Rotavirus

ROTAVIRUS STRUCTURE AND ANTIGENS Rotavirus Structure Rotaviruses (RVs) form the genus Rotavirus of the Reoviridae family of viruses. The RV genome consists of 11 segments of double-stranded RNA and is enclosed within 3 concentric protein layers (capsids). Viral protein (VP) 2 represents the core or inner capsid, whereas VP6 constitutes the middle capsid, determines serogroup antigen specificity (A–G), and is the most immunogenic protein in RV infection. Group A, B, and C RVs all have been identified in humans, with group A RV responsible for the majority of disease (1,2). Group A RVs are differentiated by serotype, which is determined on the basis of the antigens expressed on the outer viral capsid, VP7 (a glycoprotein, which defines the G-type antigens) or VP4 (a protease-cleaved protein, which defines the P-type antigens) (1,3). Analogous to the classification of influenza viruses, RV classification is a binary system that includes both the VP4 and VP7 types (3,4). The serotype and genotype of a particular G-type antigen usually match, and therefore RVs generally are referred to by their serotype alone (eg, G1, G2, G3, etc) (1). On the other hand, because there are greater numbers of, and variation in, P-type genotypes than serotypes, concordance for P-type antigens is rare. Accordingly, P-type antigens are denoted with their genotype referenced in brackets (eg, P[4], P[8]) (1). G and P proteins induce neutralising antibodies involved in protective immunity. The segmented nature of the RV genome allows for gene reassortment upon co-infection of a single cell with 2 different RV strains. This property has the potential to generate many combinations of G and P proteins (theoretically, 211 different combinations), and thus creates serotype diversity. In practice, however, the number of G and P combinations is less than the theoretical number of possible reassortants (Table 1).TABLE 1: Diversity of rotavirus strains circulating in humansDistribution of Rotavirus Serotypes The range of RV strains cocirculating in Europe is diverse, with predominant and emerging strains varying between regions and from year to year (2,5). At present in Europe, 5 common G/P combinations of RV predominate, namely G1P[8], G2P[4], G3P[8], G4P[8], and G9P[8] (2,5,6). There are at least 15 G types and 28 P types circulating (7–11). Analysis of strains collected between 1973 and 2003 indicates that G1P[8] was the most prevalent, being the strain responsible for 69.4% of infections in Europe (6). Surveillance for cocirculating RV strains has been initiated by the European Rotavirus Network in 11 European countries and data will be collected for the 3 years before introduction of RV vaccines and for 3 years afterward (12). The distribution of RV G genotypes in Europe in the 2004–2005 RV season is shown in Figure 1; G1, G2, G3, G4, and G9 represented >98% of rotavirus gastroenteritis (RVGE) cases in children (5). The significance of these strains for vaccination recently was reviewed by Desselberger et al (2). All of the serotypes studied cause disease in humans, with no clear evidence of any consistent strain-specific differences in the severity of disease or the age of person affected (2,5), but several reports indicate that, on average, the duration of symptoms and severity of RVGE are greater than for non–RV-associated gastroenteritis in age-matched subjects (13,14).FIG. 1: The distribution of circulating rotavirus G genotypes (corrected for participation rates and sampling fraction) in 2004–2005 among children with rotavirus gastroenteritis in 7 European countries (Belgium, France, Germany, Italy, Spain, Sweden, and the United Kingdom) (5).DISEASE Transmission and Pathology RVs are most commonly transmitted via the faecal/oral route; there is also evidence to suggest that they can be transmitted in respiratory droplets (15,16). Transmission by the respiratory route does not, however, imply virus multiplication in the respiratory tract. Rather, RVs are swallowed with mucus, which also protects against gastric acidity. RVs and live oral RV vaccines are relatively acid labile, and thus need to be buffered to withstand gastric acidity (17). After ingestion, RVs infect the mature enterocytes at the tip of the villi in the small intestine. Virus particles replicate in the cell cytoplasm and are then shed, permanently damaging the cell and preventing effective uptake of fluids and nutrients. Secretory crypt cells proliferate extensively to compensate, leading to substantial fluid and electrolyte secretion into the gut lumen. Cellular dysfunction also can prevent the expression of some digestive enzymes, and owing to the increased osmotic potential resulting from the accumulation of sugars within the small intestine, may exacerbate fluid loss (18). The excess volume of fluid excreted in this manner results in the clinical manifestation of diarrhoea. The pathophysiology of RV diarrhoea is therefore a combination of osmotic and secretory mechanisms (17). The mechanism by which RV causes vomiting is not known. It is now recognised that RV can spread from the intestine into the circulation, with RV antigen and RNA detected in blood (19). There is, however, no evidence that RV multiplies in extraintestinal sites (19). Although the clinical significance of systemic spread is largely unknown, it should be noted that RV has been detected in cerebrospinal fluid and also has been associated with central nervous system disease symptoms in case reports (20). Other disease associations include a possible role of RV infection in the induction of 2 autoimmune diseases, type 1 diabetes, and coeliac disease. For type 1 diabetes, an Australian group first reported molecular mimicry between RV surface protein VP7 and immunodominant epitopes of 2 autoantibodies that are markers for T cell–mediated destruction of pancreatic islet cells, namely tyrosine phosphatase IA-2 and GAD65 (21). Subsequently Australian investigators found an association between RV seroconversion and increased antibody levels to IA-2, insulin, and GAD65 in children at risk for diabetes who were followed up from birth (22). However, a Finnish study failed to confirm this finding (23,24). Likewise, in coeliac disease, molecular mimicry has been reported between RV VP7 and transglutaminase. A subset of antitransglutaminase antibodies from sera of patients with acute coeliac disease recognise RV VP7, suggesting a possible involvement of RV infection in the pathogenesis of the disease (25). Furthermore, a prospective study of US children with coeliac disease–risk human leucocyte antigen alleles followed up from infancy to 24 months of age identified multiple RV infections as a risk factor for coeliac disease autoimmunity (26). The role of RV vaccination in relation to these newly identified, yet hypothetical, risk associations is totally unknown. On the one hand, RV vaccination (especially repeated dosing) could be a similar risk factor to wild-type RV infection. On the other hand, RV vaccination with attenuated strains—including a mild and subclinical RV infection—will protect against severe RVGE, and therefore against the development of autoimmunity. Symptoms and Diagnosis A sudden onset of symptoms typically manifests in children 1 to 2 days after infection with RV. The clinical picture of RVGE is characterised by 4 to 7 days of acute febrile illness, vomiting, and watery, nonbloody diarrhoea. This combination can lead to rapid dehydration without appropriate intervention. Secondary infections with RV are clinically milder or asymptomatic (27,28). Adults more frequently experience asymptomatic RV infection than children; however, when symptoms are reported in adults, the most common are diarrhoea, fever, headache, malaise, nausea, or cramping (27). Despite adults having milder symptoms of RV, they are still infectious, and thus can transmit the infection to susceptible children (27,29). Laboratory diagnosis of RV infection using enzyme immunoassay involves detection of RV VP6 group A–specific antigen in diarrhoeal stools (7,30–34). Alternatively, the immunological detection of RV VP6 antigen is through the capture of antigen from faeces by anti-VP6–specific antibodies immobilised onto latex (reverse passive latex agglutination test), nitrocellulose, or nylon (rapid chromatographic assays). For enzyme immunoassay, captured antigen is detected by a second labelled anti-VP6–specific antibody and chromatographic assays, or by visible agglutination of latex particles. It is also known that demonstration of seroconversion in rotavirus immunoglobulin (Ig) A or IgG antibodies can detect more cases of RV than antigen detection. Serological diagnosis is seldom applied in the clinical setting. Antigen enzyme immunoassay detection usually is used as the laboratory determinant for a case definition of RVGE in a clinical setting, and subsequently for determination of vaccine efficacy in clinical trials of RV vaccines. However, the insensitivity of antigen detection techniques leads to an underestimation of RV disease and asymptomatic RV infection (35). Using real-time polymerase chain reaction, more RV infections will be detected due to its superior sensitivity (30,33–35), but it can be difficult to differentiate a causal association with symptoms from prolonged shedding using this method. Outcome of Infection Although RVGE is regarded as a self-limiting disease, the combination of watery diarrhoea, vomiting, and fever often can lead to rapid dehydration. Rehydration therapy is used for the treatment of dehydration associated with RVGE. This can be given orally, or in serious cases intravenously. In the developing world, lack of access to rehydration therapy and other medical care leads to the considerable mortality rates associated with RVGE (36). In industrialised countries, a delay in treatment can result in death, although this is rare (3,37). Upon resolution of gastroenteritis symptoms, RV antigen is typically shed in the stools for approximately 1 week (38). However, shedding may continue for many months in immunocompromised people (39). IMMUNITY Although not completely understood, protective RV immunity is considered to be multifactorial, achieved through the combined action of secretory antibodies and humoral and cell-mediated immunity (40). IgA and IgG antibodies specific for structural proteins VP7, VP4, and VP6, as well as an enterotoxin called nonstructural protein 4, are involved. Quantitatively, the strongest antibody response is against VP6 (41). Possible immune correlates of protection for naturally occurring RV disease include neutralising antibodies against VP7 (G-type antigen) and VP4 (P-type antigen), as well as high levels (geometric mean anti-RV serum titre >1:800) of serum IgA antibody (largely against VP6 antigen) (16,40,42,43). However, none of these correlates—neither alone nor in combination—provide complete assurance of clinical protection. Immune protection can be homotypic or heterotypic. In the case of RV, homotypic protection prevents infection with the same virus G-serotype, whereas heterotypic protection is associated with prevention of infection by a virus that is of a different G-type. Many RVs of different G-types express other proteins that are serologically or genotypically identical. This results from the relative lack of diversity, when compared with G-types, among P-types, VP6 subgroups, and nonstructural protein 4 genotypes circulating in the human population (Fig. 2). The relative importance of homotypic versus heterotypic protection against rotavirus infection still is not fully elucidated. The best available evidence comes from empirical observations of natural RV infections and RV vaccine trials.FIG. 2: Antigens shared among common rotavirus strains. Shared antigens in columns are represented by identical shading or patterns. P[8] commonly is associated with viral protein (VP) 6 subgroup II and nonstructural protein 4 (NSP4) B, whereas P[4], P[6], and P[11] are associated with subgroup I and NSP4 A.Immune Protection Arising From Natural Rotavirus Infection Natural infection with RV reduces the frequency of subsequent RV episodes, and particularly protects against clinically significant RV disease (28,42,44,45). The first study to show that asymptomatic neonatal RV infection protects against clinically severe RV diarrhoea during the first 3 years of life was published in 1983 (46). A study in Mexican children also showed that RV symptoms are the most severe with the first RV infection, and decrease in severity with increasing number of RV infections (28,42). Two natural infections with RV provided complete protection against moderate-to-severe RV diarrhoea and 3 natural infections with RV protected against RV diarrhoea (28,42). Both symptomatic and asymptomatic infection provided comparable protection against future episodes of RVGE (28). Immune Protection Arising From Vaccination Early studies with candidate animal (bovine and rhesus) RV vaccines demonstrated that heterologous RVs infect the human intestine and induce similar immune protection to that observed after natural RV infection (47–52). Protective efficacy against moderate-to-severe disease was between 80% and 90% (47–52). These early candidate vaccines demonstrated proof of principle for vaccination with a live heterologous virus. Animal–human reassortant RV vaccines, as well as vaccines derived from human RV strains, may perform better than early animal vaccines because they express human RV VP7 and/or VP4 antigens. Evidence from efficacy trials of the new rotavirus vaccines indicates that protective efficacy after 2 or 3 doses of live oral RV vaccines is largely similar to that induced by 2 natural infections with RV (53). EPIDEMIOLOGY In developing countries, RVGE is responsible for >500,000 deaths each year, likely related to limited access to medical provisions (36,54). However, in Europe and other industrialised regions, death from RVGE is rare (3,37). In these regions, the greatest health care burden of RVGE is the large number of medical visits and hospitalisations (3,37). RV disease therefore puts considerable pressure on medical resources and the economy, as well as reducing the quality of life of infected children and their families. Age of Infection Primary infection with RV peaks between 9 and 23 months of age (36,55–57), with most children infected, with or without evidence of symptoms, before 5 years of age (36). Disease Incidence in Europe In Europe and other regions with a temperate climate, RV has a seasonal distribution, with disease peaks occurring in winter and early spring. A south to north pattern of disease incidence has been reported in Europe. The RV season reportedly begins earlier in Spain, and progressively spreads northward to France, the United Kingdom, and eventually the Netherlands and Finland over the following months (1). Medical Burden of Rotavirus Disease Community-acquired Infection In European Union (EU) countries, there are approximately 25 million children younger than 5 years, and 3.6 million births per year (37). The primary burden of RV disease is on child health. It is estimated that annually in Europe almost 700,000 children younger than 5 years will visit a medical practitioner as outpatients because of RV-associated disease (37). In day care centre attendees, the median incidence of RVGE is 1.21/1000 child-days in children younger than 4 years (Table 2) (58–62). This number is higher than the median incidence of RVGE in the general population of children under 5 years in the EU (0.43/1000 child-days) (37), or calculated prospectively for placebo recipients younger than 2 years in RV vaccine trials in industrialised countries (0.37/1000 child-days; Table 3) (47,49–51,63–74), suggesting that children who attend day care centres may be at an increased risk of being infected with RV and developing RVGE.TABLE 2: Incidence of rotavirus gastroenteritis in day care centres in industrialised countries (58–62)TABLE 3: Prospectively determined incidence of rotavirus gastroenteritis in infants up to 2 years old enrolled in control groups (placebo recipients) of randomised clinical trials of historical RV vaccines in industrialised countries (47,49–51,63–74)In Europe, children who are hospitalised with severe RVGE represent the greatest medical burden of RV infection. A recent review using methodology based on the Centers for Disease Control and Prevention model estimates that almost 90,000 children younger than 5 years will be admitted to hospital with RVGE each year in the EU. This translates to a cumulative incidence rate of 1 in 54 children experiencing a severe RV infection requiring hospitalisation by the age of 5 years (37). Infants at ages 6 to 23 months have the greatest risk for being hospitalised with RVGE (75). Using the same Centers for Disease Control and Prevention model, it is estimated that 200 to 250 deaths are caused each year by severe RV in children ages 5 years or younger in the EU (37). However, limited data are available to demonstrate direct evidence for the actual number of deaths occurring as a result of RVGE in the EU. Adults exposed to infected children have a high risk for contracting RV. Although infection in adults is more frequently asymptomatic, RV is still shed in their stools and can therefore be transmitted to children (27,29). Elderly adults and people with compromised immune systems are at greater risk of developing severe symptoms of RV infection and experiencing prolonged viral dissemination (27). As a these more often medical (27). Infection RV infection represents a particular for children in Europe. from studies in European and other industrialised countries indicate the rate of incidence of RVGE during hospitalisation is to child-days of hospitalisation for infants younger than 2 years, and to child-days of hospitalisation for children younger than 5 years In a review of studies of RV disease burden in Europe, it was estimated that for 4 children years of age admitted to hospital with RVGE, there was 1 case of RVGE Burden of Rotavirus Disease the burden of disease, direct and and other to and should be considered medical include visits to the and and the of laboratory other direct include and child include loss of to attend to a child and and other can include as a of of disease, quality of and family and A study in Finland in the that direct medical for of the of RV disease of were related to hospitalisation In other countries, the relative importance of direct medical compared with loss of and may be In to due to loss of for were of the of disease, whereas for general for of the of disease For Germany, it was estimated that of the RV burden was related to the of to and to For Europe, the mean per case hospitalised from RVGE was estimated to be by estimated number of RVGE hospitalisations in children younger than 5 years of age in the EU 25 countries from (37), the of hospitalisation The burden on health care resources is particularly during the when the seasonal of RV cases with a in incidence of other as influenza and respiratory virus RV infection a substantial burden to owing to hospital and the of to new to prevent of the virus In during each case of RV infection an of direct medical compared with hospital that not infected with RV The distribution of among countries, on the of health and In Europe, the estimated 700,000 cases of symptomatic RV disease requiring medical are responsible for considerable among infants and children younger than 5 years of This translates to 1 symptomatic RV infection for 7 children each year, and of these symptomatic cases 1 in 54 children will hospital treatment (37). the development of vaccines against RV, RVGE has been to the most common in infants and children within the EU (37). This a new to prevent severe cases of RVGE in Europe. of of Group is the and of a European rotavirus strain by and is also the of a study by For both of these is by the Protection has also and for from has for and from and has been the in clinical trials for and has been the of vaccine studies for and (a for which the of the of and from and has been the in studies by and has also for and and from and has for and from and from and and for from and is a of the Rotavirus of group have to attend from in the of from and has been a to and a for studies by and has also from and has and/or for from France, and has or from Italy, and and to attend from and has been the of studies by and has also been a to and and has from these to attend