The transcription factor NF-kappaB plays a pivotal role in gene expression of inflammatory mediators such as cytokines or adhesion molecules. NF-kappaB-mediated transcriptional activation of these genes is inhibited by nitric oxide (NO) in a variety of cells, including monocytes. Morphine mediates NO release in a naloxone antagonizable manner in monocytes and neutrophils.
The influence of morphine on NF-kappaB activation was investigated in a whole-blood flow cytometric assay. A specific antibody against the p65 subunit of NF-kappaB was used and detected by fluoresceine-isothiocyanate-labeled anti-immunoglobulin G. Nuclei were stained with propidium iodide. Leukocyte subpopulations were evaluated by gating on neutrophils and monocytes. The median fluorescence channel was determined. Different morphine concentrations (50 nm, 50 microm, 1 mm) and incubation intervals (10-150 min) were used.
Morphine inhibits lipopolysaccharide-induced NF-kappaB nuclear binding in human blood neutrophils and monocytes in a time-, concentration-, and naloxone-sensitive-dependent manner. Similar effects were achieved with the NO donor S-nitroso-N-acetyl-pencillamine and the antioxidant N-acetyl-cysteine. The NO synthase inhibitors Nomega-nitro-l-arginine-methyl-esther and Nomega-nitro-l-arginine completely abolished the morphine-induced attenuation of NF-kappaB nuclear binding, demonstrating that the inhibitory action is mediated by NO release.
Morphine causes immunosuppression, at least in part, via the NO-stimulated depression of NF-kappaB nuclear binding.
MORPHINE has been shown to be an effective immunomodulator, acting, in part, by stimulating nitric oxide (NO) production in neutrophils, monocytes, and endothelial cells. 1–3The NO synthase (NOS) inhibitor N ω-nitro-l-arginine (NLA) as well as the opiate antagonist naloxone abrogates morphine-induced NO production in these cells, 1demonstrating the specificity of this process. The significance of these observations is increased because NO also inhibits gene expression in inflammation in a variety of cells, including neutrophils and monocytes, 4–10suggesting that this linkage may be the basis of the immune downregulating actions of morphine.
NF-κB is a transcription factor that plays a crucial role in cytokine- and lipopolysaccharide (LPS)-induced gene activation during inflammatory events. 11Incubation of vascular endothelial cells with NO donors inhibited tumor necrosis factor (TNF)-α–induced NF-κB activation by induction and stabilization of the NF-κB inhibitor IκBα. 12In addition, NO donors can directly inhibit the DNA binding activity of NF-κB. 13
Taken together, few studies have demonstrated the activation of NF-κB in human blood monocytes and neutrophils in a whole-blood assay. Although morphine 14and NO 15have been reported to inhibit leukocyte function, i.e. , adherence, phagocytosis, and motility, 16the exact intracellular molecular processes in the opiate-mediated downregulation is not known. In the present report, we determined the influence of morphine on the activation of NF-κB in human blood neutrophils and monocytes. Our data indicate that morphine attenuates LPS-induced NF-κB activation by a NO-dependent and naloxone-sensitive mechanism. NO donors mimicked the inhibitory effects of morphine on nuclear binding of NF-κB, whereas NOS inhibitors such as NLA and N ω-nitro-l-arginine-methyl-esther (l-NAME) abolished morphine effects on NF-κB activation. Thus, morphine seems to invoke an intracellular molecular cascade, extending into nuclear events.
Materials and Methods
After an internal review board approved the project, informed consent was obtained from 13 healthy male volunteers. Ten milliliters of venous blood was collected from each individual. The donors were nonsmokers and had no history of infection or allergy, and they had never been subjected to immunosuppressive therapy. The mean age of the group was 28 ± 3 yr, and the mean weight was 81 ± 6 kg.
Incubation of Whole Blood and Stimulation with LPS
Whole blood was collected in EDTA and aliquoted in tubes (Falcon tubes; Becton Dickinson, Heidelberg, Germany). A total of 100 μl of whole blood was treated with either morphine sulfate (50 nm, 50 μm, 1 mm) or with the NO donor S -nitroso-N -acetyl-pencillamine (SNAP) in a water bath under steady shaking for different time intervals as indicated. To investigate the effects of NOS inhibitors, NLA (0.5 mm) or l-NAME (1 mm) was added to the blood and incubated for 5 min at 37°C in a water bath before treatment with 1 mm morphine sulfate. Opiate receptor specificity was tested by preincubation of samples with naloxone (0.5 μm and 0.5 mm) before morphine treatment. Control tubes were incubated with 0.9% saline. To induce NF-κB, blood cells were stimulated with 100 ng/ml LPS for 30 min in a 37°C water bath. Controls were set up with phosphate-buffered saline. All aliquots were treated with 2 ml of a commercially available lysing solution (FACS Brand Lysing Solution; Becton Dickinson) for 30 min at room temperature for lysis of erythrocytes. The cells were centrifuged at 200 g for 5 min, and the supernatant was discarded.
Preparation of Cells for Flow Cytometry Detection
Preparation of nuclei for flow cytometric analysis was performed as previously described. 17Briefly, cells were washed once with phosphate-buffered saline before using staining reagents contained in a commercially available DNA staining kit (Cycletest Plus DNA Reagent Kit; Becton Dickinson). After centrifugation, the aliquots were resuspended in 3 ml citrate buffer and centrifuged at 300 g for 5 min. The supernatant was discarded, and 1.5 ml citrate buffer was added, followed by centrifugation at 300 g for 5 min. After discarding the supernatant, cells were resuspended in a mixture of 250 μl solution A (trypsin in a spermine tetrahydrochloride detergent buffer) and 200 μl of solution B (trypsin inhibitor and ribonuclease A in citrate stabilizing buffer with spermine tetrahydrochloride) for 10 min at room temperature. A total of 40 μl NF-κB polyclonal rabbit antibody (Santa Cruz Biotechnology, Heidelberg, Germany) was then added to the aliquots for a 10-min incubation period at room temperature. A total of 2.5 μl of a fluorescein isothiocyanate–conjugated (FITC) antirabbit monoclonal antibody (Sigma Chemicals, Deisenhofen, Germany) was added and incubated for further 10 min at room temperature. A total of 200 μl of a cold propidium iodide solution was added to the aliquots and incubated for 10 min.
Flow Cytometric Analysis
The flow cytometer (FACS-Calibur; Becton Dickinson) was equipped with forward-scatter and side-scatter light detectors that allow discrimination of cell size and complexity. It also has an optical filter to detect propidium iodide–stained nuclei. These nuclei emit fluorescent light between 580 and 650 nm that can be detected by the fluorescence-2 filter. The fluorescence-1 detector was used to detect NF-κB activation by measuring emission of green fluorescence at 530 nm corresponding with FITC staining. Fluorescence caused by unspecific binding of the secondary antibody was excluded by incubation of whole blood with FITC-labeled immunoglobulin G alone. Unspecific binding of the p65-antibody was ruled out by western blot of cytosolic and nuclear extracts of leukocytes. The set up was performed using commercially available DNA particles (DNA Quality Control Particles Kit; Becton Dickinson). A total of 20,000 events were recorded for each sample. To evaluate leukocyte subpopulations to which nuclei belonged, we analyzed a forward-scatter/side-scatter dot plot. Polymorphonuclear and mononuclear cells were analyzed for FITC staining using histogram analysis of the fluorescence-1 parameter. The median of intensity fluorescence was used as an indicator for the intensity of nuclei fluorescence.
Statistical Interpretation of Data
Statistical analysis was performed using the Friedman test. To isolate the group or groups that differ from the others, a pairwise multiple comparison procedure was used. Dunnett’s method was chosen because it allows multiple comparisons against a control group. P < 0.05 was considered significant.
Morphine Alone Has No Effect on NF-κB Activation
Three different concentrations of morphine were used to determine if there is an effect on NF-κB activation. None of the concentrations tested (50 nm, 50 μm, 1 mm) had any effect on NF-κB activation compared with control values (data not shown). Incubation with the NO donor SNAP (1 mm and 10 μm) alone also failed to induce nuclear binding of NF-κB. Subsequent stimulation with LPS (100 ng/ml) was essential to activate NF-κB translocation to the nucleus. As shown in figure 1, the percentage of cells showing nuclear binding of NF-κB increased from 2% to 88% in neutrophils and from 2% to 86% in monocytes, respectively, after stimulation with LPS (100 ng/ml;P < 0.05).
Morphine Attenuates LPS-induced Nuclear Binding of NF-κB in a Time- and Concentration-dependent Manner
Two series of experiments were performed to establish time- and concentration-dependent effects of morphine. After a 10-min preincubation with morphine (50 nm and 50 μm), only a slight reduction of NF-κB activation was observed after stimulation with LPS (fig. 2). However, after preincubation with 50 nm morphine for 2.5 h, the median of channel fluorescence was reduced from 486 to 433 in neutrophils and from 287 to 247 in monocytes, respectively (fig. 1;P < 0.05), indicating the inhibition of LPS-induced NF-κB nuclear binding. This effect became more pronounced at higher concentrations of morphine (50 μm and 1 mm).
Preincubation with morphine at different time intervals showed that morphine-induced effects on attenuation of NF-κB activation increased with longer preincubation times (fig. 2). Because morphine is known to modulate the activation of constitutive NOS in endothelial cells, 18we surmise that also in blood cells an ongoing NO release at low levels may be responsible for a stronger attenuation of NF-κB nuclear binding after a 2.5-h preincubation period, as was also suggested by the SNAP experiments. The decrease in median of channel fluorescence correlated with a lower percentage of cells exerting nuclear binding of NF-κB.
NO Donors Attenuate LPS-induced Nuclear Binding of NF-κB
As was shown in previous studies, NO donors such as sodium nitroprusside or SNAP can induce NO-associated changes in various types of cells. 1Our data show that 1 μm SNAP seems to be more effective than 50 μm morphine in inhibiting NF-κB activation (fig. 3), because its effect in the nanomolar range is comparable to the one observed with 50 μm morphine. However, it has to be taken into consideration that NO release induced by 1 μm SNAP is more than twice as high as the one induced by 1 μm morphine (unpublished data).
Morphine Attenuation of LPS-induced Nuclear Binding of NF-κB Is Abolished by NO Antagonists
To determine whether morphine attenuates NF-κB activation via NO release, the NOS inhibitors l-NAME and NLA were administered to the cells 10 min before incubation with morphine. l-NAME (1 mm) as well as NLA (0.5 mm) completely inhibited morphine-induced attenuation of NF-κB nuclear binding (fig. 4). These data demonstrate that morphine exerts its effects on NF-κB activation by inducing NO release in both cell types.
Quantification of Morphine-induced Effects by Comparison with N -acetyl-cysteine
Conventional methods to determine nuclear binding of NF-κB only allow a semiquantitative estimation of NF-κB activation. In this study, the efficacy of morphine in reducing NF-κB activation was compared with N -acetyl-cysteine (NAC). As previously described, NAC completely aborted NF-κB activation at a concentration of 30 mm 17and inhibits NF-κB–dependent gene transcription in vivo . 19The effect of 10 mm NAC on nuclear binding of NF-κB was comparable to the one observed after preincubation with 50 μm morphine (fig. 4). These data suggest that morphine in the micromolar range can inhibit NF-κB activation as effective as NAC in the millimolar range.
The Effects of Morphine Are Mediated by a Classical μ Opiate Receptor
Because naltrindole was found to have little affinity for the opiate receptors present on neutrophils 20while naloxone was very effective in reversing morphine induced changes, we chose naloxone to investigate whether morphine acts via a classical μ-subtype opiate receptor. Preincubation with naloxone at a 10-fold higher concentration than morphine completely reversed morphine-induced attenuation of NF-κB nuclear binding after LPS stimulation (fig. 5). Naloxone alone did not influence NF-κB activation after incubation of whole blood with LPS. We thus conclude that morphine attenuates LPS-induced NF-κB activation by the μ3opiate receptor coupled to NO release on human granulocytes and monocytes because opioid peptides do not release NO or bind to this opiate receptor subtype that these immunocytes contain. 20,21
The present study demonstrates the following: (1) morphine does not alter basal NF-κB nuclear binding without subsequent stimulation; (2) LPS stimulates NF-κB nuclear binding; (3) morphine inhibits LPS-induced NF-κB nuclear binding in a naloxone antagonizable manner; (4) NO donors also inhibit LPS-induced NF-κB binding; and (5) NOS inhibitors block morphine inhibition of LPS-induced NF-κB binding; demonstrating that morphine initiates its actions via constitutive NOS (cNOS) stimulation, since its NO-stimulating activity occurs immediately. Taken together, we surmise that morphine, in part, causes immunosuppression via the NO-stimulated depression of NF-κB nuclear binding.
The significance of NF-κB in immune modulation is evident. It is a DNA-binding factor that plays an essential role in the activation of several inflammatory mediators such as adhesion molecules, 22TNF-α, 23interleukin (IL)-8, 24IL-1β, IL-2, and IL-6. 25The NF-κB complex consists of two heterodimers, termed p50 (NF-κB1) and p65 (RelA). In most types of cells, the NF-κB-inhibitor IκBα is phosphorylated and proteolytically degraded within minutes on activation by inflammatory agents such as TNF-α. 26The p50–p65 complex is released, migrates to the nucleus, binds to the promoter region of target genes, and subsequently induces gene transcription. 27In addition, the NF-κB/Rel transcription factor family also plays a crucial role in cytokine-induced gene activation during inflammatory responses or after exposure to bacterial LPS, 11,28,29as also noted in the present study.
In addition to NF-κB immune modulation, NO also has been reported to modulate the downregulation of immunocytes, 16such as inhibiting the expression of proinflammatory mediators 30and adhesion molecules, 8and the phenotypic reversion of activated cells to a shape that is round and nonmotile. 1,16Although many NO effects are coupled to cyclic guanosine monophosphate–dependent pathways, 31recent data have demonstrated that inhibitory actions of NO on immunocytes are mediated by inhibiting the binding of NF-κB to specific DNA binding motifs in the promoter region of proinflammatory cytokine genes, 12linking NO and NF-κB in a cascading process.
Interestingly, in both neutrophils and monocytes, morphine stimulates cNOS-derived NO release that initiates the NO-associated cell rounding and the loss of immunocyte adherence. 1,20In the same reports, opioid peptides were demonstrated to be immunocyte-excitatory, and they did not liberate NO. Binding studies on monocytes and neutrophils demonstrated that they expressed the opiate alkaloid–selective and opioid peptide–insensitive opiate receptor subtype, designated μ3. 20Furthermore, we demonstrated that morphine activates cNOS, a calcium/calmodulin-dependent enzyme, by increasing intracellular calcium levels. 18In these studies, it is also important to note that NO donors, i.e. , SNAP, mimic the effects of morphine and that the morphine-mediated processes are naloxone-sensitive, indicating that this action is mediated via the μ3receptor subtype.
In other reports, by different laboratories, NO donors have also been shown to reduce LPS-elicited NF-κB activation. 32–34Here, the end results include the inhibition of inducible NOS (iNOS) mRNA, suggesting that NO may limit its own production by a negative feedback mechanism based on an inhibition of NF-κB activation that results in diminished expression of iNOS as a NF-κB–dependent gene. 6,35Thus, morphine-induced NO release via activation of cNOS may at least partially contribute in the regulation of NF-κB–dependent transcription and may thus limit inflammatory processes.
To our knowledge, this study constitutes the first demonstration that morphine inhibits NF-κB activation in human neutrophils and monocytes by a NO-coupled mechanism. In other studies, human peripheral-blood mononuclear cells exposed to NO-generating compounds exhibited enhanced glucose uptake and TNF-α secretion. 36Furthermore, nuclear binding of NF-κB was induced by SNAP and sodium nitroprusside. These results are contradictory to the inhibitory effects of NO donors on NF-κB nuclear binding found in the present study. However, because of different roles of leukocyte subpopulations in inflammation, NO may activate other signaling pathways in peripheral-blood mononuclear cells and may thereby finally lead to an activation of NF-κB in these cells.
In peritoneal mouse macrophages, nanomolar morphine concentrations were found to increase LPS-induced NF-κB activation, 37which correlated with an increased in TNF-α and IL-6 synthesis. However, micromolar concentrations led to a reduction of NF-κB activation and suppressed TNF-α and IL-6 gene transcription. 37These results are contradictory to the concentration-dependent effects of morphine established in our study. The different cell types used may contribute to these contrasting results, because the expression of NF-κB subunits p50 and p65 changes during differentiation of monocyte-derived macrophages. 38In addition, a 100-fold higher LPS concentration was used, which may override low-dose morphine effects. Attenuation of NF-κB activation induced by nanomolar and even micromolar concentrations of morphine is of profound clinical significance, because serum concentrations up to 5 μm were determined after high-dose treatment with morphine. 39
Another outcome of the present study is that the flow cytometric detection and quantification of NF-κB nuclear binding methodology allows for a rapid and highly reproducible determination of NF-κB activation in different leukocyte populations. 17Because a whole-blood assay is used, preactivation of leukocytes by cell separation is avoided. Although, in isolated cells, interactions between different cell populations as well as the influence of plasma proteins cannot be determined, the whole-blood assay provides a comparatively physiologic surrounding. Conventional techniques such as western blot or electric mobility shift assay require at least 5 × 106cells to ensure collection of a certain amount of proteins in nuclear extracts, whereas the flow cytometric assay can be performed with as little as 100 μl of whole blood. Furthermore, preparation of nuclear extracts from neutrophils is delicate, because the presence of certain antiproteinases is required to avoid proteolytic degradation of NF-κB. 40In accordance with previous results, 41,42we could rule out unspecific binding of the polyclonal antibody against p65 used in our flow cytometric assay by western blot of nuclear and cytosolic extracts of leukocytes, which revealed a single band corresponding to a 65-kd protein. Assay standardization with stained nuclei ensures exclusive determination of nuclear NF-κB.
In summary, our data demonstrate that in human neutrophils and monocytes, morphine attenuates NF-κB activation by a NO-coupled mechanism. This signaling pathway may at least partly mediate morphine-induced immunosuppression.