Genetic characterization of group A rotavirus in children with acute gastroenteritis in Kericho County Referral Hospital, Kenya

¹School of Science and Technology, University of Kabianga, Kericho, Kenya, ²Centre for Virus Research, Kenya Medical Research Institute (KEMRI), Nairobi, Kenya

^&Corresponding author
Beth Khayeli Nyamanga, School of Science and Technology, University of Kabianga, Kericho, Kenya

Introduction    

Acute gastroenteritis amongst children below 5 years is normally attributed to rotavirus infections [1]. A rotavirus particle is comprised of a capsid enclosing a dual-stranded ribonucleic acid (dsRNA), a characteristic found in the family Reoviridae [2]. Rotavirus ribonucleic acid (RNA) comprises 11 gene segments bounded by three concentric shells. Rotavirus infections are at least experienced once in most children around the world before or at the age of five years old. As infections keep recurring, immunity develops thus adults are least affected. Between 1996 and 2005 five strains of rotavirus “A” accounted for 90% of infections in children five years and below, that is genotype G1- G4 and G9 in the US. As of 2016 acute gastroenteritis due to rotavirus resulted in the death of 128500 youngsters under the age of 5 years of which over 100,000 happened in sub-Saharan Africa [3]. Group A of rotavirus contributes to more than 90% of infections thus making it the most prevalent in humans. Rotavirus infections of Group B strains are rare but associated with infections in Asia and China [4]. Children in Kenya have exhibited Group A, B, and C species of rotavirus strains [5]. It has been shown that there is a great variety of strains of rotavirus present in Africa [6]. Rotavirus is stable and remains viable in a locality indefinitely if the area is not disinfected. It is spread through the oral-fecal pathway. The rotavirus particles invade and destroy cells of the epithelial layer that lines the small intestine resulting in gastroenteritis [7]. The World Health Organization Strategic Advisory Group of Experts recommended the assimilation of the rotavirus vaccine into the vaccination programs for youngsters aged five years and under in 2009 [8]. In Kenya, this vaccination was introduced in 2014. Rotarix Vaccine was Kenya´s vaccine of choice administered in all public health facilities twice, at eight weeks and 16 weeks.

Globally, there is a general reduction in the number of hospitalizations more so, in first-world countries that experience fewer infant mortalities due to their effective immunization programs, compared to third-world countries that have higher infant mortalities [9]. In Kenya, a 15.2% prevalence had previously been observed confirming a higher prevalence within the age range of 13-24 months during the cold and wet months [10]. On the contrary, the areas covered by vaccines are less likely to be affected owing to the socio-economic, environmental, and cultural beliefs in addition to implementation challenges. This has led to great variations in existing and re-emerging rotavirus strains particularly, in children who had already been vaccinated [11]. This raises concern on whether the strain in the Rotarix vaccine which is a single rotavirus variant of live weakened human 89-12 (type G1P1A [8]) evoked an efficient broad defense, against additional strains that were not part of the vaccine. Hence there is a necessity for constant investigation of rotavirus strains so as to guide the efficacy of the vaccines in use. The effects of vaccine introduction in Kenya have been documented [12], but not to the extent of including its effect on the existing strains, demographics, and socio-economic aspects of agricultural communities especially one such as in Kericho county. It is with this view that the genetic variations and frequency of group A rotaviruses in children of five years and below exhibiting acute gastroenteritis at Kericho County Referral Hospital were assessed. The county experiences extreme cold and wet climate which has been attributed to high transmission of infectious diseases.

Despite the inclusion of Rotarix in the National Vaccination Program there are bound to be some effects on the levels of morbidity, mortality, and the types of strains existing in a population after vaccine introduction. This might be due to environmental influences. There is seasonality in rotavirus infections in Kenya. These infections increase in frequency such as between January to March as well as from June to September [10]. Rotavirus strains are bound to also be affected since most viruses tend to evolve in a way that aims to enable them to evade the immune system. Most rotavirus infections are normally not diagnosed since the clinical spectrum of signs and symptoms are like other gastroenteritis-related infections. Vaccine introduction is also affected by the population it has been introduced to. This also includes the prevalence of other health conditions, social economic factors, education, cultural beliefs, sanitation, and environmental interactions that the population experiences and may likely influence the change in the rotavirus gene strain. The purpose of this study was to determine the prevalent strains existing among children visiting Kericho County Hospital with acute gastroenteritis. The general objective of the study was to determine the prevalence and genetic diversity of group A rotavirus in children with acute gastroenteritis attending Kericho County Referral Hospital. The specific objectives were to determine the prevalence of gastroenteritis caused by rotavirus in children of five years and below in Kericho County Referral Hospital and to determine the genetic diversity of circulating rotavirus strains amongst children of five years and below in Kericho County Referral Hospital.

Methods     

The study design: this was a hospital-based cross-sectional study. The study site was Kericho County Referral Hospital in Kericho County South of Rift Valley.

Participant selection: the study participants consisted of children aged five years and below, visiting the paediatric clinic, exhibiting symptoms of acute gastroenteritis at the hospital. They were selected using consecutive sampling until the required sample size was reached. Vaccination against rotavirus was confirmed by the immunization booklet or word of mouth from parent/ guardian. The consent was obtained from the guardians in written form with an outline of the process the samples would undergo for the research.

Sample size estimation: in prevalence studies, sample size formula for sample size determination was determined as per calculations by Charan & Biswas, 2013 [13]. A 95% Confidence Interval set at 1.96 and an assessed prevalence for rotavirus infection amid children in Kenya of 14.5% as calculated and observed by Muendo et al. 2018 [14]. The final sample size obtained was 200 participants.

Sample collection: stool samples were collected from patients presenting symptoms of gastroenteritis. They were frozen at -20°C waiting for transportation in cooler boxes to Kenya Medical Research Institute. Laboratory work was carried out with Biosafety Level 2 procedures in line with the WHO Biosafety manual. The inclusion criterion was children below 5 years showing symptoms of acute gastroenteritis. The exclusion criterion was bloody diarrhea and children above the age of 5 years old.

Specimen processing: a small amount (roughly 10gm) of solid fecal matter or 700µl of loose fecal aliquots were collected in cryovials and put in storage at low temperatures of -20°C pending analysis. The fecal aliquots obtained from patients with symptoms of gastroenteritis were used to prepare 1ml of 10% (w/v) fecal suspension in phosphate-buffered saline (PBS), balanced salt solution (BSS), M199, or 0.01M Tris solution (pH 7.5, 14.5mM NaCl, 10mM CaCl2). This was vortexed and clarified through centrifugation. The supernatant formed was collected and stored at -70°C for the longer term [15]. Rotavirus Enzyme-Linked Immunosorbent Assay Kit was used to confirm Group A rotavirus antigens in the collected samples. The procedure was carried out as per the instructions in the WHO manual for Rotavirus detection and Characterization methods. The results were read spectrophotometrically at 450nm. The Negative Control value was less than 0.150 absorbance units. The Positive Control value was greater than 0.500 absorbance units [15].

The positive samples further underwent Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS PAGE). It first entails the removal of Rotavirus RNA from the fecal suspension through the use of Phenol- Chloroform. This process was then followed by Silver Staining of the Double Strand RNA in gels through the use of Silver Nitrate which will assist in viewing the RNA bands of the rotavirus genome. This will assist in finding out whether the strains obtained are long or short electropherotypes [15]. The remaining fecal aliquots of positive samples had to undergo RT-PCR amplification by first extracting rotavirus RNA using QIAGEN Kit. The use of the QIAGEN kit combines the selective binding characteristics of a silica-based membrane with the rapidity of microspin and is highly appropriate for the instantaneous processing of multiple samples. This is followed by RT-PCR for G genotyping as per the World Health Organisation protocols made available by the West African Regional Rotavirus Laboratory. RT-PCR of rotavirus dsRNA included: dsRNA denaturation, Reverse transcription, and amplification of the newly formed cDNA. Consensus primers sBeg 9 (GGCTTTAAAAGAGAGAATTTC)- position 1-21 and End 9 (GGTCACATCATACAATTCTAATCTAAG) - position 1062-1036 for rotaviruses that belong to group A [16]. sBeg and End 9 will be used in the first cycles of RT-PCR amplification. In RT-PCR for P Genotyping, VP4 genotype amplification was carried out using consensus primer con3 and a mixture of specific primers 1T-1, 2T-1, 3T-1, 4T-1 and 5T-1 to define rotavirus VP4 genotypes [17]. Positive and negative controls were included.

The PCR fragments were run on 2% TAE agarose gel at 100 volts with suitable molecular weight markers to define the genotype of the rotavirus strain [15]. In verification, sequencing was done on genotype-specific yields or a piece of the VP7 or VP4 gene once amplification is completed. The ability to check for infections through purification and sequencing of products of diverse sizes sequestered from agarose gel was an added advantage of sequencing genotype-specific PCR yields [18]. As for the VP7 gene, a range of consensus primer pairs were defined together with beg9/end9 and VP7-F/VP7-R and degenerate versions, 9con1/9con2, and 9con1L/VP7-R deg. Consensus primers for VP4 gene fragments included con2/con3, Hum Com5/ Hum Com3, and VP4-F/VP4-R. The PCR conditions consisted of denaturation which took place at 95°C for a duration of five minutes and this included 40 cycles, next was denaturation at 94°C for a minute and annealing at 55°C for a duration of one minute. The last step was elongation at 72°C for one minute, and afterward a last extension step at 72°C for ten minutes. Once sequencing is done, the strain genotypes are identified by matching the genes of strains recognized as VP4 or VP7 types by the Gen Bank database 40.

Data analysis

The circulating strains of rotavirus were characterized and the results were compared to previous data collected from the populations before the rotavirus vaccine was introduced. Determination of demographic characteristics of children below five years presenting with rotavirus-related gastroenteritis was done by finding the percentage of the infected cases among the different ages. This was compared to previous studies carried out in Kenya on the same to enable finding out if there is a shift in the ages that normally experience rotavirus infections. This will enable this study to pinpoint the ages vulnerable to infections due to rotavirus [19]. In the determination of the occurrence of rotavirus-related gastroenteritis among children below five years, a percentage was calculated from the number of rotavirus-positive samples against the sum total of fecal samples collected during the study. This was also compared to previously done hospital cross-sectional studies in other hospitals. Determination of genetic diversity of circulating strains of rotavirus among children five years and below, RNA extraction, Reverse Transcription, G and P-typing of Rotaviruses was done followed by sequencing of the VP7 and VP4 gene of the rotavirus positive strains. This was later followed by sequence and phylogenetic analysis based on VP4 and VP7 Genes [20].

Ethical considerations: the approval to carry out the research was obtained from the University of Kabianga Institutional Ethics Review Committee, Ethics Clearance Number: IERC/2020/005. The research permit was obtained from the National Commission of Science, Technology, and Innovation, Research License No: NACOSTI/P/21/9085. Permission was also obtained from the County commissioner of Kericho County, the Director of Education, and the Director of Health. Permission was also obtained from the Hospital's administration and Ethics committee. The study was explained to the participants before seeking written informed consent from them. Assistance was also rendered to the guardians who were unable to read. Participation was voluntary and respondents were liberal to withdraw at any stage of the study. Confidentiality was ensured by not having any sort of identification on the data collection tools.

Results     

Sociodemographic characteristics of the participants and their guardians: Table 1, a total of 200 children participated in the study. Of the 200 samples, 106 (53%) were from males whereas 94 (47%) were from females. The most affected age was the 34-38 age group at 26.5%. The least affected age groups were 19-23 (3%) followed by those of less than 8 months (3.5%). The most affected demographics were children with parents or guardians who had secondary school education (51%). The most affected social economic status was the housewives (46.5%).

Prevalence of rotavirus among the participants: twenty-three samples (23%) tested positive for rotavirus infection (11.5%). The rates of infection were as shown in Figure 1. Rotavirus RNA electrophoretic patterns were observed in these samples as 11 segments. Twenty-two (96%) of the isolated strains were long strains except for one (4%) which did not exhibit the segments. The short electropherotypes were not found in the present study.

Genetic analysis of the rotavirus strains: of the 23 samples positive for rotavirus, 16(69.57%) were G-type (VP associated), 2(8.7%) were non-typeable, and 5 (21.74%) were G-negative. When the G genotype was characterized using G1-G4, G6, G8, and G9 primers, only the G3 genotype was found (Figure 2). The P genotypes (VP4 associated) were found in 15/23 (65.22%) The P genotype was characterized using P[4], P[6], P[8], P[9], and P[10] specific primers. Of the 23 samples the P genotypes present were as follows P[6] was one (4.35%), Mixed P[6]P[8] was one (4.35%), P[8] was in eight samples (34.78%), non-typeable genotypes were in five samples (21.74%) and those negative for P genotype were eight (34.78%) (Figure 3).

Combination of G and P types: during the typing of assays it was also found that a specific G type would continuously be associated with a P-type; namely, G3, would coexist with P[8] that is G3P[8] which occurred in 8/23 (34.78%) and only once did a mixed infection occur of G3P[6]P[8] -1/23 (4.35%) (Table 2).

Limitations of the study: the study had several limitations: it was difficult to make accurate conclusion due to results obtained during a relative short period of time and the small sample size. This scope was also limited due to lack of published data on rotavirus infections at Kericho County Referral Hospital. Moreover, only patients at the hospital setting were recruited in this study. In order to investigate the true burden of viral gastroenteritis, community based studies including children in baby care centres and hospital based settings is necessary. Despite these limitations, this study provides the most comprehensive data to date on characterization of rotavirus strains at Kericho County Referral Hospital.

Discussion     

The highest prevalence of infection was identified within the age group 21-30 months old children at 26%. The lowest prevalence was among children within the age group 2-4 months post-vaccination. This is different compared to the results observed by C. Muendo et al. 2018 [21] where the highest infections were observed in the age groups of 17- 24 months. This is also different from 13-24 months at 41.4%, observed by J. Gikonyo et al. 2019 [22]. There is an upward shift in the occurrence of infection from lower age groups to higher age groups. This means the infections are occurring in ages that are not covered by the vaccine. This was also evident in the trends of infection observed in a study carried out before vaccine introduction that indicated the highest infections occurred at 6-17 months [23].

The prevalence rate of group A rotavirus in this study was 11.5% which is less compared to the overall rotavirus prevalence that was 14.5% in a study carried out at Kenyatta National Hospital in Nairobi [24]. This is also lower as compared to the prevalence of 15.2% obtained in a study in Nairobi county in 2019 [22]. This can be deduced that rotavirus vaccination is effective even in rural areas of counties deemed to be based more on agricultural backgrounds. This is also the general evidence in the long run especially in the times immediately after vaccine introduction and studies four to five years after an example being 31.5% in 2017 [11] as compared to recent studies. This shows a great improvement in the effect of the introduction of the rotavirus vaccine as compared to studies carried out pre-vaccine introduction. An example would be the prevalence of Group A rotavirus in a study carried out at Kiambu County Hospital between July 2009 and June 2014 which was at 27.5% [25]. These observations are in line with studies carried out to observe infections pre and post-rotavirus vaccination introduction which indicate a gradual decline in rotavirus prevalence as the systems of vaccination programs improve and awareness increases.

The presence of a specific electropherotype pattern has long since been a well-thought-out diagnostic for the occurrence of individual Groups A, B, and C rotaviruses. Group A rotaviruses that have tested positive through ELISA have a distinct pattern once they undergo PAGE and silver staining. This enabled the study to ensure that the samples were of group A rotavirus. All samples presented 11 segments of dsRNA pattern similar to the positive Group A rotavirus controls that consisted of long and short strain samples with the exception of one which despite being positive in ELISA, did not give any result. There are two types of electropherotype patterns of group A rotavirus that are long and short electropherotypes. Long electropherotypes are linked to G1, G3, G4, and G9 whereas the short electropherotypes are mostly linked to G2. The characteristic grouping of rotavirus RNA segments of 4-2-3-2 was observed. This indicated that the samples were positive for Group A. It should be noted that the electropherotype pattern suggests but does not confirm a given genotype [26].

Studies on Rotavirus have been done in various parts of the world. It has been shown that G1 was the most predominant strain worldwide according to U. Desselberger 2014 and Seheri et al. 2018. This was followed by others G2-G4. In this study, the most predominant strain was G3 at a percentage of 69.57%. There were also combination infections such as G3P [8] at 34.79%, and G3P [6]P[8] at 4.35%. This is in line with the discoveries of Steele and Ivanoff, in 2003 that with each outbreak new rotavirus strains. Our findings are new compared to those by Kiulia et al 2006 which indicated that the most predominant strain in the Meru North district in Kenya was G9. This also indicates a shift in the strains present from the ones that existed before vaccine introduction in Kenya to other strains post-vaccine introduction. In this case, the most prevalent in such areas such as Kiambu County Hospital was G1P[8] (28%), after that G9P[8] (12%), G8P[4] (7%), G1P[4] (5%), G9P[4] (4%), and G12P[6] (3%). In the yearly change of G and P genotypes, a major change from G9P[8] to G1P[8] was observed in 2012 [23]. On the contrary in this study, the most predominant are G3 strains thus indicating serotype placement or shift from the strains included in the vaccine to ones not in the vaccine currently being used at the hospital.

These findings have implications, it is not clear whether the existing monovalent Rotarix vaccine that comprises of G1, P[8] live attenuated strain [27] or Rota Teq vaccine comprising of serotypes G1-G4 [27] will provide a cross-protection counter to emergent strains and other unknown strains in Kenya. Genotype P[4] and G3/G9P[8] combinations identified highlight the likelihood of rotaviruses going through re-assortment [28]. This is also evident with the study carried out in peri-urban regions in Central Kenya that observed a shift in the strains found among infected children, there was an increased detection of G2P[4], G3P[6] and G3P[8], which corresponded temporally with the scheduling of the vaccine introduction [25]. It should be noted that the vaccines in use might not be catering to the emerging strains.

Conclusion     

Gastroenteritis due to Rotavirus is one of the major causes of severe dehydrating diarrhea in children and is prevalent at Kericho County Referral Hospital. The G3 genotype was the main serotype affecting the children in the study and the most prevalent in circulation. Long electropherotypes were more than short electropherotypes. Epidemiologic observation of rotavirus is necessary for understanding the burden of rotavirus-caused gastroenteritis in Kericho County and its impact on the national health cost in Kenya. A variation in the genetic diversity of rotavirus strains was observed, there was a down surge in the prevalence of G1P[8] after vaccination and an increase in G2P[4] by 12.2% and G9P[8] by 20.4% contrary to what had been reported previously. This indicates that strains may be diversified due to climatic conditions, social economic factors, and even infrastructure among other factors. Arising from the findings, it is recommended that Rotavirus surveillance programs should be considered in the future, consideration of seasons considering climate and weather inconsistencies. Considering the observation of a shift of infection to older age groups, the introduction of vaccination into older age groups (ages above 21 months) is advisable. Due to the prevalence of the G3 strains which are not part of the makeup of the Rotarix vaccine, it is also advisable for the government to include other vaccines that may wholly cover other upcoming strains of Rotavirus to increase the efficiency of the vaccination programs.

Competing interests     

The authors declare no competing interests.

Authors' contributions     

All authors were key in the development of the study and supervision at all steps including the development of the manuscript. Beth Khayeli Nyamanga: conceived the study idea, designed the study, collected, analyzed, and interpreted data, and also developed and reviewed the manuscript for publication. James Nyangao: guided the data collection. Carlene Sang: guided the data analysis. Janeth Kombich: guided the designing of the study, supervised data collection, analysis, and interpretation, and reviewed the manuscript for publication. Raphael Lihana: guided the designing of the study, supervised data collection, analysis, and interpretation, drafted the manuscript, and reviewed it for publication. All the authors have read and agreed to the final manuscript.

Acknowledgments     

We acknowledge all participants in the study together with their guardians. We also acknowledge Cheruiyot Sang, Rose Kimeto, Vivian Musimbi, and Sarah Nyamanga who were the research assistants for collecting data.

Tables and figures     

Table 1: demographic characteristics of children with acute gastroenteritis visiting Kericho County Referral Hospital, Kenya (N=200)

Table 2: G and P genotype combinations detected in children of five years and less in Kericho County Referral Hospital.

Figure 1: percentage of rotavirus infections in each age range

Figure 2: group A Rotavirus, G-genotypes identified among children aged five years and below in Kericho County Referral hospital

Figure 3: group A Rotavirus, P-Genotypes identified among children aged five years and below in Kericho County Referral hospital

References