New oxidative transformations of phenolic and indolic oxazolines: An avenue to useful azaspirocyclic building-blocks

Article abstract of DOI:10.1021/jo000341v

The oxidative cyclization of a phenolic amide to a spirolactam has long been regarded as an 'impossible' reaction, because exposure of the substrates to a variety of oxidants results in formation of spirolactones with consequent loss of the amine segment. We recently communicated that this heretofore unknown transformation may be achieved by oxidation of oxazoline analogues of phenolic and indolic amides. Herein, we provide full details of our work.

Full text of DOI:10.1021/jo000341v

J . Org. Chem. 2000, 65, 4397-4408
4397
New Oxid a tive Tr a n sfor m a tion s of P h en olic a n d In d olic
Oxa zolin es: An Aven u e to Usefu l Aza sp ir ocyclic Bu ild in g Block s
Norbert A. Braun,† Malika Ousmer,‡ J onathan D. Bray,‡ Denis Bouchu,‡ Karl Peters,§,|
Eva-Maria Peters,§,| and Marco A. Ciufolini*,†,‡
Department of Chemistry, MS 60, Rice University, 6100 Main Street, Houston, Texas 77005-1892,
Laboratoire de Synthe`se et Me´thodologie Organiques, Universite´ Claude Bernard Lyon 1 et EÄ cole
Supe´rieure de Chimie, Physique, Electronique de Lyon, 43, Boulevard du 11 Novembre 1918,
69622 Villeurbanne cedex, France, and Max-Planck-Institut fu¨r Festko¨rperforschung, Heisenbergstrasse 1,
70506 Stuttgart, Germany
Received March 9, 2000
The oxidative cyclization of a phenolic amide to a spirolactam has long been regarded as an
“impossible” reaction, because exposure of the substrates to a variety of oxidants results in formation
of spirolactones with consequent loss of the amine segment. We recently communicated that this
heretofore unknown transformation may be achieved by oxidation of oxazoline analogues of phenolic
and indolic amides. Herein, we provide full details of our work.
Sch em e 1
In tr od u ction
A number of recently discovered natural products
display novel architectures embodying variations on the
theme of spiroheterocycle 1 (Scheme 1).1 The assembly
of this apparently simple structure, in a format consonant
with the conditions imposed by the target molecules, may
result in fairly elaborate routes if standard synthetic
methodology is utilized.
Sch em e 2
An especially direct avenue to subtarget 1 would
materialize if a phenolic amine such as 2 could be induced
to undergo oxidative cyclization to 4, which could then
be reduced to 1. The oxidative cyclization of amide 3 to
spirolactam 5 would also be serviceable as an entry to 5
(Scheme 2).
The desirability of the transformations of Scheme 2
was surely recognized as early as 1987, when Kita and
collaborators disclosed a pioneering study of the oxidation
of phenolic amides with iodobenzene diacetate (“DIB”).2
However, these workers observed that compounds 3 are
converted to lactones 7 under oxidative conditions, and
not to spirolactams 5. Seemingly, it is the oxygen atom
of the amide that intercepts the electrophilic intermediate
arising through interaction of the phenol with DIB. The
presumed primary product, iminolactone 6 (Scheme 3),
Sch em e 3
is rapidly hydrolyzed during workup to yield 7. The Kita
results have been fully confirmed in the course of this
study.
* To whom correspondence should be addressed. E-mail: ciufi@cpe.fr.
Fax: (int) 33 (0)4 72 43 29 63.
† Rice University.
‡ Universite´ Claude Bernard Lyon 1 et EÄ cole Supe´rieure de Chimie,
Physique, Electronique de Lyon.
The preferential formation of spirolactones over spiro-
lactams in Kita-type oxidations is likely due to an
electronic effect. Amide resonance causes accumulation
of negative charge on the carbonyl oxygen, which becomes
basic and nucleophilic, while the nitrogen atom is actu-
ally electron-deficient and thus unable to express nu-
cleophilicity. Knapp encountered an analogous difficulty
in his study of iodolactamization of olefins (cf. 8 f 9,
Scheme 4), but he was able to harness the effects
responsible for the reactivity of the oxygen atom of an
amide and cause them to operate in favor of the nitrogen
atom, by engaging an imino analogue of the amide, such
as an imidate, as the nucleophile in such reactions (cf.
§ Max-Planck-Institut fu¨r Festko¨rperforschung.
| To whom correspondence regarding X-ray data should be ad-
dressed.
(1) E.g., FR901483: (a) Sakamoto, K.; Tsujii, E.; Abe, F.; Nakanishi,
T.; Yamashita, M.; Shigematsu, N.; Izumi, S.; Okuhara M. J . Antibiot.
1996, 49, 37 (isolation). (b) Snider, B. B.; Lin, H.; Foxman, B. M. J .
Am. Chem. Soc. 1999, 121, 7778 (synthesis). TAN-1251: (c) Shirafuji,
H.; Tsubotani, S.; Ishimaru, T.; Harada, S. World Patent WO 91/13887
(1991) to Takeda Chemical Industries, Ltd. (isolation). (d) Nagumo,
S.; Nishida, A.; Yamazaki, C.; Murashige, K.; Kawahara, N. Tetrahe-
dron Lett. 1998, 39, 4493 (synthesis). Cyclindricines: (e) Blackman,
A. J .; Li, C.; Hockless, D. C. R.; Skelton, B. H. Tetrahedron 1993, 49,
8645 (isolation). (f) Molander, G. A.; Ro¨nn, M. J . Org. Chem. 1999, 64,
5183 (synthesis).
(2) Tamura, Y.; Yakura, T.; Haruta, J .-I.; Kita, Y. J . Org. Chem.
1987, 52, 2, 3927.
10.1021/jo000341v CCC: $19.00 © 2000 American Chemical Society
Published on Web 06/16/2000

4398 J . Org. Chem., Vol. 65, No. 14, 2000
Braun et al.
Sch em e 4
Sch em e 6a
Sch em e 5
a
Reagents and conditions: (a) excess amine, no solvent, 150
°C, 95%; (b) THF, reflux, 94%; (c) pyridine, 96%; (d) CH2Cl2, rt,
78%; (e) K2CO3, MeOH, rt, 89%.
Sch em e 7
10 f 11).3 Application of similar logic ultimately allowed
us to realize the transformation depicted in Scheme 2.
In this paper, we provide a full account of studies directed
toward the development of this chemistry.4 The tech-
niques developed in the course of this investigation also
proved to be applicable to indolic (as opposed to phenolic)
systems,5 thereby permitting the creation of novel het-
erocyclic systems. The new transformations are likely to
facilitate the synthesis of structures of the type 1 to a
substantial extent.6
interest here is that formation of 13 involved transetheri-
fication of intermediate imino ethyl ether 19 during
deacetylative release of the phenol with K2CO3/MeOH.
Treatment of 12-14 with DIB in trifluoroethanol
(“TFE”) under Kita conditions8 afforded none of the
desired 16. Compound 12 furnished a ca. 3:1 mixture of
bicyclic amines 20 (major product, Scheme 7) and 21
(minor). The structure of 21, a nicely crystalline material,
was confirmed by X-ray crystallography.9 By contrast,
imidazoline 14 afforded a ca. 5:1 mixture of compounds
22 and 23. These results are consistent with a mecha-
nism in which trifluoroethanol or acetate ion/acetic acid
released from DIB intercept an electrophilic intermediate
produced through interaction of the phenol with the
oxidant. This intermediate is naively represented in
Scheme 8 as structure 24. The presumed primary prod-
ucts thus obtained, dienones 25-26, then undergo rapid
Michael cyclization to 20-23.
The reason nucleophilic capture of electrophile 24 by
the nitrogen atom is not competitive with the pathways
leading to presumed intermediates 25-26 remains un-
clear. Reactions involving DIB develop significant acid-
ity: possibly, basic groups such as amine or amidine exist
in protonated form during the reaction. This would
suppress the nucleophilicity of the N atoms and it would
not allow them to compete effectively with the solvent
or acetate ion for 24. Conduct of the reaction in the
presence of basic agent or acid scavengers, a plausible
Ba ck gr ou n d
Plausible avenues to the desired subgoals were initially
explored by studying the oxidative cyclization of amine
12, imino ether 13, and imidazoline 147 to spirocycles of
general structure 16 (Scheme 5). Phenolic imines 15 (R
) aryl, alkyl) were also considered as interesting sub-
strates for the new transformation; however, their prepa-
ration proved to be troublesome and they were not
further investigated. By contrast, the synthesis of 12-
13 was straightforward (Scheme 6). The only point of
(3) (a) Knapp S. In Advances in Heterocyclic Natural Product
Synthesis; Pearson, W. H., Ed.; J AI Press: Greenwich, CT, 1996; Vol.
3, p 57. It is well recognized that imidates react preferentially at
nitrogen with
a variety of electrophiles; cf.: (b) Tennant, G. In
Comprehensive Organic Chemistry; Barton, D. H. R., Ollis, W. D., Eds.;
Pergamon Press: Oxford, UK, 1979; Vol. 2, Chapter 8, pp 385-590,
see esp. pp 490 ff. (c) Kantlehner, W. In Comprehensive Organic
Synthesis; Trost, B. M., Ed.; Pergamnon Press: Oxford, UK, 1991; Vol.
6, Chapter 2.7, pp 485-599, see esp. pp 529 ff. and references cited in
b-c.
(4) Braun, N. A.; Ciufolini, M. A.; Peters, K.; Peters, E.-M. Tetra-
hedron Lett. 1998, 39, 4667.
(5) Braun, N. A.; Bray, J .; Ciufolini, M. A. Tetrahedron Lett. 1999,
40, 4985.
(6) Alternative methods for the preparation of related spirohetero-
cycles: (a) Bryce, M. R.; Gardiner, J . M.; Horton, P. J .; Smith, S. A. J .
Chem. Res., Synop. 1989, 1. (b) Boivin, J .; Yousfi, M.; Zard, S. Z.
Tetrahedron Lett. 1997, 38, 5985. (c) Cossy, J .; Bouzide, A. Tetrahedron
1997, 53, 5775.
(8) Kita, Y.; Tohma, H.; Kikuchi, K.; Inagaki, M.; Yakura, T. J . Org.
Chem. 1991, 56, 435.
(7) McFarland, J . W.; Howes, H. L., J r. J . Med. Chem. 1972, 15,
365.
(9) Peters, K.; Peters, E.-M.; Braun, N. A.; Ciufolini, M. A. Z.
Kristallogr. NCS 1999, 214, 273.

Oxidative Transformations of Phenolic and Indolic Oxazolines
J . Org. Chem., Vol. 65, No. 14, 2000 4399
Sch em e 8
Sch em e 10
Sch em e 9
carbonyl oxygen of the amide, followed by hydrolysis
during workup (cf. Scheme 3). However, the above
scenario fails to account for formation of imide 29. At this
time, we believe that a conformational effect controls the
genesis of 27-29.
remedy for the above difficulties, proved to be entirely
inappropriate. Insoluble bases, such as NaHCO3, had no
effect on the reaction, probably because the rate at which
they remove the acid present in solution it quite slow in
a purely organic medium. Likewise, acid scavengers such
as propylene oxide were ineffective. Soluble bases, such
as triethylamine, promoted the formation of intractable
mixtures. Numerous experiments, as well as evidence
accumulated by other workers in the field, suggest that
the acidity of the medium may be important for the
success of DIB oxidations. Indeed, adjuvants such as
TMSOTf have been utilized in other types of DIB-
promoted processes.10 It is noteworthy that in the above
reactions a presumed electrophilic intermediate arising
from the phenol is intercepted by TFE, an allegedly
nonnucleophilic solvent. We have observed a number of
similar events in the course of this study (vide infra). This
is seemingly the first report describing the capture of
reactive intermediates formed during oxidation of aro-
matic substrates with hypervalent iodine reagents by
TFE. Use of hexafluoro-2-propanol, instead of TFE, as
the solvent for the reactions of Scheme 7 delivered only
acetate products to the complete exclusion of materials
resulting from capture of electrophilic intermediates by
the solvent. The use of hexafluoro-2-propanol for DIB
reaction was first described by Kita.
The behavior of imino ether 13 differed significantly
from that of 12/14. Reaction with DIB afforded a nearly
equimolar mixture of lactone 27 and amide 28, ac-
companied by a small quantity of the noteworthy imide
29 (Scheme 9). No products of trifluoroethoxylation were
apparent in this case. A superficial explanation for the
formation of the observed products may invoke initial
decomposition of the starting 13 through alcohol ex-
change with the solvent, so that methanol would be
liberated in the reaction medium.11 Methoxylated com-
pounds 28-29 could then be accounted for by capture of
intermediates of the type 24 by free MeOH, while lactone
27 would result through interception of 24 by the
Molecular mechanics calculations (MM+ force field)12
indicate that the most stable conformer of iminoether 30,
a computationally better tractable mimic of 13, is 30a
(Scheme 10). In this zero-energy conformation, the N-
alkyl group is trans to the oxygen atom. Structure 30d
corresponds to the most energetic conformer of the
iminoether, about 4.4 kcal/mol above 30a , while 30b is
estimated to be about 3 kcal/mol more energetic than
30a . Isomerization of 30a to 30b/30d may be problematic,
because the rate of E-Z isomerization of imines, oximes,
and related species in which the nitrogen atom is formally
an sp2 hybrid is notoriously slow.13 Scheme 10 also shows
that cyclization of 13 through N-capture of an electrophile
of the type 24 is possible only from conformers 13b or
13d . The slow kinetics of iminoether isomerization, in
combination with the significant energy demand of 13b/
13d relative to 13a , may create such a small population
of reactive conformers that the desired mode of cycliza-
tion of 13 is essentially suppressed. Under these condi-
tions, reactive species 24 could be captured by the oxygen
atom of the imidate to give oxonium ion 31. This
intermediate may react with acetate ion (path a , Scheme
11) to form iminolactone 32, which would be rapidly
hydrolyzed to 27 upon workup. Alternatively (path b),
31 could fragment to nitrilium ion 33, which may be
intercepted by the solvent or by acetate ion to yield 34
or 35, respectively. Both the new iminoether 34 and
iminoester 35 would be easily hydrolyzed to 28 upon
workup; however, compound 35 could also undergo an
NfO acetyl wanderung. This would account for forma-
tion of imide 29, variable amounts of which could also
suffer hydrolysis during workup to yield more 28.
products of hydroxylation of the substrate (cf. 25/26, Z ) H) would
also have resulted if sufficient moisture were present in the system,
as it happens indeed when similar phenols of the type are exposed to
DIB in moist trifloroethanol.
(12) All computational work described herein was carried out with
the PC-based Hyperchem package, available from Hypercube, Inc.,
Ontario, Canada.
(10) Kita, Y.; Egi, M.; Okajima, A.; Ohtsubo, M.; Takada, T.; Tohma,
H. J . Chem. Soc., Chem. Commun. 1996, 1491.
(11) It may also be surmised that adventitious moisture might have
promoted hydolysis of the imino ether, resulting in liberation of MeOH.
However, it is not likely that the reaction medium contained enough
water to permit formation of 27-29 in the observed yields. In addition,
(13) Cf., e.g.: (a) Lambert, J . B.; Takeuchi, Y. Cyclic Organonitrogen
Stereodynamics; VCH: New York, NY, 1992; pp 76 ff. (b) Tennant, G.
In Comprehensive Organic Chemistry; Barton, D. H. R., Ollis, W. D.,
Eds.; Pergamon Press: Oxford, UK, 1979; Vol. 2 (Sutherland, I. O.,
Ed.), Chapter 8, pp 385-590. See especially pp 396-7 and references
therein.

4400 J . Org. Chem., Vol. 65, No. 14, 2000
Braun et al.
Sch em e 11
Sch em e 13
ylphosphine oxide is sometimes troublesome. The oxazo-
line may also be made by cyclization of a preformed
N-hydroxyethyl amide 39 with the Burgess reagent,15 as
described by Wipf,16a but in this case the phenol must be
blocked as the acetate ester; otherwise the yield of
oxazoline drops to less than 10%. It appears that the free
phenol reacts with the Burgess reagent to form water-
soluble sulfate esters that are not readily cleaved back
to desired 38, that are easily lost upon workup, and from
which retrieval of 38 is difficult. Compound 40 may be
safely converted to 38 by treatment with methanolic K2-
CO3.
Sch em e 12a
Addition of 38 to a solution of DIB in trifluoroethanol
at 25 °C (Kita conditions, Scheme 13) induced rapid
conversion to spirolactam 41 in about 50% crude yield
(NMR). The yield dropped when the reaction was run at
higher (>25 °C) or lower (0 °C) temperatures, or when
solid DIB was added to a solution of substrate. The
balance of the starting oxazoline was converted to
complex oligomeric materials. It must be noted that
meticulous adherence to the workup technique prescribed
by Kita is essential in order to obtain maximum yields
of product. The reaction must thus be quenched with solid
NaHCO3, then filtered, and concentrated. A standard
aqueous workup protocol results in much lower yields.
As noted earlier, conduct of the reaction in the presence
of (insoluble) NaHCO3 or (soluble) propylene oxide had
no effect on product yields, whereas complex, intractable
mixtures resulted in the presence of triethylamine, and
the yield of 41 dropped to less than 5% (NMR). It is
noteworthy that oxidation of 38 with PhI(OCOCF3)2
(“PIFA”) produced only small amounts of 41, accompa-
nied by much polymeric material. Once again, adjuvants
such as NaHCO3, propylene oxide, or triethylamine had
no beneficial effect on yields or product distribution.
a
Reagents and conditions: (a) MeCN/pyr, Et3N, rt, 60-70%;
(b) aq NaOH, 90-95%; (c) BOP-Cl, Et3N, CH2Cl2, 0 °C to rt, 70-
75%; (d) THF, 70 °C (tube), 65-75%; (e) K2CO3, rt, 1 h, 70-80%.
The above results suggested that a cyclic analogue of
an imino ether, e.g., an oxazoline, may resolve the
conformational difficulties presumed to be at the root of
the behavior of 13. This proved to be the case.
Oxid a tion of P h en olic Oxa zolin es
Spirolactam 41 thus obtained underwent spontaneous
Michael cyclization to a single diastereomer of tricyclic
compound 42 upon standing. The stereochemistry of this
material was ultimately confirmed by X-ray diffractom-
Oxazoline 38 (Scheme 12) was prepared in one step
from commercial 3-(4-hydroxyphenyl)propionic acid and
L-phenylalaninol, as detailed by Vorbru¨ggen.14 A signifi-
cant advantage of this method is that no protection of
the phenol was necessary during the reaction. However,
separation of the oxazolines from coproduced triphen-
(15) (a) Atkins, G. M.; Burgess, E. M. J . Am. Chem. Soc. 1968, 90,
4744. (b) Burgess, E. M.; Penton, H. R., J r.; Taylor, E. A.; Williams,
W. M. Organic Syntheses; Wiley: New York, 1988; Collect. Vol. VI, p
788.
(14) (a) Vorbru¨ggen, H.; Krolikiewicz, K. Tetrahedron Lett. 1981,
22, 4471. (b) Vorbru¨ggen, H.; Krolikiewicz, K. Tetrahedron 1993, 49,
9353.
(16) (a) Wipf, P.; Miller, C. P. Tetrahedron Lett. 1992, 33, 907. (b)
Wipf, P.; Kim, Y.; Goldstein, D. M. J . Am. Chem. Soc. 1995, 117, 11106.
(c) Wipf, P.; Li, W. J . Org. Chem. 1999, 64, 4576.

Oxidative Transformations of Phenolic and Indolic Oxazolines
J . Org. Chem., Vol. 65, No. 14, 2000 4401
Sch em e 14
Sch em e 16
Sch em e 15
serious syn-pentane-like interaction between axial sub-
stituents.16 Even a molecular mechanics simulation of the
transition states for cyclization of 41 revealed a prefer-
ence in favor of the structure shown (∆E ≈ 1.1 kcal/
mol),19a,b which was thus predicted to be favored both on
kinetic and on thermodynamic grounds.20
etry of the hydrogenated derivative of 42, ketone 43,17
but it was anticipated to be as depicted in Scheme 13 on
several grounds. First of all, a molecular mechanics study
(MM+ force field)12 of ring system 44 revealed an innate
preference for conformer 44a over 44b (∆E ) 12.9 kcal/
mol; Scheme 14, H atoms not shown for greater clarity).
Although in 44b the nitrogen atom is equatorial and the
ethano branch of the γ-lactam is axial in the cyclohex-
enone unit, both morpholine and cyclohexenone rings
may adopt a chair conformation. By contrast, placement
of ethano segment at the equatorial position of the
cyclohexenone and amide nitrogen at the axial position
creates an extremely strained conformer 44b, wherein
the rigidity of the amide group forces the morpholine
subunit into a boat form. Cyclization of 41 can produce
diastereomeric compounds 42 (observed) and 45 (not
observed, Scheme 15), which on the basis of the above
calculations would be expected to exist largely, if not
exclusively, as conformers of the type 44a . The N-
acylmorpholine subunit in these molecules is conforma-
tionally similar to an N-acylpiperidine. It is well-known
that, in order to minimize severe nonbonding interactions
with the amide carbonyl group,18 substituents adjacent
to the N atom of N-acylpiperidines strongly favor the
axial position, even if a syn-pentane-like interaction with
other axial substituents develops as consequence. The
benzyl substituent occupies the axial position (favored
in this case) in 42, whereas it is forced to an equatorial
position (unfavorable) in 45. Compound 42 was thus
anticipated to form as the major product. This conclusion
was further supported by molecular mechanics calcula-
tions with 42 and with its methyl analogue 46. In both
cases, the best diastereomer (∆E ≈ 1.4 kcal/mol)19a is the
one in which all substituents R to nitrogen, namely the
methyl or benzyl group as well as the cyclohexenone ring
branch, are axial in the morpholine ring; this despite the
Contact with silica gel during chromatographic puri-
fication of 41 accelerated conversion to 42. This propen-
sity to cyclize was observed in all subsequent compounds
of analogous structure, and while cyclization might be
useful as a means to differentiate the diastereotopic π
bonds of the dienone system, in our case this event was
undesirable. Cyclization was readily suppressed by acetyl-
ation of 41 without prior purification. It should be noted
that bases stronger than pyridine must be avoided during
acetylation. Thus, exposure of 41 to triethylamine, used
in conjunction with Ac2O or AcCl, promoted rapid forma-
tion of 42, while acetylation with Ac2O/pyridine/DMAP
proceeded without incident. The yield of 48 from 41 for
the two step sequence, DIB oxidation/acetylation, was
47% after chromatography. The X-ray crystal structure
of acetate 48 appears in Scheme 16.
The new reaction was explored with representative
oxazolines 51-53 (Scheme 17), which were obtained
either by the Vorbru¨ggen method from an appropriate
carboxylic acid/1,2-amino alcohol pair (51, 53), or by the
Wipf procedure (52). Tyrosine-derived compounds 49-
50 were stereochemically labile in the presence of acids.
Even exposure to silica gel promoted erosion of optical
integrity. This well-known propensity of oxazolines16
ruled against extensive purification prior to oxidation.
Oxidative cyclization of 51 and acetylation afforded 54
in 42% chromatographed yield. Thus, substitution on the
oxazoline ring constitutes neither a requirement for, nor
an obstacle in, oxidative cyclization. However, the sub-
strate may not contain additional functionality capable
of competing for an electrophilic species of the type 24
via a 5- or 6-centered transition state. For instance,
oxazoline 52, obtained from N-BOC-tyrosine, cyclized to
afford 55 in only 22% yield. The low yield of this reaction
is attributable, at least in part, to participation of the
carbamate in the capture of electrophile 57 (Scheme 18).
Workup of the resulting 58 and consequent hydrolytic
cleavage of the oxazoline may then lead to a variety of
byproducts. In sharp contrast, cyclization of tosylamide
53 proceeded in 41% yield, since the sulfonyl protecting
group is insufficiently nucleophilic to interact with a
positive species such as 57.
(17) Peters, K.; Peters, E.-M.; Braun, N. A.; Ciufolini, M. A. Z.
Kristallogr. NCS 1999, 214, 555.
(18) Cf., e.g.: (a) Chow, Y. L.; Colo´n, C. J .; Tan, J . N. S. Can. J .
Chem. 1968, 46, 2821. (b) Lunazzi, L.; Cerioni, G.; Foresti, E.;
Macciantelli, D. J . Chem. Soc., Perkin Trans. 2 1980, 717. See also:
J ohnson, F. Chem. Rev. 1968, 68, 375.
(19) (a) Calculated by molecular mechanics (MM+). This probably
represent a lower limit for this value, since the MM+ force field tends
to underestimate DE’s for conformers of highly strained systems such
as the ones discussed herein. (b) Approximate transition state struc-
tures were created by fixing the distance between the reacting atoms
at 2.5 Å and by allowing the remainder of the molecular framework to
relax in the MM+ force field.
(20) Implicit in this surmise was the assumption that the revers-
ibility of the conjugate addition of the OH group to the dienone would
probably lead to a thermodynamic product.

4402 J . Org. Chem., Vol. 65, No. 14, 2000
Braun et al.
Sch em e 17a
Sch em e 19
Sch em e 20
a
Reagents and conditions: (a) Et3N, 1:1 pyridine-MeCN, rt,
69% for 51, 44% for 53; 36% for 52 over four steps through the
Wipf procedure (see text); (b) TFE, rt, 42% for 54, 22% for 55, 41%
for 56.
Sch em e 18
approximate transition-state structures 62 and 63.18,19
This suggested that, e.g., oxazine 64 may undergo oxida-
tive cyclization more efficiently than oxazoline 51. How-
ever, treatment of 64 as detailed earlier afforded 65 in
only marginally better yield (48% instead of 42%, Scheme
20). One may thus conclude that the moderate efficiency
of these processes is probably not attributable to difficul-
ties hampering nucleophilic capture of the electrophilic
intermediate. In addition, many DIB-promoted transfor-
mations of complex phenolic substrates reportedly pro-
ceed in about 50% yield,8,16 signaling that innate limita-
tions of DIB as an oxidant may be responsible for such
moderate efficiencies.
In all cases described so far, the newly formed spiro-
lactam is a five-membered ring. The reactivity of sub-
strate 66 was examined in order to establish whether six-
membered spirocycles may also be obtained by the new
reaction. Oxidative cyclization to 67 did occur upon
exposure of 66 to DIB, but in a disappointing 17%
chromatographed yield. Evidently, the kinetics of forma-
tion of the incipient six-membered ring is significantly
slower than that of the five-membered ring, leaving the
door open to myriads of side reactions. Finally, several
experiments were conducted in an attempt to induce
bimolecular capture of the electrophilic intermediate
resulting from activation of the phenol with nucleophiles
such as acetamidine, imidazole, sodium azide, lithium
cyanide and phenylboronic acid (Scheme 21). These
reactions were carried out with substrate 68 in the
presence of excess external nucleophile under conditions
otherwise identical to those detailed above. All such
attempts resulted in complex mixtures containing none
of the desired products 69.
In an effort to improve the yields of these reactions,
we briefly examined the effect of alternative solvents, of
other oxidizing agents, and of structural variations in the
substrates on overall efficiency. The substantial cost of
TFE was a major incentive to explore other solvents to
conduct the new transformation, but in complete accord
with earlier observations by Kita, only hexafluoro-2-
propanol (even more costly than TFE) emerged as an
alternative. No reaction whatever occurred in CH2Cl2 or
in polar, aprotic solvents such as acetonitrile, ni-
tromethane, or DMF.
Experiments in which compound 53 was exposed to the
Dess-Martin periodinane in MeCN returned largely
unchanged21 starting oxazoline, even after long contact
time. A recent synthesis of discorhabdin C has shown
that CuCl2 induces oxidative phenolic coupling of sensi-
tive intermediates more efficiently than DIB or PIFA.22
However, reaction of 53 with CuCl2 gave none of the
desired spirocycle. It thus appears that oxidation of 38
to 41 is currently possible only with DIB.
A molecular mechanics study of presumed reaction
intermediates in our oxidative processes indicated that
oxazolinium species 59 (Scheme 19) contains 2.9 kcal/
mol more strain energy than oxazinium ion 60. A similar
energy difference (∆E ) 2.7 kcal/mol) was estimated for
A useful transformation of the dienones obtained by
the new reaction involves hydrogenation to cyclohex-
anone derivatives. Appearances notwithstanding, this
(21) Some degradation of the substrate became evident after 12 h
of exposure to the Dess-Martin oxidant.
(22) Aubart, K. M.; Heathcock, C. H. J . Org. Chem. 1999, 64, 16.

Oxidative Transformations of Phenolic and Indolic Oxazolines
J . Org. Chem., Vol. 65, No. 14, 2000 4403
Sch em e 21
Sch em e 23
Sch em e 24a
Sch em e 22
operation is not trivial, due to the propensity of the
substrates to undergo reductive aromatization under
hydrogenolytic conditions. To illustrate (Scheme 22),
hydrogenation (1 atm) of 48 at room temperature in ethyl
acetate with 10% Pd(C) resulted in a product mixture
composed of 60% of desired 70 and 40% of 71. Reductive
aromatization was the sole observed event when 5%
Rh(C) was employed as the catalyst. By contrast, clean
conversion of 48 to 70 occurred with 1% Pt(C) or PtO2
(Adams catalyst). It seems likely that the sequence of
events leading to aromatization involves either initial
oxidative addition of zerovalent metal to the dienone
C-N s bond or electron transfer from the zerovalent
metal to the dienone, followed by hydrogenolysis of the
intermediate organometallic complex. The ease of oxida-
tive addition or electron transfer would be expected to
be a function of the oxidation potential of the metal.
Indeed, the extent of aromatization seems to correlate
with the reduction potential of Rh (0.600 V), Pd (0.951
V), and Pt (1.118 V).23 We also note that use of Pt(C) as
a catalyst may promote variable degrees of reduction of
the ketone to an alcohol. This process is slower than
dienone hydrogenation and it is generally easy to control
by monitoring of the course of the reduction (TLC) and
avoiding prolonged reaction times. In any event, overre-
duction is all but suppressed by the use of PtO2 as the
catalysts.
a
Reagents and conditions: (a) CH2Cl2, Et3N (95% for 77, 51%
for 78, 88% for 79); (b) THF, 70 °C, sealed tube (64% for 80, 53%
for 81, 65% for 82).
philically activated indole. This observation suggested
that indolic oxazolines may undergo oxidative cyclization
in the same manner as their phenolic counterparts.
The Vorbru¨ggen protocol gave unsatisfactory results
in the indole series, but the Wipf method afforded the
desired oxazolines in fair yield. Accordingly, indolic acids
74-76 were coupled with (S)-phenylalaninol under the
influence of BOP-Cl, and the resulting hydroxyethyl
amides were cyclized to oxazolines with the Burgess
reagent (Scheme 24).
The behavior of indolic substrates toward DIB proved
to be quite sensitive to the nature of substituents present
on the ring. To illustrate, oxidation of 80 with 1 equiv of
DIB under Kita conditions indeed afforded a 1:1 mixture
of diastereomers 84 and 85 of the expected product in
48% cumulative yield after chromatography. Compounds
84-85 probably resulted from initially formed spiro
intermediates 83, which subsequently underwent rapid
intramolecular nucleophilic addition of the alcohol to the
imino function (Scheme 25).25 Heterocycle 84 was chro-
matographically faster moving than 85, so that the two
could be readily separated by preparative TLC (100%
Et2O). In a like manner, reaction of tryptophane-derived
81 furnished a 1:1 mixture of 86 (faster, 5:1 Et2O-
hexane) and 87 (slower) in 40% cumulative chromato-
graphed yield. The stereochemistry of all products rests
on NOE interactions (2D NOESY) observed as shown for
example in Scheme 26 with 84 and 85.
Oxid a tion of In d olic Oxa zolin es
It is well established that DIB and other hypervalent
iodine reagents readily attack the indole nucleus.24
Indeed, we found that exposure of 72 to DIB in TFE
under Kita conditions affords 73 (Scheme 23), albeit in
low yield. Once again, allegedly nonnucleophilic CF3CH2-
OH had acted as a nucleophilic trap toward the electro-
In contrast to the above substrates, reaction of 2-meth-
yl indole derivative 82 with one molar equivalent of DIB
under the same conditions produced quinonimine bis-
(trifluoroethyl)monoketals 89 (faster, 100% Et2O) and 90
(23) Vanysek P. Electrochemical Series. In Handbook of Chemistry
and Physics, 1st student version; Weast, R. C., Ed.; CRC Press: Boca
Raton, FL, 1987; pp D91-D98.
(24) E.g.: Awang, D. V. C, Vincent, A. Can. J . Chem. 1980, 58, 1589.
Moriarty, R. M.; Sultana, M. J . Am. Chem. Soc. 1985, 107, 4559.
(25) Example of an analogous process: Pihko, P. M.; Koskinen, Ari
M. P.; Nissinen, M. J .; Rissanen, K. J . Org. Chem. 1999, 64, 652.

4404 J . Org. Chem., Vol. 65, No. 14, 2000
Braun et al.
Sch em e 25
Sch em e 28
monoketal. Regardless of the precise sequence of events
involved in the genesis of 89-90, it is interesting that
long cherished beliefs regarding the nonnucleophilicity
of TFE were once again clashing against experimental
reality.
Sch em e 26
It is also worthy of note that conversion of 82 to 89-
90 corresponds to a six-electron oxidation, whereas only
1 equiv of DIB (a two-electron oxidant) had been used in
this experiment. However, exposure of 82 to 3 equiv of
DIB in an effort to improve yields had only a marginal
effect on the outcome of the reaction: the yield of
products increased to only about 30%. Similarly, control
experiments in which only 0.8 equiv of DIB were used
relative to 82 again produced 89-90 as the sole identifi-
able products and in about 20-25% yield. These results
suggest that oxidation of the presumed primary products
88 to quinonimine monoketals probably occurs much
faster than oxidative cyclization of the starting oxazoline
82.
Sch em e 27
Additional variability was observed in the reaction of
acid 91, alcohol 92, amine 93, and amide 94 with DIB.
Parallel with the behavior of phenolic analogues of these
compounds led us to infer that heterocycles 95-97 should
emerge from these reactions (Scheme 28). However,
compounds 91-93 produced complex mixtures of oligo-
meric materials upon reaction with DIB, whereas amide
94 underwent clean acetoxylation at C-2 of the indole to
give derivative 26 in 57% chromatographed yield (Scheme
28). No identifiable products arising through addition of
TFE to the substrates were observed in these reactions.
It is apparent that the variable reactivity of indolic
substrates must be attributed to subtle electronic differ-
ences that defy a simplistic explanation.
In summary, a heretofore “impossible” reaction, the
oxidative cyclization of phenolic ω-arylalkanoic carboxa-
mides to spirolactams, may be achieved by the use of DIB
via oxazoline intermediates. An “indolic” variant the new
spirolactam synthesis is also possible, with the proviso
that indolic substrates tend to show variable reactivity
toward DIB. These new transformations embody a fur-
ther aspect of the rapidly growing field of organic
hypervalent iodine chemistry27 and should be quite useful
in the synthesis of a range of nitrogenous substances,
including heterocyclic natural products and intermedi-
ates for medicinal chemistry research.
(slower) in a 1:1.2 ratio and in a modest 23% chromato-
graphed yield (Scheme 27).26 The balance of starting 82
advanced to intractable polymeric materials. We presume
that an initial reaction of 82 with DIB must have given
rise to intermediate 88, a formal aniline derivative that
may have undergone further oxidation to a quinonimine
(26) For related quinonimine syntheses see ref 10 as well as: (a)
Barret, R.; Daudon, M. Tetrahedron Lett. 1991, 32, 2133-2134. (b)
Kita, Y.; Egi, M.; Okajima, A.; Ohtsubo, M.; Takada, T.; Tohma, H.
Chem. Commun. 1996, 1491-1492. (c) Related spirocycles formally
derived from indoles: Rodriguez, J . G.; Urrutia, A.; de Diego, J . E.;
Martinez-Alcazar, M. P.; Fonseca, I. J . Org. Chem. 1998, 63, 4332-
4337.
(27) (a) Nicolaou, K. C.; Zhong, Y.-L.; Baran, P. S. Angew. Chem..
Int. Ed. Engl. 2000, 39, 622, 625. (b) Varvoglis, A. Hypervalent Iodine
in Organic Synthesis, 1st ed.; Academic Press: San Diego, CA, 1996.
(c) Varvoglis, A. Tetrahedron 1997, 53, 1179-1255. (d) Wirth, T.; Hirt,
U. H. Synthesis 1999, 1271. (d) Pelter, A.; Elgendy, S. M. A. J . Chem.
Soc., Perkin Trans. 1 1993, 1891, and references therein

Oxidative Transformations of Phenolic and Indolic Oxazolines
J . Org. Chem., Vol. 65, No. 14, 2000 4405
7.02 (d, 2H, J ) 8.8), 7.04-7.31 (m, 5H), 7.16 (d, 2H, J ) 8.8);
13C 21.6 (q), 28.7 (q), 38.5 (t), 42.1 (t), 50.0 (d), 67.4 (d), 73.0
(t), 80.2 (s), 121.7, 127.0, 128.9, 129.6 and 131.0 (all d), 134.3,
138.0 and 150.0 (all s), 166.7 (s), 169.83 (s).
Exp er im en ta l Section
Exp er im en ta l P r otocols. Unless otherwise noted, NMR
spectra (δ, ppm) were recorded in CDCl3. Coupling constants
J are in Hz. Multiplicities are reported as follows: “s” (singlet),
“d”, “dd” (doublet, doublet of doublets), “t” (triplet), “q”
(quartet), “m” (multiplet), “c” (complex), “br” broad. IR spectra
(cm-1) were obtained from films deposited on NaCl plates. Low-
and high-resolution mass spectra (m/e) were obtained in the
EI (70 eV) mode or, alternatively, in CI (CH4), or FAB (Cs+)
mode if so specified. Optical rotations were measured in CHCl3,
with concentrations, c, expressed in g/100 mL. All reactions
were run under argon and monitored by TLC. The following
compounds were prepared as described in the literature:
N-(tert-butoxycarbonyl)-L-tyrosine;28 N-tosyl-L-tyrosine;29 4-(4-
hydroxy-phenyl)butyric acid;30 3-(4-acetoxyphenyl)propionic
acid and 4-(4-acetoxyphenyl)butyric acid;31 methyl 3-(4-hy-
droxyphenyl)propionate;32 Na-tosyl-L-tryptophan;33 3-(2-meth-
yl-indol-3-yl)propionic acid;34 L-phenylalaninol;35 Burgess
reagent.15b All other reagents and solvents were commercial
products used as received except: THF (freshly distilled Na/
benzophenone), CH2Cl2, Et3N (distilled CaH2). The following
abbreviations are used: BOP-Cl, P,P-bis(2-oxo-3-oxazolidinyl)-
phosphinic chloride; DMAP, 4-(dimethylamino)pyridine; TFE,
2,2,2-trifluoroethanol.
Oxa zolin es by th e Wip f Meth od . A degassed (argon)
solution of a hydroxyethyl amide (2.0 mmol) of a p-ac-
etoxyarylpropionic or p-acetoxyarylbutyric acid and Burgess
reagent (0.6 g, 2.4 mmol) in THF (10 mL) was heated at 70 °C
(oil bath temperature) in a pressure Pyrex tube for 2 h, and
then it was allowed to cool to rt. Addition of Et2O caused
precipitation of triethylammonium salts. The suspension was
filtered over silica gel (Et2O), and the filtrate was evaporated.
If necessary, the residue was purified by column chromatog-
raphy.
(S)-2-[3-(4-Acetoxyph en yl)pr opyl]-4-ben zyl-2-oxazolin e:
68% yield from N-[(S)-1-benzyl-2-hydroxyethyl]-4-(4-acetoxy-
phenyl)butyramide; pale yellow oil; [R]20D ) -11.5° (c ) 1.02);
1H 1.91-2.00 (m, 2H), 2.26-2.31 (m, 2H), 2.28 (s, 3H), 2.60-
2.71 (m, 3H), 3.08 (dd, 1H, J ) 5.1, 13.8), 3.92 (dd, 1H, J )
7.3, 8.4), 4.13 (dd, 1H, J 1 ) J 2 ) 8.4), 4.36 (m, 1H), 6.99 (d,
2H, J ) 8.6), 7.18 (d, 2H, J ) 8.6), 7.18-7.32 (m, 5H); 13C
20.9, 27.2, 27.3, 34.3, 41.5, 66.9, 71.2, 121.2, 126.2, 128.3, 129.1,
129.2, 137.7, 138.9, 148.7, 167.5, 169.4; IR 1765, 1670; MS 337
[M]+; HRMS calcd for C21H23NO3 337.1678, found 337.1684.
Dea cetyla tion of th e P r otected Oxa zolin es. A mixture
of acetylated oxazoline (2.0 mmol), K2CO3 (42 mg, 0.3 mmol),
and MeOH (3 mL) was stirred at rt for 1 h and then it was
filtered over silica gel (1 g; MeOH) and evaporated. If neces-
sary, the residue was purified by column chromatography or
recrystallization.
(S )-4-Be n zyl-2-[2-(4-h yd r oxyp h e n yl)e t h yl]-2-oxa zo-
lin e (38): 73% yield from (S)-2-[2-(4-acetoxyphenyl)ethyl]-4-
benzyl-2-oxazoline; colorless crystals; mp 141-143 °C (recrd
1
MeOH/Et2O); [R]20 ) -26.8° (c ) 1.16, MeOH); H (DMSO-
D
d6) 2.36-2.41 (m, 2H), 2.59 (dd, 1H, J ) 7.1, 13.6), 2.67-2.72
(m, 2H), 2.80 (dd, 1H, J ) 6.1, 13.6), 3.83 (dd, 1H, J ) 7.3,
8.1), 4.14 (dd, 1H, J ) 8.1, 9.4), 4.25 (m, 1H), 6.64 (d, 2H, J )
8.5), 6.98 (d, 2H, J ) 8.5), 7.18-7.29 (m, 5H); 13C (DMSO-d6)
29.7, 30.8, 41.4, 66.8, 71.1, 115.3, 126.3, 128.3, 129.3, 129.4,
130.6, 138.5, 156.1, 165.9; IR (KBr) 3420, 1665; MS 281 [M]+;
HRMS calcd for C18H19NO2 281.1416, found 281.1416.
(S)-4-Ben zyl-2-[(S)-1-[N-(ter t. bu toxyca r bon yl)a m in o]-
2-(4-h yd r oxyp h en yl)et h yl]-2-oxa zolin e (52): 81% yield
from (S)-2-[(S)-2-(4-acetoxyphenyl)-1-[N-(tert. butoxycarbonyl)-
amino]ethyl]-4-benzyl-2-oxazoline; colorless, viscous oil. This
material was used without purification because of facile
epimerization: 1H 1.43 (s), 2.60 (dd, 1H, J ) 8.4, 13.6), 2.97-
3.07 (m, 3H), 4.05 (dd, 1H, J ) 7.0, 8.0), 4.23 (dd, 1H, J 1 ) J 2
) 8.0), 4.33 (m, 1H), 4.61 (m, 1H), 5.14 (d, 1H, J ) 8.0, NH),
6.64 (d, 2H, J ) 8.3), 6.95 (d, 2H, J ) 8.3), 7.14-7.32 (m, 5H);
IR 3300, 1720, 1670; MS 396 [M]+; HRMS calcd for C23H28N2O4
396.2049, found 396.2044.
(S )-2-[2-(4-Ace t oxyp h e n yl)e t h yl]-4-b e n zyl-2-oxa zo-
lin e (40): 74% yield from N-[(S)-1-benzyl-2-hydroxyethyl]-3-
(4-acetoxyphenyl)propionamide; colorless oil; [R]20D ) -9.9° (c
) 2.18); 1H 2.28 (s, 3H), 2.53-2.66 (m, 3H), 2.91-2.99 (m, 2H),
3.07 (dd, 1H, J ) 5.2, 13.6), 3.94 (dd, 1H, J ) 7.2, 8.4), 4.16
(dd, 1H, J 1 ) J 2 ) 8.4), 4.35 (m, 1H), 7.01 (d, 2H, J ) 8.5),
7.17-7.32 (m, 5H), 7.22 (d, 2H, J ) 8.5); 13C 20.2, 28.9, 30.7,
41.0, 66.4, 70.8, 120.8, 125.7, 127.7, 128.5, 128.6, 137.2, 137.5,
148.4, 166.2, 168.6; IR 1765, 1665; MS 323 [M]+; HRMS calcd
for C20H21NO3 323.1521, found 323.1520.
(S)-4-Ben zyl-2-[3-(4-h yd r oxyp h en yl)p r op yl]-2-oxa zo-
lin e (66): 81% yield from (S)-2-[3-(4-acetoxyphenyl)propyl]-4-
benzyl-2-oxazoline; yellow, viscous oil; [R]20D ) -4.9° (c ) 2.08);
1H 1.88-1.98 (m, 2H), 2.26-2.32 (m, 2H), 2.57 (t, 2H, J ) 7.4),
2.66 (dd, 1H, J ) 8.4, 13.7), 3.13 (dd, 1H, J ) 4.8, 13.7), 3.98
(dd, 1H, J ) 7.3, 8.7), 4.17 (dd, 1H, J 1 ) J 2 ) 8.7), 4.41 (m,
1H), 6.71 (d, 2H, J ) 8.3), 6.97 (d, 2H, J ) 8.3), 7.17-7.31 (m,
5H); 13C 27.2, 27.5, 34.2, 41.4, 66.3, 71.6, 115.3, 126.5, 128.5,
129.2, 129.4, 132.2, 137.3, 154.8, 169.4; IR 3320, 1655; MS 295
[M]+; HRMS calcd for C19H21NO2 295.1572, found 295.1572.
Oxa zolin es by th e Vor br u1 ggen P r oced u r e. A solution
of Ph3P (2.4 g, 9.0 mmol) in 1:1 MeCN/pyridine (5 mL; warming
is necessary to dissolve the PPh3) was added dropwise over 2
h to a solution of phenolic acid (3.0 mmol), amino alcohol (3.0
mmol), Et3N (0.9 g, 9.0 mmol), and CCl4 (1.8 g, 12.0 mmol) in
1:1 MeCN/pyridine (5 mL). The mixture was stirred at rt for
14 h, then it was evaporated and the residue was processed
as follows. Wor k u p P r oced u r e for 4-Un su bstitu ted Ox-
a zolin es a n d N-Tosyl Com p ou n d s. The residue was parti-
tioned between CH2Cl2 and 0.5 M aqueous NaOH. The organic
phase was discarded, and the water layer was extracted with
more CH2Cl2. The extracts were again discarded. The aqueous
phase was layered with EtOAc and acidified (cooling, stirring)
to pH 6 with solid NH4Cl (6 g). Addition of some 0.2 M aqueous
AcOH may be necessary to reach pH 6 (for C-2′ unsubstituted
compounds, it is possible to acidify with 3 M HCl instead of
NH4Cl/AcOH). The layers were separated, and the aqueous
was extracted with EtOAc. The combined extracts were dried
(MgSO4) and concentrated. The product was purified by
filtration over silica gel. Wor k u p P r oced u r e for Ot h er
Oxa zolin es. The residue was dissolved in Et2O and basified
to pH 8-9 with concentrated aqueous NH4OH solution. The
organic layer was separated, and the water phase was
(S)-2-[(S)-2-(4-Acetoxyp h en yl)-1-[N-(ter t-bu toxyca r bo-
n yl)a m in o]eth yl]-4-ben zyl-2-oxa zolin e: 67% yield from (S)-
N-[(S)-1-benzyl-2-hydroxyethyl]-3-(4-acetoxyphenyl)-2-[N-(tert-
butoxycarbonyl)amino]propionamide; colorless oil; [R]20
)
D
-28.0° (c ) 0.25); 1H 1.43 (s, 9H), 2.27 (s, 3H), 2.59 (dd, 1H, J
) 8.1, 13.7), 2.89-3.11 (m, 3H), 4.01 (dd, 1H, J ) 6.9, 8.2),
4.19 (dd, 1H, J 1 ) J 2 ) 8.2), 4.29 (m, 1H), 4.63 (m, 1H), 5.10
(d, 1H, J ) 7.9, NH), 6.99 (d, 2H, J ) 8.5), 7.14 (d, 2H, J )
8.5), 7.14-7.32 (m, 5H); 13C 21.5, 28.7, 38.5, 41.9, 50.1, 67.4,
72.8, 80.2, 121.8, 127.0, 128.9, 129.6, 130.9, 134.3, 138.0, 150.0,
166.7, 169.9; IR 3420, 1770, 1715, 1670; MS 439 [M + H]+,
438 [M]+; HRMS calcd for C25H30N2O5 438.2155, found 438.2154.
(S)-2-[(R)-2-(4-Acetoxyp h en yl)-1-[N-(ter t-bu toxyca r bo-
n yl)am in o]eth yl]-4-ben zyl-2-oxazolin e. This material, [R]20
D
) -21.1° (c ) 0.29), was separated from the major diastere-
omer above by preparative TLC (1:1 Et2O/hexane): 1H 1.42
(s, 9H), 2.28 (s, 3H), 2.46 (dd, 1H, J ) 8.8, 13.2), 2.90-3.15
(m, 3H), 3.97 (dd, 1H, J ) 7.4, 8.8), 4.23 (dd, 1H, J 1 ) J 2
8.8), 4.32 (m, 1H), 4.63 (m, 1H), 5.21 (d, 1H, J ) 7.4, NH),
)
(28) Keller, O.; Keller, W. E.; van Look, G.; Wersin, G. Org. Synth.
1984, 63, 160.
(29) Fischer, E.; Lipshitz W. Ber. Dtsch. Chem. Ges. 1915, 48, 360.
(30) Yi, C. S.; Martinelli, L. C.; DeWitt Blanton, C., J r. J . Org. Chem.
1978, 43, 405.
(31) Winter, M. Helv. Chim. Acta 1961, 44, 2110.
(32) Herbert, R. B.; Kattah, A. E. Tetrahedron 1990, 46, 7105.
(33) Corey, E. J .; Loh, T.-P. J . Am. Chem. Soc. 1991, 113, 8966.
(34) Harley-Mason, J . J . Chem Soc. 1952, 2433.
(35) Gage, J . R.; Evans, D. A. Organic Syntheses; Wiley: New York,
1993; Collect. Vol. VIII, p 528.

4406 J . Org. Chem., Vol. 65, No. 14, 2000
Braun et al.
extracted with Et2O. The combined extracts were dried
(MgSO4) and evaporated. It may be necessary to redissolve
the crude product in Et2O and filter off the precipitated Ph3Pd
O prior to further purification.
2-[2-(4-Hydr oxyph en yl)eth yl]-2-oxazolin e (51): 68% yield
from 3-(4-hydroxyphenyl)propionic acid and ethanolamine
after purification by preparative TLC (100% EtOAc); Rf ) 0.26;
colorless crystals; mp 123-125 °C dec; 1H (DMSO-d6) 1.68 (s,
OH), 2.39-2.44 (m, 2H), 2.70-2.75 (m, 2H), 3.64-3.70 (m, 2H),
4.12-4.18 (m, 2H), 6.66 (d, 2H, J ) 8.5), 6.99 (d, 2H, J ) 8.5);
13C (DMSO-d6) 29.7, 30.8, 54.0, 66.8, 115.3, 129.2, 130.8, 156.0,
166.4; IR (KBr) 3420, 1660; MS 191 [M]+; HRMS calcd for
C11H13NO2 191.0946, found 191.0946.
(S)-4-Ben zyl-2-[(S)-1-[N-(4-m eth ylp h en yl)su lfon yla m i-
n o]-2-(4-h yd r oxyp h en yl)eth yl]-2-oxa zolin e (53): 44% yield
from N-tosyl-L-tyrosine and (S)-2-amino-3-phenyl-1-propanol
after filtration through a short plug of silica (100% Et2O), not
extensively purified because of facile epimerization; pale yellow
foam; [R]20D ) -30.7° (c ) 0.87); 1H 2.16 (dd, 1H, J ) 8.5, 13.6),
2.39 (s, 3H), 2.72 (dd, 1H, J ) 5.5, 13.6), 2.93 (d, 2H, J ) 5.5),
3.86 (dd, 1H, J ) 6.7, 7.8), 4.04 (dd, 1H, J 1 ) J 2 ) 7.8), 4.07
(m, 1H), 4.33 (m, 1H), 5.78 (d, 1H, J ) 9.2, NH), 6.52 (d, 1H,
J ) 8.5), 6.87 (d, 1H, J ) 8.5), 7.03 (d, 2H, J ) 8.5 Hz), 7.19-
7.27 (m, 5H) 7.72 (d, 2H, J ) 8.5); 13C 21.5, 38.9, 41.1, 52.3,
66.2, 72.6, 115.4, 126.7, 127.3, 128.6, 128.9, 129.5, 130.5, 137.0,
137.2, 143.4, 155.4, 166.9; IR 3280, 1665; MS (CI) 451 [M +
H]+; HRMS calcd for C25H26N2O4S 450.1613, found 450.1618
Oxid a tive Cycliza tion of P h en olic Oxa zolin es. A solu-
tion of DIB (0.4 g, 1.2 mmol) in TFE (5 mL) was added
dropwise over 5 min to a solution of the oxazoline (1.0 mmol)
in TFE (5 mL). The mixture was stirred for 30 min at room
temperature (argon), and then it was treated with solid
NaHCO3 (0.3 g). The resulting suspension was filtered over
glass wool and concentrated. The crude product was im-
mediately taken up in anhydrous pyridine (0.8 g, 10.0 mmol)
and treated with Ac2O (1.0 g, 10.0 mmol) and DMAP (6.1 mg,
50 mmol)] at rt for 12 h with good stirring. Finally, the mixture
was evaporated and the residue was purified by chromatog-
raphy and/or recrystallization.
13.2), 3.07 (dd, 1H, J ) 6.1, 12.4), 3.15 (m, 1H), 3.25 (dd, 1H,
J ) 7.0, 12.4), 4.29 (dd, 1H, J ) 4.6, 11.2), 4.57 (dd, 1H, J )
8.3, 11.2), 5.23 (m, 1H), 6.04 (dd, 1H, J ) 1.9, 9.9), 6.12 (dd,
1H, J ) 2.8, 9.9), 6.24 (dd, 1H, J ) 1.9, 10.2), 6.83 (dd, 1H, J
) 2.8, 10.2), 7.18-7.32 (m, 5H), 7.42 (d, 2H, J ) 7.9), 8.04 (d,
2H, J ) 7.9); 13C: 20.9, 21.7, 25.1, 35.6, 57.3, 59.9, 62.5, 126.9,
127.6. 128.5, 128.9, 129.8, 130.1, 130.2, 136.1, 137.5, 145.5,
148.5, 148.8, 169.6, 169.8, 170.2, 184.1; IR 1745, 1705, 1675;
MS 550 [M]+; HRMS calcd for C27H28N2O6S, 550.1774, found
550.1769.
1-(2′-Acet oxyet h yl)-1-a za sp ir o[4.5]d eca -6,9-d ien e-2,8-
d ion e (54): 42% yield from 2-[2-(4-hydroxyphenyl)ethyl]-2-
oxazoline; chromatography 100% EtOAc; Rf ) 0.27; pale yellow
oil; 1H 1.96 (s, 3H), 2.14 (t, 2H, J ) 8.1), 2.52 (t, 2H, J ) 8.1),
3.23-3.27 (m, 2H), 4.03-4.07 (m, 2H), 6.27 (d, 2H, J ) 9.9),
6.75 (d, 2H, J ) 9.9); 13C 20.7, 28.8, 30.0, 40.1, 61.6, 62.0, 129.9,
148.9, 170.4, 174.9, 183.9; IR 1745, 1700, 1675; MS 249 [M]+;
HRMS calcd for C13H15NO4 249.1001, found 249.1000.
(2S,5S,10R)-2-Ben zyl-1-a za -4-oxa t r icyclo[8.3.0.05,10]-
tr id ec-8-en e-7,13-d ion e (42). A solution of 1-[(S)-2′-acetoxy-
1′-benzyl-ethyl]-1-azaspiro[4.5]deca-6,9-diene-2,8-dione (170.0
mg, 0.5 mmol) in MeOH (3 mL) containing K2CO3 (7 mg, 50
mmol) was stirred for 2 h at rt, and then it was filtered over
silica gel (1 g; MeOH) and concentrated to yield 0.14 g (94%)
1
of 42: colorless oil; [R]20 ) -104.6° (c ) 0.70); H 2.11-2.19
D
(m, 2H), 2.44-2.71 (m, 2H), 2.75-2.78 (m, 2H), 2.82 (ddd, 1H,
J ) 1.1, 5.2, 13.1), 2.92 (dd, 1H, J ) 10.3, 13.1), 3.40 (ddd,
1H, J ) 1.1, 2.6, 11.8), 3.75 (m, 1H), 3.81 (d, 1H, J ) 11.8),
4.27 (ddd, 1H, J ) 2.6, 5.2, 10.3), 6.11 (d, 1H, J ) 10.3), 6.72
(dd, 1H, J ) 2.9, 10.3), 7.20-7.33 (m, 5H); 13C 29.3, 29.9, 37.5,
40.7, 49.8, 59.3, 67.1, 81.1, 126.7, 128.1, 128.6, 129.4, 137.4,
150.3, 173.2, 194.3; IR 1665; MS 297 [M]+; HRMS calcd for
C
18H19NO3 297.1365, found 297.1363.
Rep r esen ta tive P r oced u r e for Hyd r ogen a tion of Di-
en on es: 1-[(S)-2′-Acetoxy-1′-ben zyleth yl]-1-a za sp ir o[4.5]-
d eca -2,8-d ion e (70). A solution of 48 (340 mg, 1 mmol) in
EtOAc (10 mL) containing suspended 1% Pt on 60-100 mesh
carbon (80 mg) was stirred at rt under H2 (balloon). The
reaction was followed by TLC to avoid reduction of the ketone,
and it was complete after 20 min. The catalyst was filtered
off and the solvent evaporated to yield 310 mg (90%) of 70:
1-[(S)-2′-Acet oxy-1′-b en zylet h yl]-1-a za sp ir o[4.5]d eca -
6,9-d ien e-2,8-d ion e (48): 47% yield from (S)-4-benzyl-2-[2-
(4-hydroxyphenyl)ethyl]-2-oxazoline; chromatography 1:1 EtOAc/
colorless crystals; mp 119-121 °C; [R]20 ) -21.7° (c ) 0.23);
D
Et2O, Rf ) 0.46; colorless needles; mp 163-65 °C (recrd EtOAc/
1H 1.18-1.37 (m, 2H), 1.77-1.84 (m, 1H), 1.93-2.25 (m, 5H),
2.03 (s, 3H), 2.37-2.46 (m, 4H), 3.01 (dd, 1H, J ) 5.0, 12.9),
3.32 (m, 1H), 3.49 (dd, 1H, J ) 9.7, 12.9), 4.42 (dd, 1H, J )
1
Et2O); [R]20 ) +31.4° (c ) 0.44); H 1.85-2.07 (m, 2H), 2.05
D
(s, 3H), 2.48-2.56 (m, 2H), 2.87 (dd, 1H, J ) 4.9, 13.2), 3.12
(m, 1H), 3.47 (dd, 1H, J ) 10.6, 13.2), 4.38 (dd, 1H, J ) 5.4,
11.3), 4.53 (dd, 1H, J ) 8.1, 11.3), 5.14 (dd, 1H, J ) 3.1, 10.1),
5.85 (dd, 1H, J ) 2.1, 10.1), 6.21 (dd, 1H, J ) 2.1, 10.1), 6.78
(dd, 1H, J ) 3.1, 10.1), 7.12-7.32 (m, 5H); 13C 21.0, 30.0, 30.5,
34.5, 57.0, 62.8, 63.5, 127.1, 128.6, 129.2, 129.6, 129.9, 137.9,
148.6, 149.2, 170.2, 175.0, 184.2; IR 1735, 1690, 1670; MS 339
[M]+; HRMS calcd for C20H21NO4 339.1471, found 339.1471.
(3S,1S′)-1-(2′-Acetoxy-1′-ben zyleth yl)-3-[N-(ter t-bu toxy-
ca r b on yl)a m in o]-1-a za sp ir o[4.5]d eca -6,9-d ien e-2,8-d i-
on e (55): 22% yield from (S)-4-benzyl-2-[(S)-1-[N-(tert-butoxy-
carbonyl)amino]-2-(4-hydroxyphenyl)ethyl]-2-oxazoline; chro-
matography 1:1 EtOAc/hexane, Rf ) 0.40; colorless crystals;
6.6, 11.2), 4.56 (dd, 1H, J ) 6.8, 11.2), 7.14-7.29 (m, 5H); 13
C
20.9, 28.4, 33.3, 34.5, 34.9, 37.6, 37.7, 54.7, 62.7, 64.1, 126.8,
128.5, 129.6, 138.5, 170.6, 175.1, 208.5; IR 1740, 1715, 1680;
MS 343 [M]+; HRMS calcd for C20H25NO4 343.1784, found
343.1784.
Am id e F or m a tion in th e In d ole Ser ies. A cold (0 °C)
solution of an indolic acid (3.0 mmol), BOP-Cl (0.8 g, 3.0 mmol),
and Et3N (0.3 g, 3.0 mmol) in CH2Cl2 (6 mL) was stirred for
30 min prior to addition of (S)-2-amino-3-phenyl-1-propanol
(0.45 g, 3 mmol). A solution of Et3N (0.3 g, 3.0 mmol) in CH2-
Cl2 (1.2 mL) was then introduced dropwise during 2 h, with
continued stirring at 0 °C. The mixture was stirred an
additional 2 h at 0 °C and 8 h at rt, and then it was quenched
with H2O and treated as follows. Reactions involving tryp-
tophane derivatives were acidified to pH 6 with saturated NH4-
Cl solution. Reactions involving other indolic acids were
acidified to pH 1 with 4 M aqueous HCl. In all cases, the
organic layer was separated and the water phase was ex-
tracted with more EtOAc. The combined extracts were washed
(saturated aqueous NaHCO3), dried (MgSO4), and evaporated.
mp 97-99 °C (Et2O/hexane); [R]20 ) -20.8° (c ) 1.48); 1H
D
1.46 (s, 9H), 1.79 (dd, 1H, J ) 11.2, 12.7), 2.01 (s, 3H), 2.50
(dd, 1H, J ) 8.3, 12.7), (dd, 1H, J ) 4.8, 13.2), 3.12 (m, 1H),
3.41 (dd, 1H, J ) 10.9, 13.2), 4.35 (dd, 1H, J ) 5.2, 11.4), 4.36
(m, 1H), 4.52 (dd, 1H, J ) 8.8, 11.4), 4.85 (br d, 1H, J ) 10.2),
5.23 (d, 1H, J ) 4.8, NH), 5.95 (dd, 1H, J ) 1.8, 10.2), 6.19
(dd, 1H, J ) 1.8, 10.2), 6.79 (dd, 1H, J ) 3.1, 10.2), 7.12-7.36
(m, 5H); 13C 20.9, 28.2, 34.8, 57.4, 60.3, 62.6, 80.6, 127.3, 128.6,
128.9, 129.8, 130.9, 137.5, 147.7, 148.3, 155.7, 170.2, 172.5,
184.0; IR 3335, 1740 1710, 1695, 1675; MS (CI) 455 [M + H]+;
HRMS calcd for C25H30N2O6 454.2104, found 454.2092.
(3S,1S′)-1-(2′-Acetoxy-1′-ben zyleth yl)-3-[N-a cetyl-N-(4-
m et h ylp h en yl)su lfon yla m in o]-1-a za sp ir o[4.5]d eca -6,9-
d ien e-2,8-d ion e (56): 41% yield from (S)-4-benzyl-2-[(S)-1-
[N-(4-methylphenyl)sulfonylamino]-2-(4-hydroxyphenyl)ethyl]-
N-[(S)-1-Ben zyl-2-h yd r oxyet h yl]-3-(in d ol-3′-yl)p r op i-
on a m id e (77): 95% yield from 3-(indol-3′-yl)propionic acid;
1
colorless foam; [R]20 ) -16.1° (c ) 4.98); H 2.46 (t, 2H, J )
D
7.4), 2.66 (dd, 1H, J ) 7.4, 13.5), 2.72 (dd, 1H, J ) 6.6, 13.5),
3.01 (t, 2H, J ) 7.4), 3.39 (dd, 1H, J ) 5.1, 11.4), 3.47 (dd, 1H,
J ) 3.7, 11.4), 4.09 (m, 1H), 6.03 (d, 1H, J ) 8.1), 6.80 (d, 1H,
J ) 2.2), 7.04-7.21 (m, 7H), 7.30 (d, 1H, J ) 8.1), 7.54 (d, 1H,
J ) 7.4), 8.54 (s, 1H, NH ind.); 13C 21.2, 36.6, 37.1, 52.5, 63.4,
111.3, 114.1, 118.4, 119., 121.7, 121.8, 126.4, 126.9, 128.4,
2-oxazoline; chromatography 1:1 EtOAc/hexane; pale yellow
1
foam; [R]20 ) -22.4° (c ) 1.25); H 2.03 (s, 3H), 2.29 (s, 3H),
D
2.45 (dd, 1H, J ) 4.1, 13.2), 2.47 (s, 3H), 2.53 (dd, 1H, J ) 9.9,

Oxidative Transformations of Phenolic and Indolic Oxazolines
J . Org. Chem., Vol. 65, No. 14, 2000 4407
129.1, 136.21, 137.5, 173.6; IR 3325, 1625; MS (CI) 323 (100)
[M + H]+; HRMS calcd for C20H22N2O2 322.1681, found
322.1665.
over glass wool and evaporated. The crude product was
purified by column chromatography as detailed below.
7,8-Ben zo-2-b en zyl-1,6-d ia za -4-oxa t r icyclo[7.3.0.05,9]-
d od ec-7-en -12-on e. Obtained in 48% overall yield from (S)-
4-benzyl-2-[2-(indol-3′-yl)ethyl]-2-oxazoline as a 1:1 mixture
(1H NMR) of (2S,5S,9S) and (2S,5R,9R) diastereomers (sepa-
rated by chromatography, 100% Et2O). (2S,5S,9S)-Isom er
(S)-N-[(S)-1-Ben zyl-2-h yd r oxyet h yl]-3-(in d ol-3′-yl)-2-
[N-(4-m eth ylp h en yl)su lfon yla m in o]p r op ion a m id e (78):
51% yield from N-tosyl-L-tryptophan (chromatography 3:1
Et2O/EtOAc, Rf ) 0.27; recrd EtOAc/Et2O); colorless crystals;
mp 183-185 °C; [R]20 ) -17.1° (c ) 0.86, dioxane); 1H
(84): Rf ) 0.56; 26% yield; colorless oil; [R]20 ) -53.3° (c )
D
D
(DMSO-d6) 2.26 (s, 3H), 2.45 (dd, 1H, J ) 7.4, 14.7), 2.66-
2.75 (m, 2H), 2.89 (dd, 1H, J ) 5.5, 13.6), 3.08 (dd, 1H, J )
5.5, 10.3), 3.17 (dd, 1H, J ) 4.8, 10.3), 3.69 (m, 1H), 3.89 (m,
1H), 6.90 (m, 1H), 6.99-7.05 (m, 2H), 7.08 (d, 2H, J ) 7.4),
7.16-7.20 (m, 3H), 7.24-7.28 (m, 3H), 7.33 (d, 1H, J ) 8.1
NH), 7.41 (d, 2H, J ) 7.4), 7.74 (d, 1H, J ) 8.1, NH), 7.82 (d,
1H, J ) 8.1), 10.71 (s, 1H, NH ind.); 13C (DMSO-d6) 21.1, 29.0,
36.4, 52.5, 57.3, 61.8, 109.4, 111.3, 118.2, 118.3, 120.8, 124.0,
126.1, 126.4, 127.3, 128.2, 129.1, 129.3, 136.1, 138.1, 139.1.
142.1, 170.4; IR 3330, 1650; MS 491 [M]+; HRMS calcd for
1.61); 1H 2.16 (m, 1H), 2.30 (ddd, 1H, J 1 ) 9.6, J 2 ) J 3 ) 11.8),
2.53 (m, 1H), 2.72-2.85 (m, 2H), 3.36-3.46 (m, 2H), 3.68 (m,
1H), 3.96 (dd, 1H, J ) 3.7, 13.2), 4.53 (s, 1H, NH), 5.08 (s,
1H), 6.63 (d, 1H, J ) 8.1), 6.74 (dd, 1H, J 1 ) J 2 ) 7.4 Hz),
7.06 (d, 1H, J ) 7.4), 7.11 (dd, 1H, J ) 7.4, 8.1), 7.15-7.28
(m, 5H); 13C 29.7, 34.2, 35.0, 53.3, 60.9, 67.2, 93.3, 109.1, 119.4,
121.7, 126.3, 128.4, 129.6, 129.6, 130.2, 138.0, 147.7, 175.3;
IR 3320, 1680; MS (CI) 321 [M + H]+; HRMS calcd for
C
20H22N2O2 320.1525, found 320.1516.
(2S,5R,9R)-Isom er (85): Rf ) 0.30; 22% yield; colorless
C
27H29N3O4S 491.1879, found 491.1865.
crystals; mp 209-211 °C (recrd Et2O/hexane); [R]20D ) +16.0°
(c ) 0.78); 1H 2.05 (ddd, 1H, J ) 3.0, 8.9, 12.5), 2.27 (ddd, 1H,
J ) 9.6, 10.3, 12.5), 2.41-2.51 (m, 2H), 2.68-2.81 (m, 2H),
3.40 (dd, 1H, J ) 4.0, 12.1), 3.56 (dd, 1H, J ) 3.3, 12.1), 4.28
(m, 1H), 4.56 (s, 1H, NH), 4.71 (s, 1H), 6.75 (d, 1H, J ) 8.1),
6.83 (dd, 1H, J 1 ) J 2 ) 7.4), 7.11-7.26 (m, 7H); 13C 29.2, 33.4,
36.9, 50.3, 62.0, 66.7, 93.6, 110.2, 119.9, 122.8, 126.4, 128.4,
129.4, 130.9, 137.7, 147.4, 174.7; IR (CDCl3) 3300, 1680; MS
(CI) 321 [M + H]+; HRMS calcd for C20H22N2O2 320.1525,
found 320.1528.
N-[(S)-1-Ben zyl-2-h yd r oxyeth yl]-3-(2′-m eth yl-in d ol-3′-
yl)p r op ion a m id e (79). 88% yield from 3-(2-methyl-indol-3′-
yl)propionic acid; pale yellow oil; [R]20 ) -16.9° (c ) 0.94);
D
1H 2.33 (s, 3H), 2.44 (t, 2H, J ) 7.0), 2.56-2.68 (m, 2H), 3.01-
3.04 (m, 2H), 3.34 (dd, 1H, J ) 5.5, 11.4), 3.43 (dd, 1H, J )
3.3, 11.4), 3.99 (m, 1H), 5.48 (d, 1H, J ) 7.4, NH), 6.98-7.27
(m, 8H), 7.47 (d, 1H, J ) 6.6), 7.96 (s, 1H, NH ind.); 13C 11.5,
20.4, 36.7, 37.5, 52.8, 63.9, 110.0, 110.4, 117.7, 119.2, 121.1,
126.5, 128.2, 128.5. 129.1, 131.6, 135.3 137.4, 173.6; IR 3290,
1650; MS (CI) 337 [M + H]+, 336 [M]+; HRMS C21H24N2O2
calcd 336.1838, found 336.1840.
7,8-Ben zo-2-ben zyl-11-[N-(4-m eth ylph en yl)su lfon ylam i-
n o]-1,6-d ia za -4-oxa t r icyclo[7.3.0.05,9]d od ec-7-en -12-on e:
41% yield from (S)-4-benzyl-2-[(S)-2-(indol-3′-yl)-1-[N-(4-me-
thylphenyl)sulfonylamino]ethyl]-2-oxazoline as a 1:1 ratio
(NMR) of (2S,5S,9S,11S) and (2S,5R,9R,11S) diastereomers
(separated by chromatography, 5:1 Et2O/hexane).
P r ep a r a tion of In d olic Oxa zolin es. A degassed (argon)
solution of an indolic amide (1.5 mmol) and Burgess reagent
(0.4 g, 1.7 mmol) in THF (10 mL) was heated at 70 °C in a
pressure Pyrex tube for 2 h. The mixture was cooled to rt,
diluted with Et2O (precipitate of ammonium salts), and filtered
through a short plug of silica gel (Et2O). Concentration afforded
material of sufficiently good quality that no further purification
was required before oxidation.
(2S,5S,9S,11S)-Isom er (86): Rf ) 0.37; 19% yield; colorless
1
oil; [R]20 ) -3.6° (c ) 1.49); H 2.25 (dd, 1H, J ) 7.4, 14.0),
D
2.43 (s, 3H), 2.70 (dd, 1H, J ) 9.0, 14.0), 2.81 (dd, 1H, J ) 9.2,
13.8), 3.40 (dd, 1H, J ) 6.3, 11.3), 3.54 (d, 1H, J ) 11.3), 3.65
(dd, 1H, J ) 7.2, 9.2), 3.82-3.91 (m, 2H), 4.42 (s, 1H, NH),
4.95 (s, 1H), 5.23 (d, 1H, J ) 1.8, NHTs), 6.64 (d, 1H, J ) 8.1),
6.83 (ddd, 1H, J 1 ) 0.8, J 2)J 3 ) 7.4), 7.06 (d, 1H, J ) 7.4),
7.11-7.34 (m, 8H), 7.77 (d, 2H, J ) 8.5); 13C 21.6, 34.3, 39.0,
53.5, 54.9, 62.7, 68.5, 90.7, 109.8, 120.4, 123.1, 126.5, 127.4,
128.4, 128.5, 129.0, 129.1, 129.9, 134.8, 135.7, 144.0, 148.3,
(S)-4-Ben zyl-2-[2-(in dol-3′-yl)eth yl]-2-oxazolin e (80): 64%
yield from N-[(S)-1-benzyl-2-hydroxyethyl]-3-(indol-3′-yl)pro-
pionamide; colorless oil; [R]20 ) -11.1° (c ) 1.16); 1H 2.60
D
(dd, 1H, J ) 8.1, 13.2), 2.67-2.72 (m, 2H), 3.07 (dd, 1H, J )
5.2, 13.2), 3.10-3.15 (m, 2H), 3.97 (dd, 1H, J ) 7.1, 8.3), 4.17
(dd, 1H, J 1 ) J 2 ) 8.3), 4.38 (m, 1H), 6.99 (d, 1H, J ) 2.2),
7.11-7.37 (m, 8H), 7.63 (d, 1H, J ) 7.4), 8.24 (s, 1H, NH); 13
C
171.9; IR 3385, 1695; MS 489 [M]+; HRMS calcd for C27H27
N3O4S 489.1722, found 489.1720.
-
21.7, 28.9, 41.7, 67.1, 71.5, 111.1, 115.0, 118.7, 119.2, 121.4
121.9, 126.4, 127.2, 128.4, 129.2, 136.2, 137.9, 167.8; IR 3415,
1660; MS (CI) 305 [M + H]+; HRMS calcd for C20H20N2O
304.1576, found 304.1564.
(2S,5R,9R,11S)-Isom er (87): Rf ) 0.18; 22% yield; color-
less foam; [R]20 ) +4.2° (c ) 2.11); 1H 2.26 (dd, 1H, J ) 10.8,
D
12.5), 2.43 (s, 3H), 2.47 (dd, 1H, J ) 11.0, 13.4), 2.56 (dd, 1H,
J ) 7.7, 12.5), 2.68 (dd, 1H, J ) 4.6, 13.4), 3.43 (dd, 1H, J )
3.7, 12.4), 3.56 (dd, 1H, J ) 2.8, 12.4), 4.11 (m, 1H), 4.22 (m,
1H), 4.64 (s, 1H, NH), 4.68 (s, 1H), 5.60 (s, 1H, NHTs), 6.74-
6.82 (m, 2H), 6.98 (d, 1H, J ) 7.0), 7.05-7.32 (m, 8H), 7.76 (d,
2H, J ) 8.1); 13C 21.5, 36.6, 43.1, 50.2, 52.2, 62.0, 68.5, 93.6,
110.5, 120.0, 122.5, 126.6, 127.2, 128.5, 129.1, 129.3, 129.7,
129.9, 136.1, 137.0, 143.9, 147.6, 170.5; IR 3380, 1695; MS 489
[M]+; HRMS calcd for C27H27N3O4S 489.1722, found 489.1726.
2-Ben zyl-10-bis(2,2,2-tr iflu or oeth oxy)-5-m eth yl-1,6-d i-
a za -4-oxa tetr a cyclo[11.3.0.05,1307,12]h exa d ec-6,8,11-tr ien -
16-on e. Obtained in 23% overall yield from (S)-4-benzyl-2-[2-
(2′-methyl-indol-3′-yl)ethyl]-2-oxazoline as a 1.2:1 mixture (1H
NMR) of (2S,5S,9S) and (2S,5R,9R) diastereomers (separated
by chromatography with 100% Et2O).
(S)-4-Ben zyl-2-[(S)-2-(in dol-3′-yl)-1-[N-(4-m eth ylph en yl)-
su lfon yla m in o]eth yl]-2-oxa zolin e (81): 54% yield from (S)-
N-[(S)-1-benzyl-2-hydroxyethyl]-3-(indol-3′-yl)-2-[N-(4-meth-
ylphenyl)sulfonylamino]propionamide; pale yellow foam; 1H
2.06 (dd, 1H, J ) 7.4, 13.2), 2.35 (s, 3H), 2.64 (dd, 1H, J ) 4.4,
13.2), 3.18 (d, 2H J ) 5.9), 3.76 (dd, 1H, J ) 5.9, 7.4), 3.86
(dd, 1H, J 1 ) J 2 ) 7.4), 3.90 (m, 1H), 4.34 (m, 1H), 5.64 (d,
1H, J ) 8.8), 6.97 (d, 1H, J ) 8.1), 7.02-7.06 (m, 2H), 7.11-
7.25 (m, 6H), 7.30 (d, 2H J ) 8.1), 7.43 (d, 2H J ) 8.1), 7.64
(d, 2H, J ) 8.1), 8.57 (s, 1H, NH ind.); IR 3400, 1665; MS 473
[M]+; HRMS calcd for C27H27N3O3S 473.1773, found 473.1774.
(S)-4-Ben zyl-2-[2-(2′-m eth yl-in d ol-3′-yl)eth yl]-2-oxa zo-
lin e (82): 65% yield from N-[(S)-1-benzyl-2-hydroxyethyl]-3-
(2′-methyl-indol-3′-yl)propionamide; colorless oil; [R]20D ) -7.5°
1
(c ) 0.95); H 2.36 (s, 3H), 2.50-2.61 (m, 3H), 3.01-3.06 (m,
(2S,5S,13S)-Isom er (89): Rf ) 0.46; minor product, 9%
1
3H), 3.95 (dd, 1H, J ) 6.6, 8.8), 4.15 (dd, 1H, J 1 ) J 2 ) 8.8),
4.35 (m, 1H), 7.08-7.32 (m, 8H), 7.52 (dd, 1H, J ) 2.2, 5.9),
7.91 (s, 1H, NH); 13C 11.5, 20.8, 29.2, 41.6, 67.2, 71.5, 110.2,
110.3, 117.9, 119.1, 120.9, 126.4, 128.4, 128.4, 129.2, 131.2,
135.2, 137.9, 167.8; IR 3405, 1660; MS (CI) 319 [M + H]+;
HRMS calcd for C21H22N2O 318.1732, found 318.1731.
yield as a brownish oil; [R]20 ) -19.3° (c ) 0.40); H 1.66 (s,
D
3H), 1.91 (dd, 1H, J ) 8.1, 11.8), 2.33 (m, 1H), 2.52-2.83 (m,
2H), 2.76 (dd, 1H, J ) 11.0, 13.2), 2.97 (dd, 1H, J ) 2.2, 13.2),
3.51 (d, 1H, J ) 13.2), 3.63-3.71 (m, 2H) 3.83-4.04 (m, 4H),
6.03 (d, 1H, J ) 2.4), 6.48 (dd, 1H, J ) 2.4, 10.2), 6.85 (d, 1H,
J ) 10.2), 7.19-7.30 (m, 5H); 13C 14.1, 31.3, 31.9, 35.8, 52.4,
Oxid a tion of In d olic Oxa zolin es. A solution of DIB (370.0
mg, 1.1 mmol) in TFE (5 mL) was added dropwise over 5 min
to a solution of indolic oxazoline (1.0 mmol) in TFE (5 mL),
and the resulting mixture was stirred for 30 min at rt. Solid
NaHCO3 (0.25 g) was added, and the suspension was filtered
58.5, 60.5 (J CF ) 36.2), 64.9, 96.2, 104.1, 119.4, 123.2 (J CF )
278.1), 125.9, 126.6, 128.5, 129.9, 136.9, 137.5, 147.1, 161.7,
174.7; 19F (188.31 MHz, CDCl3) -74.48 (br s); IR 1685; MS
(CI) 531 [M + H]+; HRMS (CI) calcd for C25H25F6N2O4 [M +
H]+ 531.1719, found 531.1724.

4408 J . Org. Chem., Vol. 65, No. 14, 2000
Braun et al.
(2S,5S,13S)-Isom er (90): Rf ) 0.21; major product, 14%
tion (C-1007), the CNRS, the MENRT, and the Re´gion
Rhoˆne-Alpes for support of our research program. We
also thank Dr. Bernard Fenet for his valuable assistance
with NOESY measurements and Laurence Rousset for
her help with mass spectral measurements. M.A.C. is
a Fellow of the Alfred P. Sloan Foundation (1994-1998).
1
yield as a brownish oil; [R]20 ) +13.3° (c ) 0.67); H 1.64 (s,
D
3H), 1.95 (m, 1H), 2.29 (dd, 1H, J ) 10.3, 13.2), 2.34-2.55 (m,
2H), 2.67 (m, 1H), 2.90 (dd, 1H, J ) 10.3, 13.2), 3.25 (dd, 1H,
J ) 4.4, 13.2), 3.52 (dd, 1H, J ) 6.6, 13.2), 3.94-4.11 (m, 4H),
4.29 (m, 1H), 6.19 (d, 1H, J ) 2.7), 6.61 (dd, 1H, J ) 2.7, 10.3),
6.93 (d, 1H, J ) 10.3), 7.12-7.29 (m, 5H). 13C: 14.1, 31.3, 31.9,
38.8, 51.2, 60.5, 60.6 (J CF ) 36.2), 60.7, 64.8, 96.2, 103.8, 122.2,
123.2 (J CF ) 278.1), 125.9, 126.8, 128.6, 128.8, 136.4, 137.2,
147.8, 162.8, 175.2; 19F (188.31 MHz, CDCl3) -74.40, -74.47;
IR 1695; MS (CI) 531 [M + H]+; CI-HRMS (CI) calcd for
C25H25F6N2O4 [M + H]+ 531.1719, found 531.1716.
Su p p or tin g In for m a tion Ava ila ble: Procedures for the
preparation of several intermediates not included in the above
Experimental Section and full physical data for these. This
material is available free of charge via the Internet at
http://pubs.acs.org.
Ack n ow led gm en t. We thank the NIH (CA-55268),
the NSF (CHE 95-26183), the Robert A. Welch Founda-
J O000341V