Introduction

Avian leukosis (AL), which includes both malignant and benign tumorigenic diseases caused by avian leukosis virus (ALV), has caused immense economic losses in the poultry industry worldwide [1, 2]. ALV is divided into 11 subgroups (designated A–K) based on cross-neutralization and variations in the gp85 glycoprotein, which is a group-specific antigen [3]. Avian leukosis virus subgroup J (ALV-J) is the most prevalent subgroup in chickens, and its members exhibit stronger pathogenicity and vertical transmission ability in poultry than those of the other subgroups [4,5,6]. ALV-J infection was a common problem in the poultry industry during 2003–2010, causing myelocytoma in broilers and hemangioma in layers [7,8,9,10,11]. Because of the widespread distribution of ALV-J strains and a lack of organized control measures for local-breed flocks, eradication of ALV in China remains a major challenge. Since 2018, an especially tumorigenic strain of ALV-J has emerged in imported broilers from China, causing severe osteomas of the keel and ribs, leading to massive mortality and morbidity and becoming a major concern for poultry health [3, 11]. Therefore, many domestic farms have started testing for ALV infection [3].

Rubin et al. defined four serological classes of susceptible birds: viremia, no antibody (V + A−); no viremia, with antibody (V − A+); viremia, with antibody (V + A+); and no viremia, no antibody (V − A−) [2, 12]. Another study showed that roosters spread ALV-J to hens by insemination, ultimately resulting in vertical transmission [13]. However, those researchers did not elaborate on the ALV infection status of semen from males or the role of males in the transmission of ALV. ALV virus was isolated using DF-1 cells, and p27 antigens in cloacal swabs or semen were detected directly by enzyme-linked immunosorbent assay (ELISA). However, the accuracy of the ELISA method for detecting the p27 antigen in cloacal swabs or semen has not been determined.

ALV-J can be detected in Hy-line brown roosters, suggesting that roosters are infected by ALV-J and may play a crucial role in viral transmission in poultry. Here, we determined that ALV-J-infected semen can infect hens through artificial insemination. Consequently, ALV-J was transmitted vertically to eggs, leading to an increase in the ALV infection rate on the farm. In addition, we also determined four different ALV-J infection states in roosters and evaluated the reliability of different methods for detecting ALV-J in semen samples. Taken together, our data provide a basis for detecting and eradicating ALV in poultry.

Materials and methods

Ethics statement

All chicken studies were performed according to the Institutional Guidelines of China Agricultural University (approval SKLAB-B-2010-003) and were approved by the Beijing Association for Science and Technology of China (approval SYXK, Beijing, 2007–0023).

Sample origin

Forty-six anticoagulant-containing blood samples were collected aseptically from the wing veins, and 46 semen samples were obtained from 46 30-week-old Hy-line roosters in Hebei Province, China, during November 2020. Ninety-two anticoagulant-containing blood samples from the wing vein and 92 albumen samples from breeding eggs were collected from the same 92 30-week-old indigenous hens on an experimental farm (Supplementary Table S1). The plasma was extracted from the blood samples as reported previously [9]. In addition, 10 albumen samples from breeding eggs of five ALV plasma-positive hens (2 eggs/hen) were collected for virus isolation.

Virus isolation and identification

Semen samples were diluted five times with PBS containing 2% penicillin and streptomycin. After gently mixing upside down, the supernatant was collected by centrifugation to precipitate the cells at 3,000 × g for 2 min. Albumen samples were diluted four times with DMEM in 5-ml syringes. The virus was isolated from the plasma, semen, and albumen samples by inoculating DF-1 cells as described previously, thus confirming ALV infection [9]. The ALV group-specific antigen p27 was detected in the culture supernatant by ELISA (IDEXX, Westbrook, ME, USA; #99–0925).

Indirect immunofluorescence assay (IFA)

Cells were washed with PBS and fixed in a cold acetone–alcohol mixture (3:2) for 5 min. They were then incubated with the mouse anti-ALV-J monoclonal antibody JE9 at 37°C for 60 min, followed by incubation with goat anti-mouse IgG antibody conjugated with fluorescein isothiocyanate (Sigma, St. Louis, MO, USA) at 37°C for another 60 min. Finally, the cells were observed under a fluorescence microscope.

Primers and polymerase chain reaction (PCR) amplification

DNA was extracted from DF-1 cells according to the instructions of the manufacturer of the DNA Extraction Kit (Tiangen, China; #DP304-03). A pair of universal primers targeting env and LTR, yielding an amplicon of about 2,200 bp, was used to detect exogenous ALVs as reported previously (ALV-F, GATGAGGCGAGCCCTCTCTTTG; ALV-R, TGTGGTGGGAGGTAAAATGGCGT) [11]. The PCR conditions used with the ALV-F/R primers were 95°C for 5 min, 31 cycles of 95°C for 50 s, 55°C for 40 s, and 72°C for 140 s, followed by a final elongation step of 10 min at 72°C. The PCR product was analyzed by 0.8% agarose gel electrophoresis in Tris-acetate-EDTA buffer. The amplicons were cloned, and the positive clones were confirmed by Sanger sequencing. All isolates obtained from positive samples were sequenced. The GenBank accession numbers of the isolates from this study are MW476816-MW476829 (Supplementary Table S2).

Sequence analysis

The sequences were aligned with other ALV-J reference sequences retrieved from the National Center for Biotechnology Information database (Supplementary Table S3). Phylogenetic analysis was performed by the neighbor-joining method with 1,000 bootstrap replicates, using MEGA ver. 6.0 [14].

Results

Increase in the ALV-positive rate of the flock through introduction of ALV-J-infected Hy-line brown roosters

Before introducing Hy-line brown roosters, the positive rate of virus isolation and p27 in albumen for the flock was less than 1% for two consecutive generations, and peak egg production was about 70%. However, after introducing the roosters, the positive rate of virus isolation in plasma and p27 in albumen increased to 5.4% (5/92) and 7.6% (7/92), respectively (Supplementary Table S1). Peak egg production dropped to about 65% without any other clinical signs. Ten albumen samples from five hens with positive plasma (two eggs per hen) were collected for virus isolation, and only one albumen sample was positive. These results indicate that the introduction of Hy-line brown roosters infected with ALV-J increased the positive rate of ALV in the flock.

ALV infection status in semen and plasma from roosters

Out of 46 roosters, 14 tested positive for ALV in their plasma or semen (30.43%). Of those, 12 were positive in plasma (26.09%), eight were positive in semen (17.39%), and six were positive in plasma and semen (Table 1). According to the virus isolation results from semen and plasma of the 46 roosters, there were four different ALV infection states. Six roosters were viremic, with positive semen (P + S+, 13.04%); six were viremic, with negative semen (P + S-, 13.04%); two showed no viremia, but positive semen (P-S+, 4.35%), and the remaining 32 roosters were double-negative (P-S-, 69.57%) (Table 2).

Table 1 ALV positive rates in samples subjected to virus isolation (VI) and ELISA
Full size table
Table 2 ALV infection status in the same rooster, determined by virus isolation
Full size table

High proportion of false-positives in semen tested directly by ELISA

In addition to virus isolation, the virus was detected directly in the semen by ELISA. The results showed that 11 semen samples were positive, but only six of those semen samples were from roosters whose plasma or semen samples were positive by virus isolation (Supplementary Table S1). These data suggest that ELISA method is not reliable for detection of ALV in semen, possibly due to nonspecific binding of unknown components of the semen to the p27 antigen.

Identification and sequence comparisons of ALV-J isolates from roosters, hens, and albumen

Fourteen positive samples from roosters, hens, and albumen were subjected to virus isolation and sequence analysis. Phylogenetic analysis based on the gp85 gene suggested that all of the isolates belonged to the ALV-J subgroup, but they formed a new branch within subgroup J, and their sequences were 49.6–94.8% identical to domestic and foreign reference sequences (Fig. 1). The sequence differences were mainly concentrated in the hr1, hr2, and vr2 regions. In addition, IFA was performed, and the results were consistent with the phylogenetic analysis, as clear green fluorescence was observed using monoclonal antibody JE9 (Fig. 2).

Fig. 1
figure 1

Phylogenetic analysis based on gp85 gene sequences of the isolates and referential strains. The phylogenetic tree was generated using MEGA5.1 software (bootstrap method with 1,000 replicates). ALV-A to E, K, and J are divided and indicated at the right. The source of each ALV-J reference strain is shown in parentheses. A green diamond indicates that the strain was isolated from rooster semen. A red triangle indicates that the strain was isolated from rooster plasma. A blue square indicates that the strain was isolated from hen plasma and albumen. The sources of the reference sequences and isolates are indicated. Abbreviations: USA, United States; UK, United Kingdom; CHN, China; RUS, Russia; EGY, Egypt; NGA, Nigeria

Full size image
Fig. 2
figure 2

Immunofluorescence images of DF-1 cells strained using the monoclonal antibody JE9. (A) Negative control, 100×. (B) DF-1 cells infected with semen, 100×

Full size image

Comparison of the gp85 nucleotide sequences in semen and plasma samples from the same rooster showed that they were > 98.6% identical in five of the six roosters, but those of the remaining rooster were only 96% identical to each other and 94.1% identical to those of the other five (Fig. 3). The virus isolates from the roosters, hens, and albumen samples showed 94.1–100% sequence identity in the gp85 gene. The sequence identity between the isolates from albumen and plasma from the same hen was 99.3%, and it was 98.7–99.8% between the isolates in the above five semen and plasma samples from the hen, suggesting that the virus was likely transmitted to the hens and their offspring through insemination by the roosters.

Fig. 3
figure 3

Sequence identity values for comparison of the 14 isolates, calculated using MegAlign in Lasergene 7.0

Full size image

Discussion

Since the first strain of ALV-J, HPRS-103, was isolated from meat-type breeder chickens in the UK in 1988, it has rapidly spread around the world, causing severe economic losses in the poultry industry [4, 15]. As the most pathogenic ALV subgroup, ALV-J induces immunosuppression and malignant or benign tumorigenic diseases in chickens, including hemangiomas, myelomas, and fibrosarcomas [2, 16,17,18]. An outbreak of hemangiomas associated with ALV-J was reported in commercial layer chickens between 2006 and 2010 in China [7, 19]. However, the virus has been successfully eradicated at the pedigree and multiplier levels by implementing an ALV purification project by a few major breeding companies. Therefore, there have been few reports of ALV-J infections in layers in recent years. The outbreak of ALV-J in broiler breeders during 2018–2019 was caused by introduction of broiler breeders from overseas [20]. Sun et al. discovered that coinfection with avian hepatitis E virus and avian leukosis virus subgroup J was the cause of an outbreak of hepatitis and liver hemorrhagic syndrome in a Hy-line brown layer chicken flock in China in 2017 [21]. In this study, we found subclinical infection of ALV-J in the roosters of Hy-line layers, indicating the importance of continued monitoring of roosters for ALV using the random sampling method.

Different methods have been used to detect the virus in various rooster and hen samples from the same flock. ELISA is used to detect viruses directly in semen, but in this study, only 50% of the ELISA results were consistent with the virus isolation results. Virus isolation is the only reliable method of confirming an ALV-J infection, whereas ELISA can produce both false-positive and false-negative results. At the same time, different infection states were identified in semen and plasma samples from the same rooster, namely, viremia but semen positive (V + S+, 13.04%), no viremia, but semen positive (V-S+, 4.35%), viremia, but semen negative (V + S-, 13.04%), and double-negative (V-S-, 69.56%). The positive rate of virus isolation from plasma was higher than that from semen, suggesting that both semen and plasma samples should be tested for the virus when attempting to eliminate ALV from roosters.

In experiments with specific-pathogen-free chickens, Li reported that ALV-J can be transmitted in semen to hens by insemination [13], but this resulted in ALV-J antibody production without viremia in the hens and their offspring. This can be explained by the low virulence of the ALV-J strain used in this experiment. The current study was based on infected clinical flocks. ALV-J was isolated from hens and breeding eggs. Although the positive rate of virus isolation in the albumen samples was low, one ALV-J strain was detected and sequenced. The isolates from albumen and the plasma from the same hen showed 99.3% sequence identity to each other and 98.7–99.8% identity to isolates from five semen samples and to isolates from plasma samples from the hen, demonstrating that they may have been derived from the same source. Thus, ALV-J from roosters can be transmitted to hens and their offspring through insemination.

Previous studies have demonstrated that ALV-J displays a high level of genetic variation and recombination [22, 23], which allows for the development of new variants with changes in antigenicity, tissue tropism, host range, and pathopoiesis [24]. Gao et al. reported that ALV-J is subjected to greater selection pressure in the ovarian follicles of hens, which promotes the evolution of the virus [25]. In this study, the ALV-J in the semen of roosters had only 96% sequence identity to that in the plasma, indicating that the rooster semen may have been under heavy selection pressure that promoted the evolution of ALV-J, which is why the isolates were located in a new branch in the phylogenetic tree when compared with the reference strains.

In summary, we isolated ALV-J from Hy-line brown roosters and determined the complete chain of transmission of ALV-J from roosters to hens through insemination and then to the offspring by vertical transmission. The use of ELISA to detect the ALV in semen did not provide reliable results. There were four ALV-J infection states in the plasma and semen of roosters, and it is therefore necessary to perform virus isolation from both semen and plasma. We speculate that the reason why sporadic cases of ALV-J infection in laying hens still occurs is incomplete elimination of the virus from roosters.