[CANCER RESEARCH 54, 1465-1471. March 15. W4] AA1, A Newly Synthesized Monovalent Lipophilic Cation, Expresses Potent in Vivo Antitumor Activity Xiaocheng Sun, James R. Wong, Kaili Song, Julian Hu, Keith D. Garlid, and Lan Bo Chen1 Dana-Farber Cancer Institute, Hazard Medical School, Boston, Massachusetts 02115 [J. R. W., L. B. C.j, and Department ¡ifPharmacology, Medical College of Ohio, Toledo, Ohio 43614 ¡X.S., K. S., J. H., K. D. G.J ABSTRACT Certain lipophilic cations have been reported to display anticarcinoma activities because of their selective uptake and retention by mitochondria of cancer cells. Thus, these agents may comprise a unique class of agents directed against carcinoma. After screening more than 1000 lipophilic cations, we found that the monovalent lipophilic cation, 2,6-bis(4-amino- phenyl)-4-[4-(dimethylamino)phenyl]thiopyrylium chloride (AA1), dis played remarkable anticarcinoma activity both in vitro and in vivo. Unlike most other lipophilic cations, AA1 is stable and displays minimal light sensitivity. /// vitro testing showed that AA1 was 10 times more toxic to the carcinoma cell line (VI than to the normal epithelial cell line CV-1. In vivo animal experiments showed that AA1 significantly prolonged the survival of mice implanted with tumors. For C57BL x DBA/2 F, mice implanted with the mouse bladder carcinoma cell line, MB49, the treat- ed:control ratio was 344%. For Swiss nulnu mice implanted i.p. with the human melanoma cell line, LOX, the treated:control ratio was 341%. The most significant observation was obtained with Swiss nulnu mice that were implanted ¡.p.with the human ovarian cell line, OVCAR-III. The treat- ed:control ratio in this situation was greater than 450%. In all these tumor models, AA1 produced minimal toxicities. AA1 exhibited little inhibition of electron transport in isolated rat liver mitochondria; however, it inhibited mitochondria! ATPase with 50% inhibitory concentration of 6 /UM.Com pared with previously reported anticarcinoma lipophilic cations such as rhodamine 123 and dequalinium chloride, AA1 appeared to display more effective in vivo anticarcinoma activity. Thus, AA1 could be considered for further clinical development as a candidate for anticarcinoma chemo therapy. INTRODUCTION One of the recent advances in the field of anticarcinoma chemo- therapeutic agents is the finding that certain lipophilic cations such as rhodamine 123 (1-4), dequalinium chloride (5-8), /V"-bis(2-ethyl-l, 3-dioxolane)-krytocyanine (9), Victoria blue BO (10), and tetraphe- nylphosphonium (11) are more selectively toxic to carcinoma cells than their normal counterparts. This in vitro selectivity of lipophilic cations between cancer cells and normal cells has been found to range from 10- to 100-fold. It has been hypothesized that the difference in transmitochondrial membrane potentials between normal cells and carcinoma cells may be responsible for the increased uptake and prolonged retention of these delocalized lipophilic cations, thus ac counting for their in vitro selective toxicities to carcinoma cells (1-6, 12). The in vivo anticarcinoma activities of these delocalized lipophilic cations have been examined. For example, rhodamine 123 was able to prolong the median survival of mice implanted with Ehrlich ascites tumor and the mouse bladder carcinoma cell line MB49 by 260 and 179%, respectively (3). Later experiments with other lipophilic cat ions showed that dequalinium chloride appeared to be a more potent anticancer agent with a treatedrcontrol ratio of 252% for mice im planted i.p. with MB49 cells (5). These encouraging results confirmed Received 12/17/92; accepted 1/18/94. 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. ' To whom requests for reprints should be addressed, at Dana-Farber Cancer Institute. Harvard Medical School, 44 Binney Street, Boston, MA 02115. the hypothesis that lipophilic cations may comprise a novel class of anticarcinoma agents. Because these agents are localized at the mito chondria, it is possible that these compounds derive their anticarci noma activity from inhibition of energy supplies rather than from the inhibition of DNA, RNA, or protein synthesis, which are common mechanisms of most chemotherapeutic agents in use today. Several laboratories have used lipophilic cations to enhance the effects of existing anticancer modalities. For instance, Powers et al. (13) showed that rhodamine 123 can be used to sensitize glioma cells to laser therapy. Oseroff et al. (9) showed that /V'-bis(2-ethyl-l,3- dioxolane)-krytocyanine can be used for photodynamic therapy. Teicher et al. (14) showed that rhodamine 123 and a rhodamine 123-cisplatinum complex were able to sensitize cells to hyperthermia and radiation therapy. Thus, delocalized lipophilic cations, in addition to their own intrinsic anticarcinoma effects, may also comprise a new class of sensitizers for hyperthermia, radiation, or photodynamic therapies. We undertook the screening of more than 1000 delocalized li pophilic cations in vitro to look for agents that were selectively toxic to carcinoma cells. Some of these compounds were synthesized fol lowing the study of structure-effect relationships using results ob tained with in vitro or in vivo assays. Many of these lipophilic cations were able to inhibit the proliferation of cancer cells in vitro. However, most of these compounds had limitations regarding solubility, stabil ity, light sensitivity, or excessive in vivo toxicity. One of these agents, AA1 (2, 6-bis(4-aminophenyl)-4-[4-(dimethylamino)phenyl] thiopyrylium chloride (molecular structure as shown in Fig. 1), appeared to display marked anticarcinoma effects, without the aforementioned shortcom ings. As shown in Fig. 1,AA1 has a highly delocalized positive charge in the structure and thus would be accumulated selectively by the mitochondria of carcinoma cells as proposed previously (1, 2, 6, 8). We will show in this paper that: (a) in in vitro testing, AA1 was 10-fold more toxic to carcinoma cells (CX-1) than to normal cells (CV-1); (b) in in vivo testing, AA1 remarkably prolonged the survival of mice in three different tested tumor models; and (c) electron trans port in isolated, normal rat liver mitochondria was minimally affected by AA1, whereas mitochondrial ATPase activity was strongly inhib ited by AA1. Many chemotherapeutic agents in use today have low selectivities between carcinoma cells and normal cells (15). Moreover, because of intrinsic or acquired resistance to the existing anticancer agents (16, 17), there is a need for novel anticancer agents, especially those with mechanisms different from current existing anticancer drugs. Li pophilic cations such as AA1, which target the mitochondria, are thus very attractive candidates as novel anticancer agents. MATERIALS AND METHODS Cell Cultures. CX-1 cells were grown in 50% RPM1 1640 and 50% DMEM2 (GIBCO Laboratories, Grand Island, NY) supplemented with 5% calf -The abbreviations used are: DMEM. Dulbecco's modified Eagle's medium; KOAc, potassium acetate; IC.si», 50% inhibitory concentration; MDR, multidrug resistance; T:C, treatedicontrol ratio. 1465 Research. on November 14, 2015. © 1994 American Association for Cancer cancerres.aacrjournals.org Downloaded from