Quantitative in vivo comparisons of the Fc gamma receptor-dependent agonist activities of different fucosylation variants of an immunoglobulin G antibody.

Although it has been shown that functions of immunoglobulin G (IgG) antibodies (Abs) that depend on binding to certain Fc gamma receptors (Fc gamma R) can be influenced by Fc glycan fucosylation, quantitative in vivo analyses comparing the effects of different levels of fucose are still lacking. We used a simple mouse model to compare Fc gamma R-dependent T cell activation induced by different fucosylation variants of a hamster/human IgG1 chimeric version of anti-mouse CD3 monoclonal Ab, 145-2C11 (2C11). Initial studies supported the expectation that this agonist activity by 2C11 was a reflection of Fc gamma R binding, including comparisons of human IgG1 and IgG4 variants of 2C11 that showed the IgG4 to be dramatically less active at inducing T cell activation. Dose-response analyses in mice then showed that a sample of the human IgG1 version of 2C11 Ab in which 40% of the Fc glycans in the population of Ab molecules were fucosylated was 3-5 times more potent than a sample with 90% of its Fc glycans fucosylated. A sample with 10% fucosylation showed the same activity as the 40% fucosylated sample, revealing that complete absence of fucose was not necessary to achieve maximal Fc function in this model. In vitro binding to recombinant mouse Fc gamma Rs by the 2C11 variants revealed interesting relationships between fucose content and receptor affinity, and suggested the involvement of Fc gamma RIV in mediating 2C11 activity in vivo. These analyses showed that low-fucose human IgG1 Abs indeed show greater Fc gamma R-dependent activities in mice, but that Abs with moderate levels of fucose may be just as potent as Abs with very low or no fucose.

Ex vivo activated OVA specific and non-specific CD4+CD25+ regulatory T cells exhibit comparable suppression to OVA mediated T cell responses.

CD4+CD25+ regulatory T cells (Tr) are important in maintaining immune tolerance to self-antigen (Ag) and preventing autoimmunity. Reduced number and inadequate function of Tr are observed in chronic autoimmune diseases. Adoptively transferred Tr effectively suppress ongoing autoimmune disease in multiple animal models. Therefore, strategies to modulate Tr have become an attractive approach to control autoimmunity. Activation of Tr is necessary for their optimal immune regulatory function. However, due to the low ratio of Tr to any given antigen (Ag) and the unknown nature of Ag in many autoimmune diseases, specific activation is not practical for potential therapeutic intervention. It has been shown in animal models that once activated, Tr can exhibit immune suppression in a bystander Ag-non-specific fashion, suggesting the effector phase of Tr is Ag independent. To investigate whether the immune suppression by activated bystander Tr is as potent as that of the Ag specific Tr, Tr cells were isolated from BALB/c or ovalbumin (OVA) specific T cell receptor (TCR) transgenic mice (DO11.10) and their immune suppression of an OVA specific T cell response was compared. We found that once activated ex vivo, Tr from BALB/c and DO11.10 mice exhibited comparable inhibition on OVA specific T cell responses as determined by T cell proliferation and cytokine production. Furthermore, their immune suppression function was compared in a delayed type hypersensitivity (DTH) model induced by OVA specific T cells. Again, OVA specific and non-specific Tr exhibited similar inhibition of the DTH response. Taken together, the results indicate that ex vivo activated Ag-non-specific Tr are as efficient as Ag specific Tr in immune suppression, therefore our study provides additional evidence suggesting the possibility of applying ex vivo activated Tr therapy for the control of autoimmunity.

Addition of an extra immunoglobulin domain to two anti-rodent TNF monoclonal antibodies substantially increased their potency.

The functional valency of a monoclonal antibody (mAb) has important influences on such things as antigen avidity, Fc-mediated immune effector functions, and clearance of immune complexes. cV1q, a neutralizing rat/mouse chimeric anti-mouse tumor necrosis factor (TNF) monoclonal antibody (mAb), and Rt108, a neutralizing mouse anti-rat TNF (anti-raTNF) mAb, appear to be functionally monovalent for TNF-binding despite containing two antigen binding sites. The functional monovalency of these two independent anti-rodent TNF mAbs is presumably a result of steric hindrance from one TNF molecule binding to one Fab arm that prevents binding of a second TNF molecule to the other Fab arm. To test whether this steric hindrance could be overcome by introducing extra space and flexibility between the Fab arms, these mAbs were engineered to contain an extra CH1 immunoglobulin domain between the CH1 and hinge domains of their heavy chains. In vitro binding data showed that, compared to the original mAbs, the modified mAbs (S-mAbs) had greater capability of binding two TNF molecules simultaneously. In vitro activity assays showed that, compared to the original mAbs, the S-mAbs had significantly greater TNF-neutralization potency, with the S-mAb version of cV1q (S-cV1q) being 200-fold more effective at blocking mouse TNF (muTNF) and the S-mAb version of Rt108 (S-Rt108) being 20-fold more effective at blocking raTNF. Similar results were observed in vivo, where S-cV1q was between 100- and 500-fold more protective than cV1q in mice challenged with endotoxin. These data reveal that introduction of another constant region immunoglobulin domain into two unrelated mAbs dramatically enhanced their neutralization potency. Other mAbs may also show more potent activity using this engineering approach, particularly mAbs that recognize homopolymeric antigens.