Evidence for lipid-associated serine proteases and metalloproteases in human bronchoalveolar lavage fluid

The multiple facets of alpha-1-antitrypsin

Alpha-1-antitrypsin (AAT) is recognised as a potent inhibitor of serine proteinases. Genetic deficiency is associated with several neutrophilic diseases including severe emphysema. This is believed to reflect the loss of inhibition of neutrophilic serine proteinases that then result in local tissue damage (the proteinase/antiprotease hypothesis). In recent years the role of AAT in the control of inflammatory and immunological processes has become identified. Although in some instances this may reflect its ability to control pro-inflammatory effects of serine proteinases it has also become recognised that it has non proteolytic mediated functions. This poly functional role is starting to become recognised offering a possibility of its use as a therapeutic agent in many clinical disease areas. The current review explores both the traditional and non-traditional function of AAT through published literature.

A quantitative structure-activity relationship study on matrix metalloproteinase inhibitors: Piperidine sulfonamide aryl hydroxamic acid analogs

A quantitative structure-activity relationship (QSAR) study has been made on a series of piperidine sulfonamide aryl hydroxamic acid analogs acting as matrix metalloproteinase (MMP) inhibitors. The inhibitory potencies of the compounds against two MMPs, MMP-2 and MMP-13, are found to be significantly correlated with the hydrophobic properties of the molecules, suggesting that in both enzymes the hydrophobic interaction is playing a dominant role.

Inhalation of organic dusts and lipopolysaccharide increases gelatinolytic matrix metalloproteinases (MMPs) in the lungs of heaves horses

We report the effects of mouldy hay/straw exposure, inhaled hay dust suspension (HDS) and inhaled lipopolysaccharide (LPS) on bronchoalveolar lavage fluid (BALF) gelatinolytic matrix metalloproteinase (MMP) levels and degree of activation in healthy (n = 6) and heaves- (previously termed chronic obstructive pulmonary disease) affected (n = 6 or 7) horses. Gelatinolytic MMPs in BALF were quantified by zymography, and gelatinases were shown by Western immunoblotting to be MMP-2 and MMP-9. Hay/straw and HDS challenges increased BALF total gelatinolytic activity only in heaves horses, with the majority of gelatinolytic activity comprising pro- and active MMP-9. The 5 h duration hay/straw challenge increased BALF gelatinolytic MMP activity in heaves horses at 5 and 24 h after the start of this challenge, with activity returning to baseline by Day 4. In contrast to hay/straw and HDS challenges, LPS inhalation increased BALF gelatinolytic MMP activity in both groups. For all challenges, absolute BALF neutrophil counts were highly significantly correlated (P<0.0001) with levels of proMMP-9 and active MMP-9, but not with levels of MMP-2 (P>0.05). As gelatinolytic MMPs are pro-inflammatory agents, they may contribute to lung dysfunction and tissue destruction in heaves horses exposed to airborne organic stable dusts.

MMP-9 as a marker of inflammation in tracheal epithelial lining fluid (TELF) and in bronchoalveolar fluid (BALF) of COPD horses

Gelatinolytic activity was analysed to study whether elevated activity previously found at the tracheal level of the respiratory tract of horses with chronic obstructive pulmonary disease (COPD) could also be found at the lower part of the respiratory tract. Furthermore, presence and significance of the gelatinolytic matrix metalloproteinases (MMPs) MMP-2 and MMP-9 in respiratory secretions of healthy and COPD horses were determined. Elevated gelatinolytic matrix metalloproteinases were detected in bronchoalveolar and tracheobronchial secretions from COPD horses. The main pathologically elevated MMP was characterised to be MMP-9. Significantly increased MMP-9 activities as measured by gelatin zymography and Western blotting were found in all the respiratory samples from COPD horses compared to healthy horses. Elevation of active MMP-9 paralleled with increased gelatinase-associated lipocalin levels. Bronchoalveolar lavage fluid (BALF) epithelial cells, macrophages, neutrophils and lymphocytes expressed MMP-9 immunoreactivity demonstrated by immunocytochemistry and MMP-9 mRNA was expressed by bronchial epithelial cells of lung tissue section shown by in situ hybridisation. MMP-2 seems not to play a major role in chronic lung inflammation. No clear differences in MMP-2 or MMP-14 (a potent MMP-2 activator) levels were found when comparing the samples from COPD or healthy horses. These results suggests that MMP-9 could serve as a potential diagnostic marker for the active ongoing tissue remodelling in the acute phase of equine COPD. Increased gelatinolytic activity could be found at both tested respiratory tract levels. Therefore, tracheal epithelial lining fluid (TELF) samples can usefully serve as diagnostic material for detection of increased levels of the main gelatinolytic MMP, MMP-9, representing the entire diseased lung.

Concentrations of elastinolytic metalloproteinases in respiratory tract secretions of healthy horses and horses with chronic obstructive pulmonary disease

OBJECTIVES

To determine whether samples of tracheal epithelial lining fluid (TELF) obtained from horses have elastinolytic activity characteristic of metalloproteinases, to compare elastinolytic activity in TELF obtained from healthy horses and horses with chronic obstructive pulmonary disease (COPD), and to determine whether chemically modified tetracycline-3 (CMT-3) inhibits elastinolytic activity in TELF ANIMALS: 10 horses with COPD and 10 healthy control horses.

PROCEDURE

Zymography and fluorometry were used to measure elastinolytic activity, and EDTA was used to inhibit elastinolytic activity and verify that the activity was attributable to metalloproteinases. Possible inhibition of elastinolytic activity with CMT-3 was studied in vitro.

RESULTS

Elastinolytic activity was found in TELF obtained from all horses, and this activity was significantly higher in TELF obtained from horses with COPD than in TELF obtained from healthy horses. For all samples, EDTA and CMT-3 inhibited elastinolytic activity.

CONCLUSIONS AND CLINICAL RELEVANCE

Elastinolytic activity is detectable in TELF obtained from horses and seems to be attributable to metalloproteinases. Elastinolytic activity in TELF is significantly inhibited by CMT-3. Elastinolytic activity in TELF can be detected by means of zymography or fluorometry. Increased elastinolytic activity may reflect destruction of pulmonary tissue in horses with COPD. Chemically modified tetracyclines such as CMT-3 may provide an additional treatment possibility for horses with COPD.

Role of neutrophils and alpha1-antitrypsin in coal- and silica-induced connective tissue breakdown

Mineral dusts produce emphysema, and administration of dust to rats results in the rapid appearance of desmosine and hydroxyproline in lavage fluid, confirming that dusts directly induce connective tissue breakdown. To examine the role of neutrophils and alpha1-antitrypsin (alpha1-AT) in this process, we instilled silica or coal into normal rats or rats that had been pretreated with antiserum against neutrophils. One day after dust exposure, lavage fluid neutrophils and desmosine and hydroxyproline levels were all elevated; treatment with antiserum against neutrophils reduced neutrophils by 75%, desmosine by 40-50%, and hydroxyproline by 25%. By 7 days, lavage fluid neutrophils and desmosine level had decreased, whereas macrophages and hydroxyproline level had increased. By ELISA analysis, lavage fluid alpha1-AT levels were increased four- to eightfold at both times. On Western blot, some of the alpha1-AT appeared as degraded fragments, and by HPLC analysis, 5-10% of the methionine residues were oxidized. At both times, lavage fluid exhibited considerably elevated serine elastase inhibitory capacity and also showed elevations in metalloelastase activity. We conclude that, in this model, connective tissue breakdown is initially driven largely by neutrophil-derived proteases and that markedly elevated levels of functional alpha1-AT do not prevent breakdown, thus providing in vivo support for the concept of quantum proteolysis proposed by Liou and Campbell (T. G. Liou and E. J. Campbell. Biochemistry 34: 16171-16177, 1995). Macrophage-derived proteases may be of increasing importance over time, especially in coal-treated animals.

Human Neutrophil Elastase Activates Human Factor V but Inactivates Thrombin-Activated Human Factor V

The effect of human neutrophil elastase (HNE) on human factor V (F.V) or α-thrombin–activated human factor V (F.Va) was studied in vitro by prothrombinase assays, sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), and NH2 -terminal sequence analysis. Incubation of F.V (600 nmol/L) with HNE (2 nmol/L) in the presence of Ca2+ resulted in a time-dependent increase in its cofactor activity. In contrast, treatment of F.Va (600 nmol/L) with HNE (60 nmol/L) in the presence of Ca2+ resulted only in a time-dependent decrease in its cofactor activity. Under the conditions of these experiments, the maximum extent of F.V activation accomplished by incubation with HNE was approximately 65% to 70% of that observed with α-thrombin in presence of Ca2+. The extent of both the HNE-dependent enhancement in F.V cofactor activity and the HNE-dependent decrease in F.Va cofactor activity was not influenced by the addition of phosphatidylcholine/phosphatidylserine (PCPS) vesicles (50 μmol/L). The HNE-derived cleavage products of F.V, which correlated with increased cofactor activity, as demonstrated by SDS-PAGE under reducing conditions, were different from those generated using α-thrombin. Treatment of F.V (600 nmol/L) with HNE (2 nmol/L) in the presence of Ca2+ resulted in the production of three closely spaced doublets of: 99/97, 89/87, and 76/74 kD whose appearance over time correlated well with the increased cofactor activity as judged by densitometry. Treatment of F.Va (600 nmol/L) with HNE (60 nmol/L) in the presence of Ca2+ resulted in the cleavage of both the 96 kD heavy chain and the 74/72 kD light chain into products of: 56, 53, 35, 28, 22, and 12 kD. Although densitometry indicated that both the heavy and light chains of F.Va were hydrolyzed by HNE, cleavage of the 96 kD heavy chain was more extensive during the time period (10 to 30 minutes) of the greatest loss of F.Va cofactor activity. NH2 -terminal sequence analysis of F.V treated with HNE indicated cleavage at Ile819 and Ile1484 under conditions during which the procofactor expressed enhanced cofactor activity in the prothrombinase complex. NH2 -terminal sequence analysis of F.Va treated with HNE indicated cleavage at Ala341, Ile508, and Thr1767 under conditions, which the cofactor became inactivated, as measured by prothrombinase activity. The activation and inactivation cleavage sites are close to those cleaved by the physiological activator and inactivator of F.V and F.Va, namely α-thrombin (Arg709 and Arg1545) and Activated Protein C (APC) (Arg306 and Arg506), respectively. These results indicate that HNE can generate proteolytic products of F.V, which initially express significantly enhanced procoagulant cofactor activity similar to that observed following activation with α-thrombin. In contrast, HNE treatment of F.Va resulted only in the loss of its cofactor activity, but again, this is similar to that observed following inactivation by APC.

Human Neutrophil Elastase Activates Human Factor V but Inactivates Thrombin-Activated Human Factor V

The effect of human neutrophil elastase (HNE) on human factor V (F.V) or α-thrombin–activated human factor V (F.Va) was studied in vitro by prothrombinase assays, sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), and NH2 -terminal sequence analysis. Incubation of F.V (600 nmol/L) with HNE (2 nmol/L) in the presence of Ca2+ resulted in a time-dependent increase in its cofactor activity. In contrast, treatment of F.Va (600 nmol/L) with HNE (60 nmol/L) in the presence of Ca2+ resulted only in a time-dependent decrease in its cofactor activity. Under the conditions of these experiments, the maximum extent of F.V activation accomplished by incubation with HNE was approximately 65% to 70% of that observed with α-thrombin in presence of Ca2+. The extent of both the HNE-dependent enhancement in F.V cofactor activity and the HNE-dependent decrease in F.Va cofactor activity was not influenced by the addition of phosphatidylcholine/phosphatidylserine (PCPS) vesicles (50 μmol/L). The HNE-derived cleavage products of F.V, which correlated with increased cofactor activity, as demonstrated by SDS-PAGE under reducing conditions, were different from those generated using α-thrombin. Treatment of F.V (600 nmol/L) with HNE (2 nmol/L) in the presence of Ca2+ resulted in the production of three closely spaced doublets of: 99/97, 89/87, and 76/74 kD whose appearance over time correlated well with the increased cofactor activity as judged by densitometry. Treatment of F.Va (600 nmol/L) with HNE (60 nmol/L) in the presence of Ca2+ resulted in the cleavage of both the 96 kD heavy chain and the 74/72 kD light chain into products of: 56, 53, 35, 28, 22, and 12 kD. Although densitometry indicated that both the heavy and light chains of F.Va were hydrolyzed by HNE, cleavage of the 96 kD heavy chain was more extensive during the time period (10 to 30 minutes) of the greatest loss of F.Va cofactor activity. NH2 -terminal sequence analysis of F.V treated with HNE indicated cleavage at Ile819 and Ile1484 under conditions during which the procofactor expressed enhanced cofactor activity in the prothrombinase complex. NH2 -terminal sequence analysis of F.Va treated with HNE indicated cleavage at Ala341, Ile508, and Thr1767 under conditions, which the cofactor became inactivated, as measured by prothrombinase activity. The activation and inactivation cleavage sites are close to those cleaved by the physiological activator and inactivator of F.V and F.Va, namely α-thrombin (Arg709 and Arg1545) and Activated Protein C (APC) (Arg306 and Arg506), respectively. These results indicate that HNE can generate proteolytic products of F.V, which initially express significantly enhanced procoagulant cofactor activity similar to that observed following activation with α-thrombin. In contrast, HNE treatment of F.Va resulted only in the loss of its cofactor activity, but again, this is similar to that observed following inactivation by APC.

Proteinase/proteinase inhibitor imbalance in sputum sol phases from patients with chronic obstructive pulmonary disease. Suggestions for a key role played by antileukoprotease

In order to characterize the imbalance between proteinases and proteinase inhibitors in sputum sol phases, we studied 25 patients (mean age, 59 +/- 11 yr) with exacerbated chronic obstructive pulmonary disease (COPD). An aliquot of sputum was used for bacteriologic determinations, and the remainder was centrifuged in order to obtain gel and sol phases. On the basis of the bacteriologic data, patients were divided into colonized patients (14) and noncolonized patients (11). All of the major inhibitors were immunologically detectable in sol phases without a significant difference between colonized and noncolonized patients (alpha 1-proteinase inhibitor [alpha 1-PI], 2.56 microM +/- 0.53 microM and 2.39 microM +/- 0.72 microM; alpha 2-macroglobulin [alpha 2-MG], 0.21 microM +/- 0.07 microM and 0.16 microM +/- 0.05 microM; antileukoprotease (ALP), 1.78 microM +/- 0.57 microM and 1.53 microM +/- 0.6 microM, respectively [mean +/- SE]). With regard to proteinase activities, both free elastase-like and free chymotrypsin-like activities were detectable in the majority of patients (15/25) (0.59 microM +/- 0.15 microM and 0.74 microM +/- 0.15 microM for elastase-like activity [ELA], and 0.010 microM +/- 0.003 microM and 0.017 microM +/- 0.007 microM for chymotrypsin-like activity [CLA], respectively [mean +/- SE]). The inhibitory profile of proteinase activities, performed by means of a panel of inhibitors, allowed us to assign specific activities mainly to neutrophil elastase and cathepsin G (Cat G). Next we looked at the relationships between inhibitors and proteinase activities. We found a significant negative correlation between neutrophil elastase activity and ALP (r = -0.58; p < 0.01). In confirmation of this suggestion, sol phases were divided into samples (15) with detectable ELA (> 0.50 microM) and samples (10) with no detectable ELA (< 0.18 microM). Levels of alpha 1-PI and alpha 2-MG did not differ significantly between the two groups, whereas ALP values were higher in the group with no detectable ELA (3.12 microM +/- 0.69 microM) than in the other group (0.58 microM +/- 0.21 microM; p < 0.001). We conclude that most sputum sol phases from patients with exacerbated COPD have a high burden of free neutrophil elastase and Cat G. Antileukoprotease seems to be the major naturally occurring inhibitor effective in the modulation of proteinase activities in bronchial secretions under these conditions.

Alterations in cysteine proteinase content of rat lung associated with development of Pneumocystis carinii infection

The rate of hydrolysis of three cysteine-type proteinase substrates, N-benzyloxycarbonyl-Arg-Arg-4-methyl-7-coumarylamide (AMC) (cathepsin B), Arg-AMC (cathepsin H), and N-benzyloxycarbonyl-Phe-Arg-AMC (cathepsin L), were determined in rat lung throughout the time course of the induction of Pneumocystis carinii infection by immunosuppression. Cathepsin B-like and cathepsin L-like activities fell below control values initially, but from week 8 of the immunosuppressive treatment significant increases above the control were noted. Cathepsin H-like activity was greater than control levels from week 3, and by week 12 it was 7,600% of the mean control value. When compared with the relative degree of infection, as assessed from the number of cysts present in lung impression smears, cathepsin B-like and cathepsin L-like activities were significantly increased only at heavy parasite burdens while cathepsin H-like activity displayed a close correlation with parasite number (r = 0.884; P less than 0.001). Activity was detected in lysates of purified P. carinii with all three substrates. Treatment of heavily infected animals with co-trimoxazole cleared the lungs of P. carinii, and this was accompanied by a marked reduction in proteinase activity, in particular, cathepsin H-like activity, which fell from 108- to 3-fold the mean control value following drug treatment. Analysis of cathepsin H isozyme patterns by fluorography following isoelectric focusing revealed differences between treated and control lung samples. In the immunosuppressed group, there was a time-dependent increase in the intensity of some of the bands observed in the controls and an appearance of several novel bands which corresponded to bands observed in lysates of P. carinii. It is likely, therefore, that the increased proteinase activity observed in the treated group is due, at least in part, to isozymes from P. carinii; consequently, cathepsin H-like activity might be of use diagnostically in the identification of P. carinii infection and in the estimation of parasite burden.

Plasma leucocyte elastase concentrations in smokers

The associations between cigarette smoking, plasma leucocyte elastase concentration, peripheral leucocyte count and FEV1 were examined in 148 men, 72 of whom were current cigarette smokers, 40 of whom were ex-smokers, and 36 who had never smoked. All men were part of a long-term survey. Smokers had significantly higher plasma leucocyte elastase concentrations than ex-smokers or those who had never smoked. Mean current FEV1 was lower, and the annual decline in FEV1 in the preceding 10 years was faster in smokers than the other two groups. A few smokers had slight increases in serum C-reactive protein concentrations. Although peripheral blood leucocyte counts were higher in smokers than in non-smokers or ex-smokers, no association was found in any of the three groups of men between plasma elastase concentration and peripheral leucocyte count, nor between either of these two variables and annual decline in FEV1 or current level of FEV1. There was also no relation between plasma elastase concentration and reported daily cigarette consumption or mixed expired carbon monoxide in smokers. The results indicate that some male smokers have increased in vivo release of elastase from peripheral blood neutrophils at a time when there is no evidence of acute infection. Because leucocyte elastase is a strong candidate for pulmonary tissue damage, further studies of the mechanisms that increase plasma concentrations are indicated.

Serine proteinase requirement for the extra-cellular metabolism of pulmonary surfactant

Pulmonary surfactant as lavaged from the alveoli exists in at least three structural subtypes, lamellar body-like, tubular myelin and vesicular forms that can be separated on the basis of their buoyant densities. Previous studies have suggested that surfactant is secreted in the lamellar body form and metabolized through the other subtypes in sequence. This metabolic sequence can be reproduced in vitro by cyclic expansion and contraction ('cycling') of the surface area of nascent surfactant at 38 degrees C. Cycling of nascent secretion, which is predominantly of lamellar body-like buoyant density, rapidly converted it to the buoyant density of tubular myelin and then to that of the vesicular subtype. We examined the role of proteinases in the conversion of nascent surfactant subtypes in vitro. Addition of metallo-, cysteine- and acid-proteinase inhibitors to the cycling mix did not inhibit the conversion of tubular myelin to vesicular subtype. However, a variety of serine proteinase inhibitors inhibited the formation of vesicular subtype. Their inhibitory effect was dose-related and most marked for alpha 1-antitrypsin where a concentration equal to that found in the alveolar fluid lining layer resulted in 50% inhibition of the generation of light subtype, suggesting physiological relevance. The enzyme(s) responsible for promoting the generation of light subtype was sedimentable and therefore presumably in particulate form. By differential centrifugation of lung secretions it was separable from alveolar macrophages and partially separable from surfactant itself. It has not been identified, nor has its substrate. We conclude that in vitro cycling provides a model for the study of alveolar surfactant metabolism and that the conversion of tubular myelin to vesicular forms of surfactant requires serine proteinase activity.