- Email: [email protected]
Effects of midazolam on equine innate immune response: a flow cytometric study
Veterinary Immunology and Immunopathology 95 (2003) 11–19
Effects of midazolam on equine innate immune response: a flow cytometric study C. Massoco, J. Palermo-Neto* Applied Pharmacology and Toxicology Laboratory, Departamento de Patologia, Faculdade de Medicina Veterina´ria e Zootecnia, School of Veterinary Medicine, Universidade de Sa˜o Paulo, Av. Prof. Dr. Orlando Marques de Paiva 87, CEP, 05508-900 Sa˜o Paulo, SP, Brazil Received 15 November 2002; received in revised form 10 March 2003; accepted 16 April 2003
Abstract Benzodiazepines (BDZ) are among the most frequently used class of psychotropic drugs employed in veterinary medicine in Brazil and worldwide due to their anxiolytic, muscle relaxant and anticonvulsant effects [J. Clin. Pharmacol. 33 (1993) 124]. Peripheral benzodiazepine receptor (PBR) sites were described in peripheral organs, endocrine steroidogenic tissues and immune organs and cells. Midazolam is a mixed-type agonist of PBRs. The present study is focused on the effects of midazolam on equine peripheral blood neutrophils, peritoneal macrophages and cortisol levels in plasma. Adult horses were treated with a single dose of midazolam (0.06 or 0.1 mg/kg) or with 0.9% NaCl. Immune cells were collected 24 h after treatment for flow cytometry analysis of Staphylococcus aureus-induced phagocytosis and oxidative burst. Plasma cortisol concentration was measured 30, 90, 180 and 360 min after midazolam treatment. Midazolam induced a dose-dependent reduction on: (1) peripheral blood neutrophil and peritoneal macrophage oxidative burst; (2) the capacity of both peripheral blood neutrophils and peritoneal macrophages to phagocyte S. aureus. Increments on plasma cortisol concentration were not found after 0.06 and 0.1 mg/kg of midazolam. The effects on oxidative burst of neutrophils and macrophages from horses treated with midazolam were interpreted as a consequence of an impairment of S. aureus-induced phagocytosis. The present data suggest that midazolam, most probably acting on PBRs present on equine macrophage and neutrophil membranes, might have changed some mechanisms related to both phagocytosis and oxidative burst. These results support the use of flow cytometry to identify functional properties and dysfunction of equine immune cells. They also confirm the notion that changes in the functional capacity of the immune system may represent an important hazard of exposure to drugs or chemicals. # 2003 Elsevier B.V. All rights reserved. Keywords: Midazolam; Horses; Flow cytometry; Macrophage; Neutrophil; Phagocytosis; Oxidative burst
1. Introduction Benzodiazepines (BDZ) are among the most frequently used class of psychotropic drugs employed in veterinary medicine in Brazil and worldwide due to *
Corresponding author. Tel.: þ55-11-30917957; fax: þ55-11-30917829. E-mail address: [email protected] (J. Palermo-Neto).
their anxiolytic, muscle relaxant and anticonvulsant effects (Ruiz et al., 1993). These BDZ effects are a consequence of its action on high-affinity receptors present in the central nervous system (CNS), namely the GABAA receptors. Nevertheless, besides the central receptors described for BDZ, peripheral-type binding sites (peripheral benzodiazepine receptor, PBR) have also been identified in peripheral organs (Braestrup and Squires, 1977), endocrine steroidogenic
0165-2427/$ – see front matter # 2003 Elsevier B.V. All rights reserved. doi:10.1016/S0165-2427(03)00097-7
12
C. Massoco, J. Palermo-Neto / Veterinary Immunology and Immunopathology 95 (2003) 11–19
tissues (Papadopoulos, 1993) and immune organs and cells, such as polymorphonuclear and mononuclear cells (Zavala and Lenfant, 1984; Ferrarese et al., 1990). Subsequent research on PBR showed they markedly differ from GABAA receptors both physiologically and pharmacologically (Casellas et al., 2002). An altered PBR expression was reported on human polymorphonuclear neutrophils of patients with chronic granulomatous disease (Zavala and Lenfant, 1987). This inherited phagocyte disorder is characterized by lack of superoxide anion production leading to recurrent bacterial and fungal infectious. Treatment of mice with BDZ impaired the capacity of peritoneal and spleen phagocytes to produce reactive oxygen species (ROS), IL-1, TNF-a and IL-6 (Zavala et al., 1990). These key mediators of inflammation are macrophage derived and implicated in host defense mechanisms against pathogens and tumor cells. Indeed, diazepam treatment was reported to decrease macrophage spreading and phagocytosis in mice (Massoco and Palermo-Neto, 1999) and to decrease resistance to Mycobacterium bovis infection in hamsters (Righi et al., 1999). An impaired host resistance to Trichinella spiralis was also reported in rats after diazepam treatment (Schulumpf et al., 1994). Stimulation of PBRs present in adrenal cells was shown to increase glucocorticoid production (Cavallaro et al., 1992). This hormone is known to have potent immunomodulatory properties (McEwen et al., 1997). Surgery is a traumatic procedure associated with transitory alterations in the levels of stress hormones, such as cortisol and b-endorphin (Noel et al., 1972; Cohen et al., 1981). Abdominal wall surgery in rats was associated to an elevation in the density of PBRs, a possible indication that PBRs are sensitive to acute surgical stress (Okun et al., 1988). Altogether, these data strongly suggest a relevant involvement of PBR with both surgical stress and innate immune responses. Peritoneal macrophages and blood neutrophils are known for their phagocytic capabilities and also for their secretory activities. The biologically active substances released by these immune cells are components of important homeostatic processes such as those related to host defense against microorganisms (Speert, 1991) and tumor cells (Rees and Parry, 1991). The cellular oxidative burst that takes place in immune cells during the induction of phagocytosis is known to be a good index of innate immune cell responses
against external or internal challenges to host integrity (Nathan, 1987). Flow cytometric methods have been widely used to study macrophage and neutrophil phagocytic activity in humans (Bassoe et al., 1983a,b; Rothe and Valet, 1990) and animals (Saad and Hageltorn, 1985; Johannisson et al., 1995). Cytometric methods were also considered to be a suitable and more useful methodology to study cellular oxidative burst, compared to other biochemical techniques, since it allows the differentiation between intracellular and extracellular oxidative events (Bass et al., 1983). Moreover, the use of flow cytometric methods precludes the necessity of previous immune cell separation. It has been previously described that such separation procedures are not only time consuming but might also induce per se cellular oxidative burst (Fearon and Collins, 1983). Furthermore, flow cytometry also provides information about differentiation and proportion of cells that phagocyte fluorescent-labeled yeast and bacteria (Johannisson et al., 1995). Although BDZs are used in equine medicine, evidence supporting possible relationships between BDZ treatments, such as those performed with midazolam during anesthetic procedures, and innate immune response are not hitherto investigated. Therefore, the present experiment was designed to analyze, in horses, the effects of a single and acute midazolam treatment on macrophage and neutrophil phagocytosis, and oxidative burst, using a flow cytometric method. Plasma cortisol concentrations were also analyzed in horses after midazolam treatment.
2. Materials and methods 2.1. Animals Healthy adult male and female horses (400–500 kg body weight) of various breeds and aging 6–15 years were used. The horses were individually maintained in stalls (4 m 4 m) under natural conditions of light, temperature and humidity, with access to commercial chow, salt, water, and supplemented with Bermuda grass hay (1–2% of their body weight per day). Animals were housed and used in accordance with the guidelines of the Committee on Care and Use of Animal Resources of the School of Veterinary
C. Massoco, J. Palermo-Neto / Veterinary Immunology and Immunopathology 95 (2003) 11–19
13
Medicine, University of Sa˜ o Paulo, Sa˜ o Paulo. Each animal was used only once.
blue exclusion method. The number of macrophages was adjusted to 2 105 cells/100 ml.
2.2. Drugs and reagents
2.5. Flow cytometry
Midazolam (Dormire1, Crista´ lia do Brasil S/A) was intravenously administered to the horses at two dose levels: 0.06 and 0.1 mg/kg. This dose is in the same range of that used in standard anesthetic protocols for horses (Luna et al., 1992). Saline solution (0.9% NaCl) was used as control. 20 ,70 -Dichlorofluorescein diacetate (DCFH-DA, Molecular Probes) was employed for flow cytometry (25 mM in ethanol). This reagent was kept frozen (20 8C) and protected from light, and dissolved in phosphate-buffered saline (PBS) immediately before use. Staphylococcus aureus (ATCC 25923) labeled in a 5% solution of propidium iodide (PI) (Sigma) was employed to analyze phagocytosis.
A flow cytometer (FACS Calibur, Becton Dickinson Immunocytometry Systems, San Jose, CA, USA) interfaced with a Macintosh G4 computer was used. Data from 10,000 events was collected in list mode and analyzed in Cell Quest (Becton Dickinson Immunocytometry Systems). Discrete cell populations were identified based on their properties on forward scatter (FSC)/side scatter (SSC) plots, mechanically sorted (FAC-Scan, Becton Dickinson Immunocytometry Systems) and evaluated through light microscopy after staining in Giemsa. Data from neutrophils and peritoneal macrophages was collected applying gates that sorted out lymphocyte and monocyte clusters. Fluorescence data was collected on log scale. Green fluorescence from DCFH was measured at 530 30 nm (FL1 detector) and red fluorescence from propidium iodide-labeled S. aureus was measured at 585 42 nm (FL2). PI and DCFH fluorescence were analyzed after fluorescence compensation to correct for any cross over between the PI and DCFH signals.
2.3. Sampling of peripheral blood neutrophils Blood was withdrawn from the jugular vein into lithium heparin Vacutainer tubes (Becton Dickinson) and subsequently kept on ice. Blood samples (100 ml) were directly used to analyze phagocytic activity and oxidative burst of neutrophil as described below. Blood used for cortisol levels was collected from the same vein but stored in Vacutainer tubes in the presence of EDTA. 2.4. Retrieval of peritoneal macrophages Peritoneal cells were obtained through abdominal paracentesis. Horses were individually restrained in open-sided stocks and their ventral abdomen clipped and cleaned; 2% lidocaine was administered intradermally and subcutaneously (5–10 cm caudal to the xyphoid process). A stab incision was made through the skin and abdominal fascia using a scalpel blade; a sterile teat canulla was subsequently bluntly inserted through the rectus abdominis into the peritoneal cavity that allowed peritoneal fluid collection (100–250 ml) made by gravity into sterile 50 ml polypropylene conical centrifuge tubes. These tubes were immediately cooled in ice baths and subsequently centrifuged (250 g for 10 min at 4 8C). Cell pellets were washed three times in sterile, ice-cold phosphate-buffered saline. Cell viability was estimated using the trypan
2.6. Oxidative burst and phagocytosis Quantification of phagocytosis and oxidative burst was estimated by mean PI and DCFH fluorescence cell, respectively. Briefly, 100 ml whole blood or peritoneal macrophage (2 105 cells/100 ml) was mixed with 200 ml of DCFH-DA (0.3 mM) in PBS and 100 ml PI-labeled S. aureus in polypropylene tubes. Samples were incubated under agitation at 37 8C for 20 min. Reactions were stopped by adding 2 ml of cold EDTA solution (3 mM) in order to terminate phagocytosis. After centrifugation (250 g for 10 min), erythrocytes were lysed from all samples with sterile 0.2% NaCl (2 ml per tube) for 20 s. Immediately after that, 1.6% NaCl sterile solution (2 ml) was added to each sample to restore isotonicity. Samples were then centrifuged (250 g for 10 min) and the cell pellets resuspended in 1 ml of cold EDTA (3 mM) for flow cytometry. Direct measurements of mean fluorescence of green and red channels were recorded as oxidative burst and phagocytosis, respectively, as proposed by Hasui et al. (1989). The percentage of phagocytosis
14
C. Massoco, J. Palermo-Neto / Veterinary Immunology and Immunopathology 95 (2003) 11–19
(percentage of neutrophil or macrophage which ingested bacteria) is expressed as the number of neutrophils or macrophages with red fluorescence divided by the total number of cells (multiplied by 100). 2.7. Determination of plasma cortisol concentration Cortisol is the most abundant circulating steroid secreted by horses, being considered a good indicator of adrenocortical function in this animal species (James et al., 1970). Cortisol was determined using commercial kits (Coat-A-Count). This procedure is based on a solidphase radioimmunoassay in which 125 I-labeled cortisol competes for a fixed time with cortisol in the sample for antibody sites. Plasma samples were assayed directly without chemical extraction or purification. The limit of cortisol detection was 0.05 mg/dl and the intra-assay variation coefficient was 6.2%. 2.8. Experimental design Two experiments were done in accordance with GLP standardized protocols. In each experiment, two identical replications were performed to determine the effects of midazolam on innate immune response and on plasma cortisol concentration. In each replication, 16 horses were divided at random into three groups, one control (group C) and two experimental (groups E1 and E2) with 4, 6 and 6 animals, respectively. Animals of groups E1 and E2 were treated
with a single 0.06 and 0.1 mg/kg dose of midazolam, respectively. Animals of group C received an identical volume of 0.9% NaCl (5.0 ml). In the first experiment, blood samples and peritoneal cells were collected from animals 24 h after treatment and were used to determine both oxidative burst and phagocytosis on peripheral blood neutrophils and peritoneal macrophages, as described above. In the second experiment, blood samples were taken from animals of all groups immediately before, and 30, 90, 180 and 360 min after midazolam or control solution for determination of plasma cortisol concentration. 2.9. Statistical analysis Bartlett’s test showed that the data concerning oxidative burst, phagocytosis and plasma cortisol concentration were parametric (P < 0:05). Thus, one-way Analysis of Variance, followed by the Tukey–Kramer test for comparison of cell fluorescence means was used. The StatPac Statistic Analysis Package was used throughout and the level of significance set at P < 0:05.
3. Results 3.1. Oxidative burst and phagocytosis Typical side scatter versus forward scatter cytograms of equine peripheral blood leukocyte (A) and
Fig. 1. Typical side scatter (SSC) and forward scatter (FSC) cytograms of equine peripheral blood leukocyte (PBML) (A) and peritoneal cells (B). In (A) the region R1 corresponds to peripheral blood neutrophils and in (B) the region R2 contains peritoneal macrophages confirmed by sorting and staining.
C. Massoco, J. Palermo-Neto / Veterinary Immunology and Immunopathology 95 (2003) 11–19
15
Table 1 Effects of midazolam on oxidative burst of peripheral blood neutrophils of horses Condition
PBS DCFH S. aureus
Before treatment
After treatment
C
E1
E2
C
E1
E2
2.53 0.1 34.9 13 63.2 19
3.83 0.9 19.8 15 63.2 12
2.57 0.4 27.2 19 56.7 14
4.04 1.5 30.6 12 49.2 14
3.23 0.3 18.5 23 47.8 9.2*
4.25 0.47 28 11.3 36.5 6.2**
C: animals treated with 0.9% NaCl (5.0 ml); E1 and E2: animals treated with 0.06 and 0.1 mg/kg of midazolam, respectively. Gate analysis was set on peripheral blood neutrophil using forward and side scatter gating. Values represent mean fluorescence intensity in gated peripheral blood neutrophils. Data are represented as mean S:D. * P < 0:01 in comparison to animals from group C. ** P < 0:05 in comparison to animals from group E1.
peritoneal fluid derived cells (B), representative of those taken from horses of all groups, revealed two distinct cell populations (R1 and R2), are shown in Fig. 1. Cytograms, such as these described above, were used to analyze individual populations, after mechanical cell sorting, staining in Giemsa and evaluated by light microscopy. Each distinct population had over 98% of neutrophils or macrophages, respectively. As seen, the percentage of neutrophils present on peripheral blood was higher than that of macrophages in the peritoneal fluid. Indeed, R2 cell population in Fig. 1 is less discrete than R1. Tables 1 and 2 show the effects of midazolam on oxidative burst responses of peripheral blood neutrophil and peritoneal macrophage, respectively. Analysis of variance showed significant differences for both peripheral blood neutrophils (F2;15 ¼ 15:2; P < 0:05) and peritoneal macrophages (F2;15 ¼ 17:5; P < 0:05). As expected, fluorescence was not observed in unloaded cells (PBS). An increment in green fluorescence
was observed after DCFH load, allowing satisfactory measurements of midazolam effects on oxidative responses in peripheral blood neutrophils and peritoneal macrophages. Incubation with PI-labeled S. aureus and DCFH induced an increment in the level of green fluorescence in both neutrophils (F2;15 ¼ 1:61; P < 0:05) and macrophages (F2;15 ¼ 3:44; P < 0:05) retrieved from animals of all groups before midazolam treatment; this was most probably a reflex of the increment in cellular oxidative burst that followed S. aureus phagocytosis. This oxidative burst, however, was reduced (P < 0:05) by previous midazolam administration in a dose-dependant manner. It is important to note that midazolam effects were observed only in cells challenged with S. aureus. The values of oxidative burst measured in animals of groups E1 and E2 immediately before midazolam treatment were not different from those in animals of group C taken immediately before, or 24 h after 0.9% NaCl administration.
Table 2 Effects of midazolam on oxidative burst of peritoneal macrophages of horses Condition
PBS DCFH S. aureus
Before treatment
After treatment
C
E1
E2
C
E1
E2
2.85 0.1 137 9.2 223 20
4.56 0.8 125 14 245 13.2
3.62 0.2 119 22 232 13.7
5.32 1.6 130 11.2 238 11.2
3.45 0.9 122 19 178 17*
4.23 0.6 109 12 156 11.2**
C: animals treated with 0.9% NaCl (5.0 ml); E1 and E2: animals treated with 0.06 and 0.1 mg/kg of midazolam, respectively. Gate analysis was set on peritoneal macrophage using forward and side scatter gating. Values represent mean fluorescence intensity in gated peritoneal macrophages. Data are represented as mean S:D. * P < 0:01 in comparison to animals from group C. ** P < 0:05 in comparison to animals from group E1.
16
C. Massoco, J. Palermo-Neto / Veterinary Immunology and Immunopathology 95 (2003) 11–19
Fig. 2. Effect of midazolam on phagocytosis of peripheral blood neutrophil (A) and peritoneal macrophage (B) of horses. Twenty-four hours after 0.9% NaCl or midazolam (0.06 or 0.1 mg/kg), the mixture of 100 ml of heparinized whole blood or peritoneal macrophage (2 105 cells/ 100 ml), 200 ml of 0.3 mM DCFH-DA in PBS and 700 ml of PI-labeled S. aureus was incubated for 30 min at 37 8C. The cells were then analyzed by flow cytometry. Striped and open bars refer to the percentage of phagocytic activity measured before and after midazolam or 0.9% NaCl, respectively. C: animals treated with 0.9% NaCl (5.0 ml); E1 and E2: animals treated with 0.06 and 0.1 mg/kg dose of midazolam, respectively. Data are mean S:D. () P < 0:01 and (#) P < 0:05 compared to animals from groups C and E1, respectively.
Data on phagocytic capacity of peripheral blood neutrophils and peritoneal macrophages are depicted in Fig. 2A and B, respectively. An increment in red fluorescence was induced by S. aureus in both peripheral blood neutrophils and peritoneal macrophages. This increment in fluorescence allowed satisfactory analysis of midazolam effects on phagocytosis. Thus, midazolam was able to decrease phagocytosis in both peripheral blood neutrophils (F2;15 ¼ 24:3; P < 0:05) and peritoneal macrophages (F2;15 ¼ 19:3; P < 0:05) in a dose-dependent manner. As for the oxidative burst response, no differences were found in the capacity of phagocytosis among animals of groups C treated or not with 0.9% NaCl and those measured in animals of groups E1 and E2 before midazolam treatment. Table 3 Effects of midazolam on cortisol plasma concentration of horses Time after treatment (min)
Cortisol (mg/dl)
0 30 90 180 360
6.8 6.2 7.2 6.8 5.3
C
E2
E1
1.5 1.3 1.7 0.8 1.6
6.1 5.9 6.2 5.3 5.4
1.2 0.4 2.4 2.1 1.0
6.7 8.6 7.6 6.1 5.4
2.5 2.8 3.9 4.0 1.9
C: animals treated with 0.9% NaCl (5.0 ml); E1 and E2: animals treated with 0.06 and 0.1 mg/kg of midazolam, respectively. Data are represented by mean S:D.
3.2. Determination of plasma cortisol concentration As observed in Table 3, midazolam (0.06 and 0.1 mg/kg) did not change plasma cortisol concentration of horse. Indeed, no differences were detected among cortisol levels from horses of groups E1 and E2 in relation to those measured in saline-treated animals 30, 90, 180 or 360 min after midazolam treatment (F2;15 ¼ 9:2; P > 0:05).
4. Discussion The present findings demonstrate that midazolam treatment decreased equine innate immune response in a dose-dependent manner. It was observed 24 h after a single (0.06 and 0.1 mg/kg) midazolam treatment: (1) a dose-dependent reduction on peripheral blood neutrophil and peritoneal macrophage oxidative burst after PI-labeled S. aureus challenge; (2) a reduction in the capacity of both peripheral blood neutrophils and peritoneal macrophages to phagocyte PI-labeled S. aureus. Changes in plasma cortisol concentration were not found after two midazolam doses. These results are relevant not only because innate immune response plays a significant role in host defense against microbes (Speert, 1991) and tumor cells (Rees and Parry, 1991) but also because midazolam is frequently
C. Massoco, J. Palermo-Neto / Veterinary Immunology and Immunopathology 95 (2003) 11–19
used in equine anesthetic protocols (Luna et al., 1992). As far as it is of our knowledge this is the first description of midazolam effects on neutrophil and macrophage function in horses. The respiratory burst of neutrophils and macrophages represents a sequence of events, which is essential for the intracellular killing of bacteria. Active oxygen derivatives (superoxide, hydrogen peroxide, hydroxyl radical, singlet oxygen) generated during the above reactions are of great interest in the study of leukocyte oxidative metabolism. The present study demonstrates that the oxidative burst expressed in the generation of hydrogen peroxide by stimulated macrophages and neutrophils can be quantitatively monitored using 20 ,70 -dichlorofluorescein diacetate (DCFH-DA), PI-labeled S. aureus and flow cytometry. Brandt and Keston (1965) originally described a fluorometric method for the measurement of hydrogen peroxide in aqueous solution. This method was based on the oxidation of non-fluorescent DCFH-DA to a fluorescent 20 ,70 -dichlorofluorescein by hydrogen peroxide and peroxidase. The fluorescent reagent is unable to diffuse outside the cell due to its polarity. Therefore, it is possible to detect DCFH-DA oxidation by neutrophils and macrophages using single cell analyses by flow cytometry, as in the present study. It was shown that DCFH-DA oxidation assay is related quantitatively to the concentration of hydrogen peroxide generated (Hirabayashi et al., 1985). Formation and release of hydrogen peroxide in phorbol myristate acetate- and PI-labeled S. aureus-stimulated polymorphonuclear leukocytes was observed at rates that closely paralleled those associated with phagocytosis (Root et al., 1975; Hasui et al., 1989). Therefore, the decrease we now report for the oxidative burst of neutrophils and macrophages of horses after treatment with midazolam might have been a consequence of an impairment of PI-labeled S. aureus-induced phagocytosis. In mice, complex immunological responses, such as humoral responses, are significantly increased by i.p. administration of low doses (1 mg/kg) of peripheral (Ro 5-4864, PK 111 95) and mixed-type (diazepam) PBR ligands. On the other hand, the central BDZ receptor agonist clonazepam showed no effects, nor did its antagonist Ro 15-1788 (Zavala and Lenfant, 1984) suggesting a specific functional role for the PBR in immune cells. Diazepam binding to PBR was found
17
on immune cells such as macrophages (Zavala and Lenfant, 1984) and lymphocytes (Rocca et al., 1993). Activation of PBRs was reported to change the expression and release of several cytokines (Schulumpf et al., 1994) and cell products (Torres et al., 2000), whose effects strongly influence the immune/inflammatory response. Furthermore, PBRs were also described in adrenal cells and have been related to endogenous glucocorticoid production (Papadopoulos, 1993). Thus, diazepam might decrease innate immune response acting directly on immune cells and/or indirectly, through glucocorticoid secretion (Lazzarini et al., 2001). Midazolam was reported to be a mixed-type PBR agonist at doses similar to those used in the present experiment (Taupin et al., 1991). Thus, the findings now been reported on equine innate immunity might be attributable to a possible midazolam effect on PBR present on neutrophils and macrophages. This hypothesis agrees with those already reported by our laboratory for diazepam (Massoco and Palermo-Neto, 1999) and elsewhere for other BDZ ligands (Zavala and Lenfant, 1987). Increments on equine cortisol plasma concentration were not found in the present experiment after midazolam treatments. Thus, an indirect effect for midazolam on neutrophil and macrophage activities via an increment in plasma glucocorticoid concentration seems unlikely. This latter hypothesis could be relevant since PBR stimulation within the inner mitochondrial membrane of adrenal gland cells was reported to increase the biosynthesis of glucocorticoids (Cavallaro et al., 1992). An involvement of PBRs on human immune cell oxidative burst was already reported (Zavala, 1997). The production of oxygen derivatives is dependent on activation of the multimolecular complex NADPHoxidase. A functional link between PBR and NADPHoxidase activation was supported by experiments showing that a monoclonal antibody recognizing the PBR produced a concentration-dependent stimulation of the oxidative burst and enhanced the response to f-Met-Leu-Phe (FMLP) in normal human neutrophils (Zavala et al., 1991). Up-regulation of PBRs has also been associated with maturation of cells such as HL-60, U939 and THP1 into fully competent phagocytes, suggesting that these processes may be interconnected (Canat et al., 1993; Ishiguro et al., 1987). Therefore, it seems reasonable to suggest that midazolam, most
18
C. Massoco, J. Palermo-Neto / Veterinary Immunology and Immunopathology 95 (2003) 11–19
probably acting on PBRs present on macrophage and neutrophil membranes, might have changed some relevant mechanisms related to phagocytosis in these cells, decreasing their oxidative burst. Among other mechanisms, a possible midazolam effect on Ca2þ mobilization emerges as a reliable candidate. Ca2þ mobilization was reported to be critical not only to PBR-mediated responses (Mestre et al., 1985) but also to both phagocytosis (Bengtsson et al., 1993) and NADPH-oxidase activation (Hamilton and Adams, 1987). The flow cytometric method we describe here has been successfully employed to analyze the effects of midazolam on some macrophage and neutrophil functions. Particularly relevant in the context of the present experiment was the fact that this methodology allowed the discrimination between intracellular and extracellular oxidative events, therefore generating data regardless of the absolute cell number. It was possible to identify and evaluate the response of discrete cellular populations without prior separation, which is time consuming and able to alter cell function. Our present data agrees with that reported elsewhere (Raidal et al., 1998) and strongly suggests the use of flow cytometry to identify functional properties of immune cells from both blood and fluid withdrawn from peritoneal cavity of horses. Furthermore, it raises concerns about the safety of using midazolam for anesthetic procedures in horses, particularly in immunocompromised animals. They also support the notion that changes in the functional capacity of the immune system may represent an important hazard of exposure to drugs or chemicals.
Acknowledgements This research, which is part of the Ph.D. thesis by Cristina Massoco to the Department of Pathology of the School of Veterinary Medicine of the University of Sa˜ o Paulo, was sponsored by FAPESP (99/04228-7) and CNPQ (149318/1999). The authors would like to thank the Regimento da Cavalaria ‘‘9 de Julho’’ (Sa˜ o Paulo, SP) that supplied the horses and to MS Maria Aparecida Dalboni and Dr. Miguel Cendoroglo (Fundac¸a˜ o Oswaldo Ramos-UNIFESP) for providing us with PI-labeled Staphylococcus aureus and for technical assistance with the flow cytometer.
References Bass, D.A., Parce, J.W., Dechatelet, L.R., Szejda, P., Seeds, M.C., Thomas, M., 1983. Flow cytometric studies of oxidative produce formation by neutrophils: a graded response to membrane stimulation. J. Immunol. 130, 1910–1917. Bassoe, C.F., Laerum, O.D., Solberg, C.O., Haneberg, B., 1983a. Phagocytosis of bacteria by human leucocytes measured by flow cytometry. Proc. Soc. Exp. Biol. Med. 174, 182–186. Bassoe, C.F., Laerum, O.D., Glette, F., Hopen, G., Haneberg, B., Solberg, C.O., 1983b. Simultaneous measurement of phagocytosis and phagosomal pH by flow Cytometry: role of polymorphonuclear neutrophilic leukocyte granules in phagosome acidification. Cytometry 4, 254–262. Bengtsson, T., Jaconi, M.E.E., Gustafson, M., Magnusson, K.E., Theler, J.M., Lew, D.P., Stendhal, O., 1993. Actin dynamic in human neutrophils during adhesion and phagocytosis is controlled by changes in intracellular free calcium. Eur. J. Cell. Biol. 62, 49–58. Braestrup, C., Squires, R.F., 1977. Specific benzodiazepine receptors in rat brain characterizatized by high-affinity [3 H] diazepam binding. Proc. Natl. Acad. Sci. U.S.A. 74, 3805–3809. Brandt, R., Keston, A.S., 1965. Synthesis of diacetyldichlorofluorescein: a stable reagent for fluorometric analysis. Anal. Biochem. 11, 6. Canat, X., Guillaumont, A., Bouaboula, M., Poinot-Chazel, C., Derocq, J.M., Carayon, P., Le Fur, G., Casellas, P., 1993. Peripheral benzodiazepine receptor modulation with phagocyte differentiation. Biochem. Pharmacol. 46, 551–554. Casellas, P., Galiegue, S., Basile, A.S., 2002. Peripheral benzodiazepine receptors and mitochondrial function. Neurochem. Int. 40, 475–486. Cavallaro, S., Korneyev, A., Guidotti, A., Costa, E., 1992. DBIprocessing products, acting at the mitochondrial DBI receptor, mediate ACTH-induced steroidogenesis in rat adrenal. Proc. Natl. Acad. Sci. U.S.A. 89, 10598–10602. Cohen, M., Pickar, D., Dubois, M., Roth, Y.F., Naber, D., Bunney Jr., W.E., 1981. Surgical stress and endorphins (letter). Lancet 1, 213–214. Fearon, D.T., Collins, L.A., 1983. Increased expression of C3b receptors on polymorphonuclear leukocytes induced by chemotactic factors and purification procedures. J. Immunol. 130, 370–375. Ferrarese, C., Appolonio, I., Frigo, M., Perego, M., Pierpaoli, C., Trabucchi, M., Frattola, L., 1990. Characterization of peripheral benzodiazepine receptors in human blood mononuclear cells. Neuropharmacology 29, 375–378. Hamilton, T.A., Adams, D.O., 1987. Molecular mechanism of signal transduction in macrophages. Immunol. Today 5, 151–158. Hasui, M., Hirabayashi, Y., Kobayashi, Y., 1989. Simultaneous measurement by flow cytometry of phagocytosis and hydrogen peroxide production of neutrophils in whole blood. J. Immunol. Methods 117, 53–58. Hirabayashi, Y., Taniuchi, S., Kobayashi, Y., 1985. A quantitative assay of oxidative metabolism by neutrophils in whole blood using flow cytometry. J. Immunol. Methods 82, 253–259.
C. Massoco, J. Palermo-Neto / Veterinary Immunology and Immunopathology 95 (2003) 11–19 Ishiguro, K., Taft, W.C., de Lorenzo, R.J., Sartorelli, A.C., 1987. The role of benzodiazepine receptors in the induction of differentiation of HL-60 leukemia cells by benzodiazepine and purines. J. Cell. Physiol. 131, 226–234. James, V.H.T., Horner, M.W., Moss, M.S., Rippon, A.E., 1970. Adrenocortial function in the horse. J. Endocrinol. 48, 319–335. Johannisson, A., Gro-Ndahl, G., Demmers, S., Jensen-Waern, M., 1995. Flow-cytometric studies on the phagocytic capacities of equine neutrophils. Acta Vet. Scand. 36, 553–562. Lazzarini, R., Malucelli, B.E., Palermo-Neto, J., 2001. Reduction of acute inflammation in rats by diazepam: role of peripheral benzodiazepine receptors and corticosterone. Immunopharmacol. Immunotoxicol. 23, 253–265. Luna, S.P., Massone, F., Castro, G.B., Fantoni, D.T., Hussni, C.A., Aguiar, A.J., 1992. A combination of methotrimeprazine, midazolam and guaiphenesin, with and without ketamine in an anaesthetic procedure of horses. Vet. Rec. 131, 33–35. Massoco, C.O., Palermo-Neto, J., 1999. Diazepam effects on peritoneal macrophage activity and corticosterone serum levels in Balb/c mice. Life Sci. 65, 2157–2165. McEwen, B.S., Biron, C.A., Brunson, K.W., Bullock, K., Chambers, W.H., Dhabhar, F.S., Goldfarb, R.H., Kitson, R.P., Miller, A.H., Spencer, R.L., Weiss, J.M., 1997. The role of adrenocorticoids as modulators of immune function in health and disease: neural, endocrine and immune interaction. Brain Res. Rev. 23, 79–133. Mestre, M., Carrior, T., Belin, C., Uzan, A., Renault, C., Dubroeucq, M.C., Gueremy, C., Doble, A., Le Fur, G., 1985. Electrophysiological and pharmacological evidence that peripheral type benzodiazepine receptors are coupled to calcium channels in the heart. Life Sci. 36, 391–400. Nathan, C.F., 1987. Neutrophil activation on biological surfaces: massive secretion of hydrogen peroxide in response to products of macrophage and lymphocytes. J. Clin. Invest. 80, 1550– 1560. Noel, G.L., Suh, H.K., Stone, J.C., Frantz, A.G., 1972. Human prolactin and growth hormone release during surgery and other conditions of stress. J. Clin. Endocrinol. Metab. 35, 840–851. Okun, F., Weizman, R., Katz, Y., Bomzon, A., Youdim, M.B.H., Gavish, M., 1988. Increase in central and peripheral benzodiazepine receptors following surgery. Brain Res. 458, 31–36. Papadopoulos, V., 1993. Peripheral-type benzodiazepine/diazepam binding inhibitor receptor: biological role in steroidogenic cell function. Endocr. Rev. 14, 222–240. Raidal, S.L., Bailey, G.D., Love, D.N., 1998. Flow cytometric determination of oxidative burst activity of equine peripheral blood and bronchoalveolar lavage-derived leukocytes. Vet. J. 156, 117–126. Rees, R.C., Parry, H., 1991. Macrophages in tumor immunity. In: Lewis, C.E., McGee, J.O’D. (Eds.), The Macrophage. IRL Press, New York, pp. 3–423.
19
Righi, D.A., Pinheiro, S.R., Guerra, J.L., Palermo-Neto, J., 1999. Diazepam effects on Mycobacterium bovis-induced infection in hamsters. Braz. J. Med. Biol. Res. 32, 1145–1153. Rocca, P., Bellone, G., Benna, P., Bergamasco, B., Ravizza, L., Ferrero, P., 1993. Peripheral-type benzodiazepine receptors and diazepam binding inhibitor-like immunoreactivity distribution in human peripheral blood mononuclear cells. Immunopharmacology 25, 163–178. Root, R.K., Metcalf, J., Oshino, N., Chance, B., 1975. H2O2 release from human granulocytes during phagocytosis. J. Clin. Invest. 55, 945. Rothe, G., Valet, G., 1990. Flow cytometric analysis of respiratory burst activity in phagocytes with hidroethidine and 20 ,70 dichlorofluorescein. J. Leuk. Biol. 47, 440–448. Ruiz, I., Offermanns, J., Lanctot, K.L., Busto, U., 1993. Comparative study on benzodiazepine use in Canada and Chile. J. Clin. Pharmacol. 33, 124–125. Saad, A.M., Hageltorn, M., 1985. Flow cytometric characterization of bovine blood neutrophil Phagocytosis of fluorescent bacteria and zymosan particles. Acta Vet. Scand. 26, 289–307. Schulumpf, M., Lichtensteiger, W., Loveren, H., 1994. Impaired host resistance to Trichinella spiralis as a consequence of prenatal treatment of rats with diazepam. Toxicology 94, 223– 229. Speert, D.P., 1991. Macrophages in bacterial infection. In: Lewis, C.E., McGee, J.O’D. (Eds.), The Macrophage. IRL Press, New York, pp. 3–423. Taupin, V., Jayais, P., Descamps-Latscha, B., Cazala, J.B., Barrier, G., Bach, J.F., Zavala, F., 1991. Benzodiazepine anesthesia in humans modulates the interleukin-1 beta, tumor necrosis factoralpha and interleukin-6 responses of blood monocytes. J. Neuroimmunomodul. 35, 13–19. Torres, S.R., Fro¨ de, T.S., Nardi, G.M., Vita, N., Reeb, R., Ferrara, P., Ribeiro-do-Vale, R.M., Farges, R.C., 2000. Anti-inflammatory effects of peripheral benzodiazepine receptor ligands in two mouse models of inflammation. Eur. J. Pharmacol. 408, 199–211. Zavala, F., Lenfant, M., 1984. Interactions of benzodiazepines with mouse macrophages. Eur. J. Pharmacol. 106, 561–566. Zavala, F., Lenfant, M., 1987. Peripheral benzodiazepine enhance the respiratory burst of macrophage-like P388D1 cells stimulated by arachidonic acid. Int. Immunopharmacol. 9, 269–274 [Eur. J. Pharmacol. 106, 561–566]. Zavala, F., Taupin, V., Descamps-Latscha, B., 1990. In vivo treatment with benzodiazepine inhibits murine phagocyte oxidative metabolism and production of interleukin-1, tumor necrosis factor and interleukin-6. J. Pharmacol. Exp. Ther. 255, 442–450. Zavala, F., Masson, A., Brys, L., de Baetselier, P., DescampsLatscha, B., 1991. A monoclonal antibody against peripheral benzodiazepine receptor activates the human neutrophil NADPHoxidase. Biochem. Biophys. Res. Commun. 176, 1577–1583.
Effects of midazolam on equine innate immune response: a flow cytometric study C. Massoco, J. Palermo-Neto* Applied Pharmacology and Toxicology Laboratory, Departamento de Patologia, Faculdade de Medicina Veterina´ria e Zootecnia, School of Veterinary Medicine, Universidade de Sa˜o Paulo, Av. Prof. Dr. Orlando Marques de Paiva 87, CEP, 05508-900 Sa˜o Paulo, SP, Brazil Received 15 November 2002; received in revised form 10 March 2003; accepted 16 April 2003
Abstract Benzodiazepines (BDZ) are among the most frequently used class of psychotropic drugs employed in veterinary medicine in Brazil and worldwide due to their anxiolytic, muscle relaxant and anticonvulsant effects [J. Clin. Pharmacol. 33 (1993) 124]. Peripheral benzodiazepine receptor (PBR) sites were described in peripheral organs, endocrine steroidogenic tissues and immune organs and cells. Midazolam is a mixed-type agonist of PBRs. The present study is focused on the effects of midazolam on equine peripheral blood neutrophils, peritoneal macrophages and cortisol levels in plasma. Adult horses were treated with a single dose of midazolam (0.06 or 0.1 mg/kg) or with 0.9% NaCl. Immune cells were collected 24 h after treatment for flow cytometry analysis of Staphylococcus aureus-induced phagocytosis and oxidative burst. Plasma cortisol concentration was measured 30, 90, 180 and 360 min after midazolam treatment. Midazolam induced a dose-dependent reduction on: (1) peripheral blood neutrophil and peritoneal macrophage oxidative burst; (2) the capacity of both peripheral blood neutrophils and peritoneal macrophages to phagocyte S. aureus. Increments on plasma cortisol concentration were not found after 0.06 and 0.1 mg/kg of midazolam. The effects on oxidative burst of neutrophils and macrophages from horses treated with midazolam were interpreted as a consequence of an impairment of S. aureus-induced phagocytosis. The present data suggest that midazolam, most probably acting on PBRs present on equine macrophage and neutrophil membranes, might have changed some mechanisms related to both phagocytosis and oxidative burst. These results support the use of flow cytometry to identify functional properties and dysfunction of equine immune cells. They also confirm the notion that changes in the functional capacity of the immune system may represent an important hazard of exposure to drugs or chemicals. # 2003 Elsevier B.V. All rights reserved. Keywords: Midazolam; Horses; Flow cytometry; Macrophage; Neutrophil; Phagocytosis; Oxidative burst
1. Introduction Benzodiazepines (BDZ) are among the most frequently used class of psychotropic drugs employed in veterinary medicine in Brazil and worldwide due to *
Corresponding author. Tel.: þ55-11-30917957; fax: þ55-11-30917829. E-mail address: [email protected] (J. Palermo-Neto).
their anxiolytic, muscle relaxant and anticonvulsant effects (Ruiz et al., 1993). These BDZ effects are a consequence of its action on high-affinity receptors present in the central nervous system (CNS), namely the GABAA receptors. Nevertheless, besides the central receptors described for BDZ, peripheral-type binding sites (peripheral benzodiazepine receptor, PBR) have also been identified in peripheral organs (Braestrup and Squires, 1977), endocrine steroidogenic
0165-2427/$ – see front matter # 2003 Elsevier B.V. All rights reserved. doi:10.1016/S0165-2427(03)00097-7
12
C. Massoco, J. Palermo-Neto / Veterinary Immunology and Immunopathology 95 (2003) 11–19
tissues (Papadopoulos, 1993) and immune organs and cells, such as polymorphonuclear and mononuclear cells (Zavala and Lenfant, 1984; Ferrarese et al., 1990). Subsequent research on PBR showed they markedly differ from GABAA receptors both physiologically and pharmacologically (Casellas et al., 2002). An altered PBR expression was reported on human polymorphonuclear neutrophils of patients with chronic granulomatous disease (Zavala and Lenfant, 1987). This inherited phagocyte disorder is characterized by lack of superoxide anion production leading to recurrent bacterial and fungal infectious. Treatment of mice with BDZ impaired the capacity of peritoneal and spleen phagocytes to produce reactive oxygen species (ROS), IL-1, TNF-a and IL-6 (Zavala et al., 1990). These key mediators of inflammation are macrophage derived and implicated in host defense mechanisms against pathogens and tumor cells. Indeed, diazepam treatment was reported to decrease macrophage spreading and phagocytosis in mice (Massoco and Palermo-Neto, 1999) and to decrease resistance to Mycobacterium bovis infection in hamsters (Righi et al., 1999). An impaired host resistance to Trichinella spiralis was also reported in rats after diazepam treatment (Schulumpf et al., 1994). Stimulation of PBRs present in adrenal cells was shown to increase glucocorticoid production (Cavallaro et al., 1992). This hormone is known to have potent immunomodulatory properties (McEwen et al., 1997). Surgery is a traumatic procedure associated with transitory alterations in the levels of stress hormones, such as cortisol and b-endorphin (Noel et al., 1972; Cohen et al., 1981). Abdominal wall surgery in rats was associated to an elevation in the density of PBRs, a possible indication that PBRs are sensitive to acute surgical stress (Okun et al., 1988). Altogether, these data strongly suggest a relevant involvement of PBR with both surgical stress and innate immune responses. Peritoneal macrophages and blood neutrophils are known for their phagocytic capabilities and also for their secretory activities. The biologically active substances released by these immune cells are components of important homeostatic processes such as those related to host defense against microorganisms (Speert, 1991) and tumor cells (Rees and Parry, 1991). The cellular oxidative burst that takes place in immune cells during the induction of phagocytosis is known to be a good index of innate immune cell responses
against external or internal challenges to host integrity (Nathan, 1987). Flow cytometric methods have been widely used to study macrophage and neutrophil phagocytic activity in humans (Bassoe et al., 1983a,b; Rothe and Valet, 1990) and animals (Saad and Hageltorn, 1985; Johannisson et al., 1995). Cytometric methods were also considered to be a suitable and more useful methodology to study cellular oxidative burst, compared to other biochemical techniques, since it allows the differentiation between intracellular and extracellular oxidative events (Bass et al., 1983). Moreover, the use of flow cytometric methods precludes the necessity of previous immune cell separation. It has been previously described that such separation procedures are not only time consuming but might also induce per se cellular oxidative burst (Fearon and Collins, 1983). Furthermore, flow cytometry also provides information about differentiation and proportion of cells that phagocyte fluorescent-labeled yeast and bacteria (Johannisson et al., 1995). Although BDZs are used in equine medicine, evidence supporting possible relationships between BDZ treatments, such as those performed with midazolam during anesthetic procedures, and innate immune response are not hitherto investigated. Therefore, the present experiment was designed to analyze, in horses, the effects of a single and acute midazolam treatment on macrophage and neutrophil phagocytosis, and oxidative burst, using a flow cytometric method. Plasma cortisol concentrations were also analyzed in horses after midazolam treatment.
2. Materials and methods 2.1. Animals Healthy adult male and female horses (400–500 kg body weight) of various breeds and aging 6–15 years were used. The horses were individually maintained in stalls (4 m 4 m) under natural conditions of light, temperature and humidity, with access to commercial chow, salt, water, and supplemented with Bermuda grass hay (1–2% of their body weight per day). Animals were housed and used in accordance with the guidelines of the Committee on Care and Use of Animal Resources of the School of Veterinary
C. Massoco, J. Palermo-Neto / Veterinary Immunology and Immunopathology 95 (2003) 11–19
13
Medicine, University of Sa˜ o Paulo, Sa˜ o Paulo. Each animal was used only once.
blue exclusion method. The number of macrophages was adjusted to 2 105 cells/100 ml.
2.2. Drugs and reagents
2.5. Flow cytometry
Midazolam (Dormire1, Crista´ lia do Brasil S/A) was intravenously administered to the horses at two dose levels: 0.06 and 0.1 mg/kg. This dose is in the same range of that used in standard anesthetic protocols for horses (Luna et al., 1992). Saline solution (0.9% NaCl) was used as control. 20 ,70 -Dichlorofluorescein diacetate (DCFH-DA, Molecular Probes) was employed for flow cytometry (25 mM in ethanol). This reagent was kept frozen (20 8C) and protected from light, and dissolved in phosphate-buffered saline (PBS) immediately before use. Staphylococcus aureus (ATCC 25923) labeled in a 5% solution of propidium iodide (PI) (Sigma) was employed to analyze phagocytosis.
A flow cytometer (FACS Calibur, Becton Dickinson Immunocytometry Systems, San Jose, CA, USA) interfaced with a Macintosh G4 computer was used. Data from 10,000 events was collected in list mode and analyzed in Cell Quest (Becton Dickinson Immunocytometry Systems). Discrete cell populations were identified based on their properties on forward scatter (FSC)/side scatter (SSC) plots, mechanically sorted (FAC-Scan, Becton Dickinson Immunocytometry Systems) and evaluated through light microscopy after staining in Giemsa. Data from neutrophils and peritoneal macrophages was collected applying gates that sorted out lymphocyte and monocyte clusters. Fluorescence data was collected on log scale. Green fluorescence from DCFH was measured at 530 30 nm (FL1 detector) and red fluorescence from propidium iodide-labeled S. aureus was measured at 585 42 nm (FL2). PI and DCFH fluorescence were analyzed after fluorescence compensation to correct for any cross over between the PI and DCFH signals.
2.3. Sampling of peripheral blood neutrophils Blood was withdrawn from the jugular vein into lithium heparin Vacutainer tubes (Becton Dickinson) and subsequently kept on ice. Blood samples (100 ml) were directly used to analyze phagocytic activity and oxidative burst of neutrophil as described below. Blood used for cortisol levels was collected from the same vein but stored in Vacutainer tubes in the presence of EDTA. 2.4. Retrieval of peritoneal macrophages Peritoneal cells were obtained through abdominal paracentesis. Horses were individually restrained in open-sided stocks and their ventral abdomen clipped and cleaned; 2% lidocaine was administered intradermally and subcutaneously (5–10 cm caudal to the xyphoid process). A stab incision was made through the skin and abdominal fascia using a scalpel blade; a sterile teat canulla was subsequently bluntly inserted through the rectus abdominis into the peritoneal cavity that allowed peritoneal fluid collection (100–250 ml) made by gravity into sterile 50 ml polypropylene conical centrifuge tubes. These tubes were immediately cooled in ice baths and subsequently centrifuged (250 g for 10 min at 4 8C). Cell pellets were washed three times in sterile, ice-cold phosphate-buffered saline. Cell viability was estimated using the trypan
2.6. Oxidative burst and phagocytosis Quantification of phagocytosis and oxidative burst was estimated by mean PI and DCFH fluorescence cell, respectively. Briefly, 100 ml whole blood or peritoneal macrophage (2 105 cells/100 ml) was mixed with 200 ml of DCFH-DA (0.3 mM) in PBS and 100 ml PI-labeled S. aureus in polypropylene tubes. Samples were incubated under agitation at 37 8C for 20 min. Reactions were stopped by adding 2 ml of cold EDTA solution (3 mM) in order to terminate phagocytosis. After centrifugation (250 g for 10 min), erythrocytes were lysed from all samples with sterile 0.2% NaCl (2 ml per tube) for 20 s. Immediately after that, 1.6% NaCl sterile solution (2 ml) was added to each sample to restore isotonicity. Samples were then centrifuged (250 g for 10 min) and the cell pellets resuspended in 1 ml of cold EDTA (3 mM) for flow cytometry. Direct measurements of mean fluorescence of green and red channels were recorded as oxidative burst and phagocytosis, respectively, as proposed by Hasui et al. (1989). The percentage of phagocytosis
14
C. Massoco, J. Palermo-Neto / Veterinary Immunology and Immunopathology 95 (2003) 11–19
(percentage of neutrophil or macrophage which ingested bacteria) is expressed as the number of neutrophils or macrophages with red fluorescence divided by the total number of cells (multiplied by 100). 2.7. Determination of plasma cortisol concentration Cortisol is the most abundant circulating steroid secreted by horses, being considered a good indicator of adrenocortical function in this animal species (James et al., 1970). Cortisol was determined using commercial kits (Coat-A-Count). This procedure is based on a solidphase radioimmunoassay in which 125 I-labeled cortisol competes for a fixed time with cortisol in the sample for antibody sites. Plasma samples were assayed directly without chemical extraction or purification. The limit of cortisol detection was 0.05 mg/dl and the intra-assay variation coefficient was 6.2%. 2.8. Experimental design Two experiments were done in accordance with GLP standardized protocols. In each experiment, two identical replications were performed to determine the effects of midazolam on innate immune response and on plasma cortisol concentration. In each replication, 16 horses were divided at random into three groups, one control (group C) and two experimental (groups E1 and E2) with 4, 6 and 6 animals, respectively. Animals of groups E1 and E2 were treated
with a single 0.06 and 0.1 mg/kg dose of midazolam, respectively. Animals of group C received an identical volume of 0.9% NaCl (5.0 ml). In the first experiment, blood samples and peritoneal cells were collected from animals 24 h after treatment and were used to determine both oxidative burst and phagocytosis on peripheral blood neutrophils and peritoneal macrophages, as described above. In the second experiment, blood samples were taken from animals of all groups immediately before, and 30, 90, 180 and 360 min after midazolam or control solution for determination of plasma cortisol concentration. 2.9. Statistical analysis Bartlett’s test showed that the data concerning oxidative burst, phagocytosis and plasma cortisol concentration were parametric (P < 0:05). Thus, one-way Analysis of Variance, followed by the Tukey–Kramer test for comparison of cell fluorescence means was used. The StatPac Statistic Analysis Package was used throughout and the level of significance set at P < 0:05.
3. Results 3.1. Oxidative burst and phagocytosis Typical side scatter versus forward scatter cytograms of equine peripheral blood leukocyte (A) and
Fig. 1. Typical side scatter (SSC) and forward scatter (FSC) cytograms of equine peripheral blood leukocyte (PBML) (A) and peritoneal cells (B). In (A) the region R1 corresponds to peripheral blood neutrophils and in (B) the region R2 contains peritoneal macrophages confirmed by sorting and staining.
C. Massoco, J. Palermo-Neto / Veterinary Immunology and Immunopathology 95 (2003) 11–19
15
Table 1 Effects of midazolam on oxidative burst of peripheral blood neutrophils of horses Condition
PBS DCFH S. aureus
Before treatment
After treatment
C
E1
E2
C
E1
E2
2.53 0.1 34.9 13 63.2 19
3.83 0.9 19.8 15 63.2 12
2.57 0.4 27.2 19 56.7 14
4.04 1.5 30.6 12 49.2 14
3.23 0.3 18.5 23 47.8 9.2*
4.25 0.47 28 11.3 36.5 6.2**
C: animals treated with 0.9% NaCl (5.0 ml); E1 and E2: animals treated with 0.06 and 0.1 mg/kg of midazolam, respectively. Gate analysis was set on peripheral blood neutrophil using forward and side scatter gating. Values represent mean fluorescence intensity in gated peripheral blood neutrophils. Data are represented as mean S:D. * P < 0:01 in comparison to animals from group C. ** P < 0:05 in comparison to animals from group E1.
peritoneal fluid derived cells (B), representative of those taken from horses of all groups, revealed two distinct cell populations (R1 and R2), are shown in Fig. 1. Cytograms, such as these described above, were used to analyze individual populations, after mechanical cell sorting, staining in Giemsa and evaluated by light microscopy. Each distinct population had over 98% of neutrophils or macrophages, respectively. As seen, the percentage of neutrophils present on peripheral blood was higher than that of macrophages in the peritoneal fluid. Indeed, R2 cell population in Fig. 1 is less discrete than R1. Tables 1 and 2 show the effects of midazolam on oxidative burst responses of peripheral blood neutrophil and peritoneal macrophage, respectively. Analysis of variance showed significant differences for both peripheral blood neutrophils (F2;15 ¼ 15:2; P < 0:05) and peritoneal macrophages (F2;15 ¼ 17:5; P < 0:05). As expected, fluorescence was not observed in unloaded cells (PBS). An increment in green fluorescence
was observed after DCFH load, allowing satisfactory measurements of midazolam effects on oxidative responses in peripheral blood neutrophils and peritoneal macrophages. Incubation with PI-labeled S. aureus and DCFH induced an increment in the level of green fluorescence in both neutrophils (F2;15 ¼ 1:61; P < 0:05) and macrophages (F2;15 ¼ 3:44; P < 0:05) retrieved from animals of all groups before midazolam treatment; this was most probably a reflex of the increment in cellular oxidative burst that followed S. aureus phagocytosis. This oxidative burst, however, was reduced (P < 0:05) by previous midazolam administration in a dose-dependant manner. It is important to note that midazolam effects were observed only in cells challenged with S. aureus. The values of oxidative burst measured in animals of groups E1 and E2 immediately before midazolam treatment were not different from those in animals of group C taken immediately before, or 24 h after 0.9% NaCl administration.
Table 2 Effects of midazolam on oxidative burst of peritoneal macrophages of horses Condition
PBS DCFH S. aureus
Before treatment
After treatment
C
E1
E2
C
E1
E2
2.85 0.1 137 9.2 223 20
4.56 0.8 125 14 245 13.2
3.62 0.2 119 22 232 13.7
5.32 1.6 130 11.2 238 11.2
3.45 0.9 122 19 178 17*
4.23 0.6 109 12 156 11.2**
C: animals treated with 0.9% NaCl (5.0 ml); E1 and E2: animals treated with 0.06 and 0.1 mg/kg of midazolam, respectively. Gate analysis was set on peritoneal macrophage using forward and side scatter gating. Values represent mean fluorescence intensity in gated peritoneal macrophages. Data are represented as mean S:D. * P < 0:01 in comparison to animals from group C. ** P < 0:05 in comparison to animals from group E1.
16
C. Massoco, J. Palermo-Neto / Veterinary Immunology and Immunopathology 95 (2003) 11–19
Fig. 2. Effect of midazolam on phagocytosis of peripheral blood neutrophil (A) and peritoneal macrophage (B) of horses. Twenty-four hours after 0.9% NaCl or midazolam (0.06 or 0.1 mg/kg), the mixture of 100 ml of heparinized whole blood or peritoneal macrophage (2 105 cells/ 100 ml), 200 ml of 0.3 mM DCFH-DA in PBS and 700 ml of PI-labeled S. aureus was incubated for 30 min at 37 8C. The cells were then analyzed by flow cytometry. Striped and open bars refer to the percentage of phagocytic activity measured before and after midazolam or 0.9% NaCl, respectively. C: animals treated with 0.9% NaCl (5.0 ml); E1 and E2: animals treated with 0.06 and 0.1 mg/kg dose of midazolam, respectively. Data are mean S:D. () P < 0:01 and (#) P < 0:05 compared to animals from groups C and E1, respectively.
Data on phagocytic capacity of peripheral blood neutrophils and peritoneal macrophages are depicted in Fig. 2A and B, respectively. An increment in red fluorescence was induced by S. aureus in both peripheral blood neutrophils and peritoneal macrophages. This increment in fluorescence allowed satisfactory analysis of midazolam effects on phagocytosis. Thus, midazolam was able to decrease phagocytosis in both peripheral blood neutrophils (F2;15 ¼ 24:3; P < 0:05) and peritoneal macrophages (F2;15 ¼ 19:3; P < 0:05) in a dose-dependent manner. As for the oxidative burst response, no differences were found in the capacity of phagocytosis among animals of groups C treated or not with 0.9% NaCl and those measured in animals of groups E1 and E2 before midazolam treatment. Table 3 Effects of midazolam on cortisol plasma concentration of horses Time after treatment (min)
Cortisol (mg/dl)
0 30 90 180 360
6.8 6.2 7.2 6.8 5.3
C
E2
E1
1.5 1.3 1.7 0.8 1.6
6.1 5.9 6.2 5.3 5.4
1.2 0.4 2.4 2.1 1.0
6.7 8.6 7.6 6.1 5.4
2.5 2.8 3.9 4.0 1.9
C: animals treated with 0.9% NaCl (5.0 ml); E1 and E2: animals treated with 0.06 and 0.1 mg/kg of midazolam, respectively. Data are represented by mean S:D.
3.2. Determination of plasma cortisol concentration As observed in Table 3, midazolam (0.06 and 0.1 mg/kg) did not change plasma cortisol concentration of horse. Indeed, no differences were detected among cortisol levels from horses of groups E1 and E2 in relation to those measured in saline-treated animals 30, 90, 180 or 360 min after midazolam treatment (F2;15 ¼ 9:2; P > 0:05).
4. Discussion The present findings demonstrate that midazolam treatment decreased equine innate immune response in a dose-dependent manner. It was observed 24 h after a single (0.06 and 0.1 mg/kg) midazolam treatment: (1) a dose-dependent reduction on peripheral blood neutrophil and peritoneal macrophage oxidative burst after PI-labeled S. aureus challenge; (2) a reduction in the capacity of both peripheral blood neutrophils and peritoneal macrophages to phagocyte PI-labeled S. aureus. Changes in plasma cortisol concentration were not found after two midazolam doses. These results are relevant not only because innate immune response plays a significant role in host defense against microbes (Speert, 1991) and tumor cells (Rees and Parry, 1991) but also because midazolam is frequently
C. Massoco, J. Palermo-Neto / Veterinary Immunology and Immunopathology 95 (2003) 11–19
used in equine anesthetic protocols (Luna et al., 1992). As far as it is of our knowledge this is the first description of midazolam effects on neutrophil and macrophage function in horses. The respiratory burst of neutrophils and macrophages represents a sequence of events, which is essential for the intracellular killing of bacteria. Active oxygen derivatives (superoxide, hydrogen peroxide, hydroxyl radical, singlet oxygen) generated during the above reactions are of great interest in the study of leukocyte oxidative metabolism. The present study demonstrates that the oxidative burst expressed in the generation of hydrogen peroxide by stimulated macrophages and neutrophils can be quantitatively monitored using 20 ,70 -dichlorofluorescein diacetate (DCFH-DA), PI-labeled S. aureus and flow cytometry. Brandt and Keston (1965) originally described a fluorometric method for the measurement of hydrogen peroxide in aqueous solution. This method was based on the oxidation of non-fluorescent DCFH-DA to a fluorescent 20 ,70 -dichlorofluorescein by hydrogen peroxide and peroxidase. The fluorescent reagent is unable to diffuse outside the cell due to its polarity. Therefore, it is possible to detect DCFH-DA oxidation by neutrophils and macrophages using single cell analyses by flow cytometry, as in the present study. It was shown that DCFH-DA oxidation assay is related quantitatively to the concentration of hydrogen peroxide generated (Hirabayashi et al., 1985). Formation and release of hydrogen peroxide in phorbol myristate acetate- and PI-labeled S. aureus-stimulated polymorphonuclear leukocytes was observed at rates that closely paralleled those associated with phagocytosis (Root et al., 1975; Hasui et al., 1989). Therefore, the decrease we now report for the oxidative burst of neutrophils and macrophages of horses after treatment with midazolam might have been a consequence of an impairment of PI-labeled S. aureus-induced phagocytosis. In mice, complex immunological responses, such as humoral responses, are significantly increased by i.p. administration of low doses (1 mg/kg) of peripheral (Ro 5-4864, PK 111 95) and mixed-type (diazepam) PBR ligands. On the other hand, the central BDZ receptor agonist clonazepam showed no effects, nor did its antagonist Ro 15-1788 (Zavala and Lenfant, 1984) suggesting a specific functional role for the PBR in immune cells. Diazepam binding to PBR was found
17
on immune cells such as macrophages (Zavala and Lenfant, 1984) and lymphocytes (Rocca et al., 1993). Activation of PBRs was reported to change the expression and release of several cytokines (Schulumpf et al., 1994) and cell products (Torres et al., 2000), whose effects strongly influence the immune/inflammatory response. Furthermore, PBRs were also described in adrenal cells and have been related to endogenous glucocorticoid production (Papadopoulos, 1993). Thus, diazepam might decrease innate immune response acting directly on immune cells and/or indirectly, through glucocorticoid secretion (Lazzarini et al., 2001). Midazolam was reported to be a mixed-type PBR agonist at doses similar to those used in the present experiment (Taupin et al., 1991). Thus, the findings now been reported on equine innate immunity might be attributable to a possible midazolam effect on PBR present on neutrophils and macrophages. This hypothesis agrees with those already reported by our laboratory for diazepam (Massoco and Palermo-Neto, 1999) and elsewhere for other BDZ ligands (Zavala and Lenfant, 1987). Increments on equine cortisol plasma concentration were not found in the present experiment after midazolam treatments. Thus, an indirect effect for midazolam on neutrophil and macrophage activities via an increment in plasma glucocorticoid concentration seems unlikely. This latter hypothesis could be relevant since PBR stimulation within the inner mitochondrial membrane of adrenal gland cells was reported to increase the biosynthesis of glucocorticoids (Cavallaro et al., 1992). An involvement of PBRs on human immune cell oxidative burst was already reported (Zavala, 1997). The production of oxygen derivatives is dependent on activation of the multimolecular complex NADPHoxidase. A functional link between PBR and NADPHoxidase activation was supported by experiments showing that a monoclonal antibody recognizing the PBR produced a concentration-dependent stimulation of the oxidative burst and enhanced the response to f-Met-Leu-Phe (FMLP) in normal human neutrophils (Zavala et al., 1991). Up-regulation of PBRs has also been associated with maturation of cells such as HL-60, U939 and THP1 into fully competent phagocytes, suggesting that these processes may be interconnected (Canat et al., 1993; Ishiguro et al., 1987). Therefore, it seems reasonable to suggest that midazolam, most
18
C. Massoco, J. Palermo-Neto / Veterinary Immunology and Immunopathology 95 (2003) 11–19
probably acting on PBRs present on macrophage and neutrophil membranes, might have changed some relevant mechanisms related to phagocytosis in these cells, decreasing their oxidative burst. Among other mechanisms, a possible midazolam effect on Ca2þ mobilization emerges as a reliable candidate. Ca2þ mobilization was reported to be critical not only to PBR-mediated responses (Mestre et al., 1985) but also to both phagocytosis (Bengtsson et al., 1993) and NADPH-oxidase activation (Hamilton and Adams, 1987). The flow cytometric method we describe here has been successfully employed to analyze the effects of midazolam on some macrophage and neutrophil functions. Particularly relevant in the context of the present experiment was the fact that this methodology allowed the discrimination between intracellular and extracellular oxidative events, therefore generating data regardless of the absolute cell number. It was possible to identify and evaluate the response of discrete cellular populations without prior separation, which is time consuming and able to alter cell function. Our present data agrees with that reported elsewhere (Raidal et al., 1998) and strongly suggests the use of flow cytometry to identify functional properties of immune cells from both blood and fluid withdrawn from peritoneal cavity of horses. Furthermore, it raises concerns about the safety of using midazolam for anesthetic procedures in horses, particularly in immunocompromised animals. They also support the notion that changes in the functional capacity of the immune system may represent an important hazard of exposure to drugs or chemicals.
Acknowledgements This research, which is part of the Ph.D. thesis by Cristina Massoco to the Department of Pathology of the School of Veterinary Medicine of the University of Sa˜ o Paulo, was sponsored by FAPESP (99/04228-7) and CNPQ (149318/1999). The authors would like to thank the Regimento da Cavalaria ‘‘9 de Julho’’ (Sa˜ o Paulo, SP) that supplied the horses and to MS Maria Aparecida Dalboni and Dr. Miguel Cendoroglo (Fundac¸a˜ o Oswaldo Ramos-UNIFESP) for providing us with PI-labeled Staphylococcus aureus and for technical assistance with the flow cytometer.
References Bass, D.A., Parce, J.W., Dechatelet, L.R., Szejda, P., Seeds, M.C., Thomas, M., 1983. Flow cytometric studies of oxidative produce formation by neutrophils: a graded response to membrane stimulation. J. Immunol. 130, 1910–1917. Bassoe, C.F., Laerum, O.D., Solberg, C.O., Haneberg, B., 1983a. Phagocytosis of bacteria by human leucocytes measured by flow cytometry. Proc. Soc. Exp. Biol. Med. 174, 182–186. Bassoe, C.F., Laerum, O.D., Glette, F., Hopen, G., Haneberg, B., Solberg, C.O., 1983b. Simultaneous measurement of phagocytosis and phagosomal pH by flow Cytometry: role of polymorphonuclear neutrophilic leukocyte granules in phagosome acidification. Cytometry 4, 254–262. Bengtsson, T., Jaconi, M.E.E., Gustafson, M., Magnusson, K.E., Theler, J.M., Lew, D.P., Stendhal, O., 1993. Actin dynamic in human neutrophils during adhesion and phagocytosis is controlled by changes in intracellular free calcium. Eur. J. Cell. Biol. 62, 49–58. Braestrup, C., Squires, R.F., 1977. Specific benzodiazepine receptors in rat brain characterizatized by high-affinity [3 H] diazepam binding. Proc. Natl. Acad. Sci. U.S.A. 74, 3805–3809. Brandt, R., Keston, A.S., 1965. Synthesis of diacetyldichlorofluorescein: a stable reagent for fluorometric analysis. Anal. Biochem. 11, 6. Canat, X., Guillaumont, A., Bouaboula, M., Poinot-Chazel, C., Derocq, J.M., Carayon, P., Le Fur, G., Casellas, P., 1993. Peripheral benzodiazepine receptor modulation with phagocyte differentiation. Biochem. Pharmacol. 46, 551–554. Casellas, P., Galiegue, S., Basile, A.S., 2002. Peripheral benzodiazepine receptors and mitochondrial function. Neurochem. Int. 40, 475–486. Cavallaro, S., Korneyev, A., Guidotti, A., Costa, E., 1992. DBIprocessing products, acting at the mitochondrial DBI receptor, mediate ACTH-induced steroidogenesis in rat adrenal. Proc. Natl. Acad. Sci. U.S.A. 89, 10598–10602. Cohen, M., Pickar, D., Dubois, M., Roth, Y.F., Naber, D., Bunney Jr., W.E., 1981. Surgical stress and endorphins (letter). Lancet 1, 213–214. Fearon, D.T., Collins, L.A., 1983. Increased expression of C3b receptors on polymorphonuclear leukocytes induced by chemotactic factors and purification procedures. J. Immunol. 130, 370–375. Ferrarese, C., Appolonio, I., Frigo, M., Perego, M., Pierpaoli, C., Trabucchi, M., Frattola, L., 1990. Characterization of peripheral benzodiazepine receptors in human blood mononuclear cells. Neuropharmacology 29, 375–378. Hamilton, T.A., Adams, D.O., 1987. Molecular mechanism of signal transduction in macrophages. Immunol. Today 5, 151–158. Hasui, M., Hirabayashi, Y., Kobayashi, Y., 1989. Simultaneous measurement by flow cytometry of phagocytosis and hydrogen peroxide production of neutrophils in whole blood. J. Immunol. Methods 117, 53–58. Hirabayashi, Y., Taniuchi, S., Kobayashi, Y., 1985. A quantitative assay of oxidative metabolism by neutrophils in whole blood using flow cytometry. J. Immunol. Methods 82, 253–259.
C. Massoco, J. Palermo-Neto / Veterinary Immunology and Immunopathology 95 (2003) 11–19 Ishiguro, K., Taft, W.C., de Lorenzo, R.J., Sartorelli, A.C., 1987. The role of benzodiazepine receptors in the induction of differentiation of HL-60 leukemia cells by benzodiazepine and purines. J. Cell. Physiol. 131, 226–234. James, V.H.T., Horner, M.W., Moss, M.S., Rippon, A.E., 1970. Adrenocortial function in the horse. J. Endocrinol. 48, 319–335. Johannisson, A., Gro-Ndahl, G., Demmers, S., Jensen-Waern, M., 1995. Flow-cytometric studies on the phagocytic capacities of equine neutrophils. Acta Vet. Scand. 36, 553–562. Lazzarini, R., Malucelli, B.E., Palermo-Neto, J., 2001. Reduction of acute inflammation in rats by diazepam: role of peripheral benzodiazepine receptors and corticosterone. Immunopharmacol. Immunotoxicol. 23, 253–265. Luna, S.P., Massone, F., Castro, G.B., Fantoni, D.T., Hussni, C.A., Aguiar, A.J., 1992. A combination of methotrimeprazine, midazolam and guaiphenesin, with and without ketamine in an anaesthetic procedure of horses. Vet. Rec. 131, 33–35. Massoco, C.O., Palermo-Neto, J., 1999. Diazepam effects on peritoneal macrophage activity and corticosterone serum levels in Balb/c mice. Life Sci. 65, 2157–2165. McEwen, B.S., Biron, C.A., Brunson, K.W., Bullock, K., Chambers, W.H., Dhabhar, F.S., Goldfarb, R.H., Kitson, R.P., Miller, A.H., Spencer, R.L., Weiss, J.M., 1997. The role of adrenocorticoids as modulators of immune function in health and disease: neural, endocrine and immune interaction. Brain Res. Rev. 23, 79–133. Mestre, M., Carrior, T., Belin, C., Uzan, A., Renault, C., Dubroeucq, M.C., Gueremy, C., Doble, A., Le Fur, G., 1985. Electrophysiological and pharmacological evidence that peripheral type benzodiazepine receptors are coupled to calcium channels in the heart. Life Sci. 36, 391–400. Nathan, C.F., 1987. Neutrophil activation on biological surfaces: massive secretion of hydrogen peroxide in response to products of macrophage and lymphocytes. J. Clin. Invest. 80, 1550– 1560. Noel, G.L., Suh, H.K., Stone, J.C., Frantz, A.G., 1972. Human prolactin and growth hormone release during surgery and other conditions of stress. J. Clin. Endocrinol. Metab. 35, 840–851. Okun, F., Weizman, R., Katz, Y., Bomzon, A., Youdim, M.B.H., Gavish, M., 1988. Increase in central and peripheral benzodiazepine receptors following surgery. Brain Res. 458, 31–36. Papadopoulos, V., 1993. Peripheral-type benzodiazepine/diazepam binding inhibitor receptor: biological role in steroidogenic cell function. Endocr. Rev. 14, 222–240. Raidal, S.L., Bailey, G.D., Love, D.N., 1998. Flow cytometric determination of oxidative burst activity of equine peripheral blood and bronchoalveolar lavage-derived leukocytes. Vet. J. 156, 117–126. Rees, R.C., Parry, H., 1991. Macrophages in tumor immunity. In: Lewis, C.E., McGee, J.O’D. (Eds.), The Macrophage. IRL Press, New York, pp. 3–423.
19
Righi, D.A., Pinheiro, S.R., Guerra, J.L., Palermo-Neto, J., 1999. Diazepam effects on Mycobacterium bovis-induced infection in hamsters. Braz. J. Med. Biol. Res. 32, 1145–1153. Rocca, P., Bellone, G., Benna, P., Bergamasco, B., Ravizza, L., Ferrero, P., 1993. Peripheral-type benzodiazepine receptors and diazepam binding inhibitor-like immunoreactivity distribution in human peripheral blood mononuclear cells. Immunopharmacology 25, 163–178. Root, R.K., Metcalf, J., Oshino, N., Chance, B., 1975. H2O2 release from human granulocytes during phagocytosis. J. Clin. Invest. 55, 945. Rothe, G., Valet, G., 1990. Flow cytometric analysis of respiratory burst activity in phagocytes with hidroethidine and 20 ,70 dichlorofluorescein. J. Leuk. Biol. 47, 440–448. Ruiz, I., Offermanns, J., Lanctot, K.L., Busto, U., 1993. Comparative study on benzodiazepine use in Canada and Chile. J. Clin. Pharmacol. 33, 124–125. Saad, A.M., Hageltorn, M., 1985. Flow cytometric characterization of bovine blood neutrophil Phagocytosis of fluorescent bacteria and zymosan particles. Acta Vet. Scand. 26, 289–307. Schulumpf, M., Lichtensteiger, W., Loveren, H., 1994. Impaired host resistance to Trichinella spiralis as a consequence of prenatal treatment of rats with diazepam. Toxicology 94, 223– 229. Speert, D.P., 1991. Macrophages in bacterial infection. In: Lewis, C.E., McGee, J.O’D. (Eds.), The Macrophage. IRL Press, New York, pp. 3–423. Taupin, V., Jayais, P., Descamps-Latscha, B., Cazala, J.B., Barrier, G., Bach, J.F., Zavala, F., 1991. Benzodiazepine anesthesia in humans modulates the interleukin-1 beta, tumor necrosis factoralpha and interleukin-6 responses of blood monocytes. J. Neuroimmunomodul. 35, 13–19. Torres, S.R., Fro¨ de, T.S., Nardi, G.M., Vita, N., Reeb, R., Ferrara, P., Ribeiro-do-Vale, R.M., Farges, R.C., 2000. Anti-inflammatory effects of peripheral benzodiazepine receptor ligands in two mouse models of inflammation. Eur. J. Pharmacol. 408, 199–211. Zavala, F., Lenfant, M., 1984. Interactions of benzodiazepines with mouse macrophages. Eur. J. Pharmacol. 106, 561–566. Zavala, F., Lenfant, M., 1987. Peripheral benzodiazepine enhance the respiratory burst of macrophage-like P388D1 cells stimulated by arachidonic acid. Int. Immunopharmacol. 9, 269–274 [Eur. J. Pharmacol. 106, 561–566]. Zavala, F., Taupin, V., Descamps-Latscha, B., 1990. In vivo treatment with benzodiazepine inhibits murine phagocyte oxidative metabolism and production of interleukin-1, tumor necrosis factor and interleukin-6. J. Pharmacol. Exp. Ther. 255, 442–450. Zavala, F., Masson, A., Brys, L., de Baetselier, P., DescampsLatscha, B., 1991. A monoclonal antibody against peripheral benzodiazepine receptor activates the human neutrophil NADPHoxidase. Biochem. Biophys. Res. Commun. 176, 1577–1583.