Review Zebrafish as an immunological model system Jeffrey A. Yoder a,b,c, Michael E. Nielsen d, Chris T. Amemiya e, Gary W. Litman b,c,f,* a Department of Biology, University of South Florida, 4202 East Fowler Avenue, Tampa, FL 33620, USA b Immunology Program, H. Lee Moff��tt Cancer Center and Research Institute, 12902 Magnolia Drive, Tampa, FL 33612, USA c Children���s Research Institute, Department of Pediatrics, All Children���s Hospital, University of South Florida, 140 Seventh Avenue South, St. Petersburg, FL 33701, USA d Section of Fish Disease, Department of Veterinary Microbiology, Stigb��jlen 4, DK-1870 Frederiksberg, Denmark e Molecular Genetics Program, Virginia Mason Research Center, 1201 Ninth Avenue, Seattle, WA 98101-2795, USA f Department of Molecular Genetics, All Children���s Hospital, 801 Sixth Street South, St. Petersburg, FL 33701, USA Abstract Two decades of research have established the zebrafish (Danio rerio) as a significant model system for studying vertebrate development and gene structure���function relationships. Recent advances in mutation screening, the creation of genomic resources, including the Zebrafish Genome Project and the development of efficient transgenesis procedures, make this model increasingly attractive for immunological study. �� 2002 ��ditions scientifiques et m��dicales Elsevier SAS. All rights reserved. Keywords: Mutation screens Zebrafish genome project Transgenesis Developmental immunology Novel immune-type genes 1. Introduction It has been over 20 years since the broad utility of ze- brafish (Danio rerio) as a vertebrate developmental model was realized. The small size, relatively rapid life cycle and ease of breeding of zebrafish provide unique advantages over other vertebrate model systems. Furthermore, other features of zebrafish such as ex-utero development, optical clarity of early embryos, and rapid growth are being complemented by ever-expanding genetic and genomic resources, including mapping panels, EST databases, and BAC/PAC/YAC librar- ies. Most significantly, zebrafish are invaluable in terms of large-scale mutagenesis screens, which have created thou- sands of mutant lines exhibiting a range of phenotypic varia- tions, and their study has contributed greatly to our under- standing of the development and function of vascular, neuronal and other complex systems [1,2]. Zebrafish share many orthologous genes with mouse and man, which gives this species considerable relevance over other traditional developmental models such as Drosophila or Caenorhabditis elegans, which lack genes that are in- volved in some aspects of neurological processing, adaptive immunity and other functions. Furthermore, recent studies have identified several regions of the zebrafish and human genome that encode the same (or similar) genes [3���5]. These features and other developments in the cell biology and functional genomics of zebrafish are making this model in- creasingly more attractive for immunobiological investiga- tions. Despite considerable progress in identifying genes of the immune system, developing the zebrafish as a new immuno- logical model system is a significant undertaking. In consid- ering such an effort, it is important to bear in mind the critical question ���what will studies using this species offer that cannot be realized using other models?��� Zebrafish offer at least three significant advantages over mammals that can further our understanding of the immune system: (1) the development of lymphoid tissues can be examined (and po- tentially manipulated) at far earlier points in development, (2) effecting single site mutations and screening for mutant phenotypes can be achieved on a larger scale, faster and in a more cost-effective manner than with other systems, and (3) zebrafish possess at least one unusually large family of puta- tive immune genes for which a corresponding mammalian ortholog is not entirely evident [6]. The first point is illus- trated in recent studies of macrophage function in which differential interference contrast microscopy has been used to observe zebrafish macrophages engulf and destroy large amounts of injected bacteria within live embryos [7]. The * Corresponding author. Tel.: +1-727-553-3601 fax: +1-727-553-3610. E-mail address: litmang@allkids.org (G.W. Litman) Microbes and Infection 4 (2002) 1469���1478 www.elsevier.com/locate/micinf �� 2002 ��ditions scientifiques et m��dicales Elsevier SAS. All rights reserved. PII: S 1 2 8 6 - 4 5 7 9 ( 0 2 ) 0 0 0 2 9 - 1

second point is exemplified by the recent report of a series of mutations in rag11 that were identified by direct sequencing of randomly mutagenized animals [8]. The third point is illustrated by the novel immune-type receptor genes (nitrs) and is particularly relevant when considering the evolution of immune-related genes and the relationships between adap- tive and innate immune receptors [6,9]. Although these fea- tures of the zebrafish system are significant, it is neither likely nor necessarily relevant that studies in zebrafish will approach the current levels of functional immunological in- vestigation that are conducted as a matter of routine in mam- mals. Nevertheless, zebrafish hold considerable promise in terms of developmental immunology and disease resistance and susceptibility. This review will summarize the current status of knowledge regarding zebrafish as an immunological model, describe several current areas of focus and discuss future lines of investigation, with particular emphasis on integrating studies of immune function with the unique ad- vantages of the zebrafish model. 2. Adaptive immunity in zebrafish Many features of the immune system of the zebrafish resemble those of higher vertebrates. Microscopic and ultra- structural analyses suggest a general similarity between the thymus in the zebrafish and higher vertebrates [10,11]. In addition, whole mount RNA in situ hybridization with a rag1 probe defines the embryonic thymus (Fig. 1) and is coinci- dent with hybridization patterns of TCRa [12]. Whereas the presence of the thymus and rearranged T-cell antigen recep- tors (TCRs see below) has a long relatively stable phyloge- netic history, the organization of the genes encoding the B-cell antigen receptor and mechanisms whereby they so- matically diversify [13] varies between different vertebrates. Histological examinations of a variety of lower vertebrate species indicate that B cells develop and differentiate in various tissues [11]. The bony fish kidney contains erythroid, myeloid, and lymphoid elements and appears to be the he- matopoietic equivalent of bone marrow in higher vertebrates. Lymphoid cells also are found in spleen (as well as thymus) and at other sites [11]. Zebrafish will be particularly useful for investigations of how B-cell development and diversifi- cation is compartmentalized within different tissues. The structure of the immunoglobulin (Ig) M gene in ze- brafish as well as evidence for IgM heavy chain variable (V) region sequence diversity has been reported [14]. Although the studies are limited, the evidence presented thus far sug- gests a degree of germline and somatic diversity comparable to that seen in other bony fish and higher vertebrates. Studies in other bony fish have identified a second heavy chain isotype linked to the gene encoding the IgM constant region [15]. The N-terminal domain of this constant region is of the �� type and its other seven constant region domains exhibit some homology with mammalian IgD however, it is impor- tant to note that major differences exist between this gene and IgD. Neither the biological significance nor regulation of expression of the second isotype is clear. Based on the pres- ence of a related gene in pufferfish (Amemiya et al., unpub- lished) it is likely that an orthologous gene is present in zebrafish. This model will be particularly useful for issues relating to developmental expression, and the influence of specific genes in the regulation of expression of the different isotypes. Three distinct isotypes of light chain also have been identified in zebrafish [16], all of which show evidence for somatic diversification. Large numbers of sterile Ig light chain gene transcripts are expressed, as has been seen with light chain genes in other species of bony fish. Notwithstand- ing their potential biological significance, the presence of large numbers of sterile transcripts is an important consider- ation in functional interpretations of in situ hybridization data. Information acquired to date for TCRa in zebrafish is limited to sequence characterization of cDNAs (which dem- 1 According to the Zebrafish Nomenclature Guidelines (http://zfin.org/zf_info/nomen.html) zebrafish gene names are in lower case, italic (e.g. rag1) and the protein symbol is non-italic and the first letter is uppercase (e.g. Rag1). Fig. 1. Ultrastructure of embryonic thymus in zebrafish and whole mount RNA in situ hybridization for rag1. A, thymus (red arrows) of a 7-day-old zebrafish. The organ is continuous with the pharyngeal cavity. O, otic vessicle NC, notochord (resized from 400X field) (photo courtesy of A. Zapata.) B, lateral and C, ventral views of a 7-day-old zebrafish embryo that was fixed and hybridized with a rag1 riboprobe, which was detected by a colorimetric assay. The staining pattern (purple) identifies the bilateral thymus (photos courtesy of H. Kawasaki). 1470 J.A. Yoder et al. / Microbes and Infection 4 (2002) 1469���1478