PMC:5026484 / 29223-37580 JSONTXT

{"target":"http://pubannotation.org/docs/sourcedb/PMC/sourceid/5026484","sourcedb":"PMC","sourceid":"5026484","source_url":"https://www.ncbi.nlm.nih.gov/pmc/5026484","text":"The second Rab binding site has evolved by gene duplication\nStimulated by the evidence for two Rab binding sites in some bMERB domains (Mical-1, Mical-3 and EHBP1L1), we searched for crystallization conditions of these RBDs with Rabs in a 1:2 stoichiometry. Crystallization conditions were found using a complex of Rab101–175 and the RBD of Mical-1 (residues 918–1067), yielding crystals that diffracted to a resolution of 2.8 Angstrom at a synchrotron X-ray source and the resulting structure indeed showed two molecules of Rab10 bound to Mical-1 (Figure 6a). In addition to the binding site corresponding to the one previously observed in Mical-cL, an additional binding site was identified: Whereas this site is composed of the N-terminal half of the bMERB domain (α-helix 1 and the first half of α-helix 2), the Rab binding site observed in both Mical-1 and Mical-cL comprises the C-terminal half (second half of α-helix 2 and α-helix 3).\n10.7554/eLife.18675.014Figure 6. The two Rab binding sites are highly similar.\n(a) The structure of Mical-1 in complex with Rab10 shows two molecules of Rab10 bound to Mical-1 at different sites. Whereas one Rab protein binds to α-helix 1 and the first half of α-helix 2 (Mical-1918-991, binding site 1, blue), the second molecule of Rab10 binds the second half of α-helix 2 and α-helix 3 (Mical-1994-1060, binding site 2, green). Upon superimposition of both binding sites, the strong similarity becomes obvious and the helices from both binding sites adopt very similar positions. Furthermore, the interactions are highly similar in both cases as can be seen in the close-up view on the right (similar Rab-interacting residues within binding site 1 and 2 are shown in blue and green, respectively). (b) The strong conservation of interacting residues within both halves of the Mical-1 bMERB domain can also be seen in the sequence alignment of the N- and C-terminal halves. Additionally, the alignment shows that α-helix 1 and the first half of α-helix 2 (binding site 1) correspond to the second half of α-helix 2 and α-helix 3 (binding site 2), respectively (the secondary structure is indicated above and below the corresponding sequences, interacting and conserved residues within binding site 1 and 2 are highlighted in blue and green). (c) Whereas the whole bMERB domain of Mical-1 can bind two Rab molecules (left), deletion of either α-helix 1 (middle) or α-helix 3 (right) impairs binding to binding site 1 or 2, respectively. This effect could be shown both by aSEC and ITC (note the enthalpy-driven high-affinity binding site 2 and the entropy-driven lower-affinity binding site 1 that can be clearly distinguished, a schematic drawing of the different constructs is shown within the aSEC graphs). (d) Based on the observations made above, we propose that both binding sites must have evolved from a common ancestor by gene duplication of a 200 bp DNA fragment coding for the original gene product, a single α-hairpin. The fusion lead to the arrangement of the α-helices observed in bMERB domains, with the central α-helix 2 as a continuous connecting helix of both repeats, similar to the architecture of spectrin repeats.\nDOI: http://dx.doi.org/10.7554/eLife.18675.01410.7554/eLife.18675.015Figure 6—figure supplement 1. Evolution of the second binding site.\n(a) Sequence alignments of the N- (blue) and C-terminal (green) halves of human bMERB domains made with Clustal Omega (Sievers et al., 2011). The residues in Mical-1 involved in binding Rab10 are shown in bold letters and the corresponding colors blue and green for binding site 1 and 2, respectively. In order to simplify comparison of the sequences, vertical lines are shown. The approximate positions of α-helices 1, 2 and 3 are also indicated. (b) Phylogenetic tree of the aligned sequences (the N- and C-terminal halves are again colored in blue and green, respectively). Uniprot accession IDs are Q8TDZ2 (Mical-1), Q8N3F8 (Mical-L1), Q8IY33 (Mical-L2), Q7RTP6 (Mical-3), Q6ZW33 (Mical-cL), Q8NDI1 (EHBP1) and Q8N3D4 (EHBP1L1).\nDOI: http://dx.doi.org/10.7554/eLife.18675.015\n10.7554/eLife.18675.016Figure 6—figure supplement 2. Structural comparison of the individual Rab binding sites in Mical-1.\n(a) Cartoon representation of the Rab10:Mical-1:Rab10 complex (centre) and selected close-up views (left and right). Left: Whereas Lys981 and Asn982 within the binding site (BS) 1 contact Asp45 and Gln61 in Rab10, the same residues in the other Rab10 molecule are contacted by Arg1044 and Asp1045 from BS2. Right: The hydrophobic residues Leu956BS1/Leu1011BS2, Val971BS1/Leu1034BS2, Leu975BS1/Val1038BS2 and Val978BS1/Val1041BS2 contacting Ile42Rab10, Ile44Rab10, Phe46Rab10, Trp63Rab10 and Ile74Rab10 (BS1, BS2 and Rab10 denote residues within binding site 1, binding site 2 or Rab10, respectively). (b) Schematic presentation of contacts between Rab10 and BS2 (left) or BS1 (right). Hydrophobic interactions are indicated by black dashed lines, ionic interactions and h-bonds are indicated by orange dashed lines.\nDOI: http://dx.doi.org/10.7554/eLife.18675.016\n10.7554/eLife.18675.017Figure 6—figure supplement 3. Sequence alignment of the bMERB domains examined in this work.\nResidues known from the structures of Rab:bMERB complexes to make contacts with Rab proteins are highlighted in dark green (binding site 1) or dark orange (binding site 2). Conserved residues in other bMERB domains are colored light green or light orange, respectively. The approximate position of α-helices 1, 2 and 3 are indicated above the sequences. Uniprot accession IDs are Q8TDZ2 (Mical-1), Q8N3F8 (Mical-L1), Q8IY33 (Mical-L2), Q7RTP6 (Mical-3), Q6ZW33 (Mical-cL), Q8NDI1 (EHBP1) and Q8N3D4 (EHBP1L1).\nDOI: http://dx.doi.org/10.7554/eLife.18675.017\nUpon closer inspection and alignment, a strong similarity between the two Rab binding sites in Mical-1 became obvious, involving the same/similar residues both within the two different molecules of Rab10 as well as the two binding sites in Mical-1, respectively (Figure 6a). An alignment of the sequences of the corresponding N- and C-terminal halves of all different Micals (Mical-1, Mical-cL, Mical-3, Mical-L1 and Mical-L2), EHBP1 and EHBP1L1 (see Figure 6—figure supplement 1, the example for Mical-1 is shown in Figure 6b) with Clustal Omega (Sievers et al., 2011) highlights the striking similarity between the binding sites and shows the strong conservation of Rab-interacting residues within the two binding sites. A non-exhaustive list and a close-up view of several of these interactions is shown in Table 3 and Figure 6—figure supplement 2, respectively. It should be noted that use of the N- and C-terminal halves of only one of the bMERB domains was not sufficient for Clustal Omega alignment to converge and find the conserved residues within the separate halves. In contrast, the webserver HHrepID (Biegert and Söding, 2008) nicely predicted and aligned the two repeats present in Mical-1 with a p-value of 1.1–5.\n10.7554/eLife.18675.018Table 3. Non-exhaustive list of conserved interactions between Rab10 and the separate binding sites in Mical-1.\nDOI: http://dx.doi.org/10.7554/eLife.18675.018\nConsistent with the localization of the two separate binding sites within the N-terminal and the C-terminal half of the bMERB domain, respectively, deletion constructs lacking either α-helix 1 (Mical960–1067) or α-helix 3 (Mical-1918–1020) displayed a clear 1:1 stoichiometry of Rab binding both in aSEC and ITC experiments (Figure 6c). Furthermore, the ITC data allowed us to clearly allocate the high affinity binding to the C-terminal binding site and the lower affinity binding to the N-terminal binding site.\nIn summary, the strong conservation of interacting residues between both sites as well as the structural conservation of the binding sites lead us to conclude that this family of Rab binding proteins must have evolved via gene duplication (Figure 6d). Furthermore the strong conservation of interacting residues not only between the two separate binding sites in Mical-1, but also between the different bMERB domains (see alignments in Figure 6—figure supplements 1 and 3) suggests that all of these proteins contain a second (possibly low affinity) binding site. Further analysis of the Rab specificity of both sites within these proteins will therefore be of great interest.","divisions":[{"label":"Title","span":{"begin":0,"end":59}},{"label":"Figure caption","span":{"begin":943,"end":5755}},{"label":"Title","span":{"begin":976,"end":1021}},{"label":"Figure caption","span":{"begin":3226,"end":4096}},{"label":"Title","span":{"begin":3279,"end":3316}},{"label":"Figure caption","span":{"begin":4097,"end":5082}},{"label":"Title","span":{"begin":4150,"end":4219}},{"label":"Figure caption","span":{"begin":5083,"end":5755}},{"label":"Title","span":{"begin":5136,"end":5198}},{"label":"Table caption","span":{"begin":6985,"end":7166}}],"tracks":[{"project":"2_test","denotations":[{"id":"27552051-18245125-26907303","span":{"begin":6890,"end":6894},"obj":"18245125"}],"attributes":[{"subj":"27552051-18245125-26907303","pred":"source","obj":"2_test"}]},{"project":"MyTest","denotations":[{"id":"27552051-18245125-26907303","span":{"begin":6890,"end":6894},"obj":"18245125"}],"namespaces":[{"prefix":"_base","uri":"https://www.uniprot.org/uniprot/testbase"},{"prefix":"UniProtKB","uri":"https://www.uniprot.org/uniprot/"},{"prefix":"uniprot","uri":"https://www.uniprot.org/uniprotkb/"}],"attributes":[{"subj":"27552051-18245125-26907303","pred":"source","obj":"MyTest"}]}],"config":{"attribute types":[{"pred":"source","value type":"selection","values":[{"id":"2_test","color":"#93bfec","default":true},{"id":"MyTest","color":"#d9ec93"}]}]}}