C-type lectin-like domains in mammals













C-type carbohydrate recognition domains (CRDs) and C-type lectin-like domains (CTLDs)

The sugar-binding activity in C-type lectins resides in modular domains, which have been designated C-type carbohydrate recognition domains (CRDs).  C-type CRDs have a distinctive protein fold that is shared with other domains that do not necessarily bind sugars.  The designation C-type lectin-like domain (CTLD) has been proposed for this broader group of protein modules.

The search strategy for identification of CTLDs

The residues that determine the CTLD fold form a characteristic sequence motif.  This motif was originally defined in terms of invariant and highly conserved residues, but with the large number of CTLD sequences and structures now available, it is more useful to use statistical patterns to define the motif.  Such profiles have been established in all of the major protein domain databases, including Prosite (PS00615 and PS50041), Pfam (PF00059), InterPro (IPR001304) and SMART (SM00034).  Protein sequences in the UniProt protein database and the Ensembl versions of the human and mouse genomes have been screened for such motifs and the presence of CTLDs is indicated as part of the standard annotation.

Our strategy is to examine all proteins for which profile analysis indicates that they may contain CTLDs.  In some cases, further examination and screening indicates that the profile match is not significant, and these cases are dismissed as not containing CTLDs.  Such false positives usually arise in very cysteine-rich sequences that match with multiple profiles.  After removal of false positives, the remaining proteins are merged to eliminate duplicates in the various databases screened.

Classification of CTLD-containing proteins into groups

The CTLD-containing proteins have been sorted based on two independent sets of criteria.  In one approach, the sequences of the CTLDs are compared and groups are established based on cluster analysis.  Alternatively, the proteins can be sorted based on the organization of domains in the complete polypeptide.  The two approaches give essentially the same results, indicating that members of each group are derived from a common ancestor, which had already acquired the domain architecture that is characteristic of the group.  This analysis has defined 14 groups of proteins, each with unique structural organization.  This section of the animal lectins genomics resource includes information on the structure and function of proteins in each group, as well as annotated sequence alignments, and a comprehensive database for all human and mouse CTLD-containing proteins.  Links are provided to the relevant entries in genomic, nucleotide, protein and structural databases.

Identification of CTLDs likely to function as CRDs

Many CTLD-containing proteins have already been shown to bind sugars and thus to function as lectins.  Using multiple sequence alignment combined with knowledge of how Ca2+ and sugars interact with C-type CRDs, the remaining CTLDs have been screened to identify which ones are likely to bind sugars.  Predictions of which proteins are likely to bind sugars are not absolute, but provide a guide to what activities might be expected.



This page last updated:
Wednesday, 01 January 2014
Animal lectins home
Contact information: This site is supported by:
Kurt Drickamer
Division of Molecular Biosciences
Faculty of Natural Sciences
Imperial College London
Email: k.drickamer@imperial.ac.uk