Date Published: April 26, 2016
Publisher: Public Library of Science
Author(s): Johan Nordgren, Sumit Sharma, Anita Kambhampati, Ben Lopman, Lennart Svensson, Rebecca Ellis Dutch.
The notion that certain individuals appear more or less susceptible to infections or to specific microbes is not new, but, until recently, it was assumed that clinical outcome of an infection was mainly owing to virulence factors of the microorganism. Relatively little attention has been given to host genetic factors involved in innate or adaptive immunity or expression of pathogen receptors. A remarkable example of susceptibility dependence is the strong Mendelian trait resistance to the most common noroviruses among individuals with a nonsense mutation in chromosome 19. Norovirus is recognized as the leading cause of gastroenteritis worldwide, affecting children and adults alike . Noroviruses are highly contagious and genetically diverse RNA viruses, but not all individuals are susceptible to infection to the same norovirus genotypes. Presence of histo-blood group antigens (HBGAs) on gut epithelial surfaces is essential for susceptibility to many norovirus genotypes. The synthesis of these HBGAs, specifically of the ABH and Lewis families, requires the use of several fucosyl and glycosyltransferases encoded by the FUT2, FUT3, and ABH genes. Polymorphisms in these genes vary considerably depending on ethnicity, with a homozygous nonsense mutation (individuals called non-secretors) in the FUT2 gene occurring in approximately 5%–50% of different populations worldwide [3–5]. Secretor status also affects gut microbiota composition, including HBGA-expressing bacteria and bacteria inducing fucosylation in the gut. These could be intermediary factors that govern norovirus susceptibility [6–9].
Noroviruses infecting humans are highly diverse and comprise three genogroups and at least 33 genotypes, which are classified according to nucleotide identities in the major capsid protein gene. The earliest volunteer studies in the 1970s used the first isolated norovirus, the Norwalk virus (genogroup I, genotype 1, GI.1) . These early studies hinted that not all individuals were inherently susceptible and, in the last decade, challenge and outbreak studies from several countries have confirmed a genetic component to norovirus susceptibility. Moreover, different norovirus genotypes are clearly associated with different epidemiological and susceptibility patterns. The globally dominant GII.4 viruses exhibit a strong secretor specificity in vivo ; furthermore, some data suggest that these viruses may possibly be more clinically severe [11,12]. Observational studies in countries such as the United States, Ecuador, Burkina Faso, Vietnam, China, and Nicaragua have consistently identified that non-secretors are highly resistant to GII.4 and that GII.4 variants emerging over time have similar secretor specificity in vivo, although in vitro studies have shown that some GII.4 variants can bind to sugars present in non-secretors [3,13–17].
Since distribution of the ABO(H), secretor, and Lewis genotypes is strongly dependent on ethnicity, we hypothesize that molecular epidemiology differs between regions because of host population genetics. According to this logic, populations with a higher proportion of secretors would be susceptible to both secretor-independent and dependent genotypes, including GII.4 virus. This could potentially translate into a higher norovirus infection rate than in populations with a lower proportion of secretors, especially since no norovirus genotypes have been found that selectively infect non-secretors. In European-descended, Asian, and some African populations, secretors constitute approximately 80% of the population (Fig 1). In contrast, secretor prevalence in Mesoamerican populations can be as high as 95% . A high proportion of individuals genetically susceptible to GII.4 viruses might lead to higher infection rates with GII.4 among these populations. Indeed, Gll.4 noroviruses constitute 75% of infecting genotypes among children with diarrhea in community- and hospital-based studies from Nicaragua, Guatemala, and Mexico [12,24,25], a relatively higher proportion than the majority of studies from other continents (Fig 1). However, more studies are needed to address whether a higher frequency of GII.4 infections, or other secretor-dependent genotypes, also translates into higher disease burden, because the clinical relevance of many norovirus genotypes is not well established. The observation from Ecuador that norovirus gastroenteritis was similar, but genotype distribution differed between secretors and non-secretors, suggests that population-level risk may not scale in a simplistic way with secretor prevalence in all settings.
The presence of different secretor and non-secretor human HBGAs is associated with increased susceptibility to several infectious diseases such as HIV, rhinovirus, Haemophilus influenzae, Neisseria meningitidis, and urinary tract infections , thus rendering a large selection of pressure interplay for the pathogens and perhaps resulting in genetic variation in the circulating pathogens as well as the human population. A reasonable hypothesis is that since human norovirus genotypes show differences in the spectrum of HBGAs to which they bind, it is possible that populations with a large diversity of HBGAs would sustain a greater variety of norovirus genotypes. Supporting this notion, genetic and phenotypic diversity of HBGAs has been found to be higher in sub-Saharan African populations compared to many other regions . Accordingly, the norovirus genotype distribution in children is generally more diverse in Africa than other regions. For example, in the West African nation of Burkina Faso, all secretor, Lewis, and ABO phenotypes are present in relatively large proportions . A prospective study there, spanning only ten months, detected 14 different norovirus genotypes in children with diarrhea. Other African studies have found similarly high norovirus diversity (Fig 1). This can be compared to the lower genotype diversity (approximately two to eight genotypes) generally found in similar studies performed in Europe and North America, where several HBGA phenotypes are present only in small proportions. More host genetic studies, especially from sub-Saharan Africa, are needed to address this.
Norovirus vaccines are currently under development and have shown promise in early phase clinical trials that assessed protection against experimental virus challenge [27,28]. Vaccines that have entered human trials have been either monovalent (genotype GI.1) or bivalent (genotypes GI.1/GII.4). These vaccines have been shown to confer a degree of clinical efficacy against challenge with a genotype included in the vaccine. Other studies suggest that cross-reactive antibodies are produced after natural infection with certain noroviruses [29,30], but, to date, there are no clinical data on heterotypic protection from vaccination, although heterotypic blocking antibodies have been produced following intramuscular immunization with the bivalent vaccine .