and Its Inhibitor Human-Polymorphonuclear-Leucocyte Neutral Protease Studies with Fluorescein-Labelled Polymeric Collagen Fibrils as a Substrate

Transcription

1 Eur. J. Biochem. 67, (1976) HumanPolymorphonuclearLeucocyte Neutral Protease and Its Inhibitor Studies with FluoresceinLabelled Polymeric Collagen Fibrils as a Substrate Frank S. STEVEN, David W. MILSOM, and John A. A. HUNTER Department of Medical Biochemistry, University of Manchester, and Department of Rheumatology, Manchester Royal Infirmary, University of Manchester (Received February 23/April 30, 1976) Human polymorphonuclear leucocytes were obtained from the synovial fluids of patients with inflamed knee joints suffering either from Reiter's syndrome or from rheumatoid arthritis. The polymorphonuclear leucocytes were collected by gentle centrifugation followed by disruption and their subcellular fractionation by centrifugation in 0.34 M sucrose to provide a granule fraction and a postgranule supernatant fraction. 0.5 M KCl extraction of the granule fraction yielded neutral protease activity, similar to trypsin, when assayed against fluoresceinlabelled polymeric collagen fibrils. The postgranule supernatant fraction contained an inhibitor towards the neutral protease and trypsin. The inhibition of the neutral protease was found to be timedependent, this inhibition being released after h. In contrast, the inhibition of trypsin was irreversible and this property was used to devise an assay procedure for the inhibitor. Kopitar and Lebez [l] demonstrated the presence of a neutral protease associated with the granule fraction obtained from pig polymorphonuclear leucocytes. They also showed that this enzyme could be inhibited by the addition of a protein fraction present in the postgranule supernatant fraction. At the same time, a similar neutral protease was obtained from extracts of polymorphonuclear leucocges present in the synovial fluid of patients with rheumatoid arthritis [2]. This neutral protease behaved similarly to trypsin when assayed with fluoresceinlabelled polymeric collagen fibrils as substrate [2]. These collagen fibrils contained three fluoresceinlabelled amino groups located in the telopeptide regions of each tropocollagen molecule within the polymer. Kinetic studies showed that two fluoresceinlabelled groups per tropocollagen molecule were removed from the telopeptide regions by both the neutral protease and by trypsin. Since the neutral protease was assayed in the presence of the soluble fraction of the cell sonicate [2], the question arose as to how this enzyme could be active in the presence of the inhibitor described by Kopitar and Lebez [l]. This system has now been reexamined, the presence of the inhibitor has been confirmed [l] and the reason for the lack of inhibition with the polymorphonuclear leucocyte sonicates [2] has been elucidated. The present study shows this neutral protease inhibition to be time dependent and to be reversed after Enzyme. Trypsin (EC ). a period of h at 37 "C with the subsequent total recovery of neutral protease activity towards fluoresceinlabelled polymeric collagen fibrils. MATERIALS AND METHODS Crystalline trypsin (treated with L2tosylamido 2phenylethyl chloromethyl ketone) was obtained from Worthington. Synovial fluid was obtained from the knee of a patient with Reiter's syndrome; 22 ml of fluid containing a total of approximately 1.4 x lo5 polymorphonuclear leucocytes was obtained. A similar number of cells were obtained from the synovial fluid removed from the knee of a rheumatoid arthritic patient. Both preparations of cells provided a neutral protease and also an inhibitor of the enzyme. The relative activities of these two activities varied but the patterns of results obtained were identical for both preparations. In the present study the cells from the Reiter's syndrome are described. The cells were collected by lowspeed centrifugation, washed in 0.9% saline and suspended in 10 m10.34 M sucrose containing 50 units heparin/ml. The cells were homogenised by drawing vigourously through a Swinnex filter holder approximately 15 times. The homogenate was centrifuged at 600 xg for 10 min to remove unbroken cells and cell debris. The resultant supernatant fraction was centrifuged

2 166 Neutral Protease and Inhibitor Present in Human Leucocytes for 15 min at x g to provide a pellet of granules and a postgranule supernatant fraction (10 ml) comparable to those prepared by Kopitar and Lebez [l]. ENZYME ASSAY PROCEDURES Both the neutral protease and trypsin activity were assayed with fluoresceinlabelled polymeric collagen fibrils as described previously [2]. Neutral Protease and Its Inhibition Two systems were used when the actions of the neutral protease and its inhibitor were being studied. In the first, the enzyme concentration was maintained at a constant level (100 p1 of granule extract) in 5 ml of 50 mm Tris, 5 mm CaCl,, 500 mm KCI buffer ph 7.5 containing 7.0 mg fluoresceinlabelled polymeric collagen fibrils, with the addition of variable quantities of the postgranule supernatant fraction (0 100 pl). After incubation at 37 "C for 1.5 h, the quantity of solubilised fluoresceinlabelled peptides in each tube was determined by the fluorescent emission of suitable diluted (1/50) samples of the reaction mixture supernatant fraction (Fig. 1). In the second method, the effect of time of incubation at 37 "C on the solubilisation of fluoresceinlabelled peptides from the substrate by the granule extract (100 pl) was measured (a) in the absence of the inhibitor (Fig. 2A), (b) when preincubated with postgranule supernatant fraction (100 pl) at 37 "C for 1 h prior to adding to the substrate (Fig.2B) and (c) when both the enzyme and the inhibitor were added simultaneously to the substrate at 37 "C (Fig.2C). In each instance 7.0 mg of substrate was suspended in 5.0 ml of the Tris/CaCl,/KCl buffer and the samples were assayed as described above, over a period of 4 h. Trypsin and Its Inhibition Trypsin had been shown to behave similarly to the polymorphonuclear leucocyte neutral protease [2] in selectively cleaving two fluoresceinlabelled peptides per tropocollagen molecule within the polymeric collagen fibrils. It was therefore of interest to examine the effect of the postgranule supernatant fraction on the action of trypsin on the fluoresceinlabelled polymeric collagen fibrils. Two test systems were employed for this study. In the first, trypsin (1 pg) was added to a series of tubes containing 5 ml of the Tris/CaCl,/KCI buffer, 7.0 mg substrate plus the postgranule supernatant fraction to give a range of concentrations of the latter (i. e. 25, 50, 75 and 100 pl). Incubation at 37 "C was carried out for 5 h and 100pl samples were withdrawn by means of a microsyringe at hourly intervals for fluorimetric analysis (Fig. 3). A second experimental system was designed to establish the concentration of trypsin which could be completely inactivated by a known quantity of postgranule supernatant fraction. It was possible to achieve this objective by two slightly different experimental approaches. The first approach consisted of two parts, (a) a series of tubes containing 5 ml buffer, 7.0 mg substrate to which had been added a range of trypsin concentrations (0 4 pg/5 ml), and the solubilisation of fluoresceinlabelled peptides estimated at 30min intervals (Fig.4A shows the solubilisation after 2 h). An identical series of tubes (B) were prepared but with the addition of 50 pl postgranule fraction to each tube at the start of the experiment (Fig.4B shows the extent of digestion at 2 h). The second approach employed a fixed quantity of trypsin (2 pg) with the addition of increasing volumes of the postgranule supernatant fraction in 5 ml buffer plus 7.0 mg substrate. Digestion was allowed to proceded at 37 "C for 1.5 h and then the solubilisation of fluoresceinlabelled peptides measured as described above (Fig. 5). RESULTS AND DISCUSSION The Neutral Protease and Its Inhibition The granule fraction contained a neutral protease capable of attacking the telopeptide regions in fluoresceinlabelled polymeric collagen fibrils, confirming earlier reports [1,2]. When 100 pl of the granule extract was incubated with the fluorescent substrate in the presence of increasing quantities of the postgranule supernatant fraction for 1.5 h (Fig. l), the percentage inhibition of proteolytic activity was directly proportional to the amount of postgranule supernatant fraction added to the reaction mixture. From Fig.1 it can be calculated that the neutral protease present in 100 p1 granule extract was almost completely inactivated by the inhibitor present in 100 pl of the postgranule supernatant fraction, confirming the observations of Kopitar and Lebez [l]. Since the cells yielded a total of 3 ml granule extract and 10 ml postgranule supernatant fraction, it is clear that the cells contained approximately 3.3 times as much inhibitor in the cytoplasm as was found to be necessary to neutralise the activity of all the neutral protease extracted in 0.5 M KC1 from the granules of these cells. It was found, however, that if the digestion was allowed to continue at 37 "C for a further 16 h there was no measurable difference in the quantity of soluble fluoresceinlabelled peptides present in the tubes containing the enzyme and substrate only and those tubes with the addition of the postgranule supernatant fraction. It appeared that the inhibition observed after 1.5 h (Fig. 1) was reversible and enzymic

3 F. S. Steven, D. W. Milsom, and J. A. A. Hunter digestion of the substrate took place after prolonged incubation as though no inhibition had taken place previously. This observation was not made by Kopitar and Lebez [l], since these authors followed the digestion and inhibition for short periods of time only (1 2 h). The time dependence of the inhibition is shown in Fig.2. Fig.2 (A) shows the typical solubilisation of flouresceinlabelled peptides with time from the insoluble substrate by 100 pl of the granule extract. The quantity of solubilised products increased linearly with time, line (A) passing through the origin. Fig.2 (C) shows the same reaction mixture used for line (A) but with the simultaneous addition of 100 pl postgranule supernatant fraction. In this case, a linear relationship between the products formed and the reaction time is shown, but the time of the initiation of soluble product formation has been delayed by approximately 1.5 h, due to the presence of the inhibitor. After 1.5 h, it would appear from line (C) that the enzyme is released from the inhibitor and proteolysis takes place as though no inhibitor was present. This delay in enzymic digestion of the fluoresceinlabelled polymeric collagen fibrils explains why the initial inhibition observed after 1.5 h in Fig. 1 was not observed after prolonged incubation (see above). Preincubation of the granule extract with the postgranule supernatant fraction (Fig. 2B) for 1 h prior to adding the substrate resulted in a delay in the solubilisation of fluoresceinlabelled peptides by approximately 1 h. This initial delay was followed by the progressive solubilisation of products with time in a manner similar to that observed when the granule extract alone was incubated with the substrate (see Fig. 2A). This preincubation shortened the period of effective inhibition of the neutral protease by the postgranule supernatant fraction. These results suggest a temporary form of inhibition of the neutral protease (E) solubilisation of fluoresceinlabelled peptides (P) from the substrate (S) by an inhibitor (I) present in the postgranule supernatant fraction as presented in the following scheme. E+I+S inactive inhibitor active *(EI) + S Enzymeinhibitor complex I no products, temporary inhibition h37 "C I*+E+S I enzyme (ES) I enzyme substrate complex P Products measured as fluoresceinlabelled peptides._ jo: *Ot 10 t Supernatant /extract (bl/loopl) Fig. 1. Inhibition of' granule extract neutral protease (I00 pl) hy uddition of variuble quantities (0 100 plj of the postgranule supernatant,fruction. Inhibition measured after 1.5 h digestion of fluoresceinlabelled polymeric collagen fibrils and expressed as percentage activity in the complete absence of inhibitor A ;I Time (h) Fig. 2. The ejfect of incubation time on the solubilisation of' fluoresceinluhelledpeptides from the substrate in the presence of 100 pl granule extract. (A) No inhibitor added, (B) granule extract preincubated for 1 h with 100 pl postgranule supernatant fraction prior to adding to the substrate, and (C) granule extract mixed with 100 p1 postgranule supernantant fraction and simultaneously added to the substrate. The peptides in the soluble fraction were measured by their percentage fluorescent emission on scale 0.03 employing a dilution of 1/50 It was observed that in the absence of KCl the neutral protease was incapable of solubilising fluoresceinlabelled peptides from the substrate, but that on addition of KC1 to 0.1 M concentration, the digestion of the substrate took place immediately. This sug

4 168 gests that the neutral protease activity may be confined to an intracellular role and that the function of the inhibitor in the cytosol is to protect the cell contents from destruction in the event of the neutral protease escaping from the granules. The only situation in which the neutral protease would be expected to act extracellularly would be when an extracellular increase in K+ resulted from the death of a large number of cells. Inhibition of Fraction Trypsin by the Postgranule Supernatant It was observed that trypsin was irreversibly inactivated by the postgranule supernatant fraction (Fig. 3 5). Employing low concentrations of trypsin (less than 2 pg trypsin per 7.0 mg substrate in 5.0 ml reaction mixture) the solubilisation of fluoresceinlabelled peptides is linear over a period of 5 h (Fig.3A). The addition of 25, 50, 75 or 100 pl of the postgranule supernatant fraction resulted in all instances in the complete inactivation of 1 pg trypsin (Fig. 3B). There was no evidence for the reactivation of trypsin after 18 h at 37 "C, in contrast to that observed with the neutral protease extracted from the polymorphonuclear leucocyte granules (Fig. 2). It was possible to determine the quantity of trypsin which was inhibited by 50 pl of the postgranule supernatant fraction by assaying the inhibitory ability of this quantity of inhibitor against increasing concentrations of trypsin (Fig.4). Fig.4A shows the quantity of flouresceinlabelled peptides solubilised with increasing quantities of trypsin after 2 h at 37 "C; it is made up of two linear portions, indicating that above 2 pg trypsin per 7.0 mg substrate, further increase in the enzyme concentration has little effect on the quantity of products formed. Similar observations were made for collagenase [3] and a molecular explanation provided. 50 pl of the postgranule supernatant fraction (Fig. 4B) completely inhibited up to 2 pg trypsin; however, when the trypsin added to the assay system exceeds 2 pg, digestion of the substrate proceeds on a course roughly parallel to that of trypsin alone (compare the latter part of B with the first part of A in Fig.4). Similar pairs of lines were obtained after shorter periods of incubation, i. e. 0.5, 1.0 and 1.5 h, but for simplicity only one pair has been presented in Fig. 4. The results presented in Fig.4 indicate that when the quantity of trypsin exceeds 2 pg, the enzyme is in excess of the inhibitor present in 50 p1 postgranule supernatant fraction. Above 2 pg trypsin the trypsin digestion proceeds normally and is not affected by the inhibitor already in complex formation as EI. It can be stated that 50 pl of the postgranule supernatant fraction completely inactivates 2 pg trypsin under the test conditions. Neutral Protease and Inhibitor Present in Human Leucocytes t " 2001 c Y) c a' 0, LL A i A B ii.l Time (h) Fig. 3. Eflect of incubation time on the.roluhilisation of fluoresceinlubelledpeptides,f~om the substrate in thr presence of1 pg of trypsin. (A) Trypsin alone, (B) trypsin plus 50 pl postgranule supernatant fraction. Peptides measured by their percentage fluorescent emission on scale 0.03 employing a dilution of 1/50 Fig.4. The effrct of 50 pl of the postgranule superncltant fiuction on pg trypsin digestion of' fluoresceinlabelled polymeric collagen fibrils. The solubilised peptides were measured by their percentage fluorescent emission on scale 0.03 employing a dilution of 1/50 after a 2h incubation at 37 "C. (A) Addition of 04.0 pg trypsin to substrate; (B) addition of pg trypsin plus 50 p1 postgranule supernatant fraction to substrate at the beginning of the experiment. Similar graphs were obtained at each 30min interval of digestion at 37 "C; however, only graphs obtained at 2h have been presented here for simplicity in expressing the results The inhibition of 2 pg trypsin in the presence of increasing amounts of the postgranule supernatant fraction is shown in Fig. 5. It can be seen that 2 pg of

5 F. S. Steven, D. W. Milsom, and J. A. A. Hunter 169 got\ I,,d ' Supernatant (~l) Fig. 5. The irihihition of2,ug trjpsin by increusing quantities qfthepostgrnnrtle suijernatantfrac,tion. The results are presented as percentage trypsin activity in the absence of inhibitor after 1.5 h digestion at 37 c trypsin were completely inhibited by 60 pl of the postgranule supernatant fraction. A minor discrepancy between the results in Fig. 4 and Fig. 5 needs a few words of explanation. In Fig. 4, 2 pg trypsin were inactivated by 50 pl of inhibitor, whilst in Fig pl of inhibitor was required to inactivate 2 pg trypsin. The results were obtained for Fig. 4 with freshly isolated postgranule supernatant fraction; however this was stored for 3 weeks at 4 "C prior to obtaining the results presented in Fig.5. It was to be expected that some loss in inhibitory action might be observed on storage. So far it has not been possible to determine whether a single inhibitor is present in the postgranule super natant fraction capable of inactivating both the neutral protease and trypsin or whether more than one inhibitor exists. The work described in this report confirms and extends previously published data [I. 21. It is of particular interest that the inhibition reported by Kopitar has now been shown to exhibit a timedependent reversible inhibition on the neutral protease. It is also of interest that although trypsin behaves very similarly to the neutral protease [2] and is also inhibited by the postgranule supernatant fraction, the inhibition of trypsin is irreversible over 18 h at 37 "C. Clearly the cytoplasm of these cells contains a potent inhibitor system capable of inactivating a neutral protease which is normally stored within the granules of these cells. The inhibition of the neutral protease is of relatively short duration and this would explain why whole cell homogenates [2] extracted with KC1 exhibited neutral protease activity, in spite of the fact that a potent inhibitor of this activity was also present in the initial homogenate [l]. It is probable that in uivo, the release of neutral protease activity from dying polymorphonuclear leucocytes would be prevented froin causing tissue damage, by the simultaneous release of the inhibitor present in the cytoplasm. During the period of h this extracellular inhibited neutral protease in the form of an EI complex would be disposed of by phagocytosis in the normal manner by healthy polymorphonuclear leucocytes. REFERENCES 1. Kopitar, M. & Lebez, D. (1975) Eur. J. Biochiwi. 56, Steven, F. S., TorreBlanco, A. & Hunter, J. A. A. (1975) Biochim. Biophys. Actu, 405, Steven, F. S. (1976) Bioc,/irm. J F. S. Steven and D. W. Milsom, Department of Medical Biochemistry, University of Manchester Medical School, Stopford Building, Oxford Road, Manchester, Great Britain, M13 9PT J. A. A. Hunter, Department of Rheumatology, Manchester Royal Infirmary, Oxford Road, Manchester, Great Brilain, M13 9PT