An efficient synthesis of carbazole-based secretory phospholipase A 2 (sPLA2) inhibitors LSN433771 and LSN426891

Article abstract of DOI:10.1016/j.tetlet.2005.12.043

The flexible and efficient synthesis of two structurally similar carbazole derivatives is described. This general strategy features an intramolecular palladium-mediated biaryl coupling reaction to join two aromatic domains of the target molecules. Formati

Full text of DOI:10.1016/j.tetlet.2005.12.043

Tetrahedron
Letters
Tetrahedron Letters 47 (2006) 1351–1353
An efficient synthesis of carbazole-based secretory phospholipase
A2 (sPLA2) inhibitors LSN433771 and LSN426891
Scott A. May,* Thomas M. Wilson and Allison L. Fields
Chemical Product Research and Development, Eli Lilly and Company, Indianapolis, IN 46285-4813, USA
Received 24 October 2005; revised 6 December 2005; accepted 7 December 2005
Abstract—The flexible and efficient synthesis of two structurally similar carbazole derivatives is described. This general strategy fea-
tures an intramolecular palladium-mediated biaryl coupling reaction to join two aromatic domains of the target molecules. Forma-
tion of the carbazole core is accomplished via nitrene insertion. The synthesis of secretory phospholipase A2 (sPLA2) inhibitors
LSN433771 (1) and LSN426891 (2) is detailed.
Ó 2005 Elsevier Ltd. All rights reserved.
Phospholipase A2 is an enzyme critical to the formation
of arachidonic acid from membrane phospholipids as
part of the body’s inflammatory response cascade.1 Ele-
vated levels of human non-pancreatic secretory phos-
pholipase A2 (sPLA2) have been observed in patients
suffering from a number of conditions including acute
pancreatitis, respiratory distress syndrome, bacterial
peridonitis, and septic shock.2 In conjunction with a
program aimed at identifying selective sPLA2 inhibitors,
two molecules of particular interest were identified, car-
bazoles LSN433771 (1) and LSN426891 (2).
NH2
HO
O
NH2
O
R2O
O
O
O
O
N
NO2
R1
R2
R1
1, (LSN433771) , R1 = H; R2 = C6H12 3a, R1 = H; R2 = Me
2, (LSN426891), R1 = F; R2 = Ph
3b, R1 = F; R2 = Et
O
O
O
OH
HO
The initial synthetic routes toward 1 and 2 were inde-
pendent and somewhat labor intensive. Since both com-
pounds were required on large scale for toxicology
studies, an efficient and unified strategy was highly desir-
able. In this letter, we report such a unified strategy for
synthesis of both LSN433771 (1) and LSN426891 (2)
starting from 2-bromo-3-nitrobenzoic acid (6) and an
appropriately substituted phenol (5).
Br
NO2
NO2
R1
R1
5a, R1 = H
5b, R1 = F
4a, R1 = H
4b, R1 = F
6
Figure 1. Retrosynthetic analysis.
functionalized biaryls 3 could be converted into the car-
bazole core structure. This method allows for the two
aryl portions of the molecule to be assembled in the cor-
rect oxidation state. Formation of the fully functional-
ized biaryls (3) was thought to occur via amination/
alkylation of benzocoumarin 4. In this fashion, no pro-
tecting groups are required for elaboration to 1 and 2.
The synthesis of benzocoumarin could be formed
through intramolecular biaryl coupling of the corre-
sponding aryl 2-bromo-3-nitrobenzoate, which is simply
constructed by coupling the appropriate phenol 5 with
carboxylic acid 6.
The goal of this new synthetic route toward 1 and 2 was
to incorporate flexibility into the synthetic design such
that the same general strategy could be applied to both
molecules. From a processing standpoint, this new route
focused on minimizing the use of protecting groups as
well as isolation of products via crystallization in lieu
of chromatography. The strategy that was envisioned
(Fig. 1) focused on a nitrene insertion reaction wherein
*
Corresponding author. Tel.: +1 317 433 3687; fax: +1 317 276
4507; e-mail: may_scott_a@lilly.com
0040-4039/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2005.12.043

1352
S. A. May et al. / Tetrahedron Letters 47 (2006) 1351–1353
We sought to examine the intramolecular biaryl coupling
using conditions originally described by Ames and
Opalko.3 The required benzoates (7a,b) were prepared
in straightforward fashion via DCC coupling reactions
between 2-bromo-3-nitrobenzoic acid4 and phenol or
4-fluorophenol. The corresponding aryl benzoates 7a
(92%) and 7b (80%) were obtained in good yield after
crystallization from the crude reaction mixtures. The
initial coupling conditions [Pd(OAc)2, DMA, NaOAc,
PPh3, 170 °C, 2 h] with benzoate 4a afforded the desired
coumarin 4b in ꢀ50% isolated yield. While this result was
encouraging, especially in light of the modest yields
reported in the literature for these types of reactions,5
the yield was still modest. It was subsequently found that
simply removing the phosphine ligand and lowering the
temperature to 125 °C allowed for a higher yield of the
desired product. Under more optimized conditions, 7a
and 7b afforded the corresponding benzocoumarins 4a
(76%) and 4b (73%) after crystallization (Scheme 1).
With biaryl amides 3a and 3b in hand, two strategies for
carbazole formation were considered (Fig. 2). Path ‘a’
featured a Cadogen-type6 cyclization wherein heating
3a or 3b in the presence of a phosphite would afford
the carbazole core structure 10 directly via in situ nitrene
formation. Path ‘b’ featured conversion to and isolation
of azides 11 followed by thermal decomposition to
afford 10.
Since the Cadogen-type reaction appeared less lengthy,
this was examined first. Accordingly, compounds 3a
and 3b were heated to 160 °C in trimethylphosphite.
After several hours of reaction, HPLC analysis showed
near complete consumption of the starting materials
and numerous products. The same poor results were
obtained when triethylphosphite was used. However,
when the reactions were conducted in triphenylphosph-
ite [180 °C for 16 h] HPLC analysis revealed a relatively
clean reaction profile (Scheme 3). Isolation of the new
product and careful evaluation indicated that they were
not the desired products 10a and 10b, but rather dehy-
dration products 12a (40%) and 12b (56%). Despite
extensive efforts to improve the yield of this reaction,
40–55% was the optimal result in both cases and isola-
tion was challenging.
Elaboration of benzocoumarins 4a and 4b into fully
functionalized biaryls 3a and 3b is described in Scheme
2. Treatment of 4a,b with anhydrous ammonia in meth-
anol afforded phenoxy amides 8a,b, which were typically
processed as crude solutions after workup. Smooth
alkylation with ethyl or methylbromoacetate provided
biarlys 3a (74%) and 3b (77%) after simply treating the
crude reaction mixture with water and isolating the pre-
cipitate by filtration (Scheme 2). Another more stream-
lined approach involved direct treatment of the crude
benzocoumarin (4a,b) solution in DMF from the biaryl-
coupling with anhydrous ammonia. The resulting
hydroxyamides were then processed by removal of the
excess ammonia and in situ alkylation as described
above. For example, phenyl 2-bromo-3-nitrobenzoate
7a was converted into the fully functionalized biaryl 3a
in 75% yield with a single isolation.
R2O
O
CONH2
O
1 & 2
N
H
R1
a; R = H, R = Me
10
10
1
2
Path a
NH2
b; R = F, R = Et
2
2
O
R1
NO2
O
Path b
NH2
O
O
CO2R2
O
3a; R1 = H, R2 = Me
3b; R2 = F, R2 = Et
O
R
O
R
R1
Br
N3
O
NO2
NO2
CO2R2
7a, R = H
7b, R = F
4a, R = H
4b, R = F
11a; R1 = H, R2 = Me
11b; R2 = F, R2 = Et
Scheme 1. Reagents and conditions: Pd(OAc)2, DMF, DMA, NaOAc
(73–76%).
Figure 2. Strategies for carbazole formation.
NH2
NH2
O
O
O
O
R
a
R1
b
R1
NO2
NO2
NO2
OH
O
CO2R2
8a; R1 = H, R2 = Me
8b; R1 = F, R2 = Et
4a; R = H
4b; R = F
3a; R1 = H, R2 = Me
3b; R1 = F, R2 = Et
Scheme 2. Reagents and conditions: (a) NH3, MeOH, rt, 1 h; (b) methyl or ethylbromoacetate, K2CO3, DMF, rt, 15 min, 74–77%.

S. A. May et al. / Tetrahedron Letters 47 (2006) 1351–1353
1353
Intermediates 12a,b were alkylated at nitrogen with
NH2
NH2
(bromomethyl)cyclohexane to afford 13a (90%) and with
benzylbromide to afford 13,b (96%). Carbazoles 13a,b
were converted directly to the target molecule products
LSN433771 (1, 98%) and LSN426891 (2, 86%) under
the conditions described by Hall (Scheme 3).7
O
O
b
a
NH2
NO2
O
O
CO2Me
CO2Me
3a
14
The amide dehydration that was observed in Scheme 2
(3a,b!12a,b) is likely promoted by triphenylphosphate,
which is generated during the nitrene formation. Mecha-
NH2
8
nistically, this would be similar to that of POCl3 and
MeO
O
O
CONH2
likely enabled by the vigorous reaction conditions
(180 °C, 18 h). To our knowledge, however, this type of
dehydration of primary amides in the presence of phos-
phates alone has not been reported in the literature.9
O
c
N3
O
N
H
CO2Me
10a
Since the Cadogen-based pathway suffered from long
reaction times and low yields, the thermal decomposi-
tion of the aryl azides was examined as an alternative
with biaryl 3a (Scheme 4). Thus, reduction of 3a under
the standard conditions (H2, Pd/C, MeOH, 2.5 h) affor-
ded the corresponding amine 14. The crude amine was
carried directly into the azide formation [NaNO3, HCl,
0 °C, 5 min then NaN3, CH2Cl2]. After isolation of azide
11a as an organic solution in 1,2-dichlorobenzene, the
solution was heated to 165 °C. Slow nitrogen evolution
indicated decomposition to the nitrene and after 2 h
the reaction was complete. The product 10a crystallized
from the reaction solvent upon cooling to room temper-
ature and was isolated by filtration in 74% yield from
biaryl 3a.10 The synthesis of 1 could be completed by
alkylation with excess (bromomethyl)-cyclohexane and
saponification akin to the examples in Scheme 3.
11a
Scheme 4. Reagents and conditions: (a) H2, Pd/C, MeOH, rt, 2 h; (b)
NaNO2, HCl, 0 °C, 5 min then NaN3, 15 min; (c) 1,2-dichlorobenzene,
165 °C, 2 h, 74% from 3a.
to note that all intermediates can be isolated by precip-
itation from crude reaction mixtures. Further studies are
underway to expand the scope of this chemistry.
Acknowledgements
We wish to thank John P. Gardner for help during the
scale up work.
Supplementary data
In conclusion, an efficient and robust synthesis of
LSN433771 (1) and LSN426891 (2) has been described.
The new synthetic route is nine linear steps (four isola-
tions) from 2-bromo-3-nitrobenzoic acid. This new
strategy employs a palladium mediated intramolecular
biaryl coupling and allows for effective elaboration in
the core carbazole framework rapidly. It is important
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.tetlet.2005.
12.043.
References and notes
1. Tibes, Ulrich; Friebe, Walter-Gunar Expert Opinion
Invest. Drugs 1997, 6, 279–298.
NH2
R2O
2. Schevitz, R. W.; Bach, N. J.; Carlson, D. G.; Chirgadze,
N. Y.; Clawson, D. K.; Dillard, R. D.; Draheim, S. E.;
Hartley, L. W.; Jones, N. D.; Mihelich, E. D. Nature
Struct. Biology 1995, 2, 458–465, and references cited
therein.
3. Ames, D. E.; Opalko, A. Tetrahedron 1984, 1919.
4. Culhane, P. J. Org Synth. 1941, 1, 125.
5. (a) Bringmann, G.; Pabst, T.; Busemann, S.; Peters, K.;
Peters, E. M. Tetrahedron 1998, 54, 1425; (b) Rao, A. V.
Rama; Chakraborty, Tushar K.; Joshi, Subodh P. Tetra-
hedron Lett. 1992, 33, 4045.
6. Cadogen, J.; Cameron-Wood, M.; Mackie, R.; Searle, R.
J. Chem. Soc. 1965, 4831.
7. Hall, J. H.; Gisler, M. J. Org. Chem. 1976, 41, 3769–3770.
8. Rickborn, B.; Jense, F. R. J. Org. Chem. 1962, 27, 4608–
4610.
O
CN
O
O
R1
b or c
a
NO2
N
O
H
R1
CO2R2
R2O
12a; X = H, R = Me
12b; X = F, R = Et
3ab
O
CN
O
d
1 and 2
N
R1
R3
13a; R1 = H, R2 = Me, R3 = C6H12
13b; R1 = F, R2 = Et, R3 = Ph
9. The details of this reaction are currently under investiga-
tion and will be reported in due course.
10. While no alarming results were observed during this
reaction sequence, future scale up work via this route
would undoubtedly require a thorough hazard evaluation.
Scheme 3. Reagents and conditions: (a) P(OPh)3, 180 °C, 16 h, 40–
56%; (b) (bromomethyl)cyclohexane, NaI, Cs2CO3, DMF, 12 h, 65 °C,
90%; (c) benzylbromide, K2CO3, DMF, 1 h, rt, 96%; (d) KOH,
t-BuOH, 86%.