Reduction in plasma or skin alpha-tocopherol concentration with long-term oral administration of beta-carotene in humans and mice.

BACKGROUND: Beta-carotene is one of the most commonly used compounds in clinical trials of chemopreventive agents in various neoplastic diseases. Animal studies, including our own, have documented that dietary beta-carotene can reduce plasma alpha-tocopherol (vitamin E) levels, but few published studies have examined the clinical or pharmacokinetic ramifications of long-term, high-dose beta-carotene regimens on other fat-soluble vitamins such as alpha-tocopherol. PURPOSE: This study was designed to determine the effects of long-term beta-carotene supplementation on plasma concentrations of alpha-tocopherol in normal human subjects and in an experimental C3H/HeN mouse model. METHODS: In a double-blind study, 45 normal subjects were randomly assigned to receive 0 (placebo), 15, 30, 45, or 60 mg of oral beta-carotene daily for approximately 9 months. Monthly plasma samples were collected. Thirty-five C3H/HeN mice were fed a basal diet with or without beta-carotene and treated topically with or without alpha-tocopherol, except for the control mice, which received UV radiation for 27 weeks from week 3 to week 30. Plasma and dorsal skin samples were taken after 40 weeks and were analyzed for alpha-tocopherol and/or beta-carotene by high-performance liquid chromatography. RESULTS: Long-term dietary beta-carotene administration resulted in statistically significant reductions in levels of alpha-tocopherol in the skin and plasma of UV-irradiated mice. In the human study, the decrease in plasma alpha-tocopherol levels was progressive and significant between 6 and 9 months of beta-carotene dosing in all dosage groups. The greatest decrease was observed during the 9th (last) month of dosing, with a decrease of 40% from baseline. All oral beta-carotene doses (15-60 mg/d), however, resulted in similar decreases in steady-state plasma levels of alpha-tocopherol and in only small differences in beta-carotene plasma levels. CONCLUSION: Long-term oral administration of beta-carotene decreased steady-state plasma concentrations of alpha-tocopherol. The lack of a significant dose-response effect between doses of beta-carotene and alpha-tocopherol plasma levels is not unexpected, given the small differences in steady-state beta-carotene plasma levels in the four beta-carotene dose groups. IMPLICATIONS: Studies are needed to determine how long-term beta-carotene dosing influences tissue distribution of dietary alpha-tocopherol. Careful surveillance for this and other potentially harmful nutrient interactions should become part of all long-term intervention studies.

Enhancement of chemical carcinogenesis in mice by systemic effects of ultraviolet irradiation.

The present study was designed to determine the systemic influence of ultraviolet (UVB) irradiation upon subsequent carcinogenesis induced by benzo(a)pyrene. The source of UV irradiation consisted of six Westinghouse FS-40 fluorescent sunlamps. Female BALB/c mice received five 30-min dorsal UVB radiation treatments per week for 13 wk. At the end of 13 wk, irradiated and unirradiated mice received ventral applications of 0.1 or 1.0 mg of benzo(a)pyrene twice weekly for 20 or 10 wk, respectively. At 18 wk after the first benzo(a)pyrene treatment, mice receiving 0-, 0.1-, or 1.0-mg benzo(a)pyrene treatments bore 0, 12, or 29 tumors per group of 18 mice, respectively. Tumor-free survival was significantly shortened in the UV-irradiated hosts as compared with unirradiated hosts, as analyzed by the Kaplan-Meier method of survival analysis. Therefore, ultraviolet irradiation induced a systemic effect which enhanced subsequent tumor induction by benzo(a)pyrene in a manner which was dependent on the dose of benzo(a)pyrene.

Effect of retinoic acid on the late-stage promotion of transformation in JB6 mouse epidermal cells in culture.

beta-All-trans-retinoic acid (RA) inhibited the anchorage-independent growth of JB6 cells induced by either mezerein or alpha-epidermal growth factor (alpha-EGF) (a purified fraction of epidermal growth factor). The inhibition was dose dependent for alpha-EGF as well as for RA. Mezerein-induced growth in soft agar was inhibited to a greater extent by RA than was alpha-EGF-induced growth in soft agar, at similar colony yields. The extent of inhibition of anchorage-dependent growth induced by RA was similar for nontransformed JB6 cells and for alpha-EGF-transformed cells, so that transformation was shown not to influence the sensitivity of cells to retinoid inhibition of anchorage-dependent growth. RA was as effective at inhibiting anchorage-independent growth when it was applied after promoter-induced transformation as when it was applied during promoter-induced transformation. Therefore, the antiproliferative effect of RA, without an additional antitransformation effect, was sufficient to account for the reduced colony yield. These results suggest that the antipromoting action of retinoids in JB6 cells may occur by limiting proliferation, the regulation of which may be coupled with the state of differentiation of cells.

Influence of the duration of topical 13-cis-retinoic acid treatment on inhibition of mouse skin tumor promotion.

The effect of the time and duration of retinoid treatment on the inhibition of Stage II tumor promotion by 12-O-tetradecanoylphorbol-13-acetate (TPA) was studied in CD-1 mice. All mice were initiated with 400 nmol of benzo(a)pyrene and received Stage I tumor promotion (3.2 nmol of TPA twice weekly for 2 wk). Animals were then randomized into groups which received 13-cis-retinoic acid during early, middle, or late Stage II promotion. 13-cis-Retinoic acid pretreatments starting on Day 1, Wk 8, or Wk 23 of Stage II promotion resulted in 47, 28, or 19% inhibition, respectively, of TPA-induced tumor formation. One-half of the mice receiving 13-cis-retinoic acid at Day 1 or Wk 8 were removed from the retinoid treatments at Wk 23, the time of cessation of TPA promotion. The inhibition of tumor formation remained constant during the 15-wk observation period after cessation of retinoid treatment, suggesting that retinoid inhibition of mouse skin tumor promotion is stable in the absence of further promotion and preceded the step of irreversible conversion of promoter dependence to promoter independence.

Effects of dietary retinyl palmitate or 13-cis-retinoic acid on the promotion of tumors in mouse skin.

The present study was designed to determine the effects of dietary 13-cis-retinoic acid and retinyl palmitate on mouse skin tumor promotion by 12-O-tetradecanoylphorbol-13-acetate (TPA). Female CD-1 mice were initiated with 150 nmol of 7,12-dimethylbenz(a)anthracene and promoted twice weekly with 8 nmol of TPA. Diets supplemented with retinyl palmitate to yield 60,000 or 200,000 IU or 700,000 for 5 wk followed by 350,000 IU per kg of diet (700,000/350,000) fed to mice during tumor promotion resulted in 9%, 37%, and 65% inhibition of the papilloma yield, respectively, at 21 wk of promotion. Although topical applications of 13-cis-retinoic acid have been almost as effective as retinoic acid in preventing the appearance of mouse skin tumors, dietary 13-cis-retinoic acid at 200,000 or 700,000 IU per kg of diet resulted in no reduction in papilloma yield but did result in a dose-dependent decrease in the tumor burden (weight of tumors per mouse). Therefore, dietary retinyl palmitate yielded a dose-dependent inhibition of the number and weight of tumors promoted by TPA, whereas dietary 13-cis-retinoic acid resulted in a decrease in weight but not in number of tumors promoted by TPA.

Inhibitory effects of perillyl alcohol on UVB-induced murine skin cancer and AP-1 transactivation.

The monoterpene perillyl alcohol (POH) has proven efficacious against the formation and progression of a variety of cancers. In this study, we tested the ability of POH to inhibit photocarcinogenesis in a nonmelanoma model of mouse skin carcinogenesis and its ability to inhibit UVB-induced activator protein 1 (AP-1) transactivation in mouse skin and human keratinocytes. POH (10 mM) was applied topically to the ears and shaved dorsal surface of groups of 35 BALB/c mice throughout the experiment, during and after UVB treatment. Topical POH significantly inhibited tumor incidence and multiplicity, average tumor size, and the average tumor burden/mouse without any apparent toxicity. POH inhibited UVB-induced AP-1 transactivation in both cultured human keratinocytes and transgenic mice that stably express a luciferase reporter driven by AP-1 elements. The results suggest that POH might be used for chemoprevention of human skin cancer, and that inhibition of AP-1 activity is functionally related to inhibition of skin carcinogenesis.

Reduction of murine cutaneous UVB-induced tumor-infiltrating T lymphocytes by dietary canthaxanthin.

The effect of dietary canthaxanthin, retinyl palmitate, or their combination on the tumor-infiltrating T-lymphocyte response (T-TIL) in de novo murine ultraviolet type B irradiation-induced tumors was investigated to elucidate potential mechanisms of action of these compounds. We found that dietary canthaxanthin greatly reduced the number of tumor-infiltrating helper/inducer, suppressor/cytotoxic, and interleukin-2 receptor-positive T lymphocytes and also observed a concomitant statistically significant increase in tumour incidence in canthaxanthin-fed animals. The addition of retinyl palmitate to the canthaxanthin diet ameliorated this negative effect on TIL and the development of skin tumors. We conclude that dietary retinyl palmitate and canthaxanthin can modulate the host T-cell immune response within a growing tumor and may affect tumorigenicity.

Systemic modulation by ultraviolet irradiation of cutaneous N-methyl-N'-nitro-N-nitrosoguanidine-induced carcinogenesis.

Ultraviolet irradiation can systemically enhance subsequent skin cancer induction by benzo[a]pyrene, methylcholanthrene, or UV radiation. The present study was designed to determine whether UVB irradiation influences host susceptibility to the alkylating agent N-methyl-N'-nitro-N-nitrosoguanidine (MNNG). Female C3H/HeJ mice were exposed dorsally to UVB radiation from banks of 6 Westinghouse FS40 sun lamps. The mice received a total UV dose of approximately 8.1 x 10(5) J m-2 over a 15-week period. After termination of UVB treatments, ventral tumors were induced by 4 applications of 30 mumol of MNNG at 8-day intervals. At 20 weeks after the first MNNG treatment, UVB-irradiated mice had 7-fold more MNNG-induced, ventral tumors than did the unirradiated control mice (P = 0.026, Wilcoxon rank sum test). Ventral application of MNNG after cessation of dorsal UVB exposure, but before UV tumor appearance, did not influence photocarcinogenesis. These results demonstrate that UV irradiation can systemically decrease host resistance to tumor induction by the methylating agent, MNNG.

Enhanced resistance to an antigenic tumor in immunosuppressed mice by dietary retinyl palmitate plus canthaxanthin.

Retinoids and certain carotenoids, e.g., beta-carotene and canthaxanthin, have been found to prevent photocarcinogenesis in mice and also to act as immunoenhancers. The hypothesis that retinoids and carotenoids inhibit photocarcinogenesis by preventing UV induction of immunosuppression predicts that mice treated with these agents before and during periods of UV radiation treatments should be as resistant as unirradiated mice to an antigenic UV-induced tumor. To test this prediction, mice were fed 120 IU of retinyl palmitate per gram of diet, and/or 1% canthaxanthin, before UV irradiation treatments began, and during the entire experiment. After 4.95 x 10(5) Jm-2, delivered over 12 weeks, resistance of mice to antigenic UV-induced tumor implants (UV20) was studied. Dietary supplementation with retinyl palmitate plus canthaxanthin, but not with either agent alone at these doses, prevented the enhanced growth of UV20 in UV irradiated mice.

Effects of dietary retinyl palmitate and selenium on tumoricidal capacity of macrophages in mice undergoing tumor promotion.

Dietary retinyl palmitate was administered for 22-30 weeks in CD-1 mice which had been initiated with 0.15 mumol of 7,12-dimethylbenz[a]anthracene (DMBA) and promoted with 8 nmol of 12-O-tetradecanoylphorbol-13-acetate (TPA) twice weekly thereafter. This treatment resulted in a dose response in both the tumoricidal capacity of a selected number of isolated peritoneal macrophages (PM) and in skin tumor prevention. At 350 I.U./g of diet, retinyl palmitate (RP) also resulted in a 3-fold increase in the number of DM. RP significantly increased the total capacity of macrophage host defenses by increasing the number and individual capacity for cytotoxicity. Selenium (Se), at 2 parts/million in the drinking water, did not enhance PM tumoricidal capacity, although it did result in 60% reduction of mouse tumor burden.

Cumulative reduction of primary skin tumor growth in UV-irradiated mice by the combination of retinyl palmitate and canthaxanthin.

The effects of dietary supplementation with retinyl palmitate, canthaxanthin, or the combination of both, on photocarcinogenesis was determined in pigmented C3H/HeN mice. The basal diet was the American Institute of Nutrition Diet 76A, to which was added 120 IU of retinyl palmitate per g diet, 1% canthaxanthin, or the combination of both. Administration of the diets began 18 weeks before the first UVB radiation (280-320 nm) treatment and continued throughout the study. The UV source was a bank of 6 Westinghouse FS40 lamps which delivered to the mice a total dose of 9.9 x 10(5) J/m2, delivered over 24 weeks. These diets significantly reduced the tumor burden per mouse induced by UV irradiation, however they did not influence tumor incidence. The combination of retinyl palmitate plus canthaxanthin was more effective than either agent alone at reducing autochthonous tumor growth, a result which has not been previously reported.

Deficiency of UV-induced excision repair in human thymocytes.

The capacity of human thymocytes and of differentiated lymphocytes circulating in peripheral blood to perform unscheduled DNA synthesis (a measure of nucleotide excision repair) after UV irradiation was measured by radioautographic analysis. Only 4% of immature T lymphocytes, but 68% of circulating lymphocytes exhibited unscheduled DNA synthesis. When UV sensitivity of peripheral blood lymphocytes and thymocytes from the same donor were compared, the thymocytes, in each case, were significantly more UV sensitive than were the circulating lymphocytes. Peripheral blood lymphocytes from subjects undergoing halothane and morphine anesthesia during surgery showed 56% less excision repair capacity than those from unanesthetized donors. The difference occurred in the number of cells capable of repair rather than in the extent of repair synthesis per cell. Ultraviolet-induced unscheduled DNA synthesis occurred in only 3% of the thymocytes removed from rats killed by cervical dislocation. Therefore, the deficiency of excision repair was observed in rat thymocytes which had not been affected by anesthesia or surgical trauma. Since the thymus contains more than 90% immature T-cells, our results indicate that immature T-cells are deficient in nucleotide excision repair whereas the majority of mature peripheral blood lymphocytes exhibit such repair.

Prevention by alpha-difluoromethylornithine of skin carcinogenesis and immunosuppression induced by ultraviolet irradiation.

Administration of alpha-difluoromethylornithine (DFMO) to mice was found to inhibit both the cutaneous carcinogenesis and the immunosuppression induced by ultraviolet B (UVB) irradiation. BALB/cAnNTacfBR mice were given 1% F2MeOrn in their drinking water throughout the experiment. After 3 weeks, mice received UVB irradiation consisting of five 30-min exposures per week to banks of six FS40 Westinghouse sunlamps. In the photocarcinogenesis study, mice received a total dose of approximately 1273 kJ m-2. Skin cancer incidence in UV-irradiated mice was 38% 28 weeks after the first UV exposure; DFMO reduced this incidence to 9% (P = 0.025, log-rank test). Although DFMO has been demonstrated to be chemopreventive of chemical carcinogenesis, this is the first report that it is effective against cancers induced by a physical carcinogen. The immunosuppression induced by UVB irradiation prevents the host from rejecting antigenic, syngeneic UV-induced tumors, which normal mice can reject. The level of immunosuppression in UV-irradiated mice treated with DFMO was measured by a passive-transfer assay. Splenocytes from UV-irradiated mice to naive mice prevented the recipients from rejecting 20/24 UV-induced tumor challenges, whereas splenocytes from UV-irradiated mice treated with DFMO did not prevent recipients from rejecting such challenges (2/24 grew). The difference between these values was significant (P less than 0.001, two-sample test for binomial proportions). Phenotypic analysis of splenocytes used in the passive transfer, using a biotin-avidin-immunoperoxidase technique, revealed that DFMO treatment prevented the reduction of Ia expression normally seen in UV-irradiated mice. Thus, administration of DFMO reduced skin carcinogenesis and immunosuppression induced by UVB irradiation.

Enhanced growth and experimental metastasis of chemically induced tumor in ultraviolet irradiated syngeneic mice.

Recent studies have shown that ultraviolet (UV) irradiation induces a systemic effect which enhances subsequent tumor induction by benzo[a]pyrene in a manner which is dependent on the dose of benzo[a]pyrene. The present study was designed to test whether UV-B irradiation renders mice susceptible to subcutaneous or intravenous injection of a regressor tumor induced by benzo[a]pyrene. The sources of UV-B irradiation were banks of 6 Westinghouse FS-40 sunlamps, situated 20 cm above the mouse cages. Female BALB/cAnNHsd received five 30-min dorsal UV-B radiation treatments per week for 12 weeks, resulting in a total dose of approx. 6.4 x 10(5) J m-2. Two to seven days after termination of UV treatments, syngeneic regressor tumor cells (BP2) induced by benzo[a]pyrene were injected subcutaneously or intravenously into irradiated mice and unirradiated controls. By 38 days post subcutaneous implantation, 24/30 and 3/30 BP2 implants were detectable in the irradiated and unirradiated mice, respectively. Ultraviolet irradiated mice were also unable to reject lung colonies resulting from intravenous administration of BP2 cells, although they were rejected by unirradiated mice. The mean number of lung colonies per mouse was 16- to 35-fold greater in UV irradiated mice than in unirradiated controls, at 14 to 17 days post injection. Thus, UV irradiation rendered mice, with no known exposure to benzo[a]pyrene, susceptible to a subcutaneous or intravenous injection of a regressor tumor induced by benzo[a]pyrene.

Inhibition of 12-O-tetradecanoylphorbol-13-acetate-induced tumor promotion in murine skin by systemic effects of ultraviolet irradiation.

Systemic effects of UVB irradiation (280-320 nm) have been shown to prevent subsequent chemical tumorigenesis induced by an initiation-promotion protocol. The present investigation was designed to determine whether initiation or promotion is prevented by UV irradiation. Groups of 25 B6D2F1/J mice received 12 weeks of intermittent dorsal UVB radiation treatments administered before, or 3 weeks after, initiation with a single application of 7,12-dimethylbenz[a]anthracene on the ventral skin. All mice were promoted ventrally with 5 micrograms 12-O-tetradecanoylphorbol-13-acetate (TPA) applied three times weekly throughout the experiment. UV irradiation consisted of five 30-min exposures per week to a bank of 6 Westinghouse FS40 sunlamps. UV irradiation applied before or after initiation resulted in a decrease of 18-16 tumors per group of 25 mice, for a reduction of 61 and 50%, respectively, at 24 weeks after the first TPA treatment. Thus, prevention of tumor development was similar whether the UV influence was present or not during initiation. This finding suggests that the UV prevention of promotion could account for UV inhibition of skin tumors induced by an initiation-promotion regimen. Consistent with this concept, pretreatment of mice with dorsal UVB radiation was found to reduce DNA synthesis after exposure to TPA by 46%, although it did not decrease tritiated benzo[a]pyrene binding to DNA, in ventral epidermis. Thus, UVB irradiation systemically reduced TPA-induced tumor promotion in murine skin.

Reduction of interferon-gamma as a critical mechanism by which ultraviolet radiation prevents tumor rejection.

Mice irradiated with UVB, unlike nonirradiated mice, are highly susceptible to syngeneic, immunogenic tumors induced by UVB irradiation or by chemicals. We postulated that UV induced susceptibility to immunogenic tumors results from a reduction in host capacity to generate an interferon (IFN)-gamma immune response to tumor antigens. Shaved BALB/c mice were exposed to 6 x 10(5) J m-2 of UVB radiation delivered intermittently over 12 weeks. The UVB-irradiated and nonirradiated mice received intradermal injections of UVM12 or BP2 tumor cells. After 0, 1.5, 3, 7 or 21 days, draining lymph nodes were excised. Lymph node cells were incubated with UVM12 or BP2 cells that had received 2.5 Gy of gamma-radiation. After 48 h in culture, supernatants were analyzed for IFN-gamma content by enzyme-linked immunosorbent assay and cellular RNA was extracted for mRNA detection by reverse transcriptase-polymerase chain reaction analysis. At 7 days after tumor injection, draining lymph node cells from nonirradiated control mice secreted significant levels of IFN-gamma and contained at least 0.0729 amol of IFN-gamma mRNA/microgram cDNA upon in vitro exposure to gamma-irradiated tumor cells. Draining lymph node cells removed from UV-irradiated mice contained only 18% as much IFN-gamma mRNA and secreted little or no IFN-gamma when exposed to gamma-irradiated tumor cells. A single injection of antibody directed against murine IFN-gamma rendered normal mice as susceptible as UV-irradiated mice to BP2 tumor cells. Thus, chronic UV irradiation leads to an inability of host tumor draining lymph node cells to mount an IFN-gamma response to tumor antigens.(ABSTRACT TRUNCATED AT 250 WORDS)

Osteotoxicity after chronic dietary administration of 13-cis-retinoic acid, retinyl palmitate or selenium in mice exposed to tumor initiation and promotion.

In view of the clinical trials of retinoids as therapeutic agents for premalignant skin lesions, a radiographic study was undertaken to measure skeletal toxicities after chronic dietary administration of retinoids in mice exposed to tumor initiation and promotion. CD-1 mice were initiated with 0.15 moles of 7,12-dimethylbenz[a]anthracene and promoted twice daily with 8 nmoles of 12-0-tetradecanoylphorbol-13-acetate for 23 weeks. Diets were supplemented with 60 IU, 200 IU, or 700 IU of retinyl palmitate (RP) per g diet. After 5 weeks, the 700 IU of RP /g diet was lowered to 350 IU/g diet. Administration of these diets to mice during the 23 weeks of tumor promotion resulted in a 0-fold, 2-fold, or 10-fold increase in bone fractures, respectively. Osteoporotic bone lesions identified on radiographs rose 0-fold, 0-fold, and 10-fold at the respective doses, whereas metaphyseal flares increased 0-fold, 1.4-fold, and 3.6-fold. Bone deformities were augmented 0-fold, 1.8-fold and 2.9-fold at the respective doses. Addition of selenium (2 ppm in the drinking water) did not alter the bone toxicity of RP. 13-cis-retinoic acid (CRA) was less toxic at 700 IU/g diet than was RP at that dose, as evidenced by the death of 12 of 70 mice by the 6th week of dietary RP and no deaths in the 35 mice fed 700 IU CRA/g diet for 23 weeks. CRA at 700 IU/g diet resulted in 3/4 as many osteoporotic bones, 1/3 as many bone fractures, 4/5 as many metaphyseal flares, and a similar number of bone deformities as mice fed 700/350 IU/g diet. At the dose of 200 IU/g food, osteotoxicities were similar in the mice fed diets supplemented with RP and CRA. Thus, the light dose of CRA (700 IU/g diet) was less toxic than the high dose of, RP but at a lower dose (200 IU/g), CRA was as osteotoxic as was RP. Bone fractures in mice exposed to prolonged dietary administration of retinoids was a more sensitive index of retinoid toxicity than was body weight. We have detected osteotoxicity in mice at a total dose of CRA which was about twice the total dose used clinically.

Inhibition of DFMO-induced audiogenic seizures by chlordiazepoxide.

Mice given 1% alpha-difluoromethylornithine (DFMO) in the drinking water for 5 weeks developed a hyperactive behavior characterized by uncontrolled running upon stimulation with noise. The running was followed by seizures and sometimes death. These behaviors are characteristic of audiogenic seizures. Strain differences in susceptibility to DFMO-induced audiogenic seizures were observed. The order of sensitivity to this DFMO effect was: C3HeB/FEJ = C3H/HeN > CBA/J = BALB/c. Chronic DFMO treatment was found to deplete whole brain putrescine and spermidine, but not spermine nor gamma-aminobutyric acid (GABA), in the 2 strains of mice analyzed, C3H/HeN and BALB/c. The audiogenic seizures were eliminated by pretreatment with the benzodiazepine, chlordiazepoxide (Librium) (40 mg/kg, ip) 105 minutes prior to testing for seizures.