Modulation of urelumab glycosylation separates immune stimulatory activity from organ toxicity

Abstract

Checkpoint control and immunomodulatory antibodies have become important tools for modulating tumor or self-reactive immune responses. A major issue preventing to make full use of the potential of these immunomodulatory antibodies are the severe side-effects, ranging from systemic cytokine release syndrome to organ-specific toxicities. The IgG Fc-portion has been demonstrated to contribute to both, the desired as well as the undesired antibody activities of checkpoint control and immunomodulatory antibodies via binding to cellular Fcγ-receptors (FcγR). Thus, choosing IgG subclasses, such as human IgG4, with a low ability to interact with FcγRs has been identified as a potential strategy to limit FcγR or complement pathway dependent side-effects. However, even immunomodulatory antibodies on the human IgG4 background may interact with cellular FcγRs and show dose limiting toxicities. By using a humanized mouse model allowing to study the immunomodulatory activity of human checkpoint control antibodies in vivo, we demonstrate that deglycosylation of the CD137-specific IgG4 antibody urelumab results in an amelioration of liver toxicity, while maintaining T cell stimulatory activity. In addition, our results emphasize that antibody dosing impacts the separation of side-effects of urelumab from its therapeutic activity via IgG deglycosylation. Thus, glycoengineering of human IgG4 antibodies may be a possible approach to limit collateral damage by immunomodulatory antibodies and allow for a greater therapeutic window of opportunity.

Keywords: CD137; Fc-receptors; glycosylation; therapeutic antibody; urelumab.

Figure 1

Effect of treatment with urelumab antibody variants on body weight and temperature in humanized mice. Shown are relative changes in body weight (A) and body temperature (B) of humanized mice after treatment with a human IgG4 isotype control (6µg/g) or 3µg/g or 6µg/g of urelumab (αCD137) or a deglycosylated urelumab variant (αCD137 PNGaseF) until twenty days after antibody injection (hIgG4: n=5, αCD137 6µg: n=7, αCD137 PNGaseF 6µg: n=3, αCD137 3µg: n=3, αCD137 PNGaseF 3µg: n=4). Shown is the mean+/-SEM of one representative out of two independent experiments. For assessment of statistical significance a Two Way ANOVA with Tukey’s multiple comparison test was used. In (A) * indicates p<0.05 for hIgG4 (6µg/g) vs. αCD137 (6µg/g) and hIgG4 (6µg/g) vs. αCD137 PNGaseF (6µg/g). In (B) ** indicates p<0.01 for hIgG4 (6µg/g) vs. αCD137 (3µg/g), αCD137 (6µg/g) vs. αCD137 3(µg/g) and αCD137 PNGaseF (6µg/g) vs. αCD137 (3µg/g).

Figure 2

Impact of urelumab antibody dose and glycosylation on human T cells in the peripheral blood of humanized mice. Humanized mice received either 3 or 6µg/g of a human IgG4 isotype control, urelumab (αCD137) or deglycosylated urelumab (αCD137 PNGaseF) and were studied for the next twenty days after antibody injection. (A, B) Shown are the relative amounts of human CD3+ T cells (% of human CD45+ cells) and the absolute numbers of CD4+ or CD8+ T cells per ml blood of mice treated with 3µg/g (hIgG4: n= 5; αCD137: n=6; αCD137 PNGaseF: n=4) (A) or 6µg/g (hIgG4: n= 7-13; αCD137: n=10-11; αCD137 PNGaseF: n=6) (B) of the indicated antibody preparations. Coloured asterisks matching the respective colouring of the treatment group indicate significant differences at the indicated time-point of this group compared to the hIgG4 isotype control group. (C-F) Representative dot plots demonstrating the expansion of human CD3+ T cells (C) or CD4+ and CD8+ T cell subsets (D) in the peripheral blood of humanized mice before and at 8 or 13 days after injection of 6µg/g of the respective IgG4 antibodies; and quantification (E, F) of the relative abundance of human T cell subsets in mice receiving 3µg/g (hIgG4: n=6; αCD137: n=6 αCD137 PNGaseF: n=4) (E) or 6µg/g (hIgG4: n=11; αCD137: n=11 αCD137 PNGaseF: n=5) of the specified antibodies (F) at the indicated time points after injection. (G, H) CD137 expression (mean ΔMFI) on CD3+ (left panel), CD4+ (middle panel) and CD8+ (right panel) T cells in mice treated with 3µg/g (G) (hIgG4 n= 3; αCD137: n=6; αCD137 PNGaseF: n=4) or with 6µg/g (H) (hIgG4: n=7; αCD137: n=6-7; αCD137 PNGaseF: n=4-6) of the respective antibody preparations. Shown are pooled data from two to three independent experiments. Results are presented as mean +/-SEM. For statistical analysis 2-Way Anova with Tukey’s multiple comparison test was used to assess significant differences between experimental groups. *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001.

Figure 3

Effects of treatment with urelumab variants on human monocytes and NK cells in the peripheral blood of humanized mice. Humanized mice were injected with 3µg/g (hIgG4: n=3-4; αCD137: n=6; αCD137 PNGaseF: n=4) (A, C, E) or 6µg/g (hIgG4: n=7; αCD137: n=5; αCD137 PNGaseF: n=4) (B, D, F) of the human IgG4 isotype control, urelumab (αCD137), or deglycosylated urelumab (αCD137 PNGaseF) and studied for twenty days after antibody injection. (A, B) Shown are the relative changes in CD33+ monocyte numbers (left panel) as well as in CD137 expression (right panel, ΔMFI) on CD33+ monocytes at the indicated time-points after injection of 3 (A) or 6 (B) µg/g of the respective antibody preparations. (C-F) Shown are the relative changes in CD56⁺CD16^- (C, D) and CD56⁺CD16⁺ (E, F) NK cell subset abundance (left panel) as well as in CD137 expression (right panel) on the respective NK cell subsets at the indicated time-points after injection of 3 (C, E) or 6 (D, F) µg/g of the respective antibody preparations. Results are presented as mean +/-SEM. For statistical testing a 2-Way Anova with Tukey’s multiple comparison test was used. *p<0.05; **p<0.01.

Figure 4

Effect of urelumab variant treatment on T cells in primary and secondary immunological organs. Humanized mice received either 3 or 6µg/g of a human IgG4 isotype control, urelumab (αCD137) or deglycosylated urelumab (αCD137 PNGaseF) and were studied for twenty days after antibody injection. (A) Shown are representative immunofluorescent stainings of lymph node sections of mice treated with 6µg/g hIgG4 as a control or with urelumab (αCD137) identifying human T cells (hCD3), human dendritic cells (hCD11c), as well as mouse macrophages (mCD68). (B, C) Depicted are relative amounts of human CD3+ cells (B) and of different T cell subsets (C) in thymus, spleen, bone marrow, and lymph nodes of mice treated with 3µg/g of the indicated antibodies (hIgG4: n= 6; αCD137: n=6; αCD137 PNGaseF: n=4). (D, E) Depicted are relative amounts of human CD3+ T cells (D) and of different T cell subsets (E) in thymus, spleen, bone marrow, and lymph nodes of mice treated with 6µg/g of the indicated antibody variants (hIgG4: n= 11; αCD137: n=8-12; αCD137-PNGaseF: n=4-6). Shown is the combined data from two to three independent experiments and results are represented as mean +/- SEM. Statistical analysis was done by Shapiro-Wilk normality test, Kruskal-Wallis with Dunn’s multiple comparison test or 1-way ANOVA. For analysing T cell subsets a 2-Way ANOVA was performed. *p<0.05; **p<0.01; ****p<0.0001.

Figure 5

Effect of urelumab treatment dose and glycosylation on serum cytokine levels in humanized mice. Humanized mice received either 3 or 6µg/g of a human IgG4 isotype control, urelumab (αCD137) or deglycosylated urelumab (αCD137 PNGaseF) and serum samples were collected before and at 8 and 13 days after antibody injection. Shown are concentrations of serum cytokine levels before and at the indicated time-points after injection of 3 (A, C, E, G, I, K) or 6 (B, D, F, H, J, L) µg/g. Depicted are serum concentrations of IL-6 (hIgG4: n=3-8, αCD137: n=6-8, αCD137 PNGaseF: n=3-5) (A, B), MCP-1 (hIgG4: n=3-7, αCD137: n=5-7, αCD137 PNGaseF: n=3-4) (C, D), IL10 (hIgG4: n=4-9, αCD137: n=6-10, αCD137 PNGaseF: n=3-5) (E, F), IL12p40 (hIgG4: n=3-9, αCD137: n=6-9, αCD137 PNGaseF: n=3-5) (G, H), IFNγ (hIgG4: n=3-7, αCD137: n=6-8, αCD137 PNGaseF: n=4-5) (I, J), and IL17A (hIgG4: n=4-8, αCD137: n=6-9, αCD137 PNGaseF: n=4) (K, L). Results are expressed as mean +/-SD. Statistical analysis was done using ROUT outlier test (Q=1%) and Shapiro-Wilk normality test. For graphs showing serum concentration of cytokines either Friedman test with Dunn’s multiple comparison test or RM-One-Way ANOVA with Tukey’s multiple comparison test was performed. *p<0.05.

Figure 6

Impact of αCD137 treatment on kidney function. Humanized mice received either 3 or 6µg/g of a human IgG4 isotype control, urelumab (αCD137) or deglycosylated urelumab (αCD137 PNGaseF). (A, B) Shown are blood urea nitrogen (BUN) levels before (day 0) or at the indicated timepoints after treatment of humanized mice with 3 (hIgG4: n=4, αCD137: n=6, αCD137 PNGaseF: n=4) (A) or 6 (hIgG4: n=4, αCD137: n=5, αCD137 PNGaseF: n=3) (B) µg/g of the IgG4 antibody variants, as indicated. (C) Shown are hematoxylin-eosin (H–E) and Sirius red stainings of kidney sections of mice treated with 6µg/g of the indicated IgG4 antibody variants twenty days after antibody injection. Scale bars represent 100µm. Results are expressed as mean with SD. Statistical analysis was done by using Shapiro-Wilk normality test; p values were determined by using either Friedman test with Dunn’s multiple comparison analysis or One-Way-Anova with Tukey’s multiple comparison test. *p<0.05.

Figure 7

Impact of treatment with urelumab variants on liver pathology. Humanized mice received either 3 or 6µg/g of a human IgG4 isotype control, urelumab (αCD137) or deglycosylated urelumab (αCD137 PNGaseF) and liver pathology was studied twenty days after antibody injection. (A, B) Representative hematoxylin/eosin stained liver sections (A) of mice treated with 3 or 6µg/g of human IgG4 isotype control, urelumab and deglycosylated urelumab variants (scale bar 100µm) and quantification of the mean infiltrate area (B) of immune cell infiltrates (hIgG4: n=3, αCD137: n=6, αCD137 PNGaseF: n=4). (C) Shown is the percentage of mice with detectable immune cell infiltrates in the liver. (D, E) Immunofluorescent staining of liver sections (scale bar 100µm) (D) and quantification of the infiltration (E) of CD3+ T cells (upper panel) and the relative amount of CD3+CD8+ T cells within the CD3+ T cell population (lower panel) (hIgG4: n=9, αCD137: n=7, αCD137 PNGaseF: n=7, n represents the number of analysed images of two independent mice per group). Statistical testing was performed by using a Shapiro-Wilk and Kruskal Wallis test with a Dunn’s multiple comparison test. *p<0.05; **p<0.01.