The JI
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Zhao, Y.
Right arrow Articles by Hammarström, L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Zhao, Y.
Right arrow Articles by Hammarström, L.
Right arrowPubmed/NCBI databases
*Protein
The Journal of Immunology, 2003, 171: 1312-1318.
Copyright © 2003 by The American Association of Immunologists

The Porcine Ig {delta} Gene: Unique Chimeric Splicing of the First Constant Region Domain in its Heavy Chain Transcripts 1 ,2

Yaofeng Zhao3,*, Qiang Pan-Hammarström*, Imre Kacskovics{dagger} and Lennart Hammarström*

* Center for Biotechnology, Department of Biosciences at Novum, Karolinska Institutet, Huddinge, Sweden; and {dagger} Department of Physiology and Biochemistry, Faculty of Veterinary Science, Szent István University, Budapest, Hungary


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
The pig {delta} gene is located ~3.4 kb downstream of the second transmembrane exon of the µ gene and shows a similar genomic structure to its counterpart in cow with three exons encoding the CH1, CH2, and CH3 domains. The porcine genomic {delta}CH1 exon has been replaced by a recent duplication of the µCH1 and its flanking sequences, a genetic event that also led to the formation of a short switch {delta} region, immediately upstream of the {delta} gene. The {delta}CH1 exhibits a 98.7% similarity (314 of 318 bp) to the µCH1 at the DNA level, whereas the homologies between the {delta}CH2 and µCH3, and the {delta}CH3 and µCH4 are only 33.3 and 35.8%, respectively. Either of the two CH1 exons (µ and {delta}) could be observed in the expressed porcine IgD H chain cDNA sequences VDJ-µCH1-H-{delta}CH2-{delta}CH3 or VDJ-{delta}CH1-H-{delta}CH2-{delta}CH3, showing a pattern that has not been observed previously in vertebrates. In addition, transfection of a human B cell line, using artificial constructs resembling the porcine Cµ-C{delta} locus, also generated both VDJ-µCH1-{delta}CH1-H1-{delta}CH2 and VDJ -{delta}CH1-H1-{delta}CH2 transcripts. An examination of the pig {delta} genomic sequence shows a putative, second hinge region-encoding exon. Due to the lack of a normal branchpoint sequence for RNA splicing, this exon is not present in the normal pig {delta} cDNA. However, the exon could be spliced into most of the expressed transcripts in vitro in cell transfection experiments after introduction of a single T nucleotide to restore the branchpoint sequence upstream of the putative H2 exon.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
The recent discovery of a {delta} gene in artiodactyls clearly demonstrates the presence of IgD in mammalian species other than primates and rodents (1). In cows and sheep, the {delta}CH1 is highly homologous (~96%) to its respective µCH1 sequences due to a duplication of the µCH1 and its upstream sequence ~20 million years ago (1, 2). The duplicated segment was introduced into the {delta} gene, replacing the pre-existing {delta}CH1, resulting in a similar structure of the H chain of both IgM and IgD in these species (1). The duplication also involved the Sµ region and led to the formation of a short, but functional, S{delta} region, which can direct IgD class switching (1).

In the channel catfish (Ictalurus punctatus), three C{delta} genes, each linked to a µ gene or a pseudogene, have been identified. Two of them are used to generate membrane ({delta}m) and secreted ({delta}s) transcripts, respectively (3, 4). This is highly unusual, because the cDNA encoding both the membrane and secreted forms of the Ig H chain originate from the transcription of a single constant region gene through alternative RNA splicing in all known mammalian {delta} genes (5, 6, 7, 8).

In our previous publication, we showed that the first H chain constant region domain of the pig IgD was identical with the µCH1 in one animal, whereas it was unique ({delta}CH1) in another one (1). This finding suggested that, in some individuals, the µCH1 is spliced onto the {delta} sequence to produce a chimeric IgD H chain, thus showing a splicing mechanism that is similar, but not identical with, that found in lower vertebrates such as channel catfish, cod, and salmon (4, 9, 10). Therefore, it is likely that several mechanisms are involved in generating porcine IgD. The first possibility could be that, as in humans and rodents (11), the pig {delta} gene is cotranscribed together with the µ gene to yield a long primary transcript, from which a unique {delta} mRNA is generated by RNA splicing. Alternatively, posttranscriptional processing of the long primary transcript fuses the µCH1 to the {delta} sequence, resulting in a chimeric mRNA.

The second major point is that the deduced protein sequence of the porcine IgD H chain displays a shorter hinge region than those of cow, sheep, and human that are usually encoded by two exons (1, 8). Therefore, an examination of the genomic sequence of the porcine {delta} gene may help to understand the underlying reason for hinge region differences in these species. To address these questions, we have analyzed the porcine {delta} gene and its transcription in detail.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Screening for a pig C{delta} gene-positive P1-derived artificial chromosome (PAC) 4 clone

The primary and secondary DNA pools made by the Resource Center (Berlin, Germany) primary database from a pig PAC library constructed using the vector pCYPAC2 (12), were screened by PCR using the porcine {delta}CH3-derived primers: Pig-DCH3s, 5'-ACC CGG CCC GCT CGC TTT CA-3'; and Pig-DCH3as, 5'-TCG TGC CCG ACC ACA CAG GT-3'; under the following conditions: 94°C for 3 min and then 40 cycles of 94°C for 30 s, 62°C for 30 s, and 72°C for 30 s, holding at 4°C. The primers were based on the sequences described in our previous publication (1). The obtained PAC clone (TAIGP714G04138Q4) was cultured in kanamycin-containing Luria-Bertani medium, and the PAC DNA was prepared using the Plasmid Maxi kit (Qiagen, Valencia, CA).

Pulsed field gel electrophoresis (PFGE) and Southern blotting

PFGE was used to separate large DNA fragments (CHEF-DR, III system; Bio-Rad, Hercules, CA). The digested PAC DNA was run in a 1% agarose gel in 0.5x Tris-borate-EDTA buffer (6V/cm). The switch time was adjusted based on the size of the DNA, according to the manufacturer’s instruction. The separated DNA was transferred to nylon membranes for hybridization. An oligolabeling kit (Pharmacia Biotech, Uppsala, Sweden) was used to label the probe. Hybridizations were performed with ExpressHyb hybridization solution purchased from Clontech (Palo Alto, CA).

Long PCR amplifications of genomic DNA fragments and sequencing

A long PCR kit (Expand Long Template PCR System kit; Roche Diagnostics, Bromma, Sweden) was used to amplify the genomic DNA fragments from the isolated PAC clone, and these fragments were subsequently cloned into a T vector. While a ~7 kb fragment covering the region from the µ transmembrane (TM) to {delta}CH3 was amplified using primers Pig-IgM TMs (5'-CCG TCA CGC TGT TCA AGG TAG GC-3') (13) and Pig-IgD CH3as1 (5'-CAG GAG GCG GCC TTG GTG AAA GC-3'), the {delta}CH3 to {delta}TM-containing fragment was obtained using primers Pig-IgDCH3Ls (5'-CCG GGC CAT CAG GAC GCC ACC TAC A-3') and Pig-IgDTMLas (5'-CTG CTC TCC GGG GCT ACT TCA CCT G-3'). The PCR error rate of the DNA polymerase used was estimated to be 2/10.000 (14). For sequencing of the long DNA insert, an Erase-a-Base System kit (Promega, Madison, WI) was used to make a series of subclones possessing overlapping deletions.

RNA isolations and RT-PCR

Total RNA was extracted from pig peripheral blood and spleen using TRIzol (Life Technologies, Gaithersburg, MD) following the manufacturer’s instructions. Approximately 5 µg of total RNA was used to synthesize first-strand cDNA with a First-Strand cDNA synthesis kit (Amersham Biosciences, Uppsala, Sweden). The µCH1-containing cDNA fragments were directly amplified by RT-PCR using primers Pig-IgJH (5'-GGC GTT GAA GTC GTC GTG TCC T-3') and Pig-IgMCH2as (5'-GGG GGC TGC TCC TCT ACG ACT G-3'), whereas the {delta}CH1-containing cDNA fragments were generated by a nested PCR using two primer pairs: Swine-JH (5'-CCA GGC GTT GAA GTC GTC GTG T-3') and Pig-IgDCH3as (5'-GAG GAT GTC CAG GGG CGA GAA GC-3'); and Pig-Ig JH (the sequence is indicated above) and Pig-IgDHas1 (5'-TGG TGG AGA CCG TTG ACA GGG TTG G-3'). All of the PCR amplifications were conducted under conditions of 94°C for 3 min, and then 30 cycles of 94°C for 30 s, 64°C for 30 s, 72°C for 30 s, with final extension at 72°C for 7 min. The PCR products were cloned into a T vector and sequenced using T7 and Sp6 primers.

Cloning and sequencing of both the µ and {delta} CH1 genomic sequences

Pig genomic DNA samples were isolated from the spleen using a phenol extraction-based approach. The µCH1-containing genomic fragments were directly amplified using primers Pig-IgMDs (5'-CGC CTC TCA CTC CCC TTT CTC T-3') and Pig-IgMCH2as (the sequence is indicated above). The resulting PCR products were cloned into a T vector and sequenced using T7 or Sp6 primers. The {delta}CH1-containing genomic fragments were amplified using a long PCR kit (Expand Long Template PCR System kit; Roche Diagnostics), using the primers Pig-IgM TMs (5'-CCG TCA CGC TGT TCA AGG TAG GC-3') and Pig-IgDHas (5'-GTG GTG AGG GCT TTG GTG GAG AC-3'), and were directly sequenced.

Construction of plasmids

A mammalian expression vector, pCI-Neo (Promega), was used to make the constructs pDEG and pDEA. To obtain a rearranged pig VDJ sequence fused to the genomic sequence immediately downstream of the porcine JH-encoding exon, a pig VDJ fragment was first amplified from a previously cloned IgD H chain full-length-encoding cDNA (1), while a genomic fragment covering the JH exon and its downstream sequence was generated by amplification of porcine genomic DNA (15). A fusion PCR was subsequently conducted using the mixed PCR products resulting from the above two amplifications as a template to generate a fused DNA fragment containing the VDJ and 5' splicing sites of the JH-Cµ intron. The primers used were NheI-Vs (5'-TTT TGC TAG CAC TGG GTG GTC TTG TTT GCT-3'), Pig-Vas (5'-GAC ACG ACG ACT TCA ACG CC-3'), Pig-JHgs (5'-GGG CCC AGG CGT TGA AGT CG-3'), and EcoRI-JHgas (5'-TTT TGA ATT CTC CGC AGA CAC CCT CCT CAA-3'). The bold and underlined letters indicate designed restriction sites. The µCH1- and {delta}CH1-containing genomic fragments with both 5' and 3' splicing sites were amplified using primers EcoRI-pigCu1s (5'-TTT TGA ATT CTG CAT CAG ACT CGC CAG ACC-3') and KpnI-pigCu1as (5'-TTT TGG TAC CCC CCC GCC ACA GCC ACA G-3'), and KpnI-pigCd1s (5'-TTT TGG TAC CAG CTG CCT TGG GAT GGT TC-3') and SalI-pigCd1as (5'-TTT TGT CGA CCA GGA TGG GCC AGT GGG GTC-3'), respectively. All of the above amplifications were conducted using a proofreading DNA polymerase (Vent; New England Biolabs, Beverly, MA). A hinge region and {delta}CH2-spanning genomic fragment was obtained using long PCR kit and 38Q4 PAC DNA as a template, and the primers SalI-pigH-Cd2s (5'-TTT TGT CGA CGC CGC CGA TGG ACA GCC CGC-3') and NotI-pigH-Cd2as (5'-TTT TGC GGC CGC CAG GAG CCG TTG GTA TGT TC-3'). All of the resulting PCR products were first cloned into a T vector. Except for the µCH1- and {delta}CH1-containing fragments, the other two fragments were directly cloned into pCI-Neo after being released from the T vector, using the appropriate enzymes. The µCH1 and {delta}CH1 fragments were first cocloned into pGEM-3zf(+) and subsequently cloned into pCI-Neo at the EcoRI and SalI sites. After transfection of a human B cell line (BL2), RT-PCR was conducted using primers PDE-S (5'-GTG ACA GTG TTT ACG ATG GT-3') and PDE-As (5'-GGG GGA GGA GCA GGT AGA CG-3') to detect the processed RNA transcripts. The resulting PCR products were cloned into a T vector and sequenced.

Mutagenesis

Introduction of a single T into the branchpoint site of the putative hinge exon in the artificial construct was conducted by using a Quikchange XL site-directed mutagenesis kit (Stratagene, Cedar Creek, TX). The mutant plasmid was termed pDE-T. The primers used in the experiment were Pig-Hinge-Muts (5'-CAG GGG CAG AAC CCT GAC CCA CCC CCG TT-3') and Pig-Hinge-Mutas (5'-AAC GGG GGT GGG TCA GGG TTC TGC CCC TG-3').

Cell culture and transfections

A human B cell line, BL2, was used in the transfection experiments. The protocol used for cell culture and transfection has been described elsewhere (16). The cells were collected 48 h after transfection, and total RNA was extracted using a RNeasy Mini kit (Qiagen, Valencia, CA). Approximately 5 µg of RNA was used to make first-strand cDNA using a gene-specific primer PDE-Fs (5'-GCC AGG CAG GTG AAT GTA GC-3') and a First-Strand cDNA synthesis kit (Amersham Biosciences).

RT-PCR amplification of the pig IgD hinge region

To check the splicing patterns of the hinge regions in animals or human B cells, we used a nested PCR to ensure the sensitivity of detection. The primers used in the amplification of pig spleen RNA were Pig-IgJH (5'-GGC GTT GAA GTC GTC GTG TCC T-3') and Pig-IgDCH3as (5'-GAG GAT GTC CAG GGG CGA GAA GC-3'), and Pig-Hinge1s (5'-ACC CCT GGG CCA ACC CTG TCA A-3') and Pig-IgDCH2as (5'-GTA GCC TCA GCC CGC AGC CAC AG-3'), whereas the primers used in the amplification of RNA isolated from the transfected cells were Pig-IgJH and Pig-IgDCH2as, and Pig-Hinge1s and PDE-As.

Computational analysis

DNA sequence homology searches were conducted using the National Center for Biotechnology Information BLAST program. Sequence alignment and comparison was conducted using the MegAlign program (DNAstar, Madison, WI). Dot-plot comparison was performed using the same program with the following parameters: percentage, 80; window, 30; and minimum quality, 1.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Screening of porcine C{delta} gene-positive clone from a pig PAC library

Using the primers pig-IgDCH3s and pig-IgDCH3as, designed based on the sequence of the pig {delta}CH3 exon and expected to generate a 243-bp PCR product, one pig C{delta} gene-positive PAC clone, TAIGP714G04138Q4 (briefly termed 38Q4), was finally obtained from the Resource Center primary database after two rounds of PCR screening.

The NotI-digested 38Q4 PAC DNA was separated on a 1% agarose gel using PFGE and showed that the clone contained a ~150-kb insert. Southern blotting using a porcine C{delta} gene-specific probe confirmed that the C{delta} gene is indeed present in the insert of this PAC clone. Consistent with the PCR screening result using the µCH1-derived primers, a further analysis revealed that the clone lacked the pig µCH1-µCH4 sequence, but contained both the µTM1 and µTM2 exons (13). In addition, the clone was shown to be negative for both the pig {alpha} and {epsilon} genes (data not shown), indicating that the porcine Ig H chain constant region gene locus is >150 kb.

Analysis of genomic organization of porcine IgD H chain constant region gene

A ~7 kb genomic DNA fragment, spanning the pig Cµ-C{delta} intron and part of the C{delta} gene, was amplified from 38Q4 PAC DNA using primers pig-IgM TMs (13) and pig-IgD CH3as1, and cloned into the pGEM-T vector. The recombinant plasmid containing the fragment spanning the µTM1 to {delta}CH3 was named pPMD (Fig. 1A) and subsequently subcloned as three shorter fragments. Whereas the SalI-EcoRI and EcoRI-BamHI fragments were directly cloned into the pBluescript II KS (+), the BamHI-NcoI fragment was first cloned into the pGL3-Basic vector, and then the recut HindIII-BamHI fragment was cloned into the pBluescript II KS (+) vector. All three recombinant plasmids were fully sequenced and assembled together (accession number AY228505), showing that the fragment indeed spans the DNA sequence from the µTM1 to the {delta}CH3.



View larger version (20K):
[in this window]
[in a new window]
  		FIGURE 1. Physical mapping of the pig IgD H chain constant region gene. A, Sequence analysis of the two pig C{delta} gene containing genomic fragments amplified from the 38Q4 PAC clone. B, Genomic organization of the complete pig C{delta} gene. a, Switch {delta}; 1, µCH1; 2, µCH2; 3, µCH3; 4, µCH4; 5, µTM1; 6, µTM2; 7, {delta}CH1; 8, {delta}H1; 9, {delta}H2; 10, {delta}CH2; 11, {delta}CH3; 12, {delta}s; 13, {delta}TM1; and 14, {delta}TM2. The genomic structure of the pig Cµ gene is according to Ref. 13 .

 
To identify the transmembrane exons of the pig IgD, a ~4-kb DNA fragment, containing the {delta}CH3 and its downstream sequence, was amplified from the 38Q4 PAC clone, and the exon sequence encoding both the secreted and transmembrane tails (accession numbers AY228507–228508) were deduced by partially sequencing the cloned PCR product (clone pPDE; Fig. 1A).

Based on the above sequenced genomic fragments (Fig. 1A), the genomic organization of the porcine C{delta} gene was established by comparison of the genomic sequences with the cDNA sequences (B) (1), showing that the ~7 kb porcine C{delta} gene is located 3.4 kb downstream of the µTM2 exon. Like its human and bovine counterparts (1, 8), the porcine C{delta} gene has three exons encoding the CH1, CH2, and CH3 domains, a single exon for the secreted tail and two exons for the transmembrane tail (Fig. 2, A and B). The exon-intron boundaries of the gene are presented in Table I.



View larger version (42K):
[in this window]
[in a new window]
  		FIGURE 2. Sequence alignments of the pig C{delta} gene. A, {delta}s exon sequence. B, Sequence of the transmembrane exon. The sequences have been deposited into the National Center for Biotechnology Information GenBank under the accession numbers AY228507–228508.

 

View this table:
[in this window]
[in a new window]
  		Table I. Exon-intron boundaries of the pig Ig C{delta} gene

 
Dot-plot comparison of the porcine Cµ-C{delta} and JH-Cµ sequences

A dot-plot analysis of the porcine Cµ-C{delta} intron and the previously published JH-Cµ intron (13) showed long homologous DNA stretches in the two regions (Fig. 3), indicating that the {delta}CH1 and its flanking sequences were initially duplicated from the µ gene. The genomic {delta}CH1 exon is highly similar (98.7%) to the previously published µCH1 (13), as only 4-bp differences can be noted (Fig. 4), leading to 4-aa substitutions. Because the porcine JH-Cµ intron was only partially sequenced (13), it is difficult to determine the border of the duplicated fragment, although it can be deduced that the 3' end of the duplicated fragment extends into the intron between the µCH1 and µCH2. According to the comparison of the Cµ-C{delta} and JH-Cµ intron sequences, it is likely that the initial duplication included the Sµ region, because a short Sµ-like sequence could be identified upstream of the {delta} gene. However, the duplicated Sµ region has been reduced from ~3 to ~0.5 kb during evolution, but still maintains switch region-like repetitive sequences, including CTGGG (67 repeats) and CTGAG (20 repeats). The sequence might thus constitute a functional S{delta}, as in cows (1), and could theoretically direct class switching to the {delta} gene.



View larger version (37K):
[in this window]
[in a new window]
  		FIGURE 3. Dot-plot analysis of the porcine Cµ and C{delta} sequences. The sequence of the pig Cµ gene is according to Ref. 13 .

 


View larger version (39K):
[in this window]
[in a new window]
  		FIGURE 4. Comparison of the pig {delta}CH1 and its flanking sequences with those of the pig µCH1 (13 ). The short repeats are in bold and underlined.

 
Analysis of the splicing pattern of the expressed porcine IgD H chains

Our previous study on pig IgD H chain cDNAs derived from two animals generated inconsistent data with regard to the CH1 use (1). The cDNA sequences from one animal contained a sequence identical with the µCH1, whereas the cDNA sequences cloned from another showed only a 1-bp difference from the genomic {delta}CH1 derived from the 38Q4 PAC clone (1), raising the question of whether the porcine IgD H chain could be expressed as a chimeric structure. To address this point, both the IgD and IgM H chain-encoding cDNA fragments were amplified from spleen RNA of four animals (named SW1-4) by RT-PCR. In all cases, we could identify IgD H chain cDNA fragments that contained a sequence that was identical with the µCH1 (Table II). To exclude the possibility that, in some animals, the {delta}CH1 is identical with the µCH1 at the genomic level, both the genomic µCH1 and {delta}CH1 of two animals (SW1 and SW3) were also cloned and sequenced (Fig. 5), Together with the above and previous sequencing results, we found that there is only one single position (+3) that can be used to distinguish the {delta}CH1 (+3C) from the µCH1 (+3A). Fig. 5 shows the allelic polymorphism of both {delta}CH1 and µCH1 sequences identified in animals SW1 and SW3, indicating that the SW1 {delta} gene contains a unique {delta}CH1 sequence. Furthermore, there are two {delta} gene allotypes, which differ by 2 bp in their CH1 exons in SW3. Direct comparison of both the µ and {delta} genomic CH1 exon with the cloned {delta} cDNA sequences suggest that both the µ and {delta} germline CH1 exon could be included in the expressed porcine {delta} cDNA H chain, exhibiting a unique expression pattern that has not been identified in other vertebrates. In SW3, there are two germline {delta}CH1 sequences in which one allele contains +19G and 99C (termed G allele according to +19G), and the other contains +19A and +99T (termed A allele according to +19A). A detailed analysis of the 16 {delta} cDNA clones derived from SW3 showed that there were 4 µCH1, 4 G, and 8 A {delta}CH1 allele-containing clones, respectively, while the 14 {delta} cDNA clones from SW1 could be divided into 6 {delta}CH1- and 8 µCH1-containing sequences. It is noteworthy that at position +19 in the SW1 {delta}CH1 exon, the base is exclusively a G.


View this table:
[in this window]
[in a new window]
  		Table II. Inclusion of µCH1 in expressed porcine {delta} H chainsa

 


View larger version (35K):
[in this window]
[in a new window]
  		FIGURE 5. The germline sequences of both µCH1 and {delta}CH1 in SW1 and SW3 pigs. A, Sequences derived from SW1, µ1, µ2, µCH1 allotypic variants. B, Sequences derived from SW3, µ1, µ2, µCH1 allotypic variants. {delta}1, {delta}2, {delta}CH1 allotypic variants. Asterisks indicate the same sequence as the first line.

 
Based on the data derived from SW1 and SW3, the G allele appears to be less frequently included in the {delta} H chain transcripts than the A allele. To understand the mechanism underlying the co-usage of both the µCH1 and {delta}CH1 in expressed H chain cDNAs, two constructs containing the inserts that contain the µCH1 and {delta}CH1 were made using a mammalian expression vector pCI-neo (Fig. 6A). Both the {delta}CH1 sequences of the G and A alleles were introduced into the constructs (pDEG and pDEA; Fig. 6A), and the recombinant plasmids were used to transfect a human B cell line, BL2. Total RNA was purified from cultured cells 48 h after transfection and subjected to RT-PCR to analyze the splicing pattern of the designed pig gene sequences. The two constructs containing the G or A {delta}CH1 alleles were similarly processed after transcription in the BL2 cell line that removed the intron sequence of the designed gene, resulting in two forms of mature RNA transcripts. In one case, the VDJ sequence was directly spliced to the {delta}CH1 and its downstream exons, H1 and {delta}CH2, to generate a VDJ-{delta}CH1-H1-{delta}CH2 mRNA transcript (Fig. 6B). In the second case, the µCH1 and {delta}CH1 were both included, leading to a longer RNA transcript containing VDJ-µCH1-{delta}CH1-H1-{delta}CH2 (Fig. 6C).



View larger version (26K):
[in this window]
[in a new window]
  		FIGURE 6. Transcription of pig sequences in the human BL2 cell line. A, Schematic map of the constructs pDEG and pDEA (the two constructs have only a single-base difference at the +19 position of the {delta}CH1 exon). B and C, Two splicing pathways of both pDEG and pDEA in the human BL2 B cell line.

 
A second putative hinge exon is inactive due to a weak motif for RNA splicing

We have previously shown that the deduced porcine IgD H chain has a shorter hinge segment compared with those of humans, cows, and sheep (1). The pig hinge region is most likely encoded by a single exon. This was confirmed by analyzing the genomic sequence of the pig C{delta} gene. Interestingly, however, a second putative hinge region exon, corresponding to the bovine {delta}H2, was identified in the genomic sequences of the pig C{delta} gene between the pig {delta}H and {delta}CH2 exons (Fig. 7A). The exon is flanked by normal AG/GT RNA splicing sites, and putative inclusion of the exon in porcine IgD H chain cDNA causes neither a shift of the reading frame nor a premature translation termination (data not shown). However, an examination of the intron sequence upstream of the putative {delta}H2 exon revealed that there is no conserved branchpoint sequence (consensus in mammals, YNYURAY; Y = pyrimidine; N = any nucleotide; R = purine), which is essential for intron removal during RNA splicing, upstream of this putative exon (17, 18, 19, 20). Most likely, the deletion of a single T nucleotide destroys the functional branchpoint sequence (Fig. 7B), and the putative hinge-encoded exon is thus skipped during the RNA-splicing process, leading to a single exon-encoded hinge region in the porcine IgD H chain. To confirm this notion, a single T nucleotide was introduced into the branch site of the putative H2 exon in the plasmid pDEA, resulting in a modified plasmid, pDET. Both pDEA and pDET were used to transfect the human B cell line, BL2, to detect the splicing pattern of the hinge exons using a nested PCR. As expected, transcripts containing only H1 exon were identified by PCR in the transfection experiment using pDEA (containing an intact genomic sequence between the H1 and {delta}CH2 exons) (Fig. 8, Ac and Bb, 2 and 3). However, based on the intensity of PCR bands, only a minor portion of the PCR products were found to contain both the H1 and putative H2 exons (175-bp band; Fig. 8, Ab and Bb, 2 and 3). Introduction of a single T nucleotide into the branch site of the H2 exon increased the proportion of H2 exon containing mRNA transcripts. As shown in Fig. 8Bb, 4 and 5, the amount of the PCR products containing both the H1 and H2 sequences appeared to be almost equal to that of the PCR products containing only the H1 exon.



View larger version (84K):
[in this window]
[in a new window]
  		FIGURE 7. A second putative hinge exon is located upstream of the {delta}CH2. A, The pig genomic sequence from {delta}H1 to {delta}CH2. The coding sequences, putative hinge exon, and branchpoint sites are in bold and underlined. B, Comparison of the putative pig hinge exon and its flanking sequences with those of the bovine {delta}H2. Intron sequences are indicated in small letters.

 


View larger version (31K):
[in this window]
[in a new window]
  		FIGURE 8. Differential splicing of the porcine IgD hinge regions. Aa, The schematic map of both the pDEA and pDET. The difference between the two plasmids is that pDET contains an extra T nucleotide in the branch site of the H2 exon. Ab and Ac, Two different splicing patterns. Ba, PCR amplification of hinge regions from five animals; 1, 1 kb ladder; 2–6, different animal samples. Bb, PCR of hinge regions in transfection experiments; 1, 1 kb ladder; 2 and 3, pDEA transfection; and 4 and 5, pDET transfection.

 
Based on the above data, one may guess that, in some cases, the putative H2 exon sequence containing {delta} chain transcripts could be expressed in vivo. To check this possibility, we have performed a nested RT-PCR using spleen RNA isolated from five independent pigs. However, only H1 contained transcripts that could be amplified (Fig. 8Ba).


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
The porcine genomic C{delta} gene is located ~3.4 kb downstream of the µ gene in the pig Ig H chain constant region (IGHC) gene locus. Consistent with the previously published cDNA data (1), the genomic organization of the porcine {delta} gene is similar to that of its counterparts in cows, sheep, and humans (1, 8), indicating that they are true homologous genes in mammals. At the protein level, the pig IgD H chain constant region shows only 46.8 and 38.4% similarities with those of human and mouse, respectively (1). This finding may explain why previous attempts have failed to detect the pig {delta} gene within 20 kb downstream of the µ gene (15).

Analysis of the genomic sequence of the porcine {delta} gene revealed a germline {delta}CH1 exon that is highly homologous to the porcine µCH1 (13) (>98% homology). The high homology between the {delta}CH1 and µCH1 exon is likely to be due to a recent duplication of the µCH1 exon, a phenomenon that has previously been found in cows (1). It seems that the duplication in pigs took place even later in evolution than in cows, because the similarity between the pig µCH1 and {delta}CH1 is greater than it is in cows (98.7 vs 96.9%). Based on the data derived from cows and sheep, we have previously proposed two possibilities regarding the time at which the duplication occurred (1). The phylogenetic analysis using {delta} and µ gene sequences suggested that the duplication occurred independently after the speciation of cows and sheep (1). A second model supported a gene conversion event between the µCH1 and {delta}CH1 occurring after the DNA duplication that should then have taken place in the common ancestor of cows and sheep. The present data obtained are in favor of the first model, because the duplication was also found in the {delta} gene locus in pigs that diverged from cows and sheep ~60 million years ago (2). Unlike in cows, the duplicated porcine DNA fragment could be found to extend into the intron between the {delta}CH1 and {delta}H exons. The extremely high similarity between the intron sequences that are under low selection pressure also supports the notion that the duplication took place rather late in pig evolution. The question remains as to why this duplication has occurred only in artiodactyls such as cows, pigs, and sheep, but not in humans or rodents.

According to a comparison of the duplications in both pig and cows, the duplicated fragment in pigs is much less modified than in cows (1), which may allow a possibility to track how the genetic event occurred. An examination of the sequences flanking the duplicated DNA fragment revealed an 18-nt sequence immediately downstream of the duplicated DNA fragment in the µCH1-µCH2 intron (Fig. 4), which is highly similar to the primer binding site of RNA viruses such as HIV, feline immunodeficiency virus, and equine infectious anemia virus (21). The primer binding site sequences of these RNA viruses are complementary to an 18-nt RNA sequence of tRNA-Lys3, and all lentiviruses use this tRNA as a primer to initiate reverse transcription of their genomic RNA (21). A hypothesis to explain the process of the duplication in pig is that an unprocessed primary RNA from the µ gene was, by chance, transcribed into DNA with the aid of tRNA-Lys3 and reverse transcriptase. According to the data obtained from cow, the reverse transcription should be terminated at the 3' flanking region of the 5' Eµ enhancer. The resulting DNA was then reintegrated into the genome, forming a duplicated DNA fragment. Short repeats that are usually present in the insertion sites of retroelements can also be observed in the {delta}CH1 and {delta}H1 intron (Fig. 4). However, why the resulting DNA inserted into the same place, the {delta} gene, in cows, sheep, and pigs remains to be explained.

As in cows (1), the duplicated DNA fragment in the porcine {delta} gene locus also encompasses the Sµ region, although only a ~0.5-kb-long S{delta} could be observed. Potentially, the porcine {delta} gene could thus be expressed through class switch recombination. However, due to an extremely high similarity between the flanking sequences of both the Sµ and S{delta} regions, it is virtually impossible to amplify the switch fragments by using PCR because of apparent difficulties to design suitable PCR primers.

A striking characteristic of the porcine {delta} gene is that both the µCH1 and {delta}CH1 can be used in the IgD H chain transcripts. Thus, some IgD H chain transcripts contain a unique {delta} sequence, whereas in others, the µCH1 is fused to the {delta} sequence, displaying a chimeric structure that has previously been seen only in teleost fishes (4, 9, 10). As the µCH1 is highly homologous to the {delta}CH1, both at the cDNA and protein levels, the two forms of pig IgD H chains resemble variants generated from two allotypes. However, a detailed examination of both the genomic and cDNA sequences of several individual animals demonstrated the presence of chimeric IgD H chain transcripts in which the first constant region exon is spliced from the µ gene. This finding indicates that, as in humans and mice, the porcine {delta} gene may be cotranscribed together with the µ gene, generating a long primary transcript, which can be used to create different forms of mature RNA transcripts due to alternative RNA splicing. Although the chimeric IgD H chain transcripts have now been observed in both teleost fish and pigs, there is, however, an obvious difference in the two situations. In channel catfish, the inclusion of the µCH1 is required to allow normal association of catfish IgD with L chains, because the cysteine expected to form the disulfide bond with the L chain is lacking in the first C{delta} domain. In the porcine pre-mRNA transcripts consisting of both the µ and {delta} genes, there are two homologous exons, µCH1 and {delta}CH1, both containing three cysteines, from which only one is finally included in the IgD H chain transcripts. Examination of the flanking sequences of the µCH1 and {delta}CH1 exons showed that both contain exactly the same 5' and 3' intron sequences that are essential for RNA splicing (Fig. 4). One possibility to explain the inclusion of µCH1 in the IgD H chain transcripts is the positional priority of the µCH1 exon compared with that of the {delta}CH1. However, an in vitro transfection experiment using a human B cell line did not support this concept, because either both µCH1 and {delta}CH1 exon sequences or only the {delta}CH1 exon sequence were included in the mature RNA transcripts transcribed from the genetic constructs. A second possibility could be that the sequence differences between the µCH1 and {delta}CH1 exons, but not in the flanking intron, might influence the usage of the two exons in the final IgD H chains. There is recent evidence suggesting that mutations within the exon sequence itself, or rather within the exonic splicing enhancers (ESE) that are usually located in the general vicinity of splice sites (22), change the splicing pattern of genes (23, 24). The sequence differences between the pig {delta}CH1 and µCH1 exons might be located within a predicted ESE and may thus influence their splicing pattern. However, in vitro data using a human B cell line did not support this hypothesis. As mentioned above, the mature transcripts obtained from the study contain either both the µCH1 and {delta}CH1 or only {delta}CH1, and no transcripts containing only the µCH1 were identified. However, human B cells may not be identical with pig B cells, because some trans factors, which interact with the ESE in pig B cells, may not be present in the former.

At the cDNA level, a second hinge-encoding exon is not included in the transcript, leading to a short hinge peptide. One exon-encoded short hinges of IgD have previously been observed in rodents (5, 6, 7, 25). However, in the mouse and rat genomic {delta} genes, no remnants of a second hinge exon can be identified. The data obtained for the pig IgD hinge supports the notion of a gradual evolution in terms of length. Biochemically, the data derived from this study also helps to underline the importance of the branch site for RNA splicing. The transfection experiment clearly showed that the U nucleotide in the branch site (YNYURAY) was not essential for intron removal but greatly influenced its efficiency. The exclusion of the H2 exon from the porcine IgD H chain transcripts may indicate that, in addition to the weak branch site, there are still other elements inhibiting normal splicing of the H2 exon.

In summary, the data presented in this study help to understand the evolution of not only the Ig D but also the hinge region of Igs.


   Acknowledgments
 
We thank Dr. Hodjattallah Rabbani for his help in obtaining the pig PAC clone.


   Footnotes
 
1 This work was supported by the Swedish Research Council. Back

2 The sequences presented in this article have been submitted to GenBank under accession numbers AY228505–228508. Back

3 Address correspondence and reprint requests to Dr. Yaofeng Zhao, Center for Biotechnology at Novum, SE-141 57, Huddinge, Sweden. E-mail address: yafe{at}csb.ki.se Back

4 Abbreviations used in this paper: PAC, P1-derived artificial chromosome; PFGE, pulsed field gel electrophoresis; ESE, exonic splicing enhancer; TM, transmembrane. Back

Received for publication March 24, 2003. Accepted for publication May 23, 2003.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Zhao, Y., I. Kacskovics, Q. Pan, D. A. Liberles, J. Geli, S. K. Davis, H. Rabbani, L. Hammarstrom. 2002. Artiodactyl IgD: the missing link. J. Immunol. 169:4408.[Abstract/Free Full Text]
  2. Kumar, S., S. B. Hedges. 1998. A molecular timescale for vertebrate evolution. Nature 392:917.
  3. Bengten, E., S. M. Quiniou, T. B. Stuge, T. Katagiri, N. W. Miller, L. W. Clem, G. W. Warr, M. Wilson. 2002. The IgH locus of the channel catfish, Ictalurus punctatus, contains multiple constant region gene sequences: different genes encode heavy chains of membrane and secreted IgD. J. Immunol. 169:2488.[Abstract/Free Full Text]
  4. Wilson, M., E. Bengten, N. W. Miller, L. W. Clem, L. Du Pasquier, G. W. Warr. 1997. A novel chimeric Ig heavy chain from a teleost fish shares similarities to IgD. Proc. Natl. Acad. Sci. USA 94:4593.[Abstract/Free Full Text]
  5. Tucker, P. W., C. P. Liu, J. F. Mushinski, F. R. Blattner. 1980. Mouse immunoglobulin D: messenger RNA and genomic DNA sequences. Science 209:1353.[Abstract/Free Full Text]
  6. Cheng, H. L., F. R. Blattner, L. Fitzmaurice, J. F. Mushinski, P. W. Tucker. 1982. Structure of genes for membrane and secreted murine IgD heavy chains. Nature 296:410.[Medline]
  7. Zhao, Y., L. Hammarstrom. 2003. Cloning of the complete rat immunoglobulin {delta} gene: evolutionary implications. Immunology 108:288.[Medline]
  8. White, M. B., A. L. Shen, C. J. Word, P. W. Tucker, F. R. Blattner. 1985. Human immunoglobulin D: genomic sequence of the {delta} heavy chain. Science 228:733.[Abstract/Free Full Text]
  9. Hordvik, I., J. Thevarajan, I. Samdal, N. Bastani, B. Krossoy. 1999. Molecular cloning and phylogenetic analysis of the Atlantic salmon immunoglobulin D gene. Scand. J. Immunol. 50:202.[Medline]
  10. Stenvik, J., T. O. Jorgensen. 2000. Immunoglobulin D (IgD) of Atlantic cod has a unique structure. Immunogenetics 51:452.[Medline]
  11. Maki, R., W. Roeder, A. Traunecker, C. Sidman, M. Wabl, W. Raschke, S. Tonegawa. 1981. The role of DNA rearrangement and alternative RNA processing in the expression of immunoglobulin {delta} genes. Cell 24:353.[Medline]
  12. Al-Bayati, H. K., S. Duscher, S. Kollers, G. Rettenberger, R. Fries, B. Brenig. 1999. Construction and characterization of a porcine P1-derived artificial chromosome (PAC) library covering 3.2 genome equivalents and cytogenetical assignment of six type I and type II loci. Mamm. Genome 10:569.[Medline]
  13. Sun, J., J. E. Butler. 1997. Sequence analysis of pig switch µ, Cµ, and Cµm. Immunogenetics 46:452.[Medline]
  14. Pan-Hammarstrom, Q., S. Dai, Y. Zhao, I. F. van Dijk-Hard, R. A. Gatti, A. L. Borresen-Dale, L. Hammarstrom. 2003. ATM is not required in somatic hypermutation of V(H), but is involved in the introduction of mutations in the switch micro region. J. Immunol. 170:3707.[Abstract/Free Full Text]
  15. Butler, J. E., J. Sun, P. Navarro. 1996. The swine Ig heavy chain locus has a single JH and no identifiable IgD. Int. Immunol. 8:1897.[Abstract/Free Full Text]
  16. Pan, Q., C. Petit-Frere, J. Stavnezer, L. Hammarstrom. 2000. Regulation of the promoter for human immunoglobulin {gamma}3 germ-line transcription and its interaction with the 3'{alpha} enhancer. Eur. J. Immunol. 30:1019.[Medline]
  17. Moore, M. J., C. C. Query, P. A. Sharp. 1993. Splicing of precursors to mRNAs by the spliceosome. R. F. Gesteland, and J. F. Atkins, eds. In The RNA World 303. Cold Spring Harbor Lab. Press, Plainview, NY.
  18. Berglund, J. A., K. Chua, N. Abovich, R. Reed, M. Rosbash. 1997. The splicing factor BBP interacts specifically with the pre-mRNA branchpoint sequence UACUAAC. Cell 89:781.[Medline]
  19. Li, M., P. H. Pritchard. 2000. Characterization of the effects of mutations in the putative branchpoint sequence of intron 4 on the splicing within the human lecithin: cholesterol acyltransferase gene. J. Biol. Chem. 275:18079.[Abstract/Free Full Text]
  20. Liu, Z., I. Luyten, M. J. Bottomley, A. C. Messias, S. Houngninou-Molango, R. Sprangers, K. Zanier, A. Kramer, M. Sattler. 2001. Structural basis for recognition of the intron branch site RNA by splicing factor 1. Science 294:1098.[Abstract/Free Full Text]
  21. Arts, E. J., S. R. Stetor, X. Li, J. W. Rausch, K. J. Howard, B. Ehresmann, T. W. North, B. M. Wohrl, R. S. Goody, M. A. Wainberg, S. F. Grice. 1996. Initiation of (-) strand DNA synthesis from tRNA(3Lys) on lentiviral RNAs: implications of specific HIV-1 RNA-tRNA(3Lys) interactions inhibiting primer utilization by retroviral reverse transcriptases. Proc. Natl. Acad. Sci. USA 93:10063.[Abstract/Free Full Text]
  22. Berget, S. M.. 1995. Exon recognition in vertebrate splicing. J. Biol. Chem. 270:2411.[Free Full Text]
  23. Fairbrother, W. G., R. F. Yeh, P. A. Sharp, C. B. Burge. 2002. Predictive identification of exonic splicing enhancers in human genes. Science 297:1007.[Abstract/Free Full Text]
  24. Cartegni, L., S. L. Chew, A. R. Krainer. 2002. Listening to silence and understanding nonsense: exonic mutations that affect splicing. Nat. Rev. Genet. 3:285.[Medline]
  25. Sire, J., C. Auffray, B. R. Jordan. 1982. Rat immunoglobulin {delta} heavy chain gene: nucleotide sequence derived from cloned cDNA. Gene 20:377.[Medline]



This article has been cited by other articles:


Home page
 J. Immunol.Home page
Z. Wei, Q. Wu, L. Ren, X. Hu, Y. Guo, G. W. Warr, L. Hammarstrom, N. Li, and Y. Zhao
Expression of IgM, IgD, and IgY in a Reptile, Anolis carolinensis
J. Immunol., September 15, 2009; 183(6): 3858 - 3864.
[Abstract] [Full Text] [PDF]

Home page
 J. Immunol.Home page
Y. Zhao, H. Cui, C. M. Whittington, Z. Wei, X. Zhang, Z. Zhang, L. Yu, L. Ren, X. Hu, Y. Zhang, et al.
Ornithorhynchus anatinus (Platypus) Links the Evolution of Immunoglobulin Genes in Eutherian Mammals and Nonmammalian Tetrapods
J. Immunol., September 1, 2009; 183(5): 3285 - 3293.
[Abstract] [Full Text] [PDF]

Home page
 Proc. Natl. Acad. Sci. USAHome page
Y. Zhao, Q. Pan-Hammarstrom, S. Yu, N. Wertz, X. Zhang, N. Li, J. E. Butler, and L. Hammarstrom
Identification of IgF, a hinge-region-containing Ig class, and IgD in Xenopus tropicalis
PNAS, August 8, 2006; 103(32): 12087 - 12092.
[Abstract] [Full Text] [PDF]

Home page
 Proc. Natl. Acad. Sci. USAHome page
Y. Ohta and M. Flajnik
IgD, like IgM, is a primordial immunoglobulin class perpetuated in most jawed vertebrates
PNAS, July 11, 2006; 103(28): 10723 - 10728.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Zhao, Y.
Right arrow Articles by Hammarström, L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Zhao, Y.
Right arrow Articles by Hammarström, L.
Right arrowPubmed/NCBI databases
*Protein

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS