Antimicrobial peptides (AMPs) are innate host defense peptides that protect against pathogenic microbes by neutralizing toxins or via a direct killing mechanism. AMPs are classified based on their physical properties such as charge, structure, and binding motifs. Here we investigated the antimicrobial and immune-modulating effects of the Regenerating Islet-Derived Protein 3 (REG3) family and LL-37 REG3 peptides are C-type lectins and have been demonstrated to have antimicrobial activity against Gram-positive bacteria by binding to sugars on the peptidoglycan membrane of these bacteria. A similar strategy is also employed by the lectin Surfactant Protein-D which has been shown to bind and neutralize Influenza A Virus (IAV). REG3 peptides were shown to be expressed in the lungs of mice infected with IAV. We observed reduction of IAV infected cells when IAV was pre-incubated with an Escherichia coliexpressed recombinant version of human REG3A peptide. This peptide also modified interaction of IAV with primary human neutrophils. However, these effects were lost when using a mammalian cell expressed recombinant REG3A. A second member of the REG3 family, REG3G, showed minimal inhibition of IAV infection. While the mechanism remains unclear, LL-37 has demonstrated killing activity against a spectrum of microbes including IAV. Previous work from our group identified the core domain of LL-37 responsible for IAV neutralization. In addition, our group showed that LL-37 modulates interaction of IAV with neutrophils. Here we tested three modified versions of LL-37 that retain the overall size and charge of LL-37, but with modifications in the core domain reducing hydrophobicity. We observed that these mutants retain IAV killing activity across multiple strains. In addition, these mutants retain the modulation of IAV induced neutrophil responses. We also found that the compounds sodium butyrate and Entinostat, which can upregulate endogenous expression of LL-37, have variable effects in IAV infection. We believe these findings will aid in the development of LL-37 derivatives to expand the repertoire of antimicrobials.

Background: International analyses suggest that routine maternal vaccination with seasonal trivalent influenza vaccine is cost-effective, but few studies have been done in middle- to low- income countries. Method: A decision-tree analysis was modelled for the South African public healthcare setting over one year from a payer's perspective. Direct medical costs and consequences were obtained from published literature. Incremental cost effectiveness ratios (ICERs) and univariate sensitivity analyses were then measured. Discounting was excluded due to the seasonality of influenza, limiting the time horizon to a one year period. Findings: The model predicted that to avert influenza-associated hospitalisations amongst pregnant women and their infants less than six months of age, vaccination of pregnant women was not cost-effective. This was irrespective of whether the universal vaccination or HIV-targeted approach was used. A base model simulating 100% vaccine uptake predicted that seasonal vaccination of 100,000 pregnant women results in an estimated net cost of R69,118,114.05 per neonatal influenza-associated hospitalisation averted. Similarly, the model suggested that vaccinating 100,000 pregnant women would cost R1,197,779.79 per maternal hospitalisation averted. Univariate sensitivity analyses reinforced that influenza vaccination of pregnant women was not cost-effective, except when lower incidence of maternal influenzaassociated hospitalisations associated with antenatal influenza vaccination were simulated where the targeted approach became dominant. The latter analysis predicted savings of R770,530.86 per maternal influenza-associated hospitalisation averted. Interpretation: The ICERs suggest that influenza vaccination amongst pregnant women is not cost-effective in the South African public healthcare sector compared to no vaccination, with respect to averting influenza-associated hospitalisations amongst pregnant women and their infants less than six months of age. However, these estimates should be re-evaluated, pending vaccine effectiveness studies of higher methodological quality for low- and middle- income countries and using cost inputs relevant to South African public healthcare setting. This analysis may provide preliminary information regarding the upscaling of influenza vaccination amongst pregnant women as a priority in the constraints of a limited healthcare budget and careful consideration is required regarding vaccine mobilisation amongst pregnant women. / Dissertation (MSc)--University of Pretoria, 2016. / School of Health Systems and Public Health (SHSPH) / MSc / Unrestricted

Over the past decade the widespread proliferation of mobile devices and wearable technology has significantly changed the landscape of epidemiological data gathering and evolved into a field known as Digital Epidemiology. One source of active digital data collection is online participatory syndromic surveillance systems. These systems actively engage the general public in reporting health-related information and provide timely information about disease trends within the community. This dissertation comprehensively addresses how researchers can effectively use this type of data to answer questions about Influenza-like Illness (ILI) disease burden in the general population. We assess the representativeness and reporting habits of volunteers for these systems and use this information to develop statistically rigorous methods that adjust for potential biases. Specifically, we evaluate how different missing data methods, such as complete case and multiple imputation models, affect estimates of ILI disease burden using both simulated data as well as data from the Australian system, Flutracking.net. We then extend these methods to data from the American system, Flu Near You, which has different patterns. Finally, we provide examples of how this data has been used to answer questions about ILI in the general community and promote better understanding of disease surveillance and data literacy among volunteers.

Influenza causes substantial morbidity and mortality yearly in both temperate and tropical regions, as well as sporadic and potentially severe pandemics. Although vaccines for seasonal influenza exist, most options for controlling influenza outbreaks are reactive in nature. Sufficiently accurate and well-calibrated forecasts, on the other hand, could allow public health practitioners, medical professionals, and the public to respond to unfolding influenza outbreaks proactively. For example, hospitals could prepare additional beds for a predicted surge, and public health experts could redouble vaccination efforts. Recently, skillful forecasts have been developed for a range of infectious diseases, including influenza, but this work has been limited to only a few countries. In this dissertation, we explore the potential for generating accurate influenza forecasts using a publicly-available dataset of country-level epidemiologic and virologic surveillance data. In Chapter 2, we use a combined model-inference system to generate retrospective forecasts for 64 countries in both temperate and tropical climates. We show that forecast accuracy is significantly better in countries with temperate climates, and that inclusion of environmental forcing, specifically modulation of viral transmissibility due to variability of absolute humidity conditions, also improves forecast accuracy in temperate climates. In Chapter 3, we develop a metapopulation model of twelve European countries using data on international air travel and commuting. We find that this model is unable to produce more skillful forecasts than those produced for individual countries in isolation. We make recommendations for improvements in data collection and reporting that may increase the success of similar modeling efforts in the future. In Chapter 4, we assess the performance of real-time forecasts generated for 37 countries over two influenza seasons and discuss the potential for their use in public health decision making. Finally, in Chapter 5 we describe the results of a small survey of public health practitioners in the United States. We find that the majority of respondents desire more effective communication between modelers and public health practitioners, and we discuss the importance of regular and improved communication in advancing the practical use of forecasts as public health decision making tools. This dissertation advances the science of influenza forecasting by demonstrating that skillful retrospective and real-time forecasts can be generated for many countries where previous forecasting efforts are either minimal or absent. However, it is vital that data quality issues be addressed if further progress is to be made. Future work should focus in particular on climatic drivers of influenza in the tropics and subtropics, on the role of human travel at various spatial scales, and on the development of regional and local forecasting capacity. Additionally, dedicated collaboration between modelers and public health practitioners will be instrumental for motivating and informing the use of forecasts in combating influenza outbreaks.

Highly pathogenic notifiable avian influenza (HPNAI) is a disease caused by influenza A virus. It is frequently fatal in poultry. Since late 2003, disease outbreaks caused by the Asian strains of HPAI H5N1 virus have ravaged the poultry industry with the death of over 200 million birds. The epidemic has spread from Asia to Europe and more recently to Africa. To date, more than 200 human fatalities have occurred. A clear understanding of the full epidemiology of the disease at the genetic and spatial/temporal level is critical for the management, control and eventual eradication of the virus. In this study, modern tools of molecular epidemiology (Reverse-transcriptase polymerase chain reaction (RT-PCR), molecular characterization and phylogenetic analyses), Geographical Information Systems (GIS) and remote sensing, and other epidemiological tools were used to explore the outbreak of HPNAI in Nigeria. The molecular and spatial analyses both concluded that Nigeria was infected with multiple infections. The spread of primary outbreaks, which affected mainly sectors 2 and 3 of the poultry industry as described by Food and Agricultural Organisation of the United Nations, were strongly linked to trades, live bird markets, inappropriate disposal of carcasses and poorly implemented control measures. This work did not find a strong correlation between wild birds and HPNAI H5N1 in Nigeria. Some of the analyzed viruses showed genetic drift, and the implications of these for future epidemiology and ecology of avian influenza in Africa will need further evaluation. The option of vaccination and its implications were adjudged good, and its shortcomings were highlighted. Community initiative at fighting emergency diseases like HPNAI H5N1 was similarly advocated. The financial losses to the Nigerian poultry industry were estimated at around $680 million. The risk of the spread of infection was assessed using ecological niche modeling and the whole of West Africa is at risk of infection, should no concrete action be taken to halt the spread. In conclusion, useful suggestions were proffered to affected countries like Nigeria, and unaffected countries that are at risk of infection, so that Africa can be safe from the scourge of HPNAI H5N1. / Dissertation (MSc)--University of Pretoria, 2008. / Production Animal Studies / unrestricted

The H9N2 low pathogenic avian influenza virus (LPAIV) is considered a prime candidate for evolving into a highly pathogenic AIV (HPAIV) following circulation in domestic poultry, due to its widespread endemism in Eurasia and the donation of internal genes to H5 and H7 HPAIV. In this study, we investigated if a LP H9N2 virus could acquire increased pathogenicity following nineteen serial passages in week-old chickens. We show that the H9N2 virus remained low pathogenic in chickens after nineteen serial passages, evidenced by a lack of mortality in chickens despite efficient viral replication in chicken organs. An overregulation of cytokine response to infection, typical of highly pathogenic viruses, was not observed in tissue organs analyzed by quantitative PCR. The serial passaging of the H9N2 virus led to reduced virulence in chicken embryos and reduction in Hemagglutinin heat stability after passaging. The positive selection of Leucine at amino acid position 226 (L226) after 19 passages suggests the H9N2 can gain mammalian adaptation markers following circulation in domestic poultry. In addition, we carried out four experiments to determine the effectiveness of aerosol exposure of H6N1, H10N7, H10N8 and H13N6 LPAIV’s in causing infection, immune dysregulation and mortality in chickens compared to intranasal and oral inoculation routes. From our study, we observed mortality of chickens exposed to H6N1 and H10N8 viruses via aerosols. Aerosol exposure also resulted in more efficient replication in the respiratory tracts of chickens than intranasal or oral inoculated chickens. In addition, overexpression of pro-inflammatory (IL-6, IL-1ß) and antiviral (INF- γ) cytokines was observed in chickens exposed to aerosols compared to intranasal and oral inoculation. Our results show that the aerosol route of exposure is efficient at causing infection in chickens and should be factored into control and prevention strategies against AIV.

To compare the efficacy and duration of the immune response to local and systemic vaccination, Balb/c mice were vaccinated either intraperitoneally (i.p.) with an inactivated A/PR/8/34 (H1N1) vaccine or intranasally (i.n.) with a vaccinia recombinant containing the H1 gene of influenza. The i.p. inactivated vaccine stimulated high serum IgG anti-influenza titres and protected the lungs against viral challenge for the duration of the experiment (17 months). Little nasal wash IgA was induced and the noses were susceptible to challenge. Animals vaccinated i.n. with the recombinant had lower serum IgG titres and the lungs showed poor protection against challenge. Nasal wash IgA titres were higher, however, and the noses were largely protected from viral challenge for 17 months.

While memory CD4 T cells are critical for effective immunity to pathogens, the mechanisms underlying their generation are poorly defined. Although extensive work has been done to examine the role of antigen (Ag) in shaping memory formation, most studies focus on the requirements during the first few days of the response known as the priming phase. Little is known about whether or not Ag re-encounter by effector T cells (late Ag) alters CD4 memory T cell formation. Since influenza infection produces a large, heterogeneous, protective CD4 memory T cell population, I used this model to examine the role of late Ag in promoting CD4 memory T cell formation. In the experiments presented in this thesis, I demonstrate that late Ag is required to rescue responding CD4 T cells from default apoptosis and to program the transition to long-lived memory. Responding cells that failed to re-encounter Ag had decreased memory marker expression and failed to produce multiple cytokines upon re-stimulation. Ag recognition is required at a defined stage, as short-term Ag presentation provided 6 days after infection is able to restore canonical memory formation even in the absence of viral infection. Finally, I find that memory CD4 T cell formation following cold-adapted influenza vaccination is boosted when Ag is administered at this stage. These findings imply that persistence of viral Ag presentation into the effector phase is the key factor that determines the efficiency of memory generation. They also suggest that administering Ag during the effector stage may improve vaccine efficacy.

The influenza A virus is responsible for 250,000 to 500,000 deaths every year worldwide and millions more could die in the event of a serious pandemic. Vaccines against influenza have existed for long, but until today they have been limited by extensive production times and reduced cross-protection between different strains of the virus. This leads to a recurrent need to update the vaccine composition every year, which is both costly and inadequate to fight pandemics. An innovative approach that could improve the vaccine efficacy has been recently developed based on the selection of conserved influenza epitopes with potential to induce broader immune responses. The 23-amino acid extracellular domain of the M2 protein (M2e) is highly conserved among different influenza A strains and thus it seems like an ideal candidate for a universal influenza vaccine. However, due to its small size, it is a poor immunogen when used on its own. The aim of this project was to produce M2e-presenting virus-like particles (VLPs) in Nicotiana benthamiana plants via Agrobacterium-mediated transient expression. Plants are increasingly being examined as alternative recombinant protein expression systems due to their safety, scalability and rapid production times. Moreover, numerous studies suggest the use of recombinant virus-like particles (VLPs) to increase the immunogenicity of antigens. Therefore, to obtain VLPs presenting the M2e epitope, I genetically engineered several different M2e-HA fusion proteins by replacing the hemagglutinin (HA) globular head and main epitope with five tandem repeats of M2e epitope sequences (5xM2e) from human, swine, and avian origin influenza A viruses. To increase the chances of obtaining VLPs, M2e-HA fusions either contained the HA stalk domain (5xM2e-HAstalk) or simply the transmembrane region (5xM2e-HAtrans). Furthermore, the tetramerizing leucine zipper derived from the General Control Protein (GCN4) was also included in some of the constructs to promote particle formation. In total, six different M2e-HA fusions were created: 5xM2e-GCN4-HAstalk, 5xM2e-GCN4-HAtrans, 5xM2e-HAstalk, 5xM2e-HAtrans, 1xM2e-HAstalk and 1xM2e-HAtrans. The expression of these proteins was optimized in plants by testing different conditions and using three different expression vectors. Overall, I was able to show expression after only 3 days post-infiltration for most of the M2e-HA v fusion proteins utilizing the pEAQ-HT and pRIC 3.0 expression vectors whereas expression levels with pTRAc were low or non-detectable. Once the expression of the M2e-HA fusions was optimized, the two proteins with the highest potential to form VLPs were selected for further characterization (5xM2e-HAstalk and 5xM2eHAtrans). Using transmission electron microscopy to analyse purified proteins, both 5xM2eHAstalk and 5xM2e-HAtrans were shown to assemble into VLPs resembling the shape and size of native HA VLPs. These VLPs could also be observed in the apoplastic fractions of infiltrated leaves. However, due to the low number of particles observed, the successful incorporation of the M2e peptide on the surface of the particles was inconclusive, as shown by M2e-specific immuno-gold labelling experiments. Furthermore, contrarily to previous studies, co-expression of the M2e-HA fusions with the M1 protein resulted in a decrease in recombinant protein accumulation and VLP formation in our plant system. A possible inhibition mechanism by the M1 protein is discussed. In summary, this research provides preliminary data to produce universal influenza vaccines in plants. I report here for the first time that M2e fused to either the stalk or transmembrane domain of the HA protein, can self-assemble into VLPs without any other proteins, in N. benthamiana plants. Future work on the immunogenicity of the VLPs produced in this study is required to confirm their potential as a universal influenza vaccine that can be rapidly produced.

Seasonal influenza produces substantial disease within the United States every year. Despite the availability of safe and effective vaccines for influenza, millions of individuals go unvaccinated each flu season, with notable differences across racial/ethnic groups. Using the Behavioral Risk Factor Surveillance System (BRFSS), I examine vaccination rates among non-Hispanic whites and Hispanics during the 2000-2009 influenza seasons. After developing a new method that addresses shortcomings of BRFSS vaccination measures, I find that non-Hispanic whites exhibit higher vaccination rates than Hispanics. Through a series of logistic regression models I show that the disparities between non-Hispanic whites and Hispanics narrow after controlling for healthcare coverage and socioeconomic characteristics. This suggests that seasonal influenza vaccination may be improved among U.S. Hispanics by addressing structural barriers in receiving the vaccine, especially access to health care.