Agronomic advantages conferred by endophyte infection of perennial ryegrass (Lolium perenne L.) and tall fescue (Festuca arundinacea Schreb.) in Australia

Abstract

Perennial ryegrass and tall fescue are key grasses of sown pastures in the high-rainfall zone of south-eastern Australia. Ryegrass in naturalised pastures, and in sown seed, is widely infected with Neotyphodium fungal endophytes, with toxic endophyte strains occasionally causing toxicosis in livestock. Endophyte infection is also beneficial in sown grasslands, assisting ryegrass hosts to overcome biotic stresses, and tall fescue hosts to overcome biotic and abiotic stresses. We review the literature for Australia and present new data, to examine the agronomic effects of endophyte. Frequency of endophyte infection in old, perennial ryegrass pastures and ecotype-based cultivars is high and, in all pastures, increases with time, providing evidence for endophyte-infected plants having an agronomic advantage over endophyte-free plants. Within a cultivar, agronomic field experiments have compared endophyte-infected with endophyte-free swards. Endophyte significantly improved ryegrass establishment in seven of 19 measurements taken from 12 trials. In mature ryegrass pastures, over half of the experiments found advantages to endophyte infection. Tall fescues infected with a selected endophyte (‘AR542’) had improved agronomic performance relative to endophyte-free in a majority of experiments, and on occasions, the endophyte was essential for tall fescue persistence. Cultivar × endophyte interactions occurred but were inconsistent. In high-stress environments, endophyte was more important for agronomic performance than difference between cultivars. The relative importance of cultivar and endophyte is discussed, with elite cultivars that are adapted to the region and are infected with elite endophytes being the best avenue to capture the benefits and minimise detrimental endophyte effects on livestock. The major drivers are likely to be insect pests and drought, but evidence is limited.