In addition to blocking nerve transmission, local anesthetics (LA) have significant antiinflammatory properties (1). They affect lymphocytes and peritoneal macrophages (2), block neutrophil (PMN) accumulation at the site of inflammation, and inhibit free radical and mediator release (3–5). The ability of LA to impair wound healing in various models (6) may in part reflect their ability to affect inflammatory mediator signaling. However, the exact mechanisms underlying these antiinflammatory effects are poorly understood.
Lysophosphatidic acid (LPA) is an intercellular phospholipid mediator known to induce a variety of biologic responses (e.g., cell proliferation, platelet aggregation, smooth muscle cell contraction, chemotaxis and inhibition of differentiation) (7). LPA influences target cells by activating several specific G protein-coupled membrane receptors present in numerous cell types. These in turn activate a number of intracellular signaling cascades. Although the physiological functions of the compound remain to be determined, LPA can be generated by platelets, leukocytes and other cells challenged with inflammatory stimuli (8), suggesting that it may stimulate responses at sites of inflammation. Based on in vitro studies (9,10), we have hypothesized that part of the antiinflammatory actions of LA may be because of interference with LPA signaling. Using Xenopus oocytes we showed that signaling of LPA receptors was inhibited by extracellular application of lidocaine and bupivacaine (9,10). The goal of the present study was therefore to determine the effects of LA on LPA-induced PMN responses.
PMNs are of importance in the host defense. They move actively to the site of inflammation (chemotaxis), where they generate toxic oxygen metabolites. PMNs exist in one of three states: quiescent, primed, or active. Priming refers to a process whereby the response of PMNs to a subsequent activating stimulus is potentiated. Release of oxygen metabolites is markedly enhanced when activated PMNs have been primed previously (11,12). Importantly, the priming process is a critical component of PMN-mediated tissue injury both in vitro and in vivo(11,12).
The hypotheses to be tested in this study, therefore, were that LPA is able to activate and/or prime PMNs and to act as a chemotactic stimulus, and that LA can interfere with LPA signaling in PMNs.
Materials and Methods
The study protocol was approved by the University of Virginia IRB.
Preparation of Human PMNs.
Human venous blood was repeatedly obtained from healthy donors (n = 25) who had not taken any medication for at least 2 wk. Blood was heparinized (10 U/mL) and PMNs were isolated by a one-step Ficoll-Hypaque separation procedure. After 25 min of centrifugation at 1500 rpm, PMNs were washed three times with Hank’s Balanced Salt Solution (HBSS, containing 10 U/mL heparin), and centrifuged after each washing step at 1000 rpm for 10 min. PMNs were then resuspended in 5 mL HBSS and counted using a hemocytometer. Unopette Diagnostic System (Becton Dickinson, Franklin Lakes, NJ) in vitro was used for the enumeration of PMNs in the suspension, which was then diluted with HBSS to obtain a final PMN suspension of 5 × 106 cells/mL.
Preparation of Agar Plates.
PMN migration under agar was studied, as described previously (12). Chemotaxis agar was prepared by combining the following (per 100 mL): 1.2 g agarose, 10 mL heat-inactivated pooled human serum, 10 mL 10× concentrated minimal essential medium (HMEM), 80 mL sterile water, 1 mL 7.5% sodium bicarbonate, and 1 mL 1 M HEPES buffer. Hot agar was pipetted (2.5 mL per plate) into tissue culture plates (60 × 15 mm), to give a thickness of approximately 1 mm. After the layer had hardened, nine 3-mm wells were punched in the agar producing a pattern with one center well (for control solution), 4 middle cells (for PMNs), and 4 outer cells (for chemoattractant, Fig. 1A). In this manner, random migration of PMNs to the control well could be compared with directed migration toward the chemoattractant cell.
PMNs at a concentration of 5 × 106 cells/mL were incubated at 37°C with or without lidocaine (10−4 to 10−9 M) for 60 min and then centrifuged at 1000 rpm for 5 min. The cell pellets were typically approximately 25 μL in volume. After disposal of the supernatant, 7 μL of the cell pellet (approximately 1.4 × 106 PMNs) was placed in each of the PMN wells in the agar plates. The control well was filled with HBSS (as negative control), and the chemoattractant wells were filled with formyl-methionine-leucine-phenylalanine (fMLP, 10−7 M), a recognized chemoattractant (as positive control), or various concentrations of LPA (10−10 to 10−3 M). Plates were then incubated at 37°C for 3 h to allow PMNs to migrate beneath the agar. At that time, the plates were fixed with methanol (30 min) followed by phosphate-buffered formalin. PMNs were visualized using Giemsa staining, and average PMN migration (in mm) toward the chemoattractant and control wells was measured using a microscope.
Superoxide Anion (O2−) Generation
We used the cytochrome c-reduction assay to measure extracellular O2− production by activated PMNs as described previously (13). O2− generation was mea-sured spectrophotometrically as the SOD-inhibitable reduction of cytochrome c. In contrast to the indi-rect luminol assay (which measures both extracellu-lar and intracellular free radical production), the cytochrome-c assay detects extracellular O2− generated by NADPH oxidase. We measured extracellular (rather than total) PMN-mediated release of O2− because it is the release of oxygen metabolites into the extracellular milieu that may directly damage cells in the surrounding microenvironment.
O2− production was measured as change over time by the absorbance of cytochrome c at 550 nm. The reaction was performed in a spectrophotometer (Genesys 5; Spectronic Instruments, Rochester, NY). The reaction mixture was prepared by placing 700 μL buffer (HBSS + BSA 0.1%), 200 μL of PMN suspension (final concentration 106 cells/mL), 100 μL cytochrome c (from horse heart, 3.7 mg/mL) with catalase (0.14 mg/mL) in a 1-mL cuvette. The reference sample was prepared the same way but in addition, 10 μL of superoxide dismutase (SOD, 10−2 M) was added to the mixture to determine the selective contribution of O2−.
PMNs were activated with fMLP and the change in absorbance at 550 nm was followed over time for the next 14 min. The reference sample was measured immediately after, and O2– dependent cytochrome c reduction was determined by subtracting the reference value from the study sample value. O2− generation at time points 0, 14, 16, and 18 min was calculated using as conversion factor 47.4 μmol (1/21.1 mM−1 difference of extinction coefficient between oxidated and reduced cytochrome c at 550 nm) O2− per unit change in absorbance. We used a kinetic assay because the kinetics of PMN O2– release are nonlinear and vary with different priming and activating stimuli.
After addition of the PMNs to the samples, baseline activity was measured for 4 min to exclude activating effects of the isolation process. fMLP (10−8 M-10−4 M) was added to the samples. In a second experiment, platelet-activating factor (PAF), an established priming agent in vivo, was selected as a positive control for priming studies. It primes PMNs rapidly, in a receptor-mediated manner and with minimal direct activation of O2− release. Therefore, PAF (10−6 M) was administered 10 min before adding the activating agent fMLP.
Direct Activating Effects of LPA
Direct activating effects of LPA on the O2− production and its possible concentration-dependence were investigated. After addition of the PMNs to the samples, baseline activity was measured for 4 min. LPA (10−6 to 10−4 M) was then added to the samples and absorption was measured during the next 14 min.
Priming Effects of LPA
Next, the priming effects of LPA on fMLP-activation in PMNs were studied. After baseline activity was assessed, LPA (10−6 M–10−4M) and 10 min later fMLP (10−6 M) were added to the same sample.
Effects of LA on LPA Signaling
PMNs were incubated with lidocaine (10−6 M–10−4 M for 10 and 60 min), S (-)-ropivacaine (10−6 M–10−4 M for 60 min) or tetracaine (10−6 M–10−4 M for 60 min) before LPA (10−4 M, priming agent) and fMLP (10−6 M, activation agent) were added to the samples.
Agarose (Litex type HAS) and polymorph (1-step Polymorphs) were from Accurate Chemical and Scientific Corp. (Westbury, NY), HMEM and HBSS (without phenol red, with Ca2+/Mg2+) were from Whittaker Bioproducts (Walkersville, MD), HEPES (1 M), superoxide dismutase (SOD, from bovine liver), fMLP, cytochrome C (from horse heart), catalase (from bovine liver) were from Sigma Chemical (St. Louis, MO). Tissue culture plates were from Becton Dickinson and Company (Franklin Lakes, NJ). Ficoll-Hypaque and BSA (protease free bovine albumin fraction/fatty acid free) were from ICN Biomedicals Inc. (Aurora, OH). Human Albumin (25%) was bought from Bayer, Pharmaceutical Division (Elkhart, NJ). NIM was from Cardinal Associates (Santa Fe, NM). LPA (1-oleoyl-2-hydroxy-sn-glycero-3-phosphate, solution in 10 mg/mL chloroform (2.5 mL)) and PAF (1-alkyl-2-acetoyl-sn-glycero-3-phosphocholine) were from Avanti polar lipids (Alabaster, AL). Lidocaine, tetracaine and S(-) ropivacaine were from Astra Pharmaceuticals, LP (Westborough, MA).
Calculations and Statistical Analysis
Chemoattraction distance was defined as directed migration minus random migration, and is expressed as percent change of response obtained in a concurrent untreated control group. Data are reported as mean ± sem, comparisons were made by Student’s t-tests. Leukocyte metabolic activity is reported either as O2− production or as percentage change from untreated control. Groups were compared using one-way analysis of variance, followed by Student’s t-tests or a pairwise multiple comparison procedure (Dunnett’s Method), as appropriate. SigmaStat 2.0 (Jandel Scientific Corporation, San Rafael, CA) was used for all statistical analyses.
LPA induced directed migration of PMNs. There was a trend for larger LPA concentrations to induce larger directed migration (half-maximum effect obtained at approximately 10−6 M, Fig. 1B), although it does not reach statistical significance. At the largest LPA concentration tested (10−3 M), LPA was approximately as effective as fMLP 10−7 M, indicating that the compound is a moderately potent chemoattractant.
Whereas random migration was not affected by lidocaine (data not shown), directed PMN migration to-ward LPA 10−6 M was attenuated in a concentration-dependent manner by lidocaine (Fig. 1C). Directed migration was significantly attenuated at lidocaine concentrations larger than 10−7 M, suggesting that lidocaine in clinically relevant concentrations may influence LPA-induced PMN migration. At the largest lidocaine concentration tested (10−4 M), directed migration was attenuated to 15 ± 8%.
Leukocyte Metabolic Activity
Because PMNs were to be incubated in LA for significant duration before the activation assay, we determined, in pilot experiments, the effect of incubation in buffer (37°C, 0–60 min) as well as the effect of movement (by a shaking waterbath) on O2− production. Neither affected O2– production as compared with untreated control PMNs (data not shown). We also determined any interference of LPA with the cytochrome c assay. LPA did not have any significant effect on absorbance in a PMN-free solution as compared with control (data not shown).
fMLP activated PMNs, as has been reported (14). The fMLP response was concentration-dependent, with maximal responses obtained at 10−6 M; at larger concentrations, responses decreased (Fig. 2A). Maximal O2− production was reached after 16 min. When pretreated for 10 min with PAF (10−6 M), responses to fMLP were increased significantly (Fig. 2B), indicating that PMNs could be primed in our model (15). Incubation with lidocaine (10−6–10−4 M for 60 min) was without effect on fMLP-induced activation (Fig. 2C). We observed some donor-dependent variability in fMLP effect, as exemplified in Figure 2.
We next determined if LPA can activate PMNs. PMNs were added to the assay solution, and baseline activity was measured for 4 min to exclude activating effects of the isolation process itself. LPA (10−6–10−4 M) was then added to the samples and absorbance was measured for the next 14 min. LPA did not activate human PMNs as compared with unprimed and not activated control. Even at the largest concentration tested (10−4 M), superoxide production was not increased significantly: 0.2 ± 0.02 (baseline activity, control) vs 0.3 ± 0.06 (LPA) nmol/106 cells.
We then studied the ability of LPA to prime PMNs to subsequent activation by fMLP. Baseline activity was assessed, LPA (10−6 M–10−4 M) was then added to the assay solution, and 10 min later, fMLP (10−6 M) was used to activate the PMNs. Ten minutes incubation with LPA increased fMLP-induced O2− production as compared with control (Fig. 3A) in a concentration-dependent manner. Thus, LPA does not activate PMNs directly but is able to prime them effectively.
We also studied the effects of LA on fMLP-induced activation and LPA-induced priming. Lidocaine (10−6–10−4 M, 10 or 60 min) inhibited LPA (10−4 M)-primed (10 min), fMLP (10−6 M)-activated O2− generation in a concentration- and time-dependent manner (Fig. 3B). At the largest concentration tested (10−4 M), 60 min incubation in lidocaine decreased O2− production to 70 ± 4% of untreated control. Thus, lidocaine, in clinically relevant concentrations, can attenuate LPA-induced PMN priming. An example of the time course of O2− production (raw data) induced by different agents is shown in Figure 4.
We also determined the effect of the clinically used isomer S(-)-ropivacaine on LPA (10−6 M)-primed (10 min), fMLP (10−6 M)-activated PMNs. S(-)-ropivacaine inhibited to a similar degree as did lidocaine, indicating that the action is not specific to lidocaine. At the largest concentration tested (10−4 M) ropivacaine attenuated the O2− response induced by LPA + fMLP to 69 ± 4.4% of control (LPA + fMLP without LA) (Fig. 5A). To determine whether LA inhibition of LPA is a property of amide-linked LA only, we studied the effects of tetracaine (10−6–10−4 M for 60 min). It also attenuated LPA-induced priming, and at the largest concentration tested (10−4 M) attenuated the O2− response to 48.4 ± 7.0% of control (Fig. 5B). Both LA acted in a concentration-dependent manner.
We showed that LPA affects PMN migration and metabolic activity, suggesting that LPA, which is released at sites of injury from injured fibroblasts, activated platelets and PMNs (16), can act as an inflammatory mediator. We postulate that LPA, present in wounds, may help attract PMNs and may increase their response to activating stimuli. Unfortunately, the absence of selective LPA antagonists makes it difficult to test this hypothesis directly.
Both ester and amide LA, in concentrations present in blood after IV or epidural infusion, inhibited the chemotactic and priming effects of LPA. These findings are in agreement with our previous investigations, performed in Xenopus oocytes, in which we demonstrated that lidocaine and bupivacaine (albeit at significantly larger concentrations) inhibit LPA signaling in a concentration-dependent manner (9). No interaction was observed with the oocyte intracellular Ca signaling pathway, suggesting that the compounds interact with receptor and/or coupled G protein. Using the nonpermeable lidocaine analog QX314, we found that the primary binding site for LA is intracellular (10); however, we documented a secondary site for the uncharged compound benzocaine. The two sites interacted synergistically. Recently we have shown that some LA selectively interact with the Gαq G protein subunit, inhibiting signaling through receptors coupled to this protein (17). The current findings are compatible with these previous reports.
Our findings may in part explain the antiinflammatory properties of LA (1). For example, in vitro, LA were shown to inhibit the release of inflammatory mediators, such as LTB4 or IL-1a, attenuate the adhesion of PMNs to endothelium, reduce migration and therefore decrease the amount of inflammatory cells attracted to the site of inflammation. This results in a significant decrease in production of free oxygen radicals, preventing aggravation of tissue damage. These effects on the cellular level may explain the protective effects of LA obtained in vivo. Protection against chemical, hyperoxia- or endotoxin-induced lung injury, suppression of increased microvascular permeability as occurs in many inflammatory diseases (adult respiratory distress syndrome, sepsis, burns, peritonitis), and attenuation of myocardial infarction and reperfusion injury are examples of beneficial effects of LA as a result of their antiinflammatory properties. Of particular interest are LA effects on inflammatory diseases of the gastrointestinal tract (e.g., ulcerative colitis and proctitis) (1). In particular, ropivacaine has received significant attention as an inflammatory mediator in this regard, and has undergone clinical trials in inflammatory bowel disease (18). In the present study, we found ropivacaine to inhibit LPA-induced neutrophil functions as well.
Gerrard et al. (19) studied effects of LPA on PMN migration, and found no chemoattractant action when the compound was used alone. However, chemoattraction was observed with a combination of fMLP (2.4 × 10−6 M) and LPA (1.2 and 2.4 × 10−4 M). Remarkably, when fMLP was used at a smaller concentration (8 × 10−7 M), chemotaxis was enhanced. Methodologic differences may explain the difference with our findings. In our model, PMNs moved horizontally under agar, whereas Gerrard et al. (19) used a Boyden chamber where PMNs had to pass a 5-μm filter. In addition, Gerrard et al. (19) used palmitoyl-LPA, whereas we used oleoyl-LPA. Sasagawa et al. (20) studied the effects of lidocaine on fMLP (10−8 M)- or phorbol ester (50 ng/mL)-induced PMN chemotaxis and found an IC50 for lidocaine of 3.2 × 10−3 M. In contrast, tetracaine enhanced chemotactic migration of guinea pig PMN toward fMLP (5 × 10−9 M) (21). These findings suggest that the effects of LA may depend critically on the chemoattractant used.
In our study, LPA did not activate PMNs. This is in agreement with data by Chettibi et al. (22), who found LPA in concentrations below 2 × 10−5 M to be ineffective in stimulating the respiratory burst. However, at 1–2 × 10−4 M, LPA had a weak stimulatory effect, which we did not observe. In such large concentrations LPA might stimulate respiratory burst in a nonselective, receptor-independent manner. Chettibi et al. (22) found LPA to inhibit fMLP-induced metabolic burst (IC50 10−5 M). This is in contrast to our results, in which LPA acted as a priming agent for fMLP-induced O2− generation. However, whereas we studied extracellular O2− release in a kinetic assay, Chettibi et al. (22) used 7-dimethylamino-naphtalene-1,2-dicarbonic acid hydrazide-enhanced chemiluminescence, which measures intracellular as well as extracellular free radical release, and reported the total area under the curve. As with chemoattraction, the specific effects of LA on respiratory burst seem to depend critically on the activator used and on the time of incubation. Hyvonen et al. (23) showed that lidocaine (2–8 mg/mL, a concentration 1000 times larger than used in our experiments) inhibited respiratory burst of human PMNs, activated with phorbol myristate acid (8 × 10−5 M) and primed with microparticulate synthetic hydroxyapatite. The difference in required LA concentration may be explained by different ways of priming and activating PMNs, but also suggests that LPA priming is more sensitive to LA than is hydroxyapatite priming. Mikawa et al. (24) did not find any effect of clinically feasible concentrations of lidocaine and mepivacaine on production of reactive oxygen species in human PMNs. However, PMNs in their study were incubated with opsonized zymosan and concomitant application of LA for 10 min only. Similarly, Peck et al. (5) showed an effect of large concentrations of lidocaine (5.5 × 10−3 M), applied for short periods (15 min) on phorbol myristate acid (0.83 and 1.67 ng/mL)-induced O2− release. Sufficient incubation duration seems essential to demonstrate the effects of LA on PMNs.
In summary, we have shown that LPA can act as a chemoattractant and priming agent in human PMNs, and that these actions are inhibited by clinically relevant concentrations LA. In addition to advancing our understanding of LPA physiology, these findings may explain in part the antiinflammatory properties of LA.
1. Hollmann MW, Durieux ME. Local anesthetics and the inflammatory response: a new therapeutic indication? Anesthesiology 2000; 93: 858–75.
2. Dickstein R, Kiremidjian-Schumacher L, Stotzky G. Effects of lidocaine on the function of immunocompetent cells. I. In vitro
exposure of mouse spleen lymphocytes and peritoneal macrophages. Immunopharmacology 1985; 9: 117–25.
3. Hattori M, Dohi S, Nozaki M, et al. The inhibitory effects of local anesthetics on superoxide generation of neutrophils correlate with their partition coefficients. Anesth Analg 1997; 84: 405–12.
4. Cederholm I, Briheim G, Rutberg H, Dahlgren C. Effects of five amino-amide local anaesthetic agents on human polymorphonuclear leukocytes measured by chemiluminescence. Acta Anaesthesiol Scand 1994; 38: 704–10.
5. Peck SL, Johnston RB Jr, Horwitz LD. Reduced neutrophil superoxide anion release after prolonged infusions of lidocaine. J Pharmacol Exp Ther 1985; 235: 418–22.
6. Davies B, Guyuron B, Husami T. The role of lidocaine, epinephrine, and flap elevation in wound healing after chemical peel. Ann Plast Surg 1991; 26: 273–8.
7. Moolenaar WH. Lysophosphatidic acid signaling. Curr Opin Cell Biol 1995; 7: 203–10.
8. Fourcade O, Simon MF, Viode C, et al. Secretory phospholipase A2 generates the novel lipid mediator lysophosphatidic acid in membrane microvesicles shed from activated cells. Cell 1995; 80: 919–27.
9. Nietgen GW, Chan CK, Durieux ME. Inhibition of lysophosphatidate signaling by lidocaine and bupivacaine. Anesthesiology 1997; 86: 1112–9.
10. Sullivan LM, Hoenemann CW, Arledge JAM, Durieux ME. Synergistic inhibition of lysophosphatidic acid signaling by charged and uncharged local anesthetics. Anesth Analg 1999; 88: 1117–24.
11. Condliffe AM, Kitchen E, Chilvers ER. Neutrophil priming: pathophysiological consequences and underlying mechanisms. Clin Sci (Colch) 1998; 94: 461–71.
12. Frank MO, Sullivan GW, Carper HT, Mandell GL. In vitro
demonstration of transport and delivery of antibiotics by polymorphonuclear leukocytes. Antimicrob Agents Chemother 1992; 36: 2584–8.
13. Babior BM, Cohen HJ. Leukocyte function. New York: Churchill Livingstone, 1981.
14. Simchowitz L, Spilberg I. Generation of superoxide radicals by human peripheral neutrophils activated by chemotactic factor: evidence for the role of calcium. J Lab Clin Med 1979; 93: 583–93.
15. Kitchen E, Rossi AG, Condliffe AM, et al. Demonstration of reversible priming of human neutrophils using platelet-activating factor. Blood 1996; 88: 4330–7.
16. Nietgen GW, Durieux ME. Intercellular signaling by lysophosphatidate. Cell Adhes Commun 1998; 5: 221–35.
17. Hollmann MW, Wieczorek K, Berger A, Durieux ME. Local anesthetic inhibition of lysophosphatidate signaling is mediated by interference with Gaq protein function. Mol Pharmacol 2001; 59: 1–8.
18. Arlander E, Ost A, Stahlberg D, Lofberg R. Ropivacaine gel in active distal ulcerative colitis and proctitis – a pharmacokinetic and exploratory clinical study. Aliment Pharmacol Ther 1996; 10: 73–81.
19. Gerrard JM, Clawson CC, White JG. Lysophosphatidic acids: III. Enhancement of neutrophil chemotaxis. Am J Pathol 1980; 100: 609–18.
20. Sasagawa S. Inhibitory effects of local anesthetics on migration, extracellular release of lysosomal enzyme, and superoxide anion production in human polymorphonuclear leukocytes. Immunopharmacol Immunotoxicol 1991; 13: 607–22.
21. Kownatzki E, Uhrich S. Enhancement of chemotactic migration by the local anesthetic tetracaine. Immunobiology 1983; 165: 97–106.
22. Chettibi S, Lawrence AJ, Stevenson RD, Young JD. Effect of lysophosphatidic acid on motility, polarisation and metabolic burst of human neutrophils. FEMS Immunol Med Microbiol 1994; 8: 271–81.
23. Hyvonen PM, Kowolik MJ. Dose-dependent suppression of the neutrophil respiratory burst by lidocaine. Acta Anaesthesiol Scand 1998; 42: 565–9.
24. Mikawa K, Akamatsu H, Nishina K, et al. Inhibitory effect of local anaesthetics on reactive oxygen species production by human neutrophils. Acta Anaesthesiol Scand 1997; 41: 524–8.