A Novel and Sensitive Molecular Method for Nucleic Acid Discovery in CSF Samples

UoM administered thesis: Phd

  • Authors:
  • Sana Alshaikh


Encephalitis is a matter for serious public health concern because of the high morbidity and mortality associated with many cases. Epidemiological studies have shown that viral encephalitis (VE) is more common than the sum of encephalitis caused by all other pathogens. However, more than 95% of cases have no known cause. Thus, there is a significant need to develop a sensitive method for the diagnosis of these unknown cases. Previous sequence independent amplification (SIA) assays have proved successful in detecting new viruses in many biological samples but not in CSF samples. This may be due to the relatively low sensitivity of most available methods as CSF usually contains lower concentrations of pathogen than most other samples. A known problem with these types of assays is the annealing of the random primers to human DNA which facilitates preferential amplification of background human DNA. Thus, large scale sequencing is usually required to detect a virus, which in turn reduces the detection sensitivity to more than 1000 viral copies/µl, a CSF concentration that is rarely seen in cases of VE.This project was designed to develop a highly sensitive SIA assay for novel nucleic acid identification that could be used in testing CSF samples obtained from patients with neurological diseases of unknown cause.The study started with evaluation of two existing SIA assays commonly used for virus discovery; whole genome amplification (WGA) and random PCR (r-PCR). Sequential modification and adaptation of these methods was carried out to increase their sensitivity. Ultimately, a novel primer (Sa primer) that showed no binding to most human DNA sequences in GenBank was designed and synthesised. Its 3' end was tagged with 6 and 7 random nucleotides generating 2 r-primers; Sa-6 and Sa-7. The sensitivity of the r-primers was checked in a novel assay developed during this project and named Sa-SIA using known concentrations of HCMV and HSV-1. CSF samples from Malawian children were then tested using the developed assay.Results showed that adaptation of the existing WGA and r-PCR assays allowed detection of up to 1300 viral copies/µl. When the novel primers developed in this project were used in a random PCR assay (Sa-r-PCR), it was found that using Sa-6 primer 130, 13, and 1.3 HCMV copies/µl could be detected with 100, 60, and 50% efficiency respectively. When using Sa-7 primer, the same concentrations of virus were detected with 100, 42, and 28.6% efficiency. DNase-1 treatment of the samples pre-extraction resulted in an improvement in viral detection sensitivity in samples with a high background of host DNA. Starting with template concentrations of 11000, 110, 11, and 1.1 HSV-1 copies/µl, viral detection efficiency was increased from 33.3, 10, 0, and 0% to 92, 55.6, 16.7, and 0% respectively when pre-extraction DNase-1 treatment preceded Sa-r-PCR using Sa-6 primer. The final developed assay (Sa-SIA) consisted of centrifugation, DNase-1 treatment, DNA extraction, Sa-r-PCR using Sa-6 and Sa primers, gel electrophoresis, band excision, cloning, small scale sequencing (sequencing of smaller or equal to 20 positive clones from one constructed DNA library), and bioinformatics. It had a detection sensitivity of 1.3-11 viral copies/µl. When this assay was applied to stored CSF samples, one 448bp sequence was identified which gave 96% coverage with 81% identity to Torque teno midi virus-1 and 93% coverage with 81% identity to small anellovirus-2. A 236bp sequence from another CSF sample showed 66% coverage with 97% homology to an unclassified sequence previously identified in a viral genomic survey of stool sample in an earlier published study.In conclusion, the standardised method had been shown to detect 1.3 to 11 viral copies/µl of two viruses; HCMV and HSV-1. The detection of these viruses was achieved with only small scale sequencing. Application of this method to CSF samples has shown promising results. However, this method could be followed by more advanced post amplification analyses such as next generation sequencing.


Original languageEnglish
Awarding Institution
Award date1 Aug 2012