TEM4 is a Rho-specific guanine nucleotide exchange factor.

<p>A, A schematic representation of the domain structure of TEM4. <i>TEM4</i> encodes a 2063 amino acid protein that, in addition to the DH and PH domains, contains extensive N-terminal sequences (residues 1-1059) with no identifiable domains or motifs, and a C-terminal domain containing a predicted β-propeller fold (residues 1723–2039) likely consisting of seven WD40-related repeats <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0041876#pone.0041876-Kelley1" target="_blank">[24]</a>, <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0041876#pone.0041876-Chen1" target="_blank">[60]</a>, <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0041876#pone.0041876-Sondek1" target="_blank">[61]</a>. The PH domain is split by an ∼60 amino acid α-helical insert. B, TEM4 is expressed in mammalian cell lines of endothelial and non-endothelial cell lineage. C–G, Guanine nucleotide exchange of RhoA (C), RhoB (D) or RhoC (E) (2 µM) was assessed in the presence (red trace) or absence (black trace) of the isolated DH-PH domain of TEM4 (10 nM). Guanine nucleotide exchange of Rac1 (F) and Cdc42 (G) (2 µM) was assessed in the presence of the DH-PH domains of TEM4 (50 nM; red trace) and PREX1 (200 nM; green trace). Intrinsic activities of Rac1 and Cdc42 are shown as a black trace. Arrows indicate time of GEF addition. Data shown are representative of three independent experiments. Five µg of each protein used in GEF assays were resolved by SDS-PAGE and visualized by Coomassie Blue dye (inset D). H, Structure-based sequence alignment of the TEM4 PH domain. A homology model of the PH domain of TEM4 (residues 1286–1476) was generated using the protein homology/analogy recognition engine² (Phyre²) structure prediction server and then structurally aligned with the determined structures of DH-associated PH domains from SOS1 (PDB: 1AWE) <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0041876#pone.0041876-Zheng1" target="_blank">[62]</a>, TIAM1 (PDB: 1FOE) <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0041876#pone.0041876-Worthylake1" target="_blank">[63]</a> and TRIO (PDB: 1NTY) <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0041876#pone.0041876-Skowronek1" target="_blank">[64]</a> using the Vector Alignment Search Tool (VAST) <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0041876#pone.0041876-Gibrat1" target="_blank">[65]</a>. The assignments of beta strands (blue arrows, β1–β7) and the C-terminal α-helix (yellow cylinders, αC) are shown for each PH domain sequence. Secondary structure analysis predicts a ∼70 residue α-helical region (red) is inserted in the β5/β6 turn of the TEM4 PH domain. Residues within TEM4 likely utilized for phospholipid binding are marked (arrowheads). A highly conserved tryptophan residue within αC of PH domains (phenylalanine in TEM4 and Tiam1) is indicated (star). Dots mark every 10th residue. I, Protein-lipid overlay assay. Purified GST-DH-PH fusion protein was incubated with a lipid array and then detected with an anti-GST antibody. <i>LPA</i>, lysophosphatidic acid; <i>LPC</i>, lysophosphocholine; <i>PC</i>, phosphatidylcholine; <i>PS</i>, phosphatidylserine; <i>PA</i>, phosphatidic acid; <i>PE</i>, phosphatidylethanolamine; <i>S1P</i>, sphingosine 1-phosphate; <i>PI</i>, phosphatidylinositol; <i>PI(3)P</i>, phosphatidylinositol 3-phosphate; <i>PI(4)P</i>, phosphatidylinositol 4-phosphate; <i>PI(5)P</i>, phosphatidylinositol 5-phosphate; <i>PI(3,4)P₂</i>, phosphatidylinositol 3,4-biphosphate; <i>PI(3,5)P₂</i>, phosphatidylinositol 3,5-biphosphate; <i>PI(4,5)P₂</i>, phosphatidylinositol 4,5-biphosphate; <i>PI(3,4,5)P₃</i>, phosphatidylinositol 3,4,5-triphosphate.