Eur. J . Biochem. 151, 327-331 (1985) $3 FEBS 1985

Human serum a -antichymotrypsin is an inhibitor of pancreatic elastases Anne LAINE ', Monique DAVRIL ', Michel RABAUD', Dominique VERCAIGNE-MARK0 ' and Annettc HAYEM ' Institut National de la Sante et de la Recherche Medicale, Unit6 16, Lille

Institut National de la Santk et de la Recherche Medicale, Unite 8, Pcssac

(Received February 12/April 29, 1985) - EJB 85 0140

Incubation of human serum a,-antichymotrypsin with human pancreatic elastase 2 or porcine pancreatic elastase results in the complete inhibition of each enzyme as determined by spectrophotometric assays. a , - Antichymotrypsin reacts much more rapidly with the human than with the porcine enzyme. The inhibitor:enzyme molar ratio, required to obtain full inhibition of enzymatic activity, is equal to 1.25/1 when a,-antichymotrypsin reacts with human pancreatic elastase 2 while it is markedly higher with porcine pancreatic elastase (5.5/1).

Patterns obtained by SDS/polyacrylamide gel electrophoresis of the reaction products show the formation with both enzymes of an equiinolar complex ( M , near 77000) and the release of a fragment migrating as a peptide of M , near 5000. Moreover a free proteolytically modified form of a,-antichymotrypsin, electrophoretically identical with that obtained in the reaction with cathepsin G or bovine chymotrypsin, is produced in the reaction with each elastase but in a much greater amount when a,-antichymotrypsin reacts with porcine elaslase than with human elastase. As a consequence of our findings, the specificity of a,-antichymotrypsin, so far limited to the inhibition of chymotrypsin-like enzymes from pancreas and leukocyte origin, has to be extended to the two pancreatic elastases investigated in this work.

A contribution of al-antichymotrypsin to the regulatory balance between plasma inhibitors and human pancreatic elastase 2 in pancreatic diseases is suggested.

Human a,-antichymotrypsin is a glycoprotein first de- scribed about twenty years ago [l]. This plasma proteinase inhibitor is also an early-stage acute-phase plasma protein; its concentration may more than double within sixteen hours after trauma [2]. The primary role of al-antichymotrypsin seems to be the specific regulation of chymotrypsin-like enzymes [3]. Indeed ccl-antichymotrypsin forms equimolar in- hibitor-enzyme complexes with chymotrypsins from different origins [3, 41 and leukocyte cathepsin G [3, 51. We have pre- viously shown that the complexes with bovine pancreatic chymotrypsin [6] or human leukocyte cathepsin G [ S , 71 are not stable with time: they dissociate and are degraded. The only form of a,-antichymotrypsin then obtained is an inactive, modified form produced via limited proteolysis of the inhibitor. A mechanism was proposed in view of these results [6].

Some microbial proteinases [S, 91 and snake venoms metalloproteinases [ 101 convert a,-antichymotrypsin into an inactive form by limited proteolysis. The bond cleaved by these enzymes has not been identified, thus it is impossible to know if it is localized within the reactive site recently described [I 11. As far as elastases are concerned, a,-antichymotrypsin is not able to inhibit leukocyte elastase [3, 71 but this enzyme can inactivate a,-antichymotrypsin [l 11, converting it into a

Correspondence to A. Laine, Unite INSERM No 16, Place de Verdun, F-59045 Lille Cedex, France

Ahhreviufions. SDS, sodium dodecyl sulfate; PhMeS02F, phenylmethylsulfonyl fluoride; Me2S0, dimethylsulfoxide; Suc-Ala- Ala-Pro-Phe-NH-Np, succinyl-L-alanyl-L-alanyl-prolyl-L-phenylala- nyl p-nitroanilide; s~c-(Ala)~-NH-Np, succinyl-L-alanyl-L-alanyl-L- alanyl p-nitroanilide.

Enzymes. Human pancreatic elastase 2 (EC 3.4.21.36); porcine pancreatic elastase (type 1) (EC 3.4.21.36).

proteolyzed inactive form with an apparent M , of 55000 [7]. The same phenomenon was observed with Pseudomonas aeruginosa elastase [9]. While it has been previously stated [l 1, 121 that al-antichymotrypsin is not able to inhibit pancreatic elastases, in this paper we present evidence that a,-anti- chymotrypsin is an inhibitor of porcine pancreatic elastase and of human pancreatic elastase 2, and that it forms equimolar complexes with both enzymes.

According to its specificity on synthetic substrates [13], human pancreatic elastase 2 is a chymotrypsin-like enzyme. This proteinase is also able to completely solubilize fibrous elastin [14, 151.

MATERIALS AND METHODS

Human a,-antichymotrypsin was prepared as previously described [7]; it was kept frozen in 0.01 M sodium phosphate buffer (pH 7.4) containing 0.3 M NaCl and 0.02% NaN, until use. Its concentration was determined according to [16] with a correction for the carbohydrate part of the molecule (24%).

Human pancreatic elastase 2 was prepared by the proce- dure described in [14]; it was dissolved in 0.05 M sodium phosphate buffer, pH 6.5. Its concentration was measured at 280 nm using the specific absorption coefficient of 2.02 mg- . ml cm-' [14]. Suc-Ala-Ala-Pro-Phe-NH-Np (Bachem) was used as substrate for enzymatic assays [13]. Human pancreatic elastase 2 was estimated to be 70% active by rate assays using the conditions and kinetic parameters determined in [13].

Porcine pancreatic elastase (type 1) was from Choay; it was dissolved in 0.2 M Tris/HCl buffer, pH 8.0. Its concentra- tion was determined at 280 nm using the specific absorption coefficient of 2.02 mg-' . ml . cm-' [17]. Suc(Ala),-NH-NP

328

(Choay) was used as substrate for enzymatic assays according to [18]. Porcine pancreatic elastase contained 17.4 units/mg when measured on this substrate and was thus almost 100% active, according to [19].

Enzymatic assays were performed at 25°C on an Uvikon spectrophotometer (Kontron). In order to determine the stoichiometry of inhibition for each elastase, a constant amount of enzyme was incubated with increasing amounts of 2,-antichymotrypsin at 25'C. At the end of the chosen incubation time, the residual enzymatic activity was mea- sured at 410 nm after addition of an aliquot of the appropriate substrate. We checked that both enzymes kept full activity when incubated under identical conditions. To calculate the inhibitor:enzyme molar ratios throughout the experiments, we only took into account the amount of active enzyme.

The rate constant for the association of human pancreatic elastase 2 with a,-antichymotrypsin was measured under second-order conditions by allowing equivalent concentra- tions of enzyme and inhibitor to react for various periods of time before addition of Substrdte. The association reaction was started by mixing 10 pl stock solution of enzyme (1.22 pM) into 940 p1 inhibitor solution in 0.2 M Tris/HCl buffer, pH 8.0; after a given incubation time at 25"C, the reaction was stopped by addition of 50 pl 100mM solution of Suc-Ah,-Pro-Phe-NH-Np in MezSO and the residual enzyme activity was measured by recording the linear substrate hy- drolysis at 410 nm for at least 4 min. The k,,, was determined according to the simplified equation: 1/[E] = k,,, x t + I/[E],, where [El and [Elo are free and total enzyme concentrations respectively and t is the incubation time [20]. We assumed that no dissociation of the inhibitor-enzyme complex occurred in the conditions used.

Analytical polyacrylamide gel electrophoreses were carried out on slab gels as before [4]. M , values were determined in SDS/polyacrylamide gel using the Pharmacia low-M, calibra- tion kit; they were estimated to be 59000 (instead of 58000 previously determined [6]), 25000 and 26000 for a l - antichymotrypsin, human pancreatic elastase 2 and porcine pancreatic elastase respectively.

After alkaline polyacrylamide gel electrophoresis, chymo- trypsin inhibitor was made visible as previously described [21], using N-acetyl-~~-phenylalanine-b-naphthyl ester (Sigma) as substrate.

The mixtures, in different inhibitor:enzyme molar ratios to be studied by polyacrylamide gel electrophoresis, were in- cubated at 25°C. The enzymatic action was stopped at the end of incubation time by adding a 100mM PhMeS0,F (Sigma) solution made in isopropanol to have a 100-fold molar excess of PhMeS0,F over enzyme initially present.

In all the electrophoretic analyses a mixture prepared by mixing cr,-antichymotrypsin with enzyme previously inhibited with PhMeSOzF was used as control.

RESULTS

STOICHIOMETRY OF INHIBITION

First we determined the incubation time required for completion of the reaction between inhibitor and enzyme in a mixture prepared to have about 50% residual enzyme activity. A plateau was reached within less than 1 min for the a,-antichymotrypsin/human pancreatic elastase 2 mixture while more than 30 min were required for the al-anti- chymotrypsin/porcine pancreatic elastase mixture. Thus

25 -

0 -

i \ \ , I 'i :\.-. , . , , 3 4 5 6 1 8 9 10 I n h i b i t o r / e n z y m e ( m o l a r r a t i o s )

Fig. 1. Inhibition of human pancreatic elustase 2 (A-A) and porcine puncreutic elustase (@-@) by human ccl-antich~motrypsin. Increasing amounts of inhibitor wcrc added to constant amounts of enzyme and incubated at 25'C in 0.2 M Tris/HCl buffer, pH 8.0, for 10 min with human pancreatic elastase 2 (122 nM) and 1 h with porcine pancreatic elastase (83.5 nM). Residual activity was measured at 410 nm after addition of a small aliquot ofthe appropriatc substrate (concentration of 1 mM in the assay). Stock solutions (100 mM) of Suc-Ala-Ala-Pro-Phc-NH-Np and Suc(Ala),-NH-Np were prepared in MezSO and N-methylpyrralidone respectively and stored at -20 'C

enzyme assays were performed after incubation times of 10 min for the human and 1 h for the porcine elastase.

The titration curves are plotted in Fig. 1. The extrapo- lation of the linear part of the curves leads, for a null enzymatic activity, to inhibitor:enzyme molar ratios equal to 1.25/1 and 5.50/1 for human pancreatic elastase 2 and porcine pancreatic elastase respectively.

Complete elimination of enzymatic activity can be achieved even for porcine pancreatic elastase but in this case a very high inhibitor: enzyme molar ratio is required: 7.5/1.

ASSOCIATION RATE CONSTANT FOR HUMAN PANCREATIC ELASTASE 2

Inhibition of human pancreatic elastase 2 by %,-anti- chymotrypsin was too fast to obtain a time dependence in the conditions used for the determination of the stoichiometry of the reaction.

However a time-dependent inhibition could be obtained by lowering the enzyme and inhibitor concentrations and using a higher substrate concentration in order to measure accurate hydrolysis rates. In the experimental conditions used a k,,, value of 8.9 & 1.3 x lo5 M-'s-' was determined.

N o attempt was made to determine the k,,, value for the inhibition of porcine pancreatic elastase, since this reaction was not of physiological interest.

ELECTROPHORETIC ANALYSES OF THE INHIBITOR-ENZYME REACTIONS

A study, similar to the one previously carried out with cathepsin G [5], was performed with both elastases. A qualitative characterization of the reaction products was done by analyzing in both polyacrylamide gel systems (see Mat- erials and Methods) five mixtures in different in- hibitor:enzyme molar ratios.

329

Fig. 2. Electrophoretic study qf the interaction of'cc, -untichymotrypsin with human puncreutic elustuse 2 by SDS/polyucrylumide electrophorrsi.~ on (I 5-30% gradient gel. Only sample M was reduced. M, M , markers: phosphorylase b (94000), bovine albumin (67000), ovalbumin (43000). carbonic anhydrase (30000), trypsin inhibitor (20100) and cc-lactalbumin (14400). C, control mixture: ccl-antichymotrypsin + inactivated human pancrcatic elastase 2 (molar ratio = 3.6/1); E, human pancrealic elaslase 2; 1, al-antichymotrypsin; 1 - 5, mixturcs in different inhibitor: enzyme molar ratios, incubated a t 25°C for 10 min and added with PhMeS02F; molar ratios: 1, 0.45/1; 2, 0.9/1; 3, 1.8/1; 4, 3.6/1; 5 , 7.2/1. Thc arrow indicatcs the mobility of the peptide of M , near 5000. (A) The initial amount of enzyme is kept constant (0.7 pg). (B) Thc initial amount of inhibitor is kept constant (1.5 pg). The faint band remaining in slots 3, 4 and 5 in the enzyme position corresponds to the inactive molecules of the enzymc prcparation. The minor band of M , near 14000, present in enzyme-containing samples, is probably duc to thc degradation products obtained during the treatment of the enzyme (even previously inactivated) before gel electrophoresis. When the inhibitor concentration is very high (slots 5 ) some aggrcgatcs appear

CI -Antich-~~~otr.vpsin/human pancreatic elastase 2 mixtures

Mixtures in the following inhibitor: enzyme molar ratios were prepared: 0.4511; 0.9jl; 1.8jl; 3.6/1; 7.211. The concentration of enzyme was kept constant in the different mixtures by adding 0.01 M sodium phosphate buffer (pH 7.4) containing 0.3 M NaCl and 0.02% NaN, to have the same final volume in each mixture. Thus the conditions were close to those employed for titration curves, except for final concen- trations: the enzyme concentrations was 12 times higher than in spectrophotometric assays because of the protein amounts required for electrophoretical studies. Incubation was per- formed for 10 min. Fig. 2 shows the patterns obtained when the mixtures were analyzed by SDS/polyacrylamide gel electrophoresis. In Fig. 2A identical volumes of the different mixtures were analyzed, thus the initial amount of enzyme was constant (0.7 pg) in all the lanes. On the other hand, in Fig. 2 B the sample volumes were calculated to have the same initial amount of inhibitor (1.5 pg).

In all the mixtures with active enzyme a component having a M , near 77000 is observed. According to its M , it corre- sponds to an equimolar inhibitor-enzyme complex. Under reducing conditions this complex has the same apparent M , (data not shown). The complex amount in each mixture depends on the inhibitor:enzyme molar ratio. In Fig. 2A the limiting factor for the amount of complex is the enzyme concentration while in Fig. 2 B the limiting factor is the in- hibitor concentration. Complex formation is accompanied by the appearance of a fragment migrating as a peptide of M , near 5000, probably due to the limited proteolysis of the inhibitor during the complex formation (not easily visible on the photograph).

In the presence of an excess of active enzyme (slots 1 and 2) a faint band (of M , near 55000), migrating slightly faster than native a,-antichymotrypsin, is observed; it is due to free modified inhibitor, which is well visualized in an alkaline polyacrylamide gel system. Under these conditions (Fig. 3 A) this form (in slots 1 and 2) clearly migrates faster than native inhibitor (slot I) and has no inhibitory activity left upon chymotrypsin (Fig. 3B, slots 1 and 2); the active inhibitor in excess is easily visualized in slots 3, 4 and 5 (Fig. 3 B). When the two forms of a,-antichymotrypsin, active and inactive, coexist (slots 3 , 4 and 5), they are well resolved in the alkaline gel (Fig. 3) while they appear under only one diffuse band in SDS/polyacrylamide gel electrophoresis (Fig. 2).

In the alkaline polyacrylamide gel system (Fig. 3) the equi- molar complex is well visualized; it has a p mobility when compared with acontrol serum (not shown) and it is the major component in slots 1, 2 and 3.

a -Antichymotrypsin/porcine pancreatic elastase mixtures

Mixtures in the following inhibitor: enzyme molar ratios were prepared: 0.611 ; 1.211 ; 2.411 ; 4.811 ; 7.911. Incubation was performed for 1 h. Fig. 4 shows the results obtained in both polyacrylamide gel systems. Sample volumes were calculated to have a constant initial amount of al-anti- chymotrypsin in all samples analyzed. SDS/polyacrylamide gel electrophoresis patterns are presented in Fig. 4A. A band with a M , near 77000 is visible in all the mixtures with active enzyme. It is an equimolar complex. Concurrently a fragment migrating as a peptide of M , near 5000 is visualized. In alkaline polyacrylamide gel electrophoresis (Fig. 4 B) the major prod- uct is a modified form of a,-antichymotrypsin, which has no

330

Fig. 3. Interaction njccl-untichymotrypsin with human pancreatic elastase 2 studied by alkaline electrophoresis on 10% polyucrjlumiik~ gel at p H 8.3. (A) Protein staining with Coomassie brilliant blue G250. (B) Chymotrypsin inhibitory activity. I , C, 1-5 are as in Fig. 2. x,Achy, x l - antichymotrypsin. As in Fig. 2B, the amount of a,-antichymotrypsin is constant in all the samples analyzed (1.5 pg)

inhibitory activity left. Just a faint band of complex is visible in each mixture. It migrates rather more slowly than the a,-antichymotrypsin/human pancreatic elastase 2 complex seen in Fig. 3 ; this might be due to a difference between the PI values of each enzyme or to diffcrent charges masked or free in each complex.

The reaction of El-antichymotrypsin with porcine pan- creatic elastase was followed for more than 48 h: there was no recovery of enzyme activity and the complex was stable with time (data not shown).

DlSCUSSION Spectrophotometric studies brought evidence that a1 -anti-

chymotrypsin was able to completely inactivate human pancreatic elastase 2 and porcine pancreatic elastase. How- ever, very different inhibitor: enzyme molar ratios (1.25/1 and 5.5/1) were respectively obtained by extrapolation of the in- hibition curves. The same inhibitor preparation, assayed against human leukocyte cathepsin G [7] and bovine chymo- trypsin [6], led to molar ratios of 1.76/1 and 1.46/1 respectively. Thus, when inhibitor quantities required for inhibition of each enzyme are compared on a molar basis, a,-antichymotrypsin seems to be a better inhibitor of human pancreatic elastase 2 than of human leukocyte cathepsin G or bovine chymotrypsin. It does not appear to be a strong inhibitor of porcine pancreatic elastase because more than seven molecules of al-antichymotrypsin are required to completely inhibit one molecule of this enzyme.

The patterns observed in gel electrophoretic analyses clearly show that al-antichymotrypsin is able to interact with both elastases and to form with each enzyme an equimolar complex. Although in the reaction of al-antichymotrypsin with human pancreatic elastase 2 the major product obtained corresponds to an equimolar complex, there is on the contrary a noticeable and extensive production of free, modified in- active a,-antichymotrypsin during the reaction with porcine pancreatic elastase. It has been previously shown that ccl-antichymotrypsin was inactivated by porcine pancreatic

elastase by cleavage between the Ps and P4 residues of its reactive site thus behaving as a substrate in the presence of that enzyme [I I]. Our electrophoretic studies partially agree with this substrate-like behaviour but clearly demonstrate that a,-antichymotrypsin also interacts with porcine elastase by forming an equimolar complex; furthermore, a time-course study indicated that this complex was stable and that no release of active enzyme occurred even after 48 h of incubation at 25 “C. This behaviour of a,-antichymotrypsin certainly pro- vides an explanation for the high inhibitor: enzyme molar ratio needed to obtain full inhibition of porcine elastase.

As demonstrated by the present work, the inhibitory role of human a,-antichymotrypsin is not as restricted as it has been previously claimed [ l l , 121. It was known to control the activity of chymotrypsin-like proteinases from phagocytic cells, including cathepsin G and chymase, but hitherto nobody assigned to it any inhibitory function on elastase-type enzymes, even on human pancreatic elastase [22]. Porcine pancreatic elastase is known to induce peptide bond cleavage on the carboxy-terminal side of neutral aliphatic residues such as Ala, Leu, Gly, Val and Ile with a preferential reactivity on Ala-Ala bonds. Splitting by human pancreatic elastase 2, which is a chymotrypsin-like enzyme, is preferably directed towards Leu, Tyr, Phe and Met residues [13]. Although quite different in their primary specificities, both enzymes can in- teract with a,-antichymotrypsin in the area of its reactive site, which contains a leucine as P1 residue [ll], and therefore are able to form complexes with the inhibitor. The reactions we describe here differ from those of a,-antichymotrypsin with two other elastases: Pseudornonas aevuginosa elastase [9] and leukocyte elastase [7, 111; these enzymes are not inhibited by a,-antichymotrypsin but convert this inhibitor into an inactive form after limited proteolysis. This form is electrophoretically highly similar to, or identical with that produced by bovine chymotrypsin [6], human leukocyte cathepsin G [5] and by the two pancreatic elastases studied in this paper. In- vestigations are in progress to determine if the limited pro- teolysis occurs at the same site with all these enzymes. The results might allow us to elucidate the mechanisms of all these reactions.

331

Fig. 4. Analyses bj> polyacrylamide gel electrophoresis of difjli.rent tnolur ratios of' ci -antich,ymotrypsin and porcine pancreatic elastuse incubated for 1 h at 25.C befbre PhMeS02F addition. E, Porcine paiicrcatic elastase; I, a,-antichymotrypsin; C, Control mixture: xl-antichymotrypsin + inactivated porcine pancreatic elastase ([I]/[E] = 2.4/1); 1-5, different [I]/[E] molar ratios: 1, 0.6/1; 2, 1.2/1; 3, 2.4/1; 4, 4.8/1; 5, 7.9/1. (A) SDS/polyacrylamide electrophoresis on a 5 - 30% gradient gel under non-reducing conditions. The arrow indicates the mobility of the peptide of M , near 5000. (B) Alkaline electrophoresis on 10% polyacrylamide gel at pH 8.3. al-Achy, al-antichymotrypsin. The amount of al-antichymotrypsin is constant in all the samplcs analyzed (1 .5 pg)

The role of elastase as the key enzyme responsible for the destruction of vessel walls, elastolysis, thrombosis and interstitial haemorrhage has been demonstrated by studying acute experimental pancreatitis in dogs [23] . In normal human plasina all of the detectable immunoreactive elastase was shown to be bound to a,-proteinase inhibitor [24]. According to the respective molar concentrations of a,-antichymotrypsin and a,-proteinase inhibitor in normal plasma, the latter in- hibitor must play the major role in the inhibition of human pancreatic elastase 2. Under a t least two pathological

conditions, a,-proteinase inhibitor deficiency and inflam- matory process, the ability of human a,-antichymotrypsin to strongly inhibit human pancreatic elastase 2 has to be taken into account according to the following explanation. From the k,,, value determined in this paper [S.9 x l o5 M - ' s - ' ] and assuming a concentration of a,-antichymotrypsin of 0.02 mM in inflammatory plasma, a delay time of inhibition was calculated according to [25]; the value obtained, 0.28 s, in- dicates that human pancreatic elastase 2, which is massively liberated into blood during pancreatitis for example, can be inhibited very quickly by a,-antichymotrypsin. Therefore, our findings allow us to support the hypothesis that beside a,- proteinase inhibitor and a2-macroglobulin, a,-anti- chymotrypsin may contribute to the regulatory balance be- tween plasma inhibitors and human pancreatic elastase 2 in the diseases that involve this enzyme.

The skilful assistance of Marie-Paule Ducourouble is gratefully acknowledged. This work is supported by Znstitut National de la Sante Pt d~ la Rrchurclir Mtdicale (CRL no. 825044) and by la CommunautP Eirropimne du Charlion et de l'dcier (contract 724833017)-

REFERENCES 1.

2.

3.

4.

5.

6.

7.

8. 9.

10.

11 . 12. 13.

14.

15.

16.

17. 18.

19. 20.

21.

22.

23.

24.

25.

Schultze, H. E., Heide, K. & Haupt, H . (1962) Naturwissenschuf ien 49. 133.

Aronsen, K. F., Ekelund, G., Kindmark, C. 0. & Laurell, C. B. (1972) Scand. J . Clin. Lab. Invest. 29, (suppl. 124), 127-136.

Travis. J., Bowen, J . & Baugh, R. (1978) Biochemistry 17, 5651 - 5656.

Laine, A,, Davril, M., Hayem, A. & Loucheux-Lefebvre, h4. H. (1982) Biochem. Biophys. Res. Commun. 107, 337- 344.

Laine, A., Davril, M. & Hayem, A. (1982) Biochem. Biophys. Res. Cotntnun. 105. 186-193.

Laine, A,, Davril, M. & Hayem, A. (1984) Eur. J . Biochem. 140, 105 - 11 1.

Lahe, A,, Hayem, A. & Davril, M. (1984) in Marker profeins in inflammation (Arnaud, P., Bienvenu, J. & Laurent, P., eds) vol. 2, pp. 171 - 179. Walter de Gruyter, Berlin, New York.

Kress, L. F. (1983) Acta Biochim. Pol. 30, 159-164. Catanese. J . & Kress, L. F. (1984) Biochirn. Biophys. Acra 789,

Krcss. L. F. & Hufnagel, M. E. (1984) Comp. Biochem. Physiol.

Morii, M. &Travis. J. (1983) J . Biol. Chem. 258, 12749 - 12752. Travis, J . & Morii, M. (1980) Methods Enzymol. 80, 765-771. Del Mar, E. G., Largman, C., Brodrick, J . W., Fassett, M. &

Largman, C., Brodrick, J . W. & Geokas, M. C. (1976) Biorhemis-

Rabaud, M., Dabadie, P., Lefebvre, F., Desgranges, C. &

Lowry, 0. H., Rosebrough, N. J., Farr, A. L. & Randall, R. J.

Sholton, D. M. (1970) Method~y Enzymol. l Y , 113-140. Bieth, J., Spiess, B. & Wermuth, C. (1974) Biochewi. Med. 11,

Bicth, J . (1978) Front. Matrix Bid . 6, 1-82. Beatty, K., Bieth, J . &Travis, J. (1980) 1. Biol. Chem. 255, 3931 -

Laine, A. & Hayem, A. (1981) Biochim. Biophys. Acta 668,429-

Gustavsson, E. L., Ohlsson, K. & Olsson, A. S. (1980) Hoppe

Gcokas, M. C., Rinderknecht, H., Swanson, V. & Haverback, B.

Largman, C., Brodrick, J . W., Geokas, M. C., Johnson, J . H. &

Bieth, J . G. (1980) Clin. Resp. Physiol. 16 (suppl.), 183-195.

37-43.

77B, 431 -436.

Geokas, M. C. (1980) Biochemistry 19,468-472.

t r j ' 15, 2491 -2500.

Bricaud, H . (1 984) Connect. Tissue Res. 12, 165 - 174.

(1951) J . Biol. Chem. 193, 265-275.

350 - 357.

3934.

438.

Seyler '.s Z. Physiol. Chem. 361, 169 - 176.

J. (1968) Lab. Invest. 19, 235-239.

Fassett, M. (1980) Am. J . Physiol. 238, G177-GlY2.