Direct sample preparation and simultaneous perfluoroacylation - Trimethylsilylation of biogenic monoamines along with their acidic metabolites for a single step analysis by GC-MS

Abstract

The GC-MS quantification of biogenic monoamines (BMAs), together with their acidic metabolites (ACMEs), in a single step, is presented here for the first time. This novel principle is based on the exceptional reactivity of the hexamethyldisilazane (HMDS) and perfluorocarboxylic acid (PFCA) couples [1,2], resulting in the simultaneous trimethylsilylation and acylation of BMAs and ACMEs.For this basic study, tyramine (TYR), 3-methoxytyramine (3-MeTYR), dopamine (DA), epinephrine (EP), normetanephrine (NORMNE), norepinephrine (NOREP), tryptamine (T), 3,4-dihydroxyphenylalanine (L-DOPA), 5-methoxytryptamine (5-MeT), serotonin (ST), and their ACMEs, such as homovanillic acid (HVA), vanillylmandelic acid (VMA), and 5-hydroxyindoleacetic acid (5-HIAA) were selected. These three ACMEs were derived from 3-MeTYR, NORMNE and ST, respectively.The mass fragmentation properties of the fully derivatized products proved to be of stoichiometric distribution. Informative high masses were obtained: such as the molecular ions [M]+= and/or their [M-CH3]+ alternatives. The exceptions were EP and NOREP which decomposed to the same specific, abundant mass of m/z 355 representing the C7H3-tri-OTMS ions formed by the loss of their nitrogen-containing moieties. The general rule of this new principle was confirmed by using trifluoroacetic acid (TFA), pentafluoropropionic acid (PFPA), or heptafluorobutyric acid (HFBA) with HMDS in parallel tests. In all three cases, derivatives of close retention properties in a stoichiometric manner were obtained. On the basis of the optimum separation characteristics between the BMA-ACME pairs, the HMDS & PFPA couple was preferred as the reagent of choice. Method validation was carried out, both with model solutions and in the presence of the urine matrices (without any preliminary extraction). Analytical performance characteristics for the model solutions like repeatability (RSD% 3.88–6.4), linearity (R2 0.991–0.999) and limit of quantitation (LOQ 8.8–103 ng/mL) were determined. Analytical performance characteristics for urine matrices were calculated by using the standard addition method applying the urine of a healthy volunteer and also analyzing urines of patients diagnosed with neurological diseases.