Tetrahydrobiopterin Supplementation Improves Endothelial Function But Does Not Alter Aortic Stiffness in Patients With Rheumatoid Arthritis

Background Rheumatoid arthritis is a systemic inflammatory condition associated with increased cardiovascular risk that may be due to underlying endothelial dysfunction and subsequent aortic stiffening. We hypothesized that supplementation with tetrahydrobiopterin (BH 4) would recouple endothelial nitric oxide synthase and thus improve endothelial function and consequently reduce aortic stiffness. Methods and Results We conducted 2 randomized, double‐blinded, placebo‐controlled crossover studies examining 2 separate regimens: an acute regimen, with a single dose of BH 4 400 mg versus placebo (n=18), and a short‐term regimen, composed of a 1‐week treatment with BH 4 400 mg once daily versus placebo (n=15). Flow‐mediated dilatation and aortic pulse wave velocity were studied 4 times, before and after each treatment phase. Acute BH 4 supplementation led to an improvement of flow‐mediated dilatation, whereas placebo had no effect (mean±SD of effect difference 2.56±4.79%; P=0.03). Similarly, 1‐week treatment with BH 4 improved endothelial function, but there was no change with placebo (mean±SD of effect difference 3.50±5.05%; P=0.02). There was no change in aortic pulse wave velocity following acute or short‐term BH 4 supplementation or placebo (mean±SD of effect difference: acute 0.09±0.67 m/s, P=0.6; short‐term 0.03±1.46 m/s, P=0.9). Conclusion Both acute and short‐term supplementation with oral BH 4 improved endothelial function but not aortic stiffness. This result suggests that BH 4 supplementation may be beneficial for patients with rheumatoid arthritis by improving endothelial dysfunction and potentially reducing risk of cardiovascular disease. There appears to be no causal relationship between endothelial function and aortic stiffness, suggesting that they occur in parallel, although they may share common risk factors such as inflammation.

R heumatoid arthritis (RA) and other chronic inflammatory diseases are associated with increased premature mortality, mostly due to an excess of cardiovascular disease (CVD). 1,2 A number of underlying mechanisms have been proposed, including endothelial dysfunction. Reduced nitric oxide (NO) bioavailability or endothelial dysfunction is thought to be a key early step in the initiation of atherosclerosis and predicts future CVD-related events in a variety of populations, improving risk classification beyond the Framingham Heart Study scores. 3,4 We and others have demonstrated that RA patients have endothelial dysfunction, [5][6][7][8][9] which can be improved with anti-inflammatory therapies. [10][11][12] Aortic stiffening, an independent risk factor for CVD, 13 has also been implicated. Several groups have reported increased aortic pulse wave velocity (aPWV) in RA and suggested that this could be a consequence of endothelial dysfunction. 10,[14][15][16] Previous studies suggested a link between endothelial function and stiffness of muscular arteries, 17,18 but whether endothelial function is causally related to aortic stiffness is unknown.
The mechanism of endothelial dysfunction in RA is incompletely understood, but reduced tetrahydrobiopterin (BH 4 ) availability or stability 19 and increased production of reactive oxygen species have been suggested. 20 Under basal conditions, endothelial nitric oxide synthase (eNOS) is responsible for NO production, but during inflammation, inducible NOS is also expressed in arterial endothelial cells.
Inducible NOS activation impairs vasorelaxation in vitro, at least in part, by limiting availability of BH 4 to eNOS, 21 and we have demonstrated in vivo that inducible NOS activation in RA is independently related to endothelial function. 9 BH 4 deficiency is thought to lead to eNOS uncoupling, in which electrons flowing from the reductase domain to the oxygenase domain are diverted to molecular oxygen rather than L-arginine, leading to production of superoxide (O 2 À ) rather than NO. 22 Increased release of myeloperoxidase from activated neutrophils may also play an important role in endothelial dysfunction in RA. Myeloperoxidase catalytically consumes NO and increases the production of reactive oxygen species, subsequently leading to uncoupling of eNOS 20 28 and type II diabetes. 29 Oral BH 4 supplementation improves endothelial function in elderly participants, 30 in long-term smokers, 31 and in patients with hypercholesterolemia 32 and hypertension. 33 No studies to date, however, have investigated the effects of BH 4 in patients with RA. Moreover, it is unclear if improvement of endothelial function would lead to a reduction of aortic stiffness or if other mechanisms are responsible for increased aortic stiffness.
Our aim was to test the hypotheses (1) that oral BH 4 supplementation would improve endothelial function in patients with RA and (2) that an improvement of endothelial function would lead to subsequent amelioration of aortic stiffening, providing evidence about the causal relationship between endothelial function and aortic stiffness; BH 4 is thought to have little direct effect on inflammation per se.

Study Population
Overall, 33 patients with active RA (disease activity score [DAS28] >3.5) who met the 1987 American Rheumatism Association criteria were recruited from the rheumatology clinics at Addenbrooke's Hospital, Cambridge, United Kingdom. Patients with manifest CVD, untreated hypertension (blood pressure ≥140/90 mm Hg), diabetes, hypercholesterolemia (total cholesterol ≥6.5 mmol/L), or renal disease and current smokers were excluded because these conditions are associated with endothelial dysfunction and arterial stiffening. 34 To determine whether baseline endothelial function and aortic stiffness were abnormal in RA patients, 33 age-and sexmatched healthy control participants were randomly selected from our database to compare baseline endothelial function and aortic stiffness. A favorable ethics opinion was obtained from the National Research Ethics Service, and written informed consent was obtained from each participant prior any study procedures.

Pilot Proof-of-Concept Study
We conducted a small pilot study in 5 healthy nonsmoking control participants (mean age 27AE6 years) to test whether a single dose of BH 4 400 mg would improve endothelial function. We found that flow-mediated dilatation (FMD) response increased (from 5.79AE1.77% to 11.37AE4.05%; P=0.009), and there was a significant change in the baseline diameter (from 3.77AE.52 to 4.10AE0.62 mm; P=0.008); however, there was no change in mean arterial pressure (P=0.1) or aPWV (P=0.2). Data are presented as meanAESD, and significance was determined using the paired Student t test. Based on these pilot data, the 400 mg once-daily dose was deemed sufficient for the main study in RA patients and was similar to doses used previously in other conditions. 19,33 Experimental Protocol The present study was conducted in a randomized, doubleblind, crossover manner. To determine whether acute changes were sustained, patients were allocated to 1 of 2 groups: acute or short-term BH 4 supplementation.
In the acute study, participants made two 6-hour visits to the vascular laboratory ( Figure 1A). On day 1, following the baseline measurement of endothelial function, aortic stiffness, DAS28, and venous blood sampling, participants received a single oral dose of BH 4 400 mg or placebo (random assignment). At 3 hours after the administration of BH 4 or placebo, the hemodynamic measurements were repeated. A 3-hour time point was chosen based on published pharmacokinetic data. 35 One week later, the same protocol was followed, and participants again received either BH 4 or placebo.
In the short-term supplementation study, participants made 4 visits, each separated by 1 week. Participants received BH 4 400 mg once daily or placebo (random order), each for 1 week, with 1-week washout between treatments ( Figure 1B). All hemodynamic measurements were assessed at baseline and at the end of each 7-day treatment period (treatment 1, washout, and treatment 2). Blood was drawn at each time point for the measurement of biochemical markers, and DAS28 was calculated.

BH 4 Tablets
BH 4 was given in a form of sapropterin dihydrochloride (6R-BH 4 ). 6R-BH 4 is the synthetic form of naturally occurring BH 4 (BioMarin Pharmaceutical Inc) and is approved for treatment of BH 4 -responsive phenylketonuria.

Hemodynamic Measurements
All studies were conducted in a quiet, temperature-controlled room. Blood pressure was recorded in the brachial artery using a validated oscillometric technique (HEM-705CP; Omron Corp). Radial artery waveforms were obtained with a highfidelity micromanometer (SPC-301; Millar Instruments) from the wrist, and a corresponding central waveform was generated using a validated transfer function (Sphygmocor; AtCor Medical). Augmentation index (AIx), a composite measure of wave reflection, mean arterial pressure and heart rate were determined using the integrated software. Aortic and brachial pulse wave velocities were measured, as previously described. 36 Endothelial function was assessed in the brachial artery using the FMD technique. 37 Vessel diameter was measured using high-resolution vascular ultrasound (Acuson Aspen; Siemens AG) with a 10.0-MHz linear-array transducer. Brachial artery diameter was measured continuously for 1 minute at baseline and for a further 5 minutes following cuff deflation. The cuff was placed below (distal to) the ultrasound transducer and inflated to 200 mm Hg for 5 min-utes. After return to baseline, vessel diameter was again measured continuously for 5 minutes following administration of 25 lg of sublingual glyceryl trinitrate. FMD was defined as the maximum percentage increase in vessel diameter during reactive hyperemia; glyceryl trinitrate-mediated dilatation was defined as the maximum percentage increase in vessel diameter after sublingual glyceryl trinitrate. FMD recordings were analyzed offline by a blinded operator using Cardiovascular Suite Software (Quipu).

Laboratory Measurements and DAS28
Fasting lipid profile, high-sensitivity C-reactive protein (CRP), and erythrocyte sedimentation rate (ESR) were determined using standard methodology at the Biochemistry Laboratory at Addenbrooke's Hospital.
DAS28 is a validated composite disease activity score. The components of DAS28 include the number of swollen and tender joints from 28 assessed joints, ESR, and patientassessed visual analog score of overall well-being (scale 0-100). DAS28 was calculated as described previously. 38

Data Analysis
The primary outcome of the study was the difference in FMD response between BH 4 and placebo treatments. Secondary outcomes were differences in aPWV between treatments, the relationship between changes in aPWV and FMD, and the relationship between CRP and disease activity and change in FMD and aPWV.
Based on our pilot data and published data on BH 4 supplementation, we calculated that a total of 15 patients (per  Figure 1. Schema of the study design. A, All participants made 2 visits, separated by 1 week. At each time point, blood pressure, arterial stiffness, and endothelial function were assessed, and a blood sample was taken. B, All participants made 4 visits to the unit on days 0, 7, 14, and 21. At each visit, blood pressure, arterial stiffness, endothelial function, and disease activity were assessed, and a blood sample was taken. BH 4 indicates tetrahydrobiopterin. study) were needed to enter the 2-treatment crossover study to provide >80% power at P<0.05 to detect a 4.0-U difference in FMD response between treatments with an SD of 5.0. Data were analyzed using SPSS software (version 21.0; IBM Corp). Two-way repeated-measures ANOVA was used to investigate the effect of the treatment. Custom hypothesis testing (simple) of within-participant contrasts was performed for comparison of BH 4 versus placebo, in which treatment order was entered as a between-participant factor to investigate whether the order in which BH 4 and placebo were received influenced the outcome. In post hoc tests, the effect of individual treatments was determined using paired Student t tests with Bonferroni adjustment for 2 comparisons. In addition, in post hoc tests, the use of methotrexate (MTX) was entered as a between-participant factor to establish whether dyhydrofolate reductase inhibitor use influenced the BH 4 response. For the skewed variables (CRP and ESR), logtransformed values were used for the analyses. Differences between the RA and control groups were determined using the unpaired Student t test. A probability of <0.05 was considered significant. Data are given as meanAESD.
The authors had full access to the data and take full responsibility for its integrity. All authors have read and agreed to the manuscript as written.

Results
Thirty-three patients with active RA (mean DAS28 4.21AE1.06) were studied before and after oral BH 4 and placebo. The mean age of the participants was 57AE12 years, and 27 were female. Overall, 18 and 15 participants concluded the acute and short-term supplementation studies, respectively. There were no differences in age, BMI, DAS28, CRP, ESR, or blood pressure between groups; however, total cholesterol was slightly higher in the participants allocated to the short-term treatment group (4.4AE0.9 versus 5.4AE1.1 mmol/L; P=0.01). Baseline characteristics are detailed by group in Table 1.

Acute BH 4 Supplementation
The effect of BH 4 supplementation on hemodynamics and vascular function are detailed in Table 2. Following a single dose of BH 4 , there was a significant increase in the FMD response, whereas placebo had no effect (+3.57AE4.14% versus +1.01AE3.12%, between treatments; P=0.03) (Figure 2A). The order in which the treatments were given did not affect the outcome (P=0.8). Baseline diameter of brachial artery or response to sublingual glyceryl trinitrate did not change following the treatment. There were no significant differences between the changes in blood pressure, heart rate, AIx, brachial pulse wave velocity, or aPWV ( Figure 2B) following BH 4 or placebo.

Short-Term BH 4 Supplementation
The 1-week treatment with BH 4 led to a significant increase in FMD compared with placebo (+3.69AE4.90% versus +0.19AE2.51%, between treatments; P=0.02) ( Figure 2C). The order in which the treatments were given did not affect the outcome (P=0.7). Again, there was no change in blood pressure or aPWV ( Figure 2D).
Seventeen of the 33 RA patients were receiving MTX, a dyhydrofolate reductase inhibitor, at the time of the study. Post hoc testing demonstrated that MTX had no impact on the effect of BH 4 on endothelial function (P=0.34) for the drug9visit9MTX interaction. If anything, there was a nonsignificant trend demonstrating that those on MTX had greater improvement of FMD following BH 4 in comparison to those not on MTX (+4.6AE5.0% versus +2.49AE3.40%; P=0.18,). This was despite no difference in baseline FMD (P=0.9) or brachial artery diameter (P=0.98).

Discussion
The main findings of the present study are that both acute and short-term therapy with BH 4 significantly improves endothelial function, as assessed by vasodilatory response to reactive hyperemia in patients with active RA, but does not affect blood pressure, AIx, aPWV, or brachial pulse wave velocity. These findings suggest that modulation of endothelial function does not ameliorate arterial stiffness in RA.
The current study demonstrates, for the first time, that oral BH 4 supplementation improves endothelial function in patients with RA by %3.5%, restoring it to levels comparable to those of healthy controls. The improvement of FMD was inversely associated with age, suggesting that endothelial dysfunction in younger RA patients may be more reversible than in older patients. FMD has been validated recently as a surrogate marker of future cardiovascular risk. Yeboah et al demonstrated that FMD is a predictor of incident cardiovascular events and that it improves the classification of participants as low, intermediate, and high CVD risk compared with the Framingham risk score in a large populationbased cohort. 4 The mean FMD in their whole cohort was 4.4AE2.8% versus 3.4AE2.5% in incident cases, suggesting that the improvement seen in our study is likely to have a clinical impact on cardiovascular risk reduction in patients with RA.
Previous studies have demonstrated that intra-arterial administration of BH 4 improves endothelium-dependent dilatation in patients with hypercholesterolemia, 27 coronary artery disease, 28 and type II diabetes. 29 Oral BH 4 supplementation also improves endothelial function in elderly patients, 30 in long-term smokers, 31 and in patients with hypercholesterolemia 32 and hypertension. 33 Conversely, Cunnington et al showed that although oral BH 4 led to an %2.5-fold increase of Values represent meanAESD. Significance was determined using an unpaired Student t test. BMI indicates body mass index; BP, blood pressure; bpm, beats per minute; CRP, C-reactive protein; DAS28, disease activity score; DMARDs, disease-modifying antirheumatic drugs; ESR, erythrocyte sedimentation rate; NSAIDs, nonsteroidal anti-inflammatory drugs; TC, total cholesterol; TNF-a, tumor necrosis factor a. *P value for the comparison between acute and short-term study participants.
plasma BH 4 levels in patients with established coronary artery disease, it had no effect on vascular function or superoxide production because of systemic oxidation of BH 4 to BH 2 . 19 The impact of BH 4 on endothelial function has not been assessed previously in RA; however, other interventions (eg, individually designed exercise program) have been shown to improve endothelial function in RA patients. 39 Because intracellular BH 4 levels are regulated by de novo synthesis of BH 4 from guanosine triphosphate and through recycling of BH 2 to BH 4 via dyhydrofolate reductase, 40 use of dyhydrofolate reductase inhibitors (eg, disease-modifying antirheumatic drugs, MTX) may inhibit the bioavailability of BH 4 in the endothelium. Nevertheless, we found no difference in the effect of BH 4 supplementation on endothelial function between those RA patients on MTX and those not receiving MTX. A trend suggested that those patients on MTX had a greater increase in FMD following BH 4 supplementation, perhaps because of reduced levels of inflammation. Notably, our study was not powered to look at this difference. We and others have demonstrated that patients with RA have increased aortic stiffening in comparison to healthy participants, 10,[14][15][16] and that can be reversed with immunomodulatory therapy. 10,12,41 Our previous studies suggested a link between arterial stiffness and endothelial function, 17,18 but it is unclear if this relationship is causal or if they occur in parallel in response to common drivers such as increased inflammation. Interestingly, in the present study, neither a single dose nor short-term supplementation with oral BH 4 led to a change in aPWV, brachial pulse wave velocity, AIx, or blood pressure in comparison to placebo. Park et al recently demonstrated in patients with chronic kidney disease that 6R-BH 4 significantly improved AIx 42 ; however, there was also a drop in blood pressure, and it appears that the change in AIx was not corrected for change in mean arterial pressure, only for heart rate. Similarly, Porkert et al showed a 15-mm Hg reduction in systolic blood pressure following 8-week treatment with BH 4 400 mg in hypertensive patients, but this study was not placebo controlled. 33 In healthy older men, BH 4 was shown to reduce carotid artery compliance, but there was a confounding fall in systolic blood pressure. 43 In postmenopausal women, however, Moreau et al showed an improvement in carotid artery compliance and FMD response following a single dose of oral BH 4 without change in mean arterial pressure. 44 Our study also provided evidence about causality for endothelial function and aortic stiffness. We demonstrated previously that under resting conditions, local NO production modulates distensibility of muscular iliac artery in humans and that exogenous NO increases arterial distensibility 18 ; however, data relating to elastic aorta are more controversial.

Conclusion
This study shows for the first time in patients with RA that oral BH 4 improves endothelial function as measured by a response to reactive hyperemia. These findings suggest that recoupling of eNOS using cofactor BH 4 may provide a rational therapeutic approach to prevent CVD in patients with chronic inflammatory diseases. BH 4 supplementation did not appear to reduce aortic stiffness, suggesting that endothelial dysfunction is not a pathophysiology behind aortic stiffening in RA and that these conditions may just exist in parallel, both influenced by common risk factors such as inflammation and increased reactive oxygen species production.