Current Biology Vol 23 No 20R934

Histocompatibility: Clarifying FusionConfusion

In the colonial tunicate Botryllus schlosseri, a co-dominant trait determines thecapacity of adjacent colonies to fuse or reject. An innovative RNA sequencingapproach has now identified the gene that predicts the outcomes of thisnaturally occurring allograft.

Gary W. Litman* and Larry J. Dishaw

Transplantation immunology hasits foundations in over a half centuryof intensive laboratory investigationsthat initially were based on bovinedizygotic twins [1] and later inbredand congenic strains of mice.Transplantation of human organs,bone marrow and specific cellpopulations [2] as wellas pregnancy (the most commonhistocompatibility–incompatibilityscenario) [3,4] have furtherexpanded our understanding of thegenetic basis for allorecognition,defined as the ability of an organismto distinguish its own tissues fromthose of another of the samespecies. Whether or not a graft isaccepted or rejected depends injawed vertebrates on highlypolymorphic and broadly expressedgenes encoded within the majorhistocompatibility system, theprimary function of which is torespond to viral and bacterialantigens [5]; the outcomes oftransplantation are also dependenton minor histocompatibility antigens.Allograft responses are not unique tovertebrates and have beencharacterized in several othersystems, including the sessiletunicate Botryllus schlosseri, adistant relative of vertebrates. A recentstudy by Voskoboynik et al. [6]identifies the gene governingallorecognition in Botryllus,thereby clarifying much about themolecular basis of allorecognition inthis system.

The evolutionary origins ofhistocompatibility have longinterested immunologists [7], but liketracking the origins ofimmunoglobulin-mediated antibodyresponse of the adaptive immunesystem [8], the trail runs cold at the timeof the divergence of the jawedvertebrates. However, ahistocompatibility reaction that atface value bears a striking

resemblance to graft acceptance andrejection phenomena characteristicof tissue transplantation inmammals has been described inbasal chordates [9,10].Histocompatibility in colonial tunicatesis a naturally occurring phenomenonthat involves fusion or rejection ofadjacent animals (Figure 1). Thisprocess protects against stem cellparasitism (stem cells of onegenotypic background overtaking thecolonial tissue of the adjacent animal),which could severely compromisepopulation diversity and thereby theorganism’s capacity for adaptation orsurvival [11,12]. From a geneticperspective, histocompatibility in thesecolonial organisms behaves as aco-dominant autosomal trait. If twodistinct colonies share one or both‘histocompatibility’ alleles, thecolonies fuse leading to the sharingof their blood supply; if colonies donot share alleles, they reject fusionthrough the formation of inflammatorylesions.

Over three decades, the seniorauthor, Irv Weissman, his studentsand other collaborators havedeveloped the colonial tunicateBotryllus schlosseri into awell-recognized model of geneticallypredictable histocompatibility [13–16].In previous studies, it was shownthat the genetic segregation of ahighly polymorphic candidatefusibility histocompatibility gene(cFuHC) correlates strongly but notperfectly with transplantationoutcomes [14]. cFuHC encodes bothmembrane-bound and secreted forms,which recently were shown to beencoded by independent gene lociseparated by w250 base pairs [17].In a system, where direct approachesto ascertain gene function are notreadily applicable, the lingeringquestion has been whether or notcFuHC encodes a true determinant ofallorejection or whether the actualdeterminant is a closely linked locus.In response to a recent report that cast

doubt on the role of cFuHC inallorecognition [18], definitivestudies have been carried out and itnow seems that the latter possibility isthe case.The elegance of the new work

reported by Voskoboynik et al. [6] liesin the use of high-throughput, deepcoverage RNA sequencing toapproach the identity of thehistocompatibility determinant froma functional perspective, which isneither based on nor biased by earlierwork [6]. Phased, paired-end RNAsequencing analysis of bothlaboratory-reared and wild-typecolonies of Botryllus was used todefine haplotypes in such amanner whereby computationalapproaches could be used todetermine whether a particulartranscript stratifies a known fusionoutcome. In other words,extraordinarily large numbers ofexpressed sequences were queriedto determine whether or notobserved sequence polymorphismscorrelated with transplantationoutcomes. In every instance ofcolony fusion, no sequencedifferences could be identified [6].Candidate sequences also wereentirely reliable in predictingoutcomes of wild-type pairingswith laboratory strains. Theinterpretation of the earlier results [14]was compromised by anincomplete sequence across thehistocompatibility determinantregion. Using the recentlycompleted sequence of theBotryllus genome [19], theauthors show that the cFuHC locus islocated w62,000 base pairs fromthe new gene reported here. Finally,it seems that the molecular basisof urochordate histocompatibilityis at hand.Over a half century following the

discovery that colonialhistocompatibility in Botryllus isdetermined genetically, we nowknow that it is mediated byBotryllus histocompatibility factor(BHF), a three exon protein,encoding 252 residues, thatexhibits significant relatedness onlyto genes found in other tunicates [6].The outcomes of transplantationcan be predicted from sequencesin the amino-terminal portion ofBHF. In addition to the absolutegenetic predictability of BHF as the

Figure 1. Colonial fusion in Botryllus schlosseri.

The fusion or rejection of adjacent colonies of Botryllus schlosseri represents a naturallyoccurring allograft recognition reaction. The fusion reaction shown here is between geneticallyidentical colonies and includes the formation of a common tunic. All colonies sharing oneco-dominant histocompatibility allele undergo fusion. Those pairs lacking an allele in commonexhibit rejection. The work described here identifies highly polymorphic BHF (Botryllushistocompatiblity factor) as the gene governing this physiologically significant reaction.(Image: Michelle M. Osovitz.)

DispatchR935

histocompatibility determinant, itspatterns of expression duringexperimental transplantationgenerally are consistent with thesites of colony fusion or rejection.The rejection reaction can beinhibited with translation-blockingmorpholinos based on BHF. Giventhe relationship between thepolymorphic MHC and functions ofthe jawed vertebrate adaptiveimmune system, the lackof relatedness of BHF to the primarydeterminants of allorecognition injawed vertebrates comes as nosurprise. Interestingly, BHF alsobears no relatedness to thedeterminants of histocompatibility inHydractinia [20], a cnidarian,suggesting that this formof self–nonself recognition has multipleindependent origins.

Although the lack of relatednessof BHF to previously describedtransplantation determinants maydisappoint those focusing ontransplantation in humans, whatwould the findings of a molecularhomolog, for example an MHC-likemolecule, really tell us? It is far more

significant that the naturallyoccurring transplantation settingthat confronts Botryllus sp.is mediated by a unique class ofmolecule, one that lacks the structuralfeatures of a cell surface-recognitionprotein. Future investigations of BHFfunction have the potential to affordus far more important informationabout the underlying commonalityof integrated networks and signalingprocesses involved in self–nonselfrecognition as well as what mostcertainly represents an importantmechanism for preserving populationdiversity.

References1. Owen, R.D. (1945). Immunogenetic

consequences of vascular anastomosesbetween bovine twins. Science 102,400–401.

2. Brent, L. (2003). The 50th anniversary of thediscovery of immunologic tolerance. N. Engl. J.Med. 349, 1381–1383.

3. Clark, D.A., Chaouat, G., Wong, K.,Gorczynski, R.M., and Kinsky, R. (2010).Tolerance mechanisms in pregnancy: areappraisal of the role of class I paternalMHC antigens. Am. J. Reprod. Immunol. 63,93–103.

4. Guleria, I., and Sayegh, M.H. (2007). Maternalacceptance of the fetus: true human tolerance.J. Immunol. 178, 3345–3351.

5. Blander, J.M. (2008). Phagocytosis and antigenpresentation: a partnership initiated by Toll-likereceptors. Ann. Rheum. Dis. 67(Suppl 3 ),iii44–49.

6. Voskoboynik, A., Newman, A.M., Corey, D.M.,Sahoo, D., Pushkarev, D., Neff, N.F.,Passarelli, B., Koh, W., Ishizuka, K.J.,Palmeri, K.J., et al. (2013). Identification of acolonial chordate histocompatibility gene.Science 341, 384–387.

7. Burnet, F.M. (1971). ‘‘Self-recognition’’ incolonial marine forms and flowering plants inrelation to the evolution of immunity. Nature232, 230–235.

8. Litman, G.W., Cannon, J.P., and Dishaw, L.J.(2005). Reconstructing immune phylogeny:new perspectives. Nat. Rev. Immunol. 5,866–879.

9. Taneda, Y., and Watanabe, H. (1982). Studieson colony specificity in the compound ascidian,Botryllus primigenus Oka. I. Initiation of‘‘nonfusion’’ reaction with special reference toblood cells infiltration. Dev. Comp. Immunol. 6,43–52.

10. Tanaka, K. (1975). Allogeneic distinction inBotryllus primigenus and in other colonialascidians. Adv. Exp. Med. Biol. 64,115–124.

11. De Tomaso, A.W. (2006). Allorecognitionpolymorphism versus parasitic stem cells.Trends Genet. 22, 485–490.

12. Rinkevich, B. (2002). The colonial urochordateBotryllus schlosseri: from stem cellsand natural tissue transplantation to issuesin evolutionary ecology. Bioessays 24,730–740.

13. Laird, D.J., De Tomaso, A.W., andWeissman, I.L. (2005). Stem cells are units ofnatural selection in a colonial ascidian. Cell 123,1351–1360.

14. De Tomaso, A.W., Nyholm, S.V., Palmeri, K.J.,Ishizuka, K.J., Ludington, W.B., Mitchel, K., andWeissman, I.L. (2005). Isolation andcharacterization of a protochordatehistocompatibility locus. Nature 438,454–459.

15. Rinkevich, B., and Weissman, I.L. (1992).Allogeneic resorption in colonialprotochordates: consequences of nonselfrecognition. Dev. Comp. Immunol. 16,275–286.

16. Harp, J.A., Tsuchida, C.B., Weissman, I.L., andScofield, V.L. (1988). Autoreactive blood cellsand programmed cell death in growth anddevelopment of protochordates. J. Exp. Zool.247, 257–262.

17. Nydam, M.L., Netuschil, N., Sanders, E.,Langenbacher, A., Lewis, D.D., Taketa, D.A.,Marimuthu, A., Gracey, A.Y., and DeTomaso, A.W. (2013). The candidatehistocompatibility locus of a Basal chordateencodes two highly polymorphic proteins.PLoS One 8, e65980.

18. Rinkevich, B., Douek, J., Rabinowitz, C., andPaz, G. (2012). The candidate Fu/HC gene inBotryllusschlosseri (Urochordata) andascidians’ historecognition–an oxymoron? Dev.Comp. Immunol. 36, 718–727.

19. Voskoboynik, A., Neff, N.F., Sahoo, D.,Newman, A.M., Pushkarev, D., Koh, W.,Passarelli, B., Fan, H.C., Mantalas, G.L.,Palmeri, K.J., et al. (2013). The genomesequence of the colonial chordate, Botryllusschlosseri. eLife 2, e00569.

20. Cadavid, L.F., Powell, A.E., Nicotra, M.L.,Moreno, M., and Buss, L.W. (2004). Aninvertebrate histocompatibility complex.Genetics 167, 357–365.

Morsani College of Medicine, Department ofPediatrics, University of South Florida,St. Petersburg, FL 33701, USA.*E-mail: Gary.Litman@allkids.org

http://dx.doi.org/10.1016/j.cub.2013.08.048