вход по аккаунту


Nucleosome-restricted antibodies are detected before anti-dsdna andor antihistone antibodies in serum of mrl-mp lprlpr and ++ mice and are present in kidney eluates of lupus mice with proteinuria.

код для вставкиСкачать
Number 11, November 1994, pp 1684-1688
0 1994, American College of Rheumatology
Objective. To compare the humoral response to
nucleosomes with the response to their individual components (double-stranded DNA [dsDNA] and histones)
and to assess the involvement of antinucleosome antibadies in immune deposits in the kidney of MRL mice.
Methods. We used enzyme-liked immunosorbent assays of sera and kidney eluates for antibody
activity against purified nucleosomes, dsDNA, and histones.
Results. Antinucleosome antibodies emerged before anti-dsDNA and antihistone antibodies. A fraction
of antinucleosome antibodies reacted exclusively with
nucleosomes and not with their components, dsDNA and
histones. These nucleosome-restricted antibodies were
detected in the proteinuric MRL mouse kidney eluate.
Conclusion. Our findings support the notion that
nucleosomes play a major role in the emergence of
antinuclear autoantibodies and that antinucleosome antibodies might be involved in the nephritogenic process
in murine lupus.
A recent study of the genesis and evolution of antichromatin antibodies in murine lupus suggested a T
cell-dependent immunization with chromatin (4). Interestingly, nucleosomes, the fundamental repeating
units of chromatin generated by cell apoptosis, have
been detected as multimeric particles in plasma from
patients with SLE ( 5 ) . Furthermore, nucleosomes
have recently been found to be a major immunogen for
pathogenic autoantibody-inducing T cells in lupusprone mice (6). Thus, nucleosomes, which are core
histone octamers wrapped with 160-basepair DNA,
could be the autoantigen initiating the autoimmune
response to dsDN A and histones (their components)
and play an etiopathogenic role in SLE. To test this
hypothesis, we studied the humoral response to nucleosomes compared with the response to their individual
components (dsDNA and histones), and we assessed
the involvement of antinucleosome antibodies in immune deposits in the kidney of MRL mice.
Anti-double-stranded DNA (anti-dsDNA) and
antihistone antibodies have been found in murine
models of systemic lupus erythematosus (SLE) (1). A
number of experimental findings suggest that the production of these autoantibodies is antigen-driven (2,3),
but the origin and nature of the antigen(s) are elusive.
Zahir Amoura, MD: Hbpital Necker, Paris, France; Henri
Chabre, PhD: Hbpital Necker; Sophie Koutouzov, PhD: Hbpital
Necker; Chantal Lotton, MSc: Hbpital Necker; Alban Cabrespines,
PhD: Hbpital Necker; Jean-Fransois Bach, MD, PhD: Hbpital
Necker; Laurent Jacob, MD, PhD: HBpital Necker and the Hbpital
Tenon, Paris, France.
Address reprint requests to Zahir Amoura, MD, Unit6
INSERM 25, HGpital Necker, 161 Rue de Stvres, 75745 Paris,
Submitted for publication March 24, 1994; accepted in
revised form June 12, 1994.
Mice and sera. Thirty MRL-Mp-lprlfpr mice (20 females and 10 males) were obtained from Harlan France
(Gannat, France), and 15 MRL-Mp-+/+ (10 females and 5
males) and 5 C3H mice (3 females and 2 males) were
obtained from the CSAL mouse breeding colony (Orleans,
France). Mice were kept in a conventional environment and
bled from the retroorbital plexus every 15 days beginning at
age 8 weeks until 26 weeks and 32 weeks of age for lprllpr
and +/+ mice, respectively. Serum was decomplemented at
56°C for 30 minutes and stored at -20°C until used.
Antibodies. PME 77 is a murine monoclonal antidsDNA antibody (MAb) described in detail elsewhere (7),
and i5F7 is a murine antihistone H2B MAb (kindly provided
by Dr. Dan Eilat, Jerusalem, Israel). PME 77 and 15F7 were
purified on a dsDNA-cellulose column (Sigma, St. Louis,
MO) and on a protein G-Sepharose 4 fast-flow column
(Pharmacia, Uppsala, Sweden), respectively. All antibody
preparations reacted with coated nucleosomes by enzymelinked immunosorbent assay (ELISA).
Nucleosomes, dsDNA, and histones. Nucleosomes
were prepared according to the method described by Lutter
(8). Briefly, they were isolated by digesting chromatin from
mouse erythroleukemia L1210 cell nuclei with Stuphylococcus uureus nuclease. HI and nonhistone proteins were
stripped by adding 5M NaCl dropwise (final concentration
O . S 5 M ) , at 0°C. Samples of the stripped chromatin were
applied to 10-ml sucrose gradients (5-20%) prepared in 10
mM Tris, 1 mM EDTA, pH 7.4 (TE), 0.5SM NaCl, and 0.2
mM phenylmethylsulfonyl fluoride. After overnight centrifugation at 200,00Og, nucleosome fractions were collected by
spectrophotometry at optical density 260 nm (OD,,,,).
Fractions corresponding to purified mononucleosomes on polyacrylamide gel electrophoresis were extensively dialyzed against TE, concentrated, and stored on ice
for no longer than 2 weeks. Histone gel analysis (9) showed
that the nucleosome preparation consisted of equal amounts
of the 4 core histones. Lambda phage dsDNA and total
histones were obtained from Boehringer (Mannheim, Germany).
ELISA. Purified mononucleosomes, dsDNA, and histones were dissolved in phosphate buffered saline (PBS) at
final concentrations of 5, 3.75, and 2 pg/ml, respectively.
Each substrate solution (100 pl) was added to Luxlon
microtiter plates (CML, Nemours, France), or to precoated
poly-L-lysine plates when dsDNA was used as substrate. All
plates were incubated overnight at 4"C, but plates coated
with dsDNA were first incubated for 2 hours at 37°C. Wells
were washed with PBS-O.I% Tween, pH 7.4 (PBS-Tween)
and postcoated for 2 hours with 0.1 ml of PBS-10% fetal calf
serum, pH 7.4. After washing, mouse sera (1: 100 dilution) or
kidney eluates diluted in PBS-Tween were added, and
allowed to react for 2 hours.
Bound antibodies were detected with peroxidaseconjugated goat anti-y or anti-p antisera (Sigma). Binding
was measured by adding ABTS (Southern Biotechnology ,
Birmingham, AL). OD was read at 405 nm by an automated
spectrophotometer (Dynatech, Alexandria, VA). Uncoated
plates were used to determine background activity, which
was subtracted from values obtained with coated plates.
For inhibition studies, we used the same ELISA
procedures, except that the sera were incubated for 2 hours
with the appropriate inhibiting substrate before being added
to the coated plates. To determine IgG concentrations,
plates were coated with goat anti-mouse IgG and the ELISA
was performed as above. The ODs generated by sera and
kidney eluates were compared with a standard curve generated by using known serial concentrations of a purified MAb
of the IgG isotype.
Elution of immunoglobulins from kidneys. Ig was
eluted from mouse kidney as previously described (10). Briefly,
14 kidneys were pooled, cut into small pieces while still frozen,
suspended in PBS, pH 7.2, and then homogenized in a chilled
Waring blender at 0°C. The homogenate was washed 3 times
in PBS, and the pellet was resuspended and incubated for
2 hours at 37°C in 0.02M citrate buffer, pH 3.2. After elution,
the suspension was centrifuged (4,OOOg for 30 minutes), and
the eluates were neutralized with 1M Na,HPO,. After over-
night dialysis against PBS, eluates were concentrated, adjusted to a 2-ml volume, and stored at 4°C.
Adsorption of anti-dsDNA antibodies on dsDNAcellulose. Batches of dsDNA-cellulose (Sigma) were equilibrated overnight in TE and 150 mM NaCl (TE buffer), pH
7.4, at 4°C. To remove free or weakly bound dsDNA, the
dsDNA-cellulose was washed extensively with TE and 2M
NaCI, pH 8.0, and was further equilibrated in 40 ml of TE
buffer. Sera or kidney eluates were applied to dsDNAcellulose batches and left on a roller for 12 hours at 4°C.
Batches were then centrifuged for 5 minutes at 400g, and the
supernatants, which contained the unadsorbed part of the
pooled sera or kidney eluates, were tested at the same IgG
concentration before and after adsorption.
Statistical analysis. Threshold values were determined for each ELISA substrate, and were defined as the
mean + 3 SD for the group of 8-week-old lprllpr mice. These
OD cut-off values were as follows: IgG antinucleosome
0.104; IgG anti-dsDNA 0.093; IgG antihistone 0.124; IgM
antinucleosome 0.027; IgM anti-dsDNA 0.030, and IgM
antihistone 0.031. All values for the 8-week-old lprllpr mice
fell within this range.
Antinucleosome antibodies emerge before antidsDNA and antihistone antibodies in lprllpr and +/+
mice. Based on the defined threshold values, each
lprllpr and +/+ mouse (male and female) developed an
IgM and an IgG response to nucleosomes. No antinucleosome antibodies were detected in control C3H
mice. In lprllpr mice, the median age at detection of
antinucleosome activity was 14 weeks (range 10-20
weeks) for IgM and I2 weeks (range 10-18 weeks) for
IgG compared with 25 weeks (range 10-30 weeks) and
23 weeks (range 14-32 weeks), respectively, in +/+
To compare the kinetics of the autoimmune
response against mononucleosomes, dsDNA, and histones, the percentage of mice with values over the
threshold was determined for each substrate, and at
each bleeding time. For both strains of lupus-prone
mice, antinucleosome antibodies emerged before
anti-dsDNA and antihistone antibodies (Figures 1A
and A'). For example, 50% of the lprllpr mice were
positive for antinucleosome IgG at 1 1 weeks of age,
while this percentage was reached at age 14 weeks and
after age 26 weeks for anti-dsDNA and antihistone
IgG, respectively (Figure 1A). In the +/+ strain, 50%
of the mice were positive at 25, 32, and >32 weeks of
age for antinucleosome, anti-dsDNA, and anti-histone
IgG, respectively (Figure 1A').
The lag between the antinucleosome and
anti-dsDNA responses was shorter in lprllpr than in
+I+ n7
Ipr I1 1
Figure 1. Time-course comparison of antinucleosome (O), antidouble-stranded DNA (anti-dsDNA) (D),
and antihistone (*)IgG
production in lupus-prone mice. Results are expressed as the
percentage of values in 30 MRL-Mp-lpr/lpr (A) and 15 MRL-Mp+/+ (A’) mice over threshold values, at each bleeding time. Threshold values were determined for each antigen, and were the mean +
3 SD in the group of 8-week-old MRL-Mp-lpr//pr mice. Two patterns
were observed in individual mice: antinucleosome antibodies appeared before (Band B’)anti-dsDNA and/or antihistone antibodies,
or concomitantly (C and C ’ ) . Results are expressed as optical
density (O.D.) values.
+/+ mice. For example, 50% of the mice became
positive for antinucleosome IgG and then for
anti-dsDNA IgG with a lag time of 3 weeks in lprllpr
mice, compared with 7 weeks in +/+ mice. The
immune response of individual mice showed two general patterns: mice with a lag between the appearance
of antinucleosome and anti-dsDNA IgG (Figures 1B
and B’) (24 of 30 lprllpr mice and 13 of 15 +/+ mice),
and mice in which antinucleosome and anti-dsDNA
IgG appeared concomitantly (Figures 1C and C ’ ) (6 of
30 lprllpr and 2 of 15 +/+). The same pattern was
observed for the IgM response in both strains (data not
shown). It is of interest that we never detected
anti-dsDNA or antihistone antibodies without also
detecting antinucleosome antibodies.
Nucleosome-restricted antibody activity in the
serum of lupus mice. We next determined whether the
early antinucleosome activity was due to nucleosomerestricted antibodies, i.e., antibodies which bind only
to epitopes present on nucleosomes, and not isolated
dsDNA or histones. Inhibition studies were performed
on sera from 4 +/+ and 3 lprllpr mice, all of which
exhibited antinucleosome activity without
anti-dsDNA or antihistone activity. Table 1 shows that
incubation of the sera with 25 pg of purified nucleosomes inhibited about 70% of antinucleosome activity,
while this activity was virtually unaffected by incubation with 25 pg of dsDNA or histones. In contrast, as
positive controls, the antinucleosome activity of
anti-dsDNA and anti-H2B histone MAb, which bound
not only to their related antigens, but also to coated
nucleosomes, was similarly inhibited, whether they
were incubated with nucleosomes or with their specific
In another experiment, sera from 7 +/+ mice
with antinucleosome activity but no anti-dsDNA activity were pooled and adsorbed on dsDNA-cellulose.
The antinucleosome activity (0.590 OD units) remained unchanged after adsorption on DNA-cellulose
(0.610 OD units), indicating that antinucleosome antibodies without anti-dsDNA activity were clearly a
subset distinct from anti-dsDNA antibodies. As a positive control, in pooled sera from 7 lprllpr mice with
Table 1. Inhibition of antinucleosome activity by nucleosomes,
dsDNA. and histones*
units Nucleosomes dsDNA Histones
MRL-Mp- +/+
Mouse 1
Mouse 3
Mouse 4
Mouse 9
Mouse 12
Mouse 16
Mouse 19
Affinity-purified MAb
Anti-dsDNA PME 77
Antihistone 15F7
* Sera diluted 1:lOO or control monoclonal antibodies (MAb) at 1
pglml were preincubated with purified mononucleosomes, doublestranded DNA (dsDNA), or total histones (25 Fg/ml) and then tested
for antinucleosome activity by enzyme-linked immunosorbent assay
(ELISA). Values are the percentage of inhibition of antinucleosome
activity in comparison with sera incubated without inhibitors. OD =
optical density generated by sera (1: 100) or control MAb (1 pglml) in
an antinucleosome ELISA. ND = not determined.
1000 1
4 before adsorption
El after adsorDtion
Figure 2. Nucleosome-restricted antibodies in kidney eluates of 7
MRL-Mp-lpdlprmice with proteinuria. Kidney eluates were pooled
and tested in antinucleosome and anti-dsDNA ELISAs at the same
Ig concentration, before and after adsorption on dsDNA-cellulose.
See Figure 1 for definitions.
both anti-dsDNA and antinucleosome activity, this
adsorption inhibited 90% of anti-dsDNA reactivity
(OD fell from 0.687 to 0.051 units), while antinucleosome activity was only decreased by 25% (OD fell
from 0.950 to 0.713 units), suggesting that the major
part of the antinucleosome activity was not due to
antibodies with anti-dsDNA activity.
Nucleosome-restricted antibodies in kidney eluates of Iprllpr mice with proteinuria. We next examined
whether antinucleosome activity could be found in
kidney immune deposits. Seven Iprllpr mice with
proteinuria of >lo0 mgldl were selected. All had
strong serum antinucleosome and anti-dsDNA activities. A total of 160 pg of IgG was eluted from their
kidneys. The eluate had no antihistone activity by
ELISA (data not shown), but had both anti-dsDNA
and antinucleosome activity (Figure 2).
Because anti-dsDNA antibodies may have antinucleosome activity, which could thus account for the
antinucleosome activity of the eluate, we further characterized these eluates to determine whether this
antinucleosome activity was due to nucleosomerestricted antibodies and/or to bona fide anti-dsDNA
antibodies with antinucleosome activity. As shown in
Figure 2, anti-dsDNA activity was totally abolished
after adsorption of the eluate on dsDNA-cellulose,
while antinucleosome activity persisted, although at a
lower level than before adsorption.
In the present study, we characterized the
autoimmune response to purified nucleosomes and
compared it to the response against its two components, dsDNA and histones. IgM and IgG antinucleosome responses were found in both strains of lupus
mice (MRL-Mp-lprllpr and MRL-Mp- +/+), but not in
nonautoimmune mice of the same haplotype (C3H).
This antinucleosome response emerged earlier and
faster in Iprllpr mice than in +/+ mice, suggesting that
the same immunologic process occurs in both strains,
and is accelerated in Iprllpr mice by the Ipr gene, as
has been suggested for other autoantibodies (1 1). The
time-course comparison of antinucleosome, antidsDNA, and antihistone responses showed that early
antinucleosome activity developed in 80% of lupus
mouse sera, and was not associated with concomitant
anti-dsDNA or antihistone activity. This suggests that
some autoantibodies can react with purified nucleosomes but with neither of the individual components
(dsDNA or histones). Inhibition studies with nucleosomes, dsDNA, and histones demonstrated that this
early activity was specific for the nucleosome. Accordingly, dsDNA-cellulose adsorption of sera with
early antinucleosome activity did not modify this
activity, indicating that it was not due to anti-dsDNA
antibodies. Taken together, these results strongly suggest that in the serum of lupus mice, there is an
antibody subset which is directed against antigenic
determinants generated by specific interactions among
nucleosome components.
This conclusion is substantiated by the recent
report of a MAb isolated from an lprllpr mouse, which
reacts specifically with nucleosomes but not with
either of its components (12). Similarly, 7 MAb obtained from a +/+ mouse have been shown not to
react significantly with dsDNA or isolated histones,
but to bind strongly to (H2A-H2B)-DNA complexes
(3). Moreover, a recent study of BXSB and Iprllpr
mice indicates that the initial autoimmune response in
these lupus-prone mice is primarily directed against
conformational epitopes on native chromatin (4).
Conversely, several findings suggest that lupus antidsDNA and antihistone antibodies react with nucleosomes and could therefore be considered antinucleosome antibodies. Indeed, in the present study, in
which 45 lupus mouse sera were examined, antidsDNA or antihistone activity was always associated
with antinucleosome activity. Four purified murine
MAb derived from lupus mouse splenocytes (2 anti-
dsDNA and 2 antihistone H2B) and anity-purified
anti-dsDNA antibodies from the sera of 2 SLE patients
also bound to coated nucleosomes to a similar extent
(data not shown).
Taken together, these findings strongly suggest
that the antinucleosome family is a broad antibody
population that includes nucleosome-restricted antibodies (i.e., antinucleosome antibodies without antidsDNA or antihistone activity) and antinucleosome
antibodies with anti-dsDNA or antihistone activity. In
the current work, we did not study the autoimmune
response to single-stranded DNA (ssDNA) and cannot
state whether anti-ssDNA belong to the broad antinucleosome antibody family. However, this appears to
be highly unlikely since Burlingame et al (4) showed
that in several models of lupus-prone mice, antichromatin and anti-ssDNA are clearly separate antibody
To assess the possible involvement of
nucleosome-restricted antibodies in lupus nephritis,
we eluted immunoglobulins from the kidneys of lupus
mice with proteinuria. Pooled kidney eluates of lprllpr
mice exhibited both anti-dsDNA and antinucleosome
activity. Adsorption of this eluate on dsDNAcellulose fully abolished its anti-dsDNA activity, and
decreased its antinucleosome activity by 60%. This
suggests that anti-dsDNA antibodies found in kidney
deposits have antinucleosome activity. However,
since the eluate had no antihistone activity and since
40% of the antinucleosome activity in the eluate persisted after depletion of anti-dsDNA antibodies, it is
likely that at least a fraction of the antinucleosome
activity is due to nucleosome-restricted antibodies.
These results support the notion that the large antinucleosome antibody family, which includes the antidsDNA and nucleosome-restricted subsets, participates in immune deposition in lupus mice with
proteinuria. Interestingly, a recent study demonstrated that nucleosomes perfused into mice can bind
to the glomerulus (13), and thereby could mediate the
renal deposition of antinucleosome antibodies.
Our results and those reported in the literature
(4,12) strongly support the idea that nucleosomes play
a major role in the emergence of antinuclear autoantibodies. In addition, the presence of nucleosome-
restricted antibodies in pooled kidney eluates from
mice with proteinuria suggests that these autoantibodies may be involved in the nephritogenic process. In
conclusion, we believe that the wide antinucleosome
antibody population, not only the anti-dsDNA subset,
should be taken into account in lupus. This may also
shed new light on some studies limited to the antidsDNA subset in which no correlation between antidsDNA antibodies and clinical activity was found (14).
1 . Theofilopoulos AN, Dixon F: Murine models of systemic lupus
erythematosus. Adv Immunol37:269-390, 1985
2. Tillman DM, Nainn-Tsyr J, Hill RJ,Marion TN: Both IgM and
IgG anti-DNA antibodies are the products of clonally selective
B cell stimulation in (NZB X NZW)Fl Mice. J Exp Med
3. Losman MJ. Fasv TM. Novick KE. Monestier M: Monoclonal
auto-antibodies t o subnucleosomes from a MRtlMp- +/t
mouse: oligoclonality of the antibody response and recognition
of a determinant composed of histones H2A, H2B and DNA. J
Immunol 148:1561-1569, 1992
4. Burlingame RW, Rubin RL, Balderas RS, Theofilopoulos AN:
Genesis and evolution of antichromatin autoantibodies in murine lupus implicates T-dependent immunization with self antigen. J Clin Invest 91:1687-1696, 1993
5. Rumore PM, Steinman CR: Endogenous circulating DNA in
systemic lupus erythematosus: occurrences as multimeric complexes bound to histones. J Clin Invest 86:471477, 1990
6. Mohan C, Adam S, Stanik V, Datta SK: Nucleosome: a major
immunogen for pathogenic autoantibody-inducing T cells of
lupus. J Exp Med 177:1367-1381, 1993
7. Jacob L, Tron F: Monoclonal anti-DNA antibody: isotype,
specificity studies. J Immunol 128:895498, 1982
8. Lutter LC: Kinetic analysis of deoxyribonuclease I1 cleavage in
the nucleosome core: evidence for a DNA superhelix. J Mol
Biol 124:391420, 1978
9. Thomas JO, Kornberg RD: The study of histone-histone associations by chemical crosslinking. Methods Cell Biol 18:429440, 1978
10. Jacob L, Lety MA, Moqteiro RC, Jacob F, Bach JF, Louvard
D: Altered cell-surface proteincs), crossreactive with DNA, on
spleen cells of autoimmune lupic mice. R o c Natl Acad Sci
U S A 84:1361-1363, 1987
11. Izui S, Kelley VE, Masqda K, Yoshida H, Roths JB, Murphy
ED: Induction of various autoanfibodies by mutant gene lpr in
several strains of mice. J Immunol 133:227-233, 1984
12. Losman JA, Fasy TM, Npvick KE, Massa M, Monestier M:
Nucleosome-specific antibody from an autoimmune MRLMplprilpr mouse. Arthritis Rheum 36552-560, 1993
13. Coritsidis G, Beers P, Pizzuto A, Rumore P: Glomerular uptake
of nucleosomes in vivo and in vitro: a role for mesangial cells
(abstract). J Am SOCNephrol 32305, 1992
14. Isenberg DA, Shoenfeld Y, Schwartz RS: Multiple serologic
reactions and their relationship to clinical activity in systemic
lupus erythematosus. Arthritis Rheum 27: 132-138, 1984
Без категории
Размер файла
521 Кб
detected, antibodies, antihistone, dsdna, present, serum, kidney, anti, andor, lupus, nucleosome, mrl, lprlpr, mice, restricted, eluates, proteinuric
Пожаловаться на содержимое документа