Heterogeneity and Seroprevalence of a Newly Identified Avian Hepatitis E Virus from Chickens in the United States

Abstract

We recently identified and characterized a novel virus, designated avian hepatitis E virus (avian HEV), from chickens with hepatitis-splenomegaly syndrome (HS syndrome) in the United States. Avian HEV is genetically related to but distinct from human and swine HEVs. To determine the extent of genetic variation and the seroprevalence of avian HEV infection in chicken flocks, we genetically identified and characterized 11 additional avian HEV isolates from chickens with HS syndrome and assessed the prevalence of avian HEV antibodies from a total of 1,276 chickens of different ages and breeds from 76 different flocks in five states (California, Colorado, Connecticut, Virginia, and Wisconsin). An enzyme-linked immunosorbent assay using a truncated recombinant avian HEV ORF2 antigen was developed and used to determine avian HEV seroprevalence. About 71% of chicken flocks and 30% of chickens tested in the study were positive for antibodies to avian HEV. About 17% of chickens younger than 18 weeks were seropositive, whereas about 36% of adult chickens were seropositive. By using a reverse transcription-PCR (RT-PCR) assay, we tested 21 bile samples from chickens with HS syndrome in California, Connecticut, New York, and Wisconsin for the presence of avian HEV RNA. Of the 21 bile samples, 12 were positive for 30- to 35-nm HEV-like virus particles by electron microscopy (EM). A total of 11 of the 12 EM-positive bile samples and 6 of the 9 EM-negative bile samples were positive for avian HEV RNA by RT-PCR. The sequences of a 372-bp region within the helicase gene of 11 avian HEV isolates were determined. Sequence analyses revealed that the 11 field isolates of avian HEV had 78 to 100% nucleotide sequence identities to each other, 79 to 88% identities to the prototype avian HEV, 76 to 80% identities to chicken big liver and spleen disease virus, and 56 to 61% identities to other known strains of human and swine HEV. The data from this study indicated that, like swine and human HEVs, avian HEV isolates are genetically heterogenic and that avian HEV infection is enzoonotic in chicken flocks in the United States.

Human hepatitis E is an important public health disease in many developing countries, and sporadic cases of hepatitis E have also been reported in some industrialized countries including the United States (1, 16, 20, 24, 30-33, 38, 40-41, 44, 45). The disease is generally transmitted by the fecal-oral route via contaminated water (1, 10, 22, 36). Hepatitis E virus (HEV), the causative agent of hepatitis E, is a positive-sense, single-stranded RNA virus (1, 7, 16, 18). The HEV genome contains three open reading frames (ORFs). ORF1 encodes viral nonstructural proteins, ORF2 encodes the putative capsid protein, and ORF3 encodes a small protein that may be involved in virus replication (1, 42, 44, 45, 49).

Seroepidemiological studies revealed that anti-HEV antibodies are present in numerous animal species, including pigs, rodents, chickens, dogs, cows, sheep, and goats, from both developing and industrialized countries (2, 3, 8, 19, 23, 29-31, 48; N. T. Tien, H. T. Clayson, H. B. Khiem, P. K. Sac, A. L. Corwin, K. S. Myint, and D. W. Vaughn, Abstract, Am. Trop. Med. Hyg. 57:211, 1997; S. A. Tsarev, M. P. Shrestha, J. He, R. M. Scott, D. W. Vaughn, E. Clayson, S. Gigliotti, C. F. Longer, and B. L. Innis, Abstract, Am. J. Trop. Med. Hyg. 59:242, 1998), suggesting that these animal species have been exposed to HEV-like virus. The first animal strain of HEV, swine HEV, was identified and characterized in 1997 from a pig in the United States (26). The U.S. swine HEV strain is genetically more closely related to two U.S. human HEV strains than to strains of HEV from other geographic regions (7, 26-28, 38, 39). Recently, many swine HEV strains have been genetically identified from pigs in the United States, Taiwan, The Netherlands, and Japan (17, 18, 34, 36, 43, 47, 48). Phylogenetic analyses revealed that, like human HEV (25, 38, 39, 44, 45), these strains of swine HEV from different geographic regions are also genetically heterogenic (17, 18, 34, 36, 43, 47, 48). Interspecies HEV transmission has been documented: a U.S. strain of human HEV infected pigs, and swine HEV infected nonhuman primates (11, 28, 46). Recent seroepidemiological studies indicated that individuals working with swine have higher risks of HEV infection (6, 15, 21, 29, 32). Taken together, these data strongly suggest that HEV is zoonotic and that there exist animal reservoirs for HEV.

More recently, another animal strain of HEV, designated avian HEV, has been genetically identified and characterized from chickens with hepatitis-splenomegaly (HS) syndrome in the United States (13, 14). The newly discovered avian HEV is genetically related to but distinct from other known HEV strains. Unlike swine HEV, which causes only subclinical infection in pigs, avian HEV is associated with HS syndrome in chickens (13, 14). HS syndrome in chickens was first described in 1991 in western Canada and is now recognized in eastern Canada and the United States (37). HS syndrome is characterized by above-normal mortality in laying hens at the age of 30 to 72 weeks. Infected birds usually have red fluid in their abdomens and enlarged livers and spleens (37). Microscopically, liver lesions vary from multifocal patches to areas of extensive hepatic necrosis and hemorrhage (13, 37). We have now successfully reproduced HS syndrome in adult specific-pathogen-free (SPF) chickens by infection with avian HEV (F. F. Huang et al., unpublished data). Since avian HEV is a novel virus and only one strain of avian HEV has been identified in the United States thus far, the molecular and serological epidemiology of this new virus is not known. In this study, we genetically characterized 11 additional avian HEV isolates from chickens with HS syndrome in the United States and assessed the prevalence of anti-HEV antibodies in chickens of different ages from five different states.

MATERIALS AND METHODS

Clinical specimens.

Bile samples used for genetic identification and characterization of avian HEV isolates in this study were collected from 21 chickens with HS syndrome in California, Connecticut, New York, and Wisconsin from 1993 to 2001 (Table 1). Twelve samples were positive by electron microscopy (EM) examination for 30- to 35-nm avian HEV-like virus particles, and nine samples were negative by EM.

TABLE 1.

Detection of avian HEV RNA from bile samples of chickens with HS syndrome in the United States

Strain	Geographic location	Yr of isolation	Breed/age (wk) of chickens	Clinical symptom(s)	Test results
					By EM	RT-PCR with B1-B2 primers	RT-PCR with helicase F-helicase R primers
WI966B	WI	2000	Layer/60	Decreased egg production and increased mortality	+	−	+
WI966G	WI	2000	Layer/60	Decreased egg production and increased mortality	+	+	+
NY449A	NY	2000	Layer/39	Decreased egg production and increased mortality	+	+	+
CA427	CA	1993	Layer/42	Decreased egg production and increased mortality	+	−	+
CA077	CA	1993	Broiler/56	Decreased egg production and increased mortality	+	+	+
CA242	CA	1994	Broiler/63	Decreased egg production and increased mortality	+	+	+
CA518.3	CA	1997	Broiler/64	Decreased egg production and increased mortality	+	−	+
CA518.5	CA	1997	Broiler/64	Decreased egg production and increased mortality	+	+	+
CA708A	CA	1997	Broiler/56	Decreased egg production and increased mortality	+	−	−
CA889	CA	1996	Layer/51	Decreased egg production and increased mortality	+	−	+
CA697A	CA	2001	Broiler/41	Increased mortality	+	+	+
CA697B	CA	2001	Broiler/41	Increased mortality	+	+	+
CA697C	CA	2001	Broiler/41	Increased mortality	−	+	+
WI318B	WI	2000	Layer/46	Decreased egg production and increased mortality	−	+	+
CT090A	CT	2000	Layer/35	Decreased egg production and increased mortality	−	+	+
CT690.2	CT	2000	Layer/39	Decreased egg production, increased mortality, and amyloid arthropathy	−	+	+
CT691.5	CT	2000	Layer/39	Decreased egg production, increased mortality, and amyloid arthropathy	−	−	−
CA658.2	CA	2000	Broiler/55	Decreased egg production	−	−	+
CA659.1	CA	2000	Broiler/55	Decreased egg production	−	−	−
CA571.1	CA	2000	Broiler/55	Decreased egg production	−	−	−
CA078A	CA	2000	Layer/80	Has gone through HS syndrome	−	−	+

Serum samples used for the seroprevalence study were collected from 1,276 chickens of different ages from 76 chicken flocks located in California, Colorado, Connecticut, Virginia, and Wisconsin (see Table 2).

TABLE 2.

Prevalence of antibodies to avian HEV in chickens of different ages in the United States

Location	Age	No. of birds positive/ no. tested (%)	No. of flocks positive/ no. tested (%)
WI	Adulta	20/20 (100)	1/1 (100)
CO	Adult	66/278 (24)	7/7 (100)
CA	<18 wk	58/381 (15)	13/26 (50)
	Adult	138/486 (28)	27/36 (75)
VA	<18 wk	13/24 (54)	1/1 (100)
	Adult	68/70 (97)	3/3 (100)
CT	Adult	17/17 (100)	2/2 (100)

^aChickens older than 18 weeks are generally considered adults.

ELISA to detect anti-HEV in chickens.

A truncated recombinant ORF2 protein of avian HEV was expressed in Escherichia coli and purified by affinity chromatography (14). The purified avian HEV truncated ORF2 protein was used as the antigen to standardize an enzyme-linked immunosorbent assay (ELISA) to detect anti-HEV immunoglobulin G (IgG) antibodies in chickens essentially as previously described for the ELISAs used to detect anti-HEV antibodies in humans and swine (26-29, 32). The purified ORF2 antigen of avian HEV was coated onto 96-well plates. A horseradish peroxidase-conjugated rabbit anti-chicken IgG (Sigma Chemical Co., St. Louis, Mo.) was used as the secondary antibody. The cutoff value was determined on the basis of the optical density (OD) values of the 85 adult chicken sera collected from a commercial SPF flock (Charles River Laboratory Inc., Wilmington, Mass.) and from two high-health university research flocks. The cutoff value was set conservatively at 0.30, which is about 6 standard deviations above the mean OD value of the normal chicken sera. All sera from 76 chicken flocks in the five states were tested in duplicate at a dilution of 1:100 in 0.05% Tween-phosphate-buffered saline blocking buffer containing 5% nonfat dried milk and 5% goat sera. Sera from SPF chickens were used for negative controls, and convalescent-phase sera from SPF chickens experimentally infected with avian HEV were included as positive controls. In general, the OD values for the positive sera from experimentally infected chickens were above 1.0 whereas those from naturally infected birds were about 0.5.

RT-PCR assay for detection of field isolates of avian HEV.

To identify field isolates of avian HEV, it was necessary to develop a reverse transcription-PCR (RT-PCR) assay capable of detecting avian HEV strains with significant genetic variation. Our previous studies showed that the helicase gene is the most highly conserved between avian HEV and other known strains of HEV (13, 14); therefore, primers for RT-PCR were targeted in the helicase gene region. A short stretch of the helicase gene sequence of the Australian chicken big liver and spleen disease virus (BLSV) has been reported, and it had about 80% nucleotide sequence identity to avian HEV (13, 35). The primers were selected from conserved helicase gene regions based on a sequence alignment of BLSV (35), the prototype strain of avian HEV (13), and other HEV strains. The primer sequences were of BLSV origin: forward primer (B1), 5′-GCTAGGCGACCCGCACCAGAT-3′; reverse primer (B2), 5′ GGTTAGCGCAACAATAGCATG-3′. The RT-PCR assay with this set of primers was able to detect the prototype strain of avian HEV and was therefore used to identify additional field isolates of avian HEV. Briefly, total RNAs were extracted with TriZol Reagent from 100 μl of bile samples from each of the 21 chickens with HS syndrome. Total RNAs were resuspended in DNase- and RNase-free water. RT was performed at 42°C for 60 min with Superscript II reverse transcriptase (GIBCO-BRL) using the reverse primer B2. A 5-μl volume of the resulting cDNA was amplified using AmpliTaq Gold DNA polymerase (Perkin-Elmer). The PCR parameters included a denaturation at 95°C for 9 min followed by 39 cycles of denaturation for 1 min at 94°C, annealing for 1 min at 42°C, and extension for 1 min at 72°C, with one final incubation at 72°C for 7 min. Negative and positive controls were included in each set of RT-PCR amplifications. The negative control was a water sample treated in the same way as the bile samples. The positive control was a fecal stock of avian HEV with a titer of about 10⁴ genomic equivalents per ml.

Nucleotide sequencing.

The expected PCR products of field isolates of avian HEV were purified using the glassmilk procedure with a GENECLEAN kit (Bio 101 Inc., Carlsbad, Calif.). PCR products amplified from 11 chicken bile samples from chicken flocks in four states were directly sequenced at the Virginia Tech DNA Sequencing Facility. Sequences of the PCR products were determined for both DNA strands.

Development of a more sensitive RT-PCR assay for the detection of avian HEV.

The RT-PCR assay described above was based primarily on the helicase gene of BLSV (35) and the prototype strain of avian HEV (12). Therefore, it was possible that more genetically divergent field isolates of avian HEV would not be detected by the RT-PCR assay. With the availability of the sequence information from 11 additional field isolates of avian HEV, we attempted to develop a more sensitive RT-PCR assay that is capable of detecting genetically divergent field strains of avian HEV. Briefly, the resulting sequences of the 11 field isolates of avian HEV and the prototype strain of avian HEV were aligned and a set of degenerate primers were designed based on the sequence alignment: forward primer (helicase F), 5′-TGGCGCACC(T)GTT(A)TCC(T)CACCG-3′; reverse primer (helicase R), 5′-CCTCA(G)TGGACCGTA(T)ATCGACCC-3′. The parameters for the RT-PCR assay with the degenerate primers helicase F and helicase R included a denaturation at 95°C for 9 min followed by 39 cycles of denaturation for 50 s at 94°C, annealing for 50 s at 46°C, and extension for 50 s at 72°C, with one final incubation at 72°C for 7 min. Positive and negative controls were the same as the ones described above.

Sequence and phylogenetic analyses.

The primer sequences used to amplify the avian HEV isolates were excluded from the sequence and phylogenetic analyses. The resulting 330-bp sequences of the helicase genes of the 11 field isolates of avian HEV were analyzed and compared with the corresponding regions of the prototype avian HEV, BLSV, and other known strains of swine and human HEVs available in the GenBank database by the MacVector computer program (Oxford Molecular Inc.). Phylogenetic analysis was conducted with the aid of the PAUP program (David Swofford, Smithsonian Institute, Washington, D.C., distributed by Sinauer Associate Inc. Sunderland, Mass.). A heuristic search with 1,000 replicates was used to produce a phylogenetic tree.

The geographic origins and the GenBank accession numbers of the nucleotide sequences of the HEV strains used in the phylogenetic and sequence analyses are as follows: JRA1 (Japan, AP003430), SWJ570 (Japan, AB073912), swine HEV (United States, AF082843), US1 (United States, AF060668), US2 (United States, AF060669), T1 (China, AJ272108), Burma (Burma, M73218), Myanmar (Myanmar, D10330), Hyderabad (India, AF076239), Madras (India, X99441), Nepal (Nepal, AF020486), hev037 (India, X98292), Xinjiang (China, D11092), Uigh179 (China, D11093), Hetian (China, L08816), Sar-55 (Pakistan, M80581), and Mexico (Mexico, M74506).

Nucleotide sequence accession numbers.

The sequences of the helicase gene region of the 11 avian HEV isolates have been deposited in the GenBank database with accession numbers AF531898 through AF531908.

RESULTS AND DISCUSSION

Avian HEV is enzoonotic in chicken flocks in the United States.

The prevalence of anti-HEV in commercial pig herds in the United States has been reported previously (26, 29). Seropositivity for swine HEV in pigs is age dependent: most pigs younger than 2 months were seronegative, whereas most pigs older than 3 months were seropositive (26, 29). In this study, we tested 1,276 serum samples from chickens of various ages in five states for the presence of antibodies to avian HEV (Table 2). The percentage of seropositive chickens varied greatly from flock to flock, ranging from 0% to 100%. Of the 76 flocks tested, 54 (71%) were positive for anti-HEV IgG. Similar to swine HEV, the prevalence of avian HEV antibodies in adult chickens (older than 18 weeks) was higher than that in younger ones: about 17% of chickens younger than 18 weeks were seropositive, whereas about 36% of adult chickens were seropositive. However, the age-related difference in avian HEV seropositivity in chickens is less definitive than that previously observed in swine (data not shown). Several flocks of adult chickens were also found to be negative for antibodies to avian HEV. It is possible that these negative flocks may have better biosecurity measures that could prevent the transmission of avian HEV. The antibodies to avian HEV detected in some 1-day-old chickens are probably maternal antibodies, which usually wane at about 3 weeks of age. So far, avian HEV has been isolated only from chickens in the United States. The BLSV identified in Australia is genetically and antigenically related to HEV (12, 13, 14), and it is likely that the big liver and spleen disease in Australia (4, 5, 12, 35) and the HS syndrome in North America (13, 14, 37) are caused by variant strains of the same virus.

Detection of avian HEV by RT-PCR.

Twenty-one bile samples from chickens with HS syndrome (increased mortality and decreased egg production) were examined for avian HEV-like virus particles by EM and for avian HEV RNA by RT-PCR. Twelve of them were positive for 30- to 35-nm virus particles by EM. RT-PCR results with the B1 and B2 primers, based primarily on BLSV sequence, showed that 7 of the 12 EM-positive bile samples and 4 of the 9 EM-negative bile samples were positive for avian HEV RNA (Table 1). The bile samples that tested negative with the B1 and B2 primers were further tested by a more sensitive RT-PCR assay with degenerate primers helicase F and helicase R based on the sequence alignment of 11 avian HEV isolates and the prototype avian HEV. Another four EM-positive and two EM-negative bile samples turned out to be positive for avian HEV RNA (Table 1). The six additional positive bile samples detected by the more sensitive RT-PCR assay with the degenerate primers were confirmed by sequencing (data not shown). Overall, 11 of 12 EM-positive and 6 of 9 EM-negative bile samples from chickens with HS syndrome were positive for avian HEV RNA by RT-PCR, indicating that RT-PCR is a more sensitive method than EM for detection of avian HEV. The RT-PCR assay developed in this study should be very useful for future avian HEV studies.

Field isolates of avian HEV are genetically heterogenic.

It has been reported that swine and human HEV isolates from different geographic regions are genetically heterogenic (17, 39, 44). The extent of genetic variation of avian HEV is not known, since only one strain of avian HEV has been characterized. The PCR products from the 11 avian HEV isolates amplified by the B1 and B2 primers were sequenced. The 330-bp sequences within the helicase genes were analyzed and compared to each other and to the corresponding regions of the prototype avian HEV, BLSV, and other known strains of HEV. Sequence analyses revealed that the avian HEV isolates identified in this study had 78 to 100% nucleotide sequence identities to each other, 79 to 88% identities to the prototype avian HEV, 76 to 80% identities to BLSV, and 56 to 61% identities to other known strains of human and swine HEVs (Table 3). At the amino acid sequence level, the 11 avian HEV isolates identified in this study displayed 90 to 100% sequence identities to each other, 89 to 98% identities to the prototype avian HEV, 78 to 79% identities to BLSV, and 55 to 62% identities to other known strains of human and swine HEVs.

TABLE 3.

Pairwise comparison of the nucleotide sequences of the partial helicase gene of 11 avian HEV isolates identified in this study and other viruses^a

Isolate	Nucleotide sequence identity (%) to:
	Sar-55	US2	Swine HEV	Mexico	BLSV	Avian HEV	CT090A	CA518.5	CA242	CA077	CA697C	CA697B	CA697A	NY449A	WI318B	CT690.2	WI966G
T1	76	75	75	73	56	57	56	57	56	56	56	56	56	56	57	57	56
Sar-55		75	75	78	59	60	59	61	59	59	59	59	60	59	60	59	61
US2			91	72	60	57	59	59	59	60	60	60	60	60	59	59	59
Swine HEV				75	60	58	59	60	60	60	59	59	59	60	59	60	59
Mexico					59	61	60	61	59	60	59	59	60	60	60	60	60
BLSV						77	76	78	79	77	76	76	76	77	78	80	77
Prototype avian HEV							79	86	80	79	79	79	80	80	88	80	86
CT090A								80	86	94	93	93	93	96	81	83	79
CA518.5									83	80	80	80	80	83	96	81	98
CA242										86	85	85	85	88	83	83	83
CA077											97	97	96	96	81	84	79
CA697C												100	99	95	80	83	78
CA697B													99	95	80	83	78
CA697A														94	80	83	78
NY449A															83	84	81
WI318B																80	96
CT690.2																	81
WI966G

^aThe other viruses include prototype avian HEV, BLSV, and other selected strains of HEV from each of the four recognized genotypes. The isolates identified in this study are indicated in bold type.

Phylogenetic analyses revealed that all the U.S. avian HEV isolates identified in this study clustered with the prototype avian HEV and BLSV but are distinct from other known strains of HEV (Fig. 1). Minor branches, indicating heterogeneity, also exist among avian HEV isolates. Whether avian HEV represents a fifth HEV genotype or belongs to a separate genus remains to be determined.

FIG. 1.

Phylogenetic tree based on the sequences of the helicase gene of avian HEV isolates from this study and other selected HEV strains. The tree was constructed with the aid of the PAUP program by using a heuristic search with 1,000 replicates and midpoint rooting option. The scale bar indicating the numbers of character state changes is proportional to the genetic distance. The 11 avian HEV isolates from this study are indicated in boldface type.

An increasing amount of data suggested that HEV infection is zoonotic. Recently it was reported that swine veterinarians (29, 32) and other pig handlers (6) may be at higher risk of HEV infection than normal blood donors. Therefore, it will be important to determine if avian HEV infection is also zoonotic and if chicken handlers have a higher risk of HEV infection, as swine handlers do.