Comparison of Proteolytic Enzyme Activity in Pulmonary Alveolar Macrophages and Blood Leukocytes in Smokers and Nonsmokers'-
3
JAMES 0. HARRIS, GERALD N. OLSEN, JAMES R. CASTLE, and ANNE S. MALONEY
SUMMARY __________________________________________________________ Proteolysis (or more specifically, clastolysis) of the lung may be involved in the pathogenesis of pulmonary emphysema. To investigate the human alveolar macrophage as a potential mediator of lung damage, elastase-like esterase and protease activity was determined in these cells as well as in alveolar lavage fluid and in peripheral blood leukocytes. Bronchoalveolar lavage was used to obtain alveolar cells and fluid in normal volunteers who were divided into two groups according to cigarette smoking history, nonsmokers and smokers. Results of these studies revealed that human alveolar macrophages possess a high activity of both elastase-like esterase and protease. Furthermore, the alveolar macrophages of cigarette smokers had a significantly greater elastase-like esterase and protease activity than those of nonsmokers. \Vhen the 4- to 5-fold increase in the number of macrophages found in cigarette smokers is taken into account there was approximately 10 times more elastase-like esterase activity and 18 times more protease activity within macrophages in the alveolar spaces of cigarette smokers' lungs. This makes the alveolar mauophage a potent potential source of proteolytic enzymes in man.
Introduction In the wake of Laurell and Eriksson's (I) discovery of a 1 -antitrypsin deficiency and the association of this deficiency with emphysema, intense interest has developed in proteolysis as a pathogenic mechanism in chronic obstructive lung disease. Experimentally, it has been demonstrated that papain (2, 3), elastase (3, 4), leu-
(Received in original form August 22, 1974 and in revised form ]anuary 15, 1975) 1 From the Veterans Administration Hospital and the Department of Medicine, College of Medicine, University of Florida, Gainesville, Florida. 2 This research was supported in part by N"ational Institutes of Health Research Grant ES00789 and by a grant from the Florida Lung Association. 3 Requests for reprints should be addressed to James 0. Harris, M.D., Pulmonary Disease Section (IliA), Veterans Administration Hospital, Gainesville, Florida 32602. A~IERICA;\;
REVIEW OF RESPIRATORY DISEASE,
kocyte proteases (5), and alveolar macrophage proteases (5) when given intratracheally can produce in animals lung lesions resembling emphysema in man. Thus, an excess of proteolytic enzyme in the air spaces of the lung has been associated with experimental emphysema. Both experimental and clinical studies have stressed the importance of the elastolytic activity of these enzymes (3, 4, 6). A potential source of proteolytic enzymes in the air space of human lungs is the alveolar macrophage, and Janoff and associates (7) have reported that human alveolar macrophages contain a low level of elastase-like esterase. Studies of human alveolar macrophages from cigarette smokers haye shown these cells to be stimulated (8) and to contain increased amounts of lysosomal enzymes, such as lysozyme, acid phosphatase, and cathepsin (9). Because of the strong association of cigarette smoking with chronic obstructive lung disease and the observation of differences in alveolar macrophages between smokers and nonsmokers,
VOLU~IE
Ill, 1975
579
580
HARRIS, OLSEN, CASTLE, AND MALONEY
elastolytic enzyme activity of these cells should be measured. This study reports proteolytic enzyme activity, including elastase-like esterase in the alveolar macrophages, peripheral leukocytes, and the alveolar lavage fluid of cigarette smoking and nonsmoking normal subjects. Materials and Methods Pulmonary alveolar macrophages (PAM) were obtained from 21 normal male volunteers-ll cigarette smokers and 10 nonsmokers. All subjects had a normal physical examination, chest roentgenogram, and routine spirometry. Each subject underwent bronchoalveolar lavage with normal saline by the method of Finley and co-workers (10). The lavage fluid obtained was centrifuged at 250 X g for lO min. The cell button was then washed twice with 50 ml of normal saline, a cell count was done, and the cells were again centrifuged at 250 X g. The original and wash supernatants were combined and concentrated to a volume of 5 ml by pressure dialysis with an Amicon ultrafiltration cell with a retentive membrane for molecular weights of 1,000 or more. The cell button in 5 ml of saline and the concentrated lavage fluid were stored at -70' C until enzyme analysis was performed. Thirty milliliters of venous blood was drawn at the time of lavage, and white blood cells (WBCs) were separated by adding an equal volume of 6 per cent dextran (mol wt, 75,000) in normal saline and allowing the red blood cells to sediment for 30 min. The WBC-rich plasma was removed and the WBCs centrifuged, washed, resuspended in saline, counted, and stored at -70' C until analysis. All intermediate steps from obtaining lavage fluid and blood to storage of final specimens were carried out at 4' C. At the time of the first enzyme determination, the cell specimens were thawed, and the cells were disrupted with a Bronwill Biosonic III homogenizer by three 15-sec probe activations at a power intensity setting of 35 per cent. This was found consistently to disrupt 95 per cent or more of the cells. Standard commercial enzyme preparations were treated in an identical fashion, and this method of cell disruption caused no change in enzyme activity. Likewise, standard commercial enzyme preparations were subjected to the conditions of freezing, thawing, and stor-
age for up to 8 months without a change in enzyme activity. This docs not imply that the PAM enzymes behave similarly, but it does suggest that the way in which the specimens were handled did not affect PAM enzyme activity. Determinations of elastase-like esterase, protease, and lysozyme activities were performed in the whole cell homogenate of PAMs and peripheral blood \VBCs and in the concentrated lavage fluid supernatant. There was not sufficient rna terial to perform each assay on each of these materials for all subjects. Protein concentration in the cell homogenates and lavage fluid was determined by the method of Lowry and associates (11). Lysozyme activity was determined by the method of Shugar (12). Dried Micrococcus lysodeihticus ('Vorthington Biochemical Corp., Freehold, N. J.) was used as a substrate, and a unit of lysozyme activity was defined as a decrease in absorbency of 0.001 per min at pH 7.0 and 25° C. A standard curve was established with egg white lysozyme (Schwarz(Mann, Orangeburg, N. Y.); a unit of activity in the experimental material could be converted to l"g of lysozyme using this standard enzyme activity curve. Protease activity was measured at pH 3.2 and 7.0 by the method of Anson, as modified by Press and associates (13). Denatured hemoglobin (Nutritional Biochemical Corp., Cleveland, Ohio) was used as substrate, and units of enzyme activity were expressed as an extinction of the trichloroacetic acid filtrate of 1.0 in excess of the blank reading. Elastase-like esterase was determined by the method of Janoff (14) using a synthetic substrate t-henzyloxy-carbonyl-L-alanine paranitrophenyl ester (BOC-Ala-ONP) (Cyclo Chemical, Los Angeles, Calif). A standard elastase cnrve was made by observing the mean change in absorbency for three 60-scc time intervals at 25° C for standard porcine pancreatic elastase (Worthington Biochemical Corp.), (2X crystalizecl, aqueous suspension, 17 units per mg of protein) at concentrations of 5 to 25 ,ug per mi. Alveolar macrophage homogenate, WBC homogenate, and lavage fluid elastase-like esterase activities were expressed in terms of equivalent porcine elastase activity from the standard curve. Spectrophotometric determinations were made with a Gilford 2400 spectrophotometer. Two milliliters of one specimen were used to measure elastase activity by the method utilizing orcein-
TABLE 1 BRONCHOPULMONARY LAVAGE CONSTITUENTS Lavage Fluid
Nonsmokers Smokers
'±SE.
Alveolar Macrophages
Volume
Total Protein
Number
Total Protein
(m/)
(mg)
( 1 Q7)
(mg)
Mg of Protein/ 1 o7 Cells
1.16 ± 0.26 7.30±1.67
0.417 ± 0.040 0.586 ± 0.043
223 ± 24' 202 ± 13
2.23 ± 0.44 2.37 ± 0.36
2.70 ± 0.57 12.54 ± 2.37
PROTEOLYTIC ENZYME ACTIVITY IN HUMAN ALVEOLAR 1\IACROPHAGES
impregnated elastin (Worthington Biochemical Corp.) as substrate (15). A unit of activity is defined as I mg of elastin solubilized per 20 min. Enzyme units were expressed as total units, units per 10 million cells, and units per milligram of pro· tein of specimen. Results for each enzyme in each specimen were compared for significant differences between the cigarette smoking and nonsmoking groups by the Student t test for nonpaired experiments. Results
Comparison of the 2 groups of subjects reveals that the mean age was 32.6 years for smokers and 29.4 years for nonsmokers. The smoking group had smoked an average of one pack of cigarettes per day for I 6.63 years, with a range of 2. 7 5 to 48.0 pack-years. Data comparing the characteristics of the lav350
• 300
250
I
•
200
••
I
~
-;; u
9.....
ISO
E
•
en
::l
100
50
•
I••
t•
•
•
• •
1;
WBC's
PAM's
WBC's
~
Nonsmokers
•
PAM's
Smokers
Fig. I. Elastase-like esterase activity of leukocytes (WBC) and pulmonary alveolar macrophages (PAM) from cigarette smokers and nonsmokers expressed as micrograms of equivalent porcine pancreatic elastase and assayed by cslerolysis of t-benzyloxycarbonyl-L-alanine paranitrophcnyl ester. The heavy cross bar represents mean values for each column with SE indicated by the shaded area.
581
age procedure and material for the 2 groups are listed in table I. Neither the volume of lavage fluid obtained nor the total protein content of the fluid was significantly different for the 2 groups. We confirmed the previously reported increased number of PAMs in cigarette smokers (8). We used a greater volume of lavage fluid and obtained more cells in each group, but the ratio of smokers' PAlVIs to nonsmokers' PAMs remained 4.6: I as reported previously (8). This increased number of P AMs in the smoking group is highly significant and assumes great importance when considering the enzyme content in the lungs of smokers. It was reported previously that the PAMs of smokers are larger by measurement of cell diameter (8); the finding in this study of increased protein content per cell in smokers' PAMs supports this, assuming that the protein concentration is similar in the PAMs of smokers and nonsmokers. Elastase-like esterase activity is shown in figure I. The difference in WBC elastase-like esterase in the 2 groups is not significant. It should be noted that lymphocytes were not separated from granulocytes; therefore, the WBC enzyme levels are probably lower than values for granulocytes alone. This makes a quantitative comparison of WBC and PAM enzymes impossible. However, if lymphocytes are considered to possess no enzyme activity and the WBC enzyme levels are corrected for a hypothetical 1: 1 granulocyte/lymphocyte ratio, the differences in PAM and WBC elastase-like esterase are significant. The mean level of 168.84 ± SE 24.42 ~tg in smokers' PAMs is significantly greater than the mean of 78.23 ± SE I 6.59 JLg per 107 cells in nonsmokers (P 0.01). Smokers' PAMs have a mean elastase-like esterase activity of 279.72 ± SE 28.69 !Lg per mg of cell protein compared to 192.08 ± SE 37.59 in nonsmokers; this difference was just outside the 95 per cent confidence limits for significance and suggests that the increase in cellular elastase-like esterase activity in smokers' PAi\Is is related more to the general stimulus responsible for an increase in cell size than to a specific stimulus for enzyme production. Protease activity in the WBCs and PAMs is shown in figure 2; this demonstrates that PAMs are potent sources of acid hydrolytic enzymes. The most significant point demonstrated in figure 2 is the marked increase in the protease activity at pH 3.2 in the smokers' PAMs compared to the nonsmokers'. Smokers' P Al\1s contained a mean of 30.48 ± SE 2.89 units per 107 cells
=
582
HARRIS, OLSEN, CASTLE, A/I:D MALONEY
70
•
60
•
50 40
~::I ~
I
. ...
~
") 10
I
•
"' 8
:::>
~
6 4
2 0
~
WI C's pH 3.2
PAM's
W8C's
NONSMOKERS
I
• ••
PAM's pH 7.0
W8 C's pH 32
• ~ PAM's
W 8 ('s
SMOKERS.
PAM's pH 7.0
Fig. 2. Protease activity at neutral and acid pH of leukocytes (WBC) and pulmonary alveolar macrophages (PAM) assayed by hydrolysis of denatured hemoglobin. The heavy cross bar represents mean values for each column with SE indicated by the shaded area. compared to 6.80 units per 107 cells in nonsmokers', a highly significant difference (P = 0.001 ). The increase in this acid hydrolase is in excess of an increase in cell size as noted for elastase, in that the protease activity expressed as units per mg of protein is approximately twice as great in smokers' PAMs, and the difference remains highly significant (P 0.001 ). The neutral protease activity of 1.30 units per 107 cells in nonsmokers' PAMs and 0.98 units per 107 cells in smokers' PAMs was not significantly different. The WBC protease activity at both acid and neutral pH was not different in smokers compared to nonsmokers. If one assumes that these normal volunteers had normal granulocyte/lymphocy te ratios in their peripheral blood the differences m PAM protease and WBC protease at pH 3.2 in both groups was significantly different. Total macrophage proteolytic enzyme content was obtained by multiplying the per cell enzyme activity by the total number of cells per lavage. A marked difference in proteolytic enzymes between smokers and nonsmokers is illustrated in figure 3. Differences in total enzyme content for each enzyme between the 2 groups of subjects is significant (P =="' 0.01 ). Lysozyme was measured in both the PAM and peripheral WBC homogenate. Smokers' PAl\Is contained approximately twice the lysozyme activity of nonsmokers' (figure 4), a sig-
=
nificant difference (P = 0.02). When the lysozyme activity in the PAMs is expressed as micrograms per mg of cell protein, no significant difference exists. This is not unexpected, because both the lysozyme activity and protein content of smokers' PAMs are increased. Enzyme activity in the concentrated lavage supernatant is shown in figure 5. Lysozyme content was significantly increased (P = 0.05) in the lavage fluid of cigarette smokers, whereas the level of elastase-like esterase and protease was not significantly different in the 2 groups. Orcein-impregna ted elastin was also used to confirm elastolytic activity of the macrophage homogenate. Although this method is less sensitive than the method using the synthetic substrate BOC-Ala-ONP, we were able to demonstrate that I ml of PAM homogenate (approximately 1.83 X l ()7 cells) from a cigarette smoker had 2.5 units of elastolytic activity by this method. Discussion
Animal studies have demonstrated that proteolytic enzymes (2, 15), especially elastolytic enzymes, can produce experimental emphysema (3, 4). Because the specificity of elastase for alanine peptide bonds is well established (16, 17), the esterolysis of the synthetic substrate BOC-Ala-ONP is considered highly specific for elastase-like enzymes. We, therefore, attempted
583
PROTEOLYTIC ENZYME ACTIVITY I N HUMAN ALVEOLAR MACROPHAGES
440
ent in the lung at all times, and in cigarette smokers they are present in increased numbers. One reason for the lack of interest in PAMs may 420 be the report of low elastase-like esterase activity in these cells, although the authors who 400 reported this questioned their findings because of the low viability of the autopsy material they 4> 380 studied (7). In our study, using the same en= C> > zyme assay on viable PAMs, we found high elasC> tase-like esterase activity, and the PAMs of cig'360 arette smokers show a significant increase over :§ nonsmokers' PAMs. A recent study in dogs sug340 4> gests that intra-alveolar administration of en= ~ C> C> > 4> C> zymes is more damaging to lung than are en__, c; 320 zymes administered intravenously (19); there.>: ~ C> fore, the large amount of elastase-like esterase -~ I: 1,200 ~ 300 activity in PAMs makes PAMs a potent potential ;::;::; agent for proteolysis of the lung in cigarette E = smokers. This, then, makes it unnecessary to 280 800 =.. propose conditions that cause sequestration of WBCs in the lung for proteolysis to occur. 400 Smokers' PAMs are also rich sources of acid protease activity, which was demonstrated by hydrolysis of denatured hemoglobin (figure 2). pH 3.2 pH 7.0 pH 3.2 pH 7.0 Protease activity decreased greatly at pH 7.0, as Proteose Elastase Protease Elastase observed by Press and associates (I 3). These NONSMOKERS SMOKERS investiga tors, however, demonstrated considerFig. 3. Total pulmonary alveolar macrophage pro- able activity at pH 4 to 5; if another substrate tease and elastase-like esterase activity per lavage in (such as denatured albumin) is used, a greater smokers and nonsmokers. Total activity was determined by multiplying the number of macrophages proteolytic activity at pH 5 is seen, indicating obtained by lavage times the per cell enzyme activity. that activity is substrate-depend ent to some exHeight of bar represents mean values with SE indi- tent. Therefore, if lung proteins become substrates, the pH at which maximum activity cated. occurs zn vivo in man is not known, but might
::
3
JAMES 0. HARRIS, GERALD N. OLSEN, JAMES R. CASTLE, and ANNE S. MALONEY
SUMMARY __________________________________________________________ Proteolysis (or more specifically, clastolysis) of the lung may be involved in the pathogenesis of pulmonary emphysema. To investigate the human alveolar macrophage as a potential mediator of lung damage, elastase-like esterase and protease activity was determined in these cells as well as in alveolar lavage fluid and in peripheral blood leukocytes. Bronchoalveolar lavage was used to obtain alveolar cells and fluid in normal volunteers who were divided into two groups according to cigarette smoking history, nonsmokers and smokers. Results of these studies revealed that human alveolar macrophages possess a high activity of both elastase-like esterase and protease. Furthermore, the alveolar macrophages of cigarette smokers had a significantly greater elastase-like esterase and protease activity than those of nonsmokers. \Vhen the 4- to 5-fold increase in the number of macrophages found in cigarette smokers is taken into account there was approximately 10 times more elastase-like esterase activity and 18 times more protease activity within macrophages in the alveolar spaces of cigarette smokers' lungs. This makes the alveolar mauophage a potent potential source of proteolytic enzymes in man.
Introduction In the wake of Laurell and Eriksson's (I) discovery of a 1 -antitrypsin deficiency and the association of this deficiency with emphysema, intense interest has developed in proteolysis as a pathogenic mechanism in chronic obstructive lung disease. Experimentally, it has been demonstrated that papain (2, 3), elastase (3, 4), leu-
(Received in original form August 22, 1974 and in revised form ]anuary 15, 1975) 1 From the Veterans Administration Hospital and the Department of Medicine, College of Medicine, University of Florida, Gainesville, Florida. 2 This research was supported in part by N"ational Institutes of Health Research Grant ES00789 and by a grant from the Florida Lung Association. 3 Requests for reprints should be addressed to James 0. Harris, M.D., Pulmonary Disease Section (IliA), Veterans Administration Hospital, Gainesville, Florida 32602. A~IERICA;\;
REVIEW OF RESPIRATORY DISEASE,
kocyte proteases (5), and alveolar macrophage proteases (5) when given intratracheally can produce in animals lung lesions resembling emphysema in man. Thus, an excess of proteolytic enzyme in the air spaces of the lung has been associated with experimental emphysema. Both experimental and clinical studies have stressed the importance of the elastolytic activity of these enzymes (3, 4, 6). A potential source of proteolytic enzymes in the air space of human lungs is the alveolar macrophage, and Janoff and associates (7) have reported that human alveolar macrophages contain a low level of elastase-like esterase. Studies of human alveolar macrophages from cigarette smokers haye shown these cells to be stimulated (8) and to contain increased amounts of lysosomal enzymes, such as lysozyme, acid phosphatase, and cathepsin (9). Because of the strong association of cigarette smoking with chronic obstructive lung disease and the observation of differences in alveolar macrophages between smokers and nonsmokers,
VOLU~IE
Ill, 1975
579
580
HARRIS, OLSEN, CASTLE, AND MALONEY
elastolytic enzyme activity of these cells should be measured. This study reports proteolytic enzyme activity, including elastase-like esterase in the alveolar macrophages, peripheral leukocytes, and the alveolar lavage fluid of cigarette smoking and nonsmoking normal subjects. Materials and Methods Pulmonary alveolar macrophages (PAM) were obtained from 21 normal male volunteers-ll cigarette smokers and 10 nonsmokers. All subjects had a normal physical examination, chest roentgenogram, and routine spirometry. Each subject underwent bronchoalveolar lavage with normal saline by the method of Finley and co-workers (10). The lavage fluid obtained was centrifuged at 250 X g for lO min. The cell button was then washed twice with 50 ml of normal saline, a cell count was done, and the cells were again centrifuged at 250 X g. The original and wash supernatants were combined and concentrated to a volume of 5 ml by pressure dialysis with an Amicon ultrafiltration cell with a retentive membrane for molecular weights of 1,000 or more. The cell button in 5 ml of saline and the concentrated lavage fluid were stored at -70' C until enzyme analysis was performed. Thirty milliliters of venous blood was drawn at the time of lavage, and white blood cells (WBCs) were separated by adding an equal volume of 6 per cent dextran (mol wt, 75,000) in normal saline and allowing the red blood cells to sediment for 30 min. The WBC-rich plasma was removed and the WBCs centrifuged, washed, resuspended in saline, counted, and stored at -70' C until analysis. All intermediate steps from obtaining lavage fluid and blood to storage of final specimens were carried out at 4' C. At the time of the first enzyme determination, the cell specimens were thawed, and the cells were disrupted with a Bronwill Biosonic III homogenizer by three 15-sec probe activations at a power intensity setting of 35 per cent. This was found consistently to disrupt 95 per cent or more of the cells. Standard commercial enzyme preparations were treated in an identical fashion, and this method of cell disruption caused no change in enzyme activity. Likewise, standard commercial enzyme preparations were subjected to the conditions of freezing, thawing, and stor-
age for up to 8 months without a change in enzyme activity. This docs not imply that the PAM enzymes behave similarly, but it does suggest that the way in which the specimens were handled did not affect PAM enzyme activity. Determinations of elastase-like esterase, protease, and lysozyme activities were performed in the whole cell homogenate of PAMs and peripheral blood \VBCs and in the concentrated lavage fluid supernatant. There was not sufficient rna terial to perform each assay on each of these materials for all subjects. Protein concentration in the cell homogenates and lavage fluid was determined by the method of Lowry and associates (11). Lysozyme activity was determined by the method of Shugar (12). Dried Micrococcus lysodeihticus ('Vorthington Biochemical Corp., Freehold, N. J.) was used as a substrate, and a unit of lysozyme activity was defined as a decrease in absorbency of 0.001 per min at pH 7.0 and 25° C. A standard curve was established with egg white lysozyme (Schwarz(Mann, Orangeburg, N. Y.); a unit of activity in the experimental material could be converted to l"g of lysozyme using this standard enzyme activity curve. Protease activity was measured at pH 3.2 and 7.0 by the method of Anson, as modified by Press and associates (13). Denatured hemoglobin (Nutritional Biochemical Corp., Cleveland, Ohio) was used as substrate, and units of enzyme activity were expressed as an extinction of the trichloroacetic acid filtrate of 1.0 in excess of the blank reading. Elastase-like esterase was determined by the method of Janoff (14) using a synthetic substrate t-henzyloxy-carbonyl-L-alanine paranitrophenyl ester (BOC-Ala-ONP) (Cyclo Chemical, Los Angeles, Calif). A standard elastase cnrve was made by observing the mean change in absorbency for three 60-scc time intervals at 25° C for standard porcine pancreatic elastase (Worthington Biochemical Corp.), (2X crystalizecl, aqueous suspension, 17 units per mg of protein) at concentrations of 5 to 25 ,ug per mi. Alveolar macrophage homogenate, WBC homogenate, and lavage fluid elastase-like esterase activities were expressed in terms of equivalent porcine elastase activity from the standard curve. Spectrophotometric determinations were made with a Gilford 2400 spectrophotometer. Two milliliters of one specimen were used to measure elastase activity by the method utilizing orcein-
TABLE 1 BRONCHOPULMONARY LAVAGE CONSTITUENTS Lavage Fluid
Nonsmokers Smokers
'±SE.
Alveolar Macrophages
Volume
Total Protein
Number
Total Protein
(m/)
(mg)
( 1 Q7)
(mg)
Mg of Protein/ 1 o7 Cells
1.16 ± 0.26 7.30±1.67
0.417 ± 0.040 0.586 ± 0.043
223 ± 24' 202 ± 13
2.23 ± 0.44 2.37 ± 0.36
2.70 ± 0.57 12.54 ± 2.37
PROTEOLYTIC ENZYME ACTIVITY IN HUMAN ALVEOLAR 1\IACROPHAGES
impregnated elastin (Worthington Biochemical Corp.) as substrate (15). A unit of activity is defined as I mg of elastin solubilized per 20 min. Enzyme units were expressed as total units, units per 10 million cells, and units per milligram of pro· tein of specimen. Results for each enzyme in each specimen were compared for significant differences between the cigarette smoking and nonsmoking groups by the Student t test for nonpaired experiments. Results
Comparison of the 2 groups of subjects reveals that the mean age was 32.6 years for smokers and 29.4 years for nonsmokers. The smoking group had smoked an average of one pack of cigarettes per day for I 6.63 years, with a range of 2. 7 5 to 48.0 pack-years. Data comparing the characteristics of the lav350
• 300
250
I
•
200
••
I
~
-;; u
9.....
ISO
E
•
en
::l
100
50
•
I••
t•
•
•
• •
1;
WBC's
PAM's
WBC's
~
Nonsmokers
•
PAM's
Smokers
Fig. I. Elastase-like esterase activity of leukocytes (WBC) and pulmonary alveolar macrophages (PAM) from cigarette smokers and nonsmokers expressed as micrograms of equivalent porcine pancreatic elastase and assayed by cslerolysis of t-benzyloxycarbonyl-L-alanine paranitrophcnyl ester. The heavy cross bar represents mean values for each column with SE indicated by the shaded area.
581
age procedure and material for the 2 groups are listed in table I. Neither the volume of lavage fluid obtained nor the total protein content of the fluid was significantly different for the 2 groups. We confirmed the previously reported increased number of PAMs in cigarette smokers (8). We used a greater volume of lavage fluid and obtained more cells in each group, but the ratio of smokers' PAlVIs to nonsmokers' PAMs remained 4.6: I as reported previously (8). This increased number of P AMs in the smoking group is highly significant and assumes great importance when considering the enzyme content in the lungs of smokers. It was reported previously that the PAMs of smokers are larger by measurement of cell diameter (8); the finding in this study of increased protein content per cell in smokers' PAMs supports this, assuming that the protein concentration is similar in the PAMs of smokers and nonsmokers. Elastase-like esterase activity is shown in figure I. The difference in WBC elastase-like esterase in the 2 groups is not significant. It should be noted that lymphocytes were not separated from granulocytes; therefore, the WBC enzyme levels are probably lower than values for granulocytes alone. This makes a quantitative comparison of WBC and PAM enzymes impossible. However, if lymphocytes are considered to possess no enzyme activity and the WBC enzyme levels are corrected for a hypothetical 1: 1 granulocyte/lymphocyte ratio, the differences in PAM and WBC elastase-like esterase are significant. The mean level of 168.84 ± SE 24.42 ~tg in smokers' PAMs is significantly greater than the mean of 78.23 ± SE I 6.59 JLg per 107 cells in nonsmokers (P 0.01). Smokers' PAMs have a mean elastase-like esterase activity of 279.72 ± SE 28.69 !Lg per mg of cell protein compared to 192.08 ± SE 37.59 in nonsmokers; this difference was just outside the 95 per cent confidence limits for significance and suggests that the increase in cellular elastase-like esterase activity in smokers' PAi\Is is related more to the general stimulus responsible for an increase in cell size than to a specific stimulus for enzyme production. Protease activity in the WBCs and PAMs is shown in figure 2; this demonstrates that PAMs are potent sources of acid hydrolytic enzymes. The most significant point demonstrated in figure 2 is the marked increase in the protease activity at pH 3.2 in the smokers' PAMs compared to the nonsmokers'. Smokers' P Al\1s contained a mean of 30.48 ± SE 2.89 units per 107 cells
=
582
HARRIS, OLSEN, CASTLE, A/I:D MALONEY
70
•
60
•
50 40
~::I ~
I
. ...
~
") 10
I
•
"' 8
:::>
~
6 4
2 0
~
WI C's pH 3.2
PAM's
W8C's
NONSMOKERS
I
• ••
PAM's pH 7.0
W8 C's pH 32
• ~ PAM's
W 8 ('s
SMOKERS.
PAM's pH 7.0
Fig. 2. Protease activity at neutral and acid pH of leukocytes (WBC) and pulmonary alveolar macrophages (PAM) assayed by hydrolysis of denatured hemoglobin. The heavy cross bar represents mean values for each column with SE indicated by the shaded area. compared to 6.80 units per 107 cells in nonsmokers', a highly significant difference (P = 0.001 ). The increase in this acid hydrolase is in excess of an increase in cell size as noted for elastase, in that the protease activity expressed as units per mg of protein is approximately twice as great in smokers' PAMs, and the difference remains highly significant (P 0.001 ). The neutral protease activity of 1.30 units per 107 cells in nonsmokers' PAMs and 0.98 units per 107 cells in smokers' PAMs was not significantly different. The WBC protease activity at both acid and neutral pH was not different in smokers compared to nonsmokers. If one assumes that these normal volunteers had normal granulocyte/lymphocy te ratios in their peripheral blood the differences m PAM protease and WBC protease at pH 3.2 in both groups was significantly different. Total macrophage proteolytic enzyme content was obtained by multiplying the per cell enzyme activity by the total number of cells per lavage. A marked difference in proteolytic enzymes between smokers and nonsmokers is illustrated in figure 3. Differences in total enzyme content for each enzyme between the 2 groups of subjects is significant (P =="' 0.01 ). Lysozyme was measured in both the PAM and peripheral WBC homogenate. Smokers' PAl\Is contained approximately twice the lysozyme activity of nonsmokers' (figure 4), a sig-
=
nificant difference (P = 0.02). When the lysozyme activity in the PAMs is expressed as micrograms per mg of cell protein, no significant difference exists. This is not unexpected, because both the lysozyme activity and protein content of smokers' PAMs are increased. Enzyme activity in the concentrated lavage supernatant is shown in figure 5. Lysozyme content was significantly increased (P = 0.05) in the lavage fluid of cigarette smokers, whereas the level of elastase-like esterase and protease was not significantly different in the 2 groups. Orcein-impregna ted elastin was also used to confirm elastolytic activity of the macrophage homogenate. Although this method is less sensitive than the method using the synthetic substrate BOC-Ala-ONP, we were able to demonstrate that I ml of PAM homogenate (approximately 1.83 X l ()7 cells) from a cigarette smoker had 2.5 units of elastolytic activity by this method. Discussion
Animal studies have demonstrated that proteolytic enzymes (2, 15), especially elastolytic enzymes, can produce experimental emphysema (3, 4). Because the specificity of elastase for alanine peptide bonds is well established (16, 17), the esterolysis of the synthetic substrate BOC-Ala-ONP is considered highly specific for elastase-like enzymes. We, therefore, attempted
583
PROTEOLYTIC ENZYME ACTIVITY I N HUMAN ALVEOLAR MACROPHAGES
440
ent in the lung at all times, and in cigarette smokers they are present in increased numbers. One reason for the lack of interest in PAMs may 420 be the report of low elastase-like esterase activity in these cells, although the authors who 400 reported this questioned their findings because of the low viability of the autopsy material they 4> 380 studied (7). In our study, using the same en= C> > zyme assay on viable PAMs, we found high elasC> tase-like esterase activity, and the PAMs of cig'360 arette smokers show a significant increase over :§ nonsmokers' PAMs. A recent study in dogs sug340 4> gests that intra-alveolar administration of en= ~ C> C> > 4> C> zymes is more damaging to lung than are en__, c; 320 zymes administered intravenously (19); there.>: ~ C> fore, the large amount of elastase-like esterase -~ I: 1,200 ~ 300 activity in PAMs makes PAMs a potent potential ;::;::; agent for proteolysis of the lung in cigarette E = smokers. This, then, makes it unnecessary to 280 800 =.. propose conditions that cause sequestration of WBCs in the lung for proteolysis to occur. 400 Smokers' PAMs are also rich sources of acid protease activity, which was demonstrated by hydrolysis of denatured hemoglobin (figure 2). pH 3.2 pH 7.0 pH 3.2 pH 7.0 Protease activity decreased greatly at pH 7.0, as Proteose Elastase Protease Elastase observed by Press and associates (I 3). These NONSMOKERS SMOKERS investiga tors, however, demonstrated considerFig. 3. Total pulmonary alveolar macrophage pro- able activity at pH 4 to 5; if another substrate tease and elastase-like esterase activity per lavage in (such as denatured albumin) is used, a greater smokers and nonsmokers. Total activity was determined by multiplying the number of macrophages proteolytic activity at pH 5 is seen, indicating obtained by lavage times the per cell enzyme activity. that activity is substrate-depend ent to some exHeight of bar represents mean values with SE indi- tent. Therefore, if lung proteins become substrates, the pH at which maximum activity cated. occurs zn vivo in man is not known, but might
::
