The chimeric antiganglioside GM2 monoclonal antibody (MAb) KM966, which showed high effector functions such as complement-dependent cytotoxicity and antibody-dependent cellular cytotoxicity (ADCC), potently suppressed growth and metastases of GM2-positive human cancer cells inoculated into mice. To further improve the therapeutic efficacy of the anti-GM2 MAb in humans, we constructed a humanized anti-GM2 MAb, KM8969. The humanized KM8969 was more efficient in supporting ADCC against GM2-positive human cancer cell lines than the chimeric KM966, whereas complement-dependent cytotoxicity was slightly reduced in the humanized KM8969. In addition, the humanized KM8969 was shown to exert potent ADCC mediated by both lymphocytes and monocytes. To investigate the effect of the humanized KM8969 on the biological function of GM2 in the condition physiologically mimicking formation and growth of cancer masses, the heterospheroids composed of normal human dermal fibroblasts and GM2-positive human lung cancer cells were developed. Interestingly, the humanized KM8969 gave rise to growth inhibition of heterospheroids without dependence of the effector functions. Morphological and immunocytochemical analysis suggested that the inhibitory effect was due to the apoptosis of GM2-positive cancer cells in the heterospheroids. The result indicates that GM2 captured by the antibody on the cell surface loses its physiological function that plays a critical role in maintaining the three-dimensional growth of cancer cells in contact with its own cells or other type of cells in a microenvironment. The humanized KM8969, which can destroy the cancer cells via blocking functional GM2 on the cell surface as well as the effector functions, would have extraordinary potential in human cancer therapy.

Gangliosides, which constitute a class of cell membrane constituent glycolipids, are molecules composed of a carbohydrate chain with sialic acid at the cell surface and a hydrophobic ceramide in the lipid bilayers (1). It has been known that quantitative and qualitative changes occur in the expression of gangliosides through the oncogenic transformation of cells (2). Gangliosides such as GD3, GM2, and GD22 are abundantly expressed on the cell surface of certain types of human cancer and have been shown to function as effective targets for passive immunotherapy with MAbs3(3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14). Several anti-GM2 IgM antibodies were produced in mice, rats, and humans (14, 15, 16, 17, 18, 19), and a human anti-GM2 IgM antibody showed some clinical responses in the treatment of melanoma (20). Moreover, in clinical studies, vaccination with GM2-keyhole limpet hemocyanin in melanoma patients resulted in the extension of disease-free intervals and survival when correlated with high serum titer anti-GM2 IgM and IgG antibodies (21, 22). In view of the foregoing, GM2 is expected to be an ideal antigen for specific immunotherapy of human cancers with either vaccination or MAbs.

We have previously reported the construction of a first mouse/human chimeric anti-GM2 MAb KM966, which consists of the constant (C) region of hIgG1-κ and the variable (V) region of the murine anti-GM2 IgM MAb KM696 (12). The chimeric KM966 seemed to be therapeutically effective in inhibiting the growth of human lung cancer cells implanted s.c. into nude mice and metastases formation of human lung cancer cells inoculated i.v. into NK cell-depleted SCID mice (12, 23). Chimeric MAbs, however, may have the possibility of inducing a substantial human antimouse antibody response because one-third of the molecule is still of murine origin (24). To further reduce the immunogenicity of chimeric MAbs and to prolong the circulating half-life, humanized MAbs are generated by grafting CDRs of murine MAbs into the backbone of human FRs and C regions (25, 26). Humanized MAbs contain about 10% murine origin residues and 90% human origin residues, and, thus, theoretically further reduction of the immunogenicity is achieved compared with chimeric MAbs.

In the present article, we constructed a humanized anti-GM2 MAb, KM8969, and studied its antigen-binding affinity and effector functions such as CDC and ADCC. ADCC is considered to be the major mechanism through which cancer cells, on treatment with anticancer MAbs, are eliminated in vivo(27, 28). The antimetastatic effect of the chimeric KM966 in vivo was mainly due to an ADCC reaction mediated by macrophages in the NK cell-depleted SCID mice (23). Furthermore, the chimeric KM966 was very effective in the lysis of human lung cancer cell lines mediated by both lymphocytes and monocytes (29). These facts suggest that one potentially important mechanism for the in vivo anticancer effects of the chimeric KM966 is its ability to mediate ADCC. Therefore, we also examined whether the humanized KM8969 is effective in ADCC mediated by lymphocytes and monocytes against various human lung cancer cell lines including the cells with the characteristics of multidrug resistance or high metastasis.

Many studies have noted that gangliosides function as receptors and are involved in signal transduction and that some are involved in the process of cell adhesion (30, 31, 32, 33, 34, 35). In the past, Bjerkvig et al.(36) reported that murine anti-GM2 IgM antibodies induced necrosis in spheroids consisting of human glioma cells, which express high levels of GM2; however, there were no studies to demonstrate reproducibility in different in vitro culture systems and its cytotoxic mechanism. Multicellular spheroids have been shown to represent tissues and organs in a model in which the biological and morphological properties are maintained in conditions similar to those that exist in vivo(37, 38, 39, 40). In this study, heterospheroids were successfully developed by using a collagen-conjugated thermo-responsive polymer, PNIPAAm, as a cell substratum (37, 38). This method enabled us to regulate the size and the cell composition of resultant heterospheroids and to evaluate the interaction of the cancer cells with other types of cells, including fibroblasts. We investigated, with this heterospheroid culture system, the biological effects of the humanized KM8969 on GM2-positive cancer cells in the heterospheroids composed of normal human dermal fibroblasts and human lung cancer cells.

Cell Lines.

The human SCLC SBC-3 and SBC-5 cells were kindly provided by Dr. S. Hiraki (Okayama University, Okayama, Japan). The human lung squamous cell carcinoma RERF-LC-AI cells were kindly provided by Dr. M. Akiyama (Radiation Effects Research Foundation, Hiroshima, Japan). The human SCLC H69 cells were obtained from the American Type Culture Collection (Rockville, MD). The human lung adenocarcinoma PC-14 cells were kindly provided by Dr. N. Saijo (National Cancer Institute, Tokyo, Japan). The human large cell lung cancer PC-13 cells and the human stomach adenocarcinoma MKN-28 cells were obtained from Immuno Biological Laboratory (Tokyo, Japan). Two drug-resistant sublines of SBC-3 cells were obtained by culturing the cells with gradually increasing concentrations of ADM or CDDP. After 6 months, cells that grew in 100ng/ml ADM and 400ng/ml CDDP were obtained and named SBC-3/ADM and SBC-3/CDDP, respectively (41). PC-14-PM4, a variant cell line of PC-14, with higher metastatic potential to the pleural cavity, was established by repeated in vivo selection (42). Cell cultures were maintained in RPMI 1640 supplemented with 10% heat-inactivated fetal bovine serum at 37°C in a humidified atmosphere of 5% CO2 in air. Normal human dermal fibroblasts were obtained from Kurabo Industries (Osaka, Japan). Normal human dermal fibroblasts and heterospheroids were maintained in DMEM containing 10% fetal bovine serum, 20 mm HEPES, 100 units/ml penicillin, and 100 μg/ml streptomycin at 37°C in a humidified atmosphere of 5% CO2 in air.

Antibodies.

The chimeric anti-GM2 MAb KM966 was developed in our laboratory (12). Polyclonal hIgG was obtained from Organ Teknika Corporation (West Chester, PA).4

Construction of Humanized Antibody.

A detailed process of humanization of the murine anti-GM2 MAb KM696 was described previously.4 Briefly, a series of humanized anti-GM2 MAb variants that had various mutations in the amino acid residues of their FRs were constructed based on the molecular modeling analysis of their antibody V regions; and their binding affinities to GM2 were evaluated using a transient expression system in COS-7 cells. As a result, one of the humanized MAb variants, KM8969, which revealed a high binding affinity comparable to the chimeric KM966, was selected as a candidate for humanized anti-GM2 MAb; its biological activities including cytotoxic effector functions were further characterized. Stable expression and purification of the humanized KM8969 were performed as reported previously (12). In the humanization, the IgG1-κ isotype was chosen because it is the preferred human isotype for supporting potent cytotoxic effector functions.

Flow Cytometric Analysis.

For indirect immunofluorescence, the cancer cells (1 × 106) were incubated with purified MAb (50 μg/ml) at 4°C for 1 h. The cells were washed in PBS and then incubated in FITC-labeled protein A (Boehringer Mannheim, Mannheim, Germany) at 4°C for 1 h. The cells were washed in PBS and analyzed on a flow cytometer, EPICS Elite (Coulter, Hialeah, FL). Five × 103 cells were acquired by list mode and gated by forward light scatter versus side light scatter, thereby excluding dead cells and debris. For quantitating expression levels of GM2 on the cancer cells surface, flow cytometric analysis was also done using the chimeric KM966 (29).

CDC Assay.

A CDC assay was performed as reported previously (43). Briefly, the cancer cells (5 × 106 cells) were labeled with 3.7 MBq of Na251CrO4 (51Cr) at 37°C for 1 h and kept for 30 min at 4°C to remove loosely bound 51Cr after washing. Aliquots of the labeled cancer cells were distributed into 96-well U-bottomed plates (5 × 104 cells/50 μl) and incubated with serial dilutions of MAbs (50 μl) at room temperature for 30 min. After centrifugation, the supernatants were removed, and aliquots of the diluted human serum were distributed (150 μl) as a source of the complement. After a 1-h incubation at 37°C, the plate was centrifuged, and the radioactivity in the supernatants was measured using a gamma counter. The percentage of specific cytolysis was calculated from the counts of samples according to the formula:

\[\%\ specific\ lysis\ {=}\ \frac{\mathit{E}\ {-}\ \mathit{S}}{\mathit{M}\ {-}\ \mathit{S}}\ {\times}\ 100\]

where E is the experimental release (cpm in the supernatant from cancer cells incubated with antibody and complement), S is the spontaneous release (cpm in the supernatant from cancer cells incubated with medium alone), and M is the maximum release (cpm released from cancer cells lysed with 1 n HCl). To evaluate CDC, concentrations of MAbs required for EC50 were calculated.

ADCC Assay.

An ADCC assay was performed by 4-h 51Cr-release assay as reported previously (43). Briefly, aliquots of the 51Cr-labeled cancer cells as described in the CDC assay were distributed into 96-well U-bottomed plates (1 × 104 cells/50 μl) and incubated with serial dilutions of MAbs (50 μl) in the presence of human effector cells (100 μl) at 37°C for 4 h. Human PBMCs, separated from a healthy donor’s peripheral blood using Polymorphprep (Nycomed Pharma AS, Oslo, Norway) according to the manufacturer’s instructions, provided the effector cells. After centrifugation, the radioactivity in the supernatants was measured using a gamma counter. The percentage of specific cytolysis was calculated in the same way as in the CDC assay. To evaluate ADCC, concentrations of MAbs required for EC50 were calculated. Moreover, to analyze the effector cell populations in the human PBMCs that were involved in the ADCC of the humanized KM8969 against various human lung cancer cell lines, highly purified lymphocytes (>99%) and monocytes (>99%) were separated by centrifugal elutriation from human PBMCs and were used as described previously (29).

Preparation of Heterospheroids.

Heterospheroids were prepared according to the previously reported method (37, 38). A thermo-responsive polymer, PNIPAAm, was used. PNIPAAm was insoluble in water over the lower critical solution temperature (∼30°C) and was reversibly soluble below the lower critical solution temperature. The substratum, with surface area of about 9.6 cm2, for cell culture and cell detachment was prepared by coating a hydrophilic culture dish with a uniform mixture of PNIPAAm and type I collagen and was called type I substratum. The formation and maintenance of heterospheroids were carried out on a hydrophilic culture dish coated with 1% agarose (type II substratum). Normal human dermal fibroblasts were seeded on the prewarmed (37°C) type I substratum at an initial cell density of 4.0 × 105/2 ml. After 3 days of culture at 37°C, the fibroblasts proliferated to a confluent state. Then aliquots of 1 ml of prewarmed (37°C) human cancer cells suspensions were seeded on the confluent fibroblasts monolayer at a cell density of 5.0 × 105/ml After 60 min of coculture at 37°C, it was confirmed, using a phase-contrast microscope, that more than 90% of the seeded cancer cells attached to the fibroblasts monolayer. Then the culture dishes were transferred to an ambient temperature (∼25°C) and allowed to stand for about 5 min. By this procedure, the cancer cells-attached fibroblasts monolayer was completely detached from the type I substratum as a self-supporting sheet. The detached cell sheet was gently transferred into a new dish containing chilled PBS using a wide pipette tip. This process was repeated and finally carried out using chilled culture medium instead of PBS. Then the rinsed cell sheet was transferred into prewarmed (37°C) culture medium on type II substratum by the same pipetting procedure (day 0). The heterospheroids formation and culture were carried out at 37°C on the type II substratum, and the culture medium was changed every other day from day 2.

Antibody Treatment of Heterospheroids.

The heterospheroids (n = 10) were incubated with the humanized KM8969 from day 2 to day 14. The diameter of the heterospheroids cultured for 2 days ranged from 700 to 850 μm. The freshly prepared medium containing 20 μg/ml antibody was used for each medium exchange at days 2, 4, 6, 8, 10, and 12. Two control groups received either polyclonal hIgG or medium alone.

Construction of Humanized MAb with Tag.

For detecting GM2 in the heterospheroids, we constructed a humanized anti-GM2 MAb with tag composed of FLAG peptide (DYKDDDDK) on the COOH-terminal of the heavy-chain C region. Briefly, the synthetic DNA-encoding FLAG peptide was fused in-frame to the 3′-end of the heavy-chain C region cDNA. Then the modified cDNA was subcloned into the expression vector of the humanized KM8969 to replace the native heavy-chain C region cDNA. Stable expression and purification of the tagged humanized KM8969 were performed as described previously (12). The purified tagged MAb was named KM8969FLAG.

Morphological and Immunocytochemical Analysis of Heterospheroids.

Paraffin-embedded sections and frozen sections of 7- and 14-day-cultured heterospheroids were prepared as follows. For paraffin sections, the heterospheroids were fixed in a 10% formalin neutral buffer solution for 60 min at 4°C. They were dehydrated and embedded in paraffin wax. Sections were obtained by cutting around their center at a thickness of 4 μm, dewaxed and stained with H&E by standard procedures. For immunoperoxidase staining, dewaxed sections were immersed in methanol containing 0.3% H2O2 for 30 min to remove endogenous peroxidase activity. For frozen sections, the heterospheroids were snap-frozen with OCT compound in liquid nitrogen. Sections were cut at a thickness of 5 μm, air-dried for 5 min, and fixed in a 10% formalin neutral buffer solution for 15 min. For immunoperoxidase staining, sections were washed with PBS and immersed in methanol containing 0.3% H2O2 for 30 min to remove endogenous peroxidase activity.

For detecting GM2 in the heterospheroids, frozen sections of 14-day-cultured heterospheroids were incubated with 10 μg/ml of the KM8969FLAG at 37°C for 1 h. Bound KM8969FLAG was detected using biotinylated mouse anti-FLAG M2 MAb (Eastman Kodak, New Haven, CT), followed by incubation with fluorescein avidin D (Vector Laboratories, Burlingame, CA).

For detecting the humanized KM8969 in the heterospheroids, frozen sections of 7- and 14-day-cultured heterospheroids were incubated with horseradish peroxidase-labeled goat anti-hIgG (H&L) (American Qualex, San Clemente, CA) at 37°C for 1 h. Bound peroxidase was detected after incubation with 0.5 mg/ml DAB containing 0.01% H2O2 for 2 min. Sections were counterstained in hematoxylin and dehydrated before mounting.

Apoptosis on paraffin sections of 14-day-cultured heterospheroids was detected using the In Situ Cell Death Detection kit (POD, Boehringer Mannheim) as described by the manufacturer. Sections were counterstained in methyl green and dehydrated before mounting.

All of the sections were observed with phase-contrast microscope or fluorescent microscope.

Humanized MAb and Characterization of Its Binding Activity.

The binding affinities of the purified chimeric KM966 and humanized KM8969 for GM2 were measured using GM2-binding ELISA (12). From the dose titration curves of ELISA, the concentrations of MAbs corresponding to the midpoint absorbance (EC50) were 0.045 ± 0.004 and 0.036 ± 0.003 μg/ml for chimeric KM966 and humanized KM8969, respectively (data not shown). In the case of the humanized KM8969, seven residues of FRs in the variable heavy region and nine residues of FRs in the variable light region as well as the residues of each CDR of the murine MAb were transferred to human FRs to attain high binding affinity. To confirm whether the binding characteristics of parental MAb were preserved in the humanized KM8969, we performed flow cytometric analysis using GM2-positive human SCLC SBC-3 cells. As shown in Fig. 1, the humanized KM8969 bound to SBC-3 cells at a slightly higher rate than did the chimeric KM966. We also examined antigen-binding specificity by ganglioside-binding ELISA (12). The humanized KM8969 reacted strongly with N-acetyl-GM2 and N-glycolyl-GM2 but weakly with GD2 of 11 common gangliosides (GM1, N-acetyl-GM2, N-glycolyl-GM2, N-acetyl-GM3, N-glycolyl-GM3, GD1a, GD1b, GD2, GD3, GQ1b, GT1b), which was the same reactive pattern as that of the chimeric KM966 (data not shown).

CDC of Humanized KM8969.

CDC of the purified humanized KM8969 against SBC-3 cells was evaluated in the presence of various concentrations of human serum as complement (5–15%; Fig. 2). Both humanized and chimeric MAb led to an enhancement of complement activation at a serum concentration of 5%, but the increase of serum concentrations had no effect on the enhancement of antibody-dependent specific cytolysis. At a serum concentration of 15%, the concentration of the humanized KM8969 required for half-maximal CDC (EC50) was 4.41 ± 0.15 versus 1.02 ± 0.07 μg/ml for the chimeric KM966. The slight loss of CDC despite the potent binding affinity to GM2 may reflect the slightly different binding characteristics of the humanized KM8969 compared with the chimeric KM966.

ADCC of Humanized KM8969.

The ADCC of the purified humanized KM8969 against SBC-3 cells was evaluated in the presence of human PBMCs as effector cells at various E:T ratios (5:1, 10:1, and 20:1; Fig. 3). Both humanized and chimeric MAbs exhibited an enhanced cancer-cell killing even at an E:T ratio of 5:1, but the increase of E:T ratios seemed to have no effect on the enhancement of antibody-dependent specific cytolysis because of the high cytolytic activity of effector cells without MAbs. At an E:T ratio of 20:1, the EC50 were 0.17 ± 0.02 and 0.07 ± 0.02 μg/ml for chimeric KM966 and humanized KM8969, respectively. The humanized KM8969 exhibited an ADCC that was slightly higher than the chimeric KM966, presumably also reflecting different binding characteristics between humanized KM8969 and chimeric KM966. From the dose titration curves of ADCC, the optimal dose of the humanized KM8969 was determined at 1 μg/ml. To further analyze the ADCC of the humanized KM8969, lymphocytes and monocytes isolated from a healthy donor’s PBMCs were each incubated with various human lung cancer cell lines in the presence of the optimal dose of MAb (1 μg/ml) and at an E:T ratio of 20:1. As shown in Table 1, the humanized KM8969 significantly induced ADCC mediated by both lymphocytes and monocytes against SBC-5 (SCLC), H69 (SCLC), PC-14 (lung adenocarcinoma), PC-13 (large cell lung cancer), and SBC-3 (SCLC) cells, but not RERF-LC-AI (lung squamous cell carcinoma) cells. The GM2 expression levels on the cancer cells surface were significantly correlated with the ADCC of the humanized KM8969 mediated by both lymphocytes and monocytes. The humanized KM8969 mediated a higher ADCC than the chimeric KM966 against GM2-positive human lung cancer cell lines, and the results supported the enhanced ADCC of the humanized KM8969 compared with the chimeric KM966, irrespective of the cell type. It was pointed out that the ADCC mediated by lymphocytes was higher than that mediated by monocytes.

We also evaluated ADCC against multidrug-resistant lung cancer cell lines and a highly metastatic lung cancer cell line (Table 2). The humanized KM8969 significantly induced ADCC mediated by lymphocytes and monocytes against two types of multidrug-resistant lung cancer cell lines, SBC-3/ADM and SBC-3/CDDP. Moreover, the humanized KM8969 exerted potent ADCC mediated by lymphocytes and monocytes against a highly metastatic cancer cell line, PC-14-PM4.

Morphology of Heterospheroids.

The H&E staining of sections from central regions of 14-day-cultured heterospheroids are shown in Fig. 4. Fibroblasts and lung cancer SBC-3 cells were strictly localized in a heterospheroid. In the control heterospheroid with hIgG treatment, it was observed by high magnification that the thick rim cell layers of strongly H&E-stained viable cells fully covered the cells that were aggregated into masses within the heterospheroid (Fig. 4, A and B). The same result was obtained in another control heterospheroid with medium alone (data not shown). On the other hand, the rim cell layers of the heterospheroid treated with the humanized KM8969 were scarce (Fig. 4, C and D). These observations suggested that the humanized KM8969 caused growth inhibition of the cells in the rim layers of the heterospheroid.

Detection of GM2 in Heterospheroids.

To detect GM2 in the heterospheroids, we constructed KM8969FLAG that had FLAG peptide on the COOH-terminus of the heavy chain. The KM8969FLAG retained binding affinity and specificity of the humanized KM8969 (data not shown). Immunofluorescence staining using the KM8969FLAG, biotinylated anti-FLAG M2 MAb, and fluorescein-avidin D enabled us to detect GM2 specifically with high sensitivity. As shown in Fig. 5,A, the specific signals of GM2 that indicate the existence of GM2-positive SBC-3 cells were observed only in the rim layers of 14-day-cultured heterospheroids of the control group treated with hIgG. The same result was also obtained in another control group treated with medium alone (data not shown). On the other hand, the heterospheroid treated with the humanized KM8969 exhibited no positive signals of GM2 (Fig. 5 B). The results showed that the humanized KM8969 caused the disappearance of GM2-positive SBC-3 cells and prohibited the outgrowth of the rim layers in the heterospheroids that were thought to be SBC-3 cells.

Localization of Humanized KM8969 in Heterospheroids.

Immunoperoxidase-stained sections of 7- and 14-day-cultured heterospheroids are shown in Fig. 6. After 7 days of the humanized KM8969 exposure, KM8969 penetrated into the heterospheroid and probed GM2-positive SBC-3 cells that were scattered within the heterospheroid (Fig. 6, A and B). The result revealed that the binding activity of the humanized KM8969 was preserved even in the heterospheroid. After 14 days, SBC-3 cells migrated from inner cell masses to the surface of the spheroid, and the humanized KM8969 was detected in the thin rim layers of the heterospheroid with heterogeneity (Fig. 6, C and D). The heterogenic staining pattern of the humanized KM8969 suggested that the existence of GM2-negative SBC-3 cells in the heterospheroid. The sections from two control groups showed no specific staining (data not shown).

Detection of Apoptosis in Heterospheroids.

To investigate the mechanism of the growth-inhibitory effect of the humanized KM8969 on the heterospheroid, we undertook experiments to detect apoptosis in the heterospheroid using a method based on in situ DNA strand-break labeling. The heterospheroid treated for 14 days with control hIgG exhibited strong positive signals only in the central area of severe necrotic cells caused by the limited supply of oxygen and nutrients (Fig. 7, A and B). In contrast, the heterospheroid treated for 14 days with the humanized KM8969 exhibited the positive signals of apoptosis distributed heterogeneically at the periphery of the heterospheroid (which resembled the distribution of the humanized KM8969) in addition to the positive signals in the central area (Fig. 7, C and D). The results suggested that apoptosis of SBC-3 cells induced by the humanized KM8969 contributed to the growth inhibition of the heterospheroid.

In control experiments, the humanized KM8969 affected neither the growth of the heterospheroids composed of normal human dermal fibroblasts and GM2-negative human stomach adenocarcinoma MKN-28 cells nor the growth of SBC-3 cells monolayer cultures (data not shown). These results showed that the growth-inhibitory effect of the humanized KM8969 was strictly dependent on GM2 expression and the multicelluar organization of the heterospheroid.

We have humanized a murine anti-GM2 MAb in an attempt to improve its potential clinical efficacy by reducing its immunogenicity and by changing the C region to support potent CDC and ADCC. The pharmacokinetic studies of humanized MAbs and murine MAbs in monkeys revealed that the humanization resulted in a prolonged serum half-life and in a substantial reduction in immunogenicity compared with the murine MAbs (44, 45). The antihumanized KM8969 response would be theoretically directed to a conformational epitope formed by the CDR loops, so that the reduction in immune response to the KM8969 in monkeys and humans would be expected in comparison with the chimeric KM966, which has murine V regions including CDRs. There has been no comparative pharmacokinetic studies between the humanized MAbs and their counterpart chimeric MAbs. The actual advantages of the humanized MAbs in pharmacokinetics vary for each MAb and need to be evaluated in clinical studies.

The humanized KM8969 showed a binding affinity and specificity to GM2 similar to the binding affinity and specificity of the chimeric KM966 but showed a 4-fold weaker CDC than that of the chimeric KM966 (Figs. 1 and 2). On the other hand, the humanized KM8969 was slightly more efficient in mediating ADCC than the chimeric KM966 was (Fig. 3). The different activities of CDC and ADCC of the humanized KM8969 may reflect the different binding characteristics between humanized KM8969 and chimeric KM966. In fact, as a result of the humanization of MAbs, different binding characteristics from their parental MAbs have also been reported (46, 47). For example, it is possible that the humanized KM8969 may have kinetic parameters that are different from those of the chimeric KM966. An alternate explanation is that the change of the V region conformation on humanization may affect the interaction between the Fc region and components of the complement or Fc receptor on the effector cell surface. To our knowledge, the humanized KM8969 represents the first humanized anti-GM2 IgG MAb with high binding affinity and specificity. Additional studies of X-ray crystallographic structural analysis and the determination of kinetic parameters of the humanized KM8969 are needed to define the detailed binding mechanisms of anti-GM2 MAbs. These studies are now under way. The results obtained from such studies should be very useful in the humanization of other MAbs.

The humanized KM8969 was able to induce ADCC mediated by both lymphocytes and monocytes against a variety of human lung cancer cell lines in direct proportion to GM2 expression levels on the cell surface (Table 1). Previous studies suggested that the types of effector cells that mediated ADCC varied depending on the nature of the recognized antigen and MAbs (48, 49, 50). For example, a chimeric MAb ch14.18 with specificity for GD2 was found to induce ADCC by granulocytes more efficiently than that by NK cells and to have no effect on monocytes (50). Although both lymphocytes and monocytes were able to be effector cells for the KM8969-dependent cytotoxicity, ADCC with lymphocytes seemed to be higher than ADCC with monocytes at an optimal concentration of the KM8969 (1 μg/ml) and an E:T ratio of 20:1. In addition, the humanized KM8969 also induced potent ADCC mediated by both lymphocytes and monocytes against multidrug-resistant lung cancer cells and highly metastatic lung cancer cells (Table 2). The possibility of the humanized KM8969 being potently useful for overcoming multidrug resistance in cancer cells is very attractive for combination studies with conventional cytotoxic drugs. In addition, potent ADCC against highly metastatic PC-14-PM4 cells of the humanized KM8969 suggested that KM8969 also had the antimetastatic effect in vivo similar to the chimeric KM966 (23) and could eradicate the multiple organ micrometastases of human cancers.

Gangliosides are ubiquitous components in the cell surface. Physiological functions of gangliosides have been investigated, and the involvement in the signal transduction of cell growth and the process of cell adhesion has been discussed in relation to oncogenesis and cancer metastasis (30, 31, 32). Recently, Iwabuchi et al.(35) reported that the GM3-enriched membrane subfraction, termed the glycosphingolipid signaling domain, comprised a structural and functional unit for the initiation of GM3-dependent cell adhesion coupled with signal transduction in mouse melanoma B16 cells. To investigate the role of GM2 in the growth of cancer cells in the form of three-dimensional organization, heterospheroids composed of normal human dermal fibroblasts and GM2-positive human lung cancer SBC-3 cells were cultured in the presence of the anti-GM2 humanized MAb, KM8969.

We found that the growth of the heterospheroids was evidently inhibited on exposure to the humanized KM8969. We noted that the humanized KM8969 induced apoptosis against GM2-positive SBC-3 cells in the heterospheroids, and this effect may play, at least in part, a role in growth inhibition (Fig. 7). The growth-inhibitory effects of MAbs were also reported by other investigators. An anti-p185HER2 MAb inhibited the growth of p185HER2-expressing cell lines through weak agonist effects on p185HER2(51). On the other hand, an anti-CD20 MAb showed remarkable growth inhibition against CD20-positive cell lines by apoptosis induction (52). Both MAbs were fully effective in simple monolayer cultures, whereas the humanized KM8969 had no inhibitory effect against monolayer cultures of GM2-positive cell line. The results indicated that the apoptosis of SBC-3 cells was induced by the antibody-capturing by GM2 on the cell surface and by subsequent events in the heterospheroid. Although the detail of the mechanisms of the growth-inhibitory effect of humanized KM8969 remain to be elucidated, the available evidence suggests that GM2 expression on the surface of cancer cells is strongly related to the formation and growth of cancer masses. Additional studies on the cytotoxic effect of the humanized KM8969 using this heterospheroids culture should provide new insights into more effective therapies for GM2-positive human cancers. A very potent cytotoxic agent specific for the growth inhibition of cancer cells could be attained by the cytotoxic functions of the humanized KM8969 because the growth-inhibitory mechanism could enhance the effectiveness of the antibody in vivo together with CDC and ADCC.

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.

            
2

Gangliosides have been designated according to the recommendations of the IUPAC-IUB Commission on Biochemical Nomenclature (53) and to the coding system of Svennerholm (1).

      
3

The abbreviations used are: MAb, monoclonal antibody; SCID, severe combined immunodeficient; NK, natural killer; CDR, complementarity-determining region; FR, framework region; C, constant (region); V, variable (region); ADCC, antibody-dependent cellular cytotoxicity; PNIPAAm; poly-N-isopropyl acrylamide; ADM; Adriamycin; CDDP, cisplatin; CDC, complement-dependent cytotoxicity; EC50, concentration for half-maximal cytolysis; PBMC; peripheral blood mononuclear cell; hIgG, human IgG; SCLC, small cell lung cancer.

      
4

K. Nakamura, Y. Tanaka, I. Fujino, K. Shitara, and N. Hanai. Construction and characterization of a humanized anti-ganglioside GM2 monoclonal antibody, submitted for publication.

Fig. 1.

Flow cytometric analysis of MAbs binding to SBC-3 cells. SBC-3 cells (1 × 106) were incubated with 50 μg/ml MAbs, and bound MAbs were detected with fluorescein-labeled protein A. A, humanized KM8969; B, chimeric KM966; C, control hIgG.

Fig. 1.

Flow cytometric analysis of MAbs binding to SBC-3 cells. SBC-3 cells (1 × 106) were incubated with 50 μg/ml MAbs, and bound MAbs were detected with fluorescein-labeled protein A. A, humanized KM8969; B, chimeric KM966; C, control hIgG.

Close modal
Fig. 2.

CDC of MAbs against human lung cancer SBC-3 cells. 51Cr-labeled SBC-3 cells were incubated at room temperature for 30 min with either chimeric KM966 (open columns) or humanized KM8969 (closed columns) at the concentrations indicated. After washing, diluted human serum with medium was added and incubated at 37°C for 1 h. Serum dilutions were 5% (A), 10% (B), or 15% (C). After centrifugation, the radioactivity in the supernatants was measured, and the percentages of cytolysis were plotted against the concentrations of MAbs. The lysis percentages observed with diluted serum alone (striped columns) were 6.7 ± 1.8, 16.8 ± 1.2, and 28.4 ± 2.6% for dilutions of 5, 10, and 15%, respectively.

Fig. 2.

CDC of MAbs against human lung cancer SBC-3 cells. 51Cr-labeled SBC-3 cells were incubated at room temperature for 30 min with either chimeric KM966 (open columns) or humanized KM8969 (closed columns) at the concentrations indicated. After washing, diluted human serum with medium was added and incubated at 37°C for 1 h. Serum dilutions were 5% (A), 10% (B), or 15% (C). After centrifugation, the radioactivity in the supernatants was measured, and the percentages of cytolysis were plotted against the concentrations of MAbs. The lysis percentages observed with diluted serum alone (striped columns) were 6.7 ± 1.8, 16.8 ± 1.2, and 28.4 ± 2.6% for dilutions of 5, 10, and 15%, respectively.

Close modal
Fig. 3.

ADCC of MAbs against human lung cancer SBC-3 cells. 51Cr-labeled SBC-3 cells were incubated at 37°C for 4 h with either chimeric KM966 (open columns) or humanized KM8969 (closed columns) at the concentrations indicated in the presence of effector cells at E:T ratios of 5:1 (A), 10:1 (B), or 20:1 (C). Effector cells were PBMCs obtained from a healthy donor. After centrifugation, the radioactivity in the supernatants was measured and the percentages of cytolysis were plotted against the concentrations of MAbs. The lysis percentages observed with effector cells alone (striped columns) were 9.8 ± 2.0, 31.3 ± 9.3, and 55.1 ± 7.4% for E:T ratios of 5:1, 10:1, and 20:1, respectively.

Fig. 3.

ADCC of MAbs against human lung cancer SBC-3 cells. 51Cr-labeled SBC-3 cells were incubated at 37°C for 4 h with either chimeric KM966 (open columns) or humanized KM8969 (closed columns) at the concentrations indicated in the presence of effector cells at E:T ratios of 5:1 (A), 10:1 (B), or 20:1 (C). Effector cells were PBMCs obtained from a healthy donor. After centrifugation, the radioactivity in the supernatants was measured and the percentages of cytolysis were plotted against the concentrations of MAbs. The lysis percentages observed with effector cells alone (striped columns) were 9.8 ± 2.0, 31.3 ± 9.3, and 55.1 ± 7.4% for E:T ratios of 5:1, 10:1, and 20:1, respectively.

Close modal
Fig. 4.

H&E-stained sections of 14-day-cultured heterospheroids composed of normal human dermal fibroblasts and human lung cancer SBC-3 cells. The heterospheroids were incubated with the antibody from day 2 to 14. The freshly prepared medium containing 20 μg/ml of the antibody was used for each medium exchange at day 2, 4, 6, 8, 10, and 12. The heterospheroids were treated with the control hIgG (A and B) and with the humanized KM8969 (C and D). B and D were high magnification of A and C, respectively. Bar: 100 μm in A and C; 20 μm in B and D.

Fig. 4.

H&E-stained sections of 14-day-cultured heterospheroids composed of normal human dermal fibroblasts and human lung cancer SBC-3 cells. The heterospheroids were incubated with the antibody from day 2 to 14. The freshly prepared medium containing 20 μg/ml of the antibody was used for each medium exchange at day 2, 4, 6, 8, 10, and 12. The heterospheroids were treated with the control hIgG (A and B) and with the humanized KM8969 (C and D). B and D were high magnification of A and C, respectively. Bar: 100 μm in A and C; 20 μm in B and D.

Close modal
Fig. 5.

Detection of GM2 in 14-day-cultured heterospheroids. Heterospheroids were treated with 20 μg/ml antibody on the same schedule as in Fig. 4. GM2 on the heterospheroids was detected using the tagged antibody, KM8969FLAG, followed by biotinylated anti-FLAG and fluorescein-avidin D detection systems. The heterospheroids were treated with the control hIgG (A) and with the humanized KM8969 (B). Bar, 100 μm.

Fig. 5.

Detection of GM2 in 14-day-cultured heterospheroids. Heterospheroids were treated with 20 μg/ml antibody on the same schedule as in Fig. 4. GM2 on the heterospheroids was detected using the tagged antibody, KM8969FLAG, followed by biotinylated anti-FLAG and fluorescein-avidin D detection systems. The heterospheroids were treated with the control hIgG (A) and with the humanized KM8969 (B). Bar, 100 μm.

Close modal
Fig. 6.

Detection of the humanized KM8969 in 7- and 14-day-cultured heterospheroids. Heterospheroids were treated with 20 μg/ml antibody on the same schedule as in Fig. 4. The bound humanized KM8969 was detected using horseradish peroxidase-labeled anti-hIgG antibody. A and B, 7-day-cultured spheroid. C and D, 14-day-cultured spheroid. B and D were high magnification of A and C, respectively. Bar: 100 μm in A and C; 20 μm in B and D.

Fig. 6.

Detection of the humanized KM8969 in 7- and 14-day-cultured heterospheroids. Heterospheroids were treated with 20 μg/ml antibody on the same schedule as in Fig. 4. The bound humanized KM8969 was detected using horseradish peroxidase-labeled anti-hIgG antibody. A and B, 7-day-cultured spheroid. C and D, 14-day-cultured spheroid. B and D were high magnification of A and C, respectively. Bar: 100 μm in A and C; 20 μm in B and D.

Close modal
Fig. 7.

Detection of apoptosis in 14-day-cultured heterospheroids. Heterospheroids were treated with 20 μg/ml antibody on the same schedule as in Fig. 4. Apoptosis on the heterospheroids was detected using the In Situ Cell Death Detection kit, POD. Sections were also counterstained in methyl green. The heterospheroids were treated with the control hIgG (A and B) and with the humanized KM8969 (C and D). B and D were high magnification of A and C, respectively. Bar: 100 μm in A and C; 20 μm in B and D.

Fig. 7.

Detection of apoptosis in 14-day-cultured heterospheroids. Heterospheroids were treated with 20 μg/ml antibody on the same schedule as in Fig. 4. Apoptosis on the heterospheroids was detected using the In Situ Cell Death Detection kit, POD. Sections were also counterstained in methyl green. The heterospheroids were treated with the control hIgG (A and B) and with the humanized KM8969 (C and D). B and D were high magnification of A and C, respectively. Bar: 100 μm in A and C; 20 μm in B and D.

Close modal
Table 1

Effect of humanized KM8969 on human lymphocyte- or monocyte-mediated cytotoxicity against various human lung cancer cell lines

Target cellsOriginGM2 expressionaMAb (1 μg/ml)% cytotoxicityb (mean ± SDc)
LymphocyteMonocyte
RERF-LC-AI Sq.d 2.2  1.7 ± 1.2 0.1 ± 0.7 
   KM966 3.1 ± 0.5 0.6 ± 1.0 
   KM8969 2.9 ± 0.4 0.8 ± 0.5 
SBC-5 SCLC 29.1  5.5 ± 0.9 0.1 ± 1.0 
   KM966 20.8 ± 5.6 17.6 ± 1.9 
   KM8969 24.3 ± 4.7 23.0 ± 3.4 
H69 SCLC 40.3  9.5 ± 6.3 4.9 ± 2.1 
   KM966 31.7 ± 3.4 26.0 ± 2.9 
   KM8969 35.6 ± 2.9 31.1 ± 3.0 
PC-14 Adeno. 45.6  15.0 ± 0.4 1.8 ± 1.1 
   KM966 46.8 ± 0.3 30.0 ± 0.8 
   KM8969 44.9 ± 1.2 32.5 ± 0.6 
PC-13 Large 212.7  11.9 ± 1.4 1.2 ± 0.6 
   KM966 57.7 ± 1.7 45.5 ± 1.1 
   KM8969 77.7 ± 6.3 47.3 ± 3.4 
SBC-3 SCLC 250.7  12.8 ± 2.5 2.2 ± 1.0 
   KM966 68.5 ± 4.0 40.9 ± 5.0 
   KM8969 78.7 ± 3.3 48.0 ± 2.0 
Target cellsOriginGM2 expressionaMAb (1 μg/ml)% cytotoxicityb (mean ± SDc)
LymphocyteMonocyte
RERF-LC-AI Sq.d 2.2  1.7 ± 1.2 0.1 ± 0.7 
   KM966 3.1 ± 0.5 0.6 ± 1.0 
   KM8969 2.9 ± 0.4 0.8 ± 0.5 
SBC-5 SCLC 29.1  5.5 ± 0.9 0.1 ± 1.0 
   KM966 20.8 ± 5.6 17.6 ± 1.9 
   KM8969 24.3 ± 4.7 23.0 ± 3.4 
H69 SCLC 40.3  9.5 ± 6.3 4.9 ± 2.1 
   KM966 31.7 ± 3.4 26.0 ± 2.9 
   KM8969 35.6 ± 2.9 31.1 ± 3.0 
PC-14 Adeno. 45.6  15.0 ± 0.4 1.8 ± 1.1 
   KM966 46.8 ± 0.3 30.0 ± 0.8 
   KM8969 44.9 ± 1.2 32.5 ± 0.6 
PC-13 Large 212.7  11.9 ± 1.4 1.2 ± 0.6 
   KM966 57.7 ± 1.7 45.5 ± 1.1 
   KM8969 77.7 ± 6.3 47.3 ± 3.4 
SBC-3 SCLC 250.7  12.8 ± 2.5 2.2 ± 1.0 
   KM966 68.5 ± 4.0 40.9 ± 5.0 
   KM8969 78.7 ± 3.3 48.0 ± 2.0 
a

Mean fluorescence intensity analysed by FACScan using chimeric KM966 (10 μg/ml; Ref. 29).

b

Lymphocytes or monocytes were incubated for 4 h in medium with or without MAb (1 μg/ml) with 51Cr-labeled cells at E:T ratio of 20:1.

c

Mean ± SD for triplicate cultures. Data are representative of four separate experiments.

d

Sq., lung squamous cell carcinoma; Adeno., lung adenocarcinoma; Large, large cell lung cancer.

Table 2

Effect of humanized KM8969 on human lymphocyte- or monocyte-mediated cytotoxicity against multidrug-resistant or highly metastatic human lung cancer cell lines

Target cellsCharacteristicMAb (1 μg/ml)% cytotoxicitya (mean ± SDb)
LymphocyteMonocyte
SBC-3/ADM ADM-resistant  11.1 ± 1.2 0.1 ± 0.5 
 P-gpc (+), MRP (+) KM966 50.7 ± 0.1 33.0 ± 2.0 
  KM8969 68.8 ± 2.0 27.9 ± 3.7 
SBC-3/CDDP CDDP-resistant  9.6 ± 3.0 1.3 ± 0.6 
 P-gp (−), MRP (+) KM966 71.8 ± 2.9 37.4 ± 1.5 
  KM8969 76.4 ± 5.7 39.3 ± 2.2 
PC-14-PM4 Highly metastatic to pleural cavity  7.8 ± 0.9 1.3 ± 0.5 
  KM966 39.1 ± 0.8 21.9 ± 0.8 
  KM8969 40.0 ± 1.5 25.1 ± 0.9 
Target cellsCharacteristicMAb (1 μg/ml)% cytotoxicitya (mean ± SDb)
LymphocyteMonocyte
SBC-3/ADM ADM-resistant  11.1 ± 1.2 0.1 ± 0.5 
 P-gpc (+), MRP (+) KM966 50.7 ± 0.1 33.0 ± 2.0 
  KM8969 68.8 ± 2.0 27.9 ± 3.7 
SBC-3/CDDP CDDP-resistant  9.6 ± 3.0 1.3 ± 0.6 
 P-gp (−), MRP (+) KM966 71.8 ± 2.9 37.4 ± 1.5 
  KM8969 76.4 ± 5.7 39.3 ± 2.2 
PC-14-PM4 Highly metastatic to pleural cavity  7.8 ± 0.9 1.3 ± 0.5 
  KM966 39.1 ± 0.8 21.9 ± 0.8 
  KM8969 40.0 ± 1.5 25.1 ± 0.9 
a

Lymphocytes or monocytes were incubated for 4 h in medium with or without MAb (1 μg/ml) with 51Cr-labeled cells at E:T ratio of 20:1.

b

Mean ± SD for triplicate cultures. Data are representative of four separate experiments.

c

P-gp, P-glycoprotein; MRP, multidrug-resistance-associated protein; +, positive; −, negative.

We thank Dr. K. Yano for protein sequencing, Dr. K. Uchida for protein purification, and Dr. A. Saito for oligonucleotide synthesis.

1
Svennerholm L. Chromatographic separation of human brain gangliosides.
J. Neurochem.
,
10
:
613
-623,  
1963
.
2
Hakomori S. Aberrant glycosylation in cancer cell membranes as focused on glycolipids: overview and perspectives.
Cancer Res.
,
45
:
2405
-2414,  
1985
.
3
Portoukalian J., Zwingelstein G., Dore J. Lipid composition of human malignant melanoma tumors at various levels of malignant growth.
Eur. J. Biochem.
,
94
:
19
-23,  
1979
.
4
Fuentes R., Allman R., Mason M. D. Ganglioside expression in lung cancer cell lines.
Lung Cancer
,
18
:
21
-33,  
1997
.
5
Zhang S., Cordon-Cardo C., Zhang H. S., Reuter V. E., Adluri S., Hamilton W. B., Lloyd K. O., Livingston P. O. Selection of tumor antigens as targets for immune attack using immunohistochemistry. I. Focus on gangliosides.
Int. J. Cancer
,
73
:
42
-49,  
1997
.
6
Hellström I., Brankovan V., Hellström K. E. Strong antitumor activities of IgG3 antibodies to a human melanoma-associated ganglioside.
Proc. Natl. Acad. Sci. USA
,
82
:
1499
-1502,  
1985
.
7
Houghton A. N., Mintzer D., Cordon-Cardo C., Welt S., Fliegel B., Vadhan S., Carswell E., Melamed M. R., Oettgen H. F., Old L. J. Mouse monoclonal IgG3 antibody detecting GD3 ganglioside: a Phase-I trial in patients with malignant melanoma.
Proc. Natl. Acad. Sci. USA
,
82
:
1242
-1246,  
1985
.
8
Cheung N-K. V., Lazarus H., Miraldi F. D., Abramowaky C. R., Kallic S., Saarinen U. M., Spitzer T., Strandjord S. E., Coccia P. F., Berger N. A. Ganglioside GD2 specific monoclonal antibody 3F8: a Phase-I study in patients with neuroblastoma and malignant melanoma.
J. Clin. Oncol.
,
5
:
1430
-1440,  
1987
.
9
Saleh M. N., Khazaeli M. B., Wheeler R. H., Dropcho E., Lui T., Urist M., Miller D. M., Lawson S., Dixon P., Russell C. H., LoBuglio A. F. Phase I trial of the murine monoclonal anti-GD2 antibody 14G2a in metastatic melanoma.
Cancer Res.
,
52
:
4342
-4347,  
1992
.
10
Irie R. F., Morton D. L. Regression of cutaneous metastatic melanoma by intralesional injection with human monoclonal antibody to ganglioside GD2.
Proc. Natl. Acad. Sci. USA
,
83
:
8694
-8698,  
1986
.
11
Schlom J., Eggensperger D., Colcher D., Molinolo A., Houchen D., Miller L. S., Hinkle G., Siler K. Therapeutic advantage of high-affinity anticarcinoma radioimmunoconjugates.
Cancer Res.
,
52
:
1067
-1072,  
1992
.
12
Nakamura K., Koike M., Shitara K., Kuwana Y., Kiuragi K., Igarashi S., Hasegawa M., Hanai N. Chimeric anti-ganglioside GM2 antibody with antitumor activity.
Cancer Res.
,
54
:
1511
-1516,  
1994
.
13
Dohi T., Nakamura K., Hanai N., Taomoto K., Oshima M. Reactivity of a mouse/human chimeric anti-GM2 antibody KM966 with brain tumors.
Anticancer Res.
,
14
:
2577
-2582,  
1994
.
14
Nishinaka Y., Ravindranath M. H., Irie R. F. Development of a human monoclonal antibody to ganglioside GM2 with potential for cancer treatment.
Cancer Res.
,
56
:
5666
-5671,  
1996
.
15
Natoli E. J., Jr., Livingston P. O., Pukel C. S., Lloyd K. O., Wiegandt H., Szalay J., Oettgen H. F., Old L. J. A murine monoclonal antibody detecting N-Acetyl- and N-glycolyl-GM2: characterization of cell surface reactivity.
Cancer Res.
,
46
:
4116
-4120,  
1986
.
16
Miyake M., Ito M., Hitomi S., Ikeda S., Taki T., Kurata M., Hino A., Miyake N., Kannagi R. Generation of two murine monoclonal antibodies that can discriminate N-glycolyl- and N-acetyl neuraminic acid residues of GM2 gangliosides.
Cancer Res.
,
48
:
6154
-6160,  
1988
.
17
Vrionis F. D., Wikstrand C. J., Fredman P., Mansson J-E., Svennerholm L., Bigner D. D. Five new epitope-defined monoclonal antibodies reactive with GM2 and human glioma and medulloblastoma cell lines.
Cancer Res.
,
49
:
6645
-6651,  
1989
.
18
Dohi T., Ohta S., Hanai N., Yamaguchi K., Oshima M. Sialylpentaosylceramide detected with anti-GM2 monoclonal antibody: structural characterization and complementary expression with GM2 in gastric cancer and normal gastric mucosa.
J. Biol. Chem.
,
265
:
7880
-7885,  
1990
.
19
Yamaguchi H., Furukawa K., Fortunato S. R., Livingston P. O., Lloyd K. O., Oettgen H. F., Old L. J. Human monoclonal antibody with dual GM2/GD2 specificity derived from an immunized melanoma patient.
Proc. Natl. Acad. Sci. USA
,
87
:
3333
-3337,  
1990
.
20
Irie R. F., Matsuki T., Morton D. L. Human monoclonal antibody to ganglioside GM2 for melanoma treatment.
Lancet
,
1
:
786
-787,  
1989
.
21
Helling F., Zhang S., Shang A., Adluri S., Calves M., Koganty R., Longenecker B. M., Yao T-J., Oettgen H. F., Livingston P. O. GM2-KLH conjugate vaccine: increased immunogenicity in melanoma patients after administration with immunological adjuvant QS-21.
Cancer Res.
,
55
:
2783
-2788,  
1995
.
22
Ragupathi G. Carbohydrate antigens as targets for active specific immunotherapy.
Cancer Immunol. Immunother.
,
43
:
152
-157,  
1996
.
23
Hanibuchi M., Yano S., Nishioka Y., Yanagawa H., Kawano T., Sone S. Therapeutic efficacy of mouse-human chimeric anti-ganglioside GM2 monoclonal antibody against multiple organ micrometastases of human lung cancer in NK cell-depleted SCID mice.
Int. J. Cancer
,
78
:
480
-485,  
1998
.
24
Brüggemann M., Winter G., Waldmann H., Neuberger M. S. The immunogenicity of chimeric antibodies.
J. Exp. Med.
,
170
:
2153
-2157,  
1989
.
25
Jones P. T., Dear P. H., Foote J., Neuberger M. S., Winter G. Replacing the complementarity-determining regions in a human antibody with those from a mouse.
Nature (Lond.)
,
321
:
522
-525,  
1986
.
26
Riechmann L., Clark M., Waldmann H., Winter G. Reshaping human antibodies for therapy.
Nature (Lond.)
,
332
:
323
-327,  
1988
.
27
Takamuku K., Akiyoshi T., Tsuji H. Antibody-dependent cell-mediated cytotoxicity using a murine monoclonal antibody against human colorectal cancer in cancer patients.
Cancer Immunol. Immunother.
,
25
:
137
-140,  
1987
.
28
Sung M. W., Yasumura S., Johnson J. T., Van Dongen G. A., Whiteside T. L. Natural killer (NK) cells as effectors of antibody-dependent cytotoxicity with chimeric antibodies reactive with human squamous-cell carcinomas of the head and neck.
Int. J. Cancer
,
61
:
864
-872,  
1995
.
29
Hanibuchi M., Yano S., Nishioka Y., Yanagawa H., Sone S. Anti-ganglioside GM2 monoclonal antibody-dependent killing of human lung cancer cells by lymphocytes and monocytes.
Jpn. J. Cancer Res.
,
87
:
497
-504,  
1996
.
30
Hakomori S. Possible functions of tumour-associated carbohydrate antigens.
Curr. Opin. Immunol.
,
3
:
646
-653,  
1991
.
31
Cheresh D. A., Pierschbacher M. D., Herzig M. A., Mujoo K. Disialoganglioside GD2 and GD3 are involved in the attachment of human melanoma and neuroblastoma cells to extracellular matrix proteins.
J. Cell Biol.
,
102
:
688
-696,  
1986
.
32
Kaur G., Viallet J., Laborda J., Blair O., Gazdar A. F., Minna J. D., Sausville E. A. Growth inhibition by cholera toxin of human lung carcinoma cell lines: correlation with GM1 ganglioside expression.
Cancer Res.
,
52
:
3340
-3346,  
1992
.
33
Ortaldo J. R., Mason A. T., Longo D. L., Beckwith M., Creekmore S. P., McVicar D. W. T cell activation via the disialoganglioside GD3: analysis of signal transduction.
J. Leukocyte Biol.
,
60
:
533
-539,  
1996
.
34
Okada H., Wu X., Okada N. Complement-mediated cytolysis and azidothymidine are synergistic in HIV-1 suppression.
Int. Immunol.
,
10
:
91
-95,  
1998
.
35
Iwabuchi K., Handa K., Hakomori S. Separation of “glycosphingolipid signaling domain” from caveolin-containing membrane fraction in mouse melanoma B16 cells and its role in cell adhesion coupled with signaling.
J. Biol. Chem.
,
273
:
33766
-33773,  
1998
.
36
Bjerkvig R., Engebraaten O., Laerum O. D., Fredman P., Svennerholm L., Vrionis F. D., Wikstrand C. J., Bigner D. D. Anti-GM2 monoclonal antibodies induce necrosis in GM2-rich cultures of a human glioma cell line.
Cancer Res.
,
51
:
4643
-4648,  
1991
.
37
Takezawa T., Yamazaki M., Mori Y., Yonaha T., Yoshizato K. Morphological and immuno-cytochemical characterization of a hetero-spheroid composed of fibroblasts and hepatocytes.
J. Cell Sci.
,
101
:
495
-501,  
1992
.
38
Takezawa T., Yoshizato K. Organoid formation with branched capillary-like networks Kitagawa Y. Matsuda T. Iijima S. eds. .
Animal Cell Technology: Basic & Applied Aspects
,
10
:
43
-47, Kluwer Academic Publishers Dordrecht, the Netherlands  
1999
.
39
Kunz-schughart L. A., Kreutz M., Knuechel R. Multicellular spheroids: a three-dimensional in vitro culture system to study tumour biology.
Int. J. Exp. Pathol.
,
79
:
1
-23,  
1998
.
40
Sutherland R. M. Cell and environment interactions in tumor microregions: the multicell spheroid model.
Science (Washington DC)
,
240
:
177
-184,  
1988
.
41
Shrivastava P., Hanibuchi M., Yano S., Parajuli P., Tsuruo T., Sone S. Circumvention of multidrug resistance by a quinoline derivative, MS-209, in multidrug-resistant human small-cell lung cancer cells and its synergistic interaction with cyclosporin A or verapamil.
Cancer Chemother. Pharmacol.
,
42
:
483
-490,  
1998
.
42
Yano S., Nokihara H., Hanibuchi M., Parajuli P., Shinohara T., Kawano T., Sone S. Model of malignant pleural effusion of human lung adenocarcinoma in SCID mice.
Oncol. Res.
,
9
:
573
-579,  
1997
.
43
Ohta S., Honda A., Tokutake Y., Hanai N. Anti-tumor effects of a novel monoclonal antibody with high binding affinity to ganglioside GD3.
Cancer Immunol. Immunother.
,
36
:
260
-266,  
1993
.
44
Stephens S., Emtage S., Vetterlein O., Chaplin L., Bebbington C., Nesbitt A., Sopwith M., Athwal D., Novak C., Bodmer M. Comprehensive pharmacokinetics of a humanized antibody and analysis of residual anti-idiotypic responses.
Immunology
,
85
:
668
-674,  
1995
.
45
Co M. S., Baker J., Bednarik K., Janzek E., Neruda W., Mayer P., Plot R., Stumper B., Vasquez M., Queen C., Loibner H. Humanized anti-Lewis Y antibodies: in vitro properties and pharmacokinetics in rhesus monkeys.
Cancer Res.
,
56
:
1118
-1125,  
1996
.
46
Carter P., Presta L., Gorman C. M., Ridgway J. B. B., Henner D., Wong W. L. T., Rowland A. M., Kotts C., Carver M. E., Shepard H. M. Humanization of an anti-p185HER2 antibody for human cancer therapy.
Proc. Natl. Acad. Sci. USA
,
89
:
4285
-4289,  
1990
.
47
Tempest P. R., Barbanti E., Bremner P., Carr F. J., Ghislieri M., Rifaldi B., Marcucci F. A humanized anti-tumor necrosis factor-monoclonal antibody that acts as a partial, competitive antagonist of the template antibody.
Hybridoma
,
13
:
183
-190,  
1994
.
48
Nishioka Y., Sone S., Heike Y., Hamada H., Ariyoshi K., Tsuruo T., Ogura T. Effector cell analysis of human multidrug-resistant cell killing by mouse-human chimeric antibody against P-glycoprotein.
Jpn. J. Cancer Res.
,
83
:
644
-649,  
1992
.
49
Masucci G., Wersall P., Ragnhammer P., Mellstedt H. Granulocyte-monocyte-colony-stimulating factor augments the cytotoxic capacity of lymphocytes and monocytes in antibody-dependent cellular cytotoxicity.
Cancer Immunol. Immunother.
,
29
:
288
-292,  
1989
.
50
Barker E., Mueller B. M., Handgretinger R., Herter M., Yu A. L., Reisfeld R. A. Effect of a chimeric anti-ganglioside GD2 antibody on cell-mediated lysis of human neuroblastoma cells.
Cancer Res.
,
41
:
144
-149,  
1991
.
51
Kern J. A., Torney L., Weiner D., Gazdar A., Shepard H. M., Fendly B. Inhibition of human lung cancer cell line growth by an anti-p185HER2 antibody.
Am. J. Respir. Cell Mol. Biol.
,
9
:
448
-454,  
1993
.
52
Taji H., Kagami Y., Okada Y., Andou M., Nishi Y., Saito H., Seto M., Morishima Y. Growth inhibition of CD20-positive B lymphoma cell lines by IDEC-C2B8 anti-CD20 monoclonal antibody.
Jpn. J. Cancer Res.
,
89
:
748
-756,  
1998
.
53
IUPAC-IUB Commission on Biochemical Nomenclature (CBN). The nomenclature of lipids.
Eur. J. Biochem.
,
79
:
11
-21,  
1977
.