Differential Regulation by Cytokines of Constitutive

Differential Regulation by Cytokines of Constitutive and Stimulated Secretion of von Willebrand Factor From Endothelial Cells By Ewa M. Paleolog, David C. Crossman, John H. McVey, and Jeremy D. Pearson We examined the effect of cytokines on basal and agoniststimulated release of von Willebrand factor (vWf) by human endothelial cells. Treatment of endothelial cells for up to 48 hourswith human recombinant or purifiedinterleukin 1 (IL-1) or human recombinant tumor necrosisfactor-a (TNF-a) did not significantlyaffect constitutive secretion of vWf or intracellular levels of vWf, although basal prostacyclin (PGI,) production was markedly enhanced. In contrast, both IL-1 and TNF-a modulated vWf release in response to thrombin or phorbol ester. Pretreatment of endothelial cells for 2 hours with either cytokine enhanced by up to threefold the stimulatory effect of a subsequent 60-minute exposure to thrombin. Addition of cycloheximide(5 pg/mL)

during the preincubationabolishedthis enhancement. Moreover, if the cytokine pretreatment time was extended to 24 hours, agonist-stimulated vWf release was significantly suppressed. Cytokine treatment for 2 or 24 hours had no detectable effect on levels of vWf messenger RNA. The effects of cytokines were not the result of contamination with bacterial lipopolysaccharideand were not attributable to endothelial cell injury. These results show that cytokines have little or no direct effect on vWf release from endothelial cells but can significantly modulate its acute release in response to other stimuli in a complex time- and dose-dependent manner. 0 1990 by The American Society of Hematology.

T

We have proposed'' that the acute and transient elevations of plasma vWf levels observed during infection may be the result of modulation of either constitutive or stimulated endothelial vWf release by circulating mediators of the acute phase response, such as IL-1 or TNF. Cytokines induce significant changes in endothelial cell function, including stimulation of prostacyclin (PGI,) s y n t h e ~ i s ,induction ~ ~ , ~ ~ of tissue factor procoagulant activity" and enhancement of leukocyte adhesion to endothelial cells.29Evidence regarding the effect of cytokines on vWf release from endothelial cells is conflicting, however. De Groot et a13' reported that pretreatment of endothelial cells for 12 to 48 hours with recombinant human or murine IL-1 decreased both constitutive and stimulated release of vWf, with a concomitant reduction in the amount of vWf stored in the Weibel-Palade organelles, whereas Zavoico et a131 failed to observe any effect of prolonged IL- 1 treatment on either basal or agonist-induced secretion of vWf. In contrast, Giddings and reported that vWf release from endothelial cells incubated for up to 4 hours with purified human IL-1 was significantly higher than release from control cells. Similar observations were also independently reported by two other groups of investigator~.~~.~~ Because such discrepancies may relate to the conditions of the individual studies and especially to the time of incubation with IL-1, we investigated further the effect of cytokines on release of vWf. In particular, we assessed the effect of different cytokine preparations and different cytokine treatment times on both constitutive and agonist-stimulated secretion of vWf by human vascular endothelial cells.

HE ACUTE INFLAMMATORY response to tissue injury caused by infection or trauma consists of several biochemical and clinical alterations, including fever, leukocytosis, increased synthesis of hepatic acute phase proteins such as C-reactive protein (CRP), and changes in plasma metal and steroid concentration^."^ Considerable evidence now shows that many of the aspects of the acute phase response can be attributed to release of a family of monocyte-derived peptides, which include interleukin (IL-1), IL-6, and tumor necrosis factor (TNF).'.'' We recently showed that in addition to the welldocumented components of the acute phase response, a significant elevation occurs in the levels of plasma von Willebrand factor during the course of a range of bacterial, viral, and parasitic infections." von Willebrand factor (vWf) is a large adhesive glycoprotein, circulating in the plasma as a series of heterogeneous multimers of molecular weights 0.4 to 20 x lo6, which plays an important role in the interaction of platelets with the blood vessel wall, as well as serving as the stabilizing carrier protein for coagulation factor VIII.'2-15 The predominant site of vWf synthesis is the vascular endothelium,16 which is the major source of plasma vWf in v ~ v o . ' ~ Cultured vascular endothelial cells secrete vWf constitutively.I6In addition to this continuous release of vWf, rapid secretion of preformed vWf, localized in intracellular storage organelles termed Weibel-Palade b o d i e ~ , 'can ~ ~ be ' ~ provoked in vivo and in vitro by a number of agents, including thrombin, 4-phorbol 12-myristate 13-acetate (PMA), calcium ionophore A23 187, fibrin, and

From the Section of Vascular Biology. and the Haemostasis Research Group, MRC Clinical Research Centre, Harrow, Middlesex. England. Submitted February 22.1989; accepted October 6,1989. Address reprint requests to Ewa M . Paleolog. PhD, Section of Vascular Biology, MRC Clinical Research Centre, Watford Rd, Harrow, Middlesex HA1 3UJ. England. The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C.section 1734 solely to indicate this fact. o I990 by The American Society of Hematology. 0006-497119017503-OO39%3.00/0 688

MATERIALS AND METHODS

Cell culture. Endothelial cells from human umbilical veins were isolated and cultured as described by Jaffe et Except where noted, for our experiments first-passage endothelial cells were harvested with 0.1% trypsin/0.02% EDTA and plated in 16-mm diameter wells (GIBCO, Uxbridge, England) at approximately 8 x lo4 cells per well, in Medium 199 supplemented with 20% calf serum, 90 pg/mL heparin, and 20 pg/mL endothelial cell growth supplement (ECGS) (prepared from bovine brain36). Visually confluent monolayers were formed after overnight incubation. Release of vWf. At confluence, culture medium was removed from each well and replaced with Medium 199 containing 20 mmol/L HEPES and 20% calf serum, with or without the appropriBlood, Vol75, No 3 (February I ) , 1990: pp 688-695

689

REGULATION BY CYTOKINES OF vWF SECRETION

ate agent. Cell supernatants were sampled at various time intervals. In certain experiments, endothelial cells were then transferred to serum-free Medium 199 containing 20 mmol/L HEPES for 60 minutes and subsequently incubated at 37OC in the above medium, with or without agonist. Cell supernatants were sampled after another 60 minutes. Release of vWf into the cell supernatants was measured by an enzyme-linked immunosorbent assay (ELISA) using a modification of the technique described by Short et a].’’ Mouse monoclonal anti-vWf CLBRAg35, used as the solid phase, was a gift from Dr J.A. Van Mourik.” Samples were incubated for 1 hour at 20° to 22OC with the coating antibody, after which rabbit polyclonal antihuman vWf (Dako, High Wycombe, England) was added, followed by goat anti-rabbit IgG conjugated to alkaline phosphatase (Sigma, Poole, UK). 4-Methylumbelliferyl phosphate (Koch-Light, Haverhill, England) was used as the substrate, and the fluorescence was read in a Titertek Fluoroskan. Dilutions of pooled normal human plasma containing 2.4 IU/mL vWf (by refrence to an international standard provided by Dr T. Barrowcliffe, National Institute for Biological Standards and Control, UK) were used to construct standard curves. The lower limit of detection was 2.5 x IU/mL. Measurement of intracellular vWf levels. After cytokine treatment as described and removal of cell supernatants, the endothelial cell monolayers were washed twice with serum-free Medium 199 and lysed with 0.1% Triton X-100 (Sigma) in phosphate-buffered saline (PBS). Levels of vWf in cell lysates were assayed by ELISA as described. Measurement of PGI,. Production of PGI, was measured using a specific radioimmunoassay for 6-keto-prostaglandin F,,, the stable hydrolysis product of PG12.” Cytotoxicity assay. Cellular protein synthesis after cytokine treatment was assessed using incorporation of ’H-leucine, as described by Morgan.“’ As a positive control a cytotoxic anti-MHC class I antibody was used (MAS1 14b, Seralab, Crawley, UK). RNA extraction, electrophoresis. and hybridization. RNA was extracted by a modification of the method of Auffray and Ro~geon.~’ Seven milliliters 3 mol/L Li C1, 6 mol/L urea at 4OC was added to the culture surface of a 75-cm2 flask of cells and left on ice for 5 minutes. The resulting lysate was stored at 4OC overnight to precipitate the RNA, which was collected by centrifugation, redissolved in 10 mmol/L Tris-HC1 (pH 7.5) containing 0.5% sodium dodecyl sulfate (SDS), and incubated for 30 minutes at 37OC with .proteinase K (100 pg/mL). The RNA was then extracted three times with phenol/chloroform (I:]), precipitated in ethanol, and redissolved in water. Total RNA, 10 pg per lane, was electrophoresed in denaturing 1.1% agarose/formaldehyde gels4, and transferred overnight using 20 x SSC (3 mol/L NaCI, 300 mmol/L sodium citrate) to GeneScreen membranes (NEN-Dupont, Stevenage, UK). The RNA was fixed to the membrane by ultraviolet cross-linking at 254 nm, 120,000 pJ, followed by baking at 8OoC for 2 hours. RNA was prehybridized for 2 hours at 65OC and then hybridized overnight, at 45OC when cDNA probes were used and at 65OC when anti-sense RNA probes were used. Prehybridization and hybridization were both carried out in 50% formamide, 50 mmol/L sodium phosphate, 5 x SSC, 1% SDS, 5 x Denhardt’s solution, 200 pg/mL denatured sheared salmon sperm DNA, and 100 pg/mL yeast tRNA. Blots were then washed twice, for 30 minutes each time, in 0.1 x SSC and 0.5% SDS at 55OC for cDNA probe hybridization and at 75OC for anti-sense RNA probe hybridization. The membrane was then exposed to Kodak XARS film (Hemel Hempstead, UK) at -7OOC with image intensifying screens. Filters were stripped of the previously hybridized probe by washing at 8OoC for 30 minutes in 70% formamide, 1% SDS.

The human cDNA probe for vWf (pvWf 1,100) used was a gift from Dr H. Pannekoek of the Central Laboratory of the Netherlands Red Cross Blood Transfusion Service, Am~terdam.~’ It was labeled with 32P-dATP (3,000 Ci/mmol/L; Amersham, UK) by random primer extensi0n.4~Human cDNA for tissue factor representing the first 1,350 base pairs (bp) of the full-length CDNA;~a gift from Genentech, San Francisco, CA, was subcloned into pSP64 and linearized with SacI. Single-stranded, anti-sense RNA probes were synthesized with SP6 RNA polymerase using ”P-dUTP (800 Ci/mmol/L; Amersham) as recommended by Promega Biotec. A 1,350 bp Psi1 fragment of a murine a-actin cDNA was subcloned into pGEMl (Promega Biotec). This was linearized with Hind 111, and single-stranded, anti-sense, 32P-labeledRRA probes were made as described but with T7 RNA polymerase. Materials. Experiments were performed with (a) human recombinant IL-la (HRIL-la) (provided by John Budd, Roche Products, Welwyn, England), 700 U/pmol, 4 x lo7 U/mg; (b) HRIL-I@ (Genzyme), 1,750 U/pmol, lo8 U/mg; (c) purified IL-1 obtained from Staphylococcus albus-stimulated human monocytes (Ultrapure IL-1, Genzyme: HN IL-1), 1.4 x lo5U/pmol, 8 x lo9U/mg; (d) purified human placental IL-I (gift from Dr R. B ~ n n i n g ~ ~ : HPIL-I) 2,500 LAF U/mL; and (e) HRTNF-a (Genzyme), 720 U/pmol, 2 x lo7 U/mg. Human a-thrombin, cycloheximide, polymyxin B, and PMA were purchased from Sigma; L-[4,5-3H(N)]leucine (1 .O mCi/mL, 5.0 Ci/mmol) was from NEN. RESULTS

Eflect of cytokines on constitutive release of v WJ Treatment of endothelial cells for 24 hours with recombinant IL-la had no significant effect on constitutive release of vWf into the culture medium (Fig 1). In contrast, basal synthesis of PGI, increased in the presence of IL-1 in a dose-dependent manner (Fig 1). Similarly, when endothelial cells were treated for 2 to 24 hours with a fixed dose (10 U/mL) of either placental IL-1 (HPIL-1) or recombinant IL-la, no difference was detected in the amount of vWf secreted by control and cytokine-treated cells (Fig 2a). In the same experiment, an increase in basal PGI, synthesis was observed as early as 6h after addition of IL-1 (Fig 2b). Comparable results were obtained using endothelial cells incubated with recombinant IL-1/3 or monocyte-derived IL-1 (HN IL-1)

r200

1001

--

-150

0

x

E

-3 5 E

-B -E

50-

--

-100 -@ N

- 50 0-

a.

-0 0

5

10

20

IL1 concn(U/ml) Fig 1. Effect of IL-1 on constitutive release of vWf and synthesis of PGI,. Endothelial cells were incubated for 24 hours with HRIL-la at the concentration indicated. Results are means i SD of three determinations and are representative of four similar experiments.

690

PALEOLOG ET AL

15

15-

I

0 10

--

. I

E

3

-E

55

0

I

z)

-E 'c

3

5-

0-

Control

HRlLlff

10-

\

HRlLlP

30,

HNlLl

Fig 2. Effect of different Dreoarations of IL-1 on release of vWf . . and synthesis of PGI,. Endothelial cells were incubated for 24 hours with either recombinant or purified IL-1 (10 U/mL). Results are means f SD of three determinations and are representative of four similar experiments.

(Fig 3). In other experiments, no effect of any human IL-1 preparation tested (at 520 U/mL) on constitutive secretion of vWf was detected even after 48-hour treatment (not shown). Incubation of endothelial cells for 24 hours with recombinant TNF-a, at doses up to 100 U/mL, also failed to alter release of vWf, although synthesis of PGI, was again significantly stimulated a t all concentrations of TNF-a (Table 1). Moreover, addition of a combination of submaximal doses of recombinant T N F - a and recombinant I L - l a was without effect on basal secretion of vWf although it had an additive effect on PGI, release (Table 1). If primary rather than subcultured endothelial cells were used, similar increases in basal PGI, production were observed after treatment with IL-1 for I24 hours, again with no detectable effect on constitutive release of vWf (not shown). Effect of cytokines on stimulated release of vWJ Stimulation of endothelial cells with thrombin (0.1 to 10 U/mL) induced release of vWf into the culture medium at levels of 2 to 20 mU/ 1O5 cells in 60 minutes. Preincubation of endothelial cells for 2 hours with recombinant IL-I@ increased the amount of vWf released by the cells in response to a subsequent 60-minute exposure to 0.1 to 10 U/mL thrombin, with a substantial elevation in the maximal secretory response induced by thrombin (Fig 4). Similar results were obtained using purified rather than recombinant IL-1. The extent of this enhancement of vWf release in response to a subsequent stimulus was dependent on the dose of IL-1 used, reaching a maximum with 30 U/mL IL-1 (not shown), and also on the time of pretreatment with IL-1, reaching a maximum with a preincubation time of 2 hours. Thereafter, vWf release in response to thrombin decreased to control levels within 4 hours (Fig 5). Addition of cycloheximide (5 pg/mL) during the pretreatment period completely abolished the effect of the cytokine (Fig 5 ) . Pretreatment of human umbilical vein endothelial cells (HUVEC) with TNF-a induced a similar enhancement of thrombin-stimulated vWf release. The effects of TNF-a and

B

-- 20.

E ---

(3,

I=

-

N

c3

a

10-

0-

1

I

2

24 Incubation time (h)

Fig 3. Effect of incubation time on IL-1-induced changes in PGI, production. Endothelial cells were incubated for the time indicatedwithout).( or with either recombinant (01or purified (0) IL-1. (A) Constitutive release of vWf. (6) Production of PGI,. Results are means f SD of three determinations and are representative of three similar experiments.

IL-1 were not synergistic (Table 2). Pretreatment of endothelial cells with IL-1 or TNF-a also increased the amount of vWf secreted in response to a subsequent challenge by PMA (1 to 100 ng/mL), with a similar enhancement of the maximal secretory response (not shown). Table 1. Effect of TNF on Basal Release of vWf and PGI, VWf

Treatment ~

No addition TN F 10 U/mL 5 0 U/mL 100 U/mL 10U/mL

(mUlmL)

PGI, (nalmL1

~~

30.6 f 8.7

+ HRIL-la 1 U/mL

26.5 f 8.4 36.6 f 0.9 34.7 0.2 31.7 + 12.6

5.8 f 0.4 14.4 f 1 18.8 & 2.1 18.2 f 3.2 21.5 f 2.6

Endothelial cells were incubated for 24 hours with HRTNF-a at the concentrations indicated. The results are the mean of three determinations f SD.

69 1

REGULATION BY CYTOKINES OF VWF SECRETION

20

40

15

30

;

-

I I I I I I

-.E ... 3

E

I

5

-

.

10

E

-2

I

20 I

'lL\

5

\

5

1

1

I

2

10

0

I

3

I

4

I

5

I

6

+'

P' 24

Pre-treatment time (HI

+ +- I 0

0.1

1.o Thrombin concn (Ulm 1)

Fig 5. Effect of IL-1 pretreatment time on thrombin-induced vWf release. Endothelial cells were pretreated without (0)or with (0) 10 U/mL HRIL-la for the time indicated and subsequently challenged for 1 hour with 7.5 U/mL thrombin. (*) In the sample indicated, cycloheximide ( 5 pg/mL) was added during the pretreatment period. Results are means i SD of three determinations and are representative of three similar experiments.

10

Fig 4. Effect of IL-1 pretreatment on thrombin-induced vWf release. Endothelial cells were pretreated for 2 hours without (0) or with (0) 30 U/mL HRIL-ID and subsequently challenged with the indicated dose of thrombin for 1 hour. Results are means f SEM of 4 t o 27 determinations, combined from five similar experiments.

In contrast, if the cytokine pretreatment time was extended to 24 hours, significant suppression of thrombinstimulated vWf release was observed (Figs 5 and 6). The extent of the decrease in stimulated vWf release was dependent on the dose of cytokine (Fig 7). Again comparable results were obtained using PMA as the stimulus for release of vWf (not shown). Eflect of IL-1 on intracellular levels of vWf: Treatment of HUVEC with IL-1 for 2 to 24 hours did not significantly alter the cellular levels of vWf: eg, in three similar experiments, the intracellular levels of vWf after 24-hour incubation with HRIL-ID were 14.34 2 4.07 mU/105 cells, as compared with 11.24 2.37 mU/105 cells in control cultures (means f SD, n = 18). Ejiect of cytokines on stimulated PCI, synthesis. As described, treatment with IL- 1 and TNF-a significantly enhances basal synthesis of PGI,. In addition, cytokine pretreatment markedly increased PGI, synthesis in response to a subsequent dose of thrombin, although the pattern of this enhancement differed strikingly from the effect of cytokines on stimulated release of vWf. As shown in Fig 8, 2-hour pretreatment of endothelial cells with IL- 1 had no significant effect on thrombin-induced PGI, synthesis. An increase in the amount of PGI, synthesized was observed only after 6-

hour pretreatment with IL-1 and persisted even after a 24-hour preincubation time with IL-I, with a clear elevation in the maximal thrombin-induced response. Effects of cytokines on cell viability. Treatment of endothelial cells for 24 hours with either IL-1 (1 to 20 U/mL) or TNF-a (5 to 25 U/mL) did not significantly alter incorporation of 3H-leucine (Table 3). Higher doses of TNF-a (50 to 100 U/mL), but not of IL-1, significantly stimulated protein synthesis, an effect observed more clearly if 'H-leucine incorporation was measured for the 24-hour period subsequent to cytokine treatment (Table 3). The effects of cytokines on stimulated release of vWf and PGI, were not altered when 10 pg/mL polymyxin B was added (not shown). Effect of cytokines on v Wf mRNA. Northern blot analysis (Fig 9) shows an mRNA transcript for vWf of approximately 9 kilobases, which is in agreement with the original Table 2. Effect of Cytokine Pretreatment on Subsequent Agonist-Stimulated vWf Release v w f (mU/mL)

Treatment

No addition IL-1 1 U/mL TNF 10 U/mL TNF 10 U/mL IL-1 1 U/mL TNF 10 U/mL IL-1 1 U/mL 5 pg/mL cycloheximide

+ +

+

10.3 & 3 13.7 f 0.8 17.4 f 0.5 19.7 f 2.7 8 i 2.7

Endothelial cells were incubated for 2 hours with the appropriate concentration of HRIL-1a and/or HRTNFa and subsequently challenged for 1 hour with 3.75 U/mL thrombin. Release of vWf in the absence of thrombin was 2.9 f 1.9 mU/mL. Results are the mean of three determinations f SD.

692

PALEOLOG ET AL

..

25

2.0

IL1 A

20

TNF

0 Control

15

--

1.5

. E c1 0

--

10

N

c3

a

0.5

5

0

I+'

0

I

1

1

0.1

1

10

Thrombin concn (Ulml) Fig 6. Effect of IL-1 and TNF 24-hour pretreatment on thrombininduced vWf release. Endothelialcells were incubatedfor 24 hours with or without 10 UlmL HRIL-la or 50 U/mL HRTNF-a and subsequently challenged for 1 hour with the indicated dose of thrombin. Data are means f SD of three determinations (six for untreated cells) and are representative of six similar experiments.

de~cription.~' Stimulation of endothelial cells with HRIL- l p for 2 or 24 hours had no effect on the level of vWf or a-actin mRNAs. Substantial induction of procoagulant tissue factor mRNA was effected by the 2-hour IL-1 treatment, however, but this response had decayed by 24 hours. DISCUSSION

The host response to acute inflammation is characterized by a coordinated series of metabolic reactions, predominantly, although not exclusively, within the liver. In a variety of experimental systems, monocyte-derived products have

0

I+' 0

I 10

I

I

0.1

1 Thrombin concn (Ulml)

Fig 8. Effect of IL-1 pretreatment on thrombin-induced PGI, production. Endothelialcells were preincubated with or without 10 UlmL HRIL-la for the times indicated and challenged with the indicated dose of thrombin for 1 hour. Data are means f SD of three determinations (nine for untreated cells) and are representative of three similar experiments.

been shown to elicit many of these acute phase reactions. For example, TNF-a or IL-1 regulates the synthesis of positive (serum amyloid A) and negative (albumin) acute phase reactant^.^.^' We recently showed that circulating vWf, a plasma protein of nonhepatic origin, is significantly increased during the acute phase response." Exposure to cytokines is known to regulate the release of several other biologically active mediators from endothelium, including PGI,,26tissue Table 3. Effect of Cytokines on Cell Viability

Bt

Treatment

A*

No addition IL-1 (U/mL) 1 5 10 20 TNF (U/mL) 5 25 50 100

100 f 10

100 f 5

101 105 104 100

f6

106 f 106 f 113 f 120 f

107 105 112 115

f

f 6 f

8

f9

113 f lo$ 129 lo$ 133 f 11$ 143 f 14$

6

f 12 f f

12 12 12 15

6$ 7f

~~

Fig 7. Effect of IL-1 concentration on the decrease in thrombininduced vWf release. Endothelial cells were pretreated for 24 hours with 0 t o 10 U/mL HRIL-la and subsequently challenged for 1 hour with 7.5 UlmL thrombin. Results are means f SD of three determinations. This experiment was repeated once with similar results.

All results are a percentage of the 3H-leucineincorporated by untreated cells in 24 hours and are the mean f SD of 8 to 10 determinations (25 for controls). Mean incorporation of 3H-leucine by untreated cells was 9,072 f 900 cpm for experiment A and 17,539 f 9 2 0 for experiment 8. * A Incorporation of 3H-leucine in the presence of HRIL-lo( and HRTNF-a. t B : Incorporation of 3H-leucineafter pretreatment of endothelial cells for 2 4 hours with IL-1 and TNF. $Significantdifferences from control values: P .O1.

-=

693


vWF

-

Tissue Factor

-2hn-cult

IL1

28s

18s

2lhrs-

cant

IL1

Actin

Fig 9. Northern Mot arulysis of total RNA (10 ~ I b n ofrom ) HUVEC probed for vWf. tissue factor. nnd win. Cells were exposed for 2 or 2 4 hours to vehicle nlono (lor).. 1 and 3)or to 30 UlmL IL-1s (Innos 2 and 41. The actin transcript is 2.1 kb.

plasminogen activator and plasminogen activator inhibitorI." and 11.-1 itself," but reports describing their effects on endothelial vWf release arc contradictory."' '' We show that treatment of HUVEC for up to 48 hours with human recombinant TNF-a or various human I L - I preparations (purified or recombinant) has no significant effect on cumulative constitutive release of vWf or on intracellular levels of vWf. I n the same experiments, each cytokinc preparation markedly enhanced synthesis of PGI,. These observations agree with the recent results of 7avoico et al." who also failed to observe any effect of HRIL-I on constitutive vWf secretion under conditions in which PGI, synthesis was significantly increased. These results are at variance. however. with previous reports'?'" in which a transient two- to fourfold elevation of vWf release occurred when endothelial cells were treated with purified human 11.-1. in the absence of other agonists. for 30 minutes to 4 hours. This release appeared to result from granular secretion because it was accompanied by loss of the granular fluorescent locali7ation of vWf" and was not sensitive to cycloheximide pretreatment." Wecannot explain the discrep ancy bctween this and our own results. although further inconsistencies exist in the available data. Thus. Schorer et all' reported lipopolysaccharide as well as IL-l to bc capable of inducing vWf release. whereas Harlan et al" were unable to detect (as were we. in a limited number of unpublished experiments) any effect of lipopolysaccharide on vWf sccretion. In addition. Giddings and Shall.'' although they reported increased release to IL-I,failed to induce release with PMA. which has been well characterized by other investiga-

tors?'.'' and by us as an active secretagogue. Finally, despite the considerable evidence for multiple alterations in endothclial cell properties in response to cytokines.'h'.'".'' until now the functional effects of cytokines on this cell type. as on others, appear to bc RNA and protein-synthesis dependent and not immediate in onset. In contrast to other investigators."." De Groot et al" reported a progressive decrease in constitutive vWf release to the medium and to the extracellular matrix on treatment with HRIL-IH or mouse recombinant IL-I in 48 hours and correlated this with a decrease in vWf mRNA levels. In addition. lower cellular levels of vWfafter 48-hour treatment with human purified IL-I were noted by Schorcr et al." The results shown in Fig 6 indicate that when measured over 60 minutes without added agonist, vWf release is reduced after 24-hour pretreatment with IL-l and. in additional cxperiments (not shown), we showed that this effect persisted with pretreatment times 572 hours. confirming the observations of De Groot et al."' Our major new finding. however. is that pretreatment of endothelial cells with cytokincs modulates rapid release of preformed vWf in a time-dependent manner. Thus. pretreatment of endothelial cells for 2 hours significantly augmented vWf release induced by thrombin or PIMA.This enhancing effect required protein synthesis. After 24-hour pretreatment of endothelial cells with 11.-1 or TNF-a. however. stimulated vWf release was reduced. in agreement with the results reported by De Groot et al." although 7avoico et al" did not report reduction of stimulated vWf release after 24-hour treatment with IL-I. The effects ofcytokines on vWf release were not the result of contamination with bacterial lipopolysaccharide and arc not attributable to endothelial cell injury. Northern blot analysis showed no detectable alteration in vWf mRVA levels after either 2- or 24-hour stimulation by II.-I. De Groot et aI"' reported a modest reduction in vWf mRNA after 24-hour IL- I treatment. whereas Reinders et al" reported no change in vWf mRNA levels up to 48 hours after PIMA stimulation. Our results. however. mainly show that at 2 hours, when vWf secretion was substantially enhanced in response to an agonist and easily detectable elevation of another mRNA (tissue factor) was observed. vWf mRNA levels were not changed. Because cytokine pretreatment for 2 to 4 hours increases the maximal ability of an agonist to secrete vWf. while steady-state levels of vWf mRNA and constitutive vWf release remain unchanged. we conclude that the cytokines act by facilitating the transduction pathways that couple agonist-binding to the exocytotic event causing secretion of granular vWf. These pathways (like those utilized by activated cytokine receptors) are largely unknown. Their definition and their links with the pathways leading to stimulation of PGI, synthesis require further study. The time courses of cytokine-induced changes in vWf secretion and PGI, synthesis are clearly different. Rossi et al and Dejana et al"." previously reported that basal PGI, synthesis wasenhanced after pretreatment with cytokines for 24 hours. We have extended these results and confirmed the recent findings of 7 ~ v o i c oet al'' by showing that PGI,

694

PALEOLOG ET AL

release, either basal or in response to an agonist (Fig 8), is significantly enhanced after 6 hours of treatment with cytokine, and this enhancement is greater after 24 hours. Earlier studies have led to conclude that cytokine treatment of endothelial cells alters their properties in a procoagulant direction, owing to the reciprocal actions of IL-1 or TNF-(Y on the procoagulant and fibrinolytic activities of endothelial cell^,*^.^'*^^ and our own results also indicate that an increased ability to secrete vWf may accompany or precede these changes. At later times, however, when vWf release is unaltered or depressed and PGI, release is enhanced, an antithrombotic balance is restored. This investigation was initiated by our finding that vWf is elevated as part of the acute phase response in humans," which led us to propose that IL-1 mediates vWf release. This hypothesis is in part substantiated by the present results. We have been able to discern effects of IL-1 or TNF-a on

agonist-stimulated vWf secretion but were unable to detect direct effects of the cytokines alone on vWf release. More recently, several of the actions of IL-1 were shown to result from stimulated release of IL-6 (eg, recombinant IL-6 regulates the transcriptional activation of a set of acutephase genes in a human hepatoma cell line9 and stimulates CRP synthesis in cultured human hepatocytes''). Endothelial cells produce IL-6,56*57 which is stimulated within 4 hours by pretreatment with IL- 1 .56 Although exogenous IL-6 did not alter endothelial cell production of PGI, or procoagulant it will be of interest to determine whether IL-6 affects vWf release. ACKNOWLEDGMENT

We thank Stephen Coade for technical assistance and the personnel of the Delivery Suite of Northwick Park Hospital for their continued expert help in collecting umbilical cords.

REFERENCES

1. Tillet WS, Francis T: Serological reactions in pneumonia with a non-protein somatic fraction of Pneumococcus. J Exp Med 52:561, 1930 2. Kushner I: The phenomenon of the acute phase response. Ann NY Acad Sci 389:39,1982 3. Pepys MB, Baltz ML: Acute phase proteins with special reference to C-reactive protein and related proteins (pentaxins) and serum amyloid A protein. Adv Immunol34:141, 1983 4. Dowton SB, Colten H R Acute phase reactants in inflammation and infection. Semin Haematol25:84, 1988 5. Dinarello CA: Interleukin 1 and the pathogenesis of the acute phase response. N Engl J Med 31 1:1410,1984 6. Perlmutter DH, Dinarello CA, Punsal PI, Colten HR: Cachectin/tumour necrosis factor regulates hepatic acute phase gene expression. J Clin Invest 78:1349, 1986 7. Beuscher HU, Fallon RJ, Colton HR: Macrophage membrane interleukin 1 regulates the expression of acute phase proteins in human hepatoma hep3B cells. J Immunol 139:1896,1987 8. Gauldie J, Richards C, Harnish D, Lansdorp P, Baumann H: Interferon @,/B-cell stimulatory factor type 2 shares identity with monocyte-derived hepatocyte-stimulating factor and regulates the major acute phase protein response. Proc Natl Acad Sci USA 84:7251, 1987 9. Morrone G, Gilbert0 G, Oliviero S, Arcone R, Dente L, Content J, Cortese R: Recombinant interleukin 6 regulates the transcriptional activation of a set of human acute phase genes. J Biol Chem 263:12554,1988 10. Moshage HJ, Roelofs HMJ, Van Pelt JI, Hazenberg BPC, Van Leguwen MA, Limburg PC, Aarden LA, Yap SN: Effect of interleukin 1, interleukin 6 and its interrelationship on the synthesis of serum amyloid A and C-reactive protein in primary cultures of adult human hepatocytes. Biochem Biophys Res Commun 155:112, 1988 11. Pottinger BE, Read RC, Paleolog EM, Pearson JD: Von Willebrand factor is an acute phase reactant in man. Thromb Res 53:389, 1989 12. Tschopp TB, Weiss HJ, Baumgartner H R Decreased adhesion of platelets to subendothelium in von Willebrand disease. J Lab Clin Med 83:296,1974 13. Hoyer LW: The factor VI11 complex: Structure and function. Blood 58:1,1981 14. Stel HV, Sakariassen KS, De Groot PG, Van Mourik JA, Sixma JJ: von Willebrand factor in the vessel wall mediates platelet adherence. Blood 65:85, 1985

15. Turitto VT, Weiss HJ, Zimmerman TS, Sussman 11: Factor VIII/von Willebrand factor in subendothelium mediates platelet adhesion. Blood 65:823, 1985 16. Jaffe EA, Hoyer LW, Nachman RL: Synthesis of von Willebrand factor by cultured human endothelial cells. Proc Natl Acad Sci USA 71:1906,1974 17. Bowie EJW, Solberg LA, Fass DN, Johnson CH, Knutson GJ, Stewart ML, Zoecklein LJ: Transplantation of normal bone marrow into a pig with severe von Willebrand's disease. J Clin Invest 718:26, 1986 18. Wagner DD, Olmsted JB, Marder VJ: Immunolocalisation of von Willebrand protein in Weibel-Palade bodies of human endothelial cells. J Cell Biol 95:355, 1982 19. Ewenstein BM, Warhol MJ, Handin RI, Pober JS: Composition of the von Willebrand factor storage organelle (Weibel-Palade body) isolated from cultured human umbilical vein endothelial cells. J Cell Biol 104:1423,1987 20. Levine JD, Harlan JM, Harker LA, Joseph ML, Counts RB. Thrombin-mediated release of factor VI11 antigen from human umbilical vein endothelial cells in culture. Blood 60531, 1982 21. Loesberg C, Gonsalves MD, Zandbergen J, Willems C, Van Aken WG, Stel HV, Van Mourik JA, De Groot PG: The effect of calcium on the secretion of factor VIII-related antigen by cultured human endothelial cells. Biochim Biophys Acta 763:160, 1983 22. Reinders JH, De Groot PG, Gonsalves MD, Zandbergen J, Loesberg C, Van Mourik JA: Isolation of a storage and secretory organelle containing von Willebrand protein from cultured human endothelial cells. Biochim Biophys Acta 804:361, 1984 23. Sporn LA, Marder VJ, Wagner DD: Inducible secretion of large, biologically potent von Willebrand factor multimers. Cell 46:185, 1986 24. Hamilton KK, Sims PJ: Changes in cytosolic Ca2' associated with von Willebrand factor release in human endothelial cells exposed to histamine. J Clin Invest 79:600, 1987 25. Ribes JA, Francis CW, Wagner DD: Fibrin induces release of von Willebrand factor from endothelial cells. J Clin Invest 79:117, 1987 26. Rossi V, Breviario F, Ghezzi P, Dejana E, Mantovani A: Prostacyclin synthesis induced in vascular cells by interleukin 1. Science 229:174, 1985 27. Kawakami M, Ishibashi S, Ogawa H, Murasei T, Takaku F, Shibata S: Cachectin/TNF as well as interleukin 1 induces prostacyclin synthesis in cultured vascular endothelial cells. Biochem Biophys Res Commun 141:482,1986


28. Bevilacqua MP, Pober JS, Majeau GR, Fier SW, Cotran RS, Gimbrone MA. Recombinant tumour necrosis factor induces procoagulant activity in cultured human vascular endothelial cells: Characterisation and comparison with the actions of interleukin 1. Proc Natl Acad Sci USA 83:4533, 1986 29. Bevilacqua MP, Pober JS, Wheeler ME, Cotran RS, Gimbrone MA: Interleukin 1 activation of vascular endothelium. Effects on procoagulant activity and leukocyte adhesion. Am J Pathol 121:304, 1985 30. De Groot PG, Verweij CL, Nawroth PP, De Boer HC, Stern DM, Sixma JJ: Interleukin 1 inhibits the synthesis of von Willebrand factor in endothelial cells which results in a decreased reactivity of their matrix towards platelets. Atherosclerosis 7:605, 1987 31. Zavoico GB, Ewenstein BM, Schafer AI, Pober JS: IL-1 and related cytokines enhance thrombin stimulated PGI, production in cultured endothelial cells without affecting thrombin-stimulated von Willebrand factor secretion or platelet-activating factor biosynthesis. J Immunol 142:3993, 1989 32. Giddings JC, Shall L: Enhanced release of von Willebrand factor by human endothelial cells in culture in the presence of phorbol myristate acetate and interleukin 1. Thrombos Res 47:259, 1987 33. Booth F, Allington MJ, Cederholm-Williams SA: An in vitro model for the study of acute release of von Willebrand factor from human endothelial cells. Br J Haematol67:71, 1987 34. Schorer AE, Moldow CF, Rick ME: Interleukin 1 or endotoxin increases the release of von Willebrand factor from human endothelial cells. Br J Haematol67:193, 1987 35. Jaffe EA, Nachman RL, Becker CG, Minick CR: Culture of human endothelial cells derived from umbilical cord veins. Identification by morphologic and immunologic criteria. J Clin Invest 52:2745, 1973 36. Maciag T, Weinstein R: Preparation of endothelial cell growth factor, in Barnes DW, Sirbasku DA, Sat0 GH (ed): Cell culture methods for molecular and cell biology, vol 1. New York, NY, Liss, 1984 37. Short PG, Williams CE, Picken AM, Hill FGN: Factor VI11 related antigen: An improved ELISA. Med Lab Sci 35:351, 1982 38. Stel HU, Sakariassen KS, Scholte BJ, Veerman ECI, Van der Kwast TH, De Groot PG, Sixma JJ, Van Mourik JA: Characterisation of 25 monoclonal antibodies to factor VI11 von Willebrand factor: Relationship between ristocetin-induced platelet aggregation and platelet adherence to subendothelium. Blood 63:1408, 1984 39. Ager A, Gordon JL, Moncada S, Pearson JD, Salmon JA, Trevethick MA: Effects of isolation and culture on prostaglandin synthesis by porcine aortic endothelial and smooth muscle cells. J Cell Physiol 1109, 1982 40. Morgan DML: Oxidised polyamines and the growth of human vascular endothelial cells. Prevention of cytotoxic effects by selective acetylation. Biochem J 242:347, 1987 41. Auffray C, Rougeon F: Purification of mouse immunoglobulin heavy-chain messenger RNAs from total myeloma tumor RNA. Eur J Biochem 107:303,1980 42. Manitais T, Fritsch EF, Sambrook J: Molecular Cloning-A

695

Laboratory Manual. Cold Spring Harbor, NY, Cold Spring Harbor Laboratory, 1982 43. Verweij CL, Hofker M, Quadt R, Briet E, Pannekoek H: RFLP for a human von Willebrand factor (vWf) cDNA clone, pvWF 1100.Nucleic Acids Res 13:8289, 1985 44. Feinberg AP, Vogelstein B: A technique for radiolabelling DNA restriction endonuclease fragments to high specific activity. Anal Biochem 132:6,1983 45. Fisher KL, Gorman CA, Vehar GA, OBrien DP, Lann RM: Cloning and expression of human tissue factor cDNA. Thromb Res 48:89, 1987 46. Minty AJ, Caravatti M, Robert B, Cohen A, Daubas P, Weydert A, Gros F, Buckingham ME: Mouse actin messenger RNAs. J Biol Chem 256:1008,1981 47. Bunning RAD, Crawford A, Richardson HJ, Opdenakker G, Van Damme J, Russell RGG: Interleukin 1 preferentially stimulates the production of tissue-type plasminogen activator by human articular chondrocytes. Biochim Biophys Acta 924:473,1982 48. Verweij CL, de Vries CJM, Distel B, van Zonneveld A-J, Van Kessel AG, Van Mourik JA, Pannekoek H: Construction of cDNA for human von Willebrand factor using antibody probes for colony screening and mapping of the chromosomal gene. Nucleic Acids Res 13:4699, 1985 49. Reinders JH, Vervoorn RC, Verweij CL, Van Mourik JA, De Groot PG: Perturbation of cultured human vascular endothelial cells by phorbol ester or thrombin alters the cellular vWf distribution. J Cell Physiol 133:79, 1987 50. Ramadori G, Sipe JD, Dinarello EA: Pretranslational modulation of acute phase hepatic protein synthesis by recombinant murine interleukin 1 and purified human interleukin 1. J Exp Med 162:930, 1985 51. Bevilacqua MA, Schleef RR, Gimbrone MA, Loskutoff DJ: Regulation of the fibrinolytic system of cultured human vascular endothelium by interleukin 1. J Clin Invest 78587, 1986 52. Libby P, Ordovas J, Auger KR, Robbins AH, Birinyi LK, Dinarello C A Endotoxin and tumour necrosis factor induce interleukin 1 gene expression in adult human vascular endothelial cells. Am J Pathol 124:179, 1986 53. Harlan JM, Harker LA, Striker GE, Weaver LJ: Effect of lipopolysaccharide on human endothelial cells in culture. Thromb Res 20:15, 1983 54. Dejana E, Breviario F, Balconi G, Rossi V, Remuzzi G, De Gaetano G, Mantovani A: Stimulation of prostacyclin synthesis in vascular cells by mononuclear cell products. Blood 64:1280, 1984 55. Nawroth PP, Handley DA, Esmon CT, Stern DM: Interleukin 1 induces endothelial cell procoagulant while suppressing cellsurface anticoagulant activity. Proc Natl Acad Sci USA 83:3460, 1986 56. Sironi M, Breviario F, Proserpio P, Biondi A, Vecchi A, Van Damme J, Dejana E, Mantovani A: Interleukin 1 stimulates interleukin 6 production in endothelial cells. J Immunol 142543, 1989 57. Norioka K, Hara M, Karigai M, Kitani A, Hirose T, Suzuki K, Kawakami M, Tabata H, Kawagoe M, Nakamura H: Production of B cell stimulatory factor-2/interleukin 6 activity by human endothelial cells. Biochem Biophys Res Commun 153: 1045, 1988