From structure to the dynamic regulation of a molecular switch: A journey over 3 decades

Abstract

It is difficult to imagine where the signaling community would be today without the Protein Data Bank. This visionary resource, established in the 1970s, has been an essential partner for sharing information between academics and industry for over 3 decades. We describe here the history of our journey with the protein kinases using cAMP-dependent protein kinase as a prototype. We summarize what we have learned since the first structure, published in 1991, why our journey is still ongoing, and why it has been essential to share our structural information. For regulation of kinase activity, we focus on the cAMP-binding protein kinase regulatory subunits. By exploring full-length macromolecular complexes, we discovered not only allostery but also an essential motif originally attributed to crystal packing. Massive genomic data on disease mutations allows us to now revisit crystal packing as a treasure chest of possible protein:protein interfaces where the biological significance and disease relevance can be validated. It provides a new window into exploring dynamic intrinsically disordered regions that previously were deleted, ignored, or attributed to crystal packing. Merging of crystallography with cryo-electron microscopy, cryo-electron tomography, NMR, and millisecond molecular dynamics simulations is opening a new world for the signaling community where those structure coordinates, deposited in the Protein Data Bank, are just a starting point! It is difficult to imagine where the signaling community would be today without the Protein Data Bank. This visionary resource, established in the 1970s, has been an essential partner for sharing information between academics and industry for over 3 decades. We describe here the history of our journey with the protein kinases using cAMP-dependent protein kinase as a prototype. We summarize what we have learned since the first structure, published in 1991, why our journey is still ongoing, and why it has been essential to share our structural information. For regulation of kinase activity, we focus on the cAMP-binding protein kinase regulatory subunits. By exploring full-length macromolecular complexes, we discovered not only allostery but also an essential motif originally attributed to crystal packing. Massive genomic data on disease mutations allows us to now revisit crystal packing as a treasure chest of possible protein:protein interfaces where the biological significance and disease relevance can be validated. It provides a new window into exploring dynamic intrinsically disordered regions that previously were deleted, ignored, or attributed to crystal packing. Merging of crystallography with cryo-electron microscopy, cryo-electron tomography, NMR, and millisecond molecular dynamics simulations is opening a new world for the signaling community where those structure coordinates, deposited in the Protein Data Bank, are just a starting point! The scientific community is enormously grateful to the visionary leaders of the early crystallography community who realized how important it would be to have an international resource where all new structures would be deposited and validated. Max Perutz, Michael Rossmann, and Fred Richards were among the early pioneers who championed the PDB concept. In 1971 the Protein Data Bank (PDB) was officially announced in Nature New Biology (1Crystallography: Protein Data Bank.Nat. New Biol. 1971; 233: 223Crossref PubMed Scopus (39) Google Scholar) as a joint venture between Brookhaven National Laboratory under the direction of Walter Hamilton and the Cambridge Crystal Data Center, founded by Olga Kennard, with an initial holding of seven protein structures. Helen Berman, who was also involved with the initial establishment of the PDB, led the transformation of the PDB into a modern, worldwide database in 1998. The determination and dedication of these two pioneering women, along with the visionary leadership of Perutz, Richards, Rossmann, and others, was essential to the founding of the PDB and to the development of the modern PDB. From an initial concept as a static archive to the development of a dynamic research tool, the development of the PDB allows the research community to leverage the investments made by many funding agencies across disciplinary and national boundaries and solidifies the concept that these structures belong to the community and not just to the individual laboratory that solved the structure. It represents an exciting commitment to science and to the scientific community. Of course, no one envisioned the ways in which this community would explode along with the technologies in computing over the ensuing decades. We describe here the impact that this vision has had on the signaling community using cAMP-dependent protein kinase (PKA) as a model. We not only highlight the wide-ranging benefits that have evolved but also emphasize future challenges and new opportunities to build on for the future as our understanding of proteins evolves from those early days of the first crystals of hemoglobin, lysozyme, and lactate dehydrogenase. The structure of the PKA catalytic (C) subunit was our first entry into the PDB in 1991. Taylor had been trained in protein chemistry and structural biology at the MRC Laboratory of Molecular Biology in Cambridge, and Michael Rossman was one of her major mentors in her early career when she was sequencing lactate dehydrogenase. Lynn Ten Eyck with his long-term commitment to public data bases was also a part of the team as was Janusz Sowadski who led the crystallography. So, there was never a question in their minds not to share this structure with the signaling community that was just beginning to make inroads into the structure of protein kinases, which are now recognized to represent one of the largest gene families and were already associated with many diseases. Ours was the first protein kinase structure to be solved, and even before the structure was published, we were anxious to share this new information with our colleagues such as Bruce Kemp and others. A core philosophy of my laboratory has been that one needs a structure before one can truly begin to understand function, and so for me having a structure was a starting point. Although we did not fully appreciate it at the time, we were very fortunate with that first structure to have captured an active and fully phosphorylated protein kinase (Fig. 1). The bilobal fold was novel as was the ATP-binding pocket (2Knighton D.R. Zheng J.H. Ten Eyck L.F. Ashford V.A. Xuong N.H. Taylor S.S. Sowadski J.M. Crystal structure of the catalytic subunit of cyclic adenosine monophosphate-dependent protein kinase.Science. 1991; 253: 407-414Crossref PubMed Google Scholar). Although ATP was present in the buffer of that initial structure and the position of the pseudosubstrate peptide derived from the heat-stable protein kinase inhibitor was clearly defined (3Knighton D.R. Zheng J.H. Ten Eyck L.F. Xuong N.H. Taylor S.S. Sowadski J.M. Structure of a peptide inhibitor bound to the catalytic subunit of cyclic adenosine monophosphate-dependent protein kinase.Science. 1991; 253: 414-420Crossref PubMed Google Scholar), the position of the nucleotide was only predicted to be localized to the cleft between the two lobes. In the subsequent structure when we used a 10-fold excess of Mg2+ over ATP we were able to trap both the ATP along with two Mg2+ ions and the peptide (Fig. 2A) (4Zheng J. Knighton D.R. Ten Eyck L.F. Karlsson R. Xuong N. Taylor S.S. Sowadski J.M. Crystal structure of the catalytic subunit of cAMP-dependent protein kinase complexed with MgATP and peptide inhibitor.Biochemistry. 1993; 32: 2154-2161Crossref PubMed Google Scholar). This represented a novel ATP-binding site that is defined by the glycine-rich loop (G-Loop) (Fig. 2D) that embraces the of ATP and to it at the of the cleft as as to position the for to a protein It is from the A of motif A in and PubMed Google Scholar) or the motif that had been by for ATP in and in the of structures in Biol. PubMed Scopus Google Scholar). So, we captured an of information in those first we also or to appreciate of the essential of the Although both ions were captured in that structure, we did not appreciate the of the as the that is essential for of ATP and for the of the nucleotide (Fig. Taylor S.S. A a for from protein kinase PubMed Scopus Google bound to of a catalytic the structure and dynamics of kinase for PubMed Scopus Google Scholar). protein kinase structures still not or the of that was also from this structure was an of the and of this protein kinase structure and the of the concept that protein kinases had evolved not to be but to be dynamic molecular as a of kinase The PKA not only defined a novel fold but also defined a novel ATP-binding site that was from the The of the that embraces the nucleotide over many decades. the site defined in (4Zheng J. Knighton D.R. Ten Eyck L.F. Karlsson R. Xuong N. Taylor S.S. Sowadski J.M. Crystal structure of the catalytic subunit of cAMP-dependent protein kinase complexed with MgATP and peptide inhibitor.Biochemistry. 1993; 32: 2154-2161Crossref PubMed Google Scholar) the of and that the nucleotide in a the is essential for the of ATP in the protein kinase inhibitor and for Taylor S.S. A a for from protein kinase PubMed Scopus Google bound to of a catalytic the structure and dynamics of kinase for PubMed Scopus Google Taylor S.S. of the nucleotide and in cAMP-dependent protein PubMed Scopus Google Scholar). that the ATP the Taylor S.S. Ten Eyck L.F. A for the of active protein PubMed Scopus Google Scholar), the Taylor S.S. Ten Eyck L.F. of active and protein kinases a PubMed Scopus Google Scholar), and the N. Taylor S.S. the structural of protein kinase Biol. PubMed Scopus Google Scholar) as as from the and J. Taylor S.S. that protein kinase A PubMed Scopus Google Scholar). This the allostery that the J. Taylor S.S. A dynamic core allostery in protein PubMed Scopus Google S.S. Protein of dynamic regulatory PubMed Scopus Google Taylor S.S. allostery in protein PubMed Scopus Google Scholar). of the kinase glycine-rich loop (G-Loop) is a of the protein to the in the ATP and is part of the provides and is part of that is by the This motif an and a In kinases such as PKA the is as a The across and the catalytic the to the loop site in and the is the catalytic that is at the site of In to kinases such as which have evolved to be the protein kinases have evolved to be molecular are understand and fully appreciate the that the of an active kinase many protein kinase structures as as an that the static structures that are captured in crystal (Fig. 1). the our structure of that first protein kinase was solved there were many protein kinases, all deposited in the PDB, and all were solved by molecular were active were a to of the for we used a tool, Taylor S.S. Ten Eyck L.F. of active and protein kinases a PubMed Scopus Google Scholar). This where we active and kinases, led to the of a that to be the of active The were part of of the such as the motif (Fig. that the catalytic loop and the motif (Fig. that the in the as as the and from the The essential of the is that it is in to of the were in the that is in a to the and is by the of ATP Taylor S.S. Ten Eyck L.F. A for the of active protein PubMed Scopus Google Scholar). The of this core for allostery was only fully when was able to the dynamics of these J. Taylor S.S. and in protein kinase A by PubMed Scopus Google J. Taylor S.S. A dynamic core allostery in protein PubMed Scopus Google Scholar). The that our and what we learned over the 3 since that initial structure is in our understanding of the active site cleft and ATP is in which also of the that are by all protein From the first structures we defined all of the (Fig. 2A) and the (Fig. but we to appreciate the that embraces the active site with the captured at the of the cleft between the two and the in a fully for to a protein at the of the cleft (Fig. This of the nucleotide with the (Fig. 2D) is from the structures that both the and the where it is the that is at the of the captured in our first structures were the two (Fig. we to appreciate the of the Taylor S.S. A a for from protein kinase PubMed Scopus Google Scholar) as essential for and for the of ATP bound to of a catalytic the structure and dynamics of kinase for PubMed Scopus Google Taylor S.S. of the nucleotide and in cAMP-dependent protein PubMed Scopus Google Scholar). subsequent of the those of the structural and how the kinase is in that are to Taylor S.S. of a protein PubMed Scopus Google Scholar). The of the that the kinase to of the are captured by and in the these we are able to a dynamic of that static structure that was captured in the we are also that kinase of as as these are so of kinases is a protein kinases Biol. PubMed Google Scholar). In the of PKA there are on the R. The protein kinase of the PubMed Scopus Google Scholar) (Fig. that we which in so many ways has as a for the protein kinase is the subunit which was in of for the catalytic subunit of cAMP-dependent protein PubMed Google Scholar). The subunit is and localized to and Molecular of a protein kinase from a for the catalytic subunit of cAMP-dependent protein PubMed Google Scholar). In the subunit in for a of the catalytic subunit of cAMP-dependent protein Biol. PubMed Google Scholar), is in in many at a the gene also has many at the first of the cAMP-dependent protein kinase (PKA) catalytic subunit PubMed Scopus Google novel of the gene of cAMP-dependent protein Biol. PubMed Scopus Google Scholar) (Fig. and these are in a of the between and are localized to the and the (Fig. that both and the kinase core that is by all protein with the PKA regulatory where there are that for it is that the of are also of PKA in the for is for by such as and are in is in The that is by these that these proteins be as which to the of the important the kinase signaling community is how to the of In the of PKA we are fortunate in that the kinase is by cAMP-binding regulatory that are by all protein kinases are into an by that are to the kinase to the of that for the For this of the kinases in the PDB represent kinase that are so many of the regulatory that are in the and that the kinase is opening a new for the signaling community in that it is us to at these full-length we the and protein kinases with the kinases there is a The protein kinases in the of to be into an by their and of the only In of their of the for in part of the and of the it has been to trap this of this essential information is when one only at the kinase one the loop in a disordered In the kinases are first as an active kinase by the that can kinases as as N. N. Taylor S.S. The of kinase is a regulatory PubMed Scopus Google Scholar). The active kinase is in an that can be by such as and For PKA there are and to the in a a of that all to by S.S. R. of macromolecular from Biol. PubMed Scopus Google Scholar). The are as a by an which the two cyclic nucleotide in in In to the the as a for A kinase proteins which an that the kinase to the in and Biol. PubMed Scopus Google Scholar). In this one signaling to a Although we have captured that the it has been to the of the even when are how our understanding of the PKA evolved we as an beginning with the structure of an that defined the cyclic Xuong N.H. Ten Eyck Taylor S.S. subunit of protein kinase Structure of with PubMed Google Scholar). The subsequent structure of the full-length J. Taylor S.S. PKA structure an with for and PubMed Scopus Google Scholar) made us appreciate the of a novel motif that we had originally attributed to crystal packing disease mutations in are our of this motif and for PKA The of the is and it has also been the by a of that crystallography Xuong N.H. Ten Eyck Taylor S.S. subunit of protein kinase Structure of with PubMed Google Scholar), R. of an for the cAMP-dependent of protein kinase A as by the of PubMed Scopus Google Scholar), and Taylor S.S. to and proteins to and with PubMed Scopus Google Scholar). It is also the subunit that is in such as and J. R. and in the gene regulatory subunit of protein kinase A PubMed Scopus Google J. of mutations a molecular but Biol. PubMed Scopus Google Scholar). In our initial structure of a of the subunit we captured both Xuong N.H. Ten Eyck Taylor S.S. subunit of protein kinase Structure of with PubMed Google Scholar) and with the protein in a of gene in Ten Eyck L.F. Taylor S.S. The signaling PubMed Scopus Google Scholar). an A that is to an motif that is by two and The which the motif for is in a loop between and and also a that is an site for the The subsequent structures of two us to appreciate the that when the is the it in an Xuong N.H. Taylor S.S. Crystal structure of a between the catalytic and regulatory of PubMed Scopus Google Taylor S.S. structure a for cAMP-dependent PubMed Scopus Google Scholar). We a that our to This us to the of the that are associated with that to the of PKA (Fig. Taylor S.S. Ten Eyck L.F. A for cyclic nucleotide Biol. PubMed Scopus Google Scholar). In to a understanding of the of the in the the led to the of an important and novel motif that had been in our initial of the and that also the from protein Taylor S.S. Ten Eyck L.F. A for cyclic nucleotide Biol. PubMed Scopus Google Scholar). This to as the motif (Fig. an at the and a A is in the that the two are two in is to the that the to the and the the to the Although two mutations are in the first we did not fully appreciate the significance of the motif we solved the structure of the full-length subunit J. Taylor S.S. PKA structure an with for and PubMed Scopus Google Scholar). we that the by two was an of crystal packing. with the full-length in did we appreciate that the motif was a and important between the two The significance of this motif has been by new structures of the full-length and by disease mutations that are this motif as as mutations that the of the motif in protein:protein The between the two was in the initial structure of the and was in all of our subsequent structures of we attributed this to crystal packing (Fig. In this structure the the motif to the crystal packing (Fig. It was not we solved the structure of the full-length J. Taylor S.S. PKA structure an with for and PubMed Scopus Google Scholar) (Fig. that we that this motif was a of the and for the structure of the in with the which are fully J.M. Taylor S.S. J. among structures of the and protein kinase A regulatory subunit of the Biol. PubMed Scopus Google Scholar). the structure we to appreciate that the not only the motif but also the that the motif (Fig. this is by between the two (Fig. In the structure of the only seven were by packing the (Fig. in the full-length an (Fig. The of this was first by R. Taylor S.S. between in the regulatory subunit of protein kinase as by Biol. PubMed Scopus Google Scholar) and by Taylor S.S. of PKA of cyclic nucleotide PubMed Scopus Google Scholar), which to the of the first captured the of the and the two and how the were by the clearly that the motif in the is a of the of the full-length The structure of the full-length the of the and an for the mutations J. Taylor S.S. PKA structure an with for and PubMed Scopus Google Scholar). the packing of the two in is a of the full-length the is in this structure to to and the of the that the of the full-length Although the and the are present in this these regions are disordered so the between these and the is not the between the two the of the in to the of and It also to the significance of the The the of the that the motif that is and part of the The intrinsically disordered that the inhibitor site The packing of the also to the of the two mutations in this motif J. Taylor S.S. PKA structure an with for and PubMed Scopus Google Scholar). In the these two are the why the for is from to to to with of these two a at the such as or to the as by and the is to to of the full-length (Fig. us to at the that we had in our structures but attributed to crystal packing Xuong N.H. Ten Eyck Taylor S.S. subunit of protein kinase Structure of with PubMed Google Scholar). to important interfaces from crystal packing is a and which is an of crystal with have us to of crystal packing as a treasure chest us what is The that one is and one is is not a Crystal packing us what is and when these interfaces are with disease mutations these crystal packing interfaces can us which are the relevance of these interfaces can be by From we in the of that there is an of structures not a Taylor S.S. the of a PubMed Scopus Google Scholar). In to the at the which is a in the and in the subunit by two J. Taylor S.S. of the two of the regulatory subunit of cAMP-dependent protein kinase Biol. PubMed Google Scholar), the between the two are and not trap the by The relevance is captured by In the of the in the are to be in the J.M. Taylor S.S. S.S. of PKA regulatory subunit protein J. PubMed Google R. of protein kinase A is by subunit Biol. PubMed Scopus Google Scholar). The can be by or by so is to be in the and in the on data and the structure of a new of an of has been and this on the N. J. J. J. J. Taylor S.S. protein kinase A by of regulatory as by PubMed Google Scholar). The data of this that to and these to regions that both the and regions of the motif in (Fig. This has a for the and is to a where the is in a that it in an active A new was also for the subunit that with Taylor S.S. J. of a PKA regulatory subunit and PubMed Scopus Google Scholar). of the also a and in a of The of that is in the a for these The of the allows us to appreciate the of and these are when one the of the The motif is a in the it is not a in the which both have an on with between the two the of the motif in the of the have been solved, and in the between the two in the two (Fig. J. J.H. N. Taylor S.S. PKA ATP as an inhibitor that with the PubMed Scopus Google Scholar). This is also in the that is with the that is to the of a protein that is the of a Taylor S.S. of the PKA with the or PubMed Scopus Google Scholar). It is that this is also in of the of the subunit that at the (Fig. initial structures that ATP is also in these which are to by J. J. and S.S. ATP is to is clearly an essential of of the The motif is a of all (Fig. but the motif not the of PKA The and of the PKA is major concept that we and what we have learned over the since those first PKA structures were solved and be in with in a in in the motif in the is not part of the it is to where it is to with proteins that to PKA signaling that are at by The of the on our early data J.M. Taylor S.S. J. among structures of the and protein kinase A regulatory subunit of the Biol. PubMed Scopus Google Taylor S.S. J. in the structures of protein kinase A Biol. PubMed Scopus Google Scholar), is the of the and our crystal structure of this first captured the between the two (Fig. Taylor S.S. Structure and allostery of the PKA PubMed Scopus Google Scholar). It why the be captured in an but the full-length even the is not This structure also clearly that the is made of the loop in and the of the the motif is not part of this but is to with structure is the first structure of PKA J.H. J. Taylor S.S. of the PKA the protein and in Biol. PubMed Scopus Google Scholar). Although the is this structure the of the full-length where the the to the is between the two (Fig. This the at the of to with and is to the to in to the and the J. Taylor S.S. PKA to PubMed Scopus Google Scholar). The of the would also be to the in this model. The structure of the and dynamics of the an between the two and allows us to appreciate the of the between the two and why it is so to the dynamics and of this it why crystallography is not to the dynamics of the between the and subunits. understanding of the of PKA signaling the of of us a protein from our a of as a as a dynamic molecular as a of as a of protein:protein into of these and a of for Although we originally of proteins as we now appreciate that are dynamic and that regions are to and on the of or signaling The of these to as disordered can be protein PubMed Scopus Google Scholar) and are of the of biological signaling The crystal structure the information a starting for to their to the exciting that are these structures in the PDB allows to the information that is captured in that structure, and even today we are still new that PKA structure. This has been an journey over 3 to the essential of how the protein kinases have evolved to be one of the important molecular in biology and to understand how are in to biological many when this is even The signaling of the can be so by a in a which is only one in the we not have this so without the PDB as a resource where all data were It has been and to be an essential for all of us in the signaling and structural biology Although we that was solved with that first structure of the PKA we now that it was only the we are still only our journey at challenges can now be with with crystallography and NMR, and with our new It this it us to the the to appreciate new ways that with the of and our is to understand biology and to understand how biology is by This is what we are to and understand with our structures.

Topics & Concepts