| The anaphylaxis as immediate hypersensitivity is the most common allergic reactions in clinical settings, and it has been defined as type I hypersensitivity in classical immunology theory. The pathogenesis of anaphylaxis can be divided into two stages that consist of the sensitization and the effect. The sensitization stage represents the allergen stimulation to immune system of host to produce antibodies. The effect stage stands for the pathological symptoms induced by exposure to the same allergen again, in which the biological mediator such as histamine, leukotrienes, platelet activating factor (PAF) and prostaglandins release and can cause a systemic or local reactions including increased vascular permeability, reduced blood pressure, mucus secretion, smooth muscle contraction and tissue damage. Based on whether the priming process is needed, two types of animal models, the active and the passive allergic model, have been widely adopted. In active anaphylaxis model, antibody against allergen is produced in host; while in passive model, the prepared antibody should be given into host before antigen challenge. For passive anaphylaxis model, moreover, the priming process or immunization with allergen is unnecessary, the quanlity and quantity of antibody are easy to design or monitor, which make the experiment process more simple to operate, and also can shorten the experimental time.At present, more and more therapeutic monoclonal antibodies (mAbs) have been used or to be applied in clinical. More than30therapeutic mAbs have been approved by FDA, and more than300mAbs are in clinical trials. However, it was found that allergic reactions are the main adverse effects for almost all therapeutic mAbs, which makes the security as important consideration both in screening and adminatration of therapeutic mAbs. In classical immunology theory, the antibodies responsible for most allergic reactions are IgE. However, in recent years, a large number of studies have revealed that IgG antibodies could also mediate anaphylaxis, which implies that apart from IgE-mediated classical pathway of anaphylaxis, an IgG mediated alternative pathway independent of IgE antibodies also exists. In fact, all therapeutic mAbs are IgG, and more and more of the therapeutic mAbs are humanize or human antibodies, which can induce IgG-mediated allergic reactions, including the most serious and life-threatening anaphylactic shock. In experimental settings, researchers could take advantage of the model of passive systemic anaphylaxis (PSA) to simulate and study anaphylactic shock that is induced by infusion of mAbs.By now, the factors to trigger IgG pathway remains unclear. It is of notable that only a fraction of IgG antibody-antigen complexes or IgG aggregates, not all, can trigger PSA. However, nobody notice or study what kinds of IgG ICs can trigger PSA, and how to estimate the ICs being capable of initiating PSA. Furthermore, most of the studies use single antigen-antibody complex to trigger the PSA, in which haptens such as DNP (dinitrophenyl), TNP (trinitrobenzene) and penicillin are used rather than native antigens that are more important in clinical, and many studies also mainly focus on the effector mechanisms of anaphylaxis rather than the trigger or initiation. Our study uses36ICs containing eight natural and recombinant antigens and27corresponding IgG mAbs to trigger PSA, and five experiments were used to test whether an IC can mediate PSA or not.It is accepted that IgG ICs can bind with FcyRIII and FcyRIV on effector cells by cross-linking in vivo to promote the release of PAF by effector cells, causing increased vascular permeability and decreased blood pressure. However, there are still controversies for effector cells of IgG pathway. The different considerations on effector cell types mainly include macrophages, basophils, monocytes and neutrophils. In this study, we would use antigen-antibody complexes containing four natural and recombinant antigen and antibodies of IgG1, IgG2a and IgG2b to set up PSA model, and adopt both BalB/c and C57BL/6mice for in vivo observation of the role of various cells in IgG pathway.In the investigation of the effector cells responsible for IgG pathway, we accidentally discovered a soluble form of lymphocyte antigen6G (soluble Ly6G, sLy6G), which has never been reported before. Moreover, a sandwich ELISA system has been developed for sLy6G detection, and a preliminary study of its function has been conducted. It is also worth pointing out that the anti-granulocyte-differentiation antigen (Grl) antibody recognizes a common Grl epitope of lymphocyte antigen6C (Ly6C) and Ly6G. To be specific, Ly6C, MW14-17kDa, is mainly expressed on monocytes and macrophages, while Ly6G, MW25kDa, is mainly expressed on neutrophils.Part I. Only special IgG immune complexes could induce PSA-the definition of Anaphylaxis-inducing immune complexes (Ai-ICs)In previous work,10IgG mAbs against ovalbumin(OVA) were generated, and three of them (2C2,5G12,6H4) reacting with OVA could induce PSA. The futher identification by Western Blot and ELISA revealed that mAb2C2,5G12,6H4,1A6and5B8are actually antibodies reacting with ovomucoid (OVM), an ingredient mixed in OVA. By sensitization in vivo with100μg2C2, merely5μg OVM challenge could induce obvious PSA symptom. Accompanied by plasma exudation in the ears, challenge by ICs prepared in vitro show similar effects compared to traditional PSA protocol in which antigen challenge is preformed2h after antibody priming.In the model with mAb-OVM, only3of the7mAbs could induce PSA alone, while the other4mAbs (1A6,5B8, No.7, No.9) could not. Thereafter, by using combinations or mAb cocktail of these4mAbs incapable of inducing PSA alone, we found that3(No.7+1A6, No.7+5B8, No.7+No.9) of the6combinations could induce PSA with OVM, while the other3could not (No.9+1A6, No.9+5B8,1A6+5B8). Then, by using36IgG antibody-antigen complexes consisted of8natural and recombinant protein antigen with27corresponding mAbs, we found that14of the36IgG ICs could trigger PSA while the remaining22ICs could not.In this part, we successfully established PSA model using14IgG antibody-antigen complexes, and confirmed that only a fraction of these complexes could trigger PSA (14/36). Therefore, we defined IgG ICs capable of inducing PSA as Ai-ICs, in order to distinguish from the non-anaphylaxis-inducing immune complexes (nAi-ICs), which are incapable of triggering PSA.Part II. Characterization of Ai-ICsThere were four in vitro and in vivo experiments adopted for the characterization of Ai-ICs, which would be expected to provide a new safety evaluation for development and adminastration of therapeutic monoclonal antibodies.(1) Sandwich ELISA:all anti-OVM mAbs or mAbs cocktails being capable of forming Ai-ICs can be used to establish sandwich ELISA in vitro with OVM. The verification using36IgG antibody-antigen complexes showed that all Ai-ICs could form sandwich ELISA in vitro (14/36), and all the ICs incapable of forming ELISA (17/36) could not induce PSA. Notablely, there were also some ICs capable of forming ELISA, but being incapable of inducing PSA (5/36). (2) Native-PAGE:the ICs containing mAbs with OVM, HSA, AOPP and HbA2were subjected to Native-PAGE analysis, the results showed that all Ai-ICs manifested as a high band with low charge-mass ratio. However, some nAi-ICs capable of forming sandwich ELISA also manifested as high band, while nAi-ICs incapable of forming sandwich ELISA did not.(3) Passive cutaneous anaphylaxis (PCA):plasma exudation in the paw was observed after2C2intracutaneous priming followed by OVM i.v. challenge or direct intracutaneous challenge with preformed2C2-OVM ICs. The verification using36IgG antibody-antigen complexes showed that all Ai-ICs could induce PCA (14/36), while nAi-ICs could not (22/36).(4) Induction of neutrophil CD16/32down-regulation:we observed considerable down-regulation of blood neutrophil CD16/32expression after2C2-OVM IC stimulation in vivo and in vitro. Further investigations confirmed that2C2-OVM IC could also induce decreased CD16and CD32expression on human blood neutrophils. The verification using36IgG antibody-antigen complexes showed that all Ai-ICs could induce down-regulation of mouse blood neutrophil CD16/32expression in vivo and in vitro (14/36), while nAi-ICs could not (22/36).The above work demonstrated that Ai-ICs should recognize multiple epitopes on antigen molecule to form cross-linking, and could manifest as high band in Native-PAGE, induce PCA and down-regulate neutrophil CD16/32expression in vivo and in vitro. These indexes would be expected to be used as the safety criteria for the developments and clinical applications of therapeutic antibodies.Part III. The factors influencing PSA triggered by Ai-ICs and the effector mechanism of IgG pathwayIn this part, we found that the occurrence and severity of PSA is associated with the instantaneous antigen concentration in the peripheral blood in the presence of antibody. Four antigen-antibody systems were used for this phenomenon. OVM was extremely unstable in plasma and was quickly removed. The PSA symptoms were reduced or even undetectable with delayed2C2injection, since OVM was extremely unstable in plasma and was quickly removed0,10,20minutes after mAb2C2injection. Similarly, recombinant human hemoglobin zeta (rHb zeta) were also found unstable in plasma and reduced PSA symptoms could be observed following delayed3H9+4D11injection. However, HSA were relatively stable in plasma, and PSA symptoms remained unchanged following delayed2E9+3A5injection. These results indicate that the different antigen cleaning rate in blood depends on half-life of each protein, which would impact the instantaneous antigen concentration that is associated with the occurrence of PSA.The study of the effector mechanism of IgG pathway was mainly focused on the exploration of cells and mediators that played pivotal roles in PSA. By depletion or blockage of basophils, neutrophils and monocyte/macrophages using Mar-1, anti-Gr1, GdCl3, Clo-lipo, CPM etc, we found that macrophages were the major effector cells mediating IgG pathway, while hemocytes including basophils, neutrophils, monocytes were not. Similarly, we also found that PAF, rather than histamine, was the major mediator for IgG pathway by using blocking agents CV-3988and Triprolidine. Moreover, PSA symptoms remained the unchanged in complement C3-/-mice and mice subjected to splenectomy, while Syk was proved to play an important role in the IgG pathway.Part IV. The discovery of soluble Ly6G and the establishment of its detection systemThis work is based on an unexpected discovery in the study of monocyte/macrophage (Mo/Mcφ) function. After Mo/Mφ depletion by using Clodronate encapsulated liposomes (Clo-lipo), Gr1staining of murine blood neutrophils was considerably down-regulated, while the staining of CD11b and CD16/32was uninfluenced. Blood neutrophil staining remained unchanged after in vitro stimulation with Clo-lipo, indicating the down-regulated Grl staining induced by Clo-lipo in vivo may be not directly induced by Clo-lipo impacting on neutrophils. Further investigations showed that GdCl3, an inhibitor of Mo/Mφ function, could also induce down-regulated neutrophil Grl staining, highly suggesting that the depletion of Mo/Mφ was associated with down-regulated neutrophil Grl staining. Then, we found that the staining of neutrophil Grl was homogenized in mixed blood collected from mice subjected to PBS encapsulated liposomes (PBS-lipo) and mice subjected to Clo-lipo, and neutrophils with different Gr1expression could not be observed. Moreover, after plasma was removed from blood collected form mice subjected to Clo-lipo, the staining of neutrophil Grl on blood cells immediately recovered, indicating that an unknown substance existed in plasma blocked the binding between anti-Grl antibody and Grl expressed on neutrophil surface. Additionally, the exchange of plasma and blood cells collected from mice subjected to PBS-lipo and Clo-lipo also confirmed the existence of this substance in plasma.The soluble form of Ly6G (sLy6G) was confirmed by Dot-blot and Western blot, which showed that the plasma acquired a peculiar25kDa protein in mice subjected to Clo-lipo. Furthermore, we established a sandwich ELISA system for the detection of sLy6G in plasma collected from mice subjected to Clo-lipo by using anti-Grl against Ly6G and Ly6C as capture antibody and anti-Ly6G as detect antibody.The binding assay using plasma collected from mice subjected to Clo-lipo and peritoneal cells proved that sLy6G specifically bind macrophages in vitro. Besides, the in vivo experiments, plasma sLy6G was unstable in the in mice subjected to PBS-lipo injected with the plasma collected from mice subjected to Clo-lipo, indicating that the clearance of sLy6G was Mo/Mφ dependent. In summary, based on PSA model established by using36IgG antibody-antigen complexes, we characterized the initiator of IgG pathway-Ai-ICs. The mAbs or mAb cocktails in Ai-ICs should recognize multiple epitopes on antigen molecule, antibody and antien constructing Ai-ICs should be capable of forming sandwich ELISA in vitro, manifesting as high band in Native-PAGE, inducing PCA and down-regulating neutrophil CD16/32expression in vivo and in vitro. These indexes are expected to be used as the safety criteria for the developments and clinical applications of therapeutic antibodies. The occurrence and severity of PSA is associated with the instantaneous antigen concentration in the peripheral blood in the presence of antibody. IgG pathway is triggered by Ai-ICs, and PAF is as major effector mediator and macrophages are as the major effector cells. It is also independent of white blood cells, complement C3and intact spleen structure, while Syk was found playing an important role. Moreover, we discovered an soluable form of Ly6G and established its sandwich ELISA detection system, which is never reported before. Further studies confirmed that sLy6G specifically bound Mo/Mφ and was removed from plasma in normal mice through a Mo/Mcp dependent pathway. |
PDF Full Text Request