Modulation of the Catalytic Activity of the Src Family Tyrosine Kinase Hck by Autophosphorylation at a Novel Site in the Unique Domain*

Autophosphorylation is a key event in the activation of protein kinases. In this study, we demonstrate that autophosphorylation of the recombinant Src family kinase Hck leads to a 20-fold increase in its specific enzymatic activity. Hck was found to autophosphorylate readily to a stoichiometry of 1.3 mol of phosphate per mol of enzyme, indicating that the kinase autophosphorylated at more than one site. Solid phase sequencing and two-dimensional mapping of the phosphopeptide fragments derived from the autophosphorylated enzyme revealed that the kinase can undergo autophosphorylation at the following two sites: (i) Tyr-388, which is located to the consensus autophosphorylation site commonly found in the activation loop of many protein kinases, and (ii) Tyr-29, which is located in the unique domain of Hck. Hck purified from mouse bone marrow-derived macrophages could also autophosphorylate in vitro at both Tyr-388 and Tyr-29, indicating that naturally occurring Hck can also autophosphorylate at Tyr-29. Furthermore, Hck transiently expressed in human embryonic kidney 293T cells was found to be phosphorylated at Tyr-29 and Tyr-388, proving that Hck can also undergo autophosphorylation at both sitesin vivo. The recombinant enzyme carrying the mutation of Tyr-388 to Phe was also able to autophosphorylate at Tyr-29, albeit at a significantly slower rate. A 2-fold increase in the specific enzymatic activity was seen with this mutant despite the stoichiometry of autophosphorylation only approaching 0.2 mol of phosphate per mol of enzyme. This indicates that autophosphorylation of Tyr-29 contributes significantly to the activation of Hck. Regulation of the catalytic activity by phosphorylation of Tyr-29 in the unique domain may represent a new mechanism of regulation of Src family tyrosine kinases.

The Src family of protein tyrosine kinases consists of nine members documented to participate in a variety of cellular functions such as cellular growth and differentiation (see Refs. 1 and 2 for review). The catalytic activity of the Src family kinases is indispensable to their ability to carry out their cellular functions. Members of the Src family kinases share the same overall structural organization of domains as follows: (i) an N-terminal fatty acid-acylation domain, (ii) a unique domain, (iii) a Src homology 3 (SH3) 1 domain, (iv) a Src homology 2 (SH2) domain, (v) a catalytic domain, and (vi) a C-terminal regulatory domain. Phosphorylation of a conserved tyrosine residue within the C-terminal regulatory domain of Src family kinases by another protein tyrosine kinase called the C-terminal Src kinase (CSK) or its cellular homologue (Chk) leads to inactivation of the kinase. In contrast, autophosphorylation of a conserved tyrosine residue within the kinase domain leads to activation. The recently determined crystal structures for Src and Hck provide the structural basis for the inactivation of Src family kinases by phosphorylation of the C-terminal regulatory tyrosine residue (3)(4)(5)(6). The inactive conformation of Src or Hck is stabilized by a tripartite intramolecular interaction involving binding of the phosphorylated C-terminal regulatory tyrosine residue to the SH2 domain, and binding of the SH3 domain to a linker sequence between the SH2 and catalytic domains (SH2-CD linker) of the kinase that is capable of adopting a poly-proline type II helical conformation. Such interactions impart conformational constraints on the kinase domain such that it is unable to bind ATP, thus maintaining the enzyme in an inactive conformation. Activation of the enzyme may be achieved either by dephosphorylation of the C-terminal regulatory tyrosine residue (Tyr-499 in murine Hck) by a tyrosine phosphatase, engagement of the SH2 domain by a tyrosine-phosphorylated protein, or binding of the SH3 domain to a specific PXXP motif-containing cellular protein (7,29).
In addition to regulating the catalytic activity, the SH2 and SH3 domains of Src family tyrosine kinases are also involved in mediating their physical association with other proteins (8 -10). Such interactions are important for association of the kinases with substrates and targeting the kinases to their specific subcellular localizations.
The unique domain is also implicated as contributing significantly to the differences in regulatory properties and functions of the Src family kinases, as it is the region of the enzymes containing most sequence variation between different Src family members. The unique domain of Lck, for example, mediates its interaction with the transmembrane receptor-like proteins CD4 and CD8 in T-lymphocytes (11). The unique domain of Src can be phosphorylated by cAMP-dependent protein kinase (12), protein kinase C (13), or p34 Cdc2 (14) under various conditions. The functional significance of phosphorylation by these serine/ threonine kinases is not yet known, although phosphorylation by p34 Cdc2 was suggested to enhance activation by dephosphorylation of the C-terminal regulatory tyrosine of Src by the upstream protein tyrosine phosphatase (29).
In this paper, we report that autophosphorylation of Hck occurs at a novel site in addition to the consensus autophosphorylation site at Tyr-388 2 of the activation loop of the catalytic domain (27). This novel site is identified to be Tyr-29 2 in the unique domain. Autophosphorylation of Tyr-29 and Tyr-388 is shown to follow an intermolecular mechanism. Analysis of the activities and stoichiometric levels of phosphorylation of recombinant Hck mutants, containing a Tyr to Phe mutation at either Tyr-388 or Tyr-29, showed that phosphorylation at Tyr-29 contributes to the activation of Hck by autophosphorylation. In vitro autophosphorylation of Tyr-29 in Hck isolated from macrophages and in vivo phosphorylation of Tyr-29 in Hck overexpressed in the human embryonic kidney 293T cells were also demonstrated, indicating that autophosphorylation of Tyr-29 in Hck is of physiological significance.

EXPERIMENTAL PROCEDURES
Materials-The Sephadex G-25 (Fine) gel filtration matrix, the fast protein liquid chromatography Mono-Q anion (HR5/5), and the Mono-S (HR5/5) columns were from Amersham Pharmacia Biotech. Hydroxylapatite was from Bio-Rad. The preparative reversed phase Econosil C 18 high performance liquid chromatography column was from Alltech (Deerfield, IL), and the analytical Vydac reverse phase C 18 high performance liquid chromatography column was from Separation Group Inc. The polyclonal ␣-Hck peptide antibody (M28) was Santa Cruz Biotechnology (Santa Cruz, CA). The rabbit polyclonal ␣-Hck antibody was raised against a GST fusion protein containing the unique and SH3 domains (amino acid residues 1-129) of Hck. Recombinant Hck was expressed and purified as described by Sicilia et al. (15). Immunoblot analysis using anti-phosphotyrosine antibodies revealed that less than 5% of the purified Hck was tyrosine-phosphorylated (15). Oligonucleotides were purchased from GeneWorks, Inc. (South Australia).
Determination Hck, respectively) were withdrawn at various time points, and the autophosphorylation reaction was stopped by the addition of 5ϫ SDS-PAGE sample buffer. The aliquots were then run on a 7.5% SDS-polyacrylamide gel. The gel was dried and then analyzed by autoradiography. The stoichiometry of autophosphorylation of Hck was then determined by cutting out from the dried gel the bands of radioactively labeled Hck and subjecting them to liquid scintillation counting. Results of the experiment are presented in Fig. 1B and Fig. 6, A and B.
For determination of the kinase activity, aliquots of 10 l were taken at the same time intervals and added to 50 l of the dilution buffer (25 mM Hepes, pH 7.0, 5% Nonidet P-40, 1 mM EDTA, 0.1 mg/m benzamidine, and 20% glycerol). From this dilution, 10 l of the diluted kinase was added to 15 l of assay mix with the final mixture containing 100 M ATP (specific radioactivity of approximately 600 cpm/pmol), 20 mM Tris-HCl, pH 7, 10 mM MgCl 2 , 1 mM MnCl 2 , and 50 M Na 3 VO 4 , and 300 M [Lys 19 ]Cdc2-(6 -20)-substrate peptide. The reaction was allowed to continue for 10 min before being stopped with 10 l of 50% acetic acid. Aliquots of 26 l of the stopped reaction mix were then spotted onto P81 filter paper and the papers washed 5 times in 300 ml each of 0.5% v/v phosphoric acid. The papers were washed a final time in acetone and dried for liquid scintillation counting. Results of the kinase assay are presented in Fig. 1C and Fig. 6, C and D.
Phosphopeptide Mapping-Recombinant wild type, [Y29F]Hck, or [Y388F]Hck were allowed to autophosphorylate at 30°C for 60 min in the kinase assay buffer and 50 M [␥-32 P]ATP (2000 cpm/pmol). The autophosphorylation reaction was stopped by the addition of 5ϫ SDS-PAGE sample buffer, prior to running on a 7.5% polyacrylamide gel. Following transfer to a nitrocellulose filter, the bands corresponding to the two isoforms of radioactively phosphorylated Hck were excised and blocked with 0.1% polyvinylpyrrolidone 40 in 100 mM acetic acid, washed with 8 ϫ 1 ml of Milli-Q H 2 O, and digested overnight with 29 l of 0.2 mg/ml of L-1-tosylamido-2-phenylethyl chloromethyl ketonetreated trypsin (Worthington) in 100 mM NH 4 HCO 3 , pH 7.8, acetonitrile (95:5, v/v) at 37°C for 15 h (30). After digestion, the supernatant was lyophilized three times following the addition of 1 ml of Milli-Q H 2 O and the resulting dry material dissolved in 10 -29 l of water. Samples were spotted onto crystalline cellulose Machery-Nagel TLC plates. Separation of the proteolytic fragments was accomplished by (i) thin layer electrophoresis (TLE) in the first dimension and (ii) by thin layer chromatography in the second dimension as described previously (15,18). The radioactive tryptic fragments were located by autoradiography. The identity of the novel autophosphorylation site was verified by demonstrating the co-migration of a 32 P-labeled enzymatically phosphorylated synthetic phosphopeptide standard corresponding to residues 26 -38 of Hck (pY-29 standard) (Fig. 3, B and D). Hck, purified by immunoprecipitation from bone marrow-derived macrophages, was also phosphorylated in vitro and subjected to phosphopeptide mapping ( Fig.  4)  Culture and Preparation of Bone Marrow-derived Macrophages-Bone marrow-derived macrophages were prepared from bone marrow precursors isolated from CBA mouse femurs as described (20).
Cell Culture, Transient Transfection, and Metabolic Labeling of Human Embryonic Kidney 293T Cells-Human embryonic kidney (HEK) 293T cells were maintained in DMEM supplemented with 10% fetal calf serum. HEK 293T cells (10 6 cells per 10-cm tissue culture dish) were transfected for 4 h with 10 g of transient mammalian expression vector pCDM8 (Invitrogen) plasmid cloned into the XbaI site with either wild type Hck or the activated Hck mutant with Tyr-499 replaced by phenylalanine in the presence of 50 g/ml polyethyleneimine. For metabolic labeling of transfected HEK 293T cells with [ 32 P]orthophosphate, the cells were washed twice with phosphate-free DMEM 65 h posttransfection and then incubated for 1 h in 5 ml of phosphate-free DMEM containing 1% dialyzed fetal calf serum. The medium was then removed and replaced with 5 ml of phosphate-free DMEM containing 1% dialyzed fetal calf serum and 1 mCi of [ 32 P]orthophosphate, and the 2 The numbering system for the amino acid residue is based upon the sequence of the 56-kDa isoform of mouse Hck (GenBank TM accession number NM010407) with the novel autophosphorylation site, the consensus autophosphorylation site, and the C-terminal regulatory phosphorylation being Tyr-29, Tyr-388, and Tyr-499, respectively. For the 59-kDa isoform, these phosphorylation sites correspond to Tyr-50, Tyr-409, and Tyr-520. cells were incubated for 3 h. The cells were washed three times with ice-cold phosphate-buffered saline and then lysed as described below.
Cell Lysis and Immunoprecipitation of Hck-The bone marrow-derived macrophages and the HEK 293T cells overexpressing Hck were washed three times with ice-cold phosphate-buffered saline and then lysed in Triton X-100 lysis buffer (TLB; 1% Triton X-100, 25 mM Hepes, pH 7.4, 137 mM NaCl, 10% glycerol, 10 mM MgCl 2 , 1 mM EDTA, 1 mM vanadate, 40 mM ␤-glycerophosphate, 50 mM NaF, 0.1 mM Pefabloc, 10 M leupeptin, 10 M pepstatin A, and 10 g/ml aprotinin). Aliquots of cell lysate were precleared with 40 l of a 1:1 slurry of protein A-Sepharose in TLB and then incubated at 4°C overnight with 5 g of anti-Hck antibody. Immune complexes were collected on protein A-Sepharose beads and washed five times with TLB. The beads were then heated for 5 min in SDS-PAGE sample buffer prior to electrophoresis on a 7% SDS-polyacrylamide gel.
Solid Phase Sequencing of Phosphopeptides-Hck (0.9 g) was autophosphorylated at 30°C for 60 min to a stoichiometry of 1.2 mol of phosphate/mol of kinase by incubation in the kinase assay buffer and 29 M [␥-32 P]ATP (2900 cpm/pmol). The reaction was terminated by the addition of 5ϫ SDS sample buffer. Hck and other components of the reaction mixture were separated by SDS-PAGE. After drying the gel, the location of the autophosphorylated protein bands was determined by autoradiography and excised. The gel slices were washed several times with milli-Q H 2 O (3 ϫ 1.5 ml) for 1 h before drying under vacuum. The gel piece was rehydrated with 1 ml of 0.5 M Tris-HCl, pH 8.0, 0.1 mM CaCl 2 , 6 g/ml sequencing grade trypsin (Promega) before incubation at 37°C for 24 h. The supernatant was then collected, and peptides were separated by reversed phase HPLC on a 5-m Waters Delta-Pak ODS column using a 0.1% trifluoroacetic acid in H 2 O, 0.1% trifluoroacetic acid in acetonitrile gradient. Absorbance was monitored at 214 nm, and the entire column eluate was collected in 1-min fractions for subsequent Cerenkov counting. Fractions containing 32 P radioactivity were then spotted onto a disc of Millipore Sequelon-AA TM membrane in 5-l aliquots and dried. The peptide sample was then covalently linked to the disc by the addition of 6 l of coupling buffer (10 mg/ml 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide hydrochloride, 160 mM MES, pH 4.35, 0.4 M guanidinium hydrochloride, and 20% acetonitrile) and incubated for 1 h at 37°C. The disc was then washed twice with 5 ml of 20% acetonitrile (in H 2 O) before being loaded onto an Applied Biosystems 477A Protein Sequenator. One-third of the resulting PTH-derivative sample for each cycle was then subjected to PTH-derivative analysis with an on-line model 120 PTH-derivative analyzer, and the remaining fraction was collected for scintillation counting (21). Results of the solid phase sequencing experiment are shown in Fig. 2.
Construction of the Hck Baculovirus Vectors for the Expression of Hck Carrying the Y29F and Y388F Mutations-Two polymerase chain reactions were carried out using full-length Hck cloned into the XbaI site of the transient mammalian expression vector pCDM8 (Invitrogen) as the template. The first reaction for the production of the Y29F mutant Hck was carried out using primer 1 (5Ј-TCGAAATTAATACGACTCACTAT-AGGGAGA) derived from the pCDM8 vector, 5Ј to the Hck sequence, and primer 2 (5Ј-CTGTGTTTGTGCCGGATCCCACG) containing the required A to T mutation for the Y29F mutation of Hck. The changed residue is denoted in boldface italic. The amplified PCR product of the first reaction was used as the megaprimer and together with primer 3 (5Ј-CACACCACAGAAGTAAGGTTCCTTCACAA) derived from the pCDM8 sequence to produce the full-length Hck containing the Y29F mutation in the second PCR. A similar procedure was used to produce the Y388F mutant of Hck with primer 2 replaced by another primer corresponding to the region of the Hck genetic sequence encoding for Tyr-388 and containing the correct mutation for the conversion of Tyr-388 to Phe (5Ј-ATGAGTTCACAGCTCGGGA-AGGAG). The changed residue is denoted in boldface italic.
The resultant full-length polymerase chain reaction products were digested with XbaI, purified, and ligated directly into the pBacPAK9 (CLONTECH) transfer vector pre-digested with XbaI. Clones containing the vectors in the correct orientation were determined by restriction digestion analysis.
Generation Hck baculovirus, and cells were harvested 2 days following infection. All extraction and purification procedures were carried out at 4°C. Cells were spun down at 1000 ϫ g for 5 min, washed once with Grace's serum-free medium (22), then homogenized in lysis buffer containing 25 mM Hepes, pH 7.0, 5% Nonidet P-40, 1 mM EDTA, 0.1 mg/m benzamidine, 0.1 mg/ml phenylmethylsulfonyl fluoride, and 1 mM dithiothreitol. The homogenate was clarified by centrifuging at 100,000 ϫ g for 40 min. The recombinant Hck was purified by sequential column chromatography as described in Ref. 15.
Production of the pGEX6p-3 Vector for the Expression of Hck-(1-91)-A polymerase chain reaction was carried out using full-length Hck cloned into the XbaI site of the transient mammalian expression vector pCDM8 as the template. The reaction was carried out using primer 1 (5ЈCCCGAATTCGCTGGGG GGTCGGTCTAGCTGCGAG) and primer 2 (5ЈGCGGATCCCTCGAGTGAATAGCCTCATAGTCGTACAG). The restriction sites for EcoRI and XhoI are in boldface italic. The resulting polymerase chain reaction product was purified using the QIAquick PCR purification kit (Qiagen) and then digested with EcoRI and XhoI. The digested product was again purified using the QIAquick PCR purification kit and directionally cloned into pGEX6p-3 (Amersham Pharmacia Biotech) expression vector pre-digested with EcoRI and XhoI.
Expression and Purification of Hck-(1-91)-Escherichia coli DE3 cells were transformed with pGEX6p-3 expression vector cloned with the sequence encoding for Hck-(1-91) and expanded to a 1.5-liter culture in Luria-Bertani medium containing 100 g/ml ampicillin. The induction of GST-Hck-(1-91) expression was carried out as described previously (15). The purified GST-Hck-(1-91) fusion protein was dialyzed against 2ϫ 1 liter of cleavage buffer containing 50 mM Tris-HCl, pH 7.0, 150 mM NaCl, 1 mM EDTA, and 1 mM dithiothreitol. It was digested with the Precision Protease according to the manual supplied by the manufacturer (Amersham Pharmacia Biotech). GST cleaved from Hck-(1-91) was removed by reapplying the digest to a glutathione-Sepharose column pre-equilibrated with the cleavage buffer. The column flow-through containing the Hck-(1-91) was applied to a fast protein liquid chromatography Mono-S column pre-equilibrated with washing buffer containing 25 mM Hepes, pH 7.0, 1 mM EDTA, 1 mg/ml phenylmethylsulfonyl fluoride, and 1 mM dithiothreitol, and the bound protein was eluted using a 10-ml linear gradient of 0 -0.6 M NaCl at a flow rate of 0.5 ml/min. Concentration Dependence of Autophosphorylation at Tyr-388 and Tyr-29 -Autophosphorylation reactions were carried out in the presence of the kinase assay buffer and 50 M [␥-32 P]ATP using constant amounts of Y29F and Y388F Hck (0.55 and 3.2 pmol respectively). However, final reaction volumes were varied such that the effect of differing enzyme concentrations on autophosphorylation could be determined for each mutant. Final enzyme concentrations were 114, 57, 28.5, 14.7, and 7.13 nM for [Y29F]Hck, and 1, 0.5, 0.25, 0.125, and 0.0625 M for [Y388F]Hck. Autophosphorylation reactions were carried out at 30°C for 15 min. Reactions were stopped by the addition of 5ϫ SDS-PAGE sample buffer prior to running on a 7.5% polyacrylamide gel. Bands were visualized by autoradiography and excised from the gel for 32 P liquid scintillation counting. The specific radioactivity of the [␥-32 P]ATP was used to determine the rate of autophosphorylation in the unit of picomoles of phosphate incorporated per min. Less than 5% of the Hck mutant was autophosphorylated at the end of the reaction, indicating that the initial velocity of autophosphorylation was measured. The experiment was repeated five times. In all five assays, the same trend of change of the autophosphorylation velocities versus enzyme concentrations was observed. The results shown in Fig. 7 represents the data obtained from one of the assays.
Phosphorylation of Hck-(1-91) by Full-length Hck-3.4 pmol of wild type, recombinant Hck was incubated in the presence of 100 M [␥-32 P]ATP with 0, 0.036, 0.073, 0.15, 0.29, and 1.46 nmol of Hck-(1-91) in the kinase assay buffer for 30 min in a total volume of 25 l. 10-l aliquots were taken from each reaction mixture and added to 10 l of sample buffer prior to running on a 10% polyacrylamide gel. Since the wild type Hck can autophosphorylate at both Tyr-29 and Tyr-388, a primary antibody specific for phosphotyrosine 29 (anti-pY-29 Hck antibody) was used to quantitate the rate of Tyr-29 phosphorylation in intact Hck and Hck-(1-91). Following transfer to nitrocellulose, the blot was probed with a primary antibody specific for Hck phosphorylated at Tyr-29 (anti-pY-29 Hck antibody). The blot was then washed and probed with anti-rabbit horseradish peroxidase-conjugated secondary antibody, and the protein bands were visualized by enhanced chemiluminescence (Amersham Pharmacia Biotech). Results of the experiment are shown in Fig. 13.
Production of the Phospho-Tyr-29-specific Hck Polyclonal Anti-body-A phosphopeptide raised against the sequence CASKTEPSAN-QKGPV(pY)VPDPTSSSKLGGKK, encompassing the region of the unique domain of Hck containing Tyr(P)-29, was synthesized, coupled to keyhole limpet hemocyanin (Calbiochem), and used as the antigen for the production of polyclonal antibody. Crude serum was first purified by ammonium sulfate precipitation according to procedures detailed in Ref. 23. The non-phosphorylated version of the peptide antigen was synthesized, purified, and covalently coupled to Affi-Gel-10 (Bio-Rad). The non-phosphorylated peptide column was used to deplete from the antiserum of the antibody against the non-phosphorylated version of the peptide. The antiserum was applied to the non-phosphorylated peptide column, washed (10 mM Tris, pH 7.5, 0.5 M NaCl), and eluted with 100 mM glycine pH 2.5. The flow-through fraction, containing the antibody against the pY-29 peptide was collected and saved for the next affinity column step. A second affinity column was produced by covalently coupling the phosphorylated peptide used as the antigen to Affi-Gel-10 (Bio-Rad). The flow-through fraction from the first affinity column step was preincubated with 1 mM phosphotyrosine and 1 mM non-phosphorylated peptide prior to application to the column to minimize binding of the nonspecific, phosphotyrosine targeting antibody. After washing, the bound antibody was eluted with 100 mM glycine, pH 2.5. As shown in Fig. 10, the antibody eluted from the Tyr(P)-29 peptide column binds to Hck only when it is phosphorylated at Tyr-29. This antibody is called the anti-pY-29 Hck antibody. Characterization of the Anti-pY-29 Hck Antibody-0.98 g of recombinant Hck was incubated in the presence of kinase assay buffer and 100 M [␥-32 P]ATP in a total volume of 25 l. 5-l aliquots were removed at 0, 15, 30, and 60 min, and the reaction was stopped by the addition of 30 l of 5ϫ SDS-PAGE sample buffer prior to running 10 l (containing 55 ng of Hck) of this in duplicate on 7.5% polyacrylamide gels. As controls, 22.5 ng of pp60 c-Src and 300 ng of pp53/56 Lyn were incubated in the presence of kinase buffer, 100 M [␥-32 P]ATP, and the presence or absence of 225 ng of CSK in a final volume of 5 l. The reactions were incubated for 1 h at 30°C and then stopped by the addition of 17.5 l of SDS-PAGE sample buffer prior to running 2ϫ 10-l aliquots (containing 10 ng of phosphorylated c-Src or 133.3 ng of phosphorylated Lyn) on the same duplicate 7.5% SDS-PAGE gels used for the Hck autophosphorylation time course samples. One of the gels was transferred to a nitrocellulose membrane, and this was then incubated with the purified anti-Hck pY-29 antibody. The blot was then washed, incubated with sheep anti-rabbit secondary antibody conjugated to horseradish peroxidase, and developed with an enhanced chemiluminescence kit according to the specifications of the manufacturer. The second gel was dried and analyzed by autoradiography. Results of the experiment are shown in Fig. 10, A (24). HEK 293T cells (10-m culture dish) were transfected for 4 h with a total to 12 g of plasmid in the presence of 50 mg/ml polyethyleneimine (25).

Comparison of the Amounts of Endogenous Hck in Bone Marrowderived Macrophage and the Bac1.2F5 Macrophage Cell Line with Those in the Transfected HEK 293T Cells Overexpressing Wild Type Hck and Its
Mutants-Crude cell lysate containing 17-100 g of total proteins from each cell type was analyzed by immunoblotting with the anti-Hck antibody. The amount of Hck in the cell lysate was quantitated by comparing the immunoreactive signal with those of known amounts of purified recombinant Hck (data not shown). The amounts of endogenous Hck in Bac1.2F5 cells and bone marrow-derived macrophages are, respectively, 0.01 g per mg of total proteins and 0.06 g per mg of total proteins. For the transfected HEK 293T cells, the levels of Hck and its mutants are 1.4 g per mg of total proteins, 0.46 g per mg of total proteins, and 0.47 g per mg of total proteins, respectively, for cells overexpressing the wild type Hck, [Y499F]Hck, and [Y29F]Hck. Thus, the levels of Hck in the lysates of the transfected cells are 2-14-fold higher than those of endogenous Hck in the macrophages. Given that the efficiency of transfecting the HEK 293T cells was approximately 20%, the levels of Hck in the successfully transfected HEK 293T cells would be 10 -70-fold above the endogenous Hck levels in bone marrow-derived macrophages and Bac1.2F5 cells. Aliquots of cell lysate containing 1.5 mg of protein were incubated with rat anti-Hck monoclonal antibody for 3 h at 4°C. The reaction was then cleared for nonspecific complexes by incubation with protein A-Sepharose for an additional 1 h at 4°C with mixing. The supernatant was then incubated with protein G-Sepharose beads for 90 min at 4°C with mixing. The immunoprecipitates were collected by centrifugation and washed with five times with 1 ml of lysis buffer. The washed immunoprecipitates were heated at 95°C for 5 min in 1ϫ SDS-PAGE sample buffer and then fractionated on 10% SDS-polyacrylamide gels, transferred to nitrocellulose, and analyzed by Western blot, sequentially probing with anti-phosphotyrosine, anti-pY-29 Hck, and anti-Hck antibodies. Results of the experiment are shown in Fig. 11.

RESULTS
Activation and Determination of Stoichiometry of Autophosphorylation of Wild Type Hck-Hck requires autophosphorylation in order to become fully activated. Upon incubation with ATP and Mg 2ϩ , recombinant wild type Hck readily undergoes autophosphorylation, achieving a stoichiometry of 1.3 mol of phosphate per mol of kinase after 40 min (Fig. 1, A and B). Concomitant with the increase in autophosphorylation was a corresponding 20-fold increase in specific activity of Hck (Fig.  1C). The finding that Hck achieves a stoichiometry of autophosphorylation greater than 1 mol of phosphate per mol of kinase clearly suggests that Hck is capable of autophosphorylating at more than one site.
Identification of Tyr-388 and Tyr-29 as the in Vitro Autophosphorylation Sites of Hck-Phosphoamino acid analysis of recombinant Hck that had been allowed to autophosphorylate to a stoichiometry of 1.5 mol of phosphate/mol of protein revealed that phosphorylation of Hck had only occurred on tyrosine (data not shown). As shown in Fig. 2A, when tryptic digests of both the 56-and 59-kDa isoforms of autophosphorylated Hck were subjected to reversed phase HPLC, the presence of two radioactively labeled phosphopeptides (I and II) in the digests were detected (Fig. 2A). Again, this finding suggests that Hck is capable of autophosphorylation at two sites.
Tryptic peptides I and II ( Fig. 2A) were subjected to solid phase sequencing in order to allow the sites of autophosphorylation in Hck to be identified. For phosphopeptide I, most of the radioactivity appeared in the seventh cycle of Edman degradation (Fig. 2B). As Hck-(382-391) is the only theoretical tryptic fragment derived from Hck with a tyrosine at the seventh position (Table I), phosphopeptide I is most likely the phosphorylated form of the Hck-(382-391) fragment. This fragment contains Tyr-388, which is known to correspond to the consensus autophosphorylation site found in all Src family kinases; therefore, further characterization of this site was not undertaken.
Solid phase sequencing of phosphopeptide II released most of the radioactivity at the fourth cycle of Edman degradation (Fig.  2C). Hck-(26 -36) with sequence GPVYVPDPTSSSK is the only predicted fragment generated from exhaustive tryptic digestion of Hck that has a tyrosine (Tyr-29) as the fourth residue from the N terminus (Table I). The identity of phosphopeptide II as the phosphorylated form of the Hck-(26 -36) fragment was further verified by identifying the PTH-derivative released in each cycle of Edman degradation. The sequence obtained was GPVXVPX in which "X" represents a cycle in which no PTHderivative was identified. The fourth cycle of Edman degradation would have released the phosphorylated PTH-Tyr-29 that broke down during the procedure and therefore could not be seen. However, the significant burst of 32 P radioactivity recorded for this fourth cycle establishes that Tyr-29 was phosphorylated (Fig. 2C). These data strongly suggest that Tyr-29 is an autophosphorylation site of Hck.
To confirm further Tyr-29 as one of the in vitro autophosphorylation sites, the ability of phosphopeptide fragment II to co-migrate with a radioactively phosphorylated synthetic pY-29 peptide standard GPV(pY)VPDPTSSSK when subjected to two-dimensional phosphopeptide mapping was examined. As shown in Fig. 3, two closely migrating but distinguishable radioactive spots corresponding to the tryptic fragments of the two autophosphorylation sites were detected in the phosphopeptide maps of both the 56-and 59-kDa isoforms of phosphorylated Hck (Fig. 3, A and C). Upon mixing of the pY-29 peptide standard with the tryptic digest of each isoform of phosphorylated Hck, spot II was found to co-migrate with the standard, indicated by its increase in intensity (Fig. 3, B and  D). As Tyr-388 is the other major phosphorylation site in the autophosphorylated Hck (Fig. 2), spot I should correspond to the pY-388-containing tryptic fragment.
Taken together, these results unequivocally identify Tyr-29 in the unique domain as one of the autophosphorylation sites in both isoforms of Hck. This is the first observation that an Src family kinase is capable of autophosphorylating a tyrosine residue in its unique domain.
Hck Isolated from Macrophages Also Undergoes Autophosphorylation at Tyr-29 -Since autophosphorylation of Hck at Tyr-29 was identified in experiments that used recombinant Hck expressed in Sf9 cells, we wanted to establish if Hck that The two purified radioactive fragments (I and II) were analyzed by solid phase sequencing. B, the profile of radioactivity associated with the PTH-derivative generated in each cycle of Edman degradation of phosphopeptide fragment I. C, the profile of radioactivity associated with the PTH-derivative generated in each cycle of Edman degradation of phosphopeptide fragment II. PTEYIQSVLDDFYTATESQYQQQP had been expressed under normal physiological conditions (e.g. in macrophages) also undergoes autophosphorylation at Tyr-29 and Tyr-388 in vitro. Accordingly, Hck was immunoprecipitated from bone marrow-derived mouse macrophages and allowed to autophosphorylate in vitro in the presence of [␥-32 P]ATP. Analysis of the autophosphorylation reaction by SDS-PAGE followed by autoradiography revealed the presence of two major radioactive bands, corresponding to the 56-and 59-kDa isoforms of Hck (data not shown). The radioactive bands corresponding to the two Hck isoforms were separately excised from the gel and digested with trypsin. Since the amount of protein present in the tryptic digests was considerably less than that obtained in experiments where we had used recombinant Hck expressed in Sf9 cells, we were unable to purify the radioactively labeled peptides by HPLC for direct amino acid sequencing. Consequently, the sites of phosphorylation in Hck were determined by subjecting the tryptic digests to two-dimensional phosphopeptide mapping. Two radioactive spots were detected on the two-dimensional TLE/TLC map for both the 56-and 59-kDa isoforms of Hck (Fig. 4 A and C). Of the two labeled spots, spot II co-migrated with the pY-29 phosphopeptide standard (Fig. 4, B and D Hck were allowed to autophosphorylate and were subsequently analyzed by two-dimensional phosphopeptide map-ping with reference to the wild type enzyme (Fig. 5). Wild type Hck produces two spots corresponding to phosphorylation of Tyr-29 and Tyr-388 (Fig. 5A), whereas the [Y29F]Hck and [Y388F]Hck mutants autophosphorylate at a single site corresponding to Tyr-388 (Fig. 5B) and Tyr-29 (Fig. 5C), respectively. To prove that the sites of phosphorylation of [Y29F]Hck and [Y388F]Hck correspond to Tyr-388 and Tyr-29, respectively, tryptic digests of these autophosphorylated mutants were mixed with a tryptic digest of the wild type Hck (Fig. 5, D and E). As expected, mixing the tryptic digests of phosphorylated [Y29F]Hck and wild type Hck leads to an increase in intensity of the spot corresponding to the fragment containing Tyr(P)-388 with respect to that containing Tyr(P)-29 (Fig. 5D), whereas mixing the tryptic digests of phosphorylated [Y388]Hck and wild type Hck leads to an increase in the intensity of the spot corresponding to the fragment containing Tyr(P)-29 with respect to that containing Tyr(P)-388 (Fig. 5E). This proves that the sites phosphorylated upon autophosphorylation of [Y29F]Hck and [Y388F]Hck correspond to Tyr-388 and Tyr-29, respectively. A faint spot is evident in the phosphopeptide maps above the origin. This spot is likely to be derived from Tyr(P)-499, suggesting that a very low level of autophosphorylation also occurred at Tyr-499. shown to increase to 1 mol of phosphate incorporated per mol of enzyme over the 50-min time course (Fig. 6A). Concomitant with this was a 15-fold increase in catalytic activity peaking at 30 min (Fig. 6C).

Modulation of Tyrosine Kinase Activity of Hck by Phosphorylation of Tyr-388 and
In contrast, the rate of [Y388F]Hck autophosphorylation was much slower than that of [Y29F]Hck autophosphorylation. The stoichiometry of phosphorylation of [Y388F]Hck was found to approach 0.2 mol of phosphate incorporated per mol of enzyme over the 50-min autophosphorylation time course (Fig. 6B). The results indicate that Tyr-29 was autophosphorylated with an efficiency much lower than that of Tyr-388 autophosphorylation. Intriguingly, even though Tyr-29 does not reside in the catalytic domain, an approximately 2-fold increase in catalytic activity was also evident over the time course of autophosphorylation (Fig. 6D) (Fig. 7). Autophosphorylation at Tyr-388 (Fig. 7A) and at Tyr-29 (Fig. 7B)  concentration-dependent, intermolecular mechanism.
Hck in Macrophages Is Predominantly in the Down-regulated, Tyr-499-phosphorylated Form-In order to determine if autophosphorylation of Hck at Tyr-29 also occurs in vivo, cells from a mouse macrophage cell line (RAW264.7 cells) were metabolically labeled with 32 P i so as to allow endogenous Hck to incorporate [ 32 P]phosphate. Hck was then immunoprecipitated from lysates of the 32 P-labeled cells and subjected to two-dimensional phosphopeptide mapping. Only a very faint radioactive spot located to the right of the origin, in the region where tryptic phosphopeptides containing Tyr-388 and Tyr-29 (i.e. spots I and II) migrate, was detected (Fig. 8). Therefore, phosphorylation at Tyr-29 could not be verified. We also attempted to demonstrate in vivo autophosphorylation of endogenous Hck at Tyr-29 using the anti-pY-29 Hck antibody. The antibody failed to detect any Tyr-29 phosphorylation. The results suggest that the Tyr-29 of Hck is predominantly in the non-phosphorylated form.
Four strong radioactive spots to the left of the origin are obvious in the phosphopeptide map shown in Fig. 8. As indicated by comparison of the phosphopeptide maps of wild type Hck and [Y499F]Hck phosphorylated in vivo (Fig. 9), the four spots correspond to the tryptic fragments containing Tyr(P)-499.
Thus the phosphopeptide map shown in Fig. 8 suggests that the majority of Hck in RAW264.7 cells is phosphorylated at Tyr-499. Tyr-499 phosphorylation keeps the kinase in an inactive state and does not permit autophosphorylation at Tyr-388 and Tyr-29. We are currently attempting to identify physiological conditions that lead to activation of Hck by inducing dephosphorylation of Tyr-499 and autophosphorylation of both Tyr-388 and Tyr-29.
Phosphorylation at Tyr-29 Occurs in Vivo-As an alternative approach to determine if Hck undergoes autophosphorylation at Tyr-29 in vivo, an activated form of Hck was created by replacing the regulatory tyrosine at position 499 with phenylalanine ([Y499F]Hck). This mutant along with wild type Hck were transiently expressed in HEK 293T cells. The cells were then metabolically labeled with [ 32 P]phosphate. Wild type and [Y499F]Hck were immunoprecipitated from lysates of the labeled cells and subjected to two-dimensional phosphopeptide mapping. As shown in Fig. 9, two radioactive spots that exhibited electrophoretic and chromatographic properties identical to spots I and II in Fig. 3 were detected.
When the tryptic digests of the phosphorylated wild type Hck and [Y499F]Hck were mixed with the pY-29 phosphopeptide standard, co-migration of the standard with the putative pY-29-containing tryptic fragment of Hck occurred, conclusively proving that phosphorylation of Tyr-29 occurs in vivo.
Four additional radioactive spots were found in the tryptic phosphopeptide map of the wild type Hck phosphorylated in The immunoprecipitate was analyzed by SDS-PAGE and transferred to nitrocellulose, and the radioactively labeled Hck was processed for two-dimensional phosphopeptide mapping. The four-pronged arrow denotes spots resulting from phosphopeptide fragments containing the phospho-Tyr-499 of Hck. The directions of thin layer electrophoresis and thin layer chromatography are denoted, as is the faint spot corresponding to phosphorylated Tyr-29 or Tyr-388. vivo (Fig. 9B), but not in the tryptic phosphopeptide map of the [Y499F]Hck (Fig. 9A), suggesting that the four radioactive fragments contain the in vivo phosphorylated Tyr-499 residue. The multiple phosphopeptide fragments generated from C-terminal regulatory tyrosine (Tyr-499) in Figs. 8 and 9 are attributed in part to the presence upstream from Tyr-499 of two Arg-Pro bonds that are known to be refractory to tryptic cleavage (26). Four conspicuous radioactive spots of similar pattern are also found in the phosphopeptide map (Fig. 8) of Hck isolated from RAW 264.7 cells, further confirming that Tyr-499 of Hck is phosphorylated in this cell line.
Characterization of the anti-pY-29 Hck-specific Polyclonal Antibody-Since Hck consists of three tyrosine phosphorylation sites (i.e. Tyr-29, Tyr-388, and Tyr-499), an antibody, which specifically cross-reacts with Tyr(P)-29, would be an effective reagent for monitoring the phosphorylation status of Hck. This antibody termed anti-pY-29 Hck antibody, was produced by immunizing rabbits with a phosphopeptide derived from the region of murine Hck encompassing Tyr(P)-29. Following affinity purification, the resultant antibody was characterized (Fig. 10). A time course of autophosphorylation of Hck was carried out, with aliquots at various time points taken for immunoblot analysis. These samples were analyzed in conjunction with samples of two other Src family kinases, Lyn and c-Src which had been either C-terminally phosphorylated by CSK or autophosphorylated. Autoradiography of the phosphorylated kinases after SDS-PAGE revealed increasing Hck phosphorylation with the time course and strong levels of phosphorylation of both C-terminal tail phosphorylated and autophosphorylated Lyn and Src (Fig. 10A). A protein contaminant of approximately 38 kDa present in some of our Hck preparations is also seen to be strongly phosphorylated. Analysis of a duplicate blot using the anti-pY-29 Hck antibody shows increasing immunoreactivity with Hck as the time course of autophosphorylation continues. No immunoreactivity is seen with the zero time point of the Hck autophosphorylation time course or with the C-terminally phosphorylated or autophosphorylated Lyn and c-Src (Fig. 10B). This shows that the Tyr(P)-29 specific antibody reacts primarily with autophosphorylated Hck and exhibits no cross-reactivity with two other members of the Src family, be they either C-terminally phosphorylated or autophosphorylated. Since the amino acid sequences around the consensus Src family kinase autophospho- To show further that the anti-pY-29 Hck antibody can specifically recognize the Tyr-29-phosphorylated form of Hck and that it reacts equally well with Hck expressed in mammalian cells, [Y29F]Hck and [Y499F]Hck were expressed in HEK 293T cells. The constitutively active [Y499F]Hck was used to ensure autophosphorylation was stimulated (Fig. 11). The cells expressing [Y499F]Hck, [Y29F]Hck, or carrying a vector only were lysed, and Hck was initially immunoprecipitated with a rat monoclonal Hck antibody. This immunoprecipitate was eluted and separated by SDS-PAGE prior to transfer to a membrane and sequentially probed with anti-phosphotyrosine, anti-pY-29 Hck, and anti-Hck antibodies (Fig. 11) Anti-phosphotyrosine is seen to react strongly with both [Y499F]Hck and [Y29F]Hck immunoprecipitates, whereas the anti-pY-29 Hck antibody reacted strongly only with the [Y499F]Hck immunoprecipitate. The anti-Hck blot reacted with immunoprecipitates from both mutants. Bands corresponding to the heavy chain immunoglobulin are evident in the anti-phosphotyrosine and anti-pY-29 Hck blots. Similar procedures were used to examine the phosphorylation status of endogenous Hck in the RAW264.7 murine macrophage cell line, and we failed to detect any significant Tyr-29 phosphorylation (data not shown). Our inability to detect Tyr-29 phosphorylation is likely to be the result of the relatively poor sensitivity of the anti-pY-29 Hck antibody and the very low abundance of active Hck which is Hck at Low Concentrations Can Autophosphorylate at Tyr-29 -The in vitro and in vivo data presented so far unequivocally demonstrate that Hck can autophosphorylate at Tyr-29. However, the in vitro phosphorylation of Hck was carried out at relatively high concentrations (ϳ100 -400 nM) of Hck. Furthermore, in vivo phosphorylation of Tyr-29 was detected only when Hck was overexpressed in HEK 293T cells, and the levels of expression of recombinant Hck and its mutants in HEK 293T cells were 10 -70-fold higher than those of endogenous Hck in macrophages (see under "Experimental Procedures" for details of Hck concentration estimation). For these reasons, there is a possibility that the phosphorylation of Tyr-29 is a "forced" autophosphorylation event that only occurs at high Hck concentrations. To rule out this possibility, we examine if Hck at lower concentrations (e.g. low nM to pM concentrations) can undergo autophosphorylation at Tyr-29. As shown in Fig. 12, Tyr-29 autophosphorylation at Hck concentrations ranging from 0.46 to 3.7 nM was readily detected by the anti-pY-29 Hck antibody, indicating that autophosphorylation at Tyr-29 can occur at both high and low concentrations of Hck in vitro.
It is generally believed that Src family kinases are recruited to specific subcellular localizations through binding of the unique, SH2, and SH3 domains to cell surface receptors and other cellular proteins. Conceivably, recruitment of the kinases to specific subcellular compartments can significantly increase their effective concentrations and hence enhance their autophosphorylation. However, we do not know the effective concentrations of Src family kinases undergoing autophosphorylation in the specific subcellular compartments. It is therefore not known if the concentration ranges of Hck we chose to carry out the experiments shown in Figs. 1, 6, 7, and 12 reflect the actual concentrations at which Hck undergoes autophosphorylation in specific subcellular compartments in vivo. The result presented in Fig. 12, however, does suggest that Tyr-29 phosphorylation is not a forced autophosphorylation event.
Hck-(1-91) Is a Poor Substrate for Phosphorylation by Fulllength, Recombinant Hck-Although Fig. 7 demonstrates that autophosphorylation of Tyr-29 follows an intermolecular mechanism, it is not clear if structural integrity of Hck is necessary for the efficient autophosphorylation of Tyr-29. To ascertain the structural requirement for efficient autophosphorylation of Tyr-29, the relative efficiencies of Tyr-29 phosphorylation in intact Hck and in the Hck-(1-91) fragment by intact Hck were compared.
As evident from its significant reactivity with the anti-pY-29-specific antibody (Fig. 13), wild type Hck efficiently autophosphorylates at Tyr-29. In contrast, at least a 40-fold excess of Hck-(1-91) was required before low levels of Tyr-29 phosphorylation was apparent. This illustrates that structural requirements necessary for the efficient phosphorylation of Tyr-29 are not met in the isolated unique domain of murine Hck. Regions in addition to the unique domain appear to be required for efficient phosphorylation to occur at this site. DISCUSSION The recent study by Moarefi et al. (7), showing activation of Hck by SH3 domain displacement, used a truncated form of Hck without the fatty acid acylation domain and unique domain. As such, they would only have noticed phosphorylation at Tyr-388, and phosphorylation at Tyr-29 of the unique domain would have gone unnoticed. The use of intact recombinant Hck and Hck isolated from macrophages enabled us to demonstrate autophosphorylation of Hck at both Tyr-29 and Tyr-388. More importantly, our demonstration of the in vivo phosphorylation of Hck at Tyr-29 indicates that phosphorylation of this novel site is of physiological significance. Furthermore, evidence supporting the notion that phosphorylation at Tyr-29 may contribute to the regulation of catalytic activity is provided in Fig. 6. Since activation of protein kinases by auto- Tyr-29 could also likely contribute to the multiple immunoreactive bands. As phosphorylation at more than one site can generate a higher degree of gel mobility shift (32), the slowest migrating band marked by an asterisk is probably the 59-kDa isoform of [Y499F]Hck phosphorylated at both Tyr-388 and Tyr-29. B, anti-pY-29 Hck immunoblot of the Hck mutants after immunoprecipitation. The multiple immunoreactive bands of [Y499F]Hck indicates phosphorylation of both isoforms at Tyr-29. Whereas some [Y499]Hck molecules were autophosphorylated either at Tyr-388 or at Tyr-29, some molecules were expected to be autophosphorylated at both Tyr-388 and Tyr-29. Phosphorylation at both sites could give rise to a higher degree of gel mobility shift and hence additional immunoreactive bands. C, anti-Hck immunoblot of the whole cell lysate. Again, the two 56-and 59-kDa isoforms of the Hck mutants and the mobility shift resulting from phosphorylation of the two isoforms contributed to the multiple immunoreactive bands. FIG. 12. Tyr-29 phosphorylation can occur at low Hck concentrations. Wild type Hck at concentrations ranging from 0.46 to 3.7 nM was allowed to autophosphorylate by incubation at 30°C with (ϩATP) and without (ϪATP) 100 M ATP in the Kinase Assay Buffer in 80 l for 30 min. After SDS-PAGE and electrotransfer of the sample to a nitrocellulose, the blot was probed with the anti-pY-29 Hck antibody. The immunoreactive bands show the extent of Tyr-29 autophosphorylation. phosphorylation normally results from phosphorylation of a conserved tyrosine and/or a threonine residue in the activation loop or the kinase domain, activation of Hck by phosphorylation of Tyr-29 in the unique domain may represent a new and novel mechanism of activation of protein kinases (27). Elucidation of the structural basis of activation of Hck by Tyr-29 phosphorylation may reveal this new activation mechanism.
The modulation of catalytic activity of Hck by phosphorylation at Tyr-29 shows that there must be some cross-talk between the unique and catalytic domains. As the recent crystal structure determination of Hck also used the truncated mutant excluding the unique domain, it is not known how phosphorylation at Tyr-29 may impart conformational change such that catalytic activity may be enhanced. From the known structural data of Hck, it is likely that the unique domain makes contact with the SH3 and/or the catalytic domain. Therefore, it is possible that phosphorylation at Tyr-29 may produce structural changes in the unique domain which is then transmitted to either or both of these domains. As Moarefi et al. (7) have shown, displacement of the SH3 domain from the SH2 kinase domain linker serves to activate the kinase, which is likely to occur through providing for rotation of the ␣-C helix such that ion pairing occurs between Glu-303 and Lys-288. Consequently, ATP binding is made feasible which in turn allows for autophosphorylation to occur. It is possible that phosphorylation at Tyr-29 may impart conformational change to the SH3 domain that will further destabilize the tripartite interactions among the SH3 domain, the SH2-CD linker, and the catalytic domain. As the enzyme can be visualized as being in a fluid state of conformational equilibrium, such a change may act to move the equilibrium more toward the active conformation.
Phosphorylation at Tyr-29 of the isolated recombinant unique domain of Hck by full-length Hck is shown to occur with low efficiency (Fig. 13). It is likely that the unique domain must be situated in the context of the intact enzyme to provide for the structural requirements for efficient phosphorylation of Tyr-29. It may be that the intermolecular interaction between two molecules allowing for the autophosphorylation of Tyr-388 places the unique domain in the correct structural alignment for the efficient phosphorylation of Tyr-29. Activation of the enzyme by autophosphorylation at the consensus site would further increase the propensity for phosphorylation at Tyr-29. Alternatively, interactions made with the unique domain by adjacent domains such as the SH3 and catalytic domains may provide for enhanced affinity of Tyr-29 for the active site of Hck.
Interestingly, upon sequence comparison between different Src family kinases, the YXXDPT motif corresponding to 29 YVP-DPT in murine Hck was found to occur also in Lyn ( 32 YVRDPT), Fgr ( 32 YYPDPT), Fyn ( 29 YGTDPT), and Yrk ( 29 YDPDPT). As this motif is repeated several times in the Src family of kinases and occurs at either residue number 29 or 32 in each case, it may be of general regulatory significance for the members in which it occurs.
Similar to most tyrosine phosphorylation sites in protein tyrosine kinases, the phospho-Tyr-29 may also act as a docking site for specific cellular proteins containing SH2 and/or protein tyrosine-binding (PTB) domains. Comparison of the sequence around phospho-Tyr-29 with those of known SH2 domain and PTB domain binding motifs does not reveal any significant homology. Some degree of homology can be found between the GPVpY 29 sequence and the consensus NPXpY motif recognized by the Shc PTB domain (28). Structural analysis of the Shc PTB domain bound to the ligand reveals that the conserved Pro residue is essential for the formation of the ␤-turn recognized by the Shc PTB domain. As the GPVpY 29 motif of Hck is predicted to have a propensity to form a ␤-turn, it is possible that this motif is the docking site of an as yet to be identified PTB domain. Elucidation of the cellular functions of the phosphorylation of Tyr-29 in addition to modulation of activity awaits the isolation and identification of the putative cellular protein(s) which selectively bind to the sequence around Tyr-29.