1 Introduction
The human kinome constitutes a large superfamily of essential enzymes with more
than 500 family members [1]. Kinases can be grouped in two main classes based
on their catalytic activity on serine/threonine and on tyrosine residues. Based on
their primary sequence and conserved structural features, the human kinome
has been classified into eight major kinase groups, namely, the AGC (protein
kinase A, G and C), CaMK (calcium/calmodulin-dependent kinases), CMGC
(cyclin-dependant kinases, MAP kinases, glycogen synthase kinases, casein
kinases 2), TK (tyrosine kinases), STE (homologues of yeast sterile 7), CK1
(casein kinases), TKL (tyrosine kinase-like) and the RCG (receptor guanylate
cyclases) kinase families [1]. In addition, a large number of kinases share only
weak sequence homology with any of these major groups and have been classified
as “other” and atypical kinases. While the group of other kinases are typical protein
kinases, the group of atypical kinases lack canonical sequence motifs of the kinase
catalytic domain. Some of the atypical kinases have therefore been reassigned
as non-kinase proteins, whereas there are also several recent additions to this
group such as the FAM20 kinases [2–4]. Interestingly, around 10% of all human
kinases are considered catalytically inactive and have been classified as
pseudokinases. Pseudokinases share a typical kinase domain fold, but they lack at
least one conserved structural motif which is considered important for catalytic
activity [5, 6]. Pseudokinases have essential signalling function despite their
lack of catalytic activity by acting as scaffolding proteins and allosteric regulators
of catalytically active kinases.
Many protein kinases are deregulated in human disease which made protein
kinases major drug targets. However, only few have been targeted to date offering
huge opportunities for future drug development efforts [7, 8]. The high sequence
homology within the kinase ATP-binding site which is the target of most kinase
inhibitors poses however challenges on the development of selective inhibitors.
On the other hand, design efforts for selective inhibitors are now facilitated
by the large number of crystal structures that are now available covering about
40% of the kinase family. The protein data bank (PDB; http://www.rcsb.org/)
and the KLIFS database (http://klifs.vu-compmedchem.nl/) [9] currently list 4,521
crystal structures covering 293 kinases. Current strategies for the development of
selective inhibitors comprise now several structure-based strategies such as covalent
targeting of unique cysteine residues with in the ATP-binding site [10, 11], allosteric
inhibitors [12, 13] and conventional inhibitors with good shape complementarity
[14]. Here we review structural features important for kinase catalytic function
and regulation as well as strategies for structure-based kinase inhibitor design.
2
S. Röhm et al.
The human kinome constitutes a large superfamily of essential enzymes with more
than 500 family members [1]. Kinases can be grouped in two main classes based
on their catalytic activity on serine/threonine and on tyrosine residues. Based on
their primary sequence and conserved structural features, the human kinome
has been classified into eight major kinase groups, namely, the AGC (protein
kinase A, G and C), CaMK (calcium/calmodulin-dependent kinases), CMGC
(cyclin-dependant kinases, MAP kinases, glycogen synthase kinases, casein
kinases 2), TK (tyrosine kinases), STE (homologues of yeast sterile 7), CK1
(casein kinases), TKL (tyrosine kinase-like) and the RCG (receptor guanylate
cyclases) kinase families [1]. In addition, a large number of kinases share only
weak sequence homology with any of these major groups and have been classified
as “other” and atypical kinases. While the group of other kinases are typical protein
kinases, the group of atypical kinases lack canonical sequence motifs of the kinase
catalytic domain. Some of the atypical kinases have therefore been reassigned
as non-kinase proteins, whereas there are also several recent additions to this
group such as the FAM20 kinases [2–4]. Interestingly, around 10% of all human
kinases are considered catalytically inactive and have been classified as
pseudokinases. Pseudokinases share a typical kinase domain fold, but they lack at
least one conserved structural motif which is considered important for catalytic
activity [5, 6]. Pseudokinases have essential signalling function despite their
lack of catalytic activity by acting as scaffolding proteins and allosteric regulators
of catalytically active kinases.
Many protein kinases are deregulated in human disease which made protein
kinases major drug targets. However, only few have been targeted to date offering
huge opportunities for future drug development efforts [7, 8]. The high sequence
homology within the kinase ATP-binding site which is the target of most kinase
inhibitors poses however challenges on the development of selective inhibitors.
On the other hand, design efforts for selective inhibitors are now facilitated
by the large number of crystal structures that are now available covering about
40% of the kinase family. The protein data bank (PDB; http://www.rcsb.org/)
and the KLIFS database (http://klifs.vu-compmedchem.nl/) [9] currently list 4,521
crystal structures covering 293 kinases. Current strategies for the development of
selective inhibitors comprise now several structure-based strategies such as covalent
targeting of unique cysteine residues with in the ATP-binding site [10, 11], allosteric
inhibitors [12, 13] and conventional inhibitors with good shape complementarity
[14]. Here we review structural features important for kinase catalytic function
and regulation as well as strategies for structure-based kinase inhibitor design.
2
S. Röhm et al.