Mammalian heparanase: what is the message?

Abstract

Heparan sulphate proteoglycans are ubiquitous macromolecules of cell surfaces and extracellular matrices. Numerous extracellular matrix proteins, growth factors, morphogens, cytokines, chemokines and coagulation factors are bound and regulated by heparan sulphate. Degradation of heparan sulphate thus potentially profoundly affects cell and tissue function. Although there is evidence that several heparan sulphate-degrading endoglucuronidases (heparanases) might exist, so far only one transcript encoding a functional heparanase has been identified: heparanase-1. In the first part of this review, we discuss the current knowledge about heparan sulphate proteoglycans and the functional importance of their versatile interactions. In the second part, we summarize recent findings that have contributed to the characterization of heparanase-1, focusing on the molecular properties, working mechanism, substrate specificity, expression pattern, cellular activation and localization of this enzyme. Additionally, we review data implicating heparanase-1 in several normal and pathological processes, focusing on tumour metastasis and angiogenesis, and on evidence for a potentially direct signalling function of the molecule. In that context, we also briefly discuss heparanase-2, an intriguing close homologue of heparanase-1, for which, so far, no heparan sulphate-degrading activity could be demonstrated.

1

Heparan sulphate (HS) chains are covalently bound to specific serine residues of the proteoglycan core protein. After assembly of the tetrasaccharide linkage region (Xyl- Gal-Gal-GlcA) and the first GlcNAc residue, further HS polymerization takes place by alternate addition of GlcA and GlcNAc to the non-reducing end of the growing chain. While chain elongation is ongoing, the nascent polysaccharide forms a substrate for the N-deacetylase/N-sulphotransferases, which control all subsequent modifications (C5-epimerization of GlcA to IdoA by an epimerase and O-sulphations by 6-O- and 3-O-sulphotransferases). Each sugar residue is depicted by a symbol shown below the scheme. NA, NA/NS and NS domains are defined according to the degree of GlcNAc sulphation. Also shown are regions that are implicated in the binding of specific ligands, such as FGF-1/FGF-2 and antithrombin (adapted from [2]).

2

Predicted structure and processing of human heparanase-1. Heparanase-1 is synthesized as an inactive precursor of ∼65 kD that subsequently undergoes proteolytic cleavage, yielding ∼8 and ∼50 kD protein subunits that heterodimerize to form the active enzyme. The protein contains a putative N-terminal signal peptide (SP, Met¹-Ala³⁵), a C-terminal hydrophobic region (HR, Pro⁵¹⁵-Ile⁵³⁴), five cysteine residues (black C's) and six N-linked oligosaccharides (orange balloons). Heparanase-1 uses a general acid catalysis mechanism for the hydrolysis of the HS-chains, requiring two critical residues, a proton donor (Glu²²⁵) and a nucleophile (Glu³⁴³) (red asterisks).

3

HS-sequences recognized and cleaved by human heparanase-1. (A) One candidate human heparanase-1 cleavage site, as identified by Pikas et al.[70]. The heparin-derived octasaccharide, which binds antithrombin (binding sequence corresponds to units 2–6, between brackets), is cleaved by human heparanase-1 at a single site (arrow), yielding a GlcA at the newly formed reducing terminus (endo-β-D-glucuronidase activity). The 2-O-sulphate group on a hexuronic acid residue located two monosaccharide units away from the cleavage site is essential for substrate recognition (shown in the largest font size). The 3-O-sulphate group on the GlcN residue, required for high affinity binding of antithrombin, but of no particular relevance in the present context, is shown in italics. (B) A target octasaccharide sequence for human heparanase-1, as identified by Okada et al.[73]. Although longer oligosaccharides are most likely better substrates for the cleavage by human heparanase-1, the minimum size required for recognition is only a trisaccharide (indicated between brackets). A highly sulphated structure is crucial for enzyme action, and the GlcNS structure on the reducing side and GlcN(6S) structure on the non-reducing side of the cleavage site are considered as important for the substrate recognition (shown in the largest font size). The additional 2-N-sulphate group on the non-reducing GlcN or 6-O-sulphate group on the reducing GlcN appear to promote cleavage by heparanase-1 (shown in the middle font size). The GlcN 3-O-sulphate structure (shown in italics) promotes cleavage when it resides in a relatively low sulphated sequence but inhibits cleavage when located in a highly sulphated sequence.

4

Regulation of heparanase-1 mRNA expression in different cell types. Heparanase-1 mRNA expression by RNA polymerase II (RNA pol II) is positively regulated by transcription factors such as Sp1, GA-binding protein (GABP), Ets1 and Ets2, early growth response-1 (EGR1), nuclear factor-kappa B (NF-κB), oestrogen and glucose (or transcription factors induced by the latter two). Expression is negatively regulated by the transcription factor p53 and by DNA-methylation.

5

Four different splice-variants of human heparanase-2. Based on the predicted amino acid sequence, heparanase-2 contains two potential N-glycosylation sites (orange balloons) that are shared by all four isoforms, whereas the B-part contains three additional potential N-glycosylation sites. Moreover, six cysteine residues (black C's) occur in the pro-protein, one in the predicted signal peptide (SP), one in the A-part, one in the B-part and three in the C-terminal part that is present in all isoforms. The red asterisks indicate the two glutamates that correspond with the critical active sites of human heparanase-1 (according to Dürr and David; patent WO 02/04645).

6

Amino acid alignment of human heparanase-1 with human heparanase-2AB. The predicted signal peptides are shown in brown, the ∼8 and the ∼6 kD fragment of heparanase-1 and the A- and B-parts of heparanase-2AB are shown in blue, green, orange and purple respectively. The potential N-glycosylation sites and cysteine residues are underlined, and the two critical active sites of human heparanase-1 and the corresponding glutamates of heparanase-2 are marked in red.