HIV gp120 induces endothelial dysfunction in tumour necrosis factor-alpha-activated porcine and human endothelial cells

Abstract

Aims: The aim of this study was to determine direct effects and potential molecular mechanisms of HIV gp120, a viral envelope glycoprotein, on endothelial function.

Methods and results: Fresh porcine coronary artery rings and human coronary artery endothelial cells (HCAECs) were treated with recombinant HIV gp120 for 16 h with or without pretreatment with tumour necrosis factor-alpha (TNF-alpha) (8 h). With a myograph tension analysis, HIV gp120 with TNF-alpha pretreatment significantly decreased endothelium-dependent vasorelaxation in response to bradykinin in porcine coronary artery rings compared with untreated control vessels. In addition, HIV gp120 with TNF-alpha pretreatment significantly reduced endothelial nitric oxide synthase (eNOS) expression-both mRNA and protein levels-in porcine coronary artery rings and HCAECs compared with untreated controls. Furthermore, TNF-alpha pretreatment substantially increased intercellular adhesion molecule-1 (ICAM-1) expression in artery rings and HCAECs. Anti-gp120 or anti-ICAM-1 antibody significantly blocked these effects of HIV gp120. Silencing of ICAM-1 by siRNA oligonucleotides significantly blocked the effect of gp120 on eNOS downregulation in TNF-alpha-pretreated HCAECs.

Conclusion: HIV gp120 and TNF-alpha synergistically reduce eNOS expression and cause endothelial dysfunction in both porcine coronary arteries and HCAECs. ICAM-1 induced by TNF-alpha pretreatment may mediate HIV gp120-induced endothelial dysfunction, which suggests a novel molecular mechanism of HIV gp120-ICAM-1 interaction inducing endothelial dysfunction.

Figure 1

Effects of HIV gp120 and TNF-α on the vasomotor function and eNOS expression in porcine coronary arteries. Endothelium-intact porcine coronary artery rings were treated with TNF-α for 8 h and/or gp120 for 16 h. (A) Maximal contraction of the vessel rings in response to U46619 (3 × 10⁻⁸ mol/L) was analysed (n = 8). (B) Relaxation concentration response curves were generated by five cumulative additions of the endothelium-dependent vasodilator bradykinin (10⁻⁹, 10⁻⁸, 10⁻⁷, 10⁻⁶, and 10⁻⁵ mol/L) at 3 min intervals (n = 8, ANOVA). (C) eNOS mRNA levels were analysed by conventional RT–PCR (n = 7). (D) eNOS protein levels were analysed by western blot (n = 3). The bands were analysed and quantified by densitometry. The ratio of eNOS/GAPDH (mRNA) and ratio of eNOS/β-actin (protein) were evaluated. Results are means ± SEM. ANOVA test was used. *P < 0.05 vs. controls.

Figure 2

Effects of HIV gp120 and TNF-α on the immunoreactivity of eNOS and ICAM-1 in porcine coronary arteries. Porcine coronary artery rings were treated with TNF-α for 8 h and/or HIV gp120 for 16 h. Porcine coronary artery rings were fixed in formalin and embedded in paraffin. Immunostaining was performed. Brown colour represents positive staining of immunoreactivity. (A) eNOS immunoreactivity. (B) ICAM-1 immunoreactivity. Magnification ×400.

Figure 3

Effects of HIV gp120 on eNOS expression and NO release in TNF-α-pretreated HCAECs. HCAECs were incubated in EGM-2 with TNF-α for 8 h and followed by HIV gp120 or heat-inactivated HIV gp120 for 16 h. (A) eNOS mRNA expression was examined by real-time RT–PCR. GAPDH was used as an internal control. n = 3. (B) ICAM-1 mRNA expression was examined by real-time RT–PCR. GAPDH was used as an internal control. n = 3. (C) eNOS and ICAM-1 protein levels were examined by western blot. β-Actin served as a loading control. The bands were analysed and quantified by densitometry. The ratio of eNOS/β-actin and ICAM-1/β-actin was calculated. n = 3. (D) NO levels released from HCAEC cultures were analysed by nitrite assay. Specific eNOS inhibitor l-NAME was used. n = 3. Error bar indicates SEM. ANOVA test was used. *P < 0.05 and **P < 0.01 vs. controls. ^#P < 0.05 vs. TNF-α pretreatment.

Figure 4

Blocking effects of anti-gp120 and anti-ICAM-1 antibodies on HIV gp120-induced decrease in vasomotor function and eNOS expression in TNF-α-pretreated porcine coronary arteries. Porcine coronary artery rings were treated with TNF-α for 8 h, and followed by HIV gp120 alone, HIV gp120 plus anti-gp120 antibody or HIV gp120 plus anti-ICAM-1 antibody treatment for additional 16 h. Vehicle (PBS) treatment served as a negative control. (A) Maximal contraction of the vessel rings in response to thromboxane A2 analogue U46619 (3 × 10⁻⁸ mol/L) n = 4. (B) Relaxation concentration response curves were generated by five cumulative additions of the endothelium-dependent vasodilator bradykinin (10⁻⁹, 10⁻⁸, 10⁻⁷, 10⁻⁶, and 10⁻⁵ mol/L) at 3 min intervals (n = 4, ANOVA). (C) eNOS mRNA levels were examined by conventional RT–PCR (n = 4). (D) eNOS protein levels were determined by western blot (n = 3). The bands were analysed and quantified by densitometry. The ratio of eNOS/GAPDH (mRNA) and ratio of eNOS/β-actin (protein) were evaluated. Results are means ± SEM. ANOVA test was used. *P < 0.05 vs. controls. ^#P < 0.05 vs. gp120 and TNF-α pretreatment.

Figure 5

Effects of HIV gp120, TNF-α, anti-gp120, and anti-ICAM-1 antibodies on eNOS expression in HCAECs. Cells were pretreated with TNF-α for 8 h and followed by HIV gp120, HIV gp120 plus anti-gp120 antibody or HIV gp120 plus anti-ICAM-1 antibody treatment for additional 16 h. (A) eNOS mRNA levels were examined by conventional RT–PCR and densitometry analyses (n = 4). (B) eNOS mRNA levels were determined by real-time RT–PCR analysis (n = 3). GAPDH was used for the loading control. (C) eNOS protein levels were examined by western blot and densitometry analyses (n = 3). (D) ICAM-1 mRNA levels were examined by conventional RT–PCR and densitometry analyses (n = 3). Results are means ± SEM. ANOVA test was used. *P < 0.05 vs. controls. ^#P < 0.05 vs. gp120 and TNF-α pretreatment.

Figure 6

Expression of ICAM-1 and HIV receptors/coreceptors and role of ICAM-1 silencing in HCAECs. (A) HCAECs were treated with or without TNF-α (2 ng/mL) for 8 h, and the expression of ICAM-1 and HIV receptors or coreceptors including CD4, CXCR4, CCR5, and DC-SIGN was studied by flow cytometry analysis. (B) HCAECs were transiently transfected with human ICAM-1 siRNA or scramble siRNA and pre-incubated with TNF-α (2 ng/mL) for 8 h, and then treated with or without HIV gp120 for 16 h. ICAM-1 mRNA levels were examined by real-time RT–PCR. GAPDH served as an internal control (n = 3). (C) eNOS mRNA levels were examined by real-time RT–PCR. GAPDH served as an internal control (n = 3). (D) eNOS protein levels were examined by western blot. β-Actin was served as a loading control. Error bar indicates SEM. ANOVA test was used. **P < 0.01 vs. controls.