Articles

The kinase p38 activated by the metabolic regulator AMPK and scaffold TAB1 drives the senescence of human T cells © 2014 Nature America, Inc. All rights reserved.

Alessio Lanna1, Sian M Henson1, David Escors1,2 & Arne N Akbar1 In T lymphocytes, the mitogen-activated protein kinase (MAPK) p38 regulates pleiotropic functions and is activated by canonical MAPK signaling or the alternative activation pathway downstream of the T cell antigen receptor (TCR). Here we found that senescent human T cells lacked the canonical and alternative pathways for the activation of p38 but spontaneously engaged the metabolic master regulator AMPK to trigger recruitment of p38 to the scaffold protein TAB1, which caused autophosphorylation of p38. Signaling via this pathway inhibited telomerase activity, T cell proliferation and the expression of key components of the TCR signalosome. Our findings identify a previously unrecognized mode for the activation of p38 in T cells driven by intracellular changes such as low-nutrient and DNA-damage signaling (an ‘intrasensory’ pathway). The proliferative defect of senescent T cells was reversed by blockade of AMPK-TAB1–dependent activation of p38. T cell proliferative responses are essential for the lifelong preservation of adaptive immunity. Optimal activation of T cells requires both engagement of the T cell antigen receptor (TCR) and costimulatory signals1; however, progressive loss of the costimulatory receptors CD27 and CD28 occurs during the end-stage differentiation of T cells toward senescence that correlates well with loss of proliferation2. The mechanisms that regulate the loss of proliferative potential in highly differentiated T cells are poorly understood. The aim of this study was to identify mechanisms involved in the end-stage differentiation of human T cells and whether intervention is possible to restore proliferative activity in these cells. The loss of surface CD27 followed by loss of CD28 expression during the differentiation of human CD4+ T cells3 enables the identification of undifferentiated T cells that have high proliferative activity and the longest telomeres (CD27+CD28+); intermediate differentiated cells that have reduced proliferation and telomeres of intermediate length (CD27−CD28+); and end-stage or senescent T cells that proliferate poorly and have the shortest telomeres (CD27−CD28−)3. Senescent CD27−CD28− CD4+ T lymphocytes accumulate substantially in old humans and in patients with chronic viral infections and those with autoimmune disorders3–7. The senescence characteristics of these cells include short telomeres, low telomerase activity and reduced proliferative capability8 that is due in part to a spontaneous but unexplained increase in activity of the mitogen-activated protein kinase (MAPK) p38 (ref. 9). Nevertheless, CD27−CD28− CD4+ T cells exhibit potent effector functions and cannot be viewed as a dysfunctional population per se3–7. Here we investigated the regulatory network upstream of p38 in the senescent CD27−CD28− CD4+

T cell subset. The p38 family comprises the p38α, p38β, p38γ and p38δ isoforms that regulate many different functions in a cell-type– and stimulus-dependent manner10. T lymphocytes express mainly the p38α isoform and, to a lesser extent, the p38β and p38δ isoforms10. The canonical MAPK cascade is the main mechanism for the activation of all four p38 isoforms in mammalian cells, including T cells10–12. Signaling through the MAPK pathway activates p38 by dual phosphorylation at Thr180 and Tyr182 in response to environmental stress, inflammation, DNA damage12 and also the engagement of costimulatory receptors13. In addition, T cells have an alternative pathway for the activation of p38 in which antigenic stimulation via the TCR induces phosphorylation of p38α and p38β at Tyr323 by TCR-proximal tyrosine kinases14, followed by autophosphorylation at Thr180 and Tyr182 (ref. 14). We found here that instead of using the canonical or alternative pathway, freshly isolated CD27−CD28− CD4+ T lymphocytes engaged a distinct mechanism whereby the intracellular metabolic sensor AMPK triggered autophosphorylation of p38 via the scaffold protein TAB1. Furthermore, signaling through this pathway downregulated the expression of central components of the TCR signalosome, a previously unidentified characteristic of T cell senescence. Activation of p38 through AMPK-TAB1 was induced not only by endogenous DNA damage but also by a decrease in intracellular concentrations of glucose. Thus, unlike the canonical and alternative pathways for the activation of p38 that respond to external cues (for example, cytokines or activation of the TCR), T cells have an ‘intrasensory’ pathway for the activation of p38 that senses intracellular changes, such as glucose deprivation and genotoxic stress. Once triggered, this

1Division

of Infection and Immunity, University College London, London, UK. 2Navarrabiomed-Fundacion Miguel Servet, Complejo Hospitalario de Navarra, Pamplona, Spain. Correspondence should be addressed to A.N.A. ([email protected]). Received 21 February; accepted 29 July; published online 24 August 2014; doi:10.1038/ni.2981

nature immunology  aDVANCE ONLINE PUBLICATION



Articles pathway with the osmotic stressor sorbitol (data not shown). Thus, despite the constitutive increase in p38 activity (Fig. 1b), human senescent CD27−CD28− CD4+ T cells did not regulate p38 signaling by the canonical MAPK cascade. Antigenic stimulation through the TCR activates p38 by an alternative pathway in which a tyrosine kinase cascade that is dependent on the kinase Lck and is mediated by the signaling kinase Zap70 phosphorylates p38 at Tyr323 (ref. 14). The membrane-associated scaffold molecule DLG1 binds p38 and coordinates this process18. Because persistent antigenic stimulation drives the differentiation of human T cells in vivo2,3, we investigated whether the alternative pathway for the activation of p38 via TCR signaling was responsible for inducing p38 signaling in CD27−CD28− CD4+ T cells. However, we found that the CD27−CD28− CD4+ population did not phosphorylate p38 on the alternative Tyr323 site before or after stimulation with antibody to the invariant signaling protein CD3 (anti-CD3) (Fig. 1g). In contrast, undifferentiated CD27+CD28+ CD4+ T cells had abundant phosphorylation of p38 at Tyr323 upon activation with anti-CD3 (Fig. 1g). We next investigated expression of elements of the TCR signalosome that constitute the alternative pathway for p38 activation in CD27+CD28+, CD27−CD28+ and CD27−CD28− CD4+ T cells. We found that in contrast to the CD27+CD28+ and CD27−CD28+ CD4+ subsets, freshly isolated CD27−CD28− CD4+ T cells did not express Lck, Zap70 or DLG1 (Fig. 1h,i). In addition, these cells also lacked other essential downstream components of the TCR signaling machinery, such as the adaptors Lat and SLP-76, and had very low expression of the phospholipase PLC-γ1 (Supplementary Fig. 1a,b). The loss of TCR signaling machinery in senescent CD27−CD28− CD4+

8.2

87.8

4

10

3

10

2

10 0

4

0.01 2

0 10

CD27

3

10

4

10

5

10

p-p38(T180,Y182) MFI (fold)

5

10

CD27+CD28+ CD27–CD28+ CD27–CD28–

c

d CD27 + CD28 +

3.0

**

MKK3

2.0

*

MKK6

1.0 0

p-MKK3, MKK6 GAPDH

– +

– –

CD27+CD28+ CD27–CD28+ CD27–CD28– 1.5 1.0

f CD27–CD28–

**

0.5 0

e

ND

75

CD27–CD28– + PMA 73 CD27+CD28+ 38 + + CD27 CD28 + PMA 230 0 10

2

10

3

10

4

p-p38(T180,Y182)

10

5

p-p38(T180,Y182) MFI (fold)

b

p-MKK3, MKK6 (fold)

a

Count

RESULTS p38 activation without canonical or alternative pathways The relative expression of CD27 and CD28 identified three distinct populations in primary human CD4+ T cells (Fig. 1a). CD27−CD28− T cells are known to exhibit low telomerase and proliferative activity upon activation3, which are characteristics of end-stage differentiation or senescence3,4,9,15. The CD27−CD28− subset of CD4+ T cells also exhibited endogenously elevated (spontaneous) phosphorylation of p38 (at Thr180 and Tyr182) that was significantly (2- to 2.5-fold) higher than that of the CD27+CD28+ or CD27−CD28+ T cell subset in vivo (Fig. 1b). We investigated whether the canonical MAPK cascade, which regulates the activation of p38 in response to stress stimuli and the engagement of costimulatory receptors12,13, was responsible for the endogenous activation of p38 observed in CD27−CD28− CD4+ T cells. Freshly isolated human CD27−CD28− CD4+ T cells did not express or activate either MKK3 or MKK6 (Fig. 1c,d), the direct upstream regulators of p38 in this signaling cascade12. Similarly, these cells did not activate MKK4 (data not shown), an activator of the MAPK Jnk16 that in some conditions may also activate p38 (ref. 10). Furthermore, the phorbol ester PMA, a well-established agonist of the canonical MAPK cascade17, failed to enhance p38 activity in CD27−CD28− CD4+ T cells but induced a significant increase in the phosphorylation of p38 in the undifferentiated CD27+CD28+ subset (Fig. 1e,f). We made similar observations when we induced the canonical p38

CD28

–

8.0

–

CD27 CD28 – – CD27 CD28 + PMA + + CD27 CD28 CD27+CD28+ + PMA

**

6.0 4.0

**

2.0 0

i

CD27+CD28+ –

+

CD27 CD28 Figure 1  Spontaneous activation of p38 in the absence of upstream canonical and CD27–CD28– alternative pathways in senescent T cells. (a) Expression of CD27 and CD28 by human CD27 – – + + + CD28 – – + + 1.5 CD4 T cells (n = 9 donors). Numbers in quadrants indicate percent cells in each throughout. ** * p-p38(Y323) (b) Phosphorylation of p38 at Thr180 and Tyr182 (p-p38(T180,Y182)) in subsets of CD4 + 1.0 T cells (n = 9 donors) defined by expression of CD27 and CD28 (key; gated as in a), assessed p38 0.5 by phosphorylation-specific flow cytometry and presented as mean fluorescence intensity (MFI) Anti-CD3 – + – + relative to that in CD27+CD28+ cells, set as 1. (c) Immunoblot analysis of total MKK3 and 0 MKK6 and of MKK3 and/or MKK6 phosphorylated at Ser189 and Ser206 (p-MKK3,MKK6) 1.5 ** in subsets of CD4+ T cells gated as in a (above lanes); GAPDH serves as a loading control CD27 + + – – – – * + + + – – CD28 + throughout. (d) Endogenous activation of MKK3 and/or MKK6 in subsets of CD4 + T cells 1.0 Lck (n = 4 donors) gated as in a, assessed as phosphorylated MKK3 and/or MKK6 versus total 0.5 GAPDH. (e) Phosphorylation of p38 at Thr180 and Tyr182 in CD27 −CD28− and CD27+CD28+ + Zap70 CD4 T cells before and after (+ PMA) treatment for 60 min with PMA (20 ng/ml), assessed 0 by phosphorylation-specific flow cytometry. Numbers above lines in plots indicate mean 1.5 ** DLG1 fluorescence intensity throughout. (f) Summary of results obtained as in e, assessed as in * + + 1.0 b and presented relative to those of CD27 CD28 cells without PMA treatment, set as 1. GAPDH (g) Immunoblot analysis of total p38 and p38 phosphorylated at Tyr323 (p-p38(Y323)) in 0.5 A B A B A B isolated CD27−CD28− and CD27+CD28+ CD4+ T cells before (Anti-CD3 −) or after (Anti-CD3 +) ND 0 Donor activation for 30 min with anti-CD3 (10 µg/ml). (h) Immunoblot analysis of total Lck, Zap70 and DLG1 in subsets of CD4+ T cells (from two donors (A and B)) gated as in a. (i) Expression of Lck (top), Zap70 (middle) and DLG1 (bottom) in subsets of CD4 + T cells (n = 4 donors) defined as in a, presented relative to total GAPDH. ND, not detected. *P < 0.01 and **P < 0.001 (one-way analysis of variance (ANOVA) for repeated measures with a Bonferroni post-test correction). Data are representative of nine experiments (a), four independent experiments (c), three independent experiments (e,g) or four experiments (h) or are pooled from nine (b), four (d,i) or three (f) independent experiments (error bars (b,d,f,i), s.e.m.).

Lck (fold)

g

Zap70 (fold)

h

DLG1 (fold)

© 2014 Nature America, Inc. All rights reserved.

pathway inhibits T cell proliferation and telomerase activation via p38 signaling, and this can be reversed by inhibition of AMPK, TAB1 or p38 itself.



aDVANCE ONLINE PUBLICATION  nature immunology

Articles

AMPKα

– +

– –

Ctrl 47.5 AMPK act 103.5 2

3

p-p38

TRAF6

4

0 10 10 10 10

5

2.5

**

2.0 1.5 1.0 0.5 0

250k 200k 150k 100k 50k 0k

22

2

3

4

5

f

577

g

392 2

3

4

0 10 10 10 10

p-p38 (T180,Y182) shCtrl

shAMPKα

shTAB1

Steady state GFP+ 806

GFP

5

h

i

1.5 1.0

** * *

0.5 0

shCtrl

T cells was also associated with defective calcium influx upon stimulation with anti-CD3 (Supplementary Fig. 1c,d), which indicated that the loss of TCR signaling molecules impaired TCR responsiveness. Together these data indicated that the spontaneous increase in p38 activity observed in CD27−CD28− CD4+ T cells was not regulated by either the canonical MKK signaling pathway or the alternative TCR-mediated signaling pathway, in contrast to the regulation of p38 in nonsenescent CD4+ T cells11,19. p38 activation is triggered by AMPK and mediated by TAB1 In HEK293 human embryonic kidney cells, allosteric autophosphorylation of p38 at Thr180 and Tyr182 is induced independently of upstream MKKs via the scaffold molecule TAB1 (ref. 20). We investigated whether TAB1 signaling was involved in the constitutive activation of p38 observed in senescent CD27 −CD28− CD4+ T cells. Two TAB1 splice variants (α or β, with a molecular mass of 60 kilodaltons (kDa) or 50 kDa, respectively) bind to and activate p38 by an identical mechanism 21. Using a monoclonal antibody directed against the p38-binding domain common to both isoforms, we found that CD27−CD28− CD4+ T cells expressed the 50-kDa isoform of TAB1 (Fig. 2a), which does not bind to or regulate the kinase TAK1, its binding partner21. In addition, CD27−CD28− CD4+ T cells also did not express TAK1 itself (Fig. 2a). Conversely, nonsenescent T cell subsets expressed TAK1 and also full-length TAB1, which binds to TAK1 (Fig. 2a). Because activation of the TAB1 pathway requires nondegradative ubiquitination by the upstream E3 ubiquitin ligase TRAF6 (ref. 20), we examined the expression of TRAF6 in the nonsenescent and senescent CD4 + T cell subsets. CD27−CD28− CD4+ T cells lacked TRAF6 expression (Fig. 2a), which indicated that the TRAF6-TAK1 pathway was not active in these cells. AMPK is activated in response to low intracellular concentrations of glucose, and activated AMPK has been shown to bind to TAB1 in ischemic mouse cardiomyocytes22. This molecule is activated by nature immunology  aDVANCE ONLINE PUBLICATION

p-p38(T180,Y182) MFI (fold)

shTAB1 shTAB1 AMPK act Figure 2  Spontaneous activation of p38 triggered by AMPK and mediated AMPK act shAMPKα shAMPKα GFP+ + by TAB1 in senescent T cells. (a,b) Immunoblot analysis of TAB1 GFP 1,742 1.5 1.5 *** (right margin, alternative isoforms), TAK1 and TRAF6 (a) and of total 18,494 *** *** 1,217 AMPKα and AMPKα phosphorylated at Thr172 (p-AMPKα) (b) in freshly 1.0 1.0 8,013 isolated CD4+ T cells sorted into subsets by expression of CD27 and CD28 926 * 5,816 2 3 4 5 (gated as in Fig. 1a). A and B (below lanes, a), two donors. (c) Phosphorylation 0.5 0.5 0 10 10 10 10 2 3 4 5 0 10 10 10 10 + + + p-p38 of p38 at Thr180 and Tyr182 in CD27 CD28 CD4 T cells with (AMPK act) (T180,Y182) p-ATF2(T71) 0 0 or without (control (Ctrl)) treatment for 60 min with the AMPK activator A-769662 (150 µM), assessed by phosphorylation-specific flow cytometry. (d) Summary of results obtained as in c, presented relative to those of control (untreated) cells, set as 1. (e) Expression of the GFP reporter (left) in purified CD27−CD28− CD4+ T cells (n = 9 donors) 96 h after transduction of shCtrl, shTAB1 or shAMPKα (confirmation of knockdown, Supplementary Fig. 2), and steady-state phosphorylation of p38 at Thr180 and Tyr182 (middle) in the GFP + populations of CD27−CD28− CD4+ T cells outlined at left. Right, summary of data in middle, presented relative to those of cells transduced with shCtrl, set as 1. Number in outlined area (left) indicates percent GFP+ cells. (f) Phosphorylation of p38 at Thr180 and Tyr182 in the GFP + subset of CD27−CD28− CD4+ T cells transduced as in e (colors match those in key in g) and treated for 60 min with A-769662 (150 µM), assessed by phosphorylation-specific flow cytometry. (g) Summary of results obtained as in f, presented as in e (right). (h) Phosphorylation of ATF2 at Thr71 (p-ATF2(T71)) in cells as in f. (i) Summary of results obtained as in f, presented as in g. *P < 0.05, **P < 0.01 and ***P < 0.001 (paired Student’s t-test (d) or ANOVA for repeated measures with a Bonferroni post-test correction (e (right), g,i)). Data are representative of four experiments with four (a) or three (b) donors or three independent experiments (c,e (left, middle), f,h) or are pooled from three independent experiments (d,e (right), g,i; error bars (d,e,g,i), s.e.m.). Count

© 2014 Nature America, Inc. All rights reserved.

shCtrl

0 10 10 10 10

A B A B A B Donor

e

p-p38(T180,Y182) MFI (fold)

p-AMPKα TAK1

+ +

Ctrl AMPK act

p-ATF2(T71) MFI (fold)

CD27 CD28

d

Count

50

c

Count

TAB1

b (kDa) 60

SSC

+ + – – – – + + + + – –

p-p38(T180,Y182) MFI (fold)

CD28

Count

a CD27

phosphorylation at Thr172 in its catalytic α-subunit to shift metabolism toward catabolic reactions23. In addition, AMPK inhibits mTOR activity in T cells24,25. We investigated whether AMPK regulated p38 activity in CD27−CD28− CD4+ T cells via TAB1. Freshly isolated human CD27+CD28+, CD27−CD28+ and CD27−CD28− CD4+ T cells all expressed AMPKα, but only the CD27−CD28− CD4+ T cell subset showed spontaneous activation of AMPK (phosphorylation of AMPKα; Fig. 2b). To determine whether AMPK was acting upstream of p38 in T cells, we treated CD27+CD28+ CD4+ T cells, which had no baseline AMPK activity (Fig. 2b) and exhibited low spontaneous phosphorylation of p38 (Fig. 1b), with A-769662, a selective agonist of AMPK26. This agonist induced significant phosphorylation of p38 at Thr180 and Tyr182 in CD27+CD28+ CD4+ T cells (Fig. 2c,d). Despite the high baseline p38 activity of senescent CD27−CD28− CD4+ T cells, A-769662 also enhanced phosphorylation of p38 at Thr180 and Tyr182 in these cells (data not shown). To directly investigate the role of signaling via TAB1 and AMPK in the activation of p38, we used lentiviral vectors (pseudotyped with glycoprotein G of vesicular stomatitis virus) coexpressing a reporter gene encoding green fluorescent protein (GFP) and short hairpin RNA (shRNA) targeting either the catalytic α-subunit of AMPK (shAMPKα) or a region common to both the 60- and 50-kDa isoforms of TAB1 (shTAB1) (Supplementary Fig. 2a–d). We used an irrelevant vector expressing shRNA with a scrambled sequence (shCtrl) as control (Supplementary Fig. 2c,d). We transduced isolated CD27−CD28− CD4+ T cells with these vectors 48 h after activating the cells with anti-CD3 and recombinant human interleukin 2 (IL-2). At 4 d after transduction, we examined changes in the phosphorylation of p38 at Thr180 and Tyr182 in GFP+ T cells by phosphorylationspecific flow cytometry (Fig. 2e). Under steady-state conditions, GFP+ T cells in which expression of either AMPKα or TAB1 was silenced had significantly lower p38 activity than that of cells transduced with shCtrl (Fig. 2e), which indicated that both AMPK and TAB1 regulated the activation of p38 in senescent human CD4+ T cells. 

Articles

0 10 10 10 10

p-ATM (S1921)

–

+

–

3

p-γ-H2AX

2

GAPDH

0

50

CD27 CD28

**

30 20

ATMi 1,201 2

10

3

0 10 10 10

0

i

1.5

Ctrl 1,983

40

p-AMPKα (T172)

4

5

10

Ctrl ATMi

g

h CD27–CD28– Ctrl 160

**

1.0 0.5

ATMi 69 2

3

4

0 10 10 10 10

0

j

k

1.5

p-p38(T180, Y182)

5

Ctrl ATMi

**

1.0 0.5 0

l

shCtrl

We next transduced CD27−CD28− CD4+ T cells with shCtrl, shTAB1 or shAMPKα and treated them for 1 h with the AMPK agonist A-769662, followed immediately by analysis of the activation of p38 in GFP+ T cells by phosphorylation-specific flow cytometry (Fig. 2f,g). The AMPK agonist–induced activation of p38 was inhibited in the CD27−CD28− CD4+ T cells (Fig. 2f,g), which confirmed that this process was AMPK dependent. The AMPK agonist– induced activation of p38 was also inhibited in CD27−CD28− CD4+ T cells transduced with shTAB1 (Fig. 2h,i); this indicated that AMPK required TAB1 to activate p38 in these cells. Silencing of the expression of either AMPKα or TAB1 in AMPK agonist–activated CD27−CD28− CD4+ T cells also inhibited phosphorylation of the p38 substrate ATF2 at residue Thr71 (Fig. 2h,i). Similarly, agonist-driven activation of AMPK failed to activate p38 in CD27+CD28+ CD4+ T cells in which expression of either AMPKα or TAB1 was silenced (data not shown). Thus, activation of AMPK induced p38 signaling in a TAB1-dependent way in senescent human CD4+ T cells. DNA-damage and low-nutrient sensing by AMPK activates p38 Cellular AMPK is activated mainly in response to decreased intracellular ATP23. However, we did not find a lower cellular concentration of ATP in freshly isolated senescent human CD27−CD28− CD4+ T cells than in the CD27+CD28+ or CD27+CD28− subset (data not shown). Because only end-stage–differentiated CD27−CD28− CD4+ T cells exhibited spontaneous AMPK activity, we investigated whether various end-stage–differentiation features of these cells triggered activation of AMPK. We found evidence of endogenous DNA damage27 in CD27−CD28− CD4+ T cells, including spontaneous phosphorylation of ATM, an apical kinase involved in the DNA-damage response, at residue Ser1921 (Fig. 3a,b), and of its downstream target γ-H2AX at Ser139 (Fig. 3c,d), which has been used as an indicator of active DNA-damage foci28. Inhibition of ATM activity for 1 h by the selective 

p-p38(T180,Y182) MFI (fold)

Figure 3  DNA-damage and low-nutrient signals converge at AMPK, CD27+CD28+ shTAB1 Glu+ + + Glu– CD27 CD28 Glu– shAMPKα which drives TAB1-dependent activation of p38 in T cells. + Glu+ GFP 2.0 ** 1.5 (a) Phosphorylation of ATM at Ser1921 (p-ATM(S1921)) in ex vivo, 43.5 *** 606 unstimulated CD27−CD28− and CD27+CD28+ CD4+ T cells (n = 4 1.5 – 396 Glu 1.0 donors), assessed by phosphorylation-specific flow cytometry. 81.5 1.0 256 (b) Summary of results obtained as in a, presented relative to those 2 3 4 5 2 3 4 5 0.5 0 10 10 10 10 0 10 10 10 10 0.5 of CD27+CD28+ cells, set as 1. (c) Immunoblot analysis of γ-H2AX p-p38(T180, p-p38(T180, + phosphorylated at Ser139 (p-γ-H2AX) in freshly isolated CD4 T cells Y182) Y182) 0 0 sorted into subsets by expression of CD27 and CD28 (gated as in Fig. 1a). (d) Frequency of cells containing phosphorylated γ-H2AX (p-γ-H2AX+ cells) among CD27−CD28−, CD27−CD28+ and CD27+CD28+ CD4+ T cells (n = 4 donors), assessed by phosphorylation-specific flow cytometry and presented relative to that among CD27 +CD28+ cells, set as 1. (e) Constitutive phosphorylation of AMPKα at Thr172 (p-AMPKα(T172)) in CD27−CD28− CD4+ T cells with (ATMi) or without (Ctrl) treatment for 60 min with the selective ATM inhibitor KU-55933 (10 µM), assessed by phosphorylation-specific flow cytometry. (f) Summary of results obtained as in e, presented relative to those of untreated (control) cells, set as 1. (g) Phosphorylation of p38 at Thr180 and Tyr182 in cells as in e. (h) Summary of results obtained as in g, presented as in f. (i) Phosphorylation of p38 at Thr180 and Tyr182 in CD27 +CD28+ CD4+ T cells cultured for 18 h in the presence (Glu+) or absence (Glu−) of glucose, assessed by phosphorylation-specific flow cytometry. (j) Summary of results obtained as in i, presented relative to those of cells cultured with glucose, set as 1. (k) Phosphorylation of p38 at Thr180 and Tyr182 in the GFP + subsets of CD27+CD28+ CD4+ T cells transduced with shCtrl, shTAB1 or shAMPKα (colors match key in l) and deprived of glucose for 18 h, assessed by phosphorylation-specific flow cytometry. (l) Summary of results obtained as in k, presented relative to those of cells transduced with shCtrl, set as 1. *P < 0.05, **P < 0.01 and ***P < 0.001 (paired Student’s t-test (b,d,f,h,j) or ANOVA for repeated measures with a Bonferroni post-test correction (l)). Data are representative of four experiments (a), two independent experiments (c) or three independent experiments (e,g,i,k) or are pooled from four experiments (b) or three independent experiments (d,f,h,j,l; error bars, s.e.m.). Count

© 2014 Nature America, Inc. All rights reserved.

–

p-p38(T180,Y182) MFI (fold)

5

–

CD28 +

f –

Count

4

CD27 +

e

Count

3

*

CD27+CD28+ – + CD27 CD28 CD27–CD28–

p-AMPKα(T172) MFI (fold)

2

5

d

p-p38(T180,Y182) MFI (fold)

CD27+CD28+ 207

CD27–CD28–

c

Count

Count

CD27–CD28– 656

CD27+CD28+

% p-γ-H2AX+ cells (fold)

b p-ATM(S1921) MFI (fold)

a

ATM inhibitor KU-55933 inhibited constitutive phosphorylation of AMPKα at Thr172 and of p38 at Thr180 and Tyr182 in CD27−CD28− CD4+ T cells (Fig. 3e–h), which indicated that ATM acted ‘upstream’ of both molecules in these cells. We also found a greater abundance of reactive oxidative species (ROS) in these cells (Supplementary Fig. 3a). Because genotoxic stress by ROS induces DNA damage 29 and because ROS are a known activator of AMPK30, we assessed the effect of the ROS scavenger superoxide dismutase on the activation of ATM, AMPK and p38. Such scavenging of ROS impaired the activation of all three molecules in CD27−CD28− CD4+ T cells (Supplementary Fig. 3b,c), which suggested that ROS may have triggered the endogenous DNA-damage response that led to the activation of p38 in senescent human T cells. Because AMPK acts as a sensor of low glucose concentrations in cells23,24, we investigated whether deprivation of glucose also induced the AMPK-dependent activation of p38 in nonsenescent T cells with intact DNA and low ROS production (Fig. 3a–d and Supplementary Fig. 3a). Depriving cells of glucose resulted in enhanced phosphorylation of p38 in undifferentiated CD27+CD28+ CD4+ T cells (Fig. 3i,j). Following glucose deprivation, the phosphorylation of p38 was inhibited more effectively in cells transduced with shAMPKα or shTAB1 than in cells transduced with shCtrl (Fig. 3k,l). Together these data supported a model in which signals from both genotoxic pathways and nutrient-deprivation pathways converged at the point of AMPK in T cells, which subsequently activated p38 through TAB1. AMPK recruits p38 to TAB1 and causes p38 autophosphorylation To further delineate the mechanistic details of the AMPK-TAB1-p38 pathway, we performed immunoprecipitation studies of primary human CD27+CD28+ CD4+ T cells activated by the AMPK agonist A-769662. We coimmunoprecipitated proteins from these cells with anti-TAB1, followed by immunoblot analysis, and found that aDVANCE ONLINE PUBLICATION  nature immunology

Articles a

b

c

© 2014 Nature America, Inc. All rights reserved.

agonist-driven activation of AMPK promoted the recruitment of p38 to TAB1 by 2.5- to 3-fold (Fig. 4a,b). Furthermore, the proteins coimmunoprecipitated with anti-TAB1 showed specific enrichment for phosphorylated AMPKα and p38 (Fig. 4a), which suggested the existence of an active AMPK-TAB1-p38 complex. Similarly, the proteins coimmunoprecipitated from glucose-deprived, undifferentiated CD27+CD28+ CD4+ T cells with anti-TAB1 also showed enrichment for phosphorylated AMPKα and p38 (data not shown). Notably, activated AMPKα and p38 were constitutively present among proteins coimmunoprecipitated from CD27−CD28− CD4+ T cells with anti-TAB1 (Fig. 4c); in these cells, the AMPK-TAB1-p38 pathway was constitutively active (as shown above). Because TAB1 mediates autophosphorylation of p38 (ref. 20), a modification that relies on the intrinsic activity of the kinase, and because p38 is not a direct substrate for AMPK, these data suggested that AMPK may have induced autophosphorylation of p38 by eliciting its binding to TAB1. We therefore used anti-p38 to immunoprecipitate p38 from AMPK agonist–activated CD27+CD28+ CD4+ T cells transduced with shCtrl, shTAB1 or shAMPKα and assessed the kinase activity of p38 in the absence of an added substrate in vitro to analyze autophosphorylation of p38, as described14,18. With this assay, we detected activated p38 in immunoprecipitates of CD27+CD28+ CD4+ T cells transduced with shCtrl, and this activity was enhanced by the addition of ATP in vitro (Fig. 4d). Conversely, the addition of ATP to proteins coimmunoprecipitated with anti-p38 from CD27+CD28+ CD4+ T cells transduced with shTAB1 or shAMPKα did not increase p38 activity in vitro (Fig. 4d). This indicated that the autophosphorylation of p38 required AMPK and TAB1 to form a complex together, with which p38 interacted in response to activation by AMPK. To confirm that the results presented above were indicative of autophosphorylation of p38 (and not an event mediated by an unrelated kinase coimmunoprecipitated with p38), we added the ATP competitor SB-203580, which acts as an inhibitor of p38, directly to the in vitro kinase reaction itself. SB-203580 prevented the enhanced phosphorylation of p38 in vitro in response to exogenous ATP nature immunology  aDVANCE ONLINE PUBLICATION

d

e

p-p38 (fold)

p-p38 (fold)

TAB1-bound p38 (fold)

AMPK act – + Figure 4  AMPK-triggered recruitment of p38 to TAB1 causes CD27 + – CD28 + – autophosphorylation of p38. (a) Immunoblot analysis (IB) of total IP: TAB1 –AMPK act IP: TAB1 IB: p38 p38, p38 phosphorylated at Thr180 and Tyr182, total AMPKα, +AMPK act IB: p-AMPKα IP: TAB1 AMPKα phosphorylated at Thr172 and total TAB1 in CD27 +CD28+ * 3.0 IB: p-p38 IP: TAB1 CD4+ T cells activated for 2 h with the vehicle dimethyl sulfoxide IB: p-p38 (DMSO) (−) or the AMPK agonist A-769662 (150 µM) (+), followed by IP: TAB1 2.0 (kDa) IB: AMPKα immunoprecipitation (IP) with anti-TAB1 (top group), and immunoblot 60 IP: TAB1 1.0 analysis of total p38 in whole-cell lysates (WCL) of the cells above, as IP: TAB1 IB: TAB1 IB: p-AMPKα input control (bottom). (b) Binding of p38 to TAB1 in cells as in a, 0 50 presented relative to that in cells treated with DMSO (−AMP act), set IP: TAB1 IB: TAB1 as 1. (c) Immunoblot analysis of phosphorylated AMPKα and p38 and total TAB1 (right margin, alternative TAB1 isoforms) in freshly WCL isolated CD27+CD28+ or CD27−CD28− CD4+ T cells, assessed after p38 immunoprecipitation with anti-TAB1. (d) Autophosphorylation of p38 IP: TAB1 IP: p38 in purified CD27+CD28+ CD4+ T cells transduced with shCtrl, shTAB1 4.0 4.0 or shAMPKα and, 96 h later, treated for 2 h with A-769662 (150 µM), ** * –ATP –ATP followed by immunoprecipitation with anti-p38 and incubation +ATP 3.0 3.0 +ATP for 30 min with (+ATP) or without (−ATP) ATP (200 µM), assessed by in vitro kinase assay based on an enzyme-linked immunosorbent 2.0 2.0 assay (absorbance at 450 nm) and presented relative to that in cells transduced with shCtrl and not treated with ATP, set as 1. 1.0 1.0 (e) Autophosphorylation of p38 in CD27 +CD28+ CD4+ T cells activated for 2 h with A-769662 (150 µM), followed by immunoprecipitation with 0 0 anti-TAB1 and incubation for 30 min with or without ATP (as in d) in the shCtrl shTAB1shAMPKα No p38i +p38i absence (No p38i) or presence (+p38i) of the p38 inhibitor SB-203580 (10 µM). *P < 0.01 and **P < 0.001 (paired Student’s t-test (b) or one-way ANOVA for repeated measures with a Bonferroni post-test correction (d,e)). Data are representative of four experiments with four individual donors (a) or two independent experiments (c; cells pooled from two individual donors for CD27−CD28− cells to obtain sufficient cells for assay) or are pooled from four independent experiments (b; error bars, s.e.m.) or three experiments with three individual donors (d,e; error bars, s.e.m. of triplicate wells).

(but not the baseline kinase activation in the cells) among proteins coimmunoprecipitated with anti-TAB1 from CD27+CD28+ CD4+ T cells after activation by AMPK (Fig. 4e). We obtained similar results with BIRB 796, a different inhibitor of p38 activity (data not shown). Thus, the p38 recruited to the AMPK-TAB1 complex autophosphorylated itself. Silencing of AMPKa or TAB1 restores proliferation Loss of telomerase activity in senescent T cells 3,8,15 limits their proliferative potential after antigenic challenge both in vitro and in vivo31,32 and is mediated in part by activation of p38, as shown by ‘rescue’ with a p38-selective small-molecule inhibitor9,33. To determine whether silencing of TAB1 or AMPKα restored these functions in senescent T cells, we transduced activated, purified CD27−CD28− CD4+ T cells with shAMPKα or shTAB1 (Supplementary Fig. 4a). Because the activity of telomerase in T lymphocytes specifically requires re-expression of its catalytic subunit hTERT upon activation34, we measured hTERT expression by both immunoblot analysis and quantitative PCR. The expression of hTERT was significantly higher in CD27−CD28− CD4+ T cells transduced with shAMPKα or shTAB1 than in cells transduced with shCtrl, as assessed 1 week after activation with anti-CD3 and recombinant human IL-2 (Fig. 5a); this led to increased telomerase activity in these cells at the 1-week time point, as assessed by a telomeric repeat–amplification protocol (Fig. 5b). To assess the in vivo effect of enhanced telomerase activity in CD27−CD28− CD4+ T cells in which expression of either TAB1 or AMPKα was silenced, we measured telomere length by fluorescence in situ hybridization coupled to flow cytometry, as described15. Four weeks after transduction of shCtrl, shTAB1 or shAMPKα (Supplementary Fig. 4a), knockdown of either TAB1 or AMPKα induced a significant extension of telomeric length (0.5–0.7 kilobases) in the GFP+ CD27−CD28− CD4+ T cells compared with that of cells transduced with shCtrl (Fig. 5c,d). Furthermore, transduction of shTAB1 or shAMPKα significantly enhanced the short-term 

2.0 1.0

4.0 3.0

*** **

2.0

shCtrl shTAB1 shAMPKα

1.0 0

0

c

shCtrl shTAB1 shAMPKα

+

GFP 6,590 5,995 4,900 2

3

4

0 10 10 10 10

Cy5-telomere

*** **

0.8 0.6 0.4

shCtrl shTAB1 shAMPKα

0.2

e

0

ND

*** **

5.0 4.0 3.0

shCtrl shTAB1 shAMPKα

2.0 1.0 0

Figure 5  Silencing of the expression of AMPKα or Reactivation TAB1 restores telomerase and proliferation in shCtrl shCtrl shCtrl 1.5 senescent T cells. (a) Immunoblot analysis of shTAB1 4.0 ** 100 GFP+ ** shTAB1 shTAB1 ** shAMPKα hTERT (left) and quantitative PCR analysis of shAMPKα 80 3,085 1.0 shAMPKα 3.0 ** * hTERT mRNA (right) in CD27−CD28− CD4+ T cells 60 4,105 * 2.0 6,396 transduced with shCtrl, shTAB1 or shAMPKα and 40 0.5 * 1.0 analyzed 96 h later (experimental design, 20 Supplementary Fig. 4a); results at right were 0 0 0 10 20 30 40 0 102 103 104 105 normalized to those of mRNA from the housekeeping CellTrace Time (d) gene GAPDH and are presented relative to those of Violet cells transduced with shCtrl, set as 1. (b) Telomerase − − + activity in CD27 CD28 CD4 T cells transduced and cultured as in a, assessed by a telomeric repeat–amplification protocol assay (as absorbance at 450 nm) and presented relative to that in cells transduced with shCtrl, set as 1. (c) Abundance of telomeres in the GFP+ population of CD27−CD28− CD4+ T cells transduced as in a, assessed by fluorescent in situ hybridization coupled to flow cytometry with a telomere-specific probe conjugated to labeled with indodicarbocyanine (Cy5 telomere). (d) Change in telomere length (∆ bp) in CD27−CD28− CD4+ T cells transduced as in a and subjected to two rounds of activation with anti-CD3 and IL-2 during culture and assessed after 30 d (experimental design, Supplementary Fig. 4a), as in c and presented relative to that of cells transduced with shCtrl, set as 0. (e,f) Proliferative activity of CD27−CD28− CD4+ T cells transduced and cultured as in a, analyzed by incorporation of [3H]thymidine (e) or as dilution of the fluorescent dye CellTrace Violet (f). (g) Summary of results in f, presented relative to those of cells transduced with shCtrl, set as 1. (h) Replicative lifespan of CD27−CD28− CD4+ T cells activated with anti-CD3 and IL-2 and transduced as in a (experimental design, Supplementary Fig. 4a), then cultured for 30 d with reactivation (downward arrows) every 10 d with anti-CD3 and IL-2, assessed as cumulative population doubling (PD). *P < 0.01 and **P < 0.001 (one-way ANOVA for repeated measures with a Bonferroni post-test correction). Data are representative of two independent experiments (a, left) or three experiments (c,f) or are pooled from three experiments with three individual donors (a (right), b,d,e,g,h; error bars, s.e.m. (d,g) or s.e.m. of triplicate wells (b,e,h)).

1.0 0.5 0

* *

6.0 4.0 2.0 0

c

α-CD3 + α-CD28

* * *

* * *

*

AMPK act Ctrl p38 inhib AMPK act + p38 inhib AMPK act

p-AMPKα 1.5 Lck Zap70 SLP-76

10

20 Time (d)

30

d

1.0 0.5 0

**

NS

SLP-76 MFI (fold)

NS

To further delineate the role of p38 activation in the proliferation of human senescent T cells, we investigated the effect of p38 on the cellcycle machinery. Because blockade of p38 increased the rate of DNA synthesis (an event that characterizes actively cycling cells in S phase), we activated CD27−CD28− CD4+ T cells with anti-CD3 and recombinant human IL-2 and investigated the expression of several factors that regulate the transition from G1 phase to S phase. Inhibition of p38 induced coordinated downmodulation of the cell-cycle inhibitors p27 and p21, upregulation of the D cyclins, enhanced inactivation of the transcription factor Rb (by lowering the abundance of active, underphosphorylated Rb) and dampened activation of the tumor suppressor p53 (Supplementary Fig. 5 and data not shown), which are all well-established events that promote the G1-to-S transition35.

Lat MFI (fold)

**

b 8.0 Cumulative PD

Telomerase activity (fold)

1.5

Ctrl AMPK act AMPK act + p38 inhib p38 inhib

h

Cumulative PD

Count

proliferation of CD27−CD28− CD4+ T cells following activation for 7 d with anti-CD3 and recombinant human IL-2, as assessed by either incorporation of thymidine (to assess new DNA synthesis) or dyedilution analysis (Fig. 5e–g), as well as their long-term expansion capacity (after 30 d in culture) (Fig. 5h), compared with that of cells transduced with shCtrl. As a control, inhibition of p38 by p38-specific shRNA or by blockade with either of two different specific inhibitors (BIRB 796 or SB203580) resulted in enhanced telomerase activity and short-term and long-term proliferation of senescent T cells (Supplementary Fig. 4b–g and data not shown), which confirmed the results of published studies9. Thus, the defective proliferation and telomerase activity of human senescent T cells was reversed by the inhibition of AMPK and TAB1 upstream of p38.

a

g

CellTrace Violet MFI (fold)

f

© 2014 Nature America, Inc. All rights reserved.

5

d

3

3.0

b

[ H] thymidine uptake (fold)

shCtrl shTAB1 shAMPKα

Telomere length (∆ bp)

**

Count

hTERT

4.0

Telomerase activity (fold)

sh C t sh rl TA sh B1 AM PK α

a

hTERT mRNA (fold)

Articles

1.5 1.0

**

Ctrl AMPK act AMPK act + p38 inhib p38 inhib

NS

0.5 0

Figure 6  Activation of p38 by a specific AMPK agonist reproduces senescent features in nonsenescent T cells. (a) Telomerase activity in CD27+CD28+ CD4+ T cells activated with anti-CD3 and anti-CD28 and cultured for 3 d with DMSO (Ctrl), the AMPK activator A-769662 (150 µM) alone (AMPK act) or the p38 inhibitor BIRB 796 (500 nM) alone (p38 inhib) or both A-769662 and BIRB 796 together (AMPK act + p38 inhib), assessed as in Figure 5b and presented relative to that in cells cultured with DMSO, set as 1. (b) Replicative lifespan of CD27+CD28+ CD4+ T cells activated and cultured as in a (assessed and presented as in Fig. 5h). (c) Immunoblot analysis of AMPKα phosphorylated at Thr172 (as a pharmacological activation control) and of total Lck, Zap70 and SLP-76 in CD27+CD28+ CD4+ T cells activated as in a and cultured for 3 d with DMSO (far left lane) or 100 or 150 µM (wedge) A-769662. (d) Expression of Lat and SLP-76 in CD27+CD28+ CD4+ T cells cultured for 3 d as in a, assessed by flow cytometry. *P < 0.01 and **P < 0.001 (one-way ANOVA for repeated measures with a Bonferroni post-test correction). Data are pooled from three experiments with three individual donors (a,b; error bars, s.e.m of triplicate wells) or three independent experiments (d; error bars, s.e.m.) or are representative of four independent experiments (c).



aDVANCE ONLINE PUBLICATION  nature immunology

Articles

© 2014 Nature America, Inc. All rights reserved.

We also obtained similar results when we ‘tuned’ p38 activity by silencing the expression of either AMPKα or TAB1 via transduction of shRNA in CD27−CD28− CD4+ T cells (data not shown), which indicated that the activation of p38 induced by AMPK-TAB1 signaling inhibited cell-cycle progression at the G1-to-S checkpoint. These data suggested that p38 activated by AMPK-TAB1 inhibited T cell proliferation by direct inhibition of cyclins, induction of cyclin inhibitors and downstream activation of p53. p38 activation induces senescence characteristics in nonsenescent T cells To further link the AMPK-TAB1 pathway to the spontaneous activation of p38, we activated nonsenescent CD27+CD28+ CD4+ T cells with the AMPK agonist A-769662 in the presence or absence of the selective p38 inhibitor BIRB 796. A-769662 inhibited the telomerase activity, proliferation and population expansion of CD27+CD28+ CD4+ T cells activated with anti-CD3 and anti-CD28, while the addition of BIRB 796 restored these functions (Fig. 6a,b and data not shown). We obtained similar results in CD27+CD28+ CD4+ T cells in which expression of p38 was silenced via shRNA (Supplementary Fig. 6a,b). All these observations were reproduced in glucosedeprived CD27+CD28+ CD4+ T cells, and they were counteracted in part by inhibition of p38 (Supplementary Fig. 6c–e). However, cells did not survive the prolonged absence of glucose after 5 d (ref. 36) (data not shown). In addition, activation of AMPK by the agonist A-769662 induced a dose-dependent loss of TCR signaling molecules in CD27+CD28+ CD4+ T cells activated with anti-CD3 and anti-CD28 (Fig. 6c and data not shown), and this was prevented in part by inhibition of p38 (Fig. 6d). Together these data showed that AMPK inhibited the proliferation of human T cells and telomerase via activation of p38, a process that required its allosteric binding to TAB1. Furthermore, nonsenescent T cells activated p38 and downregulated the proliferation and telomerase activity when AMPK was activated chemically or by glucose deprivation. DISCUSSION In this study we investigated the mechanism by which p38 is activated in senescent human CD4+ T cells. In T cells, activation of p38 has been documented as being induced either by the canonical MAPK cascade or the alternative TCR signaling pathway in response to environmental stress or TCR activation, respectively10,11,19. Here we found that senescent human CD4+ T cells lacked essential ‘upstream’ components of both the canonical pathway and the alternative pathway and instead activated p38 by a previously unrecognized mechanism dependent on AMPK-TAB1. This occurred in vivo in response to endogenous signaling of a DNA-damage response by the apical kinase ATM, in response to genotoxic stress. However, glucose deprivation also triggered the energy gauge AMPK and led to activation of p38 via the scaffold protein TAB1 in nonsenescent CD4+ T cells that had no DNA damage. Therefore, signaling through this AMPK-TAB1 pathway for p38 activation was induced in undifferentiated T cells by low nutrient availability or in senescent T cells by DNA damage. This led to the downregulation of key signaling molecules of the TCR complex and inhibition of both telomerase activity and proliferation. The scaffold molecule TAB1 was initially identified as a TAK1 binding partner37. Subsequently, it was found that TAB1 also induces allosteric autophosphorylation of p38 in the HEK293 cell line independently of TAK1 (refs. 20,21). However, the biological relevance of TAB1-dependent activation of p38 has remained uncertain. In a model of mouse ischemic cardiomyocytes, both TAB1 and AMPK were reported to act upstream of p38 (ref. 22), but two subsequent nature immunology  aDVANCE ONLINE PUBLICATION

studies with the same experimental system have challenged that earlier report38,39. The reason for the evolution of different modes for the activation of p38 remains unknown, but our findings suggest that they may enable the same signaling molecule to exert opposite effects on T cell function. We confirmed the published observation that ‘alternative’ DLG1-mediated activation of p38 downstream of the TCR supports proliferative activity in nonsenescent CD4+ T cells17,18; however, p38 activated via AMPK-TAB1 inhibited the proliferation of both senescent human CD4+ T cells and nonsenescent human CD4+ T cells. Furthermore, a published report has shown that p38 activated by a ‘canonical’ pathway and p38 activated by an ‘alternative’ pathway counter-regulate each other in controlling the effector functions of T cells by exerting different substrate specificity40. Also, we note that the mapped TAB1-p38 interaction site41 impedes access to the ‘alternative’ regulatory phosphorylation site of p38 at Tyr323, which suggests that scaffold molecules may also be involved in the determination of downstream substrate specificity. One unexpected observation was that key components of the TCR signaling machinery, such as Lck, Zap70, DLG1, Lat and SLP-76, were naturally lost in senescent CD4+ T cells, which also lose expression of CD27 and CD28 (refs. 8,35). Collective loss of these signaling molecules would exacerbate the inhibition of proliferation of these cells by decreasing their capacity for TCR activation. Of note, blocking p38 in highly differentiated T cells did not restore expression of the TCR signalosome (data not shown), which suggested that other p38independent mechanisms were involved. At present, p38 inhibitors have been developed for use in treating diseases with an inflammatory etiology42. When tested in clinical trials, those compounds have shown a low therapeutic index and engagement of escape or bypass mechanisms; the latter effect is possibly due to the regulation of diverse cellular functions by p38 activation 42. Because inhibition of p38 activity by silencing of the expression of either AMPK or TAB1 in vitro enhanced the proliferation of human T cells, we suggest the possibility of specific inhibition of this pathway as a strategy for enhancing the function of cells of the immune system in vivo. The crystal structure of the docking site for TAB1 on p38 has been reported41 and is conserved in both the 60-kDa and 50-kDa variants of TAB1 (ref. 21). Targeting this restricted, MKK-independent interaction may provide a selective way for dampening p38 activity, as it would spare ubiquitous activation by the MAPK cascade10. One caveat is that the enhancement of telomerase and proliferation in highly differentiated T cells that have evidence of DNA damage may be tumorigenic35. The risk versus possible benefit of selective therapeutic ‘tuning’ of telomerase in human T lymphocytes in vivo remains to be determined. We have therefore described an ‘intrasensory’ pathway in T cells that integrates nutrient deprivation and DNA-damage–sensing signals to activate p38. This may be the prototype of diverse convergent signaling pathways that link senescence, immunomodulation, metabolism and cancer. It is possible that AMPK-TAB1 signaling may be involved with the activation of p38 that is associated in the senescence of various cell types, such as fibroblasts and stem cells from skeletal muscle, in both mice and humans43,44; this will require further study. Methods Methods and any associated references are available in the online version of the paper. Note: Any Supplementary Information and Source Data files are available in the online version of the paper.



Articles Acknowledgments We thank S.S. Marelli for discussions. Supported by the Medical Research Council (A.L.), the Biotechnology and Biological Science Research Council (BB/J006750/1 to A.N.A. and S.M.H.) and the Instituto de Salud Carlos III, Spain (D.E.). AUTHOR CONTRIBUTIONS A.L. conceived of and performed the study, analyzed and interpreted the data and wrote the paper; S.M.H. performed experiments; D.E. provided lentiviral tools; A.N.A. wrote the paper and provided overall direction; and all authors read and approved the final version of the manuscript. COMPETING FINANCIAL INTERESTS The authors declare no competing financial interests.

© 2014 Nature America, Inc. All rights reserved.

Reprints and permissions information is available online at http://www.nature.com/ reprints/index.html. 1. Huang, Y. & Wange, R.L. T cell receptor signaling: beyond complex complexes. J. Biol. Chem. 279, 28827–28830 (2004). 2. Akbar, A.N., Beverley, P.C. & Salmon, M. Will telomere erosion lead to a loss of T-cell memory? Nat. Rev. Immunol. 4, 737–743 (2004). 3. Fletcher, J.M. et al. Cytomegalovirus-specific CD4+ T cells in healthy carriers are continuously driven to replicative exhaustion. J. Immunol. 175, 8218–8225 (2005). 4. Henson, S.M. et al. KLRG1 signaling induces defective Akt (ser473) phosphorylation and proliferative dysfunction of highly differentiated CD8+ T cells. Blood 113, 6619–6628 (2009). 5. Libri, V. et al. Cytomegalovirus infection induces the accumulation of short-lived, multifunctional CD4+CD45RA+CD27+ T cells: the potential involvement of interleukin-7 in this process. Immunology 132, 326–339 (2011). 6. Warrington, K.J., Vallejo, A.N., Weyand, C.M. & Goronzy, J.J. CD28 loss in senescent CD4+ T cells: reversal by interleukin-12 stimulation. Blood 101, 3543–3549 (2003). 7. Vallejo, A.N. CD28 extinction in human T cells: altered functions and the program of T-cell senescence. Immunol. Rev. 205, 158–169 (2005). 8. Weng, N.P., Akbar, A.N. & Goronzy, J. CD28− T cells: their role in the age-associated decline of immune function. Trends Immunol. 30, 306–312 (2009). 9. Di Mitri, D. et al. Reversible senescence in human CD4+CD45RA+CD27− memory T cells. J. Immunol. 187, 2093–2100 (2011). 10. Ashwell, J.D. The many paths to p38 mitogen-activated protein kinase activation in the immune system. Nat. Rev. Immunol. 6, 532–540 (2006). 11. Rudd, C.E. MAPK p38: alternative and nonstressful in T cells. Nat. Immunol. 6, 368–370 (2005). 12. Chang, L. & Karin, M. Mammalian MAP kinase signalling cascades. Nature 410, 37–40 (2001). 13. Dodeller, F. & Schulze-Koops, H. The p38 mitogen-activated protein kinase signaling cascade in CD4 T cells. Arthritis Res. Ther. 8, 205 (2006). 14. Salvador, J.M. et al. Alternative p38 activation pathway mediated by T cell receptorproximal tyrosine kinases. Nat. Immunol. 6, 390–395 (2005). 15. Plunkett, F.J. et al. The loss of telomerase activity in highly differentiated CD8+CD28−CD27− T cells is associated with decreased Akt (Ser473) phosphorylation. J. Immunol. 178, 7710–7719 (2007). 16. Yan, M. et al. Activation of stress-activated protein kinase by MEKK1 phosphorylation of its activator SEK1. Nature 372, 798–800 (1994). 17. Jirmanova, L., Sarma, D.N., Jankovic, D., Mittelstadt, P.R. & Ashwell, J.D. Genetic disruption of p38α Tyr323 phosphorylation prevents T-cell receptormediated p38α activation and impairs interferon-γ production. Blood 113, 2229–2237 (2009). 18. Round, J.L. et al. Scaffold protein Dlgh1 coordinates alternative p38 kinase activation, directing T cell receptor signals toward NFAT but not NF-κB transcription factors. Nat. Immunol. 8, 154–161 (2007).



19. Rincon, M. & Davis, R.J. Choreography of MAGUKs during T cell activation. Nat. Immunol. 8, 126–127 (2007). 20. Ge, B. et al. MAPKK-independent activation of p38α mediated by TAB1-dependent autophosphorylation of p38α. Science 295, 1291–1294 (2002). 21. Ge, B. et al. TAB1β (transforming growth factor-β-activated protein kinase 1-binding protein 1β), a novel splicing variant of TAB1 that interacts with p38α but not TAK1. J. Biol. Chem. 278, 2286–2293 (2003). 22. Li, J., Miller, E.J., Ninomiya-Tsuji, J., Russell, R.R. III & Young, L.H. AMP-activated protein kinase activates p38 mitogen-activated protein kinase by increasing recruitment of p38 MAPK to TAB1 in the ischemic heart. Circ. Res. 97, 872–879 (2005). 23. Hardie, D.G. AMP-activated protein kinase: an energy sensor that regulates all aspects of cell function. Genes Dev. 25, 1895–1908 (2011). 24. Rolf, J. et al. AMPKα1: a glucose sensor that controls CD8 T-cell memory. Eur. J. Immunol. 43, 889–896 (2013). 25. Finlay, D. & Cantrell, D.A. Metabolism, migration and memory in cytotoxic T cells. Nat. Rev. Immunol. 11, 109–117 (2011). 26. Sanders, M.J. et al. Defining the mechanism of activation of AMP-activated protein kinase by the small molecule A-769662, a member of the thienopyridone family. J. Biol. Chem. 282, 32539–32548 (2007). 27. d’Adda di Fagagna, F. Living on a break: cellular senescence as a DNA-damage response. Nat. Rev. Cancer 8, 512–522 (2008). 28. Mondal, A.M. et al. p53 isoforms regulate aging- and tumor-associated replicative senescence in T lymphocytes. J. Clin. Invest. 123, 5247–5257 (2013). 29. Passos, J.F. & Von Zglinicki, T. Oxygen free radicals in cell senescence: are they signal transducers? Free Radic. Res. 40, 1277–1283 (2006). 30. Blattler, S.M., Rencurel, F., Kaufmann, M.R. & Meyer, U.A. In the regulation of cytochrome P450 genes, phenobarbital targets LKB1 for necessary activation of AMP-activated protein kinase. Proc. Natl. Acad. Sci. USA 104, 1045–1050 (2007). 31. Hodes, R.J., Hathcock, K.S. & Weng, N.P. Telomeres in T and B cells. Nat. Rev. Immunol. 2, 699–706 (2002). 32. Reed, J.R. et al. Telomere erosion in memory T cells induced by telomerase inhibition at the site of antigenic challenge in vivo. J. Exp. Med. 199, 1433–1443 (2004). 33. Lanna, A. et al. IFN-alpha inhibits telomerase in human CD8+ T Cells by both hTERT downregulation and Induction of p38 MAPK signalling. J. Immunol. (2013). 34. Akbar, A.N. & Vukmanovic-Stejic, M. Telomerase in T lymphocytes: use it and lose it? J. Immunol. 178, 6689–6694 (2007). 35. Akbar, A.N. & Henson, S.M. Are senescence and exhaustion intertwined or unrelated processes that compromise immunity? Nat. Rev. Immunol. 11, 289–295 (2011). 36. Chang, C.H. et al. Posttranscriptional control of T cell effector function by aerobic glycolysis. Cell 153, 1239–1251 (2013). 37. Shibuya, H. et al. TAB1: an activator of the TAK1 MAPKKK in TGF-β signal transduction. Science 272, 1179–1182 (1996). 38. Jacquet, S. et al. The relationship between p38 mitogen-activated protein kinase and AMP-activated protein kinase during myocardial ischemia. Cardiovasc. Res. 76, 465–472 (2007). 39. Jaswal, J.S., Gandhi, M., Finegan, B.A., Dyck, J.R. & Clanachan, A.S. p38 mitogenactivated protein kinase mediates adenosine-induced alterations in myocardial glucose utilization via 5′-AMP-activated protein kinase. Am. J. Physiol. Heart Circ. Physiol. 292, H1978–H1985 (2007). 40. Alam, S.M. et al. Counter-regulation of T cell effector function by differentially activated p38. J. Exp. Med. 211, 1257–1270 (2014). 41. De Nicola, G.F. et al. Mechanism and consequence of the autoactivation of p38α mitogen-activated protein kinase promoted by TAB1. Nat. Struct. Mol. Biol. 20, 1182–1190 (2013). 42. Lee, J.C. et al. Inhibition of p38 MAP kinase as a therapeutic strategy. Immunopharmacology 47, 185–201 (2000). 43. Tivey, H.S., Brook, A.J., Rokicki, M.J., Kipling, D. & Davis, T. p38 (MAPK) stress signalling in replicative senescence in fibroblasts from progeroid and genomic instability syndromes. Biogerontology 14, 47–62 (2013). 44. Bernet, J.D. et al. p38 MAPK signaling underlies a cell-autonomous loss of stem cell self-renewal in skeletal muscle of aged mice. Nat. Med. 20, 265–271 (2014).

aDVANCE ONLINE PUBLICATION  nature immunology

ONLINE METHODS

© 2014 Nature America, Inc. All rights reserved.

Cells. Heparinized peripheral blood samples were obtained from healthy volunteers (40–65 years of age; median age, 52 years; male, 55%, and female, 45%; n = 57). All samples were obtained with the approval of the Ethical Committee of Royal Free and University College Medical School and voluntary informed consent was obtained in accordance with the Declaration of Helsinki. Donors did not have any comorbidity, were not on any immunosuppressive drugs and retained physical mobility and lifestyle independence. The CD4+ T cell subsets defined by expression of CD27 and CD28 were isolated from peripheral blood mononuclear cells as described15. Immunoblot analysis. Purified CD4+ T cells were sorted ex vivo immediately after separation into subsets defined by expression of CD27 and CD28. Alternatively, purified human highly differentiated CD27−CD28− CD4+ T cells were harvested 4 d after transduction (described below). Lysates from 2 × 106 cells were obtained and analyzed as described33. Membranes were probed with anti-p38 (9212), anti-MKK3 (5674), anti-MKK6 (9264), antibody to MKK3 and/or MKK6 phosphorylated at Ser189 and Ser207 (22A8), anti-Lck (2752), anti-Zap70 (99F2), anti-TAB1 (C25E9), anti-TAK1 (5206), anti-TRAF6 (D21G3), antibody to AMPKα phosphorylated at Thr172 (40H9), anti-AMPKα (D5A2), antibody to γ-H2AX phosphorylated at Ser139 (20E3), anti-Lat (9166), anti-SLP-76 (4958), anti-PLC-γ 1 (2822) and antiGAPDH (14C10; all from Cell Signaling); antibody to p38 phosphorylated at Tyr323 (PP3411; ECM-Biosciences); and anti-hTERT (H-231) and anti-DLG1 (SAP-972D11; both from Santa Cruz Biotechnology). All immunoblots were developed with a ECL Prime Western Blotting Detection Kit according to the manufacturer’s instructions (GE Healthcare). Immunoprecipitation. For coimmunoprecipitation analysis, cell lysates were prepared with ice-cold HNGT buffer (50 mM HEPES, pH 7.5, 150 mM EDTA, 10 mM sodium pyrophosphate, 100 mM sodium orthovanadate, 100 mM sodium fluoride, 10 mg/ml aprotinin, 10 mg/ml leupeptin and 1 mM phenylmethylsulfonyl fluoride). Lysates from 20 × 106 purified primary human CD27+CD28+ CD4+ T cells were incubated overnight at 4 °C on a rotary shaker with antibody to TAB1 (C25E9; Cell Signaling) or p38 (9212; Cell Signaling). For the minor CD28−CD27− CD4+ T cell subset, cells from two separate donors were pooled to obtain sufficient cells for analysis. Extracts were then incubated for 3 h at 4 °C with protein A–protein G–conjugated agarose beads (Santa Cruz Biotechnology). Samples were washed and then were analyzed by immunoblot. Coimmunoprecipitated proteins were detected with mouse anti-rabbit IgG (conformation-specific antibody; L27A9; Cell Signaling) or mouse antibody to rabbit IgG light chain (L57A3; Cell Signaling), followed by a secondary anti-mouse IgG antibody (7076; Cell Signaling) and ECL Prime Western detection kit (GE Healthcare). In vitro kinase assay. Proteins immunoprecipitated with anti-TAB1 (C25E9; Cell Signaling) or anti-p38 (9212; Cell Signaling) were obtained as described above. Samples were washed twice in lysis buffer and twice in kinase buffer (both from Cell Signaling). Kinase reactions were incubated for 30 min at 30 °C in a the presence or absence of 200 µM ATP (Cell Signaling). In some experiments, the ATP competitor and allosteric p38 inhibitor SB-203580 (10 µM) was added directly to the in vitro kinase reaction. Total p38 and p38 phosphorylated at Thr180 and Tyr182 were detected by enzyme-linked immunosorbent assay with a PhosphoTracer ELISA Kit according to the manufacturer’s instructions (Abcam). In vitro kinase assays are presented as proportional to the absorbance at 450 nm. Detection of intracellular ROS levels and calcium flux. After cell surfaces were staining with anti-CD27 (337169; BD Biosciences) and anti-CD28 (348040; BD Biosciences), human CD4+ T cells were incubated with dihydroethidium or the Ca2+ indicator Fluo-4-AM, respectively, for analysis of intracellular levelsof ROS or calcium according to the manufacturer’s instructions (Invitrogen Molecular Probes). Cells were analyzed by flow cytometry with an LSR Fortessa (BD Biosciences). Lentiviral vector design. We generated the pHIV1-SIREN-GFP system used for knockdown of gene expression as follows. The lentivector pSIN-GFP45

doi:10.1038/ni.2981 

was digested with EcoRI and BamHI, and the promoter of the gene encoding phosphoglycerate kinase (PGK) was amplified by polymerase chain reaction (PCR) for introduction of EcoRI-BamHI restriction sites. Then, the SFFV promoter from the plasmid pSIN-GFP was replaced with the promoter of the gene encoding PGK, cloned by blunt-end ligation, to generate the plasmid pSIN-PGKp-GFP. This vector was used as a backbone. A cassette containing the promoter of the gene encoding the human small nuclear RNA U6 was cloned in this backbone downstream the cPPT sequence between ClaI-EcoRI restriction sites (extended vector map, Supplementary Fig. 2a). A BamHI site was introduced downstream of the U6 promoter to clone the shRNA of interest between BamHI and EcoRI restriction sites. The U6-shRNA cassette was placed upstream the PGK promoter and GFP. All constructs were engineered by standard cloning techniques. The following shRNA sequences were used: CCTAAGGTTAAGTCGCCCTCG (shCtrl), GGCGGTCCTTCTCAACAACAAG (shTAB1) and ATGATGTC AGATGGTGAATTT (shAMPKα), as described21,46,47. The shRNA sequence used for knockdown of TAB1 targets a region common to both the 60-kDa isoform and the 50 kDa isoform of TAB1. The shRNA target used for knockdown of p38 (shp38) was GTACTTCCTGTGTACTCTTTA. Lentiviral particles (from lentivirus pseudotyped with glycoprotein G of vesicular stomatitis virus) were produced by transient cotransfection of three plasmids into HEK293 cells as described38 and the particles were concentrated 100-fold by ultracentrifugation through a 20% sucrose cushion in phosphate-buffered saline (PBS), following resuspension in PBS containing 10% glycerol. Lentiviral preparations were then ‘titrated’ for GFP expression by serial dilution in HEK293 cells by flow cytometry. Cell cultures and lentiviral transduction of primary human T lymphocytes. Cells were cultured at 37 °C in a humidified 5% CO2 incubator in RPMI-1640 medium supplemented with 10% heat-inactivated FCS, 100 U/ml penicillin, 100 mg/ml streptomycin, 50 µg/ml gentamicin and 2 mM l-glutamine (all from Invitrogen). Purified, highly differentiated CD27−CD28− human CD4+ T cells were activated in the presence of plate-bound anti-CD3 (0.5 µg/ml; purified OKT3; 86022706; Sigma-Aldrich) plus recombinant human IL-2 (10 ng/ml; R&D Systems) and then, at 48 and 72 h after activation, were transduced with pHIV1-Siren lentiviral particles (multiplicity of infection, 10). For long-term cultures, transduced cells were reactivated every 10 d. Relatively undifferentiated CD27+CD28+ CD4+ T cells were cultured and transduced as their CD27−CD28− counterparts were, but cells were activated by plate-bound anti-CD3 (0.5 µg/ml; 86022706; Sigma-Aldrich) plus anti-CD28 (0.5 µg/ml; 37407; MAB342; R&D Systems). Quantitative PCR (real-time analysis). At 4 d after transduction of lentiviral vector, samples from 5 × 105 viable highly differentiated CD27−CD28− CD4+ T cells were resuspended in TRIzol (Ambion). For cDNA synthesis, RNA was retro-transcribed with M-MuLV Reverse Transcriptase (New England Biolabs) and random primers. The expression of p38 (Hs00176247_m1), AMPKα (Hs01562315_m1), TAB1 (Hs00196143_m1) and hTERT (Hs00972656_m1) was detected by quantitative PCR with Taqman probes (Applied Biosystems identifiers in parentheses). Transcript expression was normalized to that of the housekeeping gene GAPDH by the change-in-cycling-threshold method (∆∆Ct) for the calculation of values supplied by Applied Biosystems. Signaling studies. Cells were fixed for 10 min at 37 °C with warm Cytofix Buffer (BD Biosciences). Cells were then permeabilized for 30 min at 4 °C with ice-cold Perm Buffer III (BD Biosciences) and were incubated for 30 min at room temperature with phycoerythrin-conjugated antibody to p38 phosphorylated at Thr180 and Tyr182 (612565), phycoerythrin-conjugated antibody to γ-H2AX phosphorylated at Ser139 (562377), phycoerythrin-conjugated antibody to underphosphorylated Rb (550502) and antibody to Rb phosphorylated at Ser807 and Ser811 (558549; all from BD Biosciences); and antibody to AMPKα phosphorylated at Thr172 (40H9), antibody to ATM phosphorylated at Ser1921 (D25E5), antibody to ATF2 phosphorylated at Thr71 (9221), antibody to p53 phosphorylated at Ser46 (2521), anti-cyclin D1 (DCS6) and anti-p27 (D69C12; all from Cell Signaling). Primary unconjugated antibodies were subsequently probed for 30 min at room temperature in the dark with a secondary antibody, either phycoerythrin-conjugated goat anti–rabbit IgG

nature immunology

(P-2771MP; Life Technologies) or fluorescein isothiocyanate-conjugated goat anti–rabbit IgG (65-6111; Life Technologies). Cells were washed in Stain Buffer (BD Pharmingen) and were immediately analyzed with an LSR Fortessa (BD Biosciences). In some experiments, cells were preincubated for 60 min with the AMPK agonist A-769662 (150 µM; Tocris Bioscience), the ATM inhibitor KU-55933 (10 µM; Calbiochem) or PMA (phorbol 12-myristate 13-acetate; 20 ng/ml; Sigma) before being fixed. Data were analyzed with FlowJo software (TreeStar). For signaling studies with transduced cells, events were gated on the GFP+ compartment.

© 2014 Nature America, Inc. All rights reserved.

Measurement of telomerase activity. Telomerase activity was assessed with a TeloTAGGG telomerase ELISA kit according to the manufacturer’s instructions (Roche) and extracts of 2 × 103 viable transduced CD27−CD28− CD4+ T cells. CD27−CD28− CD4+ T cells were counted by Trypan blue exclusion (Sigma) and analysis of the proliferation marker Ki67 as described33. Telomerase activity was determined as absorbance at 450 nm. Assessment of telomere length. CD27−CD28− CD4+ T cells transduced with shCtrl, shAMPK or shTAB1 were activated by culture for 30 d with anti-CD3 (0.5 µg/ml; purified OKT3; 86022706, Sigma-Aldrich) plus recombinant human IL-2 (10 ng/ml; R&D Systems), then their telomere length was measured by a modified version of a published fluorescence in situ hybridization–flow cytometry method15. The length of telomeres in transduced GFP+ CD27−CD28− CD4+ T cells is presented as the difference in base pairs (∆) for GFP+ cells in which AMPKα or TAB1 was silenced over a control GFP+ CD27−CD28− CD4+ population transduced with shCtrl. Proliferation assays. At 4 d after transduction of lentiviral vector, the proliferation of activated CD27−CD28− or CD27+CD28+ CD4+ T cells was assessed

nature immunology

by overnight probing with [3H]thymidine. Short-term proliferative activity is presented as incorporation of [3H]thymidine (in c.p.m.). Alternatively, immediately after being sorted, purified, highly differentiated CD27−CD28− human CD4+ T cells were stained with a CellTrace Violet Cell Proliferation Kit and were analyzed by flow cytometry 96 h after transduction (day 7), according to the manufacturer’s instructions (Invitrogen). In some experiments with early differentiated CD27+CD28+ CD4+ T cells, proliferation was assessed by staining for the cell cycle–related nuclear antigen Ki67 after 48 h of activation with anti-CD3 (0.5 µg/ml; 86022706, Sigma-Aldrich) plus anti-CD28 (0.5 µg/ml; 37407; MAB342; R&D Systems), as described32. Population doubling was calculated as follows: log10 (number of cells counted after population expansion) − log10 (number of cells seeded) / log10 2. Statistical analysis. Graphpad Prism was used for statistical analysis. For pairwise comparisons, a paired Student’s t-test was used. For three matched groups, one-way ANOVA for repeated measures using a Bonferroni post-test correction was used. Differences with a P value of