Circulating rotaviral RNA in children with rotavirus antigenemia
© Ahmed et al.; licensee BioMed Central Ltd. 2013
Received: 9 July 2012
Accepted: 18 January 2013
Published: 1 February 2013
Skip to main content
© Ahmed et al.; licensee BioMed Central Ltd. 2013
Received: 9 July 2012
Accepted: 18 January 2013
Published: 1 February 2013
Rotavirus antigenemia is a common phenomenon in children with rotavirus diarrhea, but information is scarce on aspects of this phenomenon, such as genotype specificity, presence of intact viruses and correlation between genomic RNA and antigen concentration. Such information may help in understanding rotavirus pathogenesis and eventually be useful for diagnosis, treatment and prevention.
Serum samples were collected from children who presented at hospitals with diarrhea. Antigenemia was present in 162/250 (64.8%) samples from children with rotavirus diarrhea. No specific rotavirus genotype was found to be associated with antigenemia. Rotavirus particles could not be found by electron microscopy in concentrated serum from children with high levels of antigenemia. In passaged rotavirus suspension a significant correlation (r = 0.9559; P = 0.0029) was found between antigen level and viral copy number, but no significant correlation (r = 0.001480; P = 0.9919) was found between antigenemia level and viral copy number in serum. When intact rotavirus was treated with benzonase endonuclease, genomic double-stranded (ds) RNA was not degraded, but when sera of patients with antigenemia were treated with benzonase endonuclease, genomic dsRNA was degraded, indicating genomic dsRNA was free in sera and not inside virus capsid protein.
Antigenemia is present in a significant number of patients with rotavirus diarrhea. Rotavirus viremia was absent in the children with rotavirus diarrhea who participated in our study, and was not indicated by the presence of antigenemia. The significance of circulating rotavirus antigen and genomic dsRNA in serum of patients with diarrhea deserves further study.
Globally, every year rotavirus infects 114 million children and accounts for about 453,000 deaths mainly in developing countries [1, 2]. RVA is a non-enveloped virus with a triple-layered capsid containing 11 segments of double stranded RNA genome. The nucleotide sequence of VP7 and VP4 characterize the G and P genotypes, respectively . These proteins are used in a binary classification system, 27 G and 35 P types thus far has been identified . In general G1 through G4 and G9 are the most common types causing human infection . A variety of diseases have been found to be associated with rotavirus [6, 7]. Detection of rotavirus antigen and/or RNA in the central nervous system, heart, liver, testes, kidneys, bladder, liver biopsy from infant with cholestatic disease, respiratory secretions, lung cells, microvasculatures of heart and serum of those patients have provided indirect evidences of rotavirus or its components as a causative agent . How rotavirus or its components reach other parts of the human body from the gastrointestinal tract is not known. Saulsburry et al. first reported the presence of rotavirus antigenemia in immunodeficient children with chronic rotavirus diarrhea . Early work suggested that rotavirus antigenemia was the result of host immunological defects , but recent reports suggest that rotavirus antigenemia is commonly observed in immunocompetent children with rotavirus diarrhea.
Therefore, antigenemia could be at the center of the pathogenesis of various extraintestinal infections with rotavirus. However, several key characteristics of this antigenemia are unknown, such as whether antigenemia is associated with certain genotypes of rotavirus, whether antigenemia consists of intact virus particles, or whether there is any correlation between the concentrations of virus particles and genomic RNA in serum. Information on all these points will assist our understanding of the pathogenesis of rotavirus infection and may open new avenues for diagnosis, treatment and prevention of the disease. The aim of the present study was therefore to obtain further information about the characteristics of rotavirus antigenemia.
A total of 250 serum samples were available from patients with rotavirus diarrhea, and antigenemia was detected in 162 (64.8%) of these. Of the 250 serum samples, 169 were from Bangladeshi children: in this group, antigenemia was detected in 123 (72.8%) samples. Eighty-one samples were from Turkish children, and antigenemia was detected in 39 (48.1%) of these samples. A total of 126 samples were available from patients with diarrhea caused by agents other than rotavirus: rotavirus antigenemia was detected in 37 (29.4%) of these samples. Of these 126 serum samples, 7 were from Bangladeshi children, of which 5 (71.4%) showed antigenemia and 119 were from Turkish children, of which 32 (26.9%) showed antigenemia.
Viral particles with characteristics suggestive of rotavirus were not found in any of the samples. This result indicates that intact rotavirus particle is absent in the serum of patients with rotavirus antigenemia.
Serum samples from forty-nine patients were available to determine the correlation between antigenemia level determined by ELISA and the corresponding genomic copy number determined by RT-qPCR. Except one sample from a Turkish child all samples were rotavirus RNA positive. The mean copy number and OD values were 1.24 × 108 copies/μl (95% CI, 0 – 2.89 × 108 copies/μl) and 0.5264 (0.3866 – 0.6663), respectively. The correlation was not significant (Figure 1, Lower panel): the Spearman r was 0.001480 (95% CI −0.2878 – 0.2905) and the P value was 0.9919.
The frequency of antigenemia in children with rotavirus diarrhea that was determined in the present study is within the range observed in other studies [9–13]. Although the prevalence of rotavirus diarrhea in Bangladesh and Turkey is similar [14, 15], it is difficult to say why in Bangladeshi children the frequency of antigenemia was higher than Turkish children. Detection of rotavirus antigenemia is influenced by several factors such as increased acute-phase serum immunoglobulin G antibody titer , days after onset of diarrhea [11, 12] and stool viral antigen levels . Although in this study the frequency of antigenemia found in the rotavirus-negative diarrhea patients seems to be high compared with previous studies [13, 16], comparison is difficult as studies on rotavirus antigenemia in children with rotavirus-negative diarrhea are scarce. The high frequency found in this study may reflect factors such as differences in study population, the high prevalence of rotavirus diarrhea in the study children or sensitivity of the method utilized to detect antigenemia. It is possible that the study children had had a rotavirus infection and recovered before collection of serum but continued to have diarrhea for some reason, because although antigen levels peak 1–3 days after the onset of symptoms and the serum of some children remains positive beyond 1 week, a decline in serum antigen values to baseline over a 3–4 week interval has been observed . Rotavirus antigenemia has even been found in children with no gastroenteritis : a recent study revealed that 18.5% of the healthy children have rotavirus in their stool . Therefore, detecting several cases of rotavirus antigenemia in children without rotavirus diarrhea is not uncommon. In the present study the total number of such samples from Bangladeshi children was small and therefore the high positive antigenemia rate in those children can be almost ignored. Reviewing the serum sample collection period of rotavirus negative Turkish children showed that the majority of the antigenemia positive samples were from the rotavirus season which might be responsible for high frequency of antigenemia in this group of children.
In the case of rotavirus grown in the MA104 cell line, there may only be one infectious virus particle in 46,000 virus particles . If there are a low number of infectious virus particles during rotavirus antigenemia, then serum culture will be unsuccessful. In the present study, two different approaches were undertaken to determine whether intact rotavirus is present in the serum of patients with antigenemia. In one approach, serum samples with high levels of antigenemia were observed under the electron microscope to obtain direct evidence of the presence of intact rotavirus particles. In the other approach, we predicted that the presence of intact rotavirus particles in serum samples of patients with antigenemia will protect genomic RNA from endonuclease and there will be a correlation between serum antigen level and viral copy number. Benzonase endonuclease was used for the endonuclease protection assay. Benzonase endonuclease degrades both DNA and RNA, whether single-stranded, double-stranded, linear, circular or supercoiled (http://wolfson.huji.ac.il/purification/PDF/DNA_Removal/NOVOGEN_Benzonase.pdf).
The use of electron microscopy in serological identification of viruses is an established technique. Two major limitations of electron microscopic detection of extracellular viruses are the requirement for a relatively high concentration of virus (105 to 107/ml, depending on size) and the presence of background materials (such as serum proteins) that may mask the virus particles . The concentrations of virus in our samples were more than adequate to be observed under the electron microscope. The ELISA OD values of the serum samples used for electron microscopy correspond to more than 5 × 105 rotavirus particles/ml. In addition, the samples were concentrated 20× concentrated by ultracentrifugation, which should make the concentration of any viral particles present in the serum at least 1 × 107 rotavirus particles/ml. Had there been viremia during rotavirus diarrhea, these serum samples would have contained enough rotavirus particles to be visualized by electron microscopy. Our experience with cultured rotavirus particles indicates that the characteristic features of rotavirus particles are unlikely to be missed by electron microscopy, even in the presence of background materials, because they cannot mask all the virus particles.
In this study, we provided evidence that the ELISA OD of intact virus correlates significantly with the copy number. However, serum ELISA OD values did not correlate with serum virus copy numbers, which suggested that intact virus particles are not present during rotavirus antigenemia. The absence of intact virus particles in the serum of patients with antigenemia was confirmed by showing degradation of viral RNA in serum by benzonase endonuclease treatment. In an intact virus particle, the protein coat protects the genome from degradation by nucleases, therefore absence of this protection during treatment with benzonase endonuclease demonstrated that viremia was absent during rotavirus diarrhea.
There is only one study that showed the presence of viremia in all of the antigenemia-positive and 22% of the antigenemia-negative children tested by culturing their sera . Virus culture may be more sensitive than electron microscopy, but PCR after benzonase treatment of serum as used in this study is more sensitive than virus culture. The present study provides evidence that free viremia is absent in patients with rotavirus antigenemia, possibly indicating a role for differences in population genetics in causing viremia.
Previous results confirmed that rotavirus G genotypes of the fecal and serum samples from the same patient were identical . Therefore, we used the genotypes found in stool as representative of the genotypes causing antigenemia in the corresponding serum samples. We found no significant differences in genotype distribution between antigenemia positive and negative patients, indicating that it is unlikely that particular genotypes are capable of causing antigenemia. Similar results were found by Fujita et al. , which together with our data suggests that there is no particular genotype of rotavirus that is more likely to cause antigenemia.
In conclusion, there was no specific rotavirus genotype associated with antigenemia, and viremia was absent in children with rotavirus antigenemia. Further study is needed to verify whether there is cell-associated viremia in children with rotavirus antigenemia.
The research proposal was approved by the ethics review board of the Faculty of Medicine, Gazi University and the ethics committee of the Institute of Child and Mother Health (ICMH). The verbal consent of the child’s guardian was obtained prior to sample collection. Since a fraction was obtained from samples sent to the laboratory of routine diagnostic purposes therefore informed written consent was not obtained and it was not required by the ethics review boards.
Children enrolled in this study were less than 5 years old. This is part of studies on rotavirus diarrhea previously carried out in Turkey and Bangladesh [14, 15]. Between September 2004 and December 2005, samples were collected from patients who attended at Gazi University Hospital and the Ministry of Health Ankara Educational and Research Hospital in Ankara, Turkey with watery diarrhea. During July 2005-June 2006, samples were collected from children who attended at the affiliated hospital of ICMH, Matuail, Dhaka, Bangladesh with watery diarrhea. A case of diarrhea was defined as passing looser than normal stool three or more times, during a 24 hour period. One stool and one serum sample were collected from each patient and stored at −80°C until use.
Rotavirus antigen was detected from a 10% stool dilution in phosphate-buffered saline by enzyme immunoassay (Rotaclone, Meridian Diagnostics Inc., Cincinnati, Ohio, USA), according to the manufacturer’s instructions.
Rotavirus antigenemia was detected by a previously published method . A twofold dilution of serum was made with phosphate-buffered saline (pH 7.2), and 100 μl of this was used for rotavirus antigen detection by Rotaclone. Absorbance of 0.15 or more was used as the criterion for positive reactions.
Four serum samples with relatively high optical density (OD) were selected during the detection of antigenemia for observation by electron microscopy. Samples 184 and 109 (OD values 2.333 and 1.874) were obtained from two Bangladeshi children with rotavirus diarrhea and antigenemia. Samples AHP20 and GUP30 (OD values 0.409 and 0.373) were obtained from two Turkish children with rotavirus diarrhea and antigenemia. Serum was concentrated by ultracentrifugation: 400 μl of serum was mixed with an equal volume of 0.9% NaCl solution and laid on a 30% sucrose cushion. Using a Beckman 100.2 rotor, samples were centrifuged at 50,000 rpm for 1 hour at 4°C. The pellet was resuspended in 20 μl 0.9% NaCl solution, and 5 μl of this was applied to a carbon-coated copper grid and then stained with 2% uranyl acetate for 30 sec. The excess of fluid was removed by a piece of filter paper. Specimens were observed under a JEM 100C electron microscope (JEOL Ltd., Tokyo, Japan) operated at 80 kV. The grids were observed blinded at a magnification of × 36,000 or greater .
To determine whether there is a correlation between ELISA OD and virus genome copy number when intact virus particles are present in a fluid, cultured rotavirus strain SA11 suspended in phosphate-buffered saline (PBS) was serially diluted twofold in PBS, and then ELISA OD was determined as described above and the copy number in each diluted sample was determined as described below. The correlation between ELISA OD and copy number was determined as described in the statistical analysis section.
Rotavirus copy number was determined by reverse transcription quantitative polymerase chain reaction (RT-qPCR) . Nucleic acid was extracted from 140 μl of serum using the QIAamp Viral RNA Mini kit (Qiagen Co. Ltd., Tokyo, Japan). RT-qPCR with VP6-specific primers was carried out using the QuantiTect SYBR Green RT-PCR kit (Qiagen) and a LightCycler 480 (Roche, Mannheim, Germany) real-time PCR thermal cycler. The primers VP6- F (sense) (5′ − GACGGVGCRACTACATGGT − 3′) (nucleotides [nt] 747–766) and VP6-R (antisense) (5′ − GTCCAATTCATNCCTGGTGG − 3′) (nt 1126–1106) amplified a 379-bp region (nt 747–1126) of the VP6 gene . The genomic dsRNA was extracted from cultured rotavirus strain SA11 by the phenol − chloroform − isoamyl alcohol method and run on a 0.8% agarose gel. Segment 6 was cut out, and then dsRNA was extracted and purified by QIAquick Gel Extraction kit (Qiagen) according to the manufacturer’s instructions. A standard curve was generated by amplifying a known amount of dsRNA . One copy of rotavirus equals 1.29 × 10−12 μg of segment 6 dsRNA. A serial 10-fold dilution of extracted and purified VP6 was used for calibration using the LightCycler 480 software and a reference gene for normalization between PCR runs. The crossing point, defined as the PCR cycle at which the amplification signal intersects the log linear region, was used as a direct correlate of the amount of rotavirus RNA present in the serum. Each sample was measured three times and the average was taken to calculate the copy number.
To determine whether antigenemia consists of intact rotavirus particles, sera were treated with benzonase endonuclease (Novagen EMD Chemicals, San Diego, CA, USA), and then VP6-specific RT-PCR was performed to determine the presence or absence of genomic RNA. In the presence of intact virus particles, the genomic RNA will not be degraded by endonuclease and the target gene will be amplified by RT-PCR. However, if naked RNA is present, it will be degraded and RT-PCR will show a negative result. To determine the optimum conditions for this assay, different amounts (7.3 × 107, 9.0 × 105 and 8.5 × 103) of rotavirus SA11 strain and corresponding genomic RNA in distilled water were treated with 100 U of benzonase for 22 hours at 37°C according to the manufacturer’s instructions. The reaction was stopped by adding EDTA at a final concentration of 2 mM to inactivate by the benzonase. Specimens handled similarly without benzonase treatment were used as a control. Nucleic acid was extracted using a QIAamp viral RNA minikit according to the manufacturer’s instructions and subjected to RT-PCR for the detection of the VP6 gene using primers and methods described elsewhere [14, 24].
To determine the effects of rotavirus genotype distribution on antigenemia, the genotype distribution of rotavirus found in stool was compared in patients with and without antigenemia. Genotypes were determined by RT-PCR as described in our previous studies [14, 15].
The Mann–Whitney test and Spearman rank correlation, appropriate for experimental data that display a non-Gaussian distribution , were used to calculate two-tailed P values. The Pearson correlation was used for experimental data that displayed a Gaussian distribution. The χ2 test was used to determine whether genotype distribution differed between patients with and without antigenemia. All statistical analyses were performed using Prism 5 software (GraphPad Software Inc., La Jolla, CA, USA).
We thank Mr. Akitoyo Ichinose, Institute of Tropical Medicine, Nagasaki University and Ms. Aiko Yasuda, Research Promotion Institute, Oita University for their excellent technical assistance for electron microscopy. This research was supported by a Grant-in-Aid Scientific Research B from Japan Society for the Promotion of Sciences (grant number: 17406013). The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.