Domain structures of mammalian F-box proteins. Domains identified by the Hidden Markov Model algorithms of SMART or PFam include F-box motif (F), WD40 repeat (WD), leucine-rich repeat (L), transmembrane domain (T), F-box-associated domain (FBA), between-ring domain (IBR), domain in carbohydrate binding proteins and sugar hydrolases (CASH), kelch repeat (K), calponin homology domain (CH), domain found in cupin metalloenzyme family (Jmjc), domain present in PSD-95, Dlg, and ZO-1 (PDZ), zinc-binding domain found in Lin-11, Isl-1, and Mec-3 (Lim), HNH nuclease family (HNHc), novel eukaryotic zinc-binding domain (CHORD), and tetratrico peptide repeat (TPR). The following domains were found via the Structural Classification of Proteins (SCOP) database, which can be used to predict protein sequences that can adopt known protein folds: ApaG-like, which is structurally similar to bacterial ApaG; Apolipophorin, the apolipophorin-III-like fold; Ubl, the ubiquitin-like fold; TDL, which is Traf-domain like; RNI-like, which may form structure similar to that of leucine-rich repeats in placental RNase inhibitor; and RCC1, which is a possible regulator of chromatin condensation-1 fold.

Phylogenetic tree depiction of interrelationships between human F-box proteins. The tree is generated from the pairwise ZEGA distances (Abagyan and Batalov 1997) within the set of amino acid sequences comprising the F-box domain only by the neighbor-joining method (Saitou and Nei 1987) as adapted in ICM software (Molsoft LLC; http://www.molsoft.com).

Phylogenetic trees for FBXW, FBXL, and FBA-domain-containing subfamilies of F-box proteins, along with orthologous sequences from D. melanogaster and C. elegans. Only the contiguous portions of the sequence corresponding to the F-box domain followed by the indicated protein interaction domain were included and aligned.