Untargeted analysis of the airway proteomes of children with respiratory infections using mass spectrometry based proteomics

The upper airway – which consists mainly of the naso- and oro-pharynx - is the first point of contact between the respiratory system and microbial organisms that are ubiquitous in the environment. It has evolved highly specialised functions to address these constant threats whilst facilitating seamless respiratory exchange with the lower respiratory tract. Dysregulation of its critical homeostatic and defence functions can lead to ingress of pathogens into the lower respiratory tract, potentially leading to serious illness. Systems-wide proteomic tools may facilitate a better understanding of mechanisms in the upper airways in health and disease. In this study, we aimed to develop a mass spectrometry based proteomics method for characterizing the upper airways proteome. Naso- and oropharyngeal swab samples used in all our experiments had been eluted in the Universal Transport Media (UTM) containing significantly high levels of bovine serum albumin. Our proteomic experiments tested the optimal approach to characterize airway proteome on swab samples eluted in UTM based on the number of proteins identified without BSA depletion (Total proteome: Protocol A) and with its depletion using a commercial kit; Allprep, Qiagen (cellular proteome: Protocol B, Ci, and Cii). Observations and lessons drawn from protocol A, fed into the design and implementation of protocol B, and from B to protocol Ci and finally Cii. Label free proteome quantification was used in Protocol A (n = 6) and B (n = 4) while commercial TMT 10plex reagents were used for protocols Ci and ii (n = 83). Protocols Ci and ii were carried out under similar conditions except for the elution gradient: 3 h and 6 h respectively. Swab samples tested in this study were from infants and children with and without upper respiratory tract infections from Kilifi County Hospital on the Kenyan Coast. Protocol A had the least number of proteins identified (215) while B produced the highest number of protein identifications (2396). When Protocol B was modified through sample multiplexing with TMT to enable higher throughput (Protocol Ci), the number of protein identified reduced to 1432. Modification of protocol Ci by increasing the peptide elution time generated Protocol Cii that substantially increased the number of proteins identified to 1875. The coefficient of variation among the TMT runs in Protocol Cii was