Adrenalectomy Abrogates Reduction of 12-O-Tetradecanoylphorbol-13-acetate-induced Extracellular Signal-regulated Protein Kinase Activity in the Epidermis of Dietary Energy-restricted SENCAR Mice: Implications of Glucocorticoid Hormone¹

A substantial body of literature demonstrates that DER³ is among the most potent and reproducible cancer preventive regimens in animals (1). Previous studies in our laboratory have found that DER is a potent inhibitor of skin tumor promotion in the DMBA-initiated and TPA-promoted two-stage mouse skin carcinogenesis model (2). Several lines of evidence have indicated that the cancer prevention effect of DER may be mediated by GCH (1). Increased cortical adrenal gland activity in underfed animals was reported previously (3), and several studies suggested a critical role of the adrenal gland in the cancer-inhibitory effect of DER (4, 5, 6). Among the adrenal gland hormones, GCH is consistently elevated by DER in numerous animal studies (7), including our recent studies on SENCAR mice used in the two-stage skin carcinogenesis model (8). The importance of GCH in mediating DER inhibition of skin carcinogenesis is further demonstrated by the findings that GCH and its synthetic analogues are potent inhibitors of skin tumor promotion by TPA (9, 10, 11). GCH may exert its skin tumor-inhibitory effects, at least in part, by inhibiting DNA synthesis and cell proliferation in skin (9, 12). In addition, the relative potency of synthetic and physiological GCHs as anti-inflammatory agents is correlated with their potency in the inhibition of tumor promotion (10).

ERK plays an important role in cell proliferation and tumor development (13). We have reported recently that TPA treatment resulted in a profound induction of ERK activity in the epidermis of SENCAR mice, whereas the ERK induction was abolished in mice prefed 40% DER diet. c-Jun NH₂ terminal kinase and p38 kinase activity, the other two mitogen-activated protein kinases, were not changed by TPA or DER (14). Thus, based on the GCH inhibition of skin tumor promotion (9, 10, 11), the requirement of adrenal glands for skin tumor inhibition by food restriction (5) and the elevation of GCH in DER animals (1, 7), we propose that adrenal gland hormones, especially GCH, may be important mediators in the DER inhibition of ERK activation, a potential mechanism for DER inhibition of skin tumor promotion.

To address the hypothesis that DER inhibition of ERK activity requires intact adrenal glands and that GCH from the adrenal gland contributes to the DER inhibition of ERK activity in mice with intact adrenal glands, two studies were conducted using female SENCAR mice. In study I, Adx was performed before the initiation of DER to assess the role of the adrenal gland in the DER inhibition of ERK activity induced by a single TPA treatment. Study II was conducted by adding CORT, the major GCH in mice, to the drinking water of Adx mice with the hypothesis that addition of corticosterone would restore the DER inhibitory effect of ERK in Adx mice. In addition, mouse skin was initiated with DMBA and promoted with multiple TPA treatments [DMBA-TPA (two times/week for 8–10 weeks)] in study II, and ERK activity was measured at 48 h or 1 h after the last TPA treatment to evaluate the basal as well as acute ERK induction in the epidermis by this classic DMBA-TPA carcinogenic regimen.

Female SENCAR mice were obtained at 5–6 weeks of age from the NIH facilities at Frederick, MD. They were housed individually and kept on a 12-h light/12-h dark cycle. One week after their arrival, mice were shaved and subjected to either Adx or sham operation. The mice were anesthetized with sodium pentobarbital, and the adrenals were removed via a dorsal approach as described earlier (15). Sham-operated mice underwent the same procedure to expose the adrenals without their removal. The mice were given 0.5 ml of sterile saline i.p. after surgery to prevent dehydration. Mice were allowed two weeks of recovery before the wound clips were removed and the initiation of the chemical or experimental diet treatments. At the time of sacrifice, the kidney and surrounding tissues were visually examined to verify complete removal of the adrenals in Adx mice. Mice with any trace of adrenal tissue at the site were excluded from the experiment.

Two separate studies were conducted. In study I, mice were divided into four groups based on their surgery and diet: (a) sham/AL, which were allowed free access to modified AIN-93 diet; (b) sham/DER, which received a daily aliquot of a specific AIN formulation diet that restricted energy intake from fat and carbohydrate to 60% of the AL consumption, i.e., 40% DER; (c) Adx/AL; or (d) Adx/DER. Mice were given the experimental diets for 8–10 weeks and were terminated 1 h after a single treatment of 3.2 nmol of TPA (in 200 μl of acetone) or 200 μl of acetone (vehicle control) on the shaved dorsal skin. In study II, Adx mice were given either saline (0.9% NaCl) or 60 μg/ml CORT in the drinking water for the duration of the experiment. The concentration of CORT given to the Adx mice was selected according to a recent study showing that addition of 60 μg/ml CORT in the drinking water effectively restored the peak plasma CORT level in the Adx mice to that of the sham/DER mice (16). Mouse skin was initiated with 10 nmol of DMBA (in 200 μl of acetone) 2 weeks after the surgery and was followed by twice weekly 3.2 nmol of TPA treatment that lasted 8–10 weeks. Vehicle control groups were treated with 200 μl of acetone in place of DMBA and TPA application throughout the study. At the end of the study, mice were killed between 9 and 10 a.m., and at 48 h or 1 h after the last TPA treatment to assess ERK activity in the epidermis at the time when the next TPA treatment would be applied in the DMBA-TPA treatment regimen (48 h after the last treatment) and at the time a maximal induction of ERK was expected (1 h after TPA). Thus, based on the surgery applied, TPA treatment, diet, and drinking fluid that they were assigned to, mice were divided into 10 treatment groups as shown in Table 1. DER mice were subjected to 20% DER diet during the first 2 weeks of the DER regimen to increase their survival rate and were then given 40% DER diet for the remaining 8–10 weeks of the experiment. The composition of experimental diets was described earlier (14, 17). Except for fat and carbohydrate, DER mice were given an equivalent amount of all dietary components as their AL counterparts. Food intakes and body weights were recorded weekly. Feed was provided to the AL mice weekly and to the DER mice in daily aliquots about 30 min before the lights were turned off.

ERK activity was determined using an immune complex kinase assay as described earlier (14). Briefly, mouse skin was removed after decapitation and blood collection and was immediately immersed in liquid nitrogen. Epidermis was scraped from the frozen skin using a single-edge razor blade and was lysed in 1 ml of cold lysis buffer (14). Homogenates were centrifuged, and total protein concentration was determined by standard BCA protein assay (Bio-Rad, Hercules, CA). ERK protein was immunoprecipitated with anti-ERK rabbit polyclonal antibody (Santa Cruz Biotechnology, Inc., Santa Cruz, CA) that was prebound to protein A-agarose beads. The immunoprecipitates were collected by centrifugation, followed by extensive washes in cold lysis buffer and kinase buffer (14). Kinase activity was assayed for 30 min at 30 °C in the presence of 11 μg of MBP (Life Technologies, Inc., Grand Island, NY), 7 μCi of [γ-³²P]ATP (NEN Life Science, Boston, MA), and 15 μl of kinase buffer. The reaction products were subjected to electrophoresis after the completion of an assay. The radioactive, labeled, phosphorylated MBP bands were visualized and quantified by a PhosphorImager (Molecular Dynamics, Sunnyvale, CA) analysis of the dried gel. ERK activity in each epidermal sample was represented by the density of the radioactive band and was normalized by the background radioactive reading on the blot.

All mice were sacrificed between 9 and 10 a.m. Mice were quickly decapitated upon removal from their cages to avoid a hypothalamic-pituitary-induced elevation of GCH in mice with intact adrenals. The blood was collected in heparinized centrifuge tubes, and the plasma was separated and stored at −70°C for future simultaneous assay. Plasma CORT concentrations were measured by radioimmunoassay according to the manufacturer’s instructions (ICN, Costa Mesa, CA).

Body weight and plasma CORT were analyzed by single-factor ANOVA, and differences were tested by t test, based on residual mean square from the ANOVA. ERK activity as measured by radioactive band densitometry was statistically analyzed by t test to evaluate the difference between two groups and then converted to percentages of the acetone-treated controls for presentation.

The average body weights of the mice in study II are shown in Fig. 1. Mice fed AL diet gained weight gradually until the end of the study. There was no significant difference in the body weight of the AL mice in different treatment groups at the end of the study. The body weight of the DER mice dramatically dropped 1 week after they started 40% DER diet (week 8) and maintained a lower weight until the end of the experiment. At the end of the study, there was a 25% decrease in the body weight of the DER mice compared with the AL mice (24.2 g ± 0.7 versus 31.9 g ± 0.4; P < 0.001, ANOVA test). There was no significant difference among different DER groups through the experiment. Similar body weight changes in AL (Sham/AL and Adx/AL) and DER (Sham/DER and Adx/DER) mice were observed in study I (data not shown).

To evaluate the role of the adrenal gland in the DER inhibition of TPA-induced ERK activity, in study I we performed Adx before the administration of DER diet and measured ERK activity 1 h after a single dose of 3.2 nmol of TPA treatment of the mouse skin (Fig. 2). ERK activity was elevated after TPA treatment in sham-operated AL mice (Fig. 2,A, Lane 3) and in Adx mice (Lanes 4 and 6) but not in sham-operated DER mice (Fig. 2,A, Lane 5). As expected, DER or Adx had no effect on the ERK activity in the acetone-treated mice. Thus, ERK activity in each treatment group was expressed as the percentage of the average ERK activity in the acetone-treated groups across the surgery and diet treatment. There was nearly a 70% reduction of the TPA-induced ERK activity in sham/DER mice compared with sham/AL mice (P < 0.05, t test), whereas an inhibition of ERK activity by DER was not observed in Adx mice (Fig. 2 B). Thus, Adx abrogated DER inhibition of TPA-induced ERK activity.

In study II, ERK activity was assessed in the mouse skin that was treated with the DMBA-initiation and TPA-promotion (two times/week for 8–10 weeks) regimen (Fig. 3). The basal ERK activity measured 48 h after the last TPA treatment in the sham-operated DMBA-TPA-treated mice displayed a 2-fold but not significant increase compared with acetone-treated controls. This basal level of ERK activity was not significantly changed by DER. On the other hand, there was a nearly 5-fold increase of ERK activity measured 1 h after the last TPA treatment in AL mice compared with acetone-treated controls and a 2.5-fold increase compared with the level measured at 48 h after TPA treatment (P < 0.05). However, as expected, ERK was not induced in the epidermis of DER mice. Similar to study I, DER markedly inhibited TPA-induced ERK activity in sham-operated mice (P < 0.05), whereas Adx abolished the DER inhibition of TPA-induced ERK activity. Most importantly, addition of CORT to Adx mice markedly suppressed ERK activity to the level of the sham/DER mice, irrespective of diet treatment (P < 0.05). These results provide evidence that CORT is a potent inhibitor of DMBA-TPA-induced ERK activity and is a major hormone that contributes to the inhibition effect of DER on TPA-induced ERK activity in mice with intact adrenal glands.

Plasma CORT concentrations of the experimental mice were measured in blood samples collected at the end of study II at 9–10 a.m. (Fig. 4). DER increased the CORT level by 1.5-fold over AL/sham (P < 0.05). Adx significantly decreased CORT levels compared with sham-operated DER mice (P < 0.05). Addition of CORT to the drinking water restored the hormone level to that of the sham/DER group, irrespective of diet treatment.

DER is a potent inhibitor of carcinogenesis in a wide variety of organ sites including skin and is associated with other beneficial effects such as an increase in longevity (7). The precise mechanisms by which DER inhibits cancer development have not been fully elucidated. Our laboratory has recently reported evidence that DER strikingly inhibited ERK activity in the epidermis of SENCAR mice after a single promoting dose of TPA treatment (14). Because ERK is important in tumor development (18, 19), understanding the mechanisms by which DER inhibits ERK activity may provide significant insights in skin cancer prevention. Studies reported here indicate that the adrenal gland, and more specifically, increased plasma CORT in the DER mice are necessary components in the DER effect. Results from study I are consistent with our previous study, which showed that application of a single promoting dose of TPA led to a marked in vivo ERK induction in the epidermis of SENCAR mice and that the induction was abrogated by DER (14). As an extension, in study II we investigated the ERK activity in the mouse skin initiated with DMBA and promoted with TPA treatment, a short-term procedure that mimics the two-stage skin carcinogenesis model. There was no statistically significant induction of basal ERK activity as measured 48 h after the last TPA treatment. However, a nearly 5-fold induction over acetone controls was detected 1 h after the last TPA treatment in the sham/AL mice. To our knowledge, this is the first time that the effect of DMBA-TPA treatment on the ERK activity in vivo was reported. Most importantly, the TPA-induced ERK activity using this regimen was also inhibited by DER in the sham-operated mice, providing further evidence that DER may prevent skin carcinogenesis through the inhibition of ERK induction by TPA.

The adrenal gland has been implicated as a mediator of the beneficial effects of DER in a number of studies (4, 5, 6). Adx stimulated carcinogen-induced mammary tumor growth (4) and the tumor-inhibitory effect of food restriction in skin and lung carcinogenesis models were reversed by Adx (5, 6). In particular, GCH, a major adrenal cortical hormone, has been found elevated in food-restricted or energy-restricted animals (1, 7) and underfed humans (20). These data indicate that increased circulating GCH is an adaptation of animals or humans in response to reduced food or energy intake, and GCH may play a role in mediating the beneficial effect of DER. In this study, we found that the plasma CORT levels of the DER mice were 1.5-fold higher than those of AL fed mice. This elevation was attenuated compared with an earlier study (8), where we found a 10-fold rise of plasma CORT levels at 7 a.m. in the mice prefed 40% DER for 8–10 weeks. The differences may be caused by the different experimental protocols and the different time of the day when the mice blood samples were collected. It is well established that the physiological GCH levels display a circadian rhythm in rodents, being low in the early morning, gradually increasing in the daylight hours, and reaching the peak value in late afternoon (21). Compared with other studies in our laboratory, the basal CORT level in Sham/AL mice measured between 9 and 10 a.m. in this study was higher than the value measured early in the morning (8), and the hormone levels in Adx mice supplemented with GCH were lower than those measured in late afternoon (16). It should be noted that we detected CORT in Adx mice that were not given the hormone supplementation, possibly produced by other tissues in the mice such as the cardiovascular tissue and brain (22, 23). Interestingly, there was also a small but significant elevation of CORT by DER in the NaCl/Adx mice in comparison with NaCl/Adx/AL mice. The mechanisms by which DER increases the GCH level are not known.

GCHs and glucocorticoid analogues possess marked anti-inflammatory and antiproliferative effects, and these effects have been correlated with their potency in the inhibition of tumor promotion (9, 10). It has been shown that GCHs can suppress the growth of certain lung cancers (24) and inhibit hyperplasia and neoplasia in a number of systems including several types of leukemia and lymphoma (25). Studies on mouse skin carcinogenesis models have demonstrated that when applied topically or orally, glucocorticoids can inhibit skin tumor promotion by croton oil (26). Other investigations found that synthetic GCH such as dexamethasone and fluocinolone acetonide were potent inhibitors of TPA-induced promotion of skin carcinogenesis (10).

Data reported here indicated that GCH may target the ERK signal transduction pathway to mediate the tumor-inhibitory effect of DER. Supporting this notion, a recent study in our laboratory demonstrated that adrenalectomy reversed DER inhibition of TPA-induced AP-1:DNA binding, and supplementation of CORT restored the inhibition effect (27). On the other hand, the nuclear localization of GCH receptor, an indirect measurement of GCH receptor activation, was not changed by DER (8). Thus, the observed inhibition of AP-1:DNA binding activity in DER mice may not be attributable to interference by activated GCH receptor. But rather this reduced AP-1:DNA binding activity may be attributable to reduced AP-1 constituent proteins because of reduced ERK1/2 signaling. Taken together, these data provide strong evidence that the adrenal glands play a critical role in mediating the tumor-inhibitory effect of DER in mouse skin, and one of the major responsible adrenal gland secretary products is CORT.

The fact that Adx abrogated the inhibitory effect of DER on TPA-induced ERK in the epidermis was striking. The adrenal gland secretes a number of hormones that regulate stress and metabolism. Considering the importance of GCHs in maintaining energy metabolism in the body (28), it is not surprising that CORT is a main factor in the adrenal glands that mediates the mouse response to reduced dietary energy intake. Epinephrine and nonepinephrine play important roles in mediating energy metabolism in response to acute stress, and aldosterone is an important mediator of salt and water balance (28). Currently, there is no evidence that supports their involvement in the tumor development and their roles in DER, if any, are not elucidated.

Although the data on ERK activity in study II strongly suggest the involvement of CORT in mediating the DER inhibition effect, it should be noted that it is unlikely that the effects of DER are mediated via GCHs alone. Other hormones may also be involved (29, 30). In addition, DER may act concomitantly on metabolic pathways regulated not only by corticosterone but also by other hormones or growth factors that are modified in response to DER (7). Furthermore, removal of the adrenal gland probably induced a profile of hormone changes involved in the hypothalamic-pituitary-adrenal gland axis, including some hormones that may have an impact on the TPA-induced ERK activation in the adrenalectomized DER mice observed in this study. However, CORT clearly played a central role in the DER inhibition of ERK1/2, as apparent from the restoration of inhibition in the corticosterone supplemented groups.

In summary, our data provide the first evidence that DER acts to inhibit ERK activity via a mechanism involving adrenal glucocorticoid hormones, suggesting a mechanism by which DER inhibits skin carcinogenesis.

We thank Dr. Douglas Lewis for assistance in editing.