Modeled secondary and tertiary structure of Cas domains. (A) Superposition of a model of the SH3 region of human HEPL with the crystal structure of the same domain from p130Cas. The HEPL sequence alignment with the template had an e value of 4.00e⁻³², 52% sequence identity, 70% positives, and no gaps, comprising residues 13-79 (3-69 in the template sequence). The backbone is shown in green for both proteins. For the side chains, oxygen atoms are red and nitrogen atoms are blue. Carbons in p130Cas are purple and carbons in HEPL are green. (B) Superposition of models of the serine-rich region of human p130Cas (purple ribbon) and human HEPL (green ribbon). HEPL alignment with the template showed an e value of 7.00e⁻⁵³, 28% sequence identity, 48% positives, 8% gaps and contained the 432-591 residue range (4-161 in the template). Because the p130Cas and HEPL helix regions are highly similar, only the HEPL ribbons are shown. Two divergent loop regions are shown with side chains. Oxygen atoms are red and nitrogen atoms are blue. Carbons in p130Cas are purple and carbons in HEPL are white. (C) Predicted Cas family carboxy-terminal secondary structure. Red bar, predicted helical regions (with lighter shades corresponding to lower certainty of prediction); green bar, β -sheets; gray bar, ambiguous.

HEPL localizes to focal adhesions and promotes cell spreading. (A) HA-HEPL (green) colocalizes with paxillin (red) at focal adhesions. Scale bar, 10 μm. (B) Relative spreading (expressed in pixels) in cells transfected with HA-HEPL, HA-HEF1, HA-Δ BioB (a negative control), or HA vector. Bars indicate SE. ***p < 0.001 for increased spreading on fibronectin or plastic versus negative controls. (C) Cells transfected with HA-HEPL, HA-HEF1, or HA-negative vector control were stained with antibody to HA (green) or to Y^397-FAK (red). Scale bars, 10 μm. (D) Two-hybrid measurement of HEPL-SH3 (H-SH3) interaction with FAK. Graph, relative activation of a lacZ reporter, expressed in arbitrary units. Inset, relative activation of an auxotrophic reporter (growth on medium lacking leucine) for the constructs indicated. DBD, DNA-binding domain; AD, activation domain. BioB, PTB, and Raf are used as negative controls. Bars, SE. (E) Cells transfected with HA-HEPL, HA-FAK, HA-FAK, or vector were used for immunoprecipitation with antibody to total FAK, followed by Western blot with antibodies to the proteins indicated. (F) HEPL and HEF1 were immunoprecipitated from lysates of cells held in suspension (Su) or plated on fibronectin (Pl), either untreated or treated with the Src family inhibitor PP2. Western analysis for phospho-tyrosine (Ph-Y) content of immunoprecipitated HEPL and HEF1 is shown.

Definition of Cas family members. (A) Phylogenetic tree of bilateralia (only branches relevant to analysis detailed; dotted lines indicated intermediate branching not shown). Arrow 1, emergence of Cas ancestor, arrow 2, emergence of 4 family members (∼500 million years ago). (B) Cas protein dendrogram indicating rate of divergence. All sequences analyzed were full length. Dm, Drosophila melanogaster; Dp, Daphnia pulex (daphnia); Am, Apis mellifera (bee); Sp, Strongylocentrotus purpuratus (sea urchin); Dr, Danio rerio (zebrafish); Tn, Tetraodon nigroviridis (pufferfish); Xl, Xenopus laevis; Gg, Gallus gallus (chicken); Bt, Bos taurus (cow); Md, Monodelphis domestica (opossum); Cf, Canis familiaris (dog); Mm, Mus musculus (mouse); Rn, Rattus norvegicus (rat); MAm, Macaca mulatta (macaque); Pt, Pan troglodytes (chimpanzee); Hs, Homo sapiens. (C) Domain structure of Cas family proteins, including SH3 domain (SH3) preceded by a short region with no defined functional elements (black box), SH2-binding site rich domain (SH2bd), serine-rich region encompassing α-helices (in p130Cas; SRR), and conserved C-terminal (C-term) domain.

HEPL is most abundant in blood cells and lung tissue. (A) Relative abundance of HEPL transcript assessed by RT-PCR in cDNA prepared from primary human tissues, expressed in arbitrary units (AU). Four human cell lines are included for reference of expression levels. (B) Expression of HEPL and other Cas mRNAs in human cell lines. cDNA made with 100 ng (shown) or 20 ng (similar results; not shown) of total RNA of each sample was used for relative quantification of the transcripts using Taqman assays. The figure displays intensity-coded cycle threshold (Ct) values, reflecting the number of PCR cycles required for a significant fluorescence level above the baseline: Ct values of ≥35 indicate little or no gene expression. (C) Detection of HEPL and other Cas transcripts in nontransformed, immortalized, and transformed ovarian cell lines and tumors; data prepared as in B. (D) HA-HEPL or HA-fused to a negative control protein (Materials and Methods) were transfected into 293T cells, and whole cell lysates (WCL) were used directly for Western analysis with antibodies to HEPL or actin (as loading control) or for immunoprecipitation (IP) with antibody to HA before Western analysis with anti-HEPL. HEPL doublet seen preferentially in WCL over IP likely reflects a population of mixed HA-HEPL and HEPL with HA cleaved because of proteolysis. (E) Antibody to HEPL was used to probe immunoprecipitates from cell lysates as indicated; WCL of 293T cells transfected with HA-HEPL is a sizing control. (F) IPs of K562 or HOP-62 cells transfected with siRNA to HEPL (siHEPL) or control scrambled (Scr) siRNA were probed with antibody to HEPL; WCL from 293T expressing HA-HEPL provides a sizing control in each case. (G) Western blots of 293 cells expressing HA-HEPL (HP), -HEF1 (H1), -p130Cas (p130), or -negative control (−), sequentially probed with antibodies indicated.

Depletion of HEPL influences FAK activation, cell spreading, and motility. (A) Relative spreading (expressed in pixels) in cells transfected with siRNA to HEPL, HEF1, or scrambled control (Scr). Bars, SE. **p < 0.01; ***p < 0.001; ns, nonsignificant (p > 0.05). HEPL-depleted cells staining weakly (20%) or more intensely (80%) for Y³⁹⁷-FAK are presented grouped, and separately. For this and subsequent panels in this figure, a Dharmacon Smartpool containing four HEPL targeting siRNAs was used. We have separately confirmed results using individual deconvoluted siRNAs to HEPL (Supplementary Figure 2) and shown that siHEPL1 and siHEPL4, but not siHEPL2 and siHEPL3, deplete HEPL and generate equivalent results. (B and C) Confocal images of HOP62 cells transfected with siRNA to HEPL, HEF1, or scrambled control (Scr) were stained with antibody to Y³⁹⁷-FAK (B) or paxillin (C). Two different representative HEPL-depleted cells, corresponding to the 20 and 80% phenotypes, are shown. All scale bars, 20 μm. (D) The migration of individual cells was tracked in movies taken from 36 to 48 h after treatment with siRNAs as indicated. Cells were binned into different groups based on average velocity over 12 h. For each group ∼80 cells were analyzed in two independent experiments. (E) The ability of siRNA-treated cells to maintain directional movement was calculated from the movies analyzed in D. D/T indicates the average radial distance (D) individual siRNA-treated cells move from position at the beginning of the assay in comparison to the total (T) distance the cells move during the period of observation. ***p < 0.001; other values not significant versus Scr. (F) Proliferation of K562 cells treated with siRNA to HEPL (siHEPL) or control scrambled (Scr) siRNA, as assessed by Alamar blue staining 1, 2, or 3 d after depletion, with data represented using Tukey box-plots. Statistical significance was assessed by use of a GEE-estimated model assuming a Gamma distribution with log link and an exchangeable correlation over time. (G) Guava analysis of cell cycle compartmentalization of Scr or siHEPL-transfected cells analyzed in F. Note that the high percentage of cells in S is characteristic of K562 cells. (H) Analysis of apoptosis in K562 cells treated with siRNA to HEPL or control scrambled (Scr) siRNA, or with dasatinib as a positive control. Apoptosis detected by Apoptag is reported in arbitrary units (AU); difference between dasatinib and other samples is statistically significant (p < 0.001). (I) Western blot of full-length and cleaved (**) gelsolin in cells transfected 3 d previously with Scr or HEPL siRNA, or treated with dasatinib or etoposide as positive controls to induce apoptosis. β-actin provides a loading control. Similar results were obtained for PARP cleavage (not shown).