The Carbohydrate-Active Enzyme (CAZy) database is a knowledge-based resource specialized in the enzymes that build and breakdown complex carbohydrates and glycoconjugates. As of September 2008, the database describes the present knowledge on 113 glycoside hydrolase, 91 glycosyltransferase, 19 polysaccharide lyase, 15 carbohydrate esterase and 52 carbohydrate-binding module families. These families are created based on experimentally characterized proteins and are populated by sequences from public databases with significant similarity. Protein biochemical information is continuously curated based on the available literature and structural information. Over 6400 proteins have assigned EC numbers and 700 proteins have a PDB structure. The classification (i) reflects the structural features of these enzymes better than their sole substrate specificity, (ii) helps to reveal the evolutionary relationships between these enzymes and (iii) provides a convenient framework to understand mechanistic properties. This resource has been available for over 10 years to the scientific community, contributing to information dissemination and providing a transversal nomenclature to glycobiologists. More recently, this resource has been used to improve the quality of functional predictions of a number genome projects by providing expert annotation. The CAZy resource resides at URL: http://www.cazy.org/.

Family GH13, also known as the alpha-amylase family, is the largest sequence-based family of glycoside hydrolases and groups together a number of different enzyme activities and substrate specificities acting on alpha-glycosidic bonds. This polyspecificity results in the fact that the simple membership of this family cannot be used for the prediction of gene function based on sequence alone. In order to establish robust groups that show an improved correlation between sequence and enzymatic specificity, we have performed a large-scale analysis of 1691 family GH13 sequences by combining clustering, similarity search and phylogenetic methods. About 80% of the sequences could be reliably classified into 35 subfamilies. Most subfamilies appear monofunctional (i.e. contain enzymes with the same substrate and the same product). The close examination of the other, apparently polyspecific, subfamilies revealed that they actually group together enzymes with strongly related (or even sometimes virtually identical) activities. Overall our subfamily assignment allows to set the limits for genomic function prediction on this large family of biologically and industrially important enzymes.

The recently identified etiological agent of the severe acute respiratory syndrome (SARS) belongs to Coronaviridae (CoV), a family of viruses replicating by a poorly understood mechanism. Here, we report the crystal structure at 2.7-Å resolution of nsp9, a hitherto uncharacterized subunit of the SARS-CoV replicative polyproteins. We show that SARS-CoV nsp9 is a single-stranded RNA-binding protein displaying a previously unreported, oligosaccharide͞oligo-nucleotide fold-like fold. The presence of this type of protein has not been detected in the replicative complexes of RNA viruses, and its presence may reflect the unique and complex CoV viral replication͞transcription machinery.I n 2003, a human coronavirus (CoV) was identified as the causative agent of a form of atypical pneumonia: severe acute respiratory syndrome-CoV (SARS-CoV) (1-5). Coronaviridae have the longest known single-stranded (ss)RNA genome (27-31.5 kb), with a complex genetic organization and sophisticated replication͞transcription cycle (6, 7). Twenty-eight proteins are predicted to be encoded by the SARS-CoV genome (8, 9). The nonstructural (nsp) or ''replicase'' proteins of CoVs are derived from an unusually large replicase gene of Ͼ20 kb that consists of two large ORFs (ORFs 1a and 1b). Translation of this replicase gene from the incoming genomic RNA is the first step in CoV genome expression and includes a Ϫ1 ribosomal frameshift to express the ORF1b-encoded polypeptide. Translation products are the pp1a polyprotein (Ͼ4,000 amino acids) and the C-terminally extended pp1ab polyprotein (Ͼ7,000 amino acids), which are both cleaved by two or three ORF1a-encoded viral proteinases (10). Most of these replicase cleavage products assemble into a membrane-associated viral replication͞ transcription complex. Among other components, this complex includes a set of relatively small polypeptides (nsp6 to nsp11) encoded by the 3Ј region of ORF1a, for which no predicted nor proven function has been assigned. For the mouse hepatitis CoV, several of these cleavage products were reported to colocalize with other components of the viral replication complex in the perinuclear region of the infected cell (11), suggesting their involvement (directly or indirectly) in viral RNA metabolism.As part of a viral structural genomics program (12), we have cloned the 28 gene products of SARS-CoV and expressed them either as full-length proteins or as (predicted) functional domains. The determination of the three-dimensional structures of these gene products is expected to facilitate and accelerate discovery of drugs against this emerging and life-threatening pathogen. Furthermore, structural homology search is becoming a powerful method to infer biochemical and͞or biological function of previously uncharacterized proteins. We report here the crystal structure of nsp9, one the SARS-CoV uncharacterized nonstructural protein, as well as evidence for its function as an ssDNA͞RNA-binding protein. Materials and MethodsCrystallization, Structure Determination, and Refinement. SARS...