Cell News 2/2014
17
Research news
reversible, and dose-dependent inhibition of single components
in the cell. However, highly selective agonist and antagonist
compound of protein kinases are difficult to develop since the
500 protein kinases contain highly homologous ATP-binding
pockets. Recently, combining protein engineering and organic
synthesis called chemical genetics promote the generation of a
target small molecule inhibitor. This approach ensures that only
a specific protein is targeted by the synthesized small molecule
(Dar et al., 2008).
Protein Kinases Share a Conserved Catalytic Core
Protein kinases in eukaryotes share a highly conserved catalytic
core known as the protein kinase domain, which consists of ap-
proximately 250-300 amino acid residues (Hanks and Hunter,
1995). The protein kinase domain is located in the cytoplasm
either as an element of a multidomain protein or in transmemb-
rane receptor kinases. The extracellular domains of the receptor
kinases are quite divergent and carry oligosaccharide chains to
promote the surface residency of these receptors. This diver-
gence would allow these proteins to respond to a wide range of
external signals. The eukaryotic genome contains a huge prote-
in kinase superfamily that can be divided into 10 groups (AGC,
CAMK, CMGC, TK, TKL, STE, CKI, RGC, atypical and others). These
groups have furthermore been subdivided into 134 families and
201 subfamilies (Manning et al., 2002). The close analysis of pro-
tein kinases showed twelve highly conserved subdomains in the
primary structure of the protein kinase superfamily. These sub-
domains mediate a catalytic function by binding ATP, substrate
binding and catalysis (Madhusudan et al., 1994) (Fig. 1A). The
first protein kinase structure solved in an inactive, phosphorylate
state is the cAMP-dependent protein kinase PKA (Knighton et
al., 1991). Solving the protein kinase structures revealed similar
three-dimensional structures to PKA, which is expected by the
high degree of sequence homology. The structure visualized by
X-ray crystallography showed bilobal architecture, with an ATP
binding pocket and an active site located within a deep cleft
sandwiched between the two lobes. The smaller N-terminal lobe
(N-lobe) is consistent of
β
-sheets and the larger C-terminal lobe
(C-lobe) is essentially
α
-helical. The two segment lobes are con-
nected by one stretch of polypeptide known as the hinge region.
The important part of the C-lobe is called the activation seg-
ment, or A-loop (Fig. 1B). This is where the phospho-acceptor
site binds, which causes small deviations of the activation seg-
ment, and determines if a kinase is a serine/threonine or tyro-
sine kinase. Also it dictates the preferred or allowed consensus
sequence for a kinase. For example, the activation loop of the
proline-directed kinases such as cyclin-dependent kinase 2 CDK2
and mitogen-activated protein kinase ERK2 is in such a confor-
mation that it can accommodate a proline next to the site to be
phosphorylated in their substrates. To enhance specificity, these
primary recognition mechanisms are often supplemented by se-
condary interactions (De Bondt et al., 1993; Zhang et al., 1994).
Substrate Targeting Mechanisms Based on Short
Peptide Motifs: A-Loop Activation is the Primary
Determinants for Substrate Recognition
Regulatory mechanisms are revealed by structural information
for many protein kinases and showed that the primary determi-
nants reflect the positioning of active-site elements within the
catalytic domain. Furthermore, secondary determinants are con-
stituted by external unfolded linear polypeptide elements when
the substrate protein is not in complex with a protein kinase do-
main. The first type of regulation is called the pseudo-substrate
inhibition when protein kinases are inactivated by a regulatory
C-terminal polypeptide stretch and involve the kinase domain in
complex with the ATP and peptide substrate-binding sites. In this
type of regulation, the secondary targeting is peptide-directed
and the recognition of the substrate is only dependent on a sig-
nature sequence motif. There are examples of catalytic activity
inhibition of protein kinases, for which available structural in-
formation exists. The C-terminal regulatory segment of the cal-
cium/calmodulin-dependent protein kinase I (CaMKI) contains
both autoinhibitory and calcium/calmodulin binding sites, which
associate with the kinase domain to generate an open inactive
conformation (Yokokura et al., 1995) (Fig. 2A). The activation of
CAMKI is caused by rearrangement of the intramolecular associ-
ation of the inhibitory element with the protein kinase domain.
Here, CaMKI is autoinhibited by a pseudosubstrate following
binding of calmodulin to the C-terminal inhibitory element of
CAMKI in a calcium-dependent manner. The full activation of
CaMKI is accomplished by phosphorylation of the A-loop threo-
nine residue (Yokokura et al., 1995).
The juxtamembrane region of receptor tyrosine kinases achieves
a second type of regulation. Receptor tryrosine kinases (RTKs) are
multidomain proteins and contain an N-terminal extracellular
domain, which binds an extracellular ligand, a transmembrane
segment and a cytoplasmic region containing a tryrosine kinase
domain with catalytic activity. The perception of a ligand signal
via RTKs extracellular domain would induce oligomerization and
cause tyrosine kinase activation by intermolecular autophospho-
rylation leading to a cellular response. This type of regulation is
found in the Eph receptor tyrosine kinases (Dodelet and Pasqua-
le, 2000). The domain structure of Eph receptors consists of an
N-terminal extracellular ligand-binding domain, a cysteine-rich
region, two fibronectin type III repeats, a single transmembrane
domain, a juxtamembrane segment, a tyrosine kinase domain, a
sterile alpha motif (SAM) domain, and a C-terminal PDZ domain
binding motif. Here, a highly conserved motif with two phospho-
regulatory tyrosine residues is located in the Eph receptor juxta-
membrane region. The activation of the Eph tyrosine kinase do-
main is regulated by the phosphorylation of the juxtamembrane
region. The Eph kinase catalytic activity is inhibited when the
two key regulatory tyrosines are not phosphorylated. Upon phos-
phorylation of these tyrosine residues, the Eph protein kinase do-
main gets fully activated (Fig. 2B) (Wybenga-Groot et al., 2001).
