Intramolecular sila-matteson rearrangement: A general access to silylated heterocycles

Article abstract of DOI:10.1021/ol300598s

A series of new silylated heterocycles has been efficiently prepared using an intramolecular silicon version of the Matteson rearrangement, providing two isomers of binuclear heterocycles. This method applies to a large variety of substrates, a direct rel

Full text of DOI:10.1021/ol300598s

ORGANIC
LETTERS
2012
Vol. 14, No. 8
2074–2077
Intramolecular Sila-Matteson
Rearrangement: A General Access
to Silylated Heterocycles
Cyril Franc-ois,† Thomas Boddaert,† Muriel Durandetti,*,† Olivier Querolle,‡
Luc Van Hijfte,‡ Lieven Meerpoel,‡ Patrick Angibaud,‡ and Jacques Maddaluno*,†
ꢀ
CNRS UMR 6014 “COBRA” & FR 3038 “INC3M”, Universite de Rouen and INSA de
Rouen, 76821 Mont St. Aignan Cedex, France, and Department of Medicinal Chemistry,
Janssen, Campus de Maigremont, BP615, 27106 Val de Reuil, France
muriel.durandetti@univ-rouen.fr; jmaddalu@crihan.fr
Received March 8, 2012
ABSTRACT
A series of new silylated heterocycles has been efficiently prepared using an intramolecular silicon version of the Matteson rearrangement,
providing two isomers of binuclear heterocycles. This method applies to a large variety of substrates, a direct relationship between the Hammett
constants of the aromatic substituents and the isomer ratio being observed. Complementary experiments suggest that a common penta-
organosilicate species is involved.
Replacing a carbon atom with a silicon can be regarded
as a way to develop innovative new drugs.1 Despite the
large similarities between these elements, significant ad-
vantages can be returned.2 Thus, the effect of a C/Si swap
on the physiological and biological properties of known
drug skeletons has been widely investigated in the past
two decades,3 and several bioactive silacycles have been
synthesized. For example, sila-haloperidol 1, a dopamine
receptor antagonist, displayshigher subtypeselectivity and
a different metabolism pattern compared to its carbon
analogue.4 The tetrahydrosilaisoquinoline 25 showed psy-
chotropic activity, while the disila-bexarotene 36 was stud-
ied for its retinoid agonist potency (Figure 1). If several
compounds contain a heteroatomꢀsilicon bond, hetero-
cyclic molecules bearing a tetraorganosilicon moiety in-
corporated in a cycle are much less described.7
†
ꢀ
Universite de Rouen and INSA de Rouen.
(5) Lukevics, I. S. E.; Germane, S.; A. Zablotskaya, A. Chem.
Heterocycl. Compd. 1997, 33, 234–238.
‡ Janssen.
(1) Meanwell, N. A. J. Med. Chem. 2011, 54, 2529–2591.
(2) (a) Mills, J. S.; Showell, G. A. Expert Opin. Invest. Drugs 2004, 13,
1149–1157. For an example of a more stable silicon analogue than its
ꢀ
´
parent all-carbon compound, see: (b) Dıez-Gonzalez, S.; Paugam, R.;
Blanco, L. Eur. J. Org. Chem. 2008, 3298–3307.
€
(6) Bauer, J. B.; Lippert, W. P.; Dorrich, S.; Tebbe, D.; Burschka, C.;
Christie, V. B.; Tams, D. M.; Henderson, A. P.; Murray, B. A.; Marder,
T. B.; Przyborski, S. A.; Tacke, R. ChemMedChem. 2011, 6, 1509–1517.
(7) (a) Rousseau, G.; Blanco, L. Tetrahedron 2006, 62, 7951–7993.
For recent examples: (b) Blaszykowski, C.; Brancour, C.; Dhimane,
A. L.; Fensterbank, L.; Malacria, M. Eur. J. Org. Chem. 2009, 1674–
1678. (c) Furukawa, S.; Kobayashi, J.; Kawashima, T. J. Am. Chem.
(3) (a) Bains, W.; Tacke, R. Curr. Opin. Drug Discovery Dev. 2003, 6,
526–543. (b) Franz, A. K. Curr. Opin. Drug Discovery Dev. 2007, 10,
€
ꢀ
Soc. 2009, 131, 14192–14193. (d) Hernandez, D.; Nielsen, L.; Lindsay,
ꢀ
´
K. B.; Lopez-Garcıa, M. A.; Bjerglund, K.; Skrydstrup, T. Org. Lett.
654–671. (c) Tacke, R.; Bertermann, R.; Burschka, C.; Dorrich, S.;
€
€
Fischer, M.; Muller, B.; Meyerhans, G.; Schepmann, D.; Wunsch, B.;
Arnason, I.; Bjornsson, R. ChemMedChem 2012, 7, 523–532.
2010, 12, 3528–3531. (e) Kirpichenko, S. V.; A. I. Albanov, A. I.
J. Organomet. Chem. 2010, 695, 663–666.
(8) (a) Aoyama, T.; Sato, Y.; Suzuki, T.; Shirai, H. J. Organomet.
Chem. 1978, 153, 193–207. (b) Hwu, J. R.; King, K. Y. Chem.;Eur. J.
2005, 11, 3805–3815. (c) Alberico, D.; Rudolph, A.; Lautens, M. J. Org.
Chem. 2007, 72, 775–781.
€
(4) (a) Tacke, R.; Popp, F.; Muller, B.; Theis, B.; Burschka, C.;
Hamacher, A.; Kassack, M. U.; Schepmann, D.; Wunsch, B.; Jurva, U.;
€
Wellner, E. ChemMedChem 2008, 3, 152–164. (b) Johansson, T.;
Weidolf, L.; Popp, F.; Tacke, R.; Jurva, U. Drug Metab. Dispos. 2010,
38, 73–83.
r
10.1021/ol300598s
Published on Web 04/05/2012
2012 American Chemical Society

Scheme 2. First Cyclization Attempts and Proposed
Pentavalent Silicate Species
Figure 1. Examples of bioactive silacycles.
Increasing interest in silylated derivatives and a lack of
general synthetic procedures prompted us to focus on the
preparation of the hydrosilaquinoline 4a (Z = N-Boc) and
silachroman 5a (Z = O) moieties, obtained previously in
moderate yields and under drastic conditions.8 Our initial
retrosynthetic route is outlined in Scheme 1: the expected
products 4a and 5a could be formed by an intramolecular
cyclization of the precursors 6 and 7, respectively. These
substrates would in turn be prepared from aniline and
phenol derivatives 8 and 9 in the presence of bis-
(chloromethyl)dimethylsilane 10.
We hypothesized that a common pentaorganosilicate spe-
cies 11 could explain this result. The latter would evolve by
either migration of the aromatic ring (path a, Scheme 2) or
the CH2ꢀSi bond (path b). Such a mechanism parallels the
reactivity of the R-halosilanes with that of the R-halobo-
ronic esters, classically employed in the Matteson rearrange-
ment (Scheme 3).10
Scheme 3. Matteson and Sila-Matteson Rearrangements
Scheme 1. Initial Synthetic Route for the Preparation of
Hydrosilaquinoline 4a and Silachroman 5a
The precursors 6 and 7 were prepared in good yields
from 8 and 9 (72% and 82% yields, respectively). Then, the
halogenꢀlithium exchange and subsequent nucleophilic
substitution were performed at ꢀ40 °C using n-butyl-
lithium in tetrahydrofuran. Surprisingly, the intramolecu-
lar cyclization furnished not only the desired products 4a
and 5a but also their regioisomers 4b and 5b.8a,9 In both
series, the cyclization proceeds in good yield (about 85%)
and low selectivity (4a/4b = 45:55 and 5a/5b = 60:40,
Scheme 2).
This reactivity has been previously proposed for
silicon in the case of the nucleophilic addition of
halide or alkoxide11 and was briefly evoked for a carbon
nucleophile.12
Next, various experimental conditions were screened.13
In tetrahydrofuran, none of the following parameters
seemed to exert a significant influence on the ratio of the
(9) For previous syntheses of related compounds, see: (a) Miller,
D. J.; Showell, G. A.; Conroy, R.; Daiss, J.; Tacke, R.; Tebbe, D. Amedis
Pharmaceuticals Ltd. (U.K.), PCT Int. Appl. WO/2005/005443A1,
2005. (b) Greenlee, W. J.; Zhu, Z.; Asberom, T.; Huang, X.; Josien, H. B.
Schering Corp. (USA), PCT Int. Appl. WO/2009/061699A1, 2009.
(10) (a) Matteson, D. S. Chem. Rev. 1988, 89, 1535–1551. (b) Matteson,
D. S. Tetrahedron 1998, 54, 10555–10607 and references cited therein.
For recent applications, see: (c) Molander, G. A.; Hiebel, M.-A. Org.
Lett. 2010, 12, 4876–4879. (d) Porcel, S.; Bouhadir, G.; Saffon, N.;
Maron, L.; Bourissou, D. Angew. Chem., Int. Ed. 2010, 49, 6186–
6189. (e) Sonawane, R. P.; Jheengut, V.; Rabalakos, C.; Larouche-
Gauthier, R.; Scott, H. K.; Aggarwal, V. K. Angew. Chem., Int. Ed.
2011, 50, 3760–3763. For recent silicon-based anion-relay rearrange-
ments, see: (f) Smith, A. B., III; Tong, R.; Kim, W. S.; Maio, W. A.
Angew. Chem., Int. Ed. 2011, 50, 8904–8907. (g) Zheng, P.; Cai, Z.;
Garimallaprabhakaran, A.; Rooshenas, P.; Schreiner, P. R.; Harmata,
M. Eur. J. Org. Chem. 2011, 5255–5260. (h) Smith, A. B., III; Hoye,
A. T.; Martinez-Solorio, D.; Kim, W. S.; Tong, R. J. Am. Chem. Soc.
2012, 134, 4533–4536.
(11) (a) Corey, J. Y.; Corey, E. R.; Chang, V. H. T.; Hauser, M. A.;
Lelber, M. A.; Reinsel, T. E.; Riva, M. E. Organometallics 1984, 3, 1051–
1060 and references cited therein. (b) Damrauer, R.; Danahey, S. E.;
Yost, V. E. J. Am. Chem. Soc. 1984, 106, 7633–7634. (c) Sans, E. A.;
Shechter, H. Tetrahedron Lett. 1985, 26, 1119–1122. (d) Hudrlik, P. F.;
Abdallah, Y. M.; Kulkarni, A. K.; Hudrlik, A. M. J. Org. Chem. 1992,
57, 6552–6556. (e) Eisch, J. J.; Chiu, C. S. Heteroat. Chem. 1994, 5, 265–
274. (f) Hijji, Y. M.; Hudrlik, P. F.; Hudrlik, A. M. Chem. Commun.
1998, 1213–1214. (g) Allen, M. J.; Aprahamian, S. L.; Sans, E. A.;
Shechter, H. J. Org. Chem. 2002, 67, 3561–3574.
(12) (a) Shiragami, H.; Kawamoto, T.; Imi, K.; Matsubara, S.; Utimo-
to, K.; Nozaki, H. Tetrahedron 1988, 44, 4009–4022. (b) Matsumoto, K.;
Oshima, K.; Utimoto, K. Tetrahedron. Lett. 1990, 31, 6055–6058.
(13) Comparative results are detailled in the Supporting Information.
Org. Lett., Vol. 14, No. 8, 2012
2075

isomers: (i) temperature; (ii) structure of the organometal-
lic (n-BuLi, t-BuLi, MeLi, i-PrMgCl); (iii) halogen on the
aromatic ring (iodine or bromine); (iv) leaving group on
the silicon side chain (chlorine, iodine, mesylate, or
tosylate). These observations suggested that the isomer
ratio is mainly governed by the intrinsic reactivity of the
pentaorganosilicate species 11 and, more precisely, by the
respective stabilities of the ArꢀSi and CH2ꢀSi bonds.14
We first noted that the electron-withdrawing effect of the
trifluoromethyl or of the halogen substituents favors the
heterolytic cleavage of the ArꢀSi bond in the pentaorga-
nosilicate species, increasing the amount of the a-isomer
(entries 2ꢀ9). The delicate balance between inductive and
mesomer effects probably explains that similar ratios were
obtained with the 5-CF3 (12) and 5-Cl (13) substituents
(entries 2 to 4), as well as 4-CF3 (15) and 4-F (16) (entries 7
to 9). Similarly, the 5-F (14) affords modest selectivity
because of the contradictory resonance donor and induc-
tive attractor effects it generates (entry 6).16a
Table 1. Aromatic Substituents Effect
In contrast, the pure electron-donating character of
substituents in the para position activates the aromatic
ring and disfavors the ArꢀSi bond heterolytic cleavage,
leading to b-isomers predominantly (entries 11, 12, and
14). The low donating effect of a methyl substituent in the
meta position leads to a negligible influence on the isomer
ratio. Finally, and as pointed out by Schlosser et al.,16b the
competition between the attractor inductive and donor
resonance effects associated to the meta methoxy group
could explain that the a-isomer is favored in this case (entry
10). Such results suggest that the limiting step is the CꢀSi
bond heterolytic cleavage rather than the displacement of
the leaving group by the nucleophile. A similar observation
was published before by Allen et al.11g
It was tempting at this stage to correlate these results to
the σ Hammett constants.17 Gratifyingly, the plot of the
log(ratio b/a) against σ led to a satisfying linearity
(correlation coefficient = 0.940) on a relatively large scale
of σ values (ꢀ0.83 to þ0.54) and for substituents in the
meta as well as the para position (Figure 2).
starting
material
prod.
yields
[%]c
entry
1
R
cond.a
A
(a/b)b
H
6
4a/4b
84
52
72
74
81
74
76
47
82
79
73
92
89
93
45:55
2
5-CF3
ꢀ
12
ꢀ
A
B
A
B
A
B
A
B
B
B
B
B
B
22a/22b
86:14
3
22a/22b
90:10
4
5-Cl
13
ꢀ
23a/23b
83:17
5
ꢀ
23a/23b
83:17
6
5-F
14
15
16
ꢀ
24a/24b
57:43
7
4-CF3
4-F
25a/25b
89:11
8
26a/26b
78:22
9
ꢀ
26a/26b
82:18
10
11
12
13
14
4-MeO
5-MeO
5-Me2N
4-CH3
5-CH3
17
18
19
20
21
27a/27b
61:39
28a/28b
25:75
29a/29b
10:90
30a/30b
43:57
31a/31b
37:63
Figure 2. Plot of log(b/a) against Hammett σ constants.
a Condition A: n-BuLi (1.2 equiv), ꢀ40 °C. Condition B: t-BuLi
(2.4 equiv), ꢀ78 °C. b Ratio a/b determined from the crude 1H NMR
spectra. c Isolated yields.
Next, a study on the influence of the protecting group
borne by the nitrogen atom was undertaken (Table 2). If
both carbamates 6 and 32 led to disappointing selectivities
(entries 1ꢀ2), the mesyl and tosyl protectediodoanilines 33
To evaluate the influence of the aromatic substituents on
the ratio of the two isomers, the intramolecular cycliza-
tion was applied to substrates incorporating electron-rich
or -deficient aromatics (Table 1).
For all substituents, the silylated heterocycles were
obtained in good chemoselectivies and yields, as long as
n-BuLi (Condition A) was replaced by t-BuLi(ConditionB).15
(15) Most of the time, n-BuLi (ꢀ40 °C, 1.2 equiv) was replaced by
t-BuLi (ꢀ78 °C, 2.4 equiv) to avoid the WurtzꢀFittig type addition of
the aryllithium on the cogenerated butyl halide: (a) Bailey, W. F.;
Longstaff, S. C. Chimica Oggi (Chemistry Today) 2001, 19, 28–32. (b)
Clayden, J. Organolithiums: Selectivity for Synthesis; Pergamon: Oxford,
U.K., 2002; pp 111ꢀ142.
(16) (a) Schlosser, M.; Maccaroni, P.; Marzi, E. Tetrahedron 1998,
54, 2763–2770. (b) Faigl, F.; Marzi, E.; Schlosser, M. Chem.;Eur. J.
2000, 6, 771–777.
(14) Couzijn, E. P. A; Ehlers, A. W.; Schakel, M.; Lammertsma, K.
J. Am. Chem. Soc. 2006, 128, 13634–13639.
(17) (a) Hammett, L. P. J. Am. Chem. Soc. 1937, 59, 96–103.
(b) Hansch, C.; Leo, A.; Taft, R. W. Chem. Rev. 1991, 91, 165–195.
2076
Org. Lett., Vol. 14, No. 8, 2012

and 34 afforded almost exclusively the b-isomer (entries
3ꢀ4). This striking effect can be understood in light of the
recent results by Chataigner et al. suggesting that the
efficient π-delocalization of the nitrogen lone pair occur-
ring toward the carbonyl group of the carbamate fully
deactivates the donor character of the nitrogen and there-
fore the aromatic ring.18 Note that, in the case of the mesyl
protected substrate 34 (entry 4), a competitive deprotona-
tion and subsequent intramolecular nucleophilic substitu-
tion of the chlorine afford a cyclic sila-sulfonamide,19
decreasing the yield in 37. Thus, the nitrogen substituent
can also be used as a lever to control the selectivity.
Scheme 4. Evidence for a Hypervalent Silicon Pathway
Both isomers 5a and 5b were obtained in a ratio compa-
rable to that of Scheme 2 (50:50 vs 60:40), hinting at the
formation of a similar hypervalent species. However, it
remains unclear at this stage whether 11 is a transition state
or an intermediate in this transformation. Calculations are
in progress to answer this question and to evaluate to
which extent the Berry pseudorotation21 can influence the
selectivity.
Table 2. Electron-Withdrawing Protecting Group Effect
In summary, we have developed a general and efficient
access to silylated heterocycles through an original “sila-
Matteson” type rearrangement. The influence of both the
aromatic ring substituents and nitrogen protecting group
on the ratio of the isomers was established, as well as the
verisimilitude of a mechanism involving a hypervalent-
silicon species. In addition, a good correlation between the
isomer ratio and the Hammett constants of the aromatic
substituents could be established. We hope these observa-
tions will promote the application of this new reaction to
little known families of silaheterocycles.
starting
material
prod
yields
[%]b
entry
1
R0
(a/b)a
Boc
6
4a/4b
84
70
63
28
45:55
2
3
4
Alloc
Ts
32
33
34
35a/35b
48:52
36a/36b
0:100
Acknowledgment. C.F. and T.B. are grateful to Janssen
for PhD and Postdoctoral stipends. We also acknowledge
Ms
37a/37b
26:74
ꢀ
the Region de Haute-Normandie and the CRUNCh inter-
regional program for their support of our research projects.
a Ratio a/b determined from the crude 1H NMR spectra. b Isolated
yields.
Supporting Information Available. Experimental meth-
ods for the preparation of compounds 4ꢀ7 and 12ꢀ38
In further attempts to evaluate the mechanism of this
reaction and inparticularthe likelihoodof a pentaorgano-
silicate species such as 11, the siladihydrobenzofuran
3820 was reacted with the lithium carbenoid generated
1
and their characterization. Copies of H and 13C NMR
spectra of the new compounds (4ꢀ7 and 12ꢀ38). This
material is available free of charge via the Internet at
http://pubs.acs.org.
from chloroiodomethane and MeLi LiBr at ꢀ100 °C
3
(Scheme 4).12b
(21) (a) Holmes, R. R. Chem. Rev. 1990, 90, 17–31. (b) Couzijn, E. P.
A; Slootweg, J. C.; Ehlers, A. W.; Lammertsma, K. J. Am. Chem. Soc.
2010, 132, 18127–18140. (c) Moberg, C. Angew. Chem., Int. Ed. 2011, 50,
10290–10292.
ꢀ
(18) Chataigner, I.; Panel, C.; Gerard, H.; Piettre, S. R. Chem.
Commun. 2007, 3288–3290.
(19) For structure, see Supporting Information.
€
(20) Boge, O.; Nietzschmann, E.; Heinicke, J. Phosphorus, Sulfur
Silicon Relat. Elem. 1992, 71, 25–29.
The authors declare no competing financial interest.
Org. Lett., Vol. 14, No. 8, 2012
2077