Avian Hepatitis E Detections in the United States, from Diagnostics to Koch’s Postulates Open Access

Avian hepatitis E virus (aHEV) is a nonenveloped, single-stranded RNA virus (1) categorized as an Orthohepevirus B species within the family Hepeviridae (2). This virus causes hepatitis-splenomegaly syndrome (3), also denominated big liver and spleen disease (BLS) in Australia (4). The transmission of the virus mainly is by the fecal oral route (5), and the infection has been successfully reproduced via nasal and oral route inoculation in specific-pathogen-free (SPF) chickens (6).

In layers, aHEV increases weekly mortalities to 0.3%–1% on average and decreases egg production by approximately 20% (3). It is important to notice that this reduction in egg production is a consequence of the increased mortality although effects on the “hen housed egg production” might not be noticed if one focuses on the “hen per day egg production.” Most of the birds in the flock would appear normal, while the mortality will show livers that are mildly to severely enlarged and exhibit a friable texture and white mottling and may have subcapsular hematomas or attached blood clots on the surface (4,5). Additionally, complex infections are associated with fowl cholera or infectious bronchitis virus and some aHEV cases present regressive ovaries (7).

The first report of this virus causing reduced egg production and mortality in 42- and 56-wk-old laying hens was in California in 2001 (8). Another study in the United States in healthy adult layers between 16 and 104 wk reported a seroprevalence of 44.8% for anti-aHEV IgY antibodies. In addition, aHEV RNA was detected in 62.9% of pooled fecal samples (7). These results in addition to the results published by Huang and collaborators (5) provide a good idea of the presence and prevalence of this virus in poultry populations in the United States. Considering that this virus can be present in flocks without causing clinical problems (7) it is important to evaluate and understand the current epidemiology of this pathogen.

The diagnosis of aHEV is difficult due to the lack of an efficient virus recovery and amplification system. A study in 2022 reported the highest replication efficacy after intravenous inoculation of SPF chicken embryos in comparison to yolk sac inoculations (9), and, although studies report the isolation of the virus in eggs or cell cultures, methodologies are complicated and yields are very low (3,9). Considering aHEV isolation is not generally successful, diagnostics have been based on serology and molecular biology. In serology, ELISA is the most popular. Two indirect ELISA commercial kits have been designed to identify antibodies against aHEV (4).

Regarding molecular detection, reverse transcriptase polymerase chain reaction (RT-PCR) has been the most effective technique to detect segments of the aHEV genome. Numerous RT-PCR strategies such as RT-qPCR and nested RT-PCR have been documented (6,10,11,12,13,14,15). Among these different molecular techniques, SYBR Green RT-qPCR is characterized by its simplicity, low cost, fast detection times, and high sensitivity and specificity. However, only one validated test is reported in the literature (16). The molecular typing of aHEV is challenging due to limited information in sequence repositories. As of April 2024, only 20 complete sequences of the avian hepatitis E genome originating from different parts of the world have been deposited in GenBank, obstructing the understanding of the molecular epidemiology and distribution of this disease.

To overcome these challenges, experimental infections administering isolates or infected tissues or fluids in SPF chickens are an important step in fulfilling Koch’s postulates and validating diagnostic methods. A previous study showed that 60-wk-old SPF chickens infected with aHEV by the oronasal or intravenous route developed liver lesions such as subcapsular hemorrhages, hepatomegaly, and slight elevation of liver enzymes, providing valuable insights into the understanding of the pathobiology of the disease (14).

The objective of this report is to (a) describe the case history, (b) understand the seroprevalence of aHEV, (c) report on the development of an RT-qPCR for the identification and quantification of aHEV RNA in livers, spleens, and gallbladders, (d) report on the isolation of aHEV strains from field samples in SPF embryonated eggs, (e) report on molecular characterization of the strains associated with outbreaks in commercial egg flocks, and (f) report on the real effect of the virus in chickens by fulfilling Koch’s postulates.

During 2023, in the states of South Carolina, Florida, Arkansas, and Minnesota, various commercial table egg layer flocks between 40 and 52 wk of age, housed in conventional cages, reported mortality showing gross pathological changes resembling enlarged, tan, and hemorrhagic livers, splenomegaly, and blood-tinged exudate in the coelomic cavity (Fig. 1). These flocks also presented an increased weekly mortality averaging from 0.3% to 1% for a period of 8–12 wk after the onset of clinical signs. The hen mortality affected parameters such as hen-housed egg production (HHEP). This was confusing for some farmers that were used to looking at the hen daily egg production, which is not affected in these cases because mortality gets discounted from the flock size number every day.

To understand the seroprevalence of the disease in commercial layers, historical serum samples (n = 210) were analyzed for aHEV antibody detection with ELISA from flocks in the above-mentioned states with ages ranging between 24 and 105 wk. In addition, one bile fluid pool and tissue samples from 17 livers, 13 gallbladders, and 20 spleens were collected from flocks in South Carolina, Arkansas, Minnesota, and Florida from birds with ages ranging between 45 and 73 wk old to detect the virus with electron microscopy and RT-qPCR and to attempt virus isolation in SPF eggs.

Liver samples (n = 3) positive to aHEV by RT-qPCR were processed for negative contrast electron microscopy and analyzed using the described methodology (17). The identification of the virus was carried out via comparison of structural characteristics with the standard features established in the Report of the International Committee on Taxonomy of Viruses (18). One out of three RT-qPCR–positive liver samples showed viral particles that had a capsid with icosahedral symmetry, lacked an envelope, and exhibited a diameter of 27 to 32 nm (Fig. 2).

Serum samples were evaluated for aHEV antibodies using the commercial BLS ELISA kit (BioChek, UK, Ltd) using the manufacturer's guidelines. The cutoff for positivity of the test is 390 GMT. Out of 210 serum samples, 64 tested positive for aHEV antibodies, for a calculated seroprevalence of 30.5%. The seroprevalence differed with the age of the flock, with 18.4% (7/38) positivity between 20 and 50 wk of age; 38.8% (49/126) between 51 and 80 wk, and 17.4% (8/46) between 81 and 110 wk (Fig. 3).

Livers (n = 19), gallbladders (n = 13), bile (n = 1 pool), and spleens (n = 20) from flocks ranging from 45 to 73 wk old were tested for avian hepatitis E using an in-house SYBR-Green RT-qPCR described below. Viral RNA extraction was conducted through a hybrid approach involving TRIzol (Invitrogen, Carlsbad, CA) and the Direct-zol RNA Miniprep Plus Kit (Zymo Research, Orange, CA). Primers were designed using 22 full-genome sequences of aHEV published in GenBank aligned using the Clustal Omega1.2.2 method in Geneious Prime (version 2022.0.5; Biomatters Inc., San Diego, CA). Primers were designed to amplify a relatively conserved 191 bp portion of ORF3 ranging from positions 4724 to 4914 on GenBank accession EF206691 (Fig. 4). The forward primer (F1) sequence is 5′-GCTCATGCTYGCWATGTGYT-3′, and the reverse primer (R1) sequence is 5′-CTACRTCTGGTACCGTGC-3′. Primer efficiency was analyzed using the Multiple Primer Analyzer online tool (Thermo Fisher Scientific®, Waltham, MA) to detect self-dimers and determine optimal annealing temperatures. In silico specificity was tested with BLAST-NCBI to examine potential target sequences and determine alternative binding sites to other poultry pathogens, host, or common contaminants. While no dimers were detected, the nucleotide BLAST determined that these primers may nonspecifically bind to numerous sequences, including both Gallus gallus and Homo sapiens in several loci. It was therefore important to analyze any RT-qPCR results with melt curve analysis. Using the publicly available sequences, it was determined that melt curves should peak at an average of 88.3 C with a possible range from 87.1 to 89.1 C. Custom gBlocks (IDTH®, Coralville, IA) that synthesized to the target range on EF206691 were used to standardize the RT-qPCR and determine the limit of detection.

The QuantiTect® SYBR® Green PCR kit (Qiagen, Germantown, MD) was used according to the manufacturer’s instructions with a final primer concentration of 0.5 μM. The overall reaction comprised 25 μl of 2 × QuantiTect SYBR Green PCR Master Mix, 2.5 μl each of forward and reverse primers at 10 µM/μl, 0.5 μl of QuantiTect-RT Mix, 14.5 μl of DEPC H₂O, and 5 μl of template. The reactions were performed using a CFX96 Real Time System (Biorad, Hercules, CA) with the following thermal profile: reverse transcription for 30 min at 50 C, polymerase activation for 15 min at 95 C, and 45 cycles of 15 sec denaturation at 94 C, 30 sec annealing at 52 C, 30 sec extension at 72 C, and a 15 sec dimer denaturation at 83 C with data acquisition. Following amplification, melt curve analysis was performed between 65 and 95 C in 0.2 C increment data acquisition steps. Using 10-fold dilutions of the synthesized gBlocks from 10⁹ to one copy, the RT-qPCR efficiency was determined to be 81.8%, and the correlation coefficient (R²) was 0.963. The RT-qPCR sensitivity allowed us to consistently detect 100 or more starting template sequences, though several standards with 10 template sequences were detectable. Results showed a higher viral load (P < 0.05) in gallbladders compared to liver and spleen samples (Fig. 5).

RT-qPCR positive samples were selected for genotyping by either targeted or total RNA sequencing. An RT-PCR was designed to amplify the complete genome of aHEV from South Carolina and Arkansas samples. The forward primer F2 5′-TGGACGTCTCGCAGTTTGCAGAGTCCA-3′ covered nucleotides 31-52 (EF206691), and primer R2 5′-ACAATGCCCGAGATGGGAGGATTTC-3′ covered nucleotides 6641–6665 (EF206691). Primers were designed following the methodology previously described. Again, there is a possibility for the primers to nonspecifically bind to the host, and the forward primer has the potential to self-dimerize. For the reaction we used the SuperScript IV One Step RT-PCR system (Invitrogen®, Waltham, MA) according to the manufacturer’s instructions with primer final concentration of 0.5 µM and the following thermal profile: reverse transcription for 10 min at 50 C, RT inactivation for 2 min at 98 C, 35 cycles consisting of (1) denaturation for 10 sec at 98 C, (2) annealing at 72 C for 10 sec, and (3) extension at 72 C for 3.5 min, and a final extension step for 5 min at 72 C. Amplification of the products could not be confirmed by gel electrophoresis.

For the remaining samples, Florida and Minnesota, we used the Maxima™ H Minus Double-Stranded cDNA Synthesis kit (Thermo Fisher Scientific®, Waltham, MA) to generate double-stranded cDNA from total RNA according to the manufacturer’s instructions. RT-PCR and cDNA products were purified using Agencourt AMPure XP beads (Beckman Coulter, Indianapolis, IN). Subsequently, the amplicons were assigned barcodes via the Native Barcoding Expansion EXP-NBD104 and EXP-NBD114 (Oxford Nanopore Technologies [ONT], Oxford, UK). Samples were combined and a sequencing library created utilizing the Ligation Sequencing Kit SQK-LSK109 (ONT). The library was loaded onto ONT R9.4.1 flow cells and run 24 hr using a MinION sequencer (Mk1B). During the sequencing process, super-accuracy base calling was utilized, and a minimum read quality of Q10 was maintained via MinKNOW software.

Base-called reads that passed quality control were trimmed of barcodes and adapters with Porechop (https://github.com/rrwick/Porechop) using default settings. Trimmed reads were mapped to a reference strain (EF206691) using the Geneious mapper in Geneious Prime (version 2022.0.5, Biomatters Inc., San Diego, CA) with five-iteration fine tuning, minimum mapping quality of 30, trimming to reference, and consensus and contig output. Samples from South Carolina, Arkansas, and Minnesota presented an adequate number of filtered reads, while Florida samples displayed a higher number of filtered reads. The South Carolina samples had the highest mean genome coverage (Table 1). All sequences were submitted to GenBank and can be retrieved using the following accession numbers: PQ799636 (Arkansas), PQ799637 (Florida), PQ799638 (Minnesota), and PQ799639 (South Carolina).

The four obtained sequences were compared with a set of 22 aHEV genomic sequences obtained in GenBank and distributed in the seven genotypes described for aHEV in chickens (19). Sequences were aligned using Geneious Prime 2020.1.1 through the MAFFT plugin 2.0.5 (20). Phylogenetic trees were built using the maximum-likelihood method based on the GTR GAMMAI model with 1000 bootstrap replicates utilizing the RaxML plugin (21) in Geneious Prime. South Carolina and Arkansas samples contained viral sequences that were identified as genotype 3 with a homology of 93% to strain MW924815.1. This strain was associated with clinical cases of liver swelling, hemorrhages, and mortality increases at 20 wk of age in China’s Shandong Province in 2020 (22). Sequences obtained from samples sourced from farms in Florida and Minnesota were classified within genotype 2 showing a homology of 90.4% and 95.5% to strain EF206691.1. This strain was collected from healthy chickens in the United States in 2002 (Fig. 6) (23).

Inocula were prepared from either aHEV-positive livers and gallbladders or bile. Livers and gallbladders were macerated and suspended in 1:10 wt/vol of tryptose phosphate broth (TPB) solution containing antibiotic/antimycotic. Bile was diluted 1:10 in TPB + antibiotic/antimycotic. The suspensions were sonicated and then centrifuged at 5000 g for 5 min at 4 C, and the supernatant was collected and left to cool on ice for 1 min. Samples were centrifuged again at 8000 g for 20 min at 4 C. Inocula were administered intravenously in 11-day-old embryonated SPF chicken eggs according to Clavijo et al. (24). Inoculated eggs were incubated for 9 days, and mortality was recorded daily by candling. At 9 days postinoculation, viable eggs were refrigerated at 4 C for 4 hr. Embryos were extracted and livers, gallbladders, and intestines harvested and processed as described above for the initial inoculum preparation. This harvest was used as the inoculum for the next passage. The presence of the virus was assessed in the initial inoculum and after the first and second passage using RT-qPCR. The virus concentration after a single passage was less than the estimated concentration in the original bile and liver-gallbladder sample. After the second passage, we were unable to detect the virus by RT-qPCR in embryonic tissues (Table 2).

To confirm the etiology of these cases, we designed an experiment to fulfill Koch’s postulates. The University of California is an AAALAC-accredited institution. Animal experiments were performed under the IACUC authorization no. 22572.

Bile and liver samples positive to aHEV by RT-PCR were processed as shown in the inoculum preparation for virus isolation and administered to SPF chicks hatched at our facility. Chicks were randomly divided into three separate groups, with the two experimental groups containing 15 birds each, while the control group contained nine birds. One of the experimental groups was challenged with liver extracts, while the other group was challenged with bile extracts, both containing 10⁵ aHEV genomic equivalents per 200/µl in the first and second challenges. In the third challenge, the bile and liver extracts contained 10⁸ aHEV genomic equivalents per 200/µl. The virus was given in three separate times and in different routes i.e., oral, oro-cloacal, and intravenous at 4.5, 7.5, and 10 wk of age, respectively. The inocula at 4.5 and 7.5 wk of age were identified as genotype 2 originating from Florida, and the one used at 10 wk was identified as genotype 3 from Arkansas.

Before inoculation 10 cloacal swabs and 10 blood samples were collected to confirm the absence of aHEV and anti aHEV antibodies. These samples were repeated at 5 and 11 days postchallenge. Necropsies were performed at 15 days post-second challenge and 21 days post-third challenge. Liver and spleen samples were collected for histopathology. Liver, spleen, gallbladder, and bile were collected for viral load assessments. Liver and spleen to body weight ratios were calculated. Liver enzyme function was assessed by blood biochemistry assessments such as albumin/globulin ratio, total protein, and aspartate aminotransferase (AST).

All birds tested negative to the presence of antibodies against aHEV in blood at all time points (data not shown). Birds tested negative to the presence of aHEV before challenge. Very low viral loads were detected after the first challenge in cloacal swabs (data not shown). The most relevant findings were found in samples obtained on the second necropsy (21 days after the third challenge), particularly pertaining to blood biochemistry and organ/body mass ratio. Regarding blood biochemistry, there was a statistically significant decrease in the albumin/globulin ratio in the group of birds challenged with bile compared to the control group (Fig. 7A). In addition, total serum protein was elevated (P < 0.05) in the bile-challenged birds compared to the control group (Fig. 7B). On the other hand, AST levels were within normal limits (60–220 UI) in all the birds (25).

An increase in the liver/body mass ratio was detected in the bile-challenged compared to the control group (P < 0.05) (Fig. 7C). Gross pathology changes were observed in both infected groups, while the control group did not exhibit gross lesions (Fig. 8A). The liver-extract–challenged group showed pale liver discoloration with friable texture, slight enlargement, and rounded edges (Fig. 8B). The liver from birds challenged with bile presented similar liver features of pale discoloration, friable texture, slight enlargement, and rounded edges (Fig. 8C). Regarding histopathology, the group inoculated with liver homogenate showed lesions including mild to moderate focally extensive hemorrhage in the liver (Fig. 9A). One liver had focal vasculitis with infiltrations of lymphocytes and granulocytic cells (Fig. 9B). The spleen of birds inoculated with the liver homogenate ranged from no lesions to mild to moderate capsular thickening. Occasionally, small deposits of amorphous eosinophilic material within the white and red pulp were seen. This material was not positive for amyloid with Congo red staining. On the other hand, the group inoculated with bile conjugate showed lesions including mild hepatic necrosis, mild to moderate lymphoplasmacytic inflammation in portal regions, and mild to moderate bile duct hyperplasia (Fig. 10A,B). In the spleens, follicular hyperplasia and occasional deposition of acellular eosinophilic material was seen (Fig. 10C,D). This material was not positive for amyloid with Congo red staining.

Avian hepatitis E is an emerging and significant concern in adult laying hens associated with increased mortality and decreased HHEP due to accumulated mortality (7). Despite studies conducted more than 10 yr ago reporting high seroprevalence rates and viral RNA detection, the current epidemiology and impact of the disease in laying hens remains poorly understood (7). With the current data, it is impossible to establish patterns associated with age and productivity or have a clear understanding of the pathobiology of the disease.

This case report aims to address this knowledge gap through diagnostic and characterization work using samples obtained from laying hens exhibiting clinical issues related to mortality and drops in egg production. Our investigations were successful in detecting viral particles that resemble aHEV using electron microscopy. If these viral particles are indeed aHEV, this suggests an abundance of aHEV in livers considering the low sensitivity of the technique. Previous studies have reported a seroprevalence of aHEV-specific antibodies of ∼30% in commercial laying hens in the United States (10). These results are comparable to our findings. Higher anti-aHEV antibody titers were detected between 51 and 80 wk of age, which coincides with the post-egg production peak period. This result suggests highest infection close to peak of production and a potential stress-related susceptibility. In addition, concomitant infection with other pathogens such as Campylobacter hepaticus, Clostridium perfringens, and Escherichia coli cannot be ruled out. In fact, through our sequencing efforts, partial sequences of the mentioned pathogens were encountered in varying quantities in several of the samples. Challenge studies using multiple pathogens might help clarify this potential pathogen synergy. The seroconversion patterns found can be related to a cumulative exposure of the virus (26). The decrease in anti-aHEV antibody titers after 80 wk of age can be related to immune senescence (27) due to lymphocytes’ and macrophages’ lessened capabilities to respond to infections. More studies examining the seroprevalence of different flocks in different geographical locations in association with the detection of the pathogen as well as investigations of seroconversion patterns in different genetic lines are needed to better understand the infection dynamics.

A higher viral presence, assessed by RT-qPCR, was observed in the gallbladder in comparison to liver and spleen, while the liver exhibited a higher viral load than spleen indicating its potential for diagnostic purposes. These results are consistent with aHEV pathogenesis. After replication in the liver, aHEV is released into the gallbladder from hepatocytes and then excreted in feces (3). The high-aHEV viral RNA load detected in gallbladders indicates that this organ acts as a reservoir of the virus after replication in the liver and serves as a useful sample for virus detection in infected flocks.

aHEV is a poorly understood pathogen due to the lack of a good culture system for virus isolation and replication. Among the tested isolation strategies, the one that seems most successful in the literature is the intravenous inoculation using SPF chicken embryos (9). In practice, we observed a reduction of more than 50% of the virus RNA after the first passage and a complete absence of detectable virus after the second and third passages. These results reflect the lack of adaptation of the virus to this system. In addition, a breed susceptibility to this pathogen cannot be discarded.

Seven genotypes have been described for aHEV (19). In the United States only genotype 2 has been reported. We detected genotype 2 in samples from flocks located in Florida and Ohio and genotype 3 viruses in flocks located in South Carolina and Arkansas. This is the first detection of aHEV genotype 3 in the United States. Genotype 3 viruses were previously reported in chicken flocks in Hungary (15) and laying hens with clinical signs including drops in egg production in China (22). While no clear indication of vertical transmission exists, the introduction of genotype 3 into the United States may have occurred through the importation of parent stocks or fertile eggs from countries positive to this genotype. Another possibility of introduction is wild birds that might act as carriers of these viruses (28). Epidemiological surveillance to track the circulation aHEV and the emergence of genotypes might clarify the virus dynamics in the chicken population.

Koch’s postulate experiments showed effects in the albumin/globulin ratio and total serum protein concentrations. The albumin/globulin ratio reduction indicates inflammatory changes in the liver, which results in elevated globulin levels in response to inflammatory cytokines (25). In addition, we also noticed a total serum protein increase, which can be associated with an acute inflammatory response (25). On the other hand, AST levels were within normal limits (60–220 UI) in all the birds (25). These results are compatible with previous studies that found no significant changes in AST levels from chickens infected with aHEV (14). Regarding gross pathology, some livers of the infected birds were enlarged, pale, and friable and had rounded edges. The group infected with the bile contents had more severe lesions compared to the liver homogenate–inoculated birds. This might be due to changes in the virus when it is deposited in the bile after replicating in the liver that might increase viral pathogenicity. Regardless, birds in both groups showed liver hemorrhages. Hepatic necrosis was also seen at the single-cell level, which suggests an acute nature of the lesions and may be related to a low challenge dose. While we attempted Congo red staining for the detection of amyloid, results were negative. Amyloid development is usually a subacute or chronic lesion and backs up the theory that the lesions observed were acute.

These results, in addition to histopathology observations and biochemical changes in serum, confirm the pathobiology of this virus in which primary replication occurs in the liver (3). The low viral RNA detected in blood and swabs as well as the absence of antibodies after challenge with aHEV requires further investigation. A breed susceptibility might be associated with viral replication, pathology, and prevalence in egg-laying flocks. It is not uncommon to have diverse susceptibilities to infection in different commercial chicken lines. These susceptibilities might be associated with the predominant MHC B haplotypes and are worth investigating (29). In addition, there is currently no complete understanding of aHEV replication or the virus's ability to evade or suppress immune responses. More studies are required to understand the pathobiology of the disease. This case report contributes valuable information on seroprevalence, diagnostic techniques, and detection of new genotypes to the United States and insights into viral isolation and pathology in chickens. Further research is required to enhance our understanding of aHEV in commercial layers.

We thank the Poultry Medicine Laboratory at UC Davis for outstanding technical assistance, Hendrix Genetics, Merck Animal Health, the Utrecht University Poultry Medicine Team, and the Royal GD Poultry Department.

aHEV =

avian hepatitis E virus;

AST =

aspartate aminotransferase;

BLS =

big liver and spleen disease;

HHEP =

hen-housed egg production;

RT-PCR =

reverse transcriptase polymerase chain reaction;

SPF =

specific-pathogen-free;

TBP =

tryptose broth peptone