Long-term melatonin supplementation does not recover the impairment of natural killer cell activity and lymphocyte proliferation in aging mice

Life scicnns, Vat. 61, No. 9, pp. 857-W&,1997 cqyligbt Q 1997 F!kvkr scienceInc. Printedin the USA. AUrigbl0reurvcd cn?A-z%/w 317.00 + .Mt

ELSEVIER

PII soO24-3205(97)00587-0

LONG-TERM MELATONIN SUPPLEMENTATION DOES NOT RECOVER THE IMPAIRMENT OF NATURAL KILLER CELL ACTIVITY AND LYMPHOCYTE PROLIFERATION IN AGING MICE

Mauro Provinciali, Giuseppina Di Stefkno, Daniele Buiian, Stefania Stronati, Nicola Fabris* Immunology Center, Gerontology Research Department, INRCA, Ancona, Italy *Institute of Haematology, University of Pavia, Pavia, Italy (Received in fmal form May 27, 1997) Summary In this study we evaluated the effect of long-term melatonin (MEL) treatment on the cytotoxic activity and number of natural killer (NK) cells and the proliferative response of spleen lymphocytes to phytohemagglutinin (PHA) or interleukin-2 (IL-2) in old mice Seventeen-eighteen month-old Balb/c mice were supplemented with MEL (405Opg/day/mouse) and sacrificed after eight months. The MEL supplementation was unable to recover the low levels of both endogenous and IL-2-induced NK cell activity fwnd in old untreated mice. Also the NK cel1 number was unaf&cted by MEL treatment. The spleen lymphocyte proliferative response to both PHA and IL-2 was not different in old MEL-treated compared to old untreated mice. These results indicate that long-term MEL supplementation does not recover the age-related deterioration of NK cell activity and lymphocyte proliferative capacity. Key Words: melatonin, natural killer ceils, lymphocyte proliferation, mice, aging, neuroimmunomodulation A considerable amount of evidence has demonstrated that the pineal gland, via its hormone melatonin (MEL) enhances immune fimctions either at central or peripheral level (I-3).The administration of pineal extracts or MEL produces thymic hyperplasia, increases the antibody response, the proliferative response to ConA, and the antigen presentation by macropbages (4-7). Conversely, surgical or pharmacological pinealectomy determines thymic atrophy, depression of humoral and cell mediated immune responses, and decrease of IL-2 production (4,8,9). Influence of the pineal gland and melatonin on NR cells has been suggested both in mice and humans (10-12). However, these effects have been inconsistent, possibly because of experimental paradigm and length of treatment. Chronic MEL administration to healthy subjects (daily for 60 days) augmented NK cytotoxicity and the number of NK effecters (10). In another study, MEL treatment of healthy volunteers for either 15 or 28 consecutive days did not modify their NK cell activity and number and the IL-2-induced generation of LAK cells (11). In mice, effectors’fiom the spleen of pinealectomized animals displayed reduced NK activity and IL-2 production, which were restored by acute (a single injection), but not chronic (9 consecutive days), treatment with melatonin (9). ComWJodnce: Dr. Mauro Prok&li, Immunology Center., Gerontol Res Dept, MRCA, Via Bkuelli 8,60121 ~COM, Italy, Fax : +71-20679, E-mail: [email protected]

858

Melatonin Supplementation

in Aging

Vol. 61, No. 9, 1997

The imrnunoenhancing effects of MEL are particularly relevant in old ages since plasma MEL levels are decreased with advancing age both in man and rodents (12,13) and the recovery of MEL turnover results in a significant restoration of several immune parameters depressed in aging (1417). In fact, it has been demonstrated that MEL administration recovers age-related thymic involution (14) reduced thyrnic endocrine activity and T-lymphocyte subsets (15) impaired in old animals (16). We recently antibody response, T helper activity and IL-2 production demonstrated that MEL treatment prevents age-related thymus involution through regulation of thymocyte apoptosis (17) The aim of this paper was to investigate the efficacy of long-term MEL treatment in very old mice on the cytotoxic activity and number of NK cells and the proliferative capacity of spleen lymphocytes in response to PHA or IL-2

Methods Animals and experimentaldesign. Balblc inbred male mice were used at the age of 2-3 (young) and

17-18 (pre-senescent) months. Young Balb/c inbred mice are reported to have normal circadian fluctuations in their plasma melatonin (3,5). The mice were housed conventionally in plastic non galvanised cages (4-6 mice per cage) and fed with standard pellet food (NossanJtaly) and tap water “ad libitum”. Pre-senescent (17-18 months of age) inbred Balb/c male mice were treated with melatonin for eight months and then sacrificed at 25-26 months of age. In our housing conditions the life span of Balb/c mice ranged between 28 and 29 months (mean life span = 23 + 1 months). The animals were maintained on a 12 h light/l2 h dark cycle from 7-00 a.m. to 7:00 p.m., at constant temperature (20 f 1’C) and humidity (50 f 5%). Darkness and light exposure were controlled by a fixed timer governing 2 standard fluorescent fixtures (Philips TDL 36 W/84). 7 p.m light off, 7 a.m. light on. Melatonin treatment.Melatonin

(MEL), was purchased from Sigma (St. Louis, MO,USA). 100 mg of MEL were dissolved in 1 ml absolute ethanol and mrther diluted to 100 ml distilled water. MEL stock solution was prepared freshly twice a week. 5 ml of the MEL stock solution were diluted to 500 ml tap water The final concentration of MEL in the drinking tap water was 10 @ml Control mice received tap water containing 0.0001% ethanol. MEL was administered in the drinking tap water with a fixed darkness cycle: both control and melatonin-containing opaque bottles were administered from 6:00 p.m. to 8:30 a m of next day MEL was administered every day for 8 months. Water intake was monitored at monthly intervals. The intake range was 4-5 ml/day/mouse (40-50 pg melatonin/day/mouse).

Preparation and culture conditions of spleen cells and thymocytes. Spleen and thymus were teased through a 60 mesh sieve in Ca*‘and Mg2’ -free phosphate buffered saline (PBS, GIBCO, Gaithersburg, Md, USA) solution. Spleen cells were then fractionated on Iympholyte M (Cedarlane, Canada) and mononuclear cells separated by density gradient centrimgation (500 g, 20 min.). Cells from the interface of the gradients were washed twice with PBS and resuspended in RPMI 1640 containing penicillin (100 U/ml) and streptomycin (100 l&ml). To evaluate IL-2-induced NK cell activity, spleen lymphocytes were incubated overnight at 37°C in 5 % CO, at a concentration of 3

x 106/rnl in RPMI 1640 plus 10% foetal calf serum (FCS, GIBCO) with or without 1000 U/ml rIL2 (Proleukin, EuroCetus, The Netherlands).

Vol. 61, No. 9,1997

Melatonin Supplementatianin Agiog

859

Target cells. The cytotoxic activity of NK cells was tested against the YAC-1 tumor cell line. These cells were maintained in continuous culture throughout the study in medium containing RPMl 1640 (GIBCO), 10% decomplemented foetal calf serum (FCS), 100 @ml streptomycin, 100 U/ml penicillin (all from GIBCO).

Cytotoxic assq. NK cell assay was performed by a fluorimetric method as recently reported ( 18). Briefly, a stock solution of carboxyfluorescein diacetate (c’FDA, Molecular Probes, Oregon, USA) (2Omg/ml acetone, stored at -2O’C) was diluted in phosphate buffered saline (PBS) to give tinal concentration of 75ygAnl The NK sensitive cell line YAC-1 was used as target cell. Tumor cells were washed twice with PBS and then labelled with CODA by resuspending the ceils in I ml of working solution and incubating at 37” in a humidified 5% CO2 incubator for 30 min. Target cells were then washed 3 times in PBS containing 1% BSA (Sigma, USA) and resuspended in KPMI +lO% FCS at a concentration of 1~10~ /ml. 1~10~ #DA-labelled tumor target cells were incubated with effector cells in 200 pl total volume in 96-well round microtiter plates (Nunc, W.G.). Effector : target cell ratios from 1OO:l to 12.5:l were tested in triplicate. The plates were kept at 37” in a humidified 5% CO, incubator for 3 h and then centrifbged at 700 x g for 5 min. The supematant was separated from the cellular fraction by rapidly inverting the plate and flicking the supematants out. Then, 1OOpl of 1% triton Xl00 in 0.05 M Borate buffer, pH 9.0 was added to each well. The plate was kept for 20 hr at 4°C to allow for solubilitation and then was read for fluorescence with a Titertek Fluoroskan II (Flow Laboratories, USA). The percentage of specific lysis was calculated as fallows :

% Specific Lysis = [(Fmed -Fexp)/ Fmed] x 100 where F represent the fluorescence of the solubilized cells after the supematant has been removed; med=F from target incubated in medium alone ; exp= F from target incubated with effector cells. Lytic units (LUZJ /lo7 cells) were calculated by using a computational method (19). One LU corresponded to the number of effector cells required to produce 20% of specific lysis. h?K cell count. To evaluate NK cell number, spleen lymphocytes were stained with rabbit antiA&Ml monoclonal antibody (Wake Co., Dallas, Tex., USA) at a l/40 dilution. FITC-conjugated sheep anti-rabbit IgG (Serotec, Otiord, UK) was used as second reagent. 2 xf06 spleen lymphocytes were labelled with 20 pl of AsGMl MoAb in a final volume of 120 nl of RPMI 1640 with 10% heat inactivated FCS (Gibco) for 30 min. in ice, At the end of the incubation, cells were washed twice in PBS, the sumatant was removed, and 100 ul of the FITC-conjugated sheep antirabbit IgG was added. After 30 min. incubation in ice, cells were washed twice in PBS, resuspended in Isoton II (Coulter, Euro Diagnostics, GMBH) and immediately analysed with a Coulter Epics V flow cytometer. Assay of theproliferatjve reqonse. A volume of 100 ul of the cell suspension (3 x lO”/ml in RPMJ 1640 plus 10% FCS) was delivered in each well of a microculture plate (Nunk, West Germany) Phytohemagglutinin (PHA, Seromed, Biochrom KG) and IL-2 were added in the amount of 10 ul/well at concentrations of 24 &ml and 1000 U/ml respectively. Atter incubation at 37°C in a humidified atmosphere with 5% CO2 for 48 h, f &i of 3H-thymidine (spec. Act. 2000 mCi/mM, Amersham) was added to each well. Cells were collected 20 h later by an automatic cell harvester (Skatron, Norway) and radioactivity was determined using a scintillation counter (TricarbPackard). Each cell culture was performed in triplicate and the proliferative response to each stimulant was expressed as counts/mm per culture.

&Xl

Melatonin Supplementation in Aging

Vol. 61, No. 9, 1997

Staristicul analysis. Results are expressed as mean * SD. Statistical significance was determined by using analysis of variance (ANOVA). When significant differences were found statistical analysis was made by paired Student t-test with the Bonferroni adjustement. Differences were considered statistically significant when p< 0.05.

Results Table I shows the effect of long-term MEL treatment on NK cell activity of spleen lymphocytes from old mice. NK cell cytotoxicity was found to be lower in old mice than in young ones in the four experiments performed with a mean of 2.6 + 0.65 fold reduction. The MEL supplementation for 8 months to old mice did not mod@ their low endogenous levels of NK activity. A small but significant increase of NK cell activity was found only in one of four experiments (Exp 4, p
Exp Exp Exp Exp

1 2 3 4

Young 24.59 f 3.93b 14.10 _+3.50 13.70 t 2.00 19.00 + 3.3 1

TREATMENT

ON NK CELL ACTIVITY

NK cell activity (L&/l Old 14.24 f 1.05” 5.51 + 2.90’ 4.20 + 2.77’ 6.61 * 1.59’

0’) Old MEL-treated 14.45 f 3.54 4.31 + 3.50 4.60 f 1.78 9.70 * 0.50

Mean 8.60 i 5.41 17.93 f 4.72 7.04 k 4.29 ‘MEL treatment was started in 17-18 month-old mice and terminated in 25-26 month-old mice. kesults are the mean + SD. of 3-4 young, 4-5 old and 6-7 old-MEL treated mice examined in each experiment cp
Melatonin Supplementation in Aging

Val. 61, No. 9, 1997

TABLE II EFFECT OF MELATONIN

TREATMENT

ON NK CELL AND THYMOCYTE

NUMBER Spleen lymphocytes

A&Ml+

ceils (%)

(x 1o6 cells) Young 58.80 i 7.39 22.21 +_ 1.30” Old 60.20 f 17.71 25.21 f 7.60 Old MEL-treated 60.62 k 10.87 26.88 + 7.49 ‘Results are the mean + SD. of 3 young. 5 old, and 6 old MEL-treated mice. ‘p
Thymocytes (x ! O6cells) 94 60 c 9.04 3.32 5 0.53

8.90 f 1.40b

60

50

.$ > ‘i: : -

4o

30

5

20

10

0

Young

Oid

Old MEL-treated

Fig. 1 Effect of long-term MEL supplementation on IL2-induced NK cell activity. lymphocytes were incubated overnight at 37°C in 5% COZ at a concentration lO”/ml in RPMI 1640 plus 10% FCS with or without 1000 U/ml IL-2.

Spleen of 3 x

862

Melatoain Supplementation in Aging

TABLE III EFFECT OF MELATONIN TREATMENT PROLIFERATIVE CAPACITY

Vol. 61, No. 9, 1997

ON LYMPHOCYTE

Old MEL-treated Old 5421 f 3077 5165 i- 3885 20227,6157 20481 + 8308 41404 k 14961 37641 -t 8863 Exp 2 6981 f 1337 5532 + 3375 17210+10096 8481 f 3426 50093 + 9675 51275 It 10308 ‘MEL treatment was started in 17- 18 month old mice and terminated in 25-26 month-old mice. kesults are the mean + SD. of 3-4 young, 4-5 old and 6-7 old-MEL treated mice examined in each experiment Exp 1

Culture conditions None PHA IL-2 None PJU n-2

Young” 1887 * 89gb 42293 + 8606 70030 i 7610 5430 + 2300 38870 f 6500 79520 +10522

Discussion Clinical and experimental studies have demonstrated the role of MJX in increasing nmnune responses and in recovering immunofkctioning when depressed by aging or drugs (l-9). MEL administration has been &cm-n to recover age-related thyrnic invokion (I4), reduced thyrnic endocrine activity and T-lymphocyte subsets (15), impaired antibody response, T helper activity and IL-2 production in old animals (16). We recently showed that MEL treatment prevents age-related thymus involution through regulation of thymocyte apoptosis (17). The results reported in this paper demonstrate that long-term MEL supplementation is unable to recover the low levels of NK cell activity and the reduced lymphocyte proliferative response to PHA or IL-2 present in old untreated mice, The effect of in vivo MEL treatment on NK cell activity has been described in previous studies with conflicting results. In humans, chronic MEL administration to healthy subjects augmented NK cytotoxicity and number in one study (2 mg/day for 60 days)(lO), and was not effective in another (10 mg/day for 15-28 days)( 11). Acute MEL administration (1 or 100 mg given orally) did not modi$ spontaneous NK activity (10). In mice, the reduction of NK cell activity determined by pinealectomy, was prevented by acute (So-100 mgKg through single injectlon) but not chronic (1 O-50 mg/Kg for 9 consecutive days) MEL treatment (9) Our study reports, for the first time, the effect of long-term MEL supplementation showing the ineffectiveness of MEL in recovering low levels of endogenous NK cell activity and reduced NK cell activation by IL-2 in old mice The ineffectiveness of our MEL treatment in increasmg NK activity does not seem to be related to the MEL concentration used since MEL supplementations either lo-fold lower (I 1) or IO-50-fold higher (9) than our MEL dose were also ineffective in modulating NK cell activity Only in one study chronic MEL supplementation was successful in increasing NK activity (10) In this study MEL was administered at very low doses (2 mg/day) for 60 days in six heahhy subjects UnfortunateIy, the values of NK activity were reported as the mean of three individual determinations obtained at days 20, 40 and 60 from MEL treatment and it is not possible to analyse the effective levels of NK activity present at day 60 In our study we performed a h4EL supplementation for eight months, i.e., for a longer length of time than those used in the above reports The effectiveness of this protocol of MEL treatment was previously demonstrated in a paper describing the partial recovery of thymic involution and the modulation of thymocyte

Vol. 61, No. 9, 1997

Melatonio Supplementation in Aging

863

apoptosis in mice supplemented with MEL for eight months (17). The recovery of thymus cellularity was also checked and observed in the present study, further emphasising the relevant role exerted by MEL on the thymus gland rather than on peripheral immune e&ctors. On the whole, MEL supplementation does not seem able to modulate the levels of NK activity and even short term experiments, in which acute MEL administration of 1 or 100 mg MEL (10) or 30 mg/Kg MEL (9) was performed, were ineffective. Only extremely high MEL doses modulated NK cytotoxicity after acute MEL administration (9). We also demonstrated the lack of effects of MEL treatment on the proliferative response of spleen lymphocytes to PHA or IL-2. In a recent paper, Mocchegiani et al. (15) reported the increase of spleen cell proliferative response in mice chronically supplemented with MEL. However, some differences are present in this paper in comparison with our experimental design such as the length of treatment (4 months rather than 8 months), the age at which the animals were sacrificed (22 months rather than 25-26 months), and the population of spleen cells used (unpurified spleen cells rather than spleen purified lymphocytes). The ineffectiveness of MEL in modulating NK cell activity and lymphocyte proliferation, in the presence of an increased thymus cellularity, clearly demonstrates the efficacy of MEL supplementation performed in our study. The causes responsible for the ineffectiveness of MEL treatment in recovering NK cell activity and lymphocyte proliferation remain to be investigated. Certainly, they are not related to intrinsic and irreversible alterations of these immune parameters, since other kind of neuroendocrine approaches have been shown to recover them (19,20). The lack of MEL receptors on immune cells does not seem to be involved, since the existence of MEL bindiig sites has been demonstrated in human blood lymphocytes and rat thymocytes (21,22). Also the occurrence of tolerance or downregulation of MEL receptors is unliiely since in this case we should have found no effects of MEL, whereas in our study MEL was able to increase thymus weight and cellularity. Furthermore, the presence of MEL receptors on immune cells does not represent a necessary requirement for MEL action. In fact, we (17) and others (15) have demonstrated that in vivo MEL effect on immune system may be mediated by regulation of zinc and/or glucocorticoid turnover and that in vitro MEL does not affect immune functions. The involvement of indirect mechanisms in the action of MEL on immune system might explain the different effect of the pineal hormone on thymus or peripheral immune efficiency. Another possible explanation may be related to the crucial action exerted by MEL on specific rather than non specific immune effecters, as demonstrated by the effect of MEL supplementation on the thymus gland, as we (17) and others (14,15) have shown and on B and T cell tunctions (6,16,20,23). Recent evidence emphasising the role of T-derived cytokines (both Thl and Th2) as the main mediators of the immunological effect of MEL clearly support this possibility (23,24).

Acknowledgements

This work was supported in part by CNR Progetto Finalizzato Invecchiamento and by the Italian Health Ministry target project “Applicazioni neuroimmunologiche nella malattia neoplastica dell’anziano”. We thank Ms. B. Bartozzi for her technical assistance and Mr. G. Bernardini for performing the flow cytometric analysis.

864

Melatoain Supplementation

in Aging

Vol. 61, No. 9, 1997

References

1. W. PIERPAOLI, and G.J.M. MAESTRONT, Immunol. L&t. 16 355-362 (1987) 2. J.M. GUERRER 0, and R.J. REITER, Endocrinol. Res. 18 91-113 (1992). 3. G.J.M MAESTRONI, and A CONTI, Melatonin: Biosvnthesis. Phvsiolonical Effects. and Clinical Aunlications. Y. Hing-Sing and R.J. Reiter (Eds.), 290-309, CRC Press Boca Raton Ann Arbor London Tokyo ( 1993). 4. C. CSABA, and P. BARATH, Endocrinol. Exp. 9 59-67 (1975). 5. G.J.M. MAESTRONI, A. CONTI, and W. PIERPAOLI, Clin. Exp. Immunol. 68 384-391 (1987) 6. M.C. CAROLEO, D FRASCA, G. NISTIC@ and G DORIA, Immunopharmacology 23 81-89 (1992) 7. C. PIOLI, M.C CAROLEO, G. NISTIC@ and G DORIA, Int J. Immunopharmac 15 463-468 (1993). 8. B.D. JANKOVIC, K. ISAKOVIC, and S. PETROVIC, Immunology 8 l-5 (1970) 9 V DEL GOBBO, V. LIBRI, N. VILLANI, R. CALId, and G. NISTICi), lnt J Immunopharmacol 11 567-569 (1989). 10.A ANGELI, G GATT& M L SARTORI, D DEL PONTE, and R.CERlGNOLA, The uineal piand and cancer D. Gupta, A Attanasio and R J Reiter (Eds), 145-156, Muller and Bass, Tubingen ( 1988). 1l.G.J M MAESTROM and A CONTI, Psvchoneuroimmunolog II, R Ader, D L. Felten, and N Cohen (Eds), 495-5 15. Academic Press, San Diego (199 1). 12.1. IGUCHI, K I KATO, and H. IBAYASHI, J. Clin. Endocrinol. Metab. 55 27-29 (1982) 13.R.J. REITER, C M. CRAFT, J E. JR JOHNSON, T S.,KING, B.A. RICHARDSON. G M. VAUGHAN, and M.K. VAUGHAN, Endocrinology 109 1295-1297 (1980). 14.W PIERPAOLI, and W. REGELSON, Proc. Nat1 Acad. Sci. USA 91 787-791 (1994) 15.E. MOCCHEGIANI, D. BULIAN, L. SANTARELLI, A TIBALDI, M. MUZZIOLI, W PIERPAOLI, and N. FABRIS, J. Neuroimmunol. 53 189-201 (1994). 16.M.C. CAROLEO, G. DORIA, and G. MSTICd, Ann. N.Y. Acad. Sci. 719 343-352 (1994). 17.M. PROVINCIALI, G. DI STEFANO, D BULIAN, A. TIBALDI, and N. FABRIS, Mech. Ageing Develop. 90 1- 19 (1996) 18.M. PROVINCIALI, G. DI STEFANO, and N FABRIS, J. Immunol. Meth. 155 19-24 (1992) i9.N. FABRIS and M. PROVINCIALI, Natural Immunity, D. Nelson (Ed), 306-347, Academic Press, Australia (1989). 20.M. PROVINCIALI, M. MUZZIOLI, G. DI STEFANO, and N. FABIUS, Nat. Immun Cell Growth Regul. 10 226-236 (1991) 21.M.A. LOPEZ-GONZALES, L.R CALVO, C OSUNA and J.M. GUERRERO, J Pineal Res 12 97-104 (1992) 22.A MARTIN-CACAO, M.A LOPEZ-GONZALES, J.R. CALVO, J.J SEGURAAND, and J.M GUERRERO, Melatonin and the nineal aland. Y Touitou, J Arendt, and P Pevet (Eds ), 7982, Elsevier, Amsterdam, (1993). 23 G.J.M MAESTRONI and A CONTI, J. Neuroimmunol. 28 167-l 76 (1990). 24.G.J.M h4AESTROM, J Pineal Res 18 84-89 (1995).