Sildenafil improves vascular endothelial function in patients with cystic fibrosis

Abstract

Cystic fibrosis (CF), characterized by defective CFTR function, is associated with multiple systemic complications, including vascular dysfunction. Sildenafil, a phosphodiesterase type 5 inhibitor, not only enhances nitric oxide (NO) metabolism but has been shown to improve CFTR functionality as well. Thus, sildenafil has been proposed as a therapy to improve vascular health in CF; however, its potential therapeutic role has yet to be determined. We sought to investigate the effect of sildenafil on endothelial function in patients with CF. Patients with CF completed a randomized, double-blind, placebo-controlled, crossover study with an acute dose of sildenafil (50 mg) or placebo followed by a 4-wk open-label extension with sildenafil (20 mg/day). Flow-mediated dilation (FMD) was used to evaluate endothelial function before and after treatments. In addition, phosphorylated endothelial NO synthase (pNOS3) and total NOS3 protein expression was determined from endothelial cells that were exposed to plasma from the patients before and after 4 wk of sildenafil treatment. No changes (P ≥ 0.110) in endothelial function were observed after the acute dose of sildenafil. However, FMD significantly (P = 0.029) increased after 4 wk of treatment (∆FMD: 1.5 ± 2.2%). Moreover, pNOS3 protein expression significantly (P = 0.013) increased after 4 wk of treatment (∆pNOS3: 0.31 ± 0.39 arbitrary units) and was associated (r = 0.593, P = 0.033) with the change in FMD. These data suggest that 4 wk of sildenafil treatment can improve vascular endothelial function in patients with CF, likely through an increase in NOS3 phosphorylation.

NEW & NOTEWORTHY Findings from the present study demonstrate, for the first time, significant improvement of endothelial function in patients with cystic fibrosis treated with sildenafil that is associated with greater phosphorylation of endothelial nitric oxide synthase. These results support the use of sildenafil as a potential novel therapy for this patient population.

INTRODUCTION

Cystic fibrosis (CF), the most common, life-shortening genetic disorder among Caucasians, is due to a defective CF transmembrane conductance regulator (CFTR) protein (23). Although pulmonary dysfunction is the most commonly associated feature of the disease, multiple systemic complications, including gastrointestinal, endocrine, and musculoskeletal, among others, have also been described in CF (21, 41). In addition, our group recently described microvascular (47) and conduit vascular (42) dysfunction in patients with CF. Notably, these vascular impairments were identified in patients with mild pulmonary obstruction and could, potentially, contribute to the development of exercise intolerance, a predictor of mortality in CF independent of lung function (39).

Chronic inflammation and systemic oxidative stress are established phenotypes in patients with CF (54, 58) and are thought to contribute to the multiorgan pathophysiology (13, 54) and decline in pulmonary function over time (7) in CF. An environment enriched in reactive oxygen species and inflammatory mediators impacts the normal functionality of the endothelium, mediating a reduction in nitric oxide (NO) synthesis through different mechanisms, including a reduction in endothelial NO synthase (NOS3) phosphorylation (35). Endothelial NO production is essential for the appropriate regulation of vascular tone, particularly endothelium-mediated vasodilation, which has significant implications for systemic blood flow regulation in CF (53). However, despite evidence of impaired endothelial function in CF, current therapeutic interventions have neglected to address this systemic consequence.

Sildenafil, a phosphodiesterase type 5 (PDE5) inhibitor, has been approved by the United States Food and Drug Administration for the treatment of both pulmonary arterial hypertension and erectile dysfunction (57, 59). Sildenafil reduces cGMP breakdown, thereby increasing the sensitivity of vascular smooth muscle to endogenous and exogenous NO, which promotes vasodilation (5). Indeed, improvements in endothelial function have been reported after the use of sildenafil in different populations (12, 22, 48). In addition to the specific vascular benefits, PDE5 inhibition has anti-inflammatory properties (50) and is able to rescue F508del-CFTR trafficking (33, 43). These compelling results have led to the proposal that sildenafil might be a therapy to improve vascular function and the functionality of the CFTR channel in CF (33). However, whether sildenafil improves vascular endothelial function in patients with CF remains to be elucidated. Thus, this proof-of-concept study sought to test the hypothesis that oral ingestion of sildenafil would improve endothelial function in patients with CF through an increase in phosphorylation of NOS3.

MATERIAL AND METHODS

Experimental design.

All patients presented to the Laboratory of Integrative Vascular and Exercise Physiology (LIVEP) at Augusta University for a randomized, double-blind, placebo-controlled, crossover study with an acute dose of sildenafil or placebo followed by a 4-wk open-label extension with sildenafil (clinicaltrials.gov no. NCT02057458). The study followed the principals of the Declaration of Helsinki and was approved by the Institutional Review Board of Augusta University (no. 10-07-019).

All participants reported to the LIVEP four times: a preliminary visit and three separate experimental visits. During the preliminary visit, informed consent was obtained and anthropometrics and pulmonary function were assessed. All participants and parents of minors were informed of the objectives and possible risks of the study before written consent/assent was obtained. For the three experimental visits and in order to follow endothelial function assessment requirements (24), participants reported to the LIVEP at 8 AM after an overnight fast and having abstained from tobacco, caffeine, and vigorous physical activity for 24 h and from vitamin supplements for 72 h before all the visits. All patients were asked to adhere to the timing of their daily airway clearance and inhaled treatments and remained on their prestudy medications throughout the study. At the time of individual completion, no patients were on CFTR modulator therapies. For the randomized, double-blind, placebo-controlled, crossover study, patients reported to the LIVEP on two separate occasions (experimental visits 1 and 2) separated by ≥1 wk to avoid possible carryover effects. During both visits, endothelial function was evaluated before (pre) and 60 min after (post) an acute oral dose of sildenafil (50 mg) or placebo. On the day after experimental visit 2, patients started a short-term treatment of 20 mg sildenafil three times daily (total of 60 mg daily). After 4 wk of treatment with sildenafil, followup testing of vascular endothelial function and a blood draw were completed within 48 h of the patient’s last treatment dose. Treatments were blinded and dispensed by the Augusta University Research Pharmacy following a randomization scheme with a block size of 2. None of the investigators involved in the study knew the details of the series. Adherence to treatment was verified by pill count by the pharmacy. Although sildenafil safety has been previously evaluated in patients with CF (50), adverse events, vital signs, and clinical laboratory values were monitored to verify safety of treatment.

Patients.

Figure 1 shows the clinical trial flowchart for the present study. Nineteen patients were initially recruited, and all but one patient were eligible to participate. Each participant had a clinical diagnosis of CF based on both a positive sweat test and genotype analysis. The majority of patients were homozygous F508del; however, one patient was F508del/621+1G→T and another patient was F508del/G551D. Exclusion criteria for all patients were based on the safety of sildenafil and possible interference with the vascular assessments and consisted of 1) forced expiratory volume in 1 s (FEV₁) < 50% predicted, 2) resting O₂ saturation < 90%, 3) clinical diagnosis of pulmonary hypertension or cardiovascular disease (CVD), 4) use of vasoactive medications, 5) diagnosis of sleep apnea or sleep disorders, or 6) pregnancy or being a self-reported smoker.

Fig. 1.

Clinical trial flowchart.

Demographic characteristics and clinical laboratory values.

During the preliminary visit, participants completed a standard anthropometric assessment of height, weight, and calculated body mass index. Standard biochemical values for lipids (total cholesterol, high-density lipoproteins, low-density lipoproteins, and triglycerides) and glucose concentration were also evaluated using a Cholestech LDX analyzer (Alere, Providence, RI). Hemoglobin and hematocrit values were obtained using a HemoPoint H2 analyzer (Stanbio Laboratory, Boerne, TX). Concentrations of high-sensitivity C-reactive protein and estradiol were obtained using standard core laboratory techniques (LabCorp, Burlington, NC).

Pulmonary function test.

In all patients, pulmonary function was assessed using a complete lung function testing device (EasyOne Pro LAB, ndd Medical Technologies, Andover MA) to evaluate FEV₁, forced vital capacity (FVC), FVC/FEV₁, and forced expiratory flow at 25–75% of pulmonary volume, according to American Thoracic Society guidelines (1). European Respiratory Society Global Lung Function Initiative spirometric reference standards (44) were used to determine the percent predicted data set.

Vascular endothelial function.

Endothelial function was evaluated noninvasively using the brachial artery flow-mediated dilation (FMD) technique in accordance with the most recent methodological recommendations (24, 46). Briefly, simultaneous B-mode and blood velocity profiles of the brachial artery were evaluated by ultrasound imaging (Logiq 7, GE Medical Systems, Milwaukee, WI) using a 12-MHz linear transducer. After acquisition of baseline values, a forearm occlusion cuff (Hokanson, Bellevue, WA), placed immediately distal to the medial epicondyle, was rapidly inflated to 250 mmHg for 5 min (E-20 rapid cuff inflator, Hokanson) to induce arterial occlusion and then deflated to induce reactive hyperemia of the brachial artery. R-wave gating (AccuSync 72, AccuSync Medical Research, Milford, CT) was used to capture end-diastolic arterial diameters for automated offline analysis of brachial artery vasodilation (Medical Imaging Applications, Coralville, IA). The greatest 5-s diameter average after cuff release was used as the peak response. Shear rate was calculated as mean blood velocity × 8/vessel diameter (46); cumulative shear rate [area under the curve (AUC)] was also calculated, following the trapezoidal rule, every 4 s for the first 20 s after cuff release and every 5 s for the rest of the data collection period (24). FMD was expressed as the percent increase in peak diameter from baseline diameter and also relative to shear rate (FMD/shear), as previously recommended (24).

Collection of blood samples.

Venus blood samples (~30 ml) from the antecubital vein were collected into EDTA-containing or serum-separating Vacutainer collection tubes (Becton Dickinson, Franklin Lakes, NJ) after an overnight fast at baseline and after 4 wk of treatment with sildenafil. All blood samples were centrifuged at 3,000 rpm for 10 min at 4°C to obtain plasma. Samples were stored at −80°C until further analysis.

Endothelial cell proliferation.

Cryopreserved human aortic endothelial cells (HAoEC; PromoCell, Heidelberg, Germany) were cultured on tissue culture-treated dishes (Corning, Corning, NY) at 37°C in 5% CO₂ and 95% humidity. The recommended culture medium was endothelial cell growth medium MV2 (PromoCell) supplemented with 0.05 ml/ml heat-inactivated FCS, 5 ng/ml epidermal growth factor, 10 ng/ml basic fibroblast growth factor, 20 ng/ml insulin-like growth factor, 0.5 ng/ml VEGF-165, 1 μg/ml ascorbic acid, 0.2 μg/ml hydrocortisone (PromoCell), and 1% penicillin-streptomycin (ThermoFisher, Waltham, MA). Cells were not used beyond passage 6 and were allowed to reach confluence before experiments.

Endothelial cell experiments.

To investigate possible mechanisms related to the change in FMD and considering that in vitro endothelial cells preserve PDE5 normal activity levels (15), we evaluated the impact of sildenafil treatment on NO metabolism by exposing endothelial cells to plasma from the patients before and after treatment. HAoEC were cultured in 24-well plates for 24 h (100,000 cells/well) at 37°C in a 5% CO₂ environment. Cells were then incubated for 8 h in endothelial cell growth medium MV2 supplemented with 20% (vol/vol) plasma from patients with CF before and after 4 wk of treatment with sildenafil. In the same experiment, HAoECs cultured without plasma supplementation were used as a negative control. After incubation, cells were prepared for Western blot analysis of protein expression (see below). All treatments were completed at the same time of day, minimizing the potential confounding effects of diurnal variation.

Western blot analysis.

An ultrasonic processor was used to homogenize HAoEC samples in 50 μl of RIPA buffer containing phosphatase (PhosStop, Roche, Basel, Switzerland) and protease inhibitor cocktails (Complete Mini EDTA free, Roche). Individual samples for every patient containing 40 μg protein were separated by SDS precast gels (Bio-Rad, Hercules, CA) and transferred to commercial nitrocellulose transfer packs (Trans-Blot Turbo Transfer System, Bio-Rad). Blocking was completed using blocking buffer (Odyssey, LI-COR Biotechnology, Lincoln, NE) for 60 min at room temperature. Membranes were incubated overnight at 4°C with specific primary antibodies for phosphorylated NOS3 (pNOS3; 1:500 dilution, Abcam, Cambridge, UK) and total NOS3 (1:200 dilution, Santa Cruz Biotechnology, Santa Cruz, CA). Bound primary antibody was detected using IRDye secondary antibody (LI-COR Biotechnology). Membranes were scanned with an infrared imaging system (Odyssey CLx, LI-COR Biotechnology). Blots were subsequently quantified using imaging densitometry (ImageJ, National Institutes of Health, Bethesda, MD). To control for loading error, proteins were expressed relative to total β-actin abundance (1:20,000 dilution, Sigma-Aldrich, St. Louis, MO).

Statistical analysis.

Data were analyzed using SPSS (version 23, IBM, Somers, NY) and expressed as means (SD) unless otherwise noted. Significance was set at P < 0.05. The Shapiro-Wilk test was used to analyze the normality of the measurement distribution. Repeated-measures ANOVA was used to evaluate the effects of acute treatment (sildenafil vs. placebo) and time (pre vs. post) for all vascular endothelial function outcomes evaluated during the acute study (experimental visits 1 and 2). Paired t-tests were performed to identify differences between baseline (before the last acute experimental visit) and after 4 wk of sildenafil treatment (baseline vs. 4 wk). Pearson’s correlation coefficients (r) were used to examine the relationships between endothelial function and NOS3 protein expression. Effect sizes for FMD and FMD/shear responses after 4 wk of treatment are reported as Cohen's d values to represent small (Cohen's d = 0.2), medium (Cohen's d = 0.5), and large (Cohen's d = 0.8) effect sizes (10). To confirm reproducibility and reliability of the endothelial function assessments, data from both acute experimental visits before treatment (pre experimental visit 1 and pre experimental visit 2) were compared. To verify reproducibility, five of the six following criteria were required for the main outcomes: 1) a nonsignificant (P > 0.05) paired t-test, 2) a significant intraclass correlation coefficient, 3) a significant r value, 4) a coefficient of variation (CV = SD/mean × 100) of <35%, 5) CV′ [CV′ = (100 × SD)/(mean + 100)] of <3%, and 6) a Bland-Altman plot with not more than one value outside the 95% confidence interval.

RESULTS

Patient characteristics and clinical laboratory values.

Demographic characteristics and laboratory values for patients during the preliminary visit and after 4 wk of treatment with sildenafil are shown in Table 1. Of the 18 patients recruited, 13 patients completed both acute experimental visits. From the 5 patients lost to followup during the acute study, 3 patients, along with the other 13 patients, participated in the open-label 4-wk extension with sildenafil. A total of 15 patients with CF [8 male patients and 7 female patients, 23 (11) yr of age] then completed the 4-wk treatment with sildenafil. Study discontinuation of the different patients was due to time commitment and travel distance. No serious adverse events were reported, and the side effects were usually mild and consistent with those previously reported and indicated on the product label. No differences were identified in any of the participant characteristics between visits. Treatment did not impact pulmonary function (P ≥ 0.354) or clinical laboratory values (P ≥ 0.194). Treatment with sildenafil was well tolerated among all patients throughout the study. No differences in O₂ saturation were observed during the acute treatments or the open-label study.

Table 1.

Demographic characteristics and clinical laboratory values in patients with cystic fibrosis during the preliminary visit and after 4 wk of sildenafil treatment

Variable	Preliminary Visit	4-wk Sildenafil	P Value
Demographic characteristics
n	15
Sex (male/female patients), n	8/7
Age, yr	23 ± 11
Height, cm	160 ± 13
Weight, kg	54 (17)	54 (17)	0.945
Body mass index, kg/m2	20.4 (4.0)	20.6 (3.9)	0.921
Body fat, %	28.5 (7.9)	28.5 (8.0)	0.934
O2 saturation, %	98 (1)	98 (1)	0.670
Systolic blood pressure, mmHg	108 (11)	110 (10)	0.274
Diastolic blood pressure, mmHg	65 ± 7	68 (4)	0.090
Pulmonary function
FVC, liters	3.35 (0.92)	3.36 (0.93)	0.986
FEV1
Liters	2.59 (0.83)	2.64 (0.84)	0.563
%Predicted	80 (16)	81 (14)	0.588
FEV1/FVC, %	77 (8)	76 (7)	0.741
FEF25–75, l/s	2.41 (1.28)	2.29 (1.32)	0.354
Clinical laboratory values
Total cholesterol, mg/dl (<200 mg/dl)	134 (33)	139 (31)	0.257
High-density lipoprotein, mg/dl (>40 mg/dl)	42 (12)	47 (10)	0.194
Low-density lipoprotein, mg/dl [<130 mg/dl]	72 (28)	75 (30)	0.486
Triglycerides, mg/dl (<150 mg/dl)	99 (39)	84 (31)	0.179
Total cholesterol/high-density lipoprotein (<5)	3.5 (1.0)	3.3 (1.1)	0.756
Glucose, mg/dl (70–110 mg/dl)	97 (18)	95 (12)	0.711
Hemoglobine A1c (<6%)	5.5 (0.4)	5.6 (0.5)	0.426
High-sensitivity C-reactive protein, mg/l (<3 mg/l)	2.7 (4.9)	1.4 (1.7)	0.361
Hemoglobin, g/dl (12–18 g/dl)	14.7 (1.9)	14.7 (2.0)	0.989
Hematocrit, % (37–54%)	44.1 (4.0)	44.5 (4.2)	0.607

Values are means (SD). Clinical laboratory values reference ranges are denoted in parentheses. FVC, forced vital capacity, FEV₁, forced expiratory volume in 1 s; FEF_25–75, forced expiratory flow at 25–75% of pulmonary volume.

Acute vascular endothelial function responses.

Parameters of the FMD test before and after the acute dose of sildenafil and placebo are presented in Table 2. No statistically significant differences (all variables P ≥ 0.110) were observed in any of the variables evaluated in the present study after a single oral dose of the PDE5 inhibitor compared with the placebo.

Table 2.

Endothelial function in patients with cystic fibrosis before and after acute sildenafil and placebo

	Sildenafil		Placebo
Variable	Pre	Post	Pre	Post	P Value
Diameter, mm
Baseline	3.05 (0.61)	3.05 (0.58)	3.04 (0.67)	3.00 (0.64)	0.350
Peak	3.29 (0.62)	3.24 (0.58)	3.26 (0.65)	3.19 (0.65)	0.110
Absolute change	0.23 (0.12)	0.24 (0.13)	0.23 (0.11)	0.22 (0.11)	0.848
FMD, %	7.8 (4.3)	7.3 (4.8)	7.7 (4.3)	6.6 (3.7)	0.776
Shear, s−1, AUC	49,384 (23,214)	43,899 (27,793)	50,265 (23,373)	42,328 (17,335)	0.760
FMD (%)/shear (s−1, AUC)	0.17 (0.07)	0.17 (0.08)	0.16 (0.05)	0.16 (0.08)	0.867
Time to peak, s	46 (20)	45 (24)	43 (15)	41 (21)	0.862

Values are means (SD). FMD, flow-mediated dilation; AUC, area under the curve.

Short-term vascular endothelial function responses.

Figure 2 shows brachial artery baseline diameter, FMD, and FMD normalized for shear rate in patients before and after 4 wk of sildenafil treatment. Baseline diameter was similar (P = 0.294; Fig. 2A) before and after treatment. Although there were no differences (P = 0.796) in shear rate between visits, the FMD response (t₁₄ = 2.47, P = 0.029, Cohen’s d = 0.66; Fig. 2B) and FMD normalized for shear rate (t₁₄ = 2.17, P = 0.049, Cohen’s d = 0.58; Fig. 2C) were significantly improved after 4 wk of sildenafil treatment. Additional parameters of FMD testing are shown in Table 3. Peak diameter (P = 0.036) and absolute change in diameter (P = 0.012) increased significantly, but there was no difference (P = 0.646) in time to peak vasodilation after treatment with sildenafil.

Fig. 2.

Vascular endothelial function at baseline and after 4 wk of sildenafil treatment in patients with cystic fibrosis. A: brachial artery baseline diameter. B: flow-mediated dilation (FMD). C: FMD normalized for shear rate. AU, arbitrary units. Values are means ± SE. *P < 0.05 vs. baseline.

Table 3.

Endothelial function in patients with cystic fibrosis before and after 4 wk of sildenafil treatment

Variable	Baseline	4-wk Sildenafil	P Value
Diameter, mm
Baseline	3.00 (0.59)	3.06 (0.60)	0.294
Peak	3.22 (0.60)	3.30 (0.60)	0.035*
Absolute change	0.21 (0.12)	0.27 (0.09)	0.012*
Shear rate, s−1, area under the curve	50,656 (21,929)	49,227 (18,866)	0.796
Time to peak, s	49 (21)	45 (16)	0.646

Values are means (SD).

^*Statistically significant.

NOS3 protein expression.

Protein expression for pNOS3 and total NOS3 in HAoEC exposed to plasma from patients with CF at baseline and after 4 wk of treatment with sildenafil is show in Fig. 3. A significant (t₁₄ = 2.88, P = 0.013, Cohen’s d = 0.77) increase in pNOS3 protein expression was observed after 4 wk of sildenafil treatment, whereas total NOS3 content was unchanged (t₁₄ = 1.69, P = 0.116, Cohen’s d = 0.45) after sildenafil treatment. In addition, a significantly (t₁₄ = 4.02, P = 0.001, Cohen’s d = 1.08) greater ratio of pNOS3 to total NOS3 was observed in endothelial cells after 4 wk of sildenafil treatment.

Fig. 3.

Nitric oxide synthase (NOS3) protein expression in human aortic endothelial cells exposed to plasma from patients with cystic fibrosis at baseline and after 4 wk of sildenafil treatment. A and C: representative Western blots for phosphorylated NOS3 (pNOS3) and total NOS3 with their corresponding β-actin protein expression from 2 patients at baseline (BL) and after 4 wk of treatment with sildenafil (SIL). B and D: quantitative analysis of pNOS3 and total NOS3 expressed relative to β-actin. E: ratio of pNOS3 to total NOS3. AU, arbitrary units. Values are means ± SE. *P < 0.05 vs. baseline.

Relationships between endothelial function and NOS3.

A significant (r = 0.593, P = 0.033) positive association was identified between baseline FMD and the ratio of pNOS3 to total NOS3. In addition, there were strong positive correlations between the change in pNOS3 after 4 wk of sildenafil treatment and both the change in FMD (r = 0.770, P = 0.002) and change in FMD normalized for shear rate (r = 0.789, P = 0.001).

Reproducibility of endothelial function.

Reproducibility of endothelial function in patients with CF is shown in Table 4. An average of 23 days separated pre experimental visit 1 and pre experimental visit 2. Mean values for baseline diameter, peak diameter, FMD, and FMD normalized for shear rate were similar (P > 0.05) between both visits. In addition, all endothelial function variables met at least five of the six criteria for reproducibility (see Statistical analysis) within this time period.

Table 4.

Reproducibility of endothelial function assessments in patients with cystic fibrosis

			Between-Day Pre Visit 1 vs. Pre Visit 2
Variable	Pre Visit 1	Pre Visit 2	t-test	CV	CV′	ICC	r	BA plot
Diameter, mm
Baseline	3.07 (0.53)	3.04 (0.64)	0.558*	19.9*	0.01*	0.997, P < 0.01*	0.979, P < 0.01*	X*
Peak	3.32 (0.61)	3.21 (0.63)	0.576*	18.7*	0.02*	0.985, P < 0.01*	0.970, P < 0.01*	X*
FMD, %	7.6 (4.1)	7.5 (4.3)	0.615*	52.0	2.07*	0.940, P < 0.01*	0.982, P < 0.01*	X*
FMD (%)/shear (s-1, AUC)	0.16 (0.06)	0.15 (0.07)	0.507*	39.2	0.05*	0.854, P < 0.01*	0.825, P < 0.01*	X*

Values are means (SD). Reproducibility was evaluated using t-test statistic (P value), coefficient of variation (CV = SD/mean × 100), CV′ [CV′ = (100 × SD)/(mean + 100)], intraclass correlation coefficient (ICC), Pearson’s correlation coefficient (r), and Bland-Altman (BA) plot. X in the BA cell indicates an acceptable plot. FMD, flow mediated dilation.

^*Acceptable reproducibility.

DISCUSSION

The present study investigated the hypothesis that sildenafil would improve vascular endothelial function in patients with CF through an increase in NO bioavailability. For the first time, we demonstrated that 4 wk of treatment with sildenafil can improve endothelial function in patients with CF, likely by increasing phosphorylation of the NOS3 enzyme in endothelial cells. These proof-of-concept data provide unique evidence supporting the novel use of sildenafil as a potential therapy to improve vascular health in CF.

Effects of sildenafil on vascular endothelial function in patients with CF.

Dysfunction of the CFTR channel contributes to multiple systemic disorders in CF, including vascular dysfunction in the conduit arteries and microcirculation (42, 47). Indeed, a dysfunctional vascular network can negatively impact appropriate blood flow regulation in this patient population (53) and, consequently, contribute to the development of exercise intolerance, a predictor of mortality in CF independent of lung function (36). In addition, vascular dysfunction and its association with increased CVD risk (26) are areas of emerging concern in CF, as the life expectancy of patients continues to increase with advances in medical therapies (45). Because a reduction in NO bioavailability is a key marker of endothelial dysfunction, strategies that enhance vasodilatory capacity may positively impact multiple clinical outcomes in CF.

PDE5 inhibitors, through their dual role as potentiators of NO synthesis (43) and modulators of CFTR functionality (33), exhibit great potential as a possible therapy for CF-associated vascular dysfunction (37). The present study provides the first evidence that short-term treatment with sildenafil can improve vascular endothelial function in patients with CF. Specifically, 4 wk of PDE5 inhibition with sildenafil promoted a significant improvement in vascular endothelial function, as evidenced by a 1.5% increase in FMD (Fig. 2). Although some concerns have been raised about the clinical applicability of FMD, this test represents a noninvasive assessment of NO bioavailability that is associated with cardiovascular health (26). Indeed, 1% changes in FMD have been associated with strong clinical benefits in the cardiovascular system (26). Results from the present study using a within-subject experimental design highlight the potential impact of sildenafil treatment on vascular health in patients with CF.

The importance of the FMD test as a biomarker of vascular health through NO metabolism has been previously reported (14, 20, 26) and is associated with other invasive techniques (3). Several technical and methodological issues must be carefully considered when the FMD test is performed (46), and the same methodology and the same cohort must be considered when absolute values are extrapolated from different studies (4, 6); the reproducibility of the FMD test has been demonstrated in several different populations (9, 11, 19). In addition, we have also evaluated the reliability of this technique in CF. Accordingly, between-day reproducibility for FMD and all FMD parameters (Table 4) met the robust reproducibility criteria. These additional results reinforce our treatment findings, ruling out possible inherent variations in the vasculature during a short period of time and asserting that 4 wk of sildenafil treatment promotes a significant improvement in vascular endothelial function in patients with CF.

It is perhaps not surprising that a single dose of sildenafil had no impact on vascular function in CF, since the efficacy of the PDE5 antagonists seems to be compromised by prior NO-related cGMP formation (8). Indeed, in a previous study (22), a single 100-mg dose of sildenafil, double the acute dose used in the present study, also did not impact vascular functionality in healthy participants or patients with coronary artery disease. These findings support the notion that a repetitive stimulus to increase basal NO activity, such as a short-term treatment, seems to be necessary to produce changes in conduit arterial vasodilation, as previously suggested (22, 34). Future studies are warranted to determine if short-term treatment with sildenafil can improve functionality of the microvasculature, as well as other important dysfunctions described in CF.

Effects of sildenafil on NOS3.

Although the most well-established function of the CFTR protein is the transport of Cl⁻, the CFTR channel is able to regulate other key molecular mechanisms. Specifically, in endothelial cells, the CFTR acts as a sensor of different mechanical stimuli (i.e., blood flow) and triggers multiple signaling pathways (49, 52), including phosphorylation and activation of the NOS3 enzyme (51). Indeed, CFTR antagonists are able to reduce phosphorylation of NOS3 in endothelial cells. This reduction in phosphorylation ameliorates NO synthesis (51) and promotes greater oxidative stress and vascular dysfunction (29, 56), conditions that are prevalent in CF (30, 42).

PDE5 inhibitors are well known for promoting NO metabolism through different mechanisms, including an increase in the activation of NOS3 in vivo (5), and through the use of in vitro experiments with endothelial cells (15, 17, 25). Our results favor this general position, documenting that short-term treatment with sildenafil in patients with CF enhanced phosphorylation of NOS3 in endothelial cells. It is important to note that patients with CF exhibit dysfunctional hepatic metabolism (28) and a faster elimination of sildenafil than controls, which could impact the metabolism of the treatment (50). Our findings agree with previous investigations that described the ability of sildenafil to recouple NOS3, increase its phosphorylation, and restore endothelial NO synthesis (5). Without a doubt, other potential mechanisms may also be related to the treatment response. Indeed, different PDE5 inhibitors, including sildenafil, are also able to exert anti-inflammatory effects, observed specifically in CF (32, 43, 50), that may also contribute, at least partially, to preserve NO metabolism (16). Notably, our data show that the change in pNOS3 after treatment with sildenafil was strongly associated with the improvement in endothelial function in our patients. Indeed, an even stronger positive association was identified between the change in the active form of NOS3 and the change in normalization of FMD for shear stress after treatment. Although these results are from non-CF endothelial cells, they are consistent with previous investigations (31, 51) suggesting that PDE5 inhibition may be able to normalize endothelial cell function through NOS3 activation.

Clinical relevance.

It is well established that endothelial dysfunction strongly contributes to the development of CVD risk in different populations (55). Specifically, the 1.5% improvement in FMD after 4 wk of sildenafil treatment could not only be associated with lower risk to develop cardiovascular manifestations (26) but is estimated to result in a ~19% reduction in CVD risk (26). In the CF patient population, however, the improvement in vascular function may have additional implications related to blood flow regulation and exercise capacity. Given the positive association between endothelial function and exercise capacity in patients with CF (42), our data suggest that sildenafil-mediated improvements in endothelial function may also lead to improved exercise capacity in this population, which may have significant implications for quality of life and survival (18, 36, 38, 39). Although the lack of placebo treatment in the short-term response must be considered in interpretation of the results of the present study, data from this study suggest that sildenafil is able to improve NO bioavailability and vascular endothelial function in CF.

Conclusion.

In summary, this is the first proof-of-concept study to evaluate the effect of sildenafil on vascular health in patients with CF. Findings from the present investigation demonstrate that 4 wk of treatment with sildenafil significantly improves endothelial function in patients with CF. In addition, the increase in endothelial function was associated with greater NOS3 phosphorylation in endothelial cells. Furthermore, we provide evidence for the reproducibility of FMD in patients with CF. The present results support the use of sildenafil as a potential novel therapy in patients with CF to improve NO bioavailability and vascular endothelial function in this population. Future studies are warranted to determine if longer-term treatment with sildenafil can lead to sustained improvements in vascular function in this population and to further our understanding of the mechanism(s) by which sildenafil impacts vascular function in CF.

GRANTS

This work was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grant R21 DK-100783 (to R. A. Harris).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

P.R.M., N.S., and R.A.H. performed experiments; P.R.M., N.S., G.C., and R.A.H. analyzed data; P.R.M., M.A.T., G.C., and R.A.H. interpreted results of experiments; P.R.M. and R.A.H. prepared figures; P.R.M., M.A.T., and R.A.H. drafted manuscript; P.R.M., N.S., M.A.T., G.C., K.T.M., C.F., and R.A.H. edited and revised manuscript; P.R.M., N.S., M.A.T., G.C., K.T.M., C.F., and R.A.H. approved final version of manuscript; K.T.M., C.F., and R.A.H. conceived and designed research.