|Home | About | Journals | Submit | Contact Us | Français|
Defining the pathways required for keratinocyte cell migration is important for understanding mechanisms of wound healing and tumor cell metastasis. We have recently identified an α6β4 integrin-Rac1 signaling pathway via which the phosphatase Slingshot (SSH) activates/dephosphorylates cofilin, thereby determining keratinocyte migration behavior. Here, we assayed the role of 14-3-3 isoforms in regulating the activity of SSH1. Using amino or carboxy terminal domains of 14-3-3ζ we demonstrate that in keratinocytes 14-3-3ζ/τ heterodimers bind SSH1, in the absence of Rac1 signaling. This interaction leads to an inhibition of SSH1 activity, as measured by an increase in phosphorylated cofilin levels. Overexpression of the carboxy terminal domain of 14-3-3ζ acts as a dominant negative and inhibits the interaction between 14-3-3τ and SSH1. These results implicate 14-3-3ζ/τ heterodimers as key regulators of SSH1 activity in keratinocytes and suggest they play a role in cytoskeleton remodeling during cell migration.
Keratinocyte motility is a multistep process that requires changes in their cytoskeleton, in their adhesion, and in the extracellular matrix to which they adhere. Recent studies have suggested that the migration and invasion of tumor cells is controlled, in part, by α6β4 integrin and its ligand, laminin-332 [1–5]. Moreover, results from several laboratories have demonstrated that keratinocyte motility behavior is dependent on α6β4 integrin/Rac1 signaling [5, 6]. This pathway signals to the actin severing protein, cofilin, by promoting its dephosphorylation/activation at serine residue 3 . In keratinocytes, we have previously presented evidence that cofilin is dephosphorylated by the Slingshot (SSH) family of phosphatases . Moreover, we have demonstrated that the linear migration of keratinocytes is determined by α6β4 integrin-mediated activation of SSH proteins .
SSH protein activity is regulated by 14-3-3 proteins [9, 10]. 14-3-3 proteins are a family of highly conserved proteins that are abundantly expressed in eukaryotic cells (reviewed in ). They inhibit SSH phosphatase activity by binding phosphorylated serine residues and sequestering SSH proteins into the cytoplasm [9, 12]. Four of the seven 14-3-3 isoforms have been shown to bind to phosphorylated SSH proteins in a variety of cell types . Our previous results demonstrated that loss of α6β4 integrin signaling or inhibiting signals along the pathway leads to an increase in 14-3-3/SSH interaction .
Over 200 proteins have been shown to interact with 14-3-3 proteins. These interactions all occur in the groove that forms between the amino and carboxy terminal domains of the folded 14-3-3 monomer [13–15]. The crystal structures of 14-3-3ζ, τ and σ have been resolved and reveal that all isoforms possess a similar tertiary structure [14, 16, 17]. Therefore, the sequence within the phosphopeptide binding groove is highly conserved. However, there is a level of specificity since not all isoforms bind the same phosphorylated proteins [18–20]. The crystal structures of the 14-3-3 isoforms have also revealed that these proteins are dimers. Each isoform, except for σ which preferentially forms homodimers, has the ability to form homo- and heterodimers [21–23]. It is through interactions of the N-terminal helices of each 14-3-3 monomer that homo- or heterodimerization occurs [14, 16, 22–24]. Therefore, particular heterodimer pairing may regulate ligand specificity.
We previously reported that all three SSH family members interact with 14-3-3 proteins in keratinocytes . Therefore, in this study we sought to extend our analysis by determining which specific 14-3-3 heterodimer binds SSH1 in keratinocytes. During the course of our study we identified an interaction between 14-3-3ζ/τ heterodimers and SSH1 protein in keratinocytes, and demonstrate that this interaction is a key regulator of cofilin activity.
Human epidermal keratinocytes, immortalized with Human Papilloma Virus genes E6 and E7, were previously described . The cells were maintained in defined keratinocyte serum-free medium supplemented with a 1% penicillin/streptomycin mixture (Invitrogen Corp.) and grown at 37 °C. The Rac1 inhibitor NSC23766 was obtained from Calbiochem and added to growth medium for 18 hrs. at a concentration of 50 μM. Mouse monoclonal antibody specific for 14-3-3τ was obtained from Abcam (Cambridge, MA). Rabbit polyclonal antibody specific for 14-3-3ζ and a rabbit pan-14-3-3 antibody were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). Mouse monoclonal antibody against the V5 epitope tag was purchased from Invitrogen. Mouse monoclonal antibody against the HA tag (HA.11) was purchased from Covance (Emeryville, CA). Rabbit polyclonal antibody against Ser-3-phosphorylated cofilin was purchased from Cell Signaling Technology (Beverly, MA). Rabbit monoclonal antibody against β-actin was obtained from Epitomics, Inc. (Burlingame, CA).
Adenovirus encoding the human wildtype SSH1 protein was described previously . 14-3-3ζ cDNA was amplified by PCR using a human heart cDNA library (from Stratagene) as a template and cloned into pcDNA3 expression vector (Invitrogen Corp.). Additional sequences encoding the HA tag were added to the amino-terminus to generate pHA14-3-3. Two 14-3-3ζ deletion constructs were subsequently generated by PCR to encode amino acids 1–140 (pHA14-3-3-ΔC) and amino acids 140–245 (pHA14-3-3-ΔN). These cDNAs were subcloned into the polylinker of pENTR4 vector (Invitrogen Corp.) and subsequently placed into the pAD/CMV/V5-DEST vector using LR recombination according to the manufacturer’s protocol. All constructs were verified by sequencing. Adenovirus was amplified as previously described . Keratinocytes were infected at a multiplicity of infection of 1:50 in cell medium.
Whole cell extracts from approximately 0.5 × 106 cells were prepared by solubilization in 1% SDS, 8 M urea, 10% glycerol, 5% β-mercaptoethanol, 25 mM Tris-HCl, pH 6.5. The proteins were separated by SDS-PAGE, transferred to PVDF membrane and processed for immunoblotting as previously described . For immunoprecipitation analyses, cells treated with or without Rac1 inhibitor were extracted in IP lysis buffer previously described . Cell lysates were centrifuged for 10 min. at 12,000 × g. One tenth of the lysate was saved as input for immunoblotting. 14-3-3 proteins were resolved on a 15% polyacrylamide gel, whereas V5 tagged SSH1 was resolved on a 7.5% polyacrylamide gel. For immunoprecipitation of V5 tagged SSH1, anti-V5-conjugated agarose beads (Sigma-Aldrich) were added to the lysate and incubated for 1.5 hrs. at 4 °C on a rocking platform. Immunoprecipitation of the amino- or carboxy-terminal halves of HA tagged 14-3-3ζ (ΔC or ΔN, respectively) was performed by incubating the cell lysate with the HA monoclonal antibody overnight at 4 °C on a rocking platform. Mouse IgG agarose beads were added to the lysate/antibody mix and incubated for 1.5 hrs. at 4 °C on a rocking platform. The beads were washed with phosphate-buffered saline and collected by centrifugation. Following the last wash, sample buffer (10% glycerol, 1% SDS, 0.2M Tris-HCl, pH 6.8, 5% β-mercaptoethanol) was added to the bead pellet. The sample was boiled for 10 min., processed for SDS-PAGE, transferred to PVDF, and processed for immunoblotting as above. Immunoblots were scanned and quantified using MetaMorph Imaging System.
Our previous results have demonstrated that Rac1 signaling leads to the activation of SSH proteins in migrating keratinocytes . In contrast, inhibition of Rac1 signaling leads to an increase in 14-3-3/SSH interactions . Several 14-3-3 isoforms, including β, γ, τ, and, ζ have been shown to bind SSH1 and inhibit its phosphatase activity . However, the exact 14-3-3 isoforms that make up the homo- or heterodimers that bind SSH1 in keratinocytes have not been determined. Thus, to identify the specific 14-3-3 isoforms that bind SSH1 in skin cells, we immunoprecipitated V5 tagged SSH1 from keratinocytes treated with a Rac1 inhibitor and probed the precipitate with 14-3-3 isoform specific antibodies. Our results indicate that SSH1 interacts with the endogenous 14-3-3ζ and τ isoforms upon Rac1 inhibition (Figure 1A).
We next investigated whether 14-3-3ζ and τ isoforms form homo- or heterodimers when interacting with SSH1. Previous results using mutant 14-3-3 proteins in which either the carboxy-terminal or amino-terminal residues are removed, indicate that the amino-terminus of 14-3-3 is required for dimerization . Thus, we used 14-3-3ζ amino- and carboxy-terminal deletion mutants, 14-3-3ζΔN and 14-3-3ζΔC consisting of either amino acid residues 140–245 or residues 1–139 respectively, as tools to determine whether homodimers of 14-3-3ζ or heterodimers of 14-3-3ζ/τ exist in keratinocytes. Specifically, 14-3-3ζΔC should form dimers whereas 14-3-3ζΔN should not.
Expression of either HA tagged 14-3-3ζΔC or 14-3-3ζΔN was induced in keratinocytes via adenoviral delivery. The overexpressed proteins were immunoprecipitated with antibodies against the HA tag and the lysates were probed with antibodies that specifically recognize endogenous 14-3-3ζ or 14-3-3τ. 14-3-3ζΔC interacts with the 14-3-3τ isoform in extracts of our skin cells (Figure 1B, top panel). As expected, very little interaction between 14-3-3ζΔN and the 14-3-3τ isoform is detected (Figure 1B, top panel). In contrast, neither 14-3-3ζΔN nor 14-3-3ζΔC co-precipitate with endogenous 14-3-3ζ (Figure 1B, bottom panel). These data suggest that the overexpressed 14-3-3ζ protein fails to form homodimers with endogenous 14-3-3ζ in keratinocytes. Rather, 14-3-3ζΔC heterodimerizes with endogenous 14-3-3τ in epidermal cells. Therefore, our results imply that 14-3-3ζ and τ exist as heterodimers in keratinocytes.
We next wanted to assess whether 14-3-3ζΔC/τ heterodimers interact with SSH1. Therefore, keratinocytes expressing HA tagged 14-3-3ζΔC or 14-3-3ζΔ N and V5 tagged SSH1 protein were treated with a Rac1 inhibitor to induce assembly of SSH1/14-3-3 complexes. Tagged 14-3-3ζΔC or 14-3-3ζΔN was immunoprecipitated from treated cells using an antibody against HA (Figure 2A). The lysate was then probed with antibodies that specifically recognize the 14-3-3τ isoform or the V5 tag. A complex between 14-3-3ζΔC, tagged SSH1, and endogenous 14-3-3τ is detected in cells expressing 14-3-3ζΔC (Figure 2A).
To further verify that 14-3-3ζΔC/τ heterodimers bind SSH1 in keratinocytes, tagged SSH1 protein was immunoprecipitated from cells treated with a Rac1 inhibitor using V5 antibody-conjugated agarose beads. The keratinocytes were induced to express V5 tagged SSH1 and 14-3-3ζΔC or 14-3-3ζΔN. The lysate was then probed with an antibody that specifically recognizes the 14-3-3τ isoform (Figure 2B). Binding of tagged SSH1 to endogenous 14-3-3τ is detected in cells expressing 14-3-3ζΔC (Figure 2B). Taken together, our results demonstrate that 14-3-3ζ/τ heterodimers bind to SSH1 in keratinocytes.
As mentioned earlier, the 14-3-3ζΔN construct lacks the amino terminal residues important for dimerization. Thus, as expected, 14-3-3ζΔN does not co-precipitate with endogenous 14-3-3τ from extracts of keratinocytes (Figure 1B and and2A).2A). However, monomers of 14-3-3ζΔN bind SSH1, this result being consistent with published data indicating that monomeric 14-3-3ζ can interact with phosphopeptide targets (Figure 2A). Moreover, our finding reveals that the carboxy terminal half of 14-3-3ζ contains a binding site for SSH1, which can mediate 14-3-3/SSH1 interaction independent of 14-3-3 dimerization.
Expression of 14-3-3ζΔN inhibits the binding of endogenous 14-3-3τ to SSH1 (Figure 2B). In addition, we examined whether expression of 14-3-3ζΔN impacts the interaction of SSH1 with other 14-3-3 isoforms. Keratinocytes expressing V5 tagged SSH1 and HA tagged 14-3-3ζΔC or 14-3-3ζΔN were treated with Rac1 inhibitor and processed for immunoprecipitation using V5 antibody-conjugated beads, as described above. The precipitate was then probed for endogenous 14-3-3/SSH1 interactions with an antibody that recognizes multiple isoforms of 14-3-3, including β, γ, η, σ, θ, and ζ. As expected, expression of 14-3-3ζΔC has no affect on SSH1/14-3-3 interaction (Figure 2C). In contrast, expression of 14-3-3ζΔN abolishes the interaction between SSH1 and 14-3-3 (Figure 2C). Therefore, these results imply that 14-3-3ζΔN competes with multiple 14-3-3 isoforms for binding to SSH1.
During the course of the above studies, we measured the expression of endogenous 14-3-3τ and 14-3-3ζ in keratinocytes expressing 14-3-3ζΔC or 14-3-3ζΔN. Surprisingly, the level of endogenous 14-3-3τ expression decreases in the presence of 14-3-3ζΔN, although endogenous 14-3-3ζ protein levels are not affected when 14-3-3ζΔC or 14-3-3ζΔN is expressed in keratinocytes (Figure 3). These results suggest that expression of 14-3-3ζΔN regulates the level of endogenous 14-3-3τ and thus, the binding of 14-3-3τ to SSH1.
Previous results from our laboratory have demonstrated that Rac1 and SSH activities positively regulate cofilin activity in keratinocytes [5, 8]. Treatment of cells with a Rac1 inhibitor increases phospho-cofilin (inactive cofilin) levels in keratinocytes . Similarly, keratinocytes expressing 14-3-3ζΔC treated with Rac1 inhibitor have increased levels of inactive/phospho-cofilin compared to untreated keratinocytes (Figure 4). This is consistent with the finding that 14-3-3ζΔC can heterodimerize with 14-3-3τ, bind SSH1, and inactivate it. In contrast, keratinocytes expressing 14-3-3ζΔN treated with Rac1 inhibitor exhibit a decrease in phospho-cofilin (inactive) levels compared to cells expressing 14-3-3ζΔC treated with Rac1 inhibitor (Figure 4). Therefore, binding of 14-3-3ζΔN to SSH1 is insufficient to inactivate SSH1 and increase phospho-cofilin levels. Indeed, these data indicate that 14-3-3 dimers, and not monomers of 14-3-3 proteins, regulate SSH1 activity. Furthermore, they imply that 14-3-3ζΔN acts in a dominant negative manner with regard to SSH1 activity and the consequent state of cofilin phosphorylation. Our results suggest that it does so via two mechanisms. Specifically, 14-3-3ζΔN competes with endogenous 14-3-3 proteins for binding to SSH1 and also regulates the stability of endogenous 14-3-3 proteins. Generally, 14-3-3 proteins are dimers and are thermodynamically stable. It has been suggested that once 14-3-3 dimers are formed, the partners do not readily exchange . Thus, one consequence of 14-3-3ζΔN expression is an alteration in endogenous 14-3-3τ stability, which, in turn, affects the pairing of 14-3-3ζ/τ into heterodimers. A similar mechanism, termed “isotype interference” has been suggested by others , where the expression level of a certain isoform affects the combination of homo- or heterodimer pairs.
In summary, we have provided new insight into the molecular mechanism via which SSH regulates cofilin activation and the consequent actin cytoskeleton remodeling that occurs during cell migration in wound healing and cancer. The results we have presented demonstrate that 14-3-3ζ preferentially interacts with 14-3-3τ in keratinocytes when binding to SSH1. Moreover, 14-3-3ζ/τ heterodimers inhibit SSH1 phosphatase activity in the absence of Rac1 signaling, leading to an increase in the level of phosphorylated cofilin. The balance of 14-3-3, SSH and cofilin activity is clearly a crucial checkpoint in skin cell migration.
This work was supported by a National Institutes of Health Grant (RO1 AR054184) to JCRJ. KK was supported by a National Institutes of Health Training Grant (T32 HL076139).
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.