Novel imidazole-based histamine H3 antagonists

Article abstract of DOI:10.1016/j.bmcl.2008.11.114

A novel series of imidazole containing histamine H₃ receptor ligands were investigated and found to be potent functional antagonists. After improving the stability of these molecules towards liver microsomes, these compounds were found to have

Full text of DOI:10.1016/j.bmcl.2008.11.114

Bioorganic & Medicinal Chemistry Letters 19 (2009) 903–907
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Novel imidazole-based histamine H3 antagonists
*
Jill A. Jablonowski , Kiev S. Ly, Michael Bogenstaetter , Curt A. Dvorak, Jamin D. Boggs, Lisa K. Dvorak, Brian
Lord, Kirsten L. Miller, Curt Mazur , Sandy J. Wilson, Timothy W. Lovenberg, Nicholas I. Carruthers
Johnson & Johnson Pharmaceutical Research & Development, L.L.C., 3210 Merryfield Row, San Diego, CA 92121, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
A novel series of imidazole containing histamine H3 receptor ligands were investigated and found to be
potent functional antagonists. After improving the stability of these molecules towards liver microsomes,
these compounds were found to have no appreciable affinity for CYP P450s. Subsequent in vivo experi-
ments showed significant brain uptake of (4-chloro-phenyl)-[2-(1-isopropyl-piperidin-4-ylmethoxy)-3-
methyl-3H-imidazol-4-yl]-methanone 22.
Received 9 October 2008
Revised 25 November 2008
Accepted 26 November 2008
Available online 6 December 2008
Ó 2008 Elsevier Ltd. All rights reserved.
Keywords:
Histamine H3
Benzoylimidazole
Imidazole
Metabolic stability
Blood–brain barrier penetration
Biogenic amines accomplish a myriad of functions throughout
the body, where their interdependent relationships and varying
affinity for natural receptors and transporters allow these small
molecules to elicit dramatic physiological responses. Histamine
can affect conditions including allergic rhinitis through the hista-
mine H1 receptor1 and gastric acid secretion through the hista-
mine H2 receptor.2 The actions of histamine on the H3
receptor3 and the recently discovered H4 receptor4 are predomi-
nantly in the central nervous system (CNS) and immune systems,
respectively. Following its characterization by Schwartz,3 the
function of the H3 receptor was extensively evaluated. Later,
cloning of the H3 receptor allowed researchers to apply high
throughput screening techniques to this target to identify novel
structures.5 Since this discovery a number of pharmaceutical
companies have prepared small molecule H3 antagonists, some
of which have entered clinical trials for diseases such as ADHD,
excessive daytime sleepiness and allergic rhinitis.6
The early development of therapeutic agents acting at the
histamine H3 receptor was particularly challenging. In order to
maintain both affinity and efficacy, retention of the imidazole
moiety of the endogenous ligand was thought to be necessary.
Several of the first 1-H-4-substituted imidazole H3 antagonists,
thioperamide (2), ciproxifan (3) and clobenpropit (4), showed
unfavorable interactions with cytochrome (CYP) P450s.7,8 Initial
S
N
N
N
H
HN
Thioperamide
O
2
N
NH
S
N
N
O
HN
H
Cl
HN
Ciproxifan
3
Clobenpropit
4
approaches to replacing the imidazole included those from
Ganellin and coworkers9 and separately from the James Black
Foundation10 led to 5 and 6.
Cl
O
N
N
NH2
NH
N
NO2
6
5
JB 98064
UCL 1972
guinea pig H3 Ki = 2.0 nM
rat H3 Ki = 39 nM
High throughput screening of our corporate compound collection
identified several series of lead compounds including the imidazo-
pyridines (7), the indolizidines (8) and the N-methylimidazoles
(9) some of which have been previously described.11,12
* Corresponding author.
E-mail address: jjablono@its.jnj.com (J.A. Jablonowski).
Present address: The Boston Consulting Group Exchange Place 31[st] Floor
Boston, Massachussets 02109.
Present address: Antisense Drug Discovery, Isis Pharmaceuticals Inc., 1896
Rutherford Road, Carlsbad, CA 92008, USA.
Of these templates, 9 represented a potent imidazole containing
functional H3 antagonist. In this structure the tri-substituted imid-
azole is presumably behaving in a different capacity to the imidaz-
ole present in compounds based on the endogenous ligand.13 Thus
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.11.114

904
J. A. Jablonowski et al. / Bioorg. Med. Chem. Lett. 19 (2009) 903–907
ing an additional methylene and replacing the Cl with a Br shows
n-Bu
n-Bu
N
N
significantly reduced affinity with Ki = 70 nM. This result suggests
that the longer chain length is responsible for the diminished bind-
ing affinity, assuming that the change from Br to Cl has little effect
on binding affinity. This assumption is based on a study of the ben-
zoyl substituents in the single methylene analogs (Table 5). In this
study no significant changes in affinity were observed over a wide
range of substituents including halogens, electron-withdrawing
and electron donating groups, Table 5, with one exception.16
Synthesis: The H3 ligands presented here were prepared via sev-
eral routes as shown in Schemes 1–6.17 Intermediate 34 was pre-
pared in 66% yield by treatment of 2-thio-1-methylimidazole
(33) with t-BuLi (2 equiv) followed by reaction with 4-chlorobenz-
aldehyde as described in Scheme 1.18,19 Alkylation of 34 with
1-bromo-3-chloropropane followed by oxidation of the benzyl
alcohol with MnO2 provided alkyl chloride 35, which was subse-
quently converted to analogs 9 and 14 by reaction with the appro-
priate amine. Along similar lines, intermediate 34 was also
converted to compounds 13 and 16. Benzyl ketone 9 was converted
to its corresponding oxime 10 via treatment with NH2OH/HCl and
reduced with either NaBH4 or under Wolff–Kischner conditions to
provide analogs 11 and 12, respectively. Intermediate 34 was also
converted to sulfone 39 which led to compounds 18 and 22.20
An alternative route to the sulfur and oxygen linked analogs is
shown in Scheme 2.21 Treatment of commercially available
N-methylimidazole (36) with n-BuLi followed by diphenyldisulfide
gave 1-methyl-2-phenylsulfanyl-imidazole. When this compound
was subjected to LTMP (lithium 2,2,6,6-tetramethylpiperidine) in
a mixture of THF and DME at À78 °C for 15 min followed by
quenching with 4-chlorobenzaldehye, 2-thiophenylimidazole
intermediate 37 was prepared. Stepwise oxidation of 37 with
MnO2 followed by MCPBA gave sulfone 39. Both sulfur (15 and
17) and oxygen (21) linked analogs were prepared from this
intermediate.
S
O
N
N
hH3 Ki = 4 µM
7
O
hH3 Ki = 12 nM
8
H
N
H
N
S
N
Cl
N
hH3 Ki = 5 nM
9
we set out to explore the structure activity relationship (SAR)
around this lead and establish whether the structure was free of
the aforementioned liabilities and therefore a viable drug template.
Results: Our investigations focused on optimizing the key frag-
ments required for biological activity.14 We found that the car-
bonyl could be converted to an oxime (10) without loss of
affinity but that the corresponding alcohol 11 and methylene 12
exhibited reduced affinity, Table 1.
Replacement of the terminal nitrogen with a carbon (13) elim-
inated affinity, however replacing the dimethylamino group with a
piperidinyl group (14), or introducing a conformationaly restricted
linker (15) provided compounds with similar affinity for the recep-
tor, a result that is consistent with that observed in our previously
described efforts.15 However, either shortening, compare 9 with
16, or lengthening the distance, compare 14 with 17, between
the imidazole ring and the terminal amine led to a reduction in
affinity (Table 2).
While many of the analogs we prepared, including the HTS hit
(9), showed high affinity for the H3 receptor, many were unstable
when exposed to rat liver microsomes (RLM), as shown in Table
3. Interestingly these compounds proved to be much more stable
in the presence of human liver microsomes (HLM). For example
thiopropylpiperidine analog 14 was readily degraded in the pres-
ence of RLM with a t1/2 = 7 min and displayed a HLM t1/2
=
Table 2
149 min. Since our in vivo models are rodent based, we hoped to
increase the stability of these molecules to RLM by changing the
atom linking the basic amine tail to the aromatic core of the mol-
ecule. Thus we prepared the oxygen (18), nitrogen (19) and carbon
linked (20) analogs.
Binding and functional antagonist data for the human histamine H3 receptor for
compounds 13–17
O
N
R
Of these analogues, the oxygen linked compound 18 offered the
optimal combination of affinity and stability. Since replacing sulfur
with oxygen reduced the distance between the terminal amine and
the benzoylimidazole we prepared several compounds to explore
this area, Table 4. For this study we focused our attention on mod-
ifications to the isopropylpiperidine 15.
Replacing the sulfur linkage of 15 with oxygen afforded 21,
which did not significantly alter affinity. Insertion of one methy-
lene into the linker of compound 21 provided analog 22 and these
two compounds were found to be equipotent. Compound 23 bear-
N
Cl
b
Compound
R
hH3 Kia (nM)
hpA2
13
>5000
5
S
S
14
15
16
6 ( 2)
2 ( 0.3)
71 ( 11)
50d
7.7c
9.3
7.3
nd
N
Table 1
Binding and functional antagonist data for the human histamine H3 receptor for
compounds 9–12
N
N
S
S
X
N
S
N
N
Cl
b
Compound
X
hH3 Kia (nM)
hpA2
9
O
5 ( 0.8)
2 ( 0.05)
18 ( 5)
20 ( 1)
8.7c
8.4
17
N
10
11
12
N(OH)
OH, H
H, H
S
8.4c
8.2
a
Values are means of at least three experiments in triplicate, standard error of
the mean is given in parentheses.
a
Values are means of at least three experiments in triplicate, standard error of
the mean is given in parentheses.
b
Values are based on a single determination unless stated otherwise.
b
c
Values are based on a single determination unless stated otherwise.
Average of two runs.
Average of two runs.
Single determination.
c
d

J. A. Jablonowski et al. / Bioorg. Med. Chem. Lett. 19 (2009) 903–907
905
Table 3
Binding and functional data for the human histamine H3 receptor and half-life (t1/2) as predicted by liver microsomal assays for compounds 14, 18–20
O
N
R
N
Cl
b
Compound
R
hH3 Kia (nM)
hpA2
RLM stability t1/2 (min)
HLM stability t1/2 (min)
14
6 ( 2)
7.7c
7
149
S
N
N
N
N
18
19
14 ( 1.5)
50 ( 3.5)
24 ( 9)
7.9c
7.4
55
4
>180
>180
179
O
H
N
20
8.0c
4
a
Values are means of at least three experiments in triplicate, standard error of the mean is given in parentheses.
Values are based on a single determination unless stated otherwise.
Average of two runs.
b
c
Table 4
Table 5
Binding and functional antagonist data for the human histamine H3 receptor for
compounds 15, 21–23
Binding and functional data for the human histamine H3 receptor for compounds 22,
24–32
O
N
R
O
N
N
O
N
X
X
N
b
Compound
X
R
hH3 Kia (nM)
hpA2
b
Compound
X
hH3 Kia (nM)
hpA2
22
24
25
26
27
28
29
30
31
32
4-Cl
H
2-Cl
3-Cl
3,5-di Cl
4-Br
4-CF3
4-NO2
4-F
3 ( 1)
9.2c
9.5
N
15
Cl
2 ( 0.3)
9.3
1 ( 0.3)
1 ( 0.1)
5 ( 0.6)
4 ( 0.3)
3 ( 0.9)
4 ( 0.6)
4 ( 1)
11.1d
9.7
S
9.0
21
22
Cl
Cl
Br
10 ( 5)
3 ( 1)
8.2c
9.3d
7.1
N
9.1
9.3c
9.0d
9.5
O
O
3 ( 0.4)
9 ( 2)
4-iPr
8.9
N
a
Values are means of at least three experiments in triplicate, standard error of
the mean is given in parentheses.
b
Values are based on a single determination unless stated otherwise.
N
23
70 ( 13)
c
Average of four runs.
d
Average of three runs.
O
a
Values are means of at least three experiments in triplicate, standard error of
the mean is given in parentheses.
b
Values are based on a single determination unless stated otherwise.
Average of three runs.
The nitrogen linked compound 19 was prepared using the pre-
viously described 2-Cl-imidazole intermediate (42) as shown in
Scheme 5.22
c
In order to evaluate the effect of substitution on the benzoyl
ring, we prepared intermediate 43 as shown in Scheme 6. This
compound was then converted into analogs 25–27 and 29–32.
The low yields shown in Scheme 6 can be attributed to incomplete
reaction in step d, as we recovered a significant amount of imidaz-
ole 43.
Biological results and discussion: Compound 22 was profiled in
detail, in particular with respect to establishing whether the struc-
ture suffered from the aforementioned liabilities often associated
with imidazoles. When exposed to RLM and HLM, compound 22
was found to be quite stable with t1/2 > 180 min in both RLM and
Compounds 23 and 28 were prepared using similar chemistry,
from intermediate 40 as shown in Scheme 3. In addition, com-
pound 28 was converted to 24 under hydrogenation conditions.
Deprotonation of N-methylimidazole (36) followed by quench-
ing with 1-chloro-3-iodobutane and subsequent treatment of the
alkyl chloride intermediate with piperidine provided compound
41. The benzoyl moiety was then installed by deprotonation with
LTMP at À78 °C, and subsequent reaction with 4-chlorobenzalde-
hyde. Oxidation with MnO2 gave compound 20 as shown in
Scheme 4.

906
J. A. Jablonowski et al. / Bioorg. Med. Chem. Lett. 19 (2009) 903–907
O
N
O
N
k
O
N
18
N
SO2n-Pr
Cl
N
Cl
O
N
39
N
O
l
N
h, i, j
Cl
22
OH
O
O
N
N
SH
SH
N
N
S
S
Cl
NR2
a
b, c
d
e
N
N
N
N
10
11
12
Cl
Cl
Cl
33
34
f
35
9 NR2 = N(CH3)2
g
N
m, c
n , c
14 NR2
=
13
16
Scheme 1. Reagents and conditions: (a) i—t-BuLi (2 equiv), THF, À78 °C for 15 min then warm to 0 °C for 30 min; ii—4-Cl-PhCHO, THF, À78 °C, 66%; (b) K2CO3, Br-(CH2)3-Cl,
DMF, acetone, 69%; (c) MnO2, DCM, rt, 1 h, (81% for 35), (86% for 13), (87% for 16); (d) HNR2, K2CO3, acetone, 60 °C, 16 h, (94% for 14), (6% for 9); (e) NH2OH-HCl, CH3CH2OH,
pyridine, 80 °C, 16 h, 14%; (f) NaBH4, CH3OH, 16 h, 60%; (g) n-butanol, KOtBu, NH2NH2, 120 °C, 16 h, 45%; (h) Br(CH2)2CH3, K2CO3, DMF, 71%; (i) MnO2, DCM, 16 h, 29%; (j) m-
CPBA, DCM, rt, 93%; (k) 3-piperidin-1-yl-propan-1-ol, NaH, THF, 0 °C–rt, 16 h, 52%; (l) (1-isopropyl-piperidin-4-yl)-methanol, NaH, THF, 0 °C–rt, 16 h, 51%; (m) K2CO3, Cl-
(CH2)3CH(CH3)2, DMF, acetone, 23%; (n) K2CO3, Cl-(CH2)2N(CH3)2, acetone, 69%.
e
15
O
O
H
N
O
OH
N
N
Cl
a
N
N
N
SO2Ph
SPh
f
a, b
Cl
c, d
Cl
17
21
N
36
N
N
Cl
N
Cl
g
42
19
37
39
Scheme 5. Reagents and conditions: (a) 3-Piperidin-1-yl-propylamine, DIPEA, THF,
reflux, 16 h, 16%.
Scheme 2. Reagents and conditions: (a) i—n-BuLi, THF, 20 min, À78 °C; (ii) Ph2S2,
À78 °C for 15 min then warm to rt over 45 min, 82%; (b) (i) 2,2,6,6-tetramethyl-
piperidine (TMP), THF, DME, n-BuLi, À78 °C, 15 min; (ii) 4-Cl-PhCHO,THF, À78 °
C–rt, 16 h, 69%; (c) MnO2, DCM, rt, 1 h, 87%; (d) m-CPBA, diethyl ether, rt, quant.; (e)
1-isopropyl-piperidine-4-thiol, THF, NaH, 0 °C–rt, 16 h, 5%; (f) i—1,3-dioxolane-2-
propanethiol, NaH, THF, 78%; ii—PPTS, acetone, H2O, reflux, 18 h, quant.;
iii—piperidine, Na(OAc)3BH, HOAc, DCE, 97%; (g) 1-isopropyl-piperidin-4-ol, NaH,
THF, 0 °C—rt, 16 h, 54%.
O
N
N
N
O
N
O
a, b, c,
d, e
33
R
N
N
43
25 R = 2-Cl (12%)
26 R = 3-Cl (5%)
27 R = 3,5-diCl (27%)
29 R = 4-CF3 (10%)
30 R = 4-NO2 (7%)
31 R = 4-F (5%)
e
O
23
28
N
SO2n-Pr
N
SH
a, b, c, d
f
N
N
Br
g
24
32 R = 4-iPr (5%)
33
40
Scheme 6. Reagents and conditions: (a) i—n-BuLi, THF, 20 min, À78 °C; ii—Ph2S2,
À78 °C for 15 min then warm to rt over 45 min, 82%; (b) m-CPBA, CH2Cl2, 0 °C–rt,
82%; (c) 2-(1-isopropyl-piperidin-4-yl)-methanol, NaH, THF, reflux, 36 h, 63%; (d)
i—2,2,6,6-tetramethylpiperidine, THF, DME, n-BuLi, À78 °C; ii—then add substituted
benzaldehyde) in THF and let warm to rt over 16 h; (e) MnO2, CH2Cl2. Yield shown is
the yield after steps d and e.
Scheme 3. Reagents and conditions: (a) i—t-BuLi (2 equiv), THF, À78 °C for 15 min
then warm to 0 °C for 30 min; ii—4-Br-PhCHO, THF, À78 °C, 57%; (b) Br(CH2)2CH3,
K2CO3, DMF, 95%; (c) MnO2, DCM, 16 h, 86%; (d) m-CPBA, DCM, rt, quant; (e) 2-(1-
isopropyl-piperidin-4-yl)-ethanol, NaH, THF, 0 °C–rt, 16 h, 67%; (f) 2-(1-isopropyl-
piperidin-4-yl)-methanol, NaH, THF, 0 °C–rt, 16 h, 51%; (g) H2, Pd/C, ethanol, 90%.
HLM assays. Additionally, analog 22 was evaluated as a substrate
for the major cytochrome P450 enzymes (1A2, 2C9, 2C19, 2D6
and 3A4) and exhibited IC50s > 10 lM in every case. In functional
assays, 22 was approximately ten fold less potent at the rat recep-
tor when compared to the human receptor (rat pA2 = 8.0 vs human
pA2 = 9.2) and in broad selectivity screening (>50 receptors and/or
ion channels) 22 only showed weak affinity for the sigma receptor
O
N
N
3
N
N
3
a, b
c, d
36
N
N
Cl
20
41
Scheme 4. Reagents and conditions: (a) i—n-BuLi, THF, 30 min, À78 °C; ii—
Cl(CH2)4I, THF, À78 °C for 3 h then warm to rt over 16 h, 38%; (b) piperidine,
K2CO3, DMF, 53%; (c) i—2,2,6,6-tetramethylpiperidine, THF, n-BuLi, À78 °C, 3 h; ii—
4-Cl-PhCHO, THF, À78 °C–rt, 8 h, 56%; (d) MnO2, DCM, 16 h, 29%.
(77% at 10
When22was administered to rats (10 mg/kg, ip) maximumbrain
and plasma concentrations of 17 and 2.6 M, respectively, were ob-
lM).
l

J. A. Jablonowski et al. / Bioorg. Med. Chem. Lett. 19 (2009) 903–907
8. Yang, R.; Hey, J. A.; Aslanian, R.; Rizzo, C. A. Pharmacology 2002, 66, 128.
907
served after 60 min. Significant amounts of 22 (3.8 lM) were also
9. Ganellin, C. R.; Leurquin, F.; Piripitsi, A.; Arrang, J.-M.; Garbarg, M.; Ligneau, X.;
Schunack, W.; Schwartz, J.-C. Arch. Pharm. (Weinheim) 1998, 331, 395.
10. Linney, I. D.; Buck, I. M.; Harper, E. A.; Kalindjian, S. B.; Pether, M. J.; Shankley,
N. P.; Watt, G. F.; Wright, P. T. J. Med. Chem. 2000, 43, 2362.
11. Shah, C.; McAtee, L.; Breitenbucher, J. G.; Rudolph, D.; Li, X.; Lovenberg, T. W.;
Mazur, C.; Wilson, S. J.; Carruthers, N. I. Bioorg. Med. Chem. Lett. 2002, 12, 3309.
12. (a) Bonaventure, P.; Letavic, M.; Dugovic, C.; Wilson, S.; Aluisio, L.; Pudiak, C.;
Lord, B.; Mazur, C.; Kamme, F.; Nishino, S.; Carruthers, N.; Lovenberg, T.
Biochem. Pharmacol. 2007, 73, 1084; (b) Stocking, E. M.; Letavic, M. A. Curr. Top.
Med. Chem. 2008, 8, 988.
13. Kovalainen, J. T.; Christiaans, J. A. M.; Ropponen, R.; Poso, A.; Peraekylae,
M.; Vepsaelaeinen, J.; Laatikainen, R.; Gynther, J. J. Am. Chem. Soc. 2000,
122, 6989.
14. Barbier, A. J.; Berridge, C.; Dugovic, C.; Leposky, A. D.; Wilson, S. J.; Boggs, J.;
Aluisio, L.; Lord, B.; Mazur, C.; Pudiak, C. M.; Langlois, X.; Xiao, W.; Apodaca, R.;
Carruthers, N. I.; Lovenberg, T. W. Br. J. Pharmacol. 2004, 143, 649.
15. Dvorak, C. A.; Apodaca, R.; Barbier, A. J.; Berridge, C. W.; Wilson, S. J.;
Boggs, J. D.; Xiao, W.; Lovenberg, T. W.; Carruthers, N. I. J. Med. Chem.
2005, 48, 2229.
seen as long as 48 h after dosing. A rat pharmacokinetic study
(10 mg/kg, po, iv) provided additional in vivo parameters such as
an oral half-life (po t1/2) = 9 h, oral bioavailability (%F) = 10%, clear-
ance (CL) = 9.3 ml/min/kg and volume of distribution (Vd) = 5.2 L/kg.
These observations indicate that 22 was free of the metabolic issues
which have been observed of many imidazole-based ligands.
In summary we investigated a novel series of imidazole con-
taining histamine H3 antagonists. They were found to be potent
functional antagonists and we were able to significantly improve
their stability towards liver microsomes. Following in vivo admin-
istration compound 22 achieved high brain concentration.
References and notes
1. Barger, G.; Dale, H. H. J. Physiol. 1910, 41, 19.
16. Compound 25 exhibited a disconnect between receptor binding affinity and
functional activity for which we have no explanation.
17. Bogenstaetter, M.; Carruthers, N. I.; Jablonowski, J. A.; Lovenberg, T. W.; Ly, K. S.
Chem. Abstr. 2002, 137, 279192. WO 2002079168, 2002.
18. Ohta, S.; Yamamoto, T.; Kawasaki, I.; Yamashita, M.; Katsuma, H.; Nasako, R.;
Kobayashi, K.; Ogawa, K. Chem. Pharm. Bull. 1992, 40, 2681.
2. Black, J. W.; Duncan, W. A. M.; Durant, G. J.; Ganellin, C. R.; Parsons, M. E. Nature
1972, 236, 385.
3. Arrang, J.-M.; Garbarg, M.; Schwartz, J.-C. Nature 1983, 302, 832.
4. Liu, C.; Ma, X.-J.; Jiang, X.; Wilson, S. J.; Hofstra, C.; Blevitt, J.; Pyati, J.; Li, X.;
Chai, W.; Carruthers, N.; Lovenberg, T. W. Mol. Pharmacol. 2001, 59, 420.
5. Lovenberg, T. W.; Roland, B. L.; Wilson, S. J.; Jiang, X.; Pyati, J.; Huvar, A.;
Jackson, M. J.; Erlander, M. G. Mol. Pharmacol. 1999, 55, 1101.
6. Celanire, S.; Wijtmans, M.; Talaga, P.; Leurs, R.; de Esch, I. J. P. Drug Disc. Today
2005, 10, 1613.
19. Phillips, B. T.; Claremon, D. A.; Varga, S. L. Synthesis 1990, 761.
20. Jarosinski, M. A.; Anderson, W. K. J. Org. Chem. 1991, 56, 4058.
21. Shapiro, G.; Marzi, M. Tetrahedron Lett. 1993, 34, 3401.
22. Mani, N. S.; Jablonowski, J. A.; Jones, T. K. J. Org Chem. 2004, 69, 8115.
7. Murray, M. Drug Metab. Rev. 1987, 18, 55.