The global population is rapidly urbanising, with East Africa experiencing some of the fastest rates of urban growth. Urbanisation drives changes in diet, including increased consumption of animal source products (ASPs), and livestock value chains are becoming increasingly long and complex to meet these demands. This may place urban consumers at increased risk of food-borne infectious diseases. Evaluation of food-borne disease (FBD) risk to urban consumers in developing countries has been hampered by a lack of data on the composition of urban diets, and a lack of methodologies to systematically assess risk along food value chains which are typically informal, unregistered, and unregulated. This research used a value chain risk assessment approach (VCRA) to evaluate food-borne infectious disease risks along dairy value chains supplying Moshi Municipality, the regional capital of Kilimanjaro, Northern Tanzania. Our findings demonstrated that by far the most frequently consumed products were unpackaged milk and mtindi (fermented milk). While there was some role for urban livestock keepers in supplying these products to their communities directly, most of the milk and mtindi sold within Moshi originated with milk produced by smallholder farmers in rural areas surrounding the towns. Both the milk and mtindi value chains involved similar value chain nodes and actors, with a large degree of overlap between the formal and informal sectors and little to no formal education or training on milk handling and hygiene for chain participants. VCRA identified the bulking, wholesale and retail stages of the value chain as potential hotspots for introducing infectious disease risk. Consumers were well informed about many of the FBD risks posed by milk, and took active steps to mitigate these risks by boiling before consumption; however they perceived mtindi as posing a lower risk and were unable to mitigate risks with any preparatory step as mtindi is consumed as purchased. The highest risk to consumers was estimated to be posed by mtindi rather than milk, particularly mtindi made from leftover unsold milk, as this milk had a high risk of contamination. More studies are needed to investigate the infectious hazards present in both mtindi and other fermented milk products which are consumed widely across the region. The practice of valorising leftover ASPs as alternative products for human consumption may represent a particular source of FBD risk to urban consumers in developing countries.

Rhipicepahlus (Boophilus) microplus is an obligate feeding, hard tick of great economical importance in the cattle industry. Every year billions of dollars of loss is attributed to R.(B) microplus, mainly through loss of cattle due to pathogens transmitted such as Babesia and Anaplasma, but also through damage to hides from blood-feeding. There is conflicting evidence regarding the taxonomic status of R.(B) microplus, however the most recent published research has been in support of the reinstatement of R.(B) australis as a species distinct from R.(B) microplus. The way in which some members of the scientific community have responded to the designation of separate species has implications for vaccine and acaricide research. In this study, we aimed to resolve the taxonomic status of Rhipicephalus (Boophilus) microplus, using morphological and phylogenetic approaches. 1,650 Rhipicephalus (Boophilus) microplus ticks from Australia, Thailand, South Africa, North and Central America and South America were used in this study. 340 specimens consisting of 170 R.(B) annulatus (USA) and 170 R.(B) decoloratus (South Africa) were also used. To maximize the information obtained from morphological observations, three methods were used; a binary scoring system based on previously described features, a standard morphometric method, and the more novel approach of geometric morphometrics. For the phylogenetic analysis three genes were used; the mitochondrial gene COX1 and two functional nuclear genes; Bm86 and βAOR. Morphological scoring is the process of assigning a binary value to any feature as being present or absent, or satisfying a logical comparator. For this study the scoring matrix was based on previously described sets of morphological criteria used for discriminating among species. Each of the populations for which samples were obtained was tested using four two-way analyses, each of which was designed to test whether a sample should be classified as one of two possible species: R.(B) australis versus R.(B) microplus; R.(B) microplus versus R.(B) annulatus; R.(B) microplus versus R.(B) decoloratus; and R.(B) annulatus versus R.(B) decoloratus. The scoring system was highly repeatable for the differentiation of males and females of R. (B) annulatus and R.(B) decoloratus from both of R.(B) microplus and R.(B) decoloratus. However, in the case of R.(B) australis and R.(B) microplus, clear differentiation was not achieved for either male or female ticks. Among females, the Australian population were classified almost evenly as R.(B) australis and R.(B) microplus, with 8 individuals showing a mixture of features and therefore not able to be classified. Ticks from the rest of the regions were mainly classified as R.(B) microplus, which is to be expected as R.(B) australis is reported in Australia. However, only the Mozo isolates were classified as solely R.(B) microplus. The remaining regions included several ticks with mixed features. Six ticks from South Africa, and four of the Juarez isolate were classified as R.(B) australis. Among the males an entirely different pattern emerged. Most male ticks from all geographical locations were classified as either R.(B) australis or showed a mixture of both features, with only a small number scoring as R.(B) microplus. Morphometrics is the linear measurement from one anatomical landmark to another and is a widely-used technique for quantifying phenotypic variation. Twelve features based on previous morphometric work were used. The results obtained from this study varied according to stage and sex. For the larvae, the Fisher Pairwise comparison showed that the Australian ticks tended to have a shorter body length, idiosoma length and narrower scutum width. Among the remaining morphological features, there were no consistent patterns in the different populations and species. A principal components analysis (PCA) was undertaken and in PC1 the strongest feature was scutum length and hypostome length. In PC2 the strongest feature was idiosoma length. The PCA of larval stage ticks didn’t provide conclusive evidence that R.(B) australis is a distinct species from R.(B) microplus, and there was no obvious grouping based on region at all, even when including R.(B) decoloratus and R.(B) annulatus. In relation to the male ticks studied, the Fisher Pairwise comparison and the PCA showed that Australian males (presumed R.(B) australis) were significantly different from the other isolates. As with the larvae, no patterns were seen in the other populations, based on species or region. In PC1 palpal length measures were the strongest features for differentiation and in PC2 the length of the ventral basis capituli had the strongest effect. The adult female samples yielded a mixed result. There was no real trend in the size of Australian ticks observed from the Fisher Pairwise comparison. However, R.(B) decoloratus tended to be smaller for most of the morphological features tested and R.(B) annulatus tended to be larger. This observation was inconsistent with the results from the PCA, in which there was grouping of Australian ticks. Measures of palpal length, width of the basis capituli and the length of the dorsal basis capituli were the strongest for differentiation in PC1. In PC2 the length of the ventral basis capituli was the strongest feature for differentiating populations. Geometric morphometrics is the quantitative representation of shape using coordinates in the form of landmarks, instead of measurements and is intended to give the shape of the feature independent of size. Hence it is useful for eliminating the effect of size distortion occurring with physiological changes. Geometric morphometric analysis did not clearly and consistently enable the differentiation of any of the populations of ticks in this study. Each feature differed among samples in different sets of pairwise relationships. Mitochondrial cytochrome oxidase subunit I gene (COX1) has been presented as a suitable mitochondrial gene to clarify complex groupings that were not resolved when using other mitochondrial genes. COX1 has also been proposed to be the main gene for differentiating between R.(B) microplus and R.(B) australis. The aims of this study were to confirm whether COX1 can be used to resolve complex relationships within the R.(B) microplus clade and to determine whether there is justification for the view that R.(B) australis is a distinct species from R.(B) microplus. Maximum likelihood trees were constructed with a Bootstrap analysis. A relaxed clock Bayesian analysis was then undertaken to estimate topology and divergence timings, using three ticks found in amber covering three genera: Amblyomma, Hyalomma and Ixodes to calibrate the clock. These analyses suggest that R.(B) microplus is a clade, containg five subspecies including R.(B) annulatus, R.(B) australis, and three, regionally based clades of R.(B) microplus: 1. All the South and Central American isolates together with isolates from Cambodia, Thailand, and some of those from Malaysia; 2. Indian and the remaining Malaysian isolates; 3. Most of the Chinese isolates. R.(B) decoloratus shares a common ancestor with R.(B) microplus and R.(B) annulatus however it is clearly divergent, appearing to be more related to R. bursa. All proposed groups of R.(B) microplus also appear to have evolved within the same time scale (within the last 20 million years). Bm86 is the name given to a midgut glycoprotein that is the target antigen of the only commercially available vaccine against ticks. All the prior work on this gene has been conducted using cDNA and suggests a high degree of sequence variation and the presence of different isoforms. The aim was to use genomic DNA to examine the regional variation in the Bm86 sequences and to determine whether Bm86 variation segregated according to the recently proposed taxonomic re-classification of R.(B) microplus and R.(B) australis. / After extensive optimization, it was found that all primer sets, including those previously published and those designed in this project, failed on the extracted genomic DNA from all isolates. High variability in the published cDNA sequences indicated an extremely high mutation rate, which could potentially be linked to variation in the function of the protein and its utility as a vaccine immunogen. Analysis of sequence alignments from publicly available databases did not allow grouping of samples by either geographical location or proposed taxon. These findings are in apparent contradiction to claims by other researchers that regional variation in the efficacy of the vaccine is associated with regional variation in sequence.

Johne’s disease, caused by Mycobacteria avium subsp. paratuberculosis (MAP), plagues cattle and dairy farmers worldwide. Infected animals suffer from chronic granulomatous enteritis, reduced fertility, decline in milk production and emaciation. It is spread through colostrum, milk and faeces. Johne’s disease has also become strongly associated with human Crohn’s disease due to the similarity in symptoms and the presence of MAP in samples taken from Crohn’s disease patients. Currently there is not cure for Johne’s disease, therefore, routine testing and isolation or culling of diseased animals are measures used to prevent infection. At present, the gold standard test for MAP is by faecal culture, which can take up to 4 months to reach a diagnosis. The low-cost ELISA test has a shorter diagnosis duration, however, it uses blood or milk samples for testing which do not correlate with the bacterial shedding in faecal samples. As a result of this mismatch in bacterial shedding between blood, milk and faeces, ELISA testing lacks in both sensitivity and specificity to MAP when compared to faecal culture and PCR tests. PCR can be sensitive and specific for MAP testing, although it is currently the most expensive test for MAP detection within the UK. Reducing the cost of PCR testing was one of the motivating factors during the development of this assay and device. In this thesis, surface acoustic waves (SAW) have been used to create, develop and test a diagnostic device and the assay for Johne’s disease. SAW is becoming an increasingly popular tool in the field of diagnostic devices due to its multi-functionality which allows for many pieces of laboratory equipment to be consolidated into one device. In using SAW there was no longer a need for laboratory equipment usually used to perform a DNA extraction because SAW could be used to mix, heat and move the sample droplet. Traditional, laboratory-based faecal DNA extraction was adapted for use with SAW. To date, faecal DNA extraction using SAW has not been published. The SAW-driven, droplet-based assay developed during this project used 90% less DNA extraction reagents than the traditional tube-style method. Due to the thermal resistant nature of MAP it is particularly difficult to lyse during DNA extraction. The newly created SAW faecal DNA extraction was used to retrieve and clean DNA sufficiently for successful PCR to follow. In the context of faecal samples, this was particularly challenging due to the PCR inhibitors found in these complex samples. To investigate the effectiveness of the SAW DNA extraction, K-10 strain of genomic MAP DNA, MAP cell cultures and pre-tested bovine faecal samples were tested to prepare the MAP DNA for amplification and detection. DNA extraction was followed by SAW PCR. The sensitivity and specificity of the SAW PCR was ensured by using both IS900 and F57 target sequences. IS900 is repeated 17 times in the MAP genome thereby providing sensitivity down to 102 CFU/g. F57 is only repeated once therefore providing specificity. Specificity of the assay was further improved by using TaqMan probes to quantify the PCR. In order to keep this PCR-based assay stable in the absence of cold-chain storage, the disaccharide sugar trehalose was added to the PCR reagents and freeze dried to determine its ability to maintain performance. These experiments enhanced the future portability of this assay. It was found that reagents maintained activity for at least 41 days after freeze drying and this duration is expected to be extended. During this thesis, a newly developed acoustically driven DNA extraction and qPCR was integrated onto one device which had the capability of detecting as few as 5 MAP genomes. This novel proof-of-concept research, lays the foundation for an acoustically driven portable device for use on faecal samples and in this case, for the detection of Johne’s disease.

Research was undertaken to investigate the demographics of the pet population in a local community. A sample of the general dog population and owners was then compared with a sample of relinquished dogs and their surrendering owners, to identify dog and household characteristics associated with canine relinquishment to an animal welfare centre (AWC). The investigation was carried out in Strathclyde Region, Scotland. A cross-sectional telephone survey was conducted to establish the pet owning status of a randomly selected sample of households. The dog and cat populations were then estimated by extrapolation of results to local census data. Using an area based approach, the association between deprivation, community setting and pet ownership were assessed using available census data. A Geographical Information System approach was applied to the data to display the spatial distribution of pet ownership in Strathclyde Region. Dog owning households identified by the telephone survey were invited to participate in a second study. This case-control study compared information regarding dog and household characteristics of these successful owners (controls), with a sample of unsuccessful dog owners and their pets (cases). The control sample was assessed by mailed questionnaire. Case households were selected by AWC staff at the time of relinquishment and data collected using self-administered questionnaires. Results revealed that 36.1% of households were pet owners, with dogs and cats being the most prevalent species owned. The canine population of Strathclyde Region was estimated to be 248,649 and the feline population estimated to be 170,044. These dogs and cats were owned by an estimated 185,589 and 121,235 households, respectively. Deprivation and urban communities were negatively associated with pet ownership. A sample of 360 of the dog owning households agreed to participate in the second study. Based on a response rate of 89.2%, 321 of these households returned usable questionnaires. Comparison of data from these questionnaires with data obtained from 49 case households revealed that several factors were associated with relinquishment. Although certain dog characteristics were identified as important, most of the predisposing characteristics were owner-related. Uneducated, inexperienced dog owners who impulsively acquired their dog for little cost, were more likely to relinquish their pet after a short duration of ownership because of inappropriate care expectations, lack of planning and the dog failing to meet their expectations. Surprisingly, dog behavioural problems were no more prevalent in relinquished animals. These data were entered into a multivariable logistic regression model. Variables retained in the final model associated with relinquishment included ownership of a small mammal pet, no history of previous pet ownership, young dogs, mixed breed dogs, lack of veterinary care, short duration of residence in the present home and absence of a garden. The study identified several risk factors associated with canine relinquishment, many of which could potentially be modified to decrease the numbers of animals abandoned at AWCs. Use of the multivariable logistic regression model could enable the assessment of the likelihood of future relinquishment. In particular, application of the model in AWCs could decrease the number of adopted dogs that are subsequently relinquished.

The under-reporting of cases of infectious diseases is a substantial impediment to the control and management of infectious diseases in both epidemic and endemic contexts. Information about infectious disease dynamics can be recovered from sequence data using time-varying coalescent approaches, and phylodynamic models have been developed in order to reconstruct demographic changes of the numbers of infected hosts through time. In this study I have demonstrated the general concordance between empirically observed epidemiological incidence data and viral demography inferred through analysis of foot-and-mouth disease virus VP1 coding sequences belonging to the CATHAY topotype over large temporal and spatial scales. However a more precise and robust relationship between the effective population size (N_e) of a virus population and the number of infected hosts (or 'host units') (N) has proven elusive. The detailed epidemiological data from the exhaustively-sampled UK 2001 foot-and-mouth (FMD) epidemic combined with extensive amounts of whole genome sequence data from viral isolates from infected premises presents an excellent opportunity to study this relationship in more detail. Using a combination of real and simulated data from the outbreak I explored the relationship between N_e, as estimated through a Bayesian skyline analysis, and the empirical number of infected cases. I investigated the nature of this scaling defining prevalence according to different possible timings of FMD disease progression, and attempting to account for complex variability in the population structure. I demonstrated that the variability in the number of secondary cases per primary infection R_t and the population structure greatly impact on effective scaling of N_e. I further explored how the demographic signal carried by sequence data becomes imprecise and weaker when reducing the number of samples are described, including how the extent of the size and structure of the sampled dataset impact on the accuracy of a reconstructed viral demography at any level of the transmission process. Methods drawn from phylodynamic inference combine powerful epidemiological and population genetic tools which can provide valuable insights into the dynamics of viral disease. However, the strict and sensitive dependency of the majority of these models on their assumptions makes estimates very fragile when these assumptions are violated. It is therefore essential that for these methods to be applied as reliable tools supporting control programs, more focused theoretical research is undertaken to model the epidemiological dynamics of infected populations using sequence data.

Leptospirosis is an important but neglected zoonotic disease that is often overlooked in Africa. Although comprehensive data on the incidence of human disease are lacking, robust evidence of infection has been demonstrated in people and animals from all regions of the continent. However, to date, there are few examples of direct epidemiological linkages between human disease and animal infection. In East Africa, awareness of the importance of human leptospirosis as a cause of non-malarial febrile illness is growing. In northern Tanzania, acute leptospirosis has been diagnosed in 9% of patients with severe febrile illness compared to only 2% with malaria. However, little is known about the relative importance of different potential animal hosts as sources of human infection in this area. This project was established to investigate the roles of rodents and ruminant livestock, important hosts of Leptospira in other settings, in the epidemiology of leptospirosis in northern Tanzania. A cross-sectional survey of rodents living in and around human settlements was performed alongside an abattoir survey of ruminant livestock. Unusual patterns of animal infection were detected by real-time PCR detection. Renal Leptospira infection was absent from rodents but was detected in cattle from several geographic areas. Infection was demonstrated for the first time in small ruminants sub-Saharan Africa. Two major Leptospira species and a novel Leptospira genotype were detected in livestock. L. borgpetersenii was seen only in cattle but L. kirschneri infection was detected in multiple livestock species (cattle, sheep and goats), suggesting that at least two distinct patterns of Leptospira infection occur in livestock in northern Tanzania. Analysis of samples from acute leptospirosis in febrile human patients could not detect Leptospira DNA by real-time PCR but identified social and behavioural factors that may limit the utility of acute-phase diagnostic tests in this community. Analysis of serological data revealed considerable overlap between serogroups detected in cattle and human leptospirosis cases. Human disease was most commonly attributed to the serogroups Mini and Australis, which were also predominant reactive serogroups in cattle. Collectively, the results of this study led to the hypothesis that livestock are an important reservoir of Leptospira infection for people in northern Tanzania. These results also challenge our understanding of the relationship between Leptospira and common invasive rodent species, which do not appear to maintain infection in this setting. Livestock Leptospira infection has substantial potential to affect the well-being of people in East Africa, through direct transmission of infection or through indirect effects on food production and economic security. Further research is needed to quantify the impact of livestock leptospirosis in Africa and to develop effective interventions for the control of human and animal disease.

Feline immunodeficiency virus (FIV) is a naturally occurring lentivirus of domestic cats, which shares many similarities with its human counterpart, human immunodeficiency virus (HIV). FIV infects its main target cell, the CD4+ T lymphocyte, via interactions with its primary receptor CD134 (an activation marker expressed on activated CD4+ T lymphocytes), and, the chemokine receptor CXCR4. According to the different ways in which FIV isolates interact with CD134, FIV may be categorised into two groups. The first group contains strains that tend to dominate during the earlier phase of infection, such as GL8 and CPG41. These strains are characterized by their requirement for an additional interaction with the second cysteine rich domain (CRD2) of the CD134 molecule and are classified as “CRD2-dependent” strains. The second group, on the other hand, contains either laboratory-adapted isolates or isolates that emerge after several years of infection, such as PPR or the GL8 variants that emerged in cats 6 years post experimental infection and were studied in this thesis. These isolates are designated “CRD2-independent” as they can infect target cells without interacting with CRD2 of the CD134 molecule. This study provides the first evidence that FIV compartmentalisation is related to FIV-CD134 usage and the tissue availability of CD134+ target cells. In tissue compartments containing high levels of CD134+ cells such as peripheral blood and lymph nodes, CRD2-dependent viruses predominated, whereas CRD2-independent viruses predominated in compartments with fewer CD134+ cells, such as the thymus. The dynamics of CD4+CD134+ T lymphocytes at different stages of FIV infection were also described. The levels of CD4+CD134+ T lymphocytes, which were very high in the early phase, gradually decreased in the later phase of infection. The dynamics of CD4+CD134+ T lymphocyte numbers appeared to correlate with FIV tropism switching, as more CRD2-independent viruses were isolated from cats in the late phase of infection. Moreover, it was observed that pseudotypes bearing Envs of CRD2-dependent variants infected CD134+ target cells more efficiently than pseudotypes bearing Envs of CRD2-independent variants, confirming the selective advantage of CRD2-dependent variants in environments with high levels of CD134+ target cells. In conclusion, this study demonstrated that target cell types and numbers, as well as their dynamics, play important roles in the selection and expansion of FIV variants within the viral quasispecies. Improved understanding of the roles of target cells in FIV transmission and pathogenesis will provide important information required for the development of an improved, more successful protective FIV vaccine and will provide insight into the development of effective vaccines against other lentiviral infections such as HIV.

Teladorsagia circumcincta is a common abomasal parasite of sheep in temperate regions and is one of the major causes of parasitic gastroenteritis (PGE) in growing lambs. Control of infection is achieved using anthelmintic drugs; however, this practice is rapidly becoming unsustainable due to widespread anthelmintic resistance within the T. circumcincta population. Sheep can acquire protective immunity against this parasite; immunity involves local and systemic antibodies and immune cells which can impair worm growth and fecundity and lead to expulsion of the parasites from the abomasum. Vaccination against this parasite is therefore a feasible option of control. A recent study showed that a recombinant vaccine cocktail containing 8 T. circumcincta antigens significantly reduced the faecal egg count and worm burdens of immunised sheep, compared to an adjuvant-only control group. However, the recombinant antigens induced a suboptimal antibody response to the recombinant antigens. This suggests that differencies between the native antigens and their recombinant versions may exist, possibly due to variations in structure and/or post-translational modifications (PTMs). The main aim of this work was to use a novel expression system, the free living nematode Caenorhabditis elegans, to generate alternative recombinant versions of two of the T. circumcincta antigens used in the 8-antigen vaccine, cathepsin F (Tci-CF-1) and monocyte Migration Inhibitory Factor (Tci-MIF-1). This was achieved by micro-injection of C. elegans worms with plasmids containing the cDNA sequences of Tci-cf-1 and Tci-mif-1 followed by purification of recombinant Tci-CF-1 and Tci-MIF-1 from the transformed worms. Immune recognition, enzyme activity and biological effects on sheep cells of the recombinant antigens were characterised. The results show that immunisation-induced antibodies bind to native Tci-CF-1 purified from T. circumcincta L4 ES, whereas infection-induced antibodies were unable to bind the recombinant Tci-CF-1 versions. Further characterisation of recombinant Tci-CF-1 versions expressed in C. elegans or Pichia pastoris showed that in order to be enzymically active, these proteins require cleavage of the pro-peptide by an exogenous enzyme and that some differences were present in the glycosylation of the recombinant versions and native Tci-CF-1. Characterisation of both recombinant Tci-MIF-1 versions showed that although both are enzymically active, neither showed a significant inhibitory effect on the migration of sheep monocytes or on the activation of sheep macrophages in vitro compared to unstimulated controls. It is speculated that Tci-MIF-1 may be involved in T. circumcincta larval development rather than host immunosuppression.

Feline Osteoarthritis (OA) is a pathological change of a diarthrodial articulation which primarily occurs in older cats. The aims of this study were:- 1) to define the radiographic features of OA in the cat for each individual appendicular joint; 2) to relate the radiographic features to the gross pathologic and histopathologic features; 3) to explore underlying causes of OA in cats, 4) to identify the presence of Protease Activated Receptor-2 (PAR-2) and matriptase in feline articular cartilage and synovial membrane and to determine their role in OA pathogenesis. The present study has defined five radiographic features of OA for each appendicular joint:- presence of osteophytes, enthesiophytes, areas of abnormal mineralisation,synovial effusion and joint remodelling. The study furthermore suggested that increases in radio-opacity beneath the semilunar notch, along the femoral trochlea, beneath the tibial plateau and on the femoral head/neck are also important radiographic features. The radiographic prevalence was highest in the elbow (23.9%, 93/389) and stifle (23.9%,93/389) joints, followed by the hip (21.1%, 82/389), tarsal (17.7%, 69/389), shoulder(6.7%, 27/389) and carpal (6.4%, 25/389) joints. The results from this study demonstrate that the presence of a radiographically apparent supinator sesamoid bone(SSB), meniscal mineralisation (MM) and two fabellae are related to cartilage pathology and can be indicators of OA. Prevalence rates for gross pathology changes were highest in the elbow (20.2%,102/506) joint, followed by the stifle (19.6%, 99/506), hip (18.4%, 93/506), shoulder (17.8%, 90/506), tarsal (15.0%, 76/506), and carpal (9.1%, 46/506) joints. Eight key gross pathologic features were identified- cartilage discolouration, cartilage fibrillation,cartilage ulceration, cartilage erosion, osteophytes, thickening of joint capsule, synovium discolouration and joint remodelling. The radiographic and gross pathologic total scores were positively correlated in each appendicular joint and the joint most likely to have cartilage damage without radiographic evidence of OA is the shoulder (71.1%, 64/90) followed by the elbow (39.1%, 9/23), hip (32.4%, 11/34), stifle (26.1%,6/23), carpal (23.1%, 21/91) and tarsal (14.9%, 7/47) joints. Four possible underlying conditions that lead to secondary OA were identified:- radioulnar incongruity, hip dysplasia (HD), cranial cruciate ligament (CCL) disease and primary meniscal mineralisation. The identification of PAR-2 and matriptase proteins and gene expression in feline articular tissues is a novel and important finding supporting the hypothesis that serine proteases are involved in the articular cartilage degradation seen in feline OA.

No description available.