Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
alanine aminotransferase
fulminant hepatic failure
Guillain-Barré syndrome
hepatitis E virus
human immunodeficiency virus
open reading frame
For many years hepatitis E virus (HEV) was thought to be largely restricted to endemic developing countries and only of clinical relevance in developed countries in returning travelers. In the last few years this notion has been shown to be mistaken, as hepatitis E is endemic in many developed countries. In such settings it is largely a porcine zoonosis caused by HEV genotype 3 (HEV3) and HEV genotype 4 (HEV4), and human infection is surprisingly common. Hepatitis E is, therefore, an infection of global significance and is the commonest cause of acute viral hepatitis worldwide.
Hepatitis E was not recognized as a distinct disease until 1980, when waterborne epidemics of hepatitis in India previously thought to have been caused by hepatitis A virus (HAV) were shown to have occurred in persons who were already immune to HAV. Three years later the etiologic agent (tentatively designated as epidemic non-A non-B hepatitis , or enterically transmitted non-A non-B hepatitis ) was transmitted to a volunteer following ingestion of a pooled fecal extract from Soviet troops with unexplained hepatitis stationed in Afghanistan. The viral genome was cloned and sequenced in 1990, and the virus was renamed hepatitis E virus .
In retrospect it is probable that most waterborne clinical hepatitis occurring in developing countries in the first half of the twentieth century or earlier was hepatitis E and not hepatitis A. For more than 20 years following its discovery, HEV was thought to be largely restricted to endemic areas in Asia, Africa, and Mexico, where it causes sporadic cases of hepatitis and more dramatic outbreaks involving thousands or tens of thousands of cases. More recently it has become clear that locally acquired hepatitis E is a health issue also in developed countries.
According to the International Committee on Taxonomy of Viruses, HEV is classified in the family Hepeviridae . There are several genotypes found in mammals, birds, and fish. A classification that divides the family Hepeviridae into two genera, Orthohepevirus and Piscihepevirus , has been proposed. Within the genus Orthohepevirus , four species have been designated that infect mammalian species ( Orthohepevirus A , Orthohepevirus C , and Orthohepevirus D ) and avian species ( Orthohepevirus B ). Orthohepevirus A contains HEV strains that are able to infect humans and these are divided into four major genotypes. HEV genotype 1 (HEV1) and HEV genotype 2 (HEV2) have been recovered only from humans, whereas HEV3 and HEV4 have been recovered from humans and a number of other animal species such as swine, wild boar, and deer. Other genotypes, including strains from rabbits, wild boars, and camels, also belong to Orthohepevirus A . All genotypes of HEV infecting humans belong to the same serotype. The species Orthohepevirus B (infecting chicken), Orthohepevirus C (infecting rat and ferret), and Orthohepevirus D (infecting bat) and the genus Piscihepevirus (infecting trout) contain strains that do not infect humans.
HEV is a small spherical virus, has an icosahedral capsid, and is quasi-enveloped. Whereas the virus sheds in feces and in the external environment is naked, the virus circulating in blood is associated with lipids. The naked particle is approximately 27 nm to 34 nm in diameter and the density is 1.27 g/cm 3 in sucrose density gradients. By contrast, the particle associated with lipids (quasi-enveloped HEV [eHEV]) is larger and the density is 1.15 g/cm 3 . The virions produced in cell culture systems are also associated with lipids. The host-derived membranes of eHEV protect the virion from antibody-mediated neutralization and could play an important role in cell entry.
The HEV genome is a single-stranded, positive-sense RNA of approximately 7.2 kb. The genome consists of a short 5′ noncoding region that is capped with 7-methylguanosine, three open reading frames (ORFs), and a short 3′ noncoding region that is terminated by a stretch of adenosines ( Fig. 36-1 ).
ORF1 encodes a nonstructural protein of approximately 1700 amino acids containing several functional domains. These include a methyltransferase/guanyltransferase responsible for capping the viral genome, a papain-like cysteine protease, a macrodomain, a helicase exhibiting RNA 5′-triphosphatase, nucleoside triphosphatase, and RNA unwinding activity, and an RNA-dependent RNA polymerase. A variable region containing a proline-rich hinge, called the polyproline region , is located between the protease and the macrodomain (X region). The PPR is an intrinsically disordered region where segments of human gene have been identified in vitro and in vivo . This region could be involved in virus adaptation. A greater heterogeneity of polyproline region quasispecies in the acute phase of HEV infection was found in immunocompromised patients who develop chronic infection than in individuals with resolving infection.
ORF2 encodes the viral capsid protein of 660 amino acids that forms the capsid through dimerization. Each monomer contains a shell domain, a middle domain, and a protruding domain. During assembly, capsid monomers self-assemble into dimers and subsequently into decamers that encapsidate the viral RNA. The virion-sized capsid is a T = 3 icosahedral capsid, whereas the recombinant HEV capsid protein can self-assemble into a T = 1 viruslike particle that retains the antigenicity of the native HEV virion. Immunologic and structural studies of the capsid protein have contributed to the development of hepatitis E vaccine. A high heterogeneity of HEV quasispecies in the region encoding the middle and protruding capsid domains was associated with chronic evolution of HEV infection in immunocompromised patients.
ORF3 encodes a small protein of 133 (for HEV3) or 114 amino acids that is involved in the release of virus from cells. The ORF3 protein interacts with the capsid protein in a manner dependent on phosphorylation of ORF3 at Ser80. The ORF3 protein also contains a PSAP motif that specifies interactions with host endosomal complex required for transport I proteins involved in the budding of many enveloped viruses.
Despite the recent development of cell culture systems and animal models, the HEV life cycle is still poorly understood ( Fig. 36-2 ). HEV particles attach to liver cells via heparin sulfate proteoglycans and other potential receptors and then undergo clathrin-mediated endocytosis. Heat shock protein 90 and tubulin seem to be involved in the early intracellular trafficking of the virus, but where and how the virus becomes uncoated and its RNA released are unknown. The virus RNA is directly translated into ORF1 polyprotein. Whether the polyprotein is cleaved into its separate functional units or functions as a single multidomain protein remains to be answered. The viral methyltransferase, protease, helicase, and polymerase activities are used to replicate the genomic RNA into negative-sense RNA intermediates that then serve as templates for the synthesis of genomic and subgenomic positive-sense RNAs. The subgenomic RNA is translated into the ORF2 and ORF3 proteins. The positive-sense genomic RNA is packaged into progeny virions. The ORF3 protein contains a PSAP motif that is necessary for the release of virions from infected cells. Tumor susceptibility gene 101 and the enzymatic activities of vacuolar protein-sorting protein 4A and vacuolar protein-sorting protein 4B are both involved in this process. HEV may use the multivesicular body pathway for virion release. HEV particles could acquire a membrane when they are released with internal vesicles from the multivesicular body via the cellular exosomal pathway.
The variability of the nucleotide sequences among the four major HEV genotypes is approximately 22% to 27%. These genotypes have been divided into several subgenotypes : 5 for HEV1 (1a, 1b, 1c, 1d,1e), 2 for HEV2 (2a and 2b), 10 for HEV3 (3a, 3b, 3c, 3d, 3e, 3f, 3g, 3h, 3i, and 3j), and 7 for HEV4 (4a, 4b, 4c, 4d, 4e, 4f, and 4g). On the basis of comparisons of full-length genome sequences, HEV3 variants were divided into two major clades corresponding to HEV3a, HEV3b, HEV3c, HEV3i, and HEV3j, and HEV3e, HEV3f, and HEV3g. Rabbit strains with close sequence homology to HEV3 have recently been described in both rabbits and humans. In addition, fragments of human genes have recently been found inserted into the polyproline region of HEV RNA obtained from chronically infected patients (part of the S19 ribosomal gene, tyrosine aminotransferase, or inter-alpha-trypsin inhibitor) or from cell culture systems (portions of the S17 ribosomal gene). These strains with inserted sequences replicated better in vitro .
Worldwide, hepatitis E occurs in two different patterns that correlate with the distribution of infecting viral genotypes ( Fig. 36-3 ). HEV1 is endemic in Asia and parts of Africa, is limited to humans, and is waterborne, often causing large outbreaks. HEV2 has similar epidemiology but is geographically more restricted to Africa and Latin America. HEV3 and HEV4 are zoonotic, affecting a wide range of mammalian species, particularly pigs. HEV3 is very widely dispersed, whereas HEV4 is most common in East Asia. The zoonotic genotypes generally result in sporadic cases of hepatitis E rather than outbreaks. Given the varied routes of transmission and geographic distribution of the different viral genotypes, it is not surprising that the epidemiology of the infections differs by genotype. The reported hepatitis E epidemiology depends very much on testing practices as well as the underlying pattern of infection. In less developed countries, diagnostic tests may not be performed because of lack of access to healthcare, or the costs of testing and reporting systems may not be robust. In more developed countries, lack of awareness among clinicians or lack of available assays mean that reported rates underestimate the true incidence. With these provisos in mind, the situation in different geographic areas is described in the following sections ( Table 36-1 ).
HEV1 and HEV2 | HEV3 and HV4 | |
---|---|---|
Geographic distribution | Asia (HEV1) Africa (HEV1 and HEV2) Mexico (HEV2) |
Worldwide, including developed countries (HEV3) Japan, China, Europe (HEV4) |
Source of infection | Obligate human pathogen | Zoonotic: primary host is pigs; deer, wild boars, rabbits, and shellfish are other hosts implicated in human infection Blood supply |
Route of infection | Orofecal via infected water | Oral via consumption of infected pork Parenteral, iatrogenic: human to human via blood products |
Infection from blood supply | Low | Possible |
Outbreaks | Yes | No |
Intrafamilial spread | Maybe | No |
Clinical attack rate | 1 : 2 | <1 : 10 |
Demographics | Mainly affects young adults (male-to-female ratio 2 : 1) | Mainly affects older men (median age 63 yr, male-to-female ratio 3 : 1) |
Chronic infection | No | Yes, in the immunocompromised Rapidly progressive liver disease if untreated in some: 10% of those infected are cirrhotic within a few years. Viral clearance is usually achieved with reduction in immunosuppression and/or a 3-month course of ribavirin monotherapy |
Can second HEV infections occur? | Yes, but not well documented | Yes Not well documented for HEV3 Well documented in HEV4: more likely in women who have a milder hepatitis than that seen in primary infections |
Clinical course | Self-limiting hepatitis in most | Self-limiting hepatitis in most |
Pregnancy | Mortality 25% | Increased mortality not seen |
Underlying chronic liver disease | Increased mortality | Increased mortality |
Neurologic sequelae | Yes, poorly documented | Yes |
Hepatitis E first came to prominence as the cause of large waterborne outbreaks of acute hepatitis and jaundice in the Indian subcontinent and China. In these areas there is thought to be continuous low-level transmission interspersed with episodes of intense transmission via fecally contaminated drinking water when sanitation breaks down (e.g., during heavy rains). Waterborne infection results in the exposure and infection of large numbers of susceptible people in a short time and results in dramatic outbreaks, such as the Kanpur (India) epidemic in 1991 when an estimated 79,000 people were affected.
Direct person-to-person spread of the virus seems less efficient than waterborne spread because secondary cases are rarely detected among household contacts of individuals with hepatitis E during outbreaks. In Africa, HEV1 and HEV2 have wide distributions that may overlap. Both genotypes have been recorded from outbreaks among displaced people as well as sporadic cases. Although waterborne transmission may account for many of these cases, there is evidence from a large outbreak of HEV1 infection in Uganda that intrahousehold transmission does occur.
During outbreaks of hepatitis E, the attack rate of clinically apparent disease is reported to be between 1% and 15% in India, but much higher rates have been observed elsewhere; for instance, during a Ugandan outbreak the rate was 30.5%. The proportion of asymptomatic cases during outbreaks is unclear but the rate of infection is likely to be much higher than the clinical attack rate. Previous exposure and subsequent immunity must determine the proportion of the exposed population susceptible to infection, and this probably plays a large role in determining the attack rate. Reinfection does occur, possibly when antibody levels fall below a critical value.
During outbreaks the age group most affected is young adults, particularly males. For instance, during the Kanpur epidemic, 81% of affected individuals were aged 10 to 39 years and the odds ratio for infection in males versus females was 1.73. Serious infection, with associated high mortality, disproportionately affects pregnant women and people with preexisting liver disease (see the section entitled “ Complications of Hepatitis E Virus Infection ”). This predilection for pregnant women is a specific feature of hepatitis E, which may point to the cause of outbreaks both in the past and in the present. Outside these special patient groups, mortality rates are generally low, but in vulnerable groups they may be higher; in an outbreak among displaced persons in Sudan, the death rate was 17.8%.
Hepatitis E has been recognized in travelers returning from endemic countries since it was first described, and was regarded as an uncommon imported infection in developed countries until approximately 15 years ago. The detection of HEV3 and HEV4 strains from individuals with hepatitis in the United States, Japan, China, and Europe with no history of recent travel made it clear that infection could be locally acquired in these regions. At approximately the same time, genetically very similar strains were detected in the same areas from other animals, including wild and farmed pigs and deer. Latterly, HEV has been detected in retail meat products in these countries. In a few cases, foodborne transmission has been proven when the same virus strains have been detected in epidemiologically linked foods and infected humans. Implicated foods include raw deer meat, pig liver sausage, wild boar meat, and shellfish. Direct animal contact may be the source of infection in some cases, evidenced by higher rates of anti-HEV antibody in pig farmers and veterinarians. Environmental contamination with animal feces may be another source, either directly via contaminated water or indirectly via contaminated food crops ( Fig. 36-4 ). However, there is little direct evidence of either route, and in the great majority of cases the source of infection is undetermined. A few case clusters have been reported in relation to common food sources or exposures but there is very little evidence of person-to-person spread, except via the human blood supply. For these reasons HEV3 and HEV4 are believed to be largely zoonotic in origin.
As noted already, virtually all hepatitis E cases in developed countries are sporadic and zoonotic. As with infection in endemic areas, males are affected more frequently than females (male-to-female ratio >3.1) and the mean age of those affected (>50 years) is much higher than in endemic areas. There are no reports of excess mortality in pregnant women with HEV3 or HEV4 infection. In a small minority of cases HEV has been transmitted from human to human via blood, as HEV RNA has been detected in blood products in many countries and transfusion-related infection has clearly been documented in recipients of blood products ( Tables 36-2 and 36-3 ). Transfusion-transmitted HEV accounts for approximately 1% of human infections with HEV in England.
Country/Region | HEV RNA–Positive Blood Donors | HEV IgG Seroprevalence (%) | Reference |
---|---|---|---|
Netherlands | 1 : 600 1 : 2671 |
27.0 | Zaaijer Slot et al. |
Midi-Pyrénnées, southwestern France * | 1 : 1595 | 52.5 | Gallian et al. Mansuy et al. |
France * | 1 : 2218 | Gallian et al. | |
Germany | 1 : 1200 1 : 4525 |
29.5 | Vollmer et al. Baylis et al. Wenzel et al. |
Japan | 1 : 1781 | Fukuda et al. | |
China † | 1 : 1493 | 32.6 | Guo et al. |
* Deconstructed solvent-detergent treated mini pools.
† Of those with HEV viremia, 57% were genotype 1 and 43% were genotype 4.
Country | HEV RNA–Positive Blood Donors | HEV IgG Seroprevalence (%) | Reference |
---|---|---|---|
England | 1 : 2848 1 : 7000 |
* 12.0 16.0 |
Hewitt et al. Ijaz et al. Beale et al. Dalton et al. |
Sweden | 1 : 7986 | Baylis et al. | |
Austria | 1 : 8416 | 13.5 | Fischer et al. |
United States | Nil Nil * |
16.0 | Baylis et al. Xu et al. |
Scotland | 1 : 14,520 | 4.7 | Cleland et al. |
Australia | Nil † | 6.0 | Shrestha et al. |
New Zealand | NA | 4.0 | Dalton et al. |
Fiji | NA | 2.0 ‡ | Halliday et al. |
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here