Gentamicin and Tobramycin Binding to Human Serum In Vitro

Journal of Analytical Toxicology, Vol. 28, April 2004 Gentamicin and TobramycinBindingto Human Serum In Vitro David N. Bailey* and John R. Briggs Division of Laboratory Medicine, Departmentof Pathology, University of California, San Diego, San Diego, California I Abstract [ The binding of gentamicin and tobramycin to human serum was in vitro using equilibrium dialysis of pooled human serum supplemented to various concentrations of either drug. O n l y minimal and variable non-specific binding was noted for each drug: gentamicin, less than 1 5 % and tobramycin, less than 3 0 % . Conventional Scatchard analysis conducted over an array of drug concentrations failed to identify any specific binding proteins. studied Introduction Gentamicin and tobramycin are popular intravenously administered aminoglycosideantibiotics. Despitetheir widespread use, the binding of these drugs to human serum has been only poorly studied to date, and the few reports in the literature are discrepant. In one paper (1), gentamicin binding in vitro has been reported as none while in others it has been reported as up to 10% (2), up to 30% (3), up to 5% (4), and up to 23% (5). Similarly, tobramycin binding has been reported as none (6) and as up to 10% (2,3). These studies have utilized several conventional binding-experiment methodologies including ultracentrifugation (5), equilibrium dialysis (4,7), and ultrafiltration (6). The binding characteristics (e.g., concentration of binding sites, binding affinity constants) of these drugs have not been reported. This study was performed to determine the binding of gentamicin and tobramycin to human serum in vitro and to attempt to characterize the binding of each of the two drugs. Experimental Supplies Gentamicin sulfate, tobramycin, and pooled human serum (clotted human male whole blood) were purchased from Sigma * Author to whom correspondenceand reprint requestsshouldbe addressed: David N. Bailey, M.D., Division of Laboratory,Medicine, Departmentof Pathology, UCSD MedicalCenter,200 West Arbor Drive, San Diego,California92103-8320. E-mail: . Chemical Company(St. Louis,MO). Dialysiscells,5 mL on each side of the membrane, were obtained from Scientific Products Division of Baxter Diagnostics (McGawPark, IL). Dialysismembranes, molecular weight cutoff 6000 and 0.073-mm thickness, were purchased from Scienceware, Division of Fisher Scientific Company (Pequannock, NJ). Solutions The pooled human serum was supplemented with magnesium chloride, 39.1 g/L, to raise the magnesium concentration from 0.008 g/L to a physiologicalconcentration of 0.018 g/L because the binding of gentamicin is known to decrease with increasing concentrations of ionized calcium and magnesium (4,7). The calcium in the pooled serum was already at a physiological concentration of 0.091 g/L. To 50 mL of the magnesium-supplemented pooled human serum was added 500 mg of either gentamicin sulfate or tobramycin to yield a nominal concentration of 10,000 mg/L of either drug. Similarly, to 50 mL of phosphate buffer (0.1 tool/L, pH 7.4) was added 500 mg of either gentamicin sulfate or tobramycin to yield a concentration of 10,000 mg/L of either drug. Each solution was then sequentially diluted with either serum or phosphate buffer, respectively,to yield nominal drug (gentamicin or tobramycin) concentrations of 5000, 2500, 1000, 500, 250, 100, 50, 25, 10, and 5 mg/L. Whereas all but the lower two concentrations are non-therapeutic levels,they were used in order to permit subsequent Scatchard analysis of binding, including an attempt to saturate any binders that might be detected. Equilibriumdialysis Each drug solution in serum was dialyzed against the same drug concentration in buffer on a platform rocker for 24 h at 37~ Each concentration was studied in triplicate (i.e., three dialysis chambers). At the end of each dialysis,samples (serum and buffer) were withdrawn from the respective sides of the chambers and diluted appropriately with pooled human serum in order to yield concentrations within the assayable peak therapeutic range for gentamicin and tobramycin: 10,000 mg/L (lO00-folddilution); 5000 mg/L (500fold dilution); 2500 mg/mL (250-fold dilution); 1000 mg/L (lO0-fold dilution); 500 mg/L (50-fold dilution); 250 mg/L Reproduction (photocopyin 8) of editorial content of this journal is prohibited without publisher's permission. 187 Journal of Analytical Toxicology,Vol. 28, April 2004 (25-fold dilution); 100 mg/L (10-fold dilution); 50 mg/L (5-fold dilution); 25 mg/L (2.5-fold dilution); 10 mg/L (no dilution); and 5 mg/L (no dilution). Drug analyses Gentamicin was measured by a commercial enzyme immunoassay ("EMIT 2000 Gentamicin Plus Assay," Syva Company, Dade Behring Inc., Cupertino, CA), which uses glucose-6-phosphate dehydrogenase as the enzyme and monitors the absorbance increase as NAD is converted to NADH.Tobramycin was measured by a commercial particle-enhanced turbidometric inhibition immunoassay (Beckman Coulter Inc., Fullerton, CA). Both assays were automated on the Beckman Synchron CX7 Delta System (Beckman Instruments Inc., Brea, CA). The within-run analytical imprecision of each assay was less than 5%. Calculations The post-dialysis drug concentrations in serum represented "total" ("bound" plus "free") drug, and those in buffer represented "free" drug. Hence, the difference between the drug concentrations in serum and those in the respective buffer represented "bound" drug. Resuffs from the triplicate studies of each concentration were averaged. The imprecision was 5-10%. The percent binding was calculated by the following formula: % Bound = 100% x (Cs eru m - CBuffer)/Cseru m In attempt to determine the binding parameters, Scatchard analysis was done by performing least-squares regression analysis of the ratio of the concentration of bound drug: concentration of free drug (Y-axis) versus the concentration of bound drug (X-axis). By this method, the negative slope represents the affinity constant of association (Ka), and the Xintercept represents the concentration of Table I. Gentamicin Binding to Human Serum* binding sites (Bo) (8,9). This methodology has been utilized for many years in the author's Nominal Drug Drug Concentration laboratory (10). Bound/Free Gentamicin Concentration Concentration of Bound Drug % Concentration in Serum (mg/L) in Buffer (mg/L) Drug (mg/t) (Bound + Free) (Free) (mg/L) Concentration Bound 10,000 5000 2500 1000 500 250 100 50 25 10 5 6800 3450 1750 640 325 155 71 36.5 16 7.1 3.7 6600 3350 1575 640 320 152.5 61 31.5 15.8 6.1 3.3 200 100 175 0 5 2.5 10 5 0.2 1 0.4 0.03 0.03 0.11 0.00 0.02 0.02 0.16 0.16 0.01 0.16 0.12 2.9% 2.9% 10.0% 0.0% 1.5% 1.6% 14.1% 13.7% 1.2% 14.1% 10.8% * Experimentaldatarepresentthe averageof triplicates.The triplicateexperimentshad an imprecisionof 5-10%. Table II. Tobramycin Binding to Human Serum* Nominal Drug Drug Concentration Tobramycin Concentration Concentration of Bound Bound/Free Concentra (...truncated)