| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Cellular Cytotoxicity Laboratory, The Austin Research Institute, Heidelberg, Victoria, Australia; and
Rotary Bone Marrow Research Laboratory, Royal Melbourne Hospital, Parkville, Victoria, Australia
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
receptor I 
-chain fusion (scFv anti-CEA) receptor and
demonstrated high-level expression of this chimeric receptor in naive
mouse T lymphocytes by retroviral gene transduction. These
gene-modified CTL were able to lyse CEA+ targets and
secrete high levels of IFN- following Ag stimulation. Depletion
studies demonstrated that specific tumor cell cytotoxicity was mediated
by gene-modified CD8+ T cells. Importantly, in increasingly
stringent tests of efficacy in vivo, transduced CTL were sequentially
shown to reject CEA+ colon carcinoma cells in a Winn assay
and then reject established s.c. colon carcinoma in scid
or syngeneic mice. Furthermore, using gene-targeted and scFv
anti-CEA receptor-transduced donor CTL, perforin and IFN- were
demonstrated to be absolutely critical for the eradication of colon
carcinoma in mice. In summary, we have developed a highly efficient
gene transfer system for evaluating chimeric receptor expression in
cytotoxic lymphocytes. This series of experiments has revealed the
utility of scFv anti-CEA chimeras in providing mouse T cells the
capacity to reject colon carcinoma in an Ag- and perforin-specific
manner. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To address these problems, one alternative approach explored in recent years has been to genetically manipulate CTL ex vivo to express receptors for tumor-associated Ags (TAA)3 that can activate CTL effector function upon Ag ligation (3, 4, 5, 6, 7). This alternative approach has several advantages over strategies relying strictly upon cellular or humoral immunity. First, CTL and other effector lymphocytes can be redirected to a variety of tumor types, particularly those for which very few specific CTL clones have ever been isolated. Second, gene-engineered T cells kill tumor cells in a non-MHC-restricted and tumor Ag-specific manner, and thus this approach is not compromised, like others, by tumor loss of MHC alleles. Third, CTL grafted with chimeric receptors can be targeted not only to protein TAA, but also to tumor-associated gangliosides and carbohydrates (8).
Several single chain variable domain (scFv) receptors reactive to TAA
on ovarian, breast, and colon carcinoma have been functionally
expressed in mouse T cell lines such as MD45 and in both mouse and
human lymphocytes (4, 5, 6, 7, 8, 9, 10). Ag-specific activation via
these chimeric scFv receptors resulted in cytokine release and specific
lysis of tumor cells in mice (4, 6, 11). Despite these
advances, progress toward understanding the important features and
limitations of this approach have been constrained by the lack of a
robust system for efficiently expressing chimeric genes in primary
lymphocytes. In this study, we have developed such a robust system and
used it to demonstrate the efficacy of a scFv- chimeric receptor
recognizing carcinoembryonic Ag (CEA) to redirect T cell-mediated
rejection of colon carcinoma. The general utility and effectiveness of
this system of gene transduction was highlighted by the efficacy of
redirected T cells to reject tumor in a variety models requiring
stringent CTL/tumor localisation, specificity, and potency.
Furthermore, transduction of T lymphocytes from various gene-targeted
mice enabled us to begin to dissect which effector and inflammatory
cytokine molecules are critical, such as perforin and IFN-,
respectively, in achieving tumor eradication. This protocol and these
experiments provide the basis on which to rationally test and improve
gene-modified lymphocyte therapies of cancer.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The human colorectal carcinoma cell lines COLO 205 and Lovo, mouse (C57BL/6) colon adenocarcinoma MC-38 and its CEA+ transfectant MC-38-CEA2 (12) (kindly provided by Dr. Jeff Schlom, National Institutes of Health, Bethesda, MD), and the C57BL/6 sarcoma cell line 24JK (kindly provided by Dr. Patrick Hwu, National Institutes of Health) were maintained in RPMI 1640 or DMEM at 37°C and 5% CO2 supplemented with the following additives: 10% (v/v) FCS, 2 mM glutamine, 100 U/ml penicillin, and 100 µg/ml streptomycin (Life Technologies, Grand Island, NY). The retroviral packaging cell lines GP+E86 and PA317 and the fibroblast cell line NIH3T3 were cultured in DMEM with additives. PA317 cells transduced with recombinant retroviral DNA were maintained in DMEM supplemented with 0.5 mg/ml G418 (Life Technologies). Transduced T cells were cultured in DMEM containing 100 U/ml of human recombinant IL-2 (kindly provided by Chiron, Emeryville, CA).
Mice
Inbred C57BL/6 (B6), BALB/c, and BALB/c scid/scid
(scid) mice were purchased from The Walter and Eliza Hall
Institute of Medical Research, Melbourne, Australia. C57BL/6
gld (Fas ligand (FasL) mutant) and C57BL/6 TNF-deficient
(B6.TNFo) mice (13) were obtained
from the Centenary Institute of Cancer Medicine and Cell Biology
(Sydney, Australia). C57BL/6 IFN--deficient
(B6.IFN-o) (14) mice were
obtained from Genentech (South San Francisco, CA). These gene targeted
mice and C57BL/6 perforin-deficient (B6.Po) mice
(15) were bred at the Austin Research Institute Biological
Research Laboratories. Mice of 4–8 wk of age were used in all
experiments that were performed according to animal experimental ethics
committee guidelines.
Chimeric receptor gene construction
A 767-bp fragment of DNA coding for scFv of anti-CEA and a
marker epitope from c-myc was amplified by PCR from the
vector MFE-23 (16) and subcloned into
XbaI/BstEII-digested pRSVscFvR (a kind gift
from Zelig Eshhar, Weizmann Institute, Rehovot, Israel). The scFv
anti-CEA construct is composed of the VH and
VL regions of the anti-CEA mAb joined via a
flexible linker; a membrane proximal hinge region of human CD8 and the
transmembrane and cytoplasmic regions of the human FcR chain. The
scFv anti-CEA chimeric receptor was then digested with
SnaB1/XhoI and subcloned into the
HpaI/XhoI restriction sites of the retroviral
vector pLXSN (a kind gift from Dusty Miller, Seattle, WA) containing
the murine Moloney leukemia virus long repeat and a neomycin resistance
gene under the control of an SV40 promoter.
Retroviral gene transfer of primary mouse T lymphocytes
The LXSN retroviral vectors (plus the anti-CEA scFv
receptor) were transfected into the amphotrophic PA317 packaging cell
line by CaPO4- precipitation.
The supernatants were then collected and used to infect the ecotrophic
packaging cell line GP+E86 for 24 h in the presence of 4 µg/ml
polybrene (Sigma, St. Louis, MO). Stable ecotrophic packaging cell
lines were obtained after G418 selection. High receptor-expressing
clones were isolated by RNA Slot-blot analysis as described previously
(17), and viral titer was determined on the basis of
neomycin resistance of infected NIH3T3 cells. GP+E86 clones producing
107 cfu/ml were used for transduction of mouse
naive T lymphocytes. Spleen cells from mice were initially depleted of
RBC by hypotonic lysis with NH4Cl and enriched by
passing through a nylon wool syringe as described previously
(18). Enriched T lymphocytes (2 x
106) were then cocultivated for 72 h with
3 x 105 viral-producing packaging cells in
DMEM supplemented with 4 µg/ml polybrene, 5 µg/ml PHA (Sigma), and
100 U/ml IL-2. Following cocultivation, T cells were separated from
adherent packaging cells, washed with DMEM, and cultured in DMEM
supplemented with 100 U/ml rIL-2. T cells were then analyzed for
transduction efficiency by flow cytometry and used for in vitro assays
and in vivo experiments. In some experiments,
CD8+ cells were depleted (to <2%
CD8+) from T cells by immunomagnetic beads
(anti-Lyt2; Dynal Australia, Carlton, Australia).
Flow cytometry
Detection of cell-surface chimeric receptor expression on mouse T lymphocytes was achieved by indirect immunofluorescence with a c-myc tag Ab purified from supernatants of mouse 9E10 cells (19), followed by staining with a PE-labeled anti-mouse Ig mAb (Becton Dickinson, San Jose, CA). Cell-surface phenotyping of transduced cells was determined by direct staining with quantum-red-labeled anti-CD4 (clone H129-19; Sigma) and anti-CD8 (clone 53-6.7; Sigma) and PE-labeled anti-TCRß (clone H57-597, PharMingen, San Diego, CA) mAbs. Briefly, 5 x 105 transduced T cells were incubated with 100 µl of appropriately diluted Ab at 4°C for 30 min, washed twice in DMEM, and analyzed for Ab binding by flow cytometry using a FACScan (Becton Dickinson).
Ag-specific binding, cytotoxicity, and IFN- secretion
The binding capacity of gene-modified mouse T lymphocytes was
determined in a rosetting assay as described (10). The
cytotoxicity capacity of transduced T cells was determined in a 6-h
51Cr release assay as described (9).
Mouse IFN- secretion by scFv-modified mouse T lymphocytes upon CEA
ligation was detected by ELISA. Mouse T cells (transduced with LXSN
alone or LXSN plus anti-CEA scFv receptor) were cultured with
either 5 x 105 Lovo
(CEA+) or 24JK (CEA-)
sarcoma cells for 24 h in 24-well plates at a 1:1 E:T ratio.
Following incubation, supernatants were harvested and spun to remove
cell debris, and levels of IFN- were measured by ELISA (PharMingen)
according to the suppliers specifications.
Adoptive transfer
Two different adoptive transfer models were employed. First,
mouse 24JK sarcoma cells (106) and/or human COLO
205 colon carcinoma cells (5 x 106) were
injected s.c. into opposite flanks of groups of five scid
mice. Spleen T lymphocytes (5 x 106) from
B6, B6.Po, B6.gld,
B6.IFN-o, or B6.TNFo
mice (transduced with LXSN vector alone or LXSN plus scFv anti-CEA)
were injected i.v. groups of five scid mice at 6 h (day
0), days 0 and 1, or day 3 after tumor inoculation. Second, mouse MC-38
colon adenocarcinoma cells (5 x 106) and/or
their CEA+ transfectants, MC-38-CEA2 (5 x
106) were injected s.c. into opposite flanks of
groups of five B6 mice. Spleen T lymphocytes (5 x
106) from B6 mice (transduced with LXSN vector
alone or LXSN plus scFv anti-CEA) were injected i.v. into groups of
5 B6 mice at 6 h (day 0), days 0 and 1, or day 3 after tumor
inoculation. In both models above, subsequent tumor growth was
monitored daily and measured by a caliper square along the
perpendicular axes of the tumors. The data were recorded as the mean
tumor size (product of the two perpendicular diameters) ±
SEM.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The scFv anti-CEA chimeric receptor gene was composed of the
VH and VL regions of the
anti-CEA mAb joined via a flexible linker; a membrane proximal
hinge region of human CD8 and the transmembrane and cytoplasmic regions
of the human FcR chain. It was subcloned into the retroviral vector
pLXSN and high titer virus-producing GP+E86 clones were used to
transduce enriched naive mouse T cells from BALB/c mice as described in
Materials and Methods. Viral titers of GP+E86 clones ranged
from 5 x 106 to 2 x
107 cfu/ml determined after infection of NIH3T3
fibroblast cells, and thus subsequent T cell infections were performed
using the highest virus-producing clone. Naive mouse T cells were
transduced by coculturing with virus in the presence of activating
agents PHA and IL-2 and polybrene for 72 h. Following
cocultivation, expression of the scFv anti-CEA receptor on the
surface of the T cells was determined by flow cytometry using an
anti-c-myc tag mAb. High-level expression of the scFv
anti-CEA receptor (between 40 and 70% for many infections) was
detected on T cells (59 ± 10%, n = 10) (Fig. 1
A). Expression of the scFv
receptor could not be detected on T cells transduced with the LXSN
retroviral vector alone (Fig. 1B). Consistent with
preferential proliferation of CD8+ spleen T cells
in response to PHA/IL-2, the transduced T cell population consisted of
predominantly
TCRß+CD8high and
TCRß+CD4-CD8low/-
T cells (mean CD8high T cells 86 ± 11%,
n = 10).
|
In all in vitro and in vivo assays where the specificity of mouse
T cells expressing scFv anti-CEA receptor against human
CEA-expressing tumor cells was examined, a suitable control human
CEA- tumor (i.e., that was nonadherent and grew
well in scid mice) was not available. Consequently, we used
CEA- mouse 24JK sarcoma cells with the rationale
that, if anything, they would be more likely to interact
non-specifically with mouse effector CTL than a human tumor target. The
ability of mouse T cells expressing scFv anti-CEA to specifically
bind to CEA-expressing target cells was first demonstrated in a
rosetting assay. Transduced BALB/c T cells conjugated with the
CEA+ colon carcinoma cell line Lovo (44 ±
2%, n = 3) (Fig. 1
D), but not the
CEA- sarcoma cell line 24JK (Fig. 1F). T cells transduced with the control LXSN vector were
unable to conjugate to either cell line (Fig. 1, C and
E). The ability of transduced mouse BALB/c T cells to
secrete IFN- was demonstrated after specific interaction with
CEA+ Lovo target cells (1062 ± 15 pg/ml,
n = 3), but not CEA- 24JK
sarcoma cells (<20 pg/ml). T lymphocytes transduced with LXSN vector
alone secreted < 20 pg/ml IFN- after interaction with Lovo or
24JK cells.
We then evaluated the ability of transduced T cells to mediate specific
target cell lysis in a 6-h 51Cr release assay.
BALB/c T lymphocytes expressing the scFv anti-CEA receptor were
able to lyse the CEA+ colon carcinoma cells Lovo
and COLO 205, but not the CEA- 24JK sarcoma cell
line (Fig. 2A). T cells
transduced with the LXSN vector were unable to lyse Lovo, COLO 205, or
24JK cells (Fig. 2A). Lysis of both COLO 205 and Lovo (data not shown)
was CEA specific, because intact anti-CEA mAb, but not an isotype
control mAb, inhibited lysis by gene-modified T cells (Fig. 2B). Moreover, scFv-redirected T cell-mediated lysis was
delivered by the CD8+ T cell population, because
depletion of CD8+ cells with magnetic beads
completely abrogated lysis of both COLO 205 cells (Fig. 2B)
and Lovo cells (data not shown). B6 T lymphocytes that expressed the
scFv anti-CEA receptor equivalently to BALB/c T cells (Fig. 3A) were next examined in a
syngeneic setting. These effectors were able to lyse the
CEA+ MC-38-CEA2 mouse colon adenocarcinoma cells,
but not the CEA- parental MC-38 cells (Fig. 2C). T cells transduced with the LXSN vector were unable to
lyse MC38-CEA2 or MC-38 cells (Fig. 2C).
|
|
CTL-mediated lysis occurs via two major pathways: either involving
granule exocytosis (perforin and granzymes) or a FasL/Fas interaction
(20). We have previously demonstrated that CTL cell lines
engineered with scFv can be activated to lyse colon carcinoma cells in
a FasL-dependent manner (9). To determine what mechanism
of lysis is used by spleen T cells engineered with a scFv chimera,
enriched spleen T cells from B6, B6.Po, and
B6.gld (FasL mutant) were transduced with scFv anti-CEA
receptor. Flow cytometry determined a high level of receptor expression
(40–68% between infections) in T lymphocytes from B6 (49 ± 4%,
n = 5), B6.gld (45 ± 4%,
n = 5), and B6.Po (40 ±
3%, n = 5) (Fig. 3, A, C, and
E). Transduced T cells expressing scFv anti-CEA receptor
from all strains of mice consisted of similar ratios of
TCRß+ CD8high and
TCRß+CD4-CD8low/-
cells, were demonstrated to equivalently bind to
CEA+ colon carcinoma cell lines, and comparably
secreted IFN- after specific Ag stimulation (data not shown).
Importantly, transduced T cells from both B6 and B6.gld, but
not B6.Po, mice were able to specifically lyse
both CEA+ targets, Lovo and COLO 205, in a 6-h
51Cr release assay (Fig. 3, B,
D, and F). By contrast, transduced T cells were
unable to lyse the CEA- sarcoma cell line, 24JK.
Collectively, these results demonstrate specific Ag-dependent lysis of
colon carcinoma cells and that perforin is absolutely critical for
lysis by redirected T lymphocytes and triggered by ligation via
chimeric scFv anti-CEA.
Efficacy of redirected T cells after adoptive transfer into scid mice bearing tumors
The simplest in vivo assay to determine the ability of transduced
mouse T cells to control tumor growth is a Winn assay in
scid mice (21). B6 T cells transduced with scFv
anti-CEA receptor rejected the growth of COLO 205 tumor cells, but
not 24JK cells, at an E:T ratio of 1:1 in all mice (data not shown).
Preincubation of COLO 205 cells with pLXSN-transduced T cells did not
inhibit tumor cell growth. To further test the efficacy of redirected T
cells in a more stringent model requiring T cell localization to the
site of tumor challenge, transduced T cells were adoptively transferred
i.v. into scid mice with established s.c. colon carcinoma.
Transduced B6 T cells were injected at either 6 h (day 0), days 0
and 1, or on day 3 after COLO 205 tumor inoculation. COLO 205 tumors
were measurable within 2 days of tumor inoculation. Adoptive transfer
of T cells transduced with LXSN vector alone had no effect on COLO 205
tumor growth (Fig. 4). By contrast, mice
treated on days 0 and 1 with T cells expressing scFv anti-CEA
receptor completely rejected COLO 205 tumor growth in two of five mice
and partially inhibited tumor growth in the remaining three mice (Fig. 4). In addition, COLO 205 tumor growth was temporarily delayed (for 3
days) in groups of mice receiving T cells expressing scFv anti-CEA
receptor on day 0 or day 3 alone. Escaping tumors still retained
significantly high levels of CEA expression (data not shown). These
initial data suggested that multiple i.v. injections or larger numbers
of gene-modified CTL would be required to completely regress
established tumors.
|
To test whether the granule exocytosis or FasL pathway was
critical for tumor rejection in vivo, transduced
B6.Po and B6.gld T cells were
transferred. Adoptively transferred B6.Po T cells
expressing scFv anti-CEA receptor did not inhibit the growth of
s.c. COLO 205 tumors, indicating that perforin was critical for
effective tumor rejection (Fig. 5). T
cells from B6.gld mice that were transduced with scFv
anti-CEA receptor inhibited the growth of COLO 205 tumors to a
similar degree to transduced B6 T cells. These data suggested that FasL
was not an important CTL effector molecule in the control of this colon
carcinoma (in support of the in vitro data), despite the fact that in
vitro ligation of Fas can mediate the death of COLO 205 tumor cells
(P. K. Darcy, unpublished data).
|
in control of colon carcinoma in
scid mice
To examine whether effector cell inflammatory mediators were also
important in redirected T cell-mediated rejection of colon carcinoma, T
cells from B6.IFN-o and
B6.TNFo mice were transduced with scFv
anti-CEA receptor. High-level expression of the scFv anti-CEA
receptor was observed in B6. IFN-o (42 ±
3%, n = 5) and B6.TNFo (45
± 2%, n = 5) as demonstrated in B6 T cells.
Transduced B6.TNFo T cells inhibited the growth
of COLO 205 tumors to a similar degree to transduced B6 T cells. These
data suggested that effector cell TNF was not important in the control
of this colon carcinoma. By contrast, a notable observation supported
by a repeat experiment (data not shown, also n = 5) was
that despite early treatment and reduction in tumor growth,
scFv-transduced T cells from B6.IFN-o mice
were unable to cause tumor eradication. These data suggested that
effector cell IFN- may play a minor but significant role in the
control of colon carcinoma (Fig. 5).
Ag-specific control of colon carcinoma in scid mice
To demonstrate Ag-specific tumor eradication, additional
experiments were performed with mice inoculated with
CEA+ COLO 205 tumor cells in the right flank and
CEA- 24JK tumor cells in the left flank. T cells
expressing scFv anti-CEA receptor specifically reduced COLO 205
tumor growth (two of five eradicated) (Fig. 6A), while 24JK grew
unaffected (Fig. 6B). Control non-transduced, but activated,
T cells did not reduce COLO 205 or 24JK tumor growth. Nevertheless,
perforin-dependent eradication of some CEA+
tumors in a s.c. site upon early treatment was impressive evidence that
the transduced CTL could localize to the site of tumor inoculation.
|
Two additional experiments were performed to assess the possible
limitations of efficacy of T cells expressing scFv anti-CEA
receptor. In the first, multiple adoptive transfers were provided in an
attempt to eradicate all COLO 205 tumors. Four sequential transfers of
redirected T cells enabled four of five mice to survive tumor free;
however, this was not a large improvement over two sequential transfers
producing three tumor-free mice in a group of five (Fig. 7A). Due to the nature of i.v.
injection, the maximum number of sequential T cell transfers
practically achieved was four (from day 0 to 3). In a second
experiment, three sequential injections of 5 x
106 CEA-specific T cells were compared with one
single bolus dose of 1.5 x 107 T cells
(Fig. 7B). Not surprisingly, a larger dose of cells earlier
(6 h after tumor inoculation) was more effective than the same total
dose delivered over three successive days (6 h, day 1, day 2). In all
experiments to date, the indications have been that greater tumor
inhibition can be achieved with earlier and larger T cell transfers.
These data are consistent with a previous study using T cells
transgenic for a scFv (22), where the prolonged survival
of the experimental group was proportional to the number of
anti-tumor T cells given.
|
To further test the efficacy of redirected mouse T cells in a more
immunologically stringent model, transduced T cells were adoptively
transferred i.v. into syngeneic B6 mice with established s.c. mouse
colon carcinoma. Ag-specific efficacy was again examined by inoculating
mice with CEA+ MC-38-CEA2 tumor cells in the
right flank and CEA- MC-38 tumor cells in the
left flank. B6 T cells expressing anti-CEA receptor administered on
days 0 and 1, or as late as day 2, specifically inhibited MC-38-CEA2
tumor growth (two of five eradicated) (Fig. 8A), while the neighboring
MC-38 tumor grew unaffected (Fig. 8B). Control LXSN vector
alone transduced T cells did not reduce MC-38-CEA2 or MC-38 tumor
growth (Fig. 8, A and B).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We have demonstrated the efficacy of the scFv anti-CEA- receptor
to redirect CTL to CEA-expressing colon carcinoma in three settings in
vivo. From the Winn assay to targeting established s.c. syngeneic colon
carcinoma, each model was increasingly stringent and biologically
relevant. Indeed, the efficacy of scFv-engineered and -transferred T
cells in our study has demonstrated that these effector T cells can
even eradicate colon carcinomas that grow quite rapidly in a syngeneic
host. Furthermore, there was no requirement for the addition of
exogenous IL-2. In the adoptive transfer models (in scid and
syngeneic mice), the specificity of the redirected CTL was demonstrated
by the rejection of CEA+ colon carcinoma and the
normal tumor growth of CEA- tumor inoculated on
the opposite flank. This is the first study to demonstrate in this
fashion the targeted specificity of CTL-expressing scFv. Further
specificity controls included the inability of either T cells activated
with IL-2/PHA (as for infection) or T cells infected with the LXSN
retroviral vector alone to affect tumor growth. In all models
evaluated, CTL-expressing scFv anti-CEA rejected colon carcinoma in
a CEA-specific manner. Total tumor eradication in mice correlated with
both the commencement of treatment and the number of transduced T cells
administered.
Only one other study has shown that mouse T lymphocytes retrovirally
transduced with a chimeric scFv can reject tumors in vivo
(25). However, it is important to note two points when
examining that study: 1) large numbers (up to 5 x
107 total) of transduced T cells were injected
directly into the established tumor; and 2) the tumors expressing human
erbB-2 receptor only grew in BALB/c mice that had been tolerized with
Ag as newborns. In this sense, our study has more definitively
demonstrated the ability of transferred T cells expressing scFv to
specifically localize and prevent the growth of a s.c. syngeneic tumor
that grows rapidly in the naive host. Moreover, we have demonstrated
the robustness and utility of our system by determining critical
molecules involved in the rejection process. In both xenogeneic and
syngeneic settings, donor-transduced T cells have been adoptively
transferred from gene-targeted mice. Consistent with other studies that
suggested that CTL-mediated tumor regression or disease protection was
perforin dependent (31, 32, 33), we have clearly shown that
gene-modified T cells expressing scFv anti-CEA- mediated
cytotoxicity and tumor rejection via a perforin-dependent pathway. By
contrast, transduced T cells from FasL mutant or TNF-deficient mice
were able to reject colon carcinomas cells following adoptive transfer.
This is not to say that in some models FasL or TNF might be important
effector molecules in the rejection process. Sherman et al. have
demonstrated that a fusion protein consisting of a scFv
Her-2/neu-specific mAb linked to IL-2 could mediate formation of stable
conjugates between T cells expressing the IL-2 receptor and tumor cells
expressing Her-2/neu, resulting in lysis through the FasL-Fas pathway
(34). In addition, we have previously shown that the mouse
hybridoma cell line MD45 expressing either the scFv anti-CEA- or
anti-erbB2- could specifically lyse target cells in vitro by a
FasL-dependent mechanism (9, 10). Yet clearly the most
appropriate test of effector mechanism is in vivo using primary T cells
and solid syngeneic tumors.
One interesting observation realized from the in vivo analysis of
various gene-targeted T cells was the importance of IFN- in
achieving effective tumor eradication. Clearly IFN--deficient T
cells could inhibit colon carcinoma growth, but complete tumor
eradication was not achieved. There has been previous evidence to
suggest that secretion of T cell cytokines such IFN- may enhance the
lytic capacity of effector cells through up-regulation of both Ag and
Fas expression on colon carcinoma cells (35). Indeed, we
have demonstrated that human IFN- treatment of COLO 205 increases
CEA and Fas expression (9). The exact role of IFN- in
tumor eradication remains unclear, but herein effector cell mouse
IFN- exerted its action against human colon carcinomas growing in
scid mice. In the known absence of murine IFN- activity
on human IFN- receptors and a lack of mature T and B cells in the
scid mouse, these experiments raise the possibility of an
indirect mechanism involving either an innate arm of immunity or
effects of IFN- on tumor angiogenesis. It remains to be determined
whether the addition of exogenous mouse IFN- can enhance redirected
T cell lysis in our models. Whatever the mechanism, it would appear Ag
specific because coinoculated control CEA-
tumors grew unaffected. It had been postulated that grafted CTL might
secrete important cytokines following Ag ligation (22) and
clearly our data support the concept that such activity was important
for tumor eradication.
It has also been noted that to be effective in adoptive immunotherapy, scFv expressing T cells would need to be preactivated (22). In our preparation, T cells have to be preactivated to get them cycling for retroviral transduction. Although the levels of scFv receptor expressed on T cells did not approximate endogenous TCR/CD3 complex, these cells appeared relatively able to mediate cytotoxicity and cytokine secretion after scFv receptor ligation. While adhesion and accessory molecules clearly play an important role in T cell activation and redirected T cell-mediated lysis (36), these molecules appeared to be adequately expressed in as much as scFv anti-CEA-expressing mouse CTL were effective against both CEA+ human colon carcinoma and mouse adenocarcinoma cell lines in vitro and in vivo. The importance of molecules such as T cell CD2, CD11a/CD18, and CD54 will now be evaluated in our models using gene-targeted mice as donors or blocking mAbs in vivo. It appeared, at least for this scFv receptor, that transduced effector cells must eventually die rather than differentiating into memory cells, because mice (scid or B6) surviving tumor in the first instance, subsequently succumbed to a secondary s.c. challenge (data not shown). Future studies using other scFv receptors better designed to improve the survival and differentiation of transduced T cells may be required to improve anti-tumor efficacy and resist tumor rechallenge.
The issue regarding the most optimal signaling domain for any given
population of effector cells remains controversial. Many studies have
only assessed the functional capacity of scFv receptors linked to
either , 
, or downstream intracellular protein tyrosine kinases
in vitro (7, 10, 37, 38, 39). In this study, we have
demonstrated that the FcR- signaling chain could induce in vivo
effector function in purified primary mouse CTL. Tumor studies are
currently in progress to compare the efficacy of FcR- and
TCR/CD3- signaling domains linked to the same scFv when expressed in
mouse CTL.
Generally CEA serves as a good model target TAA for testing new immune therapies (40, 41). However, soluble monovalent CEA Ag may bind to scFv anti-CEA receptor-grafted CTL without cross-linking the receptor, thereby preventing tumor cell recognition and rendering the CTL refractory to Ag-driven activation. This limitation may restrict the scFv receptor approach to patients expressing low levels of soluble CEA, and indeed our previous studies in vitro have shown such a correlation between inhibition of lytic capacity and soluble CEA levels (9). Examining this problem in experimental settings in vivo is more problematic; however, the development of scFv that preferentially recognize the membrane-bound form of CEA over the soluble form (42) offers an alternative solution. Studies will now be conducted using these robust CEA+ xenogeneic and syngeneic tumor rejection models to further evaluate the mechanism of tumor rejection and to optimize the homing, proliferation, and persistence of retrovirally transduced mouse T cells.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Phillip Darcy, Cellular Cytotoxicity Laboratory, The Austin Research Institute, Studley Road, Heidelberg, 3084, Victoria, Australia. E-mail address:
3 Abbreviations used in this paper: TAA, tumor-associated Ag; CEA, carcinoembryonic Ag; FasL, Fas ligand; IFN-o, IFN--deficient; Po, perforin-deficient; scFv, single chain variable domain; TNFo, TNF-deficient.
Received for publication October 6, 1999. Accepted for publication January 21, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
gene redirected human T lymphocytes produce cytokines, specifically lyse tumor cells, and recycle lytic activity. J. Immunol. 157:836.[Abstract]
and tumor necrosis factor in organogenesis and spatial organisation of lymphoid tissue. Eur. J. Immunol. 278:2600.
genes. Science 259:1739. receptor I to redirect cytotoxic T lymphocytes to tumours. J. Leukocyte Biol. 60:721.[Abstract]
or subunits of the immunoglobulin and T-cell receptors. Proc. Natl. Acad. Sci. USA 90:720. chain. J. Exp. Med. 178:361. receptor-mediated T-cell lysis of tumor cells is coregulated by adhesion and accessory molecules. Int. J. Cancer 77:181.[Medline]
or signalling chains. Int. J. Mol. Med. 2:99.[Medline]
receptor redirected T cells: implications for cellular immunotherapy of CD20+ malignancy. Biol. Blood Marrow Transplant. 4:75.[Medline]
This article has been cited by other articles:
|
|
 |
|
  M. Moeller, M. H. Kershaw, R. Cameron, J. A. Westwood, J. A. Trapani, M. J. Smyth, and P. K. Darcy Sustained Antigen-Specific Antitumor Recall Response Mediated by Gene-Modified CD4+ T Helper-1 and CD8+ T Cells Cancer Res., December 1, 2007; 67(23): 11428 - 11437. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Hombach, H. Kohler, G. Rappl, and H. Abken Human CD4+ T Cells Lyse Target Cells via Granzyme/Perforin upon Circumvention of MHC Class II Restriction by an Antibody-Like Immunoreceptor J. Immunol., October 15, 2006; 177(8): 5668 - 5675. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H.-R. Jiang, D. E. Gilham, K. Mulryan, N. Kirillova, R. E. Hawkins, and P. L. Stern Combination of Vaccination and Chimeric Receptor Expressing T Cells Provides Improved Active Therapy of Tumors J. Immunol., October 1, 2006; 177(7): 4288 - 4298. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Moeller, N. M. Haynes, M. H. Kershaw, J. T. Jackson, M. W. L. Teng, S. E. Street, L. Cerutti, S. M. Jane, J. A. Trapani, M. J. Smyth, et al. Adoptive transfer of gene-engineered CD4+ helper T cells induces potent primary and secondary tumor rejection Blood, November 1, 2005; 106(9): 2995 - 3003. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. M. Haynes, J. A. Trapani, M. W. L. Teng, J. T. Jackson, L. Cerruti, S. M. Jane, M. H. Kershaw, M. J. Smyth, and P. K. Darcy Rejection of Syngeneic Colon Carcinoma by CTLs Expressing Single-Chain Antibody Receptors Codelivering CD28 Costimulation J. Immunol., November 15, 2002; 169(10): 5780 - 5786. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. M. Haynes, J. A. Trapani, M. W. L. Teng, J. T. Jackson, L. Cerruti, S. M. Jane, M. H. Kershaw, M. J. Smyth, and P. K. Darcy Single-chain antigen recognition receptors that costimulate potent rejection of established experimental tumors Blood, October 16, 2002; 100(9): 3155 - 3163. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. M. J. Niederman, Z. Ghogawala, B. S. Carter, H. S. Tompkins, M. M. Russell, and R. C. Mulligan Antitumor activity of cytotoxic T lymphocytes engineered to target vascular endothelial growth factor receptors PNAS, May 14, 2002; 99(10): 7009 - 7014. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Hombach, C. Heuser, T. Marquardt, A. Wieczarkowiecz, V. Groneck, C. Pohl, and H. Abken CD4+ T Cells Engrafted with a Recombinant Immunoreceptor Efficiently Lyse Target Cells in a MHC Antigen- and Fas-Independent Fashion J. Immunol., July 15, 2001; 167(2): 1090 - 1096. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. J. Smyth, J. M. Kelly, V. R. Sutton, J. E. Davis, K. A. Browne, T. J. Sayers, and J. A. Trapani Unlocking the secrets of cytotoxic granule proteins J. Leukoc. Biol., July 1, 2001; 70(1): 18 - 29. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. M. Haynes, M. B. Snook, J. A. Trapani, L. Cerruti, S. M. Jane, M. J. Smyth, and P. K. Darcy Redirecting Mouse CTL Against Colon Carcinoma: Superior Signaling Efficacy of Single-Chain Variable Domain Chimeras Containing TCR-{{zeta}} vs Fc{{epsilon}}RI-{{gamma}} J. Immunol., January 1, 2001; 166(1): 182 - 187. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) and COLO 205
(•, dashed line) but not the CEA- 24JK sarcoma cell line
(
). T cells cocultured with GP+E86 cells producing the pLXSN
retrovirus alone were unable to lyse Lovo (
), COLO 205 (
), or
24JK (
). B, BALB/c T cells expressing scFv
anti-CEA or pLXSN were tested for specific lysis of COLO 205; an
E:T ratio of 20:1 is shown. In some wells, the assay was performed in
the presence of 1 µg/ml anti-CEA mAb or an isotype control Ab
(+control mAb). Transduced T cells after depletion of the
CD8+ T cell population (<2% CD8+) were also
examined (CD8+ depl.). The experiment was performed twice
and error bars represent the SE of triplicate samples.
C, Enriched splenic T cells from B6 mice were transduced
with the scFv anti-CEA-
