Hit-to-lead evaluation of a novel class of sphingosine 1-phosphate lyase inhibitors

Article abstract of DOI:10.1016/j.bmcl.2016.03.043

Inhibition of sphingosine-1-phosphate lyase has recently been proposed as a potential treatment option for inflammatory disorders such as multiple sclerosis, rheumatoid arthritis, and inflammatory bowel disease. In this report we describe our hit-to-lead evaluation of the isoxazolecarboxamide 6, a high-throughput screening hit (in vitro IC₅₀ = 1.0 μM, cell IC₅₀ = 1.8 μM), as a novel S1P lyase inhibitor. We were able to establish basic structure-activity relationships around 6 and succeeded in obtaining X-ray structural information which enabled structure-based design. With the discovery of 28, enzyme activity was quickly improved to IC₅₀ = 120 nM and cell potency to IC₅₀ = 230 nM. The main liability in the established isoxazolecarboxamide hit series was determined to be metabolic stability. In particular we identified that future lead-optimization efforts to overcome this problem should focus on blocking the N-dealkylation on the secondary amine.

Full text of DOI:10.1016/j.bmcl.2016.03.043

Accepted Manuscript
Hit-To-Lead Evaluation of a Novel Class of Sphingosine 1-Phosphate Lyase
Inhibitors
Jurgen Dinges, Christopher M. Harris, Grier A. Wallace, Maria A. Argiriadi,
Kara L. Queeney, Denise C. Perron, Eric Dominguez, Tegest Kebede, Kelly E.
Desino, Hetal Patel, Anil Vasudevan
PII:
S0960-894X(16)30263-3
DOI:
http://dx.doi.org/10.1016/j.bmcl.2016.03.043
BMCL 23686
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
13 January 2016
11 March 2016
12 March 2016
Please cite this article as: Dinges, J., Harris, C.M., Wallace, G.A., Argiriadi, M.A., Queeney, K.L., Perron, D.C.,
Dominguez, E., Kebede, T., Desino, K.E., Patel, H., Vasudevan, A., Hit-To-Lead Evaluation of a Novel Class of
Sphingosine 1-Phosphate Lyase Inhibitors, Bioorganic & Medicinal Chemistry Letters (2016), doi: http://dx.doi.org/
10.1016/j.bmcl.2016.03.043
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Graphical Abstract
To create your abstract, type over the instructions in the template box below.
Fonts or abstract dimensions should not be changed or altered.
Hit-To-Lead Evaluation of a Novel Class of Sphingosine
1-Phosphate Lyase Inhibitors
Jurgen Dinges, Christopher M. Harris, Grier A. Wallace, Maria A. Argiriadi, Kara L. Queeney, Denise C.
Perron, Eric Dominguez, Tegest Kebede, Kelly E. Desino, Hetal Patel, and Anil Vasudevan

Bioorganic & Medicinal Chemistry Letters
Hit-To-Lead Evaluation of a Novel Class of Sphingosine 1-Phosphate Lyase
Inhibitors
Jurgen Dingesa*, Christopher M. Harrisb, Grier A. Wallacec, Maria A. Argiriadic, Kara L. Queeneyb, Denise
C. Perronb, Eric Dominguezb, Tegest Kebedeb, Kelly E. Desinoa, Hetal Patela, and Anil Vasudevana
a
AbbVie Inc., 1 North Waukegan Road, IL 60064, U.S.A.
AbbVie Bioresearch Center, 100 Research Drive, Worcester, MA 01605, U.S.A.
b
c AbbVie Bioresearch Center, 381 Plantation Street, Worcester, MA 01605, U.S.A.
ARTICLE INFO
ABSTRACT
Article history:
Received
Revised
Accepted
Available online
Inhibition of sphingosine-1-phosphate lyase has recently been proposed as a potential treatment
option for inflammatory disorders such as multiple sclerosis, rheumatoid arthritis, and
inflammatory bowel disease. In this report we describe our hit-to-lead evaluation of the
isoxazolecarboxamide 6, a high-throughput screening hit (in vitro IC50 = 1.0 μM, cell IC50 = 1.8
μM), as a novel S1P lyase inhibitor. We were able to establish basic structure-activity
relationships around 6 and succeeded in obtaining X-ray structural information which enabled
structure-based design. With the discovery of 28, enzyme activity was quickly improved to IC50
= 120 nM and cell potency to IC50 = 230 nM. The main liability in the established
isoxazolecarboxamide hit series was determined to be metabolic stability. In particular we
identified that future lead-optimization efforts to overcome this problem should focus on
blocking the N-dealkylation on the secondary amine.
Keywords:
Sphingosine 1-phosphate lyase
Hit-to-Lead
Inhibitor
Parallel synthesis
Structure-based design
2009 Elsevier Ltd. All rights reserved.
Egress of lymphoyctes from secondary lymphoid organs is
regulated by a gradient of sphingosine-1-phosphate (S1P) (1,
Figure 1).1 S1P lyase catalyzes the irreversible catabolism of S1P
(1) and is critical to maintaining that S1P gradient. Genetic
deletion and chemical inhibition of S1P lyase elevates S1P levels
in lymph nodes, resulting in arrest of lymphocyte trafficking and
immunosuppression.2,3 This blockade of lymphocyte function
could be efficacious in T lymphocyte-mediated disorders such as
multiple sclerosis, rheumatoid arthritis, and inflammatory bowel
disease. The first reported functional S1P lyase inhibitors, 4’-
deoxypyridoxine
(DOP)
(2),
2-acetyl-4-(1R,2S,3R,4-
tetrahydroxybutyl)-1H-imidazole (THI) (3) and related
compounds2,4 block the conversion of Vitamin B6 into the active
pyridoxal-4-phosphate (PLP) (4) cofactor required for S1P lyase
activity.5,6 However, they may have limited clinical utility due to
either reversal of inhibition from dietary B6 or from co-inhibition
of other PLP-dependent enzymes. Benzylphthalazine 5 was
recently reported as the first direct S1P lyase inhibitor that was
efficacious in an animal model.7 The same class of phthalazines
was previously investigated as Smoothened antagonists for solid
tumors8, but it has not been reported if they are proceeding
through clinical development. An additional class of
benzimidazole sulfonamide S1P lyase inhibitors has also been
published.9
In this report, we describe our hit-to-lead efforts around the
active site-directed isoxazolecarboxamide S1P lyase inhibitor 6
(IC50 = 1 μM), which was identified from high-throughput
screening. Our main objectives were first to quickly establish the
basic structure-activity relationship (SAR) around 6 and to
demonstrate the viability of the generated hit series for further
lead optimization. The screening hit 6 is very amenable to the
Figure 1. Structures of sphingosine-1-phosphate (S1P) (1), 4’-deoxy-
pyridoxine (DOP) (2), 2-acetyl-4-(1R,2S,3R,4-tetrahydroxybutyl)-1H-
imidazole (THI) (3), the pyridoxal-4-phosphate (PLP) (4) cofactor, and
benzylphthalazine 5.

production of hit-explosion libraries, and we also aimed to
provide co-crystal structures of selected inhibitors bound to S1P
lyase to enable structure-based design. Our second objective was
to optimize cell potency to IC50 <500 nM, and our third objective
was to identify the main liabilities in this novel hit series and to
provide a starting point on how those could be resolved.
For the synthesis of 6, 2-bromoacetophenone (7) was treated
with hexamethylenetetramine and hydrochloric acid (Delepine
reaction)10 to introduce a primary amino group in 85% yield over
two steps (Scheme 1). The intermediate was then protected as a t-
butyl carbamate. Conversion of the keto group in 8 into a primary
amine was afforded via the formation of a ketoxime, followed by
hydrogenation. The protected diamine 9 was acylated with 5-
methylisoxazole-3-carboxylic acid and the Boc protecting group
was removed under acidic conditions. After reductive amination
Scheme 2. Reagents and conditions: a) Boc2O, Et3N, CH2Cl2, 12 h, 93%;
b) isoindoline-1,3-dione, DEAD, PPh3, THF, 12 h; c) H2NNH2 ∙ H2O, EtOH,
90oC, 2 h, 49% over 2 steps; d) 2,5-dimethyl-4-methoxy-benzaldehyde,
NaBH3CN, MeOH, 12 h, 43%; e) 2 N HCl, EtOAc, 4 h, 91%.
of 10 with 2,5-dimethyl-4-methoxy-benzaldehyde,
6 was
obtained as a pale yellow oil. The isoxazolecarboxamide 6
contains a chiral center. The enantiomers were separated using
chiral chromatography and the absolute configuration of the two
enantiomers was determined by X-ray crystallography. The S-
enantiomer (IC50 = 485 nM) is 2-fold more potent than the
racemate, while the R-enantiomer is inactive.
back-filled to establish systematic structure-activity relationships
(for example ortho, meta and para substitution of a phenyl ring
with the same substituent). The library sizes were between 50
and 100 members. In the first library, 5- and 6-membered
aromatic and aliphatic heterocyclic acids and small aliphatic
acids with polar functionalities were utilized for the modification
of the heterocyclic amide portion of 6 (R1). Table 1 shows that
potency was very specific for the isoxazole-3-carboxamide and
just switching the heteroatoms (14) already led to a 4-fold loss in
activity. The 5-methyl substituent was also important. The des
methyl analog 15 only had an IC50 of about 6.5 μM. Larger
and/or more polar substituents always resulted in a substantial
loss in potency, with 16 giving the best result. Only the small and
hydrophobic cyclopropyl analog 17 was able to improve activity
over the parent compound. The next library aimed to probe
substitutions on the phenyl ring (R2) in the center of 6 and to
replace the phenyl group with 5- and 6-membered aromatic
heterocycles. Only compounds with 6-membered aryl groups
were active. The representative triade of methoxyphenyl analogs
in Table 1 (18-20) establishes that para-substitution is favored.
The most potent members of this library were the 4-fluoro phenyl
compound 21 which displayed an IC50 of 0.72 μM and the 4-
pyridyl analog 22 with an IC50 of 0.85 μM. In a third library, we
investigated the substitution pattern on the benzyl group (R3) and
also replaced it with benzylic 5- and 6-membered aromatic and
aliphatic heterocycles, as well as small aliphatic cyclic and
acyclic systems. It turned out that the 2,5-dimethyl-4-methoxy-
benzyl moiety was already optimized and removal of just one or
both methyl groups (represented by 23) resulted in a complete
loss of activity. Later, the X-ray structure of 6 bound to SPL
showed that the two methyl groups orient the benzyl group in the
active site.
Scheme 1. Reagents and conditions: a) hexamethylenetetramine, NaI,
EtOH, 24 h; b) 6N HCl, EtOH, reflux, 5 h, 85% over 2 steps; c) Boc2O,
NaHCO3, MeOH, 12 h, 75%; d) NH2OH ∙ HCl, pyridine, EtOH, 3 h, 66%; e)
H2 (40 psi), 10% Pd/C, MeOH, 5 h, 90%; f) 5-methylisoxazole-3-carboxylic
acid, HATU, DIEA, CH2Cl2, 12 h, 68%; g) 2N HCl, MeOH, 12 h, 97%; h)
2,5-dimethyl-4-methoxy-benzaldehyde, NaBH3CN, MeOH, 12 h, 49%.
The described synthesis pathway was used for the production
of hit-explosion libraries in which the benzyl (Table 1: R3) and
phenyl moieties (R2) were varied. For the diversification of the
amide portion R1, the synthesis was revised as shown in Scheme
2. The starting 2-amino-2-phenylethanol (11) was Boc-protected.
Mitsunobu reaction with isoindoline-1,3-dione followed by
hydrazinolysis provided the diamine 12 in moderate yield.
Reductive amination with 2,5-dimethyl-4-methoxy-benzaldehyde
and Boc-deprotection gave the desired core 13.
Access to the X-ray crystal structure of 6 bound to S1P lyase in
the presence of the PLP (4) cofactor,19 allowed us to utilize
structure-based design for further optimization efforts. Based on
Figure 2 it could be speculated that the 3- and 4-positions of the
central phenyl ring (R2) of 6 were directed towards the position
of the phosphate moiety in S1P (1), which established a
hydrogen-bonding network with PLP (4) and S1P lyase.
Introduction of a side chain with a polar functionality in those
positions could have the potential to form a similar network. To
test this hypothesis, the 4-bromophenyl derivative of 6 was
synthesized according to Scheme 1, starting from the 4-
bromophenyl analog of 8. This core was then used to produce
parallel synthesis libraries based on Negishi and Sonogashira
couplings in which the types and tether lengths, and the polar
functionalities of the attached side chains were explored.
Comparison of compounds 24 and 25 in Table 1 reveals that 4-
For the general design of the hit-explosion libraries, the core
molecules were enumerated with accessible custom and
commercially available monomers. The virtual reaction products
were then triaged based on their drug-like properties (rule-of-
five,11 golden triangle,12 three-75 rule,13 fraction of sp3 centers,14
number of rotatable bonds and polar surface area,15,16 number of
aromatic rings,17 and sweet spot18) and selected monomers were

substitution is preferred, with 25 achieving a clear potency
improvement (IC50 = 150 nM). The alkyne 26 was equipotent and
an X-ray structure of this compound bound to S1P lyase
uncovered that it was indeed forming the predicted hydrogen-
bonding network.
Table 1. Enzyme potencies of selected S1P lyase inhibitors. Detailed assay conditions were described previously.19
Compound R1
R2
R3
IC50 [μM]
≡ 5-methyl-isoxazole-3-yl
phenyl
≡ 4-methoxy-2,5-
dimethylbenzyl
6
1.0
14
phenyl
phenyl
phenyl
phenyl
4-methoxy-2,5-dimethylbenzyl
4-methoxy-2,5-dimethylbenzyl
4-methoxy-2,5-dimethylbenzyl
4-methoxy-2,5-dimethylbenzyl
3.8
6.4
15
16
17
9.2
≡ 5-cyclopropyl-isoxazole-3-yl
0.73
5-methyl-isoxazole-3-yl
5-methyl-isoxazole-3-yl
2-methoxyphenyl
3-methoxyphenyl
4-methoxyphenyl
4-fluorophenyl
4-pyridyl
4-methoxy-2,5-dimethylbenzyl
4-methoxy-2,5-dimethylbenzyl
4-methoxy-2,5-dimethylbenzyl
4-methoxy-2,5-dimethylbenzyl
4-methoxy-2,5-dimethylbenzyl
>50
7.3
18
19
20
21
22
5-methyl-isoxazole-3-yl
5-methyl-isoxazole-3-yl
5-methyl-isoxazole-3-yl
5.0
0.72
0.85
5-methyl-isoxazole-3-yl
4-methoxy-2,5-dimethylbenzyl
9.6
24
5-methyl-isoxazole-3-yl
4-methoxy-2,5-dimethylbenzyl
0.15
25
5-methyl-isoxazole-3-yl
4-methoxy-2,5-dimethylbenzyl
4-methoxybenzyl
0.15
26
5-methyl-isoxazole-3-yl
5-methyl-isoxazole-3-yl
phenyl
phenyl
>50
1.0
23
27
5-methyl-isoxazole-3-yl
phenyl
0.12
28
≡ 4-ethoxy-2,5-
dimethylbenzyl
5-methyl-isoxazole-3-yl
5-cyclopropyl-isoxazole-3-yl
5-methyl-isoxazole-3-yl
phenyl
0.26
4.6
29
30
31
4-fluorophenyl
phenyl
4-ethoxy-2,5-dimethylbenzyl
0.59
5-methyl-isoxazole-3-yl
phenyl
1.2
32
The X-ray structures further indicated that the methoxy group
d) and prepared monomers by alkylating (Cs2CO3, KI, DMF,
80oC, 12h) the resulting phenol with a set of solubilizing groups
(alkyl halides of various tether lengths with polar functional
groups and aliphatic or aromatic heterocycles). Several library
members were able to maintain the potency of the original
on the 2,5-dimethyl-4-methoxy-benzyl moiety of 6 could be
further extended. On this side of the enzyme, the binding pocket
is solvent exposed. We therefore hydrolyzed the methoxy group
of 2,5-dimethyl-4-methoxy-benzaldehyde (BBr3, CH2Cl2, r.t., 2

HT-ADME profiling of the synthesized compounds revealed
high unbound clearance as the biggest liability in the
isoxazolecarboxamide series. The results of the metabolite
identification for 6 in the presence of human (HLM) or rat liver
microsomes (RLM) are shown in Figure 3A. The numbers
indicate the potential prevalence of each metabolite, based solely
on the peak height in the LC/MS spectrum after a 60 min.
incubation period (1 = largest peak). Figure 3B displays the
predicted sites of metabolism of 6 using the MetaSite software
(Molecular Discovery).21 The prediction correlated very well
with the experimental data, which allowed us to pre-screen
newly-designed compounds that were intended to block the
metabolism sites.
Figure 2. Overlay of the X-ray crystal structure of compound (S)-6
(orange) and the pyridoxal-4-phosphate (4) cofactor (blue) bound to S1P
lyase (grey) (PDB code 5EUE) and a model of S1P (1) (purple) bound to S1P
lyase.19
screening hit, as compound 27 demonstrates (Table 1), but none
of them led to an activity improvement. The overlay with the
natural substrate shows that S1P (1) only has an aliphatic chain in
this position. We therefore also alkylated the generated phenol
with aliphatic alkyl halides and used those intermediates in our
library synthesis. This exercise was more successful, alkyl groups
up to C6 led to a potency improvement, with the sweet spot being
at n-butyl (28: IC50 = 117 nM). More details on the X-ray
crystallography are reported in an accompanying paper.19
To finalize our initial potency optimization efforts, we were
interested to see whether the established SAR information was
additive. To keep the size of the target molecule in a drug-like
range, we combined the cyclopropylisoxazole moiety from
compound 17 with the 4-fluorophenyl portion of 21 and the
ethoxy-dimethylbenzyl group of 29. Disappointingly, the
resulting product 30 only had an IC50 of 4.6 μM. The SAR
therefore didn’t appear to be additive.
Figure 3. Metabolite identification for 6 in the presence of human
(HLM) or rat (RLM) liver microsomes (the numbers indicate the prevalence:
1 = highest). B) Metasite prediction of the likelihood of sites in 6 to be
metabolized (black = most likely, light gray = least likely).
All compounds with enzyme potencies of <1 μM were tested
in a cell assay.20 Starting from the parent compound 6, which
exhibited a cell activity of 1.8 μM (Table 2), cell penetration
appears to be very sensitive to the cyclopropyl group in 17 and
substitutions on the central phenyl ring (21, 22, 25 and 26). In
particular 22 with its basic 4-pyridine, which could be positively
charged, had a poor activity in whole cells. On the other hand, the
O-alkyl analogs were much better tolerated with 28 achieving a
greatly improved cell potency of 230 nM.
The first compound that we prepared to reduce the rate of
metabolism aimed to block the O-dealkylation on the benzyl
residue. Applying the strategy that was used for the synthesis of
28, 4-hydroxy-2,5-dimethylbenzaldehyde was alkylated with
2,2,2-trifluoroethyl trifluoromethanesulfonate (pyridine, CH2Cl2,
0oC, 30 min) and the resulting intermediate was converted to 31.
Compared to the parent 6, the compound did not improve
metabolic stability in either HLM and RLM, but its enhanced cell
potency of 240 nM allowed it to achieve an improved multi-
parameter optimization (MPO) score22,23 with a cell IC50
∙
Table 2. Cell activity, metabolic stability and multi-parameter
optimization (MPO) scores for selected S1P lyase inhibitors.
CLint,HLM unbound of only 51 (Table 2). Additional compounds that
were investigated included the already available 4-fluorophenyl
compound 21, in the hope that it could block the metabolism in
the 4-position of the central phenyl ring. We further envisioned
that replacing the 5-methyl group in the benzyl moiety of 6 with
a trifluoromethyl substituent could prevent any oxidation in this
position. The respective benzaldehyde was generated from
Cell IC50 CLint,HLM unbound
CLint,RLM unbound
[L/h/kg]
cell IC50 ∙
CLint,HLM unbound
Cmpd.
[μM]
[L/h/kg]
1.8
1.3
5.5
7.2
25
110
-
300
100
-
-
-
-
-
210
170
18
21
720
190
-
1600
1100
-
-
-
-
-
6
(S)-6
17
21
22
25
26
28
29
31
32
33
34
-
3100
580
-
-
-
methyl
5-iodo-4-methoxy-2-methylbenzoate
by
4.4
3.7
trifluoromethylation following Huber et al.24,25 in HMPA26 and a
reduction oxidation sequence (LiAlH4/Dess-Martin). The
obtained building block was used to produce 32. The cyclopropyl
group in 17 could have the potential to reduce metabolism on the
methyl group off the isoxazole ring. Unfortunately none of those
modifications were able to lower unbound metabolic clearance
and their weak cell potencies ultimately led to high MPO scores.
0.23
0.93
0.24
2.7
n.d.
n.d.
-
-
600
1100
38
34
51
450
-
-
n.d: not determined due to low in vitro potency.

Finally we looked at abolishing the N-dealkylation by
subsequently changing the two methylene groups in 6 into
carbonyl functionalities (Figure 4: 33 and 34). To obtain the
carboxamide 33, 2,5-dimethyl-4-methoxy-benzaldehyde was
oxidized (Jones reagent, acetone, 50oC, 12 h) to the carboxylic
acid and was coupled with 10 (HATU, DIEA, CH2Cl2, 12 h).
While this change led to a loss of potency (IC50 = 12.4 μM), it
was able to significantly improve the metabolic stability both in
HLM (CLint,unbound = 18 L/h/kg) and RLM (CLint,unbound = 38
L/h/kg).
Acknowledgments
All authors are employees of AbbVie. The design, study
conduct, and financial support for this research was provided by
AbbVie. AbbVie participated in the interpretation of data,
review, and approval of the publication.
We thank the chemists at WuXi AppTec who prepared the
compounds described in this article and the structural chemistry
group at AbbVie for the acquisition of NMR and MS spectra.
References and notes
1.
2.
Cyster, J. G.; Schwab, S. R. Ann. Rev. Immunol. 2012, 30, 69.
Schwab, S. R.; Pereira, J. P.; Matloubian, M.; Xu, Y.; Huang, Y.;
Cyster, J. G. Science, 2005, 309, 1735.
3.
4.
Vogel, P.; Donoviel, M. S.; Read, R.; Hansen, G. M.; Hazlewood,
J.; Anderson, S. J.; Sun, W.; Swaffield, J.; Oravecz, T. PloS One,
2009, 4, e4112.
Bagdanoff, J. T.; Donoviel, M. S.; Nouraldeen, A.; Carlsen, M.;
Jessop, T. C.; Tarver, J.; Aleem, S.; Dong, L.; Zhang, H.; Boteju,
L.; Hazelwood, J.; Yan, J.; Bednarz, M.; Layek, S.; Owusu, I. B.;
Gopinathan, S.; Moran, L.; Lai, Z.; Kramer, J.; Kimball, S. D.;
Yalamanchili, P.; Heydorn, W. E.; Fraier, K. S.; Brooks, B.;
Brown, P.; Wilson, A.; Sonnenburg, W. K.; Main, A.; Carson, K.
G.; Oravecz, T.; Augeri, D. J. J. Med. Chem. 2010, 53, 8650.
Coburn, S. P.; Mahuren, J. D. J. Biol. Chem. 1976, 251, 1646.
Phillips, J. A.; Paine, A. J. Xenobiotica, 1990, 20, 555.
Weiler, S.; Braendlin, N.; Beerli, C.; Bergsdorf, C.; Schubart, A.;
Srinivas, H.; Oberhauser, B.; Billich, A. J. Med. Chem. 2014, 57,
5074.
5.
6.
7.
Figure 4. The carboxamides 33 and 34.
8.
Miller-Moslin, K.; Peukert, S.; Jain, R. K.; McEwan, M. A.;
Karki, R.; Llamas, L.; Yusuff, N.; He, F.; Li, Y.; Sun, Y.; Dai, M.;
Perez, L.; Michael, W.; Sheng, T.; Lei, H.; Zhang, R.; Williams,
J.; Bourret, A.; Ramamurthy, A.; Yuan, J.; Guo, R.; Matsumoto,
M.; Vattay, A.; Maniara W.; Amaral, A.; Dorsch, M.; Kelleher, J.
F. J. Med. Chem. 2009, 52, 3954.
Kashem, M. A.; Wa, C.; Wolak, J. P.; Grafos, N. S.; Ryan, K. R.;
Sanville-Ross, M. L.; Fogarty, K. E.; Rybina, I. V.; Shoultz, A.;
Molinaro, T.; Desai, S. N.; Rajan, A.; Huber, J. D.; Nelson, R. M.
Assay Drug Dev. Technol. 2014, 12, 293.
To prepare the amino acid amide 34, 2,5-dimethyl-4-
methoxy-benzaldehyde was converted to the corresponding
benzylamine (NH2OH ∙ HCl, pyridine, 12h, then Zn powder,
HCl, THF, 80oC, 30 min.) and was coupled with Boc-protected
phenyl glycine (HATU, Et3N, CH2Cl2, 12 h). Amide formation
according to Scheme 2 led to the desired product. As with the
prior compound, 34 could not match 6 in its S1P lyase inhibitory
activity (IC50 = 39.2 μM), but clearly improved the stability of
the compound in the presence of liver microsomes (CLint,HLM
unbound = 21 L/h/kg and CLint, RLM unbound = 34 L/h/kg). We therefore
conclude that blocking the N-dealkylation via modification of
either methylene group in 6 has the highest potential to reduce
the rate of metabolism for the isoxazolecarboxamide-based S1P
lyase inhibitors.
9.
10. Henry, R. A.; Hollins, R. A.; Lowe-Ma, C.; Moore, D. W.; Nissan,
R. A. J. Org. Chem. 1990, 55, 1796.
11. Lipinski, C. A. Drug Disc. Today: Technol., 2004, 1, 337.
12. Johnson, T. W.; Dress, K. R.; Edwards, M. Bioorg. Med. Chem.
Lett. 2009, 19, 5560.
13. Hughes, J. D.; Blagg, J.; Price, D. A.; Bailey, S.; DeCrescenzo, G.
A.; Devraj, R. V.; Ellsworth, E.; Fobian, Y. M.; Gibbs, M. E.;
Gilles, R. W.; Greene, N.; Huang, E.; Krieger-Burke, T.; Loesel,
J.; Wager, T.; Whiteley, L.; Zhang, Y. Bioorg. Med. Chem. Lett.
2008, 18, 4872.
14. Lovering, F.; Bikker, J.; Humblet, C. J. Med. Chem. 2009, 52,
6752.
15. Veber, D. F; Johnson, S. R.; Cheng, H. Y.; Smith, B. R.; Ward, W.
W.; Kopple, K. D. J. Med. Chem. 2002, 45, 2615.
16. Kelder, J.; Grootenhuis, P. D. J.; Bayada, D. M.; Delbressine, L.
P. C.; Ploemen, J. P. Pharm. Res. 1999, 16, 1514.
17. Ritchie, T. J.; Mcdonald, S. J. F.; Young, R. J.; Pickett, S. D. Drug
Disc. Today, 2011, 16, 164.
In summary, the isoxazolecarboxamide S1P lyase inhibitor 6
was very amenable to high-speed parallel synthesis, which
allowed us to quickly establish consistent and interpretable SAR
information that spans an enzyme potency range of more than
two orders of magnitude. In addition, we succeeded in
determining the binding mode of 6 and some of its analogs in the
active site of S1P lyase through X-ray crystallography and
demonstrated that it is possible to obtain inhibitors with
improved potencies via structure-based design. With the n-butyl
ether 28 (cell IC50 = 230 nM) and the trifluoroethyl ether 31 (cell
IC50 = 240 nM) we further established that the cell potencies of
18. Hann. M. M.; Keserü, G. M. Nat. Rev. Drug Disc. 2012, 11, 355.
19. Argiriadi, M. A.; Banach, D.; Radziejewska, E.; Marchie, S.;
DiMauro, J.; Hutchins, C.; Queeney, K.; Judge, R. A.; Dinges, J.;
Dominguez, E.; Wallace, G.; Harris, C. M. Bioorg. Med. Chem.
Lett. in press.
20. Cellular potency was determined by measuring the metabolism of
NBD-labeled sphingosine into NBD-hexadecenoic acid, a stable
metabolite of NBD-hexadecenal. HEK 293 H Cells were stably
transfected with wild-type, full-length human S1P Lyase. Cells, 5
x 104 per well, were incubated on poly-D-lysine coated 96 well
plates overnight in 100 l Hybridoma SFM media and 1% FBS, at
37°C and 8% CO2. Cells were pre-treated for 30 min with various
concentrations of inhibitor, added in 12.5 μL/well of 10x stocks in
10% DMSO. Next, 12.5 μL/well of a solution containing 50 μM
NBD-sphingosine (Avanti), 5 μM PMA, 10 mM Na3VO4, and 10
mM NaF was added, and cells were incubated 3 h as above. Cell
viability was measured by addition of 10 μl PrestoBlue (Invitrogen
#A13262) for 30 min and subsequent measurement of absorbance
the hit series can be optimized into a desirable range (cell IC50
<
500 nM). Finally we identified metabolic stability as the biggest
liability of the isoxazolecarboxamides and were able to pinpoint
that N-dealkylation of the secondary amine was the main culprit.
Based on the above results, we concluded that our S1P lyase
inhibitors had potential and that it should be possible to
overcome the metabolic liability problem. We therefore declared
the hit series to be viable and recommended it for further lead-
optimization efforts.

at 570 nm. Following that, cells were lysed by addition of 100
μL/well of 0.1% formic acid in acetonitrile, and cells were
centrifuged for 10 min at 1000 rpm. 96-well C18 resin plates (The
Nest Group, Incl, cat# SNS SS18VL) were pre-equilibrated with
200 μL 70% acetonitrile/30% H20, and excess solvent was
removed by centrifugation for 2 min at 2,000 rpm. 50 μl of
supernatant from cell lysates were added to these C18 plates. Plates
were centrifuged to remove solvent and washed with 2 cycles of
200 μL 70% acetonitrile/30% H20) followed by centrifugation.
Elution of NBD-hexadecenoic acid was accomplished by addition
of 200 μL 0.1% formic acid in acetonitrile and collection of the
material eluted upon centrifugation. Analyte concentration was
measured by fluorescence (λex = 460 nm, λem = 531 nm) on an
Envision plate reader. IC50 values were analyzed by least-means-
squares fitting to the following equation: Percent Activity =
100%/(1+([inhibitor]/IC50)). Assay precision was ascertained
from a set of 28 compounds with repeat data by computing a
Test/Re-test Mean Significant Ratio of 2.5, according to published
calculations.27
21. Zamora, I.; Fontaine, F.; Serra, B.; Plasencia, G. Drug Disc.
Today, Technol. 2013, 10, e199.
22. Gao, H.; Yao, L.; Mathieu, H. W.; Zhang, Y.; Maurer, T. S.;
Troutman, M. D.; Scot, D. O.; Ruggeri, R. B.; Lin, J. Drug Metab.
Dispos. 2008, 36, 2130.
23. Liu, X.; Chen, C.; Hop, C. E. C. A. Curr. Topics Med. Chem.
2011, 11, 450.
24. Huber, J. D.; Bentzien, J.; Boyer, S. J.; Burke, J.; De Lombert, S.;
Eickmeier, C.; Guo, X.; Haist, J. V.; Hickey, E. R.; Kaplita, E.;
Karmazyn, M.; Kemper, R.; Kennedy, C. A.; Kirrane, T.;
Madwed, J. D.; Mainolfi, E.; Nagaraja, N.; Soleymanzadeh, F.;
Swinamer, A.; Eldrup, A. B. J. Med. Chem. 2012, 55, 7114.
25. Chen, Q. Y.; Wu, S. W. J. Chem. Soc., Chem. Commun. 1989,
705.
26. Kunitomo, J.; Yoshikawa, M.; Fushimi, M.; Kawada, A.; Quinn, J.
F.; Oki, H.; Kokubo, H.; Kondo, M.; Nakashima, K.; Kamiguchi,
N.; Suzuki, K.; Kimura, H.; Taniguchi, T. J. Med. Chem. 2014,
57, 9627.
27. Iversen, P. W.; Beck, B.; Chen, Y. F.; Dere, W.; Devanarayan, V.;
Eastwood, B. J.; Farmen, M. W.; Iturria, S. J.; Montrose, C.;
Moore, R. A.; Weidner, J. R.; Sittampalam, G. S. In HTS Assay
Validation; Sittampalam, G. S.; Gal-Edd, N.; Arkin, M.; Auld, D.;
Austin, C.; Bejcek, B.; Glicksman, M.; Inglese, J.; Lemmon, V.;
Li, Z.; McGee, J.; McManus, O.; Minor, L.; Napper, A.; Riss, T.;
Trask, O. J.; Weidner J., Eds.; Assay Guidance Manual: Bethesda
Maryland, 2004.