Although numerous animal studies have indicated that morphine compromises the in vivo immune status (1–3), this compound is an important analgesic for optimal pain management, especially in cancer patients. The reason for encouraging the use of morphine is that, because its immunomodulations are probably mediated through central mechanisms such as the hypothalamo-pituitary-adrenocortical responses (4,5), its optimal pain relief potentially counteracts such negative effects by stabilizing the stress responses in painful conditions (6).
In recent years, however, several investigators have independently reported direct effects of morphine on apoptosis in cultured human peripheral blood lymphocytes (PBL). Nair et al. (7) and Singhal et al. (8) demonstrated increased DNA degradation in human PBL that were cultured with morphine. Yin et al. (9) reported rather different results. They did not detect direct apoptosis but found increased Fas expression in morphine-treated PBL that showed apoptotic cell death when co-cultured with Fas ligand (FasL) or Fas-specific antibodies. Despite discrepancies in the methodologies and results, these reports clearly indicated a possibility that morphine directly triggers the apoptotic process in human PBL and raised a serious concern about the use of clinical opioid-analgesic strategies, although they could effectively attenuate pain-related stress responses. Regarding this issue, we present contradictory results based on a quantitative determination of apoptosis-related substances using flow cytometry (FCM) in morphine-treated human PBL.
With approval of our institutional human investigation committee, venous blood samples were collected in heparin from healthy subjects after written informed consent was obtained from all individuals. RPMI-1640 medium (Invitrogen, Carlsbad, CA) supplemented with 5% (vol/vol) heat inactivated fetal calf serum (Dainippon, Osaka, Japan), 0.5% HEPES, 0.2% NaHCO3, and 40 μg/mL of gentamicin was used for the cell culture. Heparinized blood was diluted two times with the medium and then poured over 20 mL of Ficoll-Paque Plus (Amersham Biosciences, Wikströms, Sweden). Peripheral blood mononuculear cells (PBMCs) were isolated by density gradient centrifugation at 400 g for 30 min at 20°C. Isolated PBMCs were washed three times with the medium, and more than 95% cell viability was confirmed by trypan blue staining. The PBMCs (each of 2 × 105 cells/mL) were incubated in a medium containing etoposide (10−4 M; topoisomerase II inhibitor for a positive control; MBL, Nagoya, Japan), morphine hydrochloride (10−8 M–10−4 M; Sankyo, Osaka, Japan), or vehicle (RPMI-1640) for 48 h at 37°C in a humidified atmosphere, containing 95% air/5% CO2.
For the detection of apoptosis, the PBMCs were double-stained after incubation with FITC-conjugated annexin V (MBL), which interacts with cell-surface phosphatidylserine (PS), and 7-amino actinomycin D (7-AAD) (Beckman Coulter, Tokyo, Japan), which passes across the injured cell membrane and binds to degraded DNA, according to the manufacturer’s instructions. Briefly, PBMCs were centrifuged (500g for 5 min), and each cell pellet was incubated with 20 μL of annexin V, 85 μL of binding buffer (MBL), and 10 μL of 7-AAD for 15 min at room temperature in the dark. FCM analysis was performed on all samples, which had been resuspended in 400 μL of the binding buffer. Samples from 10 healthy donors (5 men and 5 women, 25–35 yr of age) were analyzed in this and the following detections of Fas, Bcl-2, and Caspase-3.
Fas (CD95) expression on the PBL surface was detected by a direct immunofluorescence method with FITC-conjugated anti-CD95 monoclonal antibody (IMMUNOTECH, Marseille, France). After incubation, PBMCs were centrifuged (500g for 5 min), and the supernatants were discarded. Each sample was incubated with 20 μL of anti-CD95 antibody for 30 min at room temperature in the dark, washed twice with phosphate-buffered solution (PBS), and analyzed on a flow cytometer. As a negative control, mouse immunoglobulin (Ig)G1-FITC (IMMUNOTECH) was applied for the determination of the percentage of cells labeled because of nonspecific binding.
To detect intracellular Bcl-2, PBMCs were fixed and permeabilized using a commercial cell-preparation kit (IntraStain, Dako, Glostrup, Denmark) according to the manufacturer’s instructions. Briefly, PBMCs were centrifuged (500g for 5 min), and each cell pellet was fixed in 100 μL of Reagent A for 15 min and then permeabilized in 100 μL of Reagent B for 15 min at room temperature in the dark. Bcl-2 immunostaining was performed simultaneously with the cell permeabilization using 20 μL of FITC-conjugated anti-human Bcl-2 monoclonal antibody (Dako). The samples were washed twice with PBS and analyzed on a FCM. Mouse IgG1-FITC (IMMUNOTECH) was used for a negative control.
Intracellular Caspase-3 activity was measured using PhiPhiLux-G1D2 (Calviochem, San Diego, CA). PhiPhiLux is a peptide substrate for Caspase-3 that has been conjugated to two fluorophores (G1D2). The substrate contains the amino acid sequence GDEVDGI, with the caspase cleavage site underlined. The cleaved PhiPhiLux-G1D2 substrate emits a green fluorescence. After incubation, PBMCs were centrifuged (500g for 5 min), and the supernatants were discarded. Each pellet was resuspended with 50 μL of 10 μM of PhiPhiLux-G1D2 substrate solution and incubated for 1 h in 95% air/5% CO2 at 37°C. After washing once with ice cold flow cytometry dilution buffer (Calviochem), the cells were analyzed on a flow cytometer.
To evaluate the effects of the Fas-mediated pathway on the apoptosis of morphine-exposed lymphocytes, PBMCs were isolated from eight healthy donors (30–42 yr old, including three subjects who were involved in the previous series of experiments) and co-incubated with an anti-Fas monoclonal antibody (FasAb) (20 μg/mL; clone CH-11; MBL) that had been verified to have cytolytic activity on human cells expressing Fas. These PBMCs were co-cultured with morphine (10−5 and 10−4 M) or etoposide (10−4 M), as described above.
Each labeled PBMC was examined on a Coulter EPICS XL series FCM (Beckman Coulter, Tokyo, Japan). Before measurement, the optical path was adjusted by testing with FlowCheck (Beckman Coulter). The result of one-half CV should be in the range of <1.5%. Data acquisition and analysis were performed with an EPICS System II (Beckman Coulter). Lymphocytes were gated from the analyses based on the two-dimensional profiles of forward scatter and side scatter. The fluorescence intensity was measured in 10,000 cells. The fluorescence intensity of FITC and the cleaved PhiPhiLux-G1D2 substrate was measured at the fluorescence 1 (FL1) channel (band pass filter: 525 ± 15 nm), whereas the fluorescence intensity of 7-AAD was measured at the FL4 channel (band pass filter: 675 ± 15 nm).
Statistical analysis was performed by repeated-measure analysis of variance. All values are mean ± sd. Statistical significance was defined as P < 0.05.
Figure 1 shows representative distributions of the fluorescence intensity in the PBLs that were labeled with annexin V and 7-AAD after co-culture with etoposide (10−4 M), morphine (10−6 M), or vehicle analyzed by FCM. The 24-h co-culture with etoposide clearly increased both annexin V-positive (exposing PS) and 7-AAD-positive (expressing degraded DNA) cells, which was not observed with vehicle or morphine.
The changes in annexin V-positive/7-AAD-negative cell rates, which indicates early apoptosis (10), during co-culture with etoposide (10−4 M), morphine (10−8 – 10−4 M), or vehicle are shown in Figure 2. The cells incubated with etoposide did not survive for 48 h. The rate of the labeled cells after 12- and 24-h incubation, respectively, was 30.2% ± 3.8% and 41.5% ± 8.0% in the etoposide (P < 0.0001 versus the morphine and vehicle in both time points), 4.4% ± 1.20% and 4.2% ± 0.8% in the morphine (all data combined with no interconcentration difference), and 4.2% ± 0.9% and 4.7% ± 1.1% in the vehicle cultures. The difference in the expression rates was not statistically significant between the morphine and the vehicle cultures during 48 h. There was no difference in the rate of annexin V positive/7-AAD positive cells, indicating late apoptosis or necrosis (10) between these cultures, although a significant increase was observed in the co-culture with etoposide during 24 h-incubation (data not shown).
Figure 3 shows surface Fas expression, intracellular Bcl-2 activity, and Caspase-3 activity. Etoposide significantly increased Fas-positive cells during 24 h of the incubation (37.2% ± 0.8% at 6 h, 45.6% ± 14.6% at 12 h, and 44.1% ± 9.0% at 24 h; P < 0.0001 versus the morphine and the vehicle cultures). Compared with these, the expression rates of Fas were stable ranging from 30.2% ± 4.0% to 32.1% ± 13.9% in the morphine-treated cells (10−5 and 10−4 M, data combined with no interconcentration difference) and from 28.3% ± 10.1% to 34.5% ± 7.7% in the vehicle-treated cells (ns) during 48 h of incubation (Fig. 3A). Intracellular Bcl-2-positive cells decreased from 95.8% ± 0.4% to 84.1% ± 8.7% between 12 and 24 h of incubation with etoposide (difference was not statistically significant). However, neither the morphine nor the vehicle cultures affected the Bcl-2 expression rates, which were distributed in a narrow range from 91.0% to 97.6% during 48 h of incubation (Fig. 3B). The rates of the PBLs that expressed activated intracellular Caspase-3 were significantly increased in the etoposide cultures after 12 h of incubation (11.7% ± 3.8% at 12 h and 27.1% ± 15.6% at 24 h; P < 0.0001 versus the morphine and the vehicle cultures). Compared to these, the rates remained <2% in both of the morphine and vehicle cultures without interculture difference throughout the 48-h incubation (Fig. 3C).
The effects of the FasAb treatment on the rates of annexin V-positive/7-AAD-negative PBLs during co-culture with morphine or etoposide are shown in Figure 4. In the PBLs co-incubated with morphine and FasAb, the rate of annexin V positive/7-AAD negative (early apoptotic) cells ranged from 3.1% ± 1.2% to 6.2% ± 3.2% during 48 h with no statistical difference in each sampling period. These expression rates were not different from those in the cultures with vehicle, vehicle with FasAb, or morphine alone in any sampling period. No significant changes in the rate of annexin V positive/7-AAD positive (late apoptotic or necrotic) cells were observed in these cultures either (data not shown). Compared with these results, etoposide significantly increased the apoptotic cell rates after 12 h both in the cultures with and those without FasAb (P < 0.0001 versus each of the morphine or vehicle cultures at 12 and 24 h) and eliminated all cells by 48 h. Although the difference did not reach statistical significance, the annexin V positive/7-AAD negative cell rates tended to be more frequent in the etoposide cultures with FasAb (FasAb+; 52.9 ± 13.9 versus FasAb−; 41.5% ± 8.0%) at 24 h. In this assay, the expression rates of Fas in the cultures without FasAb ranged from 28.3% ± 10.1% to 34.5% ± 7.7% in the vehicle and morphine (ns among the treatments, morphine concentrations, and sampling periods) and from 31.1% ± 3.1% (3 h) to 45.6% ± 14.6% (12 h) in the etoposide (P < 0.0001 versus the morphine and the vehicle cultures). These results were consistent with the ranges of the data presented in Figure 3A.
In the first series of this study, cells cultured with morphine hydrochloride that were simultaneously labeled with annexin V and 7-AAD were examined by FCM. The concentrations of morphine in the culture medium (10−8 – 10−4 M) used in this study were relevant to the therapeutic plasma range (<10−5 M) of this compound in cancer pain management (11). However, no treatment effect of morphine on the labeled cell numbers was found at any concentrations applied throughout the incubation (Fig. 1). Loss of the plasma membrane integrity occurs early in the apoptotic process. This event results in the exposure of PS residues at the outer plasma membrane leaflet (12). Annexin V was shown to interact strongly and specifically with PS (13), which led to the general use of labeled annexin V in apoptosis research. Combining annexin V staining with DNA staining (7-AAD), which has become a widely used methodology for detecting apoptosis, enables possible discrimination between early apoptotic cells (annexin V-positive/7-AAD-negative) and late apoptotic or necrotic cells (annexin V-positive/7-AAD-positive) (10). In fact, in this study, etoposide clearly increased the rates of the cells that were labeled with these specific substances, and the cells did not survive for 48 h of incubation. Therefore, our findings do not support those of previous reports (7,8) suggesting that morphine directly induces apoptosis in human PBL in vitro.
To confirm these findings, we next looked at the major determinants relating to the apoptotic process including surface Fas expression, intracellular Bcl-2 activity, and Caspase-3 activity in morphine-treated PBLs (Fig. 3). Fas (CD95/Apo1) is a transmembrane protein belonging to the tumor necrosis factor superfamily of receptors that regulates cellular immune response and homeostasis. Binding of specific ligands to Fas in various immune events initiates cell apoptosis (14). Yin et al. (9) demonstrated that morphine increased Fas expression in human PBL resulting in promotion of apoptotic cell death when co-incubated with FasL or Fas-specific antibodies. Bcl-2 is an integral membrane protein located mainly on the mitochondrial outer membrane. Although the role of Bcl-2 in human PBL apoptosis is not entirely understood, it has been demonstrated that a decrease in intracellular Bcl-2 results in apoptosis in specific cell lines through the efflux of cytochrome c from the mitochondria and Caspase-3 activation (15). Caspase-3 plays a key role in apoptotic cell death. Although no previous report has indicated the direct involvement of morphine in the process, various apoptotic signals can activate this enzyme resulting in DNA degradation (16). In our second series of experiments, the FCM identified rates at which PBL expressed each molecule-like immunoreactivity were consistent with previously reported ranges (17,18). Furthermore, etoposide clearly modulated the expression of these molecules in cultured human PBL, suggesting that apoptotic processes progressed during co-culture with this agent. However, we found no treatment effects of morphine on any of these determinants (Figs. 2 and 3).
The third series of experiments were performed to elucidate the putative involvement of the Fas-mediated system on the morphine-primed PBL apoptosis by using FasAb that has been verified to induce apoptosis in human cells expressing Fas. However, we found no effect of FasAb on the rate of annexin V-positive (PS-exposing) cells in the cultures with morphine (Fig. 4). This result seems to be reasonable because we found no modulation by morphine on the expression of Fas in PBLs (Fig. 3A).
Morphine-induced lymphocyte apoptosis was previously claimed based on two independent nonquantitative measurements, electrophoresis on extracted DNA in PBMCs (7), and microscopic analysis of DNA staining in purified T cells (8). Both of these detected DNA degradation in the cells during morphine exposure. In contrast, a quantitative FCM analysis performed by Yin et al. (9) failed to reconfirm such a phenomenon when PBL were co-cultured with morphine alone. The primary FCM results in our study agree with the latter findings, suggesting that differences in the quantification of the nuclear events in part affected the conclusions drawn regarding apoptosis. In addition, our data were obtained from PBLs that were sorted using light scatter-based FCM gating. Because this may include some erroneous T-cell discrimination, differences in the analyzed cell populations among previous studies and our study may alternatively explain the discrepant results.
There are also apparent differences between the findings by Yin et al. (9) and our findings. These authors reported increased Fas expression in PBL during morphine treatment resulting in apoptosis in the presence of FasL or Fas-specific antibodies, which we failed to observe (Fig. 3A and 4). It might be possible that these inconsistencies result from differences in the morphine compounds (morphine sulfate in the previous versus morphine hydrochloride in the present study), quantification methods for Fas (Northern-blot in the previous versus FCM in the present study), and cytolytic activities in Fas-specific antibodies. However, these all seem to be unlikely for the following reasons. Differences in pharmacological effects between morphine sulfate and morphine hydrochloride have not been shown. Our FCM analysis clearly detected increased Fas expression in the etoposide-treated PBLs (Fig. 2), and FasAb (clone CH-11) that we used increased etoposide-induced PBL apoptosis (Fig. 4). Therefore, the precise reasons for the discrepant results remain unclear.
We should note that there are certain limitations in interpreting the present results. First, our sample sizes may not have been statistically large enough to detect the effects of morphine on PBL apoptosis, although they clearly distinguished the apoptotic events in the etoposide culture from those in the vehicle or morphine cultures. Second, our results were obtained solely from healthy donors. Because cancer pain patients experience various disease-specific immunomodulations, as well as those from long-term opioid treatments, immune cells under such abnormal conditions may behave differently. Nonetheless, the present results at the minimum considerably weaken the simple hypothesis that morphine induces or primes apoptosis in human lymphocytes resulting in undesirable immunosuppression. Further studies with larger numbers of subjects, including clinical patients, are warranted before any definitive conclusions can be made regarding the justifications for opioid therapy.
In conclusion, co-culture with morphine did not influence the surface PS exposure, nuclear degradation, Fas expression, intracellular Bcl-2 expression, or Caspase-3 activation in human PBL that were gated using light scatter-based FCM analyses. The treatment with the cytolytic FasAb had no effects on the apoptotic cell rates in morphine-exposed PBLs. Our results provided no supportive evidence for the previously reported morphine-induced lymphocyte apoptosis in humans in vitro. We believe that the choice of opioids for optimal pain management should not be discouraged until further studies clarify this issue.
The authors thank Brent Bell for reading the manuscript.
1. Lorenzo P, Portolés A Jr, Beneit, JV, et al. Physical dependence to morphine diminishes the interferon response in mice. Immunopharmacology 1987;14:93–9.
2. Starec M, Rouveix B, Sinet M, et al. Immune status and survival of opiate- and cocaine-treated mice infected with Friend virus. J Pharmacol Exp Ther 1991;259:745–50.
3. Freier DO, Fuchs BA. Morphine-induced alterations in thymocyte subpopulations of B6C3F1 mice. J Pharmacol Exp Ther 1993;265:81–8.
4. Sei Y, Yoshimoto K, McIntyre T, et al. Morphine-induced thymic hypoplasia is glucocorticoid-dependent. J Immunol 1991;146:194–8.
5. Fuchs BA, Pruett SB. Morphine induces apoptosis in murine thymocytes in vivo but not in vitro: involvement of both opiate and glucocorticoid receptors. J Pharmacol Exp Ther 1993;266:417–23.
6. Page GG, Blakely WP, Ben-Eliyahu S. Evidence that postoperative pain is a mediator of the tumor-promoting effects of surgery in rats. Pain 2001;90:191–9.
7. Nair MP, Schwartz SA, Polasani R, et al. Immunoregulatory effects of morphine on human lymphocytes. Clin Diagn Lab Immunol 1997;4:127–32.
8. Singhal PC, Kapasi AA, Reddy K, et al. Morphine promotes apoptosis in Jurkat cells. J Leukoc Biol 1999;66:650–8.
9. Yin D, Mufson RA, Wang R, Shi Y. Fas-mediated cell death promoted by opioids. Nature 1999;397:218.
10. Cornelissen M, Philippé J, De Sitter, S, De Ridder, L. Annexin V expression in apoptotic peripheral blood lymphocytes: an electron microscopic evaluation. Apoptosis 2002;7:41–7.
11. Takahashi M, Ohara T, Yamanaka H, et al. The oral-to-intravenous equianalgesic ratio of morphine based on plasma concentrations of morphine and metabolites in advanced cancer patients receiving chronic morphine treatment. Palliat Med 2003;17:673–8.
12. Fadok VA, Voelker DR, Campbell PA, et al. Exposure of phosphatidylserine on the surface of apoptotic lymphocytes triggers specific recognition and removal by macrophages. J Immunol 1992;148:2207–16.
13. Koopman G, Reutelingsperger CP, Kuijten GA, et al. Annexin V for flow cytometric detection of phosphatidylserine expression on B cells undergoing apoptosis. Blood 1994;84:1415–20.
14. Curtin JF, Cotter TG. Live and let die: regulatory mechanisms in Fas-mediated apoptosis. Cell Signal 2003;15:983–92.
15. Chinnaiyan AM, Orth K, O’Rourke K, et al. Molecular ordering of the cell death pathway: Bcl-2 and Bcl-xL function upstream of the CED-3-like apoptotic proteases. J Biol Chem 1996;271:4573–6.
16. Enari M, Sakahira H, Yokoyama H, et al. A caspase-activated DNase that degrades DNA during apoptosis, and its inhibitor ICAD. Nature 1998;391:43–50.
17. Kawakami A, Eguchi K, Matsuoka N, et al. Expression and function of Fas and Fas ligand on peripheral blood lymphocytes in normal subjects. J Lab Clin Med 1998;132:404–13.
18. Aiello A, Delia D, Borrello MG, et al. Flow cytometric detection of the mitochondrial Bcl-2 protein in normal and neoplastic human lymphoid cells. Cytometry 1992;13:502–9.