|Home | About | Journals | Submit | Contact Us | Français|
For over a decade, p130Cas/BCAR1, HEF1/NEDD9/Cas-L, and Efs/Sin have defined the Cas (Crk-associated substrate) scaffolding protein family. Cas proteins mediate integrin-dependent signals at focal adhesions, regulating cell invasion and survival; at least one family member, HEF1, regulates mitosis. We here report a previously undescribed novel branch of the Cas protein family, designated HEPL (for HEF1-Efs-p130Cas-like). The HEPL branch is evolutionarily conserved through jawed vertebrates, and HEPL is found in some species lacking other members of the Cas family. The human HEPL mRNA and protein are selectively expressed in specific primary tissues and cancer cell lines, and HEPL maintains Cas family function in localization to focal adhesions, as well as regulation of FAK activity, focal adhesion integrity, and cell spreading. It has recently been demonstrated that upregulation of HEF1 expression marks and induces metastasis, whereas high endogenous levels of p130Cas are associated with poor prognosis in breast cancer, emphasizing the clinical relevance of Cas proteins. Better understanding of the complete protein family should help inform prediction of cancer incidence and prognosis.
The Cas (Crk-associated substrate) scaffolding protein family contains three defined members: p130Cas/BCAR1 (Sakai et al., 1994 ; Brinkman et al., 2000 ); HEF1/Cas-L/NEDD9 (Kumar et al., 1992 ; Law et al., 1996 ; Minegishi et al., 1996 ), and Efs/Sin (Ishino et al., 1995 ; Alexandropoulos and Baltimore, 1996 ). Elevated expression of p130Cas/BCAR1 has been linked to poor prognosis in breast cancer (van der Flier et al., 2000 ), whereas overexpression of HEF1/NEDD9/Cas-L recently been found to potently induce metastatic melanoma (Kim et al., 2006 ). Mechanistically, the best-studied functions of the Cas family proteins include regulation of attachment-dependent survival signaling or anoikis and regulation of cell motility and invasion, although there is evidence for additional roles for some of these proteins in control of cell cycle, growth factor signaling, cell differentiation, and bacterial and viral infection (reviewed in Defilippi et al., 2006 ; Singh et al., 2007 ). These many functions reflect the ability of the Cas proteins to interact with multiple partner proteins, as the predominant structural feature of Cas proteins is their possession of numerous protein interaction domains (discussed in O'Neill et al., 2000 ), allowing them to act as scaffolding proteins for different functional complexes.
An important current issue in understanding cancer pathogenesis is that of why different oncogenes and tumor suppressors are selectively targeted in tumors arising from different tissue sources. For example, although elevation of HEF1/Cas-L/NEDD9 induces metastasis in melanomas, reduced levels of the same gene have been reported in breast cancers that metastasize aggressively to the lung (Minn et al., 2005 ; see also O'Neill et al., 2007 for discussion). Undoubtedly, the differing physiology and complement of expressed genes in differing precursor cell types imposes distinct requirements for the type of genetic or epigenetic change required to make a cell cancerous. For protein families, another relevant issue is likely to be that for a given cell type, the expression of one family member may condition the impact of modulating the expression of a paralogous family member with overlapping biological activities. The complexity of cellular signaling networks currently emerging through systems-level analysis (Mak et al., 2007 ) emphasizes the importance of exactly defining the composition, expression, and functional properties of protein family groups.
In this study, we have identified a previously unreported but evolutionarily conserved member of the Cas group, which we have termed HEPL (HEF1-Efs-p130Cas-like). We show that the HEPL mRNA and protein are expressed in cultured cell lines and tumors, and that HEPL has biological activities similar to those of other family members in influencing cell attachment and movement. The identification of HEPL provides an important context for further studies of this increasingly important protein group.
Details of sequence collection and processing are provided in the legend to Supplementary Figure 1. Dendrograms showing family relationships were displayed using the Treeview program. The HEPL SH3 domain was modeled using as template the high-resolution crystal structure (1.1Å) of the SH3 domain of p130Cas (PDB code 1WYX; Wisniewska et al., 2005 ). The rat p130Cas structure (PDB code 1Z23 (Briknarova et al., 2005 ) was used to model the four-helix bundle region. Homology modeling was initiated using a multiple-round PSI-BLAST (Altschul et al., 1997 ) sequence search using HEPL and p130Cas sequences to build profiles. The profiles were used to identify suitable templates in the Protein Data Bank (PDB; Berman et al., 2000 ). The profile/template match was refined using secondary structure predictions from PSIPRED (Jones, 1999 ). Conserved backbone and side chain residues were copied from the template structure, whereas divergent residues were rebuilt using SCWRL3 (Canutescu et al., 2003 ). LOOPY (Xiang et al., 2002 ) was used to build loops at points of insertion and deletion. Molecular graphics and 3D structural manipulation was performed using Chimera (Pettersen et al., 2004 ). The Cas multiple sequence alignment was overlaid with secondary structure prediction rendered by MolIDE (Canutescu and Dunbrack, 2005 ).
Total RNA was isolated using an RNeasy kit (Qiagen, Chatsworth, CA). Contaminating DNA was removed using TURBO DNA-free (Ambion, Austin, TX). RNA was quantified using the Agilent 2100 BioAnalyzer (Agilent Technologies, Santa Clara, CA) in combination with a RNA 6000 Nano LabChip (Agilent Technologies). See Supplementary Tables 1 and 2 for technical details of PCR assays. Ambion's First Choice human total RNA survey panel was used as a source of RNA from 20 different normal tissues.
HEPL was cloned using conventional molecular biology techniques by combining sequences from Human MGC verified full-length cDNA (Clone 5205865, Open Biosystems, Huntsville, AL) and human genomic DNA. Hemagglutinin (HA)-epitope tagged HEF1, HEPL, FAK, and negative controls (empty vector or Δ BioB, an extensively truncated Escherichia coli BioB) were expressed from pcDNA3.1-6HA for transfections. Cell lines were cultured under standard conditions, in DMEM or in RPMI-1640 plus 10% fetal bovine serum (FBS) supplemented with antibiotics, as specified by the ATCC (Manassas, VA). Scrambled (control) small interfering (siRNA) and siRNA duplexes against HEPL (NM_020356) and HEF1 were made by Dharmacon Research (Lafayette, CO). HEPL-directed siRNAs were used both as a Smartpool and four individual deconvoluted sequences, as described in Results. Plasmid transfections were done using LipofectAMINE-Plus reagent (Invitrogen, Carlsbad, CA) and siRNA transfections using the Cell Line Nucleofector Kit V from Amaxa Biosystems (Gaithersburg, MD).
Rabbit polyclonal antibody to HEPL was generated using a peptide corresponding to HEPL amino acids 773-786 (by Zymed Laboratories, San Francisco, CA). Antibody was purified from sera using the NAb Protein A Spin Purification Kit (Pierce Biotechnology, Rockford, IL). Other antibodies included anti-HA mAb (Santa Cruz Biotechnology, Santa Cruz, CA), anti-paxillin and anti-p130Cas (BD Transduction Laboratories, Carlsbad, CA), anti-HEF1 (2G9; Pugacheva and Golemis, 2005 ), anti-FAK[pY397] (Biosource, Nevelle, Belgium), anti-gelsolin (BD Biosciences, San Jose, CA), Alexa Fluor 488– and 568–conjugated anti-mouse (Molecular Probes, Eugene, OR), and anti-mouse and anti-rabbit IgG antibodies conjugated to HRP (Amersham Biotech, Buckinghamshire, England). For immunoprecipitations, transfected cells were lysed in M-PER Mammalian Protein Extraction Reagent (Pierce Biotechnology) and immunoprecipitated with either anti-HA or anti-HEPL Abs, using Immobilized Protein A/G Agarose (Pierce Biotechnology). To establish HEPL and FAK interaction, HA-epitope tagged HEF1, HEPL, FAK, and negative control (Δ BioB, an extensively truncated E. coli BioB) were expressed from pcDNA3.1-6HA for transfections in 293T cells and immunoprecipitated with anti-FAK mAb, clone 4.47 (Millipore, Bedford, MA).
To study cell adhesion–dependent tyrosine phosphorylation, trypsinized HOP-62 cells were either maintained in suspension in serum-free medium for 45 min at 37°C or subsequently were replated on fibronectin (4 μg/cm2; Chemicon International, Temecula, CA)-coated dishes for 30 min. Experiments were performed in parallel in the presence or absence of 10 μM PP2 (Calbiochem, San Diego, CA). Cell lysates were prepared using M-PER Mammalian Protein Extraction Reagent supplemented with protease inhibitors and Halt Phosphatase Inhibitor Cocktail (Pierce Biotechnology), immunoprecipitated with antibodies to HEPL or HEF1, and immunoblotted with anti-phosphotyrosine mAb (BD Transduction Laboratories).
The modified Interaction Trap form of two-hybrid system was used to study HEPL protein interactions, using reagents and approaches as described in (Serebriiskii et al., 2002 ). The LexA-fused HEPL SH3 domain (aa 1-148) was used to assess interactions with B42 activation domain-fused FAK C-terminus (aa 688-997). LexA fused to the SHC PTB domain and to a B42-ΔBioB, and B42-fused Raf and B42-ΔBioB were used as nonspecific negative controls. Expression of all protein fusions was analyzed by Western blot.
Cells were fixed in 4% paraformaldehyde for 10 min, permeabilized in 0.2% Triton X-100 for 5 min and blocked with 3% BSA in PBS. After incubation with primary antibodies, cells were stained with either Alexa Fluor 488– or 568–conjugated secondary antibodies. Epifluorescence microscopy was performed using an inverted Nikon TE300 microscope (Melville, NY). Confocal microscopy was performed using a Radiance 2000 laser scanning confocal microscope (Carl Zeiss, Thornwood, NY). All images were acquired as 12-bit images with a Spot RT monochrome camera (Diagnostic Instruments, Sterling Heights, MI). For cell spreading analysis, cells were transfected with indicated plasmids or siRNAs for 18–48 h before fixation, as indicated. Anti-paxillin mAb was used to mark focal adhesions and outline cells. Cell area measurements were made using MetaMorph or MetaVue software (Molecular Devices, Universal Imaging, Downingtown, PA) software to score pixels within cell perimeters.
To measure motility, movement of siRNA-treated HOP62 cells plated in six-well tissue culture dishes was monitored with a Nikon TE300 microscope using 10× NA 0.25 PlanA objective, and images were collected with CCD video camera (Roper Scientific, Trenton, NJ) at 20-min intervals over a 12-h period and then digitized and stored as image stacks using MetaMorph software. Velocity and persistence of migratory directionality (D/T) were determined by tracking the positions of cell nuclei using the Track Point function of MetaMorph.
Apoptosis was measured using an APOPercentage apoptosis assay kit (Biocolor, Belfast, Northern Ireland, United Kingdom) and Western blot to measure appearance of cleaved gelsolin. Cell cycle compartmentalization was measured using a Guava Personal Cell Analysis (PCA) System (Guava Technologies, Hayward, CA). Treatment for 48 h with 200 μM etoposide (Sigma-Aldrich, St. Louis, MO) or 10 nM dasatinib (a gift of Dr. Andrew Godwin) was used as positive control for apoptosis assays. All calculations of statistical significance were made using the GraphPad InStat software package (San Diego, CA) and STATA software (StataCorp, College Station, TX). Approaches included unpaired t tests, ANOVA analysis, and generalized linear models estimated using generalized estimating equations (GEE).
Using the p130Cas, HEF1, and Efs protein and mRNA sequences in reiterative BLAST analysis against genomic sequences and EST resources, we searched for Cas-related sequences in an evolutionarily diverse group of organisms (Figures 1, A and B), and Supplementary Figure 1). No protein strongly related to the Cas family was identified in Saccharomyces cerevisiae or Caenorhabditis elegans, whereas a single ancestral family member was detected in arthropods, echinoderms, and primitive chordates. The family branches in gnathostomes to produce the three previously characterized mammalian family members. Unexpectedly, we detected a completely novel member of the Cas family that was conserved as an ortholog group from gnathostomes through mammals. We have designated this ortholog group as HEPL. Dendrogram analysis indicates that the HEPL does not diverge at a significantly more rapid rate than other Cas branches, suggesting that it maintains a biological function, rather than representing a pseudogene (Figure 1B).
Human HEPL localizes to chromosome 20q13.31 and is annotated in Unigene as C20orf32. Comparison of the human HEPL protein sequence with the other three human Cas family members (Supplementary Figure 1) shows overall identity with other family members up to 26% and similarity up to 42%. Human HEPL is 786 amino acids (aa), versus 870 aa for p130Cas, 834 aa for HEF1, and 561 aa for Efs. For the three well-studied Cas proteins, a highly conserved amino-terminal SH3 domain is followed by a moderately conserved region encompassing multiple SH2-binding sites, a serine-rich region shown to encompass a four-helix bundle in p130Cas (Briknarova et al., 2005 ), and a well-conserved carboxy-terminal domain that contributes to focal adhesion targeting (Nakamoto et al., 1997 ; Harte et al., 2000 ; O'Neill and Golemis, 2001 ; Figure 1C). Although the HEPL proteins maintain a recognizable SH3 domain, this domain has the lowest overall similarity to other SH3 domains in the Cas group. Human HEPL has a limited number of candidate SH2-binding sites, in contrast to p130Cas and HEF1. The HEPL carboxy-terminus has a short region of detectable Cas family homology (res ~670 to the carboxy-terminus), but otherwise lacks obvious similarity at the level of primary sequence. Flanking the carboxy-terminal domain, HEPL also lacks a YDYVHL sequence conserved among the other three Cas family proteins (bold, Supplementary Figure 1) that is an important binding site for the Src SH2 domain (Tachibana et al., 1997 ), possibly suggesting reduced functionality.
To better assess whether the predicted HEPL proteins maintain important features of the Cas family, we used molecular modeling to compare the Cas proteins based on predicted secondary and tertiary structure, using structures of p130Cas as templates (Briknarova et al., 2005 ; Wisniewska et al., 2005 ). Figure 2A demonstrates that HEPL and p130Cas are predicted to fold almost identically within the SH3 domain. Further, despite only 28% primary sequence identity, the predicted secondary structure for residues 432-591 of HEPL is extremely similar to that for residues 449-610 for p130Cas, implying a well-conserved fold (Figure 2B). At present, no adequate template exists in PDB to create a tertiary model for the Cas carboxy-terminus. However, comparison of the predicted secondary structure for the four Cas proteins reveals a strikingly similar periodicity of α -helices and β-sheets (Figure 2C) that is again compatible with the idea of a conserved tertiary structure.
The evolutionary conservation of HEPL suggested that it encoded a functional protein product rather than an unexpressed pseudogene. To test this directly, we first used quantitative RT-PCR to analyze HEPL expression in mRNAs prepared from 20 human tissues (Figure 3A). HEPL was most abundant in lung and spleen and was detected at lower levels in additional tissues. These results were in accord with online resources in NCBI/Unigene (Wheeler et al., 2004 ) that estimate the relative abundance of mRNAs based on their frequency of isolation in sequencing of tissue-specific libraries (results not shown). We next used quantitative RT-PCR to analyze mRNAs prepared from a panel of commonly used laboratory cell lines derived from diverse cell lineages (but enriched in carcinomas and leukemias, in accord with the expression prediction), to establish the general abundance of HEPL mRNA in cultured cells.
HEPL mRNA was detected in the majority of the cell lines (15 of 26), with highest levels of HEPL in the leukemia and ovarian cell lines (Figure 3B). As reference, HEF1 and p130Cas were readily detected in most of the cell lines examined, although p130Cas was not detected in most of the lymphoma/leukemia cell lines analyzed. In contrast, although very abundant in one breast carcinoma cell line (T47D), the Efs mRNA was only detected in 6 of the 26 cell lines assessed. Because increased expression of some Cas proteins has been linked to cancer progression, we investigated the relative expression of HEPL mRNA in a series of normal primary human ovarian surface epithelial (HOSE) cells, human SV40-immortalized ovarian (HIO) cell lines, and primary ovarian tumor tissue (Figure 3C). From this preliminary analysis, HEPL mRNA expression levels did not correlate significantly with ovarian transformation status.
To analyze HEPL at the protein level, we cloned the gene and prepared antibody against HEPL-derived peptide sequences that specifically recognized overexpressed epitope-tagged HEPL (Figure 3D). Using this antibody, we have found that HOP-62, K562, and SR, cell lines predicted by mRNA analysis to contain relatively abundant levels of HEPL, contained a protein species of ~105 kDa, whereas lower levels of a similarly migrating species are detected in a number of other cell lines (Figure 3E). This species was removed by treatment of cells with an siRNA targeted to HEPL (Figure 3F). Together, these data indicate that HEPL is a bona fide new member of the Cas family. Based on our analysis to date, antibodies to the more widely studied Cas family members p130Cas and HEF1 do not cross-react with HEPL (Figure 3G), suggesting the presence of HEPL may mask phenotypes associated with depletion of other family members.
The best-defined action of Cas family proteins is as intermediates in integrin-dependent attachment signaling, regulating cell attachment, spreading, and migration. Although antibody to endogenous HEPL worked poorly in immunofluorescence analysis, HA-HEPL transfected into MCF7 cells (which express a low level of HEPL mRNA) colocalized with paxillin at focal adhesions (Figure 4A), comparable to other Cas proteins. HA-HEPL–transfected cells spread to a greater degree than control or vector-transfected cells, but to a lesser degree than cells transfected with HA-HEF1 (Figure 4B). Also suggesting a less potent action for HA-HEPL than HA-HEF1, levels of Y397-phosphorylated (activated) FAK were strongly increased at focal adhesions in HA-HEF1 cells. Interestingly, in HA-HEPL transfected cells, only a subpopulation of the cells (~15–20%) showed increased levels of Y397-phosphorylated FAK, whereas the remainder of the population remained at the levels of the negative control cells (Figure 4C).
Our modeling experiments had suggested that the HEPL SH3 domain would bind FAK (Figure 2A). Using a two-hybrid approach, we confirmed a direct interaction between the HEPL SH3 domain and the FAK C-terminus (Figure 4D). This result paralleled previous demonstrations of interactions between the HEF1 and other Cas family SH3 domains with proline-rich SH3 domain–binding motifs in this region (Polte and Hanks, 1995 ; Law et al., 1996 ) and suggested that overexpressed HEPL activates FAK based on direct interaction. Finally, like HEF1, full-length HEPL was immunoprecipitated with antibody to FAK from cells transfected with plasmids expressing each protein (Figure 4E). A common characteristic of Cas family proteins is their phosphorylation by Src family kinases during the cell attachment process, which provides a docking site for FAK and contributes to assembly of signaling complexes at focal adhesions (O'Neill et al., 2000 ). HEPL lacks the YDYVHL site that coordinates docking with the Src SH2 domain for other family members (Tachibana et al., 1997 ). Nevertheless, attachment-induced Src phosphorylation of HEPL was observed (Figure 4F), suggesting that additional interactions between HEPL, FAK, and Src are sufficient to drive this modification.
We next analyzed focal adhesions and cell spreading in HOP-62 cells from which HEPL had been depleted by siRNA (Figure 5A). In the total population of siRNA transfected cells, HEPL depletion reduced cell spreading (p < 0.01), although to a significantly smaller degree than HEF1 depletion (p < 0.001). On closer inspection, the HEPL-depleted population differed from both scrambled siRNA control- and HEF1-depleted cells in its heterogeneous nature. A subpopulation of ~20% of HEPL-depleted cells showed reduced and/or differentially localized staining for Y397-phosphorylated FAK (Figure 5B), with residual staining diffusely distributed at the cell periphery rather than in discrete focal adhesions. Subsequent staining with antibody to paxillin (Figure 5C) revealed a similar population of ~20% of cells that had greatly reduced paxillin staining. Reanalyzing spreading data for strongly versus weakly Y397-FAK staining HEPL-depleted cells (Figure 5A) revealed a clear segregation of weak staining with reduced spreading. By contrast, almost all cells with depleted HEF1 were less spread than Scr control cells and had reduced Y397-phosphorylated FAK, to the same degree as the 20% of responsive HEPL-depleted cells.
An important function of Cas proteins is regulation of cell migration (Klemke et al., 1998 ; van Seventer et al., 2001 ; Fashena et al., 2002 ). Analysis of live cell images of HEPL- or HEF1-depleted cells indicated that depletion distorted migration profiles relative to Scr-depleted controls. HEF1 depletion uniformly reduced cellular velocity (p < 0.001, Figure 5D). Interestingly, the phenotype observed with HEPL depletion was more complex. Two distinct HEPL-depleting siRNAs caused appearance of a population of slow-moving cells, although for one of the siRNAs the effect was not statistically significant (velocity <9 μm/h, p < 0.05 and p = 0.15). However, the HEPL-depleting siRNAs each also unexpectedly caused appearance of a faster-moving (velocity <18 μm/h) group of cells, corresponding to ~15% of the population (p < 0.05 and p < 0.02). Greater velocity has been reported to be associated with greater cell spreading in some cell types with manipulated Cas proteins (e.g., Fashena et al., 2002 ) for HEF1 in MCF7 cells), although there are examples of cell movement where a connection between spreading and velocity is not observed (Friedl and Wolf, 2003 ). No “highly spread” population was detected with HEPL-depleted cells (results not shown). As further difference from the Cas group, whereas depletion of HEF1 reduced directionality of movement of cells, no such effect was seen with HEPL depletion (Figure 5E).
Some members of the Cas family, such as HEF1, also play important roles in regulation of apoptosis and proliferation (e.g., Law et al., 2000 ; O'Neill and Golemis, 2001 ; Pugacheva and Golemis, 2005 ; Dadke et al., 2006 ). siRNA depletion of HEPL from K562 cells led to a slightly slower accumulation of cells over 3 days (Figure 5F), although the general compartmentalization of cells in the G1, S, and G2/M phases of cell cycle was not significantly affected (Figure 5G). HEPL depletion did not influence the level of apoptotic cells in the population (Figure 5. H and I). As siRNA depletion rarely exceeds 90–95%, a definitive determination that HEPL does not affect cell cycle or apoptosis requires a gene knockout; however, at present, the most demonstrable activity of HEPL is at focal adhesions, as with p130Cas.
Although the Cas protein family has been studied for over a decade, this study represents the first extensive examination of the Cas proteins utilizing genomic resources. The newly identified HEPL proteins define a legitimate novel branch of the Cas family. HEPL proteins conserve important functional domains required for interaction with FAK, for targeting to focal adhesions, and for regulating cell spreading and FAK activation. Although human HEPL is expressed in a relatively limited subset of cell types relative to p130Cas and HEF1, based on mRNA analysis it appears to be as prevalent as Efs/Sin in cultured cell lines.
HEPL does not appear to be as biologically active as HEF1, based on a number of criteria presented above. Particularly in control of FAK activation, only a minority of cells respond either to overexpressed HEPL or to depleted HEPL, under conditions where almost all cells respond to similarly manipulated HEF1. siRNA depletion typically introduces siRNA into >90% of cells, and our analysis of HEPL siRNA-transfected cells confirmed >75% depletion in practice, excluding the trivial explanation of incomplete depletion. Rather, we expect the difference may relate to cell-specific variability in the intrinsic expression level of the additional Cas family members within HOP-62 cells: single cell analyses are beginning to demonstrate that this is an important property governing average gene expression in cell populations (e.g., Levsky and Singer, 2003 ; Mar et al., 2006 ). We propose that typically within cells expressing multiple Cas family members, HEPL may make a minor contribution to regulation of cell growth properties. Part of the reduced biological activity of HEPL may arise from lack of a key motif for Src recognition (YDYVHL; Tachibana et al., 1997 ). We have shown that HEPL is still phosphorylated by Src family kinases during cell attachment, suggesting this motif is not essential for an interaction with Src, presumably because of the presence of multiple interaction interfaces joining HEPL, FAK, and Src; however, Cas proteins also reciprocally contribute to Src activation in the attachment process (e.g., Alexandropoulos and Baltimore, 1996 ), and this function may be limited.
Intriguingly, HEPL activity qualitatively differs from other Cas proteins in at least one important way, in the regulation of migration. Although loss or depletion of p130Cas and HEF1 reduces cell migration (e.g., Natarajan et al., 2006 ), HEPL depletion induced faster migration in at least a subset of cells. The reason for this is so far unknown; however, an intriguing possibility is that through possession of some but not all Cas family functions, HEPL may weakly oppose the action of other Cas family proteins via action as a “dominant negative.” Particularly in a cell background low in other Cas proteins, HEPL may be important. Hence, it is important for experiments involving knockdown or knockout of Cas family proteins to subsequently consider HEPL status in interpreting phenotypes. Separately, for each of the Cas proteins, some interacting partners have been described unique to that family member; the interaction profile of HEPL has not yet been explored, but may include novel interactors and intracellular roles.
Intriguingly, the region of chromosome 20 encompassing HEPL is included as an amplicon in many solid tumors (Dessen et al., 2002). It is hence possible that as with p130Cas and HEF1 (Singh et al., 2007 ), altered expression of HEPL contributes to the pathogenesis of cancers or other diseases. Our data demonstrate that HEPL overexpression is sufficient to increase cell spreading and FAK activation, phenotypes associated with increased tumor invasiveness. Based on its expression profile, in nontransformed cells HEPL may be most relevant to the normal function of the hematopoietic system and the lung. As all the Cas proteins have the potential to interact with multiple partner proteins, sometimes in large complexes, the presence of additional family members might also be expected to induce cell- and tissue–type differences in complex assembly and stoichiometry. Clearly, future studies of the Cas group should consider the possible role of redundant HEPL function in evaluating knockdown, knockout, or overexpression phenotypes. In sum, this study suggests ample new ground for further investigation.
We thank E. D. Cohen and Andrew Godwin (both of the Fox Chase Cancer Center) for ovarian normal cells and cell lines and Daniel Bassi, Robert Page, and Andres Klein-Szanto (all of the Fox Chase Cancer Center) for ovarian tumors. We thank Edna Cukierman for technical guidance and for many meaningful discussions regarding migration and attachment assays. We are grateful to Ian Ochs and Nicolas Day for help with cloning and two-hybrid analysis. We appreciate the technical help from the FCCC centralized Imaging Facility for Epifluorescence and Confocal Microscopy. This work and the authors were supported by National Institutes of Health (NIH) RO1s CA63366 and CA113342, by funding from the Susan B. Komen Foundation, by Tobacco Settlement funding from the State of Pennsylvania (E.A.G.), and by NIH core grant CA-06927 and support from the Pew Charitable Fund to Fox Chase Cancer Center. D.D. was a recipient of the Plain and Fancy Fellowship of the Fox Chase Cancer Center Board of Associates.
This article was published online ahead of print in MBC in Press (http://www.molbiolcell.org/cgi/doi/10.1091/mbc.E07-09-0953) on February 6, 2008.