Use this URL to cite or link to this record in EThOS:
Title: Diagnosing tuberculosis using nanomaterial-based detection of a host gene expression signature
Author: Gliddon, Harriet Dulcie
ISNI:       0000 0004 7657 1571
Awarding Body: Imperial College London
Current Institution: Imperial College London
Date of Award: 2017
Availability of Full Text:
Access from EThOS:
Access from Institution:
Tuberculosis (TB) diagnosis remains a major public health challenge. Recent gene expression studies of TB have used microarray analysis to identify host whole blood gene expression signatures comprising relatively few transcripts that are able to distinguish TB from other diseases with similar presentation (OD), and TB from latent TB infection (LTBI), regardless of HIV infection status. These gene expression signatures represent promising diagnostic biomarkers for TB. However, the feasibility of measuring gene expression at the point-of-care has been questioned due to technical and cost constraints. This work presents the identification of two reduced gene expression signatures for TB/OD and TB/LTBI, which each consist of just four transcripts, and therefore represent feasible diagnostic TB biomarkers. The validation of these TB-specific gene expression signatures was performed using RT-qPCR and digital PCR. Both platforms showed excellent agreement with the results of the microarray analysis and verified the robustness of the signatures in classifying patients with TB, LTBI or OD. In order to translate these gene expression signatures into diagnostic assays for TB, nanomaterials were explored as a means of detecting the transcripts comprising the gene expression signatures for TB. The field of nanodiagnostics, in which nanoscale phenomena are linked to the presence of specific bio-analytes, has given rise to sensors with extremely low limits of detection. Nanomaterials offer many signalling mechanisms, strong signal intensities, finely tuneable surface chemistries, extremely large surface areas and multiplexing capabilities. A nucleic acid sensor based on graphene oxide quenching of fluorescent dyes was developed that could successfully detect synthetic RNAs in a monoplex format, with nanomolar limits of detection. The use of a cholesterol moiety to anchor DNA probes to graphene oxide was investigated in order to prevent nonspecific desorption of probes in the absence of target RNA. An alternative detection system relied on the strand displacement of a quencher-labelled DNA probe by target RNA, which then modulates quantum dot (QD) fluorescence emission. QDs are nanoparticles that exhibit extraordinarily bright fluorescence emission and are ideal for multiplexing due to their broad excitation and narrow emission spectra. This sensor design was first developed in a direct detection format that allowed multiplexed, sub-picomole (nanomolar) limits of detection of synthetic target RNA in pure buffer and in the presence of purified RNA from whole blood. To improve the sensitivity of this QD-based direct detection assay, an alternative format was also designed, which relies on an isothermal amplification-assisted detection assay that can potentially give much lower limits of detection. This work develops host gene expression signatures as promising new biomarkers for TB, and the potentially powerful application of nanomaterials for their detection.
Supervisor: Levin, Michael ; Stevens, Molly Sponsor: Medical Research Council
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral