Pathogenicity and genomic characterization of Brazilian fowl adenovirus serotypes 1 and 11

TL;DR

This study examines the pathogenicity and genetic makeup of two Brazilian fowl adenovirus isolates, revealing their tissue effects and genetic relationships with global strains.

Contribution

The study provides the first evaluation of pathogenicity and genomic features of Brazilian FAdV isolates.

Findings

FAdV-11 showed higher virulence and affected multiple tissues, including liver and pancreas.

Genomic analysis revealed BRMSA3761 as a putative recombinant related to European and Middle Eastern strains.

Both isolates were detected in all tissues and showed prolonged cloacal shedding in infected chickens.

Abstract

Fowl adenoviruses (FAdVs) are important pathogens affecting poultry worldwide. Different serotypes cause distinct disease syndromes in domestic chickens, including gizzard erosions (GE), inclusion body hepatitis (IBH), and hepatitis-hydropericardium syndrome (HHS). Despite the economic impact, the evaluation of pathogenicity and genomic features of Brazilian isolates has not yet been assessed. This study aimed to investigate the pathogenic characteristics and whole-genome features of two FAdV Brazilian isolates, BRMSA3762 (FAdV-1) and BRMSA3761 (FAdV-11). For this, pathogenicity was evaluated through inoculation of day-old specific-pathogen-free chickens and embryos. Infected birds were macroscopically and histologically examined in trachea, heart, proventriculus, gizzard, liver, pancreas, duodenum, kidney, and bursa of Fabricius. Also, we analyzed viral distribution in tissues and cloacal virus shedding through real-time PCR. Furthermore, whole-genome sequencing enabled phylogenetic reconstruction and recombination analyses. Results revealed that FAdV-1 was pathogenic, causing severe erosions in the gizzard and a slight alteration in the heart tissue, while FAdV-11 exhibited higher virulence and affected several tissues, including pancreas, liver, gizzard, bursa, and heart. Both viruses were detected in all tissues 8 days post-inoculation, and viral excretion through cloaca occurred throughout the entire experimental period. Differences were established in viral titers, with FAdV-11 demonstrating a higher viral load in tissues and cloacal shedding. In addition, both isolates demonstrated to be pathogenic in SPF chicken embryos. Genomic analysis revealed that BRMSA3762 clustered with sequences from other countries in the Americas, showing no evidence of recombination. Surprisingly, BRMSA3761 clustered with sequences isolated in Europe and the Middle East and was identified as a putative recombinant showing a genetic arrangement closely related to Iranian and Lebanese sequences. Collectively, our findings provide important insights into the pathogenicity of Brazilian fowl adenovirus isolates and contribute to the current understanding of their genetic diversity.

Figures

Pathogenicity analysis in SPF chickens. (A) Contrast in clinical appearance between a non-infected and FAdV-infected bird. (B) Survival rates of SPF chickens after 8 dpi. (C) Significant macroscopic alterations observed in the liver of a FAdV-11 infected bird (left) compared to the liver of the negative control bird (right). (D) Severe erosive lesion (circled) in the gizzard of a FAdV-1 infected bird (left) in contrast to the gizzard of the control bird (right).Fig 1 dummy alt text

Histopathological findings in tissues from chickens infected with FAdV-1 or FAdV-11. Gizzard: open arrows indicate mononuclear inflammatory cell infiltration in the mucosa, while solid arrows indicate degeneration of the koilin layer, characteristic of gizzard erosion. Liver: solid arrows indicate viral intranuclear basophilic inclusion bodies in hepatocytes, and open arrows indicate areas of hepatocyte degeneration. Pancreas: diffuse degeneration and necrosis of acinar cells in the FAdV-11 group. Bursa of Fabricius: lymphoid rarefaction associated with increased interfollicular space in the F

Absolute quantification of FAdV in SPF chickens. (A) Viral DNA copy numbers among various tissues from FAdV-1 and FAdV-11 infected chickens at 8 dpi. (B) Daily viral shedding through cloacal swabs of different chicken groups.Fig 3 dummy alt text

Phylogenetic analysis of FAdVs whole-genome sequences. A) Maximum likelihood tree of genotype 1, showing the position of the Brazilian sequence (green circle). Clades with SH-aLRT ≥ 90 are indicated by gray circles at the internal nodes. The trees were rooted using the midpointing rooting method. B) Phylogenetic tree of genotype 11. The most basal sequence of this tree is not shown due to its long divergence from the most recent common ancestor. The full tree is available in Supplementary Figure 1. C) Recombination pattern of sequence BRMSA3761, isolated in Brazil. D) Detailed view of the clad

Full text

Introduction

Fowl adenovirus (FAdVs) belongs to the family Adenoviridae, genus Aviadenovirus, whose members were previously classified into 5 species (A to E) and 12 serotypes (Benko et al., 2022). Recently, the International Committee on Taxonomy of Viruses (ICTV) renamed species introducing binomial latinized names: Aviadenovirus ventriculi (FAdV-A), A. quintum (FAdV-B), A. hydropericardii (FAdV-C), A. gallinae (FAdV-D) and A. hepatitidis (FAdV-E) (ICTV, 2024). The genome of such virus consists of linear double-stranded DNA of approximately 45 kb. FAdVs are morphologically non-enveloped icosahedral viruses with nucleocapsids containing hexon, penton, and fiber (Harrach et al., 2019).

FAdVs are widely distributed across the globe, circulating between both wild and domestic bird populations (Fitzgerald, 2020). Transmission can occur horizontally and vertically. Infected birds can excrete the virus for several weeks, consequently, increasing their capacity to spread within commercial poultry systems (Grgic et al., 2006; Grafl et al., 2012).

Infections may be subclinical or cause several clinical diseases. Their pathological presentations depend on several factors, including immune competence, age, co-infecting agents, animal genetics, flock management, and strain virulence (Hess, 2020; Schachner et al., 2018). FAdVs can promote a range of clinical manifestations, primarily leading to three distinct disease syndromes in chickens during infection: gizzard erosions (GE), inclusion body hepatitis (IBH), and hepatitis-hydropericardium syndrome (HHS) (Fitzgerald, 2020; Grafl et al., 2012). These complexes are mainly associated with different strains. FAdV-11 (A. gallinae) serotype and serotypes 8a and 8b (A. hepatitidis) are commonly associated with IBH disease outbreaks (Sadekuzzaman et al., 2024; Wang et al., 2020). Since 1993, FAdV-1 (A. ventriculi) has been identified as the primary cause of gizzard erosion (GE) (Tanimura et al., 1993; Ono et al., 2001; Lindgren et al., 2022). Additionally, HHS is linked to A. hydropericardii (serotype 4) (Sun et al., 2019; Niu et al., 2022).

IBH is characterized by massive liver necrosis, hepatomegaly, and basophilic intranuclear inclusions (Wang et al., 2020; Niu et al., 2018; Oliver-Ferrando et al., 2017). Therefore, IBH poses as a systemic affection, committing several other tissues such as kidney and pancreas (Schachner et al., 2021). GE is marked by affecting the koilin layer and underlying mucosa, promoting defects/erosions associated with ulcers and inflammation, respectively (Lindgren et al., 2022; Grafl et al., 2018; Okuda et al., 2006). These diseases can cause economic impacts on broilers and layer flocks at different ages, due to a rise in mortality, reduction of performance and condemnations at the slaughterhouse (Matos et al., 2016b).

Prophylaxis of FAdVs infection poses a challenge for the poultry industry. Vaccination can be made by a wide range of vaccine types including live or inactivated, virus-like particles, subunit vaccines and autogenous formulations (De Luca and Hess, 2025; Schachner et al., 2018). However, there is no cross-protection among all different serotypes and the availability of formulations against most FAdVs is limited in many countries (De Luca and Hess, 2025).

Throughout the years, an increase in outbreaks in poultry flocks was noticed, thus establishing FAdVs as an important cause of economic losses for the global poultry industry (Schachner et al., 2021; Kiss et al., 2021). In Brazil, as in many other countries, this problem has been a growing challenge in recent years (Salvador et al., 2025; Batista et al., 2024; Marín et al., 2022; Roppa et al., 2022; De la Torre et al., 2018; Mettifogo et al., 2014; Pereira et al., 2014). Furthermore, detailed genomic information of FAdV’s is limited by the small number of available complete genome sequences from field isolates (Jakab et al., 2023). Such limitations restrict the understanding of their molecular diversity, evolutionary relationships, and potential pathogenic differences among isolates.

Although fowl adenovirus has gained importance due to the growing impact on the Brazilian poultry industry, comprehensive studies addressing the pathogenicity and the genomic characterization are still lacking. Thus, the present study aimed to evaluate the pathogenic properties of two Brazilian isolates of fowl adenovirus, FAdV-1 and FAdV-11, in specific-pathogen-free (SPF) chicken embryos and chickens. In addition, their genomic features and phylogenetic analyses have been addressed.

Materials and methods

Samples origin and collection history

The FAdV sample BRMSA3762 was detected in tracheas of 11-day-old chickens primarily suffering from IBV GI-23 infection in 2022, while the FAdV sample BRMSA376 was detected in 2024 from liver samples of 12-day-old chickens, during an outbreak with findings suggestive of adenovirus infection (IBH). All samples were collected from broiler chicken farms of Paraná state from southern Brazil and kept at −80°C in the Laboratory of Animal Health and Genetics at Embrapa Swine and Poultry/Concordia/Brazil.

Polymerase chain reaction (PCR) and Partial sequencing

The partial sequence of the 52 K and pIIIa genes were amplified by PCR using the primer set, 52K-F and 52K-R, following the method described by Günes et al. (2012). Partial sequencing aimed to identify sample serotypes and was carried out by the amplification of a Hexon gene of FAdV previously described by Meulemans et al. (2001). PCR product purification was achieved using GFX PCR DNA and Gel Band Purification Kit (Cytiva, Marlborough, MA, USA), following manufacturer's instructions. Partial sequencing was performed at the ACTGene Laboratory (Alvorada, RS, Brazil) using the automatic sequencer ABI-PRISM 3100 Genetic Analyzer armed with 50 cm capillaries and POP6 polymer (Applied Biosystems, Waltham, MA, USA). The electropherograms obtained were edited through MEGA 11 (Tamura et al., 2021). Consensus was analyzed using the Basic Local Alignment Search Tool (BLAST) to determine their similarity with other sequences downloaded in the NCBI GenBank database.

Virus Isolation and Titration in specific pathogen-free chicken embryonated eggs (SPF-CEE)

Suspect tissues were homogenized in phosphate-buffered saline (PBS) (10–20% w/v), centrifuged, and processed in 0.22-μm filters (EMD Millipore Corp., Billerica, MA, USA). After, 0,2 mL of tissue suspensions were inoculated in 9-day-old SPF CEE via the allantoic cavity. SPF CEE were incubated for 7 days at 37°C with daily monitoring by candling and dead embryos were stored under refrigeration until evaluation. On the seventh day, all embryos (died and survived) were macroscopically analyzed, searching for alterations or lesions. Viral titers as 50% embryo infectious doses (EID_50_) were determined following the Reed and Muench method (Reed and Muench, 1938). In order to discard co-infections, the samples were tested by PCR for ARV, IBV, IBDV, CIAV, and Mycoplasma sp. Viruses with hemagglutination activity were excluded by hemagglutination assay (HA). Also, BRMSA3762 sample was previously treated with chloroform following the methodology described by Feldman and Wang (1961), to inactivate enveloped viruses, such as IBV.

Pathogenicity assessment in SPF chicken embryos and Viral re-isolation

Ten 9-day-old SPF CEE were inoculated with 10^3.75^ EID_50_ of virus via the allantoic cavity, for each viral isolate. As a control, another group was inoculated with sterile PBS (mock). All CEE were monitored daily and after 7 days the embryos were macroscopically analyzed, screening for alterations and lesions. Liver tissues were collected with 10% neutral phosphate-buffered formalin, for histopathological examination. In order to confirm the virus infectivity post-infection, re-isolation of FAdV from infected groups was conducted using tissue samples from gizzard and liver. Samples were pooled from three birds, resulting in three pools per group (n = 9), using the same inoculation methodology as described for viral isolation.

Pathogenicity analysis in day-old SPF chickens

27 one-day-old SPF chickens are randomly allocated in 3 groups (n = 9 per group). Groups 1 and 2, were inoculated with 200 μl containing 10 ^3.75^ EID_50_ by the ocular-nasal route of isolates BRMSA3762 and BRMSA3761, respectively. Group 3 was inoculated with the same amount of phosphate buffered saline (PBS), as negative control. All groups were maintained in individual negative-pressure isolators with controlled environments and free intake of water and food. Animals were daily monitored for 8 days post-inoculation (dpi), observing clinical signs and mortality. To assess viral shedding, cloacal samples were collected daily (1 to 8 dpi) from five individually identified birds in each group. After collection, all swabs were immersed in 1 mL of sterile PBS for further analysis. At 8 dpi, all the surviving birds were humanely euthanized by rapid cervical disarticulation and necropsied searching for macroscopic alterations. Trachea, heart, proventriculus, gizzard, liver, pancreas, duodenum, kidney and bursa of Fabricius were collected for histopathological analysis and quantification of the viral load.

Histopathology

Tissues were collected with 10% buffered formalin, embedded in paraffin, sectioned with a microtome, colored using hematoxylin-eosin method (H&E), and evaluated with optical microscopy. Histological evaluation was conducted by classifying lesions in scores of severity (0-4): 0, was assigned to samples with no significant lesion observed, 1 for mild lesions, 2 for moderate lesions, 3 for severe, and 4 when marked injuries were observed.

Cloacal viral load and Tissue distribution by qPCR

To detect virus shedding, cloacal swabs were pooled per day (5 per group) and 200 μL of each pool was separated for extraction. Viral load assessment in organs was standardized using 25 mg of each tissue sample (trachea, heart, proventriculus, gizzard, liver, pancreas, duodenum, kidney, and bursa of Fabricius). Extractions of samples total DNA in both assays were conducted using the Purelink Genomic DNA Mini Kit (Invitrogen, Carlsbad, CA, USA) following manufacturer's protocol. FAdV DNA detection was performed using SYBR green dye with primers that amplify the highly conserved 52 K region (52k-fw: 5ʹ-ATG GCK CAG ATG GCY AAG G-3ʹ and 52k-rv: 5ʹAGC GCC TGG GTC AAA CCG A-3ʹ), previously described by Günes et al. (2012) with slight modifications. The total reaction volume contained 20 μl, compound of 10 μl of 2x GoTaq® qPCR Master Mix (Promega™, Madison, WI, USA), 1 μl (10 μM) of each primer, 2 μl of DNA and 6 μl of nuclease-free water. The real-time PCR was performed using the ABI 7500 fast instrument (Applied Biosystems Inc., Foster City, CA, USA), under the following conditions: 2 min at 95°C, followed by 40 cycles of 15 s at 95°C, 30 s at 60°C, and 30 s at 72°C, and a melting step between 60 and 95°C. A standard template was constructed using amplicons of the region of the 52 K and pIIIa genes (Günes et al., 2012), purified with GFX PCR DNA and Gel Band Purification Kit (Cytiva, Marlborough, MA, USA), following manufacturer's instructions. Purity of template was verified utilizing a NanoDrop spectrophotometer (Thermo Fisher Scientific, Carlsbad, CA, USA), and quantification was performed in Qubit™ 2.0 Fluorometer (Invitrogen™, Thermo Fisher Scientific, Waltham, MA, USA). The known quantity of DNA copies was calculated through the web tool DNA copy Number and Dilution Calculator by Thermo Fisher Scientific. To generate a standard curve tenfold dilutions were assembled, and a predetermined cycle threshold (Ct) value was established with samples considered positive if amplification was detected within 32 cycles as described by Günes et al. (2012). Additionally, the specificity of the amplification product was verified through melting curve analysis and agarose gel electrophoresis.

Whole-genome sequencing and assembly

The nucleic acids were extracted from the samples with the commercial MagMAX™ CORE nucleic acid purification kit (Applied Biosystems, Waltham, Massachusetts, USA), using the automated KingFisher™ Duo Prime equipment (Thermo Fisher Scientific Inc., Waltham, Massachusetts, USA). The shotgun metagenomic library was performed using the Illumina DNA Prep kit, according to the manufacturer’s recommendations, and the sequencing was performed on the Illumina Nextseq 1000 platform using NextSeq 1000/2000 P1 Reagents kit (Illumina Inc., San Diego, California, USA).Genome assemblies were performed using Geneious Prime version 2025.0 (Biomatters Ltd.). Sequencing reads from sample BRMSA3761 were aligned to the reference sequence MK757569, whereas reads from sample BRMSA3762 were aligned to the reference sequence AAU46933 (Supplementary data 1).

Phylogenetic reconstruction and Recombination analysis

Hexon gene sequences from the two samples in this study were aligned with reference sequences from all FAdVs genotypes (Supplementary Data 1). Following genotype determination, whole-genome alignments were generated in MAFFT v7.490 (Katoh and Standley, 2013), using all available sequences of that same genotype from public databases. Maximum likelihood phylogenetic reconstruction was performed in IQ-TREE v. 2.0.3 (Nguyen et al., 2015), employing the most suitable evolutionary model selected by ModelFinder based on BIC criterion. Statistical support for the clades was estimated using the SH-like approximate Likelihood-Ratio Test (SH-aLRT) with 1,000 replicates. Finally, trees were visualized with the ggtree package in R. Recombination detection was applied on whole-genome alignments to understand recombination patterns. Initially, a full exploratory recombination scan was performed using seven different methods in RDP v.5.34 (Martin et al., 2015). Only events detected with statistical significance by three or more methods were considered for further analysis. After identifying potential minor and major parental sequences, the recombination breakpoints were determined using Simplot v.3.5.1 (Lole et al., 1999), and a recombination plot was generated in R.

Statistical analysis

Data analyses were performed using GraphPad Prism 7.0.0 software. Results were tested by one-way analysis of variance (ANOVA) and Tukey's multiple-comparison post-test. A statistically significant p-value of < 0.05 was considered.

Ethics statement

The in-vivo experimentation was developed in the Laboratory of Animal Health and Genetics at the Embrapa Swine and Poultry, Concordia- SC, Brazil. All assays and procedures were conducted in accordance with the ethical principles and guidelines for animal experimentation recommended by the Brazilian College of Animal Experimentation. (COBEA). The project was approved by the Embrapa Swine and Poultry Ethical Committee for Animal Experimentation (CEUA/CNPSA), protocol number 22/2024 on 9 May 2024.

Results

Virus isolation and titration

FAdV samples were successfully isolated from biological samples in SPF CEE after three or four passages. Infected embryos showed dwarfism, hyperemia, and several hepatic lesions. The presence of FAdV DNA was confirmed by PCR. Virus peak titers for FAdV-1 and FAdV-11 samples were 10^3.75^ EID_50_ and 10^4.8^ EID_50_, respectively. Both isolates were negative for other tested pathogens, thus confirming the effectiveness of chloroform inactivation.

Genotyping of isolates

Phylogenetic analysis based on the hexon gene sequences showed that the sample BRMSA3762 isolated from tracheal swabs belongs to species Aviadenovirus ventriculi, serotype 1 (FAdV-1), demonstrating high identity with chicken embryo lethal orphan (CELO) reference strain and other FAdV-1 strains. Differently, the isolate BRMSA3761 associated with the IBH outbreak, showed high similarity to A. gallinae, serotype 11 (FAdV-11) strains, deposited in NCBI GenBank database.

Pathogenicity assessment in SPF Chicken Embryos

After 7 days, the embryos inoculated with FAdV-1 demonstrated an accumulated mortality rate of 20% (2/10). Macroscopically embryos showed dwarfism (10/10), hemorrhagic (3/10), hepatic alterations (7/10), and opacity/thickening of chorioallantoic membrane - CAM (10/10). Meanwhile, the FAdV-11 assessment exhibited a higher mortality of 50% (5/10), with all deaths occurring between 6 and 7 dpi. Embryos revealed similar alterations with dwarfism (9/10), hemorrhagic (6/10), and several hepatic lesions (10/10), however without any abnormality in CAM. No alterations were observed in the control group (PBS inoculated). Histopathology of livers revealed differences between viruses. FAdV-1 had multifocal areas of dissociation, degeneration, and mild necrosis of hepatocytes. Otherwise, FAdV-11-infected embryos demonstrated a severe and diffuse degeneration of hepatocytes with numerous basophilic intranuclear inclusion bodies.

Pathogenicity assessment in day-old SPF Chickens

Clinical signs and severe gross pathology*.* Pathogenicity assessment was conducted utilizing day-old SPF chicks inoculated through the oral/nasal route and monitored for 8 days. Similar clinical signs were observed in both infected groups, however, they were more pronounced in FAdV-11 inoculated birds. Diarrhea was noted at 3 dpi, followed by additional signs starting at 4 dpi, including hunched posture, instability of standing, ruffled feathers, and depression (Fig. 1A). Groups only demonstrated differences in mortality rates. The group infected with FAdV-11 had a higher mortality rate, with three birds (33.3%) dying, one at 5 dpi and two at 7 dpi. In comparison, the FAdV-1 group had two deaths (22.2%), also occurring at 5 and 7 dpi. No mortality or clinical signs were observed in the control group (Fig. 1B).Fig. 1. Pathogenicity analysis in SPF chickens. (A) Contrast in clinical appearance between a non-infected and FAdV-infected bird. (B) Survival rates of SPF chickens after 8 dpi. (C) Significant macroscopic alterations observed in the liver of a FAdV-11 infected bird (left) compared to the liver of the negative control bird (right). (D) Severe erosive lesion (circled) in the gizzard of a FAdV-1 infected bird (left) in contrast to the gizzard of the control bird (right).Fig 1 dummy alt text

At necropsy (8 dpi), chickens of the FAdV-1 group demonstrated marked gross lesions in gizzard with regions of severe erosion (Fig. 1D). Also, liver alterations were observed in two animals, characterized by slight yellowing. In contrast, FAdV-11 group birds showed marked and extended alterations in liver, including hepatomegaly and yellow-green appearance (Fig. 1C). Additionally, half of the animals of the group presented slight erosive lesions in the gizzard. No significant gross lesions were observed in other analyzed organs or in the control chicks.

Histopathological lesions. Microscopic alterations were classified into five scores of severity (0-4): normal (0), mild lesions (1), moderate lesions (2), severe lesions (3), and marked injuries (4) (Fig. 2). Gizzard from FAdV-1 infected chickens revealed mild to marked degeneration and necrosis of the koilin layer, associated with mucosal infiltration of inflammatory cells. In addition, a large number of basophilic intranuclear inclusions were visualized in glandular epithelium. Interestingly, three birds from the group presented mild lymphocytic and heterophilic infiltration in heart epicardium. Conversely, FAdV-11 demonstrated a more systemic presentation. As expected, livers presented an extensive degeneration of hepatocytes, associated with periportal inflammatory infiltration. Furthermore, more than half of the group presented mild to moderate lymphoid rarefaction in the bursa follicles, associated with increased interfollicular spaces due to fibroblast proliferation and heterophilic infiltration. Unexpectedly, pancreas was the most severely affected tissue, demonstrating a diffuse degeneration and necrosis of acinar cells and mild heterophilic infiltration. In addition, some birds exhibited mild degeneration of the koilin layer, accompanied by slight heterophilic infiltration in the gizzard. Interestingly, four chickens presented slight infiltration of mononuclear cells in myocardium. Typical basophilic intranuclear inclusions were found in liver, pancreas, and gizzard. According to histopathological analysis, the highly affected tissues in order were pancreas and liver, followed by bursa of Fabricius, gizzard, and heart, with average lesion scores of 3.0, 2.55, 0.66, 0,55, and 0.44, respectively, in FAdV-11 infected group. By contrast, in the FAdV-1 infected group, only gizzard and heart demonstrated significant alterations, with scores of 2.22 and 0.33, respectively. Other less relevant microscopic lesions found present in very few birds were also described in Table 1.Fig. 2. Histopathological findings in tissues from chickens infected with FAdV-1 or FAdV-11. Gizzard: open arrows indicate mononuclear inflammatory cell infiltration in the mucosa, while solid arrows indicate degeneration of the koilin layer, characteristic of gizzard erosion. Liver: solid arrows indicate viral intranuclear basophilic inclusion bodies in hepatocytes, and open arrows indicate areas of hepatocyte degeneration. Pancreas: diffuse degeneration and necrosis of acinar cells in the FAdV-11 group. Bursa of Fabricius: lymphoid rarefaction associated with increased interfollicular space in the FAdV-11 group. Heart: open arrows indicate inflammatory cell infiltration, characterized by heterophils in the epicardium (FAdV-1) and mononuclear cells in the myocardium (FAdV-11).Fig 2 dummy alt textTable 1Histopathologic scores from lesioned tissues.Table 1 dummy alt textTissuesGroups and lesion scoresFAdV-1FAdV-1101234Mean ± SD01234Mean ± SDHeart630000.33 ± 0.50540000.44 ± 0.53Gizzard211322.22 ± 1.56531000.55** ± 0.73Liver81°0000.11 ± 0.3302°3132.55 ± 1.24Pancreas710000.13 ± 0.3501214*3.00** ± 1.20Bursa801000.22** ± 0.67522000.66** ± 0.87Died bird at 5 dpi. *Died bird at 7 dpi. (Pancreas from 5 dpi birds was not evaluated).

The analysis of viral load, shedding, and tissue distribution. To assess virus distribution and shedding in chickens, the viral copy numbers were measured by real-time PCR. Tissues analyzed were trachea, heart, proventriculus, gizzard, liver, pancreas, duodenum, kidney, and bursa of Fabricius. No DNA viral copies were detected in the negative control group (data not shown). At 8 days post-infection, FAdV DNA was detected in all tissues tested in both groups. FAdV-1 demonstrated a significantly higher viral replication in gizzard, when compared with other tissues (p < 0.05). Otherwise, the FAdV-11 group exhibited a notably higher viral load in most analyzed tissues, with the highest levels detected in the gizzard, liver, duodenum, and pancreas (Fig. 3A). Viral shedding was carried out based on daily collected cloacal swabs. FAdV-11 exhibited an increased viral DNA copy number excreted through cloaca, reaching its peak at 7 dpi. Conversely, FAdV-1 group viral shedding increased exponentially till 4 dpi, then maintained a similar viral shedding rate up to the 8th day post-inoculation (Fig. 3B).Fig. 3. Absolute quantification of FAdV in SPF chickens. (A) Viral DNA copy numbers among various tissues from FAdV-1 and FAdV-11 infected chickens at 8 dpi. (B) Daily viral shedding through cloacal swabs of different chicken groups.Fig 3 dummy alt text

Viral re-isolation. Virus isolation post-challenge was conducted utilizing tissue samples from macroscopically mostly affected organs at necropsy (gizzard and liver). Therefore, pools of three birds (3 pools/group) were separately inoculated in SPF CEE. As a result, FAdV was recovered from all tissue pools, causing typical embryo alterations at first passage in both samples, FAdV-1 and FAdV-11. These results confirmed viral replication in gizzard and liver at 8 dpi.

Analysis of complete sequences of BRMSA3761 and BRMSA3762 isolates

The complete nucleotide sequences were submitted to GenBank under the accession numbers: PX448739 (BRMSA3761) and PX481163 (BRMSA3762). Phylogenies were reconstructed for whole-genome sequences of genotypes 1 and 11, including all publicly available sequences for these two genotypes (Fig. 4). The tree for genotype 1 shows that the Brazilian sequence (BRMSA3762) clusters within a well-supported clade composed primarily of sequences from other countries in the Americas (Fig. 4A). This clade is supported by a long branch that diverges from a common ancestor shared with a primarily European clade. We found no evidence of recombination in BRMSA3762. In the other hand, the genotype 11 tree shows the Brazilian sequence (BRMSA3761) clustering within a cluster composed of sequences isolated in Europe and the Middle East (Fig. 4B and Supplementary Figure 1). Curiously, the Brazilian sequence was identified as a putative recombinant, showing a genetic arrangement evolutionarily related to a sequence from Iran and a final minor segment derived from a Lebanese sequence (Fig. 4C and 4D). The grouping of the Brazilian sequence within this clade is most likely explained by the evolutionary relatedness of a major segment of its genome to a sequence from Iran.Fig. 4. Phylogenetic analysis of FAdVs whole-genome sequences. A) Maximum likelihood tree of genotype 1, showing the position of the Brazilian sequence (green circle). Clades with SH-aLRT ≥ 90 are indicated by gray circles at the internal nodes. The trees were rooted using the midpointing rooting method. B) Phylogenetic tree of genotype 11. The most basal sequence of this tree is not shown due to its long divergence from the most recent common ancestor. The full tree is available in Supplementary Figure 1. C) Recombination pattern of sequence BRMSA3761, isolated in Brazil. D) Detailed view of the clade highlighted in (B). The geographical origin of FAdV sequences is indicated by the legend to the right of the figure.Fig 4 dummy alt text

Discussion

Fowl adenovirus (FAdVs) are important viruses in the poultry industry and have been recognized as primary pathogens worldwide. Currently, they pose significant challenges due to impacts on poultry health and productivity (Hess, 2020; Fitzgerald, 2020). FAdV-11 has been identified in several countries associated with IBH outbreaks (Wang et al., 2020; Niu et al., 2018; Oliver-Ferrando et al., 2017; Niczyporuk and Czekaj, 2018; Schachner et al., 2018), including Brazil (Batista et al., 2024; Salvador et al., 2025). Additionally, previous studies have demonstrated a high prevalence in Brazilian commercial flocks (Pereira et al., 2014; De la Torre et al., 2018). FAdV-1 has been described as the etiology of erosions in gizzard (GE), although the number of reports is lower when compared with other FAdVs affections. In Brazil, a recent outbreak was described by Roppa et al. (2022).

It is well established that different FAdVs strains exhibit varying levels of pathogenicity and virulence, potentially influencing the infection outcome and cell tropism (Schachner et al., 2018; Hess, 2020). Considering this, the evaluation of phylogenetic relationships and pathogenic properties of FAdVs strains is crucial to understanding their role in disease outbreaks. Our study focused on the analysis of two strains, BRMSA3762 (FAdV-1) and BRMSA3761 (FAdV-11), isolated from broiler flocks in Brazil. To our knowledge, this is the first report of whole-genome sequencing and pathological characterization of FAdV Brazilian strains.

Phylogenetic analyses revealed that, as expected, BRMSA3762 clustered with sequences from the Americas, sharing the highest homology (99.94 - 99.97%) with the North American strains D2342/2/2/13/US (OP985612) and D3150/9/15/US (OP985625). A recent study developed for Jakab et al. (2023) analyzed 40 FAdV-1 genomes collected over a 12-year period from 15 different countries and based on genomic similarity, the sequences were classified into 11 groups (GT-I to GT-XI). In this context, our sequence would be grouped within group GT-VII, along with nine other isolates, including D2342/2/2/13/US and D3150/9/15/US. Interestingly, all isolates in this group display a similar pathological background to our strain, marked by the absence of clinical signs or gizzard erosion, and in some cases, by co-infection with IBV.

Surprisingly, the isolate BRMSA3761 clustered with sequences from Europe and the Middle East, exhibiting the highest sequence similarity with Iranian (99.24%) and Lebanese (98.82%) strains: Iran/UT-Kiaee/2018 (MK757569) and D2291/2/3/13/LB (PP471933). We also detected a clear recombination signal in our isolate, resulting from a genetic exchange between these two strains. Transmission pathways of the FAdV-11 strain to Brazil are still not understood. However, a previous study by our group (Trevisol et al., 2023), demonstrated that the novel exotic variant of IBV GI-23 responsible for outbreaks in Brazilian poultry farms, has clustered (based on partial sequencing) with strains from Europe (Poland) and Middle East (Israel), suggesting a possible common source for intercontinental spread.

The pathogenicity assessment was conducted in a day-old SPF chick model to represent early-life exposure and potential vertical transmission scenarios. Important differences in pathological presentation, virulence, viral shedding, and tissue tropism were observed between the strains. During the necropsy, macroscopic alterations were observed only in gizzards and livers of infected birds. FAdV-1 group demonstrated prominent gizzard erosion associated with typical basophilic intranuclear inclusions in glandular epithelium. Our results are in accordance with classical GE description (Grafl et al., 2018; Ono et al., 2007; Okuda et al., 2006; Schachner et al., 2018; Lindgren et al., 2022). It is important to highlight that lesions in gizzard can lead to tissue dysfunction, reducing grinding efficiency and motility, as well as promoting a decrease in food ingestion (Lindgren et al., 2022). We hypothesize that these digestion and consumption changes could also influence the severity of clinical signs and lethality observed in the FAdV-1 group.

Surprisingly, superficial focuses of erosion in gizzard were also observed in four birds of the FAdV-11 infected group. Lesions were histologically similar to the FAdV-1 group, including basophilic intranuclear inclusions. However, they exhibited a significantly less severity degree. Interestingly, FAdV–11 in experimental studies has never been directly associated with gizzard erosion. Nevertheless, Steer et al. (2015) demonstrated the presence of FAdV-11 and FAdV-8b in gizzard, even in the absence of gross and histological lesions, indicating that FAdVs have a predilection for epithelial lining of the gizzard and may play a hole contributing to their infectivity and shedding after resolution of typical hepatic lesions. Our microscopical findings associated with a high viral load in gizzard and successful viral recovery, suggest that FAdV-11 could replicate and generate lesions in this tissue. This also indicates that gizzard evaluation should be included in future pathogenicity assays investigating other FAdV serotypes, beyond FAdV-1.

Also, it should be noted that gizzard damage in both infected groups was only found in survival birds at 8 dpi, and no histopathological lesions were observed in early-dead birds. This pathologic pattern suggests that the onset of gizzard erosions occurs after 7th day post-inoculation and probably will become more severe in later stages of infection. Our findings are consistent with those previously demonstrated by Grafl et al. (2013), in an experimental study evaluating AGE pathological progression in different days, up to 17th dpi.

In our study, FAdV-11 infected birds developed apparent hepatitis, as related to the flock outbreak source. Furthermore, microscopic findings agree with typical IBH descriptions (Wang et al., 2020; Zhang et al., 2023; Joshi et al., 2022). In addition, histological evaluation identified a remarkable pancreatic affection, with degeneration and necrosis of acinar cells, inflammation, and a large number of basophilic intranuclear inclusions. Interestingly, pancreas presented the greatest mean degree of lesions during the histological evaluation, demonstrating an unexpected high tropism to this tissue.

Recently, IBH has been recognized as a disease involving a severe metabolic impairment, besides its pathogenesis associated with tissue damage and functional disruption of liver and pancreas, changing its focus as mainly hepatic pathology (Schachner et al., 2021). Matos et al. (2016b) revealed that A. gallinae tropism in both liver and pancreas could create severe metabolic impairment. In addition, severe liver damage causes a dysfunction in protein synthesis, which can impact erythrocytes formation, consequently reducing hematopoiesis capacity (Joshi et al., 2022). This association between metabolic and hematopoietic alterations may contribute to explain the development of clinical signs and mortality.

We also identified bursal damage, which indicates the involvement of primary lymphoid organs in FAdV-11 pathogenesis. In this sense, several authors show the immunological impairment of FAdVs infections, thus viral replication in bursa of Fabricius and thymus (Zhang et al., 2023; Steer et al., 2015). By contrast, in the FAdV-1 infected group, only a single bird displayed mild hepatic microscopic changes and a moderated bursal rarefaction, while another showed slight pancreatic alteration during the period, thus we disregarded the significance between infection and these alterations.

Some birds of both groups presented slight lymphocytic infiltration in their hearts. These alterations were observed in a reduced percentage of infected birds, approximately 33% (FAdV-1) and 44% (FAdV-11). Despite this, we decided to consider these alterations relevant in our experimental assessment based on the fact that FAdV-1 and FAdV-11 groups were affected in different regions of their hearts, such as epicardium and myocardium, respectively, and no similar lesions were found in the control group. Heart alterations are consistently described during HHS outbreaks, normally associated with FAdV-4 (Sun et al., 2019; Niu et al., 2022). However, the hepatic pathogenesis in IBH cases could be also responsible for inflammatory and degenerative processes in heart tissue (Schachner et al., 2018). Descriptions relating FAdV-11 as responsible for heart damage are less common (Joshi et al., 2022; Zhao et al., 2015). Furthermore, De la Torre et al. (2018) identified FAdV-11 as an etiological agent related to the IBH—HSS outbreak in Brazil. However, according to our knowledge, this is the first report of FAdV-1 associated with heart histopathological alteration.

The viral load in organs was consistent with the degree of damage found by histopathology. Viral DNA was detected in all tested tissues corroborating previous studies showing several tissue distribution of FAdVs during infection (Steer et al., 2015; Niczyporuk and Czekaj, 2018; Li et al., 2022; Wang et al., 2023; Zhang et al., 2023). At 8 days post-infection, the dynamics of viral distribution were completely different between infected groups. FAdV-1 demonstrated high tropism for gizzard tissue. Thus, it had a low viral load in most tissues, when compared with FAdV-11. As expected, a marked difference between strains was observed in the liver. However, both groups presented similar high viral replication rates in gizzards. Other authors observed similar results in experimental studies when comparing FAdV-1 and FAdV-11 strains (Steer-Cope et al., 2017; Steer et al., 2015). We also noted an evident contrast between groups, regarding viral replication in pancreas and duodenum tissues.

Infected birds shed virus by cloaca throughout the entire experimental period (1 to 8 dpi). FAdV-11 inoculated chickens shed a higher number of viral particles compared to the FAdV-1 strain from 5 days post-infection onward. Excretion patterns were consistent with those previously described (Li et al., 2022; Oliver-Ferrando et al., 2017; Matos et al., 2016b).

FAdV-11 infected birds showed more pronounced clinical signs during the assessment period. Mortality rates post-infection, range from 33,3% to 22,2% in FAdV-11 and FAdV-1 infected groups, respectively, with deaths occurring between 5 and 7 dpi. Differently, studies conducted by Sadekuzzaman et al. (2024) and Zhao et al., (2015) observed that FAdV-11 experimentally infection via oral/nasal, utilizing similar doses in day-old chickens 10^4.5^ EID_50_ and 3-wk-old SPF chickens 10^3.5^ EID_50_, respectively, induced mortality rates less than 10%. However, in natural IBH outbreaks, mortality variates among 10% and can reach 30% (Schachner et al., 2018). Other experimental studies have also demonstrated that FAdV-1 is a less lethal serotype, describing higher survival rates with reduced or no deaths during infection (Niczyporuk and Czekaj, 2018; Grafl et al., 2018). Interestingly, Mirzazadeh et al. (2021) found a similar mortality (25%) in day-old chicks orally inoculated, however utilizing a higher dose (10^7^ TCID_50_) of FAdV-1 isolate. It is recognized that FAdVs virulence can be influenced by numerous factors, including genetic differences among SPF chicken lineages and virus pathogenicity, as earlier demonstrated by Cook (1974). In this sense, we hypothesized that the increased mortality of our study was mainly correlated with the virulence of the isolates and chickens' level of susceptibility.

Outcomes found in SPF CEE pathogenicity assessment were consistent with results found in day-old chickens. FAdV-11 was more virulent, inducing a lethality of 50% in embryos. Conversely, FAdV-1 only caused two deaths, representing 20% mortality in the assay. Gross lesions in both groups were similar, including embryos with dwarfism, hemorrhagic, and hepatic alterations. However, the FAdV-1 strain also induced opacity and thickening of CAM. As expected, histopathological analysis of livers revealed differences in severity of lesions. The FAdV-1 group presented multifocal areas of dissociation, degeneration, and mild necrosis of hepatocytes. In contrast, the FAdV-11 group exhibited a more pronounced damage, with severe and diffuse degeneration of hepatocytes, in association with numerous basophilic intranuclear inclusion bodies. Our findings corroborate previous descriptions of FAdVs isolation, with similar gross and microscopic lesions observed during isolation of different serotypes such as FAdV-2 (Joubert et al., 2014), FAdV-4 (Yu et al., 2018), FAdV-8a (Nuñez et al., 2024), FAdV-8b (Joubert et al., 2014; Sadekuzzaman et al., 2024), FAdV-9 (Alemnesh et al., 2012) and FAdV-11 (Abghour et al., 2019; Qiao et al., 2024; Sadekuzzaman et al., 2024).

Different than other studies, our findings showed a lower lethality in embryo chickens. We hypothesized that contrasts in mortality rates found in embryos may be caused by differences in the quantities of virus inoculated. Most studies had inoculated untitrated suspensions (Abghour et al., 2019; Nuñez et al., 2024; Sadekuzzaman et al., 2024) or used higher viral titers (Qiao et al., 2024; Yu et al., 2018; Alemnesh et al., 2012; Joubert et al., 2014).

In summary, our results showed that Brazilian strains BRMSA3761 and BRMSA3762 were pathogenic to SPF day-old chickens and embryos. Both FAdV strains demonstrated several tissue distribution and cloacal shedding during the entire experiment period. Moreover, FAdV serotype 11 exhibited greater virulence as evidenced by clinical signs, higher mortality, and broader tissue distribution and lesions. The complete genomic sequences of Brazilian strains represent a valuable addition to the existing FAdV-11 and FAdV-1 genome database. Additionally, whole-genome analysis provided relevant insights into the distribution and recombination patterns suggesting a potential intercontinental transmission. Our findings advance the understanding of the pathogenicity patterns and genomic characteristics of Brazilian FAdV strains, providing essential information for infection, diagnosis and control, supporting the development of disease mitigation strategies in poultry production systems.

CRediT authorship contribution statement

Gabriel S. Zani: Writing – review & editing, Writing – original draft, Visualization, Validation, Software, Project administration, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Paulo A. Esteves: Visualization, Validation, Supervision, Investigation, Funding acquisition, Conceptualization. Luizinho Caron: Visualization, Validation, Supervision, Investigation, Funding acquisition, Conceptualization. Iara M. Trevisol: Writing – review & editing, Visualization, Validation, Supervision, Resources, Project administration, Funding acquisition, Formal analysis, Conceptualization. Amanda O. Barbosa: Methodology, Investigation. Marcos A.Z. Mores: Methodology, Investigation, Formal analysis, Data curation. Dennis M. Junqueira: Software, Methodology, Investigation, Data curation. Meriane Demoliner: Software, Methodology. Fernando R. Spilki: Writing – review & editing, Software, Resources, Methodology. Geferson Fischer: Writing – review & editing. Marcelo de Lima: Writing – review & editing, Visualization, Supervision, Resources, Project administration, Methodology, Investigation, Funding acquisition, Data curation, Conceptualization.

Disclosures

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.