NOVEMBER/DECEMBER 1997 Volume 8, Number 6 © Copyright 1997 by the American Chemical Society
COMMUNICATIONS Listeriolysin O Potentiates Immunotoxin and Bleomycin Cytotoxicity David E. Kerr,† George Y. Wu,‡ Catherine H. Wu,‡ and Peter D. Senter*,† Bristol-Myers Squibb Pharmaceutical Research Institute, 3005 First Avenue, Seattle, Washington 98121, and Division of Gastroenterology-Hepatology, Department of Medicine, University of Connecticut Health Center, 263 Farmington Avenue, Farmington, Connecticut 06030-1845. Received July 3, 1997X
Antitumor immunotoxins were formed by covalently attaching the ribosome-inactivating protein ricin A chain (RA) to the antitumor antibodies BR96 and L6. In vitro cytotoxicity assays established that BR96-RA was cytotoxic to H2987 human lung adenocarcinoma cells (IC50 ) 6 nM), while L6-RA exhibited very low levels of cytotoxic activity (18% cell kill at 67 nM). The virulence factor from the intracellular pathogen Listeria monocytogenes, listeriolysin O (LLO), was able to potentiate the cytotoxicity of BR96-RA and L6-RA by 120- and >1340-fold, respectively, resulting in IC50 values of approximately 50 pM. LLO also potentiated the cytotoxicity of the peptide anticancer drug bleomycin by a factor of >2500 but had no effect on the cytotoxic activities of the anticancer drugs cytarabine and etoposide phosphate. In addition, LLO did not potentiate the cytotoxic activity of unconjugated ricin A chain or L6-RA on H2987 cells that were saturated with L6 prior to conjugate treatment. These results are attributed to LLO-induced alteration of the intracellular trafficking of molecules that are incorporated into acidic vesicles.
INTRODUCTION
A great deal of research has been directed toward the use of antitumor monoclonal antibodies (mAb1) for the delivery of ricin A chain (RA) and other toxins to tumors (reviewed in refs 1-3). Therapeutic efficacy with such immunotoxins requires that sufficient quantities bind to cell surface receptors, undergo intracellular uptake, and * Author to whom correspondence should be addressed. † Bristol-Myers Squibb. ‡ University of Connecticut. X Abstract published in Advance ACS Abstracts, October 15, 1997. 1 Abbreviations: mAb, monoclonal antibody; RA, ricin A chain; IC50, concentration resulting in 50% cell death; LLO, listeriolysin O.
S1043-1802(97)00124-9 CCC: $14.00
then access the cytosol where the toxin exerts its activity (4, 5). Entry into the cytosol has been identified as a common limitation in immunotoxin efficacy (4-6), and a number of methods have been described to enhance cytosolic delivery. These include the use of lysosomotropic amines such as ammonium chloride (7), ionophores such as monensin (8, 9), and, more recently, a 25 amino acid peptide derived from protein G of vesicular stomatitis virus (10). These agents reduce immunotoxin degradation by raising lysosomal pH (7), by altering protein trafficking within the cell (8), or by destabilizing intracellular membranes in lower pH environments (10). Thus, there is considerable evidence that improvements in immunotoxin activity can result if the process of cytosolic delivery is enhanced. Listeriolysin O (LLO) is a major virulence factor from © 1997 American Chemical Society
782 Bioconjugate Chem., Vol. 8, No. 6, 1997
Kerr et al.
the intracellular pathogen Listeria monocytogenes. Upon uptake of the bacteria into target cells, secreted LLO disrupts endosomes and lysosomes, allowing the bacteria to enter the cytoplasm and grow (11-14). One of the key differences between LLO and other hemolytic proteins is that its pH optimum is close to 5.5, and at pH 7 the activity is almost undetectable (11). For these reasons, it seemed reasonable that LLO might potentiate immunotoxin activity by facilitating its transport out of acidic vescicles. The same might be true for anticancer drugs that have restricted intracellular access. Here, we describe the activities of LLO in combination with immunotoxins and clinically approved anticancer drugs. We show that LLO can have a pronounced effect on the cytotoxic activities of selected antitumor agents. EXPERIMENTAL PROCEDURES
Materials. The H2987 human lung carcinoma cell line has been described before (15). Bleomycin sulfate, cytarabine, bovine serum albumin, and dithiothreitol were purchased from Sigma (St. Louis, MO), and etoposide phosphate was provided by Bristol-Myers Squibb. The mAbs BR96 (16) and L6 (17) were conjugated to deglycosylated ricin A chain (Inland Laboratories, Austin, TX) through a disulfide bond using previously described methods (16). Isolation of LLO. The hemolytic L. monocytogenes strain, MAC, a LLO hypersecretor, was a kind gift from Dr. Daniel A. Portnoy. MAC cells were grown overnight in 10 mL of brain heart infusion broth (Difco) at 37 °C, and the culture was continued in 1 L of LB medium at 37 °C with shaking until A660nm was ∼0.6. After centrifugation (10000g, 15 min at 4 °C), the supernatant was concentrated at 4 °C to 200 mL using an Amicon CH2PRS-spiral cartridge concentrator with an Amicon S1Y30 spiral cartridge. The concentrated supernatant was applied to a 40 mL Macro-Prep ceramic hydoxyapatite column (80 µm particles, Bio-Rad) that was equilibrated with 10 mM potassium phosphate buffer at pH 6.8 at 4 °C. LLO was eluted with 400 mM potassium phosphate buffer at pH 6.8. The hemolysis assay on human erythrocytes at pH 6.0 was performed as previously described (11), and it was found that the purified protein contained 167 000 units/mg. In Vitro Cytotoxicity. H2987 cells in Iscove’s modified Dulbecco’s medium supplemented with 10% fetal bovine serum containing penicillin (60 µg/mL) and streptomycin (100 µg/mL) were plated into 96-well-microtiter plates at 104 cells/well and allowed to adhere overnight at 37 °C. The cells (previously untreated or saturated with unconjugated L6 at 1 mg/mL) were then treated with L6-RA or BR96-RA at various concentrations at 37 °C in the presence or absence of LLO at 1 µg/mL. Control cells were treated with LLO or ricin A chain alone. After 18 h, the plates were washed three times with Iscove’s modified Dulbecco’s medium, incubated for an additional 8 h, and pulsed for 7 h with [3H]thymidine (1 µCi/well). The plates were frozen at -20 °C and thawed, and the cells were harvested on a Tomtec 96-well harvester. Radioactivity was counted on a Wallac beta plate counter. In a separate experiment, cells were treated with bleomycin, cytarabine, or etoposide phosphate in the presence or absence of 1 µg/mL LLO. After 2 h, the plates were washed, incubated overnight at 37 °C, and pulsed for 7 h with [3H]thymidine (1 µCi/well). The plates were then treated as above. RESULTS AND DISCUSSION
RA was conjugated to the antitumor antibodies L6 (17) and BR96 (16) through a disulfide-containing linker as previously described (16). The conjugates thus formed
Figure 1. Cytotoxic effects of L6-RA and BR96-RA in the presence or absence of LLO as determined by the inhibition of [3H]thymidine into DNA: (A) H2987 cells incubated for 18 h with L6-RA or BR96-RA ( LLO at 1 µg/mL, washed, and pulsed with [3H]thymidine; (B) cytotoxic effects of LLO without immunoconjugate treatment; (C) H2987 cells incubated for 18 h with RA or L6-RA ( LLO at 1 µg/mL and treated as in (A). Some of the cells were treated with unconjugated L6 (1 mg/mL) and incubated at 4 °C for 30 min, followed by L6-RA ( LLO. The data represent the mean of three samples ( standard deviations.
were capable of binding to cell surface L6 and BR96 antigens, which are present in high density on the H2987 human lung adenocarcinoma cell line. These particular antibodies were selected for the studies described here since they display significantly different characteristics upon antigen binding. BR96 is known to be very rapidly internalized into cells (16), while L6 remains primarily extracellular but gradually internalizes and is degraded (18, 19). It was therefore not surprising to find that BR96-RA was cytotoxic to H2987 cells (IC50 ) 6 nM), while L6-RA exhibited very low levels of activity (18% cell kill at 67 nM) (Figure 1A). Once internalized into antigen-positive cells, the potency of immunotoxins is often limited by their ability to escape from acidic intracellular compartments to the cytosol. Agents that are capable of facilitating this transfer have been shown to lead to enhanced levels of cytotoxicity (7-10). Toward this end, we explored the activities of LLO, a 58 kDa protein that enables the pathogenic bacterium L. monocytogenes to escape from acidic vesicles into the cytosol of mammalian host cells
Communications
Bioconjugate Chem., Vol. 8, No. 6, 1997 783
LLO at a completely nontoxic concentration (1 µg/mL) was able to enhance the activity of bleomycin by 2500fold, resulting in an IC50 of 8 nM. The effect of LLO was not general, in that the protein had no effect on the activities of etoposide phosphate and cytarabine (Figure 2B). Etoposide phosphate, being negatively charged, is expected to have limited intracellular access and has been shown to be significantly less cytotoxic than etoposide (25). These results, taken together with those shown in Figure 1, are therefore in general agreement with previously studies showing that LLO has an intracellular site of activity rather than a nonspecific mechanism involving disruption of the plasma membrane (11-14). The data reported here indicate that LLO complements other known immunotoxin potentiators (7-10) and has the additional property of being able to significantly increase the potency of bleomycin, a clinically approved anticancer drug. It may be possible to exploit these activities by combining the RA conjugates, bleomycin, or other appropriate anticancer drugs with a mAb-LLO conjugate that binds to the target cell population of interest. Such studies will be the subject of future investigations. ACKNOWLEDGMENT
Figure 2. Cytotoxic effects of (A) bleomycin and (B) cytarabine and etoposide phosphate in the presence or absence of LLO as determined by the inhibition of [3H]thymidine into DNA. H2987 cells were incubated for 2 h with bleomycin, cytarabine, or etoposide phosphate ( LLO at 1 µg/mL, washed, incubated further, and then pulsed with [3H]thymidine.
(11-14). Purified LLO at 0.1, 1, and 10 µg/mL was noncytotoxic to H2987 cells (Figure 1B). When LLO at 1 µg/mL was combined with BR96-RA or L6-RA, the resulting activity far exceeded that of either immunotoxin alone (Figure 1A). In both cases, the IC50 was reduced to 50 pM, corresponding to 120- and >1340-fold increases in activity for BR96-RA and L6-RA activity, respectively. LLO did not significantly enhance the activity of ricin A chain alone, a molecule that does not show detectable binding to H2987 cells (Figure 1C). In this same experiment, it was found that the effect of LLO on L6-RA cytotoxic activity could be abrogated by saturating the cells with unconjugated L6 prior to the addition of LLO and the L6-RA chain conjugate. LLO therefore appears to mediate its effects on molecules that bind to cell surface receptors and are then actively transported inside cells. These results led us to explore whether LLO could potentiate the activity of some clinically approved antitumor agents. Bleomycin is used for the treatment of lymphoma, squamous cell carcinoma, and testicular cancer (20). The activity of this drug on mammalian cells has been attributed to the induction of DNA (20) and RNA (21) strand breaks. It has previously been reported that bleomycin binds in a saturable manner to a membrane protein that is thought to play a role in its internalization (22). This process is most likely inefficient, on the basis of the finding that bleomycin potency is greatly increased when target cell membranes are permeablized by electroporation (23, 24). In a 2 h exposure assay on H2987 lung adenocarcinoma cells, bleomycin had an IC50 value of 20 µM (Figure 2A).
We thank Nathan Siemers, Jacques Garrigues, and Cherie Walton for their valuable contributions to the research described here. This work was supported in part by the U.S. Public Health Service, NIDDK Grant DK42182 (G.Y.W.), and a grant from TargeTech, Inc./ Immune Response Corp. (C.H.W.). G.Y.W. and C.H.W. hold equity in the Immune Response Corp. LITERATURE CITED (1) Vitetta, E. S. (1994) From the basic science of B cells to biological missiles at the bedside. J. Immunol. 153, 14071420. (2) Press, O. W. (1991) Immunotoxins. Biotherapy 3, 65-76. (3) Pastan, I., Pai, L. H., Brinkmann, U., and FitzGerald, D. (1996) Recombinant immunotoxins. Breast Cancer Res. Treat. 38, 3-9. (4) Byers, V. S., Pawluczyk, I. Z., Hooi, D. S., Price, M. R., Carroll, S., Embleton, M. J., Garnett, M. C., Berry, N., Robins, R. A., and Baldwin, R. W. (1991) Endocytosis of immunotoxin791T/36-RTA by tumor cells in relation to its cytotoxic action. Cancer Res. 51, 1990-1995. (5) Bilge, A., Howell-Clark, J., Ramakrishnan, S., and Press, O. W. (1994) Degradation of ricin A chain by endosomal and lysosomal enzymessthe protective role of ricin B chain. Ther. Immunol. 1, 197-204. (6) Kornfeld, S. B., Leonard, J. E., Mullen, M. D., and Taetle, R. (1991) Assessment of ligand effects in intracellular trafficking of ricin A chain using anti-ricin hybridomas. Cancer Res. 51, 1689-1693. (7) Carayon, P., Bord, A., Gaillard, J. P., Vidal, H., Gros, P., and Jansen, F. K. (1993) Ricin A-chain cytotoxicity depends on its presentation to the cell membrane. Bioconjugate Chem. 4, 146-152. (8) Marsh, J. W. (1989) Cellular processing of a ricin-antibody conjugate. J. Biol. Chem. 264, 10405-10410. (9) Dosio, F., Franceschi, A., Ceruti, M., Brusa, P., Cattel, L., and Colombatti, M. (1996) Enhancement of ricin toxin A chain immunotoxin activity: synthesis, ionophoretic ability, and in vitro activity of monensin derivatives. Biochem. Pharmacol. 52, 157-166. (10) Chignola, R., Anselmi, C., Dalla Serra, M., Franceschi, A., Fracasso, G., Pasti, M., Chiesa, E., Lord, J. M., Tridente, G., and Colombatti, M. (1995) Self-potentiation of ligand-toxin conjugates containing ricin A chain fused with viral structures. J. Biol. Chem. 270, 23345-23351. (11) Geoffroy, C., Gaillard, J. L., Alouf, J. E., and Berche, P. (1987) Purification, characterization, and toxicity of the
784 Bioconjugate Chem., Vol. 8, No. 6, 1997 sulfhydryl-activated hemolysin listeriolysin O from Listeria monocytogenes. Infect. Immun. 55, 1641-1646. (12) Mengaud, J., Vincente, M.-F., Chenvert, J., Pereira, J. M., Geoffrey, C., Gicquel-Sanzey, B., Baquero, F., Perez-Diaz, J.C., and Cossart, P. (1988) Expression in Escherichia coli and sequence analysis of the listeriolysin O determinant of Listeria monocytogenes. Infect. Immun. 56, 766-772. (13) Cossart, P., Vicente, M. F., Mengaud, J., Baquero, F., Perez-Diaz, J.-C., and Berche, P. (1989) Listeriolysin O is essential for virulence of Listeria monocytogenes: direct evidence obtained by gene complementation. Infect. Immun. 57, 3629-3636. (14) Bielecki, J., Youngman, P., Connelly, P., and Portnoy, D. A. (1990) Bacillus subtilis expressing a haemolysin gene from Listeria monocytogenes can grow in mammalian cells. Nature 345, 175-176. (15) Svensson, H. P., Vrudhula, V. M., Emswiler, J. E., MacMaster, J. F., Cosand, W. L., Senter, P. D., and Wallace, P. M. (1995) In vitro and in vivo activities of a doxorubicin prodrug in combination with monoclonal antibody β-lactamase conjugates. Cancer Res. 55, 2357-2365. (16) Hellstro¨m, I., Garrigues, H. J., Garrigues, U., and Hellstro¨m, K. E. (1990) Highly tumor-reactive, internalizing, mouse monoclonal antibodies to Ley-related cell surface antigens. Cancer Res. 50, 2183-2190. (17) Hellstro¨m, I., Horn, D., Linsley, P., Brown, J. P., Brankovan, V., and Hellstro¨m, K. E. (1986) Monoclonal mouse antibodies raised against human lung carcinoma. Cancer Res. 46, 3917-3923. (18) Mattes, M. J., Griffiths, G. L., Diril, H., Goldenberg, D. M., Ong, G. L., and Shih, L. B. (1994) Processing of antibodyradioisotope conjugates after binding to the surface of tumor cells. Cancer 73, 787-793.
Kerr et al. (19) Schmidberger, H., King, L., Lasky, L. C., and Vallera, D. A. (1990) Antitumor activity of L6-ricin immunotoxin against the H2981-T3 lung adenocarcinoma cell line in vitro and in vivo. Cancer Res. 50, 3249-3256. (20) Chabner, B. A., and Myers, C. E. (1993) Antitumor antibiotics. Cancer: Principles and Practice of Oncology (V. T. J. DeVita, S. Hellman, and S. A. Rosenberg, Eds.) pp 374376, J. B. Lippincott Co., Philadelphia. (21) Hecht, S. M. (1994) RNA degradation by bleomycin, a naturally occurring bioconjugate. Bioconjugate Chem. 5, 513526. (22) Pron, G., Belehradek, J., Jr., and Mir, L. M. (1993) Identification of a plasma membrane protein that specifically binds bleomycin. Biochem. Biophys. Res. Commun. 194, 333337. (23) Shimizu, N., and Kawazoe, Y. (1996) A new method for permeabilization of cultured cells without cell damage. Biol. Pharm. Bull. 19, 484-486. (24) Tounekti, O., Pron, G., Belehradek, J., Jr., and Mir, L. M. (1993) Bleomycin, an apoptosis-mimetic drug that induces two types of cell death depending on the number of molecules internalized. Cancer Res. 53, 5462-5469. (25) Senter, P. D., Saulnier, M. G., Schreiber, G. J., Hirschberg, D. L., Brown, J. P., Hellstrom, I., and Hellstrom, K. E. (1988) Anti-tumor effects of antibody-alkaline phosphatase conjugates in combination with etoposide phosphate. Proc. Natl. Acad. Sci. U.S.A. 85, 4842-4846.
BC970124+