A universal GlycoDesign for lysosomal replacement enzymes to improve circulation time and biodistribution

Abstract

Currently available enzyme replacement therapies for lysosomal storage diseases are limited in their effectiveness due in part to short circulation times and suboptimal biodistribution of the therapeutic enzymes. We previously engineered Chinese hamster ovary (CHO) cells to produce α-galactosidase A (GLA) with various N-glycan structures and demonstrated that elimination of mannose-6-phosphate (M6P) and conversion to homogeneous sialylated N-glycans prolonged circulation time and improved biodistribution of the enzyme following a single-dose infusion into Fabry mice. Here, we confirmed these findings using repeated infusions of the glycoengineered GLA into Fabry mice and further tested whether this glycoengineering approach, Long-Acting-GlycoDesign (LAGD), could be implemented on other lysosomal enzymes. LAGD-engineered CHO cells stably expressing a panel of lysosomal enzymes [aspartylglucosamine (AGA), beta-glucuronidase (GUSB), cathepsin D (CTSD), tripeptidyl peptidase (TPP1), alpha-glucosidase (GAA) or iduronate 2-sulfatase (IDS)] successfully converted all M6P-containing N-glycans to complex sialylated N-glycans. The resulting homogenous glycodesigns enabled glycoprotein profiling by native mass spectrometry. Notably, LAGD extended the plasma half-life of all three enzymes tested (GLA, GUSB, AGA) in wildtype mice. LAGD may be widely applicable to lysosomal replacement enzymes to improve their circulatory stability and therapeutic efficacy.

Comparison of efficacy between subcutaneous and intravenous application of moss‐aGal in the mouse model of Fabry disease

“…While enzyme activities are lower in the heart than in the kidney after sc injection (11% vs. 21%, Figure S1), the corresponding Gb3 reduction is more efficient in this organ (33% vs. 8%, Figure S4). This observation is in line with reports by others indicating that Gb3 clearance is least effective in the kidneys of Fabry mice 23,24 …”

“…This observation is in line with reports by others indicating that Gb3 clearance is least effective in the kidneys of Fabry mice. 23,24 Use of higher dose could increase moss-aGal's bioavailability (Figure S1). Results from the initial efficacy study show dose dependency; with the strongest Gb3 reduction obtained for 10 mg/kg (Figure S3), consistent with the highest enzyme activities for this dose.…”

Glycoengineered recombinant alpha1-antitrypsin results in comparablein vitroandin vivoactivities to human plasma-derived protein

“…Beyond the ability to perfectly capture human-like glycan structures, glycoengineering CHO cells can also be utilized to produce therapeutic proteins having optimized glycoprofiles that impart improved binding, heightened specificity or better pharmacokinetic profiles. Modifying the native glycosylation profiles of many biologics encompassing monoclonal antibodies, protein vaccines and hormones have resulted in better therapeutic outcomes(Chen et al, 2023; Lusvarghi et al, 2023; Newby, Allen, & Crispin, 2023; Rocamora et al, 2023), and tools are emerging to enable more predictive glycoengineering of therapeutics. (Spahn, Hansen, Kol, Voldborg, & Lewis, 2017; Spahn & Lewis, 2014; Yang et al, 2015) In the case of A1AT, different diseases states of inflammation and malignancy have been shown to be associated with alternative glycoforms of the protein suggesting that variations in glycosylation could play a role in tuning the protein’s immunomodulatory properties(McCarthy et al, 2014).…”