Riluzole prodrugs for melanoma and ALS: Design, synthesis, and in vitro metabolic profiling

Article abstract of DOI:10.1016/j.bmc.2012.07.004

Riluzole (1) is an approved therapeutic for the treatment of ALS and has also demonstrated anti-melanoma activity in metabotropic glutamate GRM1 positive cell lines, a mouse xenograft assay and human clinical trials. Highly variable drug exposure following oral administration among patients, likely due to variable first pass effects from heterogeneous CYP1A2 expression, hinders its clinical use. In an effort to mitigate effects of this clearance pathway and uniformly administer riluzole at efficacious exposure levels, several classes of prodrugs of riluzole were designed, synthesized, and evaluated in multiple in vitro stability assays to predict in vivo drug levels. The optimal prodrug would possess the following profile: stability while transiting the digestive system, stability towards first pass metabolism, and metabolic lability in the plasma releasing riluzole. (S)-O-Benzyl serine derivative 9 was identified as the most promising therapeutically acceptable prodrug.

Full text of DOI:10.1016/j.bmc.2012.07.004

Bioorganic & Medicinal Chemistry 20 (2012) 5642–5648
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Riluzole prodrugs for melanoma and ALS: Design, synthesis, and in vitro
metabolic profiling
Mark E. McDonnell a, Matthew D. Vera a, Benjamin E. Blass a, Jeffrey C. Pelletier a, Richard C. King b,
Carmen Fernandez-Metzler b, Garry R. Smith a, Jay Wrobel a, Suzie Chen c, Brian A. Wall c, Allen B. Reitz a,
⇑
a Fox Chase Chemical Diversity Center, Inc., 3805 Old Easton Road, Doylestown, PA 18902, USA
b PharmaCadence Analytical Services, LLC, 2880 Bergey Road, Suite AA, Hatfield, PA 19440, USA
c Department of Chemical Biology, Ernest Mario School of Pharmacy, 164 Frelinghuysen Road, Rutgers University, Piscataway, NJ 08854, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Riluzole (1) is an approved therapeutic for the treatment of ALS and has also demonstrated anti-mela-
noma activity in metabotropic glutamate GRM1 positive cell lines, a mouse xenograft assay and human
clinical trials. Highly variable drug exposure following oral administration among patients, likely due to
variable first pass effects from heterogeneous CYP1A2 expression, hinders its clinical use. In an effort to
mitigate effects of this clearance pathway and uniformly administer riluzole at efficacious exposure lev-
els, several classes of prodrugs of riluzole were designed, synthesized, and evaluated in multiple in vitro
stability assays to predict in vivo drug levels. The optimal prodrug would possess the following profile:
stability while transiting the digestive system, stability towards first pass metabolism, and metabolic
lability in the plasma releasing riluzole. (S)-O-Benzyl serine derivative 9 was identified as the most prom-
ising therapeutically acceptable prodrug.
Received 31 May 2012
Revised 27 June 2012
Accepted 4 July 2012
Available online 21 July 2012
Keywords:
Riluzole
Prodrug
Cancer
Melanoma
Cyp1A2
Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction
cells.11 These results support the notion that high glutamate re-
lease may promote tumor cell growth.1–7
The glutamatergic system plays a key role in tumor biology.1–7
The growth inhibition of various human tumors including colon
adenocarcinomas, breast, gliomas, and lung carcinomas by specific
antagonists to ionotropic glutamate receptors has been reported.2,6
Furthermore, implanted glioma cells with high levels of glutamate
release showed a distinct growth advantage.2 We have evaluated
riluzole (1; Scheme 1), the only currently FDA-approved drug for
the treatment of amyotrophic lateral sclerosis (ALS),8 for potential
anti-cancer activity, as it inhibits glutamate release, thus lowering
the potential for glutamatergic system signaling and function.9,10
Riluzole blocks the release of glutamic acid and thus lessens the
response mediated by the metabotropic glutamate receptor GRM1
(mGluR1) among other receptors that recognize glutamic acid as a
cognate ligand. We have shown a reduction in the number of viable
cells after treatment with riluzole in two different GRM1-positive
melanoma cell lines (C8161 and SKMEL187) while two different
GRM1-negative human melanoma cell lines (UACC930 and C81-
61) or normal human melanocytes (HEM) were much less sensitive
under similar conditions.11 Furthermore, we have also demon-
strated a correlation between a decrease in levels of released glu-
tamate and number of viable cells in riluzole-treated C8161
Metastatic melanoma has few treatment options. Surgical
resection is the primary modality for cutaneous malignant mela-
noma but results in local recurrence rates as high as 50%. For a long
time the therapeutic standard of care was dacarbazine, a highly
cytotoxic drug with severe side effects including vomiting, head-
ache and hair loss.12,13 Treatment with dacarbazine has a median
progression-free enhancement of survival time of only 1.5 months.
Recently, two new drugs showed survival benefit in randomized
phase III clinical trials. Ipilimumab, a human monoclonal anti-
immunosuppressive antibody against the immune check-point
CTLA-4 inhibitory receptor enhanced life expectancy and
vemurafenib, a small molecule inhibiting mutated BRAF, improved
F3CO
S
F3CO
S
N
Cyp1A2
NH2
NHOH
N
Riluzole, 1
2
HO
OH
OH
CO2H
F3CO
S
N
UDPGT
NH
O
O
3
⇑
Corresponding author. Tel.: +1 215 589 6435; fax: +1 215 589 6335.
E-mail address: AReitz@fc-cdci.com (A.B. Reitz).
Scheme 1. Riluzole (1), and metabolities 2 and 3.
0968-0896/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmc.2012.07.004

M. E. McDonnell et al. / Bioorg. Med. Chem. 20 (2012) 5642–5648
5643
overall survival for several months in most melanoma patients
O
with mutated BRAF.14,15 Despite these advances, the clinical re-
sponses are not durable and relapse of melanoma is a near-cer-
tainty. There is a need for new melanoma treatments that are
both disease-modifying and effective in treating patients that are
refractory to current treatments. Riluzole, a generally non-toxic
drug for which there is significant clinical data available, is the only
FDA-approved treatment for ALS.16 We tested the efficacy of riluz-
ole in melanoma patients with a poor prognosis and severely lim-
ited treatment options.17 Riluzole demonstrated efficacy in this
patient group in recently completed phase 0 and phase 2 clinical
trials.17 These results, along with the mild side-effect profile that
riluzole has shown among ALS patients, suggests this drug has sig-
nificant potential for use as an improved treatment for metastatic
melanoma.
F3CO
F3CO
R
S
N
S
N
i, ii
NH
NH2
Scheme 2. Preparation of riluzole prodrugs. Reagents and conditions (i) RCO2H,
EDCI, DCM; or ROCOCl, Et3N, DCM; or RCO2COR’, DMF, then R1R2NH, HATU, Et3N,
DMF (ii) TFA, DCM, 2 h if deprotection is required.
Esterases would also act on succinates 20 and 21 and amidases
on 22 and 23 to afford resulting acid 19 (Scheme 3, Eq. 4), which
would then cyclize to succinic anhydride and riluzole.25 Amines
24 and 25 would release 1 in a pH dependent manner (Scheme 3,
Eq. 3),26 and metabolic reduction of ortho-nitrophenyl amide ana-
log 26 to the corresponding amino derivative would liberate riluz-
ole as well (Scheme 3, Eq. 5).27
The clinical use of riluzole is characterized by extensive hepatic
metabolism and an exceptionally high level of patient-to-patient
variability in drug exposure, due to variable first pass elimination
effects caused by heterogeneous expression of the cytochrome
P450 isoform CYP1A2.18–21 N-Hydroxylation of riluzole by CYP1A2
to give 2 is quickly followed by O-glucoronidation to 3 and subse-
quent elimination (Scheme 1).21 In order to circumvent the first
pass metabolism by CYP1A2 we devised a prodrug approach where
the riluzole exocyclic nitrogen is masked in such a way as not to be
recognized by CYP1A2 as a substrate for oxidation. A successful
prodrug would be stable in the gastrointestinal (GI) tract upon oral
administration, stable to first pass hepatic metabolism and then
cleaved in plasma or at the target tissue to liberate riluzole. Herein
we report the identification and selection of a riluzole prodrug
suitable for in vivo testing based on the design, synthesis, and rapid
in vitro evaluation of the 23 potential riluzole prodrugs shown in
Figure 1.
To determine suitability as prodrugs, compounds 4–26 were
first tested for stability in various in vitro assays including simu-
lated gastric fluid (SGF), simulated intestinal fluid (SIF), mouse
and human liver microsomes (mLM, hLM), as well as mouse and
human plasma (mPS, hPS).28 Mouse assays were included to pre-
dict in vivo stability for eventual murine xenograft assays. The data
are shown graphically in Figure 2.29 Individual assay results indi-
cate disappearance of prodrug, as a percent of the original concen-
tration, after the incubation times indicated. In order to be
considered for further study, compounds should be stable in simu-
lated gastric fluid (SGF) and simulated intestinal fluid (SIF) as well
as liver microsomes (LM) to predict intact absorption into the plas-
ma. Instability in plasma was assumed to predict in vivo release of
the active species required for tumor exposure. Compounds were
found to be stable in the SGF assay, however, amino acid deriva-
tives 4–6, morpholino succinamide 23, and 4-aminobutyrate
analog 25 were rapidly degraded in the SIF assay. Interestingly,
c
-aminobutyrate analogue 24 is significantly more stable than
2. Results and discussion
substituted aminobutyrate 25. This stability difference was attrib-
uted to the Thorpe-Ingold effect in which the gem-dialkyl substitu-
tion enhances intramolecular cyclization.30 Stability in the LM
assays suggest first pass liver metabolism will be insignificant
and systemic exposure will be high.31 Hence, derivatives 6, 10,
13, 14, 18–21, 24 and 26 were excluded for further consideration
due to instability to LMs. Finally, varying levels of instability were
seen with 5, 6, 9 and 20 in the mouse and human plasma assays
suggesting the parent substance, riluzole, was being released via
plasma amidases.
The preparation of prodrugs from amine-containing drugs and
drug candidates is well described in the literature.22–24 Scheme 2
shows the preparation of potential riluzole-derived prodrugs
involving conversion of the exocyclic amine to single alpha-amino
amides (4–10), carbamates (11–18), succinamides (19–23) and
amide linkages from c-aminobutyric acids (24, 25). Plasma bound
amidases and esterases would be expected to cleave 4–10 and 11–
18, respectively, to give riluzole directly (Scheme 3, Eqs. 1, 2).23
O
O
O
NH2
R
F3CO
F3CO
F3CO
R
O
OR
S
N
S
N
S
N
NH
NH
NH
4
R = (S)-Me
11 R = Me 15 R = s-Bu
12 16
R = n-Hx
8 R = (R)-Bn
19 R = OH
20
5 R = (R)-Me
6 R = (S)-i-Bu
9
R = Et
R = (S)-CH2OBn
R = OMe
21 R = O-t-Bu
22 R = NHCH2-3-pyridyl
13 R = Pr
14 R = Bu
17 R = (CH2)2NMe2
18 R = (CH2)3NMe2
10 R = (R)-CH2OBn
7
R = (S)-Bn
23
R = NH(CH2)2-1-morpholinyl
O
O
O
NH2
F3CO
F3CO
F3CO
NH2
NH
NO2
S
N
S
S
N
NH
N
NH
24
25
26
Figure 1. Potential prodrugs of riluzole prepared and evaluated for in vitro stability properties.

5644
M. E. McDonnell et al. / Bioorg. Med. Chem. 20 (2012) 5642–5648
O
NH2
R'
metabolic
reduction
Eq. 1
Eq. 5
O
NH
plasma
NH2
O
amidase
NO2
(ref. 27)
NH
(ref. 23)
NH
plasma
O
O
R'
O
esterase
S
Eq. 2
F3C
NH2
(ref. 23)
N
NH
pH
dependant
X R'
O
OH
O
(ref. 25)
O
O
Eq. 4
(ref. 26)
NH
NH
O
NH2
R'
X= O, plasma
esterase
Eq. 3
NH
X = NH, plasma
amidase
(ref. 23)
Scheme 3. Enzymatic and chemical mechanisms for the release of riluzole from various prodrug classes.
Figure 2. Stability results for potential prodrugs 4–26 evaluated in SGF, SIF, human plasma (hPS), mouse plasma (mPS), human liver microsomes (hLM) and mouse liver
microsomes (mLM). Vertical bars represent percent disappearance of prodrug after 60 min incubation times (30 min incubation for microsome assays). See Supplementary
data for actual data in tabular form. [Intended for color reproduction].
Comparison of all data indicated that 9 had the most promising
balance of stability in the intestinal and microsomal assays, and
desired lability in the plasma assays. Therefore, this compound
was subjected to more extensive analysis to determine half-lives
of prodrug disappearance in SGF, SIF, hLM and hPS assays (Table 1).
Riluzole was also detected in the hPS assay. The data from these
studies support compound 9 as a candidate prodrug that will tra-
verse the GI tract and not be metabolized by the liver.
Riluzole is predominantly metabolized by the CYP1A2 iso-
zyme19 which is expressed heterogeneously in the population. This
leads to variable bioavailability and drug exposure among patients
with one report indicating riluzole exposure per kilogram of body
evaluation, we wanted to ensure that this compound itself was
not a substrate for oxidative metabolism by CYP1A2 compared to
riluzole. Therefore, both riluzole and 9 were subjected to stability
assays in liver microsomes from two individual donors expressing
high and low CYP1A2 activity, respectively. Microsome baseline
activity was determined by disappearance of the CYP1A2 substrate
positive control phenacetin accompanied by the appearance of
acetaminophen. As expected, riluzole was metabolized to a greater
extent in high activity CYP1A2 microsomes (47% remaining at
30 min)20 than in low CYP1A2 activity microsomes (100% remain-
ing at 30 min). Prodrug 9 was not metabolized to a significant ex-
tent in either microsome samples with or without the co-factor
NADPH (P70% remaining at 30 min), indicating the prodrug is a
poor substrate for the CYP1A2 isozyme (Table 2).
weight at 48.7
l
g-h/L/Kg with a high standard deviation of 40.9
(189 patients).18 Before considering prodrug
9
for in vivo

M. E. McDonnell et al. / Bioorg. Med. Chem. 20 (2012) 5642–5648
5645
Table 1
Evaluation of compound 9 in various in vitro human stability assays.32 Values are half-lives of prodrug disappearance in the indicated media.
O
NH2
(S)
F3CO
S
N
NH
O
9
Simulated gastric fluid
>60 min
Simulated intestinal fluid
98 min
Human liver microsomes
>60 min
Human plasma stability
ꢁ300 mina
a
Riluzole formation occurred during the experiment.
Table 2
Human liver microsome stability of riluzole 1 and prodrug 9 in liver microsomes with high and low Cyp 1A2 activity based on phenacetin metabolism to acetaminophen
Microsomes
NADPH
Time (min)
Riluzole 1a(%)
Compound 9a(%)
High Cyp1A2 Activityb
(+)
(+)
(ꢂ)
(+)
(+)
(ꢂ)
0
30
30
0
30
30
100
47c
100
100
100
91
100
70
77
100
78
76
Low Cyp1A2 Activityb
a
b
c
Compounds tested at 1
l
M concentration, n = 3.
Cyp1A2 activity level was based on 10 fold kinetic conversion ratio of phenacetin to acetaminophen in microsomes.
Accompanied by substantial formation of N-hydroxyriluzole.
4.2. General synthetic method of amide analogs 4–10, 24–26
6-Trifluoromethoxy-benzothiazol-2-ylamine (1, 100 mg,
3. Conclusion
The only FDA-approved drug for ALS, riluzole, has shown
promising activity in GRM1 positive melanoma cell assays,
refractory melanoma clinical studies and in an in vivo xenograft
assay. To stabilize highly variable exposure levels of the drug
resulting from patient to patient heterogeneous CYP1A2 expres-
sion, putative prodrugs were prepared and evaluated in intesti-
nal, microsome, and plasma stability assays. O-Benzylserine
derivative 9 emerged from this process as a preferred prodrug
candidate and was further characterized in SGF, SIF, liver micro-
some, and plasma stability assays. The presence of riluzole was
confirmed in the human plasma assay indicating the prodrug is
indeed cleaved to provide riluzole. Additional evaluation for
metabolism in microsome assays with both high and low CYP1A2
activity levels indicated 9 was most likely not a CYP1A2 sub-
strate. Therefore, compound 9 is expected to be a riluzole pro-
drug that could display consistent plasma exposure levels upon
in vivo administration and is therefore an ideal candidate for
testing in murine xenograft efficacy models following evaluation
in cell based melanocyte assays, for which data will be reported
in due course.
0.42 mmol), a carboxylic acid (0.43 mmol), and EDCI (123 mg,
0.65 mmol) were combined in methylene chloride (5 mL) and stir-
red 4 days. The reaction was washed with 0.1 N HCl (2 ꢀ 10 mL),
dried over magnesium sulfate, filtered, and then concentrated.
Chromatography on silica gel afforded the t-Boc protected com-
pounds as well as compound 26 directly. Compounds 4–10, 24,
and 25 were dissolved in 4 N HCl and 1,4-dioxane and stirred 3 h
at ambient temperature. The reactions were then concentrated to
afford the products.
4.3. (S)-2-Amino-N-(6-trifluoromethoxybenzothiazol-2-yl)-
propionamide (4)
Yield: 71%; white powder. 1H NMR (300 MHz, DMSO-d6) d 8.57
(s, 1H), 8.18 (s, 1H), 7.88 (d, J = 8.8 Hz, 1H), 7.46 (d, J = 8.8 Hz, 1H),
4.78–3.63 (m, 1H), 1.46 (d, J = 7.1 Hz, 3H). HPLC–MS m/z (M+)
305.9; Tr = 3.58 (95%).
4.4. (R)-2-Amino-N-(6-trifluoromethoxybenzothiazol-2-yl)-
propionamide (5)
4. Materials and methods
4.1. General
Yield: 59%; white powder. 1H NMR (300 MHz, DMSO) d 8.57 (s,
1H), 8.18 (s, 1H), 7.88 (d, J = 8.8 Hz, 1H), 7.46 (d, J = 8.8 Hz, 1H),
3.98 (bs, 3H), 1.46 (d, J = 7.1 Hz, 3H). HPLC–MS m/z (M+) 305.9;
Tr = 3.57 (>95%).
All chemicals and anhydrous solvents were purchased from Sig-
ma–Aldrich (St. Louis, Mo.) and used without additional purifica-
tion. General solvents and reagents were purchased from Fisher
Scientific. 1H NMR spectra were obtained on a Varian Mercury
300-MHz NMR. Purity (%) and mass spectral data were determined
with a Waters Alliance 2695 HPLC/MS (Waters Symmetry C18,
4.5. (S)-2-Amino-3-methyl-N-(6-trifluoromethoxybenzothiazol-
2-yl)-butyramide (6)
Yield: 66%; white powder. 1H NMR (300 MHz, DMSO-d6) d
8.55 (s, 1H), 8.18 (s, 1H), 7.88 (d, J = 8.8 Hz, 1H), 7.46 (d,
J = 8.9 Hz, 1H), 4.00 (m, 1H), 3.43 (m, 2H), 3.16 (s, 4H), 2.26
(d, J = 6.4 Hz, 1H), 1.29–0.63 (m, 6H). HPLC–MS m/z (M+)
333.9; Tr = 3.92 (>95%).
4.6 ꢀ 75 mm, 3.5
lm) with a 2996 diode array detector from
210–400 nm; the solvent system is 5–95% MeCN in water over nine
mins using a linear gradient and retention times are in minutes.
Yields refer to the amount of product after purification and are
not optimized.

5646
M. E. McDonnell et al. / Bioorg. Med. Chem. 20 (2012) 5642–5648
4.6. (S)-2-Amino-3-phenyl-N-(6-trifluoromethoxybenzothiazol-
2-yl)-propionamide (7)
solid collected by filtration and dried under vacuum to afford
product.
Yield: 83%; beige powder. 1H NMR (300 MHz, CD3OD) d 7.90 (s,
1H), 8.05–7.47 (m, 2H), 7.81 (d, J = 8.9 Hz, 1H), 7.34 (dt, J = 9.7,
7.8 Hz, 2H), 7.34 (dt, J = 9.7, 7.8 Hz, 2H), 4.40 (dd, J = 8.1, 6.3 Hz,
1H), 3.46–3.30 (m, 2H), 3.19 (dd, J = 14.0, 8.2 Hz, 1H). HPLC–MS
m/z (M+) 381.9; Tr = 4.27 (>95%).
4.14. (6-Trifluoromethoxybenzothiazol-2-yl)-carbamic acid
methyl ester (11)
Yield: 47%; white powder. 1H NMR (300 MHz, DMSO-d6) d 8.22
(s, 1H), 7.89 (d, J = 8.8 Hz, 1H), 7.52 (d, J = 9.0 Hz, 1H), 3.97 (s, 1H),
3.44 (s, 3H). HPLC–MS m/z (M+) 292.9; Tr = 5.26 (94%).
4.7. (R)-2-Amino-3-phenyl-N-(6-trifluoromethoxybenzothiazol-
2-yl)-propionamide (8)
4.15. (6-Trifluoromethoxybenzothiazol-2-yl)-carbamic acid
ethyl ester (12)
Yield: 72%; white powder. 1H NMR (300 MHz, DMSO-d6) d 8.75
(bs, 2H), 8.31 (s, 1H), 7.99 (d, J = 8.9 Hz, 1H), 7.59 (d, J = 8.8 Hz, 1H),
7.42 (m, 4H), 4.53 (bs, 1H), 3.44 (s, 2H), 3.32 (m, 2H). HPLC–MS m/z
(M+) 381.9; Tr = 4.21 (>95%).
Yield: 67%; white powder. 1H NMR (300 MHz, DMSO-d6) d 8.23 (s,
1H), 7.90 (d, J = 8.8 Hz, 1H), 7.52 (d, J = 8.7 Hz, 1H), 4.40 (q, J = 7.1 Hz,
2H), 1.42 (t, J = 7.1 Hz, 3H). HPLC–MS m/z (M+) 306.9; Tr = 5.62 (96%).
4.8. (S)-2-Amino-3-benzyloxy-N-(6-trifluoromethoxybenzothi-
azol-2-yl)-propionamide (9)
4.16. (6-Trifluoromethoxybenzothiazol-2-yl)-carbamic acid
propyl ester (13)
Yield: 66%; white powder. 1H NMR (300 MHz, DMSO-d6) d 8.66
(s, 1H), 8.19 (s, 1H), 7.88 (d, J = 8.9 Hz, 1H), 7.47 (d, J = 8.9 Hz, 1H),
7.35–7.18 (m, 5H), 4.56 (q, J = 12.3 Hz, 2H), 4.44 (s, 1H), 4.11–3.77
(m, 2H). HPLC–MS m/z (M+) 411.9; Tr = 4.40 (95%).
Yield: 74%; white powder. 1H NMR (300 MHz, DMSO-d6) d 12.15
(s, 1H), 8.10 (s, 1H), 7.77 (d, J = 8.8 Hz, 1H), 7.39 (d, J = 8.6 Hz, 1H),
4.18 (t, J = 6.7 Hz, 2H), 2.57–2.45 (m, 2H), 0.95 (t, J = 7.4 Hz, 3H).
HPLC–MS m/z (M+) 320.9; Tr = 5.98 (95%).
4.9. (R)-2-Amino-3-benzyloxy-N-(6-trifluoromethoxybenzothi-
azol-2-yl)-propionamide (10)
4.17. (6-Trifluoromethoxybenzothiazol-2-yl)-carbamic acid
butyl ester (14)
Yield: 48%; buff powder. 1H NMR (300 MHz, DMSO-d6) d 8.37 (d,
J = 103.5 Hz, 1H), 8.20 (s, 1H), 7.89 (d, J = 8.8 Hz, 1H), 7.48 (d,
J = 9.6 Hz, 1H), 7.40–7.18 (m, 5H), 4.57 (q, J = 12.4 Hz, 2H), 4.42
(m, 2H), 4.06–3.82 (m, 1H), 3.50 (s, 1H). HPLC–MS m/z (M+)
411.9; Tr = 4.43 (>95%).
Yield: 56%; white powder. 1H NMR (300 MHz, DMSO-d6) d 12.14
(s, 1H), 8.09 (s, 1H), 7.77 (d, J = 8.8 Hz, 1H), 7.39 (d, J = 7.4 Hz, 1H),
4.21 (t, J = 6.7 Hz, 1H), 1.73–1.31 (m, 2H), 1.32 (t, J = 6.9 Hz, 2H),
0.88 (t, J = 6.7 Hz, 3H). HPLC–MS m/z (M+) 334.9; Tr = 6.36 (95%).
4.18. (6-Trifluoromethoxybenzothiazol-2-yl)-carbamic acid
isobutyl ester (15)
4.10. 5-Aminopentanoic acid (6-trifluoromethoxybenzothiazol-
2-yl)-amide (24)
Yield: 66%; white powder. 1H NMR (300 MHz, DMSO-d6) d 12.16
(s, 1H), 8.08 (s, 1H), 7.76 (d, J = 8.8 Hz, 1H), 7.38 (d, J = 7.2 Hz, 1H),
3.97 (dd, J = 18.1, 6.6 Hz, 2H), 1.96 (dt, J = 13.5, 6.8 Hz, 1H), 1.05 (d,
J = 6.6 Hz, 3H). HPLC–MS m/z (M+) 334.9; Tr = 6.33 (>95%).
Yield: 77%; white powder. 1H NMR (300 MHz, CD3OD) d 8.00–
7.65 (m, 2H), 7.34 (d, J = 8.9 Hz, 1H), 3.72 (t, J = 6.6 Hz, 2H), 3.2–
2.89 (m, 2H), 2.71 (t, J = 7.0 Hz, 2H), 2.17–1.93 (m, 2H), 1.93–1.73
(m, 2H). MS m/z (M+Na+) 342.0; Tr = 3.45 (>95%).
4.19. (6-Trifluoromethoxybenzothiazol-2-yl)-carbamic acid
hexyl ester (16)
4.11. 2-(1-Aminomethyl-cyclohexyl)-N-(6-trifluoromethoxy-
benzothiazol-2-yl)-acetamide (25)
Yield: 72%; white powder. 1H NMR (300 MHz, DMSO-d6) d 8.12
(s, 1H), 7.79 (d, J = 8.8 Hz, 1H), 7.42 (d, J = 7.4 Hz, 1H), 4.24 (t,
J = 6.7 Hz, 2H), 1.67 (dd, J = 14.2, 6.8 Hz, 2H), 1.35 (m, 6H), 0.91
(t, J = 6.7 Hz, 3H). HPLC–MS m/z (M+) 362.9; Tr = 6.97 (95%).
Yield: 34%; buff powder. 1H NMR (300 MHz, DMSO-d6) d 12.70 (s,
1H), 8.88 (s, 1H), 8.14 (s, 1H), 7.97 (s, 1H), 7.85 (d, J = 8.8 Hz, 1H), 7.44
(d, J = 8.8 Hz, 1H), 3.01 (d, J = 5.6 Hz, 4H), 2.90 (s, 3H), 2.80 (s, 3H),
2.74 (s, 3H), 1.47 (d, J = 8.8 Hz, 3H). MS m/z (MH+) 388.0; Tr = 4.03
(96%).
4.20. (6-Trifluoromethoxybenzothiazol-2-yl)-carbamic acid 2-
dimethylamino-ethyl ester (17)
4.12. 2-(2-Nitrophenyl)-N-(6-trifluoromethoxybenzothiazol-2-
yl)-acetamide (26)
Yield: 13%; white powder. 1H NMR (300 MHz, CD3OD) d 7.85 (s,
1H), 7.75 (d, J = 9.1 Hz, 1H), 7.42–7.27 (m, 1H), 4.86 (s, 6H),
4.64 (m, 2H), 3.58 (m, 2H). HPLC–MS m/z (M+) 349.8;
Tr = 3.58 (>95%).
Yield: 38%; white powder. 1H NMR (300 MHz, DMSO-d6) d 12.79
(s, 1H), 8.12 (d, J = 8.2 Hz, 2H), 7.84 (d, J = 8.8 Hz, 1H), 7.75 (d,
J = 7.4 Hz, 1H), 7.62 (d, J = 5.6 Hz, 2H), 7.42 (d, J = 8.7 Hz, 1H),
4.30 (s, 2H). MS m/z (M+) 397.9; Tr = 5.77 (>95%).
4.21. (6-Trifluoromethoxybenzothiazol-2-yl)-carbamic acid 3-
(dimethylamino)propyl ester (18)
4.13. General synthetic method for carbamate analogs 11–18
Yield: 14%; beige powder. 1H NMR (300 MHz, DMSO-d6) d 10.33
(s, 1H), 8.10 (s, 1H), 7.77 (d, J = 8.8 Hz, 1H), 7.39 (d, J = 8.7 Hz, 1H),
4.26 (dd, J = 18.9, 12.6 Hz, 2H), 3.16 (d, J = 15.6 Hz, 2H), 2.76 (d,
J = 5.0 Hz, 6H), 2.30–1.77 (m, 2H). HPLC–MS m/z (M+) 363.9;
Tr = 3.71 (>95%).
6-Trifluoromethoxybenzothiazol-2-ylamine (1, 100 mg, 0.42
mmol), chloroformate (0.74 mmol), and triethylamine (64 mg,
0.64 mmol) were combined in methylene chloride (3 mL) and stir-
red 24 h at ambient temperature. The reaction was concentrated.
The residue was treated with methanol/water (1:1, 5 mL) and the

M. E. McDonnell et al. / Bioorg. Med. Chem. 20 (2012) 5642–5648
5647
4.22. General synthesis method of amide analogs 19–21
concentration of 0.01 mg/mL. The samples were thoroughly mixed
and maintained at 37 °C for 60 min. Each sample was injected con-
6-Trifluoromethoxy-benzothiazol-2-ylamine (1, 100 mg, 0.42
secutively onto an Agilent 1100 system (Luna C18, 3 lm,
mmol),
a
carboxylic acid (0.43 mmol), and EDCI (123 mg,
50 mm ꢀ 3 mm; 1 mL/min; mobile phase of 0.1% trifluoroacetic
acid in water/0.1% trifluoroacetic acid in acetonitrile) after a
60 min period. The percent prodrug conversion was calculated by
comparing the area of prodrug compound versus riluzole gener-
ated. The identities of the parent compounds and conversion prod-
ucts were confirmed by LC/MS.
0.65 mmol) were combined in methylene chloride (5 mL) and stir-
red 4 days. The reaction was washed with 0.1 N HCl (2 ꢀ 10 mL),
dried over magnesium sulfate, filtered, and then concentrated.
Chromatography on silica gel afforded compounds 20, and 21.
Compound 21 was dissolved in dichloromethane and trifluoroace-
tic acid (1:1). After stirring 2 h the solvents were removed in vacuo
to yield compound 19.
5.2. Plasma stability
4.23. N-(6-Trifluoromethoxybenzothiazol-2-yl)-succinamic acid
(19)
Assessment of plasma stability was carried out by individual
incubations of drug candidates in fresh mouse or human control
plasma at a concentration of 1 lM for 1 h at 37 °C. After which,
Yield: 71%; white powder. 1H NMR (300 MHz, DMSO-d6) d 8.11 (s,
1H), 7.81 (d, J = 8.8 Hz, 1H), 7.41 (d, J = 8.8 Hz, 1H), 2.72 (d, J = 6.8 Hz,
2H), 2.68–2.44 (m, 2H). HPLC–MS m/z (M+) 334.8; Tr = 4.60 (>95%).
the samples were de-proteinized by addition of 2 volumes of ace-
tonitrile containing 0.1% formic acid and internal standard, vortex
mixed for 2 min and centrifuged at 4000 rpm for 10 min to pellet
precipitated protein. The resulting supernatants containing the
drug candidates were diluted fivefold with water containing 0.1%
formic acid and submitted to LC–MS/MS analysis. All determina-
tions were done in triplicate. Plasma stability was expressed as
percent of control remaining.
4.24. N-(6-Trifluoromethoxybenzothiazol-2-yl)-succinamic acid
methyl ester (20)
Yield: 77%; white powder. 1H NMR (300 MHz, DMSO-d6) d 12.54
(s, 2H), 8.12 (s, 1H), 7.82 (d, J = 8.8 Hz, 1H), 7.42 (d, J = 8.8 Hz, 1H),
3.60 (d, J = 9.2 Hz, 3H), 3.03–2.58 (m, 2H), 1.76 (t, J = 6.6 Hz, 2H).
HPLC–MS m/z (M+) 348.9; Tr = 5.23 (>95%).
5.3. Metabolic stability
4.25. N-(6-Trifluoromethoxybenzothiazol-2-yl)-succinamic acid
tert-butyl ester (21)
In vitro metabolic stability was determined in pooled mouse
or human liver microsomes (BD Gentest) at a protein concentra-
tion of 0.5 mg/mL in reaction buffer (100 mM KH2PO4, pH 7.4 and
12 mM MgCl2). Each drug candidate was added to a final concen-
Yield: 33%; light yellow powder. 1H NMR (300 MHz, DMSO-d6) d
12.50 (s, 1H), 8.10 (s, 1H), 7.81 (d, J = 8.6 Hz, 1H), 7.40 (m, 1H), 3.59
(t, J = 6.6 Hz, 2H), 1.75 (t, J = 6.6 Hz, 2H), 1.37 (s, 9H). HPLC–MS m/z
(MH+-C4H9) 334.8; Tr = 6.08 (95%).
tration of 1 lM. This mixture was pre-warmed to 37 °C for 10 min
prior to starting the reaction with the addition of b-nicotinamide
adenine dinucleotide 20-phosphate reduced (NADPH) to a final
concentration of 1 mM. A parallel incubation lacking NADPH
served as the control. After incubation for 30 min at 37 °C, the
reactions were quenched by the addition of acetonitrile contain-
ing 0.1% formic acid and internal standard, vortex mixed for
2 min and centrifuged at 4000 rpm for 10 min to pellet the pre-
cipitated protein. The resulting supernatant containing the drug
candidate and its potential metabolites was diluted fivefold with
water containing 0.1% formic acid and submitted to LC/MS/MS
analysis. Metabolic stability was expressed as percent of control
remaining.
4.26. Synthesis of amide analogs 22 and 23
Maleic anhydride (1 mmol) and an amine (1 mmol) were stirred
in DMF (1 mL) 24 h. 6-Trifluoromethoxy-benzothiazol-2-ylamine
(1, 1 mmol), HATU (1 mmol), and Et3N (1 mmol) were the reaction
was stirred an additional 24 h. The reactions were purified using
reverse phase chromatography to afford the desired compounds.
4.27. N-Pyridin-3-ylmethyl-N0-(6-trifluoromethoxybenzothi-
azol-2-yl)-succinamide (22)
Yield: 16%; beige powder. 1H NMR (300 MHz, DMSO-d6) d 12.50
(s, 1H), 8.84–8.42 (m, 2H), 8.10 (d, J = 7.4 Hz, 2H), 7.81 (d, J = 8.8 Hz,
2H), 7.71 (d, J = 7.9 Hz, 1H), 7.35 (dd, J = 36.7, 27.4 Hz, 1H), 4.40 (d,
J = 5.8 Hz, 2H), 2.78 (t, J = 6.7 Hz, 2H), 2.57 (t, J = 6.7 Hz, 2H). HPLC–
MS m/z (M+) 424.9; Tr = 3.49 (95%).
5.4. Cyp1A2 metabolic stability
Microsomes were purchased from Invitrogen as frozen 20 mg/
ml solutions. Two individual donor microsome lots were used
which had a wide range of Cyp1A2 activity as measured by Invitro-
gen using a phenacetin dealkylation activity assay. Lot# HU8045
was determined to have a Vmax for the demethylation of phenace-
4.28. N-(2-Morpholin-4-yl-ethyl)-N0-(6-trifluoromethoxyben-
zothiazol-2-yl)-succinamide (23)
tin of 94.7
lmol/min and Lot# HU8022 was determined to have a
Yield: 9%; white powder. 1H NMR (300 MHz, DMSO-d6) d 12.51
(s, 1H), 9.61 (s, 1H), 8.23 (s, 1H), 8.11 (s, 1H), 7.82 (d, J = 8.9 Hz, 1H),
7.42 (d, J = 8.9 Hz, 1H), 3.96 (s, 2H), 3.64 (s, 3H), 3.44 (m, 6H), 3.18
(m, 3H), 2.76 (d, J = 6.4 Hz, 2H), 2.53 (dd, J = 15.1, 4.8 Hz, 1H). MS
m/z (M+) 446.9; Tr = 3.66 (95%).
Vmax of 1190
l
mol/min. Individual microsome lots were used to
determine the metabolic disappearance of Riluzole and Riluzole
prodrug 9 using the incubation method from the Metabolic Stabil-
ity section above.
5.5. LC–MS/MS analysis
5. In vitro stability assays
An aliquot from each incubation was analyzed by LC/MS/MS
with SRM detection in the positive ionization mode using an ABS-
ciex API 5500 QTrap Mass Spectrometer interfaced via the ABSciex
Turbo V IonSpray source (ESI) to an Eksigent ExpressHT LC system.
Best peak shape and separation from interfering matrix species
5.1. Stability in Simulated Gastric Fluid (SGF) and Simulated
Intestinal Fluid (SIF)
The compounds were prepared in
appropriate test component (SGF, SIF) and acetonitrile to a final
a 9:1 mixture of the
was afforded by an Eksigent 3C18-CL-300, 3
l
, 50 ꢀ 1 mm column.

5648
M. E. McDonnell et al. / Bioorg. Med. Chem. 20 (2012) 5642–5648
9. Kretschmer, B. D.; Kratzer, U.; Schmidt, W. J. Naunyn-Schmiedebergs Arch
Pharmacol. 1998, 358, 181.
10. Wang, S. J.; Wang, K. Y.; Wang, W. C. Neuroscience 2004, 125, 191.
11. Namkoong, J.; Shin, S. S.; Lee, H. J., et al Cancer Res. 2007, 67, 2298.
12. Shuff, J. H.; Siker, M. L.; Daly, M. D.; Schultz, C. J. Clin Plas. Surg. 2010, 37, 147.
13. Trinh, V. A. Am. J. Health Syst. Pharm. 2008, 65, 3.
A fast gradient, from 15 to 85% organic in 2.5 min, with run time of
5.0 min, and flow rate of 50 L/min was utilized. Peak areas were
integrated using MultiQuant v2.0 software from ABSciex.
l
Acknowledgments
14. Sondak, V. K.; Flaherty, L. E. Nat. Rev. Clin. Oncol. 2011, 8, 513.
15. Luke, J. J.; Hodi, F. S. Clin. Cancer Res. 2012, 18, 9.
16. Bensimon, G.; Lacomblez, L.; Meininger, V. N. Engl. J. Med. 1994, 330, 585.
17. Yip, D.; Le, M. N.; Chan, J. L.-K.; Lee, J. H.; Mehnert, J. A.; Yudd, A.; Kempf, J.;
Shih, W. J.; Chen, S.; Goydos, J. S. Clin. Cancer Res. 2009, 15, 3896.
18. Groeneveld, G. J.; Van Kan, H. J. M.; Kalmijn, S., et al Neurology 2003, 61, 1141.
19. van Kan, H. J. M.; Groeneveld, G. J.; Kalmijn, S.; Spieksma, M.; van den Berg, L.
H.; Guchelaar, H. J. Brit. J. Clin. Pharmacol. 2004, 59, 310.
This work was funded by a Grant from the National Cancer
Institute of the National Institutes of Health (R43 CA156781-01).
Supplementary data
20. Sanderink, G. J.; Bournique, B.; Stevens, J.; Petry, M.; Martinet, M. J. Pharmacol.
Exp. Ther. 1997, 282, 1465.
21. Martinet, M.; Montay, G.; Rhodes, G. Drugs Today 1997, 33, 587.
22. Ettmayer, P.; Amidon, G. L.; Clement, B.; Testa, B. J. Med. Chem. 2004, 47, 2393.
23. Rautio, J.; Kumpulainen, H.; Heimbach, T., et al Nat. Rev. Drug Disc. 2008, 7, 255.
24. Testa, B. Biochem. Pharmacol. 2004, 68, 2097.
Supplementary data associated with this article can be found, in
the online version, at http: //dx.doi.org/10.1016/j.bmc.2012.
07.004.
25. Fredholt, K.; Mork, N.; Begtrup, M. Int. J. Pharm. 1995, 123, 209.
26. Nam, N.-H.; Kim, Y.; You, Y.-J.; Hong, D.-H.; Kim, H.-M.; Ahn, B.-Z. Bioorg. Med.
Chem. 2003, 11, 1021.
27. Sykes, B. M.; Atwell, G. J.; Hogg, A.; Wilson, W. R.; O’Connor, C. J.; Denny, W. A.
J. Med. Chem. 1999, 42, 346.
28. Di, L.; Kerns, E. H. Curr. Opin. Drug Disc. Devel. 2005, 8, 495.
29. Baudy, R. B.; Butera, J. A.; Abou-Gharbia, M. A.; Chen, H.; Harrison, B.; Jain, U.;
Magolda, R.; Sze, J. Y.; Brandt, M. R.; Cummons, T. A.; Kowal, D.; Pangalos, M.
N.; Zupan, B.; Hoffmann, U. M.; May, M.; Mugford, C.; Kennedy, J.; Childers, W.
E., Jr. J. Med. Chem. 2009, 52, 771.
References and notes
1. Cavalheiro, E. A.; Olney, J. W. Proc. Natl. Acad. Sci. USA 2001, 98, 5947.
2. Takano, T.; Lin, J. H.; Arcuino, G.; Gao, Q.; Yang, J.; Nedergaard, M. Nat. Med.
2001, 7, 1010.
3. Kalariti, N.; Pissimissis, N.; Koutsilieris, M. Expert Opin. Investig. Drugs 2005, 14,
1487.
4. Ackermann, J.; Beermann, F. Pigment Cell Res. 2005, 18, 315.
5. Rzeski, W.; Ikonomidou, C.; Turski, L. Biochem. Pharmacol. 2002, 64, 1195.
6. Rzeski, W.; Turski, L.; Ikonomidou, C. Proc. Natl. Acad. Sci. USA 2001, 98, 6372.
7. Stepulak, A.; Sifringer, M.; Rzeski, W., et al Proc. Natl. Acad. Sci. USA 2005, 102,
15605.
30. Jung, M. E.; Piizzi, G. Chem. Rev. 2005, 105, 1735.
31. Mandagere, A. K.; Thompson, T. N.; Hwang, K.-K. J. Med. Chem. 2002, 45, 304.
32. Assays performed by Absorption Systems, Inc. Exton, PA.
8. Doble, A. Neurology 1996, 47, 233S.