Melatonin decreases cell proliferation and transformation in a melatonin receptor-dependent manner

Cancer Letters 151 (2000) 133±143 www.elsevier.com/locate/canlet

Melatonin decreases cell proliferation and transformation in a melatonin receptor-dependent manner q Marla P. Jones a, Melissa A. Melan b, Paula A. Witt-Enderby a,* a

Department of Pharmacology and Toxicology, Duquesne University, Pittsburgh, PA 15282, USA b Department of Biology, Duquesne University, Pittsburgh, PA 15282, USA

Received 22 July 1999; received in revised form 15 September 1999; accepted 1 November 1999

Abstract There are con¯icting claims for the role of melatonin in oncogenesis. In addition, the mechanism(s) underlying melatonin's effects in oncogenic processes is (are) unknown. In this study, the effects of melatonin exposure on cell proliferation and transformation were assessed in NIH3T3 cells transfected with either the human mt1 (NIH-mt1) or MT2 (NIH-MT2) melatonin receptors. The effects of melatonin exposure on proliferation was assessed by direct cell counts and [ 3H]thymidine uptake assays. The effect of chronic melatonin pretreatment on transformation was assessed by focus assays. In both NIH-mt1 and NIH-MT2 cells, melatonin pretreatment decreased cell proliferation and transformation. Control (NIH-neo) cells did not show this effect. However, as revealed by the [ 3H]thymidine uptake assays, an increase in DNA synthesis occurred in NIH-mt1 cells, whereas no increase occurred in the NIH-MT2 or NIH-neo cells. Upon examination of melatonin receptors, a decrease in the function of both mt1 and MT2 receptors occurred. These data suggest that perhaps an attenuation of receptor-mediated processes are involved in the anti-proliferative and anti-transformation capabilities of melatonin in NIH3T3 cells. In addition, based on the [ 3H]thymidine assays, receptor mediated signal transduction mechanisms may slow the growth of cells via actions on the cell cycle. The results from this study shed new insight on the putative mechanisms underlying melatonin's effects on cell proliferation and transformation and lends support for a protective role of melatonin in oncogenesis. q 2000 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Melatonin; Human mt1 and MT2 melatonin receptors; Transformation; Proliferation; Desensitization; Cancer

1. Introduction The effects of melatonin in oncogenesis is unclear, however, a majority of the studies conclude that the q The nomenclature used for classi®cation of melatonin receptors throughout this manuscript was approved by the nomeclature committee of the International Union of Pharmacology as published in: The IUPHAR Compendium of Receptor Characterization and Classi®cation. IUPHAR Media, London. * Corresponding author. Tel.: 11-412-396-4346; fax: 11-412396-5599. E-mail address: [email protected] (P.A. Witt-Enderby)

hormone has a protective role in the modulation of cancer [1±4]. It is believed, though not conclusively proven, that there may be a link between the function of the pineal gland (which synthesizes and secretes melatonin) and tumor formation [5±7]. That is, in patients af¯icted with certain types of cancers including leukemia, lymphoma, breast, or melanoma, hypertrophy of the pineal gland occurs [8]. Also, in many studies, those af¯icted with cancer show higher daily blood levels of melatonin as compared to healthy individuals [9±11]. These studies suggest that melatonin synthesis and secretion from the pineal gland is

0304-3835/00/$ - see front matter q 2000 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0304-383 5(99)00394-8

134

M.P. Jones et al. / Cancer Letters 151 (2000) 133±143

increased in response to the development of cancer and, therefore, acts as a protective mechanism to control neoplasia [9,10]. Although not tested experimentally, it is suggested that a feedback mechanism may occur between the pineal gland and proliferating cells. It is possible that the proliferating cells, but not resting cells, may be releasing (unknown) factors that trigger melatonin synthesis and/or secretion from the pineal gland [9]. However, this is merely speculation and further studies need to be performed to test whether this occurs. The effects of melatonin on tumor formation in rats show similar results as humans. Pinealectomized rats have a higher incidence of dimethylbenzanthracene (DMBA)-induced tumor formation compared to intact rats. In addition, administration of melatonin to intact or pinealectomized rats reduces the incidence of DMBA-induced tumor formation when compared to rats given DMBA alone [12,13]. Thus, melatonin may reduce the formation of tumors. The mechanism underlying melatonin's inhibitory effects in oncogenesis may operate through its effects on cell proliferation. In one study using rats af¯icted with R3327H Dunning prostatic adenocarcinoma, injection with melatonin results in a decrease in tumor weight and an increase in the doubling time of the tumor [14]. Similarly, melatonin exposure decreases the proliferation of SK-N-SH neuroblastoma cells [15], rat hepatoma AH130 cells [16], PC12 cells [17], human M-6 malignant melanoma cells [3], MCF-7 human breast cancer cells [2,18,19], and in ®broblasts derived from the skin of patients affected by systemic sclerosis [20]. However, decreases in cell proliferation may not always underlie the oncostatic actions of hormones or drugs. For example, treatment of Syrian hamster embryo (SHE) cells with estrogens (17b-estradiol and diethylstilbestrol) and anti-estrogens (tamoxifen, toremifene and ICI 164,134) increases the number of colony forming units (indicative of cellular transformation), without a concomitant increase in cell number [21]. Recently, however, it has been shown that melatonin decreases the invasive and metastatic properties of human MCF7 breast cancer cells [22]. Even though numerous studies suggest that melatonin may play a protective role in oncogenesis, there is evidence to support an adverse effect of melatonin within the body. For example, administration of phar-

macological doses of melatonin to C3H/He female mice results in their premature death due to the formation of reproductive tract tumors [23]. Furthermore, administration of melatonin during the morning has a stimulatory effect on tumor formation in mice bearing ®brosarcoma ascties or Ehrlich solid tumors, while late afternoon administration has an inhibitory effect [5]. Although these effects of melatonin within the body are unclear, they are most likely mediated through receptors. Activation of the mt1 melatonin receptor (formerly known as Mel1a) [24] by melatonin results in an inhibition of forskolin-induced cAMP formation [24±26], an inhibition of protein kinase A [25,27] and an inhibition of the phosphorylation of cyclic AMP response element binding protein [25,27]. Similarly, activation of the MT2 melatonin receptor (formerly known as Mel1b) [28] results in an inhibition of cAMP accumulation [28]. In support of melatonin receptor-mediated effects in oncogenesis, it is known that certain cancerous cell lines contain melatonin receptors including human MCF-7 breast cancer cells [29], mouse N1E115 neuroblastoma cells (unpublished data) and human B(E)C2 neuroblastoma cells (unpublished data). Besides plasma membranebound receptors, nuclear receptors may also be involved in melatonin's effects on oncogenesis as reviewed [30]. However, how these nuclear receptors mediate such effects remains unclear. Thus, melatonin, acting through its receptor(s) and ultimately via its signal transduction pathways may underlie the oncostatic effects of melatonin on cells. To date, it is dif®cult to ascertain whether such effects of melatonin are mediated through receptors due to: (1) the unavailability of high-af®nity and subtype-selective melatonin receptor antagonists and (2) because melatonin can penetrate cells [31], and produce its effects in a receptor-independent manner [32]. Since melatonin has the ability to exert its effects with or without a receptor, melatonin's effects on oncogenic mechanisms are dif®cult to interpret. Until selective antagonists are developed, another approach to use to study the mechanisms underlying melatonin's effects on cell proliferation and transformation is by use of transfected cell lines as models. Therefore, the direct effects of melatonin exposure on cells either containing no melatonin receptors (receptor-independent effects) or on cells expressing de®ned

M.P. Jones et al. / Cancer Letters 151 (2000) 133±143

melatonin receptor subtypes (receptor-dependent effects) can be determined. In the past, most of the research performed on cellular transformation or proliferation used, as their models, humans, animals or cells already in a cancerous state [3,16,22,29]. Since melatonin may be able to promote oncogenesis in normal animals [23] and also because healthy individuals take melatonin to promote good health, it is necessary to examine the effects of pharmacological levels of melatonin in `normal' tissue or cells. As a result, the goal of our study was to examine the effects of pharmacological concentrations of melatonin on cellular transformation and proliferation in a non-cancerous cell line and to determine whether melatonin receptors played a role.

2. Materials and methods 2.1. Development of the cell lines In this study, NIH3T3 cells were chosen because they display a well-characterized phenotype when transformed [33,34]. Their signal transduction capabilities are also well-de®ned as others have used these cells for studying muscarinic cholinoceptors [34], serotonin 5HT2C receptors [35] and melatonin receptors [36]. At the beginning of each experiment, NIH-3T3 cells were initially grown to 60% con¯uence in 10-cm dishes in Dulbecco's modi®ed Eagle's medium (DMEM), (Gibco-BRL, Grand Island, NY) supplemented with 10% donor calf serum (GibcoBRL, Grand Island, NY) and 1% penicillin/streptomycin (Gibco-BRL, Grand Island, NY). The plates of NIH-3T3 cells were then co-transfected with the pSV2-neo plasmid (Clontech, Palo Alto, CA) alone (NIH-neo) or with the human mt1 cDNA [24] (NIHmt1) or the human MT2 melatonin receptor cDNA [28] (NIH-MT2), by the method of lipofection as described previously [26], using lipofectaminee (Gibco-BRL, Grand Island, NY). Stable cell lines were maintained for 2±3 weeks in DMEM supplemented with 10% donor calf serum and 1% penicillin/ streptomycin and geneticin at 300 mg/ml (GibcoBRL, Grand Island, NY) in an incubator at 378C with 10% CO2 until appropriate assays were conducted. Cells conferring resistance to geneticin

135

were then screened for melatonin receptor expression by their ability to bind 2-[ 125I]iodomelatonin. 2.2. Pretreatment conditions In all experiments and following transfection, cells (NIH-neo, NIH-mt1, and NIH-MT2) were subjected to a two week pretreatment with either vehicle (0.1% ethanol) or melatonin (1, 10, or 100 mM) in DMEM containing 10% donor calf serum, 1% penicillin/streptomycin, and geneticin. DMEM containing geneticin and either vehicle or melatonin and geneticin was replaced three times weekly and cells were incubated at 378C in a 10% CO2 atmosphere. 2.3. Transformation assessment To determine the effects of chronic melatonin pretreatment on cellular transformation, focus assays were performed as described previously with modi®cation [34]. Following transfection of NIH3T3 cells with the various plasmids, cell cultures were then pretreated with vehicle or melatonin. Following approximately two weeks of treatment, the stably transfected cells were ®xed in cold (2208C) methanol and stained with 0.05% Giemsa (Sigma Chemical Co., St. Louis, MO). The number of foci present on the plate after the pretreatment period were scored and then expressed as a percentage of their control (vehicle-treated) cells. 2.4. Cell proliferation assessment 2.4.1. Direct cell counts To determine the effects of melatonin pretreatment on cell proliferation, cells were counted. Two days post transfection, cells (NIH-neo, NIH-mt1, and NIH-MT2) were plated at equal density (20 000± 70 000 cells/plate) and then subjected to the treatment conditions. At the end of two weeks, the stably transfected cells were counted on a hemocytometer and cell viability was also assessed using trypan blue (Sigma Chemical Co., St. Louis, MO). The data were then expressed as a percentage of their control (vehicle-treated) cells. 2.4.2. [ 3H]Thymidine uptake assays To determine the effects of melatonin pretreatment on DNA synthesis, [ 3H]thymidine uptake assays were

136

M.P. Jones et al. / Cancer Letters 151 (2000) 133±143

performed on cells as described previously with slight modi®cation [35]. Two days post transfection, cells (NIH-neo, NIH-mt1, and NIH-MT2) were plated at equal density (20 000±70 000 cells/plate) and then subjected to the treatment conditions. On day 12 of the treatment with vehicle or melatonin, transfected cells were lifted into 12 ml of DMEM and 10 ml of cells were plated into 24-well dishes containing 500 ml DMEM. Once again, pretreatment of the cell cultures was resumed as before except that the media was supplemented with 1 mCi/ml [ 3H]thymidine (0.25 mCi/ml, NEN/DuPont, Boston, MA). Following two days of incubation with [ 3H]thymidine, the media was aspirated and cells were washed with phosphate-buffered saline (137 mM NaCl, 2.7 mM KCl, 8 mM Na2PO4, 0.62 mM KH2PO4, pH 7.4). The cells were then ®xed in 5 ml methanol (2208C) and washed with tap water. Nuclei and cytoskeletal remains were solubilized in 500 ml 0.2 N NaOH/1% SDS for 2±3 h at room temperature. The solubilized lysate was added to vials containing 5 ml of scintillation cocktail and radioactivity was quanti®ed by liquid scintillation counting. Data were expressed as a percentage of their control (vehicle-treated) cells. 2.5. Autoradiography To assess whether the incorporation of [ 3H]thymidine was into DNA speci®cally, autoradiography was performed. Brie¯y, cells (NIH-neo, NIH-mt1 and NIHMT2) were treated with melatonin and [ 3H]thymidine exactly as described in section 2.4.2, except that cells were plated on slides. Following the pretreatment conditions, cells were washed repeatedly with phosphate buffered saline and ®xed with ice-cold (2208C) methanol. Next, the slides were dipped in Autoradiography Emulsion (NTB-2) (Eastman Kodak, Rochester, NY) and kept in complete darkness at 48C for 12 weeks. Next, the slides were developed in D-19 Developer (Eastman Kodak, Rochester, NY) and stained with Toluidine Blue O (Sigma ChemicalCo., St. Louis, MO). 2.6. Radioligand binding assays 2.6.1. Competition binding To assess whether the af®nity of melatonin for the human mt1 or MT2 melatonin receptor expressed in NIH3T3 cells was similar to the af®nity of melatonin for receptors reported elsewhere, competition assays were performed exactly as described previously [26].

All data were ®t by non-linear regression analysis (least squares ®t) with a one-site ®t model using GraphPad Prism Software (GraphPad Inc., San Diego, CA). 2.6.2. Total binding To determine the effects of melatonin pretreatment on receptor density, total binding assays were conducted using the radioligand 2-[ 125I]iodomelatonin. Following treatment, cells were washed with phosphate buffered saline, lifted in buffer, and pelleted by centrifugation. The cells were resuspended in Tris (50 mM, (pH 7.4)) and then added to tubes containing 500 pM of 2-[ 125I]iodomelatonin in the absence (total) or presence (non-speci®c) of melatonin (1 mM) in a ®nal reaction volume of 0.26 ml. Cells were then incubated for 1 h at room temperature and harvested by ®ltration over glass ®lters presoaked in polyethylenimine and counted in a gamma counter. All reactions were run in duplicate. The data were ®rst expressed as femtomoles of receptor per milligram of protein and then normalized against their control (vehicle-treated) cells. Protein determinations were made by the method of Bradford using BioRad protein assay reagents (BioRad, Hercules, CA). 2.7. Cyclic AMP accumulation assays To determine the effects of melatonin pretreatment on cAMP accumulation, assays were performed on NIH-mt1 and NIH-MT2 cell lines according to the methods described previously [26], except that cells were ®rst pretreated for 13 days in a 10-cm plate with vehicle or melatonin (1 mM) and then split into a 24well dish and the treatment with either vehicle or melatonin resumed. The next day, cell cultures were labeled with 2 mCi/ml [ 3H]adenine (1 mCi/ml, NEN, DuPont, Boston, MA) in DMEM still containing vehicle or melatonin (1 mM) for 5 h. All data were normalized against the forskolin response and each data set was ®t by non-linear regression analysis (least squares difference and one-site ®t) using GraphPad Prism software. 2.8. Statistical analysis Parametric data obtained from the focus assays, direct cell counts, [ 3H]thymidine uptake assays and total binding assays were analyzed by one-way

M.P. Jones et al. / Cancer Letters 151 (2000) 133±143

ANOVA, followed by Newman Keul's post hoc. For the cAMP accumulation assays, potency (IC50) values were generated from each individual data set, and then averaged with the other potency values obtained under the same treatment conditions. Ultimately, the potency of melatonin for the mt1 and MT2 receptors in vehicle-treated cells was compared to the potency of melatonin for the receptors pretreated with melatonin by Student's unpaired t-test. Signi®cance was de®ned as P , 0:05. 3. Results 3.1. Characterization of NIH3T3 cells transfected with the human mt1 and MT2 melatonin receptor

137

NIH3T3 cell cultures with the neomycin resistance plasmid alone followed by pretreatment with vehicle or with melatonin for 2 weeks exhibited no formation of foci (Fig. 1A,D). In contrast, transfection of NIH3T3 cell cultures with the human mt1 or MT2 receptor cDNA followed by pretreatment with vehicle induced the formation of foci. Pretreatment of NIHmt1 or -MT2 cell cultures with melatonin (100 mM) decreased the number of foci formed (Fig. 1B±D; Table 2). Differences in the transfection ef®ciency of the plasmids into the cells could not account for the differences in transformation between the NIHneo and NIH-mt1 and NIH-MT2 cell lines as the transfection ef®ciency for all cell lines was approximately 50% (data not shown).

Expression of the human mt1 melatonin receptor in NIH3T3 cells did not change the pharmacology or function of these receptors when compared to receptors expressed in other tissues or cells [37] (Table 1). Similarly, expression of the human MT2 melatonin receptor in NIH3T3 cells yielded similar pharmacology and function of the MT2 receptor previously reported [36] (Table 1). In NIH-neo cells, no total speci®c binding of 2-[ 125I]iodomelatonin occurred and no inhibition of forskolin-induced cAMP accumulation by melatonin (0.1 pM±100 mM) occurred (data not shown). These data con®rm the absence of endogenous melatonin receptors in NIH3T3 cells.

3.3. The effects of melatonin pretreatment on cellular proliferation in NIH-neo, -mt1, and -MT2 cells.

3.2. The effects of melatonin pretreatment on cellular transformation in NIH-neo, -mt1 and ±MT2 cells using focus formation assays

3.3.2. [ 3H]Thymidine uptake assays Melatonin pretreatment produced variable changes in [ 3H]thymidine uptake in the cell lines. No change in [ 3H]thymidine uptake occurred in NIH-neo cells during any of the melatonin pretreatment conditions compared to control (vehicle-treated) cells. However, in NIH-mt1 cells, an increase in [ 3H]thymidine uptake

Melatonin pretreatment induced variable cellular transformation between the three cell lines (NIHneo, NIH-mt1, and NIH-MT2). Transfection of

3.3.1. Direct cell counts Following melatonin (1, 10 and 100 mM) pretreatment, no change in NIH-neo cell number occurred compared to control (vehicle-treated). In both NIHmt1 and NIH-MT2 cells, a decrease in cell number occurred following 100 mM melatonin pretreatment (Fig. 2). This decrease in cell proliferation was not due to cell death as cell viability was greater than 93% after all pretreatments (data not shown).

Table 1 Characterization of NIH3T3 cells transfected with the human mt1 or MT2 melatonin receptor

NIH-neo NIH-mt1 NIH-MT2

Af®nity of melatonin (log Ki ^ SEM)

n

Receptor density (fmol/mg protein ^ SEM)

n

Potency of melatonin (log IC50 ^ SEM)

n

Not detected 29.658 ^ 0.512 M 211.33 ^ 1.71 M

3 3 3

Not detected 23 ^ 9 179 ^ 70

4 4 4

No effect 29.362 ^ 0.968 M 29.290 ^ 0.237 M

3 3 4

138

M.P. Jones et al. / Cancer Letters 151 (2000) 133±143

Fig. 1. The effects of melatonin pretreatment on focus formation in NIH-neo, -mt1, and -MT2 cells. (A) Photomicrograph of NIH3T3 cells transfected with the pSV2-neo plasmid demonstrating that no formation of foci occurred following pretreatment with either vehicle or melatonin. (B and C) Photomicrographs of NIH3T3 cells transfected with the pSV2-neo plasmid and either the human mt1 melatonin receptor cDNA (B) or the human MT2 melatonin receptor cDNA (C), demonstrating that following pretreatment with vehicle, an increase in the formation of foci occurred. However, following pretreatment with 100 mM melatonin, a decrease in the formation of foci occurred. Data are from one representative experiment repeated ®ve times. C ˆ vehicle; 1 ˆ 1 m M melatonin; 10 ˆ 10 m M melatonin; 100 ˆ 100 mM melatonin. (D). The following graph shows that a signi®cant decrease in focus formation occurred in NIH-mt1 and -MT2 cells following 100 mM melatonin pretreatment. a ˆ P , 0:05 when compared to untreated cells; b ˆ P , 0:05 when compared to cells pretreated with 1 mM melatonin; c ˆ P , 0:05 when compared with cells pretreated with 10 mM melatonin. Each bar represents the mean ^ SEM of six experiments.

occurred following pretreatment with 10 mM melatonin, which returned to control levels following 100 mM melatonin pretreatment. In contrast to the NIHmt1 cells, NIH-MT2 cells exhibited a decrease in [ 3H]thymidine uptake following 10 and 100 mM melatonin pretreatment compared to control (vehicle-treated) cells (Fig. 3). The uptake of [ 3H]thymidine was speci®cally into the DNA in all cells as revealed by autoradiography. As shown, the uptake of [ 3H]thymidine in NIH-mt1 cells following pretreatment with vehicle or 10 mM melatonin was speci®cally into the DNA and not into the surrounding cytoplasm (Fig. 3, inset).

3.4. The effects of melatonin pretreatment on mt1 and MT2 melatonin receptors 3.4.1. Total 2-[ 125I]iodomelatonin binding Following melatonin (1, 10, and 100 mM) pretreatment in the NIH-mt1 and NIH-MT2 cell lines, a decrease in 2-[ 125I]iodomelatonin binding occurred in both NIH-mt1 cells and NIH-MT2 cells (Table 2). 3.4.2. Cyclic AMP accumulation assay Melatonin (0.1 pM±100 mM) inhibited forskolininduced cAMP accumulation in NIH-mt1 cells following pretreatment with vehicle with high

Vehicle 1 mM Mel 10 mM Mel 100 mM Mel

Vehicle 1 mM Mel 10 mM Mel 100 mM Mel

Vehicle 1 mM Mel 10 mM Mel 100 mM Mel

NIH-neo

NIH-mt1

NIH-MT2

100 108 ^ 8 85 ^ 10 30 ^ 4 b,c,d

100 120 ^ 20 103 ^ 19 37 ^ 9 b,c,d

100 77 ^ 12 62 ^ 9 69 ^ 15

Proliferation total # cells (% vehicle ^ SEM)

100 96 ^ 6 62 ^ 11 52 ^ 9 b,c

100 121 ^ 14 102 ^ 6 34 ^ 6 b,c,d

# background # background # background # background

Transformation total # foci formed (% vehicle ^ SEM)

b

Data represent the mean ^ SEM from 3±10 individual experiments. P , 0:05 when compared to vehicle-treated cells. c P , 0:05 when compared to cells treated with 1 mM melatonin. d P , 0:05 when compared to cells treated with 10 mM melatonin

a

Treatment

Cell line

100 120 ^ 25 61 ^ 16 b,c 30 ^ 10 b,c

100 158 ^ 23 232 ^ 45 b,c 118 ^ 18 d

100 102 ^ 12 88 ^ 8 91 ^ 12

DNA synthesis [ 3H]thymidine uptake (% vehicle ^ SEM)

100 24 ^ 16 b 22 ^ 14 b 11 ^ 7 b

100 6 ^ 2b 0.8 ^ 0.3 b,c 1 ^ 0.3 b,c

Not detected Not detected Not detected Not detected

Total 2[ 125I]iodomelatonin binding (% vehicle ^ SEM)

Table 2 The effects of melatonin pretreatment on cell growth, transformation and melatonin receptor pharmacology and function a

42 ^ 15 0b ± ±

40 ^ 8 21 ^ 9 b ± ±

No effect No effect No effect No effect

Ef®cacy of melatonin (% maximal inhibition ^ SEM)

29.290 ^ 0.2370 0b ± ±

29.362 ^ 0.968 25.667 ^ 0.600 b ± ±

No effect No effect No effect No effect

Potency of melatonin (log IC50 ^ SEM) (M)

M.P. Jones et al. / Cancer Letters 151 (2000) 133±143 139

140

M.P. Jones et al. / Cancer Letters 151 (2000) 133±143

Fig. 2. The effect of melatonin pretreatment on cell proliferation in NIH-neo, -mt1, and -MT2 cells. The graph depicts the total number of cells following pretreatment with either vehicle or melatonin. Decreases in ®nal cell number were apparent in the NIH-mt1 and NIH-MT2 cells following exposure to melatonin. No decreases in cell number occurred in NIH-neo cells. a ˆ P , 0:05 when compared to untreated cells; b ˆ P , 0:05 when compared to cells pretreated with 1 mM melatonin; c ˆ P , 0:05 when compared to cells pretreated with 10 mM melatonin. Each bar represents the mean ^ SEM of 8±10 individual experiments.

potency. However, compared to control cells, pretreatment of NIH-mt1 cells with melatonin (1 mM) for 2 weeks and then rechallenge with melatonin (0.1 pM±100 mM) resulted in a loss of potency and ef®cacy of melatonin at these receptors. Similarly, melatonin (0.1 pM±100 mM) inhibited forskolininduced cAMP accumulation in vehicle-treated NIH-MT2 cells with high potency and ef®cacy. However, following pretreatment with melatonin (1 mM) for 2 weeks and then rechallenge with melatonin (0.1 pM±100 mM), no inhibition of forskolin-induced cAMP accumulation occurred at any concentration of melatonin tested (Table 2). 4. Discussion This study has demonstrated that chronic pretreatment of `normal' cells transfected with each of the melatonin receptor subtypes with pharmacological concentrations of melatonin: (1) reduced cellular transformation, (2) reduced cell proliferation, and (3) either reduced or stimulated DNA synthesis depending upon which melatonin receptor subtype was expressed in NIH3T3 cells. Thus, the effect of melatonin pretreatment on these processes was found to be receptor-dependent. In addition, these

studies demonstrated that these preteatment conditions resulted in (4) a decrease in 2-[ 125I]iodomelatonin binding and (5) attenuated melatonin-mediated inhibition of cAMP accumulation in NIH-mt1 and MT2 cells. The effect of melatonin on the proliferation of NIHmt1 and -MT2 cells supports existing evidence in which SK-N-SH neuroblastoma cells [15], rat AH 130 hepatoma cells [16], and PC12 cells [17], human M-6 malignant melanoma cells [3], and human MCF-7 breast cancer cells [2,18,19] show decreases in proliferation following melatonin exposure. Perhaps, the anti-proliferative effects of melatonin are due to factors released from proliferating cells but not from resting cells [9]. However, alterations in the cell cycle may also be responsible for the decrease in cell proliferation. In another study, it is believed that a delay in the movement from G0/G1 to the Sphase occurs in MCF-7 cells following melatonin pretreatment [16,38]. Although we did not assess

Fig. 3. The effect of melatonin pretreatment on [ 3H]thymidine uptake in NIH-neo, -mt1 and -MT2 cells. The following graph depicts the amount of [ 3H]thymidine uptake following pretreatment with either vehicle or melatonin. An increase in [ 3H]thymidine uptake occurred in NIH-mt1 cells following pretreatment with 10 mM melatonin whereas a decrease in uptake occurred in NIH-MT2 cells following pretreatment with 10 and 100 mM melatonin. No change in [ 3H]thymidine uptake occurred in NIH-neo cells following pretreatment with melatonin. a ˆ P , 0:05 when compared to untreated cells; b ˆ P , 0:05 when compared to cells pretreated with 1 mM melatonin; c ˆ P , 0:05 when compared to cells pretreated with 10 mM melatonin. Each bar represents the mean ^ SEM of 4±10 individual experiments performed in duplicate. Inset: autoradiography of NIH-mt1 cells following pretreatment with vehicle or 10 mM melatonin. As shown, the uptake of [ 3H]thymidine was speci®cally into the DNA and not into the surrounding cytoplasm.

M.P. Jones et al. / Cancer Letters 151 (2000) 133±143

speci®c points along the cell cycle, our studies could involve a cell cycle phenomenon except that the delay in the cell cycle appears to occur in the progression from the G2 to the M-phase in NIH-mt1 cells. This idea is supported from evidence generated through [ 3H]thymidine assays. [ 3H]Thymidine assays are an indirect method used to assess cell death or proliferation because [ 3H]thymidine is only incorporated into DNA by cells during the S-phase [39]. In our study and only in NIH-mt1 cells, an increase in [ 3H]thymidine uptake, speci®cally into the DNA, was observed following 10 mM melatonin pretreatment without a concomitant increase in cell number. Thus, more cells may be arresting at the G2 checkpoint following melatonin pretreatment and not entering cell division. Thus, because these cells are unable to proceed from S to G2 and then to mitosis, the result would be less cell proliferation by the end of two weeks. This inability to divide was most apparent following 100 mM melatonin pretreatment in the NIH-mt1 cells and may re¯ect a concentration-dependent effect of melatonin on factors within the cell (i.e. cyclin-dependent protein kinases or cyclins) [40], regulating cell cycle phenomenon. This, however, was not assessed in this study but will be investigated in the future. In contrast to the NIH-mt1 cells, pretreatment of NIH-MT2 cells with melatonin (10 mM) resulted in a decrease in DNA synthesis without a change in cell number. Perhaps in these cells, melatonin is producing effects on the cell cycle similar to MCF-7 cells. That is, following melatonin pretreatment, a delay in the cell cycle at a different checkpoint like G0/G1 to the S-phase may be occurring [38]. In addition, because melatonin can enhance or suppress DNA synthesis in NIH3T3 cells in a receptor-dependent manner, then it is possible that receptor-mediated signal transduction processes are capable of regulating cell cycle phenomena uniquely depending upon which melatonin receptor is present. The ability of signal transduction components to regulate cell proliferation are now being explored in MCF-7 breast cancer cells [29] and in prostatic smooth muscle cells [41]. The role of signal transduction processes in cell cycle regulation of these cells will be examined in more detail in the future. Transfection of NIH3T3 cells with either the human mt1 or MT2 receptor cDNA induced transformation of vehicle-treated cells as evidenced by the

141

increase in the formation of foci. No increases in focus formation were observed in NIH3T3 cells transfected with the neomycin resistance plasmid alone. These data suggest that simple expression of either the mt1 or MT2 melatonin receptor alone in the absence of a high concentration of melatonin is suf®cient to produce a transformed phenotype. In addition, this transformation is suppressed only when cells are exposed to high levels of melatonin. Previous experiments examining melatonin's effects on transformation yield similar results. For example, rats with R3327H Dunning prostatic adenocarcinoma show a reduction in tumor growth following melatonin injections [14]. Similarly, administration of melatonin reduces the number of mammary tumors as well as DMBA-induced tumors in rats [12,13]. These data also support the ®ndings found in the rat where decreased levels of melatonin predisposes the rat (with intact melatonin receptors) to DMBA-induced tumors [12,13]. A possible mechanism underlying melatonin's inhibitory effects on tumor formation in the rat could be due to its effects on cell motility. In a previous study both in vitro and in vivo, melatonin pretreatment of MCF-7 cells results in a reduction of the invasiveness of these cells possibly due to an increase in cell-to-cell adhesion [22]. This increase in cell-to-cell contact may reduce metastases and ultimately tumor formation [22]. Since melatonin's effects on transformation and cell proliferation were found to be receptor-dependent in our studies, the consequences of this same exposure on mt1 and MT2 melatonin receptor density and function were also examined. Following melatonin exposure, a decrease in 2-[ 125I]iodomelatonin binding was observed for both the mt1 and MT2 melatonin receptors. This decrease in binding was not due to residual melatonin on the receptor because vehicle-treated cells exposed to 100 mM melatonin for 2 min and then washed in parallel with the other plates exhibited no decrease in 2-[ 125I]iodomelatonin binding (data not shown). Besides a loss of 2-[ 125I]iodomelatonin binding, a loss of melatonin potency and ef®cacy at inhibiting cAMP accumulation at these receptors also occurred following melatonin exposure. From these studies, it appears that a decrease in receptor function may underlie the inhibitory effects of melatonin on cellular transformation and proliferation in both NIH-mt1 and NIH-MT2 cells. One explanation for

142

M.P. Jones et al. / Cancer Letters 151 (2000) 133±143

this could be due to melatonin's effects on another signal transduction mechanism potentially involved in transformation and proliferation, such as the mitogen activated protein kinase signal transduction cascade. In another melatonin receptor model, melatonin has been shown to inhibit forskolin-induced MAP Kinase activity [42]. Perhaps, then, desensitization of the melatonin receptors coupled to an inhibition of the cAMP-dependent signal transduction cascade, resulting in increases in cAMP and PKA, may stimulate the MAP Kinase cascade leading to a regulation of genes involved in the prevention of cellular transformation. By contrast, then, a lack of activation of MAP Kinase via activation of intact melatonin receptors (i.e. vehicle-treated receptors) due to decreases in cAMP and PKA activity may inhibit genes involved in the prevention of transformation and proliferation. These studies may begin to explain the mechanisms underlying the protective effects of melatonin in vivo. In conclusion, this study has demonstrated that exposure to high concentrations of melatonin decreases cell proliferation and transformation in NIH3T3 cells expressing the human mt1 or MT2 receptor. Thus, these processes are affected in a receptor-dependent manner. In addition, the same exposure also attenuated the function of both melatonin receptor subtypes. Perhaps melatonin, via an attenuation of its signal transduction pathway, results in a disruption in cell cycle processes within NIH3T3 cells at speci®c points along the cell cycle (i.e. possibly at the G2 checkpoint). The results from this study shed new insight on the putative mechanisms underlying melatonin's ability to reduce cell proliferation and transformation and lends support for a protective role of melatonin in oncogenesis. Acknowledgements The authors would like to thank Ms Ann Gorman for her help in the cell viability and autoradiographic studies. This work was supported by Faculty Start-up funds to PAW-E and by NIH R15NS37672 to MAM.

[2]

[3]

[4] [5] [6]

[7]

[8]

[9]

[10]

[11] [12]

[13]

[14] [15]

References [1] T.M. Molis, L.L. Spriggs, S. Hill, Modulation of estrogen

[16]

receptor mRNA expression by melatonin in MCF-7 human breast cancer cells, Mol. Endocrinol. 8 (1994) 1681±1690. Y. Furuya, K. Yamamoto, N. Kohno, Y. Ku, Y. Saitoh, 5Fluorouracil attenuates an oncostatic effect of melatonin on estrogen-sensitive human breast cancer cell (MCF7), Cancer Lett. 81 (1994) 95±98. S.W. Ying, L.P. Niles, C. Crocker, Human malignant melanoma cells express high af®nity receptors for melatonin: antiproliferative effects of melatonin and 6-chloromelatonin, Eur. J. Pharmacol. 246 (1993) 89±96. A. Brzezinski, Melatonin in humans, N Engl. J. Med. 336 (1997) 186±195. H. Bartsch, C. Bartsch, Effect of melatonin on experimental tumors under different photoperiods and times of administration, J. Neural Transm. 52 (1981) 269±279. C. Bartsch, H. Bartsch, A. Karenovics, H. Franz, G. Peiker, D. Mecke, Nocturnal urinary 6-sulphatoxymelatonin excretion is decreased in primary breast cancer patients compared to agematched controls and shows negative correlation with tumorsize, J. Pineal Res. 23 (1997) 53±58. C. Bartsch, I. Kvetnoy, T. Kvetnaia, H. Bartsch, A. Molotkov, H. Franz, Nocturnal urinary 6-sulfatoxymelatonin and proliferating cell nuclear antigen-immunopositive tumor cells show strong positive correlations in patients with gastrointestinal and lung cancer, J. Pineal Res. 23 (1997) 90±96. P. Lissoni, S. Viviani, E. Bajetta, R. Buzzoni, A. Barreca, R. Mauri, M. Resentini, F. Morabito, D. Esposti, G. Esposti, F. Fraschini, A clinical study of the pineal gland activity in oncologic patients, Cancer 57 (1986) 837±842. P. Lissoni, S. Crispino, S. Barni, A. Sormani, F. Brivio, F. Pelizzoni, A. Brenna, G. Bratina, G. Tancini, Pineal gland and tumor cell kinetics: serum levels of melatonin in relation to Ki-67 labeling rate in breast cancer, Oncology 47 (1990) 275± 277. L. Dogliotti, A. Berruti, T. Buniva, M. Torta, A. Bottini, M. Tampellini, M. Terzolo, R. Faggiuolo, A. Angeli, Melatonin and human cancer, J. Steroid Biochem. Mol. Biol. 37 (1990) 983±987. G.J.M. Maestroni, A. Conti, Melatonin in human breast cancer tissue: association with nuclear grade and estrogen receptor status, Lab. Invest. 75 (1996) 557±561. L. Tamerkin, M. Cohen, D. Roselle, C. Reichert, M. Lippman, B. Chabner, Melatonin inhibition and pinealectomy enhancement of 7,12-dimethylbenz(a)anthracene induced mammary tumors in the rat, Cancer Res. 41 (1981) 4432±4436. D.E. Blask, S.M. Hill, K.M. Orstead, J.S. Massa, Inhibitory effects of the pineal hormone melatonin and underfeeding during the promotional phase of 7,12-dimethylbenz(alpha)anthracene-induced (DMBA)-induced mammary tumorigenesis, J. Neural Transm. 67 (1986) 125±138. R. Philo, A.S. Berkowitz, Inhibition of dunning tumor growth by melatonin, J. Urol. 139 (1988) 1099±1102. S. Cos, R. Verduga, C. Fernandez-Viadero, M. Megias, D. Crespo, Effects of melatonin on the proliferation and differentiation of human neuroblastoma cells in culture, Neurosci. Lett. 266 (1996) 113±116. G. Cini, M. Coronnello, E. Mini, B. Neri, Melatonin's growth-

M.P. Jones et al. / Cancer Letters 151 (2000) 133±143

[17]

[18] [19] [20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28]

inhibitory effect on hepatoma AH 130 in the rat, Cancer Lett. 125 (1998) 51±59. J.C. Mayo, R.M. Sainz, H. Uria, I. Antolin, M.M. Esteban, C. Rodriguez, Inhibition of cell proliferation: a mechanism likely to mediate the prevention of neuronal cell death by melatonin, J. Pineal Res. 25 (1998) 12±18. D.E. Blask, S.M. Hill, Effects of melatonin on cancer: studies on MCF-7 human breast cancer cells in culture, J. Neural. Transm. 21 (1986) 433±449. S. Cos, E.J. Sanchez-Barcelo, Melatonin inhibition of MCF-7 human breast-cancer cell growth: in¯uence of cell proliferation rate, Cancer Lett. 93 (1995) 207±212. A.M. Carossino, A. Lombardi, M. Matucci-Cerinic, A. Pignone, M. Cagnoni, Effect of melatonin on normal and sclerodermic skin ®broblast proliferation, Clin. Exp. Rheumatol. 14 (1996) 493±498. T. Tsutsui, S. Taguchi, Y. Tanaka, J.C. Barrett, 17b-estradiol, diethylstilbestrol, tamoxifen, toremifene, and ICI 164,384 induce morphological transformation and aneuploidy in cultured Syrian hamster embryo cells, Int. J. Cancer 70 (1997) 188±193. S. Cos, R. Fernandez, A. Guezmes, E.J. Sanchez-Barcelo, In¯uence of melatonin on invasive and metastatic properties of MCF-7 human breast cancer cells, Cancer Res. 58 (1998) 4383±4390. W. Pierpaoli, A. Dall'ara, E. Pedrinins, W. Regelson, The pineal control of aging he effects of melatonin and pineal grafting on the survival of older mice, Ann. N. Y. Acad. Sci. USA 621 (1991) 291±313. S.M. Reppert, D.R. Weaver, T. Ebisawa, Cloning and characterization of a mammalian melatonin receptor that mediates reproductive and circadian responses, Neuron 13 (1994) 1177±1185. P.J. Morgan, P. Barrett, H.E. Howell, R. Helliwell, Melatonin receptors: localization, molecular pharmacology and physiological signi®cance, Neurochem. Int. 24 (1994) 101± 146. P.A. Witt-Enderby, M.L. Dubocovich, Characterization and regulation of the human ML1A melatonin receptor stably expressed in chinese hamster ovary cells, Mol. Pharmacol. 50 (1996) 166±174. P.A. Witt-Enderby, M.I. Masana, M. Dubocovich, Physiological exposure to melatonin supersensitizes the cyclic adenosine 3 0 ,5 0 -monophosphate-dependent signal transduction cascade in chinese hamster ovary cells expressing the human mt1 melatonin receptor, Endocrinology 139 (1998) 3064±3071. S.M. Reppert, C. Godson, C.D. Mahle, D.R. Weaver, S.A. Slaugenhaupt, J.F. Gusella, Molecular characterization of a second melatonin receptor expressed in human retina and

[29]

[30] [31] [32] [33] [34] [35]

[36] [37] [38] [39]

[40]

[41]

[42]

143

brain: the Mel1b melatonin receptor, Proc. Natl. Acad. Sci. USA 92 (1995) 8734±8738. P.T. Ram, T. Kiefer, M. Silverman, Y. Song, G.M. Brown, S.M. Hill, Estrogen receptor transactivation in MCF-7 breast cancer cells by melatonin and growth factors, Mol. Cell. Endocrinol. 141 (1998) 53±64. W.S. Baldwin, J.C. Barrett, Melatonin: receptor-mediated events that may affect breast and other steroid hormonedependent cancers, Mol. Carcinog. 21 (1998) 149±155. L.M.E. Finocchiaro, G.C. Glikin, Intracellular melatonin distribution in cultured cell lines, J. Pineal Res. 24 (1998) 22±34. G. Benitez-King, F. Anton-Tay, Calmodulin mediates melatonin cytoskeletal effects, Experientia 49 (1993) 635±641. J. Jainchill, S.A. Aaronson, G.J. Todaro, Murine sarcoma and leukemia viruses: assay using clonal lines of contact-inhibited mouse cells, J. Virol. 4 (1969) 549±553. J.S. Gutkind, E.A. Novotny, M.R. Brann, K. Robbins, Muscarinic acetylcholine receptor subtypes as agonist-dependent oncogenes, Proc. Natl. Acad.Sci. USA 88 (1991) 4703±4707. R.S. Westphal, E. Sanders-Bush, Differences in agonist-independent and -dependent 5-hydroxytryptamine 2c receptormediated cell division, Mol. Pharmacol. 49 (1996) 474± 480. C. Godson, S.M. Reppert, The mel1a melatonin receptor is coupled to parallel signal transduction pathways, Endocrinology 138 (1997) 397±404. M.L. Dubocovich, Melatonin receptors: are there multiple subtypes, Trends Pharmacol. Sci. 16 (1995) 50±56. S. Cos, J. Recio, E.J. Sanchez-Barcelo, Modulation of the length of the cell cycle time of MCF-7 human breast cancer cells by melatonin, Life Sci. 58 (1996) 811±816. A. Najid, M.-H. Ratinaud, Comparative studies of steroidogenesis inhibitors (econazole, ketoconazole) on human breast cancer MCF-7 cell proliferation by growth experiments, thymidine incorporation and ¯ow cytometric DNA analysis, Tumori 77 (1991) 385±390. B. Alberts, D. Bray, J. Lewis, M. Raff, K. Roberts, J.D. Watson, The cell-division cycle, in: Molecular Biology of the Cell, 3rd Edition, Garland Publishing, Inc, New York, 1994, pp. 863±910. J.H. Guh, T.L. Hwang, F. Ko, S. Chueh, M. Lai, C. Teng, Antiproliferative effect in human prostatic smooth muscle cells by nitric oxide donor, Mol. Pharmacol. 53 (1998) 467± 474. D.G. Hazlerigg, M. Thompson, M.H. Hastings, P.J. Morgan, Regulation of mitogen-activated protein kinase in the pars tuberalis of the ovine pituitary: interactions between melatonin, insulin-like growth factor-1, and forskolin, Endocrinology 137 (1996) 210±218.