Interactions between Escherichia coli and the New Zealand native plants Leptospermum scoparium and Kunzea robusta : A dissertation submitted in partial fulfilment of the requirements for the Degree of Doctor of Philosophy at Lincoln University

Elevated concentrations of human pathogens in waterways is a global environmental issue and can result in widespread human disease and suffering. The provenance of these pathogens is usually from biological wastes, specifically faecal material from humans and livestock. In New Zealand, Escherichia coli is used as a reference/indicator organism to indicate pathogen contamination in waterways, with the drinking water standard set at < 1 colony forming unit (cfu)/100 ml and < 260 cfu/100 ml for swimming, following weekly testing. In 2017, the media reported that over half of NZ’s rivers are unfit for swimming and in 2016, livestock contaminated drinking water with Campylobacter jejuni resulted in some 5500 people becoming ill in Havelock North, the largest recorded Campylobacter outbreak in world history. In 2016, a study showed that the NZ native plants Leptospermum scoparium (J.R. G.Forst.) and Kunzea robusta (A. Rich.) increased the decimal reduction time of E. coli in biosolids amended soil by 90% within five to eight days of bacterial inoculation. Other studies have shown that the antibacterial properties of these plants (with their essential oils and honey) extend to Candida albicans and Salmonella typhimurium. Potentially, L. scoparium and K. robusta could be planted in areas receiving human or animal wastes to intercept human pathogens before they enter waterways. However, the mode of action by which these plants increase microbial die-off is unclear. Nor is it understood how these plants affect bacterial fluxes in field conditions, where soils may receive high rates of rainfall or irrigation. This thesis aimed to determine the antimicrobial properties of the roots and leaves of L. scoparium and K. robusta and to determine whether the provenance of the plants can affect antimicrobial activity. Furthermore, the thesis sought to determine the fluxes of bacteria in soils planted with L. scoparium and K. robusta. Throughout the thesis, E. coli was used as an indicator organism for pathogens. Lolium perenne (L.) was used as a comparison plant, as this species is present in 70% of NZ pasturelands. Laboratory experiments, using nutrient broth and optical density measurements, revealed that after 24 hours, water extracts (containing 50 – 400 mg/l total organic carbon) from L. scoparium and K. robusta reduced E. coli ATCC13706 growth to just 13% - 25% of the control (no extract). In contrast, L. perenne leaves extracts significantly increased the growth of bacteria by 63% compared to the control. Root extracts of L. scoparium and K. robusta inhibited growth more strongly than the leaf extracts. There were significant differences in the inhibitory activities of plants from different provenances, indicating that either genetic or environmental conditions can affect the antimicrobial activities of these species. Glasshouse experiments using repacked 10 l lysimeters showed that under high flow conditions (14 mm/day), both L. scoparium and K. robusta exacerbated leaching of E. coli compared to L. perenne. This was attributed to increased preferential flow caused by the taproot systems of these species. Under low-flow conditions (< 8 mm/day), there was negligible bacterial leaching under all species tested. The examination of soil pore water under L. scoparium and L. perenne revealed that seven days after inoculation, E. coli numbers under L. scoparium were significantly lower (1.3 x103 cfu/ml) than under L. perenne (5.9 x 103 cfu/ml). Similar results were obtained for two soil types used: A pallic brown soil (Pawson Silt Loam) and a recent soil (Lismore stony silt loam), but with higher overall leaching from the stony silt loam. A field experiment, near Duvauchelle on the Pawson silt loam also showed decreased E. coli numbers under L. scoparium and K. robusta, nine days after the inoculation. In contrast, there was little die-off under L. perenne. However, this experiment did not reveal whether the disappearance of E. coli under L. scoparium and K. robusta was due to increased die-off or increased preferential flow that removed the E. coli from the top soil. This thesis shows that in some cases, strategic plantings of L. scoparium and K. robusta may improve the quality of NZ’s freshwater resources. The efficacy of planting these species is dependent on the provenance of the plants and the climatic conditions. Although L. scoparium and K. robusta increased leaching of E. coli under high-flow conditions, this may be beneficial in some cases, such as the protection of surface waterways where the trees may result in increased infiltration and decreased surface run off of bacteria. Future work should focus on determining the effect of plants age and rhizosphere chemistry on microbial properties and elucidating whether the regional differences between L. scoparium and K. robusta are due to the genetic or environmental differences.