Conformational Studies and Atropisomerism Kinetics of the ALK Clinical Candidate Lorlatinib (PF-06463922) and Desmethyl Congeners

Article abstract of DOI:10.1002/anie.201509240

Lorlatinib (PF-06463922) is an ALK/ROS1 inhibitor and is in clinical trials for the treatment of ALK positive or ROS1 positive NSCLC (i.e. specific subsets of NSCLC). One of the laboratory objectives for this molecule indicated that it would be desirable

Full text of DOI:10.1002/anie.201509240

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201509240
German Edition: DOI: 10.1002/ange.201509240
Anti-Cancer Drugs
Conformational Studies and Atropisomerism Kinetics of the ALK
Clinical Candidate Lorlatinib (PF-06463922) and Desmethyl
Congeners
Jeff Elleraas, Jason Ewanicki, Ted W. Johnson, Neal W. Sach, Michael R. Collins,* and
Paul F. Richardson*
Abstract: Lorlatinib (PF-06463922) is an ALK/ROS1 inhib-
itor and is in clinical trials for the treatment of ALK positive or
ROS1 positive NSCLC (i.e. specific subsets of NSCLC). One
of the laboratory objectives for this molecule indicated that it
would be desirable to advance a molecule which was CNS
penetrant in order to treat brain metastases. From this
emphasize these differences.[2] Evaluation of the properties of
the individual enantiomers has long been positioned as
a critical step in the development process, which in turn has
led to the trend in which the vast majority of marketed
compounds are currently single enantiomers.[3] More recently,
the phenomenon of atropisomerism has been recognized as
becoming more prevalent in the discovery and development
of new molecular entities.[4] The hypothesis for this shift has
been attributed to the relationship between the design of
more compact and conformationally constrained inhibitors
with the availability of new effective synthetic methods for
the construction of the sterically hindered bonds contained
within these molecules.[5] However, atropisomerism brings an
additional layer of complexity to the development process.[6]
Unlike enantiomers, which are usually relatively stable, and
racemize via a bond-breaking and making process, atro-
pisomers interconvert through an intramolecular dynamic
process that in the majority of cases involves bond rotation.
Given that bond rotation is time-dependent, the racemization
half-lives for atropisomers vary from seconds to years
depending on numerous factors including steric hindrance,
temperature, electronics, solvent, etc. Furthermore, since the
biological properties of the molecule may differ between the
atropisomers,[7] it is critical to determine the racemization
half-life (preferably under physiological conditions) in order
to follow the best strategy for how to progress a molecule
(single atropisomer, or mixture) through a programꢀs devel-
opment cascade. Alternatively, a series of design strategies
have been proposed to modify a molecular structure to
a related analogue, while possessing the desired activity
against the biological target and mitigating the issue of
atropisomerism.[8] Herein, we will demonstrate one such
strategy in which the introduction of a stable stereogenic
center prevents the inter-conversion of atropisomers leading
to a single desired conformation being observed for the
molecule of interest.
perspective,
a macrocyclic template was attractive for
a number of reasons. In particular, this template reduces the
number of rotatable bonds, provides the potential to shield
polar surface area and reinforces binding through a restricted
conformation. All of these features led to better permeability
for the molecules of interest and thus increased the chance for
better blood brain barrier penetration. With a CNS penetrant
molecule, kinase selectivity is a key consideration particularly
with regard to proteins such as TrkB, which are believed to
influence cognitive function. Removal of the chiral benzylic
methyl substituent from lorlatinib was perceived as not only
a means to simplify synthetic complexity, but also as a strategy
to further truncate the molecule of interest. Examination of the
NMR of the desmethyl analogues revealed that the compound
existed as a mixture of atropisomers, which proved separable
by chiral SFC. The individual atropisomers were evaluated
through a series of in vitro assays, and shown to have
a favorable selectivity profile when compared to lorlatinib.
The challenge to develop such a molecule lies in the rate at
which the atropisomers interchange dictated by the energy
barrier required to do this. Here, we describe the synthesis of
the desmethyl macrocycles, conformational studies on the
atropisomers, and the kinetics of the interconversion. In
addition, the corresponding conformational studies on lorla-
tinib are reported providing a hypothesis for why a single
diastereomer is observed when the chiral benzylic methyl
group is introduced.
I
n drug discovery, it is well established that the separate
enantiomers of a chiral molecule can differ significantly in
properties such as biological activity, toxicity and metabo-
lism,[1] and several high profile examples have served to
In addition to the potential complication of developing an
atropisomeric compound, a further obstacle is the initial
detection of atropisomers, and there are cases in which
a molecule has been developed as a racemic mixture of
atropisomers, the existence of which were only later revealed
through chiral detection experiments.[9] With the growing
knowledge base, atropisomerism within a lead molecule can
be anticipated through the instinctive recognition of struc-
tural motifs, which lead to hindered axial rotation, or through
computational predictions of the likelihood of a molecule
presenting atropisomerism.[10] In the cases in which less
[*] J. Elleraas, J. Ewanicki, Dr. T. W. Johnson, N. W. Sach, M. R. Collins,
Dr. P. F. Richardson
Oncology Medicinal Chemistry, Pfizer, La Jolla
10770 Science Center Drive, San Diego, CA 92121 (USA)
E-mail: michael.collins@pfizer.com
paul.f.richardson@pfizer.com
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201509240.
3590
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 3590 –3595

Angewandte
Communications
Chemie
intuitive molecular fragments leading to conformational-
based chirality are present, numerous analytical tools (NMR,
X-ray, chiral HPLC) or even biological assays have been used
to determine the existence of atropisomerism.[7,11]
The evaluation of macrocyclic structures as potential drug
candidates is currently enjoying a significant renaissance.[12,13]
In particular, these structural motifs are attractive with the
perspective of being able to constrain the compound into the
binding conformation, and pick up additional protein–ligand
interactions. As part of our next generation anaplastic
lymphoma kinase (ALK) inhibitor project[14] for the treat-
ment of non-small cell lung cancer (NSCLC), we sought to
develop a small molecule inhibitor that combined brain
availability with broad-spectrum ALK potency for the treat-
ment of patients who exhibit tumor progression while on
crizotinib,[15] overcoming resistance mutations[16] and brain
metastases.[17] This initiative led to the design and discovery of
the macrocyclic inhibitor, lorlatinib (2.[18] In addition, the
macrocyclic framework enables the number of rotatable
bonds to be reduced leading to a more compact structure with
increased buried surface area compared to the corresponding
acyclic analogues, which we postulated would provide per-
meability advantages. In this initial disclosure, we highlighted
a matched-pair analysis between a pair of macrocyclic
inhibitors with and without the chiral methyl substituent at
the C-10 position (Figure 1). At that time, we also made the
qualitative observation regarding the desmethyl compound
existing as a pair of atropisomers, which can be separated by
chiral supercritical fluid chromatography (SFC), though inter-
conversion occurs on standing in solution over several weeks.
Herein, we describe the synthesis of the direct desmethyl
analogue of our lead compound, lorlatinib, and provide
a comparison of the biological activities of the pair of
atropisomers as well as an in-depth analysis of the racemiza-
tion kinetics, energetics,
Figure 1. 12-membered macrocyclic ALK inhibitors differing by methyl
substitution.
whereas nevirapine racemizes rapidly at room temperature
and as such is developed as a single entity.[21] As noted, the
strategy employed herein to progress the lead compound
involved introduction of a second chiral center into the
molecule, which brings about a favorable shift in the
conformational equilibrium.
Scheme 1 highlights the synthesis of the desmethyl macro-
cycle 1a/1b. The synthesis of the Boc-protected amino ester 6
utilized a Suzuki reaction where aryl bromide 4 was refluxed
in an aqueous methanolic suspension in the presence of
palladium acetate, cataCXium A, cesium fluoride and
diboron pinacol ester. The pyrazole bromide 5 was subse-
quently added to the crude reaction mixture and refluxed
overnight. Conversion of the Boc-protected amino methyl
ester 6 to the amino acid 7 was accomplished by hydrolysis of
the methyl ester followed by an acid mediated removal of the
Boc protecting group. Lactamization of the amino acid 7 to
the racemic atropisomer 1a/1b was accomplished by slowly
adding a solution of the amino acid 7 and DIEA in DMF to
a solution of HATU and HOBt in DMF.
and the mechanism.
The inter-conversion
of atropisomers within
macrocycles
and
medium-sized
rings
takes place not through
a single-bond rotation,
but instead a “ring flip”
occurs to transform one
enantiomer into the
other.[19] The rate of
this inter-conversion is
again dependent on
a
variety of factors
most notably steric hin-
drance and ring size.
Numerous examples of
atropisomerism exist in
medium-size rings and
macrocycles. For exam-
ple, telenzepine[20] rep-
resents a marketed com-
pound in which the atro-
Scheme 1. Synthesis of racemic atropisomer 1a/1b. TFA=trifluoroacetic acid, DCM=dichloromethane,
pisomers are stable, HATU=O-(7-azabenzotriazol-1-yl)-tetramethyluronium hexafluorophosphate, HOBt=1-hydroxybenzotriazole.
Angew. Chem. Int. Ed. 2016, 55, 3590 –3595
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3591

Angewandte
Communications
Chemie
Often the first sign of existence of atropisomers is
provided by the NMR, and this was the case for compound
1a/1b. 1H NMR reveals the presence of enantiotopic protons
as shown in Figure 2. For lorlatinib (2), the flipped conformers
are diasteromeric, and the presence of a single conformer is
confirmed by the presence of a single set of diastereotopic
protons.
expectations. All compounds showed low in vitro clearance in
human liver microsomes. In addition, the compounds were
highly permeable and showed low efflux ratios consistent with
Pgp substrate incompatibility and increased potential for CNS
exposure. Kinase and general selectivity were paramount
given the required high tissue distribution of these analogues.
TrkB was used as a selectivity surrogate based on its high ATP
binding site sequence homology to ALK and its putative role
in memory and cognition. In addition, achieving selectivity
over TrkB was deemed to be critical given the potential
adverse CNS-related effects due to TrkB inhibition.[23] The
increased selectivity ratios for 1a (831-fold) and 1b (244-fold)
relative to 2 (33-fold) were initially intriguing and prompted
further investigation.
Although the atropisomers 1a and 1b are enantiomeric
and energetically redundant (Figure 3, top panel), the tran-
sition barrier from one conformation to the other determines
racemization rates and impacts time-dependent analytical
methods. The 12-membered ring allows a small space for the
inversion, but several atoms must pass through the inside of
the ring in order to arrive at each distinct conformation. This
passage of atoms through a tight space gives rise to the higher
energy barrier and atropisomerism.
Figure 2. 1H NMR comparison of compound 1a/1b and 2
(PF-06463922).
The addition of a methyl group to form compound 2/2’
changes the conformational dynamics drastically. Figure 3
(bottom panel) highlights two distinct conformers of methyl
analogue 2/2’. The benzylic center in 2 minimizes A-1,3 strain
while 2’ maximizes A-1,3 strain.[24] Hypothetically, even if the
inversion barrier were low enough to allow access to both
conformers, the large energetic ground state difference
(+ 8.5 kcalmolÀ1) would drive the equilibrium completely to
the bioactive conformer 2. However, this is not the probable
mechanistic scenario which more likely involves the late
transition state during the cyclization step driving the product
to the desired structure 2 where it is unable to access the
inverted conformation due to high energetic penalties.[25]
One of the most common meth-
With this information in hand, we were able to investigate
the separation of compounds 1a and 1b using chiral SFC.[22]
This enabled the individual isomers to be evaluated sepa-
rately in a series of biological assays as described in Table 1.
Compounds 1a, 1b, 2 and 3 were submitted to an in vitro
panel assessing ALK-L1196M potency, CNS (central nervous
system) exposure, ADME, and selectivity (Table 1). Com-
pound 1a had comparable biochemical and cell potency to 2,
while 1b appears to be slightly less active. Although 1a and
1b were separated by chiral chromatography, it is likely that
significant racemization occurred prior to and/or during
testing, providing values more consistent with racemic
ods for determining activation ener-
gies involves determining the coa-
Table 1: Potency, ADME and selectivity of macrocyclic analogues.
Compound
ALK-L1196M pALK-L1196M LogD[a] HLM
MDR
TrkB Ki [nm]
lescence temperature of two NMR
resonances of the interconverting
species.[26] Variable temperature
NMR was carried out on Com-
pound 1a (see Supporting Informa-
tion). As can be seen, no coales-
cence was observed up to 1108C,
and as such an alternative method
was utilized for examining the acti-
vation energy through an Eyring
plot.[27]
[b]
Ki [nm]
cell IC50 [nm]
Clint,app
BA/AB (ratio)[c] (selectivity)[d]
1a
1b
1.0
2.0
0.70
30
131
21
1.9
1.9
2.3
<8
<8
<8
27.0/13.7 (2.0) 831 (831”)
33.0/13.4 (2.5) 488 (244”)
28.0/19.3 (1.5) 23 (33”)
2 (lorlatinib,
PF-06463922)
3
273
ND
2.2
<8
23.5/12.1 (1.9) ND
[a] LogD was measured at pH 7.4. [b] Clint,app refers to the total intrinsic clearance obtained from scaling
in vitro half-lives in human liver microsomes (HLM). [c] MDCK-MDR1 BA/AB efflux at 2 mm substrate
concentration and pH 7.4. [d] TrkB Ki and ratio relative to ALK-L1196M Ki.
mixtures. Indeed, modeling of 1b is completely inconsistent
with optimal binding and would be expected to provide
significant losses in potency relative to 1a, certainly more
than the modest decrease observed (Figure 3, top panel).
Compound 3, the enantiomer of 2, highlights the significance
of the methyl chirality, losing several hundred fold in
biochemical potency relative to the more active enantiomer
2. The desmethyl analogues were slightly less lipophilic (logD
1.9) than the methyl analogue (logD 2.3), in line with
We then envisioned that we would employ the segmented
flow technology for the measurement of the racemization
kinetics.[28] However, despite the advantages of this method-
ology with both the automation and the unprecedented
temperature control offered by the Accendo Conjure,[29] there
are drawbacks particularly given the introduction of a dis-
connect between the emergence/processing of segments from
the instrument and the actual chiral analysis. Although, for
molecules with a high barrier to inter-conversion (such as (S)-
3592
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 3590 –3595

Angewandte
Communications
Chemie
pre-determined intervals. In addition, to eliminate initial drift
after sample preparation, the iChem Explorer is capable of
dispensing the required volume of solvent (in this case IPA
(isopropyl alcohol)) to the solid material at the outset of the
experiment. From our range finding studies, we decided to
evaluate 25, 45 and 658C with automated sampling at
5 minute intervals. The in-line HPLC was set up with
a variation of the chiral method developed for SFC in order
to determine the ee at each time point, and the data able to be
evaluated in real time. Sampling was continued so that the
reactions were monitored for at least two to three half-lives to
ensure that the inter-conversion approximates to a first-order
reaction. From these experiments, the rate constant of
racemization (krac) can be determined at each of the temper-
atures evaluated by plotting ln (% ee) vs. t in which the slope
provides the rate. The half-life can then be determined from
the equation t1/2 = ln2/krac. Examples of the iChem data, and
the kinetics plot for the 658C run is provided in Figure 4 with
Table 2 providing the data for the three temperatures
evaluated.
Figure 3. Top panel: Enantiomeric low-energy conformations of 1a and
1b. Bottom panel: Low-energy bioactive conformation of methyl
analogue 2 with ALK-L1196M KD surface (PDB 4CLJ). Diastereomeric
ring-flipped, locally minimized conformation (2’) with additional strain
of 8.5 kcalmolÀ1 relative to locally minimized conformation 2. Com-
pounds were locally minimized and energetic differences calculated
using the OPLS2001 force field.
BINOL), the potential errors from this disconnect are
practically non-existent and this method is unique in offering
the ability to accelerate the kinetics through heating solvents
above the boiling point. For molecules though in which t1/2 is
in the range of hours, delays in sample analyses and back-
ground epimerization of stock solutions can introduce
a degree of error. The first step in carrying out our analyses
was to determine which measurements to determine our
kinetic analyses at, and how best to conduct these experi-
ments to minimize the potential errors.
Table 2: Measured first-order rate constants for the racemization of
compound 1a in IPA.
Entry
T [8C]
krac [minÀ1
]
R2
t1/2 [min]
1
2
3
25
45
65
0.0002
0.0096
0.1356
0.9674
0.9926
0.9966
3466
72.2
5.11
Our initial range finding experiments at 25, 80, 110 and
1408C confirmed this hypothesis, and indicated that we would
need an alternative method for collecting data on the kinetics
of the atropisomerization. Although, the 258C sample after
1 h showed 98% ee, all three of the other samples had
racemized completely. Switching to 25, 40 and 558C, and
evaluating after 24 h intervals indicated that the 558C sample
was racemic after 24 h. However, the 25 and 408C samples
showed ee values of 84.2% and 27.9% respectively. Allowing
these samples to stir at temperature for a further 24 hours
lead to values of 70% (258C) and 12.4% (408C), whereas
after 72 hours, the 408C sample had racemized whilst the
258C sample showed an ee of 34%. This gave us a temperature
range to work within as well as indicating that regular in-line
sampling would be optimal in order to obtain the most
accurate data.
Figure 5 shows the Eyring plot derived from the exper-
imental kinetic data. Using the three points, the Eyring
equation calculates DH° = 32.1 kcalmolÀ1 and DS° =
0.0243 kcalmolÀ1 K. These parameters provide a calculated
Gibbs energy of activation DG° to be 24.6 kcalmolÀ1 at
physiological temperature (378C), which corresponds to a t1/2
of approximately 6 hours in this medium.[31]
With this data in hand, it was also important to establish
the stability of the macrocyclic compound, lorlatinib (2) and
determine the chiral integrity. An analogous experiment was
set up on the iChem Explorer heating a sample of lorlatinib
(spiked with the enantiomer 3 to ensure both compounds are
detected in the chiral assay). After 24 hours at 658C in IPA,
no chiral erosion of the sample material was detected (see
Supporting Information).[32]
The iChem Explorer is ideally suited for measuring the
kinetics of the atropisomerization as this instrument adds
heating and stirring capabilities to an HPLC instrument
enabling chemistry to be carried out with automated in-line
sampling. The instrument is compatible with a range of
Agilent HPLC instruments, and enables heating up to 1508C
with stirring rates of up to 1200 rpm using a magnetic stirrer
bar placed in a vial.[30] Typically, a 2 mL vial is used to evaluate
the chemistry making this a relatively material-sparring
method for the evaluation of the kinetics. In contrast to the
segmented-flow method previously described, only one
aliquot of material needs to be prepared for each temperature
evaluated with the instrument capable of taking samples at
During our search for broad-spectrum, brain-penetrant
ALK inhibitors, we discovered, initially by H NMR, that
some 12-membered macrocycles displayed atropisomerism.
These analogues displayed excellent in vitro potency and
ADME properties, but were not progressed due to the
complications associated with atropisomerism. In addition,
a unique use of iChem Explorer with in-line sampling allowed
us to define half-lives and transition state barrier energies for
interconversion of the atropisomers. The chiral benzylic
methyl of lorlatinib rigidifies the 12-membered template
and does not display complicating conformational effects.
Calculations confirmed these observations based on ground-
state energies. Lorlatinib also shows excellent in vitro and
Angew. Chem. Int. Ed. 2016, 55, 3590 –3595
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3593

Angewandte
Communications
Chemie
for the macrocycles reported in this manuscript. We acknowl-
edge Hieu Lam and Dac Dinh for the enzyme-based assay
data, and Laura Lingardo and Sergei Timofeevski for the cell-
based data. In addition, we also acknowledge chemists at
Peakdale and Wuxi for carrying out the synthetic chemistry
reported herein.
Keywords: atropisomers · conformational analysis · EML4-ALK ·
macrocycles · racemization
How to cite: Angew. Chem. Int. Ed. 2016, 55, 3590–3595
Angew. Chem. 2016, 128, 3654–3659
[1] M. Eichelbaum, A. S. Gross, Adv. Drug Res. 1996, 28, 1 – 64.
[2] a) R. R. Shah, J. M. Midgley, S. K. Branch, Adverse Drug React.
Toxicol. Rev. 1998, 17, 145 – 190; b) G. Blaschke, H. P. Kraft, K.
Fickentscher, F. Kohler, Arzneim.-Forsch. 1979, 29, 1640 – 1642;
c) S. Fabro, R. L. Smith, R. T. Williams, Nature 1967, 215, 296 –
297; d) M. Reist, P. A. Carrupt, E. Francotte, B. Testa, Chem.
Res. Toxicol. 1998, 11, 1521 – 1528; e) T. Eriksson, S. Bjorkman,
B. Roth, A. Fyge, P. Hoglund, Chirality 1995, 7, 44 – 52; f) H.
Ashrafian, J. D. Horowitz, M. P. Frenneaux, Cardiovasc. Drug
Rev. 2007, 25, 76 – 97.
[3] a) I. Agranat, H. Caner, J. Caldwell, Nat. Rev. Drug Discovery
2002, 1, 753 – 768; b) M. Rouhi, Chem. Eng. News 2003, 81, 56 –
61; c) Department of Health & Human Services, Food and
DrugAdministration: FDAꢀs Policy Statement for the Develop-
ment of New Stereoisomeric Drugs, Washington, DC, 1992:
http://www.fda.gov/Drugs/
GuidanceComplianceRegulatoryInformation/Guidances/
ucm122883.htm; d) S. Miller, 18th International Symposium on
Chirality, Busan, Korea, June 26, 2006: http://www.fda.gov/
downloads/AboutFDA/CentersOffices/CDER/ucm103532.pdf.
[4] Atropisomers are enantiomers (or diastereomers) that result
from slow axial rotation that can, in principle, interconvert
thermally, for which the half-life of interconversion is ꢀ 1000 s or
longer, thus allowing analytical separation. M. Oki, Top. Stereo-
chem. 1983, 14, 1 – 81; Compounds 1a, 1b, 2 and 3 are examples
of bridged bi-(hetero)aryl-lactams. See G. Bringmann, T.
Gulder, T. A. M. Gulder, M. Breuning, Chem. Rev. 2011, 111,
563 – 639.
Figure 4. a) Representative iChem Explorer sampling data and b) first-
order kinetics plot for the racemization of compound 1a in IPA at
658C.
[5] R. Adams, H. C. Yuan, Chem. Rev. 1933, 12, 261 – 338.
[6] S. R. LaPlante, L. D. Fader, K. R. Fandrick, D. R. Fandrick, O.
Hucke, R. Kemper, S. P. F. Miller, P. J. Edwards, J. Med. Chem.
2011, 54, 7005 – 7022.
[7] A. Zask, J. Murphy, G. A. Ellestad, Chirality 2013, 25, 265 – 274.
[8] J. Clayden, W. J. Moran, P. J. Edwards, S. R. LaPlante, Angew.
Chem. Int. Ed. 2009, 48, 6398 – 6401; Angew. Chem. 2009, 121,
6516 – 6520.
[9] I. Alkorta, J. Elguero, C. Roussel, N. Vanthuyne, P. Piras, Adv.
Heterocycl. Chem. 2012, 105, 1 – 188.
[10] S. R. LaPlante, P. J. Edwards, L. D. Fader, A. Jakalian, O. Hucke,
ChemMedChem 2011, 6, 505 – 513.
[11] J. E. Smyth, N. M. Butler, P. A. Keller, Nat. Prod. Rep. 2015, 32,
1562 – 1583.
[12] Macrocycles in drug discovery have been defined as a ring
system consisting of 12 or more atoms.
Figure 5. Eyring plot for the atropisomerization of compound 1a in
IPA.
in vivo properties, and is currently in Phase I/II clinical
trials.[33]
[13] a) E. M. Driggers, S. P. Hale, J. Lee, N. K. Terrett, Nat. Rev. Drug
Discovery 2008, 7, 608 – 624; b) E. Marsault, M. L. Peterson, J.
Med. Chem. 2011, 54, 1961 – 2004; c) F. Giordanetto, J. Kihlberg,
J. Med. Chem. 2014, 57, 278 – 295; d) S. Vendeville, M. D.
Cummings, Annu. Rep. Med. Chem. 2013, 48, 371 – 386.
[14] a) H. Mano, Cancer Sci. 2008, 99, 2349 – 2355; b) A. T. Shaw, B.
Solomon, Clin. Cancer Res. 2011, 17, 2081 – 2086; c) M. Soda,
Y. L. Choi, M. Enomoto, S. Takada, Y. Yamashita, S. Ishikawa, S.
Fujiwara, H. Watanabe, K. Kurashina, H. Hatanaka, M. Bando,
Acknowledgements
We are grateful to Ben Burke for carrying out a further
computational assessment of the conformational preferences
3594
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 3590 –3595

Angewandte
Communications
Chemie
S. Ohno, Y. Ishikawa, H. Aburatani, T. Niki, Y. Sohara, Y.
Siesler, S. M. S. V. Wardell, N. Boechat, V. Dabbene, S. L.
Sugiyama, H. Mano, Nature 2007, 448, 561 – 566; d) K. Kinosh-
ita, N. Oikawa, T. Tsukuda, Annu. Rep. Med. Chem. 2012, 47,
281 – 293.
Cuffini, J. Mol. Struct. 2007, 828, 201 – 210.
[22] a) T. Q. Yan, F. Riley, L. Philippe, J. Davoren, L. Cox, C. Orozco,
B. Rai, M. Hardink, J. Chromatogr. A 2015, 1398, 108 – 120;
b) T. J. Ward, Anal. Chem. 2006, 78, 3947 – 3956.
[23] V. K. Gupta, Y. You, V. B. Gupta, A. Klistorner, S. L. Graham,
Int. J. Mol. Sci. 2013, 14, 10122 – 10142.
[24] R. W. Hoffmann, Chem. Rev. 1989, 89, 1841 – 1860.
[25] a) V. Martí-Centelles, M. D. Pandey, M. I. Burguete, S. V. Luis,
Chem. Rev. 2015, 115, 8736 – 8843; b) J. Blankenstein, J. Zhu,
Eur. J. Org. Chem. 2005, 1949 – 1964.
[15] a) J. J. Cui, M. Tran-Dube, H. Shen, M. Nambu, P. P. Kung, M.
Pairish, L. Jia, J. Meng, L. Funk, I. Botrous, M. McTigue, N.
Grodsky, K. Ryan, E. Padrique, G. Alton, S. Timofeevski, S.
Yamazaki, Q. Li, H. Zou, J. Christensen, B. Mroczkowski, S.
Bender, R. S. Kania, M. P. Edwards, J. Med. Chem. 2011, 54,
6342 – 6363; b) J. J. Cui, M. McTigue, R. Kania, M. Edwards,
Annu. Rep. Med. Chem. 2013, 48, 421 – 434.
[16] R. Katayama, T. M. Khan, C. Benes, E. Lifshits, H. Ebi, V. M.
Rivera, W. C. Shakespeare, A. J. Iafrate, J. A. Engelman, A. T.
Shaw, Proc. Natl. Acad. Sci. USA 2011, 108, 7535 – 7540.
[17] a) P. S. Steeg, K. A. Camphausen, Q. R. Smith, Nat. Rev. Cancer
2011, 11, 352 – 363; b) I. Zhang, N. G. Zaorsky, J. D. Palmer, R.
Mehra, B. Lu, Lancet Oncol. 2015, 16, e510 – e521.
[26] a) http://chemnmr.colorado.edu/manuals/DNMR_Calculations.
pdf; b) K. D. Zimmer, R. Shoemaker, R. R. Ruminski, Inorg.
Chim. Acta 2006, 359, 1474 – 1478.
[27] Alternatively, the Arrhenius equation can be used to calculate
the relationship between the rate constant (k) to the activation
energy (Ea) and the pre-exponential factor (A).
[18] a) T. W. Johnson, P. F. Richardson, S. Bailey, A. Brooun, B. J.
Burke, M. R. Collins, J. J. Cui, J. G. Deal, Y. L. Deng, D. Dinh,
L. D. Engstrom, M. He, J. Hoffman, R. L. Hoffman, Q. Huang,
R. S. Kania, J. C. Kath, H. Lam, J. L. Lam, P. T. Le, L. Lingardo,
W. Liu, M. McTigue, C. L. Palmer, N. W. Sach, T. Smeal, G. L.
Smith, A. E. Stewart, S. Timofeevski, H. Zhu, J. Zhu, H. Y. Zou,
M. P. Edwards, J. Med. Chem. 2014, 57, 4720 – 4744; b) H. Y.
Zou, Q. Li, L. D. Engstrom, M. West, V. Appleman, K. A. Wong,
M. McTigue, Y. L. Deng, W. Liu, A. Brooun, S. Timofeevski,
S. R. McDonnell, P. Jiang, M. D. Falk, P. B. Lappin, T. Affolter,
T. Nichols, W. Hu, J. Lam, T. W. Johnson, T. Smeal, A. Charest,
V. R. Fantin, Proc. Natl. Acad. Sci. USA 2015, 112, 3493 – 3498.
[28] J. E. Davoren, M. W. Bundesmann, Q. T. Yan, E. M. Collantes, S.
Mente, D. M. Nason, D. L. Gray, ACS Med. Chem. Lett. 2012, 3,
433 – 435.
[29] http://www.accendocorporation.com///file1.lookchem.com/html/freearticle/2022081100/68502446/Conjure2.html.
[30] http://www.ichemexplorer.com/.
[31] Note there can be significant differences between t1/2 depending
on the media in which this is determined. Researchers at BMS
have demonstrated this for BMS-207940 reporting a 30-fold
decrease in half-life in going from aqueous media to human
plasma. Y. S. Zhou, L. K. Tay, D. Hughes, S. Donahue, J. Clin.
Pharmacol. 2004, 44, 680 – 688.
[32] With the conformational stability of compound 2 established at
658C over a 24 hour period, it can be concluded that the barrier
ˇ
[19] See for example, a) E. Van den Berge, J. Pospísil, T. Trieu-Van,
L. Collard, R. Robiette, Eur. J. Org. Chem. 2011, 33, 6649 – 6655;
b) N. Z. Burns, I. N. Krylova, R. N. Hannoush, P. S. Baran, J. Am.
Chem. Soc. 2009, 131, 9172 – 9173.
to interconversion will significantly exceed 30 kcalmolÀ1
[33] http://clinicaltrials.gov/ct2/show/NCT01970865.
.
[20] P. Eveleigh, E. C. Hulme, C. Schudt, N. J. M. Birdsall, Mol.
Pharmacol. 1989, 35, 477 – 483.
Received: October 21, 2015
[21] a) O. Abrahao-Junior, P. G. B. D. Nascimento, S. E. Galembeck,
J. Comput. Chem. 2001, 22, 1817 – 1829; b) A. P. Ayala, H. W.
Revised: November 27, 2015
Published online: February 15, 2016
Angew. Chem. Int. Ed. 2016, 55, 3590 –3595
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3595