Microbial Pathogenesis 1990 ; 9 : 117-125
Short communications Inhibition of polymorphonuclear leukocyte function by Legionella pneumophila exoproducts Narendra N . Sahney,' Beverly C . Lambe,' * James T . Summersgi112 and Richard D . Miller 2 t 'Department of Microbiology and Immunology, School of Medicine, and 2Section of Infectious Diseases, Department of Medicine, University of Louisville, Louisville, Kentucky 40292, U .S .A . (Received February 22, 1990; accepted in revised form April 30, 1990)
Sahney, N . N . (Dept of Microbiology and Immunology, School of Medicine, University of Louisville, Louisville, Kentucky 40292, U .S .A .), B . C . Lambe, J . T . Summersgill and R . D . Miller . Inhibition of polymorphonuclear leukocyte function by Legionella pneumophila exoproducts . Microbial Pathogenesis 1990 ; 9 : 117-125 . Total exoproducts (relative molecular mass > 10000) from wild-type strains of Legionella pneumophila markedly inhibited human polymorphonuclear leukocyte (PMN) superoxide anion generation, at sub-lethal concentrations, in response to four stimuli [1 .7, 0, 0 .6 and 3 .4% of control for zymosan activated particles (ZAP), phorbol myristate acetate (PMA), calcium ionophore (A 23187), and formyl-methionyl-leucyl-phenylalanine (fMLP), respectively] . PMN chemotaxis towards fMLP and spontaneous migration, were also dramatically inhibited (2 .8 and 2 .9% of buffer-treated controls, respectively) . In contrast, total exoproducts from the cas-1 strain of L . pneumophila, a protease-deficient mutant generated by ethyl methane sulfonate mutagenesis, failed to inhibit PMN superoxide production in response to ZAP and PMA and only partially inhibited PMN response to A 23187 and fMLP . PMN spontaneous migration was unaffected by treatment with total exoproducts from the mutant, while directed chemotaxis was partially inhibited (51 .4%) . These data demonstrated that L . pneumophila total exoproducts, primarily protease had significant inhibitory effects on normal PMN function and may play an important contributory role in the pathogenesis of legionnaire's disease . Key words: Legionella pneumophila ; protease ; PMN ; superoxide anion generation ; chemotaxis .
Introduction Legionella pneumophila is the etiological agent of legionnaire's disease (LD), most often presenting as a life-threatening bronchopneumonia . The pathogenesis of LD involves the intracellular multiplication of the organism in alveolar macrophages as demonstrated in lung specimens from patients with LD,' in the lungs of experimentally infected guinea pigs' and within in vitro cultured alveolar macrophages .' A major feature of L . pneumophila is its ability to inhibit phagosome-lysosome fusion in permissive monocyte cultures in vitro' and its resistance to killing by human polymorphonuclear
Present address : Johns Hopkins Medical Institutions, Meyer Building, Room B 1-193, 600 N . Wolfe St ., Baltimore, MD 21205, U .S .A. t Author to whom correspondence should be addressed . 0882-4010/90/080117+09 $03 .00/0
© 1990 Academic Press Limited
1 18
N . N . Sahney et al.
Table 1 Enzymatic activity in culture supernatants of Legionella pneumophila strains Knoxville-1, cas-1 mutant and Ver5e Enzyme activity Strains
Proteases
PL b
AP`
RNAse`
Knoxville-1 cas-1 Ver5e
353 .3 3 .1 360 .0
14 15 23
0 .299 0 .299 0 .240
12 15 18
a
Units of protease activity ; hide powder azure substrate . Phospholipase A/esterase ; zones of hydrolysis in Tween40 agar (mm diameter) . `Acid phosphatase ; units of activity, p-nitrophenylphosphate substrate . 'Ribonuclease ; zones of hydrolysis in RNA agar (mm diameter) . b
leukocytes (PMN) .' The mechanisms by which these responses are induced and the virulence factor(s) involved have not been elucidated . L . pneumophila produces a wide array of extracellular enzymes and toxins (exoproducts) including a cytotoxin s several proteases,' phospholipase C, 8 phospholipase A/esterase, 9 acid phosphatases, 10 amylase,' ribonuclease s and beta lactamase .' It has previously been shown that the cytotoxin 11 - 12 and acid phosphatase 10 interfere with the oxidative burst of PMN . Recently, purified protease from L . pneumophila has been demonstrated to have haemolytic and cytotoxic activity . 13 Taken as a whole, this information suggests that these exoproducts may play a role in the pathogenesis of LD . In the present study, ethyl methane sulfonate mutagenesis was employed and a protease-deficient mutant of L . pneumophila was isolated and characterized . This study compared the protease-deficient mutant and the wild-type strain by examining the effects of their total exoproducts (relative molecular mass > 10000) on normal PMN superoxide anion generation and chemotaxis . Results Isolation and characterization of mutant Mutagenesis was carried out by ethyl methane sulfonate (EMS) treatment and a total of 18500 colonies from the virulent L . pneumophila Ver5e and avirulent Knoxville-1 strains were screened for the absence of proteolytic activity on casein-containing agar plates . One mutant (cas-1) was isolated from the Knoxville-1 strain exhibiting an apparent mutation rate of 0 .007%, whereas no mutants were isolated from the virulent Ver5e strain . A total of 7500 colonies from the two strains were screened for loss of phospholipase A/esterase activity on Tween-40 agar plates, however no mutants deficient in this exoproduct were isolated . The cas-1 mutant was stable after 20 weekly passages on BCYE with no apparent reversions to wild-type protease production . Enzymatic activities assayed in culture supernatants (Table 1), or the 20 enzymes of the API-ZYM system (Analytab Products, Plainview, New York ; positive reactions for alkaline phosphatase, acid phosphatase, butyrate esterase, caprylate esterase/lipase, leucine aminopeptidase, valine aminopeptidase, cystine aminopeptidase and phosphoamidase) failed to demonstrate any other major difference between the cas-1 mutant and wild-type avirulent Knoxville-1 strain (or the virulent Ver5e strain) . The
1 19
Inhibition of PMN by L . pneumophila exoproducts
140 •
(a )
120 •
100 80 60 40 20 0 Z
0-
2 Q a NJ
U)
U)
0
0
0
U 2
Wild-type
z a a a NI
N
z
0
C1.
0
a a
U
N
0 0
0 U
•
z C U) • 2 o a a_ U
2
Wild-type
Protease mutant
Protease mutant
140 0 0
120
• 100 0 0,0 80
v• x•
60 40
aL CL
•
20 0 z
f a
a J F
U, 0 0
0
v
z a 2 J a -2
U) C U
Protease mutant Wild-type Fig . 1 . Effects of ConS from wild-type and protease-deficient mutant strains of Legionella pneumophila and HiConS, on superoxide anion generation by PMN in response to (a) ZAP ; (b) PMA ; (c) A23187 ; (d) fMLP . Superoxide anion generation was determined by adding ConS to 2 .0x106 PMN/ml and stimulating with ZAP, PMA, A23187, of fMLP (details in Materials and methods) . Data shown represent means and standard deviations of six and three experiments with ConS of wild-type and mutant strains, respectively, and expressed as a percentage of each stimulus (0) . Viability of PMN treated with ConS was > 95% .
only significant difference was that the cas-1 mutant was a low protease producer (0 .9% of wild-type) . Superoxide anion generation Concentrated supernatants (ConS) from the mutant and wild-type strains were added to washed PMN which were then exposed to one of four stimuli . In all experiments, ConS was added in an amount equivalent to 75 units of phospholipase A/esterase activity (used as an internal standard) with the average amount of protease present in the ConS being 1316 units for the Knoxville-1 and Ver5e strains and only 4 .2 units for the mutant . Additionally, the amount of ConS used was based on concentrations that did not affect overall PMN viability, as determined by trypan blue dye exclusion . Controls consisted of PMN treated with each stimuli (represented as the 100% control) and buffer alone . As shown in Fig . 1, PMN exposed to the wild-type ConS and then stimulated with zymosan activated particles (ZAP), phorbol myristate acetate (PMA), calcium ionophore (A23187), or formyl-methionyl-leucyl-phenalanine (fMLP) showed almost total inhibition of superoxide anion generation (1 .7, 0, 0 .6 and 3 .4% of controls
1 20
N . N . Sahney et al.
respectively, P < 0 .05) . In contrast, PMN exposed to ConS from the cas-1 mutant and stimulated as described showed only a modest inhibition of superoxide anion levels with A 23187 and fMLP (61 .5 and 49 .5% of controls respectively, P < 0 .05), and no statistically significant inhibition in superoxide anion generation with ZAP and PMA (61 .1 and 89% of controls respectively, P < 0 .5) . No significant differences were noted between the heat-inactivated ConS (HiConS) and untreated controls with any of the four stimuli used (87 .5, 64 .2, 78 .7 and 116% of ZAP, PMA, A 23187 and fMLP controls respectively, P < 0 .5) . The inhibitory effect of ConS (wild-type) on superoxide anion generation was also partially reversible, since PMN preincubated for 30 min with ConS, then washed twice in HBSS and stimulated in the presence of cytochrome C showed a significant recovery (78 .5% of control) . However, 10-fold higher concentrations of ConS resulted in greater inhibition (33 .6% of control) . Measurement of maximal reduction of ferricytochrome C by the addition of sodium dithionite revealed that inhibition of PMN superoxide anion generation upon exposure to ConS from wild-type was not due to substrate limitation . Total reduction of cytochrome C exceeded the levels of superoxide dismutase-inhibitable superoxide anion generated in all cases (data not shown) . Chemotaxis Under agarose chemotaxis by PIVIN was determined using fM LP as the chemoattractant . As shown in Fig . 2(a), PMN pretreated for 30 min with ConS of the wild-type strains showed a dramatic decrease in chemotactic ability (chemotactic differential 2 .8% of buffer control, P < 0 .05) . In contrast, PMN pretreated with ConS from the mutant were considerably less inhibitory (chemotactic differential 51 .4% of control, P < 0 .05) . Pretreatment of PMN with heat-inactivated ConS also showed a similar partial inhibition (chemotactic differential 47 .8% of control, P < 0 .05) . As a control, pretreatment with YE broth medium alone had no effect on chemotactic differential (99 .6% of buffer control) . Spontaneous migration of PMN was also inhibited markedly when pretreated with ConS from wild type strains as shown in Fig . 2(b) (2 .9% of control, P < 0 .05) . Interestingly, PMN pretreated with ConS from the mutant had no significant inhibitory effect on spontaneous migration (96 .0% of control, P < 0 .5), while the HiConS showed a small but significant inhibition (84 .1% of buffer control, P < 0 .05) .
Discussion Based on enzymatic activities present in culture supernatants, the cas-1 mutant used in this study closely resembled the parental strains, with the major difference being that the mutant had low levels of protease activity in culture supernatants (none in cell lysates) . This suggested an alteration in some aspect of protease production, and not in secretion . Virulence testing of cas-1 mutant was not reported as part of the present study because the mutant was derived from an avirulent parent (Knoxville1) . However, a naturally occurring (environmental) protease-deficient strain of L . pneumophila isolated by our laboratory has been found to be avirulent for guinea pigs (unpublished data) . Since L . pneumophi/a produces several extracellular enzymes and toxins, the aim of this study was to examine the effects of these total exoproducts on two parameters of PMN functions, the oxidative and chemotactic response to exogenous stimuli . The exoproducts were tested in concentrated form (relative molecular mass > 10000) which excluded the low molecular weight cytotoxin . In addition, the concentrations
1 21
Inhibition of PMN by L . pneumophila exoproducts
120 100 80 60 40 20 0
II Buffer Cons
Buffer Cans HiConS
Medium
120
T
O U
100
O
o 80 0
2 T E
60
0
40
G7 0 0 `0
n
20 0 Buffer Cons Wild-type
Buffer ConS HiConS
Medium
Protease mutant
Fig . 2 . Effects of ConS of wild type and mutant strains of L . pneumophi/a, HiConS and YE broth medium on PMN (a) chemotaxis ; (b) spontaneous migration . PMN were treated with either ConS, HiConS, buffer or medium for 30 min . Under agarose chemotaxis was performed in response to fMLP 10 -7 M . Chemotaxis differential was determined by directed minus spontaneous migration of PMN . Data shown represent mean and standard deviations of nine experiments performed in triplicate and expressed as a percentage of buffer treated controls (0) . Viability of pretreated PMN was > 95% .
used were chosen so as to be sub-lethal for PMN viability and thus may represent more relevant concentrations actually observed in vivo . Analysis of superoxide anion generation data revealed that ConS from wild-type strains completely inhibited PMN superoxide anion generation in response to four stimuli used (ZAP, PMA, A 23187 and fMLP) . This inhibitory effect was not due to limitations in substrate availability as determined by excess reducible cytochrome C present after the addition of sodium dithionite . The amount of ConS used minimally affected the cytochrome C and thus was not sufficient to explain the complete inhibition . Also, the inhibitory effect of ConS was only partially reversible following washing . No statistically significant inhibition of PMN superoxide anion generation was seen after treatment with ConS from the mutant for two of the stimuli (ZAP and PMA), suggesting that the protease can completely diminish PMN oxidative response at these sub-lethal concentrations . Saha et a/. 1 ° demonstrated that an acid phosphatase
1 22
N . N . Sahney et al .
isolated from L . micdadei could block PMN superoxide ani, on generation . However, in the present study both wild-type and mutant strains contained equivalent amounts of acid phosphatase . In a similar study the exoprotease from Pseudomonas aeroginosa has also been shown to inhibit PMN superoxide anion generation ." Partial inhibition of superoxide anion generation by ConS from mutant when A 23187 and fMLP were used as stimuli suggested that other (presumably heat-labile) exoproducts also contributed to the inhibitory effect on PMN function . The differences between the four stimuli in PMN superoxide anion generation following treatment with ConS from the mutant is not totally unexpected, since the four stimuli used in the study can activate PMN via different pathways . Spontaneous migration of PMN was also severely inhibited (3% of control) by ConS from wild-type, but was minimally inhibited with ConS from mutant (96% of control), which indicated that the inhibitory effect on PMN spontaneous migration was solely due to the protease . The ability of L . pneumophila protease to completely inhibit PMN spontaneous migration suggested that protease may interfere not only with sensory and transducing mechanisms (possibly cleaving fMLP receptors), but that it may also alter some aspect of the effector mechanisms involved in PMN locomotion (at sublethal concentrations) . The PMN chemotactic response to the chemoattractant fMLP was also dramatically inhibited by ConS from the wild-type strains (chemotactic differential 2 .8% of control), while this inhibition was only partial with both ConS from mutant and HiConS (chemotactic differential 51 .4 and 47 .8% of controls respectively) . This suggested that other components present in the exoproducts (presumably heatstable) have inhibitory effects on PMN chemotactic function as well . In a similar study, complete inhibition of PMN chemotaxis has been reported to occur following treatment with an exoprotease from Pseudomonas aeroginosa . 14 Overall, analysis of the data in this study suggested that L . pneumophila exoproducts inhibit PMN superoxide anion generation and chemotaxis . At the sub-lethal concentrations used, the protease had a major role in the inhibitory effect on these PMN functions . Of course, the protease is the major secretory protein produced by this bacterium 15 and would be present in the highest concentration in the ConS relative to the other secretory products . Nevertheless, considerable evidence has been accumulating concerning the role of protease in the pathogenesis of LD . Studies have documented the production of protease by L . pneumophila in the lungs of guinea pigs with experimentally induced LD . 16 Baskerville et al." and Conlan et al.' have reported that purified protease aerosolized into the lungs of guinea pigs induced lesions resembling those seen in experimentally induced guinea pig legionellosis and also from lung biopsy of humans with naturally acquired disease . Antibody to the protease has been detected in convalescent-phase sera from LD patients, 18 and vaccination of guinea pigs with the protease has been shown to induce protective immunity . 19 Purified protease has also been shown to inhibit human natural killer (NK) cell activity . 20 The results of the present study support the hypothesis that L . pneumophila exoproducts (particularly protease) may play an important contributory role in the pathogenesis of LD by interfering with phagocytic responses and in resistance to killing in vivo by PMN and possibly macrophages . The effect of purified L . pneumophila protease on those same PMN and macrophage functions is currently being investigated .
Materials and methods Bacteria/ strains and media . Legione/la pneumophi/a strain Ver5e (serogroup-1) was isolated from an environmental sample (cooling tower water) on buffered charcoal yeast extract (BCYE) agar medium prepared from Legionella Agar Base (Difco Laboratories, Detroit, Michigan),
Inhibition of PMN by L . pneumophila exoproducts
123
supplemented with 0 .01% (w/v) L-cysteine and 0 .05% ferric pyrophosphate . This isolate has been well characterized in our laboratory27 and is virulent for guinea pigs based on an LD 50 of < 10' organisms by intraperitoneal injection . L . pneumophi/a strain Knoxville-1 (serogroup-1) is an extensively laboratory-adapted (avirulent) isolate originally obtained from R . Weaver (Centers for Disease Control, Atlanta, Georgia) . All bacteria were stored at -70°C in Tryptic soy broth (Difco) containing 20% (w/v) glycerol . Casein agar (screening medium for protease) contained 1 .0% yeast extract, 1 .7% Difco-agar, 1 .0% ACES, 1 .0% casein, 0 .5% corn starch, supplemented with 0 .04% L-cyteine and 0.025% ferric pyrophosphate . Tween-40 agar (screening medium for phospholipase A/esterase) contained 1 .0% yeast extract, 1 .7% Difco-agar, 1 .0% ACES, 0 .075% Tween 40, 2 .0% charcoal, supplemented with 0 .04% L-cysteine and 0 .025% ferric pyrophosphate . Ethyl methane su/fonate mutagenesis . L. pneumophila strains Knoxville-1 and Ver5e were grown in yeast extract (YE) broth containing 1 .0% yeast extract (Difco) and 1 .0% ACES (Research Organics, Cleveland, Ohio) supplemented with L-cysteine and ferric pyrophosphate as described above . Cultures were grown to mid-to-late exponential phase, centrifuged at 10000xg for 10 min at 25°C and resuspended in 10 ml YE broth . Ethyl methane sulfonate was added to the suspension at a final concentration of 15 µg/ml and the culture was incubated for 1 h at 35°C on a rotary shaker . Cells were centrifuged as described above, washed once in YE broth and resuspended in 20 ml YE broth . A portion of these cells was diluted in distilled water and plated on casein and Tween-40 media so as to obtain approximately 100 colonies per plate (10x150 mm Petri dish) . An additional 20 ml of YE broth was added to the remaining cells, incubated for 1 h at 35°C, and then used directly to inoculate fresh YE broth . These cultures were incubated at 37°C with shaking overnight, then diluted and plated onto casein media and Tween-40 media . Casein plates were incubated at 37°C in 5% CO 2, examined daily for 4 days, and colonies lacking a zone of white precipitate were isolated and characterized . Tween-40 agar plates were incubated for 7-10 days at 37°C in 5% CO 2 and examined daily for colonies lacking an oily sheen on the surface of agar plates . Enzyme assays . Protease activity was measured by the method described by Thompson et a/. 22 Briefly, 100 µl of sample was added to 20 mg hide powder azure (Sigma Chemical, St Louis, Missouri) suspended in 5 ml of 100 mm sodium phosphate buffer (pH 6 .0) and incubated at 35°C for 60 min on a rotary shaker . Samples were filtered (Whatman number 1 filter) and the absorbance of the soluble fraction was measured at 595 nm in an Ultrospec II (LKB, Biochrom, Cambridge, U .K .) spectrophotometer . Each unit was arbitrarily defined as proteolysis resulting in a change of absorbance of 0 .001 per h . The phospholipase A/esterase of L . pneumophila has activity on phospholipids (e .g . phosphatidylinositol), as well as on general esterase substrates (T . C . Thorpe and R . D . Miller, unpublished data) . In this study, phospholipase A/esterase activity was determined by adding 20 µI of sample to 2 .5 ml 10 µnn Tris-HCI buffer (ph 8 .0) containing 0 .2% sodium deoxycholate . p-Nitrophenyl caprylate (PNP-caprylate ; Sigma ; 20 µl) was then added as an esterase substrate and incubated for 5 min at 37°C . Dimethyl sulfoxide (DMSO ; 1 ml) was added and the absorbance read at 415 nm . One unit of activity was defined as 1 nmcl of PNP released/min . Phospholipase A/esterase activity was also measured using an agar plate assay containing 0 .2% deoxycholate, 1 .0% Tween-40 (esterase substrate), 0 .01% sodium azide, 1 .7% Difco-agar in 0 .01 M Tris-HCI buffer (pH 8.0) . Zones of oily sheen radiating from wells containing the enzyme were then measured . Acid phosphatase activity was assayed by the method of Saha et al." Briefly, samples were incubated with PNP-phosphate in 0 .15 M citrate buffer (pH 4 .8) for 3 h . Sodium hydroxide was added and the absorbance measured at 420 nm . Ribonuclease activity was detected on plates containing 0 .2% ribonucleic acid (Type II-s from Toru/a yeast cells) in 0 .1 M sodium phosphate buffer (pH 6 .0) and 1 .7% Difco-agar. After 96 h incubation at 35°C, plates were flooded with 1 M HCI and zones of clearing were measured . Concentrated supernatants (ConS) . Each isolate was inoculated into 100 ml YE broth at 30 Klett units and incubated in a rotary shaker at 37°C . After 18 h, cells were removed by centrifugation at 10000 xg for 15 min at 4°C . Culture supernatant (50 ml) was then concentrated to 5 ml by ultrafiltration in a Centricell 60 (Polysciences, Warrington, Pennsylvania) with a 10000 relative molecular mass cut-off . Half of the concentrated supernatant (ConS) was heattreated by autoclaving for 90 min (HiConS), and any precipitate formed was separated by centrifugation and discarded .
124
N . N . Sahney et al.
Neutrophil isolation. Whole blood (75 ml) was collected from healthy volunteers in tubes containing heparin (10 units/ml) . PMN were isolated by buoyant density centrifugation using Neutrophil Isolation Medium (Los Alamos Diagnostics, Los Alamos, New Mexico) and washed twice in Hanks balanced salt solution (HBSS) . The final pellet containing > 95% PMN was resuspended in HBSS to a cell density of 6 .0x10 6 /ml . Cell viability was determined by trypan blue dye exclusion and was always > 95% . Superoxide anion assays . Levels of PMN superoxide dismutase inhibitable superoxide anion generation were assayed by the method described by Summersgill et al. 21 ConS, HiConS, or buffer was added to 0 .5 ml PMN (2 .0x106 /ml) in the presence of 240 µM cytochrome C (Type III, Sigma) . Superoxide dismutase (Sigma) was added to matched control tubes to a final concentration of 50 pg/ml . Following the addition of either 100 µl zymosan activated particles $ (ZAP) (Los Alamos Diagnostics), 3 .24x10_ M phorbol myristate acetate (PMA ; Sigma), 2 x 10 -6 M calcium ionophore (A 23187 ; Sigma) or 10 -6 M formyl-methionyl-leucyl-phenylalanine (fMLP ; Sigma), all tubes were incubated at 37°C for 90 min with 8 rpm rotation . Icecold HBSS was then added to each tube and all tubes were centrifuged at 6900xg for 5 min at 4°C . Absorbance of supernatants was determined at 550 nm . Values were corrected for amounts of cytochrome C reduced in the presence of superoxide dismutase and converted to nanomoles of reduced cytochrome C using the extinction coefficient E 550 = 2 .1 x 104 /M/cm . 23 Superoxide anion generation was calculated as nanomoles of superoxide anions released/ min/2 .0x10 6 cells, and all results were expressed as percentage of control (for each stimulus used) . In order to determine maximal reduction of ferricytochrome C, PMN superoxide anion generation was determined as described above, and sodium dithionite (Sigma ; 50 mg) was then added to all tubes and the absorbance was read at 550 nm . Total reduction of ferricytochrome C was calculated as described above . Chemotaxis . Agarose medium (Accurate Chemical and Scientific Corporation, Westbury, New York; 0 .024 g/ml) was prepared and 5-ml aliquots dispensed into 60x15 mm tissue culture plates (No . 3002, Falcon Plastics, Oxnard, California), and wells (3 mm in diameter) punched . Plates were preincubated in 5% CO 2 for 1 h at 37°C, followed by the aspiration of all moisture from the wells . HBSS (10 µl), PMN (10 µl) (pretreated with ConS, HiConS, YE broth medium or HBSS buffer for 30 min) and fMLP 10 - ' M (7 ul) were added to the appropriate wells . Plates were incubated for 2 h in 5% CO 2 at 37°C, fixed with methanol and Giemsa stained . PMN migration was measured by light microscopy at a magnification of 400x, using an ocular micrometer . Chemotactic differential was calculated as directed migration minus spontaneous migration and expressed as a percent of the HBSS-treated controls . Viability of PMN, before and after pretreatment, was determined by trypan blue exclusion and was always > 95% . Statistical analysis. Data were compared for significance by Student's t-test (paired) and expressed in P-values . Differences were considered significant at the level of P < 0 .05 .
This work was supported by Public Health Service grants Al 23567 and A122139 (to R .D .M .) from the National Institute of Allergy and Infectious Diseases .
References 1 . Chandler FW, Hicklin MD, Blackmon JA . Demonstration of the agent of legionnaires' disease in tissue . N Eng J Med 1977; 297 :1218-20 . 2 . Chandler FW, McDade JE, Hicklin MD, Blackmon JA . Pathologic findings in guinea pigs inoculated intraperitoneally with the Legionnaires' disease bacterium . Ann Intern Med 1979 ; 90 : 671-5 . 3 . Jacobs RF, Locksley RM, Wilson CB, Haas JE, Klebanoff SJ . Interactions of primate alveolar macrophages and Legionella pneumophila . J Clin Invest 1984 ; 73 :1515-23 . 4 . Horwitz MA. The legionnaire's disease bacterium (Legion/la pneumophi/a) inhibits phagosomelysosome in human monocytes . J Exp Med 1983; 158 : 2108-26 . 5 . Horwitz MA, Silverstein SC . Interaction of the legionnaires' disease bacterium (Legionella pneumophila ) with human phagocytes . 1 . L . pneumophila resists killing by polymorphonuclear lekocytes, antibody and complement . J Exp Med 1981 ; 153 : 386-97 . 6 . Friedman RL, Iglewski BH, Miller RD . Identification of a cytotoxin produced by Legionella pneumophila . Infect Immun 1980; 29 : 271-4 . 7 . Conlan JW, Baskerville A, Ashworth LAE . Separation of Legionella pneumophila proteases and
Inhibition of PMN by L . pneumophila exoproducts
8. 9. 10 .
11 .
12 . 13 . 14 . 15 . 16 .
17 . 18 . 19 .
20 . 21 . 22 . 23 .
125
purification of a protease which produces lesions like those of legionnaires' disease in guinea pig lung . J Gen Microbiol 1986 ; 132 : 1565-74 . Baine WB . Cytolytic exotoxin and phospholipase C activity in Legionella species . J Gen Microbiol 1985 ; 131 : 1383-91 . Thorpe TC, Miller RD . Extracellular enzymes of Legionella pneumophila. Infect Immun 1981 ; 33 : 6325. Saha AK, Dowling JN, LaMarco KL et al Properties of an acid phosphatase from Legionella micdadei which blocks superoxide anion production by human neutrophils . Arch Biochem Biophys 1985 ; 243 : 150-60 . Friedman RL, Lochner JE, Bigley RH, Iglewski BH . The effects of Legionella pneumophila toxin on oxidative processes and bacterial killing of human polymorphonuclear leukocytes . J Infect Dis 1982 ; 146 :328-34 . Lochner JE, Bigley RH, Iglewski BH . Defective triggering of polymorphonuclear leukocyte oxidative metabolism by Legionella pneumophila toxin . J Infect Dis 1985 ; 151 : 42-6 . Keen MG, Hoffman PS . Characterization of a Legionella pneumophila extracellular protease exhibiting hemolytic and cytotoxic activities . Infect Immun 1989; 57 : 732-8 . Kharazmi A, Doring G, Hoiby N, Valerius NH . Interaction of Pseudomonas aeroginosa alkaline protease and elastase with human polymorphonuclear leukocytes in vitro . Infect Immun 1984; 43 : 161-5 . Horwitz MA . Characterization of avirulent mutant Legionella pneumophila that survive but do not multiply within human monocytes . J Exp Med 1987 ; 166 :1310-28 . Williams A, Baskerville A, Dowsett AB, Conlan JW. Immunocytochemical demonstration of the association between Legionella pneumophila, its tissue-destructive protease, and pulmonary lesions in experimental legionnaires' disease . J Pathol 1987 ; 153 : 257-64 . Baskerville A, Conlan JW, Ashworth LAE, Dowsett AB . Pulmonary damage caused by a protease from Legionella pneumophila . Br J Exp Pathol 1986 ; 67 : 527-36 . Quinn FD, Keen MG, Tompkins LS . Genetic, immunological and cytotoxic comparisons of Legionella proteolytic activities . Infect Immun 1989 ; 57 : 2719-2725 . Blander SJ, Horwitz MA . Vaccination with the major secretory protein of Legionella pneumophila induces cell-mediated and protective immunity in a guinea pig model of legionnaires' disease . J Exp Med 1989 ; 169 :691-705 . Rechnitzer C, Diamant M, Pedersen BK . Inhibition of human natural killer cell activity by Legionella pneumophila protease . Eur J Clin Microbiol Infect Dis 1989 ; 8 : 989-92 . Summersgill JT, Raff MJ, Miller RD . Interactions of virulent and avirulent Legionella pneumophila with human polymorphonuclear leukocytes . Microbial Pathogenesis 1988 ; 5 : 41-7 . Thompson MR, Miller RD, Iglewski BH . In vitro production of an extracellular protease by Legionella pneumophila . Infect Immun 1981 ; 34 : 299-302 . Massey V . The microestimation of succinate and the extinction coefficient of cytochrome C . Biochem Biophys 1959 ; 34 : 255-6 .
Short communications Inhibition of polymorphonuclear leukocyte function by Legionella pneumophila exoproducts Narendra N . Sahney,' Beverly C . Lambe,' * James T . Summersgi112 and Richard D . Miller 2 t 'Department of Microbiology and Immunology, School of Medicine, and 2Section of Infectious Diseases, Department of Medicine, University of Louisville, Louisville, Kentucky 40292, U .S .A . (Received February 22, 1990; accepted in revised form April 30, 1990)
Sahney, N . N . (Dept of Microbiology and Immunology, School of Medicine, University of Louisville, Louisville, Kentucky 40292, U .S .A .), B . C . Lambe, J . T . Summersgill and R . D . Miller . Inhibition of polymorphonuclear leukocyte function by Legionella pneumophila exoproducts . Microbial Pathogenesis 1990 ; 9 : 117-125 . Total exoproducts (relative molecular mass > 10000) from wild-type strains of Legionella pneumophila markedly inhibited human polymorphonuclear leukocyte (PMN) superoxide anion generation, at sub-lethal concentrations, in response to four stimuli [1 .7, 0, 0 .6 and 3 .4% of control for zymosan activated particles (ZAP), phorbol myristate acetate (PMA), calcium ionophore (A 23187), and formyl-methionyl-leucyl-phenylalanine (fMLP), respectively] . PMN chemotaxis towards fMLP and spontaneous migration, were also dramatically inhibited (2 .8 and 2 .9% of buffer-treated controls, respectively) . In contrast, total exoproducts from the cas-1 strain of L . pneumophila, a protease-deficient mutant generated by ethyl methane sulfonate mutagenesis, failed to inhibit PMN superoxide production in response to ZAP and PMA and only partially inhibited PMN response to A 23187 and fMLP . PMN spontaneous migration was unaffected by treatment with total exoproducts from the mutant, while directed chemotaxis was partially inhibited (51 .4%) . These data demonstrated that L . pneumophila total exoproducts, primarily protease had significant inhibitory effects on normal PMN function and may play an important contributory role in the pathogenesis of legionnaire's disease . Key words: Legionella pneumophila ; protease ; PMN ; superoxide anion generation ; chemotaxis .
Introduction Legionella pneumophila is the etiological agent of legionnaire's disease (LD), most often presenting as a life-threatening bronchopneumonia . The pathogenesis of LD involves the intracellular multiplication of the organism in alveolar macrophages as demonstrated in lung specimens from patients with LD,' in the lungs of experimentally infected guinea pigs' and within in vitro cultured alveolar macrophages .' A major feature of L . pneumophila is its ability to inhibit phagosome-lysosome fusion in permissive monocyte cultures in vitro' and its resistance to killing by human polymorphonuclear
Present address : Johns Hopkins Medical Institutions, Meyer Building, Room B 1-193, 600 N . Wolfe St ., Baltimore, MD 21205, U .S .A. t Author to whom correspondence should be addressed . 0882-4010/90/080117+09 $03 .00/0
© 1990 Academic Press Limited
1 18
N . N . Sahney et al.
Table 1 Enzymatic activity in culture supernatants of Legionella pneumophila strains Knoxville-1, cas-1 mutant and Ver5e Enzyme activity Strains
Proteases
PL b
AP`
RNAse`
Knoxville-1 cas-1 Ver5e
353 .3 3 .1 360 .0
14 15 23
0 .299 0 .299 0 .240
12 15 18
a
Units of protease activity ; hide powder azure substrate . Phospholipase A/esterase ; zones of hydrolysis in Tween40 agar (mm diameter) . `Acid phosphatase ; units of activity, p-nitrophenylphosphate substrate . 'Ribonuclease ; zones of hydrolysis in RNA agar (mm diameter) . b
leukocytes (PMN) .' The mechanisms by which these responses are induced and the virulence factor(s) involved have not been elucidated . L . pneumophila produces a wide array of extracellular enzymes and toxins (exoproducts) including a cytotoxin s several proteases,' phospholipase C, 8 phospholipase A/esterase, 9 acid phosphatases, 10 amylase,' ribonuclease s and beta lactamase .' It has previously been shown that the cytotoxin 11 - 12 and acid phosphatase 10 interfere with the oxidative burst of PMN . Recently, purified protease from L . pneumophila has been demonstrated to have haemolytic and cytotoxic activity . 13 Taken as a whole, this information suggests that these exoproducts may play a role in the pathogenesis of LD . In the present study, ethyl methane sulfonate mutagenesis was employed and a protease-deficient mutant of L . pneumophila was isolated and characterized . This study compared the protease-deficient mutant and the wild-type strain by examining the effects of their total exoproducts (relative molecular mass > 10000) on normal PMN superoxide anion generation and chemotaxis . Results Isolation and characterization of mutant Mutagenesis was carried out by ethyl methane sulfonate (EMS) treatment and a total of 18500 colonies from the virulent L . pneumophila Ver5e and avirulent Knoxville-1 strains were screened for the absence of proteolytic activity on casein-containing agar plates . One mutant (cas-1) was isolated from the Knoxville-1 strain exhibiting an apparent mutation rate of 0 .007%, whereas no mutants were isolated from the virulent Ver5e strain . A total of 7500 colonies from the two strains were screened for loss of phospholipase A/esterase activity on Tween-40 agar plates, however no mutants deficient in this exoproduct were isolated . The cas-1 mutant was stable after 20 weekly passages on BCYE with no apparent reversions to wild-type protease production . Enzymatic activities assayed in culture supernatants (Table 1), or the 20 enzymes of the API-ZYM system (Analytab Products, Plainview, New York ; positive reactions for alkaline phosphatase, acid phosphatase, butyrate esterase, caprylate esterase/lipase, leucine aminopeptidase, valine aminopeptidase, cystine aminopeptidase and phosphoamidase) failed to demonstrate any other major difference between the cas-1 mutant and wild-type avirulent Knoxville-1 strain (or the virulent Ver5e strain) . The
1 19
Inhibition of PMN by L . pneumophila exoproducts
140 •
(a )
120 •
100 80 60 40 20 0 Z
0-
2 Q a NJ
U)
U)
0
0
0
U 2
Wild-type
z a a a NI
N
z
0
C1.
0
a a
U
N
0 0
0 U
•
z C U) • 2 o a a_ U
2
Wild-type
Protease mutant
Protease mutant
140 0 0
120
• 100 0 0,0 80
v• x•
60 40
aL CL
•
20 0 z
f a
a J F
U, 0 0
0
v
z a 2 J a -2
U) C U
Protease mutant Wild-type Fig . 1 . Effects of ConS from wild-type and protease-deficient mutant strains of Legionella pneumophila and HiConS, on superoxide anion generation by PMN in response to (a) ZAP ; (b) PMA ; (c) A23187 ; (d) fMLP . Superoxide anion generation was determined by adding ConS to 2 .0x106 PMN/ml and stimulating with ZAP, PMA, A23187, of fMLP (details in Materials and methods) . Data shown represent means and standard deviations of six and three experiments with ConS of wild-type and mutant strains, respectively, and expressed as a percentage of each stimulus (0) . Viability of PMN treated with ConS was > 95% .
only significant difference was that the cas-1 mutant was a low protease producer (0 .9% of wild-type) . Superoxide anion generation Concentrated supernatants (ConS) from the mutant and wild-type strains were added to washed PMN which were then exposed to one of four stimuli . In all experiments, ConS was added in an amount equivalent to 75 units of phospholipase A/esterase activity (used as an internal standard) with the average amount of protease present in the ConS being 1316 units for the Knoxville-1 and Ver5e strains and only 4 .2 units for the mutant . Additionally, the amount of ConS used was based on concentrations that did not affect overall PMN viability, as determined by trypan blue dye exclusion . Controls consisted of PMN treated with each stimuli (represented as the 100% control) and buffer alone . As shown in Fig . 1, PMN exposed to the wild-type ConS and then stimulated with zymosan activated particles (ZAP), phorbol myristate acetate (PMA), calcium ionophore (A23187), or formyl-methionyl-leucyl-phenalanine (fMLP) showed almost total inhibition of superoxide anion generation (1 .7, 0, 0 .6 and 3 .4% of controls
1 20
N . N . Sahney et al.
respectively, P < 0 .05) . In contrast, PMN exposed to ConS from the cas-1 mutant and stimulated as described showed only a modest inhibition of superoxide anion levels with A 23187 and fMLP (61 .5 and 49 .5% of controls respectively, P < 0 .05), and no statistically significant inhibition in superoxide anion generation with ZAP and PMA (61 .1 and 89% of controls respectively, P < 0 .5) . No significant differences were noted between the heat-inactivated ConS (HiConS) and untreated controls with any of the four stimuli used (87 .5, 64 .2, 78 .7 and 116% of ZAP, PMA, A 23187 and fMLP controls respectively, P < 0 .5) . The inhibitory effect of ConS (wild-type) on superoxide anion generation was also partially reversible, since PMN preincubated for 30 min with ConS, then washed twice in HBSS and stimulated in the presence of cytochrome C showed a significant recovery (78 .5% of control) . However, 10-fold higher concentrations of ConS resulted in greater inhibition (33 .6% of control) . Measurement of maximal reduction of ferricytochrome C by the addition of sodium dithionite revealed that inhibition of PMN superoxide anion generation upon exposure to ConS from wild-type was not due to substrate limitation . Total reduction of cytochrome C exceeded the levels of superoxide dismutase-inhibitable superoxide anion generated in all cases (data not shown) . Chemotaxis Under agarose chemotaxis by PIVIN was determined using fM LP as the chemoattractant . As shown in Fig . 2(a), PMN pretreated for 30 min with ConS of the wild-type strains showed a dramatic decrease in chemotactic ability (chemotactic differential 2 .8% of buffer control, P < 0 .05) . In contrast, PMN pretreated with ConS from the mutant were considerably less inhibitory (chemotactic differential 51 .4% of control, P < 0 .05) . Pretreatment of PMN with heat-inactivated ConS also showed a similar partial inhibition (chemotactic differential 47 .8% of control, P < 0 .05) . As a control, pretreatment with YE broth medium alone had no effect on chemotactic differential (99 .6% of buffer control) . Spontaneous migration of PMN was also inhibited markedly when pretreated with ConS from wild type strains as shown in Fig . 2(b) (2 .9% of control, P < 0 .05) . Interestingly, PMN pretreated with ConS from the mutant had no significant inhibitory effect on spontaneous migration (96 .0% of control, P < 0 .5), while the HiConS showed a small but significant inhibition (84 .1% of buffer control, P < 0 .05) .
Discussion Based on enzymatic activities present in culture supernatants, the cas-1 mutant used in this study closely resembled the parental strains, with the major difference being that the mutant had low levels of protease activity in culture supernatants (none in cell lysates) . This suggested an alteration in some aspect of protease production, and not in secretion . Virulence testing of cas-1 mutant was not reported as part of the present study because the mutant was derived from an avirulent parent (Knoxville1) . However, a naturally occurring (environmental) protease-deficient strain of L . pneumophila isolated by our laboratory has been found to be avirulent for guinea pigs (unpublished data) . Since L . pneumophi/a produces several extracellular enzymes and toxins, the aim of this study was to examine the effects of these total exoproducts on two parameters of PMN functions, the oxidative and chemotactic response to exogenous stimuli . The exoproducts were tested in concentrated form (relative molecular mass > 10000) which excluded the low molecular weight cytotoxin . In addition, the concentrations
1 21
Inhibition of PMN by L . pneumophila exoproducts
120 100 80 60 40 20 0
II Buffer Cons
Buffer Cans HiConS
Medium
120
T
O U
100
O
o 80 0
2 T E
60
0
40
G7 0 0 `0
n
20 0 Buffer Cons Wild-type
Buffer ConS HiConS
Medium
Protease mutant
Fig . 2 . Effects of ConS of wild type and mutant strains of L . pneumophi/a, HiConS and YE broth medium on PMN (a) chemotaxis ; (b) spontaneous migration . PMN were treated with either ConS, HiConS, buffer or medium for 30 min . Under agarose chemotaxis was performed in response to fMLP 10 -7 M . Chemotaxis differential was determined by directed minus spontaneous migration of PMN . Data shown represent mean and standard deviations of nine experiments performed in triplicate and expressed as a percentage of buffer treated controls (0) . Viability of pretreated PMN was > 95% .
used were chosen so as to be sub-lethal for PMN viability and thus may represent more relevant concentrations actually observed in vivo . Analysis of superoxide anion generation data revealed that ConS from wild-type strains completely inhibited PMN superoxide anion generation in response to four stimuli used (ZAP, PMA, A 23187 and fMLP) . This inhibitory effect was not due to limitations in substrate availability as determined by excess reducible cytochrome C present after the addition of sodium dithionite . The amount of ConS used minimally affected the cytochrome C and thus was not sufficient to explain the complete inhibition . Also, the inhibitory effect of ConS was only partially reversible following washing . No statistically significant inhibition of PMN superoxide anion generation was seen after treatment with ConS from the mutant for two of the stimuli (ZAP and PMA), suggesting that the protease can completely diminish PMN oxidative response at these sub-lethal concentrations . Saha et a/. 1 ° demonstrated that an acid phosphatase
1 22
N . N . Sahney et al .
isolated from L . micdadei could block PMN superoxide ani, on generation . However, in the present study both wild-type and mutant strains contained equivalent amounts of acid phosphatase . In a similar study the exoprotease from Pseudomonas aeroginosa has also been shown to inhibit PMN superoxide anion generation ." Partial inhibition of superoxide anion generation by ConS from mutant when A 23187 and fMLP were used as stimuli suggested that other (presumably heat-labile) exoproducts also contributed to the inhibitory effect on PMN function . The differences between the four stimuli in PMN superoxide anion generation following treatment with ConS from the mutant is not totally unexpected, since the four stimuli used in the study can activate PMN via different pathways . Spontaneous migration of PMN was also severely inhibited (3% of control) by ConS from wild-type, but was minimally inhibited with ConS from mutant (96% of control), which indicated that the inhibitory effect on PMN spontaneous migration was solely due to the protease . The ability of L . pneumophila protease to completely inhibit PMN spontaneous migration suggested that protease may interfere not only with sensory and transducing mechanisms (possibly cleaving fMLP receptors), but that it may also alter some aspect of the effector mechanisms involved in PMN locomotion (at sublethal concentrations) . The PMN chemotactic response to the chemoattractant fMLP was also dramatically inhibited by ConS from the wild-type strains (chemotactic differential 2 .8% of control), while this inhibition was only partial with both ConS from mutant and HiConS (chemotactic differential 51 .4 and 47 .8% of controls respectively) . This suggested that other components present in the exoproducts (presumably heatstable) have inhibitory effects on PMN chemotactic function as well . In a similar study, complete inhibition of PMN chemotaxis has been reported to occur following treatment with an exoprotease from Pseudomonas aeroginosa . 14 Overall, analysis of the data in this study suggested that L . pneumophila exoproducts inhibit PMN superoxide anion generation and chemotaxis . At the sub-lethal concentrations used, the protease had a major role in the inhibitory effect on these PMN functions . Of course, the protease is the major secretory protein produced by this bacterium 15 and would be present in the highest concentration in the ConS relative to the other secretory products . Nevertheless, considerable evidence has been accumulating concerning the role of protease in the pathogenesis of LD . Studies have documented the production of protease by L . pneumophila in the lungs of guinea pigs with experimentally induced LD . 16 Baskerville et al." and Conlan et al.' have reported that purified protease aerosolized into the lungs of guinea pigs induced lesions resembling those seen in experimentally induced guinea pig legionellosis and also from lung biopsy of humans with naturally acquired disease . Antibody to the protease has been detected in convalescent-phase sera from LD patients, 18 and vaccination of guinea pigs with the protease has been shown to induce protective immunity . 19 Purified protease has also been shown to inhibit human natural killer (NK) cell activity . 20 The results of the present study support the hypothesis that L . pneumophila exoproducts (particularly protease) may play an important contributory role in the pathogenesis of LD by interfering with phagocytic responses and in resistance to killing in vivo by PMN and possibly macrophages . The effect of purified L . pneumophila protease on those same PMN and macrophage functions is currently being investigated .
Materials and methods Bacteria/ strains and media . Legione/la pneumophi/a strain Ver5e (serogroup-1) was isolated from an environmental sample (cooling tower water) on buffered charcoal yeast extract (BCYE) agar medium prepared from Legionella Agar Base (Difco Laboratories, Detroit, Michigan),
Inhibition of PMN by L . pneumophila exoproducts
123
supplemented with 0 .01% (w/v) L-cysteine and 0 .05% ferric pyrophosphate . This isolate has been well characterized in our laboratory27 and is virulent for guinea pigs based on an LD 50 of < 10' organisms by intraperitoneal injection . L . pneumophi/a strain Knoxville-1 (serogroup-1) is an extensively laboratory-adapted (avirulent) isolate originally obtained from R . Weaver (Centers for Disease Control, Atlanta, Georgia) . All bacteria were stored at -70°C in Tryptic soy broth (Difco) containing 20% (w/v) glycerol . Casein agar (screening medium for protease) contained 1 .0% yeast extract, 1 .7% Difco-agar, 1 .0% ACES, 1 .0% casein, 0 .5% corn starch, supplemented with 0 .04% L-cyteine and 0.025% ferric pyrophosphate . Tween-40 agar (screening medium for phospholipase A/esterase) contained 1 .0% yeast extract, 1 .7% Difco-agar, 1 .0% ACES, 0 .075% Tween 40, 2 .0% charcoal, supplemented with 0 .04% L-cysteine and 0 .025% ferric pyrophosphate . Ethyl methane su/fonate mutagenesis . L. pneumophila strains Knoxville-1 and Ver5e were grown in yeast extract (YE) broth containing 1 .0% yeast extract (Difco) and 1 .0% ACES (Research Organics, Cleveland, Ohio) supplemented with L-cysteine and ferric pyrophosphate as described above . Cultures were grown to mid-to-late exponential phase, centrifuged at 10000xg for 10 min at 25°C and resuspended in 10 ml YE broth . Ethyl methane sulfonate was added to the suspension at a final concentration of 15 µg/ml and the culture was incubated for 1 h at 35°C on a rotary shaker . Cells were centrifuged as described above, washed once in YE broth and resuspended in 20 ml YE broth . A portion of these cells was diluted in distilled water and plated on casein and Tween-40 media so as to obtain approximately 100 colonies per plate (10x150 mm Petri dish) . An additional 20 ml of YE broth was added to the remaining cells, incubated for 1 h at 35°C, and then used directly to inoculate fresh YE broth . These cultures were incubated at 37°C with shaking overnight, then diluted and plated onto casein media and Tween-40 media . Casein plates were incubated at 37°C in 5% CO 2, examined daily for 4 days, and colonies lacking a zone of white precipitate were isolated and characterized . Tween-40 agar plates were incubated for 7-10 days at 37°C in 5% CO 2 and examined daily for colonies lacking an oily sheen on the surface of agar plates . Enzyme assays . Protease activity was measured by the method described by Thompson et a/. 22 Briefly, 100 µl of sample was added to 20 mg hide powder azure (Sigma Chemical, St Louis, Missouri) suspended in 5 ml of 100 mm sodium phosphate buffer (pH 6 .0) and incubated at 35°C for 60 min on a rotary shaker . Samples were filtered (Whatman number 1 filter) and the absorbance of the soluble fraction was measured at 595 nm in an Ultrospec II (LKB, Biochrom, Cambridge, U .K .) spectrophotometer . Each unit was arbitrarily defined as proteolysis resulting in a change of absorbance of 0 .001 per h . The phospholipase A/esterase of L . pneumophila has activity on phospholipids (e .g . phosphatidylinositol), as well as on general esterase substrates (T . C . Thorpe and R . D . Miller, unpublished data) . In this study, phospholipase A/esterase activity was determined by adding 20 µI of sample to 2 .5 ml 10 µnn Tris-HCI buffer (ph 8 .0) containing 0 .2% sodium deoxycholate . p-Nitrophenyl caprylate (PNP-caprylate ; Sigma ; 20 µl) was then added as an esterase substrate and incubated for 5 min at 37°C . Dimethyl sulfoxide (DMSO ; 1 ml) was added and the absorbance read at 415 nm . One unit of activity was defined as 1 nmcl of PNP released/min . Phospholipase A/esterase activity was also measured using an agar plate assay containing 0 .2% deoxycholate, 1 .0% Tween-40 (esterase substrate), 0 .01% sodium azide, 1 .7% Difco-agar in 0 .01 M Tris-HCI buffer (pH 8.0) . Zones of oily sheen radiating from wells containing the enzyme were then measured . Acid phosphatase activity was assayed by the method of Saha et al." Briefly, samples were incubated with PNP-phosphate in 0 .15 M citrate buffer (pH 4 .8) for 3 h . Sodium hydroxide was added and the absorbance measured at 420 nm . Ribonuclease activity was detected on plates containing 0 .2% ribonucleic acid (Type II-s from Toru/a yeast cells) in 0 .1 M sodium phosphate buffer (pH 6 .0) and 1 .7% Difco-agar. After 96 h incubation at 35°C, plates were flooded with 1 M HCI and zones of clearing were measured . Concentrated supernatants (ConS) . Each isolate was inoculated into 100 ml YE broth at 30 Klett units and incubated in a rotary shaker at 37°C . After 18 h, cells were removed by centrifugation at 10000 xg for 15 min at 4°C . Culture supernatant (50 ml) was then concentrated to 5 ml by ultrafiltration in a Centricell 60 (Polysciences, Warrington, Pennsylvania) with a 10000 relative molecular mass cut-off . Half of the concentrated supernatant (ConS) was heattreated by autoclaving for 90 min (HiConS), and any precipitate formed was separated by centrifugation and discarded .
124
N . N . Sahney et al.
Neutrophil isolation. Whole blood (75 ml) was collected from healthy volunteers in tubes containing heparin (10 units/ml) . PMN were isolated by buoyant density centrifugation using Neutrophil Isolation Medium (Los Alamos Diagnostics, Los Alamos, New Mexico) and washed twice in Hanks balanced salt solution (HBSS) . The final pellet containing > 95% PMN was resuspended in HBSS to a cell density of 6 .0x10 6 /ml . Cell viability was determined by trypan blue dye exclusion and was always > 95% . Superoxide anion assays . Levels of PMN superoxide dismutase inhibitable superoxide anion generation were assayed by the method described by Summersgill et al. 21 ConS, HiConS, or buffer was added to 0 .5 ml PMN (2 .0x106 /ml) in the presence of 240 µM cytochrome C (Type III, Sigma) . Superoxide dismutase (Sigma) was added to matched control tubes to a final concentration of 50 pg/ml . Following the addition of either 100 µl zymosan activated particles $ (ZAP) (Los Alamos Diagnostics), 3 .24x10_ M phorbol myristate acetate (PMA ; Sigma), 2 x 10 -6 M calcium ionophore (A 23187 ; Sigma) or 10 -6 M formyl-methionyl-leucyl-phenylalanine (fMLP ; Sigma), all tubes were incubated at 37°C for 90 min with 8 rpm rotation . Icecold HBSS was then added to each tube and all tubes were centrifuged at 6900xg for 5 min at 4°C . Absorbance of supernatants was determined at 550 nm . Values were corrected for amounts of cytochrome C reduced in the presence of superoxide dismutase and converted to nanomoles of reduced cytochrome C using the extinction coefficient E 550 = 2 .1 x 104 /M/cm . 23 Superoxide anion generation was calculated as nanomoles of superoxide anions released/ min/2 .0x10 6 cells, and all results were expressed as percentage of control (for each stimulus used) . In order to determine maximal reduction of ferricytochrome C, PMN superoxide anion generation was determined as described above, and sodium dithionite (Sigma ; 50 mg) was then added to all tubes and the absorbance was read at 550 nm . Total reduction of ferricytochrome C was calculated as described above . Chemotaxis . Agarose medium (Accurate Chemical and Scientific Corporation, Westbury, New York; 0 .024 g/ml) was prepared and 5-ml aliquots dispensed into 60x15 mm tissue culture plates (No . 3002, Falcon Plastics, Oxnard, California), and wells (3 mm in diameter) punched . Plates were preincubated in 5% CO 2 for 1 h at 37°C, followed by the aspiration of all moisture from the wells . HBSS (10 µl), PMN (10 µl) (pretreated with ConS, HiConS, YE broth medium or HBSS buffer for 30 min) and fMLP 10 - ' M (7 ul) were added to the appropriate wells . Plates were incubated for 2 h in 5% CO 2 at 37°C, fixed with methanol and Giemsa stained . PMN migration was measured by light microscopy at a magnification of 400x, using an ocular micrometer . Chemotactic differential was calculated as directed migration minus spontaneous migration and expressed as a percent of the HBSS-treated controls . Viability of PMN, before and after pretreatment, was determined by trypan blue exclusion and was always > 95% . Statistical analysis. Data were compared for significance by Student's t-test (paired) and expressed in P-values . Differences were considered significant at the level of P < 0 .05 .
This work was supported by Public Health Service grants Al 23567 and A122139 (to R .D .M .) from the National Institute of Allergy and Infectious Diseases .
References 1 . Chandler FW, Hicklin MD, Blackmon JA . Demonstration of the agent of legionnaires' disease in tissue . N Eng J Med 1977; 297 :1218-20 . 2 . Chandler FW, McDade JE, Hicklin MD, Blackmon JA . Pathologic findings in guinea pigs inoculated intraperitoneally with the Legionnaires' disease bacterium . Ann Intern Med 1979 ; 90 : 671-5 . 3 . Jacobs RF, Locksley RM, Wilson CB, Haas JE, Klebanoff SJ . Interactions of primate alveolar macrophages and Legionella pneumophila . J Clin Invest 1984 ; 73 :1515-23 . 4 . Horwitz MA. The legionnaire's disease bacterium (Legion/la pneumophi/a) inhibits phagosomelysosome in human monocytes . J Exp Med 1983; 158 : 2108-26 . 5 . Horwitz MA, Silverstein SC . Interaction of the legionnaires' disease bacterium (Legionella pneumophila ) with human phagocytes . 1 . L . pneumophila resists killing by polymorphonuclear lekocytes, antibody and complement . J Exp Med 1981 ; 153 : 386-97 . 6 . Friedman RL, Iglewski BH, Miller RD . Identification of a cytotoxin produced by Legionella pneumophila . Infect Immun 1980; 29 : 271-4 . 7 . Conlan JW, Baskerville A, Ashworth LAE . Separation of Legionella pneumophila proteases and
Inhibition of PMN by L . pneumophila exoproducts
8. 9. 10 .
11 .
12 . 13 . 14 . 15 . 16 .
17 . 18 . 19 .
20 . 21 . 22 . 23 .
125
purification of a protease which produces lesions like those of legionnaires' disease in guinea pig lung . J Gen Microbiol 1986 ; 132 : 1565-74 . Baine WB . Cytolytic exotoxin and phospholipase C activity in Legionella species . J Gen Microbiol 1985 ; 131 : 1383-91 . Thorpe TC, Miller RD . Extracellular enzymes of Legionella pneumophila. Infect Immun 1981 ; 33 : 6325. Saha AK, Dowling JN, LaMarco KL et al Properties of an acid phosphatase from Legionella micdadei which blocks superoxide anion production by human neutrophils . Arch Biochem Biophys 1985 ; 243 : 150-60 . Friedman RL, Lochner JE, Bigley RH, Iglewski BH . The effects of Legionella pneumophila toxin on oxidative processes and bacterial killing of human polymorphonuclear leukocytes . J Infect Dis 1982 ; 146 :328-34 . Lochner JE, Bigley RH, Iglewski BH . Defective triggering of polymorphonuclear leukocyte oxidative metabolism by Legionella pneumophila toxin . J Infect Dis 1985 ; 151 : 42-6 . Keen MG, Hoffman PS . Characterization of a Legionella pneumophila extracellular protease exhibiting hemolytic and cytotoxic activities . Infect Immun 1989; 57 : 732-8 . Kharazmi A, Doring G, Hoiby N, Valerius NH . Interaction of Pseudomonas aeroginosa alkaline protease and elastase with human polymorphonuclear leukocytes in vitro . Infect Immun 1984; 43 : 161-5 . Horwitz MA . Characterization of avirulent mutant Legionella pneumophila that survive but do not multiply within human monocytes . J Exp Med 1987 ; 166 :1310-28 . Williams A, Baskerville A, Dowsett AB, Conlan JW. Immunocytochemical demonstration of the association between Legionella pneumophila, its tissue-destructive protease, and pulmonary lesions in experimental legionnaires' disease . J Pathol 1987 ; 153 : 257-64 . Baskerville A, Conlan JW, Ashworth LAE, Dowsett AB . Pulmonary damage caused by a protease from Legionella pneumophila . Br J Exp Pathol 1986 ; 67 : 527-36 . Quinn FD, Keen MG, Tompkins LS . Genetic, immunological and cytotoxic comparisons of Legionella proteolytic activities . Infect Immun 1989 ; 57 : 2719-2725 . Blander SJ, Horwitz MA . Vaccination with the major secretory protein of Legionella pneumophila induces cell-mediated and protective immunity in a guinea pig model of legionnaires' disease . J Exp Med 1989 ; 169 :691-705 . Rechnitzer C, Diamant M, Pedersen BK . Inhibition of human natural killer cell activity by Legionella pneumophila protease . Eur J Clin Microbiol Infect Dis 1989 ; 8 : 989-92 . Summersgill JT, Raff MJ, Miller RD . Interactions of virulent and avirulent Legionella pneumophila with human polymorphonuclear leukocytes . Microbial Pathogenesis 1988 ; 5 : 41-7 . Thompson MR, Miller RD, Iglewski BH . In vitro production of an extracellular protease by Legionella pneumophila . Infect Immun 1981 ; 34 : 299-302 . Massey V . The microestimation of succinate and the extinction coefficient of cytochrome C . Biochem Biophys 1959 ; 34 : 255-6 .
