IgG subclass co-expression brings harmony to ... - Wiley Online Library

Immunology and Cell Biology (2016) 94, 949–954 & 2016 Australasian Society for Immunology Inc. All rights reserved 0818-9641/16 www.nature.com/icb

PERSPECTIVE

IgG subclass co-expression brings harmony to the quartet model of murine IgG function Andrew M Collins A model of murine IgG function is presented in which the co-expression of the IgG subclasses is a central feature, class switching occurs before the commencement of somatic hypermutation, and there is little switching between subclasses. It is named the quartet model to emphasize the harmony that comes from the simultaneous presence of the four subclasses. In this model, IgG3 and IgG2b antibodies are particularly important early in the response, when T-cell help may be limiting. IgG3 initiates inflammation through complement fixation, whereas IgG2b provides early FcγR-mediated effector functions. As T-cell help strengthens, IgG2a antibodies increase the power of the response, whereas IgG1 production helps limit the inflammatory drive and limits immunopathology. The model highlights the fact that murine IgG subclasses function quite differently to human IgG subclasses. This allows them to serve the special immunological needs of a species that is vulnerable because of its small size. Immunology and Cell Biology (2016) 94, 949–954; doi:10.1038/icb.2016.65; published online 9 August 2016

INTRODUCTION: THE PARADOX OF OPPOSING ANTIBODY FUNCTIONS The identification of four human1 and four murine IgG subclasses2,3 has encouraged a belief that there is a correspondence between the subclasses of the two species, though even early studies recognized that there may be none.3 Instead, the sharing of four IgG subclasses by humans and mice, and also by rats,4 appears to be a coincidence (Figure 1). Other mammalian species express widely varying numbers of IgG subclasses. There is only one IgG subclass in the rabbit,5 whereas there are six in the pig,6 seven in the horse7 and eight subclasses have been reported in the elephant.8 Nevertheless, comparisons between the human and murine subclasses remain important, for such comparisons could provide insights into the immunological functions of the different IgG subclasses of both species. There is one reported feature of the immune systems of mice and humans that may be confounding our understanding of humoral immunity. Models of antibody regulation emphasize the way cytokines direct the immune system towards the production of particular IgG subclasses,9,10 but the reality of the immune response is that the simultaneous production of multiple subclasses is routinely seen. In fact, it is a paradox of humoral immunity that this occurs despite some of the subclasses having essentially opposing functions. For example, the murine IgG subclasses include IgG2a and IgG2b that fix complement11 and bind to all activating FcγR12 (Table 1). The murine IgG1 subclass, on the other hand, does not fix complement, and only binds well to the inhibitory FcγRIIb.12 Its contribution to protective immunity is therefore unclear.

SOMATIC HYPERMUTATION AND THE HUMAN IGG SUBCLASSES We recently proposed a new model to explain how the human IgG subclasses may function together, despite their varying and even opposing properties.13 The model was developed in response to intriguing observations of somatic point mutations in immunoglobulin heavy chain IGHV genes. There are significant differences in the number of somatic point mutations that are seen in human IGHV genes when they are expressed in association with the different IgG constant region genes. In fact, the ranking of the mean number of mutations in IgG subclass-associated IGHV genes corresponds to the position of each constant region gene within the human immunoglobulin heavy chain gene locus.14–16 That is, IgG3-associated IGHV genes are the least mutated IgG-associated IGHV gene sequences, IgG1 and IgG2 sequences carry progressively more mutations, and IgG4-associated IGHV genes are the most mutated sequences of all. This observed ranking of mutation numbers is not easily explained if class switching is simply the outcome of T-cell commands. Rather, the data suggest that there is a general tendency for human B cells to switch sequentially between each of the IgG subclass-encoding IGHG genes (Figure 2). Although there is presently no direct evidence that class switching moves progressively through all four IGHG genes, there is evidence for switching from one IGHG gene to the next downstream IGHG gene. Remnants of IgG3 switch regions have been found in IgG1 þ B cells and remnants of IgG1 switch regions have been found in IgG2 þ B cells.17 If such sequential switching commonly involves all four IGHG genes, it would have important consequences. Such sequential switching would result in predictable differences in the relative antigen-binding affinities of the different IgG subclasses.

School of Biotechnology and Biomolecular Sciences, The University of New South Wales, Sydney, NSW, Australia Correspondence: Dr AM Collins, School of Biotechnology and Biomolecular Sciences, The University of New South Wales, Sydney, NSW 2052, Australia. Email: [email protected] Received 3 June 2016; revised 12 July 2016; accepted 14 July 2016; published online 9 August 2016

The quartet model of murine IgG function AM Collins 950

When previous reports of the properties of the human IgG subclasses were reviewed in the context of relatively predictable and ordered switching, and of the resulting relative affinities of the subclasses, it became clear that some of their opposing properties could be reconciled. This led to the development of the temporal model of human IgG function.13 The model was originally formulated to describe the process of class switching in response to a single encounter with antigen, but it may also describe the outcome of repeated encounters with antigen. Memory B cells include cells that have switched to each of the IgG subclasses.17 Over a lifetime, repeated encounters with a particular antigen may also move antigen-selected memory cells along the sequential pathway.18 SOMATIC HYPERMUTATION AND THE MURINE IGG SUBCLASSES We recently investigated somatic point mutations in mouse IGHV genes,19 seeking evidence of sequential class switching between the IgG-encoding constant region genes of the mouse. As class switching can only proceed from 5′ to 3′ through the heavy chain constant region gene locus, sequential class switching should lead cells carrying relatively unmutated IGHV genes in association with upstream constant region genes to give rise to cells expressing more mutated IGHV genes in association with downstream constant region genes. The order of the mouse IgG-encoding constant region genes on chromosome 12 (from 5′ to 3′) is IGHG3, IGHG1, IGHG2B and IGHG2A or IGHG2C.20 (The IgH-1b haplotype of BALB/c mice and mice of many other strains includes the IGHG2A gene but not the IGHG2C gene, whereas the IgH-1a haplotype of the C57BL/6, NOD and SJL strains includes the IGHG2C gene but not the IGHG2A gene.21) If sequential switching is common, IgG3-associated IGHV genes would on average be the least mutated sequences in very large

Figure 1 The genomic positions of IgG-, IgA- and IgE-encoding constant region genes of the human, mouse and rat: three species that each express four IgG subclasses. The evolutionary relationships between the rat and mouse gamma genes are indicated with shading. The most active complement fixing and FcγR-binding subclass of each species (*), and the subclass of each species that binds the inhibitory FcγR (**) are indicated. Presumed rat IgG subclass functions are based upon comparative studies of amino-acid sequences,4 and on basophil degranulation assays.69

sequence datasets. IgG2a- and IgG2c-associated IGHV genes would be the most mutated sequences. This was not seen in the mutation data. In our study, IGHV sequences associated with the IgG1-encoding IGHG1 gene were the most highly mutated sequences, while the 3′ gene-encoded IgG2a and IgG2c sequences were the least mutated sequences.19 Although inferences must be drawn with caution, the mutation data suggest not simply a lack of sequential switching from IgG1 to the downstream IgG2b subclass, but that IgG1+ B cells rarely switch to either of the downstream IgG subclasses. In fact, there may be little switching between any of the IgG subclasses. The mouse response may instead be dominated by direct switching of IgM+ B cells to each of the IgG subclasses. Another striking feature of the mutation data was that even IgG1-associated IGHV genes had relatively few mutations, and a high percentage of all murine IgG sequences were germline encoded. Class switch recombination and somatic hypermutation are often seen as processes that occur together, but these data confirm the observations of others that class switching in the mouse must often occur before the induction of somatic hypermutation.22,23 In our study, a staggering 44% of the IgG2c-associated IGHV genes were unmutated.19 The lowest average number of mutations was seen in IgG2c-associated IGHV genes (mean: 2.6 mutations), and even the most highly mutated IgG1-associated genes were only slightly more mutated (mean: 4.2 mutations). This level of mutation is not substantially different from previous reports of 6–10 mutations in selected cells of the germinal center reaction.22,23 These and other studies have described this as a high rate of mutation, but it is a relatively low rate compared with what is seen in human sequences. The mean number of mutations in unselected human sequences ranges from ~ 18 mutations in IgG3-associated IGHV genes, to ~ 27 mutations in IgG4-associated IGHV genes.14,15 Broadly neutralizing human anti-HIV antibodies have been reported with over 90 mutations in their heavy chain IGHV genes,24,25 and very few human IgG sequences are unmutated.

Figure 2 The temporal model of human IgG class function, highlighting sequential switching between the four human IgG subclasses.13 The affinities shown were inferred from the analysis of mutations in a large data set of IgG-associated IGHV gene sequences.

Table 1 Properties of the mouse IgG subclasses Complement fixation

Activating FcγR binding

IgG3c

++

−

−

−

+

IgG1

−

+

++

0.1

++

IgG2b IgG2a

++ ++

+++ +++

++ +

7 69

+ +

aThe

A/I ratioa

Relative affinityb

A/I ratio is calculated from the highest binding affinity of each subclass for an activating FcγR (A), and from the binding affinities of the subclasses for the inhibitory FcγRIIb (I).12 from the analysis of mutations in a large data set of heavy chain IGHV gene sequences.19 subclasses are listed in the order in which they are found in the genome, from 5′ (IgG3) to 3′.

bInferred cThe

Inhibitory FcγRIIb binding

Immunology and Cell Biology


The relative lack of somatic point mutations in mouse IGHV genes may be an inevitable consequence of the small size of the mouse. A mouse has a mass that is ~ 1/3000 that of a human, and the murine clonal burst size should therefore be correspondingly small. As somatic point mutations accumulate at the rate of approximately one mutation per cell division within the germinal center reaction,26 the extent of somatic point mutation should reflect the burst size. The capacity of the murine antibody response to undergo affinity maturation must be reduced, in comparison with a large species such as the human, as a consequence of the relative lack of somatic point mutations in the mouse. The specificities encoded by the germline genes may therefore be particularly important for immunological protection of the mouse. Indirect evidence of this comes from early reports that the primary murine anti-viral response includes highaffinity antibodies,27 and from observations that the germline-encoded T15 idiotype provides the mouse with high-affinity protection against pneumococcal infections.28 This is not to deny the capacity for mutations to deliver improved protection for the mouse. Even a single mutation can substantially improve the binding affinity of some murine antibodies.29 THE MOUSE IGG RESPONSE USUALLY INVOLVES ALL SUBCLASSES The relatively low affinity of the murine IgG response that is implied by the low levels of mutation seen in murine antibody genes means that a greater number of antibody forming cells will be required to achieve antigen clearance than is the case in species that produce higher-affinity antibodies. Given the small clonal burst size of the mouse, this must be achieved through the activation of a large population of naive B cells. Although it is possible that such a cell population could be narrowly focused on particular antigenic targets, it is more likely that a large population of low-affinity B cells will have a very broad range of fine specificities, and each cell within this B-cell population would have to compete for T-cell help. The complexity of the B-cell response may be matched by complexity in the T-cell response, but in the early phase of an immune response, T-cell help is likely to be limiting, and intermolecular T-cell help cannot be assured.30,31 Stochastic processes may determine whether individual B cells receive T-cell help, or whether they differentiate before engaging with T cells. This could help explain a puzzling feature of the murine IgG response. Much more so than is the case in the human, the typical mouse IgG response includes isotypes that have been associated with the T-independent response, as well as isotypes that have been associated with the T-dependent response. The production of all four IgG subclasses has been reported after animals have been challenged with widely varying antigens including DNA vaccines,32 recombinant Ascaris suum antigens,33 recombinant Plasmodium antigens34 and bacterial adhesins,35 as well as after infection with bacteria36 and a range of viruses.37–39 The four subclasses have also been reported in response to T-independent antigens. The T-independent response leads to the production of IgG3 and IgG2b antibodies,40 but the earliest studies of the anti-carbohydrate response show that it typically involves all the IgG subclasses.41,42 The proportions of the different subclasses will certainly vary as a result of the prevailing cytokine environment. Switching to IgG2b is promoted by transforming growth factor-β.40 The production of IgG1 increases in response to the Th2 cytokine interleukin-4, and the production of IgG2a increases in response to the Th1 cytokine interferon-γ.9,10 In reality, cytokine production is usually less polarized

than the classic Th1/Th2 model suggests,43 and in the presence of interleukin-4 and interferon-γ, both IgG1 and IgG2a are produced.44 The cytokine milieu in which a response develops will depend upon the migration and interactions of many cell populations within the organized lymphoid tissues. Class switching has certainly been observed early during T-dependent responses in the germinal center reaction,45,46 but class switching is not confined to follicular B cells. Switching to IgG1 has been reported as part of the extrafollicular T-independent response of marginal zone B cells.47 Switching to IgG3 has been seen in extrafollicular B-cell blasts48 and switching to IgG2c is seen in the complex extrafollicular response to Salmonella infection.49 In different circumstances, the balance of these isotypes may vary, but no matter where and how the mouse antibody response develops, the isotypes of these disparate responses must still work together in the same way. THE QUARTET MODEL OF MURINE IGG FUNCTION Discussions of murine IgG subclass function usually emphasize the properties of each subclass, as if they act on their own. However, when considered in isolation, the properties of murine IgG1 and IgG3 antibodies are quite mysterious. IgG1 antibodies are often reported to be a key antibody isotype of the Th2 response. These antibodies can neutralize toxins and viruses through steric hindrance, but they are unable to fix complement or to trigger FcγR-mediated effector functions (Table 1). The production of IgG1 can therefore achieve relatively little if such antibodies are produced on their own. IgG3 antibodies have long been considered to be a key isotype produced in response to carbohydrates and other T-independent antigens.50 Complement fixation by IgG3-containing immune complexes may result in the recruitment of inflammatory cells, but these cells would be unable to engage in FcγR-mediated effector functions, for IgG3 antibodies do not bind to FcγR. IgG3 antibodies directed against pathogen-associated carbohydrate targets would therefore be ineffective if murine IgG3 was produced on its own. On the other hand, if the four murine isotypes are co-expressed, much of the confusion surrounding murine IgG1 and IgG3 function is resolved. The co-expression of all four IgG subclasses is a central feature of a new model of mouse IgG function that is presented as Figure 3. The quartet model emphasizes direct class switching of IgM þ B cells to each of the four IgG subclasses, early in a response, and before the commencement of somatic point mutations. It also highlights a general lack of class switching between the IgG subclasses. It is named the quartet model to emphasize the harmony that results from the co-expression of the subclasses. If there is little class switching between the IgG subclasses, all the IgG þ cells of a murine clone are likely to express a single IgG subclass, whereas different clones are likely to express different subclasses. The fine specificities of the antibodies of the different subclasses may therefore not be shared. This is different to the situation in the human. High-throughput sequencing of human antibody gene sequences has clearly shown that clones of human B cells routinely include multiple IgG isotypes.51 Human antibodies of different isotypes will therefore compete for antigen binding. In the mouse, however, there should be less competition between isotypes. In the quartet model, murine IgG3 antibodies ensure that IgG-mediated effector functions begin to operate early in a response, even if T-cell help is lacking. IgG3 antibodies fix complement well,11 triggering a cascade of inflammatory events (Table 1). Although IgM antibodies also serve this role, the tendency of IgG3 antibodies to self-aggregate may give these antibodies additional powers.52 Immunology and Cell Biology


Figure 3 The quartet model of murine IgG function, highlighting the direct switching of IgM+ B cells to each of the four murine IgG subclasses. Co-expression of all four subclasses is an essential feature of the model, with the relative proportions of the different isotypes being influenced by the cytokine milieu.

Multivalent binding of pentameric IgM requires antigenic determinants to be appropriately spaced on, for example, the surface of a bacterial invader.53 As a consequence of their ability to self-aggregate, bivalent murine IgG3 antibodies may be able to engage in multivalent binding of high functional affinity to antigenic determinants that are spaced more variably on the surface of a pathogen. In some respects, this parallels the situation seen in the human. The flexibility that results from the long hinge of human IgG3 seems to allow these early human antibodies to bind antigen with high functional affinity.54 The IgG2a, IgG2b and IgG2c subclasses all have similar functions. IgG2a and IgG2c are generally assumed to be identical in function, but as IgG2c function has not been formally characterized, its functions are not considered further here. IgG2b antibodies are usually considered to be part of the T-independent response.40 In the quartet model, IgG2b antibodies work in partnership with IgG3 antibodies. Like IgG3 antibodies, they are conspicuous in the early response, when T-cell help may be limiting. Both these subclasses arise from IgM+ cells in a divisionlinked manner, and the probabilities of switching to each of the isotypes are independent.40 Cells will therefore usually switch directly to IgG3 or IgG2b, but will sometimes switch indirectly to IgG2b via IgG3. Early switching to IgG2b should ensure early recruitment of FcγR-mediated effector functions, for IgG2b binds with moderate affinity to all the activating FcγRs.55 The IgG2a response was linked to viral infection in early reports,56 and subsequently its production was shown to be upregulated by interferon-γ.57 IgG2a has the same general properties as IgG2b, and such redundancy of function might be expected from a gene duplication event such as the one that gave rise to IgG2a and IgG2b. In the quartet model, the generation of IgG2a antibodies strengthens antigen clearance, for IgG2a is associated with much stronger FcγR-mediated activity.12 IgG2a is therefore well suited to drive viral clearance through the mechanism of antibody-dependent cellular cytotoxicity.58,59 IgG1 has properties that are distinct from those of the other murine subclasses, though strangely it is sometimes chosen for study as if it is Immunology and Cell Biology

the prototypical IgG subclass.22 Mouse IgG1 does not fix complement and only engages with the inhibitory FcγRIIb (Table 1). In the quartet model, the function of murine IgG1 is to limit IgG-driven inflammatory processes through engagement with FcγRIIb. It therefore has functions that are similar to those of human IgG4, as recently highlighted by others.60 Class switching to murine IgG1 and to human IgG4 is promoted by interleukin-4, and further downstream switching leads from mouse IgG1 and human IgG4 to IgE.61,62 This linking of anti-inflammatory IgG isotypes with the IgE response in both the mouse and the human is curious. It may be that some downregulation of IgG activity is necessary for the full engagement of IgE-mediated defences. In the human, IgG4 production is strongly associated with persisting antigen.63 In the quartet model, IgG1 production is seen much earlier in a response. This early pathway to IgE production and IgE-mediated effector function is in keeping with the general thesis that a small and vulnerable species such as the mouse must rapidly marshal its defences. We have argued that the high affinity of human IgG4 antibodies is critical to their ability to limit and even to switch off IgG-mediated inflammatory processes.13 Although mouse IgG1 antibodies are relatively unmutated in comparison with human antibodies, they do carry the highest mean number of mutations of the murine IgG subclasses.19 It is therefore likely that IgG1 antibodies have the highest mean antigen-binding affinity, and this has been reported in the past.64 The high affinity of the IgG1+ B-cell population may also be boosted by class switching to IgG1 that results directly from highaffinity antigen binding.65 This affinity-based cytokine-independent recruitment of cells into the IgG1 compartment could ensure that suitable IgG1 antibodies are always present and able to mitigate some of the dangers associated with IgG3 antibodies. Murine IgG3 antibodies can cause renal disease by precipitating in the glomerular capillaries, as a result of their tendency to self-aggregate.66 IgG1 antibodies offer protection against such IgG3-mediated immune complex immunopathology.60 Strait and colleagues hypothesize that this may be a result of the inflexibility of IgG1 antibodies, leading to monovalent rather than bivalent binding. In combination with the higher affinity of IgG1 antibodies, this should increase the likelihood of steric hindrance to IgG3 binding.60 CONCLUDING REMARKS Mammals share a common immunological inheritance. The mammalian inheritance of the mouse includes an antibody gene locus with a tendency for gene duplication. Four murine IgG subclasses have resulted from this, but the ways in which the subclasses work together may be unique to the mouse. Based in part on an analysis of IgG-associated IGHV gene mutations, a model of IgG function has been developed that features the simultaneous production of all four IgG subclasses. Switching to IgG occurs early in the response, before the acquisition of somatic point mutations, and there is very little switching between the different IgG subclasses. Such a system would be in keeping with the biological needs of the species. A mouse lacks large metabolic reserves and is vulnerable to dehydration.67 It experiences sickness without the social support that can sustain humans through prolonged illness. A sick mouse will soon be a dead mouse, unless the adaptive immune system responds quickly and effectively. IgG-mediated defences must therefore be rapidly engaged. The early recruitment of an IgG response of adequate affinity may be at least as important to a mouse as the crafting of a superior response over time, through the process of somatic hypermutation. The prevention of useless competition


between the different IgG subclasses may also be important, as this too would tax the resources of the mouse. In the quartet model, the IgG subclasses work in harmony, rather than in competition. The nature of the murine IgG response may therefore be better understood if the actions of antibodies of all four subclasses are considered together. The mutational evidence that led to the quartet model was obtained from studies of young, unimmunized mice, living in the clean conditions of an animal holding facility. It will be important for future work to analyze patterns of mutation in different strains of mice and in wild mice, of different ages, and faced with a more natural antigenic burden. The clonal burst size of mice will need to be confirmed, using techniques such as assays of kappa-deleting recombination excision circles.68 It will also be important to re-explore events within the organized lymphoid tissues, to determine whether the movement of cells and their phenotypic development can be linked with particular isotypes, or whether these processes are quite independent of class switching. Whatever the outcomes of these studies, it remains clear and incontrovertible that the IgG subclasses of the mouse work together in ways that are distinct from those of the human IgG subclasses. CONFLICT OF INTEREST The author declare no conflict of interest.

1 Ballieux RE, Bernier GM, Tominaga K, Putnam FW. Gamma globulin antigenic types defined by heavy chain determinants. Science 1964; 145: 168–170. 2 Fahey JL, Wunderlich J, Mishell R. The immunoglobulins of mice. Ii. Two subclasses of mouse 7s gamma-2-globulins: gamma-2a- and gamma-2b-globulins. J Exp Med 1964; 120: 243–251. 3 Grey HM, Hirst JW, Cohn M. A new mouse immunoglobulin: IgG3. J Exp Med 1971; 133: 289–304. 4 Bruggemann M. Evolution of the rat immunoglobulin gamma heavy-chain gene family. Gene 1988; 74: 473–482. 5 Mage RG, Lanning D, Knight KL. B cell and antibody repertoire development in rabbits: the requirement of gut-associated lymphoid tissues. Dev Comp Immunol 2006; 30: 137–153. 6 Butler JE, Wertz N. Antibody repertoire development in fetal and neonatal piglets. XVII. IgG subclass transcription revisited with emphasis on new IgG3. J Immunol 2006; 177: 5480–5489. 7 Lewis MJ, Wagner B, Woof JM. The different effector function capabilities of the seven equine IgG subclasses have implications for vaccine strategies. Mol Immunol 2008; 45: 818–827. 8 Guo Y, Bao Y, Wang H, Hu X, Zhao Z, Li N et al. A preliminary analysis of the immunoglobulin genes in the African elephant (Loxodonta africana). PLoS ONE 2011; 6: e16889. 9 Snapper CM, Mond JJ. Towards a comprehensive view of immunoglobulin class switching. Immunol Today 1993; 14: 15–17. 10 Mosmann TR, Coffman RL. TH1 and TH2 cells: different patterns of lymphokine secretion lead to different functional properties. Ann Rev Immunol 1989; 7: 145–173. 11 Neuberger MS, Rajewsky K. Activation of mouse complement by monoclonal mouse antibodies. Eur J Immunol 1981; 11: 1012–1016. 12 Nimmerjahn F, Ravetch JV. Divergent immunoglobulin g subclass activity through selective Fc receptor binding. Science 2005; 310: 1510–1512. 13 Collins AM, Jackson KJ. A temporal model of human IgE and IgG antibody function. Front Immunol 2013; 4: 235. 14 Jackson KJ, Wang Y, Collins AM. Human immunoglobulin classes and subclasses show variability in VDJ gene mutation levels. Immunol Cell Biol 2014; 92: 729–733. 15 van Zelm MC, Bartol SJ, Driessen GJ, Mascart F, Reisli I, Franco JL et al. Human CD19 and CD40L deficiencies impair antibody selection and differentially affect somatic hypermutation. J Allergy Clin Immunol 2014; 134: 135–144. 16 Wang Y, Jackson KJ, Chen Z, Gaeta BA, Siba PM, Pomat W et al. IgE sequences in individuals living in an area of endemic parasitism show little mutational evidence of antigen selection. Scand J Immunol 2011; 73: 496–504. 17 Berkowska MA, Driessen GJ, Bikos V, Grosserichter-Wagener C, Stamatopoulos K, Cerutti A et al. Human memory B cells originate from three distinct germinal center-dependent and -independent maturation pathways. Blood 2011; 118: 2150–2158. 18 van Zelm MC. B cells take their time: sequential IgG class switching over the course of an immune response? Immunol Cell Biol 2014; 92: 645–646.

19 Collins AM, Wang Y, Roskin KM, Marquis CP, Jackson KJ. The mouse antibody heavy chain repertoire is germline-focused and highly variable between inbred strains. Phil Trans R Soc B 2015; 370: 1676. 20 Honjo T, Kataoka T. Organization of immunoglobulin heavy chain genes and allelic deletion model. Proc Nat Acad Sci USA 1978; 75: 2140–2144. 21 Morgado MG, Cam P, Gris-Liebe C, Cazenave PA, Jouvin-Marche E. Further evidence that BALB/c and C57BL/6gamma 2a genes originate from two distinct isotypes. EMBO J 1989; 8: 3245–3251. 22 Hara Y, Tashiro Y, Murakami A, Nishimura M, Shimizu T, Kubo M et al. High affinity IgM(+) memory B cells are generated through a germinal center-dependent pathway. Mol Immunol 2015; 68: 617–627. 23 Jacob J, Kelsoe G, Rajewsky K, Weiss U. Intraclonal generation of antibody mutants in germinal centres. Nature 1991; 354: 389–392. 24 Wu X, Yang ZY, Li Y, Hogerkorp CM, Schief WR, Seaman MS et al. Rational design of envelope identifies broadly neutralizing human monoclonal antibodies to HIV-1. Science 2010; 329: 856–861. 25 Wu X, Zhou T, Zhu J, Zhang B, Georgiev I, Wang C et al. Focused evolution of HIV-1 neutralizing antibodies revealed by structures and deep sequencing. Science 2011; 333: 1593–1602. 26 Victora GD, Nussenzweig MC. Germinal centers. Annu Rev Immunol 2012; 30: 429–457. 27 Roost HP, Bachmann MF, Haag A, Kalinke U, Pliska V, Hengartner H et al. Early high-affinity neutralizing anti-viral IgG responses without further overall improvements of affinity. Proc Nat Acad Sci USA 1995; 92: 1257–1261. 28 Briles DE, Forman C, Hudak S, Claflin JL. Anti-phosphorylcholine antibodies of the T15 idiotype are optimally protective against Streptococcus pneumoniae. J Exp Med 1982; 156: 1177–1185. 29 Allen D, Simon T, Sablitzky F, Rajewsky K, Cumano A. Antibody engineering for the analysis of affinity maturation of an anti-hapten response. EMBO J 1988; 7: 1995–2001. 30 Alam S, Knowlden ZA, Sangster MY, Sant AJ. CD4 T cell help is limiting and selective during the primary B cell response to influenza virus infection. J Virol 2014; 88: 314–324. 31 Sette A, Moutaftsi M, Moyron-Quiroz J, McCausland MM, Davies DH, Johnston RJ et al. Selective CD4+ T cell help for antibody responses to a large viral pathogen: deterministic linkage of specificities. Immunity 2008; 28: 847–858. 32 Kurupati P, Ramachandran NP, Poh CL. Protective efficacy of DNA vaccines encoding outer membrane protein A and OmpK36 of Klebsiella pneumoniae in mice. Clin vaccine immunol 2011; 18: 82–88. 33 Islam MK, Miyoshi T, Tsuji N. Vaccination with recombinant Ascaris suum 24-kilodalton antigen induces a Th1/Th2-mixed type immune response and confers high levels of protection against challenged Ascaris suum lung-stage infection in BALB/c mice. Int J Parasitol 2005; 35: 1023–1030. 34 Cheong FW, Fong MY, Lau YL, Mahmud R. Immunogenicity of bacterial-expressed recombinant Plasmodium knowlesi merozoite surface protein-142 (MSP-142). Malaria J 2013; 12: 454. 35 Esgleas M, Dominguez-Punaro Mde L, Li Y, Harel J, Dubreuil JD, Gottschalk M. Immunization with SsEno fails to protect mice against challenge with Streptococcus suis serotype 2. FEMS Microbiol Lett 2009; 294: 82–88. 36 Byrd W, Cassels FJ. Long-term systemic and mucosal antibody responses measured in BALB/c mice following intranasal challenge with viable enterotoxigenic Escherichia coli. FEMS Immunol Med Microbiol 2006; 46: 262–268. 37 Allen SJ, Mott KR, Ghiasi H. Involvement of STAT4 in IgG subtype switching and ocular HSV-1 replication in mice. Mol Vision 2010; 16: 98–104. 38 Honda-Okubo Y, Ong CH, Petrovsky N. Advax delta inulin adjuvant overcomes immune immaturity in neonatal mice thereby allowing single-dose influenza vaccine protection. Vaccine 2015; 33: 4892–4900. 39 Slifka MK, Ahmed R. Long-term antibody production is sustained by antibody-secreting cells in the bone marrow following acute viral infection. Ann NY Acad Sci 1996; 797: 166–176. 40 Deenick EK, Hasbold J, Hodgkin PD. Switching to IgG3, IgG2b, and IgA is division linked and independent, revealing a stochastic framework for describing differentiation. J Immunol 1999; 163: 4707–4714. 41 Sarvas HO, Aaltonen LM, Peterfy F, Seppala IJ, Makela O. IgG subclass distributions in anti-hapten and anti-polysaccharide antibodies induced by haptenated polysaccharides. Eur J Immunol 1983; 13: 409–414. 42 Moreno C, Esdaile J. Immunoglobulin isotype in the murine response to polysaccharide antigens. Eur J Immunol 1983; 13: 262–264. 43 Kelso A, Groves P, Ramm L, Doyle AG. Single-cell analysis by RT-PCR reveals differential expression of multiple type 1 and 2 cytokine genes among cells within polarized CD4+ T cell populations. Int Immunol 1999; 11: 617–621. 44 Hasbold J, Hong JS, Kehry MR, Hodgkin PD. Integrating signals from IFN-gamma and IL-4 by B cells: positive and negative effects on CD40 ligand-induced proliferation, survival, and division-linked isotype switching to IgG1, IgE, and IgG2a. J Immunol 1999; 163: 4175–4181. 45 De Silva NS, Klein U. Dynamics of B cells in germinal centres. Nat Rev Immunol 2015; 15: 137–148. 46 Pape KA, Kouskoff V, Nemazee D, Tang HL, Cyster JG, Tze LE et al. Visualization of the genesis and fate of isotype-switched B cells during a primary immune response. J Exp Med 2003; 197: 1677–1687. 47 Kretschmer B, Weber J, Hutloff A, Fleischer B, Breloer M, Osterloh A. Anti-CD83 promotes IgG1 isotype switch in marginal zone B cells in response to TI-2 antigen. Immunobiol 2015; 220: 964–975.


The quartet model of murine IgG function AM Collins 954 48 Marshall JL, Zhang Y, Pallan L, Hsu MC, Khan M, Cunningham AF et al. Early B blasts acquire a capacity for Ig class switch recombination that is lost as they become plasmablasts. Eur J Immunol 2011; 41: 3506–3512. 49 Di Niro R, Lee SJ, Vander Heiden JA, Elsner RA, Trivedi N, Bannock JM et al. Salmonella infection drives promiscuous B cell activation followed by extrafollicular affinity maturation. Immunity 2015; 43: 120–131. 50 Perlmutter RM, Hansburg D, Briles DE, Nicolotti RA, Davie JM. Subclass restriction of murine anti-carbohydrate antibodies. J Immunol 1978; 121: 566–572. 51 Looney TJ, Lee JY, Roskin KM, Hoh RA, King J, Glanville J et al. Human B-cell isotype switching origins of IgE. J Allergy Clin Immunol 2016; 137: 579–586. 52 Greenspan NS, Cooper LJ. Intermolecular cooperativity: a clue to why mice have IgG3? Immunol Today 1992; 13: 164–168. 53 Tobita T, Oda M, Azuma T. Segmental flexibility and avidity of IgM in the interaction of polyvalent antigens. Mol Immunol 2004; 40: 803–811. 54 Roux KH, Strelets L, Michaelsen TE. Flexibility of human IgG subclasses. J Immunol 1997; 159: 3372–3382. 55 Bruhns P, Jonsson F. Mouse and human FcR effector functions. Immunol Rev 2015; 268: 25–51. 56 Coutelier JP, van der Logt JT, Heessen FW, Warnier G, Van Snick J. IgG2a restriction of murine antibodies elicited by viral infections. J Exp Med 1987; 165: 64–69. 57 Finkelman FD, Katona IM, Mosmann TR, Coffman RL. IFN-gamma regulates the isotypes of Ig secreted during in vivo humoral immune responses. J Immunol 1988; 140: 1022–1027. 58 Fang Y, Banner D, Kelvin AA, Huang SS, Paige CJ, Corfe SA et al. Seasonal H1N1 influenza virus infection induces cross-protective pandemic H1N1 virus immunity through a CD8-independent, B cell-dependent mechanism. J Virol 2012; 86: 2229–2238.


59 Jegaskanda S, Weinfurter JT, Friedrich TC, Kent SJ. Antibody-dependent cellular cytotoxicity is associated with control of pandemic H1N1 influenza virus infection of macaques. J Virol 2013; 87: 5512–5522. 60 Strait RT, Posgai MT, Mahler A, Barasa N, Jacob CO, Kohl J et al. IgG1 protects against renal disease in a mouse model of cryoglobulinaemia. Nature 2015; 517: 501–504. 61 Erazo A, Kutchukhidze N, Leung M, Christ AP, Urban JF Jr, Curotto de Lafaille MA et al. Unique maturation program of the IgE response in vivo. Immunity 2007; 26: 191–203. 62 Jabara HH, Loh R, Ramesh N, Vercelli D, Geha RS. Sequential switching from mu to epsilon via gamma 4 in human B cells stimulated with IL-4 and hydrocortisone. J Immunol 1993; 151: 4528–4533. 63 Aalberse RC, Stapel SO, Schuurman J, Rispens T. Immunoglobulin G4: an odd antibody. Clin Exp Allergy 2009; 39: 469–477. 64 Sarvas HO, Seppala IJ, Tahtinen T, Peterfy F, Makela O. Mouse IgG antibodies have subclass associated affinity differences. Mol Immunol 1983; 20: 239–246. 65 Turner ML, Corcoran LM, Brink R, Hodgkin PD. High-affinity B cell receptor ligation by cognate antigen induces cytokine-independent isotype switching. J Immunol 2010; 184: 6592–6599. 66 Abdelmoula M, Spertini F, Shibata T, Gyotoku Y, Luzuy S, Lambert PH et al. IgG3 is the major source of cryoglobulins in mice. J Immunol 1989; 143: 526–532. 67 Schmidt-Nielsen K. Scaling. Why is Animal Size so important? Cambridge University Press: New York, 1984. 68 van Zelm MC, Szczepanski T, van der Burg M, van Dongen JJ. Replication history of B lymphocytes reveals homeostatic proliferation and extensive antigen-induced B cell expansion. J Exp Med 2007; 204: 645–655. 69 Philips JR, Brouwer W, Edwards M, Mahler S, Ruhno J, Collins AM. The effectiveness of different rat IgG subclasses as IgE-blocking antibodies in the rat basophil leukaemia cell model. Immunol Cell Biol 1999; 77: 121–126.