ARTICLE IN PRESS Developmental & Comparative Immunology Developmental and Comparative Immunology 31 (2007) 889–902 www.elsevier.com/locate/devcompimm Identiﬁcation of immune-relevant genes in histoincompatible rejecting colonies of the tunicate Botryllus schlosseri Matan Orena,b,Ã, Jacob Doueka, Zvi Fishelsonb, Baruch Rinkevicha a Israel Oceanographic and Limnological Research, National Institute of Oceanography, P.O. Box 8030, Haifa 31080, Israel b Department of Cell and Developmental Biology, Sackler School of Medicine, Tel Aviv University, Tel Aviv 69978, Israel Received 22 November 2006; received in revised form 16 December 2006; accepted 18 December 2006 Available online 24 January 2007 Abstract The colonial ascidian Botryllus schlosseri manifests a unique allorecognition system that is controlled by a single histocompatibility haplotype, the Fu/HC locus. When two allogeneic incompatible colonies come into direct contact, they develop inﬂammatory-like rejection lesions, called points of rejection (POR). While screening for differentially expressed genes during POR formation, we developed and analyzed a cDNA library of expressed sequence tags (ESTs) with 1693 unique ESTs that were clustered and assembled into 217 contigs and 1476 singlets. About 51% of these ESTs showed high similarity (E-value p0.005) to known database sequences, of which 123 matches were identiﬁed as immune-relevant genes encoding for stress proteins, pattern recognition receptors and complement proteins, proteases and protease inhibitors, cell adhesion and coagulation proteins, cytokine-related proteins, programmed cell death and proteasome-associated proteins. This ﬁrst EST wide-screening analysis of the Botryllus allorecognition effector arm reveals a complex innate immune system, hallmarked by a whole genome response to allorecognition challenge. r 2007 Published by Elsevier Ltd. Keywords: Allorecognition; Botryllus; Expressed sequence tags; Immunity; Subtraction library; Urochordata 1. Introduction nerve tube and segmented musculature in their tadpole larva ), and therefore have been con- Ascidians are sessile marine invertebrates of the sidered as a legitimate subject for evolution studies subphylum Urochordata, occupying a key phyloge- . This is further supported by the availability of netic position in the origin of the vertebrates . As improved research tools, such as the publication of chordates, this group of organisms features basic the draft genome from the solitary ascidian Ciona morphological characteristics, similar to those ob- intestinalis . served in the vertebrates (tail, notochord, dorsal One colonial urochordate group, the botryllid ascidians (family Styelidae, subfamily Botryllinae), ÃCorresponding author. Israel Oceanographic and Limnologi- exhibits a unique and well-deﬁned allorecognition cal Research, National Institute of Oceanography, P.O. Box phenomenon, which has become a model system in 8030, Haifa 31080, Israel. Tel.: +972 4 8565273; fax: +972 4 8511911. the study of the evolution of immunity . A typical E-mail addresses: firstname.lastname@example.org, email@example.com botryllid colony consists of few to several thousands (M. Oren). morphologically and genetically identical modular 0145-305X/$ - see front matter r 2007 Published by Elsevier Ltd. doi:10.1016/j.dci.2006.12.009 ARTICLE IN PRESS 890 M. Oren et al. / Developmental and Comparative Immunology 31 (2007) 889–902 units, each called a zooid, which are typically tag (EST) screening analysis that aims to reveal arranged in oval, serpentine or star-shaped struc- transcriptionally modulated genes during Botryllus tures, termed systems, embedded within a translu- rejection process. cent and gelatinous matrix, the tunic. All zooids within a colony are connected to each other via a 2. Methods ramiﬁed vascular system that projects along the peripheral colonial zones through numerous blind 2.1. Allorecognition assay termini, referred to as ampullae. Ampullae are the sites for self–nonself-recognition and allorecogni- Four Fu/HC incompatible colonies of B. schlos- tion effector-arm expression in botryllid ascidians seri, originating from specimens collected some [6,7]. In the model species, Botryllus schlosseri years ago at Monterey Marina, California, were allorecognition processes begin when the tunic edges maintained under constant conditions, in the ﬂow of two allogeneic disparate colonies make contacts. through facility at the National Institute of Ocea- These contacts lead further to either fusion or nography. Two similar size subclones were prepared rejection between the colonies, a phenomenon that from each genotype, as described . Four is controlled by a single highly polymorphic subclones, one from each genotype, were used for histocompatibility haplotype termed the Fu/HC the allorecognition assays by positioning them in locus, which contains multiple co-dominantly, ex- two allorecognition incompatible pairs, each pair on pressed alleles [8,9]. Fusion may occur between two a different glass slide. The subclones were juxta- different genotypes sharing one or both Fu/HC posed in a distance of less than 1 mm. This alleles, whereas rejecting colonies share no Fu/HC permitted the development of a full repertoire of alleles. Fused colonies form a chimeric entity, interactions between growing edges. When the ﬁrst leading to morphological resorption of one of the point of rejection (POR) had been identiﬁed partners  followed by development of somatic (24–36 h after initiation; Fig. 1), the allogeneic and germ cell parasitism . Allogeneic incompa- challenged subclones were harvested for RNA tible encounters develop dark-golden to brown extraction. The four allogeneic naive subclones of cytotoxic lesions that are usually followed by the same genotypes were harvested at the same time formation of necrotic areas (called points of (illustrated schematically in Fig. 2). rejection or PORs; Fig. 1) and in many cases, amputation of interacting ampullae. Interactions 2.2. RNA extraction terminate by disconnections between contacting colonies, in many cases in concert with a retreat Total RNA was extracted separately from the growth phenomenon . Although scientists have interacting pairs and from their naive homologues recently claimed to have isolated a candidate Fu/ by EPICENTER MasterPureTM RNA Puriﬁcation HC locus , the genes participating in fusion or kit. The integrity of the total RNA was veriﬁed by rejection processes have remained unknown. It agarose gel electrophoresis. should be noted, that although characteristics of botryllid ascidians allorecognition resemble some 2.3. Subtraction library characteristics of the vertebrates immunity , it is regarded as innate and distinct from the adaptive First strand cDNA synthesis was performed on graft rejection of the vertebrates. Observations 500 ng from each, experiment and naive total RNA, suggest that allorecognition in botryllid ascidians by Super SMARTTM PCR cDNA synthesis kit is a blood-borne phenomenon initiated by humoral (Clontech) according to manufacturer’s instruc- factors . To date, attempts to identify these tions. cDNA subtraction was done in one direction proteins have failed. Several Botryllus genes, which (naive transcriptosome was subtracted from the could be related to its innate immunity (which are challenged transcriptosome) by Clontech PCR- prevalent or upregulated at sites of interactions), SelectTM cDNA Subtraction Kit according to were identiﬁed in botryllid ascidians [16–23], but no manufacturer’s instructions. All PCR reactions of extensive screening of the allorecognition machinery the process were done on a Perkin-Elmer GeneAmp has been made. Based on a speciﬁc allorecognition PCR System 5700 using Clontech original primers. assay, developed previously  we present here, for The ampliﬁed fragments were cloned in pTZ57R/T the ﬁrst time, a comprehensive expressed sequence vector (Fermentas; the proportion of white to blue ARTICLE IN PRESS M. Oren et al. / Developmental and Comparative Immunology 31 (2007) 889–902 891 Fig. 1. B. schlosseri colonies during rejection: (A) general picture am. ampulla z. zooid and (B) enlarged picture of the marked area. por. point of rejection. ia. Interacting ampulla. Scale bar indicates 50 mm. clones was about 80%). A total of 2304 individual processing was done by SequencherTM program positive clones were selected and re-cultured in 96 (Gene Codes Corporation). It included trimming wells LB plates with 100 mg/ml ampicillin. Sequen- of low quality sequence ends, pTZ57R vector cing of the PCR products was performed in the Max sequences and of the Clontech primers Planck Institute for Molecular Genetics, Berlin, (50 -CCCGGGCAGGT-30 , 50 -GCGGCCGAGGT-30 ). Germany. Further cleaning was done by a dedicated Perl script, which included removal of CDS5 (AC- 2.4. Sequences processing and assembly GCGGG) and AC from the 50 ends and their complementary sequence from the 30 ends; trimming Sequences analysis was performed at the Bioin- polyA ends from the 30 ends and polyT from the 50 formatics Support Unit at the Ben-Gurion Uni- ends; and rejecting cleaned sequences shorter than versity of the Negev, Israel. Initial sequence 50 bp. Finally, manual check was done to ensure ARTICLE IN PRESS 892 M. Oren et al. / Developmental and Comparative Immunology 31 (2007) 889–902 Colony A Colony B Subcloning Subcloning Naive Rejection assay Naive Total RNA Tester cDNA Total RNA Driver cDNA Fig. 2. The cDNA source of B. schlosseri subtraction library—experiment scheme. that all remaining sequences were of high quality. matches if their Blastx E-value cut-off was lower The cleaned sequences were assembled into singlets than 0.005 (only the top match for each sequence is or contigs using CAP3 software . revealed). All sequences with E-value above this threshold were considered sequences with no 2.5. Blast analysis database match. Processed sequences (contigs and singlets) longer 2.6. RT PCR than 100 base pairs were merged into one Fasta ﬁle and submitted to database search using Blastx RNA was extracted from interacting and naive algorithm. The blast search was done separately colonies (N ¼ 4) as described in Section 2.2. First against both SwissProt and GenBank databases on strand cDNA was synthesized by ﬁrst strand DNA all unique sequences. Sequences were considered as synthesis kit (Fermentas). The PCR ampliﬁcation ARTICLE IN PRESS M. Oren et al. / Developmental and Comparative Immunology 31 (2007) 889–902 893 Table 1 Primers used for RT PCR Gene match name 50 -primer 30 -primer Annealing temperature (1C) AsMASPb (seq. 39) 50 -GGCCTTGGTCAGTCTTCCTC-30 50 -TGACTTTCATCAACAGTTCGG-30 53 Cortical granule lectin 50 -AACCCTGGAAATCGTGAGC-30 50 -CGTTCGGTATTCCTTTCGG-30 53.3 (seq. 33) C-type lectin 2 (seq. 32) 50 -ACCGGTATTAGTGGCTGGAAG-30 50 -TCGTCATAACCAACCCACAAG-30 55 Humoral lectin 50 -TACCGCTGTGACCAACTTTG-30 50 -CCTGGTGCTAAGACGTTTCC-30 55.4 (seq. 31) Rhamnospondin 2 50 -GGCACGATCACCTACACCAG-30 50 -ACGTAAGTTTGAATGCCAGGTC-30 55.5 (seq. 37) Selectin P (seq. 73) 50 -GCCTAAATGCACCCTGATCAC-30 50 -GCGCGGGACATACAATCAG-30 56.2 BCAM (seq. 70) 50 -GATCCTTTGGCCGTGCATG-30 50 -CACAAGGCAGCACCAAGGTC-30 57 Rhamnospondin 1 50 -ATGTATGTGCACCGTCCTGTC-30 50 -ACGGAGGAGAGTGTTGGCAC-30 57.2 (seq. 36) 40S ribosomal protein 50 -TTGTGACGGGTAACCAGATTC-30 50 -CTTGTAGTAGCGAGCGAGACG-30 57.2 S13 (Ctrl) BS cytoplasmic actin 50 -GTAGGTAGTCTCGTGAATTC-30 50 -CACGCCATCTTGCGTCTGGA-30 57.2 (ref gene) was performed using designed sets of primers (IDT sequences]/Total number of sequences) was calcu- Inc.) as depicted in Table 1. In addition to the lated to be 21.6%. speciﬁc sets of primers, actin primers were added to all samples in order to amplify fragments of a 3.2. Functional categories of EST matches known database Botryllus actin (accession number: AY159281), which served as a reference gene. The The BLASTX analysis revealed that 51% (861) of PCR reaction was carried out for 30 cycles (94 1C, the ESTs had high similarity to known genes (E- 1 min; 53–57.2 1C, 1 min; 72 1C, 1 min) following by value p0.005) whereas the remaining 49% (832) additional 5 min at 72 1C. The RNA samples were had low or no similarity (E-value 40.005). The 861 normalized by a separate RT PCR reaction (ctrl) sequences with signiﬁcant SwissPort matches were using two sets of primers to amplify fragments of classiﬁed into functional categories according to both housekeeping genes actin and a library 40S their blast results (Fig. 4): 336 ESTs (39%) were ribosomal protein S13 (accession number: classiﬁed as genes with unknown function; 123 EH017358). This reaction was carried out for 27 (14%) matched cell/organism defense genes; 112 cycles with same parameters as noted above. Four (13%) with gene expression and protein biosynth- microliters of each PCR product was analyzed in esis genes; 57 (7%) with cell structure and motility 1.2% agarose/EtBr gel alongside DNA marker. genes; 57 (7%) with cell communication and signaling genes; 149 (17%) with metabolism genes; 3. Results and discussion 20 (2%) with transport-related genes; and 7 (1%) with cell cycle genes. Since urochordate genes are 3.1. General library statistics under-represented in protein databases, the distri- bution pattern revealed here might not be fully The quality of 2161 ESTs (93.8% of total ESTs) representative of the tunicate transcriptome. The was found adequate for assembly. The ESTs relatively high expression ratio of genes related to length before assembly ranged between 50 and known cell and organism defense mechanisms 1000 bp (Fig. 3) with an average length of 350 bp indicates the high efﬁciency of the allorecognition and standard deviation of 171. These sequences assay performed. It is also possible that part of the were assembled into 1693 ESTs from which 217 39% ESTs characterized as genes with unknown are contigs and 1476 are singlets. Redundancy functions may be linked to the allorecognition ([Total number of sequencesÀnumber of unique effector arm. In addition, genes of the metabolism ARTICLE IN PRESS 894 M. Oren et al. / Developmental and Comparative Immunology 31 (2007) 889–902 Distribution of Sequences Length 600 500 400 Frequency 300 200 100 0 0-100 101-200 201-300 301-400 401-500 501-600 601-700 701-800 801-900 901-1000 Sequence Length (bp) Fig. 3. Distribution of Botryllus subtraction library EST lengths. Gene expression and protein Cell cycle biosynthesis 1% 13% Unknown function 39% Metabolism 17% Transport 2% Cell structure and Cell motility communication 7% and signaling 7% Cell/organism defense 14% Fig. 4. Distribution of 874 SwissPort EST matches (E-value p0.005) by major functional categories. category include many of the mitochondrial elec- and actin (structure and motility category) and tron transport chain such as ATP synthases, transport proteins (transport category). It is, there- cytochromes B and C, and NADH dehydrogenase, fore, proposed that whole genome response to possibly indicating high energy consuming processes allorecognition challenge is a hallmark of this during the allorecognition assay. Other genes phenomenon. The category of cell/organism de- prominent to the subject of allorecognition are the fense-related genes includes 123 ESTs, matching transcription and translation regulators (in the gene to 110 database entries, among them only six expression and protein biosynthesis category), B. schlosseri known sequences (Table 2). All 123 kinases (cell communication and signaling cate- ESTs were deposited in the GeneBank (accession gory), cytoskeleton genes such as myosin, tubulin numbers: EE743534 to EE743656). ARTICLE IN PRESS M. Oren et al. / Developmental and Comparative Immunology 31 (2007) 889–902 895 Table 2 B. schlosseri cell/organism defense EST matches Functional Seq. no. GeneBank Gene match SwissPort ID Gene match name Organism E-value classiﬁcation accession number Response to stress 1 EE743591 Q90Z20_ANAPL Heat-shock protein 90 alpha A. platyrhynchos 0 2 EE743602 Q6K045_PAROL Heat-shock protein 90 beta P. olivaceus 3.00EÀ32 3 EE743613 Q7YZJ0_9METZ Heat-shock protein 90 S. fuscus 2.00EÀ23 4 EE743624 Q6USB9_9BIVA Heat-shock protein 90 C. farreri 3.00EÀ16 5 EE743635 Q9NAX9_PARTI Heat-shock protein 70 P. trichosuri 3.00EÀ47 6 EE743646 Q95V47_ARTSF Heat-shock protein 70 A. franciscana 1.00EÀ45 7 EE743534 Q2MJK5_HALDI Heat-shock protein 70 H. discus hannai 4.00EÀ43 8 EE743545 HSP70_PYRSA Heat-shock protein 70 P. salina 5.00EÀ04 9 EE743556 GRP75_RAT Heat-shock protein 70 R. norvegicus 1.00EÀ37 10 EE743564 CH10_RAT Heat-shock protein 10 R. norvegicus 2.00EÀ09 11 EE743565 Q5RLR3_PIG Heat-shock protein 10 S. scrofa 2.00EÀ07 12 EE743566 Q3UDS0_MOUSE Heat-shock protein 8 M. musculus 5.00EÀ25 13 EE743567 Q27569_DROHY Cu, Zn superoxide dismutase D. hydei 2.00EÀ13 14 EE743568 Q7PXV9_ANOGA Cu, Zn superoxide dismutase A. gambiae 2.00EÀ10 like (ENSANGP00000018136) 15 EE743569 Q210I6_RHOPB Glutathione peroxidase R. palustris 2.00EÀ13 16 EE743570 OXR1_RAT Oxidation resistance protein 1 R. norvegicus 5.00EÀ09 17 EE743571 Q5DWF4_9CNID Putative 8-lipoxygenase-allene C. viridis 9.00EÀ35 oxide synthase fusion protein 18 EE743572 Q53GB9_HUMAN Microsomal glutathione S- H. sapiens 3.00EÀ25 transferase 3 variant 19 EE743573 DPYS_HUMAN Dihydropyrimidinase H. sapiens 4.00EÀ07 20 EE743574 Q6I666_CITLA Cytochrome P450 like_TBP C. lanatus 6.00EÀ11 21 EE743575 CP3AE_CAVPO Cytochrome P450 3A14 C. porcellus 3.00EÀ05 22 EE743576 O04892_TOBAC Cytochrome P450 like_TBP N. tabacum 2.00EÀ04 23 EE743577 Q8CD72_MOUSE NIMA (Nek1) M. musculus 6.00EÀ74 Pattern recognition 24 EE743578 C1QBP_BOVIN Complement component 1 Q B. taurus 3.00EÀ04 receptors/ subcomponent-binding protein complement proteins 25 EE743579 Q9BP40_HALRO Complement factor B H. roretzi 5.00EÀ03 26 EE743580 Q966W1_HALRO Ficolin 4 H. roretzi 9.00EÀ43 27 EE743581 Q95P98_HALRO Ficolin 3 precursor H. roretzi 8.00EÀ37 28 EE743582 Q95P99_HALRO Ficolin 2 precursor H. roretzi 2.00EÀ45 29 EE743583 * * * 1.00EÀ24 30 EE743584 Q29041_PIG Ficolin S. scrofa 3.00EÀ11 31 EE743585 HMCT_BOMMO Hemocytin precursor (humoral B. mori 2.00EÀ07 lectin) 32 EE743586 Q8AXR8_ANGJA C-type lectin 2 A. japonica 1.00EÀ11 33 EE743587 Q69HM9_CIOIN Cortical granule lectin-like C. intestinalis 7.00EÀ18 34 EE743588 Q5IWS5_HUMAN Intelectin 1 H. sapiens 1.00EÀ30 35 EE743589 Q5PPM0_XENLA XEEL protein X. laevis 2.00EÀ10 36 EE743590 Q1PG16_HYDSY Rhamnospondin 1 H. 2.00EÀ03 symbiolongicarpus 37 EE743592 Q1PG02_HYDSY Rhamnospondin 2 H. 4.00EÀ10 symbiolongicarpus 38 EE743593 Q6NVT6_XENTR Calreticulin X. tropicalis 9.00EÀ23 39 EE743594 O01655_HALRO AsMASPb H. roretzi 3.00EÀ33 Proteases 40 EE743595 Q27458_BOTSH Serine protease B. schlosseri 4.00EÀ27 41 EE743596 * * * 6.00EÀ22 42 EE743597 * * * 2.00EÀ09 43 EE743598 * * * 6.00EÀ20 44 EE743599 Q8T3A0_CIOIN Putative coagulation serine C. intestinalis 1.00EÀ10 protease 45 EE743600 DPP4_PIG Dipeptidyl peptidase 4 (T-cell S. scrofa 4.00EÀ20 activation antigen CD26) 46 EE743601 Q3V5L6_POLMI Trypsin P. misakiensis 2.00EÀ73 47 EE743603 O01310_BOTSH Trypsinogen B. schlosseri 3.00EÀ43 48 EE743604 * * * 9.00EÀ19 49 EE743605 * * * 1.00EÀ33 50 EE743606 O01309_BOTSH Trypsinogen B. schlosseri 8.00EÀ37 51 EE743607 O17439_9ASCI Chymotrypsinogen B. villosa 2.00EÀ16 52 EE743608 * * * 2.00EÀ20 53 EE743609 Q63ZG7_XENLA Cathepsin A like (LOC494810 X. laevis 2.00EÀ07 protein) ARTICLE IN PRESS 896 M. Oren et al. / Developmental and Comparative Immunology 31 (2007) 889–902 Table 2 (continued) Functional Seq. no. GeneBank Gene match SwissPort ID Gene match name Organism E-value classiﬁcation accession number 54 EE743610 Q864S1_BOVIN Cathepsin C B. taurus 2.00EÀ35 55 EE743611 Q45HJ6_9TREM Cathepsin D-like aspartic O. viverrini 2.00EÀ38 protease 56 EE743612 Q6A1I0_SUBDO Cathepsin L S. domuncula 2.00EÀ07 57 EE743614 Q6WNW6_BRABE Cathepsin L B. belcheri 4.00EÀ03 tsingtaunese 58 EE743615 Q2V9 Â 2_9METZ Cathepsin L H. perlevis 6.00EÀ23 59 EE743616 Q5U4I5_XENLA Aspartyl aminopeptidase X. laevis 1.00EÀ11 (LOC495491 protein) 60 EE743617 Q5U2N2_RAT Predicted ubiquitin-speciﬁc R. norvegicus 2.00EÀ16 protease 14 61 EE743618 Q6PGT1_XENLA Legumain like (MGC64351 X. laevis 2.00EÀ11 protein) 62 EE743619 Q504N0_MOUSE Carboxypeptidase A2, M. musculus 8.00EÀ31 pancreatic 63 EE743620 Q364L6_9GAMM Peptidyl-dipeptidase A Shewanella sp. 7.00EÀ27 precursor ANA-3 Protease inhibitors 64 EE743621 ALST_TAMSI Alpha-1-antitrypsin-like P. misakiensis 7.00EÀ06 protein CM55ST precursor 65 EE743622 Q25B53_BRALA Serpin 1 precursor B. lanceolatum 2.00EÀ20 66 EE743623 Q27HF2_BOVIN Squamous cell carcinoma B. taurus 4.00EÀ04 antigen recognized by T-cells 1- like protein 67 EE743625 ILEU_HORSE Serpin B1 E. caballus 1.00EÀ29 68 EE743626 Q2ES50_DABRR Kunitz protease inhibitor-I D. russelli 9.00EÀ03 69 EE743627 Q1LXL1_BRARE Kunitz protease inhibitor-I B. rerio 3.00EÀ18 Cell adhesion/ 70 EE743628 Q99K86_MOUSE Bcam protein M. musculus 5.00EÀ05 coagulation 71 EE743629 SCRB2_MOUSE Scavenger receptor class B M. musculus 4.00EÀ21 member 2 (LIMP II) 72 EE743630 Q3U5F6_MOUSE Selectin, endothelial cell M. musculus 7.00EÀ09 73 EE743631 Q5TI49_HUMAN Selectin P H. sapiens 7.00EÀ19 74 EE743632 Q2XNL5_CHICK Tumor necrosis factor- G. gallus 1.00EÀ10 inducible protein 6 75 EE743633 Q6WB00_TRYCR Mucin-like protein T. cruzi 2.00EÀ03 76 EE743634 Q26878_TRYCR Mucin-like protein T. cruzi 1.00EÀ03 77 EE743636 Q4CUU5_TRYCR Mucin TcMUCII, putative T. cruzi 4.00EÀ03 78 EE743637 MUA1_XENLA Integumentary mucin A.1 X. laevis 5.00EÀ10 precursor 79 EE743638 CLD19_MOUSE Claudin-19 M. musculus 1.00EÀ18 80 EE743639 Q1ZXQ3_DICDI Annexin VII D. discoideum 2.00EÀ11 81 EE743640 Q6NVI8_XENTR Annexin A7 X. tropicalis 9.00EÀ09 82 EE743641 Q2F5T8_BOMMO Annexin isoform 1 B. mori 4.00EÀ13 83 EE743642 Q7T391_BRARE Annexin 11a, isoform 2 B. rerio 6.00EÀ25 84 EE743643 Q6PVV9_CIOIN SPARC C. intestinalis 1.00EÀ28 85 EE743644 O76470_LYTVA Echinonectin L. variegatus 6.00EÀ07 86 EE743645 * * * 3.00EÀ10 87 EE743647 * * * 6.00EÀ07 88 EE743648 Q5S3N1_SALSA Zonadhesin-like S. salar 2.00EÀ20 89 EE743649 Q2PC93_CHICK SCO-spondin G. gallus 5.00EÀ41 90 EE743650 Q7QCP4_ANOGA Notch like A. gambiae 5.00EÀ03 (ENSANGP00000010271 protein) 91 EE743651 Q69HL6_CIOIN Polydomain protein-like C. intestinalis 2.00EÀ08 92 EE743652 O18977_BOVIN Tenascin-X B. taurus 4.00EÀ06 93 EE743653 Q769I3_CIOIN Von Willebrand Factor like 1 C. intestinalis 2.00EÀ40 94 EE743654 * * * 1.00EÀ39 95 EE743655 Q769I2_CIOIN Von Willebrand Factor like 2 C. intestinalis 1.00EÀ20 96 EE743656 Q9BLJ1_CIOIN Ci-META1 C. intestinalis 2.00EÀ03 97 EE743535 * * * 5.00EÀ16 98 EE743536 Q9BLJ2_CIOIN Ci-META2 C. intestinalis 9.00EÀ11 99 EE743537 * * * 6.00EÀ10 100 EE743538 Q8MVQ1_9ASCI Vwa1 protein B. villosa 3.00EÀ06 101 EE743539 * * * 1.00EÀ10 102 EE743540 APLP_LOCMI Apolipophorins precursor L. migratoria 9.00EÀ12 103 EE743541 Q5YD83_BIOGL Fibrinogen related domain B. glabrata 4.00EÀ17 variant 3 ARTICLE IN PRESS M. Oren et al. / Developmental and Comparative Immunology 31 (2007) 889–902 897 Table 2 (continued) Functional Seq. no. GeneBank Gene match SwissPort ID Gene match name Organism E-value classiﬁcation accession number Cytokins and 104 EE743542 Q9R175_RAT Interferon-inducible protein 16 R. norvegicus 7.00EÀ04 cytokine regulation 105 EE743543 Q4H3B3_CIOIN Interferon regulatory factor C. intestinalis 2.00EÀ17 like protein 106 EE743544 Q6GL57_XENTR Interleukin enhancer binding X. tropicalis 1.00EÀ36 factor 3 107 EE743546 Q49CF6_BOVIN Myeloid differentiation B. taurus 3.00EÀ10 primary response protein (MyD88) Programmed cell 108 EE743547 Q7T166_BRARE Death-associated protein 1a B. rerio 2.00EÀ07 death (Fragment) 109 EE743548 Q3HL92_AIPPA Caspase-like protein A. pallida 9.00EÀ06 110 EE743549 PARP1_CRIGR Poly [ADP-ribose] polymerase C. griseus 1.00EÀ34 1 (PARP-1) Tumor proteins 111 EE743550 Q6P3J1_BRARE Tumor protein D52-like 2, 2.00EÀ28 isoform 2 112 EE743551 Q3TWS8_MOUSE Tumor protein, translationally- M. musculus 2.00EÀ15 controlled 1 113 EE743552 ZG14_HUMAN Gastric cancer antigen Zg14 H. sapiens 2.00EÀ27 114 EE743553 BCL7B_HUMAN B-cell CLL/lymphoma 7 H. sapiens 5.00EÀ20 protein family member B Proteasome- 115 EE743554 P90725_BOTSH LMP7-like protein B. schlosseri 1.00EÀ53 associated proteins 116 EE743555 * * * 1.00EÀ38 117 EE743557 LAMP3_HUMAN LAMP-3 H. sapiens 2.00EÀ05 Other 118 EE743558 Q17280_BOTSH FK506-binding protein B. schlosseri 1.00EÀ34 (FKBP) 119 EE743559 P90724_BOTSH Soluble immunoglobulin B. schlosseri 9.00EÀ53 molecule homologue, IG_BOTSC 120 EE743560 Q9CYV7_MOUSE Helicase, lymphoid speciﬁc M. musculus 3.00EÀ34 121 EE743561 Q5TBV0_HUMAN Esterase D H. sapiens 1.00EÀ18 122 EE743562 Q6T4Q2_PYRCO Putative senescence-associated P. communis 2.00EÀ31 protein 123 EE743563 Q1VTK4_9FLAO Cold-shock protein P. torquis 1.00EÀ10 *Different singlet correlated with the same SwissPort match as the one above it. 3.3. Differential expression of lectins shock proteins (seq. 1–12), including HSPs 90, HSPs 70, and two small HSPs (10 and 8 kDa). HSPs are Lectins were chosen for the RT PCR ampliﬁcation members of the chaperone family that reach high since this protein family is highly represented in our expression levels under diverse environmental stress library and its members have been shown to conditions, such as heat . In vertebrates, HSPs participate in innate immune response as described play a role in inﬂammation [27,28] and cell death below. Eight lectins from the library were differentially , immune response to pathogens and detection ampliﬁed by RT PCR using independent RNA from of stressed or damaged cells , all of which can be naive and rejecting Botryllus colonies. The results relevant to Botryllus allogeneic rejection responses. (Fig. 5) revealed speciﬁc ampliﬁcation of lectins in It is interesting to note, that although two HSP70 allogeneic challenged colonies, further conﬁrming the genes had been sequenced from B. schlosseri , speciﬁcity of this EST library to the rejection process. neither of them matched the HSPs expressed here, reﬂecting a wider repertoire of HSP70 genes in this 3.4. Stress proteins species. Sequences 13–23 match proteins that are involved Sequences 1–23 in Table 2 show high similarity to in the protection of cells against oxidative stress and stress-related proteins. Among them are 12 heat- other cytotoxic agents. Oxidative stress is known to ARTICLE IN PRESS 898 M. Oren et al. / Developmental and Comparative Immunology 31 (2007) 889–902 Ctrl. N R N R N R N R N R N R N R N R N R Actin (ref.) ESTs Rhamnospondin 2 Rhamnospondin 1 Cortical granule C-type lectin 2 Humoral lectin 40S ribosomal Seclectin P AsMASPb BCAM protein S13 lectin Fig. 5. RT PCR differential expression of lectins in rejecting (R) and naive (N) Botryllus colonies. An actin fragment (321 bp) served as the reference gene and was expressed in all samples, whereas the approx. 200 bp lectin EST fragments were expressed only in the rejecting colonies. RNA samples were normalized by two housekeeping genes actin and 40S ribosomal protein S13 (Ctrl). be a major cause for cell damage in Botryllus . acids, lipoproteins, peptidoglycans (PGN) and Reactive oxygen metabolites (ROMs) such as (1-3) b-D-glucans , known as pathogen-asso- hydrogen peroxide, hydroxyl radicals and singlet ciated molecular patterns (PAMPs). The two oxygen can be products of innate immunity reac- collagenous lectins, mannose-binding lectins tions, such as respiratory burst and prophenolox- (MBLs) and ﬁcolins, have been proven to be pattern idase activating system expressed in the immune recognition receptors (PRRs) that recognize circulating cells of Botryllus, the morula cells . PAMPs  and initiate the lectin complement We have identiﬁed two new putative Cu and Zn pathway through MBL-associated serine proteases; superoxide dismutases (SOD; seq. 13, 14) and a MASPs . In this study, we have identiﬁed eight glutathione peroxidase (seq. 15), known antioxi- lectins (seq. 31–38), an ascidian MASP (AsMASPb; dants that are capable of reducing ROMs such as seq. 39) and four different ﬁcolins (seq. 26–30), of free hydrogen peroxide to water. SOD was found to which three matched ascidian ﬁcolins (ﬁcolins 2–4). inhibit the proPO cytotoxic activity in Botryllus C1q binding protein (seq. 24) and Factor B (seq. 25) [31,32] and to inactivate superoxide anion products are the only representatives found of the classical in H. roretzi hemocytes . Other genes that may and alternative complement pathways, respectively. participate in protection from oxidative damage are The high number of different PRRs: lectin and the oxidation resistance protein 1,8-lipoxygenase- ﬁcolin matches together with several serine pro- allene oxide synthase fusion proteins and the teases (including MASP) implies a possible existence glutathione S-transferase 3 variant (matching to of a lectin-based opsonization system as suggested seq. 16–18). Sequences 19–22 show similarity to for other ascidians . In this alloreactivity detoxiﬁcation proteins including three different context, it is possible that the C1q, ﬁcolins and cytochrome P450 proteins. Sequence 23 matches lectins play a role in opsonization of dead cells at NIMA (Nek1), a protein involved in the response to the POR and in the inﬂammatory process . DNA damage and ionized radiation stress. Although we have identiﬁed some of the comple- ment lectin pathway components, as well as C1q 3.5. Pattern recognition receptors and complement- binding protein and Factor B matches from the associated genes classical and alternative pathways, there is still no evidence for the existence of a full complement Since, so far there has not been any evidence to system as in vertebrates. This is due to the absence the existence of speciﬁc adaptive immunity in of downstream complement components (C3–C9) in invertebrates , it is assumed that these organisms our library. Matches for most of the complement are equipped with innate immune responses that proteins were predicted based on C. intestinalis draft rely on detection of conserved surface epitopes, genome sequence , but classical complement including lipopolysacharides (LPS), lipoteichoic functionality has not been proven yet. Two putative ARTICLE IN PRESS M. Oren et al. / Developmental and Comparative Immunology 31 (2007) 889–902 899 complement proteins have been identiﬁed pre- coagulation-related protein matches in our library viously in Botryllus: a complement control protein must not be ignored and should be examined in the containing a SCR domain  and a Botryllus context of Botryllus immune system. C-type lectin . Both are not presented in our EST list. It should be noted that our subtraction 3.8. Cytokine-related proteins approach only permits identiﬁcation of genes that undergo up-regulation in their level of transcription. Sequences 104–107 share similarity with cytokine It is possible that products of other genes, such as interacting proteins. Two are interferon-related complement C3, play a role in the alloreactivity (interferon-inducible protein 16 and interferon process but their level of synthesis is not modiﬁed. regulatory factor like protein) and two are inter- Function of complement C3 of H. roretzi in leukin-related (interleukin enhancer binding factor 3, opsonization has been demonstrated . Myeloid differentiation primary response protein; MyD88). The identiﬁcation of these four proteins 3.6. Proteases and protease inhibitors implies a possible existence of the two major cytokines in Botryllus. To date, none of the above proteins or Sequences 40–63 match 19 different proteases, any other interferon or interleukin-related genes have including eight serine proteases (seq. 40–53) from been identiﬁed in Botryllus. which ﬁve are new and three have been identiﬁed before [16,19]. Sequences 64–69 show similarity to six 3.9. Programmed cell death and tumor proteins new serine protease inhibitors (Serpins). Interestingly, all serine proteases identiﬁed in the EST library Sequences 108–110 share similarity with pro- match sequences from ascidian origin. Another grammed cell death proteins and sequences protease group represented here is the cathepsin 111–114 with tumor proteins, some of which posses group (seq. 54–58) including representative of cathe- anti-apoptotic properties (such as tumor protein, psins A, C, D and L. The cathepsins are lysosomal translationally controlled 1). Programmed cell death cysteine proteases that can degrade and digest events are important and integral parts of Botryllus products of phagocytosis. However, they are also cyclic blastogenesis, the weekly developmental known to be involved in programmed cell death . process characteristic to this group of organisms , but apoptosis may be also involved in Botryllus 3.7. Cell adhesion and coagulation immune system. Sequences 70–103 show similarity to cell adhesion 3.10. Proteasome-associated proteins and coagulation proteins. Among them are several cell adhesion proteins that may play a major role in Analysis reveals expression of LMP proteins in inﬂammation, such as members of the selectin (seq. the allorecognition EST library. LMPs are subunits 72, 73) and mucin (seq. 75–77) protein families. In of the immunoproteasome that are regulated by mammals these proteins help in leukocytes recruit- interferon g and tumor necrosis factor . Two ment by promoting their adherence to blood vessel putative LMP matches have been identiﬁed here: a walls, enabling their penetration into the inﬂamed Botryllus LMP 7  (seq. 115–116) and a new LMP tissue. Similarly, during Botryllus rejection re- 3 candidate (seq. 117). Proteasomes are essential sponse, blood cells inﬁltrate through the vasculature protein degradation machinery found in all eukar- endothelium into the tunic, where they participate yotic cells, including ascidians  and involved in in the formation of PORs . This process apoptosis . The role of LMPs and, in general, resembles the inﬂammatory response in higher proteasomes in alloresponsiveness still awaits clar- vertebrates. To date there is no evidence for the iﬁcation. Recently, upregulation of LMP7 expres- existence of coagulation system in ascidians as sion was described in epithelial and endothelial cells known for vertebrates. Nevertheless, ascidians of ischemia kidney . paralogs and/or constituent domains seem to be present for all vertebrates key coagulation factors 3.11. Other . We have identiﬁed 19 coagulation-related protein matches (seq. 85–103), of which 9 are from A previously identiﬁed Botryllus soluble immu- C. intestinalis genome. The presence of so many noglobulin  was expressed in our EST library ARTICLE IN PRESS 900 M. Oren et al. / Developmental and Comparative Immunology 31 (2007) 889–902 (seq. 119). This protein could reveal primitive roles hallmarked by a whole genome response to allor- of immunoglobulin molecules in botryllid ascidians ecognition challenge. In order to analyze further alloresponses. The recently sequenced soluble im- gene expression patterns during the rejection munoglobulin that was deﬁned as the Fu/HC process, we will apply microarray technology based candidate of B. schlosseri  (though its function on our ESTs library. has yet to be revealed) is not represented in the EST library. Another interesting EST match is the Botryllus FKBP (seq. 118), which was identiﬁed Acknowledgments previously . Vertebrate FKBP is a target for immunosuppressive agents FK 506 and its function This study was supported by a grant from the in ascidians have yet to be clariﬁed. Botryllus Marine Genomics Europe Network of Excellence putative CD94/NKR-P1-related receptor, recently (EDD Node), from the United States–Israel Bi- isolated from rejecting Botryllus colonies by non- National Science Foundation (2003-010) and partly biased screening strategy of differential display from the Israel Academy of Science (550-06). We library , was not found in our EST list. thank V. Caspi from the Bioinformatics Support Unit at the Ben-Gurion University of the Negev for 4. Conclusion dedicated computer work and G. Paz for assistance in manuscript preparation. While allorecognition and the complex effector mechanisms involved are well described in the vertebrates, invertebrate’s allorecognition is char- References acterized only on the morphological level. More- over, no adult vertebrates undergo (as in the case of  Berrill NJ. The origin of vertebrates. London: Oxford Botryllus) natural transplantation in the wild (ex- University Press; 1955.  Kowalevsky A. In Entwicklungsgeschichte der einfachen cept of pregnancy in mammals). The remarkable Ascidien. Mem Acad St Petersburg 1866;10:1–19. morphologically deﬁned and predicted allorejection  Satoh N, Jeffery WR. Chasing tails in ascidians: develop- process in botryllid ascidians makes the colony mental insights into the origin and evolution of chordates. allorecognition assay highly suitable for elucidating Trends Genet 1995;11:354–9. various aspects of ascidians’ immunity. Developing  Dehal P, Satou Y, Campbell RK, Chapman J, Degnan B, De Tomaso A, et al. The draft genome of Ciona intestinalis: ESTs for the analysis of whole-genome expression insights into chordate and vertebrate origins. Science towards a speciﬁc biological challenge has been 2002;298:2157–67. proven to be an efﬁcient tool in identiﬁcation of new  Rinkevich B. Primitive immune systems: are your ways my immune genes , including immune-related asci- ways? Immunol Rev 2004;198:25–35.  Taneda Y, Watanabe H. Studies on colony speciﬁcity in the dian genes . The cDNA subtraction technology compound ascidian, Botryllus primigenus Oka, I: initiation permitted us to screen large quantities of ESTs with of ‘‘nonfusion’’ reaction with special reference to blood cells high ratio of relevance to the biological process of inﬁltration. Dev Comp Immunol 1982;6:43–52. ascidian allo-rejection. In our functionality analysis,  Taneda Y, Saito Y, Watanabe H. Self or non-self we have focused on currently known immune/ discrimination in ascidians. Zool Sci 1985;2:433–42. defense-related database gene matches, although  Sabbadin A. Le basi genetiche della capacita di fusione fra colonie in Botryllus schlosseri (Ascidiacea). Atti Accad Naz matches of other categories can be equally impor- Lincei Rend 1962;32:1031–5. tant to the process. Other potentially relevant ESTs  Rinkevich B, Porat R, Goren M. Allorecognition elements may prove to be hidden in the 819 ESTs (48%) that on a urochordate histocompatibility locus indicate unprece- show low or no similarity, or within the 343 ESTs dented extensive polymorphism. Proc R Soc London B (20%) that match with genes with unknown 1995;259:319–24.  Rinkevich B, Weissman IL. A long-term study of fused function. The last category can be of particular subclones of a compound ascidian: the resorption phenom- interest for gene functionality studies in ascidians, enon. J Zool (London) 1987;213:717–33. because it can help to reveal the functions of these  Stoner DS, Rinkevich B, Weissman IL. Heritable germ and genes in higher vertebrates. We found 123 ESTs somatic cell lineage competitions in chimeric colonial protochordates. Proc Natl Acad Sci USA 1999;96:9148. (14%) that show similarity to cell/organism defense  Rinkevich B, Weissman IL. Retreat growth in the ascidian genes. As expected, no classical elements of acquired Botryllus schlosseri: the consequences of non-self recogni- immunity has been found. Nevertheless, the overall tion. In: Grosberg RK, editor. Invertebrate historecognition. ﬁgure drawn is of complex innate immune system, New York: Plenum Press; 1988. p. 93–109. ARTICLE IN PRESS M. Oren et al. / Developmental and Comparative Immunology 31 (2007) 889–902 901  De Tomaso AW, Nyholm SV, Palmeril KJ, Ishizukal KJ,  Stewart GR, Young DB. Heat-shock proteins and the Ludington WB, Mitchel K, et al. Isolation and characteriza- host–pathogen interaction during bacterial infection. Curr tion of a protochordate histocompatibility locus. Nature Opin Immunol 2004;16:506–10. 2005;38:454–9.  Ballarin L, Cima F, Floreani M, Sabbadin A. Oxidative  Scoﬁeld VL, Schlumpberger JM, West LA, Weissman IL. stress induces cytotoxicity during rejection reaction in the Protochordate allorecognition is controlled by an MHC-like compound ascidian Botryllus schlosseri. Comp Biochem gene system. Nature 1982;295:499–502. Physiol C 2002;133:411–8.  Watanabe H, Taneda Y. Self and non-self recognition in  Ballarin L, Cima F. Phenoloxidase and cytotoxicity in the compound ascidians. Am Zool 1982;22:775–82. compound ascidian Botryllus schlosseri. Dev Comp Immu-  Muller WZ, Pancer Z, Rinkevich B. Molecular cloning and nol 1998;22:479–92. localization of a novel serine protease from the colonial  Abe Y, Ishikawa G, Satoh H, Azumi K, Yokosawa H. tunicate Botryllus schlosseri. Mol Mar Biol Biotechnol Primary structure and function of superoxide dismutases 1994;3:70–7. from the ascidian Halocynthia rortezi. Comp Biochem  Fagan M, Weissman I. Sequence and characterization of Physiol B 1999;122:321–6. two HSP70 genes in the colonial protochordate Botryllus  Rinkevich B. Invertebrates versus vertebrates innate im- schlosseri. Immunogenetics 1996;44:134–42. munity: in the light of evolution. Scand J Immunol  Pancer Z, Cooper EL, Muller WE. A tunicate (Botryllus 1999;50:456–60. schlosseri) cDNA reveals similarity to vertebrate antigen  Iwanaga S, Lee BL. Recent advances in the innate immunity receptors. Immunogenetics 1996;45:69–72. of invertebrate animals. J Biochem Mol Biol 2005;38:128–50.  Pancer Z, Leuck J, Rinkevich B, Steffen R, Muller I, Muller  Holmskov U. Collections and ﬁcolins: humoral lectins of the WE. Molecular cloning and sequence analysis of two innate immune defense. Annu Rev Immunol 2003;21: cDNAs coding for putative anionic trypsinogens from the 547–78. colonial Urochordate Botryllus schlosseri (Ascidiacea). Mol  Fujita T, Endo Y, Nonaka M. Primitive complement Mar Biol Biotechnol 1996;5:326–33. system-recognition and activation. Mol Immunol 2004;41:  Pancer Z, Leuck J, Rinkevich B, Steffen R, Muller I, Muller 103–11. WE. Cloning of sponge (Geodia cydonium) and tunicate  Endo Y, Takahashi M, Fujita T. Lectin complement system (Botryllus schlosseri) proteasome subunit epsilon (PRCE): and pattern recognition. Immunobiology 2006;211:283–93. implications about the vertebrate MHC-encoded homologue  Kuraya M, Ming Z, Liu X, Matsushita M, Fujita T. Speciﬁc LMP7 (PRCC). Biochem Biophys Res Commun 1996;228: binding of L-ﬁcolin and H-ﬁcolin to apoptotic cells leads to 406–10. complement activation. Immunobiology 2005;209:689–97.  Pancer Z, Diehl-Seifert B, Rinkevich B, Muller WE. A novel  Pancer Z, Gershon H, Rinkevich B. Cloning of a urochor- tunicate (Botryllus schlosseri) putative C-type lectin date cDNA featuring mammalian short consensus repeats features an immunoglobulin domain. DNA Cell Biol (SCR) of complement-control protein superfamily. Comp 1997;16:801–6. Biochem Physiol B 1995;111(4):625–32.  Khalturin K, Becker M, Baruch R, Bosch TCG. Urochor-  Nonaka M, Azumi K, Ji X, Namikawa-Yamada C, Sasaki dates and the origin of natural killer cells: identiﬁcation of a M, Saiga H, et al. Opsonic complement component C3 in the CD94/NKR-P1-related receptor in blood cells of Botryllus. solitary ascidian, Holocynthia roretzi. J Immunol 1999;162: Proc Natl Acad Sci 2003;100:622–7. 387–91.  Pancer Z, Gershon H, Rinkevich B. cDNA cloning of a  Chwieralski CE, WElte T, Buhling F. Cathepsin-regulated putative protochordate FK506-binding protein. Biochem apoptosis. Apoptosis 2006;11:143–9. Biophys Res Commun 1993;197:973–7.  Rinkevich B, Taratakover S, Gershon H. Contribution of  Rinkevich B. Morphologically related allorecognition assays morula cells to allogenic responses in the colonial urochor- in botryllid ascidians. In: Stolen JS, Fletcher TC, Smith SA, date Botryllus schlosseri. Mar Biol 1998;131:227–36. Zelikoff JT, Kaattari SL, Anderson RS, Soderhall K, Weeks ¨ ¨  Jiang Y, Doolittle R. The evolution of vertebrate blood Perkins BA, editors. Techniques in ﬁsh immunology, 4: coagulation as viewed from a comparison of puffer ﬁsh and immunology and pathology of aquatic invertebrates. Fair sea squirt genomes. Proc Natl Acad Sci 2003;24(100): Haven, NJ: SOS Publishers; 1995. p. 17–21. 7527–32.  Huang X, Madan A. CAP3: a DNA sequence assembly  Cima F, Basso G, Ballarin L. Apoptosis and phosphatidyl- program. Genome Res 1999;9:868–77. serine-mediated recognition during the take-over phase of  Ellen A, Nollen A, Morimoto RI. Chaperoning signaling the colonial life-cycle in the ascidian Botryllus schlosseri. Cell pathways: molecular chaperons as stress-sensing ‘‘heat- Tissue Res 2003;312:369–76. shock’’ proteins. J Cell Sci 2002;115:2809–16.  Loukissa A, Cardozo C, Altschuller-Felberg C, Nelson JE.  Quintana FJ, Cohen IR. Heat shock proteins as endogenous Control of LMP7 expression in human endothelial cells by adjuvants in sterile and septic inﬂammation. J Immunol cytokines regulating cellular and humoral immunity. Cyto- 2005;175:2777–82. kine 2000;12:1326–30.  van Eden W, van der Zee R, Prakken B. Heat-shock proteins  Kawahara H, Sawada H, Yokosawa H. The 26 S protea- induce T-cell regulation of chronic inﬂammation. Immunol- some is activated at two points in the ascidian cell cycle. ogy 2005;5:318–30. FEBS Lett 1992;310:119–22.  Yanari MA, Liu J, Zheng Z, Vexler ZS, Lee JE, Fiffard RG.  Orlowski RZ. The role of the ubiquitin-proteasome pathway Antipoptotic and anti-inﬂammatory mechanisms of heat in apoptosis. Cell Death Differ 1999;6:303–13. shock protein protection. Ann N Y Acad Sci 2005;1053:  Ostrowska H, Kruszewski K, Kasacka I. Immuno-protea- 74–83. some subunit LMP7 is up-regulated in the ischemic kidney in ARTICLE IN PRESS 902 M. Oren et al. / Developmental and Comparative Immunology 31 (2007) 889–902 an experimental model of renovascular hypertension. Int glabrata hemocytes. Dev Comp Immunol 2005;29: J Biochem Cell Biol 2006;38(10):1778–85. 393–407.  Mitta G, Galinier R, Tisseyre P, Allienne JF, Girerd-  Davidson B, Swalla BJ. A molecular analysis of ascidian Chambaz Y, Guillou F, et al. Gene discovery and expression metamorphosis reveals activation of an innate immune analysis of immune-relevant genes from Biomphalaria response. Development 2002;129:4739–51.
Pages to are hidden for
"Developmental and Comparative Im"Please download to view full document