An investigation of the inhibition of non-typeable Haemophilus influenzae, by substances secreted by Haemophilus haemolyticus

Non-typeable Haemophilus influenzae (NTHi) are Gram-negative bacteria that reside in the upper respiratory tract of humans. As well as being part of the normal flora, they are opportunistic pathogens, causative of non-invasive respiratory tract infections, and an occasional cause of invasive infections. In the Australian aboriginal population, a high prevalence of otitis media (OM) in children is a factor that contributes significantly to their educational and economic disadvantage, and NTHi is the most important cause of recurrent OM cases. In the worldwide adult population, NTHi causes approximately 25% of the “exacerbations” of chronic obstructive pulmonary disease (COPD), which ultimately lead to death in most cases. NTHi is transmitted by respiratory secretions, initially colonising the pharynx, and in the case of disease, moves to the lower respiratory tract (COPD) or middle ear (OM). Efforts to treat and prevent NTHi infection have been hampered by antibiotic resistance and difficulty in finding suitable vaccine targets, and alternative treatment or prevention strategies are therefore desired. Bacterial therapy (probiotics) for the prevention of respiratory tract infections, based on co-colonizing bacteria, has been a topic of research in the 21st century, and a commercial preparation of Streptococcus salivarius (BLIS K12\(^{TM}\)) is now available for the prevention of S. pyogenes infections, and another S. salivarius preparation (24SMB) is patented for prevention of S. pneumoniae infections. However, there have so far been no co-colonizing probiotics developed specifically for the prevention of NTHi infections. Haemophilus haemolyticus are bacteria that are closely related to NTHi, also colonize the upper respiratory tract of humans, but are not opportunistic pathogens of the respiratory tract, and only very rarely causative of invasive infections in the immunocompromised host. Recently, pre-treatment of epithelial cells with H. haemolyticus was shown to prevent NTHi attachment, and as a result, H. haemolyticus was suggested as a possible agent of bacterial therapy (Pickering et al., 2016). In this project, the NTHi-inhibitory activity of H. haemolyticus was investigated, with a focus on secreted substances. Initially, SYBR Green real-time (RTPCR) assays were developed for the identification of the two species, which were then applied to a collection of isolates from Australian pathology laboratories. The H. haemolyticus isolates were then screened for inhibition of NTHi, using co-culture and agar well diffusion assays. Two H. haemolyticus isolates (BW1 and RHH122) were identified as consistent producers of an NTHi inhibitory substance (NIS), and the physical and chemical properties of the NIS produced by BW1 were characterised and published. The NIS produced by BW1 (BW1-NIS) and RHH122 (RHH122-NIS) were identified, by mass spectrometry analysis and comparison with published whole genome sequences, as identical to a putative 27 kDa OMPA-like protein in H. haemolyticus isolate M19107. A recombinant version of this H. haemolyticus NIS (Hh-NIS), made in E. coli from the Hh-NIS ORF without signal peptide, also inhibited NTHi. With in silico structural modelling, similarities were identified between Hh-NIS and the haemoglobin-haptoglobin utilisation protein A (hpuA) in Kingella denitrificans and Neisseria gonohorreae. The spectrum of absorbance of Hh-NIS identified haem as a likely component, and with cation exchange chromatography, Hh-NIS was separated into a holoprotein fraction, containing haem as a ligand, and an apoprotein fraction, from which haem was absent. As the activity of both fractions against NTHi was similar, and the addition of haemin neutralised the activity of Hh-NIS, haem was considered not to be a co-factor required for activation of an enzyme, but that it had another role. The addition of purified Hh-NIS holoprotein to broth cultures, resulted in inhibition of NTHi growth, and promotion of growth of an Hh-NIS producing strain of H. haemolyticus (BW1), but only in haemin-limited conditions. A model mechanism for activity was described, whereby Hh-NIS is a haemophore that binds haem, making it available to producing strains, but unavailable to NTHi. To further understand the biological function and importance of Hh-NIS to H. haemolyticus, construction and testing of a knockout mutant, the elucidation of the control of Hh-NIS secretion and utilisation, and in vivo testing in animal models are recommended. With regards to its application as a probiotic for NTHi control in humans, in the continued absence of an effective vaccine, in vivo testing of H. haemolyticus strains which produce Hh-NIS is recommended, although as haem-acquisition genes are associated with pathogenicity, a cautionary approach should be taken