0192-0561/90 $3.00 + .00 Pergamon Press plc. International Society for Immunopharmacology.

Int. J. lmmunopharmac., Vol. 12, No. 7, pp. 809-819, 1990. Printed in Great Britain.

EFFECTS OF PROCAINAMIDE HYDROXYLAMINE ON GENERATION OF REACTIVE OXYGEN SPECIES BY MACROPHAGES A N D PRODUCTION OF CYTOKINES L. E. ADAMS,* S. M. ROBERTS,-? J. M. CARTER,* J. F. WHEELER,* H. W. ZIMMER,:~ R. J. DONOVAN-BRAND* and E. V. HESS*~ *Division of Immunology, Department of Medicine, University of Cincinnati Medical Center; *Center for Environmental Toxicology, University of Florida, Gainesville, FL; and *Department of Chemistry, University of Cincinnati, Cincinnati, OH, U.S.A.

(Received 22 February 1990 and in final form 13 April 1990)

-A series of experiments was conducted to examine the effects of the N-oxidized metabolite of procainamide, procainamide hydroxylamine (PAHA), on reactive oxygen species (ROS) production by macrophages in vitro, as well as on the release of the cytokine interleukin-1 (IL-1). Results with PAHA were compared with those from the parent compound, procainamide, and in some cases with other procainamide metabolites such as N-acetylprocainamide or nitrosoprocainamide. The effects of PAHA on ROS production by mouse and rat macrophages were complex, resulting in both stimulatory and inhibitory activity depending upon the PAHA concentration and whether macrophages were resting or elicited. The primary effect of PAHA appeared to be a stimulation of ROS production. Monocytes pretreated with PAHA (20/aM) depressed the responsiveness of lymphocytes in co-culture to a T-cell mitogen (conconavalin A) but not a B-cell mitogen (lipopolysaccharide). This effect was inhibited when monocyte pretreatment with PAHA was accompanied by the antioxidants, catalase or superoxide dismutase. IL-1 production by rat adherent splenocytes was unaffected by PAHA in concentrations that were not cytotoxic. These observations suggest that the oxidative metabolism of procainamide to PAHA may result in enhanced production of ROS by macrophages contributing its toxicity to lymphocytes.

Abstract

Several drugs are capable of eliciting symptoms in humans resembling systemic lupus erythematosus. Procainamide is one of the most important of these drugs because of its widespread use as an antiarrhythmic agent and comparatively high incidence of lupus-like symptoms in patients receiving chronic procainamide therapy (Hess, 1987; Kosowsky, Taylor, Lown & Ritchie, 1973; Blomgren, Condemi & Vaughn, 1972; Henningsen, Cederberg, Hanson & Johansson, 1975). Though the mechanism(s) by which drugs such as procainamide can initiate an autoimmune syndrome is unclear, several recent studies have focused on the hypothesis that procainamide influences the immune system primarily through an N-oxidation metabolite. Procainamide has been shown to be oxidatively metabolized to an arylhydroxylamine by both rat and human liver (Uetrecht, Sweetman, Woosley & Oates, 1984; Budinsky, Roberts, Coats, Adams & Hess, 1987). This procainamide hydroxylamine

metabolite may be further oxidized to the nitroso derivative (Uetrecht, 1985). Both procainamide hydroxylamine (PAHA) and nitrosoprocainamide (NOPA) have been shown in in-vitro studies to be much more toxic to lymphocytes than the parent drug, procainamide (Rubin, Uetrecht & Jones, 1987; Wheeler, Lunte, Heineman, Adams & Hess, 1988; Adams, Sanders, Budinsky, Donovan-Brand, Roberts & Hess, 1989). In contrast to the outcome of N-oxidation, if the primary aromatic amine of procainamide is instead N-acetylated, the resulting metabolite (N-acetylprocainamide) has been shown to have comparatively little effect on lymphocyte function or viability (Adams et al., 1989; Roberts, Adams, Donovan-Brand, Budinsky, Skoulis, Zimmer & Hess, 1989) and clinically its use is not associated with the drug-related lupus syndrome (Kluger, Drayer, Reidenberg & Lahita, 1981; Atkinson, Lertero, Kushner, Chao & Nevin, 1983; Roden, Reele,Higgins, Smith, Oates & Woosley,

§Author to whom correspondence should be addressed at Division of Immunology, Mail Location 563, University of Cincinnati Medical Center, 231 Bethesda Ave., Cincinnati, OH 45267, U.S.A. 809

810

L. E. ADAMSet al.

1980; Lahita, Kluger, Drayer, Koffler & Reidenberg, 1979). A clear picture as to how the N-oxidation metabolites of procainamide such as PAHA or NOPA could trigger an autoimmune response has not yet emerged. There are numerous examples of compounds which influence the various specific aspects of immune function, yet are not associated with induction of the symptoms or serology of lupus. Therefore it is reasonable to suspect that the events required to produce the lupus-like syndrome are complex, involving more than one component of the immune system. We and others have been evaluating the reactivity of these metabolites towards endogenous macromolecules in an effort to determine whether metabolite binding could alter their antigenicity (Uetrecht, 1985; Wheeler, Lunte, Zimmer & Heineman, 1990). Effects of the PAHA metabolite on lymphocyte function have also been examined (Wheeler et al., 1988; Adams et al., 1989; Roberts et al., 1989). However, almost nothing is known about potential effects of procainamide metabolites on monocytes/macrophages. The macrophage (M+) has a major role in the immune response both as an essential accessory cell and as a regulatory cell that can either suppress or enhance immune reactions. Two important mechanisms for M+ effects on other ceils are through release of cytokines and reactive oxygen species (ROS). Elicited M+ induced by Corynebacterium p a r v u m can produce and release large amounts of oxygen intermediates that are deleterious to other cells (Johnston, Godzik & Cohn, 1978; Metzger, Hoffeld & Oppenheim, 1980). Exposure of normal human or murine leukocytes, as well as various cell lines, to N-oxidative procainamide metabolites such as PAHA and NOPA in vitro has been shown to cause cytotoxicity and DNA strand breaks (Rubin et al., 1987; Adams et al., 1989; Roberts et al., 1989). The mechanism(s) whereby PAHA and NOPA cause a dose-dependent reduction in mitogen-induced DNA synthesis and a loss of lymphocyte viability is unclear, but recent evidence suggests a major role for ROS produced in vitro by activated monocyte/M+ (Adams et al., 1989). The purpose of the present study was to compare the effects of PAHA and the parent compound procainamide on M+ function in vitro. The effects of procainamide metabolites on the release of ROS by murine M+ was of particular interest. The influence of procainamide metabolites on cytokine release was also studied, involving a cytokine derived primarily from macrophages (interleukin 1, IL-1). The results

indicate that PAHA is capable of producing dosedependent changes in macrophage ROS release, but it had limited effect on IL-1 secretion. EXPERIMENTAL PROCEDURES

Animals

Nine to 12 week old (150-200 g) male Lewis rats were purchased from Charles River Breeding Labs (Wilmington, MA) and used for all rat experiments. Adult male C3H/HEJ (H-2k) mice (Jackson Laboratories, Bar Harbor, ME) were used for thymocyte and spleen cell donors for IL-1 assays of cell-free supernatants collected from rat cells preincubated with test compounds. Adult male DBA/1 (H-2q) and DBA/2 (H-2d) mice (Jackson Laboratories, Bar Harbor, ME) were used for the induction of ROS production by splenic adherent cells following preincubation or co-culture with varying concentrations of test compounds. All animals were housed in AAALAC-approved quarters with free access to food and water until the time of sacrifice. Test chemicals

Procainamide hydroxylamine (PAHA) and nitrosoprocainamide (NOPA) were synthesized as described previously (Budinsky et al., 1987; Wheeler e t a l . , 1988). Procainamide hydrochloride (PA), catalase (bovine liver), and superoxide dismutase (SOD) were purchased from Sigma Chemical Co. (St Louis, MO), and N-acetylprocainamide hydrochloride (NAPA) was a gift from Parke Davis Research Division, Warner Lambert Company (Morris Plains, N J). Preparation o f cells'

Spleens were harvested from individual rats or mice as described previously (Litwin, Bash, Adams, Donovan & Hess, 1979; Wheeler et al., 1988) using chilled Ca 2~- and Mg2+-free Hank's balanced salt solution (HBSS) (Gibco, Grand Island, NY) containing antibiotic and antimycotic mixture. Single cell suspensions were prepared from a pool of 4 - 5 spleens and the cells were washed three times with cold RPMI-1640 medium containing Hepes, 2 mM L-glutamine, and antibiotic and antimycotic mixture. Assay f o r R O S release by naive and elicited cells To produce activated cells, Lewis rats or DBA/I and DBA/2 mice were injected (i.p.) with 300 ~1 C. p a r v u m (CN 6134; Wellcome Research Laboratories, Bechenham, England) or sterile distilled water

PAHA on Reactive Oxygen Species Generation by Macrophages three days prior to sacrifice. The spleens from these animals were harvested and single cell suspensions were prepared in cold HBSS. The cells were washed three times and resuspended in RPMI-1640 medium containing 2007o fetal calf serum (FCS). The monocyte/M~ population was enriched by a standard procedure of adherence to plastic. Briefly, 2 0 - 2 5 × 106 splenic cells were incubated at 37°C for 1.5 h in a humidified atmosphere of 507o CO2/95070 air. The adherent cells were recovered by washing with warm RPMI medium containing 1007o FCS and 0.2507o trypsin (M.A. Bioproducts, Walkersville, MD), followed by a final wash with chilled serumfree medium. The recovered cells were extensively washed, resuspended in RPMI-1640 plus 1007o FCS (complete medium), and assayed for viability by trypan blue dye exclusion and purity using nonspecific esterase staining (Li, Lam & Yam, 1973). The quantitative nitroblue tetrozolium (NBT) reduction assay was performed as previously described (Adams et al., 1989) using minor modifications of the method of Baehner & Nathan (1968) and Rook and coworkers (Rook, Steele, Umar & Dockrel, 1985). In brief, "elicited" and "resident" adherent cells (monocytes/Md~) were recovered, as above, from individual tissue culture flasks, washed three times in phenol red-flee HBSS and adjusted to 3 × 106/ml. Fifty percent of the cells were preincubated with medium alone or with varying concentrations of PAHA, NOPA, PA, or NAPA and were washed with HBSS before the assay. The remainder of the cells were coincubated with and without superoxide dismutase (SOD) and these compounds during the NBT assay phase to assess the effect of direct addition of the compounds on NBT reduction and solubilization. The assay was performed in triplicate using cells alone (resting), cells incubated with latex spherules (0.8/a; Sigma Chemical Co., St Louis, MO) for assessing phagocytosis, and cells incubated with phorbol myristate acetate (PMA; Sigma Chemical Co., St Louis, MO.) at a final concentration of 12.5 ng/ml.

Preparation o f effector and target cells f o r co-culture Spleens were harvested from naive DBA/1 mice as described above. Enriched adherent cell populations were obtained by incubation on Petri dishes (Falcon #1058, Fisher Scientific, Cincinnati, OH). The nonadherent target cells were recovered and washed three times in HBSS and adjusted to 4 × 106/ml in complete RPMI-1640 medium. The non-adherent cells were placed onto F i c o l l - H y p a q u e gradients and the mononuclear cells were recovered at the

811

interface. The recovered cells were passed over Sephadex G-10 columns as previously described (Adams et al., 1989). The G-10 passed cells were washed three times with plain RPMI-1640 medium and resuspended at 4 × 106/ml in complete RPMI1640 medium. A 50/A volume of cell suspension (2 x 105 cells) was transferred to each in series of wells of a flat-bottom microtiter plate. The adherent cells (monocyte/M+) were recovered, washed, and adjusted to 8 × 105 or 1.6 × 10 6 cells/ml. The monocyte/M+ were transferred to a series of sterile 15 ml plastic tubes for pre-incubation with RPMI-1640 media alone or varying dilutions of antioxidants including catalase (233- 930 units/ml) and SOD (158-625 units/ml). The remainder of the cells were transferred to a series of tubes containing varying dilutions of procainamide-related compounds. Both sets of tubes were incubated for 30 min at 37°C. Following incubation, the cells containing the procainamide-related compounds were washed, resuspended to volume in complete RPMI-1640 medium, and transferred to the microtiter wells containing the lymphocytes. The target to effector cell ratios were 10:2 to 10:4. The monocyte/M~ that were preincubated with the antioxidants were centrifuged at 1500 rev./ min × 10 rain, the supernatants decanted, and the cells resuspended in media alone or varying concentrations of the procainamide compounds. These cells were further incubated for 30 min at 37°C. Following incubation, the cells were centrifuged, washed and resuspended to volume in complete RPMI-1640 medium. These effector cells were dispensed into microtiter wells containing lymphocytes to equal an effector-to-target cell ratio of 2:10. Both sets of co-cultures were incubated at 37°C for 2 h with 5 0 CO2/95070 air. Cell viability was assessed using trypan blue dye exclusion. Lastly, media alone or varying dilutions of conconavalin A (Con A) or lipopolysaccharide (LPS; 055:B5; Difco Laboratories, Detroit, MI) were added to appropriate wells. The cultures were incubated at 37°C in a 5°70 CO2/95070 air humidified atmosphere for 3 days and pulsed with 3H-thymidine (0.25/aCi/well) for 16 h before being harvested with an automatic cell harvester (MASH II, Whitaker MA Bioproducts, Walkersville, MD).

lnterleukin 1 assay The ability of whole spleen cells pretreated with PAHA, procainamide, or vehicle alone to produce and secrete IL-1 was assessed as described in detail elsewhere (Gilman, Daniels, Wilson, Carlson &

812

L. E. ADAMSet al.

Lewis, 1984). Briefly, spleen cell suspensions were prepared in supplemented RPMI-1640 complete medium. Splenic cells (5 x 10~) were cultured for 24 h with 10/~g/ml LPS and the supernatant fluids were recovered and passed through a 0.45/~ Millipore filter. IL-1 activity was assayed using a standardized thymocyte costimulatory technique employing C3H/HEJ thymocytes stimulated with submitogenic concentrations of phytohemagglutinin (PHA; 5/~g/ml; Burroughs-Wellcome Co., Research Triangle Park, NC) as described by Scala and Oppenheim (1983). One IL-I unit was defined as twice the counts/min 3H-thymidine incorporated by thymocytes cultured in control medium plus PHA, or the data were calculated on the basis of isotope incorporation of a standard rIL-1B activity (Genzyme, Boston, MA).

Statistical analysis o f data Significant differences between or among experimental groups were determined by either the twotailed Student's t-test (two groups) or analysis of variance using Dunnett's multiple comparison of treatment groups against a control. Data were stored and analyzed on a CLINFO data system. Results are presented as mean _+ S.D. Statistical significance equal P