NORWALK AND RELATED VIRUSES (CALICIVIRIDAE)

History

Diarrheal illnesses of humans have been documented since the dawn of history. In spite of the discovery during the past century of numerous enteric bacteria and parasites, the etiology of a major portion of diarrheal illnesses remained unknown until the 1970s. It was assumed by exclusion that viruses were responsible for a significant portion of these illnesses. Thus, volunteer studies were initiated in the 1940s and 1950s because most viruses could not be propagated in vitro or in an animal model. These studies attempted to determine if illness could be induced with a presumptive nonbacterial viral agent by feeding bacteria-free fecal filtrates, derived from adults with acute nonbacterial gastroenteritis, to adult volunteers. Illness was induced demonstrating that the nonbacterial suspensions were indeed etiologic agents of acute gastroenteritis. The filtrates were named after the source of the original specimen, e.g. Marcy (Marcy State Hospital, Utica, New York), Niigata (Niigata Prefecture, Japan) and FS (Family Study, Cleveland, Ohio). These agents were passaged serially to other volunteers and also characterized antigenically in crosschallenge studies. However, because none of the infectious agents could be identified further in vitro or in an animal model, the etiology of acute nonbacterial gastroenteritis remained elusive. In the 1950s and 1960s, the introduction of tissue culture technology led to the discovery of over 100 new viruses, many of which were isolated in feces, but none of these new viruses could be implicated as an important etiologic agent of acute gastroenteritis. In the late 1960s, human embryonic organ cultures of the respiratory tract were used to successfully isolate several fastidious respiratory coronaviruses. Prompted by this success, renewed efforts were made to isolate viruses from fecal specimens of patients with viral gastroenteritis by utilizing human embryonic intestinal organ culture; these studies failed to identify an etiologic agent. It was considered that perhaps the lack of success was due to the absence of an infectious agent in the fecal specimen used as inoculum. Fecal suspensions prepared in the 1940s and 1950s, as described above, that were known to be infectious in challenge studies were either exhausted or not available at that time for evaluation with these newer techniques. Thus, in the early 1970s, a ‘second generation’ of volunteer studies was initiated. In one of these, a filtrate made from a rectal swab specimen from a secondary case of gastroenteritis from an outbreak of ‘winter vomiting disease’ that affected about 50% of the students and teachers in an elementary school in Norwalk, Ohio, was found to induce gastroenteritis in adult volunteers. This illness-inducing filtrate was serially passaged to other volunteers and was designated as the Norwalk agent. Attempts to cultivate or identify an infectious agent in vitro with the newer cell or intestinal organ culture techniques or in animals were unsuccessful.

In 1972, application of immune electron microscopy (IEM) led to the discovery of 27 nm virus-like particles in an infectious stool derived from the Norwalk outbreak and to the etiologic association of these virus-like particles with the Norwalk, Ohio outbreak. The 27 nm Norwalk particles thus became the first virus to be implicated as the cause of epidemic nonbacterial gastroenteritis; it was discovered without the use of any in vitro or animal model system. IEM also led to the discovery of other gastroenteritis agents such as the Montgomery County, Hawaii and Snow Mountain agents. Approximately 1 year after the discovery of the Norwalk virus, rotaviruses, the major etiologic agents of acute gastroenteritis of infants and young children, were discovered by electron microscopic techniques, again without the benefit of any in vitro system or animal model.

Taxonomy and Classification

The classification of viruses responsible for acute gastroenteritis was first based on morphology. For example, Norwalk virus was the prototype of a group of agents initially called small round-structured viruses (SRSVs). Recently, rapid advances in molecular biology have allowed these viruses to be classified based on their genome characteristics and most of the previously named SRSVs belong to the Caliciviridae.

The family Caliciviridae contains four genera: Lagovirus, ‘Norwalk-like viruses’, ‘Sapporo-like viruses’ and Vesivirus. The human caliciviruses responsible for epidemic gastroenteritis belong to the genus ‘Norwalk-like viruses’ (type species: Norwalk virus) and ‘Sapporo-like viruses’ (type species: Sapporo virus) (Table 1 ).

Table 1.

Human calicivirus classification and antigenic relationships among the caliciviruses

		Antigenic relationships
Genus	Virus straina	Determined by serologic IEM and/or crosschallenge studies in volunteers as described in the text	Determined by ELISA with hyperimmune antisera raised against recombinant Grimsby, Norwalk and Mexico virus-like particles
‘Norwalk-like viruses’	Norwalk	Distinct serotype	Distinct type
	Hawaii	Distinct serotype	Unknown
	Snow Mountain	Distinct serotype	Unknown
	Lorsdale	Unknown	Unknown (probably same as Grimsby)
	Southampton	Unknown	Unknown
	Mexico	Unknown	Distinct type
	Grimsby	Unknown	Distinct type
‘Sapporo-like viruses’	Sapporo	Distinct serotype	Unknown
	Manchester	Unknown	Unknown

^aOnly a representative subset of virus strains in each genus (‘Norwalk-like viruses’ and ‘Sapporo-like viruses’). See Green et al. (1999) for complete listing of strains.

Properties of the Virion

These nonenveloped viruses contain a genome of single-stranded RNA of positive polarity. They have a diameter of 27–40 nm by negative stain electron microscopy and a buoyant density of 1.33–1.41 g ml⁻¹. These viruses, which are shed in the feces of individuals with gastroenteritis, cannot be propagated in cell cultures.

The virions are composed of a single capsid protein. Structural analysis of virus particles from stool is limited by the low number of particles present in these samples. However, by negative stain electron microscopy, the Norwalk virus has an indistinct ‘feathery’ outer edge and an indistinct surface substructure, although there is a suggestion of indentations on its surface (Fig. 1A ). Expression of the capsid protein using the baculovirus system results in the self-assembly of the Norwalk virus protein into virus-like particles (VLPs) (Fig. 1B). By negative stain electron microscopy, these VLPs have a similar morphology to the native virus. The structure of these VLPs has been resolved by electron cryomicroscopy and computer image processing. The capsid exhibits a T = 3 icosahedral symmetry. The major structural protein folds into 90 dimers that form a shell domain from which arch-like capsomers protrude. A key characteristic of this architecture is 32 cup-shaped depressions at each of the icosahedral fivefold and threefold axes.

Figure 1.

Negative stain electron micrographs of (A) Norwalk virus purified from stool of a volunteer (♯547) given Hu/NVL/NV/1968/US, and (B) recombinant Norwalk virus-like particles produced and purified from insect cells infected with a recombinant baculovirus expressing ORF 2 and ORF 3 of Norwalk. Bar = 100 nm.

Genome Organization

In 1990, the genome of Norwalk virus was cloned and characterized. More recent studies have characterized and sequenced completely other viruses belonging to one of the two human calicivirus genera, either in the ‘Norwalk-like viruses’ (Norwalk virus, Lordsdale virus and Southampton virus) or in the ‘Sapporo-like viruses’ (Manchester virus). These viruses contain a positive-sense polyadenylated single-stranded RNA of approximately 7.6 kb (Fig. 2 ).

Figure 2.

Genome organization of human caliciviruses. (A) Genome organization of Norwalk virus. The first ORF encodes the nonstructural proteins; ORF 2 encodes the capsid protein; and ORF 3 encodes a small basic protein. (B) Genome organization of Manchester virus (Liu et al, 1997). ORF 1 encodes the nonstructural proteins and the capsid protein, followed by ORF 2 (the homologue of the Norwalk virus ORF 3) encoding a small basic protein. ORF 3 is encoded by an out-of-frame sequence within ORF 1.

The genome of ‘Norwalk-like viruses’ is organized in three major open reading frames (ORFs). For Norwalk virus, the first ORF at the 5′ end encodes a large polyprotein of 1738 amino acids (aa) with a predicted molecular weight of 193.5 (193.5 K). This polyprotein contains short motifs of similarity with the 2C (helicase), 3C (cysteine protease), and 3D (RNA-dependent RNA polymerase) proteins of picornaviruses ( Fig. 3 ). Thus, the 5′ end of the genome of the ‘Norwalk-like viruses’ codes for a precursor of the nonstructural proteins. ORF 2 encodes a 530aa (56.6 K) protein, the capsid protein. The ORF 2 protein expressed in insect cells self-assembles into virus-like particles as explained above (Fig. 1B). ORF 3 at the 3′ end of the genome is predicted to code for a small protein of 212aa (22.5 K) with a very basic charge (isoelectric point of 10.99). The ORF 3 protein does not have sequence similarity with any other proteins in the GenBank and its function remains unknown, although recently the ORF 3 protein has been found in VLPs expressed from cDNA constructs that contain both ORF 2 and ORF 3.

Figure 3.

The genome of ‘Sapporo-like viruses’ is organized slightly differently. For Manchester virus, the first ORF codes for the nonstructural proteins as well as the capsid protein, which is found in-frame at the end of the nonstructural proteins. This genome organization is similar to that found in the animal calicivirus rabbit hemorraghic disease virus (RHDV) belonging to the genus Lagovirus (Fig. 3). ORF 2 encodes a predicted small highly basic protein of unknown function, similar to ORF 3 for Norwalk virus. Manchester virus contains a third ORF within the capsid protein which could encode another small basic protein. The significance of this ORF is unclear as it is not seen in any of the other calicivirus genomes sequenced thus far, and the small protein it potentially encodes shows no sequence homology to other viral proteins in the database.

Serologic and Phylogenetic Relationships

Four distinct serotypes of the ‘Norwalk-like viruses’ (Norwalk, Hawaii, Snow Mountain and Taunton) and one serotype of the ‘Sapporo-like viruses’ ( Sapporo) were originally described by serologic IEM studies employing virus particles shed in stools as antigen and paired sera of infected individuals as the source of antibody. These serotype designations assumed that antibody reactivity by IEM reflects the reactivity of neutralizing antibodies. This may not be the case. Therefore, serotypes remain to be more clearly defined, a goal that is unable to be achieved due to the lack of a cultivation system. Early volunteer studies also examined the ability of different viruses to induce crossprotection. Based on these studies in addition to IEM, Norwalk, Hawaii and Snow Mountain viruses were defined as separate serotypes (Table 1). More recently, the antigenic relationships between these viruses have been examined by ELISA using hyperimmune antisera raised against virus-like particles. By this method, Norwalk, Mexico and Grimsby viruses are antigenically distinct.

The genetic relationships between the viruses within the calicivirus and the picornavirus families are being unraveled by sequence analysis of nucleic acids amplified directly from stools using reverse transcription–polymerase chain reaction (RT-PCR). Sequence comparisons of the full-length capsid protein of a number of viruses belonging to the four genera of the calicivirus family and of picornaviruses can be used to generate phylogenetic relationships but detailed understanding of the biological significance of these relationships awaits successful cultivation of these fastidious viruses.

Host Range and Virus Propagation

Attempts to cultivate the Norwalk and related viruses in tissue or organ culture have been unsuccessful. Efforts to find an animal model that develops illness after virus administration also have failed. However, chimpanzees become infected subclinically following administration of the Norwalk virus by the alimentary route; they shed soluble Norwalk antigen in the feces and develop serologic evidence of infection by IEM and radioimmunoassay (RIA).

Epidemiology

Human caliciviruses are the major known etiologic agents of acute nonbacterial gastroenteritis that cause clearly defined outbreaks that affect adults, school-aged children and family contacts; infrequently, these outbreaks involve infants and young children. Outbreaks occur in all seasons of the year in various communal settings, including schools, recreational and swimming facilities, cruise ships, restaurants, families and nursing homes. In an analysis of 74 gastroenteritis outbreaks from 1976 to 1980 (most of which were selected because they were nonbacterial), 42% were associated with Norwalk virus. These were minimal estimates because assays to detect evidence of infection were not available for most members of this group of agents until recently. Recent epidemiologic data using new assays to detect human caliciviruses indicate almost all (>90%) of reported gastroenteritis outbreaks in the United States and the Netherlands are caused by these viruses.

The prevalence of serum antibody to Norwalk virus in individuals in developed and developing countries is markedly different: in developing countries, antibody is acquired early in life and ≥90% of individuals have antibody by 2 years of age. In the United States, such antibody is acquired slowly during childhood and more rapidly in adulthood so that, by age 50, at least 80% of individuals have antibody. Although Norwalk virus may be associated with mild gastroenteritis in infants and young children in developing countries, there is no evidence that it causes severe, life-threatening diarrhea in this age group.

Transmission

Human caliciviruses are highly contagious. Transmission of infection occurs by the fecal–oral route. The virus is found in the feces and has been detected in vomitus. A common source of infection, such as contaminated water or food, is frequently described. Person-to-person spread has also been demonstrated. Airborne transmission has been proposed because of the rapid spread of illness when a common source is not identified. Recent studies have shown that the virus may also be spread by individuals with subclinical infections, and virus can be shed for a longer time than previously recognized (≥ 14 days postinfection).

Pathogenicity

Volunteers inoculated with Norwalk or Hawaii viruses develop characteristic transient histopathological mucosal lesions of the upper small intestine that includes broadening and blunting of the villi along with microvillus shortening on an intact mucosa. Mononuclear cell infiltration of the lamina propria and cytoplasmic vacuolization are also observed. Brush-border enzyme levels (alkaline phosphatase, sucrase and trehalase) are decreased during acute illness. Adenylate cyclase levels in the jejunal mucosa are not increased. The gastric and rectal mucosa are not affected histologically. Nausea and vomiting, common characteristics of this infection, may be caused by a delay in gastric emptying.

Clinical Features of Infection

Norwalk virus illness may begin abruptly with vomiting or diarrhea, or both. The spectrum of illness may vary widely in individual patients. For example, in experimentally infected adults, one volunteer vomited 20 times and required parenteral fluid therapy, whereas a second volunteer had no vomiting but eight diarrheal stools. The relative frequency of these and other symptoms was recently described for 50 volunteers challenged with Norwalk virus and is similar to those seen in natural outbreaks and in infection with other human caliciviruses. Of 50 volunteers orally administered Norwalk virus, 41 (82%) became infected; of these infections, 68% were symptomatic and 32% asymptomatic (Table 2 ). The most common symptoms with clinical illness are nausea, malaise and abdominal cramps. Diarrhea is usually watery without mucus, blood or leukocytes. Norwalk illness is usually mild, lasting 12–48 hours. It should be noted, however, that infrequently severe Norwalk virus gastroenteritis has been observed in middle-aged patients; in addition, Norwalk virus gastroenteritis has also been a contributing factor to the death of elderly, debilitated individuals.

Table 2.

Response of 50 adult volunteers (19–39 years old) given Norwalk virus

Response	% of infected volunteersa(n=41)	% of uninfected volunteers (n=9)
Seroconversion	98	0
Antigen excretion	88	0
Infection	100	0
Asymptomatic	32	0
Symptomatic	68	0
Symptoms with clinical illness (n = 28)
Diarrhea	86	0
Vomiting	57	0
Nausea	96	10
Abdominal cramps	96	0
Headache/bodyache	96	40
Chills	36	0
Fever (>37.8°C)	32	0

Data from Graham et al (1994).

^aInfection determined by antigen shedding and/or antibody response.

Immune Response

The mechanism of immunity to human calicivirus remains an enigma as it deviates from the characteristic pattern observed with most acute infectious illnesses in several ways: (1) adults are highly susceptible to both naturally occurring and experimentally induced Norwalk virus illness, as approximately 50% of unselected volunteers become ill after challenge; (2) the presence of preexisting serum antibody (measured by ELISA) does not appear to correlate with protection; and (3) short-term immunity has been demonstrated but long-term immunity may be absent. For example, 12 volunteers who were challenged with Norwalk virus on two separate occasions showed markedly different clinical responses: six became ill after initial and subsequent rechallenge 27–42 months later, whereas the others failed to develop illness following either challenge. Serum antibody did not correlate with protection. Paradoxically, individuals who became ill possessed higher levels of serum antibody to Norwalk virus than those who were resistant to challenge. The lack of correlation of serum antibody with protection may reflect the fact that the serum antibody detected was ELISA binding antibody and not neutralizing antibody. Correlation between antibody and protection may be seen when local intestinal neutralizing titers can be measured.

Diagnosis

A specific diagnosis of Norwalk virus gastroenteritis in an individual patient cannot be made on the basis of clinical signs and symptoms. Until recently, diagnosis remained essentially a research problem because reagents were not readily available. Diagnosis is made by detection of Norwalk virus in a stool specimen or the demonstration of a fourfold or more antibody increase in a patient's paired sera. Detection of a serologic response is more efficient than virus detection for demonstrating evidence of Norwalk virus infection. Early assays for these viruses included IEM (for the entire group), and RIA and/or ELISA for the Norwalk, Hawaii and Snow Mountain viruses, as well as immune adherence hemagglutination assay (IAHA) for the Norwalk virus. The cloning of Norwalk virus provided a major impetus to the study of the natural history of Norwalk virus infection because of the potentially unlimited source of antigen in the form of recombinant baculovirus-expressed virus-like particles (Fig. 1b). This capsid antigen shares antigenic specificity with native Norwalk virus by IEM and ELISA. The availability of animal antisera to recombinant Norwalk VLPs has also enabled the development of an ELISA for virus antigen detection. The antigen ELISA is highly sensitive but also highly specific. Similar antigen ELISAs using VLPs from Grimsby, Hawaii, Snow Mountain, Mexico and Sapporo viruses are being developed, as well as other assays that can detect more broadly cross-reactive antigenic epitopes. Until these assays become widely available, detection of the viral genome by RT-PCR is a favored method for diagnosis. Because diagnostic reagents previously were not available, it was suggested from analysis of numerous Norwalk virus outbreaks that a provisional diagnosis of Norwalk virus infection can be made if the following criteria are met: (1) bacterial or parasitic agents are not detected; (2) the incubation period is 24–48 hours; (3) vomiting occurs in at least 50% of ill individuals; and (4) the mean or median duration of illness is 12–60 hours.

Treatment

There is no specific antiviral therapy for Norwalk virus gastroenteritis: treatment therefore consists of fluid and electrolyte replacement therapy. Oral administration of isotonic fluids is usually sufficient for replacement of fluid and electrolyte loss. However, parenteral administration of fluids may be required if the vomiting or diarrhea is too severe to be managed by oral fluid replacement. Oral administration of bismuth subsalicylate can significantly reduce the severity and duration of abdominal cramps in volunteers challenged with Norwalk virus; the median duration of gastrointestinal symptoms was reduced from 20 hours to 14 hours, but this treatment did not affect significantly the number, weight and water content of stools and the level of virus excretion.

Prevention and Control

Because of the highly infectious nature of Norwalk virus infection, careful handwashing and effective disposal of contaminated material should help to reduce transmission. Careful attention to hygienic precautions in the preparation of food and monitoring the purity of drinking water or swimming facilities should also be practiced. Although the mechanisms of immunity to Norwalk virus remain unclear, the increasing recognition of the importance of these infections suggests that a vaccine might be useful if it is shown to be effective. Recombinant Norwalk virus-like particles represent one vaccine candidate. These VLPs are highly immunogenic when administered orally or intranasally to mice without adjuvant and they have also been shown to be safe and immunogenic when administered orally to volunteers. It is likely that evaluation of the immune response to these particles will help us understand immunity to these enteric infections.

Future Perspectives

Following the discovery of Norwalk virus and other human caliciviruses, a major impediment to their characterization was the inability to propagate them in tissue culture. Thus, all studies relied on the availability of particle-positive stools as the source of antigen. In spite of this limitation, important advances were made in the elucidation of the epidemiology, natural history, immunology and characterization of this group, with special emphasis on the prototype Norwalk strain. The cloning, characterization and expression of human calicivirus genomes represent another major breakthrough in the study of these viruses. Recent epidemiologic studies (allowed by new diagnostic assays) indicate that the epidemiology of infections with these viruses is not well understood, and infections are more common than realized previously. We predict that as human caliciviruses are studied, they will be found to play a significant role(s) in both acute and chronic human illness whose etiologies are currently unknown.

See also:

DIAGNOSTIC TECHNIQUES | Detection of Viral Antigens, Nucleic Acids and Specific Antibodies; DIAGNOSTIC TECHNIQUES | Isolation and Identification by Culture and Microscopy; Enteric Viruses.