MOLECULAR CHARACTERIZATION OF PREVIOUSLY NON-TYPEABLE ROTAVIRUS STRAINS IN GHANA

ABSTRACT Background Rotavirus is associated with severe infantile diarrhoea requiring hospitalization. In order to control severe disease caused by rotavirus infection, vaccination is considered an essential strategy. Currently, two rotavirus vaccines (Rotarix™ and RotaTeq™) have been included in the Expanded Program on Immunization of most African countries to help reduce the high disease indections. Ghana has over the past 20 years been involved in the surveillance of rotavirus diseases by isolating rotaviruses from diarrhoeic stools and determining the strain types as part of the build up to vaccine introduction. During rotavirus burden of disease surveillance between 2007 and 2011, three hundred and one rotavirus samples could not genotype with the pool of primers recommended for use within the WHO Regional Reference Laboratory network. Samples with sufficient stools were characterized. Aim The study sought to characterize and identify primer mismatches for nontypeable rotavirus strains at the Regional Rotavirus Reference laboratory in Ghana. Methodology One hundred and eleven G and 89 P-types had sufficient stool and were characterized by sequencing methods. To confirm previous EIA results, 10% of randomly selected nontypeable samples were reanalyzed using the commercially available DAKO IDEIA kit (Dako Diagnostics, Cambridgeshire, UK). To assess the integrity of RV dsRNA, Polyacrylamide Gel Electrophoresis (PAGE) was carried out. Viral ribonucleic acids were extracted by the phenolchloroform-guanidine-isothiocyanate method and one step RT-PCR targeting the VP7 and VP4 genes of RVA was carried out using PCR primer pairs, Beg9/End9 and 9Con1/9Con2 for VP7 genes and Con2/Con3 and VP4F/VP4R for VP4 genes. The RT-PCR products were purified with the QIAquick PCR purification kit (Qiagen/Westburg), sequenced with the ABI PRISM® BigDye Terminator v.3.1 Ready Reaction mix. Amplicons were purified using ethanol-sodium acetate precipitation method. Sequences were subjected to BLAST (Basic Local Alignment Search Tool) and subsequently characterized using the web-based genotyping tool, RotaC v2.0. Results The G-types were determined for 74 out of the 111 samples that had sufficient stool to work with. The G-types detected were as follows: G1; 31 (41.8%), G2; 9 (12.1%), G3; 11 (14.8%), G6; 1 (1.4%), G8; 2 (2.7%), G9; 14 (18.9%) and G12; 6 (8.1%). Sequence analyses of the Ghanaian VP7 isolates and cognate genes of known ancestral and contemporary reference strains revealed the accumulation of point mutations in the antigenic regions. The P-types were successfully characterized for 57 of the 89 samples that had sufficient stool. Forty-eight (84.2%) of these were identified as P[8]a strains of which 5 (10.4%) were characterized as the rare OP354-like human rotavirus P[8]b subtype. Other strains detected in the study were P[4], 1 (1.8%), P[6], 7 (12.3%) and P[14], 1 (1.8%). P[14] was found in combination with G6. In Silico PCR performed with in-house primers failed to genotype Ghanaian P[8] isolates. However, In Silico PCR with published primers, Rev compl 1T-1D and P8b-MMC38 successfully genotyped the Ghanaian P[8]a and P[8]b isolates respectively