Development and validation of an ultrahigh performance liquid chromatography-high resolution tandem mass spectrometry assay for nine toxic alkaloids from endophyte-infected pasture grasses in horse serum.

Endophyte fungi (e.g. Epichloë ssp. and Neotyphodium ssp.) in symbiosis with pasture grasses (e.g. Festuca arundinacaea and Lolium perenne) can produce toxic alkaloids, which are suspected to be involved in equine diseases such as fescue toxicosis, ryegrass staggers, and equine fescue oedema. The aim of this study was, therefore, to develop and validate a quantification method for these and related alkaloids: ergocristine, ergocryptine, ergotamine, ergovaline, lolitrem B, lysergic acid, N-acetylloline, N-formylloline, peramine, and paxilline in horse serum. Horse serum samples (1.5mL) were worked up by solid-phase extraction (OASIS HLB). The extracts were analyzed by ultra high performance liquid chromatography-high resolution tandem mass spectrometry (UHPLC-HRMS/MS). Chromatographic separation was achieved by gradient elution with ammonium formate buffer and acetonitrile on a RP18 column (100×2.1mm; 1.7μm). HRMS/MS detection was performed on a QExactive Focus instrument with heated positive electrospray ionization and operated in the parallel reaction monitoring (PRM) mode. Method validation included evaluation of selectivity, matrix effect, recovery, linearity, limit of quantification (LOQ), limit of detection (LOD), accuracy, and stability. With exception of lolitrem B solid phase extraction yielded high recoveries (73.6-104.6%) for all analytes. Chromatographic separation of all analytes was achieved with a run time of 25min. HRMS/MS allowed sensitive detection with LODs ranging from 0.05 to 0.5ng/mL and LOQs from 0.1 to 1.0ng/mL. Selectivity experiments showed no interferences from matrix or IS, but N-acetylloline and N-formylloline were found to be ubiquitous in horse serum. Newborn calf serum was therefore used as surrogate matrix for the validation study. Calibration ranges were analyte-dependent and in total covered concentrations from 0.1 to 50ng/mL. Lolitrem B and paxilline could be sensitively detected, but did not meet quantification requirements. For the other analytes, accuracy and precision were shown for 3 different concentrations (QC low, medium, high) with acceptable bias (-10, 5%-7.9%) and precision (CV 2.6%-12.5%). Matrix effects varied from 55.0% to 121% (RSD 7.8-18.5%) and were adequately compensated by IS. Matrix effects of N-acetylloline and N-formylloline could only be estimated in newborn calf serum (n=1) and ranged from 52.5-88.3%. All analytes were stable under autosampler conditions and over 3 freeze and thaw cycles. Applicability was proven by analyzing authentic horse serum samples (n=24). In conclusion, the presented method allows a sensitive detection of ergocrisitine, ergocryptine, ergotamine, ergovaline, lolitrem B, lysergic acid, N-acetylloline, N-formylloline, peramine, and paxilline in horse serum and reliable quantification of all but lolitrem B and paxilline.