HER-2/neu (p185HER2) oncogene represents an attractive target for antibody-mediated immunotherapy. The major problem of combining chemotherapy and immunotherapy is the severe side effects that limit the use of doxorubicin (Doxo) as a cytotoxic drug. We have used virosomes (Vir; reconstituted fusion-active viral envelopes) as a new drug delivery system and have shown that Vir are capable of binding and penetrating into tumor cells, delivering cytotoxic drugs. We have additionally demonstrated that conjugating Fab′ fragments of an antirat Neu (anti-rNeu) monoclonal antibody to Vir selectively and efficiently inhibits tumor progression of established rNeu-overexpressing breast tumors. Fab′-Doxo-Vir combine the antiproliferative properties of the monoclonal antibody and the cytotoxic effect of Doxo in vivo. Furthermore, Fab′-Doxo-Vir significantly inhibit tumor formation at a tumor load representing metastatic spread. These results indicate that Vir conjugated with an antibody against a tumor antigen are a promising new selective drug delivery system for the treatment of tumors expressing a specific tumor antigen.

Among various antigens specific for tumor cells, HER-2/neu, a member of the epidermal growth factor receptor family encoding a transmembrane growth factor receptor (1), is an attractive target antigen. The overexpression of HER-2/neu protein has been observed in various tumors, especially in 20–30% of breast cancers (2, 3, 4). Moreover, HER-2/neu overexpression occurs in the primary tumor as well as in metastatic sites (5). 4D5, a murine mAb4 directed against the extracellular domain of human HER-2/neu protein, has been shown to elicit receptor internalization and ultimately to inhibit proliferation of HER-2/neu-overexpressing breast cancer cells in vitro and in breast cancer xenografts (6, 7, 8, 9). Another anti-HER-2/neu mAb (clone 7.16.4), initially raised against the ectodomain of rNeu, was shown to share an epitope with 4D5 and to inhibit tumor formation of HER-2/neu-overexpressing tumor cells (10, 11, 12, 13, 14).

On the basis of observations of a synergistic effect between anti-HER-2/neu mAb and chemotherapeutic drugs (8), liposomes have been used as a drug carrier system to achieve increased therapeutic effectiveness while at the same time reducing the systemic side effects of cytotoxic drugs. Advances in the development of liposomes have shown that PEG conjugated to the liposomal surface achieved a steric stabilization that avoided clearance by the reticuloendothelial system, prolonged the circulation time, and enhanced tumor extravasation of drugs (15, 16, 17). By the conjugation of tumor-specific antibodies to the surface of liposomes (immunoliposomes), tumor cells were targeted specifically, and the therapeutic effect of encapsulated anticancer drugs was increased (18). Whether immunoliposomes targeted at HER-2/neu would enter tumor cells is dependent upon the surface density of HER-2/neu as well as the receptor internalization (19, 20, 21).

In the present study, we have investigated whether Vir could be used as a drug carrier system into cells. The advantage of using Vir is the intrinsic capability of influenza virus to enter any mammalian cells triggered by its HA. This effect is conserved if the native HA is inserted into the virosomal lipid bilayer (22, 23). Moreover, anti-rNeu-Fab′ fragments were conjugated to the virosomal surface by a linker with a long PEG spacer arm, thus enabling Vir to target rNeu-expressing cancer cells. We analyzed the binding and internalization of Fab′-Vir to rNeu+ tumor cells and compared it with unconjugated Vir. We demonstrated the cytotoxic effect of Fab′-Vir in vitro. Furthermore, we have shown a significant inhibition of tumor progression in mice with established tumors treated with Doxo encapsulated in Fab′-Vir (Fab′-Doxo-Vir). The same virosomal construct was capable of successfully inhibiting tumor formation in mice through early treatment after tumor inoculation. Our results demonstrate that Vir conjugated with a tumor-specific antibody are a new and efficient drug delivery system with tumor-specific targeting.

Chemicals.

C12E8, PC solution, 5(6)-carboxyfluorescein N-succinimidyl ester, and Doxo·HCl were purchased from Fluka (Buchs, Switzerland). PE was purchased from Avanti Polar-Lipids (Alabaster, AL). Bio-Beads SM-2 were purchased from Bio-Rad (Richmond, CA). NHS-PEG-MAL (Mr 2000) was purchased from Shearwater Polymers (Huntsville, AL).

Animals and Cell Lines.

Mouse mammary tumor virus (MMTV)/r-Neu:FVB mice transgenic for the rNeu protein (rNeu-TG) were purchased from Charles River (Sulzfeld, Germany). Animal care was in accordance with institutional guidelines. The rNeu+ breast cancer cell line NF 9006, derived from a rNeu-TG mouse, and the rNeu breast cancer cell line M/BB 659, derived from a c-myc-TG mouse, have been described previously (24).

Production of Anti-rNeu mAb and Fab′ Fragments.

Hybridoma cells producing the mAb 7.16.4 were kindly provided by M. Greene (University of Pennsylvania, Philadelphia, PA; Ref. 25). The mAb 7.16.4 was purified through a protein G column. F(ab′)2 was produced using partial digestion with pepsin and further reduced to Fab′ fragments in 30 mm cysteine, 100 mm Tris (pH 7.6) for 20 min at 37°C as described previously (26).

Synthesis of PE-PEG-Fab′.

To perform a site-directed conjugation of Fab′ fragments to keep the antigen binding site available, the Fab′ fragments were conjugated to the linker NHS-PEG-MAL containing a long PEG spacer arm. Thus, 100 mg of NHS-PEG-MAL were dissolved in 3 ml of anhydrous methanol containing 10 μl of triethylamine. Then, 45 mg of dioleoyl phosphatidylethanolamine dissolved in 4 ml of chloroform and methanol (1:3, v/v) were added to this solution. The mixture was triggered under nitrogen for 3 h at room temperature. Methanol:chloroform was removed under decreasing pressure, and the products were redissolved in chloroform. The solution was extracted with 1% NaCl to remove unreacted material and water-soluble byproducts. The PE-PEG-MAL was further purified by silic acid chromatography (27) and dissolved in Tris-HCl buffer (100 mm, pH 7.6) containing 10 mg/150 μl of C12E8. To this solution, the Fab′ fragments were added at a Fab′:PE-PEG-MAL ratio of 1:10. The solution was stirred at room temperature, and additional C12E8 was added to obtain a 1% C12E8 solution. Unreacted PE-PEG-MAL was removed by the addition of 400 μl of washed, moist Thiopropyl-Sepharose 6B. After a 3-h incubation, the gel was removed by centrifugation. PE-PEG-Fab′ solution was filtered through a 0.2 μm filter and stored as a 0.01 m C12E8 detergent solution.

Preparation of Vir and Fab′ Vir.

HA from the A/Singapore/6/86 strain of influenza virus was isolated as described previously (23, 28). Supernatant containing solubilized HA trimer (2.5 mg/ml) in 0.01 m C12E8 was used for the production of Vir. PC (38 mg) in chloroform was added to a round-bottomed flask, and the chloroform was evaporated by a rotary evaporator. The supernatant (4 ml containing 10 mg of HA) and 3.6 ml of PE-PEG-Fab′ (containing 4 mg of Fab′ fragments) were added to this flask. Under gentle shaking, the PC film covering the glass wall of the flask was solubilized by the C12E8-containing mixture. The detergent of the resulting solution was removed by extraction with sterile Biobeads SM-2. After this procedure, a slightly transparent solution of Fab′-Vir was obtained (1 ml of Fab′-Vir contained 1.3 mg of HA, 5 mg of PC, and 0.18 mg of Fab′ fragments). The FITC-Vir and Fab′-FITC-Vir were prepared by the same method, with the exception that carboxyfluorescein N-succinimidyl ester coupled to PE was built into the lipid bilayer. The procedure for the production of Vir was the same without PE-PEG-Fab′.

Encapsulation of Doxo into Vir and Fab′-Vir.

Doxo was loaded into Vir through a proton gradient generated by Vir-entrapped ammonium sulfate as described (29). In brief, to load Vir with ammonium sulfate, an ammonium sulfate solution (4.17 g/ml) was added to the Vir solution, sonicated for 1 min, and dialyzed (Spectra/Por 2.1; Biotech DispoDialyzers, molecular weight cutoff: 15,000; Spectrum Medical Industries, Houston, TX) against 1 liter of PBS containing 5% glucose for 24 h at 4°C. To prepare the Doxo loading solution, 10 mg of Doxo were dissolved in 3 ml of water and sterilized (0.2 μm filter); then 750 μl of sterile 5× concentrated PBS and 5% glucose were added. Afterward, the Viro solution and Doxo loading solution (2:1, v/v) were incubated for 10 h at 33°C and further incubated overnight at 28°C. Nonencapsulated Doxo was separated from the Viro by gel filtration on a High Load Superdex 200 column (Pharmacia, Uppsala, Sweden). The void volume fractions containing Fab′-Vir with encapsulated Doxo were eluted with 5% glucose in PBS and collected. The Doxo concentration was determined by absorbance at 480 nm. Vir preparations contained 150 μg/ml Doxo.

The mean diameter of the Vir was determined by photon-correlation spectroscopy with a Coulter N4Plus Sub-Micron-Particle Size Analyzer (Miami, FL). The proper expression of viral fusogenic activity of the Vir was measured as described previously (30).

Immunofluorescent Labeling.

For binding, rNeu+/rNeu breast cancer cell lines (NF 9006 and M/BB 659) were incubated at 4°C for 30 min with FITC-conjugated anti-rNeu-Vir (Fab′-Vir) and FITC-Vir. The fluorescence was analyzed by flow cytometry on a FACScan (Becton Dickinson, Heidelberg, Germany).

For internalization, cells in suspension were incubated with 1 μg/ml FITC-conjugated Fab′-Vir or Vir for 30 min at 4°C and then washed in PBS. Internalization was performed in 1 ml of complete medium for 1 h at 37°C. Control samples were incubated at 4°C for 30 min with anti-Neu mAb Ab-4 (Oncogene Science, Tarzana, CA) and then washed with PBS, followed by incubation with FITC-conjugated goat antimouse IgG (Southern Biotechnology, Birmingham, AL). Internalization was performed for 1 h at 37°C in medium. Samples were washed and fixed in 3% paraformaldehyde. Fixed cells were permeabilized in 0.2% Triton X-100 for 15 min and stained for F-actin with rhodamine:phalloidin (1:100; Molecular Probes, Leiden, the Netherlands). After staining, the cells were centrifuged on slides, and preparation mounted in PBS:glycerol (2:1; Calbiochem, Lucerne, Switzerland).

For CLSM, a Microradiance system from Bio-Rad combined with an inverted Nikon microscope (Eclipse TE3000) was used (Lasers, Ghe/Ne 543 nm and Ar 488 nm). Optical sections at intervals of 0.3 μm were taken with a ×100/1.4 Plan-Apochromat objective. Image processing was done on a Silicon Graphics workstation using IMARIS, a three-dimensional multichannel image processing software (Bitplane AG, Zurich, Switzerland).

Colorimetric Cell Cytotoxicity Study.

Cytotoxic activities of the conjugates were tested by an XTT assay for measuring cell proliferation (31). Briefly, cells (10,000) of both cell lines were seeded in 96-well plates overnight in DMEM with 10% FCS. Cells were then cultured in fresh medium with various concentrations of anti-rNeu antibody (7.16.4), Fab′-Vir, and empty FITC-Vir for 48 h. XTT solution was added according to the manufacturer’s directions (Roche Diagnostics, Rotkreuz, Switzerland). After 4 h of incubation at 37°C, the absorbance was measured using an ELISA reader. Each value represents mean ± SE of three samples.

Treatment of Established Tumors.

Tumor implantation and therapeutic treatment were performed after anesthesia with i.p. injection of medetomidine hydrochloride (Domitor; Orion, Espoo, Finland; 500 μg/kg body weight), climazolamum (Climasol; Gräub, Bern, Switzerland; 5 mg/kg), and fentanyl citrate (Fentanyl-Janssen; Janssen-Cilag, Baar, Switzerland; 50 μg/kg). Mice were shaved at the injection sites, and rNeu+ (NF 9006) and rNeu (M/BB659) cells at a concentration of 2 × 106 were injected s.c. Treatment was started when palpable tumors of at least 5 mm in diameter had formed. Injections into the tail vein of 200 μl of Doxo-Vir (Doxo, 150 μg/ml), Fab′-Doxo-Vir (same Doxo concentration; Fab′ at 182 μg/ml), Fab′-Vir (Fab′ at 182 μg/ml), and free Doxo at a concentration of 150 μg/ml were performed three times/week for the whole observation period. Tumors were measured every 3–4 days, the length and width of each tumor were measured with Vernier calipers. Tumor volume was calculated using the formula: π/6 × largest diameter × (smallest diameter)2.

Treatment of Recently Implanted Tumors.

For the micrometastatic stage of tumor formation, 0.2 × 106 tumor cells were injected s.c., and treatment was started 3–5 days later. Again, different Vir combinations were compared with mice injected with free Doxo and controls. After nine injections (over 3 weeks), the treatment was stopped in all groups. Tumor formation was assessed by palpation, followed by measurement of tumor size as described above. Tumor formation was defined as tumor size beyond possible regression (90 mm3). Mice were sacrificed when tumors exceeded institutional guidelines.

Histological Analysis.

Mice were shaved at the injection sites, and cell suspensions from rNeu+/rNeu breast tumors (2 × 106 in 100 μl of PBS) were injected s.c. in the back. Mice were sacrificed at the indicated time points, and tissues were collected for histological analysis. Deparaffinized slides were stained with H&E.

Anti-rNeu Fab′-Vir Bind to Murine Breast Cancer Cells and Are Efficiently Internalized.

The produced Vir (reconstituted fusion-active viral envelopes), composed of a single phospholipid bilayer and densely covered with HA spikes, were relatively homogeneous in size, ranging from 80 to 200 nm. The PE-PEG-MAL cross-linker was used to conjugate anti-rNeu Fab′ fragments by their free thiol group to the maleimido group. The long PEG spacer arm of the cross-linker allowed an extended and site-directed binding of Fab′ molecules and thus prevented blockage of the antigen binding sites by the neighboring HA trimers. The chosen ratio of total phospholipid:PE-PEG-Fab′ resulted in 100–150 Fab′ fragments/Vir.

To be effective as a drug delivery system, Vir have to bind to tumor targets. FITC-labeled, unconjugated Vir and FITC-labeled Vir conjugated with Fab′ fragments of an anti-rNeu mAb (Fab′-Vir) were analyzed for their binding capacity to breast cancer cell lines. As depicted in the FACS histograms in Fig. 1,A, the NF9006 (rNeu+) breast cancer cell line expressed significant levels of rNeu on the cell surface in comparison with the negative rNeu expression of M/BB 659 breast cancer cells (Fig. 1,C) when the specific anti-rNeu mAb (clone 7.16.4) was used. In contrast to the specific binding with anti-rNeu mAb, Vir showed an increased binding to breast tumor cells independent of their expression of rNeu on the cell surface. We observed that cells overexpressing rNeu had an augmented binding of Fab′-Vir compared with the unconjugated FITC-labeled Vir (Fig. 1,B). Although FITC-Vir and Fab′-Vir showed a strong binding to rNeu breast cancer cells, there was no difference between Fab′-Vir and FITC-Vir (Fig. 1 D). These results indicated that Vir were capable of binding to the tumor cell membranes and that Fab′-Vir showed an increased binding to rNeu+ breast cancer cell lines through the specific binding of anti-rNeu Fab′ fragments.

CLSM was used to study the internalization and distribution of FITC-conjugated virosomes in tumor cells after binding. NF9006 and M/BB 659 breast cancer cells were incubated at 37°C for 1 h with FITC-conjugated Fab′-Vir and Vir. CLSM with a three-dimensional reconstruction showed that in both cell lines, bound Vir were efficiently internalized, as evidenced by the large aggregates of FITC fluorescence observed within the breast cancer cell line (Fig. 2, B and D). There was no visible difference in internalization between Fab′-Vir and Vir (data not shown). By contrast, in NF9006 cells, FITC-conjugated anti-rNeu mAb (clone 7.16.4) was predominantly localized in the membrane or in its close proximity (Fig. 2,A), whereas no FITC fluorescence was observed in M/BB 659 tumor cells (Fig. 2 C). These results suggest that Vir may represent a novel carrier system to deliver encapsulated drugs into the cytosol of solid tumors.

In Vitro Cytotoxicity Studies of Fab′-Vir.

To clarify whether Vir containing anchored Fab′ fragments of an anti-rNeu mAb might have an antiproliferative activity, we investigated the effect of empty Vir, Fab′-Vir, and anti-rNeu mAb on breast cancer cells in vitro. Cells were cultured in the presence of increasing concentrations of anti-rNeu mAb, Fab′-Vir, and Vir, and proliferation of the cells was assessed by the colorimetric XTT assay (10). As shown in Fig. 3, neither rNeu+ nor rNeu breast cancer cells were affected in their proliferation, where different concentrations of empty Vir were added to the cultures. In contrast, mAb 7.16.4 was capable of specifically inhibiting proliferation of rNeu+ breast cancer cells in a dose-dependent manner, whereas the rNeu breast cancer cells were only marginally affected in their proliferation. Monovalent Fab′ fragments are known to be much less effective in the inhibition of proliferation (15); however, the conjugation of monovalent anti-rNeu Fab′ fragments to the surface of Vir showed an important antiproliferative effect on rNeu+ breast cancer cells. Whereas the addition of 10 μg/ml of intact anti-rNeu mAb induced >90% growth inhibition, the addition of 50 μg/ml of Fab′-Vir was necessary to induce a 50% proliferation inhibition. The antiproliferative effect of Fab′-Vir was specific for rNeu+ cells because no inhibitory effect was seen in cultures with rNeu breast cancer cells. Because Vir showed no inhibitory or toxic effect on breast cancer cells in vitro and Fab′-Vir exhibited a dose-dependent inhibition of proliferation of rNeu+ cells, we conclude that the virosomal lipid envelope with inserted HA was not cytotoxic to breast cancer cells.

Treatment of Established Tumors.

In a first set of experiments, we wanted to clarify whether the observed in vitro binding and internalization of Vir into tumor cells also corresponded to an enhanced delivery of encapsulated cytotoxic drugs in vivo. The therapeutic effect on breast tumor implants of i.v.-injected free Doxo was compared with Doxo-encapsulated in Vir (Doxo-Vir). In a second set of experiments, the specific drug targeting with immunovirosomes (Fab′-Doxo-Vir) was tested for an increased therapeutic effect. Mice were inoculated s.c. with 2 × 106 rNeu+ and rNeu tumor cells. After tumor formation (5-mm diameter), treatment was started with i.v. injections of either free Doxo, Doxo-Vir, Fab′-Doxo-Vir, or Fab′-Vir every 3–4 days. Control groups received no treatment. As shown in Fig. 4,A, there was a marginally significant decrease in tumor progression over time in mice treated with Doxo-Vir compared with the control groups. However, there was no significant difference in tumor progression in mice treated with free Doxo or Doxo-Vir. In contrast, as shown in Fig. 4, B and C, tumor progression was almost completely inhibited in the group of mice treated with the Fab′-Doxo-Vir, which was significantly more effective than Doxo-Vir. To determine whether the effect of Fab′-Doxo-Vir was dependent on targeting Doxo to the tumor cells or on the antibody blocking effect of the Fab′ fragments, mice with established tumors were also treated with Fab′-Vir. There was a significant inhibition on tumor progression when Fab′-Doxo-Vir were used for treatment compared with the group of mice where only Fab′-Vir was injected, demonstrating an additive effect of targeting Doxo to the tumor cells by Fab′ fragments on Vir (Fig. 4 C).

The specificity of rNeu targeting was further assessed by treating mice with established rNeu (M/BB 659) tumors with the different Viro compounds. As shown in Fig. 4 D, the treatment of mice with Doxo-Vir, Fab′-Vir, and Fab′-Doxo-Vir had no significant effect on rNeu tumor progression over time as compared with untreated control groups. These results demonstrated that in all experiments, rNeu+ tumor growth was specifically and efficiently suppressed in mice treated with Fab′-Doxo-Vir as compared with mice treated with free Doxo, Doxo-Vir, and Fab′-Vir. Although these results presented a superiority of Fab′-Doxo-Vir in suppressing tumor growth of rNeu+ tumors, these immunovirosomes were unable to completely eradicate established tumors.

Cytotoxicity of Anti-rNeu Immunovirosomes Containing Doxo.

To gain additional insight into the histological pattern induced by the different Vir treatments on inoculated tumors, histopathological sections were prepared at different time points after tumor injection. At days 5–7 after rNeu+ breast cancer cell implantation, tumors were excised and analyzed. In untreated mice, the border of the tumor was well demarcated from the neighboring s.c. normal tissue (Fig. 5,A). Tumor cells appeared uniform with a large, slightly granular cytoplasm, and some nuclei showed mitosis, but no inflammatory infiltrates were detected. In mice treated with Doxo-Vir, again the tumor border was well delineated from the underlying normal tissue, and there was hardly any inflammatory infiltrate in the surrounding s.c. tissue nor any necrotic tumor cells visible (Fig. 5,B). In contrast, in mice treated with Fab′-Vir, a large amount of necrotic cells was found predominantly in the center of the tumors (Fig. 5,C). The surviving tumor cells appeared in contiguous groups surrounded by a significant infiltrate of granulocytic cells also infiltrating the underlying s.c. tissue. In mice treated with Fab′-Doxo-Vir, moreover, most of the tumor cells were necrotic and replaced by an inflammatory infiltrate composed mainly of granulocytes and eosinophils. There was also an impressive granulocytic infiltrate in the vicinity of surviving tumor conglomerates (Fig. 5 D). In contrast, mice injected with the rNeu breast cancer cell line (M/BB 659) showed no necrosis in any of the Vir-treated animals, and some inflammatory infiltration surrounding the tumors was mainly seen in Fab′-Vir- and Fab′-Doxo-Vir-treated mice (data not shown). These results confirmed the effect seen in vivo by Fab′-Doxo-Vir-treated mice on established rNeu+ tumors.

Treatment of Recently Implanted Tumors.

As an alternative evaluation of the efficacy of our virosomal carrier system, we investigated the long-time protection from tumor formation in animals with recently implanted tumors. Therefore, treatment was started 3–5 days after s.c. injection of 2 × 105 Neu+ breast cancer cells into mice. All mice were treated with nine injections of virosomal compounds for 3 weeks, and tumor formation was monitored in the following weeks. Within 4 weeks after tumor cell inoculation, all mice of the control group and groups treated with free Doxo and Doxo-Vir developed tumors and had to be sacrificed because of excessive tumor load (Fig. 6). The median time to tumor formation in mice treated with Doxo-Vir was 20 days and did not significantly differ from mice treated with free Doxo (20 days) or the control group (17 days; P > 0.4–0.8). Whereas mice treated with Fab′-Vir had no significant (P > 0.3) difference in the median time to tumor formation compared with the control group, we noticed that 20% of mice did not develop tumors during the observation period of >90 days. Mice treated with Fab′-Doxo-Vir showed a significant increase (P < 0.005) in time to tumor formation with 90% of mice having no tumor at >90 days after tumor cell inoculation. These data suggest that Fab′-Doxo-Vir are highly efficient in delivering cytotoxic drugs to tumor cells and preventing tumor formation in recently implanted tumors.

The data presented here demonstrate that Vir can be used as a new drug carrier system and be selectively targeted to breast tumors to deliver cytotoxic drugs. Anti-rNeu Vir (Fab′-Vir) had the advantage to bind to the ubiquitous sialic acid residues on the cell surface by HA and retained the binding capacity to rNeu receptors by Fab′ fragment of the mAb (clone 7.16.4). Our data of FACS analysis demonstrated that site-directed conjugation of Fab′ fragments to a cross-linker with a long PEG spacer arm resulted in specific binding of Fab′-Vir to rNeu-expressing cancer cells. By CLSM, we demonstrated that Viro bound to tumor cells became rapidly internalized. In contrast, internalization of targeted liposomes (immunoliposomes) were shown previously to occur slowly, dependent upon the cell surface density of HER-2/neu and the internalization rate of receptor after cross-linking with anti-HER-2/neu mAb (19, 20, 21). The Fab′ fragments of the anti-rNeu mAb retained the antiproliferative effect when coupled to Vir and were linked to the ability to cause receptor internalization (10). Thus far, evidence for an in vivo cytotoxic effect of Doxo-containing liposomes was mostly based on the release of drugs from liposomes into the extracellular space (32, 33). Internalizing epitopes, such as HER-2/neu, are thought to be more efficient at increasing the intracellular drug concentration because entry of the drug into the cells is not dependent upon passive diffusion alone. However, the appropriate selection of mAb and the targeted antigen are crucial for the success of intracellular drug delivery (15, 21). Vir, in contrast, are rapidly and efficiently internalized through receptor-mediated endocytosis and are trapped in the endosomes, where a pH change from 5 to 6 triggers the fusion of the virosomal membrane with the endosomal membrane (22, 34). As a consequence, the encapsulated Doxo is delivered into the cytoplasm.

The in vivo efficacy of Vir loaded with Doxo (Fab′-Doxo-Vir) demonstrated a significant decrease of tumor progression in large established tumors compared with tumor-bearing mice receiving free Doxo or Doxo-Vir (drug-loaded, nontargeted Vir). Although our study confirms the specificity and the increased therapeutic index of Fab′-Doxo-Vir in vivo on rNeu established tumors, it is important to emphasize that both in our work and in previous studies with immunoliposomes, the established tumors were not eradicated completely (35, 36). The inability of immunoliposomes to obliterate established tumors was attributed in part to the specific binding to cell surface receptors at the periphery of the tumor and the ineffectiveness of liposomes to penetrate tumor cells (18, 37). We showed that virosomes have a higher capacity to enter tumor cells; however, the large amount of cells in an established tumor may not be completely attainable for Vir. In contrast, the treatment with targeted Vir of small poorly vascularized tumors prevented the formation of established tumors, thus suggesting that treatment with Vir may be effective in early metastatic spread or combined with other tumor-reducing modalities, such as surgery, radiotherapy, and chemotherapy.

This study also provided some impact with respect to animal models for studying anticancer therapeutics. Thus far, experiments mainly in SCID mice were used for testing anti-HER-2/neu-coated liposomal compounds (15). A considerable difference exists between the tolerated dose of Doxo between SCID mice (2–3 mg/kg) and normal mice (6 mg/kg; Ref. 38). Because SCID mice have a defect in DNA repair mechanisms, these models may not be the appropriate mouse models to test immunotherapeutic strategies. We have used an immune competent mouse strain and used syngeneic breast tumor cell lines derived from the same transgenic mouse (24). In our experiments, mice were injected with 7.5 mg/kg virosomal Doxo, a dose required for successful therapeutic outcome.

Vir may have therapeutic potential, and this report has established a specific targeting by coupling Fab′ fragments to Vir. Our results provide evidence that internalization of Vir occurred, and the encapsulated drug was subsequently delivered to the targeted tumor cells.

Fig. 1.

Vir binding to rNeu+ (NF9006) and rNeu (M/BB659) tumor cell lines. Histograms represent rNeu+ and rNeu tumor cell lines stained with the specific rNeu mAb, followed by FITC-conjugated goat-antimouse IgG (rNeu) or with an isotype-matched primary mAb, followed by FITC-conjugated secondary IgG (neg; A and C). For Vir-binding assays, cells were incubated with either FITC-labeled Vir conjugated with Fab′ fragments of anti-rNeu mAb (Fab′-Vir) or with unconjugated FITC-labeled Vir (Vir.; B and D). Results are expressed as mean fluorescence intensity.

Fig. 1.

Vir binding to rNeu+ (NF9006) and rNeu (M/BB659) tumor cell lines. Histograms represent rNeu+ and rNeu tumor cell lines stained with the specific rNeu mAb, followed by FITC-conjugated goat-antimouse IgG (rNeu) or with an isotype-matched primary mAb, followed by FITC-conjugated secondary IgG (neg; A and C). For Vir-binding assays, cells were incubated with either FITC-labeled Vir conjugated with Fab′ fragments of anti-rNeu mAb (Fab′-Vir) or with unconjugated FITC-labeled Vir (Vir.; B and D). Results are expressed as mean fluorescence intensity.

Close modal
Fig. 2.

Analysis of internalization in CLSM. rNeu+ and rNeu tumor cells were stained with FITC-labeled anti-rNeu mAb (A and C) or FITC-labeled Fab′-Vir (B and D, green) and rhodamine-labeled F-actin (red). Three-dimensional reconstruction: XY projection, A′, B′, C′, D′; XZ projections, A″, B″, C″, D″. Arrows mark the position of projections.

Fig. 2.

Analysis of internalization in CLSM. rNeu+ and rNeu tumor cells were stained with FITC-labeled anti-rNeu mAb (A and C) or FITC-labeled Fab′-Vir (B and D, green) and rhodamine-labeled F-actin (red). Three-dimensional reconstruction: XY projection, A′, B′, C′, D′; XZ projections, A″, B″, C″, D″. Arrows mark the position of projections.

Close modal
Fig. 3.

Antiproliferative activity of Fab′-Vir on rNeu+/rNeu tumor cells in vitro. rNeu+ (closed symbols) and rNeu (open symbols) tumor cells were cultured in monolayers for 48 h with Vir, Fab′-Vir, and anti-rNeu mAb (7.16.4) at the doses indicated. The proliferation of the cells after culture with the different compounds was measured by a colorimetric cytotoxicity assay (XTT). The results shown (means; bars, SE) represent four independent experiments. □, mAb clone 7.16.4; ○, Fab′-Vir; ▿, Vir.

Fig. 3.

Antiproliferative activity of Fab′-Vir on rNeu+/rNeu tumor cells in vitro. rNeu+ (closed symbols) and rNeu (open symbols) tumor cells were cultured in monolayers for 48 h with Vir, Fab′-Vir, and anti-rNeu mAb (7.16.4) at the doses indicated. The proliferation of the cells after culture with the different compounds was measured by a colorimetric cytotoxicity assay (XTT). The results shown (means; bars, SE) represent four independent experiments. □, mAb clone 7.16.4; ○, Fab′-Vir; ▿, Vir.

Close modal
Fig. 4.

Effect of Vir treatment on established rNeu+ and rNeu tumors. rNeu+ tumor cells (2 × 106; A–C) were injected s.c., and treatment was started when tumor size reached 5 mm in diameter. Therapeutic i.v. injections were performed every 3–4 days, and tumor size was assessed twice a week by measuring the length and width of each tumor with Vernier calipers. Tumor volume was calculated using the formula π/6 × largest diameter × (smallest diameter)2. A, treatment with free Doxo and Doxo-Vir compared with control groups. B, Doxo-Vir and Fab′-Doxo-Vir compared with control groups. C, Fab′-Vir and Fab′-Doxo-Vir compared with control groups. These results show the means; bars, SE. ○, control groups; ▾, free Doxo; ♦, Doxo-Vir; •, Fab′-Vir; ▪, Fab′-Doxo-Vir. D, the treatment with injections of Fab′-Doxo-Vir, Doxo-Vir, and free Doxo (all 150 μg/ml Doxo) was evaluated in mice with established rNeu tumors. Tumor sizes are shown at days 8 and 21 after tumor inoculation. □, control groups; , Doxo-Vir; , Fab′-Vir; ▪, Fab′-Doxo-Vir. These are results of four to six experiments, each including three to five mice/group. Bars, SE.

Fig. 4.

Effect of Vir treatment on established rNeu+ and rNeu tumors. rNeu+ tumor cells (2 × 106; A–C) were injected s.c., and treatment was started when tumor size reached 5 mm in diameter. Therapeutic i.v. injections were performed every 3–4 days, and tumor size was assessed twice a week by measuring the length and width of each tumor with Vernier calipers. Tumor volume was calculated using the formula π/6 × largest diameter × (smallest diameter)2. A, treatment with free Doxo and Doxo-Vir compared with control groups. B, Doxo-Vir and Fab′-Doxo-Vir compared with control groups. C, Fab′-Vir and Fab′-Doxo-Vir compared with control groups. These results show the means; bars, SE. ○, control groups; ▾, free Doxo; ♦, Doxo-Vir; •, Fab′-Vir; ▪, Fab′-Doxo-Vir. D, the treatment with injections of Fab′-Doxo-Vir, Doxo-Vir, and free Doxo (all 150 μg/ml Doxo) was evaluated in mice with established rNeu tumors. Tumor sizes are shown at days 8 and 21 after tumor inoculation. □, control groups; , Doxo-Vir; , Fab′-Vir; ▪, Fab′-Doxo-Vir. These are results of four to six experiments, each including three to five mice/group. Bars, SE.

Close modal
Fig. 5.

Histological analysis of rNeu+ tumor injection sites. Paraffin sections from tumor injection sites at day 5 after injections of 2 × 106 rNeu+ tumor cell line (NF9006) into control animals (A), mice treated with Doxo-Vir (B), mice treated with Fab′-Vir (C), and mice treated with Fab′-Doxo-Vir (D) were used. Treatment was performed every second day.

Fig. 5.

Histological analysis of rNeu+ tumor injection sites. Paraffin sections from tumor injection sites at day 5 after injections of 2 × 106 rNeu+ tumor cell line (NF9006) into control animals (A), mice treated with Doxo-Vir (B), mice treated with Fab′-Vir (C), and mice treated with Fab′-Doxo-Vir (D) were used. Treatment was performed every second day.

Close modal
Fig. 6.

Effect of Vir treatment on recently implanted tumors. rNeu+ tumor cells (2 × 105) were injected s.c., and treatment was started 3–5 days later. i.v. injections of the different Vir formulations, such as Doxo-Vir, Fab′-Doxo-Vir, Fab′-Vir, and free Doxo (all at a concentration of 150 μg/ml doxorubicin), were performed at the indicated time points (arrows) for a period of 3 weeks. Tumor size was assessed twice a week during therapy and weekly in the follow-up period. Tumor formation (defined as volume >90 mm3) in the different groups and the follow-up of >12 weeks are shown. Statistical analysis using Mann-Whitney rank test was performed. ○, control groups; ▾, free Doxo; ♦, Doxo-Vir; •, Fab′-Vir; ▪, Fab′-Doxo-Vir.

Fig. 6.

Effect of Vir treatment on recently implanted tumors. rNeu+ tumor cells (2 × 105) were injected s.c., and treatment was started 3–5 days later. i.v. injections of the different Vir formulations, such as Doxo-Vir, Fab′-Doxo-Vir, Fab′-Vir, and free Doxo (all at a concentration of 150 μg/ml doxorubicin), were performed at the indicated time points (arrows) for a period of 3 weeks. Tumor size was assessed twice a week during therapy and weekly in the follow-up period. Tumor formation (defined as volume >90 mm3) in the different groups and the follow-up of >12 weeks are shown. Statistical analysis using Mann-Whitney rank test was performed. ○, control groups; ▾, free Doxo; ♦, Doxo-Vir; •, Fab′-Vir; ▪, Fab′-Doxo-Vir.

Close modal

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.

1

This work was supported by Swiss National Science Foundation Grant 31-59151.99, Swiss Cancer League Grant 268-1-1996, and the Helmut Horton Stiftung (to N. W.) and the Beatrice Borer Stiftung, Rolex, Switzerland (to E. W.). C. D. G. is supported by the Swiss National Science Foundation 32-44482.95.

4

The abbreviations used are: mAb, monoclonal antibody; rNeu, rat Neu; Vir, virosome(s); Doxo, doxorubicin; C12E8, octaethylenglycol mono(n-dodecyl) ether; PE, 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine; PEG, polyethylene glycol; PC, 3-sn-phosphatidylcholine; NHS-PEG-MAL, N-hydroxysuccinimidyl (maleimido) polyethylene glycol; HA, hemagglutinin; XTT, sodium 3′-(1-phenylaminocarbonyl-3,4-tetrazolium)-bis (4-methoxy-6-nitro) benzene sulfonic acid hydrate; CLSM, confocal laser scanning microscopy; FACS, fluorescence-activated cell sorter; SCID, severe combined immunodeficient.

We thank C. Hammer for excellent technical assistance and R. Mühlbauer for statistical help.

1
Karunagaran D., Tzahar E., Beerli R. R., Chen X., Graus-Porta D., Ratzkin B. J., Seger R., Hynes N. E., Yarden Y. ErbB-2 is a common auxiliary subunit of NDF and EGF receptors: implications for breast cancer.
EMBO J.
,
15
:
254
-264,  
1996
.
2
Yamamoto T., Ikawa S., Akiyama T., Semba K., Nomura N., Miyajima N., Saito T., Toyoshima K. Similarity of protein encoded by the human c-erbB-2 gene to epidermal growth factor receptor.
Nature (Lond.)
,
319
:
230
-234,  
1986
.
3
Kern J., Schwartz D., Nordberg J., Weiner D., Greene M., Torney L., Robinson R. p185 expression in human lung adenocarcinomas predicts shorten survival.
Cancer Res.
,
50
:
5184
-5187,  
1990
.
4
Slamon D. J., Godolphin W., Jones L. A., Holt J. A., Wong S. G., Ullrich A., Press M. F. Studies of the HER-2/neu proto-oncogene in human breast and ovarian cancer.
Science (Wash. DC)
,
244
:
707
-712,  
1989
.
5
Niehans G. A., Singleton T. P., Dykoski D., Kiang D. T. Stability of HER-2/neu expression over time and at multiple metastatic sites.
J. Natl. Cancer Inst.
,
85
:
1230
-1235,  
1993
.
6
Lewis G., Figari I., Fendly B., Wong W., Carter P., Gorman C., Shepard H. Differential responses of human tumor cell lines to anti-p185HER2 monoclonal antibodies.
Cancer Immunol. Immunother.
,
37
:
255
-263,  
1993
.
7
Shepard H. M., Lewis G. D., Sarup J. C. Monoclonal antibody therapy of cancer: taking the HER2 protooncogene to the clinic.
J. Clin. Immunol.
,
11
:
117
-121,  
1991
.
8
Baselga J., Norton L., Albanell J., Kim Y-M., Mendelson J. Recombinant humanized anti-HER-2 antibody (Herceptin) enhances the antitumor activity of paclitaxel and doxorubicin against HER2/neu overexpressing human breast cancer xenografts.
Cancer Res.
,
58
:
2825
-2831,  
1998
.
9
Carter P., Presta L., Gorman C., Ridgway J. B., Henner D., Wong W. L., Rowland A. M., Roth C., Caver M. E., Shepard H. M. Humanization of an anti-p185HER2 antibody for human cancer therapy.
Proc. Natl. Acad. Sci. USA
,
89
:
4285
-4289,  
1992
.
10
Zhang H., Wang Q., Montone K., Peavey J., Drebin J., Greene M., Murali R. Shared antigenic epitopes and pathobiological functions of anti-p185 Her2/neu monoclonal antibodies.
Exp. Mol. Pathol.
,
67
:
15
-25,  
1999
.
11
Drebin J. A., Link V. C., Greene M. I. Monoclonal antibodies specific for the neu oncogene product directly mediate anti-tumor effects in vivo.
Oncogene
,
2
:
387
-394,  
1988
.
12
Drebin J. A., Link V. C., Stern D. F., Weinberg R. A., Greene M. I. Down-modulation of an oncogene protein product and reversion of the transformed phenotype by monoclonal antibodies.
Cell
,
41
:
695
-706,  
1985
.
13
McKenzie S., Marks P., Lam T., Morgan J., Panicolli D. L., Trimpe K., Carney W. Generation and characterization of monoclonal antibodies specific for human neu oncogene product p185.
Oncogene
,
4
:
543
-548,  
1989
.
14
Hudziak R., Lewis G., Winget M., Fendly B., Shepard H., Ullrich A. p185 HER2 monoclonal antibody has antiproliferative effects in vitro and sensitizes human breast tumor cells to tumor necrosis factor.
Mol. Cell. Biol.
,
9
:
1165
-1172,  
1989
.
15
Park J., Hong K., Carter P., Asgari H., Guo L., Keller G., Wirth C., Shalaby R., Kotts C., Wood W., Papahadjopoulos D., Benz C. Development of anti-p185 HER2 immunoliposomes for cancer therapy.
Proc. Natl. Acad. Sci. USA
,
92
:
1327
-1331,  
1995
.
16
Papahadjopoulos D., Allen T., Gabizon A., Mayhew E., Huang S., Lee K-D., Woodle M., Lasic D., Redemann C., Martin F. Sterically stabilized liposomes: improvements in pharmacokinetics and antitumor therapeutic efficacy.
Proc. Natl. Acad. Sci. USA
,
88
:
11460
-11464,  
1991
.
17
Gabizon A. Selective tumor localization and improved therapeutic index of anthracyclines encapsulated in long-circulating liposomes.
Cancer Res.
,
52
:
891
-896,  
1992
.
18
Sugano M., Egilmez N., Yokota S., Chen F., Harding J., Huang S., Bankert R. Antibody targeting of doxorubicin-loaded liposomes suppresses the growth and metastatic spread of established human lung tumor xenografts in severe combined immunodeficient mice.
Cancer Res.
,
60
:
6942
-6949,  
2000
.
19
Tagliabue E., Centis F., Campiglio M., Mastroianni A., Martignone S., Pellegrini R., Casalini P., Lanzi C., Menard S., Colnaghi M. I. Selection of monoclonal antibodies which induce internalization and phosphorylation of p185HER2 and growth inhibition of cells with HER2/NEU gene amplification.
Int. J. Cancer
,
47
:
933
-937,  
1991
.
20
Sarup J. C., Johnson R. M., King K. L., Fendly B. M., Lipari M. T., Napier M. A., Ullrich A., Shepard H. M. Characterization of an anti-p185HER2 monoclonal antibody that stimulates receptor function and inhibits tumor cell growth.
Growth Regul.
,
37
:
72
-82,  
1991
.
21
Kirpotin D., Park J., Hong K., Zalipsky S., Li W-L., Carter P., Benz C., Papahadjopoulos D. Sterically stabilized anti-HER2 immunoliposomes: design and targeting to human breast cancer cells in vitro.
Biochemistry
,
36
:
66
-75,  
1997
.
22
Stegmann T., Morselt H., Booy F., van Breemen J., Scherphof G., Wischut J. Functional reconstitution of influenza virus envelopes.
EMBO J.
,
6
:
2651
-2659,  
1987
.
23
Waelti E. R., Glück R. Delivery to cancer cells of antisense L-myc oligonucleotides incorporated in fusogenic, cationic-lipid-reconstituted influenza-virus envelopes (cationic virosomes).
Int. J. Cancer
,
77
:
728
-733,  
1998
.
24
Morrison B. W., Leder P. neu and ras initiate murine mammary tumors that share genetic markers generally absent in c-myc and int-2 initiated tumors.
Oncogene
,
9
:
3417
-3426,  
1994
.
25
Drebin J. A., Link V. C., Weinberg R. A., Greene M. I. Inhibition of tumor growth by a monoclonal antibody reactive with an oncogene-encoded tumor antigen.
Proc. Natl. Acad. Sci. USA
,
83
:
9129
-9133,  
1986
.
26
Shahinian S., Silvius J. R. A novel strategy affords high-yield coupling of antibody Fab′ fragments to liposomes.
Biochim. Biophys. Acta
,
1239
:
157
-167,  
1995
.
27
Martin F., Papahadjopoulos D. Irreversible coupling of immunoglobulin fragments to preformed vesicles.
J. Biol. Chem.
,
257
:
286
-288,  
1982
.
28
Skehel J. J., Schild G. C. The polypeptide composition of influenza-A virus.
Virology
,
44
:
396
-408,  
1971
.
29
Gabizon A., Shiota R., Papahadjopoulos D. Pharmacokinetics and tissue distribution of doxorubicin encapsulated in stable liposomes with long circulating times.
J. Natl. Cancer Inst.
,
81
:
1484
-1488,  
1989
.
30
Hoekstra D., Deboer T., Klappe K., Wilschut J. Fluorescence method for measuring the kinetics of fusion between biological membranes.
Biochemistry
,
23
:
5675
-5681,  
1984
.
31
Jost L., Kirkwood J., Whiteside T. Improved short-and long-term XTT-based colorimetric cellular cytotoxicity assay for melanoma and other tumor cells.
J. Immunol. Methods
,
147
:
153
-165,  
1992
.
32
Horowitz A., Barenholz Y., Gabizon A. In vitro cytotoxicity of liposome-encapsulated doxorubicin: dependence on liposome composition and drug release.
Biochim. Biophys. Acta
,
1109
:
203
-209,  
1992
.
33
Nässander U., Steerenberger P., De Jong W., Van Overveld W., Te Boekhorst C., Poels L., Jap P., Storm G. Design of immunoliposomes directed against human ovarian carcinoma.
Biochim. Biophys. Acta
,
1235
:
126
-139,  
1995
.
34
Matlin K. S., Reggio H., Helenius A., Simons K. Infectious entry pathway of influenza virus in a canine kidney cell line.
J. Cell Biol.
,
91
:
601
-613,  
1981
.
35
Yuan F., Leunig M., Huang S. K., Berk D., Papahadjopoulos D., Jain R. K. Microvascular permeability and interstitial penetration of sterically stabilized (Stealth) liposomes in a human tumor xenograft.
Cancer Res.
,
54
:
3352
-3356,  
1994
.
36
Lopes de Menezes D. E., Pilarski L. M., Allen T. M. In vitro and in vivo targeting of immunoliposomal doxorubicin to human B-cell lymphoma.
Cancer Res.
,
58
:
3320
-3330,  
1998
.
37
Moase E. H., Qi W., Ishida T., Gabos Z., Longenecker B. M., Zimmermann G. L., Ding L., Krantz M., Allen T. M. Anti-MUC-1 immunoliposomal doxorubicin in the treatment of murine models of metastatic breast cancer.
Biochim. Biophys. Acta
,
1510
:
43
-55,  
2001
.
38
Williams S., Alosco T., Mayhew E., Lasic D., Martin F., Bankert R. Arrest of human lung tumor xenograft growth in severe combined immunodeficient mice using doxorubicin encapsulated in sterically stabilized liposomes.
Cancer Res.
,
53
:
3964
-3967,  
1993
.