- Email: [email protected]
From the Loading Dock to the Boardroom: A New Job for Akt Kinase
Immunity
Previews Given the pro-B cell-specific and Pax5 dependency of the PAIR sequence, as measured by transcriptional activity and pro-B cell preferential binding of CTCF and cohesin to these same sequences, coupled with known functional requirement of Pax5 in distal VH recombination, Ebert et al. (2011) propose that PAIRs may represent the long sought cis-regulatory elements critical for Igh looping to ensure utilization of distal VHs in VHDJH rearrangements. In the context of such a model, Pax5 might activate PAIRs, perhaps via antisense transcription, allowing PAIRs to form the base of rosette loops that bring distal VHs into V(D)J recombination complex with the DJH, iEm, and potentially other elements via CBEs in the VH to DH intergenic region and/or in the region downstream of the CHs. Testing of such models will require deleting these proposed regulatory elements and mutating factor binding sites within them to test effects on V(D)J recombination. As noted by Ebert et al. (2011), this will be challenging given the repetitive nature of the VH locus as well as potential redundancy. It may also be further complicated if there
are many different VH loops formed across the Igh locus, with only one or a few in a given progenitor B cell. Such experimental difficulties might require, if possible, truncating the VH portion of Igh. Another very important experiment will be to mutate the more unique potential looping elements (e.g., CBEs) in the downstream portions of Igh to test effects on distal VH gene utilization. Finally, given that the large human IGH must also have mechanisms to ensure utilization of distal versus proximal VHs, it will be important to determine whether PAIR-related sequence exist in the human. Overall, the finding of the PAIR sequences in the upstream portion of the Igh locus provides an important focus for future studies aimed at solving the long-standing mystery of how the Igh locus organizes itself during V(D)J recombination to ensure broad VH utilization and a diverse antibody repertoire.
Degner-Leisso, S.C., and Feeney, A.J. (2010). Semin. Immunol. 22, 346–352. Ebert, A., McManus, S., Tagoh, H., Medvedovic, J., Salvagiotto, G., Novatchkova, M., Tamir, I., Sommer, A., Jaritz, M., and Busslinger, M. (2011). Immunity 34, this issue, 175–187. Fuxa, M., Skok, J., Souabni, A., Salvagiotto, G., Roldan, E., and Busslinger, M. (2004). Genes Dev. 18, 411–422. Jhunjhunwala, S., van Zelm, M.C., Peak, M.M., Cutchin, S., Riblet, R., van Dongen, J.J., Grosveld, F.G., Knoch, T.A., and Murre, C. (2008). Cell 133, 265–279. Jhunjhunwala, S., van Zelm, M.C., Peak, M.M., and Murre, C. (2009). Cell 138, 435–448. Malynn, B.A., Yancopoulos, G.D., Barth, J.E., Bona, C.A., and Alt, F.W. (1990). J. Exp. Med. 171, 843–859. Subrahmanyam, R., and Sen, R. (2010). Semin. Immunol. 22, 337–345.
REFERENCES
Yancopoulos, G.D., Desiderio, S.V., Paskind, M., Kearney, J.F., Baltimore, D., and Alt, F.W. (1984). Nature 311, 727–733.
Corcoran, A.E. (2010). Semin. Immunol. 22, 353–361.
Yancopoulos, G.D., Blackwell, T.K., Suh, H., Hood, L., and Alt, F.W. (1986). Cell 44, 251–259.
From the Loading Dock to the Boardroom: A New Job for Akt Kinase Sonia Feau1 and Stephen P. Schoenberger1,* 1Laboratory of Cellular Immunology, La Jolla Institute for Allergy and Immunology, 9420 Athena Circle, La Jolla, CA 92037, USA *Correspondence: [email protected] DOI 10.1016/j.immuni.2011.02.013
A new report in Immunity shows that, rather than driving the metabolic changes required for proliferation, Akt controls the gene expression programs that determine whether activated CD8+ T cells differentiate into memory or effector cells. The multifunctional cytokine interleukin-2 (IL-2) can drive metabolic changes needed to insure that activated CD8+ T cells can rapidly expand in response to infection as well as develop the effector functions needed to eradicate the threat posed by infected host cells. Additionally, the capacity to produce IL-2 is a central feature of the long-lived memory population that persist following contraction of
the effectors. A key goal for immunologists has therefore been to identify the pathways and players through which signals emanating from a single cytokine receptor can lead to the downstream gene expression events required to coordinate the metabolic and differentiation needs of an expanding clone. An important early ‘‘branch point’’ for this division of labor is thought to occur with the IL-2-
mediated generation of the lipid second messenger phosphatidylinositol 3,4,5-trisphosphate (PI(3,4,5)P3), the product of phosphatidylinositol 3-kinases (PI3K). This phosporylated lipid recruits the specific signaling intermediates via specialized lipid-binding pleckstrin homology (PH) domains and itself becomes activated to initiate further signaling cascades. Two key players in this process
Immunity 34, February 25, 2011 ª2011 Elsevier Inc. 141
Immunity
Previews are Akt (protein kinase B, PKB) and PDK1 (3-phosphoinositide-dependent protein kinase 1), both which can be recruited to the plasma membrane by PI(3,4,5)P3) and activate downstream kinases. In the case of PDK1, Akt itself is a substrate for PDK1-dependent activation via phosphorylation at its T308 site. Akt has traditionally been associated with metabolic pathways in T cells, based mostly on studies in which it was targeted to the membrane by myristoylation and manifested a corresponding effect on growth and survival of proliferating thymocytes (Juntilla and Koretzky, 2008). A similar role in mature T cells was suggested by studies using constitutively active Akt that could induce growth and survival of CD4+, but remarkably not CD8+, T cells, raising the possibility that these subsets are wired differently with respect to the downstream consequences of activating the PI3K-Akt pathway (Finlay and Cantrell, 2010). In this issue of Immunity, Macintyre et al. (2011) reveal that, in contrast to controlling metabolism in CD8+ T cells, Akt has a direct role in regulating whether their differentiation leads them to either the memory or effector fate. To investigate the role of Akt in CD8+ T cells, Macintyre et al. used an in vitro model in which T cell receptor (TCR)-stimulated cells are cultured in IL-2 in order to sustain the signals that lead effector cell generation (Manjunath et al., 2001). Using either pharmacologic or genetic inhibition of the p110d subunit of PI3K that activates Akt or an allosteric inhibitor of all 3 Akt isoforms, Macintyre et al. found that, in the absence of Akt function, TCR-stimulated CD8+ T cells could both proliferate and survive in response to IL-2. Subsequent studies using a genetically targeted model showed that IL-2-driven proliferation is actually mediated by PDK1, not Akt, as deletion of PDK1 in CD8+ T cells rendered them unable to proliferate or sustain glucose uptake in response to IL-2. Remarkably, although the PDK1-deficient CD8+ T cells failed to divide in response to IL-2, they did not die, suggesting that survival is mediated by yet another distinct pathway. These findings reveal that Akt controls neither IL-2-mediated proliferation nor glucose uptake in this system and raised questions as to its actual function in enabling the CD8+ T cell response. To further
explore this, Macintyre et al. performed microarray analysis to profile geneexpression patterns from Akt-inhibited and PDK1 null CD8+ T cells. In support of their biological data, they found that genes encoding proteins involved in glucose uptake and lipid metabolism were overrepresented among genes regulated by PDK1 independent of Akt. Surprisingly, the Akt-inhibited CD8+ T cells induced genes that are important for memory function in CD8+ T cells. These include cytokine receptors that are important for their homeostatic survival (IL-7R), as well as the chemokine receptors (CCR7) and adhesion molecules (CD62L/L-selectin) that direct migration of T cells to lymph nodes and retain them there. The ‘‘memory-like’’ transcriptional profile was subsequently confirmed at the level of CD8+ T cell function in vivo by showing that inhibition of Akt or PI3K p110d changed the differentiation fate of CD8+ T cells that were originally destined to become effectors, in effect ‘‘reprogramming’’ them to become memory phenotype cells by restoring the expression of CCR7 and CD62L and thereby enabling these cells to home to secondary lymphoid organs in adoptive transfer experiments just as bona fide memory cells would be expected to do. Concomitant with induction of genes related to memory function, Macintyre et al. also show that Akt inhibition caused a decrease in the expression of genes encoding molecules critical to CD8+ T cell effector function, including Interferon gamma (IFN-g), Fas ligand (FasL), several of the granzymes, as well as perforin. The picture emerging from these findings is that IL-2-driven Akt activation simultaneously coordinates a program of gene expression related to the development of effector functions while repressing genes involved in determining the memory phenotype, a far cry from its expected role in control of metabolism. Macintyre et al. delved further into the underlying mechanism and found that a common feature of the Akt-responsive genes is their regulation by the Foxo transcription factors, mammalian orthologs of the ‘‘Forkhead box’’ factors first identified in Caenorhabditis elegans in programming resistance to oxidative stress (Burgering, 2008). Foxo factors have been shown to control highly specialized programs of
142 Immunity 34, February 25, 2011 ª2011 Elsevier Inc.
gene expression in the adaptive immune response of mammals through both transactivation and repression, including those involved in the differentiation, survival, and metabolism of lymphocytes (Hedrick, 2009). As with many transcription factors, the activity of the Foxos is regulated via their phosporylation-dependent nuclear exclusion through binding to cytoplasmic 14-3-3 proteins. Previously, work in human T cells had established that Akt phosphorylates and inactivates key Foxo proteins in T cells through this mechanism, and a more recent report has shown that Foxo proteins can regulate expression of the same ‘‘memory phenotype’’ molecules such as IL-7R, CCR7, and L-selectin that Macintyre et al. identified in their study in a process that involves kruppel-like factor-2 (KLF2) as a critical target gene for their induction (Fabre et al., 2005; Kerdiles et al., 2009). In an opposing role in regulating memory genes, Macintyre et al. also found that Foxo factors are responsible for the capacity of Akt to maintain key effector molecules with the demonstration that a mutant form of Foxo3a lacking sites required for phosphorylation by Akt prevented CD8+ T cells from producing IFNg in response to TCR signaling. The cumulative data in this and recent allied reports suggest that the role of Akt in guiding the fate of CD8+ T cells is one of coordination of mutually exclusive programs of gene expression. Strong and/or sustained Akt signals will favor the development of effectors in TCR-stimulated CD8+ T cells by promoting genes that contribute to effector functions while repressing those associated with memory. The inhibition studies performed my Macintyre et al. suggest that when these signals are weak, or perhaps antagonized at various levels, the differentiation process will lead to the generation of memory cells capable of localizing secondary lymphoid organs. Remarkably, both outcomes appear to be controlled by a common mechanism involving specific members of the Foxo family of transcription factors whose Akt-regulated presence within the nucleus can either induce or repress the relevant target genes (Figure 1). These findings shed new light on a unique link established by a kinase that likely found its original function in the pan-cellular process of metabolism that evolved to direct specialized
Immunity
Previews system used by Macintyre et al. may play a role in generating the range of functional phenotypes that can emerge from a single CD8+ T cell precursor primed in intact animals by a complex pathogen (Stemberger et al., 2007). Accordingly, the functional heterogeneity found in effector versus memory phenotype cells may turn out to reflect specific differences in the content or activity of Foxo factors or in the range of cofactors that modify their range of target genes. In any case, these new findings are certain to generate much discussion in the field and give new perspective to ongoing experiments into the molecular regulation of immune memory.
IL-2
PI3K
PI(3,4,5)P3 PH
PH
PDK1
AKT P
AGC kinases Proliferation Glucose uptake
p110δ
P
-
+
14-3-3 P
P
P
P
P
P
Foxo
Foxo JNK kinase? PTEN? Phosphatases?
Nucleus Foxo
Genes that encode: KLF2 CD62L CCR7 CD27 No Foxo IL-7R IL-6R Foxo
REFERENCES Burgering, B.M. (2008). Oncogene 27, 2258–2262. Genes that encode: IFN-γ Perforin FasL Granzymes IL-12Rβ
Figure 1. Akt-Dependent Regulation of Effector and Memory Cell-Associated Genes in CD8+ T Cells IL-2 signals act through PI3K to generate PI(3,4,5)P3 and recruit PDK1 and Akt to the membrane where they become activated by phosphorylation. Active PDK1 can then work through downstream members of the AGC kinase family to provide for the metabolic needs of proliferation. PDK1 also phosphorylates Akt, which can in turn phosphorylate and inactivate Foxo transcription factors on multiple residues and facilitate binding to cytosolic 14-3-3 proteins, which prevents translocation into the nuclues. In the nucleus, Foxo transcription factors can induce transcription of memory phenotype genes, including Klf2 and its target genes involved in trafficking and homeostatic survival, whereas its exclusion from the nucleus in stimulated CD8+ T cells results in repression of KLF-2 target genes and promotes expression of genes associated with effector functions.
programs of genes expression that are key to the functions of effector and memory CD8+ T cells. The new perspective offered by the results of Macintyre et al. were based in an in vitro experimental system, however, which lacks
many of the additional signals found under in vivo priming conditions that could possibly counteract or modify the TCR + IL-2 signals used to stimulate CD8+ T cells in this report. As such, these additional signals missing from the in vitro
Fabre, S., Lang, V., Harriague, J., Jobart, A., Unterman, T.G., Trautmann, A., and Bismuth, G. (2005). J. Immunol. 174, 4161–4171. Finlay, D., and Cantrell, D. (2010). Ann. N Y Acad. Sci. 1183, 149–157. Hedrick, S.M. (2009). Nat. Immunol. 10, 1057– 1063. Juntilla, M.M., and Koretzky, G.A. (2008). Immunol. Lett. 116, 104–110. Kerdiles, Y.M., Beisner, D.R., Tinoco, R., Dejean, A.S., Castrillon, D.H., DePinho, R.A., and Hedrick, S.M. (2009). Nat. Immunol. 10, 176–184. Macintyre, A.N., Finlay, D., Preston, G., Sinclair, L.V., Waugh, C.M., Tamas, P., Feijoo, C., Okkenhaug, K., and Cantrell, D.A. (2011). Immunity 34, this issue, 224–236. Manjunath, N., Shankar, P., Wan, J., Weninger, W., Crowley, M.A., Hieshima, K., Springer, T.A., Fan, X., Shen, H., Lieberman, J., and von Andrian, U.H. (2001). J. Clin. Invest. 108, 871–878. Stemberger, C., Huster, K.M., Koffler, M., Anderl, F., Schiemann, M., Wagner, H., and Busch, D.H. (2007). Immunity 27, 985–997.
Immunity 34, February 25, 2011 ª2011 Elsevier Inc. 143
Previews Given the pro-B cell-specific and Pax5 dependency of the PAIR sequence, as measured by transcriptional activity and pro-B cell preferential binding of CTCF and cohesin to these same sequences, coupled with known functional requirement of Pax5 in distal VH recombination, Ebert et al. (2011) propose that PAIRs may represent the long sought cis-regulatory elements critical for Igh looping to ensure utilization of distal VHs in VHDJH rearrangements. In the context of such a model, Pax5 might activate PAIRs, perhaps via antisense transcription, allowing PAIRs to form the base of rosette loops that bring distal VHs into V(D)J recombination complex with the DJH, iEm, and potentially other elements via CBEs in the VH to DH intergenic region and/or in the region downstream of the CHs. Testing of such models will require deleting these proposed regulatory elements and mutating factor binding sites within them to test effects on V(D)J recombination. As noted by Ebert et al. (2011), this will be challenging given the repetitive nature of the VH locus as well as potential redundancy. It may also be further complicated if there
are many different VH loops formed across the Igh locus, with only one or a few in a given progenitor B cell. Such experimental difficulties might require, if possible, truncating the VH portion of Igh. Another very important experiment will be to mutate the more unique potential looping elements (e.g., CBEs) in the downstream portions of Igh to test effects on distal VH gene utilization. Finally, given that the large human IGH must also have mechanisms to ensure utilization of distal versus proximal VHs, it will be important to determine whether PAIR-related sequence exist in the human. Overall, the finding of the PAIR sequences in the upstream portion of the Igh locus provides an important focus for future studies aimed at solving the long-standing mystery of how the Igh locus organizes itself during V(D)J recombination to ensure broad VH utilization and a diverse antibody repertoire.
Degner-Leisso, S.C., and Feeney, A.J. (2010). Semin. Immunol. 22, 346–352. Ebert, A., McManus, S., Tagoh, H., Medvedovic, J., Salvagiotto, G., Novatchkova, M., Tamir, I., Sommer, A., Jaritz, M., and Busslinger, M. (2011). Immunity 34, this issue, 175–187. Fuxa, M., Skok, J., Souabni, A., Salvagiotto, G., Roldan, E., and Busslinger, M. (2004). Genes Dev. 18, 411–422. Jhunjhunwala, S., van Zelm, M.C., Peak, M.M., Cutchin, S., Riblet, R., van Dongen, J.J., Grosveld, F.G., Knoch, T.A., and Murre, C. (2008). Cell 133, 265–279. Jhunjhunwala, S., van Zelm, M.C., Peak, M.M., and Murre, C. (2009). Cell 138, 435–448. Malynn, B.A., Yancopoulos, G.D., Barth, J.E., Bona, C.A., and Alt, F.W. (1990). J. Exp. Med. 171, 843–859. Subrahmanyam, R., and Sen, R. (2010). Semin. Immunol. 22, 337–345.
REFERENCES
Yancopoulos, G.D., Desiderio, S.V., Paskind, M., Kearney, J.F., Baltimore, D., and Alt, F.W. (1984). Nature 311, 727–733.
Corcoran, A.E. (2010). Semin. Immunol. 22, 353–361.
Yancopoulos, G.D., Blackwell, T.K., Suh, H., Hood, L., and Alt, F.W. (1986). Cell 44, 251–259.
From the Loading Dock to the Boardroom: A New Job for Akt Kinase Sonia Feau1 and Stephen P. Schoenberger1,* 1Laboratory of Cellular Immunology, La Jolla Institute for Allergy and Immunology, 9420 Athena Circle, La Jolla, CA 92037, USA *Correspondence: [email protected] DOI 10.1016/j.immuni.2011.02.013
A new report in Immunity shows that, rather than driving the metabolic changes required for proliferation, Akt controls the gene expression programs that determine whether activated CD8+ T cells differentiate into memory or effector cells. The multifunctional cytokine interleukin-2 (IL-2) can drive metabolic changes needed to insure that activated CD8+ T cells can rapidly expand in response to infection as well as develop the effector functions needed to eradicate the threat posed by infected host cells. Additionally, the capacity to produce IL-2 is a central feature of the long-lived memory population that persist following contraction of
the effectors. A key goal for immunologists has therefore been to identify the pathways and players through which signals emanating from a single cytokine receptor can lead to the downstream gene expression events required to coordinate the metabolic and differentiation needs of an expanding clone. An important early ‘‘branch point’’ for this division of labor is thought to occur with the IL-2-
mediated generation of the lipid second messenger phosphatidylinositol 3,4,5-trisphosphate (PI(3,4,5)P3), the product of phosphatidylinositol 3-kinases (PI3K). This phosporylated lipid recruits the specific signaling intermediates via specialized lipid-binding pleckstrin homology (PH) domains and itself becomes activated to initiate further signaling cascades. Two key players in this process
Immunity 34, February 25, 2011 ª2011 Elsevier Inc. 141
Immunity
Previews are Akt (protein kinase B, PKB) and PDK1 (3-phosphoinositide-dependent protein kinase 1), both which can be recruited to the plasma membrane by PI(3,4,5)P3) and activate downstream kinases. In the case of PDK1, Akt itself is a substrate for PDK1-dependent activation via phosphorylation at its T308 site. Akt has traditionally been associated with metabolic pathways in T cells, based mostly on studies in which it was targeted to the membrane by myristoylation and manifested a corresponding effect on growth and survival of proliferating thymocytes (Juntilla and Koretzky, 2008). A similar role in mature T cells was suggested by studies using constitutively active Akt that could induce growth and survival of CD4+, but remarkably not CD8+, T cells, raising the possibility that these subsets are wired differently with respect to the downstream consequences of activating the PI3K-Akt pathway (Finlay and Cantrell, 2010). In this issue of Immunity, Macintyre et al. (2011) reveal that, in contrast to controlling metabolism in CD8+ T cells, Akt has a direct role in regulating whether their differentiation leads them to either the memory or effector fate. To investigate the role of Akt in CD8+ T cells, Macintyre et al. used an in vitro model in which T cell receptor (TCR)-stimulated cells are cultured in IL-2 in order to sustain the signals that lead effector cell generation (Manjunath et al., 2001). Using either pharmacologic or genetic inhibition of the p110d subunit of PI3K that activates Akt or an allosteric inhibitor of all 3 Akt isoforms, Macintyre et al. found that, in the absence of Akt function, TCR-stimulated CD8+ T cells could both proliferate and survive in response to IL-2. Subsequent studies using a genetically targeted model showed that IL-2-driven proliferation is actually mediated by PDK1, not Akt, as deletion of PDK1 in CD8+ T cells rendered them unable to proliferate or sustain glucose uptake in response to IL-2. Remarkably, although the PDK1-deficient CD8+ T cells failed to divide in response to IL-2, they did not die, suggesting that survival is mediated by yet another distinct pathway. These findings reveal that Akt controls neither IL-2-mediated proliferation nor glucose uptake in this system and raised questions as to its actual function in enabling the CD8+ T cell response. To further
explore this, Macintyre et al. performed microarray analysis to profile geneexpression patterns from Akt-inhibited and PDK1 null CD8+ T cells. In support of their biological data, they found that genes encoding proteins involved in glucose uptake and lipid metabolism were overrepresented among genes regulated by PDK1 independent of Akt. Surprisingly, the Akt-inhibited CD8+ T cells induced genes that are important for memory function in CD8+ T cells. These include cytokine receptors that are important for their homeostatic survival (IL-7R), as well as the chemokine receptors (CCR7) and adhesion molecules (CD62L/L-selectin) that direct migration of T cells to lymph nodes and retain them there. The ‘‘memory-like’’ transcriptional profile was subsequently confirmed at the level of CD8+ T cell function in vivo by showing that inhibition of Akt or PI3K p110d changed the differentiation fate of CD8+ T cells that were originally destined to become effectors, in effect ‘‘reprogramming’’ them to become memory phenotype cells by restoring the expression of CCR7 and CD62L and thereby enabling these cells to home to secondary lymphoid organs in adoptive transfer experiments just as bona fide memory cells would be expected to do. Concomitant with induction of genes related to memory function, Macintyre et al. also show that Akt inhibition caused a decrease in the expression of genes encoding molecules critical to CD8+ T cell effector function, including Interferon gamma (IFN-g), Fas ligand (FasL), several of the granzymes, as well as perforin. The picture emerging from these findings is that IL-2-driven Akt activation simultaneously coordinates a program of gene expression related to the development of effector functions while repressing genes involved in determining the memory phenotype, a far cry from its expected role in control of metabolism. Macintyre et al. delved further into the underlying mechanism and found that a common feature of the Akt-responsive genes is their regulation by the Foxo transcription factors, mammalian orthologs of the ‘‘Forkhead box’’ factors first identified in Caenorhabditis elegans in programming resistance to oxidative stress (Burgering, 2008). Foxo factors have been shown to control highly specialized programs of
142 Immunity 34, February 25, 2011 ª2011 Elsevier Inc.
gene expression in the adaptive immune response of mammals through both transactivation and repression, including those involved in the differentiation, survival, and metabolism of lymphocytes (Hedrick, 2009). As with many transcription factors, the activity of the Foxos is regulated via their phosporylation-dependent nuclear exclusion through binding to cytoplasmic 14-3-3 proteins. Previously, work in human T cells had established that Akt phosphorylates and inactivates key Foxo proteins in T cells through this mechanism, and a more recent report has shown that Foxo proteins can regulate expression of the same ‘‘memory phenotype’’ molecules such as IL-7R, CCR7, and L-selectin that Macintyre et al. identified in their study in a process that involves kruppel-like factor-2 (KLF2) as a critical target gene for their induction (Fabre et al., 2005; Kerdiles et al., 2009). In an opposing role in regulating memory genes, Macintyre et al. also found that Foxo factors are responsible for the capacity of Akt to maintain key effector molecules with the demonstration that a mutant form of Foxo3a lacking sites required for phosphorylation by Akt prevented CD8+ T cells from producing IFNg in response to TCR signaling. The cumulative data in this and recent allied reports suggest that the role of Akt in guiding the fate of CD8+ T cells is one of coordination of mutually exclusive programs of gene expression. Strong and/or sustained Akt signals will favor the development of effectors in TCR-stimulated CD8+ T cells by promoting genes that contribute to effector functions while repressing those associated with memory. The inhibition studies performed my Macintyre et al. suggest that when these signals are weak, or perhaps antagonized at various levels, the differentiation process will lead to the generation of memory cells capable of localizing secondary lymphoid organs. Remarkably, both outcomes appear to be controlled by a common mechanism involving specific members of the Foxo family of transcription factors whose Akt-regulated presence within the nucleus can either induce or repress the relevant target genes (Figure 1). These findings shed new light on a unique link established by a kinase that likely found its original function in the pan-cellular process of metabolism that evolved to direct specialized
Immunity
Previews system used by Macintyre et al. may play a role in generating the range of functional phenotypes that can emerge from a single CD8+ T cell precursor primed in intact animals by a complex pathogen (Stemberger et al., 2007). Accordingly, the functional heterogeneity found in effector versus memory phenotype cells may turn out to reflect specific differences in the content or activity of Foxo factors or in the range of cofactors that modify their range of target genes. In any case, these new findings are certain to generate much discussion in the field and give new perspective to ongoing experiments into the molecular regulation of immune memory.
IL-2
PI3K
PI(3,4,5)P3 PH
PH
PDK1
AKT P
AGC kinases Proliferation Glucose uptake
p110δ
P
-
+
14-3-3 P
P
P
P
P
P
Foxo
Foxo JNK kinase? PTEN? Phosphatases?
Nucleus Foxo
Genes that encode: KLF2 CD62L CCR7 CD27 No Foxo IL-7R IL-6R Foxo
REFERENCES Burgering, B.M. (2008). Oncogene 27, 2258–2262. Genes that encode: IFN-γ Perforin FasL Granzymes IL-12Rβ
Figure 1. Akt-Dependent Regulation of Effector and Memory Cell-Associated Genes in CD8+ T Cells IL-2 signals act through PI3K to generate PI(3,4,5)P3 and recruit PDK1 and Akt to the membrane where they become activated by phosphorylation. Active PDK1 can then work through downstream members of the AGC kinase family to provide for the metabolic needs of proliferation. PDK1 also phosphorylates Akt, which can in turn phosphorylate and inactivate Foxo transcription factors on multiple residues and facilitate binding to cytosolic 14-3-3 proteins, which prevents translocation into the nuclues. In the nucleus, Foxo transcription factors can induce transcription of memory phenotype genes, including Klf2 and its target genes involved in trafficking and homeostatic survival, whereas its exclusion from the nucleus in stimulated CD8+ T cells results in repression of KLF-2 target genes and promotes expression of genes associated with effector functions.
programs of genes expression that are key to the functions of effector and memory CD8+ T cells. The new perspective offered by the results of Macintyre et al. were based in an in vitro experimental system, however, which lacks
many of the additional signals found under in vivo priming conditions that could possibly counteract or modify the TCR + IL-2 signals used to stimulate CD8+ T cells in this report. As such, these additional signals missing from the in vitro
Fabre, S., Lang, V., Harriague, J., Jobart, A., Unterman, T.G., Trautmann, A., and Bismuth, G. (2005). J. Immunol. 174, 4161–4171. Finlay, D., and Cantrell, D. (2010). Ann. N Y Acad. Sci. 1183, 149–157. Hedrick, S.M. (2009). Nat. Immunol. 10, 1057– 1063. Juntilla, M.M., and Koretzky, G.A. (2008). Immunol. Lett. 116, 104–110. Kerdiles, Y.M., Beisner, D.R., Tinoco, R., Dejean, A.S., Castrillon, D.H., DePinho, R.A., and Hedrick, S.M. (2009). Nat. Immunol. 10, 176–184. Macintyre, A.N., Finlay, D., Preston, G., Sinclair, L.V., Waugh, C.M., Tamas, P., Feijoo, C., Okkenhaug, K., and Cantrell, D.A. (2011). Immunity 34, this issue, 224–236. Manjunath, N., Shankar, P., Wan, J., Weninger, W., Crowley, M.A., Hieshima, K., Springer, T.A., Fan, X., Shen, H., Lieberman, J., and von Andrian, U.H. (2001). J. Clin. Invest. 108, 871–878. Stemberger, C., Huster, K.M., Koffler, M., Anderl, F., Schiemann, M., Wagner, H., and Busch, D.H. (2007). Immunity 27, 985–997.
Immunity 34, February 25, 2011 ª2011 Elsevier Inc. 143