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 Fleurant, M.
Right arrow Articles by Hsu, E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Fleurant, M.
Right arrow Articles by Hsu, E.
The Journal of Immunology, 2004, 173: 5574-5582.
Copyright © 2004 by The American Association of Immunologists

Shark Ig Light Chain Junctions Are as Diverse as in Heavy Chains1

Marshall Fleurant*, Lily Changchien*, Chin-Tung Chen*, Martin F. Flajnik{dagger} and Ellen Hsu2,*

* Department of Physiology and Pharmacology, State University of New York Health Science Center, Brooklyn, NY 11203; and {dagger} Department of Microbiology and Immunology, University of Maryland, Baltimore, MD 21201


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
We have characterized a small family of four genes encoding one of the three nurse shark Ig L chain isotypes, called NS5. All NS5 cDNA sequences are encoded by three loci, of which two are organized as conventional clusters, each consisting of a V and J gene segment that can recombine and one C region exon; the third contains a germline-joined VJ in-frame and the fourth locus is a pseudogene. This is the second nurse shark L chain type where both germline-joined and split V-J organizations have been found. Since there are only two rearranging Ig loci, it was possible for the first time to examine junctional diversity in defined fish Ig genes, comparing productive vs nonproductive rearrangements. N region addition was found to be considerably more extensive in length and in frequency than any other vertebrate L chain so far reported and rivals that in H chain. We put forth the speculation that the unprecedented efficiency of N region addition (87–93% of NS5 sequences) may be a result not only of simultaneous H and L chain rearrangement in the shark but also of processing events that afford greater accessibility of the V or J gene coding ends to terminal deoxynucleotidyltransferase.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Antigen receptor diversity, the basis of adaptive immunity, is generated in part by a mechanism of somatic recombination that brings together the different gene segments to be expressed in lymphocytes (1). Any tetrapod Ig is encoded by a H chain locus and one of 1–3 L chain loci. The Ag-binding portion is generated by rearrangement among the multiple tandemly duplicated V, D, and J gene segments; VL to JL recombination occurs for L chain and VH, D, and JH for H chain. Combinatorial diversity thus results from different combinations of the gene segments. Although the sites of the DNA double-strand breaks are precisely determined by the recombination signal sequences (RSS),3 the ends undergo modifications entailing nucleotide deletion and addition before rejoining, resulting in sequence and sequence length differences at the VH/D, D/JH, and the VL/JL joints; this is called junctional diversity (for a review, see Ref. 2).

In contrast, cartilaginous fishes such as sharks and skates, representatives of the oldest jawed vertebrates that possess Ig and TCR gene systems, carry 100–200 Ig loci ("clusters"), each one consisting of a few gene segments (VH-D-D-J-Cµ, VL-J-CL) (for a review, see Ref. 3). These rearrange only within a cluster and independently of other clusters (4). This is an early, alternative form of the Ag receptor gene organization. Another unusual feature of the Ig clusters in cartilaginous fishes is that many of them contain partially or fully assembled germline VHD, VHDD, VHDDJ, and VLJ sequences (5). Many of the fused V(D)J sequences are in-frame and potentially functional. Thus, it would seem that for shark H and L chains, combinatorial diversity does not exist and junctional diversity is restricted. Nonetheless, when shark Ig sequences are examined, it is clear that there is in fact a diverse Ab repertoire. We have been investigating the compensations for these apparent restrictions by systematically characterizing the germline Ig genes and analyzing the transcribed products.

We have defined three L chain isotypes in the nurse shark, Ginglymostoma cirratum, called NS3, NS4, and NS5 (6, 7), and they are the respective homologues of type II, III, and I L chains in the horned shark (3, 8) (Fig. 1). NS4, the {kappa} L chain homologue (9), is the largest IgL gene family and consists of ~60 loci, 6 of which carry fused VJ but the rest can rearrange (10). In contrast, all six NS3 L chain clusters carry germline-joined VJ (11); we had asked whether these preassembled Ig sequences remain invariant throughout the life of the shark. The NS3 genes in fact undergo extensive Ag-dependent hypermutation, more than half of the changes consisting of side-by-side substitutions extending 2–4 bp (11); the tandem mutations may arise by a pathway different from that which generates point mutations. Possibly hypermutations en bloc partly offset the other limitations on the shark primary repertoire.



View larger version (17K):
[in this window]
[in a new window]
  		FIGURE 1. Gene organizations of L chain types in nurse shark and horned shark. The three L chain types in nurse shark (left) are compared with their horn shark homologue (right). The clusters in horn shark are all split (Type I, Type III) or all germline joined (Type II) (3 8 ), in contrast to the "mixed" arrangements in nurse shark NS4 and NS5 (Refs. 10 and 11 and this study).

 
The third L chain type in nurse shark, called NS5, has been identified and characterized (6, 7). It is now found that the cDNA sequences are all encoded by a small family consisting of one germline-joined cluster and two clusters capable of rearrangement. Together with the NS4 L chain, this is another instance where clusters with joined as well as split gene organizations of an IgL isotype has been found, supporting the idea of rare VJ recombination taking place in the shark germ cells (10). NS5 is a very small family of genes, and the same genes are carried by several sharks from the wild. Since rearrangement can occur at only two of the NS5 loci, it was possible to examine junctional diversity in the NS5 L chains. Productive as well as nonproductive rearrangements were isolated and compared. Fish Ig junctional diversity has not been analyzed with respect to a defined germline gene, and in the shark L chain genes we have found it far more extensive than what has been observed in any other vertebrate L chains.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Animals

Nurse sharks were captured off the coast of the Florida Keys and bled from the caudal sinus into heparinized syringes. Whole blood was separated by Ficoll into erythrocyte and peripheral blood leukocyte fractions from which genomic DNA and total RNA were isolated, respectively. Blood samples were obtained from two adult individuals, shark-Y and shark-J; the epigonal organ was removed from adult shark-33. For one experiment, genomic DNA was extracted from lymphocyte-enriched shark-J PBL. In this study, the PBL fraction was resuspended to 3 x 106 cells/ml in RPMI 1640 medium containing 10% FCS adjusted to elasmobranch osmolarity with the addition of 350 mM urea and 200 mM NaCl. The cells were adhered to plastic for 2 h at 27°C and 5% CO2; most of the adherent cells had a thrombocyte morphology (12). The nonadherent cells were collected, centrifuged at 300 x g for 10 min, and the pellets were resuspended in lysis buffer for preparation of genomic DNA.

Libraries

Erythrocyte DNA from shark-Y was digested with Sau3AI and cloned into Super Cos1 (Stratagene, La Jolla, CA). A Sau3AI partial library was constructed from shark-Y genomic DNA (Lofstrand Labs, Gaithersburg, MD). A cDNA library was prepared from the epigonal organ RNA of adult shark-33 ({lambda}ZAP ExpressXR; Lofstrand Labs).

Probes to the V and C regions of NS5-2 were used to screen both the genomic and cDNA libraries. Hybridization conditions for the library filters and the genomic Southern blots were as follows: hybridization (0.2% polyvinylpyrrolidone, 0.2% Ficoll, 0.2% BSA, 50 mM Tris (pH 7.5), 0.1% sodium pyrophosphate, and 1% SDS) with 5 x 105 cpm/ml random primed DNA (Roche, Basel, Switzerland) at 68°C overnight, followed by three washes (2x SSC, 0.1% SDS) at 55°C, and final rinse in 0.2x SSC at room temperature before autoradiography.

Polymerase chain reaction

Oligonucleotide primers specific to NS5-2 framework (FR) 1 (codons 19–25, NS5-2: 5'-CGGTTGAACTGTGCGTT-3'), J gene segment (NS5-J: codons 97–103, 5'-CGAACGGTCAGGGCAGT-3'), and C region (NS5-3: codons 106–111, 5'-AGGACAGATGGTTTCCG-3') were synthesized (Invitrogen Life Technologies, Grand Island, NY) and used to amplify sequences from cDNA primed with oligo(dT) (11). PCR cycling parameters were: 96°C for 1 min, 55°C for 2 min, 72°C for 2 min for 39 cycles, and a final cycle with the 72°C step extended to 15 min. Alternative program included 55°C for 30 s and 72°C for 30 s, and the PCR products are distinguished in the text as series 1 and 3, respectively. The RNA source for RT-PCR was shark-Y PBL. For experiments using genomic DNA from PBL, primers to the NS5-2 leader intron (NS5LI: 5'-CTGGCAAAGTCACTCAAAGTG-3') and to the region downstream from the J gene segment (NS5JI: 5'-AGAGAACAGAATCGACACTGG-3') were synthesized; nonrearranged NS5-2 was anticipated to be 880 bp, in contrast to rearranged VJ of ~486 bp. In this case, the shorter annealing and elongation times were chosen for preferential amplification of the shorter sequences. PCR products were isolated and ligated into pGEM-T-Easy (Promega, Madison, WI). The sequencing was performed by the DNA Sequencing Lab at Rockefeller University (New York, N.Y.).


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Germline NS5 genes

A full-length NS5 L chain sequence was initially isolated from a splenic cDNA library and characterized as the nurse shark homologue of the horn shark "type I" L chain (6). Cosmid and genomic libraries were generated from shark-Y erythrocyte DNA and screened with a probe generated from this cDNA.

One germline gene, NS5-2, was obtained from a cosmid clone, and three other different ones were isolated from the phage library (Fig. 2). Each phage carried one V gene segment, one J and one C gene. Both the regulatory (octamer, splice sites, RSS) and coding sequences (split leader, V gene, J gene), in content and position, are typical of germline L chain genes from other species. The NS5-2 locus overall is ~6.4 kb from the octamer to the TGA of the C region. It is similar in organization although larger in size than a horn shark type I L chain locus (2.7 kb) (14). Upon a database search, the closest match to NS5-2 was horn shark type I L chain with 73% identity in the V gene segment (74% amino acid) and 83% identity in the C gene (76% amino acid). In mammals, matches with {kappa} and {lambda} L chain amino acid sequences occurred with 38–42% identity.



View larger version (90K):
[in this window]
[in a new window]
  		FIGURE 2. Nucleotide sequences of germline NS5 genes. Comparison of NS5-2, NS5-48, NS5-16, and NS5-37 genomic sequences showing split leader with intron, V gene segment, and J gene segment. NS5-16 is germline joined. The J gene segment of NS3-37 has not been determined. Coding regions are underlined, regulatory elements such as the octamer sequence, splice sites, and the RSS on either side of the V and J gene segments are bolded. The V region is divided into FR and CDR regions according to Kabat et al. (13 ). The numbering of V gene residues was done according to homology with human L chain sequences. Residues that are additional in the NS5 sequence (28A–B, 52A–D, 65A–B, 95A–B) were placed according to probable loop location and as established by Kabat et al. (13 ). The sequence TGTGT in 5' of the NS5-37 V gene segment indicates its insertion in the location indicated (accession nos. AY720853–AY720856).

 
After extensive screening, only three other NS5 genes were repeatedly obtained. NS5-48 is very closely related to NS5-2, with 98% nucleic acid identity in V. NS5-16 is germline joined, but its V sequence shows 82% identity with NS5-2 compared with other germline-joined genes of the other L chain isotypes (54% identity with NS3-23 (11); 56% identity with NS4-S4 (10)). NS5-37 is a pseudogene that is 83% identical to NS5-2 in the V gene segment but carried noncanonical RSS (CATAATG—12 bp spacer—ACAACAAGG) and octamer (ATTTGCAG). Because of substitutions in the leader methionine and the acceptor splice site homologous to NS5-2, the deduced leader sequence in NS5-37 has been shifted to compose a possible functional transcript (Fig. 3), although none has been found by RT-PCR.



View larger version (33K):
[in this window]
[in a new window]
  		FIGURE 3. Genomic Southern blotting of nurse shark DNA. Top, Map of NS5-2 V and J gene segments. Filled boxes represent (left to right) leader, V gene, and J gene segments; filled oval is the octamer sequence. The restriction enzyme sites are shown along with the sizes of the fragments generated by double digestions of SspI with HindIII (1136 bp, 791 bp) or with XhoI (1197 bp, 730 bp). The locations of the two probes, V and Int, are shown as open bars. Bottom, Restriction enzyme analyses of the NS5 genes. Genomic RBC DNA from shark-Y, shark-33, and shark-J (shown) were digested with restriction enzymes, electrophoresed on a 1.5% Tris-borate-EDTA gel, transferred onto nylon filters, and hybridized with the Int probe (left panel) which detects only NS5-2 and NS5-48 or V probe (right panel) which cross-hybridizes with all four NS5 V genes. The restriction enzymes are indicated over the lanes and the DNA markers({lambda}HindIII and 100 bp marker) at the right sides of each panel. To the right of the autoradiograms are interpretations depicting NS5-2 (filled bar), NS5-48 (open bar), NS5-37 (slashed), and NS5-16 (vertical stripes). The NS5-16 band at 9 kb in the SspI lane was faint though distinct on the autoradiogram but could not be rendered visible in the scan.

 
The four genomic constant (C) region exons were isolated from the cosmid and the phages (accession nos. AY720857–AY720860); those of the three functional genes match the C region sequences from cDNA clones (data not shown). The C region of NS5-2 contains six amino acid differences from NS5-48 (93.6% nt identity) and 10 differences from NS5-16 (90.8% nt identity). The genomic C region of NS5-37 carries a 1-bp deletion generating a nonsense frameshift, supporting the idea that it is a pseudogene. Excluding the codon affected, there are 27 aa differences (82.5% nt identity) between the C regions of NS5-2 and NS5-37.

Rearrangement occurs only at two loci

The NS5-2V probe cross-hybridizes with all NS5 sequences and detects few fragments by genomic Southern blotting, supporting the impression that there are four NS5 gene clusters in total. Since many mutant cDNA sequences were obtained (see below), experiments were performed to show that only two genes rearrange and no additional, related sequences participated in the diversification of the NS5 L chain repertoire.

The NS5-2 and -48 V genes carry a HindIII site at the FR2/CDR2 border and an XhoI site in FR3; there were 7 (23%) and 3 (10%) sequences with changes in the respective restriction enzyme sites of 31 NS5-2 RT-PCR mutant sequences. If these changes, all of them different, were derived from germline templates by gene conversion, we would accordingly detect genomic NS5 counterparts without internal HindIII or XhoI sites. The following genomic Southern analyses show that only two such genes exist in the shark germline.

A genomic SspI fragment of 1927 bp encompasses the NS5-2 V and J gene segments (Fig. 3, top). Digestion with both SspI and HindIII produces fragments of 1136 and 791 bp, of which the latter is detected with a probe hybridizing to the intervening DNA (Fig. 3, middle, Int probe, lane Ssp/H). Double digestion with SspI and XhoI produced two fragments of 1197 and 730 bp; the latter is detected (Fig. 3, panel Int probe, lane Ssp/X). The second SspI fragment at ~2500 bp is the NS5-48 gene (Fig. 3, panel Int probe, lane SspI) and results with the double digests parallel those of NS5-2, as anticipated from analyses with the NS5-48 phage lysates (data not shown). The interpretation of these results is shown next to the gels, differentiating NS5-2 (filled bars) from NS5-48 (open bars).

The NS5-2 V probe, cross-hybridizing with all four NS5 genes, detected NS5-2 (1927 bp) and NS5-48 (~2500 bp), NS5-37 (~3 kb), and NS5-16 (>9 kb) in the SspI lane (Fig. 3, middle, panel V probe). After double digestion with SspI/HindIII, the latter two NS5 genes (NS5-37 is slashed bar, NS5-16 is vertically striped bar in the interpretation), neither of that have internal HindIII sites remain singly hybridizing fragments. The fragments expected from NS5-2 (1136 and 791 bp) and NS5-48 (~1200 and 1250 bp) are detected by the V probe in Fig. 3, center, lane Ssp/H.

Thus, there is no indication of additional NS5 genes in the nurse shark genome. Shark-Y, shark-33, and shark-J have identical patterns of hybridization with probes to the C region, V region, and the intervening sequence (data not shown); only the SspI fragment of NS5-16 in shark-Y is smaller, at 4 kb instead of >9 kb. Nonmutant cDNA or PCR-amplified genomic sequences (below) show that all three animals express the same NS5 genes. In this study, unique, rearranging germline Ig genes have been characterized, and their rearranged products can be unequivocally identified and examined.

Rearranged NS5 sequences

In the course of screening a primary adult epigonal cDNA library from shark-33, 50 phages were initially selected that hybridized to NS5 probes. Of the four NS5 genes characterized from the genomic library, only NS5-2 (25 phages), NS5-48 (14 phages), and NS5-16 (11 phages) were found. Most of the cDNA sequences carried mutations in the V region, but their C regions matched the genomic counterparts. An additional 31 NS5-2 cDNA sequences were obtained by RT-PCR from PBL RNA of shark-Y, and their junctions are shown in Fig. 4(top).



View larger version (26K):
[in this window]
[in a new window]
  		FIGURE 4. Junctional diversity in NS5-2 sequences. RT-PCR was performed on adult shark-Y PBL (top) total RNA compared with PCR products from shark-J PBL genomic DNA (bottom). The PCR clones were sequenced in both directions, but only the CDR3 are shown. The reference sequence consists of the coding V gene and J gene segment flanks underlined in the CDR3 region. The V gene flank, as shown, begins with the codon 95A, GCT; the 96th codon in J is TTT (numbering according to Ref. 13 ). Mutations are shown, dots indicate identity with germline sequence; dashes indicate deletions. Additions in the form of P region (italicized) and N region are as indicated. At the right, columns indicate total number of substitutions per sequence, and CDR3 size if the rearrangement is in-frame. The nonproductive sequences are labeled "out."

 
The very unusual feature is the frequency and extent of N region addition found in the NS5-2 cDNA sequences; 28 of 31 RT-PCR sequences contain 1–10 bp of nontemplated sequences, with an average length of 4 bp (Fig. 4, top). In addition, analysis of the sequences from the epigonal cDNA library showed that 35 of 39 of NS5-2/-48 sequences had N regions or at least 90% of the L chains, which is an extraordinarily high frequency for L chains.

To see whether the selection of B cell populations may have skewed the cDNA representation of NS5 and therefore of the CDR3 and N region content, PCR was performed on PBL DNA from shark-J to amplify the recombined genomic NS5-2 genes. The primers were located in the leader intron and in the intervening sequence 3' of the J gene segment, and PCR products of 880 and ~486 bp were selected by gel purification. Both were cloned and sequenced. The larger fragments were unrearranged NS5-2, and in 19 sequences there were 4 differences (4 of 16,017 bp, 0.025%), which were the result of Taq misincorporation.

The experiment was intended to ascertain whether the resting B cell population carried a different N region addition frequency. However, we harvested many nonfunctional rearrangements in the process, which turned out to be more informative (indicated as "out" in Fig. 4). Of 29 rearranged sequences, 23 were nonfunctional, of which 22 contained N region addition (Fig. 4, bottom), and these were even more extensive, ranging from 1 to 18 bp, averaging at 7.3 bp. Thus, the prevalence of N region addition is inherent in the rearrangement process.

The proportion of genomic NS5-2 nonproductive rearrangements obtained was much higher than expected if only NS5-2-expressing cells were sampled. It is quite possible that some of the nonfunctional NS5-2 were aborted rearrangements originated from NS5-48- and NS5-16-expressing cells, perhaps even some NS3- or NS4-expressing cells. However, the large number of productive rearrangements in the cDNA compared with genomic DNA is due to RT-PCR amplification of the mRNA population, to which activated B cells contribute disproportionally more functional Ig transcripts. In contrast, the PCR results from genomic DNA reflected individual cells, and perhaps in shark-J the majority was resting B lymphocytes. This idea is supported by the observation that most of the cDNA sequences are mutants (28 of 31), suggestive of activation at some stage, compared with few mutants among the genomic sequences (6 of 29; see "total changes" in Fig. 4).

Most of the NS5 sequences from the cDNA library as well as the RT-PCR products were mutants, carrying point and tandem substitutions that were previously described (11); the nonmutant V regions matched the defined NS5 V gene segments. Like in mammals, the nurse shark cluster/allele with a nonfunctional rearrangement remains active; transcription is ongoing (cDNA e1–9, e1–13 in Fig. 4) and hypermutation can occur (cDNA e1–9; genomic DNA clones GNS5-6, -13, -15, -22, and -42). An in-depth analysis comparing mutations in productive and out-of-frame NS5 rearrangements is currently in progress. In the present report, we confirm that the nature of the mutations in all three NS5 genes, germline rearranged or not, is very similar to the germline-rearranged NS3.

Processing of NS5 flanks

In shark-Y PBL and shark-J PBL, respectively, 3.4 ± 2.2 and 3.3 ± 2.2 bp were deleted from the NS5-2 V gene segment end, whereas 1.0 ± 1.3 and 1.5 ± 2.6 bp were deleted from the NS5-2 J flank (Fig. 4). This is at once visually obvious from the larger number of J flanks carrying P region. A similar result was obtained from 28 shark-33 NS5-2 epigonal cDNA sequences (data not shown). Since all of these 88 sequences consist of both in-frame and nonproductive rearrangements and originate from three individuals, this trimming asymmetry is intrinsic to the processing of the NS5-2 gene. Additional evidence also comes from the NS5-48 gene, which carries the same flanks as NS5-2. Of 23 NS5-48 cDNA sequences that were eventually obtained from the epigonal cDNA library, there is deletion of 4.0 ± 2.2 bp from the V flank and 2.2 ± 1.6 bp from the J flank (data not shown).

N region is present in almost all of the NS5 sequences, in both adults and in neonatal pups (our unpublished results). This is not unexpected, because terminal deoxynucleotidyltransferase (TdT) has been detected in very young elasmobranchs (15). The extent of N region addition appears to increase somewhat during ontogeny, although the average CDR3 size is not appreciably different (31.7 bp in pups compared with 33.9 bp in adults). The nature of the N region addition is tallied in Table I. Overall, the higher GC nucleotide content of the N regions is similar to other systems and is ~75%. The exception is the group of genomic sequences consisting mostly of out-of-frame CDR3, which shows a GC content of 64%. Because of this discrepancy, we report here some preliminary data from other work on shark-J to clarify this point: the N regions of productive rearrangements contain 75% GC, whereas the N regions from nonproductive joins, isolated from the same PBL cDNA batch, contain 67% GC. Thus, the "true" GC content of nurse shark N regions is lower than in the selected B cell population.


View this table:
[in this window]
[in a new window]
  		Table I. Nature of N region additions in NS5 junctions

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Germline-joined genes

The NS5 L chain is the second nurse shark Ig isotype for which we have found gene clusters with heterogeneous organization of V and J gene segments. That is, like the NS4 isotype, some NS5 loci contain split V and J gene segments that rearrange in somatic cells while others carry preassembled VJ sequences. The NS4 V genes are closely related (up to 99% identity), enabling the construction of a molecular clock to estimate dates of the recombination events that produced the six germline-joined genes; they are thought to have occurred over a period of 0–49 million years ago (10). At 82% identity with NS5-2, the rearranged NS5-16 V gene is a member of the same family and definitely not a translocation from the NS3 or NS4 loci, but it diverged a long time ago from the NS5 split genes. The closest horn shark match, gene HFL141 (accession no. X15315), is 70% identical to NS5-16 and 75% identical to NS5-2 over codons 1–92 (numbering according to Ref. 13). Since the horn shark homologues, type I L chain, are all reported to be split (3, 8), the germline rearrangement of the NS5-16 gene occurred sometime after divergence of nurse shark, 180 million years ago.

Of the three L chain isotypes in nurse shark, there are >50 split and 6 rearranged NS4 genes, 6 rearranged NS3 genes, and 3 split and 1 rearranged NS5 genes. Thus, of >70 L chain gene clusters, only 13 carry germline-joined VJ genes. This situation is in contrast to that of sandbar shark, where the NS3 homologue, also germline joined, occurs in greater numbers (16). This suggests that, generally, germline-joined Ig is not deleterious, as the proportion fluctuates among shark species. However, the fact that extensive N region addition occurs throughout ontogeny may create a certain niche for those sequences with a deviant fixed-length CDR3. NS5-2 CDR3 sequences average 11 codons (Fig. 4) compared with 6 codons in the fixed NS5-16, and no CDR3 of <9 codons has been found in productively rearranged NS5–2, even in the neonates. Thus, combinations of H chain with NS5-16 may generate a very different Ag-combining site from other NS5, providing a nonoverlapping spectrum of specificities.

L chain N region addition as frequent as in H chain

As shown in Table II, the frequency and length of N region addition in nurse shark L chains exceeds what has been reported in other vertebrate species. This observation is not restricted to productive, in-frame rearrangements that may have undergone selection; nonproductive NS5-2 also show the same characteristics. Although the N regions in neonatal sequences are somewhat shorter, the frequency of N region addition is the same as those from adult PBL. Since this is also a characteristic of the joints from another nurse shark L chain isotype, NS4 (Table II), it is probable that the preponderance of N region addition is integral to the Ig rearrangement process in nurse shark B cells.


View this table:
[in this window]
[in a new window]
  		Table II. N region addition at L chain junctions from different species

 
The frequency of N region addition in adult vertebrate H chain is >70% (24), and in nurse shark H chain rearrangements at one locus, V18, it is 100% (C.-T. Chen and E. Hsu, unpublished results). The extent of N region addition in 13 nurse shark H chain sequences averages 3.2 ± 1.9 bp/junction (3.8 ± 1.9 bp for V/D1; 2.4 ± 2.0 bp for D1/D2; 3.5 ± 1.7 bp for D2/JH), and 35 of 39 junctions (90%) contain N region. Thus, the shark L chain coding ends are as accessible to TdT as any H chain, possessing a similar extent and length of N region addition. This is in great contrast to other vertebrate systems, where the paucity of N region addition in L chains is due to the decrease in TdT levels following successful H chain rearrangement (24).

The simplest interpretation adhering to the mammalian model is that shark L chain rearrangement follows the expression of a functional H chain protein coupled to surrogate L chain, but that TdT is expressed throughout the shark pro- and pre-B cell stages. So far, no homologue of surrogate L chain has been discovered outside of mammals; nor are such sequences evident in the current Fugu rubripes database (http://fugu.hgmp.mrc.ac.uk). We would like to put forth a different idea: that in shark, H and L chains are rearranging at the same time and are therefore subject to the same availability of processing factors. This may be a peculiarity of the multiple cluster organization, with 70 L chain clusters and ~50 H chain clusters (M. F. Flajnik and E. Hsu, unpublished results); an extended expression of RAG needed for a stepwise succession of H chain followed by L chain recombination may not be optimal to obtain a clonally expressed cell surface receptor. At least among the L chain genes there probably is not a hierarchy of activation: the abundance of nonproductive NS5 genomic rearrangements might be interpreted as argument for more than one L chain cluster and its allele activated and rearranging in any precursor lymphocyte. That is, those nonproductive NS5-2 may be largely from B cells expressing functional NS5-48, NS5-16, NS4 or NS3.

Another important factor in the formation of the junction is the sequence of the coding ends, which influences the resolution of the hairpin rearrangement intermediate; this aspect is discussed below.

Processing of coding region ends

The many N region-containing sequences may reflect shark TdT activity patterns or TdT accessibility to the coding ends. So far, it is only known that the presence of shark TdT, as in higher vertebrates, is limited to developing lymphocytes in the thymus and epigonal organs, and the signal decreases with age, probably in parallel with lymphopoiesis (25). However, it has been demonstrated that simply increasing levels of TdT does not affect the nature of the coding joints formed (26), and extended expression of TdT throughout B cell maturation, induced in TdT-transgenic mice, increased both the frequency and average number of N nucleotides, although only to 73% of {lambda}-chains and 44% of {kappa}-chains (27).

Coding-end hairpins from a variety of V, D, and J gene segments are invariably nicked to form potential 3' overhanging ends, and Schlissel (28) showed that hairpin intermediates of Ig gene segments are cleaved close to the coding end, mostly at a preferred site 2–4 bp from the end, depending upon the gene (or coding end sequence). The consistent pattern of nucleotide deletion of 3–4 bp in NS5 V and 1–2 bp in the J coding ends in three sharks probably reflects such a sequence-dependent component in processing. Just as the sequence composition of the coding end may influence the nicking site, it also determines the way the opened hairpin is processed. Many studies have shown that particular Ig genes/coding ends are prone to certain patterns of nucleotide deletion and P region formation, and this has been attributed to the coding-end sequence or its tertiary structure (29, 30, 31, 32). We suggest that, similarly, so is accessibility to TdT and that this influences the frequency of N region addition in shark L chains that is so much higher than in other animals.

CDR3 diversity

The NS5-2 CDR3 lengths range from 9 to 13 codons (Fig. 4), which is a fairly wide spectrum for L chain when compared with mouse {kappa}, where 90% of the CDR3 are 9 codons long. We have shown that in NS5, this is due entirely to processing of the coding ends. The nurse shark NS4 isotype makes up 80–90% of the L chains (our unpublished observations) and its CDR3 spectrum is similar, ranging from 8 to 14 codons (7), although some of this may be due to a 1- to 2-codon heterogenity among the various germline V gene segment flanks (10).

The amino acid sequence at the NS5-2 junctions is fairly variable. The most frequent CDR3 size is 11 codons (13 of the 29 in-frame CDR3 in Fig. 4), and among these 13 CDR3, there are only 2 that are the same (Table III), and the most variable position, 95B, is occupied by 9 different amino acids among the 13 sequences. By comparison, in 74 Xenopus rho L chain rearrangements, 89% of CDR3 are 9 codons long (Table II), and there is only a 2-aa variation at the joint, proline or glutamine, a result of uniform trimming patterns and little N region addition (17).


View this table:
[in this window]
[in a new window]
  		Table III. Translation of CDR3 junctions of NS5–2 cDNA from Fig. 4

 
The V region composition of 64 adult H chain cDNA clones has been recently described in detail (33), and their CDR3 mean length is 11.6 codons, with a range of 4–18 codons. This is somewhat longer than mouse (2–19 codons, mean 8.7 codons) (34) and Xenopus (3–12 codons, mean 8.6 codons) (35), whose respective major L chain isotypes are {kappa} and rho, as described above. The greater length of the average shark H chain CDR3 certainly is a result of rearrangement of the two germline D gene segments per locus, and perhaps one might speculate about a more prominent role for CDR3, in H or L chain, when there is only one VH gene family in nurse shark (33) and other elasmobranchs (3).

Conclusion

At first glance, one expects a more limited primary repertoire in cartilaginous fishes because the cluster organization prevents combinatorial diversity and the presence of preassembled V sequences would reduce overall junctional diversity. However, we have found an especially efficient hypermutation mechanism where one-half the substitutions are adjacent, virtually guaranteeing amino acid replacements with every hit (11). In this study, we report that processing of coding ends in the nurse shark L chains generates greater junctional diversity than observed in other vertebrate L chains.


   Acknowledgments
 
We thank Louis Du Pasquier for reading this manuscript and David Feliciano and Becky Lohr for technical assistance.


   Footnotes
 
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1 This work was supported in part by grants from the National Science Foundation (MCB 0080098) and the National Institutes of Health (GM068095). Back

2 Address correspondence and reprint requests to Dr. Ellen Hsu, Department of Physiology and Pharmacology, State University of New York, 450 Clarkson Avenue, Brooklyn, NY 11203. E-mail address: hsue{at}hscbklyn.edu Back

3 Abbreviations used in this paper: RSS, recombination signal sequence; TdT, terminal deoxynucleotidyltransferase. Back

Received for publication March 8, 2004. Accepted for publication August 30, 2004.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Tonegawa, S.. 1983. Somatic generation of antibody diversity. Nature 302:575.[Medline]
  2. Lewis, S. M.. 1994. The mechanism of V(D)J joining: lessons from molecular, immunological, and comparative analyses. Adv. Immunol. 56:27.[Medline]
  3. Yoder, J. A., G. W. Litman. 2000. Immune-type diversity in the absence of somatic rearrangement. Curr. Top. Microbiol. Immunol. 248:271.[Medline]
  4. Hinds, K. R., G. W. Litman. 1986. Major reorganization of immunoglobulin VH segmental elements during vertebrate evolution. Nature 320:546.[Medline]
  5. Kokubu, F., R. Litman, M. J. Shamblott, K. Hinds, G. W. Litman. 1988. Diverse organization of immunoglobulin VH gene loci in a primitive vertebrate. EMBO J. 7:3412.
  6. Greenberg, A. S.. 1994. Evolution of the antigen receptor family. Doctoral thesis University of Miami, Miami, FL.
  7. Lee, S. S., A. Greenberg, E. Hsu. 2000. Evolution and somatic diversification of immunoglobulin light chains. Curr. Top. Microbiol. Immunol. 248:285.[Medline]
  8. Rast, J. P., M. Anderson, T. Ota, R. T. Litman, M. Margittai, M. J. Shamblott, G. W. Litman. 1994. Immunoglobulin light chain class multiplicity and alternative organizational forms in early vertebrate phylogeny. Immunogenetics 40:83.[Medline]
  9. Greenberg, A. S., L. Steiner, M. Kasahara, M. F. Flajnik. 1993. Isolation of a shark immunoglobulin light chain cDNA clone encoding a protein resembling mammalian {kappa} light chains: implications for the evolution of light chains. Proc. Natl. Acad. Sci. USA 90:10603.[Abstract/Free Full Text]
  10. Lee, S. S., D. Fitch, M. F. Flajnik, E. Hsu. 2000. Rearrangement of immunoglobulin genes in shark germ cells. J. Exp. Med. 191:1637.[Abstract/Free Full Text]
  11. Lee, S. S., D. S. Tranchina, Y. Ohta, M. F. Flajnik, E. Hsu. 2002. Hypermutation in shark immunoglobulin light chain genes results in contiguous substitutions. Immunity 16:571.[Medline]
  12. Walsh, C. J., C. A. Luer. 1998. Comparative phagocytic and pinocytic activities of leucocytes from peripheral blood and lymphomyeloid tissues of the nurse shark (Ginglymostoma cirratum Bonaterre) and the clearnose skate (Raja eglanteria Bosc). Fish Shellfish Immunol. 8:197.
  13. Kabat, E. A., T. T. Wu, H. M. Perry, K. S. Gottesman, C. Foeller. 1991. Sequences of Proteins of Immunological Interest 5th Ed. U.S. Department of Health and Human Services, Washington, DC.
  14. Shamblott, M. J., G. W. Litman. 1989. Genomic organization and sequences of immunoglobulin light chains in a primitive vertebrate suggest coevolution of immunoglobulin gene organization. EMBO J. 8:3733.[Medline]
  15. Miracle, A. L., M. K. Anderson, R. T. Litman, C. J. Walsh, C. A. Luer, E. V. Rothenberg, G. W. Litman. 2001. Complex expression patterns of lymphocyte-specific genes during the development of cartilaginous fish implicate unique lymphoid tissues in generating an immune repertoire. Int. Immunol. 13:567.[Abstract/Free Full Text]
  16. Hohman, V. S., D. B. Schuchman, S. F. Schluter, J. J. Marchalonis. 1989. Genomic clone for the sandbar shark {lambda} light chain: generation of diversity in the absence of gene rearrangement. Proc. Natl. Acad. Sci. USA 90:9882.
  17. Ji, Y., S. Desravines, E. Hsu. 1999. Junctional diversity in Xenopus immunoglobulin light chains. Mol. Immunol. 36:1159.[Medline]
  18. Sehgal, D., G. Johnson, T. T. Wu, R. G. Mage. 1999. Generation of the primary antibody repertoire in rabbits: expression of a diverse set of Ig{kappa}-V genes may compensate for limited combinatorial diversity at the heavy chain locus. Immunogenetics 50:31.[Medline]
  19. Victor, K. D., K. Vu, A. J. Feeney. 1994. Limited junctional diversity in {kappa} light chains. Junctional sequences from CD43+B220+ early B cell progenitors resemble those from peripheral B cells. J. Immunol. 152:3467.[Abstract]
  20. Bentolila, L. A., G. E. Wu, F. Nourrit, M. Fanton d’Andon, F. Rougeon, N. Doyen. 1997. Constitutive expression of terminal deoxynucleotidyltransferase in transgenic mice is sufficient for N region diversity to occur at any Ig locus throughout B cell differentiation. J. Immunol. 158:715.[Abstract]
  21. Klein, U., K. Rajewsky, R. Kuppers. 1999. Phenotypic and molecular characterization of human peripheral blood B-cell subsets with special reference to N-region addition and J{kappa}-usage in V{kappa}J{kappa}-joints and {kappa}/{lambda}-ratios in naive versus memory B-cell subsets to identify traces of receptor editing processes. Curr. Top. Microbiol. Immunol. 246:141.[Medline]
  22. Bridges, S. L., Jr. 1998. Frequent N addition and clonal relatedness among immunoglobulin {lambda} light chains expressed in rheumatoid arthritis synovia and PBL, and the influence of V{lambda} gene segment utilization on CDR3 length. Mol. Med. 4:525.[Medline]
  23. Jenne, C. N., L. J. Kennedy, P. McCullagh, J. D. Reynolds. 2003. A new model of sheep Ig diversification: shifting the emphasis toward combinatorial mechanisms and away from hypermutation. J. Immunol. 170:3739.[Abstract/Free Full Text]
  24. Gilfillan, S., C. Benoist, D. Mathis. 1995. Mice lacking terminal deoxynucleotidyl transferase: adult mice with a fetal antigen receptor repertoire. Immunol. Rev. 148:201.[Medline]
  25. Rumfelt, L. L., D. Avila, M. Diaz, S. Bartl, E. C. McKinney, M. F. Flajnik. 2001. A shark antibody heavy chain encoded by a nonsomatically rearranged VDJ is preferentially expressed in early development and is convergent with mammalian IgG. Proc. Natl. Acad. Sci. USA 98:1775.[Abstract/Free Full Text]
  26. Lieber, M. R., J. E. Hesse, K. Mizuuchi, K., and M. Gellert. 1988. Lymphoid V(D)J recombination: nucleotide insertion at signal joints as well as coding joints. Proc. Natl. Acad. Sci. USA 85:8588.
  27. Bentolila, L. A., S. Olson, A. Marshall, F. Rougeon, C. J. Paige, N. Doyen, G. E. Wu. 1999. Extensive junctional diversity in Ig light chain genes from early B cell progenitors of µ MT mice. J. Immunol. 162:2123.[Abstract/Free Full Text]
  28. Schlissel, M. S.. 1998. Structure of nonhairpin coding-end DNA breaks in cells undergoing V(D)J recombination. Mol. Cell. Biol. 18:2029.[Abstract/Free Full Text]
  29. Meier, J. T., S. M. Lewis. 1993. P nucleotides in V(D)J recombination: a fine-structure analysis. Mol. Cell. Biol. 13:1078.[Abstract/Free Full Text]
  30. Boubnov, N. V., Z. P. Wills, D. T. Weaver. 1995. Coding sequence composition flanking either signal element alters V(D)J recombination efficiency. Nucleic Acids Res. 23:1060.[Abstract/Free Full Text]
  31. Nadel, B., A. J. Feeney. 1997. Nucleotide deletion and P addition in V(D)J recombination: a determinant role of the coding-end sequence. Mol. Cell. Biol. 17:3768.[Abstract]
  32. Ezekiel, U. R., T. Sun, G. Bozek, U. Storb. 1997. The composition of coding joints formed in V(D)J recombination is strongly affected by the nucleotide sequence of the coding ends and their relationship to the recombination signal sequences. Mol. Cell. Biol. 17:4191.[Abstract]
  33. Rumfelt, L. L., R. L. Lohr, H. Dooley, M. F. Flajnik. 2004. Diversity and repertoire of IgW and IgM VH families in the newborn nurse shark. BMC Immunol. 5:8.[Medline]
  34. Wu, T. T., G. Johnson, E. A. Kabat. 1993. Length distribution of CDRH3 in antibodies. Protein Struct. Funct. Genet. 16:127.
  35. Desravines, S., E. Hsu. 1994. Measuring CDR3 length variability in individuals during ontogeny. J. Immunol. Methods 168:219.[Medline]



This article has been cited by other articles:


Home page
 J. Immunol.Home page
V. Lee, J. L. Huang, M. F. Lui, K. Malecek, Y. Ohta, A. Mooers, and E. Hsu
The Evolution of Multiple Isotypic IgM Heavy Chain Genes in the Shark
J. Immunol., June 1, 2008; 180(11): 7461 - 7470.
[Abstract] [Full Text] [PDF]

Home page
 J. Immunol.Home page
E. Hsu and M. F. Criscitiello
Diverse Immunoglobulin Light Chain Organizations in Fish Retain Potential to Revise B Cell Receptor Specificities
J. Immunol., August 15, 2006; 177(4): 2452 - 2462.
[Abstract] [Full Text] [PDF]

Home page
 J. Immunol.Home page
K. Malecek, J. Brandman, J. E. Brodsky, Y. Ohta, M. F. Flajnik, and E. Hsu
Somatic Hypermutation and Junctional Diversification at Ig Heavy Chain Loci in the Nurse Shark
J. Immunol., December 15, 2005; 175(12): 8105 - 8115.
[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 Fleurant, M.
Right arrow Articles by Hsu, E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Fleurant, M.
Right arrow Articles by Hsu, E.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS