Evaluation of gene expression in human lymphocytes activated in the presence of melatonin

International Immunopharmacology 2 (2002) 885 – 892 www.elsevier.com/locate/intimp

Evaluation of gene expression in human lymphocytes activated in the presence of melatonin Enrica Capelli a,*, Ilaria Campo a, Simona Panelli b, Giuseppe Damiani c, Maria Grazia Santagostino Barbone d, Adele Lucchelli d, Mariaclara Cuccia a a Dipartimento di Genetica e Microbiologia, Universita` di Pavia, Via Ferrata 1, 27100-Pavia, Italy Istituto di Patologia Generale Veterinaria, Universita` di Milano, Via Celoria 10, 20133-Milan, Italy c IDVGA-CNR, Palazzo LITA, Via F.lli Cervi 93, 20090-Milan, Italy d Dipartimento di Farmacologia Sperimentale e Applicata, Facolta` di Farmacia, Universita` di Pavia, Viale Taramelli 12, 27100-Pavia, Italy b

Received 22 June 2001; received in revised form 25 September 2001; accepted 20 January 2002

Abstract The effect of melatonin on the expression of genes previously correlated to T lymphocyte activation (HLA-DRB, thymosin beta 10 (h-Tim)) and to Lymphokine Activated Killer (LAK) activity (h-Tim, Tumour Rejection Antigen (TRA 1), nRap 2) was investigated in phytohemagglutinin (PHA)-stimulated human lymphocyte cultures. The aim was to find an enhancing effect of this substance on anti-tumoral immune defences as suggested by studies on tumour progression in mice and clinical immunotherapy trials in humans. mRNA obtained from melatonin-treated and -untreated PHA-stimulated lymphocytes was retrotranscribed and amplified by RT-PCR using primers based on the sequences of the selected genes. The results suggest that melatonin does not increase T and LAK cell responses: in fact, a reduction in the transcription of all the considered genes was observed. These data are correlated with the antiproliferative effect of melatonin observed in in vitro treated lymphocytes. D 2002 Elsevier Science B.V. All rights reserved. Keywords: Melatonin; T lymphocytes; LAK cells

1. Introduction It has been suggested that melatonin enhances immune system function and melatonin treatment in mice leads to an increase in thymus weight and in tritiated thymidine incorporation into T lymphocytes [1]. From in vivo studies, it is known that melatonin binds to specific receptors on T CD4 positive lym-

*

Corresponding author. Tel.: +39-382-505528; fax: +39-382528496. E-mail address: [email protected] (E. Capelli).

phocytes and stimulates the production of cytokines, including IL-4 and GM-CSF [2]. According to recent reports [3,4,5], melatonin seems to stimulate anti-tumour defences and is able to counteract the growth of both spontaneous and experimental tumours. In humans, immunotherapy trials with melatonin in addition to administration of IL-2 or Lymphokine Activated Killer (LAK) cells or radiotherapy have demonstrated a general positive effect on behaviour and a reduction of tumour progression. [2,6]. It is well known that tumour immunity is supported by T cytotoxic CD8 positive cells, NK and LAK cells

1567-5769/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved. PII: S 1 5 6 7 - 5 7 6 9 ( 0 2 ) 0 0 0 2 1 - 8

886

E. Capelli et al. / International Immunopharmacology 2 (2002) 885–892

[7,8,9]. LAK cell differentiation is induced by IL-2. LAK cells probably represent many cell types belonging to T or NK subsets that are specifically stimulated by IL-2 and have a common cytolytic anti-tumoral activity. These cells are also present in phytohemagglutinin (PHA)-stimulated cultures because of the secretion of IL-2 by stimulated T helper lymphocytes. Previously, we have associated the expression of particular genes with LAK cell differentiation in a human experimental cell model in vitro [10]. The genes whose up-regulation we have associated with such a differentiation model code for Tumour Rejection Antigen (TRA 1), thymosin beta 10 (h-Tim) and for a new gene named nRap 2 [11]. This gene codes for a protein highly homologous to the Ras-related protein Rap 2A. The aim of the present work is to investigate the interference of melatonin with the expression of these genes together with RAD 51 (which codes for the human homologous of bacterial Rec A), GAPDH (which codes for the enzyme glyceraldehyde-3-phosphate dehydrogenase) and HLADRB (which codes for the b subunit of the class II HLA-DRB molecule). These genes, known to be associated with cell activation [12,13,14], were studied concomitantly. For this purpose, we used as a model peripheral blood mononucleated cells (PBMC) stimulated in vitro with phytohemagglutinin (PHA) and cultured in the presence and in the absence of melatonin. Gene expression was studied by RT-PCR using specific primers based on the selected genes.

2. Materials and methods 2.1. Cells Peripheral blood mononucleated cells (PBMC) from six healthy unrelated subjects were obtained by centrifugation of heparinized blood on a 1077 density gradient (Istopaque, Gibco). The human cell line Chang (ATCC CL13) was routinely maintained in RPMI 1640 medium (Gibco) supplemented with 10% foetal calf serum (FCS). 2.2. Lymphocytes cultures PHA-stimulated cells: PBMCs from each subject were treated with phytohemagglutinin (PHA Gibco,

1% in RPMI 1640 medium, 1  106 cells/ml) for 96 h. After 2 days, the non-adherent cells were centrifuged and fresh medium, containing PHA, was added. IL-2-stimulated cells: PBMCs were pulsed with PHA (1%) for 12 h, then the medium was discarded and fresh medium containing recombinant IL-2 (Eurocetus, 50 pg/ml), supplemented with 10% FCS, was added. Finally the cell suspension was distributed in 24 multiwell plates (1.6 mm diameter/well, 1  106 cells/ml/well) and the cultures were incubated at 37 jC in a CO2 humidified cell incubator for 8 days. Every 2 days the medium was replaced with fresh medium containing IL-2. 2.3. Melatonin treatment Lymphocytes were stimulated with PHA for 24 h, distributed in microwells conical plates (0.2  106 cells) in complete medium containing melatonin (10 3 – 10 12 M) and incubated for 24, 48 and 72 h in the presence of the drug. PBMC cultures without melatonin were set up in the same way and served as controls. 2.4. Mitotic index The percentage of mitoses in unstimulated and PHA-stimulated lymphocyte cultures treated with melatonin at different concentrations and in control cultures were assessed 48 and 72 h after starting the treatments. Five hours before cell harvesting, colcemid (1 ng/ml) was added to all cultures. Finally, the cells from each culture were harvested, briefly fixed (5 min) with a methanol –acetic acid solution (3:1) and dropped onto slides. The smears thus obtained were air dried and stained with a Giemsa solution (10% in distilled water buffered with pH 6.8 Sorensen solution). Mitotic indexes were calculated based on an average number of 1000 cells screened per sample, in duplicate. 2.5. Apoptotic index The percentage of apoptotic cells was determined in parallel to the evaluation of the mitotic index in PBMC derived cultures and in the human continuous cell line Chang (ATCC CL13). As a positive

E. Capelli et al. / International Immunopharmacology 2 (2002) 885–892

control, cultures treated with Etopside (Vepesid) were used. 2.6. Cytofluorimetric analysis Pools of the lymphocytes suspensions from treated and untreated cultures were labeled with DAPI and analyzed by a flow cytometer (FACStar Becton-Dickinson, Mountain View, CA). 2.7. Statistical analysis Statistical comparisons for mitotic and apoptotic indexes were performed by means of the Fisher’s exact test. 2.8. RNA isolation and cDNA synthesis Total cellular RNA was isolated from PHA-stimulated cells cultured in the presence and in the absence of melatonin, for 48 h (see Section 2.3), from unstimulated control and IL-2-treated cells by the guanidinium thyocianate method as described elsewhere [15]. The concentration of the purified RNA was evaluated spectrophotometrically. Samples were then reversetranscribed using the anchored primer T18VN (V = A, C or G; N = A, C, G or T) in the presence of Maloney murine leukaemia virus (MMLV) using the first-strand cDNA synthesis kit (Amersham Pharmacia Biotech), according to the manufacturer’s specification. 2.9. RT-PCR As control we amplified the constitutively expressed 18S rRNA gene, using primers 18SU (ACCTGGTTGATCCTGCCAGT) and 18SL (AATTACCGCGGCTGCTGGCT). The selected genes were amplified by using the following couples of primers: RAD 5V(GGCCCAGGTAGCGGCAAAAC) and RAD 3V(AGCAGCTGGAGGAGGCGGGA), GAPDH 5V(TGAAGGTCGGTGTCAACGGATTTGGC) and GAPDH 3V(CATGTAGGCCATGAGGTCCACCAC), TRA1 5V(CAGCAGAAGACACAACAGAAGA) and TRA1 3V(TTCCTGTGACCCATAATCCC), h-Tim 5V (GGAACGAGACTGCACGGATT) and h-Tim 3V (GCTCGTGTCCATCTTGCA), DRB 5V(CAAYGG(CAAYGGGACSGAGCGGGT Y=T or C; S=C or G)

887

and DRB 3V(ACCCCGTAGTTGTGTCTGCA), nRap 2 5V(AGGGTCGTCTTACAAAATGA) and nRap 2 3V(ATTGCATGTTGCTTATTAGGA). Amplification reactions were performed in a 10-Al volume containing: 2 Al cDNA, 67 mM Tris –HCl, pH 8.8, 16 mM (NH4)2SO4, 0.01% Tween 20, 1.5 mM MgCl2, 200 AM of each dNTP, 100 ng of each primer, 0.5 units of Taq polymerase (Eurobio). Reactions were assembled at 0 jC, primers were denatured, and the tubes quickly transferred to the thermal cycler (Gene Amp PCR System Perkin Elmer 2400) preheated at 90 jC. A first denaturation step at 94 jC was followed by a specific amplification protocol. The conditions were: 35 cycles at 92 jC for 30 s, 55 jC for 30 s, 72 jC for 1 min for 18S rRNA primers; 35 cycles at 92 jC for 30 s, 60 jC for 30 s, 72 jC for 1 min for RAD primers; 40 cycles at 92 jC for 30 s, 58 jC for 30 s, 72 jC for 1 min for GAPDH primers; 35 cycles at 92 jC for 30 s, 60 jC for 30 s, 72 jC for 1 min for h-Tim primers; 35 cycles at 92 jC for 30 s, 55 jC for 30 s, 72 jC for 1 min for TRA 1 primers; 40 cycles at 92 jC for 30 s, 59 jC for 30 s, 72 jC for 1 min for DRB primers; 35 cycles at 92 jC for 30 s, 56 jC for 30 s, 72 jC for 1 min for nRap 2 primers. Amplification products were separated by agarose gel electrophoresis for 2 h at 5 V/cm in a 2% (w/v) agarose gel (Seakem GTG-FMC) with TBE buffer containing 0.5 Ag/ml (w/v) of ethidium bromide, as described [16]. As molecular weight markers, the Hyper Ladder I Bioline (TRA 1, h-Tim, RAD 51 and GAPDH) and the 100 bp Ladder Gibco (DRB and Rap) were used.

3. Results The concentrations of melatonin and the experimental conditions were set up in preparatory experiments in which the ability of melatonin to support lymphocyte proliferation was tested. The ability of the drug to stimulate PBMCs or to increase proliferation of PHA-stimulated cells was determined from literature showing an increase in thymus weight and in 3H thymidine incorporation in thymic lymphocytes [1]. Peripheral blood mononucleated cells were treated in vitro for 24, 48 and 72 h with melatonin (10 12 – 10 3 M). Examination of stained smears

888

E. Capelli et al. / International Immunopharmacology 2 (2002) 885–892

(see Materials and methods) from each culture did not reveal the presence of mitoses or blasts. Moreover, a progressive degeneration of cells was observed, with the increasing time of incubation, indicating progressive death of the lymphocytes. In contrast, in the control cultures treated with PHA, the presence of mitoses and blasts was observed. The same results were obtained on testing the ability of conditioned media from melatonin-treated cultures to induce lymphocyte proliferation. The ability of melatonin to enhance the stimulation of activated T lymphocytes was tested on PHA-stimulated PBMC cultures. The drug (10 6 – 10 3 M) caused a decrease in the percentage of mitoses and an increase in the percentage of apoptotic cells (Fig. 1). Lower concentrations (10 12 –10 7) were ineffective (data not shown). These results were confirmed using as target the tumorigenic human cell line Chang. The cytofluorimetric analysis of PBMC treated with melatonin confirmed that the drug has an antiproliferative effect. In both the presence and the absence of melatonin, unstimulated cells produced signals indicative of a lower DNA content than cells treated with PHA (Fig. 2). In the PHA-stimulated cultures, a peak typical of cells undergoing duplication of the DNA (G2 and M phases) is well evident, and this is reduced in the presence of melatonin. Based on the data obtained in the preparatory steps, we performed a molecular analysis using PBMC cultures stimulated with PHA in the presence of 10 5 M melatonin for 48 h. At this concentration, the percentage of mitoses was reduced to 40– 50% of

Fig. 1. Percentages of mitoses (broken line) and apoptoses (continuous line) in PHA-stimulated cultures of lymphocytes treated with melatonin at different concentrations.

the control and the apoptotic index was significantly increased (2.3% vs. 1.3% of the control; p < 0.05). The cell models considered were: unstimulated lymphocytes (PBL), IL-2-stimulated lymphocytes (IL-2), PHA-stimulated lymphocytes (PHA) and lymphocytes stimulated with PHA in the presence of melatonin (MLT). Total RNA was obtained and then mRNA was reverse-transcribed (see Materials and methods). As controls of cDNA synthesis and concentration, we amplified the housekeeping transcript of the ribosome subunit 18S. The RT-PCR amplification pattern obtained using the RNA 18 S primers demonstrated that the expected band (480 bp) was present with the same intensity in all the RNA preparations. Analysing the electrophoretic patterns obtained following RTPCR on the selected genes, we can observe (Fig. 3a) that a band corresponding to the TRA 1 gene (256 bp) is present in IL-2-stimulated cells, as expected. The same band is also present in the cells treated with PHA only and disappears when cells are treated with PHA and melatonin. The same pattern is obtained in the case of the HLA-DRB gene: the specific band of 200 bp is obtained from cells treated with IL-2 or PHA and is not detectable when cells are treated with PHA plus melatonin (Fig. 3b). h-Tim and nRap 2 were previously shown to be specifically expressed in IL-2 induced LAK cells [11]. The electrophoretic pattern shows that h-TIM (250 bp) is expressed at high level in IL-2-treated cells, a faint band is observed for PHAstimulated cells and the signal is absent in the cells treated with PHA plus melatonin (Fig. 3c). The 277-bp band corresponding to the nRap 2 sequence is present both in IL-2-stimulated and in PHA-treated cells, and disappears in the cells treated with PHA in the presence of melatonin (Fig. 3d). Using the primers specific for the RAD 51 gene, the expected band of 400 bp is detectable under all conditions, but a reduction of the signal is observed in cells treated with PHA plus melatonin, compared to cells treated with PHA alone (Fig. 3e). The RT-PCR amplification pattern obtained with the primers based on the sequences of the GAPDH gene confirms that melatonin induces a decrease in cell activation. In fact, the expected band (982 bp) that is observed for unstimulated PBMC disappears in IL-2-treated or in PHA-treated cells and again appears in the cells treated with PHA plus melatonin (Fig. 3f).

E. Capelli et al. / International Immunopharmacology 2 (2002) 885–892

) or treated (MLT+) with melatonin (10

5

M, 48 h culture).

889

Fig. 2. Cytofluorimetric analysis of unstimulated and PHA-stimulated PBMC cultures untreated (MLT

890

E. Capelli et al. / International Immunopharmacology 2 (2002) 885–892

Fig. 3. Differential amplification pattern (RT-PCR) obtained using specific primers for TRA 1 (a), DRB (b), h-Tim (c), nRap 2 (d), RAD 51 (e), GAPDH (f) genes. The numbers at the sides of the bands correspond to their lengths in base pairs. Lane 1: freshly isolated lymphocytes. Lane 2: IL-2-activated cells. Lane 3: PHA-treated cells. Lane 4: PHA-treated cells in the presence of melatonin (10 5 M). Lane M: molecular weight marker (Hyper ladder I BioLine for TRA 1, h-Tim, RAD 51, GAPDH genes; 100 bp Ladder Gibco for DRB and nRap 2 genes).

E. Capelli et al. / International Immunopharmacology 2 (2002) 885–892

4. Discussion In the present study, the possible interference of the pineal hormone melatonin with activation of cells involved in the immune response was investigated. For this purpose, we considered the expression of genes known to be associated with lymphocyte activation. The variation in gene expression induced by melatonin in human peripheral blood mononucleated cells stimulated in vitro with PHA was studied. Moreover, we compared the results with those previously obtained in a model based on IL-2 induced LAK cell subset [10]. The results suggest that melatonin caused a general decrease in mRNA synthesis, since the expression of all the genes considered, except GAPDH, was reduced when the cells were treated with melatonin. The increase in the GAPDH gene expression is in agreement with the above data. This gene codes for the glycolytic enzyme glyceraldehyde3-phosphate dehydrogenase (GAPDH)/uracil DNA glycosilase (UDG), that is a cell-cycle-dependent protein. GAPDH/UDG gene expression is markedly reduced when DNA synthesis is maximal [12]. We consider particularly significant the melatonininduced decrease in TRA1 and nRap 2 genes transcription because these genes have been previously associated with the differentiation of lymphocyte subsets with LAK activity. TRA 1 codes for a protein that is a molecular chaperon involved in the mounting of antigenic peptides to MHC class I proteins. This gene is the human homologue of the murine gp96 gene and codes for a heat shock protein of the endoplasmic reticulum. The complex between TRA 1 and specific peptides elicits CD8+ cytolytic T lymphocyte response and tumour immunity. Therefore, exogenous administration of TRA 1 obtained from tumour or infected cells can be a useful method of vaccination against cancer or infectious disease [17]. TRA 1 plays a crucial, direct role in antigen presentation, as a part of a larger molecule on cell surface [18] and also indirectly influences the presentation of endocellular antigens by MHC molecules [19]. The presence of TRA 1 on activated lymphocytes enhances the peptide repertoire normally presented by MHC molecules and agrees with the proposed role of this protein in the priming of CD8 positive cells and in tumour processes. Moreover, TRA 1 might be involved with a more stringent control of proliferating lymphocytes to cope

891

with the possibility of unwanted tumour transformations [17]. The nRap 2 gene has been identified in LAK cells [10]. Sequence analysis revealed that this gene is a novel member of the Ras proto-oncogene family and the coded protein presents a high homology with the Rap 2A isoform [11]. Under our experimental conditions, PHA-activated lymphocytes as well as IL-2-activated lymphocytes expressed both TRA 1 and Rap 2 genes. This is to be expected, because of the presence, in PHA-treated cultures, of IL-2-dependent lymphocyte subsets that are maintained following secretion of the interleukin by T helper cells. In cells treated with PHA and melatonin, the levels of transcription of TRA 1 and nRap 2A genes were markedly reduced when compared to those detected in lymphocyte cultures treated with PHA alone. These results suggest that melatonin may antagonize the effect of IL-2. Withregardtotheexpressionoftheothergenesconsideredinthisstudy(RAD51,h-Tim,HLA-DRB),melatonin inducedageneralreductionintheirexpression.Thesedata correlatewiththefailureofmelatonintostimulatePBMC and with its ability to reduce proliferation in PHA-treated cells,asevaluatedbymeasurementofmitoticindexesand cytofluorimetric analysis. Moreover, a slight, dosedependent, induction of apoptosis in melatonin-treated cells was observed, indicating a cytotoxic effect of this hormone. A possible biological consequence of this phenomenonisafurtherreductionofproliferativeactivity. We therefore suggest, from our data and because of the anti-tumour effect of melatonin described by other researchers [20], that this molecule can slow down tumour growth through a simple cytostatic action. We used an activation model of in vitro cultured lymphocytes: these cells are known to express surface receptors for melatonin [2]. According to the literature [21], binding of melatonin to these receptors causes the release of opioid peptides that act as mediators of lymphocyte activation. For this reason, we tested the ability of conditioned medium to stimulate freshly isolated PBMC. The results obtained in our model are not in agreement with the activation theory. The molecular results could be related to the interaction of melatonin with nuclear receptors or could be due to a general effect of melatonin on synthetic activities. The effect of melatonin on the expression of the GAPDH/UDG gene suggests a further mechanism:

892

E. Capelli et al. / International Immunopharmacology 2 (2002) 885–892

because of the involvement of this enzyme in the assembly of intracellular calcium, melatonin, via GAPDH, could have a role in the inter-nucleosome digestion of DNA that is typically activated in the apoptotic cells [22]. References [1] Maestroni GJ. The immunoendocrine role of melatonin. J Pineal Res 1993;141:1 – 10. [2] Gonzales-Haba MG, Garcia-Maurino S, Calvo JR, Goberna R, Guerrero JM. High affinity binding of melatonin by human circulating T lymphocytes (CD 4+). FASEB J 1995;9:1331 – 5. [3] Blask DE, Sauer LA, Dauchy RT, Holowachuk EW, Ruhoff MS, Kopff HS. Melatonin inhibition of cancer growth in vivo involves suppression of tumor fatty acid metabolism via melatonin receptor-mediated signal transduction events. Cancer Res 1999;59:4693 – 701. [4] Bartsch C, Bartsch H. Melatonin in cancer patients and in tumour-bearing animals. Adv Exp Med Biol 1999;467:247 – 64. [5] Lissoni P, Barni S, Mandala M, Ardizzoia A, Paolorossi F, Vaghi M, et al. Decreased toxicity and increased efficacy of cancer chemotherapy using the pineal hormone melatonin in metastatic solid tumour patients with poor clinical status. Eur J Cancer 1999;35:1688 – 92. [6] Lissoni P, Barni S, Meregalli S. Modulation of cancer endocrine therapy by melatonin: a phase II study of tamoxifen alone. Br J Cancer 1995;71:854 – 6. [7] Hokland M, Kjaergaard J, Kuppen PJ, Nannmark U, Agger R, Hokland P, et al. Endogenous and adoptively transferred A-NK and T-LAK cells continuously accumulate within murine metastases up to 48 h after inoculation. In Vivo 1999;13:199 – 204. [8] Li Q, Chang AE. Adoptive T-cell immunotherapy of cancer. Cytokines Cell Mol Ther 1999;5:105 – 17. [9] Weiss L, Reich S, Slavin S. Allogeneic cell therapy in murine B-cell leukemia (BCL1). The role of non-activated and rIL-2 activated CD4+ and CD8+ T cells in immunotherapy for leukemia. Cytokines Cell Mol Ther 1999;5:153 – 8. [10] Damiani G, Capelli E, Comincini S, Mori E, Panelli S, Cuccia M. Identification of mRNA differentially expressed in lymphocytes following Interleukin-2 activation. Exp Cell Res 1998;245:27 – 33.

[11] Panelli S, Damiani G, Paltrinieri S, Capelli E, Sartorelli P, Cuccia M. A novel member of the Ras proto-oncogene family: domain architecture, tissue-specific expression and evolutionary conservation. II Congr FISV, Riva del Garda (TN) 2000; 245:56. [12] Tashiro S, Kotomura N, Shinohara A, Tanaka K, Ueda K, Kamada N. S phase specific formation of the human Rad51 protein nuclear foci in lymphocytes. Oncogene 1996;12: 2165 – 70. [13] Mansur NR, Meyer-Siegler K, Wurzer JC, Sirover MA. Cell cycle regulation of the glyceraldehyde-3-phosphate deidrogenase/uracil DNA glycosylase gene in normal human cells. Nucleic Acids Res 1993;21:993 – 8. [14] Mortarini R, Belli F, Parmiani G, Anichini A. Cytokine-mediated modulation of HLA-class II, ICAM-1, LFA-3 and tumorassociated antigen profile of melanoma cells. Comparison with anti-proliferative activity by rIL1-h, rTNF-a, rIFN-g, rIL4 and their combinations. Int J Cancer 1990;45:334 – 41. [15] Chomiczinski P, Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate – phenol – chloroform extraction. Anal Biochem 1987;172:156 – 9. [16] Sambrook J, Fritsch EF, Maniatis T. Molecular cloning: a laboratory manual. Cold Spring Harbor, NY: Col Spring Harbor Laboratory Press, 1989. [17] Heike M, Noll B, Meyer zum Buschenfelde KH. Heat shock protein – peptide complexes for use in vaccines. J Leukoc Biol 1996;60:153 – 8. [18] Altmeyer A, Maki RG, Feldweg AM, Heike M, Protopov VP, Mansur SK, et al. Tumor-specific cell surface expression of the KDEL containing endoplasmic reticular heat shock protein gp96. Int J Cancer 1996;69:340 – 9. [19] Li Z, Srivastava PK. Tumor rejection antigen gp96/grp94 is an ATPase: implication for protein folding and antigen presentation. EMBO J 1993;12:3143 – 51. [20] Jones MP, Melan MA, Witt-Enderby PA. Melatonin decreases cell proliferation and transformation in a melatonin receptordependent manner. Cancer Lett 2000;151:133 – 43. [21] Maestroni GJ, Zammaretti F, Pedrinis E. Hematopioietic effect of melatonin involvement of type 1 kappa-opioid receptor on bone marrow macrophages and interleukin-1. J Pineal Res 1999;27:145 – 53. [22] Redd JC. Bcl-2 and the regulation of programmed cell death. J Cell Biol 1994;124:1 – 6.