The Effector Functions of Antibodies

Abstract

“Antibody” is one of those specialists' words that became common language. Everyone knows that antibodies protect against infectious diseases, especially since the COVID-19 pandemic swept across the world. Everyone, or almost, even knows what antibodies look like. Their anthropomorphic Y shape has become an iconic symbol that most societies of immunology have included in their logo. What antibodies actually are, however, is not so clear in everyone's mind, as judged by media which often confuse serum with vaccine. What antibodies do and how they work is another question. Their well-known ability to recognize specific antigens with each of their two “arms” is often thought to be enough to protect; even by scientists, sometimes by immunologists. Whatever how, antibodies protect, and when they have pathogenic effects, these are viewed as the unfortunate consequences of targeting errors such as in allergic and autoimmune diseases, or collateral damages such as in inflammatory diseases.

Antibodies are also well known as tools. Due to their exquisite specificity, antibodies have proven unrivaled diagnostic tools and they are used in a variety of techniques adopted by all medical disciplines and beyond. Due to their powerful biological properties, antibodies have been increasingly used as therapeutic tools with amazing efficiencies. This is not new: antibodies saved thousands of children from diphtheria and many more wounded soldiers from tetanus at the beginning of the 20th century, when they were nothing but elusive substances in immune serum. They are well-known molecules now and, as serum therapy for deadly infectious diseases yesterday, humanized monoclonal antibodies have provided long-sought cures for cancers with a poor prognosis today. Not without side effects, though. But antibodies can be engineered genetically to enhance their expected effects and to decrease their unwanted effects.

Why, therefore, put together another series of review articles on such well-known molecules? As stated in its title, this volume of Immunological Reviews is focused on the effector functions of antibodies. Antibodies are bi-functional molecules: They can not only recognize antigens; they can also act on them. How they do so is poorly known by immunologists, except those who work specifically on the subject. Yet, antibodies are the main effectors of adaptive immunity, at least quantitatively: 10 mg/mL IgG and 2–3 mg/mL IgA circulate in the blood stream—and much more are present in tissues since 80% immunoglobulin-secreting plasma cells of the whole body produce mucosal IgA. How do these antibodies deal with pathogens and commensals? How can they both prevent infections and tolerate microbiotas without inducing devastating inflammatory reactions? How antibodies induced by vaccines exert their protective effects?

This volume deals with the effector functions of antibodies not only in health, but also in disease. If they protect against infectious disease, antibodies can also cause diseases, and these do not result from mistakes of the immune system. IgG antibodies that recognize the same antigen can indeed be both protective and pathogenic. IgA antibodies can have both pro- and anti-inflammatory effects. The pathogenic role of IgE antibodies is well known, but their protective role remains hypothetic. What makes antibodies protective and/or pathogenic? How can they protect, cure, make sick, and sometimes kill? How can one explain so many, in some cases opposite, effects?

One explanation is that antibodies have no effector functions per se. They can do nothing but bind. They bind specifically to antigens via their Fab “arms,” but they bind also to effector systems via their Fc “leg.” Doing so, they bring antigens close to effectors that can act on them by a variety of mechanisms. It follows that the effector functions of antibodies are not theirs. They are those of a variety of effectors that antibodies recruit and activate. This volume discusses these mechanisms, the recruitment of effectors by antibodies, their activation, and their effects on antigen. Its purpose is to help the reader understand how antibodies work, how they can exert their many effects, for better and for worse, how they can protect and how they can make sick, how they can be used too, and tailored to achieve specific therapeutic effects.

When Paul Ehrlich forged the word in 1891 [1], the meaning of “antibody” was far from clear. What were these soluble substances that nobody had seen, whose existence was inferred from biological activities that appeared in the serum of immunized animals, and whose ability to protect against deadly diseases made them “the magic bullets of immunity” [2]? What did “anti” and “body” mean and what did they designate when they were associated to form a new word? What do they tell us today of what we keep calling “antibodies”?

In 1891, Ehrlich was studying the biological activities of immune sera that Shibasaburō Kitasato and Emil von Behring had just found to protect against diphtheria or tetanus [3]. These diseases had recently been shown to be caused by toxins secreted by the responsible bacteria, and antitoxins—a word first used in Italian (antitossine) by Guido Tizzoni and Giuseppina Cattani in April 1891—present in the serum of immunized animals were thought to account for the observed protection. The German word “Antikörper” (antibody)—derived from anti-toxischer Körper (anti-toxic body)—was coined by Ehrlich in October 1891 on the model of “antitoxin” [1]. An antitoxin was a body (Körper) that acted on a toxin; an antibody was therefore a body that acted on something. In “antibody,” “body” was initially the subject of an action. It designated the antibody itself. Yet, “body” could designate the opposite and be the object on which an antibody acts; in other words, it could designate an antigen [4]. The meaning of “antibody” was indeed ambiguous: It was a body anti-(another) body (The confusion was even worse when antibodies were directed against antibodies. Antibodies can be antigenns too…). Today, “body” seems to have lost its initial meaning, possibly because we learnt that an antibody is an immunoglobulin with a well-known structure. Paradoxically, an antibody stopped being a body when it acquired a material existence, and “body” designates now what it acts on. An antibody is an anti-body.

Whatever the body, what does the prefix “anti” mean in “antibody”? “Anti” combines three notions: binding, specificity, and antagonism. First of all, Ehrlich thought that antitoxins present in the serum of immunized animals neutralized toxins when binding to them. “Corpora non agunt nisi fixata” [5], he said, which translates “bodies (corpora) do not act if they do not bind” or simply “bodies must bind in order to act.” Binding was necessary for antitoxins to neutralize toxins.

Second, Ehrlich understood the neutralization of a toxin like the neutralization of an acid by a base. Unlike the acid–base reaction, however, toxin neutralization was specific: anti-tetanus toxin neutralized tetanus toxin but not diphtheria toxin, and anti-diphtheria toxin neutralized diphtheria toxin but not tetanus toxin. Earlier, Ehrlich had studied the ability of colored chemicals to stain specific cells and tissues. The reason why tissues were specifically stained, he thought, was that they possessed “side chains,” to which specific stains could bind. Likewise, antitoxins could neutralize specific toxins because they bound to specific side chains borne by these toxins as a key fits in a specific lock only, and doing so, they prevented toxins from acting [6]. To illustrate the specificity of the toxin–antitoxin reaction, Ehrlich borrowed the analogy used by Emil Fischer to illustrate the specificity of enzymes for their substrates [7]. Like the specificity of a key for a lock, that of antitoxins for toxins was explained by a spatial complementarity.

Antitoxins could therefore (1) bind, (2) specifically to toxins, and doing so, (3) neutralize them. The third meaning of “anti” is against. Literally, “to neutralize” means the act of making a substance neutral. Again, it refers to acids and bases that form neutral salts when they combine. It also means to render something ineffective. By extension, it applies to potentially harmful things or persons, and in military parlance, it means to kill when applied to enemies. Antitoxins render toxins harmless. To explain this effect, Ehrlich hypothesized that toxins, “… unite with certain chemical groupings in the protoplasm of cells, […] and that this chemical union represents the prerequisite and cause of the disease” [6], and he called such chemical groupings “poison receptors” (Italics in citations were in Ehrlich's original text). He concluded “that the group in the protoplasm, the cellular receptor, must be identical to the antitoxin which is contained in solution in the serum of immunized animals” [6]. Antitoxins were soluble toxin receptors, and neutralization was the result of a competition between cell-bound and soluble receptors for the toxin.

It was soon noticed that all kinds of bodies including those that had no toxicity could induce antibodies as toxins induced antitoxins: “Even the genuine protein substances of animal and plant organisms are able, irrespective of whether they have a toxic effect or not, to produce antibodies,” Ehrlich said [6]. Such antibodies had no toxicity to neutralize. They were nevertheless directed against bodies. As toxins, these “genuine protein substances” had “receptors” through which they could act (because corpora non agunt nisi fixata), and these receptors could be released as antibodies, in the same way as toxin receptors could be released as antitoxins. Antibodies had no direct effect on the bodies against which they were directed, but they could prevent their action by competing with corresponding receptors for the same bodies. “Anti” had the same antagonistic meaning in “antibody” and in “antitoxin,” and today's antibodies keep being directed against what they recognize. One notices that B cells' BCRs from which they derive have the same specificity as antibodies, but they are merely antigen receptors, whereas antibodies are directed against antigen. Antibodies are definitely antibodies.

Thus, specific antibodies could be directed against any of a large number of molecules, whether from microbes or from normal cells, whether toxic or not toxic, against a variety of bodies. After the general name “antibody,” the general name “antigen” started to be used to designate collectively all these bodies against which antibodies can be directed. The term was first used in French by Ladislas Deutsch in 1899 [8]: “Substances immunogènes ou antigènes,” that is, substances that lead to the formation of “immune bodies.” In Deutsch's mind, immune bodies (antibodies) were not the result of an immune response, but that of the transformation of antigens. “Immunogenic substances or antigens”: an antigen was what could generate antibodies. The suffix “gen” can indeed mean either what is at the origin of something such as in “fibrinogen,” or what triggers the production of something such as in “pathogen” [4]. The second meaning is of course the one that is used today.

If the word “antibodies” designates what can bind to bodies, the word “antigens” therefore does not designate the bodies to which antibodies bind. It designates the bodies that can induce the production of antibodies. The chemical notion of interaction between two substances was replaced by the biological notion of stimulation/response. Indeed, it is in response to a stimulation by an antigen that the organism produces antibodies against this antigen. “Anti” does not have the same meaning In the word “antigen” and in the word “antibody.” In “antigen,” it implies no affinity, no binding, no antagonism. It is nothing but an abbreviation that stands for “antibody.” As for “gen,” it means “which generates.” Therefore, if an antibody is a molecule that can bind to an antigen, an antigen is a molecule that can induce the production of an antibody. Why so much confusion?

The situation was indeed confusing at the beginning of the 20th century. The property of antibodies to bind to antigens was inferred rather than observed. Its consequences were easier to see than binding itself. Antibodies were therefore commonly named after their biological activities. Besides antitoxins that neutralized toxins, “bacteriolysins” dissolved bacteria, “agglutinins” agglutinated particular antigens such as bacteria or erythrocytes, “precipitins” made soluble antigens precipitate at the bottom of test tubes, “hemolysins” (from the Greek haima, blood) induced the lysis of red blood cells, “opsonins” (from the Greek opson, which prepares food) enabled the phagocytosis of particulate antigens, and we keep talking of “cold agglutinins” in medical practice. Antibodies were what they did. However, since antibodies could act on specific antigens, they could also be named by their specificity: anti-tetanus antibodies, anti-ovalbumin antibodies, autoantibodies, anti-SARS-CoV-2 antibodies, etc., irrespectively of their biological properties. The word “antibody” started to mean both antigen-specific and biologically active molecules when two scientific issues were clarified. The first issue was raised by Ehrlich himself: How could one explain the infinite (or almost) number of possible antibodies if there are specific antibodies for all antigens and if the number of antigens is infinite (or almost)? The second issue was as follows: How could antibodies have so many biological activities? Are these activities those of one antibody that can exert many, or those of many antibodies that can exert, each, one only?

The first issue, that is, that of the diversity of antibodies and of their generation was clarified first by Macfarlane Burnet's clonal selection theory in the 1950s [9], then by the elucidation of the genetic mechanisms at the origin of antibody diversity by Susumu Tonegawa in the early 1980s [10]. This issue is not the subject of this volume. The second issue, namely, the cellular and molecular bases of the effector functions of antibodies, is. It took a long time, and many investigations led independently by several groups worldwide for this issue to be progressively clarified. What was found is that most effector functions of antibodies are not those of antibodies themselves. They are those of effector systems that antibodies use when they act. Effector systems are of two kinds: molecules and cells. Molecules used by antibodies are primarily those of the complement system. Cells used by antibodies are primarily those of the myeloid lineage. The involvement of both molecules and cells depends on the Fc portion of antibodies. Via their Fc portion, antibodies can bind to complement components in solution, to Fc Receptors (FcRs) expressed by myeloid cells and, when bound to complement components, to complement receptors (CRs). Altogether, complement, FcR- and CR-expressing cells can exert numerous actions on antigens recognized by the Fab portions of antibodies. The result of Fab-mediated antigen recognition and Fc-mediated activation of effector mechanisms is a wide array of biological effects that affect a variety of antigen-bearing molecules, cells, or organisms. Some effects are protective whereas others are pathogenic; others are neither protective nor pathological, but regulatory (Figure 1).

The review articles contained in this volume provide in-depth discussions of the right part of Figure 1, that is, of the effector functions of antibodies in health and disease. These 28 reviews are organized in five sections.

This volume was initially planned to have a sixth section entitled “Antibodies and vaccines” in which knowledge on vaccine-elicited antibodies would have been reviewed. Indeed, with the exception of BCG, the protection conferred by all efficient vaccines depends on antibodies. It would have been interesting to know (1) which isotypes of antibodies are induced by efficient vaccines, which mechanisms they use to protect and against which diseases, and (2) how can one manufacture specific vaccines that induce the desired antibodies at the right time in the right place. One can guess that vaccines against systemic or mucosal, viral, or bacterial infections may not involve the same antibodies and the same effector mechanisms, and that they should be engineered differentially in order to preferentially induce the appropriate antibodies. I was not able to build up this section based on available literature on the subject. The new science of immunity which bloomed at the dawn of the 20th century aimed at explaining protective immunity conferred by Pasteurian vaccines. Immunology described exquisitely specific recognition means and a plethora of powerful effector mechanisms that were integrated in a sophisticated biological system, but how vaccines work remains poorly known, more than one century later.

I will end by mentioning that 82 authors and coauthors from 14 countries wrote the 28 review articles gathered in this volume. They all played the game proposed by Immunological Reviews, that is, reviewing their own work and thoughts in the context of others' findings rather than providing comprehensive reviews on a subject. As a result, they all collaborated to build a unique collection of original publications written in different styles, in which a variety of points of view on the same topic—including some redundancies that offer different ways of presenting things—complement, comfort, and confront each other.

The author declares no conflicts of interest.