Document Type Master's Dissertation Author Stewart, Katherine Anne URN etd-04222008-145431 Document Title Strain differentiation of Citrus tristeza virus isolates from South Africa by PCR and microarray Degree MSc (Microbiology) Department Microbiology Supervisor
Advisor Name Title Prof G Pietersen Supervisor Keywords
- Citrus tristeza virus
- amino acid sequences
Date 2007-09-06 Availability unrestricted AbstractThe aim of this study was to characterize strains used in the cross-protection scheme in South Africa by establishing Polymerase Chain Reaction (PCR) systems aimed at differentiating the strains by targeting the conserved p23 gene and the variable 5' half of the Citrus tristeza virus (CTV) genome. Two cross-protecting sources GFMS 12 and GFMS 35; and eight single aphid sub-isolates were tested and classified into strain types or genotypes. An oligonucleotide microarray system was developed to differentiate T30 and T36 strains of CTV. The establishment and development of such tests will enable the South African Citrus Industry to better select mild strains for cross-protection and determine which strains are present in citrus growing areas so as to better understand the dynamics of the disease.
The first aim was to characterise the p23 gene of possible mild-strain cross-protection isolates in South Africa (RSA) and compare them to known isolates worldwide. Isolates were amplified with bi-directional RT-PCR, sequenced and phylogenetic analysis performed. The predicted amino acid sequences were compared for areas of possible variability for further strain differentiation. A bi-directional PCR developed by Sambade et al. (2003) was established that targets differences in amino acid positions 78-80 of the p23 gene and allows discrimination of isolates into mild, atypical and severe groups. The group designations of RSA isolates 390-3 and 390-5 were atypical; 390-4, 389-4 and 389-3 were mild; GFMS 35 had mild and atypical isolates; GFMS12, 12-7 and 12-9 had mild and severe isolates and; 12-5 was severe. The three main clusters on the phylogenetic tree confirmed the group designations of these isolates. Isolates in the atypical group were more diverse than ones in the mild or severe groups. There were 53 polymorphic sites within the amino acid sequences of p23 gene of the RSA and reference isolates, of which 4 distinct regions showed variability. The amino acid region 78-80 was confirmed as being very useful in grouping these isolates as mild, severe or atypical. The PCR system was robust, reproducible and has potential in the RSA Citrus industry as a screening tool in selecting mild strains for cross-protection and in detecting mixed strains in isolates.
The secondary aim was to establish the 23 primer pair PCR system developed by Hilf et al. (2000) to differentiate isolates as T36, T30, VT or T3 genotypes. Each isolate was tested with RT-PCR using 23 individually optimised genotype specific primer sets (Hilf et al., 2000). The most common genotype detected was T30 and the least common was T3. The GFMS 35, T30 plant and 389-3 isolates had a homogenous T30 genotype profile and isolate 12-5 had a VT genotype profile. The 389-4, 390-3 and 390-4 isolates had a predominantly T30 genotype profile and isolates 12-7 had a predominantly VT genotype profile. Isolate GFMS 12 had a mixed genotype profile indicative of a mixed infection while isolates 390-5 and 12-9 appeared to have mixed genotypes of VT, T30 and T36. Isolates 390-3 390-4 and 390-5 had no amplification within regions 4-7 and appear to be highly variable isolates or possible recombinants. The T3 genotype specific markers were found in region 2 of a few isolates and could be a cross-reacting primer set to the T3o genotype. It is useful for homogenous strains in determining the genotypes, molecular marker information, possible variability or recombination and for approving isolates for mild strain cross-protection. Potential drawbacks of the system include non-amplification of regions; cross-reacting primers; difficulty in optimising; and secondary structures. It was difficult to objectively draw conclusions if an isolated had mixed genotypes since mixed genotype amplifications were not consistently found in all regions targeted.
The third aim was to develop an oligonucleotide (oligo) microarray system to differentiate mild T30 and severe T36 strains. The 5' half of the CTV genome was Cy3 5'-end labelled and amplified. Oligos were designed to be T36-strain specific with a Tm above 60 °C and if possible a GC content above 65 %; and differed in amount and position of mismatches to strain T30. A standard operating protocol was set up by testing different labelling methods, hybridization mixes and washing steps. The array was tested using individual T30 and T36 strains as templates at 42, 52 and 60 °C. Experimental variation was quantified and normalised. The secondary structures of the hybridizing amplicons were determined by mfold (Zuker et al., 2003). Some oligos were specific at 42 °C and others at 52 °C. The hybridization allowed a clear differentiation of strain T36 with 13 of the T36-specific oligos at their optimal hybridization temperature. A few oligonucleotides showed cross-hybridization to strain T30 and were not used in further analysis. Oligonucleotides with 21 % or more mismatches were successful oligos, whereas ones that had 18 % or fewer mismatches had cross-hybridization. Some oligos were modified to include Locked Nucleic Acid (LNA) instead of DNA in an attempt to increase specificity with two of them having increased specificity compared to the unmodified DNA oligonucleotides. The successful differentiation by hybridization to strain specific oligos opens paths for highly parallel, yet specific assays for strain differentiation of CTV strains and a more thorough insight into the future strains circulating in RSA.
© University of Pretoria 2006
Filename Size Approximate Download Time (Hours:Minutes:Seconds)
28.8 Modem 56K Modem ISDN (64 Kb) ISDN (128 Kb) Higher-speed Access 00dissertation.pdf 5.32 Mb 00:24:37 00:12:39 00:11:04 00:05:32 00:00:28 01article.pdf 356.60 Kb 00:01:39 00:00:50 00:00:44 00:00:22 00:00:01