- Email: [email protected]
Biotransformation of para-aminobenzoic acid and salicylic acid by PMN
FreeRadical Biology& Medicine, Vol. 14, pp. 27-35, 1993 Printed in the USA. All rights reserved.
0891-5849/93 $6.00 + .00 Copyright © 1993PergamonPress Ltd.
Original Contribution BIOTRANSFORMATION OF PARA-AMINOBENZOIC AND SALICYLIC ACID BY PMN
ACID
ARTHUR L. SAGONE,JR., ROSE MARIE HUSNEY, and W . BRUCE DAVIS Division of Hematology and Oncology, and Pulmonary and Critical Care, Department of Internal Medicine, The Ohio State University, Columbus, OH 43210, USA (Received 4 March 1992; Revised 9 July 1992; Accepted 30 July 1992) Abstract---Para-aminobenzoic acid (PABA) is an essential cofactor for the production of folic acid in bacteria and has mild anti-inflammatory activity. We have recently reported that salicylic acid and benzoic acid are oxidized by stimulated granulocytes Polymorphonuclear Neutrophils (PMN). The oxidation of salicylate appears mediated by a potent oxygen metabolite generated during the respiratory burst which is dependent primarily on superoxide (02) for its production. These background studies with the salicylate group of drugs suggested that PABA might be similarly metabolized by PMN. In these studies, we demonstrate that PABA is metabolized by stimulated PMN. However, in contrast to the biochemical mechanism involved in the metabolism of salicylate, our scavenger studies indicate that PABA is metabolized primarily by the myeloperoxidase pathway. Our results may explain the mild anti-inflammatory actions of the drug and suggest that the degradation of PABA by PMN at an inflammatory site may limit the availability of PABA for bacterial growth. Keywords--Para-aminobenzoic acid, Salicylic acid, Hydroxyl radical, Myeloperoxidase, Biotransformation of xenobiotics, Anti-inflammatory drugs, Reactive oxygen species, Activated granulocytes, Free radicals
rine. 9-12 The possible products include a chloramine, a hydroxylamine, or chlorination of the benzene ring. Second, Chance demonstrated almost 30 years ago that PABA is oxidized directly by a complex of H202 and horseradish peroxidase (Compound I).13 This observation also raised the possibility that a similar mechanism for the oxidation of PABA might occur in stimulated PMN. The metabolism of PABA by stimulated granulocytes could destroy its availability for normal folic acid metabolism by bacteria. ~4 If so, the degradation of PABA by granulocytes might represent an unrecognized microbicidal mechanism for granulocytes. In these studies we have demonstrated that PABA, similar to salicylic acid, is decarboxylated by stimulated granulocytes. However, in contrast to the mechanism involved in the metabolism of the salicylate, our scavenger studies indicate that PABA is metabolized primarily by the myeloperoxidase pathway.
INTRODUCTION
Para-aminobenzoic acid (PABA) is an essential cofactor for the production of folic acid in microorganisms. This compound is also known to have mild anti-inflammatory activity and has been reported to decrease the fibrosis occurring in scleroderma and Peyronie's disorder) '2 We have recently reported that benzoic acid and salicylic acid undergo decarboxylation and hydroxylation by stimulated granulocytes. 3-6 Our data suggest that the reaction is mediated by a potent radical which is dependent primarily on superoxide for its production. These background studies with the salicylate drugs suggested that PABA might be similarly metabolized by granulocytes. However, we also felt that PABA might be metabolized by the myeloperoxidase pathway. First, PABA has an amine group, and we postulated that this compound might be oxidized by the hypochlorous acid (HOCI) produced by the myeloperoxidase pathway in a manner similar to the metabolism of d a p s o n e 7,s o r tau-
MATERIALS AND METHODS
Address correspondence to: Dr. Arthur L. Sagone, Jr., The Ohio State University, N1011 Doan Hall, 410 West 10th Avenue, Columbus, OH 43210.
p-amino[carboxyl-14C]benzoic acid (specific activity = 56.6 or 54.5) was obtained from Amersham, 27
28
A.L. SAGONE,JR. el aL
Arlington Heights, IL, and [carboxyl-~4C]salicylic acid (specific activity 56.6) from ICN Radiochemicals, Irvine, CA. Methionine, sodium azide, sodium benzoate, glutathione (GSH), dimethyl sulfoxide (DMSO), taurine, catalase (14,100 U/mg protein), PABA (free acid, purity 99+%), para-aminosalicylic acid, sodium salt (PAS), superoxide dismutase (SOD) #S-8254 (2725 U/mg protein) and #S-2515 (3250 U/ mg protein), phorbol myristate acetate (PMA), and zymosan (lot #71F-8086) were obtained from Sigma Chemical Co., St. Louis, MO. Dimethylthiourea (DMTU), sodium hypochlorite (NaOC1), and sodium salicylate were from Aldrich Chemical Co., Milwaukee, WI. Mannitol was obtained from Abbott Laboratories, North Chicago, IL, and phenol (molecular biology grade) from Fisher Scientific, Fair Lawn, NJ. Dulbecco's phosphate buffered saline (DPBS) was from GIBCO, Grand Island, NY.
Cell preparation Fresh blood was obtained for each experiment from healthy volunteers who had taken no medications for at least 72 h. Purified neutrophil suspensions were isolated by dextran sedimentation and Ficoll-hypaque density gradient centrifugation, as previously described. 3-6
Metabolic studies We determined the rate of decarboxylation of [Carboxyl-~4C]salicylic acid, ([14C]salicylate), and of p-amino[carboxyl-14C]benzoic acid ([14C]PABA) by suspensions of unstimulated and stimulated neutrophils. As in previous studies, 3'5 an ionization chamber electrometer apparatus was used to continuously monitor the production of ~4COz from this substrate. Purified neutrophils (5 × 106/ml) were resuspended in Dulbecco's phosphate-buffered saline containing 50 mg/dl of glucose, and 4-ml aliquots of the suspension were placed into 25-ml triple-arm distilling flasks containing either 1 uCi of [~4C]salicylic acid or 2 #Ci of [~4C]PABA (in 0.1 ml or 0.2 ml normal saline, respectively). The flasks with salicylate also contained unlabeled substrate. The final cell concentration was 4 x 106]ml. The final substrate concentration for the salicylate was 104 #M and for the PABA, 8 #M. After incubating the suspensions at 37 ° for 45 to 50 min, DPBS alone or an equal volume of various free radical inhibitors (in DPBS) were added, and 10 to 15 min later, PMA (in DPBS-0.10% DMSO) was added to stimulate the neutrophils. The final concentration of PMA was 50 ng/ml, while the final concentration of the original diluent, DMSO, was 0.37 raM. In experi-
ments where opsonized zymosan was used to stimulate the cells, the particles were added in a saline suspension such that the final concentration was 1 mg/ ml. The amount of ~4CO2 thus generated from the labeled compounds was calculated from the peak millivolt value achieved and was expressed as nanomoles 14CO2 produced per hour per 107 cells.
Experiments with cell free systems We have previously published evidence that salicylate reacts with HOC1 in a cell-free system. 6 In these experiments, we determined the capacity of solutions of hypochlorite to decarboxylate [~4C]salicylate. Initial studies established that HOCI decarboxylates [~"C]salicylate (0.1 mM) and that the rate of the reaction increased as the concentration of HOC1 was increased from 0.005 mM to 0.5 mM. Following this, additional experiments were done to determine the relative effects of scavengers on the reaction. In these flasks, similar quantities of radioactive and unlabeled salicylate were added as in the previously described experiments, along with 4 ml of DPBS. After incubating several minutes at 37°C with stirring to establish zero baseline, scavengers were added; then 0.1 cc of NaOC1 was added. The final concentration of HOCI used for these experiments was 0.05 mM.
Neutrophil stable oxidants and hypochlorous acid (HOC1) production Stimulated granulocytes release a stable oxidant into the supernatant, which has been identified primarily as taurine chloramine produced by the reaction of HOCI with cellular taurineJ ° These experiments determined the effect of salicylate or PABA on the production of stable oxidants and HOC1 by stimulated neutrophilsJ ~'12 The cell incubations were prepared in three-neck flasks in a fashion similar to that of the metabolic studies, except that radioactive substrates were not added to the incubations. For the experiments estimating total HOC1 production, taurine (10 or 15 mM final concentration) was also added to the flasks to trap the HOCI as taurine chloramine. The cells were preincubated for 1 h at 37°C with stirring in the presence of a range of concentrations of either salicylate or PABA (or DPBS alone). This was done on the ion chamber apparatus at the same airflow as for the metabolic studies. After the preincubation, either PMA (50 ng/ml final concentration) or zymosan (1 mg/ml final concentration) was added and incubated for 1 h. The reactions were terminated by placing the flasks into an ice bath and by immediately adding 25 ug/ml of catalase. Cell-free superna-
Biotransformation of PABA by PMN
29
1412 10 (n l.J
0 >
8
u
..I ..I
6 4
PMAAde~d d
2 0 0
20
40
60
80
100
120
140
160
180
MINUTES Fig. 1. The metabolism of para-aminobenzoic acid by granulocyte suspensions. The results indicate a typical experiment which is representative of six done. The curve was derived by computer analysis using the indicated data points (rq) from the continuous recorder tracing of the electrical output of the electrometer. The y axis indicates the millivolt signal generated by ~4CO2in the ionization chamber. The x axis indicates the time in minutes. The addition of PMA resulted in a prompt increase in the electrical signal.
tants were obtained by centrifugation of the suspensions at 27,700 × g. The capacity of aliquots to oxidize I- to I2 was determined as previously described.~l'~2 The results were expressed as nanomoles of I2 produced per 10 7 cells, ll'12
Statistics Data were analyzed according to the test for independent samples. Results were expressed as mean ___SD. RESULTS
Unstimulated suspensions of granulocytes did not oxidize significant amounts of PABA. The addition of PMA resulted in a prompt production of CO/, which was maximal at approximately 30 min (see Fig. 1). Similar results were found when zymosan was used as a stimulus (data not shown). These observations indicate that both zymosan and PMA-stimulated granulocytes decarboxylate PABA. The maximum rate of CO2 production from PMN suspensions stimulated with PMA was 0.78 +__0.36 (SD) nanomoles per 10 7 cells per hour, approximately two- to threefold the rate of cells stimulated with zymosan, which was 0.29 ___0.10 (SD) nanomoles per 10 7 cells per hour.
Effect of scavengers on the decarboxylation of [14C]PABA by either zymosan or PMA The addition of superoxide dismutase to the suspensions prior to incubation with PMA enhanced PABA oxidation twofold (see Table 1). Catalase, a scavenger ofHzO2, and two HOCI scavengers, methionine and taurine, were relatively potent inhibitors of the reaction, while DMSO had no effect. Azide also inhibited the reaction. The results of scavenger studies done with zymosan-stimulated cells were less clear. The addition of superoxide dismutase, catalase, and two scavengers of HOC1 did not appreciably affect the rate of decarboxylation of PABA by zymosan-stimulated granulocytes (Table 1). Sodium azide (10 -4 M) did produce an approximately 40% suppression of the reaction. Similar to the results with PMA, DMSO did not impair the reaction. However, DMTU, a nonspecific free radical scavenger,12 did impair the decarboxylation 89%.
Effect of scavengers on the decarboxylation of [t4C]salicylate by PMA-stimulated PMN We have previously reported the effects of scavengers on the oxidation of salicylate by zymosan and PMA on stimulated PMN. 5,6 However, in our prior
30
A . L . SAGONE, JR. et al. Table 1. Effect of Scavengers on the Rate o f 14CO2Production from a 0.008 m M Solution of [~4C]PABA by Stimulated Granulocytes
Scavenger Azide(0.1 mM) SOD (10 #g/ml) Catalase (100 ~g/ml) Methionine (1 mM) Taurine (5 mM) DMSO (30 mM) D M T U (10 mM)
PMA
Percent o f ControP P Value
33.6_+ 11.2(3) 209.3_+76.8(3) 26.3 _+ 13.4 (3) 32.7 _+ 3.1 (3) 52.7 _+ 5.7 (3) 115.7 (1) --
<0.02 <0.01 < 0.01 < 0.01 < 0.02
Zymosan
P Value
61.7_+ 8.7(3) 95.0_+ 4.1 (3) 90.7 _+ 28.8 (4) 121.6 _+ 19.5 (3) 103.1 _+ 18.2 (3) 91.7 (1) 11.4 (1)
<.1 NS NS NS NS
a The results were derived from paired experiments. The number of experiments done is indicated in parentheses. The mean maximal rate of 14CO2 production for PMA-stimulated cells was 0.78 _+ 0.36 nanomoles per l07 cells per hour (six experiments). The mean maximal rate of ~4CO: production for zymosan-stimulated cells was 0.29 _+ 0. l0 nanomoles per 10 7 cells, per hour (seven experiments). See Fig. 1.
studies, we did not evaluate extensively the effect of scavengers on the decarboxylation of salicylate induced by PMA stimulated PMN. The unexpected effect of a number of scavengers on the decarboxylation of PABA by PMA-stimulated PMN prompted us to do similar scavenger studies with salicylate as a substrate. The results are given in Table 2. Similar to our prior results, SOD inhibited salicylate decarboxylation. This effect is opposite to that observed in our experiments with PABA (see Table 1). Catalase also inhibited the reaction. However, three scavengers of HOCI (i.e., GSH, taurine, and methionine) had no effect. Mannitol, a scavenger of "OH, was similarly ineffective. Phenol, in an equimolar concentration to salicylate, decreased the reaction 72%. DTMU, which is a known scavenger of "OH and HOC1, was ineffec-
tive at a 0.1 mM concentration, and at a 1 mM concentration produced only 18% inhibition (p > .3). A 100-fold higher concentration of DMTU (10 mM) than salicylate was required to produce significant inhibition (62%) (p < .01). Both DMSO and benzoate, when used in high concentrations, inhibited the reaction approximately 50%. PABA in an equimolar concentration produced 25% inhibition (p > .3), while a 1-mM concentration decreased the rate of oxidation 66%. Overall, the results of these experiments with PMA are similar to those which we have previously reported for zymosan-stimulated granulocytes. However, there is one difference. In these experiments, we did demonstrate some specific inhibition of the reaction by catalase. In contrast, we did not observe a major inhibitory effect of catalase in our studies of
Table 2. Effect of Scavengers on the Rate of Decarboxylation o f a 0.1 m M Solution o f Salicylate by PMA-Stimulated Granulocytes Scavenger Catalase Heat-inactivated catalase SOD GSH Taurine Methionine Mannitol Phenol Phenol DMTU DMTU DMTU DMSO DMSO Benzoate PABA PABA
Concentration 25 #g/ml 25 t~g/ml 10 #g/ml 1 mM 5 mM 1 mM 40 m M 0.01 m M 0.1 m M 0.1 m M 1 mM 10 m M 20 m M 100 m M 20 m M 0.1 m M 1.0 m M
Percent of ControP 40.1 _+ 5.6 82.7 -+ 5.6 44.7 _+ 12.5 100.3 _+ 30.2 116.8 _+ 33.1 102.0 _+ 30.0 114.0 _+ 19.3 37.8 (1) 28.3 (1) 98.2 _+ 16.4 82.6 _+ 26.6 38.0 -+ 10.8 53.9 (2) 48.0 _+ 3.9 40.0 ( 1 ) 74.1 _+ 13.7 33.7 _+ 9.9
(5) (3) (4) (4) (3) (4) (3)
P Value
(3)
< .01 <. 1 < .02 NS NS NS NS --NS > .3 < .01 -< .05
(3) (6)
< .3 < .05
(3) (4) (3)
--
a The mean maximum rate of salicylate oxidation for the controls was 4.8 _+ 2.2 nanomoles per hour per 10 7 cells (27 experiments). The results were derived from paired experiments. The number of experiments done is indicated in parentheses.
Biotransformation of PABA by P M N Table 3. Effect of Scavengers on the Rate of CO2 Production from Salicylate (0. l raM) by HOCI (0.05 mM) Concentration of Additive
Percent of ControP
P Value
Mannitol (40 mM) Benzoate (1 mM) (20 mM) DMSO (0.37 mM) b (10 mM) (20 raM) (100 mM) Phenol (0.01 mM) (0.1 mM) PABA (0.1 raM) PAS (0.1 mM) GSH (1 mM) Methionine (1 raM) Taurine (5 mM) SOD (10 #g/ml) Heat-inactivated SOD (10 #g/ml) Catalase (25 ~tg/ml) Heat-inactivated catalase (25 ~g/ml)
98.5 + 9.3 (3) 94.3 (1) 89.7 _+ 20.6 (3) 28.3 (2) 12.2 (2) 8.8 (2) 2.1 (1) 59.0 (2) 39.9 (2) 12.0 (2) 27.1 ( 1) 0 (3) 0.2 _+ 0.4 (3) 3.5 + 0.3 (3) 84.7 _+ 17.8 (4)
NS -NS --------< .001 < .01 < .02 > .1
80.1 _+ 6.8 (3) 50.6 + 3.4 (4)
< .01 < .05
59.5 + 8.2 (3)
< .05
a The mean maximal rate of decarboxylation o f a 0.1 m M solution of salicylate by a 0.05 m M concentration of HOCI was 21.1 + 2.4 nanomoles/h for nine experiments. The number of experiments done with the scavenger is indicated in the parentheses. b A final concentration of 0.37 m M DMSO was present in the cell experiments in which PMA was used as a stimulus, since this compound was used to dissolve the PMA.
zymosan-stimulated granulocytes or in our experiments involving the hydroxylation of salicylate by PMA-stimulated granulocytes. 5'6 The inhibition ofsalicylate oxidation by SOD confirms the importance of O~ in the biotransformation of this drug. The inhibition of salicylate oxidation by catalase and SOD supports the hypothesis that the decarboxylation ofsalicylate may be mediated in part by "OH produced by a Haber-Weiss type (superoxide-driven Fenton) reaction. However, in view of the problematic nature of scavenger studies in a system as complex as the cell, our results do not exclude the possibility that more than one mechanism may be involved in the decarboxylation of salicylate by PMA-stimulated PMN (i.e., one dependent on 0 2 and a second dependent on H202) or the possibility that the oxidizing species dependent on 02- for its production may not be "OH but rather another ROS of similar reactivity.
Effect of scavengers on salicylate decarboxylation by HOC1 The addition of HOC1 (final concentration of 0.05 mM) to a 0.1-mM solution of salicylate stimulated decarboxylation. As indicated in Table 3, mannitol and benzoate did not impair the oxidation of salicylate by HOCI. DMSO inhibited the reaction at rela-
31
tively low concentrations. PABA (0.1 mM) inhibited the reaction 88%. Para-aminosalicylic acid (PAS), a structurally similar compound, inhibited the reaction 73%. Phenol also inhibited the reaction but was somewhat less potent than PABA or PAS. Catalase and SOD inhibited the reaction 15.3% and 49.4%, respectively. However, the effects were nonspecific, since similar results were observed with the heat-inactivated enzymes. The magnitude of nonspecific inhibition by catalase was somewhat unexpected. Three compounds known to react rapidly with HOC1, namely methionine, GSH, and taurine, inhibited the reaction completely. These experiments confirm our prior observation, which indicates that HOCI can metabolize salicylate. 6 However, the results of the scavenger studies in this cell-free system suggest that a reaction of HOCI with salicylate is not a major mechanism for the decarboxylation of salicylate by PMA stimulated PMN. First, GSH, methionine, and taurine, in concentrations known to impair HOC1 production by PMA-stimulated PMN and which ablated salicylate decarboxylation in the cell-free system, did not inhibit salicylate oxidation by PMA-stimulated PMN (comparison of Tables 2 and 3). Second, the inhibition of the reaction by SOD in the cell system is also hard to resolve with a primary role of HOC1 in the decarboxylation ofsalicylate, since the production of HOC1 is dependent primarily on the oxidation of C1- by H202 and myeloperoxidase (MPO) (Compound I). Finally, a low concentration of DMSO (0.37 mM) inhibited salicylate oxidation by HOCI approximately 70%. This observation suggests that the rate constants for the reaction of DMSO and salicylate with HOCI are similar. Wasil et al. have previously suggested that while DMSO can react with HOCI in a cell-free system, the reaction may be too slow for a major interaction of DMSO with HOC1 in a biological system) 5 Our observations suggest that this may also be true for salicylate, unless high concentrations of this drug are present. Further, as discussed under Materials and Methods, DMSO was present in a final concentration of 0.37 mM in our cell experiments with PMA since this compound was required to dissolve the PMA. This concentration of DMSO appears sufficient to prevent most of the potential decarboxylation of salicylate by the HOCI produced by the PMN under our experimental conditions.
Effect of PABA and salicylate on the recovery of stable oxidants The effect of PABA on the production of stable oxidants by stimulated PMN is given in Table 4. As
32
A. L. SAGONE, JR. el aL Table 4. Effect of PABA on the Production of Stable Oxidants by Stimulated Cellsa
Source of Supernatant Stim. cells alone Stim. cells + 0.1 mM PABA Stim. cells + 0.5 mM PABA Stim. cells + 1.0 mM PABAb
PMA Stimulated
Zymosan Stimulated
34.1 + 9.7 (4) 31.2+10.1(4) 26.9 + 5.6 (3) 18.3 _+ 4.2 (4)
24.8 + 6.6 (3) 19.4_+7.0(3) 15.2 + 6.3 (3) 10.7 _+ 4.9 (3)
Results indicate the nanomoles ___SD of 12 produced by l0 7 cells in 1 h. Since 1 mol oftaurine chloramine oxidizes 2 mols of I-, the amount of 12 produced equals the amount of taurine chloramine generated by the cells. The number of experiments done with each concentration is indicated in parentheses. b p value for both zymosan and PMA < .05.
indicated, a 0.1-mM concentration of PABA did not alter the recovery of stable oxidants. A 1-mM concentration, however, was associated with a decreased recovery from both zymosan and PMA-stimulated PMN suspensions. This observation also appears to exclude the conversion of PABA to a stable chloramine by stimulated PMN. The observation that salicylate can be metabolized by HOCI in a cell-free system prompted us to do additional studies concerning the effects of salicylate on the activity of the MPO pathway. Therefore, we quantitated the effects of salicylate on the production of HOCI as determined by the taurine assay (Table 5). While a 0. l-mM concentration of salicylate did not definitely impair the production of taurine chloramine by zymosan or PMA-stimulated PMN, higher concentrations did have a significant effect. A 10-mM concentration inhibited the production of taurine chloramine almost completely. A 1-mM concentration inhibited the reaction 70% when PMA was used as a stimulant and 50% when zymosan was the stimulus. Salicylate had similar effects on the production of stable oxidants (see Table 5). In contrast, there was no effect of salicylate (0.1 to 10 mM) on the recovery of taurine chloramine when added directly to supernatants containing taurine chloramine (i.e., supernatants recovered from granulocyte suspensions which had been incubated with PMA for 1 h). This observation indicates that salicylate does not react directly with taurine chloramine or interfere with the I- assay which is used for its quantitation. DISCUSSION
These data indicate that zymosan- and phorbolstimulated granulocytes decarboxylate para-aminobenzoic acid. The stimulation of the cells by PMA was associated with a higher rate of reaction than that of cells stimulated by zymosan. The scavenger experi-
ments done with zymosan were not definitive. There was some inhibition of the reaction by azide but no effect of superoxide dismutase, catalase, methionine, or taurine. DMTU, a nonspecific ROS scavenger, did impair the reaction. This suggests that the reaction is mediated by a reactive oxygen species (ROS) produced during the respiratory burst. We felt that the failure of our scavengers to impair PABA decarboxylation might be due to their inability to penetrate to the cellular sites of PABA oxidation in concentrations adequate to inhibit the reaction. Therefore, we studied PMA-stimulated PMN since no phagolysosome develops in PMN following stimulation by this agent. Under these circumstances, a significant portion of the reaction is probably occurring at the cell membrane. The scavenger studies done with the PMA-stimulated granulocytes suggest that the main mechanism for the reaction involves myeloperoxidase. There was an inhibition with azide as well as two scavengers, methionine and taurine, which are known to be metabolized by the myeloperoxidase pathway. Similarly, catalase was able to inhibit the reaction, while the addition of superoxide dismutase actually stimulated the rate of decarboxylation. These results are consistent with the metabolism of PABA by the myeloperoxidase pathway) 2 The reasons for the difference noted in the scavenger studies done with PMA and zymosan are not entirely clear. It seems likely that the biochemical mechanism for the oxidation of PABA by zymosan-stimulated PMN is similar to that of PMA. However, we cannot exclude the possibility that PABA decarboxylation occurs by another mechaTable 5. Effect of Salicylate on the Production of Stable Oxidants and Taurine Chloramine by Stimulated PMN a Stimulant Source of Supernatant Stim. cells Stim. cells+0.1 mMsalicylate Stim. cells + 1.0 mM salicylate Stim. cells + 10 mM salicylate Stim. cells + taurine Stim. cells + taurine + 0.1 mM salicylate Stim. cells + taurine + 1.0 mM salicylate Stim. ceils + taurine + 10.0 mM salicylate
PMA 18 18 2 5 73.2
+ 5 (3) _+ 3(3) +_ 2 (3) + 1 (3) + 32.6 (4)
Zymosan 21 _+ 4 (3) 18 _+ 4(3) 9 + 11 (3) 5 _+ 1 (3) 194.0 (2)
52.8 _+ 33.8 (4)
198.0 (2)
21.5 _+ 2.5 (4)
96.5 (2)
8.0 + 3.9 (4)
8.0 (2)
a The concentration of taurine used for these experiments was 15 mM with PMA as stimulant, l0 mM with zymosan as stimulant. The results are given as nanomoles of 12 -- SD produced by l07 cells in 1 h and equal the amount of taurine chloramine formed. The number of experiments done with each concentration is indicated in parentheses. PMA experiments done with the 1 mM and l0 mM concentrations of salicylate are significantly lower than controls--p values < .05.
Biotransformation of PABAby PMN nism in zymosan-stimulated PMN even though these cells are generating substantial amounts of H O C I . 9-12 This possibility seems less likely. 9-12 Our results indicate that the mechanism for the biotransformation of PABA by stimulated PMN differs from that of benzoic and salicylic acid, since we found that the decarboxylation and hydroxylation of both of these latter substrates was inhibited by superoxide dismutase, while the addition of this enzyme enhanced PABA oxidation. Recently, Dull et al. reported that the oxidative deamination of 5-aminosalicylic acid (ASA) by activated PMN could be inhibited by the addition of superoxide dismutase to the incubation) 6 Therefore, in spite of the structural similarity of PABA to these compounds, it appears to be metabolized by a different mechanism. Recent studies in our laboratory have provided further evidence that the oxidation of PABA and its structurally similar analog, para-aminosalicylic acid (PAS), are oxidized by the MPO pathway. We have demonstrated that stimulated PMN generate products from PABA and PAS which can be detected by high-pressure liquid chromatography (HPLC). The primary product appears to involve chlorination at the 3 position of the benzene ring) 7 Similar to decarboxylation, the formation of the product in PMA cultures can be enhanced by the addition of SOD and inhibited by the addition ofcatalase and HOC1 scavengers. Additional experiments have demonstrated that an identical product can be detected in a cell-free system by the reaction of HOC1 with PABA (unpublished observations) and PAS) 7 The addition of PABA to both zymosan- and PMA-stimulated granulocytes decreased the recovery of stable oxidant activity in the supernatants of the suspensions when higher concentrations of the PABA were used. These observations indicate that PABA impairs the production of stable oxidants by PMN. Further, these findings suggest that PABA is not metabolized to a stable chloramine as is the case with amino acids such as taurine. 9-12 Similar to PABA, we demonstrated that higher concentrations of salicylate impair the production of hypochlorous acid, as measured by the taurine chloramine assay, and of stable oxidants. Similar results have been reported by Shacter et al. with murine neutrophils.t8 This finding is somewhat unexpected and suggests that the anti-inflammatory effect of salicylate may be partly due to its ability to decrease the production of HOCI by PMN as well as the production of another potent ROS. As discussed, a reaction of salicylate with HOC1 or chloramines does not appear to relate to the decarboxylation or hydroxylation of salicylate by PMN. Therefore, the mechanism by which salicylate impairs
33
HOC1 production by PMN may involve a reaction which is not associated with the oxidation of salicylate. In this regard, Kettle and Winterbourn 19 have reported that several anti-inflammatory drugs, including salicylate, decrease the production of HOC1 from H202 and MPO (Compound I) by promoting the irreversible formation of Compound II. This mechanism seems consistent with the impaired production of HOC1 by salicylate observed in our experiments but requires further testing. The capacity of salicylate to impair HOCI production again raises the issue of whether or not salicylate oxidation by PMN is dependent on the MPO pathw a y . 5'6 However, at this point the mechanism for the oxidation of salicylate by activated PMN remains problematical. As discussed in this study and our prior ones, we demonstrated that stimulated PMN decarboxylate and hydroxylate salicylate. This oxidation of salicylate by PMN is analogous to its oxidation by "OH in other systems. 5'6'2° We were unable to impair salicylate hydroxylation by PMN with deferoxamine. 6 This observation suggests that the oxidation of salicylate measured under our experimental conditions cannot be explained simply on the basis of iron contamination in our buffer resulting in the artifactual production of "OH from H202 by a Fenton type of reaction. 2t'22 However, while our results are consistent with the hypothesis that the oxidation of salicylate by stimulated PMN might be mediated by "OH, our results have not conclusively proven this point; nor do they prove that the mechanism involves the Haber-Weiss reaction as classically defined, since a number of our observations are not entirely consistent with this model. First, while we have consistently observed that SOD inhibits salicylate oxidation, indicating that 02 appears to be required for this reaction, we have been unable to prove that H202 is essential for the reaction. 5'6 Second, we have been unable to impair salicylate oxidation by mannitol, even under experimental conditions in which a substantial portion of the reaction should be occurring extracellulady (i.e., PMA stimulation of PMN). Finally, the molar ratio of 2,3-dihydroxybenzoic acid to 2,5-dihydroxybenzoic acid predicted by the random attack of "OH on salicylic acid is approximately 1:1.23 While we did detect some 2,3-dihydroxybenzoic acid in our cell studies, the amount was small, and the ratio of 2,3-dihydroxybenzoic acid to 2,5-dihydroxybenzoic acid was only 1:10, 6 a pattern of hydroxylation inconsistent with the primary oxidation of salicylate by free " O H . 23
Babior has previously suggested that alkyl hydroperoxides might be related to the oxidation of organic compounds by PMN. 24 This raises the possibility that
34
A.L. SAGONE,JR. et al.
the oxidation of salicylate is mediated by an alkoxyl radical (RO") produced from an organic peroxide by a ferrous iron complex or iron containing protein. 25,26 Such a reaction could be mediated by MPO if 02 reduces the Fe +÷+ in this enzyme to Fe +÷ following the metabolic burst. In this regard, 02 is known to form a complex with peroxidase forming an oxyperoxide (Compound III). 27 Another possible mechanism for the oxidation of salicylate by PMN is the production of ferryl iron in MPO or another iron protein during the metabolic burst. It has recently been reported that salicylate can be hydroxylated by H202 in the presence of myoglobin, most likely due to the production of ferryl iron. 28 A similar oxidation of iron may occur during the interaction of H202 with a peroxidase (Compound I). 2s-3° However, as discussed previously, we were unable to detect the hydroxylated products of salicylate in a cell-free system using the H202-MPO enzyme system. 6 Further, our inability to reliably decrease salicylate oxidation by catalase is not consistent with this mechanism. At this point, the biochemical mechanism for the oxidation of salicylate remains unclear. Further, while a single mechanism may be involved, salicylate oxidation may be mediated by more than one mechanism, and by more than one ROS, including "OH produced by a Haber-Weiss type of reaction or another mechanism. In summary, our data supports the following conclusions. First, PABA is metabolized by activated granulocytes by the MPO pathway. This may explain in part the reported anti-inflammatory properties of this drug in scleroderma and Peyronie's disorderJ ,2 The possibility that rapid degradation of PABA by activated PMN at an inflammatory site may be sufficient to limit the production of folic acid by microorganisms in vivo needs additional study since it may represent an unrecognized microbicidal mechanism. Second, our studies confirm that the decarboxylation of salicylate is dependent on 02 and cannot be inhibited by HOC1 scavengers. Therefore, in spite of the structural similarities between PABA and salicylate, these drugs appear to be metabolized differently by stimulated granulocytes. These differences appear to reflect a number of variables such as the different rate constants of the drug for reaction with the ROS produced by phagocytes, and possibly their cellular distribution. Finally, our studies indicate that concentrations of salicylate which occur in vivo impair the production of HOCI and chloramines by PMN. Therefore, the
anti-inflammatory effect of this drug may also be related to its capacity to impair the MPO activity of PMN as well as scavenge other ROS produced during the metabolic burst. The mechanism of the decreased HOC1 production by salicylate does not appear to involve a direct reaction ofsalicylate with hypochlorous acid or chloramine and is consistent with a direct inhibition of MPO activity, as suggested recently by Kettle and Winterbourn. 19 Acknowledgement - - Supported by grants from The Ohio Cancer Research Associates (OCRA), the NCI RO1-CA-32321, and The American Lung Association.We would like to thank Barbara Baker for her help in typing this manuscript.
REFERENCES 1. Zarafonetis, C. J. D. The treatment of scleroderma: Results of potassium para-aminobenzoate therapy in 104 cases. In: Mills, C.; Moyer, J. H., eds. Inflammation and diseases of connective tissue. Philadelphia: W. B. Saunders Co.; 1961:688-696. 2. Zarafonetis, C. J. D. Antifibrotic therapy with potaba. Am. J. Med. Sci. 248:550-561; 1964. 3. Sagone, A. L., Jr.; Decker, M. A.; Wells, R. M.; Democko, C. A new method for the detection of hydroxyl radical production by phagocytic cells. Biochem. Biophys. Acta. 628:90-97; 1980. 4. Alexander, M. S.; Husney, R. M.; Sagone, A. L., Jr. Metabolism of benzoic acid by stimulated polymorphonuclear cells. Biochem. Pharmacol. 35:3649-3651; 1986. 5. Sagone, A. U, Jr.; Husney, R. M. Oxidation of salicylates by stimulated granulocytes: Evidence that these drugs act as free radical scavengers in biological systems. J. Immunol. 138:2177-2183; 1987. 6. Davis, W. B.; Mohammed, B. S.; Mays, D. C.; She, Z.-W.; Mohammed, J. R.; Husney, R. M.; Sagone, A. L., Jr. Hydroxylation ofsalicylate by activated neutrophils. Biochem. Pharmacol. 38:4013-4019; 1989. 7. Uetrecht, J.; Zahid, N.; Shear, N. H.; Biggar, W. D. Metabolism ofdapsone to a hydroxylamine by human neutrophils and mononuclear cells. J. Pharmacol. and Exp. Therap. 245:274279; 1988. 8. Uetrecht, J. P. Drug-induced agranulocytosis and other effects mediated by peroxidases during the respiratory burst. In: Sbarra, A. J.; Strauss, R. R., eds. The respiratory burst and its physiological significance. New York: Plenum Press; 1988:233-243. 9. Weiss, S. J.; Klein, R.; Slivka, A.; Wei, M. Chlorination of taurine by human neutrophils. J. Clin. Invest. 70:598-607; 1982. 10. Thomas, E.; Grisham, M. B.; Jefferson, M. M. Myeloperoxidase-dependent effect of amines on functions of isolated neutrophils. J. Clin. Invest. 72:441-454; 1983. 11. Davis, W. B.; Husney, R. M.; Wewers, M. D.; Herzyk, D. J.; Sagone, A. L., Jr. Effect of O2 partial pressure on the myeloperoxidase pathway of neutrophils. J. Appl. Physiol. 65:19952003; 1988. 12. Sagone, A. L., Jr.; Husney, R. M.; Wewers, M. D.; Herzyk, D. J.; Davis, W. B. Effect of dimethylthiourea on the neutrophil myeloperoxidase pathway. J. Appl. Physiol. 67:10561062; 1989. 13. Chance, B. The kinetics and stoichiometry of the transition from the primary to the secondary peroxidase peroxide complexes. Arch. Biochem. Biophys. 41:416-424; 1952. 14. Mandell, G. L.; Sande, M. A. Antimicrobial agents. In: Gilman, A. G.; Goodman, L. S.; Rail, T. W.; Murad, F., eds.
Biotransformation of PABA by PMN
15.
16.
17. 18.
19. 20.
21.
Goodman and Gilman's the pharmacological basis of therapeutics. New York: Macmillan Publishing Co.; 1985:1095-1114. Wasil, M.; HalliweU, B.; Grootveld, M.; Moorhouse, C. P.; Hutchison, D. S. C.; Baum, H. The specificityof thiourea, dimethylthiourea and dimethyl sulphoxide as scavengers of hydroxyl radicals. Biochem. J. 243:867-870; 1987. Dull, B. J.; Salata, K.; Van Langenhove, A.; Goldman, P. 5aminosalicylate: Oxidation by activated leukocytes and protection of cultured cells from oxidative damage. Biochem. Pharmacol. 36:2467-2472; 1987. She, Z.-W.; Davis, W. B. Use of para-aminosalicylic acid as a trapping agent for hypochlorous acid. Am. Rev. Respir. Dis. 143:742; 1991 (abstract). Shacter, E.; Lopez, R. L.; Sangeeta, P. Inhibitionof the myeloperoxidase-H202-C1- system of neutrophils by indomethacin and other non-steroidal anti-flammatory drugs. Biochem. Pharmacol. 41:975-984; 1991. Kettle, A. J.; Winterbourn, C. C. Mechanism of inhibition of myeloperoxidase by anti-flammatory drugs. Biochem. Pharmacol. 41:1485-1492; 1991. Maskos, Z.; Rush, J. D.; Koppenol, W. H. The hydroxylation of the salicylateanion by a fenton reaction and r-radioanalysis: A consideration of the respective mechanisms. Free Radic. Biol. Med. 8:153-162; 1990. Kaur, H.; Fagerheim, 1.; Grootveld, M.; Puppo, A.; Halliwell, B. Aromatic hydroxylation of phenylalanine as an assay for hydroxyl radicals: Application to activated human neutrophils and to the heme protein leghemoglobin. Anal, Biochem. 172:360-367; 1988.
35
22. Winterbourn, C. C. Myeloperoxidase as an effective inhibitor of hydroxyl radical production. J. Clin. Invest. 78:545-550; 1986. 23. Ingelman-Sundberg, M.; Kaur, H.; Terelius, Y.; Persson, J.-O.; Halliwell, B. Hydroxylation of salicylate by microsomal fractions and cytochrome P-450. Biochem. J. 276:753-757; 1991. 24. Babior, B. M. Oxygen-dependent microbial killing by phagocytes. N. Eng. J. Med. 298:659-668; 1978. 25. Winston, G. W.; Cederbaum, A. I. Oxidative decarboxylation of benzoate to carbon dioxide by rat liver microsomes: A probe for oxygen radical production during microsomal electron transfer. Biochemistry 21:4265-4270; 1982. 26. Winston, G. W.; Harvey, W.; Bed, L.; Cederbaum, A. I. The generation of hydroxyl and alkoxyl radicals from the interaction of ferrous bipyridyl with peroxides. Biochem. J. 216:415421; 1983. 27. Halliwell, B.; Gutteridge, J. M., eds. Free radicals in biology and medicine. Oxford: Clarendon Press; 1989:86-187. 28. Galaris, D.; Mira, D.; Sevanian, A.; Cadenas, E.; Hochstein, P. Co-oxidation of salicylateand cholesterol during the oxidation of metmyoglobin by H202. A rch. Biochem. Biophys. 262:221231; 1988. 29. Davies, M. J. Detection ofmyoglobin-derived radicals on reaction of metmyoglobin with hydrogen peroxide and other peroxidic compounds. Free Rad. Res. Comm. 10:361-370; 1990. 30. Puppo, A.; Halliwell,B. Formation of hydroxyl radicals in biological systems. Does myoglobin stimulate hydroxyl radical formation from hydrogen peroxide? Free Rad. Res. Comm. 4:415-422; 1988.
0891-5849/93 $6.00 + .00 Copyright © 1993PergamonPress Ltd.
Original Contribution BIOTRANSFORMATION OF PARA-AMINOBENZOIC AND SALICYLIC ACID BY PMN
ACID
ARTHUR L. SAGONE,JR., ROSE MARIE HUSNEY, and W . BRUCE DAVIS Division of Hematology and Oncology, and Pulmonary and Critical Care, Department of Internal Medicine, The Ohio State University, Columbus, OH 43210, USA (Received 4 March 1992; Revised 9 July 1992; Accepted 30 July 1992) Abstract---Para-aminobenzoic acid (PABA) is an essential cofactor for the production of folic acid in bacteria and has mild anti-inflammatory activity. We have recently reported that salicylic acid and benzoic acid are oxidized by stimulated granulocytes Polymorphonuclear Neutrophils (PMN). The oxidation of salicylate appears mediated by a potent oxygen metabolite generated during the respiratory burst which is dependent primarily on superoxide (02) for its production. These background studies with the salicylate group of drugs suggested that PABA might be similarly metabolized by PMN. In these studies, we demonstrate that PABA is metabolized by stimulated PMN. However, in contrast to the biochemical mechanism involved in the metabolism of salicylate, our scavenger studies indicate that PABA is metabolized primarily by the myeloperoxidase pathway. Our results may explain the mild anti-inflammatory actions of the drug and suggest that the degradation of PABA by PMN at an inflammatory site may limit the availability of PABA for bacterial growth. Keywords--Para-aminobenzoic acid, Salicylic acid, Hydroxyl radical, Myeloperoxidase, Biotransformation of xenobiotics, Anti-inflammatory drugs, Reactive oxygen species, Activated granulocytes, Free radicals
rine. 9-12 The possible products include a chloramine, a hydroxylamine, or chlorination of the benzene ring. Second, Chance demonstrated almost 30 years ago that PABA is oxidized directly by a complex of H202 and horseradish peroxidase (Compound I).13 This observation also raised the possibility that a similar mechanism for the oxidation of PABA might occur in stimulated PMN. The metabolism of PABA by stimulated granulocytes could destroy its availability for normal folic acid metabolism by bacteria. ~4 If so, the degradation of PABA by granulocytes might represent an unrecognized microbicidal mechanism for granulocytes. In these studies we have demonstrated that PABA, similar to salicylic acid, is decarboxylated by stimulated granulocytes. However, in contrast to the mechanism involved in the metabolism of the salicylate, our scavenger studies indicate that PABA is metabolized primarily by the myeloperoxidase pathway.
INTRODUCTION
Para-aminobenzoic acid (PABA) is an essential cofactor for the production of folic acid in microorganisms. This compound is also known to have mild anti-inflammatory activity and has been reported to decrease the fibrosis occurring in scleroderma and Peyronie's disorder) '2 We have recently reported that benzoic acid and salicylic acid undergo decarboxylation and hydroxylation by stimulated granulocytes. 3-6 Our data suggest that the reaction is mediated by a potent radical which is dependent primarily on superoxide for its production. These background studies with the salicylate drugs suggested that PABA might be similarly metabolized by granulocytes. However, we also felt that PABA might be metabolized by the myeloperoxidase pathway. First, PABA has an amine group, and we postulated that this compound might be oxidized by the hypochlorous acid (HOCI) produced by the myeloperoxidase pathway in a manner similar to the metabolism of d a p s o n e 7,s o r tau-
MATERIALS AND METHODS
Address correspondence to: Dr. Arthur L. Sagone, Jr., The Ohio State University, N1011 Doan Hall, 410 West 10th Avenue, Columbus, OH 43210.
p-amino[carboxyl-14C]benzoic acid (specific activity = 56.6 or 54.5) was obtained from Amersham, 27
28
A.L. SAGONE,JR. el aL
Arlington Heights, IL, and [carboxyl-~4C]salicylic acid (specific activity 56.6) from ICN Radiochemicals, Irvine, CA. Methionine, sodium azide, sodium benzoate, glutathione (GSH), dimethyl sulfoxide (DMSO), taurine, catalase (14,100 U/mg protein), PABA (free acid, purity 99+%), para-aminosalicylic acid, sodium salt (PAS), superoxide dismutase (SOD) #S-8254 (2725 U/mg protein) and #S-2515 (3250 U/ mg protein), phorbol myristate acetate (PMA), and zymosan (lot #71F-8086) were obtained from Sigma Chemical Co., St. Louis, MO. Dimethylthiourea (DMTU), sodium hypochlorite (NaOC1), and sodium salicylate were from Aldrich Chemical Co., Milwaukee, WI. Mannitol was obtained from Abbott Laboratories, North Chicago, IL, and phenol (molecular biology grade) from Fisher Scientific, Fair Lawn, NJ. Dulbecco's phosphate buffered saline (DPBS) was from GIBCO, Grand Island, NY.
Cell preparation Fresh blood was obtained for each experiment from healthy volunteers who had taken no medications for at least 72 h. Purified neutrophil suspensions were isolated by dextran sedimentation and Ficoll-hypaque density gradient centrifugation, as previously described. 3-6
Metabolic studies We determined the rate of decarboxylation of [Carboxyl-~4C]salicylic acid, ([14C]salicylate), and of p-amino[carboxyl-14C]benzoic acid ([14C]PABA) by suspensions of unstimulated and stimulated neutrophils. As in previous studies, 3'5 an ionization chamber electrometer apparatus was used to continuously monitor the production of ~4COz from this substrate. Purified neutrophils (5 × 106/ml) were resuspended in Dulbecco's phosphate-buffered saline containing 50 mg/dl of glucose, and 4-ml aliquots of the suspension were placed into 25-ml triple-arm distilling flasks containing either 1 uCi of [~4C]salicylic acid or 2 #Ci of [~4C]PABA (in 0.1 ml or 0.2 ml normal saline, respectively). The flasks with salicylate also contained unlabeled substrate. The final cell concentration was 4 x 106]ml. The final substrate concentration for the salicylate was 104 #M and for the PABA, 8 #M. After incubating the suspensions at 37 ° for 45 to 50 min, DPBS alone or an equal volume of various free radical inhibitors (in DPBS) were added, and 10 to 15 min later, PMA (in DPBS-0.10% DMSO) was added to stimulate the neutrophils. The final concentration of PMA was 50 ng/ml, while the final concentration of the original diluent, DMSO, was 0.37 raM. In experi-
ments where opsonized zymosan was used to stimulate the cells, the particles were added in a saline suspension such that the final concentration was 1 mg/ ml. The amount of ~4CO2 thus generated from the labeled compounds was calculated from the peak millivolt value achieved and was expressed as nanomoles 14CO2 produced per hour per 107 cells.
Experiments with cell free systems We have previously published evidence that salicylate reacts with HOC1 in a cell-free system. 6 In these experiments, we determined the capacity of solutions of hypochlorite to decarboxylate [~4C]salicylate. Initial studies established that HOCI decarboxylates [~"C]salicylate (0.1 mM) and that the rate of the reaction increased as the concentration of HOC1 was increased from 0.005 mM to 0.5 mM. Following this, additional experiments were done to determine the relative effects of scavengers on the reaction. In these flasks, similar quantities of radioactive and unlabeled salicylate were added as in the previously described experiments, along with 4 ml of DPBS. After incubating several minutes at 37°C with stirring to establish zero baseline, scavengers were added; then 0.1 cc of NaOC1 was added. The final concentration of HOCI used for these experiments was 0.05 mM.
Neutrophil stable oxidants and hypochlorous acid (HOC1) production Stimulated granulocytes release a stable oxidant into the supernatant, which has been identified primarily as taurine chloramine produced by the reaction of HOCI with cellular taurineJ ° These experiments determined the effect of salicylate or PABA on the production of stable oxidants and HOC1 by stimulated neutrophilsJ ~'12 The cell incubations were prepared in three-neck flasks in a fashion similar to that of the metabolic studies, except that radioactive substrates were not added to the incubations. For the experiments estimating total HOC1 production, taurine (10 or 15 mM final concentration) was also added to the flasks to trap the HOCI as taurine chloramine. The cells were preincubated for 1 h at 37°C with stirring in the presence of a range of concentrations of either salicylate or PABA (or DPBS alone). This was done on the ion chamber apparatus at the same airflow as for the metabolic studies. After the preincubation, either PMA (50 ng/ml final concentration) or zymosan (1 mg/ml final concentration) was added and incubated for 1 h. The reactions were terminated by placing the flasks into an ice bath and by immediately adding 25 ug/ml of catalase. Cell-free superna-
Biotransformation of PABA by PMN
29
1412 10 (n l.J
0 >
8
u
..I ..I
6 4
PMAAde~d d
2 0 0
20
40
60
80
100
120
140
160
180
MINUTES Fig. 1. The metabolism of para-aminobenzoic acid by granulocyte suspensions. The results indicate a typical experiment which is representative of six done. The curve was derived by computer analysis using the indicated data points (rq) from the continuous recorder tracing of the electrical output of the electrometer. The y axis indicates the millivolt signal generated by ~4CO2in the ionization chamber. The x axis indicates the time in minutes. The addition of PMA resulted in a prompt increase in the electrical signal.
tants were obtained by centrifugation of the suspensions at 27,700 × g. The capacity of aliquots to oxidize I- to I2 was determined as previously described.~l'~2 The results were expressed as nanomoles of I2 produced per 10 7 cells, ll'12
Statistics Data were analyzed according to the test for independent samples. Results were expressed as mean ___SD. RESULTS
Unstimulated suspensions of granulocytes did not oxidize significant amounts of PABA. The addition of PMA resulted in a prompt production of CO/, which was maximal at approximately 30 min (see Fig. 1). Similar results were found when zymosan was used as a stimulus (data not shown). These observations indicate that both zymosan and PMA-stimulated granulocytes decarboxylate PABA. The maximum rate of CO2 production from PMN suspensions stimulated with PMA was 0.78 +__0.36 (SD) nanomoles per 10 7 cells per hour, approximately two- to threefold the rate of cells stimulated with zymosan, which was 0.29 ___0.10 (SD) nanomoles per 10 7 cells per hour.
Effect of scavengers on the decarboxylation of [14C]PABA by either zymosan or PMA The addition of superoxide dismutase to the suspensions prior to incubation with PMA enhanced PABA oxidation twofold (see Table 1). Catalase, a scavenger ofHzO2, and two HOCI scavengers, methionine and taurine, were relatively potent inhibitors of the reaction, while DMSO had no effect. Azide also inhibited the reaction. The results of scavenger studies done with zymosan-stimulated cells were less clear. The addition of superoxide dismutase, catalase, and two scavengers of HOC1 did not appreciably affect the rate of decarboxylation of PABA by zymosan-stimulated granulocytes (Table 1). Sodium azide (10 -4 M) did produce an approximately 40% suppression of the reaction. Similar to the results with PMA, DMSO did not impair the reaction. However, DMTU, a nonspecific free radical scavenger,12 did impair the decarboxylation 89%.
Effect of scavengers on the decarboxylation of [t4C]salicylate by PMA-stimulated PMN We have previously reported the effects of scavengers on the oxidation of salicylate by zymosan and PMA on stimulated PMN. 5,6 However, in our prior
30
A . L . SAGONE, JR. et al. Table 1. Effect of Scavengers on the Rate o f 14CO2Production from a 0.008 m M Solution of [~4C]PABA by Stimulated Granulocytes
Scavenger Azide(0.1 mM) SOD (10 #g/ml) Catalase (100 ~g/ml) Methionine (1 mM) Taurine (5 mM) DMSO (30 mM) D M T U (10 mM)
PMA
Percent o f ControP P Value
33.6_+ 11.2(3) 209.3_+76.8(3) 26.3 _+ 13.4 (3) 32.7 _+ 3.1 (3) 52.7 _+ 5.7 (3) 115.7 (1) --
<0.02 <0.01 < 0.01 < 0.01 < 0.02
Zymosan
P Value
61.7_+ 8.7(3) 95.0_+ 4.1 (3) 90.7 _+ 28.8 (4) 121.6 _+ 19.5 (3) 103.1 _+ 18.2 (3) 91.7 (1) 11.4 (1)
<.1 NS NS NS NS
a The results were derived from paired experiments. The number of experiments done is indicated in parentheses. The mean maximal rate of 14CO2 production for PMA-stimulated cells was 0.78 _+ 0.36 nanomoles per l07 cells per hour (six experiments). The mean maximal rate of ~4CO: production for zymosan-stimulated cells was 0.29 _+ 0. l0 nanomoles per 10 7 cells, per hour (seven experiments). See Fig. 1.
studies, we did not evaluate extensively the effect of scavengers on the decarboxylation of salicylate induced by PMA stimulated PMN. The unexpected effect of a number of scavengers on the decarboxylation of PABA by PMA-stimulated PMN prompted us to do similar scavenger studies with salicylate as a substrate. The results are given in Table 2. Similar to our prior results, SOD inhibited salicylate decarboxylation. This effect is opposite to that observed in our experiments with PABA (see Table 1). Catalase also inhibited the reaction. However, three scavengers of HOCI (i.e., GSH, taurine, and methionine) had no effect. Mannitol, a scavenger of "OH, was similarly ineffective. Phenol, in an equimolar concentration to salicylate, decreased the reaction 72%. DTMU, which is a known scavenger of "OH and HOC1, was ineffec-
tive at a 0.1 mM concentration, and at a 1 mM concentration produced only 18% inhibition (p > .3). A 100-fold higher concentration of DMTU (10 mM) than salicylate was required to produce significant inhibition (62%) (p < .01). Both DMSO and benzoate, when used in high concentrations, inhibited the reaction approximately 50%. PABA in an equimolar concentration produced 25% inhibition (p > .3), while a 1-mM concentration decreased the rate of oxidation 66%. Overall, the results of these experiments with PMA are similar to those which we have previously reported for zymosan-stimulated granulocytes. However, there is one difference. In these experiments, we did demonstrate some specific inhibition of the reaction by catalase. In contrast, we did not observe a major inhibitory effect of catalase in our studies of
Table 2. Effect of Scavengers on the Rate of Decarboxylation o f a 0.1 m M Solution o f Salicylate by PMA-Stimulated Granulocytes Scavenger Catalase Heat-inactivated catalase SOD GSH Taurine Methionine Mannitol Phenol Phenol DMTU DMTU DMTU DMSO DMSO Benzoate PABA PABA
Concentration 25 #g/ml 25 t~g/ml 10 #g/ml 1 mM 5 mM 1 mM 40 m M 0.01 m M 0.1 m M 0.1 m M 1 mM 10 m M 20 m M 100 m M 20 m M 0.1 m M 1.0 m M
Percent of ControP 40.1 _+ 5.6 82.7 -+ 5.6 44.7 _+ 12.5 100.3 _+ 30.2 116.8 _+ 33.1 102.0 _+ 30.0 114.0 _+ 19.3 37.8 (1) 28.3 (1) 98.2 _+ 16.4 82.6 _+ 26.6 38.0 -+ 10.8 53.9 (2) 48.0 _+ 3.9 40.0 ( 1 ) 74.1 _+ 13.7 33.7 _+ 9.9
(5) (3) (4) (4) (3) (4) (3)
P Value
(3)
< .01 <. 1 < .02 NS NS NS NS --NS > .3 < .01 -< .05
(3) (6)
< .3 < .05
(3) (4) (3)
--
a The mean maximum rate of salicylate oxidation for the controls was 4.8 _+ 2.2 nanomoles per hour per 10 7 cells (27 experiments). The results were derived from paired experiments. The number of experiments done is indicated in parentheses.
Biotransformation of PABA by P M N Table 3. Effect of Scavengers on the Rate of CO2 Production from Salicylate (0. l raM) by HOCI (0.05 mM) Concentration of Additive
Percent of ControP
P Value
Mannitol (40 mM) Benzoate (1 mM) (20 mM) DMSO (0.37 mM) b (10 mM) (20 raM) (100 mM) Phenol (0.01 mM) (0.1 mM) PABA (0.1 raM) PAS (0.1 mM) GSH (1 mM) Methionine (1 raM) Taurine (5 mM) SOD (10 #g/ml) Heat-inactivated SOD (10 #g/ml) Catalase (25 ~tg/ml) Heat-inactivated catalase (25 ~g/ml)
98.5 + 9.3 (3) 94.3 (1) 89.7 _+ 20.6 (3) 28.3 (2) 12.2 (2) 8.8 (2) 2.1 (1) 59.0 (2) 39.9 (2) 12.0 (2) 27.1 ( 1) 0 (3) 0.2 _+ 0.4 (3) 3.5 + 0.3 (3) 84.7 _+ 17.8 (4)
NS -NS --------< .001 < .01 < .02 > .1
80.1 _+ 6.8 (3) 50.6 + 3.4 (4)
< .01 < .05
59.5 + 8.2 (3)
< .05
a The mean maximal rate of decarboxylation o f a 0.1 m M solution of salicylate by a 0.05 m M concentration of HOCI was 21.1 + 2.4 nanomoles/h for nine experiments. The number of experiments done with the scavenger is indicated in the parentheses. b A final concentration of 0.37 m M DMSO was present in the cell experiments in which PMA was used as a stimulus, since this compound was used to dissolve the PMA.
zymosan-stimulated granulocytes or in our experiments involving the hydroxylation of salicylate by PMA-stimulated granulocytes. 5'6 The inhibition ofsalicylate oxidation by SOD confirms the importance of O~ in the biotransformation of this drug. The inhibition of salicylate oxidation by catalase and SOD supports the hypothesis that the decarboxylation ofsalicylate may be mediated in part by "OH produced by a Haber-Weiss type (superoxide-driven Fenton) reaction. However, in view of the problematic nature of scavenger studies in a system as complex as the cell, our results do not exclude the possibility that more than one mechanism may be involved in the decarboxylation of salicylate by PMA-stimulated PMN (i.e., one dependent on 0 2 and a second dependent on H202) or the possibility that the oxidizing species dependent on 02- for its production may not be "OH but rather another ROS of similar reactivity.
Effect of scavengers on salicylate decarboxylation by HOC1 The addition of HOC1 (final concentration of 0.05 mM) to a 0.1-mM solution of salicylate stimulated decarboxylation. As indicated in Table 3, mannitol and benzoate did not impair the oxidation of salicylate by HOCI. DMSO inhibited the reaction at rela-
31
tively low concentrations. PABA (0.1 mM) inhibited the reaction 88%. Para-aminosalicylic acid (PAS), a structurally similar compound, inhibited the reaction 73%. Phenol also inhibited the reaction but was somewhat less potent than PABA or PAS. Catalase and SOD inhibited the reaction 15.3% and 49.4%, respectively. However, the effects were nonspecific, since similar results were observed with the heat-inactivated enzymes. The magnitude of nonspecific inhibition by catalase was somewhat unexpected. Three compounds known to react rapidly with HOC1, namely methionine, GSH, and taurine, inhibited the reaction completely. These experiments confirm our prior observation, which indicates that HOCI can metabolize salicylate. 6 However, the results of the scavenger studies in this cell-free system suggest that a reaction of HOCI with salicylate is not a major mechanism for the decarboxylation of salicylate by PMA stimulated PMN. First, GSH, methionine, and taurine, in concentrations known to impair HOC1 production by PMA-stimulated PMN and which ablated salicylate decarboxylation in the cell-free system, did not inhibit salicylate oxidation by PMA-stimulated PMN (comparison of Tables 2 and 3). Second, the inhibition of the reaction by SOD in the cell system is also hard to resolve with a primary role of HOC1 in the decarboxylation ofsalicylate, since the production of HOC1 is dependent primarily on the oxidation of C1- by H202 and myeloperoxidase (MPO) (Compound I). Finally, a low concentration of DMSO (0.37 mM) inhibited salicylate oxidation by HOCI approximately 70%. This observation suggests that the rate constants for the reaction of DMSO and salicylate with HOCI are similar. Wasil et al. have previously suggested that while DMSO can react with HOCI in a cell-free system, the reaction may be too slow for a major interaction of DMSO with HOC1 in a biological system) 5 Our observations suggest that this may also be true for salicylate, unless high concentrations of this drug are present. Further, as discussed under Materials and Methods, DMSO was present in a final concentration of 0.37 mM in our cell experiments with PMA since this compound was required to dissolve the PMA. This concentration of DMSO appears sufficient to prevent most of the potential decarboxylation of salicylate by the HOCI produced by the PMN under our experimental conditions.
Effect of PABA and salicylate on the recovery of stable oxidants The effect of PABA on the production of stable oxidants by stimulated PMN is given in Table 4. As
32
A. L. SAGONE, JR. el aL Table 4. Effect of PABA on the Production of Stable Oxidants by Stimulated Cellsa
Source of Supernatant Stim. cells alone Stim. cells + 0.1 mM PABA Stim. cells + 0.5 mM PABA Stim. cells + 1.0 mM PABAb
PMA Stimulated
Zymosan Stimulated
34.1 + 9.7 (4) 31.2+10.1(4) 26.9 + 5.6 (3) 18.3 _+ 4.2 (4)
24.8 + 6.6 (3) 19.4_+7.0(3) 15.2 + 6.3 (3) 10.7 _+ 4.9 (3)
Results indicate the nanomoles ___SD of 12 produced by l0 7 cells in 1 h. Since 1 mol oftaurine chloramine oxidizes 2 mols of I-, the amount of 12 produced equals the amount of taurine chloramine generated by the cells. The number of experiments done with each concentration is indicated in parentheses. b p value for both zymosan and PMA < .05.
indicated, a 0.1-mM concentration of PABA did not alter the recovery of stable oxidants. A 1-mM concentration, however, was associated with a decreased recovery from both zymosan and PMA-stimulated PMN suspensions. This observation also appears to exclude the conversion of PABA to a stable chloramine by stimulated PMN. The observation that salicylate can be metabolized by HOCI in a cell-free system prompted us to do additional studies concerning the effects of salicylate on the activity of the MPO pathway. Therefore, we quantitated the effects of salicylate on the production of HOCI as determined by the taurine assay (Table 5). While a 0. l-mM concentration of salicylate did not definitely impair the production of taurine chloramine by zymosan or PMA-stimulated PMN, higher concentrations did have a significant effect. A 10-mM concentration inhibited the production of taurine chloramine almost completely. A 1-mM concentration inhibited the reaction 70% when PMA was used as a stimulant and 50% when zymosan was the stimulus. Salicylate had similar effects on the production of stable oxidants (see Table 5). In contrast, there was no effect of salicylate (0.1 to 10 mM) on the recovery of taurine chloramine when added directly to supernatants containing taurine chloramine (i.e., supernatants recovered from granulocyte suspensions which had been incubated with PMA for 1 h). This observation indicates that salicylate does not react directly with taurine chloramine or interfere with the I- assay which is used for its quantitation. DISCUSSION
These data indicate that zymosan- and phorbolstimulated granulocytes decarboxylate para-aminobenzoic acid. The stimulation of the cells by PMA was associated with a higher rate of reaction than that of cells stimulated by zymosan. The scavenger experi-
ments done with zymosan were not definitive. There was some inhibition of the reaction by azide but no effect of superoxide dismutase, catalase, methionine, or taurine. DMTU, a nonspecific ROS scavenger, did impair the reaction. This suggests that the reaction is mediated by a reactive oxygen species (ROS) produced during the respiratory burst. We felt that the failure of our scavengers to impair PABA decarboxylation might be due to their inability to penetrate to the cellular sites of PABA oxidation in concentrations adequate to inhibit the reaction. Therefore, we studied PMA-stimulated PMN since no phagolysosome develops in PMN following stimulation by this agent. Under these circumstances, a significant portion of the reaction is probably occurring at the cell membrane. The scavenger studies done with the PMA-stimulated granulocytes suggest that the main mechanism for the reaction involves myeloperoxidase. There was an inhibition with azide as well as two scavengers, methionine and taurine, which are known to be metabolized by the myeloperoxidase pathway. Similarly, catalase was able to inhibit the reaction, while the addition of superoxide dismutase actually stimulated the rate of decarboxylation. These results are consistent with the metabolism of PABA by the myeloperoxidase pathway) 2 The reasons for the difference noted in the scavenger studies done with PMA and zymosan are not entirely clear. It seems likely that the biochemical mechanism for the oxidation of PABA by zymosan-stimulated PMN is similar to that of PMA. However, we cannot exclude the possibility that PABA decarboxylation occurs by another mechaTable 5. Effect of Salicylate on the Production of Stable Oxidants and Taurine Chloramine by Stimulated PMN a Stimulant Source of Supernatant Stim. cells Stim. cells+0.1 mMsalicylate Stim. cells + 1.0 mM salicylate Stim. cells + 10 mM salicylate Stim. cells + taurine Stim. cells + taurine + 0.1 mM salicylate Stim. cells + taurine + 1.0 mM salicylate Stim. ceils + taurine + 10.0 mM salicylate
PMA 18 18 2 5 73.2
+ 5 (3) _+ 3(3) +_ 2 (3) + 1 (3) + 32.6 (4)
Zymosan 21 _+ 4 (3) 18 _+ 4(3) 9 + 11 (3) 5 _+ 1 (3) 194.0 (2)
52.8 _+ 33.8 (4)
198.0 (2)
21.5 _+ 2.5 (4)
96.5 (2)
8.0 + 3.9 (4)
8.0 (2)
a The concentration of taurine used for these experiments was 15 mM with PMA as stimulant, l0 mM with zymosan as stimulant. The results are given as nanomoles of 12 -- SD produced by l07 cells in 1 h and equal the amount of taurine chloramine formed. The number of experiments done with each concentration is indicated in parentheses. PMA experiments done with the 1 mM and l0 mM concentrations of salicylate are significantly lower than controls--p values < .05.
Biotransformation of PABAby PMN nism in zymosan-stimulated PMN even though these cells are generating substantial amounts of H O C I . 9-12 This possibility seems less likely. 9-12 Our results indicate that the mechanism for the biotransformation of PABA by stimulated PMN differs from that of benzoic and salicylic acid, since we found that the decarboxylation and hydroxylation of both of these latter substrates was inhibited by superoxide dismutase, while the addition of this enzyme enhanced PABA oxidation. Recently, Dull et al. reported that the oxidative deamination of 5-aminosalicylic acid (ASA) by activated PMN could be inhibited by the addition of superoxide dismutase to the incubation) 6 Therefore, in spite of the structural similarity of PABA to these compounds, it appears to be metabolized by a different mechanism. Recent studies in our laboratory have provided further evidence that the oxidation of PABA and its structurally similar analog, para-aminosalicylic acid (PAS), are oxidized by the MPO pathway. We have demonstrated that stimulated PMN generate products from PABA and PAS which can be detected by high-pressure liquid chromatography (HPLC). The primary product appears to involve chlorination at the 3 position of the benzene ring) 7 Similar to decarboxylation, the formation of the product in PMA cultures can be enhanced by the addition of SOD and inhibited by the addition ofcatalase and HOC1 scavengers. Additional experiments have demonstrated that an identical product can be detected in a cell-free system by the reaction of HOC1 with PABA (unpublished observations) and PAS) 7 The addition of PABA to both zymosan- and PMA-stimulated granulocytes decreased the recovery of stable oxidant activity in the supernatants of the suspensions when higher concentrations of the PABA were used. These observations indicate that PABA impairs the production of stable oxidants by PMN. Further, these findings suggest that PABA is not metabolized to a stable chloramine as is the case with amino acids such as taurine. 9-12 Similar to PABA, we demonstrated that higher concentrations of salicylate impair the production of hypochlorous acid, as measured by the taurine chloramine assay, and of stable oxidants. Similar results have been reported by Shacter et al. with murine neutrophils.t8 This finding is somewhat unexpected and suggests that the anti-inflammatory effect of salicylate may be partly due to its ability to decrease the production of HOCI by PMN as well as the production of another potent ROS. As discussed, a reaction of salicylate with HOC1 or chloramines does not appear to relate to the decarboxylation or hydroxylation of salicylate by PMN. Therefore, the mechanism by which salicylate impairs
33
HOC1 production by PMN may involve a reaction which is not associated with the oxidation of salicylate. In this regard, Kettle and Winterbourn 19 have reported that several anti-inflammatory drugs, including salicylate, decrease the production of HOC1 from H202 and MPO (Compound I) by promoting the irreversible formation of Compound II. This mechanism seems consistent with the impaired production of HOC1 by salicylate observed in our experiments but requires further testing. The capacity of salicylate to impair HOCI production again raises the issue of whether or not salicylate oxidation by PMN is dependent on the MPO pathw a y . 5'6 However, at this point the mechanism for the oxidation of salicylate by activated PMN remains problematical. As discussed in this study and our prior ones, we demonstrated that stimulated PMN decarboxylate and hydroxylate salicylate. This oxidation of salicylate by PMN is analogous to its oxidation by "OH in other systems. 5'6'2° We were unable to impair salicylate hydroxylation by PMN with deferoxamine. 6 This observation suggests that the oxidation of salicylate measured under our experimental conditions cannot be explained simply on the basis of iron contamination in our buffer resulting in the artifactual production of "OH from H202 by a Fenton type of reaction. 2t'22 However, while our results are consistent with the hypothesis that the oxidation of salicylate by stimulated PMN might be mediated by "OH, our results have not conclusively proven this point; nor do they prove that the mechanism involves the Haber-Weiss reaction as classically defined, since a number of our observations are not entirely consistent with this model. First, while we have consistently observed that SOD inhibits salicylate oxidation, indicating that 02 appears to be required for this reaction, we have been unable to prove that H202 is essential for the reaction. 5'6 Second, we have been unable to impair salicylate oxidation by mannitol, even under experimental conditions in which a substantial portion of the reaction should be occurring extracellulady (i.e., PMA stimulation of PMN). Finally, the molar ratio of 2,3-dihydroxybenzoic acid to 2,5-dihydroxybenzoic acid predicted by the random attack of "OH on salicylic acid is approximately 1:1.23 While we did detect some 2,3-dihydroxybenzoic acid in our cell studies, the amount was small, and the ratio of 2,3-dihydroxybenzoic acid to 2,5-dihydroxybenzoic acid was only 1:10, 6 a pattern of hydroxylation inconsistent with the primary oxidation of salicylate by free " O H . 23
Babior has previously suggested that alkyl hydroperoxides might be related to the oxidation of organic compounds by PMN. 24 This raises the possibility that
34
A.L. SAGONE,JR. et al.
the oxidation of salicylate is mediated by an alkoxyl radical (RO") produced from an organic peroxide by a ferrous iron complex or iron containing protein. 25,26 Such a reaction could be mediated by MPO if 02 reduces the Fe +÷+ in this enzyme to Fe +÷ following the metabolic burst. In this regard, 02 is known to form a complex with peroxidase forming an oxyperoxide (Compound III). 27 Another possible mechanism for the oxidation of salicylate by PMN is the production of ferryl iron in MPO or another iron protein during the metabolic burst. It has recently been reported that salicylate can be hydroxylated by H202 in the presence of myoglobin, most likely due to the production of ferryl iron. 28 A similar oxidation of iron may occur during the interaction of H202 with a peroxidase (Compound I). 2s-3° However, as discussed previously, we were unable to detect the hydroxylated products of salicylate in a cell-free system using the H202-MPO enzyme system. 6 Further, our inability to reliably decrease salicylate oxidation by catalase is not consistent with this mechanism. At this point, the biochemical mechanism for the oxidation of salicylate remains unclear. Further, while a single mechanism may be involved, salicylate oxidation may be mediated by more than one mechanism, and by more than one ROS, including "OH produced by a Haber-Weiss type of reaction or another mechanism. In summary, our data supports the following conclusions. First, PABA is metabolized by activated granulocytes by the MPO pathway. This may explain in part the reported anti-inflammatory properties of this drug in scleroderma and Peyronie's disorderJ ,2 The possibility that rapid degradation of PABA by activated PMN at an inflammatory site may be sufficient to limit the production of folic acid by microorganisms in vivo needs additional study since it may represent an unrecognized microbicidal mechanism. Second, our studies confirm that the decarboxylation of salicylate is dependent on 02 and cannot be inhibited by HOC1 scavengers. Therefore, in spite of the structural similarities between PABA and salicylate, these drugs appear to be metabolized differently by stimulated granulocytes. These differences appear to reflect a number of variables such as the different rate constants of the drug for reaction with the ROS produced by phagocytes, and possibly their cellular distribution. Finally, our studies indicate that concentrations of salicylate which occur in vivo impair the production of HOCI and chloramines by PMN. Therefore, the
anti-inflammatory effect of this drug may also be related to its capacity to impair the MPO activity of PMN as well as scavenge other ROS produced during the metabolic burst. The mechanism of the decreased HOC1 production by salicylate does not appear to involve a direct reaction ofsalicylate with hypochlorous acid or chloramine and is consistent with a direct inhibition of MPO activity, as suggested recently by Kettle and Winterbourn. 19 Acknowledgement - - Supported by grants from The Ohio Cancer Research Associates (OCRA), the NCI RO1-CA-32321, and The American Lung Association.We would like to thank Barbara Baker for her help in typing this manuscript.
REFERENCES 1. Zarafonetis, C. J. D. The treatment of scleroderma: Results of potassium para-aminobenzoate therapy in 104 cases. In: Mills, C.; Moyer, J. H., eds. Inflammation and diseases of connective tissue. Philadelphia: W. B. Saunders Co.; 1961:688-696. 2. Zarafonetis, C. J. D. Antifibrotic therapy with potaba. Am. J. Med. Sci. 248:550-561; 1964. 3. Sagone, A. L., Jr.; Decker, M. A.; Wells, R. M.; Democko, C. A new method for the detection of hydroxyl radical production by phagocytic cells. Biochem. Biophys. Acta. 628:90-97; 1980. 4. Alexander, M. S.; Husney, R. M.; Sagone, A. L., Jr. Metabolism of benzoic acid by stimulated polymorphonuclear cells. Biochem. Pharmacol. 35:3649-3651; 1986. 5. Sagone, A. U, Jr.; Husney, R. M. Oxidation of salicylates by stimulated granulocytes: Evidence that these drugs act as free radical scavengers in biological systems. J. Immunol. 138:2177-2183; 1987. 6. Davis, W. B.; Mohammed, B. S.; Mays, D. C.; She, Z.-W.; Mohammed, J. R.; Husney, R. M.; Sagone, A. L., Jr. Hydroxylation ofsalicylate by activated neutrophils. Biochem. Pharmacol. 38:4013-4019; 1989. 7. Uetrecht, J.; Zahid, N.; Shear, N. H.; Biggar, W. D. Metabolism ofdapsone to a hydroxylamine by human neutrophils and mononuclear cells. J. Pharmacol. and Exp. Therap. 245:274279; 1988. 8. Uetrecht, J. P. Drug-induced agranulocytosis and other effects mediated by peroxidases during the respiratory burst. In: Sbarra, A. J.; Strauss, R. R., eds. The respiratory burst and its physiological significance. New York: Plenum Press; 1988:233-243. 9. Weiss, S. J.; Klein, R.; Slivka, A.; Wei, M. Chlorination of taurine by human neutrophils. J. Clin. Invest. 70:598-607; 1982. 10. Thomas, E.; Grisham, M. B.; Jefferson, M. M. Myeloperoxidase-dependent effect of amines on functions of isolated neutrophils. J. Clin. Invest. 72:441-454; 1983. 11. Davis, W. B.; Husney, R. M.; Wewers, M. D.; Herzyk, D. J.; Sagone, A. L., Jr. Effect of O2 partial pressure on the myeloperoxidase pathway of neutrophils. J. Appl. Physiol. 65:19952003; 1988. 12. Sagone, A. L., Jr.; Husney, R. M.; Wewers, M. D.; Herzyk, D. J.; Davis, W. B. Effect of dimethylthiourea on the neutrophil myeloperoxidase pathway. J. Appl. Physiol. 67:10561062; 1989. 13. Chance, B. The kinetics and stoichiometry of the transition from the primary to the secondary peroxidase peroxide complexes. Arch. Biochem. Biophys. 41:416-424; 1952. 14. Mandell, G. L.; Sande, M. A. Antimicrobial agents. In: Gilman, A. G.; Goodman, L. S.; Rail, T. W.; Murad, F., eds.
Biotransformation of PABA by PMN
15.
16.
17. 18.
19. 20.
21.
Goodman and Gilman's the pharmacological basis of therapeutics. New York: Macmillan Publishing Co.; 1985:1095-1114. Wasil, M.; HalliweU, B.; Grootveld, M.; Moorhouse, C. P.; Hutchison, D. S. C.; Baum, H. The specificityof thiourea, dimethylthiourea and dimethyl sulphoxide as scavengers of hydroxyl radicals. Biochem. J. 243:867-870; 1987. Dull, B. J.; Salata, K.; Van Langenhove, A.; Goldman, P. 5aminosalicylate: Oxidation by activated leukocytes and protection of cultured cells from oxidative damage. Biochem. Pharmacol. 36:2467-2472; 1987. She, Z.-W.; Davis, W. B. Use of para-aminosalicylic acid as a trapping agent for hypochlorous acid. Am. Rev. Respir. Dis. 143:742; 1991 (abstract). Shacter, E.; Lopez, R. L.; Sangeeta, P. Inhibitionof the myeloperoxidase-H202-C1- system of neutrophils by indomethacin and other non-steroidal anti-flammatory drugs. Biochem. Pharmacol. 41:975-984; 1991. Kettle, A. J.; Winterbourn, C. C. Mechanism of inhibition of myeloperoxidase by anti-flammatory drugs. Biochem. Pharmacol. 41:1485-1492; 1991. Maskos, Z.; Rush, J. D.; Koppenol, W. H. The hydroxylation of the salicylateanion by a fenton reaction and r-radioanalysis: A consideration of the respective mechanisms. Free Radic. Biol. Med. 8:153-162; 1990. Kaur, H.; Fagerheim, 1.; Grootveld, M.; Puppo, A.; Halliwell, B. Aromatic hydroxylation of phenylalanine as an assay for hydroxyl radicals: Application to activated human neutrophils and to the heme protein leghemoglobin. Anal, Biochem. 172:360-367; 1988.
35
22. Winterbourn, C. C. Myeloperoxidase as an effective inhibitor of hydroxyl radical production. J. Clin. Invest. 78:545-550; 1986. 23. Ingelman-Sundberg, M.; Kaur, H.; Terelius, Y.; Persson, J.-O.; Halliwell, B. Hydroxylation of salicylate by microsomal fractions and cytochrome P-450. Biochem. J. 276:753-757; 1991. 24. Babior, B. M. Oxygen-dependent microbial killing by phagocytes. N. Eng. J. Med. 298:659-668; 1978. 25. Winston, G. W.; Cederbaum, A. I. Oxidative decarboxylation of benzoate to carbon dioxide by rat liver microsomes: A probe for oxygen radical production during microsomal electron transfer. Biochemistry 21:4265-4270; 1982. 26. Winston, G. W.; Harvey, W.; Bed, L.; Cederbaum, A. I. The generation of hydroxyl and alkoxyl radicals from the interaction of ferrous bipyridyl with peroxides. Biochem. J. 216:415421; 1983. 27. Halliwell, B.; Gutteridge, J. M., eds. Free radicals in biology and medicine. Oxford: Clarendon Press; 1989:86-187. 28. Galaris, D.; Mira, D.; Sevanian, A.; Cadenas, E.; Hochstein, P. Co-oxidation of salicylateand cholesterol during the oxidation of metmyoglobin by H202. A rch. Biochem. Biophys. 262:221231; 1988. 29. Davies, M. J. Detection ofmyoglobin-derived radicals on reaction of metmyoglobin with hydrogen peroxide and other peroxidic compounds. Free Rad. Res. Comm. 10:361-370; 1990. 30. Puppo, A.; Halliwell,B. Formation of hydroxyl radicals in biological systems. Does myoglobin stimulate hydroxyl radical formation from hydrogen peroxide? Free Rad. Res. Comm. 4:415-422; 1988.