AHS disease forms and pathogenesis
African horsesickness has four clinicopathological forms: horsesickness fever, cardiac form, pulmonary form and mixed form. Horsesickness fever is characterised by a remittent mild to moderate fever and is often subclinical with no associated mortality. It occurs frequently following infection with less virulent strains of virus or when some degree of immunity exists (for example horses that have been vaccinated with or exposed to a heterologous serotype). This is the only form of disease exhibited by zebra. The cardiac (sub-acute) form is characterised by fever (lasting 3-6 days), subcutaneous oedema of the head, neck and chest and supraorbital fossa, and haemorrhages of the eyes and ventral tongue surface. Death is as a result of cardiac failure. Mortality rates with this form range from 50 to 60%. The pulmonary (peracute) form develops rapidly (1 to 3 days), with marked depression, fever, increased respiratory rate, severely laboured breathing, profuse sweating and coughing spasms. A frothy, serofibrinous fluid may exude from the nostrils. The mortality rate in horses with this form is high, exceeding 95%, and death is due to a lack of oxygen. The mixed form of AHS is the most common and is a combination of the cardiac and pulmonary forms with mortality rates between 50 and 90% (Coetzer & Guthrie, 2004; Mellor & Hamblin, 2004).
The form of disease in infected horses may be the result of a variety of factors, including the route of infection, tropism of sub-populations of virus particles, host immune status, permissivity and genetic susceptibility (Laegreid et al., 1993). However, the principal determinant of the clinical form of the disease was shown in experimentally infected naive horses to be the viral virulence phenotype (Laegreid et al., 1993). The clinical and molecular basis of AHS pathogenesis is not well understood. The key pathological features of AHS are oedema, effusion and haemorrhage, and the clinical signs are thought to develop as a result of damage to the circulatory and respiratory systems (Mellor & Hamblin, 2004). On infection of the vertebrate host initial virus multiplication occurs in local lymph nodes and the virus is then spread throughout the body via the circulatory system. This primary viraemia enables the virus to infect target organs; namely the lungs, spleen, other tissues of the lymphoid system, and various endothelial cells (EC). Replication of the virus in these tissues then leads to secondary viraemia that usually coincides with the onset of fever (Mellor & Hamblin, 2004). In experimentally infected horses, with the peracute form of disease, viral antigen is found mainly in the cardio-vascular and lymphatic systems. In these horses AHSV antigen was located primarily in EC of capillaries, and small venous and arteriolar vessels, particularly of cardiopulmonary tissues suggesting that EC are the main target of AHSV infection during the late stages of this form of the disease (Wohlsein et al., 1997). AHSV has been shown to infect pulmonary microvascular EC, causing pronounced cell swelling, discontinuity of the plasma membrane and loss of structural detail in severely affected cells (Laegreid et al., 1992a). Skowneck et al. (1995) suggest that it is the loss of the EC barrier function that results in the development of the prominent pathological features. Damage to EC and loss of integrity of intercellular junctions could result in loss of EC barrier function and the subsequent development of oedema (Laegreid et al., 1992a; Skowronek et al., 1995; Gomez-Villamandos et al., 1999). Virulence variants of AHSV have also been shown to differ in their ability to infect and damage cultured EC (Laegreid et al., 1992b).
AHSV NS1 is encoded by segment 5. NS1 is highly conserved; a comparison of NS1 sequences from AHSV serotypes 4, 6 and 9 revealed 95 to 98% amino acid homology (Maree & Huismans, 1997). The protein is synthesised in large amounts (up to 25% of the total viral proteins in BTVinfected cells) and forms the characteristic tubules that are observed in orbiviral infected cells throughout the infection cycle (Huismans & Els, 1979). NS1 contains several conserved cysteine residues that are believed to be of functional significance in forming the highly ordered helical assemblies. Of the 16 cysteine residues found in AHSV NS1 9 are conserved at the same site in BTV NS1 (Maree & Huismans, 1997). Mutational analysis of BTV NS1 revealed that both the conserved cysteines at residues 337 and 340 as well as the N- and C-termini are essential for tubule formation (Monastyrskaya et al., 1994). The exact function of NS1, and the tubular structures formed by this protein, is not yet clear but studies on BTV NS1 suggest an involvement in virus morphogenesis (Owens et al., 2004). The use of the tubular structures of NS1 as an immunogen delivery system is being investigated (Ghosh et al., 2002a; Ghosh et al., 2002b; Murphy & Roy, 2008).
AHSV NS2 is encoded by segment 8 and is highly conserved (Van Staden & Huismans, 1991). The protein forms dense granular bodies in recombinant baculovirus infected cells (Uitenweerde et al., 1995). These structures are similar to the virus inclusion bodies (VIBs) observed in the cytoplasm of AHSV infected cells. VIBs are the virus factories where viral replication and core assembly occur. NS2 is therefore believed to be involved in virus assembly by forming a matrix within the cytoplasm to which cores, viral proteins and ssRNA rapidly associate and where core assembly occurs. AHSV NS2 furthermore has ssRNA binding ability and is therefore thought to be directly involved in viral mRNA recruitment to these inclusion sites (Uitenweerde et al., 1995). BTV NS2 has been found to recognise different RNA structures and to contain several RNAbinding domains. This may be the basis for the discrimination between viral RNAs and may explain how a single copy of each RNA segment is selected for incorporation into cores during virus assembly (Fillmore et al., 2002; Butan et al., 2004; Lymperopoulos et al., 2006). NS2 is the only phosphorylated viral protein in infected cells (Devaney et al., 1988). In BTV phosphorylation of NS2 is necessary for VIB formation. De-phosphorylation of the protein is thought to result in the disassembly of VIBs allowing the release of assembled cores for the attachment of the outer capsid proteins and subsequent viral release (Modrof et al., 2005; Kar et al., 2007).
Infected Vero cell viability and protein synthesis levels
As estimating CPE can be subjective, an additional quantitative measure of cell viability was obtained using the CellTiter-BlueTM cell viability assay (Promega). The results are shown in Fig. 2.7A. AHSV-2 induced rapid cell death, with only 51±1.1% viable cells remaining at 24 h p.i. compared to AHSV-3 (81±3.6%, P < 0.01) and AHSV-4 (93±1.2%, P < 0.01), these values declined further resulting in 15±0.5%, 22±4.4% (P < 0.01) and 36±4.9% (P < 0.01) cell viability respectively at 48 h p.i.. All the reassortants, except R4-210, displayed an intermediate phenotype between the respective parental strains. R4-210 did not induce significant cell death, with 85±2.5% viable cells still remaining at 48 h p.i., which corresponded to the lack of CPE previously observed (Fig. 2.6). There was no clear link between the phenotypes of reassortant viruses and the origin of their NS3 proteins.
As an additional measure of the metabolic activity of infected cells, pulse-labelling with [35S]methionine was used to monitor total intracellular protein synthesis over an 18 h period as compared to uninfected cells (Fig. 2.7B), before cells started showing distinct CPE. Protein synthesis in AHSV-2 and AHSV-3 infected cells declined rapidly to 34.9±0.72% and 37.4±6.2% respectively of the uninfected control at 18 h p.i. In AHSV-4 infected cells protein synthesis fluctuated closer to that in uninfected Vero cells, at levels of 76.8±2.3% of the control. The values for four of the five reassortants were intermediary to those of the two parental strains. Protein synthesis levels in cells infected with R3-25,10 and R3-23,5 decreased to 35.8±0.12% and 37.9±3.4% compared to the control, and for R4-24,5,7,10 and R4-27,10 the corresponding values were 54.5±2.4% and 65±12.3%. Only in the case of R4-210 did the exchange of the single segment apparently have no effect, resulting in a profile very similar to AHSV-4 with an endpoint value of 83.3±8.8% protein synthesis. Significant correlation was found between the percentage viable cells and percentage protein synthesis (r = 0.76, p = 0.028). For each of the AHSV strains the decrease in protein synthesis in the period up to 18 h p.i. (Fig. 2.7B) corresponded to the observed subsequent decrease in cell viability as measured by the CellTiter-BlueTM cell viability assay (Fig. 2.7A).
CHAPTER 1: LITERATURE REVIEW
1.2. AFRICAN HORSESICKNESS (AHS)
1.3. AFRICAN HORSESICKNESS VIRUS (AHSV)
1.4. BLUETONGUE VIRUS NS3
1.5. ROTAVIRUS NSP4
1.6. GENOME SEGMENT REASSORTMENT
CHAPTER 2: GENOME SEGMENT REASSORTMENT IDENTIFIES NS3 AS A KEY PROTEIN IN AHSV RELEASE AND MEMBRANE PERMEABILISATION
2.2. MATERIALS AND METHODS
CHAPTER 3: COMPARISON OF THE CYTOTOXICITY AND MEMBRANE PERMEABILISING ACTIVITY OF AHSV, BTV AND EEV NS3 AND IDENTIFICATION OF DOMAINS IN AHSV NS3 THAT MEDIATE THESE ACTIVITIES
3.2. MATERIALS AND METHODS
CHAPTER 4: COMPARISON OF THE NS3 PROTEINS OF AHSV-2, AHSV-3 AND AHSV-4
4.2. MATERIALS AND METHODS
CHAPTER 5: CONCLUDING REMARKS