Ascorbate (ascorbic acid, vitamin C) is essential for both plants and mammals. Ascorbate is a reducing agent capable of donating electrons, enabling it to perform a range of biochemical functions, such as scavenging reactive oxygen species, assisting enzymatic activity, and reducing higher oxidative states of iron (Fe). In plants, ascorbate is the most abundant water-soluble antioxidant and plays a key role in many fundamental processes, such as photosynthesis, stress tolerance, and the transport of Fe. In humans, ascorbate is an essential micronutrient that must be obtained through diet and takes part in a range of important physiological functions, such as collagen synthesis, epigenetic programming, and Fe uptake in human digestion. Several pathways towards ascorbate biosynthesis have been proposed in plants, but there is only definitive evidence for the L-galactose pathway. The GDP-L-galactose phosphorylase (GGP or vtc2/5) gene encodes the first-committed and rate-limiting enzymatic step of the L-galactose pathway and represents the most promising candidate for increasing ascorbate biosynthesis in plants. In addition to transcriptional regulation, the translation of GGP is regulated through a highly conserved, cis-acting upstream open reading frame (uORF) in the 5’ leader sequence of the GGP mRNA. Developing strategies to increase ascorbate biosynthesis in rice (Oryza sativa L.) and wheat (Triticum aestivum L.), two of the world’s most important staple crops, has the potential to improve both food security and crop productivity. As part of this PhD project, two distinct metabolic engineering strategies were employed to increase ascorbate concentrations in rice: (i) constitutive overexpression of the OsGGP coding sequence (35S-OsGGP plants), and (ii) CRISPR/Cas9-targeted mutagenesis of the OsGGP uORF (uorfOsGGP mutants). Ascorbate levels were negligible in both 35S-OsGGP and uorfOsGGP brown rice, likely due to the decline of ascorbate levels in maturing grain reported in cereals; highlighting the challenge of increasing ascorbate levels in cereal species, such as rice. Ascorbate concentrations were significantly increased in germinated brown rice and tissues of 35S-OsGGP plants at the vegetative growth phase. In contrast, foliar ascorbate concentrations were significantly reduced in 35S-OsGGP plants at the reproductive growth phase. This was dependent on homozygosity of the 35S-OsGGP transgene and was associated with a significant reduction in endogenous OsGGP transcript levels, suggesting the occurrence of gene silencing. Foliar ascorbate concentrations were significantly increased in uorfOsGGP mutants, without any changes to OsGGP transcript levels, attributed to alleviated ribosomal stalling on the OsGGP uORF and enhanced translation of the GGP major ORF. Editing the GGP uORF represents an effective transgene-free strategy to increase ascorbate concentrations not only in rice, but other species. Challenging convention, automated imaging revealed that neither the 35S-OsGGP nor the uorfOsGGP plants displayed increased salt tolerance at the vegetative growth phase, despite having elevated ascorbate levels. Ascorbate concentrations were positively correlated with ferritin concentrations in Caco-2 cells—an accurate predictor of Fe uptake in human digestion—exposed to in vitro digests of null segregant and 35S-OsGGP brown rice and germinated brown rice, suggesting that ascorbate-enriched crops may be able to improve Fe bioavailability in human diets. Grain Fe concentrations were not changed in the uorfOsGGP mutants, indicating that ascorbate may not facilitate the transport of Fe into developing rice grain. Next, this PhD project identified six TaGGP genes in the hexaploid bread wheat genome, each with a highly conserved uORF in the 5’ leader sequence. Phylogenetic analyses demonstrated that the TaGGP genes separate into two distinct groups and identified a duplication event of the GGP gene in the ancestor of the Brachypodium/Triticeae lineage. A microsynteny analysis revealed that the TaGGP1 and TaGGP2 subchromosomal regions have no shared synteny suggesting that TaGGP2 may have been duplicated via a transposable element. A transcript analysis of the TaGGP genes identified that the TaGGP1 homoeologs were broadly expressed across different tissues and developmental stages and that the TaGGP2 homoeologs were highly expressed in anthers. Finally, transient transformation of the TaGGP coding sequences in Nicotiana benthamiana significantly increased foliar ascorbate concentrations more than five-fold, confirming their activity toward ascorbate biosynthesis in planta. The six TaGGP genes and uORFs identified in this study present an opportunity to fine-tune ascorbate biosynthesis in this important staple crop.

In natural communities, species may evolve characteristics in response to a variety of complex interactions involving heterospecific or conspecific members. Symbioses between arthropod predators, traditionally considered to be predatory and thus antagonistic, provide intriguing opportunities to explore this topic. Recent studies of interspecific interactions among certain spider symbiotic systems reveal outcomes ranging from araneophagy to mutualism. The latter is apparently unusual among any predators, and may reflect the unappreciated role of visual and chemical signalling as prey-luring traits in symbioses. In this thesis, I used a tractable symbiosis allowing long-term trials, which comprises web-building host spiders whose mobility is low and their associates whose behaviour is easily observed, to develop an understanding of the evolutionary significance of symbioses among arthropod predators. First, as orb-weavers are generally the focal species in these symbiotic systems, I investigated the deceptive signal and trait evolution of these web-building spiders. By performing field experiments conducted under contrasting light conditions with paper model spiders, I examined the visual signal design for prey luring. The results show that both the colour (yellow) and pattern (yellow and black mosaic) are essential for luring prey in a high ambient light environment. Subsequently, using a phylogenetic comparative approach, I investigated the association between prey viewing environment and the ventral signal of orb-weavers among 63 species and found that yellow mosaic patterns are more prevalent for web-building spiders living in high ambient light intensity. Combined, my data indicate that high contrast yellow mosaic colour patterns are highly effective in luring prey for orb-weavers occurring under high ambient light conditions, suggesting that prey colour preferences may be important in the evolution of visually mediated prey-luring systems. Second, to highlight the direct and indirect ecological effects in multispecies arthropod symbioses, I conducted field experiments that manipulated the presence of web-building and non-web-building guests to test the synergistic and antagonistic effects of these inquilines on their hosts across a broad latitudinal range. I discovered that Cyrtophora hosts may promote their foraging success by forming heterospecific aggregations with web-building Leucauge guests. Specifically, a Cyrtophora host web complex intercepted more prey when Leucauge guest webs were attached, and thus these hosts experienced higher weight gain. However, non-web building Argyrodes guests imposed fitness cost on their hosts, but only when Leucauge guests were excluded. This pattern was consistent among different host-guest pairs and over an extensive spatial distribution, in which the negative interaction between Cyrtophora host and Argyrodes guest was mitigated by the mutualism between web-building host and its guest – Leucauge orb-weaver. This suggests that the outcome of interspecific interactions between arthropod predators may be more associated with biotic factors – the presence of a third partner species, rather than an abiotic factor – environmental gradients. Lastly, I investigated the underlying mechanism facilitating the establishment of symbioses between arthropod predators. Using a series of field experiments performed on guest species that form distinct interspecific interactions with their host, I evaluated if the settlement pattern of Argyrodes guest spiders is affected by the nature of their relationship with, and the cues from the host and conspecifics. The field surveys confirmed that more guests settled onto the webs with a host and fewer conspecifics, and that the effect of conspecifics was stronger for kleptoparasites than mutualists. This suggests that intraspecific competition may be stronger in kleptoparasitism. Using behavioural assays with y-maze olfactometers, I revealed the role of volatile odours in intra- and interspecific communication. Both A. kumadai from Japan and A. fissifrons from Taiwan preferred host related odours, but klepoparasitic A. kumadai appeared to be repelled by the odours of conspecifics, and the mutualistic A. fissfrons was indifferent to the odours. The differences in cue discrimination ability correspond with the density of sensilla on the front legs of these guest species. These results highlight how the nature of species interactions can act as a selection pressure on not only the strength of cues for partnership recognition but on odorant perception traits.

Theoretical and empirical studies demonstrate that female moths vary their investment into pheromone-releasing behaviour according to both biotic factors (including age, mating status and competitive signalling) and abiotic factors (including temperature, host plant quality, and photoperiod). This suggests that female moths are capable of sophisticated strategic adjustment of their pheromone production, yet the impact of variation in pheromone output on male arrival rates and male preferences in moths are relatively unexplored. Furthermore, studies documenting sexual chemical communication in moths have focused primarily on the chemical nature of the signal, and have largely ignored female signalling strategies or chemical receiving structures (antennae), despite the costs of signalling and of maintaining signal receiving structures. In Chapter 1, I provide a synthetic review of the results of my research on the gumleaf skeletonizer moth, Uraba lugens, in the context of the broader literature, and argue that while females strategically adjust their sex pheromone signalling behaviour and hence their attractiveness, males strategically balance their investment between longevity and antennal morphology, which reflects their mate searching capacity. In the context of chemical communication in moths, these females calling (pheromone-releasing) strategies reduce the likelihood of mating failure and allow males to maximise their encounter rates with females. In Chapter 2, I explored how adult age influences the calling behaviour of virgin female U. lugens over four continuous ten-hour scotophases (dark periods). I found that female U. lugens alter their calling behaviour with age, but in contrast with theoretical predictions and empirical observations in other species, older females were less likely to call and spent less time calling than younger females. Older females, however, commenced calling earlier in the scotophase, suggesting a strategic shift, potentially to avoid competition from younger females. Behavioural assays with y-maze olfactometers showed that males prefer the pheromones produced by younger females, and that pheromone quality likely plays a role in this choice. In Chapter 3, I explored how juvenile population density influences pheromone output in female U. lugens. I found that female U. lugens facultatively adjust their calling behaviour in response to socio-sexual cues: females that eclosed from high juvenile population densities started calling earlier and spent more time calling than individuals eclosed from low juvenile population densities. Juvenile density also affected female pheromonal attractiveness: males prefer the pheromones produced by females reared at high juvenile densities. Females are likely to benefit from this strategic investment: increased investment into chemical signalling (at high densities) suggests that females compete with conspecific neighbouring signallers in order to avoid mating failure. In Chapter 4, I explore how juvenile diet influences reproductive investment of both female (the quality of female sex pheromone) and male (pre- and post-copulatory) adults. I found that the effect of juvenile diet on adult fitness depended upon adult sex: in females, diet influenced body size, while in males diet influenced longevity. Juvenile diet also affected female pheromonal attractiveness: males tended to prefer the pheromones produced by females reared on host plants which has been supplemented with a fertiliser. Finally, host plant species affected male pre-copulatory investment: males reared on Eucalyptus camaldulensis have longer antennae but less dense sensilla than when reared on a different Eucalypt species, although there was no difference in the testes size of males reared on the two different species. In Chapter 5, I explore how upregulation of immunity affects male antennal functional morphology, female pheromone quality, and other life-history traits. I found that immune activation affected male, but not female signalling investment: immune challenged males had a lower density of sensilla on their antennae, but female pheromonal attractiveness was not affected by their immune status. Nevertheless, immune activation reduced female investment into ovary mass, and the longevity of adult males and females increased following an immune challenge. In conclusion, my different experiments consistently reveal that females alter their sex pheromone production and releasing behaviour for mate attraction in order to avoid mating failure, while males balance the resources allocated to longevity and chemoreception, which improves the likelihood of locating mates.

Plants synthesise a vast range of secondary metabolites that are stored in specialised cells or organs. The presence of sub-dermal glands rich in volatile terpene essential oils is characteristic of the trees of the genus Eucalyptus (Myrtaceae). Recent studies showed that non-volatile compounds (NVCs), particularly monoterpene acid glucose esters (MAGEs), co-occur with volatile components in Eucalyptus foliar oil glands. The principal aim of this thesis is to characterise MAGEs and other non-volatiles localised to Eucalyptus oil glands and to explore their relationships to the co-housed oil components. Glandular extracts from a range of Eucalyptus species belonging to the two major subgenera, Symphyomyrtus and Eucalyptus, were investigated in Chapter 2. Non-volatiles were extracted from enzymatically isolated glands and analysed using high-performance liquid chromatography (HPLC) and mass spectrometry (LC-MS). MAGEs were identified based largely on diagnostic mass spectral fragmentation patterns. MAGEs were the dominant NVC in Symphyomyrtus species. In contrast, the sampled subgenus Eucalyptus species lacked MAGEs, but were rich in phenolics. Volatile oil components were also analysed using gas chromatography with flame ionisation detection and mass spectrometry (GC-FID and GC-MS). Monoterpenes and sesquiterpenes were identified and quantified for each species. In addition, cell suspensions of E. polybractea were successfully established from leaf-derived callus as a potential tool to investigate the biosynthesis of MAGEs. Glandular extracts from subgenus Eucalyptus species rich in non-volatiles containing phenolic moieties were further analysed in Chapter 3. A suite of unsubstituted B-ring flavanones was identified as the dominant glandular NVCs. In addition, flavones, flavanone-O-glucosides, flavanone-b-triketone conjugates, triketone heterodimers and chromone-C-glucosides were also identified. Flavanones were quantified and species-specific variations were observed. This chapter also showed that flavanones are exclusively localised to the glands rather present throughout leaf tissues. A positive correlation was observed between some monoterpenes and sesquiterpenes with total flavanones and particularly with pinostrobin. Interestingly, b-triketones were also found in the volatile extracts of glands from E. suberea and E. brevistylis. The presence of glandular b-triketones was further explored in Chapter 4. A major discovery of this thesis was the occurrence of two gland types in E. brevistylis that differ in their colour and importantly, in their metabolic contents. ‘Sesquiterpene glands’, which are translucent-white in appearance, contained sesquiterpene alcohols. ‘Triketone glands’, which are golden-brown in appearance, contained mostly the b-triketone conglomerone and sesquiterpene hydrocarbon caryophyllenes in lower abundance. None of the glands contained the NVCs identified from the other species in this thesis. The results were consistent in trees from a natural population of E. brevistylis and in glasshouse-grown seedlings and saplings. In addition, ‘Triketone glands’ seem to develop earlier than ‘sesquiterpene glands’ in leaf ontogeny. This is the first identification of such metabolic differentiation of embedded glands from any tree species. The work presented in this thesis revealed many novel aspects of oil glands, particularly in relation to their non-volatile and volatile constituents and how they are related to one other. Overall, the findings of this thesis contribute significantly to the knowledge on Eucalyptus foliar oil gland chemistry and biology.

Human iron (Fe) and zinc (Zn) deficiencies are among the most prevalent nutritional disorders in the world and manifest as a range of health issues including fatigue, impaired cognitive development and increased mortality. Micronutrient supplements and fortificants are frequently used to increase human Fe and Zn intakes yet these strategies require continuous investment and frequently miss rural populations; improving the density and/or bioavailability of Fe and Zn in staple crops (a process known as biofortification) represents a powerful alternative. Bread wheat (Triticum aestivum L.) is cultivated on more land than any other crop and is processed into a range of food products to supply ~20% of the daily calories consumed by humans. The wheat grain is mostly comprised of starch and protein, with micronutrients such as Fe and Zn concentrated in the outer aleurone layer of the grain. In the aleurone layer, Fe and Zn are complexed to compounds that inhibit their absorption (bioavailability) in the human gut and the aleurone layer is removed during grain milling to produce white flour. White flour (representing the inner wheat endosperm) contains low concentrations of dietary Fe and Zn, and high consumption of either wholemeal flour or white wheat flour coincides with high prevalences of human Fe and Zn deficiencies. Generating Fe and Zn biofortified wheat through conventional breeding in modern wheat cultivars is hindered by inherently low grain Fe and Zn concentrations and a lack of genetic variation for these traits. Nicotianamine (NA) is a low-molecular weight metal chelator present in all higher plants with high affinity for Fe2+, Zn2+, and other divalent metal cations. In graminaceous plant species such as wheat, NA serves as the biosynthetic precursor to 2’-deoxymugineic acid (DMA), a root-secreted mugineic acid family phytosiderophore that chelates ferric iron (Fe3+) in the rhizosphere for subsequent uptake by the plant. Both NA and/or DMA are the major chelators of Fe within white wheat flour, and NA is known to enhance Fe bioavailability in cereal grain. For these reasons, increasing the biosynthesis of NA/DMA through upregulation of nicotianamine synthase (NAS) genes has emerged as a popular strategy for Fe and Zn biofortification of cereal crops. In this study we employed constitutive expression (CE) of the rice (Oryza sativa L.) nicotianamine synthase 2 (OsNAS2) gene in bread wheat to upregulate the biosynthesis of NA and DMA, and evaluate plant growth, grain nutrition and food processing properties of CE-OsNAS2 wheat. Our lead CE-OsNAS2 wheat transgenic event (CE-1) demonstrated higher concentrations of Fe, Zn, NA and DMA in wholemeal flour, white flour and white bread, altered distribution of Fe in the grain, and higher white flour Fe bioavailability relative to null segregant (NS) control. Protein composition, dough rheology and breadmaking properties were similar between CE-1 and NS white flour, and a chicken (Gallus gallus) feeding study over a period of six weeks demonstrated that chickens consuming CE-1 white flour had improved Fe status, intestinal morphology and microbial populations relative to chickens that consumed NS white flour. Multi-location confined field trial (CFT) evaluation over three field seasons demonstrated no differences between CE-1 and NS agronomic performance apart from plant height. Throughout all CFTs, grain yield was negatively correlated with grain Fe, Zn, and protein concentrations yet not correlated with grain NA and DMA concentrations. White flour Fe bioavailability was positively correlated with white flour NA concentrations, and we determined NA to be the strongest enhancer of in vitro Fe bioavailability identified to date. Together these results suggest the proportion of Fe that is chelated to enhancers of bioavailability (such as NA and DMA) should be prioritized in future crop biofortification efforts and highlight new strategies for developing Fe and Zn biofortified wheat as a more nutritious staple food.

The standard correlative species distribution model accounts for the effect of the environment on species distributions without explicitly accounting for the effect of species interactions. Community ecology, in contrast, studies species co-occurrence patterns without accounting for the co-occurrence that could be explained by the shared response of species to the environment. Joint species distribution models (JSDMs) are a relatively recent development in the ecological literature that extend the single species distribution model framework to model multiple species simultaneously while accounting for species co-occurrence, and are an exciting prospect for bridging the gap between these two disciplines. Research into JSDMs is still in its early stages as a field and we have identified six outstanding problems in the literature that are potentially hampering the wider uptake of JSDMs by the wider ecological community. The confluence of statistical and computational advancements has led to the rapid development of JSDMs, and the proposal of several models and implementations. These models are statistically complex but, while well defined as stand-alone studies, are described using a diverse array of model notation, terminology, and symbols. This often conflicting notation makes comparing the different JSDM implementations a daunting task to most ecologists who are trying to identify which model is best suited to their purpose. In this thesis I develop and present a clearly-defined, singular notation for the description of JSDMs and use it to elucidate the similarities and differences between seven different JSDM implementations. All of the newly proposed JSDM implementations have been presented on completely different datasets to each other and thus any measure of inferential performance is not directly comparable. Without a direct comparison potential practitioners are unable to make an informed decision about the performance of these models. In this thesis I conduct a direct comparison of seven JSDM implementations on six datasets to assess their inferential performance. I found that all JSDMs identified similar species-environment relationships to each other. All JSDMs identified species co-occurrence patterns in the same direction but with different strengths and uncertainties. The newly proposed JSDMs have been implemented using a variety of software with different default or suggested parameters for model fitting. In addition to being fit to datasets of a varying size it is difficult to directly compare the computational performance of different implementations. Potential practitioners are going to be influenced by factors such as how fast a model can run and the size of dataset it is capable of feasible fitting to. In this thesis I perform a direct comparison of seven JSDMs on six datasets to directly assess their computational performance. I found a greater difference in computational performance between JSDMs than for inferential performance. I found evidence that scaling issues due to the size of datasets, and thus the number of parameters to estimate, was more prevalent for some implementations than others. A comparison of effective sample size in their respective Markov chain Monte Carlo regimes also indicated that some implementations are far more computationally efficient than others. The use of JSDMs for prediction has only recently begun to be addressed in the literature, but has almost exclusively focussed on environment-only prediction types akin to those of single species models. The multivariate nature of JSDMs, however, allows for new, community-level predictions to be used. To date none of these prediction types have a formal definition for JSDMs. In this thesis I present four types of prediction that can be applied with JSDMs. Marginal predictions are environment-only predictions that do not account for the residual associations between species. Joint predictions are a community-level prediction that can return either a probability of observing a particular assemblage or generate a series of plausible assemblages at a site. These two prediction types can then be conditioned upon the known occurrence states of one or more species at a site, and thus present conditional marginal and conditional prediction types respectively. Single-species models have been subject to substantial research into the evaluation of their prediction, but this has seen minimal to date for JSDMs. In this thesis I present five classes of evaluation metrics that assess different aspects of species distributions and the community assemblage process. Threshold-independent and threshold-dependent metrics operate at the species-level, community dissimilarity and species richness metrics operate at the community-level, and likelihood-based metrics which assess model fit. The literature on JSDM prediction has to date been limited in its comparisons of predictive performance. Studies have either compared a limited number of models, a subset of evaluation metrics, and/or generally only considered environment-only prediction types. In this thesis I present a comparison of predictive performance for six JSDMs and two stacked single-species models. I fit the models to 22 real and simulated datasets, predict with the four prediction types I have defined for JSDMs, and evaluate these predictions with up to 32 metrics from five classes. I found that likelihood-based metrics suggested the JSDMs were better fit to the data but that the other metric classes generally showed them to underperform compared to the standard stacked single species model for all prediction types. The stacked single species model with the SESAM constraint was found to consistently outperform all other models. This suggests an interesting avenue of future research by applying this constraint to JSDM predictions.

The strategy of conserving threatened species away from mainland threats on offshore ‘island arks’ is particularly common in Australia and New Zealand but has been used worldwide to conserve a broad range of taxa. This strategy, however, has consequences for both the species being conserved and the recipient community. The effects of invasive species can be complex, and the assisted colonisation of species to islands—a species invasion of sorts—could cause a myriad of unintended and unpredictable consequences, for both the introductee, and its recipient community. In this thesis, I set out to examine the consequences of ‘island arks’ as a conservation measure for both the endangered species being conserved and the recipient island community into which it is introduced. To do this, I exploit the introductions of an endangered marsupial predator, the northern quoll (Dasyurus hallucatus), to a number of ‘island arks’ in the Northern Territory, Australia. Northern quolls have suffered dramatic recent declines throughout northern Australia due to the invasion of a toxic invasive cane toads (Rhinella marina). In 2003, to prevent their extinction in the Northern Territory, insurance populations of quolls were established on two toad- and predator-free islands (Astell and Pobassoo) off North East Arnhem Land. In 2017, we established a new ‘island ark’ population of toad-smart quolls on Indian Island, near Darwin. These ‘island ark’ populations provided fertile ground to explore the consequences of island isolation on quolls (Astell Island) and also the impacts quolls have as a novel predator on an ‘island ark’ (Indian Island). In my first data chapter (Chapter 2), I document an unsuccessful reintroduction of northern quolls from Astell Island to southern Kakadu National Park on the Australian mainland. Although quolls that were trained to be ‘toad-smart’ survived longer than quolls that received no training, the reintroduction ultimately failed because of predation by dingoes (a native top-order predator). The aim of this study, and of my entire PhD at the time, was to determine whether toad-aversion training could instil multi-generational recovery from toads in northern quolls. The failure of this reintroduction due to predators, rather than cane toads, motivated a change in perspective and direction. I began to question whether the conservation of quolls on a predator-free ‘island ark’ may have resulted in unintended maladaptation for mainland reintroduction. Indeed, a follow up study (Chapter 3) determined, that after only 13 generations isolated on this ‘island ark’, quolls did not appear to recognise the scent of dingoes and cats; predators with which quolls had coevolved alongside on the Australian mainland for at least 3500 and 130 years, respectively. Unlike their isolated island conspecifics, wild-caught and captive-born mainland quolls (from populations with continuous exposure to predators) both recognised and avoided the scent of cats and dingoes. I also found (Chapter 4) that using instrumental conditioning to re-establish antipredator behaviours in ‘island ark’ conserved quolls was not straightforward: training protocols that were logistically and ethically feasible did not work, suggesting that the protocols were inadequate or that it is difficult to train individuals that have totally lost predator recognition. A conservation introduction of quolls to an additional island (Indian Island) then provided a unique opportunity for a detailed investigation of the direct and indirect impacts of an introduced novel predator on a recipient ‘island ark’ community. In Chapter 5, I first establish some basic properties of the system: that behaviour of the prey species is both repeatable, and not correlated with their detection rate. This is important given that individuals are assumed to have homogenous detection probabilities in most population estimation models. And in Chapter 6, I document a trophic cascade driven by quolls that elicits shifts in both population size and behaviour of naive prey (grassland melomys). After the arrival of quolls on Indian Island, the survival and abundance of grassland melomys (Melomys burtoni) rapidly declined, and behavioural traits of populations invaded by quolls changed. Although quoll-invaded melomys populations remained consistently more neophobic than quoll-free populations, boldness in quoll-invaded melomys converged with that of quoll-free populations through time. Quoll-exposed melomys populations exhibited lower per-capita seed take rates, and rapidly developed an avoidance of seeds associated with quoll scent, with discrimination playing out over a spatial scale of tens of metres. My study of this ‘island ark’ provides a rare insight into the rapid ecological and behavioural shifts enacted by prey to mitigate the impacts of a novel predator and shows that trophic cascades can be strongly influenced by behavioural as well as numerical responses. Globally, we face a rapidly accelerating extinction crisis. In many cases in situ conservation is difficult, if not impossible. Isolationist conservation measures, where threatened species are removed from the threatening process—captive populations, conservation fences, and ‘island arks’—are often necessary to halt the extinction of the most critically endangered species. We must, however, undertake these actions with a long-term plan, and with awareness of potential consequences. Central to such considerations is the phenotype: we need to move beyond population size as a sole monitoring criterion. The value of more quolls, for example, is dramatically undermined if increased population size is associated with a shift towards maladaptive phenotypes. Similarly, the trophic cascade that results from the introduction of a new predator is not driven solely by the predator’s demographic impact, but also by the behavioural shifts in prey resulting from that impact. Phenotypes, particularly behavioural phenotypes, can change rapidly in response to conservation actions. We would be wise to use that knowledge, or risk unanticipated consequences.

While understanding insecticide resistance in Drosophila melanogaster is informative for controlling pest insects that threaten agricultural yields and vector deadly diseases, it also serves as a powerful model of microevolution which can be interrogated with an exceptionally powerful genetic toolkit. The Drosophila Genetic Reference Panel (DGRP) provides the opportunity to study population-genetic signatures of natural selection in individuals that can be repeatedly measured for a range of phenotypes. In this work, genomic and transcriptomic data from the DGRP are compared with phenotypes from nine insecticidal compounds against the background of genome-wide signals of selection. The two most prominent signatures of selection in the population are attributable to insecticides from a single, widely-used chemical class, the organophosphates. Evidence suggests that insecticide-based selection is limited to these two loci, however the genetic bases of insecticide phenotypes appear to be complex. Insecticide-associated variation includes both structural effects through amino acid substitution and chimeric gene formation, and regulatory effects on transcript abundance by cis- and trans-acting factors. Resistance mechanisms exhibiting pleiotropic effects on insecticides from different chemical classes is found to be rare; one such case is correlated with constitutive, modular regulation of oxidative stress-related transcripts, the genetic basis of which is mapped to multiple trans-acting factors. Comparisons of the results from the DGRP with diverse population genomics data suggests that the outcomes of these analyses are applicable to populations of D. melanogaster worldwide.

Leaderless secretory proteins (LSPs) are proteins that are secreted yet lack the classical canonical signal peptide sequences and, therefore, by definition are undergoing unconventional protein secretion (UPS). We cannot necessarily assume all such proteins found in plant secretomes are LSPs due to the high possibility of cellular contamination arising from cellular disruption, leading to a mix of LSPs and contaminant proteins that are both characterised by the absence of signal peptides. The aims of this Thesis are to use computational biology methods to identify LSPs. In Chapter 1 the current knowledge of secretory pathways (LSPs and UPS) in plants, motifs for secretion, the difficulties of isolating the plant secretome (cell wall/apoplastic fluids) without contamination, and the lack of appropriate bioinformatics tools for plants to distinguish them from LSPs is reviewed. Chapter 2 evaluates a commonly used computational tool, SecretomeP which was trained on mammalian data and is the most widely used prediction tool for LSPs, including in plant studies. Exploring whether this tool is applicable to plants required using conventionally secreted proteins as a proxy, and evaluations were made on SecretomePs premise that conventional and unconventional secretory proteins will share properties. By removing the signal peptide from sequence data, the research shows a bias in scores due to the signal peptide, and only a marginally higher true positive rate compared to false positives. The use of the tool on further plant studies was not recommended and suggested previous inferences of plant LSP status based solely on SecretomeP predictions needed to be re-evaluated. This work was published as ”Better Than Nothing? Limitations of the Prediction Tool SecretomeP in the Search for Leaderless Secretory Proteins (LSPs) in Plants” (Lonsdale et al., 2016) Chapter 3 details the creation of a putative LSP database for Arabidopsis thaliana by taking the entire proteome and applying a workflow of collating annotation, literature observation and experimental relationships between proteins. This results in a framework to identify proteins without signal peptides that have been observed (unclassified set) or unobserved (non-secreted set) and compare them to conventionally secreted proteins that have been observed (secreted set) or unobserved (theoretically secreted sets). Protein-protein interactions, GO term and PFAM distributions that are similar between the secreted and unclassified protein sets are used to create confidence lists of putative LSPs. In Chapter 4 new prediction tools were created. Firstly, a SecretomeP-like tool is created using the properties of the observed secreted set, with signal peptides removed. Lessons from Chapter 2 on SecretomeP allowed the bias to be minimized. Secondly, tools were created based on the candidate list from the putative LSP database. Each tool is based on a Random Forest (RF) using protein-derived features and trained on subsets of the LSP database. Consensus predictions between them on new data was used to identify further LSP candidates. Additional reporting tools provide a convenient way to map data from the LSPDB onto new sequences, and the results of these tools on LSPs from plants and other organisms shown. In Chapter 5 additional putative candidates for LSPs in Arabidopsis and other species were identified by applying the tools and databases to additional data excluded from the original database created in Chapter 3, including recent cell wall proteomes, secretory pathway experiments and analysis of extracellular vesicles. In conclusion, Chapter 6 discusses the limitations of the computational methods developed, and suggestions for improvement to predict LSPs in plants for further experimental investigation and confirmation of location.

Over fourty species of invertebrate pests can be pests of emerging crop seedlings in south-eastern Australia. Some species are a common and widespread issue for farmers, while others only sporadically cause damage. The inability to predict potential pest problems is a key reason why insecticides are commonly applied to fields prior to sowing (applied as seed treatments or bare-earth sprays) and/or post-sowing (foliar sprays) in a prophylactic manner. This thesis investigates factors which predict the risk of pest problems arising from two common pests and two emerging sporadic pests of crop seedlings. The common pests are the earth mites, the redlegged earth mite, Halotydeus destructor and the blue oat mite, Penthaleus destructor. The emerging pests are the Portuguese millipede, Ommatoiulus moreleti and the common pillbug, Armadillidium vulgare. These species are a curious problem as they generally feed on decaying plant matter and have only recently been recognised as emerging pests. Sampling was conducted to assess the capacity to predict the abundance of the damaging autumn generation of the major pest mites, H. destructor and P. major, at a field level. Across the fields sampled, abundances of both mites showed limited association with a range of common agronomic and environmental field variables of known biological importance. The abundance of H. destructor could however be divided in to risk categories when fields were grouped in to categories of field type. These categories were valid across two years and two farming regions. The abundance of P. major, by contrast showed a regional association. The first pest step in understanding the pest status of O. moreleti and A. vulgare was to understand their ability to damage common crop seedlings. Under shadehouse conditions, this study found O. moreleti has a limited ability to feed on a range of crop seedlings. A. vulgare can damage a wide range of crops, although some crops are only susceptible in the first days of seedling establishment. This finding was supported by pest reports from fields across southern Australia over the past decade. Lupinus angustinus was the most susceptible crop seedlings for O. moreleti. This appears to be due to the removal of deterrent chemicals in commercial breeding as well the thickness or strength of leaf tissue. The fitness benefits of feeding on seedlings appear to vary between crop seedlings. In controlled studies, the level damage was found to be influenced by the lifestage and sex of the individual, while the presence of crop stubbles as alternate food sources were not found to influence the level of damage in controlled studies. To understand the role of water stress on the pest status of A. vulgare, I conducted controlled trials on the influence of dry environmental conditions on the level of feeding damage A. vulgare caused to canola. I hypothesised water stress would result in individuals increasing their feeding on plants, as they searched for water. Conversely, I found individuals kept in environments with low ambient humidity caused less damage. Field monitoring observations were made over the course of the thesis to provide the basis for further research on the population dynamics of O. moreleti. While stubble is known to be important in influencing population sizes, I also found evidence that soil-type has an important effect on O. moreleti numbers. Interestingly, abundances were found to vary greatly between years in some crop fields. This appears to be due to seasonal population dynamics, with O. moreleti leaving fields and, in some cases, not returning. The research conducted in this thesis points to ways to guide monitoring and sustainable management practices. Farm managers should use the association of damaging abundances of H. destructor with field-type to predict pest risk at a farm level. The research on A. vulgare and O. moreleti helps to explain why these species are a sporadic pest risk, given they are limited in their ability to damage crop seedlings. This knowledge will assist in the development of pest management programs for O. moreleti and A. vulgare. The studies on the association of environmental and agronomic factors should be used to guide future research to understand the many remaining knowledge gaps.

Nutrient excess derived from anthropic impacts on coastal ecosystems can have harmful consequences at the economic and ecosystem levels. Traditionally, companies in charge of wastewater treatment in coastal cities have implemented infrastructure improvements to try and reduce eutrophic conditions that arise due to an imbalance between their nutrient inputs and the subsequent consumption within the coastline ecosystems. In this thesis I explored the potential of macroalgae as bioremediation agents as an alternative to engineering solutions. Chapter 2 is focused on the selection of candidates for bioremediation before the wastewater discharge reaches the coastline. Chapter 3 explores the evaluation of macroalgae species and biomass amount present at zones with nutrient excess as an essential step to assess the possibilities that seaweed cultivation can entail at an economic level. Chapters 4 and 5 delve into the consequences that the drift algae cultivation has for the associated fauna and nutrient cycles. In Chapter 2, I investigated the competitiveness and production properties of three species of filamentous macroalgae easily found in temperate environments. The results found competitiveness to be the preferable feature when selecting a candidate species for freshwater bioremediation. In addition, I suggested two different bioremediation systems. The first option is the use of monocultures of two different species, one selected for warm seasons (Oedogonium) and another for cold seasons (Stigeoclonium). The second option is the establishment of bi-cultures of both species with varying dominances depending on the light and temperature conditions of temperate environments. In Chapter 3, I estimated the biomass and seaweed macroalgae composition at three locations in a semi-enclosed Bay in southern Australia. The results showed a 13-times decrease in drift macroalgae density compared to what was recorded 20 years ago, in 1996. However, the biomass retained the red algae dominance and remained abundant, especially during conditions that favoured the appearance of blooms. Isotope analyses showed that these macroalgae fed on anthropogenic nitrogen in areas close to nutrient sources, making them a preferable alternative for nutrient removal through harvest. This chapter also explored the potential co-benefits of the encountered seaweed. Chapter 4 showed how seaweed drifting mats gave shelter to a large number of invertebrates and ichthyofauna; however no individual was noxious/invasive or belonged to a special protection category. Drift macroalgae was not a key habitat for the fauna, since all individuals were abundant in sandy or reef habitats in the area. Finally, in Chapter 5 I studied how the drift mats affected the process of denitrification in the bay. The results obtained from benthic chamber experiments showed how perennial mats located in areas close to the nutrient loadings gave a buffering effect that helped maintain high denitrification rates under unfavourable seasonal periods (staying up to 70% higher in winter compared to controls). Overall, my findings provide a novel approach to select freshwater macroalgae candidates for bioremediation in temperate environments and give information regarding the variable effects on associated organisms and nutrient cycles derived from marine drift algae harvest. The findings of my thesis will be widely applicable to decision-making processes involving bioremediation mediated by fresh- and saltwater macroalgae.

Fungal pathogens are ubiquitous and are equally proficient in infecting both plants and animals raising the constant concern for public health as well as the economic impacts on agricultural production. Among the huge cohort of human fungal pathogens currently present, one is Talaromyces marneffei, an opportunistic fungus that mainly infects immunocompromised people. This pathogen is unique from others in its order as it is the only species that has the capacity to switch between two different cellular morphologies (known as dimorphism). At 25oC, it grows in a saprophytic multinucleate hyphal form which under favourable conditions undergoes asexual development to produce conidia, through specialised structures called conidiophores. These conidia are the infectious agents of this pathogen. Upon inhalation of these conidia by the host, these are phagocytosed by alveolar macrophages resulting in a temperature shift to 37oC. This shift in temperature induces the dimorphic switch and the conidia germinate into a uninucleate yeast form which is the pathogenic state of this fungus. The aim of this study is to investigate how T. marneffei survives and replicates inside the hostile environment of the host after being phagocytosed and to determine the importance of cell wall in this process. For this purpose, we developed a phagocytic model to mimic the early events in the host-pathogen interaction while analysing the host response to the invading pathogen and the counterattack mechanisms adopted by the pathogen to survive within the hostile environment. This study showcased the ultimate fate of T. marneffei inside the host cell, focussing on important events like phagocytosis, onset of germination, neutralization of phagolysosomes and replication inside the macrophages. This process of T. marneffei survival inside host also highlighted the importance of cell wall for a successful infection as difference in the cell wall architecture led to modulation of host responses against it. The overall cell wall architecture of T. marneffei in different cellular morphologies was determined with focus on beta-1,3;1,4 glucans (Mix linkage beta glucan) and alpha-(1,3) glucan with their respective role in pathogenicity. Given the importance of cell wall in survival of the pathogen inside the host, deletion mutants were also generated for genes involved in the biosynthesis of these cell wall components. This study revealed how disruption in the cell wall layering pattern rendered them susceptible against the applied stress signifying the importance of cell wall in survival and as a potential antifungal drug target. Overcompensation by upregulation of other cell wall genes on loss of a particular cell wall component also displayed the importance of intact cell wall for growth and replication. Overall this study has revealed how T. marneffei survives, germinates and replicates inside the host macrophages while highlighting the significance of the cell wall in its growth and pathogenicity with focus on two essential cell wall components beta-1,3;1,4 glucans (Mix linkage beta glucan) and alpha-(1,3) glucan.