Reaction of electron-deficient N-sulfinylanilines with chiral α-hydroxy acids: A new process for the synthesis of enantiomerically pure α-hydroxy amides

Article abstract of DOI:10.1016/S0040-4039(00)01011-X

A practical procedure to prepare enantiomerically pure α-hydroxy amides from chiral α-hydroxy acids and electron-deficient anilines via N-sulfinylaniline derivatives has been developed. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(00)01011-X

Tetrahedron Letters 41 (2000) 6017±6020
Reaction of electron-de®cient N-sul®nylanilines with chiral
a-hydroxy acids: a new process for the synthesis of
enantiomerically pure a-hydroxy amides
Ramakrishnan Chidambaram,a,* Jason Zhu,a Kumar Penmetsa,b David Kronenthala
and Joydeep Kanta
aDepartment of Process Research, Bristol-Myers Squibb, Pharmaceutical Research Institute, One Squibb Drive,
PO Box 191, New Brunswick, New Jersey 08903-0191, USA
bDepartment of Analytical Research and Development, Bristol-Myers Squibb, Pharmaceutical Research Institute,
One Squibb Drive, PO Box 191, New Brunswick, New Jersey 08903-0191, USA
Received 20 April 2000; revised 7 June 2000; accepted 8 June 2000
Abstract
A practical procedure to prepare enantiomerically pure a-hydroxy amides from chiral a-hydroxy acids
and electron-de®cient anilines via N-sul®nylaniline derivatives has been developed. # 2000 Elsevier Science
Ltd. All rights reserved.
As part of an ongoing e€ort to develop a practical synthesis of a novel RAR-g speci®c retinoid
agonist (1), an ecient coupling reaction between the homochiral a-hydroxy acid 2 and the
electron-de®cient aniline 3 was needed.
A number of methods are known for the conversion of enantiomerically pure a-hydroxy acids
to the corresponding amides.1 The most prevalent is in-situ conversion of the hydroxy acid to the
bis-trimethylsilyl derivative followed by conversion to the acid chloride using oxalyl chloride and
treatment with the appropriate amine.2 In our hands, under a variety of conditions, the major
product from this protocol was the a-chloro-amide 4.3 Alternatively, activation of acid 2 via the
dioxolanedione4 5 or acetonide5 6 and subsequent treatment with aniline 3 a€orded the desired
amide 9 (Table 1) in low yields. Standard coupling procedures using carbodiimides in conjunction
* Corresponding author.
0040-4039/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)01011-X

6018
with 1-hydroxybenzotriazole (HOBT)6 and DMAP a€orded only trace amounts of the product.
Employment of carbonyldiimidazole,7 benzotriazol-1-yloxytris(dimethylamino)phosphonium
hexa¯uorophosphate (BOP),8 or 2-ethoxy-1-ethoxycarbonyl-1,2-dihydroquinoline (EEDQ)9 as
coupling agents was also unsuccessful. Activation of the a-hydroxy acid with isobutylchloro-
formate or pivaloyl chloride gave low yields of the desired compound, presumably due to the
poor nucleophilicity of the aniline 3.10
Table 1
Coupling reactions between chiral a-hydroxy acids and N-sul®nylanilines
In view of the unsuccessful coupling reactions with the activated carboxylic acid, reaction
between the a-hydroxy acid and an `activated' form of the aniline was pursued. Kim and Shin
have reported the preparation of racemic a-hydroxy amides in high yields by simply admixing
a-hydroxy acids and N-sul®nylanilines derived from aniline, p-toluidine, and p-chloroaniline.11
The reaction is reported to proceed via protonation of the N-sul®nylaniline followed by reaction
with the hydroxy acid leading to intermediate 8 (Scheme 1).

6019
Scheme 1.
In our case, treatment of the electron-de®cient aniline 3 with SOCl2 using the protocol described
by Kim12 gave N-sul®nylaniline 7 (Table 1) in 95% yield. However, treatment of 7 with 2 under
the reported conditions (no added catalyst) gave only unreacted starting materials. On the other
hand, contrary to Kim's report that the reaction does not proceed in the presence of a basic catalyst,
we observed rapid coupling when the reaction was performed in the presence of bases such as
triethylamine or diisopropylethylamine. In this case, the base catalyzes the initial reaction of the
hydroxy acid with the relatively electrophilic N-sul®nylaniline in contrast to the acid-catalyzed
process depicted above. Unfortunately, the product isolated from these reactions generally had a low
ee (<80%), presumably due to racemization at the chiral center in the intermediate corresponding
to 8. After screening a variety of other bases, we found that the reaction in the presence of weakly
basic 1,2,4-triazole (pKBH+ 2.3) a€orded the desired product in high yield without racemization.13
Thus, treatment of 7 with 2 in the presence of 1,2,4-triazole in dichloromethane a€orded the
desired coupled product 9 (Table 1) in 77% yield and with >99% ee. This methodology was
extended to couple other homochiral a-hydroxy acids with N-sul®nylanilines derived from
electron-de®cient anilines in high yields and without detectable racemization (Table 1).14
In conclusion, we have developed an ecient, racemization-free coupling procedure to prepare
enantiomerically pure a-hydroxy amides which are otherwise dicult to prepare using standard
coupling procedures.
Typical procedure (preparation of 9): To a 500 mL, three-necked ¯ask equipped with a
mechanical stirrer, gas inlet adapter and temperature probe was charged chiral hydroxy acid 2 (20 g,
76.3 mmol), 1,2,4-triazole (7.4 g, 106 mmol) and dichloromethane (150 mL) under an atmosphere
of argon. The reaction mixture was stirred at ambient temperature until a clear solution was
obtained. The solution was cooled to 0ꢀC and N-sul®nylaniline 715 (23 g, 107 mmol) was added to
the reaction mixture. The reaction mixture was stirred for 3 h at 0ꢀC. After completion of the
reaction (HPLC), the dichloromethane was evaporated. The residual solid was dissolved in a 2:1
mixture of heptane:methyl t-butyl ether and washed with 3N HCl, saturated NaHCO3 and brine.
Removal of solvent followed by crystallization (cyclohexane) provided 24 g of 9 (77% yield).
Compounds 10, 11, 13 and 15 were puri®ed by chromatography.
References
1. Bodanszky, M. Principles of Peptide Synthesis; Springer-Verlag: New York, 1984, pp. 9.

6020
2. (a) Kelly, S. E.; LaCour, T. G. Synthetic Communications 1992, 22, 859. (b) Smith, M. Tetrahedron Lett. 1989, 30,
313. (c) Rosen, T.; Watanabe, M.; Heathcock, C. H. J. Org. Chem. 1984, 49, 3657. (d) Wissner, A.; Grudzinskas,
C. V. J. Org. Chem. 1978, 43, 3972.
3. No attempts were made to determine the enantiopurity of 4.
4. Toyooka, K.; Takeuchi, Y.; Kubota, S. Heterocycles 1989, 29, 975.
5. Khalij, A.; Nahid, E. Synthesis 1985, 1153.
6. Konig, W.; Geiger, R. Chem. Ber. 1970, 103, 788.
7. Staab, H. A. Liebigs Ann. Chem. 1957, 609, 75.
8. Bates, A. J.; Galpin, I. J.; Hallett, A.; Hudson, D.; Kenner, G. W.; Ramage, R. Helv. Chim. Acta 1975, 58, 688.
9. Belleau, D.; Malek, G. J. Amer. Chem. Soc. 1968, 90, 1651.
10. (a) Vaughan Jr., J. R.; Osato, R. L. J. Amer. Chem. Soc. 1951, 73, 5553. (b) Zaoral, M. Coll. Czech. Chem.
Commun. 1962, 27, 1273.
11. Shin, J. M.; Kim, Y. H. Tetrahedron Lett. 1986, 27, 1921. The acids used in this study were racemic with the
exception of d-tartaric acid. In this case, the enantiomeric purities of the product amides were not reported.
12. Kim, Y. H.; Shin, J. M. Tetrahedron Lett. 1985, 26, 3821.
13. 1,2,4-Triazole was added to a mixture of the hydroxy acid 2 and N-sul®nylaniline 7 in CD2Cl2 in an NMR tube. A
1
new peak was observed by H NMR at ꢀ 4.2, which, by CS Chem Draw Pro, estimates were suggestive of an
intermediate such as 8 (methine proton). This peak was not observed in the absence of 1,2,4-triazole.
14. All new compounds were fully characterized by NMR, HRMS and/or CHN analysis. Chiral purity of compounds
was determined using chiral HPLC: Chiralpack AD, 4.6Â250 mm ID, 10 mM particle diameter column; 95%
1
hexane/5% isopropanol as mobile phase. Compound 2: H NMR (CDCl3, 500 MHz) ꢀ 1.25±1.27 (overlapping
singlets, 12H), 1.67 (s, 4H), 5.19 (s, 1H), 7.17 (dd, J=8, 2 Hz, 1H), 7.30 (d, J=8 Hz, 1H), 7.35 (d, J=2 Hz, 1H).
1
Elemental analysis: theoretical: C=73.05, H=8.45; found: C=73.25; H=8.36. Compound 7: H NMR (CDCl3,
500 MHz) ꢀ 3.93 (s, 3H), 7.80±7.90 (m, 2H), 8.14 (dd, J=8, 8 Hz, 1H) HRMS: calculated: 215.0052; found:
215.0052. Compound 9: 1H NMR (CDCl3, 500 MHz) ꢀ 1.25±1.30 (overlapping singlets, 12H), 1.68 (s, 4H), 3.51 (d,
J=2 Hz, 1H), 3.89 (s, 3H), 5.17 (d, J=2 Hz, 1H), 7.22 (dd, J=8, 2 Hz, 1H), 7.33 (d, J=8 Hz, 1H), 7.4 (d, J=2 Hz,
1H), 7.75 (dd, J=11, 2 Hz, 1H), 7.80 (d, J=11 Hz, 1H), 8.44 (apparent t, J=8 Hz, 1H), 8.86 (d, J=2 Hz, 1H)
1
HRMS (M+H): calculated: 414.2081; found: 414.2066. Compound 10: H NMR (CDCl3, 300 MHz) ꢀ 3.34 (d,
J=3.Hz, 1H), 3.9 (s, 3H), 5.26 (d, J=3 Hz, 1H), 7.3±7.45 (m, 3H), 7.46±7.55 (m, 2H), 7.76 (dd, J=11, 2 Hz, 1H),
7.81 (d, J=8 Hz, 1H), 8.45 (t, J=8.1 Hz, 1H), 8.81 (broad s, 1H) HRMS: calculated: 303.0907; found: 303.0940.
Compound 11: 1H NMR (CDCl3, 300 MHz) ꢀ 2.61 (d, J=4 Hz, 1H), 2.97 (dd, J=14, 9 Hz, 1H), 3.38 (dd, J=14, 4
Hz, 1H), 3.91 (s, 3H), 4.48 (ddd, J=9, 4, 4 Hz, 1H), 7.19±7.4 (m, 5H), 7.76 (dd, J=12, 2 Hz, 1H), 7.85 (d, J=9 Hz,
1H), 8.53 (apparent t, J=8 Hz, 1H), 8.89 (broad s, 1H) HRMS: calculated: 317.1063; found: 317.1064. Compound
1
12: H NMR (CDCl3, 300 MHz) ꢀ 7.93 (d, J=9 Hz, 2H), 8.30 (d, J=9 Hz, 2H) HRMS: calculated: 303.0907;
1
found: 303.0940. Compound 13: H NMR (DMSO-d6, 300 MHz) ꢀ 2.9 (dd, J=14, 8 Hz, 1H), 3.09 (dd, J=14, 4
Hz, 1H), 4.34 (m, 1H), 6.02 (broad d, J=5 Hz, 1H), 7.19±7.35 (m, 5H), 8.01 (d, J=9 Hz, 2H), 8.25 (d, J=9 Hz,
1
2H), 10.39 (s, 1H) HRMS: calculated: 286.0954; found: 286.0946. Compound 14: H NMR (CDCl3, 300 MHz) ꢀ
7.30 (dd, J=9, 2 Hz, 1H), 7.53 (d, J=2 Hz, 1H), 8.35 (d, J=9 Hz, 1H) HRMS: calculated: 206.9312; found:
206.9318. Compound 15: 1H NMR (CDCl3, 300 MHz) ꢀ 1.2±1.3 (overlapping s, 12H), 1.68 (s, 4H), 3.22 (broad s,
1H), 5.19 (s, 1H), 7.2±7.3 (m, 2H), 7.3±7.41 (m, 1H), 8.35 (d, J=9 Hz, 1H), 8.77 (broad s, 1H) HRMS: calculated
(using 37Cl): 407.1262; found: 407.1226.
15. A general procedure for preparing N-sul®nylanilines from the corresponding aniline is described in Ref. 12.