Tosylates in palladium-catalysed coupling reactions. Application to the synthesis of arylcoumarin inhibitors of gyrase B

Article abstract of DOI:10.1016/S0040-4039(99)02351-5

The palladium-catalysed coupling reaction between tosylate derivatives and organostannanes has been investigated as a methodology for carbon-carbon bond formation. Aryl substituents have been successfully incorporated even in highly functionalised coumarin structures to afford new analogues of the antibiotic novobiocin. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(99)02351-5

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 1543–1547
Tosylates in palladium-catalysed coupling reactions. Application
to the synthesis of arylcoumarin inhibitors of gyrase B
Laurent Schio,∗ Fabienne Chatreaux and Michel Klich
Medicinal Chemistry, Hoechst Marion Roussel, 102 Route de Noisy, 93235 Romainville Cedex, France
Received 16 November 1999; accepted 12 December 1999
Abstract
The palladium-catalysed coupling reaction between tosylate derivatives and organostannanes has been investi-
gated as a methodology for carbon–carbon bond formation. Aryl substituents have been successfully incorporated
even in highly functionalised coumarin structures to afford new analogues of the antibiotic novobiocin. © 2000
Elsevier Science Ltd. All rights reserved.
Keywords: antibacterials; coumarin; sulfonyl compounds; palladium catalyst; organotin reagents.
Novobiocin 1 and clorobiocin 2 are coumarin-derived antibiotics which act as competitive inhibitors
of the bacterial ATP binding gyrase B subunit, blocking negative supercoiling of relaxed DNA.1
These naturally produced antibacterials are extremely active against Gram-positive bacteria especially
methicilin-resistant staphylococcus strains.2 However, their poor pharmacokinetic properties as well as
toxicity issues3 have prevented them from pharmaceutical applications as antibiotics. Consequently,
we have embarked on an extensive programme of chemical modifications4 aiming to improve the
pharmacological and pharmacokinetic properties of this class of compounds.
We report here an efficient methodology for the preparation of 4-arylcoumarin analogues 4. The
key step relies on an unexpected palladium-catalysed coupling reaction between p-toluenesulfonate
derivatives 3 and aryltrialkylstannanes 5.5
∗
Corresponding author. Tel: +00 33 1 49 91 55 80; fax: +00 33 1 49 91 50 87; e-mail: laurent.schio@hmrag.com (L. Schio)
0040-4039/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(99)02351-5
tetl 16279

1544
The leaving entity of choice in transition metal-catalysed coupling reactions with organometallic
reagents is often the triflate group.6 However, we could not find a satisfactory methodology for preparing
the corresponding trifluoromethanesulfonate analogue of 3b (R1=CO2Bn) because of stability issues. We,
therefore, turned our attention to a more stable sulfonate group with the idea that the electron withdrawing
coumarin moiety could decrease the electronic density of the C–OSO2 bond allowing palladium insertion.
Preliminary coupling reactions were attempted with the coumarin-derived tosylate 6 using standard Stille
conditions (see Table 1).7
Table 1
Palladium-catalysed coupling reactions between tosylate 6 and various organostannanes 5

1545
We were delighted to observe that organostannane 5a underwent the coupling reaction with an
acceptable yield (60%). Furthermore, the scope of the tosylate-assisted coupling reaction was extended
to functionalised aryl reagents (5b–d). In addition, the reaction took place effectively with heterocycle-
derived stannanes (5f, 5g) albeit occurring in low yield from the pyridine derived reactant 5e. Among
the organostannanes investigated, only 5h failed to give the desired coupled compound (7h). No further
studies were carried out to optimise these results. The developed methodology was applied promptly to
the synthesis of 4-arylcoumarin analogues of novobiocin.
The preparation of the key coumarin intermediates 9 and 15 from dihydroxycoumarin 8 is depicted in
Scheme 1.
Scheme 1.
It is worthy of note that tosylation of 8 in pyridine occurred predominately at the O-4 position
whereas silylation using triethylamine led exclusively to the 7-silyl ether derivative 11. Introduction
of the carboxy group at the C-3 position was realised by Fries-type rearrangement of the 4-acyloxy
derivative 12.4a,8 Tosylation of 13 was successful in dichloromethane in the presence of triethylamine.
The 7-O t-butyldiphenylsilyl ether was then cleaved cleanly using HF/KF in a water–THF mixture9 to
afford 15.
The desired α-glycosides 17 and 18 were obtained by coupling L-rhamnopyranoside 164c with the
coumarins 9 and 15 using Mitsunobu’s conditions (see Scheme 2).10 Tosylate derivatives 17 and 18 were
then submitted to cross-coupling with organostannanes 5c and 5e in the presence of palladium catalysts.
The catechol moiety was successfully introduced at the coumarin C-4 position from tosylate 18 to
afford the corresponding analogue 19 of novobiocin.11 The yield obtained with Pd(Ph3P)4 as catalyst was
satisfactory considering the high level of functionalisation as well as the unprotected groups in the starting
substrate structure 18. On the other hand, the coupling reaction performed with the pyridine-derived
stannane 5e led to partial degradation of tosylate 17 when catalysed with Pd(Ph3P)4 or PdCl2(Ph3P)2.
Finally, the pyridine containing derivative 20 was accessible in conditions generating in situ zero-valence
palladium (Pd(OAc)2/Ph3P).12
In conclusion, we have developed a new variation to the methodology for carbon–carbon bond forma-
tion by a palladium-catalysed coupling reaction. In this respect, we have shown that the p-toluenesulfonyl
group constitutes an efficient alternative to the widely used trifluoromethanesulfonyl moiety as a leaving
group. Ease of preparation as well as chemical stability contribute to the attractiveness of the tosylate
group over triflate. The developed methodology has been applied to the convergent synthesis of 4-
arylcoumarin analogues of novobiocin in the search for new antibacterial clinical candidates.

1546
Scheme 2.
Acknowledgements
We are grateful to Didier Babin (Medicinal Chemistry) for assistance and to Didier Ferroud (Chemical
Development) for fruitful discussions.
References
1. (a) Lewis, R. J.; Singh, O. M.; Smith, C. V.; Skarzynski, T.; Maxwell, A.; Wonacott, A. J.; Wigley, D. B. EMBO J. 1996, 15,
1412–1420. (b) Ali, J. A.; Jackson, A. P.; Howells, A. J.; Maxwell, A. Biochemistry 1993, 32, 2717–2724. (c) Gellert, M.;
O’Dea, M. H.; Itoh, T.; Tomizawa, J. I. Proc. Natl. Acad. Sci. U.S.A. 1976, 73, 4474–4478.
2. (a) For reviews on resistance in bacteria, see: (a) Resistance to Antibiotics, in Science 1994, 264, 360–393. (b) Smith, T. L.;
Jarvis, W. R. Microbes and Infection 1999, 1, 795–805.
3. Goranov, I.; Avramova, K. Probl. Khig. 1975, 1, 165–170.
4. (a) Laurin, P.; Ferroud, D.; Klich, M.; Dupuis-Hamelin, C.; Mauvais, P.; Lassaigne, P.; Bonnefoy, A.; Musicki, M. Bioorg.
Med. Chem. Lett. 1999, 9, 2079–2084. (b) Laurin, P.; Ferroud, D.; Schio, L.; Klich, M.; Dupuis-Hamelin, C; Mauvais,
P.; Lassaigne, P.; Bonnefoy, A.; Musicki, B. Bioorg. Med. Chem. Lett. 1999, 9, 2875–2880. (c) Ferroud, D.; Collard, J.;
Klich, M.; Dupuis-Hamelin, C.; Mauvais, P.; Lassaigne, P.; Bonnefoy, A.; Musicki, B. Bioorg. Med. Chem. Lett. 1999, 9,
2881–2886.
5. To the best of our knowledge only one application of other arenesulfonates in palladium-catalysed coupling reactions has
been described: Badone, D.; Cecchi, R.; Guzzi, U. J. Org. Chem. 1992, 57, 6321–6323.
6. Ritter, K. Synthesis 1993, 735–762, and references cited therein.
7. (a) Echavarren, A. M.; Stille, J. K. J. Am. Chem. Soc. 1987, 109, 5478–5486. (b) Scott, W. J.; Stille, J. K. J. Am. Chem. Soc.
1986, 108, 3033–3040. (c) Following experimental observations, yields were higher when the reactions were conducted in
the dark.
8. See, for example: (a) Kappe, T.; Schnell, B. J. Heterocyclic Chem. 1996, 33, 663–670. (b) Tabuchi, H.; Hamamoto, T.; Miki,
S.; Tejima, T.; Ichihara, A. Tetrahedron Lett. 1993, 34, 2327–2330. (c) Eisenhauer, H. R.; Link, K. P. J. Am. Chem. Soc.
1953, 75, 2044–2045.
9. To be published elsewhere.
10. (a) For the synthesis of glycosides via the Mitsunobu reaction, see: Roush, W. R.; Lin, X. F. J. Org. Chem. 1991, 56,
5740–5742, and references cited therein. (b) The Mitsunobu reaction has been widely used in our programme for creating the
glycosidic linkage between various saccharides and hydroxylaryl moieties. Usually equivalents of reagents (Ph3P/DEAD)
were necessary for reaction completion. Consequently, laborious purification steps were required to isolate the desired
α-glycoside from complex crude mixtures. For example, pure 17 was obtained after several triturating manipulations in
dichloromethane and ether.
11. From 17, the coupling reaction with 5c realised in the same experimental conditions occurred in 34% yield.
12. Preparation of 20: a mixture of tosylate 17 (129 mg, 0.21 mmol), organostannane 5e (93 mg, 0.25 mmol), Pd(OAc)2 (9.7
mg, 0.019 mmol) and Ph3P (10.4 mg, 0.039 mmol) was stirred overnight in refluxing dioxane (5 ml). After concentration at

1547
reduced pressure the reaction mixture was purified by flash chromatography on silica gel (hexane:ethyl acetate: 30:70 v/v)
to afford 20 (30 mg, 27% yield). TLC, SiO2, hexane:ethyl acetate: 30:70 (v/v), Rf=0.27; IR (CHCl3, ν cm−1): 3608 (OH),
3448 (C_C-NH), 1714, 1696 (C_O), 1603, 1567, 1491 (C_C). 1H NMR (300 MHz, CDCl3, ppm): δ 9.07 (m, NH), 8.78
(m, 1H), 7.88 (t, J=8 Hz, 1H), 7.6–7.4 (m, 3H), 7.06 (d, J=9 Hz, 1H), 6.76 (m, 1H), 6.18 (s, 1H), 5. 92 (m, 1H), 5.44 (s, 1H),
5.37 (dd, J=10 Hz, J=3 Hz, 1H), 3.94 (m, 1H), 3.70 (m, 1H), 3.49 (s, 3H), 3.46 (t, J=9 Hz, 1H), 2.40 (s, 3H), 2.33 (s, 3H),
1.26 (d, J=6 Hz, 3H). MS m/e (relative intensity): 521 (MH+, 15), 268 (53), 254 (35), 108 (100).