The inter-heavy chain disulfide bonds of IgG4 are in equilibrium with intra-chain disulfide bonds

https://doi.org/10.1016/S0161-5890(01)00050-5Get rights and content

Abstract

Unlike other immunoglobulin G (IgG) subclasses, IgG4 antibodies in plasma have been reported to be functionally monovalent. In a previous paper, we showed that the apparent monovalency of circulating IgG4 antibodies is caused by asymmetry of plasma IgG4  a large fraction has two antigen-binding sites resulting in bispecificity. We postulated that the generation of bispecific antibodies was caused by a post-secretion mechanism, involving the exchange of IgG4 half-molecules (i.e. one heavy and one light chain). This hypothesis was based on the observed instability of the inter-heavy chain disulfide bonds of IgG4. To investigate this instability, we constructed IgG4 mutants and analyzed the covalent interaction between the heavy chains by sodium dodecyl sulfate-poly acrylamide gel electrophoresis (SDS-PAGE) under non-reducing conditions. The mutation to serine of one of the hinge cysteines involved in the inter-heavy chain bond formation, Cys226, resulted in a more stable rather than a more labile inter-heavy chain linkage. Moreover, we confirmed that mutating the IgG4 hinge sequence Cys-Pro-Ser-Cys to the IgG1 hinge sequence Cys-Pro-Pro-Cys also markedly stabilizes the covalent interaction between the heavy-chains. These two observations suggested an explanation for the observed instability of the inter-heavy chain disulfide bonds: the formation of an alternative, intra-chain cystine. Obviously, this intra-chain cystine cannot be formed in the mutant where Cys226 is replaced by Ser, and cannot easily be formed in the mutant with the IgG1 hinge sequence (Cys-Pro-Pro-Cys) due to the restricted torsional freedom of prolines. We, therefore, postulate that the lack of a covalent heavy-chain interaction in a subpopulation of IgG4 reflects an equilibrium between inter- and intra-chain cystines. Based upon the published structure of the IgG4-related hinge-deleted IgG1 myeloma protein Mcg, we propose a model for the two forms of IgG4 and for the half-molecule exchange reaction, which might result in the formation of bispecific IgG4 antibodies.

Introduction

Several investigators observed instability of the inter-heavy chain disulfide bonds of human immunoglobulin G4 (IgG4). This was shown by the presence of IgG4 half-molecules (existing of one heavy and one light chain) when the molecule was analyzed on non-reducing sodium dodecyl sulfate-poly acrylamide gel electrophoresis (SDS-PAGE; Petersen and Dorrington, 1974, Colcher et al., 1989, King et al., 1992). However, under non-denaturing conditions IgG4 half-molecules cannot be detected, because these half-molecules associate by strong non-covalent interactions. A single amino acid substitution in the core-hinge region (see Table 1 for the aminoacid sequences of the human IgG1 and IgG4 hinge regions), Cys-Pro-Pro-Cys-Pro (IgG1) instead of Cys-Pro-Ser-Cys-Pro (IgG4), resulted in disappearance of IgG4 half-molecules (Angal et al., 1993) when analyzed on non-reducing SDS-PAGE.

Human serum IgG4 antibodies are unable to precipitate purified antigens (van der Zee et al., 1986). The inability to form large immune-complexes is due to inability of IgG4 to cross-link two antigens, so human IgG4 is behaving as a monovalent antibody (Aalberse et al., 1983, van der Zee et al., 1986). In a previous paper, we showed that this monovalency of serum derived (polyclonal) IgG4 was caused by asymmetry of plasma IgG4. A large fraction of plasma IgG4 molecules have two different antigen-binding sites, resulting in bispecificity (Schuurman et al., 1999). We postulated that the bispecificity of IgG4 was induced by a post-secretion mechanism  the instability of the inter-heavy chain disulfide bonds of IgG4 renders IgG4 susceptible to the exchange of half-molecules. This hypothesis would explain the apparent monovalency of IgG4  exchange in a pool of polyclonal IgG4 would result in bispecific antibodies, behaving as monovalent antibodies.

In this report, we present chimeric IgG4 antibodies with mutations affecting the assembly. These antibodies were expressed to study the cause of the instability of the disulfide bonds in the IgG4 hinge.

Section snippets

Monoclonal chimeric IgG4 antibodies

The mouse/human chimeric IgG4 antibody (hIgG4-Dp2A) was described previously (Schuurman et al., 1997). This chimeric antibody is directed to the house dust mite allergen Der p 2 and originates from the mouse hybridoma cell line 2B12 (Ovsyannikova et al., 1994). The Der p 2-specific chimeric antibodies have a (mouse) kappa light chain.

IgG4 mutants: M1, M2 and M3

The positions of the mutations introduced in the CH1 and hinge domains of IgG4 are indicated in Fig. 1. The amino acid numbering was according to the EU-numbering.

IgG4 mutant antibodies: description and binding characteristics

To study the assembly and the instability of the disulfide bonds in the IgG4 molecule three IgG4 mutant antibodies were developed. These chimeric IgG4 mutant antibodies are directed to the house dust mite allergen Der p 2. In Fig. 1, the positions of the mutations are indicated (amino acid positions are according to EU-numbering).

In mutant 1 (M1) cysteine 131, involved in the inter-heavy–light chain disulfide bond, was replaced by serine. In mutant 2 (M2), cysteine 226, involved in an

Discussion

In this report, chimeric mutant antibodies were expressed to study the assembly of the IgG4 molecule and the cause of the instability of the disulfide bonds in the IgG4 hinge. In the past, kinetic studies on the inter-chain disulfide bond formation of the IgG4 molecule showed that the inter-heavy chain disulfide bonds were relatively slowly formed and were unstable (Petersen and Dorrington, 1974). Instability of the inter-heavy chain disulfide bonds was also observed by other investigators (

Acknowledgements

This work was supported by a research grant from the Netherlands Asthma Foundation grant no. 91.35.

References (20)

There are more references available in the full text version of this article.

Cited by (0)

View full text