Human plasma contains low, but physiologically significant, concentrations of morphine that can increase following trauma or exercise. We now demonstrate that normal, human white blood cells (WBC), specifically polymorphonuclear cells, contain and have the ability to synthesize morphine. We also show that WBC express CYP2D6, an enzyme capable of synthesizing morphine from tyramine, norlaudanosoline, and codeine. Significantly, we also show that morphine can be synthesized by another pathway via l-3,4-dihydroxyphenylalanine (l-DOPA). Finally, we show that WBC release morphine into their environment. These studies provide evidence that 1) the synthesis of morphine by various animal tissues is more widespread than previously thought and now includes human immune cells. 2) Moreover, another pathway for morphine synthesis exists, via l-DOPA, demonstrating an intersection between dopamine and morphine pathways. 3) WBC can release morphine into the environment to regulate themselves and other cells, suggesting involvement in autocrine signaling since these cells express the μ3 opiate receptor subtype.
The immune-regulatory effects of opioid chemical messengers are well established (1, 2). Exogenously administered opiates are immunosuppressive, inhibiting both cellular and humoral responses via their cytokine-like effects in both the CNS and the periphery (2). In vitro exposure of human monocytes and granulocytes to morphine results in immunosuppression, as evidenced by a marked reduction in chemotaxis, phagocytosis, and responsiveness of cells to signal molecules (3). Furthermore, receptors for many of the opioid peptides as well as morphine have been shown to be expressed outside the CNS (2). In addition, endogenous morphine-like compounds have been identified in diverse organisms, including humans, and are released in response to surgical, pathogenic, and psychological stress (4, 5, 6). In this regard, because endogenous morphine is constitutively expressed and overexpressed following trauma, we have surmised that this opiate alkaloid normally down-regulates immune, vascular, and neural responsiveness on a basal level. In addition, it also limits microenvironmental noise and postinflammatory immune responsiveness (7, 8, 9).
Our laboratory recently demonstrated that human monocytes and granulocytes express a novel μ opiate receptor, μ3, which is morphine selective but opiate peptide insensitive (10). This G protein-coupled receptor, an alternatively spliced variant of the μ receptor gene, mediates downstream signaling via constitutive NO synthase-derived NO release (10).
Human plasma contains low concentrations of morphine, which can increase following trauma or exercise (4, 5, 6). Potential sources of plasma morphine include the adrenal gland and/or brain. Another important potential source is white blood cells (WBC;3 leukocytes) since morphine is found in plasma constitutively (4). Although the presence of morphine in human plasma has been demonstrated, the source and biosynthetic pathways have not been identified. We therefore undertook this study to 1) determine whether human WBC have the capacity to synthesize morphine; 2) identify the biosynthetic pathways for morphine synthesis in these cells; and 3) determine whether such synthesis can result in modulation of immune activity in a physiologically relevant manner.
Materials and Methods
Human blood was obtained from the Long Island Blood Services (Melville, NY). Human heparinized whole blood was immediately separated using 1-Step Polymorphs (Accurate Chemical and Scientific) gradient medium. Five milliliters of heparinized blood was layered over 5 ml of Polymorphs in a 14-ml round-bottom tube and then centrifuged for 35 min at 500 × g in a swinging bucket rotor at 18°C. After centrifugation, the lower band consisting of polymorphonuclear cells (PMN) were harvested in 14-ml tubes and then washed with 10 ml of PBS (Invitrogen Life Technologies) by centrifugation for 10 min at 400 × g. In addition, residual RBC were lysed using ACK lysing buffer (0.15 M NH4Cl2, 1 mM KHCO3, 0.1 mM Na2EDTA (ph 7.2)) (Current Protocols in Immunology).
A two-way ANOVA was used for statistical analysis after precursor exposure to the cells. Each experiment was performed four times. The mean value was combined with the mean value taken from four other replicates. The SEM represents the variation of the mean of the means. All drugs were purchased from Sigma-Aldrich, except bufuralol, which was purchased from BD Biosciences Clontech. The medium containing the PMN was then separated after and before precursor exposure at varying concentrations for 1 h. Cells were washed and the endogenous morphine content was determined.
The morphine extraction, isolation, and identification and quantification protocols were performed as described elsewhere in great detail for the past 15 years (11). Several blank HPLC purifications were performed between each sample to prevent residual morphine contamination remaining on the column. Furthermore, invertebrate molluscan mantle tissue was run as a negative control, demonstrating a lack of contamination (12). All solutions, medium etc., were also examined for any presence of morphine. The results of these tests revealed a lack of morphine contamination.
Nanoflow electrospray quadrupole time-of-flight mass spectrometry (Q-TOF-MS) offers an effective way of demonstrating endogenous alkaloids in biological tissues (11, 13, 14). Q-TOF-MS was performed on a Q-TOF hybrid system (Q-Tof2, Micromass). One microliter of acetonitrile/water/formic acid (30:69:1, v/v/v) of the PMN HPLC extraction was loaded in a gold-coated borosilicate capillary (Proxeon). Authentic morphine standard was used as a control. This sample was sprayed at a flow rate of ∼30 nl/min, giving extended analysis time in which MS spectra, as well as MS/MS spectra, were obtained. During MS/MS or tandem MS, fragment ions are generated from a selected precursor ion by collision-induced dissociation. The collision energy is typically varied between 20 and 35 eV so that the parent ion is fragmented into a number of different daughter ions, allowing a comparison to be made between the experimental and authentic samples.
CYP2D6 molecular demonstration
Human heparinized whole blood obtained from volunteer blood donors (Long Island Blood Services) was prepared as described above.
Isolation of total RNA
PMN (5 × 106) were pelleted by centrifugation and total RNA was isolated with the RNeasy Mini kit (Qiagen). Pelleted cells were resuspended in 500 μl of RLT buffer and homogenized by passing the lysate five times through a 20-gauge needle fitted to a syringe. The samples were then processed according to the manufacturer’s instructions. In the final step, the RNA was eluted with 50 μl of RNase-free water by centrifugation for 1 min at 10,000 rpm.
First-strand cDNA synthesis was performed using random primers (Invitrogen Life Technologies). One microgram of total RNA was denatured at 95°C and reverse transcribed at 40°C for 1 h using Superscript III RNase H-RT (Invitrogen Life Technologies). Ten microliters of the RT product was added to the PCR mix containing specific primers for the CYP2D6 gene and Platinum TaqDNA polymerase (Invitrogen Life Technologies. The forward primer sequence was 5′-AGGTGTGTCTCGAGGAGCCCATTTGGTA-3′ and reverse primer was 5′-GCAGAAAGCCCGACTCCTCCTTCA-3′. The PCR was denatured at 94°C for 5 min, followed by 40 cycles at 95°C for 1 min, 60°C for 1 min, and 72°C for 1 min and then an extension step cycle at 72°C for 10 min. PCR products were analyzed on a 2% agarose gel (Sigma-Aldrich) stained with ethidium bromide. The expected sizes of the PCR products were 700 bp, 300 bp, and others as described by Zhuge and Yu (17).
Sequencing of the PCR product
The 300-bp bands obtained after running the agarose gel were excised and purified with the Qiaquick gel extraction kit (Qiagen). The PCR product (200 ng) and the forward primer were then sent to Lark Technologies for direct sequencing.
Computer-assisted cell activity analysis
PMN, obtained as described earlier, were also processed for image analysis of cell conformation (18). The morphological measurements of PMN are based on cell area and perimeter determinations by the use of image analysis software (Compix). Form-factor (FF) calculations were performed as previously described (19, 20, 21). The observational area used for measurement determinations and frame grabbing was 0.4 μm in diameter. The computer-assisted image analysis system (Zeiss Axiophot fitted with Nomarski and phase-contrast optics) was the same as previously described (19, 20, 21).
The cells were analyzed for conformational changes indicative of either activation (amoeboid and mobile) or inhibition (round and stationary) (19, 20, 21). The lower the FF number, the longer the perimeter and the more amoeboid the cellular shape. The proportion of activated cells was determined as previously described (19, 20, 21). The activated state (amoeboid conformation) of human and vertebrate cells is correlated to biochemical adhesion molecule alterations (19, 20, 21) as well as cytokine production (19, 20, 21).
All pharmacological agents were purchased from Research Biochemicals or Sigma-Aldrich.
In control (vehicle-exposed) WBC, morphine was identified at a level of 12.33 ± 5.64 pg/million cells ± SEM (Fig. 1). Mass spectrometric analysis demonstrated the molecular mass of the single charged ion of morphine, which is identical with authentic morphine (Fig. 1, top and middle). It is important to note that these cells were extensively washed in serum-free RPMI 1640, limiting any plasma morphine that may be found on the cells. However, it is possible that the cells nonspecifically accumulated morphine from plasma. To determine whether WBC contain morphine due to endogenous synthesis, cells were incubated with specific morphine precursors, including tyramine, previously shown to be present in whole animals (12, 16, 22) and previously shown to increase morphine synthesis in healthy neural tissue (16, 22). Tyramine, norlaudanosoline tetrahydropapaveroline (THP), reticuline, and l-3,4-dihydroxyphenylalanine (l-DOPA) significantly increase WBC morphine concentrations above those found in untreated cells in a concentration-dependent manner (ANOVA test, p < 0.001; Fig. 1, bottom). Morphine concentrations in cells incubated with precursors were 90.25 ± 10.42, 136.04 ± 8.71, and 146.5 ± 12.43 pg/million cells ± SEM after 1 h for THP, reticuline, and l-DOPA, respectively. Furthermore, we found that morphine concentrations increased with exposure to precursors in a concentration-dependent manner (Fig. 1, bottom). However, WBC incubated with the serotonin precursors (23), either tryptophan or 5-hydroxytryptophan, in the range of 1 nM to 10 μM did not exhibit endogenous morphine level increases or decreases, demonstrating a level of specificity in morphine synthesis (data not shown). Regardless of the concentration, these results demonstrate that WBC contain low but physiologically significant quantities of morphine and that exposure of these cells to morphine precursors increases morphine synthesis.
To identify a specific population of WBC capable of synthesizing morphine, we more closely examined PMN. Morphine was found in these cells at the level of 11.2 ± 4.21 pg/million cells ± SEM (Fig. 2). Exposing PMN to morphine precursors including tyramine, at levels previously shown to increase WBC morphine concentrations, resulted in a statistically significant increase in morphine concentrations in PMN (Fig. 2).
To determine whether CYP2D6 is involved in morphine synthesis in PMN, we incubated PMN with tyramine and the CYP2D6 substrate bufuralol and found significantly diminished synthesis of morphine (p < 0.001 compared with precursor augmentation levels; Fig. 2, top). Additionally, the CYP2D6 inhibitor, quinidine, blocked morphine synthesis when tyramine, THP, or codeine was exposed to PMN, further demonstrating CYP2D6 presence and modulation (Fig. 2, middle). In addition, we found that CYP2D6 is expressed in these cells, as exemplified by RT-PCR expression analysis that amplified a 300-bp fragment corresponding to the enzyme in PMN (Fig. 2, bottom right and left). Sequence analysis of this fragment demonstrates 100% homology with human CYP2D6 (Fig. 2, bottom right). These results demonstrate that CYP2D6 is expressed in human PMN and that it is involved in morphine synthesis.
Possible roles for PMN-derived morphine include self-regulation as well as regulation of cells in the local environment or distantly. It was of interest therefore to determine whether morphine found in PMN would also be found in the PMN incubation medium following exposure to morphine precursors (Fig. 3, top). We found that following exposure to morphine precursors, concentrations of morphine detected in the medium increased significantly compared with untreated cells (Fig. 3). Before incubation with PMN, the incubation medium was found to contain morphine at the lower detection limit of the assay (Fig. 3).
To examine a possible physiological role of PMN-derived morphine, we then incubated precursor-treated PMN with other PMN that had been exposed to different experimental protocols and evaluated their activity level via computer-assisted image analysis. Specifically, we aimed to examine the effect of precursor-treated PMN on PMN that had been activated by IL-1β (Fig. 3, bottom). Untreated PMN exhibit an activity level of 7.3 ± 2.1% activated (FF >0.6) compared with 43.4 ± 5.7% for cells treated with IL-1β (2 ng/ml) after 1 h. PMN incubated with l-DOPA (10−6 M; 106 cells) exhibited a 3.7 ± 0.4% level of activation. After washing PMN separately and mixing the populations (l-DOPA treated and IL-1β treated) in a 1:1 ratio, the percentage of activated cells decreased to 12.5 ± 3.7% in the mixed PMN population (Fig. 3, bottom). In performing the same experiment but cotreating the IL-1β cells with naloxone and then mixing them with the l-DOPA-treated PMN, the activity level was 35.2 ± 6.3% activated, indicating that morphine mediated this reduced level of activation since naloxone significantly blocked its action.
In the current report, we demonstrate that morphine is present and can be synthesized in human WBC and PMN from its precursors, l-DOPA, reticuline, THP, and tyramine, in a concentration-dependent manner. In part, this finding provides further evidence for this signaling molecule acting in a hormonal capacity as well as acting within the confines of an autocrine loop, since PMN express the μ3 opiate receptor subtype that is coupled to constitutive NO release (10, 20, 24, 25). We envision the autocrine process to be inhibitory in nature since morphine exposure to monocytes, granulocytes, and platelets results in constitutive NO release, resulting in inactive rounded cells, which do not adhere (3, 7, 8, 26, 27, 28, 29, 30, 31, 32). Taken together, we surmise endogenous morphine normally performs this immune down-regulating function.
The CYP2D6 isoenzyme, found on chromosome 22, is part of the CYP2 family and is expressed in neural, immune, and hepatic tissues (33). CYP is involved with the metabolism of many endogenous compounds such as biogenic amines, steroids, leukotrienes, fatty acids, and PG, where it may generate inactive or active compounds (33). The CYP system is involved in the oxidative metabolism of pharmaceutical compounds as well. For example, anti-arrhythmics, β blockers, antipsychotics, vasodilators, and, as demonstrated in the present report (Fig. 4), analgesics are also substrates for this enzyme (33).
In this report we show that CYP2D6 appears to act at critical steps of the morphine biosynthetic pathway in PMN (12, 34, 35) as others have done with different animal tissues (36, 37, 38). We show that bufuralol, a CYP2D6 substrate, as well as a CYP2D6 inhibitor, diminished morphine synthesis in PMN exposed to tyramine. This provides additional support for the ability of this tissue to synthesize morphine via dopamine since we and others have shown that this enzyme is expressed in PMN (7, 38, 39). Therefore, our data show that a key enzyme family, which is also present in mammalian brain, is expressed in WBC and is capable of synthesizing morphine via tyramine (Fig. 4). These results, taken together with previous studies show that both brain and WBC express both CYP2D6 and the μ3 receptor subtype (10, 34, 39).
The morphine-enhancing effect of l-DOPA and its key position in animals in both the dopamine and morphine biosynthesis pathways suggest that morphine may be involved in a regulatory step (7, 12) (Fig. 2). The observation that l-DOPA can also lead to increased PMN morphine concentrations suggests that an additional pathway for morphine synthesis exists via tyrosine (7). It is important to note that WBC also contain dopaminergic signaling components (40, 41, 42), supporting this hypothesis.
Reticuline and THP have both been shown to be endogenously synthesized in animal tissues (12, 16, 22), supporting their role in the present pathway as precursors of morphine synthesis. Interestingly, each is believed to exert specific pharmacological effects. However, given their presence in the biochemical morphine synthesis pathway their independent actions must be reexamined (12, 15, 16).
We further demonstrate that PMN exposed to morphine precursors eventually release morphine into their environment, which can influence the activity state of the same cells as well as other cells not exposed to the precursors. The mechanism of morphine release is unknown, however. Excited cells may elaborate a factor that specifically stimulates the release of morphine from WBC. However, in cells not exposed to the precursors morphine can still be found in the environment, suggesting that once synthesis starts it is designed to be released since this may only occur at appropriate times, i.e., the presence of high levels of these precursors. Further demonstrating that this is the case in the ability of naloxone to block morphine’s action on activated PMN in the vicinity of morphine precursor-loaded WBC. What, however, is the physiological function of low plasma concentrations of morphine? Perhaps, it provides a basal level that may help down-regulate immune and vascular tissues tonally via NO, preventing their activation by nonspecific microenvironmental noise (43).
In terms of an autocrine role for morphine, endogenously synthesized morphine could act via μ3 receptors expressed on PMN (10). In addition to demonstrating the immunomodulating effect of morphine on PMN at the cellular level, our laboratory recently identified, using expression microarray analysis, that morphine altered expression of RNA in key pathways such as TNF signaling, IL-2 signaling, apoptosis, and oxidative stress response (44). Furthermore, in malignant histiocytosis, which is characterized by hyperactive immune cells, a condition that results in death, the μ3 opiate receptor subtype was not expressed on human immunocytes, resulting in cells which could not be down-regulated by morphine (45). In general, morphine tended to down-regulate proinflammatory gene expression and up-regulate genes associated with signaling and immune down-regulation. These observations support the evidence that suggests morphine’s role in down-regulating tissue excitation via a constitutive NO synthase-derived NO pathway (10). The presence of opiate alkaloids in the circulation and of special opiate receptors on immunocytes, shown in both vertebrates and invertebrates, provides a direct role for these compounds in both autocrine and immunomodulatory pathways.
Dopamine can significantly effect the immune system (reviewed in Ref.46). Depending on the environment (in vitro vs in vivo) and cell type, dopamine has activating and suppressive effects on cytokines such as IL-1β, IL-2, IL-6, TNF, and IFN. How these effects are mediated is unknown; however, our data suggest a potential role for morphine signaling in some of these processes.
In addition to a direct metabolic link, there are alternative hypotheses regarding the interaction between dopamine and morphine. For example, dopamine may be acting via cell surface dopamine receptors, supported by the observation that dopamine receptor antagonists can block morphine-induced immunomodulation (47). The time course for these observed changes is likely much slower than the metabolic link between dopamine and morphine that we have described in this study.
In summary, our data support the coupling of the morphine and dopamine biosynthetic pathways, a finding that has tremendous biomedical significance that crosses many physiological systems. Further studies that more clearly elucidate the potential interplay between these two regulatory chemical messengers and their effects on the immune system are currently under way.
The authors have no financial conflict of interest.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
This work was supported in part by grants from the National Institute of Mental Health (Grant 47392) and National Institute on Drug Abuse (Grant 09010).
Abbreviations used in this paper: WBC, white blood cell; PMN, polymorphonuclear cell; l-DOPA, l-3,4-dihydroxyphenylalanine; FF, form factor; Q-TOF, quadrupole time-of-flight; MS, mass spectrometry.