Veterinary and Agricultural Sciences Collected Works - Theses
Now showing items 13-24 of 51
Discovery of natural product scaffolds with anthelmintic activity against Haemonchus contortus
The present thesis focused on discovering new candidate compounds in natural products against economically important parasitic worms of livestock to circumvent current problems associated with resistance to commercially-available drugs (anthelmintics) (cf. Chapter 1). One of the most pathogenic parasites of small ruminants, Haemonchus contortus (barber’s pole worm), was used as the screening tool. This whole-organism bioassay was used to identify active compounds/extracts that reduced motility and/or development or altered the morphology (phenotype) of H. contortus larvae in vitro. Selected compounds were purified from active extracts using a bioassay-guided fractionation approach. The screening of two compound libraries containing a total of 1,000 natural products, natural product-inspired or synthetic compounds identified 34 compounds with notable activity against H. contortus (Chapters 2 and 3). Most of the ‘hits’ (n=32) identified from the first library were analogues of arylpyrrole (a natural product scaffold), while deguelin and rotenone (two plant rotenoids) were identified from the second library. Encouraged by these findings, a library of 7,500 extracts from different plant species was screened (Chapter 4). This screen identified three active extracts from the leaves and roots of Cryptocarya novoguineensis and the roots of Piper methysticum. Bioassay-guided fractionation of active extracts yielded four known alpha-pyrones, namely goniothalamin from C. novoguineensis, and dihydrokavain, desmethoxyyangonin and yangonin (= kavalactones) from P. methysticum. Three kavalactones induced a lethal ‘evisceration’ phenotype in treated larvae of H. contortus, and had limited toxicity on mammalian epithelial (MCF10A) cells. This is the first report on the activity of such natural compounds on a parasitic nematode of animals. The work was extended to explore compounds from Australian marine species. A library of 2,000 extracts from marine invertebrates was screened (Chapter 5) and identified three active extracts from marine sponges - one from Monanchora unguiculata and two extracts from Haliclona sp. The compounds with moderate activity against H. contortus were fomiamycalin from Monanchora unguiculata and halaminol A from Haliclona sp., although the activity of these compounds was not entirely selective. Subsequently, a structure-activity relationship (SAR) investigation of one of the identified natural product scaffolds, alpha-pyrone was conducted (Chapter 6). This work identified three analogues (designated W-408, W-415 and W-417) with anthelmintic activity similar to, or greater than the synthetic parent kavalactones (desmethoxyyangonin or yangonin), with the most potent analogue W-408 achieving a marked increase in potency (7-fold) against and selectivity (> 21) for H. contortus. Taken together, this drug discovery effort identified or purified 40 natural products representing distinct chemical classes (Chapters 2-5), and the SAR investigation of the alpha-pyrone scaffold identified three key analogues with enhanced potency and/or selectivity (Chapter 6). Collectively, the findings of this thesis indicate that some of the identified natural product scaffolds have the potential to be developed as ‘lead’ candidates. Developing such anthelmintic candidates via future chemical optimisation, efficacy and safety assessments, broad spectrum activity assessments, and target identification represents an exciting prospect (cf. Chapter 7) and, if successful, could pave the way to subsequent pre-clinical and clinical evaluations.
Compressive fatigue life and micromorphology of equine metacarpal subchondral bone
Bone fatigue injuries are an important cause of lameness in racehorses and as such are an animal welfare issue and a cause of losses to the industry. The metacarpal condyles are commonly affected by stress fractures and palmar osteochondral disease, a fatigue injury of subchondral bone. Bone fatigue injuries should be preventable through training adjustments. Knowledge about the behaviour of equine bone, and particularly subchondral bone, under cyclic loading is important to inform training recommendations. Computational models have been developed to investigate the effects of training changes on stress fracture development but require equine subchondral bone material properties as an input to meaningfully model metacarpal condylar fatigue injuries. A method for compressive fatigue testing of subchondral bone was developed and the fatigue life of equine subchondral bone determined. The fatigue life curve followed a power law, similar to cortical and cancellous bone. Subchondral bone stiffness increased during fatigue testing over hundreds to thousands of cycles, which was initially attributed to a method artefact. Initial specimen stiffness was positively associated with actual density but not with horse and training related factors. Micro-CT was used to investigate the micromorphology of the palmar metacarpal subchondral bone. Micromorphology parameters were analysed with principal component analysis and mixed-effects linear regression models. The largest differences in micromorphology were observed in untrained horses between the age of 16 and 20 months. In racehorses in training, age and duration of a training period had no influence on micromorphology. Horses with palmar osteochondral disease had more pores in cross-section than those without. Tissue mineral density increased, and bone volume fraction decreased with increasing distance from the articular surface. Several alterations to the fatigue testing method were explored to accommodate the curved articular surface and eliminate the stiffness increase during compression-compression cyclic loading. The stiffness increase during the first 50 cycles was demonstrated to be reversible and no microdamage or trabecular compaction was detected in specimens loaded up to 200 cycles. Stiffness increase during compression-compression fatigue testing at a physiological frequency is thus likely the result of normal viscous material properties of subchondral bone. Finally, compressive fatigue life was positively associated with bone volume fraction in the deeper layers of subchondral bone. Initial stiffness was positively associated with tissue mineral density in the deeper layers and bone volume fraction in the superficial layer. Both maximum stiffness and cycles to maximum stiffness were positively associated with fatigue life. Cycles to 10% reduction of maximal stiffness correlated strongly with cycles to gross fracture. The results showed that subchondral bone’s viscous material properties led to a prolonged stiffness increase during cyclic compressive loading at physiological frequencies. Further research is required to investigate the physiological implications of this finding. Subchondral bone sclerosis measured as bone volume fraction is positively associated with compressive fatigue life and thus has a protective effect on subchondral bone. Further research is required to reconcile this finding with the common colocation of fatigue damage in sclerotic subchondral bone of racehorses.
A study into the genotypes of Escherichia coli from pyometra affected dogs
Pyometra is one of the most common reproductive diseases in female intact dogs. It affects nearly 25% of all intact female dogs before they reach 10 years of age. This disease not only affects the animal’s health but also its breeding value. The predominant pathogen that has been isolated from dogs with pyometra is Escherichia coli (E. coli). The research described in this thesis focused on characterising the particular strains found in dogs affected by pyometra and comparing them with those found in the rectum of control dogs. The objectives were to identify differences regarding the genotype, the presence of genes transcribing for uropathogenic virulence factors (UVF) and the bacteria’s abilities to form biofilms. The initial aim of investigating differences between bacteria isolated from young and old pyometra patients was hindered by the limited availability of E. coli strains from young dogs. To achieve these aims uterine, vaginal and rectal samples were collected from 32 pyometra patients and the patients’ age was recorded. Rectal samples were collected from 45 clinically healthy dogs to serve as controls. Samples were analysed by polymerase chain reactions for genotyping by Clermont’s scheme and evaluation of UVF gene presence. A subset of samples was further investigated to analyse their biofilm forming potentials using crystal violet assays. Finally, the ability of biofilm producing bacteria to withstand antimicrobial treatment was evaluated using assays to determine the minimum inhibitory concentration and time-kill curve studies. These investigations demonstrated that there is a significantly different UVF gene profile in E. coli isolated from pyometra patients when compared to control dogs. In contrast, there were only few differences between the uterine, vaginal and rectal samples taken from the same dogs. Similarly, little difference was detected between phylogenetic groups with the majority of all samples being classified as B2. Isolates with strong biofilm forming potential were only identified from those collected from pyometra patients but not control dogs and they were subsequently shown to be more resistant to antimicrobials. These studies have shown that there are major differences between the E. coli strains present during pyometra and those present as part of the normal microflora in the intestinal tract. It is also the first time that the biofilm forming potential was evaluated for bacteria involved in pyometra. It becomes apparent that care must be taken in the medical treatment of pyometra patients to consider the impact of biofilms on their efficacy.
Investigation into the Morphology of the Canine Sacrum and its Relationships with Selected Structures of the Vertebral Column and the Hind Limb
In dogs, the sacrum consists of three fused vertebrae (Evans & De Lahunta, 2013) and recently the population of greyhounds in Victoria, Australia have shown a remarkably high incidence of sacrocaudal fusion in which the sacrum consists of four fused vertebra (Oheida, Philip, Yen, & Davies, 2016). Scientific data and the available literature regarding how the occurrence of sacrocaudal fusion might influence the morphology of other parts of the locomotory system in dogs is not clear enough to explain why sacrocaudal fusion appears to be so prevalent in the Victorian greyhound population. This study set out to explore the hypothesis that there are measurable differences in the morphology of the sacrum (S. Weight, S. Length, and S. Width) between greyhounds with standard (3 fused vertebrae) and those with different types of fused sacra (4 fused vertebrae, Type B, C, and D). Also, it was hypothesised that there is an association between the sex and body mass of the greyhounds and morphology of the sacrum (S. Weight, S. length, and S. Width). In addition, this study included an exploration into the influence of sacrocaudal fusion on the morphology of related anatomical structures within the spine such as the L.7 vertebra, and selected bones in hind limbs such as the calcaneus, talus, and patella. The sacra were collected from 171 greyhounds and classified using two systems. For the first system, sacra were classified based on the number of fused vertebrae and the type of fusion into four types: Type A (standard sacra), Type B (complete fusion between the transverse processes and between the bodies of the S3 and Ca1 vertebrae), Type C (fusion only between the transverse processes of the S3 and Ca1 vertebrae), and Type D (fusion only between the bodies of the S3 and Ca1 vertebrae). For the second classification system, sacra were classified based on the presence, reduction, or absence of the median sacral crest between the spinous processes of the S1 and S2 vertebrae into the following three types: Type F (a complete fusion or crest), Type R (reduction in the height of the crest between successive spinous processes), and Type N (absence of the crest). The length, width, and weight of sacra, calcanei, tali and patellae were recorded and compared between the greyhounds with standard and fused sacra. Also, the length of the L.7 vertebra and the angle of the lumbosacral junction and the angle of the spinous process of the 1st sacral vertebra were compared between the greyhounds with standard and fused sacra. There were variable numbers of the population in various chapters because some bones were too difficult to measure or because of breakage of bones during processing. The exact sample numbers are indicated in each table illustrating the results. The results of the studies showed that there was an increase in the convexity of the curvature of the pelvic surface of the sacrum and an increase in the roughness of the articular surface of the sacral wing in association with the occurrence of sacrocaudal fusion. Sacrocaudal fusion was found in 71 out of 171 greyhounds (41%). Overall, the mean weight, length, and width of the fused sacra were found to be significantly greater (P < 0.01) than standard sacra. Among the fused sacra, 13.5% (23/71) were Type B, 21.1% (36/71) Type C, and 6.4% (11/71) Type D. Type B sacra (complete fusion) were the heaviest (P <0.001), broadest (P <0.001), and longest (P <0.001) sacra compared to Type A, C or D sacra. Regarding standard and different types of fused sacra, there was no association between the sex, body mass and the occurrence of sacrocaudal fusion. In addition, different scenarios were suggested for the evolution of sacrocaudal fusion in the greyhound. The first scenario was termed “Continuous Dependent Sacrocaudal Fusion”, in which it was proposed that fusion occurred first between the S3 and Ca1 vertebral bodies, followed by the occurrence of fusion (Type C), with a final result of complete fusion (Type B). The second scenario was termed “Continuous Independent Sacrocaudal Fusion”, in which the sacrum might change from Type A to either Type C or D, and then further evolve into Type B (complete fusion). The third proposed scenario was termed “Non-continuous Independent Sacrocaudal Fusion”, in which a sacrum may evolve directly from Type A (standard) to be Type B, C or D without going through the possible sequences assumed in the first or second scenarios. The reduction (R) or absence (N) of the median sacral crest between the spinous processes of the S1 and S2 vertebrae, which has been observed in association with the occurrence of sacrocaudal fusion, allowed us to suggest this new classification of sacra in greyhounds. This classification divided sacra into three different types (F, R, and N) and the percentage of each was found to be significantly different between standard and fused sacra (P <0.001) with the fused sacra more commonly showing the N type, and not associated with sex or body mass. In addition, the angle of the spinous process of the 1st sacral vertebrae of sacra with median crest type N was statistically significantly less (more upright) than those in sacra with median crest type F (P <0.042). Investigating the morphology of the L.7 vertebra as an anatomical structure articulating directly with the sacrum and lumbosacral junction, showed that the L.7 length was significantly (P ≤ .008) longer and the angle of the lumbosacral junction in greyhounds with fused sacra was significantly increased (P <0.028) in dogs with sacrocaudal fusion compared with those with standard sacra and this variation was not associated with sex (P <0.765) or body mass (P <0.802). The results showed that the right and left calcanei, tali, and patellae in greyhounds with standard and fused sacra were anatomically similar. Among all greyhounds, asymmetry between some of measurements was found, including in the width of the calcaneus (P <0.009) and the talus (P <0.025) and the length of the calcaneus (P <0.001). Measurements were higher in those bones from left hind limbs. Studying specifically the greyhounds with standard sacra also showed asymmetry in the length of the calcaneus (P <0.008) and again those calcanei from left hind limbs (L.C.L) were longer than those from right hind limbs (R.C.L). For greyhounds with fused sacra, asymmetry was found in the width of the talus (P <0.024) and again those of left hind limbs were wider than those from right hind limbs. Regarding the influence of sacrocaudal fusion on bones, there were no significant differences in the mean of measurements in bones between greyhounds with standard and those with fused sacra except for the mass of the right (M.R.C, P <0.003) and left (M.L.C, P <0.006) calcaneus. Also, the measurements of these limb bones from greyhounds with fused sacra were heavier than those with standard sacra. This thesis provides quantitative data about the occurrence of sacrocaudal fusion and is the first study to investigate the morphological anatomy of the two types of sacrum: standard or fused, as well as selected bones in the vertebral column and hind limbs of greyhounds, such as the L.7 vertebra, calcaneus, talus, and patella. In summary, sacrocaudal fusion in greyhounds was associated with a change in the morphology of the sacrum, which included an increased angle of the spinous process in the 1st sacral vertebra, a reduction or absence of the median sacral crest between the spinous processes of the S1 and S2 vertebrae, and changes in related structures, such as an increase in the length of the L.7 vertebra and angle of the lumbosacral junction. Furthermore, it seems that sacrocaudal fusion has a type of influence on certain hind limbs bones such as the mass of the right and left calcanei. These variations in the anatomical features of the sacrum between dogs with or without sacrocaudal fusion may influence the effect of forces applied to the sacral region and also the incidence of injuries. In conclusion understanding the morphological variations of different types of sacra, in addition to the data provided by this thesis, is establishing a solid base for further research, which might help to explain why sacrocaudal fusion is so common in the Victorian greyhound racing population. Overall, this work provides an anatomical foundation for understanding the function of the sacrum in greyhounds and highlights how investigation into the sacral anatomy and related structures of the greyhound body may further the current understanding of dog locomotion.
Comparative analysis of the faecal virome of dogs with various inflammatory intestinal diseases
Little is known about the population of viruses present in the intestine of dogs, the virome, despite advances in molecular biology techniques. The viral flora in the intestine of dogs has been mostly studied in disease settings, targeted to known pathogenic viruses. There is currently a gap in our knowledge of the faecal canine virome present in healthy conditions. This study aimed to identify and characterise the virome present in faeces of healthy dogs compared to the virome present in dogs with acute diarrhoea and dogs with chronic enteropathy. Faecal samples were collected from 8 healthy dogs, 8 dogs with acute diarrhoea and 8 dogs with chronic enteropathy. A viral enrichment protocol, using a mixture of endonucleases and bacterial filtration was used to enrich viral DNA and RNA. The total DNA and RNA isolated from the stool samples were amplified using a sequence-independent single-primer amplification (SISPA) protocol and subsequently sequenced by next-generation sequencing using the Illumina MiSeq platform at the Australian Genome Research Facility. Two bioinformatic pipelines were used to analyse the viral population present in each sample. After selecting high-quality reads (HQRs) and removing dog and bacterial sequences, sequence information was compared against CAMERA viral reference database in one pipeline and against the NCBI non-redundant nucleotide database in the other. Faecal samples from all groups showed a large occurrence of bacteriophages, from different families mainly under the order Caudovirales. Eukaryotic viruses were also identified. Consistent results, obtained from both bioinformatic analysis pipelines, were found for only 8 viral eukaryotic families. Interestingly, sequences with high similarity to viruses within the Astroviridae and Caliciviridae families were identified only in samples from the dogs with acute diarrhoea. Sequences similar to those of Picornaviridae were identified in one dog with acute diarrhoea and in one dog with chronic enteropathy. Only one group of sequences similar to a known virus family with known pathogenic members, Reoviridae, was identified in all groups (healthy and diarrhoeic samples). In conclusion, the largest proportion of viruses identified belonged to bacteriophages. The eukaryotic virome detected in this study contained viral sequences from a range of different virus families, including some with known enteric pathogens. These results were complemented with the complete genome characterisation of a canine kobuvirus, a canine astrovirus and a canine rotavirus. Furthermore, a prevalence analysis of these viruses, in a wider population, was performed. This project gives the first description of the canine faecal virome of healthy dogs and dogs with chronic enteropathy, thus contributing new knowledge that complements what was already known about the faecal virome of dogs with acute diarrhoea. Results from this study indicate that metagenomic analyses are useful for the investigation of viral populations in the faeces of dogs with various inflammatory intestinal diseases. Studies to further elucidate the epidemiological and biological relevance of these findings are warranted.
Studies on gastrointestinal nematodes of alpacas in Australia
The Australian alpaca industry has grown remarkably since the re-introduced of alpaca into the country in the 1980s. This relatively new and small industry faces various challenges that affect the overall production (fibre, meat etc.) of animals. For example, production losses and deaths associated with gastrointestinal nematodes (GINs) have become one of the major concerns for alpaca producers in Australia and elsewhere as there is a scarcity of reliable information about the epidemiology of GINs in alpacas, worm control practices used by alpaca farmers and the efficacy status of widely used unregistered anthelmintics. The present thesis aimed to (i) assess worm control practices used by Australian alpaca farmers, (ii) determine the epidemiology GINs and (iii) the efficacy of commonly used anthelmintics against GINs of alpacas in Australia. A comprehensive literature review along with seven results chapters and a general discussion chapter are included in this thesis. In Chapter 2, an online questionnaire survey was conducted to assess worm control practices used by Australian alpaca farmers. Results showed that more than half of the respondents perceived that GINs was an important health problem. Macrocyclic lactones (MLs) were the most commonly used anthelmintics in alpacas. Almost half of the respondents (47%) used anthelmintics at the dose rate recommended for sheep. The majority of alpaca farmers were unaware of pasture spelling and one-third of respondents had shared paddocks with ruminants. This study led to subsequent studies to understand the epidemiology of GINs of Australian alpacas using sensitive diagnostic methods and assess the efficacy of commonly used anthelmintics against GINs of Australian alpacas. In chapters 3 and 4, sensitive diagnostic methods were developed to determine the faecal eggs counts and identify nematode genus/species in the faeces of alpacas. In chapter 3, a newly developed diagnostic method, the FECPAKG2, was compared to a routinely used method, the McMaster technique, for the counting of GIN eggs in the faeces of alpacas. Data revealed moderate to good agreement between the two methods. This was the first study to assess the agreement of measurements between two methods for estimating nematode eggs in the faeces of alpacas. In Chapter 4, a molecular diagnostic tool based on multiplexed-tandem PCR (MT-PCR) was developed. This tool is faster and is capable of identifying common nematode genera/species of alpacas, including Camelostrongylus mentulatus which was not possible using traditional larval culture method. The epidemiology of GINs of alpacas in Australia was assessed using cross-sectional (Chapter 5) and longitudinal studies (Chapter 6). A range of GINs are prevalent in Australian alpacas with variable worm burdens in different climatic zones and seasons. The results of both studies were comparable. Both studies showed an overall prevalence of GINs in SACs from 61 – 66%. Weaners had the highest prevalence in both studies ranging from 73 - 80%. However, the pattern of prevalence was not same across the climatic zones. In cross-sectional study, the highest prevalence of GINs (77%) were observed in the summer rainfall zone, whereas in longitudinal study the winter rainfall zone had the highest prevalence (68%). In addition, a mixed-effects zero-inflated negative binomial (ZINB) regression model has been used to design parasite control interventions. To assess worm burden and the spectrum of GINs infecting Australian alpacas, 100 gastrointestinal tracts of alpacas were examined (Chapter 7). Results revealed a mean burden of 1,300 worms, with the highest burden of 29,000 worms. Nineteen different species of GINs were identified from Australian alpacas, including three camelid specific nematodes: Graphinema auchenia, Camelostrongylus mentulatus and Trichuris tenuis. Haemonchus contortus was the most prevalent nematode followed by C. mentulatus and Trichostrongylus spp. In Chapter 8, the efficacy of commonly used anthelmintics against GINs of alpacas was assessed. The faecal egg count reduction tests were conducted on 20 alpaca farms across the country. The results showed that a commercially available combination of levamisole, closantel, albendazole and abamectin was the most effective dewormer followed by single anthelmintic such as, monepantel, moxidectin, closantel, fenbendazole and ivermectin. Haemonchus spp. were the most commonly resistant nematodes. This was the first comprehensive study to investigate the efficacy of commonly used anthelmintics against GINs of alpacas. This study provides significant information on GINs of Australian alpacas. Results of this study advance our knowledge on the epidemiology and control of GINs and efficacy status of most commonly used anthelmintics which could be used to develop control strategies against GINs of alpacas in Australia.
The effect of diet on hormone levels in horses and ponies and in vitro effects of insulin on lamellar tissue
Equine laminitis is a multifactorial condition leading to the rotation of the distal phalanx. It is widely accepted that endocrinopathic is the most common form of the condition and that Equine Metabolic Syndrome (EMS) is a common predisposing factor associated with laminitis, of which, the pony is often more susceptible than the horse. Recent studies have shown that hyperinsulinaemia plays a significant role in the development of endocrinopathic laminitis, however the direct causal link between hyperinsulinaemia and laminitis is still not fully understood. The aim of the studies presented in this thesis aimed to further investigate this link and to determine what other factors may contribute to hyperinsulinaemia in the horse. The in vitro studies presented in this thesis have shown that as the concentration of insulin increases, so does the rate of cellular proliferation of the equine lamellar epithelial cells. It was shown that whilst there are no insulin receptors present at the site of the pathological lesions associated with laminitis, the insulin-like growth factor-1 (IGF-1) receptor (IGF-1R) is present. And due to the structural similarities between IGF-1 and insulin, insulin can act on the IGF-1R causing similar cellular cascades mediated through the ERK 1/2 pathway, namely proliferation. This activity was confirmed through western blotting techniques and also through blocking of the IGF-1R which led to a decrease in the proliferation of lamellar epithelial cells. Circulating IGF-1 levels were also measured in three different equine breeds, two of which have been shown in the past to be predisposed to the development of EMS, the pony and the Andalusian horse, as well as the particularly insulin sensitive Standardbred horse. The levels of IGF-1 were measured during periods of obesity and weight and whilst it was shown that ponies had higher IGF-1 concentrations than Standardbreds, there was no significant difference in IGF-1 concentration between the Andalusian and Standardbred. This suggests that not all breeds of horse predisposed to endocrinopathic laminitis will have increased IGF-1 concentrations and thus, circulating IGF-1 may not play a significant role in the development of endocrinopathic laminitis. Finally, the studies presented in this thesis have shown that the enteroinsular axis may be an important mechanism to consider in the development of hyperinsulinaemia. The studies presented in this thesis show that GIP did not correlate strongly with insulin release, however previous studies by our group found that GLP-1 did. Further to these findings, the studies of this thesis, that localise the incretin releasing L and K cells in the intestinal tract of horses, suggest that incretin release and thus, insulin release may be heightened in response to NSCs digestion, the fermentation of fructans and/or volatile fatty acid increases.
Molecular epidemiological investigations of Enterocytozoon bieneusi
Enterocytozoon bieneusi is a microsporidian microorganism that causes intestinal disease in animals including humans. Although E. bieneusi has been discovered and investigated in numerous countries around the world, at the commencement of this PhD project, there was no record of E. bieneusi in animals in Australia and no information on the prevalence and genetic make-up of E. bieneusi populations in this country. Thus, the present project explored E. bieneusi of a number of species of wild and domestic animals as well as humans in parts of Australia using a molecular approach, in order to improve our knowledge of the epidemiology of this microsporidian. Nested-PCR-based sequencing of the internal transcribed spacer (ITS) and/or small subunit (SSU) of nuclear ribosomal DNA was used for the detection of E. bieneusi DNA in faecal samples and the genotypic characterisation of this microbe. In the first instance, three species of wild deer (Cervus elaphus, Dama dama and Rusa unicolor; n = 610) and three species of wild marsupials (Macropus giganteus, Vombatus ursinus and Wallabia bicolor; n = 1365) in Melbourne's water catchment areas were investigated (Chapters 2 and 3). Here, E. bieneusi was detected in 4.1% of sambar deer and 1.4% of marsupials. Phylogenetic analyses of sequence data showed that genotypes D, J, MWC_d1, MWC_d2 and Type IV (in sambar deer) and MWC_m1 and NCF2 (in marsupials) clustered with E. bieneusi genotypes recorded previously in humans, indicating the zoonotic potential of these genotypes. Studies of cattle on farms located near Tarago reservoir (n = 471; Chapter 4) in Victoria, and alpacas in the states of New South Wales, Queensland, South Australia, Victoria and Western Australia (n = 81; Chapter 5) revealed relatively high prevalences of E. bieneusi in both cattle (10.4%) and alpacas (9.9%). The phylogenetic analysis of ITS sequence data revealed six genotypes (BEB4, I and J, TAR_fc1, TAR_fc2 and TAR_fc3) in cattle and three (ALP1, APL3 and P) in alpacas, all recognised to have zoonotic potential, indicating that these herbivores might act as reservoirs for human infection. Subsequently, a study of companion animals (cats and dogs; n = 514) identified genotype D, which is commonly detected in humans (Chapter 6). However, surprisingly, an investigation of humans (n = 605) in the states of Queensland and Western Australia (Chapter 7) did not identify the genotypes previously found in cats, dogs and wildlife in the state of Victoria, although, surprisingly, a known genotype (i.e., ALP1) recorded in farmed alpacas (Chapter 5) was identified in humans for the first time (Chapter 7). This latter finding raises questions about the transmissibility of this genotype from alpaca to humans. Overall, this thesis has recorded E. bieneusi, for the first time, in a variety of animals in Australia, and provides a first glimpse of the epidemiology of this microsporidian in this country (Chapter 8). The findings of this thesis indicate that the animals studied here, including eastern grey kangaroos, swamp wallabies and wombats; sambar deer, cattle and alpacas; and cats and dogs, can all carry E. bieneusi genotypes that might infect humans via water, food and/or the environment. To provide further evidence to support this proposal and much deeper insights, well-designed large-scale (temporal and spatial) epidemiological studies are required. Population genomics of E. bieneusi using advanced next-generation sequencing and informatics tools could significantly support this endeavour. Other future investigations might also, for example, attempt to establish in vitro culture and in vivo systems for E. bieneusi for controlled studies of the biology of distinct genotypes of this microbe as well as approaches for assessing the viability and infectivity of E. bieneusi and anti-infectives against this microbe. Clearly, we are only beginning to understand some aspects of this enigmatic pathogen - E. bieneusi.
Molecular pathogenesis of infectious disease in domestic animals
Mycoplasma species are free living bacteria with minimal gene sets, and have been used as model organisms to study the basic requirements for life. M. gallisepticum and M. bovis are important pathogens of birds and cattle and cause considerable economic losses in production animals, in part because of the chronic nature of the diseases they cause in their hosts. The limitations of current methods for detection, treatment and control underline the need for more effective vaccines to control these pathogens. A better understanding of the functions of genes of mycoplasmas is required to identify targets to create live attenuated vaccines. Although the genome sequences of M. gallisepticum and M. bovis have been completed and could lead to the identification of such potential targets, the limited tools available for genetic engineering of mycoplasmas prevents us from ascertaining many of the functions of genes in these organisms. Genetic loci that control CRISPR-based immunity are predicted to be present in mycoplasmas chromosomes, usually in non-coding regions. The M. gallisepticum genome contains a set of cas genes coding for putative type II-A Cas proteins, as well as four CRISPR arrays, one of which is potentially functional. In M. gallisepticum, a putative PAM sequence, NNNAAAA, located downstream of the spacers has been proposed. However, an in silico search of protospacers during the studies described in this thesis suggested that the potential PAM sequence may be TANN and may be located upstream of the spacers. Endogenous CRISPR/Cas systems have been exploited for genomic engineering in a number of bacteria. This opens the possibility of performing targeted mutagenesis of M. gallisepticum genome, by directing its endogenous CRISPR/Cas system to specific sites. In the studies described here, three different CRISPR arrays were constructed with or without the putative PAM sequence identified by in silico analyses. As a proof of concept, the ksgA sequence was targeted because mutations in this gene confer resistance to the antibiotic kasugamycin, allowing development of an efficient screening method using this readily detectable phenotype. The sequence analysis of PCR products showed that there was a higher frequency of indels in ksgA amplicons derived from M. gallisepticum transformed with plasmids containing a synthetic CRISPR array targeting ksgA than in M. gallisepticum transformed with a control plasmid lacking the CRISPR arrays. This indicated that the CRISPR/Cas system in M. gallisepticum could be used to introduce indel mutations into a targeted location. Furthermore, a single spacer in the CRISPR array could also direct the induction of mutations in M. gallisepticum strain S6. This prompted the design of a second type of synthetic CRISPR array that targeted ksgA with one spacer and another target gene with the other spacer. The rationale for this approach was to use kasugamycin resistance as a marker for concurrently induced mutations in the second target. Because they lack enzymes for de novo synthesis of nucleic acid precursors, mycoplasmas rely on their intrinsic nuclease activity to scavenge nucleotides from the environment. The orthologue of the M. bovis major nuclease gene in M. gallisepticum, mnuA, was selected as the second target gene and a spacer was synthesised that targeted a sequence with the PAM sequence identified by other studies on the M. gallisepticum CRISPR-Cas system. Following transformation with an assembled CRISPR array targeting both the ksgA and mnuA genes, kasugamycin resistant colonies were tested for nuclease activity. These transformants had a complete or partial loss of nuclease activity compared to negative control transformants. This suggested that the assembled CRISPR array and the endogenous Cas system suppressed the MnuA nuclease activity. Interestingly, sequence analyses of the mnuA coding sequence of the mutants found that the Cas proteins had not affected the gene itself. However, RT-qPCR analyses of the mnuA mRNA in the CRISPR array transformants found that they had no detectable mnuA mRNA, even though this mRNA was detectable in M. gallisepticum transformed with the negative control plasmid, and that the region of the mRNA 3’ to the site targeted by the spacer was less abundant than the 5’ end of the mnuA mRNA in the CRISPR mutants, suggesting that the mRNA was being cleaved by the M. gallisepticum Cas protein (MgaCas9) at the site targeted by the guide RNA transcribed from the synthetic CRISPR array. The similarities between the amino acid sequence of MgaCas9 and that of the Staphylococcus aureus SauCas9, which is capable of targeting both DNA and RNA, suggest the possibility that MgaCas9 may be able to cleave both DNA and RNA. Further experiments will be required to further characterise the MgaCas9 protein and explore its dependence on one or more PAM sequences. Metabolomics has recently been shown to be a useful tool for determining gene function in M. gallisepticum and M. bovis. The simple genomes of mycoplasmas make them ideal candidates for metabolomic studies. Because of the genome reduction that has occurred in mycoplasmas during their evolution, many biosynthetic pathways are missing resulting in their dependence on transporter proteins to provide nutrients from extracellular sources to supply their metabolic pathways. Therefore, many of the nutrient transport systems in mycoplasmas are likely to be essential for virulence, and thus are potentially targets for development of attenuated vaccines. Here, the polar metabolomes of two transposon mutants were characterised and compared to those of the relevant M. gallisepticum and M. bovis wild type strains. The first mutant selected for this study was a ΔmalF strain of M. gallisepticum with a transposon inserted into the malF gene. The MalF protein is predicted to be a component of an ABC transport system and has been reported to be essential for persistence and pathogenicity of M. gallisepticum strain Ap3AS in vivo. Bioinformatic studies and database searches confirmed that MalF was a component of an ABC transporter protein. Comparison of the steady state metabolite profile of the ΔmalF mutant to that of the wild type strain using gas chromatography/mass spectrometry (GC/MS) and liquid chromatography/mass spectrometry (LC/MS) suggested that MalF was involved in glycerol transport. Labelling experiments utilising 13C-glycerol were used to demonstrate a change in glycerol metabolism in the absence of MalF. To assess the regulatory role of MalF, RT-qPCR studies were carried out on selected genes predicted to be involved in glycerol metabolism, demonstrating upregulation of expression of gtsA and the PTS system in the malF mutant, suggesting the predicted transporter encoded by gtsA and the PTS system were compensating for the loss of MalF, although the 13C- glycerol labelling experiments suggest that GtsA is more likely to be importing glycerol-3-phosphate than glycerol. The ΔmalF mutant produced reduced levels of hydrogen peroxide (H2O2) in vitro than the wild type, presumably due to the changes in glycerol metabolism, which may explain the reduced virulence of this strain in vivo that has been described previously. Electron microscopy demonstrated that the cells of the malF mutant were significantly larger and more elongated, and suggested the loss of the electron dense attachment organelle in the ΔmalF mutant, which may also explain its reduced capacity to colonise and attach to tracheal mucosa in studies described previously. Another mutant selected for metabolomic studies was a strain of M. bovis containing a transposon inserted into a putative component of a phosphotransferase system (PTS) transporter (MBOVPG45_0735). Bioinformatic analyses demonstrated more than 90% similarity to a putative pentitol phosphotransferase enzyme IIA component of M. agalactiae. The steady state metabolite profile of the ΔMBOVPG45_0735 mutant and that of the wild type were compared using both GC/MS and LC/MS, and 13C-glucose labelling experiments were also performed. The significantly lower levels of galactitol and sorbitol in the ΔMBOVPG45_0735 mutant suggested that MBOVPG45_0735 might be involved in transport or metabolism of these sugar alcohols. However, because of technical limitations, very few phosphorylated sugars were detected in the steady state analysis, and 13C labelling was not detected in phosphorylated sugars on either of the platforms, restricting assessment of the PTS function(s). In conclusion, the studies described in this thesis demonstrated the potential for use of the endogenous CRISPR/Cas system of M. gallisepticum to induce targeted mutations in putative virulence genes. This is a significant advance in the tools available to genetically manipulate mycoplasmas, which will progress understanding of the molecular pathogenesis of this organism, as well as assist in studies to synthesise a minimal bacterial genome based on this species. Furthermore, these studies show that metabolomics has potential to assist in understanding gene function in mycoplasmas, a key step forward in designing effective vaccines.
Molecular pathogenesis of Mycoplasma bovis
Mycoplasma bovis is an important pathogen in cattle production systems worldwide and causes significant economic losses. It causes mastitis, pneumonia, otitis media, keratoconjunctivitis and polyarthritis in affected cattle. Antibiotics have limited efficacy in treating M. bovis infections so vaccines are considered the method most likely to achieve effective control of the disease it causes. As M. bovis lacks a cell wall, lipoproteins, which are exposed on the cell surface, play a key role in its survival in the host through their role in the acquisition of essential nutrients, in adherence to cell surfaces, in colonisation of the host and in evasion of the immune system. Mycoplasmas are the simplest free-living organisms, and are generally regarded as having evolved to retain only the essential machinery necessary for survival in their hosts. This study sought to understand the functions of selected membrane genes of M. bovis and their roles in adhesion, colonisation and immune evasion. The gross pathology of respiratory disease caused by M. bovis is characterised by the presence of areas of caseous necrosis and purulent exudates in the affected pulmonary tissues. Calves experimentally infected with M. bovis have an infiltration of neutrophils into the bronchioles. The studies described in this thesis showed that these neutrophilic infiltrates contained neutrophil extracellular traps (NETs) by detecting changes in the morphology of neutrophils in these sites, and free histones and neutrophil elastase, which are derived from the nuclear components of NETs. Further investigations of the induction of NETs by M. bovis were conducted in vitro using neutrophils isolated from the blood of healthy cattle. These studies demonstrated that M. bovis does induce the formation of NETs by neutrophils, but that the stability of these NETs was compromised by the product of the major membrane nuclease gene mnuA. The MnuA nuclease is the major nuclease of M. bovis detected in vitro. A fluorescence assay was developed to quantify NET formation when neutrophils were stimulated with M. bovis strain PG45 and mutants derived from it. The formation of NETs in bovine neutrophils was dependent on the presence of a relatively high concentration of M. bovis (a multiplicity of infection of 100), but NETs were degraded at very low concentrations of M. bovis, demonstrating the potency of the MnuA gene product in the degradation of NETS and its likely role in enabling M. bovis, and possibly other bacteria at the sites of infection, to escape from NETs. Ovine neutrophils were also shown to release NETs after stimulation with M. bovis and the MnuA nuclease also degraded these NETs. M. bovis-induced NETosis was found to result from stimulation of an alternative pathway to the classical NET induction pathway. It was dependent on neutrophil elastase and the ERK pathway, but was induced independently of the formation of reactive oxygen species (ROS), or the protein kinase C (PKC) and MAPK pathways. These studies suggest that the MnuA nuclease assists M. bovis in evading the innate immune system and in the establishing chronic infection. Mycoplasmas must adhere to epithelial cell surfaces to successfully colonise their host. To determine the role of selected genes of M. bovis in adhesion, an in vitro adhesion assay was developed using the Madin-Darby bovine kidney (MDBK) cell line. The adhesion of wildtype M. bovis strain PG45 was compared to mutants in which a transposon had been inserted into the open reading frames MBOVPG45_0416 or MBOVPG45_0565. The products of both of these genes have been found to have fibronectin binding activity in earlier studies. Both mutants had significantly reduced binding to MDBKs compared to the wildtype strain. In addition, adhesion of wildtype PG45 to MDBK cells was significantly inhibited by treatment of the M. bovis PG45 wildtype strain with polyclonal sera raised against the products of MBOVPG45_0565 or MBOVPG45_0416. Taken together, these findings are consistent with a significant role of these gene products in the adhesion of M. bovis to cell surfaces. In conclusion, the studies described in this thesis have demonstrated the likely role of three membrane genes of M. bovis in its survival in the host and have highlighted some of the mechanisms that contribute to the virulence of this pathogen. This in turn has enhanced our understanding of the pathogenesis of the disease caused by this organism and will assist in the design of vaccines and diagnostic tools that can be used to more effectively control disease caused by M. bovis
Assessment of treatment outcomes and symmetric dimethylarginine in hyperthyroid cats treated with a fixed dose of radioiodine
Background: Hyperthyroidism and chronic kidney disease (CKD) are common in geriatric cats and often occur concurrently. Among other effects, hyperthyroidism causes protein catabolism and increases renal blood flow and the glomerular filtration rate (GFR). These effects render traditional renal markers insensitive for the detection of CKD in cats with uncontrolled hyperthyroidism. Early identification of occult CKD may influence the prognosis, monitoring and therapy of hyperthyroidism or CKD. Symmetric dimethylarginine (SDMA) is a new serum renal biomarker that is stable, easily measured and has been recently validated for use in cats. SDMA correlates well with GFR and is not affected by muscle mass. SDMA is more sensitive than serum creatinine in detecting early CKD in non-hyperthyroid cats, but our understanding of its performance in hyperthyroid cats is in its infancy. In Australia, the gold standard treatment for cats with benign hyperthyroidism (i.e. thyroid hyperplasia or adenoma) is a fixed dose of orally administered radioiodine, regardless of the serum total thyroxine (TT4) concentration at the time of diagnosis. The development of iatrogenic hypothyroidism after radioiodine treatment is detrimental to renal function and may negatively affect long-term survival. As a result, the goal of radioiodine treatment is to maximise the number of cats that achieve euthyroidism while limiting the development of hypothyroidism. Currently, information regarding long-term outcomes in cats treated with an oral fixed dose of radioiodine is limited and there are no practical markers to detect occult CKD in these cats. Objectives: This study aimed to describe the treatment outcomes following oral administration of a fixed dose (138 MBq; 3.7 mCi) of radioiodine in a sample of hyperthyroid cats and examine the correlation between serum TT4 concentrations before and after treatment. A second aim was to measure the serum SDMA concentration in hyperthyroid cats before and after treatment to determine the relationships between SDMA, creatinine and TT4 concentrations at both time points. Method: In a sample of cats previously treated with a fixed dose of oral radioiodine, TT4 concentration and renal clinical pathological parameters were documented before and after radioiodine treatment. Logistic regression was used to assess the relationship between TT4 concentrations before and after treatment. The differences in pre- and post-treatment variables for cats that had TT4 concentrations below or within the reference interval were also assessed. Renal parameters including urine specific gravity and serum SDMA, urea and creatinine concentrations were measured before and three months after receiving a fixed dose of oral radioiodine treatment in prospectively enrolled hyperthyroid cats. The Pearson correlation coefficient was used to determine the association between treatment and serum TT4, SDMA and creatinine concentrations. To assess agreement between serum SDMA and serum creatinine regarding categorisation of CKD staging, Goodman and Kruskal’s gamma statistic was used. Results: Of 162 cats, 133 (82%) had TT4 concentrations within the reference interval after treatment. Four (3%) and 25 (15%) cats had TT4 concentrations above and below the reference interval after treatment, respectively. The severity of hyperthyroidism at diagnosis, as measured by the percentage elevation of TT4 concentration above the reference interval, had no impact on the odds of cats having TT4 concentrations below the reference interval after treatment (OR = 1.00; 95% CI 0.96–1.05; p = 0.93). Over the follow-up period after radioiodine treatment, serum SDMA increased in 51 of 74 (69%) cats, whereas serum creatinine increased in 78 of 80 (98%) cats. A moderate correlation between serum SDMA and creatinine was seen after treatment (r = 0.523, p < 0.001) but not before treatment (r = 0.523, p = 0.23). Where assessable after treatment, serum SDMA and creatinine did not agree in the staging of cats with kidney dysfunction based on the International Renal Interest Society CKD guidelines. No significant correlation was seen between serum SDMA and TT4 concentration at any time point. Conclusions and relevance: When using an orally administered fixed dose of radioiodine for the treatment of feline hyperthyroidism, TT4 concentrations at diagnosis cannot be used to predict TT4 concentrations after treatment. The proportion of cats with TT4 concentrations below the reference interval after treatment was 15%. Further work is required to optimise oral radioiodine dosing to maximise euthyroidism. Extra-renal factors associated with hyperthyroidism appear to hinder the ability of SDMA to identify cats with occult CKD before treatment. Further work is required to elucidate the mechanism and impact on the performance of SDMA as a diagnostic test of renal function in hyperthyroid cats.
Response of subchondral bone and articular cartilage to joint surface incongruity: an equine osteochondral defect model
Many diseases and injuries of synovial joints result in a focal loss of articular surface congruity. These conditions are known to lead to degenerative changes within the affected joint and subsequent osteoarthritis (OA). Incongruity due to loss of articular cartilage is also a feature of OA itself. However there is little work specifically investigating how the tissues of a synovial joint respond to loss of congruity. The aim of this study was to observe how articular cartilage and subchondral bone (SCB) respond to a focal loss of surface congruity using an equine model. It was hypothesised that the degree of remodelling and bone volume of SCB is dependent on local loading of the overlying joint surface and that intact cartilage would respond to changes in local surface pressure in a way that attempts to restore congruity. An osteochondral defect was created in one surface of the midcarpal joint of six adult horses. Two weeks post-operatively the animals began an eight-week treadmill training program. Following this, osteochondral samples were collected from the site of the defect and immediately adjacent to it in the radial carpal bone (Cr), and from the unloaded site directly opposing the defect and a loaded site immediately adjacent to this in the third carpal bone (C3). Control samples were collected from equivalent sites in the sham-operated contralateral limb and from un-operated, untrained control animals. Undemineralised samples were imaged with microCT and backscattered scanning electron microscopy (BSEM) and stained with Masson’s Goldner trichrome to measure cartilage thickness and osteoid. Decalcified cryosections were stained for TRAP to identify osteoclasts. Cartilage was examined with routine histology and differential interference contrast microscopy. In SCB immediately below the Cr lesion in treated joints bone volume fraction (BV/TV) was 32% lower and bone formation and erosion were markedly increased, whereas in the deeper bone formation was increased, erosion activity became less prominent and BV/TV did not change. In the unloaded region of C3 opposing the defect in treated joints the hyaline cartilage (HC) was 50% thicker, while SCB displayed a mild increase in remodelling activity with a net decrease in BV/TV of 5.3%. In the loaded region adjacent to the lesion in Cr HC was 30% thinner in treated joints while there was no change in SCB parameters. At the loaded site of C3 the HC thickness was unchanged while the calcified cartilage in treated joints was 40% thicker than in controls. In the underlying SCB in treated joints there was a mild reduction in remodelling activity and unchanged BV/TV. In this equine osteochondral defect model the articular cartilage and SCB of adult synovial joints responded in a focal manner to changes in joint surface congruity. Local unloading due to a loss of surface contact resulted in thickened HC and, in the underlying SCB, increased remodelling activity and loss of bone volume. Focally increased loading resulted in thickened calcified cartilage and decreased SCB bone remodelling activity. These focal responses may play a role in normal joint homeostasis and in the pathogenesis of OA.