Protein kinase Cδ (PKCδ) has been implicated to play a crucial role in cell proliferation, differentiation and apoptosis. In this study, we have investigated the role of PKCδ in cell motility using Madin-Darby canine kidney cells. Overexpression of PKCδ promoted membrane protrusions, concomitant with increased cell motility. By contrast, suppression of PKCδ expression by RNA interference inhibited cell motility. Moreover, a fraction of PKCδ was detected at the edge of membrane protrusions in which it colocalized with adducin, a membrane skeletal protein whose phosphorylation state is important for remodeling of the cortical actin cytoskeleton. Elevated expression of PKCδ correlated with increased phosphorylation of adducin at Ser726 in intact cells. In vitro, PKCδ, but not PKCα, directly phosphorylated the Ser726 of adducin. Finally, we demonstrated that overexpression of both adducin and PKCδ could generate a synergistic effect on promoting cell spreading and cell migration. Our results support a positive role for PKCδ in cell motility and strongly suggest a link between PKCδ activity, adducin phosphorylation and cell motility.
A dynamic movement of membrane protrusions consisting of filopodia and lamellipodia is thought to be important for cell motility, which requires fluctuating remolding of the cortical actin cytoskeleton (Lauffenburger and Horwitz, 1996; Mitchison and Cramer, 1996; Welch et al., 1997). Adducins (α, β, and γ isoforms) are a family of membrane skeletal proteins known to play a crucial role in assembly of the membrane skeleton (Gardner and Bennett, 1987; Hughes and Bennett, 1995). The α and γ isoforms of adducins (103 kDa and 84 kDa, respectively) are ubiquitously expressed in most tissues, whereas the β isoform (97 kDa) has a more restricted pattern of expression, abundant in erythrocytes and the brain (Bennett et al., 1988; Dong et al., 1995; Gardner and Bennett, 1986). The adducin isoforms are closely related in amino acid sequences and domain organization containing an N-terminal head domain, a neck domain, and a C-terminal tail domain (Dong et al., 1995; Joshi and Bennett, 1990; Joshi et al., 1991). They assemble into heteromeric complexes composed of either α and β, or α and γ, through interactions in the globular Nterminal head domains (Bennett et al., 1988; Dong et al., 1995; Hughes and Bennett, 1995). The C-terminal tail domain contains a highly basic stretch of 22 amino acids with sequence similar to a domain in the myristoylated alanine-rich C kinase substrate (MARCKS) (Dong et al., 1995; Joshi et al., 1991). The MARCKS-related domain is required for interactions of adducin with actin and spectrin (Li et al., 1998). Adducin binds to the barbed ends (Kuhlman et al., 1996; Li et al., 1998) and the sides (Mische et al., 1987) of actin filaments, thereby promotes the association of spectrin with actin filaments to form a spectrin-actin meshwork beneath the plasma membrane (Bennett et al., 1988; Gardner and Bennett, 1987). The phosphorylation of adducin in vitro by protein kinase C (PKC) diminishes its interaction with actin and spectrin (Matsuoka et al., 1998), rendering it possible that adducin may be involved in exposing the barbed ends of actin filaments in motile cells. Consequently, adducin may be implicated in the development of actin assembly sites (Falet et al., 2002; Ichetovkin et al., 2002). Although the Ser726 of adducin was demonstrated to be the major phosphorylation site for PKC (Matsuoka et al., 1998), the specificity of the PKC isozymes for this phosphorylation event has not been clarified.
PKC comprises a group of serine/threonine protein kinases that participate in the control of a wide variety of cellular functions (Newton, 1995; Nishizuka, 1995). So far, at least ten isozymes in the mammalian PKC family have been identified that can be classified into three subgroups based on their ability to be activated by Ca2+ and diacylglycerol (DAG). The classical PKCs (α, βI, βII, γ) are activated by both Ca2+ and DAG. The activation of the novel PKCs (δ, ϵ, η, θ) is Ca2+ independent but DAG dependent. The atypical PKCs (ζ, ι/λ) are both Ca2+ and DAG independent. In addition, PKCμ and PKCν are considered by some to constitute a fourth class and by others to comprise a distinct family called protein kinase D (Johannes et al., 1994; Valverde et al., 1994). All PKC isozymes are composed of an N-terminal regulatory domain and a Cterminal catalytic domain. The regulatory domain contains two key elements: an autoinhibitory sequence (pseudosubstrate) and one or two membrane-targeting modules (C1 and C2 domains). The C1 domain is a cysteine-rich region of approximately 50 residues, which binds DAG or phorbol ester. In classical and novel PKCs, it is present as a tandem repeat, named C1A and C1B. The C2 domain is an independent membrane-targeting module found in classical and novel PKCs. The C2 domain in classical PKCs binds phosphatidylserine and Ca2+; however, the C2 domain (or C2-like domain) in novel PKCs does not bind Ca2+ (Newton, 2001). The C2 domain of PKCδ does not bind to phospholipids (Stahelin et al., 2004), instead it mediates the interaction of PKCδ with other cellular proteins, including actin (Lopez-Lluch et al., 2001), GAP-43 (Dekker and Parker, 1997) and probably SRBC [serum deprivation response factor (sdr)-related gene product that binds to c-kinase] (Izumi et al., 1997). More recently, the C2 domain of PKCδ was demonstrated to serve as a phosphotyrosine-binding domain through which PKCδ interacts with a Src-binding glycoprotein, CDCP1 (Benes et al., 2005).
PKCδ is the most thoroughly studied member of the novel PKC subfamily (Gschwendt, 1999), which has been implicated to play a role in cell proliferation (Ashton et al., 1999; Fukumoto et al., 1997; Kitamura et al., 2003; Li et al., 1996), differentiation (Corbit et al., 1999; Mischak et al., 1993; Pessino et al., 1995), apoptosis (Brodie and Blumberg, 2003; Kajimoto et al., 2004; Zhong et al., 2002), and tumor suppression (La Porta and Comolli, 1995; Lu et al., 1997; Reddig et al., 1999). Increasing evidence also indicates that PKCδ is involved in the control of cell motility. Smooth muscle cells isolated from the aortic medium of PKCδ-deficient mice were reported to have less organized actin cytoskeleton and reduced migration (Li et al., 2003) compared with that observed in wild-type animals. Activation of PKCδ was shown to be required for vascular endothelial-growth-factorstimulated migration of endothelial cells (Gliki et al., 2002) and play a major role in epidermal-growth-factor-induced contractility and motility of fibroblasts (Iwabu et al., 2004). Moreover, PKCδ was reported to be essential for the migration and metastatic potential of mammary tumor cell lines (Kiley et al., 1999a; Kiley et al., 1999b; Kruger and Reddy, 2003). In the above studies, PKCδ seems to play a positive role in cell motility. Intriguingly, PKCδ could also have a negative role on cell motility. Jackson et al. (Jackson et al., 2005) showed that mouse embryo fibroblasts derived from PKCδ knockout mice had increased cell motility relative to their wild-type counterparts and that expression of PKCδ in human breast cancer BT-549 cells suppressed their migration. Therefore, PKCδ may have either positive or negative effects on cell motility depending on the cellular context. In this study, we have used Madin-Darby canine kidney (MDCK) cells as a model to examine the role of PKCδ in cell motility and its link with the membrane skeletal protein adducin.
Overexpression of PKCδ in MDCK cells induces the formation of lamellipodia
To examine the potential involvement of PKCδ in the control of cell motility, the effect of the selective PKCδ inhibitor rottlerin on the motility of MDCK cells was first evaluated. At concentrations as low as 5 μM, rottlerin completely inhibited the motility of MDCK cells (data not shown). Subsequently, an overexpression strategy was used to examine the role of PKCδ in cell motility. GFP-fused PKCδ (GFP-PKCδ) and its kinase-deficient (kd) mutant were stably overexpressed in MDCK cells. The level of GFP-PKCδ was approximately tenfold that of the endogenous PKCδ (Fig. 1A), correlated with an increase in the total PKCδ activity of the cells (Fig. 1B). The activity of GFP-PKCδ was increased by PMA, but inhibited by rottlerin (Fig. 1C), indicating that GFP-PKCδ is not a constitutively active form. Although PKCδ has been shown to play a role in cell proliferation (Ashton et al., 1999; Fukumoto et al., 1997; Kitamura et al., 2003; Li et al., 1996) and apoptosis (Brodie and Blumberg, 2003; Kajimoto et al., 2004; Zhong et al., 2002), it is not the case in our study. Overexpression of GFP-PKCδ or its kd mutant in MDCK cells neither caused their death nor disturbed their cell cycle progression (data not shown). Notably, the colonies formed by the cells overexpressing GFP-PKCδ, but not its kd mutant, exhibited a more spread morphology with evident lamellipodia at the periphery (Fig. 1D). When plated on collagen-coated glass coverslips, the cells overexpressing GFP-PKCδ wt frequently displayed a phenotype of motile cells with more active membrane protrusions in which the cortical actin filaments were enriched (Fig. 1E). The fluorescence of GFP-PKCδ was detected at the Golgi-like region and at the edge of the membrane protrusions (Fig. 1E). Since the localization of PKCδ at the membrane protrusions might be relevant to its ability to regulate cell motility, it is important for us to ascertain what we observed was not an artifact as a result of the GFP-fused PKCδ construct. For this purpose, FLAG epitope-tagged PKCδ (FLAG-PKCδ) was expressed in MDCK cells (Fig. 1F) and its localization was examined by immunofluorescent staining with the monoclonal anti-FLAG. Like GFP-PKCδ, a fraction of FLAG-PKCδ was found to localize at the cell periphery (Fig. 1G).
Overexpression of PKCδ in MDCK cells promotes their spreading and migration
The effect of GFP-PKCδ overexpression on the spreading and migration of MDCK cells was measured. The onset of the cells stably expressing GFP-PKCδ spreading on collagen was at least 20 minutes earlier than that in control MDCK cells (Fig. 2A). One hour after plating, more than 80% of the cells expressing GFP-PKCδ were spread, which was at least double that of the control cells. Two hours after plating, all of the PKCδ-overexpressed cells and the control MDCK cells were spread, but half of the cells expressing the PKCδ kd mutant retained a rounded shape. Moreover, the migratory rate of the PKCδ-overexpressed cells was approximately three times faster than that of the control MDCK cells, measured by a Trans-well cell migration assay (Fig. 2B). Of note, the expression of the GFP-PKCδ kd mutant significantly (∼60%) suppressed MDCK cell migration through the membrane (Fig. 2B). To further confirm the role of PKCδ in cell motility, a wound-healing assay was performed and monitored by time-lapse microscopy. We found that the motility of the PKCδ-overexpressed cells was indeed faster than the control MDCK cells (Fig. 2C). The motility of MDCK cells and the PKCδ-overexpressed cells was measured at an average speed of 17 μm/hour and 33 μm/hour, respectively. Under these conditions, the cells expressing the GFP-PKCδ kd mutant were defective in migration toward the wound (Fig. 2C). These results, together suggest that PKCδ may play a role in promoting cell spreading and migration and that the PKCδ kd mutant can exert a dominant-negative effect on both events.
siRNA-mediated knockdown of endogenous PKCδ in MDCK cells suppresses their migration
Next, the role of PKCδ in promoting cell motility was examined by siRNA strategy. For this purpose, we cloned and sequenced the canine PKCδ cDNA from MDCK cells. The nucleotide sequence of canine PKCδ is 90%, 89% and 88% identical to the sequences of human, rat and mouse PKCδ, respectively. The canine PKCδ cDNA encodes 674 amino acid residues, which is 87% identical to human PKCδ and 86% identical to rat and mouse PKCδ. The residues (Thr505, Ser643 and Ser662) known for regulatory phosphorylation of human and murine PKCδ are all conserved in canine PKCδ. Based on the nucleotide sequence of canine PKCδ, three sets of oligonucleotides were designated for establishing a doxycycline-inducible expression of PKCδ siRNA in MDCK cells. The siRNA derived from residues 143-162 of canine PKCδ partially suppressed the expression level of endogenous PKCδ in MDCK cells. 72 hours after doxycycline addition, the extent of PKCδ knockdown was estimated to be 70%; this correlated with a decrease in the total PKCδ activity of the cells (Fig. 3A). The expression levels of PKCα and PKCϵ remained unchanged by induction of PKCδ siRNA (Fig. 3A), supporting the specificity of the PKCδ siRNA in this system. More importantly, the PKCδ knockdown in MDCK cells was correlated with a suppression (∼60%) of their migration (Fig. 3B), further supporting a role of PKCδ in promoting cell migration.
PKCδ, but not PKCα, phosphorylates Ser726 of α-adducin in vivo and in vitro
Since active formation of lamellipodia is the most striking effect of PKCδ overexpression on the morphology of MDCK cells (Fig. 1D), it is possible that PKCδ may facilitate cortical actin remodeling, leading to lamellipodia formation. Among the molecules known to regulate remodeling of the cortical actin cytoskeleton, the membrane skeletal protein adducin was reported to be a substrate for PKC (Kaiser et al., 1989; Ling et al., 1986; Matsuoka et al., 1996; Matsuoka et al., 1998; Waseem and Palfery, 1990). However, the specificity of the PKC isoforms to phosphorylate adducin remains uncertain. To examine the role of PKCδ in this event, GFP-PKCδ and its kd mutant were transiently expressed in COS cells and their effect on the Ser726 phosphorylation of endogenous α-adducin was measured. Our result showed that the Ser726 phosphorylation of α-adducin in COS cells was enhanced by GFP-PKCδ but not its kd mutant (Fig. 4A). Consistently, stable overexpression of GFP-PKCδ, but not its kd mutant, in MDCK cells enhanced the Ser726 phosphorylation of α-adducin (Fig. 4B), which could be further increased by PMA treatment but inhibited by rottlerin (Fig. 4C). Moreover, a fraction of GFP-PKCδ was found to colocalize with adducin and its phosphorylated form at the edge of the lamellipodia (Fig. 4D). To examine whether PKCδ could directly phosphorylate adducin in vitro, histidinetagged adducin proteins were initially expressed in bacteria. Unfortunately, adducin proteins expressed in bacteria were very insoluble (data not shown). To overcome this obstacle, GFP-α-adducin and its mutants (S716A, S726A and S716A/S726A) were transiently overexpressed in HEK293 cells, purified by immunoprecipitation, and used as a substrate in an in vitro kinase assay for PKCδ (Fig. 4E). Ser716 and Ser726 are in the phosphorylation motif for PKC (Matsuoka et al., 1996; Matsuoka et al., 1998). The result showed that PKCδ was capable of phosphorylating purified GFP-adducin in vitro (Fig. 4F). Mutation at Ser716 did not have an impact to the phosphorylation of α-adducin by PKCδ. However, mutation at Ser726 of α-adducin largely (∼80%) prevented phosphorylation by PKCδ (Fig. 4F). These results together indicate that Ser726 of α-adducin serves as the major phosphorylation site for PKCδ both in vivo and in vitro. To examine the specificity of PKCδ to phosphorylate adducin at Ser726, GFP-PKCδ and GFP-PKCα were transiently expressed in COS cells and their effect on adducin phosphorylation was evaluated. The result showed that the Ser726 phosphorylation of α-adducin was only enhanced by GFP-PKCδ, but not by GFP-PKCα, in COS cells (Fig. 5A). The results from the in vitro kinase assays revealed that PKCα phosphorylated the S716A mutant and the S726A mutant to a similar extent to that of wt α-adducin (Fig. 5B). These results suggest that PKCδ, but not PKCα, may be the kinase responsible for Ser726 phosphorylation of α-adducin.
PKCδ-mediated phosphorylation of α-adducin promotes cell spreading and migration
To examine the biological significance of PKCδ-mediated phosphorylation of α-adducin, FLAG-PKCδ was stably co-expressed with GFP-α-adducin or its S726A mutant in MDCK cells. Accordingly, the level of the Ser726 phosphorylation of adducin in cells overexpressing FLAG-PKCδ was higher than that in the neomycin-resistant control cells (Fig. 6A). In addition, FLAG-PKCδ was found to colocalize with GFP-α-adducin at the leading edge of the migratory cells (Fig. 6B). Expression of GFP-α-adducin alone, but not its S726A mutant, in MDCK cells promoted their spreading and migration. More importantly, co-expression of FLAG-PKCδ with GFP-α-adducin synergistically promoted both events. By contrast, the adducin S726A mutant failed to collaborate with FLAG-PKCδ to promote MDCK cell spreading (Fig. 6C) and migration (Fig. 6D). These results together suggest that the phosphorylation of adducin at its Ser726 by PKCδ may be important for cell spreading and cell migration.
PKCδ has been implicated in the control of many aspects of cell behavior, including cell proliferation, differentiation, apoptosis, and tumorigenesis (Ashton et al., 1999; Brodie and Blumberg, 2003; Corbit et al., 1999; Fukumoto et al., 1997; Kajimoto et al., 2004; Kitamura et al., 2003; La Porta and Comolli, 1995; Li et al., 1996; Lu et al., 1997; Mischak et al., 1993; Pessino et al., 1995; Reddig et al., 1999; Zhong et al., 2002). However, PKCδ is often found to have either positive or negative impact to a given cell function, when it is studied in different types of cells. For example, increased PKCδ activity is associated with increased cell proliferation in some cells (Kitamura et al., 2003; Li et al., 1996), but it suppresses cell proliferation in others (Ashton et al., 1999; Fukumoto et al., 1997). Similarly, the role of PKCδ in cell motility has been equivocal. For instance, smooth muscle cells isolated from the aortic media of PKCδ knockout mice have a less organized actin cytoskeleton and reduced migration (Li et al., 2003), whereas mouse embryo fibroblasts derived from PKCδ knockout mice have increased cell motility (Jackson et al., 2005). In addition, elevated expression of PKCδ increases the motility and metastatic potential of rat mammary tumor cells (Kiley et al., 1999a; Kiley et al., 1999b), whereas it suppresses the migration of human breast cancer cells (Jackson et al., 2005). Currently, there is no good explanation for these conflicting data. It certainly requires further study before we can really understand how PKCδ operates inside the cell. Our results here support a positive role for PKCδ in cell motility and strongly suggest a link between PKCδ activity, adducin phosphorylation and cell motility.
Adducin has been implicated in the development of actin assembly sites (Falet et al., 2002; Ichetovkin et al., 2002). In particular, phosphorylation of adducin by PKC is crucial for platelet activation (Barkalow et al., 2003; Gilligan et al., 2002). In resting platelets, most of the actin is in the monomeric state, and some exists as short actin filaments with capping proteins preventing polymerization into longer filaments. When platelets are activated, most of the G-actin polymerizes into Factin, causing extensions of the platelet membrane into filopodia and lamellipodia (Hartwig et al., 1999). Dissociation of phosphoadducin releases spectrin from actin, facilitating centralization of spectrin, and leads to the exposure of actin filament barbed ends, which may then participate in converting the resting platelet's disc shape into its active, spreading phenotype (Barkalow et al., 2003). In this study, we found that PKCδ overexpression in MDCK cells promoted membrane protrusions (Fig. 1), and a fraction of PKCδ colocalized with adducin and phosphoadducin at the edge of membrane protrusions (Fig. 4). Moreover, we demonstrated that PKCδ was capable of phosphorylating α-adducin at Ser726 both in vitro (Fig. 4) and in intact cells (Figs 4 and 6). Taken together, our results raise the possibility that in epithelial cells, PKCδ-mediated phosphorylation of adducin may cause the exposure of actin filaments barbed ends for actin polymerization, thereby facilitating cell motility.
Another interesting finding in this report is that PKCδ, but not PKCα, is responsible for Ser726 phosphorylation of adducin (Fig. 5). Although PKC has long been thought to phosphorylate adducin at its Ser726 (Ling et al., 1986; Matsuoka et al., 1996; Matsuoka et al., 1998), the specificity of the PKC isozyme responsible for this event has not been clarified. In previous studies, the activators (e.g. phorbol 12-myristate 13-acetate) of PKC were often used to establish a correlation between PKC activity and adducin phosphorylation (Kaiser et al., 1989; Waseem and Palfery, 1990), and the mixture of PKC isozymes from tissue extracts were used to demonstrate an in vitro phosphorylation of adducin by PKC (Ling et al., 1986; Matsuoka et al., 1996; Matsuoka et al., 1998). However, neither approach could specifically determine which type of the PKC isozymes is responsible for phosphorylating adducin at Ser726. In this report, we found that elevated expression of PKCδ, but not PKCα, induced an increase in the Ser726 phosphorylation of adducin (Fig. 5A). Furthermore, the mutation of adducin at the Ser726 largely (∼80%) prevented phosphorylation by PKCδ, but had no impact on the phosphorylation of adducin by PKCα (Fig. 5C). These results therefore suggest that PKCδ is likely to be the PKC isozyme responsible for phosphorylating Ser726 of adducin. Since PKCα phosphorylated the Ser716 mutant and the Ser726 mutant of adducin to an extent similar to the wild type (Fig. 5C), this suggests that PKCα may phosphorylate adducin at residue(s) other than Ser716 and Ser726.
Cell migration is the consequence of coordinated regulation between cell adhesion, cell contractility and actin cytoskeletal remodeling. Our results in this study suggest that PKCδ may enhance cell motility, at least partially, through its ability to phosphorylate adducin, leading to reorganization of the cortical actin cytoskeleton. However, it has previously been proposed that PKCδ may regulate cell motility through its effect on cell adhesion (Barry and Critchley, 1994) and contractility (Iwabu et al., 2004). In fact, we already examined those possibilities in MDCK cells. We found that GFP-PKCδ did not localize at the focal adhesions of MDCK cells and that the adhesive strength of the cells to collagen, the expression level of the collagen receptors (integrins α2β1 and α3β1) and the phosphorylation and activity of focal adhesion kinase, a major downstream effector for most integrins, did not change upon PKCδ overexpression or siRNA-mediated knockdown (data not shown). Furthermore, the phosphorylation of myosin light chain, an indicator for contractile force, was not affected by PKCδ expression in MDCK cells either (data not shown). These results therefore do not support the idea that PKCδ promotes MDCK cell migration through either focal adhesion assembly or cell contractility. Since the formation of membrane protrusions is most striking when PKCδ is overexpressed in MDCK cells, it is more likely that PKCδ-promoted motility of MDCK cells occurs through its effect on facilitating remodeling of the cortical actin cytoskeleton. In addition to adducin, many other molecules, such as cortactin, WASP and Arp2/3 (Small et al., 2002; Uruno et al., 2001), are known to involve in the formation of lamellipodia. Therefore, it remains possible that PKCδ may interact with and phosphorylate those molecules, thereby facilitating the remodeling of the cortical actin cytoskeleton for membrane protrusions.
Localization of PKCδ at the cell periphery, in particular at the leading edge of motile cells, has never been described previously. This finding sheds light on a new direction for us to investigate how PKCδ regulates cell motility. Several questions relevant to this phenomenon remain to be answered. For example, how does PKCδ target to the membrane cortical regions? Does PKCδ crosstalk with the Rho GTPase family proteins for regulating the actin cytoskeleton? Does PKCδ generally participate in growth-factor-induced cell motility though its effect on the cortical actin cytoskeleton? Experiments are in progress to test these possibilities.
Materials and Methods
The polyclonal anti-PKCδ (C-20), polyclonal anti-PKCϵ (C-15), polyclonal anti-adducin (H-100), and polyclonal anti-phospho-adducin (Ser726) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). The monoclonal anti-PKCα was purchased from BD Transduction Laboratories (Lexington, KY). The monoclonal anti-GFP was purchased from Roche. The monoclonal anti-FLAG M2 and antitubulin, type I collagen, myelin basic protein (MBP), tetramethyl rhodamine isothiocyanate (TRITC)-conjugated phalloidin, and Protein A-Sepharose were purchased from Sigma-Aldrich. Fetal bovine serum and LipofectAMINE were purchased from Invitrogen Life Technologies. Rottlerin, phorbol-12-myristate-13-acetate (PMA), G418 sulfate, and doxycycline were purchased from Calbiochem (San Diego, CA). The plasmid pEGFP-N1-PKCδ was kindly provided by Peter M. Blumberg (National Institutes of Health, Bethesda, MD) and was described previously (Wang et al., 1999). The plasmid pcDNA3-FLAG-PKCδ was kindly provided by Ushio Kikkawa (Kobe University, Japan) and was described previously (Konishi et al., 2001). The plasmid pEGFP-N1-PKCα was kindly provided by Dominique Joubert (Institut de Génomique Fonctionnelle, France) and was described previously (Quittau-Prevostel et al., 2004). The plasmids pEGFP-C1-α-adducin and pEGFP-C1-α-adducin (S716A/S726A) were kindly provided by Vann Bennett (Duke University, Durham, NC). All mutagenesis was carried out using QuikChange site-directed mutagenesis kit (Stratagene). The mutagenic primers used are: PKCδ K376R (kinase-deficient): 5′-GATAAGTACTTTGCAAATCAGGTGTCTGAAGAAGGACG-3′; α-adducin S716A: 5′-GGTCTCCAGGCAAGTCCCCGGCCAAAAAGAAGAAGAAGTTCCG-3′; α-adducin S726A: 5′-GTTCCGTACCCCGGCCTTTCTGAAGAAGAG-3′. The positions of substituted codons are underlined. The desired mutations were confirmed by dideoxy DNA sequencing, a service provided by the Biotechnology Center of National Chung Hsing University, Taiwan.
Cloning and sequencing of canine PKCδ cDNA from MDCK cells
Total RNA was extracted from MDCK cells using a RNA isolation kit from Promega, and then the first strand of cDNA was reverse transcribed by Moloney murine leukemia virus reverse transcriptase (Invitrogen Life Technologies) using random hexamers as the primers (Promega). Subsequently, a 1739-nucleotide fragment of canine PKCδ cDNA was amplified by the polymerase chain reaction (PCR) with the primers forward 5′-TGCAGCCTCAGGCCAAGGT-3′ (equivalent to nt. 323-342 of the human PKCδ cDNA, accession number BC043350) and reverse 5′-GAGTCGATGAGGTTCTTGTC-3′ (equivalent to nt. 1961-1980 of the human PKCδ cDNA sequence). These two primers were designated based on the highly conserved nucleotide sequence among human, rat, and mouse PKCδ cDNA. Once the nucleotide sequence of the PCR-amplified DNA fragment was determined, another two sets of primers were designated to amplify the DNA fragments covering the 5′ and 3′ ends of the canine PKCδ cDNA by the PCR. For the 5′ end of the canine PKCδ cDNA, the primers used are: forward 5′-TGCAGGCCCCACCATGGCGCC-3′ (the initiation codon is underlined) and reverse 5′-TAGCCTTGCTTGTTGAGGC-3 (equivalent to nt. 560-578 of the canine PKCδ cDNA sequence). For the 3′ end of the canine PKCδ cDNA, the primers used are: forward 5′-ACCCTTCAGGCCAAAGTGAA-3′ (equivalent to nt. 1854-1873 of the canine PKCδ sequence) and reverse 5′-ATAACTACATTCAAGTAATGA-3 (based on the 3′-noncoding sequence of human PKCδ cDNA). The nucleotide sequence of the PCR-amplified fragments covering the full-length canine PKCδ cDNA was determined and submitted to GenBank with accession number AY825360.
Small interfering RNA (siRNA) for PKCδ
A 20-nucleotide sequence spanning nt. 144-163 of the canine PKCδ cDNA was selected for generating PKCδ-specific siRNA. The oligonucleotide 5′-GATCCCCGCCCACCATGTACCCTGAGTTCAAGAGACTCAGGGTACATGGTGGGCTTTTTGGAAA-3′ (nt. 144-163 of canine PKCδ are underlined) and its complementary strand were synthesized, annealed and cloned into the pSuperior vector (OligoEngine, Seattle, WA). The pSuperior vector is a tetracycline-regulated expression vector that utilizes the Tet operator. Tetracycline regulation in the pSuperior vector is based on the binding of tetracycline to the Tet repressor and derepression of the promoter controlling expression of the gene of interest. For the control siRNA, the oligonucleotide 5′-GATCCCCGTCCACATTCGACGCCCACTTCAAGAGAGTGGGCGTCGAATGTGGACTTTTTGGAAA-3 and its complementary strand were synthesized, annealed and cloned into the pSuperior vector. The resulting plasmid pSuperior-control-siRNA and pSuperior-PKCδ-siRNA were then used to generate control MDCK cells or those whose PKCδ expression can be suppressed upon addition of tetracycline or its analog doxycycline.
Cell culture and transfections
MDCK cells, COS cells, and human embryonic kidney (HEK) 293 cells were maintained in Dulbecco's Modified Eagle's Medium (Invitrogen Life Technologies) supplemented with 10% fetal bovine serum and cultured at 37°C in a humidified atmosphere of 5% CO2 and 95% air atmosphere. To generate stable MDCK cells that express PKCδ-specific siRNA upon addition of doxycycline, MDCK cells were transfected with pSuperior-PKCδ-siRNA and pcDNA6/TR (Invitrogen Life Technologies) at a ratio of 1:10 using LipofectAMINE. The plasmid pcDNA6/TR expresses Tet repressor. Stable cell clones were selected in the medium containing 0.5 mg/ml of neomycin G418 and screened for suppression of PKCδ expression by adding doxycycline (2 μg/ml) in the medium. 72 hours after induction of PKCδ siRNA, the level of PKCδ was analyzed by immunoblotting with polyclonal anti-PKCδ. Several cell clones whose PKCδ expression can be suppressed upon doxycycline addition were selected for further analysis.
To generate MDCK cells stably expressing GFP-PKCδ or FLAG-PKCδ, MDCK cells were transfected with pEGFP-N1-PKCδ or pcDNA3-FLAG-PKCδ using LipofectAMINE. Two days after transfection, cells were selected in medium containing 0.5 mg/ml neomycin G418. After approximately 10 days, neomycinresistant cells were pooled and analyzed for exogenous PKCδ expression by immunoblotting using anti-GFP or anti-FLAG. To generate MDCK cells stably expressing both of FLAG-PKCδ and GFP-adducin, MDCK cells that had already expressed FLAG-PKCδ were transfected with pEGFP-C1-adducin or pEGFP-C1-adducin-S726A. Two days after transfection, the cells were harvested and replated on 100-mm dishes at an appropriate density in the medium containing 0.5 mg/ml of G418. After approximately 14 days, cell colonies that emitted green fluorescence under a fluorescent microscope were picked using cloning cylinders and screened for expression of GFP-adducin by immunoblotting using anti-GFP. For transient transfection, COS cells were grown on 60-mm dishes for 18 hours, and then transfected with the plasmids as indicated using LipofectAMINE. 48 hours after transfection, cells were lysed in 1% Nonidet P-40 lysis buffer and the phosphorylation of adducin at the Ser726 was analyzed by immunoblotting with polyclonal anti-phospho-adducin.
Immunoprecipitation and immunoblotting
Cells were lysed in 1% Nonidet P-40 lysis buffer (1% Nonidet P-40, 20 mM Tris-HCl, pH 8.0, 137 mM NaCl, 10% glycerol and 1 mM Na3VO4) containing protease inhibitors (1 mM phenylmethylsulfonyl fluoride, 0.2 trypsin inhibitory units/ml aprotinin, and 20 μg/ml leupeptin). The lysates were centrifuged for 10 minutes at 4°C to remove debris, and the protein concentrations were determined using the Bio-Rad protein assay (Hercules, CA). For immunoprecipitation, aliquots of cell lysates were incubated with l μg of polyclonal anti-PKCδ, or 0.4 μg of monoclonal anti-GFP for 1.5 hours at 4°C. Immunocomplexes were collected on protein-A Sepharose beads. For monoclonal antibodies, protein-A Sepharose beads were coupled with rabbit anti-mouse IgG (1 μg) before use. The beads were washed three times with 1% Nonidet P-40 lysis buffer, boiled for 3 minutes in SDS sample buffer, subjected to SDS-polyacrylamide gel electrophoresis, and transferred to nitrocellulose (Schleicher and Schuell). Immunoblotting was performed with appropriate antibodies using the Amersham Biosciences enhanced chemiluminescence system for detection. Chemiluminescent signals were detected and quantified by Fuji LAS-3000 luminescence image system.
In vitro protein kinase assay
To perform in vitro kinase assays for PKC, the immunoprecipitates by anti-PKCδ, anti-PKCα or anti-GFP were washed three times with 1% NP-40 lysis buffer and once in 25 mM Tris buffer. Kinase reactions were carried out in 40 μl of kinase buffer (25 mM Tris-HCl, pH 7.5, 10 mM MgCl2, 1 mM CaCl2, 1 mM dithiothreitol, 1.25 μg phosphatidylserine) containing 10 μCi of [γ-32P]ATP (3000 Ci mmol-1; PerkinElmer Life Sciences) and MBP or purified GFP-adducin protein at 25°C for 20 minutes. Reactions were terminated by addition of SDS sample buffer, and the 32P-incorporated proteins were fractionated by SDS-polyacrylamide gel electrophoresis and visualized by autoradiography. The radioisotope activity was quantified using a phosphoimager system (Pharmacia).
Purification of GFP-adducin proteins
To obtain purified GFP-adducin proteins, HEK293 cells were transiently transfected with pEGFP-C1-adducin, pEGFP-C1-adducin-S716A, pEGFP-C1-adducin-S726A or pEGFP-C1-adducin-S716A/S726A using LipofectAMINE. 24 hours after transfection, cells were serum starved for 16 hours and then lysed in 1% Nonidet P-40 lysis buffer. The GFP-adducin proteins were immobilized on protein-A Sepharose with monoclonal anti-GFP, eluted in 0.1 M citric acid (pH 3.0), and neutralized in Tris-HCl buffer (pH 9.0). The purified proteins were fractionated by SDS-polyacrylamide gel electrophoresis and visualized by Coomassie Blue staining.
Cell spreading assay
MDCK cells were collected by trypsinization and suspended in serum-free medium at 5×104 cells/ml. 2 ml of cell suspension was added to a 60-mm dish that had been coated with collagen (10 μg/ml) and blocked with bovine serum albumin. Cells were allowed to spread, and spread cells were scored under a phase-contrast microscope at indicated intervals. Cells with extended processes and not phase bright were defined as spread cells.
Trans-well cell migration assay
MDCK cells were collected by trypsinization and suspended in serum-free medium at 2×105 cells/ml. Migration assays were carried out in a Neuro Probe 48-well chemotaxis chamber (Cabin John, MD), as described previously (Lai et al., 2000). Briefly, the medium containing type I collagen (10 μg/ml) was added to the lower chamber. The lower and upper chambers were separated by a polycarbonate membrane (8-μm pore size, Poretics, Livermore, CA). Cells were allowed to migrate for 6 hours at 37°C in a humidified atmosphere containing 5% CO2. The membrane was fixed in methanol for 10 minutes and stained with modified Giemsa stain (Sigma-Aldrich) for 1 hour. Cells on the upper side of the membrane were removed by cotton swabs. Cells on the lower side of the membrane were counted under a light microscope. Each experiment was performed in triplicate.
Wound healing assays and time lapse microscopy
MDCK cells were grown on glass coverslips with 0.17 mm in thickness and 42 mm diameter. The monolayer of cells was wounded by manual scratching with a pipette tip. Cells on the microscope stage were maintained at 37°C with a humid CO2 atmosphere in a micro-cultivation system (model POC-R, Zeiss) with temperature and CO2 control devices (tempcontrol 37-2 digital and CTI controller 3700 digital, Zeiss). Cells were monitored under differential interference contrast (DIC) optics on an inverted Zeiss microscope (Axiovert 100) using Zeiss 40× LD Achroplan objective. Time-lapse sequential micrographs were captured every 5 minutes using a cooled CCD camera (CoolSNAP fx, Roper Scientific) and analyzed by Meta Imaging Series™ software (version 4.5) from Universal Imaging Corporation (West Chester, PA).
Laser-scanning confocal fluorescent microscopy
Cells were plated on 12-mm collagen-coated glass coverslips for 24 hours, fixed for 15 minutes in phosphate-buffered saline containing 4% paraformaldehyde, and permeabilized in phosphate-buffered saline containing 0.04% Triton X-100 for 15 minutes. Coverslips were stained with primary antibodies at 4°C for overnight and followed by TRITC-conjugated secondary antibodies (Jackson ImmunoResearch Laboratories) at 4 μg/ml for 120 minutes. The primary antibodies used in immunofluorescence staining were diluted before use; monoclonal anti-FLAG (1:500), polyclonal anti-adducin (1:100), and polyclonal anti-phospho-adducin (Ser726) (1:100). TRITC-conjugated phalloidin at 2 μM was used to stain actin filaments. Coverslips were mounted in anti-fading solution and viewed using a Zeiss LSM510 laser-scanning confocal microscope image system with a Zeiss 63× Plan-Apochromat objective. Wavelengths 488 nm and 543 nm were used to excite GFP and TRITC, respectively. A beam path filter (BP 505-530 nm) and a long path filter (LP 560 nm) were used to acquire images for the emission from GFP and TRITC, respectively, in a multi-track channel mode.
Student's t-tests were used to determine whether there was a significant difference between two means (P<0.05); statistical differences are indicated with an asterisk.
We thank P. M. Blumberg for pEGFP-PKCδ, U. Kikkawa for pcDNA3-FLAG-PKCδ, D. Joubert for pEGFP-PKCα, and V. Bennett for pEGFP-adducin and its S716A/S726A mutant. This work was supported by National Science Council, Taiwan, Grants NSC94-2320-B-005-010, NSC95-2320-B-005-003 and NSC95-3112-B-005-001.