Relationship between arachidonic acid metabolism

0065-4299/88/020115-09$ 1.50+0.20/0 © 1988 Birkhiiuser Verlag, Basel

Agents and Actions, vol. 25, 112 (1988)

Relationship between arachidonic acid metabolism, myeloperoxidase activity and leukocyte infiltration in a rat model of inflammatory bowel disease N. K. Houghton-Smith 1 , J. L. Wallace 2 and B. J. R. Whittle 1 1

Department of Mediator Pharmacology, Wellcome Research Laboratories, Beckenham, Kent BR3 3BS, U.K. ' Dr. J. L. Wallace, Department of Physiology, Botterell Hall, Queen's University, Kingston, Ontario, K7L 3N6, Canada

Abstract The relationship between 14 C-arachidonic acid (1 4 C-AA) metabolism, myeloperoxidase activity (MPO) and leukocyte infiltration was studied in a chronic model of inflammatory bowel disease, induced by a single intrarectal application of the hapten, trinitrobenzene sulphonic acid (lNB). The colonic damage produced by lNB was accompanied, after 12-36 hours, by a marked increase in MPO, which was directly correlated to leukocyte infiltration, assessed histologically. There was also a marked increase in the metabolism of 14 C-AA, by homogenates of inflamed colon, to 12-, 15-HETE and 6-keto-PGF1 a as indices of Iipoxygenase and cyclo-oxygenase metabolism respectively. However, a further increase in MPO-cellular infiltration, between 36-72 hours after lNB, was accompanied by a reduction in 12- and 15-HETE formation. The increase in MPO-cellular infiltration was maintained for up to 3 weeks, at which time both 12-, 15-HETE and 6-keto-PGF 1 a formation had returned to control levels. These results suggest that these AA metabolites have a greater importance in the acute phase of the inflammatory response induced by lNB compared to the later chronic phase.

Introduction A model of inflammatory bowel disease (IBD) can be induced in rats by a single instillation of the hapten, trinitrobenzene sulphonic acid (lNB), into the rat colon [1]. The resulting chronic ulceration and inflammation produced by 1NB persists for up to five weeks. Using this model we have previously shown that the gross macroscopic ulceration and inflammation of the rat colon, present three weeks after 1NB, was accompanied by an increase in the colonic synthesis of 6-keto prostaglandin (PG) F 1 a and leukotriene (L1) 8 4 [2]. In addition, although anti-inflammatory drugs substantially reduced the formation of eicosanoids, including LTB 4 by the inflamed colon, they had no significant effects on the macroscopic

ulceration and inflammation induced by 1NB. This dissociation between chronic ulceration and inflammation and eicosanoid formation does not, however, preclude a role for the eicosanoids in the initial development of the inflammation induced by 1NB. In clinical studies, during relapse of IBD, there is an increase in the formation of both cyclo-oxygenase and lipoxygenase metabolites of arachidonic acid [3-6]. Relapse in IBD is also characterised by mucosal and submucosal leukocyte infiltration [7]. Since leukocytes are a predominant source of arachidonate lipoxygenase metabolites [8, 9] it is important to establish the relationship between the infiltration of these inflammatory cells and the formation of arachidonic acid metabolites.

116 In

~h~

present study, the development of the comflammation induced by 1NB has been investigated. Leukocyte infiltration into the colon wa~ . assessed both histologically and as the actlVlty of the enzyme myeloperoxidase. Myeloperoxidase (MPO) is a haemoprotein which catalyses the peroxidation of a wide variety of substrates and is involved in intracellular bacterial killing [10]. The enzyme, characteristically found in the azurophil granules, has been used as a ma!ker. for ~olymorphonuclear leukocyte infiltratwn m a dtverse number of tissues including human ski~ [11], rabbit eye [12], dog heart [13] and also m an acute model of intestinal inflammation in the rat [14]. The metabolism of 14 C-arachidonic acid by the rat colon has also be~n studied and the relationship between changes m arachidonic acid metabolism and leukocyte infiltration investigated. Some of this work was presented to the British Society of Gastroenterology [15].


Myeloperoxidase activity

1om~

Methods Time course of the colonic ulceration and inflammation induced following intra-colonic TNB administration

Inflammation of the colon was induced in female Wistar rats (200-250 g) by a single intra-colonic administration of 1NB. Groups of rats were anaesthetised with ether, and a 1NB solution (2~ ~g, diss.olved in 0.25 ml of 30% ethanol) admtmstered mto the colon via the anus using a rubber catheter (8 em long, external diameter 2 mm). The rats were killed at 1 hour to 3 weeks after 1NB. The distal 6-8 em of colon was removed, opened longitudinally and washed to remove luminal contents. The excised colon was pinned out flat on a wax block flooded with 0.9% saline. Small sections of colon were taken from two distinct areas from each colon and placed in storage fixative (glutaraldehyde I% in 4% formaldeh~de) for. histology. A larger segment(> 50 mg) was tmmedtately frozen (- 20 ° C dry ice) for subsequent estimation of MPO activity. A third segment (> 200 mg) was placed on ice for incubation with .14 C-AA (as described below). When gross colome damage was apparent, the segments oftissue were taken from the areas involved.

Myeloperoxidase assay

MPO was measured using a method similar to that described by Bradley et al. [11]. Segments of colon (stored at -20°C) were homogenised (Ultratu.rrax, 30 se.c) in 0.5% hexadecyltrimethyl-ammomum bromtde (HTAB) in 50 mM potassium phosphate buffer (pH 6.0) to give a 50 mg ml-1 w /v suspension. Aliquots (I ml) of the suspension were frozen (on dry ice granules, -20°C) and thawed (immersion in warm water 37 °C) three times. Following centrifugation (10000 g, 2 min at 4 oq, aliquots of supernatant ( 100 !J.}) were mixed in a cuvette with 2.9 ml of 50 mM phosphate buffer (pH 6.0) containing 0.167 mg ml-1 of 0-dianisid!ne dihydrochloride and 0.0005% hydrogen peroxtde, and the change in absorbance at 460 nm measured (Beckman, spectrophotometer model 25). One unit of MPO activity was defined as that degrading one micromole of peroxide per minute at 25 °C [11]. Rat peritoneal neutrophils

Rats were anaesthetised with ether and 20 ml of a I% glycogen (Type II, Oyster, Sigma) solution injected intra peritoneally (21 G needle). After 4 hours, the rats were again anaesthetised with ether, and 20 ml of heparinized saline (25 units ml-1 ) injected into the peritoneum. The rat abdomen was gently manipulated to mix the peritoneal con~ents and after careful laparotomy, the fluid was wtthdrawn. The peritoneal fluid was collected (4 oq and after centrifugation (1000 rpm, 15 min, 4 °C), the pellet resuspended in heparinized saline. Further serial dilutions of cells were made in isotonic saline. In one experiment, resident peritoneal cells were collected from anaesthetised rats by injecting 20 ml ofheparinised saline. Histological assessment of inflammation

Segments of rat colon which had been kept in glut~raldehyde (1 ~in 4% formaldehyde) storage fixative (pH 7.2, 4 C) were processed for plastic (S~urr r~sin) histology and sections (1 J.UD.) stamed wtth methylene blue-azure 11-basic fuchsin. Two coded histological sections from each colon were examined using light microscopy. Cellular

117

Agents and Actions, vol. 25, 1/2 (1988)

infiltration was assessed on a randomised blind basis and depending on the extent of infiltration, each section given a score of 0 (normal), 1 (minimal), 2 (moderate) or 3 (marked) and the mean histological score calculated.

flamed tissue were expressed as the mean ± SEM of(n) experiments. Statistical significance was calculated using Students' t-test for unpaired data (two-tailed). The level of statistical significance was taken asp< 0.05.

Metabolism of1 4 C-arachidonic acid

Results

Incubation and extraction

Morphological and histological changes to the rat colon following TNB administration

The fresh colonic tissue was homogenised (Ultraturrax, 30 sec) in Tris buffer (50 mM, pH 7.4 at 4°C) to give a 100 mg ml- 1 w/v suspension. Aliquots of homogenate (2 ml) were incubated with 14 C-AA (0.5 [!Ci, 2.6 [.lg) for 30 min at 37 oc in a shaking water bath. The unmetabolised 14 C-AA and its metabolites were extracted using the method described previously [16]. After acetone (2 volumes, -20 oq extraction, the supernatant was washed with hexane (2 volumes). The upper hexane layer was discarded, and the remaining sample acidified (pH 3.5) with formic acid and extracted into chloroform. The chloroform extract was dried (50 o C) under N2 , redissolved in chloroform: methanol (2: 1) and applied to multi-lane 1LC plates (Whatman LK 50).

TLC In each experiment, duplicate TLC plates were prepared and the radiolabelled products identified and localised using two solvent systems as previously described in detail [5]. Solvent Systems A (ether : hexane : acetic acid, 60: 40: 1 v /v /v) facilitates the separation of the lipoxygenase derived mono-hydroxyeicosatetraenoic acids (HE1E's) and the di-hydroxyeicosatetraeonoic acids (including LTB 4 ) from the prostaglandins and thromboxane (which run as a single band). Solvent System B (the organic phase of ethyl acetat: trimethylpentane: water: acetic acid, 110: 50: 100: 20, v /v /v /v), separates the individual prostaglandins from thromboxane and the hydroxyeicosatetraenoic acids.

The normal rat colon had a white, pearly, semitranslucent appearance. Histologically, the epithelium was intact, and the glandular mucosa organised and filled with dark-staining mucus, as shown in Figure 1. The muscularis mucosa, the submucosa and the muscularis were clearly delineated. The colonic mucosa was stained yellow with 1NB one hour after administration. Histologically, there was almost complete necrosis of the mucosa, submucosal oedema and some cellular infiltration. By 24 hours the colonic mucosa was disrupted and bleeding was apparent. The cellular infiltration and oedema of the submucosa was increased, although some restoration of the mucosa was apparent. The gross damage to the colon had become segmented 54 hours after 1NB, the damaged area being covered in adherent faeces. Histologically, there was continued restoration of the mucosa, but the glands were disorganised. There was a reduction in submucosal oedema, but the cellular infiltration into the mucosa and submucosa had increased further. The areas of colonic damage had formed into discrete linear ulcers one week after 1NB. Bands of macroscopically normal tissue alternated with areas of ulcerated and inflamed tissue. As described previously [1, 2] the ulceration and inflammation persisted for up to 3 weeks after 1NB. The histological features of the more chronic phase of the model were of destorted glandular architecture and mucosal and sub-mucosal leukocyte infiltration. Myeloperoxidase activity in rat peritonea/leukocytes

Calculations and statistical analysis

Differences in metabolism of 14 C-arachidonic acid and in MPO activity between control and in-

The rat peritoneal leukocytes recovered following glycogen challenge were 95% polymorphonuclear and the activity of MPO was directly related to

118


CONTROL

6HR

24HR

54 HR

2WKS

Figure I Histological section rat colon and inflamed colons taken at various times after intra-colonic TNB (20 mg in 0.25 ml of 30% ethanol). Indicated on the control colon are, (a) mucosa, (b) muscularis mucosa, (c) submucosa and (d) muscularis. At 6 hours after TNB, there is marked mucosal necrosis and submucosal oedema; by 24 hours there is some restoration of the mucosa but a marked submucosal cellular infiltration and oedema. At 54 hours the submucosal oedema was reduced but the cellular infiltrate was greater. At 2 weeks after TNB. the glands remain disorganised and there was a marked cellular infiltrate (methylene blue-azure II basic fuchsin stain , x 80 mag).

the leukocyte concentration (r=0.99, p~

...

15

16-Keto PGF1a c .5! 10

I

.

r! >

c 0

"

~

5

.,. ·:•. 121

~·

0

v

15

.,s

10

~

5

>~

IHETE I

121

0 0

6h 12h

24h

54h

36h

72h

Time after TNB

Figure3 Changes in 14 C-arachidonic acid (1 4 C-AA) metabolism and myeloperoxidase activity (MPO) in the rat colon following lNB administration. Metabolites of 14 C-AA co-migrating on 1LC with 12/15- HE1E (HE1E) and 6-keto-PGF1a were formed by homogenates of rat colon and are expressed as% conversion(% of total dpm on T.L.C. plate). MPO activity is expressed as multiples of· control. The results are the mean± SEM of (n) rats per group. Controi 14 C-AA conversion is the mean (single line) and SEM (stippled area of 16 experiments and statistically sign.ifically significant changes are shown by* p< 0.05, ** p< O.DI, *** p < 0.001.

vent system B for TLC separation of the metabolites. The profile of changes were the same using either system (Figure 4). The formation of 6-keto-PGF 1 a had a different profile (Figure 3). Following an initial significant (p < 0.01) fall, there was a gradual increase in 6-keto-PGF1 a which was maintained between 36 and 72 hours after lNB. The formation of 6-keto-

PGF 1 a then returned to control levels after one week, but fell below the level of the control, 2 and 3 weeks after lNB. The metabolism of 14 C-AA by the rat colon to other products following lNB administration were inconsistent. Generally, the cyclo-oxygenase metabolites showed a similar pattern to 6-keto-PGF 1 a but because of considerable variation, there was no consistent profile.

121

Agents and Actions, vol. 25, 112 (1988) Solvent syl18m A

In this series of experiments, the changes in MPO activity, expressed as multiples of control, were similar to those in which the cellular infiltration was also assessed histologically. The initial increase in MPO activity levelled off between 12 hours and 36 hours and after a further increase, which reached a maximum 72 hours after 1NB (17.2 ± 2.7 X control, n = 15), there was a gradual reduction in activity (Fig. 3). However, even 3 weeks after 1NB, the MPO activity was several fold greater than control (5.9 ± 2.6 X control, n=4).

20

16

Discussion

4

36

54

72

Time after TNB (hrs)

Solvent syl18m B

20

16

... ·

·; u ;,f.

:·{.:

12

.:b::

8

4

6

12

54

72

Time after TNB(hn)

Figure4 Changes in 14 C-AA metabolism to 12/15- HElE by control rat colon (filled columns) or colon taken at various times after TNB administration (stippled columns). The indentity of 12/ 15-HElE was confirmed by the separation of duplicate samples, using two solvent systems; Solvent system A (ester:hexane:acetic acid, 60:40: I, v/v/v). Solvent system B (organic phase), ethyl acetate: trimethylpentane: water: acetic acid, 110:50: 100:20), v/v/v/v). The results expressed as % conversion are the mean ±SEM of3-4 rats per group.

The structural damage to the rat colon following 1NB administration was accompanied by marked changes in cellular infiltration and metabolism of 14 C-AA. The damage to the colon was characterised by an early necrosis of the colonic mucosa which was accompanied by submucosa oedema and cellular infiltration which was predominantly polymorphonuclear but also included mononuclear cells. There was a gradual repair of the damaged mucosa, but the glandular architecture remained distorted and leukocyte infiltration was maintained, reaching a maximum three days after 1NB administration. After one to three weeks, the colonic damage resolved as discrete linear ulcerations interspersed with areas of macroscopically normal tissue. The major changes in 14 C-AA metabolism by the rat colon occurred soon after 1NB. The increased formation of HETE from 14 C-AA, used as an index oflipoxygenase activity, was accompanied by an increase in MPO activity, which was directly correlated with an increase in cellular infiltration into the rat colon. Similar increases in 14C-AA metabolism were reported in a model of acute inflammation of the rat colon, induced by injection of acetic acid [ 17). In that study, the major polymorphonuclear leukocyte cellular infiltration, observed 24 hours after acetic acid injection, was also accompanied by an increase in the metabolism of 14 C-AA to both cyclo-oxygenase and lipoxygenase products. As well as an increase in 12-HETE and 15-HETE formation by scrapings of rat colonic mucosa, an increase in the formation of 5-HETE and LTB 4 could also be detected in that study. Therefore, as in the acute phase of the present study, a large infiltration of pre-

122 dominantly polymorphonuclear leukocytes into the rat colon was related to increased 14 C-AA metabolism to lipoxygenase products. The absence of 5-HE1E and LTB 4 formation from 14 C-AA by the inflamed colon in the present study may reflect the tissue preparation used. Homogenisation of the tissue, which was necessary to obtain consistent and adequate metabolism of the 14 C-AA, may have led to the destruction of 5-lipoxygenase enzymes. Indeed, disruption of rabbit neutrophils can result in the formation of 15-Iipoxygenase products at the expense of the 5-lipoxygenase, which in whole cells are the major lipoxygenase metabolites formed [18]. In contrast to these early initial increases in 14 C-AA metabolism, the formation of HE1E was reduced to control levels 72 hours after 1NB, even though there was a further increase in MPO activity and cellular infiltration. Hence, in the presence of a considerable cellular infiltration, the levels of the major arachidonic acid products ofleukocytes were not elevated. Although in the present study the cell type composing the infiltrate have not been quantitated, the initial increase in infiltration could represent the acute increase in polymorphonuclear leukocytes which were the major cell type and, which may be the source of the HE1E formation from 14 C-AA. The subsequent increase in MPO activity in the 1NB model may represent a later mononuclear cell infiltration, more characteristic of chronic inflammation. Similar changes in cellular infiltration have been described in the carrageenin sponge model of inflammation in the rat [19]. Following subcutaneous implantation of a sterile polyester sponge, there was an initial infiltration of polymorphonuclear leukocytes which, after reaching a maximum after 2 days, levelled off or began to fall. However, the total leukocyte infiltration continued to increase, due to a later but marked infiltration of mononuclear cells, which persisted for at least a week. In the present study, resident leukocytes from the rat peritoneum, which are approximately 95% mononuclear cells, had approximately one fifth of the MPO activity of peritoneal polymorphonuclear leukocytes, illicited by glycogen. In addition in another study, in which MPO activity was used to assess acute intestinal inflammation in rats and hamsters, the peritoneal macrophages also had MPO activity of approximately one

Agents and Actions, vol. 25, I /2 ( 1988)

tenth that of peritoneal polymorphonuclear leukocytes [14]. Therefore, these studies suggest that MPO activity is not specific for polymorphonuclear leukocytes, and that mononuclear cells may also contribute to the increased MPO activity in inflamed rat colon. If the formation of HE1E by rat colon were related purely to an early infiltration by polymorphonuclear leukocytes, this would not explain why, in the presence of a further increase in leukocyte infiltration, the formation of HE1E was reduced. A possible explanation may derive from an earlier observation that polymorphonuclear leukocytes from carrageenin sponge exudates collected more than 8 hours after implantation, have a reduced capacity to synthesize LTB 4 when stimulated with ionophore, even in the presence of exogenous arachidonic acid [20]. The modulation of LTB4 synthesis may have been due to inhibitory factor(s) in the sponge exudate, since, in washed cells, response to stimulation was restored. Similar factors modulating the synthesis of HE1E may be present in the inflamed rat colon and may in some way be related to the arrival of mononuclear cells. It is interesting that in the present study the formation of 6-keto-PGF1 a was maintained at a time when HE1E formation was reduced, since prostacyclin has been shown to inhibit polymorphonuclear cell infiltration in vitro [21] and reduce leukocyte margination in the microcirculation [22] and therefore may modulate the accumulation of inflammatory cells. Alternatively, increases in MPO activity have been reported to directly modulate leukocyte function [23]. In the present study increases in the metabolism of 14 C-AA, particularly the lipoxygenase products, occured soon after 1NB administration and had returned to control levels after only a week. In contrast, however, the substantial cellular infiltration into the colon following 1NB was maintained for at least 3 weeks. These results indicate that if involved in this model of IBD, such eicosanoids maybe more important in the acute stages of the inflammation produced by 1NB. In order to determine the involvement of the eicosanoids, particularly the lipoxygenase products, studies are required using more sophisticated chromatography into the identity of the lipoxygenase products formed and into the effects of anti-inflammatory drugs on the development of this model of IBD. Of particular interest would


be the effects of specific inhibitors oflipoxygenase on the colonic inflammation and cellular infiltration produced by 1NB. The use of MPO activity as a marker of cellular infiltration will be a convenient and valuable tool in evaluating the activity of anti-inflammatory drugs. Received 26 October 1987; accepted 26 February 1988

References [I] G. P. Morris, L. Rebeiro, M. M. Herridge, M. Szewezuk and W. Depew, An animal model of chronic granulomatous inflammation of the stomach and colon. Gastroenterology 86, All88 (1984). [2] N. K. Bough ton-Smith, J. L. Wallace, G. P. Morris and B. J. R. Whittle, The effects of anti-inflammatory drugs on eicosanoid formation in a chronic model of inflammatory bowel disease in the rat. Br. J. Pharmac 94, 65-72 (1988). [3] D. W. Harris, P.R. Smith and C. H. J. Swan, Determination of prostaglandin synthetase activity in rectal biopsy material and its significance in colonic disease. Gut 19, 875-877 (1978). [4] P. Sharon, M. Ligumsky, D. Rachmilewitz and U. Zor, Role of prostaglandins in ulcerative colitis, enhanced production during active disease and inhibition by sulfasalazine. Gastroenterology 75, 638-640 (1978). [5] N. K. Boughton-Smith, C. J. Hawkey and B. J. R. Whittle, Biosynthesis of lipoxygenase and cyclo-oxygenase products from f 4 C]- arachidonic acid by human colonic mucosa. Gut 24, 1176-1182(1983). [6] P. Sharon, W. F. Stenson, Enhanced synthesis of leukotriene B. by colonic mucosa in inflammatory bowel disease. Gastroenterology 86, 453-60 (1984). [7] S. B. Kirsner and R. G. Shorter, Recent developments in non-specific inflammatory bowel disease. Part I. New Eng. J. Med. 306,755-785 (1982). (8] G. A. Higgs, R. M. J. Palmer, K. E. Eakins and S. Moncada. Arachidonic acid metabolism as a source of inflammatory mediators and its inhibition as a mechanism of action for anti-inflammatory drugs. Molec. Aspects Med. 4, 275-301 (1981). [9] B. Samuelsson, Leukotrienes mediators of immediate hypersensitivity reactions and inflammation. Science 220, 568-575 (1983). [ 10] S. J. Klebanoff and R. A. Clark, The neutrophil: Function and clinical disorders. North Holland, Amsterdam 1978. [II] P. P. Bradley, D. A. Priebat, R. D. Christensen and G. Rothstein, Measurement of cutaneous inflammation. Esti-

123 mation of neutrophil content with an enzyme marker. J. Investigative Dermatol. 78, 206-209 (1982). (12] R. N. Williams, C. A. Paterson, K. E. Eakins and P. Bhattacheijee, Quantification of ocular inflammation: evaluation of polymorphonuclear leucocyte infiltration by measuring myeloperoxidase activity. Cur. Eye Res. 2, 465-470 (1983). [13] K. M. Mullane, R. Kraemer and B. Smith, MyeloperoxidLzse activity as a quantitative assessment of neutrophil infiltration in ischaemic myocardium. J. Pharmacological Methods 14, 157-167 (1985). [14] J. E. Krawisz, P. Sharon and W. F. Stenson, Quantitative assay for acute intestinal inflammation based on myeloperoxidase activity. Gastroenterology 87, 1344-1350 (1984). (15] N. K. Boughton-Smith, J. L. Wallace and B. J. R. Whittle, Arachidonic acid metabolism and leukocyte infiltration, as measured by myeloperoxidase activity, in a model of IBD. Gut 26, A1148 (1985). (16] J. A. Salmon, P. M. Simmons and R. M. J. Palmer, A radioimmunoassay for leukotriene B4 • Prostaglandins 24, 225-235 (1982). ( 17] P. Sharon and W. F. Stenson, Metabolism of arachidonic acid in acetic acid colitis in rats. Similarity to human inflammatory bowel disease. Gastroenterology 88, 55-63 (1985). [18] S. Narumiya, J. A. Salmon, F. H. Cottee, B. C. Wheatherley, R. J. Rower, Arachidonic acid 15-lipoxygenase from rabbit peritoneal polymorphonuclear leucocytes. J. Bioi. Chern. 256, 9583-9592 (1981). [19] G. A. Higgs and K. G. Mugridge, Inhibition of mononuclear leukocyte accumulation by arachidonic acid lipoxygenase inhibitor BW755C Advances in Prostaglandin, Thromboxane and Leukotriene Research, Vol. 12. (Ed. B. Samuelsson, R. Paoletti and P. Ramwell) pp. 19-23, Raven Press, New York (1983). [20] P. M. Simmons, J. A. Salmon and S. Moncada, The release of leukotriene B 4 during experimental inflammation. Biochem. Pharmacol. 32, 1353-1359 (1983). (21] B. B. Weksler, J. M. Knapp and E. A. Jaffe, Prostacyclin (PG/2 ) synthesised by cultured endothelial cells modulates polymorphonuclear leukocyte function. Blood 50, 287 (1977). [22] G. A. Higgs, Prostaglandins and the microcirculation. In Cardiovascular pharmacology of the prostaglandins. (Eds. A. G. Herman, P.M. Vanhoutte, H. Denolin and A. Goossens) pp 315-325, Raven Press New York (1982). [23] 0. Stendohl, B. I. Coble, C. Dahlgen, J. Hed, and L. Molin, Myeloperoxidase modulates the phagocytic activity of polymorphonuclear neutrophil leukocytes. Studies with cells from a myeloperoxidase deficient patient. J. Clin. Invest. 73, 366-373 (1984).