Inflammation, Vol. 14, No. 5, 1990
RETINOIDS INHIBIT PHOSPHOLIPASE A 2 IN H U M A N SYNOVIAL FLUID AND ARACHIDONIC ACID RELEASE FROM RAT PERITONEAL MACROPHAGES WILLIAM
C. H O P E ,
FIEDLER-NAGY,
BHAVNA
J. P A T E L ,
CHRISTA
a n d B A R B A R A H. W I T T R E I C H
Department of Allergy and Inflammation Research Hoffmann-La Roche Inc. Nutley, New Jersey 07110
Abstract--Retinoids have demonstrated antiinflammatory activity in certain animal models and human disease states. The mechanism by which retinoids elicit this activity is unknown. Some retinoids are known to inhibit aracbidonic acid (AA) release and metabolism in intact cells in vitro. Retinoids may exert their antiinflammatory effects by inhibiting phospholipase A2 (PLA2) and the resultant production of inflammatory AA metaboliteg. Retinoids were evaluated in vitro as inhibitors of the PLA2 activity in human synovial fluid (HSF-PLA~). Of the naturally occurring, nonaromatic retinoids tested, all-trans-retinal, all-trans-retinoic acid (all-trans-RA) and 13cis-RA were the most potent inhibitors (ICso s 6-15/zM), whereas all-trans-retinol was much less potent. Of the synthetic aromatic retinoids and arotinoids examined, the free carboxylic, sulfonic, and sulfinic acid forms were more than 15-fold more potent inhibitors of HSF-PLA2 than their corresponding ethyl esters. These retinoids also were evaluated as inhibitors of calcium ionophore A23187-induced AA release from rat peritoneal macrophages. All-trans-RA and 13-cis-RA were potent inhibitors of AA release from these cells (ICs0 s 4 t~M), while the other natural retinoids were inactive. Of the aromatic retinoids and arotinoids tested, the free acid forms (ICso s 2-6/zM) were 5- to 21-fold more potent inhibitors of AA release from the macrophages than their corresponding ethyl esters. The potencies of the arotinoids as inhibitors of HSF-PLA2 appeared to correlate with their potencies as inhibitors of AA release from A23187-stimulated rat peritoneal macrophages. These data support the hypothesis that one possible mechanism for the known antiinflammatory activity of some retinoids may be by inhibition of phospholipase A2.
INTRODUCTION R e t i n o i d s are v i t a m i n A a n a l o g s that h a v e p r o f o u n d b i o l o g i c a l activities. S e v eral r e t i n o i d s h a v e b e e n r e p o r t e d to h a v e a n t i i n f l a m m a t o r y activity in certain a n i m a l m o d e l s o f arthritis, s u c h as a d j u v a n t - i n d u c e d a n d s t r e p t o c o c c a l cell w a l l 543 0360-3997/90/1000-0543506.00/0 9 1990PlenumPublishingCorporation
544
Hope et al.
induced arthritis in rats (1, 2). Some retinoids also have been shown to possess antiinftammatory activity in man by their ability to modulate inflammatory diseases of the skin (3, 4). It has been reported, for example, that retinoid treatment can inhibit neutrophil accumulation in cutaneous disorders such as psoriasis
(5). The mechanism of the antiinflammatory activity of retinoids has not yet been defined. One possible mechanism for this action of retinoids may be by direct inhibition of phospholipase Az (PLA2), which hydrolyzes 2-arachidonylcontaining phospholipids in cells. This inhibition would lead to the decreased availability of free unesterified arachidonic acid (AA), from which inflammatory metabolites such as leukotrienes and prostaglandins are produced (6). Extracellular PLA2, as well as cellular PLA2, may contribute to the development of inflammatory diseases. For example, elevated levels of proinflammatory PLA 2 activity have been detected in the synovial fluid of patients with rheumatoid arthritis (7). In order to study the role of PLA2 in this inflammatory disease, an assay for the soluble PLA 2 activity in the synovial fluid from rheumatoid arthritic patients was utilized. This assay for human synovial fluid phospholipase A 2 (HSF-PLAz) was used in conjunction with an assay for cellular PLA 2 in rat peritoneal macrophages to evaluate the activity of retinoids and other test compounds as inhibitors in these systems. Retinoids are often referred to as first-, second-, or third-generation retinoids (8). These three structural classes of retinoids are designated as: (I) naturally occurring nonaromatic retinoids, (II) second-generation synthetic aromatic retinoids, and (III) third-generation synthetic polycyclic aromatic retinoids, referred to as arotinoids. Some of the first-generation, naturally occurring retinoids have been reported to inhibit human nonpancreatic PLAzs (9, 10) and snake venom PLA2s (11, 12). Representative compounds from each of the three classes of retinoids were studied for their ability to inhibit HSF-PLA2 activity in our assay system. In addition to the test retinoids, several known inhibitors of PLAzs also were evaluated as inhibitors of HSF-PLA2, for comparative purposes. The known compounds included two reversible PLA2 inhibitors, Upjohn U-10029A, which is a substituted 3-amino-1,2-propanediol (13), and mepacrine (14), and two irreversible PLA2 inhibitors, p-bromophenacyl bromide (15) and manoalide (16). Retinoids have been reported to block both the release and metabolism of AA from rat peritoneal macrophages stimulated by either calcium ionophore A23187, opsonized zymosan, or 12-O-tetradecanoyl-13-acetate (11, 17, 18). Several first- and second-generation retinoids have also been shown to inhibit ionophore A23187-induced leukotriene B4 generation in rat peritoneal polymorphonuclear leukocytes (19). Some of the third-generation retinoids have been suggested to be equipotent dual inhibitors of cyclooxygenase and lipoxygenase pathways of AA metabolism in human platelets and neutrophils (5, 20).
Retinoid Inhibition of Phospholipase A2
545
It is p o s s i b l e t h a t r e t i n o i d s m a y b e a c t i n g at a n e a r l i e r e n z y m a t i c step in t h e p a t h w a y s o f A A m e t a b o l i s m in t h e s e cell t y p e s , s u c h as at t h e l e v e l o f P L A z catalyzed release of AA from membrane-associated phospholipids. This paper r e p o r t s t h e ability o f t h r e e c l a s s e s o f r e t i n o i d s to i n h i b i t b o t h P L A 2 in h u m a n s y n o v i a l fluid a n d A A r e l e a s e f r o m rat p e r i t o n e a l m a c r o p h a g e s .
MATERIALS
AND METHODS
Materials. Synovial fluids from patients with rheumatoid arthritis were supplied by Dr. I.A. Jaffe (Columbia Presbyterian Medical Center, New York, New York). The fluids were centrifuged to remove cells, and the cell-free supematants were pooled. The resultant pool (total protein concentration of 44 mg/ml) was utilized as the enzyme source in the PLAz assay. The substrate for the PLA2 assay was a suspension of [1-14C]oleate-labeled, autoclaved Eseherichia coli (E. coli, specific activity of 0.484 mCi/mmol of phospholipid phosphorus) obtained from Dr. Richard Franson (Medical College of Virginia, Richmond, Virginia). Approximately 95 % of the radiolabeled oleate of the substrate was located at the sn-2 position of membrane phospholipids. The stock buffer solution, 1 M sodium N-2-hydroxyethylpiperazine-N'-2-ethanesulfonate (sodium HEPES, pH 7.4), was obtained from Flow Laboratories (McLean, Virginia). The retinoids and [1-14C]arachidonie acid were synthesized at Hoffmann-La Roche Inc. (Nutley, New Jersey) and were stored desiccated at - 2 0 ~ under argon and protected from light. Upjohn U-10029A, which is 3-[(3-trifluromethyl)phenyl]-l-[methyl(phenylmethyl)amino]-2-propanol,was kindly provided by Dr. Donald Wallach of the Upjohn Company (Kalamazoo, Michigan). Manoalide was a gift from Dr. Paul Scheuer (University of Hawaii at Manoa, Honolulu, Hawaii). Tissue culture reagents were obtained from Grand Island Biological Co. (Grand Island, New York). Calcium ionophore A23187 and other reagents were obtained from 'Sigma Chemical Co. (St. Louis, Missouri). Human Synovial Fluid Phospholipase A2 Assay. The assay for HSF-PLA2 activity, a modification of the method of Franson et al. (21), monitored the release of [1-t4C]oleic acid from the sn-2 position of membrane-associated phospholipids of [1-~4C]oleate-labeled, autoclaved E. coli. The assay was conducted using the radiolabeled E. coli substrate in excess at a final concentration of 20,000 dpm/ml. This was equivalent to 18.2 ~tM of cell membrane phospholipid phosphorus and 2 x 109 autoclaved E. coli/ml. The final optimal conditions which were developed for the standard assay of HSF-PLA 2 inhibitors (in a total volume of 0.5 ml of reaction mixture) included: substrate (20,000 dpm/ml); enzyme (0.10 or 0.15 % HSF, v/v, which was equivalent to 44 or 66 p,g protein/ml); 2 mM CaCI2; 150 mM Na+; 50 mM sodium HEPES buffer, pH 7.3; and 1% dimethyl sulfoxide (DMSO, used to solubilize test inhibitors), in the presence or absence of inhibitor. The reaction was initiated by the addition of HSF-PLA2 and duplicate samples of the mixture were incubated for 30 rain at 37~ The reaction was terminated by the addition of 2.5 ml of chloroform-methanol (1:1.5, v/v). A modified method of Bligh and Dyer (22) for the extraction of lipids was conducted by the further additions of 0.5 ml of chloroform and 1 ml of water with mixing. After centrifugation, the lower chloroform phase was isolated and evaporated to dryness under nitrogen. The lipid residue was redissolved in 50 #1 of a solution containing carder oleic acid (0.2 mg/ml of chloroformmethanol, 9 : 1, v/v). The whole lipid extract was applied to a preactivated (30 min at 110~ silica gel-impregnated glass fiber thin-layer chromatography sheet (ITLC type SG sheet from Gelman Sciences Inc., Ann Arbor, Michigan) and chromatographed for 6 rain using hexane-acetic acid (100:1, v/v) as the developing solvent. This TLC system resolved the enzymatically released
546
H o p e et al.
[14C]oleic acid from the unreacted [ ~4C]phospholipids. The unsaturated lipids were located by exposure to iodine vapor. The oleic acid zone and phospholipid zone were cut out, shaken with 2 ml of ethanol-water (80 : 20, v/v) and 15 ml of Aquasol and counted for radioactivity. A control incubation of substrate in the absence of HSF-PLA2 was performed in each experiment. The HSF-PLA 2 activity was corrected for this nonenzymatic substrate hydrolysis. These optimal conditions resulted in approximately 18% hydrolysis of substrate (corrected for a substrate blank of
RETINOIDS INHIBIT PHOSPHOLIPASE A 2 IN H U M A N SYNOVIAL FLUID AND ARACHIDONIC ACID RELEASE FROM RAT PERITONEAL MACROPHAGES WILLIAM
C. H O P E ,
FIEDLER-NAGY,
BHAVNA
J. P A T E L ,
CHRISTA
a n d B A R B A R A H. W I T T R E I C H
Department of Allergy and Inflammation Research Hoffmann-La Roche Inc. Nutley, New Jersey 07110
Abstract--Retinoids have demonstrated antiinflammatory activity in certain animal models and human disease states. The mechanism by which retinoids elicit this activity is unknown. Some retinoids are known to inhibit aracbidonic acid (AA) release and metabolism in intact cells in vitro. Retinoids may exert their antiinflammatory effects by inhibiting phospholipase A2 (PLA2) and the resultant production of inflammatory AA metaboliteg. Retinoids were evaluated in vitro as inhibitors of the PLA2 activity in human synovial fluid (HSF-PLA~). Of the naturally occurring, nonaromatic retinoids tested, all-trans-retinal, all-trans-retinoic acid (all-trans-RA) and 13cis-RA were the most potent inhibitors (ICso s 6-15/zM), whereas all-trans-retinol was much less potent. Of the synthetic aromatic retinoids and arotinoids examined, the free carboxylic, sulfonic, and sulfinic acid forms were more than 15-fold more potent inhibitors of HSF-PLA2 than their corresponding ethyl esters. These retinoids also were evaluated as inhibitors of calcium ionophore A23187-induced AA release from rat peritoneal macrophages. All-trans-RA and 13-cis-RA were potent inhibitors of AA release from these cells (ICs0 s 4 t~M), while the other natural retinoids were inactive. Of the aromatic retinoids and arotinoids tested, the free acid forms (ICso s 2-6/zM) were 5- to 21-fold more potent inhibitors of AA release from the macrophages than their corresponding ethyl esters. The potencies of the arotinoids as inhibitors of HSF-PLA2 appeared to correlate with their potencies as inhibitors of AA release from A23187-stimulated rat peritoneal macrophages. These data support the hypothesis that one possible mechanism for the known antiinflammatory activity of some retinoids may be by inhibition of phospholipase A2.
INTRODUCTION R e t i n o i d s are v i t a m i n A a n a l o g s that h a v e p r o f o u n d b i o l o g i c a l activities. S e v eral r e t i n o i d s h a v e b e e n r e p o r t e d to h a v e a n t i i n f l a m m a t o r y activity in certain a n i m a l m o d e l s o f arthritis, s u c h as a d j u v a n t - i n d u c e d a n d s t r e p t o c o c c a l cell w a l l 543 0360-3997/90/1000-0543506.00/0 9 1990PlenumPublishingCorporation
544
Hope et al.
induced arthritis in rats (1, 2). Some retinoids also have been shown to possess antiinftammatory activity in man by their ability to modulate inflammatory diseases of the skin (3, 4). It has been reported, for example, that retinoid treatment can inhibit neutrophil accumulation in cutaneous disorders such as psoriasis
(5). The mechanism of the antiinflammatory activity of retinoids has not yet been defined. One possible mechanism for this action of retinoids may be by direct inhibition of phospholipase Az (PLA2), which hydrolyzes 2-arachidonylcontaining phospholipids in cells. This inhibition would lead to the decreased availability of free unesterified arachidonic acid (AA), from which inflammatory metabolites such as leukotrienes and prostaglandins are produced (6). Extracellular PLA2, as well as cellular PLA2, may contribute to the development of inflammatory diseases. For example, elevated levels of proinflammatory PLA 2 activity have been detected in the synovial fluid of patients with rheumatoid arthritis (7). In order to study the role of PLA2 in this inflammatory disease, an assay for the soluble PLA 2 activity in the synovial fluid from rheumatoid arthritic patients was utilized. This assay for human synovial fluid phospholipase A 2 (HSF-PLAz) was used in conjunction with an assay for cellular PLA 2 in rat peritoneal macrophages to evaluate the activity of retinoids and other test compounds as inhibitors in these systems. Retinoids are often referred to as first-, second-, or third-generation retinoids (8). These three structural classes of retinoids are designated as: (I) naturally occurring nonaromatic retinoids, (II) second-generation synthetic aromatic retinoids, and (III) third-generation synthetic polycyclic aromatic retinoids, referred to as arotinoids. Some of the first-generation, naturally occurring retinoids have been reported to inhibit human nonpancreatic PLAzs (9, 10) and snake venom PLA2s (11, 12). Representative compounds from each of the three classes of retinoids were studied for their ability to inhibit HSF-PLA2 activity in our assay system. In addition to the test retinoids, several known inhibitors of PLAzs also were evaluated as inhibitors of HSF-PLA2, for comparative purposes. The known compounds included two reversible PLA2 inhibitors, Upjohn U-10029A, which is a substituted 3-amino-1,2-propanediol (13), and mepacrine (14), and two irreversible PLA2 inhibitors, p-bromophenacyl bromide (15) and manoalide (16). Retinoids have been reported to block both the release and metabolism of AA from rat peritoneal macrophages stimulated by either calcium ionophore A23187, opsonized zymosan, or 12-O-tetradecanoyl-13-acetate (11, 17, 18). Several first- and second-generation retinoids have also been shown to inhibit ionophore A23187-induced leukotriene B4 generation in rat peritoneal polymorphonuclear leukocytes (19). Some of the third-generation retinoids have been suggested to be equipotent dual inhibitors of cyclooxygenase and lipoxygenase pathways of AA metabolism in human platelets and neutrophils (5, 20).
Retinoid Inhibition of Phospholipase A2
545
It is p o s s i b l e t h a t r e t i n o i d s m a y b e a c t i n g at a n e a r l i e r e n z y m a t i c step in t h e p a t h w a y s o f A A m e t a b o l i s m in t h e s e cell t y p e s , s u c h as at t h e l e v e l o f P L A z catalyzed release of AA from membrane-associated phospholipids. This paper r e p o r t s t h e ability o f t h r e e c l a s s e s o f r e t i n o i d s to i n h i b i t b o t h P L A 2 in h u m a n s y n o v i a l fluid a n d A A r e l e a s e f r o m rat p e r i t o n e a l m a c r o p h a g e s .
MATERIALS
AND METHODS
Materials. Synovial fluids from patients with rheumatoid arthritis were supplied by Dr. I.A. Jaffe (Columbia Presbyterian Medical Center, New York, New York). The fluids were centrifuged to remove cells, and the cell-free supematants were pooled. The resultant pool (total protein concentration of 44 mg/ml) was utilized as the enzyme source in the PLAz assay. The substrate for the PLA2 assay was a suspension of [1-14C]oleate-labeled, autoclaved Eseherichia coli (E. coli, specific activity of 0.484 mCi/mmol of phospholipid phosphorus) obtained from Dr. Richard Franson (Medical College of Virginia, Richmond, Virginia). Approximately 95 % of the radiolabeled oleate of the substrate was located at the sn-2 position of membrane phospholipids. The stock buffer solution, 1 M sodium N-2-hydroxyethylpiperazine-N'-2-ethanesulfonate (sodium HEPES, pH 7.4), was obtained from Flow Laboratories (McLean, Virginia). The retinoids and [1-14C]arachidonie acid were synthesized at Hoffmann-La Roche Inc. (Nutley, New Jersey) and were stored desiccated at - 2 0 ~ under argon and protected from light. Upjohn U-10029A, which is 3-[(3-trifluromethyl)phenyl]-l-[methyl(phenylmethyl)amino]-2-propanol,was kindly provided by Dr. Donald Wallach of the Upjohn Company (Kalamazoo, Michigan). Manoalide was a gift from Dr. Paul Scheuer (University of Hawaii at Manoa, Honolulu, Hawaii). Tissue culture reagents were obtained from Grand Island Biological Co. (Grand Island, New York). Calcium ionophore A23187 and other reagents were obtained from 'Sigma Chemical Co. (St. Louis, Missouri). Human Synovial Fluid Phospholipase A2 Assay. The assay for HSF-PLA2 activity, a modification of the method of Franson et al. (21), monitored the release of [1-t4C]oleic acid from the sn-2 position of membrane-associated phospholipids of [1-~4C]oleate-labeled, autoclaved E. coli. The assay was conducted using the radiolabeled E. coli substrate in excess at a final concentration of 20,000 dpm/ml. This was equivalent to 18.2 ~tM of cell membrane phospholipid phosphorus and 2 x 109 autoclaved E. coli/ml. The final optimal conditions which were developed for the standard assay of HSF-PLA 2 inhibitors (in a total volume of 0.5 ml of reaction mixture) included: substrate (20,000 dpm/ml); enzyme (0.10 or 0.15 % HSF, v/v, which was equivalent to 44 or 66 p,g protein/ml); 2 mM CaCI2; 150 mM Na+; 50 mM sodium HEPES buffer, pH 7.3; and 1% dimethyl sulfoxide (DMSO, used to solubilize test inhibitors), in the presence or absence of inhibitor. The reaction was initiated by the addition of HSF-PLA2 and duplicate samples of the mixture were incubated for 30 rain at 37~ The reaction was terminated by the addition of 2.5 ml of chloroform-methanol (1:1.5, v/v). A modified method of Bligh and Dyer (22) for the extraction of lipids was conducted by the further additions of 0.5 ml of chloroform and 1 ml of water with mixing. After centrifugation, the lower chloroform phase was isolated and evaporated to dryness under nitrogen. The lipid residue was redissolved in 50 #1 of a solution containing carder oleic acid (0.2 mg/ml of chloroformmethanol, 9 : 1, v/v). The whole lipid extract was applied to a preactivated (30 min at 110~ silica gel-impregnated glass fiber thin-layer chromatography sheet (ITLC type SG sheet from Gelman Sciences Inc., Ann Arbor, Michigan) and chromatographed for 6 rain using hexane-acetic acid (100:1, v/v) as the developing solvent. This TLC system resolved the enzymatically released
546
H o p e et al.
[14C]oleic acid from the unreacted [ ~4C]phospholipids. The unsaturated lipids were located by exposure to iodine vapor. The oleic acid zone and phospholipid zone were cut out, shaken with 2 ml of ethanol-water (80 : 20, v/v) and 15 ml of Aquasol and counted for radioactivity. A control incubation of substrate in the absence of HSF-PLA2 was performed in each experiment. The HSF-PLA 2 activity was corrected for this nonenzymatic substrate hydrolysis. These optimal conditions resulted in approximately 18% hydrolysis of substrate (corrected for a substrate blank of
