A complex pharmaceutical raw material was characterized by means of reversed-phase gradient elution. By varying gradient steepness and mobile-phase pH, it was possible to optimize band spacing so as to separate 16 impurities or degradation products from the drug substance. Computer simulation was useful in interpreting these complex chromatograms and determining the maximum number of peaks that could be separated in this way. A marginal separation of all 17 sample components could be obtained, but the resulting method was quite pH-sensitive and therefore not very rugged. As an alternative, a rugged method was developed that separates the drug substance from all other sample components. The present study also describes how present computer simulation software for isocratic separation can be used to predict resolution for gradient elution runs as a function of pH.

Isocratic optimization of coumarin, potassium sorbate, ascorbic acid, sodium benzoate, vanillin, ethylvanillin, methylparaben, ethylparaben, sodium saccharin

Computer simulation by DryLab G/plus has proved to be an invaluable tool in the rapid development of analytical HPLC methods for antibiotics. A glycopeptide antibiotic under study at this Research Center was taken into consideration as a case study. Retention data from two preliminary experiments have allowed us to perform several simulations which greatly shortened the time normally required for the identification of the optimum gradient conditions. The “key steps” in the simulation process have been experimentally verified and a more than satisfactory agreement between calculated and experimental retention times was consistently found.

A complex mixture of arachidonic acid metabolites was separated by reversed-phase HPLC using a multi-step gradient, which was modelled by computer-assisted HPLC method development. The metabolites were extracted on-line on a precolumn connected to the analytical column in the same HPLC system. The predictions of the resolution and also the retention times calculated by computer simulation were very accurate when compared with the corresponding experimental run (maximum deviation 0.86%). An appropriate HPLC method and additionally an on-line extraction procedure could be developed with just three experimental HPLC runs. This method could be useful for evaluating the concentrations of arachidonic acid metabolites involved in inflammatory diseases.

Peptide and protein samples are often complex mixtures that contain a number of individual compounds. The initial HPLC separation of such samples typically results in the poor resolution of one or more band pairs. Various means have been suggested tor varying separation selectivity so as to minimize this problem. In this study of a tryptic digest of recombinant human growth hormone, the simultaneous variation of temperature and gradient steepness was found to be a convenient and effective means of varying selectivity and optimizing the separation. The use of computer simulation greatly facilitated this investigation.

Changes in band spacing as a function of temperature and/or gradient steepness were investigated for four peptide or protein samples. Reversed-phase HPLC in a gradient mode was used to separate tryptic digests of tissue plasminogen activator and calmodulin. Additionally, a synthetic peptide mixture and a storage protein sample from wheat were studied. Simultaneous changes in gradient steepness and temperature were found to provide considerable control over band spacing and sample resolution.

The effects of temperature and gradient steepness on selectivity in these systems appear to be complementary. Simultaneous optimization of both temperature and gradient steepness thus represents a powerful and convenient means of controlling band spacing and separation. Because of the complexity of these sample chromatograms, computer simulation proved to be a useful tool in both interpreting these experiments and in optimizing final separations.