Nei

SSD

a

ARRAA

KNCESKF

1

sctcppl(rat

h0

Molecular Immunology 71 (2016) 42–53

Contents lists available at ScienceDirect

Molecular Immunology

j ourna l ho me pa g e : www.elsev ier .com/ locate /mol imm

eem leaf glycoprotein promotes dual generation of central andffector memory CD8+ T cells against sarcoma antigen vaccine tonduce protective anti-tumor immunity

arbari Ghosh, Madhurima Sarkar, Tithi Ghosh, Ipsita Guha, Avishek Bhuniya, Akata Saha,hayani Dasgupta, Subhasis Barik, Anamika Bose1, Rathindranath Baral ∗,1

epartment of Immunoregulation and Immunodiagnostics, Chittaranjan National Cancer Institute (CNCI), 37, S. P. Mukherjee Road, Kolkata 700026, India

r t i c l e i n f o

rticle history:eceived 27 November 2015eceived in revised form 8 January 2016ccepted 20 January 2016vailable online 3 February 2016

eywords:eem leaf glycoproteinentral memory CD8+ T cellsffector memory cellsarcoma antigenLF2

a b s t r a c t

We have previously shown that Neem Leaf Glycoprotein (NLGP) mediates sustained tumor protection byactivating host immune response. Now we report that adjuvant help from NLGP predominantly generatesCD44+CD62LhighCCR7high central memory (TCM; in lymph node) and CD44+CD62LlowCCR7low effectormemory (TEM; in spleen) CD8+ T cells of Swiss mice after vaccination with sarcoma antigen (SarAg).Generated TCM and TEM participated either to replenish memory cell pool for sustained disease free statesor in rapid tumor eradication respectively. TCM generated after SarAg + NLGP vaccination underwentsignificant proliferation and IL-2 secretion following SarAg re-stimulation. Furthermore, SarAg + NLGPvaccination helps in greater survival of the memory precursor effector cells at the peak of the effectorresponse and their maintenance as mature memory cells, in comparison to single modality treatment.Such response is corroborated with the reduced phosphorylation of FOXO in the cytosol and increasedKLF2 in the nucleus associated with enhanced CD62L, CCR7 expression of lymph node-resident CD8+ T

+

OXO cells. However, spleen-resident CD8 T memory cells show superior efficacy for immediate memory-to-effector cell conversion. The data support in all aspects that SarAg + NLGP demonstrate superiority thanSarAg vaccination alone that benefits the host by rapid effector functions whenever required, whereas,central-memory cells are thought to replenish the memory cell pool for ultimate sustained disease freesurvival till 60 days following post-vaccination tumor inoculation.

© 2016 Elsevier Ltd. All rights reserved.

. Introduction

Cancer vaccinology deals with the awakening of the immuneystem to cancer by presenting antigens. These antigens are asso-iated with tumor cells and participating in either prevention ofhe cancer development (prophylactic cancer vaccines) or existingancer treatment (therapeutic cancer vaccine). In use for cancerrevention, vaccines must elicit long term memory without theotential of causing autoimmunity (Finn, 2003). The major prob-

em in developing an efficient cancer vaccine is the lack of TSAsantigens present only on tumor cells) and the weakness of immune

esponses against TAAs (antigens present mostly on tumor cells butlso on some normal cells), usually recognized by the immune sys-em as self-antigens (Buonaguro et al., 2011; Cunto-Amesty et al.,

∗ Corresponding author. Fax: +91 033 2475 7606.E-mail address: baralrathin@hotmail.com (R. Baral).

1 Both authors contributed equally to this work.

ttp://dx.doi.org/10.1016/j.molimm.2016.01.007161-5890/© 2016 Elsevier Ltd. All rights reserved.

2003; Kaech et al., 2002). Nevertheless, in the last 20 years, sev-eral different vaccines from whole tumor cells or tumor–cell lysateshave been evaluated in preclinical models and clinical trials. Vac-cine adjuvants trigger the activation and maturation of dendriticcells (DCs), so that DCs loaded with antigen, migrate to the proximallymph nodes and acquire the ability to optimally present antigensfor initiation of de novo T cell responses (Dubensky and Reed, 2010;Mohan et al., 2013).

A productive encounter of naïve CD8+ T cells to antigen stim-ulation follows a prototypical, tri-phasic response consisting of:(i) activation phase; (ii) death phase and iii. immunologic mem-ory phase (Klebanoff et al., 2006). According to Sallusto et al.(2004); Sallusto et al. (1999), memory T cells are divided broadlyinto central (TCM) and effector (TEM) memory T cells. TCMs areantigen-experienced cells that constitutively express CD62L and

CCR7, two surface molecules necessary for cellular extravasation inhigh endothelial venules and migration to T-cell zones of periph-eral lymph nodes. Whereas TEMs are antigen-experienced T cellsthat have these markers significantly downregulated and popu-

Immu

leaab(uentsmgt

camCwpipee

2

2

LwdaCa(gsiDRd

2

bMtppet1imgm

2

at

S. Ghosh et al. / Molecular

ate to the peripheral tissues, such as, the liver and lung (Masopustt al., 2001). Adoptively transferred self/tumor-reactive CD8+ TCMsre observed to be superior mediator of antitumor immunity ton established cancer compared to TEM cells, when given in com-ination with a systemically administered tumor antigen vaccineKlebanoff et al., 2005). TCM cells have greater proliferative capacitypon antigen re-encounter compared with TEM cells (Bachmannt al., 2005), thereby, leading to generation of a larger absoluteumber of terminally differentiated effector T cells that can infil-rate the peripheral sites to mediate antigen clearance. These datauggest that TCM may be more potent on a per cell basis inediating antigen clearance compared with TEM cells. Therefore,

eneration of TCM should be an important immunologic end-pointo consider in future preventive and therapeutic vaccine trials.

In the present study, we have reported that vaccination with sar-oma antigen (SarAg), along with neem leaf glycoprotein (NLGP),

global immunomodulator, generates antigen-specific centralemory CD8+ T cells (TCM) in lymph nodes and effector memory

D8+ T cells (TEM) in spleen during vaccination-post vaccinationindow. Furthermore, this vaccination schedule shows significantrotection against growth of sarcoma tumor expressing SarAg by

nducing effective conversion of memory-to-effector cells. Superiorrotection from tumor growth, along with tumor free survival tillxperiment termination, provides evidence in favor of the adjuvantfficacy of NLGP.

. Materials and methods

.1. Antibodies and reagents

RPMI-1640 and Fetal Bovine Serum (FBS) were purchased fromife Technologies (NY, USA). Lymphocyte separation media (LSM)as procured from MP Biomedicals, Irvine, CA, USA and HiMe-ia, Mumbai, India. Fluorescence conjugated different anti-mousentibodies (CD44, CD127, CD69, Ly6C-FITC conjugated and CD8,D62L, CCR7, Granzyme B-PE conjugated), purified anti-mousentibodies (CD8, Ki67, KLF2, FOXO3, pFOXO3, pAKT (ser), pAKTthr), pmTOR) were procured from either BD-Pharmingen or Biole-end (San Diego, CA, USA) or Santa Cruz (Dallas, Texas, USA). TMBubstrate solutions (for ELISA), CytoFix/CytoPerm solutions (forntracellular staining) were procured from BD-Pharmingen, Saniego, CA, USA. LDH cytotoxicity detection kit was purchased fromoche Diagnostics, Mannheim, Germany. RT-PCR primers wereesigned and procured from MWG Biotech AG, Bangalore, India.

.2. Neem leaf glycoprotein (NLGP)

Extract from neem (Azadirachta indica) leaves was preparedy the method as described previously (Chakraborty et al., 2010).ature leaves of same size and color (indicative of same age),

aken from a standard source, were shed-dried and pulverized. Leafowder was soaked overnight in phosphate-buffered saline (PBS),H 7.4. Supernatant was collected by centrifugation at 1500 rpm,xtensively dialyzed against PBS, pH 7.4 and concentrated by Cen-ricon membrane filter (Millipore Corporation, MA, USA) with0 kDa molecular weight cut-off. Purified NLGP was checked for

ts quality by electrophoresis and HPLC using routine laboratoryethods. Biological activity of purified NLGP was checked by tumor

rowth restriction assay before use. The protein concentration waseasured by Folin–Lowry method (Lowry et al., 1951).

.3. Sarcoma antigen preparation

Peritoneally grown sarcoma cells were collected, washed in PBSnd then lysed by five freeze (in liquid nitrogen)-thaw (at roomemperature) cycles (Mallick et al., 2014). Total cell disruption

nology 71 (2016) 42–53 43

was microscopically validated. Disrupted cells were sonicated andlysates were centrifuged at 15,000 × g (30 min, 4 ◦C). The super-natant was recovered as SarAg. The protein concentration wasmeasured using Folin’s phenol reagent (Lowry et al., 1951) andstored at −20 ◦C. Similarly, melanoma antigen (MelAg) was pre-pared from B16 melanoma cells grown in culture.

2.4. Mice and immunization

Female Swiss mice (Age: 4–6 weeks; Body weight: 24–27 g)were obtained from the Institutional Animal Care and MaintenanceDepartment of Chittaranjan National Cancer Institute, Kolkata, andmaintained under standard laboratory conditions. Autoclaved drypellet diet and water were given ad libitum. Mice (n = 6) were vac-cinated sub-cutaneously with the SarAg and SarAg + NLGP in thelower right flank. Vaccination was done at an interval of seven daysfor 4 weeks’ time period (Kalli et al., 2013; Hariharan et al., 1995).One group of mice was kept as vaccine free control (PBS treated).Animal experiments were performed according to the guidelinesestablished by the Institutional Animal Care and Ethics Committeeof CNCI, Kolkata, India, following their approval.

2.5. Tumor inoculation in vaccinated mice

Three groups of Swiss mice (n = 6 in each group) that were vac-cinated with PBS, SarAg and SarAg + NLGP, were inoculated withsarcoma cells (1 × 106 cells/mice) in the lower right flank (the siteof immunization) 30 days after the completion of the vaccinationschedule. Growth of solid tumor (in mm3) and survivability wasmonitored bi-weekly in these three cohorts of mice (PBS, SarAgand SarAg + NLGP) by caliper measurement using the formula:(width2 × length)/2. Tumors were monitored till day 60 from theday of tumor inoculation. Tumors were checked macroscopicallyby caliper measurement regularly and tumor size was recorded asa tumor area (in mm3) and mice were sacrificed if tumors becameulcerated or reached a size >250 mm2 within 120 days (Saha et al.,2006). At experiment termination tumors were checked micro-scopically after histological preparation.

2.6. Generation and culture of bone marrow derived DCs andantigen pulsing

Primary bone marrow-derived DCs (BmDCs) were obtainedfrom mouse bone marrow (from tibia and femurs) precursoraccording to the protocol described (Mallick et al., 2014). Tissuepieces were minced through a nylon mesh into a single-cell suspen-sion. Next, erythrocytes were lysed by resuspending the cell pelletin a hypotonic buffer (9.84 g/l NH4Cl, 1 g/l KHCO3, 0.1 mM EDTA)and incubating the cell suspension for 10 min on ice. The cells werewashed and cultured in a six-well plate at 2 × 106 cells/well withRPMI-1640 containing rmGM-CSF (10 ng/ml) and rmIL-4 (5 ng/ml).On day 6 of culture, non-adherent cells obtained from these cul-tures were considered to be immature bone marrow-derived DCs.Immature BmDCs (1 × 106 cells/ml) on day 8 were incubated withSarAg/MelAg (5 �g/ml of culture) for overnight in same culture con-dition. MelAg was used as an unrelated, non-specific control to seethe antigen specificity of the generated TCM phenotype cells forSarAg.

2.7. CD8+CD62Lhigh T cell (central memory phenotype)purification

CD8+CD62Lhigh T cells were purified from the single cell sus-pension of TDLN and spleen using Magnetic Assisted Cell Sorter(MACS) according to the manufacturer’s instruction (Barik et al.,2013a). In brief, CD8+ T cells were purified by magnetic bead

4 Immunology 71 (2016) 42–53

absolCfle

2s

stclawmflaahonfld

2

vh1Iwcwufa

2

gfl(2tctiwa

2

(c1l4

Table 1Primer sequences of various cytokine genes studied.

Name Primer sequences (5′-3′) Product size

�-Actin-F CAACCGTGAAAAGATGACCC 228 bp�-Actin-R ATGAGGTAGTCTGTCAGGTC

IL-7Ra-F CCACAATGAGTGCCCTACCT 238 bpIL-7Ra-R GACCGGACAGACACTCCAAT

Eomesodermin-F CCACTACAATGTTTTCGTGG 217 bpEomesodermin-R TTGTTGTTGTTTGCACCTTT

4 S. Ghosh et al. / Molecular

ttached anti-CD8 antibody. The isolated cells were labelled withiotinylated anti-CD62L antibody followed by incubation withtreptavidin microbeads. The cell suspension was then loadedn a MACS column and allowed to pass through. The cellu-ar fraction that stuck to the walls of the tube is the requiredD8+CD62Lhigh T cell fraction. The purity of cells was checked byow cytometry and cell preparation with >90% purity was taken forxperiment.

.8. Flow cytometric analysis of cells from lymph nodes andpleen

Single cells were prepared from the harvested lymph nodes andpleens of mice from the 3 groups (PBS, SarAg, SarAg + NLGP) asep-ically. T cell surface phenotypic markers were assessed by flowytometry after labeling of cells (1 × 106) with mouse fluorescenceabelled antibodies (CD8, CD44, CD62L, CD69, Ly6C and CD127)s per manufacturer’s recommendation. After labeling, cells wereashed in FACS buffer (PBS with 1% FBS). Similarly, intracellularolecules, like, IFN�, Granzyme B were stained with anti-mouse

uorescence labelled antibodies using cytofix-cytoperm reagentss described before (Goswami et al., 2014; Mallick et al., 2013b). Inll flow cytometric staining, cells were fixed with 1% paraformalde-yde in PBS and cytometry was performed with Cell Quest softwaren a FACS Caliber (Becton Dickinson, Mountainview, CA). Suitableegative isotype controls were used to rule out the backgrounduorescence. Percentage of each positive population and MFI wereetermined by using quadrant statistics.

.9. Detection of antigen specific CD8+ T cells

Splenocytes and lymph node cells were prepared from the har-ested spleens and lymph nodes. The cells were stimulated for 6r in vitro with/without SarAg (5 �g/ml) in RPMI-1640 containing0% FBS and brefeldin A (1 mg/ml). Brefeldin A was used to detect

FN� intracellularly in CD8+ T cells by flow cytometry. The cellsere harvested and stained with anti-mouse CD8-PE first, then,

ells were permeabilized with CytoPerm wash buffers and stainedith anti-mouse IFN�. The gate for IFN�+ cells was selected on annstimulated sample for each mouse. This value was subtractedrom the antigen stimulated values to determine the frequency ofntigen specific CD8+ T cells (Hamilton and Harty, 2002).

.10. Proliferation of TCM cells in in vitro

The TCM cells (CD8+CD62Lhigh T cells) were purified from threeroups of mice and co-cultured with the SarAg/MelAg pulsed DCsor 72 hr at 37 ◦C with 5% CO2. After incubation, the cells were pel-eted down and 70–80% chilled ethanol was added to the pellet1–5 × 107 cells) with vortexing and then incubated at −20 ◦C for

hr. The cells were washed twice with staining buffer and cen-rifuged for 10 min at 200 × g. Cells were then resuspended to aoncentration of 1 × 107 cells/ml and transferred into each sampleube to add anti-Ki67 antibody and mixed gently. The tubes werencubated at room temperature for 20–30 min in dark. Cells were

ashed and 0.5 ml of staining buffer was added to each tube andnalyzed flow cytometrically.

.11. Intracellular staining of phosphorylated proteins

The lymph node and spleen cells were stimulated with SarAg1 �g/ml) for 30 min at 37 ◦C with 5% CO2. After stimulation, the

ells were washed and fixed in fresh 2% paraformaldehyde for0 minutes at room temperature (Krutzik and Nolan, 2003), fol-

owed by permeabilization of cells with 90% methanol for 1 h at◦C. Next, the cells were stained with anti-mouse purified pAKT

Bcl-6- F AGCAAGAACGCCTGCATCCTTC 417 bpBcl-6- R CATCTCTGTATGCTGTGGCGACTG

(thr308), pAKT (ser473), pmTOR, FOXO3, KLF2 antibodies. After30 min of incubation, cells were stained with secondary anti-rabbit-FITC and anti-goat-PE antibodies. The cells were then analysed withCell Quest software on a FACS Calibar (Becton Dickinson, Mountain-view, CA).

2.12. Isolation of RNA and RT–PCR analysis

CD8+CD62Lhigh T cells were purified from lymph nodes andspleens by MACS as mentioned above. Total RNA was isolatedusing the Trizol reagent (Invitrogen, USA). The cDNA synthesis wascarried out using RevertAidTM first strand cDNA synthesis kit (Fer-mentas, K1622) following the manufacturer’s protocol and PCRwere carried out using gene-specific oligonucleotide primers, aslisted in Table 1. PCR products were identified by image analysissoftware for gel documentation (Gel DocTM XR + system, BioRad)following electrophoresis on 1.5% agarose gels, stained with ethid-ium bromide.

2.13. Cytosolic and nuclear lysate preparation

Single cells from lymph nodes were prepared in chilled PBSand then centrifuged at 3000 rpm for 10 min. The cell pellet wasre-suspended in ice-cold nuclear extraction buffer and incubatedfor 1 h at 4 ◦C. The cells were then centrifuged at 6000 rpm for5 min. The supernatant was isolated as the cytosolic fraction. Theremaining pellet was dissolved in nuclear extraction buffer andkept in vortex for 30 min at 4 ◦C. The suspension was centrifuged at12,000 rpm for 10 min. The supernatant was collected as the nuclearfraction (Dimauro et al., 2012). The separation was validated bywestern blot analysis of ‘house keeping marker’ proteins (nuclearHISTONE H1 and cytosolic GAPDH).

2.14. Western blot analysis

Nuclear and cytosolic lysate (50 �g) were separated on 12.5%SDS-polyacrylamide gel and transferred onto a nitrocellulose mem-brane for Western Blotting. Incubation was performed for differentprimary antibodies, e.g., histone h1, GAPDH, KLF2, FOXO3 andpFOXO3 and the procedure was followed as published (Barik et al.,2013a; Banerjee et al., 2014).

2.15. Immunofluorescence studies

For immunofluoroscence analysis, CD8+ T cells from lymphnodes were harvested to prepare cytospin slides. Cells were fixedwith 2% paraformaldehyde and then permeabilized with 0.1%Triton-X-100. All washing steps were performed using 0.5% BSA

in PBS while blocking steps were carried out using 2% BSA in PBS.For detection of the presence of pFOXO3, cells were incubated withpurified anti-mouse pFOXO3 antibody, followed by FITC conjugatedanti-rabbit secondary antibody. All sections were counterstained

S. Ghosh et al. / Molecular Immunology 71 (2016) 42–53 45

Fig. 1. Generation of antigen specific CD8+ T cell response after sarcoma antigen (SarAg) vaccination with/without NLGP. (A) Vaccination schedule of Swiss mice with SarAgwith/without NLGP. (B) Percent positive CD8+ T cells during and after SarAg and SarAg + NLGP vaccination in lymph nodes and spleen, as determined by flow cytometry.(C) Ex vivo cytotoxicity of sarcoma and carcinoma cells by CD8+ T cells from lymph nodes and spleens of the PBS, SarAg and SarAg + NLGP vaccinated mice as determinedby LDH release assay. (D) Percent positive antigen experienced CD8+ cells expressing CD44 in lymph nodes and spleens as studied by flow cytometry in three groups ofmice. (E) Frequencies of antigen specific CD8+ T cells expressing IFN� after in vitro antigenic stimulation with SarAg and CarAg in the presence of brefeldinA in lymph nodesand spleens as monitored by flow cytometry. Bar diagram represents mean ± SD of three individual observations from each group at each time point. ***p < 0.001, **p < 0.01,*

wD

2

mtM

2

SPem

2

etdnu

p < 0.05.

ith DAPI and then mounted. Images were acquired using LeicaM 1000, Fluorescent Microscope (Leica, BM 4000B, Germany).

.16. Cytotoxicity assay

Cellular cytotoxicity of lymph node and spleen cells was deter-ined by measuring lactate dehydrogenase (LDH) released by

arget cells using a commercially available kit (Roche Diagnostics,annheim, Germany) (Barik et al., 2013a; Mallick et al., 2013b).

.17. Detection of cytokine secretion by ELISA

IL-2 secretion from TCM phenotype cells after co-culture witharAg/MelAg pulsed DCs for 72 h was assessed by ELISA (BD-harmingen, OptEIATM) as per manufacturer’s instruction (Barikt al., 2013a) and optical density was measured at 450 nm usingicroplate reader (BioTek Instruments Inc., Vermont, USA).

.18. Statistical analysis

All results represent the average of separate in vivo and in vitroxperiments. In each experiment a value represents the mean of

hree individual observations and is presented as mean ± standardeviation (SD) or standard error (SE) and p < 0.05 is considered sig-ificant. Statistical significance was established by student’s t-testsing INSTAT 3 Software (Graphpad Inc., USA).

3. Results

3.1. Vaccination with SarAg along with NLGP adjuvant generatessuperior antigen specific CD8+ T cell response

Two groups of mice were vaccinated with SarAg andSarAg + NLGP along with a PBS treated control group (Fig. 1A). Sta-tus of CD8+ T cells was studied in vaccination and post-vaccinationperiod in lymph nodes and spleens, two important secondary lym-phoid organs. In lymph nodes, the frequencies of CD8+ T cellswere increased gradually and reached a peak on day 30, 2 daysafter the completion of vaccination (Fig. 1B). Upregulated CD8+

T cell response was superior in SarAg + NLGP vaccinated groupsthan those mice group vaccinated with SarAg alone, however, noresponse was observed in PBS control mice. As assessed on day 60post vaccination schedule, a decline in the number of CD8+ T cellswas observed in both the groups. The decline might correspond tothe contraction phase of the CD8+ T cell immune response. On thecontrary, in spleen a peak response was observed on day 10 (aftertwo vaccinations) that yielded highest frequencies of CD8+ T cellsin the SarAg + NLGP vaccinated mice (Fig. 1B). The post vaccina-tion contraction phase was observed in both the vaccinated groups.Repetition of the similar experiment yielded identical result.

Further, antigen specific cytotoxic ability of the CD8+ T cells afterSarAg and SarAg + NLGP vaccination was checked toward sarcomaand carcinoma cells. In case of both lymph node and spleen, extent

of cytotoxicity was significantly greater toward antigen positivesarcoma than carcinoma cells (Sarcoma vs Carcinoma, 80% vs 30%)(Fig. 1C). Combination of NLGP with SarAg vaccine provided supe-rior cytotoxicity of CD8+ T cells over SarAg alone toward sarcoma

46 S. Ghosh et al. / Molecular Immunology 71 (2016) 42–53

Fig. 2. Enhanced generation of SarAg specific CD8 memory phenotypic cells by NLGP adjuvant. (A) Gating strategy of lymphocytes and CD8+ cells in flow cytometry. (B)Expression of CD69 on CD8+ T cells in lymph nodes and spleens of mice at different time points, as studied by flow cytometry. (C) Expression of Ly6C in CD8+ T cells inlymph nodes and spleens of mice, as determined by flow cytometry. (D) Surface expression of IL-7R� on CD8+ T cells in lymph nodes and spleens of mice, as studied byflow cytometry. Representative histogram represents memory phenotype cells on day 60. (E) Transcriptional level analysis of IL-7R� gene at different time points in CD8+ Tc CR ano p < 0.0

otTcndTwwca(ot

ells of lymph nodes and spleens from representative mice, as determined by RT-Pf three individual observations from each group at each time point. ***p < 0.001, **

n post vaccination days 10, 30 and 60 (n = 3 in each group at eachime point). Next, we studied the antigen priming status of CD8+

cells with CD44 marker expression on lymph node and spleenells (n = 3 in each group at each time point) (Fig. 1D). In lymphodes, the antigen priming is significantly higher on day 30 anday 60 in the vaccinated mice with more number of CD8+CD44+

cells in SarAg + NLGP cohort. A similar trend of CD44 expressionas found in spleen, but maximum number of CD8+CD44+ T cellsas detected on day 10. Such antigen (SarAg) specificity of CD8+ T

ells was studied by SarAg specific IFN� release assay after in vitro

ntigen stimulation, using carcinoma antigen (CarAg) as controlFig. 1E) (n = 3 in each group at each time point). The frequencyf antigen specific CD8+ T cells is higher in SarAg + NLGP cohorthan SarAg group on day 30/60 in lymph nodes and day 10/30/60

d bar diagram shows mean relative expression. Bar diagram represents mean ± SD1, *p < 0.05.

in spleens, suggesting more antigen specific CD8+ T cells in bothimmune organs after SarAg + NLGP vaccination.

3.2. Vaccination with SarAg along with NLGP adjuvant generatesenhanced number of memory phenotypic cells

In absence of any inflammation or post antigenic stimulationphase, the CD8+ memory T cells generated are maintained in a qui-escent state (with minimum activation) (Alp et al., 2015). To studythis particular aspect of memory T cells in our settings, the acti-

vation status of CD8+ T cells (as gated in Fig. 2A) was examinedand found that in SarAg and SarAg + NLGP cohort the number ofCD8+CD69+ T cells is higher on day 10 and 30 in lymph nodes aswell as spleens (during and just after the end of vaccination) (n = 3

Immu

ihiilo1iwmoigt2cecnutstdvtctg

3gn

ttSL(clwpCiS(acdcCaoCIirTcMdtpv

S. Ghosh et al. / Molecular

n each group at each time point). Whereas on day 60, the observedigh CD69 expression was downregulated on CD8+ T cells more

ntensely in mice of SarAg + NLGP cohort than SarAg mice groupndicating low activation or resting state of the CD8+ T cells in bothymph nodes and spleen (Fig. 2B). Next, we studied the expressionf Ly6C, a good marker of memory CD8+ T cells (Walunas et al.,995; Hanninen et al., 2011), in all the three cohorts (Fig. 2C) (n = 3

n each group at each time point). The number of CD8+Ly6C+ T cellsas increased on days 10 and 30 in both vaccinated cohorts and wasaintained till day 60 in these two immune organs. The number

f CD8+Ly6C+ T cells is higher in SarAg + NLGP cohort than SarAg,ndicating more number of memory phenotype cells in the formerroup of mice. Furthermore, memory CD8+ T cells require IL-7 forheir survival and maintenance (Bradley et al., 2005; Scluns et al.,000; Nanjappa et al., 2008; Dong et al., 2012), thus, memory pre-ursors are identified as CD8+IL-7R�+. Accordingly, we studied thexpression of IL-7R� (CD127) on CD8+ T cells during and after vac-ination (Fig. 2D) (n = 3 in each group at each time point). In lymphodes, the number of CD8+CD127+ T cells was significantly upreg-lated in SarAg + NLGP than SarAg cohort, which was maintainedill day 60 in the former. A similar observation was found in case ofplenic CD8+ T cells. The CD127 expression is also validated in theranscriptional level by RT–PCR and results show the similar dayependent increment of IL-7R� in lymph nodes and spleens fromaccinated mice groups (n = 3 in each group at each time point), par-icularly in those injected with SarAg + NLGP (Fig. 2E). Therefore, itan be concluded that at this point more number of memory pheno-ypic CD8+ T cells are generated in SarAg + NLGP cohort than SarAgroup as CD69lowLy6ChighCD127high phenotype.

.3. Vaccination with SarAg along with NLGP adjuvant facilitateseneration of central memory phenotype of CD8+ T cells in lymphodes with more effector memory phenotypes in spleen

Memory phenotype cells are broadly classified into two sub-ypes: central memory (TCM) and effector memory (TEM) based onheir migration patterns (Klebanoff et al., 2006; Sallusto et al., 2004;allusto et al., 1999; Masopust et al., 2001; Bouneaud et al., 2005;anzavechhia and Sallusto, 2005). As reported by Klebanoff et al.2006) and Sallusto et al. (2004); Sallusto et al. (1999) in TCMells the migration molecules like CD62L and CCR7 are upregu-ated, whereas, those are downregulated in TEM cells. Therefore,

e studied the expression of these molecules on the antigenrimed CD8+ T cells (CD8+CD44+). In lymph nodes, the number ofD8+CD44+CD62L+ cells (representing TCM phenotype) was signif-

cantly upregulated in SarAg + NLGP cohort with modest increase inarAg monotherapy group on day 30 post vaccination time frameFig. 3A, 3C) (n = 3 in each group at each time point) that, declinesgain on day 60 probably coinciding with the decrease in CD8+ Tell frequency. However, the MFI values gradually increases fromay 0 to day 60 with greater expression of CD62L in SarAg + NLGPohort than SarAg (data not shown). In spleen, the number ofD8+CD44+CD62L+ T cells increases on day 10 post vaccinationnd then decreases gradually toward day 60 (with more numberf CD8+CD44+CD62L+ TCM cells in SarAg + NLGP cohort) (Fig. 3A,). Next, we studied the expression of CCR7 on CD8+CD44+ T cells.

n lymph nodes, the frequencies of CD8+CD44+CCR7+ T cells (n = 3n each group at each time point) increase on day 30 and theneduce on day 60 in both the vaccinated cohorts. However, theCM phenotype cells were more in SarAg + NLGP than only SarAgohort till day 60 (Fig. 3D, F). In spleens, the number as well asFI of CD8+CD44+CCR7+ T cells is high on day 10 and gradually

ecreases toward day 60 (Fig. 3D, F). Effector memory pheno-ype cells (CD8+CD44+CD62L− and CD8+CD44+CCR7−) were morerominent in spleen on days 10, 30 and 60 during SarAg + NLGPaccination in comparison to lymph nodes (Fig. 3B, E).

nology 71 (2016) 42–53 47

The central memory CD8+ T cells were characterized further onthe basis of upregulation of transcription factors, like, eomesoder-min (eomes) and bcl-6 (Banerjee et al., 2010; Ichii et al., 2004).RT-PCR analysis (n = 3 in each group) demonstrated that eomesand bcl-6 were expressed in greater extent in the CD8+CD62Lhigh

T cells from SarAg + NLGP vaccinated mice over SarAg cohort onday 60, where less increment was noted. No such changes weredemonstrated in non-vaccinated controls (Fig. 3G).

3.4. SarAg with NLGP vaccination generates central memoryphenotype of lymph node CD8+ T cells in association withdownregulation of mTOR/AKT and upregulation of KLF2/FOXO3

In quiescent state of a cell on post vaccination day 60, CD8+

T cells maintain their CD62LhighCCR7highIL7R�high status throughKLF2 and FOXO3 signaling in nucleus (Navarro and Cantrell, 2014;Michelini et al., 2013; Cao et al., 2010). After its phosphorylationwith help of AKT/mTOR signaling FOXO moves to cytoplasm. Foxotranscriptional factors are nuclear but when phosphorylated theyexit the nucleus and form a complex with 14-3-3 in the cyto-sol thereby terminating their transcriptional activity (Finlay andCantrell, 2011). Accordingly, it was observed that on post vac-cination day 60, lymph node cells from SarAg + NLGP vaccinatedmice showed downregulated phosphorylation of mTOR than SarAgcohort (Fig. 4A). Further, we studied the expression of pAKT (ser)(Fig. 4B) and pAKT (thr) (Fig. 4B) in CD8+ T cells. No change inexpression was observed between SarAg and SarAg + NLGP groupsfor pAKT (ser), however, the pAKT (thr) expression was reduced inSarAg + NLGP cohort than SarAg vaccinated mice. Such downregu-lated phosphorylation of mTOR and AKT (thr) might be associatedwith reduced phosphorylation of FOXO (Navarro and Cantrell,2014). In the nucleus, FOXOs induce expression of transcriptionfactor KLF2, which directly regulates transcription of genes encod-ing CD62L and CCR7 that control T cell trafficking. Flow cytometricanalysis demonstrated increased KLF2 expression (Fig. 4C) andreduced pFOXO3 (Fig. 4D) expression on day 60 in mice vacci-nated with SarAg + NLGP (n = 3 in each group). Moreover, westernblot analysis showed upregulated expression of FOXO3 and KLF2in nucleus and reduced pFOXO3 levels in the cytoplasm in theSarAg + NLGP vaccinated mice than the SarAg cohort in CD8+ T cells(Fig. 4E) (n = 3 in each group). Moreover, the immunofluorescencestudies show decreased expression of pFOXO3 in the cytoplasm ofCD8+ T cells from SarAg + NLGP vaccinated mice than SarAg cohort(Fig. 4F). High nuclear expression of FOXO3 as well as KLF2 canbe corroborated with the increased CD62L and CCR7 expression incentral memory CD8+ T cells from the lymph nodes in SarAg + NLGPvaccinated mice. Similar observation was not detected in spleen(data not shown).

3.5. Vaccination with SarAg with NLGP adjuvant induces betterproliferation of central memory cells and operates effectorresponse on antigen re-encounter

The TCM cells are best characterized by their proliferative poten-tial on antigen-reencounter (Klebanoff et al., 2005; Zaph et al.,2004). Therefore, on post vaccination day 60 (which probably coin-cides with the beginning of the memory phase), the MACS purifiedCD8+CD44highCD62Lhigh TCM population from lymph node andspleen cells was studied for proliferation in vitro. Accordingly, TCMphenotype cells from vaccinated mice of both cohorts were co-cultured with SarAg and MelAg (unrelated antigen) pulsed ex-vivogenerated mDCs. The MelAg was used as an unrelated, non-specific

control to see the antigen specificity of the generated TCM phe-notype cells for SarAg. In both lymph nodes and spleens, TCMcells from SarAg + NLGP cohort showed superior SarAg specificproliferation than SarAg group. Such antigen specific prolifera-

48 S. Ghosh et al. / Molecular Immunology 71 (2016) 42–53

Fig. 3. Enhanced generation of central memory phenotype of CD8+ T cells by NLGP. (A–C) Bar diagrams and representative diagrams showing surface expression of CD44 andCD62L, as studied by flow cytometry in lymph nodes and spleens of PBS, SarAg and SarAg + NLGP vaccinated mice. (D–F) Bar diagrams and representative diagrams showingsurface expression of CCR7 as studied by flow cytometry in lymph nodes and spleens. Representative histogram represents TCM population on day 60. (G) Transcriptionall +CD62r e poin

tAprog

hgeipnootctoe

evel analysis of eomesodermin and bcl-6, as performed by RT–PCR on day 60 in CD8epresents mean ± SD of three individual observations from each group at each tim

ion was significantly less when MelAg pulsed DCs were used.gain, the frequencies of proliferating Ki67+ TCM cells are morerominent in case of lymph nodes than spleen. This is further cor-oborated with the increased IL-2 secretion from lymph node T cellsf SarAg + NLGP vaccinated mice than SarAg group (n = 4 in eachroup) (Fig. 5A).

Antigen primed memory T cells show a more pronounced andeightened response on secondary challenge with the same anti-en, thereby, providing protective immunity to the host (Robertst al., 2005; Bachman et al., 2005). Thus, two vaccinated mice werenoculated with sarcoma cells (SarAg+) on day 60 post vaccinationeriod (Fig. 5B) and tumor growth was monitored in vaccinated andon-vaccinated mice (Fig. 5B). No tumor was visible in both cohortsf vaccinated mice (SarAg and SarAg + NLGP) till day 102 (all micef PBS group died at this time point so tumor growth was moni-ored till this day). Superior protection was noticed in SarAg + NLGP

ohort where none of the mice developed tumor (0/6) till the end ofhe experiment i.e., day 120 (day 60 post tumor inoculation). On thether hand, tumor appeared in 4/6 mice from SarAg cohort (n = 6 inach group) (Fig. 5B).

L+ T cells from lymph nodes and spleens and mean relative expression. Bar diagramt. ***p < 0.001, **p < 0.01, *p < 0.05.

The memory CD8+ T cells so generated by SarAg + NLGP vaccina-tion impart protection to the host by means of effector functions,as described (Lalvani et al., 1997). In order to identify the rationalebehind tumor growth restriction in vaccinated mice in our set-tings, the memory–effector conversion (by downregulated CD62Land increased Granzyme B expression) was next studied andmice from SarAg and SarAg + NLGP cohorts were sacrificed on day67 post vaccination as outlined in Fig. 5B1 (to study immedi-ate memory-to-effector conversion) to harvest lymph nodes andspleens. Alterations of CD62L and GranzymeB were noted signifi-cantly in SarAg + NLGP cohort than SarAg alone (n = 6 in each group)(Fig. 5C). The CD62L downregulation was noted in both lymphnodes and spleen, whereas, Granzyme B expressed more intenselyin spleen than lymph nodes, exhibiting more effector functionsin spleen. Next, the remaining mice were sacrificed at the end oftumor growth monitoring on day 120 post vaccination (to study late

memory-to-effector conversion, in relation to tumor free status). Inlymph nodes (now the tumor draining lymph node) of SarAg + NLGPvaccinated mice, the frequencies of CD8+ T cells with CD62L expres-sion was higher than SarAg treated mice, whereas, the Granzyme

S. Ghosh et al. / Molecular Immunology 71 (2016) 42–53 49

Fig. 4. Enhanced TCM generation in lymph node depends on upregulated nuclear KLF2 expression and reduced cytosolic FOXO3 phosphorylation. (A–D) Intracellularexpression of pmTOR, pAKT ser, pAKT thr, KLF2 and pFOXO3 in CD8+ T cells on day 60, as determined by flow cytometry in lymph nodes of PBS, SarAg and SarAg + NLGPvaccinated mice. Bar diagram represents mean ± SD of three individual observations on day 60. *p < 0.05, **p < 0.01. Representative histogram in each case is presented. (E)Protein level expression of KLF2, FOXO3 and pFOXO3, as determined by western blot analysis in cytosolic and nuclear cell fractions. Representative blot and mean relativee HISTI toplas

Beaecnlpp

4

mdeoSientwNb

xpression are presented from three groups of mice as described in A. GAPDH andmmunofluorescence study of CD8+ T cells showing expression of pFOXO3 in the cy

expression was prominent in spleen of the same cohort (n = 6 inach group) (Fig. 5D). This coincides with the fact that lymph nodesre primarily the niche for TCM cells (Klebanoff et al., 2006; Sallustot al., 2004; Sallusto et al., 1999). Therefore, at this point it can beoncluded that on post vaccination tumor inoculation the TCM phe-otype cells are proliferating thereby increasing their number in

ymph nodes, whereas, the immediate effector functions are morerominent in spleens compared to lymph nodes (due to more TEMopulations).

. Discussion

We have reported earlier that NLGP restricts the growth ofurine sarcoma and melanoma significantly in CD8+ T cell depen-

ent manner (Mallick et al., 2013b; Barik et al., 2013a,b; Banerjeet al., 2014). In that case, NLGP may help to present antigen (SarAgr MelAg) to CD8+ T cells by influencing APCs (Goswami et al., 2010;arkar et al., 2008) to induce effector functions. In an effort to exam-ne the adjuvanicity of NLGP further to generate central and/orffector memory T cell functions that may participate for mainte-ance of disease free survival and to protect the host from further

umor occurence, Swiss mice were vaccinated with SarAg with orithout NLGP weekly for four weeks in total. NLGP or its precursorLP was proved to be effective as adjuvant to enhance the anti-ody response against different tumor antigens and its adjuvant

ONE H1 were used as controls for cytosolic and nuclear fractions respectively. (F)m in the lymph node of SarAg and SarAg + NLGP vaccinated mice.

efficacy is comparable to Freund’s adjuvant or QS-21 (Baral et al.,2005; Mandal-Ghosh et al., 2007; Sarkar et al., 2008). Additionally,NLGP/NLP is non-toxic and inexpensive (Haque et al., 2006; Mallicket al., 2013a). Here, the status of CD8+ T cells in context of memorygeneration after completion of the SarAg −/+ NLGP vaccination pro-tocol was studied. We have found that following vaccination of micewith SarAg, CD8+ T cell response was upregulated in presence ofNLGP adjuvant, which was maintained till day 30 post vaccinationperiod and declined subsequently as monitored on day 60. RobustCD8+ T cell response following four NLGP vaccinations was reportedearlier (Mallick et al., 2013), however, present results suggest thatdecline in CD8+ T cell pool after termination of vaccination mighthave effector and/or memory functions. Antigen specific effectorfunctions after SarAg + NLGP vaccination were maintained till day60 in lymph node and splenic cells, as denoted by greater cyto-toxicity to SarAg+ sarcoma cells, rather than killing of the SarAg−

carcinoma cells. NLGP helps in obtaining antigen experience of con-cerned T cells which was reflected in CD8+CD44high cells. In courseof this examination, interestingly, we have found greater frequencyof antigen specific CD8+ T cells in SarAg + NLGP cohort than SarAgtreated group on day 60 (Fig. 6A).

Based on this preliminary observation, we checked the acti-

vation status of involved CD8+ T cells. These cells possess highactivation initially on day 10, however, significantly declined after-wards on day 60. Activation in terms of CD69 expression was

50 S. Ghosh et al. / Molecular Immunology 71 (2016) 42–53

Fig. 5. Superior tumor protection and proliferation of TCM cells on SarAg re-encounter in presence of NLGP. (A) Proliferation of Ki67 expressing CD8+CD62Lhigh TCM cellsand IL-2 secretion from lymph nodes and spleens (n = 4 in each group) after their co-culture with SarAg/MelAg pulsed DCs in vitro for 72 h. **p < 0.01, *p < 0.05. (B) On day60, following 4 weeks of vaccination schedule, mice of three groups (n = 6 in each case) were inoculated with sarcoma and tumor growth was monitored till the end of thee nd tumt tracelt ined b

rarcoStctCrctm

CCcTnoStAcS

xperiment (day 120). Diagrammatic representation of the experimental design aumor relapsed mice within day 102–120. (C, D) Surface expression of CD62L and inhree groups of mice (n = 6 in each case) on days 67 and 120 respectively, as determ

eciprocally correlated with Ly6C expression, denoting the gener-tion of memory phenotypes. Maintenance of these memory cellsequires cytokines, like, IL-7, as confirmed by high expression ofoncerned receptors (IL-7R�) in the memory precursor cells. As webtained CD69lowLy6ChighCD127high CD8+ T cells specifically afterarAg + NLGP vaccination, the question was next addressed to knowhe nature of these antigen experienced CD44+ memory cells. Inomparison between SarAg and SarAg + NLGP cohort, it was dis-inctly observed that greater number of CD62L and CCR7 expressingD8+ T cells was detected in the latter group indicating crucialole of NLGP in generation of central memory cell pool. Furtheronfirmation was obtained by observing the greater expression ofranscription factor eomesodermin and bcl-6, markers of central

emory cells.Experimental evidences suggest the generation of

D62LhighCCR7highCD127high CD8+ T cells, those are alsoD44+CD69low, thus indicating the maintenance of sustainedentral memory pool following SarAg + NLGP vaccination. TheseCM phenotypes were maintained by FOXO dependent KLF2uclear expression, as we obtained greater nuclear expressionf KLF2 and reduced cytosolic phosphorylation of FOXO3a inarAg + NLGP vaccinated cohort. This quiescent situation is main-

ained by NLGP adjuvant assisted vaccination that suppresses theKT/mTOR-induced phosphorylation of FOXO and its subsequentytoplasmic translocation. It is important to mention here thatarAg + NLGP vaccination resulted in less phosphorylation of AKT

or growth restriction curve are presented. Table in B represents tumor free andlular expression of Granzyme B on CD8+ T cells from lymph nodes and spleens fromy flow cytometry. **p < 0.01, *p < 0.05.

and mTOR on day 60 post vaccination period than SarAg vacci-nation only. A major role for AKT in the context of TCR signallingis to control the phosphorylation and localization of the Foxotranscription factors, including FOXO1, FOXO3a and FOXO4 (Finlayand Cantrell, 2011). The Foxo transcription factors are nuclear andactive in quiescent T cells, but when phosphorylated by AKT theyexit the nucleus and form a complex with 14-3-3 proteins in thecytosol, thereby, terminating their transcriptional activity (Navarroand Cantrell, 2014; Michelini et al., 2013). Accordingly, quiescentstate is maintained better by KLF2/FOXOs after SarAg + NLGPvaccination (Fig. 6B).

Above discussion establishes the fact that SarAg + NLGP vaccinegenerates superior central memory response than SarAg vaccineonly. To ascertain the functional utility of such memory response,related (SarAg) and unrelated (MelAg) antigens were presented toCD8+CD44highCD62Lhigh TCM phenotype cells via DCs to observetheir proliferation status. NLGP assisted SarAg vaccine showedgreater proliferation, as indicated by more number of proliferativeKi67+ cells, after in vitro exposure of TCM cells to related antigenonly. This in vitro observation provides clue that SarAg + NLGP vac-cine generated memory response might be effective at the timeof tumor initiation for tumor eradication. Further confirmation of

this speculation appeared in in vivo study after inoculation of sar-coma cells in both groups of vaccinated mice on day 60 (whenmice possessing good number of TCM phenotypes) and main-tained till day 120. Although, initially both groups of vaccinated

S. Ghosh et al. / Molecular Immunology 71 (2016) 42–53 51

Fig. 6. Schematic representation of generation of memory CD8+ T cell. (A) Generation of effector cells, memory precursors and two subsets of memory T cells (centralmemory, TCM; effector memory, TEM) in lymph nodes and spleen from SarAg + NLGP and SarAg immunized group on day 60. (B) In case of TCR triggering, FOXOs migrate tocytosol, undergo phosphorylation by 14-3-3. This inhibits expression of KLF2 and transcription of its downstream genes CD62L, CCR7 and IL7R�. But in the absence of TCRt g to eN phost

mlgnstefiwweCswpi

riggering (as in memory T cells), FOXOs are retained in the nucleus thereby leadinLGP mediated increased expression of cell trafficking molecules is by the inhibited

he nucleus.

ice showed tumor restriction in comparison to control mice, inater phase tumor appears in 4 mice among 6 in SarAg vaccineroup, whereas, no tumor was detected in SarAg + NLGP vacci-ated mice. This is in concert to the observations that antigenpecific memory CD8+ T cells can rapidly acquire cytotoxic func-ion upon re-exposure to antigen (Klebanoff et al., 2005; Barbert al., 2003). The experiment was terminated on day 120 to ful-l the ethical requirement as tumor of mice from untreated groupas greater than 250 mm2 in either direction. Tumor restrictionas well correlated with the effector profile of CD8+ T cells as

valuated by upregulation of Granzyme B and downregulation ofD62L. As studies were undertaken in lymph nodes and spleens,ubstantial difference in the number of proliferating effector cells

as noted that can be concluded as on tumor inoculation the TCMhenotype cells are proliferating, thereby, increasing their number

n lymph nodes, the primary niche of TCM cells. However, the CD62L

xpression of KLF2 and transcription of its target genes required for cell trafficking.phorylation of FOXO in the cytoplasm and increased FOXO and KLF2 expression in

downregulation and Granzyme B expression is more prominentin spleen than lymph node. The fact should be mentioned herethat our immunization schedule with adjuvant help from NLGPgenerates both TCM and TEM populations in lymph nodes andspleens. However, the TCM population was predominant in lymphnodes than spleen and the reverse is true for TEM. This observa-tion can be corroborated with high effector GranzymeB expressionin spleen and upregulated CD62L expression in lymph nodes onday 120 in SarAg + NLGP vaccinated mice at the termination ofexperiment. This is probably responsible for the tumor free stateof SarAg + NLGP cohort in post vaccination, post tumor inocula-tion phase. This observation is supported by the fact that TEM cellspresent an immediate defence, while TCM cells proliferate in the

secondary lymphoid organs thereby producing a supply of neweffectors (Roberts et al., 2005; Harris et al., 2002).

5 Immu

tfttceva

C

A

ILNata5cm

R

A

B

B

B

B

B

B

B

B

B

B

C

C

C

2 S. Ghosh et al. / Molecular

In conclusion, it can be stated that adjuvant help from non-oxic NLGP (Mallick et al., 2013a) may generate robust effectorunctions of CD8+ T cells during SarAg vaccination that can be main-ained as antigen specific memory in quiescent state after end ofhe immunization period. Most effectively, such memory responsean be converted to effector response on further sarcoma antigenicxposure. Thus, NLGP may be considered as an ideal vaccine adju-ant with SarAg vaccine and, hopefully, such adjuvanicity would bepplicable for other vaccines too.

onflict of interest

None.

cknowledgements

We acknowledge Dr. Jaydip Biswas, Director, CNCI, Kolkata,ndia, for providing necessary facilities. Thanks to Dr. Subrataaskar, Burdwan University, India, for his help in characterization ofLGP. The work was partially supported by Department of Sciencend Technology, New Delhi (DST/INSPIRE FELLOWSHIP/2011/188o SG and DST Young Scientist Scheme SB/YS/LS-289/2013 to AB)nd Indian Council of Medical Research, New Delhi (grant no.9/6/2011/BMS/TRM). The funders had no role in study design, dataollection and analysis, decision to publish, or preparation of theanuscript.

eferences

lp, O.S., Durlanik, S., Schulz, D., McGrath, M., Grun, J.R., Bardua, M., Ikuta, K.,Sgouroudis, E., Riedel, R., Zehentmeier, S., Hauser, A.E., Tsuneto, M., Melchers,F., Tokoyoda, K., Chang, H.D., Thiel, A., Radbruch, A., 2015. Memory CD8+ T cellscolocalize with IL-7+ stromal cells in bone marrow and rest in terms ofproliferation and transcription. Eur. J. Immunol. 45, 975–987.

achmann, M.F., Wolint, P., Schwarz, K., Oxenius, A., 2005. Recall proliferationpotential of memory CD8+ T cells and antiviral protection. J. Immunol. 175,4677–4685.

anerjee, A., Gordon, S.M., Intlekofer, A.M., Paley, M.A., Mooney, E.C., Lindsten, T.,Wherry, E.J., Reiner, S.L., 2010. Cutting edge: the transcription factoreomesodermin enables CD8+ T cells to compete for the memory cell niche. J.Immunol. 185, 4988–4992.

anerjee, S., Ghosh, T., Barik, S., Das, A., Ghosh, S., Bhuniya, A., Bose, A., Baral, R.,2014. Neem leaf glycoprotein prophylaxis transduces immune dependent stopsignal for tumor angiogenic switch within tumor microenvironment. PLoS One9, e110040.

aral, R., Mandal, I., Chattopadhyay, U., 2005. Immunostimulatory neem leafpreparation acts as an adjuvant to enhance the efficacy of poorly immunogenicB16 melanoma surface antigen vaccine. Int. Immunopharmacol. 5, 1343–1352.

arber, D.L., Wherry, J.E., Ahmed, R., 2003. Cutting edge: rapid in vivo killing bymemory CD8 T cells. J. Immunol. 171, 27–31.

arik, S., Banerjee, S., Mallick, A., Goswami, K.K., Roy, S., Bose, A., Baral, R., 2013a.Normalization of tumor microenvironment by neem leaf glycoproteinpotentiates effector T cell functions and therapeutically intervenes in thegrowth of mouse sarcoma. PLoS One 8, e66501.

arik, S., Banerjee, S., Sarkar, M., Bhuniya, A., Roy, S., Bose, A., Baral, R., 2013b.Neem leaf glycoprotein optimizes effector and regulatory functions withintumor microenvironment to intervene therapeutically the growth of B16melanoma in C57BL/6 mice. Trials Vaccinol.

ouneaud, C., Garcia, Z., Kourilsky, P., Pannitier, C., 2005. Lineage relationships,homeostasis, and recall capacities of central and effector memory CD8 T cellsin vivo. J. Exp. Med. 201, 579–590.

radley, L.M., Hanes, L., Swain, S.L., 2005. IL-7: maintaining T-cell memory andachieving homeostasis. Trends Immunol. 26, 172–176.

uonaguro, L., Petrizzo, A., Tornesello, M.L., Buonaguro, F.M., 2011. Translatingtumor antigens into cancer vaccines. Clin. Vac. Immunol. 18, 23–34.

ao, Z., Sun, X., Icli, B., Wara, A.K., Feinberg, M.W., 2010. Role of Kruppel-like factorsin leukocyte development function, and disease. Blood 116, 4404–4414.

hakraborty, K., Bose, A., Chakraborty, T., Sarkar, K., Goswami, S., Pal, S., Baral, R.,2010. Restoration of dysregulated CC chemokine signaling formonocyte/macrophagechemotaxis in head and neck squamous cell carcinoma

patients by neem leaf glycoprotein maximizes tumor cell cytotoxicity. Cell.Mol. Immunol. 7, 396–408.

unto-Amesty, G., Monzavi-Karbassi, B., Luo, P., Jousheghany, F., Kieber-Emmons,T., 2003. Strategies in cancer vaccines development. Int. J. Parasitol. 33,597–613.

nology 71 (2016) 42–53

Dimauro, I., Pearson, T., Caporossi, D., Jackson, M.J., 2012. A simple protocol for thesubcellular fractionation of skeletal muscle cells and tissue. BMC Res. Notes 5,513–517.

Dong, H., Franklin, N.A., Roberts, D.J., Yagita, H., Glennie, M.J., Bullock, T.N.J., 2012.CD27 stimulation promotes the frequency of IL-7 receptor-expressing memoryprecursors and prevents IL-12-mediated loss of CD8 + T cell memory in theabsence of CD4 + T cell help. J. Immunol. 188, 3829–3838.

Dubensky Jr., T.W., Reed, S.G., 2010. Adjuvants for cancer vaccines. Semin.Immunol. 22, 155–161.

Finn, O.J., 2003. Cancer vaccines: between the idea and the reality. Nat. Rev.Immunol. 3, 630–641.

Finlay, D., Cantrell, D., 2011. Metabolism: migration and memory in cytotoxic Tcells. Nat. Rev. Immunol. 11, 109–117.

Goswami, K.K., Barik, S., Sarkar, M., Bhowmick, A., Biswas, J., Bose, A., Baral, R.,2014. Targeting STAT3 phosphorylation by neem leaf glycoprotein preventsimmune evasion exerted by supraglottic laryngeal tumor induced M2macrophages. Mol. Immunol. 59, 119–127.

Goswami, S., Bose, A., Sarkar, K., Roy, S., Chakraborty, T., Sanyal, U., Baral, R., 2010.Neem leaf glycoprotein matures myeloid derived dendritic cells and optimizesanti-tumor T cell functions. Vaccine 28, 1241–1252.

Hamilton, S.E., Harty, J.T., 2002. Quantitation of CD8+ T cell expansion, memory,and protective immunity after immunization with peptide-coated dendriticcells. J. Immunol. 169, 4936–4944.

Hanninen, A., Maksimow, M., Alam, C., Morgan, D.J., Jalkanen, S., 2011. Ly6Csupports preferential homing of central memory CD8+ T cells into lymphnodes. Eur. J. Immunol. 41, 634–644.

Hariharan, K., Braslawsky, J., Black, A., Raychaudhuri, S., Hanna, N., 1995. Theinduction of cytotoxic T cells and tumor regression by soluble antigenformulation. Cancer Res. 55, 3486–3489.

Harris, N.L., Watt, V., Ronchese, F., Le Gros, G., 2002. Differential T cell function andfate in lymph node and nonlymphoid tissues. J. Exp. Med. 195, 317–326.

Haque, E., Mandal, I., Pal, S., Baral, R., 2006. Prophylactic dose of neem (Azadirachtaindica) leaf preparation restricting murine tumor growth is nontoxic,hematostimulatory and immunostimulatory. Immunopharmacol.Immunotoxicol. 28, 33–50.

Ichii, H., Sakamoto, A., Kuroda, Y., Tokuhisa, T., 2004. Bcl-6 acts as an amplifier forthe generation and proliferative capacity of central memory CD8+ T cells. J.Immunol. 173, 883–891.

Kaech, S.M., Wherry, E.J., Ahmed, R., 2002. Effector and memory T-celldifferentiaion: implications for vaccine development. Nat. Rev. Immunol. 2,251–262.

Kalli, F., Machiorlatti, R., Battaglia, F., Parodi, A., Conteduga, G., Ferrara, F., Proietti,M., Tardito, S., Sanguineti, M., Millo, E., Fenoglio, D., De Palma, R., Inghirami, G.,Filaci, G., 2013. Comparative analysis of cancer vaccine settings for theselection of an effective protocol in mice. J. Transl. Med. 1, 1–11.

Klebanoff, C.A., Gattinoni, L., Restifo, N.P., 2006. CD8+ T cell memory in tumorimmunology and immunotherapy. Immunol. Rev. 211, 214–224.

Klebanoff, C.A., Gattinoni, L., Torabi-Parizi, P., Kerstann, K., Cardones, A.R.,Finkelstein, S.E., Palmer, D.C., Antony, P.A., Hwang, S.T., Rosenberg, S.A.,Waldmann, T.A., Restifo, N.P., 2005. Central memory self/tumor reactive CD8+T cells confer superior anti-tumor immunity compared with effector memoryT cells. Proc. Natl. Acad. Sci. U. S. A. 102, 9571–9576.

Krutzik, P.O., Nolan, G.P., 2003. Intracellular phospho-protein staining techniquesfor flow cytometry: monitoring single cell signaling events. Cytometry A 55,61–70.

Lalvani, A., Brookes, R., Hambleton, S., Britton, W.J., Hill, A.V., McMichael, A.J., 1997.Rapid effector function in CD8+ memory T cells. J. Exp. Med. 186, 859–865.

Lanzavechhia, A., Sallusto, F., 2005. Understanding the generation and function ofmemory T cell subsets. Curr. Opin. Immunol. 17, 326–332.

Lowry, O.H., Rosenbrough, N.J., Farr, A.L., Randall, R.J., 1951. Protein measurementwith the Folin phenol reagent. J. Biol. Chem. 193, 265–275.

Mallick, A., Ghosh, S., Banerjee, S., Majumder, S., Das, A., Mondal, B., Barik, S.,Goswami, K.K., Pal, S., Laskar, S., Sarkar, K., Bose, A., Baral, R., 2013a. Neem leafglycoprotein is nontoxic to physiological functions of swiss mice and spraguedawley rats: histological, biochemical and immunological perspectives. Int.Immunopharmacol. 15, 73–83.

Mallick, A., Barik, S., Goswami, K.K., Banerjee, S., Ghosh, S., Sarkar, K., Bose, A.,Baral, R., 2013. Neem Leaf glycoprotein activates CD8+ T cells to promotetherapeutic anti-tumor immunity inhibiting the growth of mouse sarcoma.PLoS One 8, e47434.

Mallick, A., Barik, S., Ghosh, S., Roy, S., Sarkar, K., Bose, A., Baral, R., 2014.Immunotherapeutic targeting of established sarcoma in Swiss mice bytumor-derived antigen-pulsed NLGP matured dendritic cells is CD8+ T-celldependent. Immunotherapy 6, 821–831.

Mandal-Ghosh, I., Chattopadhyay, U., Baral, R., 2007. Neem leaf preparationenhances Th1 type immune response and anti-tumor immunity against breasttumor associated antigen. Cancer Immun. 7, 8–17.

Masopust, D., Vezys, V., Marzo, A.L., Lefrancois, L., 2001. Preferential localization ofeffector memory cells in nonlymphoid tissue. Science 291, 2413–2417.

Michelini, R.H., Deodens, A.L., Goldwrath, A.W., Hedrick, S.M., 2013. Differentiationof CD8 memory T cells depends on Foxo1. J. Exp. Med. 210, 1189–1200.

Mohan, T., Verma, P., Rao, D.N., 2013. Novel adjuvants and delivery vehicles forvaccine development: a road ahead. Indian J. Med. Res. 138, 779–795.

Nanjappa, S.G., Wallent, J.H., Morre, M., Suresh, M., 2008. Effects of IL-7 on memoryCD8 T cell homeostasis are influenced by the timing of therapy in mice. J. Clin.Invest. 118, 1027–1039.

Immu

N

R

S

S

S

Walunas, T.L., Bruce, D.S., Dustin, L., Loh, D.Y., Bluestone, J.A., 1995. Ly6C is amarker of memory CD8+ T cells. J. Immunol. 155, 1873–1883.

S. Ghosh et al. / Molecular

avarro, M.N., Cantrell, D.A., 2014. Serine-threonine kinases in TCR signaling. Nat.Immunol. 15, 808–814.

oberts, A.D., Ely, K.H., Woodland, D.L., 2005. Differential contributions of centraland effector memory T cells to recall responses. J. Exp. Med. 202, 123–133.

aha, A., Baral, R.N., Chatterjee, S.K., Mohanty, K., Pal, S., Foon, K.A., Primus, F.J.,Krieg, A.M., Weiner, G.J., Bhattacharya-Chatterjee, M., 2006. CpGoligonucleotides enhance the tumor antigen-specific immune response of ananti-idiotype antibody-based vaccine strategy in CEA transgenic mice. CancerImmunol. Immunother. 55, 515–527.

allusto, F., Lenig, D., Forster, R., 1999. Two subsets of memory T lymphocytes withdistinct homing potentials and effector functions. Nature 401, 708–712.

allusto, F., Geginat, J., Lanzavecchia, A., 2004. Central memory and effectormemory T cell subsets: function: generation and maintenance. Annu. Rev.Immunol. 22, 745–763.

nology 71 (2016) 42–53 53

Sarkar, K., Bose, A., Chakraborty, K., Haque, E., Ghosh, D., Goswami, S., Chakraborty,T., Laskar, S., Baral, R., 2008. Neem leaf glycoprotein helps to generatecarcinoembryonic antigen specific antitumor immune responses utilizingmacrophage mediated antigen presentation. Vaccine 26, 4352–4362.

Scluns, K.S., Kieper, W.C., Jameson, S.C., Lefrancois, L., 2000. Interleukin-7 mediatesthe homeostasis of naïve and memory CD8 T cells in vivo. Nat. Immunol. 1,426–432.

Zaph, C., Uzonna, J., Beverly, S.M., Scott, P., 2004. Central memory T cells mediatelong-term immunity to Leishmania major in the absence of persistentparasites. Nat. Med. 10, 1104–1110.