Cortactin is a scaffolding protein that is targeted to sites of actin polymerization, including lamellipodia in migrating cells, cell-cell junctions in epithelial cells, growth cones of neurons, podosomes of osteoclasts, invadopodia of tumor cells, and sites of actin rearrangement induced by pathogenic bacteria and viruses (Daly, 2004; Lua and Low, 2005; Pfaff and Jurdic, 2001). Although it is phosphorylated in response to a number of stimuli including growth factors, bacterial invasion, platelet activation, transformation and integrin-mediated adhesion, the only sites of phosphorylation that have been characterized to date are Y421, Y466 and Y482 (Src sites that regulate cell migration) and S405 and S418 (ERK sites that regulate N-WASP-mediated actin polymerization) (Campbell et al., 1999; Huang et al., 1998; Martinez-Quiles et al., 2004; Weed and Parsons, 2001). To better understand how the phosphorylation of cortactin contributes to the dynamic regulation of actin structures, we mapped the repertoire of phosphorylation sites within cortactin by mass spectrometry (MS). Herein, we report the identification of 17 new sites of phosphorylation (12 serine, four threonine and one tyrosine residues). The distribution of phosphorylation sites on well-conserved residues within cortactin supports a role for these modifications in the regulation of cortactin function.
Insights into the role of phosphorylation come from the analysis of the domain structure of cortactin which includes an N-terminal acidic (NTA) domain, a repeat domain (comprised of 6.5 copies of a 37 amino acid sequence), an α-helical and proline-rich (HPR) domain and a C-terminal Src homology 3 (SH3) domain (Fig. 1). The NTA domain of cortactin nucleates actin polymerization through direct interaction with the Arp2/3 complex (Weed et al., 2000). The NTA domain shares homology with the acidic domains found in other Arp2/3-binding proteins, and mutation of the conserved tryptophan (W22A) or the two adjacent aspartic acids (D20G and D21A) within the NTA of cortactin disrupts the interaction with the Arp2/3 complex and inhibits the nucleation of new actin filaments (Uruno et al., 2001; Weaver et al., 2003). The MS analysis showed that the conserved DDW sequence was flanked by three sites of phosphorylation (S11, T13 and T24) (Fig. 2A). Peptides containing phosphorylated S11 or T13 were among the most abundant peptides identified in this analysis (Table 1). Residues S11 and T24 are conserved between mouse, human, frog and fish cortactin sequences; residue T24 is maintained in fly and sponge (Table 2). The proximity of these phosphorylation sites to the DDW motif indicates that phosphorylation of S11, T14 and/or T24 contributes to the regulation of the association of cortactin with the Arp2/3 complex.
The repeat domain binds directly to filamentous (F)-actin, providing a branch point for nascent actin polymerization (Wu and Parsons, 1993). This domain contained seven sites of phosphorylation (S135, S150, S172, T182, S209, S282 and S283). Fig. 2B shows the alignment of the individual Repeats, with the phosphorylated sites shown in red. The same conserved serine (SAVG) was phosphorylated in four of the six repeats. Since the repeat domain is both necessary and sufficient for F-actin binding, we suggest that modification of sequences within this domain influences the structure of each repeat and plays a role in regulating the binding of cortactin to actin filaments.
Within the HPR domain, we detected seven sites of phosphorylation (S322, S345, S348, S407, S454, T468 and Y475) (Table 2). The peptide that was phosphorylated on S322 was one of the most abundant peptides identified. An additional site was mapped in peptides containing S417, S418 and Y421, but the data were ambiguous with respect to which residue(s) was phosphorylated. Previous studies have shown that cortactin is phosphorylated on S418 and Y421 (Campbell et al., 1999; Huang et al., 1998). It is possible that phosphorylation of these residues in the HPR contributes to conformational regulation of the SH3 domain or accessibility to the SH3 domain. The SH3 domain binds a number of effector molecules, including proteins that regulate actin polymerization (WIP, N-WASP,), endocytosis and vesicle trafficking (dynamin-2), GTPase regulators (FGD1, BPGAP1) and adaptor and/or scaffold proteins (CD2AP, CortBP1, ZO-1 and Shanks) (Daly, 2004; Lua and Low, 2005; Weed and Parsons, 2001). No sites of phosphorylation were identified in the C-terminal SH3 domain.
The MS analysis described herein adds to the repertoire of post-translational modifications that occur in cortactin. Previous studies had mapped sites of phosphorylation to S405, S418, Y421, Y466 and Y482 (Campbell et al., 1999; Huang et al., 1998). Of these, S418 and Y421 were identified as potential sites in the current study; the other sites were not seen under the conditions of these experiments. However, 17 new sites of phosphorylation were identified that were found in all domains of cortactin except the SH3 domain. Including the previously identified sites, 22 sites of phosphorylation in cortactin are currently known. The task is now to determine the differential regulation and functional consequences of these modifications in cortactin.
Materials and Methods
HEK293 cells were transfected with FLAG-cortactin plasmid (5 μg/107 cells) using Polyfect (QIAGEN). The cells were treated for 30 minutes with the serine/threonine phosphatase inhibitor calyculin (0.1 μM), the tyrosine phosphatase inhibitor sodium ortho-vanadate (5 mM) or were processed without any inhibitor treatment. FLAG-cortactin was captured from clarified cell extracts made in CSK-NP buffer [150 mM NaCl, 5 mM MgCl2, 20 mM Tris-HCl pH 7.4, 10% glycerol, 10 mM sodium ortho-vanadate, 2 mM sodium pyrophosphate, 1% NP-40, 1× EDTA-free protease inhibitors (Roche)] using M2-agarose affinity resin (anti-FLAG antibody conjugated to agarose, Sigma). After 2 hours of incubation at 4°C, the M2 agarose beads were sequentially washed with CSK-NP buffer, medium-salt buffer (150 mM NaCl, 5 mM MgCl2, 20 mM Tris-HCl pH 7.4) and high-salt buffer (1 M NaCl, 5 mM MgCl2, 20 mM Tris-HCl pH 7.4) to remove non-specific binding proteins. The cortactin protein was eluted with 0.3 mg/ml of FLAG peptide in MS buffer. Typically 30-40 pmoles of purified cortactin were recovered from 107 cells, and 2-10 pmoles of protein were used to initiate C18 reverse-phase (RP)-HPLC analysis and mass spectrometry. Aliquots of purified cortactin protein were digested with either chymotrypsin or Lys-C followed by chymotrypsin to maximize the coverage of peptides analyzed in the mass spectrometer. The resulting peptides were separated by RP-HPLC coupled to online detection by nanoflow LC, microcapillary ESI tandem mass spectrometry (LTQ-FTMS, Thermo Electron, San Jose, CA). In some cases enrichment of phosphopeptides was performed with immobilized metal affinity chromatography (IMAC) (Ficarro et al., 2002), prior to RP chromatography. In those cases, 10-30 pmoles of protein were loaded onto IMAC columns.
Individual MS/MS spectra from each analysis were searched against a FASTA database containing the protein sequence for mouse cortactin using the SEQUEST database-searching algorithm. The SEQUEST parameters file included a differential search for +80 atomic mass unit (amu) on serine, threonine and tyrosine residues to identify sites of phosphorylation. Oxidized methionine was identified by a differential search of +16 amu on the methionine residue. Static modification of +57 amu on cysteine was due to treatment with the protein with DTT-iodoacetamide before the proteolytic digest. For IMAC analyses, static modifications of +14 amu on glutamate, aspartate and the C-terminus of the peptide were included to account for conversion of carboxylic acid groups to methyl esters. Peptides that scored above Xcorr 3.0 and met the criteria described by the Yates group (Washburn et al., 2001), were deemed as valid hits. All hits that included a phosphorylation site were verified by manual interpretation of the corresponding MS-MS spectra.
Acknowledgements
The data discussed herein were obtained as a part of a study supported by the Cell Migration Consortium. A summary of the data and methods utilized, as well as analyses of other migration related proteins, can be accessed at www.cellmigration.org. P.R.G. is a member of the Research Career at the Centro de Virologia Animal-CONICET, Argentina. J.T.P. acknowledges additional support for these studies from NIH-NCI CA 29243.
