Identification of divergent WH2 motifs by HMM-HMM alignments
© Weiß and Schultz; licensee BioMed Central. 2015
Received: 4 August 2014
Accepted: 14 January 2015
Published: 24 January 2015
The actin cytoskeleton is a hallmark of eukaryotic cells. Its regulation as well as its interaction with other proteins is carefully orchestrated by actin interaction domains. One of the key players is the WH2 motif, which enables binding to actin monomers and filaments and is involved in the regulation of actin nucleation. Contrasting conserved domains, the identification of this motif in protein sequences is challenging, as it is short and poorly conserved.
To identify divergent members, we combined Hidden-Markov-Model (HMM) to HMM alignments with orthology predictions. Thereby, we identified nearly 500 proteins containing so far not annotated WH2 motifs. This included shootin-1, an actin binding protein involved in neuron polarization. Among others, WH2 motifs of ‘proximal to raf’ (ptr)-orthologs, which are described in the literature, but not annotated in genome databases, were identified.
In summary, we increased the number of WH2 motif containing proteins substantially. This identification of candidate regions for actin interaction could steer their experimental characterization. Furthermore, the approach outlined here can easily be adapted to the identification of divergent members of further domain families.
Actin is of vital importance for a wide range of cellular processes. Thus both, its conversion from the globular (G-actin) to the filamentous form (F-actin) and its interaction with other proteins has to be finely tuned. Actin interaction domains play key roles in both processes. This includes, among others, the Wasp-homology domain 2 (WH2) motif , which not only binds to G-actin and sometimes F-actin, but is also involved in the nucleation of new actin fibres . It exists additionally in proteins as single or repeated modules, and some WH2 repeats appear to promote interactions with F-actin [2,3]. Contrasting archetypal domains, which are structurally conserved, the WH2 motif is intrinsically disordered. Only when interacting with partners such as actin or neighbouring domains in autoinhibited multimodular proteins, WH2 motifs take on a defined structure . Probably as a result of this structural variability, also its sequence is extremely divergent. Furthermore, with an originally reported length of about 35 amino acids , WH2 domains are comparably short. These two features make their identification with standard sequence analysis approaches challenging. Possibly as a result of their divergence, WH2 motifs reported in the SMART database  range from 12 to 23 AA underlining the difficulties in their identification. Still, delineation of WH2 motifs in a protein would be a considerable benefit as (i) the function of this protein can be considered as actin binding and (ii) experimental characterization of actin interaction can focus on these specific regions.
One of the most sensitive approaches for the identification of divergent homologs is HHblits, which enables the iterative comparison of a single HMM against a database of HMMs . Thus, when searching against standard protein sequence databases, these have be converted into a HMM database. Usually, this is achieved by sequence based clustering using programs like kClust . Here, we suggest a different, more biologically driven approach. Instead of clustering complete sequence databases de novo, we are relying on annotated orthology relationships. We speculated that the addition of noise by combining sequences from different species will highlight conservation patterns which can be picked up by HHblits. Still, by focusing on orthologs, the domain structure of the encoded proteins should be conserved.
All genomes from the ensembl vertebrate, metazoan, plant, protists and fungi databases, respectively, were analyzed . For each data set, orthology information as annotated by Ensembl was extracted. In the case of the vertebrate data set, the pre-calculated alignments of ortholog groups were downloaded. For the other data sets, each ortholog group was aligned using muscle . Subsequently, a HMM was calculated by the hhblitsdb script of the HHSuite package . As queries, WH2 alignments from SMART , Pfam  and Prosite  were downloaded and translated to HMMs. These were used as input for HHblits searches against the ortholog HMMs databases.
For manual evaluation, a database of 12 Drosophila species currently annotated in Ensembl was built. Following the approaches of domain databases like SMART and Pfam, we used this data set to determine the E-value cutoff. As the first true negative hit had an E-value of 3.5 and the last true positive one of 1.7, we set the gathering cutoff to 2. Of the currently annotated eight WH2 motifs in Drosophila melanogaster six were identified when combining searches with the three query HMMs. Interestingly, all eight were identified when searching against the metazoan data set which also included the Drosophila data. This indicates that sufficient variation within the orthologous groups is needed for the identification of conserved signals. Not unexpectedly, the identified sequences differed between the three query HMMs. Thus, also the quality of the domain alignment influences search results.
Summary of identified WH2 domains (E-value cutoff 2/0.05)
Annotated not found
In further ortholog groups WH2 motifs were not consistently predicted by Ensembl. For these, the WH2 motif is now reliably annotated for all members spanning the according position. A typical example is Spire . So far, its WH2 motif was annotated only for a few species including D. melanogaster. We now identified the motif also in the vertebrate proteins including mouse, rat and human. Lastly, we identified a significant similarity to thymosine-β4 domain, which has a WH2 motif like structure despite the very poorly conserved sequence. Indeed, it is mainly the C-terminus which displays commonly conserved positions, whereas the N-terminus varies strongly (Figure 1). Still, the identification of a significant similarity corroborates a suggested common evolutionary origin .
In summary, our approach was able to identify a substantial number of WH2 motifs so far overlooked in standard genome databases. Admittedly, this is influenced by the chosen E-value cutoff. Still, we think that this is a valid approach, which is also implemented in Pfam as gathering threshold and in SMART as domain specific cutoff . Differences between the analyzed data sets indicate that the degree of conservation versus variation is of importance for the sensitivity of the approach. Thus one might have to adopt the considered species depending on the domain. In general, the advantage of relying on orthologs instead of de-novo clustering is that the sets are based on a biological assumption - common origin based on speciation - and therefore, the domain content should be conserved. In sequence clustering, this will be only achievable by strict length cutoffs, which might influence the results of clustering. Thus, the approach presented here could be easily adapted for the identification of divergent family members of further domains.
JS was partially supported by the DFG SPP 1464: Principles and evolution of actin-nucleator complexes. The publication was partially funded by the German Research Foundation (DFG) and the University of Würzburg in the funding program Open Access Publishing.
- Paunola E, Mattila PK, Lappalainen P. WH2 domain: a small, versatile adapter for actin monomers. FEBS Lett. 2002;513:92–7.PubMedView ArticleGoogle Scholar
- Qualmann B, Kessels MM. New players in actin polymerization–WH2-domain-containing actin nucleators. Trends Cell Biol. 2009;19:276–85.PubMedView ArticleGoogle Scholar
- Renault L, Deville C, van Heijenoort C. Structural features and interfacial properties of WH2, β-thymosin domains and other intrinsically disordered domains in the regulation of actin cytoskeleton dynamics. Cytoskeleton. 2013;70:686–705.PubMedView ArticleGoogle Scholar
- Letunic I, Doerks T, Bork P. SMART 7: recent updates to the protein domain annotation resource. Nucleic Acids Res. 2012;40:302–5.View ArticleGoogle Scholar
- Remmert M, Biegert A, Hauser A, Söding J. HHblits: lightning-fast iterative protein sequence searching by HMM-HMM alignment. Nat Methods. 2011;9:173–5.PubMedView ArticleGoogle Scholar
- Hauser M, Mayer CE, Söding J. kClust: fast and sensitive clustering of large protein sequence databases. BMC Bioinformatics. 2013;14:248.PubMed CentralPubMedView ArticleGoogle Scholar
- Flicek P, Amode MR, Barrell D, Beal K, Billis K, Brent S, et al. Ensembl 2014. Nucleic Acids Res. 2014;42(Database issue):D749–55.PubMed CentralPubMedView ArticleGoogle Scholar
- Edgar RC. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res. 2004;32:1792–7.PubMed CentralPubMedView ArticleGoogle Scholar
- Söding J. Protein homology detection by HMM-HMM comparison. Bioinformatics. 2005;21:951–60.PubMedView ArticleGoogle Scholar
- Finn RD, Bateman A, Clements J, Coggill P, Eberhardt RY, Eddy SR, et al. Pfam: the protein families database. Nucleic Acids Res. 2014;42(Database issue):D222–30.PubMed CentralPubMedView ArticleGoogle Scholar
- Sigrist CJ, de Castro E, Cerutti L, Cuche BA, Hulo N, Bridge A, et al. New and continuing developments at PROSITE. Nucleic Acids Res. 2013;41(Database issue):D344–7.PubMed CentralPubMedView ArticleGoogle Scholar
- Crooks GE, Hon G, Chandonia J-M, Brenner SE. WebLogo: a sequence logo generator. Genome Res. 2004;14:1188–90.PubMed CentralPubMedView ArticleGoogle Scholar
- Toriyama M, Shimada T, Kim KB, Mitsuba M, Nomura E, Katsuta K, et al. Shootin1: a protein involved in the organization of an asymmetric signal for neuronal polarization. J Cell Biol. 2006;175:147–57.PubMed CentralPubMedView ArticleGoogle Scholar
- Shimada T, Toriyama M, Uemura K, Kamiguchi H, Sugiura T, Watanabe N, et al. Shootin1 interacts with actin retrograde flow and L1-CAM to promote axon outgrowth. J Cell Biol. 2008;181:817–29.PubMed CentralPubMedView ArticleGoogle Scholar
- Schultz J, Terhoeven N. The bilaterian roots of cordon-bleu. BMC Res Notes. 2013;6:393.PubMed CentralPubMedView ArticleGoogle Scholar
- Wellington A, Emmons S, James B, Calley J, Grover M, Tolias P, et al. Spire contains actin binding domains and is related to ascidian posterior end mark-5. Development. 1999;126:5267–74.PubMedGoogle Scholar
- Edwards J. Are beta-thymosins WH2 domains? FEBS Lett. 2004;573:231–2. author reply 233.PubMedView ArticleGoogle Scholar
- Ponting CP, Schultz J, Milpetz F, Bork P. SMART: identification and annotation of domains from signalling and extracellular protein sequences. Nucleic Acids Res. 1999;27:229–32.PubMed CentralPubMedView ArticleGoogle Scholar
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.