0143-4179/90/0016-0025/$10.00
Neuropeptides (1990) 16,25-32 0 Longman Group UK Ltd 1990
Neuropeptide Regulation of Ceil-Mediated Cytotoxicity Against Human Tumor Cells E. A. F. van TOL, H. W. VERSPAGET, Department
of Gastroenterology
C. B. H. W. LAMERS
and Hepatology,
University Hospital, Leiden, The Netherlands
Abstract-The effects of the brain-gastrointestinal polypeptide neurotransmitters bombesin, substance P, neurotensin, and somatostatin-14 on cytotoxicity of peripheral blood mononuclear cells against K-562 and CaCo-2 tumour cells were investigated. Bombesin significantly stimulated cytotoxicity against CaCo-2 target cells (lo-‘*, lo-‘OM and 10m6M) and against K-562 target cells (lo-‘* and lo-“M) in the short 4 hour assay. Substance P showed a tendency to stimulate cytotoxicity at higher concentrations but the changes observed did not reach significance because of large inter-individual variation of responsiveness. Neurotensin did not influence cytotoxicity against either target cell lines. Somatostatin was found to have no influence on cytotoxicity of peripheral blood mononuclear cells but was the only peptide tested which markedly increased chromium release by target cells alone. These findings support the idea that brain-gastrointestinal neuropeptides can play a part in tumour cytotoxicity.
Introduction Immune processes are complicated networks in which interaction of lymphoid and non-lymphoid cells with the nervous and endocrine systems provide adequate modulation of immune responsiveness (l-4). The immune microenvironment has been shown to be composed of lymphocytes, non-lymphoid accessory cells, hormones provided
Date received 18 September 1989 Date accepted 30 October 1989 Correspondence: Dr H.W. Verspaget, M.Sc., Ph.D., Dept. Gastroenterology and Hepatology, University Hospital, Building 1, C4-P012, Rijnsburgerweg 10, 2333, AA, Leiden, The Netherlands.
by capillary systems and innervating neurons, as well as lymphokines, monokines and other cytokines. There is increasing evidence that bidirectional communication pathways exist between the immune and neuroendocrine systems examplified by immunoregulatory circuits between hypothalamus, pituitary gland, adrenals and immunocompetent cells (5). Lymphokines produced by monocytes, for instance, have been shown to stimulate the pituitary gland resulting in an increased ACTH release. This hormone on its turn increases corticosteroid secretion by the adrenal cortex which then resulted in stimulation of peripheral blood mononuclear cells. Moreover, specific receptors for neuropeptides and hormone have been characterized on immunocompetent cells (6).
26 Currently the interplay between hormones or neuropeptides and the immune response in the gastrointestinal tract is an area of major interest (7,s). Because of the abundant peptidergic innervation of the gastrointestinal tract, neuropeptides may therefore be involved in several local immune functions. Substance P, the most extensively studied peptide, has been shown to enhance IgA and IgM synthesis by ConA stimulated murine lymphocytes (9) and to enhance proliferation of human and murine lymphocytes (10, 11). This effect on immunoglobulin synthesis was found to be isotype specific and depended largely on the origin of the B-cells. Moreover, substance P has been shown to stimulate macrophage and neutrophil phagocytosis (12) to cause histamin release from both peritoneal and mucosal mast cells (13), to induce lymphokine production (14)) and macrophage tumoricidal activity (15). Bombesin was found to have weak immunostimulatory effects, as determined by mitogen induced lymphocyte proliferation studies, mixed lymphocyte cultures (lo), and macrophage function analysis (15). Up till now there have been no reports on a regulatory role of neurotensin in immune responses. Somatostatin, which acts both as neurotransmitter and as paracrine and hormonal substance in the digestive tract, has been reported to modulate several immune functions, e.g. lymphocyte proliferation (lo), immunoglobulin synthesis (9), and secretory activity of mast cells and basophils (13). Vasoactive intestinal peptide (16, 17), and to a lesser extent somatostatin (18), are the only peptides from the gastrointestinal tract that were found to modulate spontaneous cell mediated cytotoxicity against tumour cells in vitro. In the present study a group of neurotransmitters from the digestive tract, i.e. bombesin, substance P, neurotensin, as well as somatostatin were studied for their effects on the spontaneous cell mediated cytotoxicity against human tumour cells in vitro. Materials and methods Culture media Human tumour cell lines K-562 and CaCo-2 were cultured in RPM1 1640 medium with 20mM
NEUROPEPTIDES
Hepes-buffer (Gibco; Paisley, Scotland) supplemented with fetal calf serum (FCS: Gibco), 10% for K-562 and 20% for CaCo-2, penicillin (lOOU/ ml; Gibco), streptomycin (100 &ml; Gibco), gentamycin (50 pg/ml; Flow Laboratories; Irvine, Scotland), amphotericin B (2.5 kg/ml; Gibco) and glutamine (2 mM; Flow Laboratories). Effector cells Peripheral blood mononuclear cells (PBMC) were obtained by isolation from heparinized venous blood from healthy volunteers. After densitygradient centrifugation on Ficoll-Isopaque (l.O79g/ml) at room temperature, the cells were washed three times in culture medium with 10% FCS. Viability of the cells, as determined by dye-exclusion tests with Trypan-Blue, was consistently above 98%. The effector cells were brought to a final concentration of 2.5 x lo6 cells/ml. Target cells The human erythroleukemia cell line K-562 and the human colon carcinoma cell line CaCo-2 were used as targets in the cytotoxicity assay. The cells were cultured in 25 cm* tissue-culture flasks in the respective culture media at 37°C in a 5% CO2/95% air incubator. They were kept in growth phase by transferring 1:3 twice a week. Cytotoxicityassay Target cells were harvested and labelled with sodium-51chromate (51Cr, lOOl.~Ci/5 x lo6 cells) for 1 hour at 37°C while agitating to improve labelling efficiency. After washing, the cells were resuspended in the appropriate concentration of 1 x 16 cells/ml. Viable target cells were quantitated by Trypan-Blue exclusion. The cytotoxicity assay was performed as published previously (19) in a 5O:l effector to target (E:T) ratio for all incubations. The cells were supplemented with 50~1 medium or the same volume of the peptide solution. Target cells alone were incubated in medium to measure spontaneous release of 51Cr or with the detergent saponine to cause maximal lysis and chromium release during the incubation period. The assay was performed in triplicate in a final volume of 200@well. After the 4 and 18 hour incubation period the plates were centrifuged and
27
NEUROPEPTIDES AND CYTOTOXICITY
supernatants were collected with a Skatron harvesting system. The released 51Cr label was counted in a gamma-counter (Compugamma; LKB, Finland). Cellular cytotoxicity was calculated by the following formula: cytotoxicity (%) = experimental release - spontaneous release
x
loo
maximal release - spontaneous release
Peptide incubations
For all experiments highly purified synthetic peptide preparations were used; bombesin (Sigma, St. Louis, USA), substance P (Boehringer Mannheim Biochemicals, Mannheim, FRG), neurotensin (Novabiochem, Ltiufelfingen, Switzerland) and somatostatin-14 (UCB Bioproducts, Bruxelles, Belgium). The lyophilized neuropeptides were freshly dissolved in culture medium without additives. Cytotoxicity assays were performed in the presence of lo-‘*, lo-“, 10T8 and 10m6M bombesin, substance P or somatostatin, or 10-16, 10-14, lo-‘*, lOlo, 10p8, 10e6M neurotensin. In each assay, the influence of peptides on spontaneous release of “Cr label from target cells after both incubation periods was evaluated. Statistical analysis
All data are expressed as mean + standard error of the mean (SEM). The statistical significances of the differences were evaluated with Friedman’s analysis of variance for repeated measurements and the paired non-parametrical two tailed Wilcoxon’s rank-sum test.
Results The potency of neuropeptides in modulating the cellular cytotoxic response is best illustrated by expressing the effect as mean relative change of cytotoxicity. Bombesin significantly stimulated cytotoxicity against both CaCo-2 and K-562 target cells only in the short 4 hour assay (Fig. 1). The stimulation of cytotoxicity against K-562 cells occurred preferentially at low concentrations whereas against CaCo-2 cells a more gradual
increase of stimulation to the higher concentrations was found. This transient stimulatory effect of bombesin on cytotoxicity of PBMC in the short 4 hour assay resulted in no alteration of cytotoxicity in the extended 18 hour assay against both cell lines. In contrast, substance P incubations revealed differences in the modulation of cytotoxic activity against both cell lines (Fig. 2). In the 4 hour assay against CaCo-2 target cells a tendency to dose dependent stimulation was observed reaching its maximum at 10e8M, whereas in the 18 hour assay a marginal inhibition was found. Substance P did not change the cytotoxic activity of PBMC against K-562 cells in both the 4 and 18 hour assay. Because of the large variations of responsiveness between the individuals tested with substance P, the increase in cytotoxicity did not reach significance although these changes were in the order of magnitude as those found in the bombesin experiments. Again with substance P, like with bombesin, we found no influence on cytotoxicity in the prolonged 18 hour assay against both K-562 or CaCo-2 target cells. The influences of neurotensin on cytotoxicity could be described as a non-significant tendency to enhancement. Only a marginal increase of cytotoxicity was found after 18 hour of incubation against K-562 target cells (Fig. 3). Somatostatin did not alter the cytotoxic response of PBMC against both K-562 or CaCo-2 target cells. Moreover, this was the only peptide in this study that showed a dose dependent enhancement of chromium release when incubated with target cells alone as control experiment. When these target effects of somatostatin were taken into account, inhibition of cytotoxicity did not reach significance even at a concentration of 10m8 or 10e6M against CaCo-2 targets (Fig. 4).
Discussion The results presented in this study indicate that neuropeptides are directly involved in the modulation of cellular cytoxicity in vitro. Among the peptides that were investigated, bombesin was the most potent stimulator of cytotoxicity especially against CaCo-2 colon carcinoma cells. Bombesin
28
NEUROPEI’TIDES
Increoee in cytotoxicity
X
50
+CaCo-2
lml
-CcoCo-2 4h -- K-662 16h .. K-662 4h
lOE-12
lM-10
lOE-8
NE-6
Concentration BBS(M)
Fig. 1 Effect of bombesin incubation on spontaneous cell mediated cytotoxicity of PBMC against CaCo-2 or K-562 target cells expressed as relative percentual change. Significant differences of bombesin incubations from those of the control (0) experiments are denoted by asterisks (n = 12, (*) = p < 0.06, * = p c 0.05, ** < 0.02, *** = p < 0.01). The mean percentual variation of the individual specific cytotoxicity is indicated by the shaded area.
showed a gradual increase of the cytotoxic response with the concentration added. The observed stimulation of cytotoxicity probably reflects the presence of high affinity receptors on either effector or both effector and target cells. The latter might be explained by the differences observed between K-562 and CaCo-2 cells with respect to their susceptibility to lysis. Bombesin levels in blood are hardly detectable, which might explain the sensitivity of our effector cells to very low concentrations in vitro, mimicking high plasma levels. Notably, the CaCo-2 cells were shown to be more sensitive to bombesin with the higher concentrations. Lymphoid cells in the lamina propria and Peyer’s patches may become exposed to the higher concentrations of the neu-
ropeptide that were tested, when they are in the close vicinity of nerve endings found to be present in the gut wall (20, 21). Functionally relevant peptide receptors, which have been characterized on immunocompetent cells, might reflect the role of both nervous and endocrine systems in regulating the immune response. Although these receptors on lymphoid cells have been characterized for substance P, VIP, somatostatin, P-endorphins and enkephalins by binding studies (6), this has not yet been reported for bombesin. Substance P is relatively potent for stimulation of lymphocyte proliferation, immunoglobulin synthesis, neutrophil activity and mediator release from mast cells (9, 11, 13). It plays a role, not only
29
NEUROPEPTIDES AND CYTOTOXICITY
in these inflammation and allergic events but also in local gut immunity. Because of the connection of substance P containing nerves and gut mucosal tissue, it can locally interact with immunocompetent cells in mucosal tissue or gut associated lymphoid tissue (20, 21). Surprisingly, only cytotoxicity against CaCo-2 target cells was substantially increased at the higher peptide concentrations. The almost similar distribution of bombesin and substance P containing nerves in the gut could be related to the similar stimulatory effects that were observed when using these peptides in vitro against CaCo-2 colon tumour cells. Moreover, receptor binding studies indicated that both the bombesin and substance P receptor share specificity for the binding of both peptides (22). Indeed there are indications for the presence of
specific substance P receptors on immunocompetent cells in gut tissue, e.g. on T- and B-lymphocytes from Peyer’s patches (23). Somatostatin by itself had no effect on cell mediated cytotoxicity when present during the assay, although stimulation of cytotoxicity has been reported when effector cells were preincubated with somatostatin (18). Still somatostatin might interfere with the action of other peptides, since it is well known for its antagonistic effects on gastrointestinal hormone function as well as on exocrine secretion (24). Somatostatin did stimulate the chromium release by target cells in a dose dependent fashion. The underlying mechanism for this phenomenon is unclear but it could be a direct toxic effect of somatostatin on the cells. Moreover, binding of somatostatin to the tumour cells might
X Increare in cytotoxicity SO-
40--
30-+ coca-2 M--
1Bh
- Coca-2 4h -- K-562 1Bh . . K-562 4h
lOE-10 Concentration
lOE-6
lM-8 Subrrtoncr
P(M)
Fig. 2 Influence of substance P incubation on spontaneous cell mediated cytotoxicity of PBMC against CaCo-2 or K-562 target cells expressed as relative percentual change (n = 12). The mean percentual variation of the individual specific cytotoxicity is indicated by the shaded area.
30
NEUROPEF’TIDES
60
0
4h
60
q 16h
0
IN 1OE-161CE-141oE-121E10 1M-6 lOE-6 ConcaltrotimNmrotmsin M)
Cl
IN IOE-161E-14lOE-12loE-10 N-E-8ME-6 Concmtr.tlm NwrotPMInW
Fig. 3 Spontaneous cell mediated cytotoxicity of PBMC after incubation with neurotensin against CaCo-2 (a) and K-562 (b) target cells (n = 12).
interfere with the membrane architecture resulting in increased permeability of the cells. Neurotensin did not influence cytotoxicity of PBMC against the tumour target cells in our assay, even though a broader concentration range of the peptide has been tested. Thus bombesin, and possibly substance P, can modulate spontaneous cell mediated cytotoxity against tumour cells. Although the precise mechanism of stimulation of cytotoxicity is not yet unravelled we found that the effector cells influenced by these peptides are undoubtedly the NK-cells. The alterations of cytotoxicity are only seen in the 4 hour assay where cytotoxicity was found to be mediated for over 95% by CD16+ NK-cells as demonstrated by selective depletion of these cells (unpublished data). The observed effects are rapid, i.e. only in the short assay and at low concentrations. What the in vivo importance of such a rapid transient stimulation may be, remains unclear. In this way for instance neuropeptides may rapidly intervene with cellular cytotoxic effector mechanisms. However, it cannot be excluded that these peptides play a regulatory role by stimulating accessory cells to produce certain cytokines which ultimately results in stimulation of cytotoxic effector cells.
In the present studies large inter-individual variations of responsiveness were observed. Therefore we would like to emphasize that the modulatory role for these neuropeptides in the immune response should not be _generalized.
talk-2 cl
conomtrotionSST(M) Fig. 4 Effect of different concentrations somatostatin on the spontaneous cell mediated cytotoxicity of PBMC against CaCo-2 target cells after 18h incubation and on the spontaneous release of chromium label by CaCo-2 target cells (n = 9).
NEUROPEPTIDES
AND CYTOTOXICITY
Different factors may influence the basal state of the mononuclear cells which were isolated. For instance hormone and receptor levels may interfere with NK-cell activity and therefore change their susceptibility for modulation. In conclusion, the results in this study suggest a role for neuropeptides of the gastrointestinal tract in modulating the cellular immune response. This may especially be of importance in gastrointestinal pathology where impaired NK-cell activity has been reported in relation to several diseases (25, 26).
Acknowledgements We are indebted to Marij Mieremet-Ooms and Annie van der Zon for their careful technical assistance and Susan Vervoordeldonk and Ciske Widijanto for their participation in the project. We also thank Loes Niepoth and Nelia Koek for preparing the manuscript and Rent? Kuijppers for his contribution to the designing of the figures.
References 1. Payan,
2.
3.
4.
5.
6.
D. G., McGillis, J. P., Goetzl, E. J. (1986). Neuroimmunology. Advances in Immunology 39: 299323. Bloom, F. E. (1985). Neuropeptides and other mediators in the central nervous system. Journal of Immunology 135: 743-745. Besedovsky, H. O., delRey, A. E., Sorkin, E. (1985). Immune-neuroendocrine interactions. Journal of Immunology 135: 750-754. Blalock, J. E., Harbour-McMenamin, D., Smith, E. M. (1985). Peptide hormones shared by the neuroendocrine and immunologic systems. Journal of Immunology 135: 858-861. McEwen, B. (1987). Influences of hormones and neuroactive substances on immune function. In: The neuroimmune-endocrine connection. Cotman C. W. (ed.) Raven Press, New York p. 71-92. Bost, K. L. (1988). Hormone and neuropeptide receptors on mononuclear leukocytes. In: Neuroimmunoendocrinology. Blalock J. E.,Bost, K. L. (eds). Progress in allergy. Karger, Basel. p. 68-83. Bienenstock, J., Perdue, M., Stanisz, A., Stead, R. (1987). Neurohormonal regulation of gastrointestinal immunity. Gastroenterology 93: 1431-1434. O’Dorisio, M. S. (1986). Neuropeptides and gastrointestinal immunity. American Journal of Medicine 81: 74-82. Stan&, A. M., Befus, D., Bienenstock, J. (1986). Differential effects of vasoactive intestinal peptide, substance P, and somatostatin on immunoglobulin synthesis and proliferations by lymphocytes from peyer’s patches, mesen-
31 teric lymph nodes, and spleen. Journal of Immunology 136: 152-156. 10. Krco, C. J., Gores, A., Go, V. L. W. (1986). Gastrointestinal regulatory peptides modulate mouse lymphocyte functions under serum-free conditions in vitro. Immunological Investigations 15: 103-l 11.
II. Payan, D. G., Brewster, D. R., Goetzl, E. J. (1983). Specific stimulation of human T lymphocytes by Substance P. Journal of Immunology 131: 1613-1615.
r2. Bar-Shavit, Z., Goldman, R., Stabinsky, Y., Gottlieb, P.,
Fridhin, M., Teichberg, V. I., Blumberg, S. (1980). Enhancement of phagocytosis - a newly found activity of substance P residing in its N-terminal tetrapeptide sequence. Biochemical and Biophysical Research Communications 94: 1445. Shanahan, F.. Denburg, J. A., Fox, J. Bienenstock, J., Befus, D. (1985). Mast cell heterogeneity: effects of neuroenteric peptides on histamine release. Journal of Immunology 135: 1331-1337.
14. Lotz, M., Vaughan, J. H., Carson, D. A. (1988) Effect of neuropeptides on production of inflammatory cytokines by human monocytes. Science 241: 1218-1221. 15. Peck, R. (1987). Neuropeptides modulating macrophage function. In: Neuroimmune interactions. Proceedings of the second international workshop on neuroimmunomodulation. Jankovic, B. D., Markovic, B. M., Spector, N. H. (eds). Annals of the New York Academy of Sciences 496: 264-270. 16 Rola-Pleszczynski, M., Bolduc, D., St-Pierre, S. (1985). The effects of vasoactive intestinal peptide on human natural killer cell function. Journal of Immunology 135: 2569-2573. 17. Drew, P. A., Shearman, D. J. C. (1985). Vaso-active intestinal peptide: a neurotransmitter which reduces human NK cell activity and increases Ig synthesis. Australian Journal of Experimental Biology and Medical Science 63: 313-318. 18, Pawhkowski, M., Zelazowski, P., Stepien, H. (1989). Enhancement of human lymphocyte natural killer activity by somatostatin. Neuropeptides 13: 75-77. 19. Aparicio-Pages, M. N., Verspaget, H. W., Pena, A. S., Weterman, I. T., Mieremet-Ooms, M. A. C., van der Zen, J. M., de Bruin, P. A. F., Lamers, C. B. H. W. (1988) Natural, lectin- and phorbol ester-induced cellular cytotoxicity in Crohn’s disease and ulcerative colitis. Journal of Clinical Laboratory Immunology 27: 109-113. 20. Ferri, G-L., Adrian, T. E., Ghatei, M. A., O’Shaughnessy, D. J., Probert, L., Lee,Y. C.,Buchan, A. M. J., Polak, J. M., Bloom, S. R. (1983). Tissue locahzation and relative distribution of regulatory peptides in separated layers from the human bowel. Gastroenterology 84: 777-786. 21. Lluis, F., Thompson, J.C. (1988). Neuroendocrine potential of the colon and rectum. Gastroenterology 94: 832-844. 22. Jensen, R. T.. Heinz-Erian, P., Mantey, S., Jones, S. W., Gardner, J. D. (1988). Characterization of ability of
32 various substance P antagonists to inhibit action of bombesin. American Journal of Physiology 254: G883-890. 23. Stat&, A. M., Scicchitano, R., Dazin, P., Bienenstock, J., Payan, D. G. (1987). Distribution of Substance P receptors on murine spleen and Peyer’s Patch T and B cells. Journal of Immunology 139: 749-754. 24. Reichlin, S. (1983). Somatostatin. New England Journal of Medicine 309: 1556-1563.
NEUROPEPTIDES
25. Ginsburg, C. H., Dambrauskas, J. T., Ault, K. A., Falchuk, Z. M. (1983). Impaired natural killer cell activity in patients with inflammatory bowel disease: evidence for a qualitative defect. Gastroenterology 85: 846-851. 26. Gibson, P. R., Jewell, D. J. (1986). Local immune mechanisms in inflammatory bowel disease and colorectal carcinoma. Natural killer cells and their activity. Gastroenterology 90: 12-19.
Neuropeptides (1990) 16,25-32 0 Longman Group UK Ltd 1990
Neuropeptide Regulation of Ceil-Mediated Cytotoxicity Against Human Tumor Cells E. A. F. van TOL, H. W. VERSPAGET, Department
of Gastroenterology
C. B. H. W. LAMERS
and Hepatology,
University Hospital, Leiden, The Netherlands
Abstract-The effects of the brain-gastrointestinal polypeptide neurotransmitters bombesin, substance P, neurotensin, and somatostatin-14 on cytotoxicity of peripheral blood mononuclear cells against K-562 and CaCo-2 tumour cells were investigated. Bombesin significantly stimulated cytotoxicity against CaCo-2 target cells (lo-‘*, lo-‘OM and 10m6M) and against K-562 target cells (lo-‘* and lo-“M) in the short 4 hour assay. Substance P showed a tendency to stimulate cytotoxicity at higher concentrations but the changes observed did not reach significance because of large inter-individual variation of responsiveness. Neurotensin did not influence cytotoxicity against either target cell lines. Somatostatin was found to have no influence on cytotoxicity of peripheral blood mononuclear cells but was the only peptide tested which markedly increased chromium release by target cells alone. These findings support the idea that brain-gastrointestinal neuropeptides can play a part in tumour cytotoxicity.
Introduction Immune processes are complicated networks in which interaction of lymphoid and non-lymphoid cells with the nervous and endocrine systems provide adequate modulation of immune responsiveness (l-4). The immune microenvironment has been shown to be composed of lymphocytes, non-lymphoid accessory cells, hormones provided
Date received 18 September 1989 Date accepted 30 October 1989 Correspondence: Dr H.W. Verspaget, M.Sc., Ph.D., Dept. Gastroenterology and Hepatology, University Hospital, Building 1, C4-P012, Rijnsburgerweg 10, 2333, AA, Leiden, The Netherlands.
by capillary systems and innervating neurons, as well as lymphokines, monokines and other cytokines. There is increasing evidence that bidirectional communication pathways exist between the immune and neuroendocrine systems examplified by immunoregulatory circuits between hypothalamus, pituitary gland, adrenals and immunocompetent cells (5). Lymphokines produced by monocytes, for instance, have been shown to stimulate the pituitary gland resulting in an increased ACTH release. This hormone on its turn increases corticosteroid secretion by the adrenal cortex which then resulted in stimulation of peripheral blood mononuclear cells. Moreover, specific receptors for neuropeptides and hormone have been characterized on immunocompetent cells (6).
26 Currently the interplay between hormones or neuropeptides and the immune response in the gastrointestinal tract is an area of major interest (7,s). Because of the abundant peptidergic innervation of the gastrointestinal tract, neuropeptides may therefore be involved in several local immune functions. Substance P, the most extensively studied peptide, has been shown to enhance IgA and IgM synthesis by ConA stimulated murine lymphocytes (9) and to enhance proliferation of human and murine lymphocytes (10, 11). This effect on immunoglobulin synthesis was found to be isotype specific and depended largely on the origin of the B-cells. Moreover, substance P has been shown to stimulate macrophage and neutrophil phagocytosis (12) to cause histamin release from both peritoneal and mucosal mast cells (13), to induce lymphokine production (14)) and macrophage tumoricidal activity (15). Bombesin was found to have weak immunostimulatory effects, as determined by mitogen induced lymphocyte proliferation studies, mixed lymphocyte cultures (lo), and macrophage function analysis (15). Up till now there have been no reports on a regulatory role of neurotensin in immune responses. Somatostatin, which acts both as neurotransmitter and as paracrine and hormonal substance in the digestive tract, has been reported to modulate several immune functions, e.g. lymphocyte proliferation (lo), immunoglobulin synthesis (9), and secretory activity of mast cells and basophils (13). Vasoactive intestinal peptide (16, 17), and to a lesser extent somatostatin (18), are the only peptides from the gastrointestinal tract that were found to modulate spontaneous cell mediated cytotoxicity against tumour cells in vitro. In the present study a group of neurotransmitters from the digestive tract, i.e. bombesin, substance P, neurotensin, as well as somatostatin were studied for their effects on the spontaneous cell mediated cytotoxicity against human tumour cells in vitro. Materials and methods Culture media Human tumour cell lines K-562 and CaCo-2 were cultured in RPM1 1640 medium with 20mM
NEUROPEPTIDES
Hepes-buffer (Gibco; Paisley, Scotland) supplemented with fetal calf serum (FCS: Gibco), 10% for K-562 and 20% for CaCo-2, penicillin (lOOU/ ml; Gibco), streptomycin (100 &ml; Gibco), gentamycin (50 pg/ml; Flow Laboratories; Irvine, Scotland), amphotericin B (2.5 kg/ml; Gibco) and glutamine (2 mM; Flow Laboratories). Effector cells Peripheral blood mononuclear cells (PBMC) were obtained by isolation from heparinized venous blood from healthy volunteers. After densitygradient centrifugation on Ficoll-Isopaque (l.O79g/ml) at room temperature, the cells were washed three times in culture medium with 10% FCS. Viability of the cells, as determined by dye-exclusion tests with Trypan-Blue, was consistently above 98%. The effector cells were brought to a final concentration of 2.5 x lo6 cells/ml. Target cells The human erythroleukemia cell line K-562 and the human colon carcinoma cell line CaCo-2 were used as targets in the cytotoxicity assay. The cells were cultured in 25 cm* tissue-culture flasks in the respective culture media at 37°C in a 5% CO2/95% air incubator. They were kept in growth phase by transferring 1:3 twice a week. Cytotoxicityassay Target cells were harvested and labelled with sodium-51chromate (51Cr, lOOl.~Ci/5 x lo6 cells) for 1 hour at 37°C while agitating to improve labelling efficiency. After washing, the cells were resuspended in the appropriate concentration of 1 x 16 cells/ml. Viable target cells were quantitated by Trypan-Blue exclusion. The cytotoxicity assay was performed as published previously (19) in a 5O:l effector to target (E:T) ratio for all incubations. The cells were supplemented with 50~1 medium or the same volume of the peptide solution. Target cells alone were incubated in medium to measure spontaneous release of 51Cr or with the detergent saponine to cause maximal lysis and chromium release during the incubation period. The assay was performed in triplicate in a final volume of 200@well. After the 4 and 18 hour incubation period the plates were centrifuged and
27
NEUROPEPTIDES AND CYTOTOXICITY
supernatants were collected with a Skatron harvesting system. The released 51Cr label was counted in a gamma-counter (Compugamma; LKB, Finland). Cellular cytotoxicity was calculated by the following formula: cytotoxicity (%) = experimental release - spontaneous release
x
loo
maximal release - spontaneous release
Peptide incubations
For all experiments highly purified synthetic peptide preparations were used; bombesin (Sigma, St. Louis, USA), substance P (Boehringer Mannheim Biochemicals, Mannheim, FRG), neurotensin (Novabiochem, Ltiufelfingen, Switzerland) and somatostatin-14 (UCB Bioproducts, Bruxelles, Belgium). The lyophilized neuropeptides were freshly dissolved in culture medium without additives. Cytotoxicity assays were performed in the presence of lo-‘*, lo-“, 10T8 and 10m6M bombesin, substance P or somatostatin, or 10-16, 10-14, lo-‘*, lOlo, 10p8, 10e6M neurotensin. In each assay, the influence of peptides on spontaneous release of “Cr label from target cells after both incubation periods was evaluated. Statistical analysis
All data are expressed as mean + standard error of the mean (SEM). The statistical significances of the differences were evaluated with Friedman’s analysis of variance for repeated measurements and the paired non-parametrical two tailed Wilcoxon’s rank-sum test.
Results The potency of neuropeptides in modulating the cellular cytotoxic response is best illustrated by expressing the effect as mean relative change of cytotoxicity. Bombesin significantly stimulated cytotoxicity against both CaCo-2 and K-562 target cells only in the short 4 hour assay (Fig. 1). The stimulation of cytotoxicity against K-562 cells occurred preferentially at low concentrations whereas against CaCo-2 cells a more gradual
increase of stimulation to the higher concentrations was found. This transient stimulatory effect of bombesin on cytotoxicity of PBMC in the short 4 hour assay resulted in no alteration of cytotoxicity in the extended 18 hour assay against both cell lines. In contrast, substance P incubations revealed differences in the modulation of cytotoxic activity against both cell lines (Fig. 2). In the 4 hour assay against CaCo-2 target cells a tendency to dose dependent stimulation was observed reaching its maximum at 10e8M, whereas in the 18 hour assay a marginal inhibition was found. Substance P did not change the cytotoxic activity of PBMC against K-562 cells in both the 4 and 18 hour assay. Because of the large variations of responsiveness between the individuals tested with substance P, the increase in cytotoxicity did not reach significance although these changes were in the order of magnitude as those found in the bombesin experiments. Again with substance P, like with bombesin, we found no influence on cytotoxicity in the prolonged 18 hour assay against both K-562 or CaCo-2 target cells. The influences of neurotensin on cytotoxicity could be described as a non-significant tendency to enhancement. Only a marginal increase of cytotoxicity was found after 18 hour of incubation against K-562 target cells (Fig. 3). Somatostatin did not alter the cytotoxic response of PBMC against both K-562 or CaCo-2 target cells. Moreover, this was the only peptide in this study that showed a dose dependent enhancement of chromium release when incubated with target cells alone as control experiment. When these target effects of somatostatin were taken into account, inhibition of cytotoxicity did not reach significance even at a concentration of 10m8 or 10e6M against CaCo-2 targets (Fig. 4).
Discussion The results presented in this study indicate that neuropeptides are directly involved in the modulation of cellular cytoxicity in vitro. Among the peptides that were investigated, bombesin was the most potent stimulator of cytotoxicity especially against CaCo-2 colon carcinoma cells. Bombesin
28
NEUROPEI’TIDES
Increoee in cytotoxicity
X
50
+CaCo-2
lml
-CcoCo-2 4h -- K-662 16h .. K-662 4h
lOE-12
lM-10
lOE-8
NE-6
Concentration BBS(M)
Fig. 1 Effect of bombesin incubation on spontaneous cell mediated cytotoxicity of PBMC against CaCo-2 or K-562 target cells expressed as relative percentual change. Significant differences of bombesin incubations from those of the control (0) experiments are denoted by asterisks (n = 12, (*) = p < 0.06, * = p c 0.05, ** < 0.02, *** = p < 0.01). The mean percentual variation of the individual specific cytotoxicity is indicated by the shaded area.
showed a gradual increase of the cytotoxic response with the concentration added. The observed stimulation of cytotoxicity probably reflects the presence of high affinity receptors on either effector or both effector and target cells. The latter might be explained by the differences observed between K-562 and CaCo-2 cells with respect to their susceptibility to lysis. Bombesin levels in blood are hardly detectable, which might explain the sensitivity of our effector cells to very low concentrations in vitro, mimicking high plasma levels. Notably, the CaCo-2 cells were shown to be more sensitive to bombesin with the higher concentrations. Lymphoid cells in the lamina propria and Peyer’s patches may become exposed to the higher concentrations of the neu-
ropeptide that were tested, when they are in the close vicinity of nerve endings found to be present in the gut wall (20, 21). Functionally relevant peptide receptors, which have been characterized on immunocompetent cells, might reflect the role of both nervous and endocrine systems in regulating the immune response. Although these receptors on lymphoid cells have been characterized for substance P, VIP, somatostatin, P-endorphins and enkephalins by binding studies (6), this has not yet been reported for bombesin. Substance P is relatively potent for stimulation of lymphocyte proliferation, immunoglobulin synthesis, neutrophil activity and mediator release from mast cells (9, 11, 13). It plays a role, not only
29
NEUROPEPTIDES AND CYTOTOXICITY
in these inflammation and allergic events but also in local gut immunity. Because of the connection of substance P containing nerves and gut mucosal tissue, it can locally interact with immunocompetent cells in mucosal tissue or gut associated lymphoid tissue (20, 21). Surprisingly, only cytotoxicity against CaCo-2 target cells was substantially increased at the higher peptide concentrations. The almost similar distribution of bombesin and substance P containing nerves in the gut could be related to the similar stimulatory effects that were observed when using these peptides in vitro against CaCo-2 colon tumour cells. Moreover, receptor binding studies indicated that both the bombesin and substance P receptor share specificity for the binding of both peptides (22). Indeed there are indications for the presence of
specific substance P receptors on immunocompetent cells in gut tissue, e.g. on T- and B-lymphocytes from Peyer’s patches (23). Somatostatin by itself had no effect on cell mediated cytotoxicity when present during the assay, although stimulation of cytotoxicity has been reported when effector cells were preincubated with somatostatin (18). Still somatostatin might interfere with the action of other peptides, since it is well known for its antagonistic effects on gastrointestinal hormone function as well as on exocrine secretion (24). Somatostatin did stimulate the chromium release by target cells in a dose dependent fashion. The underlying mechanism for this phenomenon is unclear but it could be a direct toxic effect of somatostatin on the cells. Moreover, binding of somatostatin to the tumour cells might
X Increare in cytotoxicity SO-
40--
30-+ coca-2 M--
1Bh
- Coca-2 4h -- K-562 1Bh . . K-562 4h
lOE-10 Concentration
lOE-6
lM-8 Subrrtoncr
P(M)
Fig. 2 Influence of substance P incubation on spontaneous cell mediated cytotoxicity of PBMC against CaCo-2 or K-562 target cells expressed as relative percentual change (n = 12). The mean percentual variation of the individual specific cytotoxicity is indicated by the shaded area.
30
NEUROPEF’TIDES
60
0
4h
60
q 16h
0
IN 1OE-161CE-141oE-121E10 1M-6 lOE-6 ConcaltrotimNmrotmsin M)
Cl
IN IOE-161E-14lOE-12loE-10 N-E-8ME-6 Concmtr.tlm NwrotPMInW
Fig. 3 Spontaneous cell mediated cytotoxicity of PBMC after incubation with neurotensin against CaCo-2 (a) and K-562 (b) target cells (n = 12).
interfere with the membrane architecture resulting in increased permeability of the cells. Neurotensin did not influence cytotoxicity of PBMC against the tumour target cells in our assay, even though a broader concentration range of the peptide has been tested. Thus bombesin, and possibly substance P, can modulate spontaneous cell mediated cytotoxity against tumour cells. Although the precise mechanism of stimulation of cytotoxicity is not yet unravelled we found that the effector cells influenced by these peptides are undoubtedly the NK-cells. The alterations of cytotoxicity are only seen in the 4 hour assay where cytotoxicity was found to be mediated for over 95% by CD16+ NK-cells as demonstrated by selective depletion of these cells (unpublished data). The observed effects are rapid, i.e. only in the short assay and at low concentrations. What the in vivo importance of such a rapid transient stimulation may be, remains unclear. In this way for instance neuropeptides may rapidly intervene with cellular cytotoxic effector mechanisms. However, it cannot be excluded that these peptides play a regulatory role by stimulating accessory cells to produce certain cytokines which ultimately results in stimulation of cytotoxic effector cells.
In the present studies large inter-individual variations of responsiveness were observed. Therefore we would like to emphasize that the modulatory role for these neuropeptides in the immune response should not be _generalized.
talk-2 cl
conomtrotionSST(M) Fig. 4 Effect of different concentrations somatostatin on the spontaneous cell mediated cytotoxicity of PBMC against CaCo-2 target cells after 18h incubation and on the spontaneous release of chromium label by CaCo-2 target cells (n = 9).
NEUROPEPTIDES
AND CYTOTOXICITY
Different factors may influence the basal state of the mononuclear cells which were isolated. For instance hormone and receptor levels may interfere with NK-cell activity and therefore change their susceptibility for modulation. In conclusion, the results in this study suggest a role for neuropeptides of the gastrointestinal tract in modulating the cellular immune response. This may especially be of importance in gastrointestinal pathology where impaired NK-cell activity has been reported in relation to several diseases (25, 26).
Acknowledgements We are indebted to Marij Mieremet-Ooms and Annie van der Zon for their careful technical assistance and Susan Vervoordeldonk and Ciske Widijanto for their participation in the project. We also thank Loes Niepoth and Nelia Koek for preparing the manuscript and Rent? Kuijppers for his contribution to the designing of the figures.
References 1. Payan,
2.
3.
4.
5.
6.
D. G., McGillis, J. P., Goetzl, E. J. (1986). Neuroimmunology. Advances in Immunology 39: 299323. Bloom, F. E. (1985). Neuropeptides and other mediators in the central nervous system. Journal of Immunology 135: 743-745. Besedovsky, H. O., delRey, A. E., Sorkin, E. (1985). Immune-neuroendocrine interactions. Journal of Immunology 135: 750-754. Blalock, J. E., Harbour-McMenamin, D., Smith, E. M. (1985). Peptide hormones shared by the neuroendocrine and immunologic systems. Journal of Immunology 135: 858-861. McEwen, B. (1987). Influences of hormones and neuroactive substances on immune function. In: The neuroimmune-endocrine connection. Cotman C. W. (ed.) Raven Press, New York p. 71-92. Bost, K. L. (1988). Hormone and neuropeptide receptors on mononuclear leukocytes. In: Neuroimmunoendocrinology. Blalock J. E.,Bost, K. L. (eds). Progress in allergy. Karger, Basel. p. 68-83. Bienenstock, J., Perdue, M., Stanisz, A., Stead, R. (1987). Neurohormonal regulation of gastrointestinal immunity. Gastroenterology 93: 1431-1434. O’Dorisio, M. S. (1986). Neuropeptides and gastrointestinal immunity. American Journal of Medicine 81: 74-82. Stan&, A. M., Befus, D., Bienenstock, J. (1986). Differential effects of vasoactive intestinal peptide, substance P, and somatostatin on immunoglobulin synthesis and proliferations by lymphocytes from peyer’s patches, mesen-
31 teric lymph nodes, and spleen. Journal of Immunology 136: 152-156. 10. Krco, C. J., Gores, A., Go, V. L. W. (1986). Gastrointestinal regulatory peptides modulate mouse lymphocyte functions under serum-free conditions in vitro. Immunological Investigations 15: 103-l 11.
II. Payan, D. G., Brewster, D. R., Goetzl, E. J. (1983). Specific stimulation of human T lymphocytes by Substance P. Journal of Immunology 131: 1613-1615.
r2. Bar-Shavit, Z., Goldman, R., Stabinsky, Y., Gottlieb, P.,
Fridhin, M., Teichberg, V. I., Blumberg, S. (1980). Enhancement of phagocytosis - a newly found activity of substance P residing in its N-terminal tetrapeptide sequence. Biochemical and Biophysical Research Communications 94: 1445. Shanahan, F.. Denburg, J. A., Fox, J. Bienenstock, J., Befus, D. (1985). Mast cell heterogeneity: effects of neuroenteric peptides on histamine release. Journal of Immunology 135: 1331-1337.
14. Lotz, M., Vaughan, J. H., Carson, D. A. (1988) Effect of neuropeptides on production of inflammatory cytokines by human monocytes. Science 241: 1218-1221. 15. Peck, R. (1987). Neuropeptides modulating macrophage function. In: Neuroimmune interactions. Proceedings of the second international workshop on neuroimmunomodulation. Jankovic, B. D., Markovic, B. M., Spector, N. H. (eds). Annals of the New York Academy of Sciences 496: 264-270. 16 Rola-Pleszczynski, M., Bolduc, D., St-Pierre, S. (1985). The effects of vasoactive intestinal peptide on human natural killer cell function. Journal of Immunology 135: 2569-2573. 17. Drew, P. A., Shearman, D. J. C. (1985). Vaso-active intestinal peptide: a neurotransmitter which reduces human NK cell activity and increases Ig synthesis. Australian Journal of Experimental Biology and Medical Science 63: 313-318. 18, Pawhkowski, M., Zelazowski, P., Stepien, H. (1989). Enhancement of human lymphocyte natural killer activity by somatostatin. Neuropeptides 13: 75-77. 19. Aparicio-Pages, M. N., Verspaget, H. W., Pena, A. S., Weterman, I. T., Mieremet-Ooms, M. A. C., van der Zen, J. M., de Bruin, P. A. F., Lamers, C. B. H. W. (1988) Natural, lectin- and phorbol ester-induced cellular cytotoxicity in Crohn’s disease and ulcerative colitis. Journal of Clinical Laboratory Immunology 27: 109-113. 20. Ferri, G-L., Adrian, T. E., Ghatei, M. A., O’Shaughnessy, D. J., Probert, L., Lee,Y. C.,Buchan, A. M. J., Polak, J. M., Bloom, S. R. (1983). Tissue locahzation and relative distribution of regulatory peptides in separated layers from the human bowel. Gastroenterology 84: 777-786. 21. Lluis, F., Thompson, J.C. (1988). Neuroendocrine potential of the colon and rectum. Gastroenterology 94: 832-844. 22. Jensen, R. T.. Heinz-Erian, P., Mantey, S., Jones, S. W., Gardner, J. D. (1988). Characterization of ability of
32 various substance P antagonists to inhibit action of bombesin. American Journal of Physiology 254: G883-890. 23. Stat&, A. M., Scicchitano, R., Dazin, P., Bienenstock, J., Payan, D. G. (1987). Distribution of Substance P receptors on murine spleen and Peyer’s Patch T and B cells. Journal of Immunology 139: 749-754. 24. Reichlin, S. (1983). Somatostatin. New England Journal of Medicine 309: 1556-1563.
NEUROPEPTIDES
25. Ginsburg, C. H., Dambrauskas, J. T., Ault, K. A., Falchuk, Z. M. (1983). Impaired natural killer cell activity in patients with inflammatory bowel disease: evidence for a qualitative defect. Gastroenterology 85: 846-851. 26. Gibson, P. R., Jewell, D. J. (1986). Local immune mechanisms in inflammatory bowel disease and colorectal carcinoma. Natural killer cells and their activity. Gastroenterology 90: 12-19.
