A switch in nucleotide affinity governs activation of the Src and Tec family kinases

von Raußendorf, Freia; de Ruiter, Anita; Leonard, Thomas A.

doi:10.1038/s41598-017-17703-5

Download PDF

Article
Open access
Published: 12 December 2017

A switch in nucleotide affinity governs activation of the Src and Tec family kinases

Scientific Reports volume 7, Article number: 17405 (2017) Cite this article

2328 Accesses
15 Citations
Metrics details

Subjects

Abstract

The Tec kinases, closely related to Src family kinases, are essential for lymphocyte function in the adaptive immune system. Whilst the Src and Abl kinases are regulated by tail phosphorylation and N-terminal myristoylation respectively, the Tec kinases are notable for the absence of either regulatory element. We have found that the inactive conformations of the Tec kinase Itk and Src preferentially bind ADP over ATP, stabilising both proteins. We demonstrate that Itk adopts the same conformation as Src and that the autoinhibited conformation of Src is independent of its C-terminal tail. Allosteric activation of both Itk and Src depends critically on the disruption of a conserved hydrophobic stack that accompanies regulatory domain displacement. We show that a conformational switch permits the exchange of ADP for ATP, leading to efficient autophosphorylation and full activation. In summary, we propose a universal mechanism for the activation and autoinhibition of the Src and Tec kinases.

Slow phosphorylation of a tyrosine residue in LAT optimizes T cell ligand discrimination

Article 14 October 2019

Phosphorylation of CrkL S114 induced by common gamma chain cytokines and T-cell receptor signal transduction

Article Open access 20 August 2021

THEMIS is a substrate and allosteric activator of SHP1, playing dual roles during T cell development

Article 04 January 2024

Introduction

The Tec family kinases Btk and Itk are essential for B- and T-lymphocyte development in the adaptive immune system. Engagement of the B- or T-cell receptor triggers a cascade of intracellular phosphorylation events, initiated by phosphorylation of ITAM motifs in the cytoplasmic tails of activated receptors by Src family kinases^1,2,3. The phosphorylated ITAMs recruit the tandem SH2 domain tyrosine kinases Syk and ZAP-70 in B- and T-cells respectively, which leads to activation of Tec family tyrosine kinases upon their PI3K-dependent recruitment⁴. Activated Btk and Itk phosphorylate phospholipase C-γ^5,6, which triggers calcium flux in the cell by producing the second messenger inositol-1,4,5-trisphosphate (IP3). Dysregulation of Btk in humans causes the heritable disorder X-linked agammaglobulinemia⁷, characterised by a lack of circulating antibodies and a corresponding susceptibility to opportunistic infection. Itk dysfunction, on the other hand, specifically affects the development of T-cells and has been implicated in a number of lymphoproliferative disorders^8,9,10, as well as playing a role in HIV replication and infectivity^11,12.

Whilst the Tec and Src kinases are often referred to as separate families, they are in fact very closely related, belonging to the same branch of the tyrosine kinase phylogenetic tree¹³ and sharing in common a conserved SH3-SH2-kinase domain module. Src kinases are localised to the plasma membrane via lipid modification of their N-termini, whereas Tec kinases are regulated by the lipid second messenger PIP₃, responding specifically via their N-terminal PH domain¹⁴. Src and Tec kinases are activated by engagement of intracellular ligands for their SH2 and SH3 domains, events that promote activation loop phosphorylation and downstream signalling^15,16,17. Deletion of the PH domain or mutation of the phosphotyrosine-binding SH2 domain of Itk results in a loss of Itk activation^16,18.

The structures of Src, Hck, and Abl kinases show a conserved intramolecular assembly of their regulatory SH3, SH2, and kinase domains that maintains the kinase in an autoinhibited conformation^19,20,21. While Src and Hck are proposed to be maintained in this conformation by intramolecular engagement of their phosphorylated C-terminal tail^20,21 and N-terminal myristoylation of Abl stabilises a similar conformation¹⁹, the Tec kinases notably lack either feature, leading to proposals that the PH domain might fulfil this function^22,23. However, the recent crystal structure of the SH3-SH2-kinase domains of Btk shows a similar domain arrangement to Src in the absence of both the PH domain and a C-terminal tail²³. The interface between the regulatory and kinase domains is mediated by a variable linker between the SH2 and kinase domains that forms a network of interactions essential for maintaining the inactive state. Displacement of the SH3 domain of Hck activates the kinase by relieving these inhibitory constraints^15,17 and promoting a conformational switch in the N-terminal end of the catalytic domain²⁴.

Activation loop autophosphorylation regulates the activity of Src tyrosine kinases^17,25,26. In the Tec kinases, activation loop phosphorylation in cells depends on the activity of upstream Src kinases^27,28,29,30, presumably via recruitment and activation of the kinase by SH2 and SH3 domain-ligands. Like the Src kinases, the Tec kinases robustly autophosphorylate in vitro ^16,23,31. However, while the Src kinases have been extensively studied, the Tec kinases have been suggested to be mechanistically distinct, despite their evolutionary similarity.

We set out to determine the structure and mechanism of activation of the T-cell kinase Itk. To our surprise, we discovered that both Itk and Src bind ADP significantly stronger than the non-hydrolysable ATP analogue, AMPPNP. We found that activation loop phosphorylation or C-terminal tail deletion had no effect on the nucleotide binding preference or autoinhibited conformation of Src in solution. We describe the crystal structure of autoinhibited Src bound to ADP and use molecular dynamics to rationalise the binding preference for ADP. We show that Itk adopts a stable, globular conformation in the absence of its PH domain that is well described by the autoinhibited state of Src. We demonstrate the allosteric activation of both Itk and Src by mutation of a conserved hydrophobic stack between the regulatory and kinase domains, which converts the kinase from an ADP- to an ATP-binding conformation that undergoes robust auto-phosphorylation. In summary, we present a general mechanism for the allosteric activation and autoinhibition of both Src and Tec family tyrosine kinases.

Results

Autoinhibited Itk and Src kinases preferentially bind ADP

Sequence and structural homology of Itk to Src (Fig. 1a) indicated that its biochemical properties, structure, and mechanism of activation were likely to also be similar. The first crystal structures of Src and Hck in their autoinhibited conformations suggested that nucleotide occupancy of the active site promoted ordering of the activation loop in its inactive conformation^20,21. Subsequent structures of Src and Btk under different crystallisation conditions, however, have revealed an ordered conformation of the activation loop independent of nucleotide and Mg²⁺ binding^23,32.

To characterise the nucleotide binding of Itk and Src, we purified recombinant wild type and mutant constructs of Itk and Src containing their respective SH3, SH2, and kinase domains (from here on referred to as Itk^32K, Src^32K, and Src^32KΔC, Fig. 1a). We confirmed the mass and homogeneity of isolated species by high-resolution anion exchange chromatography, intact mass spectrometry, and Western blotting with phospho-specific antibodies (Supplementary Figs 1–3). We performed fluorescence anisotropy measurements using Mant-labelled nucleotides. Structural modelling of Mant-ADP bound to Src did not reveal any steric clashes or additional contacts with the Mant moiety (Supplementary Fig. 4a,b). To avoid autophosphorylation during the course of the titration, we used the non-hydrolysable ATP analogue AMPPNP. Itk^32K binds ADP with a K _d of 23.7 ± 1.7 μM, while AMPPNP was estimated to bind with 15-fold lower affinity (Supplementary Fig. 4c) (in the absence of saturating data for AMPPNP, the upper asymptote of the curve was set to the same value as for ADP based on the maximum anisotropy achievable being the same in each case). Src^32K bound ADP approximately 6-fold more tightly than AMPPNP (Supplementary Fig. 4d).

Since the concentration of ATP in the cell is typically higher than ADP^33,34, we performed nucleotide competition assays in order to determine which nucleotide is bound in vivo. We measured the concentrations of ADP or AMPPNP required to displace Mant-ADP. The equilibrium inhibition constant for displacement of ADP by AMPPNP in the case of Itk^32K is 7.90 mM, approximately a 41-fold excess over the concentration of ADP required (Fig. 1b). Src^32K exhibits a similar 19-fold preference for ADP over AMPPNP while activation loop phosphorylated Src^32K exhibits identical behaviour, indicating that activation loop phosphorylation alone does not change the binding preference of Src for ADP (Fig. 1c).

The finding that Itk and Src preferentially bind ADP is qualitatively supported by thermal stability measurements showing that ADP stabilises Itk^32K and Src^32K to a greater extent at the same nucleotide concentrations (Supplementary Fig. 4e,f), irrespective of activation loop phosphorylation (Supplementary Fig. 4 g). We noted that the thermal stability of Src^32K initially decreases at low concentrations of ATP, before rising again at higher concentrations. This was reproducible and correlates with autophosphorylation of Tyr416 during the course of the assay (Supplementary Fig. 4h).

To investigate the preferred binding of ADP at atomic resolution we determined the crystal structure of tail-phosphorylated Src in the presence of ADP and magnesium. Src^32K crystallised in the same conditions as previously reported for the nucleotide-free protein and in the same space group (Supplementary Fig. 5a)²¹. The overall structure agrees closely with that of a previous Src structure in complex with AMPPNP (2SRC), in which the activation loop is fully ordered³² (Supplementary Fig. 5b). A difference density map of the nucleotide-binding site shows clear electron density for the adenosine, both alpha and beta phosphates, and a single magnesium ion (Fig. 1d). The magnesium ion is coordinated between the alpha and beta phosphates and Asn391 and Asp404. The beta phosphate is directly coordinated by Arg388, the guanidinium group of which is stabilised by a hydrogen bond network between Asn391, Asp386 and Arg388 (Supplementary Fig. 5c,d). For comparison, the coordination of AMPPNP in the active conformation of insulin receptor tyrosine kinase (IRK) is shown in Supplementary Fig. 5e,f.

While the position of the alpha phosphate is superimposable with that of AMPPNP and the protein has, in principle, a large enough pocket in which to accommodate an additional phosphate, our binding experiments indicate a strong preference for ADP over AMPPNP (Fig. 1b,c). In order to rationalise the energetic basis for ADP over ATP binding, we computed the relative binding free energies between ADP and ATP using molecular dynamics simulations^35,36,37. Figure 1e shows a schematic overview of the thermodynamic cycle used in this approach (see also Supplemental Text). The perturbation of ADP to ATP in the bound state resulted in a free energy difference of -262.9 kJ mol⁻¹ compared with −281.8 kJ mol⁻¹ in the free state (Fig. 1e). This results in a relative binding free energy for ADP over ATP of −18.9 kJ mol⁻¹, confirming our experimental findings.

Itk adopts a Src-like autoinhibited conformation

C-terminal tail phosphorylation of the Src kinases Src, Hck, and Lck has previously been proposed to stabilise and maintain the inactive, autoinhibited conformation of the kinase^20,21,32,38, while the PH domain of Btk has been proposed to stabilise its inactive conformation^22,23. Having observed that Itk^32K binds and is stabilised by ADP in the absence of a C-terminal tail, we investigated whether the C-terminal tail of Src influences its nucleotide binding or conformation in vitro, using a construct of Src lacking its C-tail (Src^32KΔC) and a SH2 domain mutant of Src that is incapable of binding phosphotyrosine peptides (Src^32K R175L, Supplementary Fig. 3a,b)³⁹. Src^32KΔC binds ADP with approximately 23-fold higher affinity than AMPPNP (Fig. 2a), indicating that tail dephosphorylation does not affect the preference of Src for ADP over ATP (Src^32K exhibits 19-fold stronger binding to ADP, Fig. 1c). Thermal stability measurements support the stabilisation of Src^32KΔC by ADP (Supplementary Fig. 6a).

To address whether Itk^32K, Src^32K, Src^32KΔC, and Src^32K R175L adopt the same intramolecular conformation, we collected small-angle X-ray scattering (SAXS) data on all four proteins in solution in the presence of ADP and magnesium (Fig. 2b). The scattering curves and pair distribution functions of Itk^32K, Src^32K, Src^32KΔC, and Src^32K R175L are superimposable (Fig. 2b,c). The structural parameters of the four particles are summarised in Fig. 2d and Supplementary Fig. 6b-e. Fitting of the theoretical scattering curve for autoinhibited Src (PDB entry: 2SRC)³² to the experimental scattering of Src^32K confirms that Src adopts an identical conformation in solution (χ² = 0.22, Supplementary Fig. 6f). Iterative ab initio calculation of the molecular envelopes from the pair-distribution functions converged on a core electron density distribution in each case, which is well fitted by the crystal structure of Src. Superposition of the molecular envelopes (Fig. 2e,f, Supplementary Fig. 6f) indicates that these proteins adopt a very similar shape and conformation in solution and this is confirmed by the close agreement between the experimental scattering curves (Itk^32K χ² = 2.39, Src^32KΔC χ² = 2.36, Src^32K R175L χ² = 1.31; 0.16 ≤ q ≤ 4 nm⁻¹).

Having observed that activation loop phosphorylation did not influence the nucleotide binding of Src^32K (Fig. 1c), we next addressed whether it altered the overall conformation of Src^32K. The experimental scattering curve and pair distribution function indicate that, at the resolution of the SAXS data (nominally 18 Å), this protein adopts the same conformation as Src^32K in solution (Supplementary Fig. 6g-i). Together, these findings indicate that Itk^32K adopts a Src-like autoinhibited conformation in solution and that the C-terminal tail is dispensable for assembly of the closed conformation of Src.

An allosteric switch triggers nucleotide exchange

The inactive conformation of Src is stabilised by a conserved hydrophobic stack on the distal surface of the N-lobe of the kinase domain in which a hydrophobic side chain is contributed by the linker between the regulatory domains and the kinase domain^40,41. Sequence analysis shows that this residue is Leu, Trp, or Phe in all Src, Abl, Tec, Csk, and Fer family kinases, which derive from the same branch of the human kinome¹³ (Fig. 3a, Supplementary Fig. 7a). The inactive conformation is stabilised by the insertion of the side chain of the invariant tryptophan (W260) into a hydrophobic pocket formed between the αC helix and strand β4 of the kinase N-lobe (Supplementary Fig. 7b).

By contrast, active Lck, lacking its regulatory domains and SH2-kinase inter-domain linker, exhibits a collapsed hydrophobic stack (Supplementary Fig. 7c). The conformational change is propagated to an inward rotation of the αC helix and formation of the salt bridge between Glu288 and Lys273 that characterises the active state of protein kinases⁴². The side chain of the invariant tryptophan (W238) is reoriented to form a hydrogen bond between the amide nitrogen of its indole ring and the main chain carbonyl of Leu303, while Lys293 in the αC helix maintains a stacking interaction with Trp238 by reorienting its side chain; together, these interactions stabilise the active conformation of the αC helix.

To investigate whether disruption of the hydrophobic stack could trigger an allosteric switch from the inactive to active conformation, we mutated the central hydrophobic amino acid in the stack contributed by the SH2-kinase linker (Itk Leu351, Src Leu255) (Figs 1a, 3a). The mutant proteins and their phosphorylated species were validated by high-resolution anion-exchange chromatography, immunoblotting, and mass spectrometry (Supplementary Fig. 1c-e, Supplementary Fig. 3c-e). Due to hyperphosphorylation of purified Itk^32K L351A, unphosphorylated protein was produced by in vitro dephosphorylation and validated by mass spectrometry (Supplementary Fig. 1f).

Mutation of Leu351 to alanine causes a reduction in compactness of Itk^32K as judged by a 10% increase in the radius of gyration and the maximum dimension of the particle (Fig. 3b), consistent with displacement of the inhibitory SH3-kinase interaction. Guinier analysis of the low angle portion of the scattering curve shows the clear increase in R_g between wild type Itk^32K and Itk^32K L351A (Supplementary Fig. 6c, Supplementary Fig. 7d). To investigate the consequences of the L351A mutation in more detail, we calculated the ab initio molecular envelope for Itk^32K L351A and compared it to the crystal structure of autoinhibited Src (ΔC-tail). We could not fit the data with the structure of Src (χ² = 35.29), so we turned to rigid body modelling of the individual domains to try to better fit the scattering curve. Allowing only the SH3 domain to move did not result in an acceptable fit, but by allowing both the SH2 and SH3 domains to move freely, we were able to fit the scattering with a reasonable chi-square (Supplementary Fig. 7e). The best fit to the scattering shows a significant rotation and translation of both domains with respect to the kinase domain (Fig. 3c). Our results suggest that the L351A mutation destabilises the inhibitory interactions of both the SH3 and SH2 domains with the kinase domain, but does not lead to a wholesale displacement of the regulatory domains from the kinase domains.

To establish whether hydrophobic stack disassembly triggers a switch in nucleotide binding of the kinase domain, we measured the binding of both unphosphorylated and activation loop phosphorylated (pY511) Itk^32K L351A to ADP and ATP. While the affinity for ADP remained unchanged, we observed that Itk^32K L351A now binds ATP with comparable affinity to ADP (Fig. 3d). Activation loop phosphorylation did not significantly change the affinity for ATP, but slightly lowered the affinity for ADP (Supplementary Fig. 7f), perhaps by relieving the coordination of the beta phosphate by Arg388 in the C-lobe.

To test whether Src exhibits the same switch in nucleotide binding affinity as Itk, we tested the equivalent mutation in Src lacking its C-terminal tail (Src^32KΔC L255A). Due to limiting protein concentrations, we were unable to obtain a binding curve by fluorescence anisotropy, so we instead used thermal stability measurements to investigate its nucleotide-binding properties. Determination of accurate binding constants from thermal stability measurements requires knowledge of the enthalpy of unfolding. However, for a given protein, the enthalpy of unfolding is independent of the ligand being tested, so comparison of the concentrations of nucleotide required for half maximal stabilisation is valid⁴³. In contrast to Src^32K, Src^32KΔC L255A binds ATP approximately three times better than ADP and is stabilised to a greater extent by ATP than ADP (Fig. 3e). The affinities for ADP and ATP that we obtain by fluorescence anisotropy (Itk) or by thermal stability measurements (Src) are comparable to those obtained for the isolated Src kinase domain by isothermal titration calorimetry⁴⁴. Taken together, our results demonstrate that disruption of the hydrophobic stack in Itk and Src triggers a conformational switch in the kinase domain that favours ATP binding.

Nucleotide exchange permits trans-autophosphorylation of Itk and Src

Having observed that hydrophobic stack disassembly promotes ATP-binding in both Itk and Src, we determined whether this also promotes autophosphorylation and activation against an exogenous substrate. Whilst wild type Itk^32K is a very poor kinase against Tyr511, Itk^32K L351A autophosphorylates very efficiently (Fig. 4a). To demonstrate that Src exhibits the same behaviour, we evaluated Tyr416 autophosphorylation of Src^32K, Src^32KΔC, and Src^32KΔC L255A in vitro. Src^32KΔC, lacking its C-tail, exhibits only modestly higher autophosphorylation (1.8-fold). However, disruption of the hydrophobic stack in Src results in an 11-fold activation of the kinase and very efficient autophosphorylation (Fig. 4b). Notably, the slow initial rate of autophosphorylation is dramatically accelerated in both Itk^32K L351A and Src^32KΔC L255A, consistent with a lowering of the activation energy barrier to the conformational change between inactive and active states. While Itk^32K is a very poor kinase against a peptide derived from Cdk1, Itk^32K L351A readily phosphorylates it (Fig. 4c). The kinetics can be fitted with an exponential function in which the lag phase describes the activation of the kinase by autophosphorylation. Similarly, Src^32K is a relatively inefficient kinase, and while removal of its C-tail is modestly activating, mutation of the hydrophobic stack converts it into a very efficient enzyme (Fig. 4d).

Discussion

Previous crystal structures of Src and Hck in complex with AMPPNP both resolved the adenosine, alpha and beta phosphates, and a single magnesium ion in the electron density, but observed poorly defined density and an uncoordinated, solvent accessible conformation for the terminal gamma phosphate, respectively. In solution, however, both Itk and Src bind to ADP with approximately an order of magnitude greater affinity. Whilst the crystal structure of Src in complex with ADP shows minimal differences to AMPPNP-bound Src, molecular dynamics calculations of the free energy of binding reinforce the strong preference for ADP.

Cellular concentrations of ADP have been measured at between 100 μM and 1.3 mM in both bacterial⁴⁵ and mammalian cells^46,47, while a carefully controlled mass spectrometry study found that the ratio of ATP to ADP in rat kidney tissue varied between 1.69 and 2.65³⁴. In E. coli, the ratio of ATP to ADP in glucose-fed, exponentially growing cells is approximately 17:1⁴⁵, which likely places an upper limit on the ratio. Given that a 9–23 fold molar excess of ATP is required to displace 50% of Mant-ADP from Src/Itk in vitro and independent measures of the cellular ATP/ADP ratio vary between 1.7 and 17^33,45,46,47, the majority of inactive Src or Itk is likely to be ADP-bound in the cell under all metabolic conditions.

Interestingly, Src lacking its C-terminal tail retains its binding preference for ADP, while solution X-ray scattering measurements indicate that Src^ΔC and an SH2 domain mutant of Src both adopt the same conformation as Src and Itk, suggesting that the role of the C-tail in Src is not to lock the closed conformation. A previous SAXS and molecular dynamics study of Hck found that the solution scattering of dephosphorylated Hck could be modelled by more than 85% of Hck molecules in the autoinhibited conformation⁴⁸. A separate molecular dynamics study of Src found that dephosphorylation of the C-terminal tail promoted higher flexibility in the SH2 and SH3 domains⁴⁹ but, significantly, did not reveal any domain displacements. Rather, this study suggests that the energy barrier to Src activation is lowered by tail dephosphorylation, consistent with the very modest effect of tail deletion on in vitro activity. These observations are also consistent with an enzymological study which found that tail phosphorylation of Src does not affect the K _m for either ATP or substrate peptide⁵⁰ and a study on Lck that found that tail phosphorylation did not contribute to inhibition of activation loop-phosphorylated Lck⁵¹. Although a structural study found that Src adopts an open conformation in the absence of tail phosphorylation⁵², this conclusion was based on crystallisation of Src in the presence of the inhibitor imatinib, which promotes a conformation of the kinase domain resembling that of active Lck (r.m.s.d. of 1.2 Å over all atoms)³⁸.

In the Abl kinases, which lack a C-terminal tail, docking of the myristoylated N-terminus to a hydrophobic pocket on the C-lobe of the kinase domain was proposed to stabilise the autoinhibited conformation⁵³, though a more recent SAXS study showed that a point mutant in the pocket that prevents myristoyl binding does not lead to disassembly of the regulatory domains⁵⁴. Furthermore, recent investigations of ancestral Src kinase in unicellular choanoflagellates and filastereans have revealed that, while the C-terminal tail is present, kinase activity is not suppressed by Csk^55,56, suggesting that phosphorylation of the C-terminal tail was a later evolutionary adaptation.

All of these observations are contrasted by cellular studies demonstrating that deletion of the tail or mutation of its phosphorylation site leads to hyperactivation of Src and cellular transformation^{57,58,59,60,61}. Whilst the C-terminal tail of Src is of undoubted importance in the cell, we find that it has no effect on Src conformation and only a very modest effect on Src activity in vitro. Indeed, one can dispense with either the C-tail or the capacity to bind it, and Src still adopts the autoinhibited conformation. Therefore, although v-Src is transforming in cells, it is unlikely due to the intrinsic disassembly of its autoinhibited conformation. Indeed, the SH2 domain mutant of c-Src incapable of engaging its C-tail is not transforming^62,63. Likewise, the Tec kinases Btk and Itk require a functional SH2 domain for their activation, despite lacking a C-tail^16,64. Notably, while constitutively membrane localised Btk has transforming activity, mutation of its SH2 domain abrogates this activity⁶⁴.

A distinguishing feature of the Src kinases is the presence of a membrane anchor, in contrast to the Tec kinases, which are transiently recruited upon PIP₃ production. Consequently, Src kinases are, on average, exposed to a higher density of SH2 and SH3 domain ligands. In vitro, the phosphorylated C-terminal tail of Src inhibits engagement of the SH2 domain by exogenous ligands supplied in trans by a factor of ~200⁵⁰. We therefore favour a mechanism in which the phosphorylated C-terminal tail acts as a competitive inhibitor of SH2 domain engagement with activating ligands in the cell, rather than a physical latch that maintains the closed conformation.

Another widely used proxy for Src activation is activation loop phosphorylation. We show, however, that activation loop phosphorylation of Src alone does not change its overall conformation or binding preference for ADP, indicating that the regulatory and kinase domains are still assembled in the autoinhibited conformation. Whilst the electrostatic network stabilising the inactive conformation of the activation loop is undoubtedly important⁶⁵, its disruption does not appear to lead to a conformational switch in either the regulatory domains or the nucleotide binding site of the kinase domain, an observation consistent with a recent molecular dynamics study⁶⁶.

The invariant tryptophan in the N-lobe of the kinase domain has previously been the focus of efforts to understand the allosteric activation of these kinases. While mutation of Trp260 in Src and Hck results in their hyperactivation^40,67, mutation of the equivalent residue in Btk (Trp395) and Itk (Trp356) renders them inactive^68,69,70. Whilst these apparently opposite observations have been used to argue that these kinases are mechanistically distinct, Trp395 of Btk is observed in the same, active conformation as Trp260 of Src and Trp238 of Lck, supporting the notion that they employ similar mechanisms to activate their respective kinase domains⁷¹. Recent molecular dynamics studies have also demonstrated the importance of the invariant tryptophan in stabilising the active conformation of both Src and Tec kinases^69,72. The invariant tryptophan therefore fulfils a dual role in stabilising both the inactive and active conformations.

In order to avoid compromising the catalytic machinery in Src and Itk, we mutated a hydrophobic residue in the inter-domain linker that acts as a ‘molecular glue’ between the kinase and regulatory domains. Critically, this residue inserts its side chain into a hydrophobic stack on the back of the N-lobe of the kinase domain, which stabilises the inactive conformation of the kinase^24,40,73. Mutation of this residue has previously been shown to abrogate Csk mediated inhibition of Src in a yeast cell-based assay⁴¹. We observed a dramatic switch in affinity for ATP, which is accompanied by efficient activation loop autophosphorylation and attainment of full catalytic activity. That Src and Itk bind ATP in their active conformations is not surprising, but the selective binding of ADP by the inactive kinases has not been previously observed. We speculate that this may serve to prevent promiscuous kinase activity and restrict kinase activation exclusively to situations in which the regulatory domains, and specifically the SH3 domain, have been engaged by intracellular ligands. As such, one could regard the hydrophobic stack as the ultimate regulatory node that controls the activation of Src and Tec family kinases, irrespective of activation loop phosphorylation or C-terminal tail dephosphorylation.

Whilst mutation of the linker led to a dramatic increase in binding affinity for ATP, it resulted in a very modest conformational change in Itk, consistent with regulatory domain displacement, but not complete disassembly of the regulatory and kinase domains. A putative conformation of the mutant Itk shows the SH2 domain to be in close proximity to the N-lobe of the kinase domain, though we cannot be certain of the respective domain arrangement given the resolution limits of the SAXS data. Intriguingly, the SH2 domain of c-Abl was found to enhance kinase activity through an interaction with the kinase N-lobe that stabilises the active conformation of the αC-helix, a mechanism that was also demonstrated for Fes⁷⁴. Whilst these structures are reminiscent of active Csk⁷⁵, the SH2 domain actually adopts distinct positions and makes different contacts with the kinase N-lobe in each structure. It is therefore unclear how and to what extent the SH2 domain may modulate catalytic activity through the N-lobe in the various family members.

Our data are consistent with the following model (Fig. 5): In the cytosol, inactive Src and Tec kinases exist bound to ADP. Activation is driven by regulatory domain displacement by specific interactions with SH2 and SH3 domain ligands. Whilst it is clear from our data that SH3 domain displacement is a prerequisite for Src and Itk activation, it may not necessarily be the case that wholesale displacement of both SH2 and SH3 domains is required in all circumstances. SH3 domain displacement is accompanied by collapse of the hydrophobic stack on the distal surface of the N-lobe of the kinase domain and an inward rotation of the αC helix to its active conformation; repositioning of the αC helix breaks the salt bridge between Glu310 in the helix and Arg409 in the activation loop, expelling the activation loop from the catalytic cleft in the process. A salt bridge between Glu310 in the αC helix and Lys295 in the N-lobe stabilises the active conformation and creates the binding site for ATP and a second magnesium ion. The αC helix is further stabilised by the formation of a hydrogen bond between the indole ring of Trp260 and the main chain carbonyl of Leu325. In the active conformation, Src binds ATP with equal affinity to ADP, but ADP is rapidly exchanged for ATP due to the high ratio of ATP to ADP in the cell. Once loaded with ATP, Src is primed for catalytic activity, but must first immobilise its activation loop to create a surface for substrate docking. Autophosphorylation proceeds rapidly, driving immobilisation of the phosphorylated activation loop on the C-lobe of the kinase. Since ADP release is rapid and non-limiting for phospho-transfer⁷⁶ iterative cycles of substrate phospho-transfer will be ensured as long as the regulatory domains remain engaged with their cognate ligands. While the core catalytic machinery is conserved between Itk and Src, additional regulatory features restrict their activities. In the case of Itk, the N-terminal PH domain restricts activity to PIP₃-containing membranes. In the case of Src, which has an N-terminal membrane anchor, we propose that its C-terminal tail prevents spurious activation in the context of a high density of potentially activating ligands at the membrane.

In summary, we propose a general mechanism for the allosteric regulation of the Src and Tec kinases. We describe the basic unit of kinase control and show that the inactive state is bound and stabilised by ADP. Based on the strong sequence and structural homologies between the Src, Abl, Tec, Csk, and Fer kinases, we predict that the mechanisms of autoinhibition and kinase activation are also conserved. We propose that nature has evolved various sub-family specific adaptations, such as the C-terminal tail of Src or PH domain of Itk, that further refine the control and localisation of enzymatic activity.

Methods

Antibodies

The following antibodies were used in this study: Src (Cell Signaling Technology, 32G6); Src pY416 (Cell Signaling Technology, D49G4); Itk (Cell Signaling Technology, 2F12); Btk pY551 (Itk pY511) (BD Pharmingen clone 24a).

Protein expression and purification

Human Itk^171–620 (Itk^32K), mouse Src^85-535 (Src^32K), and mouse Src^85-523 (Src^32KΔC) as well as their respective point mutants were expressed as GST fusions in baculovirus-infected Sf9 insect cells. For the generation of Tyr527-phosphorylated Src^85-535, the protein was co-expressed with human Csk using pFastBac Dual (Invitrogen). Proteins were purified by sequential glutathione-sepharose affinity chromatography, TEV cleavage, high-resolution anion exchange, and size exclusion chromatography in a final buffer of 50 mM Tris, pH 7.5, 200 mM NaCl, 1 mM TCEP, 1% (v/v) glycerol.

Mass spectrometry

500 ng protein was dissolved in 20 mM Tris 7.6, 50 mM KCl, 1 mM TCEP and loaded on an Eksigent MicroLC column (ChromXP C4, 3 μm particle size, 300 Å pore diameter, dimensions 50 × 0.5 mm) using a Dionex Ultimate 3000 HPLC system (Thermo Scientific). The proteins were separated on a gradient from 5% to 90% acetonitrile in water and 0.1% formic acid for 30 min at a flow rate of 30 µL/min. Mass spectra were recorded on a Sciex TripleTOF 5600 instrument equipped with a Turbo V DuoSpray ion source in positive mode. Data was analysed in PeakView version 2.1 (Sciex) using the Bio Tool Kit to reconstruct the uncharged average protein mass. All other samples were diluted in 0.1% formic acid (FA) to a concentration of 10 ng/μL and 50 ng were loaded on an Aeris Widepore C4 column, 3.6 µm particle size, dimensions 2.1 × 150 mm (Phenomenex), using a Dionex Ultimate 3000 HPLC system (Thermo Scientific) with a working temperature of 55 °C, 0.1% FA as solvent A, 90% acetonitrile, 0.08% FA as solvent B. Proteins were separated in a 6 min gradient from 10 to 70% solvent B at a flow rate of 300 μL/min. Mass spectra were recorded on a Waters Synapt G2-Si equipped with a ZSpray ESI source. Data were analysed in MassLynx V 4.1 using the MaxEnt 1 process to reconstruct the uncharged average protein mass.

Differential scanning fluorimetry

Thermal stabilities as a function of nucleotide concentration were measured by differential scanning fluorimetry (DSF). Samples contained 0.1 mg/ml protein in 50 mM Tris pH 7.5, 150 mM NaCl, 1 mM TCEP, and 5 mM MgCl₂ with ADP, ATP, or AMPPNP concentrations ranging from 0–16 mM. Samples were measured in triplicate using a BioRad iQ™5 Multicolor Real-Time PCR Detection System. For detection of phosphorylated protein, samples were diluted and 1 ng and 15 ng were loaded to SDS-PAGE for Src and Itk proteins, respectively. Phosphorylation was measured by immunoblot.

Fluorescence anisotropy

The binding constants for ADP and AMPPNP were determined for Src and Itk proteins by fluorescence anisotropy, using 2′/3′-Mant-labelled nucleotides (Jena Bioscience). Increasing concentrations of Itk^32K, Itk^L351A or Src^32K were incubated with 40 μM Mant-nucleotide in 30 mM Tris pH 7.5, 200 mM NaCl, 5 mM MgCl₂, and 1 mM TCEP. Measurements were made with a Perkin Elmer LS50B fluorimeter (λ_ex = 355 nm, λ_em = 447 nm) at 20 °C. For each data point 50–60 measurements, each with an integration time of 1 s, were averaged. For competition nucleotide displacement assays, 40–80 μM of protein was incubated with 40 μM Mant-ADP until equilibrium was established. Unlabelled nucleotide was titrated into the reaction and fluorescence anisotropy monitored. The equilibrium inhibition constant for the unlabelled nucleotide was calculated from the fitted curve.

Small-angle X-ray scattering (SAXS)

SAXS data were collected on 0.5–4 mg/ml samples of Itk or Src in 50 mM Tris, pH 7.6, 200 mM KCl, 1 mM TCEP, 1% (v/v) glycerol, 2 mM ADP, 2 mM MgCl₂ on BioSAXS beamline BM29, ESRF. Data reduction and analysis was performed using the BsxCuBE data collection software and the ATSAS package⁷⁷. Scattering data for Itk^32K L351A and Src^32K R175L were collected using an online size exclusion chromatography setup on BM29. The proteins were applied to a Superdex 200 column equilibrated in 20 mM Tris, pH 8.0, 100 mM NaCl, and 1 mM TCEP and images were acquired every second for the duration of the size exclusion run. Buffer subtraction was performed by averaging 50 frames either side of the peak. The program AutoGNOM⁷⁸ was used to generate the pair distribution function (P(r)) for each isoform and to determine D _max and R _g from the scattering curves (I(q) vs. q) in an automatic, unbiased manner. Ab initio molecular envelopes were computed by 20 iterative cycles of simulated annealing starting with a dummy atom model in DAMMIF⁷⁹. The models were aligned, averaged, and filtered using DAMAVER⁸⁰. The final models for Itk and Src were superimposed using SUPCOMB⁸¹. The theoretical scattering curve for Src was calculated from the PDB entry 2SRC using CRYSOL⁸². The agreement between the experimental scattering curves for Src^32K, Itk^32K, Src^32KΔC and Src^32K R175L was evaluated in terms of the chi-squared value using DATCMP. For Itk^32K L351A, rigid body modelling was performed using CORAL⁷⁷ with 2SRC.pdb Δ521–533 (C-tail) as the starting model. The rigid domains of Src were defined as follows: SH3 (84-139), SH2 (148–246), kinase (260-520). The linker residues were implemented in CORAL as dummy residues. Iterative runs of CORAL were performed in which either the SH3 domain alone or both the SH3 and SH2 domains were allowed to move, while the kinase domain was fixed.

Crystallization, data collection, and refinement

Src^32K at 2.6 mg/ml was co-crystallized with 2 mM ADP and 5 mM MgCl₂ in 50 mM PIPES, pH 6.5, 1.2 M sodium tartrate, and 20 mM DTT. Crystals were cryoprotected in reservoir solution plus 22–24% glycerol and plunge frozen in liquid nitrogen. Data were collected at 100 K on ID30A3, ESRF, Grenoble. Crystals grew in space group P2₁2₁2₁ as previously reported for the human apo-protein²¹, but with altered unit cell parameters in b and c (a = 51.01 Å, b = 82.97 Å, c = 105.05 Å; α = β = γ = 90°). The structure was solved by molecular replacement using 2SRC as a search model. Residues 205–212 in a loop on the surface of the SH2 domain and residues 110–117 of the SH3 domain were slightly shifted from their position in 2SRC due to lattice contacts and were therefore rebuilt. The structure was refined by iterative cycles of simulated annealing all-atom refinement in refmac. ADP and a single magnesium ion were built into clear electron density in the nucleotide-binding site and the structure refined to a final R _free = 26.8, R = 20.8 at 2.42 Å resolution.

Molecular dynamics simulations

Molecular dynamics simulations were performed using the GROMOS simulation package³⁵, using the GROMOS 54A7 parameter set to describe molecular interactions³⁶. Relative binding free energies between ADP and ATP were computed using extended thermodynamic integration³⁷ from simulations of the ligands free in solution or bound to the protein. For a more detailed description of the simulation setup, the reader is referred to the supplementary material.

Autophosphorylation assay

Autophosphorylation reactions were performed in a total volume of 100 μl containing 67 nM or 1 μM Src^32K or Itk^32K respectively in 50 mM Tris pH 7.5, 100 mM NaCl, 5 mM MgCl₂, 1 mM TCEP, and 1 mM ATP and incubated at 30 °C. 5 μl samples were taken at each time point and loaded to SDS-PAGE. Phosphorylation was measured by immunoblot. Data were fitted in Origin using a sigmoidal function to model the slow first step (autophosphorylation by unphosphorylated kinase), the fast second step (trans-phosphorylation of unphosphorylated kinase by phosphorylated kinase), and the stationary phase in which substrate (the kinase itself) has been consumed (phosphorylated).

Kinase assay

Kinase assays were performed using the ADP-Glo kit (Promega) according to the manufacturer’s instructions. Kinase reactions contained 500 nM purified kinase, 50 μM SRC substrate (SignalChem #S30-58), and 1 mM ATP in 20 mM Tris pH 7.5, 100 mM NaCl, 1 mM TCEP, 2 mM MgCl₂, and 0.05 mg/ml BSA. Reactions were incubated at RT for the specified time and luminescence was measured using a TECAN Infinite F500 plate reader.

Data availability

Crystallographic coordinates have been deposited in the Protein Data Bank with accession code 6F3F.

References

Chu, D. H., Morita, C. T. & Weiss, A. The Syk family of protein tyrosine kinases in T-cell activation and development. Immunol Rev 165, 167–180 (1998).
Article CAS PubMed Google Scholar
Mocsai, A., Ruland, J. & Tybulewicz, V. L. The SYK tyrosine kinase: a crucial player in diverse biological functions. Nat Rev Immunol 10, 387–402, https://doi.org/10.1038/nri2765 (2010).
Article CAS PubMed PubMed Central Google Scholar
Wang, H. et al. ZAP-70: an essential kinase in T-cell signaling. Cold Spring Harbor perspectives in biology 2, a002279, https://doi.org/10.1101/cshperspect.a002279 (2010).
Article PubMed PubMed Central Google Scholar
Andreotti, A. H., Schwartzberg, P. L., Joseph, R. E. & Berg, L. J. T-cell signaling regulated by the Tec family kinase, Itk. Cold Spring Harbor perspectives in biology 2, a002287, https://doi.org/10.1101/cshperspect.a002287 (2010).
Article PubMed PubMed Central Google Scholar
Takata, M. & Kurosaki, T. A role for Bruton’s tyrosine kinase in B cell antigen receptor-mediated activation of phospholipase C-gamma 2. The Journal of experimental medicine 184, 31–40 (1996).
Article CAS PubMed Google Scholar
Liu, K. Q., Bunnell, S. C., Gurniak, C. B. & Berg, L. J. T cell receptor-initiated calcium release is uncoupled from capacitative calcium entry in Itk-deficient T cells. The Journal of experimental medicine 187, 1721–1727 (1998).
Article CAS PubMed PubMed Central Google Scholar
Tsukada, S. et al. Deficient expression of a B cell cytoplasmic tyrosine kinase in human X-linked agammaglobulinemia. Cell 72, 279–290 (1993).
Article CAS PubMed Google Scholar
Streubel, B., Vinatzer, U., Willheim, M., Raderer, M. & Chott, A. Novel t(5;9)(q33;q22) fuses ITK to SYK in unspecified peripheral T-cell lymphoma. Leukemia 20, 313–318, https://doi.org/10.1038/sj.leu.2404045 (2006).
Article CAS PubMed Google Scholar
Dierks, C. et al. The ITK-SYK fusion oncogene induces a T-cell lymphoproliferative disease in mice mimicking human disease. Cancer research 70, 6193–6204, https://doi.org/10.1158/0008-5472.CAN-08-3719 (2010).
Article CAS PubMed Google Scholar
Huck, K. et al. Girls homozygous for an IL-2-inducible T cell kinase mutation that leads to protein deficiency develop fatal EBV-associated lymphoproliferation. The Journal of clinical investigation 119, 1350–1358 (2009).
Article CAS PubMed PubMed Central Google Scholar
Hussain, A. et al. TEC family kinases in health and disease–loss-of-function of BTK and ITK and the gain-of-function fusions ITK-SYK and BTK-SYK. The FEBS journal 278, 2001–2010, https://doi.org/10.1111/j.1742-4658.2011.08134.x (2011).
Article CAS PubMed Google Scholar
Tarafdar, S., Poe, J. A. & Smithgall, T. E. The accessory factor Nef links HIV-1 to Tec/Btk kinases in an Src homology 3 domain-dependent manner. J Biol Chem 289, 15718–15728, https://doi.org/10.1074/jbc.M114.572099 (2014).
Article CAS PubMed PubMed Central Google Scholar
Manning, G., Whyte, D. B., Martinez, R., Hunter, T. & Sudarsanam, S. The protein kinase complement of the human genome. Science 298, 1912–1934 (2002).
Article ADS CAS PubMed Google Scholar
Yang, W. C., Ching, K. A., Tsoukas, C. D. & Berg, L. J. Tec kinase signaling in T cells is regulated by phosphatidylinositol 3-kinase and the Tec pleckstrin homology domain. J Immunol 166, 387–395 (2001).
Article CAS PubMed Google Scholar
Briggs, S. D., Sharkey, M., Stevenson, M. & Smithgall, T. E. SH3-mediated Hck tyrosine kinase activation and fibroblast transformation by the Nef protein of HIV-1. J Biol Chem 272, 17899–17902 (1997).
Article CAS PubMed Google Scholar
Ching, K. A. et al. TCR/CD3-Induced activation and binding of Emt/Itk to linker of activated T cell complexes: requirement for the Src homology 2 domain. J Immunol 165, 256–262 (2000).
Article CAS PubMed Google Scholar
Moarefi, I. et al. Activation of the Src-family tyrosine kinase Hck by SH3 domain displacement. Nature 385, 650–653, https://doi.org/10.1038/385650a0 (1997).
Article ADS CAS PubMed Google Scholar
Ching, K. A., Kawakami, Y., Kawakami, T. & Tsoukas, C. D. Emt/Itk associates with activated TCR complexes: role of the pleckstrin homology domain. J Immunol 163, 6006–6013 (1999).
CAS PubMed Google Scholar
Nagar, B. et al. Structural basis for the autoinhibition of c-Abl tyrosine kinase. Cell 112, 859–871 (2003).
Article CAS PubMed Google Scholar
Sicheri, F., Moarefi, I. & Kuriyan, J. Crystal structure of the Src family tyrosine kinase Hck. Nature 385, 602–609, https://doi.org/10.1038/385602a0 (1997).
Article ADS CAS PubMed Google Scholar
Xu, W., Harrison, S. C. & Eck, M. J. Three-dimensional structure of the tyrosine kinase c-Src. Nature 385, 595–602 (1997).
Article ADS CAS PubMed Google Scholar
Saito, K., Scharenberg, A. M. & Kinet, J. P. Interaction between the Btk PH domain and phosphatidylinositol-3,4,5-trisphosphate directly regulates Btk. J Biol Chem 276, 16201–16206, https://doi.org/10.1074/jbc.M100873200 (2001).
Article CAS PubMed Google Scholar
Wang, Q. et al. Autoinhibition of Bruton’s tyrosine kinase (Btk) and activation by soluble inositol hexakisphosphate. Elife 4, https://doi.org/10.7554/eLife.06074 (2015).
Banavali, N. K. & Roux, B. The N-terminal end of the catalytic domain of SRC kinase Hck is a conformational switch implicated in long-range allosteric regulation. Structure 13, 1715–1723, https://doi.org/10.1016/j.str.2005.09.005 (2005).
Article CAS PubMed Google Scholar
Sun, G., Sharma, A. K. & Budde, R. J. Autophosphorylation of Src and Yes blocks their inactivation by Csk phosphorylation. Oncogene 17, 1587–1595, https://doi.org/10.1038/sj.onc.1202076 (1998).
Article CAS PubMed Google Scholar
Sun, G. et al. Effect of autophosphorylation on the catalytic and regulatory properties of protein tyrosine kinase Src. Arch Biochem Biophys 397, 11–17, https://doi.org/10.1006/abbi.2001.2627 (2002).
Article CAS PubMed Google Scholar
Rawlings, D. J. et al. Activation of BTK by a phosphorylation mechanism initiated by SRC family kinases. Science 271, 822–825 (1996).
Article ADS CAS PubMed Google Scholar
Heyeck, S. D., Wilcox, H. M., Bunnell, S. C. & Berg, L. J. Lck phosphorylates the activation loop tyrosine of the Itk kinase domain and activates Itk kinase activity. J Biol Chem 272, 25401–25408 (1997).
Article CAS PubMed Google Scholar
Gibson, S., August, A., Branch, D., Dupont, B. & Mills, G. M. Functional LCK Is required for optimal CD28-mediated activation of the TEC family tyrosine kinase EMT/ITK. J Biol Chem 271, 7079–7083 (1996).
Article CAS PubMed Google Scholar
August, A., Sadra, A., Dupont, B. & Hanafusa, H. Src-induced activation of inducible T cell kinase (ITK) requires phosphatidylinositol 3-kinase activity and the Pleckstrin homology domain of inducible T cell kinase. Proc Natl Acad Sci USA 94, 11227–11232 (1997).
Article ADS CAS PubMed PubMed Central Google Scholar
Dinh, M. et al. Activation mechanism and steady state kinetics of Bruton’s tyrosine kinase. J Biol Chem 282, 8768–8776, https://doi.org/10.1074/jbc.M609920200 (2007).
Article CAS PubMed Google Scholar
Xu, W., Doshi, A., Lei, M., Eck, M. J. & Harrison, S. C. Crystal structures of c-Src reveal features of its autoinhibitory mechanism. Mol Cell 3, 629–638 (1999).
Article CAS PubMed Google Scholar
Berg, J., Hung, Y. P. & Yellen, G. A genetically encoded fluorescent reporter of ATP:ADP ratio. Nature methods 6, 161–166, https://doi.org/10.1038/nmeth.1288 (2009).
Article CAS PubMed PubMed Central Google Scholar
Klawitter, J., Schmitz, V., Klawitter, J., Leibfritz, D. & Christians, U. Development and validation of an assay for the quantification of 11 nucleotides using LC/LC-electrospray ionization-MS. Analytical biochemistry 365, 230–239, https://doi.org/10.1016/j.ab.2007.03.018 (2007).
Article CAS PubMed Google Scholar
Schmid, N., Christ, C. D., Christen, M., Eichenberger, A. P. & van Gunsteren, W. F. Architecture, implementation and parallelisation of the GROMOS software for biomolecular simulation. Comput Phys Commun 183, 890–903, https://doi.org/10.1016/j.cpc.2011.12.014 (2012).
Article ADS CAS Google Scholar
Schmid, N. et al. Definition and testing of the GROMOS force-field versions 54A7 and 54B7. European biophysics journal: EBJ 40, 843–856, https://doi.org/10.1007/s00249-011-0700-9 (2011).
Article CAS PubMed Google Scholar
de Ruiter, A. & Oostenbrink, C. Extended Thermodynamic Integration: efficient prediction of lambda derivatives at non-simulated points. Journal of Chemical Theory and Computation (in press) (2016).
Yamaguchi, H. & Hendrickson, W. A. Structural basis for activation of human lymphocyte kinase Lck upon tyrosine phosphorylation. Nature 384, 484–489, https://doi.org/10.1038/384484a0 (1996).
Article ADS CAS PubMed Google Scholar
Bibbins, K. B., Boeuf, H. & Varmus, H. E. Binding of the Src SH2 domain to phosphopeptides is determined by residues in both the SH2 domain and the phosphopeptides. Mol Cell Biol 13, 7278–7287 (1993).
Article CAS PubMed PubMed Central Google Scholar
Gonfloni, S. et al. The role of the linker between the SH2 domain and catalytic domain in the regulation and function of Src. EMBO J 16, 7261–7271, https://doi.org/10.1093/emboj/16.24.7261 (1997).
Article CAS PubMed PubMed Central Google Scholar
Gonfloni, S., Frischknecht, F., Way, M. & Superti-Furga, G. Leucine 255 of Src couples intramolecular interactions to inhibition of catalysis. Nature structural biology 6, 760–764, https://doi.org/10.1038/11537 (1999).
Article CAS PubMed Google Scholar
Endicott, J. A., Noble, M. E. & Johnson, L. N. The structural basis for control of eukaryotic protein kinases. Annu Rev Biochem 81, 587–613, https://doi.org/10.1146/annurev-biochem-052410-090317 (2012).
Article CAS PubMed Google Scholar
Matulis, D., Kranz, J. K., Salemme, F. R. & Todd, M. J. Thermodynamic stability of carbonic anhydrase: measurements of binding affinity and stoichiometry using ThermoFluor. Biochemistry 44, 5258–5266, https://doi.org/10.1021/bi048135v (2005).
Article CAS PubMed Google Scholar
Foda, Z. H., Shan, Y., Kim, E. T., Shaw, D. E. & Seeliger, M. A. A dynamically coupled allosteric network underlies binding cooperativity in Src kinase. Nature communications 6, 5939, https://doi.org/10.1038/ncomms6939 (2015).
Article ADS CAS PubMed PubMed Central Google Scholar
Bennett, B. D. et al. Absolute metabolite concentrations and implied enzyme active site occupancy in Escherichia coli. Nature chemical biology 5, 593–599, https://doi.org/10.1038/nchembio.186 (2009).
Article CAS PubMed PubMed Central Google Scholar
Traut, T. W. Physiological concentrations of purines and pyrimidines. Molecular and cellular biochemistry 140, 1–22 (1994).
Article CAS PubMed Google Scholar
Veech, R. L., Lawson, J. W., Cornell, N. W. & Krebs, H. A. Cytosolic phosphorylation potential. J Biol Chem 254, 6538–6547 (1979).
CAS PubMed Google Scholar
Yang, S., Blachowicz, L., Makowski, L. & Roux, B. Multidomain assembled states of Hck tyrosine kinase in solution. Proc Natl Acad Sci USA 107, 15757–15762, https://doi.org/10.1073/pnas.1004569107 (2010).
Article ADS CAS PubMed PubMed Central Google Scholar
Young, M. A., Gonfloni, S., Superti-Furga, G., Roux, B. & Kuriyan, J. Dynamic coupling between the SH2 and SH3 domains of c-Src and Hck underlies their inactivation by C-terminal tyrosine phosphorylation. Cell 105, 115–126 (2001).
Article CAS PubMed Google Scholar
Boerner, R. J. et al. Correlation of the phosphorylation states of pp60c-src with tyrosine kinase activity: the intramolecular pY530-SH2 complex retains significant activity if Y419 is phosphorylated. Biochemistry 35, 9519–9525, https://doi.org/10.1021/bi960248u (1996).
Article CAS PubMed Google Scholar
Nika, K. et al. Constitutively active Lck kinase in T cells drives antigen receptor signal transduction. Immunity 32, 766–777, https://doi.org/10.1016/j.immuni.2010.05.011 (2010).
Article CAS PubMed PubMed Central Google Scholar
Cowan-Jacob, S. W. et al. The crystal structure of a c-Src complex in an active conformation suggests possible steps in c-Src activation. Structure 13, 861–871, https://doi.org/10.1016/j.str.2005.03.012 (2005).
Article CAS PubMed Google Scholar
Nagar, B. et al. Organization of the SH3-SH2 unit in active and inactive forms of the c-Abl tyrosine kinase. Mol Cell 21, 787–798 (2006).
Article CAS PubMed Google Scholar
Badger, J. et al. c-Abl Tyrosine Kinase Adopts Multiple Active Conformational States in Solution. Biochemistry 55, 3251–3260, https://doi.org/10.1021/acs.biochem.6b00202 (2016).
Article CAS PubMed PubMed Central Google Scholar
Schultheiss, K. P., Craddock, B. P., Suga, H. & Miller, W. T. Regulation of Src and Csk nonreceptor tyrosine kinases in the filasterean Ministeria vibrans. Biochemistry 53, 1320–1329, https://doi.org/10.1021/bi4016499 (2014).
Article CAS PubMed PubMed Central Google Scholar
Schultheiss, K. P., Suga, H., Ruiz-Trillo, I. & Miller, W. T. Lack of Csk-mediated negative regulation in a unicellular SRC kinase. Biochemistry 51, 8267–8277, https://doi.org/10.1021/bi300965h (2012).
Article CAS PubMed PubMed Central Google Scholar
Takeya, T. & Hanafusa, H. Structure and sequence of the cellular gene homologous to the RSV src gene and the mechanism for generating the transforming virus. Cell 32, 881–890 (1983).
Article CAS PubMed Google Scholar
Iba, H., Jove, R. & Hanafusa, H. Lack of induction of neuroretinal cell proliferation by Rous sarcoma virus variants that carry the c-src gene. Mol Cell Biol 5, 2856–2859 (1985).
Article CAS PubMed PubMed Central Google Scholar
Coussens, P. M., Cooper, J. A., Hunter, T. & Shalloway, D. Restriction of the in vitro and in vivo tyrosine protein kinase activities of pp60c-src relative to pp60v-src. Mol Cell Biol 5, 2753–2763 (1985).
Article CAS PubMed PubMed Central Google Scholar
Courtneidge, S. A. Activation of the pp60c-src kinase by middle T antigen binding or by dephosphorylation. EMBO J 4, 1471–1477 (1985).
CAS PubMed PubMed Central Google Scholar
Cooper, J. A., Gould, K. L., Cartwright, C. A. & Hunter, T. Tyr527 is phosphorylated in pp60c-src: implications for regulation. Science 231, 1431–1434 (1986).
Article ADS CAS PubMed Google Scholar
Hirai, H. & Varmus, H. E. Mutations in src homology regions 2 and 3 of activated chicken c-src that result in preferential transformation of mouse or chicken cells. Proc Natl Acad Sci USA 87, 8592–8596 (1990).
Article ADS CAS PubMed PubMed Central Google Scholar
Yeo, M. G. et al. Src SH2 arginine 175 is required for cell motility: specific focal adhesion kinase targeting and focal adhesion assembly function. Mol Cell Biol 26, 4399–4409, https://doi.org/10.1128/MCB.01147-05 (2006).
Article CAS PubMed PubMed Central Google Scholar
Li, T. et al. Constitutive membrane association potentiates activation of Bruton tyrosine kinase. Oncogene 15, 1375–1383, https://doi.org/10.1038/sj.onc.1201308 (1997).
Article CAS PubMed Google Scholar
Ozkirimli, E. & Post, C. B. Src kinase activation: A switched electrostatic network. Protein Sci 15, 1051–1062, https://doi.org/10.1110/ps.051999206 (2006).
Article CAS PubMed PubMed Central Google Scholar
Shukla, D., Meng, Y., Roux, B. & Pande, V. S. Activation pathway of Src kinase reveals intermediate states as targets for drug design. Nature communications 5, 3397, https://doi.org/10.1038/ncomms4397 (2014).
Article ADS PubMed PubMed Central Google Scholar
LaFevre-Bernt, M. et al. Intramolecular regulatory interactions in the Src family kinase Hck probed by mutagenesis of a conserved tryptophan residue. J Biol Chem 273, 32129–32134 (1998).
Article CAS PubMed Google Scholar
Joseph, R. E., Min, L. & Andreotti, A. H. The linker between SH2 and kinase domains positively regulates catalysis of the Tec family kinases. Biochemistry 46, 5455–5462, https://doi.org/10.1021/bi602512e (2007).
Article CAS PubMed Google Scholar
Chopra, N. et al. Dynamic Allostery Mediated by a Conserved Tryptophan in the Tec Family Kinases. PLoS Comput Biol 12, e1004826, https://doi.org/10.1371/journal.pcbi.1004826 (2016).
Article PubMed PubMed Central Google Scholar
Guo, S., Wahl, M. I. & Witte, O. N. Mutational analysis of the SH2-kinase linker region of Bruton’s tyrosine kinase defines alternative modes of regulation for cytoplasmic tyrosine kinase families. International immunology 18, 79–87, https://doi.org/10.1093/intimm/dxh351 (2006).
Article CAS PubMed Google Scholar
Marcotte, D. J. et al. Structures of human Bruton’s tyrosine kinase in active and inactive conformations suggest a mechanism of activation for TEC family kinases. Protein Sci 19, 429–439, https://doi.org/10.1002/pro.321 (2010).
CAS PubMed PubMed Central Google Scholar
Meng, Y. & Roux, B. Computational study of the W260A activating mutant of Src tyrosine kinase. Protein Sci 25, 219–230, https://doi.org/10.1002/pro.2731 (2016).
Article CAS PubMed Google Scholar
Williams, J. C. et al. The 2.35 A crystal structure of the inactivated form of chicken Src: a dynamic molecule with multiple regulatory interactions. J Mol Biol 274, 757–775, https://doi.org/10.1006/jmbi.1997.1426 (1997).
Article CAS PubMed Google Scholar
Filippakopoulos, P. et al. Structural coupling of SH2-kinase domains links Fes and Abl substrate recognition and kinase activation. Cell 134, 793–803, https://doi.org/10.1016/j.cell.2008.07.047 (2008).
Article CAS PubMed PubMed Central Google Scholar
Ogawa, A. et al. Structure of the carboxyl-terminal Src kinase, Csk. J Biol Chem 277, 14351–14354, https://doi.org/10.1074/jbc.C200086200 (2002).
Article CAS PubMed Google Scholar
Lieser, S. A. et al. Phosphoryl transfer step in the C-terminal Src kinase controls Src recognition. J Biol Chem 280, 7769–7776, https://doi.org/10.1074/jbc.M411736200 (2005).
Article CAS PubMed Google Scholar
Petoukhov, M. V. et al. New developments in the program package for small-angle scattering data analysis. J Appl Crystallogr 45, 342–350, https://doi.org/10.1107/S0021889812007662 (2012).
Article CAS PubMed PubMed Central Google Scholar
Petoukhov, M. V., Konarev, P. V., Kikhney, A. G. & Svergun, D. I. ATSAS 2.1 - towards automated and web-supported small-angle scattering data analysis. Journal of Applied Crystallography 40, S223–S228, https://doi.org/10.1107/S0021889807002853 (2007).
Article CAS Google Scholar
Franke, D. & Svergun, D. I. DAMMIF, a program for rapid ab-initio shape determination in small-angle scattering. J. Appl. Crystallogr. 42, 342–346 (2009).
Article CAS PubMed PubMed Central Google Scholar
Volkov, V. V. & Svergun, D. I. Uniqueness of ab initio shape determination in small-angle scattering. J. Appl. Crystallogr. 36, 860–864, https://doi.org/10.1107/s0021889803000268 (2003).
Article CAS Google Scholar
Kozin, M. B. & Svergun, D. I. Automated matching of high- and low-resolution structural models. J. Appl. Crystallogr. 34, 33–41, https://doi.org/10.1107/s0021889800014126 (2001).
Article CAS Google Scholar
Svergun, D., Barberato, C. & Koch, M. H. J. CRYSOL - a Program to Evaluate X-ray Solution Scattering of Biological Macromolecules from Atomic Coordinates. J. Appl. Cryst. 28, 768–773 (1995).
Article CAS Google Scholar

Download references

Acknowledgements

We would like to thank Markus Hartl (MFPL) for assistance with mass spectrometry, and Milla Neffling and Christian Baumann from Sciex for their assistance in setting up the instrument for intact mass measurements. We thank Adam Round for BioSAXS beam line support at the ESRF, and Joan Lopez Arolas and Christian Fercher for help with X-ray data collection. We thank Chris Oostenbrink (BOKU) for advice and assistance with molecular dynamics simulations. This work was funded by a start-up package from the MFPL and a grant from the Austrian Science Fund (FWF; P 28135) to T.A.L. F.v.R. was supported by a DOC Fellowship from the Austrian Academy of Sciences.

Author information

Authors and Affiliations

Department of Structural and Computational Biology, Max F. Perutz Laboratories (MFPL), Campus Vienna Biocenter 5, 1030, Vienna, Austria
Freia von Raußendorf & Thomas A. Leonard
Institute of Molecular Modeling and Simulation, University of Natural Resources and Life Sciences (BOKU), 1190, Vienna, Austria
Anita de Ruiter
Department of Medical Biochemistry, Medical University of Vienna, 1090, Vienna, Austria
Thomas A. Leonard

Authors

Freia von Raußendorf
View author publications
You can also search for this author in PubMed Google Scholar
Anita de Ruiter
View author publications
You can also search for this author in PubMed Google Scholar
Thomas A. Leonard
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

T.A.L. and F.v.R. designed, and F.v.R. performed, all lab-based experiments reported in this manuscript. T.A.L. refined the structure of Src in complex with ADP. A.d.R. performed and analysed the molecular dynamics simulations. T.A.L. wrote the manuscript with help from F.v.R.

Corresponding author

Correspondence to Thomas A. Leonard.

Ethics declarations

Competing Interests

The authors declare that they have no competing interests.

Additional information

Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Supplementary Information

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

von Raußendorf, F., de Ruiter, A. & Leonard, T.A. A switch in nucleotide affinity governs activation of the Src and Tec family kinases. Sci Rep 7, 17405 (2017). https://doi.org/10.1038/s41598-017-17703-5

Download citation

Received: 15 June 2017
Accepted: 29 November 2017
Published: 12 December 2017
DOI: https://doi.org/10.1038/s41598-017-17703-5

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.