Chemistry & Biology
A General Framework for Inhibitor
Resistance in Protein Kinases
Deborah Balzano,1,3Stefano Santaguida,1Andrea Musacchio,1,2,* and Fabrizio Villa1,*
1Department of Experimental Oncology, European Institute of Oncology, Via Adamello 16, 20139 Milan, Italy
2Max Planck Institute of Molecular Physiology, Otto-Hahn-Strasse 11, 44227 Dortmund, Germany
3Present address: Structural Biology and Biocomputing Programme, Spanish National Cancer Research Centre (CNIO), c/Melchor Fdez.
Almagro 3, 28029 Madrid, Spain
*Correspondence: firstname.lastname@example.org (A.M.), email@example.com (F.V.)
Protein kinases control virtually every aspect of
normal and pathological cell physiology and are
considered ideal targets for drug discovery. Most
kinase inhibitors target the ATP binding site and
interact with residue of a hinge loop connecting
the small and large lobes of the kinase scaffold.
Resistance to kinase inhibitors emerges during
clinical treatment or as a result of in vitro selection
approaches. Mutations conferring resistance to
ATP site inhibitors often affect residues that line the
inhibitor binding. Here, we show that mutations at
two specific positions in the hinge loop, distinct
from the previously characterized ‘‘gatekeeper,’’
have general adverse effects on inhibitor sensitivity
in six distantly related kinases, usually without con-
sequences on kinase activity. Our results uncover a
unifying mechanism of inhibitor resistance of protein
kinases that might have widespread significance
for drug target validation and clinical practice.
In eukaryotes, protein kinases are implicated in nearly any
signaling pathway under both physiological and abnormal
conditions (Cohen, 2002). In human cancer, deregulation of
protein kinase activity often correlates with disease progres-
sion. Accordingly, various members of the protein kinase family
are recognized as targets for anticancer therapy, and small-
molecule kinase inhibitors represent an important class of
anticancer agents (Bikker et al., 2009; Karaman et al., 2008;
Knight and Shokat, 2005; Zhang et al., 2009). The small-mole-
cule inhibitor imatinib (also known as Gleevec), for instance,
has revolutionized the treatment of chronic myeloid leukemia
(CML) (Capdeville et al., 2002). In CML patients, a chromosomal
translocation leads to the creation of the BCR-ABL gene.
c-Abl is a tyrosine kinase whose activity in normal cells is tightly
regulated. Within the Bcr-Abl fusion protein, Abl activity be-
comes constitutive (Suryanarayan et al., 1991; Wong and Witte,
Imatinib is a potent and selective active-site inhibitor of Bcr-
Abl. Its remarkable clinical efficacy, however, is counteracted
by the emergence of clinical resistance (Gorre et al., 2001). The
most common mechanism for imatinib resistance is the emer-
gence of mutations in the Abl kinase domain (Krishnamurty
and Maly, 2010). Over 50 different point mutations in the c-Abl
kinase domain have been detected in imatinib-resistant CML
patients (Melo and Chuah, 2007). Several mutations affect
residues in the vicinity of the ATP binding site, such as Y253
and E255 in the phosphate-binding loop (P loop) (Figure 1A).
Additional hot spots of mutations are in the hinge loop, which
connects the small and large lobes and provides a scaffold for
adenine binding. T315Abl, also known as the gatekeeper residue
(Branford et al., 2002), is a prominent hot spot of mutation in the
hinge loop (Figure 1A). Gatekeeper mutations disrupt the inhib-
ertiesof theenzyme, and therefore satisfy the tumorcells’ addic-
tion to Abl activity while rendering them resistant to imatinib
(Gorre et al., 2001). Remarkably, mutations of the gatekeeper
residues of the ALK tyrosine kinase confer resistance to the
small-molecule inhibitor crizotinib in EML4-ALK tumor cells
(Choi et al., 2010). Similarly, EGFR kinase mutations at the gate-
keeper confer resistance to erlotinib and gefitinib in lung tumors
(Kobayashi et al., 2005; Pao et al., 2005).
Imatinib is a so-called type II inhibitor. Type II inhibitors induce
a distinct conformation of the activation segment. Besides occu-
pying the ATP site, they exploit a hydrophobic pocket created by
this rearrangement (Schindler et al., 2000). Type I inhibitors, on
the other hand, bind exclusively in and around the ATP binding
site (Liu and Gray, 2006; Okram et al., 2006). Type II inhibitors
are generally sensitive to mutations in the gatekeeper residue,
whereas type I inhibitors are usually, although not universally,
less sensitive to gatekeeper mutations.
Indeed, type I inhibitors VX-680 and dasatinib target the gate-
keeper and other imatinib-resistant mutants of Abl (Carter et al.,
2005; Shah et al., 2004). Nevertheless, certain imatinib-resistant
hinge loop mutants of Abl, most prominently F317L, are also
resistant to dasatinib (Aguilera and Tsimberidou, 2009; Soverini
et al., 2006). Interestingly, an in vitro screen aiming to identify
itor, converged on Y156AurB, corresponding to F317Abl(Girdler
et al., 2008) (Figures 1A and 1B). Another position in the hinge
loop, G321Abl, is found mutated in cases of clinical resistance
to imatinib (Melo and Chuah, 2007). Mutations at the equivalent
position in Aurora B, G160AurB, also confer resistance to
966 Chemistry & Biology 18, 966–975, August 26, 2011 ª2011 Elsevier Ltd All rights reserved
ZM447439 and other type I inhibitors of Aurora B (Girdler et al.,
These previous results suggest the interesting possibility that
mutations affecting the hinge loop of protein kinases have the
potential to confer inhibitor resistance independently from the
specific active-site inhibitor used (i.e., whether type I or type II)
and independently of the specific chemical scaffold. To test
this idea systematically, we introduced individual amino acid
substitutions into the hinge region of six distant protein kinases
and determined the inhibitor sensitivity of the mutant kinases
against structurally distinct inhibitors (Table 1; see Figure S1
available online). Here, we report the results of our efforts.
A Data Set of Distant Protein Kinases
To generate a sufficiently diverse data set, we selected six
kinases belonging to different groups in the classification of the
eukaryotic protein kinases (ePK) fold (Manning et al., 2002).
The six kinases included (1) c-Abl [belonging to the tyrosine
kinase (Hung et al., 2007) subgroup]; (2) c-Src (also in the TK
group); (3) casein kinase 1 [CK1d, belonging to the casein kinase
7, Sterile 11, Sterile 20 (STE) group]; (5) phosphorylase kinase
[Phkg, belonging to the calcium/calmodulin-dependent protein
kinase (CAMK) group]; and Haspin, which does not belong to
Thus, the six kinases in our data set are phylogenetically distant.
We did not include kinases of the group containing the PKA,
PKG, and PKC families (AGC group) because of previous
evidence that Aurora B, a member of this group, becomes resis-
tant to several active-site inhibitors when mutated at the same
positions (Girdler et al., 2008). Additional criteria for their selec-
tion are (1) that recombinant, active versions of the kinases can
be produced relatively straightforwardly; (2) that the kinases
are structurally characterized (see Table 1); and (3) that one or
more public domain small-molecule inhibitors are available for
each of them (Table 1; Figure S1).
A sequence alignment of the hinge loop of the six kinases is
shown in Figure 1B. A light blue dot at the bottom of the align-
ment marks the ‘‘gatekeeper’’ residue. The red dots mark
Phe317 and Gly321 of c-Abl, and homologous residues in the
other five kinases in our data set (we will use Abl numbering to
refer to these positions unless otherwise indicated). The Kinase
Sequence Database (KSD; http://sequoia.ucsf.edu/ksd/) pro-
vides exhaustive sequence information on protein kinases. In
kinase sequence alignment from the KSD, the gatekeeper
residue is marked in red. The positions corresponding to
Phe317Abland Gly321Ablare found in most cases at positions +2
and +6 from the gatekeeper, respectively. In all six kinases in our
data set, there is a glycine residue at the position equivalent to
Figure 1. Structural Basis of Inhibitor Resistance
(A) Ribbon representation of a portion of the c-Abl kinase domain including
the small lobe and part of the helical array of the large lobe (pdb id: 2fo0). The
position of inhibitor-resistant mutants discussed in the text is mapped onto
(B) Multiple sequence alignment of the hinge region of kinases analyzed in this
study. The positions of hinge loop residues mutated in this study and of the
gatekeeper are highlighted with orange and blue spheres, respectively.
(C) Ribbon representation of the active site and hinge loop of Abl with bound
imatinib (blue, pdb id: 2fo0) and of Haspin with bound 5-iodotubercidin (red,
pdb id: 3iq7). Residues selected as candidates for this study are represented
as sticks. Molecular models were rendered with pymol (www.pymol.org).
See also Figure S1.
Table 1. Protein Kinases and Inhibitors Employed in This Study
Protein KinaseGroup Family PDBInhibitor
c-AblTK Abl 2fo0 VX-680
c-SrcTKSrc 1fmk VX-680
Haspin Other Haspin2wb8 5-Iodotubercidin
Pak5 STE STE202f57 Staurosporine
CAMKPHK 1phk Staurosporine
Chemistry & Biology
Chemistry & Biology 18, 966–975, August 26, 2011 ª2011 Elsevier Ltd All rights reserved 967
Gly321Abl(Figure 1B). As far as position Phe317Ablis concerned,
Pak5 and Haspin also have phenylalanine, Phkg and CK1d have
leucine, and c-Src has tyrosine (Figure 1B). We mutated these
two positions individually in each of the six kinases. The mode
are illustrated in Figure 1C.
Inhibitor Resistance from Mutating Phe317Abl
To determine the effects of mutations on kinase activity, we puri-
fied wild-type and mutant versions of our kinases from bacteria
and subjected them to in vitro kinase assays with suitable
significantly from kinase to kinase. For instance, mutations of
F317Ablto C, H, or Y (indicated as F317Abl-X, where X is the
amino acid substituent) left the catalytically activity essentially
unaltered. Conversely mutations of G321Ablto E, M, or V almost
completely abrogated catalytic activity (Figure 2A; summarized
in Figure S2).
Imatinib resistance of gatekeeper mutants of Abl (e.g.,
T315Abl-I) can be overcome by using VX-680, originally
described as an Aurora family inhibitor (Carter et al., 2005).
In line with this idea, we found that VX-680 inhibits the imati-
nib-insensitive mutant T315Abl-I to levels comparable to those
of wild-type Abl (Figures 2B–2D). Remarkably, the F317Abl-C
and F317Abl-H mutants were strongly resistant to VX-680, with
a 15-fold increase of the half-maximal inhibitory concentration
(Figure 2B).F317Abl-H was alsosignificantly resistant to imatinib,
albeit to a lower degree compared with T315I (Figure 2D).
We also tested our mutants against the second-generation Abl
inhibitors dasatinib and nilotinib. F317Abl-C and F317Abl-H
mutants were strongly resistant to dasatinib and nilotinib
(Figures 2E and 2F). These observations suggest that at least
one hinge loop mutant, F317Abl-H, confers resistance to at least
four unrelated inhibitors without grossly perturbing the catalytic
activity of the mutated kinase.
An Inhibitor-Resistant Form of c-Src
We next asked whether these findings could be extended to
other tyrosine kinases. The effects on inhibitor binding from
mutating the gatekeeper residue of Src have been characterized
(Blencke et al., 2004), but the effects from mutating Y340Srcand
Figure 2. c-Abl VX-680-Resistant Mutants
(A) Activity test on wild-type c-Abl and the indicated mutants.
(B–F) Wild-type Abl and the indicated Abl mutantswere tested inthe presence of increasing concentrations of(B and C) VX-680, (D) imatinib, (E) dasatinib, and (F)
nilotinib. Activity in the presence of DMSO solvent control (3% v/v) is also shown. Fractional residual activity (Fractional res. act.), from at least two independent
experiments, expressed as mean and standard deviation [arbitrary unit (AU)] was then quantified and plotted against the inhibitors concentration. CBB:
Coomassie brilliant blue, a P-Tyr: antibody recognizing phosphotyrosine residues. The condition marked as ‘‘ni’’ indicates absence of the inhibitor. Here and
elsewhere in this work, this condition is assigned fractional activity of 1.0 as expected in the absence of inhibitors.
See also Figure S2.
Chemistry & Biology
968 Chemistry & Biology 18, 966–975, August 26, 2011 ª2011 Elsevier Ltd All rights reserved
G344Src, equivalent to F317Ablor G321Abl, areunclear. Wethere-
fore mutated Y340Srcand G344Srcas indicated in Figure 3A. In
contrast with c-Abl, substitution at G344Srcdid not grossly affect
the catalytic activity. Substitution Y340Src-H or Y340Src-A
caused strong resistance toward VX-680 and PP2 (Hanke
et al., 1996), with an approximately 80-fold increase in the half
maximal inhibitory concentration (Figures 3B–3D). Importantly,
neither mutation significantly affected the KMfor ATP, which
was very similar to that of the wild-type kinase (Figure 3E).
cellular activity of c-Src kinase in presence of a specific inhibitor.
To activate c-Src tyrosine kinase, the carboxy-terminal Tyr527
residue, which negatively regulates c-Src upon C-terminal
c-Src kinase (CSK)-mediated phosphorylation (Cooper and
MacAuley, 1988), was mutated to phenylalanine. Mutants
Y527F, Y340A-Y527F, and Y340H-Y527F of Src were transiently
expressed in HeLa cells (Figure 3F) and c-Src kinase activity
associated with each kinase was then measured by immunoblot
analysis with an antiserum specifically recognizing c-Src-medi-
ated autophosphorylation on its tyrosine residue 416 (Smart
et al., 1981). VX-680 largely suppressed autophosphorylation
of the constitutively active Y527F Src mutant. In contrast, the
Src Y340H-Y527F double mutant revealed marked resistance
even when the cells were treated with 50 mM VX-680 (Figure 4B).
The Y340A-Y527F mutant showed milder levels of resistance
than expected based on the in vitro experiments.
Substitutions of Glycine Render CK1d, Phkg, and Pak5
Resistant to Their Respective Chemical Inhibitors
Mutation of residues at the position equivalent to G321Ablsignif-
icantly affected the kinase activity of CK1d and Pak5 (Figures 4A
and 4G). The mutations, however, resulted in marked resistance
to their cognate inhibitors (Figures 4B, 4C, 4H, and 4I). Specifi-
cally, the inhibitors D4476 and staurosporine (Rena et al.,
2004; Tamaoki et al., 1986) were largely inert toward the
G86CK1dand G529Pak5mutants. D4476 inhibited wild-type
CK1d with a half-maximal concentration near 3 mM (Figures 4B
and 4C), whereas the G86CK1d-D and G86CK1d-E mutants were
essentially not inhibited at concentrations of D4476 as high as
30 mM. Similar effects were observed with the F525Pak5and
G529Pak5mutants. For instance, the G529Pak5-D mutant was
largely insensitive to concentrations of staurosporine up to
Figure 3. Cellular Resistance Formation
of c-Src Tyrosine Kinase Mutants
(A) Activity test on wild-type c-Src and the indi-
(B–D) Wild-type Src and the indicated mutants
were tested in the presence of increasing
concentration of (B and C) VX-680 or (D) PP2. The
DMSO solvent control (3% v/v) is indicated.
Fractional residual activity (Fractional res. act.)
from at least two independent experiments ex-
pressed as mean and standard deviation [arbitrary
unit (AU)] was then quantified and plotted against
the inhibitors concentration.
(E) c-Src ADP formation in presence of different
concentration of ATP was monitored at different
time points in a luminescence-based assay (left
panel). The maximal velocity (Vmax) obtained was
plotted against different ATP concentrations to
obtain the KMvalue for the wild-type protein and
inhibitor-resistant mutants (right panel). Each data
point was collected in duplicate and kinetic
parameters were obtained using GraphPad Prism
(F) HeLa cells transiently expressing FLAG-Src-
Y527F, Src-Y340H-Y527F, and Src-Y340A-Y527F
were serum starved for 16 hr. Cells were treated
with the indicated VX-680 concentrations for
90 min prior to serum stimulation and subsequent
lysis. c-Src tyrosine kinase in total lysates was
analyzed in parallel by immunoblotting with anti-
pTyr416 Src family kinase-specific antibody, anti-
flag antibody for transfection efficiency, and
anti-vinculin loading control. Densitometric quan-
tification was performed with ImageJ and the
activity normalized against the transfection effi-
ciency and with the loading control. Quantification
of theresidual activityis indicated below eachlane
of pTyr416. CBB = Coomassie brilliant blue;
a P-Tyr = antibody recognizing phosphotyrosine
Chemistry & Biology
Chemistry & Biology 18, 966–975, August 26, 2011 ª2011 Elsevier Ltd All rights reserved 969
7 mM, and the G529Pak5-N mutant was only slightly less effective
and comparable to F525Pak5-L (Figures 4H and 4I).
of inhibitor resistance (Figures 4E and 4F). G110Phkg-N had the
most penetrant effects, being completely resistant to stauro-
sporine at all tested concentrations.
Analysis of the ‘‘Phe’’ mutants (i.e., in the position equivalent
to Phe317 of Abl, see Figure 1B) in Phkg and Pak5 (Figures 4E
and 4H) suggested the possibility that mutations at this position
might also have the potential to confer resistance to a generic
kinase inhibitor such as staurosporine. To assess the generality
of this idea, we extended the analysis to c-Abl and c-Src
(Figure S3). Indeed, the F317Abl-C and F317Abl-H of Abl were
strongly resistant to staurosporine, while Y340Src-A and
Y340Src-H were moderately resistant to it.
Validating 5-Iodotubercidin as a Haspin Inhibitor
Haspin is an atypical kinase (Eswaran et al., 2009; Villa et al.,
2009). Its kinase domain occupies the C-terminal part of the
molecule. The large N-terminal region has unknown function
(Higgins, 2003). It is a nuclear protein that associates with
Figure 4. Inhibitor Resistance in CK1d,
Phkg, and Pak5
(A) Activity test on wild-type human CK1d.
(B and C) Wild-type CK1d and the indicated
mutants were tested in the presence of increasing
concentration of D4476.
(D) Activity test on wild-type human CK1d.
(E and F) Wild-type Phkg and the indicated
mutants were tested in the presence of increasing
concentration of staurosporine.
(G) Activity test on wild-type human Pak5.
(HandI)Wild-type Pak5andtheindicated mutants
were tested in the presence of increasing
concentration of staurosporine. In all experiments,
the DMSO solvent control (3% v/v) is also indi-
cated. Fractional residual activity (Fractional res.
act.) from at least two independent experiment
expressed as mean and standard deviation [arbi-
trary unit (AU)] was then quantified and plotted
against the inhibitors concentration. CBB, Coo-
massie brilliant blue; pho-b,phosphorylase b;
See also Figure S3.
chromosomes to phosphorylate T3 of
histone H3 (P-T3-H3) (Dai et al., 2005).
This modification has been recently
implicated in centromere recruitment of
the Aurora B-containing chromosomal
passenger complex at the centromere
(Kelly et al., 2010; Wang et al., 2010;
Yamagishi et al., 2010).
5-Iodotubercidin was initially reported
to be a potent adenosine kinase inhibitor
(Parkinson and Geiger, 1996). Recently,
it was also identified as a potent in vitro
inhibitor of Haspin (Eswaran et al., 2009).
Because 5-iodotubercidin has not been yet fully characterized
as a Haspin inhibitor, we set out to validate its potency in vitro
and in vivo. In vitro, 5-iodotubercidin inhibited the isolated
kinase domain of Haspin with an IC50of 9 nM (Figure 5A). To
determine the selectivity of inhibition, we set up an in vitro kinase
assay with several human mitotic kinases, including Aurora A,
Aurora B, Bub1, Cdk1:Cyclin B, Mps1, Nek2A, and Plk1. At
1 mM, 5-iodotubercidin failed to alter the activity of any of these
kinases (Figure 5B).
We next sought to identify Haspin mutants that are insensitive
to 5-iodotubercidin. Substitutions G609Haspin-D, G609Haspin-S,
or G609Haspin-V, while retaining a substantial fraction of kinase
activity (Figure 5C), rendered Haspin resistant to high concentra-
tion of 5-iodotubercidin (Figures 5D and 5E). Conversely, substi-
tution at F607Haspindid not affect the susceptibility of Haspin to
inhibition by 5-iodotubercidin. Decreased activity of the
G609Haspin-D or G609Haspin-S mutants was not due to impaired
ATP binding, because both mutants had a KMfor ATP that was
similar to that of the wild-type kinase (Figure 5F). Thus, these
mutants are likely affected in their ability to recognize their
To validate the effects of 5-iodotubercidin in cells, we first
determined a working concentration of 5-iodotubercidin that
Chemistry & Biology
970 Chemistry & Biology 18, 966–975, August 26, 2011 ª2011 Elsevier Ltd All rights reserved
would inhibit Haspin activity, using the levels of P-T3-H3 as a
read-out. As little as 0.5 mM 5-iodotubercidin efficiently inhibited
histone H3 Thr3 phosporylation in HeLa cells, as assessed
by indirect immunofluorescence with a specific antibody (Fig-
ure S4). This effect is consistent with those observed upon
RNAi-mediated depletion of Haspin in mammalian cells (Dai
et al., 2005).
Next, we created a Venus-Haspin fusion protein to express
the wild-type and the G609Haspin-D inhibitor-resistant mutant in
HeLa cells (Nagai et al., 2002). After treatment with 5-iodotuber-
cidin, cells were fixed and subjected to immunodecoration with
antibodies recognizing P-T3-H3 (Figure 5G). At 5 mM 5-iodotu-
bercidin, the P-T3-H3 signal observed upon transfection of
the wild-type Venus-Haspin construct was largely abrogated.
On the other hand, at 5 mM 5-iodotubercidin, the P-T3-H3 signal
was largely preserved in cells expressing the G609Haspin-D
Inhibitor resistance in protein kinases is a scourge that limits the
long-term clinical efficacy of small-molecule inhibitors. On the
other hand, inhibitor resistance can be exploited as a tool for
target identification and validation, and to evaluate the mecha-
nism of action of new inhibitors. Here, we validate a general
strategy to generate an inhibitor-resistant mutant without signif-
icant prior knowledge of their structural organization and without
using cumbersome selection strategies. The strategy stems
from previous observations that mutations at two defined posi-
tions in the hinge loop of Abl, Aurora B, and other kinases can
of the generality of the effects of these mutations on inhibitor
resistance has been carried out. Therefore, we set out to test
the hypothesis that these positions can be targeted to develop
a general strategy for inducing resistance toward active site
inhibitors. We selected a diverse set of kinases and demon-
strated that indeed the effects from mutating the hinge loop are
very general and invariably result in inhibitor resistance. The
generality of our findings is further reinforced by previous obser-
vations. Mutations at G216 in Aurora A, equivalent to G160AurB,
confer resistance to VX-680 (Scutt et al., 2009). Mutation of the
structurally related G110 in p38a leads to marked resistance
toward quinazolinone and pyridol-pyrimidine inhibitors (Fitzger-
ald et al., 2003). Furthermore, mutation of G95 in Plk4 leads to
resistance toward VX-680 and MLN8054 (Sloane et al., 2010).
Due to their generality, we are inclined to believe that our results
might have significant clinical and technological applications.
A remarkable conclusion from our studies is that the catalytic
activity of mutants at the two active-site positions we have iden-
tified is usually preserved, or only moderately reduced, although
in some cases we also observed strong adverse effects on
The mechanisms of inhibitor resistance in protein kinases are
poorly understood. In many cases, resistance-conferring muta-
tions affect residues that are distant from the active site. In these
cases, the effects of mutation on resistance are usually difficult
or impossible to rationalize. A class of mutations that received
wide attention is that affecting the gatekeeper residue. In Abl,
this mutation affects the potency of imatinib as an Abl inhibitor,
but leaves the kinase susceptible to other inhibitors, such as
VX680. The selectivity of the effects of gatekeeper mutants is
due to the specific molecular structure of imatinib, a large inhib-
itor that penetrates deeply into the kinase’s active site making
extensive contacts with the gatekeeper. Larger gatekeeper resi-
dues create a problem of steric hindrance and prevent inhibitor
binding. Binding of smaller inhibitors, such as VX-680, is unaf-
fected, because the smaller inhibitors do not make contacts
with the gatekeeper, implying that this residue does not act as
discussed here, on the other hand, are implicated as selectivity
filters of essentially any active-site ATP-competitive inhibitor,
which is why their mutation invariably results in resistance,
regardless of the type of active-site inhibitor used. These results
predict that the previously described hinge loop mutants of Abl
will be resistant not only to imatinib, but to essentially any alter-
native ATP site inhibitors of the Abl kinase, providing a strong
argument toward alternative curative strategies for patients
that relapse with such mutations.
Protein kinases are considered ideal targets for chemical
inhibition. Several classes of small-molecule kinase inhibi-
tors are available, and several kinase inhibitors are being
tested in clinical trials, most often as anticancer inhibitors
(Knight and Shokat, 2005). Inhibitor resistance arises due
to mutations in the kinase scaffold, or, in case of multido-
main kinases, also in adjacent domains (Daub et al., 2004).
Here, we have shown that mutations at either of two specific
positions in the active site of protein kinases usually results
in inhibitor resistance with relatively mild effects on kinase
activity. We demonstrate that this strategy can lead to the
isolation of inhibitor-resistant mutants of a kinase of choice
without significant priorknowledge of the structural basis of
the kinase-inhibitor pair, except that the inhibitor is a nonal-
losteric, ATP-competitive inhibitor. We also show that muta-
tion of the ‘‘Phe’’ residue within the hinge loop usually
renders the affected kinases insensitive to the unspecific
kinase inhibitor staurosporine. This strategy might become
extremely useful for the validation of new inhibitors. For
instance, we validate Haspin as a target of 5-iodotubercidin,
a compound that was previously shown to bind to Haspin
(Eswaran et al., 2009), but whose ability to target Haspin in
living cells had not been characterized. Furthermore, the
strategy might be a useful aid in the characterization of
specific phenotypes. The approach discussed here is
complementary to a widely used approach based on the
substitution of the gatekeeper residue with smaller resi-
dues, which promotes the sensitization of the mutant kinase
to specific inhibitors (Bishop et al., 2001). Thus, the
approach discussed here has the potential to become
widely applicable in kinase target validation in vitro and in
live cells and organisms.
Cell Culture, Transfection, and Synchronization
HeLa cells were grown in DME (EuroClone) supplemented with 10% fetal
bovine serum (HyClone) and 2 mM L-glutamine. FuGENE 6 (Roche Applied
Chemistry & Biology
Chemistry & Biology 18, 966–975, August 26, 2011 ª2011 Elsevier Ltd All rights reserved 971
Figure 5. 5-Iodotubercidin Is a Potent Haspin Inhibitor
(A) A kinase assay on the kinase domain of human Haspin with the indicated concentrations of 5-iodotubecidin (5-ITu). The substrate is histone H3.
(B) The indicated recombinant, purified mitotic kinases were tested with the appropriated substrates for their sensitivity to 1 mM 5-ITu. Error bars are
mean ± SEM.
(C) The activity of wild-type Haspin and of selected mutants.
(D and E) The activity of wild-type and indicated Haspin mutants was tested in presence of increasing concentration of 5-ITu. The ethanol (EtOH) solvent control
(3% v/v) is indicated. Fractional residual activity (Fractional res. act.) from at least two independent experiment expressed as mean and standard deviation
[arbitrary unit (AU)] was then quantified and plotted against the inhibitors concentration.
(F) Haspin wild-type ADP formation in presence of different concentration of ATP was monitored at different time points in a luminescence assay (left panel).
The Vmaxobtained were then plotted against different ATP concentrations to obtain the KMvalue for the wild-type protein and inhibitor-resistant mutants
Chemistry & Biology
972 Chemistry & Biology 18, 966–975, August 26, 2011 ª2011 Elsevier Ltd All rights reserved
Sciences) was used for transfection of HeLa cells following the manufacturer’s
suggestions. Nocodazole (3.3 mM) and thymidine (2 mM) were obtained from
Sigma-Aldrich. MG132 (EMD) was used at 10 mM.
Antibodies for Immunoblotting, Kinase Substrates, and Inhibitors
The following antibodies were used for immunoblotting: mouse anti-phos-
phoTyr (working dilution 1:2000; Upstate), rabbit anti-phosphoTyr416 c-Src
(working dilution 1:2000; Cell Signaling), mouse anti-flag (working dilution
1:8000; Sigma), rabbit anti-vinculin (working dilution 1:10000; Sigma). Phos-
phorylase b and a-casein were from Sigma. Histone H3 was from Roche.
5-Iodotubercidin was from Cayman Chemical. PP2, D4476, and staurospor-
ine were from Calbiochem, VX-680 and imatinib from LC Laboratories, and
dasatinib and nilotinib from Santa Cruz Biotechnology, Inc.
A segment of a human cDNA encoding CK1d1-303was subcloned with
BamHI-XhoI sites of a pGEX-6P vector (GE Healthcare) as a C-terminal fusion
to GST, with an intervening PreScission protease site. Vectors carrying cDNA
encoding Phkg2-297, Pak5425-715, and Crk120-225were obtained respectively
from the University of Dundee (UK), University of Oxford (UK), and the IFOM
institute (IT). Expression plasmid carrying Haspin452-798was described previ-
ously (Villa et al., 2009). A segment of mouse cDNA encoding c-Abl229-511
and chicken c-Src251-533were subcloned with BamHI-HindIII sites in a
pGEX-6P vector. The cDNA of the tyrosin phosphatase from yersinia pestis
YopH was cloned into a pACYC-Duet (Novagen) with NcoI-XhoI sites. For
mammalian expression the full-length sequence of chicken c-Src1-533has
been subcloned with BamHI-XhoI sites into a pcDNA3.1 (Invitrogen) as
a C terminus fusion of a flag peptide. Haspin1-798has been subcloned with
EcoRV-NotI sites into a pcDNA5-FRT/TO (Invitrogen) as a C terminus fusion
of the fluorescent protein Venus (Nagai et al., 2002). Point mutants were
created with the QuikChange kit (Stratagene) according to the manufacturer’s
suggestions. All constructs were sequence verified.
Protein Expression for Kinetic Analysis
GST fusion vectors of Crk120-225, Haspin452-798, Pak5251-533, CK1d1-303, Phkg2-
297and indicated mutants were used to transform BL21 Escherichia coli strain.
All wild-type and mutated versions were expressed and purified identically.
Expression was induced with 0.25 mM IPTG at OD600 = 0.7–0.9 and pro-
tracted for 12–16 hr at 23?C. Bacterial cells were harvested by centrifugation
at 4000 3 g for 15 min in a Beckman JLA 8.1 rotor, and the pellets were resus-
EDTA, 5% glycerol, Roche Complete Protease Inhibitor Cocktail Tablets).
Cells were lysed by sonication, and lysates were cleared by centrifugation at
23,000 3 g for 45–60 min on a JA25-50 rotor. The supernatants were incu-
bated with 300 ml of GST Sepharose Fast Flow (Amersham Biosciences) per
liter of bacterial culture previously washed with PBS and equilibrated with lysis
equilibrated in cleavage buffer (50 mM Tris [pH 7.6], 300 mM NaCl, 5% glyc-
erol, 1 mM DTT, 1 mM EDTA). All the fusion proteins were subjected to GST
cleavage with the exception GST-Crk120-225that was eluted with 20 mM
reduced glutathione. For GST cleavage, 10 units of PreScission Protease
(Amersham Biosciences) per milligram of substrate was added, and the incu-
bation was protracted for 16 hr at 4?C. The purified kinases containing frac-
tions were collected and employed for subsequent enzymatic analysis. c-
Abl229-511and c-Src251-533were cotransformed with a plasmid encoding for
the Yersinia pestis YopH tyrosine phosphatase to limit the toxicity of the
kinases catalytic activity (Seeliger et al., 2005). GST purification of c-Abl and
c-Src was then performed as previously described.
Radioactive Kinase Assays
quenched with SDS loading buffer, and resolved on 14% SDS-PAGE.
Incorporationof32Pwasvisualizedby autoradiography. Densitometry analysis
was performed using ImageJ software (National Institutes of Health).
IC50valueswerecalculatedfrom log-dose response curvesusingPrism4soft-
ware (GraphPad Software, Inc.). Reactions for CK1d were carried out ina solu-
tion containing 20 mM HEPES [pH 7.5], 10 mM MgCl2, 0.1 mM EDTA. Final
substrate concentrations in the assay were 20 mM ATP (5 mCi g-32P-ATP),
5 mM of a-casein, and 5 nM kinase. Reactions for Phkg were carried out in
a solution containing 50 mM Tris [pH 8.6], 10 mM MgCl2, 0.04 mM CaCl2. Final
of phosphorylase b, and 50 nM kinase. Reactions for Haspin were carried out
inasolution containing 25mMTris [pH 7.5],10mM MgCl2,150mM NaCl.Final
substrate concentrations in the assay were 250 mM ATP (5 mCi g-32P-ATP),
5 mM of histone H3, and 5 nM kinase. Reactions for Pak5 were carried out in
a solution containing 25 mM Tris [pH 7.5], 10 mM MgCl2, 150 mM NaCl. Final
substrate concentrations in the assay were 50 mM ATP (5 mCi g-32P-ATP),
5 mM of histone H3, and 50 nM kinase. Inhibitors were tested for their ability
to inhibit candidate’s kinases activity in vitro. Indicated amounts of inhibitors
were added to the reaction and DMSO was used as control. To test 5-iodotu-
bercidin specificity, mitotic kinases were purified and assayed as described
elsewhere (Santaguida et al., 2010).
Kinase Assays Employing pTyr-Specific Antibodies
Reactions for c-Src and c-Abl kinases were carried out in a solution containing
100 mM Tris [pH 8.0] and 10 mM MgCl2. Final substrate concentrations in the
assay were 50 mM ATP, 10 mM of GST-Crk1, and 50 nM kinase. Reaction were
initiated by the addition of ATP and MgCl2and carried out at 30?C for 30 min
and finally terminated by adding SDS-page loading buffer. Inhibition assay
reactions were then separated on a 15% SDS-PAGE gel and transferred
onto a nitrocellulose membrane. The membrane was blocked in 50 mM
Tris-HCl [pH 7.5], 150 mM NaCl, 0.1% (v/v) Tween (TBS-Tween), and 5%
(w/v) BSA for 1 hr. The membrane was then incubated with the same buffer
for 16 hr at 4?C in the presence of monoclonal antibody directed against
epitope phospho-Tyr. Detection was performed using horseradish peroxi-
dase-conjugated secondary antibody and the enhanced chemiluminescence
(ECL Amersham Pharmacia Biotech) reagent.
ADP Luminescent Assay
Kinetic analyses of Haspin452-798and c-Src221-533and the indicated mutants
were performed using a luminometric kinase assay varying the concentration
of ATP using the ADP-Glo reagents (Promega). Haspin kinase (5 nM) was as-
sayed in a reaction (10 ml) containing 25 mM Tris (pH 7.6), 10 mM MgCl2,
150 mM NaCl, 1 mM EDTA, 1 mM DTT, varied concentrations of ATP, and
5 mM of histone H3 (Roche) and followed for 15 min. c-Src kinase (50 nM)
was assayed in a reaction (10 ml) containing 100 mM Tris (pH 8), 10 mM
MgCl2, varied concentrations of ATP, and 10 mM of GST-Crk1 and followed
for 15 min. The overall rate of reaction is determined as the slope of the
decreasing phase of the reaction. Each data point was collected in duplicate
and kinetic parameters were obtained using GraphPad Prism v3.0 software.
HeLa cells were transfected with indicated Venus-Haspin constructs using the
Fugene transfection reagent according to the manufacturer’s instructions.
The inhibitor, or the equivalent volume of ethanol as a control, was added
to the tissue culture medium for 90 min and the cells were fixed using 4%
PFA in PBS. The following antibodies were used: anticentromeric antibody
1:50; Cell signaling). Cy3- and Cy5-labeled and Alexa Fluor 488-labeled
secondary antibodies for immunofluorescence were purchased from Jackson
ImmunoResearch Laboratories, Inc. and Invitrogen, respectively. DNA was
stained with DAPI. The coverslips were mounted using Mowiol mounting
media. Cells were imaged using a confocal microscope (TCS SP2; Leica)
equipped with a 63 3 NA 1.4 objective lens using the LCS 3D software (Leica).
(G) Transiently transfected HeLa cells expressing Venus-Haspin wild-type and the 5-ITu-resistant mutant G609D were treated with the indicated 5-ITu
concentrations for 90 min. Immunofluorescence images demonstrated that expression of Venus-HaspinG609D conferred inhibitor resistance to the transfected
cells, which retain the phospho-Thr3 signal in the presence of the inhibitor. CBB,Coomassie brilliant blue;32p, autoradiography32P. Scale bar represents 5 mm.
See also Figure S4.
Chemistry & Biology
Chemistry & Biology 18, 966–975, August 26, 2011 ª2011 Elsevier Ltd All rights reserved 973
Images were imported in Photoshop CS3 (Adobe Systems, Inc.), and levels
Cell Stimulation, Inhibitor Treatment, and Cell Lysis
HeLa cells were transfected with indicated flag-SrcY527F plasmid using the
Fugene transfection reagent according to the manufacturer’s suggestions.
The cells transfected with indicated flag-SrcY527F constructs were deprived
was added to the tissue culture medium 90 min prior to stimulation. The cells
were then stimulated with serum for 30 min. HeLa cells were finally harvested
by trypsinization and lysed in 50 mM HEPES (pH 7.5), 150 mM NaCl, 0.5%
NP40, 1% glycerol, 5 mM EDTA, 10 mM Na3VO4, 50 mM NaF, 20 mM
Na4-pyrophosphate, and protease inhibitor cocktail (Calbiochem) for 20 min
on ice. Cell lysates were centrifuged for 15 min at 13,000 rpm at 4?C in an
Eppendorf microcentrifuge. Protein amounts were measured with protein
assay reagent (Bio-Rad Laboratories) as specified by the manufacturer, and
equivalent amounts of total protein of each cell lysate were analyzed by
westernblotting.Densitometryanalysiswasperformed usingImageJ software
(National Institutes of Health).
Supplemental Information includes four figures and can be found with this
article online at doi:10.1016/j.chembiol.2011.04.013.
We thank the members of the Musacchio laboratory for many helpful discus-
sions and D. R. Alessi, S. Knapp, E. Maspero, G. Scita, and Z. Y. Zhang for
sharing reagents. The work was funded by the Association for International
Cancer Research and the Italian Association for Cancer Research. S.S. is sup-
ported by a fellowship from the Italian Foundation for Cancer Research. F.V. is
a former EMBO long-term postdoctoral fellow.
Received: January 16, 2011
Revised: April 10, 2011
Accepted: April 26, 2011
Published: August 25, 2011
Aguilera, D.G., and Tsimberidou, A.M. (2009). Dasatinib in chronic myeloid
leukemia: a review. Ther. Clin. Risk Manag. 5, 281–289.
Bikker, J.A., Brooijmans, N., Wissner, A., and Mansour, T.S. (2009). Kinase
domain mutations in cancer: implications for small molecule drug design
strategies. J. Med. Chem. 52, 1493–1509.
Bishop, A.C., Buzko, O., and Shokat, K.M. (2001). Magic bullets for protein
kinases. Trends Cell Biol. 11, 167–172.
Blencke, S., Zech, B., Engkvist, O., Greff, Z., Orfi, L., Horvath, Z., Keri, G.,
Ullrich, A., and Daub, H. (2004). Characterization of a conserved structural
determinant controlling protein kinase sensitivity to selective inhibitors.
Chem. Biol. 11, 691–701.
Branford, S., Rudzki, Z., Walsh, S., Grigg, A., Arthur, C., Taylor, K., Herrmann,
R., Lynch, K.P., and Hughes, T.P. (2002). High frequency of point mutations
clustered within the adenosine triphosphate-binding region of BCR/ABL in
patients with chronic myeloid leukemia or Ph-positive acute lymphoblastic
leukemia who develop imatinib (STI571) resistance. Blood 99, 3472–3475.
Capdeville, R., Buchdunger, E., Zimmermann, J., and Matter, A. (2002). Glivec
(STI571, imatinib), a rationally developed, targeted anticancer drug. Nat. Rev.
Drug Discov. 1, 493–502.
Carter, T.A., Wodicka, L.M., Shah, N.P., Velasco, A.M., Fabian, M.A., Treiber,
D.K., Milanov, Z.V., Atteridge, C.E., Biggs, W.H., 3rd, Edeen, P.T., et al. (2005).
Inhibition of drug-resistant mutants of ABL, KIT, and EGF receptor kinases.
Proc. Natl. Acad. Sci. USA 102, 11011–11016.
Choi, Y.L., Soda, M., Yamashita, Y., Ueno, T., Takashima, J., Nakajima, T.,
Yatabe, Y., Takeuchi, K., Hamada, T., Haruta, H., et al. (2010). EML4-ALK
mutations in lung cancer that confer resistance to ALK inhibitors. N. Engl. J.
Med. 363, 1734–1739.
Cohen, P. (2002). Protein kinases: the major drug targets of the twenty-first
century? Nat. Rev. Drug Discov. 1, 309–315.
Cooper, J.A., and MacAuley, A. (1988). Potential positive and negative
autoregulation of p60c-src by intermolecular autophosphorylation. Proc.
Natl. Acad. Sci. USA 85, 4232–4236.
Dai, J., Sultan, S., Taylor, S.S., and Higgins, J.M. (2005). The kinase haspin is
required for mitotic histone H3 Thr 3 phosphorylation and normal metaphase
chromosome alignment. Genes Dev. 19, 472–488.
Daub, H., Specht, K., and Ullrich, A. (2004). Strategies to overcome resistance
to targeted protein kinase inhibitors. Nat. Rev. Drug Discov. 3, 1001–1010.
Eswaran, J., Patnaik, D., Filippakopoulos, P., Wang, F., Stein, R.L., Murray,
J.W., Higgins, J.M., and Knapp, S. (2009). Structure and functional character-
ization of the atypical human kinase haspin. Proc. Natl. Acad. Sci. USA 106,
Fitzgerald, C.E., Patel, S.B., Becker, J.W., Cameron, P.M., Zaller, D., Pikounis,
V.B., O’Keefe, S.J., and Scapin, G. (2003). Structural basis for p38alpha MAP
kinase quinazolinone and pyridol-pyrimidine inhibitor specificity. Nat. Struct.
Biol. 10, 764–769.
Girdler, F., Sessa, F., Patercoli, S., Villa, F., Musacchio, A., and Taylor, S.
(2008). Molecular basis of drug resistance in aurora kinases. Chem. Biol. 15,
Gorre, M.E., Mohammed, M., Ellwood, K., Hsu, N., Paquette, R., Rao, P.N.,
and Sawyers, C.L. (2001). Clinical resistance to STI-571 cancer therapy
caused by BCR-ABL gene mutation or amplification. Science 293, 876–880.
Hanke, J.H., Gardner, J.P., Dow, R.L., Changelian, P.S., Brissette, W.H.,
Weringer, E.J., Pollok, B.A., and Connelly, P.A. (1996). Discovery of a novel,
potent, and Src family-selective tyrosine kinase inhibitor. Study of Lck- and
FynT-dependent T cell activation. J. Biol. Chem. 271, 695–701.
Higgins, J.M. (2003). Structure, function and evolution of haspin and haspin-
related proteins, a distinctive group of eukaryotic protein kinases. Cell. Mol.
Life Sci. 60, 446–462.
Hung, C.H., Thomas, L., Ruby, C.E., Atkins, K.M., Morris, N.P., Knight, Z.A.,
Scholz, I., Barklis, E., Weinberg, A.D., Shokat, K.M., et al. (2007). HIV-1 Nef
assembles a Src family kinase-ZAP-70/Syk-PI3K cascade to downregulate
cell-surface MHC-I. Cell Host Microbe 1, 121–133.
Karaman, M.W., Herrgard, S., Treiber, D.K., Gallant, P., Atteridge, C.E.,
Campbell, B.T., Chan, K.W., Ciceri, P., Davis, M.I., Edeen, P.T., et al. (2008).
A quantitative analysis of kinase inhibitor selectivity. Nat. Biotechnol. 26,
Kelly, A.E., Ghenoiu, C., Xue, J.Z., Zierhut, C., Kimura, H., and Funabiki, H.
(2010). Survivin reads phosphorylated histone H3 threonine 3 to activate the
mitotic kinase Aurora B. Science 330, 235–239.
Knight, Z.A., and Shokat, K.M. (2005). Features of selective kinase inhibitors.
Chem. Biol. 12, 621–637.
Kobayashi, S., Boggon, T.J., Dayaram, T., Janne, P.A., Kocher, O., Meyerson,
M.,Johnson, B.E.,Eck, M.J., Tenen, D.G., and Halmos, B.(2005).EGFR muta-
tion and resistance of non-small-cell lung cancer to gefitinib. N. Engl. J. Med.
Krishnamurty, R., and Maly, D.J. (2010). Biochemical mechanisms of resis-
Liu, Y., and Gray, N.S. (2006). Rational design of inhibitors that bind to inactive
kinase conformations. Nat. Chem. Biol. 2, 358–364.
The protein kinase complement of the human genome. Science 298, 1912–
Melo, J.V., and Chuah, C. (2007). Resistance to imatinib mesylate in chronic
myeloid leukaemia. Cancer Lett. 249, 121–132.
Nagai, T., Ibata, K., Park, E.S., Kubota, M., Mikoshiba, K., and Miyawaki, A.
(2002). A variant of yellow fluorescent protein with fast and efficient maturation
for cell-biological applications. Nat. Biotechnol. 20, 87–90.
Chemistry & Biology
974 Chemistry & Biology 18, 966–975, August 26, 2011 ª2011 Elsevier Ltd All rights reserved
Okram, B., Nagle, A., Adrian, F.J., Lee, C., Ren, P., Wang, X., Sim, T., Xie, Y., Download full-text
Xia, G., Spraggon, G., et al. (2006). A general strategy for creating ‘‘inactive-
conformation’’ abl inhibitors. Chem. Biol. 13, 779–786.
Pao, W., Miller, V.A., Politi, K.A., Riely, G.J., Somwar, R., Zakowski, M.F., Kris,
M.G., and Varmus, H. (2005). Acquired resistance of lung adenocarcinomas to
gefitinib or erlotinib is associated with a second mutation in the EGFR kinase
domain. PLoS Med. 2, e73.
Parkinson, F.E., and Geiger, J.D. (1996). Effects of iodotubercidin on adeno-
sine kinase activity and nucleoside transport in DDT1 MF-2 smooth muscle
cells. J. Pharmacol. Exp. Ther. 277, 1397–1401.
Rena, G., Bain, J., Elliott, M., and Cohen, P. (2004). D4476, a cell-permeant
inhibitor of CK1, suppresses the site-specific phosphorylation and nuclear
exclusion of FOXO1a. EMBO Rep. 5, 60–65.
Dissecting the role of MPS1 in chromosome biorientation and the spindle
checkpoint through the small molecule inhibitor reversine. J. Cell Biol. 190,
Schindler, T., Bornmann, W., Pellicena, P., Miller, W.T., Clarkson, B., and
Kuriyan, J. (2000). Structural mechanism for STI-571 inhibition of abelson
tyrosine kinase. Science 289, 1938–1942.
Scutt, P.J., Chu, M.L., Sloane, D.A., Cherry, M., Bignell, C.R., Williams, D.H.,
and Eyers, P.A. (2009). Discovery and exploitation of inhibitor-resistant aurora
and polo kinase mutants for the analysis of mitotic networks. J. Biol. Chem.
Seeliger, M.A., Young, M., Henderson, M.N., Pellicena, P., King, D.S., Falick,
A.M., and Kuriyan, J. (2005). High yield bacterial expression of active c-Abl
and c-Src tyrosine kinases. Protein Sci. 14, 3135–3139.
Shah, N.P., Tran, C., Lee, F.Y., Chen, P., Norris, D., and Sawyers, C.L. (2004).
Overriding imatinib resistance with a novel ABL kinase inhibitor. Science 305,
Sloane, D.A., Trikic, M.Z., Chu, M.L., Lamers, M.B., Mason, C.S., Mueller, I.,
Savory, W.J., Williams, D.H., and Eyers, P.A. (2010). Drug-resistant aurora A
mutants for cellular target validation of the small molecule kinase inhibitors
MLN8054 and MLN8237. ACS Chem. Biol. 5, 563–576.
Smart, J.E., Oppermann, H., Czernilofsky, A.P., Purchio, A.F., Erikson, R.L.,
and Bishop, J.M. (1981). Characterization of sites for tyrosine phosphorylation
in the transforming protein of Rous sarcoma virus (pp60v-src) and its normal
cellular homologue (pp60c-src). Proc. Natl. Acad. Sci. USA 78, 6013–6017.
Soverini, S., Martinelli, G., Colarossi, S., Gnani, A., Castagnetti, F., Rosti, G.,
Bosi, C., Paolini, S., Rondoni, M., Piccaluga, P.P., et al. (2006). Presence or
the emergence of a F317L BCR-ABL mutation may be associated with resis-
tance to dasatinib in Philadelphia chromosome-positive leukemia. J. Clin.
Oncol. 24, e51–e52.
Suryanarayan, K., Hunger, S.P., Kohler, S., Carroll, A.J., Crist, W., Link, M.P.,
and Cleary, M.L. (1991). Consistent involvement of the bcr gene by 9;22
breakpoints in pediatric acute leukemias. Blood 77, 324–330.
Tamaoki, T., Nomoto, H., Takahashi, I., Kato, Y., Morimoto, M., and Tomita, F.
(1986). Staurosporine, a potent inhibitor of phospholipid/Ca++dependent
protein kinase. Biochem. Biophys. Res. Commun. 135, 397–402.
Villa, F., Capasso, P., Tortorici, M., Forneris, F., de Marco, A., Mattevi, A., and
Musacchio, A. (2009). Crystal structure of the catalytic domain of Haspin, an
atypical kinase implicated in chromatin organization. Proc. Natl. Acad. Sci.
USA 106, 20204–20209.
Wang, F., Dai, J., Daum, J.R., Niedzialkowska, E., Banerjee, B., Stukenberg,
P.T., Gorbsky, G.J., and Higgins, J.M. (2010). Histone H3, Thr-3.
Wong, S., and Witte, O.N. (2004). The BCR-ABL story: bench to bedside and
back. Annu. Rev. Immunol. 22, 247–306.
Yamagishi, Y., Honda, T., Tanno, Y., and Watanabe, Y. (2010). Two histone
marks establish the inner centromere and chromosome bi-orientation.
Science 330, 239–243.
Zhang, J., Yang, P.L., and Gray, N.S. (2009). Targeting cancer with small
molecule kinase inhibitors. Nat. Rev. Cancer 9, 28–39.
Chemistry & Biology
Chemistry & Biology 18, 966–975, August 26, 2011 ª2011 Elsevier Ltd All rights reserved 975