Effect of substitution on the intramolecular 1,3-dipolar cycloaddition of alkene tethered muenchnones

Article abstract of DOI:10.1021/jo061556t

(Chemical Equation Presented) A sequence of chemoselective activation of N-acylaminoacids, muenchnone generation, intramolecular 1,3-dipolar cycloaddition, and ring opening efficiently generated functionalized polycyclic structures such as cyclopenta[b]py

Full text of DOI:10.1021/jo061556t

Effect of Substitution on the Intramolecular 1,3-Dipolar
Cycloaddition of Alkene Tethered Mu1nchnones
Guillaume Be´langer,* Myriam April, EÄ tienne Dauphin, and Ste´phanie Roy
Laboratoire de synthe`se organique et de de´Veloppement de strate´gies de synthe`se, De´partement de Chimie,
UniVersite´ de Sherbrooke, 2500 bouleVard UniVersite´, Sherbrooke, Que´bec, J1K 2R1, Canada
guillaume.belanger@usherbrooke.ca
ReceiVed July 27, 2006
A sequence of chemoselective activation of N-acylaminoacids, mu¨nchnone generation, intramolecular
1,3-dipolar cycloaddition, and ring opening efficiently generated functionalized polycyclic structures such
as cyclopenta[b]pyrroles or zwitterionic bicyclo[4.3.0]nonane or bicyclo[3.3.0]octanes in one operation
is given. These zwitterionic species were isolated for the first time and were subsequently reduced to
bicyclic aminoalcohols. The effect of the substitution of both the dipolarophile and the mu¨nchnone on
the intramolecular cycloaddition outcome was examined. It was found that either nonactivated or electron-
poor alkenes can react with the mu¨nchnone if these alkenes are tethered at position 4 on the mu¨nchnone
(2, R2 ) alkene tether), whereas only an electron-poor alkene at position 2 (2, R3 ) alkene tether) could
undergo successful cycloaddition. Also, mu¨nchnones substituted at position 2 with a phenyl (2, R3 ) Ph)
showed a dramatic increase in reactivity, whereas a phenyl at position 4 (2, R2 ) Ph) had a very limited
effect.
Introduction
substrates can be transformed into 1,4,5,6-tetrahydrocyclopenta-
[b]pyrroles, octahydrocyclopenta[b]pyrroles, and octahydroin-
doles in high yields.
Polysubstituted pyrrole or pyrrolidine containing natural
products are extremely abundant in nature. In addition to the
rich families of pyrrolizidine and indolizidine alkaloids, one
could think of aspidospermanes, strychnanes, mesembranes,
tropanes, kopsanes, and erythranes, among others. Most of these
alkaloids contain additional rings fused to the pyrrole or
pyrrolidine core, conferring a higher degree of molecular
complexity. New methods to rapidly access such cores are thus
very interesting as a synthetic entry to a wide range of alkaloids.
We report herein that the use of mu¨nchnones (1,3-oxazolium-
5-olates) trapped intramolecularly with a pendent dipolarophile
serve this purpose very well: in a single operation, acyclic
Mu¨nchnones 2, usually prepared from the amido acid
precursors 1,1 have three sites a dipolarophile tether could be
attached to, indicated as R1, R2, and R3 in structure 1 (Scheme
1). Alkene branches at these positions could lead to different
cycloadducts following paths A, B, or C. Padwa and co-workers
reported examples that showed the viability of path A.2
However, for the examples of substrates that fell in path C,
limited success was achieved.3 Finally, no intramolecular
mu¨nchnone-alkene cycloaddition following path B has been
reported so far. Paths B and C are extremely interesting and
differ substantially from path A in that the cycloadducts 5 and
(1) For a leading reference on the preparation and use of mu¨nchnones,
see: (a) Gribble, G. W. In Synthetic Applications of 1,3-Dipolar Cycload-
dition Chemistry Toward Heterocycles and Natural Products; Padwa. A.,
Pearson, W. H., Eds; John Wiley & Sons: Hoboken, NJ, 2003; pp 681-
755. For mu¨nchnones generated by palladium-catalyzed coupling of imines,
acyl chlorides, and carbon monoxide, see: (b) Dhawan, R.; Dghaym,
R. D.; Arndtsen, B. A. J. Am. Chem. Soc. 2003, 125, 1474.
(2) (a) Padwa, A.; Gingrich, H. L.; Lim, R. J. Org. Chem. 1982, 47,
2447. (b) Padwa, A.; Lim, R.; MacDonald, J. G.; Gingrich, H. L.; Kellar,
S. M. J. Org. Chem. 1985, 50, 3816.
(3) Padwa reported three examples of Path C that all gave less than 17%
yield of desired cycloadducts (see ref 2a). Their best result is given in
Scheme 15, for the formation of pyrrole 56 from mu¨nchnone 55.
10.1021/jo061556t CCC: $37.00 © 2007 American Chemical Society
Published on Web 01/18/2007
1104
J. Org. Chem. 2007, 72, 1104-1111

Cycloaddition of Alkene Tethered Mu¨nchnones
SCHEME 1. Cycloadducts from Differently Substituted
Mu1nchnones
SCHEME 3. Path A: Synthesis of Substrate 12 and Key
Step
11 in good yields (Scheme 3).4 The ester 11 was then saponified
and activated with Mukaiyama’s salt (2-chloro-1-methylpyri-
dinium iodide) to give, after only 3 h in refluxing THF, an
almost quantitative yield of the expected cycloadducts 13 and
14 (structure of 14 confirmed by single-crystal X-ray diffrac-
tion).5,6
SCHEME 2. Possible Transformations of 6
Intramolecular 1,3-Dipolar Cycloaddition Following Path
B. With the assurance that a fully saturated tether between the
mesoionic ring and the dipolarophile could efficiently lead to
the cycloadduct in yields that were at least as good as those
obtained with unsaturated tethers,2 we started the investigation
of path B. The synthesis of the model compounds 20a-d
debuted with a Boc protection of sarcosine ethyl ester hydro-
chloride (15, Scheme 4) followed by an alkylation with either
5-iodopent-1-ene or 6-iodohex-1-ene or the iodide 177 derived
from the known silyl ether 21a8 (Scheme 5). The resulting
carbamates 18a-c were cleaved and acylated with either acetyl
chloride or benzoyl chloride to give 20a-d after saponification
of the ethyl ester.
7, compared to 3 and 4, now have a bridged nitrogen with a
lone pair of electrons correctly aligned to allow for the opening
of the hemiaminal function. As a result, the zwitterionic species
6 and 8 could be used as functionalized intermediates in the
construction of complex alkaloids. They could be further
oxidized to the corresponding substituted pyrroles (as shown
with 6, Scheme 2, right), reduced to the aminoalcohols, or
eventually trapped in a Mannich cyclization with a tethered
π-nucleophile. These transformations lead to valuable building
blocks for alkaloid syntheses.
An N-phenyl-substituted substrate was also prepared. Syn-
thesis of 26 started with an N-alkylation of N-phenylbenzamide
(23) in 87% yield based on recovered starting material (Scheme
6). C-Alkylation of the resulting ester 24 and saponification gave
the desired compound 26.
When the carboxylic acids of 20a-c were activated with
DCC9 at room temperature, a sequence of three transformations
Results and Discussion
(4) When the alkylation was performed with the benzamide obtained
from benzoylation of 9, the anion of that benzamide was quenched with
product 11 as the latter was formed in the reaction mixture.
(5) An ORTEP representation of 14 can be found in the Supporting
Information. Crystallographic data for this compound have been deposited
with the Cambridge Crystallographic Data Centre (CCDC No. 628189).
The coordinates can be obtained on request from the Director, Cambridge
Crystallographic Data Centre, 12 Union Road, Cambridge, CB2 1EZ, U.K.
or on http://www.ccdc.cam.ac.uk/deposit.
(6) Cycloadducts 13 and 14 were stable and could eventually be reduced,
resulting in the opening of the bicyclic core as shown by Padwa with similar
compounds (see ref 2b).
(7) cis-Iodide 17 was isomerically pure. About 5% of the trans isomer
was obtained from the previous hydrogenation of alkyne 22a and was
separated at the iodide stage.
We investigated the three different paths of intramolecular
1,3-dipolar cycloaddition of mu¨nchnones, as depicted in Scheme
1. So far, the only reported examples following path A all
involved substrates bearing an aromatic ring between the
dipolarophile and the mu¨nchnone.2 We wanted to verify if fully
saturated tethers between the mesoionic ring and the alkene
could be used. Because the majority of the natural products that
would be accessible through our strategy do not contain an
aromatic ring, the use of saturated tethers would thus widen
the synthetic applications.
Intramolecular 3 + 2 Cycloaddition Following Path A. A
simple model 12 was elaborated from but-3-en-1-amine (9):
N-alkylation followed by benzoylation afforded the amido ester
(8) Garc´ıa-Fandin˜o, R.; Aldegunde, M. J.; Codesido, E. M.; Castedo,
L.; Granja, J. R. J. Org. Chem. 2005, 70, 8281.
(9) Potts, K. T.; Yao, S. J. Org. Chem. 1979, 44, 977.
J. Org. Chem, Vol. 72, No. 4, 2007 1105

Be´langer et al.
SCHEME 4. Path B: Synthesis of N-Methyl Substrates
20a-d
SCHEME 7. Path B: Tandem Mu1nchnone-Alkene
Intramolecular 1,3-Dipolar Cycloaddition and Ring Opening
SCHEME 5. Synthesis of Silyl Ether 22b and Iodide 17
lation of the zwitterionic cycloadducts 27a-c occurred with
these reaction conditions.10 In fact, we were able to isolate the
pure zwitterionic products from the aqueous phase, after washing
with organic solvents. Unfortunately, treatment of acetamido-
acids 20d and 20e in the same reaction conditions afforded no
cycloadduct.11 This result will be discussed in a later section
(vide infra, Scheme 13).
SCHEME 6. Path B: Synthesis of N-Phenyl Substrate 26
The zwitterionic products 27a-c could be advantageously
used: because of the presence of both an iminium ion and a
carboxylate, one could think of reducing both of these functional
groups to the corresponding aminoalcohol or trapping the
iminium ion in a Mannich addition. The reduction was ef-
fectively performed and the aminoalcohols 28b,c were isolated
upon a one-pot treatment with NaHB(OAc)3 to reduce the
iminium ion then with LiAlH4 to effect the reduction of the
carboxylate of 27b,c (Scheme 8).12 The structure of 28c (â-Ph)
was confirmed by single-crystal X-ray diffraction.13
occurred in 29-98% yield: (1) the corresponding mu¨nchnones
were generated, (2) these mu¨nchnones underwent an intramo-
lecular 1,3-dipolar cycloaddition, and (3) the resulting hemi-
aminal ring opened to the zwitterionic adducts 27a-c (Scheme
7, following the general sequence 1 f 2 f 5 f 6 of Scheme
1). Besides being the first examples of intramolecular cycload-
dition of mu¨nchnones following path B (see Scheme 1), the
most remarkable aspect of these results is that no decarboxy-
(11) In the case of the activation of 26, the generation of the mu¨nchnone
intermediate was proven by addition of DMAD, so that the corresponding
pyrrole 49c was isolated in 99% yield. For trapping of mu¨nchnones with
DMAD, see Scheme 13.
(12) When 27b was treated with an excess of LiAlH4 alone, a 2:1 mixture
of diastereomers was obtained, presumably due to a less stereoselective
initial reduction of the iminium ion with LiAlH4 than with NaHB(OAc)3.
The major diastereomers of 28b,c were assumed to come from a reduction
of the iminium from the convex face of 27b,c.
(13) An ORTEP representation of 28c can be found in the Supporting
Information. Crystallographic data for this compound have been deposited
with the Cambridge Crystallographic Data Centre (CCDC No. 628188).
The coordinates can be obtained on request from the Director, Cambridge
Crystallographic Data Centre, 12 Union Road, Cambridge, CB2 1EZ, U.K.
or on http://www.ccdc.cam.ac.uk/deposit.
(10) To the best of our knowledge, the closest reported example is the
formation of a neutral imino acid (not zwitterionic N-alkyliminium
carboxylate as in our case) that did not decarboxylate and that was obtained
in modest yield. See: Maryanoff, C. A.; Karash, C. B.; Turchi, I. J.; Corey,
E. R.; Maryanoff, B. E. J. Org. Chem. 1989, 54, 3790.
1106 J. Org. Chem., Vol. 72, No. 4, 2007

Cycloaddition of Alkene Tethered Mu¨nchnones
SCHEME 8. Reduction of the Zwitterionic Adducts 27b,c
SCHEME 10. Path C: Synthesis of Alkynyl Substrate 34
and Key Step
SCHEME 9. Path C: Synthesis of Alkenyl Substrate 31
and Key Step
phile. The desired cascade generated the pyrrole 33 in 89% yield
through the cycloadduct intermediate 35 after CO2 extrusion.17
The electron-poor tethered dipolarophiles of compounds 31
and 34 fruitfully demonstrated the possibility of effecting the
key transformation following path C. However, from these
results, we did not know if an electron-poor dipolarophile was
absolutely needed for the intramolecular cycloaddition following
path C. To address this question, a series of substrates were
prepared.18 The elaboration of the model compounds 40a-d
started with the amide coupling of acid 38 (or its acyl chloride
derivative) with amine 37, aniline, or 4-methoxyaniline (Scheme
11). The resulting amide 39 was saponified to the amido acid
40a, whereas amides 41a,b were N-alkylated with either methyl
2-bromopropionate or methyl R-bromophenylacetate then sa-
ponified to the acids 40b-d.
Intramolecular 1,3-Dipolar Cycloaddition Following Path
C. With these results in hand, we turned to the study of path C
(cf. Scheme 1). We opted to test the mu¨nchnone generation-
intramolecular 1,3-dipolar cycloaddition-ring opening cascade
with substrates 31 and 34, both prepared from silyl ether 22b14
(Schemes 9 and 10). The latter was hydrolyzed and oxidized
with use of Jones’ conditions. The resulting carboxylic acid was
coupled with methyl 2-(phenylamino)propionate15 via the acyl
chloride intermediate in 85% yield for the two steps. Compound
29 was partially hydrogenated with ethylenediamine to poison
the catalyst then quantitatively saponified to the alkene-acid
31. When the latter was activated with DCC, the formation of
the mu¨nchnone occurred, followed by intramolecular 1,3-dipolar
cycloaddition with the tethered alkene and hemiaminal cleavage.
The resulting zwitterionic species 32 spontaneously decarboxy-
lated and was further oxidized with air to the corresponding
pyrrole 33, as the only observed product.16
In another sequence, the vinylic chloride 40e and methyl enol
ether 40f were prepared by Wittig olefinations of a common
aldehyde 44 (Scheme 12). The latter was straightforwardly
derived from penta-1,5-diol (42).
When subjected to the acid activation conditions, 40a-f gave
no desired intramolecular cycloadducts 46a-f or any ring-
opened products 47a-f or further decarboxylated compounds
48a-f (Scheme 13). This contrasts with the fast and high
yielding intramolecular reactions of mu¨nchnones bearing non-
activated alkenyl dipolarophiles following path B (Scheme 7).
For path C, nonactivated alkenes (45a-d) or slightly electron-
poor chloroalkene (45e) tethered to the mu¨nchnone at the R3
position were not reactive enough. The electron-rich dipolaro-
phile of 45f did not give intramolecular cycloaddition either,
which might have been envisaged through a reverse electron
demand 1,3-dipolar cycloaddition. Nonetheless, it should be
noted that in every case the generation of the mu¨nchnone
intermediate (45a-f) unequivocally occurred within 30 min at
The structure of the pentasubstituted pyrrole 33 was con-
firmed by the preparation of the same compound through the
route depicted in Scheme 10. Although the yield for the
formation of pyrrole 33 derived previously from the vinylic
amide 31 was somewhat low, a much better conversion was
obtained when acetylenic amide 34 was used as the dipolaro-
(17) Similar intramolecular alkyne-mu¨nchnone cycloadditions following
path C were reported firstly by Kato with a meta-substituted aryle as an
unsaturated tether (Kato, H.; Wang, S.-Z.; Nakano, H. J. Chem. Soc., Perkin
Trans. 1 1989, 361) and later by Padwa (see ref 2b). Tethers containing a
ketone also proved to be successful in generating the cycloadducts (Pinho
e Melo, T. M. V. D.; Barbosa, D. M.; Ramos, P. J. R. S.; d’A Rocha
Gonsalves, A. M.; Gilchrist, T. L.; Beja, A. M.; Paixa˜o, J. A.; Silva,
M. R.; Alte da Viega, L. J. Chem. Soc., Perkin Trans. 1 1999, 1219). The
use of fully saturated tethers between the alkyne and the mu¨nchnone was
reported by Sainsbury, although modest yields were obtained (Sainsbury,
M.; Strange, R. H.; Woodward, P. R.; Barsanti, P. A. Tetrahedron 1993,
49, 2065).
(18) The preparation of other electron-poor alkenyl dipolarophiles
substituted with alkoxycarbonyl, cyano, or nitro groups failed. These
substituents did not tolerate the saponification conditions needed to generate
the required carboxylic group for the key transformation. The generation
of the carboxylic acid prior to the installation of the alkoxycarbonyl, cyano-
or nitro-substituted alkene was also unsuccessful.
(14) Guay, B.; Deslongchamps. P. J. Org. Chem. 2003, 68, 6140.
(15) Prepared by using the procedure described in: Gaertzen, O.;
Buchwald, S. L. J. Org. Chem. 2002, 67, 465. For characterization data,
see: Gately, D. A.; Norton, J. R. J. Am. Chem. Soc. 1996, 118, 3479.
(16) The following proposed mechanism for the decarboxylation and
oxidation of the zwitterionic adduct 32 was also suggested by Padwa (see
ref 2a):
J. Org. Chem, Vol. 72, No. 4, 2007 1107

Be´langer et al.
SCHEME 13. Path C: Attempted Intramolecular
SCHEME 11. Path C: Synthesis of Substrates 40a-d
1,3-Dipolar Cycloadditions
SCHEME 14. Decarboxylation of 27b and 27c
SCHEME 12. Path C: Synthesis of Substrates 40e,f
only reacted with an electron-poor dipolarophile (unsaturated
ester) and led to the corresponding pyrrole (33) after spontaneous
decarboxylation of the zwitterionic product (Scheme 9). The
difference in the reaction products (zwitterionic adduct versus
pyrrole) could be explained by looking at the temperature
required for the 1,3-dipolar cycloadditions: For path B, the
intramolecular cycloaddition occurred at room temperature
(Scheme 7). For path C, the mu¨nchnone intermediates were also
formed at room temperature (see Scheme 13, where mu¨nchnones
45a-f were trapped intermolecularly with DMAD at room
temperature) but the intramolecular 1,3-dipolar cycloaddition
required heat (75 °C). Seemingly, the resulting zwitterionic
adduct 32 was heat-sensitive and readily decarboxylated to lead
to the observed pyrrole 33 (Scheme 9). To verify this hypoth-
esized heat-sensitivity of the zwitterionic adducts, 27b and 27c
(obtained from path B) were heated to 55-60 °C and were
completely consumed within 20 min (Scheme 14). We were
able to isolate the corresponding pyrrole 51 from decarboxy-
lation and air oxidation of the zwitterionic compound 27c,16
whereas only traces of 50 were identifiable when 27b was
subjected to the same conditions.
room temperature upon activation of 40a-f with DCC,19 as
confirmed by the isolation of the corresponding pyrroles 49a-f
in good to quantitative yield when dimethyl acetylenedicar-
boxylate (DMAD) was added to the reaction mixture.20
So far, all of our results showed an important difference
between the reactivity of alkene tethered mu¨nchnones branched
at the R2 position (path B) and those branched at the R3 position
(path C). The former gave the zwitterionic cycloadducts (27a-
c) with electron-poor dipolarophiles and with unactivated
dipolarophiles (Scheme 7), whereas R3-branched mu¨nchnone
(19) Comparable results were obtained when the activation of the
carboxylic acid and generation of the mu¨nchnone were initiated with either
DCC or Mukaiyama’s salt.
(20) The generation of tetrasubstituted pyrroles by intermolecular cy-
cloaddition of mu¨nchnones with alkynyl dipolarophiles has already been
extensively documented. See ref 1.
These results also suggest that the low yield (23%) obtained
for the conversion of substrate 31 to pyrrole 33 (Scheme 9)
could be due to the decarboxylation step only, as suggested by
1108 J. Org. Chem., Vol. 72, No. 4, 2007

Cycloaddition of Alkene Tethered Mu¨nchnones
SCHEME 15. Effect of Substitution on Intramolecular 1,3-Dipolar Cycloaddition of Alkene-Tethered Mu1nchnones
the yield (23%) obtained for the decarboxylation of 27c to 51
(Scheme 14). But even though the cycloaddition step in the
formation of 33 might be high yielding, this cycloaddition
following path C is nonetheless more difficult than any
cycloaddition according to path B, as a higher temperature and
an electron-poor dipolarophile were absolutely necessary to the
success of path C. This brings another interesting question: why
are intramolecular 1,3-dipolar cycloadditions of alkene tethered
mu¨nchnones more difficult following path C than path B? To
answer this question, two distinct factors were analyzed: (1)
substitution of the dipolarophile and (2) substitution of the
mu¨nchnone.
no cycloaddition was observed for their nonactivated alkene
parents 52d and 45b (Scheme 15).
Effect of the Substitution of the Mu1nchnone on the
Intramolecular Mu1nchnone-Alkene 1,3-Dipolar Cycload-
dition. Switching from a methyl to a phenyl conferred a
dramatically different reactivity to the mu¨nchnone when that
change was made at position R3 as confirmed by the cycload-
ditions of 52d (Me) and 52b (Ph), giving 0% and 98% yield,
respectively (Scheme 15). Interestingly, the difference was
apparently not as spectacular when the same switch was made
at the R2 position since both 45b (Me) and 45c (Ph) resulted in
no cycloaddition. The beneficial effect of a phenyl at the R3
position was also exemplified in Padwa’s group by the cycload-
dition of mu¨nchnone 55, bearing an o-butenylaryl group at the
R3 position.2a This aryl substitution allowed for the intramo-
lecular cycloaddition with a nonactivated alkene (cf. 45c and
55).
Effect of the Substitution of the Dipolarophile on the
Intramolecular Mu1nchnone-Alkene 1,3-Dipolar Cycload-
dition. It is known that mu¨nchnones act as type-I dipoles
(HOMOdipole-LUMOdipolarophile control) and prefer to react with
electron-poor dipolarophiles,21 as supported by the Frontier
Molecular Orbital (FMO) theory.22 Indeed, the electron-poor
alkenes of 52a and 54 gave the desired cycloadducts whereas
Additionally and most strikingly, changing the methyl to a
phenyl at position R3 on the mu¨nchnone (cf. 52d and 52b) was
(22) (a) Dewar, M. J. S. In Molecular Orbital Theory of Organic
Chemistry; McGraw-Hill: New York, 1969. (b) Dewar, M. J. S.; Dougherty,
R. C. In The PMO Theory of Organic Chemistry; Plenum Press: New York,
1975. (c) Fleming, I. Frontier Orbitals and Organic Chemical Reactions;
John Wiley & Sons: New York, 1976.
(21) (a) Sustmann, R. Tetrahedron Lett. 1971, 2717. (b) Sustmann, R.;
Trill, H. Angew. Chem., Int. Ed. Engl. 1972, 11, 838. (c) Sustmann, R.
Pure Appl. Chem. 1974, 40, 569. (d) Huisgen, R. J. Org. Chem. 1976, 41,
403.
J. Org. Chem, Vol. 72, No. 4, 2007 1109

Be´langer et al.
SCHEME 16. Conformational Analysis of Mu1nchnone 53
remarkably more beneficial for the cycloaddition than just
adding an electron-withdrawing group on the dipolarophile (cf.
52d and 52a). The conjugation of a phenyl group with the
mu¨nchnone (i.e., on the amide carbonyl of the starting material)
might in fact play two roles: (1) it increases the reactivity of
the mu¨nchnone23 and (2) it increases the thermal stability of
the mu¨nchnone. These two roles must be amplified when the
phenyl is at the R3 position on the mu¨nchnone compared to the
R2 position in order to fit the observations discussed above.
The geometry of the reactive conformation should not play an
important role in this difference in reactivity because the
geometrical constraint imposed to the tether in the reactive
conformation is the same for the R2- and R3-tethered alkenes.
However, when examining electronic factors, a tentative ex-
planation comes up: a phenyl at position R3 can delocalize both
charges of the mu¨nchnone through linear conjugation,24 thus
increasing the overall reactivity of the mu¨nchnone,23 whereas a
phenyl at position R2 can only delocalize the negative charge
of the mu¨nchnone through cross conjugation,25 which should
lower the energy of the mu¨nchnone HOMO and render the
mu¨nchnone less reactive.23
Finally, comparison between the high yield of cycloadduct
27b through mu¨nchnone intermediate 52b and the absence of
intramolecular cycloaddition for mu¨nchnone 53 tends to indicate
a dramatic decrease in reactivity due to the addition of a phenyl
substituent on the mu¨nchnone nitrogen. By looking at the
conformations of mu¨nchnone 53, we find that the latter could
simply not adopt a conformation in which both phenyls would
be coplanar as in 53A, due to severe steric interactions between
the phenyls (Scheme 16). Instead, one of the phenyls has to be
perpendicular to the mu¨nchnone plane, as in 53B or 53C, thus
seriously hindering the approach of the alkenyl dipolarophile.26
to 45d was not sufficient to allow for an intramolecular 1,3-
dipolar cycloaddition following path C, whereas the decrease
in energy of the dipolarophile LUMO played a beneficial role
on the cycloaddition outcome (cf. 45b and 54).
Conclusions
In summary, we successfully developed two reaction se-
quences leading to either iminium-carboxylate ylides that could
be further reduced to bicyclic amino alcohols or pyrrole-
containing bicyclic systems. These motifs are extremely abun-
dant in alkaloids’ skeletons. The routes that were designed are
highly convergent and the tandem mu¨nchnone generation-
intramolecular 1,3-dipolar cycloaddition-ring opening gave
good to high yields of valuable bicycloadducts. As expected,
electron-poor alkenes are preferred for the intramolecular 1,3-
dipolar cycloaddition of alkene tethered mu¨nchnone and could
make the difference between no reaction and a successful
cycloaddition. Also, a dramatic increase of reactivity for
mu¨nchnones conjugated with an aromatic ring at the R3 position
(path B) was observed. This conjugation proved to be even more
beneficial on the cycloaddition outcome than the effect of adding
an electron-withdrawing substituent on the tethered dipolarophile
for path B, whereas conjugation of an additional phenyl at the
R2 position was not sufficient for a successful cycloaddition
following path C.
Experimental Section
rac-6-Methyl-3-phenyl-4-oxa-7-azatricyclo[4.3.0.03,7]nonan-5-
one (13) and 3-Methyl-6-phenyl-5-oxa-7-azatricyclo[4.3.0.03,7]-
nonan-4-one (14). Et3N (3.9 mL, 28 mmol) and 2-chloro-N-
methylpyridinium iodide (Mukaiyama’s salt, 0.22 g, 0.84 mmol,
dried azeotropically by coevaporations with benzene) were added
to a solution of 12 (0.208 g, 0.840 mmol) in THF (14 mL). The
reaction mixture was heated at reflux for 4 h, and then concentrated
under reduced pressure. The crude material was dissolved in EtOAc
and a saturated aqueous solution of NaHCO3 was added. The
resulting mixture was vigorously stirred for 5 min, and then the
layers were separated. The aqueous phase was extracted with
EtOAc. The organic layers were combined, washed with a saturated
aqueous solution of NaCl, dried over anhydrous Na2SO4, filtered,
and concentrated under reduced pressure. A first purification by
flash chromatography (silica gel, 30% EtOAc in hexanes) afforded
a 2:3 mixture of 13 and 14 (0.188 g, 97%). A second purification
by flash chromatography (silica gel, 5% acetone in toluene) afforded
13 (white semisolid) and pure 14 (white solid). 13: 1H NMR (300
MHz, CDCl3) δ (ppm) 7.64-7.60 (m, 2H), 7.44-7.43 (m, 3H),
2.91 (ddd, J ) 13.5, 10.5, 2.5 Hz, 1H), 2.54-2.47 (m, 2H), 2.35
(ddd, J ) 12.5, 5.5, 2.5 Hz, 1H), 2.07-1.98 (m, 1H), 1.73 (d, J )
13.5 Hz, 1H), 1.58 (ddd, J ) 11.5, 8.0, 2.5 Hz, 1H), 1.44 (s, 3H);
IR (film) ν (cm-1) 3001-2890, 1784, 1343; MS (EI) m/z (rel %)
This analysis suggests that the additional phenyl on the
nitrogen does not deactivate the mu¨nchnone by a conjugation
effect (or FMO argument) but rather by a conformational effect.
Furthermore, results of path C even tend to reveal that
substitution on the mu¨nchnone nitrogen has in fact a limited
effect on the cycloaddition, since no cycloaddition was observed
when going from an N-methyl (45a), to an N-phenyl (45c), to
an N-(4-methoxy)phenyl (45d) (Scheme 15). Apparently, the
increase in energy of the mu¨nchnone HOMO from 45a to 45c
(23) Conjugation of a phenyl with a dipole (or a dienophile) reduces the
energy of its LUMO and increases the energy of its HOMO, thus giving
rise to a more reactive dipole toward both electron-rich and electron-poor
dipolarophile (see ref 22c).
(24) Linear conjugation of the phenyl in R3 with the mu¨nchnone:
(26) Mu¨nchnone 53 intermediate was successfully trapped with DMAD
and gave the corresponding pyrrole. This constituted a proof that 53 was
indeed formed from the acid 26 activation. The reaction of 53 with DMAD,
compared to the absence of reaction with the tethered alkene, could be
understood from two standpoints: (1) DMAD is a highly activated electron-
poor dipolarophile compared to the nonactivated tethered alkene and (2)
an alkyne is generating much less steric congestion when it approaches a
hindered dipolarophile than a substituted alkene does.
(25) Cross conjugation of the phenyl in R2 with the mu¨nchnone:
1110 J. Org. Chem., Vol. 72, No. 4, 2007

Cycloaddition of Alkene Tethered Mu¨nchnones
229 [M+] (50), 201 (15), 105 (100); HRMS (EI) calcd for C14H15-
The white suspension was then filtered on a pad of Celite (Et2O
washings). The layers were separated and the organic layer was
washed with a saturated aqueous solution of NaCl, dried over
anhydrous Na2SO4, filtered, and concentrated under reduced pres-
sure. The crude material was purified by flash chromatography
(silica gel, 20 to 50% EtOAc in hexanes) to afford 28b (0.061 g,
50%) as a colorless oil: 1H NMR (300 MHz, CDCl3) δ (ppm) as
a 8:1 mixture of diastereomers, 7.39-7.21 (m, 5H), 3.88 (d, J )
11.0 Hz, 1H), 3.83 (dd, J ) 10.5, 6.0 Hz, 1H), 3.61 (d, J ) 11.0
Hz, 1H), 3.52 (d, J ) 10.5 Hz, 1H, minor diastereomer), 3.31 (d,
J ) 10.5 Hz, 1H, minor diastereomer), 2.52-2.44 (m, 1H), 2.37
(dd, J ) 12.0, 5.5 Hz, 1H), 2.34 (dd, J ) 12.0, 5.5 Hz, 1H), 2.21
(s, 3H), 2.00 (s, 3H, minor diastereomer), 1.99-1.82 (m, 2H), 1.69-
1.50 (m, 4H), 1.31 (ddd, J ) 12.0, 10.0, 8.0 Hz, 1H), 1.16 (td, J )
12.0, 6.0 Hz, 1H); 13C NMR (75 MHz, CDCl3) δ (ppm) as a 8:1
mixture of diastereoisomers, 144.4, 128.4, 128.2, 127.1, 126.9, 75.7,
69.6, 68.5, 66.9, 64.2, 44.0, 43.7, 42.6, 34.4, 33.9, 32.9, 32.2, 29.2,
26.0, 24.3; IR (film) ν (cm-1) 3406 (br), 2945, 2860, 2787, 1453,
759, 700. MS (EI) m/z (rel %) 231 [M+] (1), 200 [(M - CH2-
OH)+] (100); HRMS (EI) calcd for C15H21NO [M+] 231.1623,
found 231.1619 ( 0.0007.
NO2 [M+] 229.1103, found 229.1100 ( 0.0007. 14: mp 139-142
1
°C; H NMR (300 MHz, CDCl3) δ (ppm) 7.65-7.62 (m, 2H),
7.44-7.41 (m, 3H), 3.08 (dd, J ) 5.5, 3.5 Hz, 1H), 2.87 (ddd, J )
13.5, 10.5, 3.5 Hz, 1H), 2.66 (ddd, J ) 13.5, 9.0, 7.0 Hz, 1H),
2.00 (ddd, J ) 12.5, 5.5, 2.5 Hz, 1H), 1.70-1.58 (m, 1H), 1.52
(ddd, J ) 12.5, 9.0, 3.5 Hz, 1H), 1.49 (s, 3H), 1.19 (d, J ) 13.5
Hz, 1H); 13C NMR (75 MHz, CDCl3) δ (ppm) 174.8, 132.3, 129.8,
128.7, 127.9, 106.2, 69.1, 42.3, 41.0, 36.8, 33.0, 12.7; IR (film) ν
(cm-1) 2993-2882, 1801, 1454, 1347; MS (EI) m/z (rel %) 229
[M+] (50), 201 (15), 105 (100); HRMS (EI) calcd for C14H15NO2
[M+] 229.1103, found 229.1100 ( 0.0007.
Zwitterion 27b. DCC (0.147 g, 0.713 mmol) was added to a
solution of 20b (0.169 g, 0.646 mmol) in CH2Cl2 (6 mL). The
reaction mixture was stirred at room temperature for 30 min and
then concentrated under reduced pressure. After addition of EtOAc
and water, the layers were separated. The aqueous layer was washed
with EtOAc and concentrated under reduced pressure. The resulting
crude material was dissolved in CH2Cl2, dried over anhydrous Na2-
SO4, filtered, and concentrated under reduced pressure to afford
27b (0.155 g, 98%) as a brown hygroscopic solid: 1H NMR (300
MHz, CDCl3) δ (ppm) 7.66-7.55 (m, 5H), 3.84 (ddd, J ) 19.5,
9.5, 2.5 Hz, 1H), 3.61 (s, 3H), 3.19 (q, J ) 7.0 Hz, 1H), 3.07 (d,
J ) 19.5 Hz, 1H), 2.77 (dt, J ) 14.5, 7.0 Hz, 1H), 2.23-2.05 (m,
2H), 1.95 (dqi, J ) 12.0, 6.0 Hz, 1H), 1.61 (dqi, J ) 14.5, 7.0 Hz,
1H), 1.48 (dq, J ) 12.0, 6.0 Hz, 1H); 13C NMR (75 MHz, CDCl3)
δ (ppm) 183.2, 170.2, 134.1, 129.6, 128.6, 127.1, 96.0, 45.4, 43.2,
37.1, 34.6, 34.2, 25.8; IR (film) ν (cm-1) 3479 (br), 2944, 2869,
1631, 1448, 1367.
Acknowledgment. This research was supported by the
Natural Science and Engineering Research Council (NSERC)
of Canada, FQRNT (Que´bec), the Canadian Fund for Innovation
(CFI), and the Universite´ de Sherbrooke. Universite´ de Sher-
brooke graduate fellowships to M.A. and EÄ .D. are also gratefully
acknowledged.
rac-(1-Methyl-2-phenylhexahydrocyclopenta[b]pyrrol-6a-yl)-
methanol (28b). NaHB(OAc)3 (0.343 g, 1.62 mmol) was added to
a solution of 27b (0.127 g, 0.521 mmol) in CH2Cl2 (10 mL) at
-78 °C. The reaction mixture was allowed to warm slowly to room
temperature then was stirred at room temperature for 12 h under
nitrogen flow to remove CH2Cl2. THF (10 mL) and LiAlH4 (0.160
g, 4.22 mmol) were added. The reaction mixture was heated at
reflux for 3 h then cooled to room temperature. Distilled water
and a 1 N solution of NaOH were added to the reaction mixture.
Supporting Information Available: Experimental procedures
for compounds 10-12, 16-18c, 19b, 20a-d, 22a,b, 24-27a, 27c,
1
28c-31, 33, 34, 37, 39-41b, 43, 44, 49a-f, and 51, H and 13C
NMR spectra for compounds 10-14, 16-18c, 19b, 20a-22b, 24-
31, 33, 34, 37, 39-41b, 43, 44, 49a-f, and 51, and crystallographic
data for compounds 14 and 28c. This material is available free of
charge via the Internet at http://pubs.acs.org.
JO061556T
J. Org. Chem, Vol. 72, No. 4, 2007 1111