aturally occurring V region connected antibodies inhibit anti-dsDNA antibodyeactivity with dsDNA
atjana Srdic-Rajica,∗, Vladimir Jurisicb, Sladjana Andrejevicc, Branka Bonaci-Nikolicc,imothy Bowkerd, Daniela Concase, Radmila Metlas f
Department of Experimental Oncology, Institute of Oncology and Radiology of Serbia, Pasterova 14, 11000 Belgrade, SerbiaInstitute of Haematology, School of Medicine, University of Kragujevac, 34000 Kragujevac, SerbiaDepartment of Allergology and Immunology, Clinical Centre of Serbia, Faculty of Medicine, University of Belgrade, 11000 Belgrade, SerbiaHomerton University Hospital, Homerton Row, London E9 6SR, UKWezen Biopharmaceuticals, Sardegna Ricerche, Loc. Piscinamanna, 09010 Pula (CA), ItalyCenter for Multidisciplinary Research, Institute of Nuclear Science VINCA, 11001 Belgrade, Serbia
r t i c l e i n f o
rticle history:eceived 7 June 2011eceived in revised form 14 July 2011ccepted 19 July 2011
The production of autoantibodies against a vast array of self antigens, most notably double stranded (ds)DNA, characterized systemic lupus erythematosus (SLE). The purpose of this work is to study specificIg fractions isolated from normal human serum (NHS) and their effect on the binding of anti-double-stranded deoxyribonucleic acid (dsDNA) antibodies (Abs) to dsDNA. A fraction named immunoglobulinG (IgG)-reactive IgG was purified from total NHS IgG by absorption onto (CNBr)-activated Sepharose 4Blinked to intact IgG molecules (IgG-Sepharose column). IgG-reactive IgG was co-incubated with systemiclupus erythematosus (SLE) patient’s serum and binding of the anti-dsDNA Abs to dsDNA was measuredby enzyme-linked immunosorbent assay (ELISA). Co-incubation of SLE patient’s serum with IgG-reactiveIgG resulted in a dose-dependent reduction in binding of anti-dsDNA Abs to dsDNA. A reduction greaterthan 70% was observed at a concentration of 300 �g of IgG-reactive IgG per mL of a 400-fold dilutedSLE patient’s serum whereas total NHS IgG, at the same concentration, resulted in a 10% reduction in
binding. The purification process used to isolate IgG-reactive IgG was based on interactions between intactIg rather than on interactions between F(ab’)2 portions. IgG2 is the predominant immunoglobulin (Ig)subclass in IgG-reactive IgG. Thus, IgG2 might have an important role in the connectivity characteristicsof NHS IgG. The capacity of IgG-reactive IgG to inhibit anti-DNA Ab binding to dsDNA may have potentialapplication in the treatment of SLE. This targeted biological approach may provide an alternative strategyto immunosuppressants.
ntroduction
Despite a rigorous selection process during B cell develop-
ent in the bone marrow a where B lymphocytes with potentially
angerous self-reactivity are eliminated a significant number ofelf-reactive B cell survives to enter the peripheral compartment
(Monroe and Dorshkind 2007). In healthy individuals, a major pointat which reduction in the number of self-reactive B cells takes placeis during transition from IgM naive, to IgM memory B cells beforethe onset of somatic hypermutation. In patients with SLE, this pro-gressive decline in self-reactive cells does not occur, suggestingthat SLE is associated with a failure to establish self-tolerance dur-ing early B cell development which leads to increased numbers ofautoreactive mature naive B cells (Yurasov et al. 2005).
SLE, the most serious of the lupus disorders, is characterized bythe production of a number of autoantibodies which involves anyB cell subset including B1 cells, marginal zone (MZ) B cells, short-lived plasma cells or germinal centre-matured long-lived plasmacells (Jacobi and Diamond 2005). The most common autoantibody
seen in lupus is the antinuclear antibody (ANA). The type of ANApattern helps to determine if SLE or a related connective tissuedisease is present and anti-dsDNA antibodies are the most fre-quently detected antibodies in SLE. Current diagnostic criteria for
LE should include 4 of the ten symptoms to be present at anyime during a patient’s history among which is the dsDNA anti-ody reacting with double stranded deoxyribonucleic acid (DNA)Pearson and Lightfoot 1981). The processes that govern the gen-ration of pathogenic anti-dsDNA Abs in SLE are largely unknown.utoantibodies may appear many years before the diagnosis of SLEith subsequent progressive accumulation of specific autoantibod-
es during the development of this syndrome (Arbuckle et al. 2003).nti-dsDNA Abs may arise as a consequence of polyclonal antigen
Ag)-nonspecific disturbances of B and of T cells which leads ton extensive diversity of anti-DNA Abs, or as a consequence of Ag-riven responses being promoted by polyclonal activation (Dziarski988; Klinman et al. 1994; Singh et al. 1998). The high ratio ofeplacement to silent mutations in the complementary determin-ng regions (CDR) suggests Ag-driven expansion (Shlomchik et al.987; Eilat et al. 1989). Back mutations of SLE patient’s anti-DNA-gG demonstrates that some germline configuration of these Absave no specificity for DNA suggesting that at least some anti-DNAutoantibodies are generated from non-DNA reactive B cells duringhe normal immune response (Wellmann et al. 2005). Basic aminocids introduced by somatic mutations may contribute to anti-DNAbs binding to DNA through electrostatic interactions (Eilat et al.988) supporting Ag-driven positive selection of somatic mutations
n the gene elements that frequently encode multi-reactive IgMutoantibodies (Van Es et al. 1991). Furthermore, the reversion ofomatically mutated anti-dsDNA-IgM, from the same patient, toermline sequence, suggests that both self and non-self Ags arenvolved in B cell activation (Zhang et al. 2008).
In addition to alteration in mutational activity, SLE patientshow alteration of Ig repertoire. A bias toward VH3, VH4-34, and�1 gene family usage was shown, however the Ig gene reper-
oire is not predictive of a self-reactive antibody selection biasn SLE (Yurasov et al. 2005). No particular family of germline Venes has been demonstrated to encode anti-DNA Abs, for morehan twenty apparently different DNA antibody idiotypes haveeen identified, some of which tend to be public or cross-reactiveCRI) (Isenberg et al. 1990). The idiotypic cross-reactions of Ig fromnrelated patients suggest that in SLE autoantibodies are derivedrom related families of germline genes (Shoenfeld et al. 1983).he precise nature of Ags, self or foreign, triggering the produc-ion of anti-dsDNA Abs is not defined yet. Anti-dsDNA Abs canind to a wide variety of antigens, including nucleic acids, nucle-proteins, cell membranes, and phospholipids. There is evidenceor idiotypic and reactivity relationship between anti-DNA Abs andnti-bacterial polysaccharides (PS) (Grayzel et al. 1991; Spellerbergt al. 1995).
Anti-idiotypic Abs reactive with anti-DNA Abs were detected inera of healthy individuals as well as in SLE patients in remissionmplying that autoreactive Abs may be manipulated by a hetero-eneous population of anti-idiotypic Abs (Williams and Isenberg998; Silvestris et al. 1987; Muryoi et al. 1990).
The aims of this study were to separate naturally occurring Vegion connected Abs from NHS by exploiting their reactivity withotal intact NHS IgG, and to test their reactivity against anti-dsDNAbs present in the sera of SLE patients.
aterials and methods
atients’ sera
Sera from five Serbian patients with a confirmed diagnosis of
LE according to the American College of Rheumatology (ACR)evised criteria were studied. Patients (four female and one male,ith a mean age at 46.2 ± 13.22 years) had active disease and high
erum levels of anti-dsDNA Abs. The anti-dsDNA Ab screen ORG
logy 217 (2012) 111– 117
604S, which detect IgG, IgM and IgA isotypes, was used to mea-sure anti-dsDNA Ab level. Serum was collected from patients byperipheral venepuncture according to standard operating proce-dures. The Institution’s Ethics Committee approved the study. Fullinformed consent was obtained from all patients.
Healthy donors’ sera
Sera of 30 healthy Serbian blood donors with no history of SLEor other autoimmune disorders were studied. Serum was collectedby peripheral venepuncture according to standard operating pro-cedures. For each experiment, the pooled serum contained an equalvolume of serum from each donor. Usually 3 mL from each donorwere used. Prior to affinity chromatography procedures, serum washeat inactivated at 56 ◦C for 30 min.
Purification of NHS IgG and of the GammaBind flowthroughfraction
Two millilitres of NHS, containing about 18.0 mg of IgG, wasadded to 10 mL of binding buffer (BF) and then dialysed by centrifu-gation (MW limit 50 kDa). Alternatively, 2 mL of NHS was added to2 mL of 2× BF. Prior to loading to a 2.0 mL GammaBind G (GB) col-umn the mixture was filtered through a 0.2 �m-pore size Milliporefilter. The GB column capacity was 18 mg of human IgG per 1 mL ofdrained gel. After elution IgG containing fractions were collectedand dialysed by centrifugation and thus represents NHS IgG. How-ever, the GB flow through fraction (GBF) contains unbound IgG. Thisfraction was dialysed and concentrated (to original serum volume)by centrifugation in phosphate buffered saline (PBS) prior to deter-mining Ig class concentration by capture ELISA. GBF fractions withIgG concentrations of at least 10 �g per mL were used to separateIgG-reactive GBF Abs on IgG-Sepharose column.
Purification of IgG-reactive IgG and of IgG-reactive GBF Abs
Separation of IgG-reactive IgG from total NHS IgG affinity puri-fied on GB, and of IgG-reactive GBF Abs was performed usingan IgG-Sepharose column (NHS IgG-coupled to CNBr-activatedSepharose 4B). Samples of affinity purified IgG-reactive IgG andof IgG-reactive GBF Abs obtained from six independent runs wereconcentrated by ultrafiltration (YM-50, cut-off MW 50,000). The Igconcentration present in IgG-reactive IgG obtained from total NHSIgG and in IgG-reactive GBF Abs was adjusted to 100 �g per mL.
Detection of rheumatoid factor activity
In order to determine the rheumatoid factor (RF) activity ofIgG-reactive IgG and of IgG-reactive GBF Abs, the quantitativemeasurement of autoantibodies reactive with the Fc portion wasperformed using the RF screen ORGENTEC (ORG) 522S (DiagnostikaGmbH – GTI, Brookfield, WI). This ELISA kit is used for the quan-titative measurement of IgG, IgM and IgA RF in serum or plasma.The cut off was 25 U per mL. The RF level was determined at Ig con-centrations of 100 �g (IgG-reactive IgG) and of 50 �g (IgG-reactiveGBF Abs) per mL. The serum of a rheumatoid arthritis (RA) patient(which contained about 50 �g per mL of IgG at 200-fold dilution,assuming a serum IgG concentration of 10 mg per mL) was testedat 200-fold dilution.
Detection of anti-dsDNA Abs in SLE patients’ serum
The quantitative measurement of autoantibodies reactive withdsDNA was performed using the ELISA kit ORG 604S anti-dsDNA
IgG-reactive IgG (100 �g per mL) 29.9IgG-reactive GBF Abs (50 �g per mL) 47Total NHS IgG (50 �g per mL) 19.6
C
AI
I
wiodoramab(
ive Iive IgG or of IgG-reactive GBF Abs or both
× 100
R
tN0otsr
s
aapcatpcs
r1aoItS
dpdri
in the presence of different concentrations of IgG-reactive IgG arepresented (Fig. 3). The absorbance values for patient serum P5, mea-
1 2 4 80.00
0.25
0.50
0.75
1.00
1.25
1.50
1.75
SLE P1 se rum dilution [x1 02]
A4
50
/63
0 n
m
CB
A
0
50
100
150
200
250
SLE P1 se rum [8x1 02 dil ution]
U/m
l
P10
25
50
75
100
SLE P1 se rum [8x102 dil ution]
% o
f re
du
cti
on
Fig. 1. Binding of anti-dsDNA antibodies from SLE P1 serum on microtiter platescoated with dsDNA. (A) Binding of anti-dsDNA antibodies from SLE P1 serum atdifferent dilution expressed as absorbance A450/630 nm before [-�-] and after coin-cubation with 100 �g per mL of IgG-reactive IgG antibodies [-�-]. The horizontal
RA patient’s serum (200-fold dilution) 800
ut off was 25 U per mL and was not subtracted.
b screen (Diagnostika GmbH – GTI, Brookfield, WI) which detectsgG, IgM and IgA (cut off 25 U per mL).
nhibition of anti-dsDNA Abs binding to dsDNA
The interaction of IgG-reactive IgG and of IgG-reactive GBF Absith anti-dsDNA Abs from SLE patients was determined by assess-
ng the capacity of the IgG-reactive fractions to reduce bindingf anti-dsDNA Abs to dsDNA bound to microwells. SLE serum atefined dilution was co-incubated with defined concentrationsf IgG-reactive IgG, of IgG-reactive GBF Abs or of combined IgG-eactive IgG and IgG-reactive GBF Abs for 30 min at 37 ◦C and thendded to wells coated with dsDNA. Values greater than 25 U perL of anti-dsDNA Abs at serum dilutions of 1:100 were regarded
s positive (as recommended by the manufacturer). Reduction ininding of anti-dsDNA Abs to dsDNA was calculated as follows:
1 − anti-dsDNA Abs bound to dsDNA in the presence of IgG-reactanti-dsDNA Abs bound to dsDNA in the absence of IgG-react
esults
Naturally occurring IgG-reactive Abs were separated both fromotal IgG purified from NHS and from the GBF fraction derived fromHS. Recovery of IgG-reactive IgG from total NHS IgG was up to.7% when the ratio of loaded, to coupled IgG was molar, i.e. 8 mgf loaded IgG to 8 mg of coupled IgG. IgG-reactive IgG showed onlyraces rheumatoid factor (RF) activity at the concentrations used,uggesting that this preparation consists predominantly of an Vegion connected Ab population.
The IgG-reactive GBF Abs consisting of IgG, IgM and IgA, showedome RF activity (Table 1).
As described in Section “Materials and methods”, the level ofnti-dsDNA Abs present in SLE patient’s serum was measured using
commercial ELISA kit. The calibration curve ranged from 0 to 200 Uer mL, where the absorbance (A450/630) of the upper limit of thealibration curve varied slightly from kit to kit being between 1.2nd 1.35 A450/630. From the screening experiments we observedhat the absorbance (A450/630) of all the available SLE serum sam-les, at 100-fold dilution, exceeded that of the upper limit of thealibration curve and therefore the higher dilutions of the patient’serum was introduced.
For the sera from patients #1 (P1), #2 (P2) and #3 (P3), at theecommended 100-fold dilution, the A450/630 was 1.743, 1.649 and.440, respectively. In order to determine the SLE serum level ofnti-dsDNA Abs, and to measure the potential reduction in bindingf anti-dsDNA Abs to dsDNA after co-incubation with IgG-reactivegG, the values of the absorbance of the SLE sera had to fall withinhe range of the calibration curve. Therefore, increased dilutions ofLE sera were used.
The effect of increasing of P1 serum dilutions on binding of anti-sDNA Abs to dsDNA in the absence, and in the presence, of 100 �g
er mL of IgG-reactive IgG is shown in Fig. 1. Increasing P1 serumilution from 100-fold to 200-fold had only a minimal effect ineducing the absorbances measured, either without or with co-ncubation with IgG-reactive IgG (Fig. 1A), and thus was ineffective
logy 217 (2012) 111– 117 113
gG or of IgG-reactive GBF Abs or both)
in bringing the measured absorbance within the range of the cal-ibration curve. This is a reflection of the very high levels of theanti-dsDNA Abs in this patient’s serum. Further increases in thedilutions of P1 serum (to 400-fold and 800-fold) allowed the effectof co-incubation with 100 �g per mL of IgG-reactive IgG to becomedetectable (Fig. 1A).
At 800-fold dilution the A450/630 of P1 serum was below theupper limit of the calibration curve, both in the absence and in thepresence of IgG-reactive IgG, allowing the anti-dsDNA Ab levels tobe expressed as U per mL (Fig. 1B). Co-incubation with 100 �g permL of IgG-reactive IgG reduced the binding of anti-dsDNA Abs todsDNA by almost 60% (Fig. 1C).
Fig. 2A presents the effect of using different serum dilutions(100-, 200-, 400-fold) on the values of absorbance obtained frompatients P2, P3 and P4.
The effect of co-incubation of sera from patients P2, P3 and P4at a dilution of 400-fold with 50 �g per mL of IgG-reactive IgG onbinding of anti-dsDNA Abs to dsDNA was determined (Fig. 2B). Afterco-incubation of 50 �g per mL of IgG-reactive IgG with 400-folddiluted serum, the observed reduction in binding varied betweenpatients expressed either as U per mL or as percent reduction,Fig. 2B and C respectively.
The data from a representative experiment, using serum frompatient P5 at a fixed 400-fold dilution, in which dose-dependentinhibition of binding of anti-dsDNA Abs to dsDNA was obtained
line represents the upper limit of the calibration curve. (B) Binding of anti-dsDNAantibodies from SLE P1 serum at 800-fold dilution (A) expressed as U per mL. Bind-ing levels are presented as a mean of two independent determinations. (C) Bindingof anti-dsDNA antibodies from SLE P1 serum at 800-fold dilution (A) expressed aspercentage of reduction.
114 T. Srdic-Rajic et al. / Immunobiology 217 (2012) 111– 117
A
0.00
0.25
0.50
0.75
1.00
1.25
1.50
1.75P2
P3
P4
SLE serum dilution [x102]
A4
50
/63
0 n
m1 2 4 1 2 4 1 2 4
CB
P2 P3 P4
0
50
100
150
200
250
SLE se rum [4x1 02 dilution ]
U/m
l
P2 P3 P40
20
40
60
SLE serum [4x102 dil ution ]
% o
f re
du
cti
on
Fig. 2. Reduction of anti-dsDNA antibodies binding on microtiter plates coated with dsDNA after coincubation with IgG-reactive IgG antibodies. (A) Binding expressed asabsorbance A450/630 nm of anti-dsDNA antibodies from different SLE patient serum (P2, P3 and P4) at different dilution. The horizontal line represents the upper limit of thecalibration curve. (B) Binding expressed as U per mL of anti-dsDNA antibodies from SLE P2, P3 and P4 serum at 400-fold dilution (A) before [-�-] and after coincubationwith 50 �g per mL of IgG-reactive IgG antibodies [-�-]. Binding levels are presented as a mean of two independent determinations. (C) Percent of reduction of anti-dsDNAantibodies from SLE P2, P3 and P4 serum at 400-fold dilution (A).
1 2 40.00
0.25
0.50
0.75
1.00
1.25
1.50
1.75
SLE P5 serum dilution [x102]
A4
50
/63
0 n
m
B C
A
0 100 200 300 4000
20
40
60
80
IgG concentr ation [µg/ml]
% o
f re
du
cti
on
1 2 3 4 50.00
0.25
0.50
0.75
1.00
1.25
1.50
1.75
SLE P5 serum [4x 102 dil ution ]
A4
50
/63
0 n
m
1 2 3 4 5
Fig. 3. Effect of IgG-reactive Abs and total NHS IgG on SLE P5 serum anti-dsDNA antibodies binding to microtiter plates coated with dsDNA. (A) Binding of anti-dsDNA Absfrom SLE P5 serum at different dilution expressed as absorbance A450/630 nm. The horizontal line represents the upper limit of the calibration curve. (B) Anti-dsDNA Absbinding from SLE P5 serum at 400-fold dilution expressed as absorbance A450/630 nm before (1) and after coincubation with IgG-reactive IgG antibodies at concentrationof 100 �g per mL (2) and 300 �g per mL (3) as well as coincubation with total NHS IgG at concentration of 100 �g per mL (4) and 300 �g per mL (5). (C) Dose dependentbinding of anti-dsDNA to dsDNA expressed as percent reduction of A450/630 nm (B) for IgG-reactive Abs (-�-) and total NHS IgG (-�-). The horizontal line defines cut off valueof anti-dsDNA and IgG-reactive IgG Abs binding at concentration 300 �g per mL were in that range (B).
T. Srdic-Rajic et al. / Immunobio
0 2 4 6 8 10
0
15
30
45
60
SLE P2 serum dilution [x102]
% o
f re
du
cti
on
Fig. 4. Binding of anti-dsDNA from SLE P2 patients serum coincubated with IgG-reactive IgG and IgG-reactive GBF on microtiter plates coated with dsDNA. SLE P2serum at 400-fold dilution and 800-fold dilution was coincubated with IgG-reactiveIgG antibodies at 50 �g per mL (-�-) and IgG-reactive GBF antibodies at 4 �g per mL(G
sl
oc3gwr
iAdac(
dP
D
pwclb(
otbanGw
saT(I
SLE is a complex disease characterized by serological (e.g.
-�-). Coincubation of SLE P2 serum with mixed IgG-reactive IgG and IgG-reactiveBF antibodies at 54 �g per mL (-�-).
ured at a dilution of 400-fold, were consistently above the upperimit of the calibration curve (Fig. 3A).
Thus, Fig. 3B expresses the effect of different concentrationsf IgG-reactive IgG on binding of anti-dsDNA Abs to ds-DNA ashanges in absorbance values. IgG-reactive IgG at concentration of00 �g per mL showed the reduction in anti-DNA antibody bindingreater than 70% (Fig. 3C). In contrast, co-incubation of P5 serumith total NHS IgG at a concentration of 300 �g per mL led to a
eduction in binding of only 10%.The reduction in binding was obtained by separate co-
ncubation of IgG-reactive IgG (50 �g per mL) and IgG-reactive GBFbs (4 �g per mL) with patient P2 serum at 400-fold and at 800-foldilutions as well as with combined antibody preparations. Thereppears to be an additive effect of combined preparations, which islear at 400-fold dilution and barely detectable at 800-fold dilutionFig. 4).
Neither IgG-reactive IgG nor total NHS IgG showed binding tosDNA above the cut off value (Fig. 3B, horizontal line) as well asBS buffer alone has no effect (results not presented).
iscussion
The V region connected Abs isolated from NHS are capable ofreventing anti-dsDNA Ab binding to dsDNA when co-incubatedith SLE patients’ serum. Our results are consistent with the
oncepts that the idiotypic network may play a role in SLE by regu-ating disease activity, and that therapy of autoimmune disease cane based on manipulation of idiotype–anti-idiotype interactionsWilliams et al. 1995b; Singh et al. 1995).
IgG-reactive IgG and IgG-reactive GBF Abs were separated basedn an affinity chromatography method which exploits the interac-ion between intact antibody molecules (Metlas et al. 2009) and notetween F(ab’)2 portions, as described by other authors (Williamsnd Isenberg 1998; Silvestris et al. 1987). To obtain V region con-ected Abs, intact total IgG molecules separated from NHS onammaBind G Sepharose were used to generate the IgG-Sepharoseith which total intact NHS IgG or the GBF fraction were processed.
Affinity chromatography was carried out under rather high ionictrength conditions in order to reduce unspecific binding and/orvoid purification of low affinity Abs present in NHS, such as RFs.
hese natural RFs are generally polyreactive IgM of low affinitySoltys et al. 1997; Van Esch et al. 2001). Neither total NHS IgG norgG-reactive IgG displayed RF activity at a 100 �g per mL (Table 1).
logy 217 (2012) 111– 117 115
However, some RF activity was observed for IgG-reactive GBF Abs(Table 1) which contain IgM and IgA isotypes.
The classical concept of Ig function is that solely the variable (V)domains define antibody affinity and specificity (Lieber 2000; Liet al. 2004), whereas constant (C) domains have effector functionsnot directly affecting the specificity for the Ag. Recently this viewwas challenged by the observation that V region-identical antibod-ies with different C heavy (H) regions showed differences in themagnitude of their binding to the Ag and in fine specificity andidiotype (Reitan and Hannestad 2002; Torres et al. 2007; Torreset al. 2005). Furthermore, the CH domain can regulate immunityand tolerance to idiotypes (Ids) of the antibody V-region so that, forexample, idiotypes that have switched to IgG may become tolero-genic (Reitan and Hannestad 2002). Isotype switching, thus, maybe another mechanism for generating diversity in Ag binding. Inthis regard different CH domains can impose diverse structural con-straints on the interaction of Ab with Ag, especially multivalentAgs such as PS, thus offering an explanation for isotype restriction(McLean et al. 2002).
Considering that isotype could affect antibody specificity, it wasdecided to separate IgG-reactive, i.e. V connected, Abs by exploit-ing the interaction between intact Ig molecules present in NHS.This method of purification resulted previously in the isolationfrom NHS of IgG-reactive fractions enriched in IgG2 monomers anddimers (Metlas et al. 2009). These and the current data seem to indi-cate that among IgG subclasses IgG2 might be the most connectedIgG subclass.
IgG-reactive preparations did not show binding to dsDNA at theconcentrations used for co-incubation with SLE patient’s serum.This suggests that, under such conditions, NHS anti-DNA activitieswere not co-purified with IgG-reactive preparations in a signifi-cant amount. In contrast, binding to dsDNA was reported by otherauthors for IgG anti-F(ab’)2 obtained from NHS (McLean et al. 2002).
Idiotypic analysis, as a marker of VH and VL gene expression,showed that in SLE a variety of VH genes are used to encode autoan-tibodies reactive with dsDNA (Isenberg et al. 1990). Interestingly,anti-dsDNA public or cross-reactive idiotypes (CRI) are used also inprotective Ab responses in non-autoimmune individuals (Grayzelet al. 1991; Shoenfeld et al. 1986; Naparstek et al. 1985). However,in SLE patients such antibodies might be present at an increasedtitre. Thus, the presence of conserved V regions encoding idiotypi-cally related polyspecific Igs with the capacity to bind different Ags(such as bacterial PS and nucleic acids), was suggested.
The 16/6 and 9G4 idiotype positive Abs, encoded by a germlinegene from the human VH4 gene family, are expressed in a significantnumber of patients with SLE. However, even if a restricted V geneusage is suggested, such restriction does not appear to be skewedtowards particular gene families. Nevertheless, this observation isimportant and provides the rationale for proposing the use of anti-idiotypes as a novel therapeutic strategy to control the disease.
Idiotypes and anti-idiotypic Abs coexist in the repertoire of asingle individual. They occur spontaneously or can be induced.Anti-idiotypic IgG which reacts with anti-DNA Abs has beendetected in healthy individuals (Williams and Isenberg 1998) aswell as in SLE patients, in whom the serum titre of these genericAbs oscillates during active disease and remitting phases (Zoualiand Eyquem 1983; Williams et al. 1995a). In MRL/Mp-lpr/lpr micea mosaic of anti-DNA Ab idiotopes against which autoantibodiesare made spontaneously in SLE was identified (Ward et al. 1997).As the disease progresses, there is a switch from IgM to IgG reactingwith anti-DNA idiotopes located both in the CDR and the FR region(Ward et al. 1997).
anti-double-strand DNA), clinical (elevation of erythrocyte sed-imentation rate, leukopenia and proteinuria) and histological(immune complex deposition in kidneys) parameters. Importantly,
LE is the disease, which lacks a uniform response to therapy.orticosteroids, antimalarials (hydroxychloroquine), and cytotoxicrugs are classically used as medication in SLE (Monneaux anduller 2009). Major side effects of these conventional drugs may
everely counterbalance the clinical outcomes of treated patients.or this reason collective research has led to the development ofore targeted approaches such as monoclonal antibodies (mAbs)
ituximab and belimumab (benlysta) targeting B cells (Monneauxnd Muller 2009) for treating patients with SLE.
On the other hand naturally occurring anti-idiotypic activitygainst anti-DNA antibodies might not be restricted to B cells, i.e.
cell products as here shown, since T cells also react to V regioneterminants of antibodies.
Evidence has accumulated in support of a potential role for Tell responses to idiotypic determinants. Thus, the administrationf CDR-based peptides (pCDR1 and pCDR3) expressed by anti-NA antibodies could prevent or down-regulate experimental SLE
nduced in mice (Zinger et al. 2003).The IgG-reactive preparations tested i.e. population of Abs
hich share complementary structures with other antibodieshowed the capacity to prevent anti-DNA Ab binding to dsDNA.
simple conclusion might be that anti-DNA Abs are idiotypicallyonnected to normal Igs and that IgG-reactive Abs recognize onnti-DNA Abs structural counterparts of normal Ab. Interestingly,n spontaneous murine SLE was reported the presence of anti-diotypic Abs to anti-DNA idiotopes acting as natural mimotopesncoded by CRI (located at CDR-H2/FR-H3 junction) and express-ng dual specificity for dsDNA and anti-DNA idiotype (Eivazova et al.000). These natural anti-idiotypic Abs with dual specificity (epi-ody activity) produced in the course of the disease may drive theathologic development of SLE (Eivazova et al. 2000) and/or facili-ate removal of DNA-immune complexes from circulation (Fischelnd Eilat 1992).
onclusions
This study showed that co-incubation of SLE patients’ serumith IgG-reactive IgG at 300 �g per mL resulted in about a 75%
eduction in binding of anti-dsDNA Abs to dsDNA bound to microw-lls while total NHS IgG, at the same concentration, inhibited bynly 10% binding of anti-dsDNA Abs to dsDNA.
Our data suggest that IgG-reactive Abs might share V regionomplementarity with anti-dsDNA Abs and may inhibit the inter-ction of high titre and/or high affinity anti-dsDNA Abs with dsDNA.ue to the heterogeneous autoantibody population present in SLE,either the targeted anti-DNA Ab(s), nor the anti-idiotypic anti-sDNA Ab(s) and the idiotope involved in the interaction with
gG-reactive preparations were defined in the current study. How-ver, the finding that Abs against anti-DNA Abs from healthyubjects recognize preferentially idiotype determinants in the FRegion common among some anti-DNA Abs (Sasaki et al. 1988),nd the presence of IgG2 dimers in the IgG-reactive preparationsMetlas et al. 2009) seem to indicate that IgG2 stoichiometry mighte possibly crucial for the observed reduction in dsDNA binding.
Since the intravenous immunoglobulins contain a low amountf immunoregulatory antibodies therefore to achieved therapeuticffect a high doses of IVIg are used to treat a variety of diseases, suchs SLE. The V region connected Abs isolated from NHS, enrichedn immunoregulatory antibodies could achieved at least the sameherapeutic effect as intravenous immunoglobulins with a signifi-antly lower total immunoglobulin concentration.
ompeting interests
The authors declare that they have no competing interests.
logy 217 (2012) 111– 117
Contributions
R.M. and T.S.-R. designed the study, performed the experiments,analysed and interpreted the data, drafted the manuscript. V.J., S.A.and B.B.-N. participated in the design of the experiments, enrolledthe patients with SLE, provided the sera, revised critically themanuscript and contributed to the discussion. D.C. and T.B. revisedcritically the manuscript and added important points to the discus-sion. All the authors approved the final draft of the manuscript.
Funding
This project was funded in part by the Ministry of Science ofRepublic of Serbia (Grants Nos. III41026 and 175065).
References
Arbuckle, M.R., McClain, M.T., Rubertone, M.V., Scofield, R.H., Dennis, G.J., James,J.A., Harley, J.B., 2003. Development of autoantibodies before the clinical onsetof systemic lupus erythematosus. N. Engl. J. Med. 349, 1526–1533.
Dziarski, R., 1988. Autoimmunity: polyclonal activation or antigen induction?Immunol. Today 9, 340–342.
Eilat, D., Webster, D.M., Rees, A.R., 1988. V region sequences of anti-DNA and anti-RNA autoantibodies from NZB/NZW F1 mice. J. Immunol. 141, 1745–1753.
Eilat, D., Hochberg, M., Tron, F., Jacob, L., Bach, J.F., 1989. The VH gene sequencesof anti-DNA antibodies in two different strains of lupus-prone mice are highlyrelated. Eur. J. Immunol. 19, 1241–1246.
Eivazova, E.R., McDonnell, J.M., Sutton, B.J., Staines, N.A., 2000. Cross-reactivity ofantiidiotypic antibodies with DNA in systemic lupus erythematosus. ArthritisRheum. 43, 429–439.
Fischel, R., Eilat, D., 1992. Structure and binding properties of a monoclonal anti-idiotypic autoantibody to anti-DNA with epibody activity. J. Immunol. 149,3089–3096.
Grayzel, A., Solomon, A., Aranow, C., Diamond, B., 1991. Antibodies elicited by pneu-mococcal antigens bear an anti-DNA-associated idiotype. J. Clin. Invest. 87,842–846.
Isenberg, D., Williams, W., Axford, J., Bakimer, R., Bell, D., Casaseca-Grayson, T., Dia-mond, B., Ebling, F., Hahn, B., Harkiss, G., et al., 1990. Comparison of DNA antibodyidiotypes in human sera: an international collaborative study of 19 idiotypesfrom 11 different laboratories. J. Autoimmun. 3, 393–414.
Jacobi, A.M., Diamond, B., 2005. Balancing diversity and tolerance: lessons frompatients with systemic lupus erythematosus. J. Exp. Med. 202, 341–344.
Klinman, D.M., Shirai, A., Ishigatsubo, Y., 1994. Polyclonal B cell activation and B cellcross-reactivity during autoantibody production in systemic lupus erythemath-osus. Adv. Exp. Med. Biol. 347, 115–123.
Li, Z., Woo, C.J., Iglesias-Ussel, M.D., Ronai, D., Scharff, M.D., 2004. The generation ofantibody diversity through somatic hypermutation and class switch recombi-nation. Genes Dev. 18, 1–11.
Lieber, M., 2000. Antibody diversity: a link between switching and hypermutation.Curr. Biol. 10, R798–R800.
McLean, G.R., Torres, M., Elguezabal, N., Nakouzi, A., Casadevall, A., 2002. Isotype canaffect the fine specificity of an antibody for a polysaccharide antigen. J. Immunol.169, 1379–1386.
Metlas, R.M., Srdic, T.V., Jenabian, M.A., Elluru, S., Carta, L., Bowker, T.J., Kaveri, S.V.,Concas, D., 2009. Immunoglobulin G-reactive antibodies from sera of healthyindividuals enriched in IgG2 – therapeutic potential in HIV-1 infection. Curr.HIV Res. 7, 378–383.
Monroe, J.G., Dorshkind, K., 2007. Fate decisions regulating bone marrow and periph-eral B lymphocyte development. Adv. Immunol. 95, 1–50.
Muryoi, T., Sasaki, T., Hatakeyama, A., Shibata, S., Suzuki, M., Seino, J., Yoshinaga,K., 1990. Clonotypes of anti-DNA antibodies expressing specific idiotypes inimmune complexes of patients with active lupus nephritis. J. Immunol. 144,3856–3861.
Naparstek, Y., Duggan, D., Schattner, A., Madaio, M.P., Goni, F., Frangione, B., Stollar,B.D., Kabat, E.A., Schwartz, R.S., 1985. Immunochemical similarities betweenmonoclonal antibacterial Waldenstrom’s macroglobulins and monoclonal anti-DNA lupus autoantibodies. J. Exp. Med. 161, 1525–1538.
Pearson, L., Lightfoot Jr., R.W., 1981. Correlation of DNA-anti-DNA association rateswith clinical activity in systemic lupus erythematosus (SLE). J. Immunol. 126,16–19.
Reitan, S.K., Hannestad, K., 2002. Immunoglobulin heavy chain constant regions reg-ulate immunity and tolerance to idiotypes of antibody variable regions. Proc.Natl. Acad. Sci. U. S. A. 99, 7588–7593.
Sasaki, T., Muryoi, T., Takai, O., Tamate, E., Saito, H., Yoshinaga, K., 1988. Binding
specificity of antiidiotypic autoantibodies to anti-DNA antibodies in humans. J.Clin. Invest. 82, 748–754.
Shlomchik, M.J., Aucoin, A.H., Pisetsky, D.S., Weigert, M.G., 1987. Structure and func-tion of anti-DNA autoantibodies derived from a single autoimmune mouse. Proc.Natl. Acad. Sci. U. S. A. 84, 9150–9154.
hoenfeld, Y., Isenberg, D.A., Rauch, J., Madaio, M., Stollar, B.D., Schwartz, R.S., 1983.Idiotypic cross-reactions of monoclonal human lupus autoantibodies. J. Exp.Med. 158, 718–730.
hoenfeld, Y., Vilner, Y., Coates, A.R., Rauch, J., Lavie, G., Shaul, D., Pinkhas, J., 1986.Monoclonal anti-tuberculosis antibodies react with DNA, and monoclonal anti-DNA autoantibodies react with Mycobacterium tuberculosis. Clin. Exp. Immunol.66, 255–261.
ilvestris, F., Williams Jr., R.C., Frassanito, M.A., Dammacco, F., 1987. In vitro inhibi-tion of anti-DNA producing cells from systemic lupus erythematosus patientsby autologous anti-F(ab’)2 antibodies. Clin. Immunol. Immunopathol. 42,50–62.
ingh, R.R., Ebling, F.M., Sercarz, E.E., Hahn, B.H., 1995. Immune tolerance toautoantibody-derived peptides delays development of autoimmunity in murinelupus. J. Clin. Invest. 96, 2990–2996.
ingh, R.R., Hahn, B.H., Tsao, B.P., Ebling, F.M., 1998. Evidence for multiple mech-anisms of polyclonal T cell activation in murine lupus. J. Clin. Invest. 102,1841–1849.
oltys, A.I., Axford, J.S., Sutton, B.J., 1997. Rheumatoid factors: where are we now?Ann. Rheum. Dis. 56, 285–296.
pellerberg, M.B., Chapman, C.J., Mockridge, C.I., Isenberg, D.A., Stevenson, F.K., 1995.Dual recognition of lipid A and DNA by human antibodies encoded by the VH4-21gene: a possible link between infection and lupus. Hum. Antibodies Hybridomas.6, 52–56.
orres, M., May, R., Scharff, M.D., Casadevall, A., 2005. Variable-region-identicalantibodies differing in isotype demonstrate differences in fine specificity andidiotype. J. Immunol. 174, 2132–2142.
orres, M., Fernández-Fuentes, N., Fiser, A., Casadevall, A., 2007. The immunoglobu-lin heavy chain constant region affects kinetic and thermodynamic parametersof antibody variable region interactions with antigen. J. Biol. Chem. 282,
13917–13927.
an Es, J.H., Gmelig Meyling, F.H., Van de Akker, W.R., Aanstoot, H., Derksen, R.H.,Logtenberg, T., 1991. Somatic mutations in the variable regions of a humanIgG anti-double-stranded DNA autoantibody suggest a role for antigen in theinduction of systemic lupus erythematosus. J. Exp. Med. 173, 461–470.
logy 217 (2012) 111– 117 117
Van Esch, W.J., Reparon-Schuijt, C.C., Levarht, E.W., Van Kooten, C., Breedveld, F.C.,Verweij, C.L., 2001. Differential requirements for induction of total immunoglob-ulin and physiological rheumatoid factor production by human peripheral bloodB cells. Clin. Exp. Immunol. 123, 496–504.
Ward, F.J., Knies, J.E., Cunningham, C., Harris, W.J., Staines, N.A., 1997. Natural anti-bodies that react with V-region peptide epitopes of DNA-binding antibodiesare made by mice with systemic lupus erythematosus as disease develops.Immunology 92, 354–361.
Wellmann, U., Letz, M., Herrmann, M., Angermüller, S., Kalden, J.R., Winkler, T.H.,2005. The evolution of human anti-double-stranded DNA autoantibodies. Proc.Natl. Acad. Sci. U. S. A. 102, 9258–9263.
Williams, W.M., Isenberg, D.A., 1998. Naturally occurring anti-idiotypic antibodiesreactive with anti-DNA antibodies in systemic lupus erythematosus. Lupus 7,164–175.
Williams Jr., R.C., Malone, C.C., Huffman, G.R., Silvestris, F., Croker, B.P., Ayoub, E.M.,Massengill, S., 1995a. Active systemic lupus erythematosus is associated withdepletion of the natural generic anti-idiotype (anti-F(ab’)2) system. J. Rheuma-tol. 22, 1075–1085.
Williams, W.M., Staines, N.A., Muller, S., Isenberg, D.A., 1995b. Human T cellresponses to autoantibody variable region peptides. Lupus 4, 464–471.
Yurasov, S., Wardemann, H., Hammersen, J., Tsuiji, M., Meffre, E., Pascual, V., Nussen-zweig, M.C., 2005. Defective B cell tolerance checkpoints in systemic lupuserythematosus. J. Exp. Med. 201, 703–711.
Zhang, J., Jacobi, A.M., Wang, T., Diamond, B., 2008. Pathogenic autoantibodies insystemic lupus erythematosus are derived from both self-reactive and non-self-reactive B cells. Mol. Med. 14, 675–681.
Zinger, H., Eilat, E., Meshorer, A., Mozes, E., 2003. Peptides based on thecomplementarity-determining regions of a pathogenic autoantibody miti-gate lupus manifestations of (NZB × NZW)F1 mice via active suppression. Int.
Immunol. 15, 205–214.
Zouali, M., Eyquem, A., 1983. Idiotypic/antiidiotypic interactions in systemic lupuserythematosus: demonstration of oscillating levels of anti-DNA autoantibodiesand reciprocal antiidiotypic activity in a single patient. Ann. Immunol. (Paris)134C, 377–391.
