Human recombinant arginase enzyme reduces plasma arginine in mouse models of arginase deficiency

Abstract

Arginase deficiency is caused by deficiency of arginase 1 (ARG1), a urea cycle enzyme that converts arginine to ornithine. Clinical features of arginase deficiency include elevated plasma arginine levels, spastic diplegia, intellectual disability, seizures and growth deficiency. Unlike other urea cycle disorders, recurrent hyperammonemia is typically less severe in this disorder. Normalization of plasma arginine levels is the consensus treatment goal, because elevations of arginine and its metabolites are suspected to contribute to the neurologic features. Using data from patients enrolled in a natural history study conducted by the Urea Cycle Disorders Consortium, we found that 97% of plasma arginine levels in subjects with arginase deficiency were above the normal range despite conventional treatment. Recently, arginine-degrading enzymes have been used to deplete arginine as a therapeutic strategy in cancer. We tested whether one of these enzymes, a pegylated human recombinant arginase 1 (AEB1102), reduces plasma arginine in murine models of arginase deficiency. In neonatal and adult mice with arginase deficiency, AEB1102 reduced the plasma arginine after single and repeated doses. However, survival did not improve likely, because this pegylated enzyme does not enter hepatocytes and does not improve hyperammonemia that accounts for lethality. Although murine models required dosing every 48 h, studies in cynomolgus monkeys indicate that less frequent dosing may be possible in patients. Given that elevated plasma arginine rather than hyperammonemia is the major treatment challenge, we propose that AEB1102 may have therapeutic potential as an arginine-reducing agent in patients with arginase deficiency.

© The Author 2015. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com.

Figures

Figure 1.

Elevated plasma arginine levels in patients with arginase deficiency. (A) All plasma arginine levels for each subject are plotted (n = 22 subjects, range of 1–13 plasma arginine levels available for each subject). The black line indicates an estimated upper limit of normal (ULN). (B) Frequent hyperammonemia is not a complication observed in most patients with arginase deficiency. The total number of baseline and interim hyperammonemia episodes reported by each patient was calculated and plotted. The age range of the subjects at the most recent study visit is provided.

Figure 2.

Dose and time-course studies of AEB1102 in neonatal arginase-deficient mice. (A) Plasma arginine in neonatal Arg1−/− mice decreases in a dose-dependent manner 24 h after intraperitoneal injection of AEB1102. (B) Plasma arginine levels in neonatal Arg1−/− increases to near baseline levels between 72 and 120 h after a single intraperitoneal injection (0.05 mg/kg) of AEB1102. −/− = arginase deficient, +/+ = wild type.

Figure 3.

Repeated dose studies of AEB1102 in neonatal arginase-deficient mice. (A) After repeat dosing every 48 h, plasma arginine levels show persistent reduction on postnatal day 11 in neonatal Arg1−/− mice. Treatment was started on Day 6 with a dose of 0.04 mg/kg/dose. (B) Plasma ornithine levels on Day 11 are higher in treated versus untreated neonatal Arg1−/− mice. (C) Treated and untreated neonatal Arg1−/− mice survive until Days 10–14 which is consistent with previously published reports (26). (D) Whole brain arginine levels are significantly higher in untreated versus treated neonatal Arg1−/− mice. (E) Liver arginine levels are not significantly different between untreated versus treated neonatal Arg1−/− mice. One-way ANOVA was used with Dunnett's post-test, and all groups were compared with the −/− enzyme-treated group. Tx = treatment, ***P < 0.0001, **P < 0.01, *P < 0.05.

Figure 4.

Dose and time-course studies of AEB1102 in an adult arginase-deficient mouse model. (A) Plasma arginine begins to rise by Day 11 after tamoxifen treatment in the adult inducible arginase-deficient (Cre+) mice and continues to rise through Day 18. (B) Plasma arginine is reduced in a dose-dependent manner in the adult inducible arginase-deficient mice (Cre+) 24 h after treatment with AEB1102. For each dose, 1–2 outliers in each group did not respond to AEB1102. (C) Plasma arginine levels begin to rise by 72 h after AEB1102 delivery in the adult inducible arginase-deficient (Cre+) mice. Data for control mice are included for reference (Cre−). Doses are provided in mg/kg.

Figure 5.

Repeated dose studies of AEB1102 in an adult arginase-deficient mouse model. (A) Repeated doses of AEB1102 result in persistently decreased plasma arginine levels on Day 18 in adult inducible arginase-deficient mice (Cre+). Mice were treated with AEB1102 or PBS on Day 15 and on Day 17, and plasma was collected ∼24 h after the second dose. (B) Weight loss in adult inducible arginase-deficient mice (Cre+) treated with AEB1102 versus placebo is not significantly different at Day 18. Mice were monitored with weights daily and compassionately euthanized once weight loss was ≥20% (compared with Day 12) or on Day 18 (whichever was first). Five mice (n = 2 Cre+ and PBS-treated; n = 3 Cre+ and AEB1102-treated) were sacrificed on Day 17 because of weight loss ≥20% and were not included in the data analysis. (C) Repeated doses of AEB1102 result in persistent elevation in plasma ornithine on Day 18 in adult inducible arginase-deficient mice (Cre+). (D) Brain arginine levels in adult inducible arginase-deficient mice (Cre+) treated are not significantly different in mice treated with repeated doses of AEB1102 versus PBS. (E) Liver arginine levels are not significantly different in adult inducible arginase-deficient mice (Cre+) treated with repeated doses of AEB1102 versus PBS. All groups were compared with the Cre+ AEB1102-treated mice with a one-way ANOVA with Dunnett's post-test. ***P < 0.0001, *P < 0.01.

Figure 6.

Survival studies in adult arginase-deficient mice treated with AEB1102. (A) Survival in adult inducible arginase-deficient mice (Cre+) treated with AEB1102 versus PBS is not significantly different. (B) Plasma ammonia levels in adult inducible arginase-deficient mice (Cre+) treated with AEB1102 versus PBS are not significantly different. The dotted line represents the upper limit of the range in wild-type mice.

Figure 7.

AEB1102 dose studies in cynomolgus monkeys. (A) The concentration of AEB1102 and plasma arginine levels are shown for monkeys after single and repeat intravenous dosing at 0.1 mg/kg (semi-log scale). Doses were administered at Days 1 and 8. The solid-filled circles represent Monkey #1 and unfilled circles represent Monkey #2. (B) The concentration of AEB1102 and plasma arginine levels are shown in monkeys after single and repeat intravenous dosing at 0.3 mg/kg (semi-log scale). Doses were administered at Days 1 and 8. The solid-filled squares represent Monkey #1 and unfilled squares represent Monkey #2. (C) The concentration of AEB1102 and plasma arginine levels in monkeys are shown after single and repeat intravenous dosing at 1 mg/kg (semi-log scale). Doses were administered at Days 1 and 8. The solid-filled triangles represent Monkey #1 and unfilled triangles represent Monkey #2.