Regioselective O-alkylations of indazolinone using (cyanomethylene) triphenylphosphorane

Article abstract of DOI:10.1055/s-0029-1217954

Regioselective O-alkylation of indazolinones using (cyanomethylene)- triphenthylphosphorane (CMPP) as a Mitsunobu-type reagent is described with a variety of aliphatic alcohols. This method was also successfully applied to the N-alkylation of O-protected

Full text of DOI:10.1055/s-0029-1217954

LETTER
2673
Regioselective O-Alkylations of Indazolinone Using
(Cyanomethylene)triphenylphosphorane
Regioselective
O
-A
m
lkylation of Indazo
y
linones
U
sing CM
R
PP andall,* Florine Duval
Research Chemistry, Pfizer Global Research and Development, Ramsgate Road, Sandwich, Kent, CT13 9NJ, UK
Fax +44(1304)651821; E-mail: Amy.Randall@pfizer.com
Received 23 April 2009
single example utilizing recent developments in Mitsuno-
bu protocols.
Abstract: Regioselective O-alkylation of indazolinones using (cy-
anomethylene)triphenthylphosphorane (CMPP) as a Mitsunobu-
type reagent is described with a variety of aliphatic alcohols. This
method was also successfully applied to the N-alkylation of O-pro-
tected indazolinone. Selective N1 and N2 alkylations on indazolino-
ne have previously been described; our methodology is therefore
orthogonal to the previous precedent.
Initially we looked at the reaction of indazolinone with
cyclopentanol, revisiting some of the standard Mitsunobu
protocols,6 but despite exhaustive screening of conditions
this resulted in only trace quantities of the desired com-
pound. Even modified reagents [e.g., N,N,N′,N′-tetra-
methylazodicarboxamide (TMAD), tributylphosphine
(TBP7)] which allow deprotonation of less acidic nucleo-
philes showed no improvement.8
Key words: alkylations, heterocycles, ylides, Mitsunobu, trialkyl-
phosphoranes
To determine why these reactions were problematic at
room temperature we considered the mechanism of the
Mitsunobu reaction (see general reaction scheme in
Scheme 1). The key steps include the deprotonation of the
nucleophile (acidic partner) and the displacement of phos-
phine oxide by the nucleophilic anion.
The indazolinone ring system (Figure 1) is a versatile core
that can be functionalized on both the oxygen and nitro-
gen atoms. There is a limited amount of literature on
transformations involving indazolinones with the main ar-
eas of focus being N-alkylation with halides,1 reaction
with activated carbonyl compounds,2 and activation of the
oxygen.3 Temporary protecting groups have also been
employed on this system to control selectivity.4
The ‘nucleophile’ in a standard Mitsunobu reaction has to
have a pKa less then 11 in order for the reaction to proceed
well, and once it is above 13 the reaction will not occur.
Due to the thermal instability of the reagents these reac-
tions are also usually carried out at room temperature. We
have estimated the pKa of indazolinone to be 8.31,16 and
so in this case deprotonation should not be an issue. How-
ever, unlike Selwood,5 our reactions were not heated, and
this could be the reason that we see no reaction.
OH
O
N
NH
N
N
H
H
Figure 1 Indazolinone
Stabilized trialkylphosphoranes, such as (cyanomethyl-
ene)trimethylphosphorane (CMMP9), were designed spe-
cifically to mediate Mitsunobu-type reactions on
nucleophiles with a pKa up to 23. However, unlike tradi-
tional Mitsunobu reagents, trialkylphosphoranes are also
thermally stable,10 allowing reactions to be carried out at
elevated temperatures. Their structural similarity to the
To our knowledge only one example of a selective O-
alkylation on this system has been published.5 This was a
Mitsunobu reaction; the yield obtained was fairly poor
and heating was required in the presence of thermally un-
stable azodicarbonyldipiperidine (ADDP). We were inter-
ested to see if we could expand the scope and safety of this
O
O
O
O
O
O
Nu–H
Ph3P
N
OEt
EtO
N
H
OEt
EtO
N
N
EtO
N
N
OEt
nucleophile (acidic partner)
PPh3
PPh3
DEAD
Nu
HO
R
Nu
Nu
R
OPh3P
R
nucleophile (acidic partner)
Scheme 1 Mitsunobu reaction pathway9 under standard conditions. pKa of nucleophile must be less then 11 to allow deprotonation step to
occur
SYNLETT 2009, No. 16, pp 2673–2675
x
x
.x
x
.2
0
0
9
Advanced online publication: 04.09.2009
DOI: 10.1055/s-0029-1217954; Art ID: D10209ST
© Georg Thieme Verlag Stuttgart · New York

2674
A. Randall, F. Duval
LETTER
traditional Mitsunobu intermediates (Figure 2) suggests We were interested in testing this reagent on indazolino-
they proceed by a similar reaction mechanism.
ne. An initial trial was carried out using cyclopentanol (2
equiv) and CMPP (2 equiv) in toluene at reflux for 20
hours (Scheme 2). The reaction proceeded well giving a
48% yield of O-alkylated product that was confirmed by
NMR analysis (1a, Table 1).
CMMP is the most common trialkylphosphorane found in
the literature, and it has been used successfully on inda-
zoles,11 activated methylenes,12 and secondary amines.13
Unfortunately CMMP is very air and moisture sensitive
and is currently not commercially available. Tsunoda et
al. cited
a
single example of the use of
OH
N
O
OH
(Cyanomethylene)triphenylphosphorane (CMPP),14 and
the yield obtained was comparable to CMMP. This re-
agent is commercially available and, although still par-
tially air sensitive, should be easier to work with.
CMPP (2 equiv),
toluene, 100 °C, 20 h
N
+
48%
N
N
H
H
1a
Scheme 2 O-alkylation (step A)
CMPP
CN
CMMP
CN
O
O
Me3P
Ph3P
This was very encouraging, especially as secondary alco-
hols are traditionally more challenging in the Mitsunobu
reaction then primary alcohols.15 Interestingly, when the
same reaction was carried out at room temperature no
conversion was observed.
OEt
EtO
N
N
PPh3
Mitsunobu intemediate
Ph3P
CN
Me3P
CN
The scope of this reaction was investigated with a range
of alcohols (Table 1, O-alkylation) and appears to be fair-
ly general. Although yields are only moderate, the reac-
tions appear clean with major impurities being starting
material and bisalkylated byproduct (where reaction had
first occurred on the oxygen and then on the nitrogen). No
mono-N-alkylated products were observed (due to the
large pKa difference between OH and NH16). Ethers were
well tolerated (9a) as were tertiary amines (10a), howev-
er, decreasing yields were observed for secondary and
primary amines (11a and 12a). Sterically hindered alco-
hols also showed a reduction in yield (7a). When more re-
active alcohols were used (8a) reactions were pushed to
completion and so little starting material was observed,
however, yields were still only moderate as more bisalky-
lated byproduct was generated.
Figure 2 Comparison of traditional Mitsunobu intermediate to the
ylides generated form CMMP and CMPP
Table 1 Results of Alkylations Using CMMP
Alcohol
Yield (%),
Yield (%),
O-alkylation
N-alkylation of 1a
OH
48 1a
65 1b
31 2a
42 3a
54 2b
84 3b
OH
OH
56 4a
28 4b
OH
OH
We also carried out this reaction with the alcohol used by
Selwood (3-dimethylaminopropan-ol),5 our reaction con-
ditions offered an improvement in yield (40%).
36 5a
16 5b
The presence of bisalkylated byproduct suggested that a
second alkylation should be fairly simple, so an initial tri-
al was carried out on intermediate 1a with ethanol using
identical conditions to our O-alkylation, pleasingly it pro-
ceeded in a 54% yield (2b, Table 1).
OH
OH
20 6a17
11 6b
9 7a17
–
We subsequently investigated the scope of the reaction
with a range of alcohols on this cyclopentyl-substituted
indazolinone (Scheme 3 and Table 1, N-alkylation). This
second alkylation generally showed an improvement in
yield, as bisalkylation was no longer an issue (3a vs.3b).
OH
32 8a
44 9a
47 10a
95 8b
74 9b
53 10b
Similar trends in reactivity were observed, with the same
functional groups being tolerated. The effects of steric
hindrance seemed to be more pronounced possibly sug-
gesting that SN2 displacements are more challenging in
this case (4a vs.4b).
OH
O
N
OH
OH
31 11a
13 12a
0 11b
N
H
The success of this work is presumably due to the thermal
stability of CMPP which allows these reactions to be
OH
15 12b
H2N
Synlett 2009, No. 16, 2673–2675 © Thieme Stuttgart · New York

LETTER
Regioselective O-Alkylation of Indazolinones Using CMPP
2675
References and Notes
O
O
(1) (a) Malamas, M. S.; Millen, J. J. Med. Chem. 1991, 34,
1492. (b) Aran, V. J.; Diez-Barra, E.; De la Hoz, A.;
Sanchez-Verdu, P. Heterocycles 1997, 45, 129.
(2) Eacho, P. I.; Foxworthy-Mason, P. S.; Lin, H.; Lopez, J.;
Moslor, M. K.; Richett, M. E. WO 2004093872, 2004.
(3) Gordon, D. W. Synlett 1998, 1065.
CMPP (2 equiv),
toluene, 100 °C, 20 h
+
N
ROH
N
N
N
H
R
1a
Scheme 3 N-Alkylation of 1a (step B)
(4) Arb, V. J.; Flores, M.; Muiioz, P.; Paez, J. A.; Sanchez-
Verdul, P.; Stud, M. Liebigs Ann. 1996, 683.
(5) Selwood, D. L.; Brummell, D. G.; Budworth, J.; Burtin,
G. E.; Campbell, R. O.; Chana, S. S.; Charles, I. G.;
Fernandez, P. A.; Glen, R. C.; Goggin, M. C.; Hobbs, A. J.;
Kling, M. R.; Liu, Q.; Madge, D. J.; Meillerais, S.; Powell,
K. L.; Reynolds, K.; Spacey, G. D.; Stables, J. N.; Tatlock,
M. A.; Wheeler, K. A.; Wishart, G.; Woo, C. J. Med. Chem.
2001, 44, 78.
heated. It also facilitates the use of nucleophiles (NuH) of
higher pKa allowing the second alkylation (on N1) to pro-
ceed. CMPP offers a simple method for the regioselective
O-alkylation of indazolinone. Selective N1- and N2-alky-
lations on indazolinone have previously been described;18
our methodology is therefore orthogonal to the previous
precedent.
(6) Mitsunobu, O.; Yamanda, M.; Mukaiyama, T. Bull. Chem.
Soc. Jpn. 1967, 40, 935.
(7) Tsunoda, T.; Otsuka, J.; Yamamita, Y.; Ito, S. Chem. Lett.
1994, 539.
General Method for O-Alkylation (Compound 1a)
(8) These reactions were not heated for safety reasons and this
is probably why unlike Selwood (ref. 5) we saw no reaction.
(9) A more detailed review of the mechanism can be found in:
McNulty, J.; Capretta, A.; Laritchev, V.; Dyck, J.;
Robertson, A. J. Angew. Chem. Int. Ed. 2003, 42, 4051.
(10) Sakamoto, I.; Kaku, H.; Tsunoda, T. Chem. Pharm. Bull.
2003, 51, 474.
Indazolinone (100 mg, 0.746 mmol), cyclopentanol (203 mL, 2.24
mmol) and CMPP (450 mg, 1.49 mmol) were taken up in toluene (2
mL) and stirred at 100 °C in a Reacti-Vial for 12 h. Reaction mix-
ture was then concentrated under vacuum and crude material puri-
fied by column chromatography [7 g silica, heptane (100%) through
to heptane–EtOAc (91:1)] to give a yellow oil/foam (72 mg, 48%).
(11) Bombrun, A.; Casi, G. Tetrahedron Lett. 2002, 43, 2187.
(12) (a) Uemoto, K.; Kawahito, A.; Matsushita, N.; Sakamoto, I.;
Kaku, H.; Tsunoda, T. Tetrahedron Lett. 2001, 42, 905.
(b) Tsunoda, T.; Uemoto, K.; Ohtani, T.; Kaku, H.; Ito, S.
Tetrahedron Lett. 1999, 40, 7359.
(13) Golantsov, N. E.; Karachava, A. V.; Yurovskaya, M. A.
Chem. Heterocycl. Compd. (N.Y., NY, U.S.) 2008, 44, 263.
(14) Tsunods, T.; Ozaki, F.; Ini, S. Tetrahedron Lett. 1994, 35,
5081.
Characterization Data
1H NMR (400 MHz, CDCl3): d = 1.66 (2 H, m), 1.85 (2 H, m), 2.00
(4 H, m), 5.31 (1 H, quin, J = 4.7 Hz), 7.07 (1 H, m), 7.27 (1 H, d,
J = 8.4 Hz), 7.35 (1 H, m), 7.69 (1 H, d, J = 8.0 Hz), 9.34 (1 H, br
s). 13C NMR (400 MHz, CDCl3): d = 24.2, 33.2, 81.5, 109.8, 113.6,
119.8, 120.3, 127.9, 142.7, 157.3. ESI-MS: m/z = 203 [M + H]+.
MS: m/z calcd for C12H14N2O [M + H]+: 203.1179; found:
203.1178.
(15) (a) Manivel, P.; Rai, N. P.; Jayashankara, V. P.;
Arunachalam, P. N. Tetrahedron Lett. 2007, 48, 2701.
(b) Tsunods, T.; Ozaki, F.; Ini, S. Tetrahedron Lett. 1994,
35, 5081. (c) Zong, K.; Groenendaal, L.; Reynolds, J. R.
Tetrahedron Lett. 2006, 47, 3521.
General Method for N-Alkylation (Compound 2b)
3-Cyclopentyloxy-1H-indazole (100 mg, 0.5 mmol), EtOH (86 mL,
1.5 mmol) and CMPP (298 mg, 1.0 mmol) were taken up in toluene
(2 mL) and stirred at 100 °C in a Reacti-Vial for 12 h. Reaction mix-
ture was then concentrated under vacuum and crude material puri-
fied by column chromatography [7 g silica, heptane–EtOAc (95:5)]
to give a yellow oil/foam (61 mg, 54%).
(16) ACD PKa predictor suggests (Figure 3)
OH
O
Characterization Data
N
N
N
1H NMR (400 MHz, CDCl3): d = 1.42 (3 H, t, J = 7.2 Hz), 1.67 (2
H, m), 1.87 (2 H, m), 2.00 (4 H, m), 4.24 (2 H, q, J = 7.2 Hz), 5.32
(1 H, quin, J = 4.5 Hz), 7.02 (1 H, m), 7.23 (1 H, d, J = 8.5 Hz), 7.34
(1 H, m), 7.66 (1 H, dt, J = 8.0, 0.9 Hz). 13C NMR (400 MHz,
CDCl3): d = 15.1, 24.2, 33.2, 43.5, 81.4, 108.7, 113.8, 118.9, 120.5,
127.1, 141.1, 155.5. ESI-MS: m/z = 231 [M + H]+. MS: m/z calcd
for C14H18N2O [M+H]+: 231.1508; found: 231.1492.
N
N
N
H
H
H
pKa ~8.91
pKa ~14.00
pKa ~13.33
Figure 3
(17) NMR and LC-MS data confirm that 6a and 7a are
diasteromers. Based on mechanism inversion in assumed,
but this has not been confirmed experimentally.
(18) Aran, V. J.; Flores, M.; Munoz, P.; Paez, J. A.; Sanchez-
Verdu, P.; Stud, M. Liebig. Ann. 1996, 683.
Acknowledgment
The authors would like to thank Paul Glossop and Dr Lyn Jones for
advice and helpful discussions. We also thank Denise Burring for
structural confirmation and Dr. David Blakemore and Dr. David
Fenwick for reviewing the manuscript.
Synlett 2009, No. 16, 2673–2675 © Thieme Stuttgart · New York