Purpose: Among derivatives of α-vitamin E, α-vitamin E succinate (VES), has attracted much attention due to its potent anti–prostate cancer activity in vitro and in vivo. However, the in vivo antitumor activity of VES might be compromised if administrated orally due to the VES hydrolysis by esterases in the gastrointestinal tract.

Experimental Design: New nonhydrolyzable VES ether analogues were synthesized and their growth inhibition was screened by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide growth assay. Among them, RRR-α-tocopheryloxybutyl sulfonic acid (VEBSA) was further characterized by terminal deoxyribonucleotidyl transferase–mediated dUTP nick end labeling apoptosis assay, soft agar assay, and in vivo tumor formation.

Results: VEBSA has potent antitumor ability, albeit to a lesser extent than VES, in in vitro cultured prostate cancer LNCaP and PC3 cells. Like VES, VEBSA induced apoptosis, repressed androgen receptor protein expression, and enhanced vitamin D receptor expression, suggesting that VEBSA can go through mechanisms similar to those used by VES to inhibit the growth of prostate cancer cells in vitro. However, 6 weeks of oral consumption of VEBSA, but not of VES, reduced the tumor burden in the xenografted prostate tumors in nude mice. Furthermore, oral intake of VEBSA for 20 weeks inhibited prostate tumor growth and progression more efficiently compared with VES in the prostate cancer tumor model of TRAMP mice.

Conclusion: Oral consumption of VEBSA allows a greater anticancer activity compared with VES. Chemoprevention prefers the oral consumption of agents; the advantage of VEBSA over VES to be administrated orally will allow VEBSA to serve as an agent for both preventive and therapeutic purposes for prostate cancer.

Translational Relevance

Vitamin E succinate (VES) has attracted much attention due to its antiproliferative function in vitro. However, it only exhibits its antitumor function in vivo through i.p. injection, which is not convenient for humans, especially for chemopreventive purpose. The aim of this study was to develop new VES analogues with greater bioavailability in vivo. We found that orally administered RRR-α-tocopheryloxybutyl sulfonic acid (VEBSA) is better than VES in reducing prostate tumor burden without overt toxicity in preclinical mouse cancer models. In addition, the serum level of VEBSA is substantially higher than that of VES after mice were gavaged with the same amounts of VEBSA and VES. Those studies suggest that VEBSA could be an excellent and better agent for both chemopreventive and chemotherapeutic purposes for prostate cancer. Our study thus provides a strong basis to design protocols for prostate cancer therapy in future clinical trials.

Growing lines of evidence suggest that VES is one of the most potent α-vitamin E analogues in terms of antiproliferative activity, and the underlying mechanisms in different cancer cells have been proposed (18). Importantly, VES selectively inhibits the growth of cancer cells without affecting normal cells. This has been shown in cultured cells and animal models (3, 911). However, the efficacy of VES in mouse cancer models requires VES to be delivered by i.p. administration, not by oral gavage (12), suggesting that i.v. application is required for VES to inhibit tumor growth in humans. Therefore, VES might be inconvenient for chemopreventive or therapeutic purposes. It is hypothesized that the low efficacy of VES by oral intake might be due to low bioavailability of VES, possibly caused by the presence of esterases in the gastrointestinal tract, which hydrolyze VES to α-vitamin E and succinic acid, which do not have antiproliferative activity on cancer cells. To validate whether low bioavailability is the cause of the low efficacy of VES administrated orally, we measured and calculated the tissue concentrations of VES in mice after either oral gavage or i.p. administration. Although twice as much VES was administered by oral gavage compared with i.p. administration, the oral gavage VES resulted in lower VES amounts in serum, liver, prostate, and testis, at least by 5-fold, than those after i.p. administration (data not shown). Consistently, it has been reported that the resistance of ovarian carcinoma cp70 cells to VES-induced apoptosis was due to high levels of cellular esterase. This conclusion is based on the observation that VES did not induce apoptosis in cp70 cells unless the cells were pretreated with the esterase inhibitor bis-(p-nitrophenyl) phosphate (13). Thus, the bioavailability of VES is a major concern for its application in chemoprevention.

The aim of this study was to develop new VES analogues with greater bioavailability in vivo. We developed a number of new nonhydrolyzable ether analogues of VES with sulfonic and phosphonic moieties. After screening their effects on the growth of prostate cancer cells, we selected RRR-α-tocopheryloxybutyl sulfonic acid (VEBSA) for further characterization due to its relatively higher antiproliferative activity. VEBSA required twice the concentration relative to VES to reach a similar antiproliferative effect as VES on the growth of in vitro cultured prostate cancer cells. However, the same amount of orally administered VEBSA is better than VES in reducing prostate tumor burden in preclinical mouse cancer models. Those results suggest that VEBSA could be an excellent and better agent for both chemopreventive and chemotherapeutic purposes for prostate cancer.

Cell culture and culture conditions. The LNCaP and PC3 cells were maintained as described previously (14).

Chemicals and reagents. VES and 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) were purchased from Sigma. Phospho-Akt Ser473 antibody was from Cell Signaling Technology. Vitamin D receptor (VDR) and glyceraldehyde-3-phosphate dehydrogenase antibodies were purchased from Santa Cruz Biotechnology. Androgen receptor antibody NH27 was produced as previously reported (15).

MTT cell growth assay. MTT growth assay was done as described previously (14). IC50 was calculated by curve fitting (16).

Soft agar colony formation assay. The effects of VES and VEBSA on the survival of prostate cancer cells were determined by soft agar colony forming assay. After 3-d treatment with VES or VEBSA, prostate cancer cells were trypsinized and counted; 5,000 cells were suspended in 0.4% low melting agarose (Cambrex) and then layered on top of 1.5 mL of 0.8% agarose in six-well culture plates. Cells were incubated at 37°C in humidified incubator for 3 to 4 wk. Then the plates were stained with 0.5 mL of 0.005% crystal violet in methanol for 2 h and colonies were counted under a dissecting microscope.

Quantification of apoptosis by terminal deoxyribonucleotidyl transferase–mediated dUTP nick end labeling assay. Cell apoptosis was detected by the Fluorescent FragEL DNA fragmentation kit (EMD Bioscience), as instructed by the manufacturer. Briefly, after being treated with VES or VEBSA for 3 d, prostate cancer cells were trypsinized and counted and fixed in 4% paraformaldehyde (in PBS) for 15 min at room temperature, and then stored in 80% ethanol at 4°C overnight. Cells (1 × 106) were washed with PBS and incubated with 20 μg/mL proteinase K for 5 min at room temperature, and then incubated with Fluroescein-FragEL TdT labeling reaction mix at 37°C for 1.5 h in the dark; apoptotic cells were counted by flow cytometry (FACSCalibur, BD Company).

Western blot analysis. Western blot analysis was done as described previously (14).

In vivo tumor studies. Six-week-old male athymic nude mice (Charles River) were injected s.c. in two flanks with 1 × 106 PC3 cells resuspended in 30% Matrigel (BD). One week after cell implantation, animals were sorted randomly into three groups with six mice for each group. TRAMP mice in a pure C57BL/6 background were maintained and bred, and transgene screening was done as previously described (17). Male TRAMP mice (7-8 wk old) were randomly distributed into three groups (n = 9). VEBSA, VES, or vehicle control was then administered by oral gavage at the dosage of 200 μL of 50 mmol/L in sesame oil containing 10% DMSO. Compounds were orally administered on Monday, Wednesday, and Friday followed by a 2-d rest, for a total of 6 wk for nude mice and 22 wk for TRAMP mice. Mice were monitored weekly for body weight and tumor formation. Xenograft tumor volumes were calculated using the following formula: (length × width2)/2 (18). At necropsy, xenograft tumors were dissected out and weighed. For TRAMP mice, the genitourinary apparatus consisting of bladder, urethra, seminal vesicles, ampullary gland, and the prostate was excised and weighed. The dorsolateral prostate was excised for histologic analysis. Several nude mouse and TRAMP mouse organs, including brain, heart, lungs, liver, kidneys, spleen, gastrointestinal tract, and testis, were collected for H&E staining to assess the drug toxicity effect. All the animal procedures were reviewed and approved by the University Committee on Animal Resources at the University of Rochester (19).

Synthesis of α-vitamin E analogues. All-racemic α-tocopherol (synthetic compound) was purchased from Acros and used to synthesize the all-racemic α-tocopherol analogues (VES, VEBSA, and VEPSA). The R,R,R-α-tocopherol, used to synthesize the R,R,R-α-tocopherol analogues (RRR-VEBSA, RRR-VEPSA, RRR-VEBPA, and RRR-VEPPA), was obtained from the hydrolysis of the succinate ester of (+)-α-tocopherol succinate, which was purchased from Sigma-Aldrich.

All-racemic α-tocopherol succinate (Rac-VES) was obtained in one step via esterification of α-tocopherol with succinic anhydride in the presence of 4-(dimethylamino)pyridine (Fig. 1A, Eq. 1). The sulfonic acid analogues, VEBSA and VEPSA, were synthesized from α-tocopherol by alkylation of the alkoxide of α-tocopherol with 1,3-propane sultone and 1,4-butane sultone, respectively (Fig. 1A, Eq. 2). The phosphonic acid analogues, RRR-VEBPA and RRR-VEPPA, were synthesized from R,R,R-α-tocopherol via five-step sequences involving alkylation with 3-bromopropanol or 4-bromobutanol, respectively, followed by conversion of the terminal alcohol to an iodide. The iodide was displaced with triethylphosphite and the resulting phosphonates were subsequently hydrolyzed to produce the phosphonic acids (Fig. 1A, Eq. 3).

Fig. 1.

A, synthesis of α-vitamin E analogues. B, schematic presentation of structures of α-vitamin E, α-vitamin E ester analogue VES, and ether analogues TSE, VEBSA, VEBPA, VEPPA, and VEBPA and their relative growth effects on prostate cancer cells. After prostate cancer LNCaP cells were seeded, the cells were treated with 50 μmol/L vitamin E ether analogues for 3 d and examined by MTT growth assay; 20 μmol/L VES served as positive control.

Fig. 1.

A, synthesis of α-vitamin E analogues. B, schematic presentation of structures of α-vitamin E, α-vitamin E ester analogue VES, and ether analogues TSE, VEBSA, VEBPA, VEPPA, and VEBPA and their relative growth effects on prostate cancer cells. After prostate cancer LNCaP cells were seeded, the cells were treated with 50 μmol/L vitamin E ether analogues for 3 d and examined by MTT growth assay; 20 μmol/L VES served as positive control.

Close modal

α-Vitamin E ether analogues inhibit prostate cancer cell growth. It has been shown that VES, but not its metabolic products, α-vitamin E or succinic acid, has strong antitumor activity (2, 3). In addition, the acidic side chain is important for VES antiapoptotic effect (20, 21). Therefore, we expected that novel nonhydrolyzable ether derivatives of VES with other acidic side chains, including the sulfonic moiety (RRR-VEBSA and RRR-VEPSA) or the phosphonic moiety (VEBPA and VEPPA), would maintain VES antineoplasia activity with less degradation in vivo. First, we synthesized several vitamin E ether analogues (for the details, see Supplementary Materials and Methods). Then we screened these ether analogues by examining their growth inhibitory effect on prostate cancer cells via the MTT growth assay using 20 μmol/L VES and 50 μmol/L tocopheryloxybutyric acid (TSE) as comparisons (22). VEBSA, VEPSA, and VEBPA at 50 μmol/L have similar antiproliferative activities as VES at 20 μmol/L and TSE at 50 μmol/L in prostate cancer LNCaP cells and the repression reached ∼20% to 30% after 3 days of treatment. However, VEPPA at 50 μmol/L did not significantly inhibit LNCaP cell growth (Fig. 1B).

Racemic forms of α-vitamin E analogues exhibit similar growth inhibitory efficacy as their chiral forms. Most of the α-vitamin E supplements on the market belong to synthetic α-vitamin E. However, synthetic vitamin E is not identical to natural-source vitamin E, which occurs as a single stereoisomer, RRR-α-vitamin E. Synthetic α-vitamin E, known as all-racemic α-vitamin E (Rac-α-vitamin E), contains eight stereoisomers. It would be interesting to know whether the racemic forms of α-vitamin E analogues would have different antiproliferative activities on cancer cells compared with their chiral forms. We synthesized all-racemic VES (Rac-VES), all-racemic VEBSA (Rac-VEBSA), and all-racemic VEPSA (Rac-VEPSA) and investigated their growth effect compared with their counterpart chiral forms. As shown in Fig. 2A, both RRR-VES and Rac-VES at 20 μmol/L inhibited LNCaP cell growth by ∼25% and the growth inhibition reached ∼50% with 30 μmol/L treatment for 3 days. Similarly, there is no significant difference between the effects of RRR-VEBSA and Rac-VEBSA, or between RRR-VEPSA and Rac-VEPSA, on the growth of prostate cancer LNCaP cells (Fig. 2A). Those data indicated that racemic forms of α-vitamin E analogues have similar growth inhibitory capacity as their chiral forms in cultured cancer cells.

Fig. 2.

α-Vitamin E ether analogues inhibit prostate cancer cell growth. A, racemic forms of α-vitamin E analogues exhibit similar growth inhibitory efficacy as the chiral forms. LNCaP cells were treated with RRR-VES or Rac-VES at 20 or 30 μmol/L, RRR-VEBSA or Rac-VEBSA at 60 or 80 μmol/L, and RRR-VEPSA or Rac-VEPSA at 65 μmol/L for 3 d and examined by MTT growth assay. Columns, mean of triplicate samples; bars, SD. Student's t test was done to determine the difference of each drug at each dose and the data show no statistical differences in growth effects between racemic and chiral forms. B, IC50 concentrations of VES and VEBSA in prostate cancer cells. LNCaP and PC3 cells were treated with VES or VEBSA at a series of concentrations for 3 d. IC50 values were determined by MTT assay.

Fig. 2.

α-Vitamin E ether analogues inhibit prostate cancer cell growth. A, racemic forms of α-vitamin E analogues exhibit similar growth inhibitory efficacy as the chiral forms. LNCaP cells were treated with RRR-VES or Rac-VES at 20 or 30 μmol/L, RRR-VEBSA or Rac-VEBSA at 60 or 80 μmol/L, and RRR-VEPSA or Rac-VEPSA at 65 μmol/L for 3 d and examined by MTT growth assay. Columns, mean of triplicate samples; bars, SD. Student's t test was done to determine the difference of each drug at each dose and the data show no statistical differences in growth effects between racemic and chiral forms. B, IC50 concentrations of VES and VEBSA in prostate cancer cells. LNCaP and PC3 cells were treated with VES or VEBSA at a series of concentrations for 3 d. IC50 values were determined by MTT assay.

Close modal

VEBSA reduces the cell viability and colony-forming ability of prostate cancer cells. Due to the relatively strong antiproliferative activity, we chose RRR-VEBSA for further characterization. We calculated the IC50 value of VEBSA in 3-day growth inhibition assays of both androgen-responsive prostate cancer LNCaP cells and androgen-independent PC3 cells. As shown in Fig. 2B, VEBSA inhibited the growth of both LNCaP and PC3 cells at an IC50 value of ∼65 μmol/L, whereas the IC50 value of VES is ∼30 μmol/L for both cell lines. These data indicated that VEBSA could effectively inhibit the growth of in vitro cultured prostate cancer cells; nevertheless, VES inhibits cell growth to a greater degree than VEBSA.

The effects of VEBSA and VES on the survival of prostate cancer cells were compared by soft agar colony forming assay, which mimics the in vivo tumor growth. We used the IC50 concentrations for VES (30 μmol/L) and VEBSA (65 μmol/L) to treat prostate cancer LNCaP and PC3 cells for 3 days, and then performed soft agar assay. As shown in Fig. 3A, the surviving fraction of LNCaP cells after VES or VEBSA treatment was reduced to ∼40% and ∼60% compared with control treatment, respectively. VES and VEBSA can dramatically reduce the colonies by ∼90% and ∼95% in PC3 cells, respectively. These data indicated that VEBSA, similar to VES, exhibits inhibitory activity on the colony-forming capacity in prostate cancer cells.

Fig. 3.

VEBSA and VES reduce the colony-forming ability of and induce apoptosis in prostate cancer cells. A, VEBSA and VES reduce the colony-forming ability of prostate cancer cells. The effects of VES and VEBSA on the survival of prostate cancer LNCaP (left) and PC3 (right) cells were determined by soft agar colony forming assay. Prostate cancer LNCaP and PC3 cells were treated with 30 μmol/L VES or 65 μmol/L VEBSA for 3 d; 5,000 cells were then suspended in 0.4% low melting agarose (Cambrex) and layered on top of 1.5 mL of 0.8% agarose in six-well culture plates. Cells were incubated at 37°C in humidified incubator for 3 to 4 wk. The plates were stained with 0.5 mL of 0.005% crystal violet for 2 h and colonies were counted under a dissecting microscope. Columns, mean of at least three independent experiments; bars, SD. B, VEBSA and VES induce apoptosis in prostate cancer cells. LNCaP (left) and PC3 (right) cells were treated with VES at 30 μmol/L or VEBSA at 65 μmol/L for 3 d. Apoptotic cells were investigated by terminal deoxyribonucleotidyl transferase–mediated dUTP nick end labeling assay. Columns, mean of three independent experiments; bars, SD. *, P < 0.05; **, P < 0.01, compared with the control (one-way ANOVA, Dunnett's test).

Fig. 3.

VEBSA and VES reduce the colony-forming ability of and induce apoptosis in prostate cancer cells. A, VEBSA and VES reduce the colony-forming ability of prostate cancer cells. The effects of VES and VEBSA on the survival of prostate cancer LNCaP (left) and PC3 (right) cells were determined by soft agar colony forming assay. Prostate cancer LNCaP and PC3 cells were treated with 30 μmol/L VES or 65 μmol/L VEBSA for 3 d; 5,000 cells were then suspended in 0.4% low melting agarose (Cambrex) and layered on top of 1.5 mL of 0.8% agarose in six-well culture plates. Cells were incubated at 37°C in humidified incubator for 3 to 4 wk. The plates were stained with 0.5 mL of 0.005% crystal violet for 2 h and colonies were counted under a dissecting microscope. Columns, mean of at least three independent experiments; bars, SD. B, VEBSA and VES induce apoptosis in prostate cancer cells. LNCaP (left) and PC3 (right) cells were treated with VES at 30 μmol/L or VEBSA at 65 μmol/L for 3 d. Apoptotic cells were investigated by terminal deoxyribonucleotidyl transferase–mediated dUTP nick end labeling assay. Columns, mean of three independent experiments; bars, SD. *, P < 0.05; **, P < 0.01, compared with the control (one-way ANOVA, Dunnett's test).

Close modal

VEBSA induces apoptosis in prostate cancer cells. An effective way to control cancer cell viability is by regulating cell apoptosis. VES could induce apoptosis in many types of cancer cells, including prostate (10, 23). We were also interested in determining the VEBSA-induced apoptosis in prostate cancer cells. The prostate cancer LNCaP and PC3 cells were treated for 3 days in normal culture medium (RPMI with 10% fetal bovine serum) and terminal deoxyribonucleotidyl transferase–mediated dUTP nick end labeling assay was then done. As shown in Fig. 3B, vehicle, VES, and VEBSA treatment induced ∼0.27%, ∼1.2%, and ∼4.5% LNCaP cells to undergo apoptosis, respectively. The differences between the drug and control treatments are statistically significant. Similar results were obtained from PC3 cells. These data indicated that both VES and VEBSA significantly induced apoptosis in prostate cancer cells. VEBSA seems to have more potential proapoptotic activity compared with VES when using the individual IC50 concentration determined by cell viability in prostate cancer cells.

VEBSA regulates the expression of androgen receptor and VDR in prostate cancer cells. We reported earlier that VES inhibits prostate cancer cell growth via multiple mechanisms, including reducing the expression of androgen receptor and prostate-specific antigen and elevating VDR expression (3). Thus, we were interested in determining whether VEBSA has similar mechanisms as VES to inhibit prostate cancer cell growth. The levels of androgen receptor and VDR were examined in LNCaP and PC3 cells after treatment with VES and VEBSA using the IC50 concentrations. As shown in Fig. 4, both VES and VEBSA inhibited androgen receptor expression in LNCaP cells and enhanced VDR expression in LNCaP and PC3 cells with no influence on Akt phosphorylation level, indicating that VEBSA, similar to VES, may go through the androgen receptor/PSA and VDR pathways to control cell proliferation.

Fig. 4.

VEBSA, similar to VES, could differentially regulate the expression of androgen receptor (AR) and VDR in prostate cancer cells. Prostate cancer LNCaP and PC3 cells were treated with 30 μmol/L VES or 65 μmol/L VEBSA for 3 d. The cells were harvested and Western blot analyses done with indicated antibodies. The results represent at least two independent experiments. The quantitative data are presented by normalizing with the loading control.

Fig. 4.

VEBSA, similar to VES, could differentially regulate the expression of androgen receptor (AR) and VDR in prostate cancer cells. Prostate cancer LNCaP and PC3 cells were treated with 30 μmol/L VES or 65 μmol/L VEBSA for 3 d. The cells were harvested and Western blot analyses done with indicated antibodies. The results represent at least two independent experiments. The quantitative data are presented by normalizing with the loading control.

Close modal

Oral administration of RRR-VEBSA, but not of RRR-VES, suppresses prostate cancer tumor growth in nude mice. Next, we asked the question whether RRR-VEBSA and/or RRR-VES could reduce tumor growth in vivo. VEBSA is a new nonhydrolyzable VES ether derivative, which was designed to improve the in vivo biological effects of hydrolyzable VES. To confirm that VEBSA has a better in vivo biostability than VES via oral administration, we used the same oral dose of VEBSA and VES to treat the animal with prostate cancer.

Prostate cancer PC3 cells were injected s.c. in two flank sites of male nude mice. The mice were given VEBSA or VES by oral gavage on Monday, Wednesday, and Friday followed by a 2-day rest, for a total of 6 weeks. During the experiment, the body weights of mice were monitored weekly and did not differ significantly among the group administrated with vehicle control, VES, or VEBSA (Fig. 5A). In addition, the general appearance, skin, fur, and food consumption did not alter and histologic tissue analyses of liver, kidney, and heart remained normal (data not shown), indicating that VEBSA and VES oral gavage did not have outward toxicity in the mice. The tumor growth rate in the VEBSA treatment group is much slower than in the control group observed after the 3rd week of drug treatment (Fig. 5B). The image in Fig. 5C showed the typical tumor burden following oral administration with control, VES, and VEBSA. The tumor weights at the end of experiment were 0. 48 ± 0.13, 0.51 ± 0.22, and 0.16 ± 0.05 g from the mice gavaged with control, VES, and VEBSA, respectively. Whereas oral administration of VES showed little effect on the xenograft tumor growth, the statistical calculations suggested that the tumors in the VEBSA-treated group were significantly smaller compared with the tumors in the control group. The slightly yet insignificantly reduced body weight in mice with VEBSA treatment can be attributed to loss in tumor weight. Overall, the differences in body weights of different treatment groups are not statistically significant, indicating no observed toxicity. Taken together, oral intake of VEBSA, but not of VES, can reduce prostate cancer growth in the xenograft nude mouse cancer model.

Fig. 5.

Oral administration of VEBSA, but not of VES, inhibits the growth of PC3 tumor xenografts in athymic male mice. PC3 cells (1 × 106) mixed with 30% Matrigel were s.c. injected into each of two flanks in nude mice. One week after tumor injection, treatments were initiated. RRR-VES, RRR-VEBSA, and mock control were delivered by gavage thrice per week for a total of 6 wk. The dose for VES and VEBSA was 200 μL of 50 mmol/L in sesame oil with 10% DMSO. The control was 200 μL of sesame oil with 10% DMSO. Body weight and tumor volume were determined weekly. A, body weights of treated mice did not change significantly compared with control mice. Points, mean (n = 6); bars, SD. B, the PC3 xenograft tumors with VEBSA treatment have slower growth rate in nude mice. Tumor volume was assessed with the following formula: tumor volume = (length × width2)/2. Points, mean tumor volume; bars, SE. *, P < 0.05, compared with control at the same time point (two-way ANOVA with Bonferroni posttest). C, the PC3 xenograft with VEBSA treatment has a lower tumor weight. Six weeks after tumor implantation, mice were sacrificed and the tumors were isolated and weighed. The image represents the typical tumor formation in the nude mice. Columns, mean weight for each group; bars, SE. *, P < 0.05, compared with control (one-way ANOVA followed by Dunnett's test).

Fig. 5.

Oral administration of VEBSA, but not of VES, inhibits the growth of PC3 tumor xenografts in athymic male mice. PC3 cells (1 × 106) mixed with 30% Matrigel were s.c. injected into each of two flanks in nude mice. One week after tumor injection, treatments were initiated. RRR-VES, RRR-VEBSA, and mock control were delivered by gavage thrice per week for a total of 6 wk. The dose for VES and VEBSA was 200 μL of 50 mmol/L in sesame oil with 10% DMSO. The control was 200 μL of sesame oil with 10% DMSO. Body weight and tumor volume were determined weekly. A, body weights of treated mice did not change significantly compared with control mice. Points, mean (n = 6); bars, SD. B, the PC3 xenograft tumors with VEBSA treatment have slower growth rate in nude mice. Tumor volume was assessed with the following formula: tumor volume = (length × width2)/2. Points, mean tumor volume; bars, SE. *, P < 0.05, compared with control at the same time point (two-way ANOVA with Bonferroni posttest). C, the PC3 xenograft with VEBSA treatment has a lower tumor weight. Six weeks after tumor implantation, mice were sacrificed and the tumors were isolated and weighed. The image represents the typical tumor formation in the nude mice. Columns, mean weight for each group; bars, SE. *, P < 0.05, compared with control (one-way ANOVA followed by Dunnett's test).

Close modal

Oral administration of RRR-VEBSA has a better efficacy than RRR-VES in reducing tumor burden in TRAMP mouse prostate cancer model. To further examine if oral administration of VEBSA can also prevent prostate cancer occurrence and progression, we applied the TRAMP mouse model, which spontaneously develops tumors with stages mimicking human prostate cancer progression. TRAMP mouse contains a transgene with probasin-promoter to drive SV40 T-antigen special expression in prostate to induce spontaneous prostate cancer with an intact immune system (17). The mice were fed RRR-VES, RRR-VEBSA, and vehicle control by oral gavage for ∼22 weeks. As shown in Fig. 6A, the genitourinary tract weight at the end of experiment was 3.23 ± 1.65, 1.93 ± 0.86, and 1.44 ± 0.58 g from the mice gavaged with control, VES, and VEBSA, respectively. The statistical calculations suggested that the genitourinary tract weights in the VEBSA- and VES-treated groups were significantly less than those in the control group. Histologic examination revealed that the tumors in control mice were in the transition from the well-differentiated to the poorly differentiated stage. The VES-treated tumors were primarily at the well-differentiated stage, and VEBSA-treated tumors were in less malignant stages containing prostatic intraepithelial neoplasia and well-differentiated tumors. Our immunohistochemical staining results revealed similar levels of the SV40 T-antigen transgenes in the dorsolateral prostates of control and treatment mice, and the results clearly ruled out the possibility that these tumor burden differences were caused by an inadequate expression of SV40 T-antigen transgenes in the prostate of TRAMP mice (Fig. 6C).

Fig. 6.

VEBSA is better than VES in reducing prostate tumor progression in TRAMP mice. Male TRAMP mice (7-8 wk old) were randomly distributed into three groups (n = 9) and were orally gavaged with RRR-VES, RRR-VEBSA, or vehicle control DSMO until 30 wk of age. The dose was 200 μL of 50 mmol/L and given thrice per week. A, both VES and VEBSA reduce prostate tumor progression in TRAMP mice by oral intake. At the end of the experiment, the genitourinary tract (GU) components, including bladder, prostate, and seminal vesicles, were dissected and weighed. Columns, mean of nine mice; bars, SE. *, P < 0.05, versus control; **, P < 0.01, versus control (one-way ANOVA coupled with Newman-Keuls test). B, administrations of VES and VEBSA do not affect body weight. Points, mean body weights as a function of age (weeks); bars, SD. There are no significant weight differences between treatment groups and control group (two-way ANOVA, Bonferroni test). C, administrations of VEBSA and VES affect histologic features of prostate, but not SV40 T-antigen expression, in TRAMP mice. H&E staining, ×40; SV40 T-antigen immunohistochemical staining, ×100. D, serum level of VEBSA is higher than that of VES after oral administration of each compound for 1 mo. The VEBSA and VES were extracted from the pooled sera of at least three mice in each group and then detected by high-performance liquid chromatography (see Supplementary Methods for details).

Fig. 6.

VEBSA is better than VES in reducing prostate tumor progression in TRAMP mice. Male TRAMP mice (7-8 wk old) were randomly distributed into three groups (n = 9) and were orally gavaged with RRR-VES, RRR-VEBSA, or vehicle control DSMO until 30 wk of age. The dose was 200 μL of 50 mmol/L and given thrice per week. A, both VES and VEBSA reduce prostate tumor progression in TRAMP mice by oral intake. At the end of the experiment, the genitourinary tract (GU) components, including bladder, prostate, and seminal vesicles, were dissected and weighed. Columns, mean of nine mice; bars, SE. *, P < 0.05, versus control; **, P < 0.01, versus control (one-way ANOVA coupled with Newman-Keuls test). B, administrations of VES and VEBSA do not affect body weight. Points, mean body weights as a function of age (weeks); bars, SD. There are no significant weight differences between treatment groups and control group (two-way ANOVA, Bonferroni test). C, administrations of VEBSA and VES affect histologic features of prostate, but not SV40 T-antigen expression, in TRAMP mice. H&E staining, ×40; SV40 T-antigen immunohistochemical staining, ×100. D, serum level of VEBSA is higher than that of VES after oral administration of each compound for 1 mo. The VEBSA and VES were extracted from the pooled sera of at least three mice in each group and then detected by high-performance liquid chromatography (see Supplementary Methods for details).

Close modal

In addition, the body weights were monitored and did not show significant changes between control and treated groups (Fig. 6B). The histopathologic examination of liver, kidney, testis, and heart did not differ significantly between the experimental mice and control mice (data not shown), suggesting that oral gavage of VEBSA for ∼22 weeks did not result in toxicity in those animal models.

To further test our hypothesis that VEBSA has a better in vivo bioavailability than VES, we have developed a method to monitor the serum levels of VEBSA (the methods are shown in Supplementary data). The results in Fig. 6D showed that the serum level of VEBSA was ∼15 μmol/L, whereas that of VES was ∼2 μmol/L, after the mice orally received an individual compound for 1 month. Together, our results in Fig. 6 indicate that VEBSA does have a better bioavailability and antitumor effect compared with VES in vivo.

VES is one of the most potent α-vitamin E derivatives available on the market in term of antiproliferative activity. Several groups have developed new α-vitamin E analogues based on the structure of VES to further improve its antitumor effect, especially its proapoptotic efficacy. For example, Birringer et al. (21) modified VES functional moieties and found that α-tocopheryl maleate can induce more cells to undergo apoptosis compared with VES in in vitro cultured cancer cells. Tomic-Vaic et al. (24) developed new vitamin E amide analogues, and those amide compounds have more apoptosis induction than the ester counterparts. Shiau et al. (23) created new VES analogues with truncated side chains based on the structural model that VES can dock into the binding site of Bcl-xl. Again, these new compounds have stronger ability to induce apoptosis than VES. However, there are no reports to show that those three VES analogues have been tested in any in vivo animal prostate cancer models with an intact immune system. In general, cancer chemoprevention and therapy prefers agents that can be given orally. However, the esterase and other endogenous enzymes in the gastrointestinal tract might hydrolyze the ester and amide bonds of those three VES analogues resulting in loss of their efficacy after oral intake, thus comprising their applications for the clinical use.

The goal of this study was to develop VES ether analogues with greater bioavailability in vivo after oral intake without sacrificing its antitumor activity. We replaced the ester bond with an ether bond in the new VES analogues. The ether bond has an advantage over the ester bond because the compounds will not be cleaved by esterases in the gastrointestinal tract and become more stable in vivo. Based on previous reports, an acidic side chain seems necessary for the VES antitumor effect (21). VES is a weak acid with a low pKa (5.64; ref. 25). The pH of most tumor interstitium is typically around 6.2 to 6.5 (26), and this acidic tumor interstitium can facilitate protonation of the charged VES to enhance VES uptake into the cells (25). The concept that tumor pH controls the efficacy of weak acid has been shown in vivo for chemotherapeutics (2729). Because previously investigated VES analogues have been primarily limited to modifying carboxylic acid containing functional arms (12, 3032), we decided to investigate the ability of other acidic arms to induce antiproliferative effects in cancer cells. In other antineoplastic agents, the sulfate group and the phosphate group can make the compounds become weak acids in low pH milieu and render the compounds with greater anticancer efficacy (33, 34). It was interesting to expand our study to investigate whether α-vitamin E ether analogues with the sulfonic or phosphonic moiety have any effect on the growth of prostate cancer cells. We found that there is little apparent correlation with pKa and antiproliferative activity of the tested vitamin E analogues. If the correlation with pKa were dominant, one would expect the antiproliferative effects of phosphonic acids [the first pKa of MePO2(OH)2 is estimated at 1-3 in H2O; ref. 35], to be more similar to VES (the pKa of CH3CO2H is ∼4.8 in H2O; ref. 36), and should be better than the sulfonic acids (for example, the pKa of MeSO2OH is estimated at −2 in H2O; ref. 36). It is possible, however, that the higher acidity of the sulfonic acid analogue, VEBSA, may contribute to its greater degree of apoptosis induction compared with VES.

Most α-vitamin E supplements on the market are synthetic. However, synthetic vitamin E is not identical to natural-source vitamin E. In nature, α-vitamin E occurs as a single stereoisomer, RRR-α-tocopherol. Synthetic vitamin E, known as all-racemic α-vitamin E (Rac-α-vitamin E), contains eight stereoisomers (RRR, RRS, RSS, RSR, SSS, SRR, SSR, and SRS), which differ in whether they can be “right” or “left” (R or S) at three different places in the α-vitamin E molecule (37). In this study, we determined whether there was a distinct difference in the antiproliferative activity of the racemic analogues versus the chiral analogues in vitro. If S would not be as active as R, the racemic version of α-vitamin E analogues would be less efficient to suppress prostate cancer cell growth in vitro. We found that the chiral and racemic forms of VES and VES derivatives did not have substantial differences in suppressing prostate cancer LNCaP cell growth, suggesting that the chiral center is not important to determine antiproliferative activity in the cultured cancer cells.

However, these stereoisomers might have differential absorption and transportation efficiency through oral administration. It is also possible that some isomers might be more potent than others, or those eight isomers compete with each other to have the same antiproliferative activity as the chiral form. If eight isomers have the same activity, our data then suggest that the compounds might exhibit their activity by docking its nonpolar chroman and side chain in an achiral phospholipid membrane (much as α-vitamin E does) and disrupting membrane protein-protein interaction with its acidic functional domain (carboxylic acid or sulfonic acid) that is present in the cytoplasm of the cell. If it were docking entirely in an enzyme pocket, as Dr. Chen and coworkers proposed (23), the chirality of the side chain would very likely affect the activity of the compound. However, our results challenge this possibility. Furthermore, it remains to be tested whether these stereoisomers might have differential absorption and transportation efficiency through oral administration.

In this study, we found that VEBSA could use similar mechanisms as VES to inhibit the growth of prostate cancer cells. Those mechanisms for the VEBSA to inhibit cell viability include (a) induction of apoptosis (Fig. 3), (b) repression of androgen receptor protein expression, (c) enhancement of VDR expression (Fig. 4), and (d) modulation of cell cycle molecule expression (data not shown). This may explain why we observed that VEBSA induces more apoptosis than does VES at their IC50 concentrations. Importantly, as expected, we found that oral intake of VEBSA could reduce the prostate tumor burden both in xenograft tumors in nude mice and in the TRAMP mouse prostate cancer model. We have developed methods to monitor the biodistribution of VEBSA, and the better efficacy of VEBSA, compared with VES, to inhibit in vivo prostate tumor burden supports our hypothesis that VEBSA is a nonhydrolyzable compound with greater bioavailability than VES in vivo. Interestingly, we also found that oral intake of VES for a longer period (such as 22 weeks in TRAMP) can inhibit prostate tumors in TRAMP mice, although its efficiency is still far less than that of VEBSA. The reasonable explanation is that certain levels of VES might escape from esterase digestion in the gastrointestinal tract. The undigested VES accumulates in prostate to exhibit its antitumor function after long intake of VES (such as 22 weeks for TRAMP). However, a short time (such as 6 weeks for nude mice) does not allow in vivo accumulation of intact VES and the execution of its function. Nevertheless, our study supported our hypothesis that oral intake of VEBSA could more effectively reduce the prostate tumor burden in vivo. Further studies to investigate whether VEBSA has any effect on metastasis or angiogenesis will be done in the future.

One feature of a good antitumor agent is to exhibit antitumor activity without side effects. VEBSA can inhibit prostate tumor burden in vivo without any obvious toxicity in mice. This is supported by the observation that the body weight, general appearance, and histopathologic examination of the liver, kidney, intestine, spinal cord, prostate, testis, and jejunum did not differ significantly between the VEBSA-treated mice and vehicle control–treated mice (data not shown). The toxicity tests have been done in nude mice for 6 weeks, in C57BL/6J mice (the parental strain of TRAMP mice) for 8 weeks, and in TRAMP mice for 22 weeks. Those results from preclinical animal cancer models suggest that oral administration of VEBSA is an effective and better strategy to treat prostate cancer without observable toxicity.

Kline et al. have developed another α-vitamin E ether analogue, TEA (methyl carboxyl moiety), which was shown to inhibit the growth of breast cancer cells in vitro and reduce tumor burden and metastasis in the mouse breast tumor model (12, 30). However, it has not been tested in any in vivo prostate cancer animal model. We also examined the TEA effect on prostate cancer cells and found that TEA could potentially inhibit prostate cancer cell growth in vitro, similar to VEBSA and VES.5

5

Personal communication.

It will be interesting to investigate whether the oral intake of TEA can repress the tumor burden in a prostate tumor animal model in the future. Furthermore, it was reported that VES with a shortened side chain is associated with better activity against prostate cancer cell growth in vitro (23). We may develop a new VEBSA analogue with a shorter side chain and/or with replacement of the butyl sulfonate group with a methyl sulfonate group to test whether this new compound has more potent antitumor effects. Nevertheless, in this study, we examined a group of VES ether analogues with sulfonic moiety or phosphonic moiety and found that the oral administration of VEBSA could effectively inhibit prostate cancer growth without observed toxicity.

Taken together, a newly developed nonhydrolyzable VEBSA could serve as an alternative to VES and could potentially be applied both as a chemopreventive and a chemotherapeutic agent in prostate cancer.

No potential conflicts of interest were disclosed.

Grant support: NIH grant DK 60912, Elsa U. Pardee Foundation, Urology Research Fund from the University of Rochester, and Kapoor Fund from the University at Buffalo.

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.

Note: Supplementary data for this article are available at Clinical Cancer Research Online (http://clincancerres.aacrjournals.org/).

1
Zu K, Ip C. Synergy between selenium and vitamin E in apoptosis induction is associated with activation of distinctive initiator caspases in human prostate cancer cells.
Cancer Res
2003
;
63
:
6988
–95.
2
Neuzil J, Weber T, Schroder A, et al. Induction of cancer cell apoptosis by α-tocopheryl succinate: molecular pathways and structural requirements.
FASEB J
2001
;
15
:
403
–15.
3
Zhang Y, Ni J, Messing EM, Chang E, Yang CR, Yeh S. Vitamin E succinate inhibits the function of androgen receptor and the expression of prostate-specific antigen in prostate cancer cells.
Proc Natl Acad Sci U S A
2002
;
99
:
7408
–13.
4
Ni J, Chen M, Zhang Y, Li R, Huang J, Yeh S. Vitamin E succinate inhibits human prostate cancer cell growth via modulating cell cycle regulatory machinery.
Biochem Biophys Res Commun
2003
;
300
:
357
–63.
5
Simmons-Menchaca M, Qian M, Yu W, Sanders BG, Kline K. RRR-α-tocopheryl succinate inhibits DNA synthesis and enhances the production and secretion of biologically active transforming growth factor-β by avian retrovirus-transformed lymphoid cells.
Nutr Cancer
1995
;
24
:
171
–85.
6
Yu W, Heim K, Qian M, Simmons-Menchaca M, Sanders BG, Kline K. Evidence for role of transforming growth factor-β in RRR-α-tocopheryl succinate-induced apoptosis of human MDA-MB-435 breast cancer cells.
Nutr Cancer
1997
;
27
:
267
–78.
7
You H, Yu W, Munoz-Medellin D, Brown PH, Sanders BG, Kline K. Role of extracellular signal-regulated kinase pathway in RRR-α-tocopheryl succinate-induced differentiation of human MDA-MB-435 breast cancer cells.
Mol Carcinog
2002
;
33
:
228
–36.
8
Yu W, Liao QY, Hantash FM, Sanders BG, Kline K. Activation of extracellular signal-regulated kinase and c-Jun-NH2-terminal kinase but not p38 mitogen-activated protein kinases is required for RRR-α-tocopheryl succinate-induced apoptosis of human breast cancer cells.
Cancer Res
2001
;
61
:
6569
–76.
9
Neuzil J, Weber T, Gellert N, Weber C. Selective cancer cell killing by α-tocopheryl succinate.
Br J Cancer
2001
;
84
:
87
–9.
10
Malafa MP, Fokum FD, Andoh J, et al. Vitamin E succinate suppresses prostate tumor growth by inducing apoptosis.
Int J Cancer
2006
;
118
:
2441
–7.
11
Weber T, Lu M, Andera L, et al. Vitamin E succinate is a potent novel antineoplastic agent with high selectivity and cooperativity with tumor necrosis factor-related apoptosis-inducing ligand (Apo2 ligand) in vivo.
Clin Cancer Res
2002
;
8
:
863
–9.
12
Lawson KA, Anderson K, Simmons-Menchaca M, et al. Comparison of vitamin E derivatives α-TEA and VES in reduction of mouse mammary tumor burden and metastasis.
Exp Biol Med (Maywood)
2004
;
229
:
954
–63.
13
Anderson K, Simmons-Menchaca M, Lawson KA, Atkinson J, Sanders BG, Kline K. Differential response of human ovarian cancer cells to induction of apoptosis by vitamin E Succinate and vitamin E analogue, α-TEA.
Cancer Res
2004
;
64
:
4263
–9.
14
Ni J, Wen X, Yao J, et al. Tocopherol-associated protein suppresses prostate cancer cell growth by inhibition of the phosphoinositide 3-kinase pathway.
Cancer Res
2005
;
65
:
9807
–16.
15
Lin HK, Yeh S, Kang HY, Chang C. Akt suppresses androgen-induced apoptosis by phosphorylating and inhibiting androgen receptor.
Proc Natl Acad Sci U S A
2001
;
98
:
7200
–5.
16
Ni J, Pang ST, Yeh S. Differential retention of α-vitamin E is correlated with its transporter gene expression and growth inhibition efficacy in prostate cancer cells.
Prostate
2007
;
67
:
463
–71.
17
Greenberg NM, DeMayo F, Finegold MJ, et al. Prostate cancer in a transgenic mouse.
Proc Natl Acad Sci U S A
1995
;
92
:
3439
–43.
18
Plymate SR, Haugk KH, Sprenger CC, et al. Increased manganese superoxide dismutase (SOD-2) is part of the mechanism for prostate tumor suppression by Mac25/insulin-like growth factor binding-protein-related protein-1.
Oncogene
2003
;
22
:
1024
–34.
19
Yin Y, Ni J, Chen M, Dimaggio MA, Guo Y, Yeh S. The therapeutic and preventive effect of RRR-α-vitamin e succinate on prostate cancer via induction of insulin-like growth factor binding protein-3.
Clin Cancer Res
2007
;
13
:
2271
–80.
20
Kogure K, Hama S, Kisaki M, et al. Structural characteristic of terminal dicarboxylic moiety required for apoptogenic activity of α-tocopheryl esters.
Biochim Biophys Acta
2004
;
1672
:
93
–9.
21
Birringer M, EyTina JH, Salvatore BA, Neuzil J. Vitamin E analogues as inducers of apoptosis: structure-function relation.
Br J Cancer
2003
;
88
:
1948
–55.
22
Chang E, Ni J, Yin Y, et al. α-Vitamin E derivative, RRR-α-tocopheryloxybutyric acid inhibits the proliferation of prostate cancer cells.
Asian J Androl
2007
;
9
:
31
–9.
23
Shiau CW, Huang JW, Wang DS, et al. α-Tocopheryl succinate induces apoptosis in prostate cancer cells in part through inhibition of Bcl-xL/Bcl-2 function.
J Biol Chem
2006
;
281
:
11819
–25.
24
Tomic-Vatic A, Eytina J, Chapman J, Mahdavian E, Neuzil J, Salvatore BA. Vitamin E amides, a new class of vitamin E analogues with enhanced proapoptotic activity.
Int J Cancer
2005
;
117
:
188
–93.
25
Neuzil J, Zhao M, Ostermann G, et al. α-Tocopheryl succinate, an agent with in vivo anti-tumour activity, induces apoptosis by causing lysosomal instability.
Biochem J
2002
;
362
:
709
–15.
26
Wike-Hooley JL, Haveman J, Reinhold HS. The relevance of tumour pH to the treatment of malignant disease.
Radiother Oncol
1984
;
2
:
343
–66.
27
Kozin SV, Shkarin P, Gerweck LE. The cell transmembrane pH gradient in tumors enhances cytotoxicity of specific weak acid chemotherapeutics.
Cancer Res
2001
;
61
:
4740
–3.
28
Gerweck LE, Vijayappa S, Kozin S. Tumor pH controls the in vivo efficacy of weak acid and base chemotherapeutics.
Mol Cancer Ther
2006
;
5
:
1275
–9.
29
Wu W, Luo Y, Sun C, et al. Targeting cell-impermeable prodrug activation to tumor microenvironment eradicates multiple drug-resistant neoplasms.
Cancer Res
2006
;
66
:
970
–80.
30
Lawson KA, Anderson K, Menchaca M, et al. Novel vitamin E analogue decreases syngeneic mouse mammary tumor burden and reduces lung metastasis.
Mol Cancer Ther
2003
;
2
:
437
–44.
31
Fariss MW, Fortuna MB, Everett CK, Smith JD, Trent DF, Djuric Z. The selective antiproliferative effects of α-tocopheryl hemisuccinate and cholesteryl hemisuccinate on murine leukemia cells result from the action of the intact compounds.
Cancer Res
1994
;
54
:
3346
–51.
32
Hahn T, Szabo L, Gold M, Ramanathapuram L, Hurley LH, Akporiaye ET. Dietary administration of the proapoptotic vitamin E analogue α-tocopheryloxyacetic acid inhibits metastatic murine breast cancer.
Cancer Res
2006
;
66
:
9374
–8.
33
Lanvers-Kaminsky C, Bremer A, Dirksen U, Jurgens H, Boos J. Cytotoxicity of treosulfan and busulfan on pediatric tumor cell lines.
Anticancer Drugs
2006
;
17
:
657
–62.
34
Kawasaki Y, Suzuki J, Suzuki H. Efficacy of methylprednisolone and urokinase pulse therapy combined with or without cyclophosphamide in severe Henoch-Schoenlein nephritis: a clinical and histopathological study.
Nephrol Dial Transplant
2004
;
19
:
858
–64.
35
Freedman LD, Doak GO. The preparation and properties of phosphonic acids.
Chem Rev
1957
;
57
:
479
–523.
36
Serjeant EP, Dempsey B. Ionization constants of organic acids in solution. IUPAC Chemical Data Series No. 23. Oxford (UK): Pergamon Press; 1979.
37
Hoppe PP, Krennrich G. Bioavailability and potency of natural-source and all-racemic α-tocopherol in the human: a dispute.
Eur J Nutr
2000
;
39
:
183
–93.

Supplementary data