close

Вход

Забыли?

вход по аккаунту

?

Serum DNA binding activity in healthy subjects and in rheumatic disease.

код для вставкиСкачать
Serum DNA Binding Activity in Healthy Subjects
and in Rheumatic Disease
Peter Hasselbacher and E. Carwile LeRoy
Modifications of the Farr ammonium sulfate precipitation assay for DNA
antibodies have increased sensitivity and led to the detection of DNA
binding in the sera of healthy subjects and subjects with rheumatic disease other than systemic lupus erythematosus (SLE). The presence of
binding in immunochemically pure IgG and the pattern of inhibition by
native and denatured DNA support the thesis that DNA binding of normal,
non-SLE, and SLE sera is antibody and similar in kind although not in
degree. The presence of antibodies to DNA in normal and non-SLE sera
may limit the qualitative role of these antibodies in SLE and render DNA
antibodies similar to antiglobulin and other antinuclear antibodies in
rheumatic disease.
Antibodies to a variety of nuclear components have been consistently detected in
the sera of patients with systemic lupus
erythematosus (SLE) and related rheumatic
syndromes (1). Antibodies selective for
DNA have heretofore been detected primarily in SLE sera (2-7). The present work
describes a modification of the Farr technique for DNA antibody determination
which has increased the sensitivity of the
assay and led to the detection of DNA
binding in sera of normal subjects and of
subjects with rheumatic disease other than
SLE. The characterization of this binding
From the Department of Medicine, Columbia
University College of Physicians and Surgeons, New
York, New York.
E. CARWILE LEROY,
MD: Associate Professor of
Medicine. Columbia University College of Physicians and Surgeons: PETER HASSELBACHER, MD:
Fourth-Year Medical Student, Columbia University
College of Physicians and Surgeons.
Address reprint requests to: Dr. LeRoy, Columbia
University College of Physicians and Surgeons, 630
West 168th Street, New York, New York 10032.
Submitted for publication December 28, 1972; accepted September 10, 1973.
in normal and non-SLE sera as immunoglobulin-dependent and the specificity of
binding for native and denatured DNA are
the basis of this report. The finding of antibodies to DNA in normal and non-SLE
sera suggest that, in the pathogenesis of
SLE, DNA antibodies play a quantitative
rather than a qualitative role.
MATERIALS AND METHODS
The procedure used for the quantitation of DNA
binding was based on the technique developed by
Farr et a1 (8-9). The Farr technique depends on
the observation that DNA complexed with immunoglobulin precipitates in a 50% saturated solution
of (NH,)$O, while free DNA does not. This technique detects antibodies to DNA independent of
immunoglobulin class or functional capability (precipitating, complement-fixing, agglutinating, and
others).
Two modifications of the procedure of Pincus el
a1 (2) were used in the present work: a) 'H-labeled
E coli DNA was used instead of the "C-labeled
human tumor cell DNA used by others, and b) all
sera were heated prior to testing (56"C, 30 minutes)
to destroy complement since C l q has been shown to
bind DNA (10).
ArthrlUI and Rheumatism, Vol. 17, No. 1 (January-Februay 1974)
69
HASSELBACHER and LeROY
Preparation of SH-labeledDNA
E coli B were grown in D-M media (11). A 50-ml
starter culture was incubated overnight in a shaker
bath at 37°C and added to 950 ml prewarmed
media. After 2 hours incubation, 200 gCi of 'Hthymidine were added (final concentration 0.2 gCi/
ml) and incubation was continued for an additional
hour. The cells were harvested and washed in 50
ml of saline-EDTA (0.15 M NaCl, 0.1 M EDTA,
pH 8.0). Washed cells were lysed by resuspending
in 25 ml of saline-EDTA, adding 2 ml of 25%
sodium lauryl sulfate, heating at 60DC for 10 minutes, and cooling slowly to room temperature (12).
DNA was purified, according to the method of
Miura (13), by adding 25 ml of Tris-SDS buffer
(0.1 M Tris. 1% sodium lauryl sulfate, 0.1 M NaCl,
pH 9) to the lysed suspension, followed by the addition of 1 volume of phenol saturated just before
use with Tris-SDS buffer (80% phenol by weight)
at 0°C. The mixture was shaken in a glass-stoppered
bottle in the cold for 20 minutes, followed by lowspeed centrifugation for 10 minutes. The upper
aqueous phase was separated and clarified by centrifugation at 12,000 rpm for 10 minutes. Nucleic
acids, precipitated by the slow addition of 2 volumes of cold ethanol, were collected on a glass rod
and dissolved in 20 ml of saline-citrate (0.15 M
NaCl, 0.015 M sodium citrate). Pancreatic RNAse
(50 pg/ml previously heated at 80°C for 10 minutes to destroy contaminating DNAse) was added
and the solution incubated at 37°C for 30 minutes.
Phenol-buffer extraction of protein and DNA precipitation were repeated as before. The DNA was
washed in increasing concentrations of ethanol
(70%-80%-90%) and dissolved in 10 ml of salinecitrate (stock DNA solution). The OD, of the stock
solution was 1.44. Using the conversion factor of 50
pg DNA per OD unit, the concentration of DNA
was 72 ag/ml, a yield of 0.72 mg. The OD,:OD,
ratio was 1.85. The working activity in Bray's solution was 35,750 cpm/ml or 4950 cpm/pg DNA
counted at 30% efficiency.
DNA Antibody Procedure
The procedure used is a modification of the technique of Pincus et a1 (2). Two-tenths milliliters of
serum is heated at 56OC for 30 minutes and diluted
1:lO by adding 1.8 ml of borate-saline buffer (250
ml of 0.2 M boric acid, 40 ml of 0.2 M sodium
borate, and 5.84 g NaCl per liter; p H 8.0). To a
test tube containing 100 pliters of borate-saline
buffer and 0.1 p g of DNA (dilution of stock solu64
tion with borate-saline buffer, 1:35 v/v) is added
100 pliters of diluted serum; the serum-DNA solution is incubated at 37°C for 1 hour and at 4°C
overnight. Two hundred microliters of 100% saturated (NH,)$04 at 4°C are added, and after 50
minutes the mixture is centrifuged a t 2000 rpm for
30 minutes at 4OC. The top 0.2 ml of supernatant
is transferred to a counting vial and 0.3 ml of
water is added. The precipitate with remaining
supernatant is stirred and transferred to a second
vial. The test tube is washed with 0.3 ml of water
which is added to the second counting vial and
counted in a Packard Tri-Carb scintillation counter.
The results are expressed as percent DNA found in
the precipitate. Ninety-five percent of added label
was routinely recovered in the two vials.
Antigen Inhibition by Unlabeled DNA
For the inhibition studies reported below, varying amounts of unlabeled E coli DNA (prepared
identically to the labeled antigen) were added to
the assay tubes prior to the addition of the test
serum. Up to 1 p g of either native or heat denatured (lOO°C for 10 minutes) inhibiting DNA was
used. The conditions of the assay were otherwise
unchanged including total volume, labeled DNA
concentration, and time periods.
RESULTS
Characteristicsof the Assay
T h e bacterial DNA used in the assay
procedure had an OD,,,/OD,,,
ratio of
1.85, indicating a reasonable degree of
purity. T h e preparation was predominantly
in the native conformation with measurement of maximal thermal hyperchromicity
at 260 mp, indicating 93% nativity. T h e
specific activity of labeled DNA was 7.4
mCi/mg.
T h e control observations for the assay
procedure included: a) the absence of radioactivity in precipitates from DNA controls, and b) the complete elimination of
precipitation of counts after DNAse digestion of labeled DNA. T h e effect of concentration of labeled DNA on the assay was
Arthritis and Rheumatism, Vol. 17, No. 1 (January-February 1974)
SERUM DNA BINDING ACTIVITY
-
o----o
SLE
NORMAL
Fig 1. Effect of the concentration of SH-DNA
on DNA precipitation. SLE and normal sera were
diluted 1 : l O and used in the standard assay procedure with varying concentrations of SH-DNA.
The distinct differences in precipitation at varying concentrations of labeled DNA are apparent.
The arrow indicates the concentration of DNA
used in the usual assay.
normal sera than was seen in previous
studies (2,3,14-16). This difference may be
due to the source of the DNA and/or the
condition of the assay. The higher binding
levels of the present study are similar to
those obtained by Rothfield et al, who also
used bacterial DNA (17).
I n patients with SLE levels of DNA
binding correlated well with signs, symptoms, and serologic parameters (total hemolytic serum complement) of disease activity.
Reductions in binding were often associClinical Studies
The DNA binding of 409 consecutive ated with aggressive therapy with steroids
sera is shown in Figure 2. These data and/or other immunosuppressive agents.
demonstrate a higher level of binding in Also the values in the non-SLE patients
studied using both normal and SLE sera,
as shown in Figure 1. The assay is quite
sensitive to DNA concentration. A representative SLE serum is shown to vary from
70% to 20yo binding by changing only the
labeled DNA concentration; a normal
serum varied from 50% to 20% binding.
The arrow in Figure 1 indicates the concentration of DNA used in the usual assay
conditions.
ArthrlUs and Rheumatism, Vol. 17, No. 1 (January-February 1974)
65
HASSELBACHER and b R O Y
- ......"..
100. . . . . . a
80
c
c
60
.
-
.
a:
40
I
-
20
.
8
-
SLE
NORMAL
NON-SLE
GRAVES'
MEAN
66
18.7
30
45
STANDARD
DEWAT ION
23.8
8.6
10.7
0.0
NUMBER
I6 0
40
182
27
Hg 2. Summary of DNA binding in 409 c ns cutive s ra. The diagnos s of
selected patients in the non-SLE group are given in Table 1. These data are discussed in the text.
were not randomly distributed. Patients
with syndromes characterized bv other immunologic abnormalities had higher DNA
binding levels than normal subjects. Table
1 lists the diagnoses in patients in the nonSLE group with binding levels greater than
66
30%. Of the large number of patients in
the non-SLE group, only 10 had DNA
binding of less than 15%. Of these 10, nine
were under 18 years of age, consistent with
the suggestion that DNA antibody activity
is an acquired characteristic similar to the
Arthritis and Rheumatism, Vol. 17, No. 1 (January-Febrwy 1874)
SERUM DNA BINDING ACTIVITY
3
0.03 M Phosphate, pH = 7 . 3
. 3 M Phosphate ,
pH- 4.0
2
I
0
QD
(u
P
0
I
0
50
FRACTION NUMBER
Rg 3. Isolation of IgG with DNA binding activity from a representative nonSLE serum. Whole serum was placed on columns of DEAE cellulose and eluted
with step-wise gradients as outlined in the figures. The initial protein peaks containing prominent DNA binding activity were free of all immunoglobulins except
IgG in immunodiffusion tests. Similar elution patterns were obtained with SLE
sera, non-SLE sera (Graves’ disease), and normal sera.
cellulose column chromatography. A representative elution pattern is shown in Figure
3. The initial protein peak was found to
contain only IgG by immunoelectrophoresis as well as considerable DNA binding
activity. Similar purifications were obtained
for 2 SLE, 1 Graves’ disease (non-SLE), and
2 normal sera, demonstrating the uniform
association of DNA binding with IgG. All
The Nature of DNA Binding
sera fractionated had DNA binding activity
in Non-SLE Sera
T o characterize the binding of DNA in in the second elution peak. This peak connon-SLE sera, immunoglobulin G (IgG) tained IgG and IgM; DNA binding was not
was purified from selected sera by DEAE further characterized in this second peak.
age-related observations of antinuclear and
antiglobulin antibody activity (18). As a
further example of DNA binding in nonSLE patients, a group of patients with
Graves’ disease (studied in collaboration
with Dr. Sidney Werner) had levels of
binding above the normal range (Figure 2).
Arthritl. and Rheumatism, Vol. 17,
No. 1 (January-Februay 1974)
67
HASSELBACHER and LeROY
Table 1. Non-SLE Patients with
Greater than 30% Binding
Disorder
No. With
>30%
Total
Patients Binding
non-SLE, and SLE sera is accounted for by
the IgG fraction and support the hypothesis that antibodies to DNA are present in
other than SLE sera.
Nephrotic syndrome in heroin
users
Drug induced SLE
Discoid lupus erythematosus
Viral hepatitis
Polymyositis
Rheumatoid arthritis
Diffuse vasculitis; etiology
unknown
Nephrotic syndrome
Chronic glomerulonephritis
Acute glomerulonephritis
Comparative Antigen Specificities
of DNA Binding Sera
Varying amounts of native and denatured unlabeled DNA were added prior to
the assay of binding with native labeled
DNA. Figures 4A, B, and C are a composite
of
antigen inhibition experiments with
12
10
4
7
SLE, non-SLE, and normal sera as shown.
14
7
The three categories of sera (SLE, non6
6
SLE, and normal) showed similar patterns
of antigen specificity; when the labeled
Further evidence that normal and non- DNA was native, native DNA was a better
SLE sera contained IgG-dependent DNA inhibitor than was denatured DNA. The
binding was obtained by absorption with companion experiments using denatured
antisera specific for IgG, IgM, and p l C labeled DNA were confirmatory. RNA in
prior to testing for DNA binding. Absorp- hundred-fold excess had no effect on bindtion with appropriate volumes of anti-IgG ing.
reduced DNA binding to less than 5% in
SLE, non-SLE, and normal sera. AbsorpDISCUSSION
tion with anti-IgM caused a slight and
The present work presents evidence that
variable reduction in binding; absorption
with p l C antisera did not affect binding IgG from normal sera and the sera of other
and served as controls in these experiments. than SLE patients has binding specificity
When DNA binding was studied with vary- for native DNA. It is clear that the sensiing SH-DNA concentrations using either tivity of the Farr ammonium sulfate preserum or isolated IgG, the curves were su- cipitation assay for DNA antibodies is a
perimposable and similar to the data shown function of several variables including pH
in Figure 1. Simultaneous quantitative and ionic strength of the solution; dilution
serum protein electrophoresis was available of the test serum used; the concentration
in 22 of the non-SLE patients. The correla- of DNA present; and the physical state of
tion coefficient calculated from linear re- the nucleic acid antigen. Pincus (19) has
gression analysis of percent binding plotted demonstrated that a decrease in either pH
as a function of serum gammaglobulin con- or ionic strength of the assay system will
centration (range 1.3-3.7 g%), was 0.147. facilitate precipitation of DNA by both
No correlation was observed, suggesting SLE and non-SLE sera. This is not a surthat DNA binding by non-SLE sera is not prising finding, inasmuch as most nona non-specific property of the concentra- covalent protein-ligand interactions are
tion of gammaglobulin. These observations functions of these two variables. The pressuggest that the DNA binding in normal, ent assay was based on that of Pincus (2),
88
Arthritia and Rheumtiam, Vol. 17, No. 1 (January-February 1974)
SERUM DNA BINDING ACTIVITY
-s
Fig 4B Graves’ disease
A
n
r
loo-
E
0
Fig 4A
SLE sera
,
80-
a
Native
L
I
8
CL
a
i
4
I
60-
a
40-
8
m
60
,,’
40
/
20
,
/
0
w
/
,,
0
Denatured
m
I
B
I
R
0
1.0
0.6
0.2
INHIBITING DNA (ug)
w
Fig 4C ~ o m sera
l
B
2k
80-
u
W
0.
CL
sn
60Nutive
m
I
LL
40-
0
9’
z
0,
,t 20-
0,
-__ - - --- - --
c
-
,’
I
Demtured
-
c.l
O
-Q
0
I
I
m
r
I
I
*
A
T
-
:
I
1
1.0
0.6
INHIBITING DNA
(Ug)
Fig 4. Inhibition studies of native SH-DNA
binding by preincubation with unlabeled native
DNA. Characteristic inhibition curves are shown
for A) SLE sera, B) non-SLE sera (Graves’ disease), and c) normal sera. The pattern of inhibition was similar in the three types of sera.
h
-loo-
0,2
T
assay is the physical-chemical nature of the
antigen, a critical variable difficult to control. Using freshly prepared, native bacterial DNA of high molecular weight,
higher levels of binding were observed in
both SLE and non-SLE sera than has been
previously reported. This finding may be a
function of the ease of preparation of undamaged DNA from bacterial sources.
After several months of storage the preparation of DNA used in this study began
to show a decrease in binding, presumably
as a result of DNA denaturation. A second
fresh preparation of E coli DNA yielded
Arthritis and Rheumatism, Vol. 17, No. 1 (January-February 1974)
69
HASSELBACHER and LeROY
DNA preparation and the relationship of
undamaged antigen and high binding.
The precipitation seen in the non-SLE
subjects partially overlaps the values of
SLE patients, all of whom received immunosuppressive therapy. An effect of
therapy on DNA binding has also been
suggested by others (3). The increasing
levels of DNA binding with increasing age
in the control subjects suggests that these
antibodies are an acquired property similar
to the results of age-related studies of antiglobulin and antinuclear antibodies (18).
The ammonium sulfate assay does not
require that the binding protein be immunoglobulin; any protein with the ability
to bind DNA might facilitate precipitation.
Nonimmunoglobulin, DNA-binding proteins have been demonstrated to occur in
several animal species (20); human Clq has
also been noted to bind non-specifically to
DNA (10). Certain hyperimmune rabbit
and goat anti-human sera obtained for the
present study precipitated DNA in ammonium sulfate without added human
serum. Thus it is necessary to show that
isolated immunoglobulin can precipitate
DNA in order to conclude that the serum
contained antibody to DNA. This was done
by the demonstration of DNA binding by
fractions of serum containing only IgG and
by comparing concentration curves of normal serum and isolated IgG, both as a
function of DNA concentration and serum
protein concentration. (No differences were
detected in dilution curves using whole
serum or using isolated IgG, and no correlation was found between binding and
total gammaglobulin concentration in test
sera.)
The IgG in non-SLE sera which binds
native DNA may not be identical to the
anti-DNA antibodies of SLE in all parameters except titer. Indeed it might be predicted that the SLE immunoglobulin has a
70
higher affinity for DNA by virtue of the
selection of more specific clones during antigenic stimulation and response. Thus the
appearance of anti-DNA antibodies may
not be a de novo event, but rather the amplification of preexisting immunoglobulin
present in low titer in non-SLE sera. This
speculation has bearing on considerations
of the etiology of SLE, but does not change
the major clinical and pathogenic importance of anti-DNA antibodies in the diagnosis and management of SLE.
REFERENCES
1. Ritchie RF: The clinical significance of
titered antinuclear antibodies. Arthritis
Rheum 10: 544-552, 1967
2. Pincus T, Schur PH, Rose JA, et al: Measurements of serum DNA-binding activity in
systemic lupus erythematosus. N Engl J Med
281: 701-705,1969
3. Hughes GRV, &hen SA, Christian CL: Anti-DNA activity in systemic lupus erythematosus. Ann Rheum Dis 30: 259-264, 1971
4. Seligmann M: Mise en evidence dans le
serum de malades atteints de lupus erythemateux disdmine d'une substance determinant une reaction de prkcipitation avec
l'acide desoxyribonuclkique. C R Acad Sci
(Paris) 245: 243-245, 1957
5. Robbins WC, Holman HR, Deicher H, et
al: Complement fixation with cell nuclei and
DNA in lupus erythematosus. N Engl J Med
96: 575-579, 1957
6. Stollar D, Levine L, Marmur J: Antibodies
to denatured deoxyribonucleic acid in lupus
erythematosus. Biochim Biophys Acta 61: 718, 1962
7. T a n E,Schur PH, Carr RI, et al: Deoxyribonucleic acid (DNA) and antibodies to DNA
in the serum of patients with systemic lupus
erythematosus. J Clin Invest 45: 1782-1740,
1966
8. Farr RS: A quantitative immunochemical
measure of the primary interaction between
I*BSA and antibody. J Infect Dis 103: 239-
Arthritis and Rheumatism, Vol. 17, No. 1 (January-February 1974)
SERUM DNA BINDING ACTIVITY
262,1958
9. Wold RT, Young FE, Tan EM, et al: DNA
antibody: a method to detect its primary interaction with DNA. Science 161: 806-807,
1968
10. Agnello V, Carr RI, Koffler D, et al: Gel diffusion reactions of Clq with aggregated
globulin, DNA, and other anionic substances. Fed Proc 28: 696, 1969
11. David BD, Mingioli ES: Mutants of Escherichin coli requiring methionine or vitamin
B12.J Bacteriol60: 17-28, 1950
12. Marmur J: A procedure for the isolation of
deoxyribonucleic acid from micro-organisms.
J Molec Biol 3: 208-218, 1961
13. Miura K: Preparation of bacterial DNA by
phenol-pH9, RNAse method, Methods in
Enzymology (Vol XIIa). Edited by L Grossman, K Moldave. New York. Academic Press,
1967, pp 543-545
14. Can- RI, Koffler D. Agnello V, et al: Studies
on DNA antibodies using DNA labeled with
actinomycin-D (3H) or dimethyl (3H) sulphate. Clin Exp Immunol 4: 527-536, 1969
15. Bloomgren SE, Condemi JJ, Vaughn JH:
Procainamide-induced lupus erythematosus.
Am J Med 52: 338-348, 1972
16. Hahn BH, Sharp GC, Irvin WS, et al: Immune responses to hydralazine and nuclear
antigens in hydralazine-induced lupus erythematosus. Ann Intern Med 76: 365-375,
1972
17. Lucian0 AA, Rothfield NF: The relationship between patterns of nuclear fluorescence
and DNA-binding activity. Arthritis Rheum
15: 118, 1972 (abstr)
18. Rowley MJ, Buchanan H, MacKay IR: Reciprocal change with age in antibody to extrinsic and intrinsic antigens. Lancet 2: 2426, 1968
19. Pincus T: Immunochemical conditions affecting the measurement of DNA antibodies
using ammonium sulfate precipitation. Arthritis Rheum 14: 623-630,1971
20. Thoburn R, Hurvitz AI, Kunkel HG: Characterization of a nonantibody DNA reactive
protein in the serum of several mammalian
species. Arthritis Rheum 15: 129, 1972 (abstr)
,,-
Arthritb and Rheumatlrm, Vol. 17, No. 1 (January-February 1974)
71
Документ
Категория
Без категории
Просмотров
0
Размер файла
466 Кб
Теги
health, rheumatic, dna, serum, activity, disease, binding, subjects
1/--страниц
Пожаловаться на содержимое документа