Spontaneous and lipopolysaccharide-induced expression of procoagulant activity by equine lung macrophages in comparison with blood monocytes and blood neutrophils

Veterinary Immunology and Immunopathology, 29 ( 1991 ) 295-312

295

Elsevier Science Publishers B.V., Amsterdam

Spontaneous and lipopolysaccharide-induced expression of procoagulant activity by equine lung macrophages in comparison with blood monocytes and blood neutrophils G. Griinig a'~, C. Hulliger a, C. Winder a, M. Hermann b, T.W. Jungi c and R. v o n F e l l e n b e r g a'2 aDivision of Applied Veterinary Physiology of the Institute of Veterina~ PhysioloD, and bDepartment of Veterinary Medicine, University of Zfirich, Winterthurerstrasse260, 805 7 Ziirich, Switzerland Clnstitute of Veterinary Virology, University of Berne, Berne, Switzerland (Accepted 26 November 1990)

ABSTRACT Griinig, G., Hulliger, C., Winder, C., Hermann, M., Jungi, T.W. and von Fellenberg, R., 1991. Spontaneous and lipopolysaccharide-induced expression of procoagulant activity by equine lung macrophages in comparison with blood monocytes and blood neutrophils. Vet. hnmunol, hmnunopathol., 29:295-312. The procoagulant activity (PCA) associated with equine bronchoalveolar lavage cells was determined and compared with that expressed by peripheral b!oo.d mononuclear cells and neutrophils. Lung cell preparations from horses affected with chronic pulmonary disease were included in all experiments and there was no difference in the qualitative type of response compared with lung cells which were obtained from healthy horses. Significant amounts of PCA were expressed by cells freshly procured from bronchoalveolar lavages of healthy and diseased horses. When adherent lung cells were kept in culture for some time, cell-associated PCA slightly decreased within 4 h, reached its lowest point after approximately 24 h and rose again during the second week of culture. In contrast, freshly isolated blood mononuclear cells or neutrophils expressed little PCA. Following culture for 24 h, mononuclear cells began to express increased PCA levels. Both cultivated lung cells (comprised mainly on alveolar macrophages) and blood mononuclear cells responded to LPS by dramatically increased PCA expression, whereas neutrophils showed a small augmentation of PCA on LPS stimulation. Fresh mononuclear cells and cultivated lung cells differed in their PCA response to LPS in several respects. Blood mononuclear cells were more sensitive to LPS than lung macrophages and responded to a 100fold lower LPS concentration than the latter. Mononuclear cell-associated PCA peaked 4 h after stimulation whereas that of cultured macrophages continued to increase up to 24 h after stimulation. Lung macrophages cultured in adherence responded to LPS stimulation with a much higher PCA increase than macrophages cultured in suspension, in teflon containers. However, the culture vessel did not influence the PCA expressed by unstimulated cells. PCA expression depended to a large extent on Ipresent address: J.A. Baker Institute for Animal Health, New York State College of Veterinary Medicine, Cornell University, Ithaca, NY 14853, USA. 2Author to whom correspondence should be addressed.

0165-2427/91/$03.50 © 1991 Elsevier Science Publishers B.V. All rights reserved.

296

G. GRONIGET AL

transcription and translation, as evidenced by a 60-85% reduction of PCA in cycloheximide- or actinomycin D-treated, LPS-stimulated lung macrophages. PCA was largely cell-associated; only a small proportion of cell-associated PCA was shed into the medium. The PCA associated with mononuclear cells and with lung macrophages was tissue factor because of its dependence on clotting factor Vll and its independence from clotting factor VIII. The expression of PCA by freshly isolated cells, the lower sensitivity to LPS, and the loss of PCA in the first 24 h of cultivation are indicative of in vivo activation of lung macrophages. ABBREVIATIONS HBSS, Hank's balanced salt solution; LPS, lipopolysaccharide; PCA, procoagulant activity.

INTRODUCTION

Cell-free supernatants of equine respiratory secretions contain procoagulant activity (PCA) which is correlated with the severity of chronic pulmonary disease and with the quantity of neutrophils in the secretion specimens (Griinig et al., 1988). As in man and in experimental animals (Sitrin et al., 1983; Chapman et al., 1985), equine lung macrophages, cell-free bronchoalveolar lavage fluids (Griinig et al., 1990), and blood monocytes (Henry and Moore, 1988) express PCA. However, a kinetic analysis, a comparison between stimulated and unstimulated cells and between alveolar lavage cells and peripheral blood cells had not been performed in the equine species. In the present report, we assess the noninduced and the lipopolysaccharide (LPS)-induced PCA of freshly isolated and of cultured alveolar lavage cells, and we compare the PCA response with that of peripheral blood mononuclear ceils and neutrophiis immediately after isolation and after short-term culture. Particular attention is given to the LPS-induced PCA response by cells cultured under either adherent (polystyrene) or nonadherent (teflon) culture conditions, and it is confirmed that the cell-associated PCA represents, to a large extent, de novo synthesized tissue factor, as reported in a number of other species. The observation that lung macrophages from healthy horses and horses with chronic bronchitis express considerable levels of PCA suggests that these cells had been activated in some way in vivo. MATERIALS AND METHODS

Horses

Bronchoalveolar lavage cells were obtained from 24 medical or surgical patients referred to the veterinary clinic and from experimental horses. Seven animals had no signs of respiratory disease, and 17 horses were affected with chronic pulmonary disease. Blood cells were isolated from clinically healthy horses referred to the veterinary clinic.

PROCOAGULANT ACTIVITY OF EQUINE LUNG MACROPHAGES

297

Reagents Stock solutions of LPS from Escherichia coli 055 :B5 (Difco) were prepared in distilled water, aliquotted and kept frozen ( - 70°C) until use. Actinomycin D and cycloheximide were purchased from Sigma. Concentrations of LPS determined (Limusate test, Haemachem Inc., St. Luis) in buffers, media and sera were less than l 0 pg/ml.

Cell isolation Bronchoalveolar macrophages Thirty-four bronchoalveolar lavages were performed in 24 horses with two or three 250 ml aliquots of prewarmed (37°C) pyrogen-free 0.9% saline per lung using a modification (Griinig et al., 1990) of the technique described by Viel (1986).

PCA (activity of freshly isolated cells - 100) 3 1 0 0 -1


730 -


50O

400

300 ,^.

< ~. /

100

.

,ii"

2 0 0 ~,

i

•

,.

.

" "~7 ......

4 hours

24 hours

,

1 week

2 weeks

time of in vitro cultivation

Fig. 1. Expression of PCA by unstimulated, control lung macrophages. Lung macrophages were isolated from 26 horses and cultivated on plastic dishes for up to 2 weeks. Clotting times were converted to U of PCA according to a standard curve prepared with commercial thromboplastin. Dots represent the results of individual experiments, medians are indicated by lines.

G. GRONIC ET AL.

298

TABLE 1 Comparison of PCAs in culture supernatants with the respective cell-associated PCAs of lung macrophages PCA of cuiture-supernatants ( U )

Cell associated PCA (U) 2

Controls

LPS-stimulatedl

Controls

LPS-stimulated

0.018 0.008

0.66 0.46

3.64 3.17

0.202 0.126

0.48 0.30

3.91 1.49

Stimulation time 4 hours Means 0.012 + SD 0.004 Stimulation time 24 hours Means 0.018 _rSD 0.008

Lung cells were incubated with 10/~g/ml LPS. -'PCA was normalized for 3 X 104 macrophages in the assay. TABLE 2 Increase in PCA by stimulation with LPS' of lung macrophages cultivated in vitro for 1-3 weeks Horse no.

1 (3) 2 1 (3) 2 ( 1)

Stimulation time ( h ) 1.5 5

Increase in PCA (%) 7743 443

4(2)

24 24 ,~A g.,"IP

2100 2841 ~-,~,,~ .,J | U

5 (2) 6 (2)

24 24

197 874

Means +_SD

24

1704 1125

3 (2)

'in indicated time intervals before the end of in vitro cultivation, LPS ( 10/Jg/ml) was added to the cell culture medit, m. -'The total time (weeks) of in vitro cultivation is given in brackets. 3pCA of controls was designated 100%.

Cell processing and gradient centrifugation was carried out as described (Griinig et al., 1990). Briefly, cells were washed three times with medium ((RPMI 1640 (Gibco), supplemented with 25 mM HEPES (Sigma), 100 IU/ml penicillin, (Gibco) and 100 ~g/ml streptomycin (Gibco)) to which hyaluronidase (Boehringer) 0.06 mg/ml and DNase (Boehringer) 0.03 mg/ ml were added. Then cells were overlayered onto 4 ml aliquots of a 22% metrizamide (Nycomed) solution. The gradients were centrifuged at 4 0 0 × g for 25 min at 4 °C. Cells from the interface were recovered, washed once with

PROCOAGULANTACTIVITYOF EQUINELUNGMACROPHAGES

299

TABLE 3 Induction of procoagulant activity in lung macrophages: comparison of adherent and non-adherent cultures Horse no. PCA (U) and (viability, % ) of controls

l 2 3 4 5 6 7

PCA and (viability 1%) of LPS' stimulated cells

Stimulation index of PCA (control PCA = l )

Adherent'

Non-adherent 3

Adherent

Non-adherent

Adherent

Non-adherent

0.058 0.146 0.085 0.112 0.119 0.014 1.008

0.017 (25) 0.210 (66) 0.026 (76) 0.354 (46) 0.202 (50) 0.033 (75) 0.4794 (40) 5

1.527 3.419 0.607 1.863 4.061 0.993 6.304

0.079 (25) 1.310 (49) 0.164 (76) 1.148 (42) 1.489 (50) 0.173 (78) 0.6975 (50) 5

26.33 23.42 7.14 16.63 34.10 70.93 6.255

4.65 6.24 6.31 3.25 7.37 5.24 1.465

(65) (85) (92) (85) (78) (89) (78)

(44) (66) (83) (80) (80) (80) (64)

'Stimulation with LPS was carried out with 10 gg/ml for 24 hours. 2.3 Adherent cells were cultivated on plastic tissue culture plates. Non-adherent cells were cultivated in teflon bags (horses nos. 1-3 ), in teflon tubes (horses nos. 4-6), or in polypropylene tubes (horse no. 7). 4paired data were not distinct (P> 0.05; Dixon and Mood test (Sachs, 1984) ). SPaired data were distinct (P< 0.01; Dixon and Mood test (Sachs, 1984) ).

medium, resuspended in two 2 ml aliquots of a 10% metrizamide solution and underlayered with 2 ml aliquots of 14, 16, 18, 20, 22% metrizamide solutions. The gradients were centrifuged (400 × g for 25 min at 4 ° C ). Five cell subpopulations were recovered from the interfaces. The first three interfaces, rich in macrophages, were pooled in 32 specimens. In five specimens, the two remaining interfaces were also pooled and processed identically.

Blood cells Either buffy coats or leukocyte-rich plasmas were prepared from heparinized ( 50 U / m l ) blood. The leukocytes were centrifuged ( l 0 rain at 400 X g) and washed once with phosphate-buffered saline. Mononuclear cells were recovered from the interface of FicoU-Hypaque (Pharmacia) gradients, and washed once with phosphate-buffered saline and medium, respectively. Cell suspensions were used for the characterization of PCA with clotting factordeficient human plasmas. Mononuclear cells employed in experiments to evaluate the responses to LPS were further purified. Cells were resuspended in 10% metrizamide solution and undedayered with 4 ml aliquots of 16 and 18% metrizamide solutions, respectively. The gradients were centrifuged for 25 min at 400Xg and 4°C. The cells of the first interface were recovered, washed once, and resuspended in medium. Neutrophils were purified from erythrocyte-rich sediments of Ficoll-Hypaque gradients. Erythrocytes were lysed by hypotonic lysis. Isotonicity was restored using 10 times concentrated Hank's balanced salt solution (HBSS) (Gibco). Neutrophils were washed

'~u!uaoD) saqs!p ouaa,~ls,qod u! paanllna aaa,~ SllaD "so:tmlna luazaqpv ( I ) •p a u ! m a a l a p

aaaA~ ,~l![!qtn^ p u e a a q t u n u l l a a

'VZ)d "ltU/Sllaa 901 × I £Iolt~tu!xoaddt~ le mn!patu u! papuadsnsoa pue tun!patu ql!~a aauo paqse,~ aaOA~siIaa 'aan~,ina aalJV "ZOD %~ jo oaaqdsotule pat.j!p! -tunq e u! DoLt~ ~,t~patuaojaad aaaA~saanllna IIV "( tunt.patu alaldtuoa) (oaq!o) [I u!a!aaloqdtue Itu/~lr/g'~; pue mnaas sno~ololne (u!tu g~; 'Dogg) paleaq -aad %0I ql!~ PalUatualddns sea mn!patu aql 'uo!leA!llna Oal!A U! aod a~nllna llaD •pa~,eo!pu! sv sotu!l snO.LII~A.loJ o.tmlna olu! lnd sv,~ aapu!emaa a q l "VDd puv le!luaaajj!p '£1!l!qv!^ 'aaqtunu llaa jo uo!leu -!tuaalap aql aoj op!se los sea suo!suadsns llaa aql jo qaea jo ant.l!ll.tm ouo

"Im/Sllaa LOI X I ,qolt~tu!xoaddv le mn.tpotu u! papuodsnsaa pue tun!pare ql!,~ aa!,~,l •su~!patu ol s.mjaJ aU!l ploq a q I "sau!l ,~q palaauuoa a.m asaoq q:majo sllns -a; o q l "%001 pol~ufl.[sop s~,~ anle^ V D d sl! pue luatuaanseatu j o lu!od tl:)~a aoj popnl3u ! s e a ll,",l,11111+t'~,"~

i,-,+,+ ,-.-,~, V

"S'.'OlI~'IUO3U03.

'~~ " ",'~" '.. ,,. i . ~ , • ',J'.,

SdqJ 0

.. . ... . u...+.,.o . . ."+

. ."4+ . . 1.1 . . : . "t. . v.c . . .'-'.+ . . . .+.= ., t .+ :. 1.- ". .-.:++,~,i . . . . "1 . . .v o. +. ' -. - + . • '.: +. i . ' '.- " "+ .'-'-"

pu~ sosJoq x!s moJj pa;~los+ aao~ sa~+qdoaa+m Oun'l (8) "ScI'I Ira/+ Ol q]!~ auJ!l jo sql+uol (V) "aauapuadap asop pu+ am.tl+ "sa~+qclo,a+m +unI +to VDcl ut. as+a~aut,paanput-sd'l "+ "8!.-I oouopuodop osocI (8

oouopuodop otU!.L (~ (eJnoq) uo!iulnw!lt t o a w ! l

BOI '(IW/O) eplJUtlOOtllAIOdOdl'l

0 OOL

i

i

9-

g-

Li

8-

6°

i

OI. °

1Pi~

t

g'L

i

oog

,f

/

~"

/

2/i jjj ....

O00L

'

oogt

,/

!

/

m

~-oom

I i//

OOg i

/I

000~ //

k 000~,

//

I oog~

oog8

t

/ g

000¢

L O00g

(OOL tlOJ;Uoo |0

(OOL •

lllOJlUOO I0 ~llalloa) VOd

•w l a oINO~O o

00£

PROCOAGULANTACTIVITYOF EQUINE LUNG MACROPHAGES

30 l

TABLE 4 Reduction of LPS-mediated stimulation of PCA from lung macrophages by cycloheximide (Cyh) and actinomycin D (ActD) Horse no. and treatment of ceils

PCA ( U / 3 × 104 macrophages )

Increase of PCA (%)

Reduction of the PCA increase ( % )

0.249 2.490 0.741 0.379

1000 298 152

70 85

0.025 0.334 0.091 0.067

1336 364 268

73 80

0.034 0.937 0.381

2756 1121

59

0.087 2.046 0.560

2352 644

73

1 (4h)' control LPS a LPS+Cyh LP~ + ActD

2(4tl) control LPS LPS + Cyh LPS+ActD

3(24h) control LPS LVS+Cyh

4(24h) Control LPS LPS +ActD

~Times of stimulation are given in brackets; cells were preincubated with actinomycin D or cycloheximide for 30 minutes (experiment no. 1 ), 60 minutes (experiments nos. 2, 4), or for 180 minutes (experiment no. 3). 2The following concentratinns were used: LPS ( 10 ug/ml)" actinomycin D (5-6 .ag/ml)- cycioheximide (2 gg/ml, experiments nos. 1, 3, or 10#g/ml, experiment no. 2).

diameter 35 mm). After adherence for 2-4 h, plates were rinsed and complete medium was added. After further incubation for various times, cell culture supernatants were recovered by centrifugation for 10 rain at 400Xg. Nonadherent cells were discarded, except in experiments using neutrophils, where non-adherent cells were added to adherent cells. Prior to recovery, adherent cells were incubated with 1-2 ml of HBSS (Gibco, without calcium and magnesium, supplemented with 25 mM EDTA) on ice for approximately 15 rain. Cells were dislodged using a pasteur pipette. (2) Non-adherent cultures. Cells were cultured in teflon bags (FEP 100A, Dupont de Nemours, Geneva) as described (Jungi and Peterhans, 1988 ), in teflon flasks (Nalgene) or in polypropylene tubes (Nunc). After incubation, cells and culture supernatants were recovered by centrifugation for 6 min at 400Xg. (3) Bronchoalveolar macrophages. For adherent cultures, l ml aliquots were incubated up to 24 h without (controls) or with the addition of LPS. I

302

O. GRONIG ET AL.

PCA 2.6-

!.0 0.9-

0"8t 0.7

(u) 06 0.50.40.30.2!

01. 0.0

freshly

:~h

isolated monocytes

~,h

6h -time of

8h

II

24h

incubation

Fig. 3. Expression of PCA by control monocytes. Blood monocytes were isolated from nine horses and cultivated in teflon bags for up to 24 h. Clotting times were converted to U of PCA according to a standard curve prepared with commercial thrombop!astino The results of each horse are connected by lines. The bold line refers to medians.

Cultures which were kept for 1-3 weeks were supplemented with 1.5 ml complete medium and LPS solutions were added at indicated time points before the end of the cultivation period. For non-adherent cultures, 2 ml aliquots were incubated for 24 h without (controls) or with the addition of LPS. (4) Mononuclear cells. Aliquots of 2 ml were incubated in teflon bags without (controls) or with the addition of LPS for up to 24 h. Mononuclear cells used for the characterization of PCA with clotting factor-deficient human plasmas were incubated in 60 mm plastic dishes for 24 h with 4 ml of complete medium and l 0 gg/ml LPS. (5) Neutrophils. One millilitre aliquots were incubated in plastic dishes without (controls) or with serially diluted LPS for 4 h.

Cytologic examinations Cytocentrifuge slides were prepared as described (Griinig et al., 1990).

303

PROCOAGULANT ACTIVITY OF EQUINE LUNG MACROPHAGES

(activity

PCA (activity of controls

PCA of controls

• 100) 5so

- 100)

400

- 4oo

300

- 300

200

1200

lOO

[ lOO

o

i

;

;

.

I

;4

.

.

.

.

.

I

time of stimulation(hours)

Lipopolyuccharide(g/m!), log

A) Time dependence

B) Dose dependence

,

:o

Fig. 4. LPS-induced increase in PCA of blood monocytes. Time and dose dependence. (A) Blood monocytes were isolated from nine horses and cultivated in teflon bags for various lengths of time in the presence of 10 g/ml LPS. (B) Blood monocytes were isolated from five horses and cultivated in teflon bags for 4 h in the presence of LPS in various concentrations. A control was included for each point of measurement and its PCA value was designated 100%. The results of each horse are connected by lines. The boid iine refers to medians.

PCA determination PCA was determined as described in detail previously (Griinig et al., 1990). A standard curve was prepared with commercial thromboplastin (Baxter, Merz and Dade) and clotting times induced by 10/zl thromboplastin were designated 100 U. PCA of cells was normalized for 3 × 104 monocytes, macrophages, or neutrophils in the final assay. PCA was characterized using normal, clotting factor VII-, and clotting factor VIII-deficient human plasmas (Baxter, Merz and Dade ).

Statistical analysis Data were analysed using the Dixon and Mood test. Means and standard deviations as well as medians were determined (Sachs, 1984).

304

G. GRONIG ET AL.

TABLE 5

Characterisation of PCA's of LPS-stimulated t lung cells and of LPS-stimulated blood mononuclear cells using human plasma Normal plasma

Factor VIIIdeficient plasma

Factor VIIdeficient plasma

Cells

Thromboplastin 2

Cells

Thromboplastin

Cells

Thromboplastin

1 2 3 4 5 6 7

132.93 109.1 74.7 123.4 95.4 75.1 86.8

95.2 101.7 99.5 92.3 92.0 107.8 92.4

172.9 175.5 106.7 177.3 148.8 125.5 127.3

131.8 155.4 159.2 133.9 141.5 130.8 124.5

550.0 357.9 302.4 463.7 471.2 460.0 453.0

495.6 547.3 529.9 490.2 507.8 530.5 456.7

Means ±SD

99.6 ±23.0

97.3 +6.0

147.7 +28.5

139.6 -± 13.1

463.9 +81.5

508.3 ± 30.6

Blood mononuclear cells i 94.83 2 89.6 3 92.4 4 85.3

105.2 111.6 96.2 89.2

161.0 149.4 139.1 11[;.1

143.1 176.6 144.0 125.2

466.0 551.5 402.9 644.5

508.5 581.3 5 ! 4.6 643.3

Means ±SD

100.6 +9.9

141.9 +_!8.2

147.2 +2!.4

516.2 + ins o

561.9 ±63°5

No. of experiment

Lungcells

90.5 ±4.!

~Cells were stimulated with 10/~g/ml LPS for 24 hours. -'The amount of commercial thromboplastin was adjusted so that clotting times corresponded to clotting times induced by cell preparations. aclotting times (seconds). RESULTS

Expression of PCA by unstimulated control lung macrophages and by LPSstimulated cells Fresh lung cell suspensions recovered from density gradients comprised 46.5 _+13.9% macrophages, 29.2 _+ 10.5% lymphocytes and 18.3 +_ 17.7% neutrophils; the viability was 86.7 +_6.5%. In vitro cultivation of lung cells adhering to tissue culture dishes resulted in a further enrichment of macrophages to 65.1 +_20.2% after 4 h, 69.9+_ 15.5% after 24 h, and 93.2+_2.5% after l week of cultivation. During this time, giant cells were formed; the percentage of number of nuclei within giant cells per number of nuclei within macrophages and giant cells increased from less than 1% (freshly isolated cell suspensions) to approximately 5-30% ( 1-3 week cultures). The relative amount

PROCOAGULANT ACTIVITY OF EQUINE LUNG MACROPHAGES

305

PCA (U)

0.07! 0.060.050.040.030.020.01 0.00

controls i

freshly isolated

I

!

-~

-;

-6 LPS

( g / ml),

-~ Icg

blood neutrophils Fig. 5. Expression of PCA by freshly isolated, control and LPS-stimulated blood neutrophils. Blood neutrophils were isolated from four horses and cultivated on plastic dishes for 4 h in the presence of LPS in various concentrations or without LPS (controls). Clotting times were converted to U of PCA according to a standard curve prepared with commercial thromboplastin. The results of each horse are connected by lines. The bold line refers to medians.

nf ,e,trnphil~ declined from 8. ! _+ ! 6:4% after 4 h0 to 1.2 + 2.0% after 24 h of cultivation, and there were no neutrophils after I week of cultivation. PCA of freshly isolated macrophages varied between 0.02 and 1.1 U. PCA values of unstimulated cells fluctuated during in vitro cultivation (Fig. 1 ). However, 18 of 23 specimens lost 10-95% of their PCA within the first day of cultivation. Two of five and five of seven specimens which were cultivated for 1 or 2 weeks, respectively, had two to seven times increased PCA values compared with the PCA of freshly isolated cells (Fig. 1 ). Stimulation of freshly isolated lung cells with LPS caused an increase in PCA which was noted at the first point of measurement ( 1.5 h of incubation) and which continued until the last point of measurement (24 h of incubation, Fig. 2A). PCA production was dose dependent; the smallest dose of LPS which induced an increase in PCA was 10 -9 g/ml (Fig. 2B). Only a small portion of the cell-associated PCA was secreted into the culture supernatant as shown in Table 1. Stimulation of lung cell populations which were cultivated in vitro for 1-3 weeks with LPS also resulted in an increased expression of PCA as shown in Table 2. As outlined in Table 3, the PCA of controls was unaffected by adherent or non-adherent culture, however, LPS stimulation induced significantly higher PCA values in adherent lung macrophages. Consequently, the LPS-mediated

306

G. GRUNIG ET AL.

increase in PCA was greater in adherent lung macrophages than in non-adherent cultures. Strikingly, all the non-adherent cultures had reduced viabilities when compared with the respective adherent cells. Both cycloheximide (inhibitor of the protein synthesis) as well as actinomycin D (inhibitor of the RNA synthesis) reduced the LPS-mediated increase in PCA production by lung macrophages (Table 4). However, the reduction was not complete. Incubation with inhibitors for 24 h resulted in a marked decrease of viability from 86 to 61%.

Expression of PCA by unstimulated control blood mononuclear cells and by LPS-stimulated cells Cell suspensions of peripheral mononuclear cells were enriched for monocytes by centrifugation on two consecutive density gradients; centrifugation over Ficoll-Hypaque gradients was followed by metrizamide gradient centrifugation. The monocyte-enriched peripheral mononuclear cells comprised 60+_ 5% monocytes, 35 + 7% lymphocytes and few basophils; the viability was 90-95%. Immediately after isolation, blood monocytes expressed very low amounts of PCA (equal to or less than 0.1 U) (Fig. 3). When cultivated in teflon bags, PCA expression of control cells rose to approximately 0.4 U after 24 h of culture. As shown in Fig. 4A, stimulation of monocytes with LPS resulted in increased PCA expression as soon as 2 h after incubation began. Maximal PCA increases were observed at 4 h of incubation. The stimulation of PCA expression of blood monocytes was dose dependent (Fig. 4B). The lowest dose of LPS which induced an increase in PCA was 10-~! g/ml. As observed with macrophage cultures, only a minor portion of PCA was secreted into the culture supernatant (data not shown). PCA was detected only in culture supernatants of LPS-stimulated cells and values were 5-10 times lower than PCA measured with intact cells. Table 5 depicts experiments to characterize PCA further using human plasma. They were carded out using LPS-stimulated mononuclear cell suspensions derived from peripheral blood and LPS-stimulated lung cell preparations. Clotting times induced by these cell preparations in clotting ta~tor VIII- and clotting factor VII-deficient plasmas were comparable to times induced by commercial thromboplastin (Table 5).

Examination of blood neutrophils with regard to PCA expression Cell suspensions prepared from Ficoll-Hypaque sediments were composed of 94 + 2% neutrophils, 3-5% eosinophils and less than 1% monocytes; the viability was 95-100%. Blood neutrophils were cultivated in vitro only up to 4 h because preliminary experiments had shown that the viability of neutrophil suspensions decreased to 20% after 8 h of cultivation. Freshly isolated blood neutrophils expressed minute quantities of PCA (less than 0.01 U), as shown in Fig. 5. Incubation of neutrophils in the presence of

PROCOAGULANT ACTIVITY OF EQUINE LUNG MACROPHAGES

307

LPS resulted in a slight increase in PCA (Fig. 5 ). In other words, LPS-stimulated neutrophils expressed approximately twice as much PCA as control neutrophils and approximately four times as much PCA as freshly isolated neutrophils. Addition of 20-30% blood lymphocytes to neutrophil suspensions did not produce a better PCA response to stimulation with LPS (results not shown). DISCUSSION

For the first time, noninduced and LPS-induced PCA has been followed in freshly isolated and cultured equine alveolar lavage cells and compared with PCA expressed by fresh and cultured cells from peripheral blood. Our data suggest that lung macrophages are already partly activated when harvested. Several observations support this notion, firstly, PCA was observed spontaneously, immediately after cell isolation in lung macrophages whereas blood mononuclear cells failed to express PCA both in horses (this study) and in other species (Edwards and RicHes, 1984; Rothberger et al., 1984; Chapman et al., 1985 ). Secondly, within the first 24 h of in vitro culture, 18 of 23 unstimulated, control lung macrophage specimens displayed reduced PCA, whereas the PCA of control monocytes increased approximately six times. Thirdly, the different sensitivity to LPS of lung macrophages and blood monocytes (the minimal concentration of LPS which caused an increase in PCA was 10 - 9 g/ml and 10-iI g/ml, respectively) might reflect a preexisting state of activation of macrophages, but not monocytes. The increase in PCA of control ~onoe~es might have been mediated by LPS contamination of culture media. However, an influ~ace of serum factors on PCA production of monocytes has been reported by Edwards and Perla (1984). Moreover, human mononuclear cells isolated under strictly pyrogenfree conditions also acquire increased PCA on 24 h culture in suspension or adherence (Miserez and T.W. Jungi, unpublished data). Increases in PCA in five of 7 control lung macrophage specimens after 2 weeks of cultivation might have been the result of the release of lymphokines. The formation of giant cells during long-term culture of lung cells might be further evidence for lymphokine release; since fusion of giant cells is induced by interferon (Nagasawa et al., 1987) or by interleukin-4 (McInnes and Rennick, 1988). Alternatively, PCA increase in unstimulated monocytes or macrophages might accompany cell adaptation and maturation during in vitro culture. Lung macrophages as well as blood monocytes shed only one fifth to one tenth of the respective cell-associated PCA into culture media, similar to LPSstimulated human monocytes (Tsao et al., 1984) and LPS-stimulated rabbit alveolar macrophages (Sitrin et al., 1983 ). However, bronchoalveolar lavage fluids of horses (Griinig et al., 1990), rabbits (McGee and Rothberger, 1985 ) and man (Chapman et al., 1988 ), contain PCA in readily recognizable quan-

308

G. GRUNIG ET AL.

tities. The macrophage is regarded as the main source of PCA on the alveolar surface (McGee and Rothberger, 1985; Chapman et al., 1988 ) and may also be the source of PCA in bronchoalveolar lavage fluids. Our results confirm that equine blood neutrophils produce only marginal PCA even after stimulation with LPS. This is in accordance with previously published results regarding blood cells (Edward and RicHes, 1984; Henry and Moore, 1988). However, lung neutrophils could not be separated from lung macrophages (Griinig et al., 1990) and therefore LPS-stimulation experiments could only be carded out t, sing blood neutrophils. The slight LPSinduced increase in the PCA ofblood neutrophils might have been caused by either the contamination of cell suspensions with monocytes or the expression of a receptor for clotting factor X, which binds and activates factor X upon stimulation with ADP thereby initiating clotting (Altieri et al., 1988 ). LPS induced a dose-dependent increase in PCA expression in fresh lung macrophages, in lung macrophages after 1-3 weeks of in vitro cultivation, and in blood monocytes. An increase in PCA was observed as soon as 2 h after stimulation; the increase continued up to 24 h in lung macrophages and peaked at 4 h in blood monocytes. A further increase in the PCA of LPSstimulated monocytes was most probably masked by the increase in PCA of control monocytes which began after approximately 4 h of cultivation. Similar kinetic and dose-response curves have been obtained with human blood monocytes (RicHes et al., 1977 ), and rabbit alveolar macrophages (Sitrin et al., 1983; Maier and Hahnel, 1984). Lung macrophages of man (Chapman et al., 1985; Nakstad et al., 1987) and mice (Lyberg et al., 1983) and blood monocytes of rats (Lando and Edgington, 1985) express a relatively small in~r~a~c in n,-,. , ,___ .L ,.,, r~.P, 1,two to three fold) and arc ~css r a s~nsitive. "" ......... rlUW~V~I-~ comparison of LPS sensitivity between different publications is complicated by possible differences in the level of LPS contaminatica of media and labware. This might explain why, in contrast to our results, Iq.nry and Moore ( 1988 ) required a relatively high LPS concentration ( 10 -s g/ml) in order to observe a PCA increase in equine blood monocytes. The difference in LPS sensitivity of blood monocytes and lung macrophages could be caused by a different LPS burden in vitro. Although the same media and labware were utilized in all experiments, exposure of lung macrophages to LPS could have occurred during endoscopy: firstly, the endoscope material did not tolerate a stringent decontamination procedure, and secondly, secretions and saliva from the nasopharyngeal cavity might have been dragged along with the insertion of the endoscope. Lung macrophages might also have developed tachyphylaxis because of stimulation with LPS in vivo. Tachyphylaxis has been observed in the PCA response of mouse peritoneal macrophages which were pulsed with small doses ofLPS (Shands et al., 1988 ). Hay and straw dust as well as scraps of pelleted diets even from unspoiled specimens contain up to 200 gg LPS/g of specimen (Kamphues, 1986). While

PROCOAGULANT

ACTIVITY

OF EQUINE

LUNG

MACROPHAGES

309

handling mouldy hay, up to 6.7/zg airborn LPS are released per m 3 of air (Rylander, 1987). Therefore, the decrease in PCA of control macrophages within the first 24 h of cultivation might reflect a reduced level of LPS stimulation. Besides LPS, inhaled dust particles and spores (Griinig et al., 1986; Clarke et al., 1987) might have induced PCA expression through stimulation oflymphocytes. Adherence itself might be a factor which modulates PCA as has been demonstrated by the adherence-induced superoxide production of rat alveolar macrophages (Ryer-Powder and Forman, 1989). However, our results regarding adherence, are not fully conclusive; concurrent with significantly higher PCA after LPS stimulation, lung macrophages cultivated "'in adherence" exerted significantly greater viabilities than lung cells which were cultivated under non-adherent conditions. Therefore, better induction of PCA might be the result of greater vitality of adherent macrophages, although adherence-induced modulation of cell function cannot be ruled out. Factors which might quantitatively influence the PCA of :mstimulated controls and of LPS-stimulated cells are platelet contamination of blood monocytes (Henry and Moore, 1988 ) and severity of chronic pulmonary disease. Our task was to study qualitative and not quantitative LPS-mediated PCA responses. Because equine platelets do not increase PCA on stimulation with LPS (Henry and Moore, 1988 ), mononuclear cell preparations had not been specifically depleted of platelets (platelet to monocyte cell ratio was approximately 20:1 ). Lung cell preparations from horses affected with chronic pulmonary disease were included in all experiments and there was no difference in the qualitative type of response compared with lung cells which were oba._'~ __! ~ _ ~ LillllCll 11 U U l

L__1~11,... ll~illltlly

11. . . . . . llOl D~D

I ..1.-. 4. ,,~ ~,,,~..~e .-,l~..~tt,,,~ ~ Udtd llOi, i~tlOWll l.

The PCA of lung macrophages and blood monocytes was most probably caused by tissue factor because clotting was not induced in plasma deficient of factor VII but was induced in clotting factor VIII deficient plasma which is in accordance with previous reports (Sitrin et al., 1983; Wasi et al., 1983; Rothberger et al., 1984; Lando and Edgington, 1985; Henry and Moore, 1988 ). In contrast to equine macrophages, PCA of human alveolar macrophages in part comprises clotting factor VII (Chapman et al., 1985; Nakstad et al., 1987). Equine lung macrophages did not produce appreciable amounts of clotting factor VII either because of species differences or technical problems in detecting factor VII. The PCA response of equine lung macrophages is to a great extent dependent on de novo synthesis of protein as has been described for blood monocytes (Prydz and Allison, 1978; Levy et al., 1981; Lando and Edgington, 1985 ), and bovine alveolar macrophages (Car et al., 1988 ). However, PCA induction could not be fully prevented by cycloheximide and actinomycin D. This could be caused by a technical problem, e.g. suboptimal doses used; but because of the drug effect on viability of lung cells, doses of inhibitors could

3 !0

G. GRONIGETAL.

not be increased. Yet, a cycloheximide superinduction of procoagulant responses, probably owing to the reduced production of repressor proteins has been observed in routine peritoneal macrophages (Moon and Geczy, 1988 ). ACKNOWLEDGEMENTS

The authors wish to thank Mrs. Pletscher for graphics. This work was supported by grant number 3.872-0.86 from the Swiss National Science Foundation.

REFERENCES Aitied, D.C., Morissey, J.H. and Edington, T.S., 1988. Adhesive receptor Mac-I coordinates the activation of factor X on stimulated cells of monocytic and myeloid differentiation: An alternative initiation of the coagulation protease cascade. Proc. Natl. Acad. Sci. U.S.A., 85: 7462-7466. Car, B.D., Slauson, D.O., Suvemoto, M.M. and Neilsen, N.R., 1988. Kinetics and regulation of bovine alveolar macrophage procoa~ulant activity. J. Leuk. Biol., 44:299 (Abstract 162). Chapman, H.A., Allen, C.L., Stone, O.L. and Fair, D.S., 1985. Human alveolar macrophages synthesize factor VII in vitro. J. Clin. Invest., 75: 2030-2037. Chapman, H.A., Stahl, M., Allen, C.L., Yee, R. and Fair, D.S., 1988. Abnormalities in pathways of alveolar fibrin turnover among patients with interstitial lung disease. Am. Rev. Resp. Dis., 137: 1417-1425. Clarke, A.F., Madelin, T.M. and Allpress, R.G., 1987. The relationship of air hygiene in stables to lower airway disease and pharyngeal lymphoid hyperplasia in two groups of thoroughbred horses. Equine Vet. J., 19: 524-530. Edwards, R.L. and Perla, D., 1984. Effect of serum on monocyte tissue factor generation. Blood, 64:707-7 i 4. Edwards, R.L. and Rickles, F.R., 1984. Macrophage procoagulants. Prog. Hemostasis Thromb., 7: 183-209. Geczy, C.L., 1983. The role of clotting process in the activation of lymphokines on macrophages. Lymphokines, 8: 201-247. Griinig, G., von Fellenberg, R., Maier, R. and Corboz, L., 1986. Elastase-producing microorganisms in horse lungs: Their possible role in the pathogenesis of chronic pulmonary disease in the horse. Equine Vet. J., 18: 396-400. Griinig, G., Hermann, M., Winder, C. and von Fellenberg, R., 1988. Procoagulant activity in respiratory tract secretions from horses with chronic pulmonary disease. Am. J. Vet. Res., 49: 705-709. Griinig, G., Hulliger, C., Hermann, M., Winder, C. and von Fellenberg, R., 1990. Separation of equine bronchoalveolar lavage cells by density gradient centrifugation and expression of procoagulant activity in unpurified cell suspensions and in cell subpopulations. Res. Vet. Sci., in press. Henry, M.M. and Moore, J.N., 1988. Endotoxin-induced procoagulant activity in equine peripheral blood monocytes. Circ. Shock, 26: 297-309. Jungi, T.W. and Peterhans, E., 1988. Change in the chemiluminescence reactivity pattern during in vitro differentiation of human monocytes to macrophages. Blut, 56:213-220. Kamphues, J., 1986. Lipopolysaccharide in Futtermitteln-m~gliche Bedeutung, Bestimmung und Gehalte. Uebers. Tierernaehr., 14:13 l-156.

PROCOAGULANT ACTIVITY OF EQUINE LUNG MACROPHAGES

31 1

Lando, P.A. and Edgington, T.S., 1985. Lymphoid procoagulant response to bacterial endotoxin in the rat. Infect. Immun., 50: 660-666. Levy, G.A., Schwartz, B.S. and Edgington, T.S., 198 I. The kinetics and metabolic requirements for direct lymphocyte induction of human procoagulant monokines by bacterial lipopolysaccharide. J. Immunoi., 127: 357-363. Lyberg, T., Amlie, E., Kaplun, A. and Prydz, H., 1983. Macrophage heterogeneity in thromboplastin response. Scand. J. Immunol., 18: 235-240. Maier, R.V. and Hahnel, G.B., 1984. Microthrombosis during endotoxemia: Potential role of hepatic versus alveolar macrophages. J. Surg. Res., 36: 362-370. McGee, M.P. and Rothberger, H., 1985. Tissue factor in bronchoalveolar lavage fluids: evidence for an alveolar macrophage source. Am. Rev. Resp. Dis., 131: 331-336. Mclnnes, A. and Rennick, D.M., 1988. Interleukin 4 induces cultured monocytes/macrophages to form giant multinucleated cells. J. Exp. Med., 167:598-61 I. Moon, D.K. and Geczy, C.L., 1988. Recombinant IFN-a synergizes with lipopolysaccharide to induce macrophage membrane procoagulants. J. Immunol., 141:1536-1542. Nagasawa, H., Miyaura, C., Abe, E., Suda, T., Horiguchi, M. and Suda, T., 1987. Fusion and activation of human alveolar macrophages induced by recombinant interferon and their suppression by dexamethasone. Am. Rev. Resp. Dis., 136:916-921. Nakstad, B., Boyle,N.P. and Lyberg, T., 1987. Procoagulant activities in human alveolar macrophages. Eur. J. Resp. Dis., 71:459-47 I. Prydz, H. and Allison, A.C., 1978. Tissue thromboplastin m molecular and cellular biology. Thromb. Haemost., 39:582-591. Rickles, F.R., Levin, J., Hardin, J.A., Barr, C.F. and Conrad, M.E., 1977. Structural features of Salmonella typhimurium lipopolysaccharide required for activation of tissue factor in human mononuclear cells. J. Lab. Clin. Med., 89: 792-803. Rothberger, H., McGee, M.P. and Lee, T.K., 1984. Tissue factor activity. A marker of alveolar macrophage maturation in rabbits. Effects of granulomatous pneumonitis. J. Clin. Invest., 73: 1524-1531. Ryer-Powder, J.E. and Forman, H.J., 1989. Adhering lung macrophages produce superoxide demonstrated with desferal-Mn (VI). Free Radical Biol. Med., 6:513-518. Rylander, R., 1987. Role of endotoxins in the pathogenesis of respiratory disorders. Eur. J. Resp. Dis., 71: Supplement 154: 136-143. Sachs, L., 199.4. Angewandte Statistik: Anwendung statistischer Methoden. 6. Springer-Verlag, Berlin. Shands, J.W., 1987. Lymphocyte collaboration is not required for the induction ofmurine macrophage procoagulant by endotoxin. Thromb. Res., 46:271-279. Shands, J.W., Sunnenberg, T.D. and Lottenberg, R., 1988. Procoagulant syn:nesis by exsudate and bone marrow-derived murine macrophages. J. Leuk. Biol., 44:172-179. Sitrin, R.G., Kaltreider, H.B., Ansfield, M.J. and Webster, R.O., 1983. Procoagulant activity of rabbit alveolar macrophages. Am. Rev. Resp. Dis., 128: 282-287. Tsao, B.P., Fair, D.S., Curtiss, L.K. and Edington, T.S., 1984. Monocytes can be induced by lipopolysaccharide-triggered T-lymphocytes to express functional factor VII/VIIIa protease activity. J. Exp. Med., 159: 1042-1057. Viel, L., 1986. Structural-functional correlations of the lung in horses with small airway disease. In: E. Deegen and R.E. Beadle (Editors). Lung Function and Respiratory Diseases in the Horse. Intr.,-national Symposium, 27-29 June 1985 at Hannover, Germany. Hippatrica Verlagsgeselischaft Calw, Germany, pp. 41-45. Wasi, S., Burrowes, C.E., Hay, J.B. and Movat, H.Z., 1983. Plasminogen activator and thromboplastin activity from sheep alveolar macrophages. Thromb. Res., 30: 27-45.