Efficient generation of survivin-specific cytotoxic T lymphocytes from healthy persons in vitro: quantitative and qualitative effects of CD4+ T cells.

Abstract

For the adoptive immunotherapy and the study of cytotoxic T lymphocytes (CTLs) in human, efficient in vitro generation of CTLs is needed. However, it is still difficult to induce T cells specific for naïve antigens in vitro even though dendritic cells (DCs) as potent APCs are used. In this study, we investigated quantitative and qualitative effects of CD4+ T cells during in vitro stimulation of CD8+ T cells from healthy donors using DCs transduced with adenovirus vector expressing human survivin (Adv-survivin). CTLs were not efficiently induced in the absence of CD4+ T cells or in CD25+ depleted CD4+ T cells. When the ratio of CD4+:CD8+ T cells was quantitatively decreased from 2:1 to 1:2, proliferation of CTLs specific for survivin was gradually increased. Because DCs pulsed with HCMV pp65 protein could activate CD4+ T cells to secrete Th1 cytokines, the use of pp65 protein as an adjuvant induced higher numbers and frequencies of CTLs. Furthermore, Th1 conditioning of CD4+ T cells augmented this generation of CTLs. These results suggest that both quantitative and qualitative modulation of CD4+ T cells including the number and Th1 polarization may be required for the efficient induction of CTLs specific for tumor antigens in vitro.

Full text

Page 1

Vaccine 26 (2008) 3987–3997
Contents lists available at ScienceDirect
Vaccine
journal homepage: www.elsevier.com/locate/vaccine
Efficient generation of survivin-specific cytotoxic T lymphocytes from healthy
persons in vitro: Quantitative and qualitative effects of CD4+ T cells
Eun-Kyung Kima, Hyun-Il Choa,b, Sung Hee Yoona, Min-Ji Parka, Hyun-Jung Sohnb,
Hyung-Jin Kimc, Seong-Taek Ohc, Tai-Gyu Kima,b,∗
aDepartment of Microbiology, Catholic University of Korea, 505 Banpo-Dong, Seocho-Gu, Seoul 137-701, South Korea
bCatholic Hematopoietic Stem Cell Bank, Catholic University of Korea, 505 Banpo-Dong, Seocho-Gu, Seoul 137-701, South Korea
cDepartment of Surgery, Kang-Nam St. Mary’s Hospital, College of Medicine, Catholic University of Korea, 505 Banpo-Dong, Seocho-Gu, Seoul 137-701, South Korea
a r t i c l ei n f o
Article history:
Received 4 March 2008
Received in revised form 22 April 2008
Accepted 15 May 2008
Available online 6 June 2008
Keywords:
Cytotoxic T cells
T helper 1 cells
Survivin
a b s t r a c t
For the adoptive immunotherapy and the study of cytotoxic T lymphocytes (CTLs) in human, efficient in
vitrogenerationofCTLsisneeded.However,itisstilldifficulttoinduceTcellsspecificforna¨ ıveantigensin
vitro even though dendritic cells (DCs) as potent APCs are used. In this study, we investigated quantitative
and qualitative effects of CD4+ T cells during in vitro stimulation of CD8+ T cells from healthy donors
using DCs transduced with adenovirus vector expressing human survivin (Adv-survivin). CTLs were not
efficiently induced in the absence of CD4+ T cells or in CD25+ depleted CD4+ T cells. When the ratio of
CD4+:CD8+ T cells was quantitatively decreased from 2:1 to 1:2, proliferation of CTLs specific for survivin
was gradually increased. Because DCs pulsed with HCMV pp65 protein could activate CD4+ T cells to
secrete Th1 cytokines, the use of pp65 protein as an adjuvant induced higher numbers and frequencies
of CTLs. Furthermore, Th1 conditioning of CD4+ T cells augmented this generation of CTLs. These results
suggest that both quantitative and qualitative modulation of CD4+ T cells including the number and Th1
polarization may be required for the efficient induction of CTLs specific for tumor antigens in vitro.
© 2008 Elsevier Ltd. All rights reserved.
1. Introduction
Adoptively transferred tumor-reactive CD8+ T cell clones medi-
ated an antigen-specific immune responses characterized by the
elimination of antigen-positive tumor cells and regression of indi-
vidualmetastases[1].Recentstudyhasdemonstratedthatadoptive
transfer of vaccine-induced peripheral blood mononuclear cells
(PBMCs) after lymphodepletion could result in the persistence
of high numbers of tumor-reactive CD8+ T cells in patients with
metastatic melanoma [2]. The ex vivo production of T lympho-
cytes with specific reactivity towards tumor cells is a prerequisite
for designing effective cell therapy approaches for cancer patients.
Many studies have shown that induction of primary tumor reac-
tive T lymphocytes from na¨ ıve precursors in vitro requires strong
professional antigen-presenting cells (APCs) such as dendritic cells
(DCs) [3]. However, generating and expanding primary immune
responses from na¨ ıve precursors in vitro has generally been much
∗Corresponding author at: Department of Microbiology, Catholic University of
Korea, 505 Banpo-Dong, Seocho-Gu, Seoul 137-701, South Korea.
Tel.: +82 2 590 1216; fax: +82 2 594 7355.
E-mail address: kimtg@catholic.ac.kr (T.-G. Kim).
more difficult, as na¨ ıve T cells have more stringent activation and
costimulatory requirements.
InadditiontopotentAPCs,CD4+Tcellsarecrucialforgenerating
CD8+ T memory cells [4,5,12] and the absence of the help of CD4+ T
cellsresultsinlethargy[6]ordefectivememory[5,7]ofCD8+Tcells.
ToactivatetheCD4+TcellrequiredforCTLresponse,tetanustoxoid
(TT) and Keyhole limpet hemocyanin (KLH) were used as adju-
vants [8–10]. Successful in vitro priming of human antiviral CTLs
was induced from PBMCs of seronegative healthy donors when TT-
induced CD4+ T cell help was provided during primary stimulation
[11]. The induction of CTL lines against the melanoma tumor anti-
gens Melan-A and MART-1 was reproducibly demonstrated with
the same approach [9]. To overcome the negative role played by
certain types of activated CD4+ T cells during CTL culture, polariza-
tion of PBL with Th1 conditioning before CTL generation was used.
This procedure induced better and longer lived CTL response and
also prevented the expansion of CD4+ T cells that have down reg-
ulatory activity [13]. Depletion of CD4+CD25+ Treg cells, a major
barrier to the development of effective anti-tumor immunity, has
promoted long-lasting tumor antigen-specific CD8+ T cells [14].
Survivin, a member of the inhibitor of apoptosis proteins fam-
ily, has the capacity to suppress apoptotic cell death and regulate
cell division, and it is present during normal fetal development
0264-410X/$ – see front matter © 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.vaccine.2008.05.036

Page 2

3988
E.-K. Kim et al. / Vaccine 26 (2008) 3987–3997
but undetectable in terminally differentiated adult tissues [21].
Survivin is aberrantly expressed in most human cancers of an
epithelial, hematopoietic origin and brain tumors. In agreement
withitsfunction,over-expressionofsurvivincorrelateswithtumor
progression and constitutes a poor prognostic factor for a variety
of cancers [17–19]. It has been reported that some cancer patients
exhibit both a humoral and cellular immune response to survivin,
and several peptide epitopes from survivin that stimulate major
histocompatibility complex (MHC) class I-restricted cytotoxic T
lymphocytes (CTLs) have been identified in patients with breast
cancer, leukemia, colorectal cancer and melanoma [30–33]. Fur-
thermore, survivin-specific CTL responses can be induced in vitro
withDCspulsedwithpeptideandtransducedwithadenovirusvec-
tor and with RNA [34–37].
In the present study, we investigated the quantitative and qual-
itative effects of CD4+ T cells on induction of survivin-specific
cytotoxic T cells in vitro from healthy donors using dendritic cells
transduced with adenovirus vector expressing human survivin
(Adv-survivin). Our results emphasize that both the appropriate
ratio of CD4+ T cells and polarization to Th1 are required for the
efficient generation of CTLs in vitro.
2. Materials and methods
2.1. Isolation of cells by MACS system
Human peripheral blood was obtained from HLA-A2 six
healthy donors and mononuclear cells were isolated by using
Ficoll–Hypaque (Amersham Pharmacia Biotech, Piscataway, NJ,
USA) density gradient centrifugation. Human leukocyte antigen
(HLA)-A subtypes of the donors were determined with sequence-
basedtypinginHLAlaboratory.Consentformsandapprovalforthis
study were obtained from the donors and the institutional review
board of The Catholic University of Korea, College of Medicine.
Following density separation, CD14+ monocytes, CD4+ T and
CD8+ T cells were isolated by MACS System (Miltenyi Biotec,
Bergisch Gladbach, Germany) with anti-CD14, anti-CD4 and anti-
CD8 antibody coupled to magnetic microbeads (Miltenyi Biotec),
respectively, according to the manufacturer’s instructions. The
purity of isolated cells was over 95% by flow cytometric analysis.
To obtain CD4+CD25+ Treg-depleted CD4+ (CD25–CD4+) T cells,
CD25+ T cells were further depleted from previously isolated CD4+
T cells with anti-CD25 antibody coupled to magnetic microbeads
(Miltenyi Biotec).
2.2. Production of recombinant adenovirus encoding survivin
(Adv-survivin)
The human survivin expression vectors, based on pShut-
tle vector (Clontech, Palo Alto, CA, USA), were constructed by
a RT-PCR using the sense primer 5?-CGAACTCGAGATGGGTGCC-
CCGACGTTG-3?, and the antisense primer 5?-GCAAGGATCCTCA-
ATCCATGGCAGCCAG-3?. Subsequently, the survivin expression cas-
sette was inserted into pAdeno-XTMviral vector (Clontech, Palo
Alto,CA,USA).ThepAdenoviralvectorconstructwasdigestedwith
PacI, transfected in HEK 293 cells with Lipofectamine (Life Tech-
nologiesInc.,Rockville,MD,USA).Afterpackagingtherecombinant
adenoviral DNA, the adenovirus containing survivin (Adv-survivin)
DNA was amplified and purified from cell lysates by banding twice
in CsCl density gradients, as previously described [15]. Viral prod-
ucts were desalted and stored at −80◦C in phosphate-buffered
saline (PBS) containing 10% glycerol (v/v). The titer of viral stock
was determined using the tissue culture infectious dose (TCID50)
method.
2.3. Culture of dendritic cells (DCs) and antigen pulsing
Immature DCs were generated from CD14+ monocytes by cul-
turing in RPMI 1640 medium supplemented with 10% fetal bovine
serum, 100ng/mL GM-CSF (GM-CSF; Endogen, Woburn, MA, USA),
and 50ng/mL IL-4 (IL-4; Genzyme, Cambridge, MA, USA) every 3
daysfor6–7daysinahumidified37◦C,5%CO2incubator.Immature
DCs were harvested and then infected with Adv-survivin of MOI
500 and/or recombinant pp65 protein (Miltenyi Biotec) of 2?L/106
DCs for 3h at 37◦C. After antigen loading, DCs were matured with
100ng/mL TNF-? and 100ng/mL LPS for 24h.
2.4. Th1 polarization in vitro
ToobtaintheThelper1(Th1)cells,CD4+Tcellswerestimulated
with plate-bound anti-CD3 mAb (OKT3, 2?g/mL) and anti-CD28
mAb (2?g/mL) for 4 days under polarizing conditions. Th1 condi-
tionsincludedIL-12(5ng/mL)andanti-IL-4mAb(5?g/mL).After4
days, cells were re-stimulated with plate-bound anti-CD3 mAb for
24h then supernatants were collected and IFN-?, TNF-? and IL-4
concentrations were assessed.
2.5. Stimulation of survivin-specific human T-lymphocytes in
vitro
After antigen pulsing, matured DCs were ?-irradiated with
40Gy. This procedure totally prevented the outgrowth of DCs in
the control cultures. The CD14 negative fraction of the PBMCs was
used as source for generation CD8+ and CD4+ T cells. Several ratios
of CD4+:CD8+ T cells, purified CD8+, or CD8+ with CD25–CD4+ T
cells were co-cultivated with the DCs at final concentrations of
2.0×106/well and 1.0×105/well in 24-well plates, respectively, in
RPMI1640supplementedwith10%fetalbovineserum(Gibco-BRL),
2mM l-glutamine, and 1% penicillin–streptomycin (Cambrex). On
day 7, the cells were harvested for re-stimulation. The DCs were
plated at a density of 1.0×105cells/well and then the T cells were
added at a density of 1.0×106cells/well. 20U/mL IL-2 (Endogen)
was added from day 8 every 3 days. After three rounds of stim-
ulation, the T cells were evaluated for survivin-specific immune
responses.
2.6. ELISPOT assay
For detection of cells secreting interferon-? (IFN-?), ELISPOT
assayswereperformedwithBDElispotassaykit,andtheprocedure
was conducted according to the manufacturer’s instructions. Any-
Gen Co. Ltd. (Kwangju, South Korea) synthesized the HLA-A0201
restricted survivin-derived peptide, SVV-1 (ELTLGEFLKL) and pp65
antigen-derived peptide, PP65-1 (NLVPMVATV). Briefly, the T cells
were incubated with LCLs un-pulsed or pulsed with 10?g/mL of
SVV-1 or PP65-1 peptide in triplicates at a 10:1 ratio. The number
of spots corresponding to IFN-?-secreting cells was counted using
AID-ELISPOT-Reader (AID).
2.7.
51Cr-release assay
After three rounds of stimulation with each antigen-pulsed
DCs, specific lysis was determined by a standard51Cr-release assay
using peptide-pulsed autologous lymphoblastoid cell lines (LCLs)
as target cells. The survivin expressing glioma cell lines (T98G
and A172) were also used as target cells. The human glioblas-
toma T98G cell line (human leukocyte antigen [HLA]-A2) and A172
(HLA-A1, A3) were purchased from the American Type Culture
Collection (ATCC; Rockville, MD, USA). These cell lines were cul-
tured in DMEM (Invitrogen, Carlsbad, CA, USA) supplemented with

Page 3

E.-K. Kim et al. / Vaccine 26 (2008) 3987–3997
3989
Fig. 1. Ratios of CD4+:CD8+ T cells and proliferation of T cells in vitro. (A) Proliferative capacity of various responder T cells by Adv-survivin transduced DCs. The various
responder T cells containing purified CD8+ T cells, CD4+:CD8+ (2:1) T cells, CD25–CD4+:CD8+ (2:1) T cells, and CD4+:CD8+ (1:1) T cells from HLA-A2 donor-HB (see Fig. 1D)
were stimulated with DCs transduced with Adv-survivin for 3 weeks. One of three experiments with similar results is represented. (B) Proliferative capacity of responder T
cells by various ratio of CD4+:CD8+ T cells. The ratio of CD4+:CD8+ T cells of the responder T cells from three donors was provided with 2:1, 1:1, 1:2, 1:4, and 1:8, respectively,
then each of the responder T cells was stimulated with DCs transduced with Adv-survivin for 3 weeks. (C) The proportion of CD4+ and CD8 T+ cell frequencies by various
ratio of CD4+:CD8+ T cells in the total survivin-specific T cells using flow cytometric analysis. The data from three different donors was presented. Fig. 1D shows proliferative
capacity of survivin-specific T cells from HLA-A2 six healthy donors. The total viable cell numbers were determined by the Trypan blue dye exclusion method.

Page 4

3990
E.-K. Kim et al. / Vaccine 26 (2008) 3987–3997
Fig. 2. Induction of survivin-specific cytotoxic T lymphocytes (CTLs) using 1:2 ratio
of CD4+:CD8+ T cells in vitro. DCs were transduced with Adv-survivin, which were
used to stimulate autologous 1:2 ratio of CD4+:CD8+ T cells from donor-HJ (see
Fig. 1D). (A) IFN-? ELISPOT assay. Autologous LCLs un-pulsed (LCL/Mock) or pulsed
with a HLA-A0201-restiricted survivin-derived peptide (ELTLGEFLKL; LCL/SVV-1)
were used as antigen presenting cells for the ELISPOT assay. To determine HLA class
I restriction, cells were preincubated for 30min with W6/32, anti-HLA class I mAb.
(B) Cytotoxicity of T cells against autologous LCL (filled circles) or SVV-1 loaded LCL
(empty circles). The data from three independent T cells was statistically analyzed
via Student’s t-test.
10% heat-inactivated fetal bovine serum (FBS), 0.5mM sodium
pyruvate, 2mM l-glutamine, 0.1mM nonessential amino acid, and
antibiotics (Invitrogen). These glioma cell lines were pre-cultured
in the presence of 100U/mL of IFN-? (Endogen) for 2 days to
enhance the expression of MHC class I molecules. LCLs loaded with
the SVV-1 and PP65-1 or glioma cell lines were incubated with
100?Ci Na2[51Cr]O4/106cells for 1h at 37◦C, and then washed
four times and used as target cells (5.0×103cells/well) for effec-
tor T cells (E/T ratios ranging from 80:1 to 10:1) in 96-well round
bottom plates. After 4h of incubation at 37◦C, 150?L super-
natants from each well were harvested and the radioactivity was
measured on a gamma counter (Wallac Oy, Turku, Finland). Max-
imum release was assessed by the addition of 2% Triton X-100
(Sigma, Taufkirchen, Germany). Spontaneous release was deter-
mined in wells with labeled targets in the absence of effector
cells. Specific lysis was calculated by the formula: specific51Cr-
release=[(experimental counts−spontaneous counts)/(maximal
counts−spontaneous counts)×100%].
2.8. FACs analysis
Cells were harvested, washed and resuspended in PBS and
then incubated with various fluorochrome-conjugated antibodies
on ice for 30min in the dark. They were washed again and ana-
lyzed using the FACSCalibur apparatus (Becton-Dickinson, Franklin
Lakes, NJ, USA) and BD CellQuest software. The following mAbs
were used to determine the phenotype of the CTLs: FITC-labeled
anti-CD4 and anti-CD62L, PE-labeled anti-CD8 and anti-IFN-?, and
PE-cy5 labeled anti-CD45RA. All antibodies were purchased from
Becton–Dickinson. CD8+ T cells were analyzed for IFN-? produc-
tion by intracellular staining done directly after restimulation with
antigen-pulsed or unpulsed LCLs. Permeabilization of the cells was
done using a proprietary solution (Cytofix/CytopermTM, Becton
Dickinson), followed by staining according to the manufacturer’s
recommendations.
2.9. Cytometric bead array (CBA) multiplex assays
During stimulation of T cells with un-pulsed or antigen-pulsed
DCs, the supernatant were collected from samples at day 2 and
analyzed for cytokine contents using cytometric bead array (CBA).
Experiments were performed using the CBA Th1/Th2 kit (BD)
following the manufacturer’s instructions, and the standard was
followed as previously described [16].
2.10. Statistical analysis
Statistical analyses were performed using the software Graph
Pad Prism 5 (GraphPad Software, San Diego, CA, USA). Statistical
tests were applied to three independent experiments. Differences
among the values were determined by using both Student’s t-test
and two-way ANOVA. The significance was determined as p<0.05.
3. Results
3.1. Ratios of CD4+:CD8+ T cells and proliferation of T cells in vitro
We investigated the appropriate conditions of T cells used as
respondersforinducingantigen-specificTcellsinvitrousingPBMCs
from donor-HB (Fig. 1A). Because the ratio of CD4+:CD8+ T cells is
usually range of 2:1 and 1:1 in normal healthy persons, 2:1 and
1:1 ratios of CD4+:CD8+ T cells was cultured. To remove the nat-
ural regulatory T cells in peripheral blood, CD25+ cell depleted
CD4+(CD25–CD4+) T cells was used at 2:1 ratio similar with that
of peripheral blood. Several groups of responder T cells contain-
ing purified CD8+ T cells, 2:1 and 1:1 ratio of CD4+:CD8+ T cells,
and 2:1 ratio of CD25–CD4+:CD8+ T cells were stimulated with DCs
transduced with Adv-survivin for 3 weeks. As shown in Fig. 1A, the
proliferation of 1:1 ratio of CD4+:CD8+ T cells was at least three-
fold to fourfold higher than purified CD8+ T cells at day 21, whereas
T cells did not proliferated in 2:1 ratio of CD4+:CD8+ T cells even
though the use of CD25–CD4+ T cells.
To further determine the proper ratio of CD4+:CD8+ T cells,
responder T cells of various ratios such as 2:1, 1:1, 1:2, 1:4, and
1:8 were tested using PBMCs from more than three donors. The
proliferation was constantly increased in both 1:1 and 1:2 ratio of
CD4+:CD8+ T cells than the other groups in experiments of three
donors (Fig. 1B). Because the proportion of CD8+ T cells at day 21
was usually higher in 1:2 ratio of CD4+:CD8+ T cells than that in
1:1 ratio of CD4+:CD8+ T cells (Fig. 1C), we decided to use 1:2 ratio
of CD4+:CD8+ T cells for further experiments. These results suggest
thattheproperratioofCD4+:CD8+Tcellsismoreimportantforthe
generation of tumor antigen-specific T cells in vitro rather than the
depletion of CD4+CD25+ Treg cells.
To investigate whether this 1:2 ratio of CD4+:CD8+ T cells was
generally effective for stimulation of T cells in vitro in normal
healthy donors, six healthy donors were stimulated using DCs
transduced with Adv-survivin. As shown in Fig. 1D, T cells from
threedonorswasstronglyproliferated(morethansixfold),andtwo
donors showed intermediate proliferation (twofold to threefold).
Onedonorcouldnotmaintaintheproliferationduringculture.This

Page 5

E.-K. Kim et al. / Vaccine 26 (2008) 3987–3997
3991
Fig. 3. Activation of CD4+ T cells by DCs pulsed with the pp65 protein. (A) IFN-? ELISPOT assay. CD4+ and CD8+ T cells from HLA-A2 donor-KS (see Fig. 1D) were incubated
with DCs pulsed with either the pp65 protein of 2?L/106DCs or PP65-1 peptide of 2?g/mL for 24h. The data from three independent T cells was statistically analyzed for
results. The data was statistically analyzed via Student’s t-test; *, p<0.05. (B) Secretion of cytokines from CD4+ T cells stimulated with the pp65 protein. Quantification of
cytokines was performed with the CBA (cytometric bead array) assay from 48h culture supernatants of CD4+ T cells cocultured with unloaded or the pp65 protein loaded
DCs. The data from three independent samples was statistically analyzed via Student’s t-test. IFN-?; p=0.0053, TNF-?; p=0.009, IL-4; p=0.0075.
result showed that the potency of proliferation might be variable
according to persons and their conditions.
3.2. Characteristics of survivin-specific cytotoxic T lymphocytes
(CTLs)
Next, to investigate the immune responses of CTLs induced
from donor-HJ, IFN-? ELISPOT and chromium release assay were
performed. IFN-? secreting cells specific for SVV-1 peptide were
detected in high frequencies (1466±152 IFN-? secreting cells/105)
as shown in Fig. 2A. IFN-? secreting cells specific for SVV-1 were
remarkablydecreasedbypre-incubationwithanti-HLAclassImAb
(W6/32), indicating presentation by MHC Class I restriction (*,
p=0.0015, Fig. 2A). Survivin-specific T cells also showed specific
cytotoxicity against autologous LCLs loaded with SVV-1 (LCL/SVV-
1) (Fig. 2B).
3.3. The recombinant pp65 protein as an adjuvant for Th1
environment
To examine whether recombinant pp65 protein is able to acti-
vate CD4+ T cells, CD4+ and CD8+ T cells were stimulated with DCs
pulsed with the pp65 protein or with PP65-1 peptide. As shown in
Fig. 3A, we observed higher frequencies of IFN-? producing CD4+
T cells reactive to the pp65 protein (*, p<0.05). In contrast, CD8+
T cells only responded to PP65-1 loaded DCs (Fig. 3A). To eval-
uated that whether CD4+ T cells activated by the pp65 protein
pulsed DCs are able to secrete either T helper 1 (Th1) or T helper 2
(Th2) cytokines, the level of Th1 (IFN-?, TNF-?) and Th2 (IL-4) type
cytokines was measured from the culture supernatants of CD4+
T cells. As shown in Fig. 3B, CD4+ T cells stimulated with the pp65
protein-pulsedDCssecretedahigherlevelofIFN-?(p=0.0053)and
TNF-? (p=0.009), whereas the level of IL-4 detected was less than
un-pulsedDCs(p=0.0075).Theseresultssuggestthatthepp65pro-
teinnotonlyactivatesCD4+TcellsbutalsomakesTh1environment
in vitro such as an adjuvant.
3.4. Stimulation by DCs pulsed with Adv-survivin and
recombinant pp65 protein
To further enhance CTL generation from the donor-KS showed
intermediate proliferation (Fig. 1D), DCs pulsed with Adv-survivin
and HCMV pp65 protein (SVV+pp65) were cocultured with 1:2
ratio of CD4+:CD8+ T cells from donor-KS for 3 weeks. As shown