Discovery of an l-alanine ester prodrug of the Hsp90 inhibitor, MPC-3100

Article abstract of DOI:10.1016/j.bmcl.2015.09.053

Various types of Hsp90 inhibitors have been and continue to undergo clinical investigation. One development candidate is the purine-based, synthetic Hsp90 inhibitor 1 (MPC-3100), which successfully completed a phase I clinical study. However, further clin

Full text of DOI:10.1016/j.bmcl.2015.09.053

Bioorganic & Medicinal Chemistry Letters xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Discovery of an L-alanine ester prodrug of the Hsp90 inhibitor,
MPC-3100
Se-Ho Kim a, ,y, Rajendra Tangallapally b,y, In Chul Kim m,y, Richard Trovato c,y, Daniel Parker n,y
,
⇑
J. Scott Patton d,y, Leslie Reeves e,y, Chad Bradford f,y, Daniel Wettstein g,y, Vijay Baichwal h,y
,
Damon Papac i,y, Ashok Bajji j,y, Robert Carlson k,y, Kraig M. Yager l, ,y
⇑
a Orthobond Corporation, 675 U.S. 1, North Brunswick, NJ 08902, USA
b St Jude Children’s Research Hospital, Department of Chemical Biology, 262 Danny Thomas Place, Memphis, TN 38105, USA
c Jacobs Engineering, 155 N, 400 W., Suite 550, Salt Lake City, UT 84103, USA
d Arup Laboratories, 500 Chipeta Way, Salt Lake City, UT 84108, USA
e Genysis Labs, 391 Orange St, Salt Lake City, UT 84104, USA
f Sera Prognostics, 2749 East Parleys Way, Suite 200, Salt Lake City, UT 84109, USA
g MediProPharma, 419 Wakara Way, #207B, Salt Lake City, UT 84108, USA
h Core Diagnostics, 2458 Embarcadero Way, Palo Alto, CA 94303, USA
i Navigen, 383 Colorow Dr, Salt Lake City, UT 84108, USA
j VioGen Biosciences, 1290 West, 2320 South, Suite E, Salt Lake City, UT 84119, USA
k Karyopharm Therapeutics, 85 Wells Ave., 2nd Floor, Newton, MA 02459, USA
l Myriad Genetic Laboratories, 320 Wakara Way, Salt Lake City, UT 84108, USA
m New Hope Baptist Church, 149 Highland Ave., East Lansing, MI 48823, USA
n AvidXchange, 6415 South, 3000 East, Suite 150, Salt Lake City, UT 84121, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Various types of Hsp90 inhibitors have been and continue to undergo clinical investigation. One develop-
ment candidate is the purine-based, synthetic Hsp90 inhibitor 1 (MPC-3100), which successfully com-
pleted a phase I clinical study. However, further clinical development of 1 was hindered by poor
solubility and consequent formulation issues and promoted development of a more water soluble pro-
drug. Towards this end, numerous pro-moieties were explored in vitro and in vivo. These studies resulted
Received 5 August 2015
Revised 18 September 2015
Accepted 22 September 2015
Available online xxxx
in identification of L-alanine ester mesylate, 2i (MPC-0767), which exhibited improved aqueous solubil-
Keywords:
Prodrugs
Hsp90 inhibitors
Salt screening
Clinical candidate
Chromatography-free synthesis
ity, adequate chemical stability, and rapid bioconversion without the need for solubilizing excipients.
Based on improved physical characteristics and favorable PK and PD profiles, 2i mesylate was selected
for further development. A convergent, scalable, chromatography-free synthesis for 2i mesylate was
developed to support further clinical evaluation.
Ó 2015 Elsevier Ltd. All rights reserved.
Small-molecule drugs targeting a single altered cell signaling
pathway have been successfully developed as cancer therapeutics.
However, many of these drugs suffer reduced clinical efficacy over
time due to emergence of drug resistance.1 One approach to over-
coming this limitation is to develop agents that inhibit multiple
signaling pathways by inhibiting a critical, shared upstream
effector. Heat shock protein 90 (Hsp90) is a molecular chaperone
that represents one of the promising targets for such a strategy2
due to its crucial function in cellular homeostasis (e.g., maturation,
folding, anti-aggregation, cellular trafficking, and stabilization of
signal transduction proteins). The value of Hsp90 as an anticancer
target was largely overlooked until the pioneering work of
Neckers3 which revealed Hsp90 as the molecular target responsi-
ble for the antitumor activity of the natural product geldanamycin.
In addition to geldanamycin, the natural product radicicol has pro-
vided further anticancer target validation, and together have paved
the way for development of numerous synthetic Hsp90 inhibitor
chemotypes. To date, approximately 17 Hsp90 inhibitors have
entered clinical development.4 These clinical-stage Hsp90 inhibi-
tors target the N-terminal, nucleotide-binding site of Hsp90 thus
blocking the ATP-binding and rate-limiting hydrolysis steps in
the ATPase cycle.
⇑
Corresponding authors. Tel.: +1 732 658 6235; fax: +1 732 745 7270 (S.-H.K.);
tel.: +1 801 209 2225; fax: +1 801 584 3640 (K.M.Y.).
E-mail addresses: seho.kimkim@gmail.com (S.-H. Kim), kraig.yager@gmail.com
(K.M. Yager).
y
Former employees of Myrexis Inc., 305 Chipeta Way, Salt Lake City, UT 84108,
USA.
http://dx.doi.org/10.1016/j.bmcl.2015.09.053
0960-894X/Ó 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Kim, S.-H.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.09.053

2
S.-H. Kim et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx
The tumor selectivity of Hsp90 inhibitors can be attributed to
O
O
the high dependence of cancer cells on Hsp90 function.5 Cancer
cells are genetically unstable and rapidly adapt to stressful envi-
ronments such as nutrient deprivation, hypoxia, and accumulation
of misfolded proteins. Under these conditions, Hsp90 is upregu-
lated and exists entirely in a multi-chaperone complex which pref-
erentially stabilizes client proteins associated with oncogenic
signaling pathways, and inhibition of Hsp90 can simultaneously
deplete multiple oncogenic client proteins resulting in proteotoxic
stress and cumulative antitumor effect. Thus, Hsp90 facilitates
oncogene addiction along with non-oncogene addiction for stabil-
ity and function of cancer cells. Taken together, targeting Hsp90
could be an effective strategy for treating cancer as a single agent
or in combination with other chemotherapy drugs.4,6
NH2
N
NH2
N
N
N
N
a,b,c
N
S
N
H
N
3
OMs
Boc
N
N
Boc
4
5
Previously, we reported the discovery of 1 (MPC-3100), a
purine-based synthetic Hsp90 inhibitor that was advanced into
clinical development based on favorable pharmacokinetic (PK),
pharmacodynamic (PD), and broad-spectrum antiproliferative
activities in animal models.7 In a phase I human clinical trial, 1
was well tolerated and exhibited PK/PD profiles similar to those
observed in preclinical studies.8 However, poor aqueous solubility
d,e
O
S
O
O
O
NH2
NH2
N
N
N
Br
N
N
(7.5 lg/mL in simulated intestinal fluid (SIF); pH 6.8) led to a
N
S
Br
N
dosage form with a high percentage of solubilizing excipient
(40% Captisol) and consequently a development-limiting patient
‘pill burden’. To facilitate development of oral dosage forms, a pro-
drug approach was explored with the goal of identifying deriva-
tives of 1 demonstrating higher aqueous solubility and rapid
bioconversion in the gastrointestinal tract. Herein, we describe
the design, synthesis, and preclinical evaluation of candidate pro-
f
N
N
N
OH
O
H
drugs that culminated in identification of L-alaninate mesylate salt
2i, (MPC-0767).
6
1
To support clinical development of 1, the scalable, six-step syn-
thesis illustrated in Scheme 1 was developed and the details have
been reported previously.7 Briefly, N-alkylation of adenine (3) with
readily available mesylate ester 4 followed by C8 bromination and
displacement with potassium benzo[1,3]dioxole-5-thiolate pro-
vided thioether 5 in 48% overall yield. Selective C50 bromination
Scheme 1. Pilot plant synthesis of 1. Reagents and conditions: (a) K2CO3, 4, DMF,
35 °C, 65 h, 76%; (b) Br2, NaOAc–AcOH, MeOH/THF, 20–25 °C, 1 h, 75%; (c) benzo
[1,3]dioxole-5-thiol, K2CO3, DMF, 100 °C, 2.5 h, 85%; (d) NBS, AcOH, 20–25 °C, 3 h,
77%; (e) TFA, CH2Cl2, 20–25 °C, 3 h, then NH4OH, 77%; (f)
1-hydroxybenzotriazole hydrate (HOBt), 1-(3-dimethylaminopropyl)-3-ethylcar-
(L)-lactic acid,
bondiimide hydrochloride (EDCI), NEt3, DMF/THF, 20–25 °C, 18 h, 74%.
of benzodioxole
5 was achieved under mild conditions and
preceded deprotection of the piperidine nitrogen with TFA and
conversion to free base 6 with NH4OH. EDCI mediated coupling
of 6 with (L)-lactic acid afforded 1 in 74% yield.
product formulation campaigns and significant additional cost
and protracted timelines. Systematic and rational strategies for
optimal salt selection have been highlighted in the literature.10
Six salt forms of 2i (HCl, HBr, sulfate, mesylate, fumarate, and
maleate) were prepared using an approximately 1:1 stoichiometric
ratio of drug and acid (Table 1). Hydrochloride, hydrobromide, and
sulfate salts were found to be more hygroscopic than other salt
forms as indicated by 8–10% weight gains upon 24 h exposure to
conditions of 84% relative humidity (RH) at ambient temperature.
In contrast, the phosphate, mesylate, maleate and fumarate salts
exhibited relatively low moisture absorption with their water con-
tent staying relatively low and uniform at 3.5%, 1.9%, 3.7%, 3.1% and
0.4% for the phosphate, mesylate, fumarate, maleate, and succinate
salts, respectively). The fumarate, maleate, and succinate salts,
considered to have an acceptable hygroscopicity, displayed a lower
melting point than parent 1 (203 °C, DSC), suggesting a less stable
drug-counter ion complex.10 These and the phosphate derivative
also resulted in unacceptably low isolated yields or relatively poor
purities. Therefore, the mesylate salt was selected for further
evaluation.
The lactamide hydroxyl group of 1 presented a desirable point
of attachment for solubilizing promoieties; straightforward syn-
thetic chemistry and a physically robust bond potentially subject
to hydrolysis by gastrointestinal esterases. To this end, the series
of candidate ester prodrugs, 2a–k, were synthesized by standard
methods (Scheme 2). Three criteria were used to evaluate prodrug
suitability: (1) solubility (>1 mg/mL in SIF, pH = 6.5–6.8), (2)
acceptable oral bioavailability (F % > 40), and (3) efficacy (>40%
tumor regression in N-87 human tumor xenograft models). Two
candidates that satisfied these criteria were alaninate and lysinate
esters 2i and 2k, respectively. The lysine derivative 2k, while pro-
viding exceptional solubility, could not be developed further due to
poor physical and handling characteristics imparted by both
hygroscopic and amorphous solid forms. Attempts to overcome
these limitations via salt form variants (e.g., monohydrochloride,
dihydrochloride, mesylate and maleate) proved ineffective.
Additionally, during a maximum tolerated dose (MTD) study, one
death was observed among the five mice (nu+/–) treated with
2kꢀ3HCl (286 mg/kg po; equivalent to 200 mg/kg 1). Conversely,
2iꢀ2HCl was well tolerated when administered at 252 mg/kg po
(equivalent to 200 mg/kg dose of 1). Consequently, variants of
alaninate 2i were the subject of further investigations.
To facilitate preclinical evaluation, a scalable, chromatography-
free synthesis of 2i mesylate was developed based on EDCI-medi-
ated coupling of depsipeptide 8 and intermediate piperidine, 6
(Scheme 3). Depsipeptide 8a was prepared from L-lactic acid (7)
First, we sought to find an optimal salt form of 2i. Selection of
an optimal salt form is a critical element of early stage drug devel-
opment; a poor choice can lead to challenging scale-up and drug
and Boc-L-alanine N-succinimidyl ester according to a literature
procedure.11 Ester 8a was obtained as pale yellow oil and was used
without purification in the EDCI-mediated coupling with 6.
Please cite this article in press as: Kim, S.-H.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.09.053

S.-H. Kim et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx
3
OH
OH
O
O
a
O
Boc
OH
O
N
H
O
NH2
N
O
N
N
N
8a
7
(oil-semi solid)
S
Br
b
a-k
OH
1
O
Boc
NH(cHex)2
O
N
H
N
O
O
8b
(white solid)
O
Promoiety
O
O
2a-k
NH2
N
Promoiety =
SO3Na
PO3NH4
N
S
N
N
2b
2a
Br
OH
OH
O
N
CO2Na
c,d,e
6
+
8b
CH3SO3H H2O
O
2c
2f
O
2e
2d
2g
O
N
NH2
NH2
NH2
O
OPO3H
O
NH2
O
O
O
O
O
2h
2i
Scheme 3. Reagents and conditions: (a) Boc-L-alanine N-succinimidyl ester, DMAP
NH2
NH2
NH2
(0.1 equiv), pyridine (3.0 equiv), THF, rt, 1 d; (b) dicyclohexylamine, THF, rt then
crystallization (aq MTBE), 65% (two steps); (c) EDCI, HOBt, DMF, rt, 8–12 h, 93%; (d)
4 M HCl, then NH4OH; (e) methanesulfonic acid, THF, rt, 2 h, 75% (two steps).
O
O
2i
2j
2k
Scheme 2. Design and synthesis of prodrugs of 1. Reagents and conditions: (a) 2a,
chlorosulfonic acid, pyridine, DMF, rt, then NaOEt, 26%; (b) 2b, (1) bis(2,2,2-
trichloroethyl)phosphoryl chloride, pyridine, rt, 31%, (2) Cu(OAc)2ꢀH2O, Zn dust,
AcOH, 100 °C, then acetylacetone, DMF, rt, 28%; (c) 2c, succinic anhydride, DMAP,
pyridine, rt, then NaOMe, 17%; (d) 2d, 3-morpholin-4-yl-propionic acid HCl,
1-ethyl-3-(3-dimethylaminopropyl)carbodiimide HCl (EDCI), 4-(N,N-dimethy-
lamino)pyridine (DMAP), DMF, rt, 45%; (e) (1) 2e, (4S)-2,2-dimethyl-1,3-diox-
olane-4-carboxylic acid, EDCI, DMAP, DMF, rt, 22%, (2) AcOH/H2O (4:1, v/v), 50 °C,
34%; (f) 2f, (1) 3-dimethoxyphosphoryloxybenzoic acid, EDCI, DMAP, DMF, rt, 74%,
(2) TMSBr, CH2Cl2, rt, 39%; (g) 2g, N-Boc-4-aminobutyric acid, EDCI, DMAP, DMF,
then 4 M HCl in dioxane, 47%; (h) 2h, Boc-glycine NHS ester, NEt3, DMF, 60 °C, then
4 M HCl in dioxane, 22%; (i) 2i, (2S)-2-([(tert-butoxy)carbonyl]amino)-propionic
acid, EDCI, DMAP, DMF, rt, then 4 M HCl in dioxane, 37%; (j) 2j, (2S)-2-([(tert-
butoxy)carbonyl]amino)-3-methylbutanoic acid, EDCI, DMAP, DMF, rt, then 4 M
HCl in dioxane, 42%; (k) 2k, (S)-2,6-bis-tert-butoxycarbonylamino-hexanoic acid,
EDCI, DMAP, DMF, rt, then 4 M HCl in dioxane, 72%.
provided 2i mesylate with excellent purity (>99% by HPLC) without
chromatography.9
Karl Fisher analysis of prodrug candidate 2i mesylate indicated
a monohydrate form based on water content measurement (1.98%)
and was supported by differential scanning calorimetry (DSC)
which showed a small endothermic peak at around 95 °C.9 In
accordance with a general trend, 2i mesylate had a higher melting
point than the parent drug 1 (231 °C vs 203 °C, respectively). The
monohydrate form was found to be very stable with essentially
no water absorption after 7 days at 40 °C and 82% RH, and no mea-
surable degradation as evidenced by HPLC analysis.
The kinetic solubility of 2i mesylate was significantly greater
(ꢁ52 fold) than the parent compound 1 (10.2 vs 536
lg/mL; pH
6.5, for 1 and 2i mesylate, respectively), while the apparent perme-
ability was dramatically reduced (420 ꢂ 10-6 vs 2.7 ꢂ 10ꢃ6 cm/sec
at pH 6.2, for 1 and 2i mesylate). The low permeability of 2i mesy-
late suggests that bioconversion to parent should occur within the
intestinal lumen (or apical side) and that systematic exposure to
the prodrug should be minimal. Indeed, plasma concentration of
2i was below the limit of detection at the earliest possible time
points post oral dosing.
Unfortunately, this streamlined approach led to impure (70–90%)
2i mesylate and the introduction of preparative HPLC into the syn-
thetic sequence. Ultimately, pure depsipeptide was required and
this was achieved via crystallization of the dicyclohexylamine salt
8b from aqueous methyl t-butyl ether (MTBE). EDCI-mediated cou-
pling of 6 and 8b proceeded in good yield and high purity (>95%).
Deprotection and mesylate formation were uneventful and
Table 1
Comparison of salt forms of 2i
Hydrochloride
Hydrobromide
Sulfate
Phosphate
Mesylate
Fumarate
Maleate
Succinate
Melting point (°C)
221.92
8.0
216.68
8.4
251.26
9.7
224.97
3.5
233.44
1.9
202.92
3.7
194.75
3.1
190.21
0.4
Weight change (%)a
a
Saturated KCl solution was used to maintain 84% relative humidity (RH) conditions at room temperature. Weight change (%) of salt forms was determined after storage at
25 °C/84% relative humidity for 1 day.
Please cite this article in press as: Kim, S.-H.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.09.053

4
S.-H. Kim et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx
Table 2
carboxymethylcellulose (CMC), were similar to that of parent 1
dosed in 40% Captisol (1 F % = 67; 200 mg/kg, and 2i mesylate
% = 56; 265 mg/kg, respectively). Prodrug 2i mesylate
Half-life of 2i mesylate in EDTA plasma (min)
Monkeya
300
Human
149
Phosphate buffer (pH 7.4)
145
F
Mouse
<2
(265 mg/kg; 200 mg/kg equivalent of 1) also demonstrated compa-
rable efficacy in an N-87 xenograft tumor model to that observed
for parent 1 (46% tumor size reduction at 200 mg/kg) with no evi-
dence of weight loss in the study animals. It is tempting to specu-
late that a super-saturated solution of parent 1, produced as a
result of 2i bioconversion, may be responsible for superior absorp-
tion. Alternatively, it could be the result of precipitation of an
amorphous parent drug which would be expected to a higher
dissolution rate and thus net absorption.13 Based on these data,
a
Monkey plasma was incubated at room temperature and sampling time was
extended to 240 min.
Table 3
Half-life of 2i mesylate in liver microsomes (min)
Mousea
<2
Mouseb
<2
Humana
2
Humanb
2
L-alaninate 2i mesylate was selected for further clinical evaluation.
In summary, numerous ester derivatives of the Hsp90 inhibitor
a
In the absence of NADPH.
In the presence of NADPH.
b
1 were designed and synthesized in an effort to address drug
formulation issues imparted by its poor aqueous solubility.
Evaluations of the potential prodrugs 2a–2k both in vitro and
Table 4
Half-life of 2i mesylate in simulated gastric and intestinal fluid (min)
in vivo led to the identification of L-alaninate 2i mesylate that
satisfied the criteria of aqueous solubility, chemical stability, bio-
convertibility, and adequate PK and PD properties. Based on these
characteristics, 2i mesylate was selected for further clinical
evaluation which was supported by an efficient, scalable, and
chromatography-free synthesis.
SGF SIF
pH 1.2 + pepsin
>240
pH 1.2
>240
pH 6.8
173
pH 6.8 + pancreatin
22
Acknowledgment
Table 5
PK profiles of 2i mesylate and 1
Authors thank to Dr. Raouf Hussain and his team for analytical
support of this work.
Compound
Route
Dose
Tmax
Cmax
AUC
F
(mg/kg)
(h)
(ng/mL)
(h * ng/mL)
%
1a
iv
2.5
10
200
200
265
—
0.25
1
0.3
1
—
2106
3214
113035
35094
94194
—
Supplementary data
1b
po
po
po
po
2802
24888
7366
21562
38
67
21
56
1c
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2015.09.
053.
1d
2i mesylated
a
b
c
Formulation: DMA/PEG300/EtOH/H2O (3:40:12:45, v/v/v/v).
Formulation: DMA/PEG300 (5:95, v/v).
Formulation: 40% Captisol.
References and notes
d
Formulation: 2% CMC; analyses for parent compound 1.
1. (a) Faivre, S.; Djelloul, S.; Raymond, E. Semin. Oncol. 2006, 33, 407; (b) Aggarwal,
B. B.; Sethi, G.; Baladandayuthapani, V.; Krishnan, S.; Shishodia, S. J. Cell.
Biochem. 2007, 102, 580; (c) Komarova, N. L.; Wodarz, D. Proc. Natl. Acad. Sci. U.
S.A. 2005, 102, 9714.
2. (a) Barrott, J. J.; Haystead, T. A. J. FEBS J. 2013, 280, 1381; (b) Travers, J.; Sharp,
S.; Workman, P. Drug Discovery Today 2012, 17, 242; (c) Workman, P.; Burrows,
F.; Neckers, L.; Rosen, N. Ann. N.Y. Acad. Sci. 2007, 1113, 202.
3. Neckers, L. Curr. Top. Med. Chem. 2006, 6, 1163.
4. (a) Bhat, R.; Tummalapalli, S. R.; Rotella, D. P. J. Med. Chem. 2014, 57, 8718; (b)
Neckers, L.; Workman, P. Clin. Cancer Res. 2012, 18, 64.
Candidate prodrug 2i mesylate was evaluated in various matri-
ces to examine its chemical stability and kinetics and mechanism
of bioconversion. It was rapidly converted to parent 1 in mouse
plasma and mouse and human liver microsomes, but was very
stable in monkey and human plasma, similar to non-enzymatic
hydrolysis in phosphate buffer (Tables 2 and 3). The facile hydrol-
ysis of 2i mesylate in mouse plasma versus primate plasma can be
attributed to high plasma esterase activity in mice.12 In SIF, 2i
mesylate was stable with and without added pepsin, indicating
that 2i mesylate would not likely be subject to non-enzymatic
conversion in the stomach. In contrast, 2i mesylate was rapidly
converted to 1 in pancreatin-rich SIF (Table 4) supportive of a pri-
marily enzyme-mediated cleavage process, and the assumption
that 2i is most likely bioconverted to parent 1 by enzymes present
in the intestinal lumen prior to absorption.
5. (a) Chiosis, G.; Neckers, L. ACS Chem. Biol. 2006, 1, 279; (b) Kamal, A.; Thao, L.;
Sensintaffar, J.; Zhang, L.; Boehm, M. F.; Fritz, L. C.; Burrows, F. J. Nature 2003,
425, 407.
6. Biamonte, M. A.; Van de Water, R.; Amdt, J. W.; Scannevin, R. H.; Perret, D.; Lee,
W.-C. J. Med. Chem. 2010, 53, 3.
7. Kim, S.-H.; Bajji, A.; Tangallapally, R.; Markovitz, B.; Trovato, R.; Shenderovich,
M.; Baichwal, V.; Bartel, P.; Cimbora, D.; McKinnon, R.; Robinson, R.; Papac, D.;
Wettstein, D.; Carlson, R.; Yager, K. M. J. Med. Chem. 2012, 55, 7480.
8. Samlowski, W. E.; Papadopoulos, K.; Olszanski, A. J.; Zavitz, K.; Cimbora, D. M.;
Shawbell, S.; Balch, A.; Mather, G.; Beelen, A. Mol. Cancer Ther. 2011, 10, A96.
9. See Supporting information.
10. Bastin, R. J.; Bowker, M. J.; Slater, B. J. Org. Process Res. Dev. 2000, 4, 427.
11. (a) Katakai, R.; Kobayashi, K.; Yamada, K.; Oku, H.; Emori, N. Biopolymers 2004,
73, 641; (b) Gallo, E. A.; Gellman, S. H. J. Am. Chem. Soc. 1993, 115, 9774.
12. Bahar, F. G.; Ohura, K.; Ogihara, T.; Imai, T. J. Pharm. Sci. 2012, 101, 3979.
13. Guarino, V. R. In Prodrugs and Targeted Delivery; Mannhold, R., Kubinyi, H.,
Folkers, G., Eds.; Wiley-VCH: Weinheim, 2011; Vol. 47, pp 111–121.
The oral PK properties of 2i were assessed in mice as described
in Table 5. In these studies, the plasma concentration of parent 1
was quantified by LC–ESI–MS/MS and despite concerns over drug
absorption, the PK profiles of prodrug 2i formulated in 2%
Please cite this article in press as: Kim, S.-H.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.09.053