Search

Two dietary tenets of the free radical theory of cancer require refinement. The first was dietary reduction of vulnerable free-radical targets, e.g., polyunsaturated lipids. The second was the addition of one or more antioxidants to the diet. Further, it was reported in 1939 that high levels of dietary fat exacerbated UV-carcinogenesis. Both lines of enquiry (dietary lipid and antioxidant effects on UV-carcinogenesis) were investigated. Both dietary lipids and antioxidants modified carcinogenic expression. Increasing levels of omega-6 polyunsaturated fatty acids (PUFA) exacerbated UV-carcinogenesis. However, omega-3 PUFA dramatically inhibited carcinogenic expression. It is probable that the action of omega-6 and-3 PUFA rests with differential metabolic intermediates, both tumor promoting and immune-modulating, that each PUFA generates through lipoxygenase and cyclooxygenase pathways. Antioxidant supplementation with butylated hydroxytoluene or beta-carotene demonstrated that each exerted its own specific antioxidant mechanism(s). When introduced into the complex milieu of the cell with its own intricate and complex antioxidant defense system, detrimental effects may ensue. These results point to oversimplification of these dietary suggestions to reduce cancer risk and the necessity to refine these dietary recommendations.

Epidemiological studies suggested that above average intake of beta-carotene (betaC) might reduce cancer risks. However, clinical trials found that betaC supplementation did not reduce the occurrence of non-melanoma skin cancer and that smokers suffered a significant increase in lung cancer incidence. Further, supplementing semi-defined diets with betaC failed to provide photoprotection as reported earlier for closed-formula rations, but actually exacerbated carcinogenesis. A redox mechanism, based upon one-electron transfer rate constants, proposed interactions between tocopherol, betaC and ascorbic acid in which the carotenoid radical cation, a strongly oxidizing radical, would be repaired by ascorbic acid. If the carotenoid radical cation remained unrepaired, this strongly oxidizing species could account for the pro-carcinogenic activity of betaC. Data from nutritional studies supported an interaction of tocopherol and betaC but not with ascorbic acid. The repair of the betaC radical cation must be dependent on factors other than ascorbic acid, e.g., other carotenoids or unidentified phytochemical(s).

Previous studies have shown that a high level of dietary lipid (corn oil) exacerbates UV-carcinogenic expression in hairless mice. Furthermore, it was demonstrated that this effect occurs at the postinitiation, or promotion, stage of UV-carcinogenesis--a stage believed to be modulated immunologically. Thus, we sought to examine the influence of dietary lipid on specific immune parameters at various times within a UV-carcinogenic protocol, with the purpose of detecting potential relationships to UV carcinogenesis. Hairless mice were fed either a high- (12%, wt/wt, corn oil) or low-fat (0.75%, wt/wt, corn oil) diet for 2 weeks prior to start of the UV or experimental protocols. Animals were sensitized to dinitrochlorobenzene (DNCB) hapten and delayed-type hypersensitivity (DTH) was assessed. Delayed-type hypersensitivity was significantly suppressed (P = 0.01) in the high-fat group, even before UV irradiation. Although both groups exhibited UV-induced suppression of this response, the high-fat group was totally suppressed after 3 weeks of UV, whereas the low-fat group exhibited reactivity through week 8. The splenic T-lymphocyte (Thy 1.2+) population had declined by about 50% at the time of UV termination (11 weeks). Dietary lipid exerted no apparent influence upon this T-cell population. However, after 6 weeks of UV, I-J+ cells (a marker shown to be acquired adaptively by suppressor T lymphocytes) began to increase. By week 15 (4 weeks post-UV) I-J+ cells had increased by about 65% in the high-fat group, twice the % increase that occurred in the low-fat group. When UV-induced tumors were transplanted to recipient animals receiving various periods (0, 6, 11 weeks) of UV irradiation, no significant differences in median tumor rejection times between the two dietary groups occurred at 0 or 6 weeks. After 11 weeks of UV, the low-fat group exhibited a tumor rejection time that was comparable to that of nonirradiated animals, i.e. 21 days. However, median tumor rejection time for the high-fat group was greater than 63 days, significantly (P = 0.01) longer than that of the low-fat group. Thus, suppression of tumor rejection by high fat occurred at a time when high fat had been shown to exacerbate carcinogenic expression and when I-J+ cells had markedly increased. These data demonstrate that level of dietary lipid modulates immunoresponsiveness in UV-irradiated animals and is compatible with the thesis that immune suppression may account for the exacerbation of carcinogenic expression elicited by high dietary fat.

Pro-oxidants, reactive species and free radicals, are toxic substances that can cause oxidative damage to major constituents of biological systems. In contradistinction, antioxidants are defined as any substance that significantly prevents the pro-oxidant-initiated oxidation of a substrate. Consequently, it was suggested that it might be possible to reduce free radical damage and thus cancer risk through 3 dietary changes: (1) caloric reduction, that is, lowering the level of free radical reactions arising in the course of normal metabolism; (2) minimize dietary components that increase the level of free radical reactions (eg, polyunsaturated fats); and (3) supplement the diet with one or more free radical reaction inhibitors (antioxidants). Lipid peroxidation exemplifies the type of chain reaction initiated by free radicals in (2) and (3). Both the phenolic antioxidant butylated hydroxytoluene (BHT) and the carotenoid β-carotene can terminate such reactions and have been shown to influence ultraviolet (UV) carcinogenesis. However, there is a lack of correlation between physicochemical and patho-physiological responses in both instances. Whereas the influence on UV carcinogenesis of both antioxidants has been reported to diminish as the level of dietary fat decreases, pointing to the involvement of lipid peroxidative reactions, the mode of BHT’s action in inhibiting UV carcinogenesis appears to be related to UV dose diminution through increased spectral absorbance of the stratum corneum. β-carotene has no such effect and may actually exacerbate UV carcinogenesis under certain dietary conditions. This paradox points to the complex relationship between chemical mechanisms and biological mode of action of antioxidants. Recent clinical and experimental data suggest that antioxidant supplementation of the complex and intricately balanced natural antioxidant defense system as a cancer prevention strategy will demand extreme caution.

The role of beta-carotene as an anticancer agent has been questioned as a result of clinical trials in which the incidence of nonmelanoma skin cancer was unchanged in patients receiving a beta-carotene supplement and in beta-carotene-supplemented smokers who suffered a significant increase in lung cancer occurrence. In laboratory studies, beta-carotene-supplemented semidefined diets, in contrast to earlier studies employing commercial closed-formula diets, not only failed to provide a protective effect to ultraviolet (UV) carcinogenesis but resulted in significant exacerbation. A rationale for this distinct carcinogenic response to beta-carotene rests with the stability of the carotenoid radical cation, believed to be dependent on the presence of other antioxidants for rapid repair, and suggests that response to beta-carotene depends on the presence and interaction with other dietary factors. Here, we report that diet potentiates beta-carotene-mediated exacerbation of UV carcinogenesis. Although the dietary factor(s) responsible for this effect is unidentified, these studies underscore the potential risk of beta-carotene supplementation in free-living populations where dietary status is widely varied.

Previous studies have demonstrated that high levels of dietary fat exacerbate UV-carcinogenic expression and suppress immunoresponsiveness. The latter may account for the former response. We have explored this possibility through T-lymphocyte transfer studies. Groups of HRA.HRII-c/+/Skh hairless mice were fed isocaloric diets containing high (12%, wt./wt.) or low (0.75%) levels of corn oil and irradiated 5 days/week (1.0 J cm-2/day) for 11 weeks with filtered FS-40 sunlamps. At weeks nine and 12, enriched T-cells from high-fat donors that had received 11 weeks of UV were transferred intravenously to low-fat recipients. Median tumor times for high-fat, low-fat recipient, and low-fat groups were 15.8, 18.5, and 21.6 weeks, respectively. The significantly (P < 0.03) shortened primary tumor latent period in low-fat-fed animals resulting from transfer of relatively low levels of T-cells derived from chronically irradiated high-fat donors demonstrates that the influence of dietary fat upon UV-carcinogenic expression is, at least partially, mediated via immunologic mechanisms. Further studies suggest that fat-modulated carcinogenesis can, itself, be regulated immunologically. A soluble T-14 (mouse squamous carcinoma cell line) cell-free fraction was injected subcutaneously at axillae and inguen of animals fed the high-fat diet during the first three weeks of UV or immediately post-UV. At week four post-UV, animals were challenged with T-14 cells injected subcutaneously at both flanks. 21 days post-challenge the tumor volumes of low-fat and high-fat immunized animals were zero versus 593 mm3 for the high-fat group (P < 0.007). Such treatment significantly (P < 0.03) increases the latent period of UV-induced primary tumors as well, when compared to non-treated high-fat-fed animals.

Studies employing time-resolved techniques have shown that beta-carotene, astaxanthin, and lycopene behave quite distinctly with respect to radical quenching and stability, lycopene being the least stable. These results are compatible with the relative effects of the various carotenoids on ultraviolet (UV)-mediated carcinogenesis in mice in which a statistically significant exacerbation by beta-carotene and astaxanthin, but not by lycopene, was observed. Interactions between these carotenoids and vitamin C and E radicals not only provide a chemical basis to explain the failure of beta-carotene to provide benefit in recent clinical trials but suggest that future carotenoid supplementation studies should proceed with caution until carotenoid interactions and radical repair mechanism(s) are elucidated.

Photocarcinogenesis represents the sum of a complex of simultaneous and sequential biochemical events that ultimately lead to the occurrence of skin cancer. These events, initiated by UV radiation of appropriate wavelength, include the formation of DNA photoproducts: DNA repair; mutation of proto-oncogenes and tumor suppressor genes; UV-production of radical species with subsequent effects on mutation and extra-nuclear function; and other epigenetic events that influence the course of carcinogenesis. The epigenetic influences may include immunological responses, antioxidant defenses, and dietary factors. This review represents an effort to provide current research results in the aforementioned areas and an attempt to meld these events into a comprehensive overview of photocarcinogenesis. If effective prevention and intervention strategies for skin cancer are to developed, a more thorough understanding of the disease process is imperative.

Dietary sources of lipids containing predominantly n-3 or n-6 fatty acids (FA) have been examined for effect upon several potential pathophysiologic parameters. Epidermal, plasma, and red blood cell (RBC) membrane FA composition exhibited marked differences between animals fed the respective dietary lipid sources. Reduced levels of 18:1, 20:3 and 20:4 occurred in the n-3 FA fed animals which exhibited significantly higher levels of 20:5 and 22:6. Approximately equal levels of 18:2 were present in animals fed either diet. Despite marked differences in RBC membrane FA composition, only marginal effect upon osmotic fragility occurred. Lower levels of 20:3 and 20:4 found in n-3 fed animals could result from a deficit of elongase and/or delta 5-desaturase activity. Whether lower 20:4 levels in n-3 fed animals could rate-limit eicosanoid metabolism is unknown, but epidermal capacity to metabolise arachidonic acid in these animals was found to be closely related to n-6 FA intake. Animals fed n-3 FA exhibited markedly lower levels of plasma PGE2, even when the diet was supplemented with n-6 FA. In addition, UV-radiated animals receiving the n-3 FA source demonstrated a reduced (approximately 30%) response to inflammatory stimulus and a greater (4.5-fold) delayed hypersensitivity (DH) to dinitrochlorobenzene than animals fed the n-6 FA source. These data demonstrate that dietary lipid strongly influences tissue FA composition, eicosanoid metabolism, and, in the case of DH, at least one type of T-cell mediated immune response in UV-irradiated animals.

Several types of glass and plastic materials were shown to exhibit intense photoluminescence when irradiated with UV. Water or phosphate buffered saline (PBS) contained within vessels of the respective materials when irradiated, also demonstrated relatively long-lived luminescence. A significant percentage (30%) of cultured mammalian cells were killed when exposed to UV-irradiated glass beads. The nature of the luminescence of water or PBS, or whether this or the photoluminescence of glass is directly responsible for cell toxicity, is unknown. However, we call attention to this phenomenon as a potential complicating factor in photobiological studies.

In toto, there is strong circumstantial evidence from both experimental and clinical studies to support a role for omega-3 FA in the prevention of non-melanoma skin cancer (NMSC). In experimental animal studies there is direct evidence that dietary omega-3 FA inhibits ultraviolet radiation (UVR) carcinogenic expression, with regard to both increased tumor latent period and reduced tumor multiplicity. Equivalent levels of omega-6 FA increase UVR carcinogenic expression. Dietary omega-3 FA dramatically reduces the plasma and cutaneous pro-inflammatory and immunosuppressive PGE(2) levels in mice. Dietary omega-6 FA increases prostaglandin E synthase type 2 (PGE(2)) level. Dietary omega-3 FA significantly reduces the inflammatory response and sustains, or enhances, the delayed type hypersensitivity immune response in mice when compared to an equivalent dietary level of omega-6 FA. Supplementary omega-3 FA significantly increases the UVR-mediated erythema threshold in humans. Supplementary omega-3 FA significantly reduces the level of pro-inflammatory and immunosuppressive PGE(2) levels in Ultraviolet B-irradiated human skin.

Nonmelanoma skin cancer is the most prevalent malignant disease among light-skinned individuals in the USA, accounting for approximately 1 million new cases annually. Solar ultraviolet radiation is known to be the major cause of skin cancer. The idea that diet, particularly dietary fat, may play a role in modulating cancer incidence has been based largely upon indirect epidemiologic evidence. However, few studies have found correlations of dietary fat intake with skin cancer incidence. Nevertheless, a large body of evidence with experimental animals and one clinical intervention trial, the latter avoiding the pitfalls of many epidemiologic investigations, clearly demonstrate that the level of dietary fat intake can have a significant influence on the occurrence of skin cancer in individuals at high risk.

b-Carotene is a strong singlet oxygen quencher and, under most conditions, exhibits strong antioxidant properties. Based on these properties, and a number of epidemiological studies, it was suggested that an above average intake of the carotenoid might reduce cancer risks. Earlier studies had found that b-carotene, when added to commercial closed-formula rodent diets, provided significant photoprotection to ultraviolet light (UV) carcinogenesis. However, clinical trials found that b-carotene supplementation evoked no change in incidence of nonmelanoma skin cancer and that smokers suffered a significant increase in lung cancer incidence. Further, recent studies, employing b-carotene-supplemented semidefined diets, not only failed to find a photoprotective effect, but significant exacerbation of UV carcinogenic expression resulted. Based on the relative electron transfer rate constants for interactions between b-carotene, a-tocopherol (vitamin E), and vitamin C, a mechanism was proposed for the repair of b-carotene radical cation, a strongly oxidizing radical resulting from b-carotene interactions with many oxidizing species. It was theorized that vitamin C repaired the carotenoid radical cation. As mice have no nutritional requirement for vitamin C and smokers are known to exhibit low levels of the vitamin, it was suggested that differences in the relative levels of vitamin C in closed-formula rations (no vitamin C) in which photoprotection occurred, and semidefined diets (containing vitamin C) in which exacerbation resulted, might account for the differences in response. Hairless mice were fed b-carotene-supplemented semidefined diets containing varying levels of vitamins E and C (either increasing their concentrations or reducing them to reflect levels found in closed-formula rations) and subjected to a UV carcinogenesis protocol. Increasing levels of vitamins E and C did not ameliorate b-carotene exacerbation of UV carcinogenesis. Nor did elimination of vitamin C from the diet. Reduced levels of dietary vitamin E augmented b-carotene exacerbation of UV carcinogenic expression, suggesting vitamin E and b-carotene interaction. It is concluded that the photoprotective effect of b-carotene reported earlier by others, or the noninjurious effect of b-carotene found in our studies with closed-formula rations, might depend on interaction with other dietary factors that are either absent, or present in ineffectual concentrations, in the semidefined diet in which exacerbation of UV carcinogenesis occurs. Those factors could be other carotenoids, their isomers, or some yet unidentified phytochemical(s).

Solar ultraviolet radiation (UVR) is recognized as a major cause of non-melanoma skin cancer in man. Skin cancer occurs most frequently in the most heavily exposed areas and correlates with degree of outdoor exposure. The incidence of skin cancer is also increased by contact with photosensitizing drugs and chemicals such as psoralens, coal tars and petroleum stocks. Other substances which do not act as photosensitizers, such as immunosuppressants taken by organ transplant recipients, also increase the risk of skin cancer. The U.S. Food and Drug Administration requests, on a case-by-case basis, that risk of enhanced photocarcinogenesis is assessed for many classes of drugs. Health Canada's Therapeutic Products Programme has issued a Notice of Intent to regulate pharmaceutical products which may enhance carcinogenicity of the skin induced by ultraviolet radiation. Other national regulatory agencies review such data when they exist, but their own requirements emphasize batteries of short-term in vitro and in vivo tests. While they may support drug development strategies, short-term tests have yet to be validated as predictors of the ability of drugs or chemicals to enhance photocarcinogenesis. Published protocols now describe study designs and procedures capable of determining whether test agents enhance the rate of formation of UVR-induced skin tumors.

An hypothesis for the role of free radicals in cancer was elaborated by D. Harman in 1962 who suggested that it might be possible to reduce the extent of damage caused by free radicals through three dietary changes: (i) caloric reduction, i.e., lowering the level of free radical reactions arising in the course of normal metabolism; (ii) minimize dietary components that tend to increase the level of free radical reactions (e.g., polyunsaturated fats); and (iii) supplement the diet with one or more free radical reaction inhibitors (anti-oxidants). With respect to (ii) and (iii), lipid peroxidation exemplifies the type of chain reaction initiated by free radicals, with unsaturated fatty acids being the primary center of free radical attack. Anti-oxidants act as free radical scavengers and are able to terminate these reactions. Indeed, the phenolic anti-oxidant butylated hydroxytoluene (BHT), and the carotenoid beta-carotene, have both been shown to influence photocarcinogenesis, although the lack of correlation between physicochemical parameters and pathophysiological responses is apparent in both instances. The bimolecular rate constant for reaction of BHT with model peroxyl radicals is low while beta-carotene is highly reactive. However, both are able to efficiently inhibit lipid peroxidation reactions in biological membranes. Indeed, the influence of photocarcinogenesis by both BHT and beta-carotene is diminished as the level of dietary fat decreases, pointing to the involvement of lipid peroxidative reactions. Nevertheless, the mode of action of BHT in inhibiting photocarcinogenesis appears to be related to dose-diminution resulting from an increased spectral absorbance of the stratum corneum. On the other hand, beta-carotene has no such effect and may actually exacerbate photocarcinogenesis under certain dietary conditions. This paradox points to the complex relationship between chemical mechanisms and biological mode of action of anti-oxidants in photocarcinogenesis. Recent clinical and experimental data also suggest that supplementation of the complex and intricately balanced natural antioxidant defense system with one or more anti-oxidants as a cancer prevention strategy will demand extreme caution.

The first indication that high dietary fat intake could influence the development of ultraviolet (UV) radiation-induced skin cancer in experimental animals was reported in 1939. In the 1980s a series of animal studies showed that a high level of dietary fat intake markedly shortened the time between UV exposure and tumor appearance and increased the number of tumors that developed. Further, high levels of dietary fat affected skin cancer development at the promotional stage of UV-carcinogenesis, i.e., after the cancer causing dose of UV had been delivered. Perhaps more important, switching from a high-fat to a low-fat diet immediately after delivery of the UV-initiating dose negated the exacerbating effect of high fat intake. The latter finding suggested that dietary modification, even after a cancer-causing exposure to UV, might represent a potentially important intervention strategy in the prevention of non-melanoma skin cancer (NMSC) and provided the rationale for undertaking a dietary intervention trial. One hundred and fifteen skin cancer patients completed the 2-year clinical trial on the effect of a low-fat diet on occurrence of actinic keratosis (AK) and NMSC. Patients were randomly assigned to either continue their usual diet (control group, NI) or to adopt a diet with 20% of total caloric intake as fat (diet intervention group, DI). All patients were examined at 4-month intervals for new AK and NMSC. At baseline, the mean percent of caloric intake as fat was 40+/-4% in the NI group and 39+/-3% in the DI group. After 4 months of dietary therapy, the percent calories as fat had decreased to 21+/-7% in the DI group. The percent of calories as fat in the NI group did not drop below 37% during the study period. The cumulative number of new AK per patient from months 4 through 24 was 11.6+/-17 in the NI group and 3.2+/-6 in the DI group (P0.001). Numbers of new NMSC were analyzed in 8-month periods. There were no significant changes in NMSC occurrence in the NI group. However, NMSC occurrence in the DI group declined significantly (P0.02) in the last 8-month period. Patients in the DI group also had significantly (P0.01) fewer NMSC in the last 8-month period than did patients in the NI group (0.02 versus 0.26). Practical dietary advice, with respect to reduction of percent of calories as fat, could make an important contribution to the prevention and management of AK and NMSC.

A combination of psoralen and ultraviolet A radiation (PUVA) is widely used in the treatment of psoriasis. However, PUVA treatment increases the risk of developing skin cancer in psoriasis patients and induces skin cancer in mice. Since the DNA damage induced by PUVA is quite different from that induced by UV, we investigated whether PUVA-induced mouse skin cancers display carcinogen-specific mutations in the p53 tumor suppressor gene. The results indicated that 10 of 13 (77%) PUVA-induced skin tumors contained missense mutations predominantly at exons 6 and 7. In contrast, tumor-adjacent, PUVA-exposed skin from tumor-bearing animals did not exhibit p53 mutation in exons 4-8. Interestingly, about 40% of all mutations in PUVA-induced skin tumors occurred at 5'-TA sites, and an equal number of mutations occurred at one base flanking 5'TA or 5'-TAT sites. Since PUVA induces DNA cross-links exclusively at these sites and since UV "signature" mutations were rarely detected in PUVA-induced skin cancers, we can conclude that PUVA acts as a carcinogen by inducing unique PUVA signature mutations in p53. This finding may have implications for identifying the etiology of skin cancer in psoriasis patients who have undergone PUVA therapy.

Epidemiological studies suggest that the incidence of cancer increases with age in both human and animal populations and that declining physiologic condition associated with aging might be responsible. Experimentally, the reverse has been most often observed, that is, older animals appear less susceptible to the induction of UV-carcinogenesis. Thus, we examined several biochemical parameters of epidermal macromolecular synthesis in hairless mice in an effort to gain insight into the role these processes play in physiological aging and their relationship to carcinogenesis. SKh-Hr-1 hairless mice were randomized into two groups (UV-irradiated and non-irradiated controls) and were two months of age at the start of irradiation and biochemical analyses. The UV group received 0.028 sunburn units (SBUs) daily (5 days wk-1) for 16 months from 40 watt BZS-WLG lamps. Stratum corneum turnover rates (SCR), cell label index (CLI), protein, DNA and RNA synthesis, and ornithine decarboxylase (ODC) induction were determined at monthly intervals over a period of two years. There were no age-related tendencies observed in SCR. CLI increased with age. Chronic, low-dose UV had no effect upon either of these parameters. Epidermal capacity for DNA and protein synthesis increased with age from 2 months to 12-15 months at which time both parameters peaked and then began to decline. UV significantly reduced (P0.04) the magnitude of DNA synthetic capacity at peak periods of synthesis but had no effect upon protein synthesis. RNA synthetic rates declined with age, reaching their lowest levels at 24 months. Further, a significant reduction (P0.001) in ODC inducibility occurred with advancing age.(ABSTRACT TRUNCATED AT 250 WORDS)

To determine the segment along the carcinogenic continuum at which dietary lipid exerts its principal effect, six groups of 35 Skh-HR-1 hairless mice were placed on defined isocaloric diets containing either 0.75%, 12% corn oil or 12% menhaden oil as sources of omega-6 or omega-3 fatty acids, respectively. All animals received an 11 week course of UV-radiation from fluorescent sunlamps. Upon termination of UV, diets of some groups were crossed-over to either low fat, high fat, omega-6 or omega-3 fatty acid sources. The first tumor appeared at week 14. Life-table analysis of the tumor incidence curves and Wilcoxon tests of tumor multiplicity provided evidence that high corn oil diets significantly (P less than 0.01) enhance carcinogenic expression; that tumor enhancement by the omega-6 fatty acid source occurs during the post-initiation, or promotion, stage; that replacement with a low corn oil diet after UV-initiation will negate the exacerbating effect of high corn oil; and that an omega-3 fatty acid source inhibits UV-carcinogenesis even at high dietary levels, although not during the post-initiation stage.

A previous study using the hairless mouse model demonstrated that diets containing a fish-oil lipid source, which contained high levels of omega-3 fatty acids, markedly increased the minimum erythema dose (MED) when compared with diets containing other polyunsaturated fatty acids. To determine whether fish oil supplementation could produce a similar effect in humans, 20 subjects were randomized into two groups, a placebo group and a group receiving fish-oil supplements over a 4-week period. Results showed a small, but statistically significant, increase in MED in patients whose diet was supplemented with fish oil. Cholesterol and prostaglandin E2 levels were unchanged, while triglyceride levels were significantly decreased in the fish oil group. No significant changes in any of these parameters occurred in the placebo group.

Butylated hydroxytoluene (BHT) and certain carotenoid pigments have been found to inhibit photocarcinogenesis in animal models. In addition, BHT protects against UV-B-induced erythema and UV-B induction of ornithine decarboxylase. Studies on the photoprotective mechanism(s) of BHT suggested that changes in the physico-chemical properties of the keratin of the stratum corneum layer of skin occurred, leading to increases in UV absorption of that tissue. These changes might be exerted via the anti-radical action of BHT that retards oxidation and prevents cross-linking of the keratin chains, resulting in a diminution of UV-B radiation reaching potential target sites. The carotenoids beta-carotene, canthaxanthin and phytoene also inhibit UV-B carcinogenesis. beta-Carotene and canthaxanthin are excellent quenchers of singlet oxygen, and all three pigments can quench free radicals. beta-Carotene and canthaxanthin have been shown to quench singlet oxygen/free radical reactions in the skin of porphyric mice, and these two pigments as well as phytoene have been found to quench excited species formed on irradiation of mouse skin by UV-B.

To assess the preventive effect of a cyclooxygenase 2 inhibitor, celecoxib (Celebrex; G.D. SearleCo, Skokie, Ill), in UV-induced skin cancer in hairless mice.Randomized dose-response study. A total of 75 SKH-HR-1 female hairless mice, aged 2 months, were randomized into control, low-dose (200 mg twice daily human dose equivalent), and high-dose (400 mg twice daily human dose equivalent) celecoxib treatment groups. Animals received 1 J/cm(2) daily (5 d/wk) total irradiation. The animals were evaluated weekly for appearance of tumors, and the data were analyzed with respect to tumor latency period and tumor multiplicity using statistical software and Wilcoxon rank sum analyses, respectively. Prostaglandin E(2) levels in the blood and skin were assessed in each group.Veterans Affairs Medical Center, Research and Dermatology Services.Animals received restricted diets containing the Food and Drug Administration-approved human equivalent doses of 200 mg (low dose) and 400 mg (high dose) of celecoxib twice daily. Controls received no drug. Tumors were induced in all animals with an equivalent UV dose.Animals were evaluated weekly for the appearance of tumors, and data were analyzed with regard to tumor latency period and tumor multiplicity. Constitutive prostaglandin E(2) levels in blood and epidermis were assessed in each group.Low doses and high doses of celecoxib significantly lengthened the tumor latency period (P.03 and P.003, respectively) and reduced tumor multiplicity (P.005 and P.001, respectively) compared with controls. There were no differences in the constitutive levels of blood or epidermal prostaglandin E(2) in the low- or high-dose treated animals compared with controls when analyzed at study termination.Celecoxib is an effective and safe chemopreventive agent in UV carcinogenesis. The epidemiologic, laboratory, and animal studies of the influence of celecoxib on cancer incidence and its low association with systemic adverse effects have led to a potentially new therapeutic approach for the prevention of skin cancer.

It has been suggested that butylated hydroxytoluene (BHT)‡, a phenolic antioxidant that conveys photoprotective properties to skin, may act by altering epidermal parameters that influence the level of ultraviolet light (UV) reaching critical target sites. The potential diminution of effective UV dose could account for the observed inhibition of UV-induced ornithine decarboxylase (ODC). Thus studies were conducted to examine the possibility that BHT was capable of promoting epidermal proliferation, just as others have shown with lung epithelial cells. UV-induction of ODC, in unrestrained Skh-Hr-I mice irradiated with Westinghouse FS–20 lamps, was shown to be linearly responsive from 0.3 to 0.6, J/cm2/dose. The degree of inhibition afforded by dietary BHT was approximately constant over this dose range. In addition, intraperitoneal administration of BHT (400 mg/kg) inhibited UV-induced ODC as greatly as when provided dietarily. Greatest inhibition occurred when BHT was administered at least 46 h prior to UV exposure, with the inhibitory event(s) requiring over 20 h to become effective. However, BHT had no effect, during this time interval, upon the incorporation of [3H]thymidine or leucine into DNA and protein, respectively. Nor were BHT effects reflected in the epidermal labeling index. These data suggest that BHT does not induce epidermal proliferation. An alternate explanation, in which the antioxidant properties of BHT play a major role, is proposed.

Free radicals are chemical species characterized by an odd number of orbital electrons or by pairs of electrons of similar directional spin isolated singly in separate orbitals. Consequently most of these agents are highly reactive and usually exhibit an extremely short half-life, although due to steric and resonance effects some exceptions occur. Some radicals and their precursors, such as the diradical O2 which exists in the triplet state, represent a critical and essential element of normal metabolism of aerobic organisms where, under normal circumstances, controlled reduction of reactive oxygen species occurs via the cytochrome oxidase or cytochrome P-450 mixed function monooxygenase systems. In addition to reactive oxygen species, organisms may be subjected to a wide-range of other free radicals or their precursors, including those of both exogenous and endogenous origin. Elaborate defense mechanisms have evolved to avoid cellular damage from these highly reactive species. Enzymes, such as the superoxide dismutase, the glutathione peroxidase/reductase system, and catalase; interactions with conjugated diene systems such as those found in melanins, carotenoids, and tocopherols; and direct reduction by sulphydryl compounds, phenols, and purines represent but a few of these natural defense systems. Despite a strong rationale for considering free radicals as pathologic agents, progress in implicating these agents, or their reactions, in pathologic processes has been arduous. The fore-most hurdle to providing definitive evidence for free radical involvement rests with the highly transient nature of these species, hardly reaching measurable levels in vivo and thereby making rigorous testing of the hypothesis extremely difficult. Indeed, free radical damage has been studied, for the most part, by indirect means–usually by measurement of known free radical reaction intermediates and products from which free radical involvement is implied. Nevertheless, free radical formation has been shown to occur in UV-irradiated skin and a considerable body of circumstantial evidence has been amassed that strongly infers that these agents, or reactions initiated by them, are responsible for at least some of the deleterious effects of UV upon skin.

The effects of light on in vitro skin lipid metabolism were examined. Fresh human skin specimens were irradiated with a Xenon lamp at a level equivalent to 10 x the minimal erythema dose. After irradiation the skin specimens were incubated 6 hours in the presence of 14 C-acetate. Total lipids were extracted and the levels of acetate incorporation determined. The total lipid free sterol, neutrol fat, and phospholipid fractions exhibited lower levels of incorporation in irradiated skin than did those from nonirradiated controls. Parallel studies on skin obtained from swine indicated that reduced levels of incorporation of acetate into the various classed of lipids were the results of altered synthesis and not due to reduced isotope uptake, increased lipid breakdown, or direct effects of light per se, on the lipid moieties. Further studies with human skin demonstrated that the light susceptible site of sterol synthesis lies along the biosynthetic pathway between acetate and meyalonate. The data suggest that light interferes with sterol synthesis prior to acetyl CoA and probably involves acetate activation or the availability of substantia, acetyl CoA pools.

Human skin specimens, previously incubated with 14 C-cholesterol, were irradiated with a mercury arc lamp for 2, 10, 20 and 60 minutes. Total lipids were extracted and preliminary separation of the lipid classes obtained by thin-layer chromatography. Cholesterol-derived photoproducts were observed as polar materials remaining at the origin of the chromatogram. The occurrence of these products was further verified by silicic acid column chromatography. Among those photoproducts tentatively identified by co-chromatography with reference oxidation compounds of cholesterol were cholestan-3β, 5α, 6β-triol, 7α and β-hydroxycholesterol, 7-ketocholesterol and cholesterol α-oxide. Quantitation and relative concentrations of the photoproducts were based upon 14 C-radioactivity measurements. Formation of these products was approximately proportional to exposure time. Further identification of two of these products, 7-ketocholesterol and cholesterol α-oxide, was obtained by gas-liquid chromatography-radiotracer analysis. The latter, cholesterol α-oxide, has been reported to possess carcinogenic properties when administered to experimental animals.

The lactate dehydrogenase activity in mouse ear epidermis exposed to monochromatic wavelengths of ultraviolet light was determined. The wavelengths used in this experiment were between 260 and 380nm at intervals of 20nm. The exposure energy was 50 millijoules per cm 2 for each wavelength. Ultraviolet light exposure caused an immediate increase in enzyme activity, which remained above normal for at least 3hr post exposure. The greatest increase in enzyme activity was observed at 280nm, whereas activity remained close to the normal level at 320nm. The effect of 8-methoxypsoralen injected intraperitoneally prior to ultraviolet exposure was determined. Lactate dehydrogenase activity was significantly higher in 8-methoxypsoralen-treated mice than in untreated controls. The increase in enzyme activity associated with 8-methoxypsoralen administration was found to be maximal between 320nm and 340nm.

Biochemical parameters of normal and actinically induced tumorous skin were compared. Similar respiratory rates and respiratory quotients were observed. However, both quantitative and qualitative differences occur in these tissue's ability to incorporate14C-acetate into lipids.

An antioxidant supplemented diet provided marked systemic protection against ultraviolet light mediated erythema in hairless mice. Among the individual constituents of the diet, butylated hydroxytoluene was most effective whereas glutathione and vitamins C and E afforded negligible protection. The mixture of antioxidants, and butylated hydroxytoluene individually, demonstrated diminished, but significant, protection when applied topically. The safety of this systemic photoprotectant and its clinical relevance at present is unknown.

Several antioxidants were tested for their protective effect in Chinese hamster embryo cell against ultraviolet light (UVL)-irradiation. Ascorbic acid, DL-alpha-tocopherol, butylated hydroxytoluene, and reduced glutathione were all shown to reverse UVL-induced cytotoxicity. At concentrations tested, no protective effect was detected when any of these antioxidants were added prior to UVL irradiation.

Dietary antioxidants were effective in reducing the number and severity of ultraviolet light-induced squamous cell carcinomas in skin of hairless mice. This and previous studies suggest that antioxidants may play a role in both inhibition of the primary event(s) involved in ultraviolet light-mediated tumor induction and the subsequent development of precancerous lesions into tumors.

The presence of 5α,6α-epoxy-cholestan-3β-ol, an oxidation product of cholesterol, has been demonstrated in a variety of animal tissues. No definitive biologic role for this sterol has been forthcoming, although the physiologic implications of its occurrence are many. The presence of the epoxide in ultraviolet (UV)-irradiated skin was detected by a combination of thin layer chromatography (TLC) and gas liquid radiochromatographic (radio-GLC) techniques in 1971 and led to the suggestion that it might be responsible for the carcinogenic properties of UV. Subsequently, in vivo levels of this sterol in the skin of UV-irradiated animals were quantitated by radio-GLC after TLC separation and preparation of the radiolabeled acetate ester. Presence of the compound in mouse liver was detected by GLC of the trimethylsilyl (TMS) ether and confirmed by mass spectrometry (MS). In all cases, however, in vivo quantitations were complicated by substrate induction of sterol epoxide hydrase, an enzyme that hydrates the epoxide to form cholestan-3β,5α,6β-triol (triol). Consequently, the relationship of the sterol epoxide to UV-carcinogenesis is as yet unclear and recent studies suggest that if this sterol is involved, then further metabolism of the compound must be required. A practical means of examining the metabolites of the epoxide was developed, employing radio-TLC scanning. Using this procedure, the relationships of substrate concentration, time of reaction and pH optimum to reaction rate were determined. Although in short term in vitro incubations of liver homogenates the bulk of epoxide is hydrated to triol, several other metabolites are detectable by TLC autoradiography. Even though the case for direct involvement of cholesterol epoxide in the etiology of carcinogenesis or other diseases is equivocal, the epoxide or its metabolites may yet prove valuable as a diagnostic aid in revealing abnormal function associated with certain diseases.

The distribution of cholesterol-5alpha,6alpha-epoxide in ultraviolet light-irradiated mouse skin was determined. Highest levels of cholesterol-5alpha,6alpha-epoxide were found in the epidermis and distributed rather evenly throughout all subcellular fractions. Low but potentially significant levels of the compound diffused from the epidermis. Metabolic studies demonstrated that cholesterol-5alpha,6alpha-epoxide was converted to cholestane-3ss,5alpha,6ss-triol by an enzyme, cholesterol-5alpha,6alpha-epoxide hydrase, which was localized in the epidermal fractions sedimenting at 10,000 and 100,000 X g. Comparative studies indicated a much higher capacity for metabolism of cholesterol-5alpha,6alpha-expoide in liver than in skin. The effect of the epoxide and its metabolic product on the skin is discussed.

The activity of interferons was tested in ultraviolet light-induced skin tumors in mice. After the tumors were well established, they were injected and measured daily for 19 days. Mouse virus type (IF-alpha + IF-beta) and immune (IF-gamma) interferons were injected intralesionally into three groups of test animals and compared with a fourth group which received mock interferon (control). When used separately, virus type and immune interferons did not affect tumor growth; however, we observed regression in tumor size when the two interferons were used in combination.

The effect of several dietary antioxidant supplements upon ultraviolet light-induced ornithine decarboxylase activity was determined. Hairless mice received diets supplemented with either butylated hydroxytoluene, disulfiram, phenobarbital, glutathione (reduced), or a special antioxidant mixture for 2 weeks before irradiation with FS-20 fluorescent sun lamps. Epidermal ornithine decarboxylase activity, the induction of which is thought to be a necessary component of skin tumor promotion, was determined at designated post-irradiation periods. Significant inhibition of ornithine decarboxylase induction was found in epidermis from animals receiving diets containing butylated hydroxytoluene, the antioxidant mixture, or disulfiram whereas no significant effects were noted in animals receiving reduced glutathione or phenobarbital. Butylated hydroxytoluene, at physiological concentrations, had no effect upon ornithine decarboxylase activity when added directly to the reaction mixture. Nor did this compound, when provided in the diet of animals, evoke a notable effect upon 12-0-tetra-decanoylphorbol-13-acetate induced ornithine decarboxylase. The latter finding suggests that dietary butylated hydroxytoluene inhibition of ultraviolet light-induced ornithine decarboxylase is a response related directly to the degree of irradation insult rather than a general effect upon the processes associated with carcinogenic promotion.

Studies from this laboratory, employing a hairless mouse model, have indicated that a polyunsaturated fatty acid source rich in omega-3 (n−3) fatty acid (FA) inhibits ultraviolet (UV)-carcinogenic expression, when compared to that of diets containing predominantly n−6 fatty acids. Omega-3 FA is a poor substrate for cyclo-oxygenase, the rate-limiting step in prostaglandin (PG) synthesis—the latter, particularly PGE2, are known to influence tumor biology. Based upon this rationale, plasma and cutaneous PGE2 levels were determined from hairless mice fed diets containing either 4% or 12% corn or menhaden oil. After two weeks on the respective diets, plasma PGE2 levels of corn oil-fed animals were approximately 6-fold greater than those of the menhaden oil-fed groups. A similar response was found in the dermis. Although the relationship to carcinogenic expression is unknown, dietary n−3 FA content can have a pronounced effect upon PGE2 levels and possesses the potential for influencing other immunomodulators.

Recent reports suggest that eicosapentaenoic acid (EPA) inhibits growth of transplanted tumors and the formation of various chemically induced cancers in animals. Menhaden oil is a source of polyunsaturated lipid with a high content of EPA, an omega-3 fatty acid. We sought to explore the effects of menhaden oil on photocarcinogenesis by employing the hairless mouse/UV-carcinogenesis model. Five groups of 40 SKH-Hr-1 mice received (a) a semipurined equicaloric diet containing either 0.75% corn oil, 4% corn oil. 4% menhaden oil or 12% menhaden oil and (b) an escalating regimen of UV radiation to a cumulative dose of 70 J/cm2. Additional animals were employed to further examine the role of menhaden oil in acute cutaneous responses to UV, i.e., erythema, edema, and ornithine decarboxylase (ODC) induction. After a 2-week feeding period UV-induced ODC activity in mice receiving 12 or 20% menhaden oil was 3 to 13-fold lower than that of corn oil fed animals. Further studies showed that edema was also markedly decreased. Animals receiving menhaden oil required twice the level of irradiance to evoke a comparable erythema as that which occurred in corn oil fed animals. Menhaden oil exerted significant influence upon carcinogenic expression, as manifested in significantly longer tumor latent periods and lower tumor multiplicities when compared to corn oil fed animals. Although the mechanism(s) of these menhaden oil effects remains unknown, it is clear that this unique lipid has a pronounced influence upon cutaneous photoresponses. 15th Annual Meeting of the American Society for Photobiology, Bal Harbor, FL, June, 1987.

Several hypotheses to explain ultraviolet carcinogenesis have been advanced. One such theory contended that cholesterol was directly involved in actinic carcinogenesis. Although the hypothesis, in its original form, was generally abandoned by the scientific community, it has been revived and modified from time to time as structures and functions of steroid hormones become more clearly understood. In essence it suggests that carcinogenic substances, structurally related to steroid hormones might result from photochemical conversion of cholesterol. Although some compounds with such properties have been isolated under controlled chemical conditions, until recently the failure to find such compounds in biological systems had cast serious doubt upon the validity of this hypothesis. It has now been demonstrated that cholesterol-derived oxidation products are formed in human skin upon exposure to ultraviolet light. One of the products formed is known to possess carcinogenic properties when administered to experimental animals.

THE role of ultraviolet light and sunlight in the aetiology of certain types of human skin cancer is now well established1. The mechanism(s) by which light induces such changes, however, are not understood. One explanation2,3 of the induction of skin cancers by ultraviolet light and sunlight is that steroids in the skin are converted photochemically to carcinogenic materials, but failure to obtain such compounds by irradiation of naturally occurring sterols in biological systems has cast doubt upon the validity of such a hypothesis. Although Blum has argued against this hypothesis4, he has recently described the induction by ultraviolet radiation of a diffusible substance that mediates hyperplasia5. Here we provide evidence for the formation of a cholesterol-derived carcinogenic photoproduct in human skin.

THERE is no doubt that sunlight and ultraviolet light cause certain types of skin cancer1,2. Unna first called attention to the high incidence of skin cancer among sailors exposed to sunlight, and his observation on the relationship between the development of skin cancers and exposure to sunlight has been confirmed by many others.

Cholesterol-5alpha,6alpha-epoxide hydrase activity was 96 percent greater in skin of hairless mice that were receiving suberythemic ultraviolet light irradiation for 15 weeks than in nonirradiated controls. This enzyme system, which metabolizes cholesterol-5alpha,6alpha-epoxide (a known carcinogen), appears to be substrateinducible and is apparently responsible for the concomitant reduction of the sterol carcinogen that occurs prior to tumor induction.

An increased release of acid phosphatase from the lysosomes of UVL irradiated hairless mouse epidermis is demonstrated. The results indicate that lysosomal membrane stability is decreased when the hairless mouse is exposed to either acute or chronic UVL.

Hautbestrahlung beim Menschen mit Xenon- oder UV-Licht und die dabei beobachteten chemischen Vorgange (Hemmung von 1-14C-Azetat und Einbau in die Gesamtlipide) legen eine mogliche Relation zwischen Photoproduktbildung und lichtinduzierter Hemmung der Lipogenese nahe.