the lysine deacetylase sirtuin 1 modulates the ... · the lysine deacetylase sirtuin 1 modulates...

SC I ENCE S I GNAL ING | R E S EARCH ART I C L E

IMMUNOLOGY

1National Centre for Biological Sciences, Bengaluru, Karnataka 560065, India.2Manipal University, Manipal, Karnataka, India. 3Institute for Stem Cell Biologyand Regenerative Medicine, Bengaluru, Karnataka 560065, India.*These authors contributed equally to this work.†Present address: Department of Biotechnology, Indian Institute of Technology,Chennai, TN, India.‡Corresponding author. Email: [email protected]

Marcel et al., Sci. Signal. 10, eaah4679 (2017) 4 April 2017

2017 © The Authors,

some rights reserved;

exclusive licensee

American Association

for the Advancement

of Science.

Dow

nloaded

The lysine deacetylase Sirtuin 1 modulates thelocalization and function of the Notch1 receptor inregulatory T cellsNimi Marcel,1,2* Lakshmi R. Perumalsamy,1*† Sanjay K. Shukla,1,2* Apurva Sarin1,3‡

The ability to tune cellular functions in response to nutrient availability has important consequences for im-mune homeostasis. The activity of the receptor Notch in regulatory T (Treg) cells, which suppress the functionsof effector T cells, is indispensable for Treg cell survival under conditions of diminished nutrient supply. Anti-apoptotic signaling induced by the Notch1 intracellular domain (NIC) originates from the cytoplasm and is spa-tially decoupled from the nuclear, largely transcriptional functions of NIC. We showed that Sirtuin 1 (Sirt1), which isan NAD+ (nicotinamide adenine dinucleotide)–dependent lysine deacetylase that inhibits NIC-dependent genetranscription, stabilized NIC proximal to the plasma membrane to promote the survival and function of activatedTreg cells. Sirt1 was required for NIC-dependent protection from apoptosis in cell lines but not for the activity of theanti-apoptotic protein Bcl-xL. In addition, a variant NIC protein in which four lysines were mutated to arginines(NIC4KR) retained anti-apoptotic activity, but was not regulated by Sirt1, and reconstituted the functions of non-nuclear NIC in Notch1-deficient Treg cells. Loss of Sirt1 compromised Treg cell survival, resulting in antigen-inducedT cell proliferation and inflammation in two mouse models. Thus, the Sirt1-Notch interaction may constitute animportant checkpoint that tunes noncanonical Notch1 signaling.

from
on A
pril 4, 2017http://stke.sciencem

ag.org/
INTRODUCTION
Several lines of evidence suggest that posttranslational modificationsof key signaling nodes influence the outcome of signaling networks(1–4). Sirtuin 1 (Sirt1), an NAD+ (nicotinamide adenine dinucleotide)–dependent deacetylase, regulates signaling pathways by modifying targetproteins. Sirt1 is highly conserved and catalyzes the deacetylation of his-tones and of multiple nonhistone proteins, including Notch (5–11).Sirt1 inhibits Notch-mediated transcription either by directly repressingNotch targets (7, 12, 13) or by targeting the Notch intracellular domain(NIC) for degradation (6, 14).

The Notch pathway signals through evolutionarily conserved ele-ments to regulate diverse processes, including cell survival. The pathwayis activated upon binding of one of five ligands (Jagged 1 and 2, andDelta-like 1, 2, and 4) to the receptor proteins Notch1 to Notch4, whichstimulates successive proteolytic cleavage events that result in the releaseof NIC from the full-length receptor. In the canonical pathway, NIClocalizes to the nucleus and interacts with its transcriptional cofactorsRBPJ-k (recombination signal binding protein for immunoglobulinkappa J) and MAML1 (mastermind-like transcriptional coactivator 1)to promote the transcription of target genes. Signaling through Notch1is obligatory for the acquisition of T cell fate, although inputs from theNotch pathway are modulated during T cell development (15, 16).Emerging evidence has revealed a requirement for Notch proteins inthe effector functions of mature T cells (17–19). Studies have also high-lighted differences in the Notch dependency of natural regulatoryT (nTreg) cells and induced Treg (iTreg) cells, with Notch activity beingessential for the expression of the gene encoding Foxp3 (forkhead boxP3), a master regulator of Treg cell identity in iTreg cells but not nTreg

cells (20–24). In another difference from iTreg cells, activated nTreg cellsgenerated by being stimulated throughCD3andCD28 in vitro are char-acterized by the nonnuclear localization of NIC (25).

Sirtuin activity is stimulated in nutrient-limiting conditions (26–28).Building on previous work, which demonstrated a requirement forNotch1 activity in promoting Treg cell survival in response to nutri-ent withdrawal (25), we asked whether Sirt1 modulated NIC activity.We present evidence implicating the Sirt1-NIC signaling axis in theanti-apoptotic activity in Treg cells and other cell types. We showedthat the nonnuclear NIC localization was dependent on the deacetylaseactivity of Sirt1. We mapped the target sites of Sirt1 with recombinantNotch1 proteins to recapitulate these interactions in cell lines. Finally,we explored the consequences of this interaction for Treg cell function.

RESULTSSirt1 controls NIC localization in Treg cellsActivated mouse Treg cells, which were generated through the stimula-tion of freshly isolated CD4+CD25+Foxp3+ Treg cells with beads coatedwith antibodies against CD3 and CD28 (see Materials and Methods),survived withdrawal of the cytokine interleukin-2 (IL-2) for 6 hoursin culture (Fig. 1A). Under these conditions, Treg cells showed increasedabundance of Sirt1 (Fig. 1, B and C, and fig. S1, A and B), and treatmentwith the Sirt inhibitor CHIC-35 abrogated Treg cell survival (Fig. 1A).We then investigated potential interactions between Sirt1 andNIC. Theunusual localization of NIC in plasma membrane–proximal complexesin Treg cells is critical for its anti-apoptotic activity (25, 29). We foundthat the localization of NIC, which was detected by total internalreflection (TIRF) microscopy, was disrupted in CHIC-35–treated Treg

cells in culture, with near-complete loss of signal at the plasma mem-brane for both NIC and the protein Rictor (Fig. 1D and fig. S1C) and aconcomitant increase in the amounts of NIC and Rictor in the nucleus(Fig. 1E and fig. S1D). However, the localization of phosphatidylinositol3-kinase (PI3K), amembrane-proximal lipid kinase also associatedwithNIC,was unchanged byCHIC-35 (Fig. 1, F andG, and fig. S1, C andD),

1 of 13

http://stke.sciencemag.org/


on April 4, 2017


Dow

nloaded from

indicating that the inhibitor did not cause a generalized disruption ofprotein localization. We confirmed that the effects of CHIC-35 werenot reversed in the short-term experiments with washout cultures(fig. S1E). Furthermore, we checked the effect of CHIC-35 on the dis-tribution of NIC in activated effector T cells or iTreg cells, which arederived from naïve T cells and exhibit predominantly nuclear localiza-tion of NIC (25). We found that CHIC-35 did not change the localiza-tion of NIC in effector T cells or iTreg cells and did not modify theabundance of Foxp3 in iTreg cells (fig. S1F).

Consistent with the relocalization of NIC in activated Treg cells inwhich Sirt1 was inhibited, phosphorylation of the serine and threoninekinase Akt at Ser473, which is a consequence of nonnuclear NIC activityin Treg cells, was reduced in cells treated with CHIC-35 compared tothat in control cells (Fig. 1H). A concomitant increase in the abundanceof Hes1 protein, which is encoded by the transcriptional target of NICHes1, in Treg cells treated with CHIC-35 suggested that NIC was notdegraded, because readouts associated with its activity in the nucleuswere not compromised (Fig. 1I). Consistently, the abundance of NICin CHIC-35–treated cells was similar to that in control cells (Fig. 1J).

Sirt1 interacts with NIC in Treg cellsNext, we transducedTreg cells with retroviruses expressing short hairpinRNA (shRNA) specific for Sirt1 or a scrambled control shRNA duringactivation, which was followed by antibiotic-based selection to enrichthe transduced cells. The distribution of endogenous NIC in the twoselected groups of cells was then analyzed by confocal microscopy withan antibody specific for cleavedNotch1. Cells inwhich Sirt1was knocked


down showed increased localization of NIC to the nucleus (Fig. 2A, topright, and fig. S2).Despite having increased amounts of nuclearNIC, cellsin which Sirt1 was knocked down were viable in IL-2–containing medi-um (Fig. 2B, dark bars). However, the cells expressing Sirt1-specificshRNA (Fig. 2A, bottom right, and fig. S2) underwent apoptosis afterIL-2withdrawal (Fig. 2B),whichwas accompaniedby a loss ofNIC stain-ing intensity. Thus, the effects of shRNA-mediated Sirt1 ablation wereconsistent with those from the CHIC-35 experiments and suggested thatSirt1 promotes the accumulation of cytoplasmic NIC in Treg cells.

The possibility that NIC physically associated with Sirt1 was nexttested. In activated Treg cells cultured under conditions that increasedthe abundance of Sirt1 (cytokine withdrawal for 6 hours), Notch1coimmunoprecipitated with Sirt1, as assessed by Western blottinganalysis (Fig. 2C). To ascertain whether the NIC-Sirt1 interaction de-pended on Notch processing, which precedes the release of NIC fromthe full-length receptor, we treated cells with a g-secretase inhibitor(GSI-X) for 5 to 6 hours to inhibit Notch processing before immuno-precipitations were performed. These experiments showed that Sirt1did not associate with Notch1, because Sirt1 was not detected in thecomplex, although PI3K was detected in the same immunoprecipitatesfrom cells treated with GSI (Fig. 2D). This indicated that, unlike PI3K,Sirt1 associated only with processed Notch1. GSI did not disrupt theincreased Sirt1 abundance in IL-2–deprived cells, because the proteinwas detected in whole-cell lysates, albeit at slightly reduced amountscompared to those in control cells (Fig. 2, C and D). Immunostainingof GSI-treated and control cells with an anti-Notch1 antibody (whichwas also used for the immunoprecipitations) showed that unprocessed

Fig. 1. Sirt1 promotes the plasma membrane–proximal localization of NIC. (A) Analysis of apoptotic damage in activated Treg cells cultured with or without IL-2 for18 hours. CHIC-35 was added (where indicated) at the onset of the assay. Data are means ± SD from three experiments. (B) Treg cells activated for 48 hours werecultured for 6 hours in the absence or presence of IL-2 (T6). The cells were fixed, permeabilized, and incubated with an anti-Sirt1 antibody (green) and counterstainedwith Hoechst 33342 (blue) to detect nuclei before being analyzed by confocal microscopy. Images are representative of three experiments. Scale bar, 2 mm. (C) Treg cellstreated as described in (B) were analyzed by Western blotting with antibodies against the indicated proteins. Western blots are representative of three experiments.Right: Bar graph shows densitometric analysis of the indicated proteins. Data are means ± SD of three experiments. (D to G) Activated Treg cells were cultured with orwithout IL-2 for 6 hours in the presence or absence of 500 nM CHIC-35 and then were imaged by TIRF microscopy (D and F) and wide-field microscopy (inset) and byconfocal microscopy (E and G) after being stained with antibodies for NIC (mNIA, red) and either Rictor (green) or PI3K (green). Scale bars, 5 mm. Images are repre-sentative of three experiments. (H to J) Activated Treg cells were cultured with or without IL-2 for 6 hours in the presence or absence of 500 nM CHIC-35 before beinganalyzed by Western blotting with antibodies against the indicated proteins. Actin and tubulin were used as loading controls. Western blots are representative of threeexperiments. Right: Bar graph shows densitometric analysis of the indicated proteins. Data are means ± SD of three experiments.

2 of 13



on April 4, 2017


Dow

nloaded from

Notch1 was localized mostly to the outer margins of the cytoplasm inGSI-treated cells (Fig. 2E). This change in spatial distribution resultedfrom the inhibition of Notch processing by GSI. As expected, the as-sociation between Sirt1 and NIC was not observed in activated Treg cellscultured in the presence of IL-2 (Fig. 2F, T6), because the abundance ofSirt1 was negligible in these nutrient-rich conditions (Fig. 1C). In all


conditions, the association between NIC and PI3K was readily detected,which served as an experimental control (Fig. 2, C, D, and F). Together,these results suggest that Sirt1 tunes NIC-mediated anti-apoptotic ac-tivity in Treg cells and that its deacetylase activity is required for thisfunction. This mode of regulation was next explored in mammalian celllines in which NIC-mediated anti-apoptotic activity has been describedpreviously (30).

Sirt1-mediated deacetylation of NIC is required forNIC-mediated anti-apoptotic activityThe expression of NIC in various cell lines protects them from diverseapoptotic stimuli (30–32). We found that the NIC-mediated inhibitionof staurosporine (STS)–induced apoptosis in human embryonic kidney(HEK) cells was abrogated byCHIC-35 (Fig. 3A). Similarly, using eithershRNA or small interfering RNA (siRNA) from different sources,knockdown of Sirt1 attenuated NIC-mediated anti-apoptotic activity(Fig. 3B and fig. S3A). Consistent with previous studies (6, 14), we con-firmed that CHIC-35 enhanced NIC-mediated transcription in apromoter reporter assay using theHes1 promoter (Fig. 3C). NIC blocksapoptosis stimulated by the Bcl-2 family proteins Bak and Bax, whichform heteromeric complexes and cause the release of apoptogenic mol-ecules sequestered in themitochondria. These proapoptotic proteins arekept in check by the activity of endogenous antagonists, such as Bcl-2and Bcl-xL (33). Because the knockdown of Sirt1 abrogatedNIC activity(Fig. 3D and fig. S3B), we determined whether a generalized regulationof mitochondrial function underlies the regulation of NIC by Sirt1.However, knockdown of Sirt1 did not attenuate the anti-apoptotic activ-ity of Bcl-xL (Fig. 3D and fig. S3B).

The requirement for the deacetylase activity of Sirt1 was next as-sessed in experiments withwild-type (WT) or kinase-deficient Sirt1 pro-teins (8, 9). Sirt1 was coexpressed with NIC and Bak in separate groupsof transfected cells, which were then incubated in the presence or ab-sence of CHIC-35. The attenuation of NIC-mediated anti-apoptotic ac-tivity by CHIC-35 was rescued in cells that coexpressed Sirt1 (Fig. 3E).However, in cells expressing the kinase-deficient Sirt1_H363Y protein,the extent of apoptosis was comparable to that in cells that did not ex-press NIC (Fig. 3E). Western blotting analysis established that the WTand mutant Sirt1 proteins were present in the cells at comparableamounts (Fig. 3E) and that they did not independently block Bak-induced apoptosis (fig. S3C).

Together, the experiments with the cell lines suggest that Sirt1specifically modulates NIC-mediated anti-apoptotic activity and thatits deacetylase activity is required for this function. Although NIC isa known target of Sirt1 (6, 14), this has not been tested in the contextof NIC-mediated anti-apoptotic activity. We reasoned that specific ly-sine residues in NIC would be required for this interaction. Hence,NIC proteins with site-directed substitutions of specific lysine residues(which were likely targets of Sirt1) were screened for resistance toCHIC-35 in anti-apoptotic assays.

Specific lysine residues in NIC are targets for modificationby Sirt1A cluster of lysine residues (K2157, K2160, K2164, and K2174) in NIC1were identified as putative target sites for Sirt1 activity (Fig. 4A). Theselysine residues were sequentially modified to the nonacetylatableresidue arginine (R) to preserve charge, and the resulting NIC variantswere screened for their sensitivity to the effects of CHIC-35.Whereas allfour variantsNIC_K2164→R(NIC1KR),NIC_1KR+K2157→R+K2160→R(NIC3KR), and NIC_3KR+K2174→R (NIC4KR) inhibited apoptosis in

A B

C D

E F

No IL-2

Fig. 2. Analysis of the Sirt1-NIC interaction in Treg cells. (A) Activated Treg cellstransduced with retroviruses expressing scrambled control (Scr) shRNA or Sirt1-specifc shRNA were cultured with or without IL-2 for 6 hours. The cells were thenfixed, permeabilized, and stained with antibody specific for cleaved NIC (clone Val1744, red) and with Hoechst 33342 to stain nuclei (blue) before being analyzed byconfocal microscopy. Images are of the central plane and are representative of 15to 20 cells per experimental condition. Scale bars, 5 mm. (B) Activated Treg cellsexpressing shRNAs as described in (A) were cultured for 15 hours with or withoutIL-2 before being analyzed to assess apoptotic damage. Data are means ± SD ofthree independent experiments. Bottom: The same cells were also analyzed byWestern blotting with antibodies against the indicated proteins. Western blotsare representative of three experiments. (C and D) Activated Treg cells werecultured without IL-2 for 6 hours in the absence (C) or presence (D) of 10 mMGSI-X. Whole-cell lysate (WCL) was prepared and subjected to immuno-precipitation (IP) with antibody specific for Notch1 (mNIA) or with mIgG (isotypecontrol) and Western blotting analysis with antibodies against the indicated pro-teins. Western blots are representative of two experiments. (E) Activated Treg cellswere cultured for 6 hours in the absence of IL-2 and either with or without 10 mMGSI-X. The cells were then stained for Notch1 (clone mNIA, red) and counter-stained with Hoechst 33342 to mark nuclei (blue) before being analyzed by con-focal microscopy. Scale bars, 5 mm. Images are representative of 15 to 20 cells perexperimental condition. (F) Freshly activated Treg cells were lysed and subjectedto immunoprecipitation with anti-Notch1 antibody (mNIA) or mIgG (isotype con-trol) and then analyzed by Western blotting with antibodies against the indicatedproteins. Western blots are representative of two experiments. HC, heavy chain.

3 of 13



on April 4, 2017


Dow

nloaded from

CHIC-35–treated HEK cells, only cells expressing the NIC4KR proteinwere resistant to the effects of CHIC-35 (Fig. 4B) or the siRNA-mediated silencing of Sirt1 (Fig. 4C). NIC4KR retained the capabilityto mediate C-promoter binding factor 1 (CBF1)–dependent Hes1promoter activity, which was used as a readout of NIC-mediated tran-scriptional function, as assessed by a luciferase-based promoter reporterassay (Fig. 4D). To rule out the acquisition of any anomalous activity,we confirmed that NIC4KR-mediated anti-apoptotic activity was abro-gated by coexpressing Numb, a well-characterized negative regulator ofNotch activity (Fig. 4E). Despite the marked effects of Sirt1 on the spa-tial distribution ofNIC in Treg cells, in theHEK cell line, exogenousNICand NIC4KR were localized to the nucleus (Fig. 4F).


The NIC-Sirt1 interaction was also tested in the LTG fibroblast cellline, which is protected from apoptosis induced by nutrient withdrawalby NIC (30). The LTG cell line survives serum deprivation; however,replacing the culture medium with buffered saline stimulates apoptoticdamage. We found that nutrient deprivation–induced apoptosis wasinhibited by ectopic expression of NIC or NIC4KR (Fig. 4G). As before,NIC-mediated protection from apoptosis was attenuated by CHIC-35,whereas the anti-apoptotic function of NIC4KR was preserved in thepresence of the inhibitor (Fig. 4G). Together, these experiments suggestthat specific lysine residues in NIC1 are targets of Sirt1, because theNIC4KR variant protein was unaffected by Sirt1. Because a change inthe spatial localization of NIC by CHIC-35 was not observed in the celllines, we characterized the localization and function of NIC4KR in ac-tivated Treg cells.

Sirt1 is required for the nonnuclear localization ofNIC in Treg cellsWTTreg cells transduced with retroviruses expressing NIC or NIC4KRwere protected from apoptosis after cytokine withdrawal (Fig. 5A).CHIC-35 compromised survival in cells transduced with either emptyretrovirus or retrovirus expressing NIC, whereas Treg cells transducedwith retrovirus encoding NIC4KRwere resistant to CHIC-35 (Fig. 5A).We next compared NIC and NIC4KR in activated Treg cells generatedfrom mice with a conditional deletion of Notch1 (Notch1−/− Treg cells)under the control of the Cd4 promoter (16). We found that Notch1−/−

Treg cells underwent apoptosis after cytokinewithdrawal (Fig. 5B, inset);however, expression of either NIC or NIC4KR protected these cells fromdeath (Fig. 5B). Furthermore, CHIC-35 attenuated the anti-apoptoticactivity of NIC, but not NIC4KR (Fig. 5, A and B), confirming the Sirt1dependence of NIC function in Treg cells.

We next analyzed the localization ofNIC andNIC4KR inNotch1−/−

Treg cells. Activated Notch1−/− Treg cells transduced with retrovirusesencoding NIC or NIC4KR were cultured for a short period (5 to6 hours) with cytokine or without IL-2 in the presence or absence ofCHIC-35 before being subjected to immunohistochemical analysis witha NIC-specific antibody. We found that exogenous NIC was more orless equally distributed in the nucleus and cytoplasm in Notch1−/− Treg

cells (Fig. 5C and fig. S4A). However, treatment of IL-2–deprived cellswith CHIC-35 enriched nuclear staining of NIC (Fig. 5C and fig. S4A).On the other hand, in cells expressing NIC4KR, NIC staining was en-riched in the cytoplasm and was unchanged even when the cells werecultured with CHIC-35 (Fig. 5C and fig. S4A). As seen in other experi-ments, the localization of PI3Kwas not affected by CHIC-35 in the cellsexpressing either NIC or NIC4KR (Fig. 5D and fig. S4B). The immuno-staining analysis was quantified and indicated that ~42 and ~58% oftotal staining for NIC was detected in the nucleus and cytoplasm, re-spectively, in cells transfected with plasmid encoding NIC (Fig. 5E).In contrast, in cells expressing NIC4KR, NIC was highly enriched inthe cytoplasm (~80%of total) (Fig. 5E). The difference in the twogroupswas even more apparent in the response of the cells to CHIC-35, wherenuclear signalswere enhanced in cells expressingNIC (~79%of the totalstaining), but remained unchanged in cells expressing NIC4KR (Fig.5E). Similar results were obtained with activated Treg cells isolated fromWT C57BL/6 mice (fig. S4, C and D).

As a readout of the nonnuclear activity of NIC, and confirming thatexogenous NIC functioned similarly to endogenous NIC, the abun-dance of Akt phosphorylated at Ser473 was reduced inCHIC-35–treated,NIC-expressing cells but not in CHIC-35–treated, NIC4KR-expressingcells (Fig. 5F).Consistentwith theCHIC-35–dependentnuclear enrichment

Fig. 3. Sirt1 is required for the NIC-mediated protection of cells from apoptosis.(A and B) HEK cells transfected with plasmids expressing the indicated proteins alone(A) or in the presence of the indicated siRNAs (B) were incubated with 10 mM STS (toinduce apoptosis) in the presence or absence of 500 nM CHIC-35. The cells were thenanalyzed to determine the percentages of green fluorescent protein (GFP)–expressing cells with apoptotic nuclei. (B) Bottom: Cells transfected with the indicatedsiRNAs were analyzed by Western blotting with antibodies specific for Sirt1 and tu-bulin. Blots are representative of three experiments. (C) HEK cells were cotransfectedwith plasmids expressing red fluorescent protein (RFP) or NIC-RFP and the Hes1-Lucconstruct and the Renilla-Luc coreporter construct as described in Materials andMethods. The cells were cultured for 36 hours in the presence or absence of 500 nMCHIC-35 before the luciferase activity of the Hes1 reporter construct was measured. Dataare means ± SD of the fold induction in luciferase activity in NIC-RFP relative to RFP-expressing cells. (D) HEK cells pretreated with scrambled or Sirt1-specific siRNAs andthen transfected with plasmid encoding Bax-GFP alone or in the presence of the indi-cated plasmids were analyzed after 18 hours to determine the percentages of cells withapoptotic nuclei. Inset: Cells transfected with the indicated siRNAs were analyzed byWestern blotting with antibodies against the indicated proteins. Western blots are rep-resentative of three experiments. (E) HEK cells transfected with plasmid encoding Bak-RFP alone or together with plasmids encoding the indicated constructs were incubatedin the absence (black) or presence (gray) of 500 nM CHIC-35. Both the WT and H363YSirt1 proteins were fused to the Flag tag. The cells were then analyzed to determine thepercentages of cells containing apoptotic nuclei. Bottom: Cells transfected with plas-mids encoding the indicated proteins were analyzed by Western blotting with anti-bodies against the Flag tag and tubulin. Data in all graphs are means + SD fromthree independent experiments. *P ≤ 0.03, **P ≤ 0.001.

4 of 13



on April 4, 2017


Dow

nloaded from

ofNIC, the abundanceofHes1proteinwas increased inCHIC-35–treated,NIC-expressing, activated Treg cells, whereas there was no change inHes1abundance in cells expressing NIC4KR (Fig. 5F). Together, these datasuggest a previously uncharacterized requirement for Sirt1-dependentdeacetylation in the regulation ofNIC localization and its anti-apoptoticactivity. Previous studies, including our own work, implicated non-canonical NIC signaling in the suppressor function of Treg cells (29, 34).In a previous study, we showed that the NIC-mediated control of thesurvival and suppressor activity of Treg cells depends on its localizationin the cytoplasm (25, 29). Hence, we hypothesized that Sirt1, through itsregulation of NIC localization, would be required for the suppressor ac-tivity of Treg cells.

Sirt1 is required for Treg cell functionTreg cells function to dampen the proliferation of effector T cells andprevent pathologies, such as those associated with autoimmune re-sponses, and the suppressor activity and identity of Treg cells are prin-cipally controlled by the transcription factor Foxp3 (35, 36). However,Notch1 activity modulates the functions of nTreg cells by mechanisms


that do not target Foxp3 (29, 34). Before assessing the effects of Sirt1, wefirst tested whether Foxp3 protein abundance was independent of Sirt1.We found by confocalmicroscopy analysis that the abundance of Foxp3was similar in Treg cells expressing control shRNA or Sirt1-specificshRNA (Fig. 6A), which suggests that Foxp3 abundance in this exper-imental systemdoes not dependon Sirt1 (Fig. 6B).Hence, we proceededto test whether ablation of Sirt1 modified the suppressor activity ofTreg cells.

Established protocols (37, 38) were used to analyze the suppressorfunction of Treg cells. Briefly, responder cells (naïve T cells expressing anantigen-specific T cell receptor) and suppressor cells (Treg cells) were co-injected into immunocompetent recipient mice, and the ability of theTreg cells to suppress antigen-induced T cell proliferation was then as-sessed. In this analysis, naïve CD4+ T cells isolated from OT-II trans-genic mice (which could be identified because the cells were CD45.2+)were loaded with the fluorescent dye carboxyfluorescein diacetate suc-cinimidyl ester (CFSE) as described inMaterials andMethods. The CFSE-loaded cells were injected into congenic host mice (CD45.1+, B6SJLmice),whose cells differed from the responders only in the CD45 isoform

Fig. 4. The deacetylase activity of Sirt1 is required for the protection of cells from apoptosis. (A) Schematic of the site-directed lysine-to-arginine mutants ofhuman NIC. (B) HEK cells transfected with plasmid encoding Bak-GFP alone or together with plasmids encoding the indicated proteins were incubated in the absence(black bars) or presence (gray bars) of 500 nM CHIC-35 for 18 hours. The percentages of cells with apoptotic nuclei were then determined. (C) HEK cells pretreated withcontrol or Sirt1-specific siRNAs and then transfected with plasmids encoding the indicated constructs were treated with 10 mM thapsigargin (TG) for 18 hours. Thepercentages of cells with apoptotic nuclei were then determined. Bottom: Cells treated with the indicated siRNAs were analyzed by Western blotting with antibodiesagainst Sirt1 and tubulin. Western blots are representative of three experiments. (D) HEK cells cotransfected with the Hes1 reporter construct and the Renilla-Luccoreporter construct and plasmid encoding NIC4KR-RFP in the presence or absence of plasmid encoding dominant-negative CBF1 (DNCBF1) were subjected to lucif-erase assays. Data are means ± SD of the fold increase in promoter activity in NIC4KR-RFP relative to RFP-expressing cells. (E) HEK cells transfected with combinations ofplasmids encoding the indicated constructs were treated with 10 mM TG or vehicle control for 18 hours before the percentages of RFP-positive cells with apoptoticnuclei were determined. (F) HEK cells expressing either NIC1KR-RFP or NIC4KR-RFP (red) were stained with Hoechst 33342 to mark nuclei (blue) and then were analyzedby confocal microscopy. Images are representative of 50 cells per condition. Scale bars, 10 mm. (G) LTG cells were transfected with plasmids encoding the indicatedconstructs and treated with 10 mM TG in the absence (black) or presence of CHI-C35 for 18 hours before the percentages of RFP-positive cells with apoptotic nuclei weredetermined. Data in all bar graphs are means ± SD from three independent experiments. *P ≤ 0.03, **P ≤ 0.001.

5 of 13



on April 4, 2017


Dow

nloaded from

expressed on the cell surface. The responder T cells were co-injected with-out or with CD45.2+ Treg cells that expressed either control shRNA orSirt1-specific shRNA and had been activated in vitro. The recipientmice were challenged with ovalbumin (OVA) peptide 15 to 18 hoursafter receiving the naïve T cells. Three days later, lymphocytes isolatedfrom the draining lymph nodes of the recipient mice were analyzed fortheir proliferative status, which correlatedwith the extent of the dilutionof CFSE in the gated CD45.2+ (responder) OT-II T cells.


Inmice that did not receive Treg cells, the distribution of CFSE in theOT-II responder cells ranged across multiple populations, which wasindicative of cell proliferation (Fig. 6C). Treg cells expressing scrambledcontrol shRNA suppressed the proliferation of the responder OT-IIT cells; hence, CFSE fluorescence in those cells was notmarkedly diluted(Fig. 6C). On the other hand, in mice co-injected with Treg cells ex-pressing Sirt1-specific shRNA, the responder OT-II T cells proliferatedrobustly (Fig. 6C), in amanner comparable to the proliferation ofOT-II

Fig. 5. NIC4KR is refractory to regulation by Sirt1. (A and B) Treg cells from WT (A) or Notch1 null (Notch1−/−) mice (B) transduced with retroviruses expressing NIC orNIC4KR were cultured for 15 hours with or without IL-2 in the absence or presence of CHIC-35. The percentages of cells with apoptotic nuclei were then determined.Inset: Notch1−/− Treg cells were cultured with or without IL-2 for 15 hours before percentages of cells with apoptotic nuclei were determined. (C and D) Notch1−/− Treg cellstransduced with retroviruses expressing NIC or NIC4KR were cultured for 6 hours in IL-2–deficient medium in the presence or absence of CHIC-35, as indicated. The cells werethen stained with an antibody against processed Notch (green) (C) or with an antibody against PI3K (red) (D) and counterstained with Hoechst 33342 to detect nuclei (blue).The cells were analyzed by confocal microscopy. Images (central stack) are representative of 20 cells per experimental condition. Scale bars, 5 mm. (E) Quantification of thedistribution of NIC in the cells represented in (C). Data are means ± SD of three experiments. (F) Notch1−/− Treg cells transduced with retroviruses expressing NIC or NIC4KRwere cultured in the absence or presence of CHIC-35 for 6 hours. The cells were then analyzed by Western blotting with antibodies against the indicated target proteins.Western blots are representative of two experiments. Right: Bar graph shows densitometric analysis of the indicated proteins. Data are means ± SD of three experiments.

6 of 13



on April 4, 2017


Dow

nloaded from

Fig. 6. Sirt1 modulates Treg cell function in vivo. (A) Activated Treg cells transduced with retroviruses expressing scrambled control shRNA or Sirt1-specific shRNAwere fixed, permeabilized, and stained with anti-Foxp3 antibody (green) and with Hoechst 33342 to stain nuclei (blue) before being analyzed by confocal microscopy.Images (central plane) are representative of 100 to 120 cells. Scale bars, 5 mm. (B) Activated Treg cells transduced with retroviruses expressing scrambled control shRNAor Sirt1-specific shRNA were analyzed by microscopy. The integrated density of the Foxp3 staining intensity of images was determined with ImageJ software. Right:Lysates of the same cells were analyzed by Western blotting with antibodies against the indicated proteins. Western blots are representative of three experiments (C) CFSE-loaded CD4+ OT-II cells were injected alone or were co-injected into three mice with Treg cells transduced with retroviruses expressing scrambled or Sirt1-specific shRNAs. Themice were then challenged with antigen (mOVA), and the extent of dilution of CFSE in CD45.2+ OT-II cells recovered from the lymph nodes of the recipient mice on day 3 wasdetermined by flow cytometry. Flow cytometry dot plots are representative of two independent experiments, with two or three animals per group analyzed per experiment.Numbers in the panels indicate the percentage of the CD45.2+ OT-II population recovered. (D) CFSE-labeled, activated Notch1+/+ or Notch1−/− Treg cells (CD45.2+) were co-injected with naïve OT-II CD4+ T cells into CD45.1+ recipient mice. One day later, the mice were challenged with OVA peptide. After 48 hours, the lymph nodes of the recipientmice were harvested, and the extent of dilution of CFSE in the CD45.2+ donor Treg cells was analyzed by flow cytometry. Data are representative of two experiments. Numbersin the plots indicate the percentages of donor Treg cells recovered. (E) CFSE-labeled, activated WT Treg cells (CD45.2

+) transduced with retroviruses expressing scrambled orSirt1-specific shRNA were co-injected with naïve OT-II CD4+ T cells (CD45.2+) into CD45.1+ recipient mice. One day later, the mice were challenged with OVA peptide. After 48 hours,the lymph nodes of the recipient mice were harvested, and the extent of dilution of CFSE in the CD45.2+ donor Treg cells was analyzed by flow cytometry. Data are representative oftwo experiments. Numbers in the plots indicate the percentages of donor Treg cells recovered.

Marcel et al., Sci. Signal. 10, eaah4679 (2017) 4 April 2017 7 of 13



on April 4, 2017


Dow

nloaded from

cells that were not co-injected with Treg cells (Fig. 6C). This result isconsistent with a requirement for Sirt1 in the regulation of NIC activity.

To demonstrate a requirement for NIC in the survival of Treg cellsduring the response to antigen in vivo, we assessed whether there was achange in the recovery of Treg cells that expressed Sirt1-specific shRNA.Before assessing the requirement for Sirt1, we first asked whether theloss of Notch1 compromised the recovery of Treg cells. To this end,WT or Notch1−/− Treg cells, which were loaded with CFSE, were co-injectedwith the naïveOT-II responder T cells as before. Recipientmicewere challenged with OVA 15 to 18 hours after injection of cells. Again,by tracking the dilution of CFSE in the gated CD45.2+ donor T cells, weobserved a substantial reduction in the recovery of Notch1−/− Treg cellsas compared toWTTreg cells from the lymph nodes of recipient mice2 days after antigen challenge (Fig. 6D). Next, we performed the sameprotocol and analyzed the recovery of WT Treg cells expressing eitherscrambled or Sirt1-specific shRNA thatwere co-injectedwith respondercells into mice followed by antigen challenge. The number of activatedTreg cells expressing Sirt1-specific shRNA that were recovered wasmuch lower than that of Treg cells expressing scrambled shRNA (Fig.6E). Together with the results of the IL-2 deprivation assays, theseexperiments suggest a critical role for Sirt1 in the maintenance ofNIC activity and Treg cell function.

Sirt1 is a positive regulator of Treg cell functionTo strengthen the analysis of the regulation of the Notch-dependentTreg cell–suppressive function by Sirt1, we extended the analysis to in-clude two unrelated models of inflammation, which are, however,independent of specific antigens. These models include the Treg cell–mediated reduction in the accumulation of effector and memory T cellsubsets inmicewith a genetically targeted deletion ofNotch1 in the nTreg(Foxp3+) lineage and the induction of glucose intolerance in a mousemodel of obesity resulting from defective leptin signaling. The effectsof Treg cells in both systems were described previously (29, 39, 40).

Wepreviously reported that the deletion ofNotch1under the controlof the Foxp3 promoter (in Foxp3-Cre::Notch1lox/lox mice) causes lymphnode enlargement (29). Confirming reported observations, this resultsfrom the accumulation of CD4+CD62LlowCD44high (effector memory)and CD4+CD62LhighCD44high (central memory) T cells, as well as aconcomitant decrease in the number of CD4+CD62LhighCD44low (naïve)T cells within a few weeks of birth (Fig. 7A). This is consistent with adefect in immunoregulation or Treg cell function (29, 41). We previouslyshowed that the adoptive transfer of functional Treg cells reduces theaccumulation of effector T cells and the consequent enlargement oflymph nodes in suchmice (29). Consistent with these results, the adop-tive transfer of Notch1−/− Treg cells (7 to 8 days before analysis) intoFoxp3-Cre::Notch1lox/lox recipient mice failed to ameliorate T cell accu-mulation, whereas the transfer of Notch1-deficient Treg cells expressingexogenousNIC fused to a nuclear export signal (NIC-NES) reduced thenumber of T cells that accumulated in the lymph nodes of the recipientmice (Fig. 7B). This was also apparent from analysis of lymph node size,which remained enlarged in animals injected with Notch1−/− Treg cellsbut were much smaller in mice injected with Notch1−/− Treg cellsexpressing NIC-NES (Fig. 7C). As described earlier, WT Treg cells re-duced the number of effector T cells that accumulated in the recipientmice (Fig. 7D); however, the ablation of Sirt1 inWTTreg cells preventedthis result (Fig. 7D). This was also confirmed when the total number ofcells recovered from the lymphnodeswas compared under the differentconditions. Thus, cell recovery from lymph nodes was reduced in miceinjectedwithWTTreg cells 7 days before the analysis, whereasmore cells


were recovered from uninjected mice or from mice injected with Treg

cells expressing Sirt1-specific shRNA, wherein inflammation was notreduced (Fig. 7E).

Next, we assessed the regulation of Treg cell function by Sirt1 in amodel of the obesity-induced impaired clearance of injected glucose(glucose intolerance) from the bloodstreamof affectedmice (42). In thismodel, the transfer of WT Treg cells ameliorates glucose intolerance inthe short term in leptin receptor–deficient (Leprdb/db) mice (39, 40). Wepreviously demonstrated a role for nonnuclear NIC activity in Treg cellsin this system (29); thus, we used this model to assess a role for Sirt1 inthe regulation of Notch-dependent Treg cell function. Consistent withpublished observations, the clearance of glucose from the blood offasting Leprdb/db mice was substantially delayed in comparison to thatin heterozygous (Leprdb/+) littermate controls (Fig. 7F). In Leprdb/dbmiceinjected with Treg cells expressing scrambled control shRNA and thentested 7 to 8 days later, clearance of blood glucose was markedly im-proved and was comparable to that observed in Leprdb/+ mice at latetimes (Fig. 7F). In contrast, in mice injected with Treg cells expressingSirt1-specific shRNA, the extent of clearance of blood glucose did notdiffer from that of the Leprdb/db mice (Fig. 7F). Together, these experi-ments with adoptively transferred Treg cells in assays of immuno-suppression are suggestive of a key role for Sirt1 in the NIC-mediatedregulation of Treg function.

DISCUSSIONNotch signaling regulates diverse cellular responses during develop-ment and in adult tissues, and, unsurprisingly, aberrant signaling fromthis receptor underliesmultiple disorders (43–45). The coreNotch path-way is relatively simple and culminates in release of the active signalingintermediate NIC (46), which localizes to the nucleus unless regulatedotherwise. It is now well established that much of the complexity andversatility associated with Notch activity is brought about by its integra-tion with other (nodal) signaling networks (47, 48). Furthermore, post-translational modifications, such as acetylation, phosphorylation, andubiquitylation, are all reported to modulate Notch function, in manycontexts with immediate consequences to protein stability, but also re-sulting in regulated noncanonical outputs of NIC activity (6, 49).

Cellular signaling in response to changes in nutritional cues maylead to differentiation and cell survival or cell death. A well-establishedearly response to nutrient deprivation is the activation of the Sirt familyof proteins, which are sensitive to changes in the ratio of NADH (re-duced form of NAD+) to NAD (oxidized form of NAD+) that charac-terize nutrient stress (26–28). Consistent with the hypothesis that theactivation of Sirt1 enables the cell to adapt to situations of energy stress,the abundance of Sirt1 is generally increased in situations of low nutri-ent availability (28). Interactions with proteins such as p53 and AMPKare implicated in the regulation of Sirt1 function, all of which are asso-ciated with the response to nutrient stress. Thus, the increase in Sirt1protein abundance in Treg cells that we observed after the withdrawal ofcytokine was expected (Fig. 1). That the localization of NIC was sensi-tive to Sirt1, as demonstrated in experiments in which cells were trans-duced with a retrovirus encoding Sirt1-specific shRNA or were treatedwith a chemical inhibitor of Sirt1, prompted a more careful exam-ination of the underlying interaction between Sirt1 and NIC. Theincrease in Hes1 protein abundance in cells treated with the Sirt1 in-hibitor is consistent with published evidence that Sirt1 inhibits boththe nuclear functions of NIC and of the expression of the gene encod-ing Notch1 (6, 14).

8 of 13



on April 4, 2017


Dow

nloaded from

Our experiments demonstrated that the Sirt-mediated deacetylationofNICwas critical for its nonnuclear localization,whichhad consequencesfor Treg survival and, more generally, for NIC1-mediated anti-apoptoticactivity. Although Sirt1 resides mainly in the nucleus (50), several ofits targets are found in the cytoplasm, and Sirt1 can shuttle from thenucleus to the cytosol, for example, as a consequence of the inhibitionof insulin signaling (51). The observation that Notch activity is executedfrom a nonnuclear location is relatively recent (25, 29, 32, 52, 53).Building on earlier work that reported nonnuclear NIC-mediated activ-ity in Treg cells, we now suggest that NIC is a direct target of Sirt1, whichfavors noncanonical NIC activity. In support of this hypothesis, weshowed the association ofNICwith immunocomplexes as confirmationof Sirt1 activity in the cytoplasm (Fig. 2C). Furthermore, key functionalelements of the Sirt1-Notch1 signaling cascade were not restricted toT cells and were activated in cell lines expressing recombinant NIC.Sirt1-Notch1 interactions were inferred from the disruption of NIC lo-


calization and anti-apoptotic activity after Sirt1 inTreg cellswas perturbed,the restoration of NIC-mediated anti-apoptotic activity in a Sirt1-independent manner by recombinant NIC4KR, the demonstration thatSirt1 specifically modified NIC and not the anti-apoptotic protein Bcl-xLand that the enzymatic activity of Sirt1 was required for the regulation ofNIC, and the demonstration of the association between Sirt1 and NIC.The resistance of NIC4KR to the loss or inhibition of Sirt1 suggested thatNotch may be a direct target of Sirt1 and is consistent with earlier evi-dence of NIC modification (6).

Whether the deacetylation of NIC is linked with the ubiquitylationof the same or proximate lysine residues or to modifications such asmethylation (54) are possibilities that remain to be investigated. BothNIC and NIC4KR are equally susceptible to inhibition by Numb, a neg-ative regulator of Notch function (55), which suggests that NIC4KR isnot resistant to regulation. However, both the timing and duration ofNIC activity may be dependent on Sirt1. In addition to showing that

N1–/–T cells

Foxp3Cre-N1–/–

lymphnode

A B C

D E F

reg

Fig. 7. Notch regulation of Treg cell function is dependent on Sirt1. (A) Lymph nodes were isolated from heterozygous genetic control (Foxp3-Cre::Notch1WT/lox) andmutant (Foxp3-Cre::Notch1lox/lox) mice, and the lymph node cells were analyzed by flow cytometry to determine the percentages of T cells positive for the indicatedmarkers. (B) Foxp3 mutant mice were injected with CD4-Cre::Notch1lox/lox (N1−/−) Treg cells expressing either vector control (pBABE) or NIC-NES. Seven days later, the T cellsubsets in lymph nodes isolated from the recipient mice were analyzed by flow cytometry. (C) Representative images of whole lymph nodes isolated from three mice injectedwith Treg cells as described in (B). (D) Foxp3 mutant mice were injected with WT Treg cells transduced with retroviruses expressing scrambled or Sirt1-specific shRNAs. Sevendays later, the T cell subsets in lymph nodes isolated from the recipient mice (three per condition) were analyzed by flow cytometry. (E) Lymph nodes were isolated fromheterozygous genetic control Foxp3-Cre::Notch1WT/lox (Het) mice, Foxp3-Cre::Notch1lox/loxmice that had been left untreated (UT) or were injected 7 days before analysis with WTTreg cells transduced with retroviruses expressing scrambled or Sirt1-specific (Sirt1) shRNAs. The total numbers of cells in the isolated lymph nodes were determined with ahemocytometer. (F) Intraperitoneal glucose tolerance test was performed on Leprdb/+ (△) and Leprdb/db (▴) mice that were injected intravenously with phosphate-bufferedsaline (PBS) or WT Treg cells transduced with retroviruses expressing scrambled (▪) or Sirt1-specific shRNA (□). Data in all bar graphs are means ± SD of three independentexperiments, with two or three mice per condition in each experiment. *P ≤ 0.03, **P ≤ 0.001.

9 of 13



Sirt1 regulated NIC activity in the cytoplasm, our data also showed thatSirt1 antagonizes transcriptional regulation by NIC. It is tempting tospeculate that such a switch may tune NIC function during differen-tiation, perhaps to ensure cell survival and the associated remodelingof organelles in cells undergoing differentiation.

on April 4, 2017


Dow

nloaded from

MATERIALS AND METHODSMiceThe Notch1lox/lox and Cd4-Cre::Notch1lox/lox (Notch1−/−) mouse strainswere a gift from F. Radtke [École Polytechnique Federale de Lausanne(EPFL), Switzerland] (16). The Foxp3tm4(YFP/Cre)Ayr/J, C57BL/6J,B6SJL, OT-II, and Leprdb/dbmouse strains were obtained from the JacksonLaboratory. The Notch1lox/lox mutant mouse strain was backcrossed withC57BL/6 mice. The Notch1lox/lox and Foxp3-Cre strains were crossed togenerate Foxp3-Cre::Notch1lox/lox mice. To isolate Treg cells and other cellsubsets, we used mice at 8 to 12 weeks of age and housed them incontrolled temperature and light environments, inhighbarrier conditions,and in controlled systems (individually ventilated cages). Colonies wereroutinelymonitored for the full pathogenpanel recommendedby theFed-eration of Laboratory Animal Science Associations. Breeding colonieswere maintained in-house, and all experimental protocols were approvedby the Institutional Animal Ethics Committee (NCBS-AEC-AS-6/1/2012;INS-IAE-2016/01[N]) andcompliedwith thenormsof theCommittee forthe Purpose ofControl and Supervision of Experiments onAnimals, Gov-ernment of India.

CellsHEK 293T (HEK) and LTG (fibroblast) cell lines were maintainedat 37°C in 5% CO2 in Dulbecco’s modified Eagle’s medium. The LTGhuman fibroblast cell line was a gift from A. Rangarajan [Indian Insti-tute of Science (IISC), Bengaluru, India]. Primary T cells were culturedin RPMI 1640 supplemented with antibiotics, glutamine (Invitrogen),and 5 to 10% heat-inactivated fetal bovine serum (Invitrogen). HEKcells that were used to generate viruses for retroviral transduction werenot used beyond passage number 20.

ReagentsRecombinant IL-2 and the T cell selection kits (MagCellect) wereobtained from R&D Systems. Dynabeads Flowcomp T cell selectionkits, Lipofectamine 2000, magnetic beads coated with anti-CD3 andanti-CD28 antibodies, and Alexa Fluor–conjugated secondary antibodiesagainst mouse and rabbit immunoglobulin G (IgG) were purchasedfrom Invitrogen. All others reagents were from Calbiochem or Sigma.Stock solutions of 1 mM STS were made in ethanol, whereas stocksolutions of 100 mMCHIC-35 and 1 mMGSI-X were made in dimethylsulfoxide and stored at −20°C. Antibodies were procured from thefollowing sources: anti-Notch1 (clone mN1A) was from Santa Cruz;anti–cleaved Notch1 (Val1744), anti-Sirt1 (1F3), anti-Akt1 (C73H10),anti-PI3K (19H8), anti-Akt-pS473, and anti-Rictor were from CellSignaling Technology; anti-tubulin and anti-actin were obtained fromNeomarker; antibody against the Flag tag was obtained from Sigma;and anti-Foxp3 antibody (FJK-16) was from Ebiosciences.

T cells and retroviral transductionsCD4+CD25+ nTreg cells were isolated from murine spleens through acombination of negative selection to enrich for CD4+ cells followed bypositive selection for CD25+ cells according to the manufacturer’s in-structions (R&DSystemsor Invitrogen).TheCD4+CD25+cells thus isolated


were ~95% Foxp3+ by immunostaining and confocal microscopy–basedanalysis across multiple experiments. Cells (~2 × 106/ml) were activatedby being cocultured with beads coated with anti-CD3 and anti-CD28antibodies (20 ml/ml; Invitrogen) in 24-well plates. After 48 hours, thebeads were removed bymagnetic separation, and the activated Treg cellswere continued in culture in 50% conditioned medium containing IL-2(2 ng/ml; R&D Systems) or were used in functional assays. To generateiTreg cells, we activated CD4

+CD25− (naïve) T cells isolated by negativeselection (1 × 106/ml) with beads coated with anti-CD3 and anti-CD28antibodies (10 ml/ml; Invitrogen) in the presence of IL-2 (2 ng/ml) andtransforming growth factor–b (2 ng/ml) for 72 hours. Effector T cellswere obtained by activatingCD4+CD25−T cells (2 × 106/ml) with beadscoated with anti-CD3 and anti-CD28 antibodies (20 ml/ml; Invitrogen)for 48 hours. To generate retroviruses for infections, we packaged retro-viruses with the vector pCL-Eco (Addgene) in transfected HEK cells.Virus-containing cell culturemediumwas concentrated, and target T cellswere infected 24 hours after stimulation by spinfection in 24-well plates(at 500g for 90 min at room temperature) and then were continued inculture. After 48 hours, the cells were harvested and then cultured inmedium supplemented with IL-2 (1 mg/ml) for a further 18 to 24 hours.To select for infected cells, we cultured the cells for 48 hours inmediumcontaining the antibiotic puromycin (1 mg/ml). Live cells were selectedon day 2 by centrifugation in Ficoll. Knockdown of targeted proteinswas assessed by Western blotting analysis of cell lysates.

PlasmidsTheNIC-GFP, NIC-RFP, Bax-GFP, Bak-GFP, and Bak-RFP constructswere described previously (30, 31). The following plasmids were ob-tained as gifts: pBABE, pBABE-NIC-NES, and GFP-N1C-NES wereprovided by B. A. Osborne (University of Massachusetts, Amherst,MA, USA), and DN-CBF1 was provided by J. Aster (Harvard MedicalSchool, Boston). TheNumb-GFP constructwas obtained fromOriGene(Rockville, MD). The NIC1KR-RFP andNIC4KR-RFP constructs weremade in-house by site-directed mutagenesis of NIC-RFP using the fol-lowing primer sets: NIC_K2164R_FP: GCCCAGCAGCAGAGGCCTG-GCCTG; NIC_K2164R_RP: CAGGCCAGGCCTCTGCTGCTGGGC;NIC_K2157R,K2160R_FP: CGTGCAGGGCAAGAGGGTCCG-CAGGCCCAGCAGCAAAGG; NIC_K2157R,K2160R_RP:CCTTTGCTTGGGCCTGCGGACCCTCTTGCCCTGCACG;NIC_K2171R,K2174R_FP: CTGTGGAAGCAGGGAGGCCAGG-GACCTCAAG; NIC_K2171R,K2174R_RP: CTTGAGGTCCCTGG-CCTCCCTGCTTCCACAG. The sequences of all constructs wereverified by automated sequencing in-house. Both shRNAmir-GFPand the scrambled control GFPwere obtained fromOpenBioSystems.ON-TARGETplus SMARTpool siRNA specific for Sirt1, the scrambledcontrol siRNA, and the siRNA transfection reagent Dharmafect-1 wereobtained fromDharmacon.Murine Sirt1–specific shRNAwas obtainedfrom OriGene. Plasmids encoding Sirt1, Sirt1_H363Y, and pCL-Ecowere from Addgene.

Cell transfectionsHEK cells were plated at a density of 0.1 × 106 cells in 24-well platesand were routinely transfected when they reached 60 to 70% con-fluency with Lipofectamine 2000 according to the manufacturer’s in-structions. The LTG fibroblast cells were plated at a density of 0.2 ×106 cells in six-well plates, and once they reached 50 to 60% con-fluency, they were transfected with X-treme GENE HP or FuGENE 6HDaccording to the manufacturer’s instructions. The amounts of the plas-mids used in the transfections were as follows: GFP or RFP (0.5 mg), Notch

10 of 13



on April 4, 2017


Dow

nloaded from

constructs (2mg),Bax-GFPorBak(1mg),SIRT1constructs (3mg), shRNAmirSirt1 or control (1 mg), and pcDNA3 was used as the empty vector toequalize the amount of DNA across the different transfection groups.HEK cells were transfected with 100 nM siRNA according to the man-ufacturer’s instructions. Knockdown of the protein of interest was con-firmed by Western blotting analysis, and the cells were used forfunctional or biochemical analyses 48 hours after they were transfected.

Apoptosis assaysTo induce apoptosis, HEK or LTG cells were switched from completemedium to serum-free medium 16 to 18 hours after transfection andthen were treated with vehicle control or STS (10 mM). After a further18 to 20 hours, the cells were analyzed for apoptotic nuclear damage. Inexperiments in which the cells were transfected with plasmids encodingBax or Bak, apoptotic nuclear damage was assessed 18 to 20 hours aftertransfection. In LTG cells, growth medium was replaced with PBS, orthe cells were cultured in growth medium to stimulate neglect-induceddeath. Twenty-four hours later, the cells were analyzed for apoptoticnuclear damage. Where required, CHIC-35 (500 nM) was added atthe time of STS addition or at 6 hours after transfection. To induceapoptosis after cytokinewithdrawal, T cells (0.3 × 106/ml)were culturedwithout IL-2 in the presence or absence of 500 nMCHIC-35 for 16 to18 hours. Cells cultured with IL-2 were used as controls. To analyzenuclear damage, samples were stained for 5 min at ambient tempera-ture with the DNA binding dye Hoechst 33342 (1 mg/ml) in PBS. Thecells were then washed, and their nuclear morphology was assessedby microscopy. In HEK cells transfected with plasmids encodingfluorophore-tagged proteins, only those cells that were positive forfluorophores were scored.

Immunoprecipitation and Western blotting analysisCells (0.3 × 106 to 0.5 × 106) were lysed in an SDS lysis buffer [2% SDS,glycerol, bromophenol blue, 1 M DTT, and 1 M tris-Cl (pH 6.8) sup-plemented with aprotinin, leupeptin, and pepstatin (2 mg/ml each),1 mM phenylmethylsulfonyl fluoride, 1 mM NaF, and 1 mMNa3VO4]for 10 min at 100°C. MG-132 (1 mg/100 ml) was included in all buffersused to prepare lysates for analysis of the Notch protein, which werefrozen at −80°C soon after preparation. Whole-cell lysates were re-solved by SDS–polyacrylamide gel electrophoresis and transferred tonitrocellulose membranes (GE Healthcare), which were incubated over-night at 4°C with primary antibodies at concentrations recommendedby the manufacturers. The membranes were washed thrice with tris-buffered saline–Tween 20, followed by incubation with horseradishperoxidase–conjugated secondary antibody (CST, 1:1000 dilution) for1 hour at room temperature. Membranes were developed with theImageQuant LAS 4000 Biomolecular Imager (GE Healthcare) and quan-tified with the ImageJ software. Primary antibodies used for Westernblotting analysis were as follows: anti-mN1A, anti-Sirt1, and anti-Hes1(all at 1:500 dilution), and anti-Akt, anti-pAkt-Ser473, anti-PI3K, anti-actin, and anti-tubulin (all at 1:1000 dilution). For the immunopre-cipitations, Treg cells (4 × 106 cells) were lysed for 30 min at 4°C on arotational cell mixer in 1% NP-40 buffer (50 mM tris, 1 mM NaCl,1 mM EDTA) supplemented with the protease inhibitors describedearlier. After centrifugation at 1670g for 5 min, the supernatant wasincubated with 10 mg of anti-Notch1 primary antibody (clone mN1A)or IgG control for 1 hour at 4°C on a rotational cell mixer. WashedSepharose G plus beads (70 ml) were added to 0.5 to 1 ml of supernatant,and the suspension was held on a rotational cell mixer for 2 hours at4°C. Beads bound to complexes were washed five times with ice-cold


PBS by centrifugation at 150g. Finally, beads were boiled in SDS lysisbuffer for 10 min before use for Western blotting analysis.

Immunostaining and imaging by confocal andTIRF microscopyActivated Treg cells (0.3 × 106) were adhered to dishes coated withpoly-D-lysine (1 mg/ml in PBS for 15 min at room temperature) andfixed with 2% paraformaldehyde for 20 min at room temperature.Cells were permeabilized with 0.1% Tween 20 (for immunostainingwith the antibody to Val1744 or Foxp3) or 0.2% NP-40 (for the mNIAantibody clone) for 10 min at room temperature and then blocked with5% bovine serum albumin at room temperature. Samples were incu-bated overnight at 4°C with primary antibodies against the followingtargets: Val1744 antibody, which recognizes cleaved Notch1, PI3K,Rictor (1:100), mNIA, Sirt1 (1:50), and Foxp3 (1:100). The sampleswere then incubated for 1 hour at room temperature with the appro-priate secondary antibodies (1:1000). Confocal imaging was per-formed with the Zeiss 510 Meta confocal microscope with a 63×, 1.4numerical aperture (NA) objective, and Z stacks were taken at 1-mmthickness. TIRF microscopy imaging was performed with the NikonTE 2000 inverted microscope with a TIRF attachment for three laserlines [argon-ion (488 nm) and helium-neon lasers (543 and 633 nm)]equipped with a 100× oil objective (NA, 1.49) and a cooled charge-coupled device cascade camera (PhotoMetrics Inc.) with an ultimatepixel size of 75 nm. For two-color imaging, cells were illuminated withthe two laser lines (488 and 543 nm), and the camera was used in theelectron-multiplying gain mode.

Hes1 promoter reporter assayHEKcellswere transfectedwith 1mg of theHes1-Luc construct and 100ngof the Renilla-Luc construct in combination with 1 mg of the RFP, NIC-RFP, orNIC4KR-RFPplasmid. Eighteenhours before the cells were trans-fected, they were treated with or without 500 nM CHIC-35. Luciferaseassays were performed 36 hours after transfection with the Dual Lucif-erase Kit (Promega) according to the manufacturer’s instructions.

T cell suppressor assaysNaïve CD4+ T cells were isolated from the spleens of OT-II transgenicmice by negative selection with magnetic bead–based separationprotocols (MagCellect, R&D Systems). Freshly isolated naïve T cells(2 × 106 cells/ml in prewarmed PBS) were loaded with 5 mM CFSEand incubated for 8min at 37°C in a water bath. Cells were washedwithcomplete medium to remove excess dye before they were used in sup-pressor assays. CFSE-loaded naïve CD4+ OT-II T cells (0.5 × 106) wereco-injected with 0.3 × 106 activated Treg cells intravenously, into con-genic B6SJL hostmice. Twenty-four hours later, the recipientmice weresubcutaneously injected with 30 mg of maleylated OVA in Freund’scomplete adjuvant. Three days later, cells isolated from the lymphnodesof the recipientmicewere analyzed by flow cytometry for CFSE dilutionin the donor cells by gating on theCD45.2+ population. As described pre-viously (29), adoptive transfer of Treg cells into Foxp3-Cre::Notch1lox/lox

mice was based on a published protocol (36). Briefly, 1 × 106 activatedTreg cells expressing scrambled shRNA or Sirt1-specific shRNA wereadoptively transferred (intravenously) into 4- to 6-week-old Foxp3-Cre::Notch1lox/lox mice. Seven days later, lymph nodes (axillary andinguinal) from the recipient mice were harvested, and single-cell suspen-sions were prepared and assessed for T cell subsets. In the flow cytometricanalysis, lymphocyte populations were gated on the basis of size (for-ward scatter versus side scatter) and analyzed for cell surfacemarkers

11 of 13



htD

ownloaded from

by multicolor staining with fluorescently tagged antibodies on a BDFACSCalibur flow cytometer.

Intraperitoneal glucose tolerance testFor the intraperitoneal glucose tolerance test, 150 ml of D-glucose(Fischer Scientific, prepared in autoclaved water) was injected intra-peritoneally at a dose of 2 g/kg of mouse body weight into Leprdb/db

and Leprdb/+ mice, which had been fasted for 6 hours. Blood glucoseconcentrations were measured before the mice were injected with glu-cose. Blood samples were obtained from the tail tip at the times indi-cated in the figure legends, and blood glucose concentrations weremeasured using a handheld glucometer (Contour TS, blood glucosemeter, Bayer). In experiments involving the injection of Treg cells,0.8 × 106 activated Treg cells transduced with retroviruses expressingscrambled shRNA, Sirt1-specific shRNA, or NIC-NES, or the pBABEcontrol vector were used, as described earlier. The IPGTT was per-formed on day 8 after the transfer of cells.

Statistical analysis and data presentationData are means ± SD from three independent experiments, unlessstated otherwise. Statistical significance was calculated with thetwo-population Student’s t test. Western blotting data were processedwith Adobe Photoshop software. Image analysis was performed withImageJ software, and figures were processed and prepared with AdobePhotoshop.
tp://stke.sciencem
a

SUPPLEMENTARY MATERIALSwww.sciencesignaling.org/cgi/content/full/10/473/eaah4679/DC1Fig. S1. Inhibition of Sirt1 with CHIC-35 affects NIC localization in Treg cells.Fig. S2. Knockdown of Sirt1 affects the localization of NIC.Fig. S3. Loss of Sirt1 inhibits the anti-apoptotic function of NIC in response to multiple stimuli.Fig. S4. NIC4KR is resistant to modification by Sirt1.

on April 4, 2017

g.org/
REFERENCES AND NOTES1. K. W. Moremen, M. Tiemeyer, A. V. Nairn, Vertebrate protein glycosylation: Diversity,
synthesis and function. Rev. Mol. Cell Biol. 13, 448–462 (2012).2. A. Bononi, C. Agnoletto, E. De Marchi, S. Marchi, S. Patergnani, M. Bonora, C. Giorgi,

S. Missiroli, F. Poletti, A. Rimessi, P. Pinton, Protein kinases and phosphatases in thecontrol of cell fate. Enzyme Res. 2011, 329098 (2011).

3. M. Haberland, R. L. Montgomery, E. N. Olson, The many roles of histone deacetylasesin development and physiology: Implications for disease and therapy. Nat. Rev. Genet. 10,32–42 (2009).

4. K. K. Biggar, S. S.-C. Li, Non-histone protein methylation as a regulator of cellularsignalling and function. Nat. Rev. Mol. Cell Biol. 16, 5–17 (2015).

5. F. Yeung, J. E. Hoberg, C. S. Ramsey, M. D. Keller, D. R. Jones, R. A. Frye, M. W. Mayo,Modulation of NF-kB-dependent transcription and cell survival by the SIRT1 deacetylase.EMBO J. 23, 2369–2380 (2004).

6. V. Guarani, G. Deflorian, C. A. Franco, M. Krüger, L.-K. Phng, K. Bentley, L. Toussaint,F. Dequiedt, R. Mostoslavsky, M. H. H. Schmidt, B. Zimmermann, R. P. Brandes,M. Mione, C. H. Westphal, T. Braun, A. M. Zeiher, H. Gerhardt, S. Dimmeler, M. Potente,Acetylation-dependent regulation of endothelial Notch signalling by the SIRT1deacetylase. Nature 473, 234–238 (2011).

7. P. Mulligan, F. Yang, L. Di Stefano, J.-Y. Ji, J. Ouyang, J. L. Nishikawa, D. Toiber, M. Kulkarni,Q. Wang, S. H. Najafi-Shoushtari, R. Mostoslavsky, S. P. Gygi, G. Gill, N. J. Dyson, A. M. Näär,A SIRT1-LSD1 corepressor complex regulates Notch target gene expression anddevelopment. Mol. Cell 42, 689–699 (2011).

8. H. Vaziri, S. K. Dessain, E. Ng Eaton, S.-I. Imai, R. A. Frye, T. K. Pandita, L. Guarente,R. A. Weinberg, hSIR2SIRT1 functions as an NAD-dependent p53 deacetylase. Cell 107,149–159 (2001).

9. J. Luo, A. Y. Nikolaev, S.-i. Imai, D. Chen, F. Su, A. Shiloh, L. Guarente, W. Gu, Negative controlof p53 by Sir2a promotes cell survival under stress. Cell 107, 137–148 (2001).

10. H.-S. Kwon, H. W. Lim, J. Wu, M. Schnölzer, E. Verdin, M. Ott, Three novel acetylation sitesin the Foxp3 transcription factor regulate the suppressive activity of regulatory T cells.J. Immunol. 188, 2712–2721 (2012).


11. J. van Loosdregt, D. Brunen, V. Fleskens, C. E. G. M. Pals, E. W. F. Lam, P. J. Coffer, Rapidtemporal control of Foxp3 protein degradation by sirtuin-1. PLOS ONE 6, e19047(2011).

12. S. Hisahara, S. Chiba, H. Matsumoto, M. Tanno, H. Yagi, S. Shimohama, M. Sato, Y. Horio,Histone deacetylase SIRT1 modulates neuronal differentiation by its nucleartranslocation. Proc. Natl. Acad. Sci. U.S.A. 105, 15599–15604 (2008).

13. L. Tiberi, J. van den Ameele, J. Dimidschstein, J. Piccirilli, D. Gall, A. Herpoel, A. Bilheu,J. Bonnefont, M. Iacovino, M. Kyba, T. Bouschet, P. Vanderhaeghen, BCL6 controlsneurogenesis through Sirt1-dependent epigenetic repression of selective Notch targets.Nat. Neurosci. 15, 1627–1635 (2012).

14. M. Xie, M. Liu, C.-S. He, SIRT1 regulates endothelial Notch signaling in lung cancer.PLOS ONE 7, e45331 (2012).

15. F. Radtke, A. Wilson, G. Stark, M. Bauer, J. van Meerwijk, H. R. MacDonald, M. Aguet,Deficient T cell fate specification in mice with an induced inactivation of Notch1.Immunity 10, 547–558 (1999).

16. A. Wolfer, T. Bakker, A. Wilson, M. Nicolas, V. Ioannidis, D. R. Littman, C. B. Wilson, W. Held,H. R. MacDonald, F. Radtke, Inactivation of Notch 1 in immature thymocytes does notperturb CD4 or CD8 T cell development. Nat. Immunol. 2, 235–241 (2001).

17. A. Dongre, L. Surampudi, R. G. Lawlor, A. H. Fauq, L. Miele, T. E. Golde, L. M. Minter,B. A. Osborne, Non-canonical Notch signaling drives activation and differentiation ofperipheral CD4+ T cells. Front. Immunol. 5, 54 (2014).

18. Y. Maekawa, C. Ishifune, S.-i. Tsukumo, K. Hozumi, H. Yagita, K. Yasutomo, Notch controlsthe survival of memory CD4+ T cells by regulating glucose uptake. Nat. Med. 21, 55–61(2015).

19. K. Laky, S. Evans, A. Perez-Diez, B. J. Fowlkes, Notch signaling regulates antigen sensitivityof naive CD4+ T cells by tuning co-stimulation. Immunity 42, 80–94 (2015).

20. N. Asano, T. Watanabe, A. Kitani, I. J. Fuss, W. Strober, Notch1 signaling and regulatoryT cell function. J. Immunol. 180, 2796–2804 (2008).

21. J. B. Samon, A. Champhekar, L. M. Minter, J. C. Telfer, L. Miele, A. Fauq, P. Das, T. E. Golde,B. A. Osborne, Notch1 and TGFb1 cooperatively regulate Foxp3 expression and themaintenance of peripheral regulatory T cells. Blood 112, 1813–1821 (2008).

22. H.-F. Ou-Yang, H.-W. Zhang, C.-G. Wu, P. Zhang, J. Zhang, J.-C. Li, L.-H. Hou, F. He, X.-Y. Ti,L.-Q. Song, S.-Z. Zhang, L. Feng, H.-W. Qi, H. Han, Notch signaling regulates the FOXP3promoter through RBP-J- and Hes1-dependent mechanisms. Mol. Cell. Biochem. 320,109–114 (2009).

23. C. Mota, V. Nunes-Silva, A. R. Pires, P. Matoso, R. M. M. Victorino, A. E. Sousa, I. Caramalho,Delta-like 1-mediated Notch signaling enhances the in vitro conversion of humanmemory CD4 T cells into FOXP3-expressing regulatory T cells. J. Immunol. 193,5854–5862 (2014).

24. R. Bassil, B. Zhu, Y. Lahoud, L. V. Riella, H. Yagita, W. Elyaman, S. J. Khoury, Notch liganddelta-like 4 blockade alleviates experimental autoimmune encephalomyelitis bypromoting regulatory T cell development. J. Immunol. 187, 2322–2328 (2011).

25. L. R. Perumalsamy, N. Marcel, S. Kulkarni, F. Radtke, A. Sarin, Distinct spatial andmolecular features of notch pathway assembly in regulatory T cells. Sci. Signal. 5, ra53(2012).

26. Y. Kanfi, V. Peshti, Y. M. Gozlan, M. Rathaus, R. Gil, H. Y. Cohen, Regulation of SIRT1 proteinlevels by nutrient availability. FEBS Lett. 582, 2417–2423 (2008).

27. S. Nemoto, M. M. Fergusson, T. Finkel, Nutrient availability regulates SIRT1 through aforkhead-dependent pathway. Science 306, 2105–2108 (2004).

28. A. Chalkiadaki, L. Guarente, Sirtuins mediate mammalian metabolic responses to nutrientavailability. Nat. Rev. Endocrinol. 8, 287–296 (2012).

29. N. Marcel, A. Sarin, Notch1 regulated autophagy controls survival and suppressor activityof activated murine T-regulatory cells. eLife 5, e14023 (2016).

30. L. R. Perumalsamy, M. Nagala, P. Banerjee, A. Sarin, A hierarchical cascade activated bynon-canonical Notch signaling and the mTOR-Rictor complex regulates neglect-induceddeath in mammalian cells. Cell Death Differ. 16, 879–889 (2009).

31. H. Sade, S. Krishna, A. Sarin, The anti-apoptotic effect of Notch-1 requiresp56lck-dependent, Akt/PKB-mediated signaling in T cells. J. Biol. Chem. 279,2937–2944 (2004).

32. L. R. Perumalsamy, M. Nagala, A. Sarin, Notch-activated signaling cascade interacts withmitochondrial remodeling proteins to regulate cell survival. Proc. Natl. Acad. Sci. U.S.A.107, 6882–6887 (2010).

33. P. E. Czabotar, G. Lessene, A. Strasser, J. M. Adams, Control of apoptosis by the BCL-2protein family: Implications for physiology and therapy. Nat. Rev. Mol. Cell Biol. 15, 49–63(2014).

34. L.-M. Charbonnier, S. Wang, P. Georgiev, E. Sefik, T. A. Chatila, Control of peripheraltolerance by regulatory T cell–intrinsic Notch signaling. Nat. Immunol. 16, 1162–1173(2015).

35. S. Hori, T. Nomura, S. Sakaguchi, Control of regulatory T cell development by thetranscription factor Foxp3. Science 299, 1057–1061 (2003).

36. J. D. Fontenot, M. A. Gavin, A. Y. Rudensky, Foxp3 programs the development andfunction of CD4+CD25+ regulatory T cells. Nat. Immunol. 4, 330–336 (2003).

12 of 13

http://www.sciencesignaling.org/cgi/content/full/10/473/eaah4679/DC1



http://stke.sciencemag.

Dow

nloaded from

37. L. Klein, K. Khazaie, H. von Boehmer, In vivo dynamics of antigen-specific regulatoryT cells not predicted from behavior in vitro. Proc. Natl. Acad. Sci. U.S.A. 100, 8886–8891(2003).

38. B. J. C. Quah, H. S. Warren, C. R. Parish, Monitoring lymphocyte proliferation in vitro and invivo with the intracellular fluorescent dye carboxyfluorescein diacetate succinimidylester. Nat. Protoc. 2, 2049–2056 (2007).

39. D. Cipolletta, D. Kolodin, C. Benoist, D. Mathis, Tissular Tregs: A unique population ofadipose-tissue-resident Foxp3+CD4+ T cells that impacts organismal metabolism.Semin. Immunol. 23, 431–437 (2011).

40. M. Feuerer, L. Herrero, D. Cipolletta, A. Naaz, J. Wong, A. Nayer, J. Lee, A. B. Goldfine,C. Benoist, S. Shoelson, D. Mathis, Lean, but not obese, fat is enriched for a uniquepopulation of regulatory T cells that affect metabolic parameters. Nat. Med. 15, 930–939(2009).

41. M. E. Brunkow, E. W. Jeffery, K. A. Hjerrild, B. Paeper, L. B. Clark, S.-A. Yasayko,J. E. Wilkinson, D. Galas, S. F. Ziegler, F. Ramsdell, Disruption of a new forkhead/winged-helix protein, scurfin, results in the fatal lymphoproliferative disorder of the scurfymouse. Nat. Genet. 27, 68–73 (2001).

42. S. Winer, Y. Chan, G. Paltser, D. Truong, H. Tsui, J. Bahrami, R. Dorfman, Y. Wang,J. Zielenski, F. Mastronardi, Y. Maezawa, D. J. Drucker, E. Engleman, D. Winer, H.-M. Dosch,Normalization of obesity-associated insulin resistance through immunotherapy.Nat. Med. 15, 921–929 (2009).

43. A. L. Penton, L. D. Leonard, N. B. Spinner, Notch signaling in human development anddisease. Semin. Cell Dev. Biol. 23, 450–457 (2012).

44. J. C. Aster, In brief: Notch signalling in health and disease. J. Pathol. 232, 1–3 (2014).45. E.-B. Braune, U. Lendahl, Notch—A goldilocks signaling pathway in disease and cancer

therapy. Discov. Med. 21, 189–196 (2016).46. K. Hori, A. Sen, S. Artavanis-Tsakonas, Notch signaling at a glance. J. Cell Sci. 126,

2135–2140 (2013).47. E. R. Andersson, R. Sandberg, U. Lendahl, Urban Notch signaling: Simplicity in design,

versatility in function. Development 138, 3593–3612 (2011).48. K. G. Guruharsha, M. W. Kankel, S. Artavanis-Tsakonas, The Notch signalling system:

Recent insights into the complexity of a conserved pathway. Nat. Rev. Genet. 13, 654–666(2012).

49. M. E. Fortini, Notch Signaling: The core pathway and its posttranslational regulation.Dev. Cell 16, 633–647 (2009).

50. E. Michishita, J. Y. Park, J. M. Burneskis, J. C. Barrett, I. Horikawa, Evolutionarily conservedand nonconserved cellular localizations and functions of human SIRT proteins.Mol. Biol. Cell 16, 4623–4635 (2005).

51. M. Tanno, J. Sakamoto, T. Miura, K. Shimamoto, Y. Horio, Nucleocytoplasmic shuttling ofthe NAD+-dependent histone deacetylase SIRT1. J. Biol. Chem. 282, 6823–6832 (2006).


52. K.-S. Lee, Z. Wu, Y. Song, S. S. Mitra, A. H. Feroze, S. H. Cheshier, B. Lu, Roles of PINK1,mTORC2, and mitochondria in preserving brain tumor-forming stem cells in anoncanonical Notch signaling pathway. Genes Dev. 27, 2642–2647 (2013).

53. P. Andersen, H. Uosaki, L. T. Shenje, C. Kwon, Non-canonical Notch signaling: Emergingrole and mechanism. Trends Cell Biol. 22, 257–265 (2012).

54. K. Hein, G. Mittler, W. Cizelsky, M. Kühl, F. Ferrante, R. Liefke, I. M. Berger, S. Just,J. E. Sträng, H. A. Kestler, F. Oswald, T. Borggrefe, Site-specific methylation of Notch1controls the amplitude and duration of the Notch1 response. Sci. Signal. 8, ra30 (2015).

55. B. Giebel, A. Wodarz, Notch signaling: Numb makes the difference. Curr. Biol. 22,R133–R135 (2012).

Acknowledgments: We thank B. A. Osborne (University of Massachusetts, Amherst, USA) andJ. Aster for the gifts of plasmids. We acknowledge F. Radtke (EPFL, Laussane, Switzerland)for the gifts of the Notch1lox/lox and Cd4-Cre::Notch1lox/lox mice. The pCL-Eco construct was agift from I. Verma (Addgene, plasmid #12371). The Sirt1 and Sirt1_H363Y constructs weregifts from M. Greenberg (Addgene, plasmids #1791 and #1792). The LTG human fibroblastcell line was a gift from A. Rangarajan (IISC, Bengaluru, India). We acknowledge the AnimalCare and Resource Centre and the Central Imaging and Flow Cytometry Facility at theNational Centre for Biological Sciences (NCBS) and Institute for Stem Cell Biology andRegenerative Medicine, Bengaluru. Funding: The Animal Care and Resource Centre wasfunded in part by the National Mouse Research Resource [grant BT/PR5981/MED/31/181/2012;2013–2016 from the Department of Biotechnology (DBT)]. This work was funded byBT/PR13446/COE/34/30/2015 and BT/PR11751/AGR/36/606/2008 from the DBT, India, toA.S.; core funds from NCBS to A.S.; a Senior Research Fellowship (09/860[0137]/2012-EMR-1)from the Council of Scientific and Industrial Research, India, to N.M.; and University GrantsCommission to S.K.S. Author contributions: N.M. performed the confocal analysis,biochemical experiments, and the suppressor assays with Treg cells. L.R.P. performed theTIRF analysis and the initial functional characterization in Treg cells and in cell lines.S.K.S. generated and characterized the NIC mutants in cell lines. A.S., N.M., and L.R.P.contributed to the study design and the writing of the manuscript. Competing interests: Theauthors declare that they have no competing interests.

Submitted 30 June 2016Accepted 21 March 2017Published 4 April 201710.1126/scisignal.aah4679

Citation: N. Marcel, L. R. Perumalsamy, S. K. Shukla, A. Sarin, The lysine deacetylase Sirtuin1 modulates the localization and function of the Notch1 receptor in regulatory T cells. Sci.Signal. 10, eaah4679 (2017).
org
13 of 13

on April 4, 2017

/


(473), . [doi: 10.1126/scisignal.aah4679]10Science Signaling Apurva Sarin (April 4, 2017) Nimi Marcel, Lakshmi R. Perumalsamy, Sanjay K. Shukla andfunction of the Notch1 receptor in regulatory T cellsThe lysine deacetylase Sirtuin 1 modulates the localization and

This information is current as of April 4, 2017. The following resources related to this article are available online at http://stke.sciencemag.org.

Article Tools

http://stke.sciencemag.org/content/10/473/eaah4679article tools: Visit the online version of this article to access the personalization and

MaterialsSupplemental

http://stke.sciencemag.org/content/suppl/2017/03/31/10.473.eaah4679.DC1"Supplementary Materials"

Related Content

http://stm.sciencemag.org/content/scitransmed/7/315/315ec204.fullhttp://science.sciencemag.org/content/sci/347/6224/847.fullhttp://science.sciencemag.org/content/sci/355/6331/1320.fullhttp://stke.sciencemag.org/content/sigtrans/9/426/ra46.fullhttp://stke.sciencemag.org/content/sigtrans/8/386/ra71.fullhttp://stke.sciencemag.org/content/sigtrans/10/466/eaai8026.fullhttp://stke.sciencemag.org/content/sigtrans/7/351/ra106.full

's sites:ScienceThe editors suggest related resources on

Referenceshttp://stke.sciencemag.org/content/10/473/eaah4679#BIBLThis article cites 55 articles, 18 of which you can access for free at:

Permissionshttp://www.sciencemag.org/about/permissions.dtlObtain information about reproducing this article:

reserved. DC 20005. Copyright 2017 by the American Association for the Advancement of Science; all rightsAmerican Association for the Advancement of Science, 1200 New York Avenue, NW, Washington,

(ISSN 1937-9145) is published weekly, except the last December, by theScience Signaling

on April 4, 2017


Dow

nloaded from

http://stke.sciencemag.org/content/10/473/eaah4679

http://stke.sciencemag.org/content/suppl/2017/03/31/10.473.eaah4679.DC1

http://stke.sciencemag.org/content/sigtrans/7/351/ra106.full

http://stke.sciencemag.org/content/sigtrans/10/466/eaai8026.full



http://science.sciencemag.org/content/sci/355/6331/1320.full

http://science.sciencemag.org/content/sci/347/6224/847.full

http://stm.sciencemag.org/content/scitransmed/7/315/315ec204.full

http://stke.sciencemag.org/content/10/473/eaah4679#BIBL

http://www.sciencemag.org/about/permissions.dtl


the lysine deacetylase sirtuin 1 modulates the ... · the lysine deacetylase sirtuin 1 modulates...

Documents