No description available.

Pertussis toxin (PT) is a major virulence factor produced by Bordetella pertussis. In its detoxified form (PTd), it is an important component of acellular pertussis vaccines although some residual FT activity may be present because of the limitations of the detoxification processes used. The in vivo histamine sensitisation test (HIST) in mice is currently used for the safety testing of these vaccines to determine the level of their residual FT activity. However, an alternative test is needed because of large assay variability and ethical concerns with regard to animal usage. The main objective of this study was to search for an alternative test to the HIST. The ADP-ribosylation enzyme activity of FT is thought to be the major factor responsible for the histamine-sensitising activity detected in vivo. In the present study, the enzymatic activities in different acellular pertussis-based combination vaccine formulations were measured by a recently-developed ADF-ribosylation assay and compared with their reactivities in the HIST. The results indicated that different products showed differences in ADP-ribosylation activity and, these did not correlate with their reactivity in the HIST. FT has two functionally-distinct domains: the enzymatic A-protomer and the B- oligomer that facilitates host-cell binding and entry of FT into the cell. This dual biological function could explain why the residual enzyme activity of FT in vaccines did not fully reflect the in vivo reactivity observed by the HIST. Thus, refinement of the in vitro test to include a step which monitored the B-subunit activity of FT was attempted. A quantitative FT carbohydrate-binding assay using glycoproteins or defined oligosaccharides was developed. PT was found to bind preferentially to multiantennary N-glycans, with the highest binding towards the fully sialylated structures. In contrast, PTd lost the ability to bind to sialylated multiantennary structures but retained some capacity to bind to neutral multiantennary structures. Different vaccine preparations had different levels of PT binding activity as well as enzymatic activity. It was concluded that, although the enzymatic activity of PT plays a more important role in the death of mice in the HIST, a high binding activity of the B- subunit could increase the in vivo toxic effect by aiding the accessibility of the A- subunit to its cellular targets. A mathematical equation was devised to establish a preliminary relationship between the enzymatic, carbohydrate-binding and HIST assays in a product-dependent manner. Further studies with a larger number of vaccines are required for a more meaningful statistical analysis. However the methods form a sound basis for the future development of an alternative assay to the histamine challenge test. The in vitro assays could also be useful for investigating the mechanisms of PT detoxification. Comparisons of A- and B-subunit activities of purified PT and vaccine preparations of PTd indicated that both subunits are modified after chemical detoxification. Different vaccine products had different levels of enzymatic and binding activities and it was concluded that different detoxification procedures, as well as formulation factors, could contribute to this variation. A CHO cell clustering assay is used as an alternative in vitro test to the HIST for assessing residual PT activity at the bulk stage of vaccine production. In a parallel study to the above, comparative proteomics was used to gain insights into the mechanism of PT-induced CHO cell clustering with a view to developing a mechanistic-based alternative assay for the safety testing of pertussis-based combination vaccines. A proteomic map of CHO cells was established and PT-induced CHO cell clustering appeared to be a complex process involving subtle changes in various cellular functions, mainly related to intracellular transport, cell stress and the cell cycle. The information obtained will be useful for future studies into the possible mechanisms of the effect of PT on CHO cells.

Adenylate cyclase toxin (CyaA) toxin is an important virulence factor of Bordetella pertussis, the causative agent of whooping cough, and a potential component of acellular pertussis vaccine. The work involved the production of three purified forms of CyaA with different enzymic and invasive properties. These were: the native enzymatically-active, invasive toxin (CyaA), an invasive derivative lacking AC enzymic activity (CyaA*) and a non-acylated, non-invasive form of CyaA (proCyaA). These were expressed in E. coli BL21/DE3 as recombinant proteins. After purification by a combination of chromatographic methods (Q-and Butyl-Sepharose) their properties were investigated by several assays. The LPS content of these preparations was very low. The AC enzymic activity was assayed by a conductimetric method. CyaA and proCyaA had a high level of enzymic activity but that of CyaA* was very low. The cytotoxic properties of CyaA* and pro-CyaA towards J774.2 cells were similar and were 100-fold less than that of CyaA, but these differences were shown to some, extent to be Ca2+ -dependent for CyaA*. Caspase 3/7 activities were measured over a range of toxin concentrations. At these concentrations, neither urea buffer alone nor CyaA* induced any significant increase in caspase 3/7 from different mammalian cells. The greatest effect of CyaA was observed on J774.2 and RBL-2H3 cells where increasing concentration of toxin gave increasing activity. Different concentrations of CyaA and CyaA* were used to investigate dose-dependent effects of the toxins on phagocytosis and the oxidative burst in U937 human monoblastic cells, J774.2 mouse macrophagelike cells and fresh human granulocyte cells (whole blood used). Significant effects were seen with CyaA on both phagocytosis and oxidative burst, but CyaA* did not have a significant effect on either. The results of this first part of the study showed that both enzymatic and invasive functions are required for the cytotoxic effects of adenylate cyclase toxin. In the second part of the work, CyaA was formulated as protein-coated microcrystals (PCMCs) on the surface of microcrystals of DL-valine. The aims of this formulation were to remove the urea, normally used to stabilize the protein, and to determine the stability of the enzymic and cytotoxic activities of the protein in this form, as a dry powder and when the PCMCs were redissolved in aqueous solution. The CyaA in the PCMCs was shown not to be readily soluble in aqueous buffers, but could be resolubilised in urea buffer and retained high AC and cytotoxic activity. Many different types of PCMC formulation were prepared in attempts to increase solubility of PCMCs in aqueous solution. PCMCs were made with CyaA coprecipitated with different combination of CaM, BSA, CaCl2 or ATP and crystals were dissolved in different buffers at various pHs. The most promising results were obtained with CyaA-CaM- BSA-PCMCs where the highest levels of both AC enzymic and cytotoxic activities were seen when the PCMCs were dissolved in l00mM Bicine (pH 8). CaM alone preserved only the AC enzymic activity of PCMCs when coprecipitated with CyaA and DL-valine. AC and cytotoxic activities of CyaA were stable in PCMCs for up to a week at 37°C. CyaA-CaM-BSA-PCMCs induced a strong serum IgG response to CyaA and BSA when injected subcutaneously into mice. The results indicated that CyaA-CaM-BSA- PCMCs offer a promising way to preserve the activity and antigenicity of CyaA in a non-aqueous formulation. Neutralisation tests proved that the antibody generated from mice immunised with CyaA-BSA-CaM-PCMCs was able to neutralise the AC enzymic and cytotoxic activities. Such PCMCs could have application for presentation of protein antigens that normally require cold storage for stability.