Diminished Global Arginine Bioavailability and Increased Arginine Catabolism as Metabolic Profile of Increased Cardiovascular Risk

Abstract

Objective

We hypothesized that an integrated assessment of arginine with its catabolic products may better predict cardiovascular risks than arginine levels alone.

Background

Arginine is the sole nitrogen source for nitric oxide (NO) synthesis. The major catabolic products of arginine are ornithine and citrulline.

Methods

Plasma levels of free arginine, ornithine, citrulline and the endogenous NO synthase inhibitor asymmetric dimethylarginine (ADMA) were measured using LC/MS/MS. We examined the relationship of global arginine bioavailability ratio (GABR, defined as arginine/[ornithine+citrulline]) vs. arginine and its catabolic metabolites to prevalence of coronary artery disease (CAD) and incidence of major adverse cardiovascular events (MACE = death, myocardial infarction, stroke) over a 3-year follow-up in 1,010 subjects undergoing elective cardiac catheterization.

Results

Patients with CAD had significantly lower GABR [median(IQR); 1.06(0.75, 1.31) versus 1.27(0.96, 1.73), p<0.001] and arginine levels [mean: 68 ±20 μM versus 74 ±24 μM, p<0.001) than those without CAD. After adjusting for Framingham risk score, C-reactive protein, and renal function, lower GABR (but not arginine levels) and higher citrulline levels remained significantly associated with both prevalence of CAD [adjusted odds-ratio (OR) 3.93, p<0.001 and 5.98, p<0.001, respectively] and 3-year risk for incidence of MACE [adjusted Hazard ratio (HR) 1.98, p=0.025 and 2.40, p=0.01, respectively], and remained significant after adjusting for ADMA.

Conclusions

GABR may serve as a more comprehensive concept of reduced NO synthetic capacity compared to systemic arginine levels. Diminished GABR and high citrulline levels are associated with both development of atherosclerotic CAD and heightened long-term risk for major adverse cardiac events.

INTRODUCTION

Nitric oxide (nitrogen monoxide, NO) is an important endothelium-derived vasoactive substance critical to vascular health and homeostasis(1). Arginine serves as the sole nitrogen source for NO synthases. Nitric oxide is synthesized from arginine in a multistep reaction carried out by NO synthases, producing NO and citrulline (Figure 1). Numerous lines of evidence indicate that diminished NO bioavailability is a critical predisposing factor in development of atherosclerotic heart disease(2). Experimental animal model and preliminary clinical studies both indicate that intra-arterial or intravenous infusion of arginine improves NO production in the coronary arteries(2-6), although results of oral arginine supplementation studies have been variable(7-9). Assessing arginine bioavailability may therefore provide an important insight into cardiovascular health not achievable by directly assessing arginine levels(10), and may potentially identify patients who may receive the greatest benefit from enhancing arginine bioavailability.

Figure 1. Schematic illustration of pathways for nitric oxide production and arginine catabolism.

CPS = carbamyl phosphate synthetase; ADMA = asymmetric dimethylarginine; L-NMMA = N(G)-mono-methyl-L-arginine.

Arginine is the common substrate for both NO synthases and arginases (Figure 1)(11). In several disease states arginase levels are increased and have been proposed to limit NO formation through increased arginine consumption. This limiting of NO formation results in indirect competition with NO synthases, the formation of ornithine, and subsequently citrulline (12,13). In turn, citrulline (also a direct product of NO synthases) may convert back to arginine via the formation of argininosuccinate (primarily in the kidneys)(14). Hence, we hypothesized that the “bioavailability” of arginine may be impacted by different processes that may reduce arginine availability to NO synthases. We therefore proposed a concept of “global arginine bioavailability ratio” (GABR) to account for levels of the substrate (arginine) and its major catabolic products (ornithine and citrulline) in vivo. Specifically, we tested the hypothesis that plasma levels of GABR would be more highly predictive of the development and progression of cardiovascular diseases than plasma levels of free arginine. We verified this prediction by measuring the prevalence of significant CAD at baseline and the subsequent incidence of MACE in our study population. We further explored the relative prognostic value of GABR with asymmetric dimethylarginine (ADMA), an endogenous direct NO synthases inhibitor.

METHODS

Study population

The Cleveland Clinic GENEBANK study is a large, single-center, prospective, cohort study with plasma, serum, and DNA repository obtained from sequential subjects undergoing elective diagnostic cardiac catheterization. The Cleveland Clinic Institutional Review Board approved the study, and all participants gave written informed consent. Clinical and demographic information was collected at the time of enrollment, and data were de-identified before analysis. This study utilized plasma samples obtained from 1,010 consecutive subjects enrolled in GENEBANK. The Framingham Risk score (including diabetes) was calculated for each subject, and the glomerular filtration rate was estimated by the Modification of Diet in Renal Disease (MDRD) formula. High sensitivity C-reactive protein (hsCRP), creatinine, fasting blood glucose and lipid profiles were measured on the Abbott ARCHITECT platform (Abbott Diagnostics, Abbott Park IL). Over the 3 years following enrollment, adjudicated outcomes were ascertained for all subjects.

Data extraction and endpoints

Coronary artery disease (CAD) was defined as any clinical history of myocardial infarction (MI), percutaneous coronary intervention, coronary artery bypass surgery, acute coronary syndrome or angiographic evidence of CAD (>50% stenosis) in one or more major coronary arteries. A major adverse cardiovascular event (MACE) was defined as the occurrence of non-fatal myocardial infarction, non-fatal stoke, or death within 3 years of follow-up. All clinical outcomes were adjudicated with source documentation.

Sample collection and measurements

Samples were collected from fasting subjects on the day of elective cardiac catheterization. Plasma aliquots analyzed were isolated from whole blood collected into EDTA (lavender top) tubes which were maintained at 0-4°C immediately following phlebotomy, processed within 4 hours of blood draw, and stored at -80°C until use. Plasma arginine, ornithine, citrulline and ADMA levels were determined by stable isotope dilution HPLC with on-line electrospray ionization tandem mass spectrometry (LC/ESI/MS/MS) using an API 365 triple quadruple mass spectrometer (Applied Biosystems, Foster, CA) with Ionics 10+ redesigned source as upgrade, and interfaced with a Cohesive HPLC (Franklin, MA). ¹³C₆-Arginine was used as the internal standard and added at the time of plasma aliquot thawing. Plasma proteins were precipitated with 80% methanol and supernatant, after centrifugation at 5,000 g for 10 min, 0°C, and were collected for determination of the analyte concentration as described (15).

Statistical analysis

The Wilcoxon-Rank sum test for continuous variables and chi-square test for categorical variables were used to examine the difference between the groups. GABR was divided into quartiles for analysis. Unadjusted trends for increasing CAD, MACE at 3 years with increasing or decreasing quartiles were evaluated with the log-rank test of trend. Since diminished GABR (i.e., lower values) portend greater risk, logistic regression models were developed to calculate odds ratios (ORs) or hazard ratios (HRs) associated with the first, second, and third quartiles of GABR, which were compared with the highest quartile. Adjustments were made for Framingham Global Risk Score, creatinine clearance, and plasma hsCRP levels. Kaplan—Meier analysis with Cox proportional hazards regression was used for time-to-event analysis. A Spearman correlation coefficient was calculated to summarize the correlation between GABR and ADMA. All analyses were performed using SAS version 8.2 (Cary, NC). P values <0.05 were considered statistically significant.

RESULTS

Study Population

Table 1 illustrates the clinical characteristics of the study population, stratified according to the presence or absence of CAD. As expected, patients with CAD were more likely to be older, and have one or more cardiovascular risk factors (diabetes mellitus and hypertension). In our analysis, plasma arginine level was normally distributed, whereas ornithine, citrulline, and GABR were not.

Table 1.

Baseline Characteristics of Study Population

	No CAD (n=402)	CAD (n=608)	p-value
Demographics and Cardiovascular Risk Factors
Age (years)	61 ± 8	66 ± 10	<0.001
Female (%)	53%	52%	0.772
Diabetes (%)	15%	45%	<0.001
Hypertension (%)	54%	78%	<0.001
Framingham Risk Score (mean ± SD)	6.6 ± 3.4	8.5 ± 4.0	<0.001
Current Smokers (%)	5%	6%	0.318
Laboratory Data
LDL cholesterol (mean ± SD)	108 ± 33	100 ± 35	<0.001
HDL cholesterol (mean ± SD)	53 ± 17	44 ± 14	<0.001
Triglycerides (median/IQR)	117 (83, 168)	141 (105, 212)	<0.001
hsCRP, mg/dL (median/IQR)	2.2 (1.1, 4.6)	3.3 (1.7, 7.6)	<0.001
Creatinine clearance (mean ± SD)	99.2 ± 32.8	88.2 ± 40.0	<0.001
Arginine (μM) (mean ± SD)	74 ± 24	68 ± 20	<0.001
Ornithine (μM) (median/IQR)	35.9 (25.9, 49.1)	41.3 (29.9, 56.6)	<0.001
Citrulline (μM) (median/IQR)	17.7 (13.7, 26.0)	23.4 (17.5, 31.0)	<0.001
GABR (median/IQR)	1.27 (0.96, 1.73)	1.06 (0.75, 1.31)	<0.001

Plasma levels of arginine, GABR and CAD

Compared to those without CAD, patients with CAD had significantly lower levels of plasma arginine, but higher levels of ornithine and citrulline (Table 1). This directly leads to significantly lower median GABR in patients with versus those without CAD [median (interquartile range, IQR): 1.06(0.75-1.31) versus 1.27(0.96-1.73), p <0.001]. After adjusting for traditional risk factors, creatinine clearance, and hsCRP, GABR levels remained significantly associated with prevalent CAD [Quartile 1 compared to Quartile 4; adjusted OR(95% confidence interval, CI): 3.93(2.55-6.05) p<0.001; Table 2]. Higher ornithine and citrulline levels were also associated with higher prevalence of CAD (Table 3). In contrast, while lower plasma arginine levels were associated with higher CAD prevalence across quartiles in unadjusted analysis, low plasma arginine levels were no longer a predictor of prevalence of CAD after adjusting for traditional risk factors, creatinine clearance, and hsCRP [adjusted OR (95%CI): 1.19(0.79-1.80), p=0.40; Table 2].

Table 2.

Odds Ratio (Unadjusted and Adjusted) for Coronary Artery Disease (CAD) Prevalence and Major Adverse Cardiac Events (MACE) at 3 Years with Decreasing Quartiles of GABR and Arginine Levels

	4th quartile	3rd quartile	2nd quartile	1st quartile
GABR	>1.46	1.14-1.46	0.82-1.14	<0.82
CAD Prevalence
Unadjusted OR (95% CI)	1.0	2.41 (1.68-3.47)*	3.28 (2.26-4.75)*	3.63 (2.50-5.28)*
Adjusted OR (95% CI)	1.0	2.76 (1.86-4.09)*	3.77 (2.49-5.72)*	3.93 (2.55-6.05)*
MACE at 3 years
Unadjusted HR (95% CI)	1.0	1.67 (0.89-3.12)	2.27 (1.25-4.11)#	3.64 (2.08-6.39)*
Adjusted HR (95% CI)	1.0	1.41 (0.75-2.67)	1.42 (0.77-2.63)	1.98 (1.09-3.59)†
Arginine (μM)	>82.8	68.5-82.8	55.3-68.4	<55.3
CAD Prevalence
Unadjusted OR (95% CI)	1.0	1.69 (1.18-2.42)#	1.74 (1.21-2.49)#	1.94 (1.35-2.79)*
Adjusted OR (95% CI)	1.0	1.69 (1.12- 2.56)†	1.23 (0.82-1.85)	1.19 (0.79-1.80)
MACE at 3 years
Unadjusted HR (95% CI)	1.0	1.32 (0.75-2.32)	1.55 (0.90-2.66)	1.85 (1.10-3.14) †
Adjusted HR (95% CI)	1.0	1.17 (0.65-2.10)	1.15 (0.66-2.01)	1.22 (0.70-2.12)

^*p<0.001

^#p<0.01

^†p<0.05

Table 3.

Odds Ratio (Unadjusted and Adjusted) for Coronary Artery Disease (CAD) Prevalence and Major Adverse Cardiac Events (MACE) at 3 Years with Increasing Quartiles of Citrulline and Ornithine Levels

	1st quartile	2nd quartile	3rd quartile	4th quartile
Citrulline (μM)	<15.3	15.3-21.2	21.3-29.4	>29.4
CAD Prevalence
Unadjusted OR (95% CI)	1.0	1.50 (1.05-2.15)†	2.91 (2.01-4.20)*	3.17 (2.18-4.61)*
Adjusted OR (95% CI)	1.0	1.67 (1.14-2.46)#	3.63 (2.40-5.51)*	5.98 (3.72-9.60)*
MACE at 3 years
Unadjusted HR (95% CI)	1.0	1.61 (0.87-2.97)	2.21 (1.2-3.96)#	3.35 (1.93-5.82)*
Adjusted HR (95% CI)	1.0	1.62 (0.87-3.01)	1.89 (1.05-3.1) †	2.40 (1.34-4.30)#
Ornithine (μM)	<28.3	28.3-38.9	38.9-53.7	>53.8
CAD Prevalence
Unadjusted OR (95% CI)	1.0	1.30 (0.91-1.85)	1.67 (1.17-2.39)#	2.29 (1.58-3.32)*
Adjusted OR (95% CI)	1.0	1.35 (0.92-1.99)	2.32 (1.54-3.49)*	3.95 (2.53-6.18)*
MACE at 3 years
Unadjusted HR (95% CI)	1.0	1.12 (0.66-1.89)	1.44 (0.86-2.40)	1.54 (0.93-2.54)
Adjusted HR (95% CI)	1.0	1.06 (0.62-1.83)	1.16 (0.68-1.98)	1.23 (0.73-2.06)

^*p<0.001

^#p<0.01

^†p<0.05

Plasma arginine metabolite levels, GABR and MACE

Among the 991 subjects with 3-year follow-up, there were 126 (12.7%) incident MACE events. Analytes were again divided into quartiles for analyses of risk prediction, and the rates of non-fatal MI or stroke, all-cause mortality, as well as the composite MACE examined (Figure 2). Overall, GABR across decreasing quartiles provided consistent incremental association with increasing incident MACE. Figures 3 and 4 show the Kaplan-Meier time-to-event curves for the occurrence of MACE for each quartile of GABR and citrulline, respectively. After adjusting for Framingham Risk Score, hsCRP, and creatinine clearance, subjects with lowest absolute values of GABR highest levels of citrulline showed the greatest risk for incident MACE over the three years after angiography. In contrast, arginine and ornithine quartiles did not predict incident MACE risk at 3 years (Tables 2 and 3).

Figure 2. Distribution of myocardial infraction/stroke, all-cause mortality, and 3-year incidence of major adverse cardiac events according to global arginine bioavailability ratio (GABR) and arginine quartiles.

P-value for trend.

Figure 3.

Kaplan-Meier survival analysis for patients with 3-year incidence of major adverse cardiac events according to global arginine bioavailability ratio (GABR) quartiles

Figure 4.

Kaplan-Meier survival analysis for patients with 3-year incidence of major adverse cardiac events according to citrulline quartiles

Additional analyses were performed looking at the relationship between GABR and ADMA in predicting prevalence of CAD and incidence of MACE (3 year) within the cohort. Overall, GABR and ADMA showed a modest negative correlation (r =-0.24, p <0.001). After adjusting for ADMA, the lowest quartile of GABR remained independently predictive of both higher prevalence of CAD [adjusted OR (95%CI): 3.28(2.24-4.82), p<0.001] and higher incidence of MACE [adjusted HR (95%CI): 2.93(1.66-5.20), p<0.001] when compared to the highest quartile of GABR. Similar observations were made in the highest quartile of citrulline after adjusting for ADMA for higher incidence of MACE [adjusted HR(95%CI): 2.03(1.44-2.87), p<0.001].

DISCUSSION

The relationship between plasma arginine levels, NO biosynthesis and cardiovascular disease is complex. Arginine participates in various metabolic reactions that differ by compartments and are modulated by diet, cytokines, and hormones(14). To our knowledge, our study is the largest experience to date to address the clinical significance of assessing systemic levels of arginine metabolites for cardiovascular risks in humans. Our study demonstrates the importance of assessing catabolites of arginine metabolism, and shows the ability to predict cardiovascular disease prevalence and future risks for major adverse cardiac events when accounting for catabolic products of arginine (ornithine and citrulline). These findings were independent of traditional cardiovascular risk factors, renal function, markers of inflammation, and even when accounting for levels of the endogenous NO synthase inhibitor ADMA.

Increasing arginine as substrate for NO synthases has been postulated to increase in endothelial NO production(1). However, arginine supplementation studies have demonstrated mixed outcomes with respect to cardiovascular disease outcomes(7). The underlying mechanisms for this are not clear. Under physiologic conditions, the intracellular NO synthases should be adequately saturated with its arginine substrates. Enhanced NO production observed in response to elevations in extracellular arginine, despite high intracellular arginine concentrations, is often referred to as the “arginine paradox”(16,17). Several investigators have postulated that endogenous inhibitors of NO synthases, methylated arginine metabolites such as ADMA, may provide the explanation(16). Indeed, our group and others have demonstrated the clinical significance of elevated plasma ADMA levels in cardiovascular risks(15,18-20). In particular, data have suggested that a reduction in intracellular concentrations of competitive and non-competitive inhibitors by extracellular arginine displacement may augment NO synthase activity, thereby increasing NO production(21).

Our current data provide an alternative (or complementary) explanation to the arginine paradox, whereby a relative deficiency of arginine for NO synthases caused by augmented “catabolic pathways” of arginine (such as arginases) can in part contribute to underlying CAD and development of cardiovascular events. While in vitro studies have indicated an association between decreases in arginine concentrations, both systemic and locally, within atherosclerotic plaques, as well as impairment of the endothelial NO synthase-mediated stress responses(22), systemic arginine levels have not been correlated with prevalence of CAD(10). Also, arginine supplementation studies have thus far failed to demonstrate improvement in cardiovascular outcomes(8). Several studies have suggested a relative arginine deficiency exists in subjects with hypertension or heart failure as a result of abnormal transport mechanisms across vascular cellular membranes(13,23,24). Extracellular concentrations of arginine can also depend on catabolism of the amino acid via several pathways including augmented degradation of arginine, which may lead to relative NO deficiency and subsequent progression of cardiovascular disease.

Arginases, in particular, have been shown to be upregulated in several disease states with components of inflammation including CAD, cystic fibrosis, sickle cell disease, and asthma(25). Indeed, arginases are subject to regulation by inflammatory cytokines as well as inter-regulation by the arginine metabolites themselves(26). Increased arginase activity in endothelial cells has been argued to promote a pro-atherogenic effect because of direct reduction of endothelial cell NO production via the conversion of arginine to ornithine(2). Indeed, it has been suggested that indirect strategies for elevating arginine (such as the inhibition of arginase) could prove more effective at improving intracellular arginine bioavailability than exogenous arginine administration(2). Further support for this approach may be seen in smooth muscle cells, where heightened arginine catabolic processes could result in increased production of collagen and enhanced cell proliferation attributable to the metabolism of ornithine into proline and polyamines, respectively. The recently demonstrated association between the arginase I gene polymorphism, rs2781666, and both heightened MI risk and increased carotid intimal medial thickness is consistent with the hypothesis that arginine catabolic processes are linked to CAD pathogenesis(27). Meanwhile, the downstream effects of diminished arginine bioavailability leading to progression of cardiovascular disease have also been investigated. One suggestion has been the impact of this decreased availability of arginine on the synthesis of NO by interfering with NO synthase mRNA translation via altered phosphorylation and activity of eukaryotic initiation factor(28).

There are several limitations of our study. Only plasma samples at a single time-point were analyzed, and there were no direct physiologic vascular measures to directly link vascular functional changes with diminished GABR and cardiovascular risks. However, numerous studies have demonstrated enhanced vascular NO responses following arginine administration (3,5,7), conditions that correspondingly will increase GABR. It is also conceivable that underlying metabolic processes not considered and other medical therapies may have the potential to alter the plasma sample measurements. We also acknowledged that the proposed ratio of substrates and products of arginine metabolism as defined by GABR fails to take into account important issues such as compartmentation, and the fact that biological systems represent dynamic systems striving toward achieving steady state conditions, not equilibria. Finally, measuring the levels of arginine and its related metabolites poses challenges due to the influence of food intake and analytical constraints on amino acid levels. Nevertheless it should be emphasized that our study utilized fasting samples and quantitatively robust stable isotope dilution mass spectrometry techniques. The relatively large sample size further helps overcome these potential limitations. Nevertheless, further studies should be conduct to determine the prognostic value of GABR or citrulline levels alone in a prospective fashion under a broader range of conditions to determine its prognostic value as a biomarker and its potential to guide metabolic therapy. Better understanding of the mechanistic underpinnings of metabolic and vascular abnormalities may provide the basis for further investigations to explore the impact of a variety of treatment strategies aiming to improve arginine bioavailability and reduce cardiovascular risks.

CONCLUSION

The GABR, a novel comprehensive concept of arginine metabolism that accounts for arginine catabolic metabolites, may provide a better index of reduced NO synthetic capacity than systemic arginine levels alone. In our study population of stable patients undergoing elective coronary angiography, lower GABR and higher citrulline levels were associated with both higher prevalence of significant atherosclerotic coronary disease and higher risk of incident major adverse cardiac events during the first three years after angiography.

ABBREVIATIONS

ADMA

asymmetric dimethylarginine

CAD

coronary artery disease

GABR

global arginine bioavailability ratio

hsCRP

high-sensitivity C-reactive protein

MACE

major adverse cardiovascular events

MI

myocardial infarction

NO

nitric oxide