| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Immunology |
Departments of 1 Pathology and 2 Surgery, Leiden University Medical Center, Leiden, The Netherlands, and 3 Department of Molecular Biology, Nagoya City University, Graduate School of Medical Science, Nagoya, Japan
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In humans, there are three important mCRP expressed by tumor cells. CD46 is a cofactor for factor I-mediated cleavage of C3b and C4b, and CD55 accelerates the decay of C3 and C5 convertases (8) . Both CD46 and CD55 inhibit the formation of the chemoattractants C3a and C5a and the deposition of C3b (further converted to iC3b) on the cell surface, thereby preventing attraction and activation of effector cells expressing Fc and complement receptors and subsequent complement-mediated cellular cytotoxicity (9) . CD59 inhibits the formation of the membrane attack complex, thereby preventing direct complement-mediated lysis (8) . In addition to the above-described mCRP, rodents express an additional C3 regulatory protein Crry/p65 (Crry), whereas CD46 expression is restricted to the testes. Crry is a functional homologue of CD46 and CR1 (10) , thus acting at the level of C3, similar to human CD46 and CD55. Crry has been shown to be the most important C3 regulatory protein on tumor cells in rats (11 , 12) . Therefore, modulation of Crry on tumor cells is a suitable model for the modulation of human membrane-bound C3 regulatory proteins.
Previously, it has been shown that systemic inhibition of Crry by i.p. injection of anti-Crry mAbs in rats results in severe side effects, including endothelial damage, peritoneal hemorrhage, and death (13) , as a result of unhampered massive complement activation, due to the ubiquitous expression of Crry. Therefore, when therapeutically inhibiting Crry on tumor cells, anti-Crry mAbs cannot be administered systemically, and manipulation should be primarily restricted to the tumor cells. To obtain this goal, we have investigated the feasibility of using bispecific mAbs (bi-mAbs) that both recognize a tumor antigen and Crry in a syngeneic rat model for colorectal cancer.
Previously, we have described that bi-mAb directed against Ep-CAM and CD55 as well as anti-Ep-CAM-CVF conjugates causes increased amounts of C3 deposition on human colorectal tumor cells in vitro as compared with the anti-Ep-CAM mAb alone (14) . In the present study, to follow up on these in vitro studies, we have investigated the effect of blocking or overwhelming the effect of Crry on in vitro opsonized rat colon carcinoma cells in a syngeneic rat colorectal cancer lung metastases model. Blocking the function of Crry was achieved by using bi-mAbs, which possessed an arm recognizing a rat tumor antigen for tumor-specific homing and an arm that blocked Crry (15 , 16) . Overwhelming the function of Crry was achieved by tumor-specific targeting of cobra venom factor (CVF) with mAb-CVF conjugates (17) . Control bi-mAb and anti-Crry F(ab)2 fragments were used to show that the observed effect of bi-mAbs was indeed due to the modulation of Crry. Finally, to mimic adjuvant mAb immunotherapy a prophylactic experiment was performed, to substantiate the in vivo observations in a therapeutic setting.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Quadroma cells producing bi-mAb MG42a*5I2 and MG42a*CC52 were obtained by fusion of the respective hybridomas (18 , 19) and tested and purified as described before (16) : The fractions that were eluted from protein A with pH 5.5 were tested for bi-isotypicity with a sandwich ELISA. Fractions were incubated on a plate coated with rat antimouse IgG1 and detected with goat antimouse IgG2a conjugated with horseradish peroxidase as described before. Positive fractions were pooled and tested for purity and bi-isotypicity with SDS-PAGE on 7.5% gels under nonreducing conditions.
Bispecificity of bi-isotypic mAb was tested with a conjugation test: CC531 cells preincubated with 5I2 or CC52 F(ab)2 fragments were opsonized with bi-mAbs and respectively exposed to LT12 cells that expressed Crry but no MG4 or to CC531 cells opsonized with MG4 F(ab)2 fragments. Whereas approximately 25% of cross-linking of these cells was observed with flow cytometry, bi-mAbs were considered to contain mAbs that recognized the two antigens (16) .
Syngeneic normal rat serum was prepared by centrifugation on 4°C of coagulated blood of WAG/Rij rats.
F(ab)2 Fragments.
F(ab)2 fragments were produced as previously described previously (20)
. In short, 9 units of mAb were incubated with 1 unit of 1 M citrate and 2 units of 0.1 M HCl together with 100 units of pepsin-agarose (Sigma-Aldrich, Zwijndrecht, The Netherlands) for 2 h at 37°. The reaction was stopped with an equal volume of 3 M Tris. The solution was centrifuged for 5 min at 7,000 x g, and the F(ab)2 containing supernatant was dialyzed against PBS. No fragments other than the F(ab)2 fragments were detected in this preparation by SDS-PAGE.
Tumor Cells.
The CC531 cell line (21)
was cultured in RPMI 1640 culture, supplemented with 10% (v/v) heat-inactivated fetal bovine serum, 2 mM glutamine, 50 µg/ml streptomycin, and 50 units/ml penicillin (all from Life Technologies, Paisley, United Kingdom). Tumor cells used for in vitro experiments were kept in culture for several passages. Cells used in vivo were all from the same passage and always cultured according to a standard schedule according to which cells were passaged every 4 days and fresh medium was supplied every other day for 2 weeks.
mAb-CVF Conjugates.
CVF was isolated and tested for hemolytic activity as described before as was the conjugation of CVF to mAb (14
, 17)
. In short, 5 mg of antibody and 2.5 mg of CVF were incubated with N-succinimidyl-3(2-pyridyldithio)proprionate (Sigma Aldrich). mAb-N-succinimidyl-3(2-pyridyldithio)proprionate was reduced with DTT and allowed to form conjugates with CVF-N-succinimidyl-3(2-pyridyldithio)proprionate overnight at room temperature. Conjugates were isolated by gel filtration (S300 column; Pharmacia), and positive fractions were pooled. Purity of the conjugates was determined by SDS-PAGE on a 7.5% gel under nonreducing conditions and by high-pressure liquid chromatography with a superdex200 column (Pharmacia). Only fractions containing mAb:CVF at a ratio of 1:1 were used in the experiments. Protein concentrations were determined using a spectrophotometer using E280 = 1.2 for 1 mg/ml mAb conjugates (17)
.
Complement Activation ELISA.
Conjugate (20 µg/ml) or mAb (10 µg/ml) was coated in 0.1 M NaHCO3 buffer (pH 9.6) for 2 h at 37°C on a 96-well plate. After washing three times with PBS containing 0.01% Tween 20, nonspecific binding was blocked by incubating the wells for 30 min with 1% gelatin in PBS/Tween. After washing three times with PBS without Tween, different syngeneic normal rat serum dilutions, starting with 10%, in gelatin veronal buffered saline containing 2 mM CaCl2 and 2 mM MgCl2 (GVB++) were added as a source of rat complement and incubated for 30 min at 37°C. When the syngeneic normal rat serum was washed away, C3 deposition was detected with mouse antirat C3 conjugated to digoxigenin (ED11, a kind gift of Dr. A. Roos, Leiden University Medical Center, Leiden), diluted in PBS containing 1% BSA and incubated for 30 min. Anti-digoxigenin-horseradish peroxidase (Roche Diagnostics, Mannheim, Germany) diluted in PBS/1% BSA was added, after washing with PBS/Tween, and incubated for 1 h. The assay was developed with 1% 3,5,3',5'-tetramethyl-benzidine in 0.11 M Na-acetate (pH 5.5; 100 µg/ml) containing 0.1% freshly added H2O2. The reaction was stopped after 10 min with 50 µl of 2 M H2SO4. OD450 was determined as a measure for the amount of deposited C3.
Flow Cytometry.
CC531 cells (2.5 x 105) were incubated with mAbs in 100 µl of PBS, containing 1% BSA (fraction V; Sigma, St. Louis, MO; PBS/BSA,) for 30 min on ice. After washing the cells twice with PBS/BSA, 100 µl of syngeneic normal rat serum (10% v/v; final dilution) in RPMI 1640 were added as a source of complement and the cells were incubated for 15 min at 37°C. To analyze the binding of IgG or C3 deposition, the cells were washed twice with PBS/BSA and incubated with FITC-conjugated goat antimouse IgG or FITC-conjugated goat antirat C3, respectively (30 min on ice). After washing, propidium iodide (1 µg/ml) was added to visualize dead cells. Ten thousand living cells were examined using a flow cytometer (FACScalibur; Becton Dickinson, Mountain View, CA). The average amount of C3 deposition was determined from all living cells. Fluorescence compensation was used to correct for spectral cross-talk between fluorescent signals. Data are expressed as molecules of equivalent soluble fluorochrome values, to compare individual experiments independent from flow cytometer settings (22
, 23)
. To calculate molecules of equivalent soluble fluorochrome values a flow cytometry standardization kit, Quantum 25 FITC beads (Flow Cytometry Standards Europe, Leiden, The Netherlands) were measured with each experiment for calibration of FL1 measurements.
Hemolytic Activity.
The hemolytic activity (CH50) was determined by measuring the serum concentration and thus the activity of classical complement pathway to achieve a 50% reduction in lysis of sheep erythrocytes opsonized with rabbit Ab. This was achieved by using 100 µl (5 x 108/ml) of sheep erythrocytes coated with rabbit antigoat Ab and 100 µl of a 1/100 dilution of serum of treated rats, both diluted in D-glucose gelatin veronal buffer containing 0.5 mM MgCl2, 0.15 mM MgCl2, and 0.1 M Mg-EGTA (pH 7). After addition of 1.5 ml of PBS and centrifugation for 7 min at 2,000 rpm, the amount of released hemoglobin was measured at A412.
Animal Experiments.
Male WAG/Rij rats, a Wistar-derived inbred strain, were purchased from Charles River (Schutzfeld, Germany) and used at an average weight of 230 g. The animals had free access to standard food pellets and water. Animal experiment protocols were approved by the local university animal ethical committee. Tumor cells were harvested with a solution of 0.1% (w/v) EDTA and 0.25% (w/v) trypsin in Hanks’ buffered salt solution. After washing twice in PBS, 30 x 106 cells were incubated with 4 ml of PBS containing either mAb (150 µg/100 pmol), F(ab)2 (100 µg/100 pmol), bi-mAb (400 µg/269 pmol), or mAb conjugate (300 µg/100 pmol) for 1 h at 4°C. The ratio of mAbs and cells was comparable with the flow cytometry protocol. After incubation, cells were washed three times with PBS, adjusted to a concentration of 20 x 106 cells/ml, and kept in suspension on ice until use while occasionally shaking. Rats were anesthetized with halothane, and 200 µl of tumor cell suspension, containing 4 x 106 cells, were injected into the penile vein. In the prophylactic experiment, rats at t = –6 h received injections i.p. of 300 µg (200 pmol) of bi-mAb or 150 µg (100 pmol) of MG4 (amounts corrected for comparable binding sites to MG4) in 500 µl PBS. At t = 0 h, 4 x 106 tumor cells were injected in the penile vein. At t = 6 h, a second comparable dose of mAb was administered. Blood was drawn from the tail vein at t = 0 and t = 18 to collect peripheral blood lymphocytes and serum. At day 21, rats were sacrificed by abdominal bleeding under halothane anesthesia. Lungs received injections of 15% Indian ink (Royal Talens, Apeldoorn, The Netherlands) in water via the trachea until completely filled. Lungs were removed, washed for 5 min in water, and fixed in Fekete’s solution containing 60% ethanol, 9% formaldehyde solution (4%), and 4.5% acetic acid (24)
. After 24 h of fixation, lungs were rinsed and stored in water. All macroscopically visible tumors on the lung surface were counted for each pair of lungs by an observer blinded for the treatment of the tumor cells or rats. The number of metastases was previously shown to correlate well with the total number of tumors present in the lungs (24
, 25) . Each experiment consisted of five rats per group and was repeated three or four times (total n = 12–17/group in total). For the therapeutic experiment, n = 6 rats/group. Differences between groups were statistically analyzed using a Mann-Whitney test. Two of seven bi-mAb-injected rats were sacrificed 2 days after mAb injection to investigate biodistribution of the bi-mAb. Sections of liver, lung, colon, and skin were stained with goat antimouse-FITC, and staining levels were analyzed by confocal microscopy as described before (11)
.
Statistical Analysis.
Data are given as group means ± SD. Differences in group means were tested for significance using Mann-Whitney U test, considering P < 0.05 significant.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Using flow cytometry, opsonization with MG42a-CVF increased the amount of C3 deposition on CC531 cells by 100% compared with MG42a. The amount of C3 deposition induced by MG42a-CVF could still be increased another 300% compared with MG42a-CVF by simultaneously blocking Crry with 5I2 F(ab)2 fragments, indicating the important role of Crry in controlling complement activation at the cell surface (Fig. 1B)
.
|
. MG42a*5I2 and the two parental mAbs [MG42a and 5I2 F(ab)2] were similar in their complement-activating capacities as measured by flow cytometry. F(ab)2 of 5I2 were used, because the Fc part of 5I2 could, despite its IgG1 isotype, activate the rat complement system via the classical pathway (Ref. 26
; data not shown). That the beneficial effect of bi-mAb or MG4 + 5I2 was dependent on the complement-activating MG4 antitumor arm was supported by the observation that 5I2 F(ab)2 in combination with non-complement-activating MG4 IgG1 did not increase C3 deposition compared with MG42a alone. C3 deposition induced by bi-mAb or by both parental mAbs [MG4 IgG2a + 5I2 F(ab)2], however, was significantly higher compared with MG42a alone (Fig. 1B
, P = 0.05).
To investigate whether the observed increase in C3 deposition was due to blocking of Crry and not the effect of a higher number of Fc tails or a higher efficiency to activate complement of the mixed IgG1/2a Fc tail, a control bi-mAb directed against the MG42a and CC52 antigens was developed. CC52 is an antigen expressed on CC531 cells, with expression levels comparable with Crry (18)
. Opsonization of CC531 with MG42a*CC52 resulted in comparable amounts of C3 deposition as opsonization with MG42a (Fig. 1B)
, indicating that the high level of C3 deposition induced by MG42a*5I2 was indeed due to the inhibition of Crry. In addition to its functional capacity, bi-isotypicity and purity of the bi-isotypic mAb was confirmed by SDS-PAGE (Fig. 1C)
.
Effect of Modulating mCRP on Outgrowth of Lung Tumors.
Subsequently, the effect of modulating Crry on mAb-mediated immunotherapy was investigated in a syngeneic rat colorectal cancer lung metastases model. CC531 cells were preincubated with MG42a, bi-mAb MG42a*5I2, or MG42a-CVF conjugates and injected i.v. in rats (Fig. 2)
. As a reference, PBS-incubated CC531 tumor cells were injected, resulting in outgrowth of 291 tumors on average at the lung surface. Tumor cells preincubated with MG42a resulted in 175 lung surface tumors on average, which was not significantly different from the PBS group (Mann-Whitney, P = 0.18). Blocking Crry with bi-mAb MG42a*5I2 completely blocked tumor outgrowth in 33% of the rats over all experiments. In these animals, no tumors could be found on the lung surface or inside the lungs. In the other animals, only a few tumors were observed (ranging from 1 to 7 lung surface tumors). This was significantly less compared with the outgrowth of MG42a-incubated tumor cells (P = 0.008). Treatment with MG42a-CVF also resulted in a significantly reduced number of tumors at the lung surface (44 on average) compared with MG42a (P = 0.014), although treatment with bi-mAb was significantly more effective (P = 0.013). No visible tumors were found in other organs than the lungs.
|
, indicating that the MG4-binding part of the bi-mAb is essential for an optimal reduction of the number of metastases. Tumor cells incubated with control CC52 F(ab)2 induced 344 tumors on average, which was significantly more (P = 0.009). The latter number was comparable with the PBS control group (353 tumors on average; P = 0.917), suggesting that inhibition of Crry was also responsible for the observed effect, probably by allowing unhampered activation via the alternative pathway tick-over mechanism. Injection of CC531 cells incubated with both the parental mAbs [MG42a and 5I2 F(ab)2] resulted in outgrowth of 70 surface tumors, a significantly higher number compared with bi-mAb (P = 0.013), implicating a synergistic effect of bi-mAb compared with both of the parental mAbs (Fig. 3)
.
|
. The number of tumors observed after treatment with MG42a*5I2 (6 on average in this particular experiment) was significantly less (P = 0.014) than the number of tumors induced by MG42a*CC52-treated tumor cells (244 tumors on average). The latter number was comparable with the number of tumors on the lungs of rats that received injections of MG42a-incubated cells (228 tumors on average; P = 0.602). These experiments demonstrate that the effect of bi-mAb MG42a*5I2 was indeed due to its inhibitory effect on Crry function.
Effect of Prophylactic Therapy with bi-mAb.
Because MG42a*5I2 was most effective in inhibiting outgrowth of metastases compared with MG4-CVF, the effect of this bi-mAb on tumor outgrowth was further investigated. Clinical mAb therapy in patients will probably be most successful in an adjuvant setting (25
, 27)
; for this reason, a prophylactic experiment was performed. Rats received injections of MG42a*5I2 (300 µg/200 pmol), MG42a*CC52 (300 µg/200 pmol), MG42a (150 µg/100 pmol; equal number of MG4-binding sites between bi-mAb and mAb), or PBS 6 h before tumor cell injection and received injections of a similar dose 6 h after tumor cell injection. Control mAbs only binding to Crry [e.g., 5I2 F(ab)2] were not injected, because they were previously shown to evoke severe side effects or death when injected systemically. After 3 weeks, rats were sacrificed, and the number of metastases was counted. Similar results were obtained as observed in the preincubation experiments (Fig. 4)
. Injection of MG42a*5I2 i.p. did not visually affect the health of the rats and resulted in significantly less metastases than treatment with MG42a (18 versus 353 on average; P = 0.02). MG42a was less effective in this experimental set-up compared with the preincubation experiments. This resulted in comparable numbers of metastases in the MG42a group compared with the PBS-injected group (353 versus 487; P = 0.368). Although MG42a*CC52 showed a trend to be more effective than MG42a in preventing outgrowth (175 versus 353 tumors), the difference in the number of surface tumors was not significant (P = 0.053).
|
were readily detectable. Twelve h after the last mAb injection (t = 18 h), all mAbs could be detected in the serum at comparable levels (Fig. 5, A and B)
. In addition, binding to host tissues was investigated. Two days after injection, binding of bi-mAb could be demonstrated to liver endothelium (Fig. 5D)
. No binding was observed on lung, kidney, and skin tissue (data not shown). In the intestine, only infiltrating immune cells were observed to bind MG42a*5I2. Also, peripheral blood lymphocytes were shown to be positive for MG42a*5I2 by flow cytometry (data not shown). Biodistribution of MG42a has been described previously (11
, 18)
. MG4 was only shown to weakly bind to cells of the epithelial crypts in the intestine. To investigate whether injection of tumor cells in MG42a*5I2-injected rats or binding of MG42a*5I2 to host tissue caused complement activation, a CH50 of the serum of these rats at t = 0 h and t = 18 h was performed (Fig. 5C)
. In contrast to rats that received injections of MG42a or MG42a*CC52, the hemolytic activity of the serum of MG42a*5I2-injected rats was shown to be decreased 18 h after tumor cell injection. These results show that i.p. injection of MG42a*5I2 followed by injection of tumor cells results in complement activation, suggesting that the mechanism by which MG42a*5I2 eradicates tumor cells is indeed mediated via complement activation and inhibition of Crry.
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Because an important immune effector mechanism of therapeutic anticancer mAb is activation of the complement system, leading to direct lysis and complement-dependent cellular cytotoxicity, expression of mCRP on tumors is likely to inhibit therapeutic effects. This has previously been confirmed in vitro (3 , 14 , 28) . Predominantly mouse xenograft models have been used in the past to evaluate the effect of immunotherapeutic mAb in vivo, leading to an overestimation of the effect (5 , 29) . Previously, we have described a syngeneic rat colorectal cancer model to study the role of mCRP expression on tumor cells on mAb-mediated immunotherapy (11) . In that study, it was shown that Crry was the most important mCRP on CC531 cells when determining complement activation in vitro, induced by MG42a. When MG42a mAbs were injected in rats, bearing established CC531 liver tumors, no C3 deposition could be observed on these tumor cells in vivo (11) . This suggests that mCRP indeed inhibit complement activation of therapeutic mAbs in vivo. In the present study, we have expanded these previous observations by showing that Crry on these tumor cells prevents efficient eradication of mAb-opsonized tumor cells in vivo. Therefore it is important to inhibit or overcome the function of Crry in vivo to increase the effect of mAb-mediated immunotherapy. Previously, it has been shown that systemic complement activation after injection of mAb that inhibit Crry led to severe side effects in these rats (13 , 30) . This observation demonstrated the necessity to inhibit or overwhelm Crry in a tumor-restricted fashion (4) . This can be achieved either by using bi-mAb, directed against a tumor antigen and Crry or mAb-CVF conjugates. In vitro, opsonization of either human renal cell carcinoma or colorectal carcinoma cells with the appropriate bi-mAb (G250*anti-CD55 and anti-Ep-CAM*anti-CD55, respectively) mediated up to four times the amount of C3 deposition on these cells compared with a monospecific antitumor mAb (14 , 16) . Also, opsonization of colorectal cancer cells or neuroblastoma cells with anti-Ep-CAM-CVF and anti-GD2-CVF conjugates, respectively, resulted in vitro in increased amounts of complement activation compared with the unconjugated mAb (14 , 17) . In accordance with these in vitro studies, in the present study, it was demonstrated that in vivo inhibition of a major membrane-bound C3 regulatory protein (Crry) on tumor cells using either MG42a-CVF or bi-mAb MG42a*5I2 also substantially improved the outcome of immunotherapy compared with single mAb therapy.
The experimental conditions of this proof-of-principle study, using preincubation of the tumor cells, allowed an accurate determination of the effect of bi-mAb on tumor outgrowth. It was demonstrated that preincubation of CC531 cells with MG42a*5I2 bi-mAb could prevent tumor outgrowth, in contrast to control bi-mAb directed against MG42a and an irrelevant antigen in this model. However, under these experimental conditions no cross-reactivity of bi-mAb with Crry expressed on normal cells could occur. Because it was previously shown that injection of 1 mg of 5I2 led to 50% mortality in 4–8 h (13) , we also have investigated the effect of prophylactic treatment with MG42a*5I2. In contrast to the harmful effects of monospecific anti-Crry mAb or F(ab)2, MG4*5I2 was shown to be safe when injected i.p. in rats. The pharmacokinetics of bi-mAb were slower compared with MG4 parental mAb because bi-mAb binds to Crry on normal cells, in contrast to MG4. Despite binding to endothelium and blood cells, reflected by this undetectable serum concentration at t = 0 h of the bi-mAb, rats did not show any adverse consequences of this treatment, which supports the potential immunotherapeutic applicability of appropriate bi-mAb. The significant decrease in tumor outgrowth indicates that Crry can indeed be inhibited in a tumor-specific fashion with bi-mAb directed against a tumor antigen and a mCRP. We previously observed similar homing capacities and patterns of MG42a*5I2 compared with the parental MG42a mAb in a syngeneic solid liver tumor model, which supports an important role for the antitumor arm in accumulating the bi-mAb at the tumor cell surface and suggests potential applicability in a solid tumor model (data not shown; Ref. 11 ). Injection of MG42a-CVF in the syngeneic lung metastases model did not lead to detectable side effects, although bi-mAbs were more effective in complement activation in vitro and in reducing the number of outgrowing metastases in vivo compared with the mAb-CVF conjugates.
In a clinical setting, it might be expected that an anti-CVF response will be generated (31) , which can hamper the therapeutic effect on the long term. This problem may be circumvented by conjugation of a factor I-insensitive active part of C3b to the antitumor mAb (32) . Less immunogenicity can, however, be expected of bi-mAb compared with mAb-CVF. These arguments imply that inhibition of mCRP using bi-mAb will be the preferable immunotherapeutic approach. Ideally for clinical use, the tumor antigen-recognizing arm should be of high affinity, and the mCRP blocking arm should be of low affinity to maximize tumor specificity and minimize a systemic cross-reactivity and thus avoid complement-mediated toxicity (15) .
In conclusion, with this proof-of-principle study, which is to our knowledge the first in vivo study in a model syngeneic for mCRP and complement investigating the effect of a membrane-bound C3 regulatory protein (Crry) on mAb-mediated immunotherapy, we show that tumor-specific inhibition of Crry with bi-mAb or overwhelming the function of Crry with antitumor mAb-CVF conjugates (albeit to a lesser degree) is much more efficient in tumor eradication than conventional mAb immunotherapy in this colorectal cancer lung metastases model. Bi-mAbs, blocking the most important C3 regulatory protein specifically on tumor cells, are a promising approach to increase the clinical outcome of mAb-mediated immunotherapy.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Requests for reprints: Arko Gorter, Department of Pathology, L1-Q, Leiden University Medical Center, P.O. Box 9600, 2300 RC Leiden, The Netherlands. Phone: 31-71-526-6631; Fax: 31-71-5248158; E-mail: A.Gorter{at}lumc.nl
Received 7/16/03. Revised 2/18/04. Accepted 4/ 7/04.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Mß2-integrin glycoprotein. Crit Rev Immunol, 20: 197-222, 2000.[Medline]
-galactose antibody production in baboons: implications for pig-to-human xenotransplantation. Transplantation, 62: 678-81, 1996.[Medline]
This article has been cited by other articles:
|
|
 |
|
  C. Roesli, B. Borgia, C. Schliemann, M. Gunthert, H. Wunderli-Allenspach, R. Giavazzi, and D. Neri Comparative Analysis of the Membrane Proteome of Closely Related Metastatic and Nonmetastatic Tumor Cells Cancer Res., July 1, 2009; 69(13): 5406 - 5414. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. C. Varela, M. Imai, C. Atkinson, R. Ohta, M. Rapisardo, and S. Tomlinson Modulation of Protective T Cell Immunity by Complement Inhibitor Expression on Tumor Cells Cancer Res., August 15, 2008; 68(16): 6734 - 6742. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Imai, R. Ohta, J. C. Varela, H. Song, and S. Tomlinson Enhancement of Antibody-Dependent Mechanisms of Tumor Cell Lysis by a Targeted Activator of Complement Cancer Res., October 1, 2007; 67(19): 9535 - 9541. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Li, D. J. Allendorf, R. Hansen, J. Marroquin, D. E. Cramer, C. L. Harris, and J. Yan Combined Yeast {beta}-Glucan and Antitumor Monoclonal Antibody Therapy Requires C5a-Mediated Neutrophil Chemotaxis via Regulation of Decay-Accelerating Factor CD55 Cancer Res., August 1, 2007; 67(15): 7421 - 7430. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
) was more efficient in complement activation compared with MG42a (
). A representative experiment is shown. B, C3 deposition on CC531 cells due to bi-mAb and mAb-CVF conjugates. CC531 cells were incubated with the different mAbs as depicted on the X axis. 10% WAG/Rij rat serum in RPMI++ was added as a source of complement. The amount of deposited C3 was detected with goat antirat-C3-FITC and quantified by flow cytometry. The amount of C3 deposition is expressed in molecules of equivalent soluble fluorochrome (MESF) values to correct for interexperimental differences in settings and depicted on the Y axis. Y-axes values multiplied with 104 render actual values. The value of the control (no primary mAb) was subtracted from the presented values. Means ± SD of four independent experiments are shown. C, Both bi-mAb MG42a*5I2 and MG42a*CC52 migrate exactly in between their parental mAbs on SDS-PAGE, indicating that they consist of mixed IgG1 and IgG2a Fc tails and thus are bi-isotypic. Neither of the bi-mAb isolations were contaminated with parental mAb.
