Compared with mammals, the bird Ig genetic system relies on gene conversion to create an Ab repertoire, with inversion of the IgA-encoding gene and very few cases of Ig subclass diversification. Although gene conversion has been studied intensively, class-switch recombination, a mechanism by which the IgH C region is exchanged, has rarely been investigated in birds. In this study, based on the published genome of pigeon (Columba livia) and high-throughput transcriptome sequencing of immune-related tissues, we identified a transcriptionally forward α gene and found that the pigeon IgH gene locus is arranged as μ-α-υ1-υ2. In this article, we show that both DNA deletion and inversion may result from IgA and IgY class switching, and similar junction patterns were observed for both types of class-switch recombination. We also identified two subclasses of υ genes in pigeon, which share low sequence identity. Phylogenetic analysis suggests that divergence of the two pigeon υ genes occurred during the early stage of bird evolution. The data obtained in this study provide new insight into class-switch recombination and Ig gene evolution in birds.

As essential components of adaptive immunity in jawed vertebrates, Igs (or Abs) have a critical role in providing immunologic protection against pathogens (1, 2). The key functions of these defense molecules depend on their V and C regions. The V region, which is responsible for recognizing and binding to diverse Ags, is shaped through several molecular mechanisms, such as V(D)J recombination, somatic hypermutation, and gene conversion. In contrast, the C region specifies multiple Ig classes performing distinct effector functions. Different IgH classes (isotypes) are produced by a genetically programmed gene recombination process termed class-switch recombination (CSR); in this process, exons for Cμ can be exchanged for a downstream H chain C region (CH), such as Cγ, Cε, or Cα in mammals. Therefore, mammalian B cells can switch from expressing IgM and IgD on their surface to expressing IgG, IgE, or IgA (3).

Although the functions of Ig classes (or subclasses) and the above-mentioned molecular mechanisms have intensively been studied in mammals, especially in human and mouse, such aspects are relatively much less investigated in nonmammal tetrapods. Compared with mammals, the Ig system in avians is characterized by gene conversion usage to create the Ab repertoire, the absence of an IgD-encoding gene in many species, IgA-encoding gene inversion, and uncommon subclass diversification (47).

In mammals, many structurally intact, functional variable gene segments are subject to V(D)J recombination, whereas in avians, there is typically only a single variable gene involved in V(D)J recombination of both H chain and L chain gene loci (812). Thus, V(D)J recombination in birds can only generate very limited recombinatorial sequence diversity. In contrast, many pseudovariable gene segments upstream of the functional avian variable gene can be used as donor sequences to modify a recombined variable gene through gene conversion; moreover, this modification can occur repeatedly (4). Indeed, gene conversion to generate the Ab repertoire is the main mechanism observed in many bird species (13).

For decades, it has been thought that bird genomes lack an IgD-encoding gene, even though this gene is found in all other groups of jawed vertebrates, including cartilaginous fishes (14). Despite its absence in certain birds, such as chicken and duck, our group very recently showed that an IgD-encoding gene is present in many bird species (6, 7, 15), and four Ig classes—IgM, IgD, IgY (encoded by υ gene), and IgA (encoded by α gene)—have been found in birds to date. Although structurally different from mammalian IgG, IgY is thought to be its functional analogue in avians (16). However, in contrast to the extensive subclass diversification of IgG described in mammals, IgY subclass diversification has rarely been reported in birds (15).

IgA is responsible for mucosal immunity in avians, as it is in mammals (17). A remarkable feature of the bird IgA-encoding gene (α gene) is its transcriptional orientation: the α gene is inverted to μ and υ genes within both chicken and duck IgH loci (6, 7). Similarly, our recent work on the most ancient extant bird, the ostrich, as well as some nonavian reptiles, such as crocodilians, revealed inversion of the α gene (15, 18). These findings suggest that the genetic event that resulted in α gene inversion occurred before the evolutionary divergence of crocodilians and avians. In mammals, intrachromosomal deletional recombination is required for switching from IgM production to that of other Ig isotypes, whereas IgA production in avians or crocodilians apparently requires α gene inversion rather than the usual deletion. Regardless, CSR in these species remains poorly understood.

In addition to an inverted α gene, IgD-encoding genes have been lost in many birds, such as chicken and duck, as mentioned above (6, 7). These observations caused us to speculate that δ gene loss might be associated with α gene inversion and that both resulted from the same genetic rearrangement event. However, this hypothesis has been disproven, as our recent studies on ostrich showed that both a δ gene and an inverted α gene are present within the same IgH locus (15), suggesting that the δ gene loss and α gene inversion were independent genetic events. Similarly, subclass diversification of μ and υ genes in birds has only been observed in ostrich; only one μ gene and one υ gene are found within the IgH loci of chicken and duck, even though chicken and duck belong to Galloanserae, the members of which are closely related to ostrich (6, 7, 15, 19). Taken together, these data indicate that the IgH gene locus has experienced major genetic alterations during bird evolution.

The extant avian lineage is the most species-rich class of tetrapods, exhibiting extremely diverse morphologies and rates of diversification (20). As previously mentioned, Ig genes in different bird species exhibit significant differences. Columba livia (pigeon) belongs to Neoaves, which includes most living avians. Because the IgH gene locus in this clade has yet to be investigated (19), we present a detailed study of IgH genes in C. livia based on the published genome. Our results show that the IgH gene locus in pigeon contains a transcriptionally forward α gene and two υ genes; our data indicate that divergence of the two υ genes occurred at a very early stage of bird evolution. To the best of our knowledge, we also provide the first detailed analysis of CSR in birds.

Pigeons (C. livia) at the age of ∼1 y old were purchased from a local pet market. Total RNA from different tissues was prepared using TRNzol kit (Tiangen Biotech). Reverse transcription reactions were conducted using M-MLV Reverse Transcriptase according to the manufacturer’s instructions (Promega). Genomic DNA was isolated from liver tissues following the routine protocols. All animal studies and procedures were approved by the Animal Care and Use Committee of Henan University.

Total RNA samples extracted from the spleen and intestine were pooled together. RNA purity, concentration, and integrity were assessed using a Nanodrop, Qubit 2.0, and Agilent 2100 Bioanalyzer, respectively. The RNA library generated was examined using an Agilent 2100 and the ABI StepOnePlus Real-Time PCR System (Life Technologies) and then sequenced using the Illumina HiSeq 2000 platform (Illumina).

A Translated BLAST (tblastn) approach was used to search the complete pigeon (C. livia) genome at the National Center for Biotechnology Information. Ig genes obtained from transcriptome sequencing of pigeon and other birds were used as queries to search for candidate genome scaffolds containing Ig genes. The H chain V region (VH) gene domain (framework regions or CDRs) was classified using the ImMunoGeneTics (IMGT) numbering system (21). Potentially functional, open reading frame, and pseudo-V segments were all identified according to functionality of IMGT (22). Recombination signal sequences (RSSs) for variable, diversity, and joining gene segments were analyzed using the online program fuzznuc (http://embossgui.sourceforge.net/demo/fuzznuc.html).

PCR reactions were performed using liver genomic DNA as the template and Expand Long Range dNTPack (Roche). Amplified products were purified using Wizard SV Gel and PCR Clean-Up System (Promega) and then sequenced.

Sense and antisense primers (sequences are shown in Supplemental Table I) were designed based on the 5′-flanking and 3′-flanking regions of Sμ, Sα, and Sυ2, respectively, and used for nested PCR to amplify recombination fragments of Sμ-Sα and Sμ-Sυ2. PCR was performed using high-fidelity enzyme KOD -Plus- (Toyobo). The resulting products were cloned into the pMD19-T vector and sequenced. The junction sites of these recombined fragments were analyzed via sequence alignment with germline switch regions (S regions).

Total RNA was isolated from different tissues using an RNeasy Mini Kit (Qiagen), and the RNA quantity and quality were assessed with a Nanodrop 2000 (Thermo Fisher Scientific). First-strand cDNA was synthesized using 1 μg RNA and QuantiTect Reverse Transcription Kit (Qiagen). Quantitative RT-PCR (qRT-PCR) reactions were performed using LightCycler 480 SYBR Green I Master (Roche) under the following cycling conditions: 95°C for 5 min followed by 40 cycles of 95°C for 10 s, 60°C for 15 s, and 72°C for 10 s. Each sample was run in triplicate. The pigeon EF2A gene was chosen as the internal control. All primers used in the experiments are shown in Supplemental Table I.

Expressed pigeon VH genes were amplified using High Fidelity PCR Master (Roche) using primers designed based on the leader peptide of VH1 and Cμ genes (Supplemental Table I). The PCR products were cloned into the pMD-19 T vector (Takara) and sequenced.

DNA and protein sequence editing, alignments, and comparisons were performed using the DNASTAR program. S regions were analyzed using DotPlot comparison (MegAlign; DNASTAR) and the EMBOSS etandem program (http://www.bioinformatics.nl/emboss-explorer/). Multiple sequence alignments were performed using the ClustalW program. A phylogenetic tree was constructed using MrBayes version 3.1.2 (23) and viewed in FigTree version 1.4.1 (BEAST). The accession numbers for the sequences (http://www.ncbi.nlm.nih.gov) used are as follows: μ, nurse shark, I50731; lung fish, AAO52808.1; zebrafish, AF281480; Xenopus tropicalis, AAH89670; axolotl, A46532; lizard, ABV66128; Chinese soft-shell turtle, ACU45376; crocodile, AFZ39177; ostrich, AFA41926; emu, APB61245; chicken, P01875; duck, CAC43061; goose, KT253939; human, AAS01769; mouse, CAA24199; platypus, AAO37747; horse, AAU09792; cattle, AAN60017; α or χ, X. tropicalis, AAI57651; axolotl, CAO82107; crocodile, AFZ39174; ostrich, AFA41929; emu, KU641023; goose, KT253941; chicken, AAB22614; duck, AAK17834; human, P01876; mouse, AAB59662; pig, I47175; cattle, AAC98391; horse, AAP80145; platypus, AAL17700; δ, X. tropicalis, ABC75541; lizard, ABV75541; Chinese soft-shelled turtle, ACU45375; crocodile, AFZ39209; human, AAA52770; mouse, P01881; platypus, ACD31540; υ, X. laevis, CAA33212; X. tropicalis, AAH89679; Pleurodeles waltl, CAE02686; axolotl, S31436; salamander, AIW06020; lizard, ABV66132; gecko, ACF60236; Chinese soft-shell turtle, ACU45374; snake IgY1, AFR33842; snake IgY2, AFR33843; alligator IgY1, AFZ39169; alligator IgY2, AFZ39170; alligator IgY3, AFZ39171; emu IgY1, KU641026; emu IgY2, KU641027; chicken, CAA30161; duck, CAD57004; goose, KT253942; ostrich IgY1, AFA41930; ostrich IgY2, KM510516; zebra finch, ASU87374; γ, human, J00228; horse, AJ302055; cattle, AQT27058; mouse, J00453; platypus, AY055781; ε, human, J00222; mouse, X01857; horse, AAA85662; cattle, AY221098; and platypus, AY055780.

To analyze the IgH isotypes expressed in pigeon, total RNA samples isolated from spleen and intestine tissues were mixed together and subjected to transcriptome sequencing. A total of 28,680,100 clean reads were obtained, accounting for ∼181,936 transcript and ∼102,292 unigene sequences. Previously published sequences of ostrich and duck IgH genes were used as queries in Basic Local Alignment Search Tool searches against these unigenes. Using the transcriptome database, we identified four IgH genes: one gene each encoding IgM and IgA and two genes encoding IgY.

Both DNA and deduced amino acid sequences of the IgH genes obtained from transcriptome sequencing were used in a Basic Local Alignment Search Tool search against the pigeon genome. The entire pigeon IgH locus was found in scaffold 516 (GenBank accession no. KB375780), which spans ∼67 kb from the most 5′ VH segment to the most 3′ υ2 gene (Fig. 1A). The VH locus occupies ∼11 kb of DNA, containing 14 VH segments. Only one VH gene is potentially functional. The remaining 13 segments contain one open reading frame and 12 pseudogenes, as these segments lack a proper leader, RSS, or are truncated in the coding region (Supplemental Fig. 1A). The pseudo-VH segments are located within a short distance of each other, whereas the potentially functional VH segment (VH1) is separated by ∼0.3 kb from upstream pseudo-VH genes and is located at the most 3′-end of the VH locus (Fig. 1A). Regarding the pigeon IgH locus, two H chain D region (DH) segments and one H chain J region (JH) segment were identified in the 7-kb region between the functional VH and CH genes. Both DH segments are open in all three reading frames and are flanked by characteristic heptamers and nonamers separated by 12-bp spacers (Supplemental Fig. 1C). The JH segment and corresponding RSS are displayed in Supplemental Fig. 1C.

FIGURE 1.

(A) Physical map of the pigeon IgH gene locus. The filled bars indicate potentially functional VH genes and the open bars indicate VH pseudogenes. DH, diversity genes; JH, joining genes; M, membrane region–encoding exon; α, IgA-encoding gene; μ, IgM-encoding gene; υ, IgY-encoding gene. (B) Long-range PCR amplification of the DNA fragment between the μ and α genes. M, 1-kb DNA marker; 1, 8-kb DNA fragment.

FIGURE 1.

(A) Physical map of the pigeon IgH gene locus. The filled bars indicate potentially functional VH genes and the open bars indicate VH pseudogenes. DH, diversity genes; JH, joining genes; M, membrane region–encoding exon; α, IgA-encoding gene; μ, IgM-encoding gene; υ, IgY-encoding gene. (B) Long-range PCR amplification of the DNA fragment between the μ and α genes. M, 1-kb DNA marker; 1, 8-kb DNA fragment.

Close modal

Downstream of the JH gene, ∼47 kb of DNA consist of four C region genes arranged as μ-α-υ1-υ2 (Fig. 1A). We did not find any potential δ gene sequence in the intron between the μ and α genes (∼6.4 kb) or in the entire pigeon genome, suggesting the absence of an IgD-encoding gene in this species. These findings were consistent with our transcriptome analysis results. Similar to other bird species, each CH gene contains four constant domains and two transmembrane exons. All birds researched to date possess an inverted α gene. However, the α gene in the pigeon IgH gene locus exhibits a transcriptional orientation that is identical to all other CH genes. To confirm this, we employed long-range PCR to amplify the genomic region between the μ and α genes using primers based on the transmembrane exon of the Cμ gene and Cα1 exon, respectively. Using this approach, we amplified a DNA fragment of ∼8 kb, slightly longer than the genome size (∼7 kb) published in National Center for Biotechnology Information (Fig. 1B). This difference is very likely due to two gaps that exist in the sequenced pigeon genome between genes μ and α. The sequence of the 8-kb DNA fragment reveals that the Cα1 exon is located directly downstream of the μ gene.

The IgH C region of pigeon secretory IgM encodes 456 aa. Alignment of the pigeon μ gene (inferred amino acids) and those of other species shows a conserved structure. For example, 12 positionally conserved cysteines are distributed in four CH domains and the secretory tail, enabling intra– and inter–H chain disulfide bond formation (Supplemental Fig. 2A). Five potential N-linked glycosylation sites (N-X-S/T), N-43, N-133, N-204, N-305, and N-443, were identified throughout the entire C region. Only N-204 and N-443 are conserved among reptiles, avians, and mammals; the N-43 and N-305 sites are found exclusively in avians. Similar to other birds, the pigeon μ gene shares higher amino acid identity with the ostrich μ1 gene (54.6%) than the ostrich μ2 gene (45.6%), and our phylogenetic analysis also revealed that the pigeon μ gene is closely related to the chicken μ gene (Fig. 2). qRT-PCR analysis showed the μ gene to be highly expressed in the spleen but lowly expressed in the liver, intestine, and kidney (Fig. 3).

FIGURE 2.

Phylogenetic analysis of pigeon CH genes. The tree was constructed using amino acid sequences of entire CH domains, except that the first four CH domains were used for tetrapod IgD genes. The scale bar represents the genetic distance, and the credibility value is noted beside each node.

FIGURE 2.

Phylogenetic analysis of pigeon CH genes. The tree was constructed using amino acid sequences of entire CH domains, except that the first four CH domains were used for tetrapod IgD genes. The scale bar represents the genetic distance, and the credibility value is noted beside each node.

Close modal
FIGURE 3.

Relative expression levels of μ, α, υ1, and υ2 genes in different tissues. The data are representative of three independent experiments. The y-axis indicates fold normalized expression. The bars indicate SDs from the mean. Expression levels of the μ (A), α (B), υ1 (C), and υ2 (D) genes in different tissues.

FIGURE 3.

Relative expression levels of μ, α, υ1, and υ2 genes in different tissues. The data are representative of three independent experiments. The y-axis indicates fold normalized expression. The bars indicate SDs from the mean. Expression levels of the μ (A), α (B), υ1 (C), and υ2 (D) genes in different tissues.

Close modal

The pigeon α gene locus spans 6.4 kb, which is shorter than its counterpart gene in chicken (11.5 kb) and duck (10.9 kb) (6, 7). The transmembrane anchor of the pigeon α gene is encoded by two separate exons, a situation that is different to that in duck, whereby a single exon encodes the transmembrane region of IgA. Nonetheless, the highly conserved residues of the conserved Ag receptor transmembrane (CART) motif, which is involved in signal transduction, are present in the hydrophobic transmembrane region (24). Amino acid sequence alignment of Cα genes revealed an identical pattern of cysteine distribution among different species (Supplemental Fig. 2B). Five putative N-linked glycosylation sites were also observed in Cα2, Cα3, and the secretory tail of IgA. N-169, N-226, and N-424 are conserved in various species, whereas N-197 and N-300 are unique to pigeon (Supplemental Fig. 2B). IgA is the principal Ab class in secretions of the respiratory, gastrointestinal, and genitourinary tracts (25). Our qRT-PCR data indicate that pigeon IgA is primarily expressed in the spleen, liver, and small intestine, with the highest level of expression in the latter tissue (Fig. 3).

Within ∼25 kb downstream of the α gene, we identified two functional υ genes in the pigeon IgH gene locus. Comparison of these two υ genes revealed no significant difference in the distribution of cysteines and potential N-linked glycosylation sites, with the exception of one additional potential N-linked glycosylation site in CH3 of the υ1 gene (Supplemental Fig. 2C). However, the entire CHs of both υ genes share an overall nucleotide identity of only 51.6%, and this value drops to 36.2% at the amino acid level. To date, IgY subclass diversification in avians is only found in the ostrich and emu, two ancient extant bird species (15). Our phylogenetic analysis suggested that divergence of the two υ genes occurred in the early stage of bird evolution (Figs. 2, 4). Although the pattern of υ1 gene expression in different tissues was generally consistent with that of υ2, with both, genes were mainly expressed in the spleen, liver, kidney, and large intestine, and our qRT-PCR data showed a much higher level for the former than the latter in all tissues (Fig. 3).

FIGURE 4.

Schematic diagram of IgH gene evolution in birds and crocodiles. Genes are indicated as colored boxes. Arrows are used to indicate the transcriptional orientation of α genes.

FIGURE 4.

Schematic diagram of IgH gene evolution in birds and crocodiles. Genes are indicated as colored boxes. Arrows are used to indicate the transcriptional orientation of α genes.

Close modal

It is known that CSR occurs between S regions located upstream of each CH, except for Cδ. S regions typically consist of short, tandem-repeated sequences, with an overall length varying from ∼1 to 12 kb (3). According to these criteria, we identified and analyzed S regions for all CH genes within the pigeon IgH locus (Supplemental Fig. 3). Similar to mammals, all pigeon S regions are rich in activation-induced cytidine deaminase hotspot sites (RGYW/WRYC), and there is no apparent homology among different S regions (26, 27). Pigeon Sμ is ∼3 kb long and is composed of multiple repeat motifs ranging in size from 10 to 150 bp. Most repeats in the pigeon Sμ region contain scattered GGTT motifs, whereas AGCT motifs are abundant in mouse and human Sμ regions (28, 29). The pigeon Sα region is ∼2 kb in length, with 37 repeats of a 52-bp sequence that consists of ∼46% G:C base pairs, whereas mouse and human Sα regions consist of ∼60% G:C (28, 29). Furthermore, the sequence of the pigeon Sα region shows significant difference from that of duck (6). Although both pigeon Sυ1 and Sυ2 sequences are enriched in G:C, at ∼60%, the same degree of G:C asymmetry is observed. Pigeon Sυ1 tandem repeats show a G:C ratio of 3.1, whereas that of Sυ2 is 1.8. It is thought that a high G:C ratio in mammalian S regions is important for formation and maintenance of R-loops during germline transcription, and a low guanine density within S regions dramatically decreases CSR efficiency (30, 31), which may explain the higher level of υ1 gene expression compared with the υ2 gene. Moreover, the pigeon Sυ1 region shows large sequence variation from that of the Sυ2 region but shares 63.1% identity with the duck Sυ region (6).

It is known that CSR in humans and mice is initiated by activation-induced cytidine deaminase, which promotes conversion of dC bases to dU bases in donor and acceptor S regions (32). dU bases are removed and then converted into DNA double-strand breaks (DSBs) required for CSR (3, 32); this process can occur in two orientations and results in deletional and inversional switch recombination. However, the details of both types of CSR junctions have not been elucidated in birds. To address this question, sense primers derived from Sμ, Sα, and Sυ2 5′-flanking regions and antisense primers from 3′-flanking regions of Sμ, Sα, and Sυ2 were designed (Fig. 5A) to amplify the recombined Sμ-Sα and Sμ-Sυ2 fragments by means of two-round nested PCR. Using spleen genomic DNA as a template, multiple somatic recombined fragments derived from DNA deletion and inversion were amplified and cloned (Fig. 5B). After sequencing, we obtained 47 unique Sμ-Sα recombined fragments and 48 Sμ-Sυ2 fragments generated by DNA deletion and 30 Sμ-Sα fragments and 11 Sμ-Sυ2 fragments generated by DNA inversion (Table I). These results suggest that both DNA deletional joining and inversional joining can occur during CSR in pigeon.

FIGURE 5.

Amplification of recombined switch fragments of Sμ-Sα and Sμ-Sυ2. (A) Schematic diagram of DNA fragment deletion and inversion during IgA and IgY2 CSR. (B) PCR amplification of recombined switch fragments. (C) Three representative junction structures of the recombined switch fragments. For each alignment, the sequence of the recombined fragment is shown in the middle; the germline sequences of the two S regions that are involved are shown above and below. The arrows indicate break points, whereas boxed nucleotides denote microhomology shared by two S regions. The inserted nucleotide in the junction site is underlined.

FIGURE 5.

Amplification of recombined switch fragments of Sμ-Sα and Sμ-Sυ2. (A) Schematic diagram of DNA fragment deletion and inversion during IgA and IgY2 CSR. (B) PCR amplification of recombined switch fragments. (C) Three representative junction structures of the recombined switch fragments. For each alignment, the sequence of the recombined fragment is shown in the middle; the germline sequences of the two S regions that are involved are shown above and below. The arrows indicate break points, whereas boxed nucleotides denote microhomology shared by two S regions. The inserted nucleotide in the junction site is underlined.

Close modal
Table I.
Summary of the junctional patterns in the recombined fragments of S regions
JunctionDirect End JoiningPerfectly Matched Short HomologyInsertionsNo. of Fragments
1–3 bp4–6 bp7–9 bp1 bp2–3 bp>4 bp
Sμ-Sα (deletion) 10 (21%) 26 (55%) 0 (0%) 0 (0%) 3 (6%) 5 (11%) 3 (6%) 47 
Sμ−Sα (inversion) 6 (20%) 16 (53%) 3 (10%) 0 (0%) 0 (0%) 5 (17%) 0 (0%) 30 
Sμ-Sυ2 (deletion) 10 (21%) 26 (54%) 3 (6%) 0 (0%) 3 (6%) 4 (8%) 2 (4%) 48 
Sμ-Sυ2 (inversion) 3 (30%) 4 (36%) 0 (0%) 0 (0%) 1 (9%) 2 (18%) 1 (9%) 11 
Total junctions 29 (21%) 72 (53%) 6 (4%) 0 (0%) 7 (5%) 16 (12%) 6 (4%) 136 
JunctionDirect End JoiningPerfectly Matched Short HomologyInsertionsNo. of Fragments
1–3 bp4–6 bp7–9 bp1 bp2–3 bp>4 bp
Sμ-Sα (deletion) 10 (21%) 26 (55%) 0 (0%) 0 (0%) 3 (6%) 5 (11%) 3 (6%) 47 
Sμ−Sα (inversion) 6 (20%) 16 (53%) 3 (10%) 0 (0%) 0 (0%) 5 (17%) 0 (0%) 30 
Sμ-Sυ2 (deletion) 10 (21%) 26 (54%) 3 (6%) 0 (0%) 3 (6%) 4 (8%) 2 (4%) 48 
Sμ-Sυ2 (inversion) 3 (30%) 4 (36%) 0 (0%) 0 (0%) 1 (9%) 2 (18%) 1 (9%) 11 
Total junctions 29 (21%) 72 (53%) 6 (4%) 0 (0%) 7 (5%) 16 (12%) 6 (4%) 136 

In humans and mice, S-S recombination occurs through the nonhomologous end-joining pathway, which produces three junction structures: blunt-ended junctions, junctions with insertions, or microhomology junctions (3, 33). Comparison of the junction segment break points with reference sequences also indicated three nonhomologous end-joining junction structures (Fig. 5C) and revealed a similar junction pattern for both IgA and IgY2 CSR. Approximately 21% of the junctions exhibited direct joining of S-S regions, and 17% displayed 1–3 bp insertions. However, approximately one half of the junction sites appeared to have resulted from microhomology-mediated ligation, with 1–3 bp accounting for a major proportion (Table I). Furthermore, this junction pattern in pigeon was observed in recombination derived not only from DNA deletion but also from DNA inversion (Table I).

It is known that CSR is preceded by germline transcription, which initiates from the germline promoter and proceeds through the S region. Along with transcription, an intervening exon (I exon) located in the 5′ segment of each S region is spliced to CH exons; the S region is then removed to generate mature germline transcripts (GLTs) (3). As we demonstrated that IgA is expressed through CSR, an Iα exon is expected to be located upstream of the Sα region and spliced to Cα exons to generate mature GLTs. Although a series of primers based on the 5′-Sα flanking region and Cα2 exon were designed, no mature transcript was amplified. We therefore hypothesized that the Sα region may not be spliced in mature GLTs. To test this hypothesis, sense primers for the 3′ segment of the Sα region were designed and used in RT-PCR together with antisense primers corresponding to the Cα2 exon. Using splenic cDNA with genomic DNA removed as the template, we amplified a ∼1.5-kb DNA fragment containing the 3′-Sα region and the first two Cα exons (Fig. 6). Sequencing of this DNA fragment showed that a 700-bp intron between Cα1 and Cα2 was spliced, suggesting that this DNA fragment corresponds to mature GLTs. In addition, several pairs of primers based on the 5′ segment of the Sα region were designed, and only a short, 300-bp transcript located within proximity of the 5′-Sα region was detected by RT-PCR, whereas no transcripts were amplified using primers complementary to the region upstream of the 300-bp fragment. Primers based on the 5′-Sα region and the Cα2 exon were subsequently used to detect the complete transcripts, although no obvious fragment was amplified, which is most likely due to the long, repetitive sequences. These findings suggest that in pigeon, the Iα promoter is very likely located within the 300-bp fragment and that the intronic Sα region is not spliced in mature GLTs of the α gene.

FIGURE 6.

PCR amplification of GLTs of the α gene. M, 1-kb DNA marker; NC, negative control; 1, amplification using cDNA as the template; 2, amplification using genomic DNA as the template.

FIGURE 6.

PCR amplification of GLTs of the α gene. M, 1-kb DNA marker; NC, negative control; 1, amplification using cDNA as the template; 2, amplification using genomic DNA as the template.

Close modal

Considering that only one potential functional VH segment was found within the pigeon IgH locus, one pair of primers based on the leader peptide of the VH1 and CHs of the μ gene were synthesized and used to amplify rearranged VDJ fragments. Approximately 600-bp PCR products were cloned and sequenced. A total of 35 VH cDNA clones were identified using IMGT/V-QUEST on the IMGT Web site, 29 of which were unique sequences. The CDR3 plays an important role in recognizing and binding to Ags, and analysis of the CDR3 length in these 29 VH clones showed variation from 12 to 19 aa, with lengths of 15 aa (24.1%), 16 aa (27.6%), and 17 aa (31.0%) being the most common (Supplemental Fig. 1B). The JH segments of these 29 unique sequences were all assigned to the single germline JH segment. Additionally, small nucleotide variations were observed in some clones (with most being cytosine to thymine mutations), which may be an effect of somatic hypermutation.

In the current study, we performed a detailed analysis of IgH genes in pigeon, revealing that this species possesses four CH genes organized in the order μ-α-υ1-υ2. In this article, we provide a thorough analysis of CSR in birds and also demonstrate that divergence of the two identified pigeon υ genes occurred at a very early stage of bird evolution. Additionally, we show that the pigeon α gene has the same transcriptional orientation as the remaining genes. To our knowledge, this is the first report to show that the IgH gene locus in birds possesses a transcriptionally forward α gene.

Previous studies have reported that the α gene in some birds (ostrich, chicken, and duck) and crocodilians is inverted with respect to other CH genes, indicating that α gene inversion occurred before the divergence of crocodilians and avians (6, 7, 15, 18). As pigeon (C. livia) belongs to Neoaves, the most diverse avian clade (19), it is possible that the IgH locus in Neoaves or only in pigeon may have undergone genetic changes that resulted in a forward α gene. In addition to the change in transcriptional orientation, the genomic size of the α gene appears to be constrained in pigeon (at ∼6.4 kb) compared with its counterpart in duck (∼10.9 kb) and chicken (∼11.5 kb) (6, 7).

Because the α gene is inverted in chicken and duck, expression of IgA is apparently achieved by inversion of the α gene rather than the usual intrachromosomal deletion. However, a forward α gene was found at the pigeon IgH locus, possibly facilitating production of IgA via direct DNA deletional recombination between the Sμ and Sα regions. In fact, most chromosomal DSB ends join other DSBs without orientation specificity (34), indicating that both DNA deletional and inversional recombination can result from CSR. Moreover, CSR junctions in two orientations were found in our study, which suggests that transcriptional orientation has no effect on expression of the α gene in birds. Analysis of recombined fragment junctions revealed a similar pattern for deletional and inversional recombination generated by IgA and IgY2 CSR. Three junction structures, including blunt-ended junctions, junctional microhomology, and junctions with short insertions, were all detected, with junctional microhomology accounting for a large proportion (Table I). Although CSR junctions theoretically occur in two orientations with equal frequency, a recent study showed that CSR is programmed to occur biased heavily toward a deletional orientation in murine B cells (35). Future comparative studies on CSR of IgA in pigeon and other birds, such as chicken and duck, may provide more insight into the molecular mechanisms of CSR.

Another interesting finding in our study is the special germline transcription of the α gene. In mouse IgH genes, primary transcript splicing removes the intronic S region and joins the I exon to CH exons during germline transcription (36), although no obvious transcript of this splicing type was observed in our study. Nevertheless, germline transcription corresponding to the 3′-Sα region and Cα exons was detected, and this transcript was found to be mature. Moreover, we also detected a 300-bp transcript close to the 5′-Sα region. These findings suggest that only the intron between the Cα exons was removed but that the Sα region was retained in mature GLTs of the pigeon α gene. To our knowledge, this is the first report of this pattern of germline transcription. Although it is not clear why this pattern of germline transcription occurred and I exon was not observed, our results suggest that germline transcription rather than RNA processing is more essential for CSR initiation.

The present study also generated important findings regarding pigeon υ genes. Similar to ostrich, the υ gene in pigeon has diverged to produce two subclass-encoding genes, and phylogenetic analysis suggested that this subclass diversification occurred before the divergence of modern birds. As shown in the phylogenetic tree, pigeon IgY1 is evolutionarily closer to ostrich IgY2, whereas pigeon IgY2 is more homologous to ostrich IgY1 (Fig. 2). It is therefore reasonable to speculate that the ostrich υ1 gene is very likely located downstream of the υ2 gene and that the IgH gene locus in ostrich should be organized in the order μ1-δ-α-μ2-υ2-υ1. This configuration may represent the genomic structure of the IgH gene locus in the common ancestor of modern birds. Based on this genomic structure, δ, μ2, and υ1 genes have been lost in Galloanserae, whereas δ, μ2, and υ2 have been lost in other birds, such as zebra finch and penguin (37). At the pigeon IgH gene locus, the δ and μ2 genes have been lost, and the transcriptional orientation of the α gene was changed to the forward pattern (Fig. 4).

Despite a series of genetic events, including deletion and inversion in the CH gene locus, during evolution from Paleognathae to Neoaves, it appears that the VH gene locus has sustained fewer genetic alterations. Recent studies on ostrich, goose, zebra finch, and penguin have revealed only a single leader peptide in the expressed VH repertoire, indicating that gene conversion plays a major role in the generation of Abs in these species (3739). Similarly, we found only one potentially functional VH segment present within the pigeon IgH locus. Thus, it is very likely that the same mechanism is used to generate diversity of the pigeon VH. These data suggest that the distinct features of bird VH genes appeared very early during the divergence of avian species and are very likely shared by all avians.

In conclusion, a transcriptionally forward α gene and two subclasses of υ genes were identified in pigeon, which challenges the commonly held notion that the α gene is inverted in birds. Additionally, to our knowledge, CSR junctions in birds are described in this article for the first time, and similar junction patterns were found in both deletional and inversional recombination in pigeon IgA and IgY CSR. Our findings improve our understanding of CSR in birds and provide new insight into the evolutionary flexibility and diversity of Ig genes in tetrapods.

This work was supported by the National Natural Science Foundation of China (31472085 and 31771647) and open projects of State Key Laboratory of Agrobiotechnology (2018SKLAB6-6).

The sequences presented in this article have been submitted to the National Center for Biotechnology Information's GenBank (https://www.ncbi.nlm.nih.gov/genbank/) under accession numbers MG762738, MG762739, MG762740, MG762741, and MG762742.

The online version of this article contains supplemental material.

Abbreviations used in this article:

CH

H chain C region

CSR

class-switch recombination

DH

H chain D region

DSB

double-strand break

GLT

germline transcript

I exon

intervening exon

IMGT

ImMunoGeneTics

JH

H chain J region

qRT-PCR

quantitative RT-PCR

RSS

recombination signal sequence

S region

switch region

VH

H chain V region.

1
Flajnik
,
M. F.
,
M.
Kasahara
.
2010
.
Origin and evolution of the adaptive immune system: genetic events and selective pressures.
Nat. Rev. Genet.
11
:
47
59
.
2
Flajnik
,
M. F.
2002
.
Comparative analyses of immunoglobulin genes: surprises and portents.
Nat. Rev. Immunol.
2
:
688
698
.
3
Stavnezer
,
J.
,
J. E.
Guikema
,
C. E.
Schrader
.
2008
.
Mechanism and regulation of class switch recombination.
Annu. Rev. Immunol.
26
:
261
292
.
4
McCormack
,
W. T.
,
L. W.
Tjoelker
,
C. B.
Thompson
.
1991
.
Avian B-cell development: generation of an immunoglobulin repertoire by gene conversion.
Annu. Rev. Immunol.
9
:
219
241
.
5
Magor
,
K. E.
,
D. A.
Higgins
,
D. L.
Middleton
,
G. W.
Warr
.
1999
.
Opposite orientation of the alpha- and upsilon-chain constant region genes in the immunoglobulin heavy chain locus of the duck.
Immunogenetics
49
:
692
695
.
6
Lundqvist
,
M. L.
,
D. L.
Middleton
,
S.
Hazard
,
G. W.
Warr
.
2001
.
The immunoglobulin heavy chain locus of the duck. Genomic organization and expression of D, J, and C region genes.
J. Biol. Chem.
276
:
46729
46736
.
7
Zhao
,
Y.
,
H.
Rabbani
,
A.
Shimizu
,
L.
Hammarström
.
2000
.
Mapping of the chicken immunoglobulin heavy-chain constant region gene locus reveals an inverted alpha gene upstream of a condensed upsilon gene.
Immunology
101
:
348
353
.
8
Magor
,
K. E.
,
D. A.
Higgins
,
D. L.
Middleton
,
G. W.
Warr
.
1994
.
cDNA sequence and organization of the immunoglobulin light chain gene of the duck, Anas platyrhynchos.
Dev. Comp. Immunol.
18
:
523
531
.
9
Reynaud
,
C. A.
,
V.
Anquez
,
A.
Dahan
,
J. C.
Weill
.
1985
.
A single rearrangement event generates most of the chicken immunoglobulin light chain diversity.
Cell
40
:
283
291
.
10
Reynaud
,
C. A.
,
A.
Dahan
,
V.
Anquez
,
J. C.
Weill
.
1989
.
Somatic hyperconversion diversifies the single Vh gene of the chicken with a high incidence in the D region.
Cell
59
:
171
183
.
11
Das
,
S.
,
U.
Mohamedy
,
M.
Hirano
,
M.
Nei
,
N.
Nikolaidis
.
2010
.
Analysis of the immunoglobulin light chain genes in zebra finch: evolutionary implications.
Mol. Biol. Evol.
27
:
113
120
.
12
Bao
,
Y.
,
S.
Wu
,
Y.
Zang
,
H.
Wang
,
X.
Song
,
C.
Xu
,
B.
Xie
,
Y.
Guo
.
2012
.
The immunoglobulin light chain locus of the turkey, Meleagris gallopavo.
Vet. Immunol. Immunopathol.
147
:
44
50
.
13
Sun
,
Y.
,
Z.
Wei
,
N.
Li
,
Y.
Zhao
.
2013
.
A comparative overview of immunoglobulin genes and the generation of their diversity in tetrapods.
Dev. Comp. Immunol.
39
:
103
109
.
14
Ohta
,
Y.
,
M.
Flajnik
.
2006
.
IgD, like IgM, is a primordial immunoglobulin class perpetuated in most jawed vertebrates.
Proc. Natl. Acad. Sci. USA
103
:
10723
10728
.
15
Han
,
B.
,
H.
Yuan
,
T.
Wang
,
B.
Li
,
L.
Ma
,
S.
Yu
,
T.
Huang
,
Y.
Li
,
D.
Fang
,
X.
Chen
, et al
.
2016
.
Multiple IgH isotypes including IgD, subclasses of IgM, and IgY are expressed in the common ancestors of modern birds.
J. Immunol.
196
:
5138
5147
.
16
Lundqvist
,
M. L.
,
D. L.
Middleton
,
C.
Radford
,
G. W.
Warr
,
K. E.
Magor
.
2006
.
Immunoglobulins of the non-galliform birds: antibody expression and repertoire in the duck.
Dev. Comp. Immunol.
30
:
93
100
.
17
Magor
,
K. E.
,
G. W.
Warr
,
Y.
Bando
,
D. L.
Middleton
,
D. A.
Higgins
.
1998
.
Secretory immune system of the duck (Anas platyrhynchos). Identification and expression of the genes encoding IgA and IgM heavy chains.
Eur. J. Immunol.
28
:
1063
1068
.
18
Cheng
,
G.
,
Y.
Gao
,
T.
Wang
,
Y.
Sun
,
Z.
Wei
,
L.
Li
,
L.
Ren
,
Y.
Guo
,
X.
Hu
,
Y.
Lu
, et al
.
2013
.
Extensive diversification of IgH subclass-encoding genes and IgM subclass switching in crocodilians.
Nat. Commun.
4
:
1337
.
19
Jarvis
,
E. D.
,
S.
Mirarab
,
A. J.
Aberer
,
B.
Li
,
P.
Houde
,
C.
Li
,
S. Y.
Ho
,
B. C.
Faircloth
,
B.
Nabholz
,
J. T.
Howard
, et al
.
2014
.
Whole-genome analyses resolve early branches in the tree of life of modern birds.
Science
346
:
1320
1331
.
20
Zhang
,
G.
,
C.
Li
,
Q.
Li
,
B.
Li
,
D. M.
Larkin
,
C.
Lee
,
J. F.
Storz
,
A.
Antunes
,
M. J.
Greenwold
,
R. W.
Meredith
, et al
Avian Genome Consortium
.
2014
.
Comparative genomics reveals insights into avian genome evolution and adaptation.
Science
346
:
1311
1320
.
21
Lefranc
,
M. P.
,
C.
Pommié
,
M.
Ruiz
,
V.
Giudicelli
,
E.
Foulquier
,
L.
Truong
,
V.
Thouvenin-Contet
,
G.
Lefranc
.
2003
.
IMGT unique numbering for immunoglobulin and T cell receptor variable domains and Ig superfamily V-like domains.
Dev. Comp. Immunol.
27
:
55
77
.
22
Lefranc
,
M. P.
1998
.
IMGT (ImMunoGeneTics) locus on focus. A new section of Experimental and Clinical Immunogenetics.
Exp. Clin. Immunogenet.
15
:
1
7
.
23
Ronquist
,
F.
,
J. P.
Huelsenbeck
.
2003
.
MrBayes 3: Bayesian phylogenetic inference under mixed models.
Bioinformatics
19
:
1572
1574
.
24
Campbell
,
K. S.
,
B. T.
Bäckström
,
G.
Tiefenthaler
,
E.
Palmer
.
1994
.
CART: a conserved antigen receptor transmembrane motif.
Semin. Immunol.
6
:
393
410
.
25
Woof
,
J. M.
,
M. A.
Kerr
.
2006
.
The function of immunoglobulin A in immunity.
J. Pathol.
208
:
270
282
.
26
Dunnick
,
W.
,
G. Z.
Hertz
,
L.
Scappino
,
C.
Gritzmacher
.
1993
.
DNA sequences at immunoglobulin switch region recombination sites.
Nucleic Acids Res.
21
:
365
372
.
27
Zarrin
,
A. A.
,
F. W.
Alt
,
J.
Chaudhuri
,
N.
Stokes
,
D.
Kaushal
,
L.
Du Pasquier
,
M.
Tian
.
2004
.
An evolutionarily conserved target motif for immunoglobulin class-switch recombination.
Nat. Immunol.
5
:
1275
1281
.
28
Mills
,
F. C.
,
J. S.
Brooker
,
R. D.
Camerini-Otero
.
1990
.
Sequences of human immunoglobulin switch regions: implications for recombination and transcription.
Nucleic Acids Res.
18
:
7305
7316
.
29
Nikaido
,
T.
,
Y.
Yamawaki-Kataoka
,
T.
Honjo
.
1982
.
Nucleotide sequences of switch regions of immunoglobulin C epsilon and C gamma genes and their comparison.
J. Biol. Chem.
257
:
7322
7329
.
30
Yu
,
K.
,
F.
Chedin
,
C. L.
Hsieh
,
T. E.
Wilson
,
M. R.
Lieber
.
2003
.
R-loops at immunoglobulin class switch regions in the chromosomes of stimulated B cells.
Nat. Immunol.
4
:
442
451
.
31
Zhang
,
Z. Z.
,
N. R.
Pannunzio
,
C. L.
Hsieh
,
K.
Yu
,
M. R.
Lieber
.
2014
.
The role of G-density in switch region repeats for immunoglobulin class switch recombination.
Nucleic Acids Res.
42
:
13186
13193
.
32
Muramatsu
,
M.
,
K.
Kinoshita
,
S.
Fagarasan
,
S.
Yamada
,
Y.
Shinkai
,
T.
Honjo
.
2000
.
Class switch recombination and hypermutation require activation-induced cytidine deaminase (AID), a potential RNA editing enzyme.
Cell
102
:
553
563
.
33
Pan-Hammarström
,
Q.
,
Y.
Zhao
,
L.
Hammarström
.
2007
.
Class switch recombination: a comparison between mouse and human.
Adv. Immunol.
93
:
1
61
.
34
Chiarle
,
R.
,
Y.
Zhang
,
R. L.
Frock
,
S. M.
Lewis
,
B.
Molinie
,
Y. J.
Ho
,
D. R.
Myers
,
V. W.
Choi
,
M.
Compagno
,
D. J.
Malkin
, et al
.
2011
.
Genome-wide translocation sequencing reveals mechanisms of chromosome breaks and rearrangements in B cells.
Cell
147
:
107
119
.
35
Dong
,
J.
,
R. A.
Panchakshari
,
T.
Zhang
,
Y.
Zhang
,
J.
Hu
,
S. A.
Volpi
,
R. M.
Meyers
,
Y. J.
Ho
,
Z.
Du
,
D. F.
Robbiani
, et al
.
2015
.
Orientation-specific joining of AID-initiated DNA breaks promotes antibody class switching.
Nature
525
:
134
139
.
36
Chaudhuri
,
J.
,
F. W.
Alt
.
2004
.
Class-switch recombination: interplay of transcription, DNA deamination and DNA repair. [Published erratum appears in 2004 Nat. Rev. Immunol. 4: 655.]
Nat. Rev. Immunol.
4
:
541
552
.
37
Han
,
B.
,
Y.
Li
,
H.
Han
,
Y.
Zhao
,
Q.
Pan
,
L.
Ren
.
2017
.
Three IgH isotypes, IgM, IgA and IgY are expressed in Gentoo penguin and zebra finch.
PLoS One
12
:
e0173334
.
38
Huang
,
T.
,
M.
Zhang
,
Z.
Wei
,
P.
Wang
,
Y.
Sun
,
X.
Hu
,
L.
Ren
,
Q.
Meng
,
R.
Zhang
,
Y.
Guo
, et al
.
2012
.
Analysis of immunoglobulin transcripts in the ostrich Struthio camelus, a primitive avian species.
PLoS One
7
:
e34346
.
39
Huang
,
T.
,
K.
Wu
,
X.
Yuan
,
S.
Shao
,
W.
Wang
,
S.
Wei
,
G.
Cao
.
2016
.
Molecular analysis of the immunoglobulin genes in goose.
Dev. Comp. Immunol.
60
:
160
166
.

The authors have no financial conflicts of interest.

Supplementary data