1. Depleting Intratumoral CD4+
CD25+
Regulatory T Cells via
FasL Protein Transfer Enhances the Therapeutic
Efficacy of Adoptive T Cell Transfer
Aoshuang Chen,
1
Shanrong Liu,
1,2
David Park,
1
Youmin Kang,
1,3
and Guoxing Zheng
1
1
Department of Biomedical Sciences, College of Medicine at Rockford, University of Illinois, Rockford, Illinois; 2
Department of Histology
and Embryology, Second Military Medical University, Shanghai, China; and 3
State Key Lab for Agro-Biotechnology,
China Agricultural University, Beijing, China
Abstract
One strategy for improving adoptive therapy is precondition-
ing the host immune environment by depleting CD4+
CD25+
regulatory T cells (Treg) suppressive to antitumor responses.
Given that Treg increase, or selectively accumulate, within
tumors and are sensitive to FasL-mediated apoptosis, we test
here the hypothesis that inducing apoptosis of intratumoral
Treg using FasL may improve adoptive T cell therapy. We
show that FasL applied intratumorally via protein transfer
decreases intratumoral Treg via inducing apoptosis in these
cells. Significantly, we show that the use of FasL prior to the
infusion of tumor-reactive CD8+
T cells enhances the thera-
peutic efficacy of adoptive T cell transfer against established
tumors, which is mediated by persistent, systemic antitumor
immunity. Intratumoral FasL protein transfer also results in
neutrophil infiltration of tumor. However, we show that intra-
tumoral immunodepletion of neutrophils does not abolish the
effect of FasL on adoptive transfer. Rather, the effect of FasL is
completely abolished by cotransfer of Treg, isolated from the
tumor-draining lymph nodes. Hence, our study shows for the
first time that using FasL to predeplete intratumoral Treg
provides a useful means for optimizing adoptive therapy.
[Cancer Res 2007;67(3):1291–8]
Introduction
The limited efficacy of adoptive T cell therapy in clinical trials
highlights the need to continually improve this modality. One
strategy is to enhance the antitumor efficacy of therapeutic T cells
by enhancing their persistence, specificity, tumor-homing ability,
resistance to tolerance or immunosuppression, and/or memory
function (1).
Another strategy is to generate a host immune environment that
is conducive to the function of adoptively transferred therapeutic
T cells, using strategies such as depleting host lymphocytes (2–4).
Although how lymphodepletion enhances adoptive therapy
remains partially understood (5, 6), depletion of suppressor T cells
seems to be a major contributing factor (7). In the early 1980s,
North showed that depletion of a tumor-induced population of
suppressor T cells by cyclophosphamide could enhance the
antitumor efficacy of adoptive therapy (8). Later studies showed
that depletion of CD4+
CD25+
regulatory T cells (Treg), a recently
defined T cell subset playing a critical role in governing self-
tolerance (9–11) and inhibiting immune responses against ‘‘self-
like’’ cancer cells (12–14), could greatly improve adoptive therapy
(7, 15).
Thus far, depletion of Treg has been achieved mainly by systemic
use of chemotherapeutic agents such as cyclophosphamide (16, 17)
or anti-CD25 monoclonal antibodies (mAb). Nonetheless, such
strategies have limitations, including toxicity, induction of auto-
immunity due to the systemic elimination of Treg (18), and/or loss
of CD25-expressing effector T cells. On the other hand, previous
studies have shown that Treg increase or, even selectively
accumulate, within tumors (12, 19, 20), likely via the function of
the chemokine CCL22 produced by tumor cells and tumor-
infiltrating macrophages (12). Collectively, these studies emphasize
the need to specifically target intratumoral Treg. One strategy
would be to induce the apoptosis of intratumoral Treg locally at the
tumor site. Interestingly, it has recently been shown in vitro that
compared with CD4+
CD25À
effector T cells (Teff), Treg are highly
susceptible to FasL-mediated apoptosis (21).
Based on these findings, in the present study, we aim to test
the hypothesis that depleting intratumoral Treg using FasL, prior
to adoptive T cell transfer, may improve the antitumor efficacy
of adoptive therapy. Our results show for the first time that
predepleting intratumoral Treg using FasL provides an alternative
means for optimizing adoptive therapy.
Materials and Methods
Mice and tumor cell lines. BALB/c and DBA/2 mice (6–8 weeks of age)
were purchased from the Jackson Laboratory (Bar Harbor, ME) and used in
accordance with the institutional guidelines for animal care. EG.7 and
L5178Y tumor cell lines were purchased from American Type Culture
Collection (Manassas, VA) and maintained according to the recommenda-
tions of the supplier.
Antibodies. Anti-mouse CD4-FITC (GK1.5), anti-mouse CD25-PE (7D4),
anti-mouse CD25-PE-Cy5 (PC61.5), anti-mouse Foxp3-PE (JFK-16S), anti-
mouse Gr-1-PE (RB6-8C5), anti-human FasL mAb (NOK-1), anti-mouse Gr-1
(RB6-8C5), anti-mouse CD16/32 mAb (clone 93), various isotype controls,
and Annexin V-APC were purchased from BD Biosciences (San Diego, CA).
Isolation of T cell subsets. T cell subsets were purified by magnetic cell
sorting using various kits (Miltenyi Biotec, Auburn, CA). L5178Y tumor-
reactive CD8+
T cells were purified from bulk splenocytes using a CD8+
T cell kit. CD4+
CD25+
T cells were purified from lymph nodes draining
L5178Y tumors using a CD4+
T cell kit first, and subsequently, a CD4+
CD25+
cell kit. T cell purity of >98% was usually achieved, as determined by flow
cytometry.
Determination of FasL-induced apoptosis in vitro. Bulk spleno-
cytes from BALB/c or DBA/2 mice were depleted of RBC and incubated
(5 Â 105
/mL) for 20 h in 24-well plates with human FasL Fc fusion pro-
tein (FasL-Fc; ref. 22), in the absence or the presence of 10 Ag/mL of
Note: A. Chen and S. Liu contributed equally to this work.
Requests for reprints: Aoshuang Chen or Guoxing Zheng, Department of
Biomedical Sciences, College of Medicine at Rockford, University of Illinois, 1601
Parkview Avenue, Rockford, IL 61107. Phone: 815-395-5680; Fax: 815-395-5666; E-mail:
aoshuang@uic.edu or guoxingz@uic.edu.
I2007 American Association for Cancer Research.
doi:10.1158/0008-5472.CAN-06-2622
www.aacrjournals.org 1291 Cancer Res 2007; 67: (3). February 1, 2007
Research Article
Research.
on May 26, 2016. © 2007 American Association for Cancercancerres.aacrjournals.orgDownloaded from

2. FasL-neutralizing NOK-1 mAb or control mAb. Cells were stained with anti–
CD4-FITC, anti–CD25-PE, and 7-AAD, and analyzed for dead cells (7-AAD+
)
by flow cytometry on a FACSCalibur (BD Biosciences). CD4+
CD25À
and
CD4+
CD25+
T cell populations were gated, respectively. Data were analyzed
using the CellQuestPro software (BD Biosciences).
Depletion of intratumoral Treg cells in vivo via FasL protein
transfer. Procedures for generating palmitated protein A (PPA; ref. 23) and
using it for Fc protein transfer (24, 25) have been described previously.
Briefly, a conjugate of PPA and FasL-Fc was first generated by mixing the
components at a 1:1 ratio (w/w) in DMEM and incubating the mixture on
ice for 30 min. The mixture was injected intratumorally into mice bearing
an established intradermal L5178Y tumor. To assess the depletion of
intratumoral Treg, the next day, single cell suspensions were prepared from
tumors and stained with anti–CD4-FITC and anti–CD25-PE, and analyzed
by flow cytometry, gated on the CD4+
population.
Adoptive transfer therapy. To obtain L5178Y tumor-reactive T cells,
DBA/2 mice were immunized twice (at 1-week intervals) with 1 Â 106
mitomycin C–treated L5178Y tumor cells (s.c. injection) and, subsequently,
challenged with 1 Â 106
live L5178Y tumor cells (s.c. injection). Survivors
were used as donors for tumor-reactive T cells. One week prior to adoptive
therapy, the donors were rechallenged with live L5178Y tumor cells as
above. On the day when adoptive therapy was done, the donors were
sacrificed, and CD8+
T cells were purified from the spleen.
For adoptive therapy, tumors were established by intradermally injecting
5 Â 105
L5178Y tumor cells on the right flank of DBA/2 mice (day 1).
Palpable tumors (f5 mm in diameter) usually formed at the injection site
after 4 to 5 days. Occasionally, tumors grew into the s.c. area as a result of
bad injection; those mice were eliminated before experiments. On days 5
and 6, mice were each injected intratumorally with PPA (2 Ag) or PPA:FasL-
Fc conjugate (4 Ag total protein). On day 7, PPA-treated mice were each
injected with 4 Â 105
of L5178Y tumor-reactive CD8+
T cells (designated
as ‘‘CD8+
T cells’’). In parallel, PPA:FasL-Fc conjugate–treated mice either
received no further treatment (‘‘FasL’’) or were adoptively transferred
with the CD8+
T cells, either alone (‘‘FasL + CD8+
T cells’’) or together with
2 Â 105
Treg (‘‘FasL + CD8+
T cells + Treg’’). Treg were isolated from lymph
nodes draining progressively growing L5178Y tumors in a separate group of
tumor-bearing mice. In some experiments, FasL protein transfer and
depletion of tumor-infiltrating neutrophils were done concurrently. The
neutrophil-specific anti–Gr-1 mAb was injected intratumorally (15 Ag per
tumor) twice a day (at 12-h intervals) during the course of intratumoral
FasL protein transfer.
Tumor size was measured twice a week. Mice were euthanized when they
became moribund or when the tumors exceeded 400 mm2
in size. Cured
mice were each rechallenged (i.p.) with 5 Â 105
L5178Y tumor cells at least
8 weeks after the initial tumor inoculation.
Cytotoxic T lymphocyte assay. Bulk splenocytes from cured mice,
prepared 4 to 6 weeks after tumor rechallenge, were restimulated with
mitomycin C–treated tumor cells at a 10:1 ratio for 5 days. Viable cells (used
as effectors) were cultured with CFSE-labeled, mitomycin C–treated tumor
target cells. After 12 h, the cultures were stained with 7-AAD and analyzed
for dead target cells (CFSE+
/7-AAD+
) by flow cytometry. Specific lysis was
calculated as described previously (26).
Detection of cytokines. Bulk splenocytes from cured mice were
prepared and restimulated with L5178Y tumor cells as described above,
and the conditioned media were analyzed using a flow cytometry-based
bead array kit for Th1/Th2 cytokines (BD Biosciences), following the
manufacturer’s instructions.
Results
Treg are more sensitive than their CD4+
CD25À
counterpart
to FasL-induced apoptosis in vitro. First, under our experimental
settings, we compared the susceptibility to FasL-mediated
apoptosis in vitro of murine Treg versus their CD4+
CD25À
counterparts (naı¨ve effector T cells, Teff). Purified Treg expressed
high levels of Foxp3 and were suppressive to T cell proliferation
(data not shown). To ensure that the Treg and Teff subsets were
compared under the same conditions, bulk splenocytes from
BALB/c mice were directly incubated with recombinant FasL,
the FasL Fc fusion protein (FasL-Fc; ref. 22); apoptosis in each
subset of T cells was then determined by staining the cells with
anti-CD4, anti-CD25, and 7-AAD and analyzing by flow cytometry,
gated on CD4+
CD25À
and CD4+
CD25+
populations, respectively.
FasL effectively elicited apoptosis in Treg, but had a minimal effect
on Teff (Fig. 1A). The specificity of FasL was verified by the
blockade of apoptosis by FasL-neutralizing NOK-1 mAb, but not
by control mAb. FasL titration experiments verified that Treg
are more sensitive than Teff to FasL (Fig. 1B). A similar sus-
ceptibility to FasL was confirmed in splenocytes from another
strain of mice, the DBA/2 (Fig. 1C). Together, these results show
that Treg are more sensitive than Teff to FasL-mediated apoptosis
in vitro.
Intratumoral FasL protein transfer leads to depletion of
Treg in vivo. As a prelude to applying FasL to adoptive therapy, we
assessed the capacity of FasL to induce apoptosis in Treg in vivo in
the L5178Y lymphoma model. Consistent with prior findings by
others (27), we showed that L5178Y tumor cells in our laboratory
were resistant to FasL-induced apoptosis (Fig. 2A). Hence, the use
of the L5178Y tumor model excludes the possibility of direct killing
of the tumor cells by FasL.
To clearly define the lineage of the CD4+
CD25+
cell population
within established L5178Y tumors, we assessed the intracellular
expression of Foxp3, the Treg lineage marker, on these cells. As
shown in Fig. 2B, a majority (>85%) of the CD4+
CD25+
T cells
within tumors coexpress Foxp3, confirming that these T cells are,
indeed, Treg.
We next wanted to assess the susceptibility of these intratumoral
Treg to FasL. To minimize the toxicity associated with the use of
recombinant FasL (28, 29), we confined FasL-Fc within tumors
using the protein transfer (or protein ‘‘painting’’) method we
previously developed (24). In this method, protein A, after being
chemically derivatized with palmitate in a simple reaction, is first
incorporated into cell membranes; in turn, this membrane-
anchored PPA serves as a ‘‘trap’’ for secondarily added Fc fusion
proteins. Moreover, in the L5178Y tumor model, we showed that Fc
fusion proteins, after being preconjugated with PPA in vitro, can
be directly ‘‘painted’’ and localized for sufficiently long times (up to
18 h) in the tumor bed, thereby efficiently generating cancer
vaccines in situ (25).
Therefore, FasL-Fc preconjugated with PPA (PPA:FasL-Fc) was
injected (4 Ag total protein per tumor) intratumorally into L5178Y
tumors. Under these experimental conditions, treated mice seemed
healthy, and showed no obvious liver damage as assessed by
autopsy (data not shown). The next day, single cell suspensions
were prepared from the injected tumors and assessed for the
presence of Treg undergoing apoptosis. As shown in Fig. 2C,
compared with nontreated control tumors, FasL-treated tumors
contained significantly more abundant Treg undergoing apoptosis,
assessed by their Annexin V+
status. The result indicates that
FasL protein transfer could result in the apoptosis of intratumoral
Treg in vivo. Of note, we observed that dead Treg or Treg
undergoing apoptosis were relatively fragile and readily lysed
during the processing of single tumor cell suspensions (data not
shown); thus, quantification of these Treg can cause underestima-
tion of Treg depletion. Consequently, we chose to assess the extent
of Treg depletion via quantifying live (7-AADÀ
) Treg and Teff
populations.
Cancer Research
Cancer Res 2007; 67: (3). February 1, 2007 1292 www.aacrjournals.org
Research.
on May 26, 2016. © 2007 American Association for Cancercancerres.aacrjournals.orgDownloaded from

3. As shown in Fig. 2D, compared with nontreated or PPA-treated
tumors, FasL-treated tumors showed a decrease in the number of
live Treg and, thus, a decrease in the ratio of Treg to Teff.
Collectively, these results indicate that intratumoral FasL protein
transfer could lead to a decrease in intratumoral Treg by inducing
apoptosis in these cells. In addition, we also determined whether
increasing the quantity of PPA:FasL-Fc conjugate can improve the
extent of Treg depletion. As shown in Fig. 2E, the dose-dependent
response plateaus between 3 and 9 Ag per tumor. Hence, the
PPA:FasL-Fc conjugate was used at 4 Ag per tumor for the entire
study.
Next, we determined whether intratumoral FasL treatment can
lead to depletion of Treg systemically. At different time points after
FasL treatment, cell suspensions prepared from the tumor, tumor-
draining lymph nodes, spleen, and peripheral blood were each
quantified for live Treg. Again, similar to the results in Fig. 2, the
FasL treatment caused a significant decrease in intratumoral Treg,
with the maximal decrease (>80%) achieved at 12 h after FasL
Figure 2. FasL protein transfer leads to a decrease in intratumoral Treg in vivo. A, L5178Y tumor cells were incubated for 20 h with the indicated concentration of
FasL-Fc, and the resistance of the tumor cells to apoptosis was determined by 7-AAD staining and flow cytometry. B, single cell suspensions were prepared from
intradermal L5178Y tumors established in DBA/2 mice. Cells were first stained with anti–CD4-FITC and anti–CD25-PE-Cy5, and then stained intracellularly with
anti–Foxp3-PE. Cells were analyzed by flow cytometry, gated on CD4+
T cells. C, intradermal L5178Y tumors were established in DBA/2 mice (day 1). On day 5, mice
were nontreated or treated intratumorally with PPA:FasL-Fc conjugate (4 Ag total protein per tumor). The next day, single cell suspensions were prepared from the
tumors separately. Apoptosis of Treg was determined by staining the cells with anti–CD4-FITC, anti–CD25-PE, Annexin V-APC, and 7-AAD and analyzing by flow
cytometry, gated on CD4+
CD25+
T cells. The Annexin V+
fraction of the gated population is shown. D, intradermal L5178Y tumors were established (on day 1) and
treated (on day 5) as described in (C). The next day, single cell suspensions from tumors were quantified for CD4+
CD25+
and CD4+
CD25À
T cells by staining
with anti–CD4-FITC and anti–CD25-PE and flow cytometry, gated on CD4+
T cells. Columns, mean percentage of CD4+
CD25+
Treg (in total CD4+
population)
determined based on three independent experiments; bars, SD. E, intradermal L5178Y tumors were treated with the indicated amounts of PPA:FasL-Fc, and single cell
suspensions from the tumors were quantified for Treg, as described in (D). Columns, mean from two independent experiments; bars, SD.
Figure 1. Splenic Treg are highly
susceptible to FasL-mediated apoptosis
in vitro compared with their CD25À
counterpart (Teff). A, splenocytes from
BALB/c mice were depleted of RBC and
incubated for 20 h with 250 ng/mL of
FasL-Fc, alone or together with 10 Ag/mL
of FasL-neutralizing NOK-1 mAb or
control mAb. Cells were stained with
anti–CD4-FITC, anti–CD25-PE,
and 7-AAD, and analyzed by flow
cytometry, gated on CD4+
CD25À
and
CD4+
CD25+
populations, respectively.
B, dose-response curve of FasL-induced
apoptosis of CD4+
CD25+
Treg (.) and
CD4+
CD25À
Teff (o) populations.
C, splenocytes from DBA/2 mice were
similarly analyzed for FasL-induced
apoptosis as described in (A). Columns,
mean of three to four independent
experiments; bars, SD.
FasL Improves Adoptive T Cell Transfer
www.aacrjournals.org 1293 Cancer Res 2007; 67: (3). February 1, 2007
Research.
on May 26, 2016. © 2007 American Association for Cancercancerres.aacrjournals.orgDownloaded from

4. treatment; such a decrease of Treg was sustained even after
24 h (Fig. 3A). In addition, the FasL transfer also resulted in a
decrease (although to a significantly lesser extent) in Treg in the
tumor-draining lymph nodes, with the maximal decrease achieved
after 12 h (Fig. 3B). In comparison, at the same time points
evaluated, the same FasL treatment did not lead to a decrease in
Treg in the spleen and peripheral blood (Fig. 3C and D).
Collectively, these data indicate that intratumoral FasL protein
transfer leads to a decrease in Treg locally, but does not decrease
Treg systemically.
Predepleting intratumoral Treg using FasL enhances the
efficacy of adoptive T cell transfer. Having documented the
capacity of FasL applied intratumorally to induce apoptosis in
local Treg in vivo, we next determined whether predepleting intra-
tumoral Treg using FasL can improve the efficacy of adoptive T cell
therapy. First, in the L5178Y tumor model, we evaluated whether
therapy combining FasL with adoptive T cell transfer could elicit
more effective antitumor responses compared with adoptive therapy
alone. Then, we determined whether the effect of FasL on adoptive
therapy was attributable to the depletion of intratumoral Treg.
To that end, mice bearing established L5178Y tumors were
injected intratumorally with PPA alone or with PPA:FasL-Fc
conjugate. Subsequently, mice were adoptively transferred with
L5178Y tumor-reactive CD8+
T cells, alone or together with Treg.
Of note, Treg used in these therapy experiments were isolated
from lymph nodes draining progressively growing L5178Y tumors
in a separate group of tumor-bearing mice; these Treg expressed
Foxp3 and efficiently suppressed T cell proliferation in vitro (data
not shown).
As negative controls, nontreated mice all showed progressive
tumor growth and died, whereas a portion of mice treated with
either FasL alone or adoptive therapy alone showed delayed tumor
growth (Fig. 4A). On average, FasL treatment alone resulted in
complete tumor regression in 12% of treated mice compared with
9% in adoptive therapy alone (Table 1). In comparison, when mice
were treated with both FasL and adoptive T cell therapy, tumor
growth was significantly retarded in a substantial portion of
treated mice (Fig. 4A), and on average, complete tumor regression
was observed in 53% of treated mice (Table 1). These results
indicate that therapy combining FasL with adoptive T cell transfer
has stronger antitumor efficacy than that with either FasL or
adoptive transfer alone.
Significantly, post-FasL treatment, when Treg were cotransferred
with tumor-reactive CD8+
T cells, tumors grew aggressively, even to
a slightly greater extent than those in nontreated mice (Fig. 4A); on
average, tumor regression was observed in 7% of treated mice,
which is comparable with that (9%) in mice treated with adoptive
therapy alone (Table 1). Hence, the addition of Treg during
adoptive T cell transfer completely abolished the effect of FasL on
adoptive therapy, which points to the fact that FasL enhances the
antitumor efficacy of adoptive therapy via, at least partially,
predepleting intratumoral Treg.
Establishment of persistent, systemic immunity in mice
treated with FasL and adoptive therapy. Having documented
the local tumor regression in a substantial percentage of mice
treated with both FasL and adoptive Tcell transfer, we next assessed
the establishment of long-term, systemic antitumor immunity in
treated mice. To this end, cured mice were rechallenged with
L5178Y tumor cells injected at sites (i.p.) distant from the original
tumor at least 2 months after the initial tumor inoculation. All of
these cured mice were resistant to the rechallenge, whereas naı¨ve
mice inoculated with the same tumor cells all died of tumors
(Fig. 4B). It is especially notable that all of the cured mice subjected
to i.p. tumor rechallenge >6 months after the initial tumor cell
inoculation were still resistant to tumor (data not shown). These
rechallenge experiments point to a persistent, systemic antitumor
immunity that is established in mice treated with therapy
combining FasL with adoptive T cell transfer.
To further support the establishment of persistent, systemic
immunity, we recovered bulk splenocytes from these cured mice,
3 to 6 weeks after tumor rechallenge, and checked for cytokine and
cytotoxic T lymphocyte (CTL) responses. Upon in vitro restimula-
tion with the tumor cells, the splenocytes produced high levels of
Th1 cytokines, including interleukin 2, IFN-g, and tumor necrosis
factor a (Fig. 4C). The splenocytes showed CTL activity against the
L5178Y tumor cells (Fig. 4D). The CTL activity was L5178Y tumor
cell–specific, as the lysis of irrelevant, control EG.7 tumor cells was
at a significantly lower level. Thus, specific CTL can be recovered
from a secondary lymphoid organ distal from the tumor cells at the
treatment site.
Together, these results show a persistent, systemic antitumor
immunity that is established in mice cured by therapy combining
intratumoral FasL protein transfer and adoptive T cell transfer.
Figure 3. FasL-mediated Treg depletion occurs predominantly in the tumor
bed and tumor-draining lymph nodes. Intradermal L5178Y tumors were
established in DBA/2 mice (day 1). On day 5, mice were injected intratumorally
with PPA:FasL-Fc (4 Ag total protein per tumor). At the indicated time points
postinjection, single cell suspensions were prepared from the tumor (A),
tumor-draining lymph nodes (B), spleen (C), and peripheral blood (D). Cell
suspensions were each quantified for Treg, as described in Fig. 2. Points,
average results from two independent experiments; bars, SD.
Cancer Research
Cancer Res 2007; 67: (3). February 1, 2007 1294 www.aacrjournals.org
Research.
on May 26, 2016. © 2007 American Association for Cancercancerres.aacrjournals.orgDownloaded from

5. Depletion of tumor-infiltrating neutrophils does not dimin-
ish the effect of FasL on adoptive transfer. Although the
experiments above establish the contribution of depleting local
Treg to the antitumor efficacy, they do not exclude the possibility
that FasL may enhance the antitumor efficacy via other mechanisms.
Especially, FasL has been shown to mediate neutrophil recruit-
ment, which, under certain circumstances, can elicit or enhance
antitumor responses (27, 30, 31). Hence, we wanted to determine the
contribution of FasL-mediated neutrophil recruitment to the
increase in the antitumor efficacy. Consistent with the findings
from others (27, 30, 31), intratumoral FasL protein transfer resulted
in a significant increase of tumor-infiltrating neutrophils (Gr-1+
population); such an increase was effectively reversed by using
the neutrophil-specific anti–Gr-1 mAb, coinjected intratumorally
with PPA:FasL-Fc (Fig. 5A). FasL, however, did not significantly
change the levels of neutrophils in the blood, assessed at various
time points postinjection (Fig. 5B). These results indicate that
intratumoral FasL protein transfer, whereas increasing tumor-
infiltrating neutrophils, does not affect the quantity of neutrophil
systemically.
Next, we determined whether FasL-mediated increase in tumor-
infiltrating neutrophils could contribute to the effect of FasL on
adoptive therapy. To this end, we depleted tumor-infiltrating
neutrophils during the course of FasL treatment. As shown in
Fig. 5C, a single injection of anti–Gr-1 mAb was sufficient to
efficiently block the increase in tumor-infiltrating neutrophils for
up to 24 h. Consequently, for adoptive therapy, we injected anti–
Gr-1 mAb twice a day (at 12-h intervals) during the entire course
of the FasL treatment, to ensure complete depletion of tumor-
infiltrating neutrophils. As shown in Fig. 5D, the depletion of
tumor-infiltrating neutrophils did not abolish the antitumor
efficacy of the therapy combining FasL with adoptive T cell
transfer. On average, the therapy done in the presence of tumor-
infiltrating neutrophils and that in the absence of neutrophils show
comparable antitumor efficacies, each resulting in complete tumor
regression in f50% treated mice (Table 1, experiments 6 and 7).
Figure 4. Therapy combining FasL
and adoptive T cell transfer shows
enhanced antitumor efficacy mediated by
persistent, systemic antitumor immunity.
A, intradermal L5178Y tumors were
established in DBA/2 mice (day 1). On
days 5 and 6, tumors were each injected
once with PPA (2 Ag) or PPA:FasL-Fc
conjugate (4 Ag total protein). On day 7,
4 Â 105
splenic, tumor-reactive CD8+
T cells (isolated from a separate group of
mice immunized with the L5178Y tumor
cells) were injected intratumorally, either
alone or together with 2 Â 105
Treg
(isolated from lymph nodes draining
progressively growing L5178Y tumors in a
separate group of tumor-bearing mice).
Subsequently, tumor size was measured
and graphed for individual animals; lines,
tumor growth curve of a single animal.
B, a lethal dose (0.5 Â 106
) of L5178Y
tumor cells was inoculated (i.p.) into naı¨ve
mice (- - - -) or mice cured by therapy
combining FasL with adoptive T cell
transfer (—), 2 to 3 mo after initial tumor
inoculation. Mice were monitored for the
appearance of tumor weekly. C, bulk
splenocytes were prepared 3 to 6 wks after
the rechallenge experiments described
in (B), and restimulated in vitro with
mitomycin C–treated L5178Y tumor cells
(at a 10:1 ratio). Subsequently, media from
the cultures were quantified for interleukin
2, IFN-g, and tumor necrosis factor a by
flow cytometry. D, splenocytes from the
rechallenged mice (effector), after the
in vitro restimulation described in (C),
were incubated for 12 h with the L5178Y
(., specific target) or EG.7 (5, nonspecific
target) tumor cells at indicated effector/
target (E:T) ratios. Subsequently, CTL
activity was determined. As a negative
control, splenocytes from naı¨ve mice were
assayed for CTL activity against the
L5178Y tumor cells (5). The results are
representative of at least two independent
experiments.
FasL Improves Adoptive T Cell Transfer
www.aacrjournals.org 1295 Cancer Res 2007; 67: (3). February 1, 2007
Research.
on May 26, 2016. © 2007 American Association for Cancercancerres.aacrjournals.orgDownloaded from

6. Collectively, these data show that the reversal of FasL-mediated
neutrophil increase does not affect the antitumor efficacy of
therapy combining FasL with adoptive T cell transfer. This finding,
together with the finding that the addition of Treg during adoptive
transfer completely abolishes the effect of FasL on the antitumor
efficacy (Fig. 4), solidifies our conclusion that intratumoral FasL
protein transfer enhances the therapeutic efficacy of adoptive
therapy via, primarily, depleting Treg locally.
Table 1. Treatment of L5178Y tumors
Experiment Fraction of mice cured per treatment group
None FasL CD8+
T cells FasL + CD8+
T cells FasL + CD8+
T cells + Treg FasL + CD8+
T cells + anti–Gr-1
1 0/4 0/4 4/4
2 0/5 1/4 1/5 2/4
3 0/5 0/5 0/5 2/5 0/5
4 0/5 1/5 1/5 2/5 1/5
5 0/4 0/4 0/4 1/4 0/4
6 0/5 2/5 2/5
7 0/5 4/5 3/5
Total 0/33 2/18 2/23 17/32 1/14 5/10
Percentage 0 12 9 53 7 50
Figure 5. Depletion of tumor-infiltrating neutrophils does not abolish the effect of FasL on the therapeutic efficacy of adoptive T cell transfer. A, intradermal
L5178Y tumors were established in DBA/2 mice (day 1). On day 5, mice were injected intratumorally with PPA, PPA:FasL-Fc alone, or PPA:FasL-Fc together with
anti–Gr-1 mAb. PPA was used at 2 Ag/tumor, PPA:FasL-Fc was used at 4 Ag per tumor, and anti–Gr-1 mAb was used at 15 Ag per tumor. After 12 h, single cell
suspensions from the tumors were analyzed for neutrophils by staining with anti–Gr-1-PE and flow cytometry. Columns, average percentages of neutrophil (in total
cells within the tumor from two independent experiments); bars, SD. B, intradermal L5178Y tumors were treated by intratumoral FasL protein transfer, as described in
(A). At indicated time points posttreatment, single cells from peripheral blood of treated mice were quantified for neutrophils, as described in (A). C, intradermal L5178Y
tumors were treated with either PPA:FasL-Fc alone (.) or PPA:FasL-Fc together with anti–Gr-1 mAb (o), as described in (A). At indicated time points posttreatment,
single cell suspensions from the tumors were quantified for neutrophils, as described in (A). Points, mean of results from two independent experiments (B-C). D,
intradermal L5178Y tumors were established (day 1), and mice were divided into three groups. On days 5 and 6, mice in group 1 (control) remained nontreated; mice in
group 2 were each treated with 4 Ag of PPA:FasL-Fc once; mice in group 3 were each treated with 4 Ag of PPA:FasL-Fc once and 15 Ag of anti–Gr-1 mAb twice
(at 12-h intervals). On day 7, mice in groups 2 and 3 were injected intratumorally with 4 Â 105
L5178Y tumor-reactive CD8+
T cells. Subsequently, tumor size was
measured and graphed, as described in Fig. 4A. Data are representative of two independent experiments.
Cancer Research
Cancer Res 2007; 67: (3). February 1, 2007 1296 www.aacrjournals.org
Research.
on May 26, 2016. © 2007 American Association for Cancercancerres.aacrjournals.orgDownloaded from

7. Discussion
Treg increase, or selectively accumulate, within tumors (12, 19,
20), likely via the function of the chemokine CCL22 produced by
tumor-infiltrating macrophages (12). Hence, it is conceivable that it
may be advantageous to target intratumoral Treg locally. Support-
ing this notion, the depletion of intratumoral Treg using anti-CD25
mAb has recently been shown to unmask tumor immunogenicity,
leading to the rejection of late-stage tumors (32). In the present
study, we show for the first time that depleting Treg locally at the
tumor site using FasL likewise enhances the therapeutic efficacy of
adoptive T cell therapy.
Other than depleting Treg, FasL may elicit or enhance antitumor
responses via other mechanisms. One such mechanism is FasL-
mediated recruitment of neutrophils. Thus, we determined whether
intratumoral FasL protein transfer could enhance the antitumor
efficacy of adoptive therapy via recruiting neutrophils. Our results
show that the FasL transfer could cause significant neutrophil
infiltration of tumors (Fig. 5A). Nonetheless, depletion of tumor-
infiltrating neutrophils does not significantly decrease the anti-
tumor efficacy of therapy combining FasL with adoptive transfer
(Fig. 5D). This result is not completely unexpected. Although it has
been shown in certain systems that FasL-mediated neutrophil
recruitment can result in antitumor responses (27, 30, 31),
contradicting observations also exist in the literature. The
expression of FasL on tumor cells, although capable of inducing
intensive neutrophil infiltration of tumor and inflammation, has
been shown to fail to cause tumor rejection (33, 34). This may be
attributed to a few factors. For example, transforming growth factor
h in the tumor microenvironment has been shown to inhibit
neutrophil activation, thereby preventing tumor rejection (30). The
antitumor effect of FasL has been shown to be affected by the form
(i.e., soluble versus membrane-bound) of FasL expressed on tumors
(33). The density of FasL is another contributing factor; tumor cells
expressing high levels of FasL have been shown to impair neutrophil
activation (34). Under our experimental conditions, FasL-induced
neutrophil infiltration has a minimal effect on the antitumor
efficacy of adoptive therapy. This finding, together with the findings
that the addition of Treg during adoptive therapy completely
abolishes the effect of FasL (Fig. 4A) and that FasL does not induce
apoptosis of the tumor cells directly (Fig. 2A), indicates that
intratumoral FasL protein transfer improves the antitumor efficacy
of adoptive therapy, primarily via depleting Treg locally.
As we have previously reported, intratumorally injected palmi-
tated proteins can be retained in the tumor for up to 18 h (25). The
loss of the proteins during or after the 18-h period is likely due to
cell metabolism, internalization, and/or degradation. In addition,
after protein transfer, we did not observe a significant increase in
soluble Fc fusion protein in the blood from treated mice (data not
shown). Therefore, protein transfer may provide a safe alternative
means for applying immunoregulatory molecules locally to a
tumor, particularly for those associated with relatively severe
toxicity, such as FasL.
In this study, we observed high susceptibility to FasL-mediated
apoptosis by both Treg cultured in vitro and intratumoral Treg
in vivo. Our results are consistent with the in vitro study by
Fritzsching and colleagues showing that in both humans and mice,
CD4+
CD25+
Treg cells are highly susceptible to FasL-induced
apoptosis (21). Papiernik and colleagues, however, observed
resistance of prestimulated CD4+
CD25+
Treg to apoptosis induced
by anti-Fas mAb (35). One explanation for this discrepancy may be
that we and Fritzsching and colleagues used FasL, which might
bind and multimerize the Fas receptor differently from the anti-Fas
mAb used by Papiernik and colleagues.
In summary, the strategy described here and those developed by
others, such as the use of anti-GITR mAbs, now provide a set of
tools for targeting host Treg, in order to optimize adoptive transfer
and other immunotherapeutic modalities for cancer.
Acknowledgments
Received 7/14/2006; revised 11/13/2006; accepted 11/30/2006.
Grant support: NIH grant RO1CA92243 (A. Chen) and a grant from the Illinois
Department of Public Health (G. Zheng).
The costs of publication of this article were defrayed in part by the payment of page
charges. This article must therefore be hereby marked advertisement in accordance
with 18 U.S.C. Section 1734 solely to indicate this fact.
We thank John Javaherian for providing care for the animals, and Emily Diamond
and Janet Stull-Snow for their administrative assistance.
FasL Improves Adoptive T Cell Transfer
www.aacrjournals.org 1297 Cancer Res 2007; 67: (3). February 1, 2007
References
1. Cohen PA, Peng L, Kjaergaard J, et al. T-cell adoptive
therapy of tumors: mechanisms of improved therapeutic
performance. Crit Rev Immunol 2001;21:215–48.
2. Dudley ME, Wunderlich JR, Robbins PF, et al. Cancer
regression and autoimmunity in patients after clonal
repopulation with antitumor lymphocytes. Science 2002;
298:850–4.
3. Mintern JD, Davey GM, Belz GT, Carbone FR, Heath
WR. Cutting edge: precursor frequency affects the helper
dependence of cytotoxic T cells. J Immunol 2002;168:
977–80.
4. Rosenberg SA, Dudley ME. Cancer regression in
patients with metastatic melanoma after the transfer
of autologous antitumor lymphocytes. Proc Natl Acad
Sci U S A 2004;101 Suppl 2:14639–45.
5. Klebanoff CA, Khong HT, Antony PA, Palmer DC,
Restifo NP. Sinks, suppressors and antigen presenters:
how lymphodepletion enhances T cell-mediated tumor
immunotherapy. Trends Immunol 2005;26:111–7.
6. Gattinoni L, Powell DJ, Jr., Rosenberg SA, Restifo NP.
Adoptive immunotherapy for cancer: building on
success. Nat Rev Immunol 2006;6:383–93.
7. Antony PA, Piccirillo CA, Akpinarli A, et al. CD8+ T cell
immunity against a tumor/self-antigen is augmented by
CD4+ T helper cells and hindered by naturally occurring
T regulatory cells. J Immunol 2005;174:2591–601.
8. North RJ. Cyclophosphamide-facilitated adoptive
immunotherapy of an established tumor depends on
elimination of tumor-induced suppressor T cells. J Exp
Med 1982;155:1063–74.
9. Sakaguchi S, Sakaguchi N, Shimizu J, et al. Immuno-
logic tolerance maintained by CD25+ CD4+ regulatory T
cells: their common role in controlling autoimmunity,
tumor immunity, and transplantation tolerance. Immu-
nol Rev 2001;182:18–32.
10. Shevach EM. CD4+ CD25+ suppressor T cells: more
questions than answers. Nat Rev Immunol 2002;2:389–400.
11. Curotto de Lafaille MA, Lafaille JJ. CD4(+) regulatory
T cells in autoimmunity and allergy. Curr Opin Immunol
2002;14:771–8.
12. Curiel TJ, Coukos G, Zou L, et al. Specific recruitment
of regulatory T cells in ovarian carcinoma fosters
immune privilege and predicts reduced survival. Nat
Med 2004;10:942–9.
13. Shevach EM. Fatal attraction: tumors beckon regu-
latory T cells. Nat Med 2004;10:900–1.
14. Wang HY, Lee DA, Peng G, et al. Tumor-specific
human CD4+ regulatory T cells and their ligands:
implications for immunotherapy. Immunity 2004;20:
107–18.
15. Shimizu J, Yamazaki S, Sakaguchi S. Induction of
tumor immunity by removing CD25+CD4+ T cells: a
common basis between tumor immunity and autoim-
munity. J Immunol 1999;163:5211–8.
16. Turk MJ, Guevara-Patino JA, Rizzuto GA, Engelhorn
ME, Sakaguchi S, Houghton AN. Concomitant tumor
immunity to a poorly immunogenic melanoma is
prevented by regulatory T cells. J Exp Med 2004;200:
771–82.
17. Ghiringhelli F, Larmonier N, Schmitt E, et al.
CD4+CD25+ regulatory T cells suppress tumor immuni-
ty but are sensitive to cyclophosphamide which allows
immunotherapy of established tumors to be curative.
Eur J Immunol 2004;34:336–44.
18. Wei WZ, Morris GP, Kong YC. Anti-tumor immunity
and autoimmunity: a balancing act of regulatory T cells.
Cancer Immunol Immunother 2004;53:73–8.
19. Woo EY, Chu CS, Goletz TJ, et al. Regulatory
CD4(+)CD25(+) T cells in tumors from patients with
early-stage non-small cell lung cancer and late-stage
ovarian cancer. Cancer Res 2001;61:4766–72.
20. Liyanage UK, Moore TT, Joo HG, et al. Prevalence of
regulatory T cells is increased in peripheral blood and
Research.
on May 26, 2016. © 2007 American Association for Cancercancerres.aacrjournals.orgDownloaded from

8. Cancer Research
Cancer Res 2007; 67: (3). February 1, 2007 1298 www.aacrjournals.org
tumor microenvironment of patients with pancreas or
breast adenocarcinoma. J Immunol 2002;169:2756–61.
21. Fritzsching B, Oberle N, Eberhardt N, et al. In
contrast to effector T cells, CD4+CD25+FoxP3+ regula-
tory T cells are highly susceptible to CD95 ligand- but
not to TCR-mediated cell death. J Immunol 2005;175:
32–6.
22. Chen A, Zheng G, Tykocinski ML. Quantitative
interplay between activating and pro-apoptotic signals
dictates T cell responses. Cell Immunol 2003;221:128–37.
23. Kim SA, Peacock JS. The use of palmitate-conjugated
protein A for coating cells with artificial receptors which
facilitate intercellular interactions. J Immunol Methods
1993;158:57–65.
24. Chen A, Zheng G, Tykocinski ML. Hierarchical
costimulator thresholds for distinct immune responses:
application of a novel two-step Fc fusion protein
transfer method. J Immunol 2000;164:705–11.
25. Zheng G, Chen A, Sterner RE, et al. Induction of
antitumor immunity via intratumoral tetra-costimulator
protein transfer. Cancer Res 2001;61:8127–34.
26. Zheng G, Liu S, Wang P, Xu Y, Chen A. Arming tumor-
reactive T cells with costimulator b7–1 enhances
therapeutic efficacy of the T cells. Cancer Res 2006;66:
6793–9.
27. Seino K, Kayagaki N, Okumura K, Yagita H.
Antitumor effect of locally produced CD95 ligand. Nat
Med 1997;3:165–70.
28. Tanaka M, Suda T, Yatomi T, Nakamura N, Nagata S.
Lethal effect of recombinant human Fas ligand in mice
pretreated with Propionibacterium acnes. J Immunol
1997;158:2303–9.
29. Timmer T, de Vries EG, de Jong S. Fas receptor-
mediated apoptosis: a clinical application? J Pathol
2002;196:125–34.
30. Chen JJ, Sun Y, Nabel GJ. Regulation of the
proinflammatory effects of Fas ligand (CD95L). Science
1998;282:1714–7.
31. Shimizu M, Fontana A, Takeda Y, Yagita H, Yoshimoto
T, Matsuzawa A. Induction of antitumor immunity with
Fas/APO-1 ligand (CD95L)-transfected neuroblastoma
neuro-2a cells. J Immunol 1999;162:7350–7.
32. Yu P, Lee Y, Liu W, et al. Intratumor depletion of
CD4+ cells unmasks tumor immunogenicity leading to
the rejection of late-stage tumors. J Exp Med 2005;201:
779–91.
33. Gregory MS, Repp AC, Holhbaum AM, Saff RR,
Marshak-Rothstein A, Ksander BR. Membrane Fas
ligand activates innate immunity and terminates ocular
immune privilege. J Immunol 2002;169:2727–35.
34. Chen YL, Chen SH, Wang JY, Yang BC. Fas ligand
on tumor cells mediates inactivation of neutrophils.
J Immunol 2003;171:1183–91.
35. Banz A, Pontoux C, Papiernik M. Modulation of Fas-
dependent apoptosis: a dynamic process controlling
both the persistence and death of CD4 regulatory T cells
and effector T cells. J Immunol 2002;169:750–7.
Research.
on May 26, 2016. © 2007 American Association for Cancercancerres.aacrjournals.orgDownloaded from

9. 2007;67:1291-1298.Cancer Res
Aoshuang Chen, Shanrong Liu, David Park, et al.
Adoptive T Cell Transfer
FasL Protein Transfer Enhances the Therapeutic Efficacy of
Regulatory T Cells via+CD25+Depleting Intratumoral CD4
Updated version
http://cancerres.aacrjournals.org/content/67/3/1291
Access the most recent version of this article at:
Cited articles
http://cancerres.aacrjournals.org/content/67/3/1291.full.html#ref-list-1
This article cites 35 articles, 20 of which you can access for free at:
Citing articles
http://cancerres.aacrjournals.org/content/67/3/1291.full.html#related-urls
This article has been cited by 11 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
.pubs@aacr.orgDepartment at
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Permissions
.permissions@aacr.orgDepartment at
To request permission to re-use all or part of this article, contact the AACR Publications
Research.
on May 26, 2016. © 2007 American Association for Cancercancerres.aacrjournals.orgDownloaded from