Volume 38, number 1. 137-140 FEBS 11407 8 1992 Federation of Eurapnn Diochcmiclll Saictics 0014579Y91/sS.~
August 1992
Effect of RU 38486 on TNF production and toxicity
George Lkilr Jr.‘, Em6 Dudab and George Lakaif
Glurocorticoid stctoids provide conridcrithlc protection against the systemic toxicity of tumor nccrorir fnctore (TNF-a, cuchcxin). In animal cxpcrimcnts RU 38486 (mifcpristonc). u steroid mttigonist. increased rhc synthesis cf TNF und sensitized the aninuls IO thccytotonic action of TNF. As compared to the control und mcthylprcdnisola~ttWcd groups, mlfcprirtonc signinuntly increased the lcvcl of TNF in the scorn. liver and rplccn of lipopolysacchuridc (LPS).trcatcd ;Inimalr In tissue cultures F.11 38486 induced rhc TNF synthesis of mycloid cells nnd incrcascd the TNP production of gcnctisally modified H&a cclk. which synthclrizc TNF constitutk:y. Normul nnd tumor ccl1 cultures cxhibitcd incrcascd
sensitivity toward TNF in the presensc of mtitipt&onc.
Tumor nccrosir ktors: RU 3X486; Mifrpristonc: Endcrtoxin: Cilucosorticoid
1, INTRODUCTION
Tumor necrosis factor (TNF) shows a very high spcc- if’lti:y in its cytotoxicity towards ccrtnin tumor cells [l].
At the same time, TNF is a lymphokinc with powerful immunostimulating effects and causes severe systemic conscqucnccs ut elcvatcd lcvcls [2]. Furthermore, TNF [3.4] and other cytokincs [5] hnvc been also implicated in the pothogcncsis of septic shock, which is one of the most common causes of death in intensive care units today.
Despite advances in technology and in antibiotic therapy, the mortality rate for septic shock remains in excess of 50% [6,7]. Glucocorticoid hormones arc krlown to protect cxgcrimcntal animals very cffcctivcly against the lethal effect of bacterial cndotoxins [a], but the bcncficial influcncc of these hormones in clinical and cxpcrimcntol fomls of septic shock remains a question of d&ate [9-l I]. Antagonists of steroid hormones pro- vide ;t new framework of action not only because they arc of potential clinical USC, but more so bccausc they pcrmh molecular dissection of hormone-dependent proc&scs.
Tier antiglucocorticoid action of mifcpristonc, or RU 38486, has been reviewed elsewhere [12,13], In earlier studier, WC have shown that this glucocorticoid antago- nist scnsitizcs both Swiss albino OF1 and endotoxin low-responder, C3H!HcJ mite to cndotoxin lethality [Is]. WC have also shown that RU 38486 sensitizes en- dotoxin.tolcrant (cndotoxin-pretreated) and normal
C’~r*rc.rpor~~c~rrtcc U&PCW G. L-r Jr., Department oTSurgcry, Albert SnntGybrgyi Medical Univcnity, 6701-H Pr!csi u. 4. Szcgcd, P.O.D.
4&a, Hungary.Fax: (36) (62) 2G444.
mice to the lethal cffcct of septic and cndotoxin shock [15.16].
Tumor necrosis factor-a is the primary mediator of the pathogcncsis of cndotoxin and scptic shock [17.18].
Glucocorticoids inhibit LDS.induced TNF production [19] and TNF can stimulate pituitary adrcnocorti- cotropin secretion, resulting in the rclcasc of corticostcr- one [20,:!1]. All thcsc findings suggest an important role of the ncurocndocrinc axis in the secretion of steroid hormones and in the modulation of cytokinc produc- tion. On this basis, WC set out to study the influence of RU 38486 on TNF production and toxicity.
2. MATERIALS AND METHODS
Mnlc NMRI mice (LAW Animal HOW. CiBd6llb, Hungary).
30-35 g in weight. wcrc housed with free xccss IO pcllct food und wiltcr ut all times.
2.2, Chwricab
Mifcpristonc. RU 38486. was from Rousscl~Ucl~f. Fruncc Tt wns incorporated into phospholipid liporomcs. Multilrmcllrmr liposomcs wcrc prepared from phosphatidylcholinc (Sigma) and cholaterol (Sigma) according to :hc method drrrrib-d carlicr [22].
Mcthylprcdnirnlonc (Orion. Fintad) arid RU 38486 wcrc injcctcd intravenously just before the lipopolysaccharidc(LPS) trcalmcnt. &
chrrirltiu cob 026B6 lipopolysaccharidc (Difso, lot no. 672736) was susgcudcd in isotonic. nonpyrogcnic saline.
2.3. ClotGlg rend praducriott o/ ltutior necrosis /hctnr
TNF-a (human tumor nccroris factor-e, cachcxin) gcnc was iso- lotcd from a human gcnc library. The MspI-EcuRI fragmcnl, coding for approx. 80% of the fourth cxon of rhc TNF gent was li&Wd wirh a 102 bp synthetic oligonwlcotidc coding for the missing Nwtninal pnrt of the mature prokin. The multingconstruct wascloncd into the k?urtM lritc of tlrc cxprcssion vcclor pDRS40 (Pharmaciu Fin.? C$rm.
icnls).
137
Volnmr JOS. number 2 FEBS LE’I-I’ERS August 1992
B <u/i tcllx h&wing the ubwc plasmid produce tipprox. I my rce4TNF per liter culture iT yruwn in LB medium until the wrly stutionury phm (Mui cl al.. mnnu#ript in prCp.1.
2.4. A~riljrrr~ivrr ~ml trxxf~,b* uf htmr rrzcrur~~ favar
Rccombinnnt TNF ws purifiid by II combinirtion of ttmmonium sulfiltc prccipitatian. cdntrollcd pore ylae and hydrophobic intcnc- tion chronr~tayruphicr and FPLC (Mei ct ;II., mrnuscript in prep.).
If ncwsury. the lrrl t~ccs of LPS wccc removed by affinity chrome- togmphy Gn pblymyxin B.u&uQsc. The tlnnl product yicldcd n ringlc clcctfophorctic b;rnd of approx. 17 kD& w&P ut !c;r%t 95% pure b%
dcrrrmiwd by silver staining of SDS~palyturylamidc #clr;l, hd a WC- cik uctivity of et leerit 10 millian Ulmg nnd contlrin4 Its* thun 0. I?5 U ofcndotoxin per mg protein (mtisurcd by the LAL-Pyro&cnt nssay. Wittakcr 8ioproductr Wmtkcrrvillc. MD).
L929 ccllr wrc srown in DMEM (Scrw, Hcidclberp) with 59 serum. Supcrnutank alzsaycd for ~hc prarnrc 0f YNF wcrc cxlnct~d with chloroform to rcmovc KU 384i186 bclorc ditilyrir and conccntm- tian.
Uiosuny uf TNP WUL bnscd on its cyto\oxirity. mcnrurcd on mow LX9 tumor rcllr. in the presence of I &ml actin0mycin-0. al cxnClly 37% [ 1,231. The TNF srsritivity of normni mow 3T3 rcllr ws nwycd In the prcscncc oTO.IZS and Cl.ZS~a/ml tictittomycin-0. These cells tolcrntcd up IO IO,000 U TNF tit 37*C, in the ubscncc or mifc- prirlonc.
Killing of cells WIT trrscsscd viu the uptukc of Ncu~ral red vlt;rl dye.
The um0unt of TNF rcquircd t0 mcdilrtc the half-muxim;ll cytotox- icily CI~ L9Z9 cells +bw rniyncd ;I value br I U [1.?3].
The tirstic tlivcr. rplccn) wrsminccd nnd wnicatcd. nnd mtrnbrirnrr wcrc pcllelcd by ccntrit’ugiltiun and rcsurpcndcd by brick roniwion.
TPIF WIIP tittred in both the sapcrnatent and mcmbrunc fractions.
BZlfB moue mycloid tumor cells and TNF.producing M9 human cpithclial tumor cells (dcriv;nivcs or HcLa cells) wcrc grown in DMEM with 5% %crum, PZX8 ccllx da nat product TNF unlcu ztcti- vated by the prcscncc ol* cndotoxin or cxogcnourly rddcd TNF. M9 cells harbw a DNA conurusi conttrining the human TNF-0 ycnc under the control of the SVllO cnhwccr. und thercforc express TNF c0nstitut ‘xty. ut high Icvcl.
The comp&lranr of the mcllns uiIcr an;rlyrix or vwhncc wcrc car- ried out hy the mclhad orSchcffc. Stutisti~l xiynificuncc WI& acrcptcd iit P c Wl5. SurVivill dllte wcrc ;rnulyrcd for ~!ntistlcnl riynifcnncc by the Pishcr CXUCI test.
3. REFULTS
RU 384%d greatly increased the toxicity of hTNF both in tissue cultures and in animals. The highly scnsi- tivc mouse L919 tumor cells wcrc killed by a 2-3 times lower TNF concentration, if RU 38486 was nlso prcs- cnt. 3T3 normal mouss fibroblast. which is highly rcsis- tant to TNF, bccamc sensitive in the prcscncc of the drug (Fig. 1). Intravenous injection of 2 ,ug/lO g body weight of hT.NF. concurrently with 1 pg& body weight of LPS killed 4 of 20 animals (80% survival). but when it wus supplcmcntcd with RU 38486 trcatmcnt (1 mg) ull of the 20 animnb died (Table I). lnjcction of mcth- ylprcdnisolonc (2 mg) concurrently with hTNF + LPS + RU rcsultcd in significant protection (70% survival).
It Wi\s earlier shown that bacterial cndotoxin potcnti- atcs the toxicity of TNF 124). so that the vimultnncous prcscncc of othcrwisc innocuous amounts of LPS nnd
138
-A-A-A-A
0;
’ -01024 266 84 96 4 1
TNP CBNCENTRATION WI
Fig. I. The cffcct of RU 3W116 on the TNF-ssnritivity of 3T3 cells.
Mouse 3T3 nermrl fibroblustr were prow on microtircr plutw The cultures wrc trwtcd with liporomcs or liporomccntrappcd mifcpris- tone (40 rumI) before the cwporurc to diffcrcnt conccntntionr UT TNF, 0.125 or O.?S p&/ml uctinomycln-D ws added to purilllcl GUI- turds. (A) O-,75 p@nl actinomycin-D + RU: (0) O.llS ~,#ml nctinomy- tin-D + RU; (4 O.?StiNrnl uctinomycin-D; (0) 0.1 ZS&W nclinomy-
tin-0.
TNF
triggers lethal shock. In our cxpsrimsnts, TFN + RU gave 70% survival. but if combinsd with LPS trci\t- mcnt, the survival dropped to 0%The cffcct of RU 38486 was not lirnitcd to the in- creased TNF sensitivity of cultured cells and animals.
The &err of the drug ott TNF production wns ulso dcmonstratsd in vitro. In M9 cslls, which product hTNF (as B rssult of genetic tmnsformation with the hTNF gcnc). the TNF synthesis mtc was more thun doubled as XI rcvult of RU 38486 treatment (Fig. 2). In mouss P388 myc!;id tumor cells the regulation of the
Fig. 1. The clkct ol*RU 3Wd on thcTNF production of M9innd P388 ccllr. The cells wrc gown in DMEM. TNT: W:IS wlluctcd from lhc medium tind litcrcd ils dncrihcd in Methods. TilL: figure ~IIOWS daily production of co~Qlusnt culturcr. IO@ cell/ml. 0 = M9 cclla, grown wilhoul mikprisIonc: I. ? illld 3 = ML) cells. grown in the prcrcncc of 6. ?O ;rnd 6U yylml mifcpristonc. rcqxclivcly. PZS# CCIIS wccc ~I*OWI
in ~hc iIbsrllcc (4) ;~nd the prcxcncc (5) ol’ 4) jt@lll nriltpristonr.
Volume 30%. number 2 FEBS LETTERS August !B2
S0rlm
TNF (1 d U/m!) 2.6
2 1.6
1 0.5
0
0 1 2 3 4
time ( hours 1
0 control @MP
0 RU-38486 I RU-33483 + MP
liver TNF CU/mgI
14 12 10 3 6 4 2 6
0 1 2 3 4
time C hours 1
0 eontrsl .MP
0 RU-33486 IRU-38486 * MP
spleen
TNF Wmg) 22
24 20 16 12 8 4 Q
0 1 2 3 4
time Z hours 1
Oeantrol IrMP
0 RU-38482 1!W-38488 4 MP
Fig. 3. Thr: effect af RU 3148Gan the cndotoxin-induced TN F prodw tion in the blood (A), liver (13) ;rnd spleen (C). hninr;da wcrc givrn liporonm a!onc, RU M&b (I mg). rnctlrylprcdl~iMlonc (MB) (1 my) or RU 3&4kiB (RU) + MP intwvcnously just bcforc tltc cndotoxin challcn~,c (1 p&g). TNF ~ctiwity WY mcasmd 1, 1. 3 md 4 h ;litcr
1 PS trcittnmt. Mcunli of results on 10 mice per group.
TNF-a gcnc wns unbred. Them cells did not prods TNF without induction by LPS, TNF or phorbol cstcrs.
The prcscncc of RU 38486 ;~!onc induced the synthesis of rclntivcly high amounts ofTNF in thcscculturcs. The cffcct of RU 38486 on the rutc of LPS-in&cd TNF production in the swum, liver and spleen is shown in
Fig, 3A,B,C, At the time of pc& rcsponsc (one hour after LPS injection) RU 38486 had significtmtly in- crwcd the conccntmtion of TNF in the swum. liver und spleen OS compurcd to the control or mcthylprcdni- solonc-trcptcd groups. Mcthylprednisolonc sign&
cuntly decreased LPS-induced TNF production and completely abolished the effect of RU 38486. Neither RU 38486 nor mcrhylprcdnisolonc indused TNF pro- duction (data not shown).
4. DISCUSSION
TNF.a (cuchcctin) is il mucrophugedcrivcd peptids hormone rclcoscd in response to different stimuli. in=
eluding bacteria! LPS. It has !xcn implicated as u prin.
ciptl! mediator in septic and cndotoain shock. Scvcrrrl lines of cvidcncc huvc so f;lr indicatsd that the pituitary-
;rdrcnal axis !~trs un important part in regulating the TNF activity. Adrcnalcctomy sensitizes mice to the !c=
tha! cffcct of TNF. snd that sensitization is abolished
;lftcr addition of dcxamcthnsonc [251. Furthcrrnorc, in udrcnalcctomizcd or hypophyscctomkd mice the LPS- induced serum TNF conccntmtion rennin5 tit a high lcvcl as compnrcd to normal animals (211. In in vitro models. glucocorticoids inhibit the synthesis of TNF and another important septic shock mediator. IL-I. at
t!lc !cvc!s of both transcription and tmnslution
[ !9.26,27]. Morcovcr. in cxpcrimcntal animals dcxam- cth;lsonc protects against Icth;llity induced by TNF [28].
Our results prove that RU 38486 significzmtly increases the cndot.oxin-induced TNF lcvcls in the strum, liver and spleen and the drug wn induct TNF syrlthcsis in mycloid cells. even in the clbscncc of cndotoxin. Thcsc observations explain tlx shock-sensitizing cffcct of this compound, and suggest the invoivcmcnt of endogcnous
‘rilblc I
lnflucncc of RtJ 3WHQ OII the Icthol &at of TNI:-s IWtotrt riurviwl
rnlC (%I
stetisticr
(1) TNP 20120 100
(2) LPS 20/10 IOU
(3) TNF+LPS 16120 80 3 vs. I NS
(4) TNF+Rll I4420 70 4 vs, I P < 0.05
(5) LPS+RU 201’0 100 5 vs. 2 PIS
(6) TNF+LPS+RU O/20 0 6 vs. 3 P < 0.001 (7) TNF+LPS+RU+MP lJ!?O 70 1 vs. 6 P * 0.001 Animals wrc yivcll Li’S (1 FE’s). TNF (Z&t/lo I). RU 3X-W (1 W)
;utd hip (2 nrp) itttr:~vcnousI~. Numhr5 of sur+ors wcrc rccordcd
139
Volun~c 308. numbsr 2 FEDS LETTERS August 195Q
glucosorticoids in the regulation of TNF expression.
RU 38486 acts on the receptor lcvcl, und thus the in- creased TNF production seems EO bc mediated via the glucocorticoid receptors. This is corroborated by the observation that mcthylprcdnisolonc complctciy abal- ishcd the cffcct of RU 38486 on the production of TNF,
Our experiments revealed that RU 38436 enhances the toxic effects of hTNF in both cell cultures and cxpcr- imcntal nnimals. It increases the expression of the TNF gene in cells producing TNF and simultaneously de=
creases the tolcrancc of cells to TNF by interfering with the protcctivc cffsct of cndogcnous glusasorticoids. A recent rcccptor [29] dcmonstrntcd that TNF c~uscs abortion in pregnant mice and destroys embryonic cells in combination with IFN-r. It is conccivablc that. in addition to its action as an antiprogcstcronc, the in=
creased susceptibility of the fcttli tissues to cndogcnous TNF may also play o role in the abortogsnic effect of mifcpristonc.
Our results concerning the effect of the new glucocor- ticoid antagonist on systemic TNF production and tox- icity confirmed the important role cndogsnous gluco- corticoids play in the control of the immunopathologi- col proscsscs caused by high lcvcls of certain lympho- kincs.
REFERENCE§
[I] Carrwcll, E.A., Old, LJ., Koucl, R.L., Green. S,, Fiorc. N. and Williamson, 8. (1975) Pro-c. Natl. Acad, Ssi. USA 72.36663670.
[2] Traccy, KJ,, Vtnssara, H. and Ccnmi, A. (1989) Lancct i, 112%
1125.
[al $gc, A., Halsfcnrcn. A. and Espcvik. T. (1987) Lnsct i, 35%
(41 G&din, E., Gmu. G.E.. Daycr, 3.M.. Roux-Lobard. I? and Lamber, P.H, (1988) N. Eql. J. Med. 319. 397400.
151 Ertcl, W., Morrison, M.H.. Wang. D., 811. Z.F.. Ayi~la. A. and Chuudry, I.H. (1991) Ann. Surg, 214, 141-148,
[61 Parker, M.M., Shclnamcr, f,H. and Natanwn. C. (1988) Crit,
Qrrc Med. 15, 923-929.
[7] Parker. M.M. (1990) Infections in Medicine 5, S-9,
181 Aprwal, M.K. nnd L&&. G. Cl977) Microbior 20, 183-214, [91 Otosson, J.. Brandbcrg, A.. Erikson. B., Hedman, L., Dnwidroa,
I. and SLklcrbcrg, R, (1982) Circ. Shock 9. 571-577,
[IO] Boat, RC., Fisher. C.J., Clemmcr, T.P,. Slormm, G,J., Max, CA., Balk. RA. and The Mcthylprcdniro:onc Scvcrc Sepsis Study Graup (1987) N. En& J. Med. 317,653-658,
[I I] Hinrhaw. L.. Peduzzi. P., Youn& A.. Sprung. C.. Shatncy. C..
Wilson, M. and Haukcnson, C. (IOn) N. Enpi. J. Med. 317.
359-365.
[Ill Philibcrt. D. (1984) in: Adrenal Sfcroid Antagonism (A~nrwal, M.K., cd.) pp. 77.~97, Walter dc Oruytcr. Berlin&w York.
II31 LWr. G. and Aprwal, M.K. (1986) Biochcm, Biophyr. Rcr, Common. 134, 44-SO.
(141 kl;ur. G. and Aprwal, M.K. (1986) Biochcm. Med. Mctrb Eiol. 36. 70-74,
[IS] Ltir Jr.. G., Lbzlr, G. and A~nval. M.K. (1990) Circ. Shock 3 I I ??%I.
1161 ef;tY,“.. lzrlr, G. and A@rwal. M.K. (1992) Circ. Shock II71 Cc’rumi. A. hnd Bcutler. B. (IO&It) Immunol. Today 9, 28-31, [i8] Michic, M.R., Guillou. P.J. nnd Wilmorc. D.W. (1983) Br. J,
Surg. 76. 670-671.
[IS] Dcutlcr. 8., Krochin, N,, Milsark, l.W., Lucdkc, C. and Ccrami, A. (L986) Scicnw 232.977-980.
I201 Shi\rp. D.M,. Murtu, SO,, Pctcmn. P.K.. Newton, R.. Chao, C.
and McAllcn. K. (1919) Endocrinology 124, 3131-3133.
1211 tuckcrnmn, S.H.. ShclLaas. J. and Bullcr. L.D. (1989) Eur. J, Immunol. 19. 301-305.
1221 Vun Rooijcn. N, and Van Nicuwmcgcn, R. (1984) Cell Tier. Rcr.
238, 355-358.
1231 Aggarwal. B.B., Kohr. WJ., Nnu. P.E., Mofln~, B., Spcnscs, S-A.. Hcm~rcll, W.J., Briqman, T& Nedwin. G,L, Gocddcl.
D.A. and Hnrkins, R.N, (1985) S. Biol, Chcm. 260,2345-2354, [24) Rothrtcin. J.L. and Sshrciber, H. (1988) Proc. Null. Acad, Sci,
USA 8s. 607-611.
[ZS] Bcrtini, R.. Bianchi, M. rnd Ghczzi. P. (1988) I. Exp. Med. 167.
1708-1712.
[Xi] Knudsen, P.J.. Dinttrcllo. C.A. and Stram. T.B. (IPL17) 1. tmmu.
nol, 139.41299134.
[27] Wsngc. A. and Bnkkc. 0. (1988) Immunology 63, 299-303.
[2&t] Libcrt. C., Van Dladel. S., Brouckacrt, P. and Ficrs, W. (1991) J. Immunothcr. 10.227-235,
[29] Ficrr, W. 11991 j FEDS ht. 285, 199-212,
140
