Glycosylation of SPARC is Critical for Binding to Extracellular Matrix Proteins and Protection from Matrix Metalloproteinase Degradation

Abstract

Purpose : Secreted Protein Acidic and Rich in Cysteine (SPARC) is a matricellular glycoprotein that regulates the turnover of extracellular matrix (ECM) in trabecular meshwork (TM) and may act as a key regulatory node for the pathogenesis of primary open angle glaucoma (POAG) by modulating intraocular pressure (IOP).¹ Overexpression of SPARC causes the accumulation of ECM proteins within TM, in part via protective chaperone activity by competitively binding ECM proteins and blocking proteolytic interactions with matrix metalloproteinase-9 (MMP-9). Because SPARC has distinct glycosylation patterns based on the tissue of origin ², we hypothesize that SPARC exhibits distinct binding affinities for ECM proteins and variable chaperone activity depending on its glycosylation structure and tissue of origin.

Methods : Platelet SPARC (PL-SPARC) was commercially obtained and TM-SPARC was purified from TM cell culture using adenovirus carrying human SPARC cDNA. TM SPARC was treated with PNGase F to produce deglycosylated SPARC (DG-SPARC). Surface Plasmon Resonance (SPR) and glycosylation analysis of TM- or PL-derived SPARC was performed at Case Western Reserve University core facilities. In vitro MMP assays incubated a reaction mixture of SPARC, exogenous MMP-2 or MMP-9, collagen I or IV, and Ca²⁺(0 or 2 mM) for 6 hours at 37 C. Immunoblot analysis quantified undigested collagen compared to postive, negative, and BSA controls.

Results : Glycosylation analysis showed that TM-SPARC contained a mixture of fucosylated and non-fucosylated glycans while PL-SPARC had only fucosylated domains. SPR showed that the binding affinity of TM- and PL- SPARC to collagens was different, while DG-SPARC had no binding affinity. TM- and PL-SPARC demonstrated a chaperone effect on MMP2- and 9-mediated Col1 degradation at 2 mM calcium. Interestingly, PL SPARC showed chaperone effect on MMP9-mediated collagen I degradation at 0 mM calcium while TM SPARC did not. SPARC did not protect Col4 from degradation regardless of tissue origin, MMP, or calcium conditions. DG-SPARC showed no chaperone activity.

Conclusions : These results complement earlier work that demonstrate the chaperone activity of normally glycosylated TM SPARC. It appears that the glycosylation of SPARC is a critical factor for both ECM binding and chaperone function.

This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.

 

Fig. 1: Typical immunoblot results of DG-SPARC chaperone assay.