Disponible à / Available at permalink :

157  Download (0)

Texte intégral


- - -

- - -

Dépôt Institutionnel de l’Université libre de Bruxelles / Université libre de Bruxelles Institutional Repository

Thèse de doctorat/ PhD Thesis Citation APA:

Rosa, S. (2010). Positional cloning of the allorecognition gene alr1 in the cnidarian Hydractinia symbiolongicarpus (Unpublished doctoral dissertation).

Université libre de Bruxelles, Faculté des Sciences – Sciences biologiques, Bruxelles.

Disponible à / Available at permalink : https://dipot.ulb.ac.be/dspace/bitstream/2013/210157/5/f7040a47-7c7f-4630-9790-3a98a9ebad59.txt

(English version below)

Cette thèse de doctorat a été numérisée par l’Université libre de Bruxelles. L’auteur qui s’opposerait à sa mise en ligne dans DI-fusion est invité à prendre contact avec l’Université (di-fusion@ulb.ac.be).

Dans le cas où une version électronique native de la thèse existe, l’Université ne peut garantir que la présente version numérisée soit identique à la version électronique native, ni qu’elle soit la version officielle définitive de la thèse.

DI-fusion, le Dépôt Institutionnel de l’Université libre de Bruxelles, recueille la production scientifique de l’Université, mise à disposition en libre accès autant que possible. Les œuvres accessibles dans DI-fusion sont protégées par la législation belge relative aux droits d'auteur et aux droits voisins. Toute personne peut, sans avoir à demander l’autorisation de l’auteur ou de l’ayant-droit, à des fins d’usage privé ou à des fins d’illustration de l’enseignement ou de recherche scientifique, dans la mesure justifiée par le but non lucratif poursuivi, lire, télécharger ou reproduire sur papier ou sur tout autre support, les articles ou des fragments d’autres œuvres, disponibles dans DI-fusion, pour autant que :

Le nom des auteurs, le titre et la référence bibliographique complète soient cités;

L’identifiant unique attribué aux métadonnées dans DI-fusion (permalink) soit indiqué;

Le contenu ne soit pas modifié.

L’œuvre ne peut être stockée dans une autre base de données dans le but d’y donner accès ; l’identifiant unique (permalink) indiqué ci-dessus doit toujours être utilisé pour donner accès à l’œuvre. Toute autre utilisation non mentionnée ci-dessus nécessite l’autorisation de l’auteur de l’œuvre ou de l’ayant droit.

--- English Version ---

This Ph.D. thesis has been digitized by Université libre de Bruxelles. The author who would disagree on its online availability in DI-fusion is invited to contact the University (di-fusion@ulb.ac.be).

If a native electronic version of the thesis exists, the University can guarantee neither that the present digitized version is identical to the native electronic version, nor that it is the definitive official version of the thesis.

DI-fusion is the Institutional Repository of Université libre de Bruxelles; it collects the research output of the University, available on open access as much as possible. The works included in DI-fusion are protected by the Belgian legislation relating to authors’ rights and neighbouring rights.

Any user may, without prior permission from the authors or copyright owners, for private usage or for educational or scientific research purposes, to the extent justified by the non-profit activity, read, download or reproduce on paper or on any other media, the articles or fragments of other works, available in DI-fusion, provided:

The authors, title and full bibliographic details are credited in any copy;

The unique identifier (permalink) for the original metadata page in DI-fusion is indicated;

The content is not changed in any way.

It is not permitted to store the work in another database in order to provide access to it; the unique identifier (permalink) indicated above must always be used to provide access to the work. Any other use not mentioned above requires the authors’ or copyright owners’ permission.


© 2010 by Sabrina F. P. Rosa

Ail rights reserved.


D 03755

he Allorecognition Gene alrl in the Cnidarian Hydractinia symbiolongicarpus

A Dissertation Presented to the Faculty of Sciences of the Free University of Bmssels (Department of Molecular Biology)

In Candidacy for the Degree of Doctor of Philosophy


Sabrina F. P. Rosa

Dissertation Advisors:

Pr. Léo W. Buss Pr. Fadi G. Lakkis Pr. Stephen L. Dellaporta

Pr. Oberdan Léo

February 2010

J^iversite Libre de Bruxelles



Table of content

Acknowledgements... 5 List of figvires... 7 List of tables...9 Chapter I Hydractinia symbiolongicarpus, a cnidarian model to study the

genetics of allorecognition...10 1. The phenomenon of allorecognition in metazoans

2. Allorecognition in Hydractinia symbiolongicarpus (Cnidaria:Hydrozoa) a. The model organism

b. Allorecognition responses in Hydractinia symbiolongicarpus c. Genetics of allorecognition in Hydractinia symbiolongicarpus Chapter II Positional cloning of the alrlInterval...31

1. Introduction

2. Material and Methods 3. Results

4. Discussion

Chapter III Identification of the alrl candidate... 64 1. Introduction

2. Material and Methods 3. Results

4. Discussion


Chapter IV Natural variation of the alrl candidate CDS4...102 1. Introduction

2. Material and Methods 3. Results

4. Discussion

Appendix...129 Perspectives... 130 References... 135



I am deeply indebted to Fadi Lakkis for first introducing me to Hydractinia and supporting me financially for the last 5 years. I would also like to express my deep and sincere gratitude to my supervisors Léo Buss and Stephen Dellaporta, who welcomed me in their laboratory and gave me the great opportunity to work on this exiting project. Their guidance and assistance were an invaluable asset for the

completion of my work. I especially thank Léo for the patience he showed in revising many versions of this dissertation and attempting to teach me how to write clearly and with précision. I am also very gratefül to Oberdan Léo for accepting to be my thesis adviser in this very unusual situation. Finally, I thank my jury members for reading my thesis, notably Pr. Vanden Broeck from the Katholieke Universiteit of Leuven who generously accepted to be my extemal reader.

I could not hâve wished for better labmates and friends than the Buss and Dellaporta lab members. Matthew Nicotra took me under his wing when I first started working on the positional cloning of alrl. He and Maria Moreno taught me ail the molecular biology techniques that I needed to master in order to bring this project to completion. Maria’s infinité knowledge and kindness were a blessing ail along the years I spent at Yale University. Anahid Powell and Ivan Acosta provided great advice and are undoubtedly part of my success. Likewise, I was lucky to hâve Rafael Rosengarten as a labmate. Rafe had to endure countless computer-related questions but always helped me solve problems with a smile. Erica Westerman and Sandra Romero’s positivity always cheered me up and their friendship has been a gift. I will also remember ail the wonderful moments shared with Ana Signorovitch, Monita


Poudyal, and Andrea Gloria-Soria. Christina Glastris, KristaJoy Altland and Sarah Kirschner successively took on the important task of animal care over the years and therefore made this work possible. I was also pleased to supervise the undergraduate thesis of Justin Hayase, who helped me with the design of markers described in

Chapter II and the définition of the candidate genes genomic organization described in Chapter III.

None of this would hâve been possible without the support of Cristina Picardi, Cristian Astudillo, Dawn DeMeo, Maja Duszkiewicz, Jessica David, Kathleen

Murphy, Carlos Galo, Nathalie Vigneron, Déborah Lanterbecq, and Wilton Salazar.

Ail of them contributed to parts of my US adventure in varions and much appreciated ways. While I shared with them many moments of fun and happiness, they were also there for me in the most difficult times.

Finally, I would like to thank my parents for their continuons encouragements and support. It was of great importance to them that I succeeded in my Ph.D. and I therefore dedicate this work to them.

Financial support for this work was provided by NIH grants IR21-A 1066242 and IR56AI079103-01 (to F. G. Lakkis, S. L. Dellaporta, and L. W. Buss), NSF grant IOS-0818295 (to L. W. Buss, S. L. Dellaporta, and M. A. Moreno), and the Joint Genome Institute’s Community Sequencing Program (L. W. Buss and S. L.



List of Figures

Figure 1.1 Moiphology of (a) the body wall of an hydrozoan and (b) a

Hydractinia symbiolongicarpus colony...20

Figure 1.2 Life cycle of Hydractinia symbiolongicarpus...21

Figure 1.3 Rejection response... 23

Figure 1.4 Fusion response... 23

Figure 1.5 Transitory fusion (TF) responses... 24

Figure 1.6 Inbreeding program designed to study the transmission genetics of allorecognition in Hydractinia... 27

Figure 1.7 Genetic map of the allorecognition coraplex...29

Figure 2.1 Chromosome walk from the marker AFLP 194...47

Figure 2.2 Définition of the alrl interval in the ARC-/haplotype... 48

Figure 2.3 Comparison of the ARC-/and ARC-r alrl interval... 53

Figure 2.4 Chromosome walk from the marker AFLP 18... 54

Figure 2.5 Mapping of the proximal breakpoint of individual 431-2 to a 518 bp interval... 56

Figure 2.6 Structural variation between the ARC-/and ARC-r haplotypes in the distal end of the alrl interval... 57

Figure 3.1 Gene content of the alrl interval... 78

Figure 3.2 Candidate gene CDS1... 83

Figure 3.3 Structural variation between the ARC-/and -rhaplotypes in the CDSl région...84

Figure 3.4 Candidate gene CDS4... 86


Figure 3.5 CDS4 expression profile... 87

Figure 3.6 Alignment of CDS4 domains I and II to alr2 Ig-Iike domains and eanonical Ig domains... 91

Figure 3.7 Immunoglobulin superfamily gene complex... 92

Figure 3.8 Unrooted UPGMA tree of the Ig-Iike domains found in the Hydractinia IgSF complex... 94

Figure 4.1 Wild-type CDS4 alleles... 111

Figure 4.2 MUSCLE alignments of domain I and II in 20 CDS4 alleles...115

Figure 4.3 Amino acid sequence variabilité in 20 CDS4 alleles... 116

Figure 4.4 Exon 4 variability and alternative splicing...117

Figure 4.5 Variation in sequence encoding the signal peptide...117

Figure 4.6 Variability of the 3’ untranslated région of CDS4...118

Figure 4.7 Variability of the 5’ untranslated région of CDS4...118

Figure 4.8 Pairwise comparisons of CDS4 allele Ig-like domains I (a.) and II (b.) with members of the IgSF complex... 119

Figure 4.9 Distribution of synonymous and non-synonymous mutations in 20 CDS4 alleles...120

Appendix MUSCLE curated alignments of the 20 CDS4 alleles... 133


List of Tables

Table 2.1 Recombinants in the alrl genomic région...43

Table 2.2 Recombinant breakpoints defining the minimum alrl interval...49

Table 2.3 BAC end markers... 50

Table 2.4 Minimum tiling path of the 194 interval...52

Table 2.5 Minimum tiling path of the 18 interval (ARC-/haplotype)... 55

Table 2.6 Markers designed for the physical mapping of the alrl interval...58

Table 3.1 Potential coding sequences (pCDS’s) in the alrl genomic région...80

Table 3.2 Primers used to amplify predicted CDS’s in the alrl-f genomic région... 82

Table 3.3 In silico analyses on CDS4 domain I and II...89

Table 3.4 Ig-containing predicting genes in the 194-minimum tiling path...93

Table 4.1 RACE and RT-PCR primers used to amplify wild-type CDS4 alleles... 112

Table 4.2 Codons under positive sélection in CDS4... 121

T able 4.3 Matching CDS4 alleles in wild-type colonies... 122








The phenomenon of allorecognition in metazoans

Allorecognition is defined as the capacity to distinguish between one’s own tissue and those of conspecifics based on genetic relatedness. The phenomenon of allorecognition is ubiquitous amongst colonial invertebrates and widespread amongst aclonal taxa (Buss, 1982; Grosberg, 1988; Raftos, 1990; Sawada et al., 2004; Boehm, 2006; Khalturin and Bosch, 2007; Küm et al., 2007). Self^non-self discrimination in invertebrates has been described in nearly ail investigated species of the phylum Porifera (sponges) (Femandez-Busquets and Burger, 1999), in the phylum Bryozoa (sea mats) (Chaney, 1983; Hughes et al., 2004) and in the phylum Cnidaria,

specifically in the colonial and sedentary organisms belonging to the classes Anthozoa (sea anémones and corals) (Theodor, 1976; Hildemann et al., 1977;

Hildemann et al., 1980; Ayre and Grosberg, 1995) and Hydrozoa (Buss et al., 1984;

Lange et al., 1989; Shenk, 1991). In the phylum Chordata, allorecognition has been described in the sub-phylum Urochordata (tunicates) (Scofield et al., 1982; Saito et al., 1994; Rinkevich et al., 1995) and in jawed vertebrates. In the latter,

allorecognition is a feature of adaptive immunity and is represented by the MHC- based histocompatibility (Klein, 1986; Hemandez-Fuentes et al., 1999; Janeway Jr et al., 2005). While allorecognition responses in colonial invertebrates take the form of natural transplantation phenomena, with colonies fiising or rejecting upon growing into allogeneic contact, jawed vertebrates are aclonal and do not encounter allogeneic tissues except in the contexts of pregnancy or transplantation.

The capacity to respond to natural transplantation is of spécial importance for sessile, colonial organisms, which often inhabit densely populated benthic


communities. Being clonal and hence having no inhérent limit in size, sedentary and colonial animais are subject to intense intraspecific compétition in such environments.

Adult colonies are prone to grow in contact with conspecifics and undergo natural transplantation, and newly metamorphosed individuals are very likely to encounter conspecifics during their settlement (Yund et al., 1987; Westerman et al., 2009).

Fusion confers an immédiate increase in size and, in tum, higher fecundity and increased survivorship, and thus appears as advantageous in this type of compétitive environment (Buss, 1982). However, fusion is accompanied by a cost (Buss, 1982).

Colonial organisms possess mitotically active multipotent stem cells competent to produce gametes (Bosch and David, 1987; Bode, 1996; Müller et al., 2004). Upon fusion, stem cells from one colony invade the host colony and may contribute

disproportionately to the production of gametes from the chimera (Buss, 1982, 1983).

This process is known as germline parasitism and is well-known in hydroids,

botrylloid ascidians and cellular slime molds (Pancer et al., 1995; Stoner et al., 1999;

Strassmann et al., 2000). Germline parasitism may be ameliorated if fusion is

restricted to self or close kin, as is provided by a highly polymorphie allorecognition System (Buss, 1982, 1983, 1987).

The expression of histocompatibility in vertebrates might at first seem

anomalous, as it is difficult to imagine why évolution would hâve favored such a trait.

Natural transplantation and cell parasitism are however not exclusive to colonial invertebrates. Vertebrates also manifest several types ofnaturally occurring allogenic encounters. The two most common examples are hematopoietic chimerism in

dizygotic twins (Owen, 1945) and fetomatemal chimerism associated with pregnancy (Maloney et al., 1999; Blanchi, 2000). In these phenomena, a cell lineage originales


from kin and générâtes a chimera. While the chimeric twins are tolérant to skin graft from each other (Anderson et al., 1951; Billingham et al., 1952), human

microchimerisms are not always tolerated and might lead to autoimmunity (Bianchi, 2000; Anderson and Matzinger, 2001). Cancers transmissible upon physical contact, notably the canine venereal sarcoma in dogs and the devil facial tumor disease in Tasmanian devils (Murgia et al., 2006; Murchison, 2008), also constitute naturally occurring allografts in vertebrates. Thus, maintenance of individual integrity, in addition to defense against pathogens, may hâve contributed to the évolution of vertebrate immunity (Bumet, 1959; Buss and Green, 1985; Rinkevich, 2004).

Invertebrate allorecognition and vertebrate transplant rejection display a similar phenomenology and share the ability to discriminate between highly

polymorphie ligands. Deeper knowledge of invertebrate allorecognition responses is essential to résolve whether these phenomena are related to the evolutionary origin of vertebrate adaptive immunity (Du Pasquier, 2004; Kasahara et al., 2004). To date, no conclusive evidence has been presented that homologs of the MHC or other élément of the adaptive immune System exist in invertebrates (Azumi et al., 2003; Du

Pasquier, 2004; Du Pasquier et al., 2004; Kasahara et al., 2004; De Tomaso et al., 2005; Nicotra et al., 2009). The search for invertebrate homologs of molécules

involved in allorecognition has however mostly been based on techniques such as low stringency cloning and comparative genomics analyses of the few invertebrate

genomes available (Flajnik and Kasahara, 2001; Abi-Rached et al., 2002; Azumi et al., 2003). Molécules demonstrated to regulate invertebrate allorecognition Systems, for which straightforward comparisons of evolutionary relatedness can be made, are still very scarce, as detailed below (De Tomaso et al., 2005; Nicotra et al., 2009).


Irrespective of whether the allodeterminants of invertebrates are themselves related to molécules involved in vertebrate allorecognition, characterization of the invertebrate Systems may bear on our understanding of vertebrate innate immunity.

For example, the understanding in vertebrates of how the components of innate immunity recognize allogeneic tissue and regulate graft rejection by triggering the alloimmune adaptive response is very limited. Invertebrate allorecognition responses are assumed to be based on innate immune mechanisras (Rinkevich, 2004; Litman et al., 2005) and a growing body of literature on genes and effector pathways of innate immunity suggest that components of these Systems hâve been largely been conserved throughout évolution (Marino et al., 2002; Azumi et al, 2003; Shida et al., 2003; Rast et al., 2006; Miller et al., 2007). Thus, characterization of invertebrate

histocompatibility Systems might shed light on vertebrate immune functions even if the récognition molécules themselves are unrelated.

Genetic models to study allorecognition hâve been developed in three major animal groups: jawed vertebrates, tunicates, and cnidarians. Of these, only the vertebrate MHC-based histocompatibility is well characterized (Janeway Jr et al., 2005). In the two invertebrate model organisms, the tunicate Botryllus schlosseri and the cnidarian Hydractinia symbiolongicarpus, which will be detailed in the second part of this introduction, the genetic mechanisms underlying the allorecognition responses hâve just begun to be unveiled (De Tomaso et al., 1998; De Tomaso and Weissman, 2003b; Cadavid et al., 2004; De Tomaso et al., 2005; Nyholm et al., 2006;

Powell et al., 2007; Nicotra et al., 2009).


Botryllus schlosseri is a basal chordate from the subphylum of the tunicates, and hence represents one of the closest living invertebrate relatives to vertebrates.

Botryllus has been used as a model to study allorecognition for over a century.

Histocompatibility responses, which are highly spécifie and capable of discriminating between more than a hundred alleles (Scofield et al., 1982; Grosberg and Quinn,

1986; Rinkevich and Saito, 1992; Yund and Feldgarden, 1992; Rinkevich et al., 1995;

De Tomaso et al., 2005), are controlled by a single chromosomal interval bearing a highly polymorphie locus called the FuHc(fusion/histocompatibility) locus (De Tomaso and Weissman, 2003b; De Tomaso et al., 2005). Fusion rules are such that fusion occurs when two colonies share one or both alleles at the FuHclocus and rejection occurs when no alleles are shared. The FuHclocus was identified using a positional cloning approach. The FuHclocus was mapped to a 1.0 cM genetic interval by using partially inbred strains of Botryllushomozygous for different FuHcalleles and a strategy combining bulk segregant analysis and amplified fragment length polymorphism (AFLP) markers. A genomic walk then produced a physical map of the interval and sequencing of BAC clones spanning over 1 Mb of that interval allowed the identification of the FuHcgene (De Tomaso and Weissman, 2003a; De Tomaso and Weissman, 2003b; De Tomaso et al., 2005). FuHcencodes a 1008 amino acids transmembrane protein bearing a signal sequence, two extracellular immunoglobulin (Ig)-like domains and onw epidermal growth factor (EGF) domains, followed by a transmembrane domain and a cytoplasmic tail with no known signaling domains.

FuHcis a highly polymorphie locus and its polymorphism correlates to the outcome of allorecognition responses in laboratory fines and one field-collected colony, with fiising animais sharing at least one FuHcallele and rejecting animais sharing no allele. The protein is not similar to any known protein in other genomes and bears no


similarity to the vertebrate MHC (De Tomaso et al., 2005; McKitrick and De Tomaso, 2009).

A second gene named/ester lies within the FuHc chromosomal interval (Nyholm et al., 2006). Pester encodes a 368 amino acids transmembrane protein bearing a signal sequence, seven extracellular exons, one of which contains a SCR domain, three transmembrane helices and a short intracellular tail. Similarly to FuHc, /ester does not show similarity to known proteins from other genomes nor to

components of the vertebrate adaptive immunity. Pester also displays considérable polymorphism (Nyholm et al., 2006). In addition to sequence polymorphism,/e5^er is also somatically diversified by extensive alternative splicing (Nyholm et al., 2006;

McKitrick and De Tomaso, 2009). Six of its eleven exons can be altematively spliced in ail combinations, creating 64 possible splice variants (McKitrick and De Tomaso, 2009). The significance of the multiple splice variants is still unknown, but different variants are hypothesized to play different rôles in the allorecognition response

(Nyholm et al., 2006; McKitrick and De Tomaso, 2009). Unlike FuHc,/ester is not an allodeterminant. Indeed,/e5ter’s polymorphism does not correlate to allorecognition responses, as colonies sharing alleles can reject and colonies with no shared alleles can fiise (Nyholm et al., 2006). Pester is however shown to be involved in

allorecognition. Monoclonal antibody acute interférence and siRNA-mediated knockdown experiments suggest that/ester plays at least a dual rôle in the allorecognition response, as both an activating receptor required to initiate the allorecognition response and an inhibitory receptor required for recognizing self and inhibiting a rejection response (Nyholm et al., 2006). Pester's exact rôle and potential interaction with FuHc await to be determined.


The second invertebrate organism for which a genetic model of

allorecognition is available is Hydractinia symbiolongicarpus, and is described in details in what follows.

2. Allorecognition in Hydractinia symbiolongicarpus (CnidariarHydrozoa)

a. The model organism

The phylum Cnidaria is a basal metazoan phyla, predating the évolution of ail bilaterian animais and itself predated only by the sponges and placozoans (Chen et al., 2000; Narbonne, 2005). Genome (Sullivan et al., 2006; Putnam et al., 2007) and EST (Kortschak et al., 2003; Technau et al., 2005; Soza-Ried et al., 2009) sequencing projects, along with targeted study of spécifie gene families (Galliot, 2000;

Wikramanayake et al., 2003; Martindale et al., 2004; Seipel et al., 2004; Reber-Muller et al., 2006) hâve revealed that cnidarians display comparable molecular complexity to that of higher metazoans, including most signaling pathways and transcription factors involved in the patteming and development of bilaterians. For that reason, cnidarians hâve been used in a wide range of studies focusing on the évolution of biological processes such as development (Ferrier and Holland, 2001; Steele, 2002;

Guder et al., 2006; Hoffmeister-Ullerich, 2007; Khalturin et al., 2007), reproduction (Galliot and Schmid, 2002; Steele and Dana, 2009) and immunity in bilaterian animais, including the phenomenon of allorecognition. A model of choice to study allorecognition is the hydroid Hydractinia, which is easily cultured in the laboratory, fast growing, and has a short, inducible life cycle favorable to genetics experiments.

While over 30 Hydractinia species are known worldwide (“The Cnidarian Tree of


Life Project”, http://cnidtol.com/node/95), most of the published research on

allorecognition is based on the North American H. symbiolongicarpus, the organism used in this work.

Hydractinia symbiolongicarpusis found in the shallow water soft bottom benthos of the temperate North Atlantic Océan. The hydroid is an epibiont of the shells occupied by the hermit crab Pagurus longicarpus (Buss and Yund, 1989;

Cunningham et al., 1991) and is rarely found on other available substrates (Karlson and Shenk, 1983). A fiilly covered shell is usually occupied with only one

Hydractiniaindividual, although two large colonies can be found on a same shell in rare instances (Buss and Yund, 1988; Yund and Parker, 1989). Shells with a single colony reflects the victors in intraspecific compétition, as a high frequency of shells bear multiple recruits in close vicinity on a same shell (Yund et al., 1987).

Like ail members of the phylum Cnidaria, Hydractiniaare radially

symmetrical and diploblastic animais (Figure 1.1 a). Two well-developed cellular layers separated by a non-cellular layer, called the mesoglea, comprise ail tissue: the ectoderm (outer layer) and the endoderm (inner layer). The ectoderm contains epitheliomuscular cells, rare neurosensory cells, and cnidocytes. Cnidocytes, also called nematocytes, are the defming characteristic of the phylum Cnidaria (Hessinger and Lenhoff, 1988). They are stinging cells containing a spécial organelle, called the cnidocyst or nematocyst. A nematocyst consists of a capsule under pressure enclosing a coiled filament with barbs, which, when discharged, delivers a stinging venom that either serves to capture and paralyze preys or as a defense mechanism from predators (Tardent, 1995). The cell content of the endoderm is composed of digestive


epitheliomuscular cells and glands cells. Nerves underlie both cell layers. Interstitial cells are found in both cell layers and constitute a pool of undifferentiated stem cells that continuously give rise to the gametes, gland cells, nematocytes and nerves (David and Plotnick, 1980; Bosch and David, 1986). While many species of hydrozoans display two different morphological forms, the polyp and the médusa, Hydractinia is exclusively sessile and lacks the médusa form.

Hydractinia colonies (Figure 1.1b) are composed of polyps specialized for

feeding (gastrozooids), reproduction (gonozooids) and defense (dactylozooids). The body walls of feeding polyps surround the gastrovascular cavity, which only has one opening at the oral end. Polyps arise asexually from a basal coenosarc known as the ectodermal mat, which consists of two ectodermal cell layers encasing a network of endodermal gastrovascular canals. These canals form the gastrovascular System that allows the exchange of food particles throughout the entire colony. Structures called stolons, which are extensions of the gastrovascular canals, may emerge from the mat and elongate at their tips. Encounters between colonies, therefore, can arise either between ectodermal mats and/or peripheral stolons (Hauenschild, 1954, 1956; Millier,

1964; Buss et al., 1984; Buss and Grosberg, 1990). In the laboratory, colonies can be explanted from the hermit crab shell and tied onto a glass microscope slide with a string until they adhéré to their new substrate. After removal of the string, colonies will expand by asexual growth to eventually cover the ftill surface of the glass slide (Blackstone and Buss, 1991).

Hydractinia''% life cycle is completed in about three months under laboratory culture conditions (Figure 1.2). Hydractinia colonies are dioecious and male and


a. b.

Cillary receptor . Epltheliomuscularcell

^ Intersititial cell

^ Receptor cell

Cilium Granular sécrétions Longitudinal muscle fibrils Basal laminae Circuler muscle fibrils Nucléus Neuron

Enzymatic gland cell


gastrovascular canais

Figure 1.1 Morphology of (a) the body wall of an hydrozoan (longitudinal section), from Ruppert and Bames ( 1996) and (b) a Hydractinia symbiolongicarpus colony, modified from Buss and Blackstone (1991).

female individuals release gametes daily, 1 to 2 hours after sunrise (Ballard, 1942), or, in the laboratory, after being subjected to a light-dark régime. The gametes are

fertilized in the water column and sink to the bottom, where the embryos develop into planula larvae within 2 to 3 days. In natural conditions, crawling planulae creep onto a nearby sand grain and attach themselves to the shell of nearby hermit crabs by firing nematocysts (Weis et al., 1985; Weis and Buss, 1987). Bacteria commonly found on gastropod shells induce metamorphosis of the planulae into primary polyps (Müller, 1973; Müller and Leitz, 2002). In the laboratory, metamorphosis can be induced by replacing the bacterial due with a Chemical due, i.e. by incubating the larvae in a solution of seawater containing 53 mM césium chloride (Müller, 1973). The onset of allorecognition occurs at the late planula stage, allowing newly metamorphosed animais to respond to allogeneic encounters as soon as they settle on their new substrate (Lange et al., 1992). After metamorphosis, primary polyps develop into juvénile colonies, which grow by asexual itération until they cover the available


surface of the shell or slide they settled on. Sexual maturity is reached within 3 months, depending on the individual growth rate of the colony.

Figure 1.2 Life cycle of Hydractinia symbiolongicarpus, from Müller and Leitz (2002).

b. Allorecognition responses in Hydractinia symbiolongicarpus

When two Hydractinia colonies corne in contact with each other, three major outcomes are observed. The colonies can fuse, reject or undergo transitory fusion, an intermediate phenotype in which colonies first fiise to later separate. In a rejection response (Figure 1.3), nematocytes are recruited to the contact border and


nematocysts forai batteries on either side on the contact zone. Ectodermal cells of the opposing colonies fail to adhéré and, 2 to 3 hours post-contact, the nematocysts are discharged, causing damage to both competitors. Rejection responses between incompatible colonies can be of two kinds, passive or aggressive, depending on the colony morphology (Buss et al., 1984; Buss, 1990). Stoloniferous colonies, i.e.

colonies bearing numerous stolons, develop aggressive rejection responses (Figure 1.3 d). These responses are characterized by the appearance of hyperplastic stolons, i.e.

stolons swollen with nematocytes (Müller, 1964; Ivker, 1972; Buss et al., 1984).

Hyperplastic stolons discharge their nematocysts to effect tissue destruction (Ivker, 1972; Lange et al., 1989). Stolonless colonies, on the other hand, develop passive rejection responses (Figure 1.3 e), in which contacting mats of incompatible colonies do not fuse, but cease to grow along the contact border (Buss et al., 1984).

Whereas the rejection response involves a failure of tissues to adhéré, the fusion response (Figure 1.4) is characterized by adhesion of the ectodermal cell layers followed by endodermal canals. Gastrovascular continuity is established within 30 minutes post-contact, generating a permanent chimeric colony (Hauenschild, 1954,

1956; Müller, 1964; Buss et al., 1984; Lange et al., 1989).

Transitory fusion is an intermediate phenotype between fusion and rejection (Hauenschild, 1954, 1956; Shenk and Buss, 1991; Cadavid et al., 2004). While a wide range of intermediate phenotypes hâve been observed amongst wild type colonies, only two types of transitory fusion responses occur in Hydractinia inbred Unes

(Powell et al., 2007). In type I transitory fusion (Figure 1.5), colonies remain fused for 1 to 3 days before separating. A grey line develop across the contact zone, followed


Figure 1.3 Rejection response. (a) Nomarski interférence micrograph of stolon tip approaching opposing colony, (b) contact, (c) nematocysts discharge; (d) aggressive rejection; (e) passive rejection. (Photo crédit: a, b, c and e, R.G. Lange and G. Plickert;

d, R.K. Grosberg).

Figure 1.4 Fusion response (Photo crédit: L.F.


by occlusion of the gastrovascular canals, which then branch laterally, connect with other canals of the colony to which they belong, and form a ring canal inside both colonies. Complété tissue séparation between both colonies is achieved within 48 hours post-fiision and is permanent. In type II transitory fusion (Figure 1.5), colonies similarly remain fused for several days before showing signs of séparation, which, in


this case, consists of local loss of gastrovascular continuity and ectodermal séparation followed by refusion of the separated zone. Séparation is not permanent as the cycle of fusion-separation-refusion continues indefmitely. Séparation of colonies in transitory fusion does not involve the deployment of nematocytes and tests for alloimmune memory are uniformly négative (Poudyal et al., 2007).

Figure 1.5 Transitory fusion (TF) responses. (a) Initial fusion, (b) TF type I, (c) TF type II, séparation phase, (d) TF type II, re-fusion of a gastrovascular canal. (Photo crédit: A.E. Powell).

In the laboratory, allorecognition responses can be assessed by using three types of fiisibility assays: the colony (Millier, 1964; Buss et al., 1984) the polyp (Lange et al., 1992), and the embryonic assays (Poudyal et al., 2007). The colony assay consists in explanting a 3-5 polyps-containing piece of mat from each colony to be tested onto a glass slide and placing them 0.5 mm from each other. The colony fragments are maintained in position by a string until they adhéré to the slide 24 to 48


hours later. Colonies are observed daily until they contact each other and display their flisibility phenotype. In the polyp assay, polyps from the colonies to be tested are removed and held against each other with their eut ends in contact. If the polyps are compatible, ectodermal and endodermal cells adhéré and gastrovascular continuity is established within 12 to 24 hours. Incompatible polyps do not adhéré and separate. In the embryonic assay, molecular markers are used to detect chimerism in well-mixed embryonic chimeras generated from dissociated blastomeres. Chimerism is

undetectable by 4 weeks of âge in chimeras resulting from histoincompatible pairings.

Histocompatible pairings remain chimeric beyond 4 weeks of âge. In addition, histoincompatible chimeras display low survivorship and delays in growth (Poudyal et al., 2007).

c. Genetics of allorecognition in Hydmctinia symhiolongicarpus

The transmission genetics of allorecognition in Hydractinia has been studied for over half a century. Hauenschild (1954, 1956) was the first to address the

question. Based on sériés of breeding experiments using 4 field-collected colonies, he proposed that allorecognition responses could be explained by the existence of a single locus expressing alleles of differing effect, such that fusion occurred when a particular combination of alleles in the interacting partners exceeded a threshold value. This scheme was consistent with the bulk of his data, albeit with several exceptions. Moreover, the occurrence of transitory fusion phenotypes in Fl, F2 and F1 X F2 offspring were not consistently incorporated into the model because the intermediate phenotypes were first treated as fusions (Hauenschild, 1954) and later as rejections (Hauenschild, 1956). Du Pasquier (1974) reviewed Hauenschild’s data


under the simpler assumption that fusion arose if colonies shared an allele and rejected if they shared none. He showed that most, but not ail, inconsistencies in Hauenschild’s results could be accommodated by a model of linked loci if one assumed that some of Hauenschild’s parental colonies were recombinants. Grosberg et al. (1996) presented fusion frequencies from a sériés of wild-type crosses.

Fusibility assays were observed at weekly intervals, effectively insuring that transitory fusion phenotypes would be mis-scored. Grosberg et al. (1996) used a computer simulation to illustrate that, if one assumed multiple unlinked

allorecognition loci and an additive dosage scheme to map génotype to phenotype, that 4 or more such loci would be required to accommodate their results.

The Buss laboratory undertook a long term breeding program to generate defmed genetic stocks from wild colonies (Mokady and Buss, 1996; Cadavid, 2001;

Cadavid et al., 2004; Powell et al., 2007) (Figure 1.6). Two wild-type animais were crossed and progeny brother-sister inbred under sélection for fusibility for 8

successive générations. These animais were presumed (and later confirmed) to be fixed for the ‘y” allorecognition haplotype. A sixth génération animal from the/near isogenic line crossed to a wild type and a second round of eight générations of brother-sister inbreeding was performed under sélection for fusibility within the line but failure to fuse the original line. This line was presumed (and subsequently confirmed) to be fixed for the “r” allorecognition haplotype. A third line was generated by identifying an F2 animal from the original outcross that fused to both inbred lines. This animal was crossed into the f near isogeneic line for 4 subséquent rounds of repeated backcrosses under sélection for animais that fiised to both inbred lines. These animais were presumed (and later confirmed) to segregate for the/and r



Inbred Une 41

Figure 1.6 Inbreeding program designed to study the transmission genetics of allorecognition in Hydractinia, from Cadavid (2005). The shaded zone represents the work of Mokady and Buss (1996) and the non-shaded zone represents the continuation of the inbreeding program by Cadavid et al. (2004). Circles represent females, and squares represent males. Dot-

pattemed symbols indicate wild-type colonies; black-filled, hollow, and line-crossed symbols represent the 33 line, 41 line and congeneic line, respectively. Horizontal Unes represent matings, and vertical Unes represent offspring ffom such matings. Identification numbers indicate the génération to which a given individual belongs. WT, wild-type; BC, backcross.

These Unes were used in three sets of classical breeding experiments that differed principally in the size of the progeny populations. Mokady and Buss (1996) reported fusion and rejection frequencies in a conventional intercross/ backcross/

incross design based on the original outcross mentioned above. A total of 13 crosses were made, but ail involved small sample sizes. Results of ail 13 crosses were consistent with allorecognition segregating as a single locus trait such that colonies fuse if they share one or more alleles and reject otherwise.


A second sériés of crosses were reported by Cadavid et al. (2004). Using the near congeneic line, crosses were made and >300 progeny tested for fusibility.

Ségrégation was in accord with single locus ségrégation with the exception of a small number of animais that exhibited transitory fusion phenotypes. Genotyping using linked markers (see below) and crosses between recombinant animais established that the chromosomal interval harbored two linked loci and the transitory fusion

phenotype arises when one or the other of the two genes is rendered homozygous by recombination. The two loci were designated allorecognition 1 {alrl) and

allorecognition 2 {air2). The chromosomal interval spanning these loci was designated the allorecognition complex (ARC). Animais of the/inbred line are

homozygous for the/haplotype (f/J) and those of the r inbred line are homozygous for the r haplotype (r/r).

A third sériés of crosses (Powell et al., 2007) generated a progeny population of >3,500 animais. No additional loci were identified, but over 100 recombinant animais were recovered that allowed fürther clarification of the rules that govem fusibility within the defined laboratory lines. Colonies fuse when they share at least one allele at both alrl and alr2 loci and reject when no allele is shared at either alrl or alr2. Transitory fusion type I occurs when colony share an allele at the alr2 locus, but share no alleles at the alrl locus. Conversely, transitory fusion type II occurs when colonies share an allele at alrl, but not at alr2.

The ARC was localized to a defined physical chromosomal interval by genetic mapping (Cadavid et al., 2004; Powell et al., 2007). Molecular markers linked to the ARC were generated to this purpose (Cadavid et al., 2004; Poudyal et al., 2007;


Powell et al., 2007). Bulked segregant analysis, or BSA (Michelmore et al., 1991) is widely used to generate markers linked to a gene that segregates in a simple

Mendelian fashion. The virtue of the approach is that it can be applied to an anonymous System. This technique was developed in plants (Churchill et al., 1993;

Zhang et al., 1994; Van der Lee et al., 2001) but thereafter spread to a wider range of studies (De Tomaso and Weissman, 2003b; Cadavid et al., 2004; Kano, 2007; Baxter et al., 2008; Fiumera et al., 2009). The BSA methodology involves pooling together DNA from a single cross segregating for the trait of interest, such that each bulk segregant pool represents a different phenotype. These pools can be screened using a random marker technology, such as amplified fragment length polymorphism (AFLP) (Vos et al., 1995). Markers that are présent in a pool but absent from other pools are linked to the gene of interest with high confidence and can be used in subséquent linkage mapping analysis. Following that approach, Cadavid et al. (2004) and Powell et al. (2007) developed five AFLP markers polymorphie between pools of congeneic individuals differing in their fusibility phenotypes. Each AFLP marker was

developed as a co-dominant PCR marker and linkage analysis using marker génotype and fusibility phenotype produced a 1.7 cM genetic map (Figure 1.7) (Powell et al., 2007).

Figure 1.7 Genetic map of the allorecognition complex. Distances are in centimorgans.

Numbers represent the polymorphie AFLP markers developed from the bulked segregant analysis.

V alrl

<0.1 1.1 0.2 0.1 0.1 0.1 cM


Identification of the genes responsible for allorecognition in invertebrates bas been a longstanding goal of comparative immunology. The genetic mapping of the allorecognition interval in Hydractinia was a crucial step towards the identification of the genes controlling the allorecognition response and allowed Nicotra et al. (2009) to identify the alr2 allodeterminant by positional cloning. Positional cloning using BAC libraries built ffom the Hydractinia inbred Unes defined a 350 kb a/r2-containing physical interval, circumscribed by 6 recombination breakpoints, in which the alr2 gene was identified. Alr2, a novel hypervariable immunoglobulin superfamily receptor, encoded a 672 aa type I transmembrane protein with three extracellular immunoglobulin (Ig)-like domains. The identification of alr2 constituted a substantial achievement, as alr2 represented, with the Botryllus FuHc allodeterminant and fester, the only invertebrate allorecognition genes identified.

This dissertation focuses on the identification of the second Hydractinia allodeterminant, alrl, using a positional cloning strategy. This effort begins with only (a) a linkage map, (b) an available large-insert library, and (c) a pool of recombinant animais. Chapter II describes the isolation by positional cloning of a 300.8 kb alrl- containing chromosomal interval. The analysis of that interval for its gene content and the identification of a primary alrl candidate, CDS4, as a novel hypervariable IgSF protein, are described in Chapter III. In Chapter IV, variation of CDS4 in natural population is investigated and CDS4 sequence polymorphism is shown to predict fusibility within and between laboratory Unes and wild-type isolâtes, thus confirming the rôle of CDS4 as the alrl allodeterminant.








The increasing number of genome sequences released to public databases has made the identification of candidate genes routine work. The use of such databases, and of the many bioinformatics tools subsequently developed, has undoubtedly

accelerated the discovery of genes that can be identified based on sequence similarity.

Relying only on the similarity at the nucleic acid level has had inhérent limits as more distantly related homologues could be similar at the amino acid sequence level but hâve significantly different nucléotide sequences. Moreover, genes with the same ancestry might hâve diverged sufficiently to share very little or no sequence similarity at ail, as illustrated for example by the members of the cytokine receptor superfamily (Miyajima et al., 1992), the immunoglobulin heavy-chain variable-region (VH) gene family (Ota and Nei, 1994), the natural killer (NK)-cell receptors spécifie for major histocompatibility complex (MHC) class I molécules (Barten et al., 2001), and the bindin genes involved in sperm-egg attachment and fusion in echinoids (Zigler and

Lessios, 2003). The only indication of homology between distantly related proteins has often been shared three-dimensional structure (Murzin, 1998; Thomton et al.,


Structural similarity, and not sequence similarity, might be the rule for the highly diversified allorecognition Systems observed in different taxa (Loker et al., 2004; Litman et al., 2005; Dishaw and Litman, 2009). Different animal groups hâve evolved lineage-specific innovations in allorecognition Systems and extremely diverse modes of histocompatibility hâve been described in the three taxa (jawed vertebrates, protochordates, and cnidarians) for which genetic data is available. Polymorphie


MHC molécules, members of the immunoglobulin (Ig) superfamily, were shown to médiate histocompatibility responses in jawed vertebrates (Janeway Jr et al., 2005). In the tunicate Botryllus, two genes, FuHc (De Tomaso et al., 2005) and /ester (Nyholm et al., 2006), play a rôle in self/non-self récognition. The FuHc gene encodes a type I transmembrane protein with three extracellular domains, one epidermal growth factor (EGF)-like and two Ig-like domains, while/ester encodes a transmembrane protein with a short consensus repeat (SCR) domain. In the hydroid Hydractinia, one of the two genes mediating allorecognition responses, alr2, has been described as a type I transmembrane receptor with three extracellular Ig-like domains. The extreme

diversity displayed by these molécules explained why, for example, previous attempts to identify the allodeterminants in the tunicate Botryllus by using MHC-like

molécules in a cloning by homology approach did not reveal the molecular nature of allorecognition (Fagan and Weissman, 1998; Rinkevich, 2004). No sequence

similarity was found between the above-mentioned allorecognition molécules outside of the Ig domains. Members of the immunoglobulin gene superfamily hâve been commonly recruited throughout évolution to serve diverse immune functions (Litman et al., 2005) and hâve constituted to date the only conserved unit of these

allorecognition Systems.

Lack of conservation might doom naive candidate gene approaches. The identification of novel genes, or of distantly related homologues, required an unbiased forward genetics approach based on a known genetic System. To overcome this problem, researchers hâve employed a positional cloning strategy, in which the gene of interest was first mapped to a région of the genome by linkage analysis, and then fiirther circumscribed by the development of tightly linked molecular markers along a chromosome walk. Large-inserts genomic libraries were used to perform the


chromosome walk, which resulted in the définition of a minimum tiling path. The minimum tiling path was sequenced and genes so localized were further explored using experimental and in silico approaches.

Positional cloning has been used in many instances with success, both in the pre- and post-genomics era. The identification of human disease genes greatly contributed from the development of the positional cloning techniques. Genes responsible for simple monogenic diseases such as cystic fibrosis (Rommens et al.,

1989a; Rommens et al., 1989b) or Huntington’s disease (Richards et al., 1988), as well as genes involved in many rare syndromes were cloned using this method. A notable example of the success of positional cloning was the identification of the LPS receptor (Beutler and Poltorak, 2000a), responsible for distinguishing self from non- self in innate immune responses against bacterial pathogens (Beutler, 2003; Beutler and Poltorak, 2000b). A phenotype of résistance to the léthal effect of LPS was mapped in a mice strain by linkage analysis, which corresponded to a 30 Mb interval of DNA comprising hundreds of genes. While several attempts to find the LPS receptor were made using conventional biochemical methods and expression cDNA cloning, positional cloning provided the final solution and allowed the localization the Lps locus to a much smaller 2.6-Mb critical région (Poltorak et al., 1998), where the

LPS receptor Tlr4 was finally located.

Positional cloning continues to be the method of choice for identifying novel genes within anonymous genomes. The first invertebrate allorecognition genes that hâve been characterized to date, FuHc and /ester in the tunicate Botryllus schlosseri (De Tomaso et al., 1998; De Tomaso and Weissman, 2003a; De Tomaso and


Weissman, 2003b; De Tomaso et al., 2005; Nyholm et al., 2006), and alr2(Nicotra et al., 2009) in Hydractinia symbiolongicarpus,were identified by BSA and positional cloning. The prior success of the positional cloning strategy in identifying the alr2 gene paved the road for the isolation of the alrl locus. The positional cloning of alrl could be undertaken promptly as ail the key éléments required were availabié. Recall from Chapter I that Hydractinialines segregating for the allorecognition phenotypes and BSA-derived AFLP markers were generated and used to map the allorecognition complex (ARC). The alrl locus was shown to be located between the markers AFLP

194 and 18 (Figure 1.7). Large-insert libraries for both segregating lines were built (Nicotra et al., 2009) and crosses were made to saturate the ARC with recombination events (Powell et al., 2007). The availability of a large number of recombinants over the ARC facilitâtes the positional cloning of the alrlgene as it allows targeting the interval of interest with high précision. This chapter describes the chromosome walk and the fme-scale mapping of genetic markers leading to the isolation of a 300.8 kb genomic région containing the allorecognition déterminant alrl.


2. Material and Methods

a. Hydractinia culture and genetic nomenclature

Hydractinia symbiolongicarpus colonies generated in the previously described crosses (see Chapter I) were propagated asexually on microscope glass slides and were cleaned weekly under a dissecting microscope. The slides were maintained in the laboratory in 38-liters aquaria at a salinity of 29 parts per thousand (artificial sea water: Reef Crystals) and at a température of 18-20°C. One-fifth of the water was changed daily and the colonies were fed three times a week with 3 to 4 day-old Artemianauplii. The inbred and recombinant animais were explanted onto several glass slides for DNA extraction as needed.

The development of the inbred and congeneic lines used in this study, as well as the subséquent mapping of the allorecognition chromosomal interval, bave been described previously (Mokady and Buss, 1996; Cadavid et al., 2004; Powell et al., 2007). The two haplotypes segregating within these lines were designated/and r. The non-recombinant haplotypes are referred to as ARC// ARC-r/r (homozygotes for the/and rhaplotype respectively), or AKC-f/r(hétérozygotes). The nomenclature for the recombinant haplotypes incorporated information from the two linked loci

controlling allorecognition, alrl and alr2. An individual with one recombinant

haplotype over a non-recombinant/haplotype is written as AKC-rf/ff or ASS2-fr/ff, an abbreviated nomenclature for alrl-r alr2-f/alrl-f alr2-for alrl-f alr2-r/alrl-f alr2-f.


b. BAC library construction and screening

A BAC library (Amplicon Express, Pullman,WA) from the AKC-f/f colony 833-8, an 8'*’ génération inbred female colony (Mokady and Buss, 1996), was

available from a previous study (Nicotra, 2007). The library contained 36,864 clones with an average insert size of 126 kb. The library coverage was estimated to be 6X based on a Hydractinia haploid genome size of 750 Mb (Nicotra et al., 2009).

A PCR/colony-hybridization strategy was used to streamline the BAC library screening. The library was arrayed into ninety-six 384-well plates. A DNA pool was generated by standard alkaline lysis from each plate, each pool representing 384 colonies (Osoegawa et al., 2001). Each DNA pool was arrayed into a single 96-well plate thus containing the DNA préparations of the whole BAC library. In parallel, each 384-well plate was printed individually onto a Hybond-N nylon filter (GE Healthcare) according to the manufacturer’s conditions. Screening of the BAC library followed a two-step process. A PCR on the 96-well pooled DNA préparations was first performed to identify the 384-well plate of interest, followed by a colony

hybridization of the corresponding filter to isolate the positive BAC clones. The same PCR fragments used to screen the BAC library were isolated from gDNA and used as probes in the colony hybridization. Probes were labeled using the random primer labeling technique (Feinberg and Vogelstein, 1983). Briefly, random pDN6 primer at a concentration of 66 mM was added to 50-100 ng of purified PCR product. The DNA was denatured for 5 minutes at 95°C, following which 30 ^iCi of radioactive isotope phosphorus-32 (dATP or dCTP a-^^P depending on the nucléotide content of the probe), lOX OLB Il-a or -c buffer (100 mM MgCb, 100 mM beta-


mercaptoethanol, 200 pM of each dNTP minus dATP or dCTP, 40% glycerol, IM Tris pH7.5 to adjust to desired volume), and 5 units of Klenow polymerase were added. The DNA was incubated at 37°C for 2 hours to allow for incorporation of the labeled nucléotides. lOX spin dialysis dye (1% SDS, 0.05M EDTA, bromophenol blue crystals) was added to the reaction before passing the probe through a Sepharose CL-6B column to eliminate non-labeled nucléotides. The filters were pre-hybridized for one hour in hybridization buffer (20X SCP, 25% N-lauryl-sarcosine), and

hybridized ovemight at 65°C in the same buffer to which the labeled probe and lOOpg/ml of salmon sperm DNA were added. The next moming, the filters were washed twice for 20 minutes in non-stringent conditions using a 2X blot solution (20X SCP, 10% SDS) and once for 20 minutes in more stringent conditions using the same blot solution at a concentration of 0.2X. Filters were exposed to storage

phosphor screens for 90 minutes and imaged on a Typhoon 9400 Imager (GE Healthcare).

A 13X coverage BAC library of the second haplotype segregating within our lines, ARC-r/r, was kindly provided by Dr. Luis Cadavid from the National

University of Colombia (Schwarz et al., 2007). The ARC-^library was used for the chromosome walk and the identification of the alrl interval, whereas the ARC-r/r library was only used to identify the ARC-r/r clones corresponding to the minimum ûr/r7-/interval. Primers used to design probes on the ARC-/contig were used to amplify ARC-r spécifie probes ffom ARC-r/r genomic DNA. High-density filters from the ARC-r/r library were used for the hybridization.


c. Chromosome walk

Two bidirectional chromosome walks were initiated from the closest AFLP markers on each side of a/r7(Powell et al., 2007), AFLP 194 (Poudyal et al., 2007) and AFLP 18 (Cadavid et al., 2004). A 412 bp sequence obtained from marker AFLP

194 was used as the initial hybridization probe in order to identify the first set of BAC clones in the 194 interval (using primers FWD 5’-


TATACTGACCGGGCGGTAGTTACAC-3’). In the 18 interval, a 289 bp sequence derived from marker AFLP 18 was used as the initial probe (using primers FWD 5’- AACAGACGAAATGGGAAATC -3’ and REV 5’-AGCATAAAATACATACCACC -3’).

The overlap and orientation of the BAC clones hybridizing to the AFLP 194 and 18 probes in the First walk was determined by restriction digest fingerprinting (Marra et al., 1997). DNA was obtained by miniprep from each newly isolated clone.

Briefly, BAC clones were grown ovemight in LB medium containing 12.5 mg/ml chloramphenicol and inserts were induced to high copy number using CopyControl solution (Epicenter) according to manufacturer’s conditions. DNA was then isolated from 1ml of cells by standard alkaline lysis (Osoegawa et al., 2001). DNA pellets were resuspended in 50 pl TE and 20 pl were digested ovemight with HindlII in a 50 pl total volume, in presence of lOX spermidine. A DNA marker (10 ng/pl) was prepared by mixing equal concentrations of commercially available Ikb DNA ladder (Invitrogen), X DNA eut with HindlII, and X DNA eut with HindlII and EcoRI. Both the marker and clone digests were combined with lOX loading dye (50% glycerol, 20 mM EDTA, 10 mM Tris pH 8.0, 0.2% bromophenol blue, 0.2% xylene cyanol) and


incubated at 65°C for 15 minutes. The digests, along with 60 ng of marker, were then electrophoresed on 1% agarose gels in IX TAE buffer for 11 hours at 87 volts with a constant cooling. Gels were stained with SYBR Green I Nucleic Acid stain

(Invitrogen), scanned on a Typhoon 9400 laser scanner (GE Healthcare), and analyzed using Image 3.9 (Sulston et al., 1989). The physical mapping software FingerPrintedContigs (Soderlund et al., 1997) was used to assemble the clones into contigs with a tolérance of 7 and a cutoff value of 10'°*. The newly added clones were confirmed to belong to the région of interest by détection of the probe by PCR.

The BAC clones identified as outermost on either side of the developing contigs were end-sequenced in order to design a probe for the next walk. Clones were grown as described above and the BAC DNA was isolated by midipreparation from

100 to 500 ml culture using the Plasmid Midi Kit (Qiagen) prior to sequencing. The sequencing was performed at W.M. Keck Biotechnology Center at Yale University using TaqFS dye terminator cycle sequencing on an ABI PRISM 3730 DNA sequencer. Chromatograms were analyzed using the LaserGene software package, version 6 & 7 (DNASTAR, Inc.). Primers were then designed to the BAC-end sequences and used in PCR reactions to amplify homozygous genomic DNA from KKC-f/f and -r/r. PCR products were gel purified using QIAquick spin columns (Qiagen) before being used as a hybridization probe.

d. Marker design in the alrl genomic région

BAC-end sequences were assessed for polymorphism in order to design haplotype-specific markers necessary to orient the walk towards alrl. Genotyping at


these markers was performed by PCR amplification of genomic DNA from the colonies of interest. A total of 50 ng of genomic DNA was amplified in 25 pl reactions using Q-Taq DNA polymerase (Qiagen) according to the manufacturer’s conditions. The PCR products were assessed on 0.8% agarose gel when a major size polymorphism was observed between the two haplotypes. If the polymorphism observed was based on SNPs or small indels, the development of cleavage

amplification polymorphism (CAPs) markers (Konieczny and Ausubel, 1993) was required. Restriction digests were performed according to the manufacturer’s

conditions in a total volume of 50 pl, using 10 pl of the PCR products and 2 to 5 units of the restriction enzyme chosen to differentially eut the ARC-^and ARC-r/r PCR products. The génotypes were then visualized on agarose gel by loading 10-20 pl of digest. The type and concentration of the agarose gel used depended on the size of the PCR/digest fragments. Regular 0.8 - 2% agarose was used unless the size différence was less than 100 bp, in which case 2.5 - 4% MetaPhor agarose was used. When polymorphism was found but did not correspond to any usable restriction site or size polymorphism, the PCR products were gel purified on QIAquick spin columns

(Qiagen) and sequenced to allow the détermination of the génotypes directly from the chromatograms. If sequences were found to be monomorphic between the ARC-/and ARC-r haplotypes, the end-sequences were extended using internai sequencing primers followed by a second round of primer design, amplification and génotype testing. Molecular markers were subjected to co-segregation analysis using PCR amplification of genomic DNA from a panel of known AKC-f/f (N=7), ARC-r/r (N=10), and ARC-/r (N=10) animais. This procedure also provided further assurance that the BAC clones on whieh the new probe and marker were designed segregated with the alrl région.


e. Analysis of recombinants in the alrl genomic région

Mapping recombination breakpoints in the alrl genomic région was crucial to orient the developing contig and focus on a unidirectional chromosomal walk, as well as to identify the proximal and distal limits of the alrl interval. Recall from Chapter I that the alrl locus was mapped between markers AFLP 194 and AFLP 18. A total of 64 recombinants were available in this région (Table 2.1), originating from five mapping populations: 431 (N=l) (Cadavid et al., 2004), BCl (N=14) (Cadavid et al., 2004), BK3 (N=10) (Cadavid et al., 2004), and AP 100 (N=39) (Powell et al., 2007).

Two recombinants were localized between marker AFLP 194 and alrl (referred later as proximal recombinants), and 62 between alrl and marker AFLP 18 (referred later as distal recombinants).

The génotypes of the recombinants were determined for each of the previously mentioned BAC-end markers in order to defme the directionality of the walk. DNA from the recombinant animais was extracted using a protocol adapted from Dellaporta et al. (1983). Briefly, a piece of tissue containing about 10 polyps was scraped off the

glass slide and pulverized with liquid nitrogen, extracted with 500 pl of urea

extraction buffer (Chen and Dellaporta, 1994), and incubated at 65°C for 10 minutes.

1.5M potassium acetate (250 ul) was added and the samples placed on ice for 15 minutes, extracted with an equal volume of phénolichloroform (1:1). DNA was precipitated from the aqueous phase with 0.7 volume of isopropanol, the pellet washed with 2 volumes of 80% éthanol, and DNA resuspended in 25 pl 10 mM Tris,

1 mM EDTA pH 8.0.


Table 2.1 Recombinants in the alrl genomic région Family Génotype Génotype a

Phenotype N Colony ID Reference at 194 at 18

f/f r/r

431 f/r r/r R F 1 2 Cadavid et al. 2004

AP 100 f/f f/r F TF 21 573,732,837, 997. 1378, 1467.1563, 1619, 1689, 2058,2123,2213,2237, 2384, 2401,2513,2558, 2844,3128,3135,3620

Powell et al. 2007

f/r f/f F TF 17 279, 501,622, 1369, 1107

1279, 1322, 1458, 1832, 2110,2120, 2600, 2637 2724, 3001,3079,3519

f/f f/r F F 1 1064

BCl f/r f/f F TF 8 1,76,81, 170, 189,


Cadavid et al. 2004

f/f f/r F TF 6 90, 166, 201

290, 293, 379

BK3 f/r f/f F TF 9 50, 121,323,356, 506,

728, 839, 848, BEN-34

Cadavid et al. 2004

f/f f/r F TF 1 109

“ Phenotypes observed in fusibility assays against a homozygous AKC-f/fand ARC-r/r.

R: rejection; F: fusion; TF: transitory fusion

Internai markers were developed in the alrl interval in order to define the minimal a/r7-containing interval by fmely mapping the detected recombination breakpoints. The sequence of the ARC-/haplotype was initially used as a reference for the marker design. Primers spécifie to the ARC-/haplotype were initially tested on ARC-r/r genomic DNA. If the région was conserved and the amplification successfül, the PCR Products were gel purified and sequenced for the r haplotype, allowing for the design of polymorphie markers. If the corresponding ARC-r/r sequence was available, the marker design was based on sequence homology and the régions


containing interesting SNPs or indels were targeted. Additionally, a molecular marker unlinked to the ARC région was developed as a control to insure that recombinant alleles resulted from crossing-over events between the parental chromosomes and not from a source of contamination. A 2165 bp fragment containing the last 165 bp of exon 1, the fiill intron 1 and the first 218 bp of exon 2 of the GAPDH gene was amplified (FOR: 5’-ACTGCTTTTGAGCATGAGGAACTTG-3’; REV: 5’- ACAATGCTCATGGTTTTTGGATCAT-3’) from the genomic DNA of each

recombinant using Phusion High Fidelity DNA Polymerase (Finnzymes) according to manufacturer’s conditions. The PCR product was gel purified with QIAquick spin columns (Qiagen), and cloned into a pCR®-Blunt II-TOPO® vector (Zéro Blunt TOPO PCR cloning kit from Invitrogen). The ligation product was transformed into One Shot® TOP 10 chemically competent E. coli cells and the cells were grown ovemight at 37°C. Eight clones were checked for the presence of the insert by PCR with vector primers and the PCR products were purified using a PEG DNA

purification protocol before being sequenced. An equal volume of cold PEG (20%

polyethylene glycol, 2.5M NaCl) was added to the PCR products, which were then incubated at 37°C for 15’. The samples were centrifuged at 3000 rpm for 30’ and the pellets were washed in cold 70% éthanol before being resuspended in 25 |,i,l of lOmM Tris-Cl pH 8.5. The purified products were sequenced with the forward primer used in the PCR and the sequences were compared to the ARC-f/f and -r/r alleles.

f. Contig sequencing and analysis

The sequencing of the minimum tiling path of the alrl région cloned from the from the ARC-y^library (12 BAC clones), of the minimal alrl interval obtained from


the ARC-r/r library (3 BAC clones), and of the minimum tiling path of the walk achieved in the 18 interval (6 BAC clones), was performed through the Community Sequencing Program at the US Department of Energy’s Joint Genome Institute.

The sequences of the KRC-f/f and -r/r air J intervals were masked for repeats using RepeatMasker (http://repeatmasker.org/cgi-in/RepeatProteinMaskRequest) and Censor (Kohany et al., 2006). Similarity-based methods (BLAST suite, Altschul et al., 1997) ffom NCBI, PRSS/PRFX from the FASTA programs suite (Pearson, 1998;

Pearson and Lipman, 1988; Pearson et al., 1997) were used for the comparison of the ARC-/and -r haplotypes to design optimal polymorphie markers. VISTA Genome Browser VGB2 (Schwartz et al., 2000; Brudno et al., 2003a; Brudno et al., 2003b;

Frazer et al., 2004) and PipMaker (Schwartz et al., 2000) were used to investigate structural variation within and between the two haplotypes.



Sujets connexes :