Search

Parotid secretion was obtained under standardized conditions from 17 male students by stimulation with citric acid. The average flow rate was 0.65 ml/min. The free amino acids in these samples were determined with the aid of two-dimensional thin layer microchromatography of the dansyl derivatives and TV densitometric determination of the blackening of photographic negatives. Apart from glutamic acid, only amino acids with nonpolar residues such as glycine, alanine, leucine, isoleucine, phenylalanine and proline occurred in all 17 samples. The mean values of the concentrations lay between 0.2 mumol/l for isoleucine and 1.2 mumol/l for proline. There were great interindividual variations. (SD between 40% ad 110%). In the present range of 0.3 to 1.1 ml/min, no dependence of the concentrations on the flow rate could be demonstrated. It was attempted to draw up a standard profile by elimination of greatly deviating samples.

In order to obtain a better reproducibility in quantitative determination of amino acids in amounts of picomoles an internal standard was used. A mixture of ten amino acids was diluted and dansylated. 0.1 microliter of each solution was separated by two-dimensional thin-layer chromatography on polyamide sheets (3 x 3 cm). Each chromatogram was photographed with four different exposure times under ultraviolet light. The dark spots in the photographic negatives were measured by television densitometry. A number of factors influence the reproducibility of this method: loading volume, photographic exposure time and processing, and quenching of fluorescence. As an internal standard, which is influenced by all these factors, DANS-hydroxyproline was added to each solution. By using this standard, the variations in exposure time and fluorescence quenching can be corrected satisfactorily. The relative standard deviation of the mean value of four photographs of one chromatogram is 2%. The results on correction of the volume deviation were not so satisfying. The relative standard deviation of the mean of twelve different chromatograms can be reduced from 34% for the initially measured data, to 8% for the values corrected by the internal standard. It seems necessary to carry out three or more determinations of one solution. In this case the relative standard error of the mean is 2%.