Recombination-Activating Gene 1 and 2 (RAG1 and RAG2) in Flounder (Paralichthys Olivaceus)

Recombination-activating gene 1 and 2 (RAG1 and RAG2) in flounder (Paralichthys olivaceus)

1,2 1, 1 1 1 XIANLEI WANG ; XUNGANG TAN *; PEI-JUN ZHANG ; YUQING ZHANG and PENG XU 1Key Laboratory of Experimental Marine Biology, Institute of Oceanology, Chinese Academy of Sciences, 7 Nanhai Road, Qingdao 266071, China 2National Oceanographic Center, 88 Xuzhou Road, Qingdao, Shandong 266071, China

*Corresponding author (Email, [email protected])

During the development of B and T lymphocytes, Ig and TCR variable region genes are assembled from germline V, D, and J gene segments by a site-specific recombination reaction known as V(D)J recombination. The process of somatic V(D)J recombination, mediated by the recombination-activating gene (RAG) products, is the most significant characteristic of adaptive immunity in jawed vertebrates. Flounder (Paralichthys olivaceus) RAG1 and RAG2 were isolated by Genome Walker and RT-PCR, and their expression patterns were analysed by RT-PCR and in situ hybridization on sections. RAG1 spans over 7.0 kb, containing 4 exons and 3 introns, and the full-length ORF is 3207 bp, encoding a peptide of 1068 amino acids. The first exon lies in the 5′-UTR, which is an alternative exon. RAG2 full-length ORF is 1062 bp, encodes a peptide of 533 amino acids, and lacks introns in the coding region. In 6-month- old flounders, the expression of RAG1 and RAG2 was essentially restricted to the pronephros (head kidney) and mesonephros (truck kidney). Additionally, both of them were mainly expressed in the thymus. These results revealed that the thymus and kidney most likely serve as the primary lymphoid tissues in the flounder.

[Wang X, Tan X, Zhang P-J, Zhang Y and Xu P 2014 Recombination-activating gene 1 and 2 (RAG1 and RAG2) in flounder (Paralichthys olivaceus). J. Biosci. 39 849–858] DOI 10.1007/s12038-014-9469-1

1. Introduction two tightly linked genes in a single murine genomic DNA fragment (Schatz et al. 1989). Both of RAG1 and RAG2 are The mammalian immune system can be divided into the innate essential for V(D)J recombination (Mombaerts et al. 1992; immune system and the adaptive immune system. T and B Shinkai et al. 1992), and somatic V(D)J recombination is lymphocytes are the main features of the adaptive immune mediated by the RAG gene products, a process that occurs system and originate from common pluripotent stem cells specifically in immature lymphocytes. During this process, a (Ikuta et al. 1992). The primary immunoglobulin (Ig) and complex of RAG1 and RAG2 proteins specifically recog- antigen receptor (TCR) repertoires for B and T cells are as- nizes and binds to recombination signal sequences (RSS), sembled from the component V (variable), D (diversity), and J which flank the V, D, and J gene segments, to initiate (joining) gene segments by site-specific recombination, termed recombination by introducing DNA double-stranded breaks V(D)J recombination (Schatz et al. 1992), to form continuous (DSBs) (Shlyakhtenko et al. 2009). The recruitment of DSB functional genes as the B and T cells mature. repair proteins, including DNA PKcs, KU70, Ku80, XRCC4 V(D)J recombination is activated by the recombination- and DNA ligase 4, then mediates the imprecise joining of the activating genes, RAG1 and RAG2 (Oettinger et al. 1990), two coding ends into a coding joint, whereas the precise

Keywords. Flounder; RAG1; RAG2; thymus

Supplementary materials pertaining to this article are available on the Journal of Biosciences Website at http://www.ias.ac.in/jbiosci/ dec2014/supp/Wang.pdf http://www.ias.ac.in/jbiosci J. Biosci. 39(5), December 2014, 849–858, * Indian Academy of Sciences 849

Published online: 20 October 2014 850 Xianlei Wang et al. joining of the two signal ends produces a signal joint (Du cultured flounder. Although flounder RAG2 had been Pasquier 1992; Grawunder et al. 1998). Also, the RAG1- recombinant-expressed in E. coli (Wang et al. 2012), the RAG2 complex plays a role in immunoglobulin allelic organ- and lymphocyte-specific characterization of RAGs pairing, and in determining which recombination methods has never been performed in flounder. Thus, to elucidate will be used to repair the broken ends generated during the ontogeny of the lymphocytes and lymphoid organs in V(D)J recombination (Matthews and Oettinger 2009). flounder, RAG1 and RAG2 were cloned from flounder, and The RAG genes have been cloned from a variety of the organ- and lymphocyte-specific expression was species (Schatz et al. 1989; Carlson et al. 1991; Greenhalgh investigated. et al. 1993; Hansen 1997; Bernstein et al. 1996; Hansen and Kaattari 1995, 1996; Willett et al. 1997; Peixoto et al. 2000; 2. Materials and methods Corripio-Miyar et al. 2007; Zhang et al. 2012; Covello et al. 2013), and a striking feature of the RAG locus was that the RAG genes are highly conserved between species. In fact, in 2.1 Ethics statement all of the vertebrates for which the entire locus has been studied, RAG1 and RAG2 are tightly linked and are Experiments with flounder were performed according to the convergently transcribed (Willett et al. 1997). The genomic regulations of local and central governments. All of the structure of RAG1 was not conserved. A different number of experiments were approved by the Institutional Animal Care introns was found in the teleost RAG1 genomic coding and Use Committee of Institute of Oceanology, Chinese region; zebrafish and Japanese pufferfish each contain two Academy of Science. introns in their RAG1 coding regions (Willett et al. 1997; Peixoto et al. 2000), whereas rainbow trout has one intron in 2.2 Animals the RAG1 coding region (Hansen and Kaattari 1995). In contrast, other vertebrates, such as humans, rabbits, mice, Flounder were cultured at the Institute of Oceanology, Chi- chickens and Xenopus, have no introns in their RAG1 coding nese Academy of Sciences and a fish farm (Homey Group, region. No introns were found in the RAG2 coding region of Rongcheng City, Shandong Province) under controlled con- any of the organisms studied till date (Peixoto et al. 2000). ditions (a 14 h light:10 h dark photoperiod; temperature of The thymus, kidney and spleen are the major lymphoid 15±1°C; seawater; aeration). The fish were fed a commercial organs of teleosts, and the ontogeny of these organs in particle diet twice daily. teleosts has been widely studied in different species, including yellowtail, flounder, seabream and turbot (Willett et al. 1997; Zapata et al. 2006; Hansen and Zapata 1998; Romano 2.3 Isolation of flounder RAG1 and RAG2 genes et al. 1999; Josefsson and Tatner 1993; Pulsford et al. 1994; Padrós and Crespo 1996), using morphological methods. The genomic sequences of flounder the RAG1 and RAG2 However, immunity genes are also good markers for the genes were isolated using degenerate PCR and study of the development of the immune system. Genes, GenomeWalker methods (Zhang et al. 2006). Based on the such as Ikaros, Ig, RAG1 and RAG2, involved in the differ- conserved regions of the RAG1 and RAG2 nucleotide se- ent developmental stages of lymphocytes have been used to quences from human, mouse, chicken, Xenopus, shark, follow the development of lymphoid organs in fish (Lam zebrafish, trout and pufferfish (the RAG1 GenBank acces- et al. 2002; Danilova et al. 2004). The expression of RAG1 sion numbers are M29474, P15919, P24271, Q91829, and RAG2 are independent of one another (Carlson et al. AAB17267.1, U71093, Q91187 and AAD20561.1, respec- 1991; Greenhalgh et al. 1993; Chun et al. 1991), and yet tively; the RAG2 GenBank accession numbers are P55895, both genes are expressed together in maturing B and T M64796, P25022, Q91830, AY172838, U71094, Q91193 lymphocytes. Therefore, the expression of the RAG genes and AAD20562.1, respectively), nested degenerate PCR can be used to monitor the appearance and location of pre-B primers (R1F1, R1R1, R1F2, R1R2; R2F1, R2R1, R2F2 and pre-T cells during the development of the immune and R2R2) were designed to amplify a region from exon 4 system (Oettinger et al. 1990; Greenhalgh et al. 1993). of the RAG1 gene and a 5′ region of the RAG2 gene using Flounder is an important economic fish species in mari- genomic DNA as the template. The PCR program was 5 min culture and has been extensively cultured under intensive at 94°C, 40 cycles of 40 s at 94°C, 40 s at 58°C, and 1 min systems in the northern coastal regions of China. However, (for RAG1) or 2 min (for RAG2) at 72°C, followed by 10 min the frequent occurrence of diseases has caused enormous at 72°C. The remaining portions of RAG1 and RAG2 were losses in recent years. Therefore, to maintain the stable, isolated by several rounds of PCR using RAG1- or RAG2- healthy and sustainable development of aquaculture, more specific primers (R1F3, R1F4, R1R3, R1R4, R1R5, R1R6, attention is being given to the immune mechanisms in R1R7, R1R8; R2F3, R2F4, R2F5, R2F6, R2R3, R2R4,

J. Biosci. 39(5), December 2014 Recombination-activating genes from flounder 851

R2R5 and R2R6; table 1) together with the adapter primers and are in the units of the number of amino acid substitutions (AP1 and AP2) (Clontech, USA) from the flounder per site. The analysis involved 11 amino acid sequences for GenomeWalker libraries (Clontech, USA) (Zhang et al. RAG1 and 10 amino acid sequences for RAG2. All positions 2006). All of the fragments were cloned into the pUCm-T containing gaps and missing data were eliminated. There were vector (Sangon, Shanghai, China) and sequenced. a total of 973 for RAG1 and 508 positions for RAG2 in the final dataset. 2.4 Prediction of lymphocyte-specific transcription factors binding sites 2.6 Reverse transcription-polymerase chain reaction (RT-PCR) To analysis lymphocyte specific transcription factors binding sites in the alternatively 5′-UTR of RAG1,P-match Total RNA extraction was performed using the TRIzol re- program was used (http://www.gene-regulation.com/cgi- agent, following the manufacturer’s instructions (Invitrogen, bin/pub/programs/pmatch/bin/p-match.cgi) and database CA, USA). The total RNA from different tissues of six- were TRANSFAC® Public 6.0. month-old fish were extracted and treated with RQ1 RNase-free DNase (Promega, WI, USA) for 30 min at 2.5 Phylogenetic tree construction 37°C. The reactions were stopped with the Stop Solution for 10 min at 65°C. First-strand cDNA was then synthesized using M-MLV reverse transcriptase (Promega, Madison, WI, Sequence comparison was done using the software clustalw2 USA) and RAG1- or RAG2-specific primers (RAG1, R1F3, (http://www.ebi.ac.uk/Tools/msa/clustalw2/). Phylogenetic and R1R10; RAG2, R2F8, R2R7). The cycling conditions were molecular evolutionary analyses were conducted using MEGA 5 min at 94°C, 40 cycles of 50 s at 94°C, 30 s at 57°C, and version 6 (Tamura et al. 2013). The evolutionary history was 1 min at 72°C, followed by 10 min at 72°C. inferred using the neighbour-joining method (Saitou and Nei, 1987). The optimal trees with the sum of branch length = 1.52531262 for RAG1 and 1.92066380 for RGA2 were shown. 2.7 Confirmation of the intron/exon boundary The tree is drawn to scale, with branch lengths in the same units and the 5′-UTR of RAG1 as those of the evolutionary distances used to infer the phylogenetic tree. The evolutionary distances were computed using To verify the position of the RAG1 introns, RT-PCR was the Poisson correction method (Zuckerkandl and Pauling 1965) performed using two pairs of specific primers spanning the

Table 1. Genome Walker primers and RT-PCR primers

Primer Sequence(5′-3') Primer Sequence(5′-3')

R1F1 CACTGTGAB ATYGGCAWYGCVRC R2F1 TTHGGVCARAARGGVTGGCCDARRMG R1F2 GAGATHGGDGARGTVTAYMA R2F2 ARRMGMTCMTGYCCHACBGG R1F3 CTACAGCTTCAACTCCCAGCGCTTTG R2F3 TACTTTGGTCGTGAGCCACATGAG R1F4 AACCAACTACCTCCACAAGACCCTGG R2F4 AGGACGAGGAGGACGAATCACAGAC R1F10 TTCTCCTCTCCAGGGTCTAAGATC R2F5 GCTACTGGATCAAATGTTCCCTGGGC R1F11 TAACGCTCATTGCCTGGACCATGAAC R2F6 CCATACTACTCCACTGAGCTCCACC R1R1 GADGTRTABAGCCARTGRTGYTT R2F8 TTTGAGGCGTTGGAGTCACCTAAC R1R2 GAYTGVCKBGCRTTCATYTT R2R1 HTCCATRCAYTGBCMATGNACCC R1R3 TCCTCATTATCGGTTTGAGCTTC R2R2 TGBCMATGNACCCARTGBCC R1R4 CTTCTTCCTCAGCTGTTTATCC R2R3 AGGTCAAACAGGAACACCTGAG R1R5 GAAAGTGTTGACTCGAATGGCCAAGC R2R4 ACGCTATTCCAGCTCTCTGTGGTG R1R6 TGCAGAGAGGGAGAGGAAGAAAGACA R2R5 AAGTTCACCATCCAGCTGCAACAGAG R1R7 TCATAACACTCCCTGGACCCAAAC R2R6 GTCGACTTTCCAACAGGTGCTTGCGTC R1R8 GCAATAGGTGAGCTCTCTAACGAGAC R2R7 GCATCCTATAGGAAATGATTCC R1R10 GGCTTTAACTCCAGCTTCTACTCG AP1 GTAATACGACTCACTATAGGGC R1R11 GGCTGCCATTCAAACTGGTGAAAG AP2 ACTATAGGGCACGCGTGGT R1R14 GACACCTTTACTCTCCTCTAG Actin1 AGAGCAAGAGAGGCATCCTGAC Actin2 CGATGGGTGATGACCTGTCC

J. Biosci. 39(5), December 2014 852 Xianlei Wang et al. expected location of both introns (R1F10, R1R14; R1F11 alkaline phosphate buffer (0.1 M Tris, 0.1 M NaCl, 0.05 M and R1R11; table 1). MgCl2, and 0.1% Tween-20, pH 9.5), the samples were To obtain the 5′-UTR of RAG1,a5′ rapid amplification incubated in the dark with BCIP/NBT substrate solution of cDNA ends (RACE) library was constructed using the (Sangon, Shanghai, China) and monitored by microscopy SMART RACE cDNA Amplification Kit (Clontech, CA, until the desired level of staining was reached. To identify USA). The 5′-UTR of RAG1 was cloned using the RACE different tissues, the sections were counterstained with a program described in the SMART RACE cDNA Amplifica- water-eosin solution after hybridization. The paraffin section Kit (Clontech, CA, USA) with specific primers (R1R7 tions were photographed using a microscope (DM LB2, and R1R8; table 1) and the UPM primer (Clontech, CA, Leica) and a Nikon 4500 Digital camera. USA). 3. Results 2.8 Preparation of paraffin sections and in situ hybridization 3.1 Characterization of flounder RAG1, RAG2 and their genomic structure The plasmid clones that contained a portion of the exon 4 sequence of RAG1 (539 bp, nt 2536–3174) and a portion of The flounder RAG1 and RAG2 genes were isolated by nested RAG2 (1270 bp, nt 112–1381) were used as the templates to degenerate PCR using flounder genomic DNA and Genome generate antisense digoxigenin-labelled RNA probes with Walker libraries (see Section 2). The flounder RAG1 geno- T7 RNA polymerase. The RNA probes were generated by mic sequence was found to be approximately 4.1 kb (from in vitro transcription in the presence of digoxigenin-11- ATG to Stop Code) in size (GenBank No. KF537386) and UTP* (Roche, Mannheim, Germany). contained two introns (89 and 859 bp) and three exons (332, The in situ hybridization of paraffin sections was per- 1133 and 1742 bp) in the coding region, with conserved formed following the protocol from Lam et al. (2004) with GT…AG sequences at the exon/intron boundaries. The some modifications. The fixed fish tissues were dehydrated flounder RAG1 gene was predicted to encode a protein of sequentially in ethanol and xylene and embedded in paraffin 1068 amino acids. An alternatively spliced exon in the 5′- (Leica TP1020 and EG1160, Leica, Germany). Sections UTR was also identified; the two alternative exons were 132 (7 μm thick) were prepared (Leica RM2145, Leica, Germa- and 138 bp long (figure 1). ny), and some of the sections were hematoxylin-eosin The comparison of the predicted amino acid sequence of stained with Thermo Shandon Varistain (24-4) Automatic the flounder RAG1 with those of other species revealed that Slide Stainer (Thermo Shandon, Cheshire, England), and it shared a higher identity with teleosts than with other others were subjected to in situ hybridization. For the in situ species. For example, flounder RAG1 shared 79%, 72%, hybridization, the paraffin sections were rehydrated sequen- 71%, and 75% identity with the RAG1 sequences of fugu, tially in xylene and ethanol; after fixing for 10 min in 4% zebrafish, carp and trout, respectively. The C-terminal region PFA, the sections were washed 3 times for 5 min in PBST. of RAG1 was more highly conserved, sharing more than The sections were then treated with HCl (0.2 mol/L) for 70% identity with the other teleosts. 10 min and with proteinase K (10 μg/mL in PBS) for 10 The open reading frame of flounder RAG2 was 1602 bp min. After fixing for 5 min in 4% PFA, the sections were (GenBank No. KF537387), encoding a predicted protein of washed 5 times for 10 min in PBST and incubated for 2–4h 533 amino acids; no intron was found in the coding region. in pre-hybridization buffer (with tRNA and heparin) at 70°C Flounder RAG2 shared 82%, 69%, 70%, and 71% identity in a wet box containing 2× SSC. The samples were then with the RAG2 sequences of fugu, zebrafish, carp and trout, incubated at 70°C with a pre-heated DIG-labelled probe respectively, whereas only approximately 50% identity was (0.5–1.0 ng/mL) in hybridization buffer overnight in the observed with non-teleosts species. In addition, the non- wet box. The samples were then washed continuously for conserved sites within RAG2 were scattered throughout the 15 min (2 times) with 50% pre-hybridization mixture (no entire reading frame. tRNA and heparin) in 1× SSC and for 30 min (3 times) with The recombination-activating genes, RAG1 and RAG2, 0.2× SSC containing 0.1% CHAPS at 70°C. After the sam- were convergently transcribed (figure 1) in flounder. The ples cooled to room temperature, they were washed for 5 min 3′-UTR of RAG2 shared a common overlapping with 1× MAB (100 mM maleic acid and 150 mM NaCl, pH polyadenylation site with RAG1, and the intergenic region 7.5). After blocking for 2–4 h at RT with blocking buffer was 3128 bp long, which is considerably shorter than that of (10% goat serum and 2% BMB in 1× MAB), the samples tetrapods (approximately 10 kb) but comparable to that of were incubated with anti-digoxigenin-AP (Roche, Mann- other teleosts (2–3 kb). heim, Germany) in blocking buffer at 4°C overnight. After It is interesting to find that an alternatively spliced exon1 washing 6 times for 15 min in PBST and 10 min (2 times) in (132 bp and 138 bp) in the 5′-UTR of RAG1 (figure 1). In

J. Biosci. 39(5), December 2014 Recombination-activating genes from flounder 853

Figure 1. Schematic diagram of the structure of the RAG genes in flounder. these two exons, some lymphocyte specific transcription In situ hybridization was performed on paraffin sections factor binding sites were found such as NF-κB, C/EBP, using RAG1 and RAG2 probes and showed that both genes E47 and c-Ets-1 binding sites (figure 2A, B). were specifically expressed in the thymus and pronephros (figure 3A). The cortex was deeply stained, and many small and medium thymocytes were densely packed in a loose 3.2 Phylogenetic tree construction network. The lighter-staining medulla consisted of thymocytes scattered throughout a dense framework of stromal A comparison of the flounder RAG1 and RAG2 sequences cells (figure 5B–C). with those of other known vertebrate RAG1 and RAG2 revealed that they share the highest sequence similarity to that of RAG1 and RAG2, respectively (supplementary 4. Discussion figure 1). A phylogenetic analysis using RAG1 and RAG2 showed that flounder was more closely related to other In this study, we identified and characterized the flounder teleosts than to other species (figure 3). RAG1 and RAG2 genes. Their genomic sequences and predicted amino acid sequences were compared with those of other species, and the expression of RAG1 and RAG2 in 3.3 Expression of RAG1 and RAG2 in adult fish adult tissues was analysed by RT-PCR and in situ hybridization. The organ-specific expression in 6-month-old flounders was RAG1 and RAG2, which are convergently transcribed, are analysed by RT-PCR. The expression of RAG1 and RAG2 separated by an intergenic region that varies in size among was restricted to the pronephros (head kidney) and meso- species, for example, approximately 11 kb in humans (Homo nephros (truck kidney) (figure 4B). There is no expression in sapiens),8kbinmice(Mus musculus),5.2kbinfrogs the liver, spleen, heart, muscle or intestine. (Xenopus laevis), 2.8 kb in rainbow trout (Oncorhynchus

GCAAACTGAGACAGACTCAAACGTATACACTTGTATACGTGGCACGCAGGGCCCAGCTCCAGGGC

TTCCCTTGTGCTCCCATGCGGAGAGAACCCCCAGGGGAGCCGACGCCAGCTGAGACTGAAAAACA

(A)132bp exon-1

GAGTTCAGCAAGTGAAGTACACACAGAGAGCAGGGGGGGAAATGAGGGAGACAGCCAGATTCCAG

CCTCCCAGGGAGAAAATGTAGGACTAAAAGGATGGATAGACGTTGGACGTGTGTGCACTACAGGT

GTGCAATG

(B)138bp exon-1

Figure 2. Putative lymphocyte specific transcription factors binding sites in alternatively spliced exon1 of RAG1 ( NF-κB, C/EBP, E47, c-Ets-1).

J. Biosci. 39(5), December 2014 854 Xianlei Wang et al. (A) (B) T.rubripes O.mykiss O.mykiss D.rerio T.rubripes H.hippoglossus D.rerio C.carpio P.olivaceus C.carpio P.olivaceus

C.leucas C.plumbeus H.sapiens H.sapiens

M.musculus M.musculus X.laevis X.laevis 0.05 G.gallus 0.05 G.gallus

Figure 3. Phylogenetic tree of flounder RAG relative to RAG of other vertebrates. (A) RAG1 GenBank accession number of RAG1: Flounder (P. olivaceus, KF537386), pufferfish (T. rubripes, AAD20561.1, AF108420), trout (O. mykiss, Q91187), zebrafish (D. rerio, U71093), carp (C. carpio AAX16495.1, AY787040.1), shark (C. leucas, AAB17267.1, U62645.1), Xenopus (X. laevis, Q91829), chicken (G. gallus, P24271), mouse (M. musculus, P15919), human (H. sapiens, M29474)sea urchin (S. purpuratus, DQ082723). (B) RAG2 GenBank accession number of RAG2: Flounder (P. olivaceus, KF537387), pufferfish (T. rubripes, AAD20562.1), trout (O. mykiss, Q91193), zebrafish (D. rerio, U71094), carp (C. carpio, AAX16496.1, AY787041), shark (C. plumbeus, AY172838), Xenopus (X. laevis, Q91830), chicken (G. gallus, P25022), mouse (M. musculus, M64796) and human (H. sapiens, P55895), sea urchin (S. purpuratus, DQ082724).

mykiss) (Oettinger et al. 1990; Greenhalgh et al. 1993; (Hansen and Kaattari, 1996; Bertrand et al. 1998a). There Hansen and Kaattari 1996; Ichihara et al. 1992; was no significant similarity between the intergenic region of Greenhalgh and Steiner 1995) and 2.6 kb in zebrafish (Danio flounder and that of zebrafish (U69610.1) and trout rerio) (Willett et al. 1997; Bertrand et al. 1998b). The (U73750). However, multiple polyadenylation sites were intergenic region between flounder RAG1 and RAG2 is 3.1 identified in the intergenic region in flounder: 10 kb, which is similar in size to that of other teleosts (2–3 kb) polyadenylation signals for RAG1 and 6 polyadenylation but considerable smaller than that of tetrapods. Therefore, signals for RAG2. the size of the RAG locus intergenic region varies in diver- It might be the first time to describe an alternatively gent species. spliced exon1 (132 and 138 bp) in the 5′-UTR of RAG1 It had been suggested that the intergenic region between (figure 1). There were some lymphocyte-specific transcrip- RAG1 and RAG2 may contain information modulating RAG tion factor binding sites such as NF-κB, C/EBP, E47 and c- transcription (Bertrand et al. 1998a), and several potential Ets-1 binding sites (Kuo and Schlissel 2009; Fitzsimmons transcription-factor-binding sequences within the murine et al. 1996). Binding sites of E47 and NF-κB were only in intergenic region were identified. The presence of overlap- the 132 bp exon, while C/EBP and c-Ets-1 binding sites were ping transcripts and multiple polyadenylation sites in both in both of them. RAGs were expressed in both of B and T trout and zebrafish increases the likelihood that the expres- lymphocytes and the regulatory elements of RAGs were sion of RAG1 and RAG2 in teleosts may involve post- different in them (Kuo and Schlissel 2009). During lympho- transcriptional regulation through antisense RNA signals cyte development, RAG expression can be regulated by

Figure 4. The expression of flounder RAG1 and RAG2 in adult flounder tissues (L, liver; S, spleen; Pr, pronephros; Me, mesonephros; H, heart; M, muscle; T, intestine. The amplification cycles for RAG1, RAG2 and β-actin were 40, 40 and 30, respectively).

J. Biosci. 39(5), December 2014 Recombination-activating genes from flounder 855

Figure 5. In situ hybridization on paraffin sections of 2-month-old flounders. (A) The sections were counterstained with water-solution eosin after hybridization with the RAG1 or RAG2 probe. (B) RAG1 expression was detected in thymus. (C) RAG2 was expressed in thymus. The region deeply stained is the cortex, where many small and medium thymocytes are densely packed in a loose network. The lighter- staining region is the medulla, consisting of thymocytes scattered among the dense framework of stromal cells (g, gill; m, muscle; ph, pharynx; pr, pronephros; th, thymus). transcription factors through their binding to the binding DNA cleavage activity (Gwyn et al. 2009), can also been found sites in the RAG locus (Kuo and Schlissel 2009) and occurs as H972 and H977 in C-terminal domain of flounder RAG1. in two waves. The first wave leads to assembly the immu- The central domain (residues 528–760) of RAG2 is pre- noglobulin heavy chain (IgH) or TCRβ in pro-B cells or pro- dominantly responsible for DNA binding and displays spec- T cells, while the second wave results in catalysing Ig light ificity for the RSS heptamer. The C-terminal domain chain (L) or TCRα assembly in pre-B cells or pre-T cells (residues 761–980) is capable of dimerization and binds (Schatz 2004; Kuo and Schlissel 2009). E47/E2A had been DNA cooperatively (Arbuckle et al. 2001). Residues 315– proved to activate IgH rearrangement (Schlissel et al. 1991; 371 of the murine RAG2 protein provide the surface for Bain et al. 1994) and regulated RAG expression in B cell RAG1 interaction. This region corresponds to the last of precursors (Hsu et al. 2003). In B cells, NF-κB could regu- six predicted repeat motifs that have been proposed by late RAG transcription response to self-antigen (Verkoczy sequence analysis to fold RAG2 into a six-bladed β-propeller et al. 2005; Kuo and Schlissel 2009). Perhaps, the 132bp structure. Residue W317 within the sixth repeat is shown to exon1 is important for the first wave expression of RAG1 be critical for mediating contact with RAG1 and, concur- and its specific expression in B lymphocytes. The function rently, for stabilizing the binding to and directing the cleav- the alternatively spliced exon1in regulating the expression of age of the RSSs (Aidinis et al. 2000). The location of residue RAG1 needs further analysis. W317 was conserved in flounder as W319. The C-terminal The active site residues (D600, D708 and E962) that are region of murine RAG2 contains two conserved serine res- essential for catalytic activity in the murine RAG1 protein have idues (amino acids 375 and 411) and a threonine (amino acid been identified. In addition, there are three independent domains 490) that are thought to be involved in modulating RAG2 within the murine core RAG1 that are involved in the catalytic activity via phosphorylation but were shown to be dispens- activity, a C3HC4 subdomain in the N-terminal third of the able for in vitro extra-chromosomal recombination (Lin and murine RAG1 1040 residues, a RING finger and zinc finger Desiderio, 1994); the two conserved serine (amino acids 376 A (ZFA) in the zinc-binding dimerization domain, and a zinc and 414) and threonine (amino acid 494) residues were also finger B (ZFB) in the core region and a zinc site in the C- found in the C-terminus of flounder RAG2. terminal domain (Kim et al. 1999; Landree et al. 1999; The phylogenetic results (figure 3) demonstrated that the Fugmann et al. 2000; De and Rodgers 2004;Gwynet al. branching position of the taxa within the trees is consistent 2009). The position of the C3HC4 subdomain may have a with the known evolution of these organisms. Since the significant influence on the activities associated with the C- discovery of RAGs, RAG1 has been proven useful to estab- terminal domain of the protein (Rodgers et al. 1996). Zinc finger lish a higher-level phylogenetic inference among vertebrate B (residues 727–750) of murine RAG1 provided a dominant lineages (Groth and Barrowclough 1999; Martin 1999; Sato binding domain for recruiting RAG2 cleavage (Aidinis et al. and Suzuki 2004; Paton et al. 2003; Valbuena-Ureña et al. 2000). The putative active site residues of the flounder RAG1 2013; Groth and Barrowclough 1999;Martin1999;San protein are D633, D743 and E997, and the location of the Mauro et al. 2004). Our results confirmed that RAGs had putative C3HC4, ZFA and ZFB domains are at residues 312– the potential of being utilized as tools for phylogenetic 350, 375–396 and 762–785, respectively. The N-terminal do- comparisons within species that possess V(D)J recombina- main (residues 384–527) contains RSS nonamer-binding sites tion (Hansen and Kaattari 1995, 1996). and has a high affinity for DNA. The zinc-coordinating residues As the RAG genes were specifically expressed within H937 and H942 of mouse RAG1, which were essential for those organs undergoing V(D)J recombination, the

J. Biosci. 39(5), December 2014 856 Xianlei Wang et al. expression correlated with the location of the lymphoid organs and Technology of China-Platforms of Aquaculture Stock because the thymus, kidney and spleen are the major lymphoid Resources and National & Local Joint Engineering Laboratory organs in teleosts. In general, the thymus is the first organ to of Ecological Mariculture. We would like to thank Lijing become lymphoid, although the kidney can contain hemato- Zhang of the Yellow Sea Fisheries Research Institute for the poietic precursors but not lymphocytes earlier in development. paraffin sections. In freshwater teleosts, the spleen is the last organ to become lymphoid (Zapata et al. 2006). In trout, the highest expression of RAG2 was detected in the thymus, followed by the pronephros, with much weaker signals in the spleen, mesonephros Reference and liver (Hansen and Kaattari 1996). Zebrafish and pufferfish RAG2 are mainly expressed in the thymus and kidney (Peixoto Aidinis V, Dias DC, Gomez CA, Bhattacharyya D, Spanopoulou E et al. 2000;Lamet al. 2004). In our study, both genes were and Santagata S 2000 Definition of minimal domains of inter- identified by RT-PCR in the pronephros and mesonephros, but action within the recombination-activating genes 1 and 2 no expression of either gene was found in other tissues. Fur- recombinase complex. J. Immunol. 164 5826–5832 thermore, the expression of both genes was detected by in situ Arbuckle JL, Fauss LA, Simpson R, Ptaszek LM and Rodgers KK hybridization on paraffin sections of thymus, and there was 2001 Identification of two topologically independent domains in very faint staining in the kidney. Therefore, our results re- RAG1 and their role in macromolecular interactions relevant to vealed that the thymus and kidney most likely serve as the V(D)J recombination. J. Biol. Chem. 276 37093–37101 primary lymphoid tissues in flounder. Bain G, Maandag EC, Izon DJ, Amsen D, Kruisbeek AM, In most vertebrates, the thymus is a bilateral organ that is Weintraub BC, Krop I, Schlissel MS, et al. 1994 E2A proteins completely surrounded by a connective tissue capsule contain- are required for proper B cell development and initiation of – ing two well-defined areas, the cortex and the medulla. The immunoglobulin gene rearrangements. Cell 79 885 892 Bernstein RM, Schluter SF, Bernstein H and Marchalonis JJ 1996 cortex consists of a supporting meshwork of epithelial cells Primordial emergence of the recombination activating gene 1 and epithelial cysts that house the maturing thymocytes, and (RAG1): sequence of the complete shark gene indicates homol- the medulla consists of epithelial cells, secretory epithelial cells ogy to microbial integrases. Proc. Natl. Acad. Sci. USA 93 and epithelial cysts. In the mouse, RAG1 and RAG2 mRNA 9454–9459 were found in the embryonic thymus, suggesting that the Bertrand FE 3rd, Olson SL, Martin DA and Wu GE 1998a Se- thymic cortex, rather than the medulla, was the main site of quence analysis of the mouse RAG locus intergenic region. Dev. transcription of antigen receptor genes (Boehm et al. 1991). An Immunol. 5 215–222 analysis in zebrafish (Wei et al. 2002)usingaRAG1 probe Bertrand FE 3rd, Olson SL, Willett CE and Wu GE 1998b Se- revealed that the zebrafish cortex–medullary regionalization quence of the RAG1 and RAG2 intergenic region in zebrafish – begins between 1 and 2 weeks post-fertilization, when RAG1 (Danio rerio). Dev. Immunol. 5 211 214 expression clearly demarcated the cortex, whereas the medulla Boehm T, Gonzalez-Sarmiento R, Kennedy M and Rabbitts TH 1991 A simple technique for generating probes for RNA in situ was RAG1-negativeinthymustissue(Lamet al. 2002;Dani- hybridization: an adjunct to genome mapping exemplified by the lova et al. 2004). In the present study, the region demonstrating RAG-1/RAG-2 gene cluster. Proc. Natl. Acad. Sci. USA 88 a significant RAG signal was the cortex, in which many small 3927–3931 and medium thymocytes were densely packed in a loose net- Carlson LM, Oettinger MA, Schatz DG, Masteller EL, Hurley EA, work, whereas the medulla, consisting of thymocytes scattered McCormack WT, Baltimore D and Thompson CB 1991 Selec- among the dense framework of stromal cells, produced a tive expression of RAG-2 in chicken B cells undergoing immu- fainter signal. The flounder RAG mRNAs were detected in noglobulin gene conversion. Cell 64 201–208 the thymus and more likely demarcated the cortex, suggesting Chun JJ, Schatz DG, Oettinger MA, Jaenisch R and Baltimore D that the thymic cortex contained maturing lymphocytes. 1991 The recombination activating gene-1 (RAG1) transcript is – In conclusion, flounder RAG1 and RAG2, which are present in the murine central nervous system. Cell 64 189 200 convergently transcribed, are separated by a 3.1 kb intergenic Corripio-Miyar Y, Bird S, Treasurer JW and Secombes CJ 2007 RAG-1 and IgM genes, markers for early development of the region. The 5′-UTR of RAG1 contain an alternative exon. Our immune system in the gadoid haddock, Melanogrammus data suggested that the thymus and kidney most likely serve as aeglefinus, L. Fish Shellfish Immunol. 23 71–85 the primary lymphoid tissues in the flounder. Covello JM, Bird S, Morrison RN, Bridle AR, Battaglene SC, Secombes CJ and Nowak BF 2013 Isolation of RAG-1 and IgM transcripts from the striped trumpeter (Latris lineata), and Acknowledgements their expression as markers for development of the adaptive immune response. Fish Shellfish Immunol. 34 778–788 This work was supported by the National High Technology Danilova N, Hohman VS, Sacher F, Ota T, Willett CE and Steiner Research and Development Program of China (863 Program, LA 2004 T cells and the thymus in developing zebrafish. Dev. 2012AA092203), the National Key Basic Program of Science Comp. Immunol. 28 755–767

J. Biosci. 39(5), December 2014 Recombination-activating genes from flounder 857

De P and Rodgers KK 2004 Putting the pieces together: identifica- Lam SH, Chua HL, Gong Z, Wen Z, Lam TJ and Sin YM 2002 tion and characterization of structural domains in the V(D)J Morphologic transformation of the thymus in developing recombination protein RAG1. Immunol. Rev. 200 70–82 zebrafish. Dev. Dyn. 225 87–94 Du Pasquier L 1992 Origin and evolution of the vertebrate immune Landree MA, Wibbenmeyer JA and Roth DB 1999 Mutational system. APMIS 100 383–392 analysis of RAG1 and RAG2 identifies three catalytic amino Fitzsimmons D, Hodsdon W, Wheat W, Maira SM, Wasylyk B and acids in RAG1 critical for both cleavage steps of V(D)J recom- Hagman J 1996 Pax-5 (BSAP) recruits Ets proto-oncogene bination. Genes Dev. 13 3059–3069 family proteins to form functional ternary complexes on a B- Lin WC and Desiderio S 1994 Cell cycle regulation of V(D)J cell-specific promoter. Genes Dev. 10 2198–2211 recombination-activating protein RAG-2. Proc. Natl. Acad. Fugmann SD, Villey IJ, Ptaszek LM and Schatz DG 2000 Identifica- Sci. USA 91 2733–2737 tion of two catalytic residues in RAG1 that define a single active Martin AP 1999 Substitution rates of organelle and nuclear genes in site within the RAG1/RAG2 protein complex. Mol. Cell 5 97–107 sharks: implicating metabolic rate (again). Mol. Biol. Evol. 16 Grawunder U, West RB and Lieber MR 1998 Antigen receptor 996–1002 gene rearrangement. Curr. Opin. Immunol. 10 172–180 Matthews AG and Oettinger MA 2009 RAG: a recombinase diver- Greenhalgh P and Steiner LA 1995 Recombination activating gene sified. Nat. Immunol. 10 817–821 (RAG 1) in zebrafish and shark. Immunogenetics 41 54–55 Mombaerts P, Iacomini J, Johnson RS, Herrup K, Tonegawa S and Greenhalgh P, Olesen CE and Steiner LA 1993 Characterization Papaioannou VE 1992 RAG-1-deficient mice have no mature B and expression of recombination activating genes (RAG-1 and and T lymphocytes. Cell 68 869–877 RAG-2)inXenopus laevis. J. Immunol. 151 3100–3110 Oettinger MA, Schatz DG, Gorka C and Baltimore D 1990 RAG-1 Groth JG and Barrowclough GF 1999 Basal divergences in birds and RAG-2, adjacent genes that synergistically activate V(D)J and the phylogenetic utility of the nuclear RAG-1 gene. Mol. recombination. Science 248 1517–1523 Phylogenet. Evol. 12 115–123 Padrós F and Crespo S 1996 Ontogeny of the lymphoid organs in Gwyn LM, Peak MM, De P, Rahman NS and Rodgers KK 2009 A the turbot Scophtalmus maximus: a light and electron micro- zinc site in the C-terminal domain of RAG1 is essential for DNA scope study. Aquaculture 144 1e16 cleavage activity. J. Mol. Biol. 390 863–878 Paton TA, Baker AJ, Groth JG and Barrowclough GF 2003 RAG-1 Hansen JD and Kaattari SL 1995 The recombination activation sequences resolve phylogenetic relationships within gene 1 (RAG1) of rainbow trout (Oncorhynchus mykiss): clon- Charadriiform birds. Mol. Phylogenet. Evol. 29 268–278 ing, expression, and phylogenetic analysis. Immunogenetics 42 Peixoto BR, Mikawa Y and Brenner S 2000 Characterization of the 188–195 recombinase activating gene-1 and 2 locus in the Japanese Hansen JD and Zapata AG 1998 Lymphocyte development in fish pufferfish, Fugu rubripes. Gene 246 275–283 and amphibians. Immunol. Rev. 166 199e220 Pulsford A, Tomlinson MG, Lemaire-Gony S and Glynn PJ 1994 Hansen JD 1997 Inspection of the 3' UTR genomic region for Development and immunocompetence of juvenile flounders, RAG1 and RAG2 in rainbow trout (Oncorhynchus mykiss) Platichthys flesus L. Fish Shellfish Immunol. 4 63e78 reveals potential regulatory motifs. Dev. Immunol. 5 129–131 Rodgers KK, Bu Z, Fleming KG, Schatz DG, Engelman DM and Hansen JD and Kaattari SL 1996 The recombination activating Coleman JE 1996 A zinc-binding domain involved in the di- gene 2 (RAG2) of the rainbow trout (Oncorhynchus mykiss). merization of RAG1. J. Mol. Biol. 260 70–84 Immunogenetics 44 203–211 Romano N, Fanelli M, Maria Del Papa G, Scapigliati G and Hsu LY, Lauring J, Liang HE, Greenbaum S, Cado D, Zhuang Y Mastrolia L 1999 Histological and cytological studies on the and Schlissel MS 2003 A conserved transcriptional enhancer developing thymus of sharpsnout seabream, Diplodus puntazzo. regulates RAG gene expression in developing B cells. Immunity J. Anat. 194 39e50 19 105–117 Saitou N and Nei M 1987 The neighbor-joining method: a new Ichihara Y, Hirai M and Kurosawa Y 1992 Sequence and chromo- method for reconstructing phylogenetic trees. Mol. Biol. Evol. 4 some assignment to the 11p13-p12 of human RAG genes. 406–425 Immunol. Lett. 33 277–284 San Mauro D, Gower DJ, Oommen OV, Wilkinson M and Zardoya Ikuta K, Uchida N, Friedman J and Weissman IL 1992 Lymphocyte R 2004 Phylogeny of caecilian amphibians (Gymnophiona) development from stem cells. Annu. Rev. Immunol. 10 759–783 based on complete mitochondrial genomes and nuclear RAG1. Josefsson S and Tatner MF 1993 Histogenesis of the lymphoid organs Mol. Phylogenet. Evol. 33 413–427 in sea bream, Sparus auratus L. Fish Shellfish Immunol. 3 35e50 Sato JJ and Suzuki H 2004 Phylogenetic relationships and Kim DR, Dai Y, Mundy CL, Yang W and Oettinger MA 1999 divergence times of the genus Tokudaia within Murinae Mutations of acidic residues in RAG1 define the active site of (Muridae; Rodentia) inferred from the nucleotide sequences the V(D)J recombinase. Genes Dev. 13 3070–3080 encoding the Cytb gene, RAG 1, and IRBP. Can. J. Zool. 82 Kuo TC and Schlissel MS 2009 Mechanisms controlling expression 1343–1351 of the RAG locus during lymphocyte development. Curr. Opin. Schatz DG, Oettinger MA and Baltimore D 1989 The V(D)J Immunol. 21 173–178 recombination activating gene, RAG-1. Cell 59 1035–1048 Lam SH, Chua HL, Gong Z, Lam TJ and Sin YM 2004 Develop- Schatz DG, Oettinger MA and Schlissel SM 1992 V(D)J recombi- ment and maturation of the immune system in zebrafish, Danio nation: molecular biology and regulation. Annu. Rev. Immunol. rerio: a gene expression profiling, in situ hybridization and 10 359–383 immunological study. Dev. Comp. Immunol. 28 9–28 Schatz DG 2004 V(D)J recombination. Immunol. Rev. 200 5–11

J. Biosci. 39(5), December 2014 858 Xianlei Wang et al.

Schlissel M, Voronova A and Baltimore D 1991 Helix-loop-helix Wang X, Tan X, Zhang P and Xu Y 2012 Construction and transcription factor E47 activates germ-line immunoglobulin expression of a prokaryotic vector of recombinant olive flounder heavy-chain gene transcription and rearrangement in a pre-T- RAG2. Mar. Sci. (Chinese) 36 59–63 cell line. Genes Dev. 5 1367–1376 Wei XC, Kishi H, Jin ZX, Zhao WP, Kondo S, Matsuda T, Saito S Shinkai Y, Rathbun G, Lam KP, Oltz EM, Stewart V, Mendelsohn and Muraguchi A 2002 Characterization of chromatin structure M, Charron J, Datta M, et al. 1992 RAG-2-deficient mice lack and enhancer elements for murine recombination activating mature lymphocytes owing to inability to initiate V(D)J rear- gene-2. J. Immunol. 169 873–881 rangement. Cell 68 855–867 Willett CE, Cherry JJ and Steiner LA 1997 Characterization and Shlyakhtenko LS, Gilmore J, Kriatchko AN, Kumar S, Swanson PC expression of the recombination activating genes (rag1 and and Lyubchenko YL 2009 Molecular mechanism underlying rag2) of zebrafish. Immunogenetics 45 394–404 RAG1/RAG2 synaptic complex formation. J. Biol. Chem. 284 Zapata A, Diez B, Cejalvo T, Gutiérrez-de Frías C and Cortés A 20956–20965 2006 Ontogeny of the immune system of fish. Fish Shellfish Tamura K, Stecher G, Peterson D, Filipski A and Kumar S 2013 Immunol. 20 126–136 MEGA6: Molecular Evolutionary Genetics Analysis version Zhang XL, Lu YS, Jian JC and Wu ZH 2012 Cloning and expression 6.0. Mol. Biol. Evol. 30 2725–2729 analysis of recombination activating genes (RAG1/2) in red snap- Valbuena-Ureña E, Amat F and Carranza S 2013 Integrative per (Lutjanus sanguineus). Fish Shellfish Immunol. 32 534–543 phylogeography of calotriton newts (Amphibia, Salamandridae), Zhang Y, Tan X, Zhang PJ and Xu Y 2006 Characterization of with special remarks on the conservation of the endangered muscle-regulatory gene, MyoD, from flounder (Paralichthys Montseny brook newt (Calotriton arnoldi). PLoS ONE 8 e62542 olivaceus) and analysis of its expression patterns during embryo- Verkoczy L, Ait-Azzouzene D, Skog P, Martensson A, Lang J, genesis. Mar. Biotechnol. (NY) 8 139–148 Duong B and Nemazee D 2005 A role for nuclear factor kappa Zuckerkandl E and Pauling L 1965 Evolutionary divergence and B/rel transcription factors in the regulation of the recombinase convergence in proteins; in Evolving genes and proteins (eds) V activator genes. Immunity 22 519–531 Bryson and HJ Vogel (New York: Academic Press) pp 97–166

MS received 21 February 2014; accepted 04 July 2014

Corresponding editor: STUARTTUARTANANEWMANEWMAN

J. Biosci. 39(5), December 2014