CA2377786A1 - Methods and products for modulation of reproductive processes and for diagnosis, prognostication and treatment of related conditions - Google Patents

Abstract

The invention relates to diagnosis, prognostication and/or treatment of reproduction-associated diseases, such as endometriosis and infertility. The invention also relates to the determination and modulation of endometrial receptivity, and predicting the window of implantation of the endometrium.

Description

1. ~i5409-5 METHODS AND PRODUCTS ~'CiR MODULATION OF REPRODUCTIVE PROCESSES
AND FOR DIAGNOSIS, PROGNOSTICATION AND TREATMENT OF RELATED
CONDITIONS
The invention .relates mE:thods and products whir_h may be used to modu:Lat a .reproductive gr,~c.:esses, such as fertility. The i.nventic>n f_u.rther relates to the diagnosis, prognostication and tr.~atmernt ofv assoc gated conditions, such as endometriosis and infertility.
In <~n aspect, the invention describes that the levels of cer.i~ain markers c:orrel.ate wi.tO. conditions such as endometriosis. In an ~mxnodiment, the marker. is interleukin-1 receptor type II ( IL-~~ R:i I ) . In .-r fu:rthez: embodiment, the marker is morzocyte che:rnotactic protein-i (MCP-1) . In a 1_'~ further embodiment, the marker i_s rnac:rophage migratory inhibitory factor (MIFF . l.zr emb:~dimen~:~, tree invention provides diagnostic and prognostic methods involving determining a concentr_zt~:ion or level. of c;ne or more of: these markers in a sample. I~ an embodiment, the sample is a tissue or body fluid c:~>i.~a:irced from a suk>ject. In an embodiment, t:he subject is an animal; in a further embodiment, a mammal, ~.rn a furtrrcer embodiment, a human. In embodiments, t:he tissue or body fluid includes but is not limited to blood, serum,, plasma, per:itor~eal f luid, monoc:ytes, endometrial t__ssue, an~i endometrial ce=L~..s .
In an embodiment, an irncre<~se i.n t.hc=_ level of MCP-1 is used to diagnose endometriosis.
In an embadiment, a decrease -ire the level of IL-1RII is used t:o diagnc.~e er.domet::rios:i;s .

2 (35409-5 In embc>dimerut;, the l~-vEls of the above noted markers may be dE=term:.r~c d by dete,vmining protein level or by determining the level of= n~cl~~i~ acid (e. g. mRNA) encoding the protein markers.
In another ~:s~.ect, the =invention provides methods of treating c:ondwtion:; such as endometriosis, the method comprising the modulai i;_:n, in af~ embcdi_ment, the inhibition, of IL-1 acti~,Tity. In an embodiment, the method involves administering to a sukvjf:ct an IL,-_. antagonist:.
In anot:her .:aspect, the invention provides methods of treating c:onds_tion: such as enc~ometricosis, the method involving admini:~ terir:~:a t.o a suK:>j cec t IL-:1_RI ~ .
The det:ermir~at:;~on of thE:~ markers, particularly IL-1RII, as noted above, may also be used for predicting for 1!~ example the window of im.plantat i.or~. of t::le endometrium, and the receptiv_i_ty <:f eml:r-yon.ic .implantation.
In another ;:3s~>ect, the =,_nvention x>rovides methods and material> for cont:racepti.on, t:cr example based on the prevention of: em~:;.ryon c implant~.~ti_on.
The inventic.n further relates to compositions, comprising for ex;amplc an I:L~-1 antagonist or IL-1RII in admixture with a pharruac;eu~ic:all.y acceptable carrier.
The inventic:~n further re_L<~tes tc»~ses and commercial packages rt l Bating to true above-noted methods and 2'.~ products .
SUMMARY OF THE INVENTION
Acc:ordi.ng t~.~ ari aspect: cJf the present invention, there is pro~.Tided a method of assessing a reproduction-associated d_i_sease in a sub-ject, :;aid method comprising

3 85409-5 (a) determining a test: level of a parameter selected from the group cc~nsisti.ng u1-( i. ) MI F prote :i..o ; ~:nci, (i.i) MIF encc:~linc~ RNA
p in a ti r; sue or b:_sdy f=luid frorr~ said subject; and (b) comparing said te.t level t:o a stacodard selected from the groLZp consi::: i_ag of a corresponding level of said parameter i_n a t:essue or body fl.uia of a control.
subj ect; anc~ a c;~ rn:~ce:~pondi rng ~.evel of said parameter in a tissue or body l:Luid obtained ~=rom sa~~d subject at an earlier time; wherein an increase in said test 1_evel is indicative of rei:~:~-oduction--assocJ_~ated d:LSease.
According to another as~:~ect of the present invetion, there is prc~vided a metr~od of assessing endometrial receptivity i.:z a subjE~;:;t:, said method c:~mpr:ising:
(a) determining, in s~.ic: subject, a test level of a parameter selected from th~~~ group cc>nsist=i_nca of:
(i) MIF protEiri;
(ii) MIF' encc.::~:irig RNA;
(iiil IL.-1RI~~ ;~~r~~tein;
(iv) IL-1RII erj~~~:~dir~g RNA; an~:~
(v) LL-1RT=L ~c;ti.vi_t.y;
(b) comparing sa~_d te::t level. wi.ti'u a standa:rr: selected from the group consist:i og o_E a co.rre:pc~ruding level of raid 2.~ parameter from a :.::~nt:rol_ subject=; and a corresponding level of said parameter obtained 1 rom said subject. at an

4 85409-5 earlier time, wh<~rein a decrease cf said test level is indicative of en.:iometxi.al re:;eptivit~~.
According tc: st.ill another aspect of the prensent invention, there i_s pz:::~~~ idc~d a metrlc>c~ :~?. assessing a ~~ reproduction-~rsso.~i ate; disease ~,_n a suk» ect., said method comprising .
(a)determining a test ~..evel. of a pa:ramet::er selected from the group consisting of .
( i ) I L-1F~II. p:c~c:~teirl;
(ii) IL-iRII encoding R~dA: and ( iii ) IL-1RII activity;
in serum from sa :i c, ubj ect~; :mc~
(b)comparing said test Level to a standard selected from the group consisting c>i:_ a c:orresporzdirig .level of said 1'.i parameter ir1 ser~.nn of- a control subject; and a corresponding le~yel of said parameter in serum obtained from said subject, -its an earl:i_er° time; wherein a decrease in said test l.ev:~:L is indi~~a~.ive of reproduction-associated disea.:-;e.
Ac<:ording tc~ yet another aspect of the present invetion, there i s prcwi_ded use of IL-1RII for the prevention or treatment c~f a repzwoiuction-a.~s,!ociat~~d di.s~ease in a subject.
According tca ~~ fu.rt:her_- aspect of i~he present 2.'p invention, there is pzvevided a commercial package comprising means for assessing tx~c~ level. of a pararneter selected from the group corv,sist:ing c;:f (i) MIF ~.~rotein; and ( ii ) MI F encoc l.ing RNA
in a tissue or body fl,ai::l c;f a subject=; together with instructions for diagn:~s:._s, prognostic~ition, or both, of

5 reproduction-~.ssoci_ate:::1 alisease :un said sLibje;~t.
Acc~~rding tc::~ ;Tf~t. a further a:>pect o= the present invention, there is pr_~:.;~.~-._t~ed a cc~m:nerc~~_a1 pac)cage comprising means for assessing tr~~== t_eve l_ of a pax:anieter ;elected .from the group of:
( i ) _CL-1RII pvotein;
( ii ) IL-1RI I f=>ncoding RNA,: and (iii; IL--1RII ac:~tivrity;
in serum of a subject, t:.ogether with instructions for the diagnosis, or procanost_ic: ~tion, oz bot:h, of reproduction-associated di~~ease in .aid subject.
Acc~~rdir~g tc;t::i.ll a further aspect of the present invention, there is pr::;T,>:idec~ a comrnex:vial package comprising means for assessirng trr~_i.eve~_ ef a p~iz:ameter :-Selected from the group cont,isting c::
(i) MIF ~::rote.r:;
(ii) IL-_LRII )protein;
(iii; MIF~~ enc~:~ding RNA;
( iv) IL-1RII font-oding RNA,; ar_c~
(v) =CL-1RII a;vt:i_vi_ty;

6 85409-5 in an endometrial tis~.ue. of a subject, t=ogether with instructions 1 or the c:ievermiriata~~r. of ~,ne er~dometrial receptivity i_n said sL.bject.
According tc ~an-other as~:~c-~ct oa the present invetion, there i s prc: vi.ded a comrner~;ia.~_ package compri sing IL-1RII and instructic_;ns for the ~>reven~~icn or treatment of a reproduction-associat~:d disease in a s abject .
BRIEF DESCRIPTION OF THE DRAWINGS
Figure 1. Dis1=ributior3 ~~'}f MCP-1 ~ooncentr~ations in peritoneal fluid of normal c:ontrc:~:l_s (n = 4~~) and p~:~t:ieruts with endometriosis (n = 57 ) ~ccordinc~ t:.o st=age of disease.
Horizontal bars, Medi~~.ns of MCP-l conca_nt.rati.ons; horizontal dashed line, shutoff v~~.:Lue ( 100 pg%m-~ ) .
Figure 2. Chemotact:ic response of U93~7::ells to plasma from normal controls and p~:.tients with endometriosis. 0937 cells were stimulated with c:l:ibutryl. cyclic adenos:i_ne monophosphate to induce ditferentiat:ion and used at 6.70 x 103 cells per well. Results were exf7:ressed as mE:ean number of cells that migrate to lower side of membrane per well ~ SEM. Biologic activity of MCP-1.. was e~aaluatE~d by pre.i:ncubat:ing plasma samples with polyclom:~:l rabbit anti-MrP-7. antibody (I:500 dilution) forty 30 minut:.e=s. at 3'7°cv 1>efore add:it:ion of differentiated U937 cel;.s r_o top wells. Preim.mune rabbit serum was ass~~yed in :canoe E-ashion w_itizoa.zt~ regression of 2'~ activity. N-formyl-met:~.:i.ony:1-leucyl-pher~yl.a:lanine (FMLP) . 10-' mol/L and phc>sphate-buf'.ferec~ saline sol;ztion were used <~s positive and negative ~.~c>nt.rol.s, respectively. N, Normal; E, endometriosia. Asterisk, p : 0.i5, versus control; two asterisks, p < 0.01, ~~e.rsus contrc;:l.

7 E5409-5 Figure 3. Re~~:resentat:iv~:~ i;i...l.ust:ratio:n ~t:v MCi:'-1 immunostaining in the endomei_rium of ,nrm~il controls (A, proliferative day 9, immunostaining scow: f~ 1 ; B, sE:,c::ret:ory day 19, immunostaining score i' anci pat.:ier:ts ~r-'~~.::lv~ endometriosis (C, _'~ proliferative day S, :i.rrrrrrmunostaining score 2; D, secretory day 24, immunostaining scc:~r_~c:~ 2 i . Not:.e t:hf~ i~iown positive immunostaininc~ in endc::rnc~t.ria.l glands of women with endometriosis,. particr.::L.~rly in ttlE: secrfr.tory phase, compared with that of normal sl;.kyj ecr;t::s . Magni f :icaiv: ion, ;333 .
Figure 4. Detection of C~CP-1 mRNA :in the endometrium by in situ hybridization. Se:aians were hybricii_zed with biotin-labeled cDNA probes. E;'v_r::t:irn detection mars performed using a rabbit polycl.onal anti.--r:ic~tin antibody, a bio~tinylated goat anti-rabbit polyclonal_ antibody, and fluorescein l~ isothiocyanate-conjugated ,treptavid:i.r~, r~espe~:tively. Slides were treated with propidium iodine, which makes the nucleus visible in ye7_low-orange upon UV excitation, and mounted in the presence of an ant..:i.--nad~ng agent (p-pheny:lenediamine) .
Appearance of endometr i.a 1 glands ( d ) an<i strorna ( s ) at X167 (Al) and X666 (A2) magru::_:f::Lc.:,ation fo:liowi.ng hybridization, and staining with propidil.z.m iodine. t~dote tl:e green-yellow hybridization signal (m:row) that could only be observed at X1665 magnification (Bi 1:>.redorninant:l~r :_.n endornetrial glands.
Figure 5. Representati.~uE. illust.rati_orl of MCP-v' mRNA
expression in the endo-netrium of normal controls (A, proliferative day 13, :~.c::-~.i:~E: i; B, sec:rE,tcry day 22, score 1) and patients with endoi:net::ric>sis (C, proliferat;ive day 12, score 2; D, secretory c~a~n 2 ~~, score 2) . Note the positive green-yellow :pots (ar:,ows) in endom.etrial glands of women with endometriosis, par~t:ic.ularly in the secretory phase, compared with that cf normal sub_ect:~. Magnification, X1665.

Figure 6. Rep:resentat~ ve illustration ot= IL--1RII
immunostainir:~g in tile nu.nan endometriu~m. Sections of endometrial tissue we:ae in,.ubatf~d with mouse monoclonal anti-IL-lRiI antix:,ody (A, l:,r-:~l.iterat_ivEe day 1.::~; B, secretory day '_~ 24; original magnific2~tz.c>n, _x_6g;~ or with an equivalent concentration of normal mon.zse Ic~G~; (C _gild D, respectively;
original magr~:i_ficat=iorr, X63) . Sect. ions were then incubat=ed successively with. biot.i:nylated goat anti-mouse polyclon<~l antibody and <~vidin-b_i of inyl,3t:ec~ r:orser_ac~.isri peroxidase complex. The immunore~:<:tiorl was re~veal~=c_~ witch diaminobenzidine (brown stai_ruinc~ j and _mrrlatoxyl.in was used for counterstaining (l:.L~.ae w7t:ainv,ng) . N~~t:e the brown fine positive staining in v::.romal anc~ epit'ne~~_al c~ell.s (cellular staining) (E-H; c.rigirva.l m<~gniflcati~:~n, X26F~) , and the brown 1_'> deposit (arrow) that ~ s pr:imari 1y :Located at the apical side of glandular (E, secrE_t_c~~ry phase day %4 anti surface (F, secretory pha:>e day 1~ rp:i_t:lzeli.:an,, or r.~ore spread within the glands lumen (G, secrFt:c>ry phase d~~y i6. Positive immunostaininc~ is also, c=i;wtec~ed :ir, i_;5o1<:rt:ed stromal cel7_s (c) (G, secretory phase day .16;~ and microve=~sel=~ (v) (H, secretory pha:~e day 24 j foi_mc~ in the s.~~oma in the secretory phase of the r~ens~ruai c°~cl:Le. s -- st:roma, g = gland.
Figure 7. Dua=_ .immurnofluo:resc~ent~ staining of IL-1RII (A) and IL-1(3 (B) i.n t=he enc~orreei:::ri<~1. t:issuc~ of= ricrmal worsen. Ti:~sue 2~~ sections were succ~essi~,rE_:Ly incubated w:i_t:r~ mouse monoclonal anti-IL-1RII antibody, r_abb_ir poly<~I_ona~' anti-IL-1 antibody, and biotinylat:ed goat _:rr:t~ i-r<~bbi t arWihc,dy ~;efore being incubated sirnultaneou.l_~y-r with rhodam:ine--c:onj ugated goat anti-mouse antibody and fla:~:resc:e:in i~~ot:.h:ioc~~~~anat:e-conjugated streptavidin. Seri-al sec=t.ions incubated with normal mouse and normal_ rabbit IgGs intE,~c~ «:E t.rie hr:imary ant ibodies were included as negative c;~ntrol:~ (C anti D> (ori.ginal magnification, X160) . ~Z«tf~ the co-exp:ression of IL-1RII (red color; and II_~~-1(3 (q:rec~~n--yellow c:olor_ ) within the lumina=L
deposit in er~dometria.: glands. Data from a normal woman in the secretory phase of true menstrua'~ cy,le (day 23).
Figure 8. Re~~:resentattv~~~ ialustration of IL-1RII
_'> immunostainirrg in the t~::~dornetr_ium of women with and without endometriosis. A: Norrr.al sf::cretory day f_4, iuminal staining score 3, cel7.ular slvainirv~ score 2 in 5-_:rcma:l and epithelial cells . B: Endomet riosi > st<_rge secret:ory day 26, luminal staining score 0, cellca.:Car staining score 2 in stromal and 1C1 epithelial cea.ls. Nota the lack of en~:l,~rnetr~.a1 glands in B.
Original magnificatior x:160.
Figure 9. Wesi~ern blot: analysis :~f I::L-1RII expression in the endometrial tissue. A: f~lo.rma.l women: d<~~~ 10 (lanes l and 3) and day 24 (lanes 2 or d 4 j . B: Women wii~h endometriosis:
1 ~~ stage II, day I3 ( l.anE > 1 and 3 ) and s rage I:I, day 25 ( 7_anes 2 and 4 ) . Equal amount ~> of endometr i.a_L L~rote=ins (100 ugi lane) were subjected to sods :.z:n dodecyl sul. fa ~e-pol.yacrylamide gel electrophore~~=_s and tratos.ferred to n:it:rc>cell_ulose membranes.
IL-1RII was detected us:i~:g a mouse rrronoclonal antibody (lanes 20 1 and 2) and the immur .o:omplex was revearl_ed by chemiluminescence. No irn:rmznoreac.!=ion was observed in negative controls where the ant.i--IL-1RI:I anti.bod~,~~ was replaced by an equal concent.ratic_>n of: rncmzse i.mm;znog:Lobolint> ~f the same isotype (lanes 3 and 9.
2 C. Figure 10 . Immunohistc ~hern~i cal. detec:t:ior: of I L-1RI I in t:he endometrium. A) Positi sTe brown immozno5taining in the glands and the strorna (Day 24). B) Negative control: serial secaion from the same endometr~_c:~::L tissue ir~cubat_ed with normal mouse immunoglobuli.ns instead of the primary antibody. C-F) 3C Representative illu:~tr~i:~_Lons of ~I~--1.R:I:I intensity of staining in the endomet:rium thr~~aaghout the menstrual cycle: early proliferative phase, I:,~~e 6 (C) ; latee proliferative phase, Day 13 (D) ; midsecretory )::r.use, Day 1((E) ; ~:~nd late secretary phase, Day 2:3 (F) . Nct~e the markeet immunostainir~g in early proliferativfe (C) and. l;=ate sEecr.etc:>z:y (F) endometrial tissues and the reduc.~°d intensity of that staining :in the glands and '.~ surface epith~~liuTr~ of ni:~dsecretorse phase endcmet.rial tissue (E) . Magnificvation X2'~~~.
Figure 11. Graphical illustration of .im-nunostaining scores and their di~,tribut.ior~ according t;o the clay of the mens trual cycle . A) Luminal. sta ~ n~ ng in g l andular and surface 10 epithelia. B) Cel_lulat staining in glandular and surface epithelia. C) Cellu.laa :~t.ai_ning in the st:rorna. Stromal extracellular staininc; not represented in this figure) could not be evaluaved anc~ u,:~~ detectable onl,~ in Late secretary endometrial tissues. ~'ertical hatched lines represent threshold day; se~parat :irig t_he cy~cl.e into different expression periods according to the intensity of I~~-1RII immunostaining.
Figure 12. A) Western blot analysis of iL-1RII protein expression ire the endc:m~~t.rium throughou°_: the menstrual cycle:
Day 6 ( lanes .L and 5 ) , Clay 13 ( lar__es 2 and 6 ) , Day 19 ( lanes 3 and 7 ) , and Day 2 6 ( l~~ne:> 9 arid 8 ) . The ant ibody specifically recognized di.ff~?rent bands, the mol_ecular_ weights of which range from 68 t:o 31 kO.~. The immunoreactive bands (lanes :L-4) werc_ absent when t;he ~~rimary mouse monoclonal ant=i-IL-1R~1 antibody was replaced by an equal 2_'> concentration of norma:a. mouse Irx~:~s (.lanes 5--8 ) . Although barely detectable in tim= early ~>roliferative phase (lane 1) , IL-1RII bands were marked.l~~ .intense at tree approach of ovulation (lane 2), bLt t:~eix intensity clearly decreased in the midsecret.ory phasF ( lane 3 ) arch .inc ceased again thereafter in late secr_c~tory endon.et.r.ia=~ tissue (lane 4) . B) Nitrocellulose membrar_e--uansferred proteins were incubated with the antibody irn t: h:cE. ,absence ( lanes 7. and 2 ) or the 11_ F35409-5 presence (lan=.~s 3 and 4; of an excess of nI.:L-1RII (20 mg/ml).
Minor bands recognized by the ar:t_~body ~isap~~eared following competitive _i_nhiiitio:i 1:,y :recomi;oinant I__~-IR:11:, whereas the intensity of major hi~:tnE.,r rnolec::ular weight bands was .'~ considerably reduced. '-ndornetrial tissues were at Day 1~4 (lanes 1 and 3) and Day 28 (1_anc-s 2 and 4) of the menstrual cycle.
Figure 13. Localizatic,n of Ih-ll~.I '. mRNA _in the endometr:ium by in situ hybridization. ~eclvic>ns wE>re hyxori_d_ized with biotin-labeled cDNA :robes. ~;etecr_ion c.~f biotin was performed using a rabbit polyclonal ant=ibic~t::n antibody, a 1~i_otinylated goat anti-rabbit ~_»~lyc:lo:na_ <~nt ibody, ~~nd fl.zoresc:ein isothiocyanat.e-conjugacted stz:epta~ridin, respectively. S:Lides were treated with proL;iciurn iodi.nee, whi~:h m<~kes the nuc:Leus visible in yellow-orarcq~= oru u_Ltraviolet excitation, and mounted in th~~ pzvesencve of an ar:tztadin-g agent (p-phenylenediarnine) . Ap~~E=~arance o.f endometri.a_L glands (A) ,, surface epithelium (B) , anti strc:>ma (C) ire shown at X:16'7 (1) and X666 (2) magnif:ic<-ctic>n follows.ng hybridization and 2() staining with propidi~_nn iodine. Note th~~ green-yellow hybridization signal that could bE~ observed only at X1655 magnifi-catio:z (3), p,ec:aom:inantLy in the endometrial gl<~nds (A) and surface epithcylium (B) as compared to the strom<~ (C) .
Figure 14. Aru~lysis of ~L-:LRII rnRT~A expression by RT-PCI~. A) Graphical il1_ustratior~ c~f IL-:LRII mRNA relative levels (° of control ~ SEMI) in the er~dornetri.um throughout the menstrual cycle. B) Rep resentat.:.ve Southern blots of II_-iRII and GAPDH
(internal c,ontro7_) tr<~;~c.ript::s ~n the whole endcmetrial tissue after RT-PCR. , Positive control (;cDNA preparation from human eru~ometria_ tissue expz:essing I:L-1.RII) ; 2, negative control (PCR ir, the absence of cDNA). Tissues were at Days (d) 1_0, 1_3, 1 ', 21, and 24 of tine menstrual cycle.

12 E;5409-5 Note the elevated 7_evc=ls of 1L-:LRII mRNR in endometr.ial tissues at Day 1 . ( lat: a proliferat. ive phase) and Day 24 ( late secretory pha:~e) of. tl;e menstru~=~1 _:vcle and the decreased levels at Day 21 (midt:ecretory phase}. C) Representative '_i Southern blots of IL-i IZI T ,end G~~!?L>H fr_orv separated strornal (S) and glandular epithelial (E} cells at different days (10, 14, 17, and 2.'7) of t;he rnenst~rua~. cyc::L~~.
Figure 15. Represent:at .~~ve Gdestern blot: analysis of MIF
expression in endometz ~~7tic tissuE . 'fo':al ~>roteins were subjected to ;>DS-PACE a::a.:Ly~.is arnd GJesvern blotting using an affinity purii=led ~:~ol~c,:l~~na1 goat anti--P~1TF antibody (lanes 1-3) or an equivalent concentration of normal goat Ig instead of the primary (lanes ~l--of . Lanes 1 and 4, 10 ~g total proteins; lanes 2 and '>,, 2C; ~g total. p~~oteins; lanes 3 and 6, 40 ~g total proteins. Tile detected band has an estimated apparent molec:ula.r wei~Jt~t= c>f app>rcximafi.e:ly 1.2.5 kDa.
Figure 16. Representative after RT-PCR and Southern blot analysis of M7:F t:ransc:r_i_pts. Total RNA c>btai_ned from endometriotic tissue wa~~ reverse trap:>~~nibed, amplified with 2C MIF (upper lanes) or° G;~fI:~H (lower lanes) primers, and hybridized with 32P-labeled corresponding probes. Lane 1, positive control (cL>NA x>reparat:i.cm from t: he Lw.zman hystiocytic cell line U93~, known t.c:~ secrete MIF'; lane 2, negative control ( PCR i.n the ab;sc~nc:e of c.DNA) ; ;!.apes 3 -5, linearity test with different RT volumes.
Figure 17. Immunohistochemical analysis of MI:_, expression in endometriotic tissue (biopsy of :3 red papular endometriotic lesion from a 37-yr-old vaoman with st:.age T endometriosis~) .
Note the intense brawn.i:~ln immunc>staininc~ in the glands and cell aggregates througlrc:».zt lJhe st::rorna __r~ the E:~resence of a goat polyclonal anti-MIF antibody (A} and the absence of such staining in t: he presence of Goat :~-gGs used at concentration equivalent to that c~f t;oe ~~rima.r.~y antib:.~dy (B) (negative control) . Scale bar, _'!~~ Vim.
Figure 18. Dl:~al-immun;fiuorescent stair~.ing of MIF (A-C) and CD3 (D) , CD68 (E) or s-tNF (F) =Ln andometriot:ic: tissue. Tissue sections were inc;ubatEed with goat po:Lyclona_L anti-MIF
antibody and ~aitr: moue monoclonal anti-CD3, mouse monoclonal anti-C:D68, or rabbit ~~olyc.Lor:al ~~nti-vi~~, ani~i.body. Seci~ions were then incubated simultaneously with rhodamine-conjugated sheep antimou;~e ant:ibc:~dy and fluoresceiro isothiocyanate--conjugated donkey ant~goat antibody to detect coexpression of MIF with CD3 or CI768 c:~,- with rhoda.mine-conjugated mouse antirabbit ar~~~ibody arid fluorescein iso:.hiocyanate-conjugated donkey antigoat ant:ibc:~:~y tc> detect coex~r:~ess.:ion of MIF with 1_'i vGVF. Note the=_ expres~:i~.on of MIFF (green:: in CD3-, CD68-,, and vWF-positive T lympho<:~~tes, rnacrophagesi and endothelial cells, respect=ively (red) . .~uper~~osition of fluoresce in (green) and i:hodamine (.red) signals clearly shows coexpression (yel low :_ i_;:~n,.~l ) of MI F with CD~3 (G = A + D) , CD68 (H = B + E) , and vWF (I = C: + F) . kale bars, 20 Vim.
Figure 19. Gxvaphical :i..Lustrat:ion o:f: MT:~~, corrcentrations (ng/mg of total protein;:) ~zs measured by; ELISA in endometriotic; tissue. i~, Endometriot.~~ biopsies were classified according to heir appearance at laparoscopy (red, n = 11; typic<~l, n =- E: ; ;ahitf~, n -- 71 (A) or' to endometz:iosis stage (stage I, n = la:; stage II, n ~ 9; stage III-IV, n = 3) (B) . The box,--and--whihe:r. p:Lot was use~~ t=o illustrate the distribution of MIF ccnL..entrations. 'The box delimits values falling between the 2'~:lw~ arnd t:he 7~t:h percentiles and the horizontal line wittnir: 1~.~re box rr~fer;~ ~c:, the :median scores.
*, Significant. difference between endometriosis stages I and 14 E>5409-5 II using the unpaired t test (P < 0.05), **, Significant difference wish the rc~;:~ l.es:ions (P <. 0. ):l.) .
Figure 20. SemiqLiantit:<~ti.ve RT-I?~~R anal~Y~sis crf MIF mRNA in endometriotic: tis,suc~. The quantity of ~:rze F:'e~R products was '_i determined by der~.sitonuetri~::v ana:l.ysis of the intensity of the hybridization signa_i. 'I'rle r_eiat~ive lev<,i oi~ MIF mRNA
normalized to GAPDH mF:NA was ca_zculated, and the results were expressed as ~~ercent cvt control (~~ositi~fe control) . A, Types of endometriot=ic lesions (red, ru = 9; t~,rpical, n = 6; white, n = 6) ; B, stage of erc~ometriosi.s (c>tage I, n =- 10; stage II, n = 7; stage III-IV, r -- 3) . The box-,~rrd-whisker plot was used to illustrate t,he: c:~.is~r:ibut:icn o.f IvlF mRNA levels. The box delimits values f~::'i:Lirrc~ Letween the 25th and the 75th percentiles and the hc:ri.zontal lane w.irhs_n the box refers to 1~~ the median scores. *, ~>:igni_ficant= di_:faeence with the red lesions (P < 0.05).
Figure 21. I17_ustraticn of I:~-1R:II mRNA expression in situ in the endometrium of nor::n;~l women (A) anc:~ women with endometriosis (B) . Sect::i_ons were hybrid= zed with biotin-labeled cDNA ~>robes . Bioi~:i n detect ion was performed using a rabbit polycl.onal anti~.~:i_o~i.n antibody, a biotinylated goat anti-rabbit polyclonal antibody, and fluorescein isothiocyanate-conjugate-.d strept.<:rvid:i.n, respe~~tively.
Propidium iodine was usec:i to make them nucleus visible in 2~ yellow-orange on ultrav:i.o:LE~t Excitation. Note the appearance of endometrial glands (g;~ and stroma (s) at X666 magnification (Al and 31 i . The Plybr:i.cl:i.yat~ion signals (c~reen-yellow; arrow) were vi.:~:Lble at 1h6~ m<~gnifi.cation (A2 and B2) . Note th.e greater ~~izmber of hybr_i.d~ zati.o.~ signals i.n a part of an endometria:l c~:Lar,~d from a nor_ma.l woman (Day 13; A2) compared to that from <~ woman with stage I endometriosis (Day 14 , B2 ) .

15 Ei5409-5 Figure 22. G.rapriical vllustration of I~-1RII mRNA
hybridization scores in the endometrium of normal controls (n - 26) and of women wi.t z endometric>sis of different stages (n - 53) . A) Hyb:ridizati.c,n scores (mean ~ ,:~EM) in the glands. B) Hybridization scc:~res ~rn~,an ~_ SEM> in th.: sti°omal compartment.
* P < 0.05, ** P < 0.::1.
Figure 23. R'7:'--PCR folly o~aed by Southern blot analysis of IL-1RII transcriF~ts in tr c~ endometrial tissue. Total RNA
samples were extracteca r.rom endometriai biopsy samples of women with (n = 10) ar c1 o:f women w:it:hovt (n =- 8) endometriosis,, then reT.rerse transcribed, amplified with IL-1RII or GAPDI-I primers, and h:ybri.c~i zed wuth 3'P-labeled corresponding probes . .A; Rc_>I>.rese.zt <~tive :;out:hern blot analysis. Lanes 1 and ~.': women with stage I endometriosis, 1_'> Days 13 and 1f3 of the ;nc~:ns~:rual ~yc:l_e, ~~espectively; lanes 3 and 4: normal. women, Gays 7.2 anc~ 16 of ':he menstrual cycle, respectively. The GAP I: Ii was used as a control. B) The ~~box and whisker" plot was used to illustra=a IL-1RII mRNA levels using semiquantitative I<.'r-PCR. The box delimits values falling between the 2~'r- and the 75t'' pezcenti=es, and the horizontal line witrnin 1__he~ box rwfers rc; the median scores.
**Significant difference between tree endometriosis and control groups ( p <: 0Ø1;.
Figure 24. So7_uble IL-~.RII concentrations in the serum of 25~ control subjects (C) aru::i we>men w:~th encic:~metri~:~sis stage's I-II
and III-IV (E7:II-IV) . 'fhF~ "box and wh:i.s~~er" plot was used to illustrate the distr.il"..~l:.iorn of ::IL-l.R:I:I values. 'The box delimits valuE:s falling between the I?5''' and the 75th percentiles and the hct~:i_zontal. lane w:i_i~hi.n t:h~~ box refers to 3C the median. * P <0.U1 as r:ompared t:o control. subjects using the unpaired t: test.

Figure 25 . Ih-la (A) ~:n: Il:~-1 (3 (B) conc~.~nt r<~t ions measured in the serum of ~~ontrols (~~:;) -end women w:i_t~u different endometriosis (EI-II <:~n~:~ EIII-IV) stages. Comparison of control and E:~IIdOITletric:~s:is gr;~up:~ was performed with the '_i unpaired t te:~t .
Figure 26. Se,~um-induce~::~ MCP-1 ;:~ec:reti.oru by L937 cells. :L06 cells/well in 24-well ~~ultur~~ plates weave exposed for 24 hours at 37°C serum from normal. c:ontrol~; and women with endometriosis (5 o vi v ~:IU l ut: i.on ::.n E,I3S-f Y°ee F<PMI:
medium) ,. MCP-1C) 1 secretion (pg/ml) was measured ivy ELISA ire the culture supernatant . A) contro:~_:_~ (~;:) versus end~.ometriosis (E) patients; B) clontrols ;c~') ~:re:rsu.> endomet:rios:is stages I--II
(EI-II) and III-IV (E1I:I-I~,') . Data werE=_ presented as mean ~
SEM and analy.~ed with t:.h~s unpaired t. test, which comparE:s the 1'_. control group to each ~:ndome~riosis group. ~ P <0.05; ** P
<0.01.
Figure 27 . Efj:ect of r 1L--1RI:I ('f ug%m1 ) on s erum-induced MCP-1 secretion. Data were presented as mean ~ SEM and analyzed with the paired t test , ~ahi ch compares ;;era treated with 2f human rIL-1RII to untr~:<~ted sera ir: E,actn group. C - controls;
EI-II = endometriosis stages I and II; E;III-I'~J =
endometriosis. stage;> I:G::f:-:I G'. *' P <().(.).'~; ** P ~. 0.01.
Figure 28. Effect of h.nn<~n r_IL-l.r_a ( 1t)0 ng/ml) on serum-induced MCP-1 secxvet:.io,n., Dat:~ were presented as mean ~ SEM
25 and analyzed with the paired t test, which compares IL-lra-treated to untreated sf-m~.:~ wit=hiri each croup. ~; = controls;
EI-II = endometriosi.s ..:f::~ges I and I1; E:,I iI-I'~J =
endometriosis stages I1:~-_IV.
Figure 29. Effect of r:I:T.~--1R.II on serum-induced MCP-1 30 secretion. C =- controls; E = endometr_iosis; EI-II = stages I
and II; EIII-I:V = stagE:v CI:C--IV.

'L7 85409-5 DETAILED DESCb:IPTION O:f THE INVENTION
Study Design: H~ift~v-set4ren yat.ients with endometriosis at ~apar;~scopy done for ~r:ferti:Lity and pelvic pain were compared wit.Li 44 W :mile womE.r: with no evidence of endometriosis at tubal :..:Ltigatior~ key =l.aparoscopy. Monocyte chemotactic protein-1 concentration in the plasma was determined by enzyme-l:i.:rlkec~,v immunoscrbent assay and its biologic activity was ,w3luated by measuring monocyte chemotaxis with u:~e of ~:~ human h;~st:ioc::yt is c:eLl line (J937) .
1C Results: Mon~=>cyte chemotact:i.c protein-1 concentratior.:~ (median ,:end .range of ~~~ai.t:es) found in the plasma were r:igher in :patients wltt~ endc~metri~~sis ( 163, 0 to 788 pg/ml) thin in norona:L r_ont:rol_s (0, (7 to 355 pg/ml) . This elevation was significant only in the minimal stage of l ~~ endometriosis ( revisecA.rne= icon Fert:i L:it. y Sc>c: iety stage I ) .
However, incrE:ased chemotactic a~ctivi-y mean number of migrating cel.__s/mm' ~ ~~E:;M) was fc:~und in the stages I ( 1240 ~
141) , II (519 ~ 30) , ~n~;~ III-IV (5'.%3 .-_ a?3j c;f the disea=~e compared with normal controls (205 ~ 20. A total of 35o to 2O 44 0 of this act:iv:ity wa:~, inh.ibited in ttue presence of an antibody specific to monocyte chemota~~t:ic protein-1.
Conclusion: Endometriesi.s .is associated with increased level and acti.vit:y of mono~~ytE~ chemotactic protein-1 in the peripheral. blood. The elevation and activation of 2_'~ this cytokine could play a relevar:t role in the immunoinflamn«~tory prc::,:~sss associated with t:he disease.
Key words: Endometrio:= is, chemotac:ti c f<.-ctors, cytokine s Endomet.riosis is a gynecologic disorder characterized by the a:~ct:opic grc~~wt:h of '~is.sue similar to that 30 of the endomevrium, pr i.;t~,ar:i.ly on t:he peritoneum and the organs of the pelvic c:..avvty. A Growing body of evidence indicates that: the i.mm. r~r~ system is a* ~eoted Ln ~.aomen with endometriosis (Dmowski. 'v~J.l~. et a~_, l~~i~~l,' . Locally, an immunoinflammatory pro~:ess involving leukocyte recruitment and activation is taki.:-,::~ p7.ace (~laney E~. f. eat al, 1991.;
Taketani Y. et: al, 199:'.; Leiva M.C. et al, 1.991) .
Endometriotic lesions ,~rc~ Likely to c.;ontr.ibutF~ to the modulation of these in~..m.:mol.ogic react~:ior~s ( I saacson K. B. et al, 1989; Akoum A. et .:~1, 1995a) . :y<el_icv reflax of menstrual debris in the peritone._-3:L cavity may a.l_sc~ occur, thereby exacerbating t;he l_oc:a.l. ~rz:flammatory respc>nse (Halme J. et a.1, 1984a).
HowE:ver, t;hc:, al~e:ratiorzs i.n :immune functions observed in patients wii~h endometriosis are not restricted to 1'~ the peritonea__ cavity. ~:>ystemic .-~ltera~:_v.ons ar_ both humoral and cellular =_evels ha~.Te been reported, including elevated levels of ant_-_bodies :.pe~l:if:.i.c to endometrial. antigens (Badawy S.Z. et al, 1.~~90) and i.oc~reased acti.va~lonal. status of_ peripheral blood monocytes ('teller ,J.M. et al, 1987) .
Monocytes play a c:entr.:~:1.. role in the ma ~..nter~ance of humoral and cell-mediated imm~rii ty, and thrc».~gh a panoply of secretory products they :pan play a s:ign__ficant role in the pathogenesis of endomet~:riosis. A~::cordinc:) to recent reports, peripheral blood monoc.~~~;_,.~;~ of women wv ~lv endo:metriosis 2_'> secrete elevat=ed level s ::f p:rainflamm.~tc~~ry mediators such as interleukin-7. ( IL-1. ) ~rnc~ show thf~ ab.i ~ it y to stimulate endometrial cE~ll growt=h in vitro, whereas monocytes of normal fertile women suppress.- th~~ p.rol~.ferat:z_~:~tu of these cells (teller J.M. et a1, 1':~r3-;%; E~raun D.P. E_et a.1, 1.994) .
A new class of structurally r<,lated small molecular weight cytokines wi_t:h <i:i f fe.rent target: :elect ivity has been characterizes in the Blast f_ew yearn (S~lzall T.J., 1991).

:i9 85409-5 Monocyte chemotactic p:rcvi-e.r7--1 (MC:F~-:l. ~ has been shown tc>
exert a potent: ef fect can monocyt a c:hemoattract ion and activation (L~E:onar_c~ E. ~:~" et: al, .°~.~0; . ~~c~cord:ing to our recent data, L~1CP-1 is present in the peri tor:eal fluid of women and its level is t:i~~r.er in womc.=.n with endometriosi.s than i.n normal. contro:l.~~ (Ak:ourn R. et a:L, 199Fpa) . The objective of t:he current-- study was to examine the presence of MCP-1 in peripheral b:1 ~.~~:~d and to irnvesi=igate S.ahet:her any difference in. its level anti activity could be found between 1C women with and without endometri,:asis.
Material and Methods Sub-iects. WC;nlE'u were rr~ca:uitect into the study after they provided informed consent for a p-otocol approved by Saint-Fran~oi:> d'Assise hospital. ethi.<is committee. Subjects 1~~ with endometri_osis (n -- '.~~7j were i;ientiii.ed after they underwent laparoscopy f=or infertility and for pelvic pain.
These women of:herwise tua~:i r~o c>thPr ~>elvvc di.s~:~rders and were not taking any antiinf i<~rnmatory ::or hormonal medications at least 3 month:> before I_aparoscopy. ThP :stage of endometriosis 2C~ was determined accordirn<~ true revised c:lassif:i.:ation of t:he American Fertility Society. Control subjects (n = 44) were fertile women requesti.ri;~ tubal. litigation or reanastomo:~is and with no visible evi:~ene;e of <~ndc>me ~x i_osi.s at laparo:~copy.
Menstrual cyc__e datinc was determined ac_-.cording to the 25 regularity of the cyci y, tt-~e dat:r~ of: th<-: previous mense:~, and the levels of progesterone in the p.Lasma. The primary clinical pararleters lisved in Table 1 iruclude age, infertility, cycle pha>e, and st..aae of- endometriosis.
Coi_Lection ana pr_oces:~ ir~g of aW ooc~ samples. B7_ood 3C~ samples were drawn a f ew days before i.a~oaro::copy. For MC:P-1 assays blood was coll~:cted in sterile tubes containing ethylenediami.netetracE:t: i c ac:.id anti immec:~iately centrifuged at ;?0 85409-5 Table 1. Plasma 1_E:~vE:l~ ~:_~:f_ MCP-1 ~;io p=ic-~ograms pe:r mill_il.iter) and subject characteri;=~t i c~v> at laparosc--opy A9CT'-I ( riml) lVo. ofpatient,s._~_-9ge (yJy A7ediun and Signifecance*
(mean tSD) range Controls X44 33.75.6 0 (0-355) EndometriosisS7 31.2+7.:? 16, (0-7831 p=0.01 subjects Stage I 27 31.9~t7.1 180 ((7-788;1p=0.04 Stage II a?0 30.1+8.3 158 (0-5850 p=0.601 Stage III-1V10 31.35.5 195 (0-413 p=0.31 o Endometriosis subjects Fertile 31 2.07.1 163 (0-788) Infertile :'6 30.27.4 170(0-640) p=0.79 *Versus control:; witlu Wilc~~x, n t~::~t and F,onterr<>ni correcaion 2000g for 10 minutes a~-: ~_1"C', and the plasma aliquoted and stored at -8Q" C ~,~ntii ,:zssayed. f,or_ h<-~r rnonal assays blood was collected i.n red-toppe~~:l tubes arid sent: to the biochemistry laboratory for steroid clete~rrni.nat=ion.
MCF--1 e.nzym~: --_I.inkec~ inununosa.r~oent assay.
MCP-1_ concentration:> were me .-~s~.~rEed, as previously reported (A)<:oum 1C A. et al, 199E>a) , ,_~~z e:n:~yme-linked immun~:~sorbent wl_tio assay procedure developed iri 1=he laboratory. "his assay uses a mouse monoclonal antih ~_irnan MCP-1 antihoc~y (F;&D Systems, Minneapolis) and a ra>;. !o:i. t ~~o_Lycl.onal. <~ntz.human MCP-1 antibody previously used in our ;:~t~.zd.ie~> (Ak;>ram u. et al, 1995a;
Akoum l~~ A. et al, 1995b; Hac:hi.,ha PM. et al, 1'a9 ~) . This latter antibody does not cro::; ::~ rf~a<_t wi,=r: se,~rfy~-al cytokines that are closely related to MCF -:1, vi_ncluding MCP-2, MCP-3 interleukin-

8 (IL-8), regulated or. activation of normal T expressed and secreted (RAN'~ES) and :m:r~:rophage irlf:lamrnatoz:y protein-1 a and 20 ~i (MIP-la and MIF-1~3)l-iachicha M. et a_., 1993) . The sensitivity limit of t: tnfJ assay was 50 pexirnl with intraassay and interassay coeffic:: i_c->.nts of variation < 6~.

Mor~oc.yte cht-rn~:>t.3~>>is assay. M~~mocyte chemotaxi s was assayed with use of a Bc~ydc:n chamber (Niacleopore, Pleasant own, C:alif. ) ar~ci a human histioc:.ytic c:el:L line (U937) as reported previously Akoum A. et al., '~99Eai. Briefly, four separate pool; of p1 asrn~-~ <:c;x~respondinc~ t o th.e four grc>u~>s of the study (contro:i and endometriosis stages T, II, and III-IV) were prepared. An f_~cy.za~ vol.urne oj- plasma was taken from all the patients incluc:iec:i in eachr group. Trip:Licate samples of each pool Hlere plac:~~:c:i in the bot:torrl wells c:~f she Boyden chamber (200 pl/well_) . folyr_arbonate membranes were then fixed in placE: to separate bottom from t:op wells and 600 x 103 U937 cells in 200 pL. c:~I:' p:~hosphatr~-bufaE.rec~ saline solution containing 1% bovine s~erurn albumin were added to the upper well. After 9C) minutes ut :incubat:ie:~n at: ~.'7°f,, nonmigrati.ng cells were removed by several washed i_n phosphate-buffered saline solution, and t:n<~ membrane was nixed in absolute methanol for 7.0 minute;: at room temperature and stained with Wright-Giemsa (Fisher ..::~c::ierut.ific, Moni:::r_eai) . 'rhe number of cells migrating through each membrane was determined with a computerized image anal.ysi4; system (B:i_o~_)uant. TVTM, Meg X,, R &
M Biometrics, Nashville, Tenn.). Cells were counted three 2C times in three dif~fere::-ri:: areas r<~ndorn:l_y selected at the membrane surface, and the mean number ofi migrating cells per square centimeter ~ SEMI was detEerrr7ined ~vor t.wo independent experiments. N-formyl-methionyl-:Leucyl-1>henylalanine (Si.gma, St. Lc>uis) , a known chnruo~acti.c pept:i.de, was used as a 2'~ positive control at 1C ' moL/L, whereas phosphate-buffered saline solution containing 1'< bovine serum albumin served as negative control.
To appreciate MCP-:1 activ.i.ty :.n the plasma, samples were incubated with d~ f:Le:rent diluti.ons c;f polyclonal rabbit 30 anti-MCP-1 ant:ibody or_ w:ith equal dilu~.ons of normal. rabbit immunoglobulin for 30 m:i:nutes at 37"C before incubation with U937 cells, and the cterriotacti.c act: ivi ty was measured as described above.

?2 85409-5 Estradiol anw~ progest:E~rcr~e assays. 'The concentrations of e:>tr <:~ao:'~ and i:orc.,ge;>t:e tune in the pl_a~sma were measured by a c:ory::»:-'t.~t. ive irYun-~anoa;~say based on anti.body-coated tubes ycomrnerc:i~.r', k~~ts, C.'c~at-1~--e:ountT"', Diagnostic Products, Los AngelE:s ) . 'i'he intr<:ras:~ay coefic Tents of variation mea=cured at ~~~>ca, rnediearr~, arw:d high levE=ls of the standard curves were b~~r.ween 1 . 8 ~~ and 8 . s o foa all the immunoassays. The inten<:~r~say coefic:lent=:~ of variation were 8 . 1 o for estr~idio:l. and 10 ~ for ~>roc~estez-one .
1C Stat:istical a,oalyses. CvCP-.L concentrations found in the plasma dc. not fol.l~~>w a r:orma'~ dist_r_~bution; therefore analysis was c:onduct:ed x:>~;~ nonparametric, metho<ls. Analysi.s of intergroup differences ~~eas performed cc>nserva-tivfaly with Kruskal-Walli_~ one-way ~~rzalysis ~::pf va~-:;.ance by ranks.
1~ Individual groups were i-ompared with thc: Wil.cc~xon rank-~>um test (Mann-Writney-Wil~::~~a.~on test; , and twhe Bo:~ferroni procedure (al:~o call.ec~ L?~rnrz's mu'itip_Le comparison procedure) was applied when more t man two groups were compared.
Receiver-oper~rtor chaa:,_ic-teristic: c~urvE_5 analysis was performed 20 to examine the tradeof is betweeru s~,ns:it~vity ,:end specificity under different cutoff- ~~a:Lues. Sensi.i~.i~~_~.ty was defined as the proportion of positive ta:~st .:result: in patients who had the disease. Spec=_ficity r~x~_esented the p>:,~c>~>ortion of negative test results __n patient=:, who did not. lua~Te the disease. A
2C~ cutoff value refers tc i_he point that :~e~%arates negativE: and positive test. results. ~~, c:utoff value fc~r MCP-1 concentrations giving :.y:~t.irna:l lensit:ivi_t.y arid specificity was then selected, and the. ri,~mbe:~ otwomen with and without endometriosis with MC)=--:' clon~~ent~rat:i_ons f>elc:w or above t:he 30 cutoff value was detez:rn:ined. Comparison of patient age was performed with Student: t test betweE>n two gz:oups and by analysis of variance ~tnf~n ~ever_~r:1 groups wez:e compared.
Statistical analysis c>t morrowyte chemo-tactic activity in the <_'3 85409-5 different pools o:;- per:it::;>nea7_ fl~.~i:_l wa=> performed with.
analysis of v<~riaruce, c;'a_)_owec~ by t: he i'ukey's honestly significant difference t::f~st for. rnu:ltiple comparisons. For all analyses the aifference:=; were corasi.dered as statistically significant f~:r p valu::~:-~ ~: 0.05.
Results MCF-2 concenl.'_rations in the plasma. MCP-1 concentrations in the ra1.~sma varie:~l among pat-~ents, and their distribution .in ncarmai <:c:u~i endometric;t:i.c womeru according to the stage of she ~_~isea::~e i.s illu.~tzat:ed in Fi~:~. I. Because MCP-1 concent:cations w ~r~.~ root: normail.y cistrihuted, we determined and compare :1 i:rieir mecti,~~n:;, as shown ~n Table I.
Receiver-operator char~c;~t:eristic curve ana:lys=_s was performed, and an opti ;a:n:l. cut.of.f v,:~luE: of :100 pg/ml was selected. This cutoff era l..ue yields a sE~r.sitiv.-~ty of 65 ~ and a specificity o.f 61=~. In c>ther words, 37 of the 57 patients with endometr.iosis ( 65 .; had MCP-1 conc:entrat ~.ons >100 pg/ml, whereas 27 of the 44 c:mtrc~l. subnects 1610 herd MCP-1 concentrations <100 pg!rnl.. Static tir_.al ana:lysv~s of the results with the Wilco~<:>ri t.est indicates that MCP-1 concentrations were si.-~nificantly~ tuigher in women with endometriosis compared wi..t.h normal women with no laparoscopic evidence of the diseas:,~ ~:ont:rol(p <: C.05) . A significant difference among the c:>r~t:rol and endometriosi~ stages I, II, and III-IV gr~:ups was -~J.>o (sound ai t:er analys~_s of inter-group differences by tirE:~ Kxut>kal.-W~l.li_t> te;~t !;p < 0.05) . Post hoc comparisons of: ind:iv~i.dual_ gre>ups ~>y the W_~_1_coxon test and the procedure of Bonferrc>ni. reveal a s~.gnific<~nt elevation o.f MCP-1 concentrations crnl.,,~ i.n stage I d~~:ease compared with the control group (p < 0.05). Also, no statistically significant ccrrea_atio.n l:~etween the plasma levels of MCP-1 .?4 85408-5 and infertl.lit:y withi:~ t:he g=coup with endometriosis was observed.
Because endc:nE--li.riotic Les:i.ons are influenced by cyclic change:> in ovar i_:~rz .-;ex st:ero:i.c~, i.t was of interest to determine whether the i.a=vets of MCP-1 fc>und in the plasma varied accord__ng t:o trre phases of t: he meristrual ;:ycle. As shown in Table II, the ;a:if:ference between the levels of MCP-1 in patients w__th endorrrt r.iosis arid con -rot srzbjects was significant c>nly :in t~ L~~t:eal. phase o:E t:he menstrual cycle (p < 0. 01) , whereas ir; I::.ie fol.7_~ c:u La:r phase no significant difference wa:~ nosed. '>).:~f~rw:ise, there was r~o difference between the follicu~_ar and tut:eat plnasE~ s within each control or endometriosis grout:, rlor was there aruy correlation between MCP-1 concentrations ~;ncJ the levels of F-strad.iol (R2 - 0. 007 ) or progesterone ( R'' - ~..~ . i)10 ) found ~.n t:r~e plasma of patients .
Table 2 . Leve_Ls of est r<:~dio:l , progesterone, and MCP-1 in plasma of patients accc:>rding to pl<~;>e o~ m.enstrual cycle Estradiol ( nrollL) .~_ Pro esteronc: ( ~mollL) MCP-7 ( lml) No. of Menn ~-SF'!'-9 No. of Mcan -!-.SEM No. o/
patient nnc:l puticnt and I utient Median, range, and .s sienificancc * s , ~ .si~nilicunco* ~ s si,~Jni/icanco*
Follicular phase Controls 23 239 32 23 2.4l .l 24 I S (0-355) 29 365 51 30 1.50.3 31 165 (0-788) (p=0.11) Endometrios (p==0.04)y=0.42) is Luteal phase Controls I 36'7 18 19.73.2 18 0 (0-220) 38:3 ~ 20 25.84.6 21 16:3 (0-450) (p=0.006) Endometrios (p-0.83) (p--0.?9) is *Versus controls of same c:~,~c.: a yhase 20 Morv~ocyt.e c:hem~~>t.actic activity of MCP-1 in plasma.
The biologic activity :~fi Mi~P-1 way; eval..zated by measuring its ability to inc~ucE= monoc~yte chemc~taxis b~.~ use of the human histiocytic c~f~ll line IJ~37. Plasma from each of the four groups (contrc:>1 and rc ~~:i.sed .Americ:an Fe~t.i.lity Society stages I, II, and III:-IV) we.ra pooled arid the monoc:yte chemotac:tie activity in samples f:r~~> rn each pool w<xs assessed either i.n the presence or absence o.f -:i rabbit po~ycl c>r~a 1 anti-MCP-1 antibody. Thin antibod-~~ specifically recognizes MCP-1, as we have previousi_y rf-~port~ec:~ (~koum aj. e~l~ a_ , 199;~a; Akoum A. et al, 1995b) . St:atisti.ca:i.;~nal~~si.s of t=:die resul_ts by analysis of variance shows a si.~ n~ficant c:~iyfexverrce am~~nq the four groups of the study (x, ~. 0 . 01 ) . ~?cs t noc~ mutt iple pairwise comparisons b~~ the Tukf, n"; r~onest:ly s:igr.ific:ant differences 1C test reveal a higYier rr;~cyte cr,emc;tact=ic activity in ma endometriosis stage; Z 01240 1 41 ce=L_L.s/mm') , II. (519 cells/mm2) , and I:II-IV ( ~~23 23 ca--Jlls/mm') compared with the control group (205 2~ i cel ls/mm' ) (~> <- 0 . 01 ) ( Fi_g.
2 ) .

Preincubation of ~~he ~~~i.a~:~rna witYu anti-1'9C:P-1 antibody (1:500 dilution) resulted in >ic~nifi.c<~nt: :i.ruhibition (percent <~

inhibition, mE:an - SEM; of monocyte chemotaxis found in the stages I ( 41 0 8 0 ) , ( 35'~ 5 ~ ) and I I:L-IV ( 44 0 3 I I 0 ) of the disease (.p < O.C)1), whereas no significant inhibition in the control (--1 0 _t 9'~ was observed. Flabb.it p:reimmune serum ) 2C was assayed in the sane='La shion ~,aithout~ any detectable repression of monocyte ~nercrotact.ic actin%ity (dat<~ not shown) .

Comment In t:he current study we have .v.own that women with endometriosis had hi..ghc-:r circulating _leveis of MCP-1 compared with normal women with a normal clyneco=Logic status at laparoscopy. ~3y use of c~onservat_ve non--parametric statistical analyse:, ~~ s:icxro:~fic:ant E:1evation of MCP-1 concentration was observed only '~n st<~gEe I of the disea~>e, although there also was a trend for an z_ncre~ased level of 3C MCP-1 in the more adva:.uc-e:~ stages ( I:L and II I -IV) . The failure to det=ect a szcatiificant elevat:Lan in Lhese latter stages could k:e due tc: ,_~ Li_rn:ited st:at:i:~ti.cal. ?ower in these ?6 85409-5 series of patients. We :gave also shown t:rzat patients with endometriosis had a significant increase in c_zemotactic activity for monocyt:es :i_rr a.:l:L st:_~ges, tm.lt particularly i.n the stage I of tre disease. Furthermore, 35'; to 44~ of the monocyte chemotaxis oL.:v>erved was irlhuh_it:ed i.rn the presence of anti-MCP-1 ar_t:ibody. TEnese results inc~:ic:ate treat MCP-1 is biologically ~rctive bEc:,~use it:s pr:ima:ry biol.o~:~ic properties known to date are the ctnemoattraction arnd the activation of monocytes (Leonard E.~~. ~~~ a:L, 1990) . ~I'IiE:y also suggest that endometriosis is more active in the ear~.y stages. In this regard, however, avail .:~t.,:L<= data are sT l .J. contradictory. Some studies have ctocumerrte;:~ thmt ~>erit.oneaL fluid inflammation is inversely related to the extent of visible endometriosis (Haney A. F. et: al, 19~i. ) and triat lE:ss extensive disease may 1_'~ be more biochemically active than older implants (Vernon M.W.
et al, 1986) , whereas o~r_her :findings incaicate that peritoneal fluid concentrations of c=~remokines sash as RANTES and IL-8 correlate with the seu-e:rity of the disease (Khorram 0. et al, 1993; Ryan I.I?. et al, 1995).
Several recent studies have focused on the role of peripheral blood monoc:ytes in the pathophysiologic mechanisms of endometriosis. The~~F~ :~:e'yls seem to bce more activated in women with endometrio~:is and secrete elevated levels of IL-1 (Dmowski W. P. et al, 7 9'=~4; Haney F,. E~,. e~_ al, 1991; Taket=ani 2.'i Y. et al, 199?; Leiva M.C. et al, 2991; Isaacson K.B. et al, 1989; Akoum A. et a.1, 1995a; Halme ~T. et al, 1984a; Badawy S.Z, et al, 1990; Zelier J.M. et al, 198'J). 'they also show an altered expre:;~sion of ir,tegrin mol.ecule.:~, which play an active role i.n monocyt:e traf_f:icXing, adhesion, signal.
transduction, and act~.vation (Gebel H.M, et al, 1995).
Furthermore, monocyte:~ ~ rorn patients with endometriosi.s have been shown to sti.mulat.e endometxial cell proliferation in vitro, whereas those c:.f norma:L f:er:tile women suppress the 2i 85409-5 proliferation of endomFWrial cells (Brau;~ D.P. et al, 1994) .
On one hand, c>vr f:indirng of an increased concentration and activity of MC:P-1 in tlm periphe:ra:J. blooe~ of patients with endometriosis :may corrc.:~:,c~ra ~e these: observat ions because MCP-1 is known to exert a l:~ot~ent action on monoc:yte activation and only monoc:ytes have i~>een shown to express a significant number of rect~ptor_~s fo ~: t.hi s chemol<ine (Yo shimura T. et al, 1990 ) . On the other hard, however , t he f indinc~ of a higher chemotactic activity ire ': he peripheral blood of endometriosis women is difficult to f_,x~>lain because, according to previous studies, the percentage:: c~f per_ipr~eral :nonocytes seems to be unchanged in these wom:~~r;. (Gl.eichE=r rd. eat al, 7.984 ) .
Numf=rou s facr:ors may account. for the increased circulating level=> of ;~JC'l:'-1 i.n pat l ent::; wi th emdometriosis .
This cytokine is secrei:ecl by several t:y~;es of cells and its secretion is induced b~~e runny inf:~a~TUriatory cytokines (Leonard E.J. et al, 1990) . Accc::~rcting to our prE:vious data, endometriotic cells se~:.~~-e::te MCP-.i. i.n culture and such a secretion is t~timulate;=1 x:>y TL-1 anc_l t:umc;r necrosis factor-a (Akoum A. et al, :1995a ) . These ~.,:r~o.inf lan~mator=~ cytokine~, have been found in elevated .'bevels in the peritoneal >rluid of patients with endometr.iosis (Taketani 'f, et al, 1992). NICP-1 is also produced by eutop=is endometria~ cells and, interestingly, its secnc~~t:i.c>n wa~~ shown t_o be' ap-.regulate>d in women with endometrio::is (Akoum A. et a:, 1995b) . Activated monocytes have been sl::own t:o produc:e MCT-'-1 (Leonard E. J. et al, 1990) and may also account for its release in the peripheral blood of pG t. l ent s .
Our. results also indicate a significant elevation of MCP-1 in the lutE=a:;_t~hase of tfue ~nen;~t,rual cycle in patients wit;!, endometr:i:_,:;i:; compaz°ed wit-h control subjects. A
trend for an elevation in the follicular phase of patients a?8 85409-5 with endometri.osis ver~s°.~s contro:':.s cou:..c~ also be observed, albeit statist:ica.l.ly u:u.~ignificant. This would suggest a continuous activity of tln-' disease :regardless of the cycle phase. Beside~~, no stat::Lsi=zcally si.gnij.=icant difference in MCP-1 levels between tl~e I=o:lliculaA anc~ luteal phases was detected. Also, no c:or r-r~:l<~t:ior~ ~:~etween t.l:e le~~els of MCP-1 and those of estradi.oi c.~:r progest:erode hound in she pla~;ma of patients was noted. Th~=~7E~ result; seem t.o ru.le out any hormonal modulation of I~~9iP-1 levels =;_n t he peripheral blood because it would have been expecied on the basis of previous studies reporting higlu __~-~vels of h;~rnan serum proinflammatory cytokines such as Ih-:i ~_n the sec:retory phase of the menstrual cycle (Cannon ,:T.C~. et al, 1~~f3'~) and a reduced intraperitoneal ir~flammai:=ory reaction of ter hormonal treatment of patients ia..t:h endometri<?s=..~~ (Lei.~,ra M.C. et al, 1991; Haney A. F, et a1, '988) . It: is st=ill, luowe~;rer, to be determined whE~ther_ cir~:vi_z:~at i_ng MC:P-? .Levels varied on hormonal therapy of pa':. i.c~nt::s witt-z er,c~or-retrics:is with gonadotropin-releasing h<:>rmone ac~on:ist:s or danazol because gonadotropin-releasing lw::>rmone agonise effect: seem to be due to ovarian suF>pre:>si.on ; l~errvay A. , 199?; , whereas danazol seems to also exert a c_i:i_ rec.t inh i_bitor_v action on the immune system (Dmowsk:i W.P. e~ a1, i994j.
In t:he currew;: study we found increased MCP-1 levels and activit=y iru t:.he plasma of worriers wi.:~h endometriosis compared with normal. fE:r_t:il.e women wit=hout laparoscopi.c evidence of endometrio:_:~_s. Similar results were obtained in the peritoneal. fluid c%loaf=Lents with endomet:_riosis. Taken together, these f indiny7 rna ke plaus.ibl a Mc:P-1 as a key effector cell mediator ::..evolved i_n t:lee l>athogenesis of the disease.

The patrx~~genes:is ~:»- endometriosis, a disease widely believed to arise f.ronn an ~L>errant: growth o~ endometria:L
tissue out=side tree utr~:r~.s, is st:i~.~ unc.~ear. We have previously ob:~erved tt .-~t ::vtokirle-st=imulated endometria:L
_'p cells of women with er.c omet~riosis secrete irn vitro increased amounts of mc~nocyte ct~r~raotac:tic protein-1 (MCP-1) . Th:Ls factor may be imp>ortar~t in the recruitment and activation of peritoneal ma~lro~>hage~: ::;:served iru endornetriosis patieni=s.
The present s,~udy repc_ :r1_ '~ t:uat=, in t=he pr_ esence of the disease, such an up-rc~galation of MCP-1 expression arises in vivo and can be encour~>>e>red in sit a in ~:.he vntrauterine endometrium. In women ~~ai t h endomet rips ~:_~, MC:i~-1. expression was elevated :Ln endomf~t:ria'L. g'~arndt., both at the level of the protein (immunohistocr:emist:ry) and t=he mRNA (in situ 1'.~ hybridization i . This s~a~ observed throw<~heut= the menstrual cycle and var:Led according to thc~ stage of the disease. These findings strc:~ngly arga:.e in Lavor c:~:f t.h~~ presence of pathophysiolcc~ical ch~nyes i:n the eutop:i.c: endometrium of patients with endometrri~:osis and make pLausil>:Le MCP-1 as a key effector cell mediator involved iro t:he pathogenesis of t:he disease.
Endometrioss~ is a gynecolc~gi~al disorder characterized by the ~:uesence of endome?~rial-like tissue outside the ui=eras, m~:irly in true peritc:>nea7_ .cavity. It 2'i affects women in their r~~~roduct:ivw age, causing abdominal pain, dysmenc,~rhea, d~; :>~=>areunia, abnormal_ uterine bleed~_ng, and infertilii=y bat ca.~n al s« be asymptomatic and found __'_n women undergoing laparo;-c~~py for tubal_ ~_itic~ation. Its prevalence arr~.ong the c:c~rue:ra_i.. population is cx:ifficult to ascertain, bui= estirnate;~; suggest that tine reproductive health of as many a:~ 10% of 1 tm~ female pc~pulat ion is affected in this disorder ( Strath)a : ~ . H . ~~t a 1, 7_ 98.? ~ .

According t~<; i:vne most predom:iruant hypothesis, endometriosi~, would an i_:,a ~ rom t;~ne i.mplantat:ion .and proliferation of andorr.;i-ria:1 tissue t=lnat ~~an reach the peritoneal ca~lity by t:~..zl~:a:1 ref:Lux (ret:_vc:~aracle menstruation) ': (Sampson J. , 7.927 ) . Trv:_~; pt-,enomenor~ _~~, however, common to all menstruating women, and it. i:~ st i:1 ! uncl.e~:~r =yet how endometrial cells coul_~_i implant ~rnci p-ro~~ iferate ectopi_cal_ly only in certain p<~ti..er:t.~>. Women may have: a genetic predisposition to develop the disease (Frey C.H., 1957;
Malinak L. R. et al., 19~~1~:); . hormonal fa<aors might be involved in the maintenance and c::ievelopment of er~dometriotic lesions, as both eutopi.c and er_:ta.~;~ is endometria_~ tissues depend on ovarian steroids (DizeY~f,>ga G.S. et a_~, i980; Rock J.A. et al, 1992) . In recent years, tmst imrrmno.iocx.:,_c~al dysfunction has been invoked as a causa__ factor i_n trre cAevelopment of endometriosis, and it cnay be a cause of pain and reduced fertility in ~;ome patif_>rats (Drnow >k:i GJ.I?. et a:! , 1994) . C~ne of the most consistently rel:~orted irnmuno~ogic:al abnormalities is that of monocyte activ:t-.i_ora and rvec:ru~t=ment into the peritoneal cavity of pa.tient:> (Gel~_er ~.T.M. et al, 1987; Braun D. P. et al, 1994; Hanevy ~~. F, et a1, 11381; Halme J. et al, 1983). Activated monocy-tes / macrophages are known to secrete many angiogeni.c and ot.lr:e,~ growtri :fact:c~~-s (McLaren J. et al, 1996; Halme J. et al, 1988; Olive C).L. et ,al, 1991), which may promote the growth c:~t: endcmei~ri.a1 expl_ants and numerous proinflammatory mr.~lecu lE:e~ (Braun D. P. et al, -1996; Rana N. et al, 1996) that may exa:_c>rbate the i_nf=~~_arr nator~,~ reaction observed in tr,e peritor~~:~au_ cavity c>f endometr:'~osis patients.
We also believe that is E:mdometrio.~:is armses from the endometrium tads tissu:= would be the ::it.e of l:~iological changes that rrcay t:aci:l:i.t:<=~t;e its own development in ectop~ic locations. In our prev:ico.is st:udic~s we :>r:owed t=hat ectopic endometrial cells (isolated from endometriotic implants) secrete monoc~~te c;hemot:<:~,;t.:i:~ prot,e:~r.--:l MCP-1) in vitro in response to interleukin---1~3 and turm.~rwec:rosin factor-a (Akoum A. et al, 1995a) , cyt.c>k:::_ne~ whose .level:, are F~levated in the peritoneal fluid (Pf) a~:fv women w~_tzn E~ndc~metriosia (Fakih H.
et al, 1987; Eisermann J. et al, 1x88;.. MCP-1 is a chemokine of which the :major bicic,:~ic'a-~ property l~:nown T_o date i.s that of monocyte ac:tiv<~tion crud recruitment s nto the site of inflammation (Schall I'.,1.,, 1991; Leonard E.J. et al, 1990).
Subsequently, we found e.l_evated worrcentratior:s and biological activity of MC:P-l, bot',o _n the PF''' a:nd ~.r~e "erum (Akoum A. et al, 1996a) of pat_i_ents w:i_t~h endomet.rio:~.is. We also observed that followinc stimulat: i-t>n with r>roi_nflamsnatory c:ytokines in vitro, eutopi:~~ en;lomet~o_i-~-~:l_ epithelia7_ cells secreted MCF-l, and such a secretion w:~:~ greater in c:e~-is from women with endometriosis than in :~~E_.._1_s from women raving a normal gynecological status a!:: i..aparoscopy yAl<:cum A. et al, 1995b) .
These results make play-:s i..ble MCP-1 a=~ ar_ impo.ntant cell mediator involved in t:!re acti.vatiori of: ~;eripheral. blood monocytes and peri_tcne _a='a ma<_ rophag~~~s observed in endometriosis patients .. ~'riey also ~_~ive rise tr.~ the key question of whether, ira r_rie presc_::n~:::e of disea:>e, such an up-regulation of MCP-1 ex:>rc:essic>n may occur in w:avo and could be encountered i.~ situ in t:~ie in~ral..t~:-Sri-nE~ endometrium.
Therefore, t vE_~ objective of the pre:>ent= study was to examine thc~ in si to e~r;~~ressiorG ~f MCP-i in the endometrium of women with and with ?i:~t. enclometriosi=~ and tc> investigate whether that f~xpressioa ~::ould vary with the stage of the disease and the phases c:~tthe mer:strual. cycle.
Materials and methods Subjects Women were r~~,c:ru:i.ted into the study between February 1999 and Junc= 199E~ after they providad informed consent for a prot:oc:o7. ~~pproved by t.lue ,'>aint.-Eran~ois d' Assise Hospital Ethi :..~ ~ Ccnruni.ttee on Ht.iman Research. Women included in the study .nac~ ruo endometria~i hyperplasia or neoplasia, anal they ha~::l not:. received t~y° ant.i-in:flammatc>ry or hormonal medication du.r:i_ng a perLod o- at least 3 months before laparo~>copy. Erzc~ometriosis was diagnosed at laparoscopy fc>r infE:.rt.:i..:~vt:y and%or pelv~ c pain or at t:ubal litigation. The stage e~f endometriosis was determined according to the revi.sfecz c: s.ass.ifi.caton of the American Fertility Society (The Ameeric:an Fertv~l~_t.y Society, 1985) .
Patients with endometr:icsis (n = 4'7) had no other pelvic pathology. Contro:L sub;i cu=t.s n = 22 ) weir a fert=ile women requesting tug>al litigaa=ion anal ha~~inc~ no visible evidence of endometriosis at lapar::>:~copy. The a:ycle phase (proliferative or secretory) was dete:rmi_nE:d acct>rcl:irog tc the cycle history, progesterone levels in t:he serum, and histologic crlterla.
Table 3 summarizes the main clinical c'hara~~te.r~istics of the subjects incl;::ded in tf~c: ~~tudy: ;=cge, infertil.:~ty, cycle phase, and stage of endometriosis.
Table 3. Clinical Char~:~cteristics of Fatients at Time of Laparoscopy Number of subjects by cycle _ phase .~lge (Mean Number of E SD) _ Proliferative Secretory subjec.:_ts ~

Controls 22 32.4+6.3 9 13 Endometriosis47 31, 8-+ 1 30 i.5 (total) Stage I 18 31.3+5.1 7 11 Stage II 17 31.6+5.0 ~ 11 Stage III-IV12 32.8 ~ 4 8 6.9 Fertile 26 31.8-+6.6 12 14 Infertile 21 31.8-+3.8 5 16 Collection o.t~ Enc~'omet.:~:-i.~-i1 B.iops:ies Enca~metrial k:>> opsies were obt~.inec:~ during laparoscopy i.isinc7 a sterile Novak's cam.z.la. Specimens were placed at 4°C: in ster:.le Hanks' balanced salt sclution containing 1Cn U~'mi pE:~rv~.ci:ili.n, 100 ~rg/mL streptomycin, and 0.25 Irg/mL ar<rphotericp n, immediately transported to the laboratory, :.>nap-frozc::~r~ in liquid n:Ltrogen with 'tissue-Tek OCTTM compoun;~ (M:i.les, E:i .hart, IN) , ar.d snored at -70°C
until analyzed.
Immunohistoch~=mi stry Serial 4- tc:: 5-arm ~cryosections were first fixed in 4% formaldehyde sol.ut.i or:~ ( hi skier .;c.ientu.f ic, Montreal, Canada) for 20 minutes at room temperatm°e, then permeabilized with Tr:i I:c::~n ?~;-100~'f 1's in phos~:~hat.e-buffered saline (PBS) ~=or 20 mirw,_es at: room 'temperature), and treated with 0. 3 o H20~~ in abso:~ute methanol for 20 minutes at room temperature to elimin~~~:.a endogeneous pe:oxidase.
Immunostaining was performed using a mouse monoclonal anti-MCP-1 antibody ( 10 pg/.rL: .i.n 'aBS clonta.ini.ng 1. ~ bovine serum albumin) (R & D Systerrvs,, Minneapolis, MI~~) arid a Vectastain Elite ABC kit.z~~ (Vector: Laboratories, Burlingame, CA) and diaminobenzid~_ne (Sign':.=~,, :3t:. Louis, Mcj as c:hromogen and hematoxylin fcr counterstaining. The specificity of the immunoreactivity showrn key the anti-MCP'--.: antibody (primary 2 ~ antibody) was examined k;~,~ preabsorpt.:ion with an excess c~f MCP-1 (50 ug/mL) prior t:o incubation with endometrial tissue sections . Sect:ions i.nca. x;<~t:ed wi.t ruooat t hew pri.m;~ry antibody or with mouse immunoglobu l.in of the same :~nununoglobulin class and concentration as flue primary antibody were included as negative controls ir: a l ~. experiments . ~~s far as possible, each experiment includE~c:~ tissues from :~c>rmal ;subjects and patients with different ~~ndometries:i;~ :>t:ages.. Slides were viewed using a Leir_a ruit:~z~oscr>pe (I.~e:ica mikr~:~skopie and systeme GmbH, Model DL~~:PI=~) , and photomicrographs were made with Kodak 1(10 ASA fi-r. MCP-1 imnuunostaining was evaluated in a blinded fashion a;y two independent observers without knowledge of laparoscc~p:i c f indir~g~~ . The :i_ntensity of staining was evaluate<:l 3 tames ire 3 dif.ferent~ areas randomly selected in the section and scc, r~-~d using aru arbi ~rary scale { 0 =
absent, 1 = Eight, 2 =- »moderate, and 3 -- intense). High concordance between true t;ao observers was found as determined by the kappa ( K) measl.:r~~~ of agreement ( ~'leiss ~I . L. , 1981 ) for MCP-1 express:i_on scorE-~ i n t:he st:rc~ma ( K =- 0 . '70 ) and epithelial glands ( K -- i:~ . 8' ) .
DNA Probe Labf~ling anc? .i~~ Situ Hybridization 1~~ cDNA for hurr._~n MCP-1, a °i38-base pair fragment subcloned into the plasrn.id vector_ {pUC:lB), and the pUClB
plasmid were provided ~>~~r the American '.type Culture Collection (Rockville, MD) . ~3ioti~~--Labeled %: DNA p_~obes were prepared by nick translation from t:k~e entire pla:~rn:icvect:~:or with the MCP-2C 1 cDNA (Lawrence ~.I.B. ~-:r: <~1., 1.985) us.inc~ a Bi~:oNickTM Labeling System (Life Technoloc;~.e~:s, Burlington, Canada) . Serial.
cryosections were prepaxed and (fixed i.r: formaldehyde as described earlier. Aft.ex:~ digestion with prote:inase K (2 ug/mL
Tris/ethylenediamine ta.~~;:racetic acid; during L5 minutes at 25 37°C, sectiona were post:.--fixed in 4 o formaldehyde, acetylated by immersion i n 0 . 2_5% :~c.-.et=ic anhydride i n O . J. mo:1/L
triethanolamine, pH 8, f~:~r 10 minute:, and rinsed in PB~.
before progre~;sive dehydration i~r alc:ohc:l baths (50 to 100%) .
Sections were preclybri,,iiaec~ for' 0 minutes at 37°C with the 30 hybridization buffer c7c~voi_d of pro~~e containing 50 0 (v/v) formamide, 10~ (v/v) dr~x~:ran sulphate, G.1~ sodium dodecyl sulphate, 2X ~.SC, 1X Den'nardt:'s sol_utior. (C.Oi?o Ficoll ( Pharmacia, Q xebec, Ca:mada ) , 0 . Ci2aa human t~erum albumin (HSA) , 0. 02 o polyvinylpyroli~_.ione (Sigma) , and 40 mmol/L monosodium phosphate, pH 7) . They were then riybridized with 5 ng/u:~
biotinylated aerobe ancf dissolved in the hybridization buffer for 18 hours at 37°C .in a ;mmidif~.ed chamber. Thereafterr, slides were first: immc::rv;ed in 5ii'o formami.de; 2X SSC solui~ion (2 baths for ;? minut=e:. :each at :~r7°C) , tiler. _i_n 2X SSC, and finally in PBS cO.Cltai:a.ir~g (a . 25 ~ ~--ISA (2 baths for 5 minui_es each at room temperat;., _~c ) . E3i of in was dt->.tect=ed by a series of 45-minute incubations ..~t 3 7"~~ witru a rabbit pol.yclonal anti-biotin antibc:~dy (lo d1 Lut.ion in PBS/0.2v'-t: H:>A) (Fnzo Diagnostics, 7~ong Tsl~~ncl, New York) , a bxotinylated goat.
anti-rabbit polyclona:! ,antibody (1'=o dilution in PBS/0.250 HSA) (Vector) ,, and fll_,cor~~sc:ein ~_soth.ioc.yanate-conjugated 1'i streptavidin (0.5~ in f?~S/0.2_'a~ HA) (Life Technologies), respectively. Slides were then treat=ed ~1~~ th propidium iodine (1.5 ug/mL distilled c~~<~ter) (:>igma), which makes the nucleus visible in ye_Llow-orar;c~e upon LJV exc:itarion, and mounted with Mowiol to whi_<,h p-phen:yle:ne-diamine (~J:ic)ma), an anti-fading agent, was added at a final concentration of 1 mg/mL.
Sections were finally observed under t=he Leica microscope equipped for f=luorescence with a 100-watt. UV lamp, and photomicrographs were oc:~de with Kcdak 4()0 ASA film. As negative controls, :>ec t;:io:m> from each ::~; ssue were incubated without specij=is ._~.DNA probes or wish nonspecific DNA probes prepared by n__ck tram Laticjn from t_he p_~asmi_d vector alone, ie, without the MCP-1 c;l~l~A. The specificity of MCP-1 c: DNA
probes was al:~o examir,e;~ by northern blc>t using total RNA
extracted from endomet :r°:i.a L c;el7_.. 1i c>:r ~t~ese experiments 3' P-labeled DNA probes were prepared by ni~:k translation from the entire plasmicis (with end without MC:~~-L cDNA insert) and by nick translation r~r "c:=L:igo.l_abels_ng" (T7~uickPrimeTM Kit, Pharmacia) of the isolated MCP-1 cDNA fzagment. As far as possible, each experiment included tiss,.zes from normal subjects and ~~ati.ents ~a::th differcmt= encLc:~mel~:riosis stages.
The expressic:n of MCP--L mRIaA wa:~ c:~valuat:ed 1_r~ a blinded fashion by two indeper~d~-nt observer:> as descvribed ear:tier .'> using a similar arbit, ~r y scale. f-!i_gh int~erobserver concordance r_f~garding 1'~1~'P-_expression gating in the stroma and the gland: wa.s foi:~rn:~ ac~c:~~rd~.ng t:o ttm. K measures of agreement (0.'78 and 0. 7 %, respectively) .
Statistical Analyses MCF--1 score:. :follow an o.rdina:l scale. Therefore, statistical analyses ~,~ere perfoz:me~d conserva tive:Ly using non-parametric mel~hods. Ar:alysis of ir~tergroup differences was performed using t:~e Kr ~.z;~ka1-Wa 1 '~is one-way analysis of variance by r<~nks . Cnc.:~ ~ id~_za l groups were compared using the Wilcoxon rank--sure test (Mann-Whi~r~.ey-Wilooxon test), and the presence of tied scorE:~ was taken into ~~ccount in the determination of the um and the var.i._rnce of ranks (Siegel S.
et al, 1988 ) . The l3onferrorn:i. proc~edurc~ (also called Dunn' s multiple comparison poc~edi_zre) was appL:i_ed when more than two groups were compared. '~'lr; a correlat ion between immunohistochemistry ancy iru .situ rvybridization scores in individual tissue spec~niens was evaluated using the Spearman correlation coeff.icier t. Comparison of patient's age was performed using Stuc~er; ~~ ' s t-test. h~et:w~~~sr~. two groups and one-2_'> way analysis of varian,~:~e when several gm~oups were compared.
All analyses were perto:rmed using the statistical analysis system (SAS Institute, ~ar,~, rdC) . Diffe~~ences were considered as statistica:l.ly si.gnifi:;ant. f:or P va~~y~.ze: < 0.05.
Results Positive imrr~ur~:~:rl.is tochernic:al st=aining of MCP-._ was observed both in the :-.t~:roma and epi.t-heli_al g:Lands of the 8.7 85409-5 endometrium, and the i.~rtf:,rsity of staining widely varied among patients. St:atis:i~::al. analysis of the results with the Wilcoxon test did not r~~::~w a sigruificar_t difference in the intensity of MCP-1 immr.«5taining found in thE: stroma between women with anti wit-bout E~r.~::lometriosis, albeit i.n the presence of the disease, triere ,~~~..> a markE d tendency fc:>r an increased expression (P = 0.0518', . Iru add.it.i:>ri, r:c~ stati.stically significant difference ~.w MCP-1 irru~rmnostainin<:~ between stroma and endometrial glands i.ra t: rue ccratrol group was observed (P
0.0927), whereas ,~n th~~ E:ndometrio=~is group, MCP-1 immunostaining was sigr~i..fvicantly greater at the level of endometrial glands (P -- 0.0001) (Table 4) .

Table 4. Number of Norrru ::1. and Endomet:rwc;~sis Subjects According to the ~nten:=.~_ty cf MCi' arot=.e~-n Staining in the Endometrium Stroma c~l ands Intensity ~~:~t ~ intensity of staining _ _sta:ining 0 i . ~ - F va.Lue ) 1 ,_. 3 P ~ralue Controls 19 8 - - ' i.9 1, -Endometriosis 18 ?~ 0 'J.C75:i8y ' v'3 1.6 6 0.0001*
(total) Stage I 8 lCi _. ~) . % 107 ' 1 =3 ? 1 0 . 0048 Stage II 3 14 C~ 1. 01 ~~ 4 " - ~s ~a 5 0 . 0006*
Stage III- 7 ~ - - (J.1 zO'a" 1 F: ' - 0.0321*
Iv Fertile 1Ci 16 -- - - 1.2 7.1 3 Infertile 8 13 -- - 0.9899' :-' 1.. 'i 3 0.2886' Controls ~ - - 3 6 - -Proliferative phase Secretory 9 4 -- - 054c:y 4 F.~ _ - 0..515$
phase Endometriosis 7 1 C) - - 1 11 '' 1 Proliferative phase Secretory 11 i9 -- - 0.'7i:?4' 1 L~ l2 5 0.0760;
p h a s a _ _ _ _ ___ _ ___ _ *Comparison with control<>,rv~P :.°alues coi:~rected by the Boriferroni _'> procedure; r, cony>ari.son with :I=ert-le t.~omeo wir..~ en domet.riosis;
',comparison of the F:rolif<:r~~:.av~~ phase with rne secrer.ory phase.
Based on thi s finding, we furt:.her i nvestigated whether there was any association between MCP-1 expression by endometrial c=Lands any; ~=~ndometriosis. ST~atistical analysis of the results with the 4~i-'~~c:oxon test indi,:ate:> that the level of MCP-1 immunostaining was signif:icant.iy higher in the endometriosis group tt-.arl in the ~~c>ntrol group ( P = 0. 0001 ) .
Furthermore, when endc~rnetr.iosis patients were stratified by severity of disease (,stages I, II, and rTI-I'J), a significant difference among the four groups was observed following analysis of i.:~tergroux~ r~ifferenc:es by the Kruskal-Wail.is test (P = 0.0013). Post ho~_~ comparisons of individual groups using the Wilcoxon test and the procedure of Bonferroni show that the intensit~~' of stairiirig was higher in eaclu endometrios.is stage compared with ccantro:Ls, but the most significant elevation of hICP-l imrro.anostaini.ng was f~m.md in the milder stages ( I anc~ II ) ( P =-- C. 0048 ar,.d 0 . 000:x, respectively) (Table 4).
Star~ist.ical analysis of the data regarding the influence of ~~he rnenst real cycle ~~:~ase ,m the level of MCP-1 immunostainirv.g shows t h ~t within the control group there was no significanl~ differEncve in MCP-1 expression between the proliferative and the .secretcry phases tI? = 0.7515), whereas in the endomei~riosis c:a r~:up, there was a marked trend for an increased exp_~ession ~n the secretory phase (P = 0.0760) (Table 4) . Ors the oth~-r hared, MCP-1 exvression was significantly higher iri endometr:iosis p<it:ients than in 1_'> control subjects both i_::r; the proiifE:rat_i.ve (P <: 0.05) and the secretory ( P ~, O . 00 i ) puases of tl-~.e ~~y~~.l_e~ .
The 47 patiE_wts with enc,ometriosis were also stratified for infert ~::1_ity and MCP-l_ imrnunostaining scores were compared.. Using thc~ '~lil~~oxon test, both fertile and infertile patients with endometriosa_s nad elevated level of immunostaininc~ compare~::~ w it:h con ~rol women ( P < 0 . 01 ) .
However, no st:atistic~:..Ly significant di_fvference between fertile and infertile s~.~bjects having ervdomE:t~riosis was observed (P =- 0.2886) ('rab:Le 4) .
2_'> Representats~,rf~ examplE~s of MCf-1 immunostaining in the endometri.um of worr~eri with and without: endometriosis, according to t:he phase of the menstrual cycle, are shown in Figure 3: A, normal proaiferat:ive, day ~; B, normal secretory, day 19; C:, endometriosis pro:'__iferative, day 8; and D, endomet:rio:>is sec:ret.c~ry, day 24. No~E~ the brown positive immunostaining in women ',~it:h endom_et:rio>is, which is particularly rlarked or i he lcaminal s:i_de of endometrial elands 4 () f35409-5 in the secret=ory phaste c:~f the c~ycLe (i.mmunostaining score =
2) compared with that o~, normal subjects (irnrnunostaining score = 1) . No ircununo::e~ZCtioru was observed iru negative controls in which the anti-MC:P-1 t~ntibody w,_~~: replaced by an ~ equal content=ration o: nucuse imrnunoglobuiins of the same isotype or preabsorbe;:l with an excess of MCP-1 prior to incubation with endomc:~tr-i.al. tissue sections (data not shown) .
Expression o.f MCF~-1 m~~;~i'~, in t_he E'ndometrium The ex~7ress ~.c~,rn of MCP-1 in the enc~ometrium was also studied by i.1 situ hyx:~rldization in order to examine this site of MCP-1 synthesis an~:; t o corrlpare the 1_evels of MCP-1 mRNA in patients with ancz wittn~u.zt: endomet~_ iosis . Figure 4A show's the appearance o:fr endometo:ial. stroma and glands at X167 and X666 magnifications follow .~a hybr-:idizat:ion ,end :; gained with 1.'~ propidium iodine. The '.r~ybridi.zatic..~n signal ;green-yellow) could only bFe viaua.li:red at higher magnification (X1665) and appeared to be mainly located .into the cel_1 cytcplasm, as illustrated :i_n Figure 4fi, showi;ag a part of an endometr.ial gland of a women with er,dometrios;_s with a hybridization score equal t:~ 2.
As ~~escribecl earlier, an arbitrary score was used in order to c~uant...ify t:hc=v hybridiza:~tion :~:i.gnal, and the results were .~na7yzed ccanservatively ;sing ncnparametric analysis of v,~riante Table 5) . High .levcal_s of mRNA werf=
detected in t.:ze c~pithc~lial glands of women with endomet:riosis compared witO women w _.tr.out ev:iderrce of the disease ( P =-0. 0001) , whereas no s~ gnif icant: d~~_fference _in mRNA expression in the stroma:~ between women with and wit~hout~ endometrio;sis was noted (P -- 0.2453';. Furthermore, a significant elevation of MCP-1 expression ire e~ndometr~a~_ glands w<~:~ observed :in endometriosis stages ;f - 0.0054j, II (P = 0.0261), and I II-IV ( P = Ci . 0001 ) compared with the control group.

ThEe effect ;:f the rrlenstrual c~acle on the levels of.
MCP-1 mRNA found in tree endometriurrc was also evaluated.
Patients with endometuiosit road a higher MCP-1 mRNA
expression than contr~::.l subjects in both proliferative and secretory phases of true :menstrual cycle ~; P ~~ 0 . 0001 ) . Within the endometriosis group, a trend t=oward a higher expression was observed in t:he sta:,zretory pha::e (P - 0. ()804) , whereas in the control group, no significant difference between the proliferative~ and the s~~cretory phases aa~; noted (P =
0.7262).
Stavistical analysis of MCP-1 mRNA expression in fertile and ~.nfertile patients having endometriosis did not show any significant cl:ifferer~ce between the two groups as assessed by the Wilcot:o:n Mann--Whitney test ( P = 0 . 2904 ) .

Table 5. Number of Normal and Endometriosis :=subjects According to the Intelv:~:i.t:y c>f M~:~:P--1 mRNA Staining in the Endometrium Stroma __ __ _Clands ___ Intc~ns.ity o' Int ensity of sta_nin<t _ _ st~<ai_rv_nc; _ 0 - 7 . G' value 1 ~ . 3 P value Controls 11 11 -- - LO 4 --Endometriosis 20 - 0.2453* 1 11 17 0 0.0001*

(total) Stage I 6 11 - 0.70!:>6* 1 4 11 2 0.0054*

Stage II 13 __ 1 0.6327* - ~ 1 0.0261*

Stage III- 1 6 -- O.OO;sh* - -- 7 5 0.0001*

IV

Fertile 9 12 -- - ~ 16 _i Infertile 11 7 -- U.2701~' 7. 6 1'L3 0.29041 Controls 4 ~ _. _ _ _ Proliferative phase Secretory '7 5 - 0 . 700 ~' 11 ,._- 0.7267' - -phase Endometriosis 7 - __ 4 1 3 O

Proli.ferative phase Secretory 15 12 - 0.08E34' 1 , L'?5 0.0804$
_ phase _____ ~

*Comparison with _ ;, theBonferro ni-co ntro-L:P t,~alues wo~-rect:ed by 5 procedure; t,cotry~ari-aon r ferile worker.f~rdometii~,>sis;t,comparison wi mth of the proliferativephasevri th l r,e phase.
sE wr~:~torw Rep:resentat~v. examples of MCP-1 mRNA express=ion in the endometrium c:~f wonders w:itr~ and withov.~t endometriosis according to -she menst =ual <:yclE: phase are :shown in Figure 5:
A, normal pro.Liferati~:~r~, day 13; ~>, normal secretory, day 22;
C, endometrio;~is prol:if~::rative, da,y 12; and D, endometriosis secretory, day 25. Note the green-yell-o'a spots (arrows) in the endometri<~1 gland:, ~:>f women with endomet:riosis (scor_e =
1_'i 2) , particularly in t'tvE~ .sec~ret=ory phase, compared with that of normal suk::=j ect s ( sc or a =- l ; . A very l..ow level of hybridization was obsf~rvJed i_n nega.tive c:cantrols including the omission of k;:iotinylat ec DNA probes prepared from the pLJCl8 43 (35409-5 plasmid containing MCi-w-:i c.DNA ir~_sert or the use of biotinylated DNA probE.~~. c.;,btainec~ prom the p=Lasm.id alone. The absence of nonspecifi_.nteraction between plasrnid DNA and RNA was also con(: irmec; i;v Northerru blot using total RNA
:~ extracted fro:rt endomet:rial cells and 'rF-labe red DNA probes prepared from the pUC ': plasm:i.d, t:he i.,U'.:18 piasmid containing MCP-1 cDNA, or the is<;.lated cDNA insert. Finally, MCP-1 cDNA
probes were t:~asted on cL.r.omosomc~ preparat-i.on, and the hybridization spots wF~re localized at band 1?q11.2-q2l.l as expected (Mehrabian M. R.. et a_1_, 1991) (~:~at.a not. shown) .
Discussion In vhe presf nt study, we have sr~own that women with endometriosis had a h_~.gher level of MCP-1 expression in the eutopic endometrium a:=ompared with nor_ma.l women having a 1_'i normal gynecc:_Logi.cal ~ tutus at lapar~os~o~~py. The highest level of MCP-1 expz:ession w~ s observed in en~~c:~metr_ial. glands, although a low level c:( expression could also be observed in the stroma. In a prfaVS~ms >tudy, we found trv~at, following stimulation with proir:f.lammatory cytokines it vitro, epithelial ce=Lls isol~:t.f:~d from t:lne c-:ndometrium of women suffering from endometr:ios:is secrete hicxher levels of MCP-1 than those of normal wc:>men (Akourn A et: al., 7.995b) . The results of the present :=tudy clearly .incaicat:e that such an up-regulation of MCP-~ :synthesis and secretion arises in vivo 2~~ and can be enc:ounterec; i.n s:i ~u i:i the uterine endometrium of endometriosis patierot:. Furthermore, ~lney suggest that a process of ce~_1 acti.vat:::ion would t:~ke i~_~ace at this level.
What: are the i..mpl..icati.on:; of <5ur findings with respect to tr:e pathoprly~~iology of endometriosis?
3C Fir:~t, they ,_~ra consistent. w:i.t=h the basic and most accepted hypothesis advanced by :;ampson in 1927 (Sampson J.A. , 1927) , who defi~vec~l endornet r_ iosi..s as an ectopic growth of tissue that takes ~:m: i.c~in from, the uterine endometrium and reaches the peritonea l ~.:avi.ty by '::ubal reflui: during menstruation. Retrogr,.~de seeding i.s a common phenomenon in most menstruating wom~vr~ (Blumenkr<~tz I~'.,~T. e~ al, 1981; Halme J. et al, 1984b; Liu ::i.T.Y. et al, 1986>, and the presence of viable endometri.al se l1_~ in the pE,ritcneal cav.it:y per se is unlikely to be a causative factor. Genetic predisposition has been invoked to e:xpla ~r: why endomc_~trial ;;ell:; would implant 1U ectopically into some particular patients and not into others (Frey C.H., :L957; Mal:.na.k L.R. eat al, 198U). Hormonal factors (Dizerega G.S. et. al, 1980), and immur:ological dysfunctions observed bot~~n locally ilo. the pex:i_~.::o:neal cavity and systematically in the peripheral blood of patients may also have an impo.rtani~ rolf:~ .n the patloophysiology of the disease (Dmowski W.P. et al, _994; Gieicher N. et. al, 1987; Halme J.
et al, 1987; Vigano P.. <~t: al, 1991_; Gcster_lynck D. J. et al, 1991; Gebel H.M. et a:!., 1995). However, to develop endometriosi,s foci, tide,, "migrat:inc~" endometrial cells must have the intrinsic ab:ili.ty to implant outside: the uterus and to promote their own c:~rc:~wth. Int euestingly, recent data suggest that uterine E.~ndcmetr_ia<_-.ells have an enhanced ability to p:roliferat~~ (Wingfield M. et al, 1995) and to escape immunosurveillanc~e (SOmigliana E. et al, 1996). 'They 2~ also abnorma~~ly expre::,s aromatase,. which is involved in estrogen synthesis (Nc>>bl.e L.S, et al, 1996). The overexpression of MCP-- 7. by endomet~rial tissue together 'with these new observation:: make plausible that endometriosis could be associated wi.tln spec:if:ic; alteration, at. the level of eutopic endometrium.
Se<:ond, our findings provide an interesting contribution in the uruderstanding of numerous previously reported observat:ions oo. monocyte and macro.E~raage activation in endometricsis. Per.pLAeral. blood monocytes from women with endometriosis are morE activated and secrete elevated levels of proinflarrunatcry cy':o~::ines (Dmowski. W. P. :~t a1, 1994;
teller J.M. et al, 19f37'?.
They ~rlso :_;t_i.rnul.ate enc ometrial ::ell proliferation in vitro, whf=~rea:> tho:_e from normal fertile women suppress the proliferation of c~rdometrial_ cells (Braun D.P. et al, 1994). An increased ni:mber of a~t~.vated peritoneal macrophages i_s a~..so oi~;-c-~rvec:l in ttue disease (Haney A. F. et al, 1981; Ha:l_:me ~i . et ,:~ ~ , 1.983) . J:'hese cells secrete numerous growth factox-s and cyt..~kinr~s that may contribute to ectopic growth of endomet rial cc:==lls (Mc:l::aren vT. et <~i, 1996; Ha.lme J.
et al, 1988; dive D. ~!, et: a~., 1 9~~1 ) and perpet:uate the pelvic inflarrvnatory rc:~ction observed in endcmetriosis l.'s patients (Rang N. et ail., 199F~; FaL>ih H. et: al, 1987;
Eisermann J. et al, 1v38~j ) . MCP-1 ~_s a patent mediator o:f monocyte inf=i_ltration into tumors and tissues (Schall T.J., 1991; Leonard E.Ct. et a_, 1990), and only monocytes express a significant wamber of :receptors fcr MCP-1. ('foshimura T. et al, 1990) . Therefore, MCP-1 repre;>ents ~ plausible candidate as an importa:~t factor involved in macrophage and monocyte activation ire endometr .ic:sis. In si:~pport t:o this role, we recently reported the presence of elevated c:oncentrat.ions and biological ac~tivi.ties o~ MCP-1 in the PF and the peripheral blood of patients w:i_tLa endometr:ic;>is (Akoum A. et al, 1'~96a;
Akoum A. et a:cl, 1996bj. MCP-1 could be secreted by endometriotic: implant:. (Akoum A. et a.L, 199~ia) , by activated monocytes and mac:roph~:ges, or by c;ther types of cells such as endothelial <:~.r mesothel:i al c~e_L:Ls ~;Schal i_ T.,J. , 1991; Leonard E. J. et al, 1 990; Aric:v i A. et a:1, 1997 ) . However, i_t could also be postulated that uterine endomet r~i_al c:el 1s, by having the intrinsic pot:entia,l t:o expres:~ high 1. evel.s of MCP-1,, might be invc~Lved in ro~c:;r~ocyte ac.~tivation, arnd when reaching the peritoneal cavity by tubal refiux these cells may help to initiate mono~syte recr uit:ment and activ.~tiorl. Interestingly, according to .recent d«ta (Ota H, et al, 1.996), women with endometriosi::~ present ,gin increased infi Lt:ration of monocytes even in the eutopic erdometrium. ''his i.s in keeping with our observations <~nd suggest that Mc::P-1 ~sou'_ca be involved in enhanced monoc~yte rec_rw itment.
At the prote.:in level, MCP-1 expression was elevated in the initial stages of tue disease, ~ar~ticularly in the stage II, and decreased in more advanced stages (III-IV). The expression of MCP-1. mF:.N;~ was also si_gni*::i.cantly elevated in the initial st=ages ( I arcd .I I ) bu ~ rema iris high in more severe disease compared wii~h ~:a:~ntrol. '.auch a ~_:.screpancy in MCF?-1 protein and rnRNA express ion is difficult to explain with certainty, bui~ it migr t_ be due t.o a recinced translation of MCP-1 and/or t=o a probable degra:~ati.on of the protein in the endometrium i.n the more ,~dvanc:ec~ stages of the disease. On the other hand, our results would suggest that endometri_osis is more active in the ea:r.ly stages. Some-: studies have documented that PF inf:l.ammation is i.nversel.y related to the extent of vis__ble endc~Tretrlosis (Haney E~.F, et al, 1991) and that less ext.msi~ae disease may be more bioc:hemically active than older implants (Ve_-:lOr1 M.W. et a:1, 1986) . According to Lessey et al ;Lessey E . ~~.. e3t al, 1994 ) , the d~afect of 2~~ integrin exprE:ssion in eutopic endometr~um is inversely related to the stage c t: c~ndornetr ios.i.;~ . Lc>wever, it has also been reported that the concentrations oi. chemokines, such as interleukin-8 and RAN'I.E;;~~ ( x egulat,e~ or: activa ion normal- T
expressed and secreted), correlate with the severity of the 3C disease (Ryan I. P. et ~-~:L, '1995; ihorrarn 0. et al, 1993) .
MCP--1 expres.=_,:i_on was ruigher iri endometriosis patients than in contra.:L subjects both _n the proliferative and the secret=ory pha~~~:~ of the menst:rualcycle phase. Within the endometri.asis grovC>,, however:, there was a high trend for an increased expressic::r1 :in the secrE:to:r~,~ phase, either_ at the level of the protein c:u the mR)'dA. 'I'hf~se results reveal a process of ce_~l activ~:t: ien that: ~ccJu:r:t-hrougnout the menstrual cyc: _e in t:he andornetriarr. of pa:~t:ients but is amplified in t:he sec~rr:~:.c:.r;y phase. Moreo~.Ter, an intense MCP-1 immunostainin<~ was frc_queni::ly lo~:~at~ed in the lumen of endometrial guands in r:.!.e ~~ecrel:=~ry phase, indicating an 1C) increased release of N:::=-1 <~v thus period of the cycle. It remains uncle<~r how MC: f---1 F:xpre:>sion ,~s regulated in the endometrium and what: <:rf~ the me c.harui_sms that: gavern the increased synthesis ar c~ secret:ion af: MCi?--1 in the eutopic endometrium a._- women v, L vh endometz:iosis. Some proinflammatory 1_'~ cytokines can up-regu~ t:~ ~ = MCP-1 expres:~=~.an by endometrial cells (Akoum E~. et al, 1 99:~a) , but it r<.mains to be determined whether such-i an express i.an ~~could be modulated by ovarian sterc;ids .
In summary, w~~~ found inc.reaseci expression of MCP-1 2C) in the eutopi-c endornetrium of women witLn endametriosis compared with normal ierti7_e women wit.~~out laparoscopic evidence of endom.etric::is.. Such ari increased expression was dependent on i=he stage: c>f endometriosis and occurred throughout the menstrL G:~:l cycle. 'these f_~.r~dings strongly 2_'~ suggest that endometr~c:oJi.s .is not only a local disease restricted tr.: the peri !~~:~nea1 cavity but c.culd also be associated wil~h patho~;tmysiological c:harle:~es at the level of eutopic endorr~:etrium. >~ mc~ametria:L cells of women who can develop endometri.asis might: be functionally different from 30 those of norm<~l women. ~_?nce preservt ect~:~pically, these cells would have tl'-.e intrin~: i_:_~ ability t.o implant, proliferate, and display a different rc=spanse to st imu_Li ~oresent: in their new environment. Our data make also plausib:Le MCP-1 as a key effector cell. med:i_ateot in~~olved '~r. the. ~>athc;genesis of t:he disease.

Many of the :::,.i.o_ogical change. occurring in the endometrium during the n:u~rmtr_u.ai c;yc~.~e ~:~ear a striking resemblance to those a:;_;ociated with inflamrr~atory and reparative pre>cesses. (e~nce, it wol.zld neat be surprising to find that cytc>kinE:as krv::~~fJrn ior_ t.racir pr_o--inflarrunatory properties, such as inlverleukin-1 ( II_~-=_ ) , could play a k:ey role in the physi;~logy cW= thv~s t:i_ssue ar!d than, their action would be tightly car~tr::~1_ Lecl by local mechanisms. In the present study, immur~ohi;1=ochemical and GJestern b:Lot analyses show that in normal worrlen n = 39) , the endomf?trial ti.s~~ue expresses, in a cycle-c:lc~x:~eradent manner, the I:'~-1 receptor type II (IL-1F;II), a mc:~:_c=rule of which tr:e only biological property known to date . s t hat of ..:apt=ur ing I I~-l, inhi.bi.ting thereby its binding tc_~ t_Ine func:t-.:.oval t::ype I IL-1 receptor.
1~ IL-RII immuno:~taining ~~a.~~s partic:ular_Ly ~i-ntense within the lumen of the elands arw:I <~t °'he apical side of surface epithelium. Interestingly, the intensity of staining ways markedly less pronounc:~>.::~ iru the endomei=rium o women with endometriosis (n = '_i4), ~ disease be=L:ieved t.o arise from the abnormal deve~7_opment. of endc~metria:L i.~:i_ss,ue outside the uterus, especiall=y in .ue early stage:_> of- the disease (outages I and II) . Tr:is study ,..., the f:irst to show th~~ local expression in endometr i_::~:L t:.istsuE: cf :I~I~--_'1RII, a potent and specific down--regulate: r- ::~L I~-1 action and i.ts decreased 2~~ expression in women suf:~eri_ng from endometriosis.
Clterine endc;mer_rium, one of she most dynamic:
tissues of the human x::.>dxye, is are aclt:i_ve s ite of cytokine production anc~ action. D;~:ring each menai..rua1_ cycle and throughout the reprodl.:c:?v:LVe phase of wc~rnen' s life, the endometrial t_:ssue i.znce:rgoes a caries ot= dynamic physiological. processes ~f regenerat:ion, remodeling, and differentiation, follc~wf_~d by necrosis and menstrual shedding at the end of: the cycle should _..n~~Lant~~t::l.on not occur. It is well establis)ved t=hat t:.~aese ;:omp:Le~ everit:s are orchestrated by the coincz_c~ent vari~~tions of e~trogeu and progesterone levels in thE: periphe:c~~:'. circulation. H~:owever, many of t=he biological changes occ ~_rr: ring in the human endometrium during _'> the menstruaL_ cycle bc=ar a str_ikir,g resemblance to those associated wii~h inflan~rn,~tor_y and repar~~ ive processes. Hence, it is not sur~~rising t~> find that pro-irnf.lammatory cyto)cines can be involved at aul o~,~rirle, para.crine, anti endocrine levels in the modulation of a, ~~turzety of endomfet:rial functions (Tabibzadeh :: . , 1991; ~inu~n ~~. et al, 1998b) .
Interleukin---1 ( II_~-1 ) :is one o;'~ the major pro-inflammatory clytokine.found t=o act on <:~r~d to be produced by endometrial t::i_ssue (S:imo::n ~::. et_ al-, 199~~>>; Simon C. et <~l, 1993; Simon C:. et a:L, L998c; Tabi~~zadeh :.'. et al, 1990) .
1'_> Circulating levels of CC;-l were st.own tw be variable during the menstrual cycle ar,~.~ to reach rr~<~xima=_ levels during t=he secretory pha:~e (afteY ~~>vulation) (Cawnon J.G. et al, 1985) .
The cytokine :is ~-r_oduc:.c~ci by tr_ophoblast'~c: cells, and is believed to ~~c~t as an wmbr~ronic signal arid t:.o play an important ro7_e during t.:ne implantation process (Simon. C. et al, 1998b; Psychoyos Vie., 1993; Sheth K.'°l. et al, 1991) . IL-1 is produced 1 ocally ir, .::ndomet~rial. tissue as well, mainly in the late secretory phc~s~~ (Simon C. et .~i, 1993; Kauma S.. et al, 1990) , suggesting tf-cat beside it:s ~otent:ial role in 2_'> implantation <~nd embr~,-o:ni.c develop>ment, t: hi s cytokine may be involved in t:he inf:Lanunatory-hike process that takes place in the endometrium at th~> ~~nd of each mensrrual cycle.
Based on thc= above evidence, ~t: is reasonable to believe that endometr:i<~l t~-ssue pc>ssessa:~ the appropriate regulatory me~~hanisms that :pan operate l_ocal_l.y and mains=ain tight contr_ol~ on the ioc:al level c>f pro-inflammatory cytokines. Trv:i_s is crit_i.ca1 for maintaining the inflammatory-like process within sG physioi.oaic::~l~i.mit:s. Any defect f~ ' in such mechanis ms rr~ay t~:o endc~metrisldysfunr_tion and lc-a_a consequently 1:o endomet:.ri.um--relatc~a >rder_s affecting the d-is~_ reproductive :=unctic:>ne, i.nfertilit~,~,endc>met.riosis, ( T

_'i dysfunctional_bleedinc, ~nc:t neoplasia) .

Lit:i:le is kr-a:~wn about th.e mechanisms that modulate the expression and thf acti.ou of pro-i~~lammatory cytok -nes such as IL-1, in the f rm:x:~metriur,~. Cell ,~rtiVati.on by IL--1 results from its binding to ~~eli surfa~~e IL--1 receptor t:ype-1 1Cl (IL-1RI) that: in concc. rt w_t~~ Ih-7 receptor accessory protein (IL-lRacP) is capable ;.~ transducing thf, activation signal (Dinarello C.A., 1996; i:3oraschi C~. et a._, 1996) . Type I=~_ IL-1 receptor (IL-:LRII; ha::, in contrast to t:he type I recept:or, no signaling properties, W.zt has rec:entl..y been described as a 1'> ~~decoy receptor". The eutracellula.r domain of the recept:or can be shed f:=om the cell surface a:~ a soluble molecule that is capable of captur_ir:g IL--l, thus. preventing i.ts interaction with the fund=Tonal rFC~eptor. Thev.e stvc~i.es suggest that: IL-1RII play an importani ~ary:~iological ro.e in the regulation 20 of IL-1 action in the i_mf lammatior. sites (Colotta, F. et: al, 1993; Colotta F. et al, 1994; Boss.u P. fit al, 1995; Orlando S. et al, 199'7; Coultc_r ~.R. et al., 199'x) .
In the presernt study, we invest:.igated the expression of IL-1 RI:I in t:he endc~metria of healthy women, 25 and women wit:h endometr:iosis, a very frequent endometrium-dependent gyr:ecologic~,:1_ disorder. The disease i.s characterized by an alr;.rm~rmal development of endometrial tissue outside the utE-=ru..~s, mainly in tht_~ peritoneal cav_Lty, and associated with ar, nLmuno-ir~fl_ammat,~ry process that has 30 been describE:~d in the b~~:trl ectopic and eutopic endometrial sites (Witz C.A. et a.; , 199 ;'; McMaster 1'~I. 'I'. et al, 1998; Oral '~2 85409-5 E. et al, 199E>; Ota H. <..'_ al., 19'6; Tsf~rlg J. F. et al, 1996;
Jolicoeur C. et al., l9Cas3i .
Our study resrea:Led that IL-1RII is indeed expressed in endometria~_ tissue t~r.d .i.n a evade-dependent manner. The expression wa:> omnipre::~en~ in both epitrmlial and stromal compartments, and was more conspicuous ._n the secretory phase of the menstrual c:yc:lE-, 'f'lhe most. irnt.erl:;Ee imrnunostaining was, however, locat=ed in the: luminal side of endometrial glands and surface epithelium., Int.erestingl.y, we fc>und out that. such expression wa~~ strikirvg:Ly defic_ent. in women with endometriosis, parti.cL7..ar.ly :in t:rne sec:re;~t.ory phase of the menstrual cyc=_e .
The study provides f:or the f first tame evidence for the local expression in human endometrial tissue of the IL-1 1~~ decoy receptor, one of ~ha most specif ic::: down-regulator: of IL-1 action. Furthermore, it reveals a defect in that expression in the intrauterine end.omet.r_z.um of women suffering from endometriosis, tr:.at is, in the tis:;ue where the disease is believed t:o take or_.:igin.
Materials and Methods Study Participants Women were Y e~~:ruited into the study after they provided informed con:.-:erut. +'or a prot=oca i approved by the Saint-Fran~ois d'Assi~:e Hospital Ethics Committee on Human 2_'~ Research. Women i.nc:Luc,led in the st.uc~y ('I able 6) had no signs of endometrial hyperpiasia or neoplasia arid were not receiving any anti-:inflammatory or hormonal medication <~t least 3 months before laparoscopy. Endometriosis was diagnosed during inve;t::i.gative l.a~>arosc~~~:~y for infertility and/or pelvic pain, oi. at tubal litigation. The stage of .'~3 85409-5 endometriosis was dete _rnined accc>:rr,:)ing t c the revised classification of the %~rr:erican Fertility Society (American Fertility Society, 1.98!:-; . Pat:ient.s wit-r1 endometriosis (ri =
54) otherwise had no ov:=ner pelvic pathology. Normal women (n - 39) were fertile, rec,p.zest.ing tczbal 1_~t igati<m, and having no visible evi.denc.;e of t:~ridornE~trie~s:is at= laparoscopy.
Menstrual cycle dating ~~~as determined by menstrual history and confirmed by histr.:a.~c>gic~al exam:i_r~atic:n using t;he criteria of Noyes and c:olleague,s (idoyes R.W. eat al, 1995) .
1C Table 6. Clinical Criar,~z::t~<~r_istic~ of Pat ient.s at Time of Laparoscopy I~lumbe.- vbjec~ts of. ~~ by ~.:~~ic:le phasr~

_ F.~~E
umber ~:vt roli'er:ati "Mean Secretor y subj ec:t:- ve ' :
. 1 ',) C on t r o .__ _ 1 --l s - ----.. _ ~
'~
.
~

3~a ~ 1. t:

Endometriosis :17.
, C~

54 ?.' ~2 (total) Stage I , _?c7.1 , , l.() 7.3 ~

E..B

Stage II -~~~'I
S

1y, ~ 10 .
'l Stage III-IV ~....7i 7.'.~ Fi 9 ' _ .
(1 Fertile -::i.6 2 y H 16 .
~~

Infertile ='1.9+

3~-~ ~.,~
i . 6 Collection of Endometx.ial Ricpsies Endometrial biopsies were obtained during 1'> laparoscopy with the ~.:se o.f a Pipe~lle . ~Inimar Inc. , Prodimed, Neuilly-En-Tc:helle, Fxazr~:,~e) . ;>pecimens were placed at 4"C in sterile Hanks'' balancE~~~ salt. solution containing 100 U/ml penicillin, 1()0 ~:g/ml ~~t.reptomycir,, and (a.25 ~,ag/ml amphotericin, immedi.at e1 y transported to the laboratory, 20 snap-frozen in liquid nitrogen or embedded in Tissue-Tek OCT

compound (Miles Inc., ~,lkha:rt, ~1~1), and stored at -70°C until analyzed.
Immunohisto~~:'n~amist.ry Ser:Lal 4-~.zm c..ryosec:tions, were placed on poly-h-lysine-coated glass m:icrosc~o~e slides and fixed for 20 minutes in fc::=maldernycie [4'v in phc~sphatE~-buffered saline (PBS) ] (Fisher Scient i fic, M..;ntr~a.l, Quabec, Canada) . All incubations were perf< rn~ed at room t:emperature in a humidified cramber. Sf_ at ions were r~wn:_~e~:I in PBS, immersed in PBS-1o Triton X-100~'~M !:ox 2.0 minutes at room temperature, rinsed again :Ln PBS, ~.vc:i treated for_ ~0 minutes with hydrogen peroxide (H20~) (0.3° n absolute methanol) tc~ eliminate endogenous pe~:roxidase . After a PB~! r_ rose, immunostaining was performed using a moue monoclonal anti-human IL-1RII
antibody (R & D Systenus, Minneapolis, MN) (primary antibody), a Vectastain Mite AB(: i;it'r' (Vec:tc>r haboratozies, Burlingame, CA) and diaminobenz:idan~_= (Sigma Chemic:al_ Co., St. Louis, MO) as chromogen. Brief:Ly, after inc::ubat=ion with blocking serum for 30 minute;, tectic~r~~. wc~r_e rinsed _n PBS, incubated for 90 minutes with an appr_oi:~~i.ate and prPdeterrnined dilution of primary antibody (15 i.~;~/ml of PBS containing 1~ bovine serum albumin) , rin;~ed in PFsS, and incubat=ed for 60 minutes with the secondary antibody? cvonsisting of bi~ot:i.nylat;ed goat <~nti-mouse polyclc>nal antil:~ody. Sections were then rinsed in PBS
and the avid_i.z-bi.otin~,,~lated horseradish peroxidase comp:Lex was applied for 4 5 Tnirnut_es . After a PBS rinse followed by a 10-minute inc:ubat.ion v:~.ith diaminobenzidine: H~02 (0.5 mg/0.030 H20~ in PBS) sect.i_ons :were: washed '~n taP water , counterstained with hematoxylin, and mounted witlu Mowi:~:1 (cJalbiochem-Novabiochem (~~rp. , La Je:.I_la, CAj . Sections :incubated wi shout the primary a:c:ztibody c:r_ with nonirunune mouse serum were included as negative .~or~tro:ls ire al_L experiments. Slide, were viewed using a Leica rr~i~~roscope (I~ei~~a rnikroskopie and systeme GmbH, Model DM;Z13; Postfa~~r~, W~=~t:~l.ar, Germany) and photomicrographs were t-~~k~en with Kodak 1. C;0 ASA film (Kodak, Toronto, Ontario, Cana~c~a ) . I 1-1 R I I immunostaining was _'~ evaluated in a bl.indec.i m:~nner by two independent observers having no knowledge of l_ap~zroscor~ic findings. The inten city of staining w<~s evalu~ t~~~W t h.ree irnes ir: three different.
areas randoml_~~ selecte~c~l in the section <ind a :mean score was given using an arbitra~~y scale (0, abserut; I_, light; 2, moderate; and 3, inter-:~~~) . High c;oncorda:mce between the two observers was found a~ determined 'try t:rlc~ K measure of agreement ( rc =- 0 . 8 9 ) .
Dual Immunoflnorescen! ~>ta ~~.ning Tissue se<:t_ioris were treat=ed and incubated at room 1_'i temperature with the rc.ro.~se monoc:lc:~nal ar~t:i-IL-1.RII antibody as described earlier. ?after a PBS ri:~se, the sections were incubated for: 60 minut: e,s wit":~ a rabbit= polyclonal anti-IL-1(3 antibody dilul=ed 8: 1, t:~()~~i in PB S-1° bow-le serum albumin (R &
D Systems) , washe~:l in i'BS, incubated fo60 minutes with a biotinylated coat anti-rabbit antibody ;V'ect=.or Laboratories) diluted 1: 100 in PBS-1 '; bovine ser~~rm albumin, washed again in PBS, and fina_Lly incul;~ated simu=its.neously for 60 minus=es in the dark witr~ fluores<:ein ~scthiocyanatE,-conjugated streptavidin and a rhe~e~amine-conjugated goat: anti-mouse 2.'~ antibody (Sigma), which were used at a =final dilution of l : 100 and 1: o ) in PBS- :1°; bovine serum a=ibumin, respect=ively.
Slides were then mounre~1 with Mows.o ~ t~ which p-phenylenediarl:ine (S=igma, an anti i-fad:ing agent, was added at a final concentration lit: 1 mq/ml, then observed under. the Leica microsc~~pe equi~~ped for fluoresce;nc_:e with a 100 w<3tt UV
lamp and phot~~mic:rogr~,phs were rnac:le with Kodak 400 ASA .film.
In every expe::riment, ~~er_~tions from each endometrial tissue incubated with normal mouse and normal rabbit IaGs (used at concentrations ecluiVa.Crut t~:> thc:se~ of the primary antibodies) were included as negal:i~~F ~~ontrcl.s.
Western Blot Analysis Frozen endor~uet.rial tissues were directly homogenized with a mi.::r,:.;scal.e tissue grinder (Kontes, Vineland, NJ> in a bu~rfer containing 0.5s Triton X-100, 10 mmol/L HEPES (pH 7.4), '~50 mmol/h NaCl, '._% mmol/L
ethyleneglycol.tet:raacc:-t.:a c <~ci_d, 2rnmollL
ethylenediaminetetraac.eti.c acid, 0. 02 o NaN3 and a mix of. anti-proteases composed by 5 Lrmol%L aprotinin, 6.3 ~amol/L
leupeptin, and 3 mmol;':L, pheny:lmc:tr.y:lsulfony.l fluoride. Tissue homogenate was then iro~_v~ba~.ec3 at ~l °C for 45 minutes und~sr gentle shaking, and calirifuger~ at. 11, 000 X ~~ for 30 minutes 1.'~ to recover tree soluble: <~xt:ract, wL,~ose total protein concentration wa:~ det<:rroined using the Bio-Rad DC Protein Assay (Bio-Rad Laborat:~~r i.es Ltd. , M_ississauga, Ontario>, Canada) . Pro~:ein:> (10i) Eg) from each extract were then heat-denatured in a bc.~ilin<~ bath fo.r 3 minutes in 5X sodium dodecyl sulfate :>ample: buf:Eer (1.25 mol/:L T:ris-HCI, pH 6.8, 50o glycerol, 25~ ~3-m~=r~:aptoethanol, 10° sodium dodecyl sulfate, and 0. 0~. o brc:rro.~pheno:l blue ) , separated by sodium dodecyl sulfate-polya~: r ylam:ide ge=i.. electrophoresis in 10 0 acrylamide li_neaz: gra;i~rvt gEVl slabs, a2d transferred onto 2.'~ 0 . 45-}zm nitrocel~.ulosF: rr~embrane~~ using a l ectrophoretic transfer cell. (trans-~1<.~t, Bi.oRad. Nitr~~c:eilulose membranes were then immersed in f:F3S containing 5o skirmued milk and O.lo Tween 20 (blacking so_'.ur ~.on) for ~. hour at: 3 r°C., cut into strips, and incubated c~~Terruight at. 4°~ -,aith a monoclonal mouse anti-htz:man IL-1E<.I I antibody ( 2 }.zg/ml of blocking solution) (R and D Systems) or with normal mouse immunoglobul:ins (lgGs: ;f t~hE=' same immunoglo~:ulin class and concentration as the pr:irnary ant:i.body ;F: & D ;systems) . The specificity of: the imm~znc.~rea<aion way: al so verified by pre-absorpti.on of the antilooc:~y with <,rn excess of IL-1RII (20 ug/ml) . Thereafter, tr~f~ ::~trl~~s were ir~c:l~bated for 1 hour at 37°C with Fc-~>pec.ific ~~eroxidase-labeler goat anti-mouse antibody (1:3000 c:lilut.i.on in the biockiria solution) (~,Tac:kson ImmunoResearch Laborat~~ar.ie~ :Lnc. , West C-rove, PA) , washed three times in PBS/0.1~ Tween 20, incuhat.ed with chemiluminescE:nce reag~erz~t (Amersham, Oa~:zrill.e, Ontario, 1C Canada) for 1 minute, -z:i:_r-dried, wr_appecz in a plastic bag, and exposed t.o a Kodak X-GMAT AR film (Eastman Kodak, Rochester, NY) for 1 rr ir~ute.
Statistical Analysis IL-1RII staining scores follow an ordinal scale.
I'_~ Statistical analysis ~~rs per:formed usinct Fi;~her' s exact test (Siegel S. et al, 198;, and the Bon:fe.rroni procedure was applied when more than two groups were compared. Comparison of patient' s <~ge was L:erformed using o_ze-way analysis of variance. All. ana.lyse:=; were performed .n._~zng the statistical 20 analyses system (SAS nstitute Inc., ~a~~y, North Carolina).
Differences were cons:ic~~erecf as statisti<.:a.lly significant for P values < 0.05.
Results Po" itive imnourrohistoclierrzica.L :~taini.ng of IL-1RII
25 was observed in many compartments of en~.~ometrial tissue. In the stroma, i.mmunosta:~_n~ng was in general weak in the proliferativE: phase arid more pronc:~unced :in the secretory phase of the menstrual. cycle, mainly in isolated aggregates and microvessels (Figure 6). However, the most marked 30 staining was prirzvaril~~,, located i.n epithelial cells.
Immunostaining was ob:..erved all around cells (cellular staining), but: also hacl the appearance of an intense brown extracellular deposit ;.ru~~t= was predc>rn:i.nantly ! ocated within the lumen of t: he gland=~ and t:he apic~al. :>ide of surface epithelium (luminal stainir;g) (Figure 6).
According tc; o°c~cer~t findiric~;~, I:L-LRII is a decoy receptor that can be .r~.eleased by proteo~.ysis from the membrane-bound receptcz: extraceilular domain. The resulting soluble receptor seem; to retain the same affinity for its natural ligand IL-1~3 ,~':olotta F. et al, 1993; Colotta F. et al, 1994; Bossu F. E:et :L, 1995; ~Iymon~ ~:n.A. et al, 1995) .
Dual immunofluorescenc;=. analysis using antibodies specific to IL-1RII and I:1~-l~ clear°:l.y :~noweca thaboth antigens were co-expressed within the Luminal deposit that makes plausible the formation of :CL-1RII- _ I~~-:13 ~~omplex ( F~.gure ? ) .
1_'> Within the :ame endometr:ial ;~er_-tion, the intensity of staining varied withc:~ut any discernible or apparent order from one gland to anoi ht~~:r. Furthermore, great variations between biops:ies taker. from different women at different periods of the menstrn:a..l. cycle were notfed. The intensity of IL-1RII cellu:Lar and _i,,.zminal stair~c.ing was scored in a blinded manner by two indepenc~et observers using an arbitrary scale.
Statistical analysis c,f the data regarding the influence of the menstruaa cycle wing the Fisher's exact test showed that cellular staining was c~ffecti.vely more Lr~tense in the secretory than ir; the px-oliferative phase of the cycle both in stromal ( ~' _ ~X 1!::~ ~~' ~ and epi toel i.al ( P = 0 . 0310 ) cells (Table 7 ) . However, ora:L yu a weak tendency fo~.r an increased luminal staining in gar~.dular and surface epithelium was noted in the secretor~,~ phase of tree cycle as compared to the proliferative phase (.' -- 0. 13'70 > (Table 8 ) . Scattered brown '~9 85409-5 Table 7. Number of No r:na_L and E:ndometr~osis Patients According to the I:nt;er~.~i.~~y o~ II_-1R.I:L C:ellul.ar Immunostaining in the Stroma, and in t_.Ze Crlands and :~unface epithelium c_1 and ~nc~s Stroma su, face Intensit~~<~ ~-pvdueliun ~

sta._ni.ng Inr ensivy of s'~ ~i q _ -__ .._____-_ nin _. - J .. 3 P
0 1 3 C 1 '=
<

value* alue*

Controls 3 19 > 4 a: 13 1~ 7 Endometriosis 5 _ J 0.1120 4 ~ a:6 1 0.0660 (total) Stage I 1 9 3 0 0.7_'7':0_ ? 13 0 0.1030 Stage II . 10 0 0.3240 C 10 i 1 0.2580 Stage III- 1 6 ~ 0 ().71%0 7 ~1 8 0 0.3680 ~

IV

Fertile 1 1.0 3 0 C.1910 1 9 14 0 0.1280 Infertile 1 15 ? 0 Ci. 3~~0 ~ 'i212 1 0.3010 Controls G Za ~ 1 ~: 9 E Z
!

Proliferative phase Secretory 1 4 :_'3 2:10--'r C: 9 71 6 0.0310 phase Endometriosis 4 14 a :) _ 14 4 0 Proliferative phase Secretory 1 11. 9 0 0.0011' _ '% ~:2 1 0.0004 phase F?roliferative phase Control 2 15 1 ~ 9 (. 1 4 14 0 0.2430 14 ~~ 0 0. E>060 Endometriosis Secretory phase Control 1 4 ? 3 C~ 4 1 6 1 11 , 0 C' . 0910i , < 1 0 .
2 C)5:30 Endometriosis *Comparison with controls, P valme~~ corrected b;~ the Bonferroni procedure.
'Comparison of the p-~oliferat.ive phase with the ;euretory phase.

Table 8. Num;'»°r of Noznnal and Endometri~~sis Patients According to the Inter.s:ity of I~T~-1.RII Lumi.nal Immunosta:i.ning in the Glanda and surface Epithelium Gland::;
~.~~i ~u Mace epithe L~
~,am intr:nsity of -~_i-:wing st J

0 3 P value Controls ? ~~ ' !:i :!! 6 6 Endometri.os:is 37 ~ 6 10-~*

(total) Stage I 17 _. ~ ~_' 0.0006*

Stage II 11. (? ~ 0.0015*

Stage III-IV 9 . ~ _ Ci.0360*

Fertile 16 . (? 4 0.0010*

Infertile ~ . _ ' 4 ?~ 10 1 s*

Contro~s Proliferative 5 ... 9 phase Secretory phase :' 8 ' 9 0,_~370t Endometriosi.s Proliferative 13 ~ _.

phase Secretory phase 24 E: O '- 0.05901 Proliferative phase Control 5 .. 9 ?

Endometr.iosis I3 _ 9 0.0760 Secretory phase Control 2 8 % 4 Endometriosis i_'9 6 0 ? 8 X 10-~*

Comparison with c.>ntroa ~; only si_gr:i.ficantvalues werecorrect=ed P by the Bonferre.~.i r.:>cedur;:~.
p f Comparison. of prolifer~_'_i_:~:= phase ory phase.
the wit? the .secrat deposits, wh_i_ch were emcounterec~ ~_n stron~:a, beit :less the al markedly than in lumi :aa:i an:d glandu_Larepithelium,were also more frequent the ~~e<: retort' phasean in the in th proliferative ph ase, k;at: the differenr~edid riot reach the level of statist ical :,ic~nifican:_e 0.093Ci) (,F' _ .

Ba>ed c>n these findings and the recently reported down-regulatory propeotses of I~-1RII toward IL-1-mediated l.'s inflammation and cell ,:~c~tivatiou, we further investigated whether there way any ,~lte.ration ~_.n IL-1RII expression in the endometrium of women crith endometi°iosis, a frequent endometrium-related patrlology associated wilt-, an aberrant inflammatory process ::~b;erved not only in ectopi.c sites where endometrial tissue abn~.>rmal.ly implantsr bat even in the eutopic intrauterine endornetriunu. E~ndornetriai biopsies were collected from 54 women presenting l.a~~aJ~oscopical and histological evid~~nc:e .>:_ erudomet.r_io~;:i_:~>. As i.n normal women, cellular staining in w~::>rnen with endometriosi.s was significantly higher ~.ri thE: secretory pr~.ase than in the proliferative phase of 1=.he menstrual. cycle, b~~th in stromal and epithelial_ cells iP -- 0.0011. anct 0.0003, respectively).
However, when compares; to t: hat o ~ nc>rma_! women, this staining 1C~ showed a marked trend fc:~r a decreased :iratentsity in glandular and luminal epitheliurc (P =- 0.0660), wlereat> in the strc>ma the difference between women with and without was less evident ( P = 0 . 11;> 0 ) . F a rthermore, t:he clecrea sed immunostaininc~ observers in endorr~et.r.i_a:l. epithelial cells of 15 women with enc~ometrio:i..:~ compared to normal c:~ntrols appeared to occur in the secret o:ry ( P = C? . 0 '~30 ) rather than in the proliferative (P == 0. F(O=~J) phase of the menstrual cycle (Table 7).
Scat=tered bxoorn deposits we.rf: also observed in the 20 stroma of women with endomet:riosis. a:~ ~.r~ normal women, they were more obv=~~ous in t nE_ sec.retory phasF= of the menstrual cycle and showed a corr~~~arable le~,rel of .>taining. The most striking difference between women wi_t~~ endometriosis and normal women was howe~;er detected at the _Level of IL-1RII
2,i staining in the lumen ~of endomet:rial_ gl<~nds and the apical side of surface epit~heli.um. In fact, statistical analysis of the data (Tab:Le 8 ) showed a cons.ic~erablc~ lack of staining in women with endometrio~.i;: compared to :normal controls (P = 10-Endometriosis pati:~r~t~:> were then stratified by severity of 30 disease (stage I, II, a::ud III--IV). Comparison of individual groups using i~he Fisher' s exact test= an<:~ the procedure of Bonferroni showed that the intensity of staining was significantly lower ir~~~ach endometr_iosis stage compared to controls, but the most: ~.ignificant dec:re~ase in IL-1RII
immunostaininc~ was toy nc_.i irl the m1 l.der :st=ages ( I. and I I ', ( P =
0.0006 and 0.0015, re.pe~tively) . ~urt:~rez~more, statistical analysis of t:he data t <~kin~~ into accountv the phase of the _'~ menstrual cyc:_Le revea_ r,~ttn at the most marked drop in Ih-1RII
luminal staining .in er~,Icrn~~r::r.i.osis occ...zr~ed in the secret:ory phase (P = 8 X 10-~) . "r1s watt ai.sc> observed in all stages of endometriosis,. but wat mere pronounced :i.n the early ( I and I I ) ( P = 0 . 00a? 0 ) than i.r: trie 7_at a st::~ges of the disease ( III-IV) (P = O.OC)60) . In cc,u:t.rast, during the proliferative phase of the menstrual cycla , th~~ di_ffert~nce ~ri IL-1RII luminal immunostainin<I betweer women with and wr.thout endometriosis was perceptible, but: ci..c~ not reach the ievel_ of statistical significance (P = 0.0%~~0) . ThE: 54 pa ti.ent:s with endometx:iosis l~~ were also stratified f;~:r :infertilit.y anc.l IL--1RII lumina7_ immunostainin<I scores Uaere compared. Usvng t:he Fisher' s exact test, both fertile ancinfertz.le pat::it=nts wi_t:h endometriosis had decreased levels c:t imrnunost:aining c_-c;mpared with control women ( P = 0. 0010 and 4 ~ 10 ~~', respecti~~ely) , but more significant d=_fference::~ in :inferti 1e wc~rnen with. endometriosis was noted.
Repoesentati ~Te~ examples of Ia,-LRII= immunostaining in the endomet:rium of Va:~,mer~ with and without endometrio~~is are shown in Figure 8 (~, normal. secr_etcry, day 24; B, 2~~ endometriosis secretory%,, clay 26) . Node the fine brown immunostaining around ::;el.Ls both in the stroma and glandular epithelium, and the brosam depc>sit~ in tlnf-~ lumen of gland; in normal women. No i_mmun:a:_eaction was observed in negative controls in which the in 1. i-I:..-1RI I ant:ibc;dy was replaced by an equal concentration ~.f mouse immun«c~-~cbulins of the ~~ame isotype or pre-absorbe:I w:it: h an excess c>f IL-1RII before incubation with endomet:~ la: tissue sec-_-ons (data not shown) .

E>3 85409-5 To ~ontirm ::.mm.unohs.st~w.chemical data regarding the expression of IL-1RII i.n normal and endometriosis women and to determine whether t:l;~~ mobi.lit y of the endometrial re~septor corresponds t:c the knc:car~ mo_Lecular weight of this protein, '~ equivalent arnsunt:.s of erndomet::rial proteins were analyzed by Western blot . Enclomet~ :i~;l bic>ps Les were selec:ted from the proliferative anc3 the sec:reto:ry phases.
Figure 9 sh<w:-~ that monoclonal anti-IL-1RII
antibody reacted prima r:i 1.y wz th a 68-kd banca and a doublet of 1c) 45- and 48-kd mol.eculacr weight bands. Sixty--eight and 45 kd correspond to the rep::rted mo:lecu~_a_r weights of the membrane-bound and the so)_uble forms o:>= the :IL-1.RLI receptor, respectively (BoraseJh_ L~. et <~l, ;_996; ~c;lott:a F'. et al, 1993; Colotta F, et a-., 1994). Mir:or bands of lower molt=cular 15 weights recog~izEed spFecific:ally by the antibody have not been reported previous>ly an:~ may presunuably correspond to degradation products. Hawever, foi:- the same amount of total endometrial ~~roteins, the intensity of If~-1RLL bands waa clearly lower_ in biop.ies from women with endometriosis 20 included within the surne experiments.
Discussion In the pres~:o:t study, wE: have shown t:hat IL-11~II
expression wa,s ubiquitous throughc.~ut the endometrial tissue, and was in gE_>:-rera:l cye:le phase-dependent . The most marked 25 immunostainin~~, which ;~L:~peared rr;ieroscop:ic:ally as an extracellular brown deposit, was observr~d in the gland', lumen and the apical .aide of lumina:L epithe:Lium. The :luminal secretion most likely corresponds to the soluble form o:f IL-1RII. Western blot an<:,lysi:~ of IL--1RII in endometrial 30 biopsies have shown, n fact, the presence ofbands whoae molecular wei.~~hts are c~quival_en?~ to those reported fo.r the membrane-bouru:~ (68 kd; and the so~_uble 1,45 kd) forms of the IL-1RII receptor. NumErc:,us recent studies have reported that IL-1RII is a decoy recveptor t:hat could be released in a soluble form after en:~.yrr~atic clr~a~.Jage of the ext.racellular domain of the membrane-found receptor. The soluble receptor ~ possesses the ability tc-; bind Ii..~-~i.)3, the circulating and most active form c>f Ih-1, _ x:r-~ibiting t.lue.reby the interaction of the latter w:i_th t:he fi.nc~t:.ional 7L--1RI anc:~, ~;cnsequentl.y, IL-1-mediated cell activ.3t:ion (C;o:Lc:tt:a F. ;~t: al, 1.993; Colotta F. et al, 19':4; E3ossu >'. et a.1, 1~:~95; Symon~ J.P_. et al, 1995). Dual i:mmunofluc:rescenc:e analysis showed that IL-1RII
and IL-1(3 wez~e both e. fc,ct:ive.ly co-expressed within the luminal deposit, whicto makes plausible the formation of IL-1RII-IL-1(3 co:nplex. These finding:> might be cf interesting physiological. signi.fic:ar,ce. In fact IL-:i i_s cne of the major 1'.~ cytokines that are in~,rc~ved in the different cyclic events occurring in human enclometrium (Tabibza:~eh S., 1991; Simon C.
et al, 1998b). The ~cyt:t~Jkine has been demonstrated as a key mediator in t: he attacume:~nt of the embryo c>nt~; the endomfstrium and the impla:ntat.ion ~.rc~cess (Simon C. rat a:1, 1998b;
Psychoyos A. , 1993; St~et:h K. V, et a_L, 1991. ) . In the peripheral c:i.rcul_atior~, IL-1 leve.s increase after ovul<~tion (Cannon J.G. et al, 1~?85), and, locally within endometr:ial tissue, IL-1 :~roducticn ha:~ been >hown o considerably increase in t:~e secret:ory phase, x:eaching its maximal levels at the end of the menstrual cycle (Kauma :~. et al, 1990).
Hence the considerabli~ mpor_tanc:~e of the local availability in the endomE~trium of a rec~uiatory mech::3ni.sm such as that of IL-1RII that ~;an count e:rba:lance on- buffer the local action of IL-1 and maintain its leve:Ls wi~t.hi_n the physiological limits during the crucial per:.ic~d of_ implantation and during the inflammatory-like pro::ess that tal<>es place .in the endametrium at the end o~- eac:h me~::t rual cycle.

To investig~:te the ro:Le of IL-1RII in endometrium-related disorc~er~;, we .:assessed .Lta exprc~~ssion in the endometrium of women ::.offering from endumetriosis. The disease is as:~oci.ated w:i.t.h an immuno-_i_nflammatory proce ss observed cons :LStently ~L:c~ t'ne per.it.oneal c:avv_ty where endometrial t=issue abnormally develops (Kauma S. et al, 1990;
Witz C.A. et al, 1997; 'Jinatier D. et a.i., 1996) , and recently noticed in the eutopic intrauterine endometriurn of patients as well (Tseng J.F. et a.1, 1996; Isaat.~:s«r. K.B. et al, 1990;
Vigano P. et girl, 1998,' . Acc:ordirug to ~:~ur dat:a, the eutopic endometrium of women ~,i~h endometriosis expresses in situ increased levels of MC ~?~-:1 !Jolic:oeur C. et al, 1998) , a chemokine endowed w~_tr ~..he potent: ar~:iiity of inducing monocyte / macrophage cl~emoattract_ion and activation (Leonard 1~~ E.J. et al, 1.~~90) . Furt:herrno.re, cultu:rE~rl endometrial epithelial cells from women with endometriosis displayed an increased responsivene~s:~ to IL-~. in vi tx c~~ by secreting higher amounts of MCF~-1 than :::c~:Lls from normal women after exposure to the same concentrations of the proinflammatory cytokine (Akoum A. et ail, 1995;;: ) .. I-ios~revex , t=he caause ( s ) of such an exaggerated inflammatory reaction remain unknown. The present study shows a dramatic :Lack in I:h-1RI:I expression in endometrial tissue of women with endometriosi.:~. This ways particularly obvious at the level of _Lh-1RII luminal secretion, but: was alS~:> IlOtiCeab.:Le at the ievc~l of the cellular expression bot:i'1 ir: epithelial and stromal cells.
Western blot analysis ~of .iL~-1RII expr_es:~ion i.n endometri.al tissue confirmed :immurucoh.istochemical data as .:it showed that the intensity of the 6ri--kd and lower rnoi.ecular weight bands recognized specificall vy~ i:~y the anti-=CL--1 P.II antibody, was markedly lower in women witi: endometr_Losis as compared to normal controls.

The most stt .i~;.ing lack ~.n LL-1.RII l.uminal expression wa:> observF-d i.n the earl=~est and initial stages of the disease ( stages I ~~nd :~ I ) . (:)n one hand, t h.i.s in keeping with numerous studies i=~dicating that aadometriosis is more _'i active in the initial stages (Haney ,?~. E. et al, 1991; Vernon M.W. et al, 1986; Les~ey B. et al, 1994) and is consistent with the pattern o.f MO:'~'-1 expr.es.~iori t:~at we observed in the endometrium oi= endometr:iosis patients that increased in initial and decreased :.u-~ .l~~te endc;me~tri~osis stages (Jolicoeur C. et al, 1998). On tre otrier hand, these results suggest that abnormal_ IL-1RII Pxp.ressi_on rr~.ay 1:>e involved in the initiation of the inflammatory process in the intrauterine endometrial t~_ssue when-a tree disease .is believed to take origin. Interestingly, our study also sruowed that defective 1~~ IL-1RII expre:~sion was rru~re si.gnific:ant i.n i.nfertile thin in fertile women having endometriosis, which suggests an involvement in endometr,:i~.~si.s--associated infertility.
The mechanisms underly~na the decreased expression of IL-1RII in women wit:_~ endometr~i~:~s_Ls remain to be further 2C elucidated. The most si.c~nificant deficiency occurred at the level of IL-1F;II l.umim:a:L aecreti.~on in epithelial cells, in particular in the secrr~t=orl: phase cf the menstrual cycle.
This would suclgest an i_ruhibit=ed sheddi.nc~ of the receptor in endometriosis occurring throughout the cycle, but to a 25 greater extent: in the :~ec~ond phase. At. the present time, it is still unclE:ar what molecular and biochemical pathways could be involved in glue <~eneraticr: of :,oluble IL-RII.
According to recent da~.a, matrix metal~o-proteases rather than differential split_ i_nc~, play a key role in the production 30 of soluble decoy RII bay enzymatic cleavage from the cell surface receptor (Orlaricio ~. Et al . , 1997 ) . The observation of higher levels of ce_L l:zlar staining vri epithelial cells of women with enclomet:ri.os.i..~; irv th.e secret:ory phase of the 67 8.'409-5 menstrual cycle as com?~:a:r-ed to the proliferative phase (P =
0.00037, Tabl a 7) , makc::~; ylau.sible a potentia, inhibition of IL-1RII release from t:ne cell surface ir: t:he secretory phase.
However, our resu.l.ts a L.;:r~> show trra,: either cellular staining in epithelial cells or tr~:e intenwit.y c>f the 6~ -kd band corresponding to t:.he reported membrane-bound form of IL-1RII
receptor was reduced i a ~aomen w.it h endc>rnetriosi.s. This suggests that beyond a p~t:ential aberrar:t release of sTL-1RII
from endometrial ~,:ell~; c:~_ women wii_h enGOmetr~osis, a deficiency in IL-1RII ~:~rotein synthesis and/or a reduced IL-1RII mRNA levels or gecoEe t:.ranscr:iptic:n rriight be involved. In fact, our prelimiruary analyses ef IL-lRlI mRNA levels in the endometria of women wi~:.rn and without endometr_osis tend to support such ~. hypothe::~:is (data root: sr~own) .
In c:onc lut;i.or:, t~hi:~ i s the i ~z st study to show the expression in endometr:i.al tissue of t:rie decoy IL-1 receptor type II, a specific nav:,:~.aral inhi.t~it:on of IL-1. that plays an important role in the c°egulat:ion of_ z1~--j ~ act ivity in the uterine environment. Fn.~ra_hermore, our study revealed a 2C striking lack in f:L-lE~',l:f: exp~~ess.on i.re women :suffering from endometriosis. This ma;y represent a plausible mechanism underlying immuno-inf:l.,:-,zrrunat.ory changes c:~bserved in the eutopic endometrium of women with endometriosis as well as in endometrial t-_ssue abru:o_ma:i.ly implanted i_n ectopic sites.

Abstract Cytokines su;Li as inte~~leukiri-1 (IL--1) play a major role in the reparative and inflarrunatorv-like i?rocesses that occur in human endometri_~am during every menstrua:L cycle, but they also seem to be iwl~'~~icat:ed i n c:Vrit~~ cal z:e~producti.ve events such a~~ ov~~lati.c.->n avid impl.aritat~,yc>n. Interleuki.n-1 is tightly regulated in t ire body by a Complex network of cc>ntrol systems . In the ~>reseri;:: study, ~,ae ex~~rn~_ned the expression of IL-1RII, a nat:ura.1 speavi.f_ic inhibitor of IL-l, in the human IC endometrium arid found ~~-rro interesting distribution and temporal pattern of ex~:~Lc:~ssion throughout the menstrual cycle. Immur_oreacti.ve C:i~-1RII w<~s found in stromal as well as epithelial cells, W.z1= it was predominant w:Lthin the 1_umen of the glands and the v~;:ica.1 ss_de of snrfaCE' ~aplthelium. In I~~ situ hybridiz~ition anc=everse tran~;cri_ption-,~olymerase chain reaction (RT-F?CR) analyses showed hi.ghc=! levels of mRNA in epithelial than in str.rrr~a:1 cells. Thf~ ='~L-1RII cellular and luminal secretion foll~w~~d a regulated cycle ,phase-dependent pattern of expression. ,although eleva_c~d in the late 20 proliferative/early sec_retory phase of the menstrual cycle, IL-1RII luminal secret i_c~-n :>:ignit-icant.iy decreased in the midsecretory phase, reaclZing it:> lowesr_ levels at Day 27_, before augmenting markedly again during the late secretary phase. This pattern c~f expressi.on was less obvious at t:he 2_'> level of cellular st~a:aruing, as examined by immunohistocriemistry, but it was corr~:~borated by Western blot analysis of 7:7~-lP:iI protein and se~mi~:~uarnt.itative RT-PCR of IL-1RII mRNA in the wL~c:>l.a endometrial t:i_;~sue and separated glandular epii~helial cells. The reduced ex~>ression of IL-30 1RII within t=he implantation window suggest: the existence of accurate regu_Latory mes:~rianisrns ':hat, by down-regulating IL-1RII expression, al.LeT, i_~te IL~-1 iruhibit u_~n during this crucial period and :Eac: L l.itate il,-I pro imp l.ant.ation actions .

The elevated cxpressi<;~r «f I:h-1R'~.L observed da.zring the late secretory phase suggest; an s_nve,~v~rnent: ef II~--1.RII in ccntrol of the proinflammatory t;tat:e that ta~:es place in the endometrium d~:.ring the p.nennernstrua'~ anc~ menstrual. periods.
Introduction Interleukin-.. ;; IL-7_ ) i= the germ used 1~o describe two polypepti~les (IL-l.ra. sand IL-1~3) that: play ~~ key role in immune and inflammator~~- react:ions (Di.r~.arello ~:.A., 1996) .
Three receptors for IL--J., type I (:I:L---wRT ) , type II ( IL-1RII ) , and type III ( IL-J_ 1RI II ) , have l:>een ~..c~enti.fie~d i.n different cell types ( Di.narello .:'1. , x_996; Arend W . P. , 1991 ) . Cell activation in response to IL-1 appears t c be mediated exclusively vi.a the IL-~.'tW (Suns J.F. et al, X988; Sims J.E.
et al, 1993) , with coe:~=~;ression of a rec:eptcr accessory protein (IL-1F;-Ac I? or i.L,--1RIII) being crucial to IL-1-mediated signaling evera:"; (Greenf:ec~er :>.R. et. al, 1995;
Wesche H. et al, '_997; Iv:m°herr C. et a-.~, 1997 i . In itself, IL-1RII is nct: a signa.l.i.ng molecule but, in fact, is reported to be a decoy target: ef IL-1 (Mcl~qahar: C:. ~ . et: al, 1991.;
Colotta F. et al, 1994) . ~a.c~dition_vl-'~y, IL-R.II could be shed from the cell surface a~a sclub:Le mol.ec:ule that would then capture IL-1 ~~nd :inhibi.t: its binoling t:c> IL-1RI (Girl J.G. et al, 1990; Symons J.A. ~_ec: al, 199'p; Bossiz P. et a1, 1995) , thus suggesting an i.mp~ol:vt~<~r~:t rolE: t=c~r_ CI,-RII in regulating the biological_ activiti.~~s r_~f IL-~.
The avail~ibl.~:~ ;data ind.cate t~ruat I:L-1 is involved in the regulat:ion of F,°u:i~::mnetrial. flm:ct::ic>ns ('L'abibzadeh :~. , 1991 ) . It has been sru.>~wo that I.L-1 :i;.., :;ecret;ed by human blastocysts, and it i>thought: t: o acts a; an embryonic si-gnal 3C~ (Baranao R. I. et al, 1'x'3?; . The cyt.okine wa s also detected locally in the endometri.;~:1 tissue dm:r.i-nc~ the late secret:ory f' 0 85409-5 phase of the menstrual c:vycle (KaiimaEt al, 1990) . This ~.

suggests a role in the t=:i_ssue ne~~rosisand disintegration occurring in the endom~:= t: ri_unn at d of th<- menstrual t he en cycle in the ,~bserice o: .mpi_anta-tic~n,which is not surprising considering tr,e simila~Pi_t:y betwef:~nse processes and those trie occurring during inflar mat: ion. lv~xE:~res~>i.on of the functional receptor of I:L~-1 ( i . _=h-1 RI ) bias>er: detected in the a . ,. be human endometrium as w~_= _i_:. (Simon al, 19~~3; Bigonnesse C. et=

F. et al, 2001a) , where:=:'t. ~~ppears play a key role in the tc, implantation ~~rocess .imon C. et 1994 ) .
( ' al, Due to its p:LE::iotropic acti.v~_t y and potent proinflammatoryeffects,, f:L-J_ is t:ight:~y regulated in the body by complex control sys.t:ems. In r>articular, two :Lnhibitors participate in these r pu.i-'~atory mechan~_s;ms : true =eceptor antagonist (IL~-lra) , wf-~.i,h binds avi.d:~y to IL--1RI and prevents IL-1 binding .:mc:~ signal trarrsc~uction; and IL-1R.II, which is considered to r>e:e a natur~a.l_ scw-enger for IL-1. The IL-1RII can vE~ry eff:ic:iently bind IL-1~3, whereas its affinity for IL-la and IL-Ira i~ 10- too 1.00--fold lower_ (Boraschi D. et al, 1996) .
In view of tine major r~>le of 1I~-1 in the regulation of various endometrial aad repro~~uc:t-~ve funct:'~ons, knowledge regarding the local ava~.'~abi~it.y arid rxcc:urate production of specific inhibitors fc>.r- Il~-1 in t:he en<~ometri~zm Throughout the menstrual cycle be~,:ornes. esserlt:ial_. The expression of IL-lra in the human endom~.>t: r_ium has been previously reported (Simon C. et ~:1, 1995x) .; IL;-lra immunoreactivity was elevated during the proliferative pha:~e oi= the menstru<~l cycle, whereas endometri_a.l ce~l._:1:~ <xppeared to express intracellular 3C IL-lra (icIL-1_ra) . Th~=~ objecti.ve of the pres~snt study was to assess the local avai.l;~x:,i_Lity and expre.,sior~ of IL-1RII

throughout the normal rnerlst_rual cy~:le t:c~ further elucidate the mechanism; controli=i.ng LL-1 act.ivz.ty in the endometrium.
Materials and Methods SurJ~ect.~,. Women who pa .rtic_~1-<rt ed in the study provided infcrmed consE_~rit ~-or_ a protocol approved by they Saint-Fran~oi~> d'l~ssisE.~ flosp;~tal Et:hic-r Comrrci.ttee on Human Research. They>e women ;n -= 42) were a<~ec between 23 and 47 yr (mean ~ SD, 3~:. 6 1: 5. G ;~:~_ ) . The~y~ wE r_e i ertile, requested tubal ligation, and hac:~ a r.or_mai and regular menstrual cycle.
1C None had visible endonv:t=vial r~.yperpla:_~ia or nf.=oplasia, inflarrrmatory disease, ~.m:~ endornet:ric>s=i:~ at the time of clinical examination or laparoscopy. Women were not receiving any anti-ina:i.a:~rnmGtory or hormonal medication at least 3 mo bef_ore laparoscopy. ~1'he cyc~~e day was determined l~~ according to t;he cycle tn istory and h_lstologi.cal writeria (Noyes R.W. et~ al, 19t~'~;~ . Eighteen women wer~s in the proliferative phase ar7~-i ?4 i.n t:he secret:c>ry phase.
Col._Lection c.~fi ~ndometria l Biopsy Specimens.
Endometrial b__opsy spe.~:imens were obtained using sterile 20 pipelle (Unimar, Inc . , I~c~u~_ 1:1y-en-The i i_w, France ) . Samples were placed ate 4°;: in sl:~eril.e Hank balanced salt solut=ion (HBSS; Gibco BRL, Bur~:W gton, ON, c'anada) cc;ntaining 100 U/ml of penicillin,. 100 E.tg~'rra of streptomycin, and 0.25 ~g/ml_ of amphotericin. Samples ;~r~rc: then imme~,ti~~~tel~r transported to 2-'i the laborator~~, washes:. Twice in HBSS at 4°C, then snap-f=rozen on dry ice anti kept at --80''C in Ep~pendo-rf tubes for Western blot and reve:=se tran.cription-polymera~e chain reaction (RT-PCR) analyses or in Tissue-Tek OCT compound (Mi_les, Ins.., Elkhart, IN) :Eor imrnur.orist:.o~hen;ica7_ st~adie:>.
30 Immunohistocmemi.stry. Serial ~="_~-~m cryosections were placed on pole-L-lysine-coated glass mir:roscope slides and ~2 85409-5 fixed for 20 min in fc~r_maldel-~yde (;4° [v/v] in PBS; Fisher Scientific, Montreal, ~~, '.anada) . A.11 Lncubations were performed at room tempt~rature irz a humiatified chamber.
Sections werE~ rinsed i n PB:~, imrner~sed in PBS/la> (v/v) T_riton X-100 for 2 0 min at r~:~cw~-. temperature, rLnsec~ again in PBS, and then treaved for ~'() n~in with r~ydroga~~rl peroxide (H~>O2, 0.30 [v/v] in absc>:LutEa metl_ar~ol; to eliminate endogenous peroxidase. defter a r-'i3:; r_~nse, irrununos~:ainirng was performed using a mouse mor~oc:lor;:~i ants.-human IL-1RII antibody (primary antibody; R&I:~ Systems, l~linneapolis, MN) at 1~~ ~g,/ml in I?BS
containing 1« (w/v) B;A with a Vectastain Elite ABC kit (Vector Laboratories, Burlingame, CA) arid diaminobenzidine (Sigma Chemical fo., .'::t. Louis, r~IO) as ~:hromogen.
Sec:i~ions =ineul:~ated without= t.ne primary antibody or 1'_> with nonimmune mouse :serum were ir:.cludec~ as negative controls in all experiments. ;=Lides were viewed using a Leica microscope (rr~c~del DMRF!; Leica Mikroskopi_e and Systeme GmbH, Postfach, Wetz lar, c;e:rm~:~ny) , and photomicrographs were taken with Kodak 100 ASA film (Kodak, 'T c.ronto, GN, Granada) . The IL-1RII immunostaininc ~,aas evaluated in a blinded fashion by two independent observers slaving n.o knovaiedge of laparoscopic findings. The intensi tvy~ of_ =,tai.ning way eva:Luated three times in three dif_fererlt: ~~:eas that. werf= randomly selected in the section, and a mean score was give's using an arbitrary 2_'> scale (0 = absent, 1 -- Light; 2 - moderate, and 3 = intense).
High concordance betw~eo, the two ebser_v~~rs was found as determined by the kap~~a (K) measure o.f agreement (K = 0.89) (Armitage P. et al, 1~~'la4 ) .
GJes~_ern B_Lot .~lr_~a'ysis. Frozen endomet.rial tissues were directly homogen-i zE_ed with t.tle use c~f a microscale tissue grinder (Kontes, Vine-~am:~, TJ~J) i..n a buffer containing 0..50 (v/v) Triton X-100, I(: mM Hepes (~~H '7 ~ 4' , 1'~0 mM NaCl, 2 mM

EGTA, 2 mM EDTA, 0.02'=. (w/v) NaN3 (Shet~n KV,. et al, 1991) , and a mixture of antiprotc_°a: es composed of 5 ~M aprotinin, 63 ~M
leupeptin, and 3 mM PI~I'I-". Tissue homogenate was then incubated at 48°c~ for 4'_-; mi_ri undeo gentle shaking and centrifuged art 11 000 k: <, for .3(~ ruin to rec~:wer the soluble extract . Tot=al prote:i. r~ co~ncentrai_.i.on was determined using the Bio-Rad DC Pr_oteir t~:ssay (B:io--Rad Laboratories Ltd. , Mississauga, ON, Canac:~a~ . One-r:urndred micro«rams of protein from each extract wer_f:~ toeatec~ in a bciling bath for 5 min in 5X SDS samplf> bui=fer .;1 . f 5 M Tris--HCl [pH 6. 8 ] , 50 0 [v/v]
glycerol, 25'<> (3-mercax:ta>ethanol, x.0'0 [w/v] SDS, and 0.0:1 [w/v] bromophenoui. blur-) , separated by SDS-PACE in 10'~ (w/v) acrylamide l ineai:-grac:::i.ent si.ab gels, and transferred onto 0.45-um nitrc>cel~..uloser~oeml~ranev ;Sci~.leic:her & Schuell, l.~ Keene, NH) u.,~ing an a : et: t=rophoret__c: transfer cell (Bio-:Rad) .
Equal loading in each lane was confirmed by staining the blots with Ponceau S ( 2=. [w/v] ) . N.itroce)_lulose membra:rLes were then immersed in PBS contain_.,ng 50 (w/v) skimmed milk and O.lo (v/v) Tween aC (blocking solution) for 1 h at 37°C, cut into strips, and ncvubated c;vernignt at 48°f with a monoclonal mouse antil:urr~an Ih-1RI:'= antibody (2 ~g/ml of blocking solution; R&ie .systems) or with normal mouse immunoglobul:i_ns (Ig) of the sam~~~ ammunogiobulin class a:rld concentration as the I:~~ ima ry ant ik>ody ( R& D Systems ) . T:ize 2.~ specificity r:~f the .imt.mnoreactic;n was al:~o verified by preabsorption of the <3nt i.body witri an excess of IL-1RII (20 ~g/ml). Thez~eafter, ;:,:e s rips were in:ubated for 1 h at 37°C with Fc--specific oeroxidase-~'.abele~ goat anti-mous~s antibody (1:3000 dilut:ic,n in the blocking solution; Jackson ImmunoResearc:h Laborai:ories, In~;s:. , West Grove, PA) , washed three times .i_n PBS/0.:': ~ (v/v) Tween 20, incui:,ated for 1 min with an enhanced chem:..lumines~cence .system (BN( chemiluminesc:ence blon:tng substrate [POD]; R.oche Diagnostics, Laval, P~:a>, c~anada) , <~nd exposed to BioMax film (Eastman Kodak, Roche:~te:r, NY) fo3_ 5-30 v;ec for an optimal detection (a11_ bawds v:i. sibl a but not overexpo'~ed) .
In Siti.i Hyb.~:.ic_~ization. In situ hybridization was ~ performed as describe,:L :in .~;~r pre~,rious studies (Jolicoeur C.
et al, 1998) . Bl:iefl~-, cDNA for luuman II:.-1RII, a 1.3-kilobase .frac~:ment:, wa::; ~ub~Ylonec::~nto the plasmid vector pcDNA3 (Bossla P. et a., 1995) . B:..otin-labelE:~d cDNA probes were prepared by nick--t~ anslati:~~n .f:rom ~Ze er:tire plasmid vector with *:he IL,-1R:!:I cDNA (L.aw7:ence B. et. al, 1985) or from the pla~~mid vect~:~r alone (negative control) using a BioNick Labe.Ling SystE::~m (Gibco FsRI:,) . Serial cryosectio:ns were prepared and fixcci in formaldehyde as described earlier, then progres:~ive~.y delvyc~rated in al~zohol batr:.s (50 0-100 0 l ~ [v/v] ) . Sec~:ions werE:~ ~rehybricized with the hybridization buffer (50 0 [v/v] forrr~amide, 10« !~v/v] ;~ext.ran sulfate, 0. 1 0 SDS [w/v] , 2X SSC [single strength: 0. 15 M sodium chloride and 0 . 015 M :>odium cit: rate ] , and !..X Denhardt solution [ 0 . 02 0 (v/v) Ficoll (Amersharn F~harmaci F3.iotech, Inc. , Baie d' Jrfe, QC, Canada) , 0. 02 0 (w,'v) human serum altoamim (HSA) , 0. 02 0 (w/v) polyvinylpyroli:;lor~e (Sigma) , and 40 mM monosodium phosphate (pH 7 ) ] ) , tlierv hybridi. zed with _'i ne~/m1 of biotinylated probe in tie ~ybridi<.>.ation buffer. Biotin was detected by ~~ series a~f 45-min incubations at 3%°C with a 2~ rabbit polyc~'_ona=L ant:~,.b:ictin antik:>ody ( 1[v; v] dilution in PBS/ 0. 25 0 [w/v] HSA; E,n.zo l~iagr~o7t ics, Inc. , Farmingdale, NY) , a biotinylat:ed goat antz_-rabkvit polyc:lor~~al antibody (1 0 [v/v] dilution in PBS,'C~.25~ [w/v] HSP_; Vector Laboratories), and fluorescein isoth~_oc°yanat~e-conjvagated strept.avidin (0.50 [v/v] in PBS~'0.250 [w;'v) HSA; Ruche Diagnostics, Montreal, PQ, Canada), respecti~.;e:l.y. Slider were then treated with propidium iodine (1.5 ~~~~/ml of -~i:~tilled water; Sigma) which makes the nu<:leus vis:..ble in yellow-orange oru ultraviolet ~75 X35409-5 excitation, ,_~nd mounto:;~l with Mowi.ol (Calbiocr~.em, San Diego, CA), to which p-~>heny!.a-~rnediarnirre~ (Sigma), an antifading agent, was ac:~ded at a f=i.na 1 c:on~.verut.ratic~=i of 1 rtlg/ml .
Sections were finally cbse:rved v.:ncle-r tile Leica microscope '.~ equipped for fluc;resc~~r~eve and connected to an image analysis system (ISIS; Metasystems, Altlusshe.im, Germany). As negative controls, sec--clans from ear_h tissue were incubated without specific cDNA r~rc>bes or with nonspecific DNA probes prepared by nick-tran~::Lati;:n from the plasmid vector alone (i.e., withou-t the IL--1RII cDNA).
ReL~~°rse Trar~.sa~ription--Pc;lymerase Chain Reaction.
Total RNA was. ext.ractec:~ from endomet:rial- homogenized tissue with TrizolTM reagent <:~ca~crding to the manufacturer' s instructions (Gibco BFv~). The cDNA was synthesized using 500 1.'> ng of total cellular IIJ,'~ and 2. 5 N,M ra~ldom hexamers in 2 0 ~l of a solution containi?o~ 50 mM KC1, 10 mM Tris-HCl, 5 mM
MgCl2, 1 mM of each de:w:yribonucl.eotide triphc:~sphate (dNTP) , U of RNase inh.ibitc:r,, and 50 iJ of. reverse transcriptase using Gene Amp PCR Core K.itTr' (Perkin-Elmer, foster City, CA).
20 The reaction was incux:nted a~ 25"C for 5 min, 42°C for 30 min, and 99°C for 5 mirl.
For PCR analy:.;:is, 2-~1 aliquot.s of ~sach cDNA were amplified in ~~ final t%~,>:1..:.zme of '~0 ~1 c~orlt.ains.ng PCR buffer (10 mmol/L of Tris, 5U rnrnol./ L ot. KC:L, alnd 1.5 mmol/L of 2~ MgCl) , 0.2 mrnol/L of_ dT~l':Chs, 2. 5 i:7 of Tack DNA Polymerase (New England Biola.bs, Beverl.,~, MA, and 10c_> ~:>mol of each IL-1.RII
primer (forward primer, 'p' -TC;C A'fG T('JC AAA TC~~ TCT CTT-~' (SEQ
ID N0. : 1) ; reverse pr:imc:~~, 5' -TCC TGC c:C-~T TCA TCT CAT ACC-3' (SEQ ID N0. :2) ; ampl amer_ a=~ze, 5?6 ba:e pairs [bp] ) . To quantify the PCR produt.,t:.s r_omparativel~! and to confirm the integrity of the RNA, we rc>amplified a hcusekc~eping gene, glyceraldehyde-phospha'e dehydrogenase a:GAPDH), in a companion tube (forwan~~ primer, 5' -TGA 'I'GA CAT CAA GAA GGT
GGT GAA G-3' (SEQ TD NC_~. : ~) ; rever~~e prinuer, 5' -TCC TTG GAG
GCC ATG TGG GCC AT-3' . ~E~'Q ~:D N0. : 4 ) ; am~:>=(.imE~r~ size, 240 bp) .
Amplification. of IL-1~~::I7: was achieved with 30 cycles of 1 min .'l of denaturat~.c~n at 95'~, I min of ~:~nneal.ing at 60°C, and 1 min of primer ext.ensi~_~~~ at 72°C. Amplirication of GAPDH was achieved witr:~ 30 cycl~aa of 30 sec of der°~aturation at 95"C, 30 sec of anneai.ing at 6:~:~°~:, and 1 min of primer extension at 72°C. These optimal c:ondii:ions were de:.ermined following linearity tesvs using 1, 2, 5, anrl 10 ~.L of the RT reaction volume and 2w, 3f, anc! 35 amplification cycles.
Amplification of genon7ic DNA with these primers did not produce a signal, sugcxesting that the amplification sites crossed at le:~~st one rlt:ron/exor: xooundar-~,~.
Twe~.zty perce-nt: of the PC;R volume was then ana:Lyzed on a to (w/v) agarose gel in the presence of ethidium bromide and transferred t:.o Qiab:rane Nylon Plusrr' membz.:apes (Qiagen, Santa Clarita, CA). Membranes were dehydrated at 37°C for 30 min, prehybri.~~ized with a hybridization buffer comprised of 5X SSC, 5X De:~hardt sc:lution, 50 rlM NaH_>PO.~, 0 . 5 o SDS, 200 ~g/ml of salm~~n :>perm DNA, and 'v0','s (v!v) form amide;
hybridized in the same: buffer, but without Denhardt solution and with B2P-.Labeled I1_:-~1.RII or ~A~'D~i c:DNA; and washed with 1X
SSC, 0.2X SSC, and 0._.a (w/v) S%S, respectively, before being 2'.~ exposed to x-ray film (Eastman Kodak) for approximately 1 h.
Specificity of the amplification pzocess was verified by ~~outhern l~l.c::t hyb.ridi.zation. A negative control (PCR in the absence ov esD'NA) as well as a positive control (cDNA prepar~~tion from rnuman endometria:l tissue expressing IL-1RII) were included i.n each series of IL-1RII or GAPDH
amplification.

For eac~n enc~ometri_al biopsy, ?CR. was performed three times. ~Che quant: ity ~oi thE: F-~CR pra>duct:s was determined by densitomet.lic analog s i s~ or t=he i :nt:ens .ty of the hybridization signal. The relat:_ive ~evc~1 of IL-1RII mRPJA
_'> normalized tcGAF'DH mi-t~~ was calcu.l_ated, and the results were expressed as a percenta~ae ~of the c.c~n-troi value (positive control).
Ce1_1 Separat.i'~rz. Endometrial l,issue was minced into small pieces and dissc:;~.~:fated with ~~:oliagenase (Sigma) to separate epithelial g_<~:suds i:rom fibroblast-7_ike cells as previously reported (F~.k~.:~um A. et: a.1, 19'3.'~b) . These two cell populations were further purified using PercollTM density gradients (Arrcersham Ph~armac;ia Bi_otec:h, =ric. ) . The purity of epithelial or fibrobl~~st-like strcrmal calls was verified 1_'> morphological_Ly; immur~o~;~ytochemica.lly on coversli.p cultures using antibocz:i_es spec: f:~ c: t:o cytok:erat:in~ (epithelial cell marker) , vimentin (str o»~.al cell mG.rker) , smooth muscle cx-actin, and factor VIII i enc~othel iG.1 ~el:'~ marker) ; and by flow cytometry for the preerme of leukocytes using anti-CD45 monoclonal arvtibodies :3s previous.l_y des~,ribed (Akoum A. et al, 1995b) . C'c~lls were kept at -80"c: unt:il use.
Statistical Ar:alyses. 'rhe Ij~-iRII i.mrnunostain=Lng scores followE=d an orct:final scale. 'Iheret-ore, statistica=L
analyses were performE~d using non~;aramevric methods. The association b~E~tween in~munostain:inc:~ scorns and the periods of IL-1RII expre;ssic>n .:Ln '.he menstrual cycle as well as intergroup compar:ison:> c~f irnmunUst:.ain:ing scores were analyzed using the Fisher exact test, and t:he Bonferroni procedure was applied when more than two groups were :compared. The correlation between t:~e day of menstrual cycle and the immunohistocYu~mistry .:~cc.~res was e~~~aluatad using the Spearman correlation c~~eff icietat . ':The threshold <bays between the 78 k35409-5 different lea~°ls of staining (0, ~., 2, prod 3) were determined using the best combination of sensitivity and specificity values for a series of: cut-off days wit:~in the menstrual cycle. These thr-esho:ld days al:Lowed us to define or to delimit different expression periods. analysis of IL-11ZII
mRNA levels a.s determ.ined by serniquantitative RT-PCR was performed using one-wny ANOVA arid the '.Cakey test for post-hoc multiple com~:~arisons. wll analyses were performed using Statistical Analysis ._'ystem (SA:: Lnst:i.t ate, Inc. , Ca.ry, NC) .
Differences were considered to k>e statistically signifir_ant at P < 0.05.
Results Immunohistocr~f=mistry A rr~onoclonal antibody was asec~ to detect IL-1RII
1_'~ protein in endometria~A -issue sect:i.ons. Different concentrations of the antibody ~;5, 10, L5, and 20 ~g/ml) were tested to determine tr_e optimal :concentration to use. This experiment wa:~ perforn er; on Y'hree di. eY~ent series of biopsy specimens frorn diffE~rcril~ phases of them menstrual. cycle. A
concentration of 15 kZglrr.:l was seLec;t.ed, because it allowed an optimal ratio of speci.f:icit.y (minimal background) and sensitivity ldetectable po~~itive signal;. Examples of positive immunostainiri~.~ ~uait:ki ant: i-IL,--LIZ=~I ant:.ibody are ~~hown in Figure 10A. Immuncc~:1_a:~bl.a.:Lins of t.hf~ :;ame isotype and species, when used at an equivalent concentration as that of the antibody ;15 ~g/ml;, dil not display any detectable immunoreactivity (Fig. :I.~:)I3) . Trrununc~rf:.~a<:tive IL-1RII wa~>
detectable throughout cen:lornei=ri_a i t:i s:~ue, bot:;n in the st:roma and the glands. Brown :Lmmunostaining could be seen around 3C cells (cellular stainl.nc~; and along 1=he apical border of luminal and gI_and~.~lar ~~y~ ~t:kneliunu (:i.urn:i.nal staining) . This was also observed, alt.h~ough less markedi.~;, in microvessels and isolated aggregate :; t.hroughc:~~~t. t=he st_rorna in sections from late sec:=-etory er;dc>metrial tissues ;Fig. 10A) . It is noteworthy that lumin~. L. sec:ret:irm was :~~c~t_ uniform for all _'~ endometrial SE'Ctl.onS examined.
The IL-1RII nununo,staining way assessed semiquantitat=~vely by t~.ro independent observers in a double-blind manner, taking :~ ru~.,~ account the :int:en:>ity as well as the distribut.__on of tee immunost:.aining as described above.
Cellular and extracell~r:la:r :~tain:ing wf=:r a. scored independently in the stroma and in the glands and swr~ace epithelium.
Score distributions ac::;~.rding to the ~~a~~~ of the menstrual cycle are shown graphi:._~11y :in t'igurEs :L=_ .
To better un.~e.rstand I:.~-1R:L:L o:ycli.c va.riation:~, expression wa~~ defined throughout t=he menstrual cycle using the best combination cf;7ens=iti.v~_ty <~nc~ ~pec:i.fic.ity values for different cut-off days within the cycle. Our analysis revealed that after Dace 8, both ce11u:1_ai se~n;~it:ivity =
100. 0%, specificity = ~3~~ .'~a:, P < 0. COL; and luminal (sensitivity =- 55.6°-a, ;~p~:~cif_~city --- :L.Ot).0'>, P < 0.001) immunostaininc~ became ,~~gnl.ficant.ly detectablE~ in epithelial cells. A further increase in the intYen~ity of staining occurred after Days 21 ~~nd 22 , but teals increase was more obvious at the cellular (Day 21, sensitivity =- 88.60, specificity = 71. 4 0, F t.. 0.0~~) than at: the lurninal (Day 22, sensitivity = 86.1p, sE:~ecificity = 5().()<-, P = 0.07) level.
In stromal cells, cell u:l..a:~r. stain:i.ng remained. weak to absent throughout the whole p:rc:>:lit:eratiVe t>hase of t.tne menstrual cycle, but it increase:: sigroificantly after Day 15 at th.e beginning of the secret<:a:ry phase (.P <~ 0.001) . Extracellular staining was ~~irtually s.rnc:~etectable in the st.~roma, except weakly after Cray f.1 ire t: i:7sues f z om mica to late secretory endometria (sf=nsiti.vit y _ ~~ ! .2'0, ~s~>ec:if icity = 85. 7 0, P <
0.001).
Sta~.~istical a:ialysis c:f the d<~ta, using the Spearman correlation woh.f.f i_r_.lerlt:., rewea._ed a significant, positive correlat ion k::<.:t ween cel..lu:l_ar s~ aini.ng scores and day of the menstrual cycle, botri in epit:-relal (R = 0.59, P <
0 . 001 ) and stroma.l. (R - 0 . 9 6, P 5~ 0 . O1 ) cells . However, no positive correlation k.~e,Vween epithelial lumin.al staining and day of the menstrual c:yc=le was found (.R =- 0.17, P = 0.29), most probably because :a:f: a more 2:luctuati_ng E.x~~ression pattern. To better delineate these rf~:lationships, we determined the mean va.l_oaf~s of i.mrnunosta_~.ninq scores for each cycle day, and we found that they fo_Llow a third-order polynomial curve ( .Y = <3 -t- F3X + C'.~= + DXB? 1 F.ir~ , 11 ) . This curve shows that aft:e.r <~ rnax=imal. ir~c:_rf_~~~.e at: approximately Day 12 in the proliferative phase of the menstrual cycle, luminal staining of IL.~-:LI~II ciecl..ined gradually in the secretory pha:~e, reacr:i.n:~ its minimal level at approximately Day 22 before increasim~ ac;ain progress~.vely until the end of 2C the cycle. The midsecri:tory drag ir~ the intensity of IL-1RII
luminal immunostaining ( (Jays 1.9-~'_2 ) was statistically significant a;> compare~a i~o lat.e ~>rc;liferative/early secretory (Days 9-18; P < 0.05) anc~ late secretor<< (Days 23-28; P <
0 . O1 ) immunost:aini.ng 1e ~ik.=.,ls .
Representative examples of IL-1RII immunostaining in the endomet:rium thrc.w.agticut= thc:~ menst:x ual cycle are shown in Figure 10 f:or Days ~~; (Fig. 10c~) , 1:.3 ~ Fig. 10D) , 18 (Fig.
10E) , and 23 I Fig. LOf, ivot~e t.l~e reduction ~.~f IL-1RI:I
luminal secretion in tf-.e glands and surface epithelium of specimens at L>ay 18 as c:c>mpared to Days 13 anc:~ 23.
Western Blot ~~nalysis O1. X5409-5 To :furthe:r f::~x~~mine =II:~-1F;IT pr~:~t:ein expression throughout truf~ menstr~:,al cycle, total e.nc:~ometrial proteins were extracted and eq~_,iv.alent amounts wtrre subjected to Western blot analysis . ~:W r resi.rlt s stuowed that the antibody '.~ specifically =~ecognizt=d several major and minor bands (Fig.
12 ) . From the:>e, the c=a3 -~ arid 4 5-- kLn,~ barld~; are consistent with the commonly ,-eported rtiolecular weights of t:he membrane--bound and the soluble receptors, respectively. The immunoreactive bands were absent when t rue primary mouse 1C1 monoclonal anti-IL-1RI- antibody was re~:o.aced by an equal concentration of norma'~ mouse IgCs (Fig. 12A). Low molecular weight bands ;<45 kDa) :aisappeaz:~ed when t:he antibody wa;>
preabsorbed w=_th an excess of recombinant IL-1RII (20 ~g/ml) before being incubatecr with nitrrocel.:Lu:Lcase membrane-1~~ transferred proteins, whereas the int=en si.ty of major higher-molecular-weight bands was considerably reduced (Fig. 12B), suggesting specific int:r~ract:ion with r_hEe anti-IL-1RII
antibody. As shown in figure 12A, al:L IL-1RII bands revealed by the antibody were rr,-:~o kE~dl y intense at the a3pp:roach of 2C ovulation. Tree intensity of these band:; clearly decrea~~ed during the midsecretory phase but increased again thereafter in tissues from late sec~ret.ory endometr_:ia.
In Situ Hybridization Expression c~f:- LL-1RII rnRIVP., in the endometrium was 25 first studied by in situ hybridization to localize the site of synthesis. Figure L:~ shows the appearance of endomet.rial glands and stroma at 1 ~:~7X (A1, B:L, and C".l ) and 666X (A2, B2, and C2) magnification" 1=ollowing hybridization and staining with propidium iodine i i.at:e proliferat=v~ve endometrial tissue, 30 Day 13). The hybridize;:ion signal (green-yel:Lowj could only be visualized at .highe:r m<~gn=_fic.-~ti.on ;1665X) , as illustrated in the same figure, an::I wa:> more pronc~~anced i.n glandular (A3) and surface (33) epitluelia:i than ~.n stromal (C3) cells. No hybridizatioru was observed in negative ,-~ont_rols including the omission of ~~iotiny:La! e::I DNA pre~bes prepared from the p:Lasmid containing Ih-1RI:I cDl'ia insert= or the use of biotinylated DNA
probes obtained from tire plasmic.~ alone.
Reverse Transcription -F'clyrnerase C'.hain :~Zeact-i.on Expression e:~f IL-1R.I:I mF;NA throughout the menstrual cycle was studied by ~:e~niquantitative R'L-PCR. This was achieved by r::ormal.izir g t.hc~ Ih- LRI I mRNr'~ to the mRNA of the coamplified rousekeep:i n~:a gene GAPI>H and x:~y including an equal amount of thE:~ sane pre:~p.~~ration of posit'_ve control (RT
preparation of cDNA f.:~rrc human e.~clometrLal RNA) in every series of amplifi.catir~n=;. The c-orutro:L, which was subjected to the same e:~periment:;~_c~ondit:LOns from the amplificat:ion reaction until Sc>uthe c n blot ana Lytsis and autoradiography, allowed for rru~nit:orina ~_>f i:he ir:terassay vari.at.ion. Results were expressed as a pE=~~~enfiage of the cont.roL. value (:i.e., the amount of- IL-~1RII rn~~l.~1 rela t ive to ,ruat cf the corresponding GAPDH dl video :by the amount: of IL-1RII mRNA
relative to that of Gr~~'L~H ~.n 1_he control ~: :L00) . Result=s from 20 endorru~trial bio->sies across the menstrual cycle (Fig.
14A) show thaw: IL-IRII rruRNA level; are i_ow in early proliferative endomet.~ :i_.~~1 t:issues and foll.oca a kinetics of expression comparable t _~ that f_ovarud for the protein by immunohistochf~mi~stry ~:Luminal secretion) and Western blot analysis. Ir:c fact, ml.l~.~~ levels were elevated in late proliferative~, early ~~:ret-:ory, and lat~.a secretory endometria, x:~ut they i_.~nifiicantl~~ decreased in tissues from the midsecret.ory phases. Representative examples of RTPCR and Southern blot: analyse: ~f IL-lRI:I mRNA in tissues from different cyc~:Le phase:- are shown i.n F:igure :L4B. The RT--PCR
analysis of I:L-1RII mF:N,%~ irn separated ~.~ells .=_showing higher levels of expression n epithe.l.i.a1 than in :>trornal cells confirmed the in situ h,,~~bridizatiorl dat_~ (F:ig. 14C) .
Discussion Hurm~n endomt,_t'ium i s an act:iv~~ site of cytokine production and action. Complex interac'i.ons of epithelial, stromal, endc:vhel.ial, ::end lymphoid ce.Ll:~ occurring in the human endomet:rium as ~~e:ll_ ~~s dy:~amic change:> orchestrated in cyclic events of ce:l1 l:~:roliferatic>n, cii~i-erentiation, and shedding requ.:ire a wel:l_-elaborated netws:~rk of intercellular communicatiorv~ signa:Ls :~uc;h as cytokines. Many of these events are reminiscent .~f t=hose a~:sociated with the inflammatory and the reyarative processes, wrich make plausible the involvement of proir~flarrL-natory cytokines.
Furthermore, cvytekine:_ ~-such as IL- l, wh.~ch have been 1_'~ considered ar: im~~ort~an~t immune rnecliator_s, ai.so seem to be implicated in critica:. zepr_oductive ev:~nt:s such as ovulation and implantation, and E~~.~i.denc:e inclicat~=> that they act as endocrine and local rEc_~~lat=ors <f many cendornetrial functions (Tabibzadeh . 1991; ~imon C. et. al, 19':)fb) .
The control ~:_~r cytokine action in the endometrium may require the local av~ailabi:lity of specific regulatory mechanisms. l,or Ih-l, this i s illu s-tra.~.ed k>y local expression of IL-lra, ~ specific inhibitor t=hat binds to functional IL,--1 recept or type I ( i . a. , ~L,-1RI ) and that 2_'> blocks IL-1 binding ar.d cell sic~nal_i:ng ~;Boraschi D. et al, 1996) . Our present st iz;~y revea=:.ec3 t:n~s wxpression in the endometrium c:_ another irat~.zrai _~nr.ibit=o:_r for_ IL- I, the decoy IL-1RII, which acts di i: ferently, ~~y se~~uestrating active as well as inactive IL-I~~ a;zd, thereby, restricting the availability of the ligand for_ the fun~~t~ional receptor and inhibiting even its m~:t~~ar.ation (~ymons :.A. et al, 1995;.

Immunoreact:, ~;~~ IL-1.RII ~,ras localized throughout endometrial ':issue bot:r, in epit=~e=:_ial and stromal compartments, buts it ~.,,~;~ more oi;>v~..ous in endometrial glands and surface epitheliuru. This w~~s cionfirmed >';y ~.n situ ') hybridization showing ~G_.gh :Levels c:f IL-iRII mRhA in endometrial elands anc.; :,urface epithelium. The RT-PCR of IL-1RII mRNA expression a s eparatt. c;landu:iar e~~ithelial a::~d fibroblast-l:i_ke wells i. solated from endomet:rial tissue corroborated these fit~d~r:gs. Inrmunolocalizat:ion. further revealed two levels or staining. The first, which was located all a:rrour,d thc: cells, may correspond to the cellular membranebound IL-1RII receptor. ",he second, which was more intense and located par..~dom:inantly within the lumen of endometrial cilands anci at T_he~ apic:a_L side o:f surface epithelial cE:lls, is r r~.t probable% a lumina:i secretion corresponding to the :~c,:ub:le and .~ecretec~ form of the receptor (sIh-1RII) . Ir: fact:, rile:>tern ~.lc>t analysis of IL-1RII protein in endomc:tria:l tissue shows the presence o:f a 68-kDa band, which ~~oTvr_a>sponds ~.o the membrane-bound IL-1RII
(Colotta F. et a~., 190.4; Bora.schi D. et a1_, 1996), and <~
number of lower molece~~l,_~r weic~h~ >;>ands, which may correspond to different soluble i-orms. It is well known that sIL-:1RII
is released from the ; ernbrane-bc=>und receptor following proteolysis, and that the released solub:l.e mc;lecules keep their ability to bind anc~ neutralize IL-l, particularly the circulating form (IL- ~3i (~::«loti_a F.. et al., 1.994; Symons J.A.
et al, 1995; Bos:~u P . et a:1, 1995; Orla:zdo ,~. et al, 19'97) .
To our knowledge, a f:..i1_~ charactez:i:zation of such molecules has not been performed, but it c,:ould be proposed that other 3() forms of soluble IL-li' I: ~ could ex_l st, p.:~ssibl_y due to the presence of diffE:rent, uniderati:fied cleavage sites. Neumann et al (Neumann D. et ~cl., 2000) re~.>crted a 3~4-kDa soluble form of IL-1RII in the cult!are medium of IL-1RII-t:ransfected keratinocytes. ~~ccoroli~og to Orlando et .7_ (Orlando S. et al, 1997), lipopolysaccha-i.cze-treated monecyt=es release a 60-kDa IL-1-binding molecule; wrni~:h was u.dentified as the IL-1 decoy receptor.
Both cellul<~r and extracellular/luminal IL-1RII
immunoreactivity vari:~~ within the menst:rua.1 cycle and followed a rE:gulated ,c.yc:l.e phase-dependent pattern. In stromal cells>, the int-:=nsity of. staining remained weak to absent durincr the who:.e proli.ferat:ive phase but increased moderately after ovula:~tion. In epithet ial. cells, the regulation ap,~eared tc:: be diffe:eer~t, as immunoreactive cellular IL-1.:RII firs,- increased mode.ratel.y after Day 8 in the midproliw4rat:ive K 'ease, then nuore markedly after Day 21 in the secret.ory phasre. The IL-1RIT luminal secretion followed a mc:>re c:ompl~:x ~:inetics. Being undetectable until Day 8 of the ~~ycle, l:aminal secret=ion reached a maximum at the end of true pr olife:.rative phase, there dec~l_ined progressively until Day 22 before undertaking a final, gradual augme~ztation ~_lu:ring the late secretary phase.
The pat:te.rn of IL-1_RI:I i_mmunost:aining in endometrial epithelia cells was quite unusual, and the temporal exprn.=_ssi.on f~~:r th:is natural inizibitor of IL-1 :is rather intere~;sting. 'That the receptor expression is down-regulated in 'the rnidsEe~retory pha:~e, especially during l~he implantation window, rind up-regulated a'~ the end of the menstrual cycle sugge~;~s that T1.-n.RTI may have multiple functions in human endo:metrium. Tr~estern blot analysis appears to cc:~nfizm them results of immunohi.stc>chemical staining showing that er~dornetrial epithelial cells possess low amounts o.f IL-1RI~ in the midluteal phase. Epithelial IL-1RII expre.;ssion and distr_i.butic;n in endornetrial tissue across the menstrual <:~yc:le is remarkable for several reasons.

First, IL-lFzl~ appeared to be expressed and released precl~~minantly l~~y g~_andul~~r and .lum:inal epithelial cells rather than by : :i oma'~ cel 1.;. Thi> is of additional interest, becv~use fir.~.Y inter act:icrzs between the embryo and the endometr_i.vm occur at the levee of the luminal epithelial cells during the adhe:> ion process .
Second, the significant decli.ze in IL-1RII lurninal staining, which s ta:rte~,~ .at the bectinni_n~ of the secreto:ry phase and rea~~hed its rn.:inimr.rrri at approximately Day 22, .in the midsecretory ~~hase, i:- ~ ather ir~tE~resting arud suggests 1=hat IL-1RII secretion is u;jected to subtle chronological regulation ir. endometv~ i~l epit~helia~ ce Ll.s. This is remarkable, x~c~cause ti;e ;phase of madur_.ac:~ sec.ret:ion correlates with a putative "impl<~ntat~_on window" t~nought to exist within l.'> the midsecretc~ry phase 1_~etween Lays 18 <,rnd 22 (Tabibzadeh S. , 1998) . It is importa:~': to point: out= t~,=rt: the physiologic basis for thi:~ window has not yet been a.:learly established.
Also, no gene ral agrea::rn~~nt exists s regarc:~ing the dates o1= the endometrial receptivit ~~ pe:ricd c>r the implantation window during normal met=stru~.1_ cycJles (Ta.b.ibzadeh S., 1998) .
According to :~imon et al (;~~imon ~. et a.~_, 1998a) , the implantation process t:~.rts at Day LH i ':, which is consistent wii=h IL-1R_I::C decreased expression. According to Bergh and Navot (Bergs f'..?~. et: al, 1992 ~ , first embryonic 2'i signal detect: ion (pre~:urned window of _mplant:anon) extends between Days a?0 and 2~I. However, ot:ners have suggested that the implantation windc.~w .is confined to postovulatory Days 5-7 (Psychoyos A. ,, 1993) . l~onetheless, tYie decz:eased IL-1RI=I
expression at: that spEC:;i:fic time cE the cycle, at which embryonic attachment Grn~ imp.lantatiom rnay occur, is suggestive of: a pos sil.~ _e role for I:L-FtI I in the initial interactions between rnw~erria.l. and embryonic cells and the establishment: of an erdc:>>metrial period of receptivity.

8'7 8.5409-5 In :fact, num-:r .pus studies: indicate that IL-1 acts as an embryonic signal and is sec:ret:ed early by embryonic cells (Barano R.I. et ::~:~, 1.992; :;teeth K.~J. et al, 1991) .
Interleukin 1 appears tc: be crucial fc;x successful implantation, because t-.ht, blackacie of i_ts functional receptor (i.e., IL-1RI) in viva L::~r~events i.mp:lantation x.oy interfering with embryoni~~ attachm~;::nt: ~;.vimon f.. Eat al, 1994; Simon f, et al, 1998c) . InterestirAc_~ iy, t: he IL-1 ~-y:,tem appears to mediate the regulation o~= int.egr:Ln expressicn in human endometrium. Simon et ~~-~ have reported that binding of embryonic IL-l.a and IL-1(3 tc endometrial epithelial IL-1RI
up-regulates endometrial epithel.:ial ~3~ ::ubunit, which is considered to be a marker of endomet.r_ial receptivity (Lessey B.A. et al, 1992), anu :_acilitat:es adOesion of human lf~ blastoeyst (Si_mon C. cat: a:1, 199~'i . Serl.zm IL-L levels increase at o~rulation ~;c::anruon C1.C. eW a_., 1~>8S) ; the down-regulation of IL-1RTI 7:~c:~:lease in endc~mf,t.rial_ glands and surface epithelium suggests, therefore, the existence of accurate regu_~ata.r_y mec:hani.sms ttzat ai.Lf~viate IL-1 inhibition during this crucial per:i~d, thereby fa~;_'~litating IL-1 proimplantati.on actiors. Interestingly, our previous studies have shown that the ex:p:ression of :IL-li~I, the functional and signaling receptor .for I:h-~w, is modE:ratc:~ly up-regulated during the same perioc~~. ;Bic~onnesse H. at al, 2001a) . This a remarkable example of fine-tuned and weC_1_-orcteestrated endometrial events, true object of which is to preserve the reproductive function of this tissue.
The: dec:.rease~c:i expression of IL-1R=LI: protein, which was more perc:~~pti_ble <at the level of luminal secretion by immunohistoch~amistry, was detected by Western blot analysis and semiquantitative F<.'T-PCR. This makes an inhibition at the level of IL-1RII mRNA more likely than a trar:slational or posttranslational prot~~c::'_ ysis-depe:udent mechanism. Whether this is due to inl.ibit i.<:~:of TL-:~.R~I gene transcription or to decreased IL-:1RII mRNA st abil ity r~,rnai.r,s to be further elucidated. However, ~::r~is require,=. the identification of the regulatory mechanisms .rncaerly.ing Ih-LR;.I downregulation.
Third, IL-1R f :I expressi.ori l n endomet~rial tissue markedly increased in t:.Poe late scc:~et:c;ry phase of the menstrual cycle, loth ,:~t: t: toe lev~;~1 of= t:he protein and of the mRNA. This wa.s quite ::~i::~~T~.ous ire tl-re g7..ands and l.uminal epithelium, b;:a it was <=i i_so v.isibla i.r the st:.~-oma of some late secretory-phase bi.~,.>l.~sy speci.menso This may have a considerable ~cignificanue, because in t:he absence of implantation, the endontetria~i tissue unc~ergoe:~ a process of cell necrosis and disiut.~gration at the end of the menstrual cycle (Tabibzadeh S. , 1':39:1; . The a ~ evated exp:ressi.on of IL-1RII observed in the 1.:-rte secretory phase may, therefore, play a key ro~.e in the control c>F sur_rL an inflammatory-like process durir_g the premenstrual and men:; t.rua?. periods .
The potential involvement of IL-1RII in implantation and regulation of local in~~_amrnatory-like processes is supporter ky our recent data showing a marked deficiency in the expression of this specific inhibitor of IL-1 in the endometrium :~f women wit: endometriosis, especially those suffEr.ing from ir.fertii.ity (Akoum A. et. al, 2'_i 2001b).
Anoi_her natural inhibitor for IL-1, IL-lra, has been found ir_ the endc~mctrium, but more in the proliferative than in the secretory phase of the menstrual cycle.
Furthermore, accordincf t:o Simon et al (~imon C. et al, 1995aj , endorcretri.al cells rather express the icIL-lra. In view of the l.~~ca7.izat_.:~r and thr, t:ernpor.:~:L pattern of expression o: IL-~1RII revealed i.n our sturdy, this is quite interesting, and It sl:~c~c~ests that. IL~- :LF:IT anrl IL-lra have complementary roles ire t: ne control c:~f _;:L;--1 a.ci"ion in endometrial tissues, aon:7 1~l:at: t=r.cey exert their inhibitory effects at different a:;ud complementary levels. This is all the more pLausib:le bec:<:;~vse, accor_di.ng to Symo:ns et al (Symons J.A. et al, 2~a95) , IL-.l.l<:CI does uct: interfere with IL-lra mediated inhibition of CL-1, and the Two mol.ecules have an additive effect in inr .i..l.-~i.ting the bi.nd:ing of IL-1(3 to cell-surface IL-1 ~°eceptors .
1C) The process wlr.ereby IT.~-1R7::I expression is regulated in the endomei~riu.m is ynt estab..isue~:i, c:~nd it remains unknown at present whf~ther estr<l:~iol, progesterone, or other hormones of the reproductive c,yo,l a c:-an di rectly or indirectly affect IL-1RII expression in e:ndometrial tissue. On the other hand, it is quite possible that proinflammatory cytokines such as IL-1 and tumor necrosis factor a cTNFa), which have been found to be p.redominar~t in the endometr-ium during the late secretory phase (Kauma ... et al, 1990; Iabibzadeh S. et al, 1995 ) , can be: involveel n the u~.>-regulation of IL-1RII
2() expression. Tn fact, bath IL-1 and TNFcx have been shown to up-regulate I:L-1RII e:;pression or release in other types of cells (Shlopo v B.V. e!: a1, 2000; Yu P.W. et al, 1997; Plata-Salaman C.R. et al, 1;9'7) .
In summary, IL-1RII expression in the endometrium is an interesting and dynamic process. It is tempting to hypothesize that aber~:ar~.t expression cf such a natural inhibitor of IL-:C may >;;c.~ related or assoc:iatE:d with pathologic states or infertility. Additionally, IL-1RII
appears to be predict,.3b1 e, based on tre time in the menstrual cycle, thereby a:l.lowi__ng diagnosti~.~ use of IL-~1RII as a stage-specific marker .in this. tissue. Finally, an understanding of the function of IL-1RII throughout the menstrual cycle and the means to regulate it.s expression ma~.~ prove to be of future therapE>utic usF .

The establis.ument c:f a new vascular supply is essential for the surv:.~~a1 of endomet:rial. tissue and it~>
development in ectopi~: :Locat_LOns. We have previously shown that ectopic endometr~.-r:L ~~~a.l.:Ls r-elea~e an irtrportant mitogenic '. activity for human enci>thelial cel:Ls and identified macrophage mic)ration :.nh Lb:i.tory ract:oc MIF) ~s one of_ t:he principal bioactive mcle~~ule:~ involved .n endothelial cell proliferation., Irl thcE,>ret;ent: study, immunohistochemical and dual immunofluorescencc: analyses showed that MIF is effectively e;~pressed b:~ endomet.ri.ot:i~~ tissue, particularly in the glands,. and idErntified endothelial cells, macrophages, and T lymphoc~,rtes as ce.:L.ls markedly exp:r.~essing MIF in the stroma. Western blot analysis showed a single 12.5-kDa band corresponding to the ~:rn.7wn m~J,'~ wt of_ the molecule. The highest concentration: ~~~f MIf protein in endometriotic tissue, as measured b~. ELI~A, werEf=oun~:~ in flame-like red endometriotic lesions, ~vompared with typical black-bluish (P
< 0.01) or_ wivh white :Lesions (~ <:: 0.01, . interestingly, MIF
displayed a r:~arked ex~~ress:~on in 1_esions from the initial stage of endometriosi-; !;stage I;~ . Semic~uanti.tative R'r-)?CR
analysis of MIF mRNA ~.e~rels in the same E:ndometriotic tissues showed a patt.~=rn of el::p.ression c.onnparable with that of the protein. In view of '~s potent proinflammatory and angiogenic prv~pez:~ties, .ocal prc:~duction of MI:F within endometrial implants, particularly in those that are highly vascularized and repr~:sent:ing tile earliest: and most active forms of the disease, make plausible the involvement of this factor in the local ir.~mt:mo:inflarnmatory process observed in endometriosis and the irvitial steps of endomGtriotic tissue growth and development:.
Enc~omet.rios i.s, one of the commonest gynecological conditions, is definec:l as the pre:>ence of ti~,sue histologically si.mila:v t:c endometri.um. at sites outside the uterine cavit.v. This ::~:isease is intrigizi.ng and unique in that it is them on:l_y kr,::~w:a beni_gr3 d i.sea5c= :in which autologous cells, accord= ng t:o tre : widely a<;cepted transplantation theory (Sampson J. , 1~?'i ) , can im~:vl.ant <~nd develop in ec:topic _'> locations. The etiolc,gy of endometr:iosa_s is still not clearly defined, c:;ernet;in~ or edispos:it:io::z, environmental toxins, hormonal factors, and immune deficiency may contribute tcthe susc~ei~~t ibi.lity o.f a woman to develop t=his disease (McLa-=en J. , L.U~J~..~ ) . HowEwer, a key condition for endometrial tissue 1..o svnrv_ve and grow f-yc:topically following successful adhesion arid implantation is the establishment of an effective new b1_ooci :~upz~ay, a ~>rocess involving the generation of new blo<.~d vessels or angiogenesis (Healy D.L.
et al, 1998).
Ear:Ly and mc,st: act ive endomet ri.oti_c: 7_esions are markedly vascularized; increased vascularization is seen at the implant surface arud also in the surrounding peritoneal tissue (Wiegeerinck M.~~,. et al., 19931 su~ygesti.ng that endometriotic implant is capable c;f induci.ng its own neovasculariz~~tion by c~triving _ioc:.a_L mi-.:rovasculature. It is therefore of ~~ crucia::. interest tc> elucidate the mechanisms underlying en~~omE~trio:-s.i~~-associated angiogenesis and to identify the factors vevolved in t:.hat process. We have recently found that iru:nuc_>rtal.i.zed a:~s wel.l as L:rimary e~ztopic endometrial epithelia::. cells release an .important mitogenic activity for human en~;.c>t.helia.l ~-e:'~._ls and identified macrophage migration _inhibitory factor (MIF) as an important mediator of endot.heli~~.l cell prolu.ferat.ion (Yang Y. et ,al, 2000). Originally de:,cribed as a product of activated T
3~~ lymphocytes that inhit,a.i.t:ed the random migration of cultured macrophages, MIF is nc>w known as ~~n important. modulator for a variety of cell funct::.or:.s, includ~i_ng inflammatory and immune responses (Metz C..N. at al, 199?; Nishihira ..1., 2000) . The identification of MI:F ~as a mitogenic facvtor for endothelial cells released by ectoy=i_ c endometr ~a1 cells i.s consist:ent with recent d~~ta show:i:;o.;~r_ _impc>rtar,i~ :~c~l.e for MIF
in tumor growth-associated ang.i canes is 1::z vi.v<_~ and i n vi tro ~ >r~ in autocrine regulation cI: c?ndothelvai ce:L:.' prc>l iferation (Chesney J. et al, 199r~;, (~gawa H. et al, 2000; Shimizu T.
et al, 1999) . The presen-t: st~t~dy was undertaken to assess t:he expression of MIF in e:ur~ ornetriot:~c les:ic>ns, identify the sites) of expression, r. nd :investigate whether such an 1C~ expression varies acct; r~:l ing tc> t. he stage and the activity of the disease.
Materials and Methods Source and t~r:c~.l_i!~g of tissue. Endornetriotic t:issue specimens used in this :study were cabt<~ined f:rom women who l~~ provided informed cons~..nt~ f:or a resea:rct~ protocol approved by Saint-Fran~:oi:> d'Assis:-: ~-tosp:itai Eth:ic;s Committee on Human Research. These patient.:: were found to have endometriosis during laparoscopy or :i.:~paYvotomy, had nc~ endo.netrial hyperplasi.a c>r neopl_as i.~, and had no- rf=ceived arty 20 antiinflammat:ory or hcrrr,onal medic<~t:ic>n duri_ny a period of at least 3 months before smrgery. Enc~ome~rloSl.s was staged according to the reviseI Ame:ri_c:an Fert:.i:'.ity Society classification system ;i'~nneri_can Fert_ii:ity Society, 1985) .
The cycle phase (prol~.I_c_rati~aEe or secrf=t.ory) was determined 2~~ according to l..he patie,nr ~' cycle h isto:ry and to the serum progesterone. The mearu age was 34.6 pl~~:; or :minus 6.0 yr.
Endometriotic: biopsies= were immediately placed at 4°C in sterile HBSS (Life Tec:tur~ologies, Inc., Burlington, Ontario, Canada) containing IOC IU/m:1 penici_I_lin, 100 ~g/ml 30 streptomycin, and 0.2')..a~/ml. am~7hotericin tzansported to the laboratory at: whi~~h tr c~.;, were= immediate:':_y washed in HBS;3 at 4°C, snap frozen on dr~~r icc:, or fixed in 10, formalin.

For immunoh~.~rochemica~l analy:>is, 25 endometriotic biopsies were includec:. F.i E peen wc~x~e fa czer; at -80°C in Tissue-Tek OCR, ccmpour:rl (M:i.le:; Irlc . , ~,.lr;:hart:, IN) , and 7_0 were embedded in par_af r:in. '>?hese h:iop:~:.es were from women ~' with endometr__osis stac:~E~ L ( two typical black-blue, two red flame-like) , :stage II ;~::wo tvY~pic<~1 bl~r~.r:-blue, three white, and one from the inner w,al:L of o~Tari<~n endometrioma) , st:ages III-IV (three typical 1:>:L~~Jk-blue, three white, and nine from the inner wal.=_ of ovax::i..~;~ endometrioma) .
For Western b_1.~:~tt i.ng, EI I~>~, and RT-PCR analy~;es, 24 biopsies were taker:. These bs.opsi_es were from women with endometriosis stage I vw~~ typic:a.l b:Lac;k-blue, seven red flame-like, and three wl.i-te) , st:age II ;f: our typical black-blue, two red flame-1 ~ .kE=, and three whiff=e) , and stages III-IV
(two red flame-like arch :one whit:e) . BioL:as ie;; were all snap frozen and kept at; -8C"~:' in microcentrituge tubes (Eppendorf, Gordon Technologies Irvc., M:isss.ssauga, c)ntario, Canada) until used.
Immunohistoci'zra!nistry. Five-micrometer cryosect:ions of Optimal Cut=ting Terr~.~erat ure-frc.zcen ~~ridometri.otic lesions were mounted on poly-_i--lysin'=-coated mic,rosc:ope glass s7_ides, fixed during <?0 min ir.,~ 10~ buffered forma7_in phosphate solution ( Fisher Scier t:: i. f ia:, Mon'~~rea l, Quebec, Canada ) , and washed in PBS . Five-rr~.i_'v.rometer se'~t:i cn:.> of paraffin-embedded 2_'i tissues were mounted on poly-1-~yss_ne-crated microscope glass slides, depar<~ffinizec:l ir. t:oluene, rehy<~rated through graded solutions of ethanol Gn~:~ water_, anal wasrned in PBS. The subsequent steps were ~._:v,e same f-or- c:ryosections and paraffin-embedded tissue secti<:au=:. Briefly, aftfer permeabilization with Triton-X--100~rM (1''; in P~.S) anc:1 elimination of endogenous peroxidase wii~h H,OZ (n. ~'o in absolute methancl) , tissue sections were success:) ~aJl.y ~_ncuLvated at morn temperature for 95 x.5409-5 90 min with a goat pol .~~:::1 oval am t~ human--MIF antibody (RtxD
Systems, Minneapolis, T~~l'-J) [0.6'6 ~c~/n~1 in PB:>/0.2% BSA/O.Olo Tween 20TM (PF3S/BSA/Twe~e:r~.) ] , 90 min with a biotin-conjugated rabbit antigoat antibc:dy (Jackson ImmunoResearch Laboratories, Inc. , WF~st ~.~rove, PA) ( 1: '.~00 irn PBS/ BSA/Tween, 45 min with peroxidasE~-r~onjugate~_1 strep~~avidin (Jackson ImmunoResearclu Labor_ato:ries, Inc. ) ( 1: St)0 dilution in PBS/BSA/Tween) and 20 min diami.nobenzi d~~ne used as chrornogen ( 3 mg diamincbenzidinc--! n . 0;3 °o Hy:>~ in PBS ) . Sections were counterstaine<:~ with hE_rn<rto~;ylin arid moarit:ed in Mowiol~~M
(Calbiochem-Novab;~ocharn :~orp., L,a Joila, CA). Sections incubated with goat Ic..t:~;~. ,at the same ~~o.~cent:r_ation as the primary antibody were u,-;ed as negative c:ontro.ls in all experiments. Slides wc::re ~~j:ieweci using <., microscope 1_'i (mikroskopie and systc me ~SmbH, ro,:~de.l_ DI~IRB, Leica Corp. , Postfach, Wet zlar, Germany) and phot~~~grrzphed with 100 A:>A
film (Eastman Kodak Cc~. , Rochester, N~) .
Dua_1 immunoi .ir~orescent ~ taini.og. Cryostat and paraffin-embedded tis:_uE_- sf=ctiorns were tr-eat:ed and incubated at room temperature for 120 min with a Boat polyclonal antihuman-MIF antibod~~ tR&D Syst.ertcs, Mimr':eapolis, MN) at. 0.66 ug/m1 in PBS/f3SA/Tweer <~s ciescrv~ bed ~~ar.~_ier ( see the Immunohistochemistry m.ethodo.logy> . Att~~r a PBS/0.05% Tween 20 rinse, seci~ions were incubated at. r_oc:~m temperature for 90 2_'i min with one of the fol_.l,~wing antibodies: mouse monoclonal antihuman-CD6f3 (DAK(:> r:.'orp. Diagrnos tics (:anada Inc. , Mississauga, Ontario, ~:,inac~a) (d:iLuted .:50 in PBS/BSA/Tween) to detect mar.rophages; ur-.~use monoc.:Lon.a1 antihuman-CD3 (diluted 1:100 in PBS/~3~A,/Tween;to deaf-:ct T lymphocytes; and rabbit polyclonal antitnuman von Will_ebrand factor (vWF) (Sigma-Aldrich Cor_p. C:,a-~a~:~a :LTD, C.~akvi 1 .''Le, C)ntario, Canada) (dilut:ed 1:200 in. PBS/I3SA/'i'ween) to det~-?c;t endothelial cells.
After a subsequent wa~.:Cz in PBS/0. CS <'s 'Iwc-yen 20, tissue sections werF incubated simultaneous.Ly for 60 min at room temperature in the daz K with flt~oresceir~ isothiocyanate--conjugated donkey ant: gc at 3ni~ik:ody (,1a ;kson ImmunoResearch Laboratories, Inc. ) (cliZuted L: 50 in PBS!BSA/Tween) and rhodarnine-coryjugated :,t~c=ep antirnot~se an-~i.body (Roche Diagnostics, lava:L, Ql!.c~=t:;ec, ~'anada) (di Luted 1: 10 in PBS/BSA/TweerA) for ti,sues marked for MLf arnd CD68 or CD3 or rhodamine-conjugated rnoc.zse antiralr:bit anl~.ibody (Jackson ImmunoResearc.:h Laborat cries, :Cnc. ) (dilated L: 50 in PBS/BSA/Tweer:~) for ti~;:~c.ies marked for ~ICF arud vWF. After a final wash irv. PB~:/O.Oc;o 'Tween 2C, slide; were mounted in Mowiol containing loo para-phenylenediamine (Sigma-Aldr_ich Corp. Canada ::.~td. ) , ar antlfading agent, anc~ observed under the microsco~.~E~ (heica i;~rp. ) equiF~ped f:or fluorescence with a 1'.> 100-W UV lam~:~. Photom~ crogr_aphs were taken with 400 ASA film ( Koda k ) .
We~~tern blot !..:.ng. Protei.n extraction from endometriotic tit:sue u;~s perforrrceci according to our previously de:~cri.bed ~~r~~cedure (Bzgonne>se F. et al, 2001a), and total pro~~ein cone:ervtration was deternr~ined using the DC
protein assay° (Bi.o-Rac I~abor_at=or.ies htd., Mississauga, Ontario, Canada) . Fow '~lest~ern blc>t ana L~,~sis, denatured proteins were separated by SDS-PACE in L'>o acrylamide slab gels and tran:~ferred c,nt.o 0.95-iam nitro':-.ellulose membranes.
Equal protein loading ~,a.~s ;.onfirmed by :~tair:ing the membrane with Ponceau ;~ (20). Nitrocellulose membranes were then cut into strips and i.ncub~~t,~: 'd overnight at . °C with a polyc:Lonal goat antihuman MIF ant :it:odv (R&I>> :~ystern~ate 2 ug/ml of blocking soltavion (t).7 ~a 'Tris buffer, 0, ~)° NaCl_/0.05~ Tween 20 containinG 5o nonf<:~t= dr°~ mild: jwt~/vol] ) or with normal goat :Cg (R&D Systems) at the same concentration. Strips were then washed _n TBS-0.1'0 'fween 'Z0, incubated for 1 h at room temperature with a pez oxidase-conj ugateal rabbit; antigoai=

antibody (Jackson Immurn:~E~e.~earch Laboratories, Inc. ) , diluted 1: 10, 000 i.n the b:Loc:ka_rlc:1 sc::lut:ion, washed again in TBS-0. 1%
Tween 20, incubated fc:r 1 min with an enl-~anced chemiluminescence syste~rn upping BM chern:i.~ uminescence blotsting substrate (Roche Diagn~:>st~_ic:s), a:ld exposed tc> BioMax film (Kodak) for several tuna intervals a:L:l.owing f~~r_ an optimal detection (all bands wi:ibue but. not o~rerexposed) .
MIF ELISA. N F concent:rati.on i.n endometriotic tissue protein extract wa:~ measured by H~L,ISA according t:o a previously reported p.roc.edure ICalandra T, et al, 1995).
Briefly, this techni_qu~e ~~ses a c<:rpture mouse monoclonal antihuman-MIF antibodS-~ ( R YD :3yst:ems ) , a rabbit polycional antihuman-MIF anti_body- =or detect=ic>n, a~.kali.ne phosphatase-conjugated go~rt anti.ra:~~a~it IgGs (Cherni_con International Inc., l~~ Temecula, CA) and para;i:ltrophenyi phosphate as substrate (Sigma) . The optical c~~nsii::y wa.> me~asa~:ed ate 405 nm and MIF
concentrations were exta.~;~po1ated from <~ standard curve using recombinant human MI: F.
RT--~?CR. Tota'_ 1~NA was ext~rac.ted from endometriotic 2C~ tissue with TRIzo.L~lM rt:aqeni~ (L.ife Technologies, Inc. ) according to t:he manuf ~r;:t~.zrer' s v_nst.rucai ons . The GeneAmpTM
PCR core kit ;Perkin-F l_rrier C'orp. , Fosteir City, CA) was used to synthesize cDNA wit 1u X00 ng total. c:e~~ l.ular RNA and 2. 5 }rmol/liter random hexarncers in 2 0 ~i~ of a RT-P~~R solution ( 50 2~~ mmol/liter KC'__, 10 mmc:_i_!liter Tr-is-Hci.l_, 5 mrnol/liter MgC:l2, 1 mmol/ liter each of dM':~:,1~;;, >e~ ILi RNase :~nhi~:itor, and 50 IU
reverse transc:riptase) . The reaction was inc.:ubated at 25°C
for 15 min, 4a?°C for ~0 m:in, and 99°c:: for 5 m.in. For PC:R
analysis, we used 10~ ,>:C t=he reverse Transcription (RT) 30 reaction volume as temla_Late in a fina.L ~~olume of 50 1zL with 50 pmol of each MIF pr i_rner (:Forward p:rirner, 5' -CTCTCCGAGCTCAC;CCAGCAG- 3' (:~EQ ID I~cJ. : ~ ) ; reverse primer, 5' -CGCGTTCATGTCG'rAATAGTT- 3' (SEQ ID NC~. : 61 ; amplimer size, 255 bp) , 0.2 mmoi/lit.er dP'C(', ~~ru~°; 2. ~ iU T.ac~- DNA polymerase (QIAGEN, Santa Cl.ar:ita:,, CA. Ampl i.fic~t,ion was performed for 30 cycles composed of L min denaturation (at 94°C) , 1 min .'> annealing (at 60°C:) , < ic~i i min L:~ri-rner e~t:ens:ion (at 72°Cl .
These optimal conditicn~~ were determined by performing linearity tesi~s with ~~=.'.,, 10~, arid 20s ot~ the RT reaction volume and 2'; and 30 ~:rnplifi~~ati.or. c:.y;~le.. Amplification of genomic DNA w=ith thesF~ primers dig not ~or.cduce a signal, suggesting th<~t the arrp:l..i f icat:ion 5i te.s crossed at least= one intron/exon boundary. ~):Y the PCR volume, 20~ was fractionated by electrophoresis in :~ 1 . ~'a: agarc.se ge:l in the presence of ethidium browide and t :::<~nsfe:rred to a s~:i abrane~r'~ Nylon P7_us membrane (QIAGEN) . Tren the membrane was dehydrated at 37°C
1_'~ for 30 min, prehybrida.fed with a r:ybridlzation buffer composed of 5X SSC (0.~ mol/lit~er sod.irrm chloride and 0.015 mol/liter sodium cit=r~ ,:c-~ ) , 5X DE:.rlrardt:.' .:: solution, 50 mmol/liter NaH~POy, 0. ':« SDS, 20i) pg/ml salmon sperm DNA, and 50% formamide; hybridized wish ~''P-labeled MIf cDNA in the same buffer except Der:'ruardt' s sa:Luti.on; washed with SSC
solutions cont=aining C . :L'~ SDS, 11X SSC, 0 . 2X SSC and 0 . 1X
SSC, respectively, anc; E=:xposed t:o x-:ray t ilm (Eastman Kodak Co. ) for different t~irr=~ int:ervals al_:l_ow:i.ng fo.r an optimal detection (signals visil:~le but not ove.re?xposed) . As control, 2~~ glyceraldehyde phosph~~t:s~ c:~ehydrogena:~f= ;GAPL>H) amplification was used. For PCR ana:L~Y~s_i~s, we used :~'.5'~~ of the RT reaction volume as template i_n r f.:ir;a:L vc>:Lume o:E _'~0 p1 with 25 pmol of each primer ( j-orward ~ r:imer, 5' -':CGATGACATCAAGAAGGTGGTGAAG-3'(SEQ ID N0.::3); reverse primer, ~'-'TCCTTGGA~~GCCATGTGGGCCAT-3C~ 3' (SEQ iD N0. : 4) ; amplirner size, 290 bpj , 0.2 mmol/liter dNTPs, and 1 ~=U Vent L: VA polymerise. Amplification was performed f.or 30 c:yc:lc~~ c~f 30 sec: denarurati.o:n (at 95°C), 30 sec annealing (at 60°c :,. crud 1. m:i_n L>:r.i_rnez e~:t~ension (at ~') X35409-5 72°C) . ThesE_~ opt:.imal c°orrd:ition; were dc~t.ermined following linearity te:;ts using 1(W>>, 25° , 5C?° , and 75a of the RT
reaction volume. Spe~.vit~.city of t:he ampli.f:i_cation process was verified icy Sowr_ht~r_ra blot: hybr,idizat:i.on. A negative control (PCR .in the aL~sence of cD~.IA) as wel_:L as a positive control (cDNA preparat wc:.n prom t:hE~ huma;l hyst.iocytic ce:l1 line U937, krv«wn to st~~::nete MI)='i u,rere included in each series of MIF or GAPDH amp:lif ication. 'Tree intensity of the hybridization signals was a:~eterrr~ined by computer-assi sted densitometry, using QL,a.r~tit:y One Quanti~atic>r~ software (Bio-Rad Laboratories, Inc.t. a'he quantity «f the PCR products was determined by dens_i_tometric ar:,aly;~is of the intensity of the hybridizat=ion s:igr;al. The relat=ive ieve:l of MIF mRNA
normalized to GAfDH mF:IV~~ was ca a.cu.lated, and the results were 1-'i expressed as percent c: control (positi~.re~ control) .
Stav:istical ~-~r;alys is. Mult:.iplc-~ comparisons of MIF
protein concentration:-, as measured by ELISA, and mRNA
levels, as det~e.rrr.ined l.~y sem:iquant:itatiTre RT-PCR, in endometriotic lesions are-;:cording t:c t:h~J ~:.ype of the lesion or to endometric>:~is stagy were perf~.~rrned using one-way ANOVA and the Tukey's test. Corr:~~arison of twc> groups was performed using the unpaired t tc>;~t . ~?:11 an<al.ysc~;-> were performed using statistical analysis sy;~tern (SAS Institute, Inc., Cary, NC).
Differences were c:ons~ac~red as stati.st.:ic;ally significant: for 2~, a P value less> than 0. (i5.
Results The first ox; jc~c~.i.ve of ttli.s study was to asses>s the presence of MI=F i_rl end~::>n~et=r:iotic les~:i.on.:. Western blot analysis of prote.i.ns eL:i=racted f~~orn endometriotic tissue 3C using a goat polyclona.L anti.hurnan-MIh antibody showed a single specific l.'?.5-kL)a band corresponding t=o the known molecular weight of MIEN' (Fig. 1.'>; . RT-t'CR anc~ Southern blot analysis showed specie i_;:~ MIF t: r.anscr:ipy::, thereby confirming MIF expression in endcrn:~triotic tissue at the level of mRNA
(Fig. 16).
Immunohistoc:uern:ica:1 ar~alys:is of MI:F expression showed a specific brow:o:Lsh :irnmuruosta_Lnirag localized to specific compartments ,r i= endometricat=ic: t:issue. MIF was found to be strongl~~ expres~;~:e~ in glanrtula.r e~.>ithel.i.al cells and in cells scattered througluc.ut the st:roma (F'i.g. 1.~A) . Incubation of tissue sections witl-i normal goat -lgG: used at concentration equ~valew': t:o that of t:.he primary goat polyclonal anti-MiF ant: i_l~:~ody (negative control) ciid not result in any nonspecif:i_c: .immunostaini.nct (Fig. 17B) .
To identify ce:l_ls expressing MIF in the stroma, dual immunofluorescenc_~~:= arnalysis was performed using antibodies specific tc P~1.' F and to CD~i, CD68, and vWF.
Representative photomi~-roc~raphs exhik>it=ed in Fig. 18 shcw a marked expression of MIF in CD3-positive T lymphocytes, CD68-positive macrophages, zn;a vWF-po~;itlVE' endothE-elial cells .
To ~~:uantify '!II r' expres lion in endometriotic tissue and examine w:nether MI F' ~,~xpress.ic:>n c:crrelates with the type of endometriotic =t_esic,n and endometri.osis stacte, we measured MIF concentrations> in c~t:a:~l protf~i:u E~~:t..racts key F~LISA and determined the levels ~t mRNA in t':~e Name tis~~ues using a semiquantitat.ive RT-PC':~; ~.:;nal_ysis. A:> shown in Fig. 19A, the highest concentrations of MIF protai.n were fo~_rnd in flamelike red endometric~tic lesi ras, ~:~:ompare ~ wi.th typi<:al black bluish (P < 0.01) or with whir:e lesions (P <: Ci.01). Otherwise, no significant d:iffe:r-ence f~c:t:ween typical. and white endometriotic lesions ~.~~~as: found. Furthermore, MIF
concentrations appeared -c::o be significantly :higher in lesions from endometriosis sta~:;e I, compared with those from endometriosis stage II: (~> < C.05;i, whereas no significant difference berween st~.gE7s I and I~:I-IV ,~~ II and III-IV was noted. However, only t.r~ree biopsies fr~:~m endometri.osis stages III-It%' were :inc _Lude-:i in t:.h:i s a;>s~~y, which may have limited the power of ~:i~,atistical anal~~ses including these .'> stages (Fig. :L98) . ATra:Lys:LS of MLF ~ro'.c=:in expression according to i~he phase a;f the men~truaL cycle showed no statistically significant c~ifferen.ce between lesions from the proliferative and. the sf:~csretory pl~.asc~s.
Sem=_quantit~t:.:ive R'f-PC:".R ana.ly:ais of MIF mRNA 7_evels in the same endometric~::ic T: issues _;1-a~we<~ a pattern of expression corlparable with that of the protein, but the difference in MIF' mRNF L~svels was tc_>unc~ to be significant only between red and white endometriotic lesions (Fig. 20A).
On the other hand, ruo s:i.~~ni.f:icant~ c:~i.ffei:~ence in MIF mRNA
levels according t:o eru:viornet riosi.~ stage ( Fig. 20B) or to the menstrual cyc~_e phase ~:aa5 mated.
Discussion Endometriosi => rn:ight be a mul.ti fact.o:rial disea~ce, and its etiology :rema;_,°~s hypothet;ical. The presence of tissue structurally anal t=o some E-:xt:ent functionally comparable with the er:~iornetrium :f_oan<i ot,tside the uterine cavity suggests that ttue condition, at least peritoneal endometriosis, results. ::::nom r_mplant:at=.i.on of e:~fo:Liated endometrium following :retrograde menstruation (Sampson J., 1927 ) . Accordingly, t iut~ ect.opi.c grcwt:h and dc:~velopment of endometrial tissue Thai: i.s endowed w~~th the capability of adhering to and implan=.ing into the pei:itoneal tissue requires the :_tenesis c::f- rnew miczc>ves:~e7_~., a ~~-rocess called angiogenesis (McLaren ,.?-. , 2000; 1-fealy D. I . et al, 1998;
Folkman J., 1~~95) 1 . ::~;=,~;rEeral growttl f_ac:tors, :including acidic and basic fibroblast. g_:owth factors, platelet--derived endothelial cell growtkfactor, and vascular endothelial growth facto:~~ (Vf~GF) , r~~me the ,~~b i1 ity to sts.mul.ate vascular_ endothelial cell growil_ in vitro and in vivo (Gordon J.D. et a1, 1995) anct were fo~:r~,~ t<~ be rxL~resse~ by endometriotic lesions (McLaren J. e! ~:~J_, 1996; honnez ~:T. et al, 199;
Ferriani R.A. et al, 993; Fuj.imoto J. et al, 1999). Oar previous studies shove=d that endometriotic: :Lesicns express IL-8 (Akoum A. et al, %ii0lc), a chemokine having endothelial cell growth promoting a~:tivity .1.n v.ivo (Iloch A.E. et al, 1992). In an effort t::~ understand what enables endometrial cells to grow ect:opic~=ally in some women, we have previously assessed the ~~apabi:Lip y of enciornet:rioti.~ c:ell.s to release mitogenic activity foi endothelial. cells and identified MIF
as one of the major f~,.a::t:or s re:Leatsed by t:.hese cells in culture and having them ability to stimulate endothelial cell proliferation in vitr:i,fang Y. et al, 'UC)0) . In the present study, we found that ~~1IF' was effectively expressed by ectopic endometrial t::issue, b<~t:h at: the mF;~IA and protein levels, as assessed by R'I-PC:R anc:'Nestern blotting. Furthermore, we showed that NI:IF was lcv~:;:~te~~ in thEe glan«s as well as in cell aggregates sc:atte~red c ~.~~.>.r t~hc st:roma. Dual--immunofluorescence an~:lysis identified er:dothelial cells, macrophages, <~nd T :Lyrr,p'.nocytes as cells markedly expressing MIF in the st=soma. Fl:rthermore, MIF was found to be highly produced in the endomE~t:r.iot:.i~~ 7_eszons that were presenting 2'i noticeable vasculariz~tio:n and ,_eukocytce infiltration.
These findir:~:~s clearly c.emons~rate that MIF can be produced locally in erdt~metrioti~tissue, and that by different cel_:I_ types. Tn ~.Tif~w of it:s capability to stimulate endothelial cell growth shown in our and other previous 3C) studies (Yang Y. et a:~ , 200; C;hesney .:T. et al, 1999; Shimizu T. et al, 1999) and its faculty tc; actizurate and to inhibit macrophage migration (B.Loom 3. R, et a l., 1966; David J. R. , 1966) , it cou:Ld be su:.c~sstt~rl that MI:F lc_scal7.y produced within ~~03 85409-5 endometriotic: implant 1. ~raay contn:i~~ut:e to their growth and development either by stimazlating endt~tlielial cell proliferation or by ret:,.:.fining mamop:nages that have been ShOWn t0 SeCrE?te ~JECiF, ci pi:)tnrlt ~~r',C~1~:7~:~~?I?1C f:~3C'.tOr (MCLaI_"eW T.
_'> et al, 1996) , and numE=rt~~us growt:n :La~lt~~r-s for endometrial cells (Olive D.L. et: z~l, 1'91; ~talme 'T. et al, 1988) .
Moreover, the marked E-~l~~rE~s sign of MIF l:ay endothelial cells makes plausible than tPl:is i_a~~tor can s-ti.mulate endothelial cell proliferaticn viG autoc.rine and pa=~acrine mechanisms, amplifying thereby the- ang.i_ogeni.~,~~ process's .
It _~~s well Documented that: MIF is a major multifunctional proinf:-~amrnatory :~y~tokine. The molecule has been shown to be exprE-s;:~ed by inflanunatc>ry c:e.lls such as macrophages and lymphoc::yte~ and to stsimulate cytokine 1_'> production by these ce=~ls (Me~tz ~._.N. eve al, 1997; Nishihira J., 2000; Calandra T. <e:: al, 199<~; Bachc~r M. et al, 1996) .
Interestingly, it has l::o~en reporr_ec~ that MIFF overrides glucocorticoi.c~ inhibit::.~:;n c)f monocyte .secret:ion of TNF-a, IL-1f3, IL-6, and IL-8, wr i.c~h are i_mpcrtan~ inducers of immunological. and infW:~mmar:o.ry re:spc_mse:~ (Cal andra T. et: al, 1995) . MIF-st:imulat=ecz m:~c:rophages a:re :.r~ turn shown to secrete bioact:ive TNF-c:x and IL-iii (Caiad~wa T. at al, 199.'i) .
These data strongly sr.:gg~est. that f:IF ~:~~~uld funr_tion locally to amplify the in:f:lamma-tory :reaction mhserved in and around 2_'> endometriotic lesions, mairWy in those considered to represent the most actwae foams of t:he c~i_sease (Wiegerinck M.A. et al, 1993; Koko~~:ine i. et, a1_, 19~>'7; :>haw R.W., 1993;
blitz C.A. et al, 1997;, Endometriot i c; le:~i.ons can be c1_as~:ified according to their appearance and to their activit~~%. Tn fact, le:~ions of the peritoneal lin~:rng of= the pelvis leave various macroscopic appearancE=s, ~wl:ich rc~f:Le~:.~t their- .age and/or activity. The red su~t_:l.e lesions are more vascularized and have a higher epir_hel-i-i_1 rn~totic inctE~x t han the typical, puckered black or bluish per_iton~_~a1 :1_es:i_ons, whereas the vascularization and to_> m:Ltc>ti.c index <~re lower in the white ~~ lesions. Thu:~, red les:~:iont; are thoughi~ to c:o;rrespond to the first, active stage of _~~ray i.mplar~t.a~_LCn of: endometrial_ glands and str_oma and ~a~::~.~_Lc~i evol vre t.c~wa~-d the typical black or bluish les~_on after c~nc..l_osure beneath the peritoneal lining. The white ie: i_:.rn s, whic:n <~re be-~l.ieve~ to corre;~pond 1C~ to fibroti.c quiescent lc~.sic>ns, snow :Le~~ vas cularization and/or mitotic: activity.% anc:l reprc-.seni~ Less active forms of the disease (Wiegerinc>~: M.l~. et al, L99:3; Kokori:ne I. et: al, 1997; Nisolle M. et al, L993; Nisoll.e M. et al, 1997).
Int.erestingL,~,, cm.i.~ present s~~udy revealed that. MIF
1~~ expression is sig:nific~z~t:Ly higher i.n :reJd subtle than in typical blackblue or u~fu:i.tt:~ endometriot:LC~ lesions. The protein expression, a~ measured by E~ISA, according to endometriotic lesion t yl,:e was in keeping wit:h that of mRNA as assessed by semiquant_i.t:at.iwe RT-PCR, <~li_houdh in the lat:ter 20 case statistical sic~nit=:i~ance was c>bser~.~ed c.~nly between red and white endomet.rs_oti c les ions . a'h_is _~ndic:ates that reduced production of: MIf i.n rrcor~s advanced endometriotic lesion~r occurs at the mRNA le~~~1 arid is more likely owing to a reduced mRNA ;~ynthesi: <.nd;'or to rr,RNA instability than t:o 2_'~ translational and/or_ ~~c~~,ttransl_ational event:s. Uur findings are consistent: with tt c>:~~e reported. by :Oonnez. et al (Donnez J.
et al, 1998), who deteot.ed a difference in the expression of VEGF between different types of Lesion,>, with the early, highly vascu7_arized reel les:i.~~ns havi~mx a greater expression 30 of VEGF than i~he latex rrr:~rc.: inactive black. ~>owder-burn lesions. Thus, the h_ic~ller expression of MIF in the red lesions might reflect its role _Ln promor,~ng~'maintaining a higher degree of vascL:, L,~r development a:ud gwT~e support t:o our hypothesis that MIF p l_z~r:a an irn~~ortan~: r ole in endometriosis-associated active angi:_;genesis ana i.n:~.Lan~rnatory processes and as a marker c>j= active ~:isf:ase.
Although anG i _agenesis appears as cr.itic:al for ~~ endometriosis and that Fact::o:rs t:rrat .s t imulate angiogene:~is, such as VEGF ;Donnez :i . et a.L, 1.998 ) , <arid MIF in the present study were found ~o be: ;verexpressed :~_ri i_nit:ial endometriotic implants, the mechanis:~n:~~nderl~%irg the up-regulated expression of these ar:,c~i~~gen.ic factors _~n endometriotic cells remain to be rlore t.horc»;gh:ly invest~i_gawe=d. It is well known that endometr=_osis is t-z hormone-dependent. disease, frequently associated wit:h an imr~unoinf:lammatory pi:ocess both in eutopic and ectopic locations (D:izerega c~.>. ~_~: al, 1980; Noble L.S.
et al, 1996; Oral E. E-t. ~.1, 1.996; Ot:a H. et al, 1996;
1_'> Jolicoeur C. et al, 1~'),.; 'I'seng ~J.r'. et al, 1996) . Recent interesting ~>t:udies cia<::rly demo>nstrat:ec:~ that E2 up-regulates VEGF gene transcription in endometr:i_ai cells (Taylor R.N. et al, 2001; Lebovic D.I. ct al, 2000b) .~ncl that E2-induced gene transcription is ER dep~:.;ldent and i_s ac?: ivated through a variant estrc;gen respc:;cu~vive ~~lerner,.t:~ ioc,al:ized within this upstream region of i~hc 'vEGF gene ~:~romoter (Mueller M. D. et al, 2000) . 7.1. also a~:~:>~~ared that IL-1., a major proinflammatc;_ry cytokine whose peritonea- levels increase in endometriosis (Fa.kih I-L. et al., 198?; i'aKetani Y. et al, 1992), induce: an angsa~~en~~c phenc~type iri endometriotic cells (Lebovic D.I. et al, f000a;. Hypoxia, which up-regulates VEGF expression in enc:W rnetrial cell:>, is the>ught to be involved in endornetri~~ L angiogene~;is anti to assist revascularizavion o:E e:lesquamated endomet_rial explants when they attach av ec:topic~ >i_tcJs (Sharkey A.M. et. al, 2000) .
These factors may activate VEGF er>press~on in endometriotic implants. However, trnear role in the uv-regulation of MIF
expression irv: endometi ic~tir_ as well as ~~ndornetrial cells is unknown and currently ,n~aE~stigated iru of°,r labc:~ratory.
Furthermore, z.t woul.d loe :i_nterest:ir~g t:.o determine whether common transcript iorz :f ,~::~T,or i s ) c-,3n be involved in MIF/VEGF
production ar_d activat i_;~n of angiogenesl.s.
It is notewc;rthy that MIF was more markedly expressed in 1_esions from the initial stage of endometriosis (stage I ) , compared wi the th:e more advanced stages, which makes plausib_~e the it °,r~,:Lvemc~nt: of t.hi:~ factor in the initial steps of endornetriotic ~:::i~s:_>ue growth and development.
Recent studio:-. identit:ied MIF in normal endometrial tissue, predominant 1y i,r, tree glandular epithelium, and showed no significant differfv:n<-es i.n MIF level: across the menstrual cycle (Arcuri F. et a , 2001). cur current study with ectopic endomE~trial tissue showed no ~~i:~ni.f_Lcant difference 1.'~ in MIF protein and mRh.l~-a express_i.or~ according to the menstrual cycle phase. However, intense immunostaining for MIF was detected both in the :~tromal and the epithelial compartments, suggesting different expression of MIF in eutopic and ectopic endometri.um. ~tud:ies eval.uatin~~ MIF expression in the eutopic endometria:31 t.issi;re ;:~f women with endometrio;sis in comparison with endomE;triot:ic: tissue from the same women or from women having a n<:~rr,;al gynec~o-~ogical status are, however, in progress in our lak:>oz at~~~ry, wh:i ch may shed more light on the functional rc;le o:~ fll:f in endometrir~sis-associated angiogenesis.
In conclusi.~~>n, t;ne present study snowed MIF
expression in various c~:e.l.l types Throughout endometriotic tissue and ins marked expression in lesions representing the earliest and the most ~~:tive stages c>f the d-~sease. This suggests tha~~ MIE may ~wpresent a key e~ffector cell mediator involved in the patho~:..~1~~Y~~;ic>l.ogy of Emdc:~metri.osis through its proir.f lammat~~~ry and a ~:vc~ ic>genic act ivi t.i es .

i07 85909-5 Abstract Interleukin ~;LL-1j is ,_~ naajc~r prc:inflammatory cytokine that is belie;ne~~u1 tc> play a central :r<ale in the pathophysiology of end:~rnc~triosis. ThE: II~-I receptor type II
(IL-1RII) is ~nowru to i:~,:i.mc~t to IL-1 and to inhabit its biological effects. I:n :cur prev.i.ous st.udi.es, we showed that human endometrium exprc-:::;ses TL-1RII, anc we cbserved reduced expression of the prot~-.~.n i..n women with endometri.osis. The aim of this study was tc:> ~_nvestic~ate IL-7.RII rnRNA in the endometrial tissue of r;or_mal womc:m (r~ ~- ~6) and of patients with various degrees ca endometriosis ;r: = 5 j;; . In situ hybridization showed thar_ IL-1RI:1 mRNA expres:~ion was significantly decreased :in endometrios::,v, parvicularly during the early stages of th~, iisease stages I and II ) . This was quite obvious in both calanduiar and strc:mal calls, and it was corroborated by reverse ~ranscriptlon-pc~lymerase chain reaction analysis of IL-1RI:I mRNA iri the endornetrial tissue of women witr~ (n = 10; <~nd without (n =- 8) endometriosis.
The reduced levels of Il~-1RII mRNA i.n the endometrium of:
women suffering from Er:~cicm~:t.riosis reveals a profound dE:fect in IL-1RII gene expr.e~sion anl, :~cnse~:~uE-:ntly, a reduced capability of: endometn';.a~1 t.i.ssue tc~ down--regulate IL-1 activity. De=_'ect.ive !_T~-lR:I:I gene exprf~~;.;:ion during the early stages of endometr_i.osis (st_ages I and I ) may contributE: to 2_'~ the etiology of the d; cease.
Introduction Recf=nt evidEnce has demo>nstrated a direct relationship between ,.he immune and the reproductive systems (Adashi E.Y., 1990) . Cytok_ines produced by .Leukocytes <~nd various other cell tyke's ha.~ve a wde range of biological activities, i_:zcluding the abi..lity to rey .~1_ate immunolog.ical, neuroendocrine, and reor.od;.~ct.iva tuncti.ons (e-oetzl E.J. et al, 1992. ) . Among these ~~~,rtokine;~, the nterLeukin 1 ( IL-1 ) system appears to be relevant to various r_epr,.~ductive processes and to a brca:~ s~~ectrurn of pathopruysiological responses associated w i_l::i:~ roost: defense ~~nd i.nflammation (Dinarello C.A. , 1989; 1:~:~.rsen C. et ai_, 7.990) .
A c~~tok.ine r aiv:inc~ potent anti nflarnmatory properties, I7~-1 was shcwe to be involved in numerous immunological and repr:oc~u~:~tive activ.iti<~s occurring normally in the human endometr.:i urr, during a normal. rn.enstrual cycle or during embryonic impla;:ntat:i_on crud devel<y>ment (Simon C. et al, 1995b) . ~~ growinc; bod~.a of evidence ind,~cates that IL-1 may play an important r~~~le in the pattnophyszology of endometriosi=, a gynec:~~iog:ic:a1 c~i;ease :-;teat i.s believed to arise from ectopi c gre:~wr.h of enc~ometria_~.. t.i ssue and .i.s associated with a chrt.:~nic immunoinf_lamTnatory process (Halme J. et al, 19f.~4a; Gros.~>xinsky C.I~z. -st al, 1993; Sahakian V. et al, 1993) . I:z women ~,r.it h endometriosis, peripheral blood monocytes (teller J.M, Ft ~~1, 1'a8'') as well as peritoneal macrophages (l~Iori H. t:t al, 1992) were found to be more activated than irn norr~~ai women and to sc~c::rer.Fe elevated .levels of IL-1. Increased cc..rrcen ;ratir>n,~ of IL--7_ were detected in the peritoneal fluid :,f women suffering .f_rom endometriosis (Mori H. et al, 1992; fakih H. et al, 1987). According to our previous dat<~, IL--1 enharuces the production of monocyte chemotactic protein-1 (MCP-1.) b~r. r:um.an endome~tri.otic cells (Akoum A, et al, 1995<y ) and by r:utois endometri.al cells of women with endometrio;>i~. (:clicoe;~r C. et al, 1998) .
Moreover, these cells ai:peared t.o be rr:ore sensitive to the action of IL--1 i.r,, wom~::r~ with than i.n women w:i thout endometriosis (Akoum .~~. et al, 995b) .
The effects of IL-1 l_Lke.ly are str.i.ngently controlled l ~ vivo. ,lumber of IL-1 inhibitors that can block the activity of 1I.--~ c:n targE>t celis ha;Te >"een identified anc~ pair°t:.ia7.l;; charactE~r~zed (Larric:k ~T.W., 1989) .
Thrf~e recept:::~rv; f or IL-1., now designated as TL-1RI, IL-1RII, and IL-1P,III (moz:~e commonly called IL-1R AcP
[accessory protein] ) , IvarrE> been c:iesc:ri_'oed. The relative importance of these re:ept.ors in IL-1 signaling has been recently clarified. F. c_:l:~itical role for tine iL-_LRI and IL-1R
AcP in IL-1-i::~.duced cle:l:~- ~ cta_vati.oru has been demonstrated by several group's (Colott:~ r'. et: al., -99; Sims ~7.E. et al, 1993; Greenfeder S.A. =~t: al, 199':>) . In contrast, IL-1RII
appears to be dispensah:l_~-~ for II,-1 signaling and may act as a decoy receptor (Colott:~~ h. et: al, 199::; Sims ~.T.E. et al, 1993) . Interleuk:i_n--7. r-.~,eytor arutagori:~~>t (IL-lr<~) is another natural inhibitor of 7:C,--I, which competes with IL-la and IL-l~ 113 for IL-1. RI (G-ranowit:~ E.V. et al, ira91 ) . Results of several studies indicate tt-~at the IL-1 .:system is available locally in the human e:cndomet.ri-al. tissue and may be an important mediator. in :1 ~:~:a:i. cell.~_zlaxv :i_nt:eract=ions (Tabibzadeh S. et al, 1990; Simon ,.. et al, 1993; B_i_gonnesse F. et al, 2001a; Kauma S. et al, 1 990; Bictonnesse F,. et al, 2001b) .
However, few :atudies f:a~Je i ocuswd on t:lze possible role of this system i.n the pa t:tuophys.iology of ~=er~dometriosis.
According to Sahakian et ~.~_! (Sahakian 'J, et al, 1993) , ectopic endometrial tissue does not: express IL-lra.
2.'~ According to our previo-~m studies, the erndometrial tissue of women with er:,dometr=io::::L~; ex~rE~sse~> row l.PVels of IL-1R.II
protein compared with t:fu.at of healthy women (Akoum A. e1~ al, 2001a) . The aim of tu:.is study wa:> to investigate the expression of IL-1RII at the level of mi~NA in the endometrium of women wit;u and of ~r,romen without endornetr:i_osis. Alteration cf the IL-1RI_i gene a>:;~:~~es.siorr ma~% prov.ide mc;lecular evidence for a deficient contrc;l of IL-1 in the eutopic endometr.ial tissue of women with cndometriosis.

17.0 8.5409-5 Materials and Methods Subjects and .T~~:;sue Ccllect_c;r:. Endometrial tissues were obtained from 79 ~,ao:rnen aged betweE:r. 30 arid :36 yr who were undergoing l,aparo,:c:«pic surgery f: or infertility, pelvic pain, or tubal ligatic°i and who quad note receimed any anti-inflammatory '~~r horrnon~_l. meclicat ion cu.zring a ~:~eriod of at least 3 mo before ~_apa:~oscopy. nifty-t=r:ree women had endometriosis of trario:_a:== stages (I, II, III, and IV) according to the reviw,::~c~ hrnerican ~,er_t:,wiity Society classificatioc~ (Rock ~.~., 1995) . Twenty-six women were fertile and had no vi.,v.l:~:Le erndome~trioa_~a at laparoscopy (Table 9) . The cycle I~>:riase was c~etermi.ned according t.o the cycle history, progesterone levees in t~r~e serum, and histologic criteria. ~='girl ~raformec ~~onserit was obtained from each patient, and the st.;.zdy was approved by the ethical committee of t:he "Cent r c~ H~~spitd lien Urm~versitaire de Quebec"
(CHUQ).
Table 9. Clinical char:<~~r_.erist:ics of: pat~.ent:s at laparo;>copy.
Subje<:ts cycle by ~>ha:>e (n) Sub;~'~~t~: Proli~erati (n) Age (yr)'~ ae Secretory Controls 26 ?
3 :!-6 '~ 1 5 .
. ~;
~

Endomet.riosis '_.3 3-1 r 5. 9 29 , f.;
~

(total;

Stage I ~.'2 3~~, ; 6,;311 11 Stage II 7.9 30. .'5. ~ 1-~
5 t) Stages ITI-Iz7 7.2 ~ ;-q,'7': 7 i, Fertile 2~3 32.~ E.~ 'i 13 InfE:ertile a'a 30. -~4.'7-=3 16 ~3 2C) ° Mean ~ SD.
Endometr_ial :~arnple~, wr~rE~ r_oll>>ct.ed with a curette before laparoscopy. '7'~ne tissue was plagued in cold, sterile Hanks buffer~~~~ saline :~ontairlin~~ ainl~il:~i~Y~:i.cs, then immediately t-ansportec:i t:.c thir laboratc.~ry and snap-frozen in liquid nitrogen before x~F ~~ng stored at. -80"C.
Fluc:res~rence :boa ~~it:u Hyb.ricl.i.~;ation. The present experiments were ~>erfcrrr«:eci as prevvou~;ly desct-ibed (Jolicoeur C. et al, 1998). Brietl', bi.otiri-labeled IL-1RII cDNA probe was prepared lay ni.-ck t. c~~n-m:lat:ion (:f_awrenc_e J. r3, et al, 1985) from the entire p~-asmi::r ~:~ecto.r pc~DivA~.
Cryosections it::hickness, 5 arm) from i9 endometrial tissues (Table 9)~were :i.xed witr: formaldehyde and dehydrated with alcohol ~~efor_e be:i-net ruybrid.i.zed with 5 ng/ml of biotinylated IL-1RII c(nNA probe. Bioti.r was then detected using a rabbit antibior__i.ou antibody, ~~ ~;ictiny-'~ated goat antirabbit antibody, ar:ca i-luorese:ei-n iscthiocyanate-conjugated streptavi.din, respect-.vely. ~,'ections were finally treated with ~>rop:~di.urn a..c.~clir~e, which ma~:es the nucleus visible in yel-low-orange fel-towing u_t:.z~aviole~ excitation, and were observed under a fluorescence microscope (Leica mikroskopie and syst:erm~ ~'.~rnbH, Model DMRB, Postfach, Wet z;l.ar, Germany). Serial secti-;>ns from each tissue inr_ubated without 2C IL-1RII cDNA probe or ~a-i.~~h norcspec~~fuc: L:~NA pr~:~bes prepared from the plasmid vector '~:Lor~e were used as negative controls.
Eva_tuation c~ .Staining. Stain ~ng was evaluated using an arbit:rary scale as previous_Ly reported (Jolicoeur C.
et al, 1998) . In br.iE:~:,, c_=ac:h er~dorne~:r:ial section was gz-aded according to t_he inters:ity of staining ~.hree times in three different, randomly sE, _f:;zted area. arid :cored from 0 to 3 (0 - absent, 7. _- light, ~. -- moderate, and - intense) . Grading of each specimen was ~:e:rformed by two different observers who had no knowledge of th;e clirlica:~ star_us of t=he patients including laparos;_opic. ':liagnosis.

11~ 85409-5 Reverse Transcription-Palymeyase Chain Reactian.
Total RNA from the end'ametrial tissue of 8 normal women (3 in the proliferative pr:as fe ,rnc~~ .'~ in t:oc~ ;ec:retor,>> phase) and of women with stage I- l encxomet.r-ic>s.i: ~ 4 ir_ -i_he 5 proliferative phase anc::i ~ ~n the secvretc>ry ~>hase) was extracted using a Triz,~:::L ~M rea.gent accor~.:iing to the manufacturer'; instrucldions (Gibc:o f3RI,, Burlington, ON, Canada) . Total RTJA ('::~;)',.i ng;~ was rever:~E-~ transcribed into cDNA using 50 U of rev=:~::-;:~e trans~.:riptasE.~ in the presence of 1C~ random hexamer pr:imer~(:?.5 mM) , ciN'fPs !1 mM each) , 1 Uiml of RNase inhibitor, 10 m1~ ':rr.is-HCl, 5C~ ml~~ I;Cl, and 5 mM MgCl (Gene AmpTM PC;R Core K::.t; Perkin-Elmer, Foster City, CA) . The reaction was -_ncubatec.: <~t ~'S°C for 1.5 m~,.n, 42°C for 30 min, and 99°C for 5 min. 'C'wc~ microli.ters of the rF=verse transcription (RT) r_ea,<~-; iorl were use~~ fcor po:iymerase chain reaction (PCR;~ in a final volume of 50 ;ai with 200 pmol of each IL-1RII primer ('~': TCC ATG TGC AAA TCC TCTCTT (SEQ ID
NO . : 1 ) ; 5 ' : TCC TGC CC'C T~: A 'rCT CA'T Ai:'~ ( SEQ I D NO . : 2 ) ;
expected ampl:imer lenc::at~~-~, 576 bp) , 0.'~' rnM dN'fPs, 2 mM MgCl2, and 2.5 U of Taq po:Lyrruerase (GrovF~s R.W. et al, 1994) .
Amplificatiorv. was pert:ormec:i fa.r 3C) cycles cons.i.sting of 1 min of denaturation (94°C), 30 sec of annea.l.ing (60°C), and 1 min of primer extension (';<''"C). Glyceraldenyde phosphate dehydrogenase (GAPDH) was used as a control. Four mi_croliters of tr:e RT :react=ion were used for PCR in a final volume of 50 p1 with ;:::!o pmol of each pr :i_mer ( 5' : TGA 'r GA CAT
CAA GAA GGT GGT GAA G ( ~EQ I D Nc:) . : 3 ) ; 5' : TCC TTG GAG GCC ATG
TGG GCC AT (SEQ ID N0,.:4); ampli.mer size 240 bp), 0.2 mM
dNTPs, and 1 J of= Vent DNA Polymerase (New England Biolabs, Beverly, MA) . Amplif_.c:ation wa;~ performed fc;r 30 cycles consisting of 30 sec c:~.f denaturat=:on (95°C:) , 30 sec of annealing (6~:)°C) , and 7 min of primer ext=ension (72°C) .
These optimal conditic>n:= were determined following linearity 1i3 85409-5 tests using l, 2, 4, aue~ 8 ~.al of tree F<'1' react ion volume and 25, 30, and 35 amplifi-<:~t_:_cr~ cyc~.es. tlnvplificat_ion of genomic DNA with these I:~-~_m,ers d:id not.. produce a signal, suggesting that the amE:~.l...ifi~:.ation site:: crossed at least one intron/exon boundary. ~~ total o1v ~'0'~, of the 'CR volume was then analyzed on a 1.° (w!jr) agarosc: c~e_I_ i.n true presence of ethidium bromide and t r-ansfer_red to Qi.abraneTh' Nylon Plus membranes (Qi~~gen, S ar:r.f~ ~::~ ar_i.ta, C:P,;~ . Membr<~nea were dehydrated at 37°C for 3t:) rrai.n, p:~:ehybrucFized with a 1C hybridization buffer, y.Wridized in the same buffer (without Denhardt solution) with ''-P-iabeieci :CL-1.Z:II or GAPDH cDNA, and washed in 1X 0.15 M scdrum chloride and 0.015 M sodium citrate (SSC) , 0. 2X St' .',, arid 0. -~ a SD:~, respec'~ively, before being exposed to x-i.~a~~ :~.ilm (E~asa:.mar~ Kodak, R~:achester, NY) .
Specificity >:f° the amp_Lificav ion process was verified by Southern ~..lcu:t r-iybridizat:iona A negative control (PCR in the absence of c~ DNA) as we:ll_ a~;~ a positive control (cDNA preparation from human endomet:ria.L tissue expressing IL-1RII) were inc.ludec:: i.n c:e.~ch :yF~r.ies ~~tr IL-1RII or GAPI)H
amplification . The qa.ar~tit;y of the PCR produ.~ts was determined by densit=orr.et: r i<: arla~.: ysi:> of the .intensity of the hybridization signal. 'T'he relative levee of IL-1RII mRNA
normalized to GAPDH mFvtdA was calculated, and the results were expressed as a o of tt:e control t~~,:.lue (ioosit:ive control) .
Statistical A.rualysF~s. The intensity of IL-1RI_L mRNA
hybridizatior_ signals was expressed as arbitrary units.
Statistical analysis urt:~:~ performed by t:ne Fisher exact probability t.,sst (Guz. ck D. S . , ~! 9~~6 j , and Bonferroni correction was applie<a when more than twco groups were compared. An,~lysis of '~L-3~II mRNA _Levels as determined by semiquantitative RT-PC'R. was performed using one-way ANOVA and the Tukey tent fc:r pc_: t-race rnult_ ipl a compar:i.sons . All.

analyses were carried :-~ut: us.ing t.h~: ;>tat. istict~l Analysis System; (SAS Institm_rte, Lrn c. , ~.~ary, I~dC. D:iffe:rem:es were considered to be stat:i.c=,-:i<,.ally si_grrif=ic~ar~ at. P ~~ 0.05.
Results G Analysis of Ih-IR.II GE~.~~:~ E.~Y:p_ress.:icr~ _i.n r-he F;ndometrium by In Situ Hybridi~~ition The expressi~.>o of :IL-1RII rnR:'JA in the endometrium was studied by in sitr.; Iuv~b.ridi.zat.ion to examine the site of IL-1RII synthesis and t;-. c~c>mpare the :Le~,~els of IL-1RII mRNA
in patients w=_th and without endomet:riosi_s. Figure 21 shows the appearance of endc::metria:l. st:rorna anci glands at 666X
magnification (A1 arud 1311 fo:l.low:ing hyb~:~idization and staining with prep.idiu.rn iodine. The t:ybridization signal (green-yellow) could r..n.ly be visc.zalized at rrigher 1_'> magnification (1665X) and appearea to be located mainly in the endometrial glared::, (A'2 and E32) .
As described ear-~ier, an arbitrary score was used to quantify the IL-lRy :I rnRi~IA hyb:ri.dizat..:.on signal.
Statistical analysis of hybridizat ion sa~c:>res using the hisher exact test showed a s.7.gnificant de~cr_easc:~ i.n women with endometriosis. compareca =:~ normal. women, both in endometrial glands (P < 0.0001) arid st_roma (P < 0.006) ('fable 10) .
Furthermore, when pat::_ents with endometri_osis were grouped according to the stage: of disease, a significant decreaae in IL-1RII mRNA expressic:n in the glandular (.P < 0.0003) as well as the stromal (P < 0.0:.2) compartment was observed in stage I. In stage II, a sic:3:nifi:.ant decrease in IL-1R.II mRNA
levels was a1_so obsersrec~, t;ut o~l.v in the glands (P < 0.042) , whereas in more advanc~:eci stages (-=I:I and IV) , no statistically signific::arit difference was found. A graphical illustration of IL-lRll mRNA scores in normal controls and in women at different:. sta:,~c~~ cf: endc>metriosis is shown in figure 22.
The effect. or t:he menst_ rua). cycle oo levels of IL-1RII mRNA in the endorra:=tr~urr~ was a'~so ewal_uated. Statistical analysis of the hybrids ~:a~t:ior~ scores slic:wed no s:ignifi.cant difference between the prol.iferat.=.ive ancthe secretory phases within the centro~. or -:.he endomet.ricsis group;~. However, the decreased expression of ~L-1RLI TnRNA ob_>erVed in women with endometriosis was more not~ceabl.r: ciur_im~ the secretory phase 1C of the menstrual cycle, either l:~u t.w.e c~l.ands (F' < 0.003) or in the stroma, in wrii~.:r a:~ vi,~ati~~tica_~:1.~,~ significant difference between worrev with anc.~ women without endometriosis was seen only during t.tie secretory pruase ( P ~: 0 . 018 ) ( Table 10) .

11.6 85409-5 Table 10 . Number ~:~f suL: _j ect y acc~rcii.ng t o .intensity of the IL-1RII mRNA ::~ybr.i.dizat::i<~n si.gna~ i..n the endornetrium.
Stroman; Glands (n) Inter:::~it In tersty ,~ of ~. of stair:::z_:: st ain...r:g Mum- -._0 ..__ j ~. , _ 1 2 3 P
G (1 l.f~r _ Controls :-r~ 8 ~I ~ 0 ~ I 6 10 9 Endometriosis '_~ 31 ., 1 ,, 0. (i06Q ~:5 :12 2 0. 0001b (total) Stage I~ ~:2 16 c; ~ 0 0.012--4 l~: 3 0 0.0003b Stage IIb 79 8 .0 C (;.422 0 12 6 I 0.042b Stagas III- 1.2 7 '. C (:~.4:~(.)8 3 1 0.186 IV

Fertile ~'.3 14 '~ a 0 ~~ , 2 13 7 1 0. 030b CalB'-Infertile i~ 17 ~. 0 0.76 ~_ ~._ 5 1 0.0004b Proliferative phase Control j 2 _ _ () C 3 2 4 Endometriosis '4 13 1C n ~C~,2172 75 7 0 0.009b Serretory Phase Control ._ 6 ~I C 1 3 8 5 %

Endomet.riosis <'9 18 ... ) 0 ('.018'.. 20 5 2 0.003b ''Fisher exact test.
Comparison with controls; i va!11E?a c: _rc..~c;ted > y the I?onferroni c procedure.
Stat:isti.cal _~r~a_Lys:is c>f t: he hybridization scores according to t:he fertil:Lty status of suh7ects showed that, compared to normal ferr:.:i Lf~ women, fern :i~_ee women with 1G endometriosis had decreased expression of IL-1RII mRNA, both in the glandu__ar ( P < (:) . J 3C) ) and iru -~hf~ ~~trornal ( P < 0 . 018 ) compartments of endom~t~rial_ tissue. However, in infertile women with endometrio.ia, a significant decrease in Ih-1_RII
mRNA expression was o):~~~e rvF~d only i.n them glands ( P < 0 . 0004 ) 15 (Table 10) .
RT-PCR Analysis of IL-l.RII mRNA E~press.ior. in the Endometrium Exp:-ession c~_ IL-1RII mF;NA in t:.h.e endometrial tissue was fu:rther ev4cluated by semiquantitative RT-PCR in 8 normal contro:Ls and 1f, women with erldo~nei~riosis (stages I and 20 II). A representative RT-FCR and Southern blot analysis of IL-1RII mRNA .in the er:~dometrial tissue ~f women with 1.17 85409-5 endometriosis and of rv~t-mai c:antrals is shown in Figure 23A.
Levels of mRN.F~ we:r_e s_i:rnific:antly :i_ower in the endomet.riosis group than in the cons: rv::l_ clroup ( F <: 0. C'062) ( F lg. 23B) , which corroborates t:he in si.~-.~ hybridization data.
Discussion Our results :>r~awed a significant decrease in the levels of IL-LRII mRNh _~m the endametr=ium of patients with endometriosis. This w,u._ o)vw=ious in lulaE~ strama but was mare significant i.n the g.la ud:~ as examined by in situ hybridization. Semiquant~Lt.ative RT-1?c:R analysis of IL-1RII
mRNA levels in the end~:nr,f~trial. tis.,uc-~ a~ so showed a significant decrease in wc:~men with endometri.osis compared to normal women, which cor~obarates the ire situ hybridization data and prov__des evic_-.r~ce for a pro.faund defact in IL-1RII
gene expression in the ::.utopic endomet:r:.um of women with endometriosis,. The IT -:l.:f~:.II is syr.thes:i~.ed as a membrane bound receptor that lads t:he signal--trarasducing cytoplasmic domain found in the W.rmti~~ria7_ receptor: type I (Suns J.F. et al, 1993) . The receptor cyan be cleaved and shed from the cell surface by the proteolytic action of matrix metalloproteases (O.r_l~nc:xo ~. et al., 19a'') . Both membrane-bound and soluble forms ,~f Ih-1RII c,~r1 hind to IL-1 and prevent its interactic:n with the signal~~ransducing IL-1RI
(Colotta F. ei~ al., 19~~:3) . Our ray: silts, demonstrating a 2.'> significant decrease in IL-7.RII m)~:NA expression in the endometrial tissue of ea~omen with endome~;riosis, reveal a de-ficiency in the ability of eutapic: endometrial cells of these women to down-regulatf their response t~7 IL-1. In fact, eutopic endoraetri.al cFVl1 s ;f women with endorrietriosis appeared to be mare sensitive tc I:L-1 compared to those of normal women, and they secreted higher zmaunt:s of MCP-1 following stimulation with IL--1 in vitr« (Akoum A, et a-, 1995B).
In this wori< , we observed t~e~at: defective IL-1RII
gene expression in the end<.~met:rial tisa~ae was significant at _'~ stages I and .CI, but: r~.>-:_ in more advan~~ed stages (III and IV), of endometriosis. These results point toward a process of cell act:ivaticn t:ha:t~ -~a~:es place 'L~oc~a,~ly in the intrauterine endomet:r~ u:nr~. a ~: the ear l_ iest:: stages of the disease, and l:hey sugrev;t that; end.ometr.-iosis is more act=ive during its first stages. Available data regarding the correlation between tr:c: .°xt_ent of endometriosis and that of the chronic z_nfla~nmatcr~y~ process cbservec< in the peritoneal fluid of patients and ::_r: ~~c:v~r_~pic.- as w~~l_:_as eatopic endometrial_ tissue r_erT~a ~~n c:ont:rovers ial . It: has been reported that the conce~;~ratiorus of inflammatory cytokines, such as interleukin 8 arid RANTES ( regulated upon activat=ion, normal T cell expressed :~nc:l secreted) , ~.-orrelate with the severity of disease (Fy~~.n I.P. et al, 1t)95; Khorram 0. et al, 1993) . However, other :studies have srlown that less extensive endometriosis may be r~o:r a i>ioc:hemiaa~~l~ active than older implants (Vernon M.W. et al, 1980 , and that: peritoneal inflammation is more a<~t.ive dur3_na t:he Lnitial than the advanced stages of the ~~isease (Verno-n M.W. et al, 1986;
Haney A. F. et. al, 1991 . accorc~:ir~g to =~essey et al (Le:>sey B.A. et al, J_~~94) , t=hc: prefect: of int:egr.irv expression in the eutopic endometrial ta:>;~ue i.s iro-,rerse~~~ related to the :stage of endometricsis. Our ~~revious studies :;bowed that in :>itu expression of MCF-1., f" ~;.;ctemt chemotact:i_c and activating factor for mc;~nocyt:es ~n~:~ macroph<~yes, in the endometrial tissue was ma__kedly elevated during the initial. stages (I and II) of endomei~riosis ~;uei decreased during more advanced stages (III and IV) . L:r,terestingly, we also showed that.
defective IL-:LRII pros:cin e_xpre~~si.on in t:he c~.ndometrial tissue was m ore marked ea.~ 1_y stages of the disease :l~..lri ng tr~e~

(Akourn A. et ~~1, ~_OOla;~wivir_h iri keeping with the ,. i5 findings of the p.reser~t,:~~t=udy, we also found a :rid significant negative r_el.atior~ith MCP-1 expression i.n c~>r w the same tissues (Kha:rfi E:~t: al, ~-.
A. ~:~.)0 The data of t.xle present study suggest that the reduced expre~~sion of T:L-1F<.II in the uterine endometri.um of women with enclometri.os :.,:~ is ~~elat:ed, at. least. in part, to a defect at the mRNA I_evf=~~. , although tr_ans lational or proteolysis dependent ErlecharW sms cannc>t be ex~~luded.
However, whether such :..; c:~efec:t i~~ due tc:~ decreased mRNA
synthesis or reduced rr~l?Nl-~ stabili_ty ._s Linclear. It is also noteworthy to add that t_t1e decre,-use yn rrIRNA lt~ve_l_s in th.e endometrial tissue of women with endometriosi_s was more significant during the p~~ol._i_1=era~:i.~~~e t:l~an the ser_retory phase of the menstrual cycle, w>"ui.ch ag~,ii_rn ~._s ccnsistent with the cycle-dependent pattern. ~af defic _ent IL--1RII protein expression (Ak:oum A. en <~ i_, 2001x) . The mechanisms underlying th.t cycle-:Iependent, a!_>ei:rar:t IL-:LRI=L expression remain unknown. Howevr,. t: his rnay have an interesting significance, because it: suggest:; ~:h~~t= endomet=rial tissue debris refluxed into tE.e peritonceal_ c:a~rity at: the end of the menstrual cycle may corv~t:~~~_n low _Level:; c f IL-:LRI=L, which may make the tissue less capable of down-rec;ulati.ng =LL-1-mediated cell activati:~n a:nd le:~ca t:o an exaggerated pea~itoneal inflammatory response.
The data of tare present: study also suggest that defective IL-lRII expr_~:=>~:v>~~c;n dun ing t:he early stages of endometriosis may play a:~ rc;ie in t:he initiation of the immunoinflamm;:tory pro~:c:e:>:~ eutopically, in the uterine endometrium w-.ere the ~::..i~ease is belz_eved to originate, and ectopically, in ttze pe:~ i.;::oneal cavity wr.ere endometrial tissue is r_ef::Luxed ane: r~eveloped. mhe lack cf statistically significant c:~:iffe=rence: ~.n TL-1R_L I expre,~sion between women with and women without ~ndomet=r_os;_is during the late stages of the disease may, o:~ the other hand, be part o.f the counterregulatory o_r rf~>parative mechanisnus t:.hat stabilize endometriosis by 1i_mit L~;g c:~:Curonia inflarzunation and profound tissue damage by restr ictin~ IL-1-rnedi~,_ed proinflammatory actions (Ding=ello C.t= . 1939; Lars.en c.. et a.1, 1990) . It is interesting to note trat hlo:ri_ et: a1 (i~Iorv H. et al, 1992) demonstrated i=hat the ~~:~prE-e:~siorl c:,f IL-'~~~ by :peritoneal macrophages was elevated ~:~uring th.e irlit:i_al stages of endometriosi~~ and that: i L-i:ra, amo~.hr~r natural specific inhibitor for IL-~, w:~~~ more elevated d?.zrincx the late stages of the disease.
1_'~In conclusicry, our results demonstrate that the expression of: IL-1RII rnl~.:'~1;~ was dec_:reasf~ca in the endometrium of women with endometrie: s is, pa~r~=i cJuiar:~y during the initial stages of the disease.:'ue~h a defect:.iv~~ IL-IRII gene expression by endometn:la I c_:e:Lls pc: rots _oward a profound defects in the capabil~~_~;- of endometr:iaL cells to down-regulate IL-1. actions,tan:i_c_~:~ may ~vlay a relevant role in the initiation of the immr:rn_> inf lammatcry~ p:rc~cess associated with endometriosis, both tie intrauterine endometrium and in ir:

the peritonea=_cavity i=;.,:1.1c_>wing 1=uba:L rEeflux. It remains, 2_'~however, to determi n<:,a~ whether su~~h a def:e~~t is be due t:o transcriptiona l and/orf;ostt.rans;~riptiorual events.

Capsule: Serum .from wco~nc~~rz with endomet~~osi~: induced MCP-1 secretion by L)937 monc_:~-tic cells and sknowed a significant decrease i.n sI=L~-1RII i~:fr~~LL .
Abstract: The ability ~r:t 1~E~~r_Lphera_ k~>:Loc,d serum from women with endometriosis t.o i.~lduce rnonocyte. c,hemotactic protein-1 (MCP-1) secretion by r«~orloca,~tes was assessed, as well as the effects of recombinant ini.=E:rleukin-1.13 (rlh-113;; , IL-1 receptor antagonist (rI:L-lira) , .-~rLd solublf~ I I_,-=1. receptor type II (rIL-1RI I ) on MCP-I secret:i oru .
Design: Cultures of U9_>-% mc:~nocyti.c cell.:; exposed to serum from normal or enciomet:c~_c:~si s worr~t-yn.
Setting: Gynecology ci in:Lc and human r_~eproducUion research laboratory.
Patient (s) :79 women ha~,r:irlg endornetrios~s and. 38 normal. women having no evidence of E.~r~~:Iomet=ric.~.~is at I aparo~copy.
Intervention (_.) : Peri.pha_c:a:rl Wood obt,~ainE-d a few days before laparoscopy.
Main Outcome !Measure (s ~ ; MCP-1 secret=ion in t:lne culture medium and serum c:,oncer~t~rations ~of sIL,-1R.II, IL-1f3 and IL-la by enzyme-linked immun::~.~or_bent assay (EhISA) or by enzyme immuncmetric .ssay (D1.%1j ..
Result (s) :Serum coracenr...~°ations of: sIL-IRII were significantly lower in women with en:~i~rnetriosi,~ staae~ I-II than in normal women (P = 0.;X02), whereas those of I=L-1~ and IL-Ia were comparable in women w:irh <:~rnd witluo;_~t: enciometr_ osis. Serum of women with en~lometri.os i:_; v~nduced higher secretion of MCF~-1 by U937 cells th~.n that o_L normal women ;P = 0.018), particularly in the ira~.'t:i_al stagr~s oi= er~domet:~~iosis (stages I-II) (P = 0.002) , anc: -IL--11~IT sign:ifir_antly blocked that secretion ( P =- 0 . .)008 ) .
Conclusions: 7.'hese f_-in~aings point t=c-~wa.-cd a deficiency in the mechanisms involved ire the :~clwn-regu=iat~.c>n of iL-1 actions at systemic leve7_ and rev.-'al :,:IL-1R~I as a key factor involved in that process.
Key words: IL.--1, IL-1 rf,»::;<~~:,tor, ~:mdc:~rnet.z~iosi.s, MCP-l, peripheral blood.
Introduction Endomet:riosi;-~ is an irnrnune-related chronic inflammatory disease, ~~:naracteri~ ed by the presence of endometrial-like tissuf_: in ectopic locations, mainly i.n the peritoneal cavity, and r,:~;~oc.~iatE~c~ with increased secretion of proinflammatcry cytokiwes including IL-~, IL-6, IL-8, tumor necrosis factor-a'.~pha 1'hl~dl~'-cx) an::l MCP-1 in the peritoneal fluid (Senturk: L.M. e1. :I_, 1~~99; Mt.ala.y~_rn. N. e'_ al, 1999) .
These factors have bee~i postulated as being implicated in the development and pr_ogre::~,:>:i_oru of thle d-._:>e~~se. Iznmuno-inflammatory changes ok.:~~>~-~r_ved ir. patYent:s with endometriosis are not restricted onl ~y to the per i.toneal cavity where endometriotic lesions :~:f>_quently de~veio~~ (Senturk L.M. e:t al, 1999; Mulayim. N. et a:1.,, :1_999; Harada ~'. et al., 2001), bu.t were also detected i.n i_he eutopic endometrium (Braun D.F. et al, 1998; Sharpe-~himms H;.. L~. , 2001 ) ) , and the foer:ipheral blood (Mathur S. P. , 2000; Drrn~_~c~~~~:i. W. P. , et a~_, 1.994 . . In endometriosis, peritont~al macrophages are mare activates. and secrete elevated conce:r~tuations of proinflammatory cytok:ines (Mori H. et al, 1991; f;ana N. et al, 19~~6j . However, other reports indicate that oervpheral bl.aod monocytes from women with endometriosis ar~als~.a act i.v~,ted ('<eller ~T.M. et al, 1987; Braun D. P. et a:i , 1996) an:~ _,how '.he ability to stimulate endometrial ~:c:ll growth ~r1 ~;~i~.:°-~~~, whereas monocytes of normal feni_.ile women suppres:~ the pra~~ iferation of these '.> cells (Braun I:~. P. et a::1., 1 994 ) , lfowever, the mechanisms involved in t:he active l~ i on of t~hfJ~ a ~~e ins remain unknown.
IL--1, a maj< r pr<,i:~~flamrnat~:~ry cytokine, exerts its biological ef~=ects viG the receptor t~lpe I ( IL-1RI ) , t=he signaling functional receptor, which r~:~ expressed in different cel_L types. ;~}.~e~~ific inhibition of= IL-I is ensured by IL-lra which compete:. with IL-1 f~:~r >pecific binding to IL-1RI, without triggei~:~.ng s.igna.L transducti_oi~ and cell activation (Braun D. P. E-'~ al, 19'41 . I:G--I_RII, having no signaling properties, has been reported as another natural inhibitor of -_L-1 (Bor~~s~~hiD. et= al., 1~a96; Colotta F. et al, 1993) . It has been surgc:sted thar_ this ~ ecept~r, with a short 29-amino acid cytop-ias,~rui_:~ doznair:, tLmcW .ons as a non signal transducting cell surf a:rce or soluh:lce '''c~eac:oy" target for IL-1 (Boraschi D. et al, IU~:~v; C:"olotta I~'. e~ al, 1993) . We have previously der:ronstratei that I:L-RI:I presents defective expression in the endc:-nf-:mum ofwomen with endometriosis (Akoum A. et al, 2001~'~. Moreover, c»~:r recent findings :showed a significant inverse c.,_,rrelat:ion between the decreased expression of IL-:LRII -zu.:~ t_he irir.rc:a:~ec~ expression of MC:P-1 2~~ in the endomet:rium of wornert with er:domet:r iosis (Kharfi A. , et al, 2001) . Alt:bough tme rnec:hanisrns i.mpa.~cated in the regulation of IL-1. anc MCP-1. are cornple~., such an abnormal expression of IL-1RII p~.at into promi.nenr:e a major defect: in the mechanism: involve. in the c:ontro:l c~f lc>cal IL-I actions.
3f In ~Tiew of_ t h:~ numerous systemic i.mznuno-inflammatory changes ck>:~e.rved i n endornetmviosis, the aim of the present work was t:~ assess c-i_rc.u_L<~t~_ng levels of IL--1RII

and its liganc~ IL-1, Urn::A tc> ir;vestigate t:he putative r_o7_e of this system in endomet.~~:io.sis-ass.~ci.a~ec~ peripheral blood monocyte activation.
Material and Method:
Pat gents. Women were xecru:i.r:ec~ into the study after they provided informed ~::onsezzt for a prc;tocol_ approved by Saint-Fran~oi:~ d'Assis~.: hospital ethics committee on human research. Endometriosi ~ ~,aa:=. ideritifieci during laparoscopy for infertility and/o.r pelv~ icl pain c>r fo.r t~ubal ligation only.
1C The stage of endomet:r:i~::>~ ~s was de~t~~rrn:ined according to t:he revised classificatior: a:~:l- i'he Amer;~can I'erti.l ity Society (American fertility Sc~~:_~.ety, 1985) . Sub-~ects with endometriosis (n -- ?9j ~:o~herwise had nc> ot:her_ pelvic pathology and were not a,~k~nc~ any ar:t~:i-~ nfla.mmatory or 15 hormonal medication at aas t ~ rricmths be-fore iaparoscopy.
Control subjects (n -_ v,~,) wer_e fr.rt:ile women requesting tubal ligation and having nr~ ;ri sirs=~e e~~~idence of endomE=triosi~; at laparoscopy (Table 1.1). 'Che cycle phase (prol.iferative or secretory) wa~~ determ:irud accord.i_ng t=<:> t:he patients' cycle 20 history and to the seri_an~ progesterone. '~'hirty three of endometriosis women an,-:I i~? of contr-o=1. .<~ubject;~ were in the proliferative phase of t:Ine menstrual cycle, whereas 46 of endometriosis women an:~ ? 1 oj= cont.rul sub~ect.5 were in the secretory pha~>e, 25 Table 11. Serum source arid clinica'1_ characteristics of subjects at laparoscopy Serum Source N~ambar of pati2nt:s A_ge (mean ~ SD) Controls _ ~ 3c ~
Endometriosis ~i'? 3y ~_ =, Stage I 34e 2 ~_ Stage II 2- 31 Stages I and II 6: jy ~- r-, Stages III and IV 1r 3q ~ ._ Co1_!ection arl~ .P3-oeessznc~ o.f .talooca Samples. B1_ood samples were cYrawn a f~:~a da.ys befior_e :Laparoscopy in sterile tubes containing etY~y.l~ur~c~diaminetetracetic acid and immediately cE:ntrifuge~~~ are ~'OOOcr for 1.0 minutas at 4°C. The serum was then rec.over~=d, :>eparated =into small aliquot:s and stored at -8C°C urail ,a.:,~ay.
MCF--1 and Ii--_IR.II emyrne--1_inxec~ immanosorbent assay (EhISA) . MCP-7. concentr.~zt:ions were measured using a previously de:>cribed F;:fs_;:>A (Akoum A. et: al, 1996a) . This:
assay uses a mouse nuono:~orzal an'~i-hurn~zra MCf-i antibody (R &
D systems) anct a rabbi' Y:~oiyclon~rl ant:=~-human MCP-1 antibody.
This latter antibody dc:~c~:; not cross-react with several cytokines that are clo;;eiy related to I~~IC:P--l, inc_Luding NICP-2, MCP-3, IL-8, the prote:~.ru regulated on acts.vat_ion normal T
expressed and secreted RANTES) :end t:tze macrophage inflammatory protein-1 cx and ~ (MIP-:'_cx and MIP-1~3) (Haehicha M. et al, 199~~) .
sIL-1RII c:onc:ernt=rat:ior~:~ i_n serum were measured using an ELIS.F, devel.opcJc~ irv t:he Laborat:cry. This assay uses respectively ~. mouse m::~rioclonal anti--Y~urr~an IL--1RII antibody (R & D system's) for cal:~t:~zre, a ac>at: polyclona_'~ anti-human IL-1RII antibody (R & D svyst:erra) for det:ect.ion, peroxidase-conjugated rah>bit ants.-clc><~t. immunoglok>ul.ins (2ymed Laboratories, Inc. San Francisco, t:A) and 'TMB (3,3', 5,5',-tetramethylbenzid:ine) 't;i.o-Rad L<~bor-at~e;ries Lt:d, Mississauga, Ontario, Canada) as suk~~>I~r_ate fcr:v f:~er.oxidase. The optical density (OD) r~~as deter!rined at 450 nm, and sIL-1RII
concentration's were ca.lcu~ated b°,~ int=er~~clation from the standard curve.

2E: 85409-5 IL---ll~ a.nd I~~--la enzym<_ .immunornetr_ic: assay (EIA) .
The measurements were ,r.»~rformed ~.~:~ ing E iA cr~mmercial kit=s, accordin to l~he manu~~~~tu~er :~ ir~_~ =ru~~-ion.;
g ~ r ' ~t <~ (Cayman Chemical MI, USA).
Monc~cyte c,u~ .r.ure ar!d bic>lo~ic,3l assn-y. For these studies, we h<~ve ,zsed ~.~ hyst.iocyr.ic: ~.~e 1i line (U937) . Cells were culture <~t 37°C, '_~-" ''.G? in RPNCI medium supplement with 10o heat-inaci:ivated .fE~:,:~_L bovine :~e.rum fFB:.~) and 1%
antibiotics, and inc:ux~~~:_~d with L mM ::yv.ic adenosine monophosphate (CAMP) (~~~..c~rna, St. l~cuis) 1-or 48 hours to induce cell differentiation. ';ells were harv<.st-~ed by centrifugation, then distribut=ed in 2~l-sa~~l1 culture plat-.es at 10~' cells/ml/well- in fBS-tr:~E_c~ FtfMI medium :supplemented with different serum dilut_ions ;5, 1Ci and '>0'~) , and incubated .in duplicate for 24 hours at _,7''C. ~:u1tm:~e supernatants were then collected and frozen ~~t -80"C: until. assay. Thc.: b_ological assay was perf=ormed or! ~>>oo i ed sera f::rom norma 1 controls or from women wi.t=h endomet::-iosi.s ac,.:e.rd:ing to endometriosi:~
stage (I, II, and III-_~'J) . An equal vo.Lume c;f serum was taken from all the patients Inc_.luded iv each croup. The effect: of serum-induced MCP-1 se : _ a ~i.on was also :studied in the presence of human rIL-l-E~:L:I ('~ ug/ml) anc~ human rIL-lra 1;100 r~g/ml) (R & D systems, ~~!inneapo? is, MNi .
Statistical ~!:al,ys.is. Data are press=nted as mean ~
2~~ SEM. The unpa,'~red t test:: was used t=o compare the leve7_s of MCP-1 secretion or cir.:a.o:lat._i.ng c:ytok:ine:in women with endometriosis and conta-c1 :;object=s, ari;~ Bonf:err_oni correction was applying ~=or multi>lE~ c~ornpaz::ison:~. ~-'o compare the effect of one treatment on MCi-'--1 secretion, we used the paired t 3C~ test. Differences werF= :~on:~.idered a.~s statistically significant for P valu:~;~ 0.05.

Results We first measured sIL-1RII ~~oncentrations in the serum of nornr.al contr~~:ls and wor<~en with endorreetriosis stages I-II and III-IV. Stat::st_ical analv~sis of the results with the unpaired t test showed a significant de~.rease in the levels of sIL-1RII in the enclometriosis c:~roup as compared to the control grouch (P = 0. ~~~J ,~'E) . Furthermore, this decrease eaas found to be :>:ignificar_t in the initial stages of endometriosi:: (stages L-~II(P -- Ci.00~',' ;Figure 24) .
Statistical analysis c ~ our_ rata a.ccord:.~ng too the menstrual cycle phase showed no si.gnific:ant difference in sIL-1.RI=
levels between they prc L i fer~ative phase and the secretory phase, neithew in normal c<.>ntrol s, n~:~r_ .:.n endometriosis stages I-II o~~ III-IV. 1-3owever., .AIL--RII 1_evels were 1_', significantly higher a.n endometriosi_~~ st:.ages I-II as compared to normal controls boor in the pro-ii.ferat=ive phase (P =
0. 0184 ) and the secret:avy phase (P = 0 . ()261 ) .
We then measuoed the circu~_Lating :Levels of IL--1 in its tow forms IL-la an~,t .I=~-1(3. Figure 25 shows the 2C distribution of IL-7.a ~=m<~ .CL--1.~'> ~:onc:enl=r:ations found in normal and en:lometrios i.~~ women a~.:ccrdin.c) to t:he stage of the disease. Statz.stical aiaa i.f~sis of data with the unpaired t.
test showed no signifi~:~ant~ difference r_n IL-l.cx or. IL-l.~i concentration's between normal and endometrics:is women, 25 whether these latter were grouped t=oget:her (P = 0.955 and 0.667, respectively) o:c~ separated ~_nt~o ~ groups of endometriosis stages I-7::C i,P - 0.7.x_0 and 0.706, respectively) and III-IV ( P = 0 . 657 anc:~ 0 . 103, respect ivelyj .
Considering t:hE:e above results showing comparable 30 levels of cir~_ulating CI,-7. in normal and endometriosis women, but reduced levels of ~I:~~-iRII i;v women having the disease, we further assessed thc~ effect of- ~eri~>heral blood serum from women with and w-,.thoui c=:ndometrio;>is on monocyte activation, by measuring ~ICF-1 se~.: raetion by U~:~::~7 monocytic cells in response to t:hese sera. As shown in Figure '?E, serum from women with er~dometrio:iinduced a h.ighar secretion of MCP-1 by U937 cell: than thG:~~~ from nor.nal women (P = 0.018) .
Furthermore, this effc_.ct appeared to be significant in the initial ( P = 0 . 002 ) ray t tner than in t he mc:~re advanced ( P =
0.238) stage: of the c:iaease.
Hurrc<~n rIL-1F:.I T (_':i ug/rnl ) was then added to each pool of serum: prior tc:: incubatic~~n with 11937 cells. As shown in Figure 2.7, rIL-1.RI:I ,~.igni ficant 1y intnibit:ed MCP-1 secretion by 1J937 cells in .responsE~ to c::ontrol and to endometriosi~> worr~en-de r:i. ;reci sera . Howe~J<::r, t:he most significant i.nh.ib:itior: ::.;f MCP-1 induced secretion was observed in endom.etric: s i s stage I anc:~ I . ( P -- 0 . 0008 ) .
U93f~ cells weoe f~.rrthcer irucubat:ed for 24 hours at 37°C with each pool of ~:,~rum to which recombiruant human IL-lra (100 ng/ml) w~~s added. ::-sir data ~~e~>icteci in Figure 28 show a slight inhibit: ion of M~'P-- 1 secret ion induced by normal- as 2C well as by endometri.o~~..t> women-der:~ved ~aera. It is noteworthy that IL-lra inhibitory effect was, to some extent, more noticeable in serum fr:vm women with endometriosis stages I-II, but no statistical.l.y significant difference between IL-lra-treated and untreat~ee:~ sera was fount; (P = 0.101) .
Discussion Although ret;:c:>cJrade menstr.uat=ion is the most accepted theory for en :iornet.ri.osi, the pathogenesis of this disease is po~~~rly undee~~toc>d. Growth factors ~rnd cytokines that are secreted by a::t.a_~Tated immune cells have been implicated in the control. of the implantation and growth of endometrial cells outs:_c:le the uterine cavity, and in 12,9 85409-5 endometriosis-associat:eci c:Linical symptoms (Senturk L.M. et al, 199; Braun D. P. et. al, 1 994 ; ~lalme ~~ . et al, 1988 ) .
Cumulative evidences p:o;ant towards a relevant role for IL-1. The concent~aat.ions of ~ruis c~yLokine were found to .'> be increased :in the pf:. r:i t orue,~l ~: lu~.id ;>f women with endometriosis,, and ap~::ea.red to affect women's fertility (Fakih H. et al, 1987; Hill :~,~.A. e~ al, 1980; Taketani Y. et al, 1992) , According t:o oux, previcr:~s wtrzdie:;, both ectopic and eutopic endometria.'_ o~slls of women with endometrios~_s show an increased sent=itivity to IL-1 i~: vitro and secrete increased amounts of N~~:1~--1 in response to this cytokine (Akoum A. et ail, 1.995x; Akc>urn A. e2_ a:1,, 1995b) . Eutopic endometrial cells of wcarnen with endornel~riosis were further shown to express low lc~~.%~J1;_, of IL~-:l.F;L_L, a receptor that acts lc as a decoy fox' IL-1 an~:a L:irn:its I:h-:~-med~.ated ~~e11 activation, which suggested a defica.enr:y in their ab:.~ lity 1=a down-regulate IL-1 action (F~koum A. ~4t al, 200:Laa) . A soluble version for IL-1RII (sIL-1.RIIha~~ k:SF~en ident.if:ied in the supernatants from a number of diffe:reni~ cell types ;.'ymons J.A. et al, 1990; Giri J.~. et al, ~a90), ir: inflammatory synovial fluid (Arend W. P, et al, 199~~ ) , and i.n hurr.an sverurn (Eastgate J.A.
et al, 1990).
Based on the:-a evidences, we rrueasured circulating sIL-1RII in t~.e peripht~x~a~. blood of normal anc~ endometriosis women, and fo~.nd a sigr~a_iicant decx°ease in its levels in women having 'the ~;iisea:~~>. Furtherm_ore., IL-1RI:I levels were significantly reduced :i.rr endometriosis stages I-II, and that in both the p.roliferat:i.ve and the secretory phases of the menstrual cyc.Le, which ~>r:ovi.de e~~~ic:~er~.ce fo:r a deficiency in the regulation of IL-1 a~:tions at systemic lejrel in initial endometriosis stages.

IL--1 exists ~.r°: 2 fc>rms (IL-la and IL-l.~i) . Both forms of IL-_ trigger cell acti~~Taticn through the same functional IL-1RI:, anti ir~tera~ct wits the decc:y IL-1RII
although with dii feretvt aff:initie;~ (Borasc:h_i D. et al, 1996) .
'~ We therefore assessed the levels c.>f I:L-icx anc. IL-l~ in the peripheral blood and mound no sigr~ifican~ difference between women with arid without endometriosis, wizether these were grouped toget::zer or s~::(~az:ared :i.rotc> two ~r-oups according to endometriosi~ stages I-II and III:-:IV) . ~~'hest=~ findings point towards an unbalance in IL-1/Ii~-1RII c~ircula~ing levels in endometriosis, which rnay result in increased cell reactivity, and may account for tt;e reported act=ivaTvi.on of peripheral blood monocyt:cps in e~nc.ometr.~.osis women iaelu.er J.M. et al, 1987; Braun I:'. P. et a:i , 1996) .
1_'i To ~rerify tt:i_. hypotheses, we first exposed U937 monocytic cells to sera from normal an:~ endometriosis women, and estimated cell activation by measu.r~ng MCP-1 secretion by these cells. Our :resu~t;~ ~~7,.ear.ly siuowed that: the U93 7 monocytic cel.=.s produced larger amounts of MCP-1 in response to peripheral blood ser.;~ from women wish endometriosis. They further revea~_ed that such a serum-i.nduc:ed h9CP-1 secretion occurred in endometric;,,:Ls stages I-T::I, and that sera from women with more advanco~ endomet:rios:i_s :>tages (III-IV) had no significant ef=fect. These findings are ~_n keeping with our 2~ previous data showing ~ signifi.cant increase in MCP-1 levels in the periphera:I blood ;~f endometriosis patients, which, interestingly, was obs~.~.n~,Teci in init:i<~1 e-~r.dorne~Lriosis stages (Akoum A. et al, 1996b)" T)'3e~~ also indicate the presence in the peripheralblood of women wi;,h endometriosis of soluble mediators that. are capaal:ula of ac t:ivating monoc:ytes and stimulating MC:P-1 secrf~t~ion., and s~.aggest that such an activation prc>cess is ;;iependent. on endometriosis stage.
However, while activat=:~c~ mcnocytes are known to increase MCP-1 production (Leonard F; . ;1. et: a i , 1990 ) , it i s ~~till to be determined whether pe:iphe.ra~_ blood monocytes secrete elevated levels c~f MCI~-~ in endometriosi.s.
In orda;r to iY,vest=igate whether the reduced levels of sIL-1RII c:~:~served '.i~: the perprlerai blood of women with endometriosis may accc:unt for the increased reactivity of U937 cells t<:~~aarr_is encl.::n,etr_ic>si:~-women .:.lc>rived sera, we added human rIL-1RII to the different pools of serum prior to incubation wivh U937 cells. Interestingly, we observed that rIL-1RII sign::ificanr_1~ reduced L~CP-i ~~r~~duct=i.on by U937 cells in response to serum 'rc:m normal and endometriosis women.
However, the most s igr it i cant decrease was observed in women with endometz:iosis st~,de=s 1: and I1, whic:r~is in keeping with our findings showing reduced levels of sIL-1RII in the peripheral blood of trf-~:~e patients. 3ased on 'these findings and the fact t=hat circ:c.z.lati.ng levels of IL-l.a and IL-1(3 were comparable in women with,. and withcut endometriosis, it is therefore tempting too Ey~pothesiae that t:he enhancement of MCP-1 product~_on by mc:nc:cytes in response to endometriosis 2C~ women-derived sera i_s rn~;:re l:ikel.v duf-~ tca a dewrease in ~~IL-1RII levels than to an increase i_n t.hc_~:~e of I~-1(3 or IL-la.
However, the mechanisrr,s underlying s-IL-:~RII decreased levels in the peripheral bloo~:~ ~f women with endometriosis remain to be further clearly elu~_;:ic~ated. Gn the other hand, although IL-Ira appeared less eai~cient than rIL-1RII in inhibiting serum-induced MCP-1 sea.:2:w~t.:ion in our -in vitro assay, there was a noticeak>le tendency of inhibiticr~ in endometriosis stages I-II, which rrcakE-~> plausib 1_e a def icienc:y :in the down-regulati.on of IL-1 act:i_c~n in initial endcmetr:iosis stages of the systemic level.
In ~;ummary, ,:w_zr f findings ir~dir.ate that serum of endometriosis patient's w,:-zs able t_o i.rmluc:e MCP--1 secretion by 132 .35409-5 monocytes and re~Teal v:to.r- IL~-lRTi as a keys fac:-tor involved in that process. Monocyt~::- activation may play a significant role in endometriosis pathaip-tnysievlog~,~ as these cells were shown to stimulate endomet r:ial c:; E- 1.1 growt h and t ~:~ secr. ete numerous proinflammatory cytok.in~-: s that ma~.~ nave a deleterious effect on women's fertility. F,~rt~ermore, the reversal effect of recombinant :AIL-1RII ~r: MCP-i production by monocytes may be of a potential truerapE: u1 lc irntere:,>t .

REFERENCES
Adashi EY. The potent 1.a.'. relevancF= of cytokir~es to ovarian physiology: t=he E~mergi.nc~ role of resident ovarian cel'ys of the white bl~:~od c::e.ll. :e>c :i.es. Enc:aoc:~r R.ev 1990; 11:454-464.
Akoum A, Jolicoeur C, K)oarfi A, Aube M. Decreased expression of the decoy inte.rleui~~i r~-1 receptor type I I n human endometriosi:~. Am J F~:t;,ol 2001<a; 158:481--489.
Akoum A, Jolicoeur C, Kharfi A, Aub~ M. Normal human endometrium expresses the decoy interleukin-1. receptor type 1l) II and shows a defect~.ve expression of the receptor in women with endometriosi.s. Ar° ,T Pathol 2001b; 158:481-489.
Akoum A, Lawson C:, Mcc::oil S, V:illeneuve M. Ectopic endometrial calls rest>ond t=o estradiol by up-regulating the interleukin-1-induced :ir~terleukin-8 expression in vitro and express high .levels of interleukin-8 in vivo regardless of the menstrual cycle pl-~.ase. Mol Hum Reprod 2001c; 7:859-866 Akoum A, Lemay A, McCc:~ll ~, 'Turcot-Lemay, Maheux R. Elevated concentration and b:ioic:~gi~~ a,~tiv.ity of monocyte chemotactic protein-1 in t=he peritoneal flu-.~d of pa~~ients with 2C) endometriosis . Fertil :>teril 1996x; 66: 1 '-23.
Akoum A, Lema~~ A, Brur:e~t C, Hebert ~T. Cyt;okine-induced secretion of rnonocyte ~~l~emotac:ti~.J prote:irn-1 by human endometriotic: cells it ~_-:ult~are. ':rhe Groc:cpe d' Investigation en Gynecologie. ~~m J 0);>st:a:~t c:,yriec:ol 199:~a; x.72:594-600.
2~~ Akoum A, Lemay A, Brun..>r_ C, Hebert ,J. ;3ecretion of monocyte chemotactic ~>rotein-1 c~~Y c~;~tokine-st:imul_ated endometrial_ cells of women with ern:~ornet: ric>sis. Le c,r oupEe d' Investigation en Gynecol.ogie. Fertil :~f=f~::i=L 1.9:~5b; ~~~3: 322-~3?8.

1.34 85409-5 Akoum A, Lemay A, McC:~.'~ L :~R, Panac:li.~> I, Maheux E~. Increased monocyte chemotac,~aic ~:~rc_>tei.n-1 i_eVel. and act~..vit:y in the peripheral blood of w:.>rru:~n with enc:~omE:t.riosis. Lo Groupe d' Investigation en GyuE:colc>gie. Arr. .;. Obstet Gynecol S 1996b;175:1620-5.
American Fertility Soc:~i.ety Revise<:~ American Fertility Society classification of endc,metricsis: 1985. Fertil Steril 1985;43:351-:352.
Arcuri F, Ric:ci C, Ie1_ta F, (:intorino M, Tripodi SA, Cetin I, Garzia E, Schatz F, K:':.e.n.i P, Sar:topietro R, Paulesu L
Macrophage migration ..n:fuib.itc>ry factor in the human endometrium: axpnessi~:-n anc~ loc<~la_~zation during the menstrual cycle and early ~~regn~,n.cy. Bi.o.1 Reprod %001; F4 : 1200-120.5 Arend WP, Ma_l.yak M, Srith MF Jr, Whisenand 'TD, Slack JL, Sims JE. Binding of IL,-1 a-:pha, tL-1. beta, and IL-1 receptor antagonist by soluble IL-1 receptors an:~ levels of soluble IL-1 receptors ire sync>va al fluids. J Immunol 1994; 153: 4'766-74.
Arend WP. Interleukin--1 receptor antagonist: a new member of the interleuk:in-1 farru :Ly. .T ::lin Invest 1.991; 88: 1445-1451.
Arici A, Oral E, Attar E, 'Tazuke Sl, Ol~ve DL. Monocyte chemotactic p~-otein-1 ~.;w~ru~crutrat=:ion in ~~eritoneal fluid of women with enciometrio.::i_,: aznd its nuodula~.icn of_ expression in mesothelial cells. Fertil ~t~er.i-~ 1997; ~>J:1C?65-1072.
2_'~ Armitage P, Berry G, :_ t:.at ist ic:al. Nfet'_nod~ in Medical Research, 3rd ed. Oxford, I~onc~or:: Blackwe7..1 ;~c:i~an4: ific: Publications;
1994.
Bacher M, Met.<~ CN, Cal-~r~d.ra T, Mayor ~;, Chesney J, Lohof:f M, Gemsa D, Donnelly T, L:uca.la R. An c>ssent:ial regulatory role for macrophage mi.g.rat: i.<.:n i.nrwibito:ry f actor ir: T-cell activation. froc Nat_1. F~~~ad '~c;i ~r~ST~ 199E; '~)3: ~~ 849-7854 .
Badawy SZ, C:zenca V, v'rf~liech H. ;3t.efanu C~. E;ndometrial antibodies in serum ae~d ~~~er it:=on~~a 1. fluid of Infertile patients with and wituaut endometriosis. Fertil Steril_ 1990; 53 ; 930-~'_ .
Baranao RI, ?iazza A, Rumi L, Polak de Fried E. Predictive value of intcerleuki.n- .b in supernat<~nts of: human embryo culture. Fert:il :uteri ~uppl 19,r2; 58:53 (abstract) .
Bergh PA, Nav~~~t L. 'rhae .impact of- e:mbryo~ic; development. and endometrial m:~tux:ity < n the timinc:) of i:~n~;l.antation. Fer~il Steril 1992; 58:53?-5~~?.
Bigonnesse F, Zabelle Y, Akoum A. Tr_iphasic expression of interleukin-1 receptor type 1 in t.ne endometrium throughout the menstrual cycle of fertile and infect::i.le women. Fert~il Steri:l 2001a;!~:?9--8?
Bigonnesse F, Maroi;a c~'::, Maheux R, Akoum A. .nterleukin-1 receptor accessory prc t.-.~.-in ~ s r_c;ns titut .vely expressed in human endomet_=ium thro~.zqhout the menstrual cycle. Mol Hum Reprod 2001b; 7:333--3~'3.
Bloom BR, Bennett B. Mechanism of a reaction in vitro associated wit=h del.aycvd~-~y~~e hy~>ersensit_ivit:y. Science 7-966;
153:80-82.
Blumenkratz M~1, Glal_la~l~er N, Bashore RA.
2'. Retrogrademenstruation in women -..znde:rgo~.rig chronic peritoneal dialysis. Obst=et Gyneco:l 1981; 5? : 667-6'% 0.
Boraschi D, Bc>ssu P, Na~.:chi.a G, Ruggiero P, T:agliabue A.
Structures-function relationship in the IL-1 family. Front Sci 1996; 1:2f'0-308.

Bossu P, Viscont:i 1J, J-uc,giero P, I~Iacchia G, Muda M, Bertini R, Bizzarri C, Colagrande A, Sabbatini V, Maurizi G, Del Grosso E, Tac~liabue A,. ?,crasc:hi D., 'rransfected type II
interleukin-7. receptc_r impairs responsiveness. of human '~ keratinocyte::~ to rote. lceuki.n-1. Am J Pathol 1995; 147:1852-1861.
Braun DP, Dmcr~ski WP. ~.ndornet.rios~.s: ab:~~ormal. endometrium and dysfunctional. immune ,-espouse. t~m:r Opi~ Obstet Gynecol 1998; 10:365-~:~.
Braun DP, Gebel LI, Ha:_~se R, Rana r;1, Dmowski WP. Spontaneous and induced synthesis of cytokines by peripheral blood monocytes in patients with endometr_iosis. Fertil Steril 1996; 65: 1125-'~.
Braun DP, Mur:iana A, c=-E>>~>e.l H, Rotm<~r1 ~;::, Rana N, Dmowski WP.
Monocyte-mediated enhan~~ement of endometr_ial. ~e11 proliferation in womer- ~h~itr~ endc>rne ~r. ios.i_.. Fert.il Steri'~
1994;61:78-84.
Calandra T, Bernhagen .1, Metz CN, Spiegel LA, Bather M, Donnelly T, Cerami A, E3ocala R, MIF as ~-r glucocorticoid--induced modulator of cytokine production. Nature 1995;377:68-71.
Calandra T, Bernhagen J,, Mi.tchel.l RA, Bucala R. The macrophage is an import;~nt and pre-,fi.ou:~_~y unrecognized source of macrophage migratic;ri .in~:i~~itory factor. J hxp Med 1994; 179:1895--190?.
Cannon JG, Dinarello C",~~" :Lricreased ~:~:Lasrla interleukin-1 activity in women after ovulation. Science 1985; 227:1247-1249.

137 (35409-5 Chesney J, Mr:tz C',N, B,_i~.~Yuer M, Pence 'r, Me:inhardt A, Bucala R.
An essential roles for uiacrophag~: n:i:igrat ion ~rihibitory factor (MIF) in angi_ogernesis !:rid r.he grov~~th <,f ,-~ murine lymphoma.
Mol Med 1999;5:181-19 '~ Colotta F, Dower SK, : ircis ~.TE, Mant.ovani A. The type II
'decoy' rece~>tor: a n~:--vE-~1 regul_ate~ry pathway for interleukin 1. Immunol Tc>day 1994, 15:562--566.
Colotta F, Rce F, Muzic; M, Berti:ii R, Polentarutti N, Si:roni M, Giri JG, l7 ower:v SK, ~_-ms ~7E, ~~larvtovani A. Interleukin-1 type II receptor: a dc>~~c?y Target for IL-~ that is regul,~ted by IL-4. Science 1993; ~ 61: 472--4~?5.
Coulter KR, W~wer:vs MD, ,owe MP, Krv_oell ~I~: Extracellula:r regulation of inter:leok-an (II~)llbeta through lung epithelial cells and defective II_~-i type II receptor expression, Am J
l.'s Respir Cell Mol Bio1 9~~9, 2():964--974.
David JR. Delayed. hype:rsensitivstt% in vitro: its mediation by cell-free substances ~ ;~rmec-1 by lymphoid ael:i-antigen interaction. Proc: Nat ~':c:ad Sci_ U>A 1996; 56: ~'2-77 .
Dinarello CA. Int.erleiak_in-1 and it.s biologically related cytokines. Adv Immuno . 1 98'.x; 44:1.'>3-205.
Dinarello CA. Biologii a~as:is for interleukin--1 in disease.
Blood 1996; .37:2095-2.4 Dizerega GS, Barber Di.~, Hoc~gen GD. Endometr.iosis: role of ovarian steroids in ir; iaiation, rr>aintenanc:e and suppression.
2:~ Fertil Steri:l_ 1930; 3 :: k-49--6.':,3.
Dmowski WP, c:iebel- HM, Braun I)P. The role of c:~ell-mediated immunity in pathogenesis c( endorriet:rios:is. Ac:~ta Obstet Gynecol Scanc~ Suppl 1:x94; 159: 7-14..

Donnez J, Smoes 1', Gi :L ~ ~e.ret: ), ~:'a..ana-Roux t', Niso.lle M.
Vascular endothelial ~:(rcwth facto- (VEGF) in endometriosis.
Hum Reprod 1098; :1.3: 68~>--e~GO.
Eastgate JA, Symons ~~~'a, Duff GW. Identif:ication of an ~ interleukin-.. beta bir~~ding protein .in human F;lasma. FEBS Lett 1990;260:213-6.
Eisermann J, bast MJ, Pineda J, Odem RR, Collins JL. Tumor necrosis fact:sr fluid ~~~t women ~.znc~ergoing laF:~aroscopic surgery. Fert:il Steri=. 988; 50:5%3-579.
Fakih H, Bagg~°tt B, H~;.lt z c:~, Tsanc~ KY, :.~ee ,1C, Williamson HO.
Interleukin-1: a possa_ble role in the infertility associated with endometr°iosi.s. FF:rt z,.1 Ster:i:1 1987; 4'i :2:L:3-217.
Ferriani RA, Char~nock-~.lones DS, Prentice A, Thomas EJ, Smith SK. Immunohisvochemic~ll localivation of acidic and basic 1.'i fibroblast growth. f<:~ci ~:~rs ;un normal humaru endometrium and endometriosis and the d~stecti.on of their mRNA by polymerase chain reaction. Hum R~:prod i993;8:1~~-16.
Fleiss JL. Statistical Methods for Rates and Proportions, ed 2. New York, Viley, 1~~~3:1.
2C1 Folkman J. Aru~iogenesis in cancer, vas~:o.iar, rheumatoid and other disease.. Nat Mec; 1995;1: 27-31.
Frey CH. The ~=amilial occurrence of endornetriosis. Am J
Obstet Gynecol 1957; ~~ '3: ~~ 18-419.
Fujimoto J, ~~akaguchi fit, l~irose R, T<3m_~a 'T. Expression of 2~~ plateletderived endotrve:lia:l. cell growth fact:o:r (PD-ECGF) related to angiogenesis in ovarian endometriosis. J Clin Endocrinol Met:ab 1999; 84: 359--=i62.

Gebel HM, Rana N,, Bra .are PB, Dmows'~i VvIF. DiffE.rential expression of VLA(31 (c::C'<~°) cn monc=uytes from patients with endometriosis. Arn J R~_~~;rvc~d Irnmu:uol 19~~5; 34:317-22.
Giri JG, Newton RC. Her-~~;~ R. Identification of soluble interleukin-:L binding I:totein ire: cell-free supernatants:
evidence for soluble i.rot erle~zkir:-1 receptor. J Biol Chem 1990; 265:17416-::L7419..
Gleicher N, Dmowski W=, Siegel I, Liu TL, Fr_i.berg J, Radwanska E. Lymphocy : a subsets ire er.~'.cmet=r:ic>sis . Obstet Gynecol 1984;63-463-F~..
Gleicher N, hl-Roeiy a'=a, C:onfino E,, Fr.iberg J. T~>
endometriosis an auto in~~a:ne disE:ase? Obstet Crynecol 1987;
70:115-122.
Goetzl EJ, Sreedharan ~~ F- . Mediat=o ~s of cornmuriication and adaptation in thfe neu:;:cendocrine and immune ~>ystems. FASEB J
1992; 6:2646--2652.
Gordon JD, Shifren JL,. ~'oulk RA, ~I'aylar RN, a.'affe RB.
Angiogenesis in the h~am.an female reproductiv~:~ tract. Obstet Gynecol Surv 1995; 50: ~8t~-697, Granowitz EV, Clark B.=>, Marlr~illa ~7, D.inarello CA.
Interleukin-:L receptor ~r_tagc~nist competil~ivc=ly inhibits the binding of inter:ieuki-:u-1 tct.=he type I I inter leukin-1 receptor. J I3iol Chem 1~~91; <'?66:1~'-1147-14150.
Greenfeder SA, Nunes e, h:wee L, L<~bow M, Chizzonite RA, Ju G.
Molecular cloning and c:r,aracterizatior: of a :second subunit of the interleukin :1 rec~xp nor compl.e:~. J Bio=L Chem 1995;
270:13757-13'765.
Grosskinsky :JM, l:alme ~t. Endomet r:iosz.s : the most: response.
Bailliere' s ;Jlin Obst~et~ Caynaecol i 99 3; 7 :'701-- 7 13.

Groves RW, Sherman L, Ni:i zutani f-1, Dower SK, Kupper TS.
Detection of interleu~~~.r~-i receptors; 1.n human epidermis.
Induction of the type I ~ rE:c:eptc::-r after organ culture and in psoriasis. Arn J Pathoi 1.994; 14'_~>: ~G48-1456.
Guzick DS. B:~sic statist.:ical testing, ir~c=Ludi_ng interim analysis. Semin Repro~:~ r,ndocrinol 1996; 14:133-137.
Hachicha M, Ratha:~nasw~:~rr~..i. P, ',~chall. fJ, McCo.l1 SR. Production of monocyte :~hemotactic protein-1 in human type B
synoviocytes.. Synergi;~t.:i..c effect. ~~f tumor nec:-rosis factor alpha and interferon-yamrna. Arthr-i_tis Rheum 1.993;36:26-34.
Halme J, Becker S, Wi.o~g H.. Pcceroti.zate~ cyclic: activation of peritoneal macrophage> a_n patients with endometriosis. Am J
Obstet Gyneco_L 1~:~84a; 1.48:85--9G.
Halme J, Becker S, Har-unond MG, Ra j MHG, Raj S . Increased 1~ activation o1= pelvic r:~ac:.rophage~: 1-n infertilE: women with mild endometriosis. Am J Ok~sr_.et Gynecol 1983; 145:333-337.
Halme J, White C,. Kaunoa S, Ester ~;J, Haskill '. Peritoneal macrophages t=rom patiE_~nts with endometriosis release growth factor activ:i_ty in c~i'::-zc!. J C:lin H:~ndocr:inol Nfetab 1988; 66: 1044---1049.
Halme J. Hammond MG, I-!ulka JF, Raj SG, Talbert LM. Retrograde menstruation in healti:y women and in patients. with endometriosi::~ Obstet ;~yrnecol 1984b; 64:151-1.'4.
Halme J, Mathur :>. Locval autoimmunity in mild endometriosis.
2.'~ Int J Fert.il 198~r; 32:::.09-311.
Haney AF, Jenkins S, vv'einberg JB. The stimulus responsible for the perit:~neal fluica inflamrrlat::ion observed in infertile women with endometrio:~it-~. Ferti.l Steril 1991; 56:408-413.

Haney AF, We:i.nberg JB . Reduct~ iov~ c>f the int raperitoneal inflammation associate:~.a with endometr:iosws by treatment with medroxyproge:>terone a;:~~~t:at~~. Arr. ~ Obstet_ Gynecol 1988;
159;450-4.
.'~ Haney AF, Mu:~cato JJ, ~:~'einberg ~lB. Peritoneal fluid cel.1 populations :i_n irofert:.l.ity pati.ent:s. fertil ~teril 1981;
35:696-698.
Harada T, Iwa.oe 7:', 'rea akaw<~ N. F:o~_e of :yto,~cines in endometriosi;. Fert.il Steril 201; 76:I-10.
Healy DL, Rogers PA, ~3i _. L, W:inc)f_~_eld M. Angiogenesis: a new theory for em~ometrio~is. Hum Reprod Up:iate 1.998; 4:736-'740.
Hill JA, Anderson DJ. C:ymphocyte. ~~ctivit~,~ in the presence of peritoneal fl. aid .from fertile wc.~men and :i_nfertile women with and without E~:ndometric~sls. Am J Oi~si:et ~:ynecol 1989;161:861-4 .
Isaacson KB, Coutifar~ ;J C, Garc:ia CR, L~~tt.le CR. Produce=ion and secretion, of comp.uernent component 3 by endometriot:ic tissue. J Cli.n Endocrincl Metab 1989; 69:1.003-9.
Isaacson KB, Galman M, C'c>utifaris ~~, Ly~tl.e CR: Endometrial synthesis anc:l secretion of complement component-3 by pal=Tents with and witt~:~~ut endoniet riosis. Ferti:l ::>t::er:i_L. 1990; 53: 836-841.
Jolicoeur C, i3out:ou.il (~n, Drouin R, Para;:3s.s .C, Lemay A, Akoum A. Increased expression of monoc~yte chernotactic protein--1 in the endometrium c>f wo:ner~ with erv,dometriosi.s. Am J Patho_L
1998;152:125---133.
Kauma S, Matt D, St:ron~ :,, Eierman D, Turner 'f . Interleukin-1 beta, human l~Jukc>cyte antigen HhA--DR alpha, and transforming growth factor -beta ex~~rcession in e~ndomet:ri.um, placenta, and placental mernhranes . ~,rr~ J ubstet Gyneco 1 1.990; 163: 1430--143? .
Kharfi A, Akoum Fa. Cor-~.-plat=ion between :~ec:reased type-I=L
interleukin-~ receptor and increased rnouocyte chemotactic _'~ protein-1 exp:rest~ion :i n the endcmet:rium of women with endometriosis. Arr. J RE prod Irnmunol 2001; 45:193--199.
Khorram 0, Taylor RN, Ryan IP, Schall T~J, Landers DV.
Peritoneal f~_uid sconce rut rat:i.ons of- t:he ~~ytoki ne RANTE S
correlate with the se~;r~r ity of= ma.ometr...osis. Am J Obstet Gynecol 1993; 169: 154'- --1 54'3.
Koch AE, Polverini 1'J, hunkel SL, 1-~ar'~ow LA, DiPietro LA, Elner VM, Elner SG, Str:i_~~tE~r RM. Tnt:erLc>ukin-8 as a macrophage-derived meci.,~ for c~ f angiogenes is . Science 1992; 258 : 1798--1801 .
l~~ Kokorine I, N_-solle M, Donnez J, Eeckhout Y, ~.Jourtoy PJ, Marbaix E. Expression :>:F interstitia:L c~cllagenase (matri_x metalioproteinase-1) i w> re:i.ated t:c the activity of human endometriotic lesions. 1~~ertil. St~eril. 1.!a~~7; 68: 246--251.
Korherr C, Hof=meister N,,. Wesche H, )~'a_Lk W. A critical role 2C for interleukin-1 recel>t::~r accessory protein in interleukin-1 signaling. Eur J Immun~::~l L997; 2? : 26a?-'?6 r .
Larrick JW. N~~tive i.nt~:>>~.Leukin 1 inh=ibitors. Immunol Today 1989; 10: 61-E~6.
Larsen C, Zachariae C, C~~~.zkav da N, Anderson A, Yamada M, 25 Oppenheim J, Matsushim.=~ K. fr_oin~=larnmat;ory cy~_okines interleukin 1 and tumor necrosis factor_ induca,~ cytokine~; that are chemotacti.c for neut:rophils, 'T cel.:Ls and monocytes. Prog Clin Biol Res 1990; 34'x;:%119--431.

Lawrence JB, ;ginger Rf~:. Quanti.tative anal.ysi.s of in situ hybridizatiorv. methods fir !:he detect=ion of actin gene expression. Nucleic Acids Res 1985; 13:1777--1799.
Lebovic DI, Eentzien L', ~~'hao VA, Gar_r~~t~:: EN, Meng YG, Taylor .'> RN. Induction of an ar~:~iogenic L:~he~notypE~ in endometriotic stromal cell cultures (~y irlt:erle=~uk:irn-1bc_:ta. Mol Hum Reprod 2000a;6:269-275 Lebovic DI=, :~hifren J~ , Kyan IP, Nfueiler MD, Korn AP, Darney PD, Taylor RN.. Ovarian :steroid ~rnd c::ytoLcine modulation of human endometrial arng~ «c7enc>~; i ~~ . uurn Rep~~od :;uppl 2000b; 3 : 67-77.
Leiva MC, Hast:.y LA, Pf c~ i fer S, Mas.trc:~.i~nni L, Lyttle CR.
Increased chemotacVtic ~a~~tivity of peri-oneal_ fluid in patients with endometri.osi:~. Am ..T Obstet c~yne;~ol 1~~ 1991; 168:592-f3.
Lemay A. La place de J:r suppress.ion ovarienne dans 1e traitement de 1'endomet:riose. Ref Gynecol Obstet 1993;1:243-50.
Leonard EJ, Yoshimur_a 'a' . Human rnonocyte chemoattractant 2C protein-1 (MCf-1) . Imrr~.nlol Today 1990;:L:97-~1J1.
Lessey BA, Castelbaum A..J, Sawin SW, Buck CA, Schinnar R, Bilker W, Strom BL. Abu~rrant~ integr_in expression in the endometrium of: women w.t.~h endometr~.~os:i_s. J C:lin Endocrinol Metab 1994; 79:643-649.
2~ Lessey BA, Damjanovich 1::, (:'outifaris C, Cast.elbaum A, Al.belda SM, Buck C'A. I=ntegri.n <ac~fnesi.on rru:~lc~cule:-: in the human endometrium: c:orrelati~:7.o with thc~ rnormal and abnormal menstrual cyc7.e. J Cli:.u Invest. J..992 ; 90: 188-I_95.

Liu DTY, Hitc:hco<:~k A. E:r.dometric:.si.s; its assc:,ciation with retrograde mF~nstruatic::r~, dysmenc::r~voea, end tLibai.. pathology.
Br J Obstet Gynaecol (U86;, 93:85'x-862.
Malinak LR, Tuttram Jt.%C, Elias S. Heritable aspects of .'~ endometriosi:~ II; cliriaal cruaract:eristics of familial endometriosis. Arn J Ox~sr~et Gynec:o-~ 1980; 137:332-337.
Mathur SP. Autoinununit-y in eridometriosi.s: relevance to infertility. .elm ~I Repa .~d Itnrnuno.:l x:01)0; 44 : 89-95.
Mehrabian MRSI, Sparkes TM, F'ogelman AM, Lusi.s AJ.
Localization ~~f monocyte clnemotact.ic protein-1 gene (SC'~'A2) to human chromosome l~ra~_1..2-q21.1. c;enomics 1991; 9-200-203.
McLaren J. Vascular er:dothelial growth factor and endometriotic: anqioger.e;is. Hum Re~prod .lpdaiVe 2000; 6:45--55.
McLaren J, Prenti.ce A, ~~'harnock--Jc>nes D~~, Mi.llican SA, Muller KH, Sharkey AM, Smith 5~;. Vascular endovhelial growth factor is produced b;~ peritor~e:~,l fi_uid macrophages in endometriosis and is regulaved by oT.~arian steroids. J Clin Invest 1996; 98:482-4139.
McMahan CJ, S:Lack JL, Mosley B, Cosman J, Lupton SD, Brunton LL, Grubin Cf.,, Wignal.l ~.M, Jenki.nt. NA, i3rannan CT, Copeland NG, Huebener K, Croce CM, Cannizza.ra L.?~, Benj.amain D, Dower SK, Spriggs MK, Sims ~ a. A r~ove_L LL--:1 rE-~c~eptor, cloned from B
cells by mammalian ex~ r_ess:ion, i.s expressed in many cel l_ types. EMBO ~T 1991; 1( ::- 821--283a?.
2_'i McMaster MT, 7~im KH, ~'ay.Loz RN. Inumunob:i_ology of human pregnancy, Curr Probl Dbstet Gynecol fert~il 1998; 21:5-23.
Metz CN, Bucala R. Rope of macrophage migration inhibitory factor in the regulate or,~ o t_ the :immune ~ espouse . Adv Immunol 1997;66:197-223 Mori H, Sawairi M, Nal<:aGawa M, Itoh N, ~Iada K, Tamaya T.
Expression of interleu:k~n-1 tIL--1:~ beta messe-nger ribonucleic acid (mRNA) a;c:nd L:L-1 ~ ~:~.eptor ar~tagon~.s. mRNA in peritoneal macrophages from patients with ~:ndornetri~~sis. Fertil Steril 1992; 57:535-w542.
Mori H, Sawai..ri M, Na~:.agawa M, Itoh N, Wada K, Tamaya T.
Peritoneal fluid :intei~:leukin-:L ~_~eta and t umor necrosis factor in patients with benign gynecologic disease. Am J Reprod Immunol 1991; :? 6 : 62-'7 .
Mueller MD, V:igne JL, Minchenko A, hebovic DI, Leitman DC, Taylor RN. Regulation of: vascular endot)nel.ial growth factor (VEGF) gene t==anscript eion key estrogen reaceptors alpha and beta. Proc Nail Acad ;::,c1 USA 2000; 97: 10'a~%2-10977.
Mulayim N, Ar_Lci .A. Tk~e re.l_evance of: t_nEe peritoneal flu_~d in 1'_~ endometriosis--associatk:~~:3 infertility. Ht.zm Reprod 1999; 14:67-76.
Neumann D, Kollewe C, Martin MU, Boraschi D. The membrane form of the type II II~~--1 rec:~eptor accounts for inhibitoi:y function. J Ir~ununol 2C01~; 165:3350-335'7.
Nishihira J. MacrophaGe migration :inhibi.tory factor (MIF) its essential. role in t?:e i.mmune system and cell growth. J
Interferon Cyt:okine Re,:~ a_'()00; 20: '151.-'7E~<.
Nisolle M, Ca:~anas-Rou;, H , Anaf '~, MinE~ JM, Dc~nnez J.
Morphometric study of t:he stromal vascu_arisa~ion in peritoneal erodometriosis. E'ertil. Ster_i:L 1.993; 59:681-684.
Nisolle M, Ca:~anas-Roux F, Donnez J. Immunohistochemical analysis of pnoliferat:~.~rE:~ activit=y anc~ steroi_c~ receptor expression in per.itonea:l. and ovarian endometriosis. Fertil Steril 1997; E~E3: 91'2-91~-a Noble LS, SlICIJSOrI ER, ~Jc,~hns A, I-3u~.un SE. Aron~:atase exprf=ssion in endometrio.sis. J Cir: Endocrinc.~1 Metab 1996; 81: 174-179.
Noyes RW, Her_tig RocJk ~J, Dating tree end:~metri.al biopsy.
Fertil Steril 1995; 1:3--25.
Ogawa H, Nish:ihira J, Sato Y, Kondo M, T akahashi N, Oshima T, Todo S. An ar~~ibody fc~r: ma~~rophagE~ migration inhibitory factor suppresses tumour growth and inhabits tumour-associated ar.gioaenes_i. s. ~~rto k:ine 2000; i~ : 309-314.
Olive DL, Mon1=oya T, H i_c4hl RM, Sc;henken RS. Macrophage-conditioned media enhan~~:e c-~ndome~rial s~rcmal cell proliferation i.n vitro. Am ~J Obstet Gynecol 1991;164:953-958.
Oosterlynck D,J, Corm :i.:i;~ fJ, Wa<~r M, 'Jandeputte M, Koninckx PR. Women with endomet.r.:i,:~s~.s show a defect in natural killer activity resulting in a decreased cytotoxicity to autologous 1'~ endometrium. Fertil StF~:ri.1 1991; 56:45-'~1.
Oral E, Olive DL, Arid A. 'fhe peritoneal environment in endometriosis.. Hum Re~:r.°cd L,7pdate 1996;2:385-398.
Orlando S, Sironi M, E_i_anchi G, Drununond AH, Boraschi D, Yabes D, Mantovani_ A. E~tc.rle of rneta:ll.op:rcteases in the release of the IL-1 type II de!:r~y receptor. J 13~iol C'.h~rm 1997;
272:31764-31769.
Ota H, Igaras~hi S, Hay~.al~:;~wa M, Mat:>ui T, Tanaka i-~, Tanak:a T.
Effect of danazol on t.ue immunocompetent- cells in the eutopic endometrium in patient's w:itah endomet.riosis: a multicenter cooperative study. fert;:i.L ~teril 1~~~!6; ~~': 545-'351.
Plata-Salaman CR, Ilyi.rv SE. Interleukin-1 (IL-1) beta-induced modulation of the hypor::al::rlarnic I~-iber_a systern, tumor necrosis factoralpha, a:~n~::~ transforming growth factor-beta 1 mRNAs in obese (fa/fa) and lean Fa/Fa;c Zucker rats:

implications to IL-lbet_a feedback systems and cyt~okine-cytokine interactions. J C~eurosci r?es ~~97; 4~~: 541-550.
Psychoyos A. The implamt::at~ior~ wir~d~w: bass_c and clinical aspects . In: Nfori T, ~f'::~mi.naga T, H i.rc;i M i eds . ) , Perspectives on Assisted Repro~~ucti:url.. Kyoto, J:~p~ir~: Ares ~~erono Symposia;
1993: 57-62.
Rana N, Braun DP, Hous:iR, Gebel H, Rc>t.n!an C, Dmowski WF~.
Basal and stimulated ~;,c:~uet~ion cw cyt:oi<:ines b~.~ peritoneal macrophages in women w:i_ r:n endornet.ri.osis. Fert:Ll Steril 1996;65:925-30.
Rock JA. The revised A,nerican fer til~t~y Society classification of endometriosis: reproc~uci:bil:ity of scoring.
20LADEX Endometriosi.s ;~T::udy Croup. FE,r_ti l Steril 1995;
63:1108-1110.
Rock JA, Markham SM. Pat.:rogenesi.s of er~dometri_osis. Lancet 1992; 340:126-1271.
Ryan IP, Tsenq JF, Sch.r_ioc,k ED, Khorram 0, handers DV, Taylor RN. Interleuki.n-8 conc;~.mtration~~ are elevated in peritoneal fluid of women wit=.h er~c:lc~rnetriosis. I'ezt:i1 Steril 1995;
2C 63:929-932.
Sahakian V, Anners J, Lla:ki1_L S, Hal.mc_: ~J. Sel~°ctive localization of interl ~:::ukin-.L receptor antago:iist in eut:opic endometrium and endomet::.rioti.o implants. Fert.il Steril 1993;
60:276-279.
2~~ Sampson J. Peritoneal ~n.~:~ome~ric>sis due t.o menstrual dissemination of endornet~ria:1 t:.issue :in-.c; the peritoneal cavity. Am J 0bstet Gynecol 1.927;14: 4>'~--469.
Schall TJ. Bz.ology of t~l~~e RANTES,~S Iv> cyt:.okine family.
Cytokine 199~.;~ 3: 165-C 3.

Senturk LM, Arici A. imrrmno:logy of endornetriosis. J Reprod Immunol 1999; 43:67--8::
Sharkey AM, Day K, Mc:Ptlc_;rson A, Mal.i.k S, Licence D, Smith SK, Charnock-JonE~:~ DS. Va:~c,~.Lam endc~thc~l.:ia.1 growt=iz factor ~~ expression in human er::lomet:r:ium is regiz_~ated icy hypoxia. J
Clin Endocrinol Metab a?c,Di); 85: 402-409.
Sharpe-Timms F~L. Fndorr.ai:: r:ial anomalies :~_n women with endometriosis. Ann NY .~;~ad :~ci_ 2i)01;'~4:3:131-47.
Shaw RW. An atlas of ern:~ometwios:is. New York: Parthenon 1C Publishing Group 1993; ,.' 3-~..., .
Sheth KV, Roc~r GL, A1-a<~c~air_y S'1', Parhar RS, Hamilton CJCM, Al-Abdul Jabb~ir F. Pre~::licaion of succ:es:~fui embryo implantation by measur:i.rig interleukin-:l-alpha and immunosuppres~~ive fact~:_u:~ (s; in p~-evyrnpl_antati.o.z embryo culture 15 fluid. Fertil Ste.ril 1 99_1.,; 55: 952--95~i .
Shimizu T, Abe R, Nakam,.i:ra t-1, Ohkawara ~~, Suz~aki M, Nishihira J. High expre~>sion of m~~crophage migrat=e cn intzibitory factor in human melanoma cel.l.s .end it.s ro-ie irn tumor_ cell growth and angiogenesis. Biochem :~_Lophys Res Common 1999; 204:751-758.
2C Shlopov BV, Gumanovskayr~ MI.~, E-~ast:y KA. Autocrine regulation of collagenase 3 (matri<~ metal.loprote=i_rnase 13j during osteoarthriti:~. Arthritis Rheum '000; 4:195-?05.
Siegel S, Cast:ellan N~T: Nonparametric ,3t~atist.ics for the Behaviour of :''>cience, ~.~d :? . Edited by ,i C> Anke r, New York:, 25 McGraw-Hill, x.988.
Simon C, Cabal_lero-Camr~u P, C~arcia-V'elasco JA, Pellicer A.
Potential implications ,:~ ~ c.hemok_nE~s i.n reproductive function: an attractions idea. J Reprod IrnmunoL 1998a; 38:169-193.

Simon C, Frances A, L~c; 13Y, Mer_c:~der ?~, I:uynh T, Remohi J, Polan ML, Pel__Licer A. oT~munohist.ocnemi~~a1localization, identification and oecu?,at:ion of true i:zte.reukin-1 receptor antagonist in the humUrv~ endometrium. hlo~ Hunr Reprod 1995a;
~~ 10: 24 '2-2477 .
Simon C, Frances A, Piquette GN, El Danasouri I, Zurawski G, Dang W, Polan ML. Embr~,,;::~n.ie~ :impl_antat.ioru in mice is blocked by interleukin-1 recel:t:c:r anvaganist. Endocrinology 199~L;
134:521-528.
Simon C, Gimeno MJ, M~ rcJader A, 0' Connor JE, Remohi J, folan ML, Pellicer ~~. Embrycro:i_~~ regulation ~:~f integrins beta 3, alpha 4, and alpha ~ i n t~rurnan endome'.r Lal e~>ithelial cells in vitro. J Clip Endocrir;::>:1_ Met:ab 1.997; 8?:2607--2616.
Simon C, Moreno C, Rerr,~h.i ~I, Pei:Lice:r Ae Cytokines and embryo l~~ implantation. J Reproc LTrrrrrunol_ 1998b; 3~~:117-131.
Simon C, Pell=_cer A, fr~:Lan ML. lnter:Leu><:in-1_ system cro;~stalk between embryo and enc:>rnetr:ium 1.n imp.1<~nt:anon. Hum Reprod 1995b; 10:43--_'>4.
Simon C, Piquette GN, h:_anc_~es A, Pol.an ML:. Localization of interl_eukin-1 type 1 r ~;~eptor and inter? euki_n-1 beta in human endometrium throughout i::he menst:rual_ cy<~le. J Cl.in Endoc:rinol Metab 1993; i'7: 549-5_' 5.
Simon C, Valbuena D, Y russel J, F3erna.l. A, Murphy CR, Shaw T, Pellicer A, Polan ML. ~=nterleukin-:1 receptor antagonist 2~~ prevents embryonic ~_m~l~ntat:ion by a ~:irect effect or. the endometrial epitheiiun.. Fert:i_1. :~i-,er_i_1 1'-a~38c; 70:896-906.
Sims JE, Gayl.e MA, :>lacl, ~JL, Alderson hIR, Bird TA, Giri JG, Colotta F, Re F, Mantc:zTan:i A, Shanebeck )'., Grabstein KH, 1.50 85409-5 Dower SK. Interleuk:in l_ si~:~nalirig occur._> exclusively via the type I receptor. Proc N~~t1 Acad Sci. CJ S A 1993; 90:6155--6159.
Sims JE, March CJ, (;o_:rnan D, W:idme~:r MB, MacDonald HR, McMahan CJ, Grubin CE~,. Wignall ~~M, Jac:ksor,. ~1L, i~:a.l1 SM. cDNA
_'i expression cloning of t=lie aL-1 receptor, a member of the immunoglobuli_n superfami.ly. Science 1983; 241:585-589.
Somigliana E, Vigano 1~, i.;affuri. B, C~ua:rneri D, Busacca M, Vigna7_i M. Human endor~ev: ria.7_ strorr<~:1 ~.~:f:~_~is as a source of soluble intercellular a;~hesion mol eculc~ ; ICAM) -1 . Hum Reprod 1996; 11:1190--1194.
Strathy JH, Molgaard C H,, ~lc>u:lman CB. Enc_iomet:r_ iosis and infertility: a laparoscopic study of endometriosis among fertile and i.nfert:ile wt~~nen. fern=L Steril 1982; 38:6E~7--672.
Symons JA, Eastgate JP., Duff GW. A so.Luble ~;inding protein specific for interieukirn 1 beta is p:roc~uc:ed by activat:ed mononuclear cells. t:yt:~kine 1990; 2: 1.90-f7.
Symons JA, Young PR, Du.r:E f C:~W. Soiubl.e tape II interleuki.n 1 ( IL-1 ) receptor binds .-rnc~ blocks prc_~cesi ng o f IL-lb precursor and loses afrt_:i_~n:ity for I~~-:l. wNecept.or antagoni~>t.
Proc Natl Acac~ Sci LJ ~ A 1995; 9?:1 7:14--_C 718.
Tabibzadeh S. Human en~~~tmetr:ium: an ac:i_ive site of cytok:ine production ar.c~ action. I_nc~ocr Rev :1991.; 12:27:?-290.
Tabibzadeh S, Kaffka KL,, Satyaswaroop PC~, Kilian PL.
Interl.eukin-1 ( IL-1 ) rc:ecyn:lat=ion of hi.zman enclomet vial 2~ function: pre;~ence of ~I--1 receptor c;or relates with IL-1.-stimulated prc>staglandi_n E2 product:ion. L~ Clip Endocrinc>1 Metab 1990; 70:1000--100E:>.
Tabibzadeh S, Zupi F:, Babakn:ia A, Liu R, Marconi D, Romanini C. Site and rrcenstrual ~:~~;ale-dependent c~~>pression of proteins :L5I. 85409-5 of the tumor necrosis factor ('L'Nf) receptc~r family, and BCL-2 oncoprotein and phase---s~ecifi c prc>ducti;~n of TNFa in human endometrium. 1-hum Reprc~d 1.995; 10: ~.7~7-286.
Tabibzadeh S. Molecular control of t:he implantation window.
_'> Hum Reprod Update 1_99f: ; 4 : ,9 65-411 .
Taketani Y, Kuo TM, M:i ;.ano M. C:om~~arisori of cytokine levels and embryo tc.:xic:ity ir. 1.>s~rit.cneal fluids in infertile worsen with untreated or t:rea:trf~e:~ endometr:iosi.s. P.m !J Ubstet Gynecol 1992; 167:265--270.
Taylor RN, Lebovic DI, l-l;~rnung D, Mueller MD. Endocrine and paracrine regulation c:~i =ndometri_a.L ang:i.ogenesis. Ann N Y
Acad Sci 2007_,; 943:1.09-_.~1.
Tseng JF, Rydn IF, Mi.i _~n~ TD, Murai J'),, ~chriock ED, Landers DV, Taylor RN. Interl~uz:~in-6 secret:.ion .~:r~ vitro is up-regulated in ectopic ~srud ~~utopic endometr_ia1 stromal ce7_ls from women wi_i=h endome.'::rio~~is. ~T Cli.n Endocz:inol Metab 1996; 81: 1118--:L122.
Versos MW, Beard JS, C.ra:~vJes K, W=ikon Et;. Classification of endometriotic implant:. 1_-y mo:rpholog:i.c; appearance and capacity to synthesize prcstaglar:dirl .. E'erti.:L Steril_ 1986; 46:801-806.
Vigano P, Colombo A, ~~~:r~~ellini P, Cand_i.ani GB, Diblaso AM, Vignali M. Deuicient~ arlt::Lendometrii.zm iyrnphoc:yte-mediated cytotoxicity __n patiert::with endorne~:r:iosis. Fertil Stex:il 2~~ 1991; 56:894-f399.
Vigano P, Gaf j-uri B, ;':or~igi.iana E, Busac:w;a M, Di Blasio AM, Vignal_i M. Expression ):E intercellular adhesion molecule (ICAM)-1 mRNA and prot>.:in :s enhanced i_ru endometriosis versus endometrial ;tromal cce_is in cu:ltirre. Mo:l_ Hum Reprod 19'8;
4:1150-1156.
Vinatier D, Dufour P, Cc>sterlynok D. Tmmuno=Logical aspects of endometriosis. Hum Relnrc;d Update 1996; ?:2:371-2384.
Wesche H, Kor~lnerr C, I~r_acht M, t~'al.k W, ~esch K, Martin MU.
The interleuk_.in-1 r_EecEy:~i:.c>r accesscary protein (IL-lRAcP) is essential for IL-~:1-induced activation oi= interleukin-:1 receptor-asst:»=:iat:ed k:i nase (IRAK) and s~ress-activated protein kinasf~s (SAP r::iraases) . ~1 Biol C'nern 1997; 272:'77:27-7731.
Wiegerinck MA, Van Do~~ PA, Browns IA. 'The staging of peritoneal erudometr.ior,is by the type of active lesion in addition to t:loe re~.ise~; American fert:il Lt~y Society classification. fern s ~_~ter.i_~ 1.99~~~; 60: 46I-464.
Wingfield M, I~Iacphe:rs~~1 B, Healy L;L, Ro~~ers PAW. Cell proliferation is :incre:ased in the endomet riurrl of women with endometriosis. Fert:il Steri:l 1995; 64:34(7-346.
blitz CA, Schenken RS. Pathogenesis of endometriosis. Sernin Reprod Endocr:inol. 199~~'; 15: =L99--208.
Yang Y, Degranpre P, L.:inarf s A, Akoum A. Identification of macrophage migration s nr~ibitor-y f~:cct=or ,~s a potent endothelial cell growth-promos=ir:g agent released by er_topic human endomet::rial. cells. J 'lin Er~docrirnol Metab 2000; 85: 4721---472 7 .
Yoshimura T, =~eorlard t',~.J. Ideni~:ification of high-affinity receptors for human mc~nrcyte chemc>attractant protein-1 on human monocytE~s, J Imrro.zrl~1 1990; 1:5:292--.'.

Yoshimura T, =Leonard I,~7: Ident_.ific:at:ion of high-affinity receptors for hurr;an mc~nocyte chemoattra~~tant protein-1 on human monocytes. .J Imrm.zaol 1990, 145:29:'-297.
Yu PW, Schluder L,A, C~:uyrynski C:::J. Inter~euk.in receptor mRNA
_'> expression ire activatec:~ bovine l.eu.ko~cytc=::; in vitro. Clin Diagn Lab Imrnunol 199 ; ; 4 : 769-'7~i3.
Zeller JM, HeTllg I, R~dwan:>ka E, Dmowsk_~. WP. Enhancement. of human monocyte and per it:~ileal ma~;ropzn~~gF~ chemiluminescence activities in women w.t:li endomet::rio~~is. Am ~T Reprod Immunol Microbiol 198'7; 13:78--f3:='.

~~54 85409-5 .c~EpCJE=I~I(_'f:; :1:;TITd(-:
c>
(1) GENERAL I:NFORMAT~:ON:
(i) APPLICANT: CINI'JER:'L'if: LAVAL
(ii) TITLE OF INVET~rIc::~ M~fE?OUS P,ND PRODUCT'S OF REPRODUCTIVE
PROCESSES AND FOR DI?~GNOS1 :., T 30GNCnSTI~A'aT02~ FNO TREATMENT OR RELATED

(iii) NUMBER OF SIQUENC'-S: 6 (iv) CORRESPONDENCE ADI=uRI Sh:

15 (A) ADL>RESSEE: >>'9i:.R.'C & EvIGGIe.F;

(B) S:CREET: 1000 rug <9E. la Wuc,~et.ier~-' Ouest/St.West, Suite 3400 (C) CITY: MON'PRE~~.I~

(D) S-fF~TE: QIJEBE.' (E) C~:)C7NTR~-: CAT~I",Dry 2C1 (F) ZIP: H:SB 4W'.-(v) COMFUTf~R READABLE E'(,RM:

(A) MEDIUM TYPE: f-cp~vy disk (B) C~:)'4PDTF;R: IB~9 F~~:' co.npatik.l.e 25 (C) OfERATiNG SY'I'l,M: PC-DC)S;Mts-DOS

(D) SOFTWA'R.E: A;C'~ i (text) (vi. CURREN'=, APP:LICAT:L ?)'::~AI'A:
) (A) Af'fL:IC.?e,TION ':o'M:31~H:.: CA ',;si-7,%86 30 (=3) FI:LTN(J I:.'I_.:: 'J-MAR ~00'?

(C) CI~.?~SSII'I'~llT-L >' :

(vii PRIOR .t~PPLICATION DATA:
) (A) AfPLICi-~,TI(>N LM~EF:

35 (B) F. DING DATE:

(viii ATTORNEY/AGi_.NT INC'OP.MATION:
) (A) NA1~IE: SMARh ~, BIGGAR

(B) RFGI STRATION T\CIMBER,:

4C~ (C) RE EEREIUCE/DC> .'n:F'1' TIUMBER: 8'_x909-4 (ix) TELECOMI~IUNI~.'ATION TLJfORMATION:
(A) TILEPHcE.~NE: ; ; l ~i ) -c> 54-1501::
(B) T'_~LEFA:e:: ( 5; ' j ._ :~r;q-1 J96 4 =.
(2) INFORMATION FOR ;EQ II~ I~:O',.: l ( i ) SEQUENCE CT.~1RAC'I'ER:f:> ' l (A) LENGTEI: ?I
5C~ (B) TYPE: nucleic acid (C) STRANDEDN;v~S:
(D) TOPOLOGY:
( ii ) MOLECULE 'l YPE : DNP.
(vi) ORIGINAL SOURCE:
5~~ (A) ORGANISM: Artifici-.1 (ix) FEATURE
(C) GTHER INF03MATION: ~rime.r (xi) SEQUEDICE =):~SCRIPT:I:ON: ::'.EQ iD NO.: L:
TCCATGTGCA AATC~:'.TCTf.'I~ T 21 (2) INFORMATION FOR SEQ Ir t~l~:.:
(I) SEQCJENCE CHARACTERIST7C':>
(A) LENGTH: 2.1 (B) 'nYPE: IIllCIE?1C <iCld (C) STRANDEDNESS:
(D) TOPOLOGY:
( ii ) MOLECULE 'TYPE : DNr I O ( vi ) ORIGINAL ,3OURC:E
(A) ORGANISM: Ar~, if.ic~,a_ ( ix ) ?FEATURE
(C) OTHER INFORMATION: psrimer (xi) SEQUENCE ;DESCRIPTION: ,-:EQ ID NO.: ;?:
TCCTGCCGTT CATCTCATA:: C 21 (2) INFORMATION FOR SEQ If P~C.: 3:
2 O ( I ) SEQUENCE: CIjARACTERIS 1' I C::
(A) LENGTH: 25 (B) TYFE: nuc.Ceic acid (C) STRANDEDNESS:
(D) TOPOLOGY:
2 5 (ii) MOLECULE TYPE: DNA
( vi ) C:)RIGINAL :SOURCE::
(A) ORGANISM: Artificial ( ix ) FEP_TURE
(C) OTHER INFORMATION: I:~im~=r 30 (xi) SEQUENCE DESCRTPTIC>I'v: w~EQ ID N0. : 3:
TGATGAC'ATC AAGAJ~GGTGC~ TGAF\.:; 25 3'i (2) INFORMATION FOR SEQ ID 'NCI.: 4:
( I ) SEQUENCE CILARAC'_rERIST I C;
(A) LENGTH: '~3 (B) TYPE: nuciei:~ ._acid (C) STRANDEDNf~SS:
40 (D) TOPOLOGY:
(ii) MOLECULE :',YPE: DNA
(vi) ORIGINAL S7UR(_',E~:
(A) C>RGANISM: Art:.if ici,-al (ix) FEATURE
45 (C) OTHER INFO RMATLON: primer (xi) SEQUENCE DESCR:TPTION:: :'EQ :7D NO.: 9:
TCCTTGCiAGG CCAT~:~TGGG~_' ~"AT 23 5 Cl (2) INFORMATION FOR SEQ IIli;.: ,.
(i) SEQUENCE CHARACTERIS': L(':' (A) LENGTH: 21 (B) TYPE: nucleic acid (C) STRANDEDNESS:
( D ) ''OE'OLOCIY
(ii) MOLECULE IYPE: DNI1 (vi) ORIGINAL SOURCE:
(A) ORGANISM: Art.ific~,::
1 () ( ix ) I'EATURE
(C) OTHER INF~~RMATION: ~;arimf.,r (xi) SEQUENCE ~ESCRIPTIOC~: ~"EQ ID ~~0.: 5:
CTCTCCGAGC TCAC~CAGC:'?. G 21 ( 2 ) INFORMATION FOR SEQ I I' P:I(?. : h ( i ) SEQUENCE C ~ARACTERIS'I _ c';
(A) LENGTH: 21 2() (B) TYPE: nucleic acid (C) ~TRANDEDN~SS:
(D) 'TOPOLOGY:
(ii) MOLECULE TYPE: DNF
(vi) ORIGINAL SOURCE:
(A) ORGANISM: Ar~ifici,s (ix) FEATURE
(C) OTHER INFORMATION: pr-W ;e.r (xi) SEQUENCE IESCRIPTTOC~: ~aEQ ID NO.: 6:
CGCGTTCATG TCGT:~ATAG'I' T 2?.

Claims (34)

WHAT IS CLAIMED IS:

1. A method of assessing a reproduction-associated disease in a subject, said method comprising:
(a) determining a test level of a parameter selected from the group consisting of :
(i) MIF protein; and, (ii) MIF encoding RNA
in a tissue or body fluid from said subject; and (b) comparing said test level to a standard selected from the group consisting of a corresponding level of said parameter in a tissue or body fluid of a control subject; and a corresponding level of said parameter in a tissue or body fluid obtained from said subject at an earlier time; wherein an increase in said test level is indicative cf reproduction-associated disease.

2. The method of claim 1, wherein the subject is a mammal.

3. The method of claim 1 or 2, wherein the subject is a human.

4. The method of any one of claims 1 to 3, wherein the body fluid or tissue is endometrial tissue.

5. The method of any cane of claims 1 to 4, wherein the reproduction-associated disease is endometriosis or infertility.

6. A method of assessing endometrial receptivity in a subject, said method comprising:

(a) determining, in said subject, a test level of a parameter selected from the group consisting of:
(i) MIF protein;
(ii) MIF encoding RNA;
(iii) IL-1RII protein;
(iv) IL-1RII encoding RNA; and (v) IL-1RII activity;
(b) comparing said test level with a standard selected from the group consisting of a corresponding level of said parameter from a control subject; and a corresponding level of said parameter obtained from said subject at an earlier time, wherein a decrease of said test level is indicative of endometrial receptivity.

7. The method of claim 6, wherein the subject is in a secretory phase of a menstrual cycle.

8. The method of claim 6 or 7, wherein the subject is a mammal.

9. The method of any one of claims 6 to 8, wherein the subject is a human.

10. The method of any one of claims 6 to 9, further comprising predicting a window of implantation in accordance with said endometrial receptivity.

11. The method of any one of claims 6 to 10, wherein said test level is determined in a body fluid or a tissue obtained from said subject.

12. The method of claim 11, wherein the body fluid or the tissue is endometrial tissue.

13. A method of assessing a reproduction-associated disease in a subject, said method comprising:
(a) determining a test level of a parameter selected from the group consisting of.
(i) IL-1RII protein;
( ii ) IL-1RII encoding RNA; and (iii) IL-1RII activity;
in serum. from said subject; and (b)comparing said test level to a standard selected from the group consisting of a corresponding level of said parameter in serum of a control subject; and a corresponding level of said parameter in serum obtained from said subject an earlier time; wherein a decrease in said test level is indicative of reproduction-associated disease.

10. The method of claim 13, wherein the subject is a mammal.

15. The method of claim 13 or 14, wherein the subject is a human.

16. The methods of any one of claims 13 to 15, wherein the reproduction-associated disease is endometriosis or infertility.

17. Use of IL-1RII for the prevention or treatment of a reproduction-associated disease in a subject.

18. The use of claim 17, wherein the subject is a mammal.

19. The use of claim 17 or 18, wherein the subject is a human.

20. The use of any one of claims 17 to 19, wherein the reproduction-associated disease is endometriosis or infertility.

21. A commercial package comprising means for assessing the level of a parameter selected from the group consisting of:
(i) MIF protein; and (ii) MIF encoding RNA

in a tissue or body fluid of a subject; together with instructions for diagnosis, prognostication, or both, of reproduction-associated disease in said subject.

22. A commercial package comprising means for assessing the level of a parameter selected from the group of:
(i) IL-1RII protein;
(ii) IL-1RII encoding RNA; and (iii) IL-1RII activity;
in serum of a subject, together with instructions for the diagnosis, or prognostication, or both, of reproduction-associated disease in said subject.

23. The commercial package of claim 22, further comprising a reference sample of said parameter; and wherein said reference sample is a corresponding level of said parameter in a serum of a control subject.

24. The commercial package of claims 22 or 23, wherein the subject is a mammal.

25. The commercial package of any one of claims 22 to 24, wherein the subject is a human.

26. The commercial package of claim 21, wherein the body fluid or tissue is endometrial tissue.

27. The commercial package of claim 21, wherein tissue or body fluid is selected from the group consisting of endometrial tissue, endometrial cells, serum, plasma, peritoneal fluid, and monocytes.

28. The commercial package of any one of claims 21 to 27, wherein the reproduction-associated disease is endometriosis or infertility.

29. A commercial package comprising means for assessing the level of a parameter selected from the group consisting of:
(i) MIF protein;
(ii) IL-1RII protein;
(iii) MIF encoding RNA;
(iv) IL-1RII encoding RNA; and (v) IL-1RII activity;
in an endometrial tissue of a subject, together with instructions for the determination of the endometrial receptivity in said subject.

30. The commercial package of claim 29, further comprising a reference sample of said parameter; wherein said reference sample is a corresponding level of said parameter in an endometrial tissue of a control subject.

31. A commercial package comprising IL-1RII together with instructions for the prevention or treatment of a reproduction-associated disease in a subject.

32. The commercial package of claim 31, wherein the subject is a mammal.

33. The commercial package of claim 31 or 32, wherein the subject is a human.

34. The commercial package of any one of claims 31 to 33, wherein the reproduction-associated disease is endometriosis or infertility.