Internalizing Antibodies Are Necessary for Improved Therapeutic Efficacy of Antibody-targeted Liposomal Drugs

Introduction

Antibody-mediated targeting of liposomal anticancer drugs to epitopes expressed at the surface of cancer cells is being investigated as a means of increasing the site-specific delivery of drug to cancer cells (1, 2, 3, 4, 5) . Either internalizing or noninternalizing epitopes are possible targets for liposomal anticancer drugs conjugated to mAbs3 (immunoliposomes), but the mechanism of delivery of the drug into the cell is different in each case. When targeted liposomal drugs bind to noninternalizing epitopes, liposome contents are released over time at or near the cell surface, and the released drug will enter the cell by passive diffusion or normal transport mechanisms. Although increased concentrations of drug may be achieved at the cell surface by this mechanism, it can be argued that, in the dynamic in vivo environment, the rate of diffusion and redistribution of the released drug away from the cell will exceed the rate at which the drug enters the cell, particularly for drugs such as DXR, which have a large volume of distribution. When targeted liposomal drugs bind to internalizing epitopes, it triggers receptor-mediated uptake of the immunoliposomal drug package into the cell interior, where the drug contents are released subsequent to liposomal degradation by lysosmal and endosomal enzymes. Hence, one can hypothesize that targeting to internalizing epitopes should result in delivery of higher concentrations of drug to the cell interior than targeting to noninternalizing epitopes, resulting in improved therapeutic outcomes for liposomal drugs such as DXR that are resistant to degradation by the enzyme-rich, low pH environment of endosomes and lysosomes. Some indirect experimental evidence supports this hypothesis. Liposomes targeted to internalizing receptors have demonstrated increased therapeutic activity in some tumor models (2, 3, 4) . In other tumor models, targeted liposomes did not improve therapeutic outcome over nontargeted liposomes, which was hypothesized to be due to the lack of internalization of the drug-liposome package into the cells (6 , 7) . This study aims, in a B lymphoma model system, to directly verify the hypothesis that internalizing epitopes make better targets than noninternalizing epitopes for liposomal anticancer drugs. The binding, internalization, cytotoxicity, and therapeutic outcome of immunoliposomes targeted against CD19 (internalizing epitope) were compared with those targeted against CD20 (noninternalizing epitope). Significant improvements in therapeutic outcome were associated with the liposomes targeted against the internalizing epitope.

Materials and Methods

Materials.

HSPC and mPEG₂₀₀₀-DSPE (8) , were generous gifts from ALZA Pharmaceuticals, Inc. (Mountain View, CA). Chol was purchased from Avanti Polar Lipids (Alabaster, AL). Mal-PEG was custom synthesized by Shearwater Polymers, Inc. (Huntsville, AL). Nuclepore polycarbonate membranes (pore sizes, 0.2, 0.1, and 0.08 μm) were purchased from Northern Lipids (Vancouver, British Columbia, Canada). 2-Iminothiolane (Traut’s reagent) and MTT were purchased from Sigma Chemical Co. (St. Louis, MO). Rh-PE was obtained from Molecular Probes (Eugene, OR). RPMI 1640 (without phenol red), penicillin-streptomycin, and fetal bovine serum were obtained from Life Technologies, Inc. (Burlington, Ontario, Canada). Iodobeads were purchased from Pierce (Rockford, IL), and Bio-Rad Protein Assay Reagent was purchased from Bio-Rad Laboratories (Mississauga, Ontario, Canada). Sephadex G-50, Sepharose CL-4B, aqueous counting scintillant were purchased from Amersham Pharmacia Biotech (ASC; Baie d’Urfe, Quebec, Canada). [³H]CHE (1.48–2.22 TBq/mmol) and ¹²⁵I-NaI (185 MBq) were purchased from Mandel Scientific (Mississauga, Ontario, Canada). Goat antimouse-FITC IgG and goat antihuman-FITC IgG were purchased from Sigma Chemical Co. All other chemicals were of analytical grade purity.

Animals, Antibodies, and Cell Lines.

Female 6–8-week-old CB17/ICR SCID mice were purchased from Taconic Farms (Germantown. NY) and housed in the virus antigen-free unit of the Health Sciences Laboratory Animal Services, University of Alberta. All experiments were approved by the Health Sciences Animal Policy and Welfare Committee of the University of Alberta.

The murine mAb αCD19 was produced from the FMC63 murine hybridoma (9) and purified as described previously (10) . Rituximab, a chimeric IgG1 mAb, was used as a source of αCD20. Iodinated antibodies were used to measure coupling efficiencies and determine the amount of mAb attached to the liposomes (2) . The human Burkitt’s lymphoma cell line Namalwa (ATCC CRL 1432) was purchased from American Type Culture Collection (Manassas, VA) and cultured in suspension in a humidified 37°C incubator with a 5% CO₂ atmosphere in RPMI 1640 supplemented with 10% (v/v) fetal bovine serum, penicillin G (50 units/ml), and streptomycin sulfate (50 μg/ml). For experiments, only cells in the exponential phase of cell growth were used.

Immunophenotyping of Namalwa cells using single-color flow cytometry was performed to examine the cell surface expression of CD19 and CD20 epitopes. Namalwa cells (1 × 10⁶) were first stained with 10 μg of primary mAb followed by 20 μl of a 1:32 dilution of goat antimouse-FITC IgG for αCD19 or goat antihuman-FITC IgG for αCD20. Cell-associated fluorescence was analyzed on a Becton Dickinson FACScan using Lysis II software (Becton Dickinson, San Jose, CA). FITC fluorescence markers were excited with an argon laser (488 nm), and emitted fluorescence was detected using a 530 nm band pass filter.

Preparation of Liposomes.

Nontargeted liposomes, to be loaded with DXR for cytotoxicity and therapeutic studies or radiolabeled with [³H]CHE for binding studies, were composed of HSPC:Chol:mPEG₂₀₀₀-DSPE at a 2:1:0.1 molar ratio (SL), and targeted liposomes were composed of HSPC:Chol:mPEG₂₀₀₀-DSPE:Mal-PEG at a 2:1:0.08:0.02 molar ratio (SIL). For confocal microscopy studies, 0.1 mol% of Rh-PE was incorporated into the lipid mixture. Liposomes were prepared by hydration of thin films as described previously and extruded to mean diameter in the range of 100 ± 10 nm (11) . DXR was loaded into liposomes using the ammonium sulfate loading method (12) .

αCD19 mAb or αCD20 mAb was coupled to the terminus of the Mal-PEG at 2000:1 (lipid:protein) molar ratios, using the coupling procedure described previously (13) . Briefly, mAb (10 mg/ml) was incubated with 2-iminothiolane in O₂-free HEPES-buffered saline (pH 8.0) at a ratio of 20:1 mol/mol for 1 h at room temperature to thiolate the amino groups. At the end of the incubation, the sample was chromatographed on a Sephadex G-50 column, equilibrated with O₂-free HEPES-buffered saline (pH 7.4), and immediately incubated with liposomes in an O₂-free environment overnight with continuous stirring. To assess coupling efficiency of the antibodies, a trace amount of ¹²⁵I-labeled αCD19 or αCD20 was added to the unlabeled antibody before thiolation. A coupling efficiency of 80–90% for either antibody could routinely be achieved by this procedure, and particular attention was taken to ensure that similar antibody densities (within ± 10%) occurred at the surface of either type of immunoliposome.

Binding and Cytotoxicity of Immunoliposomes.

In vitro cell association of immunoliposomes labeled with [³H]CHE was determined, as described previously, at both 37°C and 4°C, i.e., permissive and nonpermissive temperatures for endocytosis, respectively (2) . Cell association (pmol PL/1 × 10⁶ cells) was calculated from the specific activity of the liposomes. Specific binding was determined by subtracting binding due to nontargeted liposomes from the total binding.

The in vitro cytotoxicity of free DXR, free antibodies, and various liposomal formulations of DXR was determined using the MTT tetrazolium dye reduction assay as described previously (2) . Results are expressed as IC₅₀, which was obtained graphically using SlideWrite software (Advanced Graphics Software, Encinitas, CA).

For confocal studies, Namalwa cells (1 × 10⁶) were incubated with Rh-PE-labeled liposomes either nontargeted or targeted via αCD19 or αCD20 mAbs for 1 h at 37°C or 4°C. Cells were then washed twice with cold PBS to remove unbound liposomes and resuspended in ∼0.1 ml of PBS. Cells were allowed to adhere onto poly-l-lysine-coated slides before mounting with Permaflor (Lipshaw Immunon, Pittsburgh, PA). Cells were then visualized on a ZEISS LSM 510 confocal microscope consisting of a 100W HBO mercury burner (for direct observation) and a He Ne laser with excitation at 543 nm. Emission was collected with LP560. The pinhole was adjusted to obtain 1.0 μm optical sections and images (512 × 512 pixels) were collected.

In Vivo Survival Experiments.

Namalwa cells (5 × 10⁶) in 0.2 ml of PBS were injected i.v. in the tail vein of SCID mice (5–7 mice/group). Treatments were given as a single bolus i.v. dose of 3 mg DXR/kg as free DXR, DXR-SL, DXR-SIL[αCD19], or DXR-SIL[αCD20]. The density of αCD19 or αCD20 on the liposomes was 76 μg/μmol PL (40 mAb/liposome equaling 80 antigen-binding sites) or 70 μg/μmol PL (37 mAb/liposome), respectively; i.e., each mouse received 15 μg of αCD19 or 13 μg of αCD20 conjugated to the DXR-containing immunoliposomes. As controls, the same amounts of free mAbs were administered, and empty immunoliposomes also contained comparable doses of antibody and PL. Mice were monitored daily and euthanized when they developed hind leg paralysis.

Statistical Analysis.

Comparisons of cellular binding and uptake, cytotoxicities, and therapeutic efficacies were done using one-way ANOVA with InStat software (GraphPad Software Version 3.0; GraphPad Software, San Diego, CA). The Tukey post-test was used to compare means. Data were considered significant at P < 0.05.

Results and Discussion

Immunophenotyping (data not shown) of Namalwa cells demonstrated that this cell line had a high expression of both of the B-cell differentiation antigens CD19 and CD20; CD20 had a MFI of 259 versus a MFI of 175 for CD19 against a background MFI of 15–18, indicating that CD20 had a slightly higher expression. The percentage population of gated cells that expressed these epitopes was 99.9% and 99.1% for CD19 and CD20, respectively. CD20 seemed to have a more heterogeneous distribution than CD19 with a coefficient of variation and MFI for CD20 of 78 and 259, respectively, and for CD19 of 45 and 163, respectively. The coefficient of variation of unstained cells was 44.

Antibody-mediated specific targeting effect of both SIL[αCD19] and SIL[αCD20] could be demonstrated in vitro (Fig 1A)⇓ . Binding of SIL[αCD19] to Namalwa cells reached saturation at a PL concentration of approximately 0.4 mm (Fig. 1B)⇓ . Binding of SIL[αCD20] had not reached saturation by a dose of 1.6 mm PL, which could be due to the higher expression of the CD20 epitope on the Namalwa cells and/or higher avidity of SIL[αCD20] for Namalwa cells than SIL[αCD19] (Fig. 1B)⇓ . Cellular association of SIL[αCD19] with cells at 37°C was higher than that at 4°C (Fig. 1A)⇓ . This was probably due to binding of the SIL[αCD19] to the cells via the pan-B-cell differentiation antigen, CD19, followed by receptor-mediated endocytosis and recycling of the epitope back to the cell surface, where it was available to partake in further binding and internalization events (2 , 14, 15, 16) . No significant difference in cellular association of SIL[αCD20] to cells at 37°C versus 4°C was observed.

Confocal fluorescence microscopy studies using Rh-PE-labeled liposomes showed that after a 1-h incubation at 4°C, both SIL[αCD19] and SIL[αCD20] were largely found on the cell surface, suggesting that both types of immunoliposomes could efficiently bind to the Namalwa cells (data not shown). After a 1-h incubation at 37°C, SIL[αCD20] remained largely on the cell surface, which is consistent with its poor ability to internalize (17) . SIL[αCD19], on the other hand, showed evidence of internalization, with aggregates of red fluorescence distributed throughout the cytoplasm (Fig. 2)⇓ . Little fluorescence was observed in the cells incubated with nontargeted liposomes, consistent with their low levels of nonselective binding to the cells.

Results from in vitro cytotoxicity assays demonstrated that liposomes conjugated to either αCD19 or αCD20 had similar cytotoxicities against Namalwa cells, and both had significantly higher cytotoxicities than nontargeted liposomes at an incubation time of 1 h (Table 1)⇓ . Neither drug-free immunoliposomes nor free antibodies displayed any cytotoxicity against Namalwa cells at the concentrations tested, suggesting that these antibodies were unable to signal cell growth arrest or death via cell-signaling mechanisms at the concentrations present on the immunoliposomes. The cytotoxicity of DXR-SIL[αCD19] was most likely due to the receptor-mediated endocytosis of the drug-loaded liposomes into the cells and release of the drug in the cell interior, as reported previously (11 , 18) . DXR-SIL[αCD20], on the other hand, probably produced its cytotoxicity by release of DXR from the bound liposomes at the cell surface and uptake of the released drug into the cells. In cell culture dishes, there is no opportunity for released drug to redistribute away from the cells, unlike the in vivo situation.

Table 2⇓ gives the survival times for tumor-bearing mice inoculated with Namalwa cells and treated with immunoliposomal formulations of DXR and a variety of control treatments. The most interesting observation is that treatment of mice with DXR-SIL[αCD19] resulted in significantly increased life spans relative to treatment with DXR-SIL[αCD20] (P < 0.001). This observation directly supports the hypothesis that internalizing epitopes make better targets than noninternalizing epitopes for immunoliposomal drugs. We have published evidence that, after internalization of the liposomal drug packages, the breakdown of the drug-liposome package by lysosomal and endosomal enzymes and release of drug into the cell interior are responsible for the cytotoxic effect produced by targeted liposomal DXR (11 , 18 , 19) . The higher concentrations of drugs delivered into the cell interior by this mechanism are the most probable reason for the increased life spans observed for immunoliposomes directed against internalizing versus non-internalizing epitopes. The drug released at the cell surface from DXR-SIL[αCD20], on the other hand, will be rapidly redistributed away from the target cells in vivo, and we hypothesize that the lower drug concentrations delivered to the target cells are the reason for the lesser therapeutic effect. DXR-SIL[αCD19] also increased the survival of mice to a significantly greater extent compared with DXR-SL (P < 0.001) or free DXR (P < 0.001). No significant difference was observed between mice treated with DXR-SIL[αCD20] and free DXR (P > 0.05). Mice treated with DXR-SIL[αCD20] had survival times that were marginally different from those of mice treated with DXR-SL (P < 0.05).

Injection of mice with drug-free liposomes conjugated to either αCD19 or αCD20 did not improve the survival times of mice compared with untreated controls, unlike comparable amounts of free αCD19 or αCD20 (P < 0.01). The modest cytotoxic effects of free αCD19 or αCD20 may be mediated through Fc-mediated complement-dependent cytotoxicity and/or antibody-dependent cell-mediated cytotoxicity. Drug-free immunoliposomes, despite the multivalent display of mAbs at the liposomes surface, may be less effective than a similar concentration of free antibodies because the orientation of the bound antibodies with respect to the liposome surface might shield the Fc segment and hinder its activity. Alternatively, different cellular processing pathways for the free mAbs and the immunoliposomes may account for the different effects of each.

We conclude that internalization of liposome-drug packages into the cell interior is an important factor in determining the therapeutic efficacy of immunoliposomal drugs. Internalization of antibodies or other ligands into the target cell is also required for other targeted therapeutics, such as immunotoxins and antibody-drug conjugates, and for targeted delivery of genes or viral DNA into cells (5) . Direct selection for antibodies that have efficient internalization is now possible by panning on target cells using antibody phage display libraries (20) .