|
|
||||||||


*
Institute for Immunology, Ruprecht-Karls-Universitaet, Heidelberg, Germany; and
E. Merck, Biomed Fo, Darmstadt, Germany
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
In this paper we report on functional studies based on a novel mitogenic combination of CD2 mAbs, termed M1 and M2, respectively. Neither M1 nor M2 interact with the CD58 binding site of human CD2 and thus do not block CD2-mediated adhesion processes. In addition, both Abs are strongly reactive with resting T cells and, when used in combination, can trigger unprimed pre-effector cells to undergo differentiation into cytolytic effectors. M2 was chemically conjugated with an Ab fragment directed against the epidermal growth-factor receptor (EGF-R)3 yielding the bispecific Ab (BsAb) M2xEGF-R. If BsAb M2xEGF-R is applied together with M1, effector cells are induced to selectively kill EGF-R-positive tumor targets. Employing such a combination of BsAb plus triggering Ab, it should be possible to delineate a treatment strategy in which the targeting and activation steps can be separated. This could help to focus the cytolytic potential to sites of EGF-R expressing tumor targets and at the same time should avoid lymphocyte activation at unwanted sites and thus minimalize systemic side effects of immunotherapy in vivo.
| Materials and Methods |
|---|
|
|
|---|
Human PBMC were prepared by Ficoll-Hypaque- (Pharmacia, Uppsala, Sweden) density centrifugation of heparinized whole blood. Resting human T cells were prepared as described in (5, 6) and were >95% reactive with CD3 mAbs. The following cell lines were used: human Jurkat T cell lymphoma line JM-P1 (7), human epidermoid carcinoma line A 431 (8), and human erythro-leukemia line K 562 (9). C8161 is a nonpigmented human melanoma cell line isolated from a recurrent malignant melanoma (10). This cell line was kindly provided by Dr. M. J. C. Hendrix (University of Arizona, Tucson, AZ). The cells were maintained in DMEM supplemented with 10% FCS and 2 mM L-glutamine. Single cell suspensions were derived after treatment with 0.5 mM EDTA. The T cell clone C3F2 (11) (1 x 106 cells) was grown in the presence of 40 U/ml recombinant human IL-2 (Biotest, Dreieich, Germany), 1 x 106 freshly prepared human PBMC (irradiated with 60 Gy), and 1.5 x 105 EBV-transformed B-lymphoblastoid LAZ 509 cells (irradiated with 60 Gy). After 710 days, no live PBMC and LAZ 509 cells could be detected, and the T cell clone was used for cytotoxicity assays. Tumor infiltrating lymphocytes (TIL) have been prepared as described previously from freshly resected melanoma tissue (12) and have been stimulated via mixed lymphocyte tumor cell culture (MLTC) (13). In short, the freshly isolated lymphocytes were cocultured with autologous irradiated melanoma bulk culture tumor cells in RPMI 1640 medium supplemented with 20 U/ml IL-2 (Genzyme, Cambridge, MA) and 10% FCS. Restimulation of the bulk culture was done on day 12. MLTC derived bulk culture TILs have been used to perform cytotoxicity assays.
Monoclonal Abs
Abs M1 (=AICD2.M1, IgG1), M2 (=AICD2.M2, IgG1), and AICD2.5 (IgG1) were produced in our laboratory. In short: BALB/c mice were immunized with a sonicated lysate of Sf9 cells transfected with human CD2. Spleen cells isolated from immunized animals were fused with Ag8-PAI (provided by M. Kramer, Institut für Immunologie, Heidelberg, Germany). HAT/HT selection was done according to standard protocols. Screening of CD2 specific hybridoma supernatants was done by measuring Ab binding to Jurkat T cells in cytometry analyses. Hybridomas have been subcloned twice. Other CD2 mAbs, OKT11 (IgG1; Ortho, Neckargemünd, Germany), T112 (1OLD4C1, IgG2A), T113 (1 MONO2A6, IgG3; T112, and T113 were kindly provided by Dr. E. Reinherz, Dana-Farber Cancer Institute, Boston, MA), ICRFCD2.4 (IgM), ICRFCD2.5 (IgG2A), and ICRFCD2.8 (IgG2A; ICRF mAbs were kindly provided by Dr. M. Crumpton, Imperial Cancer Research Fund, London, U.K.). Further mAbs were CD3 mAb OKT3 (IgG1; Ortho), CD58 mAb AICD58.5 (IgG2A (14)), and EGF-R mAb 425 (IgG2A (15)). Abs were used as protein A-purified murine Ig unless indicated otherwise.
Preparation of BsAb and Ab fragments
BsAb M2xEGF-R was generated by chemical recombination of Fab' fragments of CD2 mAb M2 and EGF-R mAb 425 as described by Brennan et al. (16). Briefly, Ab M2 and mAb 425 were converted into F(ab')2 fragments by limited proteolysis with pepsin. F(ab')2 fragments were purified by chromatography on protein A Sepharose. Fab' fragments were generated by a mild reduction with DTT (0.5 mM; Merck, Darmstadt, Germany). The CD2-specific Fab' fragments were modified with 5,5'-dithio-bis-2-nitro-benzoic acid (DTNB, Merck). Subsequently, bispecific F(ab')2 fragments were obtained by conjugation of the Fab'-TNB (thio-bis-2-nitro-benzoic acid) derivative with the hinge-SH groups of the mAb 425-derived Fab' fragments. Purification of intermediate products (Fab', Fab'-TNB) as well as the final purification of BsAb was performed by gel filtration on a Superdex 200 column (Pharmacia). Purity of the bispecific conjugate was assessed by 10% nonreducing SDS-PAGE and by chromatography on a hydroxylapatite column. The F(ab')2 fragments of the triggering Ab M1 were generated by limited proteolysis with papain. Fragments were purified as described.
Sheep erythrocyte rosetting assays
The CD2 molecule of human T cells binds to the CD58 molecule (T11TS) on sheep erythrocytes, thus causing the formation of rosettes of erythrocytes around T cells. Rosetting can be reduced by binding of certain CD2 mAbs to the CD2 molecule. Human T cells (2 x 105) were incubated with the respective mAb for 30 min at room temperature in a volume of 100 µl. Subsequently, 2 x 107 sheep erythrocytes were added in a volume of 400 µl. The cell mixture was sedimented for 5 min at 230 x g and incubated for 1 h at 37°C in a 6% CO2 atmosphere. Following cautious resuspension, a solution of T cell-staining crystal violet was added to facilitate counting. At least 300 T cells per culture were counted microscopically and were considered as rosetted when bound by at least three erythrocytes.
Proliferation assays
Cells were incubated at 37°C in a 6% CO2 atmosphere for 72 h in the presence of the indicated Abs in round-bottomed microtiter plates in 200 µl RPMI 1640 medium (Life Technologies, Paisley, Scotland), supplemented with 10% FCS, 100 IU/ml penicillin (Life Technologies), 100 µg/ml streptomycin (Life Technologies), and 4 mM L-glutamine (Life Technologies). In some of the experiments, immunoaffinity purified recombinant human CD58 (17) was added to the cultures. Subsequently, cells were pulsed with 37 kBq [3H]thymidine and harvested following 16 h of additional incubation.
Flow cytometry and analysis of CD2 modulation
Flow cytometry studies of Ag recognition by mono- and bispecific Abs were conducted as described elsewhere (18). Studies of CD2 modulation were performed as follows. Each well of a flat-bottom 24-well plate received 1 x 106 freshly prepared PBMC and the respective Abs in the concentrations as indicated. Cultures with BsAb were additionally mixed with 2 x 105 A 431 cells. All cultures had a final volume of 2 ml. Cultures were incubated for 16 h at 37°C in a 6% CO2 atmosphere to allow modulation before they were washed two times and incubated for 30 min at 4°C with biotinylated CD2 mAb ICRFCD2.8 (10 µg/ml) for detection of surface-CD2. The negative control sample was incubated with culture medium instead. Following two washes cells were stained for 15 min at 4°C with steptavidin-FITC conjugate (Becton Dickinson, San Jose, CA), washed again, and analyzed on an EPICS Profile flow cytometer (Coulter Electronics, Hialeah, FL).
Cytotoxicity assays
Cytotoxicity with T cell clones as effector cells was determined in standard 51Cr-release assays in round bottomed microtiter plates. Each well (triplicate cultures) received the respective Abs, effector cells, and 51Cr-labeledtarget cells at the indicated concentration or number to give a total volume of 200 µl of supplemented RPMI medium (see proliferation assays). Six wells containing target cells in culture medium only were used for determining spontaneous 51Cr-release (SR). Six wells containing target cells in 1% Nonidet P-40 detergent were used for determining maximum release (MR). Cell mixtures were sedimented (230 x g, 5 min) and incubated at 37°C in a 6% CO2 atmosphere for 4 h. Finally, cells were pelleted (450 x g, 10 min), and 100 µl of supernatant was collected from each well to determine Chromium release in a gamma counter (LKB-Wallac, Stockholm, Sweden). The percentage specific lysis was calculated as: 100 x [(experimental 51Cr release - SR)/(MR - SR)]. In experiments with PBMC as effector cells, these were incubated for 72 h with the respective Abs before 51Cr-labeled A 431 target cells were added and mixed with the effector cells. In some experiments unlabeled NK-sensitive K 562 cells were added additionally. Cell mixtures were sedimented (230 x g, 5 min) and submitted to a 4-h assay as described above. In assays with TILs, cells were incubated for the indicated times. Bispecific and monospecific Abs were added simultaneously or in two steps before supernatants were collected after a total culturing time of 18 h.
| Results |
|---|
|
|
|---|
For the design of CD2 x EGF-R-specific BsAbs we
produced anti-CD2 mAbs with a combination of favorable
characteristics. These mAbs (termed AICD2.M1 and AICD2.M2, abbreviated
M1 and M2) are specific for human CD2 as assessed by binding to
recombinant CD2 in a sandwich ELISA (data not shown). In competitive
binding ELISAs, the mAbs M1 and M2 cross-blocked with the established
anti-CD2 mAb T113 (3) in binding
to recombinant CD2 (Fig. 1
B).
However, mAbs M1 and M2 differ from mAb T113 in
that they strongly bind to CD2 on resting human T cells (Fig. 1
A), whereas the T113 mAb binds to its
CD2 epitope only upon T cell activation (3).
|
Fig. 2
A shows that the
combination of mAbs M1 and M2 induces proliferation of PBMC as well as
of purified T cells, whereas, individually, the Abs have no mitogenic
effect. In the presence of suboptimal concentrations of mAbs M1 and M2,
addition of recombinant CD58 strongly enhances this proliferative
response (Fig. 2
B). CD58 specificity of costimulation was
proven by the addition of an anti-CD58 mAb, which abrogated the
functional effect of recombinant CD58.
|
Production of BsAb M2xEGF-R
The CD2 specific mAb M2 as well as EGF-R mAb 425 were
cleaved into F(ab')2 and converted into Fab'
fragments. Conjugation of Fab' fragments of mAb 425 and TNB-derivatives
of M2 Fab' fragments resulted in CD2 x EGF-R bispecific
F(ab')2 fragments (BsAb M2xEGF-R). A yield of
purified BsAb of 2030% based on the F(ab')2
starting material was achieved and is demonstrated in Fig. 3
. High purity of the bispecific
conjugate was demonstrated by chromatography on hydroxylapatite (data
not shown) as well as in nonreducing SDS-PAGE (Fig. 3
).
|
Flow cytometry studies showed that BsAb M2xEGF-R bound to
both the CD2-positive T cell line JM-P1 as well as to the
EGF-R-positive epidermoid carcinoma line A 431 (Fig. 4
). As expected from the monovalent CD2
binding of the bispecific conjugate, JM-P1 staining by BsAb M2xEGF-R
was weaker as compared with staining by the parental CD2 mAb M2
(Fab')2. When compared with the parental EGF-R
mAb 425 (Fab')2, however, staining of A 431 cells
was only moderately reduced.
|
|
To induce lysis of tumor cells, BsAb M2xEGF-R has to
mediate binding of effector cells to target cells (targeting) and to
activate effector cells to trigger their cytolytic function. To
investigate the targeting properties of BsAbs, cytotoxicity assays were
conducted employing a cytotoxic CD8+ T cell clone
(C3F2) (11). Before assays, C3F2 cells were cultured in
the presence of IL-2 and feeder cells (irradiated PBMC plus
EBV-transformed B cells). Since in such a differentiation state further
activation is not required for cytolytic action, incubation with BsAb
alone should be sufficient to achieve tumor cell lysis. Fig. 6
shows that, whereas the parental mAbs
M2 (Fab')2 and 425 (Fab')2
were unable to mediate A431 cell lysis, BsAb M2xEGF-R induced strong
cytotoxicity indicating efficient targeting to the effector cells.
Furthermore, lysis of A 431 cells was due to the cytolytic function of
the effector cells and not to a direct toxic effect of BsAb M2xEGF-R on
the targets since the incubation of A431 cells with the BsAb in the
absence of effector cells did not induce lysis (data not shown).
|
BsAb M2xEGF-R plus trigger Ab M1 activate resting PBMC for target cell lysis
In a further set of experiments, we investigated whether
the combination of BsAb M2xEGF-R and M1(Fab')2
would be able to activate resting PBMC to undergo proliferation. As
shown in Fig. 7
, this is indeed the case.
When added individually, however, these reagents exerted no stimulatory
effect.
|
|
In a final set of experiments, we investigated whether it
is possible to target effector cells to tumor cells in a first step by
giving BsAb alone and to trigger the former in a second step, i.e.,
12 h later by M1 (Fab')2 to kill. We made
use of a human allogeneic system employing the EGF-R-positive C8161
melanoma cell line as target along with MLTC-derived freshly prepared
tumor infiltrating lymphocytes. Fig. 9
A shows that dose dependently
the combination of BsAb M2xEGF-R plus M1 (Fab')2,
when added simultaneously, induced tumor cell lysis while the
individual reagents did not. The efficiency of this TIL driven melanoma
specific lysis was similar as for the PBMC/A431 system situation
(compare Fig. 8
). Importantly, virtually analogous results were
obtained when the addition of BsAb M2xEGF-R and triggering-mAb M1
F(ab')2 was separated by 12 h (Fig. 9
B). This result opens up the possibility to employ the
CD2-based approach delineated here to separate targeting from
triggering steps in a potential therapeutic setting.
|
| Discussion |
|---|
|
|
|---|
MAbs M1 and M2 exhibited cross-blocking with the previously described comitogenic CD2 mAb T113 in binding to the CD2 molecule. However, mAbs M1 and M2 recognize the CD2 molecule on resting T cells, in contrast to T113 (3), suggesting that the novel mAbs and mAb T113 bind to overlapping but not identical epitopes (23). Recognition of CD2 on resting T cells is a prerequisite for their use as parental mAbs to form bispecific conjugates that are designed to target unprimed effector cells to particular sites in vivo.
The novel CD2 mAbs were tested for noninterference with the CD2-CD58 interaction. The rationale behind this was that bispecific Abs used in, e.g., cancer treatment should not block important T cell functions involving the CD2-CD58 interaction, such as cell-cell adhesion, recirculation of T-lymphocytes as well as costimulation in the activation of T cells through CD2 (17, 19, 24). Unlike mAb OKT11 (which had been used as parental CD2 mAb for a trispecific CD2 x CD3 x TA construct) (25) and several other CD2 specific Abs, mAbs M1 and M2 did not interfere with the CD2-CD58 interaction. Consequently, recombinant CD58 proved to be able to exert its costimulatory function in the presence of parental CD2 mAbs M1 and M2. Thus, in cancer therapy, two effects can be expected: T cells that have bound therapeutic Abs of M1 or M2 specificity but have not yet found access to tumor cells would still possess their CD2-mediated adhesion functions. Secondly, T cells which have bound to tumor cells via a CD2 x TA bispecific conjugate may receive an additional stimulus from binding of CD58 expressed on the tumor cells that would augment their cytotoxic activity.
In this regard, unlike several other costimulatory ligands (CD80, CD86, CD54), CD58 is expressed in the majority of tumor cells of various derivations investigated so far (26) and therefore could enhance the Ab-driven activation of different types of effector cells.
CD2 mAb OKT11when applied in unconjugated form blocks T cell activation
via the TCR/CD3 complex (21) through an as yet unknown
mechanism whereas as mentioned above (25) in the form of a
T111 x CD3 x target
cell trispecific Ab strongly enhances lysis. We have examined both
parental CD2 Abs M1 and M2 and other CD2 mAbs for interference with
this kind of activation. In contrast to mAbs
T112, T113, and a series of
additional CD2 mAbs, M1 and M2 did not reduce TCR/CD3-driven
proliferation and were even slightly comitogenic. As long as the
precise molecular mechanisms underlying lymphocyte triggering by mAbs
are not elucidated, explanations regarding receptor clustering
processes or intracellular events that might be responsible for these
phenomena will remain pure speculations. However, at least one point
can me made: there is ample experience from studies on large Ab panels,
e.g., CD clusters that have clearly demonstrated that functional Ab
activities often relate to molecular features of the Abs themselves
(which in spite of extensive investigations on their avidities,
isotypes, or molecules/epitopes to which they bind are far from being
understood). In this regard it should be mentioned that one of our
criteria for the selection of M1 and M2 was that they should not
interfere with the CD3 initiated mode of activation in vitro so that
negative side effects such as immunosuppression during in vivo
application of M1- and M2-derived (Fab')2
fragments and BsAbs would become rather unlikely. Perhaps more
importantly, when compared with the widely used targeting to the CD3
complex, the M1- and M2-based targeting of the CD2 molecule may have a
critical advantage: previous in vitro studies have shown that binding
of CD3 x TA BsAbs to CTL can lead to the inability of these cells
to carry out a TCR-specific target cell lysis (27). This
phenomenon may be due to inhibition of activation via the TCR/CD3
complex. Alternatively, a CD3 x TA BsAb-mediated reduction of
surface expression of the TCR/CD3 complex (modulation)
(28) is possible, which would lead to an inability to bind
MHC/Ag complexes. Moreover, exposure of T cells to CD3 mAb alone can
induce anergy (29). Indeed, we have observed that a
CD3 x EGF-R BsAb leads to a strong loss of cell surface
expression of the TCR/CD3 complex (our unpublished data). These
unwanted reactions can be excluded for BsAbs with M1- or M2-specificity
because these mAbs neither inhibit activation via TCR/CD3 (Fig. 2C
) nor
do they modulate the TCR/CD3 complex (data not shown).
With regard to the Ag on target cells, the EGF-R was chosen. The EGF-R has been reported to be highly over-expressed on malignant cells like epidermal and colorectal carcinoma cells (30) besides its normal expression on endothelial cells, fibroblasts, glial cells, and other cell types. Its high over-expression on certain tumors should allow to find a range of Ab concentrations that cause preferential damage to malignant cells. Moreover, unlike many Ags that are more specific for malignant cells, the EGF-R does not bear the disadvantage of a strong heterogeneity of expression within a given tumor (31).
Binding of particular Abs to triggering receptors can result in their total loss from the cell surface through modulation (32, 33, 34). In our experiments, coincubation of PBMC with M2 x EGF-R, trigger Ab M1(Fab')2, and A 431 target cells (imitating the potential targeting situation in vivo) led to a 41% reduction of the cell surface density of CD2. Maintained CD2 expression may enable T cells to bind further BsAb if a therapy with repeated administration of BsAb (35) is chosen.
Cytotoxicity assays with a CD8+ T cell clone showed that BsAb M2xEGF-R was able to target T cells to EGF-R-positive A 431 tumor cells with the bispecific conjugate being active at nanogram quantities. Inhibition of target cell lysis by the addition of EGF-R mAb as well as unreactivity toward an EGF-R negative cell proved the specificity of M2xEGF-R -mediated targeting.
The performance of M2xEGF-R in the activation of unprimed effector cells was tested in both proliferation and cytotoxicity assays. In these assays the BsAb has to be supplemented by the trigger Ab M1(Fab')2 to achieve activation. Proliferation assays showed that the M2 moiety in BsAb M2xEGF-R maintained its ability to stimulate lymphocytes in concert with M1(Fab')2.
Cytotoxicity assays showed that stimulation by M2xEGF-R and
M1(Fab')2 induced cytotoxic activity in human
PBMC against EGF-R-positive target cells at Ab concentrations similar
to those inducing proliferation. No previous study analyzing the
performance of a combination of CD2 x TA BsAb and CD2-trigger-Ab
(36, 37, 38) had demonstrated that such a combination of Abs
can induce cytotoxicity in resting PBMC. Given that the majority of NK
cells express functional CD2 molecules, it was not unexpected that the
cytolytic activity of PBMC was to
50% due to NK activity as judged
from cold target inhibition assays using the standard NK target
K562.
The ability of the Ab combination M2xEGF-R plus M1(Fab)2 to activate and recruit NK cells for tumor cell neutralization should be considered an important advantage over the function of BsAbs with CD3 specificity. The simultaneous recruitment of T and NK cells as indicated by our in vitro experiments may provide the basis of an enhanced NK activity because T lymphocytes can strongly support NK functions by secreting IL-2.
Further differences between CD2- and CD3-directed BsAbs exist: In
CD3 x TA F(ab')2 BsAbs the monovalency of
the anti-CD3 moiety may cause problems because effective
CD3-mediated T cell activation is dependent on cross-linking of CD3
molecules (39). In accordance with this fact, triggering
of T cells with monovalent reagents requires cross-linking of CD3
molecules by the simultaneous binding of BsAbs to CD3 on T cells and
Ags on tumor cells (40). Often tumor cells have reduced
surface densities of tumor Ags due to immunoselection (41)
or modulation (42). This may cause insufficient
cross-linking of the CD3 molecules on T cells if BsAbs are chosen that
are specific for such tumor Ags. Activation of T cells via CD2,
however, can be achieved by the use of soluble monoclonal CD2 mAbs
(Ref. 3 and Fig. 1
A) or BsAb M2xEGF-R plus mAb
M1(Fab)2 without cross-linking (Fig. 7
). Thus,
here, cross-linking cannot become a limiting factor for T cell
activation.
The "two signal hypothesis" of T cell activation (43) demands a second signal delivered by binding of ligands to accessory TCRs in addition to the first signal delivered via the TCR/CD3 complex. However, ligands which normally deliver such second signals, like the molecules of the B7 family, are not expressed on many solid tumors (44). Thus, second signals may not be sufficiently available in the case of tumors that would limit the application of CD3 x TA BsAbs. In such cases the combination of a CD2 x TA BsAb and trigger CD2 mAb may be more efficient because binding of two different CD2 Abs provides both signals required for T cell activation (45). Moreover the majority of malignant cells investigated in our laboratory express at least some CD58 molecules which amplify the CD2-dependent mode of T cell and NK cell activation (26).
The requirement of two Abs for activation of T cells via CD2 opens the
perspective of a two-step concept of tumor therapy. Lymphocytes may be
loaded ex vivo with BsAb M2xEGF-R before their injection into the blood
stream. Alternatively BsAb M2xEGF-R could be administered i.v.
Ab-coated lymphocytes or BsAb M2xEGF-R should then be targeted to tumor
sites. Importantly, effector cells are not activated during this
"targeting phase" because they lack binding of the second CD2 Ab.
The length of the targeting phase can be optimized by preceding
biodistribution analyses (currently performed in tumor xenotransplanted
mice) in which the time of maximal tumor infiltration by BsAb-armed
lymphocytes can be determined. In this work a time range of up to
12 h has been tested in vitro (Fig. 9
). The 12-h time range is far
off the overall half-life of murine F(ab')2 Ab
fragments in blood which is 68 h in experimental animals (46, 47). Thus, most of the circulating BsAb is expected to be either
localized or cleared by the time the trigger Ab is given. In the second
step ("activation phase") trigger Ab M1(Fab)2
will be delivered which should activate only those lymphocytes that
have been targeted by BsAb M2xEGF-R, thus allowing a preferential local
activation. In this concept, BsAb-mediated side effects should be
reduced by injecting the trigger Ab following a targeting phase of
optimal length. Moreover, in contrast to a previously proposed strategy
using a bispecific CD2 x TA Ab with a T113
moiety (36), the present concept based on M1 and M2 does
not need ex vivo activation of CD2-positive lymphocytes before
reinjection.
For the clinical application of CD2 x TA BsAbs a study by Riethmüller et al. (48) may be considered in which a mAb was used to target disseminated tumor cells in patients with colorectal cancer who had undergone curative surgery. The easy access to unshielded disseminated tumor cells for Abs as well as for effector cells may represent a favorable situation for the application of CD2 x TA BsAbs like M2xEGF-R especially when considering the biodistribution pattern of NK cells. In solid tumors these cells are found in rather small numbers (49), whereas disseminated cells may be ideal targets for BsAb-redirected NK cells. Finally, besides their therapeutic application in vivo M1 plus M2xEGF-R Abs could be valuable for ex vivo tumor cell purging.
To prepare future in vivo studies, Abs M2xEGF-R and
M1(Fab)2 have been successfully tested for their
in vitro performance in two-step cytotoxicity assays (Fig. 9
) and for
their in vitro function in autologous systems (W. Strittmatter, C.
Jäggle, and S. Matzku, unpublished data). Furthermore, the
kinetics of biodistribution of M2xEGF-R have been determined in
experimental animals to gain data on the optimal period of time needed
for targeting by the BsAb (W. Strittmatter, C. Jaggle, and S. Matzku,
manuscript in preparation). Mice transgenic for human CD2 (23, 50) exist and can be employed to determine the in vivo
performance of M1/M2-based combinations of CD2 x TA BsAbs and
F(ab')2 trigger Abs for tumor neutralization.
| Acknowledgments |
|---|
| Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. S. C. Meuer, Institute for Immunology, Ruprecht-Karls-Universitaet, Im Neuenheimer Feld 305, 69120 Heidelberg, Germany. E-mail address: ![]()
3 Abbreviations used in this paper: BsAb, bispecific Ab; EGF-R, epidermal growth factor-receptor; TA, tumor Ag; TIL, tumor infiltrating lymphocyte; MLTC, mixed lymphocyte tumor culture; TNB, thio-bis-2-nitro-benzoic acid. ![]()
Received for publication October 5, 1998. Accepted for publication June 2, 1999.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
K. Schlienger, C. S. Chu, E. Y. Woo, P. M. Rivers, A. J. Toll, B. Hudson, M. V. Maus, J. L. Riley, Y. Choi, G. Coukos, et al. TRANCE- and CD40 Ligand-matured Dendritic Cells Reveal MHC Class I-restricted T Cells Specific for Autologous Tumor in Late-Stage Ovarian Cancer Patients Clin. Cancer Res., April 1, 2003; 9(4): 1517 - 1527. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |