(CANCER RESEARCH 48, 3496-3501, June 15, 1988] Mediation of Reduction of Spontaneous and Experimental Pulmonary Métastases by Ricin A-Chain Immunotoxin 45-2D9-RTA with Potentiation by Systemic Monensin in Mice1 Jack A. Roth,2 Robert S. Ames, Kim Fry, Howard M. Lee, and Patrick J. Scannen Department of Thoracic Surgery [J. A. R., R. S. A., K. F.], and the Department of Tumor Biology [J. A. R., R. S. A.], The University of Texas System Cancer Center, M. D. Anderson Hospital and Tumor Institute, Houston, Texas 77030, and Xoma Corporation [H. M. L., P. J. SJ, Berkeley, California 94710 ABSTRACT We developed a model to assess the therapeutic effects of the 45-2D9- ricin A-chain immunotoxin (RTA) on pulmonary métastases.The 45- 21W mouse monoclonal antibody recognizes a M, 74,000 glycoprotein highly expressed by rat fibroblasts transformed with the Kirsten sarcoma virus (transformed rat fibroblasts). These cells metastasi/e spontaneously and form lung colonies in nu/nu and irradiated BALB/c mice. Injection i.v. of 45-21W-RTA specifically reduced formation of spontaneous pul monary métastasesand lung colonies originating from freshly disaggre gated tumor cells or cultured cells. Antibody alone or mixed with uncon- jugated ricin A chain and an immunotoxin that recognizes a melanoma- associated antigen were ineffective. Unconjugated -45-21Wantibody spe cifically blocked the 45-2D9-RTA activity in vivo. Administration of the lysosomotrophic agents ammonium chloride and chloroquine in vivo did not potentiate immunotoxin-mediated reduction in lung colonies although they were effective in vitro. Monensin potentiated 45-2D9-RTA activity in vitro and in vivo. INTRODUCTION Monoclonal antibodies that recognize tumor cells selectively could potentially function as delivery vehicles for cell toxins. The A chain of the toxin ricin, derived from the castor bean, has been extensively investigated as an immunotoxin compo nent (1). Theoretical considerations that suggest potential clin ical utility for ricin A chain immunotoxins include a low prob ability of drug resistance, the ability to couple the purified A chain, which is less nonspecific-ally toxic than whole ricin, to antibody, and an efficient catalytic mechanism of cell kill once the molecule enters the cytosol (2). Previous reports of the therapeutic efficacy of immunotoxins for solid tumors have been limited to treatment of primary s.c. and i.d.3 tumors and to local-regional administration of im munotoxin (3-6). A model closely approximating the common clinical situation of established metastatic solid tumor would be useful for assessing immunotoxin efficacy for adjuvant ther apy and provide a system for evaluating potentiating agents that may increase immunotoxin efficacy by altering host mech anisms responsible for inactivation of the toxin component. Our laboratory has developed a mouse monoclonal antibody (45-2D9), which recognizes a M, 74,000 serine-phosphorylated glycoprotein that is expressed by some ras oncogene-trans- formed cells (7, 8). The 45-2D9 antibody also recognizes an epitope on human carcinomas with minimal expression by normal human tissues (8). An immunotoxin was developed by conjugating the A chain of the toxin ricin to the antibody. This Received10/27/87;revised1/14/88;accepted3/16/88. The costs of publication of this article were defrayed in part by the payment of page charges. This anido must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. ' This study was supported in part by research grant CA4S187 from the National Cancer Institute. 2To whom requests for reprints should be addressed. 3The abbreviations used are: i.d., intradermal; PBS, phosphate-buffered saline; gp74, a M, 74,000 glycoprotein; SDS-PAGE, sodium dodecyl sulfate-polyacryl- amide gel electrophoresis: DPBS, Dulbecco's phosphate-buffered saline; DMSO, dimethyl sulfoxide; RTA, ricin A-chain toxin; TRF, transformed rat fibroblasts. immunotoxin demonstrated highly specific toxicity by killing 3 to 4 logs of target cells in vitro. In this report, we describe development of a model of both experimental and spontaneous established métastasesto test the efficacy of the systemically administered immunotoxin in vivo and to assess the effects of potentiating agents administered systemically. MATERIALS AND METHODS Purification of Monoclonal Antibody 45-21)9. The 45-2D9 antibody was purified from either ascites or spent hybridoma supernatant. For ascites production, pristane-primed BALB/c mice were inoculated i.p. with 1 x IO6 hybrids/mouse. The ascites was dialyzed against 0.14 M NaPO4, pH 8, and then centrifugea to remove debris. Ascites (70 ml total) or supernatant was then passed over a 1.5- x 30-cm column packed with a 40-ml bed of Protein A agarose (Sigma, St. Louis, MO) equilibrated with 0.14 M N;iI'O4. The column was then washed until the A28oreturned to baseline. Bound immunoglobulin was eluted with pH steps of 8.0, 5.0, 4.5, and 3.5. The 45-2D9 antibody eluted at pH 4.5 without contaminating isotypes. The purity of the eluted antibody was monitored by two-dimensional SDS-PAGE under reducing condi tions. The column was cleared with citrate buffer, pH 3.5, and reequil- ibrated with 0.14 M NaPO4, pH 8.O. Pooled fractions of antibody were dialyzed against DPBS, 200 v/v, with three changes at 24-h intervals, then aliquoted at 1 mg/vial, and lyophilized for storage. The isotype from each lot was routinely checked with reagents supplied in a murine immunoglobulin subtype identification kit (Boehringer Mannheim, In dianapolis, IN). Immunoglobulin concentration was determined by radial immunodiffusion. Preparation of Immunotoxin. Ricin A-chain and the 45-2D9 immu notoxin were prepared by Xoma Corporation (Berkeley, CA). The technique of Kernan et al. (9) was used to purify the ricin A chain and synthesize the antibody-toxin conjugate. Ricin A-chain was purified by affinity column chromatography. Enzymatic activity was assayed as the ability to inhibit protein synthesis in a cell-free reticulocyte lysate system. All lots used for conjugate preparation must have values for an inhibitory concentration for a 50% decrease in protein synthesis of 1 to 5 x 10"' g/ml. The preparation's purity was monitored by SDS- PAGE on 12.5% gels under nonreducing conditions and a contaminat ing band corresponding to native ricin was not detected. Antibody-toxin conjugates were prepared by coupling ricin A chain to antibody molecules by a disii Hide linkage. Antibody 45-2D9 (1 to 2 mg/ml) was dialyzed against 0.1 M NaPCVO.l M NaCl, pH 7.7, and a 15- to 20-fold M excess of Af-succinimidyl-3-(2-pyridyldithio) propionate was added with vigorous mixing. After incubation at room temperature for 30 min, the solution was dialyzed against two changes of PBS. After dialysis, the pyridyl-dithiopropionate group-to-antibody ratio was determined as described by Carlsson et al. (10). The ricin A- chain was reduced by addition of dithiothreitol to a final concentration of 50 HIM, followed by incubation for l h at room temperature. The solution was dialyzed against PBS (4°C)to remove residual reducing agent. The ricin A-chain was concentrated using an Amicon stirred cell with a YM10 membrane to a final concentration of 4 mg/ml. Five- or 10-fold M excess ricin A-chain was added to the pyridyl-dithiopropion- ate-antibody solution and incubated for 16 h at 4"( '. Preparations of antibody conjugates were purified from free ricin A- chain by sizing chromatography using either G-150 or ACA 44 resins. The purified conjugate was then concentrated with an Amicon YM30 3496 on June 3, 2017. © 1988 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from