Dynorphin regulates the phagocytic activity of splenic phagocytes in wall lizards: involvement of a κ-opioid receptor-coupled adenylate-cyclase–cAMP–PKA pathway

Dynorphins are endogenous opioid peptides encoded by a common proenkaphalin-B gene (hereafter referred to as preprodynorphin, PDYN) (Civelli et al., 1985) that is widely expressed in the central and peripheral nervous systems of vertebrates (Cone et al., 1982; Goldsmith et al., 1992; Dores et al., 2004; Shirayama et al., 2004; Alrubaian et al., 2006). As for various peptide hormones, dynorphins are synthesized as large, biologically inert precursor proteins that undergo post-translational modifications to produce biologically active molecules. In mammals, these active products include big-dynorphin, dynorphin A_(1–17), dynorphin A_(1–13), dynorphin A_(1–8), dynorphin B_(1–29), dynorphin B_(1–13) and α-neo-endorphin (Schwarzer, 2009). Cloning and sequence analysis of the preprodynorphin cDNA revealed that it is a highly conserved gene, although some discrepancies have been described in lower ectotherms. In fish and amphibians, the gene encoding PDYN contains an additional opioid sequence for leucine-enkephalin or isoleucine-enkephalin (Dores et al., 2004; Alrubaian et al., 2006) and for methionine-enkephalin (Danielson et al., 2002), respectively. Although the gene encoding PDYN has not been cloned to date in reptiles, different dynorphins such as α-neo-endorphin, dynorphin A_(1–17), dynorphin A_(1–8) and dynorphin B_(1–13) in the lizard Anolis carolinensis and α-neo-endorphin and dynorphin B_(1–13) in turtle and alligator have been demonstrated by Goldsmith and colleagues (Goldsmith et al., 1992) using an heterologous radioimmunoassay. Owing to its higher potency (James et al., 1984) and conserved nature, dynorphin A_(1–17) [dyn A_(1–17)] was used here to examine its immunomodulatory role in the wall lizard H. flaviviridis.

In addition to an analgesic effect (Knoll and Carlezon, 2010), a role for dynorphins in the regulation of immune responses has emerged in recent years (Casellas et al., 1991; Ichinose et al., 1995; Singh and Rai, 2010). Furthermore, functional μ-, δ- and κ-opioid receptors belonging to the G-protein-coupled receptor (GPCR) superfamily (Law et al., 2000) have been demonstrated to be expressed by immune cells (Martin-Kleiner and Gabrilovac, 1997). Interestingly, the same intracellular signaling cascade for different opioid receptors (Lawrence and Bidlack, 1993; Sharp et al., 1996; Wang et al., 2003) and multiple downstream signaling mechanisms for the same opioid receptor (Bohn et al., 2000; Kam et al., 2004) have been reported, depending on which mammalian cell lines were studied. With regard to ectotherms, currently there is only one report exploring the intracellular signaling cascade for β-endorphin action on immune cells (Kumar et al., 2011). In the present in vitro study, we aimed to understand the mechanism of action of receptor-coupled downstream signaling for dyn A_(1–17) in lizard splenic phagocytes.

The adaptive immune system in ectotherms, unlike that of endothermic vertebrates, is poorly developed (Flajnik, 1996). Therefore, innate immunity, the first line of host defense plays a crucial role in protecting ectothermic vertebrates from microbial infections. Phagocytes are an important constituent of the innate immune system. These cells possess a remarkable property to phagocytose and kill microbes. Furthermore, the phagocytic process initiates an array of events, including antigen presentation and secretion of various cytokines, which, in turn, activate the adaptive immune system. Thus, phagocytosis can be considered as a central phenomenon for the activation of the immune system during pathogenic attack. Bearing this in mind, the phagocytic activity of splenic phagocytes was selected as the parameter of study in the present investigation.

Specimens of the Indian wall lizard, Hemidactylus flaviviridis Rüppell 1835, possessing a body mass of 8–10 g (adults) were procured from suburbs of Delhi, India (28°12′ to 28°53′N, 76°50′ to 77°23′E). They were maintained in wooden cages fitted with mesh wire from the top, sides and front for proper air circulation and light. Lizards were acclimated to the laboratory conditions (12 h:12 h light:dark photoperiod, lights on at 07:00 h and room temperature) for one week before experiments. Animals were provided with live house flies as food, and water ad libitum. Only female lizards were used in the present in vitro study, as they show better immune responses than males (Mondal and Rai, 1999). The Institutional Animal Ethics Committee guidelines of the University of Delhi were followed with regard to the maintenance and killing of wall lizards. In order to minimize the effects related to diurnal variation with respect to the expression of opioid receptors, and their affinity to ligand or antagonists, lizards were vivisected between 10:00 and 11:00 h for all experiments.

All regular-use chemicals were of molecular biology grade. They were purchased from Qualigens Fine Chemicals (Mumbai, India), Central Drug House (New Delhi, India), Sisco Research Laboratories (Mumbai, India) and Merck Specialities (Navi Mumbai, India). Dyn A_(1–17), μ-receptor antagonist CTAP (H-D-Phe-Cys-Tyr-D-Trp-Arg-Thr-Pen-Thr-NH₂), δ_1,2-receptor antagonist naltrindole hydrochloride (NTI) and κ-receptor antagonist norbinaltorphimine dihydrochloride (NorBNI) were the generous gifts from the National Institute of Drug Abuse, Bethesda, MA, USA. The cell culture medium RPMI 1640, non-selective opioid receptor blocker naltrexone, adenylate cyclase inhibitor SQ 22536, protein kinase A (PKA) inhibitor H-89, phosphodiesterase inhibitor 3-isobutyl-1-methyl-xanthine (IBMX), cAMP enzyme immunoassay kit and MTT [3-(4,5-dimethyl thiazoyl-2)-2,5-diphenyl tetrazolium bromide] were purchased from Sigma-Aldrich (St Louis, MO, USA). The culture medium was supplemented with 100 μg ml^–1 streptomycin, 100 IU ml^–1 penicillin, 40 μg ml^–1 gentamicin, 5.94 mg ml^–1 HEPES buffer [N-(2-hydroxyethyl) piperazine-N′-(2-ethanesulphonic acid)] and 0.2% sodium bicarbonate. Before use, complete culture medium was made by supplementing 2% heat-inactivated fetal calf serum (FCS, Biological Industries, Kibbutz Beit, Haemek, Israel).

The wall lizards were vivisected by cervical dislocation, their spleens were dissected out, pooled in chilled phosphate-buffered saline (PBS, pH 7.2) and passed through nylon mesh (pore size 90 μm) in ice-cold complete culture medium to prepare the single-cell suspension. The total cell number was adjusted to 10⁶ cells ml^–1. For the preparation of the phagocyte monolayer, 200 μl of splenocyte suspension was flooded on each pre-washed slide. Following incubation for 90 min, the non-adhered cells were washed off with PBS. The viability of the adhered cell population was >98% as assessed by a Trypan Blue exclusion test. All the incubations were performed at 25°C in a humidified chamber or incubator maintained with 5% CO₂.

Yeast (Saccharomyces cerevisiae) cell suspension was prepared by warming commercially available baker's yeast (1.5 mg ml^–1 PBS) at 80°C for 20 min. The heat-killed yeast cell suspension was washed and resuspended in the complete culture medium.

For the phagocytic assay, 400 μl heat-killed yeast cell suspension was flooded on a phagocyte monolayer of each slide. After incubation for 90 min, non-phagocytosed yeast cells were washed off with PBS. Thereafter, the phagocyte monolayer was fixed in methanol, stained with Giemsa and mounted in DPX. Cells were observed under the microscope (Nikon E400, Japan) at a magnification of ×400. Phagocytes engulfing one or more than one yeast cell were considered as ‘positive phagocytes’. These cells possess the extending pseudopodia forming the phagocytic cup(s) to phagocytose the yeast cells (Fig. 1). Phagocytes that were not phagocytosing the yeast cell(s) were regarded as ‘negative phagocytes’. Without any predetermined sequence or scheme, ∼200 phagocytes were counted per slide. The experimenter was blind to the technical details of the slides. The percentage phagocytosis and phagocytic index were calculated using established formulae (Campbell et al., 2003): (1) percentage phagocytosis equals the number of positive phagocytes per 100 phagocytes counted; and (2) the phagocytic index is the average number of yeast cells engulfed by each positive phagocyte multiplied by the percentage phagocytosis.

Splenic phagocytes were treated with varying concentrations of dyn A_(1–17) (10^–12, 10^–11, 10^–10, 10^–9 and 10^–8 mol l^–1) for 30 min. Cells were washed with PBS and processed for the phagocytic assay. A phagocyte monolayer incubated in medium alone for the same duration was considered to function as a control. The concentration range of dyn A_(1–17) used in the present in vitro study was determined after reviewing the literature (Casellas et al., 1991; Szabo et al., 1993), whereas the optimum time required for incubation of phagocytes with dyn A_(1–17) was determined based on pilot experiments.

In order to explore the involvement of opioid and non-opioid receptors in mediating the effect of dyn A_(1–17) on phagocytosis, the non-selective opioid receptor antagonist naltrexone was used along with the most effective concentration of dyn A_(1–17). Splenic phagocytes were incubated with 10^–8 mol l^–1 naltrexone and 10^–9 mol l^–1 dyn A_(1–17), simultaneously, for 30 min. In order to compare the results between different treatment groups, phagocytes were incubated in medium, naltrexone or dyn A_(1–17) alone for 30 min as controls. After incubation, cells were washed and processed for the phagocytic assay.

Splenic phagocytes were incubated with a 10^–8 mol l^–1 concentration of selective μ-, δ- or κ-opioid receptor blocker (CTAP, NTI or NorBNI, respectively) and 10^–9 mol l^–1 dyn A_(1–17), simultaneously, for 30 min. For controls, phagocytes were incubated in medium, dyn A_(1–17) or μ-, δ- or κ-opioid receptor blocker alone for 30 min. Thereafter, phagocytes were washed and processed for the phagocytic assay.

Inhibitors of AC and PKA, SQ 22536 and H-89, respectively, were used to examine the opioid-receptor-coupled downstream signaling mechanism for dyn A_(1–17) action on phagocytosis. Splenic phagocytes were pre-incubated with different concentrations of SQ 22536 [(2.5, 5.0, 7.5 and 10.0)×10^–9 mol l^–1] or H-89 [(25, 50, 75 and 100)×10^–9 mol l^–1] for 30 min. Thereafter, the monolayer was incubated with 10^–9 mol l^–1 dyn A_(1–17) and the above concentrations of SQ 22536 or H-89, simultaneously, for 30 min. Control groups were made in which phagocytes were incubated: (a) for 60 min in medium alone, (b) for 30 min in medium alone, and then for another 30 min with 10^–9 mol l^–1 dyn A_(1–17), (c) for 60 min with 10×10^–9 mol l^–1 SQ 22536 in the case of the AC inhibitor experiment or 100×10^–9 mol l^–1 H-89 for the PKA inhibitor experiment. After incubations, the phagocyte monolayer was washed with PBS and processed for the phagocytic assay. The concentration range of SQ 22536 or H-89 was determined based on the available literature (Roy et al., 2008; Roy and Rai, 2009; Kumar et al., 2011) and pilot experiments.

The intracellular cAMP content in the dyn A_(1–17)-treated splenic phagocytes was estimated following the manufacturer's (Sigma-Aldrich) protocol. Briefly, 100 μl of splenocyte suspension was added to each well of a 96-well tissue-culture plate. After incubation for 90 min, non-adhered cells were washed off, and adhered cells were incubated with 10^–4 mol l^–1 IBMX for 30 min. Furthermore, phagocytes were incubated in PBS (control) or 10^–9 mol l^–1 dyn A_(1–17) for 30 min. The cells were washed, lysed with 0.1 mol l^–1 HCl and centrifuged at 600 g for 10 min at room temperature. The cAMP in the supernatant was estimated using a commercially available enzyme immunoassay cAMP kit.

The method of Mosmann (Mosmann, 1983) for MTT assay was followed to verify the viability of splenic phagocytes after treatment with dyn A_(1–17). In brief, a 100 μl splenocyte suspension was added to a 96-well tissue-culture plate for the preparation of a phagocyte monolayer. The non-adhered cells were washed off with PBS. Adhered phagocytes were incubated with medium alone (control) and different concentrations of dyn A_(1–17) for 30 min. Afterwards, splenic phagocytes were washed and incubated in 100 μl of medium containing 0.5 mg ml^–1 MTT for 2 h. Next, the cells were washed with PBS and permeabilized by adding 20 μl of 0.1% Triton X-100. The reduced product formazan was solubilized in 150 μl of DMSO. Following incubation for 15 min at room temperature, absorbance was measured at 570 nm using an ELISA plate reader (MS 5608A, ECIL, New Delhi, India).

The MTT assay is based on the ability of the mitochondrial enzyme succinate dehydrogenase in live cells to reduce MTT into a membrane-impermeable and insoluble purple crystal called formazan. The number of live cells is directly related to the amount of formazan produced.

Each treatment was performed in triplicate. All the experiments were repeated three times. The data of three independent experiments were pooled, analysed by one-way analysis of variance (ANOVA) followed by a Newman–Keuls multiple range test, and represented as means ± s.e.m. A Student's t-test was used to compare the means of two groups in the case of cAMP estimation.

Dyn A_(1–17) in a concentration-dependent manner inhibited the percentage phagocytosis and phagocytic index (ANOVA, P<0.001; Fig. 2). In comparison with a control, a marked (P<0.05) decrease in phagocytic activity was observed at a 10^–11 mol l^–1 concentration of dyn A_(1–17) [control vs 10^–11 mol l^–1 dyn A_(1–17), P<0.05]. The inhibition of phagocytosis induced by dyn A_(1–17) was further pronounced with an increase of concentration to 10^–9 mol l^–1. The maximum inhibitory effect of dyn A_(1–17) on phagocytosis was noticed at a concentration of 10^–9 mol l^–1. With a further increase in the concentration of dyn A_(1–17) (10^–8 mol l^–1), the inhibition of phagocytic activity was not increased [10^–8 mol l^–1 dyn A_(1–17)vs 10^–9 mol l^–1 dyn A_(1–17), not significant]. Furthermore, a MTT assay indicated that dyn A_(1–17) itself, in the range of the concentrations used, does not influence the cell viability (data not shown).

The non-selective opioid receptor antagonist naltrexone alone did not affect the phagocytic activity of splenic phagocytes (control vs naltrexone 10^–8 mol l^–1, not significant). However, it completely antagonized the inhibitory effect of dyn A_(1–17) on phagocytosis [dyn A_(1–17) 10^–9 mol l^–1vs dyn A_(1–17) 10^–9 mol l^–1 plus naltrexone 10^–8 mol l^–1, P<0.01; Fig. 3].

The μ- and δ-opioid receptor antagonists CTAP (10^–8 mol l^–1) and NTI (10^–8 mol l^–1), respectively, failed to antagonize the inhibitory effect of dyn A_(1–17) (10^–9mol l^–1) on the percentage phagocytosis and phagocytic index [10^–9 mol l^–1 dyn A_(1–17)vs 10^–9 mol l^–1 dyn A_(1–17) plus 10^–8 mol l^–1 CTAP or NTI, not significant, Fig. 4]. By contrast, the selective κ-opioid receptor antagonist NorBNI (10^–8 mol l^–1) completely antagonized the inhibitory effect of dyn A_(1–17) on phagocytosis [10^–9 mol l^–1 dyn A_(1–17)vs 10^–9 mol l^–1 dyn A_(1–17)plus 10^–8 mol l^–1 NorBNI, P<0.01], and the results were comparable to that of a control [10^–9 mol l^–1 dyn A_(1–17) plus 10^–8 mol l^–1 NorBNI vs control, not significant]. However, these antagonists alone did not have any effect on phagocytosis (control vs 10^–8 mol l^–1 CTAP or NTI or NorBNI, not significant).

Inhibitors of AC (SQ 22536) and PKA (H-89) decreased the inhibitory effect of dyn A_(1–17) on phagocytosis in a concentration-dependent manner. A significant decrease in the inhibitory effect of 10^–9 mol l^–1 dyn A_(1–17) was observed when cells were pre-treated with 5×10^–9 mol l^–1 SQ 22536 or 50×10^–9 mol l^–1 H-89 and then with 10^–9 mol l^–1 dyn A_(1–17) plus 5×10^–9 mol l^–1 SQ 22536 or 50×10^–9 mol l^–1 H-89 [10^–9 mol l^–1 dyn A(1–17) vs 10–9 mol l^–1 dyn A_(1–17) plus 5×10^–9 mol l^–1 SQ 22536 or 50×10^–9 mol l^–1 H-89, P<0.05; Fig. 5]. Dyn A_(1–17)-induced inhibition of phagocytosis was completely abolished at higher concentrations of SQ 22536 or H-89 (Fig. 5). At 7.5×10^–9 and 10×10^–9 mol l^–1 concentrations of SQ 22536 or 75×10^–9 and 100×10^–9 mol l^–1 H-89, the percentage phagocytosis and phagocytic index were found to be comparable to that of a control [10^–9 mol l^–1 dyn A_(1–17) plus 7.5 or 10×10^–9 mol l^–1 SQ 22536, or 75 or 100×10^–9 mol l^–1 H-89 vs control, not-significant].

In comparison with a control, a marked (P<0.01) increase in intracellular cAMP content was recorded in splenic phagocytes incubated with dyn A_(1–17) for 30 min (Fig. 6).

Our in vitro study, for the first time in reptiles, has demonstrated the role of dyn A_(1–17) and its downstream signaling cascade in the regulation of an innate immune function – phagocytosis. In the wall lizard H. flaviviridis, dyn A_(1–17) by acting through the κ-opioid receptor inhibited the phagocytosis of yeast cells by splenic phagocytes. This is in contrast to data reported in fish (Singh and Rai, 2010) and mammals (Ichinose et al., 1995), wherein, dyn A_(1–17) stimulated the phagocytosis of yeast cells by fish splenic phagocytes (Singh and Rai, 2010) and latex particles by murine peritoneal macrophages (Ichinose et al., 1995). With respect to receptor-mediated action, dyn A_(1–17) stimulated the phagocytosis through a κ-opioid receptor in fish only. Interestingly, in mammals, the stimulatory effect of dyn A_(1–17) was mediated through a non-opioid receptor as the non-selective opioid receptor blocker naltrexone failed to block the effect of dyn A_(1–17) on phagocytosis (Ichinose et al., 1995). Nevertheless, a specific opioid-receptor-mediated inhibitory effect of dynorphin(s) and its receptor agonist on phagocytosis has been reported in mammals. Dyn A_(1–13) and a κ-opioid receptor agonist were shown to inhibit the phagocytosis of sheep erythrocytes (Casellas et al., 1991) and Candida albicans (Szabo et al., 1993) by murine peritoneal macrophages. In both the cases, the effect was mediated through a κ-opioid receptor. The reason for the diverging results pertaining to the effect of dynorphin(s) on phagocytosis might be ascribed to the involvement of different receptors (opioid or non-opioid) and also to the specificity of target cells to be phagocytosed by the immune cells. Thus, it seems that, despite the species-specific differential effect, dynorphin(s) modulate the phagocytosis largely through opioid receptors – and particularly through κ-opioid receptors.

It is noteworthy that cDNA for opioid receptors have not been cloned to date from any reptile tissue, although they have been cloned from immune as well as non-immune cells of fish (Darlison et al., 1997; Barrallo et al., 1998; Barrallo et al., 2000; Rodriguez et al., 2000; Alvarez et al., 2006; Chadzinska et al., 2009) and mammals (Chuang et al., 1994; Graveriaux et al., 1995; Sedqi et al., 1995; Sedqi et al., 1996; Sharp et al., 1997; Sharp et al., 2000). The sequence analysis of cDNA revealed that these receptors belong to the superfamily of G-protein-coupled receptors (Sharp et al., 1997; Law et al., 2000). Generally, opioid receptors are coupled with the alpha inhibitory subunit (Gα_i) of the G-protein that, upon activation, downregulates the adenylate cyclase activity and, consequently, cAMP production (Lawrence and Bidlack, 1993; Sharp et al., 1996; Wang et al., 2003). However, an accumulation of intracellular cAMP upon the activation of opioid receptors is well established in mammalian immune cells (Aymerich et al., 1998) and cell lines (Martin-Kleiner and Gabrilovac, 1997; Heagy et al., 1999). This suggests the coupling of an opioid receptor with a stimulatory alpha subunit (α_s) of the G-protein. In our earlier in vitro study in the wall lizard, dibutyryl cAMP was shown to inhibit the percentage phagocytosis and phagocytic index at higher concentrations (Roy and Rai, 2004). Also, an increased intracellular cAMP level in cortisol-treated (Roy and Rai, 2009) and melatonin-treated (Roy et al., 2008) splenic phagocytes is reported to be directly correlated with inhibition of phagocytosis in the freshwater fish Channa punctatus. In light of these studies, we speculated that dyn A_(1–17) would have inhibited the phagocytic activity of lizard splenic phagocytes by activating the α_s subunit of the G-protein coupled to κ-opioid receptors and, consequently, by increasing the intracellular cAMP level. This assumption was verified using the inhibitor of adenylate cyclase SQ 22536, which completely averted the inhibitory effect of dyn A_(1–17) on phagocytosis. Also, an increase in the intracellular cAMP level in lizard splenic phagocytes in response to dyn A_(1–17) reconfirmed the assumption. Furthermore, the cAMP-dependent PKA pathway was shown to be involved in mediating dyn-A_(1–17)-induced inhibition of phagocytosis as the effect was fully reversed by the PKA inhibitor H-89. However, an altogether different downstream signaling cascade for dynorphins has been reported by Ichinose and colleagues (Ichinose et al., 1995) in murine peritoneal macrophages. In these cells, dynorphin stimulated the phagocytic activity by means of a non-opioid receptor-mediated calcium signaling cascade.

Although further studies are required across the vertebrates to develop a deeper insight into the evolutionary aspects of the immunomodulatory role of opioids, the dynorphin effect on phagocytosis in reptiles seem to be closer to that in mammals than in fish. The present study in wall lizards reveals the involvement of the AC–cAMP–PKA pathway in transducing the inhibitory effect of dyn A_(1–17) on phagocytosis.

 
• AC

  adenylate cyclase

 
• cAMP

  cyclic adenosine monophosphate

 
• CTAP

  H-^d-Phe-Cys-Tyr-^d-Trp-Arg-Thr-Pen-Thr-NH₂

 
• dyn A_(1–17)

  dynorphin A_(1–17)

 
• IBMX

  3-isobutyl-1-methylxanthine

 
• MTT

  3-(4,5-dimethyl thiazoyl-2)-2,5-diphenyl tetrazolium bromide

 
• NorBNI

  norbinaltorphimine dihydrochloride

 
• NTI

  naltrindole hydrochloride

 
• PKA

  protein kinase A

The authors are grateful to the National Institute of Drug Abuse, Bethesda, MD, USA, for the generous gifts of dyn A_(1–17), CTAP, NTI and NorBNI.