Synthesis of bioactive indolocarbazoles: Synthesis, nucleophilic ring-opening and chiral base desymmetrisation of a cyclic sulfate intermediate

Article abstract of DOI:10.1016/j.tetlet.2004.08.058

A number of new functionalised bridged indolocarbazole systems have been prepared by ring-opening reactions of a key cyclic sulfate intermediate, prepared from the corresponding diol by the action of sulfuryl diimidazole and DBU. The same cyclic sulfate a

Full text of DOI:10.1016/j.tetlet.2004.08.058

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 7469–7473
Synthesis of bioactive indolocarbazoles: synthesis, nucleophilic
ring-opening and chiral base desymmetrisation of
a cyclic sulfate intermediate
Christopher J. Nichols and Nigel S. Simpkins*
School of Chemistry, The University of Nottingham, University Park, Nottingham NG7 2RD, UK
Received 16 July 2004; accepted 6 August 2004
Available online 27 August 2004
Abstract—A number of new functionalised bridged indolocarbazole systems have been prepared by ring-opening reactions of a key
cyclic sulfate intermediate, prepared from the corresponding diol by the action of sulfuryl diimidazole and DBU. The same cyclic
sulfate also undergoes an unprecedented asymmetric rearrangement to a chiral ketone, on treatment with a chiral lithium amide
base.
ꢀ 2004 Elsevier Ltd. All rights reserved.
In a previous report we described the synthesis of a
small group of bridged indolocarbazoles, including alk-
ene 1 and a-hydroxyester 2 (Eq. 1),1 which are related to
the natural product K252a 3,2 but which incorporate
unnatural replacements to both the single atom bridge
(i.e., they are carbocyclic analogues) and the fused lac-
tam function.
H
N
H
N
O
O
N
N
N
N
O
O
Me
HO
Me
MeO
R
N
R
N
CO2Me
NHMe
O
O
O
O
3
4
steps
N
N
N
N
such analogues we decided that compounds derived
from alkene 1, but incorporating vicinal heteroatom
functions, as found in staurosporine 4,2 would be inter-
esting hybrid systems. An additional aim was to gener-
ate some of the key analogues, such as 2, in
non-racemic form through a desymmetrisation of alkene
1 or some derivative. Here we describe progress in both
of these areas, which has uncovered some new and novel
transformations, including a new chiral base-mediated
rearrangement.
HO
CO2Me
2 R = H, Bn
1 R = H, Bn, PMB
ð1Þ
The meso-alkene 1 was available in quantity from very
simple starting materials and provided ready access to
the K252a analogue 2, the N–H imide variant of which
(2 R = H) was found to show very promising kinase
inhibitory activity.3 In planning further exploration of
The rather poor solubility characteristics of indolo-
carbazoles makes them challenging substrates for
many types of reactions, especially if low temperatures
are usually employed. Our initial efforts to effect new
transformations of alkene 1 focused on epoxide genera-
tion, with the imide nitrogen protected as N–Bn or N–
PMB. Epoxide generation using peracids proved very
Keywords: Indolocarbazole; K252a analogue; Cyclic sulfate; Chiral
lithium amide.
*
Corresponding author. Fax: +44 115 9513564; e-mail: nigel.simp-
kins@nottingham.ac.uk
0040-4039/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2004.08.058

7470
C. J. Nichols, N. S. Simpkins / Tetrahedron Letters 45 (2004) 7469–7473
problematic (ca. 17–19% typical yield), DMDO gave a
maximum yield of 47% with 1 R = PMB, but the reac-
tion proved capricious, and complementary approaches
via bromohydrin formation also proved ineffective.
Experiments involving epoxide opening, using the mod-
est supplies available from the DMDO reaction, were
also discouraging, with azide (normally an excellent
nucleophile for epoxide opening) giving azidoalcohol
products in only very modest yields.
the ring opening occurred in high yield under conditions
which left the corresponding epoxide unreacted. The
direct installation of amine groups still proved difficult,
the modest yield obtained by reaction of 6 in neat
morpholine at reflux being one of the best results.8 In
contrast, reaction with oxygen or sulfur centred nucleo-
philes was relatively straightforward.
Given the apparent importance of the nitrogen function-
ality in bioactive systems such as staurosporine, we were
pleased to demonstrate that the azide function present in
7a could enable access to diverse nitrogen containing
products, such as 8–10 in straightforward fashion.
These problems led us to develop a complementary
approach, which relies on the sequence of alkene dihydr-
oxylation, cyclic sulfate formation and nucleophilic ring
opening (Scheme 1).
PMB
N
PMB
N
PMB
N
In contrast to the problematic epoxidation reactions,
dihydroxylation of alkene 1 at room temperature in
THF was very efficient.4 Cyclic sulfate synthesis by the
usual approach of forming the corresponding cyclic sulf-
ite with SOCl2 and then oxidation with the Sharpless in
situ RuO4 acetonitrile–water system proved low-yielding
due to problems with the second step.5 Instead, and
after considerable experimentation, we found that reac-
tion of the diol 5 with sulfuryl diimidazole with DBU as
base gave the desired sulfate directly and in excellent
yield. This combination proved uniquely effective in
our system, and may be useful in a more general sense.6
O
O
O
O
O
O
N
N
N
N
N
N
MeO2C
MeO2C
N
N
H2N
OH
N
OH
OH
N
8 71%
Reagents:
9 60%
10 95%
(7 to 8) Pd(C), H2, DMF; (7 to 9) Indium, NaI, allylbromide,
DMF 50 oC; (7 to 10) MeO2CC CCO2Me, DMF 70 o
C
Ring opening of the cyclic sulfate 6 was then carried out
under various conditions to give a range of products
7a–e in which interesting vicinal functionality had been
installed.7 The reaction of 6 with azide attested to the
excellent activating properties of the cyclic sulfate, since
Thus, azide reduction, to give the potentially versatile
primary amine 8, was possible under several types of
typical hydrogenation conditions, the imide N–PMB
PMB
N
PMB
N
PMB
N
O
O
O
O
O
O
(ii)
(i)
90%
95%
N
N
N
N
N
N
O
O
HO
OH
S
O2
1
5
6
reagent (iii)
yield (%)
X
PMB
N
O
O
NaN3, DMF, 80 o
C
N3
7a (91%)
7b (36%)
(iii)
morpholine, reflux
N
O
N
N
PhCO2NH4, DMF, 70 o
C
7c (88%)
OCOPh
SPh
PhSH, NaH, THF, RT
7d (91%)
7e (64%)
X
OH
NH4SCN, DMF, 70 o
C
SCN
7a-e
Scheme 1. Reagents and conditions: (i) OsO4, NMO, THF; (ii) Im2SO2, DBU, THF; (iii) see table; then 20% H2SO4, THF.

C. J. Nichols, N. S. Simpkins / Tetrahedron Letters 45 (2004) 7469–7473
7471
OBn
OBn
OBn
O
O
O
O
tBuOK, THF
O
O
O
O
H2SO4
SO2
THF-H2O
-
OSO3
O
OBn
OBn
OBn
11
12 86%
Scheme 2.
PMB
N
PMB
N
O
O
O
N
O
O
see text
Ph
N
Li
Ph
N
N
N
14
O
O
S
O2
6
13
Scheme 3.
being unaffected. Although amine 8 may prove useful in
reductive alkylation sequences, we also gained access to
diallyl amine 9 by means of the indium mediated Barbier
type process described by Yadav et al.9 Azide cycloaddi-
tion using dimethyl acetylenedicarboxylate as reaction
partner, one of the processes highlighted by Sharpless
in his Ôclick chemistryÕ approach,10 was also effective,
leading to triazole 10 in very high yield.
vided a 39% yield of ketone 13, following acidification.15
The enantiomeric excess of the ketone was established
by HPLC to be 87%, and the absolute configuration
was tentatively assigned as shown in Scheme 3 by corre-
lation with material obtained via asymmetric hydrobor-
ation of alkene 1.16 Unfortunately, preliminary reaction
screening has failed to uncover conditions under which
higher yields of ketone 13 can be obtained, and we have
not yet been able to test additional cyclic sulfate
substrates.
Whilst exploring the ring opening chemistry of cyclic
sulfate 6 we were attracted to a little-known transforma-
tion involving base-mediated rearrangement of a cyclic
sulfate to a ketonic product. This is exemplified by the
reaction of the cyclitol derived cyclic sulfate 11 to give
The chiral base transformation of 6 to give ketone 13 is
the first example of its type for cyclic sulfates, although
the chemistry is clearly related to the analogous trans-
formation of certain epoxides.17 Further study is
required to delineate the scope, limitations and likely
mechanism of the process.
´
the ketone 12, described by Fernandez-Mayoralas and
co-workers11 (Scheme 2).
These authors proposed a b-elimination process to give
an intermediate vinyl bisulfate, which on acidification
gave the ketone product.12 Unfortunately, application
of these conditions to our cyclic sulfate 6 gave only very
modest yields (8%) of the desired ketone 13 (which was
an important intermediate in the formation of hydroxy-
ester 2) (Scheme 3).
We are in the process of further modifying the indo-
locarbazoles described here, and initial screening results
show the corresponding deprotected imides (i.e., imide
N–H series) to be potent kinase inhibitors. Full details
of this work will be reported in due course.
Acknowledgements
Similarly, the use of LDA at low temperature also gave
very modest (ca. 5%) yields of the desired ketone, and
recovery of starting sulfate or the corresponding diol 5
predominated. Our main motivation for persisting with
this transformation was the hope that an asymmetric
variant might enable access to ketone 13 in non-racemic
form. Therefore chiral alkoxides, such as metal salts of
ephedrine derivatives were explored briefly, but to no
effect.13 Finally, we found that chiral lithium amide base
14 provided some more encouraging preliminary
results.14 Thus, addition of cyclic sulfate 6 to a solution
of lithium amide 14 and LiCl (4equiv) at ꢀ78ꢁC pro-
We gratefully acknowledge support for this work from
The University of Nottingham, Eli Lilly and Company,
Indianapolis, and NADAG, Munich.
References and notes
1. Riley, D. A.; Simpkins, N. S.; Moffat, D. Tetrahedron
Lett. 1999, 40, 3929.
2. For chemically-orientated reviews of the indolocarbazole
area, see: (a) Kno¨lker, H.-J.; Reddy, K. R. Chem. Rev.
2002, 102, 4303; (b) Bergman, J.; Janosik, T.; Wahlstro¨m,

7472
C. J. Nichols, N. S. Simpkins / Tetrahedron Letters 45 (2004) 7469–7473
N. Adv. Heterocycl. Chem. 2001, 80, 1; For more recent
9. Yadav, J. S.; Madhuri, C.; Reddy, B. V. S.; Reddy, G. S.
K. K.; Sabitha, G. Synth. Commun. 2002, 32, 2771.
10. Kolb, H. C.; Finn, M. G.; Sharpless, K. B. Angew. Chem.,
Int. Ed. 2001, 40, 2004.
11. (a) Carpintero, M.; Jaramillo, C.; Fernandez-Mayoralas,
A. Eur. J. Org. Chem. 2000, 1285; (b) See also Gourlain,
T.; Wadouachi, A.; Beaupere, D. Tetrahedron Lett. 2000,
41, 659.
12. The regiochemistry observed in the series of examples
described in Ref. 11b suggests to us that an a-lithiation-
carbenoid C–H insertion mechanism might operate, as
established for a somewhat analogous reaction involving
epoxides (see Ref. 17).
13. For asymmetric reactions using chiral alkoxides, see, for
example, (a) Jones, C. D.; Simpkins, N. S. Tetrahedron
Lett. 1998, 39, 1023; (b) Amadji, M.; Vadecard, J.;
Plaquevent, J. C.; Duhamel, L.; Duhamel, P. J. Am.
Chem. Soc. 1996, 118, 12483.
contributions, and leading references to biological areas,
see: (c) Faul, M. M.; Engler, T. A.; Sullivan, K. A.;
Grutsch, J. L.; Clayton, M. T.; Martinelli, M. J.; Pawlak,
J. M.; LeTourneau, M.; Coffey, D. S.; Pederson, S. W.;
Kolis, S. P.; Furness, K.; Malhotra, S.; Al-awar, R. S.;
Ray, J. E. J. Org. Chem. 2004, 69, 2967; (d) Zhu, G.;
Conner, S. E.; Zhou, X.; Chan, H.-K.; Shih, C.; Engler, T. A.;
Al-awar, R. S.; Brooks, H. B.; Watkins, S. A.; Spencer,
C. D.; Schultz, R. M.; Dempsey, J. A.; Considine, E. L.;
Patel, B. R.; Ogg, C. A.; Vasudevan, V.; Lytle, M. L.
Bioorg. Med. Chem. Lett. 2004, 14, 3057; (e) Zhang, L.;
Carroll, P.; Meggers, E. Org. Lett. 2004, 6, 521; (f)
Messaoudi, S.; Anizon, F.; Pfeiffer, B.; Golsteyn, R.;
Prudhomme, M. Tetrahedron Lett. 2004, 45, 4643.
3. In the form of their imide N–H derivatives many of these
systems display nanomolar inhibition of enzymes such as
protein kinase C (PKC).
4. This dihydroxylation appears in a patent, see: Roder, H.;
Lowinger, T. B.; Brittelli, D. R.; VanZandt, M. C. U.S.
Patent 6,013,646, Jan. 11, 2000.
14. For a review of chiral lithium amide base chemistry, see:
(a) OÕBrien, P. J. Chem. Soc., Perkin Trans. 1 1998, 1439;
For a special journal edition dedicated to chiral lithium
amide base chemistry, see: (b) OÕBrien, P. Tetrahedron
2002, 58, 4567–4733; Symposium in print guest editor. For
our most recent chiral lithium amide chemistry, see: (c)
Bennett, J. D.; Pickering, P. L.; Simpkins, N. S. Chem.
Commun. 2004, 1392.
5. For an excellent review of cyclic sulfate chemistry, see: (a)
Byun, H.-S.; He, L.; Bittman, R. Tetrahedron 2000, 56,
7051; For the seminal contribution, see: (b) Gao, Y.;
Sharpless, K. B. J. Am. Chem. Soc. 1988, 110, 7538.
6. The use of NaH as base seems more usual, but was
ineffective in our system, see, for examples, (a) Tewson, T.
J.; Soderlind, M. J. Carbohydr. Chem. 1985, 4, 529; (b)
Hanessian, S.; Vatele, J.-M. Tetrahedron Lett. 1981, 22,
3579.
15. A mixture of chiral lithium amide 14 and LiCl was
prepared by treatment of a solution of the appropriate
secondary amine–HCl salt (88.0mg, 0.34mmol) in THF
(1mL) at ꢀ78ꢁC under N2 with nBuLi (1.27M in hexanes,
0.52, 0.66mL). The mixture was allowed to warm to room
temperature for 0.5h before being recooled to ꢀ78ꢁC and
then a solution of cyclic sulfate 6 (50.0mg, 0.08mmol) in
THF (15mL) pre-cooled to ꢀ78ꢁC was added dropwise.
After 4h the reaction mixture was quenched with H2O
(1mL) and 20% H2SO4 (3.3mL) added. The reaction
mixture was allowed to warm to room temperature with
stirring for 18h and was then partitioned between EtOAc
(75mL) and H2O (75mL). The organic layer was washed
with H2O (75mL), saturated aqueous NaHCO3 solution
(75mL), brine (75mL), dried (MgSO4), and concentrated
in vacuo to give a yellow solid, which was purified by flash
column chromatography using a 2–10% MeOH–CH2Cl2
solvent system to give firstly ketone 13 (17.0mg, 39%),
7. Synthesis of azide 7a: Cyclic sulfate 6 (0.48g, 0.79mmol)
was dissolved in dry DMF (50mL) with stirring under N2.
NaN3 (0.52g, 7.92mmol) was added and the reaction
mixture heated to 80ꢁC. After 5h TLC showed complete
consumption of starting material and the reaction mixture
was allowed to cool to rt and concentrated in vacuo to
give a brown solid. The brown solid was redissolved in
THF (20mL), 20% H2SO4 (5mL) was added and the
resulting mixture stirred vigorously for 18h. The reaction
mixture was partitioned between EtOAc (200mL) and
H2O (100mL) and the organic layer was washed with H2O
(200mL), saturated aqueous NaHCO3 solution (150mL),
brine (2 · 200mL), dried (MgSO4) and concentrated in
vacuo to give crude azido alcohol. Purification by flash
column chromatography using 5% MeOH–CH2Cl2 as the
solvent system yielded 7a as an orange solid (0.41g, 91%);
mp 267–269ꢁC; (Found: m/z (FAB+) M+ 568.1836
C33H24N6O4 requires 568.1859); mmax (CHCl3)/cmꢀ1
3449, 2930, 2108, 1750, 1695, 1385, 1350; dH (DMSO-d6,
500MHz) 2.76 (1H, d, J 14.8, 10-Ha), 3.28 (1H, ddd, J
14.8, 7.5, 6.5, 10-Hb), 3.74 (3H, s, OMe), 3.87 (1H, dd, J
5.3, 4.8, 30-H), 4.45 (1H, dd, J 5.9, 5.3, 40-H) 4.84 (2H, s,
CH2Ar), 5.40 (1H, d, J 7.5, 20-H), 5.93 (1H, dd, J 6.5, 5.9,
50-H), 6.35 (1H, d, J 4.8, OH), 6.94 (2H, d, J 8.8, PMB),
7.39 (2H, d J 8.8, PMB), 7.45 (2H, dd, J 8.4, 7.2, 3-H, 9-
H), 7.70 (2H, dd, J 8.2, 7.2, 2-H, 10-H), 7.94 (1H, d, J 8.2,
11-H), 7.95 (1H, d, J 8.2, 1-H), 9.07 (1H, d, J 8.4, 8-H),
9.09 (1H, d, J 8.4, 4-H); dC (DMSO-d6, 125MHz) 36.3 (10-
CH2), 41.0 (CH2Ar), 56.0 (OMe), 59.6 (50-CH) 63.1 (20-
CH), 72.2 (40-CH), 83.6 (30-CH), 111.1 (CH) 111.2 (CH),
114.9 (PMB-CH), 116.6 (C), 116.8 (C), 119.5 (C), 119.6
(C), 121.6 (2CH), 122.0 (2C), 125.4 (CH), 125.5 (CH),
128.0 (CH), 128.2 (CH), 129.3 (C), 129.6 (C), 129.9 (PMB-
CH), 130.4 (C), 141.0 (C), 142.6 (C), 159.5 (C), 170.1
(CO), 170.2 (CO); m/z (FAB+) 569 (M++H, 8%), 568 (M+,
12), 419 (5), 329 (5), 289 (15), 137 (52), 108 (5), 77 (13).
8. Cyclic sulfate ring opening by heating in neat amine has
been used previously, see, for example, Hirsenkorn, R.
Tetrahedron Lett. 1990, 31, 7591.
26
½aꢁD ꢀ 62:9 (c 0.60, CHCl3); mp >280ꢁC; (Found: m/z
(FAB+) M+ 525.1663. C33H23N3O4 requires 525.1689);
m
max (CHCl3)/cmꢀ1 1753, 1697, 1384, 1349; dH (DMSO-d6,
500MHz) 2.49 (1H, dd, J 3.1, 19.4, 40-Ha), 3.07 (1H, dd, J
7.0, 19.7, 40-Hb), 3.26 (1H, dd, J 6.8, 19.4, 40-Ha), 3.53
(1H, m, 10-Hb), 3.74 (3H, s, OMe), 4.76 (2H, s, CH2PMB),
5.59 (1H, d, J 8.1, 20-H), 6.14 (1H, dd, J 6.8, 6.4, 20-H),
6.92 (2H, d, J 8.8, PMB), 7.35 (2H, d J 8.8, PMB), 7.45
(2H, dd, J 7.3, 7.9, 3-H, 9-H), 7.67 (2H, dd, J 6.7, 8.1, 2-H,
10-H), 7.89 (1H, d, J 8.4, 11-H), 7.97 (1H, d, J 8.4, 1-H),
9.04 (1H, d, J 7.9, 8-H), 9.05 (1H, d, J 7.9, 4-H); dC
(DMSO-d6, 100MHz) 37.3 (10-CH2), 40.9 (CH2PMB),
45.0 (40-CH2), 54.1 (50-CH), 56.0 (OMe), 58.3 (20-CH),
110.8 (CH), 111.0 (CH), 114.9 (PMB-CH and C), 116.6
(C), 116.9 (C) 119.3 (C), 119.9 (C), 121.6 (CH), 121.8
(CH), 121.9 (C), 125.4 (CH), 125.5 (CH), 128.1 (CH),
128.2 (CH), 128.9 (C), 129.2 (C), 129.9 (PMB-CH), 130.3
(C), 141.0 (C), 141.2 (C), 159.4 (C), 169.9 (2CO), 215.1
(CO).
Diol 5 was also recovered (5.0mg, 11%).
16. Our lack of success in obtaining a high yield and
enantiomeric excess in the asymmetric hydroboration (or
hydrosilylation) of alkene 1 was another reason that the
new cyclic sulfate rearrangement was pursued. Hydro-
boration of 1 with excess (ꢀ)-(Ipc)2BH, followed by the

C. J. Nichols, N. S. Simpkins / Tetrahedron Letters 45 (2004) 7469–7473
7473
usual oxidative work-up, gave alcohol (ꢀ)-14 in 33% ee
By analogy to norbornene and norbornadiene systems the
secondary alcohol would be expected to have the configu-
ration shown, and when this alcohol was oxidised using
Dess–Martin periodinane ketone (ꢀ)-13 was obtained,
matching the sample from the cyclic sulfate rearrange-
ment. For background, see: Brown, H. C.; Jadhav, P. K.;
Mandal, A. K. Tetrahedron 1981, 37, 3547.
maximum.
PMB
N
PMB
N
O
O
O
O
(i) (Ipc)2BH, THF, -78 oC to 0 o
(ii) H2O2, NaOH, THF
C
17. Hodgson, D. M.; Lee, G. P.; Marriott, R. E.; Thompson,
A. J.; Wisedale, R.; Witherington, J. J. Chem. Soc., Perkin
Trans. 1 1998, 2151.
N
N
N
N
HO
14 83% (33% ee)
1