Engineered binding to erythrocytes induces immunological tolerance to E. coli asparaginase

Abstract

Antigen-specific immune responses to protein drugs can hinder efficacy and compromise safety because of drug neutralization and secondary clinical complications. We report a tolerance induction strategy to prevent antigen-specific humoral immune responses to therapeutic proteins. Our modular, biomolecular approach involves engineering tolerizing variants of proteins such that they bind erythrocytes in vivo upon injection, on the basis of the premise that aged erythrocytes and the payloads they carry are cleared tolerogenically, driving the deletion of antigen-specific T cells. We demonstrate that binding the clinical therapeutic enzyme Escherichia coli l-asparaginase to erythrocytes in situ antigen-specifically abrogates development of antibody titers by >1000-fold and extends the pharmacodynamic effect of the drug 10-fold in mice. Additionally, a single pretreatment dose of erythrocyte-binding asparaginase tolerized mice to multiple subsequent doses of the wild-type enzyme. This strategy for reducing antigen-specific humoral responses may enable more effective and safer treatment with therapeutic proteins and drug candidates that are hampered by immunogenicity.

Keywords: Immunological tolerance; asparaginase; erythrocyte; protein drug.

Fig. 1. ASNase conjugation to the ERY1 peptide imparts long-term erythrocyte binding in vivo and enhanced pharmacodynamics of plasma asparagine depletion.

(A) Erythrocyte-binding ASNase is produced by chemically conjugating copies of the glycophorin A–binding peptide (ERY1) to ASNase. (B) Characterization of apparent affinity values of ERY1-ASNase conjugates functionalized with varying molar equivalents of ERY1 peptide per ASNase monomer, determined by flow cytometry (one-site binding, R² ≥ 0.98 for all curve fits). (C) Specific enzymatic activity characterization of ERY1-ASNase and wild-type (WT) ASNase, determined by a fluorogenic assay (n ≥ 4). (D) In vivo ERY1-ASNase binding specificity to erythrocytes (CD45⁻) but not to other blood cell populations. (E) ERY1-ASNase binding (right panel) and WT ASNase nonbinding (left panel) to healthy (annexin V⁻) and apoptotic (annexin V⁺) erythrocytes (CD45⁻) 24 hours after a 15-μg intravenous injection, determined by flow cytometry. (F) Cell surface binding kinetics and half-life of ERY1-ASNase bound to circulating erythrocytes after a single 15-μg intravenous injection, determined by flow cytometry (n = 3, one-phase decay, R² = 0.97). (G) Plasma asparagine (Asn) concentration of mice injected with a single 15-μg dose of either WT ASNase or ERY1-ASNase, determined by quantitative liquid chromatography–mass spectrometry (LC-MS) [data presented as means ± SEM (n = 5); repeated once; dotted lines represent homeostatic plasma asparagine concentrations (36)].

Fig. 2. Immunophenotyping demonstrates that ERY1-ASNase induces minimal changes in the immune cell repertoire.

Mice were administered two weekly intravenous (i.v.) injections of 15 μg of ERY1-ASNase or WT ASNase, and hematology measurements and immune cell analyses were performed. (A) Hematology parameters and B and T cell population enumeration in the blood (granulocyte, B, CD4 T, and CD8 T cell populations determined by flow cytometry; all others by hematology analyzer). (B and C) Immune cell population enumeration in the spleen (B) and bone marrow (C), determined by flow cytometry (all data presented as relative fold change between groups; n = 5; Student’s two-tailed t test: *P < 0.05, **P < 0.01, ***P < 0.001).

Fig. 3. Erythrocyte-binding ASNase is nonimmunogenic and acts as a tolerogen that enables follow-on treatment with WT enzyme.

(A) Time course of anti-ASNase IgG antibody development in plasma of mice administered with eight weekly 15-μg doses of either WT ASNase or ERY1-ASNase or with one or two tolerogenic doses of ERY1-ASNase followed by WT ASNase for the remaining doses (end-point IgG titers, Mann-Whitney U test: *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001). (B) Time course of anti-ASNase IgG antibody development in plasma of mice administered a dose-sparing regimen of 15 μg of either WT ASNase or ERY1-ASNase every 3 weeks for a total of four doses (end-point IgG titers, Mann-Whitney U test: *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001). (C) Immunogenicity incidence rates of the dosing regimens of ERY1-ASNase and WT ASNase (Mantel-Cox test: *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001). (D to G) End-point plasma anti-ASNase antibody titers of subclass (D) IgG1, (E) IgG2a, (F) IgG2b, and (G) IgG3 (Mann-Whitney U test: *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001). Initial dilutions were 100-fold for all samples; dotted lines indicate mean values for the naïve population (n = 149); data are presented as means ± SEM (n = 4 to 9 per group until 80 days; n = 3 to 9 from 80 days onward). (H) End-point IgG subclass profile, as calculated by normalization of the sum of all titers across subclasses (n = 3 to 9).

Fig. 4. ERY1-ASNase treatment induces antigen-specific tolerance in a nontoxic and age-independent manner.

(A) Mice were assessed for their capacity to respond to the irrelevant (nontolerized) antigen OVA after a single tolerogenic dose of ERY1-ASNase. (B) Time course of the development of anti-OVA total IgG in mice tolerized with 15 μg of ERY1-ASNase or WT ASNase and challenged with five weekly intravenous doses of OVA. (C) Mice were assessed for hematological parameters and for capacity to respond to the irrelevant antigen OVA after chronic intravenous administration with 15 μg of ERY1-ASNase or WT ASNase every 3 weeks for a total of four injections. (D) Hematological parameters at day 87 (data presented as ERY1-ASNase fold increase over WT ASNase; Student’s two-tailed t test, P > 0.05 for all parameters, n = 3 to 5). (E) Quantification of end-point antigen-specific total IgG responses to a challenge with the irrelevant T cell–dependent antigen OVA (n = 3 to 5, Mann-Whitney U test). (F and G) Time course of anti-ASNase IgG development in plasma of aged (27-week-old) mice receiving 15 μg of ERY1-ASNase or WT ASNase, represented as (F) log₁₀ titer (n = 3 to 4, Mann-Whitney U test, *P < 0.05) and (G) absorbance reading of 1:100 dilution of plasma [n = 3 to 4; analysis of variance (ANOVA), ****P < 0.0001)]; dotted lines indicate mean values for the naïve population (n = 149).