Inhibition of nuclear factor-κB activation improves non-nitric oxide-mediated cutaneous microvascular function in reproductive-aged healthy women

Abstract

The transcriptional regulator nuclear factor-κB (NF-κB) is a mediator of endothelial dysfunction. Inhibiting NF-κB with salsalate is used to investigate inflammatory mechanisms contributing to accelerated cardiovascular disease risk. However, in the absence of disease, inhibition of NF-κB can impact redox mechanisms, resulting in paradoxically decreased endothelial function. This study aimed to measure microvascular endothelial function during inhibition of the transcriptional regulator NF-κB in reproductive-aged healthy women. In a randomized, single-blind, crossover, placebo-controlled design, nine healthy women were randomly assigned oral salsalate (1,500 mg, twice daily) or placebo treatments for 5 days. Subjects underwent graded perfusion with the endothelium-dependent agonist acetylcholine (ACh, 10^-10 to 10^-1 M, 33°C) alone and in combination with 15 mM N^G-nitro-l-arginine methyl ester [l-NAME; nonselective nitric oxide (NO) synthase inhibitor] through intradermal microdialysis. Laser-Doppler flux was measured over each microdialysis site, and cutaneous vascular conductance (CVC) was calculated as flux divided by mean arterial pressure and normalized to site-specific maximum (CVC_%max; 28 mM sodium nitroprusside + 43°C). The l-NAME sensitive component was calculated as the difference between the areas under the dose-response curves. During the placebo and salsalate treatments, the l-NAME sites were reduced compared with the control sites (both P < 0.0001). Across treatments, there was a significant difference between the control and l-NAME sites, where both sites shifted upward following salsalate treatment (both P < 0.0001), whereas the l-NAME-sensitive component was not different (P = 0.94). These data demonstrate that inhibition of the transcriptional regulator NF-κB improves cutaneous microvascular function in reproductive-aged healthy women through non-NO-dependent mechanisms.NEW & NOTEWORTHY The transcription factor nuclear factor-κB (NF-κB) regulates multiple aspects of innate and adaptive immunity by encoding for genes that participate in inflammation and impact endothelial function following NF-κB inhibition with salsalate treatment. Our results show that cutaneous microvascular function is increased through non-nitric oxide (NO)-dependent mechanisms following salsalate treatment in reproductive-aged healthy women.

Keywords: NF-κB; NO-independent vasodilation; inflammation; microdialysis; salsalate.

Figure 1.

A: point-by-point acetylcholine (ACh)-induced increases in cutaneous vascular conductance (CVC_%max) during coinfusion of lactated Ringer’s (control, circles) or nonselective nitric oxide synthase (NOS) inhibition (N^G-nitro-l-arginine methyl ester, l-NAME, squares) in reproductive-aged healthy women (n = 9) following placebo (closed symbols) and salsalate (open symbols) treatments. B: l-NAME-sensitive component to ACh-induced increases in CVC following placebo and salsalate treatments. Women on hormonal contraception are depicted in open symbols. Data are presented as means ± SE and were analyzed using a two-way repeated-measures analysis of covariance (A) and Student’s paired t tests (B). *P < 0.0001 vs. placebo-control; **P < 0.0001 vs. placebo-control; †P < 0.0001 vs. placebo-l-NAME; ‡P < 0.0001 vs. salsalate-control.

Figure 2.

Curve modeling of acetylcholine (ACh)-induced increases in cutaneous vascular conductance (CVC_%max) during coinfusion of lactated Ringer’s (control, circles) or nonselective nitric oxide synthase (NOS) inhibition (N^G-nitro-l-arginine methyl ester, l-NAME, squares) in reproductive-aged healthy women (n = 9) following placebo (A) and salsalate (B) treatments. EC₅₀ (sensitivity to ACh) of the control sites are depicted in circle vertical lines, and l-NAME sites are depicted in dashed vertical lines. Individual best-fit EC₅₀ data points depict variability following placebo (C) and salsalate (D) treatments. CVC data are presented as means ± SE, and EC₅₀ data are best-fit value (SE). Data were analyzed using a two-way repeated-measures analysis of covariance. *P < 0.05 vs. control.