| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Tenovus Research Laboratory, Cancer Sciences Division, University School of Medicine, Southampton General Hospital, Southampton, United Kingdom
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, but not CD4
or CTLA-4, indicating the presence of a conventional cellular Th1
response. Furthermore, we found evidence of cross-recognition between
lymphomas (BCL1 and A20) as measured by cytotoxicity and
IFN- responses in vitro and using tumor rechallenge experiments,
suggesting common target Ags. Finally, although anti-CD40 was shown
to stimulate NK cell killing, we could find no role for these cells in
controlling tumor growth. These data underline the ability of
anti-CD40 mAb to potentiate CTL responses and the potency of
cellular immunity in eradicating large quantities of syngeneic
tumor. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CD40 is a member of the TNFR superfamily that plays a critical role in
both humoral and cellular immune responses (6). It was
initially shown to be constitutively expressed on the majority of B
cells but has since been found on a wide range of cells, including
epithelial and endothelial cells and on all APC. The natural ligand for
CD40, CD154, is a trimeric TNF-like molecule that is expressed mainly
on activated Th cells (6) and appears to provide an
important pathway by which T cells can control immune responses via the
APC. For example, several groups have shown that ligation of CD40 on
dendritic cells (DC)3
is a potent method of improving the Ag-presenting function of these
cells and of empowering them to present Ag to naive CD8 cells
(7, 8, 9). Such conditioning or licensing of DC can be
achieved with soluble CD40 ligand (CD40L) and CD40 mAb, which
are both able to provide sufficient cross-linking activity to stimulate
the CD40-expressing cells. While we still have much to learn about the
cellular responses to CD40 ligation, the more obvious changes include
increased expression of adhesion and costimulatory molecules, such as
ICAM-1, CD80/CD86, and CD40, and increased production of inflammatory
cytokines such as IL-12, IL-18, and macrophage-inflammatory
protein-1
(10). Results such as these support the view
that CD40 stimulation promotes a pattern of cytokine production that
skews toward a Th1-type immune response (11). Importantly,
CD40 mAb and soluble CD40L are already showing evidence of clinical
potential, being able to potentiate immune responses in a number of
therapeutic settings, including T cell responses to peptides and
proteins (12, 13), infectious disease (14, 15), and treating neoplastic disease in animals and patients
(16, 17, 18, 19).
We have recently shown that anti-CD40 mAb can stimulate extremely potent CTL responses against syngeneic murine lymphomas (16). These responses eradicate the existing tumors, leave mice resistant to rechallenge, and bypass the need for T cell help. In this study we extend this work to define the nature of the CD8 responses and investigate their specificity in a range of mouse lymphoma models.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
BALB/c, CBA/H and C57/Blk mice were all supplied by Harlan (Blackthorn, Oxon, U.K.) and maintained in local animal facilities.
BCL1 (20) and A31 (21)
mouse B lymphoma lines were maintained by in vivo, i.p. passage in
BALB/c and CBA mice, respectively. Spleens were removed at the terminal
stage of disease and single cell suspensions were prepared as described
previously (22). 
BCL1 cells are a
subline derived from BCL1 which grow in culture
(23). A20, EL4, and YAC cell lines (all American Type
Culture Collection, Manassas, VA) were maintained in culture using
standard medium (see below).
All cell culture was performed in supplemented DMEM (containing 2 mM glutamine, 1 mM pyruvate, 100 IU/ml penicillin and streptomycin, 2 mg/ml fungizone, and 5 or 10% FCS (Myoclone)) (Life Technologies, Paisley, U.K.), or in supplemented RPMI 1640 (Life Technologies) containing the same supplements but with the addition of 50 µM 2-ME (BDH Laboratory Supplies, Poole, U.K.).
Antibodies
mAbs used in this study were M17/5.2 (anti-LFA-1), 16-10A1
(anti-B7-1), GL-1 (anti-B7-2), R4-6A2 (anti-IFN-),
M5/114.15.2 (anti-MHC class II), 34-1-2S (anti-MHC class I,
Kd, and Dd), all obtained
from the American Type Culture Collection. In addition, 3/23
(anti-CD40) was a gift from G. Klaus (National Institute of Medical
Research, London, U.K.) (24); ID3 (anti-CD19)
was provided by D. Fearon (University of Cambridge School of Clinical
Medicine, Cambridge, U.K.) (25); and K9-18
(anti-Kd) and 19-191
(anti-Dd) were kindly provided by D. Kioussis
(National Institute of Medical Research) and YTA3.1.2 (anti-CD4)
and YTS 169 (anti-CD8) were from S. Cobbold (Sir William Dunn
School of Pathology, Oxford, U.K.) (26). The anti-B
cell receptor mAb Mc10-6A5 (anti-BCL1 Id) and
Mc39-12 (anti-µ-chain) were prepared in house and have been
described previously (27).
Anti-asialo GM1 (anti-ASGM1) antiserum for NK cell depletion was from Wako (Neuss, Germany).
Hybridoma cells were expanded in stationary culture using 5% supplemented DMEM and their IgG mAb prepared by precipitation with saturated ammonium sulfate followed by fractionation on protein A or G (Amersham Pharmacia Biotech, Piscataway, NJ) according to the manufacturer’s instructions. Four of the rat mAb (Mc10-6A5, M5/114.15.2, YTS169, and YTA3.1.2) were prepared by ion-exchange chromatography on DEAE (Whatman, Clifton, NJ) as described previously (28). The purity of all IgG preparations was checked by electrophoresis (Beckman EP system; Beckman Coulter, Palo Alto, CA) and HPLC using a Zorbax GF250 Bio Series column (Jones Chromatography, Hengoed, U.K.) (29).
Immunotherapy
Groups of age-matched mice were injected i.v. with tumor cells (BCL1, A31, A20, or EL4) on day 0 and, unless stated otherwise, treated with mAb daily for 4 days at the times indicated for each experiment. In vivo T cell depletion was as described by Cobbold et al. (26), using i.p. injection of 0.5 mg of anti-CD8 mAb (YTS 169) and/or 1 mg of anti-CD4 mAb (YTA3.1.2). The injections were repeated every 4–5 days, or as indicated, and the effectiveness of depletion was confirmed by flow cytometry on PBL.
Flow cytometric analysis of splenic lymphocytes following tumor and anti-CD40 mAb
To follow changes in splenic lymphocytes, groups of age-matched BALB/c mice were given 5 x 107 BCL1 cells on day 0 (i.v.) and then 1 mg of anti-CD40 mAb or an isotype-matched control (i.v.) on the day when the level of tumor cells in the spleen had reached 5–10% of total cells, which was typically day 4–6 post-tumor. To ensure that the tumor load was in this range at the time of treatment, parallel groups of mice were assessed for Id+ splenocytes by flow cytometry on a FACSCalibur (BD Biosciences, Mountain View, CA) using PE-anti-CD19 and FITC-anti-BCL1 Id. Changes in CD4+ and CD8+ lymphocytes were followed using FITC-anti-CD4 and PE-anti-CD8. Knowing the yield of spleen cells (normal plus tumor) and the percentage of each subset from the flow cytometric analysis we were able to estimate the total yield of each type of cell. In experiments to look at the effect of various blocking mAb on the response anti-CD40 mAb treatment, 0.5 mg of the blocking mAb was injected i.p. on the same day as the injection of anti-CD40, and then again on days 3 and 4 after the anti-CD40 mAb.
Cytotoxicity assays
A standard 4-h 51Cr release assay (30) was used to assess cytotoxic activity of splenic effectors (NK or cytotoxic CD8). Splenic homogenates were prepared from mice and, when necessary, any contaminating tumor cells were removed using FITC-anti-Id mAb, followed by anti-FITC MACS beads and LS columns (Miltenyi Biotec, Bergisch Gladbach, Germany) as described in the manufacturer’s instructions. The remaining effector cells were washed and resuspended at 2 x 107 cells/ml before being diluted for mixing at the required ratio with 51Cr-labeled target cells (final volume of 200 µl) in a U-bottom culture plate (Life Technologies). In experiments to determine the blocking activity of Abs, mAb were included at a final concentration of 50 µg/ml. The plates were centrifuged at 200 x g for 5 min at room temperature, incubated for 4 h at 37°C in a CO2 incubator, and then centrifuged at 500 x g for 5 min before finally harvesting 100 µl of the supernatant to estimate 51Cr release. All determinations were performed in triplicate. The maximum release of radioactivity was calculated using target cells to which 150 µl of 1% Nonidet P-40 had been added. The percentage of specific 51Cr release was calculated using the standard formula: percentage of specific release = [(sample release - background release)/(maximum release - background release)] x 100.
Rechallenge of immune splenocytes with tumor in vitro
Splenocytes were prepared from mice that had previously been
cured of BCL1 lymphoma by anti-CD40 mAb
between days 30 and 50 post-initial tumor challenge. A total of 5
x 105 cells were then mixed with 5 x
104 BCL1 or other BALB/c
tumor cells as stimulators in a final volume of 200 µl (supplemented
RPMI 1640) in a U- bottom culture plate and plates were centrifuged at
200 x g for 5 min at room temperature before
incubating for 48–72 h at 37°C in a CO2
incubator. The plates were then centrifuged at 500 x g
for 5 min before harvesting the supernatant to determine the level of
IFN- release by ELISA as described below.
ELISA for IFN- and IgM idiotype
ELISA were based on the method of Engvall and Perlman
(31). For the measurement of IFN-, 96-well ELISA plates
(Nunc Immunoplate; Invitrogen, Paisley, U.K.) were coated for 18 h
at 4°C with anti-IFN- mAb (4 µg/ml). After blocking with 1%
BSA in PBS, test supernatants and standards were added to the wells and
incubated for 2 h at 20°C. IFN- was detected using
biotinylated anti-IFN- (Serotec, Oxford, U.K.) followed by
avidin-HRP (BD Biosciences, Oxford, U.K.). For the measurement of serum
IgM Id, plates were coated with anti-BCL1 Id
mAb. After blocking with 1% BSA/PBS, dilutions of test sera and
standards were added to the wells and incubated for 90 min at 37°C.
IgM was then detected directly using HRP-coupled rat anti-mouse
µ-chain. To look for class switching of the IgM Id to IgG
Id, the final reagent was replaced by HRP-labeled rat
anti-mouse IgG (Jackson ImmunoResearch Laboratories, West
Grove, PA).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Recent studies have shown that anti-CD40 mAb is highly
effective in the treatment of a range of syngeneic mouse lymphomas.
Furthermore, rather than targeting the neoplastic cells directly, the
mAb has the potential to operate by activating APC and generating CTL
responses that eradicate the tumor (16). In this work we
have confirmed and extended these observations, showing that CD40 mAb
treatment remains effective even after the disease has become widely
disseminated. For example, Fig. 1
shows
changes in the level of BCL1-derived IgM Id in
groups of mice inoculated i.v. with BCL1
(105) on day 0 and treated with mAb on days
16–19. At this late stage of treatment the tumor cells were already
widely distributed throughout the secondary lymphoid organs, the bone
marrow, and the liver (data not shown), and the serum concentration of
idiotypic IgM was increasing exponentially. In mice given the control
IgG, the level of serum Id continued to rise with the growth of the
BCL1 until the experiment was terminated on day
27, at which point it had reached 
800 µg/ml. For the first 5–6
days after starting treatment with CD40 mAb, the rise in Id IgM was
very similar to or perhaps slightly sharper than that in the control
group, at which point it dropped markedly and eventually became
undetectable some 20 days after treatment had started. All unculled
mice in this treated group survived for >70 days and were resistant to
rechallenge with fresh BCL1. An alternative ELISA
format was also used to show that the loss of serum IgM Id was not the
result of class switching to IgG (data not shown).
|
shows the results obtained when
groups of mice were given between 105 and
107 A31 i.v. on day 0, followed by four mAb
treatments on days 3–6. Surprisingly, we consistently found that as
the tumor dose was increased so the efficacy of treatment
improved. Thus at the lowest cell dose (105 A31),
the anti-CD40 mAb provided 25 days protection, while at the high
cell-dose (107), all mice survived beyond 100
days (defined as long-term survivors). To investigate the generality of
this phenomenon, we repeated similar experiments with A31 and three
other lymphomas, BCL1, A20, and EL4. EL4 is a
CD40- T cell lymphoma, while the other tumors
are all CD40+ B cell lymphomas. The results in
Fig. 3 show that in each tumor the larger
the tumor inoculum the more effective the anti-CD40 mAb treatment.
This observation was particularly striking in experiments performed
with the A20 and EL4 tumors, showing that anti-CD40 mAb had no
appreciable activity against lower doses of tumor (e.g.,
105) but provided long-term protection against
most mice carrying the higher doses of tumor. In all these
immunotherapy trials mice given the largest i.v. inoculum of the tumors
plus anti-CD40 mAb were long-term survivors (>100 days without
sign of tumor). Interestingly, mice given lymphoma cells into the
peritoneal cavity or s.c. were less sensitive to this treatment even
when receiving an equivalent dose of tumor cells (data not shown). Thus
it appears that the anatomical site in which the tumor develops is very
important in these models. While it is not clear what brings about
these dramatic and somewhat counterintuitive results, it has been a
very consistent finding that larger tumor loads can be more effectively
treated, with the most effective setting being in those animals in
which the tumor cells represent 5–10% of splenic lymphocytes at
the time the mAb is given. The explanation may be provided if the CTL
response is directed at a scarce or weak tumor Ag that only reaches a
threshold detection level for recognition when the tumor is growing in
the lymphoid system at high doses.
|
|
In three B cell models we next followed changes in tumor and T
cells (CD4 and CD8) following anti-CD40 mAb treatment by flow
cytometry. This investigation was not possible in the case of the EL4
tumor, because it expresses the CD4 T cell marker and thus interfered
with such an analysis. We have shown previously that, at least in the
case of BCL1, anti-CD40 mAb treatment of
tumor-bearing mice promotes the rapid expansion of CD8 cells and does
not result in any detectable humoral response in the serum of
anti-CD40 mAb-"cured" mice as measured by flow cytometry.
Interestingly, we found no marked change in the growth rate of these B
cell tumors during the initial period following the anti-CD40 mAb
treatment despite the expression of CD40. Fig. 4A shows that for at least 4
days after the mAb the growth rate of the tumor in the control and the
anti-CD40 mAb-treated animals is essentially the same, and the
anti-CD40 mAb did not promote any marked change in the surface
phenotype of CD40+ lymphoma cells in vivo, with
the exception that CD40 became blocked and modulated by the treatment
(data not shown). It therefore appears that the therapeutic activity of
the mAb cannot be explained by a direct cytotoxic effect on the tumors.
The capacity of this treatment to deal with bulk tumor is underlined by
the fact that in these three models the tumor load can reach up to
5 x 108 cells/spleen before the therapeutic
activity takes over and the cell number starts to decline (Fig. 4A). At this point FACS analysis shows that up to half of
the splenic lymphocytes will be tumor. The results in Fig. 4B show that in all three B cell lymphoma models treatment
was followed by a sharp but short-lived rise in CD8 cells (Fig. 4B), starting 2 days after mAb treatment, which peaked at
4–6 days after treatment and then subsided over a period of between 24
and 48 h. In each case the decline in CD8 cells was preceded by an
equally rapid loss of splenic tumor cells (Fig. 4A).
Similar, but less exaggerated, changes were seen in CD4 T cells (Fig. 4C). This is consistent with T cell depletion studies which
show that removal of CD8, but not CD4, cells in these models with
anti-CD8 or anti-CD4 mAb ablates the therapeutic activity of
the anti-CD40 mAb (Ref. 16 and data not shown).
Interestingly, these studies also showed that when
BCL1 or A31 cells were inoculated without
anti-CD40 mAb (
) we consistently observed a small but
significant rise in CD8 cells, and although these responses were
obviously unable to control tumor development, they do indicate immune
recognition of these syngeneic tumors. Thus the current data support
the idea that the anti-CD40 mAb treatment boosts a weak, but
preexisting, CD8 tumor response that is achieved without the need for
CD4 Th cells.
|
. In each case the blocking mAb prevented the response
and the BCL1 tumor grew progressively. The
results in Fig. 5A show the
number of splenic tumor cells quantified each day following treatment
and blocking with anti-B7-1 and anti-B7-2. Interestingly,
blocking with either or both anti-B7-1 and anti-B7-2 mAb
completely prevented an effective response against the growing tumor
(Fig. 5A). This lack of response was reflected in the
failure of CD8+ T cells to expand in the normal
manner (Fig. 5B). In fact, when both anti-B7.1 and
anti-B7.2 mAb were administered, CD8+
cell numbers fell below the normal level. Fig. 5, C and
D, also shows that blocking mAb to LFA-1 and IFN- were
effective in inhibiting the therapeutic activity of anti-CD40 mAb.
Together these results point to the fact that CD8 T cells respond
against BCL1 using conventional adhesion and
coreceptor molecules and appear to generate a Th1 response. These same
mAb were also found to inhibit the activity of anti-CD40 mAb in
immunotherapy survival studies using BCL1 and A31
(data not shown). Interestingly, mAb against ICAM-1 and CTLA-4 failed
to block the CD40 mAb-induced responses either in the short-term
kinetic assays of the type shown in Fig. 5 or in long-term
immunotherapy.
|
Previous work has shown that anti-CD40 mAb can promote
therapeutic NK cell responses against murine tumors (32, 33). Three experimental conditions were investigated to clarify
the contribution of tumor cells and anti-CD40 to the NK cell
response: in the first group, mice were given just the
BCL1 cells (5 x 107)
on day 0; in the second, mice received anti-CD40 alone on day 4;
and in the third, mice received 5 x 107
BCL1 cells i.v. on day 0 and then anti-CD40
on day 4 when the splenic tumor load was 5%. The results in Fig. 6 show that in all three settings mice
responded with a wave of increased NK cell cytotoxic activity against
51Cr-labeled YAC targets which lasted 3 days
and peaked between 24 and 48 h after mAb treatment. Interestingly,
just giving the tumor cells alone was sufficient to promote some
increase in NK cell activity against YAC cells (Fig. 6A). We
do not have an explanation for this observation, but it was a
consistent finding and again points to immune recognition of neoplastic
cells. However, it is important to note that these mice did not develop
an appreciable cytotoxic response against BCL1
cells. In contrast, mice given either anti-CD40 mAb alone (Fig. 6B) or BCL1 and anti-CD40 mAb
(Fig. 6C) did make a small but consistent response against
BCL1 that peaked at 48 h after anti-CD40
mAb treatment. This early cytotoxic response to the
BCL1 was not mediated by CTL because it could not
be blocked by anti-CD8 mAb (Fig. 6C, open bars).
Finally, the major difference in mice treated with tumor and
anti-CD40 mAb (Fig. 6C) was that they made a
striking cytotoxic response against the BCL1
tumor at 96 h after the anti-CD40 mAb treatment. Furthermore,
this second burst of killing was important because, unlike the early
cytotoxic activity, it was blocked by anti-CD8 mAb and thus was
mediated by CD8+ CTL. These results show that
following tumor inoculation, with or without anti-CD40 mAb, mice
respond with a rapid burst of NK cell activity capable of killing YAC
cells but unable to control these syngeneic lymphomas. However, in
those mice given both tumor and anti-CD40 mAb, the NK cells are
followed at 4 days by CTL activity capable of controlling the
disease. This latter T cell killing also coincides with the increase in
CD8 T cells seen in Fig. 4
. Finally, we performed in vivo NK depletion
using anti-ASGM1 antiserum and confirmed that NK cells were not
required for the therapeutic activity of anti-CD40 mAb, at least in
the BCL1 model (data not shown). This depleting
Ab did not reduce the rapid rise in CD8 T cells seen in the spleens of
responding mice, although we did sometimes note both an acceleration of
tumor growth in anti-ASGM1-treated mice and a slight delay of 24
h in the initiation of the CTL response. Overall we can conclude that
NK cells are not an essential element in the anti-CD40 mAb-induced
response.
|
We next established that mice (BALB/c, CBA, or C57/Blk) cured of
syngeneic B or T cell lymphomas by anti-CD40 mAb treatment acquire
long-term immunity against rechallenge with the same tumors. The data
for all four models are summarized in Table I. Tumor-bearing mice were treated with
anti-CD40 mAb and then left for an extended period, 40 to >150
days, before i.v. rechallenge with 105 cells from
the same tumor that was used in the original therapy. All mice were
appreciably protected compared with naive controls; however, the data
clearly show that resistance to tumor, which was initially very high,
declines with time. For example, 67% of mice that had been treated for
the BCL1 lymphoma 40 days previously were fully
protected (>100 days) when rechallenged, while only 12% of mice
rechallenged between 80 and 149 days after initial treatment showed the
same degree of resistance. Interestingly, the EL4 tumor, which was very
dependent on tumor load when treated with anti-CD40 mAb and did not
respond at all to small tumor inocula (105) (Fig. 3), proved the most robust in terms of long-term memory. Thus all mice
(17 of 17) that survived for 60 and 79 days after first having the
tumor were completely resistant to rechallenge. Furthermore, even after
they had been left for >150 days, two-thirds were still able to resist
tumor rechallenge.
|
show very clearly that, in contrast to
CD8 depletion, depletion of CD4 cells before rechallenge did not
significantly diminish the level of protection. Hence these data make a
strong case for a CD8 T cell memory that provides long-term protection
in the absence of a CD4 response.
|
Immune splenocytes taken from mice during their response to
BCL1 are cytotoxic for this tumor and for A20
(16). In these experiments we inoculated mice with 5
x 107 BCL1 on day 0 and
then treated them on day 4 with 1 mg of anti-CD40 mAb. On day 8 or
9, when the tumor was cleared but while the effector
CD8+ T cells were still high, splenocytes were
harvested and assayed for cytotoxic activity against a range of BALB/c
tumor targets. In addition, a similar group of mice were left until
days 35–50 to measure the IFN- response to the BALB/c tumors. The
two assays could not be conducted simultaneously due to the high
background levels of IFN- production during the early
post-anti-CD40 mAb period. Fig. 8
shows the results of such an experiment. As expected, the
BCL1 tumor target itself initiated good killing
and IFN- responses; however, similar responses were also stimulated
by a second B cell tumor, A20. The level of IFN- production
stimulated by the A20 in these experiments was somewhat variable and in
the example shown did not result in a detectable increase (Fig. 8B). Overall these data indicated considerable
cross-reactivity between A20 and BCL1.
Interestingly, other available BALB/c tumors were not recognized in
these assays. Blocking studies with anti-MHC class I, -MHC class
II, -CD4, and -CD8 mAb confirmed that the killing responses were
mediated by CD8 T cells within the immune splenocytes (Fig. 8C). Finally, to confirm that this in vitro
cross-recognition extended to cross-protection in vivo, we next
performed rechallenge immunotherapy with BCL1 and
A20. Groups of mice were inoculated with either
BCL1 or A20 (5 x 107
cells/mouse) and treated with anti-CD40 mAb in the normal way to
cure the tumor. After 40–71 days they were rechallenged i.v. with
105 cells from either the same or the alternative
BALB/c tumor and then monitored for disease development. The results in
Fig. 9 clearly show that, as with the in
vitro assay, these two tumors provide almost complete cross-protection,
confirming the recognition of common CD8+ T
cell Ags.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
10% lymphoma in
their splenic lymphocytes.
We found no evidence of changes in the BCL1
lymphoma phenotype following treatment, nor did we see class switching,
which has been reported following anti-CD40 mAb treatment
(38). The immune responses were primarily restricted to
CD8+ cells and required 6 days to peak after
which the numbers of CD8+ cells declined rapidly.
During the initial priming period, the lymphoma cells grew at the same
rate in control and anti-CD40 mAb-treated mice, showing that the Ab
did not markedly influence tumor growth despite the presence of CD40 on
the B cell lymphomas. In the BCL1 lymphoma, the
anti-CD40 mAb-induced CD8 T cell response required coreceptor
engagement, including B7-1, B7-2, and LFA-1, and also needed the
production of IFN-, because blocking mAb to all these molecules
effectively prevented the therapy (Fig. 5). It is likely that responses
to the other lymphoma models have similar requirements. The blocking
activity of both anti-B7-1 and -B7-2 mAb was somewhat surprising,
especially in view of the redundant role these two molecules appear to
play in T cell responses. One possible explanation is that, despite
performing the same costimulatory function, both are required to
achieve an effective response against very weak tumor Ags. Thus
blocking either B7-1 or B7-2 with mAb is enough to reduce the levels of
costimulation below a minimum threshold required for the response. A
second possible explanation is that mAb, including anti-B7-1 and
anti-B7-2 (BCL1 and A31 lymphoma express
these two Ags), that reacts with the lymphoma cells will perturb the T
cell responses, for example, simply by altering the way in which tumor
cells are processed by APC or the way the tumor cells present their
own Ags.
Immune splenocytes taken from responding mice were cytotoxic toward the
tumor in vitro. This activity was provided mainly by CD8 T cells and
was achieved without further stimulation in culture. The
BCL1-reactive T cells also recognized A20 and
provided cross-protection, which allowed mice cured of
BCL1 or A20 to reject both these tumors when
rechallenged. Interestingly, such cross-recognition did not extend to a
range of other, nonlymphoid, BALB/c tumors, suggesting that the
recognized Ag(s) was restricted to lymphocytes. Studies are under way
to define these Ags. Finally, treatment of BCL1
with anti-CD40 mAb, and even the BCL1
lymphoma itself, promoted a wave of NK cell activity as measured by
increased killing of YAC cells. However, we found no evidence that this
played an important role in preventing lymphoma development or in
curing the mice, because NK cell depletion using a polyclonal Ab
(anti-ASGM1) did not block the efficacy of the CD40 mAb. Others
have found CD40 stimulation with mAb or CD40L can up-regulate NK cell
cytotoxic activity (32, 33), and that, at least in some
models, these can account for a major component of the therapeutic
activity. The most likely explanation for such activity is through
CD40-promoted production of Th1 cytokines such as IL-12 and
IFN-.
It is now well accepted that anti-CD40 mAb can act as a powerful adjuvant to promote vaccination. For example, CD40 ligation has been shown to promote CTL responses to peptide and protein Ags with known epitopes (12, 13), and functioned therapeutically against tumors where the targets were not defined (16, 18). It appears that in most situations these are Th1-skewed responses and that they occur without CD4 T cell help. Lefrancois et al. (13) used an OT-1 adoptive transfer system to show that agonistic anti-CD40 mAb was highly effective at promoting expansion and cytotoxic activity in CD8+ OT-1 cells when immunized with poorly immunogenic soluble OVA. The response was CD4 T cell-independent and required expression of CD40 on non-T cells, which, although not defined, were probably APC. Likewise, Diehl et al. (12) have shown that a tolerogenic response to a tumor-derived peptide could be converted to an immunogenic response simply by including anti-CD40 mAb. Again this work points to the APC, probably DC, as the target for the anti-CD40 mAb, in that its agonistic effects were still fully functional in B cell-deficient mice.
The effect of CD40 stimulation on APC, such as DC, has been extensively
studied, showing that it acts by up-regulating costimulatory molecules
and proinflammatory cytokines, thus equipping them to prime
naive CTL (10). During normal cellular responses it
appears that virus infection of APC or interaction of APC with CD40L
(CD154)-expressing CD4 Th cells are two mechanisms that equip APC to
prime CTL (39). Thus CD4 Th cells have the potential to
initiate an immune response not only through the production of
particular types of cytokine but also by way of activating signals
delivered to APC. A number of groups (7, 8, 9) have
shown that anti-CD40 mAb or soluble trivalent CD154 can substitute
for CD4 help to produce APC capable of presenting Ag effectively to
naive T cells and thereby provide a similar stimulatory function to
cognate CD4 helper cells. We propose that similar changes in APC are
occurring when anti-CD40 mAb is used to treat splenic lymphoma in
vivo. In all three B cell lymphoma models (BCL1,
A31, and A20) presented in this work we found that the immune responses
generated by CD40 mAb treatment were dependent on CD8 CTL and bypassed
the need for CD4 T cell. At least in the BCL1
model, it appears that CD4 cells are also not required in the effector
phase of the response because immune animals were still able to resist
tumor rechallenge even following CD4 T cell depletion (Fig. 7). Because
the tumor cells express surface CD4, the EL-4 T cell lymphoma could not
be investigated with anti-CD4-depleting mAb in these studies. Thus
it appears that CD40-reactive mAb is able to replace the need for a CD4
T cell response to prime CTL, as might occur in conventional antitumor
CD8 T cell response. Such results point to a lack of endogenous CD4
tumor epitopes in these tumors and are consistent with the view that
antitumor CTL responses are often thwarted due to the lack of suitable
recognition by CD4 T cells. For example, Surman et al.
(40) found that by adoptively transferring
CD4+ T cells specific for a model tumor Ag they
could promote the production of endogenous CD8+
CTL in the tumor-bearing mice and that these CTL were specific for the
same model Ag. In addition, activation of APC provides the potential to
cross-prime tumor Ags from the lymphomas into the MHC class I pathway
of the APC. Although previously it was felt that Ags entering APC from
outside the cell would only enter the MHC class II pathway, it is now
clear that under certain conditions exogenous Ags can be delivered into
the class I pathway for presentation into the CTL. The existence
and importance of such a system with tumor Ags in vivo is still
controversial, but it would provide a satisfactory explanation for the
CTL responses seen with CD40- tumors, such as
EL-4, in this work and against solid tumors described previously
(18).
Growing evidence shows that the anatomical site in which tumors develop
is very important for their immune recognition, with tumors that
develop in the lymphatics being more inclined to induce immunity
compared with those growing outside the immune system, such as under
the skin (41). Recent work from Ochsenbein et al.
(41) has shown for a range of tumors that the immune
system is more likely to respond effectively in cases where the tumor
cells "seed" into lymph nodes or grow in the spleen, while those
tumors that remain confined to a s.c. site are often ignored by the
immune system. Even immunogenic tumors, such as those transfected with
viral genes, when grown s.c. promote quite weak responses. All four of
the lymphomas used in this study are considered nonimmunogenic or only
weakly immunogenic and would not be expected to stimulate ablative T
cell responses. This is particularly the case for
BCL1 and A31, which cannot be grown in vitro and
have to be maintained by in vivo passage. However, in each case the
tumors grow, at least initially after inoculation, as dispersed cells
in the spleens surrounded by the immune system. Under such
circumstances it appears that even the slightest immune response is
sufficient for effective boosting by anti-CD40 mAb. The best
evidence we have of immune recognition comes from the observation that
T cell numbers, especially CD8+ cells, rise
slightly after tumor injection (Fig. 4) and NK cell cytotoxic activity
is promoted by a large i.v. inoculum of BCL1
(Fig. 6). Although it is not possible to grow these lymphomas at remote
sites without their rapid migration to the spleen, we have recently
found that anti-CD40 mAb treatment of large quantities of
BCL1 in the peritoneal cavity is less effective
than that seen against splenic tumors. This result is consistent with
the idea that anti-CD40 mAb-induced immunity is more effective in
situations where the immune system is in close proximity to the
neoplastic cells. Likewise, we find that anti-CD40 mAb is
relatively ineffective against solid tumors grown s.c.
(18). Recent results from this laboratory showed that
anti-CD40 mAb had only a modest therapeutic effect on the
colorectal carcinoma, CT26, grown s.c. and only in the case of a highly
immunogenic carcinoma, CMT93, did it demonstrate appreciable
effectiveness. Thus it appears that with highly immunogenic tumors
anti-CD40 mAb will boost the immune response regardless of
anatomical site, but with nonimmunogenic tumors such responses are seen
only when the tumor is surrounded by the immune system.
Such a suggestion would help explain why we only see good tumor
protection when treatment is initiated against a large tumor dose, this
being achieved either with a large inoculum or by allowing sufficient
time for in vivo tumor development before treatment. This would be
consistent with a situation where the Ag load was limiting. The
cross-reactive nature of the protection between
BCL1 and A20, the two BALB/c lymphomas, is
consistent with a limited antigenic expression. We have also found in
both these tumors that the T cell repertoire in mice cured of tumors by
anti-CD40 mAb is skewed in their TCR V
family usage (G.
Crowther, C. Patrick, P. Johnson, R. French, and M. Glennie, manuscript
in preparation), consistent with recognition of a limited number
of common tumor (lymphoma) Ags.
The current findings have important implications for human immunotherapy, showing that strong CD8 CTL responses can be generated without excessive toxicity. A major advantage with such a strategy is that the responses are potentially directed at multiple tumor Ags and are therefore less likely than single target Ags to suffer escape variants, and they can be induced without identifying the target Ags. Recently, Vonderheide et al. (19) have shown interesting results in patients using a soluble CD154 fusion protein which in phase I trials showed acceptable toxicity and some encouraging clinical responses. The question of whether it will be more effective to use a human (or humanized) IgG anti-CD40 and run the risk of generating anti-Id responses or to use a soluble recombinant ligand, such as a CD154 trimer, remains to be seen. In this respect recently work has shown that a dodecamer of CD154 is far more active than trimeric CD154 at activating cells via CD40 (42).
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Martin J. Glennie, Tenovus Research Laboratory, Cancer Sciences Division, University School of Medicine, Southampton General Hospital, Southampton SO16 6YD, U.K. E-mail address: mjg{at}soton.ac.uk
3 Abbreviations used in this paper: DC, dendritic cell; CD40L, CD40 ligand; anti-asialo GM1; anti-ASGM1.
Received for publication October 11, 2001. Accepted for publication January 11, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
/ x anti-CD2) F(ab')2 antibody. J. Immunol. 155:2960.[Abstract]
BCL1. Cancer Biother. Radiopharm. 15:571.[Medline]
This article has been cited by other articles:
|
|
 |
|
  C. L-L. Chiang, J. A. Ledermann, E. Aitkens, E. Benjamin, D. R. Katz, and B. M. Chain Oxidation of Ovarian Epithelial Cancer Cells by Hypochlorous Acid Enhances Immunogenicity and Stimulates T Cells that Recognize Autologous Primary Tumor Clin. Cancer Res., August 1, 2008; 14(15): 4898 - 4907. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. R. French, V. Y. Taraban, G. R. Crowther, T. F. Rowley, J. C. Gray, P. W. Johnson, A. L. Tutt, A. Al-Shamkhani, and M. J. Glennie Eradication of lymphoma by CD8 T cells following anti-CD40 monoclonal antibody therapy is critically dependent on CD27 costimulation Blood, June 1, 2007; 109(11): 4810 - 4815. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Illidge Anti-CD40: Janus or gatekeeper? Blood, October 15, 2005; 106(8): 2595 - 2596. [Full Text] [PDF]
|
|
|
|
|
|
M.-G. de Goer de Herve, D. Durali, T.-A. Tran, G. Maigne, F. Simonetta, P. Leclerc, J.-F. Delfraissy, and Y. Taoufik Differential effect of agonistic anti-CD40 on human mature and immature dendritic cells: the Janus face of anti-CD40 Blood, October 15, 2005; 106(8): 2806 - 2814. [Abstract] [Full Text] [PDF]
|
|
|
|
) or anti-CD40 mAb (anti-CD40,
). The symbols and error bars indicate the mean and SD for the five
mice in each group.
), or
control IgG (
) mAb, on the days indicated (arrows). From day 3 or 4,
duplicate spleens were harvested and tumor cells, CD4 cells, and CD8
cells were monitored by flow cytometry as described in Materials
and Methods. Data represent one of three similar
experiments.
), or anti-CD4 plus
anti-CD8 (
).
