- Email: [email protected]
Rotavirus in diarrheal children in rural Burkina Faso: High prevalence of genotype G6P[6]
Infection, Genetics and Evolution 12 (2012) 1892–1898
Contents lists available at SciVerse ScienceDirect
Infection, Genetics and Evolution journal homepage: www.elsevier.com/locate/meegid
Rotavirus in diarrheal children in rural Burkina Faso: High prevalence of genotype G6P[6] Johan Nordgren a,⇑,1, Isidore Juste O. Bonkoungou b,c,1, Leon W. Nitiema d,e, Sumit Sharma a, Djeneba Ouermi d, Jacques Simpore d, Nicolas Barro c, Lennart Svensson a a
Division of Molecular Virology, Department of Clinical and Experimental Medicine, Linköping University, SE-58185 Linköping, Sweden Laboratoire National de Santé Publique du Burkina Faso, 09 BP 24 Ouagadougou, Burkina Faso Laboratoire de Biologie Moléculaire, d’Epidémiologie et Surveillance des Bactéries et Virus Transmissibles par les Aliments, CRSBAN/UFR-SVT, Université de Ouagadougou, 03 BP 7021 Ouagadougou 03, Burkina Faso d Centre de Recherche Biomoléculaire Pietro Annigoni, Saint Camille CERBA/LABIOGENE, Université de Ouagadougou, 01 BP 364 Ouagadougou, Burkina Faso e Centre de Recherche en Sciences Biologiques, Alimentaires et Nutritionnelles (CRSBAN), Université de Ouagadougou, 03 BP 7021 Ouagadougou 03, Burkina Faso b c
a r t i c l e
i n f o
Article history: Received 12 June 2012 Received in revised form 9 August 2012 Accepted 14 August 2012 Available online 3 September 2012 Keywords: Rotavirus Genotypes G6P[6] Rural areas Gastroenteritis Children
a b s t r a c t Group A rotavirus (RVA) is the most common cause of severe gastroenteritis in young children globally, and responsible for a significant number of deaths in African countries. While vaccines are available, trials have shown a lesser efficacy in Africa. One of the reasons could be the prevalence and/or emergence of unusual or novel RVA strains, as many strains detected in African countries remain uncharacterized. In this study, we characterized RVA positive specimens from two remote rural areas in Burkina Faso, West Africa. In total 56 RVA positive specimens were subgrouped by their VP6 gene, and G-and P typed by PCR and/or sequencing of the VP7 and VP4 genes, respectively. Notably, we found a high prevalence of the unusual G6P[6]SGI strains (23%). It was the second most common constellation after G9P[8]SGII (32%); and followed by G1P[8]SGII (20%) and G2P[4]SGI (9%). We also detected a G8P[6]SGI strain, for the first time in Burkina Faso. The intra-genetic diversity was high for the VP4 gene with two subclusters within the P[8] genotype and three subclusters within the P[6] genotype which were each associated with a specific G-type, thereby suggesting a genetic linkage. The G6P[6]SGI and other SGI RVA strains infected younger children as compared to SGII strains (p < 0.05). To conclude, in this study we observed the emergence of unusual RVA strains and high genetic diversity of RVA in remote rural areas of Burkina Faso. The results highlight the complexity of RVA epidemiology which may have implication for the introduction of rotavirus vaccines currently being evaluated in many African countries. Ó 2012 Elsevier B.V. All rights reserved.
1. Introduction Group A rotavirus (RVA) is the leading cause of severe acute gastroenteritis in infants and young children, responsible for approximately 453,000 deaths annually (Tate et al., 2011), with more than 230,000 of the deaths occurring in sub-Saharan Africa. Five of 10 countries with the highest mortality due to RVA diarrhea are located in Africa (Tate et al., 2011). The high burden of RVA disease in children has accelerated the development of vaccines (Parashar et al., 2006; Patel et al., 2008), and the World Health Organization (WHO) now recommends the inclusion of RVA vaccines in the national immunization program ⇑ Corresponding author. Address: Division of Molecular Virology, Department of Clinical and Experimental Medicine, Medical Faculty, Linköping University, SE-581 85 Linköping, Sweden. Tel.: +46 70 2239323; fax: +46 10 31375. E-mail address: [email protected] (J. Nordgren). 1 These authors are contributed equally to this study. 1567-1348/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.meegid.2012.08.014
in countries highly affected by RVA. Currently, two vaccines, Rotarix and RotaTeq, are licensed and available in many countries. Although the vaccines have demonstrated high efficacy in industrialized countries, vaccine trials from African countries have shown a lesser efficacy (Madhi et al., 2010; Sow et al., 2012). The reasons for this are yet unknown, but a different and more diverse genetic setup of RVA strains could be one of the reasons. The RVA strain diversity in Africa is higher than from industrialized countries and also have a high degree of animal or animal derived genes (Nordgren et al., 2012; Todd et al., 2010). Some other major differences include the relatively high prevalence of the P[6] genotype as well as a high detection rate of G8 in certain countries (Armah et al., 2010; Nordgren et al., 2012; Page et al., 2010b; Todd et al., 2010). This highlights the need for understanding of the molecular epidemiological patterns of human and animal RVA strains in the continent in order to assess factors that could be important for vaccine efficacy. Few studies regarding the molecular epidemiology of RVA in rural areas in Africa have been performed (Armah et al., 2003),
J. Nordgren et al. / Infection, Genetics and Evolution 12 (2012) 1892–1898
and many RVA strains still remain uncharacterized (Todd et al., 2010). In Burkina Faso, previous studies have observed a high burden of RVA disease in children with high RVA prevalence during the cold dry season December–February (Bonkoungou et al., 2010, 2011; Nitiema et al., 2011; Steele et al., 2010). Furthermore, in a previous study from the capital Ouagadougou, RVA strains were characterized by their G and P-types (Nordgren et al., 2012). In this study we investigated RVA strain diversity in two remote rural locations situated south-west (Boromo) and north (Gourcy) of Ouagadougou. The samples were collected during the same time frame as the previous RVA study from the capital (Nordgren et al., 2012), thus enabling a direct comparison between urban and rural settings in Burkina Faso during the same RVA season. Of note is the high prevalence (23%) of the highly unusual G6P[6] rotavirus strains. The obtained results give a more thorough understanding about the complexity of RVA epidemiology in Burkina Faso and the results are discussed in the relation to vaccine efficacy. 2. Materials and methods 2.1. Study population and specimens The study was conducted in two rural areas, in north (Gourcy, distance 140 km) and south-west (Boromo, distance 185 km) of the capital Ouagadougou in Burkina Faso during January and February 2010, which is the period with highest RVA burden in Burkina Faso. A total of 80 samples, 17 from Gourcy and 63 from Boromo were collected. These regions are tropical savannah areas and the main sources of income are subsistence farming, animal husbandry and small scale trade. Fecal specimens were collected from children <5 years of age who sought medical care for acute diarrheal illness, defined as >3 liquid stools over a 24 h period. 2.2. Clinical-epidemiological assessment Clinical information was obtained by reviewing the clinical records of each case. Information regarding the age, sex, hospitalization and clinical symptoms such as fever (temperature P 38 °C), vomiting and dehydration, and the characteristics of stool were recorded for each child. All children were clinically evaluated by general practitioners following a local adaptation of the World Health Organization (WHO) strategy for the management of diarrhea. 2.3. Rotavirus antigen detection All stool samples were analyzed for RVA antigens and adenovirus serotypes 40/41with an immunochromatograpic method (SD Bioline Rota/AdenoÒ; Standard Diagnostics, Inc., Kyonggi-Do, South Korea) following the manufacturer’s instructions. 2.4. Viral RNA extraction Viral RNA was extracted from 140 ll of 10% stool suspensions following the manufacturer’s instructions using a QIAamp Viral RNA Mini Kit (Qiagen, Hilden, Germany). A total of 60 ll of purified viral RNA was collected and stored at 70 °C until reverse transcription reaction was carried out.
1893
[pd(N)6], denatured at 97 °C for 5 min, and quickly chilled on ice for 2 min, followed by addition of one Illustra Ready-To-Go RTPCR bead (GE Healthcare, Uppsala, Sweden) and RNase-free water to a final volume of 50 ll. The RT reaction was carried out for 30 min at 42 °C to produce the cDNA. 2.6. Quantification and VP6 subgrouping by real-time PCR All rotavirus positive specimens were quantified and VP6 subgrouped by the use of a LUX real-time PCR as described previously (Nordgren et al., 2010). This real-time PCR uses labeled primers with different fluorophores for each VP6 subgroup and external plasmid standards for semi-quantification (Nordgren et al., 2010). 2.7. Rotavirus G and P genotyping Rotavirus G and P genotyping was performed using semi-nested type-specific multiplex PCR’s (Gentsch et al., 1992; Gouvea et al., 1990; Iturriza-Gomara et al., 2004), able to detect seven G-types (G1, G2, G3, G4, G8, G9, G10) and six P-types (P[4], P[6], P[8], P[9], P[10] and P[11]), respectively. PCR products were examined on an agarose gel with ethidium bromide staining and visualized under UV light. The RVA G and P genotypes were determined by their specific size on the agarose gel. 2.8. PCR and sequencing of VP4 and VP7 genes of the detected RVA strains Sequencing of the PCR-amplicons for the VP7-and VP4 genes was performed for all samples positive for RVA antigen using 4 consensus primers (VP7-F, VP7-R, Con-2 and Con-3) as previously described (Iturriza-Gomara et al., 2004; Nordgren et al., 2012). The following thermal cycling conditions were used: An initial denaturation step at 94 °C for 5 min followed by 40 cycles of amplification (30 s at 94 °C, 30 s at 50 °C and 1 min 30 s at 72 °C), with a final extension of 7 min at 72 °C. Nucleotide sequencing was performed by Macrogen Inc. (Seoul, Korea). The sequencing reaction was based on BigDye chemistry, using the same primers as in the PCR as sequencing primers. 2.9. Sequence analysis Multiple sequence alignment of the obtained nucleotide sequences were performed using the ClustalW algorithm with default parameters on the European Bioinformatics Institute server. A phylogenetic analysis of the alignment file was performed using the MEGA 5.05 software package. A tree was constructed using the neighbor-joining method and Kimura two-parameter model (Kimura, 1980; Saitou and Nei, 1987). The statistical significance of the relationships obtained was estimated by bootstrap resampling analysis (1000 replications). The assignment of genetic lineages within the G and P genotypes was done using the following references: (Arora and Chitambar, 2011; De Grazia et al., 2011; Martella et al., 2006; Mascarenhas et al., 2010; Phan et al., 2007a,b; Sharma et al., 2009). 2.10. Sequence submission The partial nucleotide sequences for the RVA VP4 and VP7 genes can be found under GenBank accession numbers JX154433– JX154544.
2.5. Reverse transcription 2.11. Statistical analysis Reverse transcription (RT) was essentially carried out as described previously (Bucardo et al., 2007). Briefly, 28 ll of dsRNA was mixed with 50 pmol of random hexadeoxynucleotides
Interval data (age) was analyzed using independent t-test with 2-tailed significance, using SPSS 19.0 (SPSS Inc., Chicago, IL, USA).
1894
J. Nordgren et al. / Infection, Genetics and Evolution 12 (2012) 1892–1898
3. Results 3.1. High prevalence of rotavirus in rural Burkina Faso The samples were collected during the cold dry season in Burkina Faso (January–February), which in earlier studies was shown to be the period with highest RVA burden in children under the age of 5 (Bonkoungou et al., 2010; Nitiema et al., 2011). Of the 80 fecal specimens analyzed, 56 (70%) were found positive for RVA antigen. RVA was detected in 47/63 (74.6%) children with diarrhea from Boromo and 9/17 (52.9%) from Gourcy. Among these RVA infected children, 13/56 (23.2%) were hospitalized and most cases 52/56 (92.9%) occurred among children less than 2 years of age. 3.2. High prevalence of the unusual G6P[6] rotavirus strains and the detection of a G8P[6] strain The G and P genotype distribution was determined by the multiplex nested PCR assay and sequencing of the VP4 and VP7 genes; the nested PCR assay data was only used when sequencing was not successful, mostly in samples with low viral load. In total 44 VP4 and 49 VP7 sequences with good sequence quality were obtained. The distribution of G and P types is presented in Table 1 and shows that, notably, 13 (23%) of the 56 RVA positive children were infected with G6P[6] RVA. Five different G-types were detected with the following prevalence rates: G9 (34%), G6 (23%), G1 (21%), G2 (11%) and G8 (2%). Among the P-genotypes, P[8] was the most prevalent with 54% detection rate, followed by P[6] (30%) and P[4] (9%). The G/P constellations most frequently detected were G9P[8] (32%), G6P[6] (23%), G1P[8] (20%) and G2P[4] (9%). Mixed G or P type infections were found in 4% (n = 2) of the total specimens, whereas 5% (n = 3) and 7% (n = 4) of the RVA positive specimens could not be assigned a G-type and P-type respectively (Table 1). When stratified according to VP6 subgroups, we observed that 35.7% of the specimens (n = 20) were SGI and 64.3% (n = 36) were SGII. All SGII positive specimens were associated with P[8] genotype, whereas all SGI positive specimens were associated with P[4] or P[6] genotypes with the exception of two RVA specimens which had P[6] and SGII specificities. These two specimens were of the constellation G2P[6]SGII and G9P[6]SGII. 3.3. The unusual G6P[6]SGI strains infected younger children as compared to other strains The mean age of children infected with RVA SGI strains was 9.4 ± 7.2 months, as compared to children infected with RVA SGII, which was 14.9 ± 8.1 months (p < 0.05). The unusual G6P[6] strains infected the youngest age group 7.2 ± 5.4 months, and the most
Table 1 Distribution of G and P genotypes of RVA strains detected among children with gastroenteritis in rural areas of Gourcy and Boromo, Burkina Faso during January and February 2010. Rotavirus genotype
G1 G2 G6 G8 G9 Gmixa Gnt Total nt = Not determined. a G1/G9.
No. (%) of strains P[4]
P[6]
P[8]
P[nt]
Total
0 5 0 0 0 0 0 5 (9)
1 1 13 1 1 0 0 17 (30)
11 0 0 0 18 0 1 30 (54)
0 0 0 0 0 2 2 4 (7)
12 (21) 6 (11) 13 (23) 1 (2) 19 (34) 2 (4) 3 (5) 56 (100)
common RVA strain G9P[8] infected the oldest age-group 16.1 ± 9.1 months (p < 0.05) (Table 2). When stratified according to VP6 subgroups, we observed that SGI RVA strains significantly infected younger children as compared to SGII RVA strains (9.4 ± 7.2 vs 14.9 ± 8.1 months). 3.4. Phylogenetic analysis of the VP4 and VP7 genes We analyzed the sequences of the VP4 (n = 44) and VP7 (n = 49) genes of all RVA positive specimens containing sufficient viral load and yielding high quality sequence data (Fig. 1A and B). The phylogenetic analysis demonstrated that the VP7 genes within each Gtype cluster are highly similar to each other and independent of P genotype association, with the exception of specimen 35, a G1P[8] RVA which is relatively more different (1.2–1.5% nt difference) compared to the other detected G1P[8] RVs (Fig. 1B). In contrast, for the VP4 genes we observed different subclusters within the same P-genotype cluster, which were associated with different G-types, demonstrating a more promiscuous behavior of the VP4 gene (Fig. 1A). Within the P[6] cluster, we observed 3 subclusters, for G9P[6], G8P[6] and G6P[6] RVA strains respectively, and in the P[8] cluster we observed two subclusters, corresponding to G9P[8] and G1P[8] RVs respectively. The only exception is the VP4 gene for specimen 16, a G1P[8] strain, for which the VP4 gene is identical to the VP4 genes associated with the G9P[8] genotypes, indicating a reassortment event between G1P[8] and G9P[8] during the RVA season. Due to the shorter stretch of nt sequence obtained (500 bp) it was however omitted from Fig. 1A. The VP7 genes of the G9P[8] strains belonged to sublineage IIId, and are similar to G9 stains (GH1319) isolated in northern Ghana from 1998 to 2000 (99.2–99.5% nt identity) (Armah et al., 2003), and to strain 6222LP detected in South Africa in 1999 (97.7– 98.1% nt identity) (Page et al., 2010a). The VP7 genes of the G1P[8] and G1P[6] genotypes belonged to sublineage Ia, similar to the CMH042 strain detected in Thailand 2004 (99.0–99.1% nt identity) and to G1P[8] RVA strain SA4799DGM (98.5–98.8% nt identity) detected in South Africa in 2004. The VP4 gene associated with both the G1 and G9 genotypes belonged to lineage P[8]-III, but appeared in two different clusters (2.3–3.0% nt difference, Fig. 1A). The VP4 gene of the G9P[8] strains were highly similar to the Dhaka25-02 strain (98.9–99.3% nt identity), whereas the VP4 gene of the G1P[8] strains were highly similar to the BE00021 strain (99.3–99.4% nt identity). The VP7 genes of the G2P[4] strains, belonged to sublineage IIa, and were similar to G2 strains detected in Sierra Leone in 2005 (SL276, 98.6–98.8% nt identity) (Jere et al., 2011b), and to CH-128 (Trinh et al., 2010) detected in China in 2003 (98.2–98.4% nt identity) The P[4] gene of the G2P[4] RVA strains belonged to lineage P[4]-V and were also similar to the G2P[4] strains (99.1–99.3% nt identity to strain SL244) detected in Sierra Leone in 2005.
Table 2 Age of children infected with rotavirus stratified by VP6 subgroup and the three most common G, P and VP6 type constellations observed in this study
a b
Number
Age (months)
SGI SGII p-Valuea
20 36 0.013
9.4 ± 7.2 14.9 ± 8.1
G6P[6]SGI G1P[8]SGII G9P[8]SGII p-Valueb
13 11 18 0.005
7.2 ± 5.4 13.1 ± 8.1 16.1 ± 9.1
SGI vs SGII. G6P[6]SGI vs G1P[8]SGII and G9P[8]SGII.
J. Nordgren et al. / Infection, Genetics and Evolution 12 (2012) 1892–1898
1895
Fig. 1. (A) Phylogenetic analysis of sequences of the VP8⁄ subunit of the VP4 gene (nt 140–751). (B) VP7 gene (nt 135–796) with reference strains. The G and P types of the Burkina Faso strains are added after the isolation number and are marked with filled circles. Rotavirus strains from the capital Ouagadougou are added as a comparison and are marked with open circles. The brackets indicate the lineage of the Burkina Faso strains (in bold) and genotype of all strains, respectively.
1896
J. Nordgren et al. / Infection, Genetics and Evolution 12 (2012) 1892–1898
We further observed that the unusual G6P[6] RVA strains were nearly identical to the G6P[6] RVs detected in a previous study in Ouagadougou, Burkina Faso (Nordgren et al., 2012). These strains were related to the first human G6P[6] detected, the B1711 strain previously detected in Belgium 2002 in a child arriving from Mali (98.2–98.5% nt identity for the VP7 gene)(Rahman et al., 2003). The VP4 genes of the G6P[6] strains, belonged to sublineage Ia (96.7–97.0% nt identity to the G6P[6] strain B1711), and nearly identical to the G6P[6] strains detected in Ouagadougou. The VP4 gene of the G9P[6] and G8P[6] strains were different compared to the VP4 gene of the G6P[6] strains (97.0–97.3% and 98.7–99.0% nt identity, respectively), but belonged to the same sublineage. The VP7 gene of the G8P[6] strain, belonging to lineage G8d, was related to the DRC88 strain (98.5% nt identity), detected in the Democratic Republic of Congo in 2003, and to RVA strains SL285 and KY1646 detected in Sierra Leone 2005 (98.8% nt identity) and Kenya 1999 (98.9% nt identity), respectively (Jere et al., 2011b; Matthijnssens et al., 2006; Page et al., 2010b).
4. Discussion In this study we have characterized RVA strains from two rural areas of Burkina Faso. These settings are remote, and relatively few studies regarding molecular characterization of RVs have previously been performed from similar settings in Africa. The major finding was the high prevalence of the rare G6P[6] strains, as well as the detection of a G8P[6] strain for the first time in Burkina Faso. We observed a high diversity of RVA strains, in total 8 G/P type constellations were observed during only two months. The majority of RVA strains belonging to the G9P[8]SGII genotype (32%), with the unusual G6P[6]SGI strains being the second most common genotype in this study (23%), followed by G1P[8]SGII (20%). The high prevalence of the unusual G6P[6] strains in rural Burkina Faso is noteworthy. Previously, this strain had only been observed once in Belgium from a child arriving from Mali in 2003 (Rahman et al., 2003), and subsequently in high prevalence (13%) in a study from the capital of Burkina Faso, Ouagadougou (Nordgren et al., 2012) during a similar time-frame as this study. These findings combined may suggest that the G6P[6] strain is emerging as an important pathogen in pediatric RVA diarrhea in Burkina Faso. In this study, we first G and P typed all strains using a nested PCR approach commonly used worldwide (Iturriza-Gomara et al., 2004). When subsequently all VP4 and VP7 genes were sequenced, we observed a number of discrepancies. Three of the G6P[6] strains (n = 13) were initially PCR-typed as G1P[6] strains with the nested PCR approach (data not shown), which suggests that the prevalence of G6P[6] strains could previously have been underestimated at least in Africa. The G1 primer matched in 12 of 22 nt at the corresponding position in the VP7 gene of the G6 strains, but a low annealing temperature as well as high viral load could have led to the primer mismatch observed. However, a presence of a mixed G1/G6 infection cannot be ruled out since this would not be apparent with sanger sequencing if the different genotypes are present in highly varying concentrations. The fact that a high percentage of RVA strains in Africa remain untypeable suggests that novel genotypes such as G6P[6] could be more widespread than previously assumed, all suggesting that sequencing of RVA strains from Africa is most warranted. We observed significant differences in age-profiles, with SGI strains, especially G6P[6]SGIs infecting younger age-groups (7.2 ± 5.4 months) as compared to SGII strains; G1P[8]SGII (13.1 ± 8.1 months) and G9P[8]SGII (16.1 ± 9.1 months). Interestingly, in a study from the capital Ouagadougou during the same time frame (Nordgren et al., 2012), the opposite pattern was observed. In
Ouagadougou, the G6P[6]SGI and other SGI strains significantly infected an older age-group as compared to G9P[8]SGII strains. We compared the age-profiles of the G6P[6] strains between these two studies and observed a statistically significant difference (p < 0.05). The reason behind these differences in age-profiles between the urban and rural settings during the same time period remains to be elucidated and confirmed in more studies with more samples. Differences in RVA type specificities in different parts of the same country have been reported (Kirkwood et al., 2011). The rural areas in this study are rather remote and far from the capital; one possibility could be that during the previous RVA season, RVA strains with different G, P and VP6 type specificities were circulating in the urban and rural settings respectively, thus creating a different immunological response in the children during this RVA season. Another explanation could be a common source of infection of G6P[6] or other genotypes into a setting having a distinct ageprofile (e.g. many younger children). However, in the study from the capital, the G6P[6] strains were detected in children from many different parts of the city (data not shown), and for the rural community, the same age profiles were seen both in Gourcy and Boromo which are 170 km apart. The VP6 gene, coding for the major immunogenic protein of the RVA virion, is often not genotyped during epidemiological investigations which focuses more on G and P types. The clear clinicalepidemiological differences between VP6 types, e.g. age-profiles, that we have observed between children in this and in a previous study (Nordgren et al., 2012) suggest that it should be more investigated in surveys of RVs in the future. The G6P[6] strains were observed at different time points in the rural settings; the first G6P[6] strain was observed on the 13th of January 2010 in Gourcy, whereas the first G6P[6] in Boromo was observed the 28th of January 2010. The first G6P[6] in the capital Ouagadougou in the previous study (Nordgren et al., 2012) was observed the 14th of January 2010, similar to that of Gourcy. Moreover, the frequency of different G/P and VP6 constellations varied between the urban and rural settings, giving further strength to the assumption of different epidemiological patterns of RVs in different geographical areas of Burkina Faso. We observed low frequencies of mixed infections as compared to other RVA studies from Africa (Todd et al., 2010). This could be explained by the fact that G and P types in this study were primarily assigned by sequencing all VP7 and VP4 genes from the RVA positive samples, and only with the nested multiplex PCR assay when sequencing PCR was not possible due to low viral load in the samples. As observed previously (Nordgren et al., 2012) the multiplex nested PCR assays could overestimate the amount of mixed infections due to mispriming, hence presence of mixed RVA infections could preferably be determined by deep sequencing (Jere et al., 2011a). We performed a phylogenetic analysis of the sequenced VP4 and VP7 genes (Fig. 1A and B). We observed many sub-clusters when analyzing the VP4 genes, each belonging to a specific G-type. This was not the case for the VP7 genes, which were highly identical to each other independently of which P genotype they were associated with. For example, within the P[6] cluster we observed three subclusters, each belonging to G9P[6], G8P[6] and G6P[6] RVA strains respectively. Of note, sample 20, being genotype G9P[6] and representing a separate subcluster of P[6], was one of the only two samples that had P[6] and SGII specificities. Also, in the P[8] cluster we observed two subclusters, corresponding to G9P[8] and G1P[8]. However, the VP7 genes of the same G-type belonging to different P-types were highly identical. This indicates a genetic linkage between VP4 subclusters and different G genotypes, or even VP6 subgroups. We compared the rural rotavirus strains to those from the study in the capital (Nordgren et al.,
J. Nordgren et al. / Infection, Genetics and Evolution 12 (2012) 1892–1898
2012), and we observed that they were highly identical, thus indicating that the same strains were circulating in both the urban and rural settings, although in different proportions and in slightly different time frames (Fig. 1A and B). The VP7 gene of the predominant G9P[8] strains detected in this study were highly identical to G9 RVA strains detected in the neighboring country of Ghana (Armah et al., 2003) in the end of the 1990ths, demonstrating that this particular G9 has been present and circulating in West Africa for a long time. The emergence and continued high prevalence of G9 strains in Africa has frequently been reported (Armah et al., 2003, 2010), and the results of this study show that they play major role in pediatric rotavirus diarrhea in rural Burkina Faso. The VP7 and VP4 genes of the G2P[4] strains were similar to G2P[4] strains isolated in Sierra Leone in West Africa in 2005. In previous studies from Brazil, G2P[4] RVA strains were observed in high prevalence in a pre-vaccinated pediatric population (Gurgel et al., 2007). The G2 and P[4] types are not present in the monovalent G1P[8] vaccine, and it has been suggested that the vaccine shows a lesser efficiency to this type (Gurgel et al., 2007). During this RVA season however, the G2P[4] RVs were not highly prevalent and a study from Burkina Faso the previous year showed a similar prevalence of G2 strains (Bonkoungou et al., 2011). We further observed that the unusual G6P[6] was nearly identical to G6P[6] strains from the capital Ouagadougou, Burkina Faso, and similar to the B1711 strain previously isolated in Belgium in a child arriving from Mali (Rahman et al., 2003). The G8P[6] strain, detected for the first time in Burkina Faso, was closely related with the DRC88 and SL285 strains detected in the Democratic Republic of Congo and Sierra Leone 2003 and 2005, respectively (Jere et al., 2011b; Matthijnssens et al., 2006). G8P[6] RVs are reported frequently from several countries in Africa. During a survey in 2005 in the nearby country of Sierra Leone in West Africa, 18% of the RVA strains belonged to G8 (Jere et al., 2011b). The G6P[6] and G8P[6] share many similarities, they are both assumed to have emerged in the human population from relatively recent reassortment events from bovine G8 and G6 strains and then adapted to humans (Rahman et al., 2003). Furthermore, molecular characterization of the G6P[6] strains detected in Ouagadougou, Burkina Faso showed high homology to the G8P[6] strain DRC86 (Matthijnssens et al., 2006; Nordgren et al., 2012). The emergence and high prevalence of these G8 and G6 RVA strains in Africa raises concerns regarding the efficacy of the current vaccines for these particular genotypes, which are rare or nonexistent in other continents, and should be evaluated in the future. To conclude, this study has demonstrated a high diversity of RVA and high prevalence of the rare G6P[6] strains in remote rural areas of Burkina Faso. We observed clinical-epidemiological differences between RVA strains and also between urban and rural settings during the same time-period. This study provides new knowledge about the complexity of RVA epidemiology in Burkina Faso which can be useful for the future evaluation and/or introduction of RVA vaccines. Role of the funding source The study was supported by Swedish Research Council (Grant 10392). The study sponsor had no role in the design and conduct of the study. Acknowledgments We thank the parents and guardians of the enrolled children and the Health District Authorities in Boromo and Gourcy for their cooperation.
1897
References Armah, G.E., Steele, A.D., Binka, F.N., Esona, M.D., Asmah, R.H., Anto, F., Brown, D., Green, J., Cutts, F., Hall, A., 2003. Changing patterns of rotavirus genotypes in Ghana: emergence of human rotavirus G9 as a major cause of diarrhea in children. J. Clin. Microbiol. 41, 2317–2322. Armah, G.E., Steele, A.D., Esona, M.D., Akran, V.A., Nimzing, L., Pennap, G., 2010. Diversity of rotavirus strains circulating in West Africa from 1996 to 2000. J. Infect. Dis. 202 (Suppl.), S64–S71. Arora, R., Chitambar, S.D., 2011. Full genomic analysis of Indian G1P[8] rotavirus strains. Infect. Genet. Evol. 11, 504–511. Bonkoungou, I.J., Damanka, S., Sanou, I., Tiendrebeogo, F., Coulibaly, S.O., Bon, F., Haukka, K., Traore, A.S., Barro, N., Armah, G.E., 2011. Genotype diversity of group A rotavirus strains in children with acute diarrhea in urban Burkina Faso, 2008–2010. J. Med. Virol. 83, 1485–1490. Bonkoungou, I.J., Sanou, I., Bon, F., Benon, B., Coulibaly, S.O., Haukka, K., Traore, A.S., Barro, N., 2010. Epidemiology of rotavirus infection among young children with acute diarrhoea in Burkina Faso. BMC Pediatr. 10, 94. Bucardo, F., Karlsson, B., Nordgren, J., Paniagua, M., Gonzalez, A., Amador, J.J., Espinoza, F., Svensson, L., 2007. Mutated G4P[8] rotavirus associated with a nationwide outbreak of gastroenteritis in Nicaragua in 2005. J. Clin. Microbiol. 45, 990–997. De Grazia, S., Martella, V., Rotolo, V., Bonura, F., Matthijnssens, J., Banyai, K., Ciarlet, M., Giammanco, G.M., 2011. Molecular characterization of genotype G6 human rotavirus strains detected in Italy from 1986 to 2009. Infect. Genet. Evol. 11, 1449–1455. Gentsch, J.R., Glass, R.I., Woods, P., Gouvea, V., Gorziglia, M., Flores, J., Das, B.K., Bhan, M.K., 1992. Identification of group A rotavirus gene 4 types by polymerase chain reaction. J. Clin. Microbiol. 30, 1365–1373. Gouvea, V., Glass, R.I., Woods, P., Taniguchi, K., Clark, H.F., Forrester, B., Fang, Z.Y., 1990. Polymerase chain reaction amplification and typing of rotavirus nucleic acid from stool specimens. J. Clin. Microbiol. 28, 276–282. Gurgel, R.Q., Cuevas, L.E., Vieira, S.C., Barros, V.C., Fontes, P.B., Salustino, E.F., Nakagomi, O., Nakagomi, T., Dove, W., Cunliffe, N., Hart, C.A., 2007. Predominance of rotavirus P[4]G2 in a vaccinated population, Brazil. Emerg. Infect. Dis. 13, 1571–1573. Iturriza-Gomara, M., Kang, G., Gray, J., 2004. Rotavirus genotyping: keeping up with an evolving population of human rotaviruses. J. Clin. Virol. 31, 259–265. Jere, K.C., Mlera, L., O’Neill, H.G., Potgieter, A.C., Page, N.A., Seheri, M.L., van Dijk, A.A., 2011a. Whole genome analyses of African G2, G8, G9, and G12 rotavirus strains using sequence-independent amplification and 454(R) pyrosequencing. J. Med. Virol. 83, 2018–2042. Jere, K.C., Sawyerr, T., Seheri, L.M., Peenze, I., Page, N.A., Geyer, A., Steele, A.D., 2011b. A first report on the characterization of rotavirus strains in Sierra Leone. J. Med. Virol. 83, 540–550. Kimura, M., 1980. A simple method for estimating evolutionary rates of base substitutions through comparative studies of nucleotide sequences. J. Mol. Evol. 16, 111–120. Kirkwood, C.D., Boniface, K., Barnes, G.L., Bishop, R.F., 2011. Distribution of rotavirus genotypes after introduction of rotavirus vaccines, Rotarix(R) and RotaTeq(R), into the National Immunization Program of Australia. Pediatr. Infect. Dis. J. 30, S48–53. Madhi, S.A., Cunliffe, N.A., Steele, D., Witte, D., Kirsten, M., Louw, C., Ngwira, B., Victor, J.C., Gillard, P.H., Cheuvart, B.B., 2010. Effect of human rotavirus vaccine on severe diarrhea in African infants. New Engl. J. Med. 362, 289–298. Martella, V., Banyai, K., Ciarlet, M., Iturriza-Gomara, M., Lorusso, E., De Grazia, S., Arista, S., Decaro, N., Elia, G., Cavalli, A., Corrente, M., Lavazza, A., Baselga, R., Buonavoglia, C., 2006. Relationships among porcine and human P[6] rotaviruses: evidence that the different human P[6] lineages have originated from multiple interspecies transmission events. Virology 344, 509–519. Mascarenhas, J.D., Lima, C.S., de Oliveira, D.S., Guerra Sde, F., Maestri, R.P., Gabbay, Y.B., de Lima, I.C., de Menezes, E.M., Linhares Ada, C., Bensabath, G., 2010. Identification of two sublineages of genotype G2 rotavirus among diarrheic children in Parauapebas, Southern Para State, Brazil. J. Med. Virol. 82, 712–719. Matthijnssens, J., Rahman, M., Yang, X., Delbeke, T., Arijs, I., Kabue, J.P., Muyembe, J.J., Van Ranst, M., 2006. G8 rotavirus strains isolated in the Democratic Republic of Congo belong to the DS-1-like genogroup. J. Clin. Microbiol. 44, 1801–1809. Nitiema, L.W., Nordgren, J., Ouermi, D., Dianou, D., Traore, A.S., Svensson, L., Simpore, J., 2011. Burden of rotavirus and other enteropathogens among children with diarrhea in Burkina Faso. Int. J. Infect. Dis. 15, 646–652. Nordgren, J., Bucardo, F., Svensson, L., Lindgren, P.E., 2010. Novel light-uponextension real-time PCR assay for simultaneous detection, quantification, and genogrouping of group A rotavirus. J. Clin. Microbiol. 48, 1859–1865. Nordgren, J., Nitiema, L.W., Sharma, S., Ouermi, D., Traore, A.S., Simpore, J., Svensson, L., 2012. Emergence of unusual G6P[6] rotaviruses in children with acute gastroenteritis in Burkina Faso, 2009–2010. Emerg. Infect. Dis. 18, 589–597. Page, N., Esona, M., Armah, G., Nyangao, J., Mwenda, J., Sebunya, T., Basu, G., Pyndiah, N., Potgieter, N., Geyer, A., Steele, A.D., 2010a. Emergence and characterization of serotype G9 rotavirus strains from Africa. J. Infect. Dis. 202 (Suppl.), S55–S63. Page, N., Esona, M., Seheri, M., Nyangao, J., Bos, P., Mwenda, J., Steele, D., 2010b. Characterization of genotype G8 strains from Malawi, Kenya, and South Africa. J. Med. Virol. 82, 2073–2081. Parashar, U.D., Gibson, C.J., Bresse, J.S., Glass, R.I., 2006. Rotavirus and severe childhood diarrhea. Emerg. Infect. Dis. 12, 304–306.
1898
J. Nordgren et al. / Infection, Genetics and Evolution 12 (2012) 1892–1898
Patel, M.M., Widdowson, M.A., Glass, R.I., Akazawa, K., Vinje, J., Parashar, U.D., 2008. Systematic literature review of role of noroviruses in sporadic gastroenteritis. Emerg. Infect. Dis. 14, 1224–1231. Phan, T.G., Okitsu, S., Maneekarn, N., Ushijima, H., 2007a. Evidence of intragenic recombination in G1 rotavirus VP7 genes. J. Virol. 81, 10188–10194. Phan, T.G., Okitsu, S., Maneekarn, N., Ushijima, H., 2007b. Genetic heterogeneity, evolution and recombination in emerging G9 rotaviruses. Infect. Genet. Evol. 7, 656–663. Rahman, M., De Leener, K., Goegebuer, T., Wollants, E., Van der Donck, I., Hoovels, L.V., Van Ranst, M., 2003. Genetic characterization of a novel, naturally occurring recombinant human G6P[6] rotavirus. J. Clin. Microbiol. 41, 2088. Saitou, N., Nei, M., 1987. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol. Biol. Evol. 4, 406–425. Sharma, S., Paul, V.K., Bhan, M.K., Ray, P., 2009. Genomic characterization of nontypeable rotaviruses and detection of a rare G8 strain in Delhi, India. J. Clin. Microbiol. 47, 3998–4005. Sow, S.O., Tapia, M., Haidara, F.C., Ciarlet, M., Diallo, F., Kodio, M., Doumbia, M., Dembele, R.D., Traore, O., Onwuchekwa, U.U., Lewis, K.D., Victor, J.C., Steele,
A.D., Neuzil, K.M., Kotloff, K.L., Levine, M.M., 2012. Efficacy of the oral pentavalent rotavirus vaccine in Mali. Vaccine 30 (Suppl. 1), A71–A78. Steele, A.D., Page, N., de Beer, M., Sawadogo, S., 2010. Antigenic and molecular characterization of unusual rotavirus strains in Burkina Faso in 1999. J. Infect. Dis. 202 (Suppl.), S225–S230. Tate, J.E., Burton, A.H., Boschi-Pinto, C., Steele, A.D., Duque, J., Parashar, U.D., 2011. 2008 estimate of worldwide rotavirus-associated mortality in children younger than 5 years before the introduction of universal rotavirus vaccination programmes: a systematic review and meta-analysis. Lancet Infect. Dis. 12, 136–141. Todd, S., Page, N.A., Duncan Steele, A., Peenze, I., Cunliffe, N.A., 2010. Rotavirus strain types circulating in Africa: review of studies published during 1997– 2006. J. Infect. Dis. 202 (Suppl.), S34–S42. Trinh, Q.D., Pham, N.T., Nguyen, T.A., Phan, T.G., Yan, H., Hoang le, P., Khamrin, P., Maneekarn, N., Li, Y., Okitsu, S., Mizuguchi, M., Ushijima, H., 2010. Sequence analysis of the VP7 gene of human rotaviruses G2 and G4 isolated in Japan, China, Thailand, and Vietnam during 2001–2003. J. Med. Virol. 82, 878–885.
Contents lists available at SciVerse ScienceDirect
Infection, Genetics and Evolution journal homepage: www.elsevier.com/locate/meegid
Rotavirus in diarrheal children in rural Burkina Faso: High prevalence of genotype G6P[6] Johan Nordgren a,⇑,1, Isidore Juste O. Bonkoungou b,c,1, Leon W. Nitiema d,e, Sumit Sharma a, Djeneba Ouermi d, Jacques Simpore d, Nicolas Barro c, Lennart Svensson a a
Division of Molecular Virology, Department of Clinical and Experimental Medicine, Linköping University, SE-58185 Linköping, Sweden Laboratoire National de Santé Publique du Burkina Faso, 09 BP 24 Ouagadougou, Burkina Faso Laboratoire de Biologie Moléculaire, d’Epidémiologie et Surveillance des Bactéries et Virus Transmissibles par les Aliments, CRSBAN/UFR-SVT, Université de Ouagadougou, 03 BP 7021 Ouagadougou 03, Burkina Faso d Centre de Recherche Biomoléculaire Pietro Annigoni, Saint Camille CERBA/LABIOGENE, Université de Ouagadougou, 01 BP 364 Ouagadougou, Burkina Faso e Centre de Recherche en Sciences Biologiques, Alimentaires et Nutritionnelles (CRSBAN), Université de Ouagadougou, 03 BP 7021 Ouagadougou 03, Burkina Faso b c
a r t i c l e
i n f o
Article history: Received 12 June 2012 Received in revised form 9 August 2012 Accepted 14 August 2012 Available online 3 September 2012 Keywords: Rotavirus Genotypes G6P[6] Rural areas Gastroenteritis Children
a b s t r a c t Group A rotavirus (RVA) is the most common cause of severe gastroenteritis in young children globally, and responsible for a significant number of deaths in African countries. While vaccines are available, trials have shown a lesser efficacy in Africa. One of the reasons could be the prevalence and/or emergence of unusual or novel RVA strains, as many strains detected in African countries remain uncharacterized. In this study, we characterized RVA positive specimens from two remote rural areas in Burkina Faso, West Africa. In total 56 RVA positive specimens were subgrouped by their VP6 gene, and G-and P typed by PCR and/or sequencing of the VP7 and VP4 genes, respectively. Notably, we found a high prevalence of the unusual G6P[6]SGI strains (23%). It was the second most common constellation after G9P[8]SGII (32%); and followed by G1P[8]SGII (20%) and G2P[4]SGI (9%). We also detected a G8P[6]SGI strain, for the first time in Burkina Faso. The intra-genetic diversity was high for the VP4 gene with two subclusters within the P[8] genotype and three subclusters within the P[6] genotype which were each associated with a specific G-type, thereby suggesting a genetic linkage. The G6P[6]SGI and other SGI RVA strains infected younger children as compared to SGII strains (p < 0.05). To conclude, in this study we observed the emergence of unusual RVA strains and high genetic diversity of RVA in remote rural areas of Burkina Faso. The results highlight the complexity of RVA epidemiology which may have implication for the introduction of rotavirus vaccines currently being evaluated in many African countries. Ó 2012 Elsevier B.V. All rights reserved.
1. Introduction Group A rotavirus (RVA) is the leading cause of severe acute gastroenteritis in infants and young children, responsible for approximately 453,000 deaths annually (Tate et al., 2011), with more than 230,000 of the deaths occurring in sub-Saharan Africa. Five of 10 countries with the highest mortality due to RVA diarrhea are located in Africa (Tate et al., 2011). The high burden of RVA disease in children has accelerated the development of vaccines (Parashar et al., 2006; Patel et al., 2008), and the World Health Organization (WHO) now recommends the inclusion of RVA vaccines in the national immunization program ⇑ Corresponding author. Address: Division of Molecular Virology, Department of Clinical and Experimental Medicine, Medical Faculty, Linköping University, SE-581 85 Linköping, Sweden. Tel.: +46 70 2239323; fax: +46 10 31375. E-mail address: [email protected] (J. Nordgren). 1 These authors are contributed equally to this study. 1567-1348/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.meegid.2012.08.014
in countries highly affected by RVA. Currently, two vaccines, Rotarix and RotaTeq, are licensed and available in many countries. Although the vaccines have demonstrated high efficacy in industrialized countries, vaccine trials from African countries have shown a lesser efficacy (Madhi et al., 2010; Sow et al., 2012). The reasons for this are yet unknown, but a different and more diverse genetic setup of RVA strains could be one of the reasons. The RVA strain diversity in Africa is higher than from industrialized countries and also have a high degree of animal or animal derived genes (Nordgren et al., 2012; Todd et al., 2010). Some other major differences include the relatively high prevalence of the P[6] genotype as well as a high detection rate of G8 in certain countries (Armah et al., 2010; Nordgren et al., 2012; Page et al., 2010b; Todd et al., 2010). This highlights the need for understanding of the molecular epidemiological patterns of human and animal RVA strains in the continent in order to assess factors that could be important for vaccine efficacy. Few studies regarding the molecular epidemiology of RVA in rural areas in Africa have been performed (Armah et al., 2003),
J. Nordgren et al. / Infection, Genetics and Evolution 12 (2012) 1892–1898
and many RVA strains still remain uncharacterized (Todd et al., 2010). In Burkina Faso, previous studies have observed a high burden of RVA disease in children with high RVA prevalence during the cold dry season December–February (Bonkoungou et al., 2010, 2011; Nitiema et al., 2011; Steele et al., 2010). Furthermore, in a previous study from the capital Ouagadougou, RVA strains were characterized by their G and P-types (Nordgren et al., 2012). In this study we investigated RVA strain diversity in two remote rural locations situated south-west (Boromo) and north (Gourcy) of Ouagadougou. The samples were collected during the same time frame as the previous RVA study from the capital (Nordgren et al., 2012), thus enabling a direct comparison between urban and rural settings in Burkina Faso during the same RVA season. Of note is the high prevalence (23%) of the highly unusual G6P[6] rotavirus strains. The obtained results give a more thorough understanding about the complexity of RVA epidemiology in Burkina Faso and the results are discussed in the relation to vaccine efficacy. 2. Materials and methods 2.1. Study population and specimens The study was conducted in two rural areas, in north (Gourcy, distance 140 km) and south-west (Boromo, distance 185 km) of the capital Ouagadougou in Burkina Faso during January and February 2010, which is the period with highest RVA burden in Burkina Faso. A total of 80 samples, 17 from Gourcy and 63 from Boromo were collected. These regions are tropical savannah areas and the main sources of income are subsistence farming, animal husbandry and small scale trade. Fecal specimens were collected from children <5 years of age who sought medical care for acute diarrheal illness, defined as >3 liquid stools over a 24 h period. 2.2. Clinical-epidemiological assessment Clinical information was obtained by reviewing the clinical records of each case. Information regarding the age, sex, hospitalization and clinical symptoms such as fever (temperature P 38 °C), vomiting and dehydration, and the characteristics of stool were recorded for each child. All children were clinically evaluated by general practitioners following a local adaptation of the World Health Organization (WHO) strategy for the management of diarrhea. 2.3. Rotavirus antigen detection All stool samples were analyzed for RVA antigens and adenovirus serotypes 40/41with an immunochromatograpic method (SD Bioline Rota/AdenoÒ; Standard Diagnostics, Inc., Kyonggi-Do, South Korea) following the manufacturer’s instructions. 2.4. Viral RNA extraction Viral RNA was extracted from 140 ll of 10% stool suspensions following the manufacturer’s instructions using a QIAamp Viral RNA Mini Kit (Qiagen, Hilden, Germany). A total of 60 ll of purified viral RNA was collected and stored at 70 °C until reverse transcription reaction was carried out.
1893
[pd(N)6], denatured at 97 °C for 5 min, and quickly chilled on ice for 2 min, followed by addition of one Illustra Ready-To-Go RTPCR bead (GE Healthcare, Uppsala, Sweden) and RNase-free water to a final volume of 50 ll. The RT reaction was carried out for 30 min at 42 °C to produce the cDNA. 2.6. Quantification and VP6 subgrouping by real-time PCR All rotavirus positive specimens were quantified and VP6 subgrouped by the use of a LUX real-time PCR as described previously (Nordgren et al., 2010). This real-time PCR uses labeled primers with different fluorophores for each VP6 subgroup and external plasmid standards for semi-quantification (Nordgren et al., 2010). 2.7. Rotavirus G and P genotyping Rotavirus G and P genotyping was performed using semi-nested type-specific multiplex PCR’s (Gentsch et al., 1992; Gouvea et al., 1990; Iturriza-Gomara et al., 2004), able to detect seven G-types (G1, G2, G3, G4, G8, G9, G10) and six P-types (P[4], P[6], P[8], P[9], P[10] and P[11]), respectively. PCR products were examined on an agarose gel with ethidium bromide staining and visualized under UV light. The RVA G and P genotypes were determined by their specific size on the agarose gel. 2.8. PCR and sequencing of VP4 and VP7 genes of the detected RVA strains Sequencing of the PCR-amplicons for the VP7-and VP4 genes was performed for all samples positive for RVA antigen using 4 consensus primers (VP7-F, VP7-R, Con-2 and Con-3) as previously described (Iturriza-Gomara et al., 2004; Nordgren et al., 2012). The following thermal cycling conditions were used: An initial denaturation step at 94 °C for 5 min followed by 40 cycles of amplification (30 s at 94 °C, 30 s at 50 °C and 1 min 30 s at 72 °C), with a final extension of 7 min at 72 °C. Nucleotide sequencing was performed by Macrogen Inc. (Seoul, Korea). The sequencing reaction was based on BigDye chemistry, using the same primers as in the PCR as sequencing primers. 2.9. Sequence analysis Multiple sequence alignment of the obtained nucleotide sequences were performed using the ClustalW algorithm with default parameters on the European Bioinformatics Institute server. A phylogenetic analysis of the alignment file was performed using the MEGA 5.05 software package. A tree was constructed using the neighbor-joining method and Kimura two-parameter model (Kimura, 1980; Saitou and Nei, 1987). The statistical significance of the relationships obtained was estimated by bootstrap resampling analysis (1000 replications). The assignment of genetic lineages within the G and P genotypes was done using the following references: (Arora and Chitambar, 2011; De Grazia et al., 2011; Martella et al., 2006; Mascarenhas et al., 2010; Phan et al., 2007a,b; Sharma et al., 2009). 2.10. Sequence submission The partial nucleotide sequences for the RVA VP4 and VP7 genes can be found under GenBank accession numbers JX154433– JX154544.
2.5. Reverse transcription 2.11. Statistical analysis Reverse transcription (RT) was essentially carried out as described previously (Bucardo et al., 2007). Briefly, 28 ll of dsRNA was mixed with 50 pmol of random hexadeoxynucleotides
Interval data (age) was analyzed using independent t-test with 2-tailed significance, using SPSS 19.0 (SPSS Inc., Chicago, IL, USA).
1894
J. Nordgren et al. / Infection, Genetics and Evolution 12 (2012) 1892–1898
3. Results 3.1. High prevalence of rotavirus in rural Burkina Faso The samples were collected during the cold dry season in Burkina Faso (January–February), which in earlier studies was shown to be the period with highest RVA burden in children under the age of 5 (Bonkoungou et al., 2010; Nitiema et al., 2011). Of the 80 fecal specimens analyzed, 56 (70%) were found positive for RVA antigen. RVA was detected in 47/63 (74.6%) children with diarrhea from Boromo and 9/17 (52.9%) from Gourcy. Among these RVA infected children, 13/56 (23.2%) were hospitalized and most cases 52/56 (92.9%) occurred among children less than 2 years of age. 3.2. High prevalence of the unusual G6P[6] rotavirus strains and the detection of a G8P[6] strain The G and P genotype distribution was determined by the multiplex nested PCR assay and sequencing of the VP4 and VP7 genes; the nested PCR assay data was only used when sequencing was not successful, mostly in samples with low viral load. In total 44 VP4 and 49 VP7 sequences with good sequence quality were obtained. The distribution of G and P types is presented in Table 1 and shows that, notably, 13 (23%) of the 56 RVA positive children were infected with G6P[6] RVA. Five different G-types were detected with the following prevalence rates: G9 (34%), G6 (23%), G1 (21%), G2 (11%) and G8 (2%). Among the P-genotypes, P[8] was the most prevalent with 54% detection rate, followed by P[6] (30%) and P[4] (9%). The G/P constellations most frequently detected were G9P[8] (32%), G6P[6] (23%), G1P[8] (20%) and G2P[4] (9%). Mixed G or P type infections were found in 4% (n = 2) of the total specimens, whereas 5% (n = 3) and 7% (n = 4) of the RVA positive specimens could not be assigned a G-type and P-type respectively (Table 1). When stratified according to VP6 subgroups, we observed that 35.7% of the specimens (n = 20) were SGI and 64.3% (n = 36) were SGII. All SGII positive specimens were associated with P[8] genotype, whereas all SGI positive specimens were associated with P[4] or P[6] genotypes with the exception of two RVA specimens which had P[6] and SGII specificities. These two specimens were of the constellation G2P[6]SGII and G9P[6]SGII. 3.3. The unusual G6P[6]SGI strains infected younger children as compared to other strains The mean age of children infected with RVA SGI strains was 9.4 ± 7.2 months, as compared to children infected with RVA SGII, which was 14.9 ± 8.1 months (p < 0.05). The unusual G6P[6] strains infected the youngest age group 7.2 ± 5.4 months, and the most
Table 1 Distribution of G and P genotypes of RVA strains detected among children with gastroenteritis in rural areas of Gourcy and Boromo, Burkina Faso during January and February 2010. Rotavirus genotype
G1 G2 G6 G8 G9 Gmixa Gnt Total nt = Not determined. a G1/G9.
No. (%) of strains P[4]
P[6]
P[8]
P[nt]
Total
0 5 0 0 0 0 0 5 (9)
1 1 13 1 1 0 0 17 (30)
11 0 0 0 18 0 1 30 (54)
0 0 0 0 0 2 2 4 (7)
12 (21) 6 (11) 13 (23) 1 (2) 19 (34) 2 (4) 3 (5) 56 (100)
common RVA strain G9P[8] infected the oldest age-group 16.1 ± 9.1 months (p < 0.05) (Table 2). When stratified according to VP6 subgroups, we observed that SGI RVA strains significantly infected younger children as compared to SGII RVA strains (9.4 ± 7.2 vs 14.9 ± 8.1 months). 3.4. Phylogenetic analysis of the VP4 and VP7 genes We analyzed the sequences of the VP4 (n = 44) and VP7 (n = 49) genes of all RVA positive specimens containing sufficient viral load and yielding high quality sequence data (Fig. 1A and B). The phylogenetic analysis demonstrated that the VP7 genes within each Gtype cluster are highly similar to each other and independent of P genotype association, with the exception of specimen 35, a G1P[8] RVA which is relatively more different (1.2–1.5% nt difference) compared to the other detected G1P[8] RVs (Fig. 1B). In contrast, for the VP4 genes we observed different subclusters within the same P-genotype cluster, which were associated with different G-types, demonstrating a more promiscuous behavior of the VP4 gene (Fig. 1A). Within the P[6] cluster, we observed 3 subclusters, for G9P[6], G8P[6] and G6P[6] RVA strains respectively, and in the P[8] cluster we observed two subclusters, corresponding to G9P[8] and G1P[8] RVs respectively. The only exception is the VP4 gene for specimen 16, a G1P[8] strain, for which the VP4 gene is identical to the VP4 genes associated with the G9P[8] genotypes, indicating a reassortment event between G1P[8] and G9P[8] during the RVA season. Due to the shorter stretch of nt sequence obtained (500 bp) it was however omitted from Fig. 1A. The VP7 genes of the G9P[8] strains belonged to sublineage IIId, and are similar to G9 stains (GH1319) isolated in northern Ghana from 1998 to 2000 (99.2–99.5% nt identity) (Armah et al., 2003), and to strain 6222LP detected in South Africa in 1999 (97.7– 98.1% nt identity) (Page et al., 2010a). The VP7 genes of the G1P[8] and G1P[6] genotypes belonged to sublineage Ia, similar to the CMH042 strain detected in Thailand 2004 (99.0–99.1% nt identity) and to G1P[8] RVA strain SA4799DGM (98.5–98.8% nt identity) detected in South Africa in 2004. The VP4 gene associated with both the G1 and G9 genotypes belonged to lineage P[8]-III, but appeared in two different clusters (2.3–3.0% nt difference, Fig. 1A). The VP4 gene of the G9P[8] strains were highly similar to the Dhaka25-02 strain (98.9–99.3% nt identity), whereas the VP4 gene of the G1P[8] strains were highly similar to the BE00021 strain (99.3–99.4% nt identity). The VP7 genes of the G2P[4] strains, belonged to sublineage IIa, and were similar to G2 strains detected in Sierra Leone in 2005 (SL276, 98.6–98.8% nt identity) (Jere et al., 2011b), and to CH-128 (Trinh et al., 2010) detected in China in 2003 (98.2–98.4% nt identity) The P[4] gene of the G2P[4] RVA strains belonged to lineage P[4]-V and were also similar to the G2P[4] strains (99.1–99.3% nt identity to strain SL244) detected in Sierra Leone in 2005.
Table 2 Age of children infected with rotavirus stratified by VP6 subgroup and the three most common G, P and VP6 type constellations observed in this study
a b
Number
Age (months)
SGI SGII p-Valuea
20 36 0.013
9.4 ± 7.2 14.9 ± 8.1
G6P[6]SGI G1P[8]SGII G9P[8]SGII p-Valueb
13 11 18 0.005
7.2 ± 5.4 13.1 ± 8.1 16.1 ± 9.1
SGI vs SGII. G6P[6]SGI vs G1P[8]SGII and G9P[8]SGII.
J. Nordgren et al. / Infection, Genetics and Evolution 12 (2012) 1892–1898
1895
Fig. 1. (A) Phylogenetic analysis of sequences of the VP8⁄ subunit of the VP4 gene (nt 140–751). (B) VP7 gene (nt 135–796) with reference strains. The G and P types of the Burkina Faso strains are added after the isolation number and are marked with filled circles. Rotavirus strains from the capital Ouagadougou are added as a comparison and are marked with open circles. The brackets indicate the lineage of the Burkina Faso strains (in bold) and genotype of all strains, respectively.
1896
J. Nordgren et al. / Infection, Genetics and Evolution 12 (2012) 1892–1898
We further observed that the unusual G6P[6] RVA strains were nearly identical to the G6P[6] RVs detected in a previous study in Ouagadougou, Burkina Faso (Nordgren et al., 2012). These strains were related to the first human G6P[6] detected, the B1711 strain previously detected in Belgium 2002 in a child arriving from Mali (98.2–98.5% nt identity for the VP7 gene)(Rahman et al., 2003). The VP4 genes of the G6P[6] strains, belonged to sublineage Ia (96.7–97.0% nt identity to the G6P[6] strain B1711), and nearly identical to the G6P[6] strains detected in Ouagadougou. The VP4 gene of the G9P[6] and G8P[6] strains were different compared to the VP4 gene of the G6P[6] strains (97.0–97.3% and 98.7–99.0% nt identity, respectively), but belonged to the same sublineage. The VP7 gene of the G8P[6] strain, belonging to lineage G8d, was related to the DRC88 strain (98.5% nt identity), detected in the Democratic Republic of Congo in 2003, and to RVA strains SL285 and KY1646 detected in Sierra Leone 2005 (98.8% nt identity) and Kenya 1999 (98.9% nt identity), respectively (Jere et al., 2011b; Matthijnssens et al., 2006; Page et al., 2010b).
4. Discussion In this study we have characterized RVA strains from two rural areas of Burkina Faso. These settings are remote, and relatively few studies regarding molecular characterization of RVs have previously been performed from similar settings in Africa. The major finding was the high prevalence of the rare G6P[6] strains, as well as the detection of a G8P[6] strain for the first time in Burkina Faso. We observed a high diversity of RVA strains, in total 8 G/P type constellations were observed during only two months. The majority of RVA strains belonging to the G9P[8]SGII genotype (32%), with the unusual G6P[6]SGI strains being the second most common genotype in this study (23%), followed by G1P[8]SGII (20%). The high prevalence of the unusual G6P[6] strains in rural Burkina Faso is noteworthy. Previously, this strain had only been observed once in Belgium from a child arriving from Mali in 2003 (Rahman et al., 2003), and subsequently in high prevalence (13%) in a study from the capital of Burkina Faso, Ouagadougou (Nordgren et al., 2012) during a similar time-frame as this study. These findings combined may suggest that the G6P[6] strain is emerging as an important pathogen in pediatric RVA diarrhea in Burkina Faso. In this study, we first G and P typed all strains using a nested PCR approach commonly used worldwide (Iturriza-Gomara et al., 2004). When subsequently all VP4 and VP7 genes were sequenced, we observed a number of discrepancies. Three of the G6P[6] strains (n = 13) were initially PCR-typed as G1P[6] strains with the nested PCR approach (data not shown), which suggests that the prevalence of G6P[6] strains could previously have been underestimated at least in Africa. The G1 primer matched in 12 of 22 nt at the corresponding position in the VP7 gene of the G6 strains, but a low annealing temperature as well as high viral load could have led to the primer mismatch observed. However, a presence of a mixed G1/G6 infection cannot be ruled out since this would not be apparent with sanger sequencing if the different genotypes are present in highly varying concentrations. The fact that a high percentage of RVA strains in Africa remain untypeable suggests that novel genotypes such as G6P[6] could be more widespread than previously assumed, all suggesting that sequencing of RVA strains from Africa is most warranted. We observed significant differences in age-profiles, with SGI strains, especially G6P[6]SGIs infecting younger age-groups (7.2 ± 5.4 months) as compared to SGII strains; G1P[8]SGII (13.1 ± 8.1 months) and G9P[8]SGII (16.1 ± 9.1 months). Interestingly, in a study from the capital Ouagadougou during the same time frame (Nordgren et al., 2012), the opposite pattern was observed. In
Ouagadougou, the G6P[6]SGI and other SGI strains significantly infected an older age-group as compared to G9P[8]SGII strains. We compared the age-profiles of the G6P[6] strains between these two studies and observed a statistically significant difference (p < 0.05). The reason behind these differences in age-profiles between the urban and rural settings during the same time period remains to be elucidated and confirmed in more studies with more samples. Differences in RVA type specificities in different parts of the same country have been reported (Kirkwood et al., 2011). The rural areas in this study are rather remote and far from the capital; one possibility could be that during the previous RVA season, RVA strains with different G, P and VP6 type specificities were circulating in the urban and rural settings respectively, thus creating a different immunological response in the children during this RVA season. Another explanation could be a common source of infection of G6P[6] or other genotypes into a setting having a distinct ageprofile (e.g. many younger children). However, in the study from the capital, the G6P[6] strains were detected in children from many different parts of the city (data not shown), and for the rural community, the same age profiles were seen both in Gourcy and Boromo which are 170 km apart. The VP6 gene, coding for the major immunogenic protein of the RVA virion, is often not genotyped during epidemiological investigations which focuses more on G and P types. The clear clinicalepidemiological differences between VP6 types, e.g. age-profiles, that we have observed between children in this and in a previous study (Nordgren et al., 2012) suggest that it should be more investigated in surveys of RVs in the future. The G6P[6] strains were observed at different time points in the rural settings; the first G6P[6] strain was observed on the 13th of January 2010 in Gourcy, whereas the first G6P[6] in Boromo was observed the 28th of January 2010. The first G6P[6] in the capital Ouagadougou in the previous study (Nordgren et al., 2012) was observed the 14th of January 2010, similar to that of Gourcy. Moreover, the frequency of different G/P and VP6 constellations varied between the urban and rural settings, giving further strength to the assumption of different epidemiological patterns of RVs in different geographical areas of Burkina Faso. We observed low frequencies of mixed infections as compared to other RVA studies from Africa (Todd et al., 2010). This could be explained by the fact that G and P types in this study were primarily assigned by sequencing all VP7 and VP4 genes from the RVA positive samples, and only with the nested multiplex PCR assay when sequencing PCR was not possible due to low viral load in the samples. As observed previously (Nordgren et al., 2012) the multiplex nested PCR assays could overestimate the amount of mixed infections due to mispriming, hence presence of mixed RVA infections could preferably be determined by deep sequencing (Jere et al., 2011a). We performed a phylogenetic analysis of the sequenced VP4 and VP7 genes (Fig. 1A and B). We observed many sub-clusters when analyzing the VP4 genes, each belonging to a specific G-type. This was not the case for the VP7 genes, which were highly identical to each other independently of which P genotype they were associated with. For example, within the P[6] cluster we observed three subclusters, each belonging to G9P[6], G8P[6] and G6P[6] RVA strains respectively. Of note, sample 20, being genotype G9P[6] and representing a separate subcluster of P[6], was one of the only two samples that had P[6] and SGII specificities. Also, in the P[8] cluster we observed two subclusters, corresponding to G9P[8] and G1P[8]. However, the VP7 genes of the same G-type belonging to different P-types were highly identical. This indicates a genetic linkage between VP4 subclusters and different G genotypes, or even VP6 subgroups. We compared the rural rotavirus strains to those from the study in the capital (Nordgren et al.,
J. Nordgren et al. / Infection, Genetics and Evolution 12 (2012) 1892–1898
2012), and we observed that they were highly identical, thus indicating that the same strains were circulating in both the urban and rural settings, although in different proportions and in slightly different time frames (Fig. 1A and B). The VP7 gene of the predominant G9P[8] strains detected in this study were highly identical to G9 RVA strains detected in the neighboring country of Ghana (Armah et al., 2003) in the end of the 1990ths, demonstrating that this particular G9 has been present and circulating in West Africa for a long time. The emergence and continued high prevalence of G9 strains in Africa has frequently been reported (Armah et al., 2003, 2010), and the results of this study show that they play major role in pediatric rotavirus diarrhea in rural Burkina Faso. The VP7 and VP4 genes of the G2P[4] strains were similar to G2P[4] strains isolated in Sierra Leone in West Africa in 2005. In previous studies from Brazil, G2P[4] RVA strains were observed in high prevalence in a pre-vaccinated pediatric population (Gurgel et al., 2007). The G2 and P[4] types are not present in the monovalent G1P[8] vaccine, and it has been suggested that the vaccine shows a lesser efficiency to this type (Gurgel et al., 2007). During this RVA season however, the G2P[4] RVs were not highly prevalent and a study from Burkina Faso the previous year showed a similar prevalence of G2 strains (Bonkoungou et al., 2011). We further observed that the unusual G6P[6] was nearly identical to G6P[6] strains from the capital Ouagadougou, Burkina Faso, and similar to the B1711 strain previously isolated in Belgium in a child arriving from Mali (Rahman et al., 2003). The G8P[6] strain, detected for the first time in Burkina Faso, was closely related with the DRC88 and SL285 strains detected in the Democratic Republic of Congo and Sierra Leone 2003 and 2005, respectively (Jere et al., 2011b; Matthijnssens et al., 2006). G8P[6] RVs are reported frequently from several countries in Africa. During a survey in 2005 in the nearby country of Sierra Leone in West Africa, 18% of the RVA strains belonged to G8 (Jere et al., 2011b). The G6P[6] and G8P[6] share many similarities, they are both assumed to have emerged in the human population from relatively recent reassortment events from bovine G8 and G6 strains and then adapted to humans (Rahman et al., 2003). Furthermore, molecular characterization of the G6P[6] strains detected in Ouagadougou, Burkina Faso showed high homology to the G8P[6] strain DRC86 (Matthijnssens et al., 2006; Nordgren et al., 2012). The emergence and high prevalence of these G8 and G6 RVA strains in Africa raises concerns regarding the efficacy of the current vaccines for these particular genotypes, which are rare or nonexistent in other continents, and should be evaluated in the future. To conclude, this study has demonstrated a high diversity of RVA and high prevalence of the rare G6P[6] strains in remote rural areas of Burkina Faso. We observed clinical-epidemiological differences between RVA strains and also between urban and rural settings during the same time-period. This study provides new knowledge about the complexity of RVA epidemiology in Burkina Faso which can be useful for the future evaluation and/or introduction of RVA vaccines. Role of the funding source The study was supported by Swedish Research Council (Grant 10392). The study sponsor had no role in the design and conduct of the study. Acknowledgments We thank the parents and guardians of the enrolled children and the Health District Authorities in Boromo and Gourcy for their cooperation.
1897
References Armah, G.E., Steele, A.D., Binka, F.N., Esona, M.D., Asmah, R.H., Anto, F., Brown, D., Green, J., Cutts, F., Hall, A., 2003. Changing patterns of rotavirus genotypes in Ghana: emergence of human rotavirus G9 as a major cause of diarrhea in children. J. Clin. Microbiol. 41, 2317–2322. Armah, G.E., Steele, A.D., Esona, M.D., Akran, V.A., Nimzing, L., Pennap, G., 2010. Diversity of rotavirus strains circulating in West Africa from 1996 to 2000. J. Infect. Dis. 202 (Suppl.), S64–S71. Arora, R., Chitambar, S.D., 2011. Full genomic analysis of Indian G1P[8] rotavirus strains. Infect. Genet. Evol. 11, 504–511. Bonkoungou, I.J., Damanka, S., Sanou, I., Tiendrebeogo, F., Coulibaly, S.O., Bon, F., Haukka, K., Traore, A.S., Barro, N., Armah, G.E., 2011. Genotype diversity of group A rotavirus strains in children with acute diarrhea in urban Burkina Faso, 2008–2010. J. Med. Virol. 83, 1485–1490. Bonkoungou, I.J., Sanou, I., Bon, F., Benon, B., Coulibaly, S.O., Haukka, K., Traore, A.S., Barro, N., 2010. Epidemiology of rotavirus infection among young children with acute diarrhoea in Burkina Faso. BMC Pediatr. 10, 94. Bucardo, F., Karlsson, B., Nordgren, J., Paniagua, M., Gonzalez, A., Amador, J.J., Espinoza, F., Svensson, L., 2007. Mutated G4P[8] rotavirus associated with a nationwide outbreak of gastroenteritis in Nicaragua in 2005. J. Clin. Microbiol. 45, 990–997. De Grazia, S., Martella, V., Rotolo, V., Bonura, F., Matthijnssens, J., Banyai, K., Ciarlet, M., Giammanco, G.M., 2011. Molecular characterization of genotype G6 human rotavirus strains detected in Italy from 1986 to 2009. Infect. Genet. Evol. 11, 1449–1455. Gentsch, J.R., Glass, R.I., Woods, P., Gouvea, V., Gorziglia, M., Flores, J., Das, B.K., Bhan, M.K., 1992. Identification of group A rotavirus gene 4 types by polymerase chain reaction. J. Clin. Microbiol. 30, 1365–1373. Gouvea, V., Glass, R.I., Woods, P., Taniguchi, K., Clark, H.F., Forrester, B., Fang, Z.Y., 1990. Polymerase chain reaction amplification and typing of rotavirus nucleic acid from stool specimens. J. Clin. Microbiol. 28, 276–282. Gurgel, R.Q., Cuevas, L.E., Vieira, S.C., Barros, V.C., Fontes, P.B., Salustino, E.F., Nakagomi, O., Nakagomi, T., Dove, W., Cunliffe, N., Hart, C.A., 2007. Predominance of rotavirus P[4]G2 in a vaccinated population, Brazil. Emerg. Infect. Dis. 13, 1571–1573. Iturriza-Gomara, M., Kang, G., Gray, J., 2004. Rotavirus genotyping: keeping up with an evolving population of human rotaviruses. J. Clin. Virol. 31, 259–265. Jere, K.C., Mlera, L., O’Neill, H.G., Potgieter, A.C., Page, N.A., Seheri, M.L., van Dijk, A.A., 2011a. Whole genome analyses of African G2, G8, G9, and G12 rotavirus strains using sequence-independent amplification and 454(R) pyrosequencing. J. Med. Virol. 83, 2018–2042. Jere, K.C., Sawyerr, T., Seheri, L.M., Peenze, I., Page, N.A., Geyer, A., Steele, A.D., 2011b. A first report on the characterization of rotavirus strains in Sierra Leone. J. Med. Virol. 83, 540–550. Kimura, M., 1980. A simple method for estimating evolutionary rates of base substitutions through comparative studies of nucleotide sequences. J. Mol. Evol. 16, 111–120. Kirkwood, C.D., Boniface, K., Barnes, G.L., Bishop, R.F., 2011. Distribution of rotavirus genotypes after introduction of rotavirus vaccines, Rotarix(R) and RotaTeq(R), into the National Immunization Program of Australia. Pediatr. Infect. Dis. J. 30, S48–53. Madhi, S.A., Cunliffe, N.A., Steele, D., Witte, D., Kirsten, M., Louw, C., Ngwira, B., Victor, J.C., Gillard, P.H., Cheuvart, B.B., 2010. Effect of human rotavirus vaccine on severe diarrhea in African infants. New Engl. J. Med. 362, 289–298. Martella, V., Banyai, K., Ciarlet, M., Iturriza-Gomara, M., Lorusso, E., De Grazia, S., Arista, S., Decaro, N., Elia, G., Cavalli, A., Corrente, M., Lavazza, A., Baselga, R., Buonavoglia, C., 2006. Relationships among porcine and human P[6] rotaviruses: evidence that the different human P[6] lineages have originated from multiple interspecies transmission events. Virology 344, 509–519. Mascarenhas, J.D., Lima, C.S., de Oliveira, D.S., Guerra Sde, F., Maestri, R.P., Gabbay, Y.B., de Lima, I.C., de Menezes, E.M., Linhares Ada, C., Bensabath, G., 2010. Identification of two sublineages of genotype G2 rotavirus among diarrheic children in Parauapebas, Southern Para State, Brazil. J. Med. Virol. 82, 712–719. Matthijnssens, J., Rahman, M., Yang, X., Delbeke, T., Arijs, I., Kabue, J.P., Muyembe, J.J., Van Ranst, M., 2006. G8 rotavirus strains isolated in the Democratic Republic of Congo belong to the DS-1-like genogroup. J. Clin. Microbiol. 44, 1801–1809. Nitiema, L.W., Nordgren, J., Ouermi, D., Dianou, D., Traore, A.S., Svensson, L., Simpore, J., 2011. Burden of rotavirus and other enteropathogens among children with diarrhea in Burkina Faso. Int. J. Infect. Dis. 15, 646–652. Nordgren, J., Bucardo, F., Svensson, L., Lindgren, P.E., 2010. Novel light-uponextension real-time PCR assay for simultaneous detection, quantification, and genogrouping of group A rotavirus. J. Clin. Microbiol. 48, 1859–1865. Nordgren, J., Nitiema, L.W., Sharma, S., Ouermi, D., Traore, A.S., Simpore, J., Svensson, L., 2012. Emergence of unusual G6P[6] rotaviruses in children with acute gastroenteritis in Burkina Faso, 2009–2010. Emerg. Infect. Dis. 18, 589–597. Page, N., Esona, M., Armah, G., Nyangao, J., Mwenda, J., Sebunya, T., Basu, G., Pyndiah, N., Potgieter, N., Geyer, A., Steele, A.D., 2010a. Emergence and characterization of serotype G9 rotavirus strains from Africa. J. Infect. Dis. 202 (Suppl.), S55–S63. Page, N., Esona, M., Seheri, M., Nyangao, J., Bos, P., Mwenda, J., Steele, D., 2010b. Characterization of genotype G8 strains from Malawi, Kenya, and South Africa. J. Med. Virol. 82, 2073–2081. Parashar, U.D., Gibson, C.J., Bresse, J.S., Glass, R.I., 2006. Rotavirus and severe childhood diarrhea. Emerg. Infect. Dis. 12, 304–306.
1898
J. Nordgren et al. / Infection, Genetics and Evolution 12 (2012) 1892–1898
Patel, M.M., Widdowson, M.A., Glass, R.I., Akazawa, K., Vinje, J., Parashar, U.D., 2008. Systematic literature review of role of noroviruses in sporadic gastroenteritis. Emerg. Infect. Dis. 14, 1224–1231. Phan, T.G., Okitsu, S., Maneekarn, N., Ushijima, H., 2007a. Evidence of intragenic recombination in G1 rotavirus VP7 genes. J. Virol. 81, 10188–10194. Phan, T.G., Okitsu, S., Maneekarn, N., Ushijima, H., 2007b. Genetic heterogeneity, evolution and recombination in emerging G9 rotaviruses. Infect. Genet. Evol. 7, 656–663. Rahman, M., De Leener, K., Goegebuer, T., Wollants, E., Van der Donck, I., Hoovels, L.V., Van Ranst, M., 2003. Genetic characterization of a novel, naturally occurring recombinant human G6P[6] rotavirus. J. Clin. Microbiol. 41, 2088. Saitou, N., Nei, M., 1987. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol. Biol. Evol. 4, 406–425. Sharma, S., Paul, V.K., Bhan, M.K., Ray, P., 2009. Genomic characterization of nontypeable rotaviruses and detection of a rare G8 strain in Delhi, India. J. Clin. Microbiol. 47, 3998–4005. Sow, S.O., Tapia, M., Haidara, F.C., Ciarlet, M., Diallo, F., Kodio, M., Doumbia, M., Dembele, R.D., Traore, O., Onwuchekwa, U.U., Lewis, K.D., Victor, J.C., Steele,
A.D., Neuzil, K.M., Kotloff, K.L., Levine, M.M., 2012. Efficacy of the oral pentavalent rotavirus vaccine in Mali. Vaccine 30 (Suppl. 1), A71–A78. Steele, A.D., Page, N., de Beer, M., Sawadogo, S., 2010. Antigenic and molecular characterization of unusual rotavirus strains in Burkina Faso in 1999. J. Infect. Dis. 202 (Suppl.), S225–S230. Tate, J.E., Burton, A.H., Boschi-Pinto, C., Steele, A.D., Duque, J., Parashar, U.D., 2011. 2008 estimate of worldwide rotavirus-associated mortality in children younger than 5 years before the introduction of universal rotavirus vaccination programmes: a systematic review and meta-analysis. Lancet Infect. Dis. 12, 136–141. Todd, S., Page, N.A., Duncan Steele, A., Peenze, I., Cunliffe, N.A., 2010. Rotavirus strain types circulating in Africa: review of studies published during 1997– 2006. J. Infect. Dis. 202 (Suppl.), S34–S42. Trinh, Q.D., Pham, N.T., Nguyen, T.A., Phan, T.G., Yan, H., Hoang le, P., Khamrin, P., Maneekarn, N., Li, Y., Okitsu, S., Mizuguchi, M., Ushijima, H., 2010. Sequence analysis of the VP7 gene of human rotaviruses G2 and G4 isolated in Japan, China, Thailand, and Vietnam during 2001–2003. J. Med. Virol. 82, 878–885.