Pyrido[2,3-d]pyrimidines: Discovery and preliminary SAR of a novel series of DYRK1B and DYRK1A inhibitors

Article abstract of DOI:10.1016/j.bmcl.2013.10.055

DYRK1B is a kinase over-expressed in certain cancer cells (including colon, ovarian, pancreatic, etc.). Recent publications have demonstrated inhibition of DYRK1B could be an attractive target for cancer therapy. From a data-mining effort, the team has discovered analogues of pyrido[2,3-d]pyrimidines as potent enantio-selective inhibitors of DYRK1B. Cells treated with a tool compound from this series showed the same cellular effects as down regulation of DYRK1B with siRNA. Such effects are consistent with the proposed mechanism of action. Progress of the SAR study is presented.

Full text of DOI:10.1016/j.bmcl.2013.10.055

Bioorganic & Medicinal Chemistry Letters 23 (2013) 6610–6615
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Pyrido[2,3-d]pyrimidines: Discovery and preliminary SAR of a novel
series of DYRK1B and DYRK1A inhibitors
Kevin Anderson a, Yi Chen a, Zhi Chen a, Romyr Dominique a, Kelli Glenn d, Yang He c, Cheryl Janson c,
Kin-Chun Luk a, , Christine Lukacs c, Ann Polonskaia b, Qi Qiao a, Aruna Railkar e, Pamela Rossman a,
⇑
Hongmao Sun a, Qing Xiang c, Masha Vilenchik b, Peter Wovkulich a, Xiaolei Zhang c
a Discovery Chemistry, Hoffmann-La Roche Inc., pRED, Pharma Research & Early Development, 340 Kingsland Street, Nutley, NJ 07110, USA
b Discovery Oncology, Hoffmann-La Roche Inc., pRED, Pharma Research & Early Development, 340 Kingsland Street, Nutley, NJ 07110, USA
c Discovery Technology, Hoffmann-La Roche Inc., pRED, Pharma Research & Early Development, 340 Kingsland Street, Nutley, NJ 07110, USA
d Non-Clinical Safety, Hoffmann-La Roche Inc., pRED, Pharma Research & Early Development, 340 Kingsland Street, Nutley, NJ 07110, USA
e Pharmaceutical and Analytical R&D, Hoffmann-La Roche Inc., pRED, Pharma Research & Early Development, 340 Kingsland Street, Nutley, NJ 07110, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
DYRK1B is a kinase over-expressed in certain cancer cells (including colon, ovarian, pancreatic, etc.).
Recent publications have demonstrated inhibition of DYRK1B could be an attractive target for cancer
therapy. From a data-mining effort, the team has discovered analogues of pyrido[2,3-d]pyrimidines as
potent enantio-selective inhibitors of DYRK1B. Cells treated with a tool compound from this series
showed the same cellular effects as down regulation of DYRK1B with siRNA. Such effects are consistent
with the proposed mechanism of action. Progress of the SAR study is presented.
Received 17 September 2013
Revised 23 October 2013
Accepted 25 October 2013
Available online 1 November 2013
Keywords:
DYRK1B
Ó 2013 Elsevier Ltd. All rights reserved.
DYRK1A
Kinase-inhibitor
Anti-cancer
Pyrido[2,3-d]pyrimidine
DYRK1B/MIRK (Dual-specificity tyrosine-regulated kinase/
Minibrain-related kinase) is a member of a conserved family of
serine/threonine kinases which are activated by intra-molecular
tyrosine auto-phosphorylation. Amplification of the DYRK1B gene,
up-regulation of DYRK1B expression and/or constitutive activation
of this kinase has been observed in several different types of
cancer.1,2 Pharmacological inhibition of this kinase may enhance
the anti-tumor effect of chemotherapeutic drugs and therefore, is
an appropriate target for tumor suppression.1–11
Comparing the protein sequences of DYRK1B and 1A, there is only
one amino-acid difference in the ATP binding site.14 X-ray crystal-
lographic structure of DYRK1A indicated that the side chain of this
residue points away from the interaction with ATP.15 Thus it is ex-
pected to be difficult to identify selective inhibitors between
DYRK1B and 1A that bind in the ATP sites. We settled upon the task
of identifying compounds that inhibit DYRK1B that are also active
against DYRK1A.
We had included a DYRK1A kinase activity assay16 as a selectiv-
ity screen for our other kinase inhibitor projects. Examination of
these historical data identified 1 as a compound from our corporate
kinase inhibitor collection that showed inhibition of DYRK1A. In
our effort to explore structure activity relationship (SAR) of 1, we
looked at various ring system replacements. One such change
was replacing the quinolone ring system by pyrido[2,3-d]pyrimi-
dine. Conceptually this change reduces the lipophilicity of the mol-
ecule and also provides an easier synthesis to explore changes at
C2 of the pyrimidine ring (corresponding to the C7 position of
quinolone). It was gratifying to note that by making this change,
2 had improved potency in both the DYRK1B and 1A activity (see
Fig. 1).
DYRK1B knock-out mice are viable,12 suggesting that this
kinase is not essential for the survival of nontransformed cells. An-
other study has shown that while ovarian cancer cells are damaged
by DYRK1B kinase inhibition, normal cells are not.11 The effects of
DYRK1B inhibition in patients are unknown as there are no com-
pounds of this class reported in the open literature and there are
none in clinical development to our knowledge.
Our initial focus was to discover DYRK1B inhibitors for treat-
ment of cancer. Historically overexpression of DYRK1A is impli-
cated in Down syndrome. More recently DYRK1A has also been
implicated as a potential target for treating oncological diseases.13
Analogues of 2 can easily be synthesized by modifying known
⇑
Corresponding author.
E-mail address: kinluk23888@gmail.com (K.-C. Luk).
methods17–19 (Scheme 1). Synthesis of the DYRK1B inhibitors D
0960-894X/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2013.10.055

K. Anderson et al. / Bioorg. Med. Chem. Lett. 23 (2013) 6610–6615
6611
O
O
O
O
O
O
4
5
8
5
4
R
R
O
O
O
O
N
N
H
N
H
N
N
H
N
N
H
1'
1'
3'
3'
Oxone®
1
N
H
8
O
O
S
N
N
O
O
N
1
N
H
O
S
N
N
D4 H
D5 H
O
O
DYRK1B IC50 0.56 0.3 µM
DYRK1A IC50 0.32 0.2 µM
SW620 EC50 > 10 µM
DYRK1B IC50 0.13 0.09 µM
DYRK1A IC50 0.097 0.06 µM
SW620 EC50 3.3 0.6 µM
O
1
2
R
N
N
H
R'R"NH
R'
Figure 1. Structure of hit 1 and pyrido[2,3-d]pyrimidine 2.
N
N
N
O
D6 H
R"
Scheme 4. Modification at C2.
R4
N
O
O
R4
H
(MeO2C)2CH2
N
O
O
N
NaOMe, MeOH
was achieved by three general approaches as outlined in Schemes
1–4 The synthesis could start with base (e.g. sodium methoxide)
catalyzed condensation of an appropriately functionalized 4-ami-
no-5-formyl-pyrimidine A with dimethyl malonate in methanol
at reflux to give the substituted pyrido[2,3-d]pyrimidines 6-car-
boxylates B (Scheme 1). Compound B could be hydrolyzed (lithium
or sodium hydroxide) to give the corresponding acid C. These acids
could then be condensed with the appropriate amines to give the
desired products D.
R2
N
H
Reflux
R2
N
NH2
B
A
R4
O
O
N
OH
LiOH or NaOH
MeOH, H2O
R2
N
N
H
C
In cases where condensation of acid C with the desired amine
was difficult, the amine could be converted to mono-methyl malo-
namide E first (Scheme 2). The amide E could then be condensed
R4
O
R
RNH2, HATU
Et3N, DMF
N
N
H
R2
N
N
O
H
D
Table 1
Structural modification at C6
Scheme 1. Synthesis of pyrido[2,3-d]pyrimidines.
R
N
O
N
N
H
O
R
NH2
O
O
O
O
R
N
O
N
Cl
O
N
H
R
IC50
(
l
M) 1B
IC50
(
l
M) 1A
EC50
93
(lM) SW620
E
H
N
*
*
*
R4
H
R4
N
O
O
O
2
3
O
O
0.252
0.116
Piperidine
MeOH
R
R
O
N
N
H
+
H
Reflux
R2
N
NH2
R2
N
H
O
O
O
O
O
O
O
O
O
Cl
Cl
Cl
Cl
E
H
N
A
D
Scheme 2. Alternate synthesis of pyrido[2,3-d]pyrimidines.
0.134 0.09
0.097 0.06
3.3 0.6
O
R4
N
O
O
N
H
R
O
N
N
H
LiOH or NaOH
MeOH, H2O
4
5
>10
>10
NT
O
R2
N
H
D1
H
N
R4
O
*
R'R"NH, HATU
Et3N, DMF
R
OH
0.075 0.05
0.047 0.03
>10
O
N
N
H
O
R2
N
N
O
NH2
H
D2
H
N
*
R4
O
O
6
0.248
0.107
2.5
R
NR'R"
N
N
H
O
R2
N
N
O
O
N
H
H
D3
1B and 1A are DYRK1B and DYRK1A. NT: not tested. Assays were done in duplicates.
In cases where compound was tested multiple times, standard deviation is
included.
Scheme 3. Modification at C6.

6612
K. Anderson et al. / Bioorg. Med. Chem. Lett. 23 (2013) 6610–6615
Table 2
Structural modification at distal amide
Cl
O
R
N
N
H
O
N
N
H
O
R
IC50
2.97
(l
M) 1B
IC50
(lM) 1A
EC50
NT
(lM) SW620
H
N
*
7
8
0.78
O
H
N
*
0.048
0.18
>100
O
H
N
*
9
0.068 0.05
0.135
0.022 0.01
0.342
1.9 0.2
Cl
Cl
O
H
N
*
10
2.66
O
N
*
11
12
0.586
0.096
0.305
0.041
NT
2.4
Cl
S
O
S
H
N
*
O
H
N
*
13
14
0.04
>10
0.030
6.52
1.7
NT
O
N
*
N
N
*
N
H
N
15
16
0.009
0.008
0.83
O
H2N
H
N
*
*
0.013 0.006
0.009 0.003
0.66 0.3
O
O
NH2
H
N
17
0.021
0.015
NT
H2N
H
N
*
O
18
19
1.005
0.774
0.797
1.61
NT
NH2
H
N
*
>10
O
NH2

K. Anderson et al. / Bioorg. Med. Chem. Lett. 23 (2013) 6610–6615
6613
Table 2 (continued)
R
IC50
(
l
M) 1B
IC50
(
l
M) 1A
EC50
NT
(lM) SW620
H
N
*
*
*
20
0.615
0.301
O
O
O
NH2
NH2
NH2
H
N
21
22
23
0.017 0.003
1.15 0.34
0.011
0.009 0.002
1.03 0.09
0.012
0.45 0.10
H
N
>1
H
N
*
Cl
Cl
0.24
O
NH2
H
N
*
O
24
25
1.894
>10
1.229
4.01
>10
NT
N
H
N
*
O
H
N
*
Cl
26
0.233
0.156
1.97
O
N
1B and 1A are DYRK1B and DYRK1A. NT: not tested. Assays were done in duplicates. In cases where compound was tested multiple times, standard deviation is included.
Table 3
Structural modification at C2 and or C4
Cl
R2
N
O
O
H
N
N
N
H
O
R1
N
H
NH2
R1
R2
H
IC50
(
l
M) 1B
IC50
(
lM) 1A
EC50
0.18
(lM) SW620
N
N
N
27
<0.005
0.006
HN
*
*
NH
28
29
H
H
0.031
0.026
0.013
0.009
0.23
0.66
N
*
N
*
H2N
O
N
30
H
0.008
0.005
0.027
31
32
–CH3
–OCH3
H
CH3
0.018
0.101
0.037
0.052
0.89
3.13
1B and 1A are DYRK1B and DYRK1A. NT: not tested. Assays were done in duplicates. In cases where compound was tested multiple times, standard deviation is included.

6614
K. Anderson et al. / Bioorg. Med. Chem. Lett. 23 (2013) 6610–6615
with functionalized 4-amino-5-formyl-pyrimidine A to give the
desired product D.
An alternate route to compounds bearing a second amide (D3)
is shown in Scheme 3. These were prepared by hydrolysis with
methanolic hydroxide of the corresponding ester (D1) to the acid
D2 followed by amide formation to give D3.
Exploration of modification at C2 was achieved by starting with
the methylsulfanyl substituted analogue D4. These were oxidized
with Oxone to form the sulfonyl derivative D5. Heating D5 with
an appropriate amine provided the desired modification at C2.
The original hit 1 from a focused screen showed modest activity
The geminally disubstituted analog (24) and the 3,4-dihydro-
1H-isoquinoline-2-yl analog (25) were prepared in an effort to
reduce the flexibility of the amino-alkyl chain. This reduced the
potency against DYRK1B considerably. Also, methylation of the ter-
minal amine resulted in lower activity (26).
Table 3 summarized results around the exploration of substitu-
tion on C2 and C4 of the pyrimidine ring. X-ray structure indicated
that substitution at C2 should be pointing out towards solvent, and
thus should be able to accommodate a variety of groups. This was
used in an effort to increase aqueous solubility of this class of com-
pounds.20 All the compounds studied showed very good potency in
the biochemical and cellular assay. Compound 30 showed the best
potency with a cellular activity of SW620 EC50 of 27 nM and
DYRK1B IC50 of 8 nM. The methylation at C4 of the pyrimidine ring
(compound 32) resulted in a significant loss in the enzymatic po-
tency (compared with 16).
in inhibiting DYRK1B with an IC50 of 0.57 lM in the kinase enzy-
matic assay. Through hit-to-lead efforts, we identified the pyr-
ido[2,3-d]pyrimidine series which was further optimized towards
the discovery of potent inhibitors of DYRK1B. Tables 1–3 summa-
rize the preliminary SAR observed for this series.
Changing from the quinolone 1 to the analogous pyrido[2,3-
The DYRK1B program was supported by X-ray crystallography
using DYRK1A, which crystallizes more readily than DYRK1B, and
differs from DYRK1B in the ATP site by only one residue; Met240
is a leucine in 1B, but the side chain does not point towards the
ATP site. The co-crystal structure of DYRK1A with compound 3
(PDB code 4MQ1) bound showed that the pyridone moiety of the
core makes a pair of hydrogen bonds to the hinge of the kinase,
although the geometry of the ligand NH-Leu241 C@O hydrogen
d]pyrimidine
2 improved the potency against DYRK1B to
0.25 M. The replacement of the methyl group in 2 on the benzene
l
ring by different halogens (e.g. 3) suggested that a chlorine was a
good substituent which provided up to 30 fold increase in the cel-
lular potency against the human colon tumor cell line SW 620 with
an EC50 of 3.3 lM. Interestingly, reversing the amide functionality
as in 4 resulted in complete loss of activity. (see discussion of X-ray
crystallographic structure below). The methyl ester moiety of 3 is
unstable metabolically. Converting the ester moiety of 3 to an
amide showed minimal improvement on the DYRK1B potency
(compounds 5 and 6), however, based on the available X-ray crys-
tallographic structure of compound 3 complexed with DYRK1A
(Fig. 1) we were interested in probing more deeply this region of
the protein. (see X-ray structure comment below).
As shown in Table 2, preliminary SAR exploration (7–11) indi-
cated that the 3-chloro-benzyl benzamide moiety showed in-
creased potency relative to 3. Lengthening the carbon chain from
benzyl (9) to phenethyl (10) or introduction of N-methyl group
to the amide nitrogen (11) resulted in lower potency. It was also
observed that the phenyl group of benzyl amide moiety could be
replaced by heteroaryl rings such as 2-thiophenyl and 3-thio-
phenyl groups while maintaining potency (12 and 13). However,
replacing the amide functionality by a tetrazole ring (14) resulted
in a significant loss in potency, demonstrating the importance of
this amide group for efficient binding. Molecular modeling indi-
cated that there could be room to extend off the benzylic position
of 8. Different lengths of amino-alkyl chains (1–4 carbon) were
synthesized and tested (compounds 15–18). It was gratifying to
observe that the first compound (15) with sub-micromolar potency
in the cellular assay was obtained. The amino-alkyl chains contain-
ing one to three carbon atoms showed comparable potency
whereas the four-carbon amino-alkyl chain (18) had a significant
drop in potency. Compound 19 with the phenyl group missing
and saturation of the terminal phenyl ring to a cyclohexyl group
(20) resulted in reduction in potency, pointing to the importance
of the contribution of the aromatic group to binding.
The chiral resolution of racemic [3-(1,3-dioxo-1,3-dihydro-iso-
indol-2-yl)-3-phenyl-propyl]-carbamic acid tert-butyl ester using
super critical fluid chromatography (SFC) provided the intermedi-
ates needed for the synthesis of the two pure enantiomers of 16
by the process of Scheme 3. The R-enantiomer (21, stereo-chemical
assignment was based on observation that only R-enantiomer was
observed in crystal structure, see Fig. 2B) for the 2-carbon chain
was identified as the active compound showing very potent activ-
ity of 9 and 450 nM in the enzymatic and cellular assay respec-
tively. The corresponding S-enantiomer (22) was considerably
less active. Combining the amino ethyl chain at the benzylic posi-
tion and the 3-chloro phenyl group gave a very active compound
(23) in the cellular assay with a value of 240 nM.
Figure 2. (A) Upper, PDB code 4MQ1 X-ray crystallographic structure of compound
3 bound to DYRK1A; (B) Lower, PDB code 4MQ2 X-ray crystallographic structure of
compound 23 (racemic mixture used in crystallization, only R-isomer observed in
crystal) bound to DYRK1A.

K. Anderson et al. / Bioorg. Med. Chem. Lett. 23 (2013) 6610–6615
6615
Table 4
PK studies with compound 16
Bio-structure and molecular modeling guided optimization has re-
sulted in significant increase in potency in both enzyme and tumor
cell growth inhibition. Further optimization of potency and physi-
cal properties to allow for selection of compound(s) to demon-
strate in vivo efficacy in a relevant tumor model and proof of
concept are being planned.
Route
iv
po
Dose (mg/kg)
AUC (ng⁄h/mL)
CL (mL/min/kg)
T1/2 (h)
Vdss (mL/kg)
Bioavailability (%)
2.5
5350
8.2
7.7
1070
10
2550
ND
ND
ND
12
Acknowledgments
The authors would like to thank Mr. Gino Sasso for 1H NMR
spectral analysis and NOE studies, Mr. Vance Bell, Mr. Richard
Szypula, and Ms. Theresa Burchfield, for mass spectral analysis.
Table 5
PK studies with compounds 27 and 31 via iv dosing
Compound
27
31
References and notes
Dose (mg/kg)
AUC (ng⁄h/mL)
CL (mL/min/kg)
T1/2 (h)
2.5
2.5
7220
5.7
9.3
802
1. Friedman, E. Cancer 2010, 2, 1492.
2. Friedman, E. Int. J. Mol. Sci. 2013, 14, 5560.
3. Lee, K.; Deng, X.; Friedman, E. Cancer Res. 2000, 60, 3631.
4. Deng, X.; Ewton, D. Z.; Li, S.; Naqvi, A.; Mercer, S. E.; Landas, S.; Friedman, E.
Cancer Res. 2006, 66, 4149.
2610
15.8
8.5
Vdss (mL/kg)
1790
5. Jin, K.; Park, S.; Ewton, D. Z.; Friedman, E. Cancer Res. 2007, 6, 7247.
6. Yoshida, K. Biochem. Pharmacol. 2008, 76, 1389.
7. Deng, X.; Ewton, D. Z.; Friedman, E. Cancer Res. 2009, 69, 3317.
8. Yang, C.; Ji, D.; Weinstein, E. J.; Choy, E.; Hornicek, F. J.; Wood, K. B.; Liu, X.;
Mankin, H.; Duan, Z. Carcinogenesis 2010, 31, 552.
9. Gao, J.; Zhao, Y.; Lv, Y.; Chen, Y.; Wei, B.; Tian, J.; Yang, Z.; Kong, F.; Pang, J.; Liu,
J.; Shi, H. Cancer Cell Int. 2013, 13, 2.
10. Ewton, D. Z.; Hu, J.; Vilenchik, M.; Deng, X.; Luk, K.-C.; Polonskaia, A.; Hoffman,
A. F.; Zipf, K.; Boylan, J. F.; Friedman, E. A. Mol. Cancer Ther. 2011, 10, 2104.
11. Hu, J.; Deng, H.; Friedman, E. A. Int. J. Cancer 2013, 132, 2258.
12. Leder, S.; Czajkowska, H.; Maenz, B.; de Graaf, K.; Barthel, A.; Joost, H. G.;
Becker, W. Biochem. J. 2003, 372, 881.
bond is not ideal. Importantly, the chlorine substitution on the
phenyl makes a perfect 3.0 Å halogen bond to the backbone car-
bonyl of Glu239.21–23 It is also clear from the binding mode that
the reverse amide would not allow the ligand to take on the same
conformation in the ATP binding site, as there would be either a
carbonyl–carbonyl repulsion or the phenyl would not continue
into the pocket. The ester carbonyl oxygen picks up a hydrogen
bond from Lys188, helping to anchor the ligand.
13. Ionescu, A.; Dufrasne, F.; Gelbcke, M.; Jabin, I.; Kiss, R.; Lamoral-Theys, D. Mini-
Rev. Med. Chem. 2012, 12, 1315.
The co-crystal structure of DYRK1A with compound 23 (PDB
code 4MQ2) bound shows several important interactions that are
made when the ester is further substituted. The chloro-benzyl
group nestles into the hydrophobic pocket that is created by the
turn of the glycine-rich loop. When there is a group in this position,
the glycine-rich loop pushes away a bit, and Phe170 side chain
adopted a different rotamer, moving from a position turned into
the loop to one that is outside of the loop. When the aliphatic chain
is added, another hydrogen bond is created between the terminal
amine and the side chain of Asn292. The compound that was co-
crystallized for this structure was a racemic mixture, but only
the R-isomer was seen in the crystal.
In order to select compound(s) for in vivo studies, preliminary
pharmacokinetic (PK) studies with compound 16 were performed.
After single iv dose at 2.5 mg/kg (single bolus), or po dose at 10 mg/
kg (single gavage) in rats,24 reasonable exposure was observed (Ta-
ble 4). Compounds 27 and 31 also showed reasonable exposure
after iv dosing (Table 5),24 thus demonstrating the possibility that
compounds with usable exposure for in vivo studies can be found.
In summary, a novel series of pyrido[2,3-d]pyrimidines has
been identified as potent inhibitors of both DYRK1B and DYRK1A.
The use of 9 as a tool compound to study the effects of DYRK1B
inhibitors in biochemical studies has been published.10 This com-
pound was able to block tumor cells from undergoing reversible
arrest in a quiescent G0 state and enable some cells to exit quies-
cence. The effect was observed with DYRK1B expressing cells (such
as SW620) but not in cells that did not (such as HCT116).10
14. Leder, S.; Weber, Y.; Altafaj, X.; Estivill, X.; Joost, H. G.; Becker, W. Biochem.
Biophys. Res. Commun. 1999, 254, 474.
15. Roos, A. K.; Soundararajan, M.; Elkins, J. M.; Fedorov, O.; Eswaran, J.; Phillips, C.;
Pike, A. C. W.; Ugochukwu, E.; Muniz, J. R. C.; Burgess-Brown, N.; Von Delft, F.;
Arrowsmith, C. H.; Wikstrom, M.; Edwards, A.; Bountra, C.; Knapp, S. 2009,
Protein Data Bank entry 2WO6.
16. Caliper format DYRK1A assay was purchased from ProQinase GmbH,
Breisacher Str. 117, D-79106 Freiburg, Germany.
17. Perandones, F.; Soto, J. L. J. Heterocycl. Chem. 1998, 35, 413.
18. Experimental procedures for the synthesis, enzymatic inhibition assays and
cellular assays of compounds in this article are described in: Anderson, K.;
Chen, Y.; Chen, Z.; Luk, K.-C.; Rossman, P. L.; Sun, H.; Wovkulich, P. M. Patent
application US 20120184542 A1; Chem. Abstr. 2012, 157, 229681.
19. Compounds 2, 3, 8, and 14 were synthesized according to Scheme 1.
Compounds 5, 6, 7, 9 to 13, 15 to 26, 31, and 32 were synthesized according
to Scheme 3. Compounds 27 to 30 were synthesized according to Scheme 4.
20. Addition of the aminoalkyl group to improve aqueous solubility was
successful, for example, solubility of 2 was <1
mL, and 27 was 700 g/mL. It was only marginally helpful in the case of 31:
g/mL.
lg/mL while 16 was 590 lg/
l
2
l
21. Lu, Y.; Shi, T.; Wang, Y.; Yang, H.; Yan, X.; Luo, X.; Jiang, H.; Zhu, W. J. Med.
Chem. 2009, 52, 2855.
22. Fedorov, O.; Huber, K.; Eisenreich, A.; Filippakopoulos, P.; King, O.; Bullock, A.
N.; Szklarczyk, D.; Jensen, L. J.; Fabbro, D.; Trappe, J.; Rauch, U.; Bracher, F.;
Knapp, S. Chem. Biol. 2011, 18, 67.
23. Wilcken, R.; Zimmermann, M. O.; Lange, A.; Joerger, A. C.; Boeckler, F. M. J. Med.
Chem. 2013, 56, 1363.
24. IV formulations: 1.25 mg/mL of 16, 0.625 mg/mL of 27, 31, solutions in 2–10% v/
v co-solvents such as dimethylsulfoxide or dimethylacetamide, solubilizers
such as 10–20% v/v PEG400, 18–24% hydroxypropyl beta cyclodextrin in
aqueous medium, acetate buffer or water; PO formulation: 2 mg/mL of 16,
suspension in 2% w/v Klucel LF and 0.1% w/v Tween 80 in water.