4606 Vol. 9, 4606–4613, October 1, 2003 Clinical Cancer Research Retinoid Receptor-Dependent and Independent Biological Activities of Novel Fenretinide Analogues and Metabolites Anita L. Sabichi, Hui Xu, Susan Fischer, dependent. 4-MPR, a major metabolite of 4-HPR, lacks a Changchan Zou, Xiulan Yang, Vernon E. Steele, charged group on the terminal phenylamine ring and did Gary J. Kelloff, Reuben Lotan, and not induce retinoid receptor-dependent effects, but did in- 1 duce cell growth inhibition. Thus, 4-MPR may play a role in John L. Clifford the clinical activity of 4-HPR. This study further reveals the Departments of Clinical Cancer Prevention [A. L. S., X. H., C. Z., mechanism of action of these novel phenylretinamides and X. Y., J. L. C.], Thoracic/Head and Neck Medical Oncology [R. L.], supports continued investigation into their development as and Carcinogenesis [S. F.], The University of Texas M. D. Anderson Cancer Center, Houston, Texas 77030, and National Cancer Institute, chemopreventive drugs. Bethesda, Maryland 20892 [V. E. S., G. J. K.] INTRODUCTION Retinoids are a class of chemical compounds, which in- ABSTRACT clude active metabolites of vitamin A (retinol) as well as a Fenretinide (4-HPR) is a retinoid analogue with antitu- diverse array of synthetic derivatives. Retinoids modulate a mor and chemopreventive activities. In addition to 4-HPR, wide variety of cellular processes, including proliferation, dif- there are several other new phenylretinamides bearing hy- ferentiation, homeostasis, malignant transformation, and apo- droxyl, carboxyl, or methoxyl residues on carbons 2, 3, and ptosis (1). Retinoids also act pharmacologically to restore reg- 4 of the terminal phenylamine ring [N-(2-hydroxyphenyl)- ulation of differentiation and growth in certain premalignant and retinamide (2-HPR), N-(3-hydroxyphenyl)retinamide, N-(2- malignant cells in vitro and in vivo. It is now well established carboxyphenyl)retinamide, N-(3-carboxyphenyl)retinamide, that retinoids exert their effects primarily through nuclear reti- N-(4-carboxyphenyl)retinamide, and N-(4-methoxyphenyl)- noid receptor proteins. Retinoid receptors comprise two families retinamide (4-MPR) ]. It is hypothesized that these agents of ligand-dependent, DNA-binding, transcriptional transactiva- can act independent of the nuclear retinoid receptor path- 2 tors, RARs and RXRs, both members of the nuclear steroid way. To test this hypothesis directly, we have analyzed the hormone receptor superfamily (1, 2). Fenretinide (4-HPR) is one activity of these phenylretinamides in vitro on a panel of F9 of the most promising retinoids for chemoprevention (3–5). It murine embryonal carcinoma cell lines, which includes wild- has a favorable toxicity profile, potent apoptosis-induction ac- type (F9-WT) and mutant cells that have disrupted genes for tivity, biological activity in several preclinical systems, includ- both retinoid X receptor ␣ and retinoic acid receptor ␥ ing all-trans- and 9-cis-retinoic acid-resistant neoplastic cells, retinoid receptors (F9-KO). The F9-KO cells lack almost all and significant clinical chemopreventive activity in randomized measurable response to all-trans-retinoic acid, the primary trials for breast cancer and oral carcinogenesis (6, 7). In addition biologically active retinoid. Two distinct effects of reti- to 4-HPR, there are several other new phenylretinamides bear- namides were identified. The first is a rapid, dose-dependent ing hydroxyl, carboxyl, or methoxyl residues on carbons 2, 3, induction of cell growth inhibition (reduced cell viability), and 4 of the terminal phenylamine ring (2-HPR, 3-HPR, 2-CPR, and the second is a slower induction of differentiation and 3-CPR, 4-CPR, and 4-MPR). One of these, 4-MPR, is a princi- accumulation of cells in the G phase of the cell cycle that 1 ple metabolite of 4-HPR in rodents and humans (8–10) and is ␮ was observed with a concentration of 1 M, for only those the only one of these that lacks a charged group on the terminal phenylretinamides bearing charged (hydroxyl or carboxyl) ring. To date there have been few studies aimed at determining groups on the terminal phenylamine ring. The induction of the activity of these compounds, with the exception of 4-CPR, differentiation and G accumulation was only observed in 1 which has been used extensively in Phase II clinical trials in the F9-WT cells, indicating that this effect is receptor- China (11). The activity of this group of phenylretinamides for the growth inhibition of human oral epithelial cells (12), bladder transitional cell carcinoma cells (13), head and neck squamous cell carcinoma cells and non-small cell lung cancer cells (14), Received 7/31/02; revised 5/14/03; accepted 5/21/03. 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. 2 The abbreviations used are: RAR, retinoic acid receptor; RXR, retinoid This work was supported in part by National Cancer Institute Grants 5 X receptor; 4-HPR, N-(4-hydroxyphenyl)retinamide; 2-HPR, N-(2- R29 CA78560, and 5 U01 CA77150, and National Institute of Environ- hydroxyphenyl)retinamide; 3-HPR, N-(3-hydroxyphenyl)retinamide; mental Health Sciences Grant 5 P30 ES07784. 2-CPR, N-(2-carboxyphenyl)retinamide; 3-CPR, N-(3-carboxyphenyl)- 1 To whom requests for reprints should be addressed, at Louisiana State retinamide; 4-CPR, N-(4-carboxyphenyl)retinamide; 4-MPR, N-(4- University Health Sciences Center and Feist-Weiller Cancer Center, methoxyphenyl)retinamide; RA, all-trans retinoic acid; WT, wild-type; Department of Biochemistry and Molecular Biology, 1501 Kings Hwy., F9-KO, F9 RXR␣Ϫ/Ϫ/RAR␥Ϫ/Ϫ; DAPI, 4Ј,6-diamidino-2-phenylin- P.O. Box 33932, Shreveport, LA 71130. Phone: (318) 675-8264; Fax: (318) dole; KO, knockout; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphe- 675-5180; E-mail: [email protected]. nyltetrazolium bromide. Downloaded from clincancerres.aacrjournals.org on September 30, 2021. © 2003 American Association for Cancer Research. Clinical Cancer Research 4607 has been determined recently. These studies have shown that all each well followed by mild shaking to dissolve the insoluble of the novel phenylretinamides can induce cell death and sup- dark blue crystals. Absorbance was determined for each well press proliferation to various degrees, depending on the cell using a Biolumin 960 multiwell plate reader (Molecular Dy- type, and that these effects were often comparable with 4-HPR, namics Inc.) at an absorbance wavelength of 570 nM. Standard indicating that they may also have chemopreventive potential curves for cell number were generated for the F9-WT and (13). F9-KO cells by plating a broad range of known cell numbers Previous studies have shown that 4-HPR could induce (102-5 ϫ 104 cells/well) and performing the MTT assay in the apoptosis in cells that are resistant to the most biologically same day. Cell number for the treated cell samples was calcu- active retinoid, all-trans-RA, suggesting that this activity may lated from these curves. The percentage growth inhibition was not involve retinoid receptors in some cell types (15–19). To calculated using the equation: (1-R/C) ϫ 100, where R and C directly test the hypothesis that 4-HPR can act through a recep- represent the number of cells in retinoid-treated and control tor-independent mechanism, we used the F9 murine embryonal cultures, respectively. The cell cycle profile of F9-WT and carcinoma retinoid receptor KO system previously. The F9 cell F9-KO cells treated for 4 days with 1 ␮M retinoids was deter- line is a well-established model system for the study of RA- mined by cell cycle flow cytometry based on cellular DNA induced differentiation (20). Investigators have used homolo- content, using an Epics Profile II cell sorter (Coulter Electron- gous recombination-mediated gene targeting to generate F9 ics, Inc.), as described previously (13). The percentage of cells cells that lack expression of RARs ␣ and ␥, RXR␣, and pairwise in different phases of the cell cycle was determined from the raw combinations of RAR␣/RXR␣ and RAR␥/RXR␣ (Refs. 21, 22; data using the Epics Elite Flow Cytometry software. reviewed in Refs. 23, 24). It was determined that knocking out Western Blotting. Whole cell extracts were purified and the most abundant RXR (RXR␣) and the most abundant RAR Western blotting performed as described previously (21). Blots (RAR␥) in the same cell, abolishes almost all of the measurable were probed with a polyclonal goat antibody to mouse laminin retinoid receptor-mediated effects (22). By comparing the ef- B1 (Santa Cruz Biotechnology Inc.), stripped, and reprobed with fects of 4-HPR on cell growth and differentiation between the a mouse monoclonal antibody to ␤-actin (Amersham). Second- F9-WT and F9-KO cells, our group could show that 4-HPR had ary antibodies used were horseradish peroxidase-conjugated an- two distinct effects (25). The first was a rapid induction of tigoat or antimouse (Zymed). Chemiluminescence detection was apoptosis requiring a high concentration (10 ␮M), which was performed according to the manufacturer’s instructions (Amer- observed in both the F9-WT and F9-KO cells. The second was sham Life Science Inc.) followed by autoradiography. a slower induction of differentiation, accompanied by an accu- Immunocytochemical Staining. F9 cells were grown on mulation of cells in the G1 phase of the cell cycle, which is seen gelatinized glass coverslips in the presence or absence of the with a lower concentration (1 ␮M). This effect was only ob- indicated retinoids for 4 days. After fixation in 1% paraformal- served in the F9-WT cells, leading to the conclusion that the dehyde, coverslips were incubated with Superblock blocking rapid high concentration effects are receptor-independent and solution (Pierce) for1hatroom temperature to block nonspe- the delayed low concentration effects are receptor dependent.