Por Nelson Vaz
The idiotypic network theory (Jerne,
1974a) is now almost forty years old and has definitely failed to influence the
central tenets of immunological thinking. This is unfortunate because it was the
only theory proposing something different from the traditional description of
immunological phenomena as derived from independent actions of lymphocytes
clones. The clonal selection theory (Burnet, 1957) and all its variants are
theories about the behavior of lymphocyte clones, not about an immune system.
As soon as 1974, Jerne's visionary ideas suggested, that the problems raised by
clonal selection would be replaced by something he called "systems analysis", more than three
decades before the current flurry of interest in the still foggy notion of
"systems biology" (Cornish-Bowden, 2006; Hood et al., 2005; Keller,
2005; Kirschner, 2005; Kitano, 2002; Van Speybroeck et al., 2005). But Jerne’
predicition was not fulfilled: "systems analysis", whatever it means
now, is as yet to raise the interest of immunologists.
Jerne introduced the idea of
anti-antibodies, or anti-idiotypic antibodies, as a way to interconnect
lymphocyte clones in a web, or network. Immunological activity is usually seen
as atomized in isolated (specific) clones which either expand or are impeded to
expand. However, any “systemic” view requires connections among these
components. The simplest way to link these clones among each other would be to
display on each lymphocyte (on its clonal receptors) an exclusive, distinctive
element (an idiotope) recognizable by other lymphocytes. The resultant complex
multiconnected network of interactions was what Jerne identified as the immune
system itself, embodying the true nature of immunological activity, no longer
dispersed into independent clones (Jerne, 1972; 1973). In his 1974 annual report to the Basel
Institute of immunology, he wrote:
The known
recognition elements are the combining sites (paratopes) of the V-domains that
occur on antibody molecules and at the surface of lymphocytes. The fact that
V-domains have been shown also to carry idiotopes that are immunogenic even in
the individuals that produce these V-domains
forces us to reconsider the validity of "linear" ideas about
the immune system (i.e., the notion that the immune response of lymphocytes is
solely antigen-driven and is independent of lymphocytes and antibody molecules
that do not recognize this antigen), in favor of a "network" notion
(i.e., the idea that every lymphocyte is under the continuous surveillance of
other lymphocytes and of antibody molecules by idiotope-paratope interactions).
The fact that animals have been shown to produce anti-idiotypic antibodies to
their own antibodies proves that these anti-idiotypic antibodies were already
present, since clonal selection requires antibodies to be present prior to an
antigenic stimulus. In order to recognize all idiotopes, all antibodies must be
anti-idiotypic antibodies. Contrary to the notions of the 1960's, which regarded the huge set of paratopes of
an immune system as an "open"
set of recognizing elements, most of which would never meet a fitting antigen, we must now accept that the immune system is essentially
"closed" or self-sufficient in this respect: the paratope of any V-domain will
recognize idiotopes on many other V-domains, and the idiotope of any V-domain
will be recognized by the paratopes of many other V-domains (Jerne, 1974b). (italics added)
This notion of “operational closure”
of the immune system was not further mentioned by Jerne, neither in the papers
proposing the network theory, (Jerne, 1973; Jerne, 1974a,b,c), nor in an
important preceding paper (Jerne, 1971). A few years later, the late Francisco
Varela and I published a paper
emphasizing that the notion of closure - organizational closure -, is central for the a proper description of
the immune system. The paper received laudatory comments of Jerne, but remained
otherwise unknown (Vaz and Varela, 1978).
The tenets of the network theory has
unforeseen consequences not mentioned (or not acknowledged) by Jerne. The
operational closure of the immune system is necessary to stabilize the network
because, if there are anti-antibodies, there would be also
anti-anti-antibodies, in an infinite regression which would make the system
explode in a feed-forward cascade (Gell. & Kelus, 1967). Instead of the
systemic idea of closure, Jerne used the local notion of suppressive
interactions; he proposed that most interactions among lymphocytes were suppresive,
i.e., inhibitory and, thus, lost the opportunity to make a generalizing
proposal. This was a crucial point because, when Jerne proposed that all antibodies
(all immunoglobulins) are auto-antibodies (Jerne, 1974b, quoted above) [1], the class of non-auto-antibodies
becomes empty and the prefix “auto” becomes meaningless. If all antibodies are
auto-reactive, auto-antibodies are not a special class of entities; there is no
longer sense in auto-immune diseases and related concepts. Thus, network
theories are actually incompatible with self/nonself discrimination, a notion
that is central in the clonal selection theory. This is not usually
acknowledged.
From a strict point of view, network
notions are not even compatible with the notions of specific immune responses,
immunological memory and tolerance, notions that have been defined in terms of
clonal expansions and contractions. But Jerne’s theory failed to provide
alternative explanations for these immunological phenomena and, as a result,
was gradually forgotten (Coutinho, 1995; Eichmann, 2008).
Systems
To properly approach these issues,
we must first consider elementary definitions about networks as systems. In
their simplest definition, systems
are a collection of interconnected elements in which interfering with one
component, influences all the other components. Our main interest here are
complex, dynamic systems, such as living systems. What do complex dynamic
systems do? How to study them?
Donella Meadows, who wrote mostly
about social and economic systems, but offered general conclusions, said that systems are impossible to control and we
can only "dance" with them (Meadows, 2000). The aim to
"regulate" a system is misleading because the very notion of
regulation is synonimous with the notion
of systems. The concept of homeostasis is a second synonimous to dynamic
systems. because there are no “non-homeostatic” systems.
These ideas are formalized and
greatly expanded by Maturana in his Biology of Cognition (Maturana, 2002:
Maturana and Poerlsen, 2004; Maturana & Varela, 1980). In his way of
seeing, systems accept no inputs and have no outputs, cannot be stimulated and
do not respond to anything; they undergo perturbations, which are structural
changes either derived from their internal dynamics or resulting from their
interaction with the medium in which they operate. Perturbation is possibly a
bad name, because dynamic systems never reach in a non-perturbed state. These perturbations
require compensative changes, otherwise the system is destroyed. However,
perturbations are not inputs (stimuli) and compensative changes are not
responses (outputs), because both these terms refer to the operation of a
larger compound entity (the system) which is perturbed and undergoes
compensative changes. Stimulus and reponses belong to a way of seeing linked to
the idea of causes, whereas the
notion of systems is related the ideas such as autonomy and independent operations, and, as mentioned above, the
notion of operational closure (Vaz and Varela, 1978).
Closure does not mean isolation.
This point is frequently misunderstood even by famous network immunologists
(see Coutinho, 2003). All systems operate in a medium with which they interact
and all systems are necessarily open for
these interactions, although closed in their dynamics, their ways of
changing their structure. Systems remind us of the French proverb: "the
more it changes, the more it remains the same."
The organization of a system is that
which remains invariant while everything else in the system may change. Dynamic
systems, such as living systems, continuously replace their components, i.e.,
change their structure, without changing its organization. The organization
defines the class-identity of a system. The organization chair is a set of
relations among components of a chair which are present in all chairs
regardless of their structure and their structural changes. When these
relations are modified, chairs becomes something else. Functional descriptions
do not help us to understand the organization of systems; chair can be used as
objects for sitting as well as for many other functions (Figure 1).
Figure 1. Damian Ortega (1997), "Puente" (Bridge); Photo Eduardo Eckenfels.
Inhotin,
Contemporary
Art Museum, Brumadinho, MG, Brazil.
Living
systems are characterized by a self-maintaining dynamic organization which
Maturana has named autopoietic, or
self-producing (Maturana and Varela,
1980; 1987; Maturana and Mpodozis, 2000; Maturana, 2002; Maturana and Poerksen,
2004). In the structural domain, living organisms may be described as molecular
machines that produce themselves; and when they stop changing in this
particular manner - when they change their ways of changing - they desintegrate and die.
The
fundamental living systems are cells and unicellular organisms, first-order
autopoietic entities. In multicellular organisms, second-order autopoietic
systems, we may distinguish sub-systems,
such as the nervous and immune systems. Our aim is finding a proper way to
describe these sub-systems and say something about their organization; for
example, to ask: What remains invariant in the nervous system? What remains
constant in the immune system in spite of their coninuous activity and their
ceaseless replacement of cellular and molecular components?
The immune system
These
questions are not commonly made because immunology is entirely focused on a
special kind of changes of lymphocyte activity, called specific immune
responses, which are believed to be expressions of its physiology.On the other
hand, we are interested in constancy and conservation, on what remains
invariant during variations (Vaz, 2006; Vaz et al., 2006). Our aim is to define an organization for the
immune system, thus, that which remains invariant amid variation.
Usually, it
would be considered impossible to understand how something utterly complex is
organized. However, it is not necessary to describe all the components of
living systems and all the relations among them to postulate that they form
self-producing, self-maintaining molecular machines, organizationally closed.
It is not necessary to describe all the details of the relations among neurons
to describe the nervous system as a closed network of neuronal interactions, in
which relative states of neuronal activation can only lead to other relative
states of neuronal activation. Neuronal activities may be seen as closed upon
themselves, beginning and ending within the nervous system itself Maturana
& Varela, 1980).
It is also important to remember that
every system operates in a medium, which makes it possible and with which it
interacts. The medium in which the nervous system operates is the organism of
which it is a component. The nervous system is perturbed by the activity of the
organism and undergoes compensative structural changes which, in its turn,
trigger perturbations in the organism; which again trigger perturbations in the
nervous system, in a ceaseless dance that goes on as long as living exists.
Similarly,
the immune system may be described as a closed network of lymphocyte
interactions, in which relative states of lymphocyte activation can only lead
to other relative states of lymphocyte activation, i.e., lymphocyte activities
are closed upon themselves, and they begin and end within the immune system
itself. The immune system is in continuous interaction with the organism of
which it is a component; it is perturbed
by its own dynamics and by interactions with the organism and undergoes
compensative structural changes; in its turn, these changes trigger
perturbations in the organism, which again trigger perturbations in the immune
system, in a endless dance that goes on as long as living exists. Self/nonself
discrimination is a pseudoproblem that dominated immunology for more than half
a century.
Contradictions
Obviously,
from the traditional way of seeing all this sounds as nonsense and seems to be
in flagrant contradiction with common sense and abundant experimental evidence
showing that both the nervous and the immune system interact not only with the organism to which they
belong, but also, and perhaps mainly, with the medium in which the organism, as
a whole, operates. Is it not utterly apparent that the organism forms specific
antibodies to invading antigens? Is it
not obvious that these antibodies specifically bind to the antigen in a variety
of in vitro tests? Are not antigens believed to determine
(specify) antibody formation? Just to
mention a single example:
"...(I)n healthy
human subjects, not previously exposed to the rabies virus, nearly 2% of the
circulating B lymphocytes were committed to the production of antibodies that
bound the virus. These B cells expressed the surface CD5 molecule. The
antibodies they produce were polyreactive IgM that displayed a relatively low
affinity for the virus components. After immunization, different anti-virus
(IgG and IgA) antibody-producing cells consistently appeared in the circulation
and increased from less than 0.005% to greater than 10% of the total B cells
committed to the production of IgG and IgA, respectively. Most of such B cells
do not express CD5 and produce monoreactive antibodies of high affinity for
rabies virus." (Ueki et al., 1990).
These traditional beliefs are examples of what Maturana calls the "fallacy of instructive interactions" (Maturana, 2002). The structural changes which dynamic systems undergo in their operation cannot be and are not determined (specified) by interactions with the medium in which they operate; these interactions can only trigger structural changes which are determined (specified) by the dynamics of the system itself. Actually, it is the structure of the system that determines (specifies) which elements of the medium can interact with the system; except for destructive alterations, all the changes a system undergoes are determined (specified) by the system itself.
However, as
observers, we can describe the operation of living systems in two separate
non-intersecting domains: we may see their dynamic
organismic/cellular/molecular structure and
we may also see the organism, as a whole entity, interact with its
medium; and we may note that certain structural changes take place
concomitantly or immediatlely after certain interactions with the medium; and
we may we be misled to believe that the interactions determined (specified,
caused) the structural changes. But this is not so. (Maturana, 2002; Maturana
and Poerksen, 2004).
How can
antigens, therefore, provoke the formation of specific antibodies?
The specificity of immunological
obervations
The answer
to this question demands what is actually much more than a radical revision of
basic immunological tenets: it invites us to consider the crucial role played
by deliberate human actions coordinated by human languaging (Maturana, 2002) in generating experimental findings,
which are generally considered objective.
In our way
of seeing, "antibodies" are functional labels pasted on natural
immunoglobulins by means of serological tests designed and performed with the
specific aim of characterizing and/or quantitating their very presence. Specific antibodies are tautologies created
by our way of observing (Vaz, 2011a,b,c).
Natural
immunoglobulins and specific antibodies are described in different domains of
description. Natural immunoglobulins emerge spontaneously in an antigen-free
intracellular compartment and are then placed on the membrane of B lymphocytes,
where they become BCR (B cell clonal receptors). As BCR, these molecules eventually
bind a variety of elements present in their vicinity. Most of them will bind
molecules which are components of the organism itself, including other
immunoglobulins (Varela and Stewart, 1990), but also a variety of other
molecules (Nóbrega et al., 2002; Merbl et al., 2007). Some of these ligands,
however, may be invading molecules that were not produced by the organism, such
as materials derived from the gut, as dietary components and elements from the
bacterial flora, or, derived from infections by virus, germs or parasites. Both
kinds of reactions - i.e., those triggered by materials derived from internal
or external sources - may contribute to the activation and differentiation of
the B lymphocyte, with or without the "help" ot T lymphoctes, to produce
and secrete immunoglobulins with the same specificity of the BCR to the
extracellular fluid. Some of these immunoglobulins may then be identified in
the circulation as specific antibodies by serological testing.
In
describing this sequence of events, we moved between two separate domains of
description. Natural immunoglobulins are
produced without definite targets; on the other hand, antibody specificity is
ascribed to some of these immunoglobulins by means of tests which were designed
with exactly this directionality. In short, immunoglobulins are structural
components of the organism; specific antibodies are functional labels that
emerge in the coordinations of actions among immunologists; as entities,
antibodies are more like verbs, than nouns. And if we approach natural
immunoglobulins within this new way of seeing, some of their known properties
acquire a different meaning.
Polyreactivity and its hidden
aspects
For a long time now, it is
recognized that natural immunoglobulins may be polyreactive (Varela and
Stewart, 1990). The notion that specific antibodies admit
"cross-reactions" is present since the every origin of immunological
practice. The first and most effective anti-infectious vaccine, that against
smallpox, is a blatant example of cross-reactivity, in which immunity to the smallpox virus is achieved by exposure
to different virus: the vaccinia virus.
There are examples of clinically important serological tests which explore
cross reactions, of which the Wassermann test for diagnosis of syphilis is a
major example that motivated an important book in the philosophy of science
(Fleck, 1935); but there are several other cases, such as the Paul Bunnel
reaction for the diagnosis of infectious mononucleosis. But cross-reactivity in
serological testing is just one of the many consequences of polyreactive
immunoglobulins and not the ones we want to focus at this moment.
During B lymphocyte development,
antibodies are assembled by mechanisms of reassortment of gene segments, believed to happen at
random; this generates a vast number of different structures (V-regions), but,
due to the randomness of the process, some of the antibodies produced are
self-reactive. Is this self-reactivity
detrimental to the organism, as usually believed, or is it an essential aspect
of the organization of the immune system, as we believe?.
A possible explanation for
polyreactivity is that the antigen-binding ‘pocket’ of many antibody molecules
is more flexible than previously thought and can change conformation to
accommodate different antigens (Notkins, 2004). Polyreactive natural
immunoglobulins, specially IgM, have been indicated as the major source of a
"physiologic autoreactivity". However, this is not an adequate
description of the situation because
this auto-reactivity is granted almost as an indulgence, as accessory to a
responsiveness directed mostly outward, to external antigenic targets.
The polyreactivity of natural
immunoglobulins include other immunoglobulins among their many internal ligands
and t this has unforeseen consequences. Polyreactivity has been considered an
artifact generated during isolation procedures that partially denature
immunoglobulins (McMahon and O'Kennedy, 2000; Bouvet et al., 2001) but this is
highly unlikely. Polyreactive immunoglobulins may be collected in the fresh
supernantants of B cell cultures submitted to no further isolation procedure.
Moreover, when monoclonal immunoglobulins are tested upon complex mixtures of
ligands, they rarely form single bands, i.e., most of them react detectably
with many ligands. Thus, polyreactivity sems to be a natural property of
immunoglobulins, specially IgM, but also present in IgG and IgA.
The first and most notable
unforeseen consequence the polyreactivity of natural immunoglobulins is that
immunoglobulins, when isolated from serum, display hidden
"neo-reactivities" with a wide array of ligands, with which they did
not react while in whole serum. These "hidden" reactivities again
disappear when they are added back to the serum (Adib et al., 1990; Benerman et
al., 1993; Sigounas et al., 1994) .
The problem of polyreactive
immunoglobulins is minimized in the study of immunology because it is seen as an imprecision, a nuisance,
which generates cross-reactions as negative events in the dominant way of
conceiving specific immunity; and also the possibility of auto-reactivity,
which is usually seen as unnatural and potentially pathogenic. But polyreactive
antibodies can hardly be ignored.
Although generally considered a
minor and exceptional fraction of natural antibodies, polyreactive
immunoglobulins may actually include all immunoglobulins. A large proportion of
circulating B cells from healthy subjects are committed to the production of
IgM antibodies that are polyreactive and bound a variety of self- and exogenous
Ag. In contrast, significantly higher frequencies of cell precursors producing
monoreactive IgG autoantibodies to thyroid Ag (thyroglobulin and thyroid
microsomal Ag) and ssDNA were found in Hashimoto's disease and SLE patients,
respectively (Nakamura et al., 1988). The reactivity of IgG in normal mouse
serum to mouse actin and tubulin, DNA, and TNP groups is very low compared to
that of the IgM, but this activity was considerably increased when IgG was
separated from serum, by affinity chromatography on protein A-Sepharose,
whereas no difference in the IgM activity was observed. Addition of IgM to IgG
isolated from the same serum resulted in the inhibition of IgG binding to these
Ag (Adib et al., 1990). IgG
auto-reactivities are only marginally expressed in whole unfractionated sera
because of IgM-IgG, IgG-IgG and other, still unidentified, interactions
(Berneman et al., 1993). In other studies, human plasma showed only minimal, if
any, reactivity with a panel of antigens as measured by ELISA but IgM
affinity-purified from plasma showed much more reactivity with the same panel
of antigens. When plasma was added back to the affinity-purified IgM, the
reactivity of the IgM with antigens was completely inhibited. When the affinity-purified
IgM was affinity-purified a second time by passage through antigen-specific
columns (e.g., insulin or Fc or beta-galactosidase), the eluted antibodies
bound not only to the antigen used for purification, but also to a panel of
unrelated antigens, indicating that the antibodies were polyreactive. It is
concluded that polyreactive IgM antibodies are present in the circulation but
are masked by binding to circulating antigens (Sigounas, et al., 1994).
In short, both IgM and IgG
antibodies include a substantial proportion of polyreactive antibodies.
Furthermore, the universal existence of idiotypic connections amplify this
problem, because monospecific immunoglobulins may have direct or indirect
connections to multispecific ones.
A second but not less important
implication of polyreactivity is that immunoglobulins produced in vitro are
usually studied as if they were representative of the
immunoglobulins present in blood serum. This may be a misleading assumption
because nascent immunoglobulins have to percolate body tissues before they
accumulate in the circulation and all those binding to body components wil be
rapidly removed. The most drastic example of this situation are immunoglobulins
which have been named "autobodies", because they react with themselves
and are cleared immediately after they are secreted from their producing cells,
if they manage to escape intracellular aggregation ((Kang and Kohler,
1986; Kaveri et al., 1990; 1991).
A third most important aspect of
natural immunoglobulins is that they display robustly stable patterns of
reactivity revealed either by reactions with complex mixtures of protein
ligands (tissue and bacterial extracts) in modified forms of Immunoblotting
(Nóbrega et al., 2002), or, collections of hundreds of different purified
proteins assemble in micro-array (Merbl et al., 2007; Madi et al., 2012).
Similar findings concerning the T cell repertoire may be in the horizon (Davis,
2007; ).
These ideas are of paramount
importance since they have a profound implication on the way we think on the
physiology of immunological activity. The application of methods of lymphocyte
repertoire analysis holds a great promise in diagnosis of immunological
diseases, such as infections, allergies and autoimmune diseases.
Bibliography
Adib, M., Ragimbeau, J., Avrameas, S. and Ternynck, T.
(1990) IgG autoantibody activity in normal mouse serum is controlled by IgM. J
Immunol 145, 3807-3813.
Berneman, A., Guilbert, B., Eschrich, S. and Avrameas, S.
(1993) IgG auto- and polyreactivities of normal human sera. Molecular
Immunology 30, 1499-1510.
Bouvet, J.P., Stahl, D., Rose, S., Quan, C.P., Kazatchkine,
M.D. and Kaveri, S.V. (2001) Induction of natural autoantibody activity
following treatment of human immunoglobulin with dissociating agents. J
Autoimmun 16, 163-72
Cornish-Bowden, A. (2006) Putting systems back into systems
biology.
Perspectivesin
Biology and Medicine 49, 475-489.
Coutinho, A., Kazatchkine, M.D. and Avrameas, S. (1995)
Natural autoantibodies.
Curr Opin
Immunol 7, 812-8.
Coutinho, A. (1995) The network theory: 21 years later.
Eur.J.Immunol.
42, 3-8.
Coutinho, A. (2003) A walk with Francisco Varela from first-
to second-generation networks: in search of the structure, dynamics and
metadynamics of an organism-centered immune system. Biol Res 36,
17-26
Davis, M.M. (2007) The alphabeta T cell repertoire comes
into focus.
Immunity
27, 179-80
Eichmann, K. 2008. The Network Collective - Rise and Fall of a
Scientific Paradigm,
Berlin, Birkhauser.
Fleck, L. (1935) Genesis and development of a scientific
fact.
University
of Chicago Press, Chicago.
Gell, P. G. &
Kelus, A. S. 1967. Anti-antibodies.
Adv Immunol, 6, 461-78.
Hood, L., Heath, J.R., Phelps, M.E. and Lin, B. (2004)
Systems Biology and New Technologies Enable Predictive and Preventative
Medicine. Science 306, 640-643.
Jerne, N.K. (1971) What precedes clonal selection ? In: Ciba
Foundation Symposium, 1971 : Ontogeny of acquired immunity. Elsevier,
Amsterdam, p. 1-15.
Jerne, N.K. (1973) The immune system.
Sci.Amer.
228, 52-60.
Jerne, N.K. (1974a) Towards a network theory of the immune
system.
Ann.
Immunol. 125C, 373-392.
Jerne, N. K. (1974b)
Network notions.
Annual
Report of the Basel Institute for Immunology, 1974,, BII Report <
www.bii.ch>, 6.
Jerne, N. K. (1974c). The immune system: a web of
V-domains.
Harvey Lect, 70 Series, 93-110.
Jerne, N. K. (1984d). Idiotypic networks and other
preconceived ideas.
Immunol.Rev., 79, 1-14.
Kang, C.Y. and Kohler, H. (1986) Immunoglobulin with
complementary paratope and idiotope. J Exp Med 163, 787-96
Kaveri, S.V., Halpern, R., Kang, C.Y. and Kohler, H. (1990)
Self-binding antibodies (autobodies) form specific complexes in solution. J
Immunol 145, 2533-8
Kaveri, S.V., Halpern, R., Kang, C.Y. and Kohler, H. (1991)
Antibodies of different specificities are self-binding: implication for
antibody diversity. Mol Immunol 28, 773-8
Keller, E.F. (2005) The century beyond the gene.
J. Biosci.
30, 3–10.
Kirschner, M. (2005) The Meaning of Systems Biology.
Cell 121,
503-504.
Kitano, H. (2002) Systems Biology: A Brief Overview.
Science
295, 1662-1664.
Madi, A., S. Bransburg-Zabary
, et al. (2012). "The natural autoantibody repertoire in newborns and adults
: a current overview." Adv Exp Med
Biol. 750: 198-212.
Maturana, H. (2002) Autopoiesis, structural coupling and
cognition: a history of these and other notions in the biology of cognition. Cybernetics
& Human Knowing 9, 5-34.
Maturana, H.R. and Varela, F. (1980) Autopoiesis and
cognition: the realization of living. Reidel, Amsterdam.
Maturana, H.R. and Varela, F.J. (1987) The Tree of
Knowledge.Biological Basis of Human Understanding. New Science Library,
Boston.
Maturana, H. and Mpodozis, J. (2000) The origin of species
by means of natural drift. Revista Chilena de Historia Natural 73:261-310
(2000) 73, 261-310
see also
Maturana, H. and
Poerksen, B. (2004) From Being to Doing: The Origins of Biology of Cognition.
Carl-Auer, Heidelberg.
McMahon, M.J. and O'Kennedy,
R. (2000) Polyreactivity as an acquired artefact, rather than a physiologic property, of antibodies: evidence that
monoreactive antibodies may gain the ability to bind to multiple antigens after exposure to low pH. J Immunol Methods 241,
1-10
Merbl, Y., Zucker-Toledano,
M., Quintana, F.J., Cohen, I.R. and A, B. (2007) Newborn humans manifest
autoantibodies to defined self molecules detected by antigen microarray
informatics. J. Clin. Invest. 117, 712–718.
Nakamura, M., Burastero, S.E., Ueki,
Y., Larrick, J.W., Notkins, A.L. and Casali, P. (1988)
Probing the normal and autoimmune B cell
repertoire with Epstein-Barr virus. Frequency of B cells producing monoreactive
high affinity autoantibodies in patients with Hashimoto's disease and systemic
lupus erythematosus. J Immunol 141, 4165-72
Nobrega, A., Stransky, B., Nicolas, N. and Coutinho, A.
(2002) Regeneration of natural antibody repertoire after massive ablation of
lymphoid system: robust selection mechanisms preserve antigen binding
specificities. J Immunol 169, 2971-8
Notkins, A.L. (2004) Polyreactivity of antibody molecules.
Trends in
Immunology 25, 175-179
Sigounas, G., Kolaitis, N., Monell-Torrens, E. and Notkins,
A.L. (1994) Polyreactive IgM antibodies in the circulation are masked by antigen
binding. J Clin Immunol 14, 375-381.
Ueki, Y., Goldfarb, I.S., Harindranath, N., Gore, M.,
Koprowski, H., Notkins, A.L. and Casali, P. (1990) Clonal analysis of a human
antibody response. Quantitation of precursors of antibody-producing cells and generation
and characterization of monoclonal IgM, IgG, and IgA to rabies virus. J Exp
Med 171, 19-34
Van Speybroeck, L., De Backer, P., Van Poucke, J. and De
Waele, D. (2005) The conceptual challenge of Systems Biology. BioEssays 27,
1305–1307.
Varela, F.J. and Stewart, J. (1990) Dynamics of a class of
immune networks. I. Global stability of idiotype interactions. J.Theoret.Biol.
144, 93-101.
Vaz, N.M. and Varela, F.G. (1978) Self and nonsense: an
organism-centered approach to immunology. Med. Hypothesis 4, 231-257.
Vaz, N. (2006) Evolution and conservation ofimmunological
activity. Brazilian Journal of Medical and Biological Research 39,
1521-1524
Vaz, N.M., Ramos, G.C., Pordeus, V. and Carvalho, C.R.
(2006) The conservative physiology of the immune system.A non-metaphoric approach to immunological
activity. Clinical and Developmental Immunology 113, 133-142
Vaz, N. M. (2011a). The specificity of immunological
observations.
Constructivist Foundations, 6, 334-351.
Vaz, N. M. (2011b) Observing Immunologists.
Neurociências, 7, 140-146.
Vaz , N. M., Mpodozis, J. M., Botelho, J. F. &
Ramos, G. C. (2011c). Onde está o
organismo? - Derivas e outras histórias na Biologia e na Imunologia, Florianópolis,
editora-UFSC.
