Journal of Autoimmunity (1992) $495-509

Induction of Experimental Anti-phospholipid Syndrome Associated with SLE following Immunization with Human MonoclonalPathogenic Anti-DNA Idiotype

M. Blank, I. Krause, M. Ben-Bassat* Steinmetz

Research

Sheba Medical Belinson Medical

Unit of Autoimmune Center,

Tel-Hashomer

Center, Petach Tel-Aviv

and Y. Shoenfeld

Diseases, Department and *Department

Tiqva and the Sackler University,

of Medicine

‘B’,

of Pathology,

School of Medicine,

Israel

(Received 6 November 1991 and accepted 2 April 1992) MIV-7 is a human monoclonal antibody that binds to DNA and carries a pathogenic anti-DNA idiotype 16/6. The antibody was generated by fusing peripheral blood lymphocytes of a healthy donor which were stimulated with an anti-idiotypic antibody to B,, (a human mAb anti-mouse mammary tumor virus-MMTV). The MIV-7, in addition to being an anti-DNA antibody, also binds to MMTV glycoproteins. Following immunization into the footpad of naive BALB/c mice with MIV-7, the mice developed anti-phospholipid syndrome (APLS) and SLE. The APLS was characterized by thrombocytopenia, the presence of anticardiolipin antibodies, lupus anticoagulant (prolonged APTT), high resorption rate of fetuses and lower mean weights of the placentae and fetuses. The SLE was characterized by serological markers (e.g. antiDNA), laboratory (increased sedimentation rate and proteinuria) and histological findings (deposition of immune complexes in the glomeruli). Active immunization of mice with mouse monoclonal anti-cardiolipin antibodies led to the induction ofprimary APLS without SLE. The results add to our previous passive transfer model in which mouse monoclonal anti-cardiolipin antibody generated from immunized mice (CAM) was infused into the tail vein and also resulted in induction of pure APLS [ll]. Our results demonstrate the ability to induce secondary APLS to SLE following immunization with a pathogenic idiotype of anti-DNA antibodies and to induce primary APLS with anti-cardiolipin mAb. The existence of these experimental models may permit controlled studies of novel therapeutic models.

Correspondence to: Y. Shoenfeld, Department of Medicine ‘B’, Sheba Medical Center, Tel-Hashomer, 52621, Israel.

0896-841 l/92/040495

+ 15 $03.00/o

0 1992 Academic Press Limited

496 M. Blank et al. Introduction Systemic lupus erythematosus (SLE) is a classical autoimmune disease characterized by diverse manifestations [ 11. Recently, the association of anti-phospholipid syndrome (APLS) with SLE was emphasized [2]. APLS is defined as recurrent thromboembolic phenomena, thrombocytopenia and recurrent fetal loss associated with the presence of anti-cardiolipin antibodies (ACA) or lupus anticoagulant (LA) P-71. Elsewhere, we have reported the induction of experimental SLE in naive mice following immunization with pathogenic anti-DNA idiotypes [&lo]. In another set of experiments we have shown the ability to induce APLS in mice by passive transfer of ACA from various sources [ 111. In the following study we describe the induction of experimental APLS associated with SLE, following active immunization of mice with a new human monoclonal anti-DNA-antibody (MIV-7) carrying the pathogenic anti-DNA idiotype 16/6 [12, 131, as well as induction of primary APLS [14] with mouse monoclonal anti-cardiolipin antibodies derived from the mice with the induced syndromes. Materials and methods Mice BALB/c female mice aged 8 weeks were purchased from Tel-Aviv

University.

Monoclonal antibodies (mAb) (Table 1) MIV-7, MZV-5, MZV-2 The human mAbs were produced by the human hybridoma technique, fusing the GM-4672 human lymphoblastoid cell line with normal human peripheral lymphocytes which were immunized in vitro with an anti-idiotypic antibody to anti-mouse mammary tumor virus (MMTV) (Bll) [15, 161 named rabbit anti-B11 antibodies [17]. MIV-7 (previously named MIV-A7), an IgM mAb, reacted with the glycoprotein gp52 of the mouse mammary tumor virus (MMTV) and human mammary tumor virus (HuMTV) [ 16,171. MIV-5 (previously named MIV-E5), an IgG mAb, binds to the antigens derived from HuMTV, MMTV but does not bind to DNA. MIV-2 (previously named MIV-B2) is an IgG mAb that does not bind either to HuMTV, MMTV or to DNA. MOBD-5 MOBD-5 is a human IgM mAb generated by the human hybridoma technique, following fusion of axillary lymph node cells from a breast cancer patient with GM-4672 lymphoblastoid cell line, This mAb binds to DNA but does not recognize the antigens derived from HuMTV or MMTV. SA-1 SA-1 is a human IgM mAb derived by the hybridoma technique from a patient with polymyositis at an active stage of the disease. This mAb was found to bind DNA and to carry the common lupus anti-DNA idiotype 16/6 Id [18]. SA-2 is a human IgM

Induction of SLE and anti-phospholipid Table 1. Characteristics mAb MIV-‘7

Origin

IgM

Normal peripheral lymphocytes in vitro immunized with anti-idiotypic Ab to mouse mammary tumor virus (MMTV) and fused with human lymphoblastoid cell line GM-4672

Binding properties DNA+, 16 Id+, MMTV+, HuMTV+ DNA-, 16/6 Id-, MMTV+, HuMTV+ DNA-, 16/6 Id-, MMTV-, HuMTV-

MIV-2

16

MOBD-5

I&

Human lymph node cells from a breast cancer patient with GM-4672 lymphoblastoid cell line

DNA+, 16/6 Id-, MMTV-, HuMTV-

SA-1

IgM

DNA+, 16/6 Id+

SA-2

I&

Peripheral lymphocytes from a patient with polymyositis at an active stage of the disease with GM-4672 lymphoblastoid cell line Peripheral lymphocytes from a patient with polymyositis at remission with GM-4672 lymphoblastoid cell line

CAM

W

Mouse splenocytes from mice immunized with MIV-7 which developed SLE with secondary APLS, fused with NSO plasmacytoma cell line

cardiolipin+ DNA-, 16/6 Id-

N40

16

mAb from the same fusion of CAM mAb

cardiolipin1616 Id-

CAL

IgG

Mouse splenocytes from mice immunized with SA-1 which developed SLE only, fused with NSO plasmacytoma

cardiolipin+ DNA-‘, 16/6 Id-

DNA-,

mAb derived from the same patient as SA- 1 while in remission the Iigand

binding

capacities

497

of the mAbs employed in the experiments

Isotype

MIV-5

syndrome

of SA-1,

MIV-7,

MIV-5

16/6 Id-

DNA-,

[ 181, SA-2 mAb lacks

and does not express

the

1616 Id. Bll

Bl 1 mAb, a human mAb IgG described previously [15], reacts with MMTV glycoproteins gp52 and gp36 and with antigens derived from HuMTV. Mouse anti-1616 mAb was kindly provided by Prof. E. Mozes from the Weizman Institute, Rehovot, Israel [ 191. Direct binding of anti-1616

idiotype antibodies to mAbs MZV-7, SA-1, SA-2 and 1616

MZV-5,

MOBD-5,

The presence of the 1616 Id on MIV-7 and other mAbs was demonstrated by direct binding of the anti-idiotypic (16/6) antibody to the mAbs. Briefly, 96-well flat-

498

M. Blank et al.

bottomed ELISA plates (NUNC, Denmark) were incubated with different dilutions of the mAbs (MIV-7, MIV-5, MOBD-5, SA-1, SA-2,16/6) in 0.05 M borate buffer, pH 8.6. After 18 h incubation at 4°C the idiotype-coated plates were washed three times with PBS and blocked with 5% bovine serum albumin in PBS (100 ul per well) for 2 h at room temperature. Following three washes in 0.1 y0 PBS-Tween, a 1: 1,000 dilution of mouse or rabbit anti-16/6 Id sera (produced as described in [12]) was added to the wells. After 5 h of incubation at 4°C the wells were washed three times with 0.1 o/0PBS-Tween. Goat-anti-rabbit antibody conjugated to alkaline phosphatase (Sigma, St. Louis, MO, USA) diluted 1: 1,500 in 0.1 y0 PBS-Tween was applied to the wells and the plates were incubated for 18 h at 4°C. The plates were washed three times with 0.1 y0 PBS-Tween and alkaline phosphatase substrate was added. The plates were read by an ELISA reader at 405 nm. Induction of APLS

and SLE by immunization

BALB/c female mice were immunized into the hind footpads with 1 ug of the affinity-purified human mAb or mouse mAb in Freund’s complete adjuvant (FCA). Three weeks later a booster-injection of 1 ug of the mAb in PBS was injected into the hind footpads [&IO, 201. Detection of autoantibodies in immunized mice Mice were bled from the retro-orbital plexus and the sera were stored at -20°C. Pooled sera from mice immunized with MIV-7 and MOBD-5 mAbs which had high anti-human IgM activity were dialysed against PBS and extensively adsorbed on a human IgM Sepharose column. The flow-through fluids were stored at - 20°C. The titers of the following antibodies were determined each month: anti-dsDNA, ssDNA, poly(I), poly(G), cardiolipin, phosphatidyl serine, phosphatidyl choline, acetyl-0-alkyl (PAF), phosphatidyl ethanolamine, SS-A(Ro), SS-B(La), Sm, RNP, histones, thyroglobulin, 16/6 Id and anti-16/6 Id. All the above determinations were carried out by ELISA as detailed previously [8-10, 12,20-221. The ELISA detecting other phospholipid antibodies was performed in a similar way to the ELISA detecting ACA. Inhibition of anti-DNA and anti-cardiolipin antibodies Sera, at dilutions that gave 50% of maximum activity, were preincubated with cardiolipin (Sigma) or dsDNA (Sigma) in different concentrations (50 ug/ml-O. 1 ug/ ml). After 16 h of incubation at 4°C anti-cardiolipin activity was tested by ELISA as previously described [ 111. (The plates were developed with goat anti-mouse IgG alkaline phosphatase.) The mouse sera that were incubated in the competition studies were preabsorbed several times on plates coated with dsDNA or cardiolipin, for 30 min each time at 37°C until no binding to dsDNA or cardiolipin, respectively, was detected by ELISA assay. Percentage inhibition was calculated as follows: % inhibition =

OD control - OD with inhibitor OD control

x 100

Induction of SLE and anti-phospholipid syndrome

499

Immunoabsorptions on solid phase were carried out as follows: mouse sera were incubated for 1 h at 37°C on 24-well polystyrene plates coated with cardiolipin. The sera were sequentially transferred from well to well until no reactivity of the immunoadsorbed sera to cardiolipin was noted by ELISA. The immunoadsorptions on solid phase to dsDNA were carried out in a similar way: mouse sera were incubated on 24-well polystyrene plates coated with dsDNA for 1 h at 37°C. The sera were sequentially transferred from well to well until no reactivity of the immunoabsorbed sera to dsDNA was noted by ELISA. Detection of SLE-associatedfindings The erythrocyte sedimentation rate was determined by diluting the heparinized blood in PBS at a ratio of 1:l. The diluted blood was passed to a micro-sampling pipette, and the sedimentation was measured 6h later. The white blood cell count (WBC) was done in 0.1 y0 acetic acid. Proteinuria was measured semi-quantitatively using a combistix kit (Ames-Miles, Slough, UK). The deposition of immunoglobulins in the affected kidneys was confirmed by electron-microscopy (EM). Small pieces of renal tissues were fixed in 1 y0 osmic acid and embedded in epoxy resin. Ultrasections were double-stained with 25 g/d1 uranyl acetate and 2.8 g/d1 lead citrate and were examined with a Phillips 300 electron microscope [8]. Detection of primary anti-phospholi$id

syradrame (APLS)

Platelet counts from individual blood samples were quantified on an ELT8jWS cell counter (ELT8/WS cell counter, Coulter Electronics Ltd, UK). Anticoagulant activity was evaluated by the mixing activated partial thromboplastin time (APTT) with cephalin (IL test 97584-20 Instrumentation Laboratory, USA) as follows: plasma (separated from the mouse’s blood), which was preabsorbed with cardiolipin on solid phase, and a control plasma which was not preabsorbed, were added at a ratio of 1: 1 to cephalin and incubated for 2 min at 37°C. Then, another volume of 0.025 M CaCl, was added and the clotting time recorded. Correction of APTT was carried out by exchanging 50% of the plasma tested for normal mouse plasma. Anti-platelet antibodies were characterized by ELISA as previously described [23]. Briefly, human platelets were separated by centrifugation (170 x g, 10 min), resuspended in 0.33% NA, EDTA in phosphate buffered saline, pH 7.4 (EDTA-PBS) and centrifuged (1,300 xg 10 min). The pellet was washed three times in EDTAPBS and finally resuspended in buffer to a concentration of 1 x lo9 platelets/ml. Ninety-six-well polystyrene plates (Nunc) were coated with 1 x 10’ platelets/well and blocked with 5 o/0bovine serum for 2 h at room temperature. Following overnight incubation with tested sera, the plates were rinsed and incubated for 2 h at room temperature with anti-mouse immunoglobulin conjugated to alkaline phosphatase. The appropriate substrate was added and the results were read at O.D. at 405 nm.

Evaluation

of pregnancy outcame

The weights of the placentae and embryos were recorded. resorption of embryo in utero was calculated as follows:

The

percentage

of

500

M. Blank et al.

1.6

E c 0” * z J

I .4 1.2 I .o

g

0.8

2

0.6 0.4 0.2

ssDNA

dsDNA

Poly

(1)

Poly

(6)

Cordiolipin

Figure 1. The binding properties of SA-1, SA-2, MIV-5, MIV-7, MOBD 5 human rnAb at a concentration of 5 pg/ml to different autoantigens. The data are expressed as O.D. at 405 nm as measured by ELISA. W SA-1; q MOBDS; I3 MIV-7; H SA-2; 0 MIV-5.

O/‘R=-

R

R+F

i.e. the number of resorbed fetuses divided by resorbed plus full term fetuses. Production

of mAbs to cardiolipin

Mice injected with MIV-7 mAb which developed APLS with SLE and mice injected with SA-1 which developed SLE were sacrificed. The splenocytes were fused with NSO plasmacytoma cells at a ratio of 1O:l in the presence of PEG (polyethylene glycol) 1500, and seeded in 96 wells in HAT medium. After 2 weeks the hybridomas were screened and the positive clones were subjected twice to limiting dilution. The anti-cardiolipin mAb CAM was derived from a mouse with APLS, the anticardiolipin mAb CAL was derived from a mouse with SLE. Results Characterization

of MIV-7

mAb

MIV-7 mAb binds to dsDNA, ssDNA, poly (I) and poly (G) (Figure 1). Its binding characteristics are comparable to the anti-DNA, SA-1 and MOBD-5 mAbs. SA-2 and MIV-5 mAbs do not bind to any of the above antigens. MIV-7 mAb was found to carry the 1616 idiotype, as can be seen in Figure 2. Induction

of SLE

and APL.9

by MIV-7

mAb

Immunization of BALB/c mice with MIV-7, MIV-5, MIV-2, MOBD-5, SA-1 and SA-2 mAbs resulted in an SLE-like disease only in the mice immunized with mAbs

Induction of SLE and anti-phospholipid syndrome

o.or 0.0 I

501

1 Y

’

’

’ ’ ’

“‘I

I

I

I111l11

Antibody

I

concentration

(pg/m!

I

1ll1111

IO

1.0

0.1

)

Figure 2. 16/6 Id expression on the human MIV-7, SA-1, 16/6 MIV-5, SA-2, MIV-2 and MOBD-5 mAbs. The 16/6 Id expression on those antibodies was determined in an ELISA assay by rabbit anti-16/6 antibodies. Results areexpressedas O.D. at 405 mm-M--,MIV-7;-¤--, SA-l;-•--, 16/6;-0--, MIV-5; -A-, SA-2; -W--, MIV-2; --a--, MOBDS.

carrying the 16/6 Id, i.e. MIV-7 and SA-1. In addition, the mice immunized with MIV-7 mAb, but not SA-1, also developed clinical conditions compatible with anti-phospholipid syndrome. The mice injected with MIV-7 and SA-1 mAbs developed significantly (PC 0.05) higher titers of antibodies to dsDNA, ssDNA, poly(I), poly(G), cardiolipin and antiMIV-7 or SA-1, respectively (Table 2). Representative dilution curves (each one a mean of 10 mice) of autoantibodies to cardiolipin and dsDNA are shown in Figure 3. The titers of autoantibodies were lower in the sera of mice immunized with MIV-5, MIV-2, MOBD-5 or SA-2 mAbs. The sera of the mice immunized with MIV-7 mAb contained two kinds of anticardiolipin Abs; one which binds only to cardiolipin and another which also crossreacts with dsDNA, as can be seen from competition and immunoadsorption studies (Figure 4). However, the sera of the mice immunized with SA-1 mAb contained anticardiolipin Abs which cross-reacted with dsDNA, since preabsorption of the sera on dsDNA coated solid phase completely abrogated the anti-cardiolipin acitivity of the sera (Figure 4). The anti-cardiolipin Abs activity was confined to the IgG class. The appearance of autoantibodies in the sera of mice immunized with MIV-7 and SA-1 mAbs (each group 15 mice) was accompanied by findings compatible with SLE, such as significantly elevated ESR (P