Evaluation of the impurity profile of alcuronium by means of capillary electrophoresis.

Alcuronium, a neuromuscular blocking drug, was recently introduced to the European Pharmacopoeia. A HPLC method is described to limit the impurities of alcuronium, namely the diallylcaracurine (DAC) and the allyl-Wieland-Gumlich-aldehyde (WCA), to less than 0.5%. Since alcuronium and all impurities are quaternary salts, capillary electrophoresis (CE) is highly suitable to evaluate the impurity profile. Using 12 mM heptakis-(2,6-di-O-methyl)-beta-cyclodextrin in a 50 mM phosphate buffer at pH 5.5 or 50 mM diethanolamine buffer (pH 9.2)-acetonitrile 19:1 containing heptakis-(2,3-O-diacetyl-6-sulfo)-beta-cyclodextrin the impurities could be baseline separated and quantified. The limit of detection for DAC and WCA was found to be in the same range as found with HPLC; thus, less than 0.1% of both DAC and WCA could be detected in the solution for injection in presence of alcuronium. In injection solutions of alcuronium which were stored at higher temperatures three additional, unidentified impurities were detected. In addition, the conversion of alcuronium to DAC, occurring under acidic condition, was monitored by means of the CE method developed.

Evaluation of amino sugar, low molecular peptide and amino acid impurities of biotechnologically produced amino acids by means of CE.

As the second part of our studies on the impurity profiles of amino acids produced by biotechnological processes, a micellar electrokinetic chromatography (MEKC) was applied to determine amino sugars, low molecular peptides and amino acids as potential expected impurities at a level 0.1% w/w. 3-(4-Carboxybenzoyl)quinoline-2-carboxaldehyde (CBQCA) was found to be suitable as a labeling reagent for laser induced fluorescence (LIF) detection. The labeling reaction was optimized to achieve maximum reaction yields and reproducibility for all groups of substances. The 'level 0.1%' was chosen as quantities/concentrations in such a way that the major reagent peak becomes comparable with the impurities at this level, and the minor reagent peaks (<0.01%) are not able to cover other peaks at the level of interest. The reaction optimization and the validation studies were performed with model mixtures of representative amino sugars, low molecular peptides and amino acids. A linearity range of labeling of three orders of magnitude was achieved. The total precision of the method, including the labeling reproducibility was studied and for all test substances relative standard deviation less than 5% were obtained. The accuracy was evaluated by performing recovery experiments at three concentrations covered a 50-200% interval of the level 0.1%. Confidence intervals below +/-2.5% (lambda=0.05, n=9) of the target were found sufficient for purity tests.