PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
J Med Chem. Author manuscript; available in PMC 2010 September 24.
Published in final edited form as:
PMCID: PMC2761948
NIHMSID: NIHMS142219

α-Tocopheryl Succinate as a Scaffold to Develop Potent Inhibitors of Breast Cancer Cell Adhesion

Abstract

This study is aimed at the pharmacological exploitation of α-tocopheryl succinate (1) to develop potent anti-adhesion agents. Considering the structural cooperativity between the phytyl chain and the carboxylic terminus in determining the anti-adhesion activity, our structural optimization led to compound 5 ([2-(4,8-dimethyl-non-1-enyl)-2,5,7,8-tetramethyl-chroman-6-yloxy]-acetic acid), which exhibited an-order-of-magnitude higher potency than 1 in blocking the adhesion of 4T1 metastatic breast cancer cells to extracellular matrix proteins (IC50, 0.6 μM versus 10 μM). Evidence indicates that the ability of compound 5 to block cell adhesion and migration was attributable to its effect on disrupting focal adhesion and actin cytoskeletal integrity by facilitating the degradation of focal adhesion kinase. Interactions between tumor cells and the ECM in the tumor microenvironment have been increasingly recognized as critical modulators of the metastatic potential of tumor cells. Consequently, the ability of compound 5 to block such interactions provides a unique pharmacological tool to shed light onto mechanisms that govern cell adhesion and tumor metastasis.

Introduction

The therapeutic potential of α-tocopheryl succinate (a.k.a., vitamin E succinate; 1) in cancer treatment and prevention has been the focus of many recent investigations.1, 2 It is noteworthy that 1 suppresses in vitro and in vivo tumor cell growth without incurring significant toxicity to normal cells. A growing body of evidence indicates that 1 mediates its antitumor effect by perturbing a multitude of signaling pathways governing cancer cell growth, apoptosis, differentiation, angiogenesis, and metastasis. This broad spectrum of action in conjunction with low toxicity underlies the translational potential of 1 in cancer treatment or prevention. Of various target mechanisms reported in the literature, the inhibitory effect of 1 on cancer cell adhesion is especially noteworthy,3 which is evident by the ability of α-tocopheryloxyacetic acid, a derivative with increased metabolic stability, to suppress breast tumor growth and to reduce lung metastasis in animal models.4

Substantial evidence indicates that adhesion is critical to the development of different aspects of the malignant phenotype of cancer cells, including survival, invasion, metastasis, and drug resistance.5 Consequently, targeting adhesion or its associated pathways represents a therapeutically relevant strategy to improve the clinical outcome of many solid and hematological malignancies. Although humanized antibodies against different adhesion molecules have entered human trials,6 there exist few small-molecule cell adhesion-targeted agents. Consequently, this study was aimed at the pharmacological exploitation of 1 to develop novel compounds with high potency in inhibiting cell adhesion. This structural optimization has led to a novel class of anti-adhesion compounds with sub-micromolar potency.

Chemistry

We rationalized that the aliphatic side chain and the semisuccinate play a crucial role in mediating 1's anti-adhesion activity. To assess the involvement of the phytyl side chain, we curtailed the chain length by the incremental removal of isopranyl units from the hydrophobic terminus, yielding 2 and 3 (Fig. 1).

Figure 1
(A) Schematic representation of the course of structural optimization of compound 1 to develop anti-adhesion agents. (B) General synthetic procedure for compound 1 derivatives.

These truncated derivatives showed substantially improved potency vis-à-vis 1 in inhibiting the adhesion of 4T1 metastatic breast cancer cells to Matrigel-coated surface, with the relative potency of 2 > 3 > 1. Moreover, increasing the rigidity of the side chain by introducing a double bond into 2, resulting in compound 4, gave rise to a multifold improvement in the anti-adhesion potency. These findings underscore the pivotal role of the phytyl side chain in mediating the anti-adhesion activity, of which two isopentyl units represented the optimal chain length.

We carried out further modifications of 2 and 4 by replacing the hemisuccinate with ether-linked C2 – C5 carboxylic acids to generate two series of compounds (I: 5 and 79; II: 6 and 1012), of which the rationale was twofold. First, like the phytyl side chain, the carboxylic function is also critically involved in ligand recognition by the target protein. Second, as the hemisuccinate is susceptible to enzymatic digestion, appendage of the carboxylic function through an ether linkage would increase the in vivo metabolic stability of the resulting derivatives. Of all these derivatives, 1 was derived from (R,R,R)-α-tocopherol, while the others were synthesized from the chiral precursor (S)-6-hydroxy-2,5,7,8-tetramethyl-chroman-2-carboxylic acid methyl ester according to a general procedure described in Fig. 1B. However, all side chains used were racemic unless otherwise mentioned.

Results

Pharmacological exploitation of 1 to develop potent small-molecule inhibitors of cell adhesion

In this study, we used the 4T1 mouse mammary tumor cell line to investigate the anti-adhesion activity of 1 and its derivatives because of the high propensity of 4T1 cells to metastasize to lung, liver, bone and other sites,7-9 a characteristic shared by stage IV human breast tumor cells. As shown, 1 exhibited a modest inhibitory effect on 4T1 cell adhesion to a Matrigel-coated surface (Fig. 2B). While 1 inhibited 50% cell adhesion at 10 μM, its activity leveled off between 10 and 50 μM. Conceivably, this weak potency in conjunction with metabolic instability prohibited the clinical use of 1 in cancer therapy.

Figure 2
Suppression of breast cancer cell adhesion by compound 1 and its derivatives to a Matrigel-coated surface. (A) Effect of compound 1 and various derivatives, each at 5 μM, relative to DMSO (D) on the adhesion of 4T1 cells (*P < 0.05; ** ...

Although the molecular target by which 1 inhibited cell adhesion remains undefined, we hypothesized that the phytyl side chain and the succinate moiety were amenable to modifications to improve the anti-adhesion potency of 1. This premise was corroborated by the significantly improved potencies (P < 0.01) of 2 and, to a lesser extent, 3, of which the alkyl side chains were shortened by one and two isopranyl units, respectively (Fig. 2A).

The subsequent lead optimization of 2 was performed via two strategies: 1) inserting a double bond α-to the chromane ring to increase the rigidity of the side chain, and 2) replacing the hemisuccinate moiety with alkoxycarboxylic functions with varying chain lengths to increase metabolic stability. These modifications led to two series of derivatives, i.e., series I: compounds 4, 5, and 79; series II: compounds 6, and 1012. Many of these derivatives showed significantly improved activities relative to VES in inhibiting T41 cell adhesion (P < 0.05) (Fig. 2A). Of them, compound 5 exhibited the highest potency, followed by compound 4, with the respective IC50 values of 0.6 μM, and 1.3 μM vis-à-vis 2.5 μM for 2 (Fig. 2B). Compound 5, as distinguished by the α,β-unsaturated, truncated side chain and the ether-linked acetic acid, was three times more potent than its saturated counterpart, compound 6 (IC50, 2 μM), underscoring the importance of the rigidity of the alkyl chain in interacting with the target protein(s). Moreover, increases in the length of the alkoxy linker of 5 precipitously reduced the anti-adhesion potency. Together, this finding indicates a subtle structure-activity relationship (SAR) in the effect of these derivatives on tumor cell adhesion.

Relative to 4T1 cells, noninvasive MCF-7 and SKBR3 breast cancer cells were also susceptible, though with less sensitivity, to the anti-adhesion effect of these derivatives (Fig. 2C). For example, the IC50 value of compound 5 for these two cell lines was approximately 2 μM, a 3.3-fold increase over that of 4T1, while 1 even at 100 μM was ineffective in suppressing the adhesion of these cells. This discrepancy might be related to differences in the metastatic potential among these cell lines.

Furthermore, we obtained evidence that the ability of these derivatives to block adhesion was not due to drug-induced cell death. Despite high potency in inhibiting cell adhesion, these optimal derivatives exhibited modest activities in suppressing 4T1 cell viability in 2% FBS, with IC50 values of greater than 10 μM (Fig. 3A, left panel). As compared to 4T1 cells, MCF-10A nonmalignant breast epithelial cells exhibited a higher degree of resistance to the antiproliferative activity of these derivatives, with IC50 of 23 μM for compounds 4 and 5 (right panel).

Figure 3
(A) Dose-dependent inhibitory effects of compounds 2 and 4 - 6 on the viability of 4T1 breast cancer cells (left pane) versus MCF-10A nonmalignant breast epithelial cells (right panel). 4T1 cells were treated with individual agents in 2% FBS-supplemented ...

Based on our finding, we rationalized that the anti-adhesion effect of 1 and its derivatives was attributable to their abilities to interfere with the cellular interaction with extracellular matrix (ECM) proteins. As Matrigel, a partially defined ECM extract from mouse sarcoma, consists of a mixture of ECM components, we further investigated the effect of compound 5 vis-à-vis compound 1 on blocking the adherence of 4T1 cells to purified ECM proteins, including collagen IV, fibronectin, laminin, and vitronectin (Fig. 3B). As shown, compound 5 at 0.5 μM and compound 1 at 5, 10, and 100 μM displayed differential activities in suppressing cell adhesion among these four ECM proteins, with the relative order of laminin > fibronectin > vitronectin, collagen IV. However, compound 5 at 1 μM or above could effectively suppress the adhesion to all of these ECM proteins (P < 0.01).

Inhibition of migration and lamellipodia formation of 4T1 metastatic breast tumor cells

Pursuant to the above finding, we investigated the ability of the optimal compounds 2 and 46 vis-à-vis compound 1 to inhibit serum-induced 4T1 cell migration by using the Boyden chamber assay (Fig. 4A). As shown, compounds 46 at 5 μM were effective in inhibiting cell migration in the order of 5 > 4 > 6, in line with that of anti-adhesion (P < 0.01). In contrast, compound 2 at 5 μM and VES even at 50 μM showed little effect.

Figure 4
Compound 1 derivatives disrupt the migration and actin cytoskeletal structure of 4T1 cells. (A) Effects of compound 1 at 50 μM versus compounds 2 and 4 - 6, each at 5 μM, on 4T1 cell migration. Cell motility was analyzed by the Transwell ...

To shed light onto the cellular basis for the inhibition of tumor cell migration by these derivatives, we investigated the effect of compound 1 (50 μM) versus compounds 5 (5 μM) and 6 (10 μM) on the actin cytoskeleton in 4T1 cells by immunocytochemistry. After 4-h exposure, compounds 5 and 6 caused rapid dissolution of stress fiber and impairment of lamellipodia formation at the leading edge of 4T1 cells, while the actin stress fiber in compound 1-treated cells remained largely intact (Fig. 4B). Quantitative analysis indicates that treatment of 4T1 cells with compounds 5 (5 μM) and 6 (10 μM) led to a reduction in the fluorescent intensity of F-actin by 84% and 70%, respectively, relative to the DMSO control (P < 0.01). Furthermore, the 3-D imagining of compound 5-treated cells showed detachment from neighboring cells or the surface of the culture dish as a result of stress fiber loss (Fig. 4C). Similar to DMSO, cells treated with 50 μM compound 1 remained attached with intact actin cytoskeleton (data not shown). Moreover, we observed that treatment of 4T1 cells with compound 5 or 6 gave rise to the accumulation of small vesicles in areas surrounding the nucleus where the endoplasmic reticulum is typically located. The disintegration of the endoplasmic reticulum membrane might be associated with the drug-mediated loss of actin stress fibers.

Compound 1 derivatives target focal adhesion kinase (FAK) degradation

Considering the important role of FAK in regulating the formation of focal adhesions and actin stress fibers,10, 11 we assessed the effect of compound 5 (2.5 μM), with compound 1 (5 μM) as a negative control, on focal adhesion sites in 4T1 cells by immunostaining with anti-FAK antibodies (Fig. 5A). As shown, while FAK staining displayed a typical punctate pattern representing the focal adhesion sites in the DMSO and compound 1-treated control cells, compound 5 treatment led to loss of focal adhesion sites, paralleling that of the aforementioned stress fiber dissolution (Fig. 4B). It is noteworthy that the disruption of actin stress fibers and focal adhesion by compound 5 is reminiscent of that induced by mannitol in neuroblastoma cells.12 As mannitol-induced cytoskeletal changes were preceded by the degradation of FAK, we investigated the effects of compound 5 relative to compounds 1, 2, 4, and 6, each at 5 μM, on FAK protein stability in 4T1 cells by Western blotting (Fig. 5B). As shown, treatment with compound 5, and to a lesser extent, 4 and 6, induced FAK degradation, resulting in two major cleavage fragments with molecular masses of approximately 90 and 80 kDa. This degradation, however, was less evident or not appreciable in cells treated with compounds 1 and 2, which paralleled the respective activities in blocking 4T1 cell adhesion.

Figure 5
Effects of compound 1 versus compounds 2 and 46 on the expression of FAK, a central component of cell adhesion. (A) 4T1 cells were treated with 2.5 μM compound 5, 5 μM compound 1 (negative control), or DMSO for 1 h, fixed, and ...

Discussion

As cell adhesion has emerged as a promising therapeutic target in light of its critical role in promoting the invasive phenotype of cancer cells,5 this study aimed at the pharmacological exploitation of compound 1 to develop potent anti-adhesion agents. SAR analysis indicates that there exists a high degree of structural cooperativity between the phytyl side chain and the alkoxycarboxylic terminus of compound 1 in determining its anti-adhesion activity. Among various derivatives examined, compound 5 represented the optimal agent with an-order-of-magnitude higher potency relative to compound 1 in blocking 4T1 cell adhesion (IC50, 0.6 μM versus 10 μM) and migration. Evidence indicates that the anti-adhesion effect of compounds 1 and 5 was attributable to their ability to block cell adherence to EMC proteins. In contrast to 4T1 cells, MCF-7 and SKBR3 cells were less sensitive to the anti-adhesion activity of compound 1 and its derivatives, suggesting a correlation with the invasiveness of these cell lines. Moreover, this high anti-adhesion potency was independent of compound 5's cytotoxicity in 4T1 cells. The dissociation of these two pharmacological activities suggests a unique mode of mechanism underlying the strong activity of compound 5 in inhibiting cell adhesion.

We obtained evidence that the ability of compound 5 to block cell adhesion and migration was attributable to its effect on disrupting the formation of focal adhesion and actin cytoskeletal structures including lamellipodia and stress fibers through the stimulation of FAK degradation. This mode of action is reminiscent of that of mannitol-induced disruption of cytoskeletal structures, however, without the concurrent induction of apoptosis.12 Thus, the mechanism by which compound 5 induces FAK degradation warrants further investigation given the importance of FAK signaling in mediating tumor angiogenesis and metastasis.13 From a mechanistic perspective, compound 5 differs from other agents that target FAK proteolysis14 or FAK kinase activity,15 and might provide a useful pharmacological probe to shed light onto the cellular regulation of FAK signaling and its role in facilitating tumor progression.

Conclusion

Cell adhesion represents an important therapeutic target not only in oncology but also in acute and chronic inflammatory diseases such as inflammatory bowel diseases16 and autoimmune inflammation.17 To date, most of the therapeutic development in targeting cell adhesion focuses on the blockade of integrin-ligand interactions by using monoclonal antibodies, antisense oligonucleotides, or small interference (si)RNAs, while very few small-molecule agents have been developed.5 Our data indicate that the high potency of compound 5 in anti-adhesion was attributable to its effect on interfering with cell adherence to ECM proteins, followed by FAK degradation. Interactions between tumor cells and the ECM in the tumor microenvironment have been increasingly recognized as critical modulators of the metastatic potential of tumor cells.18 Consequently, the ability of compound 5 to block such interactions provides a unique pharmacological tool to shed light onto mechanisms that govern cell adhesion and tumor metastasis, which is currently under investigation.

Experimental Section

Chemical reagents and organic solvents were purchased from Sigma-Aldrich (St. Louis, MO) unless otherwise mentioned. Nuclear magnetic resonance spectra (1H NMR) were measured on a Bruker DPX 300 model spectrometer. Chemical shifts (δ) were reported in parts per million (ppm) relative to the TMS peak. Electrospray ionization mass spectrometry analyses were performed with a Micromass Q-Tof II high-resolution electrospray mass spectrometer. The purity of all tested compounds are higher than 95% by elemental analyses, which were performed by Atlantic Microlab, Inc. (Norcross, GA), and were reported to be within 0.4% of calculated values. Flash column chromatography was performed using silica gel (230-400 mesh). Synthesis of compounds 1 - 3 was carried out as previously described,19 and the two series of compounds: 4, 5, 79, and 6, 1012, were synthesized according to the general scheme described in Fig. 1B, which illustrates the synthesis of compound 4 as example.

Succinic acid mono-[2-(4,8-dimethyl-non-1-enyl)-2,5,7,8-tetramethyl-chroman-6-yl] ester (4)

Step a

A solution of 6-hydroxy-2,5,7,8-tetramethyl-chroman-2-carboxylic acid methyl ester (i, 41.9 mmol), t-butyl-dimethyl-silyloxy chloride (62.8 mmol), and imidazole (172.2 mmol) in 100 mL of DMF was stirred at 85 - 95 °C for 16 h, and cooled to room temperature. The solution was diluted with 200 mL of ethyl acetate, washed, in tandem, with H2O, 1% HCl, saturated NaHCO3, and brine, dried with Na2SO4, and concentrated. Purification by flash silica gel chromatography gave the product, 6-(t-butyl-dimethyl-silyloxy)-2,5,7,8-tetramethyl-chroman-2-carboxylic acid methyl ester (ii), as white solid in 99% yield. 1H NMR (300 MHz, CDCl3) δ 0.11(s, 6H), 1.04 (s, 9H), 1.62 (s, 3H), 1.86-1.76(m, 1H), 2.01(s, 3H), 2.11(s, 3H), 2.16 (s, 3H), 2.22-2.30 (m, 1H), 3.49-3.65 (m, 2H), 3.69(s, 3H), 4.22 (s, 1H).

Step b

To a stirring solution of compound ii (15g, 39.7mmol) in 150 mL of dry hexane at -60 °C was added 40 mL of 1 M di-isobutyl aluminum hydride (DIBAL-H) in hexane dropwise over a period of 90 min. The solution was stirred in ice bath for 1 h, and 100 mL of methanol followed by 75 mL of H2O was added to the solution to quench the reaction. The mixture was extracted with 90 mL of ethyl acetate/hexane (1:2) three times, and the pooled organic phase was dried and concentrated. The residue was purified by chromatography, resulting in 6-(t-butyl-dimethyl-silyloxy)-2,5,7,8-tetramethyl-chroman-2-carbaldehyde (iii) as white solid in 88% yield. 1H NMR (300 MHz, CDCl3) δ 0.11(s, 6H), 1.04 (s, 9H), 1.39 (s, 3H), 1.76-1.86 (m, 1H), 2.01 (s, 3H), 2.11 (s, 3H), 2.16 (s, 3H), 2.22-2.30 (m, 1H), 3.49-3.65 (m, 2H), 9.62 (s, 1H).

Step c

To a suspension of 1-bromo-3.7-dimethyl octanyl phosphonium (1.05 mmol) in 20 mL of anhydrous THF at 0 °C was added 1.05 mL of 1 M lithium bis(trimethysilyl)amide in THF. The mixture was stirred at 0 °C for 30 min, and compound iii (1 mmol) in 10 mL of THF was slowly added over a period of 1 h. After being stirred at room temperature for 1 h, the solution was concentrated, diluted with 20 mL of ethyl acetate, and washed, in tandem, with H2O and brine. The residue was purified by flash silica gel chromatography to afford t-butyl-[2-(4,8-dimethyl-non-1-enyl)-2,5,7,8-tetramethyl-chroman-6-yloxy]-dimethylsilane (iv) as colorless oil in 80% yield. 1H NMR (300 MHz, CDCl3) δ 0.11 (s, 6H), 0.86-0.95 (m, 6H), 1.04 (s, 9H), 1.39 (s, 1H), 1.49-1.56 (m, 1H), 1.76-1.86 (m, 1H), 2.01-2.06 (m, 2H), 2.11 (s, 3H), 2.17 (s, 6H), 2.25-2.32 (m, 1H), 2.62 (t, J = 6.60 Hz, 2H), 5.30-5.44 (m, 2H).

Step d

A solution of compound iv (0.67 mmol) in anhydrous THF (10 mL) was cooled to 0 °C, to which was added 0.3 mL of 1 M TBAF (tetrabutyl ammonium fluoride) in THF. After being stirred at 0°C for 1 h, the solution was concentrated, diluted with ethyl acetate (20 mL), washed with water and brine, and dried. Purification of the residue by flash silica gel chromatography gave 2-(4,8-Dimethyl-non-1-enyl)-2,5,7,8-tetramethyl-chroman-6-ol (v) as colorless syrup in 96% yield. 1H NMR (300 MHz, CDCl3) δ): 0.86-0.95 (m, 9H), 1.39 (s, 1H), 0.98-1.56 (m, 11H), 1.76-1.86 (m, 1H), 2.01-232 (m, 12H), 2.62 (t, J = 6.60 Hz, 2H), 4.20 (s, 1H), 5.31-5.42 (m, 2H).

Step e

A mixture of compound v (0.58 mmol), succinic anhydrate (1.16 mmol) and pyridine (93 μL) in dry CH2Cl2 (15 mL) was stirred at refluxing temperature for 16 h. The solution was cooled to room temperature, concentrated, and purified by flash silica gel chromatography to give compound 4 as white solid in 88% yield. 1H NMR (300 MHz, CDCl3) δ 0.75-0.90 (m, 9H), 1.00-1.58 (m, 12H), 1.68-1.82 (m, 2H), 1.96 (s, 3H), 2.01 (s, 3H), 2.06-2.28 (m, 4H), 2.30-2.40 (m, 1H), 2.58 (t, J = 5.94 Hz, 2H), 2.83 (t, J = 6.42 Hz, 2H), 2.94 (t, J = 5.49 Hz, 2H), 5.31-5.38 (m, 2H); HRMS exact mass of C28H42O5 (M + Na)+, 481.2930 amu; observed mass of (M + Na)+, 481.2947 amu. Anal. Calcd. (C 73.33, H 9.23, O 17.44) Found, C 73.49, H 9.46, O 17.56.

General Procedure for Preparing Ether-Linked Alkanoic Acid Derivatives of Compound v (Step f)

A mixture of compound v (0.58mmol) and NaH (0.64 mmol) in anhydrous DMF (5 mL) was stirred at 0 °C for 30 min, to which bromoalkanoic acid methyl ester (1.16 mmol) in DMF (1 mL) was added. After being stirred at room temperature for 16 h, the reaction mixture was diluted with ethyl acetate, washed with water and brine, dried, and concentrated. The crude residue was treated with 1 N NaOH in methanol for 3 h. The solution was neutralized with 1N HCl aqueous solution, concentrated, diluted with ethyl acetate, washed with water and brine, dried, and concentrated. Purification of the residue by flash column chromatography gave compounds 5, and 79 as white solid in 70% – 78% yield.

[2-(4,8-Dimethyl-non-1-enyl)-2,5,7,8-tetramethyl-chroman-6-yloxy]-acetic acid (5)

1H NMR (300 MHz, CDCl3) δ 0.64-0.87 (m, 9H), 0.94-1.56 (m, 11H), 1.70-1.83 (m, 1H), 1.89-2.06 (m, 2H), 2.10-2.34 (m, 10H), 2.48-2.63 (m, 2H), 4.35 (s, 2H), 5.41 (d, J = 18.75 Hz, 2H). 4.35 (s, 2H), 2.63-2.48(m, 2H), 2.34-2.10(m, 10H), 2.06-1.89 (m, 2H), 1.83-1.70 (m, 1H), 1.56-0.94 (m, 11H), 0.87-0.64 (m, 9H); HRMS exact mass of C26H40O4 (M + Na)+, 439.2824 amu; observed mass of (M + Na)+, 439.2840 amu. Anal. Calcd. (C 74.96, H 9.68, O 15.36); Found, C 75.13, H 9.77, O 15.47.

3-[2-(4,8-Dimethyl-non-1-enyl)-2,5,7,8-tetramethyl-chroman-6-yloxy]-propionic acid (7)

1H NMR (300 MHz, CDCl3) δ 0.65-0.85 (m, 9H), 1.04-1.56 (m, 12H), 1.74-1.81 (m, 1H), 1.90-2.41 (m, 13H), 2.54 (t, J = 6.75 Hz), 2.80 (t, J = 6.36 Hz, 2H), 3.99 (t, J = 6.3 Hz, 2H), 5. 40 (d, J = 18.75 Hz, 2H); HRMS exact mass of C27H42O4 (M + Na)+, 453.2981 amu; observed mass of (M + Na)+, 453.3013 amu. Anal. Calcd. (C 75.31, H 9.83, O 14.86); Found, C 75.23, H 9.87, O 14.67.

4-[2-(4,8-Dimethyl-non-1-enyl)-2,5,7,8-tetramethyl-chroman-6-yloxy]-butyric acid (8)

1H NMR (300 MHz, CDCl3) δ 0.68-0.90 (m, 9H), 1.04-1.56 (m, 17H), 1.74-1.81 (m, 1H), 1.90-2.07 (m, 2H), 2.09-2.40 (m, 12H), 2.58 (t, J = 5.88 Hz, 2H), 2.69 (t, J = 7.40 Hz, 2H), 3.70 (t, J = 6.0 Hz, 2H), 5.37-5.44 (m, 2H); HRMS exact mass of C28H44O4 (M + Na)+, 467.3138 amu; observed mass of (M + Na)+, 467.3161 amu. Anal. Calcd. (C 75.63, H 9.97, O 14.39); Found, C 75.58, H 9.89, O 14.33.

5-[2-(4,8-Dimethyl-non-1-enyl)-2,5,7,8-tetramethyl-chroman-6-yloxy]-pentanoic acid (9)

1H NMR (300 MHz, CDCl3) δ 0.69-0.89 (m, 9H), 0.99-1.57 (m, 14H), 1.74-2.05 (m, 6H), 2.09 (s, 3H), 2.13 (s, 3H), 2.17 (s, 3H), 2.24-2.40 (m, 1H), 2.48 (t, J = 6.69 Hz, 2H), 2.57 (t, J = 5.91 Hz, 2H), 3.66 (t, J = 5.76 Hz, 2H), 5.36-5.43 (m, 2H); HRMS exact mass of C29H46O4 (M + Na)+, 481.3294 amu; observed mass of (M + Na)+, 481.3314 amu. Anal. Calcd. (C 75.94, H 10.11, O 13.95); Found, C 75.83, H 10.06, O 13.92.

General Procedure for Hydrogenolysis (step g)

A mixture of the α,β-unsaturated acid (compounds 5 and 79; 0.37 mmol), 10% Pd/C (20 mg) in ethyl acetate was shaken under H2 at 56 psi for 16 hrs, filtered, and concentrated. Purification of the residue by flash silica gel column gave compounds 6 and 1012 as white solid in quantitative yield.

[2-(4,8-Dimethyl-nonyl)-2,5,7,8-tetramethyl-chroman-6-yloxy]-acetic acid (6)

1H NMR (300 MHz, CDCl3) δ 0.86 (dd, J = 6.57, 6.09 Hz, 9H), 1.03-1.61 (m, 19H), 1.74-1.86 (m, 2H), 2.09 (s, 3H), 2.14 (s, 3H), 2.18 (s, 3H), 2.58 (t, J = 6.69 Hz, 2H), 4.36 (s, 2H); HRMS exact mass of C26H42O4 (M + Na)+, 441.2981 amu; observed mass of C26H42O4 (M + Na)+, 441.3000 amu. Anal. Calcd. (C 74.60, H 10.11, O 15.29); Found, C 74.46, H 10.31, O 15.40.

3-[2-(4,8-Dimethyl-nonyl)-2,5,7,8-tetramethyl-chroman-6-yloxy]-propionic acid (10)

1H NMR (300 MHz, CDCl3) δ 0.83 (dd, J = 5.22, 5.09 Hz, 9H), 0.99-1.56 (m, 18H), 1.69-1.81 (m, 2H), 2.05 (s, 3H), 2.10 (s, 3H), 2.14 (s, 3H), 2.53 (t, J = 6.75 Hz; 2H), 2.80 (t, J = 6.36 Hz, 2H), 3.92 (t, J = 6.33 Hz, 2H); HRMS exact mass of C27H44O4 (M + Na)+, 455.3138 amu; observed mass of C27H44O4 (M + Na)+, 455.3152 amu. Anal. Calcd. (C 74.96, H 10.25, O 14.79); Found C 74.83, H 10.19, O 14.67.

4-[2-(4,8-Dimethyl-nonyl)-2,5,7,8-tetramethyl-chroman-6-yloxy]-butyric acid (11)

1H NMR (300 MHz, CDCl3) δ 0.86 (dd, J = 5.01, 5.84 Hz, 9H), 1.03-1.63 (m, 18H), 1.71-1.88 (m, 2H), 2.09-2.17 (m, 11H), 2.58 (t, J = 6.49 Hz, 2H), 2.68 (t, J = 7.42Hz, 2H), 3.70 (t, J = 5.98Hz, 2H); HRMS exact mass of C28H46O4 (M + Na)+, 469.3294 amu; observed mass of C28H46O4 (M + Na)+, 469.3321 amu. Anal. Calcd. (C 75.29, H 10.38, O 14.33); Found C 75.35, H 10.30, O 14.46.

5-[2-(4,8-Dimethyl-nonyl)-2,5,7,8-tetramethyl-chroman-6-yloxy]-pentanoic acid (12)

1H NMR (300 MHz, CDCl3) δ 0.86 (dd, J = 5.13, 5.96Hz, 9H), 1.05-1.62 (m, 18H), 1.70-1.88 (m, 6H), 2.09 (s, 3H), 2.13 (s, 3H), 2.17 (s, 3H), 2.48 (t, J = 7.05 Hz, 2H), 2.57 (t, J = 6.81 Hz, 2H), 3.66 (t, J = 5.92 Hz, 2H); HRMS exact mass of C29H48O4 (M + Na)+, 483.3451 amu; observed mass of C29H48O4 (M + Na)+, 483.3472 amu. Anal. Calcd. (C 75.61, H 10.50, O 13.89); Found C 75.75, H 10.66, O 13.91.

Cells and Cell Culture

4T1 murine, and MCF-7 and SKBR3 human breast cancer cells were purchased from American Type Culture Collection (Manassas, VA), and cultured in RPMI 1640 or DEME/F12 medium supplemented with penicillin-streptomycin and 10% fetal bovine serum (Invitrogen, Carlsbad, CA). MCF-10A nonmalignant breast epithelial cells were kindly provided by Professor Robert Brueggemeier (The Ohio State University), and maintained in the same medium supplemented with 5% FBS, 100 U of penicillin, 100 g/ml streptomycin, 20 ng/ml epidermal growth factor, 10 ng/ml insulin, 100 ng/ml cholera toxin, and 500 ng/ml hydrocortisone. All cell types were cultured at 37°C in a humidified incubator containing 5% CO2.

Adhesion Assay

Ninety six-well plates were coated with 12% (v/v) Matrigel (BD Biosciences), 2 μg/ml vitronectin (Invitrogen), 10 μg/ml collagen IV (Sigma-Aldrich), 10 μg/ml fibronectin (Calbiochem) or 10 μg/ml laminin (Invitrogen) at 37 °C for 1 hour, washed twice with washing buffer (0.1% BSA-containing RPMI medium) followed by blocking with 0.5% BSA-containing RPMI medium at 37°C for 60 minutes. 4T1 cells were treated with individual derivatives at the indicated concentrations at 37°C in a CO2 incubator for 60 minutes, and 2 × 104 cells in 100 μl were seeded in each well. Cells were allowed to adhere to the Matrigel-coated surface for 30 min at 37 °C, and nonadherent cells were removed by gentle washing with the aforementioned washing buffer. Adherent cells were fixed with 10% formalin, stained with 0.5% crystal violet, and dissolved in 2% SDS. Absorbance at 570 nm was measured in an ELISA plate reader (Molecular device, Sunnyvale, CA).

Migration Assay

4T1 cells were trypsinized for 5 min, washed, and suspended in 0.2% FBS-supplemented RPMI 1640 medium. Five × 104 cells in 0.5 ml of 0.2% FBS-supplemented RPMI medium containing individual test agents at the indicated concentrations were added to the upper chamber (i.e., insert) of each Transwell system (12 mm, polycarbonate, 12-μm pore, Millipore) in a 24-well plate, and incubated at 37 °C in a CO2 incubator for 30 min. The inserts were then switched to a new well containing 10% FBS-supplemented RPMI 1640 medium for 24 h. All cells in each well were fixed with 10% formalin followed by staining with 0.5% crystal violet. To quantify migrated cells, cells attached to the bottom side of the upper chamber and in the bottom of the well were wiped with a moistened cotton swab, which was then rinsed with 80 μl DDW. The cells were then dissolved by the addition of 320 μl 100% methanol. Enumeration of non-migrated cells was done by placing the chamber into 400 μl of 80% methanol, and incubating for 30 min in an orbital shaker. Absorbance at 570 nm was measured in an ELISA plate reader. Percentage of cell migration in each well was calculated using the following equation: % of migration= 100 × [(O.D. of migrated cells) - (O.D. of background)]/([(O.D. of migrated cells) - (O.D. of background)] + [(O.D. of non-migrated cells) - (O.D. of background)]). The migration activity in each treatment group is expressed as a percentage of that in the vehicle controls, which was considered to be 100%.

F-actin Immunostaining

To assess the effect of test compounds on actin cytoskeletal structures, cells were seeded onto coverslips in six-well plates and incubated overnight, followed by exposure to individual agents at the indicated concentrations for 4 h in 2% FBS-containing RPMI 1640 medium. Cells were then fixed in 3.7% formaldehyde, permeabilized with PBS containing 0.1% Triton X-100 and 0.1% BSA for 1 h, and then incubated with Alexa Fluor 488 phallotoxin staining solution (Molecular Probes, Inc., Eugene, OR) for 30 min. Nuclear counterstaining was achieved by use of 4,6-diamidino-2-phenylindole (DAPI)-containing mounting medium. Immunocytochemically labeled cells were visualized and images captured using a Zeiss microscope (LSM510) with Argon and HeNe lasers, appropriate filters (excitation wavelengths were 488 nm and 543 nm), and a 63 × 1.4 numerical aperture water immersion lens. Differences in fluorescence intensity were calculated from comparisons of the control sample with each of the treatment samples under the same threshold using MacBiophotonic ImageJ software (National Institutes of Health) and were expressed as percentages of the fluorescent intensity of the untreated control. Statistical significance was evaluated using Student's t-test and considered significant at P < 0.05.

Cell Viability Assay

Cell viability was assessed by using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium bromide (MTT) assay in six replicates in 96-well plates. The 4T1 cells were seeded at 6000 cells per well in 10% FBS-supplemented RPMI 1640 for 24 h, followed by treatments with various compounds in 2% FBS-supplemented RPMI 1640 at the indicated concentrations. Controls received DMSO at a concentration equal to that in drug-treated cells. After the end of incubation, MTT (0.5 mg/ml) in 10% FBS-supplemented RPMI 1640 was added to each well, and cells were incubated at 37°C for 2 h. Medium was removed and the reduced MTT dye was solubilized in DMSO (200 μl/well). Absorbance was determined at 570 nm by a 96-well plate reader.

Statistical analysis

The GraphPad InStat software V3.0 was used to perform all data analysis. P values were generated using the Dunnett Test for multiple comparisons to one control. Differences were considered significant at P < 0.05.

Acknowledgments

This work is supported by National Institute of Health grant CA12250, and the Lucius A. Wing Endowed Chair Fund at The Ohio State University.

Abbreviations

FBS
fetal bovine serum
FAK
focal adhesion kinase

References

1. Neuzil J, Tomasetti M, Zhao Y, Dong LF, Birringer M, Wang XF, Low P, Wu K, Salvatore BA, Ralph SJ. Vitamin E analogs, a novel group of “mitocans,” as anticancer agents: the importance of being redox-silent. Mol Pharmacol. 2007;71:1185–1199. [PubMed]
2. Wang XF, Dong L, Zhao Y, Tomasetti M, Wu K, Neuzil J. Vitamin E analogues as anticancer agents: lessons from studies with alpha-tocopheryl succinate. Mol Nutr Food Res. 2006;50:675–685. [PubMed]
3. Crispen PL, Uzzo RG, Golovine K, Makhov P, Pollack A, Horwitz EM, Greenberg RE, Kolenko VM. Vitamin E succinate inhibits NF-kappaB and prevents the development of a metastatic phenotype in prostate cancer cells: implications for chemoprevention. Prostate. 2007;67:582–590. [PubMed]
4. Hahn T, Szabo L, Gold M, Ramanathapuram L, Hurley LH, Akporiaye ET. Dietary administration of the proapoptotic vitamin E analogue alpha-tocopheryloxyacetic acid inhibits metastatic murine breast cancer. Cancer Res. 2006;66:9374–9378. [PubMed]
5. Schmidmaier R, Baumann P. ANTI-ADHESION evolves to a promising therapeutic concept in oncology. Curr Med Chem. 2008;15:978–990. [PubMed]
6. Simmons DL. Anti-adhesion therapies. Curr Opin Pharmacol. 2005;5:398–404. [PubMed]
7. Lelekakis M, Moseley JM, Martin TJ, Hards D, Williams E, Ho P, Lowen D, Javni J, Miller FR, Slavin J, Anderson RL. A novel orthotopic model of breast cancer metastasis to bone. Clin Exp Metastasis. 1999;17:163–170. [PubMed]
8. Yoneda T, Michigami T, Yi B, Williams PJ, Niewolna M, Hiraga T. Actions of bisphosphonate on bone metastasis in animal models of breast carcinoma. Cancer. 2000;88:2979–2988. [PubMed]
9. Tao K, Fang M, Alroy J, Sahagian GG. Imagable 4T1 model for the study of late stage breast cancer. BMC Cancer. 2008;8:228. [PMC free article] [PubMed]
10. McLean GW, Carragher NO, Avizienyte E, Evans J, Brunton VG, Frame MC. The role of focal-adhesion kinase in cancer - a new therapeutic opportunity. Nat Rev Cancer. 2005;5:505–515. [PubMed]
11. Mitra SK, Hanson DA, Schlaepfer DD. Focal adhesion kinase: in command and control of cell motility. Nat Rev Mol Cell Biol. 2005;6:56–68. [PubMed]
12. Kim B, Feldman EL. Insulin-like growth factor I prevents mannitol-induced degradation of focal adhesion kinase and Akt. J Biol Chem. 2002;277:27393–27400. [PubMed]
13. Mitra SK, Schlaepfer DD. Integrin-regulated FAK-Src signaling in normal and cancer cells. Curr Opin Cell Biol. 2006;18:516–523. [PubMed]
14. Ochel HJ, Schulte TW, Nguyen P, Trepel J, Neckers L. The benzoquinone ansamycin geldanamycin stimulates proteolytic degradation of focal adhesion kinase. Mol Genet Metab. 1999;66:24–30. [PubMed]
15. Roberts WG, Ung E, Whalen P, Cooper B, Hulford C, Autry C, Richter D, Emerson E, Lin J, Kath J, Coleman K, Yao L, Martinez-Alsina L, Lorenzen M, Berliner M, Luzzio M, Patel N, Schmitt E, LaGreca S, Jani J, Wessel M, Marr E, Griffor M, Vajdos F. Antitumor activity and pharmacology of a selective focal adhesion kinase inhibitor, PF-562,271. Cancer Res. 2008;68:1935–1944. [PubMed]
16. Van Assche G, Rutgeerts P. Physiological basis for novel drug therapies used to treat the inflammatory bowel diseases. I. Immunology and therapeutic potential of antiadhesion molecule therapy in inflammatory bowel disease. Am J Physiol Gastrointest Liver Physiol. 2005;288:G169–174. [PubMed]
17. Lockwood CM, Elliott JD, Brettman L, Hale G, Rebello P, Frewin M, Ringler D, Merrill C, Waldmann H. Anti-adhesion molecule therapy as an interventional strategy for autoimmune inflammation. Clin Immunol. 1999;93:93–106. [PubMed]
18. Gupta GP, Massague J. Cancer metastasis: building a framework. Cell. 2006;127:679–695. [PubMed]
19. Shiau CW, Huang JW, Wang DS, Weng JR, Yang CC, Lin CH, Li C, Chen CS. alpha-Tocopheryl succinate induces apoptosis in prostate cancer cells in part through inhibition of Bcl-xL/Bcl-2 function. J Biol Chem. 2006;281:11819–11825. [PubMed]