Near infra red (NIR) detection of sorgham ergot alkaloids in grain

Summary

Sorghum ergot is a fungal disease first identified in Australia in 1996, with the potential to affect a large proportion of Australia's sorghum crop. At first sorghum ergot was thought to be relatively harmless to livestock, but this situation changed when cases of reduced milk production in sows and dairy cows were caused by sorghum containing up to 30% ergot and 50 mg/kg ergot alkaloids (Blaney et al. 2000). Experiments conducted by DPI with funding support from PRDC, GRDC and MLA (Blaney 1999, McLennan 2000a,b) have shown that growth of Hereford steers in feedlots is inhibited with alkaloid concentrations of 1.5 mg/kg. This is roughly equivalent to the current limit for sorghum ergot in stockfood regulations of 0.3%.

As investigations proceeded, it became clear that ergot and alkaloid contents are not well correlated. This means that a sample containing the regulated limit of 0.3% ergot most commonly would contain about 1 mg/kg, but in principal might contain anything from 0.3 to 24 mg alkaloids/ kg. Looked at in reverse, the tolerated limit of 1 mg alkaloid/kg for feedlot cattle might conceivably be present in samples containing as little as 0.01% ergot! If climatic conditions increase ergot contamination in future seasons, bulk grain marketers and end-users such as feedlotters will need to detect alkaloids rather than ergot, to detect loads with higher toxicity (>1 mg/kg). One technique that might be fast enough is a physical test like Near Infra-red Spectroscopy (NIR). The quantification of ergovaline (an ergot alkaloid very closely related to dihydroergosine) in tall fescue pasture by NIR was reported by Roberts et al (1997). They reported several wavelengths from 2300 to 2400nm where absorbance was correlated with ergovaline concentrations ranging from 0.07 - 0.94 mg/kg.

The project reported investigated the possibility of measuring dihydroerosine in sorghum by similar means. The specific objective was to evaluate the possibility of using NIR to detect sorghum ergot alkaloid in sorghum grain at levels of 0.1 mg/kg and greater. If this was successful, there were plans to move into a second stage of assessing the test for use with whole grain and then to see how effective grading processes were for removing ergot from sorghum. a preliminary evaluation of the use of NIR to measure ergot alkaloid was made. Using a NIR Systems 5500 Spectrophotometer with a spinning sample cup, the maximum absorption wavelength of the pure sorghum ergot alkaloid (dihydroergosine) was determined to be 2262 nm (D. Law and B. Blaney, 2000, unpublished data). Three sorghum samples containing about 0, 10 and 40 mg/kg alkaloid were also assayed and showed differences at 2262 nm as well as other wavelengths, which was quite promising. About 34 samples of sorghum containing a range of ergot alkaloid contents were acquired from various parts of southern and central Queensland. Some had been associated with problems in piggeries and dairies, while others were from ergot-infection sorghum experimental plots. The resultant samples were analysed for ergot alkaloids in duplicate or triplicate.

The method involved a triple extraction into ethyl acetate: methylene chloride: ammonia: methanol (25:50:1:3) facilitated by an ultrasonic bath. Solvent was removed and the alkaloids were semi-purified by partition between diethyl ether and 0.5N HCl, and back-extraction into methylene chloride after rendering basic again with ammonia. After dissolution in methanol and filtration, samples were injected into a high performance liquid chromatograph using an RP-18 column and a mobile phase of 0.1% aqueous ammonium acetate: methanol: acetonitrile (50:20:30). Detection was by both ultraviolet and fluorescence. Ergot alkaloid results are based on the main alkaloid present, dihydroergosine, which accounts for about 80% of the total alkaloid content. The minor alkaloids are festuclavine and dihydroelymoclavine (subject to mass-spectral confirmation), and appear in constant proportions relative to dihydroerosine in all samples.