Potency of Antileukoprotease and at-Antitrypsin to Inhibit Degradation of Fibrinogen by Adherent Polymorphonuclear Leukocytes from Normal Subjects and Patients with Chronic Granulomatous Disease Jan Stolk, Philip Davies, Johannes A. Kramps, Joop H. Dijkman, John J. Humes, Wilson B. Knight, Barbara G. Green, Richard Mumford, Robert J. Bonney, and William A. Hanlon Department of Pulmonology, University Hospital of Leiden, The Netherlands; Merck, Sharp & Dohme Research Laboratories, Rahway, New Jersey; and Bristol-Myers, Buffalo, New York

We have studied the relative efficacy of antileukoprotease (ALP) and aI-antitrypsin (a1AT) to inhibit the degradation of substrate by polymorphonuclear leukocytes (PMN) attached onto a fibrinogen matrix. PMN elastase activity was assayed by radioimmunoassay of a specific 21-residue cleavage product from the amino terminus of the Aa chain, Aa(I-21), of fibrinogen. The adherence of PMN (1.0 x 1{)6) to a fibrinogen matrix was facilitated by incubation with recombinant tumor necrosis factor-a (1 nM). Subsequently, the cells were exposed to inhibitors before stimulation with cytochalasin Band formylmethionylleucylphenylalanine. Under these conditions, ALP inhibited Aa(1-21) formation with an IC so of 85 ± 30 nM and a,AT gave an ICso of 220 ± 98 nM (mean ± SD). The effect of oxidant production on Aa(I-21) formation was evaluated by comparing the effect of PMN from normal subjects with PMN from subjects with X-linked NADPH oxidase deficiency. Stimulation of PMN from the latter subjects in a similar fashion as described above resulted in the formation of 40 ± 4 pmol/ml Aa(1-21), or approximately twice the amount seen with cells from normal subjects. Preincubation with ALP or alAT in a concentration range between 10 to 900 nM resulted in an IC so of 50 ± 13 nM for ALP compared with 150 ± 21 nM for aIAT. Both inhibitors are more effective to prevent fibrinogen degradation caused by chronic granulomatous disease (CGD) PMN than by normal PMN despite the fact that CGD PMN generated more Aa(I-21) than did normal PMN. However, myeloperoxidase activity released by the two groups of cells was comparable. ALP was more potent than alAT to inhibit fibrinogen degradation by both CGD and normal PMN in this system. Possible reasons for the increased potency of ALP over alAT include its greater accessibility to the interface between PMN and the fibrinogen matrix because of its relative low molecular mass (12 kD) and/or its high isoelectric point (pl > 9).

Polymorphonuclear leukocytes (PMN) are thought to be responsible for tissue destruction in a number of inflammatory diseases. For example, the destruction of the extracellular matrix in lung alveolar septa in patients with inherited a)-antitrypsin (a)AT) deficiency is hypothesized to be mediated by PMN-derived proteases, particularly elastase (1). The early onset of chronic obstructive pulmonary disease observed in these patients points to the importance of main-

(Received in original form February 6, 1991 and in revised form October 23, 1991) Address correspondence to: Jan Stolk, M.D., Department of Pulmonology, C3-P, University Hospital Leiden, P.O. Box 9600, 2300 RC Leiden, The Netherlands. Abbreviations: a)-antitrypsin, a)AT; antileukoprotease, ALP; chronic granulomatous disease, COD; cytochalasin B, CB; formylmethionylleucylphenylalanine, FMLP; fluorescence units, FU; myeloperoxidase, MPO; polymorphonuclear leukocyte(s), PMN; radioimmunoassay, RIA; recombinant tumor necrosis factor-a, rTNF-a; tumor necrosis factor-a, TNF-a. Am. J. Respir. Cell Mol. BioI. Vol. 6. pp. 521-526, 1992

taining sufficient levels of alAT to protect pulmonary tissue against PMN-mediated destruction. However, a quantitative deficiency of inhibitors may not be the only factor involved in the disease. Individuals with emphysema have increased amounts of PMN in their lung alveoli (2). Several products from alveolar macrophages may be responsible for this increased number of PMN. Besides being chemotactic, one of these macrophage-derived products, tumor necrosis factor-a (TNF-a), was recently found to facilitate the attachment of PMN to fibrinogen by a process sensitive to inhibition by an anti-CD 18 antibody (3). The CDllb/CDI8 receptor ofPMN binds fibrinogen, resulting in a close cpntact between the PMN cell membrane and the matrix of fibrinogen. In this study, we have evaluated the activity of PMN elastase released from PMN onto a fibrinogen matrix by assay of its specific degradation product Aa(I-21). Under these conditions, we have evaluated the capacity and potency of two major natural inhibitors of PMN elastase in the lung, namely antileukoprotease (ALP) and a,AT, to inhibit fibrin-

522

AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 6 1992

ogen degradation. Comparison is also made between the capacity of PMN from healthy individuals and subjects with an X-linked NADPH oxidase deficiency to degrade this substrate matrix.

Materials and Methods Materials Fibrinogen, cytochalasin B (CB), and formylmethionylleucylphenylalanine (FMLP) were purchased from Sigma Chemical Co. (St. Louis, MO); recombinant TNF-a (tfNF-a) was from Genzyme (Boston, MA); bovine serum albumin, essentially fatty acid free, was from Miles Laboratories (Elkhart, IN); 3,3-dimethoxy-benzidine was from Eastman Kodak (Rochester, NY), and MeoSuc-Ala-Ala-Pro-Val-MCA was from Peninsula Laboratories (Belmont, CA). o.Al' was from Athens Research and Technology (Athens, GA). ALP was isolated from human sputum as described (4), and its absolute activity calibrated as described by Kramps and co-workers (5). Purified PMN elastase isolated from human sputum was obtained from Elastin Products (St. Louis, M0). This preparation was found to be 67 % active by weight based on its capacity to hydrolyze the synthetic substrate MeoSuc-Ala-Ala-Pro-Val-p-nitroanilide as described by Green and colleagues (6). The ALP and alAT used in these studies were titrated against a standardized preparation ofPMN elastase, the activity of which 'was assayed by its activity against MeoSuc-Ala-Ala-Pro-Val-p-nitroanilide (6). Both preparations were equally as potent at inhibiting the enzyme, at a 1:1 molar ratio, in a cell..free milieu within the time periodof the experiments described (data not shown). Isolation of Human PMN Human PMN were obtained from heparinized venous blood from nonsmoking volunteers. After dextran (3% in saline) sedimentation, PMN were separated by Ficoll-Hypaque centrifugation (7). Remaining red blood cells were lysed with a solution containing 150 mM NH 4CI, 0.11 mM EDTA, 10 mM KHC03, (pH 7.3). After centrifugation, cells were resuspended in Ca't/Mgv-frec Hanks' medium, containing 20 mM Hepes and 0.1% bovine serum albumin. Differential counting with crystal violet revealed> 95 % pure and viable PMN. Assays of PMNElastase Activity and Other PMN Enzymes Proteolysis of fibrinogen by'PMN elastase was quantified by radioimmunoassasy (RIA) (8). The peptide Aa(1-21) was assayed in a competitive RIA using a highly specific rabbit polyclonal antibody (R20). Enzymes released by PMN were measured in fluid phase, assays. Myeloperoxidase (MPO) was measured by its ability to oxidize 3,~-dimethoxy­ benzidine in a spectrophotometric assay and expressed in absorbance units at 450 nm. Released elastase in the medium was measured in a spectro-fluorometric assay, using MeoSucAla-Ala-Pro-Val-MCA. Elastase activity was expressed in arbitrary fluorescence units (FU). Fibrinogen Matrix Degradation by Purified PMN Elastase in the Absence or Presence of Elastase Inhibitors Human fibrinogen (500 III of 2.4 mg/ml in 1.5 % NaHC0 3) was coated onto 24-well plates at 37° C overnight. Protein

determination performed before and after coating revealed a 50 % coating efficiency. The wells were washed with phosphate-buffered saline, followed by addition of 475 III medium. Next, 25 III of 1.6 IlM or 16 ,aM purified human PMN elastase was added to the wells. The effect of elastase inhibitors was measured by preincubation of the coated wells with 475 III medium containing a\AT or ALP at concentrations ranging from 0.001 to 1.8 IlM followed by addition of 25 III of 1.6 IlM purified elastase. All plates were incubated for 60 min at 37° C, after which the incubations were stopped by cooling on ice. Samples were assayed for Aa(l-21) levels. Fibrinogen Degradation by PMN in the Presence of Different Secretagogues Approximately 1.0 x 1()6 PMN/well in 500 III culture medium (with Ca2+ and Mg2+) were incubated with 1.0 nM rTNF-a for up to 2.5 h at 37° C. At different time points, 100 IIIcell-free samples were assayed for Aa(1-21) formation and fluid phase elastase activity as described above. In another series of experiments, approximately 1.0 x 106 PMN in 0.5 ml medium (with Ca2+ and Mg2+) were incubated with 1.0 nM rTNF-a for 1 h, followed by stimulation with different concentrations of FMLP, ranging from 0.003 to 1.0 IlM in the presence of5 IlMCB. Formation of Aa(1-21) in the samples was assayed by RIA. Elastase and MPO activity were measured by fluid phase assays as described above. Degradation of Fibrinogen by Stimulated PMN in the Presence of Elastase Inhibitors Fibrinogen-coated wells were preincubated with 0.5 ml medium containing 1.0 nM rTNF-a and concentrations of ALP or alAT ranging from 0.01 to 0.9 IlM. PMN (1.0 x 106) were added and incubated for 2.5 hat 37° C. Cell-free medium was assayed for Aa(1-21) and fluid phase elastase. In addition, potencies of ALP and alAT to inhibit PMN elastase released by stimulation of PMN with CB and FMLP were compared in cultures in which PMN had been attached onto a fibrinogen matrix. Approximately 1.0 x 106 cells were incubated with 1.0 nM rTNF-a for 1 h to promote attachment, followed by a washing step. Next, cells were preincubated in 0.5 ml with equimolar concentrations of ALP and aIAT, ranging from 0.05 to 0.9 IlM. PMN were stimulated with 5 IlM CB and 0.5 IlM FMLP for 1 h at 37°C. The cell-free medium was assayed for Aa(1-21) generation and fluid phase elastase activity. Fibrinogen Degradation by Stimulated PMN from Chronic Granulomatous Disease (CGD) Patients' in the Presence of Inhibitors Fibrinogen proteolysis was measured in the presence of a IAT and ALP, using PMN (1.0 x 106 ) from patients with CGD. PMN from CGD patients are characterized by their inability to produce oxidants (9). PMN were isolated from blood obtained from four different CGD patients, all with X-linked NADPH-oxidase deficiency (courtesy of Dr. R. S. Weening, Department of Paediatrics, Academic Medical Centre, Amsterdam, The Netherlands). Incubation with PMN attached to fibrinogen matrix was performed as described above.

Stolk, Davies, Kramps et al.: ALP and ajAT Inhibit Fibrinogen Degradation by Adherent PMN

Statistics Because each Aa(l-21) assay had its own standard curve, it was evident that experimental conditions were stable within an assay. In contrast, biologic variability of cells obtained from an individual at different times and of cells from different individuals was rather high when cells were stimulated with TNF-a followed by CB + FMLP. Therefore, although the observations from successive doses in the same assay were highly correlated, the levels of response in some experiments varied appreciably between assays. For this reason, the Jackknife method (l0) was used to estimate the mean inhibitor concentrations at which 50 % inhibition (IC so) of proteolysis occurred.

0.5

20

0.4 15

] (5

0.3

E

0 LO

'>t

C.

~

Ec:

10

0

0.2

Q. 0

a..

i::l

«

~

5

0.1

-+-----.------.------.------,----+- 0.0 10-3 10-2 10- 1 100 101 ~M FMLP

Results Fibrinogen Degradation by Purified PMN Elastase Incubation of fibrinogen-coated wells for I h at 37° C with 0.5 ml medium containing 40 and 400 pmol purified human PMN elastase resulted in the cumulative formation of 75.9 ± 10.3 and 800 ± 20.1 pmol/ml of Aa(I-21) (mean ± SD), respectively. Preincubation of coated wells with ajAT or ALP in concentrations ranging from 0.5 to 900 pmol, followed by addition of 40 pmol elastase (which is 67% active), resulted in ICso values of 10 nM for both inhibitors. This value is slightly less than that expected for inhibition of 50 % of active enzyme. Fibrinogen Degradation by PMN Stimulated by Different Secretagogues When PMN (1.0 X 106) in 0.5 ml culture medium were incubated for 1 and 2.5 hat 37° C in fibrinogen-coated wells in the absence of secretagogues, we measured formation of 0.75 ± 0.3 and 0.9 ± 0.2 pmol/ml Aa(I-21) (mean ± SD; n = 3), respectively. Addition of rTNF-a facilitated PMN

8

6

E

:::::0

E c.

~

523

Figure 2. Formylmethionylleucylphenylalanine (FMLP)-induced

stimulation of myeloperoxidase (MPa) release and the formation of Aa(l-21) by human PMN. Attached PMN in 500 1£1 Hanks' medium were preincubated with 5.0 JLM cytochalasin B (CB). Sixty minutes after stimulation with different concentrations of FMLP, MPa activity was determined by a spectrophotometric assay (diamonds). Aa(1-21) formation (squares) after 60 min was assayed in an RIA as described in the text. Results shown are means of duplicate tests of a representative experiment.

attachment as previously reported (3). Maximal attachment was obtained at an rTNF-a concentration of 1.0nM (3). Figure 1 shows the cumulative formation of Aa(I-21) when PMN (1.0 x 106) are incubated in 0.5 ml medium containing 1.0nM rTNF-a. After 1 and 2.5 h, we measured low levels, 2.0 ± 0.1 and 7.0 ± 0.5 pmol Aa(l-21), respectively (mean ± SD; n = 3). Cell-free aliquots from each well, incubated with fluorogenic substrate, did not show any free elastase activity in the medium (data not shown). Cell-free aliquots from each well also did not show any detectable levels of MPO as also described by Hanlon and associates (3). After attachment of PMN (1.0 x 106) onto fibrinogen matrices by rTNF-a, the wells were washed and preincubated with fresh medium containing 5 p.MCB. After 5 min, different concentrations of FMLP were added. The release of elastase into the medium occurred in a concentration- and time-dependent fashion, with maximal response of approxi-

4

«i::l

0.5

30 2 25

l(5 E

50

100

150

200

Time (min)

Figure 1. Time-dependent formation of Aa(1-21) by human poly-

morphonuclear leukocytes (PMN) in response to recombinant tumor necrosis factor-a (trNF-a). Human PMN (1.0 x 1Q6/well) in Hanks' medium were incubated with 1.0 nM trNF-a in a fibrinogen-coated well (squares). At various times, the amount of the fibrinopeptide Aa(1-21) was measured by radioimmunoassay (RIA) as described in the text. Control wells (diamonds), containing PMN not stimulated with trNF-a, showed background Aa(121) formation. Values are mean ± SD of three experiments performed in duplicate.

0.4 20

0

LO

C.

~ i::l

Ec:

0.3

15

'>t

0

Q. 0

10

«

0.2

o, ~

5 -tY",--.r-rTT...-rrTT"T"T"1rrT""T"T"1.--rTTO.....-rTT"T"""+-

10

20

30 40 Time (min)

50

60

0.1

70

Figure 3. Time-dependent release of MPa (diamonds) and the formation of Aa(1-21) (squares) by PMN after stimulation with 5.0

1£M CB and 0.1 1£M FMLP. Results shown are means of duplicate tests of a representative experiment.

524

AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 6 1992

mately 700 FU at 0.1 JlM FMLP within 5 min. MPa activity had a maximal response of 0.400 absorbance units at 0.1 JlM FMLP within 20 min (Figures 2 and 3). The cumulative formation of Aa(I-21) over a l-h period of time resulted in a plateau at a concentration of 0.3 JLM FMLP, at which 17.6 ± 1.7 pmol Aa(I-21) (mean ± SO; n = 3) was formed (Figure 2). After 60 min of stimulation with 0.1 JlM FMLP, Aa(l-21) levels of 20.4 ± 2.7 pmol/ml (mean ± SO; n = 3) were seen (Figure 3). Degradation of Fibrinogen by Stimulated PMN in the Presence of Elastase Inhibitors The potencies of ajAT and ALP to inhibit elastase activity after addition of rTNF-a to PMN (1.0 X 106 ) were compared. The formation of Aa(l-21) was inhibited in a concentration-dependent fashion by both inhibitors (Figure 4). The ICso of ALP for inhibition of Aa(1-21) formation was 17 ± 3 nM (mean ± SO; n = 3). The IC so of a]AT was 60 ± 5 nM (mean ± SO; n = 3), being significantly less potent than ALP (P < 0.01). Using fluid phase assays performed with cell-free media, active elastase was not detectable in the presence of inhibitors. To further characterize the potency of ALP and a I AT, attached PMN were stimulated with CB and FMLP to release azurophilic granules. Enzyme release from cells was monitored by fluid phase for elastase and MPa in the cellfree supernatants. In the absence of inhibitors, PMN (1.0 x 106/well) stimulation resulted in the generation of 20 ± 3 pmol Aa(I-21)/ml (mean ± SO; n = 7) with 0.5 JlM FMLP after 60 min. When attached cells were preincubated before FMLP stimulation with ALP and oAl' at concentrations ranging from 0.05 to 0.9 JlM, Aa(I-21) generation was inhibited with an IC so of 85 ± 30 nM. (mean ± SO; n = 7) and 220 ± 98 nM (mean ± SO; n = 7) (Figure 5), respectively. These values were significantly different by analysis with the Jackknife method (P < 0.05). At or above the IC so concentrations, no free elastase activity was detectable in

25

20

:€ (5 E

15

0-

~

10

(:j

« 5

o

o

//

I

i i i II

'I

I

I

I i IjilJ

10-1

10°

i

II

I I Iii

10 1

pM Inhibitor

Figure 5. The effect of ALP and atAT on Aa(l-21) formation by

FMLP-stimulated PMN. ALP and ajAT were added at indicated concentrations to adhered PMN in fibrinogen-coated wells and incubated at 37° C for 15 min in the presence of CB, prior to addition ofFMLP (0.5 JLM). The cells were incubated for 60 min. Stimulation without inhibitors resulted in the formation of 20 ± 3 pmol Aa(l-21)/ml. The ICso values of ALP (diamonds) and ajAT (squares) were 0.085 ± 0.030 and 0.220 ± 0.098 JLM, respectively (mean ± SD; n = 7). MPO activity in the medium was not significantly affected by increasing inhibitor concentrations.

cell-free medium using fluid phase assays. MPa levels in all wells were comparable, demonstrating equal release of azurophilic granules from PMN in all wells (data not shown). Fibrinogen Degradation by Stimulated PMN from eGD Patients in the Presence of Inhibitors PMN (1.0 x 106) isolated from patients with CGO were stimulated with rTNF-a. This resulted in a similar efficiency of attachment to fibrinogen as obtained with PMN from normal subjects (data not shown). Subsequent stimulation with

6 50

5

:€(5

4

40

E

E ::::,.

0-

~ (:j

«

3

a

E

0-

2

~

30

20

(:j

«

o

o

/~

i I

lilj

10- 2

I

I I i 11'1

i

I

10- 1 pM Inhibitor

I J 1111'

10°

I

I I

¥

1111

10

10 1

Figure 4. The effect of antileukoprotease (ALP) and QI-antitrypsin (a]AT) on the formation of Aa(l-21) by PMN stimulated with rTNF-Q. ALP and alAT were added at indicated concentrations to 500 ~l Hanks' medium in fibrinogen-coated wells. PMN were incubated with 1.0 nM rTNF-a for 150 min. When cells were incubated without inhibitors for this period of time, 7.0 ± 0.5 pmol Aa(l:'21)/ml was generated. The ICso values of ALP and alAT were 0.017 ± 0.003 JLM (diamonds) and 0.06 ± 0.005 JLM (squares), respectively (mean ± SD; n = 3).

0

//, i I II

0

10-1

I

I

I i i III

10- 1 pM Inhibitor

I i I III

10° .

Figure 6. The effect of ALP and a]AT on Aa(1-21) formation by

FMLP-stimulated PMN from patients with chronic granulomatous disease (CDG). Experimental conditions were as described in the legend to Figure 5. Stimulation without inhibitors generated 40 ± 4 pmol Aa(l-21)/ml. The IC so values of ALP (diamonds) and a1AT (squares) were 0.050 ± 0.013 and 0.150 ± 0.021 JLM, respectively (mean ± SD; n = 3). .

Stolk, Davies, Kramps et al.: ALP and ajAT Inhibit Fibrinogen Degradation by Adherent PMN

5 p.M CB and 0.5 p.M FMLP resulted in formation of 40 ± 4 pmol Aa(1-21) in 60 min (mean ± SD; n = 4). This is twice the amount of Aa(1-21) as formed by the same number of stimulated PMN from normal subjects (Figure 5). MPO activity released into the medium by stimulated CGD cells did not differ from that released by normal cells after stimulation. Addition of ALP or a,AT at concentrations ranging from 0.01 to 0.9 p.M to attached CGD cells before FMLP stimulation resulted in a concentration-dependent inhibition of Aa(I-21) formation (Figure 6). In this series of four experiments, ALP had an IC 50 of 50 ± 13 nM, whereas a,AT had an IC50 of 150 ± 21 nM (mean ± SD) (Figure 6). This difference in IC 50 is significantly different in the Jackknife method (P < 0.05).

Discussion In the study reported here, we have evaluated the potencies of ALP and a,AT to inhibit the generation of Aa(I-21), the PMN elastase-specific cleavage product of fibrinogen (8), by adherent PMN stimulated with FMLP and CB of both normal subjects and patients with CGD. The experiments described in this report may mimic a number of events that take place during chronic inflammatory processes in the lung. The attachment of PMN, which is promoted by TNF-a, might be a relevant mechanism in vivo, as alveolar macrophages can produce increased amounts ofTNF-a in inflammatory reactions in the lung (11, 12). TNF-a facilitates PMN attachment (3), followed by the formation of an elastase-specific cleavage product of fibrinogen in a time-dependent fashion (Figure 1). The lag phase of formation of Aa(1-21) of about 25 min (Figure 1)coincides with the time PMN need for optimal attachment (3) and begin to start production of free radicals (13). From that point onward, cumulative production of Aa(1-21) could be measured. Free elastase activity could not be detected in cellfree supernatants during stimulation of PMN with rTNF-a. Nevertheless, the presence of Aa(l-21) is most readily explained by the local release of small amounts of elastase as Aa(I-21) is a specific cleavage product of fibrinogen generated by PMN elastase (14). Significant release of the azurophilic granule markers, MPO and elastase, could be obtained only after cells adhered with rTNF-a were stimulated with CB and FMLP. The combination of stimuli used in our experiments was reported by Nathan and co-workers (13, 15) to have different effects on superoxide anion production. CB inhibited superoxide production by adherent neutrophils (13), whereas adherent cells responded to rTNF-a with a massive, prolonged production of superoxide anion, which could be further potentiated by subsequent stimulation with FMLP (15). In a cell-free system, utilizing purified human PMN elastase to degrade fibrinogen, we showed that ALP and alAT were equally potent inhibitors of Aa(l-21) formation. However, when PMN were attached onto the fibrinogen matrix, ALP was found to be more efficient in the inhibition of Aa(I-21) formation than a,AT (Figures 4 and 5). We hypothesize, in accordance with Campbell and Campbell (16) and with recent results obtained by Rice and Weiss (17), that ALP gains better access to the interface between PMN and the extracellular matrix than does a,AT. Stimulation of

525

PMN by rTNF-a alone provides the most suitable circumstances to test our hypothesis. This is because elastasespecific fibrinogen fragments could be measured without detectable elastase activity in the fluid phase. This strongly suggests the existence of an interface between PMN and protein matrix, in which elastase is active (16). Under these circumstances, the IC 50 of ALP was significantly lower than that of alAT (17 versus 60 nM, respectively; P < 0.01). The experiments depicted in Figures 5 and 6 demonstrate that ALP is also significantly more potent than ajAT when attached PMN are stimulated by CB and FMLP. Under these conditions, the IC50 are higher than when TNF-a alone was used as stimulus (80 versus 17 nM for ALP; 220 versus 60 nM for a tAT) and active elastase was found in the supernatant also. The level of MPO found in cell-free supernatants of PMN preincubated with either inhibitor and then stimulated were no different from the MPO levels found in stimulated PMN not exposed to inhibitor. In Figure 6, we performed experiments with PMN from patients with CGD. PMN from these patients lack a functional membraneassociated NADPH-oxidase and therefore are unable to produce reactive oxidant intermediates (9). Thus, PMN from CGD patients are suitable to evaluate the effect of reactive oxidant intermediates on the potency of ALP and alAT. Figure 6 shows that the IC 50 of ALP and alAT (50 versus 150 nM, respectively) are lower than observed in Figure 5 (85 versus 220 nM, respectively). This observation is consistent with the finding that both inhibitors are susceptible to inactivation by oxidants from normal PMN as was previously reported by Dean and co-workers (18) and from our laboratories (7). We also observed that the amount of Aa(I-21) formed by PMN from CGD patients was approximately double that formed by PMN from normal subjects. The in-

400 c 0

:;:::;

:c 300 :c E