Characterization of impurities in josamycin using dual liquid chromatography combined with mass spectrometry.

The European Pharmacopoeia (Ph. Eur.) prescribes a selective and sensitive liquid chromatography/ultraviolet (LC-UV) method for the separation of the 16-membered ring macrolide josamycin and its related compounds. Since josamycin is obtained by fermentation, several closely related substances can be found in the sample. Several impurities have already been identified using reference substances. However, many peaks in the chromatogram cannot be correlated with known compounds or correspond to structures which were not described previously. The hyphenation of LC to mass spectrometry (MS) is a very useful tool for the characterization of impurities. The existing LC-UV method however uses non-volatile buffers, while for LC/MS a volatile mobile phase is required. In this study, each peak from the non-volatile system was collected separately and reinjected into a LC system using volatile mobile phase constituents. This way, the analyte could be separated from the buffer salts. Mass spectral data of this macrolide antibiotic were acquired on a LCQ ion trap mass spectrometer, equipped with an electrospray ionization (ESI) probe operating in the positive ion mode. The identity of the unknown compounds was deduced using the MS/MS and MS(n) collision-induced dissociation spectra of reference substances, combined with knowledge about the nature of functional group fragmentation behavior. The impurity profiling was done on 30 peaks in a josamycin bulk sample. This way, 12 compounds reported in the literature and Ph. Eur. were found in the bulk sample. Furthermore, 12 novel related substances were characterized and 18 compounds were partially characterized.

A Novel Depsipeptide Produced by Paenibacillus alvei 32 Isolated from a Cystic fibrosis Patient.

The important viscosity of the respiratory tract mucus of Cystic fibrosis (CF) patients impairs the mucociliary transport system and allows the growth of numerous micro-organisms. Among them, Pseudomonas aeruginosa and Staphylococcus aureus are known to be responsible for pulmonary infections. We imagined that CF microflora could also harbour micro-organisms naturally equipped to compete with these pathogens. A method was developed to recover these antibiotic-producing strains within 20 CF sputum. Using this approach, we have isolated an unusual Gram-positive bacterium identified as Paenibacillus alvei by Api galleries and 16S rRNA gene sequence analysis. This strain has inhibited the growth of P. aeruginosa, S. aureus and Klebsiella pneumoniae, in co-cultures. A liquid mineral medium named MODT50 was designed and optimised for the production and the recovery of the antimicrobial compounds. The supernatant has inhibited the growth of all Gram-positive strains tested, even Methicillin-resistant S. aureus. One antimicrobial compound with a peptide structure (mainly active against S. aureus, Micrococcus luteus, and Pseudomonas stutzeri) has been purified and characterised by liquid chromatography-mass spectrometry. The new active molecule (m/z 786.6) named depsipeptide L possesses a 15-guanidino-3-hydroxypentadecanoic acid side chain (m/z 298) linked on a cyclic part of four amino acids residues (Ser, two Leu/Ile, Arg). This work reports for the first time the production of such a molecule by a P. alvei strain in a mineral medium. The CF lung microflora might represent a valuable source for the discovery of new antimicrobial-producing strains.

Optimization and validation of liquid chromatography and headspace-gas chromatography based methods for the quantitative determination of capsaicinoids, salicylic acid, glycol monosalicylate, methyl salicylate, ethyl salicylate, camphor and l-menthol in a topical formulation.

Capsaicinoids, salicylic acid, methyl and ethyl salicylate, glycol monosalicylate, camphor and l-menthol are widely used in topical formulations to relieve local pain. For each separate compound or simple mixtures, quantitative analysis methods are reported. However, for a mixture containing all above mentioned active compounds, no assay methods were found. Due to the differing physicochemical characteristics, two methods were developed and optimized simultaneously. The non-volatile capsaicinoids, salicylic acid and glycol monosalicylate were analyzed with liquid chromatography following liquid-liquid extraction, whereas the volatile compounds were analyzed with static headspace-gas chromatography. For the latter method, liquid paraffin was selected as compatible dilution solvent. The optimized methods were validated in terms of specificity, linearity, accuracy and precision in a range of 80% to 120% of the expected concentrations. For both methods, peaks were well separated without interference of other compounds. Linear relationships were demonstrated with R² values higher than 0.996 for all compounds. Accuracy was assessed by performing replicate recovery experiments with spiked blank samples. Mean recovery values were all between 98% and 102%. Precision was checked at three levels: system repeatability, method precision and intermediate precision. Both methods were found to be acceptably precise at all three levels. Finally, the method was successfully applied to the analysis of some real samples (cutaneous sticks).

Identification of impurities in polymyxin B and colistin bulk sample using liquid chromatography coupled to mass spectrometry.

The European Pharmacopoeia (Ph. Eur.) describes liquid chromatography-ultraviolet (LC-UV) methods using C(18) stationary phases for the analysis of polymyxin B and colistin. Several unknown impurities were detected in commercial samples of those polypeptide complexes. However, the Ph. Eur. does not specify any related substances for polymyxin B and colistin. Since both methods use non-volatile buffers, the mobile phases were incompatible with mass spectrometry (MS). For the identification of related substances in bulk samples by LC/MS, volatile mobile phase systems were developed. However, the LC/MS methods (with volatile additives) showed inferior chromatographic separation compared to the LC-UV method (with non-volatile additives). Moreover, previously identified impurities by LC/MS could not be assigned in LC-UV methods as the separation in both systems was different. In this study, known impurities were traced in the LC-UV methods and new impurities present in polymyxin B and colistin bulk samples were characterized. To achieve this, each peak from the non-volatile system was collected separately and reinjected into an LC system with a volatile mobile phase coupled to MS. This way, collected impurity peaks were rechromatographed on a reversed phase column in order to separate the analyte from the buffer salts. Using this method, out of 39 peaks, five novel related substances were characterized in a polymyxin B bulk sample. Fourteen impurities, which were already reported in the literature were traced as good as possible in the LC-UV method. In the case of colistin, a total of 36 peaks were investigated, among which four new compounds. Additionally, 30 known impurities were traced in the LC-UV method.

Combination of a liquid chromatography-ultraviolet method with a non-volatile eluent, peak trapping and a liquid chromatography-mass spectrometry method with a volatile eluent to characterise erythromycin related substances.

The selectivity and sensitivity obtained with volatile liquid chromatographic (LC) methods are often inferior compared to non-volatile ones. However, the buffers often used in the non-volatile system are incompatible to mass spectrometry (MS). So, the characterisation of unknown peaks in a non-volatile system, based on data obtained from a volatile LC-MS method, is problematic. In this study, the unknown peaks in a non-volatile liquid chromatography coupled with ultraviolet detection (LC-UV) system were directly characterised by a volatile LC-MS system using a peak trapping technique. Each peak eluted from the non-volatile system was trapped by a switching valve and sent to a LC-MS system using a volatile mobile phase. Mass spectral data were acquired on an LCQ ion trap mass spectrometer equipped with electrospray ionisation (ESI) operated in the positive ion mode. Using this technique, the fragmentation behaviour of erythromycin and its related substances was studied and the components occurring in commercial samples were investigated. In total 25 compounds mentioned in the literature were traced. Fourteen more unknown impurities were also studied.