which attenuation of gap-junction communication with high glucose treatment disturbed normal transport of small molecules and calcium ions, leading to cell death [14, 19]. The reduction of Cx43 transcripts by AGE-BSA was partially recovered by inhibitors of each of the ERK1 and p38 MAPK. Moreover, the activation of the MAPK cascades was further confirmed by the increase of phosphorylated ERK and p38 after a short time exposure to AGE-BSA indicating that activation of both the ERK and p38 MAPK cascades is involved in this process. Despite the down-regulation of Cx43 expression by AGE-BSA can be abrogated by the inhibitors and the reversed Cx43 mainly stayed at the cell borders, the gap-junction communication was impaired. These novel findings are complementary to our current knowledge regarding the effects of diabetes on endothelial gap junctions.
The results of the present study, in conjunction with those from other in vitro experiments investigating the effect of high glucose on endothelial gap junctions [13, 14], indicate that in the circulation of diabetic individuals high glucose per se as well as the glycated albumin, which is generated in high glucose milieu, affects endothelial gap junction. Interestingly, both the actions of high glucose and the glycated albumin on endothelial gap junctions may share similar mechanisms. In cultured microvasular endothelial cells treated with high glucose, the Cx43 mRNA was reduced . The same effect was observed for the AGE-BSA-induced down-regulation of Cx43 in this study .
Phosphorylation of the members of connexin, including Cx43, is well known to play a key role in regulation of gap-junction communication . The effects of phosphorylation on the communication can be either enhancing or inhibiting. The phosphorylation status of Cx43 is believed to affect the oligomerization of Cx43 into connexons and delivery of integral connexins to justified locations [21–23]. Of the kinases involved in phosphorylation of Cx43, ERK and p38 MAPK were reported to attenuate the gap-junction communication . These data mainly came from non-endothelial cells. On the other hand, activation of the ERK and p38 MAPK cascades was also reported to affect the expression level of Cx43 protein. In cultured smooth muscle cells isolated from human saphenous vein, inhibition of ERK and p38 MAPK by using siRNA specific to the kinases were shown to attenuate the enhanced expression of Cx43 induced by angiotensin II . Considering in the present study that i) a close association between the AGE-BSA-induced reduction in the Cx43 mRNA level and the reduction of Cx43 protein is appeared; ii) the down-regulation of endothelial Cx43 transcript and protein by AGE-BSA is blocked by inhibitors of kinase (ERK and p38 MAPK) cascades; and iii) phosphorylations of ERK and p38 MAPK were activated dose-dependently, one may question whether in endothelial cells after exposure to AGE-BSA there is a link between activation of the MAPK and enhanced proteolysis of Cx43, as reported in lens epithelial cells that phosphorylation is a signal for Cx43 proteins to enter a proteasome-dependent degradation . Moreover, activation of the ERK and p38 MAPK cascades in endothelial cells exposed to AGE-BSA results in regulation of Cx43 protein in a direction opposite to that seen for smooth muscle cells . Recent findings indicated that AGE-HSA (human serum albumin) could promote endothelial progenitor cells (EPCs) senescence and apoptosis via ERK and p38 MAPK pathways, leading to the down-regulation of the number of EPCs [26, 27]. This decreases the protective capacity of EPCs on atherosclerosis in diabetic patients and increases their risks for cardiovascular diseases . Similarly, glucose, AGE, and methylglyoxal, the precursor of AGE, could enhance apoptosis of human cardiac myocytes and collagen deposition, an important event of diabetic cardiomyopathy, in rat cardiac fibroblasts accompanied by temporal activation of ERK1/2, p38 MAPK, and nuclear O-GlcAcylation [13, 29, 30]. Taken together, our findings in the present study suggested that reduced Cx43 transcription is the major mechanism underlying the down-regulation of Cx43 protein in HAEC treated with AGE-BSA. More studies are required to determine whether an increased turnover rate of Cx43 plays a significant role in this process. Although the down-regulation of Cx43 expression was abolished by inhibitors, these rescued Cx43 proteins may already lose their membrane-association and consequently be unable to assemble properly to exert their activity. It is also possible that gap junctions were induced to partially close during the blockage of those signaling pathways. These two phenomenons were already reported in cells under certain physical and chemical offenses [31, 32].
Down-regulation of Cx43 gap junctions in endothelial cells by AGE-BSA has implications in the pathogenesis of diabetic vasculopathy, which, as mentioned before, starts with endothelial dysfunction . Our recent findings indicated that in endothelial cells down-regulation of Cx43 per se activates the cells to a pathological status, in which pro-coagulatory molecules, such as plasminogen activation inhibitor 1 and von Willebrand factor, were up-regulated and the viability, angiogenic capacity, and proliferation of the cells were reduced . These changes of endothelial cells due to insufficient expression of Cx43 may further aggravate the harmful effects of AGE-BSA on the vascular cells.
This study provided the evidence that AGE-BSA down-regulated Cx43 expression in HAEC, mainly through reduced Cx43 transcription, and this process involved activation of ERK and p38 MAPK.
Supported by grants NSC-94-2312-B-020-001 and NSC99-2321-B-020-004-MY3 from the National Science Council, Taiwan and MMH-E 95003 from the Medical Research Department of the Mackay Memorial Hospital, Taiwan.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.