| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Recombinant Antibody Research Group,
Department of Tumor Progression and Immune Defense,
Department of Experimental Radiology, and
§
Department of Molecular Immunology, German Cancer Research Center (DKFZ), Heidelberg, Germany
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
chain of the TCR complex.
After production in Escherichia coli and purification,
we analyzed its affinity, stability, and pharmacokinetics, and tested
its capacity to stimulate T cell proliferation and mediate in vitro
lysis of CD19+ tumor cells. The effect of the diabody on
tumor growth was investigated in an in vivo model using immunodeficient
mice bearing a human B cell lymphoma. The CD3 x CD19 diabody
specifically interacted with both CD3- and CD19-positive cells, was
able to stimulate T cell proliferation in the presence of tumor cells,
and induced the lysis of CD19+ cells in the presence of
activated human PBL. The lytic potential of the diabody was enhanced in
the presence of an anti-CD28 mAb. In vivo experiments indicated a
higher stability and longer blood retention of diabodies compared with
single chain Fv fragments. Treatment of immunodeficient mice bearing B
lymphoma xenografts with the diabody and preactivated human PBL
efficiently inhibited tumor growth. The survival time was further
prolonged by including the anti-CD28 mAb. The CD3 x CD19
diabody is a powerful tool that should facilitate the immunotherapy of
minimal residual disease in patients with B cell leukemias and
malignant lymphomas. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To date, BsAbs have mainly been produced using murine hybrid hybridomas (5) or by chemical cross-linking (6, 7). However, the immunogenicity of BsAb derived from rodent mAbs is a major drawback for clinical use (8). They are also difficult to produce and purify in large quantities. Recent advances in recombinant Ab technology have provided several alternative methods for constructing and producing BsAb molecules (9, 10). For example, scFv fragments have been genetically fused with adhesive polypeptides (11) or protein domains (12) to facilitate the formation of heterodimers. The genetic engineering of scFv-scFv tandems linked with a third polypeptide linker has also been conducted in several laboratories (13, 14). An alternative BsAb fragment is the scFv heterodimer diabody (15). It is formed by the noncovalent association of two single chain fusion products consisting of the VH domain from one Ab connected by a short linker to the VL domain of another Ab (16, 17). The two Ag binding domains have been shown by crystallographic analysis to be on opposite sides of the diabody such that they are able to cross-link two cells (18).
We recently described the construction of a bispecific diabody with
dual specificity for both the human B cell Ag CD19 and -chain of the
CD3/TCR complex designed for the treatment of minimal residual disease
in patients with leukemias and malignant lymphomas (19).
In this study, we detail the binding characteristics of CD3 x
CD19 diabody and examine pharmacokinetics of this molecule and its in
vitro and in vivo antitumor properties.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
For all cloning steps and DNA isolation, the Escherichia
coli K12 strain XL1-Blue (Stratagene, La Jolla, CA) was used.
Plasmids pHOG3-19 and pHOG19-3 encoding the hybrid
VH3-VL19 and
VH19-VL3 scFvs
(19) were used as a source of genetic information to
construct the improved vector for expression of bispecific CD3 x
CD19 diabody. The gene coding for scFv3-19 was amplified by PCR with
the primers 5-NDE, 5'-GATATACATATGAAATACCTATTGCCTACGGC,
and 3-AFL,
5'-CGAATTCTTAAGTTAGCACAGGCCTCTAGAGACACACAGATCTTTAG. The
resulting PCR fragment was digested with NdeI and
AflII and cloned into the
NdeI/AflII-linearized plasmid pHKK
(20) (kind gift of Dr. D. Riesenberg,
Hans-Knoll-Institute, Jena, Germany) generating the vector pHKK3-19.
The gene encoding the second hybrid scFv19-3 was amplified by PCR with
the primers 5-BGL,
5'-GCACACAGATCTGAGAAGGAGATATACATATGAAATACCTATTGCCTACGGC,
and fe-2, 5'-CGAATTCTTAAGCTATTAGTGATGGTGATGGTGATGTGAG.
The resulting PCR fragment was digested with BglII and
AflII and cloned into the
BglII/AflII-linearized plasmid pHKK3-19
generating the expression vector pHKID3 x 19. To introduce a gene
encoding the Skp/OmpH periplasmic factor for higher recombinant Ab
production (21), the skp gene was amplified by
PCR with primers skp-3,
5'-CGAATTCTTAAGAAGGAGATATACATATGAAAAAGTGGTTATTAGCTGCAGG
and skp-4, 5'-CGAATTCTCGAGCATTATTTAACCTGTTTCAGTACGTCGG,
using as a template the plasmid pGAH317 (22), kindly
provided by Dr. R. Chen (Max-Plank Institute for Biology,
Tübingen, Germany). The resulting PCR fragment was digested with
AflII and XhoI and cloned into the
AflII/XhoI-digested plasmid pHKID3 x 19,
resulting in the expression plasmid pSKID2. This vector contains the
hok/sok plasmid-free cell suicide system (23)
and a tetra-cistronic operon for high level production of bispecific
heterodimer diabodies. The expression cassette is under the
transcriptional control of the wild-type lac
promoter/operator system and includes a short sequence coding for the
N-terminal peptide of ß-galactosidase (lacZ') with
a first rbs derived from the E. coli lacZ gene, followed by
genes encoding the two-hybrid scFvs and Skp/OmpH periplasmic factor
under the translational control of strong rbs from gene 10 of phage T7
(T7g10) (Fig. 1
). The composition of the constructed plasmids was
confirmed by restriction digests and dideoxynucleotide sequencing
(24).
|
For functional expression of the bispecific diabody in the
bacterial periplasm, the plasmid pSKID2 was transformed into E.
coli K12 strain RV308
(
lac
74galISII::OP308strA)
(25), kindly provided by Dr. A. Plückthun
(University of Zürich, Zürich, Switzerland). Transformed
bacteria were grown overnight in shake flasks containing 2YT medium
with 0.1 g/L ampicillin and 100 mM glucose
(2YTGA) at 26°C. Dilutions (1/50) of the
overnight cultures in 2YTGA were grown as flask
cultures at 26°C with shaking at 200 rpm. At
OD600 = 0.7, bacteria were harvested by
centrifugation and resuspended in the same volume of YTBS medium (2YT
containing 1 M sorbitol and 2.5 mM glycine betaine (26)).
Isopropyl ß-D-thiogalactoside was added to a final
concentration of 0.2 mM, and growth was continued at 23°C for 13
h. The bacterial cells were then harvested by centrifugation, and
periplasmic extracts were isolated as previously described (27, 28). Purification was achieved by immobilized copper metal
affinity chromatography (IMAC) on Cu2+-charged
chelating Sepharose (Amersham Pharmacia Biotech, Freiburg, Germany), as
previously described (28). The final purification of the
CD3 x CD19 diabody was achieved by ion-exchange chromatography on
a Mono S HR5/5 column (Pharmacia) in 50 mM imidazole-HCl, pH 6.4, with
a linear 0–1 M NaCl gradient. The purified diabody preparations were
dialyzed against PBSI buffer (15 mM sodium phosphate, 0.15 M NaCl, 50
mM imidazole, pH 7). All purification procedures were performed at
4°C. Isolation of the anti-CD19 and anti-CD3 scFv fragments
was performed as previously described (28, 29).
Flow cytometry
The human CD3+/CD19- acute T cell leukemia line Jurkat and the CD19+/CD3- B cell line JOK-1 were used for flow cytometry, as previously described (19). In brief, 5 x 105 cells in 50 µl RPMI 1640 medium (Life Technologies, Eggestein, Germany) supplemented with 10% FCS and 0.1% sodium azide (referred to as complete medium) were incubated with 100 µl of a recombinant Ab preparation for 45 min on ice. After washing with complete medium, the cells were incubated with 100 µl of 10 µg/ml anti c-myc mAb 9E10 (IC Chemikalien) in the same buffer for 45 min on ice. After a second washing cycle, the cells were incubated with 100 µl of FITC-labeled goat anti-mouse IgG (Life Technologies) under the same conditions as before. The cells were then washed again and resuspended in 100 µl of 1 µg/ml solution of propidium iodide (Sigma, Deisenhofen, Germany) in complete medium to exclude dead cells. The relative fluorescence of stained cells was measured using a FACScan flow cytometer (Becton Dickinson, Mountain View, CA).
Affinity determination
The apparent affinities of the bispecific diabody were determined from competitive inhibition assays with FITC-labeled mAb OKT3 (anti-CD3) and HD37 (anti-CD19), as previously described (19, 28). Binding affinities were calculated according to the following equation (30): Kd(I) = IC50/(1 + (FITC-mAb)/Kd(mAb)), in which I is the unlabeled inhibitor (diabody), FITC-mAb is the concentration of FITC-labeled mAb, Kd(mAb) is the binding affinity of mAb, and IC50 is the concentration of inhibitor that yields 50% inhibition of binding. Affinity constant (Kd) values of 0.8 and 0.4 nM were assigned to mAb OKT3 (31) and HD37 (19), respectively.
In vitro cell surface retention
Cell surface retention assays were performed at 37°C under conditions preventing internalization of cell surface Ags, as described (32), except that the detection of the retained diabody was performed using anti-c-myc mAb 9E10 (IC Chemikalien), followed by FITC-labeled anti-mouse IgG. Kinetic dissociation constants (koff) were calculated using a first order equation Ft = F0 x e-kt, in which Ft is fluorescence at time t, F0 is fluorescence at time 0, and k is koff. The t1/2 for dissociation of diabody was calculated from the equation t1/2 = ln2/koff.
Cell proliferation assay
Human PBL (1 x 104/well) were cultured together with irradiated (120 rad) CD19+ Burkitt lymphoma Raji cells (2.5 x 105/well) in 96-well plate for 5 days in IMEM (Sigma) supplemented with 10% human AB serum (Sigma), 25 IU/ml of human rIL-2 (Chiron, Emeryville, CA), and varying amounts of diabody. [3H]Thymidine (0.5 µCi/well) was added 12 h before harvesting. The cells were harvested, and [3H]thymidine incorporation was measured with a liquid scintillation beta counter (Beckman, Palo Alto, CA).
In vitro cytotoxicity
The efficacy of the diabody in mediating tumor cell lysis by activated human PBLs was determined using the JAM test (33) or chromium release assay. For the JAM test, 105 activated human PBLs prepared as previously described (19) were mixed in round-bottom microtiter plates with 1 x 104 of target cells (Raji or tumor cells from a CLL patient) labeled with [3H]thymidine (E:T ratio = 10:1) in 100 µl medium plus 50 µl of Ab sample. After incubating the plate at 37°C, 5% CO2 for 4 h, the cells were harvested and radioactivity was measured with a liquid scintillation beta counter. To analyze the specificity of diabody-mediated tumor cell lysis, in some experiments either mAb OKT3 (anti-CD3) or mAb HD37 (anti-CD19) was added at concentrations up to 100 µg/ml. Chromium release assays using 51Cr-labeled Raji cells were performed essentially as previously described (19).
Analysis of diabody stability in vitro
The diabody preparation was diluted (at least 20-fold) in human serum (Sigma) to a concentration of 15 µg/ml and sterilized by filtration through a Membrex 4CA filter with a void volume of 50 µl and a pore size 0.2 µm (MembraPure, Lörzweiler, Germany). One-hundred-microliter aliquots were stored under sterile conditions at 37°C. At given time points, the aliquots were frozen and kept at -80°C. Activities of the samples after storage were determined by flow cytometry using CD3+ Jurkat cells. Protein degradation in each sample was determined by 12% SDS-PAGE, followed by Western blot analysis using anti-c-myc mAb 9E10, as previously described (34).
Radioiodination
Diabody or scFv solutions (1 mg/ml in 50 mM Tris-HCl, 1 M NaCl, pH 7) were placed in glass tubes coated with 20 µg of IODO-Gen (Pierce, Rockford, IL) per mg of protein. After adding 3 mCi of Na125I (Amersham Buchler, Braunschweig, Germany), the resulting mixture was incubated for 10 min at room temperature. Unincorporated radioiodine was separated from the labeled protein by size-exclusion chromatography using Bio-Gel P6 (Bio-Rad, Munich, Germany) and PBSI as the elution buffer. The final sp. act. of anti-human CD3 scFv and the CD3 x CD19 diabody were 2.4 and 6.6 mCi/mg, respectively.
Pharmacokinetic studies
Male NMRI mice, each weighting 
40 g (24–27 animals for each
labeled protein), were injected into the tail vein with 200 µl of
PBSI containing 10 µg of human serum albumin and 5 µg of labeled Ab
fragments. At the indicated time points, animals in triplicates were
anesthetized, bled, and sacrificed in accordance with local animal
protection laws. For scFv, the blood samples were obtained at 5, 10,
20, 40, 90, 180, 360, and 1080 min after injection. The time points for
the diabody were the same except that the last animal groups were
sacrificed 24 h after injection of the labeled compound. Blood
samples were counted using a gamma counter. Blood content was corrected
for protein-bound radioactivity (see below) and expressed as a
percentage of injected dose per gram of blood (%ID/g). AUC was
calculated using GraphPad Prism (GraphPad Software, San Diego, CA) and
expressed as %ID/g x min. The
t1/2
was defined by time points 5,
10, 20, and 40 min, while the t1/2ß
was defined by time points from 1.5 to 24 h.
Analysis of diabody stability in vivo
Two hundred microliters of blood from each sacrificed animal were mixed with 10 µl of heparin (5000 IU/ml; Braun Melsungen, Melsungen, Germany), followed by sedimentation of cellular material and TCA precipitation of the supernatant. Radioactivity associated with pellets and supernatants was counted and expressed as a percentage of total radioactivity (cpm) for each specific time point. Pooled plasma samples from the earlier time points (up to 1.5 h) were also analyzed by 12% SDS-PAGE, followed by autoradiography.
Treatment of human B cell lymphoma in RAG2-deficient mice
All animal experiments were performed in accordance with the guidelines of the German Cancer Research Center and the animal-protection laws of the Bundesland Baden-Württenberg and of the Federal Republic of Germany. RAG2-deficient mice were bred and kept under specific pathogen-free conditions at the Central Animal Facilities of the German Cancer Research Center. In each experiment, cohorts of five animals were used to permit accurate comparison among differently treated groups. Mice were given s.c. 108 Raji cells. To reduce inherent NK cell activity, the mice were irradiated (300 rad) 1 day before tumor inoculation and received i.p. injections of 25 µg mAb PK136 (anti-IL-2R) all 5 days during the whole experiment. At days 5, 12, and 20 after tumor inoculation, the animals received i.v. 5 x 106 human PBLs that were preactivated in vitro by immobilized mAb OKT3 (anti-CD3), soluble mAb 15E8 (anti-CD28), and low-dose IL-2. A few hours after each PBL injection, either PBS or 50 µg of the bispecific diabody, or the quadroma-derived BsAb OKT3 x HD37 (35), or combinations of these with 50 µg of anti-CD28 mAb 15E8, respectively, were administered i.v. Tumor size was measured using a caliper every second day. The mean diameter and SE of each group were calculated. In addition, days of sacrifice were recorded and used for estimation of survival time. Animals were followed until the s.c. tumors reached a maximal tolerated size of 15 mm in diameter and were sacrificed by cervical dislocation. The data on tumor growth were evaluated by a linear mixed-effects model fit by REML (restricted maximum likelihood) (36, 37) using the software S-Plus 3.4 (Sun SPARC station). The p values were calculated for intercepts on the y-axis (p1) and for the slopes of tumor growth curves (p2). Therefore, p1 indicates the significance of the assumed start of tumor growth, and p2 the significance of differences in tumor growth rates. The p values for differences in survival times were calculated using the log rank test. The results of a statistical test were assumed to be significant if the p value was less than 5%.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To obtain a BsAb suitable for the therapy of human B cell
malignancies, we constructed a small recombinant molecule with dual
specificity for both the human B cell surface Ag CD19 and the
signal-transducing CD3 chain of the human TCR/CD3 complex. ScFv
fragments derived from the hybridomas HD37 (38) and OKT3
(39) were used to create a bispecific CD3 x CD19
diabody (19). To increase the yield of functional
heterodimer (diabody) formed in the bacterial periplasm by cosecretion
of the two hybrid scFvs
VH3-VL19 and
VH19-VL3, we
constructed an optimized expression vector pSKID2 (Fig. 1
). The vector contains several features
that improve plasmid performance and lead to increased accumulation of
functional bivalent product in the E. coli periplasm under
conditions of both shake-flask cultivation and high cell density
fermentation. These are the hok/sok postsegregation killing
system (23), which prevents plasmid loss, strong tandem
rbs (20), and a gene encoding the periplasmic factor
Skp/OmpH that increases the functional yield of Ab fragments in
bacteria (21). The use of an improved expression system
allowed us to isolate a functional bispecific diabody with a yield of
10 mg per 1 l of shake-flask culture and a purity greater than
95% (data not shown). The size-exclusion chromatography on a
calibrated Superdex 200 column revealed a predominantly dimeric form of
the diabody with only a small proportion of putative tetramers (data
not shown).
Ag-binding affinity
Flow cytometry experiments demonstrated a specific interaction of
the diabody with both human CD19+ JOK-1 and
CD3+ Jurkat cells. However, the fluorescence
intensities obtained for interaction with JOK-1 cells were
significantly higher than for Jurkat cells, reflecting the difference
in affinity values for the two Ag binding sites (Fig. 2). The CD19- and CD3-binding affinities
of the diabody were estimated by competitive binding to human JOK-1 and
Jurkat cells in the presence of either FITC-labeled mAb HD37
(anti-CD19) or OKT3 (anti-CD3). The relative affinities were
calculated from the corresponding IC50 values
(19). The affinity of the diabody for
CD19+ target B cells was 10-fold higher than its
affinity for CD3+ effector T cells (Table I). Theoretically, the cosecretion of two
hybrid scFv fragments may give rise to two types of dimer: active
heterodimers and homodimers that are probably inactive. However,
affinity measurements indicated that the diabody was mostly, if not
completely, in the active heterodimeric form. It bound to both human
CD3 and CD19 with affinities comparable with that of the parental scFv
fragments (19). To investigate the biological relevance of
the differences in affinity values for different Ags, the in vitro
retention of the diabody on the surface of both
CD3+ and CD19+ cells at
37°C was determined by flow cytometry (Fig. 3). The diabody had a relatively short
retention t1/2 on
CD3+ Jurkat cells (3 min) and 3-fold longer
t1/2 on the surface of
CD19+ JOK-1 cells, reflecting the lower
CD3-binding affinity deduced from inhibition experiments (Table I).
|
|
|
Three different in vitro tests were used for analyzing the
biological activity of the bispecific diabody. First, its ability to
stimulate T cell proliferation when bound to the surface of tumor cells
was determined. Irradiated CD19+ Burkitt lymphoma
Raji cells were treated with different concentrations of CD3 x
CD19 diabody and subsequently incubated in the presence of human PBL
prestimulated with a low dose of IL-2. In this assay, we observed a
clear dose-dependent increase of effector cell proliferation (data not
shown). The ability of the bispecific diabody to induce tumor cell
lysis by redirecting T cell-mediated cytotoxicity was investigated
using a JAM test that is based on measuring DNA fragmentation in the
target cell as a result of apoptosis (33). The death of
CD19+ Raji cells was specifically triggered by
the diabody in a concentration-dependent manner using an E:T ratio of
10:1 (Fig. 4A). Even with a
rather low E:T ratio of 5:1, we observed a fairly high specific Raji
cell killing of 46.4% with a diabody concentration of 10 µg/ml (data
not shown). A time-course measurement revealed that the maximal plateau
of tumor cell lysis was reached after 4–5 h (data not shown). The
specificity of diabody-mediated tumor cell lysis was demonstrated by
inhibition of cytolysis in the presence of parental monospecific
anti-CD3 and anti-CD19 mAbs (Fig. 4B). Furthermore,
the CD3 x CD19 diabody was not able to mediate the lysis of
CD19- Lovo cells, and a monospecific CD19
x CD19 diabody (40) was not able to mediate the lysis of
CD19+ Raji cells (data not shown).
|
A). Several authors have
shown that T cells, redirected against tumor cells with bispecific
antitumor x anti-CD3 Abs, can be effectively costimulated
using anti-CD28 mAb or antitumor x anti-CD28 BsAb to
increase their cytolytic activity (5, 41, 42). We
therefore investigated whether the addition of anti-CD28 mAb 15E8
in a concentration of 1 µg/ml to the effector cell population could
increase target cell lysis. A human B cell line Raji expressing CD19
was used as a target. The effect of the bispecific diabody was measured
using a standard 51Cr release assay based on
plasma membrane disintegration and the consequent release of cytoplasm.
A mixture of both parental scFv fragments (anti-CD3 and
anti-CD19) served as a negative control for determining the
background lysis. As shown in Fig. 5, the
anti-CD28 mAb significantly increased the lytic capacity of
redirected PBL. Moreover, the increased cytotoxicity of the diabody in
the presence of mAb 15E8 was independent of the E:T ratio (Fig. 5).
|
One of the most important applications of BsAbs is in cancer
therapy in which factors such as plasma
t1/2 and tumor penetration play an
important role. The BsAb should therefore be stable at 37°C in human
serum to have a significant antitumor effect. One might expect that the
bispecific diabody would be rather labile, as it was formed by
noncovalent interactions of two hybrid scFv molecules. We therefore
investigated the stability of the diabody when stored at a fairly low
concentration in human serum at 37°C for prolonged periods of time.
The residual activity was estimated by flow cytometry. We found that
the bispecific diabody retained 80% of its Ag-binding activity after
24-h incubation at 37°C in human serum and had a
t1/2 of almost 3 days (Table II). Western blot analyses of the samples
of the diabody showed no proteolytic degradation after 81 h of
incubation in human serum at 37°C, indicating that the loss of
activity was mainly caused by aggregation of the diabody (data not
shown).
|
To estimate the plasma clearance pharmacokinetics of the diabody
in comparison with a smaller scFv fragment, we administered
preparations of 125I-labeled anti-human CD3
scFv monomer (30 kDa) (28) and CD3 x CD19 diabody
(58 kDa) into normal mice. Blood samples were taken at various time
points. To exclude errors in calculating plasma
t1/2 due to free radioiodine resulting
from cellular, metabolic degradation of labeled protein, we first
determined the proportion of protein-bound radioactivity for each
plasma sample by TCA precipitation. The TCA-precipitable
radioactivities of scFv decreased rapidly in the blood over time; only
50% of the radioactivity was bound to protein 30 min after the i.v.
injection. In contrast, the diabody lost its label much more slowly;
more than 50% of its radioactivity could be precipitated 3 h
after injection (data not shown). The estimated
t1/2 of the diabody was more than
three times that of the scFv (Table II). Analyses of blood samples at
different time points demonstrated an extremely rapid clearance of scFv
from plasma, with an -phase (extravasation and renal clearance)
t1/2
(t1/2) of 3.5 min and a
ß-phase (catabolism and slow release of extravasated Ab from
perivascular space back into the blood)
t1/2
(t1/2ß) of 74.6 min (Table II). The diabody had an 2-fold slower first-pass clearance and a
somewhat slower catabolism (Table II). The pharmacokinetic AUC were
calculated for both Ab fragments to provide a means of estimating the
total relative dose that would be delivered in a therapeutic
application. The diabody demonstrated the best pharmacokinetic
performance; its AUC value was more than 2-fold higher than AUC values
calculated for the scFv (Table II). The differences in the
pharmacokinetic properties of these constructs correlate quite well
with their size.
Effect of diabody on xenotransplanted B cell lymphoma
To estimate the in vivo antitumor activity of the CD3 x CD19
diabody, we established a xenotransplant model of the Raji Burkitt’s
lymphoma in RAG2-deficient mice. Cohorts of five mice were inoculated
s.c. with 108 tumor cells, and at days 5, 12, and
20 received i.v. 5 x 106 human PBLs
preactivated in vitro by immobilized mAb OKT3 (anti-CD3), soluble
mAb 15E8 (anti-CD28), and low-dose IL-2. A few hours after each PBL
inoculation, the mice were treated either with no Ab, the bispecific
diabody, or BsAb OKT3 x HD37 (35) administered as a
tail vein injection. Additional groups received BsAb in combination
with anti-CD28 mAb 15E8. All the animals in the control groups
receiving only PBS or PBL developed tumors larger than 1 cm in diameter
(0.4–0.6 cm3) within 4 wk (Fig. 6). In contrast, the BsAb OKT3 x
HD37 and diabody demonstrated tumor growth suppression till days 19 and
34, respectively (Fig. 6). Administration of both BsAb and diabody in
combination with anti-CD28 mAb further prolonged the period of
tumor suppression till days 46 and 55, respectively (Fig. 6). After
reaching a certain size, the tumors grew progressively in all animal
groups. There was no significant difference in tumor growth between two
control groups receiving either PBS or activated PBL alone
(p1 = 0.0688, p2 = 0.0696; Fig. 6), indicating that under the conditions used, any allogeneic reaction
of the effector cells toward the tumor can be ignored. There was also
no significant difference in tumor growth between the animals receiving
either the quadroma-derived BsAb OKT3 x HD37 or the CD3 x
CD19 diabody, both with or without the anti-CD28 mAb 15E8
(calculated p1 values were 0.445 and 0.5301, respectively;
p2 values were 0.5734 and 0.6164, respectively; Fig. 6). The
comparison with the control groups demonstrated no significant delay in
the starting point of tumor growth in mice receiving both the BsAb and
diabody (p1 = 0.2125), but the tumors grew
significantly slower in those groups (p2 <
0.0001). In contrast, animals receiving the BsAb or diabody in
combination with anti-CD28 mAb revealed a significant delay in the
starting point of tumor growth and slower tumor growth rates as
compared with control groups (p1 < 0.0001,
p2 < 0.0001). Finally, the use of the costimulatory
mAb 15E8 led to a delay in the starting point of tumor growth in
comparison with that of animals receiving BsAb or diabody without CD28
costimulation (p1 < 0.0001), but had no effect
on the tumor growth rate (p2 = 0.1301). The
animals were sacrificed when the tumors achieved diameters of 15 mm.
Sacrifice dates were recorded, and the mean survival in each group was
calculated. Mean survival was 37.2 ± 1.5 days in the control
group receiving PBS, 38.2 ± 1.9 in the control group receiving
human PBL alone, 71.5 ± 1.5 days in the group given BsAb
OKT3 x HD37, and 75.4 ± 2.2 days in the diabody-treated
group. The addition of the anti-CD28 mAb 15E8 resulted in a
prolonged survival time for both BsAb- and diabody-treated animals. The
calculated mean survival was 87.2 ± 1.6 days in the group given
BsAb OKT3 x HD37 plus mAb 15E8 and 89.6 ± 2.2 days for
animals receiving bispecific diabody in combination with anti-CD28
mAb. The statistical analysis of survival times demonstrated
significant differences between all the therapy groups and the control
groups (p = 0.002 in each case). In contrast,
there was no significant difference in survival between the animals
treated either with BsAb OKT3 x HD37 or diabody both with and
without CD28 costimulation (p = 0.0925 and
0.1421, respectively).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
R-positive cells
(granulocytes, macrophages) by BsAbs (43, 44, 45). However,
the progress of these immunotherapeutic agents into clinical
applications has been slow, mainly due to the low yields of clinical
grade bispecific molecules and the immunogenicity of murine BsAbs. More recently, recombinant bispecific diabodies have been created that only comprise Ab V domains (16, 17, 19). They are less immunogenic than quadroma-derived BsAb and can be easily produced in bacteria in relatively high yields.
The CD3 x CD19 diabody described in the present investigation was designed for treating of minimal residual disease in patients with B cell malignancies. It bound to target B cells more strongly than to effector T cells, due to the almost 10-fold higher affinity of the anti-CD19 moiety. Such relatively strong binding to a target tumor cell and weaker binding to an effector cell may have certain advantages for tumor therapy. Artificial signaling via the CD3 Ag mimics the physiological Ag-specific activation of T lymphocytes by MHC-bound Ag. Accordingly, in a model of TCR serial triggering (46), a high off-rate of the TCR is essential because it allows a single peptide-MHC complex to engage many TCRs in successive rounds of ligation, triggering, and dissociation. Therefore, a high affinity due to a low off-rate may inhibit TCR reuse.
As expected, the diabody had lower clearance rates compared with the
smaller scFv Ab fragment. It proved to be fairly stable both in vitro
and in vivo, although it is formed by the noncovalent association of
two hybrid scFv fragments. A comparison of the diabody
t1/2 under different experimental
conditions (Table II) indicated that the therapeutic dose will be
mainly determined by its clearance from the bloodstream, and not by
degradation.
In vitro experiments demonstrated that the CD3 x CD19 diabody was able to activate T cells in the presence of irradiated tumor cells. It was also able to effectively recruit T cells for killing both CD19+ Raji cells and fresh CLL leukemia cells from a patient. The cytolytic potential of the diabody was enhanced in the presence of anti-CD28 mAb, which induces a well-known costimulatory signal (for review, see Ref. 47). T cell signaling via the TCR alone without a costimulatory signal can lead to anergy (48) or even activation-induced cell death (49). Furthermore, recent experiments demonstrated that costimulation via CD28 together with anti-CD3 x antitumor BsAb prevents apoptosis in the targeted T cells (50) and even reactivates exhausted tumor-reactive CTL in an in vivo model of human melanoma (42). Impressive results were also obtained using anti-CD28 x antitumor BsAb together with anti-CD3 x antitumor BsAb for curing human primary tumors (43) and metastases (42) in immunodeficient mice. However, it is somewhat surprising that the anti-CD28 mAb improves the efficacy of B cell lysis in our system because the Raji B cells carry both costimulatory molecules CD80 and CD86. In vitro, the anti-CD28 mAb in solution probably saturates the CD28 molecules on the effector cells faster and more efficiently than the cell-bound B7 of the target cells. A similar situation may occur in our in vivo model, in which the circulating anti-CD28 mAb can bind to the CD28 molecule more effectively than the B7 present only at the tumor site.
The ultimate goal of any antitumor immunotherapy is the in vivo eradication of tumor cells. Recently, we described a CD30 x CD16 diabody that was able to induce a marked regression of xenotransplanted human Hodgkin’s lymphoma in SCID mice due to the recruitment of human NK cells (51). In the present study, we tested the potency of a CD3 x CD19 diabody to mediate T cell-dependent tumor lysis in a fairly stringent in vivo model of immunodeficient mice bearing a s.c. growing human B cell lymphoma. The number of Raji cells used for s.c. tumor growth was much higher than in other studies (50), and we let the malignant cells grow for 5 days before starting treatment. A prolonged T cell-dependent suppression of tumor growth was induced by the diabody. Mice receiving the diabody had a longer mean survival time twice as long as the control animals. The administration of the diabody together with the anti-CD28 mAb further prolonged survival. The comparable efficiency of the diabody compared with the quadroma-derived BsAb was very surprising in view of its relatively rapid clearance through the kidneys. This is probably compensated by a better tumor penetration and more efficient induction of cell lysis.
The successful use of a recombinant anti-Id x anti-mouse CD3 scFv-scFv tandem ((scFv)2) for the in vivo treatment of a murine B cell lymphoma has recently been described (52). However, the yield of correctly folded (scFv)2 molecules is generally quite low, at least in the periplasm of bacteria (19). Therefore, for our work, we chose to make a bispecific diabody, which is formed by the noncovalent association of two scFvs and has a stable compact structure (18). To our knowledge, these results describe the first in vivo application of bispecific diabodies to redirect human cytotoxic T cells. The efficacy of the diabody’s immune recruiting capacity suggests that it can be used to replace the bispecific mAb for immunotherapy. This would facilitate larger clinical trials with more extensive cycles of treatments, because it is less immunogenic and can probably be produced and purified at relatively low cost.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 B.C. and S.M.K. contributed equally to this work.
3 Current address: Affimed Therapeutics AG, Ladenburg, Germany.
4 Address correspondence and reprint requests to Dr. M. Little, Affimed Therapeutics AG, Dr. Albert-Reimann-Strasse 2, D-68526 Ladenburg, Germany.
5 Abbreviations used in this paper: BsAb, bispecific Ab; AUC, area under curve; CLL, chronic lymphatic leukemia; RAG, recombination-activating gene; rbs, ribosome binding site; scFv, single-chain variable fragment of Ab.
Received for publication February 28, 2000. Accepted for publication April 26, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
)2 antibody containing thioether-linked Fab' fragments. J. Immunol. 139:2367.[Abstract]
. I. OR3 and autogenous negative control by repressor. J. Mol. Biol. 139:147.[Medline]
R bispecific antibodies and granulocyte colony-stimulating factor-primed neutrophils as effector cells against HER-2/neu overexpressing breast cancer. Cancer Res. 57:696.This article has been cited by other articles:
|
|
 |
|
  D. Muller, A. Karle, B. Meissburger, I. Hofig, R. Stork, and R. E. Kontermann Improved Pharmacokinetics of Recombinant Bispecific Antibody Molecules by Fusion to Human Serum Albumin J. Biol. Chem., April 27, 2007; 282(17): 12650 - 12660. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Yan, M. J. Wolff, J. Unternaehrer, I. Mellman, and M. J. Mamula Targeting antigen to CD19 on B cells efficiently activates T cells Int. Immunol., July 1, 2005; 17(7): 869 - 877. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Le Gall, U. Reusch, M. Little, and S. M. Kipriyanov Effect of linker sequences between the antibody variable domains on the formation, stability and biological activity of a bispecific tandem diabody Protein Eng. Des. Sel., April 1, 2004; 17(4): 357 - 366. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Dreier, P. A. Baeuerle, I. Fichtner, M. Grun, B. Schlereth, G. Lorenczewski, P. Kufer, R. Lutterbuse, G. Riethmuller, P. Gjorstrup, et al. T Cell Costimulus-Independent and Very Efficacious Inhibition of Tumor Growth in Mice Bearing Subcutaneous or Leukemic Human B Cell Lymphoma Xenografts by a CD19-/CD3- Bispecific Single-Chain Antibody Construct J. Immunol., April 15, 2003; 170(8): 4397 - 4402. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. M. Kipriyanov, B. Cochlovius, H. J. Schafer, G. Moldenhauer, A. Bahre, F. Le Gall, S. Knackmuss, and M. Little Synergistic Antitumor Effect of Bispecific CD19 x CD3 and CD19 x CD16 Diabodies in a Preclinical Model of Non-Hodgkin's Lymphoma J. Immunol., July 1, 2002; 169(1): 137 - 144. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. A. Cavacini, M. Duval, J. P. Eder, and M. R. Posner Evidence of Determinant Spreading in the Antibody Responses to Prostate Cell Surface Antigens in Patients Immunized with Prostate-specific Antigen Clin. Cancer Res., February 1, 2002; 8(2): 368 - 373. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Cochlovius, S. M. Kipriyanov, M. J. J. G. Stassar, J. Schuhmacher, A. Benner, G. Moldenhauer, and M. Little Cure of Burkitt's Lymphoma in Severe Combined Immunodeficiency Mice by T Cells, Tetravalent CD3 CD19 Tandem Diabody, and CD28 Costimulation Cancer Res., August 1, 2000; 60(16): 4336 - 4341. [Abstract] [Full Text]
|
|
| ||||||||||||
