Isolation and Characterization of Apoptotic ... - Science Direct

Journal of Autoimmunity (1998) 11, 19–27

Isolation and Characterization of Apoptotic Nucleosomes, Free and Complexed with Lupus Autoantibody Generated During Hybridoma B-cell Apoptosis Alban Cabrespines, Diego Laderach, Christelle Lebosse´, Jean-Franc¸ois Bach and Sophie Koutouzov INSERM U25, Hoˆpital Necker, Paris, France

Received 29 August 1997 Accepted 21 November 1997

Key words: nucleosome, apoptosis, immune complexes, anti-DNA, anti-nucleosome

Increasing evidence suggests that immune complexes made of anti-nuclear antibodies bound to nucleosomes released from dead cells play an important role in the pathogenesis of lupus nephritis. However, the nature and composition of apoptotic nucleosomes still remain elusive. Since large amounts of nucleosomes are released from cells undergoing apoptosis in hybridoma cell cultures, we used hybridomas secreting anti-DNA and anti-nucleosome antibodies grown in protein-free medium to generate nucleosome/anti-DNA and / anti-nucleosome immune complexes, as well as an irrelevant antibody hybridoma to generate free, non-complexed apoptotic nucleosomes. Hybridoma supernatants were fractionated by size-exclusion gel chromatography and eluted fractions with a ratio of A260/A280 >1.2 were pooled and analysed for DNA and histone profiles by gel electrophoresis and immunoblotting. When run on a ‘native’ gel, ‘intact’ apoptotic nucleosomes, free or within antinucleosome immune complexes, showed a strikingly reduced size compared with ‘standard’ nucleosomes prepared in vitro by endonuclease digestion of cell nuclei. Nucleosomal DNA (extracted from either free or complexed apoptotic nucleosomes) appeared as a major band of 160–180 bp, and had the size of ‘standard’ mononucleosome DNA, suggesting degradation of the histone moiety of apoptotic nucleosomes. Histone immunoblotting revealed degradation of histones H3 and H4, which was dramatically enhanced when apoptotic nucleosomes were complexed with an anti-nucleosome antibody. Our results provide direct evidence for abnormal histone composition of apoptotic nucleosomes and suggest that the fine specificity of the complexing antibody has an influence on complexed nucleosome composition. © 1998 Academic Press Limited

Introduction

autoantibodies (which recognize quaternary epitopes on the whole nuclear autoantigen, but not its individual constituents, DNA and histones) have been found to develop at the onset of the autoimmune response in murine disease, before anti-dsDNA and anti-histone antibodies [2, 3]. Studies in lupus-prone mice revealed that antibodies against the nucleosome are the result of an antigen-driven response [2], a finding which is substantiated by the presence of nucleosome-specific T-helper cells in young, preautoimmune lupus-prone mice [4]. The nucleosome thus could be the original immune stimulus that gives rise to not only nucleosome-specific autoantibodies but also to anti-DNA and anti-histone antibodies [4]. Increasing evidence suggests that nucleosomes play an important role in the effector phase of disease by

Systemic lupus erythematous (SLE) is the prototype of non-organ-specific autoimmune diseases in which autoantibodies develop against components of the cell nucleus, particularly against chromatin and its constituents, DNA and histones [1]. In recent years, it has become evident that the elementary component of the chromatin, the nucleosome, released by internucleosomal cleavage during cell apoptosis, plays a pivotal role both in the induction of disease and in the development of kidney lesions. Anti-nucleosome Correspondence to: Dr Sophie Koutouzov, INSERM U25, Hoˆpital Necker, 161 rue de Se`vres, 75015 Paris, France. Fax: 33 1 43 06 23 88. 19 0896-8411/98/010019+09 $25.00/0/au970172

© 1998 Academic Press Limited

A. Cabrespines et al.

20

targeting autoantibodies to the glomerular basement membrane. The first evidence came from experiments showing that anti-heparan sulfate cross-reactivity displayed in ELISA by anti-DNA antibodies was not exerted by the antibody itself, but was mediated by nucleosomal material complexed to the antibody [5–7]. It was subsequently shown in a rat kidney perfusion system that histones could mediate binding of subsequent perfused DNA and anti-DNA to the glomerulus, indicating in situ immune-complex formation through nucleosome-like planted antigens [8, 9]. Also, recent data have demonstrated that preformed complexes made of histone-DNA and antiDNA antibodies from SLE patients bound to rat glomeruli in vivo [10], suggesting that circulating nucleosome-containing immune complexes may target the kidney in vivo. Direct evidence of nucleosometriggered deposition of pathogenic antibodies came from experiments showing that monoclonal antinucleosome antibodies which are complexed to nucleosomes bound to the rat glomerular basement membrane (GBM) when perfused in vivo, while anti-nucleosome antibodies alone did not [11]. The positively charged histone part of the nucleosome appeared to be responsible for binding to negatively charged determinants in the GBM, namely to heparan sulfate (HS). Indeed, it was found that binding of nucleosome/anti-nucleosome immune complexes decreased after renal perfusion of heparinase [11], and that staining of HS in the GBM was almost completely absent in both human and murine lupus due to the binding of nucleosomecontaining immune complexes to HS in the GBM [12]. Nonetheless, other studies using MRL/l and human SLE sera have put forward collagen IV, another major component of the GBM, as a predominant ligand for nucleosome/anti-nuclear immune complexes [13, 14]. Interestingly, it was also shown that a relative decrease in DNA or a relative increase in histone content within the nucleosome could increase the capacity of the nucleosome/antibody complex to bind to HS in the GBM [11]. Moreover, the presence of antibody bound to the nucleosome seems to be a critical determinant, since it was found that perfusion of naked nucleosomes did not lead to significant GBM binding [10, 15]. As postulated previously, it is possible that the diversity of the targeted glomerular antigens, and thus the elicited pathogenic effects, may be related to the specificity of the complexing antibody, which would thereby influence the overall pI of the complex and the nephritogenicity [15]. The histone and DNA composition of circulating apoptotic nucleosomes in SLE [16] and of nucleosomes within immune complexes are at present unknown. To investigate the pathogenicity of nucleosomes and nucleosome-containing immune complexes, we developed a method permitting us to obtain purified apoptotic nucleosomes and nucleosome/anti-nucleosome autoantibody immune complexes. We characterized the histone and DNA composition of these apoptotic autoantigens,

whether or not lupus antibody.

they

were

complexed

with

Materials and Methods Monoclonal antibodies Two anti-nuclear mAbs were used for most of the studies described in this paper. The anti-dsDNA hybridoma H241 (IgG2a, ê) has been described in detail elsewhere and was derived from an MRL-Mplpr/lpr mouse [17]. The hybridoma 6E5 (IgG2b, ë), which secretes a monoclonal anti-nucleosome antibody, was isolated after fusion of a non-secreting myeloma line and MRl-Mp-+/+ mouse spleen cells [18]. Both mAbs were shown to bind native nucleosomes in ELISA [18, 19] and will be referred to here as ‘anti-nucleosome’ antibodies. Hybridomas were grown in RPMI glutamax (Life Technologies, Cergy Pontoise, France), supplemented with 10% FCS and antibiotics, in a 5% CO2 humidified atmosphere.

Preparation of apoptotic nucleosomes, free and complexed with anti-nucleosome mAbs Previous work has shown that mouse B-cell hybridomas grown in serum-free medium undergo apoptosis, resulting in presence of nucleosomes in the culture medium [20]. In order to obtain anti-nuclear mAbs complexed with apoptotic nucleosomal material, exponentially growing H241 and 6E5 hybridoma cells (approx. 5–7×105 cells/ml) were transferred into 75 cm2 flasks containing 20 ml of serum-free medium (PFHM II; Life Technologies) and grown for approx. 10–12 days until most (80–90%) hybridoma cells were dead (as visualized by trypan-blue exclusion). An irrelevant antibody-secreting hybridoma (antiperipherin IgG2b, k clone 72.2; a gift from Prof Boitard, Necker Hospital, Paris, France), which did not complex apoptotic nucleosomal material released in the culture medium, was grown in similar conditions in low-protein serum-free medium (HSFM; Life Technologies), and was used as a source of apoptotic nucleosomes [20]. To verify the presence of nucleosomes within the Ig immune complexes, approx. 5 ml of hybridoma culture fluid was mixed with protein A sepharose beads (Pharmacia Biotech, Uppsala, Sweden) for 30 min. The supernatant was discarded after centrifugation, and the beads washed five times with PBS before resuspension in Laemmli sample buffer [21]. After reduction for 5 min at 95°C, the supernatant was subjected to SDS-PAGE (15%) analysis followed by Coomassie blue staining.

Purification of apoptotic nucleosomes free and complexed with anti-nucleosome mAbs Hybridoma supernatants of interest were pooled to a working volume of 100–150 ml to 1,000 ml (depending

Isolation and characterization of apoptotic nucleosomes

on the hybridoma antibody secretion level; see below) and were concentrated 30–100-fold for 3–9 h in 100,000 molecular weight cut-off (MWCO) Spectra/ Por dialysis tubes (Polylabo, Strasbourg, France) using Aquacide II (Calbiochem, La Jolla, USA). Nucleosome-containing immune complexes (H241 or 6E5) or free apoptotic nucleosomes (+72.2 antibody) were further concentrated, if necessary, by ultrafiltration on 100,000 MWCO Centriplus concentrators (Amicon, Inc, Beverly, MA, USA), and then extensively dialysed against PBS, pH 7.0. Supernatant samples were applied to Superdex 200 gel filtration columns (Pharmacia) equilibrated in PBS; a 30 cm×0.8 cm2 column (HR gel) was used to resolve separation of immune complexes (approx. MW 400,000) from free, non-complexed antibodies (for H241 and 6E5 hybridomas) while a 70 cm×8 cm2 column (XK gel) was used to allow separation of free apoptotic nucleosomes (approx. MW 250,000) from the irrelevant 72.2 antibody. Fractions were collected at a flow rate of 0.3 ml/min, and the OD at 280 and 260 nm was recorded by spectrophotometry. Fractions in which material exhibited an optical density ratio for 260/280 ranging between 1.2 and 1.8 (which attested to the presence of DNA) were pooled, concentrated on Amicon PM-30 filters, dialysed against 10 mM Tris, 1 mM EDTA (TE), and 0.2 mM PMSF buffer, and were kept at 0°C until use. The protein content of the different pooled fractions was determined by the bicinchoninic acid (BCA) method using IgG as the standard protein (Pierce) and the presence of histones, either in apoptotic nucleosomes (from 72.2 supernatants) or in nucleosome/Ig complexes (from H241 and 6E5 supernatants), was determined by SDS PAGE (15%) analysis (with load of approx. 10 ìg proteins as assessed by BCA) followed by Coomassie blue staining.

Preparation of ‘standard’ nucleosomes Nucleosomes were prepared from L1210 cell nuclei according to the method developed by Lutter [22], with modifications detailed previously [23]. The purified H1-stripped nucleosome preparations contained almost pure mononucleosomal material [23], and will be referred to here as ‘standard’ nucleosomes.

Determination of nucleosome concentration in non-nuclear 72.2 hybridoma supernatants Nucleosome quantification was performed using a sandwich ELISA, as described previously [24]. Briefly, microtiter plates (CML, Nemours, France) were coated with 10 ìg/ml of an anti-nucleosome capture mAb (13G10; IgG2a) overnight at 4°C and, after washing with PBS-0.1% Tween, pH 7.4 (PBST), were postcoated for 2 h with 0.1 ml of PBS-10% FCS, pH 7.4. Serial dilutions of supernatants were then added to the wells and incubated for 2 h. Wells were washed thoroughly to remove free antigen, and the secondary anti-DNA mAb (PME 77; IgG2b) (which also

21

recognized nucleosomes in ELISA; [3, 19]) was then added to the plates and reacted for an additional 2 h incubation. After extensive washing, bound PME 77 was detected with peroxidase-conjugated goat antimouse ã 2b antiserum (Sigma, USA), and OD read at 450 nm. Pure mononucleosomes were used under the same conditions to generate a standard curve with known autoantigen concentrations.

Determination of immunoglobulin concentration in hybridoma supernatants and purified nucleosome/Ig complexes Immunoglobulin (Ig) concentrations were quantitated by IgG subclass-specific ELISA. Briefly, microtiter plates (CML) were coated overnight with goat antimouse IgG (Biosys, Compie`gne, France; 10 ìg/ml in PBS; 100 ìl) and washed with PBS-0.05% Tween prior to saturation with PBS containing 10% FCS to block non-specific binding. Serial dilutions of hybridoma supernatants or purified nucleosome/Ig complexes mAbs in PBS-0.05% Tween-10% FCS were reacted for 2 h at room temperature. Bound mAbs were detected with 1/1,000 diluted peroxidase-conjugated antimouse IgG2a or IgG2b (Southern Biotechnology Associated, Inc. Birmingham, AL, USA). Reactions were developed for 5 min with ABTS substrate (Southern Biotechnology), and OD at 405 nm was measured. Mouse IgG2a and IgG2b myeloma proteins (Southern Biotechnology) were used as standards.

Characterization of nucleosomes and nucleosomal DNA in preparations of apoptotic nucleosomes, free or complexed with antibody DNA was extracted from nucleosome and nucleosome/Ig complex preparations with chloroform/ isoamyl alcohol (25:1) and precipitated with cold ethanol. The pellets were washed with ethanol, resuspended in loading buffer (60% sucrose, 10 mM Tris, 1 mM EDTA, pH 7.4), and were run onto gels made of 4% polyacrylamide (bis(acrylamide): acrylamide ratio=1:20) and containing 40 mM Tris, 20 mM sodium acetate and 2 mM EDTA, pH 7.4 buffer [25]. These gels, which allowed DNA to remain in doublestranded form and nucleosomes to remain intact during electrophoresis (referred to as ‘native gels’) [25] were used in parallel samples of the various nucleosome-containing preparations as a control for particle integrity. Gels were stained with ethidium bromide (1 ìg/ml) and photographed using Polaroid 667 film.

Nucleosomal histone immunoblotting Samples of purified standard or apoptotic nucleosomes, free (approx. 1 ìg DNA) or complexed with antibodies (approx. 2 ìg IgG), were subjected to SDSPAGE (15%) and gels were blotted onto nitrocelulose in 50 mM Tris, 380 mM glucine, and 20% methanol for

22

A. Cabrespines et al.

by the different mABs showed (with the notable exception of the anti-nucleosome 1G2 mAb) altered histone profiles. Common to all mAbs (except 1G2) was the visible degradation of the H3 histone (MA16=PME77>6E5>H241) and the appearance of histone material migrating above and below histone H4. The different nucleosomal histone profiles shown on this gel suggest that the fine specificity of the complexing antibody may alter the histone profile of the complexed apoptotic nuclear antigen.

Figure 1. SDS-PAGE of protein isolated from protein A sepharose beads which were incubated with cell-culture supernatants from hybridoma-secreting anti-nucleosome (lane 1, 1G2; lane 2, 6E5) or anti-DNA (lane 3, PME77; lane 4, MA16; lane 5, H241) antibodies. Lanes 1–5 show the Ig, heavy and light, and core nucleosomal histones. Note on both sides distribution of histones extracted from standard H1-stripped purified nucleosomes.

2 h at 60 V. Blots were blocked for 1 h with 1% BSA in 10 mM Tris, 150 mM NaCl, and 0.05% Tween (TBST), and the nitrocellulose strips then reacted with rabbit antisera (1/500–1/2,000 in PBST) directed against individual histones (anti-H2A, anti-H2B, anti-H3 and anti-H4; a gift from Dr Sylviane Muller, ICBM, Strasbourg, France) for 1 h. After four 10-min washes with TBST, bound anti-histone antibodies were revealed with phosphatase alkaline-conjugated antirabbit IgG (Biosys) (1/1,000 in TBST). Nitrocellulose strips were developed with 5-bromo-4-chloro-3inodylphosphate/nitroblue tetrazolium (BCIP/NTB) (Sigma) in 100 mM Tris, 100 mM NaCl, and 5 mM MgC12, pH 9.5, and colour development was quenched after 1 min with 10 mM EDTA, pH 8.0.

Results Anti-dsDNA and anti-nucleosome autoantibodies complex apoptotic nucleosomes released in hybridoma culture medium From a panel of anti-nuclear mAbs developed in our laboratory, two anti-nucleosome and three antidsDNA mAbs (one of which, H241, was a gift from Dr David Stollar, Tufts University School of Medicine, Boston, MA, U.S.A.) were selected. Supernatants of apoptotic B-cell hybridomas were trapped on protein A sepharose beads, and material eluted from the beads was subjected to SDS-PAGE. The gel in Figure 1 shows that all the anti-nuclear mAbs in our panel were complexed with nucleosomal histones, indicating that all mAbs were capable of binding apoptotic nucleosomes released from dead cells. However, when compared with histones extracted from standard H1-stripped nucleosomes where the four histones (H3, H2A, H2B and H3) were present in equimolar amounts, the histone material complexed

Fractionation of the hybridoma culture fluid components by size exclusion gel chromatography and electrophoretic characteristics of the eluted material One anti-DNA (H241), one anti-nucleosome (6E5) and the non-anti-nuclear 72.2 mAbs were selected for further analytical studies. Concentrated supernatants from hybridoma 72.2 were fractionated by sizeexclusion gel chromatography (Figure 2a), and the elution profile showed three major peaks. We arbitrarily fractionated these peaks on the basis of both A260 and A280 ODs and of the A260/A280 ratio (indicative of the presence of DNA), designated as fractions 1–5 on Figure 2A. The electrophoretic patterns of the material contained in fractions 2 and 3 (no material was present in fraction 1) showed protein bands corresponding to those of the histones forming the nucleosome core (Figure 3A), with notable degradation of histone H3 and material migrating above and below H4. Histones from apoptotic nucleosomes were also visible in fractions 4 and 5, although these fractions were contaminated with immunoglobulin. A typical fractionation profile of anti-nuclear (antiDNA or anti-nucleosome) hybridoma supernatants is shown on Figure 2B, showing one major peak which we also fractionated on the basis of the A260/A280 ratio (see figure inset). SDS-PAGE of the eluted material showed recovery of immune complexes in the first three fractions (nucleosome/anti-DNA H241, Figure 3B; nucleosome/anti-nucleosome 6E5, Figure 3C), and in fraction 4 of free, non-complexed immunoglobulin. The nucleosomal histone material complexed by the anti-DNA and anti-nucleosome mAbs differed not only from histones of standard nucleosomes, but also (as shown earlier in Figure 1) from one complexing antibody to another. Notably, histone H3 was obviously more degraded within the complex made with the anti-NA (Figure 3B), which displayed more histone material migrating far below H4 than with the anti-nucleosome (Figure 3c).

Characterization of ‘native’ apoptotic nucleosomes, free and complexed with antibodies and of nucleosomal dsDNA by gel electrophoresis Purified apoptotic nucleosomes loaded on native gels, which allow nucleoprotein particles to remain intact during electrophoresis [25], migrated as a major band

Isolation and characterization of apoptotic nucleosomes

A

B Thyr 8

Absorbance units

2

5

4

IgG

Ferrit

3 6

3

260/280 nm

5

23

BSA 2.0 1.8 1.6 1.4 1.2 1.0 0 5 10 15 20 Effluent volume (ml)

4 4

2

1

2 1

0 20

3

1

24

28

32

4

2

36

40

44

48

0

0 Volume (ml)

4

8

12

16

20

24

Figure 2. Size exclusion gel chromatography of apoptotic nucleosomes (A) and nucleosome-containing immune complexes (6E5) (B). Columns of Superdex 200 were equilibrated and run in PBS. Fractions were collected and ODs were determined at 280 (- - -) and 260 (—) nm. Fractions in which material exhibited an OD ratio 260/280 ranging between 1.2 and 1.8 (see inset in B) were fractionated as indicated by vertical dashed lines on graphs, and pooled. Note at the top of (B) the elution volume of thyroglobulin (669,000), ferritin (440,000) IgG (150,000), and BSA (67,000) molecular weight markers.

Figure 3. SDS-PAGE determination of the protein content of the different pooled fractions after size exclusion gel chromatography. Protein patterns of apoptotic nucleosomes (A), and nucleosome-containing Ig complexes with anti-dsDNA H241 (B) or anti-nucleosome 6E5 (C) antibodies. Nuc represents histones extracted from standard H1-stripped nucleosomes, and M represents molecular weight markers.

(Figure 4A, lane 3), with slight contamination with oligomer species, suggesting that nucleosomes released during cell apoptosis were predominantly in the form of mononucleosomes [16]. Notably, these native apoptotic nucleosomes showed increased gel mobility (Figure 4A, lane 3) compared with native standard nucleosomes (Figure 4A, lane 1), suggesting alteration in histone and/or DNA nucleosomal moieties. DNA extracted from these apoptotic nucleosomes migrated as a narrow band of approx. 145– 160 bp (Figure 4A, lane 4), showing little, if any, degradation of DNA (compare with DNA extracted from standard mononucleosomes; Figure 4A, lane 2). The profile obtained with apoptotic nucleosomes from immune complexes formed with anti-nucleosome antibody was more complex and is shown on panel B; these native nucleosomes ran as a ladder pattern of multiple oligomers (lane 3) with two major bands, the upper corresponding to mononucleosomes (compare

with lane 1) and the lower to free DNA (compare with lanes 2 and 4). Again, as in apoptotic, free nucleosomes, the augmented mobility of the native complexed nucleosomes is indicative of strong nucleosome degradation since extracted DNA (and in particular monomeric DNA) had a length of approx. 150–160 bp (lane 4), as did DNA extracted from standard mononucleosomes (lane 2).

Fine nucleosomal histone composition of apoptotic nucleosomes, free or complexed with antibody Using specific rabbit antisera directed against individual histones (anti-H2A, anti-H2B, anti-H3 and anti-H4) we further characterized the fine histone composition of purified apoptotic nucleosomes, free or complexed with antibody by immunoblotting. As

A. Cabrespines et al.

24

dramatic degradation of nucleosomal histones, in particular of H3 and H4 (Figure 5C, lanes 1 and 4). In this case, degradation of H3 histone generated multiple products which migrated as multiple bands above and below H4.

Discussion

Figure 4. Characterization of apoptotic nucleosomes, free (A) or complexed with antibody (B) by electrophoresis on native gels. Lanes 1 & 2 in panels A and B represent intact standard nucleosomes (lane 1), and dsDNA derived from standard nucleosomes (lane 2) electrophoresed on native gels as described in Materials and Methods. Particle integrity of apoptotic nucleosomes, free (A, lane 3) or complexed with antibody (6E5 mAb; B, lane 3) was checked and compared to intact standard nucleosomes (A & B, lane 1). DNA extracted from apoptotic nucleosomes, free (A, lane 4) or complexed with antibody (with 6E5 mAb; B, lane 4) was compared to dsDNA extracted from standard nucleosomes (A and B, lane 2).

Figure 5. Histone immunoblots of standard purified H1-stripped nucleosomes (A), purified apoptotic nucleosomes (B), and purified nucleosome-containing antinucleosome (6E5) complexes (C). Reactivities with specific histone antisera: lane 1, anti-H3; lane 2, anti-H2B; lane 3, anti-H2A; lane 4, anti-H4. M, molecular weight markers; Nuc, histones from standard nucleosomes.

shown in Figure 5, there was no sign of histone degradation in the standard nucleosome (Figure 5A) which confirms the electrophoretic pattern where histones were present in equimolar amounts (Figures 1 and 33). In contrast, apoptotic nucleosomes (Figure 5B) showed, as predicted by the native gels (see above), clear degradation of histones, in particular of H3 (lane 1) whose product migrated as a large, single band just above histone H4. Faint degradations of H2B and H4 histones were also visible (lanes 2 and 4), but their degradation products migrated below H4. Interestingly, and in contrast with free apoptotic nucleosomes, apoptotic nucleosomes complexed with an antibody (the anti-nucleosome 6E5) displayed a

The present study was designed to obtain insights into the nature and composition of apoptotic nucleosomes, either free or complexed with anti-nuclear antibodies, which are the probable forms of in vivo circulating nucleosomes in lupus [16, 26, 27]. We took advantage of the fact that significant amounts of nucleosomes are released from cells undergoing apoptosis in hybridoma cell cultures [20] and that in the presence of anti-nuclear antibodies this may lead to the formation of nucleosome–autoantibody complexes [11, 23]. Thus we used this system to produce apoptotic nucleosomes, free or complexed with lupus autoantibody. However as both immunoglobulin secretion and the amount of nucleosomes released during cell apoptosis may vary from one antibodysecreting hybridoma to another, we continued with fractionating apoptotic nucleosomes, free or complexed with antibody, by size-exclusion gel chromatography. The three-step preparation used here to isolate the different nucleosomal species was simple, although yields depended on the particular nature of the different nucleosome/immune complexes. Concentration of hybridoma culture supernants on Aquacide was found to be the most suitable system, as we observed that ultracentrifugation was limited due to handling of large starting volumes of hybridoma supernatants, ranging from 100–150 ml (antinucleosome 6E5, total Ig concentration=approx. 12 ìg/ml; irrelevant hybridoma 72.2, nucleosome concentration=approx. 3-10 ìg/ml) to 1,000 ml (antiDNA H241, total Ig concentration=approx. 1.5 ìg/ ml). This step generated yields of 80–90% from starting material, as assessed by nucleosome or Ig determinations by specific ELISAs. However, due to the mandatory small samples loaded on the gel exclusion column (approx. 0.5 ml for the immune complexes with maximum amount of 2 mg of proteins), further concentration and dialysis was usually needed to supply this volume. This step constantly induced loss of material ranging from 15 to 30% (for the anti-nucleosome 6E5 complexes) to 70–80% (for the anti-DNA H241). However, since the same procedure when applied to immune complexes formed with other anti-DNA antibodies induced reasonable loss of material, we suggest that the dramatic aggregation of anti-nuclear antibodies (through nucleosome/nucleosome interactions?) seen with anti-DNA H241 is a feature of this particular antibody. Good recoveries (90–95%) were usually obtained after size-exclusion gel chromatography and we therefore assume that this three-step purification method be a useful technique for isolating apoptotic nucleosome, free or complexed with antibody.

Isolation and characterization of apoptotic nucleosomes

Apoptotic nucleosomes were recovered from hybridoma supernatants mainly in the form of mononucleosomes, which is also the predominant form of circulating nucleosomes in vivo [16]. Our results provide the first published information on the nature and composition of free, non-complexed apoptotic nucleosomes. The integrity of the entire apoptotic nucleoprotein was altered compared with that of standard nucleosomes prepared in vitro by endonuclease digestion of cell nuclei [22], as judged by the increased mobility of the whole particle in native gels. DNA and histone analysis showed a DNA size of between 145 and 160 bp and marked degradation of H3 histone, accompanied by lesser degradation of H4. All these results suggest that nucleosomes released in vivo possess an altered conformation. It has been shown that apoptosis induced the physical clustering of nucleosomes in large apoptotic ‘blebs’, containing nuclear membranes, mitochondria and fragments of endoplasmic reticulum, sites of increased generation of reactive species in apoptotic cells (which may generate fragmentation), amino-acid modification and novel proteolytic cleavage [28]. Since apoptosis is an asynchronous process, it is possible that a significant amount of the material analysed might have been present for protracted periods in the dying cultures, and that (some) of the degradation fragments of histone H3 (and H4) observed might reflect a postapoptotic process. In this regard, we found that individual culture supernatants of apoptotic nucleosomes exhibited variable histone profiles, particularly in the extent of H3 degradation; such degradation was, however, always noticeable. Thus, it is likely that cleavage of nucleosomes occurs during the apoptotic process, as has been described for other autoantigens such as the DNA-dependent protein kinase and the 70 kDa protein component of the U1 small nuclear ribonucleoprotein [29, 30]. Nonetheless, since in vivo cell apoptosis is an asynchronous process, we assume that this process may trigger the release in the circulation or in tissue targets of such altered nucleosomes. Nucleosomal antigens released by dead cells in hybridoma cell culture supernatants were found to be complexed by all the anti-nucleosome and antidsDNA monoclonal antibodies in our panel, providing further support to the idea that these antibodies are directed against the same target, through distinct binding sites, and belong to the large family of antinucleosome antibodies [2, 3, 11, 19, 31]. Our data show that, as in free, apoptotic nucleosomes, the overall distribution of nucleosomal histones within immune complexes was apparently altered, compared with the histone profile of standard nucleosomes. Interestingly, there were visible differences in histone distribution not only between anti-nucleosome and anti-dsDNA antibodies but also between antibodies of the same specificity. For instance, the distribution of nucleosomal histones within the complex formed by anti-nucleosome antibody (1G2) [18] appeared similar to that of standard nucleosomes, whereas for the complex made with another anti-nucleosome antibody, 6E5, both H3 and H4 histones were strikingly degraded, much more extensively than in free,

25

apoptotic nucleosomes. Since the nucleosomal histone pattern appeared similar in immune complexes isolated before and after purification, as judged by Coomassie staining after SDS-PAGE, it is unlikely that the histone degradation visualized in immunoblots is due to degradation which would have occurred during size-exclusion gel chromatography. Although we did not extensively analyse the fine DNA and histone composition of all immune complexes formed with the anti-DNA and anti-nucleosome antibodies in our panel, the striking diversity of the nucleosomal histone profiles seen with the different antibodies suggests that the fine specificity of the complexing antibody may influence the composition of the complexed autoantigen. One may postulate that at least two mechanisms could be operating to account for such an effect. It is conceivable that the autoantibody binding site plays a role in the conformation of the complexed autoantigen. In this regard, it is worth noting that two-anti-dsDNA antibodies from lupus mice have recently been reported to enhance the degradation of DNA by hydroxyl radical-generating systems, suggesting that these antibodies may have facilitated cleavage by distorting the DNA structure, and/or that they displayed catalytic activities, as has previously been shown for anti-DNA antibodies isolated from SLE patients [32, 33]. If this occurred for (some) anti-nucleosome autoantibodies in lupus, then the histone and DNA moieties of the complexed nucleosomes would be differentially exposed or, alternatively, protected from proteases and DNases released by dying cells in the extracelluar milieu. Interestingly, one of the anti-DNA antibodies used by Shuster et al. (H241) [32] was also used in the present study and may exert binding-site mediated distortion of the complexed nucleosomal autoantigen, as judged by the remarkable altered nucleosomal histone profile within nucleosome/H241 complexes. The potential biological consequences of altered apoptotic nucleosomes and the influence of autoantibody on autoantigen composition may be relevant in the pathophysiology of lupus. On the one hand, occurrence of histone degradation in apoptotic nucleosomes could lead to previously protected peptides being exposed to the immune system, triggering the autoimmune response [28–30, 34]. On the other hand, exposure or masking of nucleosomal histones within immune complexes could play a role in kidney pathogenicity by influencing nephritogenicity. We are currently investigating these hypotheses.

Note added in Proof During revision of this paper, an article was published that provides evidence that the specificity of the antibody bound to the nucleosome is a critical determinant for the nephritogenic potential of the nucleosome-autoantibody complex. See Van Bruggen M.C.G., Walgren B., Rijke T.P.M., et al. Antigen specificity of anti-nuclear antibodies complexed to nucleosomes determines glomerular basement

A. Cabrespines et al.

26

membrane binding in vivo. Eur. J. Immunol 1997; 27: 1564–1569. 12.

Acknowledgements Diego Laderach is a recipient of a grant from the Consejo Nacional de Investigaciones Cientificas y Tecnicas (CONICET).

13.

References

14.

1. Tan E.M. 1989. Antinuclear antibodies: diagnostic markers for autoimmune diseases and probes for cell biology. Adv. Immunol. 44: 93–151 2. Burlingame R.W., Rubin R.L., Balderas R.S., Theofilopoulos A.N. 1993. Genesis and evolution of antichromatin autoantibodies in murine lupus implicated T-dependent immunization with self antigen. J. Clin. Invest. 91: 1687–1696 3. Amoura Z., Chabre H., Koutouzov S., Lotton C., Cabrespines A., Bach J.F., Jacob L. 1994. Nucleosomerestricted antibodies are detected before anti-dsDNA and/or antihistone antibodies in serum of MRL-Mp lpr/lpr and +/+ mice, and are present in kidney eluates of lupus mice with proteinuria. Arthritis Rheum. 37: 1684–1688 4. Mohan C., Adams S., Stanik V., Datta S.K. 1993. Nucleosome: a major immunogen for pathogenic autoantibody-inducing T cells of lupus. J. Exp. Med. 177: 1367–1381 5. Brinkman K., Termaat R.M., de Jong J., Van Den Brink H.G., Berden J.H., Smeenk R.J. 1989. Cross-reactive binding patterns of monoclonal antibodies to DNA are often caused by DNA/anti-DNA immune complexes. Res. Immunol. 140: 595–612 6. Brinkman K., Termaat R., Berden J.H., Smeenk R.J. 1990. Anti-DNA antibodies and lupus nephritis: the complexity of crossreactivity. Immunol. Today 11: 232–234 7. Termaat R.M., Brinkman K., Van Gompel F., Van Den Heuvel L.P., Veerkamp J.H., Smeenk R.J., Berden J.H. 1990. Cross-reactivity of monoclonal anti-DNA antibodies with heparan sulfate is mediated via bound DNA/histone complexes. J. Autoimmun. 3: 531–545 8. Termaat R.M., Assmann K.J., Dijkman H.B., Van Gompel F., Smeenk R.J., Berden J.H. 1992. Anti-DNA antibodies can bind to the glomerulus via two distinct mechanisms. Kidney Int. 42: 1363–1371 9. Morioka T., Woitas R., Fujigaki Y., Batsford S.R., Vogt A. 1994. Histone mediates glomerular deposition of small size DNA anti-DNA complex. Kidney Int. 45: 991–997 10. Morioka T., Fujigaki Y., Batsford S.R., Woitas R., Oite T., Shimizu F., Vogt A. 1996. Anti-DNA antibody derived from a systemic lupus erythematosus (SLE) patient forms histone-DNA-anti-DNA complexes that bind to rat glomeruli in vivo. Clin. Exp. Immunol. 104: 92–96 11. Kramers C., Hylkema M.N., Van Bruggen M.C., Van de Lagemaat R., Dijkman H.B., Assmann K.J., Smeenk R.J., Berden J.H. 1994. Anti-nucleosome antibodies complexed to nucleosomal antigens show anti-DNA reactivity and bind to rat glomerular

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

basement membrane in vivo. J. Clin. Invest. 94: 568–577 Van Bruggen M.C., Kramers K., Hylkema M.N., Van Den Born J., Bakker M.A., Assmann K.J., Smeenk R.J., Berden J.H. 1995. Decrease of heparan sulfate staining in the glomerular basement membrane in murine lupus nephritis. Am. J. Pathol. 146: 753–763 Bernstein K.A., Valerio R.D., Lefkowith J.B. 1995. Glomerular binding activity in MRL lpr serum consists of antibodies that bind to a DNA/histone/ type IV collagen complex. J. Immunol. 154: 2424–2433 Lefkowith J.B., Kiehl M., Rubenstein J., Divalerio R., Bernstein K., Kahl L., Rubin R.L., Gourley M. 1996. Heterogeneity and clinical significance of glomerularbinding antibodies in systemic lupus erythematosus. J. Clin. Invest. 98: 1373–1380 Tax W.J., Kramers C., Van Bruggen M.C., Berden J.H. 1995. Apoptosis, nucleosomes, and nephritis in systemic lupus erythematosus editorial. Kidney Int. 48: 666–673 Rumore P.M., Steinman C.R. 1990. Endogenous circulating DNA in systemic lupus erythematosus. Occurrence as multimeric complexes bound to histone. J. Clin Invest. 86: 69–74 Andrzejewski C.J.R., Stollar B.D., Lalor T.M., Schwartz R.S. 1980. Hybridoma autoantibodies to DNA. J. Immunol. 124: 1499–1502 Jovelin F., Mostoslavsky G., Amoura Z., Chabre H., Gilbert D., Eilat D., Bach J.F., Koutouzov S. 1997. Early anti-nucleosome autoantibodies from a single MRL+/+ mouse: fine specificity, V-gene structure and pathogenicity. (Submitted for publication) Chabre H., Amoura Z., Piette J.C., Godeau P., Bach J.F., Koutouzov S. 1995. Presence of nucleosome-restricted antibodies in patients with systemic lupus erythematosus. Arthritis Rhuem. 38: 1485–1491 Franek F., Dolnikova J. 1991. Nucleosomes occurring in protein-free hybridoma cell culture. Evidence for programmed cell death. FEBS Lett. 284: 285–287 Laemmli U.K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227: 680–685 Lutter L.C. 1978. Kinetic analysis of deoxyribonuclease I cleavages in the nucleosome core: evidence for a DNA superhelix. J. Mol. Biol. 124: 391–420 Loutouzov S., Cabrespines A., Amoura Z., Chabre H., Lotton C., Bach J.F. 1996. Binding of nucleosomes to a cell surface receptor: redistribution and endocytosis in the presence of lupus antibodies. Eur. J. Immunol. 26: 472–486 Amoura Z., Piette J.C., Chabre H., Cacoub P., Papo T., Wechsler B., Bach J.F., Koutouzov S. Circulating nucleosome plasma levels in SLE patients. Correlation with serum anti-nucleosome antibody activity and absence of clear association with disease activity. Arthritis Rhem. (in press) Yager T.D., McMurray C.T., Van Holde K.E. 1989. Salt-induced release of DNA from nucleosome core particles. Biochemistry 28: 2271–2281 Sano H., Morimoto C. 1981. Isolation of DNA from DNA/anti-DNA antibody immune complexes in systemic lupus erythematosus. J. Immunol. 126: 538–539 Fournie G.J. 1996. Detection of nucleosome-IgG immune complexes in ascites from mice transplanted

Isolation and characterization of apoptotic nucleosomes

with anti-DNA antibody-secreting hybridomas and in plasma from MRL-lpr/lpr mice. Clin. Exp. Immunol. 104: 236–240 28. Casciola-Rosen L.A., Anhalt G., Rosen A. 1994. Autoantigens targeted in systemic lupus erythematosus are clustered in two populations of surface structures in apoptotic keratinocytes. J. Exp. Med. 179: 1317–1330 29. Casciola-Rosen L.A., Miller D.K., Anhalt G.J., Rosen A. 1994. Specific cleavage of the 70-kDa protein component of the U1 small nuclear ribonucleoprotein is a characteristic biochemical feature of apoptotic cell death. J. Biol. Chem. 269: 30757–30760 30. Casciola-Rosen L., Anhalt G.J., Rosen A. 1995. DNA-dependent protein kinase is one of a subset of

27

31.

32.

33.

34.

autoantigens specifically cleaved early during apoptosis. J. Exp. Med. 182: 1625–1634 Burlingame R.W., Boey M.L., Starkebaum G., Rubin R.L. 1994. The central role of chromatin in autoimmune responses to histones and DNA in systemic lupus erythematosus. J. Clin. Invest. 94: 184–192 Shuster A.M., Gololobov G.V., Kvashuk O.A., Bogomolova A.E., Smirnov I.V., Gabibov A.G. 1992. DNA hydrolyzing autoantibodies. Science 256: 665–667 Kubota T., Wantnabe N., Kanai Y., Stollar B.D. 1996. Enhancement of oxidative cleavage of DNA by the binding sites of two anti-double-stranded DNA antibodies. J. Biol. Chem. 271: 6555–6561 Mamula M.J. 1995. Lupus autoimmunity: from peptides to particles. Immunol. Rev. 144: 301–314 JAI 828 (CS)