The available techniques for the investigation of drug binding to plasma and tissues protein are reviewed and the advantages and disadvantages of the various techniques stated. A comparison of different plasma protein binding techniques is made which shows that the size of the unbound fraction of drug may be influenced by the method used. Protein binding may be assayed by methods including equilibrium dialysis, ultrafiltration, ultracentrifugation, gel filtration, binding to albumin microspheres and circular dichroism. Tissue binding techniques can involve testing binding to isolated organs, tissue slices, homogenates and isolated subcellular particles. Details of the available methods to compute pharmacokinetic constants are given. Stereoselective binding has been investigated for a limited number of drugs and the difference in the binding of 2 enantiomers is usually modest. The measurement of the binding constants is often required to characterise the drug-protein interaction. Mathematical and graphical methods to compute the pharmacokinetic parameters are discussed. The implications of binding on the volume of distribution and clearance of drugs are examined.

No abstract

1. Plasma albumin concentration was measured in 118 healthy subjects (aged between 18 and 87 years), in 95 renal patients with creatinine clearances between 15 and 50 ml min‐1 (aged between 14 and 79 years) and in 101 uraemic patients maintained on chronic haemodialysis (aged between 27 and 83 years). 2. There was a significant (P less than 0.001) negative correlation between albumin concentration and age in healthy subjects, but no correlation in patients with low creatinine clearance or in uraemic patients. 3. The ex vivo plasma binding of diazepam (1 microM), salicylic acid (2 mM) and digitoxin (37 nM) was studied in groups of age‐selected young and aged healthy subjects in patients with low creatinine clearance and in patients with uraemia. The unbound fractions of diazepam and salicylic acid were about double in old compared with young healthy subjects whereas they were similar in young and old patients with lowered creatinine clearance. In uraemic patients, ageing did not affect the binding of salicylic acid whereas the unbound fraction of diazepam was slightly but significantly greater in elderly subjects. The unbound fraction of digitoxin was independent of age in both healthy subjects and in those with renal disease. 4. Decreased plasma binding of diazepam and salicylic acid was partially corrected by extensive dialysis of plasma. The lower plasma binding of diazepam and salicylic acid associated with ageing may be ascribed to the effects of endogenous displacers and to hypoalbuminaemia. The influence of these two factors appears to be drug‐dependent.

Albumin was isolated from pooled fetal serum obtained at normal delivery at term and from pooled adult plasma. Albumin isolation was carried out by means of PEG precipitation followed by ion exchange chromatography on DEAE-Sephadex A 50 and then on SP-Sephadex C 50. The binding of diazepam (1 μM), salicylic acid (2 mM) and digitoxin (6 nM) to albumin (40 g/l) was measured by equilibrium dialysis at 37°C. The unbound fraction (mean ± SD) for fetal and adult albumin of diazepam was 1.86 ± 0.24 and 1.82 ± 0.15% (NS), that of digitoxin was 3.18 ± 0.27 and 3.36 ± 0.04% (NS) and that of salicylic acid was 11.65 ± 0.99 and 9.47 ± 0.75% (p < 0.05), respectively. With both fetal and adult albumin, a single class of binding sites was observed for diazepam and digitoxin, whereas two classes of binding sites were observed for salicylic acid. The number of binding sites (n, moles of drug per mole of albumin) for fetal and adult albumin was 0.83 and 1.02 for diazepam and 0.014 and 0.018 for digitoxin, respectively. For salicylic acid, n was 1.45 (fetal albumin) and 1.55 (adult albumin) for the higher affinity site, and 3.06 (fetal albumin) and 3.27 (adult albumin) for the lower affinity site. The association constant (K(a), M^-1) for diazepam was 1.36 × 10^5 (fetal albumin) and 1.00 × 10^5 (adult albumin) and that for digitoxin was 4.12 × 10^6 (fetal albumin) and 2.7 × 10^6 (adult albumin). For salicylic acid, K(a) was 38.4 × 10^3 (fetal albumin) and 35.8 × 10^3 (adult albumin) for the higher affinity site, and 2.7 × 10^3 (fetal albumin) and 4.3 × 10^3 (adult albumin) for the lower affinity site. This work shows that fetal and adult albumin have similar binding properties and corroborates our previous findings with furosemide.

1 The activities of microsomal glucuronyltransferase and thiomethyltransferase, and those of cytosolic sulphotransferase, acetyltransferase, glutathione transferase and thiomethyltransferase were measured in abnormal (cirrhosis and chronic hepatitis) and normal livers. 2 Glucuronyltransferase and sulphotransferase were investigated with 2-naphthol and ethinyloestradiol as substrates. p-Aminobenzoic acid, benzo(a)pyrene-4,5-epoxide and 2-mercaptoethanol were the substrates of acetyltransferase, glutathione transferase and thiomethyltransferase, respectively.3 Enzyme activities are expressed as nmol min-1 incubation mg-1 protein and the averages (± s.d.) are given. With 2-naphthol as substrate, the glucuronyltransferase activity was 6.55 ± 4.10 (abnormal liver, n = 33) and 7.81 ± 4.02 (normal liver, n = 26) (NS); whereas sulphotransferase activity was 0.28 ± 0.18 (abnormal liver, n = 35) and 0.68 ± 0.43 (normal liver, n = 26) (P < 0.01). Glucuronyltransferase activity towards ethinyloestradiol was 102.5 ± 56.9 (abnormal liver, n = 30) and 107 ± 59.9 (normal liver, n = 26) (NS), whereas sulphotransferase activity was 57.2 ± 36.0 (abnormal liver, n = 35) and 122 ± 67.6 (normal liver, n = 28) (P < 0.01). Acetyltransferase activity was 0.84 ± 0.83 (abnormal liver, n = 35) and 3.84 ± 1.65 (normal liver, n = 26) (P < 0.01). Glutathione transferase activity was 0.83 ± 0.68 (abnormal liver, n = 35) and 2.90 ± 1.59 (normal liver, n = 25) (P < 0.01) and thiomethyltransferase activity was 1.00 ± 0.69 (abnormal liver, n = 34) and 3.99 ± 1.49 (normal liver, n = 25) (P < 0.01). 4 Liver disease lowers the activities towards the substrates studied of sulphotransferase, acetyltransferase, glutathionetransferase and thiomethyltransferase but not that of glucuronyltransferase. Thus, overall hepatic conjugating capacity is decreased in liver injury. However, enzyme activity is substrate dependent and it is not possible to extrapolate the results for other compounds.

This large cross-sectional study suggests that postmenopausal women are at higher risk of type 2 diabetes after allowance for the effect of age. Other main determinants of risk of type 2 diabetes in women around menopause were low socioeconomic status and being overweight. Diabetes was found less frequently in those taking hormone replacement therapy.

The protein binding of furosemide was studied in the serum from 8 umbilical cords, in 51 children (aged between 2 weeks and 13.5 years) and in the plasma of 10 volunteers (aged between 28 and 42 years). The drug was added to the buffer to give a final concentration of 2 pg/ml. The unbound fraction of furosemide was 2.5 ± 0.1 % (cord serum) and 1.7 ± 0.7% (adult plasma). These figures are different at a level of 0.001. The unbound fraction of furosemide reached the adult values during the 1st year of life. A correlation (level of significance >0.01) was found between the unbound fraction and the age during the first 6 months of life. The furosemide binding kinetics were studied in 3 cord serum and in 3 adult plasma samples. The concentration of the drug in the buffer ranged between 1 and 16 pg/ml. The kinetic constants (mean ± SEM) were: association constant (K = 10^5 M^-) 2.4 ± 0.3 (cord serum), 2.0 ± 0.2 (adult plasma); the number of binding sites per gram protein (n = 10^-6) was 3.2 ± 0.5 (cord serum) and 3.9 ± 0.7 (adult plasma). When the concentration of furosemide was increased up to 200 pg/ml buffer, the free fraction of the drug increased up to 4.8 ± 0.2% (cord serum) and 2.9 ± 0.4% (adult plasma).

No abstract