PRGT: Identification of metabolites associated with severe cases of perennial ryegrass toxicosis

Summary

Over two seasons, perennial ryegrass (cultivar Samson, naturally infected with the wild-type endophyte, Neotyphodium lolii) was sampled on 5 occasions during November to May at 4 farms in Victoria and at Lincoln, New Zealand. Endophyte frequency in the populations was 77-100%.  Additionally, pasture on 20 farms with stock experiencing perennial ryegrass toxicosis (PRGT), was sampled; these pastures had endophyte infection frequencies of 95-100%. The 2010-11 season in Victoria was most atypical; pasture stayed green through summer. 
The following season was more typical. Analysis consistently showed high concentrations of lolitrem B at Lincoln which were significantly greater, by 2-3 fold, than those observed in Victorian samples of iso-genetic material and of the material associated with PRGT. PRGT samples commonly had concentrations of lolitrem B that were above tolerance levels in 2011 but in 2012, ergovaline was the dominant toxin; its concentration commonly exceeded tolerance levels. 
To explain the severe PRGT that occasionally occurs in Victoria and Tasmania (viz. heavy losses of livestock including many due to sheep and cattle crowding into dams etc.), we investigated the possible importance of other endophyte metabolites - that appeared on HPLC chromatograms of Australian PRG samples from 2002 but not observed for toxic PRG samples taken in NZ - by collecting quality samples (the 2002 samples had been stored at ambient temperature for 6 years) and thereby ruling out compounds produced by advantageous fungi/mould/bacteria growing on stored samples. 
The presence of these other unidentified peaks was confirmed on HPLC chromatograms but their expression was not significantly different between NZ and Australian grown PRG and did not correlate with any of the known indole diterpene compounds/precursors detected by LC-MS/MS analysis.  No unidentified peaks were observed in the erogvaline HPLC chromatograms for any samples on this occasion. Mass spectrometry (LC-MS/MS) was carried out to determine indole diterpenes present in the lolitrem B biosynthesis pathway and for the ergot alkaloids present in the ergovaline pathway.  The values for lolitrem B determined by LC-MS/MS correlated well with those obtained using HPLC. Significantly higher expression was observed in the New Zealand relative to Victorian samples of PRG for the indole diterpenes involved in the biosynthesis of lolitrem B, viz. paspaline, terpendole C, lolitrem E, lolitrem B and lolitrem F.  
For the ergot alkaloids, significant differences were not apparent between Vic. and NZ samples in year 1. In year 2, LC-MS/MS results showed ergovaline concentrations were greater in Vic. samples. The presence of ergotamine, ergocryptine and ergocornine (Claviceps-purpurea produced ergot alkaloids) was detected at 6/27.  These ergot alkaloids should not be overlooked as contributing to toxicosis in stock grazing PRG.