- Email: [email protected]
Morphine synergizes with lipopolysaccharide in a chronic endotoxemia model
Journal of Neuroimmunology 95 Ž1999. 107–114
Morphine synergizes with lipopolysaccharide in a chronic endotoxemia model Sabita Roy a
a,b,)
, Richard G. Charboneau b, Roderick A. Barke
b
Department of Pharmacology, Veterans Administration Medical Center and UniÕersity of Minnesota, Minneapolis, MN 55417, USA b Department of Surgery, Veterans Administration Medical Center and UniÕersity of Minnesota, Minneapolis, MN 55417, USA Received 7 July 1998; revised 2 December 1998; accepted 2 December 1998
Abstract Emergent or elective surgical procedures may be complicated by sepsis, resulting in critical illness that can lead to organ failure and death. The opioid drug, morphine is widely used to alleviate pain in post-surgical patients; however, it is well documented that chronic treatment of mice with morphine affects the proliferation, differentiation and function of immune cells. Thus, morphine might be expected to exacerbate the effects of sepsis, which also compromises the immune system. To test this notion, we investigated the effect on several immune functions of a clinical dose of morphine Ž4 mgrkg. superimposed upon a lipopolysaccharide ŽLPS.-induced infection model. Our results show that this relatively low dose of morphine, though generally having no effects on immune parameters by itself, significantly augmented LPS responses. A clinical dose of morphine Ž4 mgrkg body weight. superimposed upon an animal model of sepsis resulted in a significant increase in mortality at 48 h. In the absence of the drug, most septic animals died after 96 h. Phenotypic responses such as, decreased thymic cellularity, compromised mitogenic response and inhibition of IL-2 synthesis that are evident at 48–72 h after LPS injection appear as early as 24 h in animals that receive morphine in addition to LPS. In addition, our results show that in T cells there is a shift from TH1 type cytokine elaboration to a TH2 type cytokine elaboration in animals that receive both LPS and morphine. q 1999 Elsevier Science B.V. All rights reserved. Keywords: Morphine; Immune cells; Lipopolysaccharide; Endotoxin; THIrTH2 elaboration
1. Introduction Opioid drugs are the treatment of choice to alleviate pain associated with post-surgical trauma and terminal illnesses such as cancer. However, opioids may also affect immune functions, potentially reducing host resistance against invading pathogens. The first evidence for this interaction was obtained as early as 1889 in a guinea pig model ŽCantacuzene, 1898.. More recently, a number of investigators have demonstrated that a variety of immunological parameters are suppressed in opioid addicts ŽBrown et al., 1974; Louria et al., 1974; Wybran et al., 1979; McDonough et al., 1980.. These include the lymphocyte proliferative response to mitogens ŽBrown et al., 1974., T
)
Corresponding author. Department of Surgery, Veterans Administration Medical Center, 1 Veterans Drive, Minneapolis, MN 55417, USA. Tel.: q1-612-7252000 Ext. 5543; Fax: q1-612-7252227; E-mail: [email protected]
cell rosette formation ŽWybran et al., 1979. and the total number of circulating lymphocytes ŽMcDonough et al., 1980.. In animal models, morphine treatment has been found to increase mortality rates in experimentally infected mice ŽTubaro et al., 1985; Chao et al., 1990.. Furthermore, lymphocyte proliferative responses ŽBryant et al., 1988., NK cell cytotoxicity activity ŽLefkowit and Chiang, 1975; Shavit et al., 1986. antibody and serum hemolysin formation ŽGungor et al., 1980., the phagocytic and killing properties of peripheral blood mononuclear leukocytes ŽTubaro et al., 1985. and interleukin-2 ŽIL-2. synthesis ŽSibinga and Goldstein, 1988; Roy and Loh, 1996. are all attenuated after in vivo chronic morphine exposure to animals. While the mechanisms responsible for morphineinduced changes in the immune system are not completely understood, they may be mediated directly through opiate receptors present on lymphocytes ŽCarr et al., 1988., or indirectly through opiate receptors present in the central nervous system which activate the hypothalamic–pitui-
0165-5728r99r$ - see front matter q 1999 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 5 - 5 7 2 8 Ž 9 8 . 0 0 2 6 5 - 3
108
S. Roy et al.r Journal of Neuroimmunology 95 (1999) 107–114
tary–adrenal axis ŽGeorge and Way, 1955. to release immunosuppressive glucocorticoids ŽBryant et al., 1991.. A similar impairment of the immune system is a common manifestation in sepsis and may be the potential cause of death in some patients. Lipopolysaccharide ŽLPS.-induced endotoxemia, an experimental model of sepsis, is known to affect several parameters of the immune system. Sepsis down-regulates IL-2 gene expression in T cells ŽWood et al., 1984., induces thymic apoptosis, decreases T cell mitogenesis and impairs cell mediated immune responses ŽAbraham and Regan, 1985; Rodrick et al., 1986; Lin et al., 1993; Barke et al., 1994a,b; Christou et al., 1995.. In light of the similarity of effects of LPS and opioid drugs, it is important to examine a possible interaction between morphine and sepsis which could potentially complicate use of the drug in post-surgical patients. In this study, we examined the effect of clinical doses of morphine on an LPS-induced infection model. We report that morphine doses as low as 4 mgrkg can synergize with LPS and augment its effects. Phenotypic responses that are typically evident at 48–72 h after LPS injection appear as early as 24 h in animals that receive morphine in addition to LPS.
2. Materials and methods 2.1. Animals and treatment protocol Harlan ICR mice were divided into four groups and treated as follows: Ž1. saline injection subcutaneously Žs.c.. every 8 h Žmorphine control. and intraperitoneally Ži.p.. every 12 h ŽLPS control.; Ž2. morphine sulphate Ž4 mgrkg injected s.c. every 8 h.; Ž3. LPS Ž25 mgrkg injected i.p. every 12 h.; and Ž4. morphine sulphate Ž4 mgrkg injected s.c. every 8 h. q LPS Ž25 mgrkg injected i.p. every 12 h.. These treatments were carried out over a period of 24–96 h. In addition to identify if morphine effects were mediated by opioid receptors animals were injected with varying concentrations of either morphine alone or morphineq naltrexone Ža classical opioid antagonist.. These treatments were carried out over a period of 24–48 h. 2.2. HarÕest of peritoneal macrophages Resident peritoneal macrophages were harvested from all four groups of animals by peritoneal lavage with ice cold, calcium- and magnesium-free Dulbecco’s phosphate-buffered saline. The cells were washed, counted and adjusted to a density of 6 = 10 6 cellsrml in RPMI medium supplemented with 10% fetal calf serum, penicillin–streptomycin Ž100 unitsrml–100 mgrml.. Cells were incubated for 4 h at 378C to allow attachment of macrophages. At the end of 4 h, cells were washed 3 times with CLICKS buffer and the adherent cells were used.
2.3. Thymidine incorporation assay Thymocytes were cultured in RPMI-1640 medium containing 15% fetal bovine serum and penicillin–streptomycin Ž100 unitsrml–100 mgrml.. The cell suspension was adjusted to a final concentration of 1 = 10 6 cellsrml. To assay for cell proliferation, 100 ml of the cell suspension were added to a 96-well microtiter plate containing IL-1 Ž1 ngrml. and leukoagglutinin ŽPHA-L. Ž2.0 mgrml.. The final volume of each culture was 200 ml, and triplicate cultures were plated for each variable. After incubation for 24 h at 378C in 5% CO 2 , the cells were pulsed for an additional 24 h with 1 mCi of w3 Hxthymidine, and the radioactivity associated with the cells determined by scintillation counting. 2.4. ELISA for the detection of IL-2, IL-4, IL-6, IL-12, TNF-a and IFNg Peritoneal macrophages, thymuses and spleen from treated and control groups were harvested aseptically and placed in cold culture media ŽRPMI-1640. containing 10% fetal bovine serum and penicillin–streptomycin Ž100 unitsrml–100 mgrml.. Cells were gently teased apart and passed through a nylon mesh Ž100 mM pores. to remove all cell aggregates and connective tissue. The cell suspension was adjusted to a final concentration of 1 = 10 6 cellsrml in a 48-well plate and stimulated with IL-1 Ž1 ngrml. and Leukoagglutinin ŽPHA-L. Ž2.0 mgrml. for activation of thymic T cells and LPS Ž10 mgrml. for macrophage and splenic cell activation. After 24 h, 100 ml of culture supernatant was analyzed using cytokine specific ELISA kits ŽGenzyme, Cambridge, MA. according to the manufacturer’s instruction. 2.5. Analysis of RNA by reÕerse transcriptase polymerase chain reaction (RT-PCR) One microgram of total RNA was reverse transcribed to synthesize the first-strand cDNA Ž428C, 30 min. using random hexamers Ž2.5 mM., Moloney murine leukemia virus reverse transcriptase Ž2.5 units; Perkin-Elmer. and 1 mM each of dATP, dCTP, dGTP and dTTP in a final reaction volume of 20 ml. Following first-strand synthesis, the reaction mixture was heated Ž958C, 5 min. to inactivate the reverse transcriptase. Amplification was performed using upstream and downstream primers specific for rat IL-1a ŽClontech, Palo Alto, CA., rat IL-6 ŽClontech, Palo Alto, CA., and mouse G3PDH. The rat IL-1a primers amplifies a 623 bp fragment and the rat IL-6 primers amplifies a 614 bp fragment. The first-strand cDNA reaction mixture was added to PCR buffer containing 2.5 units AmpliTaq DNA polymerase ŽPerkin-Elmer., 2 mM MgCl2, 50 mM KCl, 10 mM Tris–HCl, and 0.10 mM of each primer in a total volume of 100 ml. PCR conditions were 948C for 30 s Ždenaturation., 608C for 30 s Žannealing.,
S. Roy et al.r Journal of Neuroimmunology 95 (1999) 107–114
and 728C for 40 s Žextension., followed by a final extension at 728C for 10 min. Pilot studies showed a good correlation between template input and intensity of amplified fragments at 25 cycles for GPGH and 30 cycles for IL-1a and IL-6. Control reactions included total RNA without reverse transcriptase or without template. PCR products were analyzed on 1.5% agarose gel and visualized by ethidium bromide staining. 2.6. Statistical analysis Survival data were calculated for each time period, with experimental groups tested using a chi-square test. All other data were tested using analysis of variance followed by hypothesis testing using Tukey’s test, where applicable. Significance was accepted at p - 0.05. 3. Results 3.1. Effect of morphine on mortality rate following LPS injection We have previously shown that injection of mice with LPS Ž25 mgrkg body weight. results in about 50% mortality in 48 h. The current studies were designed to investigate if there was a change in the mortality rate when morphine was administered concurrently into these animals. Mice were divided into four groups as described in Section 2. As shown in Fig. 1, there was no mortality in the animals that received saline or morphine alone. In
109
animals that were injected with LPS there was a 50 " 5% mortality after 48 h, in agreement with our previous studies. The mortality rate in animals that received morphine in addition to LPS was significantly higher after 48 h, about 85 " 5%. However, at 96 h the mortality in both groups ŽLPS and LPS q morphine. was similar at about 98 " 2%. To prove that morphine was indeed acting synergistically with LPS, we hypothesize that increasing the dose of morphine would further exacerbate the effects of LPS. To prove our hypothesis we injected varying doses of morphine in animals that also received 25 mgrkg LPS. We investigated the mortality rate of the animals 24 h after injections. Our results show that in animals that receive LPS alone the mortality was around 25%, 24 h after injection. However in animals that received morphine in addition to LPS Ž25 mgrkg. there was a dose dependent increase in mortality at 24 h. A 100% mortality was observed in animals that were injected with 10 mgrkg morphineq LPS Ž25 mgrkg. ŽFig. 2.. To prove that morphine was acting through an opiate receptor we injected a group of animals that received morphine Žvarying concentrations. and LPS Ž25 mgrkg. with varying concentrations of naltrexone Ža classical opiate antagonist.. The dose of naltrexone injected in each group was identical to the morphine dose. In these animals mortality was determined 48 h after injections. Our results ŽFig. 3. show that at lower concentrations of morphine Žup to 4 mgrkg. there was a significant reduction in mortality in animals that were injected with naltrexone suggesting that morphine may be acting through an opiate receptor.
Fig. 1. The effect of morphine treatment on mortality as a function of time following LPS injection. Harlan ICR mice were injected with either Ž1. saline Ževery 8 h. q saline Ževery 12 h.; Ž2. morphine sulphate Ž4 mgrkg body weight every 8 h. q saline Ževery 12 h.; Ž3. saline Ževery 8 h. q LPS Ž25 mgrkg body weight every 12 h.; Ž4. morphine sulphate Ž4 mgrkg body weight every 8 h. q LPS Ž25 mgrkg body weight every 12 h.. LPS injections were given i.p. and morphine injections were given s.c. Survival was assessed every 12 h for 120 h after LPS injection. Each point represents the mean" S.E.M. of 20 animals.
110
S. Roy et al.r Journal of Neuroimmunology 95 (1999) 107–114
Fig. 2. Dose dependent effect of morphine on mortality following LPS injection. Harlan ICR mice were injected i.p. with LPS Ž25 mgrkg body weight every 12 h. and varying concentrations of morphine Žs.c. every 8 h. as indicated in the figure. Survival was assessed every 12 h for 24 h after LPS injection. Each point represents the mean"S.E.M. of 20 animals.
However at concentrations of morphine higher than 4 mgrkg, naltrexone Žat equal doses. was unable to antagonize the effects of morphine. The mortality in this group of animal was 100%, 48 h after injection ŽFig. 3.. 3.2. Effect of morphine treatment on thymic weight Our previous studies demonstrated that peritoneal sepsis results in significant thymic involution ŽBarke et al., 1994a,b.. It is also well established that chronic morphine
Fig. 4. The effect of morphine treatment on thymic weight. Mice were injected i.p. as described in Fig. 1 above. Thymic weights were measured 48 h after LPS injection. Each point represents the mean"S.E.M. of eight animals.
treatment Ž70 mg morphine pellet implanted in mice. also results in significant reduction in thymic and splenic weight ŽRoy et al., 1992, 1995; Roy and Loh, 1996.. The results of combining the two treatments are shown in Fig. 4. The relatively low dose of morphine used Ž4 mgrkg body weight. did not affect thymic weight significantly. However, when the same concentration of morphine was used in combination with LPS there was an almost 75% de-
Fig. 3. Effect of naltrexone on morphine induced mortality following LPS injection. Harlan ICR mice were injected i.p. with LPS Ž25 mgrkg body weight every 12 h. and varying concentrations of morphine Žs.c. every 8 h. indicated in the figure. A second set of animals received naltrexone Žvarying concentrations. in addition to morphine Žvarying concentrations. and LPS Ž25 mgrkg.. The concentration of naltrexone injected was the same as the morphine injected. Survival was assessed every 12 h for 48 h after LPS injection. Each point represents the mean" S.E.M. of 20 animals.
S. Roy et al.r Journal of Neuroimmunology 95 (1999) 107–114
111
crease in thymic weight when compared to saline injected animals. Animals that were injected with LPS alone had a 50% decrease in thymic weight when compared to saline injected animals. Thus, morphine at a subthreshold dose was nevertheless able to potentiate the effect of LPS. 3.3. Effect of morphine treatment on thymocyte proliferation We next investigated the effect of morphine and LPS on thymocyte proliferation. Mice were divided into four groups and injected i.p. as described above. After 48 h, thymuses were removed from each group of animals and single cell suspensions were made and cultured in vitro with PHA Ž2 mgrml. and IL-1 Ž1 ngrml.. As shown in Fig. 5, there was a nearly 80% decrease in thymocyte proliferation in the group of animals injected with both LPS and morphine, when compared to the saline injected group. Animals that received LPS alone also showed some reduction in their proliferative capacity Ž40%., but not as dramatic as the group that was injected with both morphine and LPS. Treatment with morphine alone resulted in a small decrease Ž15%. in proliferation. 3.4. Effect of morphine treatment on thymic and peritoneal macrophage cytokine synthesis Activation of thymocytes is dependent on secretion of IL-2, an important autocrine T cell growth factor. We have previously documented that morphine treatment in vitro reduces IL-2 secretion from thymocytes in a dose-depen-
Fig. 5. The effect of morphine treatment on thymocyte proliferation. Animals were treated as described in Fig. 1 above. Cell proliferation was assayed as described in Section 2. Each point represents the mean"S.E.M. of eight animals.
Fig. 6. The effect of morphine treatment on IL-2 synthesis. Animals were treated as described in Fig. 1 above. After 96 h, thymocytes were removed from each group of animals and single cell suspensions were made. The cell concentration was adjusted to a final concentration of 1=10 6 cellsrml and plated in a 48-well plate and stimulated with IL-1 Ž1 ngrml. and PHA Ž2.0 mgrml.. After 24 h, 100 ml of culture supernatants was analyzed using a ELISA kit ŽGenzyme, Cambridge, MA.. Each point represents the mean"S.E.M. of eight animals.
dent manner ŽRoy et al., 1997.. We have also shown that thymocyte IL-2 gene expression and IL-2 synthesis are inhibited following peritoneal sepsis ŽBarke et al., 1994a,b.. Accordingly, we now determined the interaction of these two treatments. As shown in Fig. 6, treatment with morphine at 4 mgrkg body weight, a subthreshold dose in this system, did not affect IL-2 secretion significantly. However, LPS-treated animals showed reduced IL-2 in the culture supernatants Ž420 pgrml.. A further suppression in IL-2 secretion Ž187 pgrml. was seen in animals that received both morphine and LPS injections. These results support our hypothesis that the inhibition of proliferation seen in animals injected with LPS and morphine was mediated by decreased secretion of IL-2. To investigate the effect of morphine on macrophage cytokine secretion, single-cell suspensions of adherent peritoneal macrophage cells were adjusted to a final concentration of 1 = 10 6 cellsrml, and activated with LPS for 24 h. Culture supernatants were assayed for TNF-a and IL-6 ŽFig. 7. protein levels using an ELISA kit as described in the methods section. Our results show morphine synergized with LPS and increased both IL-6 and TNF-a secretion. The morphine-mediated increase in cytokine secretion was reversed in animals that received naltrexone Ž4 mgrkg. q morphine Ž4 mgrkg. q LPS Ž25 mgrkg. Ždata not shown.. Animals that received morphine alone, however, did not show any significant increase in synthesis of either cytokine.
112
S. Roy et al.r Journal of Neuroimmunology 95 (1999) 107–114
3.6. Effect of morphine treatment on pituitary IL-1 and IL-6 messenger RNA Following peritoneal sepsis, an increased gene expression of the cytokines IL-1 and IL-6 in the pituitary has
Fig. 7. The effect of morphine treatment on LPS induced TNF-a and IL-6 synthesis. Single-cell suspensions, of peritoneal macrophages from mice treated as described in Fig. 1 above, were adjusted to a final concentration of 1=10 6 cellsrml. The cells were allowed to adhere for 4 h. Nonadherent cells were removed by washing 3 times with Click’s medium. The adherent cells were further cultured with LPS Ž10 mgrml. for 24 h and their supernatants were assayed for IL-6 and TNF-a using ELISA kits ŽGenzyme, Cambridge, MA.. The Y-axis represents cytokine levels as a function of LPS treated samples Ždesignated as 100%.. Open bar represents TNF-a levels and closed bar represents IL-6 levels. Each point plotted here represents the mean of at least six independent determinations"S.E.M.
3.5. Effect of morphine treatment on TH1 r TH2 cytokine elaboration It has been shown that splenocytes from mice following burn trauma demonstrate aspects of impaired cellular immunity along with diminished production of IL-12 and IFNg, and impaired TH1 type differentiation. Similarly, in critically ill patients following severe trauma or sepsis, IL-12 and IFNg protein production are markedly impaired. We investigated the effect of morphine treatment on TH1rTH2 cytokine elaboration in our LPS mediated sepsis model. Our results show that in animals treated with both LPS and morphine there was a significant shift towards TH2 cytokine profile within 48 h ŽFig. 8B.. There was a reciprocal relationship between IFNg levels and IL-4 protein levels. In animals that were treated with LPS alone IL-4 levels were significantly lower or negligible at 48 h. The shift towards TH2 type cytokine secretions were observed in the LPS group at 96 h after LPS injections ŽFig. 8A..
Fig. 8. Effect of morphine treatment on TH1rTH2 cytokine elaboration. ŽA. Animals were injected with LPS Ž25 mgrkg. and sacrificed at 24, 48, 72 and 96 h following LPS treatment. Total splenocytes were cultured in a final concentration of 10 6 cellsrml with LPS Ž10 mgrml.. At the end of 48 h culture supernatant were assayed for IFNg, IL-12 and IL-4 using ELISA kits ŽGenzyme, Cambridge, MA.. The Y-axis represents cytokine levels as a function of LPS treated samples Ždesignated as 100%.. Standard curves were generated and was used to determine levels of specific cytokines in experimental cultures. Each point plotted represents the mean of at least six independent determinations"S.E.M. ŽB. Similar experiments were done with animals that received 4 mgrkg morphine sulfate in addition to LPS Ž25 mgrkg..
S. Roy et al.r Journal of Neuroimmunology 95 (1999) 107–114
been demonstrated ŽBarke et al., 1994a,b.. LPS is well known to increase circulating IL-1 and IL-6 levels in septic mouse models. We therefore investigated the effect of morphine superimposed on an experimental sepsis model on the inflammatory cytokine gene expression in the pituitary. Four hours following LPS injection animals were sacrificed and pituitaries harvested for messenger RNA. Total RNA from each pituitary was analyzed using RT-PCR with primers specific for rat IL-1a , IL-6 and mouse G3PDH. The results from these experiments are shown in Fig. 9. LPS injection significantly increased both IL-1a and IL-6 mRNA levels in the pituitary. The increase in IL-1a and IL-6 in the pituitary was even greater in animals that were injected with both LPS and morphine. Saline injected animals showed very low levels of IL-1a and IL-6 messages. Surprisingly, injection of morphine Ž4 mgrkg. alone increased only IL-6 transcript levels and did not change IL-1a transcript levels Ždata not shown.. The identity of the amplified bands were confirmed using full length mouse IL-1a and Il-6 cDNA Ždata not shown..
Fig. 9. The effect of morphine treatment on pituitary IL-1 and IL-6 messenger RNA. Animals were treated as described in Fig. 1 above. Six hours following LPS injection animals were sacrificed and pituitaries harvested. Total RNA from each pituitary was analyzed using RT-PCR with rat primers ŽClontech, Palo Alto, CA. specific for IL-1a , IL-6 and mouse G3PDH. PCR products were analyzed on 1.5% Agarose gels containing 0.5% ethidium bromide. 1: saline treatment; 2: LPS treatment; 3: LPSqmorphine treatment. ŽA. IL-1a mRNA RT-PCR product Ž623 bp.. ŽB. IL-6 mRNA RT-PCR product Ž614 bp.. ŽC. G3PDH Žhousekeeping gene. mRNA RT-PCR product. All treatments and reactions were performed in triplicate.
113
4. Discussion The immunosuppressive effects of chronic opioid use are fairly well established ŽRoy and Loh, 1996.. Impairment of the immune system is also a common manifestation of trauma or sepsis and a potential cause of increased morbidity and mortality. Since morphine is routinely used for pain management in critically ill septic patients, it is of great clinical relevance to determine the immune effects of morphine in septic animals. A clinical dose of morphine Ž4 mgrkg body weight. superimposed upon an animal model of sepsis resulted in a significant increase in mortality at 48 h, though even in the absence of the drug, most septic animals died after 96 h. Furthermore, thymic cellularity, mitogenic response, IL-2 synthesis and TH1rTH2 differentiation were more severely suppressed in animals that received both LPS and morphine than animals that received LPS alone. Morphine injection into septic animals also increased pituitary IL-1 and IL-6 mRNA levels significantly above those seen following LPS injection alone. Of particular interest, morphine administered at the same dose schedule to non-septic animals had little or no effect on these same immune parameters. This suggests that the relatively low and short-term doses given to patients recovering from surgery may nevertheless significantly complicate the effects of sepsis. Though many studies have demonstrated that opioids induce immunosuppression, the mechanisms underlying this action are not well understood. Two major possibilities have been suggested: an indirect mechanism, in which opioids bind to classical opioid receptors in the CNS, causing the release of catecholamines andror steroids, which then affect the immune cells Žreviewed by Roy and Loh, 1996.; and a direct mechanism, in which opioids bind to both classical and non-classical opioid receptors on the immune cells and directly modulate the function of these cells Žreviewed by Roy and Loh, 1996.. Several in vivo studies provide evidence to support the indirect mechanism of morphine action, through binding to brain receptors and modulation of the hypothalamic–hypophyseal-axis wHPAx ŽBayer et al., 1990; Bryant et al., 1991; Loyd et al., 1991; Woolf, 1992., and since the HPA has also been implicated in sepsis, this might be a logical site of convergence of these two treatments. Consistent with this conclusion, morphine injection into septic animals increased pituitary IL-1a and IL-6 mRNA levels significantly above LPS injection alone. The mechanism by which pituitary IL-6 can modulate the immune system is still speculative, but it has been shown that hypophyseal synthesis of IL-6 can influence pituitary transcription of hormones associated with the classic stress response such as ACTH, prolactin and growth hormones. Our studies also show that the effects of low concentrations of morphine Žclinical dose: 1–4 mgrkg. superimposed on LPS are reversed in animals that received naltrexone Žan opioid antagonist., suggesting that morphine at
114
S. Roy et al.r Journal of Neuroimmunology 95 (1999) 107–114
low doses may be acting through classical opioid receptors. However, our results also show that when high concentrations of morphine are used Žgreater than 4 mgrkg., injection of the same dose of naltrexone dose not reverse the effects of morphine. This results suggest that at high concentrations, morphine may be acting through, our previously characterized ŽRoy et al., 1992., non-classical opioid receptors Žnaltrexone insensitive.. Our preliminary studies show that morphine concentrations above 10 mgrkg by itself have significant immunosuppressive effects. In summary, we have shown that morphine at doses that have little or no effect on immune parameters by themselves can significantly exacerbate several important measures of immunosuppression in a septic animal model. Morphine is widely used as an analgesic for patients recovering from surgery who are at risk for sepsis, and these effects should be weighed against its beneficial value in treating these patients. Acknowledgements This work was supported by grants from NIH R29DA08188 ŽS. Roy. and the Veterans Affairs Merit Review ŽR.A. Barke.. References Abraham, E., Regan, R.F., 1985. The effects of hemorrhage and trauma on IL-2 production. Arch. Surg. 120, 1341–1344. Barke, R.A., Roy, S., Chapin, R., Charboneau, R., 1994a. The role of programmed cell death Žapoptosis. in thymic involution following sepsis. Arch. Surg. 129, 1256–1262. Barke, R.A., Roy, S., Chapin, R., Charboneau, R., 1994b. Sepsis induced release of interleukin-6 may activate the immediate early gene program through a hypothalamic–hypophyseal mechanism. Surgery 116, 141–149. Bayer, B.M., Daussin, S., Hernandez, M., Irvin, L., 1990. Morphine inhibition of lymphocyte activity is mediated by an opioid dependent mechanism. Neuropharmacology 29, 369–374. Brown, S.M., Stimmel, B., Taub, R.N., Kochwa, S., Rosenfeld, R.E., 1974. Immunological dysfunction in heroin addicts. Arch. Int. Med. 134, 1001–1006. Bryant, H.U., Yoburn, B.C., Intrissi, C.E., Bernton, E.W., Holaday, J.W., 1988. Immunosuppressive effects of chronic morphine treatment. Eur. J. Pharm. 149, 165–169. Bryant, H.U., Bernton, E.W., Kenner, J.R., Holaday, J.W., 1991. Role of adrenal cortical activation in the immunosuppressive effects of chronic morphine treatment. Endocrinology 128 Ž6., 3253–3258. Cantacuzene, J., 1898. Nouvelles recherches sur le monde de destruction des vibrions dans l’organisme. Ann. Inst. Pasteur 12, 273–300.
Carr, D.J.J., Kim, C.-H., DeCosta, B.R., Jacobson, A.E., Rice, K.C., Blalock, J.R., 1988. Evidence for a k-class opioid receptor on cells of the immune system. Cell. Immunol. 116, 44–51. Chao, C.C., Sharp, B.M., Pomeroy, C., Filice, G.A., Peterson, P.K., 1990. Lethality of morphine in mice infected with Toxoplasma gondii. J. Pharmacol. Exp. Ther. 252, 605–609. Christou, N.V., Meakins, J.L., Gordon, J., Yee, J., Hassan, Z.M., Nohr, C.W., Shizgal, H.M., Maclean, D., 1995. The delayed hypersensitivity response and host resistance in surgical patients. Ann. Surg. 222, 534–546. George, R., Way, E.L., 1955. Studies on the mechanism of pituitary– adrenal activation by morphine. Br. J. Pharm. 10, 260–264. Gungor, M., Genc, E., Sogduyu, H., Eroglu, L., Koyuncuoglu, H., 1980. Effect of chronic administration of morphine on primary immune response in mice. Experientia ŽBasel. 36, 1309–1310. Lefkowit, S.S., Chiang, C.Y., 1975. Effects of certain abused drugs on hemolysin forming cells. Life Sci. 17, 1763–1768. Lin, R.Y., Astiz, M.E., Saxon, J.C., Rackow, E.C., 1993. Altered leukocytes immunophenotypes in septic shock. Chest 104, 847–853. Louria, D.B., Hensle, T., Rose, J., 1974. The major medical complications of heroin addiction. Ann. Int. Med. 67, 1–22. Loyd, D.A., Teich, S., Rowe, M.I., 1991. Serum endorphins levels in injured children. Surg. Gyn. Obstet. 172, 449–452. McDonough, R.J., Madden, J.J., Falek, A., Shafer, D.A., Pline, M., Gordon, D., Bokos, P., Kuehnle, J.C., Mendelson, J., 1980. Alterations of T and Null lymphocyte frequencies in the peripheral blood of human opiate addicts. J. Immunol. 125, 2539. Rodrick, M.L., Wood, J.J., O’Mahony, J.B., 1986. Mechanisms of immunosuppression associated with severe nonthermal traumatic injuries in man. J. Clin. Invest. 6, 310. Roy, S., Loh, H.H., 1996. Effects of opioids on the immune system. Neurochem. Res. 21 Ž11., 1373–1384. Roy, S., Ge, B.L., Loh, L.L., Lee, N.M., 1992. Characterization of 3 H-morphine binding to interleukin-1 activated thymocytes. J. Pharm. Exp. Ther. 263, 45–451. Roy, S., Loh, H.H., Barke, R.A., 1995. Morphine induced suppression of thymocyte proliferation is mediated by inhibition of IL-2 synthesis. Adv. Exp. Med. Biol. 373, 41–48. Roy, S., Chapin, R., Cain, K., Charboneau, R., Ramakrishnan, S., Barke, R.A., 1997. Morphine inhibits transcriptional regulation of IL-2 synthesis in thymocytes. Cell. Immunol. 179, 1–9. Shavit, Y., Martin, F.C., Angarita, L.H., Gale, R.P., Liebeskind, J.C., 1986. Morphine-induced suppression of natural killer cell activity is mediated by the adrenal gland. Soc. Neurosci. Abstr. 12, 339. Sibinga, N.E.S., Goldstein, A., 1988. Opioid peptides and opioid receptors in cells of the immune system. Ann. Rev. Immunol. 6, 219–249. Tubaro, E., Avico, U., Santiangeli, C., Zuccaro, P., Cavallo, G., Pacifici, R., Croce, C., Borelli, G., 1985. Morphine and methadone impact of human phagocytic physiology. Int. J. Immunopharmacol. 7, 865–874. Wood, J.J., Rodrick, M.L., O’Mahony, J.B., 1984. Inadequate IL-2 production; a fundamental immunologic deficiency in patients with major burns. Ann. Surg. 200, 311–320. Woolf, P.D., 1992. Hormonal responses to trauma. Crit. Care Med. 20, 216–226. Wybran, J., Appelboon, T., Famey, J.P., Govaerts, A., 1979. Suggestive evidence for receptors for morphine and methionine enkephalin on normal human blood T-lymphocytes. J. Immunol. 123, 1068–1070.
Morphine synergizes with lipopolysaccharide in a chronic endotoxemia model Sabita Roy a
a,b,)
, Richard G. Charboneau b, Roderick A. Barke
b
Department of Pharmacology, Veterans Administration Medical Center and UniÕersity of Minnesota, Minneapolis, MN 55417, USA b Department of Surgery, Veterans Administration Medical Center and UniÕersity of Minnesota, Minneapolis, MN 55417, USA Received 7 July 1998; revised 2 December 1998; accepted 2 December 1998
Abstract Emergent or elective surgical procedures may be complicated by sepsis, resulting in critical illness that can lead to organ failure and death. The opioid drug, morphine is widely used to alleviate pain in post-surgical patients; however, it is well documented that chronic treatment of mice with morphine affects the proliferation, differentiation and function of immune cells. Thus, morphine might be expected to exacerbate the effects of sepsis, which also compromises the immune system. To test this notion, we investigated the effect on several immune functions of a clinical dose of morphine Ž4 mgrkg. superimposed upon a lipopolysaccharide ŽLPS.-induced infection model. Our results show that this relatively low dose of morphine, though generally having no effects on immune parameters by itself, significantly augmented LPS responses. A clinical dose of morphine Ž4 mgrkg body weight. superimposed upon an animal model of sepsis resulted in a significant increase in mortality at 48 h. In the absence of the drug, most septic animals died after 96 h. Phenotypic responses such as, decreased thymic cellularity, compromised mitogenic response and inhibition of IL-2 synthesis that are evident at 48–72 h after LPS injection appear as early as 24 h in animals that receive morphine in addition to LPS. In addition, our results show that in T cells there is a shift from TH1 type cytokine elaboration to a TH2 type cytokine elaboration in animals that receive both LPS and morphine. q 1999 Elsevier Science B.V. All rights reserved. Keywords: Morphine; Immune cells; Lipopolysaccharide; Endotoxin; THIrTH2 elaboration
1. Introduction Opioid drugs are the treatment of choice to alleviate pain associated with post-surgical trauma and terminal illnesses such as cancer. However, opioids may also affect immune functions, potentially reducing host resistance against invading pathogens. The first evidence for this interaction was obtained as early as 1889 in a guinea pig model ŽCantacuzene, 1898.. More recently, a number of investigators have demonstrated that a variety of immunological parameters are suppressed in opioid addicts ŽBrown et al., 1974; Louria et al., 1974; Wybran et al., 1979; McDonough et al., 1980.. These include the lymphocyte proliferative response to mitogens ŽBrown et al., 1974., T
)
Corresponding author. Department of Surgery, Veterans Administration Medical Center, 1 Veterans Drive, Minneapolis, MN 55417, USA. Tel.: q1-612-7252000 Ext. 5543; Fax: q1-612-7252227; E-mail: [email protected]
cell rosette formation ŽWybran et al., 1979. and the total number of circulating lymphocytes ŽMcDonough et al., 1980.. In animal models, morphine treatment has been found to increase mortality rates in experimentally infected mice ŽTubaro et al., 1985; Chao et al., 1990.. Furthermore, lymphocyte proliferative responses ŽBryant et al., 1988., NK cell cytotoxicity activity ŽLefkowit and Chiang, 1975; Shavit et al., 1986. antibody and serum hemolysin formation ŽGungor et al., 1980., the phagocytic and killing properties of peripheral blood mononuclear leukocytes ŽTubaro et al., 1985. and interleukin-2 ŽIL-2. synthesis ŽSibinga and Goldstein, 1988; Roy and Loh, 1996. are all attenuated after in vivo chronic morphine exposure to animals. While the mechanisms responsible for morphineinduced changes in the immune system are not completely understood, they may be mediated directly through opiate receptors present on lymphocytes ŽCarr et al., 1988., or indirectly through opiate receptors present in the central nervous system which activate the hypothalamic–pitui-
0165-5728r99r$ - see front matter q 1999 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 5 - 5 7 2 8 Ž 9 8 . 0 0 2 6 5 - 3
108
S. Roy et al.r Journal of Neuroimmunology 95 (1999) 107–114
tary–adrenal axis ŽGeorge and Way, 1955. to release immunosuppressive glucocorticoids ŽBryant et al., 1991.. A similar impairment of the immune system is a common manifestation in sepsis and may be the potential cause of death in some patients. Lipopolysaccharide ŽLPS.-induced endotoxemia, an experimental model of sepsis, is known to affect several parameters of the immune system. Sepsis down-regulates IL-2 gene expression in T cells ŽWood et al., 1984., induces thymic apoptosis, decreases T cell mitogenesis and impairs cell mediated immune responses ŽAbraham and Regan, 1985; Rodrick et al., 1986; Lin et al., 1993; Barke et al., 1994a,b; Christou et al., 1995.. In light of the similarity of effects of LPS and opioid drugs, it is important to examine a possible interaction between morphine and sepsis which could potentially complicate use of the drug in post-surgical patients. In this study, we examined the effect of clinical doses of morphine on an LPS-induced infection model. We report that morphine doses as low as 4 mgrkg can synergize with LPS and augment its effects. Phenotypic responses that are typically evident at 48–72 h after LPS injection appear as early as 24 h in animals that receive morphine in addition to LPS.
2. Materials and methods 2.1. Animals and treatment protocol Harlan ICR mice were divided into four groups and treated as follows: Ž1. saline injection subcutaneously Žs.c.. every 8 h Žmorphine control. and intraperitoneally Ži.p.. every 12 h ŽLPS control.; Ž2. morphine sulphate Ž4 mgrkg injected s.c. every 8 h.; Ž3. LPS Ž25 mgrkg injected i.p. every 12 h.; and Ž4. morphine sulphate Ž4 mgrkg injected s.c. every 8 h. q LPS Ž25 mgrkg injected i.p. every 12 h.. These treatments were carried out over a period of 24–96 h. In addition to identify if morphine effects were mediated by opioid receptors animals were injected with varying concentrations of either morphine alone or morphineq naltrexone Ža classical opioid antagonist.. These treatments were carried out over a period of 24–48 h. 2.2. HarÕest of peritoneal macrophages Resident peritoneal macrophages were harvested from all four groups of animals by peritoneal lavage with ice cold, calcium- and magnesium-free Dulbecco’s phosphate-buffered saline. The cells were washed, counted and adjusted to a density of 6 = 10 6 cellsrml in RPMI medium supplemented with 10% fetal calf serum, penicillin–streptomycin Ž100 unitsrml–100 mgrml.. Cells were incubated for 4 h at 378C to allow attachment of macrophages. At the end of 4 h, cells were washed 3 times with CLICKS buffer and the adherent cells were used.
2.3. Thymidine incorporation assay Thymocytes were cultured in RPMI-1640 medium containing 15% fetal bovine serum and penicillin–streptomycin Ž100 unitsrml–100 mgrml.. The cell suspension was adjusted to a final concentration of 1 = 10 6 cellsrml. To assay for cell proliferation, 100 ml of the cell suspension were added to a 96-well microtiter plate containing IL-1 Ž1 ngrml. and leukoagglutinin ŽPHA-L. Ž2.0 mgrml.. The final volume of each culture was 200 ml, and triplicate cultures were plated for each variable. After incubation for 24 h at 378C in 5% CO 2 , the cells were pulsed for an additional 24 h with 1 mCi of w3 Hxthymidine, and the radioactivity associated with the cells determined by scintillation counting. 2.4. ELISA for the detection of IL-2, IL-4, IL-6, IL-12, TNF-a and IFNg Peritoneal macrophages, thymuses and spleen from treated and control groups were harvested aseptically and placed in cold culture media ŽRPMI-1640. containing 10% fetal bovine serum and penicillin–streptomycin Ž100 unitsrml–100 mgrml.. Cells were gently teased apart and passed through a nylon mesh Ž100 mM pores. to remove all cell aggregates and connective tissue. The cell suspension was adjusted to a final concentration of 1 = 10 6 cellsrml in a 48-well plate and stimulated with IL-1 Ž1 ngrml. and Leukoagglutinin ŽPHA-L. Ž2.0 mgrml. for activation of thymic T cells and LPS Ž10 mgrml. for macrophage and splenic cell activation. After 24 h, 100 ml of culture supernatant was analyzed using cytokine specific ELISA kits ŽGenzyme, Cambridge, MA. according to the manufacturer’s instruction. 2.5. Analysis of RNA by reÕerse transcriptase polymerase chain reaction (RT-PCR) One microgram of total RNA was reverse transcribed to synthesize the first-strand cDNA Ž428C, 30 min. using random hexamers Ž2.5 mM., Moloney murine leukemia virus reverse transcriptase Ž2.5 units; Perkin-Elmer. and 1 mM each of dATP, dCTP, dGTP and dTTP in a final reaction volume of 20 ml. Following first-strand synthesis, the reaction mixture was heated Ž958C, 5 min. to inactivate the reverse transcriptase. Amplification was performed using upstream and downstream primers specific for rat IL-1a ŽClontech, Palo Alto, CA., rat IL-6 ŽClontech, Palo Alto, CA., and mouse G3PDH. The rat IL-1a primers amplifies a 623 bp fragment and the rat IL-6 primers amplifies a 614 bp fragment. The first-strand cDNA reaction mixture was added to PCR buffer containing 2.5 units AmpliTaq DNA polymerase ŽPerkin-Elmer., 2 mM MgCl2, 50 mM KCl, 10 mM Tris–HCl, and 0.10 mM of each primer in a total volume of 100 ml. PCR conditions were 948C for 30 s Ždenaturation., 608C for 30 s Žannealing.,
S. Roy et al.r Journal of Neuroimmunology 95 (1999) 107–114
and 728C for 40 s Žextension., followed by a final extension at 728C for 10 min. Pilot studies showed a good correlation between template input and intensity of amplified fragments at 25 cycles for GPGH and 30 cycles for IL-1a and IL-6. Control reactions included total RNA without reverse transcriptase or without template. PCR products were analyzed on 1.5% agarose gel and visualized by ethidium bromide staining. 2.6. Statistical analysis Survival data were calculated for each time period, with experimental groups tested using a chi-square test. All other data were tested using analysis of variance followed by hypothesis testing using Tukey’s test, where applicable. Significance was accepted at p - 0.05. 3. Results 3.1. Effect of morphine on mortality rate following LPS injection We have previously shown that injection of mice with LPS Ž25 mgrkg body weight. results in about 50% mortality in 48 h. The current studies were designed to investigate if there was a change in the mortality rate when morphine was administered concurrently into these animals. Mice were divided into four groups as described in Section 2. As shown in Fig. 1, there was no mortality in the animals that received saline or morphine alone. In
109
animals that were injected with LPS there was a 50 " 5% mortality after 48 h, in agreement with our previous studies. The mortality rate in animals that received morphine in addition to LPS was significantly higher after 48 h, about 85 " 5%. However, at 96 h the mortality in both groups ŽLPS and LPS q morphine. was similar at about 98 " 2%. To prove that morphine was indeed acting synergistically with LPS, we hypothesize that increasing the dose of morphine would further exacerbate the effects of LPS. To prove our hypothesis we injected varying doses of morphine in animals that also received 25 mgrkg LPS. We investigated the mortality rate of the animals 24 h after injections. Our results show that in animals that receive LPS alone the mortality was around 25%, 24 h after injection. However in animals that received morphine in addition to LPS Ž25 mgrkg. there was a dose dependent increase in mortality at 24 h. A 100% mortality was observed in animals that were injected with 10 mgrkg morphineq LPS Ž25 mgrkg. ŽFig. 2.. To prove that morphine was acting through an opiate receptor we injected a group of animals that received morphine Žvarying concentrations. and LPS Ž25 mgrkg. with varying concentrations of naltrexone Ža classical opiate antagonist.. The dose of naltrexone injected in each group was identical to the morphine dose. In these animals mortality was determined 48 h after injections. Our results ŽFig. 3. show that at lower concentrations of morphine Žup to 4 mgrkg. there was a significant reduction in mortality in animals that were injected with naltrexone suggesting that morphine may be acting through an opiate receptor.
Fig. 1. The effect of morphine treatment on mortality as a function of time following LPS injection. Harlan ICR mice were injected with either Ž1. saline Ževery 8 h. q saline Ževery 12 h.; Ž2. morphine sulphate Ž4 mgrkg body weight every 8 h. q saline Ževery 12 h.; Ž3. saline Ževery 8 h. q LPS Ž25 mgrkg body weight every 12 h.; Ž4. morphine sulphate Ž4 mgrkg body weight every 8 h. q LPS Ž25 mgrkg body weight every 12 h.. LPS injections were given i.p. and morphine injections were given s.c. Survival was assessed every 12 h for 120 h after LPS injection. Each point represents the mean" S.E.M. of 20 animals.
110
S. Roy et al.r Journal of Neuroimmunology 95 (1999) 107–114
Fig. 2. Dose dependent effect of morphine on mortality following LPS injection. Harlan ICR mice were injected i.p. with LPS Ž25 mgrkg body weight every 12 h. and varying concentrations of morphine Žs.c. every 8 h. as indicated in the figure. Survival was assessed every 12 h for 24 h after LPS injection. Each point represents the mean"S.E.M. of 20 animals.
However at concentrations of morphine higher than 4 mgrkg, naltrexone Žat equal doses. was unable to antagonize the effects of morphine. The mortality in this group of animal was 100%, 48 h after injection ŽFig. 3.. 3.2. Effect of morphine treatment on thymic weight Our previous studies demonstrated that peritoneal sepsis results in significant thymic involution ŽBarke et al., 1994a,b.. It is also well established that chronic morphine
Fig. 4. The effect of morphine treatment on thymic weight. Mice were injected i.p. as described in Fig. 1 above. Thymic weights were measured 48 h after LPS injection. Each point represents the mean"S.E.M. of eight animals.
treatment Ž70 mg morphine pellet implanted in mice. also results in significant reduction in thymic and splenic weight ŽRoy et al., 1992, 1995; Roy and Loh, 1996.. The results of combining the two treatments are shown in Fig. 4. The relatively low dose of morphine used Ž4 mgrkg body weight. did not affect thymic weight significantly. However, when the same concentration of morphine was used in combination with LPS there was an almost 75% de-
Fig. 3. Effect of naltrexone on morphine induced mortality following LPS injection. Harlan ICR mice were injected i.p. with LPS Ž25 mgrkg body weight every 12 h. and varying concentrations of morphine Žs.c. every 8 h. indicated in the figure. A second set of animals received naltrexone Žvarying concentrations. in addition to morphine Žvarying concentrations. and LPS Ž25 mgrkg.. The concentration of naltrexone injected was the same as the morphine injected. Survival was assessed every 12 h for 48 h after LPS injection. Each point represents the mean" S.E.M. of 20 animals.
S. Roy et al.r Journal of Neuroimmunology 95 (1999) 107–114
111
crease in thymic weight when compared to saline injected animals. Animals that were injected with LPS alone had a 50% decrease in thymic weight when compared to saline injected animals. Thus, morphine at a subthreshold dose was nevertheless able to potentiate the effect of LPS. 3.3. Effect of morphine treatment on thymocyte proliferation We next investigated the effect of morphine and LPS on thymocyte proliferation. Mice were divided into four groups and injected i.p. as described above. After 48 h, thymuses were removed from each group of animals and single cell suspensions were made and cultured in vitro with PHA Ž2 mgrml. and IL-1 Ž1 ngrml.. As shown in Fig. 5, there was a nearly 80% decrease in thymocyte proliferation in the group of animals injected with both LPS and morphine, when compared to the saline injected group. Animals that received LPS alone also showed some reduction in their proliferative capacity Ž40%., but not as dramatic as the group that was injected with both morphine and LPS. Treatment with morphine alone resulted in a small decrease Ž15%. in proliferation. 3.4. Effect of morphine treatment on thymic and peritoneal macrophage cytokine synthesis Activation of thymocytes is dependent on secretion of IL-2, an important autocrine T cell growth factor. We have previously documented that morphine treatment in vitro reduces IL-2 secretion from thymocytes in a dose-depen-
Fig. 5. The effect of morphine treatment on thymocyte proliferation. Animals were treated as described in Fig. 1 above. Cell proliferation was assayed as described in Section 2. Each point represents the mean"S.E.M. of eight animals.
Fig. 6. The effect of morphine treatment on IL-2 synthesis. Animals were treated as described in Fig. 1 above. After 96 h, thymocytes were removed from each group of animals and single cell suspensions were made. The cell concentration was adjusted to a final concentration of 1=10 6 cellsrml and plated in a 48-well plate and stimulated with IL-1 Ž1 ngrml. and PHA Ž2.0 mgrml.. After 24 h, 100 ml of culture supernatants was analyzed using a ELISA kit ŽGenzyme, Cambridge, MA.. Each point represents the mean"S.E.M. of eight animals.
dent manner ŽRoy et al., 1997.. We have also shown that thymocyte IL-2 gene expression and IL-2 synthesis are inhibited following peritoneal sepsis ŽBarke et al., 1994a,b.. Accordingly, we now determined the interaction of these two treatments. As shown in Fig. 6, treatment with morphine at 4 mgrkg body weight, a subthreshold dose in this system, did not affect IL-2 secretion significantly. However, LPS-treated animals showed reduced IL-2 in the culture supernatants Ž420 pgrml.. A further suppression in IL-2 secretion Ž187 pgrml. was seen in animals that received both morphine and LPS injections. These results support our hypothesis that the inhibition of proliferation seen in animals injected with LPS and morphine was mediated by decreased secretion of IL-2. To investigate the effect of morphine on macrophage cytokine secretion, single-cell suspensions of adherent peritoneal macrophage cells were adjusted to a final concentration of 1 = 10 6 cellsrml, and activated with LPS for 24 h. Culture supernatants were assayed for TNF-a and IL-6 ŽFig. 7. protein levels using an ELISA kit as described in the methods section. Our results show morphine synergized with LPS and increased both IL-6 and TNF-a secretion. The morphine-mediated increase in cytokine secretion was reversed in animals that received naltrexone Ž4 mgrkg. q morphine Ž4 mgrkg. q LPS Ž25 mgrkg. Ždata not shown.. Animals that received morphine alone, however, did not show any significant increase in synthesis of either cytokine.
112
S. Roy et al.r Journal of Neuroimmunology 95 (1999) 107–114
3.6. Effect of morphine treatment on pituitary IL-1 and IL-6 messenger RNA Following peritoneal sepsis, an increased gene expression of the cytokines IL-1 and IL-6 in the pituitary has
Fig. 7. The effect of morphine treatment on LPS induced TNF-a and IL-6 synthesis. Single-cell suspensions, of peritoneal macrophages from mice treated as described in Fig. 1 above, were adjusted to a final concentration of 1=10 6 cellsrml. The cells were allowed to adhere for 4 h. Nonadherent cells were removed by washing 3 times with Click’s medium. The adherent cells were further cultured with LPS Ž10 mgrml. for 24 h and their supernatants were assayed for IL-6 and TNF-a using ELISA kits ŽGenzyme, Cambridge, MA.. The Y-axis represents cytokine levels as a function of LPS treated samples Ždesignated as 100%.. Open bar represents TNF-a levels and closed bar represents IL-6 levels. Each point plotted here represents the mean of at least six independent determinations"S.E.M.
3.5. Effect of morphine treatment on TH1 r TH2 cytokine elaboration It has been shown that splenocytes from mice following burn trauma demonstrate aspects of impaired cellular immunity along with diminished production of IL-12 and IFNg, and impaired TH1 type differentiation. Similarly, in critically ill patients following severe trauma or sepsis, IL-12 and IFNg protein production are markedly impaired. We investigated the effect of morphine treatment on TH1rTH2 cytokine elaboration in our LPS mediated sepsis model. Our results show that in animals treated with both LPS and morphine there was a significant shift towards TH2 cytokine profile within 48 h ŽFig. 8B.. There was a reciprocal relationship between IFNg levels and IL-4 protein levels. In animals that were treated with LPS alone IL-4 levels were significantly lower or negligible at 48 h. The shift towards TH2 type cytokine secretions were observed in the LPS group at 96 h after LPS injections ŽFig. 8A..
Fig. 8. Effect of morphine treatment on TH1rTH2 cytokine elaboration. ŽA. Animals were injected with LPS Ž25 mgrkg. and sacrificed at 24, 48, 72 and 96 h following LPS treatment. Total splenocytes were cultured in a final concentration of 10 6 cellsrml with LPS Ž10 mgrml.. At the end of 48 h culture supernatant were assayed for IFNg, IL-12 and IL-4 using ELISA kits ŽGenzyme, Cambridge, MA.. The Y-axis represents cytokine levels as a function of LPS treated samples Ždesignated as 100%.. Standard curves were generated and was used to determine levels of specific cytokines in experimental cultures. Each point plotted represents the mean of at least six independent determinations"S.E.M. ŽB. Similar experiments were done with animals that received 4 mgrkg morphine sulfate in addition to LPS Ž25 mgrkg..
S. Roy et al.r Journal of Neuroimmunology 95 (1999) 107–114
been demonstrated ŽBarke et al., 1994a,b.. LPS is well known to increase circulating IL-1 and IL-6 levels in septic mouse models. We therefore investigated the effect of morphine superimposed on an experimental sepsis model on the inflammatory cytokine gene expression in the pituitary. Four hours following LPS injection animals were sacrificed and pituitaries harvested for messenger RNA. Total RNA from each pituitary was analyzed using RT-PCR with primers specific for rat IL-1a , IL-6 and mouse G3PDH. The results from these experiments are shown in Fig. 9. LPS injection significantly increased both IL-1a and IL-6 mRNA levels in the pituitary. The increase in IL-1a and IL-6 in the pituitary was even greater in animals that were injected with both LPS and morphine. Saline injected animals showed very low levels of IL-1a and IL-6 messages. Surprisingly, injection of morphine Ž4 mgrkg. alone increased only IL-6 transcript levels and did not change IL-1a transcript levels Ždata not shown.. The identity of the amplified bands were confirmed using full length mouse IL-1a and Il-6 cDNA Ždata not shown..
Fig. 9. The effect of morphine treatment on pituitary IL-1 and IL-6 messenger RNA. Animals were treated as described in Fig. 1 above. Six hours following LPS injection animals were sacrificed and pituitaries harvested. Total RNA from each pituitary was analyzed using RT-PCR with rat primers ŽClontech, Palo Alto, CA. specific for IL-1a , IL-6 and mouse G3PDH. PCR products were analyzed on 1.5% Agarose gels containing 0.5% ethidium bromide. 1: saline treatment; 2: LPS treatment; 3: LPSqmorphine treatment. ŽA. IL-1a mRNA RT-PCR product Ž623 bp.. ŽB. IL-6 mRNA RT-PCR product Ž614 bp.. ŽC. G3PDH Žhousekeeping gene. mRNA RT-PCR product. All treatments and reactions were performed in triplicate.
113
4. Discussion The immunosuppressive effects of chronic opioid use are fairly well established ŽRoy and Loh, 1996.. Impairment of the immune system is also a common manifestation of trauma or sepsis and a potential cause of increased morbidity and mortality. Since morphine is routinely used for pain management in critically ill septic patients, it is of great clinical relevance to determine the immune effects of morphine in septic animals. A clinical dose of morphine Ž4 mgrkg body weight. superimposed upon an animal model of sepsis resulted in a significant increase in mortality at 48 h, though even in the absence of the drug, most septic animals died after 96 h. Furthermore, thymic cellularity, mitogenic response, IL-2 synthesis and TH1rTH2 differentiation were more severely suppressed in animals that received both LPS and morphine than animals that received LPS alone. Morphine injection into septic animals also increased pituitary IL-1 and IL-6 mRNA levels significantly above those seen following LPS injection alone. Of particular interest, morphine administered at the same dose schedule to non-septic animals had little or no effect on these same immune parameters. This suggests that the relatively low and short-term doses given to patients recovering from surgery may nevertheless significantly complicate the effects of sepsis. Though many studies have demonstrated that opioids induce immunosuppression, the mechanisms underlying this action are not well understood. Two major possibilities have been suggested: an indirect mechanism, in which opioids bind to classical opioid receptors in the CNS, causing the release of catecholamines andror steroids, which then affect the immune cells Žreviewed by Roy and Loh, 1996.; and a direct mechanism, in which opioids bind to both classical and non-classical opioid receptors on the immune cells and directly modulate the function of these cells Žreviewed by Roy and Loh, 1996.. Several in vivo studies provide evidence to support the indirect mechanism of morphine action, through binding to brain receptors and modulation of the hypothalamic–hypophyseal-axis wHPAx ŽBayer et al., 1990; Bryant et al., 1991; Loyd et al., 1991; Woolf, 1992., and since the HPA has also been implicated in sepsis, this might be a logical site of convergence of these two treatments. Consistent with this conclusion, morphine injection into septic animals increased pituitary IL-1a and IL-6 mRNA levels significantly above LPS injection alone. The mechanism by which pituitary IL-6 can modulate the immune system is still speculative, but it has been shown that hypophyseal synthesis of IL-6 can influence pituitary transcription of hormones associated with the classic stress response such as ACTH, prolactin and growth hormones. Our studies also show that the effects of low concentrations of morphine Žclinical dose: 1–4 mgrkg. superimposed on LPS are reversed in animals that received naltrexone Žan opioid antagonist., suggesting that morphine at
114
S. Roy et al.r Journal of Neuroimmunology 95 (1999) 107–114
low doses may be acting through classical opioid receptors. However, our results also show that when high concentrations of morphine are used Žgreater than 4 mgrkg., injection of the same dose of naltrexone dose not reverse the effects of morphine. This results suggest that at high concentrations, morphine may be acting through, our previously characterized ŽRoy et al., 1992., non-classical opioid receptors Žnaltrexone insensitive.. Our preliminary studies show that morphine concentrations above 10 mgrkg by itself have significant immunosuppressive effects. In summary, we have shown that morphine at doses that have little or no effect on immune parameters by themselves can significantly exacerbate several important measures of immunosuppression in a septic animal model. Morphine is widely used as an analgesic for patients recovering from surgery who are at risk for sepsis, and these effects should be weighed against its beneficial value in treating these patients. Acknowledgements This work was supported by grants from NIH R29DA08188 ŽS. Roy. and the Veterans Affairs Merit Review ŽR.A. Barke.. References Abraham, E., Regan, R.F., 1985. The effects of hemorrhage and trauma on IL-2 production. Arch. Surg. 120, 1341–1344. Barke, R.A., Roy, S., Chapin, R., Charboneau, R., 1994a. The role of programmed cell death Žapoptosis. in thymic involution following sepsis. Arch. Surg. 129, 1256–1262. Barke, R.A., Roy, S., Chapin, R., Charboneau, R., 1994b. Sepsis induced release of interleukin-6 may activate the immediate early gene program through a hypothalamic–hypophyseal mechanism. Surgery 116, 141–149. Bayer, B.M., Daussin, S., Hernandez, M., Irvin, L., 1990. Morphine inhibition of lymphocyte activity is mediated by an opioid dependent mechanism. Neuropharmacology 29, 369–374. Brown, S.M., Stimmel, B., Taub, R.N., Kochwa, S., Rosenfeld, R.E., 1974. Immunological dysfunction in heroin addicts. Arch. Int. Med. 134, 1001–1006. Bryant, H.U., Yoburn, B.C., Intrissi, C.E., Bernton, E.W., Holaday, J.W., 1988. Immunosuppressive effects of chronic morphine treatment. Eur. J. Pharm. 149, 165–169. Bryant, H.U., Bernton, E.W., Kenner, J.R., Holaday, J.W., 1991. Role of adrenal cortical activation in the immunosuppressive effects of chronic morphine treatment. Endocrinology 128 Ž6., 3253–3258. Cantacuzene, J., 1898. Nouvelles recherches sur le monde de destruction des vibrions dans l’organisme. Ann. Inst. Pasteur 12, 273–300.
Carr, D.J.J., Kim, C.-H., DeCosta, B.R., Jacobson, A.E., Rice, K.C., Blalock, J.R., 1988. Evidence for a k-class opioid receptor on cells of the immune system. Cell. Immunol. 116, 44–51. Chao, C.C., Sharp, B.M., Pomeroy, C., Filice, G.A., Peterson, P.K., 1990. Lethality of morphine in mice infected with Toxoplasma gondii. J. Pharmacol. Exp. Ther. 252, 605–609. Christou, N.V., Meakins, J.L., Gordon, J., Yee, J., Hassan, Z.M., Nohr, C.W., Shizgal, H.M., Maclean, D., 1995. The delayed hypersensitivity response and host resistance in surgical patients. Ann. Surg. 222, 534–546. George, R., Way, E.L., 1955. Studies on the mechanism of pituitary– adrenal activation by morphine. Br. J. Pharm. 10, 260–264. Gungor, M., Genc, E., Sogduyu, H., Eroglu, L., Koyuncuoglu, H., 1980. Effect of chronic administration of morphine on primary immune response in mice. Experientia ŽBasel. 36, 1309–1310. Lefkowit, S.S., Chiang, C.Y., 1975. Effects of certain abused drugs on hemolysin forming cells. Life Sci. 17, 1763–1768. Lin, R.Y., Astiz, M.E., Saxon, J.C., Rackow, E.C., 1993. Altered leukocytes immunophenotypes in septic shock. Chest 104, 847–853. Louria, D.B., Hensle, T., Rose, J., 1974. The major medical complications of heroin addiction. Ann. Int. Med. 67, 1–22. Loyd, D.A., Teich, S., Rowe, M.I., 1991. Serum endorphins levels in injured children. Surg. Gyn. Obstet. 172, 449–452. McDonough, R.J., Madden, J.J., Falek, A., Shafer, D.A., Pline, M., Gordon, D., Bokos, P., Kuehnle, J.C., Mendelson, J., 1980. Alterations of T and Null lymphocyte frequencies in the peripheral blood of human opiate addicts. J. Immunol. 125, 2539. Rodrick, M.L., Wood, J.J., O’Mahony, J.B., 1986. Mechanisms of immunosuppression associated with severe nonthermal traumatic injuries in man. J. Clin. Invest. 6, 310. Roy, S., Loh, H.H., 1996. Effects of opioids on the immune system. Neurochem. Res. 21 Ž11., 1373–1384. Roy, S., Ge, B.L., Loh, L.L., Lee, N.M., 1992. Characterization of 3 H-morphine binding to interleukin-1 activated thymocytes. J. Pharm. Exp. Ther. 263, 45–451. Roy, S., Loh, H.H., Barke, R.A., 1995. Morphine induced suppression of thymocyte proliferation is mediated by inhibition of IL-2 synthesis. Adv. Exp. Med. Biol. 373, 41–48. Roy, S., Chapin, R., Cain, K., Charboneau, R., Ramakrishnan, S., Barke, R.A., 1997. Morphine inhibits transcriptional regulation of IL-2 synthesis in thymocytes. Cell. Immunol. 179, 1–9. Shavit, Y., Martin, F.C., Angarita, L.H., Gale, R.P., Liebeskind, J.C., 1986. Morphine-induced suppression of natural killer cell activity is mediated by the adrenal gland. Soc. Neurosci. Abstr. 12, 339. Sibinga, N.E.S., Goldstein, A., 1988. Opioid peptides and opioid receptors in cells of the immune system. Ann. Rev. Immunol. 6, 219–249. Tubaro, E., Avico, U., Santiangeli, C., Zuccaro, P., Cavallo, G., Pacifici, R., Croce, C., Borelli, G., 1985. Morphine and methadone impact of human phagocytic physiology. Int. J. Immunopharmacol. 7, 865–874. Wood, J.J., Rodrick, M.L., O’Mahony, J.B., 1984. Inadequate IL-2 production; a fundamental immunologic deficiency in patients with major burns. Ann. Surg. 200, 311–320. Woolf, P.D., 1992. Hormonal responses to trauma. Crit. Care Med. 20, 216–226. Wybran, J., Appelboon, T., Famey, J.P., Govaerts, A., 1979. Suggestive evidence for receptors for morphine and methionine enkephalin on normal human blood T-lymphocytes. J. Immunol. 123, 1068–1070.