V(D)J recombination in zebrafish: Normal joining products with accumulation of unresolved coding ends and deleted signal ends

Abstract

V(D)J recombination proceeds from a site-specific cleavage to an imprecise end joining, via generation and resolution of recombination ends. Although rearranged antigen receptor genes isolated from zebrafish (Danio rerio) resemble those made in mammals, differences may arise during evolution from lower to higher vertebrates, in regard to efficiency, fidelity and regulation of this recombination. To elucidate the V(D)J recombination reaction in zebrafish, we characterized recombination ends transiently produced by zebrafish lymphocytes, as well as joining products. Similar to their mammalian counterpart, zebrafish lymphocytes make perfect signal joints and normal coding joints, indicating their competent end resolution machinery. However, recombination ends recovered from the same zebrafish lymphoid tissues exhibit some features that are not readily seen in normal mammalian counterpart: deleted signal ends and accumulation of opened coding ends. These results indicate that the recombination reaction in zebrafish lymphocytes is inefficient and less stringently regulated, which may result from unstable post-cleavage complexes, and/or slow transition from cleavage to resolution. Our data suggests that the V(D)J recombination machinery may have undergone evolution selection to become more efficient in higher jawed vertebrates.

Fig. 1

Rearrangement of immunoglobulin (Ig) genes in zebrafish lymphoid tissues. A. Outline of zebrafish IgH and IgL gene loci, as well as location of primers that are listed in Table 1. These primers, denoted as small arrows, are utilized in PCR amplification of recombination ends and joints, as shown in Table 2. B. Diagram illustrating the production of signal ends (SEs), coding ends (CEs), coding joints (CJs) and signal joints (SJs), as well as their detection by genomic PCR (for CJs and SJs) and ligation-mediated PCR (LM-PCR for SEs and CEs). The V, D or J gene segments (rectangular boxes) are flanked by recombination signal sequences (RSSs, triangles). C. Analysis of IgH recombination status. Genomic DNA was prepared from spleen (S) and kidney (K), J_H-SEs were examined by a semi-nested LM-PCR (see Table 2 for primer selection). Rearranged VDJ_H- and DJ_H-CJs or DJ_H-SJs were amplified by genomic PCR. Germline heavy chain (G_H) was included as a control for DNA input.

Fig. 2

Examination of fidelity of SEs and SJs by ApaL1 digestion. A. Illustration of the susceptibility of SE LM-PCR (left panel) and SJ PCR (right panel) products to ApaL1 restriction enzyme. B. ApaL1 digestion of SE and SJ PCR products. Genomic DNA was prepared from recombinase-active murine cell lines (Sp9 and S9), and zebrafish kidney. The PCR products of SEs and SJs, derived from IgL gene loci (Danilova et al., 2000; Haire et al., 2000) were incubated overnight with or without restriction endonuclease ApaL1, and analyzed by gel electrophoresis. The down-shift of PCR products indicates a digestion by ApaL1, reflecting the presence of the perfect junction of SJs or full-length SEs. Arrow indicates expected sized of zebrafish J_L-SEs (∼280bps). C. Effect of zebrafish kidney tissues on the integrity of SEs made from a recombination-active murine cell line (SP9). Genomic DNA prepared from zebrafish kidney, K(z), a wild-type recombination-active murine cell line, Sp9 (m), and the mixture of zebrafish kidney and murine cells, Mix, were examined for mouse IgL-SEs (Vλ) or zebrafish IgL-SEs (J_L) by LM-PCR. The PCR products were treated with (+) or without (−) restriction endonuclease ApaL1 and analyzed by gel electrophoresis. The down-shift of PCR products indicates digestion by ApaL1, reflecting full-length SEs. D. Inability of nuclease inhibitor (aurintricarboxylic acid) to rescue the integrity of zebrafish J_L-SEs. LM-PCR was performed on the DNA plugs prepared from zebrafish kidney cells in the presence of aurintricarboxylic acid. The PCR products were treated with or without ApaL1 endonuclease, and separated on 1.2% agarose gel (lane 1-2). As a positive control for ApaL1 digestion, LM-PCR from mouse IgL-SE was included (lane 3-4), in which a down-shift of the band was noticed. E. IgL rearrangement in day 14 zebrafish embryos. Genomic DNA was prepared from zebrafish embryos. The PCR products amplified from VJ_L-SJs, J_L-SEs were treated with or without restriction endonuclease ApaL1.

Fig. 3

Southern blot analysis of PCR products amplified from recombination ends and joints made in zebrafish kidney and mouse bone marrow. A. Analysis of recombination ends and joints present in total mouse bone marrow cells. Genomic DNA prepared in DNA plugs was treated with or without T4 DNA polymerase before being analyzed by LM-PCR. Semi-quantitative LM-PCR (two different dilutions: 1 and 1:3) followed by Southern blot analysis was performed to detect Jκ-SEs and Jκ-CEs, VJκ-CJs and germline (G_L) controls. S9, a ts-Ab-MLV pre-B cell line, derived from DNA-PKcs deficient scid mice (Chang and Brown, 1999) serves as a positive control for the detection of unresolved opened CEs. B-C. Analysis of recombination ends and joints in zebrafish kidney. PCR and LM-PCR followed by Southern blot was performed on zebrafish DNA samples to amplify J_H-SEs, J_H-CEs, DJ_H-SJs, DJ_H-CJs and G_H of the IgH-locus (B) and J_L-SEs, J_L-CEs VJ_L-SJs, VJ_L-CJs and G_L of the IgL-locus (C).