Distributions Of Grepafloxacin And HSR-903 Into Pulmonary Epithelial Lining Fluid And Alveolar Macrophage

In this study, protonation equilibria for three amphoteric quinolone antibacterials, including grepafloxacin (GPFX), HSR-903 and ciprofloxacin (CPFX), were evaluated and compared. Then, pH-dependent behavior in both microspecies distribution and molecular electrical charge was described in detail. The UV spectrophotometry was employed to determine dissociation constants, with macro-dissociation constants describing integral acid-base property and micro-dissociation constants defining proton-binding ability of the individual functional group. The prerequisite for determining micro-protonation equilibria wasn't fulfilled for HSR-903 in most cases; the critical wavelength meeting analytic requirements at 294 nm was used instead, from which the results agreed with those of another determination. Distributions of neutral and zwitterionic microspecies reached the maximum at isoelectric point (pI), with the latter being predominant. The ratio of neutral to zwitterionic microforms was constant for a given quinolone and varied greatly among quinolones. In addition, the molecular electrical charge for quinolones was very sensitive to pH variation around physiological pH, and liposome/buffer system was undertaken to measure the drug-phospholipid membrane interactions in order to confirm this assumption. The drug-biomembrane interactions are the basis for drug transport across biomembrane and for drug approaching proteins or enzymes embedded in plasma membrane. Three distinct membrane-like systems were exploited in the present work, including n-octanol/buffer, immobilized artificial membrane (IAM) chromatography, and liposome/buffer systems. ①The apparent partition coefficient (log DO/B,pH) versus pH profiles had the shape of a parabolic curve in an n-octanol/buffer system, and reached the peak around pI for GPFX, CPFX and HSR-903, respectively. An ion-pair was formed at acidic condition, resulting in higher log DO/B,pH than predicted values. Furthermore, log DO/B,pH of amphoteric quinolones was not only influenced by intrinsic lipophilicity, but also markedly affected by their complex protonation equilibria. ②Three membrane-like systems were assessed by the comparisons of respective lipophilicity parameters. Liposome/buffer system and IAM chromatography provided equivalent measuring scale of lipophilicity, both differing much from n-octanol/buffer system. IAM surface demonstrated many properties in phospholipid density, interfacial motional features and distributions of functional groups, similar to those in liquid liposomal membrane. By contrast, bulky n-octanol/buffer system couldn't model the partitioning process in liposome/buffer system, because membrane consists of ordered hydrophilic and lipophilic regions with anisotropic properties, and because membrane interactions are the sum of diverse intermolecular forces. ③The lipophilicity of 27 structurally diverse compounds was evaluated by both n-octanol/buffer system and IAM chromatography. The ampholytes generally exhibited higher membrane affinity than expected from their hydrophobicity, which stemmed from attractive polar extra-interactions with phospholipid membrane, behavior similar to bases. This stressed the important perspective in drug rational design by introducing certain structural motifs in ionized solutes to enhance membrane affinity (log kIAM), but maintain lower hydrophobicity (log DO/B, 7.4). Higher membrane affinity is beneficial to good tissue distribution and pharmacological activity, and lower hydrophobicity leads to normal physicochemical properties and suitable water solubility. ④Partitioning thermodynamics was investigated and underlying partitioning mechanisms were explored. In n-octanol/buffer system, partitioning was entropy-dominated for most studied drugs, while partitioning of oxolinic and nalidixic acids into n-octanol phase was enthalpy-driven. In IAM chromatography, partitioning into membrane for studied solutes was enthalpy-driven, and exhibited an exothermal process. In the dipalmitoyl-phosphatidylcholine liposome/buffer system, log DL/B,7.4―1/T profiles for GPFX and quinidine showed the bell-shape form and log DL/B,7.4 reached the peak at the phase-transferring temperature. With CPFX chosen as the control, distributions into pulmonary epithelial lining fluid (ELF) and alveolar macrophage (AM) of rats in vivo for GPFX and HSR-903 were studied. RP-HPLC determination methods of three drugs in bio-samples were developed. GPFX and HSR-903 distributed far more effectively into ELF and AM than CPFX. After oral administration 8 h, (AUC)AM/(AUC)plasma was 353±57, 101±23, 12±3; (AUC)ELF/(AUC)plasma was 5.43±0.81, 3.09±0.51, 0.58±0.09 for GPFX, HSR-903 and CPFX, respectively. Based on physiological condition, pharmacokinetic model for describing distribution into ELF and AM in vivo was established, which could explicate distribution processes well and confer the valuable pharmacokinetic parameters. An asymmetrical transport between blood and ELF compartments was found for GPFX and HSR-903, favoring drugs distribution into ELF compartment and resulting in higher concentrations in ELF than those in blood. However, the concentrations in ELF and blood were comparable for CPFX. With regard to AM distribution, the uptake clearances from ELF to AM compartments of GPFX and HSR-903 were greater than the net efflux clearances of their own, giving rise to high accumulation within AM.