As a representative member of the 11 subgroups of avian leukosis virus (ALV-A–K) that have been recognized to date, ALV subgroup J (ALV-J) is an oncogenic virus that is associated with significant morbidity and causes avian leukosis, thus affecting the poultry industry worldwide (1). The genome of ALV comprises a single positive-strand linear RNA dimer, and its provirus presents a gene structure characteristic of C-type retroviruses, spanning a total length of 7–8 kilobases (2). The genes encoding the group-specific antigen (gag), polymerase (pol), and envelope glycoprotein (env) are located within the central coding region of the viral genome molecules (3). The noncoding region, mainly contained the long terminal repeat (LTR), is a pair of identical sequences of DNA situated at both ends of the genome, which comprises the 3′ unique region (U3), a short repeated region (R), and the 5′ unique region (U5) (4). These regions play crucial roles in viral replication, RNA processing, and integration of the virus into the host genomes (5, 6).

Currently, several research projects are dedicated to the elucidation of the pathogenesis of ALV-J, which is mainly related to its tumorigenic effect, immunosuppression, and decreased performance in chickens and layers (7). After the integration of the virus genome sequence into the host genome, the viral LTR plays a key role in the activation of the cellular proto-oncogenes, a process known as promoter insertion, which ultimately leads to multiorgan hemangioma and myeloid leukosis (8). Recent studies have shown that the aberrant expression of endogenous microRNAs (miRNAs), circular RNAs (circRNAs), and long noncoding RNAs (lncRNAs) caused by ALV-J infection is closely related to avian leukosis virus-induced tumorigenesis (9–12). These regulating effectors are mainly related to various activate/inactivate signaling pathways and differential expression change of genes at mRNA and protein levels.

Historically, DNA has been acknowledged as the carrier of genetic information, whereas mRNA has been identified as the central molecule that connects DNA to proteins and is crucial for the transmission of genetic information during various biological processes (13). To date, >100 chemically distinct modifications have been reported in cellular RNAs. Notably, in eukaryotic organisms, the presence of a modified 5′-cap at the 5′-end and a polyadenylate (poly(A)) tail at the 3′-end of mRNAs is crucial for the regulation of mRNA transcription (14). The prevalent alterations observed in mRNA mainly include N¹-methyladenosine (m¹A), 7-methylguanosine (m⁷G), N⁵-methylcytosine (m⁵C), and the prominent modification N⁶-methyladenosine (m⁶A) (15).

Generally, the m⁶A modification of RNAs occurs via a hetero-multimeric complex of nuclear methyltransferases (writers); in turn, this dynamic modification can be reversed by a group of demethylases (erasers), whereas a third group of proteins, known as readers, preferentially recognize the binding of m⁶A (readers) to methylated RNA and confer downstream functions (16). Therefore, aberrant m⁶A modifications and their related enzymes/proteins may be the best strategy for understanding the epigenetic mechanisms underlying the expression of important viral genes during the course of viral infection (17). Recently, there has been an increasing amount of research on the relevance of the m⁶A changes in viral replication and virus–host interactions in relation to virus induced illnesses in humans and animals (17). However, the functional characteristics of the m⁶A-mediated regulation of the RNAs/transcripts of DNA/RNA viruses remain unclear. According to the modification process of m⁶A, the core step consists in the recognition of the target genes, which calls for the accurate identification of RNA m⁶A sites (18). Numerous independent reports have revealed that the 5′-DRA*CH-3′ (where D stands for non-cytosine base; R for purine; H for non-guanine base; and A* for the methylatable adenosine) consensus motif is restricted for m⁶A binding; however, the mapping of these sites in viral RNAs/transcripts remains challenging and cumbersome (19, 20). Consequently, we aimed to use publicly accessible ALV-J genome sequences to predict m⁶A locations and study their evolutionary conservation, to stimulate research on m⁶A modification in ALV-J.

In recent years, the emergence of sequencing methods such as methylated RNA immunoprecipitation (MeRIP) and cross-linking and immunoprecipitation (CLIP) sequencing has given rise to a new layer of gene-expression regulation termed “epitranscriptomics,” which primarily focuses on the chemical alterations occurring on RNA molecules (28). However, these techniques can detect massive sequence segments containing m⁶A from the transcriptome, but cannot precisely identify the specific adenosine that undergoes methylation. Therefore, the use of computational tools to predict m⁶A sites from sequences would facilitate the rapid study of the m⁶a modification. From this viewpoint, the m⁶A motifs harbored in the genome of ALV-J strains available from GenBank were predicted using the SRAMP web-server. Consequently, more than 40 potential DRACH motifs were predicted in the genomes of these strains, among which 34.53 and 4.87% of the motifs had a “high confidence” and “very high confidence” grading, respectively. Studies related to the avian m⁶A modification revealed that “GGACU” was consistently regarded as the best and most highly enriched motif (29, 30). Moreover, “GGACU” was also preferred in the study of the transcriptome-wide dynamics of m⁶A methylation in chicken liver malignancies caused by ALV-J infection (22). Even though m6A motifs have not been determined for other subtypes of ALV, the consensus m6A sites across them would assist in unveiling the critical aspects of m6A modification in viral infection. According to the diversity analysis, the most common m⁶A-motif sequence of the DRACH motif of the ALV-J genome was also “GGACU,” which suggests the accuracy of this prediction, to some extent. Subsequently, although the 37 conserved motifs were predicted to be distributed in the U5 region of LTRs and all the major coding regions, they were enriched in the coding regions of the p27, p68, p32, and gp85 genes. This high conservation further confirmed the prediction accuracy for the m⁶A motif.

In the ALV-J genome, the gag gene is relatively conserved and mainly encodes glycosylated proteins, including p19, p10, p27, p12, and p15 (7). p19, which is alternatively referred to as the viral matrix protein (MA), plays a crucial role in the viral outgrowth process; p12, also known as the nucleocapsid protein (NC), is involved in the processing and packaging of RNA; p15 is a protease (PR) that is involved in the cleavage of the precursors of viral genome-encoded proteins; and p27 is a capsid protein (CA) that serves as a common group-specific antigen of ALV, which distinguishes ALV from other oncogenic viruses because of its high conservation (4, 31, 32). Fourteen potential m⁶A motifs were predicted in the coding region of p10, p12, and p15, and eight were predicted for p27. Although p27 is relatively conserved between the different groups of ALV, whether the m⁶A modification occurred in all ALVs via these motifs warrants further analysis.

The pol gene exhibits conservation across the different ALV subgroups and encodes two essential enzymes, i.e., the reverse transcriptase (p68) and the integrase (p32): p68 can use viral RNA as a template for reverse transcription, to produce proviral DNA, which is then inserted into the host chromosome under the integrative action of p32 (33, 34). Moreover, 14 potential m⁶A motifs were predicted, half each for p68 and p32, including the motif locate at 3740 with a conservation percentage of 95.30% and a rating of “very high confidence.” The motifs enriched in p68 and p32 also indicated that the m⁶A modification may play a vital role in the reverse transcriptase integrase process during the infection course of ALV-J.

The env gene is highly variable across subtypes and encodes glycosylated vesicle membrane proteins, including the outer membrane protein gp85 and transmembrane protein gp37, linked by hydrogen and disulfide bonds; thus, env is a key gene in ALV tumorigenesis (35). In the current study, the number of conserved motifs of the env gene was relatively low possibly because of the mutability of the gene. However, motifs located in the coding region of env were also predicted in different ALV-J strains. It remains unclear whether these non-conservedly predicted motifs are useful and correlated with differences in strain pathogenicity. gp37 protein has two structurally important hydrophobic regions for fusion with cell membranes. Therefore, it has a significant impact on the fusion of viral vesicle membranes with cell membranes; gp85 contains a viral receptor determinant that recognizes a specific receptor on the membrane of target cells for infection. As the major antigen of viruses, gp85 not only determines the subgroup identity and host range of ALV but also induces specific neutralizing antibodies in chickens (36). Furthermore, another motif located at position 3,740 with a rating of “very high confidence” was detected in the overlapped coding region of pol and env. The RNA secondary structures of this motif were predicted to be A–U pair or hairpin loop, which also indicated the complex RNA secondary structures induced by the sequences flanking each m⁶A motif. As reported previously, m⁶A stacking at the end of a hairpin loop stabilizes the loop, whereas the N⁶-methylation of an A–U pair in the middle of a helix would reduce the stability of helix folding (27). Refer to the conserved motifs predicted in ALV-J genomes, they mainly contained A–U pair or hairpin loop, and m⁶A modification of the unpaired adenosines attached in each helix have been confirmed to increase folding stability (27). Therefore, secondary structure prediction may help determine the structure–function relationships, thus promoting the understanding of the roles of m⁶A modifications and improving their prediction accuracy.

In this study, m⁶A modification motifs were predicted in different ALV-J strains, and their conservation was analyzed. Moreover, the RNA secondary structure of some representative motifs was determined. Although these motifs warrant further identification, a systematic comparative analysis would facilitate the rapid investigation of m⁶a modifications in ALV-J.

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/Supplementary material.

JJ: Investigation, Writing – original draft. XM: Data curation, Writing – original draft. SX: Software, Writing – original draft. XX: Data curation, Methodology, Writing – original draft. ZZ: Methodology, Visualization, Writing – original draft. LY: Supervision, Writing – review & editing. QX: Writing – review & editing. YB: Writing – review & editing.