Sugar-binding F-box proteins
The sugar ligand is shown in dark blue. Protein Data Bank structure ID: 1UMI.
A novel class of carbohydrate recognition domain (CRD) has recently been described in the murine F-box protein Fbs1 (Fbx2). Homologous domains present in F-box proteins from a range of animals may also function as CRDs.
The degradation pathway for misfolded glycoproteins
F-box proteins are key components of the cytoplasmic enzyme system which conjugates ubiquitin, a small polypeptide, on to amino groups in substrate proteins. Ubiquitination regulates protein turnover, location and activity. One important role of ubiquitination is to target proteins to the proteasome, a cytoplasmic enzyme complex which degrades proteins for homeostatic and regulatory purposes. The proteasome also degrades misfolded proteins. Proteins which have been translocated into the endoplasmic reticulum (ER) but fail to fold correctly cannot proceed along the secretory pathway. Misfolded proteins are identified by certain M-type lectins, which recognize the high-mannose N-linked glycans that are added to proteins during or shortly after translocation into the ER. Proteins bound by these M-type lectins are translocated back into the cytoplasm, where they are tagged by the ubiquitin ligase enzyme system to direct them to the proteasome for degradation.
F-box proteins in the ubiquitin ligase system
The ubiquitin ligase system consists of three protein complexes: E1, E2 and E3. E1 activates ubiquitin, E2 transfers ubiquitin on to target proteins, and E3 selects proteins for ubiquitination. An F-box protein is present in E3 and is responsible for substrate selection. Each F-box protein contains a conserved F-box domain, which interacts with other subunits in E3, and a substrate recognition domain, which determines the target specificity of the E3 complex. F-box proteins are classified according to the structure and specificity of the substrate recognition domain. For example, proteins in the Fbw and Fbl classes contain WD-40 and leucine-rich repeat domains respectively, and select ubiquitination targets by recognizing phosphorylated amino acids.
Ligand binding in sugar-binding F-box proteins
The substrate recognition domain in Fbs1 has an ellipsoid shape formed by a sandwich of five-stranded, antiparallel beta-sheets. One end of the ellipsoid is capped by two alpha-helices and is also the location of the domain N- and C-termini. The sugar binding site is located at the opposite end of the domain and is formed by the loops connecting beta-strands 3-4 and 9-10 (left). The binding site is specific for the N-acetylglucosamine (GlcNAc) disaccharide chitobiose. One GlcNAc residue stacks against a tryptophan aromatic ring, while the other inserts a methyl group into a hydrophobic pocket. Both residues make hydrogen bonds to the protein. Specificity for chitobiose enables Fbs1 to bind to misfolded glycoproteins that have been translocated back into the cytoplasm: a GlcNAc disaccharide is at the core of all N-linked oligosaccharides added to proteins in the ER. In folded proteins, these GlcNAc residues interact with the polypeptide and are shielded, but in unfolded proteins they may become exposed and act as a novel sugar-based degradation signal. In contrast, outer residues in N-linked glycans appear to make only minor interactions with Fbs1. The interaction between Fbs1 and the chitobiose core of N-linked glycans is involved in the binding and ubiquitination of physiological ubiquitination targets and subsequent degradation of these proteins by the proteasome.
