Total Synthesis of the Proposed Structures of Indole Alkaloids Lyaline and Lyadine

Article abstract of DOI:10.1021/jo035507h

The harman-1,4-dihydropyridines 1 and 2, which constitute the originally proposed structures for the indole alkaloids lyaline and lyadine, have been synthesized, and their NMR data have been compared with those available for the natural products. Due to the discrepancies in the spectral data, the structures of lyadine and lyaline should be revised.

Full text of DOI:10.1021/jo035507h

Tota l Syn th esis of th e P r op osed Str u ctu r es of In d ole Alk a loid s
Lya lin e a n d Lya d in e
M.-Llu¨ısa Bennasar,* Toma`s Roca, and Manuel Monerris
Laboratory of Organic Chemistry, Faculty of Pharmacy, University of Barcelona, 08028 Barcelona, Spain
bennasar@ub.edu
Received October 13, 2003
The harman-1,4-dihydropyridines 1 and 2, which constitute the originally proposed structures for
the indole alkaloids lyaline and lyadine, have been synthesized, and their NMR data have been
compared with those available for the natural products. Due to the discrepancies in the spectral
data, the structures of lyadine and lyaline should be revised.
CHART 1. Mon oter p en oid In d ole Alk a loid s fr om
P a u r id ia n th a lya lly
In tr od u ction
In 1974, Cave´ and co-workers reported the isolation
of a small subgroup of monoterpenoid indole alkaloids
from Pauridiantha lyalli (Rubiaceae),1 a Madagascan
plant whose roots have been traditionally used for their
febrifuge properties. The spectroscopic data of these
alkaloids revealed a particular skeletal arrangement,
with a fully aromatic harman nucleus connected to the
4-position of a 3,5-disubstituted pyridine ring: 1,4-
dihydropyridine in the major components lyaline (1,
Chart 1) and lyadine (2)2 and dihydro-2-pyridone in the
minor components lyalidine (3) and hydroxylyalidine (4).3
These structures suggested a clear relationship with
lyaloside (5, Scheme 1), a glucoindole alkaloid belonging
to the Vincosan biogenetic type,4 which co-occurs in the
same plant.5 Thus, the lyaline alkaloids would derive
biogenetically (or would be simple artifacts) from 5, by
hydrolysis of the glucosidic bond, followed by incorpora-
tion of ammonia into the D-ring.6 In the context of
biomimetic transformations in this field, it has been
reported that 5 can be converted into the ethylidene
hemiacetal A by treatment with â-glucosidase.7 However,
subsequent attempts to reproduce this result were un-
successful since only simple harman, a main constituent
of Pauridiantha lyalli and other Rubiaceous plants, was
SCHEME 1
recovered from the reaction mixtures.8 From a synthetic
standpoint, only the structure of lyaloside (5) has been
confirmed by total synthesis.9
The structural type proposed for lyaline (1) and lyadine
(2) attracted our interest due to the fact that N-unsub-
stituted 4-alkyl-1,4-dihydropyridines,10 carrying only one
electron-withdrawing group at the â-position, are sup-
posed to be very sensitive, perhaps too much so to
* To whom correspondence should be addressed. Tel: 34 934 024
540. Fax: 34 934 024 539.
(1) Pousset, J . L.; Levesque, J .; Cave´, A.; Picot, F.; Potier, P.; Paris,
R. R. Plantes Med. Phytother. 1974, 8, 51-56.
(2) Levesque, J .; Pousset, J . L.; Cave´, A.; Cave´, A. C. R. Acad. Sci.
Paris C 1974, 278, 959-961.
(3) Levesque, J .; Pousset, J . L.; Cave´, A. C. R. Acad. Sci. Paris C
1974, 279, 1053-1055.
(4) Kisaku¨rek, M. V.; Leeuwenberg, A. J . M.; Hesse, M. In Alka-
loids: Chemical and Biological Perspectives; Pelletier, S. W., Ed.;
Wiley: New York, 1983; Chapter 5.
(5) (a) Levesque, J .; Pousset, J .-L.; Cave´, A. C. R. Acad. Sci. Paris
C 1975, 280, 593-595. (b) Levesque, J .; J acquesy, R.; Foucher, J . P.
Tetrahedron 1982, 38, 1417-1424.
(6) This two-step biogenetic transformation is believed to operate
in other indolopyridine alkaloids. For a recent example, see: Takaya-
ma, H.; Tsutsumi, S.; Kitajima, M.; Santiarworn, D.; Liawruangrth
B.; Aimi, N. Chem. Pharm. Bull. 2003, 51, 232-233.
(7) Levesque, J .; J acquesy, R.; Merienne, C. J . Nat. Prod. 1983, 46,
619-625.
(8) Aimi, N.; Murakami, H.; Tsuyuki, T.; Nishiyama, T.; Sakai, S.;
Haginiwa, J . Chem. Pharm. Bull. 1986, 34, 3064.
(9) Aimi, N.; Seki, H.; Sakai, S. Chem. Pharm. Bull. 1992, 40, 2588-
2590.
(10) For reviews on the chemistry of dihydropyridines, see: (a)
Eisner, U.; Kuthan, J . Chem. Rev. 1972, 72, 1-42. (b) Stout, D.;
Meyers, A. I. Chem. Rev. 1982, 82, 223-243. (c) Sausins, A.; Duburs,
G. Heterocycles 1988, 27, 291-314. (d) Lavilla, R. J . Chem. Soc., Perkin
Trans. 1 2002, 1141-1156.
10.1021/jo035507h CCC: $27.50 © 2004 American Chemical Society
Published on Web 01/08/2004
752
J . Org. Chem. 2004, 69, 752-756

Synthesis of the Indole Alkaloids Lyaline and Lyadine
SCHEME 2. Syn th etic Ap p r oa ch
SCHEME 3. Syn th esis of
Ha r m a n -Dih yd r op yr id in es 12 a n d 13a
withstand the rigorous conditions when being extracted
from natural sources. The above observation together
with the fact that partial NMR data are given in the
original paper1,2 prompted us to develop a synthetic
sequence for these alkaloids in order to verify their
structure.
The literature data11 and our own experience12 in the
synthesis of 4-substituted 1,4-dihydropyridines by regi-
oselective addition of carbon nucleophiles to suitably
activated pyridines led us to envisage the use of an
organocopper derivative of harman in this reaction as a
logical approach to the target molecules. With the aim
of assembling the harman-dihydropyridine skeleton of
1 and 2 in a straightforward and convergent way, we
selected 3,5-disubstituted N-acylpyridinium salts,13 such
as those depicted in Scheme 2, as electrophilic partners.
Afterward, we should be able to carefully induce N-
deacylation to gain access to the desired N-unsubstituted
derivatives. In the lyadine (2) series, an additional
reduction of the acetyl group would be required to
complete the synthesis.
a
Reagents and conditions: (a) (i) n-BuLi, THF, 0 °C, 30 min,
Resu lts a n d Discu ssion
(ii) CuCN, THF, 0 °C, then ZnCl2, -78 °C; (b) acetyl chloride, THF,
or CH2Cl2, then TMSOTf (for 11c); (c) -78 °C to rt, 12 h.
We set out to explore the feasibility of this proposal
using Na-methylharman (6)14 as a model organometallic
precursor and easily available pyridines 8a , 8b,15 and
8c,15,16 which bear one or two electron-withdrawing
substituents at the â-positions (Scheme 3). After some
experimentation with different organometallic species,17
it was found that the mixed copper-zinc organometallic
9, prepared from 6 by sequential treatment with n-Buli,
CuCN, and ZnCl2, was the best reagent in terms of yield
and regioselectivity.18 Pyridines 8a and 8b were con-
verted into N-acetylpyridinium chlorides 11a and 11b by
treatment with acetyl chloride and then allowed to react
in situ with 9. In the 3-acetyl (c) series, better yields were
observed when using a triflate counterion, prepared by
the exchange of chloride with trimethylsilyl triflate.19 The
results of this preliminary study are summarized in Table
1. As can be observed, reaction of 9 with pyridinium salt
11a took place with moderate yield and acceptable
regioselectivity to give a 4.5:1 mixture of separable 1,4-
dihydropyridine 12a and 1,2-dihydropyridine 14a in 50%
overall yield (entry 1). Starting from 3-ethylpyridinium
salt 11b (entry 2), which bears only one acyl group at
the â-position, the regioselectivity of the addition was
clearly lower. In this case, from the 4:1:1 mixture of
dihydropyridines we were able to isolate the major
component 12b. Finally, pyridinium triflate 11c (entry
3) was a more convenient electrophilic substrate as it led
to 1,4-dihydropyridine 12c as the main product (40%).
(11) See inter alia: (a) Comins, D. L.; Abdullah, A. H. J . Org. Chem.
1982, 47, 4315-4319. (b) Comins, D. L.; Stroud, E. D.; Herrick, J . J .
Heterocycles 1984, 22, 151-157. (d) Akiba, K.; Iseki, Y. Wada, M. Bull.
Chem. Soc. J pn. 1984, 57, 1994-1999. (e) Mangeney, P.; Gosmini, R.;
Raussou, S.; Commerc¸on, M.; Alexakis, A. J . Org. Chem. 1994, 59,
1877-1888. (f) Yamada, S.; Misono, T.; Ichikawa, M.; Morita, C.
Tetrahedron 2001, 57, 8939-8949.
(12) (a) Bennasar, M.-L.; Vidal, B.; Bosch, J . J . Org. Chem. 1997,
62, 3597-3609. (b) Bennasar, M.-L.; J uan, C.; Bosch, J . Tetrahedron
Lett. 1998, 39, 9275-9278. (c) Bennasar, M.-L.; Vidal, B.; Kumar, R.;
La´zaro, A.; Bosch, J . Eur. J . Org. Chem. 2000, 3919-3925. (d)
Bennasar, M.-L.; Zulaica, E.; J uan, C.; Alonso, Y.; Bosch, J . J . Org.
Chem. 2002, 67, 7465-7474. (e) Bennasar, M.-L.; Roca, T.; Monerris,
M.; J uan, C.; Bosch, J . Tetrahedron 2002, 58, 8099-8106.
(13) For the use of an intramolecular enamide addition to a 3,5-
disubstituted N-acylpyridinium salt in the synthesis of indolopyridine
alkaloids, see: (a) Lavilla, R.; Gullo´n, F.; Bosch, J . Eur. J . Org. Chem.
1999, 373, 3-378.
(14) (a) Karrer, P.; Mu¨ller, H. J . Org. Chem. 1957, 22, 1433-1434.
(b) Seki, H.; Hashimoto, A.; Hino, T. Chem. Pharm. Bull. 1993, 41,
1169-1172.
(15) Bracher, F.; Papke, T. Monatsh. Chem. 1995, 126, 805-809.
(16) Bennasar, M.-L.; Zulaica, E.; Roca, T. Alonso, Y.; Monerris, M.
Tetrahedron Lett. 2003, 44, 4711-4744.
(18) For the use of this type of organometallic in the synthesis of
4-benzylpyridines, see: (a) Chia, W.-L.; Shiao, M.-J . Tetrahedron Lett.
1991, 32, 2033-2034. (b) Shing, T.-L.; Chia, W.-L.; Shiao, M.-J .; Chau,
T.-Y. Synthesis 1991, 849-850.
(19) Pabel, J .; Ho¨sl, C. E.; Maurus, M.; Ege, M.; Wanner, K. Th. J .
Org. Chem. 2000, 65, 9272-9275.
(17) The corresponding copper-catalyzed Grignard reagent and the
lower order cyanocuprate gave mixtures of dihydropyridines in poor
yields.
J . Org. Chem, Vol. 69, No. 3, 2004 753

Bennasar et al.
TABLE 1. Rea ction of Mixed Cop p er -Zin c
Or ga n om eta llic Der iva tives 9 a n d 10 w ith P yr id in iu m
Sa lts 11
SCHEME 5. Syn th esis of 1
pyridinium
salt
products
(ratio)
overall
yield (%)
entry organometallic
1
2
9
9
11a
11b
12a + 14a
(4.5:1)
50
40
12b + 14b
+16b
(4:1:1)
3
4
9
10
11c
11c
12ca
40
30
13c + 17c
(9:1)
5
10
11d
13d + 15d
(10:1)
30
a
Trace amounts of a regioisomeric 1,2-dihydropyridine.
tion, to give 6 and 8b. Oxidation at the benzylic position
at any stage would account for the minor formation of
19.
SCHEME 4. N-Dep r otection of
1,4-Dih yd r op yr id in es 12
At this point, access to the natural products required
the extension of the chemistry outlined above to Na-
unsubstituted harman (7, Scheme 3) or to a protected
derivative. We wanted to avoid protection techniques as
they would involve an increase in the number of reactions
to perform on sensitive dihydropyridines. Hence, we were
pleased to find that the mixed copper-zinc organome-
tallic 10, prepared in situ from 7, underwent regioselec-
tive addition to 3-acetylpyridinium salt 11c to give 1,4-
dihydropyridine 13c along with minor amounts of
regioisomer 17c in an overall yield only slightly lower
(30%) than in the above Na-methyl series (Table 1, entry
4). The reaction was satisfactorily extended to vinylpy-
ridinium salt 11d , prepared as above from pyridine 8d ,
to give 13d as the major product (30%, entry 5).
The only remaining transformation required to com-
plete the synthesis of 1 from 13d was the removal of the
N-acetyl group (Scheme 5). Given the experience acquired
in the model series, we had to find conditions mild enough
to avoid, if possible, the concomitant cleavage that had
taken place from the closely related dihydropyridine 12b.
To this end, 13d was treated with NaOMe in MeOH for
2 min. After extractive workup with AcOEt (chlorinated
solvents were deliberately excluded), the crude product
was shown by 1H NMR to be the desired harman-
dihydropyridine 1 along with minor amounts (10%) of
fragmentation products (pyridine 8d , harman 7, and
aldehyde 20 in a 4:1:3 relationship). To minimize the
undesired decomposition, the deprotection experiment
was reproduced in an NMR tube using CD3OD as the
solvent. In this manner, we were able to extensively
examine the structure of our pure synthetic product by
1H (500 MHz) and 13C NMR (100.6 MHz) with the aid of
bidimensional (HSQC, HMBC) techniques. We also mea-
sured NMR spectra in CDCl3, although in this solvent
significant fragmentation to 7, 8d , and 20 took place. Not
surprisingly, all our recorded data perfectly confirmed
This result was particularly satisfactory considering the
structural relationship of 12c with 2.
With the aim of fully understanding the harman-
dihydropyridine spectroscopic pattern, 12a -c were ex-
haustively analyzed by mono- and bidimensional (HSQC
and HMBC) NMR. A similar spectroscopic study was
performed with all intermediate 1,4-dihydropyridines
prepared in this work.
With the model 1,4-dihydropyridines 12a -c in hand,
attention was turned to the N-deprotection reaction.
Satisfactorily, treatment of 3,5-diacyl derivatives 12a and
12c with 1% KOH in MeOH for a short period of time
(15 min)20 smoothly gave 18a (85%) and 18c (80%,
Scheme 4). As expected from their substitution pattern,
18a and 18c were stable enough to be purified by column
chromatography, and they could be stored for several
months at -20 °C. However, exposure of 12b, bearing
only one acyl group at the pyridine â-position, to the
above deprotection conditions did not provide the desired
ethyldihydropyridine 18b. Instead, the crude product was
1
the structure 1. However, the comparison of our H NMR
data with those published for lyaline1,2 revealed differ-
ences in the chemical shifts of some protons of the
dihydropyridine ring and its substituents (Table 2). Most
notably, the protons of the vinyl methylene group of 1
appear much more shielded than those reported for
lyaline (entry 4).
1
shown by H NMR to be a 4:3:1 mixture of pyridine 8b,
harman 6, and aldehyde 19. This result indicated that
under the reaction conditions (or during the extractive
workup) dihydropyridine 18b undergoes cleavage of the
C-14/C-15 bond, probably through a retroaddition reac-
We then focused our attention on dihydropyridine 13c,
which was envisaged as a precursor of 2 (lyadine) by
N-deprotection followed by reduction of the acetyl group
(20) Lavilla, R.; Gotsens, T.; Guerrero, M.; Bosch, J . Synthesis 1995,
382-384.
754 J . Org. Chem., Vol. 69, No. 3, 2004

Synthesis of the Indole Alkaloids Lyaline and Lyadine
TABLE 2. Com p a r ison of th e 1H NMR Da ta of 1 w ith
Th ose Rep or ted for Lya lin e
corded spectroscopic data did not match those published
for lyadine.1,2 The protons of the hydroxyethyl chain in
the 1H NMR of the natural product, presumably recorded
in CDCl3, are reported2 to appear at δ 1.42 (18-H) and δ
5.2821 (19-H). However, our synthetic product 2 (mixture
of epimers) displayed a different chemical shift pattern
as the methyl group 18-H resonates at δ 1.11 or δ 1.34
in CD3OD and at δ 0.90 or δ 1.09 in CDCl3. The methine
proton 19-H is much more shielded, appearing at δ 4.21
or δ 4.35 in CD3OD and δ 4.26 in CDCl3.
In conclusion, we have synthesized the harman-
dihydropyridines 1 and 2, which constitute the originally
proposed structures for lyaline and lyadine. As could be
expected considering their dihydropyridine nature, both
compounds were very sensitive, being difficult to isolate
in pure form. In fact, in our hands these compounds (in
particular 2) were prone to undergo fragmentation in
solution into harman and pyridine. The incomplete NMR
data published for these alkaloids exhibit serious dis-
crepancies with our recorded data for 1 and are definitely
not in accordance with structure 2. A final decision about
the structure of these natural products requires a new
structure elucidation on authentic material.
entry
1
protona
14-H
1 (CD3OD, δ)
1 (CDCl3, δ)
lyalineb
δ
3.17
3.20
4.32
7.33
4.48
4.60
6.17
6.25
3.43
3.19
3.43
4.33
7.40
3.77
4.29
6.12
6.22
3.85
3.23
3.57
4.17
7.15
5.37
2
3
4
15-H
17-H
18-H
5
6
7
19-H
21-H
OMe
5.99
6.40
3.60
a
b
Biogenetic numbering, see Scheme 1. Reference 2. Solvent
not given.
SCHEME 6. Syn th esis of 2
Exp er im en ta l Section
Or ga n om eta llic Der iva tive 10. n-BuLi (1.6 M in hexane,
1.40 mL, 2.24 mmol) was slowly added to a cooled (0 °C)
solution of harman (7, 186 mg, 1 mmol) in dry THF (18 mL),
and the resulting red solution was stirred at 0 °C for 30 min.
After the solution was cooled to -78 °C, a suspension of CuCN
(183 mg, 2 mmol) in dry THF (10 mL) was added. The resulting
mixture was stirred until the reaction temperature rose to 0
°C and then cooled again to -78 °C. ZnCl2 (1 M in Et2O, 2
mL, 2 mmol) was added, and the mixture was stirred for 10
min. Then was used in the next reaction.
(Scheme 6). As anticipated, 13c could be easily converted
(87%) into the N-unsubstituted 1,4-dihydropyridine 21
in conditions identical (1% KOH in MeOH) to those that
had been effective for model 3,5-diacyldihydropyridines
12a and 12c. The next step was the reduction of the
vinylogous amide carbonyl group of 21, which was
expected to be problematic as it involved the suppression
of one stabilizing substituent of the dihydropyridine ring.
In accord with this expectation, preliminary attempts to
reduce 21 with NaBH4-CeCl3 in MeOH resulted in a
clean and fast fragmentation to harman 7 and pyridine
22, even at -40 °C. It is worth mentioning that aldehyde
20 was not detected.
We reasoned that CeCl3 not only could favor the
carbonyl reduction but also the leaving group ability of
the harman moiety by interaction with the basic (Nb)
nitrogen. Thus, dihydropyridine 21 was treated with
NaBH4 in MeOH without any additive at 0 °C for 15 min.
After the extractive workup and trituration with Et2O,
the crude product was shown by 1H NMR in CD3OD (400
MHz) or CDCl3 (300 MHz) to be the target harman-
dihydropyridine 2 (1:1 mixture of epimers) together with
variable amounts (10-30% depending on the assay) of 7
and 22.
Meth yl 1-Acetyl-4-[(â-ca r bolin -1-yl)m eth yl]-5-vin yl-1,4-
d ih yd r op yr id in e-3-ca r boxyla te (13d ). Acetyl chloride (36
µL, 0.51 mmol) was added to a solution of pyridine 8d (83 mg,
0.51 mmol) in anhydrous CH2Cl2 (5 mL) cooled at 0 °C. The
mixture was stirred at 0 °C for 30 min and then cooled to -78
°C. Organometallic 10 (1 mmol) was added by cannula, and
the mixture was allowed to slowly warm to rt for 12 h. The
reaction mixture was poured into a 1:1 mixture of 20% NH4-
OH and saturated aqueous NH4Cl (25 mL) and extracted with
CH2Cl2 (3 × 30 mL). The organic extracts were concentrated,
and the resulting residue was chromatographed (99:1 CH2Cl2-
MeOH) to give a 10:1 mixture of dihydropyridines 13d and
15d (59 mg, 30%). Both dihydropyridines were separated by
an additional chromatography (AcOEt). 1,2-Dihydropyridine
15d : 1H NMR (CDCl3, 300 MHz) δ 2.44 (s, 3H), 3.33 (dd, J )
3, 13.0 Hz, 1H), 3.44 (dd, J ) 9.9, 13.0 Hz, 1H), 3.83 (s, 3H),
4.14 (d, J ) 17.4 Hz, 1H), 4.50 (d, J ) 11.4 Hz, 1H), 5.97 (m,
1H), 6.16 (dd, J ) 11.4, 17.4 Hz, 1H), 6.50 (s, 1H), 7.30 (m,
1H), 7.58 (ddd, J ) 1.2, 7.2, 8.1 Hz, 1H), 7.67 (d, J ) 8.1 Hz,
1H), 7.69 (s, 1H), 7.83 (d, J ) 5.4 Hz, 1H), 8.12 (d, J ) 7.5 Hz,
1H), 8.24 (d, J ) 5.4 Hz, 1H), 9.94 (s, 1H). 1,4-Dihydropyridine
13d : 1H NMR (CDCl3, 500 MHz, 50 °C, assignment aided by
HSQC and HMBC) δ 2.28 (s, 3H, CH3CO), 3.28 (br t, 1H, CH2),
3.44 (dd, J ) 2.5, 14 Hz, 1H, CH2), 3.86 (s, 3H, OMe), 4.23
(dd, J ) 2.5, 8.5 Hz, 1H, pyr 4-H), 4.32 (br m, 1H, CH2d), 4.64
(d, J ) 10.5 Hz, 1H, CH2)), 6.18 (dd, J ) 10.5, 17.5 Hz, 1H,
CHd), 6.84 (br s, 1H, pyr 6-H), 7.28 (t, J ) 7.5 Hz, 1H, 6-H),
7.56 (t, J ) 8 Hz, 1H, 7-H), 7.63 (d, J ) 8 Hz, 1H, 8-H), 7.82
(d, J ) 5.0 Hz, 1H, 4-H), 8.05 (br s, 1H, pyr 2-H), 8.10 (d, J )
8 Hz, 1H, 5-H), 8.22 (d, J ) 5.0 Hz, 1H, 3-H), 10.04 (br s, 1H,
NH); 13C NMR (CDCl3, 100.6 MHz, assignment aided by HSQC
We extensively examined the NMR spectra of this
mixture in CD3OD. Significantly, harman 7 appeared
partially deuterated at the benzylic methyl group, thus
indicating that cleavage of the C-14/C-15 bond had also
1
taken place during the short H NMR measurement time
(approximately 6 min). Confirming this fast decomposi-
tion, dihydropyridine 2 was the minor component of the
mixture after the 13C NMR measurement time (2 h).
Despite all these difficulties, we successfully character-
ized our synthetic product 2. Disappointingly, our re-
(21) 19-H is reported to appear at δ 5.7 in ref 1.
J . Org. Chem, Vol. 69, No. 3, 2004 755

Bennasar et al.
and HMBC) δ 21.8 (CH3CO), 31.3 (pyr C-4), 43.4 (CH2), 52.3
(OMe), 111.9 (C-9), 112.7 (pyr C-3), 113.4 (CH2)), 113.5 (C-
4′), 119.9 (C-6′), 121.6 (C-4b), 121.7 (C-5′), 122.9 (pyr C-6),
123.8 (pyr C-5), 128.3 (C-7), 128.4 (C-4a), 132.0 (pyr C-2), 133.8
(CHd), 134.8 (C-9a), 137.6 (C-3), 140.4 (C-8a), 143.2 (C-1),
167.1 (CO), 168.3 (CO); HRMS calcd for C23H21N3O3 387.1583,
found 387.1584.
extracted with CH2Cl2 (6 × 20 mL). The organic extracts were
dried and concentrated. Pure 21 was obtained after flash
chromatography (9:1 AcOEt-MeOH): 63 mg (87%); mp 118-
1
119 °C; H NMR (DMSO-d6, 400 MHz, assignment aided by
HSQC and HMBC) δ 2.08 (s, 3H, COMe), 2.94 (dd, J ) 6.4,
12.8 Hz, 1H, CH2), 2.98 (dd, J ) 5.2, 12.8 Hz, 1H, CH2), 3.23
(s, 3H, OMe), 4.31 (dd, J ) 6.4, 5.2 Hz, 1H, pyr 4-H), 7.12 (s,
1H, pyr 2-H), 7.19 (dd, J ) 8, 7.5 Hz, 1H, 6-H), 7.35 (s, 1H,
pyr 6-H), 7.49 (dd, J ) 8, 7.5 Hz, 1H, 7-H), 7.58 (d, J ) 8 Hz,
1H, 8-H), 7.84 (d, J ) 5.2 Hz, 1 H, 4-H), 8.12 (d, J ) 5.2 Hz,
1H, 3-H), 8.15 (d, J ) 8 Hz, 1H, 5-H), 9.12 (s, 1H, pyr NH),
11.0 (s, 1H, NH); 13C NMR (DMSO-d6, 100.6 MHz, assignment
aided by HSQC and HMBC) δ 24.6 (COMe), 30.1 (pyr C-4),
41.5 (CH2), 50.6 (OMe), 106.0 (pyr C-3), 111.9 (C-8), 112.5 (C-
4), 115.0 (pyr C-5), 119.0 (C-6), 121.1 (C-4b), 121.5 (C-5), 126.7
(C-4a), 127.5 (C-7), 135.0 (C-9a), 135.5 (pyr C-2), 137.3 (C-3),
138.5 (pyr C-6), 140.2 (C-8a), 143.9 (C-1), 166.9 (CO), 195.3
(CO); HRMS calcd for C21H19N3O3 361.1426, found 361.1442.
Anal. Calcd for C21H19N3O3‚1/2H2O: C, 68.10; H, 5.44: N, 11.34.
Found: C, 68.08; H, 5.74; N, 11.21.
Meth yl 4-[(â-Ca r bolin -1-yl)m eth yl]-5-(1-h yd r oxyeth yl)-
1,4-d ih yd r op yr id in e-3-ca r boxyla te (2). NaBH4 (5 mg) was
added to a suspension of 1,4-dihydropyridine 21 (15 mg, 0.041
mmol) in MeOH (2 mL), and the mixture was stirred at rt for
15 min. The solvent was removed, and the resulting residue
was partitioned between saturated aqueous NaCl and CH2-
Cl2 and extracted with CH2Cl2. After concentration of the
organic extracts, the resulting residue was triturated with
anhydrous Et2O (10 mL) to give 2 (10 mg) as a white solid
(mixture of epimers, impurified with 10% of fragmentation
products 7 and 22): 1H NMR (CD3OD, 400 MHz, assignment
aided by HSQC and HMBC) δ 1.11 and 1.34 (2 d, J ) 6.4 Hz,
3H, CH3), 3.15-3.30 (m, 2H, CH2), 3.36 and 3.45 (2 s, 3H,
OMe), 3.94 and 4.20 (2 t, J ) 6 Hz, 1H, pyr 4-H), 4.21 and
4.35 (2 q, J ) 6.4 Hz, 1H, CHOH), 6.09 and 6.17 (2 s, 1H, pyr
6-H), 7.23 and 7.27 (2 s, 1H, pyr 2-H), 7.24 (m, 1H, 6-H) 7.53
(m, 1H, 7-H), 7.61 (d, J ) 8.4 Hz, 1H, 8-H), 7.90 (d, J ) 5.6
Hz, 1H, 4-H), 8.13 (m, 2H, 3-H, 5-H); 1H NMR (CDCl3, 300
MHz) δ 0.90 and 1.09 (2 d, J ) 6.6 Hz, 3H, CH3), 3.43 (m, 2H,
CH2), 3.86 and 3.87 (2 s, 3H, OMe), 4.13 (dd, J ) 3.6, 5.7 Hz,
1H, pyr 4-H), 4.26 (q, J ) 6.6 Hz, 1H, CHOH), 5.58 and 5.75
(2 br s, 1H, OH), 6.10 and 6.27 (2 d, J ) 4.2, 1H, pyr 6-H),
7.27 (m, 2H, pyr 2-H, 6-H), 7.56 (m, 1H, 7-H), 7.62 and 7.66 (2
d, J ) 8.1 and 8.4 Hz, 1H, 8-H), 7.84 (m, 1H, 4-H), 8.12 (d, J
) 7.5 Hz, 1H, 5-H), 8.25 and 8.28 (2 d, J ) 5.4 Hz, 1H, 3-H),
10.37 and 10.80 (2 br s, 1H, NH); 13C NMR (CD3OD, 100.6
MHz, assignment aided by HSQC and HMBC) δ 21.8 and 22.2
(CH3), 33.0 and 35.8 (pyr C-4), 43.6 and 43.9 (CH2), 51.3 and
51.4 (OMe), 68.7 and 70.2 (CHOH), 99.7 and 100.6 (pyr C-3),
112.8 (C-8), 114.2 (C-4), 120.3 and 121.6 (pyr C-5), 120.6 and
120.7 (C-6), 122.4 (C-4b), 122.5 (C-5), 123.6 (pyr C-6), 129.2
and 129.3 (C-7), 129.6 (C-4a), 136.9 and 137.0 (C-9a), 137.4
and 137.6 (C-3), 140.1 and 140.3 (pyr C-2), 142.3 (C-8a), 145.5
(C-1), 171.0 and 171.3 (CO); ESI-MS m/z 363, 345, 182.
Met h yl 4-[(â-Ca r b olin -1-yl)m et h yl]-5-vin yl-1,4-d ih y-
d r op yr id in e-3-ca r boxyla te (1). Na0 (5 mg) was added to a
solution of 1,4-dihydropyridine 13d (20 mg, 0.052 mmol) in
MeOH (1.5 mL). The mixture was stirred at rt for 2 min,
quenched with H2O, and extracted with AcOEt. Concentration
of extracts gave 1 (15 mg) as an orange solid (impure with
10% of fragmentation products 8d , 7 and 20, see text). The
deprotection experiment was reproduced in an NMR tube in
CD3OD: 1H NMR (CD3OD, 500 MHz, assignment aided by
HSQC and HMBC) δ 3.17 (dd, J ) 7, 13 Hz, 1H, CH2), 3.20
(dd, J ) 5.5, 13 Hz, 1H, CH2), 3.43 (s, 3H, OMe), 4.32 (dd, J )
7 and 5.5 Hz, 1H, pyr 4-H), 4.48 (d, J ) 17.5 Hz, 1H, CH2d),
4.60 (d, J ) 11 Hz, 1H, CH2d), 6.17 (dd, J ) 11, 17.5 Hz, 1H,
CHd), 6.25 (s, 1H, pyr 6-H), 7.22 (t, J ) 7.5 Hz, 1H, 6-H),
7.33 (s, 1H, pyr 2-H), 7.52 (t, J ) 7.5 Hz, 1H, 7-H), 7.60 (d, J
) 8 Hz, 1H, 8-H), 7.87 (d, J ) 5.5 Hz, 1H, 4-H), 8.11 (m, 2H,
3-H, 5-H); 13C NMR (CD3OD, 100.6 MHz, assignment aided
by HSQC and HMBC) δ 32.8 (pyr C-4), 42.9 (CH2), 51.5 (OMe),
101.4 (pyr C-3), 108.5 (CH2d), 113.0 (C-8), 114.1 (C-4), 118.0
(pyr C-5), 120.6 (C-6), 122.5 (C-4b, C-5), 127.5 (pyr C-6), 129.3
(C-7), 129.8 (C-4a), 136.4 (CHd), 136.9 (C-9a), 137.8 (C-3),
139.5 (pyr C-2), 142.3 (C-8a), 145.5 (C-1), 171.0 (CO); 1H NMR
(CDCl3, 300 MHz) δ 3.19 (dd, J ) 9.3, 13.0 Hz, 1H, CH2), 3.34
(d, J ) 13.0 Hz, 1H, CH2), 3.77 (d, J ) 17.4 Hz, 1H, CH2d),
3.85 (s, 3H, OMe), 4.29 (d, J ) 11.1 Hz, 1H, CH2d), 4.33 (d, J
) 9.3 Hz, 1H, pyr 4-H), 6.12 (dd, J ) 11.1, 17.4 Hz, 1H, CHd
), 6.22 (s, 1H, pyr 6-H), 7.28 (t, J ) 8.1 Hz, 1H, 6-H), 7.40 (s,
1H, pyr 2-H), 7.56 (ddd, J ) 0.9, 7.8, 8.1 Hz, 1H, 7′-H), 7.70
(d, J ) 7.8 Hz, 1H, 8-H), 7.82 (d, J ) 5.1 Hz, 1H, 4-H), 8.12 (d,
J ) 8.1 Hz, 1H, 5-H), 8.18 (d, J ) 5.1 Hz, 1H, 3-H), 10.65 (br
s, 1H, NH); 13C NMR (CDCl3, 75.4 MHz, intensive fragmenta-
tion) δ 30.5, 45.5, 51.6, 102.3, 109.3, 112.1, 113.2, 118.1, 119.5,
121.6, 121.6, 125.2, 127.8, 127.9, 134.6, 134.9, 137.0, 137.6,
140.4, 145.0, 169.5; ESI-MS m/z 345 (M+), 182, 163.
Meth yl 1,5-Dia cetyl-4-[(â-ca r bolin -1-yl)m eth yl]-1,4-d i-
h yd r op yr id in e-3-ca r boxyla te (13c). Acetyl chloride (36 µL,
0.51 mmol) was added to a solution of pyridine 8c (90 mg, 0.50
mmol) in anhydrous CH2Cl2 (5 mL) cooled at -10 °C, and the
mixture was stirred at -10 °C for 5 min. Trimethylsilyl triflate
(90 µL, 0.50 mmol) was added, and the mixture was stirred at
-10 °C for 1 h. The resulting pyridinium triflate (11c) was
allowed to react as above with 10 (1 mmol). After the extractive
workup and chromatography of the residue (98:2 AcOEt-
MeOH), a 9:1 mixture of 13c and 17c (60 mg, 30%) was
obtained. Pure 13c was obtained after an additional chroma-
tography (98:2 AcOEt-MeOH): 1H NMR (CDCl3, 300 MHz) δ
2.41 (sa, 3H), 2.44 (s, 3H), 3.05 (dd, J ) 8.7, 13 Hz, 1H), 3.22
(br s, 3H), 3.28 (dd, J ) 3, 13 Hz, 1H), 4.37 (dd, J ) 3, 8.7 Hz,
1H), 7.28 (ddd, J ) 0.9, 6.9, 8 Hz, 1H), 7.56 (ddd, J ) 1.2, 6.9,
8.1 Hz, 1H), 7.69 (d, J ) 8.1 Hz, 1H), 7.7-7.9 (br s, 2H), 7.82
(d, J ) 5.4 Hz, 1H), 8.11 (d, J ) 8 Hz, 1H), 8.20 (d, J ) 5.4 Hz,
1H), 10.15 (s, 1H); 13C NMR (CDCl3, 75.4 MHz) δ 21.8, 25.4,
30.7, 43.9, 51.8, 111.9, 113.4, 115.8, 119.6, 121.6, 121.7, 123.4,
128.0, 128.0, 131.1, 132.4, 134.7, 137.4, 140.3, 143.2, 166.1,
167.3, 197.7; HRMS calcd for C23H21N3O4 403.1532, found
403.1518.
Ackn owledgm en t. Financial support from the “Min-
isterio de Ciencia y Tecnolog´ıa” (Spain, Project No.
BQU2000-0785) is gratefully acknowledged. We also
thank the DURSI (Generalitat de Catalunya) for Grant
No. 2001SGR00084 and for a fellowship to M.M.
Su p p or tin g In for m a tion Ava ila ble: General experimen-
tal protocols and detailed experimental procedures for the
Meth yl 5-Acetyl-4-[(â-ca r bolin -1-yl)m eth yl]-1,4-d ih y-
d r op yr id in e-3-ca r boxyla te (21). Dihydropyridine 13c (80
mg, 0.2 mmol) was added to a solution of KOH in MeOH (1%,
6 mL) and stirring was maintained at rt for 15 min. The
reaction mixture was poured into 5% Na2CO3 (15 mL) and
1
preparation of 12a -c, 18a ,d . H and 13C NMR spectra of all
new compounds. ESI-MS spectra of 1 and 2. This material is
available free of charge via the Internet at http://pubs.acs.org.
J O035507H
756 J . Org. Chem., Vol. 69, No. 3, 2004