| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
,§,
,§
*
Laboratory of Molecular and Tumor Immunology, The Robert W. Franz Cancer Research Center, The Earle A. Chiles Research Institute, Providence Portland Medical Center, Portland, OR 97213; and
Biochemistry and Molecular Biology, Oregon Graduate Institute, and
Department of Molecular Microbiology and Immunology and
§
Oregon Cancer Center, Oregon Health Sciences University, Portland, OR 97201
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
) following vaccination
with gene-modified D5-Kd. Our data suggest that the failure
of unmodified D5 to generate therapeutic T cells is not due to an
inability to recognize tumor Ags, but, rather, to the induction of an
immune response that is ineffective in mediating tumor regression,
i.e., immune deviation. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, and TNF-ß) leads to cell-mediated immunity,
whereas type 2 cytokines (IL-4, -5, -6, -9, -10, and -13) are
associated with Ab production (reviewed in 1 . Initially reported
for CD4 T cells, it is now generally accepted that both CD4 and CD8 T
cells can be segregated into type 1 and type 2 populations based on
their patterns of cytokine release (1, 2, 3, 4). Uncommitted precursors can
differentiate along either pathway, with their final phenotype
determined by the cytokine milieu at the time they undergo activation
and differentiation (3, 5). The presence of IFN- and IL-12 tends to
promote the generation of type 1 cells while inhibiting the development
of type 2 cells, whereas IL-4 drives the development of type 2 cells
and inhibits type 1 (1, 6). The biologic importance of this dichotomy of T cell responses was first demonstrated for infectious diseases, where the generation of type 1 immune responses conferred protection or produced cure of animals infected with protozoa, bacteria, and fungi (reviewed in 1 . Type 1 responses also appear to be responsible for the destructive effects observed in various autoimmune states, e.g., experimental allergic encephalomyelitis, multiple sclerosis, insulin-dependent diabetes mellitus, and rheumatoid arthritis, as well as transplant rejection. In contrast, type 2 responses appear to be curative or protective in models of helminths and have been implicated in allergic reactions, including atopic asthma and Omenn’s syndrome.
The differential roles of type 1 and type 2 T cells in tumor-bearing
animals have only been described recently. A number of reports suggest
that there is a correlation between the generation of a type 1 response
and antitumor activity (7, 8, 9, 10, 11). In this study, we show that there is an
immune response in local lymph nodes
(LN)3 following vaccination
with what has been called a non-immunogenic tumor (D5). T cells from LN
draining a D5 tumor vaccine exhibited reduced expression of L-selectin
(CD62), a well-described marker of recently activated T cells (12, 13, 14, 15, 16).
Similar changes are observed in T cells draining a vaccine site
containing the immunogenic, gene-modified D5-Kd. This
suggests that both tumors had been recognized by T cells. T cells from
LN draining each of these vaccines were separated into populations with
reduced (L-selectinlow/-) or normal
(L-selectinhigh) expression of L-selectin, and their
antitumor properties were assessed in vivo following in vitro
activation with anti-CD3 and IL-2. Adoptive transfer of
L-selectinlow/- T cells from D5-Kd TVDLN
mediated significant (p < 0.05) regression of
pulmonary metastases, while T cells from D5 TVDLN were ineffective.
L-selectinhigh cells failed to mediate tumor regression
regardless of the tumor cells used for vaccination. In vitro activated
L- selectinlow/- T cells draining gene-modified or
unmodified D5 vaccines demonstrated tumor-specific cytokine release;
however, nontherapeutic D5 TVDLN were highly polarized toward a type 2
profile (IL-4, IL-10), whereas the therapeutic D5-Kd TVDLN
exhibited a type 1 profile (IFN-). Polarization of cytokine
production was already evident at the time of TVDLN harvest, 8 days
after vaccination. These results suggest that the therapeutic failure
of some tumor vaccine strategies may result from the generation of an
immune response that, while tumor-specific, does not lead to tumor
regression.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Female C57BL/6 (B6) mice were purchased from Jackson Laboratory (Bar Harbor, ME) and were maintained in a specific pathogen-free environment. Mice were generally 8 to 12 wk old when used for experiments. Recognized principles of laboratory animal care were followed (Guide for the Care and Use of Laboratory Animals, National Research Council, 1996), and all animal protocols were approved by The Earle A. Chiles Research Institute animal care and use committee.
Tumor cell lines
D5 is a poorly immunogenic subclone of the spontaneously arising B16BL6 melanoma (17). An early passage of the original BL6 tumor was provided by Dr. E. Gorelick and was subcloned by limiting dilution culture in Dr. S. Shu’s laboratory. D5 exhibits low to undetectable class I (H-2 Db and Kb) expression and no class II expression. MPR-4, which is a transformed prostate tumor cell line (provided by Dr. T. C. Thompson, Baylor College of Medicine, Houston, TX) from a C57BL/6 mouse (18), also has very low MHC class I expression.
Reagents
Magnetic beads conjugated with anti-CD62L (L-selectin) were provided by Miltenyi Biotec (Auburn, CA). The 145-2C11 hybridoma (anti-CD3) was a gift from Dr. J. A. Bluestone (University of Chicago, Chicago, IL). Recombinant human IL-2 was provided by Chiron (Emeryville, CA). DMRIE/DOPE lipid solution was provided by Vical (San Diego, CA).
Culture conditions
Lymphocytes and tumor cells were cultured in complete medium (CM), which consisted of RPMI 1640 (BioWhittaker, Walkersville, MD) containing 0.1 mM nonessential amino acids, 1 mM sodium pyruvate, 2 mM L-glutamine, and 50 µg/ml of gentamicin sulfate. This was further supplemented with 50 mM 2-ME (Aldrich, Milwaukee, WI) and 10% FBS (Life Technologies, Grand Island, NY). Tumor cells were harvested two or three times per week by brief trypsinization (BioWhittaker) and maintained in T-75 or T-150 culture flasks.
Immunization and tumor challenge
Mice were immunized by s.c. inoculation of 107 gamma-irradiated (10,000 rad) unmodified D5 tumor cells and were challenged 14 days later in the contralateral flank with viable D5 tumor. Mice were challenged with one of three different doses of tumor cells (103, 104, or 105). A challenge with 105 tumor cells induces tumors 100% of the time. Injection of 103 or 104 tumor cells is below the dose that results in tumors in 100% of animals within 14 days. Naive, age-matched control mice were included for each tumor dose. Results are presented both as the number of animals developing tumor and as the mean tumor size for animals that developed tumors.
Lipofection of tumor and vaccination for adoptive immunotherapy
An H-2 Kd expression plasmid with the CMV early gene promoter and no selectable marker gene was used for these studies. Plasmid DNA was prepared using Qiagen plasmid purification kits (Santa Clarita, CA). LPS was removed from the preparation by treatment with Triton X-114 as previously described (19). Tumor cells were harvested by trypsinization, washed twice with lactated Ringer’s solution (LR) and subsequently resuspended at 2 x 106 cells/ml. Lipofection was performed as follows. Plasmid DNA and lipid complex were prepared by mixing DNA (2.5 µg plasmid/106 tumor cells) and lipid (DMRIE/DOPE, Vical; 10 µg/106 tumor cells) in LR. The lipid/DNA complex was incubated for 5 min at room temperature before being gently mixed with D5 in LR to a final concentration of 1 to 2 x 106 cells/ml and incubated at room temperature for 1 h with an orbital shaker rotating at 75 rpm. After incubation, tumor cells were washed twice and resuspended in LR for injection or resuspended in CM and cultured for subsequent determination of trans-gene expression. Transfection efficiency, which was determined by flow cytometric analysis of H-2Kd expression, generally ranged from 30 to 70% 24 to 48 h after lipofection. One million unmodified D5 or H-2 Kd gene-lipofected D5 (D5-Kd) tumor cells were injected s.c. into both hind flanks of B6 mice.
Separation and activation of TVDLN
Eight to eleven days following tumor inoculation, the superficial inguinal LN were harvested, and lymphocytes were separated into L-selectinhigh and L-selectinlow/- populations using either panning or CD62L-MACS microbeads (Miltenyi Biotec). For panning, nylon wool-enriched T cells were incubated with anti-L-selectin (American Type Culture Collection, Rockville, MD; catalogue no. HB-132) for 30 min at 4°C and washed twice, and 80 x 106 cells in 4 ml of CM were plated onto T-25 flasks coated with rabbit anti-rat IgG (100 µg/ml; catalogue no. R-9255, Sigma, St. Louis, MO) and incubated for 45 min at 4°C. Nonadherent cells were recovered by vigorous pipetting and subjected to a second panning step on rabbit anti-rat IgG-coated dishes. Nonadherent cells recovered after this step were designated L-selectinlow/- cells. Adherent cells from the first and second panning steps were harvested using a cell scraper and pooled (L-selectinhigh). For magnetic bead separation, lymphocytes were resuspended at 108 cells/ml in CM, mixed with 10 µl/ml of CD62L-MicroBeads (497-01, Miltenyi Biotec), and incubated for 20 min at 4°C. Cells were washed once with 50 ml of CM, resuspended at 108/ml, and passaged over a VarioMACS magnetic depletion column (type BS 413-04 or CS 413-05, Milteyi Biotec) held by a high energy magnet (VarioMACS separator, Miltenyi Biotec). The column was washed extensively with CM, and nonadherent cells (L-selectinlow/- cells) were collected. The column was then removed from the magnetic field and washed to recover the bound population (L-selectinhigh) of LNC. Samples of unseparated, nonadherent, and bound cells were stained with either an FITC-labeled anti-rat Ig (panning) or an FITC-labeled anti-CD62L (magnetic bead separation) and analyzed on a Coulter flow cytometer for efficiency of separation.
Unfractionated, L-selectinhigh, and L-selectinlow/- cells were resuspended at 2 x 106 cells/ml in CM and cultured in 24-well plates with 50 µl of a 1/40 dilution of 2C11 ascites (anti-CD3). This dilution was determined previously to be optimal for T cell activation. After 2 days of activation, the T cells were harvested and expanded in CM containing 60 IU/ml rhIL-2 for 3 additional days, while culture supernatant was saved for determination of cytokine release. T cells were then harvested, washed twice in HBSS, counted, and used in adoptive transfer and cytokine release assays.
Adoptive immunotherapy
Experimental pulmonary metastases were established by i.v. inoculation of B6 mice with 1 to 3 x 105 D5 tumor cells. Three days later, after metastases were established, T cells were adoptively transferred i.v. Starting on the day of T cell infusion, mice received 90,000 IU of IL-2 twice daily for 4 days. Animals were sacrificed by CO2 narcosis 12 to 15 days following tumor inoculation, the lungs were resected and fixed in Feketes solution, and the number of pulmonary metastases was enumerated in a blinded fashion. Metastases that were too numerous to count accurately were assigned an arbitrary value of 250.
Measurement of cytokines
After activation and expansion, TVDLN were washed and incubated
alone or were stimulated with D5, MPR-4, or anti-CD3. T cells
(2 x 106/well) and tumor cells (2.5 x
105/well) were cultured in 2 ml in 24-well plates.
Supernatants were recovered 24 h (IFN-, IL-10) and 48 h
(IL-4) after stimulation and assayed in duplicate by ELISA using
commercially available reagents (IFN-, PharMingen (San Diego, CA) or
Genzyme (Cambridge, MA); IL-4, Genzyme; IL-10, PharMingen).
Supernatants were also recovered from the TVDLN cultures 2 days after
activation with anti-CD3 and before addition of IL-2. The
concentration of cytokine in the supernatant was determined by
regression analysis.
Statistical analysis
The significance of differences in the number of metastases between experimental groups was determined using the Wilcoxon rank sum test. Two-sided p values of <0.05 were considered significant. Each treatment group consisted of at least five mice, and no animal was excluded from the statistical evaluations. The significance of differences in cytokine secretion was determined using Student’s paired t test. Two-sided p values of <0.05 were considered significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
D5 has been called a poorly immunogenic/nonimmunogenic tumor
because mice cannot be protected from a subsequent tumor challenge by
immunization with irradiated tumor. B6 mice immunized with
107 gamma-irradiated (10,000 rad) D5 tumor cells were
challenged 14 days later with 103, 104, or
105 viable tumor cells. Previous studies determined that
105 tumor cells would induce tumors in 100% of the
animals. However, mice were also challenged with two lower dosages of
D5 to determine whether immunization would affect the host’s capacity
to reject a dose of tumor cells that were not uniformly tumorigenic and
thus provide a more sensitive indication of antitumor immunity. The
fraction of mice developing tumors at the three challenge doses and the
tumor growth curves are shown in Figure 1
. Similar percentages of naive and
immunized animals developed tumors at all challenge doses, and
inspection of the tumor growth curves revealed that immunization did
not slow tumor growth even when mice were challenged with as few as
1000 D5 tumor cells. Thus, immunization does not protect mice from
challenge with D5 tumor.
|
). In the same experiments, other mice
were vaccinated with D5 tumor cells that had been lipofected with an
allogeneic MHC class I plasmid vector, H-2 Kd. LN draining
this allo-modified vaccine (D5-Kd TVDLN) mediated
significant reduction of pulmonary metastases following activation with
anti-CD3/IL-2. These results confirm the poorly immunogenic nature
of the D5 tumor and demonstrate the capacity of alloantigen-modified
tumor vaccine to sensitize T cells capable of mediating a therapeutic
antitumor response.
|
Down-regulation of L-selectin expression is a well-established
marker for recently activated T cells and memory T cells (12, 13, 14, 15, 16).
Recently, Kagamu et al. demonstrated that tumor-reactive T cells from
an LN draining the weakly immunogenic MCA-205 tumor had reduced levels
of expression of L-selectin (22). Furthermore, they showed that all T
cells with therapeutic potential were contained in these
L-selectinlow/- TVDLN. Therefore, we attempted to identify
tumor-reactive T cells in LN draining the poorly immunogenic D5 tumor
by searching for T cells with reduced expression of L-selectin.
Although 10% of lymphocytes in LN from naive mice exhibited reduced
L-selectin expression, the L-selectinlow/- population
doubled to 20% following immunization with D5 tumor (Fig. 2, A and B).
Analysis of L-selectin on D5-Kd TVDLN revealed a similar
finding; 25% of LNC exhibited decreased expression (2C). The
down-regulation of L-selectin in both D5 and D5-Kd TVDLN
implied that T cells were activated in both instances.
|
, D through F and G through
L, illustrates the results of typical separations using
panning or anti-CD62L magnetic beads to fractionate cells based on
their L-selectin expression. L-selectinlow/- and
L-selectinhigh T cells were then activated with
anti-CD3 and expanded in IL-2. As described previously,
L-selectinlow/- T cells treated with anti-CD3 and IL-2
expanded more than L-selectinhigh cells (22) (data not
shown). In five experiments the former averaged 7.2-fold expansion
compared with 3.7-fold for the L-selectinhigh cells. Reduced L-selectin expression identifies therapeutic T cells in LN draining gene-modified vaccine
To measure directly the antitumor activity of in vitro activated
L-selectinlow/- T cells, they were adoptively transferred
into mice bearing 3-day established metastases. Even though D5 TVDLN
contained increased numbers of L-selectinlow/- cells,
presumed to be tumor-reactive, they failed to exhibit therapeutic
activity when activated with anti-CD3 and IL-2 and adoptively
transferred to tumor-bearing hosts (Table II). Similarly,
L-selectinhigh TVDLN from the same mice also failed to
exhibit any therapeutic activity. Thus, enrichment and activation of
the T cells presumed to be tumor reactive did not improve efficacy
following vaccination with unmodified D5. However, this was not the
case when TVDLN from mice vaccinated with alloantigen-modified tumor,
D5-Kd, were tested. Comparable numbers of L-
selectinlow/- T cells from D5-Kd TVDLN
exhibited significantly greater (p < 0.05)
therapeutic activity than L-selectinhigh TVDLN (Table II).
Thus, despite the existence of cells with reduced expression of
L-selectin in TVDLN from mice immunized with unmodified or
alloantigen-modified tumor, the only T cells with antitumor potential
were found in the LN draining D5-Kd.
|
To determine the reason for the differential antitumor effects of
the L-selectinlow/- cells, which we presumed contained the
recently activated tumor-specific cells, we examined the ability of
L-selectinlow/- T cells from TVDLN of D5 or
D5-Kd tumor vaccines to secrete IFN-. Tumor-specific
IFN- secretion has been associated with the therapeutic efficacy of
adoptively transferred T cells (7, 23, 24, 25). Stimulation with
anti-CD3 induced a similar level of IFN- secretion by
L-selectinlow/- T cells from LN draining either D5 or
D5-Kd vaccines. However, only T cells from the
D5-Kd TVDLN secreted IFN- specifically in response to
tumor stimulation (Fig. 3A).
IFN- was not produced when T cells were cultured alone or in the
presence of syngeneic, but unrelated, prostate cancer cells, MPR-4.
Thus, tumor-specific release of IFN- correlated with the therapeutic
activity of these T cells. L-selectinhigh TVDLN did not
secrete IFN- in response to stimulation with D5 tumor regardless of
the cells used for vaccination (Fig. 3D).
|
B). However, in contrast to IFN-,
only T cells from the D5 vaccine secreted IL-4 specifically in response
to stimulation with D5 tumor cells. This response was clearly tumor
specific, since IL-4 was not secreted by T cells cultured alone or with
the unrelated tumor MPR-4. No tumor-specific secretion of IL-4 was
observed from L-selectinhigh T cells from D5 or
D5-Kd TVDLN (Fig. 3E). To confirm the type 2
cytokine profile of L-selectinlow/- D5 TVDLN, we measured
IL-10 production (Fig. 3C). L-selectinlow/- T
cells from D5 TVDLN produced more IL-10 following stimulation with D5
than without stimulation or after stimulation with MPR-4
(p < 0.05). Interestingly, while
L-selectinlow/- T cells expanded from both D5 and
D5-Kd TVDLN released similar levels of IFN- and IL-4
after polyclonal stimulation with anti-CD3, T cells from the
D5-Kd TVDLN released less IL-10 than
L-selectinlow/- T cells from D5 TVDLN
(p < 0.05). Additional studies have found the
anti-CD3-stimulated release of IL-10 to be more variable from
experiment to experiment, and current studies are investigating whether
this is related to the development of a type 1 cytokine response.
Interestingly, L-selectinhigh T cells from D5 or
D5-Kd TVDLN released similar levels of IL-10 following
stimulation with anti-CD3 (Fig. 3F). Although the
cytokine profiles of D5 and D5-Kd TVDLN were markedly
different, the phenotypes of the T cells were indistinguishable (Table III).
|
Cytokine production by freshly isolated TVDLN
Previous studies have demonstrated that the cytokine environment
in which T cells are first activated and begin differentiation
determines the subsequent cytokine profile of those cells (1, 3).
Because noncultured TVDLN require additional activation in vitro with
tumor cells, microbial superantigens, or Abs to CD3 or the TCR to
develop therapeutic activity, we have concluded that they are not fully
differentiated effector T cells (20, 24, 26). Therefore, to determine
whether the cytokines released by these fresh cells during the in vitro
activation step might account for the generation of highly polarized
effector T cells, we examined which cytokines were released into the
culture medium during the in vitro activation step with anti-CD3.
L-selectinlow/- TVDLN from both D5 and D5-Kd
vaccines were activated as described above, and supernatants were
recovered at the end of the second day of activation. As a control, LN
from naive mice were harvested, and L-selectinlow/- cells
were isolated and cultured under identical conditions as TVDLN.
Analysis of cytokine production by D5 or D5-Kd TVDLN
revealed striking differences in the amounts of IFN- and IL-4
secreted. Figure 4 depicts the ratios of
the concentrations of IFN- and IL-4 found in the supernatant after 2
days of culture with anti-CD3 in four experiments.
L-selectinlow/- T cells from the D5 vaccine exhibited a
low ratio of IFN- to IL-4, indicating a polarization toward a type 2
response, while the ratio of cells draining the D5-Kd
vaccine more closely resembled that of naive LN cells. Thus, these
results suggest that the response of L-selectinlow/- T
cells in D5 TVDLN is already biased toward a type 2 profile by 8 days
following vaccination. Furthermore, the apparent excess in IL-4 release
during the 2-day culture with anti-CD3 may contribute to the highly
polarized tumor-specific cytokine release profile observed after the
5-day in vitro culture. A similar polarization of IL-10 secretion was
not observed; L-selectinlow/- cells from both D5 and
D5-Kd TVDLN secreted similar levels of IL-10 following
activation with anti-CD3 (data not shown). This is in agreement
with other reports that early expression of IL-4 or IFN-, but not of
IL-10, directs the subsequent polarization of primary CD4+
T cell cultures (27).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
These observations forced us to reevaluate our perspective on the immunogenicity of tumors. The molecular mechanisms for the inherent level of immunogenicity of tumors is poorly understood. There are many explanations for the poorly immunogenic phenotype. Speculation is that poorly immunogenic/nonimmunogenic tumors fail to exhibit or fail to present tumor-associated Ags. They may loose MHC class I/II molecules, lack accessory and/or costimulatory molecules, or have defects in Ag processing or transport pathways (28). Alternatively, since many of the recently described tumor-associated Ags are normal proteins with restricted tissue expression (e.g., Melan-A MART-1 and gp100), it is possible that the host may have deleted these self-Ag-reactive T cells during ontogeny and thus be tolerant to these self Ags. Another possibility is that tumor-reactive T cells in the host have been anergized. This may occur because tumor Ags are presented by tumor, or by APC, without the appropriate second signal (29, 30). All these explanations assume that there is a failure by the host to develop an antitumor immune response. Recently, immune deviation has been identified as a third mechanism, together with anergy and clonal deletion, capable of inducing tolerance in animals (31, 32). Immune deviation results in the inhibition of effector function by generating an opposing immune response against the same target Ag. The results of our study are consistent with the hypothesis that immune deviation is responsible for the poorly immunogenic nature of the D5 tumor.
Our observations support the results in animal models and a
clinical trial that demonstrated the efficacy of adoptively transferred
T cells is associated with tumor-specific release of IFN- or type 1
cytokines (11, 23, 24). Others have also observed a correlation between
the in situ induction of T cells that exhibit tumor-specific release of
type 1 cytokines, particularly IFN-, and tumor regression (7, 25).
Additionally, a recent analysis of spontaneously regressing and
progressing melanoma lesions suggested a significant correlation
between message for type 1 cytokine and tumor regression, thus further
extending the hypothesis that tumor-specific type 1 cytokine profiles
are associated with immune destruction of tumor metastases (9). This
paradigm of a therapeutic type 1 response and a nonprotective type 2
response closely resembles that observed in animal models of some
parasitic diseases (1).
The immune response to Leishmania major is a
well-studied model of immunoregulation. C57BL/6 mice, resistant to
infection with L. major, develop a Th1 response following
infection and induce CTL that eliminate the pathogen. BALB/c mice are
susceptible to infection with L. major, because Th1 effector
function is inhibited by Th2 cytokines produced in response to
infection (33, 34). Recently, it has been demonstrated that type
1-promoting cytokines such as IL-12 or, to a lesser extent, IFN-,
have therapeutic action in early infection as well as in the presence
of a disease-exacerbating Th2-type response (33, 35, 36). The
therapeutic success of IL-12 in several animal tumor models, covering
the spectrum of immunogenicity, suggests another comparison between the
immune response to this parasitic disease and cancer (7, 37, 38, 39).
Although observations made using L. major cannot necessarily
be extrapolated to tumor immunology, they provide substantial
information about the significant and diverse effects that polarized
cytokine response can mediate in vivo.
A number of studies have identified type 2 T cells infiltrating tumor
sites (40, 41). However, it has not been shown that these type 2 T
cells are specifically primed to tumor. Recently, a report by Aruga and
his colleagues documented both tumor-specific type 1 (IFN-) and type
2 (IL-10) cells in TVDLN draining the weakly immunogenic MCA-207
sarcoma (8). However, secretion of tumor-specific type 2 cytokines by
TVDLN did not block antitumor activity. The implication is that a
strong, predominantly type 1 response by the effector T cells in
MCA-207 TVDLN could overcome the possibly weaker, type 2 influence to
mediate tumor regression. Similarly, Tsung and colleagues observed that
progressively growing tumor nodules were associated with T cells
producing both IL-4 and IL-10 without evidence of IFN- production
(7). Following treatment with IL-12, IFN- expression was observed,
and tumors regressed even though IL-4 and IL-10 were still produced at
the tumor site. A similar observation has recently been reported in the
induction of autoimmune diabetes (42). These results suggest that it is
not a single cytokine that is critical to defining effector activity.
Similarly, it is probably not a single cytokine that is critical in
defining the subsequent cytokine profile of the responding T cell, but
the ratio of type 1 to type 2 cytokines present during
activation/differentiation (6). Our own data are consistent with this
model. While effector T cells exhibit a highly polarized cytokine
profile, the activation of freshly isolated TVDLN T cells with
anti-CD3 induces a cytokine milieu that includes IL-4 and IFN-.
However, the ratio of IFN- to IL-4 is substantially depressed for D5
TVDLN, while the ratio for D5-Kd TVDLN is similar to that
observed for naive LNC. These observations raise intriguing questions
about how the poorly immunogenic D5 tumor induces this alteration in
the ratio of type 1 and type 2 cytokines. The easiest explanation might
be that the D5 tumor produces type 2 cytokines that skew the antitumor
immune response in that direction. However, we have been unable to
identify any secretion of IL-3, IL-4, IL-5, or IL-10 by the D5 tumor,
negating a role for established type 2 cytokines in this effect (Fig. 3
and data not shown). Alternatively, a recent report by Rincon and
colleagues has suggested that IL-6 may also promote the polarization of
type 2 responses by inducing IL-4 release by T cells (43). Future
experiments will evaluate this possibility. Additional possibilities
are that generation of an immune response to tumor in the absence of
proinflammatory type 1 cytokines allows the induction of a type 2
response by default. In this way, the response to tumor may resemble
the immune response to self Ags in transgenic mice, where development
of a Th2 response blocks the development of self-Ag-reactive Th1 cells.
A similar pattern exists in the immune response to soluble protein Ags,
where type 2 responses predominate. Is the poorly immunogenic tumor,
which cannot act as its own APC, serving as a slow release reservoir of
soluble Ags and thereby inducing a type 2 antitumor response by
default?
If one accepts that the type 2 response is the natural default response to Ag in the absence of proinflammatory cytokines, vaccination with allo-modified D5 may provide additional stimuli that skew the response from the default type 2 pathway to a predominantly type 1 profile. We have previously demonstrated that engineering allo-MHC into tumor vaccines is more therapeutic than lipofection with other genes, and Plautz has identified a requirement for a complete functional allo-molecule in the generation of an antitumor effect (44, 45, 46). Moreover, since type 1 responses are generally associated with the anti-allo response seen in transplant rejection, it is possible that the expression of allo-MHC at the vaccine site contributes to the observed shift (47). However, it has been difficult to reproducibly identify the generation of an anti-H-2 Kd immune response in D5-Kd TVDLN (H.-M. Hu and B. A. Fox, unpublished observation).
Generation of type 1 responses is also favored when higher
concentrations of Ag are present during sensitization/activation, while
low concentrations of Ag promote type 2 responses (48, 49). Recently,
we demonstrated that the process of lipofection leads to increased
expression of endogenous MHC class I molecules on tumor cells that
normally exhibit low to undetectable levels of class I (50). This
effect could increase the concentration of tumor Ags and may promote
the development of a type 1 antitumor response. Moreover, the process
of lipofecting plasmid DNA into tumor cells inevitably carries with it
some low level of endotoxin. While most reports suggest that fairly
high levels (100 µg) of endotoxin are required to induce substantial
in vivo effects, a recent report by Tough et al. demonstrates that even
small doses of LPS (10 ng) administered iv indirectly leads to marked
proliferation of CD8+ T cells in the LN and spleen (51).
Thus, indirect activation of T cells by contaminating LPS in the
vaccine preparation may provide proinflammatory cytokines that could
polarize the antitumor response in the direction of a type 1 profile.
Additionally, bacterial DNA preparations can also induce IFN-
secretion indirectly via the stimulation of IL-12 and TNF-
secretion
by adherent cells (52). This effect is sensitive to treatment with
DNase, implying a requirement for intact DNA. The induction of this
cytokine pattern may explain the general tendency of DNA vaccines to
induce predominantly type 1 responses (53). However, DNA and LPS are
unlikely to be the sole inducers of a therapeutic immune response in
this model. Evidence from several studies confirms that LPS and DNA,
present in preparations of ineffective gene constructs and injected
intratumorally or mixed with tumor vaccines, are not sufficient to
induce highly therapeutic antitumor responses. Therefore, we propose
that these molecules may play an accessory role in promoting the
activation and differentiation of T cells skewed toward a type 1
profile.
If these observations in the D5 model can be confirmed in patients with
cancer, novel strategies may be applied to shift the nontherapeutic
profile toward an effective profile. Several studies have demonstrated
that naive and even memory Th cells with the
L-selectinlow/- phenotype can be forced to differentiate
into polarized memory effector cells of either type 1 or type 2
profiles by activation in the presence of exogenous cytokines. IFN-
and IL-12 in combination with anti-IL-4 neutralizing mAbs are
highly effective at polarizing T cells to type 1 profiles, whereas IL-4
and neutralizing anti-IFN- mAb can skew T cell responses toward
type 2 profiles. In contrast to these studies, cells from mice
chronically infected with L. major cannot be forced to
change their cytokine profile by manipulation of the cytokine milieu
(54). If patients with cancer do develop polarized tumor-specific type
2 responses it will be important to determine whether they can be
modulated toward a type 1 profile. If chronic exposure to tumor has
permanently altered the cytokine profile to that of type 2, as observed
in L. major, it will be critical to determine whether pools
of naive T cells remain that can be sensitized and skewed toward a
tumor-specific type 1 response.
In summary, the poorly immunogenic/nonimmunogenic D5 tumor for which standard vaccination strategies do not provide protection from a tumor challenge sensitizes T cells in TVDLN that exhibit a tumor-specific type 2 response following in vitro activation with anti-CD3 and IL-2. An allo-modified vaccine of the same tumor induces T cells with a tumor-specific type 1 cytokine response that exhibit therapeutic efficacy.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Bernard A. Fox, The Robert W. Franz Cancer Research Center, The Earle A. Chiles Research Institute, 4805 N.E. Glisan, Portland, OR 97213. E-mail address:
3 Abbreviations used in this paper: LN, lymph nodes; D5, B16BL6-D5; D5-Kd, D5 tumor transfected with H-2Kd plasmid DNA; L-selectinlow/-, cells with reduced expression of L-selectin; L-selectinhigh, cells with increased expression of L-selectin; TVDLN, tumor vaccine draining lymph nodes; B6, C57BL/6; CM, complete medium; LR, lactated Ringer’s solution; LNC, lymph node cells; -CD3, hamster anti-mouse CD3 
-chain.
Received for publication December 31, 1997. Accepted for publication May 19, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
in stabilizing the T helper cell type 1 and 2 phenotype. J. Immunol. 158:2648.[Abstract]
during the differentiation of cytotoxic T lymphocytes induced by allografts. J. Immunol. 148:2348.[Abstract]
and tumor necrosis factor have a role in tumor regressions mediated by murine CD8+ tumor-infiltrating lymphocytes. J. Exp. Med. 173:647. production by anti-tumor T cells. Int. Immunol. 7:1135., but not of interleukin-2 and interleukin-10, reflects later polarization of primary CD4+ T cell cultures. Eur J. Immunol. 26:1565.[Medline]
production. J. Immunol. 153:1697.[Abstract]
, but not of interleukin-4. Diabetologia 37:1154.[Medline]
ß-transgenic model. J. Exp. Med. 182:1579. production by stimulation of interleukin-12 and tumor necrosis factor-. Cell. Immunol. 167:72.[Medline]
This article has been cited by other articles:
|
|
 |
|
  K. Sasaki, X. Zhao, A. D. Pardee, R. Ueda, M. Fujita, S. Sehra, M. H. Kaplan, L. P. Kane, H. Okada, and W. J. Storkus Stat6 Signaling Suppresses VLA-4 Expression by CD8+ T Cells and Limits Their Ability to Infiltrate Tumor Lesions In Vivo J. Immunol., July 1, 2008; 181(1): 104 - 108. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. J. Stewart and S. I. Abrams Altered Immune Function during Long-Term Host-Tumor Interactions Can Be Modulated to Retard Autochthonous Neoplastic Growth J. Immunol., September 1, 2007; 179(5): 2851 - 2859. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. J. Dobrzanski, J. B. Reome, J. C. Hylind, and K. A. Rewers-Felkins CD8-Mediated Type 1 Antitumor Responses Selectively Modulate Endogenous Differentiated and Nondifferentiated T Cell Localization, Activation, and Function in Progressive Breast Cancer J. Immunol., December 1, 2006; 177(11): 8191 - 8201. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Li, W. Wojciechowski, C. Dell'Agnola, N. E. Lopez, and I. Espinoza-Delgado IFN-{gamma} and T-bet Expression in Human Dendritic Cells from Normal Donors and Cancer Patients Is Controlled through Mechanisms Involving ERK-1/2-Dependent and IL-12-Independent Pathways J. Immunol., September 15, 2006; 177(6): 3554 - 3563. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S.-C. Yang, R. K. Batra, S. Hillinger, K. L. Reckamp, R. M. Strieter, S. M. Dubinett, and S. Sharma Intrapulmonary Administration of CCL21 Gene-Modified Dendritic Cells Reduces Tumor Burden in Spontaneous Murine Bronchoalveolar Cell Carcinoma. Cancer Res., March 15, 2006; 66(6): 3205 - 3213. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Q. Li, A. C. Grover, E. J. Donald, A. Carr, J. Yu, J. Whitfield, M. Nelson, N. Takeshita, and A. E. Chang Simultaneous Targeting of CD3 on T Cells and CD40 on B or Dendritic Cells Augments the Antitumor Reactivity of Tumor-Primed Lymph Node Cells J. Immunol., August 1, 2005; 175(3): 1424 - 1432. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. J. Sussman, R. Parihar, K. Winstead, and F. D. Finkelman Prolonged Culture of Vaccine-Primed Lymphocytes Results in Decreased Antitumor Killing and Change in Cytokine Secretion Cancer Res., December 15, 2004; 64(24): 9124 - 9130. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S.-C. Yang, S. Hillinger, K. Riedl, L. Zhang, L. Zhu, M. Huang, K. Atianzar, B. Y. Kuo, B. Gardner, R. K. Batra, et al. Intratumoral Administration of Dendritic Cells Overexpressing CCL21 Generates Systemic Antitumor Responses and Confers Tumor Immunity Clin. Cancer Res., April 15, 2004; 10(8): 2891 - 2901. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. J. Dobrzanski, J. B. Reome, J. A. Hollenbaugh, and R. W. Dutton Tc1 and Tc2 Effector Cell Therapy Elicit Long-Term Tumor Immunity by Contrasting Mechanisms That Result in Complementary Endogenous Type 1 Antitumor Responses J. Immunol., February 1, 2004; 172(3): 1380 - 1390. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Hillinger, S.-C. Yang, L. Zhu, M. Huang, R. Duckett, K. Atianzar, R. K. Batra, R. M. Strieter, S. M. Dubinett, and S. Sharma EBV-Induced Molecule 1 Ligand Chemokine (ELC/CCL19) Promotes IFN-{gamma}-Dependent Antitumor Responses in a Lung Cancer Model J. Immunol., December 15, 2003; 171(12): 6457 - 6465. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. H. Poehlein, H.-M. Hu, J. Yamada, I. Assmann, W. G. Alvord, W. J. Urba, and B. A. Fox TNF Plays an Essential Role in Tumor Regression after Adoptive Transfer of Perforin/IFN-{gamma} Double Knockout Effector T Cells J. Immunol., February 15, 2003; 170(4): 2004 - 2013. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. M. Jensen, S. L. Meijer, R. A. Kurt, W. J. Urba, H.-M. Hu, and B. A. Fox Regression of a Mammary Adenocarcinoma in STAT6-/- Mice Is Dependent on the Presence of STAT6-Reactive T Cells J. Immunol., February 15, 2003; 170(4): 2014 - 2021. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Huang, F. Li, J. R. Gordon, and J. Xiang Synergistic Enhancement of Antitumor Immunity with Adoptively Transferred Tumor-specific CD4+ and CD8+ T Cells and Intratumoral Lymphotactin Transgene Expression Cancer Res., April 1, 2002; 62(7): 2043 - 2051. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. E. Curiel, C. S. Garcia, L. Farooq, M. F. Aguero, and I. Espinoza-Delgado Bryostatin-1 and IL-2 Synergize to Induce IFN-{gamma} Expression in Human Peripheral Blood T Cells: Implications for Cancer Immunotherapy J. Immunol., November 1, 2001; 167(9): 4828 - 4837. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Sato, Y. Torimoto, Y. Tamura, M. Shindo, H. Shinzaki, K. Hirai, and Y. Kohgo Immunotherapy using heat-shock protein preparations of leukemia cells after syngeneic bone marrow transplantation in mice Blood, September 15, 2001; 98(6): 1852 - 1857. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Sharma, M. Stolina, L. Zhu, Y. Lin, R. Batra, M. Huang, R. Strieter, and S. M. Dubinett Secondary Lymphoid Organ Chemokine Reduces Pulmonary Tumor Burden in Spontaneous Murine Bronchoalveolar Cell Carcinoma Cancer Res., September 1, 2001; 61(17): 6406 - 6412. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. J. Dobrzanski, J. B. Reome, and R. W. Dutton Role of Effector Cell-Derived IL-4, IL-5, and Perforin in Early and Late Stages of Type 2 CD8 Effector Cell-Mediated Tumor Rejection J. Immunol., July 1, 2001; 167(1): 424 - 434. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Winter, H.-M. Hu, K. McClain, W. J. Urba, and B. A. Fox Immunotherapy of Melanoma: A Dichotomy in the Requirement for IFN-{{gamma}} in Vaccine-Induced Antitumor Immunity Versus Adoptive Immunotherapy J. Immunol., June 15, 2001; 166(12): 7370 - 7380. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Kjærgaard, J. Tanaka, J. A. Kim, K. Rothchild, A. Weinberg, and S. Shu Therapeutic Efficacy of OX-40 Receptor Antibody Depends on Tumor Immunogenicity and Anatomic Site of Tumor Growth Cancer Res., October 1, 2000; 60(19): 5514 - 5521. [Abstract] [Full Text]
|
|
|
|
|
|
A. D. Weinberg, M.-M. Rivera, R. Prell, A. Morris, T. Ramstad, J. T. Vetto, W. J. Urba, G. Alvord, C. Bunce, and J. Shields Engagement of the OX-40 Receptor In Vivo Enhances Antitumor Immunity J. Immunol., February 15, 2000; 164(4): 2160 - 2169. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Winter, H.-M. Hu, W. J. Urba, and B. A. Fox Tumor Regression After Adoptive Transfer of Effector T Cells Is Independent of Perforin or Fas Ligand (APO-1L/CD95L) J. Immunol., October 15, 1999; 163(8): 4462 - 4472. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Matsui, J. D. Ahlers, A. O. Vortmeyer, M. Terabe, T. Tsukui, D. P. Carbone, L. A. Liotta, and J. A. Berzofsky A Model for CD8+ CTL Tumor Immunosurveillance and Regulation of Tumor Escape by CD4 T Cells Through an Effect on Quality of CTL J. Immunol., July 1, 1999; 163(1): 184 - 193. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
