- Email: [email protected]
Why Is There so Much CD45 on T Cells?
Immunity
Previews viruses. The abundant secretion of type I interferon might activate NOX2 or other components of the nicotinamide adenine dinucleotide phosphate (NADPH)-oxidase system, which produces reactive oxygen intermediates (ROIs). ROIs reduce endosomal pH, which in turn prevents complete degradation of the antigen and thereby enhances antigen crosspresentation (Savina et al., 2006). Although human pDCs do not express mannose receptor, a C type lectin that favors crosspresentation by enabling antigen uptake into specialized endosomes (Burgdorf et al., 2007), they might employ other mechanisms of receptormediated endocytosis to crosspresent antigens. This phenomenon could also explain the discrepancy between the crosspresentation of HIV peptide-lipopeptide conjugates and HIV-infected apoptotic cells observed in the present study, and the lack of crosspresentation of tumor antigens or tumor antigens delivered as immune complexes (Figure 1). It is likely that, depending on the receptor selectively engaged, the antigen reaches different compart-
ments and undergoes complete degradation and exclusive loading onto MHC class II or partial degradation and further processing into the cytosol for loading into MHC class I molecules. Accordingly, myeloid DCs are more efficient at crosspresenting HIV-gag protein when it is targeted to humanDEC205 receptor than when it is targeted to a closely related receptor like DC-SIGN (Bozzacco et al., 2007). Hoeffel et al. have returned pDCs to the center stage of antigen presentation. Clearly, it is essential to determine whether crosspresentation by human pDCs is a common event, especially during HIV infection, and whether such presentation leads to T cell stimulation or T cell anergy in vivo. Because HIV does not productively infect pDCs, but activates them through TLR7, crosspresentation by pDCs might be a very effective way to elicit anti-HIV T cell responses. REFERENCES Ackerman, A.L., Giodini, A., and Cresswell, P. (2006). Immunity 25, 607–617.
Bevan, M.J. (2006). Nat. Immunol. 7, 363–365. Bozzacco, L., Trumpfheller, C., Siegal, F.P., Mehandru, S., Markowitz, M., Carrington, M., Nussenzweig, M.C., Piperno, A.G., and Steinman, R.M. (2007). Proc. Natl. Acad. Sci. USA 104, 1289–1294. Burgdorf, S., Kautz, A., Bohnert, V., Knolle, P.A., and Kurts, C. (2007). Science 316, 612– 616. Hoeffel, G., Ripoche, A.-C., Matheoud, D., Nascimbeni, M., Escriou, N., Lebon, P., Heshmati, F., Guillet, J.-G., Gannage´, M., CaillatZucman, S., et al. (2007). Immunity 27, this issue, 481–492. Sapoznikov, A., Fischer, J.A., Zaft, T., Krauthgamer, R., Dzionek, A., and Jung, S. (2007). J. Exp. Med. 204, 1923–1933. Savina, A., Jancic, C., Hugues, S., Guermonprez, P., Vargas, P., Moura, I.C., LennonDumenil, A.M., Seabra, M.C., Raposo, G., and Amigorena, S. (2006). Cell 126, 205–218. Schnurr, M., Chen, Q., Shin, A., Chen, W., Toy, T., Jenderek, C., Green, S., Miloradovic, L., Drane, D., Davis, I.D., et al. (2005). Blood 105, 2465–2472. Shinohara, M.L., Lu, L., Bu, J., Werneck, M.B., Kobayashi, K.S., Glimcher, L.H., and Cantor, H. (2006). Nat. Immunol. 7, 498–506. Zhang, J., Raper, A., Sugita, N., Hingorani, R., Salio, M., Palmowski, M.J., Cerundolo, V., and Crocker, P.R. (2006). Blood 107, 3600–3608.
Why Is There so Much CD45 on T Cells? Rose Zamoyska1,* 1Medical Research Council National Institute for Medical Research, The Ridgeway, Mill Hill, London NW7 1AA UK *Correspondence: [email protected] DOI 10.1016/j.immuni.2007.08.009
The balance between kinases and phosphatases is crucial for regulating lymphocyte signaling. In this issue, McNeill et al. (2007) show that the transmembrane phosphatase CD45 has a role as both positive and negative regulator of T cell signaling. The presence of CD45 molecules on hematopoietic cells has long been an enigma: Why so much, why so big, why so variable, why can we not find a ligand? One of the most abundant molecules on the lymphocyte surface, CD45 was identified a number of years ago as a transmembrane phosphatase (Hermiston et al., 2003). The large extracellular domain of CD45 is notable for its highly glycosylated and sialy-
lated state, which varies depending on the inclusion or exclusion of alternatively spliced exons 4, 5, and 6. The resulting isoforms are specific not only to hematopoietic cell type, but also to the stage of differentiation and activation of the cell. Not surprisingly, therefore, it was suggested that these alternative isoforms might interact with unique ligands; however, convincing identification of ligands spe-
cific for any of the isoforms has so far defied a seemingly endless supply of research dollars and several lifespans of graduate students. The major intracellular targets of CD45 phosphatase activity are the Src-family kinases (SFK), which in T cells are predominantly the family members p56Lck (Lck) and p59Fyn (Fyn). Lck, in particular, is a primary initiator of signal transduction upon
Immunity 27, September 2007 ª2007 Elsevier Inc. 421
Immunity
Previews
Figure 1. Regulation of Lck Activity by the Opposing Actions of Kinases and Phosphases The regulatory Tyr (Y505) of Lck is phosphorylated by the kinase Csk, which maintains Lck in an inactive conformation. The dephosphorylation of Y505 by CD45 in resting T cells sustains a pool of basally active Lck. Upon T cell stimulation, this pool of Lck auto- or transphosphorylates the Tyr residue in the kinase domain (Y394), which in turn enhances Lck kinase activity. CD45 can also dephosphorylate Y394, returning Lck to the basally active state.
T cell receptor engagement, and the absence of Lck leads to a severe block in T cell differentiation and profound impairment of activation in mature T cells (Zamoyska et al., 2003). Conversely, the overexpression of Lck or mutations that disregulate Lck function can result in uncontrolled cellular activation and leukemogenesis. SFK activity is regulated, in large part, by the phosphorylation status of two key tyrosine (Tyr) residues, one in the kinase domain and one at the C terminus. Phosphorylation of the kinase domain Tyr (residue 394 in Lck) enhances kinase activity, whereas phosphorylation of the C-terminal Tyr (residue 505 in Lck) has an inhibitory effect (Palacios and Weiss, 2004). CD45 was shown to dephosphorylate the inhibitory Tyr of Lck, and CD45deficient mice have highly phosphorylated Lck Tyr-505 in the thymus and a block in differentiation. The expression of an Lck transgene containing Phe rather than Tyr at residue 505 overcame the CD45-deficient differentiation block, confirming that Lck Tyr-505 is a physiologically relevant target of CD45 activity in vivo and that dephosphorylation of this residue by CD45 is required to enable Lck activity [(Hermiston et al., 2003) and references therein] (Figure 1). In addition, it has been shown that CD45 can dephosphorylate Tyr-394 in the active site of Lck, suggesting that CD45 might also play a role in down-
regulating the activity of Lck (Ashwell and D’Oro, 1999). This dual activity of CD45, both enabling and downregulating Lck activity, is elegantly demonstrated in vivo in the paper by McNeill et al. in this issue of Immunity (McNeill et al., 2007), with some interesting consequences for thymocyte differentiation and peripheral T cell activation. The authors expressed transgenes encoding CD45RO isoforms (normally expressed by double positive [DP] thymocytes and activated T cells), which had either wild-type or disabled phosphatase activity, in mice lacking endogenous CD45. The expression of the transgenes to varying amounts provided a titration of CD45 phosphatase activity that correlated directly with cell surface expression and ranged from approximately 1% to 60% of wild-type values. Curiously, lower phosphatase activity than that obtained from normal CD45 expression was found to enhance differentiation of mature thymocytes, with maximal CD4SP and CD8SP differentiation occurring with constructs expressed at about 30% of wild-type amounts. Remarkably, as little as 3% of normal CD45 activity was sufficient to overcome the differentiation block of the CD45-deficient background and restore single positive (SP) differentiation to approximately wild-type amounts. It had been noted before that mice heterozygous for CD45 (expressing
422 Immunity 27, September 2007 ª2007 Elsevier Inc.
50% of wild-type amounts) show enhanced positive selection of T cell receptor (TCR) transgenic cells (Wallace et al., 1997). One explanation proposed for this observation was that reducing CD45 expression might benefit selection by reducing steric hindrance. CD45 has high amounts of glycosylation and sialylation and, therefore, could contribute considerable negative charge to the thymocyte surface which might potentially inhibit interaction between the TCR and selecting ligands. McNeill et al. indicate that this explanation is unlikely because the expression of a kinase inactive CD45 transgene at similar amounts neither restored nor worsened positive selection beyond that seen in the CD45deficient mice, indicating that the external domain alone contributes little to CD45 function in the thymus. In contrast, a second explanation, that CD45 can act both positively and negatively to regulate signaling, was supported by their observations that the amounts of Lck phospho-pTyr-505 as well as Lck phospho-Tyr-394 were altered by changing the expression of CD45. The authors showed there were comparable increases in the amounts of phosphorylation of both these residues at intermediate CD45 expression, suggesting that the negative influence of increased pTyr-505 might be counterbalanced by an enhanced positive contribution from the phosphorylation of pTyr-394. At very low CD45 expression, the phosphorylation of pTyr-505 became disproportionally higher than the phosphorylation of pTyr-394, resulting in decreased Lck activity. A consequence of intermediate CD45 expression was that an increased percentage of DP thymocytes had detectable phospho-Erk and phospho-PKB upon stimulation with anti-CD3, indicating they were more sensitive to TCR engagement and therefore more likely to undergo positive selection. Similar increases in sensitivity to TCR stimulation were found in peripheral T cells, which also displayed enhanced Erk phosphorylation and proliferation, from the mice expressing intermediate amounts of CD45. An interesting aside is that the efficiency of CD4SP cell differentiation did not mirror exactly the efficiency of
Immunity
Previews CD8SP differentiation, with the latter requiring higher amounts of CD45 (and, by inference, reduced Lck activity) to see equivalent enhancement of selection. These results concur with published data that showed higher Lck activity favored CD4 differentiation over CD8 differentiation [(Zamoyska et al., 2003) and references therein]. These data indicate that in T cells, sensitivity to stimulation exhibits a Gaussian distribution with respect to CD45 phosphatase activity, with peak responsiveness set just below normal CD45 expression. Why then is the expression of CD45 so high? It has been suggested that CD45 activity is regulated by dimerization because the crystal structure of the membrane-proximal regions of a related phosphatase, RPTPa, have been shown to form a dimer in which the catalytic sites of one molecule are occluded by binding a structural wedge contributed by the second partner. Similar residues are conserved in CD45, and the artificial dimerization of CD45 was shown to reduce its activity. Moreover, a knockin mouse in which a key residue required for dimerization was mutated developed a lymphoproliferative syndrome and autoimmune nephritis, indicating that dimerization is important for the normal control of peripheral T cell activation (Majeti et al., 2000). McNeill et al. argue that the dimerization model is unlikely to account for the decreased phosphatase activity they observe with increasing expression of CD45, above 40% of that expressed by wildtype cells, for two reasons. First, when
they measured phosphatase activity of isolated cell membranes, they found a linear increase in activity corresponding to CD45 expression, whereas the dimerization model predicts that phosphatase activity should plateau at a certain point, beyond which dimerization and inhibition of CD45 activity would be more likely to occur. Second, the inhibition of CD45 activity by dimerization at high concentrations should increase T cell signaling rather than decrease signaling, as was observed. Where does this leave us in terms of the outstanding questions about CD45? With respect to CD45’s abundance, the data from McNeill et al. supports the suggestion that CD45 behaves as a rheostat, regulating the threshold of activation in T cells. However, many issues still remain. These transgenic mice expressed the smallest CD45RO isoforms; would a similar titration of responsiveness be observed if the larger isoforms were expressed? Although studies of transgenic mice expressing other single isoforms at high amount showed only minor differences between them (Kozieradzki et al., 1997; Ogilvy et al., 2003), perhaps this kind of titration of expression is required to reveal unique functions. Furthermore, the analysis of whether the variable amounts of CD45 influence the distribution of molecules in the immunological synapse could provide clues as to how CD45 interacts with its target proteins. Finally, what about other hematopoeitic lineages? The knockin mutation influencing dimerization was found to cause B cell hyperresponsiveness (Hermiston
et al., 2005); how does the titration of CD45 influence B cell development and responses? Perhaps for these cells as well, less CD45 is more in terms of maximizing responsiveness; however, more CD45 might be optimal for setting the physiological threshold of activation and maintaining tolerance.
REFERENCES Ashwell, J.D., and D’Oro, U. (1999). Immunol. Today 20, 412–416. Hermiston, M.L., Tan, A.L., Gupta, V.A., Majeti, R., and Weiss, A. (2005). Immunity 23, 635– 647. Hermiston, M.L., Xu, Z., and Weiss, A. (2003). Annu. Rev. Immunol. 21, 107–137. Kozieradzki, I., Kundig, T., Kishihara, K., Ong, C.J., Chiu, D., Wallace, V.A., Kawai, K., Timms, E., Ionescu, J., Ohashi, P., et al. (1997). J. Immunol. 158, 3130–3139. Majeti, R., Xu, Z., Parslow, T.G., Olson, J.L., Daikh, D.I., Killeen, N., and Weiss, A. (2000). Cell 103, 1059–1070. McNeill, L., Salmond, R.J., Cooper, J.C., Carret, C.K., Cassady-Cain, R.L., Roche-Molina, M., Tandon, P., Holmes, N., and Alexander, D.R. (2007). Immunity 27, this issue, 425– 437. Ogilvy, S., Louis-Dit-Sully, C., Cooper, J., Cassady, R.L., Alexander, D.R., and Holmes, N. (2003). J. Immunol. 171, 1792– 1800. Palacios, E.H., and Weiss, A. (2004). Oncogene 23, 7990–8000. Wallace, V.A., Penninger, J.M., Kishihara, K., Timms, E., Shahinian, A., Pircher, H., Kundig, T.M., Ohashi, P.S., and Mak, T.W. (1997). J. Immunol. 158, 3205–3214. Zamoyska, R., Basson, A., Filby, A., Legname, G., Lovatt, M., and Seddon, B. (2003). Immunol. Rev. 191, 107–118.
Immunity 27, September 2007 ª2007 Elsevier Inc. 423
Previews viruses. The abundant secretion of type I interferon might activate NOX2 or other components of the nicotinamide adenine dinucleotide phosphate (NADPH)-oxidase system, which produces reactive oxygen intermediates (ROIs). ROIs reduce endosomal pH, which in turn prevents complete degradation of the antigen and thereby enhances antigen crosspresentation (Savina et al., 2006). Although human pDCs do not express mannose receptor, a C type lectin that favors crosspresentation by enabling antigen uptake into specialized endosomes (Burgdorf et al., 2007), they might employ other mechanisms of receptormediated endocytosis to crosspresent antigens. This phenomenon could also explain the discrepancy between the crosspresentation of HIV peptide-lipopeptide conjugates and HIV-infected apoptotic cells observed in the present study, and the lack of crosspresentation of tumor antigens or tumor antigens delivered as immune complexes (Figure 1). It is likely that, depending on the receptor selectively engaged, the antigen reaches different compart-
ments and undergoes complete degradation and exclusive loading onto MHC class II or partial degradation and further processing into the cytosol for loading into MHC class I molecules. Accordingly, myeloid DCs are more efficient at crosspresenting HIV-gag protein when it is targeted to humanDEC205 receptor than when it is targeted to a closely related receptor like DC-SIGN (Bozzacco et al., 2007). Hoeffel et al. have returned pDCs to the center stage of antigen presentation. Clearly, it is essential to determine whether crosspresentation by human pDCs is a common event, especially during HIV infection, and whether such presentation leads to T cell stimulation or T cell anergy in vivo. Because HIV does not productively infect pDCs, but activates them through TLR7, crosspresentation by pDCs might be a very effective way to elicit anti-HIV T cell responses. REFERENCES Ackerman, A.L., Giodini, A., and Cresswell, P. (2006). Immunity 25, 607–617.
Bevan, M.J. (2006). Nat. Immunol. 7, 363–365. Bozzacco, L., Trumpfheller, C., Siegal, F.P., Mehandru, S., Markowitz, M., Carrington, M., Nussenzweig, M.C., Piperno, A.G., and Steinman, R.M. (2007). Proc. Natl. Acad. Sci. USA 104, 1289–1294. Burgdorf, S., Kautz, A., Bohnert, V., Knolle, P.A., and Kurts, C. (2007). Science 316, 612– 616. Hoeffel, G., Ripoche, A.-C., Matheoud, D., Nascimbeni, M., Escriou, N., Lebon, P., Heshmati, F., Guillet, J.-G., Gannage´, M., CaillatZucman, S., et al. (2007). Immunity 27, this issue, 481–492. Sapoznikov, A., Fischer, J.A., Zaft, T., Krauthgamer, R., Dzionek, A., and Jung, S. (2007). J. Exp. Med. 204, 1923–1933. Savina, A., Jancic, C., Hugues, S., Guermonprez, P., Vargas, P., Moura, I.C., LennonDumenil, A.M., Seabra, M.C., Raposo, G., and Amigorena, S. (2006). Cell 126, 205–218. Schnurr, M., Chen, Q., Shin, A., Chen, W., Toy, T., Jenderek, C., Green, S., Miloradovic, L., Drane, D., Davis, I.D., et al. (2005). Blood 105, 2465–2472. Shinohara, M.L., Lu, L., Bu, J., Werneck, M.B., Kobayashi, K.S., Glimcher, L.H., and Cantor, H. (2006). Nat. Immunol. 7, 498–506. Zhang, J., Raper, A., Sugita, N., Hingorani, R., Salio, M., Palmowski, M.J., Cerundolo, V., and Crocker, P.R. (2006). Blood 107, 3600–3608.
Why Is There so Much CD45 on T Cells? Rose Zamoyska1,* 1Medical Research Council National Institute for Medical Research, The Ridgeway, Mill Hill, London NW7 1AA UK *Correspondence: [email protected] DOI 10.1016/j.immuni.2007.08.009
The balance between kinases and phosphatases is crucial for regulating lymphocyte signaling. In this issue, McNeill et al. (2007) show that the transmembrane phosphatase CD45 has a role as both positive and negative regulator of T cell signaling. The presence of CD45 molecules on hematopoietic cells has long been an enigma: Why so much, why so big, why so variable, why can we not find a ligand? One of the most abundant molecules on the lymphocyte surface, CD45 was identified a number of years ago as a transmembrane phosphatase (Hermiston et al., 2003). The large extracellular domain of CD45 is notable for its highly glycosylated and sialy-
lated state, which varies depending on the inclusion or exclusion of alternatively spliced exons 4, 5, and 6. The resulting isoforms are specific not only to hematopoietic cell type, but also to the stage of differentiation and activation of the cell. Not surprisingly, therefore, it was suggested that these alternative isoforms might interact with unique ligands; however, convincing identification of ligands spe-
cific for any of the isoforms has so far defied a seemingly endless supply of research dollars and several lifespans of graduate students. The major intracellular targets of CD45 phosphatase activity are the Src-family kinases (SFK), which in T cells are predominantly the family members p56Lck (Lck) and p59Fyn (Fyn). Lck, in particular, is a primary initiator of signal transduction upon
Immunity 27, September 2007 ª2007 Elsevier Inc. 421
Immunity
Previews
Figure 1. Regulation of Lck Activity by the Opposing Actions of Kinases and Phosphases The regulatory Tyr (Y505) of Lck is phosphorylated by the kinase Csk, which maintains Lck in an inactive conformation. The dephosphorylation of Y505 by CD45 in resting T cells sustains a pool of basally active Lck. Upon T cell stimulation, this pool of Lck auto- or transphosphorylates the Tyr residue in the kinase domain (Y394), which in turn enhances Lck kinase activity. CD45 can also dephosphorylate Y394, returning Lck to the basally active state.
T cell receptor engagement, and the absence of Lck leads to a severe block in T cell differentiation and profound impairment of activation in mature T cells (Zamoyska et al., 2003). Conversely, the overexpression of Lck or mutations that disregulate Lck function can result in uncontrolled cellular activation and leukemogenesis. SFK activity is regulated, in large part, by the phosphorylation status of two key tyrosine (Tyr) residues, one in the kinase domain and one at the C terminus. Phosphorylation of the kinase domain Tyr (residue 394 in Lck) enhances kinase activity, whereas phosphorylation of the C-terminal Tyr (residue 505 in Lck) has an inhibitory effect (Palacios and Weiss, 2004). CD45 was shown to dephosphorylate the inhibitory Tyr of Lck, and CD45deficient mice have highly phosphorylated Lck Tyr-505 in the thymus and a block in differentiation. The expression of an Lck transgene containing Phe rather than Tyr at residue 505 overcame the CD45-deficient differentiation block, confirming that Lck Tyr-505 is a physiologically relevant target of CD45 activity in vivo and that dephosphorylation of this residue by CD45 is required to enable Lck activity [(Hermiston et al., 2003) and references therein] (Figure 1). In addition, it has been shown that CD45 can dephosphorylate Tyr-394 in the active site of Lck, suggesting that CD45 might also play a role in down-
regulating the activity of Lck (Ashwell and D’Oro, 1999). This dual activity of CD45, both enabling and downregulating Lck activity, is elegantly demonstrated in vivo in the paper by McNeill et al. in this issue of Immunity (McNeill et al., 2007), with some interesting consequences for thymocyte differentiation and peripheral T cell activation. The authors expressed transgenes encoding CD45RO isoforms (normally expressed by double positive [DP] thymocytes and activated T cells), which had either wild-type or disabled phosphatase activity, in mice lacking endogenous CD45. The expression of the transgenes to varying amounts provided a titration of CD45 phosphatase activity that correlated directly with cell surface expression and ranged from approximately 1% to 60% of wild-type values. Curiously, lower phosphatase activity than that obtained from normal CD45 expression was found to enhance differentiation of mature thymocytes, with maximal CD4SP and CD8SP differentiation occurring with constructs expressed at about 30% of wild-type amounts. Remarkably, as little as 3% of normal CD45 activity was sufficient to overcome the differentiation block of the CD45-deficient background and restore single positive (SP) differentiation to approximately wild-type amounts. It had been noted before that mice heterozygous for CD45 (expressing
422 Immunity 27, September 2007 ª2007 Elsevier Inc.
50% of wild-type amounts) show enhanced positive selection of T cell receptor (TCR) transgenic cells (Wallace et al., 1997). One explanation proposed for this observation was that reducing CD45 expression might benefit selection by reducing steric hindrance. CD45 has high amounts of glycosylation and sialylation and, therefore, could contribute considerable negative charge to the thymocyte surface which might potentially inhibit interaction between the TCR and selecting ligands. McNeill et al. indicate that this explanation is unlikely because the expression of a kinase inactive CD45 transgene at similar amounts neither restored nor worsened positive selection beyond that seen in the CD45deficient mice, indicating that the external domain alone contributes little to CD45 function in the thymus. In contrast, a second explanation, that CD45 can act both positively and negatively to regulate signaling, was supported by their observations that the amounts of Lck phospho-pTyr-505 as well as Lck phospho-Tyr-394 were altered by changing the expression of CD45. The authors showed there were comparable increases in the amounts of phosphorylation of both these residues at intermediate CD45 expression, suggesting that the negative influence of increased pTyr-505 might be counterbalanced by an enhanced positive contribution from the phosphorylation of pTyr-394. At very low CD45 expression, the phosphorylation of pTyr-505 became disproportionally higher than the phosphorylation of pTyr-394, resulting in decreased Lck activity. A consequence of intermediate CD45 expression was that an increased percentage of DP thymocytes had detectable phospho-Erk and phospho-PKB upon stimulation with anti-CD3, indicating they were more sensitive to TCR engagement and therefore more likely to undergo positive selection. Similar increases in sensitivity to TCR stimulation were found in peripheral T cells, which also displayed enhanced Erk phosphorylation and proliferation, from the mice expressing intermediate amounts of CD45. An interesting aside is that the efficiency of CD4SP cell differentiation did not mirror exactly the efficiency of
Immunity
Previews CD8SP differentiation, with the latter requiring higher amounts of CD45 (and, by inference, reduced Lck activity) to see equivalent enhancement of selection. These results concur with published data that showed higher Lck activity favored CD4 differentiation over CD8 differentiation [(Zamoyska et al., 2003) and references therein]. These data indicate that in T cells, sensitivity to stimulation exhibits a Gaussian distribution with respect to CD45 phosphatase activity, with peak responsiveness set just below normal CD45 expression. Why then is the expression of CD45 so high? It has been suggested that CD45 activity is regulated by dimerization because the crystal structure of the membrane-proximal regions of a related phosphatase, RPTPa, have been shown to form a dimer in which the catalytic sites of one molecule are occluded by binding a structural wedge contributed by the second partner. Similar residues are conserved in CD45, and the artificial dimerization of CD45 was shown to reduce its activity. Moreover, a knockin mouse in which a key residue required for dimerization was mutated developed a lymphoproliferative syndrome and autoimmune nephritis, indicating that dimerization is important for the normal control of peripheral T cell activation (Majeti et al., 2000). McNeill et al. argue that the dimerization model is unlikely to account for the decreased phosphatase activity they observe with increasing expression of CD45, above 40% of that expressed by wildtype cells, for two reasons. First, when
they measured phosphatase activity of isolated cell membranes, they found a linear increase in activity corresponding to CD45 expression, whereas the dimerization model predicts that phosphatase activity should plateau at a certain point, beyond which dimerization and inhibition of CD45 activity would be more likely to occur. Second, the inhibition of CD45 activity by dimerization at high concentrations should increase T cell signaling rather than decrease signaling, as was observed. Where does this leave us in terms of the outstanding questions about CD45? With respect to CD45’s abundance, the data from McNeill et al. supports the suggestion that CD45 behaves as a rheostat, regulating the threshold of activation in T cells. However, many issues still remain. These transgenic mice expressed the smallest CD45RO isoforms; would a similar titration of responsiveness be observed if the larger isoforms were expressed? Although studies of transgenic mice expressing other single isoforms at high amount showed only minor differences between them (Kozieradzki et al., 1997; Ogilvy et al., 2003), perhaps this kind of titration of expression is required to reveal unique functions. Furthermore, the analysis of whether the variable amounts of CD45 influence the distribution of molecules in the immunological synapse could provide clues as to how CD45 interacts with its target proteins. Finally, what about other hematopoeitic lineages? The knockin mutation influencing dimerization was found to cause B cell hyperresponsiveness (Hermiston
et al., 2005); how does the titration of CD45 influence B cell development and responses? Perhaps for these cells as well, less CD45 is more in terms of maximizing responsiveness; however, more CD45 might be optimal for setting the physiological threshold of activation and maintaining tolerance.
REFERENCES Ashwell, J.D., and D’Oro, U. (1999). Immunol. Today 20, 412–416. Hermiston, M.L., Tan, A.L., Gupta, V.A., Majeti, R., and Weiss, A. (2005). Immunity 23, 635– 647. Hermiston, M.L., Xu, Z., and Weiss, A. (2003). Annu. Rev. Immunol. 21, 107–137. Kozieradzki, I., Kundig, T., Kishihara, K., Ong, C.J., Chiu, D., Wallace, V.A., Kawai, K., Timms, E., Ionescu, J., Ohashi, P., et al. (1997). J. Immunol. 158, 3130–3139. Majeti, R., Xu, Z., Parslow, T.G., Olson, J.L., Daikh, D.I., Killeen, N., and Weiss, A. (2000). Cell 103, 1059–1070. McNeill, L., Salmond, R.J., Cooper, J.C., Carret, C.K., Cassady-Cain, R.L., Roche-Molina, M., Tandon, P., Holmes, N., and Alexander, D.R. (2007). Immunity 27, this issue, 425– 437. Ogilvy, S., Louis-Dit-Sully, C., Cooper, J., Cassady, R.L., Alexander, D.R., and Holmes, N. (2003). J. Immunol. 171, 1792– 1800. Palacios, E.H., and Weiss, A. (2004). Oncogene 23, 7990–8000. Wallace, V.A., Penninger, J.M., Kishihara, K., Timms, E., Shahinian, A., Pircher, H., Kundig, T.M., Ohashi, P.S., and Mak, T.W. (1997). J. Immunol. 158, 3205–3214. Zamoyska, R., Basson, A., Filby, A., Legname, G., Lovatt, M., and Seddon, B. (2003). Immunol. Rev. 191, 107–118.
Immunity 27, September 2007 ª2007 Elsevier Inc. 423