The forage tree legume leucaena [Leucaena leucocephala (Lam.) de Wit ssp. glabrata (Rose) Zarate] is a high-quality ruminant feed vitally important as a source of protein for livestock production in tropical regions. However, the presence of the acutely toxic non-protein amino acid mimosine in leaves and seeds, and its breakdown to chronically toxic hydroxypyridones (DHP), was believed to limit its productivity and adversely affect animal health. The Australian development of a fermenter-cultured oral inoculum containing the DHP-degrading ruminal bacterium Synergistes jonesii in the 1980s was thought to overcome this issue; however, anecdotal evidence of leucaena toxicity symptoms occurring in inoculated animals remained. In response, a program of research was initiated to investigate the efficacy of S. jonesii as well as other methods of protection against toxicity in ruminants. Broadly, this involved two areas of research, namely the effectiveness of: (a) microbiological degradation via S. jonesii; and (b) non-microbial metabolic detoxification. Key objectives were to: (a) determine the efficacy of inoculation with S. jonesii both within Indonesia and Australia; (b) profile the toxicity status of ruminants in eastern Indonesia; (c) investigate the impact of mineral supplementation on 2,3-DHP toxicity; and (d) determine the extent of in vivo conjugation of DHP and its role in protecting ruminants against toxicity.Initially, an Indonesian study involving the transfer of rumen fluid from “protected” ruminants to naïve Bali bulls was conducted. Sequential monitoring of DHP levels in urine indicated that there was no effect of inoculation. Secondly, a controlled animal house experiment was conducted in Australia to measure the impact of high leucaena diets on production in naïve steers and the effect of inoculation with the commercial S. jonesii inoculum. Key findings were: (1) inoculation had no effect on total DHP excretion; (2) indigenous S. jonesii strains were present in naïve animals prior to inoculation; (3) DHP did not suppress thyroid hormone production; and (4) high levels of the isomer 2,3-DHP were present without accompanying signs of toxicity.The leucaena toxicity status of ruminants in eastern Indonesia was then assessed in a survey across the four islands of Lombok, Sumbawa, Sumba and West Timor. A number of different strains of S. jonesii, including the ATCC type strain (78.1), were detected, however S. jonesii was always at low population levels and always accompanied by high levels of undegraded 2,3-DHP in urine. Despite this apparent failure of S. jonesii to degrade DHP, there was no observed impact on animal productivity or health. Accordingly, to investigate non-microbial detoxification mechanisms, a feeding trial involving goats in Indonesia was conducted to investigate the effect of Iron(II) sulphate mineral supplementation as a possible pathway for reducing the toxicity of 2,3-DHP by chemically binding to the toxin. Key findings were: (1) despite high levels of 2,3-DHP ingested by goats, no clinical signs of toxicity presented; (2) mineral supplementation did not result in an improvement in animal production; (3) evidence of conjugation of 2,3-DHP was observed; and (4) goats consuming diets of 100% leucaena were highly productive.Finally, the role and mechanism of conjugation was investigated. This involved a longitudinal assessment of the toxicity status and productivity of Indonesian Bali bulls on 100% leucaena diets. High liveweight gains were recorded even though the bulls were persistently excreting high levels of 2,3-DHP. Using HPLC and HDMS it was found that up to 97% of DHP was excreted as 2,3-DHP in conjugated form. The conjugate was identified as glucuronic acid, forming a 2,3-DHP-O-glucuronide. It was concluded that conjugation played the key role in preventing DHP toxicity, as conjugation of DHP not only increases the speed of clearance of toxins from the body, but also actively detoxifies the compound. As the primary mode of toxicity of DHP is as a strong ligand (chelating with essential metals, resulting in deficiencies), the attachment of glucuronic acid inhibits this effect. Further, as metal binding occurs via the vicinal (O,O) moiety, the isomerisation of 3,4-DHP (from mimosine) to 2,3-DHP results in a weaker ligand and is therefore inherently less toxic. While the isomerisation pathway is not fully known, the detection of mono-hydroxypyridines (HP), previously unidentified, using HPLC and HDMS, could indicate possible intermediates.The findings of this research greatly advance the knowledge of how ruminants adapt to DHP in diet, with the innate mechanism of conjugation identified as the primary mode of detoxification.It is concluded that microbial detoxification is not the main pathway for toxicity control of ruminants consuming leucaena, and that inoculation with S. jonesii may no longer necessary. However, appropriate management of ruminants being introduced to leucaena diets is necessary to allow the animals to adapt and build metabolic tolerance.