Supplemental Material

This article contains the following supporting material:

• Figure S1 - Evaluation of the isoform specificity of polyclonal antibodies raised against a peptide located in the C-terminus of EHD1. GFP fusion proteins of EHD1, EHD2, EHD3 and EHD4 were expressed in HEK293 cells and the lysates were immunoblotted with anti-GFP antibodies (A) and polyclonal antibodies raised against a peptide representing the C-terminus of EHD1 (B). Note that the anti-EHD antibody predominantly recognizes EHD1 but is also able to detect EHD3 and EHD4 relatively well.
• Figure S2 - Coimmunoprecipitation of GFP-syndapin I with Flag-EHD1 coexpressed in HEK293 cells. Immunoblot analyses of immunoprecipitates (B, D) and supernatants (A, C). Analysis of the immunoprecipitates with GFP antibodies demonstrate that GFP-syndapin II-l is specifically coimmunoprecipitated along with Flag-mRme-1/EHD1 by using immobilized anti-Flag antibodies (D). Analysis of the supernatants show that they are depleted for Flag-mRme-1/EHD1 (A) and that their GFP-syndapin I levels are reduced upon specific coimmunoprecipitation with EHD1 (C).
• Figure S3 - Evaluation of putative defects of the internalization of transferrin caused upon overexpression of syndapin II-l and EHD1 as well as fragments and mutants thereof. Cells were incubated for 10 min with transferrin-Alexa FluorTM 568, washed, fixed and analyzed, as described in Materials and Methods. As the percentage of cells with inhibited endocytosis, i.e. with a lack of endosomal pattern labelled by transferrin (block), was very small except for overexpression of the G429R mutant, the categories wild-type and reduced transferrin uptake were combined (uptake) to increase the clarity of the figure. Untransfected cells (--), 10.2 ± 8.3% block; GFP, 5.6 ± 2.7% block; GFP-NPF (Sdp II-l), 4.2 ± 0.4% block; GFP-NPF (Sdp II-l) and Flag-EHD1, 6.9 ± 5.0% block; GFP-NPF*** (Sdp II-l), 4.3 ± 0.5% block; Flag-EHD1, 5.6 ± 1.0% block; Flag-EHD1 G65R, 4.8 ± 0.0% block; Flag-EHD1 G429R, 29.4 ± 2.6% block; Flag-EHD1 W485A, 11.5 ± 0.5% block; GFP-EH, 3.7 ± 0.1% block; GFP-EH and Xpress-Sdp II-l, 4.0 ± 2.2% block; Xpress-Sdp II-l, 10.3 ± 1.4% block. For each construct, 204-232 cells were scored.
• Figure S4 - Correlation of amino acid differences within aligned EH domains of Eps15, intersectin and of EHD protein family members with structural and ligand binding data described by de Beer et al. (2000) and revealed in this study. EF hands of all EH domains aligned are underlined. Note that the Eps15-EH domain 1 does not exhibit an EF hand, that the EF hand of the Eps15-EH domain 3 resides in a position different from all other EF hands in the alignment and that the EH domains of EHD proteins show the highest similarities with the EH domain 2 of Eps15. Amino acids that were revealed to contribute to the conserved NPF-binding pocket in the Eps15-EH domain 2 by de Beer et al. (2000) are in bold. Numbers of amino acids listed on top refer to the positions in the Eps15 EH domain 2 solved in complex with the NPF-containing Hrb peptide (de Beer et al., 2000). They include W169 and E170 contacting the asparagine, V151, L155 und W169 accomodating the proline and V151, L155 und L165 contacting the phenylalanine of the NPF motif. These amino acid numbers are underlined in the upper row. Amino acids important for specificity control at the +3 position, i.e. at the position C-terminal of the NPF, revealed for Eps15-EH domain 2 are in red. They include V162, E163, R167, E170 and M203. In Eps15-EH domain 2 they contribute to the interaction with the hydrophobic L following the NPF in the Hrb peptide studied by de Beer et al. (2000). In syndapins, NPF motifs are, in contrast, mostly followed by acidic residues. Amino acid differences of EHD protein EH domains and the EH domains of intersectin or Eps15 are highlighted by green backgrounds, if they are identical in at least three proteins among EHD1, 3 and 4 as well as the C. elegans ortholog and if the residue(s) present at these positions within these four EHD proteins are different from all EH domains of intersectins and Eps15. These positions (corresponding to 154, 163, 170 and 203) are likely to be important for the observed specificity of syndapin NPF motifs for the EH domains of EHD proteins. In two of these positions (154 and 203) EHD2 shows gross differences compared to all other EHD proteins. The E154W and N203S exchanges (white letter on black background) are likely to explain the observed inability of EHD2 to interact with syndapin NPF motifs.