Biochimica et Biophysica Acta 1793 (2009) 921–923

Contents lists available at ScienceDirect

Biochimica et Biophysica Acta

j ourna l homepage: www.e lsev ie r.com/ locate /bbamcr

Minireview

Getting a grip on Thy-1 signaling

Thomas H. Barker a,1, James S. Hagood b,⁎a Biomedical Engineering, GA Tech/Emory, 313 Ferst Drive, Ste 3111 Atlanta, GA 30332-0535, USAb University of Alabama-Birmingham, 1670 University Blvd., 648A VH, Birmingham, AL 35294-0019, USA

⁎ Corresponding author. Tel.: +1 205 9755224; fax: +E-mail addresses: thomas.barker@bme.gatech.edu (T

(J.S. Hagood).1 Tel.: +1 404 385 5039.

0167-4889/$ – see front matter © 2008 Elsevier B.V. Adoi:10.1016/j.bbamcr.2008.10.004

a b s t r a c t

a r t i c l e i n f o

Article history:

A recent study by Hermosil Received 19 March 2008Received in revised form 7 October 2008Accepted 8 October 2008Available online 25 October 2008

Keywords:Thy-1IntegrinNeuronAstrocyteContext-dependent signaling

la et al. [T. Hermosilla, D. Munoz, R. Herrera-Molina, A. Valdivia, N. Munoz, S.U.Nham, P. Schneider, K. Burridge, A.F. Quest, L. Leyton, Direct Thy-1/alphaVbeta3 integrin interaction mediatesneuron to astrocyte communication, Biochim Biophys Acta 1783 (2008) 1111–1120] demonstrates that Thy-1on neurons binds to αvβ3 integrin on astrocytes via a conserved RLD motif, triggering the formation of focaladhesions and stress fibers via tyrosine phosphorylation and RhoA activation. This study adds to growingevidence regarding the signaling mechanisms and biological roles of Thy-1, an important regulator ofcontext-dependent signaling. As knowledge of Thy-1 approaches its fiftieth year, it is critical to begin tosynthesize insights from different fields to determine how this heretofore enigmatic molecule modulates cellbehavior in both cis and trans.

© 2008 Elsevier B.V. All rights reserved.

Thy-1 (CD90) is a developmentally regulated, evolutionarily

conserved cell surface glycoprotein, the biological role of whichremains somewhat enigmatic despite hundreds of intriguingpublications over the past forty-five years. Part of the reason thatThy-1 is still a bit mysterious is that, though it is by no meansubiquitous, it is expressed in a heterogeneous range of cell types,including thymocytes, lymphocytes, fibroblasts, neuronal cells,hematopoietic and mesenchymal stem cells, ovarian follicular cells,and some cancer cells. It therefore has implications in a number offields, including neurobiology, immunology, oncology, stem cellbiology and wound healing. In each of these areas, investigatorshave tended to home in on one aspect of the function of Thy-1, in aparticular cell type. Thus the field suffers a bit from the well-known“blind men and the elephant” metaphor, in which no one has thebig picture. Furthermore, the Thy-1 null mouse is viable and hassubtle and diverse phenotypic features. Its characterization has beensomewhat piecemeal, in keeping with the above observations. In theminds of many, Thy-1 has been relegated to the role of a meremarker; useful for identifying some fibroblasts, stem cells andneurons (in a colorful, even whimsical way in the case of the Thy-1-Brainbow transgenic mouse [2]), but failing to enter the pantheon of“vitally important” molecules.

Part of Thy-1's scientific facelessness results from its lack of adefined ligand; its failure to associate with molecules of interest. Thearticle by Hermosilla et al [1] addresses this deficiency in part byclearly demonstrating, using a systematic, combinatorial approach,

1 205 9962333..H. Barker), jhagood@uab.edu

ll rights reserved.

heterotypic binding of neuronal Thy-1 to αvβ3 integrin on astrocytes,inducing focal adhesion formation and cytoskeletal alterations viaRhoA GTPase activation. The findings have implications for neurode-velopment and neural repair, and begin to paint a clearer picture ofThy-1's role with regard to signaling. The immunologic and non-immunologic functions of Thy-1 andwhat is known about its signalingmechanisms have been reviewed previously [3–5]; this mini-reviewwill focus on recent findings in two broad paradigms in Thy-1signaling, attempt to put these findings in a broader biological context,and suggest areas for future exploration.

Thy-1 can affect signaling heterotypically (in trans) by “effectorcell” Thy-1 binding to a Thy-1 ligand on a “target cell” (Fig. 1A). This isthe paradigm illustrated in the Hermosilla paper. Reported ligands forThy-1 trans- signaling are β2 and β3 integrins. Specifically, humandermal microvascular endothelial cell (HDMEC) Thy-1 binds to αXβ2

(p 150,95, CD11c/CD18) orαMβ2 (Mac-1, CD11b/CD18) on leukocytes,promoting their adhesion and transendothelial migration [6–8].Melanoma cells appear to bind instead via αvβ3 integrin to Thy-1 onactivated endothelium [9,10], which may facilitate melanoma metas-tasis. Thy-1 has been known to inhibit neurite outgrowth onastrocytes for many years [11]. Leyton et al initially described howthis interaction produces focal adhesion changes in astrocytes as well,and defined a role for β3 integrin in this heterotypic signaling [12].This group has continued to dissect this interaction, and the currentstudy demonstrates definitively the role of the αvβ3 heterodimer, andmore clearly defines the signaling in astrocytes downstream of αvβ3

[1]. A number of novel findings are reported. The α chain of the αvβ3

heterodimer identified by immunoprecipitation from DI TNC1 ratastrocytes was of smaller-than-expected molecular size, suggestingalternative splicing or post-translational modification. However, “full-length” αvβ3 binds to Thy-1 in vitro by surface plasmon resonance.

Fig. 1. Thy-1 (CD90) Signaling. (A) Trans Signaling: Simplified scheme of Thy-1 heterotypic signaling. Colored text indicates effector and responder cells for the downstream effectsnoted. (B) Cis Signaling: Colored dots in the light blue box correspond to colored text for the cell types in which the indicated responses have been described. For further detailsregarding the signaling pathways involved, the reader is referred to the text and to several relevant reviews [3–5].

922 T.H. Barker, J.S. Hagood / Biochimica et Biophysica Acta 1793 (2009) 921–923

The αvβ3 heterodimer is not normally expressed on adult astrocytes,but may be upregulated in response to injury. Hermosilla et al alsodemonstrate a requirement for the RLD tripeptide in Thy-1 in bindingto αvβ3 integrin. This sequence is found within a highly conservedregion (70% human–rat–mouse homology in the 10 amino acidsflanking RLD in both directions) of Thy-1, and is similar to the αMβ2-binding region in fibrinogen [13]. It is suggested that other portions ofthe Thy-1 molecule, such as the GPI anchor, are not required for trans-signaling through αvβ3, as Thy-1-Fc fusion protein on microspheresappears equally capable of mediating focal adhesion formation onastrocytes. This finding also would seem to negate the need for Thy-1in “effector” cells (e.g., neurons) to be present within lipid rafts. Thesignaling pathway stimulated by Thy-1 via αvβ3 involves focaladhesion kinase (FAK) and RhoA GTPase. Interestingly, these arepathways which we have observed in association with homotypic (incis) Thy-1 signaling in pulmonary fibroblasts [14–16], as discussedfurther below. The neuron-astrocyte interaction also involves Thy-1cis signaling in neurons (neurite outgrowth inhibition); however,the Thy-1-interacting molecules within neurons, and the downstreamsignaling pathways activated, have yet to be identified. Others haveshown that neurite outgrowth inhibition requires lipid raft integrity,as well as the Thy-1-specific GPI anchor [11].

Thy-1 also signals homotypically (in cis), through interactionwith TCR, via Thy-1 crosslinking, or through unknown mechanisms(Fig. 1B). The latter two must by necessity involve interaction withother molecules, as Thy-1 lacks a transmembrane or cytoplasmicportion. We and others have previously established Thy-1'sinteraction with members of the Src family kinases (SFKs; e.g.,via association of the GPI anchor with palmitoylated residues onSrc family members such as fyn and lyn [17]). Immunoprecipitationof Thy-1 from cell lysates co-precipitates SFKs [18]. Evidence for cissignaling of Thy-1 can be derived from pulmonary fibroblastsubpopulations sorted based on their surface expression of Thy-1.Thy-1-expressing fibroblasts display altered phenotypic responsesto growth factors, cytokines, and adhesion proteins. These

responses have been partially explained by regulation of down-stream signaling proteins such as phosphoinositide-3 kinase (PI3K)and SFKs. In response to the matricellular protein thrombospondin,Thy-1-expressing fibroblasts will disassemble focal adhesions in aPI3K- and SFK-mediated manner [15]. Lack of cell surface Thy-1abrogates this effect, which can be rescued by exogenous Thy-1expression following transfection. Furthermore, the baseline activityof the SFK-p190 Rho GTPase-activating protein (GAP) pathway isdependent on Thy-1 expression. Exogenous expression of Thy-1 onThy-1-nonexpressing fibroblasts results in inhibition of SFK activityand subsequent activation of RhoA, leading to focal adhesionassembly and stress fiber formation [14]. Equally important to Thy-1 cis signaling is the integrity of lipid raft micro- and nanodomains.Through its GPI linkage, Thy-1 has been shown to have equalmobility in cholesterol-rich lipid rafts as in the bulk plasmamembrane, which facilitates its trafficking into and out of lipid raftdomains, whereas transmembrane proteins and other moieties aremore spatially restricted [19]. This inherent mobility may play arole in Thy-1's capacity to facilitate cis signaling as demonstratedby the fact that Thy-1 localization to lipid rafts and lipid raftintegrity appear critical to Thy-1's effect on signaling. Eitherreplacement of the GPI linkage with a membrane-spanning domainor disruption of lipid rafts with methyl-beta-cyclodextrin (MβCD)disrupts Thy-1-mediated signaling [11,16]. Recently, strong evidencehas been presented that implies a role for Thy-1 in the associationof lipid rafts with the actin cortex immediately adjacent to theplasma membrane. Clustering of Thy-1 with antibody-coated goldnanoparticles followed by particle tracking demonstrates that Thy-1sorts into, resides within, and migrates out of distinct lipidnanodomains. The residence time within the nanodomains appearscontrolled by associations with the cell cytoskeleton [20]. Takentogether, these data are suggestive of a role of Thy-1 in thetrafficking and partitioning of signaling molecules into and out ofspecialized lipid rafts for higher order signaling networks asso-ciated with the cytoskeleton. In this way Thy-1 may act alternately

923T.H. Barker, J.S. Hagood / Biochimica et Biophysica Acta 1793 (2009) 921–923

as an inhibitor, by sequestering, or facilitator, by trafficking, ofsignaling molecules such as SFKs.

In order to truly understand the effects of Thy-1 on signalingnetworks in both cis and trans, significant further studies are required.As yet, no comprehensive search for cis binding partners has beenreported. While clear demonstration of the trans interaction withαvβ3 via RLD is a significant breakthrough, it is clear that Thy-1interacts with other cell surface molecules in a planar fashion. DoesThy-1 also associate with integrin in cis? The role of Thy-1 inregulating the Rho GTPases, focal adhesion assembly/disassembly,and stress fiber formation are suggestive that cis interactions withintegrin occur, but studies are required to demonstrate this, as well asto systematically determine other Thy-1 binding partners. Further-more, the mechanisms regulating Thy-1 localization and trafficking tospecialized and distinct lipid nanodomains present a significantchallenge that is likely to provide further insights into Thy-1's rolein coordinating signaling networks. Substitution of peptide sequencesthat direct GPI linkages in the endoplasmic reticulumwould allow theinterrogation of the effects of Thy-1 localization on signaling networksand downstream cell biology. The recent publications describing Thy-1's relative mobility and residence time in specific plasma membranedomains and the unexpected ability of Thy-1 to connect lipid rafts tounderlying cytoskeletal networks uncovers a completely unexploredtopic with direct relevance to Thy-1's documented activity inregulating cell cytoskeletal organization and changes.

The evidence for Thy-1 binding to integrin and regulation ofcytoskeletal signaling and organization make it ideally located toaffect critical cellular events, such as cell motility, mitotic lift-off,and potentially epithelial to mesenchymal transition. The fact thatthe molecule is shed from the surface [21,22] and could thusdirectly interact as a soluble factor, acting potentially as aninhibitor of cell adhesion, further complicates our understandingof the role of this protein in normal physiology and pathologicalprogression. What remains clear is that the expression of Thy-1 hassignificant consequences for the phenotype of many cells. Despiteour lack of understanding of the nuances of Thy-1 mechanisms, itis clear we can no longer be naïve in the assumption that Thy-1 issimply a marker. It appears instead to be an integral protein indefining cell behavior during normal homeostasis as well as duringtissue repair.

Other important aspects of Thy-1 biology not reviewed hereinclude evolution of Thy-1, regulation of Thy-1 expression, post-translational modification of Thy-1, including glycosylation andshedding, and potential interaction with lectins. Even by focusingmerely on Thy-1 signaling in cis and trans, however, it is clear thatThy-1 exerts significant effects on signaling modulating cell–celland cell–matrix interaction, via interesting and possibly uniquemechanisms involving changes in both target and effector cells.Thy-1 thus emerges as an important modulator of context-dependent signaling, which alters interaction of cells with eachother and with their immediate environment. Some of the mostchallenging problems in biomedicine today are disorders ofneurodevelopment and neural repair after injury, regulation ofstem cell pluripotency and differentiation, cancer metastasis, andprogressive fibrotic diseases. Given the fact that Thy-1 has a role inall of these processes, the era of seeing Thy-1 as a marker, or ofcompartmentalized, nearsighted analysis of its function is over. AsThy-1 enters its second half-century, it is time to start looking atthe big picture.

References

[1] T. Hermosilla, D. Munoz, R. Herrera-Molina, A. Valdivia, N. Munoz, S.U. Nham, P.Schneider, K. Burridge, A.F. Quest, L. Leyton, Direct Thy-1/alphaVbeta3 integrininteraction mediates neuron to astrocyte communication, Biochim. Biophys. Acta1783 (2008) 1111–1120.

[2] J. Livet, T.A. Weissman, H. Kang, R.W. Draft, J. Lu, R.A. Bennis, J.R. Sanes, J.W.Lichtman, Transgenic strategies for combinatorial expression of fluorescentproteins in the nervous system, Nature 450 (2007) 56–62.

[3] S.M. Haeryfar, D.W. Hoskin, Thy-1: more than a mouse pan-T cell marker, J.Immunol. 173 (2004) 3581–3588.

[4] T.A. Rege, J.S. Hagood, Thy-1, a versatile modulator of signaling affecting cellularadhesion, proliferation, survival, and cytokine/growth factor responses, Biochim.Biophys. Acta 1763 (2006) 991–999.

[5] T.A. Rege, J.S. Hagood, Thy-1 as a regulator of cell–cell and cell–matrix interactionsin axon regeneration, apoptosis, adhesion, migration, cancer, and fibrosis, FASEB J.20 (2006) 1045–1054.

[6] J. Choi, L. Leyton, S.U. Nham, Characterization of alphaX I-domain binding toThy-1,Biochem. Biophys. Res. Commun. 331 (2005) 557–561.

[7] A. Wetzel, T. Chavakis, K.T. Preissner, M. Sticherling, U.F. Haustein, U. Anderegg, A.Saalbach, Human Thy-1 (CD90) on activated endothelial cells is a counterreceptorfor the leukocyte integrin Mac-1 (CD11b/CD18), J. Immunol. 172 (2004)3850–3859.

[8] A. Saalbach, U.F. Haustein, U. Anderegg, A ligand of human Thy-1 is localized onpolymorphonuclear leukocytes and monocytes and mediates the binding toactivated Thy-1-positive microvascular endothelial cells and fibroblasts, J. Invest.Dermatol. 115 (2000) 882–888.

[9] A. Saalbach, G. Hildebrandt, U.F. Haustein, U. Anderegg, The Thy-1/Thy-1 ligandinteraction is involved in binding of melanoma cells to activated Thy-1-positivemicrovascular endothelial cells, Microvasc. Res. 64 (2002) 86–93.

[10] A. Saalbach, A. Wetzel, U.F. Haustein, M. Sticherling, J.C. Simon, U. Anderegg,Interaction of human Thy-1 (CD90) with the integrin alphavbeta3 (CD51/CD61):an important mechanism mediating melanoma cell adhesion to activatedendothelium, Oncogene 24 (2005) 4710–4720.

[11] M.-C. Tiveron, E. Barboni, F.B.P. Rivero, A.M. Gormley, P.J. Seeley, F. Grosveld, R.Morris, Selective inhibition of neurite outgrowth on mature astrocytes by Thy-1glycoprotein, Nature 355 (1992) 745–748.

[12] L. Leyton, P. Schneider, C.V. Labra, C. Ruegg, C.A. Hetz, A.F. Quest, C. Bron, Thy-1binds to integrin beta(3) on astrocytes and triggers formation of focal contactsites, Curr. Biol. 11 (2001) 1028–1038.

[13] D.C. Altieri, J. Plescia, E.F. Plow, The structural motif glycine 190-valine 202 of thefibrinogen g chain interacts with CD11b/CD18 integrin (aMb2, Mac-1) andpromotes leukocyte adhesion, J. Biol. Chem. 268 (1993) 1847–1853.

[14] T.H. Barker, H.E. Grenett, M.W. MacEwen, S.G. Tilden, G.M. Fuller, J. Settleman, A.Woods, J. Murphy-Ullrich, J.S. Hagood, Thy-1 regulates fibroblast focal adhesions,cytoskeletal organization and migration throughmodulation of p190 RhoGAP andRho GTPase activity, Exp. Cell Res. 295 (2004) 488–496.

[15] T.H. Barker, M.A. Pallero, M.W. MacEwen, S.G. Tilden, A. Woods, J.E. Murphy-Ullrich, J.S. Hagood, Thrombospondin-1-induced focal adhesion disassembly infibroblasts requires Thy-1 surface expression, lipid raft integrity, and Srcactivation, J. Biol. Chem. 279 (2004) 23510–23516.

[16] T.A. Rege, M.A. Pallero, C. Gomez, H.E. Grenett, J.E. Murphy-Ullrich, J.S. Hagood,Thy-1, via its GPI anchor, modulates Src family kinase and focal adhesion kinasephosphorylation and subcellular localization, and fibroblast migration, inresponse to thrombospondin-1/hep I, Exp. Cell Res. 312 (2006) 3752–3767.

[17] A.M. Shenoy-Scaria, J. Kwong, T. Fujita, M.W. Olszowy, A.S. Shaw, D.M. Lublin,Signal transduction through decay-accelerating factor. Interaction of glycosyl-phosphatidylinositol anchor and protein tyrosine kinases p56lck and p59fyn 1, J.Immunol. 149 (1992) 3535–3541.

[18] R.C. Henke, G.S. Seeto, P.L. Jeffrey, Thy-1 and AvGp50 signal transduction complexin the avian nervous system: c-Fyn and G alpha i protein association andactivation of signalling pathways, J. Neurosci. Res. 49 (1997) 655–670.

[19] F. Zhang, B. Crise, B. Su, Y. Hou, J.K. Rose, A. Bothwell, K. Jacobson, Lateral diffusionof membrane-spanning and glycosylphosphatidylinositol-linked proteins: towardestablishing rules governing the lateral mobility of membrane proteins, J. Cell Biol.115 (1991) 75–84.

[20] Y. Chen, W.R. Thelin, B. Yang, S.L. Milgram, K. Jacobson, Transient anchorage ofcross-linked glycosyl-phosphatidylinositol-anchored proteins depends on cho-lesterol, Src family kinases, caveolin, and phosphoinositides, J. Cell Biol. 175(2006) 169–178.

[21] J.S. Hagood, P. Prabhakaran, P. Kumbla, L. Salazar, M.W. MacEwen, T.H. Barker, L.A.Ortiz, T. Schoeb, G.P. Siegal, C.B. Alexander, A. Pardo, M. Selman, Loss of fibroblastThy-1 expression correlates with lung fibrogenesis, Am. J. Pathol. 167 (2005)365–379.

[22] A. Saalbach, T. Wetzig, U.F. Haustein, U. Anderegg, Detection of human solubleThy-1 in serum by ELISA. Fibroblasts and activated endothelial cells are a possiblesource of soluble Thy-1 in serum, Cell Tissue Res. 298 (1999) 307–315.