Some insights into the binding mechanism of the GABAA receptor: a combined docking and MM-GBSA study.

Gamma-aminobutyric type A receptor (GABAAR) is a member of the Cys-loop family of pentameric ligand gated ion channels (pLGICs). It has been identified as a key target for many clinical drugs. In the present study, we construct the structure of human 2α12ß2γ2 GABA(A)R using a homology modeling method. The structures of ten benzodiazepine type drugs and two non-benzodiazepine type drugs were then docked into the potential benzodiazepine binding site on the GABA(A)R. By analyzing the docking results, the critical residues His102 (α1), Phe77 (γ2) and Phe100 (α1) were identified in the binding site. To gain insight into the binding affinity, molecular dynamics (MD) simulations were performed for all the receptor-ligand complexes. We also examined single mutant GABA(A)R (His102A) in complexes with the three drugs (flurazepam, eszopiclone and zolpidem) to elucidate receptor-ligand interactions. For each receptor-ligand complex (with flurazepam, eszopiclone and zolpidem), we calculated the average distance between the C(α) of the mutant residue His102A (α1) to the center of mass of the ligands. The results reveal that the distance between the C(α) of the mutant residue His102A (α1) to the center of flurazepam is larger than that between His102 (α1) to flurazepam in the WT type complex. Molecular mechanic-generalized Born surface area (MM-GBSA)-based binding free energy calculations were performed. The binding free energy was decomposed into ligand-residue pairs to create a ligand-residue interaction spectrum. The predicted binding free energies correlated well (R(2) = 0.87) with the experimental binding free energies. Overall, the major interaction comes from a few groups around His102 (α1), Phe77 (γ2) and Phe100 (α1). These groups of interaction consist of at least of 12 residues in total with a binding energy of more than 1 kcal mol(-1). The simulation study disclosed herein provides a meaningful insight into GABA(A)R-ligand interactions and helps to arrive at a binding mode hypothesis with implications for drug design.

Sensitivity of P-glycoprotein tryptophan residues to benzodiazepines and ATP interaction.

Plasma membrane P-glycoprotein is a member of the ATP-binding cassette family of membrane transporters. In the present study tryptophan intrinsic fluorescence was used to understand the P-glycoprotein response to three benzodiazepines (bromazepam, chlordiazepoxide and flurazepam) in the presence and absence of ATP. Fluorescence emission spectra showed a red shift on the maximal emission wavelength upon interaction of P-glycoprotein with all benzodiazepines. Benzodiazepine association with nucleotide-bound P-glycoprotein also showed this trend and the quenching profile was attributed to a sphere-of-action model, for static fluorescence. Furthermore, quenching data of benzodiazepine-bound P-glycoprotein with ATP were concentration dependent and saturable, indicating that nucleotide binds to P-glycoprotein whether drug is present or not. These results seems in agreement with the proposal of the ATP-switch model by Higgins and Linton, where substrate binding to the transporters initiates the transport cycle by increasing the ATP binding affinity.

Effect of phosphatidylserine content on the partition coefficients of diazepam and flurazepam between phosphatidylcholine-phosphatidylserine bilayer of small unilamellar vesicles and water studied by second derivative spectrophotometry.

The affinity of the psychotropic benzodiazepine drugs diazepam (DZ) and flurazepam (FZ) to phosphatidylserine (PS) was examined since PS is abundantly contained in brain membranes. The effect of PS content on the partition coefficients (K(p)s) of these drugs between phosphatidylcholine (PC)-PS bilayer membranes of small unilamellar vesicles (SUV) and water was measured using second derivative spectrophotometry. The second derivative spectra of DZ and FZ measured in the solutions containing various amounts of PC-PS SUV clearly showed derivative isosbestic points and a distinct derivative intensity change depending on the amount of the SUV added. The derivative intensity differences (AD) of the drugs before and after addition of the SUV suspension were measured at a specific wavelength. Using the AD values, the Kp values were calculated and obtained with relative standard deviation of below 10%. The Kp values of both drugs increased according to the PS content in the PS-PC bilayer membranes of the SUV proving that both have higher affinity to the PC-PS bilayer membranes than to PC membranes. The effect was much larger for FZ, i.e., the Kp value of FZ at 30 mol% PS content increased to about five times the value for the PC SUV. This can be explained by the fact that at the experimental pH of 7.4, 80% of FZ molecules are in a cationic form (pKa=8.1), so that these molecules are highly accessible to the negatively charged PS molecules. The results support the rapid and high distribution of DZ and FZ in the central nervous system after their administration.

Determination of partition coefficients of diazepam and flurazepam between phosphatidylcholine bilayer vesicles and water by second derivative spectrophotometric method.

Second derivative spectrophotometry allowed the establishment of a simple and accurate method for the determination of partition coefficients of benzodiazepine drugs in a liposome/water system. The absorption spectra of diazepam (DZ) and flurazepam (FZ) in phosphatidylcholine (egg yolk) bilayer vesicle suspensions showed small spectral changes depending on the concentration of phosphatidylcholine vesicles. However, the intense background signals caused by the light scattering of the phosphatidylcholine vesicles made it difficult to yield a correct base line, thus the quantitative spectral data could not be obtained. In the second derivative spectra, the spectral changes were enhanced and three derivative isosbestic points were observed for each drug indicating the entire elimination of the residual background signal effects. The derivative intensity change of each drug (DeltaD) induced by its interaction with phosphatidylcholine bilayers was measured at a specific wavelength. From the relationship between the DeltaD value and the lipid concentration, the molar partition coefficients (K(p)s) of DZ and FZ were calculated and obtained with a good precision of R.S.D below 10%. The fractions of the partitioned DZ and FZ calculated by using the obtained K(p) values agreed well with the experimental values. The results prove that the derivative method can be usefully and easily applied to the determination of partition coefficients of benzodiazepines in the liposomes/water system without any separation procedures.

New sorbent for HPTLC--FTIR in situ determination of impurities in flurazepam.

An efficient HPTLC--UV/FTIR coupling procedure is presented for the separation and rapid identification of flurazepam hydrochloride and its related substances in bulk powder and capsules. An optimized mobile phase was developed for the separation on specialized plates for HPTLC--FTIR in situ measurement containing a proportion of 50% magnesium tungstate. The proposed procedure shows several advantages to the related compound test of the pharmacopoeia, e.g. baseline separation of the known impurities and detection of the substances as peaks in the UV, Gram-Schmidt or window chromatograms. Furthermore, unambiguous identification is obtained by postrun extraction of the DRIFT spectra and comparison with reference spectra in the library. Quantification of the related compounds was carried out densitometrically. This method shows that the current resurgence of interest in modern instrumental TLC is rightful based on the flexibility and efficiency of this analytical method.

Large-volume sample stacking of selected drugs of forensic significance by capillary electrophoresis.

Large-volume sample stacking capillary electrophoresis (LVSS-CE) and conventional capillary electrophoresis (CE) are compared for the separation of drugs of significance to forensic and clinical analyses. LVSS-CE for cations requires the use of an electroosmotic flow (EOF) modifier in conjunction with polarity switching to effect on-column concentration of an analyte and its subsequent migration in the capillary. The run buffer consists of 0.05 mol dm(-3) disodium tetraborate adjusted to pH 2.2 with orthophosphoric acid, and the EOF modifier is 0.002 mol dm(-3) cetyltrimethylammonium bromide. CE investigations used an identical run buffer minus the EOF modifier. LVSS-CE and CE investigations used injection times of 30 s and 3 s, respectively. Both modes of capillary electrophoresis are compared in terms of their limits of detection, efficiency, resolution and reproducibility. LVSS-CE is also applied to the analysis of a spiked urine sample.

A Fourier transform-Raman and IR vibrational study of flurazepam base and the mono- and dihydrochloride salts.

Fourier transform (FT) Raman and IR spectra of flurazepam base and the mono- and dihydrochloride salts have been recorded. The parent compounds is 7-chloro-1-[2-(diethylamino)ethyl]-5-(2-fluorophenyl)-1,3-dihydro-[2H]- 1,4-benzodiazepin-2-one and is closely related to diazepam. The spectra show characteristic features associated with both the diazepine ring and substituents. Very strong lines near 1615 cm-1 in the Raman spectra of the base and the monohydrochloride are assigned to the C = N stretch of the diazepine ring. The C = N group becomes a C = N+ group in the dihydrochloride and the frequency shifts to 1635 cm-1. A very strong absorption near 1680 cm-1 in the IR spectra of the three compounds is attributed to the C = O stretching mode. The hydrochlorides are characterized by very strong broad bands in the IR spectra between 2600 and 2200 cm-1. Various IR and Raman vibrational features serve to characterize and differentiate the salts from each other and the free base. Raman and IR spectra have also been recorded for the related compound desalkylflurazepam [7-chloro-5-(2-fluorophenyl)-1,3-dihydro-[2H]-1,4-benzodiazepin-2- one], in which a hydrogen atom replaces the (diethylamino)ethyl group at position 1 of the diazepine ring. Comparison of the spectra of this compound with those of flurazepam has enabled some vibrations of the (diethylamino)ethyl group to be identified.

Vibrational and NMR spectroscopic study of aged flurazepam mono- and dihydrochloride salts for content identity.

Archival samples of flurazepam monohydrochloride and "hydrochloride" (i.e., the dihydrochloride) were examined by Fourier transform infrared and Raman spectroscopy to determine evidence of degradation during storage for 13-15 years. No degradation of the three different batches of monohydrochloride salts was detected, but various degrees of degradation of the eight specimens of flurazepam hydrochloride diprotonated salts were indicated by enhanced intensities (IR 1635, 1509, 1226; Raman 1636, 1408, 1149 cm-1) and new features (IR 1742, 943, 755; Raman 1554, 837, 742 cm-1). All of these features, except the 1742 cm-1 IR band, were attributed to the presence of the hydrolysis product 5-chloro-2-[[2-(diethylamino)ethyl]amino]-2'-fluorobenzophenone hydrochloride whereas the 1742 cm-1 band was attributed to glycine hydrochloride, the other hydrolytic moiety. The flurazepam hydrochloride samples were also examined in deuterated dimethyl sulfoxide solution by proton nuclear magnetic resonance (1H-NMR) spectroscopy to verify the presence of the degradation products and to estimate the levels of degradation (approximately 3-36%) of the drug. IR and Raman spectra of the "benzophenone" hydrochloride in the "fingerprint" region are compared with two samples of flurazepam dihydrochloride (slightly and highly degraded) and their features discussed. Vibrational assignments are made and discussed for the observed IR and Raman wavenumbers for the "benzophenone" hydrochloride.

[Localization of two neurotropic molecules in cultured glial cells by ion microscopy]. / Localisation de deux molécules neurotropes dans les cellules gliales en culture par microscopie ionique.

The intracellular localization of two families of neurotropic drugs: flunitrazepam and flurazepam (benzodiazepine), triflupromazine and trifluoperazine (phenothiazine) has been studied by ion microscopy. The molecules have been incubated with C6 glioblastoma cells from rat origin and with astroglial primary cultures. The images of the intracellular distributions of the two drugs are easily obtained by selecting the fluorine atom of the molecules. The images obtained show that flunitrazepam and flurazepam, two drugs of the benzodiazepine group are mainly located to the nuclei, whereas triflupromazine and trifluoperazine, two phenothiazines are exclusively located inside the cytoplasm.