close

Вход

Забыли?

вход по аккаунту

?

IgG antilymphocyte antibodies in SLE detected by 125I protein A.

код для вставкиСкачать
1261
IgG ANTILYMPHOCYTE ANTIBODIES IN
SLE DETECTED BY 125I PROTEIN A
RALPH C. WILLIAMS, Jr., ARTHUR D. BANKHURST, and JEAN D. MONTARO
lZ5 I-labeled protein A was utilized to detect lymphocyte binding of IgC at 3 7 ° C in 50 sera from patients
with SLE. IgC binding was partially inhibited by preincubation of test cells with immune complexes in 50%
of sera. In lo%, cytophilic Ig was increased by immune
complex preincubation, and in 40% no appreciable
change was recorded. With purified preparations of T
and B cells, binding of S L E IgC at 3 7 ° C was equal for
both cell populations.
Systemic lupus erythematosus (SLE) is often accompanied by the presence of antibodies that show reactivity for various determinants on lymphocytes. Lymphocytotoxins occurring in this disorder have been
repeatedly described and characterized by many different groups (1-7). Cold reacting IgM antilymphocyte
antibodies occurring in SLE have recently been extenFrom the Department of Medicine, Bernalillo County Medical Center, University of New Mexico School of Medicine, Albuquerque. New Mexico.
Supported in part by USPHS Grants AM A I 13824-05 and
TOlAl 00353-04 and by a grant from The Arthritis Foundation.
Ralph C. Williams, Jr., M.D.: Professor and Chairman. Department of Medicine, University of New Mexico School of Medicine:
Arthur D. Bankhurst. M.D.: Assistant Professor of Medicine, University of New Mexico School of Medicine; Jean D. Montaiio, B.S.:
Assistant Scientist. Division of Rheumatology, University of New
Mexico School of Medicine.
Address reprint requests to Ralph C . Williams, Jr., M.D.,
Department of Medicine. University of New Mexico School of Medicine, Albuquerque, New Mexico 87131.
Submitted for publication March 12. 1976: accepted May 5 ,
1976.
Arthritis and Rheumatism, Val. 19, No. 6 (November-December 1976)
sively characterized, both with regard to their specificity
as well as their fluorescent staining properties and cytotoxic activities (8,9).
Other studies have made it clear that IgG antilymphocyte antibodies also present in serum of SLE
patients are capable of blocking mixed lymphocyte culture reactions (10,l l ) . That this activity appeared to be
directed mainly against responder cells was indicated by
the work of Wernet and Kunkel (l2,13).
The present report describes an additional
method for measuring antilymphocyte antibodies in the
serum of patients with SLE by utilizing radiolabeled
I
protein A. In addition, evidence is presented that certain
sera from patients with SLE may contain antibodies
reactive with determinants uncovered or apparently exposed on aged cells or cells stimulated to divide by
various artificial plant mitogens. Antibodies with similar
specificity have previously been described principally in
the sera of patients with infectious mononucleosis
(14,15).
Despite a good deal of recent work directed at
uncovering something about the specificity of antilymphocyte antibodies in SLE, which might provide useful
insight into pathogenesis or etiology of the disorder
itself, clear information in this regard is still not available. A clue to a possible relationship between lymphocytotoxic antibodies in relatives of patients with
SLE and concordant anti-RNA antibody has recently
been presented and may afford a fascinating link between lymphocytotoxic antibody activity and that directed against polynucleotides ( 1 6). The current study
1262
a t t e m p t e d t o a d a p t t h e IgG- reactivity of staphylococcal
protein A (17-19) for use in rapid and quantitative
identification of h u m a n antilymphocyte antibodies.
MATERIALS AND METHODS
Sera Studied
All SLE sera studied were from inpatients and outpatients of the Bernalillo County Medical Center and Albuquerque Veterans Administration Hospital. Diagnoses of SLE
were based on recently proposed ARA criteria (20). Sera were
stored in 0.2-ml aliquots at -70°C for no longer than 2-3
months before testing, and repeated refreezing and thawing
was avoided. Normal control sera were similarly collected
from healthy laboratory personnel, medical students, and
staff.
Detection of Antilymphocyte Antibody
The method to detect antilymphocyte antibody was
designed to take advantage of the reactivity of protein A with
structures on the Fc region of IgG (21) as a marker for
antibody fixed to various test cell sufaces. It was felt that such
antibody would probably have its combining sites occupied,
but might still have the opposite ends of the molecule which
bear the Fc-protein-A-reactive structures still available or accessible. The detection of cell binding of IgG antibody to test
lymphocyte surfaces by IZ5 I protein A has previously been
described by Dorval et al(22). Protein A was isolated from 50
g wet weight of Cowan I strain of S aureus by the method
described by Sjoquist et al (23). The isolated protein A was
I and thereafter used in the fashion of a radiolabeled with
labeled Coombs-type reagent to study the adherence of IgG
molecules to various test cells. Labeled protein A was shown
to retain its precipitating reactivity for IgG. It was recognized
that because Protein A reacts only with IgG-I, IgG-2. and
IgG-4 molecules, no reactivity with IgG-3 would be detected
(18).
The test procedure utilized wells in plastic “u” microtiter plates (Cooke Engineering Co.), which were prewashed
with 5% albumin in Hanks’ balanced salt solution (HBSS) to
prevent nonspecific adherence of cells or test materials. Tenmicroliter samples of sera were added to duplicate wells.
Twenty-five microliters of cells or 2.5 X 105 cells (lymphocytes
isolated by Ficoll-Hypaque and passage over nylon-wool columns) were added to each test well, and plates were incubated
30 minutes at 37°C. One hundred fifty microliters of HBSS
were added to each well and plates were then centrifuged at
600 X g for 5 minutes. Supernates were then decanted and
washing was repeated a second time. T o each well 10 pI of lZ5 I
protein A (40 ng of protein A/ml) were added and plates were
again incubated for 30 minutes at 37”C, followed by washing
three times as described above. Cell pellets were dissolved in
100 jd of NCS tissue solubilizer (Amersham/Searle) and
counted in Aquasol in a scintillation counter.
Controls always included test cells alone, similarly
processed without test serum, as well as other samples incubated with several normal sera. Positive controls included
rabbit anti-human lymphocyte antiserum followed by washing
WILLIAMS ET AL
and lZ5 I protein A . After extensive initial testing and experience with the test procedure, a positive test for lymphocytophilic IgG binding was scored when a test serum gave at least
three times the average binding seen with a panel of 10 normal
sera.
The protein A antilyphocyte antibody test described
above was designed to test cell binding by IgG, since protein A
has been shown to react only with Fc structures of IgG-I, IgG2, and IgG-4 molecules and not with IgM or IgA (18,19,2l).
The same procedure can be utilized to detect specific IgM or
total Ig binding to cell surfaces by using test serum incubation
with cells, followed by specific rabbit anti-human IgM or antiI protein A.
F(ab) antibody, washing, and application of
For the purposes of the present study, assay of only IgG
antilymphocyte antibody was performed in SLE sera.
Attention was also directed at IgG lymphocyte surface
binding after various incubations or mitogen treatment of
lymphocytes. Serum binding to test. lymphocytes was estimated before and after mitogen stimulation with phytohemagglutinin, pokeweed mitogen, and after merely 3-day incubation with medium alone.
Cell Types Tested
Lymphocytes. In most instances lymphocytes were initially utilized after harvesting from Ficoll-Hypaque gradients
and initial depletion of monocytes and macrophages by passage over nylon-wool columns. Such lymphocyte preparations
usually showed less than 2% cell staining as monocytes with
peroxidase stain (24). but they still contained 5-8% B cells or
lymphocytes with surface Ig detectable by direct immunofluorescence using pepsin-digested anti-F(ab’), reagents (25). T o
prepare T-lymphocyte enriched populations for testing, monocyte-depleted lymphocyte preparations were passed over
anti-lg columns by the procedure of Wigzell et a/ (26) to
produce effluents containing largely T cells and less than I %
admixture of lymphocytes bearing surface Ig as determined by
immunofluorescence. Cell populations enriched for B cells
were prepared by centrifugation of T cell sheep erythrocytes
through Ficoll-Hypaque gradients (27). The B-cell layer removed from these gradients generally contained only 2-10%
cells that was capable of forming T-cell rosettes.
In general the reactivity of various test sera with lymphocytes was assayed in parallel fashion using cells incubated
for 1 hour in HBSS, as well as in cells preincubated for 1 hour
in HBSS followed by testing in the presence of autologous
plasma. Also in order to ensure an adequate turnover of cell
surface structures which might obscure reactions between cell
membranes and antibodies in test serum, in some instances
lymphocytes were incubated for 18 hours overnight in a 5%
C0,-air mixture and were used in experiments as target cells
for antibody reactivity after viability testing to ensure the
presence of 95-99% living cells.
Figure 1 represents the presumed mechanism of the
I protein A reaction for detection of antilymphocyte antibody. It can be seen that adsorption of serum immune complexes to Fc or aggregate receptors on test cells could also
conceivably produce apparent increments in I protein A cell
binding. In order to explore this possibility, a series of experiments was carried out using preincubation of test cells with
aggregates of human IgG (isolated by DEAE cellulose ion
IgG ANTILYMPHOCYTE ANTIBODIES IN SLE
Labeled
Protein A
COMPLEXES
A
Fig I . Diagrammatic representation of a lymphocyte showing possible
A ( * ) might react with IgG adsorbed to cell
ways that 1251-labeledprotein
surface. Protein A may react directly with cytophilic lymphocyte-directed IgG, or with free Fc portions of IgG bound in antigen-antibody
complexes to Fc, or possibly to C3 receptors present on B cells.
1263
ing was used as criterion for positive binding, 80% o f the
50 SLE sera tested showed reproducible antilymphocyte
IgG binding in a panel of 10 normal lymphocytes of
varying HL-A phenotypes. The initial lymphocyte panels were monocyte depleted but contained both T and B
cells. In contrast, none of 30 sera from normal donors
and only 2 of 35 sera from miscellaneous hospitalized
control patients showed significant binding with the labeled protein A assay.
Figure 2 depicts a representative experiment indicating the marked lymphocyte-IgG binding as detected
by the lZsI protein A. In SLE sera showing strong
binding by labeled protein A assay, specificity of the test
system f o r IgG binding to test cells was confirmed by
positive reactions employing serum IgG fractions isolated by DEAE chromatography. In addition, pepsin
digestion of SLE IgG completely abolished lZsI protein
A binding. Initially it was felt that such results might be
explained by differences in total serum IgG between
normal and SLE sera. However Mancini radial diffusion
quantitations of serum IgG in LE sera and controls
25.0
exchange chromatography and prepared by heating t o 63°C
for 15 minutes, followed by concentration after ultracentrifugation at 35,000 rpm for 2 hours). Cells preincubated with
aggregates were subsequently reacted with various test sera,
washed, and incubated with labeled lz5 I protein A. In addition, preincubation with soluble BSA-rabbit IgG anti-BSA
antigen-antibody complexes was also carried out in many instances. Amounts of complexes added were determined from a
standard quantitative precipitin curve using bovine serum albumin and rabbit anti-BSA antiserum previously inactivated
at 56°C for 30 minutes.
More details concerning properties of heterologous
antibodies produced against altered or mitogen-stimulated
lymphocytes were further explored by immunizing rabbits
with normal human lymphocytes from a single donor after 3
days of stimulation with mitogenic amounts of PHA o r PWM.
Animals were given three injections every 2 weeks of 10 X 106
PHA-stimulated or PWM-stimulated normal human lymphocytes with equal volumes of complete Freund’s adjuvant
and bleedings that were obtained 7 weeks after the first injection. After inactivation at 56°C for 30 minutes, antisera
were first absorbed with PHA- or PWM-linked to sepharose
by the cyanogen bromide technique and then absorbed using
10-15 X 108 unstimulated cells from the original normal donor
per 0.5 ml of antiserum. Absorbed antisera were then tested by
Iz5 I protein A binding for specificity toward various mitogenstimulated cells.
RESULTS
Fifty sera from patients with active SLE were
tested. When threefold above average normal sera bind-
a
7
7
-
._
..
...._.
.,
2.;.
v:.
I. .:..,
!:: :
!Li
;<.;
7.?<?
: ..
.’.,
....
b
.
LE SERA
NORMAL SERA anti-Lymph
Fig 2 . Binding of IgG globulin to test normal lyntphocytes as detected b j
labeled protein A . Four normal sera show low or relatively negatice
binding, whereas four sera from patients with SLE show considerable
IgG binding. The positive control in this experiment was furnished by a
I :10 dilution of pooled rabbit antilymphocyte antiserum. which gave 25
X lo9 cpm binding. No significant binding of labeled protein A t o cells
alone was recorded.
WILLIAMS ET AL
1264
EU
30
n
20
0
X
EH
0
130
5
10
15
P
Fig 3. Two 1040% sucrose density gradient separations of S L E sera
showing strong IgC binding O / ' ~ Vprotein A to test lymphocytes. The
bottom ofgradients is shown to the lefr, the top at the right. Serum ELI
shows a clear 7s peak of binding activity. However serum EH shows
both 7s and high molecular weight IgG binding components.
showed no clear relationship between serum IgG concentration and antilymphocyte antibody, as detected
with the lZ5 I protein A test.
As Figure 1 illustrates, one reaction that might be
detected by the lZ5 I protein A procedure is binding
between complexes containing IgG in SLE serum and
Fc or aggregate receptors known to be present on some
test lymphocytes (28,29), as well as any small proportion
of monocytes not removed by the initial nylon-wool
column filtration. Eighteen-hour ultracentrifugation of
10 test LE sera on sucrose density gradients (30) and
subsequent testing of 7s and high molecular weight
fractions in most instances showed Iz5 I protein A antilymphocyte activity in 7s fractions. However a few
sera (20%) also showed binding of 19s and higher molecular weight materials to test lymphocytes (Figure 3).
Accordingly a series of experiments was performed by preincubating test lymphocytes with varying
concentrations of aggregated human IgG or soluble immune complexes (BSA-rabbit-anti-BSA), followed by
washing and subsequent incubation with test LE sera.
Test lymphocytes were preincubated with 10, 50, 100,
and in some cases 200 pg of aggregates or soluble complexes.
In general three distinct patterns were seen after
such experiments and they are shown in Figure 3. Some
LE sera showing high binding with cells alone showed
considerably less binding with cells preincubated with
either soluble complexes or aggregates of IgG (LE serum I , Figure 4). N o significant changes were noted in
control experiments utilizing cell preincubation with
BSA, HSA, or Bence Jones protein alone. On the other
hand some SLE sera showed a potentiation of serum
IgG binding after complex or aggregate preincubation
(LE serum 2, Figure 4). This latter pattern did not
appear to be correlated with demonstrable rheumatoid
factor activity as detected by latex fixation or Ripley cell
agglutination testing and occurred in 10% of SLE sera
tested. A third pattern exemplified by LE serum 3 in
Figure 4 showed extremely high binding of antilymphocyte antibody, which appeared to be only partially
affected by aggregate or complex preincubation of test
cells. This pattern was noted in 40% of SLE sera tested.
Experiments utilizing immune complex preincubation of
target cells followed by lupus serum were performed
with undiluted and diluted serum. No clear diminution
35
n
5 t
n
Fig 4. Results of '251protein A binding after preincubation of test
monocyte-depleted lymphocytes with 10. 50. and 100 pg of preformed
IgG aggregates or immune complexes. Test lymphocytes contained approximately 90% T cells and 10% B cells. S L E serum I shows some
degree of inhibition 0 f T protein A bindingfollowing preincubation with
complexes. This pattern was recorded in h a y of the S L E sera tested.
S L E serum 2 shows very low binding with cells alone, but an increase in
binding ajier cell preincubation with complexes. This pattern was recorded in 10% of S L E sera. S L E serum 3, showing a high degree of
binding, was only slightly diminished even by cell preincubation with 100
pg of complexes. This pattern was noted in 40% of S L E sera tesled.
1265
IgG ANTILYMPHOCYTE ANTIBODIES IN SLE
Table 1. Relative Reactivity of Sera from Normal Subjects and
Patients with SLE for T- or B-Cell Enriched Lymphocytes
Using the -1 Protein A Binding Test
LE serum
1
2
3
4
5
6
B-Cell
Enriched Preparation*
(cpm protein A )
T-cell
Enriched Preparationt
(cpm protein A )
I20,OOO
25,000
32,000
33,000
45,000
30,000
123,000
20,000
24.000
22.000
38,000
32,000
2,000
3,000
4,000
2,200
2,500
3,800
Normal serum
I
2
3
~
~~
* B-cell
enriched preparations contained 2-8% E-binding cells, 0%
monocytes. 60-80% cells with surface Ig, and 50-56% cells with EAC
receptors.
t T-cell enriched preparations contained 85-88% E-binding cells, 0%
monocytes, 0% cells with Fc receptors, 0-2% cells with EAC, and
0-I% cells with surface Ig.
or titering out of lymphocyte effect was consistently
noted when undiluted 1 : 5 or 1 : 10 dilutions of lupus
serum were used.
The pattern shown by LE serum 1 was believed
to be compatible with the presence of both complexes
and antilymphocyte antibody in such a serum, because
Iz5 I protein A binding was reproducibly inhibited by
certain concentrations of aggregate or complex preincubated with test cells. I n general about half of 20 SLE
sera tested by aggregate or immune complex preincubation of test cells showed a Iz5 I protein A binding
pattern compatible with the presence of complexes. This
modification of the test system also appeared to be
considerably more sensitive for the detection of complexes in the presence of associated antilymphocyte antibody than did ultracentrifugation on density gradient
separation and testing of individual fractions for cell
binding by lZ5 I protein A , shown in Figure 3.
A series of studies was next performed to determine relative avidity of binding of IgG in SLE sera to T
cells or B cells. As Table I demonstrates, with the labeled protein A method most SLE sera showed approximately equal binding for T-cell enriched preparations as
compared to B-cell enriched target lymphocytes. No
clearcut predominant T cell specificity of LE serum IgG
antibodies was thus apparent. These findings were also
noted even after Fc receptors on B cells had been
blocked by preincubation with IgG aggregates.
Because it was felt that SLE serum IgG binding
to lymphocytes as detected by the labeled protein A test
most probably represented a different phenomenon
from the cold reacting IgM antilymphocyte antibodies
previously described (8,9), both activities were compared in a panel of representative SLE sera. Lymphocytotoxic antibody was measured by incubation at
1 1 "C as previously described (1,2,8,9) and a panel of 25
individual lymphocytes of varying HL-A phenotype.
The result was compared to the average lZ5 I protein A
binding with lymphocytes with 10 individual donors. As
Table 2 shows, no clear correlation between the two
types of activity was recorded. Serum No. 1 gave the
highest binding of lZ5 I protein A as well as showing
broad (90%) cytotoxicity against the 25-donor panel. On
the other hand sera Nos. 7 , 8 , and 9 showed high degrees
of cytotoxicity but very low Iz5 I protein A binding.
Studies Using Normal Lymphocytes Before and
After Mitogen Stimulation
A series of experiments was next conducted to
determine whether sera from patients with SLE contained antibody with specificity for determinants somehow exposed or revealed on lymphocytes that had undergone blast transformation after mitogen stimulation.
Included in these studies as positive controls were several sera from patients with infectious mononucleosis,
shown in previous reports (14,15) to have activity for
dividing cells or in some instances B lymphoblasts.
Table 2 . Comparative Cold-Reacting Lymphocytotoxic A ntibodj
Activity and l Z 5 I Protein A IgG Binding in 9 Sera
from Patients with S L E
Average
Lymphocytotoxicity* (%)
''I Protein A
IgG Bindingt (cpm)
I
90
2
3
4
5
6
7
8
9
10
95.000
9,000
2 I.000
17,500
12.000
36,000
4,000
3.500
4,000
SLE patient
Normal serum
A
B
* Expressed
t
8
25
50
5
96
70
90
5
5
2.000
1.800
as average cold-reacting lymphocytotoxicity when serum
was tested against 25 individual donor lymphocytes.
Expressed as average binding with 10 individual donor lymphocytes.
I266
WILLIAMS ET AL
Antilymphocyte antibody activity before and after 72 hours of PHA, PWM, or 37°C incubation without
stimulation was assayed in a series of sera by means of
the standard lZ6I protein A assay. Results of typical
experiments are shown in Figure 5. I t can be seen that
both LE sera FE and DEB showed significant binding to
the unstimulated cells, as well as to the cells incubated
but not stimulated with mitogen. Cells stimulated with
PWM showed markedly enhanced uptake of IgG from
both LE sera as well as the positive control infectious
mononucleosis serum. However only LE serum FE
showed significant increase in binding with PHA-stimulated cells after 3 days in culture.
Similar experiments using 10 other sera from
patients with SLE showed marked increments in l z 5 I
protein A antilymphocyte IgG activity, particularly
binding to PWM-stimulated cells after 3 days in culture.
Four of the 10 LE sera tested also showed substantial
increments of lZ6 I protein A binding with cells cultured
alone for 3 days without mitogen. Parallel controls indicated 80-100-fold stimulation indices in mitogen-stimulated cultures and no significant stimulation without
mitogen. None of eight normal sera similarly tested
showed significant increment in binding with unstimulated but incubated or PHA- or PWM-stimulated cells.
With these techniques several attempts were
made to look for specificity of IgG antibodies in SLE
sera for cell-surface determinants unique only for dividing cells. Aliquots of test LE sera were preabsorbed with
initial test lymphocytes and then immediately tested
with initial lymphocytes as well as subsequently with
PWM- or PHA-treated cells. These techniques could not
demonstate absolute or clearcut specificity for unique
PHA- or PWM-induced determinants. However more
experiments using this approach, and possibly other
techniques such as immunofluorescence, are needed before the question of antibodies with specificity only for
determinants peculiar for PHA or PWM dividing cells
can finally be answered.
In the animal model, rabbits immunized with
PHA or PWM-stimulated human lymphocytes from a
single donor provided antisera which were then absorbed with insolubilized PHA or PWM, and finally
with normal test unstimulated lymphocytes from the
same normal donor. Absorption employed inactivated
rabbit antiserum diluted 1 : 2 with phosphate-buffered
23.0
13-
CELLS
hd
12II-
lo9m
0
X
20
CELLS NOR LE
U
E W =RUM
SRLU FE
f
CELLS LE
U
E SERIM
FE
CEUS LE
ALONE sRu(
FE
.E
INFO.
ERlM MONO
EB.
Fig 5 . Results 0 f " ~ 1protein A binding of sera from SLE patients FE and DEB, as yell as a serunr from a patient with inJectious
mononucleosis ( I m ) . The panel on rhe leJi shows binding to nortiral unstiniulated monocyte-depleted lyniphocytes. The patrern aJrer
3 days of incubation of cells without nlitogen shows increased '2slproteinA binding for borh L E sera as well as rhe Im serum. bur no
Jignificanr binding b)! normal serutn. The rwo righr-hand panels indicate niarked increase in l Z 5 l prorein A binding to PWMsritnulated cells by both LE sera and I m serum, but onljB one LE serum binding increase uith PHA-stirnulated cells. Eight nornial
sera (not shown) showed no increase in binding when compared to stiniulated cells alone.
IgG ANTILYMPHOCYTE ANTIBODIES IN SLE
saline and 10-15 X lo6lymphocytes per 0.5 ml of diluted
rabbit antiserum. Absorptions were performed at 37°C
for 1 hour and overnight at 4°C for 14 hours. With the
protein A binding method, such absorbed rabbit antisera did not show IgG reactivity for determinants
unique to dividing cells. Also, specificity for antigens on
mitogen-stimulated cells could not be confirmed by absorption experiments using normal unstimulated cells
followed by testing with mitogen-treated cells. These
results were confirmed by means of direct agglutination
and cytotoxicity with normal unstimulated as well as
mitogen-treated donor lymphocytes.
DISCUSSION
The present report documents the presence of
IgG antilymphocyte activity in the serum of many
patients with SLE. Thus it is clear that in addition to
cold-reactive IgM lymphocytophilic and lymphocytotoxic antibodies (8,9), warm-reacting lymphocyte-directed antibodies are also present and are capable of
detection with the sensitive Iz5 I protein A-binding technique. It is interesting that Winfield et a1 (9) also recorded some degree of IgG lymphocyte binding using
different methods. These antibodies appeared to be
more frequently encountered in serum from SLE
patients than the IgG antibodies previously characterized in lupus sera capable of blocking mixed leukocyte
culture (MLC) ( l 0 , l l ) . It is not clear what degree of
overlap exists because no MLC testing was done in
parallel with the present study.
A function or specific raison d’etre for the presence of an apparent wide variety of antilymphocyte
antibodies in sera of patients with SLE is not yet clear.
As an interesting parallel, high titers of IgM thymocytespecific lymphocytotoxins are noted during the course of
NZB disease, the lupus analogue in mice (31). Moreover, in NZB mice these lymphocyte-reactive antibodies
have been demonstrated to affect lymphocyte homing
patterns and distribution (32). At present there is some
initial evidence in lupus patients that antilymphocyte
antibodies may specifically relate to the course or precise
clinical manifestations of disease, as recently indicated
by several reports (5,33,34).
Alternatively, whether antilymphocyte antibodies act in any way like autologous immune regulators or immunostats has not yet been clarified. Their
presence in a wide variety of human disease states as
detected mainly by studies of lymphocytotoxins certainly suggests that they are by no means unique to SLE.
Lymphocytotoxins have now been described in associa-
1267
tion with pernicious anemia (35), multiple sclerosis (36),
Hodgkin’s disease (37), regional ileitis (38), viral infections (39), and chronic renal disease (40). The occurrence of antibodies cytotoxic for lymphocytes in such a
broad distribution of disorders suggests that they are
generated or present as a natural phenomenon occurring
during the immune response associated with heterogeneous inflammatory or immune reactions. Much like
rheumatoid factor or antinuclear antibody, lymphocytotoxins may represent an apparently self-directed immune response capable of generation by multiple stimuli
or clinical conditions.
The presence of antilymphocyte antibody in relatives of patients with SLE has recently also been linked
to the occurrence of anti-RNA antibody (l6,4l). I t is
therefore possible that alterations in lymphocyte membranes or even virally coded membrane antigens are
responsible for producing antibodies that cross-react
with similar determinants in a heterogeneous panel of
normal test lymphocytes. N o evidence for specific lupusrelated antigens has yet been produced, despite repeated
attempts to demonstrate such specificity (6,7,9,41). A
recent report by Lewis et a1 (42) might be interpreted as
evidence for possible viral antigens detectable in peripheral blood lymphoid cells of patients with active
SLE. However further confirmation and extension of
these observations are needed.
Absolute specificity of protein A only for IgG
has recently been questioned because Lind and coworkers have indicated some degree of reactivity for
certain types of IgM (43), as determined by inhibition of
binding of one monoclonal IgM protein to protein A. It
seems unlikely that this type of reaction interfered with
the present assay, because parallel studies using isolated
IgG-free IgM serum fractions from SLE sera containing
antilymphocyte antibodies failed to show significant elevations of binding in this assay system.
The present study appears to support the concept
that the lz5 I protein A binding test may be capable of
detecting y-globulin complexes, and possibly antigenantibody complexes, in the sera of some patients with
SLE. Interpretation of the results of tests involving preincubation of test cells with aggregates or soluble immune complexes followed by addition of LE serum was
complicated by the concurrent adsorption to test cells of
antilymphocyte antibody, so that the method outlined
here is not ideal for detection of complexes. In view of
the current findings with lZ5 1 protein A, it would seem
that the Raji cell method recently introduced by Theofilopoulos et a1 (44,45) may also be complicated by the
co-reaction of lupus antilymphocyte membrane anti-
WILLIAMS ET AL
I268
bodies, as well as by reactions of complexes with the
C3b receptor. Other methods for immune complex estimation, possibly those involving reaction with heterologous cells such as those described by Onyewotu and
coworkers (46), may offer another independent method
for estimation of circulating immune aggregates.
Another feature that complicates precise interpretation of some of the results presented are reports by
Dickler (47) and more recently Tyan (48) that anti-Ia
antisera in mice are capable of blocking Fc receptor
activity. Because it is possible that some SLE sera contain antibodies to cell surface structures similar to Ia
antigens, their reactivity for areas spatially closely related to Fc receptors may modify concurrent lymphocyte adsorption of immune complexes from the
same serum. Current studies are aimed at separating the
effects of complexes and antilymphocyte antibodies in
the same serum by initial ultracentrifugation of serum to
pellet complexes followed by lZ5I protein A testing.
The studies with cells stimulated by pokeweed
and phytohemagglutinin were of considerable interest
and appeared to suggest that certain SLE sera did indeed contain antibodies against structures which are
more prominent on such stimulated cells. With respect
to indications of viral pathogenesis in SLE, it is interesting that there is some preliminary evidence to indicate
that activated T cells appear to be the most suitable
hosts for certain proliferating viral agents (49,50).
Protein A binding to lymphocyte-adsorbed IgG
antibody in SLE serum showed no clear preferential
specificity for either T or B cells. In addition, no clear
correlation was noted between the presence of coldreactive IgM lymphocytotoxins and lZsI protein A binding among individual SLE sera. Antibodies with specificity for PHA- or mitogen-transformed cells are said to
show relative specificity for killer T cells in some systems
(51). This possibility must now be explored in regard to
lupus sera, which show relative increments in binding
with mitogen-transformed cells. If such specificity can
be confirmed in some LE sera, a mechanism potentially
protective against direct T cell-mediated killing of other
cells may be identified.
ACKNOWLEDGMENTS
The authors are indebted to Dr. H. Wigzell for helpful
initial discussions. We are happy to acknowledge the patience
and help of Ms. Bernadette Marquez in preparing the manuscript.
REFERENCES
I . Mittal K K , Rossen RD, Sharp JT, et al: Lymphocytotoxic
antibodies in systemic lupus erythematosus. Nature
(Lond) 225:1255-1256, 1970
2. Terasaki PI, Mottironi VD, Barnett EV: Cytotoxins in
disease: autocytotoxins in lupus. N Engl J Med 282:724728, 1970
3. Stastny P, Ziff M: Direct lysis of lymphocytes by complement in patients with systemic lupus erythematosus.
Arthritis Rheum 14:733-736, 1971
4. Stastney P, Ziff M: Antobodies against cell membrane
constituents in systemic lupus erythematosus and related
diseases. Clin Exp Immunol 8:543-550, 1971
5. Butler WT, Sharp JT, Rossen RD, et al: Relationship of
the clinical course of systemic lupus erythematosus to the
presence of circulating lymphocytotoxic antibodies. Arthritis Rheum 15:231-238, 1972
6. Lies RB, Messner RP, Williams RC Jr: Relative T-cell
specificity of lymphocytotoxins from patients with systemic lupus erythematosus. Arthritis Rheum 16:369-375,
1973
7. Messner RP, Lindstrom FD, Williams RC Jr: Peripheral
blood lymphocyte cell surface markers during the course
of systemic lupus erythematosus. J Clin Invest 52:30463056, 1973 '
8. Winchester RJ, Winfield JB, Siegal F et al: Analysis of
lymphocytes from patients with rheumatoid arthritis and
systemic lupus erythematosus: occurrence of interfering
cold-reactive anti-lymphocyte antibodies. J Clin Invest
54: 1082-1092, 1974
9. Winfield JB, Winchester RJ, Wernent P, et al: Nature of
cold-reactive antibodies to lymphocyte surface determinants in systemic lupus erythematosus. Arthritis Rheum
18:1-8, 1975
10. Williams R C Jr, Lies RB, Messner RP: Inhibition of
mixed leukocyte culture response by serum and y-globulin
fractions from certain patients with connective tissue disorders. Arthritis Rheum I6:597-605, 1973
1 1 . Wernet P, Kunkel HG: Antibodies t o a specific surface
antigen of T-cells in human sera inhibiting mixed leukocyte culture reactions. J Exp Med 138:1021-1026, 1973
12. Wernet P, Kunkel HG: Demonstration of specifiq T lymphocyte membrane antigens associated with antibodies
inhibiting the mixed leukocyte culture in man. Transplant
Proc 5: 1875- I88 1, 1973
13. Wernet P, Thoburn R, Moore A, et al: Blockage of responder and stimulator lymphocyte surface structures and
the shedding phenomenon mediated by human anti-lymphocyte sera in tissue culture. Proceedings of the Eight
Leukocyte Culture Conference. New York, Academic
Press, 1974, pp 543-552
14. Thomas DB, Phillips B: Evidence for membrane antigen(s) specific for human B lymphoblasts. Clin Exp Immunol 14:91-96, 1973
15. Thomas DB, Phillips B: Membrane antigens specific for
human lymphoid cells in the dividing phase. J Exp Med
138~64-70, 1973
IgG ANTILYMPHOCYTE ANTIBODIES IN SLE
16. DeHoratius RJ, Messner RP: Lymphocytotoxic antibodies in family members of patients with systemic lupus
erythematosus. J Clin Invest 55:l 254-1258, 1975
17. Forsgren A, Sjoquist J: ‘Protein A’ from staphylococcus
aureus. I . Pseudo-immune reaction with human y-globulin. J Immunol 972322427, 1966
18. Kronvall G, Williams R C Jr: Differences in anti-protein A
activity among IgG subgroups. J lmmunol 103:828-833,
1969
19. Kronvall G, Quie PG, Williams R C Jr: Quantitation of
staphylococcal protein A. Determination of equilibrium
constant and number of protein A residues on bacteria. J
lrnmunol 194:273-278, 1970
20. Cohen AS, Reynolds WE, Franklin EC, et al: Preliminary
criteria for the classification of systemic lupus erythematosus. Bull Rheum Dis 21:643-648, 1971
21. Kronvall G , Seal US, Finstad J, et al: Phylogenetic insight
into evolution of mammalian Fc fragment of yG globulin
using staphylococcal protein A. J lmmunol 104:140-147,
I970
22. Dorval G , Welsh K I , Wigzell H: Labeled staphylococcal
protein A as an immunological probe in the analysis of
cell surface markers. Scand J Irnmunol 3:405-41 I , 1974
23. Sjoquist J, Meloun B, and Hjelm H: Protein A isolated
from staphylococcus aureus after digestion with lysostaphin. Eur J Biochem 29572-578, 1972
24. Kaplon LS: Simplified myeloperoxidase stain using benzidine dihydrochloride. Blood 26:215-219, 1965
25. Husby G, Strickland G, Caldwell JL, et al: Localization of
T and B cells and alpha fetoprotein in hepatic biopsies
from patients with liver disease. J Clin Invest 56:l 1981209, 1975
26. Wigzell H, Sundquist KG, and Yoshida TO: Separation of
cells according to surface antigens by the use of antibody
coated columns. Fractionation of cells carrying immunoglobulins and blood group antigens. Scand J lmmunol
1175-87, 1972
27. Brown G , Greaves MF: Cell surface markers for human T
and B lymphocytes. Eur J lmmunol 4:302-310, 1974
28. Dickler HB, Kunkel HG: Interaction of aggregated yglobulin with B lymphocytes. J Exp Med 136:191-196,
1972
29. Bentwich Z, Kunkel HG: Specific properties of human B
and T lymphocytes and alterations in disease. Transplant
Rev 16:29-50, 1973
30. Kunkel HG: Macro$obulins and high molecular weight
antibodies, Plasma Proteins. Edited by FW Putman. Vol.
I . New York. Academic Press, 1960, pp 270-305
31. Shirai T, Yoshiki T, Mellors RC: Age-decrease of cells
sensitive to an autoantibody specific for thymocytes and
thymus-dependent lymphocytes in NZB mice. Clin Exp
lmmunol 12:455-465, 1972
32. Shirai T, Yoshiki T, Mellors RC: Effects of natural thymocytotoxic autoantibody of NZB mice and of specifically prepared anti-lymphocyte serum on the tissue distri-
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
1269
bution of T r - l a b e l e d lymphocytes. J lmmunol 110517523, 1973
Winfield JB, Winchester RJ, Kunkel H G : Association of
cold-reactive antilymphocyte antibodies with lyrnphopenia in systemic lupus. Arthritis Rheum 18:S87-594.
1975
Bluestein HG, Zvaifler NJ: Brain-reactive lymphocytotoxic antibodies in the serum of patients with systemic
lupus erythematosus. J Clin Invest 57509-516, 1976
Goldberg LS, Cunningham JE, Terasaki PI: Lymphocytotoxins and pernicious anemia. Blood 392462-865, 1972
Kuwert E, Bertrans J: Leukocyte iso- and auto-antibodies
i n multiple sclerosis (MS) with special regard to complement-dependent cold-reactive autolymphocytotoxins
(CoCoCy). Eur Neurol 7:65-73, 1973
Grifoni V , Del Giacco GS. Tognella S, et al: Lymphocytotoxins in Hodgkins disease. Ital J lmmunol Immunopatho1 l:21-31, 1970
Korsmeyer S, Strickland RG, Wilson ID, et al: Serum
lymphocytotoxic and lymphocytophilic antibody activity
in inflammatory bowel disease. Gastroenterology 67578583, 1974
Mottironi VD, Terasaki PI: Lymphocytotoxins in disease.
I . Infectious mononucleosis, rubella and measles. Histocompatibility Testing. Edited PI Terasaki. Baltimore, Williams & Wilkins, 1970, p 301
Ooi BS, Orlina AR, Masaitis L: Lymphocytotoxins in
primary renal disease. Lancet 2: 1348-1 350, I974
DeHoratius R, Pillarisetty R, Messner R, et al: Antinucleic acid antibodies in systemic lupus erythematosus
patients and their families. Incidence and correlation with
lymphocytotoxic antibodies. J Clin Invest 56:l 149-1 154,
1975
Lewis RM, Tannenberg W, Smith C , et al: C-type viruses
in systemic lupus erythematosus. Nature 252:78-79, 1974
Lind I . Harboe M, Fqlling 1: Protein A reactivity of t w o
distinct groups of human monoclonal IgM. Scand J Immunol 523434348, 1975
Theofilopoulos AN, Wilson CB, Blokisch V A , et al: Binding of soluble immune complexes to human lymphoblastoid cells. 11. Use of Raji cells to detect circulating immune complexes in animal and human sera. J Exp Med
140:1230-1244, 1974
Theofilopoulos AN, Wilson CB, Dixon FJ: The Raji cell
radioimmune assay for detecting immune complexes in
human sera. J Clin Invest 57:169-182, 1976
Onyewotu 11, Holborrow EJ, Johnson G D : Detection and
radioassay of soluble circulating immune complexes using
guinea pig peritoneal exudate cells. Nature 248: 156-159.
1974
Dickler HB, Sachs DH: Evidence of identity or close
association of the Fc receptor of B lymphocytes and alloantigens determined by the Ir region of the H - 2 complex. J Exp Med 140:779-796, 1924.
Tyan ML: Evidence of clustering of H-2K, H-2D, and the
1270
Fc receptor on the membranes of B cells. Proc SOCExp
Biol Med 151:526-531, 1976
49. Bloom NR, Jimenez L, Marcus PI: A plaque assay for
enumerating antigen-sensitive cells in delayed type hypersensitivity. J Exp Med 132:16-30, 1970
50. Jimenez L, Bloom BR, Blume, et al: O n the number and
WILLIAMS ET AL
nature of antigen-sensitive lymphocytes in the blood of
delayed-hypersensitive human donors. J Exp Med
1331740-751, 1970
51. Henney CS: Killer T cells. N Engl J Med 291:1357-1358,
1974 (editorial)
Документ
Категория
Без категории
Просмотров
0
Размер файла
841 Кб
Теги
detected, antibodies, antilymphocyte, protein, igg, sle, 125i
1/--страниц
Пожаловаться на содержимое документа