Effects of Glycosylation on the Enzymatic Activity and Mechanisms of Proteases

Abstract

Posttranslational modifications are an important feature of most proteases in higher organisms, such as the conversion of inactive zymogens into active proteases. To date, little information is available on the role of glycosylation and functional implications for secreted proteases. Besides a stabilizing effect and protection against proteolysis, several proteases show a significant influence of glycosylation on the catalytic activity. Glycans can alter the substrate recognition, the specificity and binding affinity, as well as the turnover rates. However, there is currently no known general pattern, since glycosylation can have both stimulating and inhibiting effects on activity. Thus, a comparative analysis of individual cases with sufficient enzyme kinetic and structural data is a first approach to describe mechanistic principles that govern the effects of glycosylation on the function of proteases. The understanding of glycan functions becomes highly significant in proteomic and glycomic studies, which demonstrated that cancer-associated proteases, such as kallikrein-related peptidase 3, exhibit strongly altered glycosylation patterns in pathological cases. Such findings can contribute to a variety of future biomedical applications.

Keywords: Michaelis constant; N-glycosylation; O-glycosylation; core glycan; enzyme kinetics; flexible loops; secreted protease; sequon; substrate recognition; turnover number.

Figure 1

Examples of the most relevant types of glycosylation according to the literature [3,27]. (A) N-glycosylation of asparagine in sequons with the consensus sequence Asn-Xaa-Ser/Thr. N-glycans are generated by trimming and extending the common precursor GlcNAc₂Man₉Glu₃. Small core glycans are mostly intermediates in mammalian glycan synthesis, but often occur in more primitive eukaryotes and insects, as used for recombinant expression. Mammalian N-glycans exhibit an enormous diversity, due to many possible combinations of branching sugars; (B) O-glycosylation at Ser and Thr is found in all kingdoms of life. There is no distinct consensus sequence, but proline-rich regions are favored, e.g., a typical O-glycan site would be Pro-Ser/Thr-Xaa-Yaa-Pro. A very common mammalian O-glycan is the mucin-type that starts with GalNAc and is extended by galactose and sialic acids or GlcNAc, with eight different cores known. In addition, the O-xylose linked, non-branched glucosamine glycans (GAG) or proteoglycans are a large and diverse glycan family. The displayed chondroitin can be phosphorylated and heavily sulfated, comprising up to fifty disaccharide units. O-GlcNAc glycans occur inside cells, even in the nucleus, while O-galactosylation is found at hydroxylysine residues (Hyl) of collagens.

Figure 2

Substrate recognition and catalytic steps in proteases. (A) Two-dimensional interaction map of the model peptide AWTRVR-SILMHY with the specificity subsites S6-S6′ of a KLK protease, calculated with the MOE software [114]. The tryptic specificity is based on the electrostatic interaction of P1-Arg and Asp189 in the S1 pocket. Chymotryptic proteases prefer hydrophobic P1 side chains, such as Tyr or Phe; (B) The aspartic protease mechanism requires a pair of Asp residues, with one being the general base that activates a water molecule as in BACE-1 [115]. After substrate binding, the nucleophilic water attacks the scissile bond between P1 and P1′ at the carbonyl C atom; (C) Three major metalloprotease mechanisms are known, such as the favored one for MMP-3 [116]; and (D) Catalysis of a serine protease with chymotrypsinogen numbering [117]. Other serine protease clans exhibit different arrangements of the triad with similar mechanisms. The catalytic triad activates the Ser Oγ as nucleophile via an acid (Asp102) and a general base (His57), which activates a water molecule for hydrolysis of the acyl intermediate [118]. In addition, cysteine protease mechanisms are related, but often require only Cys-His dyads, since the Sγ is more nucleophilic than the Ser Oγ [119]. Similarly, threonine proteases have a nucleophilic Thr Oγ, while the N-terminus and/or a Lys side chain serve as bases [120]. MEROPS lists variations of catalytic residues and rare protease types [45].

Figure 3

Selected human proteases shown as molecular surface with glycans as spheres according to the standard sugar coloring scheme (Figure 1). Active site residues are colored dark red, while inhibitors or substrates are shown as ball-and-stick models. (A) The aspartic protease cathepsin D with two N-glycans distant from the active sit cleft (1LYA/1LYB); (B) The papain-like cysteine protease cathepsin C has four N-glycans, one is located near the substrate binding region (1K3B/2DJF); (C) The metalloprotease meprin β carries seven N-glycans, with the one at Asn254 located at the substrate binding cleft near the catalytic Zn²⁺ (4GWM); (D) Carboxypeptidase N is another Zn²⁺-dependent protease, with three O-glycans distant to the active site (2NSM); (E) The trypsin-like factor VII is an important activator in the blood coagulation cascade, exhibiting an N-glycan in the catalytic domain and two O-glycans in the EGF-like domain 1 (1QFK); (F) Plasmin, another trypsin-like protease degrades fibrin clots and is O-glycosylated between kringle domains 3 and 4 (4DUR); (G) The trypsin-like KLK1 exhibits one N-linked GlcNAc of a core glycan (1SPJ), which is sufficient to rigidify the flexible 99-loop (yellow) at the active sit cleft, filled with the PPACK inhibitor of the closely related KLK2 (4NFF), in order to explain the effect of glycosylation on substrate binding. In glycan-free KLK2, the 99-loop is more flexible and open; (H) The chymotrypsin-like KLK3/PSA carries a triantennary N-glycan, which may enhance binding of natural substrates and inhibitors (3QUM/2ZCK). The flexible 99-loop resembles the one of KLK1 (Figure 3G) and is depicted in yellow; (I) DPP4 is a membrane anchored, dimeric α/β-hydrolase, with the shielded active site located inside a cavity. The glycans may prevent aggregation and could play a role in the receptor function of DPP4 (1N1M).

Figure 4

Free energy profile for substrate turnover by the serine protease trypsin [118,144,148]. Various approaches find shifted energy levels and additional energy minima of intermediates and the enzyme product complex [149,150]. The y-axis represents the Gibb´s free energy of the process (∆G), which is temperature dependent and related to the reaction enthalpy and entropy: ∆G = ∆H − T∆S. The reaction coordinate represents the progress of the reaction, not a real time process. E, S, and P denote enzyme, substrate and products, while TS1 to TS4 are transition states. Direct conversion of k_cat and K_M into free energy values is not feasible, while they show an inverse correlation, e.g., a high k_cat with a lower free activation energy ∆G^‡. Currently, the influence of glycosylation on the single mechanistic steps can only be estimated, but differences in the binding energy of the transition states can be calculated as for mutant enzymes [151]. For the substrate Bz-Pro-Phe-Arg-pNA with the k_cat/K_M values of N-glycosylated KLK2 (76,930 M⁻¹·s⁻¹) and of glycan-free KLK2 (5780 M⁻¹·s⁻¹) with ∆G = −RT·ln([k_cat/K_M]_glyc/[k_cat/K_M]) the result is−6.7 KJ·mol⁻¹ [140].

Figure 5

The effect of N-glycosylation at Asn95 on the active site conformation of KLK2 according to crystal structure derived models and kinetic data [140]. (A) Glycan-free KLK2 expressed in E. coli exhibits a wide open 99-loop and access of substrates, depicted as green ball-and-stick model bound to the specificity subsites (S4 to S2′ specificity subsites are labeled); (B) N-glycosylation at Asn95 favors a closed 99-loop, which covers the non-prime side region, left to Ser195 (dark red) in the standard orientation, which prevents substrate binding; (C) The N-glycosylated 99-loop of KLK2 opens to a lesser extent than in the glycan-free variant. Thus, substrate binding to glycosylated KLK2 requires more free energy, resulting in a lower k_on rate and higher K_M.

Figure 6

Effects of glycosylation on human proteases. After protein synthesis (A) unfolded proteases enter the endoplasmic reticulum (ER), N-glycans are linked (B); which supports folding (C); The N-glycans are trimmed and extended, accompanied by quality control (D); N-glycans are further modified and O-glycans are attached in the Golgi (E); Sorting leads to membrane anchoring or storage in secretory vesicles (F); After secretion, glycosylation prevents aggregation and unspecific binding (G); as well as proteolysis, which increases the stability and lifetime of proteases (H); Glycosylation regulates binding of: activators (I); cofactors (J) oligomer partners (K); inhibitors (L); and substrates (M); Eventually, glycosylation fine tunes turnover and kinetic parameters in enzymatic reactions (N) [183].