ADMA (asymmetric dimethylarginine), an endogenous inhibitor of nitric oxide synthase, is considered a major risk factor for cardiovascular disease and progression of renal disease. In the present study we aim to investigate the effect of acute variations in plasma glucose and insulin on plasma ADMA levels in young people with T1D (Type 1 diabetes). Fifteen young patients (ten males) with T1D, median age 18.3 (13.2–24.4) years, HbA1c (glycated haemoglobin) 9% (6.4–13.6%), underwent an overnight (18:00–08:00 hours) variable insulin infusion for euglycaemia, followed by a hyperinsulinaemic–euglycaemic clamp (08:00–12:00 hours). Blood samples were collected every 15 min for determination of ADMA, SDMA (symmetric dimethylarginine), valine, phenylalanine, arginine, creatinine and glucose. Insulin levels were assessed every 30 min. During the overnight period, glucose levels increased following the evening meal. In response to the protein intake there was a significant increase in ADMA, arginine, valine, phenylalanine and creatinine. For the remaining part of the night, glucose levels progressively decreased reaching 5 mmol/l by 04:00 hours. ADMA and SDMA did not change significantly. During the hyperinsulinaemic clamp, a significant fall in ADMA was observed, from 0.468±0.056 to 0.364±0.050 μmol/l (P<0.001). A significant fall was also found in SDMA, valine, phenylalanine, arginine and the ADMA/SDMA ratio (all P<0.001), but not in creatinine levels. No correlation was found between insulin sensitivity and ADMA. We conclude that acute changes in glycaemia do not significantly affect plasma ADMA levels whereas infusion of insulin significantly reduces ADMA, suggesting an important role for insulin in the regulation of this cardiovascular risk factor.

INTRODUCTION

ADMA (asymmetric dimethylarginine), an endogenous inhibitor of NOS (nitric oxide synthase), is now considered to be an important biomarker in the assessment of cardiovascular risk [1,2]. ADMA and its stereoisomer SDMA (symmetric dimethylarginine) are synthesized during post-translational modification of L-arginine residues by the action of protein-arginine-methyltransferases and subsequently released during proteolysis [3,4]. Early reports of increased ADMA concentrations in chronic renal failure patients on haemodialysis [5,6] and the significant decline in ADMA concentrations post-dialysis suggested an important role for the kidney in the elimination of ADMA. However, it is now recognized that approx. 80–90% of the ADMA released is metabolized to citrulline and dimethylamine by DDAH (dimethylarginine dimethylaminohydrolase) [7,8]. DDAH is expressed in a wide range of tissues [79], but the increase in plasma ADMA observed in liver failure and its rapid normalization post-liver transplantation [10,11] demonstrate the importance of the liver in ADMA metabolism and in maintaining plasma ADMA concentrations. Increased plasma ADMA is consistently linked with poor clinical outcomes [1219], being an independent risk factor for CVD (cardiovascular disease) [1214] and mortality [1517], and a predictor of progression of renal disease [18,19]. Consequently, it is essential that we understand the basic mechanisms resulting in increased plasma ADMA concentrations to enable the development of therapeutic strategies. In non-diabetic subjects there is a significant correlation between plasma ADMA concentrations and insulin resistance [2022]. In contrast, increased ADMA in women with gestational diabetes is not associated with insulin resistance [23]. However, it is possible that in insulin-resistance states the observations may be confounded by increased glucose or differences in insulin levels. Incubation of vascular cells with glucose resulted in a decline in DDAH activity and an accumulation of ADMA [24]. In streptozotocin-induced diabetic rats, plasma ADMA was increased as a result of reduced DDAH activity [25]. In contrast, in other experimental studies, no changes in ADMA [26], or even a decline in its concentrations, were associated with hyperglycaemia [27]. The latter finding could be related to an activation of hepatic DDAH by high glucose levels, an effect recently reported in diabetic rats [28] and which differs from previous findings of an inhibition of DDAH by hyperglycaemia [24,25]. In addition, in vivo, no change in ADMA was measured following a saccharose challenge [29]. In subjects with T1D (Type 1 diabetes), plasma ADMA is increased [3032], whereas in Type 2 diabetes ADMA is reduced and negatively correlated with HbA1c (glycated haemoglobin) [33]. In vitro, insulin reverses the accumulation of ADMA in endothelial cells incubated with TNF-α (tumour necrosis factor-α) by increasing DDAH activity [34] and, in vivo, ADMA is reduced during acute hyperinsulinaemia in non-diabetic subjects [35].

The predictive value of plasma ADMA concentrations is of fundamental importance in subjects with T1D, where the risk of premature heart disease and coronary events remains unacceptably high [36]. In the diabetic population, chronic elevations in ADMA levels have been suggested as a mediator of endothelial dysfunction and the associated risk of developing CVD [14,37]. However, the value of any ‘chronic marker’ of CVD risk is dependent on an understanding of factors, which acutely affect changes in plasma levels. With respect to T1D, as plasma insulin and glucose vary over any 24-h period, it is essential to understand how their acute fluctuations might modulate plasma ADMA levels.

The aim of the present study was to investigate acute changes in ADMA in a well-characterized group of adolescents and young adults with T1D in relation to glucose and insulin levels. Furthermore, we aimed to assess changes in SDMA, arginine and the amino acids valine and phenylalanine in order to determine the relationship between dimethylarginine metabolites and other insulin-sensitive measures of protein catabolism.

MATERIALS AND METHODS

Study population

Plasma samples from 15 subjects (ten males and five females) with T1D collected during a previous study [38] and stored at −80 °C were used in the present study. The aim of the original study was to determine the effect of IGF-I (insulin-like growth factor-I)–IGFBP-3 (IGF-binding protein-3) complex administration on insulin sensitivity in patients with T1D. Subjects were investigated after a 2-day course of IGF-I–IGFBP-3 complex or placebo. In the present study only samples collected after placebo treatment were used to assess 15-min plasma levels of ADMA, SDMA, arginine, creatinine, valine and phenylalanine and 30-min levels of plasma insulin.

Inclusion criteria were T1D of at least 2-year duration or C-peptide-negative, age between 13–25 years, puberty (Tanner) at least stage II, treatment with at least two insulin injections per day, and normal renal and liver function. Exclusion criteria were BMI (body mass index) >30 kg/m2, untreated hypothyroidism, chronic illness, pregnancy and malignancy or recurrent episodes of severe unexplained hypoglycaemia. All subjects underwent an overnight (18:00–08:00 hours) variable insulin infusion to establish and maintain euglycaemia (5 mmol/l), followed by a 4-h two-step hyperinsulinaemic–euglycaemic clamp.

The study was approved by the Cambridge and Northampton Local Research Ethics Committees, and written informed consent was obtained from all subjects and/or from their parents. All overnight studies were performed in Cambridge in the Addenbrooke's Clinical Research Centre, Wellcome Trust Clinical Research Facility, Cambridge, U.K.

Study protocol

Long- and intermediate-acting insulin was withdrawn and substituted with regular soluble insulin injections. Sampling and infusion cannulae were inserted, an intravenous insulin infusion started and subjects were allowed an unrestricted evening meal. The insulin infusion was continued overnight and adjusted in response to 15-min blood glucose determinations in order to achieve glucose concentrations of approx. 5 mmol/l by 04:00 hours. During the period 04:00 to 08:00 hours, the steady-state period, glucose concentrations were maintained at approximately, and did not significantly differ from, 5 mmol/l. Mean plasma free insulin and insulin requirement to maintain euglycaemia [mean insulin infusion per kg of FFM (free-fat mass) (m-units/kg of FFM)] were both used to assess insulin sensitivity during the steady-state period. At 08:00 hours, a 4-h two-step hyperinsulinaemic–euglycaemic clamp was performed. Between 08:00 and 10:00 hours, a low-dose insulin infusion (step one: insulin bolus, 2.8 m-units/kg of FFM; insulin infusion, 0.6 m-units·kg−1 of FFM·min−1) was administered and euglycaemia was maintained by an infusion of 20% dextrose. Between 10:00 and 12:00 hours, the insulin infusion rate was increased (step two: insulin bolus, 7.0 m-units/kg of FFM; insulin infusion, 1.5 m-units·kg−1 of FFM·min−1) and the infusion of 20% dextrose increased, as necessary, to maintain euglycaemia. Blood glucose measurements were performed every 5 min.

The M value was calculated as the mean glucose infusion rate per kg of FFM (mg/kg of FFM) during the steady-state periods of step one (09:30–10:00 hours) and of step two (11:30–12:00 hours), and insulin sensitivity was calculated by dividing the M value by the steady-state insulin levels and expressed as M/I.

Plasma samples were collected every 15 min for the measurement of ADMA, SDMA, creatinine, arginine, valine and phenylalanine. Plasma insulin levels were measured in samples collected every 30 min.

Anthropometry

On arrival, height and weight were determined while the subjects stood in light clothing with a wall-mounted stadiometer and an electronic scale respectively. BMI was calculated as weight divided by the square of the height. Body fat mass and the percentage of body fat were estimated using a bioelectrical impedance monitor (Biostat 1500). FFM was calculated as the total body mass subtracted by the amount of fat mass as determined by bioimpedance.

Biochemical analyses

Plasma ADMA, SDMA, creatinine, arginine, valine and phenylalanine

Plasma ADMA, SDMA, creatinine, arginine, valine and phenylalanine were measured by stable isotope dilution electrospray MS/MS (tandem MS). Plasma (50 μl) was mixed with 50 μl of deionized water containing 50 pmol of [2H6]ADMA, 50 pmol of [2H6]SDMA, 12.5 nmol of [2H3]creatinine, 12.5 nmol of [15N4]arginine, 12.5 nmol of [2H8]valine and 12.5 nmol of [2H5]phenylalanine and proteins were precipitated with 200 μl of acetonitrile (Rathburn Chemicals). The stable isotopes were sourced from CDN Isotopes except for [2H6]ADMA and [2H6]SDMA, which were synthesized (Department of Chemistry, King's College London, London, U.K.). Following mixing and centrifugation for 3 min at 21800 g, supernatants were transferred to a 96-deep-well plate. For measurement of ADMA and SDMA 5 μl of supernatant was automatically injected using an HTS PAL autosampler (CTC Analytics) into a 250 μl/min mobile-phase stream of acetonitrile/water (50:50; v/v) with 0.025% (v/v) formic acid. Chromatography was performed on a Chirobiotic T 100 mm×2.1 mm column with a 2 cm×4.0 mm guard column (Advanced Separation Technologies) and precursor/product ion pairs (m/z 203.1/46.2 and 209.1/52.2 for ADMA and m/z 203.1/172.2 and 209.1/175.1 for SDMA) were acquired in positive-ion multiple reaction monitoring mode using a Sciex API4000 (Applied Biosystems). Assay standardization was based on aqueous standards at 0.25, 1.0 and 5.0 μmol/l ADMA/SDMA stored at −80 °C. Results were calculated using Analyst version 1.4.1. Pooled and spiked plasma samples were used for internal quality control. Intra-assay variation {mean [CV% (coefficient of variation)]; n=10} was 0.37 μmol/l (2.1%) for ADMA and 0.44 μmol/l (3.5%) for SDMA. Inter-assay variation (n=57) was 0.39 μmol/l (6.8%), 1.45 μmol/l (5.5%) and 4.20 μmol/l (4.6%) for ADMA, and 0.41 μmol/l (3.2%), 1.88 μmol/l (2.4%) and 4.41 μmol/l (3.3%) for SDMA. A second injection of the supernatant (1 μl) was used for the measurement of creatinine, arginine, valine and phenylalanine. Solvent flow rate and chromatography were as above. Precursor/product ion pairs (m/z 114.2/44.2 and 117.2/47.2 for creatinine, m/z 175.2/70.1 and 179.2/71.1 for arginine, m/z 118.1/72.0 and 126.1/80.0 for valine, and m/z 165.9/120.1 and 170.9/125.1 for phenylalanine) were acquired in positive-ion multiple reaction monitoring mode. Stock aqueous creatinine standard, 10 mmol/l, was prepared by dissolving creatinine (BDH Chemicals) in 0.1 mol/l HCl and were stored at 4 °C. The concentration of the stock standard has been confirmed using the NIST (National Institute of Standards and Technology) standard reference material creatinine 914a (Laboratory of the Government Chemist, Teddington, U.K.). Assay calibrators at 50, 250 and 1000 μmol/l were prepared from the stock standard by dilution with deionized water. Pooled and spiked quality control samples with assigned values determined using EU (European Union) Community Bureau of Reference certified reference material sera (Report EUR 17115EN) CRM 573 (68.7 μmol/l), 574 (105.0 μmol/l) and 575 (404.1 μmol/l) (Laboratory of the Government Chemist, Teddington, U.K.) were included with each assay. Arginine stock aqueous standard, 10 mmol/l, was prepared from solid (Sigma–Aldrich) and stored at −80 °C. Assay calibrators at 10, 100 and 500 μmol/l were prepared from the stock standard by dilution with deionized water. A physiological fluid amino acid standard preparation (A9906; Sigma–Aldrich) was used for standardization, 25 and 250 μmol/l, of the valine and phenylalanine assays. It provided independent standards at these concentrations for both creatinine and arginine. Pooled and spiked (creatinine and arginine) plasma samples were used for internal quality control. Inter-assay variation (n=56) was 64.8 μmol/l (3.5%), 184 μmol/l (2.4%) and 496 μmol/l (3.1%) for creatinine, and 107 μmol/l (2.5%), 208 μmol/l (1.7%) and 465 μmol/l (2.8%) for arginine. Inter-assay variation (n=34) was 194 μmol/l (2.6%) for valine and 64.0 μmol/l (2.6%) for phenylalanine.

Blood glucose concentrations

Blood glucose concentrations were measured using YSI model 2300 stat plus analyser. The intra-assay CV at 4.1 mmol/l was 1.5%. The equivalent inter-assay CV at this glucose concentration was 2.8% and was 1.7% at 14.1 mmol/l.

Plasma free insulin concentrations

Plasma free insulin concentrations were measured using an ELISA (Dako) according to the manufacturer's instructions. Intra-assay imprecision (CV) was 4.3% at 82 pmol/l (14 m-units/l), 3% at 402 pmol/l (67 m-units/l) and 5.7% at 907 pmol/l (151 m-units/l). Equivalent inter-assay imprecision was 4.3, 5.1 and 5.4% respectively.

Statistical analysis

Statistical analysis was performed using SPSS version 11.5 software for Windows (SPSS). Values are expressed as means±S.D. (or range) and medians (range). P<0.05 was considered significant for all data analysis. ANOVA for repeated measurement was used to assess changes in variables during the overnight period (18:00–04:00 hours) and during the hyperinsulinaemic clamp. Differences in variables between specific time points were analysed using a paired Student's t test. Pearson correlation analysis was used to establish associations between variables of interest.

RESULTS

The clinical characteristics of the study population are summarized in Table 1.

Table 1
Characteristics of the study population
CharacteristicsMean (range)
Age (years) 18.3 (13.2–24.4) 
Gender (n) (male/female) 10/5 
Duration of diabetes (years) 8.8 (0.8–12.0) 
HbA1c (%) 9.0 (6.4–13.6) 
Insulin dose (units·kg−1 of body weight·day−11.1 (0.7–1.6) 
BMI (kg/m222.5 (18.4–25.3) 
Lean body mass (kg) 56.8 (36.6–71.0) 
CharacteristicsMean (range)
Age (years) 18.3 (13.2–24.4) 
Gender (n) (male/female) 10/5 
Duration of diabetes (years) 8.8 (0.8–12.0) 
HbA1c (%) 9.0 (6.4–13.6) 
Insulin dose (units·kg−1 of body weight·day−11.1 (0.7–1.6) 
BMI (kg/m222.5 (18.4–25.3) 
Lean body mass (kg) 56.8 (36.6–71.0) 

Overnight insulin infusion for euglycaemia (18:00–04:00 hours) (Table 2)

In response to the evening meal, the median plasma glucose increased from 11.0 (4.0–20.7) mmol/l to a peak of 13.6 (9.1–23.2) mmol/l at 20:15 hours (P=0.17), before a progressive decline to 5.2 (4.3–6.7) mmol/l by 04:00 hours (Figure 1). The median plasma insulin increased from 99 (62–511) pmol/l to 346 (120–863) pmol/l and was 73 (62–181) pmol/l by 04:00 hours. Plasma ADMA increased from 0.446±0.073 μmol/l to a peak of 0.493±0.075 μmol/l by 20:30 hours (P<0.001), but then remained constant through to 04:00 hours (Figure 1). Significant increases were observed for arginine (36.6±10.8 μmol/l to a peak of 75.9±33.8 μmol/l at 00:00 hours; P=0.001), valine (217±37 μmol/l to a peak of 283±67 μmol/l at 22:30 hours; P=0.001), phenylalanine (65.2±7.0 μmol/l to a peak of 88.8±12.7 μmol/l at 22.30 hours; P<0.001) and creatinine (59.2±9.7 μmol/l to peak, 63.0±9.3 μmol/l at 21:30 hours; P=0.05). All, except creatinine, declined significantly through to 04:00 hours, with phenylalanine at this time being significantly lower (58.8±10.9 μmol/l; P=0.002) than at 18:00 hours. Arginine was still significantly higher at 04:00 hours (52.1±12.3 μmol/l; P=0.001) compared with 18:00 hours. Plasma SDMA and the ADMA/SDMA ratio remained constant over the period 18:00–04:00 hours.

Table 2
Overnight variable insulin infusion for euglycaemia (18:00–04:00 hours)

Values are means±S.D. or medians (range).

TimeP value
18:00 hours20:15 hours*04:00 hours18:00 hours compared with 20:15 hours20:15 hours compared with 04:00 hours18:00 hours compared with 04:00 hours
ADMA (μmol/l) 0.446±0.073 0.475±0.073 0.479±0.058 <0.001 0.23 0.06 
SDMA (μmol/l) 0.411±0.056 0.408±0.062 0.401±0.052 0.80 0.06 0.32 
ADMA/SDMA ratio 1.08 (0.83–1.40) 1.11 (1.0–1.5) 1.17 (0.91–1.58) 0.09 0.90 0.06 
Arginine (μmol/l) 36.6±10.8 47.9±13.5 52.1±12.3 <0.001 0.008 0.001 
Creatinine (μmol/l) 59.2±9.7 61.0±9.8 60.6±9.0 0.34 0.10 0.12 
Valine (μmol/l) 217±37 257±48 239±44 0.003 0.004 0.13 
Phenylalanine (μmol/l) 65.2±7.0 84.2±14.0 58.8±10.9 <0.001 <0.001 0.002 
Glucose (mmol/l) 11.0 (4.0–20.7) 13.6 (9.1–23.2) 5.2 (4.3–6.7) 0.18 <0.001 0.001 
Insulin (pmol/l) 99 (62–511) 346 (120–863) 73 (62–181) <0.001 <0.001 0.05 
TimeP value
18:00 hours20:15 hours*04:00 hours18:00 hours compared with 20:15 hours20:15 hours compared with 04:00 hours18:00 hours compared with 04:00 hours
ADMA (μmol/l) 0.446±0.073 0.475±0.073 0.479±0.058 <0.001 0.23 0.06 
SDMA (μmol/l) 0.411±0.056 0.408±0.062 0.401±0.052 0.80 0.06 0.32 
ADMA/SDMA ratio 1.08 (0.83–1.40) 1.11 (1.0–1.5) 1.17 (0.91–1.58) 0.09 0.90 0.06 
Arginine (μmol/l) 36.6±10.8 47.9±13.5 52.1±12.3 <0.001 0.008 0.001 
Creatinine (μmol/l) 59.2±9.7 61.0±9.8 60.6±9.0 0.34 0.10 0.12 
Valine (μmol/l) 217±37 257±48 239±44 0.003 0.004 0.13 
Phenylalanine (μmol/l) 65.2±7.0 84.2±14.0 58.8±10.9 <0.001 <0.001 0.002 
Glucose (mmol/l) 11.0 (4.0–20.7) 13.6 (9.1–23.2) 5.2 (4.3–6.7) 0.18 <0.001 0.001 
Insulin (pmol/l) 99 (62–511) 346 (120–863) 73 (62–181) <0.001 <0.001 0.05 
*

Corresponding to the time when the glucose peak was reached.

Plasma glucose, insulin, ADMA and SDMA during the overnight period (18:00–04:00 hours)

Figure 1
Plasma glucose, insulin, ADMA and SDMA during the overnight period (18:00–04:00 hours)

Values are means, with error bars representing 1 S.E.M.

Figure 1
Plasma glucose, insulin, ADMA and SDMA during the overnight period (18:00–04:00 hours)

Values are means, with error bars representing 1 S.E.M.

Overnight steady-state period (04:00–08:00 hours) (Table 3)

Euglycaemia was maintained during the overnight steady-state period (04:00–08:00 hours). No significant differences were found in plasma ADMA, SDMA and creatinine between 04:00 and 08:00 hours, whereas valine and phenylalanine were lower at 08:00 hours. Insulin sensitivity, assessed either by mean plasma insulin levels or insulin requirement for euglycaemia, did not correlate with plasma ADMA, SDMA, valine, phenylalanine, arginine or creatinine (all P>0.05).

Table 3
Overnight steady-state period (04:00–08:00 hours)

Values are means±S.D.

Time
04:00 hours08:00 hoursP value
ADMA (μmol/l) 0.479±0.058 0.468±0.056 0.26 
SDMA (μmol/l) 0.401±0.052 0.414±0.046 0.07 
ADMA/SDMA ratio 1.21±0.19 1.14±0.17 0.31 
Arginine (μmol/l) 52.1±12.3 55.7±13.7 0.27 
Creatinine (μmol/l) 60.6±9.0 60.5±9.2 0.89 
Valine (μmol/l) 239±44 213±35 <0.001 
Phenylalanine (μmol/l) 58.8±10.9 52.9±9.8 0.001 
Glucose (mmol/l) 5.2±0.6 4.9±0.5 0.13 
Insulin (pmol/l) 82±32 115±16 0.02 
Time
04:00 hours08:00 hoursP value
ADMA (μmol/l) 0.479±0.058 0.468±0.056 0.26 
SDMA (μmol/l) 0.401±0.052 0.414±0.046 0.07 
ADMA/SDMA ratio 1.21±0.19 1.14±0.17 0.31 
Arginine (μmol/l) 52.1±12.3 55.7±13.7 0.27 
Creatinine (μmol/l) 60.6±9.0 60.5±9.2 0.89 
Valine (μmol/l) 239±44 213±35 <0.001 
Phenylalanine (μmol/l) 58.8±10.9 52.9±9.8 0.001 
Glucose (mmol/l) 5.2±0.6 4.9±0.5 0.13 
Insulin (pmol/l) 82±32 115±16 0.02 

Hyperinsulinaemic–euglycaemic clamp (08:00–12:00 hours) (Table 4)

Euglycaemia was maintained from 08:00 to 12:00 hours. Plasma free insulin increased significantly from 104 (62–296) to 465 (228–574) pmol/l by the end of the clamp. In response (Figure 2), there were significant (P<0.001) decreases in plasma ADMA (by 22%; 0.468±0.056 to 0.364±0.050 μmol/l), SDMA (by 9%; 0.414±0.046 to 0.373±0.040 μmol/l), the ADMA/SDMA ratio [1.11 (0.92–1.49) to 0.95 (0.77–1.31)], arginine (by 23%, 55.7±13.7 to 42.9±10.7 μmol/l), valine (by 33%, 213±35 to 142±21 μmol/l) and phenylalanine (by 29%, 52.9±9.8 to 37.4±8.4 μmol/l). In contrast, plasma creatinine did not change.

Table 4
Hyperinsulinaemic–euglycaemic clamp (08:00–12:00 hours)

Values are means±S.D. or median (range).

Time
08:00 hours10:00 hours12:00 hoursP value
ADMA (μmol/l) 0.468±0.056 0.436±0.065 0.364±0.050 <0.001 
SDMA (μmol/l) 0.414±0.046 0.402±0.047 0.373±0.040 <0.001 
ADMA/SDMA ratio 1.11 (0.92–1.49) 1.06 (1.02–1.22) 0.95 (0.77–1.31) <0.001 
Arginine (μmol/l) 55.7±13.7 55.1±12.7 42.9±10.7 <0.001 
Creatinine (μmol/l) 60.5±9.2 60.9±11.3 59.2±10.7 0.08 
Valine (μmol/l) 213±35 184±29 142±21 <0.001 
Phenylalanine (μmol/l) 52.9±9.8 46.9±9.4 37.4±8.4 <0.001 
Glucose (mmol/l) 4.9 (4.0–5.9) 4.9 (4.8–5.1) 4.8 (3.7–5.0) 0.19 
Insulin (pmol/l) 104 (62–296) 139 (83–674) 465 (228–574) <0.001 
Time
08:00 hours10:00 hours12:00 hoursP value
ADMA (μmol/l) 0.468±0.056 0.436±0.065 0.364±0.050 <0.001 
SDMA (μmol/l) 0.414±0.046 0.402±0.047 0.373±0.040 <0.001 
ADMA/SDMA ratio 1.11 (0.92–1.49) 1.06 (1.02–1.22) 0.95 (0.77–1.31) <0.001 
Arginine (μmol/l) 55.7±13.7 55.1±12.7 42.9±10.7 <0.001 
Creatinine (μmol/l) 60.5±9.2 60.9±11.3 59.2±10.7 0.08 
Valine (μmol/l) 213±35 184±29 142±21 <0.001 
Phenylalanine (μmol/l) 52.9±9.8 46.9±9.4 37.4±8.4 <0.001 
Glucose (mmol/l) 4.9 (4.0–5.9) 4.9 (4.8–5.1) 4.8 (3.7–5.0) 0.19 
Insulin (pmol/l) 104 (62–296) 139 (83–674) 465 (228–574) <0.001 

Plasma ADMA, SDMA, valine and phenylalanine during the hyperinsulinaemic–euglycaemic clamp (08:00–12:00 hours)

Figure 2
Plasma ADMA, SDMA, valine and phenylalanine during the hyperinsulinaemic–euglycaemic clamp (08:00–12:00 hours)

Values are means, with error bars representing 1 S.E.M.

Figure 2
Plasma ADMA, SDMA, valine and phenylalanine during the hyperinsulinaemic–euglycaemic clamp (08:00–12:00 hours)

Values are means, with error bars representing 1 S.E.M.

The pattern of the decrease in valine and phenylalanine was similar to that of ADMA and SDMA (Figure 2). A significant strong correlation was found between mean plasma ADMA concentrations at each time point during the clamp and mean valine (r=0.98; P<0.0001) as well as phenylalanine (r=0.97; P<0.0001) at corresponding time points.

Insulin sensitivity, assessed as the M/I ratio was predictably higher during step two than step one (0.21±0.01 compared with 0.016±0.008 mg·kg−1 of body weight·min−1·pmol−1·l−1; P=0.024). No differences were found in the rate of the decrease of plasma ADMA, SDMA, arginine, valine or phenylalanine between the two steps of the clamp (results not shown). Furthermore, no correlation was found between the M/I ratio and the rate of the decrease of plasma ADMA, SDMA, arginine, valine and phenylalanine during either step one or step two.

DISCUSSION

The present study utilizes samples obtained in a previous study [38] to assess the acute effect of changes in glucose and insulin on plasma levels of ADMA in a group of young patients with T1D. The majority of studies relating plasma ADMA concentrations to physiological changes have been based on a single measurement. In part, this has been due to the technical difficulty, and resultant time and costs, associated with conventional analytical techniques. The use of high-sensitivity MS/MS enables large numbers of samples to be processed economically with accuracy and precision. The present study represents the first investigation of sequential, 15 min, plasma ADMA concentrations in response to controlled significant changes in plasma insulin and glucose levels in a group of young patients with T1D.

The plasma ADMA concentration represents the balance between the rate of production and the rate of removal [8,9]. In acute circumstances the rate of production of ADMA will be, almost exclusively, dependent on the rate of proteolysis and the rate of removal will be dependent, primarily, on DDAH activity [8,9]. However, the production side of the equation will also be a function of dietary intake and a significant amount of the removal of ADMA will rely on renal function [8,9]. The present study allows us to consider all of these components, with some being better controlled than others. Throughout the present study there was no significant change in GFR (glomerular filtration rate), as assessed by plasma creatinine. The increase in plasma creatinine in the early post-prandial phase, before 22:00 hours, is to be expected following an unrestricted meal, as cooked meat contains significant quantities of creatinine, produced during cooking by dehydration of creatine, which is readily absorbed and increases plasma creatinine [39]. The increases in valine, phenylalanine and arginine are the result of dietary protein hydrolysis in the gut and absorption of the resulting amino acids. Since both ADMA and SDMA are likely to be components of dietary proteins, post-prandial increases in ADMA and SDMA might be expected. In the group as a whole, plasma ADMA did increase significantly, by 0.047 μmol/l. This is an interesting observation, as one might expect significant first-pass metabolism of ADMA in the liver. Surprisingly, plasma SDMA did not change. Since SDMA is not metabolized and is almost exclusively removed by renal clearance [40] it might be expected to increase post-prandially. However, the general ratio of ADMA to SDMA in most tissues is approx. 5:1 [41] and, as ADMA only increased by 0.047 μmol/l, SDMA might be expected to increase by 0.010 μmol/l; sufficiently small to be difficult to detect.

In the post-prandial phase, plasma glucose increased to a median value of 13.6 mmol/l and then declined progressively over the next 8 h to 5 mmol/l. Despite these physiologically major changes in plasma glucose during the overnight period, there was virtually no change in the mean plasma ADMA concentration. In this phase, absorption of ADMA is likely to be negligible and GFR, based on plasma creatinine levels, is constant, indicating that DDAH activity would be the predominant factor in determining plasma ADMA concentrations. It might be hypothesized that the rapid post-prandial increase in plasma glucose could have inactivated DDAH and this could explain the initial post-prandial rise in ADMA, but not in SDMA, and the sustained high ADMA levels during the remaining part of the overnight period, as DDAH might require a prolonged period to regain its full activity. The reported inhibitory effects of glucose on DDAH activity in cells incubated with glucose [24,25] and the protective effect of superoxide dismutase co-incubation [24] implies an oxidative mechanism for glucose-mediated increases in ADMA and offers compelling evidence for a hyperglycaemia-driven inhibitory mechanism on DDAH. However, animal studies have reported controversial results on the effect of glucose on DDAH activity. On the one hand, in streptozotocin-induced diabetic rats, increased plasma ADMA and reduced DDAH activity have been associated with hyperglycaemia. On the other hand, in other models of streptozotocin-induced diabetes, ADMA levels were unchanged [26] or even decreased [27]. The latter phenomenon could be explained by an increased DDAH activity in response to hyperglycaemia [28].

During the post-prandial period, there was a significant increase in plasma insulin levels, arising from compensatory increases in the insulin infusion rates associated with rises in glucose levels. In contrast with the significant effect of insulin during the hyperinsulinaemic clamp, the higher post-prandial insulin levels appeared not to affect plasma ADMA. A possible explanation for this observation is that the influx of dietary ADMA overwhelmed systemic clearance and thus masked the insulin effect on ADMA levels.

In sharp contrast with the benign response to glycaemia, hyperinsulinaemic–euglycaemic infusion resulted in a rapid decline in plasma ADMA and SDMA. Since SDMA is being released by proteolysis, is an end-product of metabolism cleared by the kidney [40] and there is no change in GFR, the decline is almost certainly due to reduced proteolysis. In the present small study, the effect on plasma ADMA appears to be proportionately greater, resulting in a significant decline in the ADMA/SDMA ratio. This may simply reflect the higher ADMA, compared with SDMA, content of protein [41,42], but an additional increase in DDAH activity cannot be excluded. Note that the proportional decreases in valine and phenylalanine are even greater, presumably because the insulin-dependent protein synthesis rate is increased relative to the proteolytic rate. Since dimethylarginine residues, in contrast with valine and phenylalanine, are not incorporated in de novo synthesis of proteins and they are produced as part of a post-translational methylation mainly of nuclear proteins [43], this could explain the faster decline in valine and phenylalanine. An alternative hypothesis explaining the reduced levels of dimethylarginine and amino acids during the clamp might be an insulin-mediated suppression of their cellular release or stimulation of their uptake, which might reduce their circulating levels.

The present findings are in agreement with Eid et al. [35], who demonstrated a 9% fall in plasma ADMA during a 90-min euglycaemic clamp in 20 male volunteers with borderline hypertension. They had also previously shown that, in vitro, insulin reverses the accumulation of ADMA in human umbilical vein and coronary artery endothelial cells incubated with TNF-α by increasing DDAH activity [34].

The results from the present study confirm a significant role for insulin in modulating plasma ADMA concentrations but do not help to explain the underlying therapeutic mechanism responsible for the clinical benefit of intensive insulin therapy in patients receiving intensive care [44]. It is seductive to conclude that reducing ADMA results in a favourable increase in NOS activity, but it may be that the fall in plasma ADMA simply reflects a beneficial effect of reducing protein catabolism or stimulating protein synthesis.

In non-diabetic subjects there is a significant correlation between plasma ADMA concentrations and insulin resistance [20,22], and improvement in insulin sensitivity with rosiglitazone or weight loss results in a fall in ADMA [20,22]. In obese and elderly subjects, increased ADMA appears to be related to reduced insulin sensitivity [21]. In contrast, we were unable to demonstrate any correlation between plasma ADMA and insulin sensitivity, assessed either by plasma insulin or insulin requirements to maintain euglycaemia, during the overnight steady-state period. Similarly, the rate of decline of ADMA during the hyperinsulinaemic clamp did not correlate with insulin sensitivity. These results are in agreement with results from women with gestational diabetes [23], where increased ADMA levels were not found to be associated with insulin resistance. The discrepancies between studies might be related to differences in study populations, and/or in their degree of insulin resistance and associated insulin levels, or in the methods used to assess insulin sensitivity. However, based on our present finding of a relevant reduction in methylarginine levels following the acute infusion of a high dose of insulin, it is probable that differences in plasma insulin levels in insulin-resistant subjects might have the main role in influencing the rate of synthesis and/or catabolism of methylarginines. It could be argued that changes in insulin resistance related to different pubertal stages could have been another factor influencing the results of the present study. However, our study population included subjects all in late puberty (stages IV–V), and therefore the effect of differences in insulin resistance related to puberty can be ruled out.

SDMA has been proposed as a new marker of GFR [40,45] and, by competitive inhibition of arginine transport, as a cardiovascular risk factor [46]. In the present study, plasma SDMA was independent of the large changes in glycaemia, but did decline during the hyperinsulinaemic–euglycaemic clamp. The implication is that the SDMA production rate will be dependent on protein turnover; an important consideration in any study using SDMA as an endogenous plasma GFR marker. A limitation of the present study is related to the interpretation of the effect of hyperglycaemia on dimethylarginines, since the best way to explore this effect would have been to perform a hyperglycaemic clamp. Therefore the apparent benign effect of the meal and subsequent changes in glucose levels need to be interpreted with caution. Furthermore, given that we did not specifically assess clearance of creatinine or ADMA, we cannot completely exclude the effect of variations in renal function during the study on plasma ADMA levels.

In conclusion, increases in insulin levels lead to a rapid decline in ADMA levels, which parallel other markers of inhibition of protein catabolism, such as valine and phenylalanine. Acute changes in glucose levels do not appear to alter ADMA levels, although further study is required. These ‘acute effects’ on ADMA need to be considered in the interpretation of long-term studies on the association between ADMA and the risk of complications in subjects with T1D.

Abbreviations

     
  • ADMA

    asymmetric dimethylarginine

  •  
  • BMI

    body mass index

  •  
  • CV

    coefficient of variation

  •  
  • CVD

    cardiovascular disease

  •  
  • DDAH

    dimethylarginine dimethylaminohydrolase

  •  
  • FFM

    free-fat mass

  •  
  • GFR

    glomerular filtration rate

  •  
  • HbA1c

    glycated haemoglobin

  •  
  • IGF-I

    insulin-like growth factor-I

  •  
  • IGFBP-3

    IGF-binding protein-3

  •  
  • MS/MS

    tandem MS

  •  
  • NOS

    nitric oxide synthase

  •  
  • SDMA

    symmetric dimethylarginine

  •  
  • T1D

    Type 1 diabetes

  •  
  • TNF-α

    tumour necrosis factor-α

The present study was supported in part by an ESPE (European Society for Paediatric Endocrinology) Research Fellowship (to M. L. M.), sponsored by Novo Nordisk A/S.

References

References
1
Siroen
 
M. P.
Teerlink
 
T.
Nijveldt
 
R. J.
Prins
 
H. A.
Richir
 
M. C.
 
The clinical significance of asymmetric dimethylarginine
Annu. Rev. Nutr.
2006
, vol. 
26
 (pg. 
203
-
228
)
2
Böger
 
R. H.
 
Asymmetric dimethylarginine (ADMA): a novel risk marker in cardiovascular medicine and beyond
Ann. Med.
2006
, vol. 
38
 (pg. 
126
-
136
)
3
Kakimoto
 
Y.
Akazawa
 
S.
 
Isolation and identification of NG, NG- and NG, N′G-dimethyl-arginine, N-ϵ-mono-, di-, and trimethyllysine, and glucosylgalactosyl- and galactosyl-δ-hydroxylysine from human urine
J. Biol. Chem.
1970
, vol. 
245
 (pg. 
5751
-
5758
)
4
McDermott
 
J. R.
 
Studies on the catabolism of NG-methylarginine, NG, N′G-dimethylarginine and NG, NG-dimethylarginine in the rabbit
Biochem. J.
1976
, vol. 
154
 (pg. 
179
-
184
)
5
Vallance
 
P.
Leone
 
A.
Calver
 
A.
Collier
 
J.
Moncada
 
S.
 
Accumulation of an endogenous inhibitor of NO synthesis in chronic renal failure
Lancet
1992
, vol. 
339
 (pg. 
572
-
575
)
6
Wahbi
 
N.
Dalton
 
R. N.
Turner
 
C.
Denton
 
M.
Abbs
 
I.
Swaminathan
 
R.
 
Dimethylarginines in chronic renal failure
J. Clin. Pathol.
2001
, vol. 
54
 (pg. 
470
-
473
)
7
MacAllister
 
R. J.
Parry
 
H.
Kimoto
 
M.
, et al 
Regulation of nitric oxide synthesis by dimethylarginine dimethylaminohydrolase
Br. J. Pharmacol.
1996
, vol. 
119
 (pg. 
1533
-
1540
)
8
Tran
 
C. T.
Leiper
 
J. M.
Vallance
 
P.
 
The DDAH/ADMA/NOS pathway
Atherscler. Suppl.
2003
, vol. 
4
 (pg. 
33
-
40
)
9
Vallance
 
P.
Leiper
 
J.
 
Cardiovascular biology of the asymmetric dimethylarginine: dimethylarginine dimethylaminohydrolase pathway
Arterioscler. Thromb. Vasc. Biol.
2004
, vol. 
24
 (pg. 
1023
-
1030
)
10
Mookerjee
 
R. P.
Dalton
 
R. N.
Davies
 
N. A.
, et al 
Inflammation is an important determinant of levels of the endogenous NOS inhibitor asymmetric dimethylarginine (ADMA) in acute liver failure
Liver Transplant.
2007
, vol. 
13
 (pg. 
400
-
405
)
11
Siroen
 
M. P.
Warlé
 
M. C.
Teerlink
 
T.
, et al 
The transplanted liver graft is capable of clearing asymmetric dimethylarginine
Liver Transplant.
2004
, vol. 
10
 (pg. 
1524
-
1530
)
12
Schnabel
 
R.
Blankenberg
 
S.
Lubos
 
E.
, et al 
Asymmetric dimethylarginine and the risk of cardiovascular events and death in patients with coronary artery disease: results from the Athero Gene Study
Circ. Res.
2005
, vol. 
97
 (pg. 
53
-
59
)
13
Schulze
 
F.
Lenzen
 
H.
Hanefeld
 
C.
, et al 
Asymmetric dimethylarginine is an independent risk factor for coronary heart disease: results from the multicenter Coronary Artery Risk Determination investigating the Influence of ADMA Concentration (CARDIAC) study
Am. Heart J.
2006
, vol. 
493
 (pg. 
1
-
8
)
14
Kryzanowska
 
K.
Mittermayer
 
F.
Wolzt
 
M.
Schernthaner
 
G.
 
Asymmetric dimethylarginine predicts cardiovascular events in patients with type 2 diabetes
Diabetes Care
2007
, vol. 
30
 (pg. 
1834
-
1839
)
15
Nijveldt
 
R. J.
Teerlink
 
T.
Van Der Hoven
 
B.
, et al 
Asymmetrical dimethylarginine (ADMA) in critically ill patients: high plasma ADMA concentration is an independent risk factor of ICU mortality
Clin. Nutr.
2003
, vol. 
22
 (pg. 
23
-
30
)
16
Mookerjee
 
R. P.
Malaki
 
M.
Davies
 
N. A.
, et al 
Increasing dimethylarginine levels are associated with adverse clinical outcome in severe alcoholic hepatitis
Hepatology
2007
, vol. 
45
 (pg. 
62
-
71
)
17
Zoccali
 
C.
Bode-Böger
 
S. M.
Mallamaci
 
F.
, et al 
Plasma concentration of asymmetrical dimethylarginine and mortality in patients with end-stage renal disease: a prospective study
Lancet
2001
, vol. 
358
 (pg. 
2113
-
2117
)
18
Fliser
 
D.
Kronenberg
 
F.
Kielstein
 
J. T.
, et al 
Asymmetric dimethylarginine and progression of chronic kidney disease: the mild to moderate kidney disease study
J. Am. Soc. Nephrol.
2005
, vol. 
16
 (pg. 
2456
-
2461
)
19
Ravani
 
P.
Tripepi
 
G.
Malberti
 
F.
Testa
 
S.
Mallamaci
 
F.
Zoccali
 
C.
 
Asymmetrical dimethylarginine predicts progression to dialysis and death in patients with chronic kidney disease: a competing risks modeling approach
J. Am. Soc. Nephrol.
2005
, vol. 
16
 (pg. 
2449
-
2455
)
20
Stuhlinger
 
M. C.
Abbasi
 
F.
Chu
 
J. W.
, et al 
Relationship between insulin resistance and an endogeneous nitric oxide synthase inhibitor
JAMA, J. Am. Med. Assoc.
2002
, vol. 
287
 (pg. 
1420
-
1426
)
21
Marliss
 
E. B.
Chevalier
 
S.
Gougeon
 
R.
, et al 
Elevations of plasma methylarginines in obesity and ageing are related to insulin sensitivity and rates of protein turnover
Diabetologia
2006
, vol. 
49
 (pg. 
351
-
359
)
22
McLaughlin
 
T.
Stuhlinger
 
M.
Lamendola
 
C.
, et al 
Plasma asymmetric dimethylarginine concentrations are elevated in obese insulin-resistant women and fall with weight loss
J. Clin. Endocrinol. Metab.
2006
, vol. 
91
 (pg. 
1896
-
1900
)
23
Mittermayer
 
F.
Mayer
 
B. X.
Meyer
 
A.
, et al 
Circulating concentrations of asymmetrical dimethyl-L-arginine are increased in women with previous gestational diabetes
Diabetologia
2002
, vol. 
45
 (pg. 
1372
-
1378
)
24
Sorrenti
 
V.
Mazza
 
F.
Campisi
 
A.
Vanella
 
L.
Li Volti
 
G.
Di Giacomo
 
C.
 
High glucose-mediated imbalance of nitric oxide synthase and dimethylarginine dimethylaminohydrolase expression in endothelial cells
Curr. Neurovasc. Res.
2006
, vol. 
3
 (pg. 
49
-
54
)
25
Lin
 
K. Y.
Ito
 
A.
Asagami
 
T.
, et al 
Impaired nitric oxide synthase pathway in diabetes mellitus: role of asymmetric dimethylarginine and dimethylarginine dimethylaminohydrolase
Circulation
2002
, vol. 
106
 (pg. 
987
-
992
)
26
Palm
 
F.
Buerk
 
D. G.
Carlsson
 
P. O.
Hansell
 
P.
Liss
 
P.
 
Reduced nitric oxide concentration in the renal cortex of streptozotocin-induced diabetic rats: effects on renal oxygenation and microcirculation
Diabetes
2005
, vol. 
54
 (pg. 
3282
-
3287
)
27
Palm
 
F.
Friederich
 
M.
Carlsson
 
P. O.
Hansell
 
P.
Teerlink
 
T.
Liss
 
P.
 
Reduced nitric oxide in diabetic kidney due to increased hepatic arginine metabolism: implications for renomedullary oxygen availability
Am. J. Physiol. Renal. Physiol.
2008
, vol. 
294
 (pg. 
F30
-
F37
)
28
Chobanyan
 
K.
Thum
 
T.
Suchy
 
M.
, et al 
GC-MS assay for hepatic DDAH activity in diabetic and non-diabetic rats by measuring dimethylamine (DMA) formed from asymmetric dimethylarginine (ADMA): evaluation of the importance of S-nitrosothiols as inhibitors of DDAH activity in vitro and in vivo in humans
J. Chromatogr. B Analyt. Technol. Biomed. Life Sci.
2007
, vol. 
858
 (pg. 
32
-
41
)
29
Washer
 
T. C.
Schoelzer
 
I.
Wiegratz
 
A.
, et al 
Reduction of postchallenge hyperglycaemia prevents acute endothelial dysfunction in subjects with impaired glucose tolerance
Eur. J. Clin. Invest.
2005
, vol. 
35
 (pg. 
551
-
557
)
30
Tarnow
 
L.
Hovind
 
P.
Teerlink
 
T.
Stehouwer
 
C. D.
Parving
 
H. H.
 
Elevated plasma asymmetric dimethylarginine as a marker of cardiovascular morbidity in early diabetic nephropathy in type 1 diabetes
Diabetes Care
2004
, vol. 
27
 (pg. 
765
-
769
)
31
Eid
 
H. M.
Eritsland
 
J.
Larsen
 
J.
Arnesen
 
H.
Seljeflot
 
I.
 
Increased levels of asymmetric dimethylarginine in populations at risk for atherosclerotic disease. Effect of pravastatin
Atherosclerosis
2003
, vol. 
166
 (pg. 
279
-
284
)
32
Altinova
 
A. E.
Arslan
 
M.
Sepici-Dincel
 
A.
Akturk
 
M.
Altan
 
N.
Touruner
 
F. B.
 
Uncomplicated type 1 diabetes is associated with increased asymmetric dimethylarginine concentrations
J. Clin. Endocrinol. Metab.
2007
, vol. 
92
 (pg. 
1881
-
1885
)
33
Paiva
 
H.
Lehtimaki
 
T.
Laakso
 
J.
, et al 
Plasma concentrations of asymmetric-dimethyl-arginine in type 2 diabetes is associated with glycemic control and glomerular filtration rate but not with risk factors of vasculopathy
Metab. Clin. Exp.
2003
, vol. 
52
 (pg. 
303
-
307
)
34
Eid
 
H. M.
Lyberg
 
T.
Arnesen
 
H.
Seljeflot
 
I.
 
Insulin and adiponectin inhibit the TNFα-induced ADMA accumulation in human endothelial cells. The role of DDAH
Atherosclerosis
2007
, vol. 
194
 (pg. 
1
-
8
)
35
Eid
 
H. M. A.
Reims
 
H.
Annesen
 
H.
Kjeldsen
 
S. E.
Lyberg
 
T.
Seljeflot
 
I.
 
Decreased levels of asymmetric dimethylarginine during acute hyperinsulinemia
Metab. Clin. Exp.
2007
, vol. 
56
 (pg. 
464
-
469
)
36
Orchard
 
T. J.
Costacou
 
T.
Kretowski
 
A.
Nesto
 
R. W.
 
Type 1 diabetes and coronary artery disease
Diabetes Care
2006
, vol. 
29
 (pg. 
2528
-
2538
)
37
Lajer
 
M.
Tarnow
 
L.
Jarsal
 
A.
Teerlink
 
T.
Parving
 
H. H.
Rossing
 
P.
 
Plasma concentration of asymmetric dimethylarginine (ADMA) predicts cardiovascular morbidity and mortality in type 1 diabetic patients with diabetic nephropathy
Diabetes Care
2008
, vol. 
31
 (pg. 
747
-
752
)
38
Saukkonen
 
T.
Amin
 
R.
Williams
 
R. M.
, et al 
Dose-dependent effects of recombinant human insulin like growth factor (IGF)-I/IGF binding protein-3 complex on overnight growth hormone secretion and insulin sensitivity in type 1 diabetes
J. Clin. Endocrinol. Metabol.
2004
, vol. 
89
 (pg. 
4634
-
4641
)
39
Preiss
 
D. J.
Godber
 
I. M.
Lamb
 
E. J.
Dalton
 
R. N.
Gunn
 
I. R.
 
The influence of a cooked meat meal on estimated glomerular filtration rate
Ann. Clin. Biochem.
2007
, vol. 
44
 (pg. 
35
-
42
)
40
Kielstein
 
J. T.
Salpeter
 
S. R.
Bode-Boeger
 
S. M.
Scalera
 
F.
Cooke
 
J. P.
Fliser
 
D.
 
Symmetric dimethylarginine (SDMA) an endogenous marker of renal function – a meta-analysis
Nephrol. Dial. Transplant.
2006
, vol. 
21
 (pg. 
2446
-
2451
)
41
Bulau
 
P.
Zakrzewicz
 
D.
Kitowska
 
K.
Wardega
 
B.
Kreuder
 
J.
Eickelberg
 
O.
 
Quantitative assessment of arginine methylation in free versus protein-incorporated amino acids in vitro and in vivo using protein hydrolysis and high-performance liquid chromatography
BioTechniques
2006
, vol. 
40
 (pg. 
305
-
310
)
42
Al Banchaabouchi
 
M.
Marescau
 
B.
Possemiers
 
I.
D'Hooge
 
R.
Levillain
 
O.
De Deyn
 
P. P.
 
NG,NG-Dimethylarginine. and NG, N′G-dimethylarginine in renal insufficiency
Eur. J. Physiol.
2000
, vol. 
439
 (pg. 
524
-
531
)
43
Pahlich
 
S.
Zakaryan
 
R. P.
Gehring
 
H.
 
Protein arginine methylation. Cellular functions and methods of analysis
Biochim. Biophys. Acta
2006
, vol. 
1764
 (pg. 
1890
-
1903
)
44
Siroen
 
M. P.
van Leeuwen
 
P. A.
Nijveldt
 
R. J.
Teerlink
 
T.
Wouters
 
P. J.
Van den Berghe
 
G.
 
Modulation of asymmetric dimethylarginine in critically ill patients receiving intensive insulin treatment: a possible explanation of reduced morbidity and mortality?
Crit. Care Med.
2005
, vol. 
33
 (pg. 
504
-
510
)
45
Dalton
 
R. N.
Booth
 
C.
Mallard
 
B.
Rigby
 
E.
Rigden
 
S.
Turner
 
C.
 
SDMA is an endogenous marker of glomerular filtration rate
J. Am. Soc. Nephrol.
2005
, vol. 
16
  
(Abstract)
46
Bode-Boeger
 
S. M.
Scalera
 
F.
Kielstein
 
J. T.
, et al 
Symmetrical dimethylarginine: a new combination parameter for renal function and the extent of coronary artery disease
J. Am. Soc. Nephrol.
2006
, vol. 
17
 (pg. 
1128
-
1134
)