Search for Novel Anti-tumor Agents from Ridaifens Using JFCR39, a Panel of Human Cancer Cell Lines

Abstract

To overcome the heterogeneous nature of cancer, the search for potent anti-cancer drug candidates with new modes of action is essential. For that purpose, we prepared forty-eight Ridaifens (RIDs), a novel series of tamoxifen-derivatives. Then, we screened them, searching for novel candidates for a new class of anti-cancer drug using a panel of human cancer cell lines (JFCR39) and by a binding assay to estrogen receptor α (ERα). First, the growth inhibition of the forty-eight RIDs against JFCR39 was evaluated. Forty RIDs showed higher growth-inhibitory activity than that of tamoxifen. The structure–activity relationship (SAR) study revealed that the aminoalkoxyphenyl groups at the C-1 position and the common central ethylenic bond were important in retaining a high level of growth-inhibitory activity. Subsequently, the ERα binding activity of all the RIDs was measured by a competitive binding assay. The SAR study for ERα binding activity indicated that both the phenyl group and the ethyl group at the C-2 position in the ethylenic bond were essential. Based on the screenings, we identified RID-SB1 and RID-SB8, which demonstrated potent tumor growth inhibition but had completely lost ERα binding activity. Furthermore, the COMPARE analysis using JFCR39 suggested that RID-SB1 and RID-SB8 had different molecular modes of action compared to those of the current anti-cancer drugs including tamoxifen. These results indicate that RID-SB1 and RID-SB8 are interesting candidates for novel anti-cancer agents with unique modes of action.

Tamoxifen is an antagonist of the estrogen receptor (ER). It has been used as the first-line agent for breast cancer for more than 30 years although competitive estrogen inhibitors have been developed to treat hormonally responsive breast cancer.^1–3) The accumulative risk-benefit assessment of tamoxifen therapy has established its efficacy and safety.^4–6) Many studies of the action mode of tamoxifen have been done, which have revealed that tamoxifen is not only an estrogen blocker but also a modulator of various signaling proteins located in ER-independent biological pathways; for example protein kinase C (PKC), calmodulin, transforming growth factor β and protooncogene c-myc. Interestingly, it has been suggested that other molecules, such as caspases, mitogen-activated protein kinases (MAPK), c-Jun N-terminal kinase (JNK) and p38, are involved in tamoxifen-induced apoptotic signaling.^7,8) These reports indicate that tamoxifen potentially acts on multiple targets. Therefore, tamoxifen can be used as a lead compound to produce a new generation of anti-cancer drug with a unique mode of action, which will be used for other cancers in addition to breast cancer.

We developed a new efficient synthetic method for tamoxifen via a three-component coupling reaction.⁹⁾ Then we synthesized a series of tamoxifen derivatives, designated as Ridaifens (RIDs), expecting that they would display a variety of biological activities. As expected, we demonstrated the anti-tumor activity of three RIDs, RID-B, C, and D.¹⁰⁾ In this study, we further modified the structure of tamoxifen, and obtained a new series of RIDs consisting of forty-eight RIDs over three generations of modifications.

To identify anti-cancer drug candidates that have potentially unique modes of action, we had previously established a panel of 39 human cancer cell lines, designated as JFCR39, which was coupled to an anti-cancer drug activity database and a computer algorithm COMPARE.^11–13) This system provides rich and fundamental information on the pharmacological action modes of chemicals on cancer cells. The COMPARE algorithm enabled us to identify novel drug candidates with potentially unique modes of action compared to current anticancer drugs. Moreover, it sometimes predicted the molecular targets or the modes of action of novel compounds. Indeed, by running the JFCR39 system and COMPARE, we successfully identified novel anti-cancer agents, such as a telomerase inhibitor (FJ-5002),¹⁴⁾ a topoisomerase I/II inhibitor (MS247),¹⁵⁾ a phosphatidylinositol 3-kinase inhibitor (ZSTK474),^16–18) a Golgi inhibitor (AMF-26)¹⁹⁾ and so on.

In the present study, we screened the forty-eight RIDs with the JFCR39 panel to identify compound that have potent growth-inhibitory activity and unique modes of action. We also evaluated the RIDs for their estrogen receptor (ER) binding activity to identify compounds which have lost ER binding activity. As a result, we found two RIDs, RID-SB1 and RID-SB8, that had potent growth-inhibitory activity against JFCR39 and had lost ERα binding activity.

MATERIALS AND METHODS

Chemicals

Forty-eight RIDs were synthesized over three generations of modifications. The 1st-generation of RIDs is shown in Fig. 1A. When aminoalkyl group R of the RID was R^A, R^B, R^C, R^D, R^E, R^F, R^G or R^H, each derivative was named as RID-A, RID-B, RID-C, RID-D, RID-E, RID-F, RID-G or RID-H, respectively. The 2nd-generation RIDs are based on modifications of the ethylene moiety in the 1st-generation RIDs (Fig. 1B). There are two types of the 2nd-generation RIDs, which include R^B or R^G for the aminoalkyl group R. For example, RID-SB1 has aminoalkyl group R^B as a side-chain and RID-SG1 has aminoalkyl group R^G as a side-chain. The 3rd-generation RIDs possess two identical alkyl chains (≥ C₃ (propyl) group) at the C-2 position, giving rise to symmetrical structures. There are three types of the 3rd-generation RIDs, bearing R^B, R^F or R^G for the aminoalkyl group R (Fig. 1C). 4-Hydroxytamoxifen and tamoxifen were purchased from Sigma (St. Louis, U.S.A.). The physical properties of RID-SB1 and RID-SB8 and their ¹H- and ¹³C-NMR charts were indicated in Supplemental information 1.

Fig. 1. Chemical Structures of Tamoxifen, 4-Hydroxytamoxifen and RIDs

(A) Tamoxifen, 4-hydroxytamoxifen and first-generation RIDs. (B) Second-generation RIDs. (C) Third-generation RIDs.

Cell Lines

A panel of 39 human cancer cell lines, known as JFCR39, containing the following cell lines: lung cancer, NCI-H23, NCI-H226, NCI-H522, NCI-H460, A549, DMS273 and DMS114; colorectal cancer, HCC-2998, KM-12, HT-29, HCT-15 and HCT-116, gastric cancer, MKN-1, MKN-7, MKN-28, MKN-45, MKN-74, St-4; ovarian cancer, OVCAR-3, OVCAR-4, OVCAR-5, OVCAR-8 and SK-OV-3; breast cancer, BSY-1, HBC-4, HBC-5, MDA-MB-231 and MCF-7; renal cancer, RXF-631L and ACHN; melanoma, LOX-IMVI; glioma, U251, SF-295, SF539, SF268, SNB75 and SNB78; prostate cancer, DU-145 and PC-3, was used as described previously (1–3). All the cell lines were cultured in RPMI 1640 medium supplemented with 5% (v/v) fetal bovine serum, penicillin (100 U/mL) and streptomycin (100 mg/mL) at 37°C in a humidified atmosphere containing 5% (v/v) CO₂.

Determination of Cell Growth Inhibition Profiles (Fingerprint)

Inhibition of cell growth was assessed by the change in total cellular protein following 48 h of treatment with a given test compound, and was measured using the sulforhodamine B (SRB) assay as described previously.²⁰⁾ The 50% growth inhibition (GI₅₀) value of the drug was calculated as described previously.²¹⁾ A graphic representation (termed fingerprint) of the differential growth inhibition of each compound for the cells in the JFCR39 panel was plotted based on a calculation employing a set of GI₅₀ values.²²⁾

COMPARE Analysis

We used COMPARE analysis to assess two compounds for the similarity of their mode of action based on their fingerprints. COMPARE analysis was performed by calculating the Pearson correlation coefficient (r) between the GI₅₀ mean graphs of the compounds X and Y using the following formula: r=(∑ (x_i−x_m)(y_i−y_m))/(∑ (x_i−x_m)² ∑ (y_i−y_m)²)^1/2, where x_i and y_i are Log GI₅₀ of the two compounds, respectively, for each cell line, and x_m and y_m are the mean values of x_i and y_i, respectively (n=39).²²⁾ The Pearson correlation coefficients were used to determine the degree of similarity. The greater the coefficient is, the higher the similarity between X and Y.

Estrogen Receptor Binding Affinity Assay

The ERα binding affinity assay was performed using recombinant ERα (Thermo, Massachusetts, U.S.A.) and the HitHunter Enzyme Fragment Complementary (EFC) Estrogen Receptor Assay kit (Discoverx Corporation, Fremont, CA, U.S.A.) according to the manufacturer’s protocol.²³⁾ HitHunter EFC technology is based on a genetically engineered β-galactosidase enzyme that consists of two fragments termed enzyme acceptor (EA) and enzyme donor conjugated with estradiol (ED-estradiol). Briefly, different concentrations of RIDs were added to wells containing recombinant ERα and ED-estradiol in a 384-well white plate and incubated for 1.5 h. ED-estradiol competes with RIDs in binding to ER. Any unbound ED-estradiol conjugates with EA to form an active β-galactosidase enzyme, which subsequently hydrolyzes the fluorescent substrate for luminescent detection by a microplate reader (Bio-Rad, California, U.S.A.). The excitation wavelength is 530 nm and luminosity is detected at 620 nm. The competition activity was calculated using the following formula: percent competition=(X-positive ER control)/(negative ER control-positive ER control)×100%, where X is the value of luminescent detection of RIDs. A standard curve of estradiol was run in parallel. All assays were performed in triplicates.

RESULTS

Growth-Inhibitory Activities of RIDs against JFCR39

The growth-inhibitory activities of forty-eight RIDs against JFCR39 were determined by SRB assay as described in Materials and Methods. Figure 2 shows the dose response curves of the cells in the JFCR39 panel against tamoxifen and RID-SB1 as examples. The concentration at which the cell growth is inhibited by 50% represents GI₅₀. Supplemental information 2 summarizes the GI₅₀ values of the forty-eight RIDs against each cancer cell line in JFCR39. The mean value of GI₅₀ over all 39 cell lines (designated as MG-MID) is summarized in Table 1. The MG-MID of the RIDs ranged from 0.85 µm (RID-G) to 43.7 µm (RID-SB4). Based on the MG-MID, we found that forty of the forty-eight RIDs showed higher proliferation inhibition than tamoxifen (MG-MID=7.41 µm).

Fig. 2. Dose Response Curves of (A) Tamoxifen and (B) RID-SB1 against Growth of JFCR39 Cancer Cell Lines

The growth inhibition of tamoxifen and RID-SB1 on 39 cell lines (Br, breast; CNS, central nervous system; Co, colorectal; Lu, lung; Me, melanoma; Ov, ovarian; Re, renal; St, stomach; xPg, prostate) were measured as described in Materials and Methods.

Table 1. Anti-tumor Activity and ERα Binding Capacity of RIDs

Chemical	MG-MID (µm)	ERα binding activity IC50 (nm)	Generation	Chemical	MG-MID (µm)	ERα binding activity IC50 (nm)	Generation
17β-Estradiol	—	1.89		RID-SG2	15.50	>10000	2nd
4-Hydroxytamoxifen	5.50	2.72		RID-SG3	6.46	>10000	2nd
Tamoxifen	7.41	21.1		RID-SG4	26.3	>10000	2nd
RID-A	1.48	171	1st	RID-SG7	17.4	>10000	2nd
RID-B	1.17	52.4	1st	RID-SG8	1.55	2910	2nd
RID-C	3.47	295	1st	RID-SG9	1.82	8570	2nd
RID-D	14.8	47.7	1st	RID-SG10	1.45	372	2nd
RID-E	1.20	320	1st	RID-SG11	1.55	127	2nd
RID-F	2.45	154	1st	RID-SG12	1.38	23.9	2nd
RID-G	0.85	26.7	1st	RID-SG13	1.32	234	2nd
RID-H	1.05	44.1	1st	RID-SG14	1.51	109	2nd
RID-SB1	1.38	>10000	2nd	RID-SG15	1.51	835	2nd
RID-SB2	5.40	>10000	2nd	RID-SG16	1.10	1230	2nd
RID-SB3	10.5	>10000	2nd	RID-SB22	1.51	383	3rd
RID-SB4	43.7	>10000	2nd	RID-SB23	1.70	253	3rd
RID-SB7	5.77	>10000	2nd	RID-SB17	2.00	380	3rd
RID-SB8	1.82	>10000	2nd	RID-SB24	2.63	568	3rd
RID-SB9	2.69	4640	2nd	RID-SF22	1.79	421	3rd
RID-SB10	1.51	2370	2nd	RID-SF23	6.03	656	3rd
RID-SB11	1.44	582	2nd	RID-SF17	12.0	428	3rd
RID-SB12	1.34	145	2nd	RID-SF24	17.0	1840	3rd
RID-SB13	1.31	218	2nd	RID-SG22	1.05	253	3rd
RID-SB14	1.44	280	2nd	RID-SG23	1.70	387	3rd
RID-SB15	1.62	4880	2nd	RID-SG17	1.78	515	3rd
RID-SB16	1.46	702	2nd	RID-SG24	2.04	1560	3rd
RID-SG1	2.29	>10000	2nd

The 1st-generation RIDs each have a pair of identical substituents (-OR) on the phenyl groups bound to the central ethylene moiety at the C-1 position (Fig. 1A). Here, the aminoalkoxy substituents were modified without changing the structures of the third phenyl group and the ethyl group at the C-2 position. The 1st-generation RIDs all showed higher anti-tumor activity than tamoxifen, except RID-D (MG-MID=14.8 µm). These results indicate that the modifications of aminoalkyl group R with R^A–R^H except R^D enhanced the growth-inhibitory activity.

All the 2nd- and 3rd-generation RIDs have two identical aminoalkoxy substituents on both phenyl groups at the C-1 position (Figs. 1B, C). We selected a number of representative aminoalkyl groups for the generation of further variations of RIDs (R=R^B or R^G in the 2nd-generation; R=R^B, R^F or R^G in the 3rd-generation). The central ethylenic bond was then either removed (S4 type in Fig. 1B), or modified with two substituents at the C-2 position. The S4 type RIDs showed remarkably reduced activities compared to tamoxifen (MG-MID=7.41 µm) and all other RIDs. The MG-MIDs of RID-SB4 and RID-SG4 were 43.7 µm and 26.3 µm, respectively (Table 1). These results indicate that the central ethylenic bond is important for enhancing growth-inhibitory activity. On the other hand, most of the RIDs with modifications to the two C-2 substituents on the central ethylene bond did not exhibit a significant change in their growth-inhibitory activity compared to 1st-generation RIDs. The 3rd-generation RIDs have completely symmetrical structures, in which two identical alkyl chains of various lengths (C3–C6) are bound to the central ethylenic moiety. Most of them retain high levels of growth-inhibitory activity.

Each test compound was further characterized by its GI₅₀ profile across JFCR39, which was designated as “fingerprint.” For example, the fingerprints of tamoxifen and RID-SB1 and RID-SB8 (Supplemental information 1) were shown in Fig. 3. According to the analysis by COMPARE, the correlation coefficients between the fingerprints of tamoxifen and RID-SB1, and between tamoxifen and RID-SB8 were 0.199 and 0.318, respectively. These results suggest that RID-SB1 and RID-SB8 are different from tamoxifen with respect to their mode of action. Indeed, these two RIDs exhibited no ERα binding activity (see below).

Fig. 3. Fingerprints of Tamoxifen, RID-SB1 and RID-SB8

Fingerprint shows the differential growth inhibition pattern of chemicals against JFCR39 cancer cell lines. The X-axis represents difference in logarithmic scale between mean of Log GI₅₀ values for 39 cell lines and the Log GI₅₀ for each cell line. Bars to the right of 0 indicates cell lines that are sensitive to the compound, in contrast, bars on the left of 0 means the resistance. MG-MID, mean of Log GI₅₀ values for 39 cell lines; Delta, difference between the MG-MID and the Log GI₅₀ value for the most sensitive cell line; Range, difference between the Log GI₅₀ values for the most resistant cell and the most sensitive cell line.

ERα Binding Activities of RIDs

The ERα binding activity of the forty-eight RIDs was determined by the competitive binding of ED-Estradiol to ERα. The competitive binding curves of estradiol (positive control), tamoxifen and 4-hydroxytamoxifen are shown in Fig. 4. The y-axis indicates the percent inhibition of ED-Estradiol binding to ERα. Among the RIDs, RID-G and RID-SB1 indicated the strongest and the lowest ERα binding activity, respectively (Fig. 4).

Fig. 4. Competitive Binding Curves of Estradiol, Tamoxifen, 4-Hydroxytamoxifen, RID-SB1 and RID-G

ERα competitive binding activity was measured as described in Materials and Methods. Estradiol, tamoxifen, 4-hydroxytamoxifen, RID-G and RID-SB1 are represented as ○, ◇, □, ■, ▲, respectively.

The IC₅₀ values of all of the RIDs are shown in Table 1. The relationship between chemical structure and ERα binding activity was studied here. The 1st-generation RIDs (RID-A–H) with the various substituents on the phenyl rings at the C-1 position yielded IC₅₀ values ranging from 26.6 (RID-G) to 320 nm (RID-E) (Table 1). These results indicated that the structures of substituents on the phenyl rings at the C-1 position in RIDs considerably affected the binding activity to ERα. RID-G showed ERα binding activity almost equal to tamoxifen. However, no RID compound surpassed 4-hydroxytamoxifen, an active form of tamoxifen, in ERα binding activity.

An interesting finding was observed in the ERα binding activities of the 2nd-generation RIDs. These RIDs were used to study the importance of the substituents around the C-2 position of the central ethylenic bond while retaining the two aromatic rings with identical aminoalkoxy groups at C-1. It was observed that the eight RIDs, RID-SB1, SB2, SB3 SB4, SG1, SG2, SG3 and SG4, had completely lost ERα binding activity. Structurally, these RIDs had lost either one or both of the phenyl group or the ethyl group at the C-2 position. These results clearly indicate that the phenyl and ethyl group at the C-2 position in the ethylenic bond are essential. The impact of modifications of the substituents around the ethylenic bond (RID-SB7 to SB16 and RID-SG7 to SG16) was also examined. The ERα binding activity varied significantly (23.89–>10000 nm) depending on the modification. A trend was observed whereby the RIDs having more bulky structures around the double bond (RID-SB7 to SB12 and SG7 to SG12) displayed higher ERα binding activities (Table 1).

The 3rd-generation RIDs (RID-SG22, SG23, SG17, SG24, SB22, SB23, SB17, SB24, SF22, SF23, SF17 and SF24) (Fig. 1C) and some of the 2nd-generation RIDs (RID-SG9, SG10, SB9, SB10, SF9 and SF10) (Fig. 1B) have symmetrical structures. These RIDs showed lower ER binding activity in comparison to the 1st-generation RIDs (Table 1). RID-SG22, SG23, SG17, SG24, SB22, SB23, SB17, SB24, SF22, SF23, SF17 and SF24 (3rd-generation RIDs, Fig. 1C) have longer alkyl chains (≥C₃ (propyl) group) than RID-SG9, SG10, SB9, SB10, SF9 and SF10 (2nd-generation, Fig. 1B), and exhibited greater activities than RID-SG9, SG10, SB9, SB10, SF9 and SF10.

Two RIDs with Notably High ER-Independent Anti-tumor Activities

To evaluate the RIDs in terms of growth-inhibitory activity and ERα binding activity, a scattergram of growth inhibition (x-axis) and ERα binding activity (y-axis) for the RIDs was drawn (Fig. 5). Each spot represents one RID compound and both activities varied over a wide range. Here, we focused on two RIDs, RID-SB1 and RID-SB8 (Supporting Information 1), that had both completely lost ERα binding activity, while exhibiting the high levels of growth-inhibitory activity, suggestive that they inhibited the cell growth via an ER-independent mechanism. To confirm this point, we examined the fingerprints of the two RIDs (Fig. 3). The COMPARE analysis of RID-SB1 and RID-SB8 revealed that both of them showed a very weak correlation coefficient (r < 0.4) with tamoxifen (Table 2). These results suggest that both RID-SB1 and RID-SB8 have a different pharmacological mode of action to that of tamoxifen.

Fig. 5. The Scattergram of RIDs for the Proliferation Inhibition Activity and ERα Binding Activity

The X-axis (-MG-MID) represents the proliferation inhibition of forty-eight RIDs in 39 cell lines. The Y-axis (−Log(IC₅₀)) represents the competitive ERα binding activity. Tamoxifen, RID-SB1 and SB8 are represented as ◇, □, △, respectively.

Table 2. COMPARE Analysis of RID-SB1 and RID-SB8

Chemical	Ranking	Drug or inhibitor	r	Function
RID-SB1	1	NVP-AEW541	0.505	IGF-1R inhibitor
	2	Bortezomib	0.481	Proteasome inhibitor
	3	RDEA119	0.474	MEK inhibitor
	:	:	:	:
	>100	Tamoxifen	0.199	ER antagonist
RID-SB8	1	PB28 dihydrochloride	0.563	sigma-2 Receptor agonist
	2	MEK inhibitor I	0.549	MEK inhibitor
	3	Raf1 Kinase Inhibitor I	0.549	Raf1 inhibitor
	:	:	:	:
	>100	Tamoxifen	0.314	ER antagonist

a) The ID for the chemical in JFCR39 database.

COMPARE Analysis of RID-SB1 and RID-SB8

We previously established the JFCR39 drug database and COMPARE analysis and demonstrated that this system is a powerful tool for use in identifying the molecular targets or pharmacological modes of action of novel compounds.^11–13) The JFCR39 drug database integrates the fingerprints of reference compounds including 87 standard drugs currently used in clinic and more than 1100 inhibitors of various enzymes and biological pathway. To investigate the possible modes of action of RID-SB1 and RID-SB8, we carried out the COMPARE analysis. The two RIDs did not correlate with any currently used anti-cancer drugs (r<0.4). Table 2 summarized the top three reference compounds, which were extracted from the database due to the similarity of their fingerprint profiles to RID-SB1 or RID-SB8. It was indicated that RID-SB1 was most similar to NVP-AEW541 (IGF-1R inhibitor), Bortezomib (proteasome inhibitor) and RDEA119 (MEK inhibitor). RID-SB8 was most similar to PB28 dihydrochloride (sigma-2 receptor agonist), MEK inhibitor I (MEK inhibitor) and Raf 1 Kinase Inhibitor I (Raf 1 kinase inhibitor). Although the correlation coefficients were not high enough, the targets of these reference compounds could be the targets of RID-SB1 or RID-SB8.

DISCUSSION

Tamoxifen performs its anti-tumor activity via many pathways, both ER-dependent and ER-independent. In the present study, we attempted to identify novel anti-tumor compounds among RIDs that were designed from a lead compound, tamoxifen. We synthesized forty-eight RIDs by efficient methods including the three-component coupling reaction.⁹⁾ To screen them for promising anti-tumor compounds, we used the JFCR39 panel and an ER binding assay. By this means, we found two RIDs, RID-SB1 and RID-SB8, that showed stronger growth-inhibitory activity than tamoxifen and are expected to have unique modes of action.

The forty-eight RIDs showed growth-inhibitory activities against JFCR39 over a wide range of MG-MIDs from 0.85 to 43.7 µm. Forty out of forty-eight RIDs (83%) showed higher growth-inhibitory activity than tamoxifen (MG-MID=7.41 µm). The SAR study indicated that the structures of aminoalkoxyphenyl groups at the C-1 position and the common central ethylenic double bond were important in retaining a high level of growth-inhibitory activity.

The ERα binding activity of the RIDs was spread across a huge range of IC₅₀ values from 26.7 to >10000 nm. In general, the 1st-generation RIDs exhibited ERα binding activities. Among them RID-G showed the most potent activity (IC₅₀=26.6 nm), which is close to the activity of tamoxifen (IC₅₀=21.1 nm). In contrast, a number of RIDs in the 2nd-generation were inactive with respect to ERα binding. The SAR study clearly indicated that both the phenyl and ethyl groups at the C-2 position of the central ethylenic bond were essential for ERα binding activity. In addition, there was a tendency for RIDs bearing more bulky structures around the double bond (RID-SB7 to SB12 and SG7 to SG12) to display higher ERα binding activity.

According to the above observations, the forty-eight RID compounds were profiled by growth-inhibitory activity and ERα binding activity (Fig. 5). Interestingly, these two activities didn’t correlate in the forty-eight RIDs. Some of them, such as RID-G, showed high levels in both activities, while others were active in only one. In the latter group, we focused on RID-SB1 and RID-SB8 because they were highly active in growth-inhibition but were inactive in regard to ERα binding. It was suggested that they inhibited the cell growth via an ER-independent mechanism and thus may inhibit the growth of cancer cells by a different mode of action than tamoxifen. The COMPARE analysis was therefore performed on this study, and it revealed that both RID-SB1 and RID-SB8 showed very weak correlation coefficients (r < 0.4) with tamoxifen. These results support the hypothesis above.

So far, a number of tamoxifen derivatives have been designed and synthesized with the aim of enhancing ER binding activity and reducing side-effects. However, this has not been entirely successful. For example, clomifene,²⁴⁾ ospemifene,^25,26) iodoxifene,²⁷⁾ raloxifene,²⁷⁾ arzoxifene,²⁸⁾ lasofoxifene²⁹⁾ and levormeloxifene,^30,31) possess high ER binding activity, while still having such disadvantages as biotransformation effect, biological isomerization, thromboembolic effect and agonist action in uterus.³²⁾ Most of these tamoxifen derivatives show very similar effects as tamoxifen and have been proposed considering only the effect on ER. In the present study, RID-SB1 and RID-SB8, while having completely lost ER binding activity, display approximately 5-fold higher antitumor activity than tamoxifen, indicating a non-ER target in its anti-tumor process. Such non-ER function might be useful in unveiling new targets to inhibit tumor cell proliferation. Therefore, RID-SB1 and RID-SB8 maybe novel drug candidates for breast cancer but also for other malignancies.

We finally attempted to predict the action modes of RID-SB1 and RID-SB8 by using our JFCR39 drug database and COMPARE analysis. The two RIDs did not correlate with any currently available anti-cancer drug, suggesting that they could be examples of a novel class of anti-cancer drug. The COMPARE analysis suggested some targets shown in Table 2, however, this needs to be verified by biological testing in the future.

In conclusion, we synthesized forty-eight tamoxifen-derivatives, RIDs, and screened them, searching for novel candidates of a new class of anti-cancer drug. We identified RID-SB1 and RID-SB8 as having potent tumor growth-inhibitory activity but having completely lost ERα binding activity. Based on these results and COMPARE analysis, it was suggested that RID-SB1 and RID-SB8 had unique action modes, different from those of current anti-cancer drugs including tamoxifen. RID-SB1 and RID-SB8 merit further investigation.

Grant Support

This work was supported by Grants-in-Aid for Scientific Research (A) from Japan Society for the Promotion of Science to TY (22240092), Grant-in-Aid for Scientific Research on Priority Areas from the Ministry of Education, Culture, Sports, Science and Technology to TY (11177101), and a Health and Labour Sciences Research Grants from the Ministry of Health, Labour and Welfare to IS (11103425).

Acknowledgment

We thank Dr. R. H. Shoemaker and the late Dr. K. D. Paull for the establishment of JFCR39 and COMPARE analysis, and also thank Ms. Y. Nishimura, Ms. Y. Ohashi, Ms. M. Okamura, Ms. M. Seki, Ms. Y. Mukai and Ms. N. Tamaki for technical assistance.

REFERENCES

• 1)  Grainger DJ, Metcalfe JC. Tamoxifen: teaching an old drug new tricks? Nat. Med., 2, 381–385 (1996).
• 2)  Beekman JM, Allan GF, Tsai SY, Tsai MJ, O’Malley BW. Transcriptional activation by the estrogen receptor requires a conformational change in the ligand binding domain. Mol. Endocrinol., 7, 1266–1274 (1993).
• 3)  Jordan VC. Molecular mechanisms of antiestrogen action in breast cancer. Breast Cancer Res. Treat., 31, 41–52 (1994).
• 4)  Catherino WH, Jordan VC. A risk-benefit assessment of tamoxifen therapy. Drug Saf., 8, 381–397 (1993).
• 5)  Fisher B, Costantino JP, Wickerham DL, Cecchini RS, Cronin WM, Robidoux A, Bevers TB, Kavanah MT, Atkins JN, Margolese RG, Runowicz CD, James JM, Ford LG, Wolmark N. Tamoxifen for the prevention of breast cancer: current status of the National Surgical Adjuvant Breast and Bowel Project P-1 study. J. Natl. Cancer Inst., 97, 1652–1662 (2005).
• 6)  McDonnell DP. The Molecular Pharmacology of SERMs. Trends Endocrinol. Metab., 10, 301–311 (1999).
• 7)  Mandlekar S, Kong AN. Mechanisms of tamoxifen-induced apoptosis. Apoptosis, 6, 469–477 (2001).
• 8)  Obrero M, Yu DV, Shapiro DJ. Estrogen receptor-dependent and estrogen receptor-independent pathways for tamoxifen and 4-hydroxytamoxifen-induced programmed cell death. J. Biol. Chem., 277, 45695–45703 (2002).
• 9)  Shiina I, Sano Y, Nakata K, Kikuchi T, Sasaki A, Ikekita M, Hasome Y. Synthesis of the new pseudo-symmetrical tamoxifen derivatives and their anti-tumor activity. Bioorg. Med. Chem. Lett., 17, 2421–2424 (2007).
• 10)  Shiina I, Sano Y, Nakata K, Kikuchi T, Sasaki A, Ikekita M, Nagahara Y, Hasome Y, Yamori T, Yamazaki K. Synthesis and pharmacological evaluation of the novel pseudo-symmetrical tamoxifen derivatives as anti-tumor agents. Biochem. Pharmacol., 75, 1014–1026 (2008).
• 11)  Yamori T. Panel of human cancer cell lines provides valuable database for drug discovery and bioinformatics. Cancer Chemother. Pharmacol., 52 (Suppl. 1), S74–S79 (2003).
• 12)  Nakatsu N, Nakamura T, Yamazaki K, Sadahiro S, Makuuchi H, Kanno J, Yamori T. Evaluation of action mechanisms of toxic chemicals using JFCR39, a panel of human cancer cell lines. Mol. Pharmacol., 72, 1171–1180 (2007).
• 13)  Dan S, Tsunoda T, Kitahara O, Yanagawa R, Zembutsu H, Katagiri T, Yamazaki K, Nakamura Y, Yamori T. An integrated database of chemosensitivity to 55 anticancer drugs and gene expression profiles of 39 human cancer cell lines. Cancer Res., 62, 1139–1147 (2002).
• 14)  Naasani I, Seimiya H, Yamori T, Tsuruo T. FJ5002: a potent telomerase inhibitor identified by exploiting the disease-oriented screening program with COMPARE analysis. Cancer Res., 59, 4004–4011 (1999).
• 15)  Yamori T, Matsunaga A, Sato S, Yamazaki K, Komi A, Ishizu K, Mita I, Edatsugi H, Matsuba Y, Takezawa K, Nakanishi O, Kohno H, Nakajima Y, Komatsu H, Andoh T, Tsuruo T. Potent antitumor activity of MS-247, a novel DNA minor groove binder, evaluated by an in vitro and in vivo human cancer cell line panel. Cancer Res., 59, 4042–4049 (1999).
• 16)  Dan S, Okamura M, Seki M, Yamazaki K, Sugita H, Okui M, Mukai Y, Nishimura H, Asaka R, Nomura K, Ishikawa Y, Yamori T. Correlating phosphatidylinositol 3-kinase inhibitor efficacy with signaling pathway status: in silico and biological evaluations. Cancer Res., 70, 4982–4994 (2010).
• 17)  Kong D, Dan S, Yamazaki K, Yamori T. Inhibition profiles of phosphatidylinositol 3-kinase inhibitors against PI3K superfamily and human cancer cell line panel JFCR39. Eur. J. Cancer, 46, 1111–1121 (2010).
• 18)  Yaguchi S, Fukui Y, Koshimizu I, Yoshimi H, Matsuno T, Gouda H, Hirono S, Yamazaki K, Yamori T. Antitumor activity of ZSTK474, a new phosphatidylinositol 3-kinase inhibitor. J. Natl. Cancer Inst., 98, 545–556 (2006).
• 19)  Ohashi Y, Iijima H, Yamaotsu N, Yamazaki K, Sato S, Okamura M, Sugimoto K, Dan S, Hirono S, Yamori T. AMF-26, a novel inhibitor of the Golgi system, targeting ADP-ribosylation factor 1 (Arf1) with potential for cancer therapy. J. Biol. Chem., 287, 3885–3897 (2012).
• 20)  Skehan P, Storeng R, Scudiero D, Monks A, McMahon J, Vistica D, Warren JT, Bokesch H, Kenney S, Boyd MR. New colorimetric cytotoxicity assay for anticancer-drug screening. J. Natl. Cancer Inst., 82, 1107–1112 (1990).
• 21)  Monks A, Scudiero D, Skehan P, Shoemaker R, Paull K, Vistica D, Hose C, Langley J, Cronise P, Vaigro-Wolff A, Gray-Goodrich M, Campbell H, Mayo J, Boyd M. Feasibility of a high-flux anticancer drug screen using a diverse panel of cultured human tumor cell lines. J. Natl. Cancer Inst., 83, 757–766 (1991).
• 22)  Paull KD, Shoemaker RH, Hodes L, Monks A, Scudiero DA, Rubinstein L, Plowman J, Boyd MR. Display and analysis of patterns of differential activity of drugs against human tumor cell lines: development of mean graph and COMPARE algorithm. J. Natl. Cancer Inst., 81, 1088–1092 (1989).
• 23)  Eglen RM. Enzyme fragment complementation: a flexible high throughput screening assay technology. Assay Drug Dev. Technol., 1, 97–104 (2002).
• 24)  Wallace S, Yang DJ, Deplessand E, Cherif A, Quadri S. High affinity tamoxifen derivatives. US6096874 (2000).
• 25)  Weatherman RV, Clegg NJ, Scanlan TS. Differential SERM activation of the estrogen receptors (ERalpha and ERbeta) at AP-1 sites. Chem. Biol., 8, 427–436 (2001).
• 26)  Rutanen EM, Heikkinen J, Halonen K, Komi J, Lammintausta R, Ylikorkala O. Effects of ospemifene, a novel SERM, on hormones, genital tract, climacteric symptoms, and quality of life in postmenopausal women: a double-blind, randomized trial. Menopause, 10, 433–439 (2003).
• 27)  Wallace OB, Richardson TI, Dodge JA. Estrogen receptor modulators: relationships of ligand structure, receptor affinity and functional activity. Curr. Top. Med. Chem., 3, 1663–1682 (2003).
• 28)  Suh N, Glasebrook AL, Palkowitz AD, Bryant HU, Burris LL, Starling JJ, Pearce HL, Williams C, Peer C, Wang Y, Sporn MB. Arzoxifene, a new selective estrogen receptor modulator for chemoprevention of experimental breast cancer. Cancer Res., 61, 8412–8415 (2001).
• 29)  Lewiecki EM. Lasofoxifene for the prevention and treatment of postmenopausal osteoporosis. Ther. Clin. Risk Manag., 5, 817–827 (2009).
• 30)  Mountfield RJ, Kiehr B, John BA. Metabolism, disposition, excretion, and pharmacokinetics of levormeloxifene, a selective estrogen receptor modulator, in the rat. Drug Metab. Dispos., 28, 503–513 (2000).
• 31)  Alexandersen P, Riis BJ, Stakkestad JA, Delmas PD, Christiansen C. Efficacy of levormeloxifene in the prevention of postmenopausal bone loss and on the lipid profile compared to low dose hormone replacement therapy. J. Clin. Endocrinol. Metab., 86, 755–760 (2001).
• 32)  Rivera-Guevara C, Camacho J. Tamoxifen and its new derivatives in cancer research. Recent Patents Anticancer. Drug Discov., 6, 237–245 (2011).