Genotyping of Rotavirus by Using RT-PCR Methods

There is a great diversity of rotavirus genotypes circulating worldwide, with dominant genotypes changing from year to year. Rotavirus genotyping was performed by using reverse transcription PCR with type-specifi c-primers. Since rotavirus is a RNA virus that has high mutation rate, there was a possibility of technical diffi culty in genotyping due to mutation in the primer binding sites. During Indonesian rotavirus surveillance study 2006-2009, it was reported that 17% of samples subjected for G type and 21% of samples subjected for P type were untypeable. The objective of this study was to identify genotypes of the samples that were untypeable previously using RT-PCR based on the method described by Das et al. (1994) and Gentsch et al. (1992). There were 30 samples subjected to G type and 61 samples subjected to P type to be re-typed using method described by Gouvea et al. (1990) and Simmond et al. (2008) for G and P typing, respectively. By using another set of primer, the genotype of all samples was identifi ed. This study highlights the importance of a constant reconsideration of primer sequences employed for the molecular typing of rotaviruses.


Introduction
Rotaviruses cause severe diarrhoea in children under 5 years old in developed and developing countries.Studies published between 1986 and 1999 showed that rotaviruses cause 22% of childhood diarrhoea hospitalization.This proportion increased to 39% in 2004.Approximately 611,000 children under 5 years old die every year from rotaviruses, mostly in developing countries (Parashar, et al., 2006).
Rotaviruses are 70-nm icosahedral viruses that belong to the family Reoviridae.Rotaviruses consist of seven different serogroups of A -F.The most common rotaviruses that infect humans belong to group A. The particle of a rotavirus has a capsid that surrounds the double-stranded RNA genome consisting of 11 segments.Rotavirus genome segments encode 6 structural proteins (VP1-4, VP6, VP7) and 6 non-structural proteins (NSP1-6) (Dennehy, 2008).
A rotavirus has a capsid that is composed of three concentric layers of protein.The innermost layer is composed of VP2 protein, the middle layer is composed of VP6 protein and the outer layer is composed of VP7 glycoprotein and VP4 forms spikes that protrude from the surface of outer layer.VP4 and VP7 are essential proteins that elicit immune responses and used as the basis of a binary classifi cation system.Typing based on VP7 protein (a glycoprotein = G type antigen) is known as the G-typing while based on VP4 protein (protease-sensitive protein = P-type antigen) is known as the P-typing (Greenberg and Estes, 2009).
Since rotavirus is a RNA virus that has high mutation rate, there is a great diversity of wild-type strains circulating in the world.The dominant strains always changes yearly.This changing is due to accumulation of point mutations that lead to antigenic change or new strain emergence (Kirkwood, 2010).Some hospital-based studies reported that 5 rotavirus genotypes, G1P[8], G2P[4], G3P[8], G4P[8], and G9P[9] have been the most common causes of severe disease in children worldwide (Bishop, 2009).
Performing rotavirus surveillance is important since the strain circulated in the world is changing in every time and every place.Beside providing valuable data of circulating strains that are important for vaccine development, this surveillance can tract emergent type, and changes in strain diversity when vaccines are implemented (Widdowson et al., 2009).To support rotavirus surveillance in the world, WHO provided a guideline for genotyping of rotavirus by using RT-PCR methods with different kind of primers applied in different geographical locations (WHO, 2009).
There were some studies to identify the genotype of rotavirus circulating in Indonesia, such as conducted by Soenarto et al. (2009) which based on RT-PCR method as described by Das et al. (1994) and Gentsch et al. (1992).It was reported that 17% of samples subjected for G typing and 21% of samples subjected for P typing were untypeable.The higher percentage of untypeable samples need a further study.So, the objective of this study is to identify the genotype of the samples that previously untypeable using RT-PCR based on the method described by Das et al. (1994) and Gentsch et al. (1992).

Rotavirus detection
All stool samples were tested for the presence of group A rotavirus using IDEIA TM Rotavirus (DakoCytomation) kit according to the manufacturer instructions.Rotavirus RNA was extracted from rotavirus positive stool samples and analyzed to determine both the VP7 (G-type) and VP4 (P-type) genotypes using the method described previously (Das et al., 1994;Gentsch et al., 1992).The samples that failed to yield detectable PCR product were classifi ed as untypeable samples and be used in this research.

RNA extraction
Rotavirus RNA extracted from rotavirus positive stool samples using Minikit (Qiagen) according to the manufacture instructions.

Results and Discussion
During Rotavirus Surveilance Study 2006-2009, there were 45 samples that has not been type detected for G typing using the method described by Das et al. (1994) and 86 samples for P typing using the method described by Gentsch et al. (1992).From these samples only 30 samples subjected to G typing and 61 samples subjected to P typing had suffi cient volume to be re-typed using method described by Gouvea et al. (1990) and Simmond et al. (2008) for G and P typing, respectively.Figure 1A and 1B show the representative of RT PCR results of G typing and P typing respectively.Finally, the genotype of all samples could be identifi ed and the result can be seen in Table 2.
Molecular methods for rotavirus genotyping provides increased sensitivity for G and P genotyping.The methods allows the identifi cation of putative reassortant strains.However, the reagents and methods used for genotyping needed to be closed monitored and up-dated since in nature, rotavirus is able to evolve rapidly by accumulation of point mutations through genetic drift and possibly zoonotic transmission and subsequent reassortment (Simmond et al., 2008).
Nucleotide mismatches between the VP7 gene and their primers are associated with genotyping failure (Rahman et al., 2005).When conducted a study in Bangladesh, Rahman et al. (2005) found some untypeable strains using the routine primer set described by Das et al. (1994).By sequence analysis, they found four nucleotide substitution at the G1 primer binding site which may have caused this genotyping failure.After using another primer set described by Gouvea et al. (1990), they could identify these untypeable strains.
Table 2. Distribution of G and P genotype that previously untypeable using the method described by Das et al. (1994) and Gentsch et al. (1992).

Untypeable G genotype (n)
G genotype after re-typing Untypeable P genotype (n) P genotype after re-typing  This result supporting the study conducted by Simmond et al. (2008) who designed VP4F/VP4R primers.These primers provides increased sensitivity and allows typing the samples that previously untypeable using available methods.After re-tested using VP4F/VP4R primers, 63.7% of samples that previously untypeable using con3/con2 primers can be typed.Sequence analysis, shows that there are point mutations at the P-type-specifi c primer binding sites which may caused genotyping failure.The mutation in the primer binding site that cause nucleotide mismatches between the VP4 gene and primer sequence also reported by Adah et al. (1997), Iturriza-Gomara et al. (2000), and Iturriza-Gomara et al. (2004).
Figure 2 and 3 described the location of primer that were used in this study.For G typing, the location of Das's primers are totally different with Gouvea's primers.In contrast, Simmond's primers similar with Gentsch's primers but amplify shorter fragments of VP4 when compare with Gentsch's primer.
In this study, we found that by using VP4F/VP4 primers increased the sensitivity of P genotyping, comparing with con3/con2 primers.There are some reasons why VP4F/ VP4R primers has higher sensitivity.Con3/ con2 primers were designed by Gentsch from sequences derived from a limited number of cell-culture adapted rotavirus strains available in public databases at the time of the study (Gentsch et al., 1992).VP4F/VP4R primers were designed by Simmond based on the sequences derived from over 200 strains of different P types of human rotavirus strains isolated in the last decade.VP4F/VP4R increased effi ciency of a PCR reaction since VP4F/VP4R amplifi es 663 bp, a shorter fragment of the VP4 encoding gene, compare with con2/con3 that amplifi es 876 bp (Simmond et al., 2008).
This study highlights the importance of a constant reconsideration of primer sequences employed for the molecular typing of rotaviruses.Since rotavirus is able to evolve rapidly by accumulation of point mutations, close monitoring of rotavirus genotyping methods is important.

Acknowledgment
We express our great gratitude to World Health Organization (WHO) that has funded this research.We thank to Carl Kirkwood (Murdoch Children Research Institute, Australia) for his support; Linda Oktabriana (Faculty of Medicine, Universitas Gadjah Mada), for the laboratory work; doctors, nurses, and administrative staff at all study sites for being our wonderful partners.

Table 1 .
Primers correspond to VP7 and VP4 genes for rotavirus genotyping