N U TR I TION RE S E ARCH 3 5 ( 2 0 15 ) 1 9 8–2 05
Available online at www.sciencedirect.com
ScienceDirect www.nrjournal.com
Original Research
Single-dose oral guanidinoacetic acid exhibits dosedependent pharmacokinetics in healthy volunteers Sergej M. Ostojic a, b,⁎, Aleksandra Vojvodic-Ostojic c a b c
Exercise Physiology Laboratory, Center for Health, Exercise and Sport Sciences, Belgrade, Serbia Biomedical Sciences Deparment, Faculty of Sport and Physical Education, University of Novi Sad, Novi Sad, Serbia Environmental Engineering Department, Faculty of Technology and Metallurgy, University of Belgrade, Belgrade, Serbia
ARTI CLE I NFO
A BS TRACT
Article history:
Guanidinoacetic acid (GAA), the natural precursor of creatine, has potential as a dietary
Received 16 September 2014
supplement for human nutrition, yet no data are available regarding its dose-dependent
Revised 5 December 2014
pharmacokinetic (PK) behavior. We hypothesized that a single dose of orally administered
Accepted 30 December 2014
GAA exhibited dose-dependent PK behavior in healthy volunteers. Forty-eight young adults were enrolled in a randomized, placebo-controlled, double-blind, parallel-group trial to
Keywords:
receive single oral doses of GAA (1.2, 2.4, and 4.8 g) or a placebo. Pharmacokinetic metrics for
Guanidinoacetic acid
plasma GAA and creatine were assessed immediately before (0 hours) and at 1, 2, 4, 6, 8, 12,
Pharmacokinetics
and 24 hours after GAA ingestion. The lag time appeared to be similar after the bolus
Area under the curve
ingestion of GAA (0.14 ± 0.17 hours for low-dose GAA, 0.31 ± 0.18 hours for medium-dose
Elimination half-life
GAA, and 0.38 ± 0.32 hours for high-dose GAA; P = .05). An increase in the area under the
Clearance
concentration-time curve for plasma GAA was found for the dose range tested, with 2.4and 9.3-fold increases in the area under the concentration-time curve for every 2-fold increase in the GAA dose (P < .0001). No differences were found for elimination half-time between the low-dose and medium-dose groups (2.1 hours) for the high-dose GAA regimen (P = .001). The volume of distribution was affected by the dosage of GAA applied (102.6 ± 17.3 L for lowdose GAA, 97.5 ± 15.7 L for medium-dose GAA, and 61.1 ± 12.7 L for high-dose GAA; P < .0001). Ingestion of GAA elevated plasma creatine by 80%, 116%, and 293% compared with the placebo for the 1.2, 2.4, and 4.8 g doses, respectively (P < .0001). Guanidinoacetic acid single-dose PK metrics were nonlinear with respect to dose size. Across the dose range of 1.2 to 4.8 g, systemic exposure to GAA increased in a greater than dose-proportional manner. © 2015 Elsevier Inc. All rights reserved.
1.
Introduction
Guanidinoacetic acid (GAA; also known as glycocyamine or guanidinoacetate) is a nitrogenous organic acid that occurs naturally in the human body. The primary biological role of
GAA is to serve as a direct precursor of creatine, an energy carrier, and mediator in the cell [1]. Guanidinoacetic acid is synthesized endogenously from L-arginine and glycine, mainly in the kidney [2]. Under certain circumstances (eg, kidney failure, exercise-related GAA depletion, and deficient diet),
Abbreviations: AUC, area under the concentration-time curve; GAA, guanidinoacetic acid; PK, pharmacokinetic. ⁎ Corresponding author. Center for Health, Exercise and Sport Sciences, Deligradska 27, Belgrade 11000, Serbia. Tel./fax: +381 11 2643 242. E-mail address: [email protected] (S.M. Ostojic). http://dx.doi.org/10.1016/j.nutres.2014.12.010 0271-5317/© 2015 Elsevier Inc. All rights reserved.
N U T RI TI O N RE S E ARCH 3 5 ( 2 0 15 ) 1 9 8–2 05
supplemental GAA may be “semiessential,” optimizing the body's energy use when the endogenous synthesis of GAA is limited [3]. Because GAA is more bioavailabile, stable, and cost-effective than creatine [4], it might be more suitable for use in human nutrition. In particular, GAA is more soluble in water then creatine (3600 mg/L for GAA vs 1330 mg/L for creatine). Guanidinoacetic acid is highly stable [5], whereas creatine is rapidly metabolized into creatinine when it comes into contact with water at elevated temperatures and low pH [6]. Finally, the cost of GAA powder is approximately 40% lower than the cost of creatine [7,8]. Although the first human use of supplemental GAA was reported more than 60 years ago, there is a little information available on its turnover, bioavailability, or elimination. Previous studies of dietary GAA reported its “creatine recovery effect” in both athletes and clinical patients. Studies have found that orally ingested GAA administered over the course of 6 weeks increases plasma GAA and creatine by up to 50% in male and female collegiate athletes, with no major adverse effects reported [9]. Studies have shown that supplemental GAA may increase homocysteine formation [10], although the effects of increased homocysteine on health status seem to be clinically insignificant [9], and can be corrected by homocysteine removal mechanisms [11]. Dietary GAA has been found to be beneficial for patients with heart disease, neuromuscular disorders, and chronic renal failure [12]. Scientific research has been limited with regard to the behavior of dietary GAA between the time of ingestion and subsequent elimination from the human body. A Japanese group was the first to describe the timeline of plasma concentrations and urinary excretions of GAA in humans after a single oral dose of 1 g of GAA [13]. The study suggested that GAA was absorbed well, with peak plasma GAA occurring after 2 hours (~600 μg/dL) and extra GAA excreted in the urine at a rate of up to 40 mg/h. We recently reported that an acute oral dose of GAA has a notable effect on plasma GAA, creatine, creatinine, total homocysteine, and excretion of urinary GAA and creatine in 20 healthy volunteers [3]. In our study, supplemental GAA was found to be readily bioavailable and rapidly transformed to creatine within the first 2 hours. Neither study described relevant pharmacokinetic (PK) metrics (eg, lag time, volume of distribution, elimination half-life, and clearance) after GAA ingestion, and no studies have examined the dose-dependent PK of oral GAA. Exploratory PK studies with increasing doses of GAA are needed to describe its biochemical behavior and to decide whether GAA has scientific merit for further development as a dietary supplement for humans. This study tested the hypothesis that single-dose, orally administered GAA would exhibit dose-dependent PK in healthy men and women. To test this hypothesis, we conducted a double-blind, placebo-controlled, randomized, parallel-group intervention trial, and evaluated PK metrics for plasma GAA and creatine to assess the acute effects of 1.2, 2.4, and 4.8 g GAA loads in healthy men and women. The study was designed as a phase I, first-in-humans trial to establish whether oral GAA behaves in human subjects as preclinical studies would suggest. Assessment of oral GAA dose escalation and its absorption, distribution, biotransformation, and elimination is a necessary step to clinically evaluate GAA as a potentially useful health intervention.
2.
Methods and materials
2.1.
Study population
199
Candidates for inclusion in the study were moderately physically active men and women between 20 and 25 years old. Potential participants were not admitted to the study if they met any of the following criteria: (1) a history of metabolic disease, (2) known heart disease, (3) use of any performance-enhancing drugs or dietary supplements within the 60 days before the study commenced, (4) smoking, and (5) pregnancy. Participants were fully informed both verbally and in writing about the nature and demands of the study and the known health risks, and they provided informed consent regarding their voluntary participation in the study. The participants completed a health history questionnaire covering cardiovascular disease risk factors, history and present status regarding any signs and symptoms suggestive of cardiovascular disease, history of chronic illnesses, history of surgeries and hospitalizations, history of any musculoskeletal or joint injuries, past and present habits that could affect health, current medication use, family health history, and other health history information [14]. All participants completed routine prescreening including blood and urine profiling, and they received a general medical examination during the initial recruitment. If any specific markers (eg, liver and muscle enzymes, kidney function) were above the reference values, subjects were excluded from the study. Upon initial recruitment, 48 participants (n = 48; 24 men and 24 women) met the criteria to participate in the study; the total number of participants fulfilled the optimal sample size (see below). The mean physical characteristics of participants were as follows: age 22 ± 2 years, weight 71 ± 14 kg, and height 176 ± 10 cm. Participants' nutrition was monitored before the experiment began using a standardized questionnaire [15] and computed using NutriBase software (CyberSoft Inc, Phoenix, AZ, USA). Approval of the local institutional review board (No. 05/A-2010/014) was obtained, and all the procedures performed were in accordance with the Declaration of Helsinki and the principles set forth in the Guidelines for Good Clinical Practice [16].
2.2.
Experimental protocol
Participants were randomized according to a computer-generated list in a double-blind, parallel-group design to receive single oral doses of GAA (1.2, 2.4, and 4.8 g) or a placebo (inulin). The GAA dosages chosen were the equimolar equivalent to dietary creatine amounts that appear to be therapeutically effective [17]. The placebo was administered as a control treatment to monitor the changes in GAA and creatine plasma concentrations during the day, especially the effects of creatine-containing foods on plasma concentrations. Twelve participants were allocated to each intervention group, with women having an equal probability of assignment to the groups. Groups were matched for participants' age, weight, and daily energy intakes. The participants received standardized meals on the day before the experiment. In the 24 hours before the tests, the subjects did not participate in any prolonged exercise and were not permitted to drink alcoholic or caffeinated beverages. On the day of the experiment, each participant arrived at the laboratory at 9:00 AM
200
N U TR I TION RE S E ARCH 3 5 ( 2 0 15 ) 1 9 8–2 05
after an overnight fast of between 10 and 12 hours. Participants provided a fasting blood sample (~5 mL) from a radial artery into heparinized tubes (Greiner Bio-One, Basel, Switzerland) to determine the basal level of GAA and creatine. After baseline sampling, participants received the appropriate oral dose of GAA or a placebo in soft capsules (AlzChem AG, Trostberg, Germany), with 0.5 L of tap water consumed within approximately 1 minute. After the ingestion, blood samples were collected at 1, 2, 4, 6, 8, 12, and 24 hours for quantification of plasma kinetics for GAA and creatine. Participants were sitting or lying down during most of the study period, and standardized meals (Table 1) were provided for all participants at the same time after GAA administration. Breakfast, lunch, snack, and dinner were provided according to a schedule (~11 AM, ~ 1 PM, ~ 3 PM, and ~7 PM, respectively). No creatine-rich foods (eg, meat, fish, and milk) were offered to avoid influencing the metabolites studied. After collection, blood samples were centrifuged, and the plasma was separated and stored at −20°C until the time of assay. Plasma GAA and creatine were assayed as described previously [18] using high-performance liquid chromatography (HewlettPackard, Palo Alto, CA, USA). All samples for each subject were assayed in the same run. For all values, the first reading was discarded, and the study used the mean of the next 3 consecutive readings with a coefficient of variation below 15%.
2.3.
Pharmacokinetic analyses
The concentration-time profiles for plasma GAA were fitted using a 1-compartment model for extravascular drug administration. Nonlinear fitting analysis of the compartmental model was performed using the PK add-in program PK Solver [19]. For data fitting, the Gauss-Newton algorithm was used, and initial parameters were obtained using the curvestripping technique [20]. The data were weighted by the inverse of the estimated measurement error variance. The weighted least-squares method was applied as the criteria for the best fit. Graphs of predicted concentrations vs observed data were visually examined, and values that fit poorly were excluded from further analysis. Quality assessment of the compartmental PK model for the individual's concentrationtime data was based on visual analysis, predicted concentrations vs time plots, residuals vs predicted concentrations plots, the precision of parameter estimates as correlation coefficients, and the weighted sum of squares of residuals. The 1-compartment open model with lag time for oral GAA administration assumed a first-order process of absorption and a first-order process of elimination. The following
Table 1 – Specification of meals for GAA PKs study Meal
Serving Total Carbohydrates Lipids Proteins size (g) energy (%) (%) (%) intake (kcal)
Breakfast Lunch Snack Dinner
340 546 25 665
504 705 111 558
63.5 56.1 68.0 72.5
18.2 15.9 16.6 12.7
18.3 28.0 15.4 14.8
equation describes the rate of change in GAA concentrations: C¼
FD ka exp−kel ðt−tlag Þ − exp−ka ðt−tlag Þ V ka −kel
where C is the plasma GAA concentration minus the baseline concentration (μmol/L); t is the time of oral administration (l-h−1); tlag is the lag time (h−1); F is the bioavailability; D is the dose (μmol); V is the volume of distribution (L); ka is the absorption-rate constant (h−1); and kel is the elimination-rate constant (h−1). In this equation, ka, kel, and tlag are primary parameters from which the other PK parameters were derived. The area under the concentration-time curve (AUC) was calculated using the linear-log trapezoidal method. Because the bioavailability (F) was not measured, only the apparent volume of distribution (Vd) and the apparent clearance (CL) were calculated using the following equations: Vd = V/F and CL = kel • Vd.
2.4.
Statistical analyses
Allowing for 80% power, detection of a 5% difference in peak plasma concentration of GAA between groups required an estimated 10 participants per group. This estimation was adjusted to 12 subjects per group to account for a predicted 20% dropout rate. All results were expressed as mean ± SD. Pharmacokinetic data for each group were tested using the Shapiro-Wilk test for the normality of distribution and Bartlett test for the homogeneity of the variances. After homogenous variances were verified for normally distributed data, plasma kinetic data were compared using 1-way analysis of variance (ANOVA), and the post-hoc Tukey HSD test was used to identify the differences between individual sample pairs. When nonhomogenous variances were identified, mean PK data were compared using the Kruskal-Wallis test. The GamesHowell post hoc test was then used to evaluate whether differences between any 2 groups were significant. A 2-way ANOVA design for repeated measures (treatment vs time) was used to uncover any significant differences between interventions during the 24-hour experiment. The relationship between the AUC and the dose of GAA was examined using Pearson product-moment correlation coefficient. The significance level was set at P < .05. Data were analyzed using the version 21.0 SPSS program (SPSS Inc, Chicago, IL, USA).
3.
Results
No volunteers reported any cardiovascular disease risk factors, history of chronic illnesses, or other relevant health information before intervention. Prior to the experiment, volunteers reported daily energy intakes of 3508 ± 768 kcal, composed of 63.5% ± 4.9% carbohydrates, 23.2% ± 4.8% lipids, and 13.3% ± 3.0% proteins (1.5 ± 0.4 g of proteins per kg body mass). All volunteers (n = 48) completed the PK studies, and no serious protocol violations were identified. No volunteers reported any adverse effects from the GAA supplementation. The mean basal plasma GAA concentration was 2.9 ± 0.3 μmol/L, with no differences found between groups for baseline plasma GAA (P = .53), as shown in Fig. 1. No changes
201
N U T RI TI O N RE S E ARCH 3 5 ( 2 0 15 ) 1 9 8–2 05
in serum GAA or creatine were reported after administration of the placebo throughout the study (P > .05). Two-way ANOVA analysis with repeated measures revealed a significant interaction between time and treatments (P > .0001), indicating different time-courses dependent on the GAA dosages administered. Post hoc analyses showed that differences occurred during both the absorption and elimination phases. After bolus ingestion of GAA, plasma concentration increased sharply to reach its peak in about 90 minutes for both the low-dose and medium-dose groups, whereas the peak concentration occurred 2.2 hours after participants consumed high-dose GAA (Table 2). The peak concentration was highest when high-dose GAA was ingested (418.4 ± 89.6 μmol/L), compared with mediumdose (151.3 ± 17.5 μmol/L) and low-dose (56.9 ± 9.2 μmol/L) administration. Lag time appeared to be similar in all groups after the bolus ingestion of GAA (P = .05). The mean absorption half-time was comparable in all GAA groups (P = .07). The absorption half-time in the low-dose and medium-dose groups was less than 40 minutes, whereas it reached almost 60 minutes with high-dose GAA. The AUC from zero to infinity, corrected for baseline GAA values, revealed significant differences between the 3 GAA dosages (P < .0001). No differences were found for elimination half-time between the low-dose and medium-dose groups (2.1 hours) for the high-dose GAA regimen (P = .001). The apparent clearance and apparent volume of distribution were affected by the GAA dosage; in the low-dose group, clearance and distribution volume were greater than in the medium-dose group, which in turn showed higher volume and clearance than the highdose group (P < .0001). Finally, a strong relationship was found between the AUC and the dose of GAA administered, as demonstrated by Pearson product-moment correlation
Table 2 – PKs of GAA after single-dose ingestion of GAA in different dosages 1.2 g of GAA (n = 12) Peak concentration of plasma GAA (μmol/L) Time to peak concentration (h) Lag time (h) Velocity constant of absorption (h−1) Absorption half-time (h) AUC (0 → ∞, μmol/L * h) Apparent volume of distribution (L) Velocity constant of elimination (h−1) Elimination half-time (h) Apparent systemic clearance (L/h)
2.4 g of GAA (n = 12)
4.8 g of GAA (n = 12)
56.9 ± 9.2
151.3 ± 17.5 a
418.4 ± 89.6 a,b
1.33 ± 0.49
1.33 ± 0.49
2.17 ± 0.58 a,b
0.14 ± 0.17 1.08 ± 0.21
0.31 ± 0.18 1.04 ± 0.08
0.38 ± 0.32 1.22 ± 1.21
0.66 ± 0.13
0.67 ± 0.06
0.93 ± 0.44
224.8 ± 38.0
531.4 ± 64.1 a
2092.4 ±453.2 a,b
102.6 ± 17.3
97.5 ± 15.7
61.1 ± 12.7 a,b
0.46 ± 0.07
0.41 ± 0.07
0.34 ± 0.06 a,b
1.54 ± 0.26
1.74 ± 0.30
2.10 ± 0.36 a,b
46.9 ± 8.3
39.1 ± 4.6 a
20.5 ± 4.4 a,b
Values are means ± SD. Tests of significance between treatment groups are based on 3sample unpaired ANOVA and Kruskal-Wallis model with post hoc tests (P < .05). a Significantly different from 1.2 g of GAA. b Significantly different from 2.4 g of GAA.
coefficient of r = 0.98 (P < .0001), and the trend line was best fitted exponentially: AUC for plasma GAA ð0→∞; μmol=L hÞ ¼ 102:63 • e0:652 • Dose of GAA ðgÞ Fig. 2 presents plasma concentration-time profiles for creatine after single oral administration of GAA at 3 dosage
140
500
*
* Placebo
450
2.4 g of GAA
*
350
*
4.8 g of GAA
300 250
*
200 150
2.4 g of GAA
100
4.8 g of GAA
*
80
*
60
*
*
100
40 *
50 0
1.2 g of GAA *
Creatine (µmol/L)
400
GAA (µmol/L)
Placebo
120
1.2 g of GAA
20 0
1
2
4
6
8
12
24
Time (hours) Fig. 1 – Kinetics of plasma GAA after a single-dose ingestion of 1.2 g GAA (n = 12), 2.4 g GAA (n = 12), 4.8 g GAA (n = 12), or placebo (n = 12). Values are means ± SD. Tests of significance between treatment groups are based on 2-way ANOVA design for repeated measures (treatment vs time). The asterisk (*; P < .05) denotes significant interaction between time and treatments.
0
1
2
4
6
8
12
24
Time (h) Fig. 2 – Kinetics of plasma creatine after a single-dose ingestion of GAA in dosages of 1.2 g GAA (n = 12), 2.4 g GAA (n = 12), 4.8 g GAA (n = 12), or placebo (n = 12). Values are means ± SD. Tests of significance between treatment groups are based on 2-way ANOVA design for repeated measures (treatment vs time). The asterisk (*; P < .05) denotes significant interaction between time and treatments.
202
N U TR I TION RE S E ARCH 3 5 ( 2 0 15 ) 1 9 8–2 05
levels. The relationship between the GAA dose and plasma creatine kinetics was similar to that of GAA (P < .0001). For all 3 doses, peak concentration occurred within approximately 2 hours of dosing; thereafter, plasma creatine declined and mean terminal half-life values were between 5.8 and 8.8 hours. Table 3 presents selected PK parameters of plasma creatine after oral GAA administration.
4.
Discussion
The present study was the first to describe PK outcomes after ingestion of GAA. We confirmed our hypothesis that a single dose of orally administered GAA would exhibit dose-dependent PK behavior in healthy men and women. The study found that GAA single-dose PK metrics were nonlinear with respect to dose size, and larger doses (up to 4.8 g) lead to longer absorption times, augmented bioequivalence, and prolonged elimination. Our findings suggest that GAA is subject to saturable metabolism in the dose range tested. Because the GAA transporter has the Km value for GAA of approximately 80 μmol [21], high-dose intervention and accompanying plasma concentrations of GAA (up to 418 μmol/L) might saturate the GAA transporter function in the gut and the liver, limiting intestinal absorption and biotransformation.
4.1. Guanidinoacetic acid absorption and volume of distribution The results of the present study indicated that plasma GAA concentrations had a significant treatment vs time interaction in that the 3 oral GAA dosages behaved in a dose-
Table 3 – Selected PK parameters of plasma creatine after single-dose ingestion of GAA in different dosages 1.2 g of GAA (n = 12) Baseline concentration (μmol/L) Peak concentration (μmol/L) Time to peak concentration (h) AUC (0 → ∞, μmol/L * h) Elimination half-time (h)
2.4 g of GAA (n = 12)
28.7 ± 6.9
31.5 ± 10.5
51.7 ± 7.8
67.9 ± 17.1
1.82 ± 0.87
1.33 ± 0.49
223.2 ± 173.4
311.3 ± 199.8
8.78 ± 4.86
8.79 ± 9.00
4.8 g of GAA (n = 12) 28.8 ± 8.8
113.3 ± 31.3 a,b
1.75 ± 0.45 536.9 ±264.8 a,b 5.76 ± 4.40
Values are means ± SD. All PK parameters for creatine were calculated using noncompartmental analysis with trapezoid linear interpolation method. The AUC and elimination half-life were calculated after correction for baseline creatine values. Tests of significance between treatment groups are based on 3-sample unpaired ANOVA and Kruskal-Wallis model with post hoc tests (P < .05). a Significantly different from 1.2 g of GAA. b Significantly different from 2.4 g of GAA.
dependent fashion. There was a significant increase in systemic exposure as GAA doses increased from 1.2 to 4.8 g. Assuming complete absorption in the gut for the present study, oral GAA seems to be rapidly absorbed, entering the blood circulation from the gastrointestinal tract [22]. Lag time corresponds to the time necessary to observe an initial change in the plasma GAA concentration after the bolus ingestion. Similar lag time for all 3 GAA dosages indicates that absorption began at the same time after ingestion of GAA. A short delay in GAA absorption (between 8 and 23 minutes) could be attributed to capsule disintegration and GAA distribution. Although no differences were found between groups, the lag time trends increased as the dose increased, pointing to a more unfavorable dose-to-solubility ratio for the higher GAA doses. Absorption was comparable for all GAA dosages, although a modest prolongation (about 15 minutes) of mean absorption half-time was noted for 4.8 g of GAA compared with the 2 other dosage protocols. This suggests that the GAA transport system in the gut may be saturated by a single dose above 2.4 g of GAA. That the time to reach peak concentration is approximately 60% higher for the high-dose GAA regimen compared with the medium- and low-dose regimens further confirms this hypothesis. The unfavorable dose-to-solubility ratio could also increase absorption halftime in the high-dose group. Upon absorption, the peak plasma concentration corresponds to the supplemental dose of GAA, revealing the capacity of higher doses to increase the amount of GAA reaching the body's circulation system. Participants receiving a high dose of GAA exhibited a level of peak plasma GAA that was approximately 7 times higher than in the low-dose group (418.4 vs 56.9 μmol/L, respectively). In addition, bioequivalence seems to be notably augmented as the GAA dose increases, as shown by the significantly different AUCs for the 3 GAA dosages. Further investigation is necessary to determine whether a greater plasma concentration of GAA and/or better bioequivalence contributes to a preferable physiological. Determination of the apparent volume of distribution showed a clear dose dependency; the largest mean volume (>102 L) was found in participants who received low-dose GAA. Because the volume of distribution is mainly determined by the ratio of plasma to tissue binding, the volume of distribution decreases when the substance is substantially bound to plasma proteins. High-dose GAA is particularly noteworthy for its volume of distribution, suggesting that extra GAA might bind extensively to blood proteins. An additional explanation for differences in the apparent volumes of distribution between the 3 GAA dosages relates to how much of the total GAA in the body exists outside the area sampled (plasma). If GAA is more extensively distributed into the tissues, the blood concentration is lower and the volume estimate is higher. This approach indicates that tissue delivery is potentially better when low-dose GAA is administered, compared with the other dosage protocols. Because a substance's binding with a protein generally restricts distribution of a substance from the vascular compartment to the tissue [23], dose-dependent differences in GAA's proteinbinding capacity could affect the distribution of GAA in the body. Further studies are needed to determine the degree to which GAA binds with blood proteins.
N U T RI TI O N RE S E ARCH 3 5 ( 2 0 15 ) 1 9 8–2 05
4.2.
Elimination of dietary GAA
The mean half-life of elimination for GAA was short (
Available online at www.sciencedirect.com
ScienceDirect www.nrjournal.com
Original Research
Single-dose oral guanidinoacetic acid exhibits dosedependent pharmacokinetics in healthy volunteers Sergej M. Ostojic a, b,⁎, Aleksandra Vojvodic-Ostojic c a b c
Exercise Physiology Laboratory, Center for Health, Exercise and Sport Sciences, Belgrade, Serbia Biomedical Sciences Deparment, Faculty of Sport and Physical Education, University of Novi Sad, Novi Sad, Serbia Environmental Engineering Department, Faculty of Technology and Metallurgy, University of Belgrade, Belgrade, Serbia
ARTI CLE I NFO
A BS TRACT
Article history:
Guanidinoacetic acid (GAA), the natural precursor of creatine, has potential as a dietary
Received 16 September 2014
supplement for human nutrition, yet no data are available regarding its dose-dependent
Revised 5 December 2014
pharmacokinetic (PK) behavior. We hypothesized that a single dose of orally administered
Accepted 30 December 2014
GAA exhibited dose-dependent PK behavior in healthy volunteers. Forty-eight young adults were enrolled in a randomized, placebo-controlled, double-blind, parallel-group trial to
Keywords:
receive single oral doses of GAA (1.2, 2.4, and 4.8 g) or a placebo. Pharmacokinetic metrics for
Guanidinoacetic acid
plasma GAA and creatine were assessed immediately before (0 hours) and at 1, 2, 4, 6, 8, 12,
Pharmacokinetics
and 24 hours after GAA ingestion. The lag time appeared to be similar after the bolus
Area under the curve
ingestion of GAA (0.14 ± 0.17 hours for low-dose GAA, 0.31 ± 0.18 hours for medium-dose
Elimination half-life
GAA, and 0.38 ± 0.32 hours for high-dose GAA; P = .05). An increase in the area under the
Clearance
concentration-time curve for plasma GAA was found for the dose range tested, with 2.4and 9.3-fold increases in the area under the concentration-time curve for every 2-fold increase in the GAA dose (P < .0001). No differences were found for elimination half-time between the low-dose and medium-dose groups (2.1 hours) for the high-dose GAA regimen (P = .001). The volume of distribution was affected by the dosage of GAA applied (102.6 ± 17.3 L for lowdose GAA, 97.5 ± 15.7 L for medium-dose GAA, and 61.1 ± 12.7 L for high-dose GAA; P < .0001). Ingestion of GAA elevated plasma creatine by 80%, 116%, and 293% compared with the placebo for the 1.2, 2.4, and 4.8 g doses, respectively (P < .0001). Guanidinoacetic acid single-dose PK metrics were nonlinear with respect to dose size. Across the dose range of 1.2 to 4.8 g, systemic exposure to GAA increased in a greater than dose-proportional manner. © 2015 Elsevier Inc. All rights reserved.
1.
Introduction
Guanidinoacetic acid (GAA; also known as glycocyamine or guanidinoacetate) is a nitrogenous organic acid that occurs naturally in the human body. The primary biological role of
GAA is to serve as a direct precursor of creatine, an energy carrier, and mediator in the cell [1]. Guanidinoacetic acid is synthesized endogenously from L-arginine and glycine, mainly in the kidney [2]. Under certain circumstances (eg, kidney failure, exercise-related GAA depletion, and deficient diet),
Abbreviations: AUC, area under the concentration-time curve; GAA, guanidinoacetic acid; PK, pharmacokinetic. ⁎ Corresponding author. Center for Health, Exercise and Sport Sciences, Deligradska 27, Belgrade 11000, Serbia. Tel./fax: +381 11 2643 242. E-mail address: [email protected] (S.M. Ostojic). http://dx.doi.org/10.1016/j.nutres.2014.12.010 0271-5317/© 2015 Elsevier Inc. All rights reserved.
N U T RI TI O N RE S E ARCH 3 5 ( 2 0 15 ) 1 9 8–2 05
supplemental GAA may be “semiessential,” optimizing the body's energy use when the endogenous synthesis of GAA is limited [3]. Because GAA is more bioavailabile, stable, and cost-effective than creatine [4], it might be more suitable for use in human nutrition. In particular, GAA is more soluble in water then creatine (3600 mg/L for GAA vs 1330 mg/L for creatine). Guanidinoacetic acid is highly stable [5], whereas creatine is rapidly metabolized into creatinine when it comes into contact with water at elevated temperatures and low pH [6]. Finally, the cost of GAA powder is approximately 40% lower than the cost of creatine [7,8]. Although the first human use of supplemental GAA was reported more than 60 years ago, there is a little information available on its turnover, bioavailability, or elimination. Previous studies of dietary GAA reported its “creatine recovery effect” in both athletes and clinical patients. Studies have found that orally ingested GAA administered over the course of 6 weeks increases plasma GAA and creatine by up to 50% in male and female collegiate athletes, with no major adverse effects reported [9]. Studies have shown that supplemental GAA may increase homocysteine formation [10], although the effects of increased homocysteine on health status seem to be clinically insignificant [9], and can be corrected by homocysteine removal mechanisms [11]. Dietary GAA has been found to be beneficial for patients with heart disease, neuromuscular disorders, and chronic renal failure [12]. Scientific research has been limited with regard to the behavior of dietary GAA between the time of ingestion and subsequent elimination from the human body. A Japanese group was the first to describe the timeline of plasma concentrations and urinary excretions of GAA in humans after a single oral dose of 1 g of GAA [13]. The study suggested that GAA was absorbed well, with peak plasma GAA occurring after 2 hours (~600 μg/dL) and extra GAA excreted in the urine at a rate of up to 40 mg/h. We recently reported that an acute oral dose of GAA has a notable effect on plasma GAA, creatine, creatinine, total homocysteine, and excretion of urinary GAA and creatine in 20 healthy volunteers [3]. In our study, supplemental GAA was found to be readily bioavailable and rapidly transformed to creatine within the first 2 hours. Neither study described relevant pharmacokinetic (PK) metrics (eg, lag time, volume of distribution, elimination half-life, and clearance) after GAA ingestion, and no studies have examined the dose-dependent PK of oral GAA. Exploratory PK studies with increasing doses of GAA are needed to describe its biochemical behavior and to decide whether GAA has scientific merit for further development as a dietary supplement for humans. This study tested the hypothesis that single-dose, orally administered GAA would exhibit dose-dependent PK in healthy men and women. To test this hypothesis, we conducted a double-blind, placebo-controlled, randomized, parallel-group intervention trial, and evaluated PK metrics for plasma GAA and creatine to assess the acute effects of 1.2, 2.4, and 4.8 g GAA loads in healthy men and women. The study was designed as a phase I, first-in-humans trial to establish whether oral GAA behaves in human subjects as preclinical studies would suggest. Assessment of oral GAA dose escalation and its absorption, distribution, biotransformation, and elimination is a necessary step to clinically evaluate GAA as a potentially useful health intervention.
2.
Methods and materials
2.1.
Study population
199
Candidates for inclusion in the study were moderately physically active men and women between 20 and 25 years old. Potential participants were not admitted to the study if they met any of the following criteria: (1) a history of metabolic disease, (2) known heart disease, (3) use of any performance-enhancing drugs or dietary supplements within the 60 days before the study commenced, (4) smoking, and (5) pregnancy. Participants were fully informed both verbally and in writing about the nature and demands of the study and the known health risks, and they provided informed consent regarding their voluntary participation in the study. The participants completed a health history questionnaire covering cardiovascular disease risk factors, history and present status regarding any signs and symptoms suggestive of cardiovascular disease, history of chronic illnesses, history of surgeries and hospitalizations, history of any musculoskeletal or joint injuries, past and present habits that could affect health, current medication use, family health history, and other health history information [14]. All participants completed routine prescreening including blood and urine profiling, and they received a general medical examination during the initial recruitment. If any specific markers (eg, liver and muscle enzymes, kidney function) were above the reference values, subjects were excluded from the study. Upon initial recruitment, 48 participants (n = 48; 24 men and 24 women) met the criteria to participate in the study; the total number of participants fulfilled the optimal sample size (see below). The mean physical characteristics of participants were as follows: age 22 ± 2 years, weight 71 ± 14 kg, and height 176 ± 10 cm. Participants' nutrition was monitored before the experiment began using a standardized questionnaire [15] and computed using NutriBase software (CyberSoft Inc, Phoenix, AZ, USA). Approval of the local institutional review board (No. 05/A-2010/014) was obtained, and all the procedures performed were in accordance with the Declaration of Helsinki and the principles set forth in the Guidelines for Good Clinical Practice [16].
2.2.
Experimental protocol
Participants were randomized according to a computer-generated list in a double-blind, parallel-group design to receive single oral doses of GAA (1.2, 2.4, and 4.8 g) or a placebo (inulin). The GAA dosages chosen were the equimolar equivalent to dietary creatine amounts that appear to be therapeutically effective [17]. The placebo was administered as a control treatment to monitor the changes in GAA and creatine plasma concentrations during the day, especially the effects of creatine-containing foods on plasma concentrations. Twelve participants were allocated to each intervention group, with women having an equal probability of assignment to the groups. Groups were matched for participants' age, weight, and daily energy intakes. The participants received standardized meals on the day before the experiment. In the 24 hours before the tests, the subjects did not participate in any prolonged exercise and were not permitted to drink alcoholic or caffeinated beverages. On the day of the experiment, each participant arrived at the laboratory at 9:00 AM
200
N U TR I TION RE S E ARCH 3 5 ( 2 0 15 ) 1 9 8–2 05
after an overnight fast of between 10 and 12 hours. Participants provided a fasting blood sample (~5 mL) from a radial artery into heparinized tubes (Greiner Bio-One, Basel, Switzerland) to determine the basal level of GAA and creatine. After baseline sampling, participants received the appropriate oral dose of GAA or a placebo in soft capsules (AlzChem AG, Trostberg, Germany), with 0.5 L of tap water consumed within approximately 1 minute. After the ingestion, blood samples were collected at 1, 2, 4, 6, 8, 12, and 24 hours for quantification of plasma kinetics for GAA and creatine. Participants were sitting or lying down during most of the study period, and standardized meals (Table 1) were provided for all participants at the same time after GAA administration. Breakfast, lunch, snack, and dinner were provided according to a schedule (~11 AM, ~ 1 PM, ~ 3 PM, and ~7 PM, respectively). No creatine-rich foods (eg, meat, fish, and milk) were offered to avoid influencing the metabolites studied. After collection, blood samples were centrifuged, and the plasma was separated and stored at −20°C until the time of assay. Plasma GAA and creatine were assayed as described previously [18] using high-performance liquid chromatography (HewlettPackard, Palo Alto, CA, USA). All samples for each subject were assayed in the same run. For all values, the first reading was discarded, and the study used the mean of the next 3 consecutive readings with a coefficient of variation below 15%.
2.3.
Pharmacokinetic analyses
The concentration-time profiles for plasma GAA were fitted using a 1-compartment model for extravascular drug administration. Nonlinear fitting analysis of the compartmental model was performed using the PK add-in program PK Solver [19]. For data fitting, the Gauss-Newton algorithm was used, and initial parameters were obtained using the curvestripping technique [20]. The data were weighted by the inverse of the estimated measurement error variance. The weighted least-squares method was applied as the criteria for the best fit. Graphs of predicted concentrations vs observed data were visually examined, and values that fit poorly were excluded from further analysis. Quality assessment of the compartmental PK model for the individual's concentrationtime data was based on visual analysis, predicted concentrations vs time plots, residuals vs predicted concentrations plots, the precision of parameter estimates as correlation coefficients, and the weighted sum of squares of residuals. The 1-compartment open model with lag time for oral GAA administration assumed a first-order process of absorption and a first-order process of elimination. The following
Table 1 – Specification of meals for GAA PKs study Meal
Serving Total Carbohydrates Lipids Proteins size (g) energy (%) (%) (%) intake (kcal)
Breakfast Lunch Snack Dinner
340 546 25 665
504 705 111 558
63.5 56.1 68.0 72.5
18.2 15.9 16.6 12.7
18.3 28.0 15.4 14.8
equation describes the rate of change in GAA concentrations: C¼
FD ka exp−kel ðt−tlag Þ − exp−ka ðt−tlag Þ V ka −kel
where C is the plasma GAA concentration minus the baseline concentration (μmol/L); t is the time of oral administration (l-h−1); tlag is the lag time (h−1); F is the bioavailability; D is the dose (μmol); V is the volume of distribution (L); ka is the absorption-rate constant (h−1); and kel is the elimination-rate constant (h−1). In this equation, ka, kel, and tlag are primary parameters from which the other PK parameters were derived. The area under the concentration-time curve (AUC) was calculated using the linear-log trapezoidal method. Because the bioavailability (F) was not measured, only the apparent volume of distribution (Vd) and the apparent clearance (CL) were calculated using the following equations: Vd = V/F and CL = kel • Vd.
2.4.
Statistical analyses
Allowing for 80% power, detection of a 5% difference in peak plasma concentration of GAA between groups required an estimated 10 participants per group. This estimation was adjusted to 12 subjects per group to account for a predicted 20% dropout rate. All results were expressed as mean ± SD. Pharmacokinetic data for each group were tested using the Shapiro-Wilk test for the normality of distribution and Bartlett test for the homogeneity of the variances. After homogenous variances were verified for normally distributed data, plasma kinetic data were compared using 1-way analysis of variance (ANOVA), and the post-hoc Tukey HSD test was used to identify the differences between individual sample pairs. When nonhomogenous variances were identified, mean PK data were compared using the Kruskal-Wallis test. The GamesHowell post hoc test was then used to evaluate whether differences between any 2 groups were significant. A 2-way ANOVA design for repeated measures (treatment vs time) was used to uncover any significant differences between interventions during the 24-hour experiment. The relationship between the AUC and the dose of GAA was examined using Pearson product-moment correlation coefficient. The significance level was set at P < .05. Data were analyzed using the version 21.0 SPSS program (SPSS Inc, Chicago, IL, USA).
3.
Results
No volunteers reported any cardiovascular disease risk factors, history of chronic illnesses, or other relevant health information before intervention. Prior to the experiment, volunteers reported daily energy intakes of 3508 ± 768 kcal, composed of 63.5% ± 4.9% carbohydrates, 23.2% ± 4.8% lipids, and 13.3% ± 3.0% proteins (1.5 ± 0.4 g of proteins per kg body mass). All volunteers (n = 48) completed the PK studies, and no serious protocol violations were identified. No volunteers reported any adverse effects from the GAA supplementation. The mean basal plasma GAA concentration was 2.9 ± 0.3 μmol/L, with no differences found between groups for baseline plasma GAA (P = .53), as shown in Fig. 1. No changes
201
N U T RI TI O N RE S E ARCH 3 5 ( 2 0 15 ) 1 9 8–2 05
in serum GAA or creatine were reported after administration of the placebo throughout the study (P > .05). Two-way ANOVA analysis with repeated measures revealed a significant interaction between time and treatments (P > .0001), indicating different time-courses dependent on the GAA dosages administered. Post hoc analyses showed that differences occurred during both the absorption and elimination phases. After bolus ingestion of GAA, plasma concentration increased sharply to reach its peak in about 90 minutes for both the low-dose and medium-dose groups, whereas the peak concentration occurred 2.2 hours after participants consumed high-dose GAA (Table 2). The peak concentration was highest when high-dose GAA was ingested (418.4 ± 89.6 μmol/L), compared with mediumdose (151.3 ± 17.5 μmol/L) and low-dose (56.9 ± 9.2 μmol/L) administration. Lag time appeared to be similar in all groups after the bolus ingestion of GAA (P = .05). The mean absorption half-time was comparable in all GAA groups (P = .07). The absorption half-time in the low-dose and medium-dose groups was less than 40 minutes, whereas it reached almost 60 minutes with high-dose GAA. The AUC from zero to infinity, corrected for baseline GAA values, revealed significant differences between the 3 GAA dosages (P < .0001). No differences were found for elimination half-time between the low-dose and medium-dose groups (2.1 hours) for the high-dose GAA regimen (P = .001). The apparent clearance and apparent volume of distribution were affected by the GAA dosage; in the low-dose group, clearance and distribution volume were greater than in the medium-dose group, which in turn showed higher volume and clearance than the highdose group (P < .0001). Finally, a strong relationship was found between the AUC and the dose of GAA administered, as demonstrated by Pearson product-moment correlation
Table 2 – PKs of GAA after single-dose ingestion of GAA in different dosages 1.2 g of GAA (n = 12) Peak concentration of plasma GAA (μmol/L) Time to peak concentration (h) Lag time (h) Velocity constant of absorption (h−1) Absorption half-time (h) AUC (0 → ∞, μmol/L * h) Apparent volume of distribution (L) Velocity constant of elimination (h−1) Elimination half-time (h) Apparent systemic clearance (L/h)
2.4 g of GAA (n = 12)
4.8 g of GAA (n = 12)
56.9 ± 9.2
151.3 ± 17.5 a
418.4 ± 89.6 a,b
1.33 ± 0.49
1.33 ± 0.49
2.17 ± 0.58 a,b
0.14 ± 0.17 1.08 ± 0.21
0.31 ± 0.18 1.04 ± 0.08
0.38 ± 0.32 1.22 ± 1.21
0.66 ± 0.13
0.67 ± 0.06
0.93 ± 0.44
224.8 ± 38.0
531.4 ± 64.1 a
2092.4 ±453.2 a,b
102.6 ± 17.3
97.5 ± 15.7
61.1 ± 12.7 a,b
0.46 ± 0.07
0.41 ± 0.07
0.34 ± 0.06 a,b
1.54 ± 0.26
1.74 ± 0.30
2.10 ± 0.36 a,b
46.9 ± 8.3
39.1 ± 4.6 a
20.5 ± 4.4 a,b
Values are means ± SD. Tests of significance between treatment groups are based on 3sample unpaired ANOVA and Kruskal-Wallis model with post hoc tests (P < .05). a Significantly different from 1.2 g of GAA. b Significantly different from 2.4 g of GAA.
coefficient of r = 0.98 (P < .0001), and the trend line was best fitted exponentially: AUC for plasma GAA ð0→∞; μmol=L hÞ ¼ 102:63 • e0:652 • Dose of GAA ðgÞ Fig. 2 presents plasma concentration-time profiles for creatine after single oral administration of GAA at 3 dosage
140
500
*
* Placebo
450
2.4 g of GAA
*
350
*
4.8 g of GAA
300 250
*
200 150
2.4 g of GAA
100
4.8 g of GAA
*
80
*
60
*
*
100
40 *
50 0
1.2 g of GAA *
Creatine (µmol/L)
400
GAA (µmol/L)
Placebo
120
1.2 g of GAA
20 0
1
2
4
6
8
12
24
Time (hours) Fig. 1 – Kinetics of plasma GAA after a single-dose ingestion of 1.2 g GAA (n = 12), 2.4 g GAA (n = 12), 4.8 g GAA (n = 12), or placebo (n = 12). Values are means ± SD. Tests of significance between treatment groups are based on 2-way ANOVA design for repeated measures (treatment vs time). The asterisk (*; P < .05) denotes significant interaction between time and treatments.
0
1
2
4
6
8
12
24
Time (h) Fig. 2 – Kinetics of plasma creatine after a single-dose ingestion of GAA in dosages of 1.2 g GAA (n = 12), 2.4 g GAA (n = 12), 4.8 g GAA (n = 12), or placebo (n = 12). Values are means ± SD. Tests of significance between treatment groups are based on 2-way ANOVA design for repeated measures (treatment vs time). The asterisk (*; P < .05) denotes significant interaction between time and treatments.
202
N U TR I TION RE S E ARCH 3 5 ( 2 0 15 ) 1 9 8–2 05
levels. The relationship between the GAA dose and plasma creatine kinetics was similar to that of GAA (P < .0001). For all 3 doses, peak concentration occurred within approximately 2 hours of dosing; thereafter, plasma creatine declined and mean terminal half-life values were between 5.8 and 8.8 hours. Table 3 presents selected PK parameters of plasma creatine after oral GAA administration.
4.
Discussion
The present study was the first to describe PK outcomes after ingestion of GAA. We confirmed our hypothesis that a single dose of orally administered GAA would exhibit dose-dependent PK behavior in healthy men and women. The study found that GAA single-dose PK metrics were nonlinear with respect to dose size, and larger doses (up to 4.8 g) lead to longer absorption times, augmented bioequivalence, and prolonged elimination. Our findings suggest that GAA is subject to saturable metabolism in the dose range tested. Because the GAA transporter has the Km value for GAA of approximately 80 μmol [21], high-dose intervention and accompanying plasma concentrations of GAA (up to 418 μmol/L) might saturate the GAA transporter function in the gut and the liver, limiting intestinal absorption and biotransformation.
4.1. Guanidinoacetic acid absorption and volume of distribution The results of the present study indicated that plasma GAA concentrations had a significant treatment vs time interaction in that the 3 oral GAA dosages behaved in a dose-
Table 3 – Selected PK parameters of plasma creatine after single-dose ingestion of GAA in different dosages 1.2 g of GAA (n = 12) Baseline concentration (μmol/L) Peak concentration (μmol/L) Time to peak concentration (h) AUC (0 → ∞, μmol/L * h) Elimination half-time (h)
2.4 g of GAA (n = 12)
28.7 ± 6.9
31.5 ± 10.5
51.7 ± 7.8
67.9 ± 17.1
1.82 ± 0.87
1.33 ± 0.49
223.2 ± 173.4
311.3 ± 199.8
8.78 ± 4.86
8.79 ± 9.00
4.8 g of GAA (n = 12) 28.8 ± 8.8
113.3 ± 31.3 a,b
1.75 ± 0.45 536.9 ±264.8 a,b 5.76 ± 4.40
Values are means ± SD. All PK parameters for creatine were calculated using noncompartmental analysis with trapezoid linear interpolation method. The AUC and elimination half-life were calculated after correction for baseline creatine values. Tests of significance between treatment groups are based on 3-sample unpaired ANOVA and Kruskal-Wallis model with post hoc tests (P < .05). a Significantly different from 1.2 g of GAA. b Significantly different from 2.4 g of GAA.
dependent fashion. There was a significant increase in systemic exposure as GAA doses increased from 1.2 to 4.8 g. Assuming complete absorption in the gut for the present study, oral GAA seems to be rapidly absorbed, entering the blood circulation from the gastrointestinal tract [22]. Lag time corresponds to the time necessary to observe an initial change in the plasma GAA concentration after the bolus ingestion. Similar lag time for all 3 GAA dosages indicates that absorption began at the same time after ingestion of GAA. A short delay in GAA absorption (between 8 and 23 minutes) could be attributed to capsule disintegration and GAA distribution. Although no differences were found between groups, the lag time trends increased as the dose increased, pointing to a more unfavorable dose-to-solubility ratio for the higher GAA doses. Absorption was comparable for all GAA dosages, although a modest prolongation (about 15 minutes) of mean absorption half-time was noted for 4.8 g of GAA compared with the 2 other dosage protocols. This suggests that the GAA transport system in the gut may be saturated by a single dose above 2.4 g of GAA. That the time to reach peak concentration is approximately 60% higher for the high-dose GAA regimen compared with the medium- and low-dose regimens further confirms this hypothesis. The unfavorable dose-to-solubility ratio could also increase absorption halftime in the high-dose group. Upon absorption, the peak plasma concentration corresponds to the supplemental dose of GAA, revealing the capacity of higher doses to increase the amount of GAA reaching the body's circulation system. Participants receiving a high dose of GAA exhibited a level of peak plasma GAA that was approximately 7 times higher than in the low-dose group (418.4 vs 56.9 μmol/L, respectively). In addition, bioequivalence seems to be notably augmented as the GAA dose increases, as shown by the significantly different AUCs for the 3 GAA dosages. Further investigation is necessary to determine whether a greater plasma concentration of GAA and/or better bioequivalence contributes to a preferable physiological. Determination of the apparent volume of distribution showed a clear dose dependency; the largest mean volume (>102 L) was found in participants who received low-dose GAA. Because the volume of distribution is mainly determined by the ratio of plasma to tissue binding, the volume of distribution decreases when the substance is substantially bound to plasma proteins. High-dose GAA is particularly noteworthy for its volume of distribution, suggesting that extra GAA might bind extensively to blood proteins. An additional explanation for differences in the apparent volumes of distribution between the 3 GAA dosages relates to how much of the total GAA in the body exists outside the area sampled (plasma). If GAA is more extensively distributed into the tissues, the blood concentration is lower and the volume estimate is higher. This approach indicates that tissue delivery is potentially better when low-dose GAA is administered, compared with the other dosage protocols. Because a substance's binding with a protein generally restricts distribution of a substance from the vascular compartment to the tissue [23], dose-dependent differences in GAA's proteinbinding capacity could affect the distribution of GAA in the body. Further studies are needed to determine the degree to which GAA binds with blood proteins.
N U T RI TI O N RE S E ARCH 3 5 ( 2 0 15 ) 1 9 8–2 05
4.2.
Elimination of dietary GAA
The mean half-life of elimination for GAA was short (
