Now showing 1 - 1 of 1
ItemThe role of infection in the aetiology of prostate cancerYow, Melissa-Ann ( 2015)An infectious aetiology for prostate cancer has been conjectured for decades but the evidence gained from questionnaire-based and sero-epidemiological studies is weak and inconsistent and a causal association with any infectious agent remains to be established. The work described in this thesis questions whether the inconsistency in evidence could be related to tumour heterogeneity (high-grade or low-grade) or the nature of infections. It was decided to focus on high-grade disease, as this the most aggressive form, and on evidence of persistent infection, as transient infection was considered unlikely to play a causal role. Quantitative molecular methods were used first to seek evidence of single-organism infection comparing high-grade with lower-grade tumours. The potential of targeted 16S rRNA gene sequencing and total RNA sequencing was then evaluated regarding its utility to characterise microbial communities within high-grade tumours. Archival tissue blocks were retrieved for 52 high-grade and 76 low-grade prostate cancers. DNA samples were extracted and screened by RealTime-PCR using validated and standardised assays for the following candidate organisms C. trachomatis, U. urealyticum, U. parvum, M. genitalium, BKV, HSV-1 and 2, P. acnes types IA, IB and II, T. vaginalis and XMRV DNA. Samples were screened for 16 HPV genotypes by PCR and flow cytometry. Prevalence of M. genitalium, U. urealyticum, and HSV was low and did not differ by tumour grade. The prevalence of P. acnes type IA, IB, II was higher than for other agents, however no evidence of an association was detected. Neither BKV, XMRV, T. vaginalis, U. parvum, C. trachomatis nor HPV DNA were detected. Given the sensitivity and specificity of the methods used for the candidate organisms, the absence or low levels of detectable microbial DNA indicate a low probability that candidates contributed to cancer risk. A massively parallel sequencing (MPS) approach was used to characterise any resident microbial communities and the relative abundance of bacterial constituents within tissue. DNA and RNA were extracted from 20 snap-frozen tissue samples from high-grade prostate tumours (10 high-grade prostate cancer cores and matched unaffected prostate tissue). Prior to sequencing, broad-range PCR was applied to DNA across three hypervariable regions (V2-V4) of the 16S rRNA gene to enrich for bacterial species. As 16rRNA gene sequencing is only able to detect bacterial/archael microorganisms, MPS was also applied to cDNA from total RNA that was extracted from the same tissue samples to detect other microorganisms (viral, bacterial and protozoal origin) that may be associated with prostate cancer. Sample cDNA was sequenced and the data were queried for 16S rRNA sequences and the presence of expressed viral genes. Partial 16S rRNA sequencing identified Enterobacteriaceae species common to all samples and P. acnes in 95%. Total RNA sequencing detected endogenous retroviruses that provided proof of concept. As this part of the study was exploratory, associations between the organisms identified and prostate cancer risk could not be ascertained. Further studies, specifically designed to detect associations between the disease phenotype and aetiological agents, are required.