Secretory products from human monocyte-derived macrophages enhance platelet aggregation

Secretory Products

From Human Monocyte-Derived Enhance Platelet Aggregation

Macrophages

Michael Aviram, Mira Rosenblat, and J. Gerald Brook Both macrophages and platelets play an important role in atherogenesis. We studied the effect of conditioned medium obtained from human monocyte-derived macrophages on in vitro platelet aggregation. Incubation of macrophage-conditioned medium (MCM) with platelets resulted in enhanced platelet aggregation (up to 35% difference between basal and MCM-stimulated activity), which was time dependent. This MCM effect on platelet function was increased both with time of mononuclear cell culturing (up to 10 days) and with the time of macrophage incubation in serum-free medium (up to 24 hours) prior to MCM collection. MCM from either cholesterol-loaded macrophages or from macrophages obtained from patients with familial hypercholesterolemia demonstrated a 37% and 20% increased effect, respectively, in comparison to MCM derived from normal subjects. Macrophage activation with lipopolysaccharide resulted in the harvesting of a MCM that enhanced platelet activity 60% more than MCM obtained from nonactivated cells. The active component of MCM was inhibited fivefold following heating at 100°C for 10 minutes or after treatment with trypsin or protease, but was not affected by antioxidants. MCM activation of blood platelets may be of importance in atherogenesis. Understanding the mechanisms involved may contribute to an improved appreciation of the role of both platelets and macrophages in atherosclerosis. Copyright 0 199; by W.B. Saunders Company

T

HE EARLY atherosclerotic plaque is composed of monocyte-derived macrophages (MCM), which are loaded with cholesterol and form foam cells.’ Macrophages, in addition to their phagocytic properties, are also important secretory cells; they secrete polypeptides, hormones, coagulation factors, enzymes, and various other proteins.’ Platelet activation is also associated with accelerated atherosclerosis and various blood components, including plasma lipoproteins, have been shown to modulate platelet function.3s4 It has been shown previously that platelet-conditioned medium could promote macrophage cholesterol accumulation, as well as modulate macrophage interaction with plasma lipoproteins.*~* In the present study, we sought to analyze whether conditioned medium from human monocyte derivedmacrophages can affect platelet activation, and hence address an important question regarding the pathogenesis of atherosclerosis. METHODS

Macrophages Human mononuclear cells were isolated by density gradient centrifugation’ from blood derived from fasting normolipidemic subjects (plasma cholesterol < 200 mg/dL). Twenty milliliters of blood (anticoagulated with 10 U/mL heparin) was layered over 1.5 mL of Ficoll-paque and centrifuged at 500 x g for 30 minutes at 23°C. The mixed mononuclear cell band was removed by aspiration and cells were washed twice in RPMI-1640 culture medium containing 100 U/mL penicillin, 100 ug/mL streptomycin, and 2 mmol/L glutamine. The cells were plated at 2 x lo6 mononuclear cells per 16-mm dish (Primaria, Falcon Labware, Becton Dickinson, Oxnard, CA) in the same medium in the presence of 20%

From the Lipid Research Unit, Rambam Medical Center and Rappaport Family Institute for Research in Medical Sciences,

Metabolism,

Vol 40, No 3 (March), 1991: pp 270-274

SECRETORY PRODUCTS

271

Platelets

l

l

Venous blood was collected (after a 14-hour fast) from healthy subjects who had taken no medication for at least 3 weeks preceding the study. Blood was taken into acid citric dextrose (2.5 p/L citric acid, 25 g/L sodium citrate, and 20 giL dextrose) in a 6:l volume ratio. Platelet-rich plasma (PRP) was prepared by lowspeLd centrifugation at 200 x g for 10 minutes at 23°C. Washed platelets (WP) were prepared from the PRP. Acetic acid (10 mmol/L) was added to PRP, which was then centrifuged at 1,500 x g fcr 10 minutes. The supernatant was removed and the platelet pellet resuspended in Hepes buffer (5 mmol/L Hepes, 137 mmol/L s&urn chloride, 2.7 mmol/L potassium chloride, 1.2 mmol/L magnesium chloride, 12 mmol/L sodium bicarbonate, 0.4 mmol/L sodsum biphosphate. and 0.6 mmol/L glucose, pH 7.4) at a platelet clm:entration of lO”/L. Acetic acid was added and the platelets aer: centrifuged again to obtain the washed platelet pellet.

I-

I-

l-

Ekperknental Procedure MCM was incubated with WP (3 x lO”/L) at various volume concentrations. The platelet pellet was resuspended in 100% MC’vI, in MCM diluted with RPMI-1640 medium, or in control meoium as indicated in the Results section. Platelet incubation with MCM was carried out for 15 minutes at 37°C before the addrtion of collagen(3 bg/mL, Hormone Chemie, Munchen, GMBH) induced platelet aggregation. This collagen concentration nas used, since in preliminary experiments it was found to induce

submaximal platelet aggregation (60%). Adenosine diphosphate (2 pmol/L; Sigma, St Louis, MO) and thrombin (0.1 UlmL; ParkeDavis, Morris Plains, NJ) were also used as aggregating agents. Platelet aggregation was measured as percentage of maximal amplitude using the appropriate control (without platelets) in a P.4F’ 4 computerized aggregometer (Bio-Data, Haterboru. PA). APlatelet aggregation was calculated by subtraction of the aggregation amplitude (taken its maximum) obtained after platelet incubation in control medium from the value obtained after incubation in the presence of MCM. In some experiments, concentration studies ot other platelet agonists were performed. Analysis of MCM activity as a function of macrophage maturation was performed using cells that had been cultured for different time periods (up to I( I d.ays). A comparison between the effect of MCM on platelet aggregation in WP and PRP was carried out by dilution of MCM to 50% concentration with either RPM1 medium or with PRP. respecti\ et+,.

4s

L

I

c

C

CollaQs”

Thrombin

MCM

ADP

Fig 1. The effect of MCM on platelet aggregation induced by various agonists. WP were resuspended in control medium (C) or in MCM for 15 minutes at 37°C before analysis of platelet aggregation induced by collagen (3 pg/mL), thrombin (0.1 U/mL), or ADP (2 kmol/L). Results represent mean 2 SEM of four different experiments. lP < .Ol (v C).

aggregation curve (data not shown). The effect of MCM on platelet aggregation was a threshold response phenomenon and was maximal (up to 35% enhancement) when the platelets were resuspended in 100% volume of MCM (Fig 2). Using lower concentration of collagen (0.5 p,g/mL), the dose-dependent effect was more pronounced. Platelet aggregation of 51%, 61%, 69%, and 76% was obtained with 0%, 25%, 50%, and 100% MCM volume concentration, respectively (mean of two studies). In approximately 30% of the experiments, spontaneous platelet aggregation occurred after the platelets came into contact with MCM and before

Othu Assays MCM lipid oxidation was analyzed by the thiobarbituric acid redcrive substance assay.” Protein concentration was determined by the Lowry procedure.” MCM concentration of cholesteroP4 and phospholipids” were analyzed by calorimetric assays.

Srattrtics Rtsults performed

are presented as mean 2 SEM. Statistical using nonpaired Student’s t test.

analysis

was

RESULTS

I

I

I

I

25

50

75

100

MCM

After 24 hours of incubation in a serum-free medium at 37°C) MCM obtained from human monocyte-derived macrophages significantly (P < .Ol) enhanced in vitro platelet aggregation. Platelet aggregation induced by collagen, thrombin, or ADP was increased by 12% to 21% (Fig 1). Similar patterns were demonstrated for the slope of the

Concentration

(volume

%I

Fig 2. The effect of MCM concentration on its activity on platelet aggregation. WP (3 x lO”/L) were incubated with increasing concentrations of MCM for 15 minutes at 37°C before determination of platelet aggregation induced by collagen (3 pg/mL. 0) or thrombin (0.1 U/mL, 0). Results are representative of two different experiments, which varied by less then 4%.

AVIRAM, ROSENSLAT, AND BROOK

Table 2. Effect of Time of MCM Collection After Cell Transfer to Serum-Free Medium on its Effect on Platelet Aggregation Incubation Time (h)

APlatelet Aggregation

1

2*1

2

5t2

3

24 f 4

24

27 + 4

1%)

NOTE. Platelet aggregation was calculated as described in the legend to Table 1. Results are mean 2 SEM of three different experiments.

I

I

I

1

5

10

Days

in Culture

Fig 3. The effect of time of monocyte culturing on MCM activity. MCM was obtained at the indicated time points after mononuclear ceils were plated in culture dishes. GPlatelet aggregation was calculated by subtraction of the aggregation amplitude obtained for platelets resuspended in control medium from the value obtained after platelet resuspension in MCM. Results represent mean * SEM of three different studies.

the addition of the aggregating agents. The shape of the curves with spontaneous aggregation was similar to that of collagen-induced platelet aggregation. There was no spontaneous aggregation in the control (platelets treated with medium instead of MCM). The enhancing effect of MCM on platelet aggregation was found when the MCM was prepared from freshly isolated mononuclear cells, but was increased substantially when the MCM was obtained from mature macrophage cells (up to fourfold 10 days after plating, Fig 3). The slope of the curve was shallow after day 2 in culture (Fig 3). The effect of MCM obtained before 5 days in culture (> 98% macrophages) on platelet aggregation may thus be due to lymphocyte contamination. MCM was stable for at least 7 days at both 4°C and -20°C (Table 1). MCM activity was minimal over the first 2 hours of macrophage incubation in serum-free medium and became maximal after 24 hours of incubation (Table 2). The effect of MCM on platelet activity was rapid; the maximal effect was obtained within 10 minutes of incubation of MCM with platelets, with no further increase after more extended periods of incubation (data not shown). Table 1. Effect of Storage Temperature

and Duration on Table 3. Effect of MCM From Different Macrophage Sources on

Platelet Aggregation APlatelet Aggregation

Platelet Aggregation (%)

Storage Duration (d)

storage 0

Temperature

The platelet activation effect was not present in MCM obtained from mouse peritoneal macrophages or J-774A.l murine macrophage-like cell line (Table 3). Thus, this effect was specific for macrophages derived from human monocytes. To ascertain whether MCM stimulation of platelet aggregation is affected by the presence of plasma, we compared WP with a PRP preparation. Identical volumes of MCM (50%) were used in each experiment. MCM enhanced platelet aggregation to a similar extent in both types of platelet preparations (Fig 4). The effect of MCM was greater when relatively low concentrations of collagen, as the aggregating agent, were used (Fig 4). Activation of macrophages with 100 ng/mL lipopolysaccharide following incubation for 24 hours at 37°C resulted in MCM that induced 89% ? 4% platelet aggregation in comparison to 79% * 4% and 63% ? 5% obtained for platelets treated with MCM from nonactivated macrophages and without MCM, respectively (three separate experiments). Secretory products from cholesterol-loaded macrophages and from cells derived from patients with familial hypercholesterolemia (FH) induced greater platelet activation (by 37% and 20%, respectively) in comparison to MCM derived from non-cholesterol-loaded cells obtained from normolipidemic subjects (Fig 5). MCM contained 59.5 4 199 &mL proteins, 37 r 16 ug/mL phospholipids, and 49 +- 19 pg/mL cholesterol (n = 3). No malondialdehyde could be detected in MCM by the thiobarbituric acid-reactive substances assay. Following heat treatment (100°C for 10 minutes), MCM activity was substantially inhibited by up to 3.5fold in comparison to untreated MCM (Fig 6). A similar effect was obtained when MCM was pretreated with trypsin or protease (Fig 6). However, antioxidant agents, such as butylated-hydroxy toluene (BHT) and the free radical scavenger superoxide dismutase (SOD) did not affect MCM activity (Fig 6). Both BHT and SOD did not affect MCM

1

3

4°C

23 + 3

21 L-2

22 + 3

21 +4

-20°C

23 -c 4

22 % 5

21 + 3

20 f 5

(WI %)

NOTE. Platelet aggregation induced by 3 kg/ml_ collagen was studied in WP following their resuspension

in MCM or in control medium.

APlatelet aggregation (%.) represents the difference between the effect of MCM and control different experiments.

medium.

Platelet Aggregation

MCM Concentration 7

Results are mean r SEM of three

(%)

HMDM

MPM

J-774

0

62

64

53

50

77

60

56

100

84

59

58

NOTE. Results are the mean of duplicate incubations. Abbreviations:

HMDM,

human monocyte

derived

macrophages;

MPM, mouse peritoneal macrophages; J-774, murine macrophage-like cell line.

SECRETORY PRODUCTS

273

2

1 zo-

s .G b m b z

lo-

4

‘m 5 m I I 2 4 Collagen Concentration

:

-I

I 6 @g/ml)

Fig 4. The effect of MCM on platelet aggregation in a WP or in a PRP suspension. WP (0) or PRP (0) were incubated at a concentration of 3 x lO”/L with MCM (50% volume concentration) or control medium, for 15 minutes at 37°C before analysis of collagen-induced platelet aggregation. Results represent the mean of two different experiments.

activity when added to the incubation medium before MC’M collection (data not shown). Since apolipoprotein (apo) E is secreted from macrophages,’ we have studied its effect on platelet aggregation. The addition of 0, 1, 5, and 10 kg/mL of apo E to WP resulted in collagen- (3 l&rnL) induced platelet aggregation of 60%, 58%, 62%, and 64%, respectively (mean of two experiments). DISCUSSION

The present study demonstrates for the first time that secretory products from human monocyte-derived macrophages are able to enhance in vitro platelet aggregation. The effect was probably not a result of interaction between macrophage secretory products and a specific platelet

i

Fig 5. The effect of MCM derived from cholesterol-loaded macrophages and from hypercholesterolemic patients on platelet aggregation. Macrophages (2 x lo6 per 16-mm plate) loaded with cholesterol (CHOL) by 24 hours of incubation at 37°C with 100 +g/mL of Acetyl-LDL were found to contain 61 -t 11 pg cholesterol/mg cell protein in comparison to a value of 27 2 5 kg/mg in the control cells (C) and 35 k 4 pg/mg in cells derived from patients with FH (plasma cholesterol, 381 to 569 mg/dL). Results are the mean +- SEM of three different experiments (three different subjects). l/J < .Ol (V C).

Heated

BHT

TfYP.

J

Fig 6. The effect of various MCM treatments on its activity. MCM was pretreated as follows: (1) Heat treatment (100°C for 15 minutes followed by centrifugation at 1.500 x g for 10 minutes to remove precipitated denatured protein). (2) Butylated hydroxy toluene (BHT, 25 pmol/L). (3) Superoxide dismutase (SOD, 20 rg/mL). (4) Trypsin (20 pg/mL). (5) Protease (10 pmol/L). Incubations 2 to 5 were carried out at 37°C for 15 minutes. Results represent mean + SEM of three different experiments. ‘P < .Ol (v MCM).

agonist, since three different agonists (collagen, thrombin, and ADP) induced similar effects. It thus represents a distinct platelet activation phenomenon. The effect of MCM on platelet aggregation suggests an all-or-nothing threshold response phenomenon, and also increased with time of macrophage incubation in a serum-free medium. Although MCM activity could be shown in fresh mononuclear cells, it substantially increased with time of cell culturing and macrophage formation, suggesting that this activity was associated with a factor synthesized and/or secreted mainly by mature macrophages and not by contamination of lymphocytes. It is interesting that human monocyte-derived macrophages, but not peritoneal macrophages or the cell line J-774.A1, was able to secrete an active MCM. Either the other macrophages do not secrete the MCM factor(s) involved in enhancing platelet aggregation or these cells secrete other substances that interfere with MCM activity. MCM activity was shown in both WP and PRP preparations, suggesting that plasma components do not interfere with the effect of MCM on platelet aggregation. The active component in MCM was found to be heat labile and sensitive to trypsin and protease, suggesting that it is probably a protein. Possible candidates would be prothrombin (plus prothrombinase). tissue factor, collagen, or perhaps some of the other coagulation proteins. The absence of any inhibitory effect of antioxidants on MCM activity, together with the absence of measurable malondialdehyde (MDA) in MCM, suggests that this activity is not related to oxidative substance(s) such as superoxides orhydrogen peroxides, which are released from macrophages.’ Human monocyte-derived macrophages are associated with atherosclerosis’,‘” and the observed MCM activity on platelet aggregation may be related to aiherogenesis. RcGently, MCM was shown to stimulate low-density lipoprotein and scavenger receptor activities.” Human monocytederived macrophages secrete triglyceride-rich lipoprotein that contain apo E.‘” Cholesterol loading of macrophages

AVIRAM,

274

has been shown to affect secretion of various substances such as apo E and cholestero1.‘9 However, in our hands, apo E did not affect platelet aggregation. Since the responsible substance(s) appeared to be a protein (not apo E), the possibility that the cholesterol secreted by the macrophages was responsible for the platelet-enhancing activity of MCM was not entertained. The observation that enhanced MCM activity could be found when macrophages are either preloaded with choles-

ROSENBLAT, AND BROOK

terol or activated with lipopolysaccharide, or derived from FH patients, may suggest a role for in vivo MCM action on platelet aggregation in the atherosclerotic process. Essentially all the above situations are associated with increased atherogenicity,*24 and thus increased platelet activation under these conditions could be related to atherogenesis. ACKNOWLEDGMENT

We wish to thank Miriam Haas for typing the manuscript.

REFERENCES

1. Gerrity RG: The role of the monocyte in atherogenesis I. Transition of blood-borne monocytes into foam cells in fatty lesions. Am J Path01 103:181-185,198l 2. Nathan CF: Secretory products of macrophages. J Clin Invest 79:319-3261987 3. Aviram M, Brook JG: Platelet activation by plasma lipoproteins. Prog Cardiovasc Dis 30:61-72,1987 4. Brook JG, Aviram M: Platelet-lipoprotein interaction. Semin Thromb Hemost 14:258-265,1988 5. Curtiss LK Black AS, Takagi Y, et al: New mechanisms for foam cell generation in atherosclerotic lesions. J Clin Invest 80:367-373,1987 6. Mendelsohn ME, Loscalzo J: Role of platelets in cholesteryl ester formation by U-937 cells. J Clin Invest 81:62-68,1988 7. Aviram M: Platelet modified LDL: Studies in normal subjects and in patients with homozygous familial hypercholesterolemia. Clin Biochem 20:91-96,1987 8. Aviram M, Fuhrman B, Keidar S, et al: Platelet-modified low density lipoproteins induces macrophage cholesterol accumulation and platelet activation. J Clin Chem Clin Biochem 27:3-12, 1988 9. Boyum A: Isolation of mononuclear cells and granulocytes from human blood. Stand J Clin Invest 21:77-99, 1968 (suppl 96-100) 10. Edelson PJ, Cohn JA: Purification and cultivation of monocytes and macrophages, in Bloom BR, David YR (eds): In Vitro Methods in Cell-Mediated and Tumor Immunity. San Diego, CA, Academic, 1976, pp 333-340 11. Aviram M, Williams KJ, McIntosh YA, et al: Intralipid infusion abolishes the ability of human serum to cholesterol-load cultured macrophages. Arteriosclerosis 9:67-75,1989 12. Buege JA, Aust SD: Microsomal lipid peroxidation. Methods Enzymol52:302-310,1987

13. Lowry OH, Rosebrough NJ, Farr AL, et al: Protein measurement with the Folin reagent. J Biol Chem 193:265-275,195l 14. Chiamori N, Henry RJ: Study of the ferric chloride method for determination of total cholesterol and cholesterol ester. Am J Clin Path01 31:305-309,1959 15. Takayama M, Itoh S, Nagasaki T, et al: A new enzymatic method for determination of serum choline-containing phospholipids. Clin Chem Acta 79:93-98,1977 16. Steinberg D: Lipoproteins and the pathogenesis of atherosclerosis. Circulation 76:508-512, 1987 17. Aviram M: Low density lipoprotein and scavenger receptor activities are modulated by secretory products derived from cells of the arterial wall. Metabolism 38:445-449,1989 18. Aviram M: Lipoprotein secretion by human monocyte derived macrophages. Biochem Biophys Res Commun 154:572-577, 1988 19. Basu SK, Goldstein JL, Brown MS: Independent pathway for secretion of cholesterol and apolipoprotein E by macrophages. Science 219:871-873,1983 20. Brown MS, Goldstein JL: Lipoprotein metabolism in the macrophage. Annu Rev Biochem 52:223-261,1983 21. Werb Z, Chin JR: Endotoxin suppresses expression of apoprotein E by mouse macrophages in vivo and in culture. J Biol Chem 258:10642-10648,1983 22. Faggiotto A, Ross R, Harker L: Studies of hypercholesterolemia in the nonhuman primate. I. Changes that lead to fatty streak formation. Arteriosclerosis 4:323-340,1984 23. Faggiotto A, Ross R: Studies of hypercholesterolemia in the nonhuman primate II. Fatty streak conversion to fibrous plaque. Arterosclerosis 4:341-346,1984 24. Aviram M: Effect of lipoproteins and platelets on macrophage cholesterol metabolism, in Harris (ed): Blood Cell Biochemistry. Chapter 7,1990, pp 179-208