Vol.

174,

January

No.

2, 1991

BIOCHEMICAL

BIOPHYSICAL

RESEARCH

COMMUNICATIONS

Pages 885-891

31, 1991

IGG-STIMULATED

AND LPS-STIMULATED MONOCYTES ELABORATE TRANSFORMING GROWTH FACTOR TYPE l3 (TGF-O) IN ACTIVE FORM

L. Schalch’,

‘institute

C. Rordorf-Adam*,

of Veterinary

Virology,

‘Ciba-Geigy

December

10,

J.R Dasch3, and T.W. Jungi’*

University of Berne, CH-3012 Ltd., CH-4012,

‘Celtrix, Received

AND

Berne, Switzerland

Basle, Switzerland

Palo Alto, CA 94303

1990

Mononuclear cells (MNC) stimulated either with lipopolysaccharide (LPS) or with surface-adsorbed IgG elaborated significant amounts of tumor necrosis factor (TNF) bioactivity, as well as immunoenzymatically detectable TNF-a and interleukin-10. (IL1 -6). In contrast, IgG-stimulated cells released little IL1 bioactivity, but released an IL1 inhibitor, as determined by the thymocyte costimulatory assay (LAF assay). This inhibition was not due to an inhibitory effect of cyclooxygenase products, e.g. prostaglandin-E2 in the LAF assay. In contrast, antibodies against transforming growth factor type 6 (TGF-l3), which is an important inhibitor of the LAF assay, augmented the LAF activity of supernatants from LPS-stimulated and IgG-stimulated MNC. Anti-TGF-O-modulated LAF inhibition was enhanced by acid treatment of supernatants from mononuclear cells, but not of those from purified monocytes. Antibody blocking experiments point for the first time to a TGF-9 species other than type 1 as a monocyte-derived TGF-I3 activity. Thus, TGF-6 released in active form from monocytes may be the more important antagonist of IL1 than cyclooxygenase-derived mediators. It implies that the LAF assay, in the absence of anti-TGF-6 antibodies, is an inadequate indicator of IL1 activity. 0 1991 Academic PreSS, Inc.

Interleukin properties

1 (ILl)

is a pleiotropic

and is absolutely

required

mediator

with

potent

as a costimulator

inflammatory

of some,

and

immunostimulatory

if not all T cells (1). Stimulated

phagocytes express membrane-bound ILl-cr and secrete IL16 which both mediate lymphocyte activation (2). In vivo, however, triggering of mononuclear phagocytes is often associated with a state of immunosuppression. This encompasses viral (3-5), bacterial (6) or parasital (7,8) infection as well as severe trauma (9) and is mediated, for Conventionally, this type of immunosuppression products feedback

example, by immune complexes or endotoxin. is explained by the induction of cyclooxygenase

such as prostaglandin E2 (PGE2) either concomittantly circuit by IL1 itself (10).

with IL1 induction,

We have reported that the triggering of monocytes by IgG is ensued bioactivity, as measured by the classical lymphocyte activation factor

* Corresponding

or as a negative

by a modest release of IL1 (LAF) assay (11). The same

author.

Abbreviations: ILl, interleukin-1; PGEP, prostaglandin E2; LAF, lymphocyte activating factor; TNF, tumor necrosis factor; LPS, lipopolysaccharide; MNC, mononuclear cells; TGF-13, transforming growth factor type 6.

Vol. 174, No. 2, 1991

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

type of stimulation led to the release of tumor necrosis factor (TNF) type a comparable in magnitude to the amount of TNF-a generated after lipopolysaccharide (LPS) stimulation (11). Subsequent experiments reported here showed that following IgG stimulation, more ILl-8 was produced than expected on the basis of the LAF assay. An analysisof this discrepancy provided evidence for an IL1 inhibitor in supematants of stimulated mononuclear cells (MNC). We also found that active transforming growth factor type 8 (TGF-8) is concomittantly produced by IgG stimulation and LPS stimulation of MNC, and that this cytokine is the more important IL1 antagonist in the LAF assay than PGE2 or other cyclooxygenase products.

MATERIALS

AND METHODS

Monoclonal antibodies and oharmacoloaical aaents: The monoclonal anti-TGF-8 antibody, 1Dl 1.16, and 3C7.14 were generated as described (12). 1Dl 1.16 neutralized both TGF-81 and 2, but not TGF83; 3C7.14 neutralized both TGF-82 and 3, but not TGF-81. Monoclonal3G8, directed against human Fc receptor type Ill, was used as an isotype control (mouse IgGl). PGE2 and lndomethacin were from Sigma (St. Louis,MO.) Purified porcine TGF-8 was obtained from R & D, Minneapolis. Mononuclear cell isolation and culture: MNC were isolatedfrom buffy coats of whole blood donations by a modified Ficoll-Hypaque centrifugation procedure, aimed at minimizing platelet contamination and avoiding trace contamination by endotoxin (11). In some experiments, elutriation centrifugationpurified monocytes k 98 % pure) were prepared as described (11). Cells were resuspended in RPM1 1640 containing 1 % HSA (Behringwerke, Marburg, Germany) and placed in cluster plates. These had been pretreated with 1 mg/ml IgG (standard gammaglobulin, Central Laboratory Swiss Red Cross Blood Transfusion Service, Berne, Switzerland) for 3 hr at room temparature, followed by extensive washing with phosphate-buffered saline (PBS), or had been mock-treated. Cells of mock-treated wells served as controls, or were stimulated with LPS (1 pg/ml; E. coli 0111 :B4, Sigma, Cat.-No. L 3012). Supernatants were collected at various times after onset of cultures, freed of cells by centrifugation, snap-frozen and stored at -20°C until cytokine determination. Some re-thawed supernatants were acidified (1 hr to pH2, followed by neutralization and testing of LAF activity) in order to activate TGF-8 (13). Determination of IL1 in MNC suoernatants: The classical LAF assay (14) was performed similarly as described previously (11). In brief, 1.5 x IO6 thymocytes from 4 to 8 weeks old C3H/HeJ mice were cultured for 3 days with 2.5 fig/ml phytohemagglutinin in RPM11640 s plemented as described (11). Cells were pulsed for the last 6 hr of culture with 0.3 uCi/well Y [ HImethyl thymidine, and harvested with an lnotech (Wohlen, Switzerland) cell harvester permitting the simultaneous harvesting of 96 wells. Harveyed filters were analysed for radioactivity in a Trace 96 filter counting system (Inotech), using the [ H] channel. Several concentrations of a reference preparation were included in each test. Thymocyte costimulatory (LAF) activity was calculated by parallel line analysis and expressed in arbitrary units, one unit being the LAF activity of the 1:1,28Ofold diluted reference preparation. ILI-8 as determined by a sandwich type ELISA assay, using a monoclonal anti-ILI-8 (2D8) as a catching antibody, and rabbit anti-Ill-8 (No. 203) as a detecting antibody (15). The sensitivity was 15 pg/ml, and the assay was specific for ILl-8 (15). Measurement of TNF in MNC suoernatants: TNF was determined by the classical L929 cytotoxicity assay as described (1I). Cytotoxicity was determined by measuring the reduction in the transformation of the substrate MTT to its colored formazan derivative (16). The assay has a sensitivity of 0.1 U/ml, corresponding to 2.5 pg/ml when using TNF-a obtained from the National Bureau of Standardization and Control as a reference. A kit from T Cell Sciences (Cambridge, MA) was used for determining TNF-a by an ELISA procedure. Using the standard of this kit, the sensitivity of detection was 5 pg TNF-a per ml. Control for endotoxin contamination: All reagents to which monocytes were exposed, and all supernatants of IgG-stimulated and control cells were screened for endotoxin contamination, using a kinetic, turbidimetric version of the Limulus amoebocyte lysate test performed in an ELISA reader. Limulusamoebocyte lysate was from Haemachem Inc. (St. Louis, MO). The lower detection limit was 1.6 pg/ml. All reagents were found to contain < 10 pg LPS/ml, corresponding to < 0.1 endotoxin units. IgG used for coating contained 0.02 units/ml endotoxin, but wells coated with IgG contained subdetectable amounts of LPS, since during our coating procedure, < 0.1 % of exogenously added LPS becomes polystyrene-bound in a stimulatory manner. 886

Vol.

174,

No.

BIOCHEMICAL

2, 1991

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

RESULTS Kinetics of IqG-induced cytokine

production,

cytokine

production:

using bioessays

MNC exposed

to surface-adherent

IgG were tested for

for IL1 and TNF as well as immunoenzymatic

assays for TNF-a

and ILI-0. A constant finding was the low thymocyte costimulatory activity elicited by IgG, although a considerable variation between donors was noted: In contrast, IL143 was measured in considerable quantities, often comparable to that induced by LPS (Fig. 1). Such a dissociation was restricted to IgG-stimulation and was not observed for TNF assessed regardless of the stimulus used (not shown). Dissociation

between

in a bioassay

LAF activity and ILl-8 is due to an inhibitor:

and an immunoassay,

MNC of some donors

exhibited

a

very small release of ILI-8 upon stimulation with surface IgG. The addition to these supernatants of 50 U/ml recombinant IL14 led to a smaller increase in LAF activity than expected (Fig. 2), suggesting that an inhibitor Suppression

of IL1 -8 action is also present of laGinduced

in these supernatants.

LAF activity is not due to concomittantlv

oroduced

PGE2:

MNC were

exposed to IgG or LPS in the presence or absence of the cyclooxygenase inhibitor, indomethacin, and supernatants were assessed after 8 hr and after 20 hr for IL14 and for LAF activity. Although more ILi-8 was detected in supernatants from indomethacin-treated cells than in control supernatants, PGE2 added

LAF activity in the presence of indomethacin to the LAF assay did not decrease thymidine

supernatants

or recombinant

Transforming Ill-B-driven simultaneous

IL14 were used as LAF source (Fig. 3 and data not shown).

arowth factor-01 interferes LAF assay dose-dependently addition

was not enhanced (Table 1). Moreover, incorporation, regardless of whether MNC

of a neutralizing

with the LAF assav: TGF-81 added depressed thymidine incorporation antibody,

1 Dl 1.16, dose-dependently

induced LAF inhibition, whereas an isotype control antibody reaffirms that TGF-8 is an important LAF inhibitor (17,18). Stimulated stimulated

MNC elaborate or IgG-stimulated

in small amounts to the (IDS0 0.1 rig/ml). The eliminated

(3G8) was ineffective

the TGF-Ol-

(not shown).

bioactive TGF-8: It was determined whether supernatants from LPSMNC displayed increased LAF activity when treated with TGF-8-

12

f 8\

4

Is, .!5 P z-

0

6

12

6

18

Time

Fiaure 1. Dissociation of ILl-0 not by LPS stimulated are shown.

This

or control

12

18

(hr)

production and LAF activity elaborated by IgG-stimulated MNC. Means of a representative experiment performed

887

MNC, but in triplicate

Vol.

174, No. 2, 1991 Control

BIOCHEMICAL

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

cells

no IL1 added

(1:4) (1%)

50 lU/ml

IL1 added

(1:4) (13)

IgG-stimulated

cells

no IL1 added

(1:4) (153)

50 NJ/ml

IL1 added

(1:4) (123) 1

3

2

Thymidine

uptake

(1 03CPM)

Fiaure 2. Thymocyte costimulatory activity of recombinant ILl-0 (50 lU/ml; National Bureau of Standardization and Control) added to medium or to supernatant of IgG-stimulated MNC from a donor elaborating little IL1 -0. Means of triplicates +. S.D. or a representative experiment are shown.

neutralizing subtypes

antibodies.

Fig. 4 shows that two monoclonal

dose-dependently

increased

antibodies

with specificity

LAF activity of supernatants

for distinct

from stimulated

TGF-l3

MNC, but not

those of control MNC. Another monoclonal antibody serving as an isotype control was inactive (not shown). A Higher dose of antibody was required to inhibit LPS-induced TGF-6 activity than to inhibit the IgG-induced Preliminary

TGF-l3, suggesting

evidence

that LPS induced

that stimulated

monocvtes

a higher amount

are the source

of bioactive

of bioactive

activates the inactive TGF-O precursor (13). Using this procedure, supernatants was found to increase, and this inhibition was abrogated supernatants

from purified

extent as with acidification,

Table

1.

Stimulus

monocytes

inhibited

and this inhibition

The generation presence

the LAF assay without was abrogated

Time of supernatant collection

Relative response

IL-1B in the presence of indomethacin* (n=8)

8 hr

104.6 i

39.3

TGF-0:

acid treatment to a larger

stimulated

in the

Relative LAF activity in the presence of indomethacin (n=4)

+

n.d.

LPS n.d.

20 hr

97.8 + 32.2

8 hr

145.8 + 70.0

91.3

20 hr

219.6 + 65.8

75.5 + 26.1

+ 20.9

IgG

+ % of control

(no indomethacin

present);

888

means

F S.D.

Acid treatment

LAF inhibition by MNC by lD11.16. In contrast,

by 3C7.14

of LAF activity and ILI-I? by MNC or absence of indomethacin

TGF-13 than IgG.

to the same extent than by

Vol.

174,

No.

2, 1991

BIOCHEMICAL

AND

e-0

BIOPHYSICAL

no PGE2

A

1 rig/ml

‘1

q

5 ‘I/”

‘I

RESEARCH

COMMUNICATIONS

added

J 100

25

6.25

rlL1 -p Fioure derived

1.56

(llJ/ml)

3. PGEP added to thymocyte costimulation cultures does not influence lAF activity supernatants. Means of 3 as obtained in a representative experiment are shown.

of MNC-

ID1 1.16. Antibodies were ineffective when added to supernatant-free LAF medium. This is compatible with the notion that monocytes elaborate TGF-02 and/or TGF-63 rather than TGF-61 in active form. DISCUSSION

In the present report, we provide evidence that human monocytes elaborate TGF-62 and/or 63 in active form upon stimulation with either LPS and IgG, and that this TGF-3 becomes manifest in a classical LAF assay. Moreover, small amounts of TGF-6 dramatically impaired ILI-O-induced LAF activity, confirming earlier reports (17,16) and raising the possibility that in many systems containing an IL1 inhibitor, TGF-l3 may be one active principle. A variety of other IL1 inhibitors have been

Control c-l

4.0 --

m 0 -

o---o

lD11.16

A-A

X7.14

/ I A L

3.0 --

s s 2

A-IA

2.0 --

: ‘Ei ‘C 6

I

LPS

l.O-4-l

I

1 0.0

1.0

10.0

0.0

Antibody

’

I

1.0

10.0

concentration

0.0

1.0

10.0

(,ug/ml)

Figure 4. Monoclonal antibody against TGF-81/82 (1 D11.16) or TGF82/83 (3C7.14) admixed at 10 pg/ml to 20 hr supernatants of LPS-stimulated, IgG-stimulated and control MNC dose-dependently enhance thymidine incorporation into thymocytes. Means + SD. of supernatants from 3 different donors are shown. Supernatatants were tested at 1:8. A control antibody, 3G8, failed to increase thymidine incorporation (not shown).

889

Vol.

174,

No.

2, 1991

BIOCHEMICAL

AND

BIOPHYSICAL

RESEARCH

COMMUNICATIONS

reported. A mediator referred to as IL1 antagonist, is also produced by IgG stimulation and acts as a competitive receptor antagonist for some, but not all types of IL1 receptors (19). Plasminogen activator inhibitor has been found to act as an IL1 inhibitor by preventing the protease-driven release of cell-bound

IL1 into the surroundings

(20). Other IL1 inhibitors

either the production or action of ILI, have been described products have been implied in suppression of IL1 generation

of different

molecular

size, inhibiting

(3-5,21). In addition, cyclooxygenase (22) or activity (23), and these factors

were shown to be induced by stimuli such as the ones used here (24-26). However, in the present system, an effect of cyclooxygenase products, in particular of PGE2, could be clearly ruled out. An important aspect is the cellular source variety of cell types, including monocytes

of bioactive TGF-8, since the latter may be produced by a (27), macrophages (28) and T lymphocytes (29) and may

be released by degranulating blood platelets (13,30). The differential effect obtained with antibodies specific for TGF-81/82 or TGF-82/83 as well as with the acidification procedure suggests that in mononuclear

cells,

several

cell types contribute

to TGF-8

production.

However,

the results

with

platelet- and lymphocyte-depleted, elutriated monocytes suggest that monocytes are the source of bioactive TGF other than type 1, probably TGF-82 and/or 133. The correlation between the amount of TGF-8 produced and the conditions of stimulation remain to be determined. In any case, the study suggests TGF-8

that supernatants

of stimulated

is one of the most potent

several

mechanisms,

including

(32), and deactivation IgG-stimulated In summary,

physiological

that LPS-stimulated

and IgG-induced

balance

TGF-8

strongly

agents expression

immunostimulation

many monocyte-derived

and exerts its effect by (31), antimitotic

MNC elaborate

bioactive

effects

from LPS-stimulated

activity beyond counteracting

or

IL1 activity.

TGF-8 and extend this to IgG-

provide the first evidence that TGF-82 (and/or 83) monocytes. Our demonstration that LPS-induced

with LAF activity raises the possibility

monocytes, contribute not only to immunstimulation/adjuvant of an immune response. In a given monocyte-dependent between

for TGF-8 activity.

(33). Thus, TGF-8 elaborated

experiments by stimulated interferes

be monitored

of IL1 receptor

function

cells may have immunomodulatory we confirm

should

immunosuppressive

down-regulation

of macrophage

stimulated cells. Antibody blocking is/are a major species elaborated stimulated regulation

monocytes

and immunomodulation

that Fc receptor-

activity, but also to downantigen stimulation model, the

may be under

complex

control

of

mediators.

ACKNOWLEDGMENTS This work was supported by the Central Laboratory of the Swiss Red Cross Blood Transfusion Service, Berne Switzerland, by the Swiss National Science Foundation, and by the Arbeitsgemeinschaft fur Osteosynthese, Berne, Switzerland. We gratefully appreciate the expert technical assistance by Mrs. M. Brcic and H. Pfister and the critical reading of the manuscript by Dr. E. Peterhans. REFERENCES 1. 2. 3. 4. 5.

Dinarello, C.A. (1988) FASEB J. 2, 108-l 15. Weaver, C.T. and Unanue, E.R. (1990) Immunol. Today 11,49-55. Rodgers, B.C., Scott, D.M., Mundin, J., and Sissons, J.G. (1985) J. Virol. 55,527-532. Roberts, N.J.Jr., Prill, A.H., and Mann, T.N. (1986) J. Exp. Med. 163,511-519. Locksley, R.M., Crowe, S., Sadick, M.D., Heinzel, F.P., Gardner, K.D.Jr., Grath, MS., (1988) J. Clin. Invest. 82, 2097-2105. 6. Nishihara, T., Koga, T., and Hamada, S. (1988) Infect. Immun. 56,2801-2807. 890

and Mills, J.

Vol.

174,

No.

2, 1991

BIOCHEMICAL

AND

BIOPHYSICAL

RESEARCH

COMMUNICATIONS

7. Wilson, C.S., Ellner, J.J., Dinarello, CA., Keusch, G.T., and el Kholy, A. (1990) Am. J. Trop. Med. Hyg. 42, 234-242. 8. Sileghem, M., Darji, A., Hamers, R., and De Baetselier, P. (1989) Immunology. 68, 137-139. 9. Antonacci, A.C. (1986) In Advances in Host Defense Mechanisms, Vol. 6 (J.I. Gallin and AS. Fauci, Eds.), pp. 81-109. Raven Press, New York. IO. Duclos, H., Maillot, M.C., and Galanaud, P. (1986) Ann. Inst. Pasteur. Immunol. 137, 3-10. 11. Kuhnert, P., Schalch, L., and Jungi, T.W. (1990) Clin. Immunol. Immunopathol. 57,218-232. 12. Dasch, J.R., Pace, D.R., Waegell, W., Inenaga, D., and Ellingsworth, L. (1989) J. Immunol. 142, 1536-1541. 13. Pirchler, R., Jullien, P., and Lawrence? D.A. (1986) BBRC 136, 30-37. 14. Rosenwasser, L.J. and Dinarello, C.A. (1981) Cell. Immunol. 63, 134-142. 15. Rordorf-Adam, C., Lazdins, J., Woods-Cook, K., Alteri, E., Henn, R., Geiger, T., Feige, U., Towbin, H., and Erard, S. (1989) Drugs - Exptl. Clin. Res XV, 355-362. 16. Mossman, T. (1983) J. Immunol. Methods 65, 55. 17. Wahl, S.M., Hunt, D.A., Wong, H.L., Dougherty, S., McCartney-Francis, N., Ellingsworth, L., Schmidt, J.A., Hall, G., Roberts, A.B., and Sporn, M.B. (1988) J. Immunol. 140, 3026-3032. 18. Ellingsworth, L.R., Nakayama, D., Segarini, P., Dasch, J., Carrillo, P., and Waegell, W. (1988) Cell. Immunol. 114, 41-54. 19. Hannum, C.H., Wilcox, C.J., Arend, W.P., Joslin, F.G., Dripps, D.J., Heimdal, P.L., Armes, L.G., Sommer, A., Eisenberg, S.P., and Thompson, R.C. Nature 343,336-340. 20. Goeken, N.E., Staggs, T.S., and Ballas, Z.K. (1989) J. Immunol. 143, 603-608. 21. Berman, M.A., Sandborg, Cl., Calabia, B.S., Andrews, B.S., and Friou, G.J. (1987) Clin. Immunol. Immunopathol. 42, 133-140. 22. Knudsen, P.J., Dinarello, C.A., and Strom, T.B. (1986) J. Immunol. 137, 3189-3194. 23. Hayari, Y., Kukulansky, T., and Globerson, A. (1985) Eur. J. Immunol. 15,43-47. 24. Kunkel, S.L., Chensue, S.W., and Phan, S.H. (1986) J. Immunol. 136, 186-192. 25. Monick, M., Glazier, J., and Hunninghake, G.W. (1987) Am. Rev. Respir. Dis. 135, 72-77. 26. Passwell, J.H., Dayer, J.-M., and Merler, E. (1979) J. Immunol. 123, 115-120. 27. Sage, H., Vernon, R.B., Funk, S.E., Everitt, E.A., and Angello, J. (1989) J. Cell Biol. 109, 341-356. 28. Wahl, SM., Hunt, D.A., Bansal, G., McCartney Francis, N., Ellingsworth, L., and Allen, J.B. (1988) J. Exp. Med. 168, 1403-1417. 29. Kehrl, J.H., Wakefield, L.M., Roberts, A.B., Jakowlew, S., Alvarez Mon, M., Derynck, R., Sporn, M.B., and Fauci, AS. (1986) J. Exp. Med. 163, 1037-1050. 30. Assoian, R.K., Komoriya, A., Meyers, C.A., and et al., (1983) J. Biol. Chem. 258, 7155-7160. 31. Dubois, C.M., Ruscetti, F.W., Palaszynski, E.W., Falk, L.A., Oppenheim, J.J., and Keller, J.R. (1990) J. Exp. Med. 172, 737-744. 32. Sporn, M.B., Roberts, A.B., Wakefield, L.M., and Assoian, R.K. (1986) Science 233, 532-534. 33. Tsunawaki, S., Sport?, M., Ding, A., and Nathan, C. (1988) Nature 334,260-262.

891