Immunobiology of the Complement System: An Introduction for Research and Clinical Medicine

Immunobiology of the Complement System: An Introduction for Research and Clinical Medicine provides an introduction to the complement system. The intention was to create a primer that would provide the basic knowledge of complement required for either research or clinical medicine in diseases involving the complement system. The book begins with a historical background of complement research; it introduces certain key investigators from the past who have made important contributions. Separate chapters on the basic aspects of complement function are followed by chapters on the molecular genetics of complement and the role of complement in different diseases. Key topics discussed include the activation of complement via the classical pathway and the alternative pathway; complement mediators of inflammation; opsonization and membrane complement receptors; assembly and functions of the terminal components; and complement-dependent mechanisms of virus neutralization. This book has been written primarily for students and scientists who have not been specifically trained in complement research.

• Language: English
• Publisher: Academic Press
• Release date: Jun 28, 2014
• ISBN: 9781483276397

Book preview

1

Preface

The complement system was first recognized to be important in host defense against infection in 1894, and much of the seminal work in immunology by Ehrlich, Bordet, and Metchnikoff focused on the role of antibody and complement in the process of immunity to infection. However, because of the tremendous biochemical complexity of the complement system, its exact chemical nature and mechanism of action did not begin to be uncovered until the 1960s. As in many other areas of immunology, there was more research on complement in the 1970s than in the period from 1894 to 1969, so that by 1980, all of the 20 different plasma proteins of the complement system had been described. Yet with all of this research, new questions were raised almost as quickly as some of the older ones had been answered. In particular, complement was found to be involved in many different areas of host defense, and investigators in specialties other than immunology began investigating the possible involvement of complement in other biologic systems. Although there have been many comprehensive review series written about the complement system, none of these has been written for the uninitiated who is unfamiliar with the jargon and basic technology of complement research. This is an introductory book on complement that has been written primarily for students and scientists who have not been specifically trained in complement research. Chapters on the basic aspects of complement function are followed by chapters on the molecular genetics of complement and the role of complement in different diseases. The intention was to create a primer that would provide the basic knowledge of complement required for either research or clinical medicine in diseases involving the complement system.

Gordon D. Ross

Introduction and History of Complement Research

GORDON D. ROSS,     Division of Rheumatology and Immunology, Department of Medicine and Department of Microbiology and Immunology, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27514

I THE COMPLEMENT SYSTEM

The complement system is an important part of host defense against infection that functions together with the immune response to provide the effector mechanisms necessary to initiate inflammation, kill bacteria and other pathogens, and facilitate the clearance of bacteria and immune complexes. The complement system is made up of 20 distinct plasma proteins and 9 different membrane proteins. The presence of bacteria or immune complexes triggers activation of the complement system, resulting in a sequence of biochemical reactions in which one component activates another component in a cascade fashion. Along this cascade, such functions as inflammation and phagocytosis are initiated, and the terminal event is the generation of bactericidal activity in the form of membrane-penetrating lesions. Because of the importance of complement, an inherited or acquired deficiency in any one component of the system is frequently associated with either an increased susceptibility to infection or a lupuslike syndrome thought to result from diminished clearance of circulating immune complexes. The purpose of this book is to provide an introduction to the complement system that will be useful both to basic scientists and clinicians. Descriptions of biochemical reaction pathways are followed by chapters defining the involvement of complement in various diseases. The purpose of this introduction is to provide a historical background of complement research, and to introduce certain key investigators from the past who have made important contributions. For several reasons this history has been abbreviated and many investigators from approximately 1970 onward have not been mentioned specifically by name. First, there has been far more research reported on complement from 1970 until the present than from 1888 to 1970. Second, it would have been difficult to mention any of the recent investigators in complement research without excluding many others whose research was equally deserving of mention. Finally, the history of current complement research is covered to some extent in the individual chapters.

II HISTORY OF RESEARCH ON COMPLEMENT

IIA Early Events

A series of experiments just before the turn of the century led to the recognition of the existence of the complement system. First in 1888, George Nuttall (Fig. 1) found that normal sheep blood had a mild bactericidal activity for anthrax bacilli that was lost rapidly when blood was heated to 55°C or allowed to stand for a longer time at room temperature. In 1889, Buchner confirmed this finding and named the labile serum bactericidal factor alexin. Next, in 1894, Richard Pfeiffer demonstrated that blood from guinea pigs that had recovered from cholera infection would protect normal guinea pigs from cholera infection if injected in a mixture with the live bacteria. Because in vitro tests showed that the cholera were only killed by fresh immune serum and not by heat-inactivated immune serum, he was surprised to find that injections of the heat-inactivated immune serum would protect normal guinea pigs from infections. Jules Bordet (Fig. 2) is credited with the critical experiments that identified complement in 1894. Bordet demonstrated that the activity of heat-inactivated immune serum could be restored in vitro by the addition of small amounts of fresh normal serum that by itself had no bactericidal activity. The killing of the vibrio by serum was shown to be dependent upon both a heat-stable, -sensitizing-substance present in immune serum and a heat-labile cytotoxic factor present in normal (as well as immune) serum. Later, Bordet made similar observations with serum from guinea pigs immunized with defibrinated rabbit blood. He correctly attributed the heat-labile serum hemolytic factor to the same alexin bactericidal factor described earlier by Buchner. In 1899, Paul Ehrlich (Fig. 3) proposed a scheme of humoral immunity in which he adopted the terms amboceptor for the heat-stable immune sensitizer and complement for alexin. Ehrlich hypothesized that amboceptor (antibody) had two binding sites on opposite ends of the molecule: one that bound specifically to bacteria and a second site on its opposite end that bound to complement (Fig. 4). However, Ehrlich later incorrectly proposed that blood contained several different types of  complements that each had different functions.

Fig. 1 Dr. George H. F. Nuttall (1862–1937), Quick Professor of biology at Cambridge University from 1906–1937. This photograph from 1936 was supplied by Professor R. R. A. Coombs of Cambridge University.

Fig. 2 Dr. Jules Bordet (1870–1961) of the Pasteur Institute, Paris (1894–1901); later founder of the Pasteur Institute, Brussels; Nobel Laureate in Medicine and Physiology, 1919. This photograph was given by Dr. Bordet to Professor R. R. A. Coombs in 1948, and was kindly provided by Professor Coombs.

Fig. 3 Dr. Paul Ehrlich (1854–1915), professor of experimental therapeutics, Frankfort; Nobel Laureate with Elie Metchnikoff in Medicine and Physiology, 1908. Photograph courtesy of Professor R. R. A. Coombs.

Fig. 4 Diagram of amboceptor and complement as proposed by Paul Ehrlich in 1906. a, complement; b, interbody (immune body); c, receptor; d, part of a cell; e, toxophore group of the toxin; f, haptophore group.

IIB Discovery of the Classical Pathway

The model of antibody-mediated cytotoxicity proposed by Ehrlich called for the activation of complement by the specific attachment of antibody to bacteria. Complement became attached to the amboceptor bacteria complex and completed the killing reaction. This pathway of antibody-mediated activation of complement was the first to be defined and accordingly became known as the classical pathway when an antibody-independent alternative pathway was subsequently found.

In 1907 Ferrata demonstrated that complement consisted of more than one serum component. Dialysis of serum against water at acid pH produced a euglobulin precipitate and a water-soluble albumin fraction. Although neither the albumin fraction nor the redissolved euglobulin fraction had complement activity, mixture of the two fractions restored activity. Furthermore, because complement activity was better when the euglobulin fraction was added prior to the albumin fraction, Brand deduced in 1908 that the two fractions reacted sequentially and termed them midpiece and endpiece (the frontpiece was amboceptor or antibody). By the 1920s evidence had been presented for the existence of four different serum fractions containing the complement activity of serum. Cobra venom was found to destroy a component that was distinct from midpiece and endpiece, and later, ammonia and hydrazine were shown to destroy another component. In 1939 these were named C′3 (for third component) and C′4 (for fourth component) according to their order of discovery, and midpiece and endpiece were named C′1 and C′2, respectively. It should be pointed out, however, that midpiece and endpiece together contained all of the components of complement (including C′3 and C′4), and thus were not comparable to the individual well-characterized proteins now called C1 and C2. The first progress in separating these serum fractions into functionally distinct components was reported in 1941 by Louis Pillemer (Fig. 5); these components were shown to react in the sequence C′1, C′4, C′2, and C′3.

Fig. 5 Dr. Louis Pillemer (1908–1957). Photograph from J. Immunol. 125, 472 (1980).

Further characterization of the biochemistry of the complement system was not possible until the 1960s with the development of modern methods for protein separation and characterization. Progress was also retarded in the 1950s by the belief that serum contained such small amounts of complement molecules that biochemical characterization was impossible. Particularly noteworthy during this time was the development of the one-hit theory by Manfred Mayer (Fig. 6) that proposed that lysis of an erythrocyte by complement required only a single complement lesion. This allowed the development of hemolytic titrations for the individual complement components on a molecular basis, and led the way to the eventual purification and characterization of complement. These hemolytic assay techniques were precisely defined in a textbook by Elvin Kabat and Mayer entitled Experimental Immunochemistry, which became the manual for complement research in the 1960s and early 1970s.

Fig. 6 Dr. Manfred Mayer (1916–1984). Photograph from J. Immunol. 119, 1195 (1977).

A major impetus for the purification and characterization of the components of complement was the discovery in 1960 by Hans Müller-Eberhard (Fig. 7) that a major protein component of serum, ß1C globulin, was C′3. This provided the first evidence that complement was a major component of serum and thus might be purified and chemically characterized. Attempts to purify the individual complement proteins revealed that the C′3 fraction actually consisted of several different protein components. By use of multiple precipitation and column chromatographic steps, Robert Nelson (Fig. 8) and co-workers in 1965 separated the C′3 fraction of guinea pig serum into six functionally distinct components required to produce hemolysis of sheep erythrocytes in combination with C′1, C′4, and C′2. These were named C′3c, C′3b, C′3e, C′3f, C′3a, and C′3d. Later, Müller-Eberhard and colleagues at Scripps Clinic in La Jolla, California, completed separation of the C′3 fraction of human serum, and over a period of more than 20 years worked out much of the complex biochemistry of the complement cascade. Because these additional C′3 components represented distinct proteins that were actually unrelated to C′3, it was agreed at the Complement Workshop of 1966 that they should be named according to their reaction sequence in the complement cascade reaction: C′5, C′6, C′7, C′8, and C′9. However, the original names of the first four components were retained even though their numbers did not correspond to their order of reaction. Later, in 1968, the World Health Organization (WHO) nomenclature committee officially adopted this terminology and also decided to drop the prime symbol in the names of the components (i.e., C1, C4, C2, C3, C5, C6, C7, C8, and C9). Further adding to the complexity of the complement reaction sequence was the finding of several regulatory proteins and protease inhibitors that were required to control the rate and extent of activation of the classical pathway.

Fig. 7 Dr. Hans Müller-Eberhard (1927–). Dr. Müller-Eberhard is presently at the Scripps Clinic and Research Foundation, La Jolla, California.

Fig. 8 Dr. Robert A. Nelson, Jr. (?–1983). Photograph taken about 1966 while Dr. Nelson was an investigator of the Howard Hughes Medical Institute at the University of Miami. Photograph provided by Dr. Irma Gigli, University of California at San Diego.

IIC Properdin and the Alternative Pathway

Despite the demonstration of an absolute requirement for specific antibody to initiate complement hemolysis of sheep erythrocytes, there were also several early reports that suggested that complement could be activated independently of antibody by certain strains of bacteria and yeast. As early as 1913, Browning and Mackie showed that cobra venom would trigger complement-mediated lysis of erythrocytes, and performed experiments that suggested that this involved certain serum components that were distinct from those of the classical pathway. Although this may now be viewed as the first description of the alternative pathway, the discovery of this pathway is probably more appropriately attributed to Pillemer 40 years later. Studies reported by Pillemer, Lepow (Fig. 9), and co-workers in 1954 described a novel protein in serum, termed properdin, that was said to bind to bacteria and yeast and trigger complement activation in the absence of antibody. A scheme of complement activation was proposed that not only was independent of antibody but also involved other activating components than those of the classical pathway. These were termed factors A and B, and properdin, and this pathway for initiation of the lytic action of complement was termed the properdin pathway of complement activation. However, as the techniques of protein purification and characterization were in their infancy at this time, Pillemer was unable to convince other immunologists of his era of this existence of the properdin pathway. In particular, preparations of purified properdin were shown to contain small amounts of natural antibodies to yeast and bacteria, and sensitive assays showed that small amounts of C1, C4, and C2 were consumed in the properdin pathway. Pillemer was said to have been particularly depressed after a 1957 meeting of complement investigators was reluctant to accept his properdin system. He died shortly thereafter of a barbiturate overdose that was ruled a suicide. For a short time, Lepow continued this work as the new leader of Pillemer’s laboratory and demonstrated that properdin and factor B were serum proteins that were distinct from antibody or any of the classical pathway components.

Fig. 9 Dr. Irwin Lepow (1923–1984). Photograph from J. Immunol. 125, 471 (1980).

Ten years after Pillemer’s death, the alternative pathway was rediscovered and work was begun in several laboratories to characterize the various unique serum protein components of the system and their mechanism of action. Finally, in 1980, it was possible to demonstrate a functional alternative pathway with a mixture of each of the purified protein components in the absence of serum. During the time of characterization of the alternative pathway, several different names for the unique components had evolved. Later, however, the WHO nomenclature committee decided to name the components in the manner originated by Pillemer. Thus, the serum proteins of the alternative pathway are named factors B, D, H, and I, and properdin. The protein described as factor A by Pillemer turned out to be C3 of the classical pathway, and activation of native C3 was shown to be the point of intersection in the activation of the lytic pathway of complement by the two pathways (Fig. 10). Although the protein components of the alternative pathway and their mechanism of action have been characterized, some unresolved questions remain on the exact chemical nature of bacterial cell wall and membrane structures that permit activation of the alternative pathway.

Fig. 10 The classical and alternative pathways of complement activation.

IID Assembly of the Lytic Terminal Components

One of the primary functions of the complement system is the generation of membrane-penetrating lesions in target cells, leading to cell death. This mechanism of cytotoxicity is not effective against all types of bacteria, particularly bacteria with thick cell walls that complement is unable to penetrate. However, this function of complement was one of the earliest to be observed, and consequently much of the early research on the mechanism of complement action focused on the role of complement in cellular cytotoxicity.

IID1 The Hemolytic Assay of Complement

A very important early development was the use of the sheep erythrocyte (E) hemolytic essay. Lysis of sheep E requires anti-sheep E antibody, all components of the classical pathway, and the terminal complement components. Hemolysis of sheep E resulted in a change in turbidity that could be easily visualized and a release of hemoglobin that could be precisely quantitated spectrophotometrically. Furthermore, stable intermediate complexes could be generated by sequential addition of the individual components, allowing the order of the reaction of the components and the mechanism of their activity to be defined. Any of the individual components could be measured specifically by use of a pool of components that was deficient in the component of interest. For example, C2 could be specifically assayed with an intermediate complex consisting of antibody-sensitized sheep E (abbreviated EA, for erythrocyte–antibody complex) coated with purified C1 and C4 (abbreviated EAC14), and a pool of the components containing C3–C9. In the presence of C2, the intermediate complex EAC142 was generated, so that the subsequent addition of C3 to C9 resulted in hemolysis of the sheep E.

IID2 The Membrane Attack Complex

Stimulation of either the classical or alternative pathway results in the activation of C3 and the terminal components C5–C9. The mechanism by which the terminal components cause lysis of cells has been the subject of intense research for many years. Initially it was believed that complement might cause lysis or cytotoxicity by forming an enzyme that digested cell membranes. This came partly from the recognition of the enzymatic nature of complement activation. Others also theorized that complement might form a detergent that dissolved membranes. Major advances in the understanding of this process came from the study of the mechanism of terminal component complex formation with different artificial membranes known as liposomes. A key hypothesis by Mayer was that the terminal components formed a channel through the membrane of cells that allowed the leakage of salts and the entry of water. Humphrey and Dourmashkin also used electron microscopy to visualize complement channels in cell membranes and found that they were discrete uniform structures with the appearance of doughnuts. Ion channel formation by complement resulted in lysis of red cells, because the cells rapidly took up water in an attempt to balance the relatively high ionic strength of the cell cytoplasmic components that were too large to traverse the complement channel. This process is termed colloid osmotic lysis, and the complex formed by the terminal components is known as the membrane attack complex (MAC). Lysis of bacteria with rigid cell walls also occurs but without swelling.

IIE Regulation of the Inflammatory Response

The inflammatory response was early recognized to be an important event in host defense against infection. In 1910, Friedberger demonstrated that injection of immune precipitate-treated serum into guinea pigs produced a toxic and sometimes fatal reaction that resembled anaphylaxis. The responsible serum factor was thus named anaphylatoxin. Because heat inactivation of normal serum prevented generation of serum anaphylatoxin activity, Novy and deKruif concluded in 1917 that the toxic factor might be derived from complement. Other activators of complement were also shown to generate anaphylatoxin activity in serum. These included agar (Bordet, 1913), inulin (Bordet and Zunz, 1915), and starch (Nathan, 1913). The first direct pieces of evidence for anaphylatoxin coming from complement were reported by Osier and colleagues in 1959 and by Vogt and Schmidt in 1966. Jensen showed that an anaphylatoxin of guinea pig serum was derived from C5, and concurrently, Lepow and co-workers showed that a human serum anaphylatoxin came from C3. Cochrane and Müller-Eberhard then showed in 1968 that anaphylatoxin activity could be generated from either C3 or C5 by enzymatic digestion.

Anaphylatoxins work primarily by triggering histamine release from mast cells and basophils. This results in contraction of smooth muscle and all the classical signs of inflammation (redness, pain, and swelling). C5a triggered the same functions as did C3a, and in addition also stimulated the chemotaxis of neutrophils and macrophages. Thus, bacterial infections that trigger complement activation result in release of C3a, and C5a, which play a major role in stimulating the inflammatory response and attracting neutrophils and macrophages to the site of infection.

IIF Opsonization by Complement

IIF1 Immune Adherence and Enhancement of Phagocytosis

In 1904, Wright and Douglas observed that serum caused enhanced binding of bacteria to phagocytic cells, and in 1905, Levaditi and Inmann deduced from simple experiments that this probably involved antibody and complement. This attachment of microorganisms to leukocytes led to phagocytosis, and it was shown to be more efficient with fresh immune serum than with heat-inactivated immune serum. Treatment of bacteria with serum followed by washing of the bacteria was shown to render the bacteria readily ingestible by phagocytic cells. Thus it was hypothesized that the serum coated the bacteria with complement and that this allowed subsequent ingestion of the bacteria. The term opsonization (or opsonification) was coined by Wright; it means to prepare for ingestion. However, Metchnikoff (1905) incorrectly hypothesized that the opsonic complement in serum was derived from leukocytes that were damaged during blood coagulation. This theory of complement origin was not disproven until many years later when the development of anticoagulants allowed demonstration of complement in unclotted plasma.

The binding of red cells to serum-opsonized trypanosomes was first reported in 1930 by Duke and Wallace. They found that fresh immune serum caused trypanosomes to be coated with red cells from humans or monkeys but not with red cells from guinea pigs. With immune guinea pig blood they noted that the platelets rather than the red cells became bound to trypanosomes. They showed that red cell binding to trypanosomes required a heat-stable component from immune serum as well as a heat-labile component from normal guinea pig serum, and this led Wallace and Wormall to propose a requirement for complement in 1931. By 1938, Brown and Broom showed that treatment of serum with other known complement inactivators (i.e., ammonia, cobra venom) also prevented red cell binding to trypanosomes. Later, in 1953, Nelson described red cell binding to serum-opsonized bacteria, and termed the phenomenon immune adherence. In immune blood, bacteria were observed to bind preferentially to erythrocytes rather than to leukocytes. However, after prolonged incubation of bacteria in whole blood, red cell-bound bacteria were ingested by blood leukocytes. Considerably less leukocyte ingestion was observed in leukocyte–serum mixtures devoid of red cells. Accordingly, Nelson proposed that bacteria that were bound to erythrocytes might be more easily ingested by phagocytic cells than were individual uncomplexed bacteria. Later, in 1963, Linscott and Nelson demonstrated that immune adherence required complement activation only up to the C3 stage, and postulated that erythrocytes bore immune adherence receptors

specific for the fixed C3 on bacteria. By 1967 it had been demonstrated that the phagocytosis of antibody-coated sheep erythrocytes by guinea pig polymorphonuclear leukocytes also required only the first four components of complement (C1, C4, C2, and C3), and it was proposed that phagocytic cells bore immune adherence receptors that were analogous to those of erythrocytes. The earlier finding by Duke and Wallace of guinea pig platelet adherence was explained by studies reported by Peter Henson in 1969. Guinea pig platelets, but not guinea pig erythrocytes, were found to have receptors for particle-bound C3.

IIF2 Membrane Complement Receptors

At least nine distinct types of membrane complement receptors are now recognized on different cell types. Six of these complement receptors are specific for different parts of the C3 molecule. Whereas some of these are present on phagocytic cells and can be shown to enhance the binding and phagocytosis of bacteria and other microorganisms, the functions of the other types of receptors are less well defined. The erythrocyte immune adherence receptor has a major function in the clearance of soluble complexes of antigen and antibody (immune complexes) from the blood, and as Nelson had proposed, serves to transport and present complexes to phagocytic cells. The functions of the anaphylatoxins C3a and C5a are dependent on triggering two distinct types of membrane receptors present on mast cells and leukocytes. Although little is known of the functions of lymphocyte complement receptors, several different types of experiments have suggested that complement may be in some way involved in the immune