Regulation of macrophage tumor necrosis factor production by prostaglandin E2

Vol.

737,No.l.

1986

BlOCHEMlCALANDBlOPHYSlCALRESEARCHCOMMUNlCATlONS

May29,1986

Pages

REGULATION

OF

MACROPHAGE

S.L.

TUMOR

Kunkel,

R.C.

Wiggins,

Departments University *Cetus

Received

April

Immune

11,

NECROSIS

of

PRODUCTION BY PROSTAGLANDIN E2

FACTOR S.W.

Chensue,and

J.

Pathology and Internal of Michigan Medical Ann Arbor, MI 48109

Corporation,

Palo

404-410

Larrick*

Medicine School

Alto,

California

94303

1986

We have studied the role of prostaglandin E on the modulation of tumor necrosis factor by immunologically elicited an 8 lipopolysaccharide treated murine macrophages. Indomethacin, a potent inhibitor of prostaglandin E caused a dose dependent augmentation of-jipopolysaccharide induce i; production, molar). Tumor necrosis tumor necrosis factor production (2-3 fold at 10 factor was released into the extracellular environment and no activity was found to be associated with membrane or cytosolic fractions. Prostaglandin E 4 added to the lipopolysaccharide treated cultures supprgfsed tumor necrosl factor in a dose dependent manner. In these studies, 10 molar PGE tumor necrosis factor production to basal levels. These data suggest may be a potent autoregulatory factor that dramatically influences turn0 necrosis factor production. @ 1986 Academic Press. Inc.

Macrophage-derived recognized

as

diverse

ways.

Carswell

et

oncolytic

tumor an

important

agent.

variety

of

it

cellular product

adipocytes

(2)

no

a)

that:

the

ation

human

While

it

in

is

response

to

**Abbreviations TNF,

tumor

CFA,

complete

0006-291X/86 Copyright All rights

known

used necrosis Freund's

and

endotoxin

in

to

affinity

high

of

dependent

murine

that

this

factor;

other

TNF TNF

has

as

may

been

and,

lines

(5).

and

macrophage

bacterial

d)

or

LPS,

$1.50 Inc. reserved.

lipopolysaccharide,

modulates

lines

protozoa1

CO,

by as

identified

(4),

cell

defined

an a

as

on

(2,3),

b)

c)

augments

the

prolifer-

can

elaborate

products

cyclooxygenase;

a

present lipase

cytotoxicity

(4)

in

stimulate

receptors

lipoprotein

cells

solely

paper:

adjuvant.

0 I986 by Academic Press, of reproduction in any form

cell

macrophages and

that TNF

phagocytosis

regard

increasingly

target

targets

to

example,

being of

original

production

antibody

neutrophil

the

is

variety

demonstrates

binds

suppresses

a

accurate

For

neutrophil-mediated

numerous

longer

responses.

and

of

were

evidence

polymorphonuclear

TNF

is

(TNF)**

effecting

cells

Current

macrophage

enhances

tumor

(l),

factor

monokine

Although al.

necrosis

(6),

Vol.

137,

No.

there

is

little

Considering logic

BIOCHEMICAL

1, 1986

information

the

aid

the

In

this

trations

by of

is

augmented

the

production

the

phagocytic

methacin.

control

activated

murine

cyclooxygenase

of

TNF

challenge provide

by

TNF

is

of

while

immune

in a

and

the

murine appears

low

induced molar TNF

macrophages

model to

be

is or

of

TNF

concenproduction

Furthermore,

indomethacin.

presence

may

processes.

(LPS) by

immuno-

production

LPS-induced

inhibitor

either

and

TNF

and

suppressed

activated

production.

physiologic

lipopolysaccharide

(CO)

specificity

modulate

homeostatic

that

for

COMMUNICATIONS

modulation

E2 (PGE2),

zymosan

RESEARCH

various

the

immunologically

evidence

on

macrophages

by

the

TNF

that

physiologic

demonstrate

prostaglandin

stimulus of

both

exogenous

We

demonstrates

we

of

regarding

of

paper,

BlOPHYSlCAL

factors

effects

information

understanding

production

regarding

pleomorphic

activities,

AND

TNF under

refractory

absence

to of

production the

indothat

regulatory

PGE2. MATERIALS

AND

METHODS

Animals - Female, specific pathogen-free CBA/J mice (Jackson Laboratories, Bar Harbor, ME) were used in cell experiments. Mice were maintained under pathogen-free conditions and given food and water ad libitum. Macrophage cultures - Resident peritoneal macrophages were recovered from normal mice by repeated washing of the peritoneal cavity (10 x 1 ml of RPMI1640 culture medija). This macrophage population was washed 3x and then suspended to 1 x 10 MB/ml RPM1 with 5% fetal bovine serum (FBS) containing 100 U penicillin and 100 ug streptomycin/ml. One ml of suspension was placed onto 35 mm sterile plastic culture dishes. After 2h incubation (37'C, 5% CO,,, 100% humidity), non-adherent cells were removed by two vigorous rinses. Elicited peritoneal macrophages were recruited using either light mineral oil, thioglycollate, or complete Freunds adjuvant (CFA) as follows: 2 mls of either Marco?-52 mineral oil or thioglycollate or 0.5 mls of CFA (CFA diluted 1:l with sterile saline) were asceptically administered via an intraperitoneal injection, four days post oil or thioglycollate or 14 days post CFA administration the macrophages were harvested by repeating washing of the peritoneal cavity, the cells were then washed and resuspended in complete culture media, and plated as described in the above methods for the resident macrophages. After the initial 2h adherence period, the monolayers were overlaid with 1 ml of serum-free media containing antibiotics plus the stimulus being tested for TNF induction, zymosan (250 pg/ml) or --E. coli LPS (Sigma Chemical Co., St. Louis, MO). Cell-free supernates were collected after 16h incubation. The total number of macrophages was quantitated by scraping the cells from the culture dishes and counting in a hemacytometer. All macrophage populations were >95% mononuclear phagocytes as determined by staining, morphologic, and phagocytic indices. Arachidonate metabolites and inhibitors Prostaglandin E was kindly provided by the Upjohn Co. (Kalamazoo, MI). Purity of the pros 2 aglandins was confirmed by high pressure liquid chromatography as previously described (7). The --in vitro specificity of indomethacin (Sigma Chemical Co.), as an inhibitor of products derived from the CO pathway, was assessed as previously published (8). . _ Radioimmunoassays for arachidonate metabolites - Cell-free culture medium from the various macrophage populations was subject to radioimmunoassay for 405

Vol.

137,

No.

1, 1986

BIOCHEMICALAND

BIOPHYSICAL

RESEARCH

COMMUNICATIONS

PGE (New England Nuclear, Boston, MA). Bound and free ligand were separated us1 4 g dextran-coated charcoal. Sensitivity for PGE was 8 picograms. Tumor necrosis factor assay - The activity o 3 TNF was monitored using a semi-automated L929 fibrob4ast lytic assay as previously described (9). Briefly, L929 cells (5 x 10 /O.l ml) in the presence of 1 pg/ml actinomycin D were cultured with serial 1:2 dilutions of test samples in 96-well, flat bottom microtiter plates (Costar, Cambridge, MA) at 37'C, 5% CO /95% air humidified incubator. After 18h incubation the plates were washed an ci! the remaining cells stained with crystal violet (0.5%) in methanol/water (1:4, V/V). The amount of cell lysis was determined using a micro ELISA autoreader. Units of TNF activity was defined as the reciprocal of the dilution necessary to lyse 50% of the L929 target cells. An internal standard of TNF was included in each assay; this material was a gift from Dr. Leo Lin; Cetus Corp., Emeryville, CA. Statistical analysis - The Student's t-test was used to compare control and experimental groups. Values of p > -05 was considered not significant.

RESULTS Production

of

populations. murine

TNF

by

Numerous macrophage

characteristics

elicited

studies

have

populations

mediators

and

by

is

macrophage

resting

and

morphologic

also

reflected As

macrophages.

0 .Ol

murine that

diverse

diversity

produced

resident

demonstrated

possess

This

(10).

inflammatory

various

1

.l

and in

shown

functional

the in

elicited

spectrum Figure

of 1,

10

LPS fugI

Figure

1.

Production of LPS. with

Only those

a significant

control, for

of tumor necrosis

all

30 min.

glycollate phages. separate

immunologically production

TRF activity CFA, complete elicited;

Data

factor

OIL,

represents

of

activated TRF in

was destroyed Fraund's oil

in response

by heating

elicited,

mean + SRM of

experiments. 406

macrophages

response

adjuvant

to graded

elicited;

responded

to LPS. samples

As a to 6O'C

THIO, thio-

and RES, resident triplicates

doses

from

macrothree

TNF

Vol.

137,

No.

production

varied

stimulus

elicited

dependent first

upon

the

10 ug/ml.

The only

capacity

to produce

incubated

peritoneal

macrophages

10 rig/ml

LPS.

duction

of

did

significant the

inhibitor,

production

was

concentration fold.

Indomethacin

zymosan

treated

which

a

reached LPS

had

augmenting

no

CFA macrophages.

the the

indomethacin. production. TNF

in

L929

the

cells

dose

at

dependent

An

inverse

Thus,

this

(100,OOOxg

the

relationship provided

strong

production.

407

found

to be

cytosolic

for

20 min)

of

at

(Figure

production

LPS

the

to

circumstantial

of At

of

no effect nor

under

exist

cyclo-

augmented

production had

a

between evidence

TNF this 2-3

TNF by

on macroaltered

study. PGE2

to

the

2).

was

stimulation

LPS-induced

as

augmentation

10-7M

on

tumor

population

Using

dependent

alone

appears

or

macrophages

prostaglandins.

of

resident

macrophage

concentrations

suppression

pro-

zymosan

on LPS-induced

this

induced-TNF

of

the

TNF activity.

CFA elicited

of

the

Neither

effect

with

stimulus

populations.

dose

absence

only

elicited

for

was not

Indomethacin

any

stimulus

TNF activity

a maximum

indomethacin,

production of

of

and oil

TNF by any of

of TNF, we utilized effect

up to

and then

indomethacin

defined

indomethacin,

of

TNF

Having

fashion

macrophages

phagocytic

demonstrated

inhibitor

regulatory

found

of

macrophage

macrophages

was

a significant

soluble

by ultracentrifugation

source

potential

oxygenase

suppressed

above

production.

cellular

shows

the

of the cyclooxygenase

factor

viability

production

macrophages

TNF even when incubated

the

In addition,

prepared

or sonicated

Effect

macrophages,

was

demonstrated

Resident

to be a potent

the

any of

samples

freeze-thawed

also

induce

that

significant

adjuvant that

in a linear

elicited

produce

populations.

in

membrane

phage

not

Freund's

by CFA elicited

population

of the LPS

TNF production

of LPS (10 ug/ml).

LPS proved

not

cell-associated

study

did

While

macrophage

necrosis

levels

TNF by CFA elicited

ug/ml)

elicited

nor

high

Complete

LPS and increased

macrophage

COMMUNICATIONS

concentration

augmented

TNF was thioglycollate

with

the

study.

TNF production

rig/ml

RESEARCH

both

under

demonstrated

lo-100

other

BIOPHYSICAL

upon

population

dose of LPS.

between

AND

depending

macrophages

apparent

(250

dramatically

and the macrophage

(CFA)

when

BIOCHEMICAL

1, 1986

the

Figure

production PGE2 that

2 by

and

TNF PGE2

Vol. 137,

No.

600

BIOCHEMICAL

1, 1986

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

-60

If

300 z

250

2 v) 2

200l50-

5 -

0.001

0

2

.Ol

.l

INDOMETHACIN

Figure

2.

of

induced

TNF and

methacin

induced

concomitant

the

cyclooxygenase

PGE2

production

a dose

rise

elicited

in

Data

by

by

mean

LPS

+ SEM of

had

PGE2

on

levels

with

treated

no

on the

effect from

LPS Indo-

(lOpg/ml)

triplicates

1

(ukl)

macrophages. in

alone

.l

indomethacin

reduction

production

.Ol PGE2

CFA elicited

Indomethacin

represents

.OOl

inhibitor

dependent

TNF

macrophages.

assay.

0.0001

3

(uM)

Effect

¶

03

1

+ 10 ug LPS .

CFA

IOO-

three

a CFA-

lytic

separate

experiments. Figure

Effect

3.

of

production.

PGE

had

no

of

pressive

then

2’

stimulated

CFA

dependent

effect

reduced

by

production the appears

PGE,

of

was

PGE2

macrophages.

viability

of

directly

three

TNF

levels. the

the

were

tested

L929

suppress

by in

TNF

graded

TNF

Data

were

concentrations

activity.

represents

factor

mean

PGE2

alone

f

SEM of

At

while control

exogenous

10m8#,

the

production. 408

this

PGE2

PGE2 at

low

agent

PGE2

supto

caused

production

concentrations

studies, Thus,

adding 3,

higher

The

production.

directly

Figure

production.

cells.

macrophages of

for

necrosis

confirmed

In

TNF

experiments.

tumor

shown

ug/ml)-induced

presence

assay.

separate

on

(10

adjuvant-elicited

LPS in

lytic

90%,

basal of

the

As

approximately to

to

on from

suppression

with

supernates

exogenous

effect

Freund’s

16 hours

triplicates

Effect

PGE2 on LPS

Complete

stimulated of

exogenous

a dose

of

TNF

suppressed alone molar

had

no

LPS-

effect

concentrations

was TNF on

Vol.

137,No.

BIOCHEMICAL

1, 1986

AND

BIOPHYSICAL

RESEARCH

COMMUNICATIONS

DISCUSSION Recent

evidence

participate ular

in sources

facets

the

production

more

with

gically

elicited

mouse

effects

of

down-regulate while

data

imply in

since

a

LPS

with

mediator,

TNF,

can

It

is

TNF

autoregulatory

the

as

provide

at

PGE2

production

may

be

that

cell-

LPS

treated,

to

study

to

have

with

evidence

near

that

appears

triggered

a model

cells major

response

utilized

of

These the

of

conclusion,

immunologically-activated

its

for

be

inter-

a specific immunolothe

regula-

that

physiologic

PGE2

can

concentrations,

production,

has

a significant

as

PGE2

on

which

TNF

production

are

in

accord

of

PGE2

if

to

fold

own

in

the

mediator induction

of

demonstrated

supported

by

(13)

studies

interleukin-1 of

important

macrophages

production

by when

regulate

has

resting

a classical

characteristics

production

that (12).

the

The

findings

augments

lectin-

(14).

recent in

intrigu-

biologically

study

PGE2,

macrophages

a role

is

PGE2

PGE2

ultimately

plays

production

5-7

act

monokine

may

by

and

a

by

especially

elicited

produce

production

with

TNF

a recent

TNF

is

CFA

TNF

decreases by

both

that

although PGEE

of

the

yet

regulated

hypothesis

only

a signal

clear

endogenously

Thus, not

stimulate

demonstrated of

stimulus

macrophages,

ability

production

This

produces

interferon

data

a

manner.

challenge

not

can

monocyte

In

have

We

macrophage.

effect

gamma

production

as

also

by

recombinant

leukin-1

then

inhibitor

type

LPS

but

synthesis

strating

we

of

macrophages and

production

TNF

serve

an

production.

induced

that

activated

presented

that

immune

"activated"

production.

potent

feedback

immunologically

PGE2

TNF

One

on

effect.

potentially ing

an

macrophages

TNF

a

during

report

influence

response.

products

peritoneal on

a profound

immune TNF

their

present

PGE2

indomethacin,

Our

the

the

LPS-induced

augmenting

the of

or

In

has

those

T-lymphocytes (6,ll).

TNF

of

specifically,

agonist

tory

that

various for

macrophages; acted

suggests

from feedback

(15). a

our

In

classical

laboratory

demon-

manner these

to

regulate

studies,

hormone

by

inducing

interthe

inhibitor. we

provide

evidence

macrophages

that is

409

regulated

TNF

production by

PGE2.

by

LPS-stimulated,

These

studies

also

Vol.

137,

lend

No.

1, 1986

additional

8lOCHEMlCALAND8lOPHYSlCALRESEARCHCOMMUNlCATlONS

support

macrophage

to

the

notion

that

PGE2

is

a major

autokine

determining

responsiveness. ACKNOWLEDGEMENT

This is

an

wish Kathleen

work

was

established to

supported

by

investigator

express

their

NIH of

appreciation

grants the

HL

31237

American

Heart

for

expert

the

and

HL

31963.

Association.

Dr. The

secretarial

Kunkel authors

support

of

Atkins. REFERENCES

1.

E.A., Old, L. J., Kassel, R. I., Green, S. and Williamson, 8. Proc. Natl. Acad. Sci. USA 72, 3666-3679. Beutler, 8., Mahoney, J., Letrang, N., Pekala, P. and Cerami, A. (1985). J. Exp. Med. 161, 984-995. Kawakami, M., Pekala, P.H., Lane, M.D. and Cerami, A. (1982). Proc. Natl. Acad. Sci. USA 79, 912-916. Shalaby, M.R., Aggrawal, 8.8.. Rinderknecht, E., Svedersky, L.P., Finkle, B.S. and Palladino, M.A. (1985). J. Immunol. 135, 2069-2073. Sugarman, 8.J.. Aggarwal, B.B., Hass, P.E., Figari, I.S., Palladino, M.A. and Shepard, H.M. (1985). Science 230, 943-945. Ruff, M.R. and Gifford, G.E. (1981). Lymphokines 2, 235-271. Fitzpatrick, F.A. and Bundy, G.L. (1978). Proc. Natl. Acad. Sci. USA 75, 2689-2693. Kunkel. S.L.. Chensue, S.W., Mouton, C. and Higashi, G. I. (1984). J. Clin. Carswell,

(1975).

2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13.

Invest:

74, 514-524.

Aggarwal, B.B., Moffat, B., Lee, S.H. and Lymphokines, Goldstein, A. ed. pp. Adams, D.O. and Hamilton, T.A. (1984). Old, L.J. (1985) Science 230, 630-632. Bachwich, P.R., Chensue, S.W., Larrick, Biochem. Biophys. Res. Comm. (in press). Boraschi, P. Censini, S. and Tagliabue,

and Harkens, R.N. Thymic Hormones 235-249, Plenum Press, New York. Ann. Rev. Immunol. 2,283-318. J.W. A.

and

Kunkel,

(1984).

J.

S.L.

(1986).

Immunol.

133,

764-768. 14. 15.

Nedwin, Goeddel, Kunkel,

186-192.

G.E., D.V. S.L.,

Svedersky,

(1985). J. Chensue,

L.P., Bringman, T-S., Palladino, Immunol. 135, 2492-2497. S.W. and Phan, S.H. (1986). J.

410

M.A. Immunol.

and 136,