Detection of avian leukosis virus subgroup J in chicken flocks from Malaysia and their molecular characterization

Abstract

Previously we have shown that avian leukosis virus subgroup J (ALV-J) might be present in chicken flocks from Malaysia based on serological study and also on detection of tissue samples with myelocytic infiltration. In this study, the polymerase chain reaction was used to detect ALV-J sequences from archived frozen samples. Out of 21 tissue samples examined, 16 samples were positive for proviral DNA and four samples for ALV-J RNA. However, only nine samples were found positive for myelocytic infiltration. A total of 465 base pairs equivalent to positions 5305 to 5769 of HPRS-103 from each of the viral RNA positive samples were characterized. Sequence analysis indicated that the samples showed high identity (95.9 to 98.2%) and were close to HPRS-103 with identities between 97.4 and 99.3%. This study indicates that ALV-J-specific sequences can be detected by polymerase chain reaction from frozen tissue samples with and without myelocytic infiltration.

Introduction

Avian leukosis virus subgroup J (ALV-J) is an economically important pathogen of meat-type birds. The virus was first reported in the UK in 1989 (Payne et al., 1991). ALV-J affects a wide range of domestic fowl including the turkey, but has not been reported from pheasants, quail, ducks and goose (Payne et al., 1992). Recent studies have shown that ALV-J infection in layer chickens is associated with economic losses (Gingerich et al., 2002; Binrui et al., 2004). The main clinical abnormality associated with ALV-J is myeloid leukosis, which is characterized mainly by the presence of whitish nodules, particularly at the skeletal tissues and also in the visceral organs. The nodules consist primarily of myelocytes with characteristic eosinophilic cytoplasmic granules (Fadly & Smith, 1999). However, other forms of lesions such as histiocytic sarcomatosis, pulmonary sarcoma, feather abnormalities and myelocyte infiltration in bones and periosteum of the sternum and ribs have also been reported (Arshad et al., 1997; Hafner et al., 1998; Nakamura et al., 2000; Landman et al., 2001).

The genome of the proviral form of ALV-J comprises three genes, gag, pol and env, flanked with a long terminal repeat (Bai et al., 1995a). The gag and pol genes are well conserved with those of other subgroups, with overall nucleotide identity of 96 to 97%, whereas the env gene of ALV-J is highly diverged from that of other ALVs, but more closely related to the EAV-HP family of endogenous viruses. It is thought that ALV-J originated through recombination of exogenous viruses (gag and pol genes) and endogenous viruses (env gene) (Bai et al., 1995b).

Currently, several methods are available for the diagnosis of ALV-J including histopathological examination, serological tests, in situ hybridization, polymerase chain reaction (PCR) and virus isolation (Fadly, 2000). Among them, the PCR has been found to be rapid, specific and more sensitive than most conventional diagnostic tests (Smith et al., 1998a,b; Garcia et al., 2003). In addition, the PCR has been found successful in detecting proviral DNA and viral RNA from various tissues including tumours, blood, serum and feathers (Smith et al., 1998a,b; Davidson & Borenshtain, 2002; Sung et al., 2002; Garcia et al., 2003).

The status of ALV-J in meat-type chickens in Malaysia is not known. However, a survey carried out at several broiler breeder farms in the northern part of Malaysia indicated that up to 50% of the chickens were positive for antibody against ALV-J by the age of 55 weeks (Siew, 2001). The evidence of eosinophilic myelocyte infiltration in various tissue samples further suggested that ALV-J infection might be present in Malaysia (Asiah et al., 2001). However, no further studies were carried out to confirm the presence of ALV-J from those samples. In this paper, we describe the detection and characterization of ALV-J-specific sequences from tissues samples with and without myelocytic infiltration.

Materials and Methods

Sample history and collection

A total of 21 tissue samples (brain, liver, trachea, spleen, heart, kidney and lung), of which five samples were from layer chickens age between 7 and 20 weeks and the remaining were from broiler chickens age between 3 and 6 weeks, were used in this study. The samples were kindly provided by Dr Asiah Naina Mohd of the Regional Veterinary Diagnostic Laboratory, Department of Veterinary Service. All the samples were collected or received from various chicken farms in 2000. Fresh samples were fixed in 10% formalin for histopathological examination while the remaining samples were stored at −30°C. Most of the samples had history of Newcastle disease virus outbreak, poor growth rate and feather abnormality, but no visible gross tumours were observed. However, based on histopathological reports, myelocytic infiltration was observed in samples of the lung, liver, heart, kidney and brain.

Polymerase chain reaction

Total DNA was extracted from 500 μl of a 30% tissue homogenate in phosphate buffer saline using the standard methods (Sambrook et al., 1989), while total RNA was extracted using Trizol reagent (GIBCO BRL, Life Technologies) following the methods recommended by the manufacturer. PCR detection of proviral DNA and viral RNA was performed using specific primers (H5/H7) and PCR profiles as established by Smith et al. (1998a). The RNA was treated with 1 μl DNase (1 u/μl) (Promega, USA) to degrade the DNA. Reverse transcription (RT) was performed in 50 μl reaction volume consisting of 10 μl of 5×AMV reverse transcriptase buffer (Promega, USA), 1 μl dNTP (10 mM), 1 μl AMV-RT enzyme, 1 μl RNase inhibitor, 1 μl of 0.1 M dithiothreitol, 50 pmol each of forward and reverse primers, 5 μl RNA (200 ng/μl) in total volume and nuclease-free water. The reaction mixture was then heated in a thermal cycler at 48°C for 1 h for cDNA synthesis followed by 94°C for 2 min to inactivate the AMV-RT enzyme. PCR was carried out as already described except 500 ng cDNA was used as the DNA template. The PCR products that were positive for viral RNA were purified using the Gene Clean kit (BIO101, USA) following the methods recommended by the manufacturer.

DNA sequencing

Four samples that were found positive for viral RNA, UPM/A6, UPM/A10, UPM/A17 and UPM/A18 were sequenced using ET DYE Terminator Cycle Sequencing Kit (Amersham Bioscience) in an automated DNA sequencer (ABI PRISM 377 DNA Sequencer). A total of 465 base pair (bp) nucleotide sequences from each of the PCR products were determined. The accession numbers of the sequences used in the analysis were HPRS-103 (Z46390) (Bai et al., 1995a), ADOL-HC-1 (AF97731), ADOL-7501 (AY027920), ADOL-R5-4 (AF076887) (Benson et al., 1998a), 0661 (AF247566) (Silva et al., 2000), UD5 (AF307952), UD4 (AF307951), TW99 (AF497905) (Wang & Juan, 2002), IMC10200 (AY234051), and SDC2000 (AY234052). Phylogenetic analysis was performed on 294 bp (98 amino acid residues) of the first half of the gp85 region of the env gene, along with bootstrap analysis using ClustalX version 1.83.

Results and Discussion

From the total of 21 tissue samples examined, 16 samples comprising the brain, liver, trachea, kidney, heart, lung and spleen were found positive for ALV-J proviral DNA, while only four samples (heart, brain, lung and spleen) were found positive for viral RNA. From the 16 samples positive for proviral DNA, only nine showed myelocytic infiltrations, indicating that occurrence of such lesions is not essential for the molecular detection of ALV-J. It has been shown that PCR detection of proviral DNA is more sensitive than conventional diagnostic methods (Smith et al., 1998a,b; Garcia et al., 2003). The detection of proviral DNA in various tissue samples might reflect the virus tissue tropism. Previous studies have indicated that ALV-J has high tissue tropism towards several organs such as the heart, adrenal gland, proventriculus and kidney (Arshad et al., 1999; Gharaibeh et al., 2001; Stedman et al., 2001). Davidson & Borenshtain (2002) indicated that ALV-J DNA can be detected from the spleen and liver. However, feather tips were more effective for diagnosis of naturally infected chickens. The samples that were positive for viral RNA (UPM/A6, UPM/A10, UPM/A17 and UPM/A18) were exclusively from tissues that showed myelocytic infiltration. Interestingly, the first two samples were from the brain, while the next two were heart and spleen samples, respectively. This finding also reflects the vast ability of the virus to replicate in various tissues. Previous studies have shown that viral RNA is mostly detected from the heart (Arshad et al., 1999; Stedman et al., 2001). The detection of only four samples positive for ALV-J RNA may be due to a low level of ALV-J transcript beyond the detection limit of the RT-PCR assay. It is also interesting to note that the samples were obtained from chicken flocks associated with poor growth rate since it has been shown previously that chickens infected with ALV-J have low body weight (Stedman & Brown, 1999; Landman et al., 2002).

A total of 465 bp from each of the four samples UPM/A6, UPM/A10, UPM/A17 and UPM/A18 were determined and deposited in the GenBank database under the accession numbers AY312966, AY312965, AY312967 and AY312968, respectively. The sequence length was equivalent to nucleotide positions 5305 to 5769 of HPRS-103, which includes 42 bp of the pol gene (positions 5305 to 5346), 127 bp of signal peptide cleavage (positions 5349 to 5475) and 294 bp of gp85 of the env gene (positions 5476 to 5769) (Bai et al., 1995a). Similar to HPRS-103 and other ALV-J isolates, a single nucleotide substitution from C to T at position 5346, leading to development of a premature stop codon resulting in a 22 amino acid shorter pol protein, was observed in all the samples. The presence of this premature stop codon suggested that ALV-J from different countries might have originated from a common ancestor as proposed earlier by Benson et al. (1998b).

The first 44 amino acids of gp85 region were found to be conserved but scattered amino acid substitutions were noticed in the remaining 54 amino acids (Figure 1). UPM/A6, UPM/A10, UPM/A17 and UPM/A18 have three, 10, three and eight amino acid substitutions, respectively, when compared with the HPRS-103 sequence from positions 8 to 162. UPM/A10 showed the highest amino acid substitutions; three substitutions at the pol and signal peptide region, and seven substitutions at the gp85 region, while all eight amino acid substitutions on UPM/A18 were found at the gp85 region (Figure 1). No amino acid substitution was noticed at the pol and signal peptide cleavage of UPM/A6, but three amino acid substitutions were noticed at the gp85 region. On the other hand, three amino acid substitutions were observed in the pol and signal peptide cleavage region but no substitution was seen at the gp85 region of UPM/A17. Interestingly, a cysteine residue at position 136 that was conserved in the prototype and variant ALV-J strains from the UK (Venugopal et al., 1998) and the US (Silva et al., 2000) was found different in UPM/A6, UPM/A10 and UPM/A18 (Figure 1). UPM/A6 also showed a substitution in the cysteine residue at position 124. However, the importance of these amino acid substitutions is not known.

Figure 1. Deduced amino acid sequences of UPM/A6, UPM/A10, UPM/A17, UPM/A18 and other ALV-J published sequences. The amino acid sequences correspond to positions 8 to 162 of HPRS-103 (Bai et al., 1995a), encompassing 43 amino acids of the pol gene, 14 amino acids of signal peptide cleavage and 98 amino acids of the gp85 domain of the env gene.

The percentage nucleotide sequence identities among the four samples were high, ranging from 95.9 to 98.2%. All the samples were also close to HPRS-103 isolate from the UK (with identities ranging from 97.4 to 99.3%) and the US isolates UD4, ADOL-R5-4 and ADOL-HC-1 (95.2 to 98.9%). However, when compared with ADOL-7501, another ALV-J isolate from the US, the percentage identity ranged from 83.6 to 85.0%, with UPM/A10 showing the lowest identity. Compared with HPRS-103, UPM/A6 showed the highest identity (99.3%) while UPM/A18 showed the lowest identity (97.4%). Phylogenetic analysis indicated that all the samples were close to HPRS-103 with a low genetic distance (0.00 to 0.03), whereas with ADOL-7501 they showed a high genetic distance (0.202) (Figure 2). UPM/A18 was found to be the most distantly related with genetic distance (0.03), followed by UPM/A10 (0.02), UPM/A6 (0.009) and UPM/A17 (0.00) when compared with HPRS-103. This finding remains to be confirmed by phylogenetic analysis based on the entire sequence of the env gene.

Figure 2. Phylogenetic analysis of ALV-J sequences of UPM/A6, UPM/A10, UPM/A17 and UPM/A18 compared with ALV-J isolates characterized elsewhere using the ClustalX program. The numbers at the branch points represents the result of bootstrap analysis of 1000 samples.

The detection of ALV-J from samples obtained from layer chickens (UPM/A10) is not an unusual phenomenon, since recent studies have indicated that layer chickens are also susceptible to both infection and tumour development following ALV-J infection (Gingerich et al., 2002; Binrui et al., 2004). Further studies on molecular characteristic of UPM/A10 are essential since it has been shown that ALV-J that affects layer chickens had undergone recombination with subgroup B ALV (Gingerich et al., 2002). Characterization of the entire env gene of the samples based on sequencing and expression studies is also essential due to the high genetic variability of ALV-J. However, no virus was isolated from the inoculation of the respective samples into DF-1 cells based on PCR and the indirect immunofluorescent antibody test using monoclonal antibody against ALV-J gp85 (unpublished data). The actual reason is not known, but probably the samples had lost their infectivity due to the repeated thawing–freezing process and also the long-term storage of the samples at −30°C. In summary, PCR is a convenient tool to detect ALV-J directly from field tissue samples prior to the development of myeloid tumour and/or myelocytic infiltration.

Acknowledgements

The authors would like to thank Dr Asiah Naina Mohd for providing the samples and Dr Venugopal Nair, Institute for Animal Health, Compton, UK for providing the monoclonal antibody against ALV-J gp85. This project was jointly supported by the Strengthening of Veterinary Services and Livestock Disease Control Project, Department of Livestock Service, HMG, Nepal and Government of Malaysia under Grant No. 09-02-04-0700-EA001.