Glycoside Hydrolase Family 33

This page has been approved by the Responsible Curator as essentially complete. CAZypedia is a living document, so further improvement of this page is still possible. If you would like to suggest an addition or correction, please contact the page's Responsible Curator directly by e-mail.

Author: Kyle Robinson and Junho Lee
Responsible Curator: Steve Withers

Glycoside Hydrolase Family GH33
Clan	GH-E
Mechanism	Retaining
Active site residues	Known
CAZy DB link
https://www.cazy.org/GH33.html

Substrate specificities

Sialic acids, often known as N-acetylneuraminic acids (Neu5Ac, NANA, NeuNAc, NeuNA), are a family of nine carbon monosaccharides with a carboxylate group in the carbon 1 position that occupy the terminal position of the glycans, glycoproteins, glycolipids, and polysaccharides in cells and play important roles in interactions of the cell with its environment [1]. More than 50 sialic acid derivatives have been detected in eukaryotic and prokaryotic species; the most frequently detected sialic acids have an α(2,3) or α(2,6) linkage to galactose, N-acetylgalactosamine, and N-acetylglucosamine or an α(2,8) linkage to another sialic acid [2, 3, 4]. These α(2,8) linked sialic acids can occur in polymeric form as polysialic acids, in the brain, as well as in capsular polysaccharides of some bacteria. Sialic acids are hydrolyzed by sialidases (E.C. 3.2.1.18), and these enzymes are categorized into four different glycoside hydrolase(GH) families: GH33, GH34, and GH83 families are exosialidases while GH53 is an endosialidase [5].

The GH33 family includes most bacterial and simple eukaryotic sialidases, trans-sialidases [6] and intramolecular trans-sialidases [7, 8]. Members of GH33 exhibit different preferences for the three most common sialic acid linkage types listed above, despite similar protein structure. For example, sialidases from Salmonella typhimurium LT2, Vibrio Cholerae, and Clostridium septicum, Clostridium sordellii, Clostridium chauvoei, Clostridium tertium demonstrate a higher hydrolysis activity towards α(2,3) linked substrates than α(2,6) linked substrates, while sialidases from Corynebacteriumm diphtheria and Micromonospora viridifaciens prefer to hydrolyze substrates with α(2,6) linkages [2]. One organism may produce sialidase isoenzymes with different substrate preferences. Pasteurella multocida produces two sialidases with different substrate preferences: NanH, an extracellular enzyme favouring α(2,3)-linked sialyllactose over α(2,6)-linked sialyllactose and NanB, a membrane bound enzyme that prefers α(2,6)-linked substrates over α(2,3)-linked substrates [9]. Similarly, membrane-bound NanA of Salmonella pneumoniae displays similar hydrolysis rates for sialyllactoses with α(2,3)-, α(2,6)- and α(2,8)-linkages whereas extracellular NanB from the same organism prefers α(2,3) linkage over substrates with the other two linkage types [2].

Kinetics and Mechanism

GH33 sialidases and trans-sialidases hydrolyse, or transfer sialic acids to acceptor sugars, with retention of anomeric configuration [10, 11]. The general mechanism is depicted here. Considerable debate had occurred over whether an ionic or covalent intermediate was formed. However, a covalent glycosyl-enzyme intermediate was observed on T. cruzi trans-sialidase(TcTS) by mass spectrometry using a fluorinated sialic acid analogue and the labelled amino acid residue identified as Tyr342 by peptide mapping. Subsequent crystal structures provided further insight into the mechanism of the protozoan sialidase [6, 12]. Kinetic analysis of TcTS confirmed the expected ping-pong double-displacement mechanism, and a covalent intermediate was demonstrated, without use of a fluorinated derivative, by use of mass spectrometry [13]. Subsequent structural studies of two strictly hydrolytic sialidases from T. rangelli [14] and C. perfringens [15] also characterised their covalent intermediates.

Catalytic Residues

Nucleophile

The catalytic machinery is conserved throughout this family and, as shown by studies on TcTS, three key residues are involved: Glu230, Tyr342, Asp59. Since Watts first trapped the covalent 3F-sialyl enzyme intermediate on Tyr342 in 2003, many confirmatory structures have been solved. It is likely that tyrosine has evolved as the catalytic nucleophile, rather than the carboxylate moiety found in most retaining glycosidases, in order to minimize repulsive charge-charge interactions with the C1 carboxylate of the sialic acids. The neutral tyrosine can more readily approach the anomeric centre, and is rendered more nucleophilic by an invariant glutamate, Glu230 for TcTS, that serves as a base catalyst for deprotonation/reprotonation of the tyrosine hydroxyl. See Chan et al [16] for a detailed analysis of the role of this residue in the GH33 sialidase from M. viridifaciens.

Acid Base Catalyst

The third mechanistically relevant conserved residue in the active site is Asp59, which serves as the acid/base catalyst, protonating the glycosidic oxygen as bond cleavage occurs. Following formation of the intermediate, Asp59 then serves as a base catalyst, activating the incoming water molecule (for simple hydrolytic pathways) or glycosyl acceptor (in the case of trans-sialidases) towards trans-glycosylation of the glycosyl-enzyme intermediate. This second attack leads to a double inversion at the anomeric center and a net retention of stereochemistry observed in the released product, Neu5Ac [6, 13, 15].

Three-dimensional structures

All members of the sialidase superfamily, including the members of GH34 and GH83, display a 6 bladed beta-propeller sheet catalytic domain [6], which is accepted as the canonical neuraminidase fold. The catalytic site structure is strictly conserved in all three families and contains an arginine triad which binds to the carboxylate in the C1 position of the sialic acid, a Tyr/Glu nucleophilic pair, and an aspartic acid that acts as the acid/base catalyst [5].

Bacterial sialidases may also contain a membrane binding domain, signal domain and a lectin-like domain. Although not all bacterial sialidases have a lectin domain, the lectin domain can be used to recognize the sialic acid in certain species, such as V. cholerae [17]. Also it is not uncommon for a bacterial sialidase to have a carbohydrate binding module (CBM) as one of its domains, such as in M. viridifaciens sialidase [18, 19].

Family Firsts

First stereochemistry determination: First determined by proton NMR by Friebolin et al [10, 11].
First catalytic nucleophile identification: NMR structures of Crennell and others were strongly suggestive, but mechanism was not clear. First definitively shown for the T. cruzi trans-sialidase by Watts et al through peptide mapping after labelling with 2,3-difluorosialic acid. [12].
First general acid/base residue identification: Identified by X-ray crystallography by Crennell et al [20].
First 3-D structure: First determined for the Salmonella typhimurium enzyme by Crennell and Taylor [20].

References

All Medline abstracts: PubMed