Adaptive evolution in two large families of ubiquitin-ligase adapters in nematodes and plants - PubMed (original) (raw)

Adaptive evolution in two large families of ubiquitin-ligase adapters in nematodes and plants

James H Thomas. Genome Res. 2006 Aug.

Abstract

Host-pathogen arms races can result in adaptive evolution (positive selection) of host genes that mediate pathogen recognition and defense. To identify such genes in nematodes, I used maximum-likelihood analysis of codon evolution to survey all paralogous gene groups in Caenorhabditis elegans. This survey found robust evidence of positive selection in two classes of genes not previously implicated in pathogen defense. Both classes of genes encode ubiquitin-dependent proteasome adapters, which recruit diverse substrate proteins for poly-ubiquitination and proteolysis by Cullin-E3 ubiquitin-ligase complexes. The adapter proteins are members of the F-box superfamily and the MATH-BTB family, which consist of a conserved Cullin-binding domain and a variable substrate-binding domain. Further analysis showed that most of the approximately 520 members of the F-box superfamily and approximately 50 members of the MATH-BTB family in C. elegans are under strong positive selection at sites in their substrate-binding domains but not in their Cullin-binding domains. Structural modeling of positively selected sites in MATH-BTB proteins suggests that they are concentrated in the MATH peptide-binding cleft. Comparisons among three Caenorhabditis species also indicate an extremely high rate of gene duplication and deletion (birth-death evolution) in F-box and MATH-BTB families. Finally, I found strikingly similar patterns of positive selection and birth-death evolution in the large F-box superfamily in plants. Based on these patterns of molecular evolution, I propose that most members of the MATH-BTB family and the F-box superfamily are adapters that target foreign proteins for proteolysis. I speculate that this system functions to combat viral pathogens or bacterial protein toxins.

PubMed Disclaimer

Figures

Figure 1.

Figure 1.

Schematic of ubiquitin-targeting system. The top panel shows the SCF1 (Cullin1) complex, which uses Skp-related and F-box proteins as substrate adapters. The domain marked “FTH etc.” varies depending on the specific adapter. The bottom panel shows the SCF3 (Cullin3) complex, which uses BTB proteins as substrate adapters. The domain marked “MATH etc.” varies depending on the specific adapter. (Ub) Ubiquitin.

Figure 2.

Figure 2.

Schematics of F-box and MATH-BTB protein domains. Domain schematics of the main types of proteins analyzed in this paper, including three types of F-box domain proteins and MATH-BTB domain proteins. Domains that bind Cullins are shades of blue, and domains that bind substrate are shades of yellow to brown. F-box family A2 has an additional domain (M for mariner, shown in green) of unknown function that is related to the DNA-binding domain of mariner transposases. The region of highest sequence diversity in each F-box family is labeled hypervariable.

Figure 3.

Figure 3.

_d_N/_d_S results for F-box-FTH genes. Alignment and maximum-likelihood _d_N/_d_S values for a set of 12 unstable F-box-FTH proteins (top panel) and one stable ortholog trio of F-box-FTH proteins (lower panel). The F-box domain and conserved segments (A–G) of the extended FTH domain are marked above the top alignment. The jagged line indicates the position of a possible β-strand. Blue alignment shading is proportional to the sum-of-pairs score for each amino acid residue relative to its aligned column. The histogram section of each panel shows estimated _d_N/_d_S values for each gap-free alignment column, with a red line indicating a value of 1.0. Sites under probable positive selection (P > 0.9) are marked with a red asterisk; the five sites near the C terminus have a smaller asterisk to indicate the possibility of misalignment. Evidence for positive selection remained highly significant when this section was removed prior to analysis (data not shown). To avoid investigator bias, the alignments shown were not hand-modified—a few places with possible artifactual alignment are apparent (e.g., misaligned R residues near the N-terminal end of the PFAM-designated FTH domain in the top panel). In the lower panel, black dots below the alignment indicate sites with an amino acid change in any of the three proteins in the F-box or FTH domains.

Figure 4.

Figure 4.

_d_N/_d_S results for MATH-BTB genes. Alignment and maximum-likelihood _d_N/_d_S values for a set of 10 unstable MATH-BTB proteins from C. elegans (top panel) and proteins from one stable ortholog trio, C. elegans (mel-26), C. briggsae (cb mel-26), and C. remanei (cr mel-26) (lower panel). The MATH and BTB domains are marked above the alignments. Blue alignment shading is proportional to the sum-of-pairs score for each amino acid residue relative to its aligned column. The histogram part of each panel shows estimated _d_N/_d_S values for each gap-free alignment column, with a red line indicating a value of 1.0. Sites under probable positive selection are marked with a red asterisk (P ≥ 0.9) or red square (P ≥ 0.8). In the lower panel, sites with an amino acid change in any of the three sequences in the MATH or BTB domains are marked with a black dot.

Figure 5.

Figure 5.

Structural model of MATH domain. The structure is TRAF6 with bound RANK peptide (PDB 1LB5), colored according to the degree of amino acid conservation among nematode MATH domains (Supplemental Fig. S11). Mapping from nematode MATH domains to TRAF6 is based on a 3D-PSSM structural alignment to the MATH domain of C08C3.2 (see Supplemental text S7). In the space-filled models, long regions of high conservation are dark green, long regions of diversity are yellow, sites of probable positive selection are red, and other regions are khaki. Residues in TRAF6 that were not aligned with C08C3.2 are gray. The bound RANK peptide is shown in gray-blue wireframe. The two large views are rotated nearly 180° from each other and are rotation-centered on the bound peptide (front) and the most conserved regions (back). The small space-filled side view shows the binding cleft in TRAF6 more clearly. The ribbon view shows the eight-stranded β-sandwich structure rotated slightly from the front view in order to show the β-strands more clearly; peptide binding strands are yellow, other strands are dark green, and non-strand regions are gray.

Figure 6.

Figure 6.

F-box family genome positions. The genome positions of 27 highly conserved (red) and 415 unstable (blue) F-box-containing genes in C. elegans. Striking clustering is apparent only for the unstable genes, consistent with evolution by local gene duplication. Highly conserved genes include all 23 ortholog trios plus four genes that did not have specific one-to-one orthologs because of a single duplication or loss in one species. Gene bins are 100 kb in length, and the number of genes in the largest bin on each chromosome is indicated.

References

    1. Aki M., Shimbara N., Takashina M., Akiyama K., Kagawa S., Tamura T., Tanahashi N., Yoshimura T., Tanaka K., Ichihara A., Shimbara N., Takashina M., Akiyama K., Kagawa S., Tamura T., Tanahashi N., Yoshimura T., Tanaka K., Ichihara A., Takashina M., Akiyama K., Kagawa S., Tamura T., Tanahashi N., Yoshimura T., Tanaka K., Ichihara A., Akiyama K., Kagawa S., Tamura T., Tanahashi N., Yoshimura T., Tanaka K., Ichihara A., Kagawa S., Tamura T., Tanahashi N., Yoshimura T., Tanaka K., Ichihara A., Tamura T., Tanahashi N., Yoshimura T., Tanaka K., Ichihara A., Tanahashi N., Yoshimura T., Tanaka K., Ichihara A., Yoshimura T., Tanaka K., Ichihara A., Tanaka K., Ichihara A., Ichihara A. Interferon-γ induces different subunit organizations and functional diversity of proteasomes. J. Biochem. 1994;115:257–269. - PubMed
    1. Altschul S.F., Madden T.L., Schaffer A.A., Zhang J., Zhang Z., Miller W., Lipman D.J., Madden T.L., Schaffer A.A., Zhang J., Zhang Z., Miller W., Lipman D.J., Schaffer A.A., Zhang J., Zhang Z., Miller W., Lipman D.J., Zhang J., Zhang Z., Miller W., Lipman D.J., Zhang Z., Miller W., Lipman D.J., Miller W., Lipman D.J., Lipman D.J. Gapped BLAST and PSI-BLAST: A new generation of protein database search programs. Nucleic Acids Res. 1997;25:3389–3402. - PMC - PubMed
    1. Andrade M.A., Gonzalez-Guzman M., Serrano R., Rodriguez P.L., Gonzalez-Guzman M., Serrano R., Rodriguez P.L., Serrano R., Rodriguez P.L., Rodriguez P.L. A combination of the F-box motif and kelch repeats defines a large Arabidopsis family of F-box proteins. Plant Mol. Biol. 2001;46:603–614. - PubMed
    1. Anisimova M., Bielawski J.P., Yang Z., Bielawski J.P., Yang Z., Yang Z. Accuracy and power of the likelihood ratio test in detecting adaptive molecular evolution. Mol. Biol. Evol. 2001;18:1585–1592. - PubMed
    1. Anisimova M., Bielawski J.P., Yang Z., Bielawski J.P., Yang Z., Yang Z. Accuracy and power of Bayes prediction of amino acid sites under positive selection. Mol. Biol. Evol. 2002;19:950–958. - PubMed

MeSH terms

Substances

LinkOut - more resources