Structural genomics of Caenorhabditis elegans: Crystal structure of the tropomodulin C‐terminal domain

The actin filament, an essential part of cytoskeleton, is associated with a number of functions including cellular shape and motility, cell integrity and division, endocytosis, cytokinesis, and muscle contraction. The supramolecular organization of the filament includes actin-binding proteins that regulate filament properties. The actin filament is polar, so that the barbed end is distinct from the pointed end. Tropomodulin is the pointed-end capping protein which maintains the length and dynamics of the actin filament.1 The protein was originally isolated by Fowler2 and shown to inhibit the binding of tropomyosin to actin.3 Tropomodulin's capping activity is dependent upon its association with tropomyosin. In the absence of tropomyosin, capping activity of tropomodulin is greatly reduced.1 The domain structure of the chicken erythrocyte tropomodulin has been thoroughly studied.4 Tropomodulin consists of two major domains: the N-terminal domain interacts with the N-terminal region of tropomyosin,5 whereas the C-terminal domain binds to the actin filament pointed end.6 Since crystallization of the full-length protein was unsuccessful, the crystal structure of the tightly folded C-terminal domain has been determined.7 The C-terminal domain adopts the leucine-rich repeat (LRR) fold,8 a motif frequently present in protein–protein interactions. The tropomodulin protein 1, isoform A, from Caenorhabditis elegans (Tmd1) was identified in the genome-sequence project9 and a comparison with the chicken erythrocyte tropomodulin shows only 34% sequence identity. The crystal structure of the C-terminal domain of Tmd1 (cTmd1) was determined as part of the Structural Genomics of C. elegans Project.10 The structure and putative function of cTmd1 are highly homologous to the evolutionary distant C-terminal domain of the chicken erythrocyte tropomodulin. The protein production, crystallization, and preliminary X-ray analysis have been reported previously.11 Briefly, the crystals are orthorhombic, space group P212121, a = 31.71 Å, b = 50.62 Å, c = 107.13 Å, with one protein molecule in the asymmetric unit. A native diffraction data set to 1.8 Å resolution was collected from one crystal at 100 K at the Advanced Photon Source (APS), beamline 22-ID. The wavelength of the incident radiation was 0.984 Å. The data were integrated and scaled with HKL200012 (Table I). The structure was solved by molecular replacement with Amore13 using the homologous structure of the chicken tropomodulin C-terminal domain as a search model [Protein Data Bank (pdb) code 1IO0]. The automated refinement procedure ARP/wARP14 was employed to retrace the protein chain and to build the new amino acid sequence. The model was further refined in the Crystallography & NMR System (CNS)15 and interactive sessions of model building were performed in O.16 The final model consists of 167 amino acid residues and 131 water sites; the crystallographic R-factor is 20.9% and the R-free is 23.4% with no cutoffs. The structure was validated in Whatcheck17 and deposited in Protein Data Bank under the code 1PGV. The polypeptide chain of cTmd1 consists of residues 196–392. The chain is well defined in the 2Fo–Fc map for the range 222–388. A large part of the N-terminus, including residues 196–221, is completely disordered, which was also observed in the homologous structure 1IO0. The C-terminus of cTmd1 has six additional residues and four of them (389–392) are also disordered. The overall fold and secondary structure of cTmd1 are highly similar to the chicken tropomodulin C-terminal domain [Fig. 1(a)] and the atomic coordinates of the corresponding 165 Cα atoms between the two structures superimpose with a root-mean-square (RMS) deviation of 1.24 Å. There are two residues (Arg283 and Gly284) in the third helix of cTmd1 with a significant deviation from the regular α-helix. Top: Superposition of the C. elegans and G. gallus (chicken) backbones. The C. elegans ribbon is colored by structural motif and has the secondary structural elements labeled. The chicken ribbon drawing is narrower and colored silver. Middle: Sequence alignment and information content. Only the C. elegans and chicken sequences are shown with polar residues red, hydrophobic residues black, and identities between the two boxed in gray. The information content is calculated from the BLAST alignment of 50 tropomodulin C-terminal domain sequences in the non-redundant database. The deeper blue boxes above the sequence indicate greater conservation. The labeled secondary structure elements and motifs are also shown. Each LRR motif sequence grouping begins with the pattern xLxxLxLxxNx. Bottom: Structure and information content. A front view (as above) and a back view (rotated 180° about Y) of the ribbon and surface representation of the C. elegans structure. The coloring matches the information content of the sequence alignments, with the deeper blue indicating a higher degree on sequence conservation. Representative highly conserved residues are labeled. The conserved residues form a large surface cluster on the front side. The canonical LRR fold of cTmd1 includes five β-α hairpin motifs of 28–31 residues each. The β-strands form a parallel β-sheet on one side of the molecule, whereas the helices run in a parallel array on the opposite side of the molecule (Fig. 1, Top). The LRR consensus sequence xLxxLxLxxNx… starts at the residue Asp237; x is any amino acid, L is usually leucine or a hydrophobic residue, N can be asparagine, methionine, threonine, or glutamine. Apparently, the first α-helix (223–233) of cTmd1 is not part of any LRR motif but it packed with the helical array. The last helix (363–386) is considerably longer and protrudes from the nearly box-shaped molecule. A number of crystal structures of LRR proteins are now available in the PDB with the LRR motif containing 20–30 residues and with 4–15 motif repeats. Although the secondary structure of the basic LRR motifs may adopt a simpler β hairpin form, the exposed parallel β-sheet is always conserved. There are two structures of the LRR ribonuclease inhibitor in complex with other proteins, one with ribonuclease A (pdb code 1DFJ),18 and another with angiogenin (1A4Y).19 Here, the ribonuclease inhibitor binds to the target proteins essentially through its C-terminal part and the conserved β-sheet is only partially involved. In the structure of U2B-U2A complex (1A9N)20 the LRR protein U2A interacts with the target protein U2B largely through the β-sheet. This mode of interaction was assumed for the proposed model of actin filament pointed-end capping by tropomodulin.7 We have aligned 50 tropomodulin sequences (Fig. 1, Middle) and projected the conserved residues in a color code to the molecular surface of cTmd1 (Fig. 1, Bottom). The most conserved residues are clustered in the β-surface. It seems plausible that a large protein area is conserved for potential binding to the target. However, the target recognition and energetically important binding contacts are often based on a smaller motif, for instance a relatively short peptide sequence or a group of several residues.19 The model of actin filament pointed-end capping by tropomodulin therefore requires more experimental evidence to be better established. We thank Dr. Stephen Foundling for great help in data collection. The diffraction data were collected at beamline 22-ID in the facilities of the South East Regional Collaborative Access Team (SER-CAT) at the Advanced Photon Source (APS), Argonne National Laboratory. Supporting institutions may be found at www.ser.anl.gov/new/index.html. Use of the APS was supported by the US Department of Energy, Basic Energy Sciences, Office of Science, under Contract No. W31-109-Eng-38.