Selective Inhibition of Regulatory T Cells by Targeting the PI3K–Akt

Research Article

Selective Inhibition of Regulatory T Cells by Targeting thePI3K–Akt Pathway

Rasha Abu-Eid1, Raed N. Samara2, Laurent Ozbun2, Maher Y. Abdalla2, Jay A. Berzofsky2,Kevin M. Friedman3, Mikayel Mkrtichyan1, and Samir N. Khleif1

AbstractDespite the strides that immunotherapy has made in mediating tumor regression, the clinical effects are often

transient, and therefore more durable responses are still needed. The temporary nature of the therapy-inducedimmune response can be attributed to tumor immune evasion mechanisms, mainly the effect of suppressiveimmune cells and, in particular, regulatory T cells (Treg). Although the depletion of Tregs has been shown to beeffective in enhancing immune responses, selective depletion of these suppressive cells without affecting otherimmune cells has not been very successful, and new agents are sought. We found that PI3K–Akt pathwayinhibitors selectively inhibit Tregs with minimal effect on conventional T cells (Tconv). Our results clearly showselective in vitro inhibition of activation (as represented by a decrease in downstream signaling) and proliferationof Tregs in comparison with Tconvs when treated with different Akt and PI3K inhibitors. This effect has beenobserved in both human and murine CD4 T cells. In vivo treatment with these inhibitors resulted in a significantand selective reduction in Tregs in both na€�ve and tumor-bearing mice. Furthermore, these PI3K–Akt inhibitorsled to a significant therapeutic antitumor effect, whichwas shown to beTreg dependent. Here, we report the use ofPI3K–Akt pathway inhibitors as potent agents for the selective depletion of suppressive Tregs.We show that theseinhibitors are able to enhance the antitumor immune response and are therefore promising clinical reagents forTreg depletion. Cancer Immunol Res; 2(11); 1080–9. �2014 AACR.

IntroductionCancer immunotherapy has proved successful in mediating

disease regression in patients with cancer. Several strategieshave been used effectively to generate therapeutic tumorantigen-reactive T-cell responses. These include active immu-nization (1–5), adoptive cell transfer (ACT) of lymphocytesgenetically engineered for antitumor function (6–9), and ACTof tumor-infiltrating lymphocytes (TIL; refs. 10–16). Despitethese successes, many responses are transient, and improve-ments are needed to increase durable responses in patients.

A suppressive tumor immune environment is thought tolimit the efficacy of ACT and active immunization approaches.Chief among the suppressive cells are CD4 regulatory T cells(Treg) that can be distinguished by the expression of FoxP3 andCD25HI molecules (17). An opposing action of Tregs and

cytotoxic CD8 T cells in tumor regression has been identifiedinmousemodels (18). A correlation of tumor-infiltrating Tregswith poor clinical prognosis has been demonstrated in humans(19–22). In fact, the depletion of Tregs was found to enhanceantitumor immunity and promote tumor regression in mousemodels (23–26).

Despite the evidence of the suppressive role of Tregs in thetumor microenvironment (TME), there is still a paucity ofTreg-depleting clinical reagents. Concerns over efficacy andspecificity have burdened currently available clinical reagentssuch as low-dose cyclophosphamide and the anti-CD25mono-clonal antibody (mAb) daclizumab (27). Novel clinical reagentsthat specifically target Treg persistence and survival are can-didates for tumor immunomodulation.

Several reports have shown that Tregs and conventional Tcells (Tconv) display unique signaling signatures downstreamof theT-cell receptor (TCR; refs. 28–31). The PI3K–Akt pathwayplays a critical role in the cellular response to TCR engagementand costimulation (32, 33). Active PI3K–Akt signaling results inincreased cytokine gene expression, a characteristic of pro-ductive T-cell activation (34). In addition to transmittingsignals critical for T-cell activation, many T-cell functions aregoverned by PI3K–Akt signaling, which includes proliferation,survival, migration, and metabolism (35, 36).

Because of the important role of the PI3K–Akt pathway in T-cell function (35, 36) and the reported differences betweenTregs and Tconvs downstream of the TCR (28–31), we eval-uated the effect of PI3K–Akt inhibition on Tregs and Tconvs.The impact of Akt and PI3K inhibitors on Treg and Tconv

1Georgia Regents University Cancer Center, Augusta, Georgia. 2Nation-al Cancer Institute, National Institutes of Health, Bethesda, Maryland.3bluebird bio, Cambridge, Massachusetts.

R. Abu-Eid and R.N. Samara contributed equally to this article.

Current address for R.N. Samara: Qiagen, Frederick, Maryland; currentaddress for M.Y. Abdalla, University of Nebraska Medical Center, Omaha,Nebraska.

Corresponding Author: Samir N. Khleif, Georgia Regents UniversityCancer Center, 1410 LaneyWalker Boulevard, Augusta, GA 30912. Phone:706-721-0570; Fax: 706-721-8787; E-mail: [email protected]

doi: 10.1158/2326-6066.CIR-14-0095

�2014 American Association for Cancer Research.

CancerImmunology

Research

Cancer Immunol Res; 2(11) November 20141080

on March 21, 2018. © 2014 American Association for Cancer Research. cancerimmunolres.aacrjournals.org Downloaded from

Published OnlineFirst July 30, 2014; DOI: 10.1158/2326-6066.CIR-14-0095

http://cancerimmunolres.aacrjournals.org/

activation and proliferation is assessed in vitro. We have alsoevaluated the effect and therapeutic efficacy of in vivo treat-mentwith these inhibitors on the antitumor immune response.

Materials and MethodsMice and cell linesFemale C57BL/6(H-2b) and BALB/c mice (6–10-week-old;

NCI, Frederick, MD) were housed under pathogen-free condi-tions. All procedures were carried out with approved institu-tional animal protocols. B16, CT26, and EL4 cell lines wereobtained from theATCC,which routinely authenticate and testthese cell lines (forMycoplasma, by theHoechst stain, PCR, andthe standard culture test). These cells were used within 6months of purchase. TC-1 (established by immortalizationwith the HPV16 E6 and E7 genes and its growth enhanced byTregs; refs. 37, 38) was a gift from Prof. T.C. Wu (Johns HopkinsUniversity, Department of Pathology, Baltimore, MD). Thesecells, along with B16, were authenticated and tested for mouseparvovirus (MPV) andmouse hepatitis virus (MHV) using PCRat Georgia Regents University (Augusta, GA). All tests werenegative.

ReagentsThe PI3K inhibitor wortmannin and the Akt inhibitor tricir-

ibine were obtained from Calbiochem. IC87114, a PI3K inhib-itor, and MK-2206, an Akt inhibitor, were purchased fromSelleckChem. The 9-mer synthetic peptide from HPV16E749–57, RAHYNIVTF, was obtained from Celtek Bioscience.E749–57 (100 mg/mouse) was used as a vaccine along with GM-CSF (5 mg/mouse; PeproTech), anti-CD40 (20 mg/mouse; Bio-Legend), and Incomplete Freund Adjuvant (IFA; 50 mL/mouse;Sigma). This was reported as the most effective therapeuticcombination for this vaccine (39).

Human T-cell culturesLeukapheresis products were obtained from healthy human

donors (Department of Transfusion Medicine, NIH, Bethesda,MD). Peripheral blood mononuclear cells (PBMC) were pre-pared over Ficoll-Paque Plus gradient centrifugation (GEHealthcare), and CD4þCD25HI and CD4þCD25� cells weresorted using the FACSAria II flow cytometer. The cells werethen labeled with CFSE (Life Technologies) according to themanufacturer's instructions. Fifty thousand cells were culturedwith anti-CD3/CD28–conjugated Dynabeads (Life Technolo-gies) at a 4:1 cell-to-bead ratio in RPMI-1640 supplementedwith 5% autologous serum and 100 U/mL IL2 (PeproTech) for 3days, in the presence or absence of escalating concentrations ofinhibitors. CFSE dilution was then assessed by flow cytometry.

Murine CD4 T-cell culturesMagnetic bead purification kits (Miltenyi Biotec) were used

to enrich CD4þCD25� and CD4þCD25þ T cells from murinesplenocytes following the manufacturer's instructions. Cellswere labeled with CFSE (Life Technologies) and cultured in 24-well plates at a density of 5 � 105 cells per well in RPMI-1640(Life Technologies) with 10% FCS in the presence of 10 mg/mLplate-bound anti-CD3 (BD Biosciences), 1 mg/mL soluble anti-

CD28 (BD Biosciences), and 100 IU/mL IL2 (R&D Systems).Plates were centrifuged and then incubated at 37�C, 5%CO2 for72 hours. Wortmannin (200 nmol/L), MK-2206 (2 mmol/L),IC87114 (10 mmol/L), or DMSO (carrier) were added to theculture media from the beginning. CFSE dilution was mea-sured by flow cytometry.

The phosphorylation level of S6 was assessed. Murine cellswere prepared as described above and stimulated for 15minutes. Thirty micrograms of cell lysates in RIPA buffer wasthen run on SDS-PAGE gels, transferred to polyvinylidenefluoride (PVDF) membranes probed with primary antibodies(1:1,000; anti-pS6 and anti-S6; Cell Signaling Technology) over-night at 4�C, and incubatedwith secondary antibodies (1:2,000)for 1 hour at room temperature. Chemiluminescence wasperformed with Pierce reagents, and densitometric analysiswas performed using ImageJ (NIH).

In vivo experiments to assess splenocyte compositionTumor-free na€�ve mice were injected intraperitoneally (i.p.)

with 40 mg of wortmannin, 50 mg of triciribine, or 10 mg of MK-2206 dissolved in 35% DMSO in PBS (100-mL volume). Micewere injected every other day for aweek. Splenocyteswere thenharvested, andCD4þ and Foxp3þ compositionwas assessed byflow cytometry.

IFNg ELISPOTTumor-free na€�ve mice were injected i.p. on alternate days

with 40mg of wortmannin, 50mg of triciribine, or DMSO vehiclefor a week before vaccination with E7 vaccine (E749–57, GM-CSF, anti-CD40, and IFA), which was given subcutaneously (s.c.) on days 7 and 14. One week after the second E7 vaccination,splenocytes were harvested, and the anti-E7 immune responsewas assayed by peptide restimulation and IFNg ELISPOT (BDBiosciences), according to the manufacturer's instructions.

Tumor treatmentC57BL/6 female mice were implanted with 50,000 TC-1 cells

per mouse s.c. in the right flank on day 0. The mice were thentreated with 40 mg of wortmannin or DMSO for 1 week afterpalpable tumors were detected. Mice were then vaccinatedwith E7 peptide vaccine (E749–57, GM-CSF, anti-CD40, and IFA),as described above, and tumor growth was monitored. Thesame model was used with MK-2206, in which mice werechallenged with TC-1 cells on day 0, and on days 7 and 14,mice in the appropriate groups were injected with the E7vaccine and/or MK-2206 (30 mg). The tumors were measuredand the mice were euthanized on day 21. The tumor immuneinfiltrate was then assessed by flow cytometry. A prophylactictherapeutic model was also used without vaccines. Threetumor models were used: B16 (200,000 cells/mouse, s.c.) andEL4 (100,000 cells/mouse s.c.), both in C57BL/6mice, and CT26(500,000 cells/mouse, s.c.) in BALB/c mice. Mice were treatedwith 40 mg of wortmannin, 50 mg of triciribine, or DMSO (i.p.)on days �7, �4, and �2 before s.c. tumor inoculation onday 0, and tumor growth was monitored thereafter. Ex vivo–activatedTregs (cellswere stimulated as described above for 72hours) were infused i.v. (10,000 cells/mouse) into wortmannin-treated EL4 and CT26 tumor–bearing animals on day 4 after

Foxp3þ Tregs Are Dependent on the PI3K–Akt Pathway

www.aacrjournals.org Cancer Immunol Res; 2(11) November 2014 1081




tumor inoculation, and tumor growth was also monitored.FoxP3þ T-cell infiltration into CT26 tumor was assessed byflow cytometry on days 20 and 24 following tumor inoculation.

Flow cytometry analysisStaining for surface markers (CD4 and CD8) was performed

using mAbs (BD Biosciences). Cells were incubated for 20minutes on ice in PBS, 2% BSA, and 0.1% sodium azide.Intracellular staining kits were used to stain for Foxp3 usinganti–Foxp3-PE mAb (eBioscience). Data acquisition was per-formed on FACSCalibur or FACScan cytometers (BD Bios-ciences). Results were analyzed with CellQuest (BD Bios-ciences), WinMDI (Purdue University, West Lafayette, IN), orFlowJo (TreeStar).

Statistical analysisAll statistical parameters (average values, SD, significant

differences between groups) were calculated using GraphPadPrism software. Statistical significance between groups wasdetermined by a paired t test or ANOVA with a post hoc Tukeymultiple comparison test (P < 0.05 was considered statisticallysignificant).

ResultsHuman CD4 Tregs are more dependent on PI3K–Aktsignaling for TCR-induced proliferation thanconventional CD4 T cells

Because of the vital role that the PI3K–Akt pathway plays inmany T-cell functions, including proliferation and survival, weevaluated the effect of inhibiting this pathway on human CD4þ

T-cell proliferation. CD4þ Tregs were identified as CD4þ

CD25HI, whereas conventional CD4þ Tconvs were identified asCD4þCD25�. Tregs and Tconvs were fractionated fromhumanPBMCs by fluorescence-activated cell sorting. Proliferationwas examined by dilution of CFSE after stimulation withanti-CD3/anti-CD28 in a 3-day culture in media containing100 IU/mL of IL2, with titrated amounts of inhibitors selectivefor members of the PI3K–Akt pathway.

We evaluated the effect of Akt inhibition on proliferationusing the pan-Akt inhibitors triciribine and MK-2206. A highersensitivity of Tregs to proliferative inhibition by both inhibitorscompared with Tconvs is shown in Fig. 1A and B. The averageof three experiments revealed a significant inhibition of Tregproliferation comparedwith Tconvs by triciribine (Fig. 1A) andMK-2206 (Fig. 1B) at all doses tested.

To confirm that this selective effect is not unique to Aktinhibition, we evaluated the effect of upstream PI3K inhibitionby wortmannin and IC87114 on the proliferation of Tregs andTconvs. Results consistent with a higher sensitivity of Tregswere observed. The average of three experiments demonstrat-ed that proliferation of Tregs was significantly inhibited bywortmannin comparedwith Tconvs at all doses tested (Fig. 1C)and by IC87114 at the 10 and 20 mmol/L doses (Fig. 1D).

To rule out selective drug toxicity as an explanation for theabove findings, we next evaluated the effect of PI3K–Aktpathway inhibition on CD4 T-cell subset viability. The sameTregs and Tconvs examined above were assayed for viability by

7-AAD incorporation. Culture viability was inversely related tothe amount of inhibitor present across the dose range used;however, the viabilitywas reduced equally in Tregs andTconvs.This was observed with all inhibitors (wortmannin, IC87114,triciribine, and MK-2206; data not shown).

Taken together, our data demonstrate that human Tregs aremore dependent on the PI3K–Akt signaling pathway for pro-liferation in response to anti-CD3/anti-CD28/IL2 stimulationcompared with Tconvs.

PI3K–Akt pathway is necessary for CD4 Tregmaintenance in mice

The documented dependence of murine T-cell proliferationand other vital functions on the PI3K–Akt pathway, in additionto the necessity of this pathway in human Treg proliferationshown above, prompted a similar evaluation in murine Tregproliferation. Mouse CD4þ splenocytes were fractionated intoCD25þ (Treg) andCD25� (Tconv) subsets usingmagnetic beadenrichment kits. After enrichment, Tregs and Tconvs were>92% and >95% pure, respectively, based on CD4 and FoxP3expression (data not shown). Proliferation was assayed bydilution of CFSE after stimulation with anti-CD3/anti-CD28and 3-day culture in media containing 100 IU/mL IL2 withwortmannin, MK-2206, or IC87114 (triciribine was found to betoxic to murine T cells in these prolonged in vitro cultureconditions). Similar to results obtainedwith human CD4 T-cellsubsets, data from at least two independent experimentsdemonstrated that mouse Tregs were significantly more sen-sitive to proliferative inhibition by wortmannin (200 nmol/L),IC87114 (10 mmol/L), and MK-2206 (2 mmol/L) than Tconvs(Fig. 2A).

Inhibition of PI3K–Akt signaling by inhibitors in Tregs wasalso confirmed by Western blot analysis, in which phosphor-ylation of S6 was used to identify active signaling through thePI3K–Akt pathway. Confirming the inhibition of PI3K–Aktsignaling, treatment of anti-CD3/anti-CD28/IL2–stimulatedTregs with wortmannin, triciribine, IC87114, or MK-2206resulted in a marked decrease of phosphorylated S6 in com-parison with Tconvs (Fig. 2B and C).

As a result of the significant in vitro difference between theproliferation inhibition of Tregs and Tconvs by PI3K and Aktinhibitors, the role of PI3K–Akt signaling in the in vivo cellularcomposition of spleens was examined. In these analyses, micewere treated with wortmannin (40 mg), triciribine (50 mg),MK-2206 (10 mg), or DMSO for 1 week on alternate days beforeflow cytometry analysis of their splenocyte composition. Micetreated with DMSO contained a percentage of CD4 T cellssimilar to wortmannin-, triciribine-, or MK-2206–treated mice(Fig. 3A). Similarly, spleens of DMSO and inhibitor-treatedmice exhibited no differences in the percentages of CD8 T cells(data not shown). Importantly, despite not affecting total CD4T cells, the percentage of FoxP3þ cells within the CD4þ T-cellpopulation was significantly reduced in mice treated withwortmannin, triciribine, or MK-2206 (Fig. 3B).

The in vitro dependence of Treg proliferation on the PI3K–Akt pathway, coupled with the in vivo reduction in the actualnumbers of these suppressive cells in response to PI3K and Aktinhibition, led us to evaluate the effect of these inhibitors on

Abu-Eid et al.

Cancer Immunol Res; 2(11) November 2014 Cancer Immunology Research1082




120 TconvsA

B

C

D

Tregs

TconvsTregs

TconvsTregs

TconvsTregs

Tconvs

DMSO

93

95.5

63.6

91.4

67.4 39.5 33.7 29.2 6.74

88.4 85.2 85.3 7.62

68 48.8 20.3 4.32

95.7 93.6 87.6 9.02

83.2 80 74.9 47.7 1.67

93.8 92.3 78.5 2.67

5 10 50 No stim

DMSO 0.1 1 5 No stim

Tregs

Tconvs

Tregs

Tconvs

Tregs

Tconvs

Tregs

CFSE

Co

un

ts

CFSE

Co

un

ts

CFSE

Co

un

ts

CFSE

Co

un

ts

100

80

60

40

20

0

120

100

80

60

40

20

0

None DMSO 5 10TCN (mmol/L)

TCN (mmol/L)

MK-2206 (mmol/L)

DMSO 300 800 1,000 No stim

WM (nmol/L)

IC87114 (mmol/L)

DMSO

95.2

70.1 58.1 45.2 28.8 1.45

91.9 85.3 81 7.88

2 10 20 No stim

Rel

ativ

e p

rolif

erat

ion

Rel

ativ

e p

rolif

erat

ion

120

100

80

60

40

20

0

Rel

ativ

e p

rolif

erat

ion

50

None DMSO

None DMSO

120

100

80

60

40

20

0

Rel

ativ

e p

rolif

erat

ion

None DMSO 2 10IC87114 [mmol/L]

20

300 800 1,000

0.1 1

MK2206 (mmol/L)

WM (nmol/L)

5

Figure 1. The inhibition of PI3K and Akt in human T cells selectively inhibits the proliferation of human Tregs compared with Tconvs in a dose-dependentmanner. Tregs (CD4þCD25HI) andTconvs (CD4þCD25�) were fractionated fromhumanPBMCsby cell sorting. Proliferationwas examined bydilution ofCFSEafter stimulation with anti-CD3/anti-CD28 for 3 days with titrated amounts of triciribine (TCN), MK-2206, wortmannin (WM), and IC87114. Dead cellswere excluded by 7-AAD incorporation. Left, average of three experiments, normalized to untreated controls; right, representative examples. A, a significantreduction in Treg proliferation was observed in response to Akt inhibition by triciribine compared with that of Tconvs at all doses tested: 5 mmol/L (P¼ 0.04),10 mmol/L (P¼ 0.03), and 50 mmol/L (P¼ 0.01). B, a significant reduction of Treg proliferation in response toMK-2206 treatment was observed comparedwiththat of Tconvs at the 1 mmol/L (P ¼ 0.05) and 5 mmol/L (P ¼ 0.005) doses. C, proliferation of Tregs was significantly reduced by wortmannin comparedwith that of Tconvs at all doses tested: 300 nmol/L (P ¼ 0.03), 800 nmol/L (P ¼ 0.01), and 1,000 nmol/L (P ¼ 0.002). D, Treg proliferation was significantlyreduced by IC87114 compared with that of Tconvs at the 10 mmol/L (P ¼ 0.04) and 20 mmol/L (P < 0.005) doses. �, P < 0.05.






Treg immunosuppressive ability in vivo. Mice were condition-ed on alternate days with wortmannin, triciribine, or DMSOfor a week before vaccination with HPV16 E7 peptide–basedvaccine that was given s.c. on days 7 and 14. One week after thesecond vaccination, splenocytes were harvested, and theE7-specific immune response was assayed by IFNg ELISPOTafter restimulation with E7 peptides. As expected, vaccination-induced E7-reactive T cells, and importantly addition ofwortmannin and triciribine to vaccine treatment, significant-ly increased the number of E7-reactive T cells compared withthose of the controls (Fig. 3C).

Taken together, these data demonstrate that the PI3K–Aktpathway is necessary for Treg proliferation, FoxP3þCD4þ Tregmaintenance inmice, and inhibition of this pathway resulted inan augmented T-cell response to peptide vaccination other-wise suppressed by the presence of Tregs.

PI3K–Akt inhibition mitigates tumor growthThe depletion of Tregs results in enhanced antitumor

immune responses (23). Our data so far show that PI3K andAkt inhibition results in selective in vitro inhibition of Tregsand enhanced immune response to peptide vaccination. Wetherefore tested the utility of targeting the PI3K–Akt pathway

for tumor treatment. Mice implanted s.c. with TC-1 tumor cellswere treated with wortmannin or DMSO for 1 week afterpalpable tumors were detected. Mice were then vaccinatedwith the E7 vaccine as described above and tumor growth wasmonitored. Although both E7 vaccination and wortmannintreatment significantly inhibited TC-1 tumor growth (P < 0.05and P < 0.001, respectively), the greatest impairment wasachieved with the combination of wortmannin treatment andE7 vaccination (P < 0.0001; Fig. 4A).

To evaluate the mechanism by which PI3K–Akt inhibi-tion reduced tumor growth and to minimize the directeffect of PI3K–Akt inhibitors on the tumors, a prophylactictumor model was employed using B16 and EL4 tumor cells.Mice were treated with wortmannin, triciribine, or DMSOon days �7, �4, and �2 before s.c. tumor inoculation onday 0, and tumor growth was monitored thereafter. Wort-mannin and triciribine were chosen because of their shorthalf-lives (10 minutes and 6 hours, respectively) to ensure aminimal direct cytotoxic effect of the inhibitors on thetumor cells. Tumor growth was significantly inhibited bywortmannin and triciribine compared with DMSO in theB16 (P < 0.0001; Fig. 4B) and EL4 (data not shown) tumormodels.

DMSOA

B C

Tconvs

Tregs

Co

un

ts

CFSE

Bas

al

DM

SO

WM

TCN

IC87

114

MK

-220

6

Basal

DMSOW

MTCN

IC87

114

MK-220

6

92.67%

89.76% 52.49% 60.87% 27.71%

88.45% 89.32% 84.22%

WM MK-2206 IC87114

Anti-CD3/CD28 + IL2

CD

25+

CD

25–

CD25–

CD25+

pS6

3

2

1

0

pS6

S6

S6 PS

6/S

6 p

rote

in le

vel

(no

rmal

ized

)

Figure 2. PI3K and Akt inhibition differentially inhibits the proliferation and downstream activation ofmurine Tregs comparedwith that of Tconvs. MouseCD4þ

splenocytes were fractionated into CD25þ (Treg) and CD25� (Tconv) subsets using magnetic bead enrichment kits. Proliferation was assayed by thedilution ofCFSE after stimulationwith anti-CD3/anti-CD28 for 3 dayswithwortmannin (WM; 200 nmol/L),MK-2206 (2mmol/L), IC87114 (10 mmol/L), or DMSO.Dead cells were excluded by 7-AAD incorporation. A, representative example of the selective proliferation inhibition in Tregs (bottom) by wortmannin,MK-2206, and IC87114 compared with that of Tconvs (top). B, treatment with wortmannin, triciribine (TCN), IC87114, or MK-2206 resulted in a markeddecrease of phosphorylated S6 in Tregs (CD25þ) in comparison with Tconvs (CD25�). C, densitometry of phosphorylated S6 levels to S6 ratio showedsignificantly lower levels in Tregs when compared with Tconvs when the stimulated cells were treated with PI3K–Akt pathway inhibitors. Densitometry wasperformed on gel as shown in B.

Abu-Eid et al.





These results demonstrate the effectiveness of PI3K and Aktinhibitors in suppressing tumor growth even without a tumor-specific vaccine andwithminimal direct effect on the tumor cells.

Inhibition of the PI3K–Akt pathway enhances theantitumor effect of tumor-specific vaccinesThe above results clearly show an enhanced antitumor

effect of a vaccine when combined with PI3K–Akt pathway

inhibitors. To further evaluate the effect of PI3K–Akt pathwayinhibition on the TME, we tested the effect of combining the E7vaccine with MK-2206 on tumor growth and immune-cellinfiltration in the TC-1 mouse tumor model. MK-2206 iscurrently in several clinical trials and has a long half-life(60–90 hours), and was therefore chosen to be used to assessthe effect of Akt inhibition on the TME. Following tumorimplantation (day 0), appropriate groups of mice were injectedwith the E7 vaccine and/or MK-2206. Tumors were measuredand the mice were euthanized on day 21, and the intratumorallymphocytic infiltrate was assessed. Both the vaccine and MK-2206 individually significantly reduced tumor sizes in com-parison with the nontreated controls (P < 0.05; Fig. 5A);however, when combined, they reduced the tumor volumemore profoundly comparedwith that of nontreated controls (P< 0.01) and of single vaccine treatment (P < 0.05).

The tumor tissues were then processed, and the numbersof CD8þ and Foxp3þ T cells per million tumor cells withinthe tumors were evaluated. In tumors treated with thecombination of MK-2206 with the vaccine, a significantlyhigher number of CD8þ T cells were detected in comparisonwith that of the nontreated group (P < 0.001; Fig. 5B). Thegroup treated with the combination also showed a signifi-cant reduction in Treg (Foxp3þ) infiltration in comparisonwith that of the nontreated group (P < 0.05) and the vaccinealone group (P < 0.01; Fig. 5C).

A

B

DMSO

DMSO

DMSO + E7WM + DMSO

WM

NT

TCN

WM + E7

Tum

or

area

(m

m2 )

Tum

or

area

(m

m2 )

600

500

400

300

200

100

0

300

200

100

0

0 3 6 9 12Days after tumor inoculation

15 18 21


15 18 21

TC-1

B16

Figure 4. PI3K–Akt signaling inhibition mitigates tumor growth. A, TC-1tumor-bearing mice were treated with wortmannin (WM) or DMSO for 1week after palpable tumors were detected. Mice were then vaccinatedwith E7 peptide and tumor growth was monitored. Both E7 vaccination(P < 0.05) and wortmannin treatment (P < 0.001) significantly inhibitedTC-1 tumor growth, yet the greatest impairment was achieved with thecombination of wortmannin treatment and E7 vaccination (P < 0.0001).B, mice were prophylactically treated with wortmannin, triciribine (TCN),or DMSO before s.c. B16 tumor inoculation and tumor growth wasmonitored. Tumor growth was significantly inhibited by wortmanninand triciribine compared with DMSO-treated animals (P < 0.0001).��, P < 0.01; ��, P < 0.001; ��, P < 0.0001.

A

B

CRestimulation

DMSOE7 peptide

No trea

tmen

t

DMSOW

MTCN

MK-220

6

No trea

tmen

t

DMSOW

MTCN

DMSONoneNone E7 E7 E7

TreatmentVaccination

WM TCN

MK-220

6

25

20

15

10

5

0

4

3

2

1

0

Per

cen

tag

e o

f C

D4

cells

of

tota

l sp

len

ocy

tes

Per

cen

tag

e o

f F

OX

P3+

cel

ls o

fC

D4+

cel

ls

300

200

100

0

IFN

g-se

cret

ing

cel

ls(P

er 1

06 c

ells

)

Figure 3. The PI3K–Akt pathway is essential for the in vivomaintenance ofmurine Tregs. Mice were conditioned with wortmannin (WM; 40 mg),triciribine (TCN; 50mg),MK-2206 (10mg), orDMSO for 1weekonalternatedays before flow cytometric analysis of splenocytes. Alternatively, micewere conditioned on alternate dayswith either wortmannin or DMSO for aweek before vaccination with E7. One week after E7 vaccination,splenocytes were harvested, and the E7 immune response was assayedby ELISPOT. A, mice treated with DMSO vehicle contained similarpercentages of CD4 T cells compared with wortmannin-, triciribine-, orMK-2206–treated animals (P ¼ 0.1). B, the number of FoxP3þ cells inCD4þ T cells was significantly reduced in the mice treated with theinhibitors compared with DMSO (P < 0.05). C, vaccination resulted in asignificant increase in E7-reactive T cells. Wortmannin and triciribinetreatment significantly increased the number of E7-reactive cellscompared with DMSO-treated controls. �, P < 0.05; ��, P < 0.01;��, P < 0.001; ��, P < 0.0001.






These data further emphasize the enhancement of thetherapeutic efficacy of tumor vaccines by the use of an Aktinhibitor and demonstrate an augmentation of the antitumorimmune response when the vaccine was combined with theinhibitor.

PI3K–Akt inhibition reduces tumor growth in aTreg-dependent manner

So far, we have shown that inhibiting the PI3K–Akt pathwayresults in the suppression of tumor growth, enhancement ofthe therapeutic efficacy of tumor-specific vaccine, andenhancement of the antitumor immune response (increase inintratumoral CD8þ cells and decrease in Tregs). To confirm therole of Treg depletion by PI3K–Akt inhibitors in their antitu-mor therapeutic effect, the CT26 tumor model was used.Similar to what was observed with the B16 and EL4 tumormodels, prophylactic treatment of mice with wortmannin ortriciribine before s.c. tumor inoculation resulted in a signifi-cant tumor growth inhibition by triciribine (P < 0.0001; Fig. 6A)and wortmannin (P < 0.0001; Fig. 6B) compared with that inDMSO-treated animals. However, this effect was reversedwhen the wortmannin-treated mice were reconstituted withTregs. Ex vivo–cultured Treg infusion on day 4 resulted in therestoration of tumor growth, comparable with that in DMSO-treated mice (Fig. 6B). The same effect was observed with theEL4 tumor model (data not shown).

An analysis of FoxP3þ T-cell infiltration into CT26 tumorselucidated a significant decrease in Treg numbers after wort-mannin treatment that was restored after infusion of ex vivo–grown Tregs 4 days after tumor inoculation (Fig. 6C).

Taken together, these data demonstrate that the Tregdependence on PI3K–Akt signaling can be exploited to selec-tively deplete suppressive CD4 Tregs, resulting in an enhancedcapacity to elicit an antitumor immune response when com-bined with the vaccine.

DiscussionCD4 T cells differentiate into a panoply of effector cells with

an array of diverse functions in the immune response. Pre-clinical mouse models have identified inflammatory CD4 Tcells that mediate tumor regression and CD4 Tregs thatsupport tumor growth (18).

A correlation of tumor-infiltrating Tregs with poor clinicalprognosis has been demonstrated (19–22), and the depletion ofTregs was found to enhance antitumor immunity and promotetumor regression (23–26). However, there is still a scarceness ofefficient and highly selective Treg-depleting clinical reagentsfor use in tumor immunomodulation.

Several reports have shown that Tregs and Tconvs displayunique signaling signatures downstream of TCR (28–31). ThePI3K–Akt signaling pathway is important in the cellularresponse to TCR stimulation and costimulation (32, 33). Thispathway plays a critical role in T-cell functions, includingproliferation, survival, migration, and metabolism (35, 36).

Because of the important role of the PI3K–Akt pathway inT-cell function (35, 36) and the reported differences betweenTregs and Tconvs downstream of TCR (28–31), we investigated

A

B

C

Tum

or

volu

me

(cm

3 )

Intr

atu

mo

ral C

D8+

cel

lsp

er 1

06 t

um

or

cells

(T

C-1

)In

trat

um

ora

l Fox

p3+

cel

lsp

er 1

06 t

um

or

cells

(T

C-1

)

2.0

1.5

1.0

0.5

0.0

8,000

6,000

4,000

2,000

0

2,500

2,000

1,500

1,000

500

0

NT

MK-220

6Vac

Vac +

MK-2

206

NT

MK-220

6Vac

Vac +

MK-2

206

NT

MK-220

6Vac

Vac +

MK-2

206

Figure5. Akt inhibition by MK-2206 enhances the antitumor therapeuticeffect of tumor-specific vaccine. TC-1 tumor-bearing mice wereinjected on days 7 and 14 with the E7 vaccine and/or MK-2206 (30 mg).The tumors were measured and the mice euthanized on day 21.A, the combination of MK-2206 with the vaccine had a synergisticeffect in the reduction of tumor volume in comparison with thevaccine alone (P < 0.05; combined data from two independentexperiments). B, the combination of MK-2206 with the vaccinesignificantly increased the number of tumor-infiltrating CD8þ T cells incomparison with the nontreated control (P < 0.001; representativeexample). C, the combination of MK-2206 with the vaccinesignificantly reduced the number of tumor-infiltrating CD4þFoxp3þ

Tregs in comparison with the vaccine alone (P < 0.05; representativeexample). �, P < 0.05; ��, P < 0.01; ��, P < 0.001.

Abu-Eid et al.





the impact of Akt and PI3K inhibition on Tregs and Tconvs todiscern any selective effect that inhibition of the PI3K–Aktpathway might have on these CD4 T-cell subsets.We demonstrated that several molecular inhibitors that

target PI3K and its downstream effector, Akt, selectively inhibitthe in vitro proliferation of human and murine Tregs whencompared with Tconvs. This selective decrease in Treg prolif-eration provided us with a potential strategy to modulate theTreg/Tconv balance in vivo. We found that Akt and PI3Kinhibition selectively decreased the number of Tregs in vivoboth in na€�ve and tumor-bearing mice. We found that this

translated into a therapeutic effect in which the in vivotreatment of tumor-bearing mice with PI3K–Akt pathwayinhibitors displayed a significant antitumor therapeutic effi-cacy. The inhibition of tumor growthwasmore profoundwhenAkt and PI3K inhibitors were combined with an antitumorvaccine. Analysis of the TME revealed an enhanced antitumorimmune response to the vaccine when combined with PI3K–Akt pathway inhibitors. On their own, or in combination withvaccines, the antitumor therapeutic effects of Akt and PI3Kinhibitors were found to be Treg dependent as they could bereversed by Treg reconstitution.

It is interesting to point out the apparent difference betweenthe inhibition of Akt and PI3K and the downstream inhibitionof mTOR. mTOR inhibition with rapamycin has been shown toexhibit the opposite effect as it supports the proliferation andsurvival of Tregs (data not shown; refs. 40–43) and is thereforeused as an immunosuppressant. The opposing effects of PI3K–Akt versus mTOR inhibition on Tregs may be explained by afeedback loop in which mTOR inhibition results in PI3K-dependent Akt activation, which, in turn, sustains signalingthrough mTOR (44).

It is also important to highlight the opposite role that thePI3K–Akt pathway plays in the de novo differentiation ofmouse Tregs. Active Akt signaling has been shown to be apotent suppressor of differentiation of human (45) and mouseCD4 Tregs (46).

Here, we report that PI3K–Akt inhibitors reduced tumorgrowth in several mouse models, and we have shown that thisis due to the selective inhibition of Tregs. However, theantitumor effect due to the inhibition of PI3K–Akt can beexhibited through other mechanisms such as the direct inhi-bition of tumor growth and the enhanced survival of CD8 Tcells (Akt signaling drives CD8 T-cell differentiation and limitssurvival and memory formation; ref. 47). To reduce the directeffect of the inhibitors on tumor cells, mice were treated withthe inhibitors before their inoculationwith tumor cells, and theinhibitors wortmannin and triciribine were chosen for in vivotreatment due to their short half-lives. To further demonstratethe role of selective inhibition of Tregs as opposed to enhancedCD8 effector function, mice were infused with Tregs, whichrestored tumor growth to a rate identical to that of the controlanimals. Together with the selective inhibition of in vitroproliferation in Tregs, these data strongly implicate the dis-ruption of Treg homeostasis as a mechanism of PI3K–Aktinhibitor–mediated antitumor function.

In summary, clinical modulation of the tumor immuneresponse has been limited by a lack of highly selective reagentsthat specifically target Tregs. Here, we demonstrate thatselective inhibition of human and murine Treg proliferationcould be achieved using PI3K–Akt molecular inhibitors. PI3Kand Akt inhibitors selectively disrupt the homeostasis of Tregsand result in a significant antitumor effect that is Treg depen-dent. Recent clinical studies have correlated immune dysfunc-tion after immunotherapy with an increase of Tregs (48–50).The data presented in this report demonstrate that PI3K–Aktinhibition enhances antitumor immune responses when com-bined with vaccines by selectively reducing the number ofTregs. The selective Treg effect and enhanced therapeutic

A

B

DMSONT

TCN

WMWM + Treg

NT

Tum

or

area

(m

m2 )

600

500

400

300

200

100

0

Tum

or

area

(m

m2 )

500

400

300

200

100

0

0 3

None WM WM + Treg

Day 20Day 24

6 9 12Days after tumor inoculation

15 18 21


15 18 21

CT26

CT26

C

Intr

atu

mo

ral F

oxp

3+ c

ells

per

106

tu

mo

r ce

lls (

CT

26) 2,000

1,500

1,000

500

0

Figure 6. PI3K–Akt inhibition reduces tumor growth in a Treg-dependentmanner. Mice were prophylactically treated with wortmannin (WM),triciribine (TCN), or DMSO before s.c. CT26 tumor inoculation.Ex vivo–cultured Tregswere infused inwortmannin-treatedmice on day 4after tumor inoculation. Tumor growth was monitored and the Foxp3þ T-cell infiltrate was then assessed on days 20 and 24. A, prophylactictreatment of the mice with triciribine for 1 week before s.c. tumorinoculation resulted in a significant tumor growth inhibition (P < 0.0001).B, prophylactic treatment of the mice with wortmannin for 1 week befores.c. tumor inoculation resulted in a significant tumor growth inhibitionwortmannin (P < 0.0001). This effect was reversed when Tregs werereconstituted by ex vivo–cultured Treg infusion on day 4. C, analysis ofFoxP3þ T-cell infiltration into CT26 tumor (days 20 and 24) elucidated asignificant decrease in Treg numbers per million tumor cells afterwortmannin treatment (P < 0.01), which was restored after the infusion ofex vivo–grown Tregs. ��, P < 0.01; ��, P < 0.001; ��, P < 0.0001.






potential suggests that PI3K–Akt inhibitors could be exploitedin the clinic as immune modulators in cancer therapy.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

Authors' ContributionsConception and design: R.N. Samara, L. Ozbun, M.Y. Abdalla, J.A. Berzofsky,S.N. KhleifDevelopment of methodology: R.N. Samara, L. Ozbun, M.Y. AbdallaAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): R.N. Samara, L. Ozbun, M.Y. Abdalla, M. MkrtichyanAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): R. Abu-Eid, R.N. Samara, L. Ozbun, J.A. Berzofsky,M. Mkrtichyan

Writing, review, and/or revision of the manuscript: R. Abu-Eid,R.N. Samara, J.A. Berzofsky, K.M. Friedman, M. Mkrtichyan, S.N. KhleifAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): R.N. SamaraStudy supervision: R.N. Samara, S.N. Khleif

AcknowledgmentsThe authors thank Dr. Esteban Celis and Dr. Rhea-Beth Markowitz for review

of the manuscript and valuable suggestions.

The costs of publication of this article were defrayed in part by the payment ofpage charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received May 14, 2014; revised June 23, 2014; accepted July 21, 2014;published OnlineFirst July 30, 2014.

References1. Rosenberg SA. Development of effective immunotherapy for the

treatment of patients with cancer. J Am Coll Surg 2004;198:685–96.2. Rosenberg SA, Yang JC, Restifo NP. Cancer immunotherapy: moving

beyond current vaccines. Nat Med 2004;10:909–15.3. Klebanoff CA, Acquavella N, Yu Z, Restifo NP. Therapeutic cancer

vaccines: are we there yet? Immunol Rev 2011;239:27–44.4. Kantoff PW, Higano CS, Shore ND, Berger ER, Small EJ, Penson DF,

et al. Sipuleucel-T immunotherapy for castration-resistant prostatecancer. N Engl J Med 2010;363:411–22.

5. Schwartzentruber DJ, LawsonDH, Richards JM,Conry RM,Miller DM,Treisman J, et al. gp100 peptide vaccine and interleukin-2 in patientswith advanced melanoma. N Engl J Med 2011;364:2119–27.

6. JohnsonLA,MorganRA,DudleyME,CassardL,YangJC,HughesMS,et al. Gene therapy with human and mouse T-cell receptors mediatescancer regression and targets normal tissues expressing cognateantigen. Blood 2009;114:535–46.

7. Robbins PF, Morgan RA, Feldman SA, Yang JC, Sherry RM, DudleyME, et al. Tumor regression in patients with metastatic synovial cellsarcoma and melanoma using genetically engineered lymphocytesreactive with NY-ESO-1. J Clin Oncol 2011;29:917–24.

8. Kochenderfer JN, Dudley ME, Feldman SA, Wilson WH, Spaner DE,Maric I, et al. B-cell depletion and remissions of malignancy along withcytokine-associated toxicity in a clinical trial of anti-CD19 chimeric-antigen-receptor–transduced T cells. Blood 2012;119:2709–20.

9. Kalos M, Levine BL, Porter DL, Katz S, Grupp SA, Bagg A, et al. T cellswith chimeric antigen receptors have potent antitumor effects and canestablish memory in patients with advanced leukemia. Sci Transl Med2011;3:95ra73.

10. Besser MJ, Shapira-Frommer R, Treves AJ, Zippel D, Itzhaki O,Hershkovitz L, et al. Clinical responses in a phase II study usingadoptive transfer of short-term cultured tumor infiltration lympho-cytes in metastatic melanoma patients. Clin Cancer Res 2010;16:2646–55.

11. Dudley ME, Gross CA, LanghanMM, Garcia MR, Sherry RM, Yang JC,et al. CD8þ enriched "young" tumor infiltrating lymphocytes canmediate regression of metastatic melanoma. Clin Cancer Res 2010;16:6122–31.

12. Dudley ME, Wunderlich JR, Yang JC, Hwu P, Schwartzentruber DJ,Topalian SL, et al. A phase I study of nonmyeloablative chemotherapyand adoptive transfer of autologous tumor antigen-specific T lympho-cytes in patients with metastatic melanoma. J Immunother 2002;25:243–51.

13. DudleyME, Yang JC, Sherry R,HughesMS,Royal R, KammulaU, et al.Adoptive cell therapy for patients with metastatic melanoma: evalu-ation of intensivemyeloablative chemoradiation preparative regimens.J Clin Oncol 2008;26:5233–9.

14. Rosenberg SA. Cell transfer immunotherapy for metastatic solidcancer—what clinicians need to know. Nat Rev Clin Oncol 2011;8:577–85.

15. Rosenberg SA, Kochenderfer JN. Personalized cell transfer immuno-therapy for B-cell malignancies and solid cancers. Mol Ther 2011;19:1928–30.

16. Rosenberg SA, Yang JC, Sherry RM, Kammula US, Hughes MS, PhanGQ, et al. Durable complete responses in heavily pretreated patientswith metastatic melanoma using T-cell transfer immunotherapy. ClinCancer Res 2011;17:4550–7.

17. TangQ, Bluestone JA. The Foxp3þ regulatory T cell: a jack of all trades,master of regulation. Nat Immunol 2008;9:239–44.

18. Antony PA, Piccirillo CA, Akpinarli A, Finkelstein SE, Speiss PJ, Sur-man DR, et al. CD8þ T cell immunity against a tumor/self-antigen isaugmented byCD4þ T helper cells and hindered by naturally occurringT regulatory cells. J Immunol 2005;174:2591–601.

19. Curiel TJ, Coukos G, Zou L, Alvarez X, Cheng P, Mottram P, et al.Specific recruitment of regulatory T cells in ovarian carcinoma fostersimmune privilege and predicts reduced survival. Nat Med 2004;10:942–9.

20. Fu J, Xu D, Liu Z, Shi M, Zhao P, Fu B, et al. Increased regulatory T cellscorrelate with CD8 T-cell impairment and poor survival in hepatocel-lular carcinoma patients. Gastroenterology 2007;132:2328–39.

21. Gjerdrum LM, Woetmann A, Odum N, Burton CM, Rossen K, Skov-gaard GL, et al. FOXP3þ regulatory T cells in cutaneous T-cell lym-phomas: association with disease stage and survival. Leukemia2007;21:2512–8.

22. Heimberger AB, Abou-GhazalM, Reina-Ortiz C, YangDS, SunW,QiaoW, et al. Incidence and prognostic impact of FoxP3þ regulatory T cellsin human gliomas. Clin Cancer Res 2008;14:5166–72.

23. Zhou S, Tao H, Zhen Z, Chen H, Chen G, Yang Y. Depletion of CD4þ

CD25þ regulatory T cells promotes CCL21-mediated antitumor immu-nity. PLoS ONE 2013;8:e73952.

24. Jarry U, Donnou S, Vincent M, Jeannin P, Pineau L, Fremaux I, et al.Treg depletion followed by intracerebral CpG-ODN injection inducebrain tumor rejection. J Neuroimmunol 2014;267:35–42.

25. KeenanBP, Saenger Y, KafrouniMI, Leubner A, Lauer P,Maitra A, et al.A listeria vaccine and depletion of T-regulatory cells activate immunityagainst early stage pancreatic intraepithelial neoplasms and prolongsurvival of mice. Gastroenterology 2014;146:1784–94.e6.

26. Reginato E, Mroz P, Chung H, Kawakubo M, Wolf P, Hamblin MR.Photodynamic therapy plus regulatory T-cell depletion producesimmunity against a mouse tumour that expresses a self-antigen. BrJ Cancer 2013;109:2167–74.

27. de Vries IJ, Castelli C, Huygens C, Jacobs JF, Stockis J, Schuler-Thurner B, et al. Frequency of circulating Tregs with demethylatedFOXP3 intron 1 in melanoma patients receiving tumor vaccinesand potentially Treg-depleting agents. Clin Cancer Res 2011;17:841–8.

28. Walsh PT, Buckler JL, Zhang J, Gelman AE, Dalton NM, Taylor DK,et al. PTEN inhibits IL-2 receptor-mediated expansion of CD4þCD25þ

Tregs. J Clin Invest 2006;116:2521–31.29. Crellin NK,Garcia RV, LevingsMK. Flowcytometry-basedmethods for

studying signaling in human CD4þCD25þFOXP3þ T regulatory cells.J Immunol Methods 2007;324:92–104.

30. CrellinNK,GarciaRV, LevingsMK.Altered activationof AKT is requiredfor the suppressive function of human CD4þCD25þ T regulatory cells.Blood 2007;109:2014–22.


Abu-Eid et al.




31. Carson BD, Ziegler SF. Impaired T cell receptor signaling in Foxp3þ

CD4 T cells. Ann N Y Acad Sci 2007;1103:167–78.32. Smith-Garvin JE, Koretzky GA, JordanMS. T cell activation. Annu Rev

Immunol 2009;27:591–619.33. Song J, Lei FT, Xiong X, Haque R. Intracellular signals of T cell

costimulation. Cell Mol Immunol 2008;5:239–47.34. Cantrell DA. T-cell antigen receptor signal transduction. Immunology

2002;105:369–74.35. Kane LP, Weiss A. The PI-3 kinase/Akt pathway and T cell activation:

pleiotropic pathways downstream of PIP3. Immunol Rev 2003;192:7–20.

36. Finlay D, Cantrell D. Phosphoinositide 3-kinase and the mammaliantarget of rapamycin pathways control T cell migration. Ann N Y AcadSci 2010;1183:149–57.

37. Simova J,Bubenik J, Bieblova J, RosaliaRA, Fric J,ReinisM.Depletionof T(reg) cells inhibits minimal residual disease after surgery of HPV16-associated tumours. Int J Oncol 2006;29:1567–71.

38. Lin KY, Guarnieri FG, Staveley-O'Carroll KF, Levitsky HI, August JT,Pardoll DM, et al. Treatment of established tumorswith a novel vaccinethat enhances major histocompatibility class II presentation of tumorantigen. Cancer Research 1996;56:21–6.

39. Ahlers JD, Belyakov IM, Terabe M, Koka R, Donaldson DD, ThomasEK, et al. A push-pull approach to maximize vaccine efficacy: abro-gating suppression with an IL-13 inhibitor while augmenting help withgranulocyte/macrophage colony-stimulating factor and CD40L. ProcNatl Acad Sci U S A 2002;99:13020–5.

40. Long SA, Buckner JH. Combination of rapamycin and IL-2 increasesde novo induction of human CD4(þ)CD25(þ)FOXP3(þ) T cells.J Autoimmun 2008;30:293–302.

41. Strauss L, Whiteside TL, Knights A, Bergmann C, Knuth A, ZippeliusA. Selective survival of naturally occurring human CD4þ

CD25þFoxp3þ regulatory T cells cultured with rapamycin. J Immu-nol 2007;178:320–9.

42. BattagliaM,StabiliniA,RoncaroloMG.RapamycinselectivelyexpandsCD4þCD25þFoxP3þ regulatory T cells. Blood 2005;105:4743–8.

43. Basu S, Golovina T, Mikheeva T, June CH, Riley JL. Cutting edge:Foxp3-mediated induction of pim 2 allows human T regulatory cells topreferentially expand in rapamycin. J Immunol 2008;180:5794–8.

44. Sun SY, Rosenberg LM,Wang X, Zhou Z, Yue P, Fu H, et al. Activationof Akt and eIF4E survival pathways by rapamycin-mediated mamma-lian target of rapamycin inhibition. Cancer Res 2005;65:7052–8.

45. Sun J, Dotti G, Huye LE, Foster AE, Savoldo B, GramatgesMM, et al. Tcells expressing constitutively active Akt resist multiple tumor-asso-ciated inhibitory mechanisms. Mol Ther 2010;18:2006–17.

46. Haxhinasto S, Mathis D, Benoist C. The AKT-mTOR axis regulatesde novo differentiation of CD4þFoxp3þ cells. J Exp Med 2008;205:565–74.

47. Kim EH, Sullivan JA, Plisch EH, Tejera MM, Jatzek A, Choi KY, et al.Signal integration by Akt regulates CD8 T cell effector and memorydifferentiation. J Immunol 2012;188:4305–14.

48. Saitoh A, Narita M, Watanabe N, Tochiki N, Yamahira A, Nakamura T,et al. WT1 peptide vaccination in a CML patient: induction of effectivecytotoxic T lymphocytes and significance of peptide administrationinterval. Med Oncol 2011;28:219–30.

49. Macatangay BJ, Szajnik ME, Whiteside TL, Riddler SA, Rinaldo CR.Regulatory T cell suppression of Gag-specific CD8 T cell polyfunc-tional response after therapeutic vaccination of HIV-1-infectedpatients on ART. PLoS ONE 2010;5:e9852.

50. FongB, JinR,WangX,SafaeeM, LisieroDN,Yang I, et al.Monitoring ofregulatory T cell frequencies and expression of CTLA-4 on T cells,before and after DC vaccination, can predict survival in GBM patients.PLoS ONE 2012;7:e32614.






2014;2:1080-1089. Published OnlineFirst July 30, 2014.Cancer Immunol Res Rasha Abu-Eid, Raed N. Samara, Laurent Ozbun, et al. Akt Pathway

−Selective Inhibition of Regulatory T Cells by Targeting the PI3K

Updated version

10.1158/2326-6066.CIR-14-0095doi:

Access the most recent version of this article at:

Cited articles

http://cancerimmunolres.aacrjournals.org/content/2/11/1080.full#ref-list-1

This article cites 50 articles, 20 of which you can access for free at:

Citing articles

http://cancerimmunolres.aacrjournals.org/content/2/11/1080.full#related-urls

This article has been cited by 9 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerimmunolres.aacrjournals.org/content/2/11/1080To request permission to re-use all or part of this article, use this link



http://cancerimmunolres.aacrjournals.org/lookup/doi/10.1158/2326-6066.CIR-14-0095

http://cancerimmunolres.aacrjournals.org/content/2/11/1080.full#ref-list-1

http://cancerimmunolres.aacrjournals.org/content/2/11/1080.full#related-urls

http://cancerimmunolres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://cancerimmunolres.aacrjournals.org/content/2/11/1080


Documents

Selective Inhibition of Regulatory T Cells by Targeting the PI3K–Akt