Self-nonself recognition by T and B lymphocytes and their roles in autoimmune phenomena.

1044
SELF-NONSELF RECOGNITION BY T AND B
LYMPHOCYTES AND THEIR ROLES IN
AUTOIMMUNE PHENOMENA
WILLIAM 0. WEIGLE
Evidence is presented which supports the suggestion that whether a given antigen is vulnerable to an autoimmune attack is dependent upon the specific immune
status of B cells and T cells to that antigen which, in
turn, is dictated by the concentration of self antigen in
their microenvironment. That B cells require much
higher concentrations of self antigens than do T cells
for the maintenance of tolerance is supported by data
presented using an experimental model of acquired tolerance to serum proteins. Depending on the immune
status of T cells and B cells to self antigen, the following
three models are suggested for the early events leading
to autoimmunity: 1) polyclonal activation of competent
B cells, 2) direct activation of competent T cells, and 3)
bypass of specifically tolerant T cells and activation of
competent B cells. The role of a regulatory network involving the suppressor cell circuit in the induction and
regulation of autoimmunity is discussed.
The relationship between experimentally induced tolerance to foreign antigens and naturally acquired tolerance to self has both practical and theoretical implications. In this original hypothesis, Burnet (1)
assumed that tolerance induced to foreign antigens is
the same as tolerance to one’s own body constituents.
For an animal to make an immune response to foreign
From the Department of Immunopathology, Scripps Clinic
and Research Foundation, La Jolla, California 92037.
This is publication no. 2396 from the Department of Immunopathology, Scripps Clinic and Research Foundation, La Jolla, California. Supported in part by the United States Public Health Service
Grant AI07007, American Cancer Society Grant IM-42G and Biomedical Research Support Program Grant RRO-55 14,
Address reprint requests to Dr. William Weigle, Department
of Immunopathology, Scripps Research and Clinic Foundation, La
Jolla. CA 92307.
Arthritis and Rheumatism, Vol. 24, No. 8 (August 1981)
substances such as bacteria, viruses, tumor antigens,
and yet not respond to its own body constituents, the
immune mechanism must discriminate between self and
foreign antigens. Thus, during prenatal and/or neonatal
life, before the immune mechanisms mature, animals
develop a state of immunologic unresponsiveness to
their own body constitutents, but this state does not interfere with their ability to respond as adults to foreign
antigens. There is overwhelming evidence that the development of tolerance to self components is not genetically determined, but rather the result of direct contact
between self components and specific antigen-reactive
cells. In this regard, Triplett (2) removed the hypophysis
(buccal component of the pituitary gland) from a tree
frog during early life (tadpole), allowed the gland to differentiate away from its donor, and returned it to the
mature donor whose immune system subsequently rejected the transplanted gland. Similarly, animals make
an immune response to body constituents that they lack
as a result of a genetic deficiency (3-5).
This presentation will be based, in part, on the
assumption that the immune status of T and B cells to a
self antigen is governed by the level of an antigen in
their microenvironment and that the dose of self antigen
required to maintain tolerance in B cells is considerably
greater than that required for maintenance of tolerance
in T cells. Data on which these assumptions are based
will be presented, employing a model of acquired immunologic tolerance to protein antigens, which appears
to mimic the cellular mechanisms of tolerance to self.
Evidence will be presented to support the suggestion
that tolerance to self is, in part, the result of 1) a central
unresponsiveness resulting from deletion of specific selfreactive cells, and 2) to a yet undetermined extent, the
normal immunoregulatory network of the host. In the
1045
SELF-NONSELF RECOGNITION
Table 1. Temporal patterns of immunologic unresponsiveness to
HGG in T and B cells of A/J mice'
nogenic forms of the antigen or after temporary suppression of the immune system.
Days of
Site
Induction
Maintenance
T cells
B cells
Whole animal
1
3
1
150
50
150
* Injected with 2.5 mg DHGG on day 0.
case of both acquired tolerance to foreign antigens and
tolerance to self antigens, the unresponsive state is antigen-directed, in that it results only from prior contact
with specific antigen.
Mechanisms of tolerance
Although immunologic tolerance induced by
prior exposure to antigen is defined by the inability of
the host to respond to that antigen, the cellular and subcellular events leading to the unresponsive state may
differ. In general, acquired immunologic tolerance can
be classified into two categories: peripheral inhibition
and central unresponsiveness (6). In peripheral inhibition, cells competent in respect to their immune capacity to a given antigen are present but their function is
blocked. Lymphocytes of the tolerant animal can bind
the antigen in question and the tolerant state disappears
when the cells are transferred to an irradiated host. Furthermore, at times the tolerant state is associated with a
transient appearance of either circulating or cell-associated antibody. Peripheral inhibition may not represent
a true tolerant state but suppression induced by regulatory mechanisms normally at play in controlling the immune response such as suppressor cells, antibody (including anti-idiotype) suppression, or antigen blockade.
In contrast to peripheral inhibition, central immunologic unresponsiveness is characterized by an immune state where the host is incapable of specificially
reacting with the tolerated antigen (6). Specific antigenbinding cells are not detectable and antibody-producing
cells do not appear even transiently. The subcellular
and cellular events involved in this type of tolerance are
probably identical to those at play in tolerance to self.
Suppressor cell activity may be concomitant, but not responsible for central unresponsiveness. Antigen blockade is not involved, and lymphocytes transferred from
the tolerant donor to the irradiated host remain unresponsive. Although central unresponsiveness is more
readily induced in newborn than in mature animals, it
can be induced in adult animals with either nonimmu-
An experimental model of central
unresponsiveness
The models that best represent tolerance to self
are those in which tolerance is induced in neonates injected with heterologous serum proteins and adults injected with deaggregated IgG (7). Heterologous preparations of IgG apparently owe their antigenicity to the
presence of small amounts of aggregated material. Dresser (8) was first to demonstrate that heterologous IgG
deaggregated by ultracentrifugation not only loses its
ability to induce an immune response in adult mice, but
also induces an unresponsive state to subsequent injections of immunogenic preparations (aggregated) of the
same IgG. In contrast to the monomeric deaggregated
preparations, IgG aggregated by heat (63"C for 25 minutes) is a good immunogen in most species. Human
gamma globulin (HGG) is often used to induce central
unresponsiveness, and like many other heterologous
and homologous serum proteins, rapidly equilibrates
between intra- and extravascular fluid spaces and persists in the circulation until it is slowly eliminated by
normal catabolic processes of the host. Unlike particulate antigens and most hapten conjugates, HGG readily
comes in contact with all antigen-reactive cells for a
prolonged period of time.
A single injection of 2.5 mg of deaggregated
HGG (DHGG) readily induces a complete and lasting
unresponsive state in adult A/J mice as evidenced by
the failure of these mice to respond to a subsequent injection of aggregated HGG (AHGG). As expected, antigen-binding cells disappear from the spleen shortly after
injection of DHGG, and during the period of tolerance
induction and the subsequent time that tolerance is
maintained, no antibody-producing cells (IgM or IgG)
are detectable. In this system, tolerance is not maintained by suppressor cells nor by antigen blockade, and
tolerance is not lost in cells transferred to irradiated,
syngeneic recipients. Both the T and B lymphocytes become tolerant (9), although the duration of tolerance
differs in these two cell types (Table 1). Induction of tolerance in peripheral T cells is rapid and parallels the kinetics of induction observed in intact mice (10). Peripheral B cells are only slightly slower to assume the
tolerant state. More important to self tolerance is the
marked difference in the kinetics of the spontaneous termination of the tolerant state in peripheral B and T
cells. Peripheral T cells, like the intact mice, remain tol-
WEIGLE
1046
Acetylcholine
1
)
Mechanism
I
I
of
Autoreactivity
I
I
I
I
Polyclonal Activation of B Cells
I
I
I
Specific B Cell Activation
(T Cell Bypass1
I
I
I
I
I
I
Specific T Cell (and B Cell) Activation
Figure 1. Relationship among levels ofself proteins in body fluids, immune status of T and B cells to self protein, and mechanism of autoreactivity.
erant for approximately 150 days after injection of
DHGG, although peripheral B cells return to complete
competency after approximately 50 days.
Another situation in which tolerant T cells coexist with competent B cells can be established with low
doses of DHGG. The dose of DHGG required to induce tolerance in adult T cells is 100-1,000 times less
than that required to induce tolerance in B cells (10,ll).
Similarly, doses required to induce tolerance to bovine
serum albumin (BSA) are considerably less in T cells
than in B cells (1 1,12). Thus, when central unresponsiveness is induced with small doses of antigen, B cells
remain competent, while T cells become tolerant. Similar dose response effects most likely apply to self antigens; antigens present in low concentration in the body
fluids would be expected to induce tolerance only in T
cells, and antigens in high concentrations should induce
tolerance in both T and B cells. When T cells are tolerant and B cells are competent, termination of the tolerant state can result by bypassing either the specificity or
the need for T cells, and the B cells activated by self-determinants would produce autoantibody, which may or
may not be accompanied by disease. It has been well
documented that acquired tolerance induced in experimental animals can be readily terminated, when tolerance is only at the T cell level, by immunizing with
cross-reacting antigens or by injection of B cell activators (reviewed in ref 7).
Mechanisms of autoreactivity
Although autoimmune reactivity may involve
abnormalities of any phase of the complex regulatory
system at play in the control of the immune response,
events instrumental in initiating autoimmunity are
probably dictated by both the manner in which self antigen is presented to the immune system and the immune status of T and B lymphocytes in regard to that
self antigen (13). In developing a cellular model of self
tolerance, one can safely assume, first, that self tolerance results from a central unresponsive state rather
than from peripheral inhibition; second, that concentrations of self antigens required to induce tolerance in T
and B cells differ markedly and, third, that self toler-
1047
SELF-NONSELF RECOGNITION
ance is dependent on the concentration of the self antigen in the microenvironment of the potential self reactive cells. Thus, the immune status of T and B
lymphocytes to the self antigen(s) in question may dictate the immunologic pathway of a particular autoimmune response.
A high degree of tolerance to self antigens such
as serum albumin may be present in both T and B lymphocytes, although with other antigens (certain classes
of Ig, growth hormone, and thyroglobulin) a hlgh level
of tolerance may exist in the T cells while B cells are
competent. For still other antigens (basic protein of
myelin, acetylcholine receptor, cytochrome C, and idiotypic Ig determinants) both T and B cells may be competent (Figure 1). In the case of tolerant T cells and
competent B cells, the B cells can be triggered by procedures that bypass either the need for, or specificity of, T
cells. When neither cell type is tolerant, both T and B
cells can be activated specifically, when self antigen is
presented in an effective manner. Competent B cells, of
course, are always susceptible to activation by polyclonal B cell activators.
B cells with receptors with reactivities ranging
from low to high affinity for a given self antigen escape
tolerance induction and are competent for that antigen
when it is present in low concentrations in the body
fluid. On the other hand, when a self antigen is present
in high concentration, the tolerant state is maintained in
both T and B cells having receptors of high to moderate
affinity for the antigen. However, even with these antigens, the concentration may still be too low to maintain
tolerance in those B cells with antigen-reactive receptors
of low affinity, since tolerance affects only B cells with
the higher affinity for antigen (14). On the other hand,
the affinity of these B cell receptors for self antigen is often inadequate to trigger differentiation and antibody
synthesis, albeit, polyclonal (B cell) activators could
trigger such B cells to produce low affinity antibody.
The ability of isolated microbial products as well as microbial, parasitic, and viral infections to cause polyclonal activation in vivo is well documented, and often
autoantibody has been detected as well (reviewed in ref
13).
Polyclonal activation of competent B cells possessing low affinity receptors to self antigen is a common in vitro observation. How often or if such polyclonal activation occurs in vivo is unclear, but it may be
in part responsible for low levels of autoreactive antibody found in the sera of normal individuals. Even if
this is antibody to antigen responsible for vital biologic
functions and the antibody is in sufficient concentra-
tions, it would rarely have a significant effect because of
its low avidity and low levels of reactivity. The possible
exception may be rheumatoid factor in patients with
rheumatoid arthritis. The Epstein-Barr (EB) virus has
been associated with rheumatoid arthritis in that it is instrumental in producing nuclear antigen reactive with
antinuclear antibody (15). For this reason, it was suggested that this virus may be an etiologic agent for this
disease. Epstein-Barr virus has been shown to be a potent polyclonal activator of human B cells in vitro, resulting in production of IgM antibodies to HGG (1618). Although the avidity of the antibody is low, once
the HGG-anti-HGG (IgM) complexes form, they are
biologically active and are capable of activating the
complement pathway. In all likelihood, low af€inity
antibody is produced to many other self antigens as a
result of polyclonal activation, but an autoimmune disease does not ensue because of the low affinity of the
antibody or the target antigen is either sequestered or
does not play a critical function which is readily interfered with by antibody. Furthermore, most polyclonal
activation in vivo may be transient and disappear with
elimination of the polyclonal activator before clinically
detectable tissue damage occurs.
Yet with other antigens present in extremely low
concentrations in the body fluids, a complete tolerance
is not maintained in either the T or B cells. With such
antigens, effective exposure may activate both T cells
and B cells, resulting in a typical T cell dependent antibody response. In addition to helper T cells, cytotoxic T
cells, suppressor T cells, and others may also be activated. Autoimmunity to these antigens, however, does
not usually develop because the concentration is too
low. Even when sequestered self antigens are released in
high concentrations into the microenvironment of selfreactive lymphocytes as a result of infection or other
trauma, the response is transient and probably disappears before clinical symptoms are induced. It is only
when antigen persists in an immunogenic form, e.g.,
when incorporated into complete Freund’s adjuvant
(CFA), that a progressive autoimmune response accompanied by disease is observed experimentally. A possible example of a disease following this pathway is experimental allergic encephalomyelitis (19).
Experimental autoimmune phenomena
Experimental allergic encephalomyeUtis (EAE).
Experimental allergic encephalomyelitis is a disease of
the central nervous system (CNS) induced by immuni-
WEIGLE
1048
Table 2. Antigen-binding cells in normal T and B cells to
thyroglobulin (Tg) and basic protein (BP) of myelin
Cell type
Antigen
Antigenbinding cells
Rat
T cell
B cell
BP
BP
+
+
Mouse
T cell
B cell
Species
-
+
Table 4. Transfer of EAE in rats* after suicide of T or B cells by
treatment with 1251-labeledsyngeneic basic protein (BP)
Cell transferred
B Cells
T Cells
Normal
Treated
Normal
Normal
Normal
Treated
Antibody
BP
BRBCt
EAE
+
-
+
+
+
+
+
-
-
-
* Immunized with syngeneic BP in complete Freund’s adjuvant.
t BRBC burro red blood cells.
Reprinted in part with permission (21).
zation with CNS tissue, basis protein (BP) of myelin or
either natural or synthetic polypeptides of basic protein
(reviewed in 19). In the rat, antigen-BP binding cells
are present in both the T cell and B cell compartments
(Table 2), suggesting that competent T and B cells exist
for basic protein. Furthermore, this disease has been
considered to be the result of cell-mediated immunity.
In an approach to evaluating effector and helper
T cells in the induction of this disease, the ability of sensitized lymphocytes to induce EAE when transferred to
irradiated recipients was assessed after removal of T
cells. Lymphocytes were removed from BP-sensitized
Lewis rats 9 days after immunization with BP in complete Freund’s adjuvant, at a time when T cells were no
longer required to sustain antibody production, and
then used to reconstitute syngeneic, irradiated rats.
These recipients, without any further stimulation with
basic protein, developed antibody, clinical symptoms,
and histologic lesions of EAE (20). However, prior
treatment of transferred cells with anti-thymocyte sera
plus complement circumvented all symptoms and lesions in the recipients, but had no effect on antibody
production (Table 3). Thus, although these recipients
had levels of antibody to BP equivalent to that of recipi-
Table 3. Effect of T cells on induction of experimental allergic
encephalomyelitis (EAE) in thymectomized, irradiated rats
reconstituted with primed cells*
Treatment of
transferred
cells
None
ATS + C
-
EAE
Clinically
Histologically
Serum
antibodyt
71 1o$
0/10
10/10
0/10
2.3
2.3
-_
Lewis rats were thymectomized, irradiated (900 rads), and reconstituted with 250 X lo6 lymph node and 350 X lo6 spleen cells from rats
previously sensitized 9 days before with BP-CFA. The transferred
cells were either untreated or treated with antilymphocyte serum
(ATC) + complement (C).
t Values represent the pg of BP bound/ml of serum. Mean of animals
tested.
$ Fraction of animals positive.
Reprinted in part with permission (20).
ents that received untreated lymphocytes, encephalomyelitis was not induced.
Deleting specific T cells, but not specific B cells
with basic protein heavily labeled with 1251before reconstituting thymectomized, lethally irradiated rats interferes with the induction of disease in these recipients
(2 1). Others have established previously that incubation
of heavily labeled lZ5I antigen with lymphocytes eliminates specific immunocompetent cells because of local
irradiation (22,23). Thymectomized, irradiated rats were
readily reconstituted with a mixture of purified thymus
and bone marrow cells from normal rats in that the recipients, when injected with basic protein in complete
Freund’s adjuvant, developed circulating autoantibodies, histologic lesions, and clinical symptoms of encephalomyelitis. On the other hand, neither symptoms,
lesions, nor antibody to basic protein resulted when the
reconstituting thymus cells were treated with BP heavily
labeled with Iz5I prior to transfer. This pretreatment
with lZ51-BPeliminated specific T cells and apparently
abrogated an essential factor for cell-mediated immunity. Treatment of the thymus cells with 1251-BP
also inhibited the formation of antibodies to BP, but not to
burro red blood cells, demonstrating that specific helper
T cells were deleted. On the other hand, when bone
marrow cells were treated with heavily Iz5I-labeledBP
and injected into thymectomized, irradiated recipients
along with normal thymus cells, and the recipients challenged with BP in complete Freund’s adjuvant, both
clinical symptoms of encephalomyelitis and histologic
lesions were similar to those in rats that received both
normal thymus cells and normal bone marrow cells, although the antibody formation was inhibited (Table 4).
The cellular events in experimental allergic encephalomyelitis are depicted in Figure 2.
Another example of an autoimmune disease in
which both competent T cells and competent B cells are
present to a self antigen is myasthenia gravis (reviewed
in 24). The antigen involved in this disease is the acetylcholine receptor which is present in the animals in ex-
SELF-NONSELF RECOGNITION
1049
Rat
Figure 2. Cellular events after injection of rats with syngeneic basic protein (BP) of myelin. Reprinted with permission (32).
tremely small amounts (approximately 8pg/rat). Tolerance to t h s selfantigen obviously exists in neither the T
nor the B cell. Although, as in experimental allergic encephalomyelitis, both T and B cells are activated in the
experimental model of myasthenia gravis in rats (25), it
is antibody and not cell-mediated immunity that is responsible for the disease.
Thyroiditis. An example of the bypass of tolerant
T cells in animals tolerant to self antigens in helper T
cells but not in B cells is seen with models of experimental autoimmune thyroiditis (EAT) induced with soluble
preparations of cross-reactive thyroglobulin (Tg). It appears that most animals enjoy a high degree of tolerance
to syngeneic Tg in their T cells but not in their B cells
(Figure 1). Although EAT has been more commonly
produced by immunizing animals with homologous Tg
in complete Freund’s adjuvant (26), it is readily induced
by immunization with aqueous preparations of either
chemically altered homologous Tg (27) or heterologous
Tg (28). In rabbits, this model appears to be mediated
by antibody, and the thyroid lesions are associated with
antibody-producing cells to rabbit Tg in the thyroid
gland (29).
Experimental autoimmune thyroiditis can also
be produced in mice by immunizing with aqueous prepTable 5. Transfer of thyroiditis in mice. after suicide of T or B cells
by treatment with 1251-labcledmouse thyroglobulin (Tg)
Cells transferred
Antibody
B Cells
TCells
Thyroiditis
Tg
Normal
Treated
Normal
Normal
Normal
Treated
+++
+
+++
+++
+
+++
* Immunized with aqueous heterologous Tg.
t SRBC = sheep red blood cells.
Reprinted with permission (32).
~
SRBCt
~
+++
+++
+++
~~
arations of heterologous Tg (30,3 1). When thymectomized, irradiated (900 rads) A/J mice are reconstituted
with syngeneic spleen cells or a combination of T cells
and B cells, but not B cells alone, agd immunized with
aqueous preparations of heterologous Tg, they produced antibody to both heterologous and mouse Tg and
developed thyroiditis. As in the “suicide” experiments
with EAE in rats, the selective elimination of specific T
and B cells with heavily labeled “’I-Tg further defined
the roles of T and B cells in EAT. Deleting specific B
cells, but not specific T cells, before reconstituting thymectomized, lethally irradiated mice interfered with the
induction of experimental autoimmune thyroiditis.
It is possible to inhibit both autoantibody production and development of lesions by preincubating
syngeneic Tg heavily labeled with “’1 with bone marrow cells but not with T cells (Table 5). The prevention
of both experimental autoimmune thyroiditis and production of autoantibody to syngeneic thyroglobulin by
the above approach demonstrates that the B cells or B
cell product (antibody) is involved in the induction of
this murine model of EAT (32). In both mice and rabbits immunization with heterologous Tg apparently bypasses T cell specificity, and the T cells activated by the
determinants specific for the heterologous Tg supply the
second signal needed for the differentiation of competent B cells that have reacted with self-related determinants of the heterologous Tg (Figure 3). That
heavily radiolabeled Tg eliminates only specific B cell
activity is compatible with the findings of antigen-binding B cells, but not T cells, for autologous Tg (Table 2).
In contrast to experimental autoimmune thyroiditis induced in animals by immunization with aqueous
preparations of cross-reactive thyroglobulin, the induction of this disease in mice immunized with homologous
thyroglobulin in complete Freund’s adjuvant appears to
WEIGLE
1050
Figure 3. Cellular events after injection of rabbits with altered homologous or heterologous thyroglobulin. Reprinted with permission (32).
involve specific T cell reactivity. This possibility has
been made apparent by Rose and coworkers (33) who
reported an H-2 linked genetic restriction at the T cell
level governing susceptibility to EAT in various mouse
strains. The extent of genetic regulation appears to be
dependent on the participation of genes at both the K
end (IR) and the D end. It appears possible in this
model that the involvement of T cells is the result of expansion of a “leaky” tolerance to thyroglobulin in T
cells by immunization with Tg in complete Freund’s adjuvant.
Other T cell tolerant models. Other experimental
models of autoimmunity have been reported where tolerance to a self antigen is present in the T cells, but not
in the B cells. Protein F is found in the liver of all mammalian species (34). Mice are polymorphic for protein F
in that they have one of two alleleic types, I or 11, which
have both seriologic similarities and differences. Although the mice respond only to the opposite type, they
produce antibody that reacts with both the immunizing
type and the host type (Table 6). Thus, it appears that
mice have tolerant T cells to their own protein F, but
contain B cells that are competent for their own protein
Table 6. Immune response to mouse F protein in mice possessing
different F protein serotypes
Antibody response
Strain I
Typeinjected
I
I1
Strain I1
Type1
TypeII
Type1
-
-
+
+
+
TypeII
+
-
~~
Reprinted with permission (32).
F (34). It was suggested that protein F consists of two
distinct antigenic moieties: a carrier region that constitutes the allogeneic part of the molecule and the antigenic determinant common to both type I and type I1
protein (35). Thus, induction of autoantibodies to protein F involves collaboration between the helper T cells
recognizing the allogeneic region of the molecule, acting
as a “carrier” determinant, and B cells with specificity
for the syngeneic determinant common to both types of
protein F.
In another model, autoantibodies to homologous
a-fetoprotein have been induced after injection of either
monkeys with heterologous a-fetoprotein (36) or rabbits
with altered homologous a-fetoprotein (37). Similarly,
rabbits immunized with a rabbit lactic dehydrogenase
do not make an antibody response, but when immunized with cross-reactive pig lactic dehydrogenase they
produce antibody that is also reactive with rabbit lactic
dehydrogenase (38). As with protein F and a-fetoprotein, these autoantibodies are not associated with any
disease process.
Suppressor cell circuit
The regulatory mechanisms involved in the control of normal immune responses to foreign antigens obviously also play a role in controlling autoimmune responses. Subsets of T cells comprise a network of
regulatory cells that control the normal immune responses to foreign antigens once they are initiated
(39,40). Such a regulatory network similarly monitors
the various parameters of autoimmunity and is instrumental in the clinical progression of autoimmune disease. However, it is not clear what role, if any, inducer,
SELF-NONSELF RECOGNITION
suppressor, contrasuppressor, and feedback inhibitory T
cells play in initiating autoimmunity. Evidence regarding this regulatory network and its possible involvement
in generating autoimmune diseases has been provided
by Cantor and Gershon (41), who examined this network in strains of mice predisposed to autoimmune disease.
Suppressor cell activity has also been linked to
autoimmune disease in human beings (42). If suppressor cells act as a deterrent in initiating autoimmune disease, they probably are effective only with antigens of
limited concentration in the body fluid and not in a
state of solid tolerance at the T cell level. Such antigens
may constantly generate subclinical responses that are
accompanied by suppressor cell activity. The failure to
generate suppressor cells because of abnormalities in
the regulatory network could lead to unchecked immune responses to these self antigens.
0ther regulating factors
The previous discussion postulates a number of
possibilities for the generation of autoimmunity. Thus,
it may be surprising that animals are not constantly
plagued with a battery of autoimmune disease. However, one must remember that autoantibodies or autoreactive T cells are not always the limiting factor in autoimmune disease. In addition to autoreactivity, a
progressive autoimmune disease depends on the persistence of autostimulation, avidity, and biologic activity of antibody (or reactive cells), and the logistics and
function of the self antigen. In the case of putative regulation of the immune response by antiidiotypic antibody, autoimmune reactivity may even be beneficial.
ACKNOWLEDGMENT
The author wishes to thank Janet Kuhns for secretarial expertise.
REFERENCES
Burnet F: The Clonal Selection Theory of Acquired Immunity. Cambridge, England, Cambridge University
Press, 1959, pp 126-133
Triplett EL: On the mechanism of immunologic self recognition. J Immunol 89:505-510, 1962
Erickson RP, Tachibana DL, Herzenberg LA, Rosenberg
LT: A single gene controlling hemolytic complement and
serum antigen in the mouse. J Immunol92:611-615, 1964
Cinader B, Dubiski S, Wardlaw AC: Distribution, inheritance, and properties of an antigen, MpBl, and its rela-
105 1
tion to hemolytic complement. J Exp Med 120397-924,
1964
5. Winchester G: Acquired self tolerance to the intracellular
liver protein F. J Supramol Struct (Suppl 3):274, 1979
6. Weigle WO, Chiller JM, Louis JA: Tolerance: central unresponsiveness or peripheral inhibition, Progress in Immunology. Vol. No. 3. Edited by L Brent, J Holborow.
Amsterdam, North-Holland Publishing Co, 1974, pp 187196
7. Weigle WO: Immunologic unresponsiveness. Adv Immuno1 16:61-122, 1973
8. Dresser DW: Specific inhibition of antibody production.
11. Paralysis induced in adult mice by small quantities of
protein antigens. Immunology 5:378-388, 1962
9. Chiller JM, Habicht GS, Weigle WO: Cellular sites of immunologic unresponsiveness. Proc Natl Acad Sci 65:55 1556, 1970
10. Chiller JM, Habicht GS, Weigle WO: Kinetic differences
in unresponsiveness of thymus and bone marrow in cells.
Science 1712313-815, 1971
11. Rajewsky K, Brenig C: Paralysis to serum albumin in T
and B lymphocytes in mice. Eur J Immunol 4:120-125,
1974
12. Katsura Y, Kawaguchi S, Muramatsu S: Difference in the
target cells for tolerance induction in relation to the dose
of tolerogen. Immunology 23:537-544, 1972
13. Weigle WO: Analysis of autoimmunity through experimental models of thyroiditis and allergic encephalomyelitis. Adv Immunol30:159-273, 1980
14. Theis GA, Siskind GW: Secretion of cell populations in
induction of tolerance: a m i t y of antibody formed in partially tolerant rabbits. J Immunol 100:138-141, 1968
15. Alspaugh MA, Jensen FC, Rabin H, Tan EM: Lymphocytes transformed by Epstein-Barr virus: induction of nuclear antigen reactive with antibody in rheumatoid arthritis. J Exp Med 147:1018-1027, 1978
16. Luzzati AL, Hengartner H, Schreir MH: Induction of
plaque-forming cells in cultured human lymphocytes by
combined action of antigen and EB virus. Nature
269:419420, 1977
17. Rosen A, Gergely P, Jandel M, Klein G, Britton S: Polyclonal Ig production after Epstein-Barr virus infection of
human lymphocytes in vitro. Nature 267:52-54, 1977
18. Slaughter L, Carson DA, Jensen FC, Holbrook TL,
Vaughan JH: In vitro effects of Epstein-Barr virus on peripheral blood mononuclear cells from patients with rheumatoid arthritis and normal subjects. J Exp Med
148:1429-1434, 1978
19. Paterson PY: Autoimmune neurological disease: experimental animal models and implications for multiple sclerosis, Autoimmunity. Edited by N Talal. New York, Academic Press, 1977, pp 643-691
20. Ortiz-Ortiz L, Nakamura RM, Weigle WO: T cell requirement for experimental allergic encephalomyelitis induction in the rat. J Immunol 117:576-579, 1976
21. Ortiz-OrtizL, Weigle WO: Cellular events in the induc-
1052
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
tion of experimental allergic encephalomyelitis in rats. J
Exp Med 144:604-616, 1975
Ada GL, Byrt P, Mandel T, Warner N: A specific reaction
between antigen labeled with radioactive iodine and lymphocyte-like cells from normal, tolerant and immunized
mice or rats, Developmental Aspects of Antibody Formation and Structure. Vol 2. Edited by J Sterzl, I Riha.
Prague, Academic, 1970, pp 503-516
Humphrey JH, Keller HU: Some evidence for specific interaction between immunologically competent cells and
antigens, Developmental Aspects of Antibody Formation
and Structure. Vol 2. Edited by J Sterzl, I Riha. Prague,
Academic, 1970, pp 485-502
Lindstrom J: Autoimmune response to acetylcholine receptors in myasthenia gravis and its animal model. Adv
Immunol27: 1-50, 1979
De Baets MH, Einarson B, Lindstrom JM,Weigle WO:
Cellular and humoral mechanism in experimental autoimmune myasthenia gravis. Fed Proc 40: I 139, 1981
Rose N, Witebsky E: Studies on organ specificity. V.
Changes in the thyroid glands of rabbits following active
immunization with rabbit thyroid extract. J Immunol
76:417427, I956
Weigle WO: The production of thyroiditis and antibody
following injection of unaltered thyroglobulin without adjuvant into rabbits previously stimulated with altered thyroglobulin. J Exp Med 122:1049-1062, 1965
Weigle WO, Nakamura RM: The development of autoimmune thyroiditis in rabbits following injection of
aqueous preparations of heterologous thyroglobulin. J
Immunol99:223-231, 1967
Clinton B, Weigle WO: Cellular events during the induction of experimental thyroiditis in rabbits. J Exp Med
136:1605-16 15, 1972
Nakamura RM, Weigle WO: Experimental thyroiditis in
complement intact and deficient mice following injection
of heterologous thyroglobulin without adjuvant. Proc SOC
Exp Biol Med 129:412-416, 1968
Clagett JA, Weigle WO: Roles of T and B lymphocytes in
the termination of unresponsiveness to autologous thyroglobulin in mice. J Exp Med 139:643--660, 1974
WEIGLE
32. Weigle WO: The discrimination between self and nonself,
Biology of Bone Marrow Transplantation, ICN-UCLA
Symposium on Molecular and Cellular Biology. Vol. 27.
Edited by RP Gale, CF Fox. New York, Academic Press,
Inc, 1980. pp 341-356
33. Kong Y-CM, David CS, Gerald0 AA, Elrehewy M, Rose
NR: Regulation of autoimmune response to mouse thyroglobulin: influence of H-2D end genes. J Immunol
123~15-18,1979
34. Fravi G, Lindenmann J: Induction of allogeneic extracts
of liver-specific precipitating autoantibodies in the mouse.
Nature 218:141, 1968
35. Iverson GM, Lindenmann J: The role of a camer-determinant and T cells in the induction of liver-specific autoantibodies in the mouse. Eur J Immunol 2:195-197,
1972
36. Ruoslahti E, Wigzell H: Breakage of tolerance to a-fetoprotein in monkeys. Nature 255:7 16-717, 1975
37. Ruoslahti E, Pihko H, Becker M, Makela 0: Rabbit alpha-fetoprotein: normal levels and breakage of tolerance
with haptenated homologous alpha-fetoprotein. Eur J Immunol 5:7-10, 1975
38. Rajewsky K: Rabbit antibody to pig lactic dehydrogenase
reacting with the rabbits own enzyme. Immunochem
3:487489, 1966
39. Gershon RK: Suppressor T cell dysfunction as a possible
cause for autoimmunity, Autoimmunity. Edited by N
Talal. New York, Academic Press, 1977, pp 171-181
40. Eardley DD, Hugenberger J, McVay-Boudreau L, Shen
FW, Gershon RK, Cantor H: Immunoregulatory circuits
among T-cell sets. I. T-helper cells induce other T-cell sets
to exert feedback inhibition. J Exp Med 147:1106-1I15,
1978
41. Cantor H, Gershon RK: Immunological circuits: cellular
composition, Symposium on Immune Regulation. Edited
by RK Gershon. Fed Proc 38:2058-2064, 1979
42. Reinherz EL, Schlossman SF: Characterization of regulatory T cells in man, Regulatory T cells, Biomedical Science Symposia. Edited by B Pernis, HJ Vogel. New York,
Academic Press, Inc, 1981