Plasma protein binding of quinine: binding to human serum albumin, alpha 1-acid glycoprotein and plasma from patients with malaria.

The binding of quinine to human serum albumin (HSA), alpha 1-acid glycoprotein (AAG) and plasma obtained from healthy subjects (10 caucasians and 15 Thais) and from Thai patients with falciparum malaria (n = 20) has been investigated. In healthy volunteers, plasma protein binding expressed as the percentage of unbound quinine was 7.9-31.0% (69-92.1% bound). The mean percentage of unbound quinine found with essentially fatty acid-free HSA (40 g L-1) was 65.4 +/- 1.5% (mean +/- s.d.) and was comparable with the value (66.3 +/- 3.8%, mean +/- s.d.) for Fraction V HSA (40 g L-1). This suggests that fatty acids do not influence the plasma protein binding of quinine. Binding of quinine to 0.7 g L-1 AAG was high (mean unbound 61.0 +/- 5.0%), indicating that quinine is bound primarily to AAG and albumin, although other plasma proteins such as lipoproteins may be involved. The mean percentage of unbound quinine was slightly less in caucasians (14.8 +/- 6.7% unbound), compared with healthy Thai subjects (17.0 +/- 6.7% unbound). The higher binding of quinine in caucasian subjects was associated with a higher plasma AAG concentration observed in caucasians. Mean percentage of unbound quinine was significantly lower in Thai patients with malaria (10.9 +/- 4.0%) than in the healthy Thai subjects. The increase in the extent of quinine binding corresponded with the increase in the acute-phase reactant protein, AAG in the patients with malaria. Overall, when the data were combined there was a significant correlation (r = 0.846, P < 0.005) between the binding ratio (bound/unbound) of quinine and the plasma AAG concentration.(ABSTRACT TRUNCATED AT 250 WORDS)     

Binding of catecholamines to alpha-1 acid glycoprotein, albumin and lipoproteins in human serum

The binding of catecholamines in human serum was determined by equilibrium dialysis at 37 degrees. For serum concentrations of 10-15 nM the bound fractions were 28.8 +/- 2.2%, 25.7 +/- 1.7% and 22.2 +/- 2.2% for (+/-)-isoproterenol (IPR), (+/-)-norepinephrine (NE) and (+/-)-epinephrine (EPI), respectively. At higher serum concentrations saturation occurred. Alpha-1 acid glycoprotein (AAG) possessed one high affinity binding site and approximately 10 low affinity sites. The catecholamines were bound to AAG with the same order of potency for both classes of binding sites: IPR (Kd1: 100 microM Kd2: 2.2 mM) greater than NE (Kd1: 120 microM, Kd2: 6.5 mM) greater than EPI (Kd1: 140 microM, Kd2: 14 mM). Human serum albumin (HSA) and lipoproteins (SLP) interacted with the catecholamines in a non-saturable manner. IPR showed the strongest and EPI the weakest association to both of these serum protein fractions. (-)-Propranolol was able to inhibit the binding of IPR in serum and to isolated AAG, but not to HSA or to SLP. The present results show that AAG is an important catecholamine-binding protein in human serum. AAG, but not HSA or SLP, possesses binding sites shared by adrenergic receptor stimulators and blockers.     

The plasma protein binding of metoclopramide in health and renal disease.

The plasma protein binding of metoclopramide was measured after addition of the drug (60 ng ml-1) to plasma from 18 patients with renal disease and 18 age and sex matched healthy individuals. The mean free fraction in renal disease (0.59 range 0.41-0.71) was not significantly different from controls (mean 0.6 range 0.56-0.69). In both groups the binding ratio of metoclopramide was significantly related to plasma alpha 1-acid glycoprotein (AAG) concentration but not to albumin or plasma non-esterified fatty acids concentration. Metoclopramide bound to human serum albumin (HSA) to a limited extent and to human AAG to a greater extent indicating that AAG is the major binding protein for the drug in plasma.     

In vitro protein binding of propafenone and 5-hydroxypropafenone in serum, in solutions of isolated serum proteins, and to red blood cells.

The binding of propafenone (PF) and 5-hydroxypropafenone (5-OH-PF) in serum and in solutions of isolated serum proteins was examined by equilibrium dialysis. Both PF and 5-OH-PF displayed pH-dependent binding in serum and in a solution of alpha-1-acid glycoprotein (AAG). PF displayed extensive binding to AAG (i.e., free fraction of 0.08 +/- 0.02), whereas the binding of 5-OH-PF to AAG was moderate (i.e., free fraction of 0.54 +/- 0.10). The removal of lipoproteins from serum did not alter the free fraction of PF but significantly increased the free fraction of 5-OH-PF compared with that in intact serum. Both PF and 5-OH-PF displayed concentration-dependent binding in a 19.3-mumol AAG solution. Concentration-independent binding was apparent in solutions of human serum albumin, high-density lipoproteins, low-density lipoproteins, and very low density lipoproteins over the PF and 5-OH-PF concentration ranges examined. By use of previously determined binding parameters (affinities and capacities), the binding model of PF provided an estimate of the free fraction in serum that was similar to the observed free fraction, although the free fraction of 5-OH-PF was overestimated. The distribution of PF and 5-OH-PF into red blood cells was extensive when buffer was used as the supernatant; however, when serum was used as supernatant, the amounts of PF and 5-OH-PF that were distributed into red blood cells decreased substantially. PF and 5-OH-PF interacted with all of the proteins examined.     

Protein Binding of Cocaine in Human Serum

The protein binding characteristics of cocaine have not been extensively studied. Since cocaine is related to other local anesthetic compounds which are highly protein bound, we examined the binding of cocaine in human serum using an ultrafiltration method. The free fraction averaged 0.083 ± 0.018 in the serum of 12 healthy volunteers. Binding was studied at concentrations ranging from 0.1 to 500 µg/ml and was concentration dependent, with increases being most pronounced at concentrations above 5 µg/ml. Two classes of binding sites were identified with affinity and capacity constants consistent with binding to alpha-1-acid glycoprotein (AAG) and albumin. The addition of AAG to serum resulted in a decrease in the free fraction from 0.079 to 0.041, while tris(butoxyethyl)phosphate increased the free fraction to 0.233. The binding ratio was found to be highly correlated with the AAG concentration (r = 0.89). In addition, the predicted free fraction in the absence of AAG (0.67) was in good agreement with the observed value of 0.647 in a solution of human serum albumin (4.5 g/dl). Of the metabolites of cocaine, only norcocaine displaced the parent drug from serum binding sites. These results indicate that cocaine is highly bound to serum proteins, primarily albumin and AAG. The significance of concentration-dependent binding to cocaine toxicity remains to be established.     

Distribution of 7-hydroxymethotrexate in human blood

We have examined the in-vitro distribution of 7-hydroxymethotrexate (7-OH-MTX), a cytotoxic metabolite of methotrexate (MTX), in human blood, and its protein binding in serum. The distribution of 7-OH-MTX (10(-6) M) in fresh samples of whole blood was studied at 37 degrees C and pH 7.51 +/- 0.05 (mean +/- s.d.), and its protein binding was assessed by equilibrium dialysis of serum against Krebs Ringer phosphate buffer at 37 degrees C and pH 7.41 +/- 0.07 (mean +/- s.d.). 7-OH-MTX had a mean cell/plasma concentration ratio of 0.03 (range 0-0.27, n = 18). It was extensively bound in human serum, with a bound fraction of 90.4 +/- 3.3% (mean +/- s.d.) in healthy volunteers (n = 11), and significantly lower, 82.3 +/- 4.0% (mean +/- s.d.), in hypoalbuminaemic surgical patients (n = 7). The binding of 7-OH-MTX was correlated with serum albumin (HSA) concentrations (r = 0.72, P less than 0.0007, n = 18). Blood distribution data support the contention that 7-OH-MTX has a small volume of distribution, and HSA appears to be mainly responsible for the high degree of its protein binding in serum.     

Effect of glycyrrhizic acid on protein binding of diltiazem,verapamil, and nifedipine

The effects of glycyrrhizic acid (GLZ) on protein binding of diltiazem, verapamil, and nifedipine were investigated. Protein binding studies (human serum, human serum albumin (HSA) and alpha1-acid glycoprotein (AAG)) were conducted using the equilibrium dialysis method with and without addition of GLZ. The binding parameters, such as the number of moles of bound drug per mole of protein, the number of binding sites per protein molecule, and the association constant, were estimated using the Scatchard plot. The serum binding of nifedipine, verapamil, and diltiazem was displaced with addition of GLZ, and the decreases of Ks for serum were observed. GLZ decreased the association constants of three drugs for HSA and AAG, while the binding capacity remained similar with addition of GLZ. Although the characteristics of interaction were not clear, GLZ seemed to mainly affect HSA binding of nifedipine rather than AAG binding, while GLZ seemed to affect both AAG- and HSA-bindings of verapamil and diltiazem resulting in a serum binding displacement.     

Stereoselective binding of propranolol in the elderly.

The influence of age on the stereoselective serum protein binding of propranolol was investigated. Serum was obtained from 10 young (mean age 23 +/- 2 years) and ten elderly (mean age 68 +/- 2 years) healthy male volunteers. The free fraction of propranolol (fu) was measured by equilibrium dialysis, using tritium labeled (+/-)- and (-)-propranolol. The fu values were 0.159 +/- 0.049 and 0.148 +/- 0.028 (+/-), 0.135 +/- 0.041 and 0.136 +/- 0.04 (-), 0.174 +/- 0.056 and 0.161 +/- 0.028 (+) in the young and elderly subjects, respectively. Serum alpha 1-acid glycoprotein (AAG) concentrations were 81.4 +/- 33.0 and 81.0 +/- 21.6 mg 100 ml-1 in young and elderly respectively (P greater than 0.05). Variability in AAG concentration accounted for most of the observed intersubject variability in the binding of both propranolol isomers. The stereoselective binding of propranolol does not appear to be affected by age.     

Protein binding of propranolol and verapamil enantiomers in maternal and foetal serum.

The protein binding of the enantiomers of propranolol and verapamil was measured in 19 pairs of maternal and foetal serum obtained at delivery. The binding of the enantiomers of both drugs was lower in foetal than in maternal serum. In maternal serum the mean (+/- s.d.) unbound percentages were 22.4 +/- 6.2 and 20.7 +/- 6.6 for R- and S-propranolol, and 16.8 +/- 5.5 and 22.5 +/- 6.2 for R- and S-verapamil; in foetal serum the values were 38.8 +/- 8.6 and 40.4 +/- 10.6 for R- and S-propranolol, and 34.7 +/- 10.5 and 44.8 +/- 10.7 for R- and S-verapamil. For propranolol, in maternal, but not in foetal serum, the difference between the binding of the R- and S-enantiomers was significant; the R/S ratio was significantly (P < 0.01) larger in the mother (1.099 +/- 0.072) than in the foetus (0.973 +/- 0.068). For verapamil, the difference in binding between the R- and S-enantiomers was significant in both maternal and foetal serum, but the R/S ratio was similar in mother (0.735 +/- 0.098) and foetus (0.763 +/- 0.070). Serum alpha 1-acid glycoprotein (AAG) concentrations were markedly higher and albumin concentrations slightly lower in maternal than in foetal samples. The binding of the four enantiomers in maternal and foetal serum was correlated (P < 0.001) with the AAG concentration (r propranolol: R 0.749, S 0.746; r verapamil: R 0.753, S 0.782). Our findings show that measurement of concentrations of total, unresolved drug allow a reasonably accurate assessment of transplacental gradients of individual isomers.     

Pharmacokinetics of intravenously administered 125I-labelled human alpha 1-acid glycoprotein

The volumes of distribution of many acidic drugs have been shown to be close to that of their binding protein, i.e. serum albumin. The distribution of basic drugs mainly bound to alpha 1-acid glycoprotein (AAG) can be questioned with respect to its dependency upon the distribution of this plasma protein. So, a pharmacokinetic study was performed in 7 subjects with human 125I-labelled alpha 1-acid glycoprotein. The steady-state volume of distribution was found to be 5.37 +/- 0.82L. The central volume was 3.23 +/- 0.33L, close to that of plasma volume and the peripheral volume was 2.14 +/- 0.63L. These data allowed the establishment of an equation giving access to the volume of distribution of a basic drug by relating its unbound fraction to physiological distribution of alpha 1-acid glycoprotein. The values yielded by this equation show that the actual and calculated volumes of distribution of basic drugs mainly bound to AAG are discrepant. This protein is thus not the main factor controlling the distribution of basic drugs within the body.     

In vitro protein binding behavior of dipyridamole

The in vitro protein binding behavior of dipyridamole in plasma and buffered protein solutions was investigated by equilibrium dialysis. The drug was highly protein bound (approximately 98%) in heparinized human plasma, and the extent of protein binding remained constant for drug concentrations over the range of therapeutic interest of 0.1-10 micrograms/mL. Comparable binding results were obtained with a mixture of 80 mg % of alpha 1-acid glycoprotein and 40 g/L of human serum albumin in pH 7.4 phosphate buffer solution. Pure alpha 1-acid glycoprotein (80-400 mg %) or pure human serum albumin (40 g/L) in phosphate buffer gave significantly (p less than 0.05) lower binding results, indicating that both proteins are responsible for the high binding of dipyridamole in plasma. Addition of alpha 1-acid glycoprotein to heparinized human plasma, to simulate an acute phase increase in the protein, had no effect on the fraction of free drug in plasma. Binding of dipyridamole to heparinized human plasma or human serum albumin in buffer was concentration independent through 40 micrograms/mL. The free fraction of dipyridamole increases with concentrations exceeding 40 micrograms/mL.     

Plasma protein binding of penbutolol in pregnancy

Penbutolol is a not cardioselective beta-adrenergic blocking drug; it is lipid soluble and differs in its protein binding from the other members of its group because shows linkage to alpha 1-glycoprotein, with no detectable binding to albumin. AAG levels change during pregnancy and so the binding of [3H]-penbutolol was compared in 11 pregnant patients and in 10 healthy women. Binding was obtained by ultrafiltration and measurement of the free fraction by scintillation spectrometry. The free penbutolol fraction was significantly higher in the pregnant women than in the controls (6.06 +/- 0.34 compared with 3.55 +/- 0.29, P less than 0.001). The AAG levels in the pregnant women were significantly lower (0.40 +/- 0.03 g/l) than in the controls (0.77 +/- 0.06 g/l) (P less than 0.001) which showed a significant correlation with the bound/free penbutolol ratio (r = 0.61, P less than 0.005). On the other hand there was no significant correlation with the extent of penbutolol's protein binding even though the albumin levels were lower in the pregnant women (2.83 +/- 0.17 compared with 4.86 +/- 0.17; P less than 0.001). Penbutolol's nK1a for AAG was lower in pregnant women, and this suggests that the fall in AAG levels is not the only factor involved in the reduced binding of penbutolol in pregnancy.     

The binding of phenprocoumon to human plasma proteins

By in vitro experiments with equilibrium dialysis and 5 micrograms PPC per ml an average of 99.7 +/- 0.2% was bound to serum proteins. The binding in a solution with the albumin concentration which was present in the serum samples did not differ from this binding or from the binding in plasma where coagulation was prevented with either heparin or citrate. The binding constant in albumin solutions at 37 degrees was 4.5 +/- 0.31/mol x 10(-5) and an average of 1.16 +/- 0.04 primary binding sites was found. The association constant for the PPC albumin interaction was temperature dependent and the results of thermodynamic calculations suggested a combined ionic and non-polar type of binding as the change in enthalpy contributed with 40% of change in free energy. A large positive entropy change was found (15.79 Kcal/mol/degree K). The addition of phytomenadione to albumin solutions and of menadione to plasma caused a considerable decrease in the protein bound fraction of PPC, indicating the relevance of a study concerning the possible clinical consequences of these interactions.     

Serum binding of ketoconazole in health and disease

The plasma protein binding of ketoconazole, an oral antifungal agent of a weak basic nature, was measured after the addition of the drug (10 micrograms.ml-1) to serum from 35 healthy individuals, ten patients with chronic renal disease and seven patients with hepatic cirrhosis. The percentage of free ketoconazole was markedly increased in patients with chronic renal disease and in patients with hepatic cirrhosis, when it was compared with the group of healthy volunteers (7.33 +/- 0.11 in renal patients; 6.12 +/- 1.43 in hepatic patients compared with 2.93 +/- 0.12 in healthy individuals). The binding ratio of ketoconazole in health and disease was significantly related to plasma albumin concentration, but not to plasma alpha 1-acid glycoprotein (AAG) concentration. Moreover, ketoconazole binds to isolated human serum albumin in a greater proportion but does not bind to isolated AAG indicating that human serum albumin is the major binding protein for this drug in plasma.     

Plasma protein binding of azapropazone in patients with kidney and liver disease.

1 The free fraction of azapropazone in the plasma of 37 healthy volunteers ranged from 0.0027 to 0.0070 (0.0044 +/- 0.0009, mean +/- s.d.). The principal binding protein was found to be albumin. 2 In 27 patients with various degrees of renal failure the free fraction values of azapropazone were markedly enhanced (0.0260 +/- 0.0239, mean +/- s.d.) and increased more than tenfold in some patients. There was a weak correlation (r = 0.46, P less than 0.05) between the free fraction and the clearance of endogenous creatinine. Such correlation was not found for serum creatinine, serum albumin, serum uric acid and serum urea nitrogen. 3 In 32 patients with chronic liver disease the free fraction values of azapropazone were also markedly higher (0.0210 +/- 0.0242, mean +/- s.d.) than in healthy subjects. There were statistical significant correlation between free fraction values and the prothrombin complex activity in the plasma (r = 0.40, P less than 0.05) and the total bilirubin concentration in the plasma (r = 0.90, P less than 0.001), respectively. Such correlation was not found for serum albumin, serum glutamic oxalacetic transaminase, serum gamma-glutamyl transpeptidase and serum alkaline phosphatase. 4 In patients with kidney and liver disease the free fraction values of azapropazone correlated well with those of the anticoagulant drug phenprocoumon (r = 0.93, P less than 0.001). However, the binding of the latter drug was less impaired. Bilirubin, when added in vitro, displaced both drugs from plasma proteins but this displacing effect was much smaller than the binding changes observed in patients with liver disease. 5 Kidney and liver disease caused a marked impairment of the plasma protein binding of azapropazone. In patients with kidney disease the degree of impairment of azapropazone binding cannot or only poorly (creatinine clearance) be predicted from the biochemical parameters of kidney function whereas in patients with chronic liver disease the total bilirubin concentration in the plasma may serve as an index of the binding defect.     

Binding of prazosin to alpha 1-acid glycoprotein and albumin: effect of protein purity and concentrations

Prazosin is extensively bound in human serum/plasma. In the present study a bound fraction of 93-95% was observed at 37 degrees for therapeutic drug concentrations. Both alpha 1-acid glycoprotein (AAG) and albumin (HSA) are established as transport proteins for prazosin, but their individual contribution to the extent and variability of protein binding in serum/plasma is unclear. The present study showed that AAG possesses one binding site per molecule with high affinity (Kd approximately 0.8 microM) for prazosin. HSA, essentially globulin-free, bound prazosin with lower affinity (Kd approximately 30 microM) with an average of 0.3 binding sites per molecule. However, less purified HSA, containing globulins, exhibited apparently higher affinity (Kd approximately 8 microM), but lower binding capacity (0.07 sites per molecule) for prazosin. In mixtures of highly purified proteins, the concentrations of AAG, and not HSA, determined the extent and variability of prazosin binding.     

Factors influencing the protein binding of vancomycin.

Various factors influencing the protein binding of vancomycin were examined using equilibrium dialysis method. Four per cent human serum albumin (HSA) and/or 0.08 per cent alpha-1-acid glycoprotein (AAG), dissolved in isotonic phosphate buffer, were dialyzed against isotonic phosphate buffer of pH 7.4 using Spectrapor 2 membrane. The protein binding of vancomycin to 0.08 per cent AAG was dependent on vancomycin concentrations; the values ranged from 21.1 per cent at the vancomycin concentration of 20 micrograms ml-1 to 5.30 per cent at 2400 micrograms ml-1. However, binding to 4 per cent HSA was relatively constant, 8.79 +/- 2.43 per cent over a vancomycin concentration range of 20-2400 micrograms ml-1. The values to 4 per cent HSA alone and 0.08 per cent AAG alone did not predict the greater binding of vancomycin in the presence of both proteins, especially at higher concentrations of vancomycin; the values to 4 per cent HSA with 0.08 per cent AAG were constant, 26.3 +/- 3.74 per cent, at the vancomycin concentration range of 20-2400 micrograms ml-1. This suggested an interaction between the proteins, which resulted in enhanced binding of vancomycin. The protein binding of vancomycin to 4 per cent HSA with 0.08 per cent AAG was not influenced by the different incubation temperatures (4 degrees, 22 degrees, and 37 degrees), quantities of heparin (up to 40 units ml-1) or AAG (up to 0.16 per cent), or buffers (isotonic phosphate buffer of pH 7.4, phosphate buffer of pH 7.4 and 0.9 per cent NaCl solution) at the vancomycin concentration of 80 micrograms ml-1. Vancomycin was found to be stable in human serum albumin or in isotonic phosphate buffer of pH 7.4.     

Binding of Prazosin to α1-Acid Glycoprotein and Albumin: Effect of Protein Purity and Concentrations

Abstract: Prazosin is extensively bound in human serum/plasma. In the present study a bound fraction of 93–95% was observed at 37° for therapeutic drug concentrations. Both α₁-acid glycoprotein (AAG) and albumin (HSA) are established as transport proteins for prazosin, but their individual contribution to the extent and variability of protein binding in serum/plasma is unclear. The present study showed that AAG possesses one binding site per molecule with high affinity (Kd≈0.8 μM) for prazosin. HSA, essentially globulin-free, bound prazosin with lower affinity (Kd≈30 μM) with an average of 0.3 binding sites per molecule. However, less purified HSA, containing globulins, exhibited apparently higher affinity (Kd≈8 μM), but lower binding capacity (0.07 sites per molecule) for prazosin. In mixtures of highly purified proteins, the concentrations of AAG, and not HSA, determined the extent and variability of prazosin binding.     

Variability of the serum protein binding of theophylline in patients with asthma and cystic fibrosis.

1. The serum protein binding of theophylline was studied in 28 asthmatics and 11 patients with cystic fibrosis (CF) who were receiving the drug regularly. Peak theophylline samples were collected at 2 week intervals on four occasions in each asthmatic and on three occasions in each CF patient. The binding was measured using ultrafiltration at 37 degrees C and pH 7.4. The total and free (unbound) theophylline concentrations were measured using high-performance liquid chromatography. 2. The mean free-fractions (+/- s.d.) in asthmatics (0.50 +/- 0.03) and in CF patients (0.51 +/- 0.04) were not significantly different. The intra- and inter-subject variability in the free-fraction (fu) was relatively small in both patient groups. The binding was found to be concentration-independent at serum theophylline concentrations up to 30.9 micrograms ml-1. The effects of age, gender, serum albumin and total serum protein on the free fraction were evaluated. 3. The results indicate that the binding of theophylline is similar in the two disease states. The low degree of variability in serum theophylline binding indicates that measurements of total serum theophylline concentrations will reflect unbound serum concentrations with acceptable accuracy in both patient groups studied.     

Plasma protein binding of sorafenib, a multi kinase inhibitor: in vitro and in cancer patients

Sorafenib is an orally administered multikinase inhibitor that exhibits antiangiogenic and antitumor activity. Few investigators have been able to correlate cumulative sorafenib dose or total exposure to pharmacodynamic effects. This discrepancy may be in part due to poorly understood protein binding characteristics. Since unbound drug concentrations are believed to be more relevant to pharmacological and toxicological responses than total drug, an equilibrium dialysis method using 96-well microdialysis plates was optimized and validated for determining the fraction unbound (F(u)) sorafenib in human plasma and in isolated protein solutions. Unbound sorafenib concentrations were determined in cancer patients receiving the drug orally at a dose of 400?mg and 600?mg twice daily. Sorafenib was extensively bound with mean F(u) value of 0.3% in both non-cancer and cancer patient's plasma. The binding in plasma was concentration independent, indicating a low-affinity, possibly nonspecific and nonsaturable process. In isolated protein solutions, 99.8% and 79.3% of sorafenib was bound to human serum albumin (HSA) (4?g/dL) and α(1)-acid glycoprotein (AAG) (0.1?g/dL) with binding constants of 1.24?×?10(6)?M(-1) and 1.40?×?10(5)?M(-1), respectively. In cancer patients receiving sorafenib, unbound sorafenib was not correlated with patient characteristics or laboratory values. In conclusion, sorafenib is highly protein bound in human plasma with a higher affinity towards albumin and limited free drug may be partly responsible for its borderline clinical activity.