Induction of transforming E2 production by ethanol

Abstract: and their

Gyongyi

Szabo,

Department

of Surgery,

To test mediators

our are

Bikash

hypothesis major

growth factor-beta and prostaglandin in human monocytes

K. Verma,

University

Miklos

of Massachusetts

that monocytes contributors to

Fogarasi, Medical

(MO) ethanol-

the

synthetic bacterial analog, muramyl dipeptide (P < 0.05 and P < 0.001, respectively). Ethanol also increased TGF3 production in interferon ‘y (IFNy)activated MO in response to MDP stimulus (P < 0.05). MO TGF/3 levels, however, were always lower in IFNyactivated than in non-IFN-y-activated MO after the same stimulation with ethanol plus MDP, suggesting that MO preactivation by IFNy can partially counteract the TGFf3 inducing potential of ethanol. Similar to its TGF/3inducing potential, ethanol (150 mM) had the capacity to induce PGE2 production in adherent human MO (P < 0.045). However, ethanol failed to augment MO

22]. Similar to TGF/3, PGE2 is also a MO-derived inhibitory mediator that can inhibit T cell proliferation, interleukin-2 ( IL-2) production, and IL-2 and transfernin receptor expression in T lymphocytes [23-25]. Elevated PGE2 levels are inhibitory for MO by depressing antigen presentation capacity, HLA class II expression, TNFa and IL-i production [26-28]. In addition to potent MO PGE2 secretagogues such as LPS and MDP, in vitro ethanol treatment has been reported to stimulate PGE2 production in munine macrophages [ii]. Similarities between the negative regulatory effects of MO-derived TGF/3 and PGE2 on the immune system and immune abnormalities characteristic of acute and chronic alcohol use lend support to the contention that one of the mechanisms by which ethanol might affect the immune system is via induction of these inhibitory mediators from MO. Consequently, this study was designed to investigate the capacity of ethanol to induce MO TGFI3 and PGE2 production. MO TUFI3 and PGE2 production was determined in adherence-isolated and/or interferon-’y (IFN-’y)-activated

production

induced by the PGE2 secretagogue, TGFI3 induction by ethanol was unaffected by the presence of cyclooxygenase inhibitor, suggesting that ethanol-induced MO TGFI3 production does not require MO PGE2 production. These results indicate that ethanol is a potent inducer for inhibitory MO mediators, TGF/3 and PGE2, and also has the capacity to augment MO TGFI3 production in response to subsequent stimulation. Thus, ethanol-induced elevation of MO TGF/3 and PGE2 production might contribute to decreased T cell proliferation and abnormal MO functions resulting in a depressed immune Biol. 52: 602-610; 1992. Key Words: dipeptide

activated

macrophages

after alcohol response.

.

IFN

.

J.

alcohol

exposure, Leukoc.

.

muramyl

INTRODUCTION Acute and chronic with immunodepressive rhosis have reduced [1], depressed mitogen

ethanol

treatment both are associated effects. Alcoholic patients with cirskin test reactivity to common antigens induced lymphocyte proliferation [2],

Abbreviations: Iscove’s

and reduced natural killer cell activity, polyclonal hypergammaglobulinaemia, and circulating immune complexes [3]. Cytokine and monokine levels are also dysregulated after chronic alcohol consumption contributing to some of the cellular immune defects [4-6]. Even a single, acute ethanol exposure can result in suppression of chemotactic functions of granulocytes [7], decreased monocyte (MO) phagocytosis

602

Journal

of

E. Catalano

Massachusetts

Elevated transforming growth factor-/3 (1.DFI3) production by Kupffer cells was recently implicated as a trigger for collagen deposition in alcoholic cirrhosis in a rat model [12]. In addition to its stimulatory potential on connective tissue, TGFf3 has a pleiotropic and negative regulatory effect on the immune system [13-15]. TGF1f3 suppresses many functions of T cells and MO, including T cell proliferation [16, 17], synthesis of IL-2 receptors [18], and generation of cytotoxic T cells [16]. In addition, MO 1DF3 can inhibit the release of H2O2 from peritoneal macrophages [19] and production of MO TNFa [20]. TGFf3 is known to be synthesized by a vanety of cells, including MO. MO constitutively synthesize a small amount of TGFJ3, and much greater amounts are synthesized after triggering with an activating agent such as lipopolysacchanide (LPS) or muramyl dipeptide (MDP) [21,

(MDP)

PGE2 MDP.

Donna

Worcester,

[ 8], impaired MO tumor necrosis factor a (TNFa) production [9], and elevated MO prostaglandin E2 (PGE2) synthesis [10, 11]. The mechanism for ethanol-induced depression of immune functions is yet to be investigated. Here we test our hypothesis that many components of the ethanol-induced immunoaberration are mediated by ethanol-induced MOderived inhibitory mediators, such as TGFI3 and PGE2, which have the capacity to induce immunodepression.

related immunodepression, the modulating capacity of acute ethanol treatment was assessed on the production of transforming growth factor-beta (TGFI3) and prostaglandin E2 (PGE2) by human peripheral blood MO. We demonstrate that acute in vitro treatment of adherent MO with either 50 or 150 mM ethanol induced a significant increase in the production of TGFf3 (P < 0.045 and P < 0.001, respectively). Furthermore, MO pretreatment with both 50 and 150 mM ethanol augmented TGF/3 production in response to subsequent stimulation with

and

Center,

Leukocyte

Biology

Volume

52,

December

charide; E2;

fetal

Dulbecco’s

MDP,

P1,

dine;

FBS,

modified

muramyl

propidium

iodide;

TGFfl,

transforming

bovine media;

serum;

IL-2,

IFN-y,

interleukin-2;

interferon--y;

dipeptide;

MO,

monocytes;

PGE2,

PMA,

phorbol

myristate

acetate;

growth

factor-fl;

IMDM,

LPS,

TNFa,

lipopolysacprostaglandin TdR,

tumor

thyminecrosis

factor-a. Reprint

requests:

University

cester,

of

MA

Received

1992

Gyongyi

Massachusetts

Szabo, Medical

M.D.,

Ph.D.,

Center,

Department 55

Lake

01655. June

30,

1992;

accepted

August

11,

1992.

Avenue

of Surgery, N.,

Wor-

human, peripheral blood MO in response stimulation with ethanol. The modulating ethanol treatment on MO responsiveness terial stimulation and the relationship induced MO TGF/3 and PGE2 production ligated.

to acute in vitro capacity of acute to subsequent bacbetween ethanolwere also inves-

METHODS Reagents RPMI

1640,

MEM,

and

Iscove’s

modified

Dulbecco’s

media

(IMDM) were obtained fromJRH Biosciences (Lenexa, KS) and fetal bovine serum (FBS) from Hyclone (Logan, UT). FBS for MO stimulation was selected to exclude lot numbers containing detectable ‘TUF!3 levels. Endotoxin contamination was less than 8 pg/ml in the culture media and FBS. Serum-free AVM medium was obtained from Gibco and utilized for MO stimulation under serum free conditions. Polymixin B sulfate was purchased from Gibco (Grand Island, NY) and used at 100 U/ml concentration in the MO culture media where indicated. The presence of polymixin B sulfate did not affect MO TGFI3 or PGE2 responses in our experiments. Human recombinant (Escherichia colz) IFNy, obtained from Collaborative Research, Inc. (Bedford, MA), contained 4.8 x 10 viral inhibition units per 100 ml of EMC virus propagation in WISH cells and contained 0.5 endotoxin unit/100 ml by Limulus amebocyte lysate assay. Nacetylmuramyl-L-alanyl-D-isoglutamine, Nor-MDP (compound CGP 11637), a derivative ofmuramyl dipeptide, was a generous gift from Ciba-Geigy AG. (Basel, Switzerland). Indomethacin and phorbol mynistate acetate (PMA) were purchased from Sigma (St. Louis, MO). E. coli (Oiii:B4)derived LPS was used from Difco (Detroit, MI). Highly purified human TGF/3 was obtained from Collaborative Research, Inc. (Bedford, MA). TGFf3-neutralizing antibody (R&D Systems, Minneapolis, preparation and had high titer logical activity of porcine and

MN) was a for neutralization human TGF-f31.

purified of the

regimens

such

THP-1

cell Culture

Human medium

monocytic supplemented

and

Stimulation

stimulation

was

combined

cells were cultured with 5% FBS and

in RPMI antibiotics

1640 at

Monocyte

viability

permeable

distinguished

to P1 (viable: analysis

on

an

PI) Epics

of M#{248} TGFI3

were Profile

by

Analyser

(Coulter,

fluoresFL).

Activity

bioactivity in the MO supernates was determined using the selectively sensitive MvlLu, mink lung connective tissue cell line (American Type Culture Collection, Rockville, MD: no. CCL-64), as described by Massague [32]. Briefly, 2.4 x 10 MviLu cells/well were seeded on flatTGFf3

of removal of nonadherent In most of the experiments with

THP-1

cence

might originate from the FBS, control experiments were performed by stimulating MO in serum-free medium. MOpurity after selective adherence to microexudate treated surface was >95% by FITC-labeled OKM5 staining [29]. Adherent

ethanol

“classical”

bio-

of Monocytes

at the time ethanol.

the

IgG

lion cells/well in 15% FBS containing IMDM on microexudate treated 6-well plates [31]. Nonadherent cells were removed by gentle washing with warm medium after 1.5 h adherence and MO were stimulated in IMDM supplemented with 10% FBS plus antibiotics. To rule out the possibility of nonspecific 11F$ activity in our samples that

mM

or with

of 1 x 106/ml in PBS iodide (P1) (Sigma) at a then incubated for 10-30 analysis. Cells that were and cells that were not

24 h before donating blood. Monocytes (MO) from human peripheral blood were isolated by selective adherence from Ficoll-Hypaque purified mononuclear cell preparations as previously described [29, 30]. Briefly, mononuclear cells were Ficoll-Hypaque density separated from hepaninanticoagulated peripheral blood. Cells were washed in icecold HBSS supplemented with 3% FBS and plated at 8 mil-

stimulated 25-150

alone

MO were suspended at a concentration containing 2% FBS and propidium final concentration of 25 g/ml and mm at 23#{176}Cbefore flow cytometer permeable to P1 (nonviable: PP)

Healthy normal individuals from the laboratory and hospital staff of the University of Massachusetts Medical Center participated in this study after signing a written consent form. Volunteers were nonalcoholics and did not consume alcohol

MOwere cells with

MDP

2 x 10 density. THP-1 cells were adjusted to 1 million cells per milliliter in the same medium for stimulation in 6-well plates. TGFI3 production was induced by stimulation of THP-1 cells with ethanol and other stimuli. Supennates of THP-1 cells were collected after different periods of incubation, as indicated, and TGFf3 levels were determined.

Detection Separation

as 20 g/ml

MOstimulation regimen: a combination of lymphokine [100 U(ml/106 MO) of IFN7] and bacterial stimulation (20 g/ml MDP). In those experiments, IFNy was applied at the time of removal of nonadherent cells followed by MDP stimulation 3-4 h after the initial priming of MO with ethanol or IFNy. Supernates of unstimulated and stimulated MO were collected after 16 h incubation by 10 mm treatment with 4 mM EDTA, followed by gentle scraping. The number of recovered MO was counted and this number was used as a conrection factor to normalize TGFf3 and PGE2 activity levels of MO produced by 1 million cells pen milliliter. The supernate and cell samples were kept frozen at -80#{176}C until the monokine assays were performed. Our previous studies confirmed that 16 h stimulation is an optimal stimulation period for assessment ofmonocyte TGF13 and PGE2 production [22, 29].

other

bottom bovine

96-well serum

plates (FBS)

in MEM supplemented and antibiotics [22].

with After

20

10% fetal h of cul-

tune, acid-treated (pH 2.3-2.5) MO supernates were added in duplicates to MviLu cells in twofold serial dilutions in MEM medium with 1% FBS. Acid-treatment of MO supernates was performed for 2 h at room temperature, then pH was readjusted to pH 7.2-7.4 in order to obtain the biologically active form of TGFI3 from the inactive ‘TUFf3-TGFf3binding protein complex. We have previously shown that TGFf3 activity in these acid-treated MO supernates can be completely inhibited by anti-TGFI3-neutralizing antibody (R&D Systems, Minneapolis, MN), as determined in the MviLu bioassay [22]. MvlLu cell proliferation was measured by [3H]TdR incorporation during the last 6 h of the 24 h inhibition [3H]TdR supernates formula:

assay. Sample activity was calculated from incorporation in the presence of acid-treated MO and human 113F13 standard by the following

Sample

stimulation

Szabo

et al.

act

Ethanol-induced

=

St.

act

monocyte

2Sample

EPSt

TGFI3

and

EP

PGE2

603

Where Sample act = sample activity, St = standard and EP = end point. End point was determined as the dilution of the standard or the sample resulting in 50% maximal proliferation of MviLu cells. Highly purified human TGFI3 was utilized as standard preparation for quantitation of TGFf3 activity in the MO supernates (Collaborative Research,

Inc.,

Western

Bedford,

600

0 400

,o

-0--

Exp

-.--

Exp2

-0--

Exp3

-.-

Exp4

-fr--

Exp 5

1

a.

MA).

200

LI.

Blotting

I-

MO samples for Western blot analysis described previously using AVM-serum million MO per stimulation group were SDS-PAGE sample buffer and samples mm. Equal amounts ofprotein (equivalent were loaded in each lane and determined staining

of a second,

electrophoresis

identical

the

gel.

samples

were

were stimulated as free medium. Five lysed in 0.8 ml 1% were boiled for 5 of 1 million cells) by Commassie blue

After

12%

0 0 Fig.

nates

The levels of PGE2 in the mined by an enzyme-linked ously described [29, 30].

M odulating

Effect

control S

Mean

0

±

SE

P value

n

50,

16 h culture. in the

and

150

ltFfl

MvlLu

done with signed-rank

Mean

of MO TGF3

±

utilizing

the

MOTGF

Production

of Ethanol

on

levels

in pM/iO’

MOafter

20 ig/m1

MDP

SE

n

6

Wilcoxon

stimulation

P value

d

Mean

U/mI

SE

±

IFN.y n

357

± 53 ± 80

31 11